RESEARCH IN

CHEMICAL KINETICS

Volume 3

Edited by

R.G. COMPTON and G. HANCOCK

Physical Chemistry Laboratory, University of Oxford,
Oxford 0X13QZ, U.K.

1995
ELSEVIER
Amsterdam - London - N e w York - Oxford - Shannon - Tokyo
 ELSEVIER SCIENCE ΒV
Sara Burgerhartstraat 25
P.O. Box 211,1000 AE Amsterdam, The Netherlands

ISBN: 0-444-82036-1

© 1995 Elsevier Science BV. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the
publisher, Elsevier Science BV, Copyright & Permissions Department, P.O. Box 521, 1000 AM Amsterdam,
The Netherlands.

Special regulations for readers in the U.S.A. - This publication has been registered with the Copyright Clearance
Center Inc. (CCC), 222 Rosewood Drive, Danvers MA 01923. Information can be obtained from the CCC
about conditions under which photocopies of parts of this publication may be made in the U.S.A. All other
copyright questions, including photocopying outside of the U.S.A., should be referred to the copyright owner,
Elsevier Science B.V., unless otherwise specified.

No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of
products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or

ideas contained in the material herein.

This book is printed on acid-free paper.

Printed in The Netherlands

 ν

PREFACE

This series of volumes aims to produce authoritative review articles on a wide range
of newly developing topics in the kinetics of both gaseous and condensed phases. Each article
will describe a particular area of the subject pertinent to the research interests and expertise
of the contributors, emphasising their recent contributions and putting this work in context
of others' progress in the same field. The reviews are aimed at a wide general readership in
the kinetics c o m m u n i t y and are intended to be short, topical accounts of a specific area from
the viewpoint of an expert in the field. The Editors and Elsevier are committed to rapid
publication of these volumes so as to ensure the highest possible benefit to the kinetics
community.

Richard Compton
G u s Hancock
 vii

LIST OF CONTRIBUTORS

R. Becerra,
Instituto Quimica Fisica "Rocasolano", C . S . I . C , C/Serrano, 119, 28006, Madrid, Spain.

Dario T. Beruto,
Interdepartmental Centre for Materials Engineering and Institute of Chemistry, Faculty of
Engineering, University of Genoa, Piazzale J.F, Kennedy, Fiera del Mare Pad. D, Genova,
Italy.

Joseph L. Durant Jr.,

Combustion Research Facility, Sandia National Laboratory, Livermore, Ca 9 4 5 5 1 , U S A .

Marino Giordani,
Interdepartmental Centre for Materials Engineering and Institute of Chemistry, Faculty of
Engineering, University of Genoa, Piazzale J.F. Kennedy, Fiera del Mare Pad. D, Genova,
Italy.

Nicholas J.B. Green,

Department of Chemistry, K i n g ' s College London, Strand, London W C 2 R 2 L S , United
Kingdom.

Simon M. Pimblott,
Radiation Laboratory, University of Notre D a m e , Indiana 46556, U S A .

J. T a m m ,
Institute of Physical Chemistry, University of Tartu, Tartu, E E 2 4 0 0 , Estonia.

L. T a m m ,
Institute of Physical Chemistry, University of Tartu, Tartu, E E 2 4 0 0 , Estonia.

R.W. Walker,
School of Chemistry, University of Hull, Hull, North Humberside, H U 6 7 R X , United
Kingdom.

R. Walsh,
Department of Chemistry, University of Reading, Whiteknights, P.O. Box 224, Reading R G 6
2 A D , United Kingdom.
Research in Chemical Kinetics, Volume 3
R.G. Compton and G. Hancock (editors) 1
© 1995 Elsevier Science B.V. All rights reserved.

Some Burning Problems in Combustion Chemistry

K. W. Walker

School of Chemistry, University of Hull, Hull, North Humberside, H U 6 7RX, United

Kingdom

1. INTRODUCTION

The well k n o w n [1] extreme complexity of hydrocarbon combustion processes has

rendered it extraordinarily difficult t o identify quantitative mechanisms and t o obtain rate
constant parameters even for key elementary reactions in order that important practical
systems can be modelled [2]. Very little progress has been made (over nearly 100 years) in
understanding the fundamentals of combustion chemistry by direct investigations of the
oxidation of hydrocarbons and related compounds. In the period up t o about 1970, virtually
no reliable kinetic data on the elementary processes involved w e r e obtained in this w a y despite
its prevalence in hydrocarbon oxidation studies. Even the advent of g a s chromatography and
other sensitive analytical techniques, which enabled quantitative measurement o f products in
the very early stages of oxidation, did not lead to a significantly increased bank of fundamental
kinetic and mechanistic data. [2]
In general, major advances in the understanding of combustion mechanisms have resulted
only when single elementary steps have been isolated under carefully selected and controlled
conditions, where the reaction products (often formed in multi-channels) can be identified and
monitored. [3] Obviously, a reliable and clean source of the radical is essential under the
necessary conditions, in terms of (i) the determination of a rate constant o f the reaction, and
(ii) the use of the rate constant. The extrapolation of the value of a rate constant determined
near r o o m temperature t o the region of high temperature combustion is often a dangerous
operation and frequently carried out with foolish optimism. The realisation that many
reactions have non-Arrhenius temperature coefficients, and perhaps a change of mechanism
with temperature, has meant that some key reactions have been studied over very wide
temperature [3] and pressure conditions [4]. T h e effect o f temperature on t h e mutual reaction
of H 0 radicals ( F i g . l ) merits attention and gives an early warning in connection with a
2

process of great importance in combustion. In outline, Foner and H u d s o n [5] obtained k =

9 3 1 1 9
1.8 χ 1 0 d m m o l " s" (1962) at 300 Κ supported by Paukert and Johnson [6] (2.2 χ 1 0 ,
9
1972) which when combined with T r o e ' s [7] shock-tube value o f 2 χ 1 0 at ca 1200 Κ
9 3 ± 0 3 3 1 1
produced unanimous agreement that A = 1 0 d m m o l " s ' , Ε = 0 over the temperature
range 300-2000 K. These parameters w e r e still recommended in 1984, despite some evidence
t o the contrary [8]. However, a number of w o r k e r s [3] demonstrated a marked negative
temperature dependence between 250 and 550 K, and moreover emphasised that the rate
constant w a s strongly dependent on the bath gas used and the total pressure. With T r o e ' s
2

data point regarded with some suspicion, the uncertainty in the value of the rate constant
exceeded an order of magnitude at combustion temperatures. Fortunately, Lightfoot et al [9]
demonstrated a minimum w a s reached at about 700 K, and T r o e [10] h a s obtained n e w data
up to 1100 Κ and validated his earlier work. The cause of the complex variation is
undoubtedly a mechanism change. At low temperatures, re-combination occurs with the
H 0 complex becoming unstable at higher temperatures, and an Η abstraction process then
2 4

dominates. The danger of extrapolation in such a situation is clearly apparent.

H0 2 + H0 2 + Μ -> H 0
2 4 + Μ

i
H 0 2 2 + 0 2

10.0

pre -1980 line

0 1 2 3 U
103 K / T

Figure 1 30 years of kinetic data for 2HO2 -> H2O2 + O2

(1 atmosphere pressure, mostly N2).
 3

A second c o m m o n pre-requisite is the availability of selective and very sensitive detection

of atoms and free-radicals in order t o monitor the p r o g r e s s of a reaction. Reactions of small
species important in combustion (eg. H , O, O H , H 0 , C H , N H , C H etc) have been studied
2 3

extensively because of the large variety o f w a y s t o generate t h e m (eg. microwave discharge,

laser photolysis etc[3]) and the sensitive, although essentially simple, m e t h o d of monitoring
their concentration (eg. resonance fluorescence, laser magnetic resonance[3]). Even within
the small species, there is often a marked distinction t o b e made. Whilst the kinetics of Η
abstraction from a vast range o f R H c o m p o u n d s by the highly reactive O H radical are in g o o d
order[3, 11], even t o the extent of accurate expressions for attack at specific C - H sites, there
is a dearth of information on the slower Η abstraction reactions of H , H 0 and C H . For 2 3

example, despite their key role in determining ignition time delays [12], there are n o data
points for H 0 + C H and only o n e for H 0 + C H [13] compared with the several
2 4 1 0 2 4

hundred data points for O H + C H between 240 and 2000 Κ with k k n o w n t o ca 2 0 % [3].
4

The H 0 reactions are simply t o o slow for the normal techniques. H o w e v e r , the reactions
2

largely determine the ignition time delay through the sequence,

H0 2 + RH -> H 0 2 2 + R

H 0
2 2 + Μ -> 20H + Μ

and literally the system is 'kept waiting' for the branching agent H 0 t o build up in 2 2

concentration. Even for Η + R H , there are few data, despite its importance in oxidation and
thermal cracking processes. Cohen [14] comments that "the only reliable rate constants for
Η + C H n w h e r e n > 3 is that of Baldwin and Walker obtained from their classical H + 0
2 n + 2 2 2

addition studies at about 750 K " (see later).

Reactions o f Ο a t o m s (and other radicals) are o f particular importance in high temperature
combustion processes such as oxy-acetylene flames which operate at about 3000 K. The
reactions occur in collisions with very high initial translational energies (> 50 kJ m o H ) .
3 1
Although reactions o f ground state Ο ( P ) a t o m s with translational energies < 40 kJ m o l '
have been studied extensively [16], a source of m o r e energetic a t o m s h a s only just b e c o m e
available [17].
Far less quantitative kinetic information is available on the reactions o f the larger alkyl
(and related) radicals. A b o v e 1000 K, this is not particularly a handicap because homolysis
will normally occur rapidly and most alkyl radicals will b e reduced effectively t o the
prototypical C H , C H 1-C3H7 and t - C H species which will then u n d e r g o oxidation.
3 2 5 4 9

However, b e l o w 1000 K, w h e n the larger radicals c o m e m o r e thermally stable, the lack of

data provides a very serious barrier t o detailed modelling. T h e underlying causes arise from
the significantly greater difficulty involved in generating 'clean' sources o f the larger species
under conditions suitable for studying their reactions quantitatively and from the dearth of
methods for t h e quantitative monitoring o f the radical concentrations. B e y o n d C5 alkyls, few
rate constants are available for reactions listed in Table 1. F o r example, Baldwin, Walker and
Walker [18] summarise the relatively few experimentally determined kinetic data for the
4

homolysis of large, branched alkyl radicals. Gutman and Slagle [19, 20] have used the elegant
technique of laser photolysis coupled with radical detection by photo-ionisation mass
spectrometry to study reactions of unsaturated radicals in an 0 environment. 2

2. F O U R C R U C I A L F A C T O R S IN M O D E L L I N G

Apart from the value of a rate constant, four other factors are of crucial importance to
modellers concerned with the quantitative interpretation of combustion systems,
(a) Calculation of the rate of attack at specific sites in R H is vital using expressions of the
type

+ A e E s / R T Ε Τ
overall = *p °s s " + «t ^ ^ (i)

where n , n and n refer to the number of primary, secondary and tertiary C-H bonds,
p s t

respectively, in RH. The O H + R H data are considered so reliable that realistic attempts have
been made to allow for the effect of near-neighbour groups [15]. For example, for Η
abstraction from a C H group in an alkane, it is possible to assign Arrhenius parameters for
2

E / R T
the following, inter alia, three environments, based on k = A T e ~ .

3 1 1 1 1
A / d m mol" s" K" E/kJ mol"

6
OH + CH CH CH
3 2 3 2.90 χ 1 0 1080

6
OH + CH CH CH -
3 2 2 2.90 χ 1 0 500

6
OH + -CH CH CH -
2 2 2 2.90 χ 1 0 -90

Access to this type of data is crucial when modelling combustion. For example, in
propanol combustion, the proportion of C H C H O H , C H C H O H and C H C H 0 radicals can
2 2 3 3 2

be calculated. Their reactions are significantly different, C H C H O H reacts rapidly with 0 to 3 2

give C H C H O and the inert H 0 , and the other t w o radicals undergo mainly unimolecular
3 2

decomposition. Unfortunately while the data for O H + alkanes have reached an advanced
level, those for attack at other organics are much less developed except at low temperatures
because of the importance of the O H radical in atmospheric chemistry [11]. Although Cohen
has stressed the paucity of data for H + R H , they have been efficiently used t o extrapolate the
available rate constants between 250 and 2000 Κ by use of transition state theory [14]. He
suggests that the entropy of activation per attackable Η atom increases slightly from C H t o 4

C5H12 by about 10 J K" mol" 298 Κ and then stays virtually constant at 100-103 J K"
1 1 1

1
mol" . Under these conditions the bond additively concept can be used and Cohen gives a
general expression for the reactions of Η with large (C>4) alkanes which allows for non-
Arrhenius behaviour, equation (ii).
 5

3 2 0 3 2 2
k = 5.4 χ 1 0 η Τ exp (-3540/Τ) + 4.7 χ 1 0 n T
β s exp (-2640/T)
3 0 3 1 1
+ 3.7 χ 1 0 n T * exp (-970/T) d m mol" s"
t (ii)

Walker has given a simple database expression based on results obtained from studies of
H0 2radicals with C H , H C H O , and 4[21].

9 3 1 1
k = 2.8 χ 1 0 [ n exp(-10355/T) + n exp (-8900/T) + n exp (-8060/T)] d m mol" s"
p s t (hi)

The development of expressions for other atoms and radicals, Ο + R H apart, which allow
for near-neighbour effects is still a distant prospect.
(b) A knowledge of the products of a reaction and of the branching ratios in multi-channelled
processes is essential when modelling combustion. This is particularly pertinent when one
channel will lead t o termination and another t o branching, as for example in the important high
temperature reaction between O H and C H radicals. 3

OH + CH 3 -> Η + CH OH
2

-> Η + CH 0 3

1
-> H 0 2 + CH 2

-> CH OH 3

Although the recombination reaction h a s been studied in detail [3,22], the first three
1 0 3 1
channels cannot be distinguished and have a combined rate constant o f 3.6 χ 1 0 d m mol"
1
s ' over the range 300-2000 Κ [3]. Similar problems are found with t h e reactions C H + 0 2 2

and Η + C H C O , both key reactions in acetylene oxidation. Determination of reaction

products is, however, often difficult and time-consuming.
(c) Many of the kinetic data on recombination and dissociation processes, eg. C H + C H , 3 3

have been obtained at atmospheric pressure or below. Extrapolation t o t h e higher pressures

found in practical combustion units, such as the petrol engine, is extremely difficult despite the
availability of several admirable theories of unimolecular reactions. M a n y modellers have
found the theories difficult t o apply and are grateful to T r o e [23] and others w h o have
developed reliable 'user-friendly' methods, based on fundamental theory [3].
(d) The sheer complexity of many comprehensive chemical mechanisms renders them
effectively unusable for the modelling of many practical combustion problems, despite the
huge and rapid computing facilities available. The auto-ignition of butane h a s been modelled
by use of a mechanism containing about 500 elementary reactions and 60 different species
[24]. Sensitively analysis t o isolate key processes in the mechanism under the conditions
used is vital. In principle, these reactions then b e c o m e the focus of attention for kineticists.
In the butane example, precise data are only needed for about 10 of t h e 500 reactions
involved. W a r n a t z [25] has been particularly active in this area and h a s developed computer-
aided methods of mechanism reduction. Pilling et al also provide a particularly g o o d example
of this approach [26].
6

3. GENERAL MECHANISM OF HYDROCARBON OXIDATION

It is impossible t o give a relatively simple general mechanism of hydrocarbon oxidation

under all conditions. At high temperatures (> 2000 K ) , radical-radical processes are of
particular importance because of the very high radical concentrations. H o w e v e r , it is in the
auto-ignition region (ca 600-1200 K) that chemistry plays a major role in the combustion
phenomena observed, and this review will focus sharply on this temperature range [27]. At
the lower temperatures, the chemistry is intimately concerned with cool-flames, ' k n o c k ' in
engines and combustion-generated pollution [28], and above 1000 Κ with flames and shock-
tube measurements. It is from studies in the middle region (650-820 K ) that the author has
obtained a vast amount of kinetic and mechanism data relevant to combustion chemistry. The
review will be developed around some of the more recent and important o f these studies.

Table 1
Basic mechanism for hydrocarbon oxidation between 500 and 1000 Κ

RH + 0 2 -> R + H0 2 (1)
RH -> R' + R" (1A)
X + RH -> XH + R (2)
R (+ M) -> R' + Α Β ' (+ M ) (3)
R (+ M)
-» Η + A B ( + M) (4)
R + 0 2 -> AB + H0 2 (5A)
R + o 2
-> ABO + OH (5B)
R + 0 2 -» OR OH (5C)
R + 0 2 (+ M ) -> R 0 (+ M )
2 (6)
R0 2 -> AB, ABO, OR + X (7A,B,C)
R0 + RH
2 -» ROOH R (8)
R0 2 -> QOOH (9)
QOOH -> AB, ABO, OR + X (10A,B,C)
QOOH + o -> 0 QOOH
2 (10D)
2
R0 "+ R 0 chain termination
2 2
(Π)
ROOH -> RO OH (12)
H 0 2+ Μ 2 —» 20H Μ (13)
H0 + H0
2 2 -> H 0 2 2
+ 0 2 (14)
Η + o -> OH + Ο (15)
2
H0 + RH
2 -> H 0 2 2
+ R (16)

A B , alkene; A B O , oxygenated compound; OR, oxygen-ring compound. X is a radical usually

O H or H 0 ; R is the radical formed from RH. R ' , R " and A B ' contain a smaller number of
2

carbon atoms than RH.

 7

The general mechanism within the auto-ignition region has been accepted since the late
1960s and is shown in Table 1.
Under many conditions, reaction 6 is equilibrated and may be considered as the 'heart of
combustion chemistry'.

R + 0 2 R0 2

Table 2 shows values of the equibrium constant for a range of alkyl radicals, together with
the 'ceiling temperature' when [R] = [ R 0 ] for 0 = 0.1 atmosphere. At l o w temperatures,
2 2

combustion chemistry is centred on reactions o f R 0 radicals, so that reactions (8)-(12), are 2

particularly dominant and lead to large yields of oxygenated products such as O-heterocyclic
compounds and organic peroxides [29]. A s the temperature is raised the equilibrum position
moves t o w a r d s R radicals and reactions (3)-(5) lead t o major amounts of conjugate alkenes
and homolysis p r o d u c t s [30]. The 'switch' in the chemistry is also closely related t o the well-
known occurrence of cool flames and negative-temperature coefficients in rates of reaction
[31].

Table 2
= :
Data for the equibrium R + O2 = RC<2

„ ΔΗ? 9 8 , v l t .γ Ceiling t e m p / K
kJmoH , o g K / a t m
PWQd
at O = 0 . 1 atm
2

500 Κ 800 Κ 1200 Κ

H[3] 208 16.8 8.6 4.1 1920

CH [3] 3 - 135 7.4 2.1 - 0.85 930
C H *
2 5 - 147 7.6 1.9 - 1.30 900
i-C H **3 7 - 155 7.7 1.6 - 1.75 860
t-C H * 4 9 - 153 7.1 1.1 - 2.3 820
CH CHCH **
2 2 - 75 1.6 -1.35 - 3.0 550

* I . R Slagle, Ε. Ratajczak and D. Gutman, J. Phys. Chem., 90 (1986) 402

** I . R Slagle, E. Ratajczak, M . C . Heaven, D. Gutman and A.F. Wagner, J.Am. Chem. S o c ,
1 0 7 ( 1 9 8 5 ) 1838.

Table 2 reveals t w o other features.

8

(i) The 'ceiling' temperature for H 0 radicals is considerably higher than that for R 0
2 2

radicals. In consequence, H 0 chemistry is still important at temperatures in excess of

2

1000 K.
(ii) The 'ceiling temperatures' for electron-delocalised radicals such as allyl and benzyl are
considerably lower (ca 300 K) than for alkyl radicals. A s a result, and arising from other
aspects of their reactivity, both allyl and benzyl radicals undergo radical-radical reactions
under conditions where, for example, ethyl radicals would react almost completely with 0 . 2

With a mixture containing 4, 30 and 26 Torr of C H , 0 and N , respectively at

3 6 2 2

750 K, the initial yield of h e x a - 1 , 5-diene is about 2 0 % , indicating that at least 4 0 % of allyl
radicals have recombined [32]. A s reaction (18) is also important in removing allyl radicals,
and nearly half of the C H is removed through addition reactions (19) and (20), virtually all
3 6

the allyl radicals are removed in radical-radical processes.

X + C H 3 6 -> XH + CH CHCH 2 2 (2p)

2CH CHCH 2 2 -> CH =CHCH CH CH=CH

2 2 2 2 (17)

CH CHCH
2 2 + H0 2 -> products (18)

2
OH + C H 3 6 ° > CH CHO
3 + HCHO + OH (19)

H0 2 + C H 3 6 -> C H 0
3 6 + OH (20)

4. O U T L I N E O F S O M E ' H U L L ' A P P R O A C H E S

References t o the w o r k of the Hull group will be made frequently and it is pertinent to
outline the experimental approaches used.

4.1. Addition to H + 0 mixtures at about 7 5 0 Κ

2 2

This approach has been used extensively over the last 30 years. U s e of trace amounts o f
additive ( R H ) and an aged boric-acid-coated vessel permits investigations of the oxidation of
alkanes and related compounds in the total absence of surface affects in a constant and
controllable radical environment determined by the H + 0 mixture. This is in marked 2 2

contrast to the direct oxidation method where the radical environment is controlled by the
oxidant and changes constantly as the reaction intermediates are formed and then oxidised.
There are t w o aspects of the approach.
(i) F r o m measurements o f the relative rate of consumption o f H and R H , kinetic data are 2

obtained for O H + R H and Η + R H , (and in some cases H 0 + R H ) [33,34]. A s indicated 2

earlier, Cohen [14] has commented on the value of the Η + R H data.

(ii) Detailed product analyses over a wide range of mixture composition give mechanistic and
kinetic information on the reaction of R radicals in an oxidising environment [33-35].
Limitation t o ranges of 730 - 770 Κ and 250 - 600 Torr restricts the approach, particularly for
the study of pressure effects.
 9

4.2. Decomposition of tetramethylbutane ( T M B ) in the presence of 0 2

1
The central C-C bond in T M B is sterically strained (ca 80 kJ m o l ' ) so that homolysis
occurs at a convenient rate at about 750 K. 9 9 % of the t - C H radicals formed react with 0 4 9 2

to give i-butene and H 0 . The sequence 2

TMB -> 2t-C H 4 9

t-C H4 9 + 0 2 -> i-C H 4 8 + H0 2

thus provides a clean and reliable source of H 0 radicals. The system has been used to obtain 2

the only reliable kinetic data for H 0 + alkane [36]. In the case of C H addition, the
2 2 6

experimental measurement involves only the relative rate of formation of C H and i-butene. 2 4

H0 2 + C H 2 6 -> H 0 2 2 + C H 2 5 (2e)

1 0 1 3 3 1 1
Over the temperature range 670-770 K, values of A = 1 0 d m m o l ' s ' and Ε = 86 2 e

1
kJ m o l ' w e r e obtained [37].
This system has also given the only kinetic data for the reaction involving addition of H 0 2

radicals to alkenes t o give oxiranes [38]. Related studies of the decomposition of

trimethylbutane and 2,3-dimethylbutane have given accurate values for the heats of formation
of i - C H and t - C H radicals [39-40].
3 7 4 9

H0 2 + C H n 2 n -> C H n 2 n O + OH

4.3. Decomposition of 4, 4-dimethylpentene-l in the present of 0 2

This system has given the first reliable kinetic data for the oxidation chemistry of allyl
radicals in the range 650-800 K.

CH = CHCH C(CH )
2 2 3 3 -> CH CHCH 2 2 + t-C H 4 9

Arising from the steric strain due t o the t - C H group and the latent 'stabilisation' energy 4 9

in the allyl radical, the homolysis is considerably faster than normal. A s t-butyl radicals react
uniquely with 0 ( 9 9 % ) t o give i-butene + H 0 (see above), then the chemistry of
2 2

C H C H C H radicals in the presence of 0 and a clean source o f H 0 can be studied with

2 2 2 2

ease. The rate constants of the multi-channelled reaction between H 0 + C H C H C H have 2 2 2

been obtained and it has been shown that reaction (5A) for allyl radicals is very slow [41,42].

CH CHCH 2 2 + 0 2 -> CH = C = CH
2 2 + H0 2 (5Aa)
10

4.4. Direct studies of hydrocarbon oxidation

In general, direct studies of hydrocarbon oxidation have p r o d u c e d very little reliable
kinetic and mechanistic information. However, in specific cases, under very carefully selected
conditions, particular reactions can be isolated, mechanisms elucidated, and kinetic data
obtained for key elementary reactions. O f relevance here is the oxidation o f C H over the 3 6

temperature range 650-800 Κ and at pressures n o greater than 60 Torr.

At these temperatures, the decomposition of H 0 , which corresponds to secondary 2 2

initiation is limited particularly by the use of low pressures. The C H C H C H + 0 2 2 2

C H = C H C H 0 is well to the left (Table 2) and radical branching via for example the
2 2 2

sequence

H 0 2 2 + Μ -> 20H + Μ (13)

CH = CHCH 0 2 2 2 + RH -> CH = CH CH OOH

2 2 2 + R
I
CH = CH CH 0
2 2 2 + OH

is negligible. A s discussed later, under the carefully chosen conditions, the direct oxidation of
propene can be used for the determination of the Arrhenius parameters o f the primary
initiation reaction (lpe)[32].

C
3 6 H
+ 0 2 -> CH CHCH 2 2 + H0 2 (lpe)

The direct oxidation of H C H O [43], i-butene [44] and of other c o m p o u n d s will also be
discussed later. The co-oxidation of alkenes (in the presence or absence o f tetramethylbutane
as a source of H 0 radicals, section 4.2) has given excellent and unique Arrhenius parameters
2

for the addition of H 0 radicals to alkenes to form oxiranes [38].

2

H0 2 + C H n 2 n -> C H n 2 n O + OH

In all these studies, it is possible to deal parochially with only part of the total mechanism
independently of the remainder.

5. M A J O R F E A T U R E S O F T H E R E V I E W

Attention will be focused on three areas of current interest,

(i) The kinetics of the primary initiation reaction. N o major review has been made of the
homogeneous initiation reactions (1) and (1A).

RH + 0 2 -> R + H0 2 (1)

RH -» R' + R" (1A)

 11

Modellers have demonstrated the kinetic importance of both reactions, and it is timely to
examine the available data critically, particularly as there is sufficient information for kj t o
warrant the establishment of a database.
(ii) The reaction o f C2H5 radicals with 0 h a s exercised the minds o f combustion chemists for
2

the last 30 years, and since 1984 has been the subject of a considerable number of
experimental and theoretical kinetic studies. Although there is something o f a consensus view
that the mechanism has been established, a large number of experimental observations cannot
be satisfactorily explained. It is important t o realise that C H + 0 is prototypical o f the
2 5 2

family of reactions of alkyl radicals with 0 , and indeed is clearly related t o the reaction of 0
2 2

with species such as R O , alkenyl, and hydroxyalkyl. In general, attention h a s unfortunately

been focused almost entirely on C H + 0 , and here relevant experimental results on the
2 5 2

related systems will b e presented.

(iii)Cyclisation reactions of aliphatic species are of key importance in the formation of soot
and polyaromatic hydrocarbons ( P A H ) at temperatures above about 1200 K. Although soot
formation is relatively unimportant below this temperature, the production of benzene and
related aromatics is of great concern in terms of pollution and health hazards. However, little
attention has been paid to low-temperature cyclisation of aliphatic radicals in the past.

6. P R I M A R Y I N I T I A T I O N R E A C T I O N S

Reactions (1) and (1A) are considered the normal primary initiation processes in
hydrocarbon oxidation [45] and should be distinguished from secondary initiation process
such as (13) where radicals are formed from 'stable' species produced as intermediates. (1)
and (1A) are key steps in determining the onset of second-stage ignition [46] and recently
Emdee, Brezinsky and Glassman [47], from a modelling study of the oxidation of toluene at
about 1200 K, have claimed that the rate is most sensitive t o the value of ( I t ) .

RH + 0 2
-» R + H0 2 (1)

RH -> R' + R" (1A)

H 0
2 2
+ Μ -> 20H + Μ (13)

C H CH
6 5 3 + 0 2 -» C H CH
6 5 2
+ H0 2 (it)

However, in practice, the picture is considerably more complex.

(i) Below ca 6 0 0 Κ, (1) and (1A) may occur at the surface with considerably reduced
activation energies. This has been responsible for dramatic changes in oxidation rate when
different surfaces are used [1], the frequently observed lack of reproducibility, and for
6
reported values of k j and k j which are a factor of 1 0 or m o r e higher than t h e h o m o g e n e o u s
A

value of k i [45,48].
Low-temperature studies by Dixon, Skirrow and Tipper [49] gave values of k j increasing a

6 5 5 4 3 1 1
from 7.4 χ 10" t o 2.9 χ 10" at 345 Κ and from 6.1 χ 10" t o 5.1 χ 10" d m mol" s" at 393
12

1
Κ when the surface to volume ratio of the reaction vessel w a s increased from 0.6 t o 6.1 c m ' .
9
The values are ca 1 0 times t o o high and correspond t o activation energies of about 55 kJ
1
m o l ' (see later).

CH CHO3 + 0 2 -> CH3CO + H0 2 (la)

(ii) Reactions (1) (Table 3) and (1A) are very endothermic so that they can b e totally masked
by the presence of minute amounts of impurities, by photo-initiation and secondary initiation
involving the products.

Table 3
Endothermicity of R H + Q 2 R + HQ 2

1
AH^s/kJmol"

C H
6 6 252
CH 4 235
CH CH CH
3 2 3 199
(CH ) CH 3 3 188
CH CH = CH
3 2 165
HCHO 170
CH = CHCH CH = CH
2 2 2 149

6.1. Difficulties in determining kj and k 1 A

In complex radical systems such as combustion, experience has shown that there is an
inverse relationship between the speed of the elementary reaction and the difficulty of
deterniining its rate constant. This is certainly true for k j , and several factors apart from
surface effects and endothermicity contribute to the difficulty.
(i) The consumption o f R H and the formation of products occur predominantly through
propagation processes, unless the kinetic chain length is very small.
(ii) Reactions (1) and (1A) may be completely dominated by secondary initiation extremely
early in reaction. A s h a s been shown earlier [30], for a mixture containing [ C H ] = [ 0 ] = 4 2

100 Torr over the range 750-1250 K, the rates of secondary (Rg) and primary initiation (Rp)
4 2
will become equal when 1 0 " % of C H has been converted into H 0 and 1 0 " % into H C H O .
4 2 2

HCHO + 0 2 -> HCO + H0 2 (If)

H 0
2 2 + Μ 20H + Μ (13)
 13

Although less spectacular with other alkanes, even with a tertiary C-H bond, Rp and
_ 2
become equal when 1 0 % o f i-butane is converted into H 0 . Direct determination of kj 2 2

under these conditions is not a practical proposition. A s seen later, the m o r e labile C-H bonds
in C H and aldehydes offer more promise,
3 6

(hi) Radical branching through reactions such as

Η + 0 2 -> OH + Ο

and QOOH0 2 -> t w o radicals

must be taken into account. In practice, this is extremely difficult.

6.2. Experimental determination of kj

6.2.1. Aldehyde + 0 2

Although values for k ^ have been available for many years [50], attempts t o determine kj
have proved intractable, and (1) is one of the few remaining reactions of importance in
3
combustion for which there are virtually no kinetic data. Early values [51,52] of 3.75 χ 1 0 '
3 3 1 1
and 1.0 χ 10" d m mol" s" obtained at 400 Κ for aldehyde + 0 w e r e clearly t o o high by 2

6
factors of at least 1 0 , due to surface effects and secondary initiation.
Baldwin, Walker and Langford [48] modelled the oxidation of propionaldehyde at 713 Κ
3 1 1
and obtained k = 0.076 d m mol" s" . The weakness of the aldehydic bond promotes a
l p

sufficiently high value of k that a measurable initial rate of oxidation is observed (Fig.2).
l p

Due to dissociation of the H 0 formed in the reaction, the ΔΡ-time curves are autocatalytic, 2 2

and sensitivity analysis shows that the maximum rate is determined largely by k and 1 3

1 / 2 1 / 2
k /k
2 p and the initial rate by k and k / k
1 4 . A s k is known, the maximum rate is used l p 2 p 1 4 1 3

1 / 2
to obtain k / k i and hence k can be detemiined accurately from the initial part of the ΔΡ-
2 p 4 l p

time curve. Fig. 2. shows the sensitivity of the curves t o the value of k . l p

C H CHO
2 5
+ 0 2 -> C H CO
2 5
+ H0 2 (lp)

H0 2
+ C H CHO
2 5 -> H 0 2 2
+ C H CO 2 5 (2p)

H0 2
+ H0 2 -> H 0 2 2
+ 0 2
(14)

1 0 3 1 1
Assuming a value of A = 1.0 χ 1 0 d m mol" s" (shown later t o be realistic), E = l p l p

1 1
153 kJ mol" , which allowing for experimental error is close t o Δ Η = 160±4 kJ mol" , as 1 ρ

would be expected since E _ « 0. Although k will not be seriously in error, the value may l p l p

include a contribution from secondary initiation due to the following sequence.

AH
C H +0
2 5 2 -> C H 0 2 5 2 > C H OOH 2 5 C H 0 + OH
2 5
14

time/s

Figure 2 Computed ΔΡ-time plots at 713K.

C2H5CHO = 4,θ2 = 30, N 2 = 26 Torr
k2p = 0.25 (Torr s - l ) ^ , x, expt points.
1

Curve A, kip = 1.0 χ 10"4 Torrl s-1, Curve B,

kip = 2.0 χ 10-6 Torrl s-1; Curve C,
Jtlp = 1.0x10-8 Torrl s-1.

Baldwin, Matchan and Walker [53] used the oxidation of C H C H O as a source of C H

3 3

radicals. A surprising feature of the results w a s the high yield of C H ( 6 0 % ) and C U^ ( 6 % )

4 2

in the initial products for a mixture containing 2, 30 and 28 Torr o f CH3CHO, 0 and N at
2 2

813 K. C H is formed uniquely in reaction (21), which from an analysis of the results
2 6

contributes at least 9 0 % of the radical termination. Based on a fundamental principle of chain

reactions, namely that on reaching the steady state of radical concentration, the rates of
generation and removal of radicals may be equated. This is expressed in equation (iv) where
Rp, Rg, Rfc and Rf are the rates of primary initiation, secondary initiation, radical branching and
radical teroaination, respectively

Rp + Rs + Rfc - Rt (iv)
 15

Their results show that [ H 0 ] is low so that Rg = 0 if initial rates are used (computer 2 2

interpretation shows a maximum error o f 5%). If at this stage Rj, = 0 is assumed, then with
total termination due t o reaction (21), equation (iv) may be expressed as (v).

k [CH CHO][0 ]
l a 3 2 = Re (v)

where Kq is the initial rate of formation o f ethane.

CH 3 + CH 3 + Μ -> C H
2 6 + Μ (21)

Although not reported by the authors, their results may be used t o obtain a value for k . l a

Table 4 shows the values of Rg for a range of rnixture composition at 813 K, and use of
equation (v) gives k for each mixture. The values agree t o within an order o f magnitude and
l a

tend to rise at high pressures of C H C H O ( A H ) and 0 , probably due t o secondary initiation 3 2

in the sequence

AH
CH + 0 3 2 = ^ CH 0 3 2 > CH OOH 3 -> C H 0 + OH
3

(6m) (22)

At 813 K, C H O O H will uniquely undergo homolysis, so that R*, = k

3 [ C H 0 ] [AH] 2 2 3 2

and with Rq = k 2
[ C H ] , then equation (v) b e c o m e s (vi) which can be rearranged t o give
2 J 3

(vii).

k [CH CHO][0 ] +
l a 3 2 R e
1 / 2
[CH CHO][0 ] = R
3 2 e (vi)
k
21

k + ^ 6 φ 2 R Κ = Re ( v i i )

l a 2
k/ [CH CHO][0 ]
3 2

CH 0 3 2 + CH CHO3 -> CH OOH 3 + CH CO3 (22)

The values of R / [ C H C H O [ [ 0 ] give a g o o d straight line w h e n plotted against R ^ in

e 3 2
2

3 1 1
Fig.3 and the intercept gives k = 2.4 d m mol" s" at 813K. l a

The gradient is of considerable interest. A sensible value w o u l d validate the u s e of (vii)

and permit for the first time the determination o f a reliable rate constant for Η abstraction by
C H 0 radicals. A t 60 T o r r (mostly 0 + N ) , reaction ( 2 1 ) is in its fall-off region at 813 Κ
3 2 2 2
16

Table 4
Values of k l a from C H C H O oxidation at 813 Κ
3

Mixture composition/Torr Rg/ (^la)obs^

1 3 1 1
Torr s" d m m o l ' s"

CH3CHO 0 2 N 2

0.5 30 29.5 0.00121 4.05

2.0 30 28 0.0085 7.10
8.0 30 22 0.060 12.7
2.0 3.0 55 0.00050 4.2
0.5 7.5 52 0.00023 3.1

5 10
104R /(moldm-3 -l)
e s
1/2

Figure 3 Plot to obtain ku and k22 (see text).

 17

1 0 3 1 _ 1
and k j = 1.40 χ 1 0 d m mol" s from parameters given by Baulch et al [3]. F r o m the
2

same source, K = 6.5 χ 1 0 d m m o l , and as the gradient = K ^ k 2 / k 2 i

6 m ^ 3 3 - 1
m 2
1/2 = x 4

dm m o l " s , then k 2 1.85 χ 1 0 d m mol" s . N o other experimental determination

9 / 2 3/2 _1/2
2
= 5 3 1 _1

of k 2 has been made. The nearest analogue is k 3 = 1.08 χ 1 0 d m mol" s" obtained [48]
2 2
6 3 1 1

from k / k = 43.4 ( d m mol" s" )^ at 713 Κ with k

2 3 1 4
1 / 2 3
= 1.85 χ 1 0 exp (-775/T) d m 1 1 2
1 4
9 3

1 1
mol" s" , from studies of propionaldehyde oxidation [3]. Taking A = A (per C-H bond) 2 3 2 4

9
= 2.2 χ 1 0 d m mol" s" [3], then E 3
= 45 kJ mol" . A s E = E 2 is very likely, then from
1 1
2 3
1
2 3 2

5 8 3 1 1
k2 2= 1 09 χ 1 0 (above), A = 1.5 χ 1 0 d m mol" s" . A value A / A « 15 is wholly
2 2 2 3 2 2

acceptable and the value of k 2 is validated together with the value of k above. 2 J a

H0 2 + C H CHO
2 5 -> H 0 2 2 + C H CO
2 5 (23)

H0 2 + C H 2 6 H 0 2 2 + C H 2 5 (24)

A s Δ Η = A H = E = E = 155 ± 5 kJ mol" , then from k = 2.4 d m mol" s" at 813K,

1 ρ l a l a l p
1
l a
3 1 1

3 1 1
the value at 713 Κ is 0.096 compared with k = 0.076 d m m o l ' s" . The agreement is l p

strikingly good.
The most successful and reliable study of (1) for aldehydes has been m a d e with H C H O . In
the temperature range 670-815 K, the rate of oxidation of H C H O in Pyrex vessels freshly
coated with KC1 is e x t r e m e l y reproducible [43]. Surface termination of H 0 and H 0 2 2 2

occurs efficiently in these vessels so that destruction is diffusion-controlled and independent of

surface efficiency [54,55]. Using the mixture 2, 30 and 26 Torr o f H C H O , 0 and N , 2 2

respectively, the chain length is only about 2 and can be reduced effectively t o zero when
pressures down to 0.1 Torr of H C H O are used.
In the initial stages of reaction, the basic mechanism is extremely simple.

HCHO + 0 2
-> HCO + H0 2 (if)

HCO + o 2 H0 2
+ CO (25)

H0 2
+ HCHO -> H 0
2 2
+ HCO (26)

H0 2
-> surface (27)

H 0
2 2
surface (28)

The rate of formation of C O (measured as an initial rate Rco) * g i v s en

by equation (vui).

R-CO = 2k, f [HCHO] [ 0 ] + 2 k 2 2 6 [HCHO] 2

[0 ] 2 (viii)
k
27
18

1 0 3 1 1
Reaction (25) is extremely fast ( k 5 = 1 0 d m mol" s" ) [3], and n o radical branching 2

occurs through peroxy species generated from H C O radicals. Decomposition of H C O t o

give Η atoms is also negligibly slow. The stationary concentration o f H 0 is extremely 2 2

small, so that reactions of O H radicals are negligible, particularly as initial rates are used.
Rearrangement of (viii) gives (ix).

2 k 2 o k l f
R c o /[HCHO] [0 ] = 2k 2 l f + [HCHO] (ix)
k 2 7

Plots of R C Q / I H C H O ] [ 0 ] against [ H C H O ] give excellent straight lines and the 2

intercepts are extremely well defined. Variation in the vessel diameter by a factor o f 4 gives
indistinguishable intercepts, comfirming the h o m o g e n e o u s nature of (If). Between 670 and
1 0 3 1 1 1
815 K, A = 2.04 χ 1 0
l f d m mol" s" and E = 163±6 kJ mol" [43]. It must be l f

emphasised that as experiments are carried at effectively zero chain length, kjf is determined
in a direct experiment. A s the thermochemistry and the kinetics are consistent, the values
represent the first totally reliable kinetic parameters for reaction (1).

6.2.2. Alkene + 0 2

Studies [32] of C H oxidation under carefully selected and controlled conditions have
3 6

shown that all the difficulties traditionally associated with the determination of k can be 1

reduced t o a minor role, essentially because the allyl radical p r o d u c e d in reaction (l^) is
stabilised by electron delocalisation and is hence very unreactive t o w a r d s 0 . The presence of 2

the weak allylic bond in C H also means that reaction (lpg) is relatively fast. Table 2 shows
3 6

that the C H C H C H + 0 equibrium is placed well t o t h e left above 6 0 0 K. Allylperoxy

2 2 2

1
reactions are consequently unimportant and moreover Δ Η is about + 4 0 kJ m o l ' , with k = 2 9 2 9

2.5 χ 1 0 d m mol" s" [41,56] at 750 K, a factor of 1 0 slower than the analogous reaction
2 3 1 1 6

of alkyl radicals eg (5e).

C H
2 5 + 0 2 C H
2 4 + H0 2 (5e)

[ H 0 ] and [ H 0 ] are both relatively low, and although an O H chain is important (about
2 2 2

one-third o f the propene is removed via O H addition), the rates o f initiation and termination
are not affected. The key experimental observation is that for, say, a mixture composition of
12, 30, and 18 Torr of C H , 0 and N , respectively, at 753 K, h e x a - l , 5 - d i e n e ( H D E ) is a
3 6 2 2

major initial product (Fig.4). In fact, up t o 5 0 % o f the allyl radicals formed from propene
recombine to give H D E . The basic chemistry is shown by reactions (2p), (17)-(20) given in
Section 3. Secondary initiation is negligible providing initial rates of product formation are
considered.

C H
3 6 + 0 2 -> C H
3 5 + H0 2 (lpe)

C H
3 5 + 0 2 -> CH = C = CH
2 2 + H0 2 (29)
 19

Figure 4 Pressure vs. time profiles for products from

C3H6 + O2 at743K.
C3H6 = 12, O2 = 30, N2 = 18 Torr.o, CO;x,
H C H O ; D , C2H4;A, C H C H O ; V ,
3

CH2 = CHCHO;#, hexadiene (x 5);0,

propene oxide (x 5).

Stothard and Walker [32] used equation (iv) t o obtain values for E and A . With
l p e l p e

R s = 0, assuming = 0 and that termination occurs solely through the recombination of allyl
20

radicals then k can be obtained directly from equation (x), where RHDE
l p e
l s t n e
initial rate of
formation of H D E .

^ipe [ C H ] [ 0 ] = RHDE
3 6 2 W

Table 5 shows that the values of k obtained vary by only a factor of 2 over a wide
l p e

range of mixture composition at 753 Κ Taking a mean value and combined with similar
9 2 3 1 1
results at other temperatures, preliminary values of ( A ) = 10 d m mol" s" and l p e o b s

= 1
(Elpe)obs 162 kJ mol" were reported. It is worth examining these values closely because
the study provides an almost 'ideal' example of h o w t o obtain g o o d kinetic data in complex
systems such as those found in combustion. Pertinent features may be emphasised.
(i) The system as set up is extremely simple (at least in a relative sense).
(ii) Direct measurement is made of the key experimental kinetic factor (namely here the major
chain termination process).
(in) The system and conditions are selected to exclude major complicating factors (in this case
radical branching and secondary initiation).
(iv)The relevant part of the chemistry can b e isolated as an independent measurement (here
the O H chain, although important in product formation, can b e ignored),
(v) The mechanism is tested over a wide range of mixture composition. Very frequently in
computer simulations, which apparently produce acceptable kinetic data on elementary
reactions, this is not the case.

Table 5
Determination o f k l p e from C^H^ + 0 2 reaction at 753 Κ

Mixture composition/Torr 10 ^HDE/

5
!0 Rp</
5
l° (^lpe)obs/
2

C H
3 6 O2 _N2 Torr s" 1
Torr s" 1 3
d m mol" s" 1 1

2 30 28 1.02 4.2 0.80

4 5 51 0.60 1.14 1.40
4 10 46 1.20 1.65 1.41
4 30 26 4.5 2.7 1.76
12 5 43 1.41 1.14 1.10
12 10 38 3.0 2.15 1.17
12 30 18 9.1 1.08 1.19
16 30 14 10.9 1.59 1.08

With regard t o the observed Arrhenius parameters for ( l p e ) , the only comparable kinetic
1 0 3 1 _ 1 1
data are the values of A = 2.04 χ 1 0 d m mol" s and E = 163±6 k J m o l " , given
l f l f
 21

1
earlier. Within experimental error ( E ) b = E j f = Δ Η = AHjf = 165±5 kJ mol" , as
l p e 0 s 1 ρ β

expected since t h e activation energies o f t h e reverse reactions will b e effectively zero.

A /A
l f « 10 is entirely consistent with t h e extra loss o f entropy o f activation in reaction
l p e

( l p e ) due t o electron delocalisation in t h e emerging allyl radical. T h e high degree of

consistency in all t h e data above provides strong support for the approach used. Stothard and
Walker [32] refined t h e treatment by allowing for extra termination d u e t o reactions (13) and
(30), and radical branching due t o (31).

H0 2 + H0 2 -> H 0
2 2 + 0 2 (13)

CH CHCH 2 2 + H0 2 -> + 0 2 (30)

CH CHCH 2 2 + 0 2 t w o radicals + products (31)

Again a simple clean approach w a s used. B y measurement of the initial rate o f formation
of propene oxide which is formed (here) uniquely in reaction (32), k being k n o w n very 3 2

H0 2 + C H 3 6 -> C H 0
3 6 + OH (32)

accurately [57], [ H 0 ] can b e calculated precisely. Similarly, [ C H C H C H ] can b e calculated

2 2 2

s
accurately from t h e initial rate o f formation of H D E (RHDE)- ^ * important t o stress that the
9 2 9 ± 0 4 1 3
correction amounts t o only about 2 0 % , leading t o t h e final values, A = 10 · dm l p e

1 1 -1
m o l ' s" and E ^ = 163.5±6 k J m o l .
Logically, g o o d kinetic data should be forthcoming from a similar type o f study of the
oxidation o f isobutene. T h e methylallyl ( M A ) radical produced by Η abstraction from one of
the t w o C H g r o u p s also h a s considerable difficulty in reacting, apart from t h e minor
3

homolysis t o give allene and C H . 3

In confirmation, 2,5-dimethylhexa-l,5-diene is found in yields o f u p t o 2 0 % from t h e

mixture containing 4, 30, 26 Torr o f isobutene, 0 and N , respectively [44], indicating that 2 2

M A radicals are removed mainly through radical-radical processes. Given a similar treatment
as shown above for propene, with corrections a little larger because o f somewhat higher H 0 2

96 8 ± 0 4 4 3 1 - 1
concentrations, t h e basic u s e o f equation (xi) gives A ^ = 1 0 d m m o l ' s and E ^
= 161.2±6.4 kJ m o l over t h e range 673-773 K , where RDMHDE *- 1
initial rate o f s t n e

formation of 2,5-dimethylhexa-1,5-diene.

k i i [ i - C H ] [ 0 ] = RDMHDE
b 4 8 2 (M)

CH C(CH )CH 2 3 2 -> CH =C=CH

2 2 + CH 3 (33)

(CH ) C=CH + 0 3 2 2 2 ^ CH C(CH )CH

2 3 2 + H0 2 (lib)
22

A s shown later on in Table 7, the agreement of the Arrhenius parameters with those for
H C H O + 0 and C H + 0 is remarkable, given the path degeneracy difference of 2 in
2 3 6 2

reaction ( l p e ) and ( l i b ) .

6.2.3. Difficult alkene systems

It is interesting n o w t o consider the possible determination of kj for butene-2 and butene-
1. T w o experimental features emphasise the large differences in the chemistry involved. Fig. 5
shows pressure change - time plots for the 4 alkenes under discussion. Clearly (and direct
product analysis confirms the effect), the initial rates of oxidation of butene-1 and butene-2
are considerably faster than those for i-butene and propene (50-100 times). Secondly, the
major ( 5 0 % ) initial product with both butene-1 and butene-2 is buta-l,3-diene and n o
evidence is found for any recombination product (and therefore for radical-radical processes)
for butenyl radicals. The basic mechanism (focussing on butenyl radicals) for both alkenes
shows the importance of H 0 / H 0 chemistry as reaction (5Ab) is the dominant propagation
2 2 2

process.

C H 4 8 + 0 2 -> C H 4 7 + H0 2 (lb)

C H 4 7
+ 0 2 -> CH =CH-CH=CH
2 2 + H0 2 (5Ab)

H0 2
+ C H 4 8 -> C H 4 7 + H 0 2 2 (34)

H0 2
+ H0 2 -» H 0 2 2 + 0 2 (13)

With a fuller mechanism to account for the products arising from O H and H 0 addition t o 2

the butenes, the product distribution can be modelled very successfully over a wide range of
mixture composition [58]. However, in the absence of a radical branching process, a value of
k
l b approximately 100 times higher than k is required at about 750 K. Such a factor is
l p e

totally unrealistic, and indeed experiments (discussed later) involving the addition o f butene-1
and butene-2 to C H + 0 mixtures at 713 K, show that k / k
3 6 2 « 5. Radical branching is l t l p e

undoubtedly of dominating importance. A possible outline sequence involves the addition of

H 0 radicals.
2

HO2 + C H C H - CHCH3
3 CH CH-CHCH
3 3 CH CH-CHCH
3 3

I I \

I
OOH OOH 0 2

branching
 23

This possibility is currently being tested [58]. A similar sequence would be of little
importance in the C H + 0 3 6 mixtures used because (in the early stages) the H 0
2 2

concentration is 50-100 lower, and H 0 adds more slowly t o C H [57]. The butene-1 and
2 3 6

butene-2 cases provide a clear indication o f the difficulty of detennining k j . In the absence of
the value for k , the 'high' value for k would b e acceptable. There is very little doubt that
l p e l b

a high value for k j can be obtained without any difficulty whatsoever!

200 U00 600

Time/s

Figure 5 Pressure change - time plots for the oxidation of

trans-butene-2, butene-1, i - butene and propene at
753K Alkene = 4, O2 = 30, N2 = 26 Torr.

6.2.4. Alkenyl oxidation chemistry

It is pertinent here t o comment on four types of alkenyl radical in respect of reaction with
0 .
2

(i) The electron-delocalised radical species C H C H C H and C H C ( C H ) C H which react

2 2 2 3 2

extremely slowly with 0 as shown (for allyl) by the parameters in Table 6. In consequence
2

the radicals are removed mainly through radical-radical processes.

(h) Electron-delocalised species which react relatively quickly with 0 t o give a diene. At
2

6
7 5 0 K , k « l x l 0 , which is 50-100 times lower than for the analogous reaction of alkyl
4
radicals [59] but about 1 0 higher than for allyl and methylallyl.
24

CH CHCHCH + 0
3 2 2 C H = CH-CH = C H + H 0
2 2 2

+ H0 2

CH3CHCH = C H C H + 0 3 2 C H = CH-CH=CH-CH + H 0
2 3 2

(iii) Alkenyl radicals where the free valency is not at the allyl position, for example.

CH = CHCH CH CHCH
2 2 2 3 + 0 2 -> CH = CHCH CH CH = CH + H 0
2 2 2 2 2

8 3 1 1
Here the reaction should proceed with k « 1 0 d m mol" s" , as for an alkyl radical [45]
(iv) Alkenyl radicals where the free valency occurs at a carbon a t o m in the double bond, for
example

C H = CH2 + 0 2 -> HCHO + HCO

CH C = CH
3 2 + 0 2 -> CH CO 3 + HCHO

and the chemistry is different in character.

Table 6
Arrhenius parameters for C H C H C H + 0 2 2 2 pathways [42]

3 1 1 1
pathway log ( A / d m mol" s ' ) E/kJ m o l '

CH CHCH + 0
2 2 2 - > 2 radicals + products 8.2 ± 0.6 72.5 ± 8

CH CHCH + 0
2 2 2 -> 2HCHO + CO + OH 9.7 ± 0.4 78.5 ± 4 . 5
CH CHCH + 0
2 2 2 -> C H = C = C H + H 0 2 2 2 9.0* 95*

2 3 1 1 9 0 3
* Values based on k = 2.5 χ 1 0 d m mol" s" at 753K and assumed A = 1 0 dm
1 1
m o l ' s" .

6.2.5. Alkane + 0 systems 2

Even for a tertiary C-H bond in alkanes, the value of ΔΗγ is considerably higher (ca 35 kJ
1
mol" ) than for the allylic C-H in alkenes. At 750 K, the rate constant is thus lowered by
about 270 on this account, although compensation for the formation of an electron-delocalised
radical reduces the difference to a factor of 30. Nevertheless, the reduction is very significant
 25

and any secondary initiation (eg reaction (14)) and radical branching may be greatly increased
in relative importance.
Only one reliable estimate for kj (alkane) is available in the literature. Baldwin et al [40]
studied the decomposition of 2, 3-dimethylbutane in the presence of 0 (for related work, see 2

section 4.2) in which the major initiation process is ( l A d ) . Although at low [ 0 ] a constant 2

value of kj is obtained, the observed value shows an increase at high [ 0 ] , consistent with a
A d 2

contribution from reaction ( I d ) which effectively involves only abstraction at a tertiary C-H
bond.

(CH ) CHCH (CH )

3 2 3 2 -> 2(CH ) CH 3 2 (lAd)

(CH ) CHCH(CH )
3 2 3 2 + 0 2 (CH ) CHC(CH )
3 2 3 2 + H0 2 (Id)

The treatment is a little complex [40], and values of k j = 0.135±0.011 and l c

3 1 1
0.0180±0.0027 d m mol" s are obtained at 813 and 773 K, respectively, after allowing for
the small amount of attack at the primary C-H b o n d s (assuming a path degeneracy difference
1 1 0 7
of 6). With A H = l d 185 kJ mol" assumed equal t o E , then k at 773 Κ gives A = 10 l d l d l d

3 -1 _1
d m m o l s which is at most a factor of 2 t o o high.

6.2.6. Aromatic + 0 systems 2

Emdee, Brezinsky and Glassman [47] modelled the oxidation of toluene at about 1200 Κ
and, based on a computer simulation approach with a comprehensive mechanism, claim that
1
the rate is most sensitive to the value of k . From their analysis with E = 173 kJ m o l ' l t l t

1 1 3 1 1
(their value of A H j ) , a value of A j = 3.0 χ 1 0 d m mol" s ' is obtained. This value is a
t t

factor of 100 higher than expected, based on the studies of C H + 0 and i - C H + 0 , 3 6 2 4 8 2

discussed earlier, as a similar electron delocalisation effect in all three should lead to an A
9 3 1 1
factor of ca 2 χ 1 0 d m m o l ' s ' . At 773 K, k / k = 36 from the respective Arrhenius l t l p e

parameters. This figure is extremely high and later it will be shown that it d o e s not exceed ca
2 at this temperature.

C H CH
6 5 3 + 0 2 -> C H CH6 5 2 + H0 2 (It)

6.3. C o m m e n t a r y on the values of k j

The values obtained for kj are summarised in Table 7. With the exception of that for
toluene, there is a high degree of consistency. For C H C H O , C H C H O , H C H O and T M B 3 2 5

1 0 3 1 1
all data are close t o A = 1.0 χ 1 0 d m m o l ' s ' (per C-H bond) and E j = Δ Η ^ For C H
Y 3 6

and i - C H , A j is lower by an order of magnitude as expected from the extra loss of entropy
4 8

of activation in the reactions. Although possibly fortuitous, A ^ j / A ^ ~ 2 as expected from

path degeneracy differences. The data are limited, but with some confidence it is concluded
that they are reliable, particularly those determined over a temperature range. Use of these
Arrhenius parameters should give kj accurate t o a factor of 2 between 600 and 1000 K. A
database for kj will be given later.
26

Table 7
Kinetic Parameters for R H + Q - > R + H Q 2 2

3 1 1 3 1 1 1
RH k / d m mol" s" T/K l o g ( A / d m mol" s" ) E/kJ mol" Ref

C H CHO
2 5
0.076 713 10.0 153* [48]

CH3CHO 2.4 813 10.0 150* [53]

HCHO - 673-815 10.31 163±6 [43]

C H 3 6
- 673-793 9.29 163.5±5 [32]

i-C H 4 8 - 673-793 9.68 161.2+6.4 [44]

(CH ) CHCH(CH )
3 2 3 2 0.135±0.011 813 - - [40]

0.0180±0.0027 773 10.7** 185 [40]

C H CH
6 5 3
- 1200 10.5 173t [47]

1 00 3 1 1
* A (per C - H bond) assumed = 1 0 d m mol" s"
** For R H = ( C H ) C H C H ( C H ) , Δ Η = Ε assumed
3 2 3 2

t Emdee et al [47] assume Δ Η = E.

6.4. Use of the C H + 0 system to determine k j .

3 6 2

The success in determining k j as described above and the verification o f a consistent set o f
data has led [44] t o the u s e of the 'ideal' C H ^ + 0 system t o obtain further values of k j .
3 2

When small amounts o f compound R H containing C - H bonds, which m a y b e attacked by

0 with at least equal facility as the allyl C - H bonds in C H , are added t o C H + 0
2 3 6 3 6 2

mixtures, then additional measureable primary initiation occurs through reaction (1).
Providing R h a s a fast reaction, then [allyl] » [ R ] , so that t h e recombination o f allyl radicals
to give hexadiene ( H D E ) remains the major tennination process

RH + 0 2 -> R + H0 2 (1)

2CH CHCH 2 2 -> C H = C H C H C H C H = CH2

2 2 2 (17)

Taking a simple case where (1) and ( l p e ) are the only primary initiation reactions, Rg = R^
= 0 and (17) is the only termination, then the ratio k ^ / k ^ can b e obtained very simply from
equation (xii), wher RHDE & (^HDE)O M a r e
initial rates o f formation o f H D E in the m e

presence and absence o f R H . Typically experiments are carried out with mixtures containing
4 t o 12 Torr of propene in 30 Torr o f 0 with u p t o 1 Torr o f additive in t h e range 670 - 770
2

K. Fig. 6 shows some typical results for R H = 2-methylbutene-2 at 693K.

 27

+
RHDE/(RHDE)O = 1 ki [ R H ] / k l p e [C H ]
3 6 (xii)

2MB-2 C3H6 O2 N2
0.4 12 30 18 Torr

200 U00 0 200 400 600

Time/s

Figure 6 Plots of HDE and C3H6O formation at 693 K,

with and without 2MB-2.

By measuring the yields o f propene oxide as well, an allowance can be made for radical
branching through reactions (31)[32] and for the additional radical termination via ( 1 4 ) and
(30).

H0 2 + H0 2 -> H 0
2 2 + 0 2 (14)

CH CHCH 2 2 + H0 2 -> + 0 2 (30)

28

CH CHCH 2 2 + 0 2 -> t w o radicals + products (31)

H0 2 + C H 3 6 -> C H 0
3 6 + OH (32)

The allyl and H 0 concentrations were calculated from the rates of formation of H D E and
2

propene oxide, respectively, both k and k being known accurately [32]. Table 8 1 7 3 2

summarises the values obtained simply from equation (xii), together with those where the
corrections have been made. Typically the corrections do not exceed about 2 0 % . Ingham,
Stothard and Walker [44] have shown that the data form a consistent pattern with values of
E = AH .
X X

Three points are emphasised here

(i) Where the values of k j / k are less than 5, the increase in H D E is quite small and the
l p e

accuracy is limited (ca ± 2, i.e. 4 0 % )

(ii) Within the confines indicated above the value of k / k = 5.5 ± 2 at 693 Κ obtained l i t l p e

from the addition study is in very g o o d agreement with the value of 3.7 from the separate
studies of the oxidation of propene and i-butene discussed earlier (Table 7). Validation of the
method is thus obtained.
(iii) It w a s only possible t o obtain a maximum value for k j t

C H CH
6 5 3 + 0 2 -> C H CH6 5 2 + H0 2 (It)

When 1 Torr of toluene w a s added (the limit t o preserve the mechanism) t o the mixture
specified earlier at 713 Κ an increase in R H D E perceptible leading to k / k < 2, w a s n o t
l t l p e

whereas the use of the Emdee et al [47] parameters would give about 40. However, due to
the weaker C-H bonds involved, reasonably accurate values w e r e obtained for ethylbenzene
and isopropylbenzene as shown in Table 8. Within experimental error, the weakest C-H bond
dissociation energies in toluene, ethylbenzene and i-propylbenzene are 370, 355 and 345 kJ
-1
m o l , and with Ε = Δ Η and allowing for path degeneracy, the relative values of k : k : l t i e b

k are calculated as 1 : 8 : 22 at 750 K, assuming equal A factors per C-H bond. T h e values
l p b

of k and k ^ are reasonably consistent with this prediction and imply k < k
l e b in marked l t l p e

contrast to a calculated value of k / k « 40 obtained from Arrhenius parameters suggested

l t l p e

by Emdee et al [47]. It is pertinent to point out also that the radicals C 6 H C H C H and 5 3

C H C ( C H ) do have facile reactions with 0

6 5 3 2 so that the condition [allyl] » [ R ] is 2

preserved in the addition studies. In confirmation no evidence w a s found for radical

recombination products such as C H C H ( C H ) C H C H = C H or C H C ( C H ) C H C H = 6 5 3 2 2 6 5 3 2 2

CH . 2

C H CH CH
6 5 2 3 + 0 2 -> C H CHCH 6 5 3 + H0 2 (leb)

C H CHCH
6 5 3 + 0 2 -> C H CH = CH 6 5 2 + H0 2

C H CH(CH )
6 5 3 2 + 0 2 -> C H C(CH ) 6 5 3 2 + H0 2 (lpb)
Table 8
Kinetic data for R H + 0 2 3
- > R + H 0 from C H
2 6 addition studies

3 1 1 3 1 1 1
RH k / d m m o l ' s"
1 T/K log( A / d m m o l " s" ) E/kJmol-
(corrj

C H *
3 6 9.29 ± 0 . 4 1 163.5 ± 5

i-C H * 4 8 4.8 5.5 0.0051 693 9.68 ± 0 . 4 4 161.2 ± 6 . 4

t-C H -24 8 4.0 4.9 0.0045 693 9.6 158
1- C H 4 8 11.1 12 0.011 693 9.1 146
1 _ C H 12.7 15 0.014 693 9.1 145
5 10
(CH ) C=CHCH 3 2 3 25.5 30.5 0.0283 693 9.8 151
(CH ) C=C(CH ) 3 2 3 2 117 145 0.134 693 9.9 143
= =
CH 2 CHCH CH CH 2 2 43 52.5 0.327 743 8.1 122
C H CH **
6 5 3 <2 <2 <0.004 713 9.3 163
7.1 8.3 0.0171 713 9.1 149
C H CH(CH )
6 5 3 2
16.1 18.5 0.038 713 8.8 140

* Arrhenius parameters from direct studies, see text [32,44].

** Arrhenius parameters recommended are those for C H + 0 . 3 6 2
29
30

C H C(CH )
6 5 3 2 + 0 2 C H C(CH ) = CH
6 5 3 2 + H0 2

6.5. C o m m e n t s on the value of k obtained by E m d e e et al [47]

l t

E m d e e et al modelled the separate experimental oxidations of benzene and toluene at 1000

and 1100 K, respectively. N o doubt is cast here on the actual experimental results used t o
verify the model. The data were obtained in the well-known Princeton flow reactor [60] and
included results from a fuel-lean and a fuel-rich mixture. E m d e e et al [47] d o comment that
some adjustments w e r e necessary t o the original data following recalibration o f flow-meters,
resulting in a shortening of the reaction times by up to 2 0 % . The model for toluene contained
as a subset the reactions used to model benzene oxidation, and in total 130 elementary
reactions, written as reversible steps, w e r e used. It is not p r o p o s e d t o reproduce the
mechanism here but t o comment on the weaknesses of the approach and interpretation.
(i) The test made of the model is inadequate. Only t w o mixtures w e r e modelled, the
composition of toluene being constant (almost) and the 0 pressure varied by less than a
2

factor of 2. The total pressure w a s constant at one atmosphere. Key measurements at low
extents of reaction w e r e not made. Given the number of variables in the mechanism and some
freedom to change their value, successful prediction is inevitable, even though the mechanism
may be seriously flawed.
The practice of modelling with very limited changes of mixture composition is widespread,
and to some extent arises from restrictions caused by explosion limits and cool-flames.
(ii) Classical chain reaction theory [54] makes clear that the overall rate o f oxidation is
strongly dependent on the production of n e w radicals whether through primary initiation,
secondary initiation, radical branching or some other means. It is, however, difficult in
practice t o assign that production t o a particular chemical process, as h a s been discussed
earlier. Primary initiation may well become dominated by secondary initiation or radical
branching very early in reaction, and this may be very difficult t o detect, particularly if testing
is not carried out over a very wide range of mixture composition.
Further, in any mechanism, if one initiation process dominates, then the overall rate of
oxidation must be sensitive to any change in the rate o f that process. In modelling, however,
it is extremely easy to draw an incorrect conclusion. The observation by E m d e e et al that the
rate w a s very sensitive t o k (the model predicted overall rate α k ) cannot be used t o
l t l t

confirm the importance of ( I t ) and t o determine k if the value in the model h a s been set
l t

much t o o high, unless very stringent tests are carried out usually by making major changes in
reaction conditions. What is actually being measured is the total rate o f initiation of n e w
radicals, a small part of which may be due t o reaction ( I t ) .
(hi) From the results presented by E m d e e et al, the rate o f oxidation is similar (within a factor
of 2) for both toluene and benzene despite (using their parameters) a value o f ^ / k ^ = 1.2 χ
5
1 0 at 1200 K. The 'chain length' is clearly much higher in the benzene oxidation.

C H
6 6 + 0 2 -> C H
6 5 + H0 2 (lbe)

According t o t h e mechanism, benzene is removed only by radical attack at t h e ring leading

to the formation of phenyl and phenoxy radicals
 31

OH + C
6 6 H -» H 0 2
+ C H 6 5

Η + C
6 6 H
-> H 2
+ C H 6 5

0 + C
6 6 H -> C H 0
6 5
+ Η

The phenyl radical reacts rapidly with 0 in a branching reaction t o give the 'stable' phenoxy
2

radical which is consumed in t w o competing reactions (34) which leads t o tennination and
(35) which p r o p a g a t e s the chain.

C H 6 5
+ o 2 -> C H 0
6 5
+ Ο

C H 0
6 5 + Η -> C H OH
6 5 (34)

C H 0
6 5 -> C5H5 + CO (35)

Not surprisingly both (34) and (35) are key reactions in determining the overall rate of
oxidation because the subsequent reactions of C5H5 generate the propagating radicals and
also involve radical branching reactions (eg Η + 0 - » O H + O ) and secondary initiation (eg 2

H 0 - > 2 0 H ) . In the ensuing 'long chains' the rate of generation o f n e w radicals through
2 2

these processes will far outweigh that from primary initiation, and the overall rate will be
insensitive to the value of k ^ .
According t o the mechanism proposed by E m d e e et al, Η atom abstraction occurs solely
from the side chain and toluene is removed by the following reactions, w h e r e O C H C H is a 6 4 3

methylphenoxy radical.

C H CH
6 5 3 + 02 -> C H CH
6 5 2 + H0 2 (It)
Η + C H CH
6 5 3 -> C H CH
6 5 2 + H 2

OH + C H CH
6 5 3 -> C H CH
6 5 2 + H 0
2 (36s)
Ο + C H CH
6 5 3 -> OC H CH 6 4 3 + Η

The benzyl radicals are stabilised by electron delocalisation and are removed mainly by
radical-radical processes, many of which lead t o chain termination. With their mechanism, a
short chain length is t o be expected with the overall rate very sensitive t o k . However, l t

E m d e e et al ignore Η - a t o m abstraction from the ring in toluene. Using their rate constants
and taking account o f path degeneracy then k / k = 0.61 at 1200 K, so that almost 4 0 % of
3 6 r 3 6 s

O H attack occurs at the ring. In consequence the branching reaction (37) may well be of
great importance in generating n e w centres relative t o ( I t ) and although abstraction at the ring
may be of minor importance directly in removing toluene, the subsequent reactions of the
methylphenyl radicals are probably very important in determining t h e overall rate. B y analogy,
it is the chemistry o f phenyl in benzene oxidation and o f methylphenyl in toluene oxidation
32

that are important in determining overall rate. By ignoring the ring attack in toluene, a high
value for k is inevitable. l t

OH + C H CH
6 5 3 -> C H CH
6 4 3 + H 0 2 (36r)
C H CH
6 4 3 + 02 -> OC H CH 6 4 3 + Ο (37)

(iv)Emdee et al ignore Η a t o m abstraction by H 0 radicals (38) because 'the reaction has a 2

high activation energy', although very surprisingly H 0 attack on C H is included. In fact 2 4

1 1
[63], E is only « 45kJ mol" and E - E » 60 kJ mol" . M o r e importantly, as the half-life
3 8 3 9 3 8

5
of H 0 does not exceed 10" s under their conditions, reaction (38) (and (39)) is effectively a
2 2

branching process via reaction (13).

H0 2 + C H CH
6 5 3 -> H 0
2 2 + C H CH
6 5 2 (38)
H0 2 + CH 4 -> H 0
2 2 + CH 3 (39)
H 0 2 2 + Μ -> 20H + Μ (13)

In consequence, although (38) may only play a minor role in removing toluene, the
important factor is its rate relative to the rate of the primary initiation step ( I t ) . With (38)
ignored, a higher value of k is necessary t o predict the experimentally determined rate of
l t

oxidation of toluene.
The difficulty in determining k in the w a y used by E m d e e et al may be seen by examining
l t

the basic theory underlying their approach as laid out in the classical Semenov theory [55].
Equation (xiii) may be written for a branched chain reaction with a radical concentration n,

2 2
dn/dt = θ + fh + F n - gn - yn - δ η (xiii)

where θ is the rate of initiation, fh, gn and yn are the rates of linear branching and linear
2 2
termination in the gas-phase and at the surface, respectively, and F n and δ η are the rates of
quadratic branching and mutual termination, respectively. For illustrative purposes, the
quadratic terms will be neglected so that (xiii) becomes equation (xiv), w h e r e 0 = f-g-y

dn/dt = θ + 0 n (xiv)

If 0 is positive, the kinetic features of the chain reaction lead t o an isothermal explosion with
the boundary defined by 0 = 0. If 0 is negative, a stationary centre concentration n = θ/l 0 | s

is reached, so that the stationary state rate ( w ) is given by equation (xv), w h e r e ρ is the
s

propagation constant.

w = p0/|0|
s (xv)

In connecting this abstract equation t o a real situation, the nature of ρ is not difficult t o
characterise. Uncertainties arise however in θ and I 0 | .
 33

(i) Where branching is significant, the value of I 0| may be close t o zero, so that small
changes in g, y or f may cause large proportional changes in I 0|. Indeed the well-known
sensitivity of branched chain reactions to traces of foreign materials, surface effects and to
slight changes in reaction conditions arises in this way. In the situation w h e r e a reaction is
modelled, therefore, neglect of any termination or branching reaction can have extremely
serious kinetic consequences. In practice, since termination reactions are usually well defined,
it is neglect of radical branching that causes the problem. A s discussed earlier, E m d e e et al
appear to ignore a major branching route in their toluene oxidation modelling study, leading to
the equivalent of a higher value of I 0|, so that a higher value of θ is required t o satisfy the
experimentaUy-determined rate. Parochially this is extremely difficult t o detect.
(ii) The chemical nature of the initiation rate must be examined very carefully. For propanal
oxidation at 713 K, Baldwin, Walker and Langford [48] showed that only t w o initiation
reactions need be considered.

C H CHO
2 5 + 0
2 -> C H CO
2 5 + H0 2 (lp)
H 0
2 2 + Μ -» 20H + Μ (13)

At any time, θ = 2 k [ C H C H O ] [ 0 ] + 2 k [ H 0 ] [ M ] , so that only for the initial rate

l p 2 5 2 1 3 2 2

of oxidation w h e r e [ H 0 ] = 0 does θ = 2 k [ C H C H O ] [ 0 ] . A s discussed earlier the

2 2 l p 2 5 2

oxidation is autocatalytic, and the maximum rate is absolutely independent of k for any l p

reasonable value of k [48]. Only, therefore, by modelling the rate of oxidation in the very
l p

early stages of reaction can an accurate value of k be obtained. However, in their modelling
l p

studies E m d e e et al had n o experimental data for the first 1 0 - 1 5 % o f toluene oxidation.

Beyond this stage, the relative importance of primary initiation may well b e markedly
diminished.
(iii) Finally, it is necessary to consider the comment made by E m d e e et al that an accurate
value of k can b e obtained because the calculated rate is so sensitive t o k . A s shown by
l t l t

equation (xv), changes in the value of θ reflect directly on the overall rate. H o w e v e r the
initiation rate may be a mixture of primary and secondary initiation (indeed it is almost certain
to be once the initial stages of reaction are over). Consequently, if the rate of primary
initiation (Rp) is set so high as t o dominate over legitimate secondary initiation, then variation
of Rp must have a direct effect on the rate. Errors in the E m d e e et al mechanism almost
certainly give rise to this situation.

6.6. Database for the primary initiation reaction

Without doubt it is very easy t o obtain inaccurate kinetic parameters for reaction (1), and
almost invariably the errors produce a high value for k j . T h e data presented here, however,
form a consistent set with E j = A H j within experimental error. There is strong evidence,
including experimental, t o suggest that the value of k j for toluene + 0 obtained by E m d e e et
2

all [47] from modelling studies is approximately t w o orders of magnitude t o o high between
750 and 1200 K, arising from the neglect of important branching reactions in t h e mechanism
used.
34

The data shown in Tables 7 and 8 are still relatively limited, but as (1) is a key reaction
under a wide range of oxidation conditions, there is an urgent need for a database. Data for
R H = alkane will be extremely difficult t o determine experimentally, and the following
parameters are suggested. They are based on the principle of b o n d additivity and that the
activation energy for the reverse reaction is zero.

1
A/dm 3
mol'V E/kJmol" 1

(perC-Hbond)
9
I 7.0 χ Ι Ο ΔΗ
8
II 7.0 χ 1 0 ΔΗ
7
III 7.0 χ 1 0 ΔΗ

I Localised electron in R (eg alkanes, oxygenates)

II Delocalised electron involving the loss of one rotation (eg alkenes, alkylbenzenes)
III Delocalised electron involving the loss of t w o rotations (eg C H = C H C H C H = C H )
2 2 2

The decreasing value of A between I and III is consistent with t h e extra loss o f entropy of
1 1
activation (ca 18 J K ' m o l ' per lost rotation) in the emerging electron-delocalised radicals.
The database is considered reliable up t o about 1200 Κ (factor o f 2) unless there is a
marked degree of non-Arrhenius behaviour in reaction (1). At higher temperatures, as shown
in Table 9, homolysis of R H becomes dominant.

RH -> R' + R" (1A)

6.7. Primary initiation through homolysis

Although reaction (1A) will not be reviewed in detail, it is the principal alternative primary
h o m o g e n e o u s initiation process, particularly at high temperatures. Considerable data have
been collected for (1A), particularly by Tsang [50]. In general, k can be determined much
1 A

more easily than k j , because of the absence of secondary initiation in t h e systems used.
Baldwin, Drewery and Walker, [40] have shown that the activation energy for ( 1 A ) can be
estimated with considerable accuracy in cases where no experimental data exist. Table 9 gives
Arrhenius parameters for a selection of hydrocarbons and Table 10 gives comparisons via k 1 A

and k j [ 0 ] of the relative rate of the t w o initiation reactions with [ 0 ] = 1 atmosphere at 600,
2 2

1000 and 1500 K. With linear alkanes, (1) dominates at low temperatures and, for example,
k , [ O ] / k « 1 0 for butane at 600 K. At 1000 K, reactions (1) and ( 1 A ) contribute almost
2 1 A
5

equally. A s the alkane structure becomes more branched, the strain energy in the weakest
bond increases and homolysis becomes relatively more important, despite the presence of
tertiary C-H bonds (which increases the value of kj). A s an extreme case, at 600 K,
k / k j [ 0 ] for tetramethylbutane is approximately 10, and reaches about 6000 at 1000 Κ
1 A 2

compared with a value of about 1 for C H . Even for 2,3-dimethylbutane, k ^ / k ^ O J « 30 at

2 6

1000 K. Similar effects are observed when electron-delocalised radicals are formed. With
hexa-l,5-diene, ^ « ^ [ 0 ] at 600 Κ and k ^ / k ^ O J « 700 at 1000 K. W h e r e [ 0 ] « 1
Α 2 2

atmosphere, then clearly initiation through (1A) may be even m o r e important.

 35

Table 9
Arrhenius parameters *, f for initiation processes (1) ** and ( 1 A ) J

Reaction logA 1 A E 1 A logAj

C H
2 6 -> 2 C H 3 16.7 372

CH3CH3 + 0 2 10.63 212

C H
4 1 0 ->2C H 2 5 16.5 343
CH CH CH CH + 0
3 2 2 3 2 10.45 198
(CH ) CHCH(CH ) -»2iPr
3 2 3 2 16.4 319
(CH ) CHCH(CH ) + 0
3 2 3 2 2 10.15 190
(CH ) CCH(CH ) ^
3 3 3 2 i-Pr + t-Bu 16.5 305
(CH ) CCH(CH ) + 0
3 3 3 2 2 9.85 190
( C H ) C C ( C H ) - > 2t-Bu
3 3 3 3 16.8 290
(CH ) CC(CH ) + 0
3 3 3 3 2 11.15 212
CH =CHCH CH CH=CH
2 2 2 2 2allyl 15.1 241
CH =CHCH CH CH=CH + 0
2 2 2 2 2 9.45 145
C H CH CH
6 5 2 3 C H CH + CH
6 5 2 3 15.6 306
C H CH CH
6 5 2 3 + 0 2 9.15 152
C H CH
6 5 3 -> C H C H + Η 6 5 2 15.5* 373*
C H CH
6 5 3 + 0 2 9.33 166

3 1 1 1
* High pressure values; units, d m mol" s" and kJ mol"
f Weakest Η atom abstracted, underlined
** Based on data base given in the text
% Based on data given in ref [40]
* N o t e parameters for C-H homolysis

With ethylbenzene, the A factors and the activation energies for homolysis and for
reaction with 0 are lowered by almost the same factors by electron delocalisation, so that the
2

values of k / k j [ 0 ] are essentially the same as those for C H at all temperatures. Similar
1 A 2 2 6

behaviour will be found for aliphatic alkenes such as h e x - l - e n e w h e r e E is lowered only by 1 A

electron delocalisation in the allyl radical produced. However, k ^ / k ^ O J will be much

higher for 4,4-dimethylpent-l-ene where E is lowered by both electron delocalisation and 1 A

strain energy. Toluene is an interesting case as C-H homolysis t o form t h e benzyl radical is
36

- 1
faster than C-C homolysis to give the phenyl radical, due t o the 50 kJ m o l stabilisation
energy in the benzyl radical.

CH = CHCH CH CH CH
2 2 2 2 3 -> CH CHCH + CH CH CH
2 2 2 2 3

CH C(CH ) CH CH = CH
3 3 2 2 2 (CH ) C + CH CHCH 3 3 2 2 (40)

Table 10
Relative rates* of reactions (1) and (1A) at [ 0 ] = 1 atmosphere 2

600 Κ 1000 Κ 1500 Κ

RH k
lA ki[0 ] 2
k
lA ki[0 ] 2
k
lA
k
i[0 ] 2

1 6 1 0
C
2 6 H
2.0xl0" l.lxlO" 1.8xl0" 3
2.4xl0" 3
5.5xl0 3
9.5
C H 4 1 0 3.2xl0" 1 4
1.5xl0" 9
2.9xl0" 2
l.OxlO" 2
2.8xl0 4
21
(CH ) CHCH(CH ) 3 2 3 2 4.1xl0" 1 2
8xl0" 9
0.54 2.0xl0- 2
1.9xl0 5
27
(CH ) CCH(CH ) 3 3 3 2 9.6x10-! 1
4.2xl0" 9
3.9 l.lxlO" 2
7.8xl0 5
13.5
(CH ) CC(CH ) 3 3 3 3 3.1xl0" 9
3.4xl0" 1 0
41 7.0xl0" 3
4.6xl0 6
29
6
CH =CHCH CH CH=CH
2 2 2 2 1.3xl0" 6.7xl0" 6
3.3xl0 2
0.46 5.1xl0 6
100
C6H CH CH 5 2 3 9xl0" 1 2
1.6xl0" 6
0.42 0.19 8.8xl0 4
57
C H CH **
6 5 3 l.OxlO" 7
1.6xl0" 7
1.05xl0- 5.7xl0" 4 2
3.3xl0 2
29

1
* Units s ' , reactions as specified in Table 9
** N o t e kj A values for C-H homolysis in C H 3 group

7. T H E R E A C T I O N O F A L K Y L R A D I C A L S W I T H 0 2

7.1. General introduction

It has long been established that conjugate alkenes are the major initial p r o d u c t s in the
reaction of alkyl radicals with 0 (where structurally possible) above ca 550 Κ [30, 4 5 , 6 3 . 2

64]. A s the carbon chain lengthens, increasing amounts of oxygenates, and in particular O-
heterocyclics, are observed in the initial products. At about 750 K, the % o f alkene decreases
from about 9 9 % for C H , 8 5 % ( 1 - C H ) , 6 5 % ( 1 - C H ) down t o less than 5 0 % for C and
2 5 3 7 4 9 6

C alkyls [29, 45]. Structural factors also play a part, and 2 - C H [40] and t - C H [36,65]
7 3 7 4 9

radicals both give 9 9 % of conjugate alkene because alternative reactions are very limited.
Baldwin and Walker [30] indicated in 1972 that the formation of p r o d u c t s through oxidation
of alkyl radicals is potentially extremely complex as each product can be formed in three ways,
the direct bimolecular process (6A), through R 0 decomposition (8) or via reaction (10) of 2

Q O O H radicals, which arise through Η atom transfer in R 0 radicals. 2

 37

(7) (9)
R+ 0 2 R0 2 QOOH

6 A \ | ( 8 )
^(lO)

Products

Very complex expressions may arise for the overall rate constant for product formation if
all routes contribute, particularly if (7) and (9) are both equilibrated and R 0 can form a 2

number of Q O O H radicals. For example, 2-pentylperoxy radicals can undergo 5 different Η

atom transfers, where p , s and t refer to the primary, secondary and tertiary Η atom
transferred.

M p
4 3 4 5 6 > CH CH(OOH)CH CH CH
2 2 2 3

1 > 3 t
CH CHCH CH CH
3 2 2 3 > CH C(OOH)CH CH CH
3 2 2 3

1 U 4 s
) CH CH(OOH)CHCH CH
3 2 3

U s
Ο—Ο > CH CH(OOH)CH CHCH
3 2 3

U 6 p
2 1 ) CH CH(OOH)CH CH CH
3 2 2 2

Walker et al [38,67] have provided strong evidence that formation of conjugate alkene
from R + 0 in the range 600-1000 Κ does n o t occur through Q O O H radicals, but are unable
2

to distinguish between the direct reaction (6A) and reaction (8) because under the conditions
used R and R 0 are fully equilibrated in (7). Benson [66] has argued that the rate of the
2

overall reaction (5A) varies little with alkyl structure and suggested a direct bimolecular
1
abstraction with an activation energy o f 15-25 kJ m o l ' . Baldwin, Walker and co-workers
[59] have obtained a number of values of k for C - C alkyl radicals at about 750 K. They
5 A 2 5

observed a g o o d correlation between log k at a fixed temperature and the enthalpy of 5 A

reaction, as shown in Table 11 and Fig. 7. The correlation extends quite well t o the analogous
1
reaction of pentenyl ( Δ Η = -18) and even t o the endothermic reaction ( Δ Η = + 40 kJ m o l ' ) of
allyl radicals [59, 69].

CH CHCH CH CH
2 2 2 3 + 0 2 -> CH =CHCH=CHCH
2 3 + H0 2

CH CHCHCH
3 2 + 0 2 -> CH =CH-CH=CH
2 2 + H0 2

CH CHCH
2 2 + 0 2 -> CH =OCH
2 2 + H0 2

The curve shown in Fig. 7 is consistent with minimal variation o f k at high negative values
of Δ Η and maximum variation at high positive values [59]. Fish [70] developed an extensive
theory for the production of oxygenates, and concluded that O-heterocyclic c o m p o u n d s (often
formed in 5 0 % yield with C , C alkanes) w e r e formed via Q O O H decomposition. Baldwin
7 8

and Walker [71] have examined this suggestion and obtained quantitative data for the
38

elementary reactions involved in the formation of oxiranes, oxetanes, tetrahydrofuranes and

tetrahydropyranes in the sequence

R + 0 2 R0 2 QOOH il^L O-heterocycle + O H

Table 11
Variation of rate constant for R + 0 2 - > alkene + H 0 2 with Δ Η at 753 Κ [59,69]

R* Alkene kJ k/C-Hbond ΔΗ/

3 1 1 3 1 1 1
( d m mol" s" ) ( d m m o l ' s" ) kJ mol"

CH CH3 2 C H
2 4 6.6xl0 7
2.2xl0 7
-46
CH3CHCH3 C H
3 6
1.25xl0 8
2.1xl0 7
-40
CH CHCH CH
3 2 3 C H -1
4 8 5.1xl0 7
1.7xl0 7
-38

CH3CH CH 2 2 C H
3 6
7.1xl0 7
3.6xl0 7
-54
CH3 C H C H C H 2 2 2 C H -1
4 8 1.6xl0 8
8.0xl0 7
-56
CH3CHCH CH3 2 trans-C Hg-2 4 7.8xl0 7
3.9xl0 7
-49
cis-C H -2 4 8 4.3xl0 7
2.2xl0 7
-45

(CH ) CHCH
3 2 2 i-C Hg
4 6.8xl0 7
6.8xl0 7
-62
(CH ) C CH(CH )
3 2 3 2 (CH ) C=C(CH )
3 2 3 8.4xl0 7
8.4xl0 7
-64

CH CHCHCH CH
2 2 3 CH =CHCH=CHCH
2 3 2.1xl0 6
1.05xl0 6
-18
CH CHCH
2 2 CH —C—CH
2 2 2.5xl0 2
2.5xl0 2
+40

* Η a t o m abstracted is underlined

With R and R 0 equilibrated, and considerable evidence that k

2 1 0 » k.9 [67, 7 1 ] , the rate of
formation of O-heterocycle is given by equation (xvi).

d[0-heterocycle] = _ K k [R][0 ]
7 9 2 (xvi)
dt

Following measurements of k and use of K o b s 7 (often calculated from B e n s o n ' s additivity

rules [72]), then k can be determined. 9
 39

Figure 7 Plot of log ksA (per C-H bond) for the

formation of alkenes at 753K.
Δ , tertiary; Ο , secondary; χ , primary • ,
pentenyl; # , allyl C-H.

One example, which has been examined in detail by Baldwin and Walker [71] and more
recently by Pilling [73], is the chemistry associated with the formation and removal of
neopentylperoxy radicals leading t o the formation of i-butene, acetone and 3,3-
dimethyloxetane ( D M O )
40

(a) (b) , C H
2
(CH ) CCH + 0
3 3 2 2 z ^ ^ (CH ) CCH 0
3 3 2 2 *• (CH ) Q 3 2

I CH OOH 2

/ (°) (d)

(CH ) OCH3 2 2 + CH 3 (CH ) CO + 2 H C H 0 + OH

3 2

(CH ) C-CH
3 2 2

CH O 2

(DMO)

Baldwin and Walker confirmed the following t w o relative initial rates of product formation
from studies of the addition of neopentane t o mixtures of H + 0 between 673 and 773K 2 2

d([DMO] + [ C H C O C H ] ) 3 3 = k b K ^ [Oj] ( x y i i )

d[i-C H ] 4 8 k e

and < J [ C H C O C H 3 3 3 =

d[DMO] k c

and were able t o determine kinetic data for a number of the elementary reactions involved. In
1 3
particular, w a s calculated from B e n s o n ' s additivity data [72] and k = 1.20 χ 1 0 b

1
exp(-14430/T)s- w a s obtained for the l,5p Η atom transfer involved. Pilling [73] has
repeated the measurements by a direct measurement of [OH] and confirmed the accuracy of
the kfcK /k values obtained by Baldwin and Walker. Further, his technique permits the
a e

determination of directly and his results show that it is a factor of about 10 greater than the
calculated figure, so that Baldwin and W a l k e r ' s values of k are reduced by the same factor. b

Pilling's direct determination of k is of paramount importance because of the set of self-b

consistent data obtained by Walker and co-workers [71] for a whole series of 1,4, 1,5, 1,6 and
1,7 Η atom transfers in alkylperoxy radicals, all based on calculated values of the equilibrium
constant for R + 0 R0 . Pilling's w o r k on the l,5p Η atom transfer in
2 2

neopentylperoxy radicals effectively calibrates and validates the complete set of data which
has been used extensively t o model many practical combusion problems.
Of relevance t o later discussion are the Arrhenius parameters (particularly the activation
energies) obtained for 1,4 Η atom transfers t o give oxiranes, and these are summarised in
Table 12. Further it is important t o emphasise the existence of extensive evidence that the
subsequent decomposition of Q O O H radicals is sufficiently fast that R 0 -> QOOH 2

isomerisation is effectively irreversible. [67, 7 1 , 73]

 41

7.2. T h e reaction of C H radicals with 02 5 2

Over t h e last 10 years, particular attention h a s been focussed on t h e C2H5 + 0 reaction 2

as its chemistry is regarded as simple but prototypical, although this view m a y be contested.
In 1980, Baldwin, Pickering and Walker [64] generated C2H5 radicals b y t h e addition of
C H t o mixtures of tetramethylbutane and 0 , and in confirming earlier w o r k showed that in
2 6 2

the initial stages o f reaction C H and oxirane w e r e the only products (at least 99.9%). Over
2 4

the temperature range 673 - 813 Κ they obtained Α / Α £ = 13.6 ± 2.1 and E - E =12.5 5 Α 5 5 C 5 A

1
± 1 kJmol"

C H
2 5 + 0 2 -> C H 2 4 + H0 2 (5A)

C H
2 5 + 0 2 -> C H 0
2 4 + OH (5C)

These parameters correspond t o variation in t h e initial product ratio [ C 2 H ] / [ C H 0 ] 4 2 4

from 88 at 813 Κ t o 127 at 673 K. Slagle, Feng and Gutman [74] in an elegant investigation
produced C H radicals photochemically in t h e presence o f O2 and monitored their decay in
2 5

real time mass spectrometrically. Between 300 and 9 0 0 K, t h e overall rate constant for C H 2 5

+ 0 decreased monotonically and below 500 Κ t h e rate constant increased with pressure
2

over the range 1-13 Torr. Above this temperature, C 2 H w a s the only product observed. 4

Later w o r k by Gutman et al [75] using a similar technique confirmed that t h e overall rate
constant w a s effectively independent o f temperature between 750 and 1000 K. Slagle, Feng,
and Gutman [74] argued that C 2 H formation by a direct bimolecular abstraction route w a s
4

inconsistent with a zero or slightly negative energy barrier and proposed a coupled reaction
path between 300 and 1000 Κ which also accounted for t h e observed pressure effects.

C H
2 5 + 0 2 — C H 0 *
2 5 2 -> C H 2 4 + H0 2

C2H5O2

At l o w temperatures and high pressures, C H 0 * radicals are mainly stabilised by 2 5 2

collision with Μ t o form the product (in real time) C H 02. A s t h e pressure is lowered, an 2 5

increasing proportion o f C2H5O2* radicals reform t h e reactants C2H5 + 0 and the overall 2

rate constant falls, as observed experimentally. T h e energy of C2H5O2* increases with

temperature and with it the concomitant decomposition t o give C2H . H o w e v e r Plumb and 4

Ryan [76] in their flow-tube study of C H + 0 using mass-spectrometric determination of

2 5 2

[ C H ] at 295 Κ showed that the percentage o f C 2 H formed decreased from 1 5 % t o a

2 5 4

limiting value o f 6 % as the pressure (mostly H e ) increased from 0.5 t o 10 Torr. They
concluded that C H + O2 occurs in t w o uncoupled paths, one being pressure dependent and
2 5

the other, a direct bimolecular reaction (5d), being independent o f pressure.

C H
2 5 + 0 2 + Μ = C H 0
2 5 2 + Μ
C H
2 5 + 0 2 -> C H 2 4 + H0 2
(5d)
42

However, Wallington et al [77-79] in a comprehensive series o f papers showed that the

C H yield at 298 Κ decreased with increasing pressure from 1 2 % o f the C H consumed by
2 4 2 5

8
0 at 1 Torr (mostly N ) t o 0 . 0 2 % at 600 Torr, corresponding t o a P"° dependence. They
2 2

suggested a m a x i m u m value o f 8 χ 1 0 d m m o l s for t h e direct reaction ( 6 A ) at 295 K. 5 3 _ 1 - 1

In effect C H is not formed at high pressures, which validates the view that it is produced in
2 4

the coupled set o f reactions through an excited ethylperoxy intermediate and that there is n o
evidence for the direct bimolecular reaction (6A). M c A d a m and Walker [80] produced C H 2 5

radicals by oxidising C H C H O between 593 and 753 K, and determined t h e initial product
2 5

ratio [ C H ] / [ C H ] over a range of mixture composition. With C H and C H formed

2 4 2 6 2 4 2 6

uniquely in the overall reactions ( 5 A ) and (41), k / k can b e determined from equation 5 A 4 1

(xix).

d[C H4]/d[C Ha
2 2 = k [0 ]/k [C H CHO]
5 A 2 4 1 2 5 (xix)
C H
2 + 0
5 2 -> C H 2 + 4 H0 2 (5A)
C H
2 + C H CHO
5 2 5 -> + C H CO 2 5 (41)

Use of literature values of k i , and correction for Η abstraction from t h e C H group in

4 2 5

1 7 0 5 ± 0 1 8 3 1 1
the aldehyde, gave E = -6.3±3.5 kJ mol" and A
5 A = 10 d m m o l " s" . T h e values 5 A

of k were also in g o o d agreement with those obtained by Slagle, Feng & Gutman [74]. A
5 A

later study by Gulati and Walker [81] gave a similar small negative activation energy for the
analogous reaction of i - C H radicals with 0 3 7 2

i-C H 3 7 + 0 2 -> C H 3 6 + H0 2

Gutman and co-workers ( W S S G ) [75] extended their earlier investigations on C H + 0 2 5 2

through a well designed experimental and theoretical study over the range 296-850 Κ at
pressures between 0.5 and 15 Torr, by use o f laser photolysis coupled t o a photo-ionisation
mass spectrometer for the measurement o f C H and C H product profiles in real time. 2 5 2 4

They were able t o measure both the rate constant for the overall loss o f C H and the 2 5

branching ratio F, ( [ C H ] ) / ( [ C H ] + [ C H 0 ] ) . T h e experimental data w e r e concordant

2 4 2 4 2 5 2

with those from the earlier study and with those obtained at r o o m temperature by a number of
workers [75-79]. In addition W S S G used R R K M theory t o model their experiments and the
data of independent workers, based on the mechanism proposed by Slagle, Feng & Gutman
[74].

T w o points should be emphasised.

(i) Above about 650 K, the step C H 2 5 + 0 - > C H 0 is fully equilibrated, and the overall
2 2 5 2

rate constant for consumption o f C H 2 5 is effectively independent o f pressure and temperature.

(ii) The branching ratio F is effectively unity above 650 K, but below this temperature falls as
the pressure is increased because the C 2 H 0 * radicals are increasingly stabilised by collision.
5 2
 43

Figure 8 Energy diagram for C2H5 + O2 -> C2H4 + HO2

Full line as given by Wagner et al (ref. 75).
Dashed line as amended by the author (see
text).
The potential energy surface used in the modelling is shown in Fig. 8. Following addition
of 0 to C H , the C2H5O2 radical passes through a 5-centre cyclic transition state (TS1)
2 2 5

which leads t o the ethylhydroperoxide radical C H O O H which then loses an H 0 radical,

2 4 2

and the formation of C H occurs via transition state (TS2).

2 4 T w o key quantitative
thermochemical features in the energy surface must be stressed. The enthalpy changes for the
1
transition C H + 0 - > T S 1 and for C H + 0 - > T S 2 are -16.5 and -25 kJ mol" ,
2 5 2 2 5 2

respectively at 298 K. Consequently, in the path from C H t o the final p r o d u c t s C H +

2 5 2 4

H 0 , n o barrier h a s an energy higher than the energy of the original reactants. With the
2

mechanism proposed, then at l o w temperatures and very l o w pressures, a fraction will return
to the reactants, very few will be stabilised and significant numbers will decompose t o give
C H . A s the pressure is increased, the stabilisation efficiency increases and eventually the
2 4

branching ratio F will fall t o zero. A s the temperature is raised, C H 0 * will have higher
2 5 2

energy and, for a particular pressure, the branching ratio will increase. Further the C H 0 * 2 5 2
44

radicals will increasingly return t o the reactants and eventually (above about 650 K ) an
established equilibrium C H + 0 C H 0 will b e observed. Under equihbrium
2 5 2 2 5 2

conditions, C H 0 will have sufficient energy t o decompose t o C H , so that the branching

2 5 2 2 4

factor will automatically assume a pressure-independent value of unity. This switch in

mechanism from low t o high yields of alkenes over a n a r r o w temperature range is well known
in the combustion chemistry of alkanes [ 1 , 29, 30, 45].
W S S G pointed out that the overall rate constant is particularly sensitive t o the enthalpy
change for C H 0 - > T S 1 , whereas the temperature dependence of the branching factor F
2 5 2

depends on the tightness of t h e cyclic transition state. It is clear that the rate constant is most
sensitive to A H ( C H + 0 - > T S 2 ) and the only requirement is that the potential energy of
2 5 2

TS2 does not exceed that of the reactants C H + 0 . With both barriers below Δ Η ( 0 Η Ο 2 5 2 2 5 2

- > C H + 0 ) , then when F = 1, the overall rate constant (which equals that for C H
2 5 2 2 4

formation) should be virtually independent of temperature.

Bozzelli and Dean [82] added t o the wealth of support for t h e coupled mechanism by
using a theoretical treatment which incorporated literature rate constants, transition state
theory for isomerisation, thermochemical data t o calculate reverse rate constants and
bimolecular Q R R K theory [83,84] t o calculate reaction probabilities for the initially formed
chemically activated complexes such as C H 0 * . A full range of possible reactions are 2 5 2

considered for each radical intermediate involved. Their calculated rate constants are in g o o d
agreement with experimental data [75,80], and they conclude that there is n o evidence for a
direct bimolecular C H 0 * reaction below 1000 K. 2 5 2

Bozzelli and Dean comment that "the energised hydroperoxy radical C H O O H * , if 2 4

formed, dissociates t o C H and H 0 almost completely at l o w pressures due t o the high β-

2 4 2

scission A factor (compared t o that for reverse isomerisation), whilst at high pressures large
1
fractions can be stabilised. This well is very shallow (they estimate ca 65 kJ m o l " ) , however,
and the stabilised radical again dissociates almost completely t o C H + H 0 " . This 2 4 2

statement requires clarification and appears at odds with experimental fact. In the 'real-time'
experiments of W S S G , C H formed from stabilised C H O O H radicals w o u l d not be
2 4 2 4

observed as t h e p r o c e s s would b e t o o slow. However, even in t h e experiments carried out by

Kaiser, Lorkovic and Wallington [79], where photolysis lasted up t o 2 minutes, C H yields 2 4

were reduced to zero at high pressures. At temperatures below 500 K, a barrier of 65 kJ

1
m o l ' will allow competing reactions unless the pressure is very l o w and it is likely that
removal of C H O O H by 0 addition would be dominant even at relatively l o w 0 pressures.
2 4 2 2

T w o other aspects of Bozzelli and D e a n ' s paper merit discussion here,

(i) The channels (a) and (b), with much higher energy barriers, b e c o m e competitive at about
2000 Κ (each contributing about 1 0 % t o the total rate constant)

C H 2 5 + 0 2 -> CH CHO 3 + OH (a)

C H 2 5 + 0 2 -> C H 0
2 5 + Ο (b)

At this temperature, however, the C H + 0 reaction will not b e an important step in 2 5 2

combustion mechanisms, and (a) and (b) may be ignored.

 45

(ii) Given the coupled mechanism and the proposed potential energy surface, the variation
of the predicted rate constant k £ f ° ^2^4 formation with temperature should show a
2 H 4
r

marked dependence on pressure. A s Bozzelli and Dean show, at high pressures (above 1
atmosphere) k £ is relatively low at 298 Κ ( C H 0 * stabilisation) and rises with
2 H 4 2 5 2

temperature until the branching factor F is unity, after which it falls steadily by a factor of 2-3
between 700 and 1800 K. At a lower pressure (ca 1 Torr), high yields of C H are observed 2 4

at 298K (little C H 0 * stabilisation), and the predicted value of k

2 5 2 remains virtually C 2 H 4

constant t o 700 Κ and then falls by a factor of 2-3 as the temperature is increased t o 1800 K.
The fall in k between 700 and 1800 K, which is independent of pressure, can be
C 2 H 4

associated with the thermochemistry of the coupled mechanism shown in Fig. 8.

C2H5 + O2 C H 0 2 5 2 C H400H
2

A s C H and C H 0 are considered t o be in equiUbrium, F = 1, and as the step forming

2 5 2 5 2

C H O O H is effectively irreversible, then k^ H4 ^a^b* &

2 4 activation energy for C H 2
= m t n e
2 4

1
formation will be Δ Η + E = -16.5 kJ mol" , a dominant factor in deterrnining the
3 b

temperature coefficient.

7.3. C o m m e n t s on C H + 0 2 5 2

There is a general consensus view that the coupled mechanism gives a quantitative
account of the reaction between C H + 0 over a wide pressure and temperature range.
2 5 2

Certainly below 1000 K, there are n o grounds for including the direct bimolecular route
(except Plumb and R y a n ' s limited pressure data at 298 K). However, as pointed out
repeatedly by Walker and co-workers [67, 68, 80, 8 1 , 85] there are problems, not with a
coupled mechanism, but with specific aspects of the one proposed, namely, first, the
requirement that C H formation occurs via C H O O H radicals and secondly, the
2 4 2 4

thermochemistry attributed t o the potential energy surface used by W S S G and by Bozzelli and
Dean. These concerns extend t o other alkyl + 0 systems, and it is unfortunate that they have 2

not received the detailed modelling attention given t o C H + 0 . It is appropriate t o review 2 5 2

some of the features, mostly associated with other alkyl + 0 systems, which give cause for 2

concern about the validity of the mechanism proposed by W S S G for the C H + 0 reaction. 2 5 2

7.4. Possible difficulties associated with the thermochemistry of C H + 0 2 5 2

The potential energy 'surface' favoured by Walker and co-workers is shown by the dotted
line in Fig. 8. In particular, the t w o barriers t o C H formation are considerably higher, and 2 4

critically, the transition states T S 1 and T S 2 are markedly higher than the initial energy of
C H + 0 . Clearly, such a surface would have serious implications for the validity of the
2 5 2

W S S G mechanism. Additionally a path, not considered by t h e m or Bozzelli and Dean, leads

to the formation of oxirane (dashed line).
46

7.4.1. Barrier for R 0 - > Q O O H2

It is generally accepted that O-heterocyclic c o m p o u n d s are formed in the region 600-

1000 Κ from alkyl radicals by the peroxy radical isomerisation and decomposition ( P R I D )
mechanism [67, 70, 71]

R + 0 2 4 ^ R0 2 QOOH O-heterocycle + OH

With (a) equilibrated, as observed by Gutman [75], and (b) and (c) effectively irreversible,
then the overall rate constant k for heterocycle formation is given by k
o b s = K^. From
o b s

studies (mainly) of the addition of traces of alkane to slowly reacting mixtures of H + 0 , 2 2

Walker and co-workers [71] have measured the value o f k for a full range o f 1, 4 t o 1, 7 Η
b

atom transfers in reaction (b). When originally determined via measurement o f the yields of
oxiranes, oxetanes, tetrahydrofuranes and tetrahydropyranes, respectively, the values of ^
were calculated from B e n s o n ' s additivity rules [72]. Arrhenius parameters for k are shown in b

Table 12. The values of the activation energies are determined by three factors:
(i) the activation energy for R 0 + R H (abstraction),
2

(ii) the type of Η atom transferred,

(iii)the strain in the ring transition state.
A s discussed by Baldwin et al [71], the variation of the activation energies is totally
consistent with these factors. Further, the fall in A factor (by a factor of about 10) with
increase in the size of the ring transition state by each added atom is totally consistent with the
increased loss of entropy of activation due to an extra loss of a rotation.
The consistency of the data is confirmed by Pilling's [73] w o r k on the Η a t o m transfer in
neopentylperoxy radicals. A s pointed out in Section (7.1) the value of w a s in almost
exact agreement with that obtained by Walker and co-workers, and further the value of
w a s determined experimentally. Use of this in place of the value obtained from B e n s o n ' s
1
additivity rules gives a reliable value of Ε equal to 123 ± 8 kJ mol" for the l,5p step R 0 - > 2

Q O O H for neopentylperoxy. This is significantly greater than the barrier p r o p o s e d by W S S G

for the l,4p C H 0 0 - > C H O O H step. When allowance is made for the increased strain in
2 5 2 4

1
the l,4p ring transition state, then a value of at least 150 kJ m o l ' is required for

CH CH 0
3 2 2 -> CH CH OOH
2 2

1
compared with the value of 128 kJ m o l ' used by W S S G . Clearly the formation of C H 2 4

1
should involve a strong positive activation energy (about 25 kJ m o l " ) and moreover the rate
constant would be very sensitive to the height of the barrier (Table 12). For the three cases
involving l,4p, 1,4s and l,4t Η atom transfers (AH(R+ 0 = ^ R 0 ) effectively constant)
2 2

which involve the same sized ring transition state so that the A factors should only differ
through path degeneracy, the relative rates of alkene formation should be 1:7:40 at 750 K,
compared to the very small changes shown in Table 11.
 47

Table 12
Arrhenius parameters* for R 0 2 - > Q O O H for alkyl radicals

Type of Η transfer** A/s" 1

Ε/kJmol· 1

(per C-H b o n d )

l,4p 1.4 χ 1 0 1 2
153
1 1
l,5p 1.75 χ 1 0 123
1 0
l,6p 2.2 χ 1 0 105
9
Up 2.75 χ 1 0 90

1,3s 1.15 χ 1 0 1 3
176
1,4s 1.4 χ 1 0 1 2
133
1,5s 1.75 χ 1 0 1 1
110
1,6s 2.2 χ 1 0 1 0
90
9
1,7s 2.75 χ 1 0 75

1 3
l,3t 1.15 χ 1 0 160
1 2
l,4t 1.4 χ 1 0 118
l,5t 1.75 χ 1 0 1 1
93
1 0
l,6t 2.2 χ 1 0 75
l,7t 2.75 χ 1 0 9
62

* Based on comprehensive data obtained by Walker et al [71] and Pilling's [73] absolute
data for the l,5p transfer
** See text

7.4.2. Barrier for Q O O H homolysis to form an alkene

N o experimental data exist for the decomposition Q O O H - » alkene + H 0 , but Walker 2

and co-workers [85] argue that they have determined extensive information from the reverse
addition of H 0 radicals t o alkenes. U s e of the decomposition of tetramethylbutane in the
2

presence of 0 as an H 0 radical source over the temperature range 650-790 K, together

2 2

with the measurement o f the appropriate oxirane permits extremely reliable values to be
obtained for the overall reaction. Additionally, competitive studies have been used under
conditions w h e r e the oxiranes are formed uniquely through H 0 addition t o alkenes [36, 68].
2

H0 2 + alkene —> oxirane + OH

48

The overall reaction occurs in t w o stages, first HO2 addition and then homolysis of the
peroxide bond and cyclisation, as shown for butene-2.

H0 2 + CH CH=CHCH
3 2 -> C H C H — C H C H
3 3 -> C H C H — C H C H
3 3 + OH
\ /
OOH Ο

Table 13 shows the Arrhenius parameters for a wide range o f alkenes. The activation
energies correlate extremely well with the ionisation energy of the alkenes, which provides
strong evidence that the measured parameters refer t o the H 0 addition step. Further 2

1
evidence is discussed later. A value of 72 kJ mol" for H 0 + C H - > C H O O H results in 2 2 4 2 4

the dotted line in Fig 8, and if correct rules out formation of C H via C H O O H , as 2 4 2 4

proposed by W S S G . Comparable data (Table 13) for the addition of C H 0 and i - C H 0 3 2 3 7 2

radicals t o alkenes lends further support t o the high energy barrier for H 0 addition. 2

7.5. Experiments with butane and butene-2

Scheme I summarises the possible reactions involved in the formation o f trans-butene-2
and 2,3-dimethyloxirane when n - C H and cis-butene-2 are added separately in trace
4 1 0

amounts t o slowly reacting mixtures of H + 0 at 753 Κ [67]. A s shown in Section 4 . 1 , this

2 2

system is an excellent source of H, O H and H 0 radicals. 2

Scheme I

trans-CjHt-l

H/OH/HO2 + n-C Hio

4 s-C H 4 9 cw-C Hg-2
4 + Η

o 2

toms-C Hg-24 s-C H 0 4 9 2

J (1.4s)

s-C^gOOH CW-C4H8-2 + H0 2

2,3-dimethyloxirane trans-Ctf{%-2
 49

With cis-butene-2 as additive t o a mixture containing 70, 140 and 290 Torr of 0 , H and 2 2

N , respectively, the initial product ratio [2,3-dimethyloxirane]/[trans-butene-2] = 1.85

2

compared with a value of 0.15 when butane is the additive. If trans-butene-2 is only formed
from the s - C H O O H radical, then the product ratio should be the same for both additives.
4 8

The much higher value for cis-butene-2 can only be explained if trans-butene-2 is formed
predominantly in the direct bimolecular process (rejected above) or in a reaction of s - C H 9 0 4 2

radicals which does not involve the Q O O H species s - C H O O H . The large difference in the
4 8

t w o values also confirms that the 1,4s step t o s - C H O O H must be irreversible.

4 8

Even m o r e emphatic evidence that the decomposition of Q O O H radicals does not yield
mainly alkene w a s obtained by Stothard and Walker [38] from further studies of the addition
of the separate addition of cis- and trans-butene-2 t o slowly reacting mixtures of H + 0 at 2 2

713 Κ and 753 Κ and t o tetramethylbutane + 0 2 mixtures. The relevant chemistry is shown
below for t r a n s - C H - 2 . 4 8

H0 2 + t-C H -2
4 8 ^ CH3CHCHCH3 -> cis-C H -2
4 8 + H0 2

OOH
i
CH3CH-CHCH3 + OH
\ /
Ο

If the C H 3 C H ( O O H ) C H C H radicals predominantly give alkene, then a high proportion

3

should reform t - C H - 2 , but as the trans-form is only marginally favoured [67], then the initial
4 8

product ratio [2,3 dimethyloxirane]/[cis-C H -2] should be very low. Conversely Walker and
4 8

Stothard report average values of ca 4 at 753 Κ and 5 at 713 Κ over a w i d e range o f mixture
composition. The results of m o r e recent experiments [59] at 713 Κ with c i s - C H as additive 4 8

are summarised in Table 14. Again a consistent value favouring the oxirane is obtained and is
in g o o d agreement with that from the earlier study reported above. In reality, the ratios for
both cis- and trans- additives are likely t o be considerably higher than indicated because O H
addition is important, and is well k n o w n t o be reversible under the conditions used. If
accepted, then it must be concluded that C H C H ( O O H ) C H C H radicals d e c o m p o s e almost
3 3

completely t o the oxirane.

Further, Walker & Stothard argue that as the thermochemistry o f the participating
reactions is essentially independent of the alkene system, this conclusion can be extended t o
C H O O H radicals, t o the C H + 0 system and, t o the R + 0 system in general. N o t only
2 4 2 5 2 2

does this confound the W S S G coupled mechanism, but the results also confirm that H 0 + 2

H
C n 2 n - > C H O O H is the rate-deterniining step in the formation o f oxiranes and that
n 2 n

1
therefore (Table 13) the barrier for H 0 + C H is considerably higher than 25 k J mol" as
2 2 4

proposed by W S S G .
50

Table 13

Arrhenius parameters for addition of H 0 , CH3O2 and 1-C3H7O2 t o alkenes

2

1 3 1 1
Alkene Ej/klmol" log(k/dm mol" s- ) E/kJmoH T/K

HQ [38] 2

C H 2 4 1013 9.35 71.9 670-770

C H 3 6 939 9.01 62.6 670-770
pent-l-ene 917 8.88 59.7 670-790
hex-l-ene 912 8.91 58.6 670-770
i-butene 891 8.80 53.1 670-790
cis-hex-2-ene 884 8.41 53.4 670-770
trans-hex-2-ene 884 8.41 53.4 670-770
trans-but-2-ene 881 8.61 50.0 670-790
2,3-dimethylbut-2-ene 801 8.58 35.4 650-770

CH3O2 *

2-methylbut-l-ene 880 8.6 53 373-403

2-methylbut-2-ene 837 8.2 42 373-403

2,3-dimethylbut-2-ene 801 8.1 36 373-403

i-C H 0 3 7 2 **

C H 3 6 939 8.9 68 373-408

i-C H 4 8 891 8.6 63 373-408

2-methylbut-l-ene 880 8.3 55 363-408

2-methylbut-2-ene 837 8.0 48 303-363
2,3-dimethylbut-2-ene 801 8.0 41 303-363

* Data from D. A. Osborne and D. J. Waddington, J . C h e m . S o c , Perkin Trans 2, ( 1 9 8 0 )

925.
** Data from Μ. I. Sway and D. J. Waddington, J . C h e n x S o c , Perkin Trans 2, ( 1 9 8 8 ) 139.
 51

Table 14
Initial values of [2,3-dimethyloxirane]/[cis/trans-C4H -2] from C H - 2 oxidation studies
8 4 8

Mixture/Torr

trans-C H -2
4 8 0 2 H 2 [2,3DMO]/[cis-C H -2]
4 8

Τ = 753 Κ

2.5 70 140 3.8

2.5 210 140 3.9
2.5 358 140 4.2
10 70 140 4.5
2.5 70 428 4.6
10 30 0 3.9
4 30 0 4.3

T = 713K

2.5 20 140 5.5

2.5 310 140 5.8
6.3 20 140 4.6
10 30 0 5.0
4 30 0 5.4

cis-C H -2
4 8 0 2 H 2 [2,3DMO]/[trans-C H -2]
4 8

T = 713K

2.5 70 140 3.2

10 140 140 3.4
10 30 0 3.7
4 30 0 3.7
4 3 0 4.1
52

7 . 6 . Experiments with isobutylperoxide radicals

A study of the kinetics and mechanism of Q O O H radical chemistry in the a b s e n c e of 0 2

offers considerable advantages because a number of complicating features are removed. Batt
used radical ( X ) attack on t-butylhydroperoxide in the absence of 0 t o produce 2

isobutylhydroperoxide radicals in the temperature range 370-470 K, and then measured the
yields of 2,2-dimethyloxirane and i-butene, considered t o b e formed in competing reactions of
the hydroperoxide radical. Although the results are inconclusive, t h e w o r k illustrates the type
of approach necessary t o elucidate Q O O H chemistry. Batt p r o p o s e d t h e following
mechanism.

CH 3 CH 3

X + CH3C-OOH CH C-OOH
2 + XH
CH 3 / CH 3

(CH ) C = CH + H 0
3 2 2 2 (CH ) C — C H + OH
3 2 2

V
2 6 ± 0 4
H e reported a value of l o g ( k / k ) = 1 0 exp(-2620/T) which corresponds t o
a b

[i-butene]/[2,2-dimethyloxirane] about 0.06 at 300 Κ and about 10 at 750 K. The oxirane

formation thus dominates at l o w temperature, but the evidence points t o i-butene as the major
product from isobutylhydroperoxide at 750 K, which appears t o contradict the conclusion
- 1
reached in the last section. However, B a t t ' s results suggest that E - E = 22 kJ m o l , i.e. a b

homolysis t o i-butene has the higher energy barrier, and this is not consistent with the W S S G
view of Q O O H chemistry for C H O O H . The mechanism for oxirane formation from H 0
2 4 2

addition t o C H consists o f a two-stage sequence.

2 4

k- a k b

H0 2 + C2H4 C H 4 0 0 H — C 2 H 4 O + OH
2

If, as suggested by Walker and co-workers, k » k , then step (-a) is rate detemiining and b a

1
E_ -Ea = 72 kJ mol* w h e r e E
o b s is t h e overall activation energy for oxirane formation (see
o b s

Table 13). With the W S S G view that C H O O H decomposes t o give mainly C H + H 0 , 2 4 2 4 2

then k » k , so that the addition step is equilibrated and E

a b = E_ + E - E . Assuming E - o b s a b a a

1
E = 22 kJ mol" (the value above for isobutylhydroperoxide radicals), then with E
b = 72, o b s

1 1
then E . = 94 kJ mol" , compared with a maximum value of ca 25 kJ m o l ' required by W S S G
a

(see Fig. 8).

 53

Unfortunately, the system may be m o r e complex than suggested by Batt. Abstraction of

the peroxidic Η in t-butylperoxide will p r o d u c e t-butylperoxy radicals which may contribute t o
isobutene formation.

7.7. Formation of larger O-heterocyclic rings

It is extremely unfortunate that direct experiments on C2H5 + O2 have n o t been repeated
on larger alkyl + 0 systems. Writing a similar coupled mechanism for pentene-1 from 1-
2

pentyl radicals, it is clear that stabilisation of the excited l - C 5 H 0 * radicals w o u l d occur at n 2

very much lower pressures than observed for C H 0 2 * . 2 5

I-C5H11 + 0 2 — I-C5H110 *—CH 2 2 = CHCH CH CH + H 0

2 2 3 2

Μ
f
l-C H 0
5 n 2

At low temperatures, pentene-1 would be seen only at very low pressures in time-resolved
experiments. A t high temperatures, the pentyl, pentylperoxy equihbrium w o u l d b e observed
and formation of pentene-1 through 1,4H transfer would be in direct competition with 1,5, 1,6
and 1,7 transfers leading t o the analogous larger O-ring compounds.

CH CH CH CH CH
2 2 2 2 3 -» CH (OOH)CHCH CH CH
2 2 2 3 d,4s)
1
0-0 -> CH (OOH)CH CHCH CH
2 2 2 3 (1,5s)

-> CH (OOH)CH CH CHCH

2 2 2 3 (1,6s)

-> CH (OOH)CH CH CH CH
2 2 2 2 2 (l,7p)

Table 15 shows the relative rate of the 4 isomerisations obtained from data in Table 12
and that the percentage of isomerisation corresponding t o 1,4 Η a t o m transfer is only about
4 % , whereas experimentally [87], the percentage of pentene-1 formed is at least 10 times
higher. Similarly in the oxidation of butanal at 713 K, which is an excellent source of
C H C H C H radicals, oxetane and propene are formed uniquely from a reaction between 1-
3 2 2

C H and 0 . If they are formed in l,5p and 1,4s transfers, respectively, then the product
3 7 2

ratio [ C H ] / [ o x e t a n e ] should be 1.0 compared with the observed value o f 35 [67]. Similar
3 6

observations may be made about a number of alkyl + 0 systems, and it appears that if the 2

conjugate alkene-forming process has t o pass through the very tight transition state involved
in a 1,4 Η atom transfer, then it is not possible t o account for the very high yields of alkene
observed. A possible explanation could be that all R 0 —> Q O O H isomerisations except that
2

involving a 1,4 transfer are effectively reversible, so that although the other transfers are
faster, the only effective path onwards is via the Q O O H leading t o an alkene. However, this
54

explanation would completely destroy the pattern of rate constants observed for R 0 2 ->
Q O O H (see Table 12 and ref [71]), and experimental evidence h a s shown that the
isomerisation is effectively reversible [73].

Table 15
Relative rate* of R 0 2 - > Q O O H for C H C H C H C H C H 0 radicals at 750 Κ
3 2 2 2 2 2

3 1 1
QOOH transfer k / d m mol" s" % contribution

2
CH (OOH)CHCH CH CH3
2 2 2 1,4s 7.7 χ 1 0 4

3
CH (OOH)CH CHCH CH
2 2 2 3 1,5s 3.8 χ 1 0 20.5

4
CH (OOH)CH CH CHCH
2 2 2 3 1,6s 1.2 χ 1 0 63.5

3
CH (OOH)CH CH CH CH
2 2 2 2 2 l,7p 2.2 χ 1 0 12

* Data from Table 12

7.8. Concluding remarks

The present position on the detailed mechanism for R + 0 is extremely unsatisfactory. 2

(i) Agreement exists that for C H + 0 the direct bimolecular reaction is probably 2 5 2

7 0
unimportant below 800 and accepting a maximum value of 1 0 at 7 5 0 K, values of A =
8 5 3 1 1 1
1 0 d m m o l ' s ' and Ε = 22 kJ m o l ' are suggested for this route.
(ii) A coupled mechanism written as

C H 2 5 + 0 2 ^ C H 0 * 2 5 2 ^ C H4 + H 0
2 2

C H 2 5 + 0 2

is acceptable, providing the reaction path does not pass through the C H 4 0 0 H radical. 2

Presumably, the C H 0 * would have t o undergo a concerted reaction t o form C H and

2 5 2 2 4

H0 . 2

(hi) Walker and M c A d a m [80] have suggested that 0 h a s particular characteristics as an 2

abstracting species. It is suggested that as 0 attacks an Η a t o m on the C H group in the 2 3

C H radical, the high density of unpaired electrons on the remote Ο a t o m may lead t o
2 5

bonding between this a t o m and the carbon a t o m with the localised free electron. A s the
resulting ring would be relatively unstrained, the intermediate may have a degree of stability so
that the competition between reverse decomposition and forward formation of C H could 2 4

give rise t o a negative temperature coefficient. The exact relationship between the
 55

intermediate here and that possibly undergoing a concerted reaction in (ii) is, however,
obscure.
(iv) Bozzelh and Dean [83] attempted t o remove the problem over the height of the barrier
between C2H4OOH and C2H4 + HO2 by suggesting that oxirane is not formed in the reaction
between H 0 + C H . Their suggestion that, due to a high concentration o f O H , the oxirane
2 2 4

is formed in the sequence

20H H 0
2 + Ο

Ο + C H 2 4 C H 02 4

is totally unacceptable because of the extremely high reactivity of O H radicals with alkanes
and the consequent very low concentration of O H .
The comprehensive series of studies of HO2 addition t o alkenes [68] leave absolutely no
doubt that the major product is oxirane. It is interesting t o note that Waddington and co­
workers [88, 89] have studied the addition o f C H 0 and 2-C3H7O2 radicals t o alkenes. 3 2

Oxirans are the only major product and, the activation energies for the addition are very
similar to those for HO2, as shown in Table 13.
(v) It should be noted that analogous reactions t o C H + O2 are widespread outside alkyl 2 5

systems. Table 16 gives a few simple examples where the kinetic data are available. Any
general mechanism should explain the data shown. Reactions (42)-(44) all show positive
temperature coefficients and all are sufficiently fast to be important in atmospheric chemistry.
The rate expression given for reaction (42) in Table 16 reflects a complex variation of k , 4 2

which decreases in value by about 1 5 % between 298 and 474 Κ and then increases by a factor
of 2 up t o 684 K. Grotheer et al [90] interpreted this behaviour as a change in mechanism,
with mainly an association reaction C H O H + O2 - » O2CH2OH at l o w temperatures being
2

replaced by an abstraction process at high temperatures. Reactions (43) and (44) differ in at
least t w o ways.
(i) The available data suggest a monotonic increase in k and k from r o o m temperature. 4 3 4 4

3
(ii) The values of k and k are a factor of at least 1 0 lower than k at 300 K.
4 3 4 4 4 2

CH OH
2 + 0 2 -> HCHO + H0 2 (42)
CH3O + 0 2 -> HCHO + H0 2 (43)
CH CH 03 2 + 0 2 -> CH3CHO + H0 2 (44)

A likely explanation is that the association R O + 0 - > R 0 is not favoured even at 2 3

300 K, so that the only mechanistic route for R O + O2 is through Η abstraction. If this view
is accepted then the Arrhenius parameters for reactions (43) and (44) may be taken as typical
- 1
for a direct H-abstraction route. Given that Δ Η = -46 kJ m o l for C H + 0 - > C H + 2 5 2 2 4

8 3 1 _ 1 1
H 0 , then Arrhenius parameters of A = 1.0 χ 1 0 d m mol" s and Ε = 15 kJ mol" would
2

appear appropriate for a direct bimolecular abstraction route. At 298 K, the value o f the rate
constant would be significantly below that detectable by S W W G , and Bozzelh and Dean. At
56

750 Κ, however it would contribute about 2 0 - 3 0 % t o the overall rate constant for C H 2 4

formation.

Table 16
Kinetic Data for Reactions R + 0 2 -> Ρ + H 0 * 2

R Ρ Α Ε ΔΗ k 3 0 0

1 2 1
CH OH 2 HCHO (k=1.5xl0 T )
11 9
+ 2.4x1ο exp(-2525/T) -88 5xl0
7 6
CH 0 3 HCHO 4.0xl0 8.9 -122 LlxlO
CH CH 0 3 2 CH3CHO 6.0xl0 7
6.9 -152 3.8xl0 6

3 1 1 - 1
* Units, d m mol" s" and kJ m o l ; data obtained from reference [3].

(vi) Parochially the W S S G interpretation of C H + 0 is totally consistent with the 2 5 2

experimental data. Unfortunately observations on a number of related reactions, H 0 + 2

alkene and R 0 - > Q O O H , are inconsistent with the mechanism used for C H + 0 . Clearly
2 2 5 2

theoretical treatments and time-resolved experiments on other alkyl + 0 reactions are 2

absolutely necessary for further clarification. Direct experiments on H 0 + alkene are also 2

vital.
However, the whole basis of the interpretation may have been oversimplified. Many of the
discussions of C H + 0 (and related systems) assume that only one potential energy surface
2 5 2

is involved. However, t w o recent theoretical treatments for C H 5 + 0 and CH3 + 0 2 2 2

indicate the need t o consider t w o potential energy surfaces. Fig. 9 shows a sketch of the t w o
2 2
surfaces A' and A"> suggested by Walch [91] for the production o f t w o sets of products
CH3O + Ο and H C H O + OH. H e suggests that the A " surface correlates with the addition of 2

ground state 0 , and that the Ο distant from the carbon a t o m h a s an in-plane d o u b l y -
2

occupied 2p-like orbital. In consequence, Η transfer t o form C H O O H is not favoured and 2

the only product channel available gives CH3O + O. The A ' surface corresponds to the 2

excited state of 0 and here the distant Ο a t o m has a singly occupied in plane 2-p orbital, so
2

that Η transfer is favourable and H C H O + O H may be formed. A s Fig 9 shows, formation of

H C H O + O H involves t w o potential energy surfaces with crossing from one t o the other.
Quelch, Gallo and Schaefer [92] have investigated the potential energy surface for C H + 2 5

0 and examined in detail the conformations of the ethylperoxy radical. A point very germane
2

to the present discussion is that they point out the existence of t w o states for the ethylperoxy
2 2
radical, namely a A " ground state and the A* excited state, as discussed by Walch [90] for
C H + 0 . The possibility n o w exists that the C H 5 + 0 side of one potential energy surface
3 2 2 2

is not n o w linked directly t o the addition o f H 0 t o C H , in which case a number o f the 2 2 4

apparently contradictory pieces of evidence would b e removed. Until a deeper theoretical

 57

insight into both C2H5 + 0 2 and many other similar systems is available, little progress can be
made. _

CH3 + O2 ('Ag)
" " " ^ \ / y^CH 0 + 0 3

\ J
\ 1

CH3+02 \ I ιA\
—\ \ A / \ » 1 i
\
\

\
\
\
\
\
\
\

A" surface
2 \

2
A' surface
CH"(5 + OH
2

Figure 9 Sketch of energy surfaces for CH3 + O2.

8. C Y C L I S A T I O N R E A C T I O N S O F N O N - A R O M A T I C S P E C I E S

8.1. Cyclisation at high temperatures

The formation of benzene, polyaromatic hydrocarbons ( P A H ) , aromatics in general and
soot in flames from non-aromatic sources h a s long been a subject of interest in combustion,
but no widespread acceptance of mechanism h a s emerged. Westmorland et al [93] give an
excellent summary o f the main routes suggested for t h e formation o f benzene and simple
aromatics which presumably are precursors t o soot. They range from 'polymerisation' o f
acetylene, Diels-Alder additions, recombination o f allyl radicals, radical addition t o highly
unsaturated hydrocarbons t o , more recently, the recombination of propargyl ( C H = C = C H ) 2

radicals. With advanced developments in both experimental and computer techniques,

mechanisms have been subjected t o stringent tests of the applicability in particular systems
[93]. T w o particular mechanisms merit further discussion because of the principles that
underly the proposals.
58

8.1.1. Formation of benzene from the recombination of propargyl radicals.

Stein et al [94] put forward a mechanism based on the combination of propargyl radicals
( C H = C = C H ] . They emphasise that particular attention has been given t o the addition of
2

vinyl and similar radicals to unsaturated molecular species. Although this route t o benzene is
relatively simple, at the higher temperatures of flames (> 1500 K ) w h e r e soot formation is at
its most significant, the thermochemistry involved renders many of the steps very reversible,
so dramatically reducing the rate of aromatic formation.
Stein et al consider that the propargyl radical is an 'ideal' precursor for benzene
formation.
(a) There is strong evidence for high concentrations in flames.
(b) The radical is easily formed.
(c) Arising from its structure and electron delocalisation effects, the propargyl radical is
stable and has very few possible reaction channels.
(d) Addition t o unsaturated molecules is highly reversible.
They emphasise that following the recombination of propargyl radicals, there is a
cascading series of reactions which finally lead to benzene formation, the sequence being
markedly irreversible. The overall reaction is highly exothermic with an enthalpy change of
1
about -710 kJ m o l ' . They propose a mechanism as summarised below ( t w o other
recombinations of propargyl are, of course, possible), and the potential energy diagram is
sketched in Fig 10. The cascading nature of the benzene formation is apparent. Stein et al
measured the Arrhenius parameters of the t w o steps hexa-diyne-1,5 t o 1,2
1
dimethylenecyclobutene and from the latter to benzene and furvene, and obtained k / s =
1 1 7 1 1 2 9
10 exp(-17870/T) and k/s" = 1 0 ( - 2 5 2 0 0 / T ) , respectively. With these data, Stein et al
successfully simulated the formation of benzene in acetylene flames.

2 C H = C = CH — •
2 H=C-CH -CH -C = H
2 2 »» [ C H = C = C H C H O C H ]
2 2

CH 2 1 C H 2

CH CH
I· <l
CH CH

In their own experiments in the range 520-1000 K, they showed significant differences in
product distributions when the isomerisations w e r e studied at atmospheric pressure and at
very low pressures ( V L P P ) . In particular, at l o w pressures, benzene w a s observed directly
from hexadiyne-1,5 at temperatures as low as 600 K. They attribute this t o chemical
activation whereby at the low collision frequencies of V L P P experiments, 1,2-
1
dimethylenecyclobutene is formed from hexadiyne-1,5 with about 230 kJ mol" excess energy
 59

which allows the cascade (see Fig 10) over each energy barrier right through t o benzene. At
the higher pressures, thermal stabilisation occurs and the intermediate p r o d u c t s can be
detected.

Path

Figure 10 Sketch of energy cascade from propargyl

radicals to benzene.

Westmoreland et al [93] in testing several mechanisms for benzene and soot formation
consider that propargyl recombination is feasible only for acetylene flames. They also argue
that subsequent cyclisation and necessary Η a t o m shifts (frequently a very limiting factor
because of the very high energy barriers involved) are sufficient t o rule out C H + C H and,
3 3 3 3

for that matter, C H + C H reactions as a source o f benzene.

3 3 3 4

8.1.2. Formation of benzene by chemically-activated isomerisation

A s pointed out earlier, many suggestions for benzene formation in flames involve radical
addition processes. For example, Frenklach and co-workers [95] p r o p o s e d the sequence
involving addition of 1 - C H to acetylene
4 3
60

HC = C - C H = C H + HC = C H - ^ H C ^ C - C H = C H - C H = C H - > c - C H 6 5

w a s the major route to the aromatic ring, and successfully simulated soot formation in
pyrolysis experiments of C H at 1700-2300 Κ and 7 atmospheres.
2 2

However, Westmoreland and co-workers [93] have argued that benzene is not formed by
high pressure limit radical addition processes, but by chemically-activated addition and
isomerisation reactions. In a simple form, they can be summarised below, w h e r e R is a
radical, X is an unsaturated molecule, and A and A are intermediates, A being an isomer o f
t 2 2

A,.

products
decomposition

R + X ^ = Αι* A *
2 ^ isomeric

| m J Μ
products

AI A 2

Westmoreland et al point out that the chemically-activated route is fundamentally different

from radical addition and isomerisation involving thermalised species in the following ways.
(i) Formation of assumed thermalised intermediates is impeded.
(ii) The oxidation of thermalised intermediates is averted,
(hi)New channels, not normally considered, b e c o m e available.
A s the 'normal' chemistry of A i and A does not occur, the formation of A i * leads very
2

rapidly t o A * and hence t o the final product with n o direct observation of the stable
2

intermediates A ! and A . This suggestion is not novel, as indeed the formation of C H from
2 2 4

C H + 0 occurs in a similar fashion through C H 0 , as discussed in Section 7. Only if the

2 5 2 2 5 2

pressure is sufficiently high to thermally stabilise the vibrationally-excited species will a

dramatic reduction in rate be observed.
Westmoreland et al [93] tested their chemical-activation theory on a number of flames
where benzene is formed. They concluded that only additions of vinylic I-C4H5 and I-C4H3
radicals to acetylene w e r e sufficiently fast t o account for the highest rates of formation of
benzene, by forming benzene and phenyl 'directly' (ie. without passing through stabilised
intermediates) by chemically-activated channels.

8.2. Concluding remarks on high-temperature cyclisation

It is most unlikely that any single mechanism is solely responsible for soot formation at
temperatures above 1000 K. It is sensible to assume that radical addition involving highly
unsaturated species and thermalised adducts will be important up t o about 1500 K,
particularly where electron-delocalised species are involved. A b o v e this temperature, the
recombination of propargyl radicals is a favourable route because of the 'cascading' energy
fall to the final products which are formed overall in a very exothermic sequence and thus
 61

avoid thermodynamic bottlenecks that reduce growth in normal radical-molecule additions.

Miller and Melius [96] have examined the kinetic and thermodynamic aspects of the
recombination and support this view. Cyclisation through chemical activation clearly must
occur, but is unlikely t o be important in high pressure flames even at very high temperatures.
Particularly in real fuels such as gasoline, where complex mixtures of hydrocarbons are found
it should be emphasised that highly substituted aromatics will be formed from radicals related
to propargyl and C H etc, for example as shown below.
4 3

(a) Recombination

(b) Radical addition

RCH=CH-CR'=CH + H C = C H — • RCH=CH-CR'=CH-CH=CH

Η +

Stein et al [94] pointed out that substituted hexadiynes could undergo ' c a s c a d e ' sequences
to form polyaromatic hydrocarbons.

^ C-CH -CH -C =
2 2 CH

C = C-CH 2 + CH -C =
2 CH
62

8.3. Cyclisation reactions below 1000 Κ

Although temperatures below 1000 Κ are not regarded as important for the formation of
soot, production of benzene and aromatics would be very important in t e r m s of pollution from
exhaust emissions in gasoline engines. The cyclisation mechanisms discussed above are totally
acceptable below 1000 K, but the species involved such as propargyl and highly unsaturated
radicals (eg C4H3) are not present in high concentration. Moreover, acetylene is not regarded
as a low-temperature product, presumably because (45) is the major reaction o f vinyl radicals,
rather than the high temperature processes such as (46) [3]

C H
2 3 + 0 2 -> HCHO + HCO (45)

C H
2 3 + 0 -» HC=CH + OH (46)

For the overall reaction (47), for which a very reasonable multi-stage mechanism can be
written, involving the cyclisation of the hexadienyl radical C H C H C H C H C H = C H , the 2 2 2

1
overall exothermicity is about -740 kJ mol" , which provides a strong drive for the formation
of benzene. A s pointed out earlier, the electron-delocalised allyl radical is present in high
concentration in C H oxidation, and hence in the oxidations of many longer chain alkanes.
3 6

2CH CHCH +
2 0 -> 2C H + 2H 0 2 (47) 6 6 2

Formation of cyclic compounds has been observed in the oxidation of penta-l,4-diene

( P D E ) and hexa-l,5-diene ( H D E ) [58]. In a mixture containing 4, 30 and 26 Torr of P D E ,
0 and N , respectively, at 753 K, cyclo-pentadiene is observed in an initial yield of about
2 2

2 5 % , which implies that the fully electron-delocalised pentadienyl radical rapidly cyclises.

OH + CH = CHCH CH = CH
2 2 2 -> CH CHCHCHCH
2 2 + H 0 2

2
CH CHCHCHCH
2 2 -> c-C H 5 7 ° > c-C H 5 6 + H0 2

When P D E is replaced in the mixture by H D E , then significant amounts of

cyclohexadiene, benzene and b u t a - 1 , 3-diene are observed as initial products, the combined
yield of the cyclic p r o d u c t s being about 2 0 % with t h e ratio [cyclic products]/[butadiene] « 1.
The three products are formed in the sequence as follows [33, 58].

(49) 0 2 OH/O2
CH CHCHCH CH = CH --»£- c - C H — •
2 2 2 6 9 c-HDE — • C H6
6

j(48) 7 0 % 8 0 %

C H = CHCH = C H
2 2

The t w o competing reactions (48) and (49) determine the relative yields of cyclic
compounds and butadiene, and from the initial yields o f the p r o d u c t s (allowing for the %
 63

conversions above) then k / k « 2. N o kinetic data are available for (48) and (49), but k
4 9 4 8 4 8

1 4 1 4 1
= 1 0 exp (-16600/T) s" is reasonable [39], so that k = 5.5 χ 1 0 s" at 753 Κ is estimated. 4 9

With A = 1 0 s" [72], then e

4 9
1 1
= 90 U m o l ' . e
1
may be predicted almost precisely
4 9
1
4 9

given the three contributions to its value,

1
(i) the 'normal' activation energy of a radical addition t o an alkene (ca 35 kJ m o l " ) ,
(ii) the strain energy in the transition state (effectively zero given the six-membered ring),
(hi)the loss of virtually all the delocalisation energy in a 'tight' transition state (ca 55 kJ
1
mol" ).
1
For the pentadienyl radical cyclisation, the activation energy should be about 30 kJ mol"
higher (ring strain energy), and the loss o f one less rotor in the transition state should increase
1 2 _ 1 - 1
the A factor by about 10, giving A = 1 0 s and Ε = 120 kJ m o l .
The observation that pentadienyl and hexadienyl radicals readily cyclise is of considerable
importance in the oxidation chemistry o f the larger radicals derived from alkenes. Gulati and
Walker [33] studied the addition of cyclohexane t o H + 0 mixtures at 753 Κ and reported 2 2

that the ring w a s retained in virtually all of the initial products. In particular, they did not
observe homolysis p r o d u c t s such as C H , buta-l,3-diene or butenes. Similar observations2 4

have been made when cyclopentane w a s added t o H + 0 mixtures [97]. In the case of 2 2

cyclohexane, following Η abstraction from the ring, any C H formed w o u l d arise from the 2 4

homolysis o f the C H = C H C H C H C H C H radical produced by isomerisation of the

2 2 2 2 2

c - C H radical.
6 n

(50) (51)
c-C H
6 n C H 6 n • CH - CHCH CH 2 2 2 + C2U4
(-50)

Fig. 11 shows the potential energy surface for this sequence, which is based on the k n o w n
enthalpies of the species together with
1
(i) an energy barrier o f 35 U mol" for the addition o f butenyl t o C H [98], 2 4

(ii) the same energy barrier for cyclisation of C s t l ^ t o c-C^in, based on zero ring strain
energy in the transition state (the barrier is thus effectively higher than that for radical addition
to a double bond [97]).
1 0 -1
For the cyclisation, k _ = 1 0 e x p ( - 4 2 1 0 / T ) s is very reasonable. A s the decomposition
50

of C H t o butenyl + C H (51) will have Arrhenius parameters of A

6 n 2 4 = 10 s" and e 5 1
1 3 5 1
5 1

- 1
= 125 kJ m o l , then it is clear that (50) is fully reversible (in the absence of other competing
reactions) since at 753 K, k _ » k . Walker and Handford-Styring [97] considered the
46 4 7

alternative reactions of C^Hi j radicals in the presence of one atmosphere of 0 and showed 2

that none are as fast as the cyclisation process (-50). Consequently, although isomerisation of
0c5h11 radicals t o the linear form does occur in cyclohexane oxidation [33], the reverse
reaction completely dominates all others. Only above about 1300 Κ will reaction (51), with
its higher activation energy, b e c o m e dominant.
A similar set of kinetic parameters will apply t o any radical of t h e type R C H =
C H C H C H C H C H R (the C a t o m s could b e further substituted) which will rapidly cyclise
2 2 2

even under oxidising conditions between 600 Κ and 1300 K. In the presence of 0 , a 2
64

substituted cyclohexene will be formed which will (in similar fashion t o cyclohexene oxidation
chemistry [33]) rapidly give a substituted cyclohexadiene and then a substituted benzene.
Formation of aromatics in this w a y from linear and branched alkenes may well b e very
significant in the temperature region 600-1000 K.

Figure 11 Energy path C2H4 for formation from C-C6H11

radicals.

REFERENCES

1. G.J. Minkoff and C.F.H. Tipper, Chemistry of Combustion Reactions, Butterworths,

London, 1962.
2. D.J. Hucknall, Chemistry of Hydrocarbon Combustion, Chapman and Hall, London,
1985.
3. D.L. Baulch, C.J. Cobos, R.A. Cox, C. Esser, Th.Just, J.A. Kerr, M.J. Pilling, J. Troe,
R.W. Walker, and J. Warnatz, J.Phys.Chem., Ref. Data, 21 (1992) 411 - see commentary
attached to many of the reactions covered.
 65

4. A.B. Trenwith, J . C h e m S o c , Faraday Trans. 1, 75 (1979) 614 and references therein.

5. S.N.Foner and K L . Hudson, Advances in Chemistry Series, 36 (1962), 34.
6. T.T. Paukert and H.S. Johnston, J . C h e m P h y s . , 56 (1972) 2824.
7. J.Troe, Ber. Bunsenges. Phys. C h e m , 73 (1969) 946.
8. J. Warnatz, in Combustion Chemistry (ed. W . C . Gardiner Jr.) Springer Verlag, N e w
York, 1984.
9. P.D. Lightfoot, B.Veyret and R. Lesclaux, Chem. Phys. Lett., 150 (1988) 120.
10. H. Hippler, J.Troe and J.Willner, J. Chem. Phys., 93 (1990) 1755.
11. D.L. Baulch, M . B o w e r s , D.G. Malcolm and R.T.Tuckerman, J. Phys. C h e m Ref. Data.,
15 (1986) 4 6 5 .
12. W.J. Pitz and C . K Westbrook, Combustion and Flame, 63 (1986) 113.
13. R.R. Baldwin, P.N. Jones and R.W. Walker, J. C h e m Soc., Faraday Trans., 2, 84, (1988)
199.
14. N . Cohen, Int. J. Chem. Kin., 23 (1991) 683.
15. R.W Walker, Int. J. C h e m Kin., 17 (1985) 573.
16. R.W.P. White, D.J. Smith and R. Grice, C h e m Phys. Lett., 193 (1992) 269.
17. J.E. Pollard, Rev. Sci. I n s t r u m , 63 (1992) 1771.
18. R.R. Baldwin, R.W. Walker, and R.W. Walker, J . C h e m S o c , Faraday Trans., 1, 77
(1981)2157.
19. I.R. Slagle, A. Bencsura, S.B. Xing and D. Gutman, Proc. Int. Symp. Combustion, 24
(1992) 6 5 3 .
20. I.R. Slagle, J R . Bernhardt and D. Gutman, Proc. Int. Symp. Combustion, 22 (1989) 953.
21. R.W. Walker, Proc. Int. Symp. Combustion, 22 (1989) 883.
22. Κ J. Hughes, A.R. Pereira and M.J. Pilling, Ber. Bunsenges. Phys. C h e m , 96 (1992)
1352.
23. R.G. Gilbert, K. Luther, and J. Troe, Ber. Bunsenges. Phys. C h e m , 87 (1983) 169.
24. W.J. Pitz, C.K. Westbrook, W . M . Proscia and F.L. Dryer, Proc. Int. Symp. Combustion,
20(1985) 831.
25. U. M a a s and J. Warnatz, Combustion and Flame, 74 (1988) 53.
26. A S . Tomlin, M.J. Pilling, T. Turanyi, J.H. Merkin and J. Brindley, Combustion and
Flame, 91 (1992) 107.
27. R.W. Walker, Science Progress, 74 (1990) 163.
28. C.T. B o w m a n , Fossil Fuel Combustion - A Source B o o k , (ed W. Bartok & A.F.
Sarofim), Wiley, N e w York, 1991.
29. R.T. Pollard, Comprehensive Chemical Kinetics, vol.17, (ed. C H . Bamford & C.F.H.
Tipper), Elsevier, Amsterdam, p. 358.
30. R.R. Baldwin and R.W. Walker, Proc. European Int. Symp. Combustion, 14 (1973) 2 4 1 .
31. J.F. Griffiths and S.K. Scott, Prog. Ener. Combust. Sci., 13 (1987) 161.
32. N . D . Stothard and R.W. Walker, J. C h e m Soc. Faraday Trans., 87 ( 1 9 9 1 ) 2 4 1 .
33. S.K. Gulati and R.W. Walker, J.Chem.Soc. Faraday Trans. 2, 85 (1989) 1799.
34. R.R. Baldwin, G.R. Drewery and R.W. Walker, J. Chem. Soc. Faraday Trans. 2, 82
(1986) 2 5 1 .
35. R.R. Baker, R.R. Baldwin and R.W. Walker, Combustion and Flame, 14 (1970) 3 1 .
66

36. G.M. Atri, R R Baldwin, G.A. Evans and R W . Walker, J. Chem. S o c , Faraday Trans.
1, 7 4 ( 1 9 7 8 ) 3 6 6 .
37. R R Baldwin, C.E. Dean, M R . Honeyman and R W . Walker, J. Chem. Soc. Faraday
Trans. 1, 8 2 ( 1 9 8 6 ) 89.
38. N . D . Stothard and R W . Walker, J. Chem. Soc. Faraday Trans., 86 ( 1 9 9 0 ) 2115.
39. R R Baldwin, R W . Walker and R W . Walker, J. Chem. Soc. Faraday Trans. 1, 76
(1980) 825.
40. R R Baldwin, G.R. Drewery and R.W. Walker, J. Chem. Soc. Faraday Trans. 1, 80
(1984) 2827.
4 1 . Z.H.. Lodhi and R W . Walker, J. Chem. Soc. Faraday Trans., 87 ( 1 9 9 1 ) 2 3 6 1 .
42. N . D . Stothard and R W . Walker, J. Chem. Soc. Faraday Trans., 88 (1992) 2 6 2 1 .
43. R R . Baldwin, A . R Fuller, D. Longthorn and R W . Walker, J. Chem. Soc. Faraday
Trans. 1 , 7 0 ( 1 9 7 4 ) 1257.
44. T. Ingham, R W . Walker and R.E. Woolford, Proc. Int. Symp. Combustion, 25 (1994),
to be published.
45. R W . Walker, Specialist Periodical Report, Chemical Society, Reaction Kinetics, 1,
(1975) 161.
46. J. Warnatz, Paper presented at the 12th International Symposium on G a s Kinetics,
Reading, September 1992.
47. J.L. Emdee, K. Brezinsky and I. Glassman, J. Phys. Chem., 96 (1992) 2 1 5 1 .
48. R R . Baldwin, R W . Walker and D.H. Langford, Trans. Faraday S o c , 65 (1969) 792.
49. D.J. Dixon, G. Skirrow and C.F.H. Tipper, Combustion Institute E u r o p e a n Symposium,
Academic Press, London (1973) p.94.
50. W. Tsang, Int. J. Chem. Kin., 1 (1969) 245.
51. J.B. Farmer and C A . McDowell, Trans. Faraday S o c , 4 8 ( 1 9 5 2 ) 624.
52. A. Combe, M. Niclause and M. Letort, Rev. Inst. France Petrole, 10 (1955), 786, 929.
53. R R . Baldwin, M.J. Matchan and R W . Walker, Combustion and Flame, 15 (1970) 109.
54. N. Semenov, Chemical Kinetics and Chain Reactions, Clarendon Press, Oxford, (1935)
p.41.
55. R R Baldwin and R W . Walker, Essays in Chemistry (ed J.N. Bradley, R . D . Gillard and
R F . Hudson), Academic Press, London, 3 (1972) 1.
56. Z.H. Lodhi, N . Stothard and R W . Walker, Proc. Int. Symp. Combustion, 23 (1991) 123.
57. R R Baldwin, C.E. Dean and R W . Walker, J. Chem. S o c , Faraday Trans. 2, 82 (1986)
251.
58. R W . Walker, unpublished work.
59. R R Baldwin, J.P. Bennett and R W . Walker, J. Chem. Soc. Faraday Trans. 1, 76 (1980)
2396.
60. K. Brezinsky, T.A. Litzinger and I. Glassman, Int. J. Chem. Kin., 16 ( 1 9 8 4 ) 1053.
61. W. Tsang, J. Phys. Chem., 76 (1972) 143.
62. J.A. Walker and W. Tsang, Int. J. Chem. Kin., 11 (1979) 867.
63. R R . Baldwin, D.E. Hopkins and R W . Walker, Trans. Faraday S o c , 66 (1970) 189.
64. R R Baldwin, I.A. Pickering and R W . Walker, J. Chem. Soc. Faraday Trans. 1, 76
(1980)2374.
 67

65. R.R. Baldwin, M . W . M . Hisham, A. Keen and R.W. Walker, J. Chem. Soc. Faraday
Trans. 1 , 7 8 ( 1 9 8 2 ) 1165.
66. S.W. Benson, Adv. Chem. Series, 76 (1968) 143.
67. R.R. Baker, R.R. Baldwin and R.W. Walker, J. Chem. Soc. Faraday Trans., 1, 71 (1975)
756.
68. R.W. Walker, Specialist Periodical Reports, The Chemical Society, 2 (1977) 296.
69. Z.H. Lodhi and R.W. Walker, J. Chem. Soc. Faraday Trans., 87 (1991) 6 8 1 .
70. A. Fish, Organic Peroxides, (ed. D. Swern), Wiley, N e w York, 1 (1970) 141.
71. R.R. Baldwin, M . W . M . Hisham and R.W. Walker, J. Chem. Soc. Faraday Trans. 1, 78
(1982) 1615.
72. S.W. Benson, Thermochemical Kinetics, Wiley, N e w Y o r k , 976.
73. K.J. Hughes, P.A. Halford-Maw, M.J. Pilling and T. Turanyi, Proc. Int. Symp.
Combustion, 24 (1992) 645.
74. LR. Slagle, Q. Feng and D. Gutman, J. Phys. Chem., 88 (1984) 3648.
75. A.F. Wagner, I.R. Slagle, D. Sarzynski and D. Gutman, J. Phys. C h e m , 94 (1990) 1853.
76. I.C. Plumb and K.R. Ryan, Int. J. Chem. Kin., 13 (1981) 1011.
77. T.J. Wallington, J.M. Andino, E.W. Kaiser and S.M. Japor, Int. J. Chem. Kin., 21 (1989)
1113.
78. E.W. Kaiser, L. Rimai and T.J. Wallington, J. Phys. Chem., 93 (1989) 4094.
79. E.W. Kaiser, I.M. Lorkovic and T.J. Wallington, J. Phys. Chem., 94 (1990) 3352.
80. K.G. M c A d a m and R.W. Walker, J. Chem. Soc. Faraday Trans. 2, 83 (1987) 1509.
81. S.K. Gulati and R.W. Walker, J. Chem. Soc. Faraday Trans. 2, 84 (1988) 199.
82. J.W. Bozzelh and A.M. Dean, J. Phys. C h e m , 94 (1990) 3313.
83. A.M. Dean, J. Phys. Chem., 89 (1985) 4600.
84. L.S. Kassel, J. Phys. Chem., 32 (1928) 2 2 5 , 1065.
85. R.R. Baldwin, C.E. Dean and R.W. Walker, J. Chem. S o c , Faraday Trans. 2, 82 (1986)
1445.
86. L. Batt, private communication.
87. R.R. Baldwin, J. Ρ Bennett and R.W. Walker, J. C h e m Soc. Faraday Trans. 1, 76 (1980)
825.
88. D.A. Osborne and D.J. Waddington, J. Chem. S o c , Perkin Trans. 2, (1980) 925.
89. M.I. Sway and D.J. Waddington, J. Chem. S o c , Perkin Trns. 2, (1983) 139.
90. H.H. Grotheer, G. Riekart, D. Walter and Th. Just, J. Phys. Chem., 92 (1988) 4028.
9 1 . S.P. Walch, C h e m Phys. Lett., 215 (1993) 8 1 .
92. G.E. Quelch, M . M . Gallo and H.F. Schaefer III, J. Am. Chem. S o c , 114 (1992) 8239.
93. S.E. Stein, J.A. Walker, M . M . Suryan and A. Fahr, P r o c Int. Symp. combustion, 23
(1990) 85.
94. P R . Westmoreland, A . M . Dean, J.B. H o w a r d and J.P. Longwell, J. Phys. C h e m , 93
(1989)8171.
95. M. Frenklach, D.W. Clary, W . C . Gardiner, Jr., and S.E. Stein, Proc. Int. Symp.
Combustion, 20 (1984) 887.
96. J.A. Miller and C.F. Melius, Combustion and Flame, 91 (1992) 2 1 .
68

97. S. Handford-Styring and R.W. Walker, submitted for publication in J. Chem. Soc.
Faraday Trans.
98. J.A. Kerr and S.J. M o s s (Ed), Handbook of Bimolecular and Termolecular Gas
Reactions, C R C Press, Boca Raton, Florida, Vol.II (1981).
Research in Chemical Kinetics, Volume 3
R.G. Compton and G. Hancock (editors) 69
© 1995 Elsevier Science B.V. All rights reserved.

R e a c t i o n s of N H x Species

Joseph L. Durant, Jr.

Combustion Research Facility

Sandia National Laboratory
Livermore, Ca 94551, USA

1 INTRODUCTION
The chemistry of NHx species is both interesting and important to a wide variety
of processes. In combustion, NHx species are intermediates in the conversion of
alkyl fuel bound nitrogen to NOx and N2 products. Ammonia, urea and cyanuric
acid have been used as flue gas additives to remove NOx in a variety of DeNOx
processes. " Central to this conversion is the reaction of N H 2 with NO, which
1 4

has a primary product channel forming N2 and H2O. The secondary product
channel from this reaction appears to be N 2 H + OH, which acts as a chain
branching step, allowing the DeNOx process to continue. This reaction has also
been claimed to be an important N H 2 sink in the troposphere.
In a 1979 paper, Hack, Schacke, Schroter and Wagner noted: "It was a great
success when Haber and Bosch first broke the N-N bond to produce N H 3 from N2
molecules. Nowadays the destruction of N2 and the resulting production of NOx
leads to severe problems since the production of energy by fuel-air combustion
5
has increased so rapidly in the last twenty years."
Reactions of NHx species most often involve formation of adducts, which undergo
extensive rearrangement before proceeding on to final product formation. For
example, the N H 2 + NO reaction forms an H 2 N N O adduct, which undergoes a 1-3
Η shift, a cis-trans isomerization about the NO bond, a trans-cis isomerization
about the NN bond and a four-center elimination to produce N2 + H2O products.
The reactions of NHx species also typically have more than one set of possible
products. For example, the reaction of NH with NO can yield Η + N 2 O products
or OH + N2 products. Under conditions typical in thermal DeNOx (temp = 1250
K, excess O2 present) Η atoms are rapidly converted via the Η + O2 —> OH + Ο
reaction, making the Η + N 2 O channel chain branching, and the OH + N2 channel
merely chain propagating. Additionally, N 2 O produced via the Η + N 2 O channel
reduces the DeNOx efficiency. It is clear that, in order to successfully model
systems containing NHx species, one must know something of the product
branching fractions. The presence of multiple products on the potential energy
surfaces for these reactions also means that the surfaces can be accessed from a
variety of reactants. To continue consideration of the NH + NO reaction, the
70

HNNO surface can also be accessed by the Η + N 2 O reaction. This allows

additional information to be obtained about the potential energy surface by
experimentally accessing the surface in different ways.
An additional complexity encountered in NHx reactions is the existence of
multiple energetically accessible potential energy surfaces. This may take the
form of a surface crossing connecting reactants and products or the existence of
1 3 1 3 5
multiple surfaces accessible to reactants, such as the > A', ' A" and Π surfaces
accessible to NH + Ο reactants. In the former case, calculations of reaction rate
constants become recast as calculation of rates of surface crossing. In the latter
case careful consideration has to be given to the correlations between reactants
and product asymptotes. In the case that reactants and products are connected
by multiple electronic states we lack a detailed understanding of the effect of the
excited e*ate surface on the reaction dynamics. Typically in cases of this type the
reaction is modeled by assuming that the reactive flux is separable into the
various channels, and there are no interference effects between the channels.

1.1 Scope of this review

The chemistry of NHx species has been reviewed by various authors, with various
emphasis, in the past. Lesclaux reviewed chemistry of the N H 2 radical in 1984, 6

covering sources for and detection of NH2, as well as its reactions with a variety
7
of species. Miller and Bowman produced a review focusing on combustion
chemistry of nitrogen-containing species. They discussed at some length
mechanisms for NOx formation and removal in combustion systems, as well as
producing a mechanism for nitrogenous species in combustion. Hanson and
8
Salaam also reviewed chemistry of nitrogen-containing species, focusing on a
careful examination of experimental results at high temperatures, and giving
evaluated rate constant data for reactions of species containing Ν, Η and O.

Recent years have seen a continuation in the efforts to provide compilations and
evaluations of reaction rate constants. The compilation of reaction rate
9
constants in the NIST database stands out for being current (new editions are
issued roughly yearly), and for its ease of use. It is a computerized database
allowing searches on both reactants and products, together with graphical
presentation of the cited results. It draws on experimental and theoretical
results, and includes recommended reaction rate coefficients from various
evaluations.
10 11
Baulch, et a l and Cohen and Westberg have evaluated kinetic data for
combustion modeling, including some reactions of nitrogen containing species.
12 13
There are also recent evaluations by Tsang and Herron ' focusing on the
nitrogen chemistry important for energetic materials (i.e. explosives and
propellants).
 71

14 15
IUPAC and NASA/JPL evaluations focus on atmospheric chemistry, and
provide evaluations of the experimental and theoretical results on a given
reaction. Because of their focus on atmospheric conditions the reactions selected
(and rates recommended) are appropriate for 200 < Τ < 300 Κ and pressures at
and below 1 atm. Extrapolation to combustion conditions should be done with
great care.
These sources, a i d others I may have omitted, have lessened the need for a
review to compile rate constant information, and, in many cases, to evaluate that
information. We shall instead focus on a few reactions of NHx species in an
attempt to highlight the richness of the chemistry. In particular, we will examine
the reactions of NH with O, NH with Ο, NH with NO, and NH with NO. We
2 2

shall emphasize direct determinations of reaction rate coefficients and products,

and de-emphasize results obtained by extensive modeling of complex reacting
systems (most often flames). While modeling complex systems is important
from both an engineering perspective and in highlighting regions of ignorance
with respect to elementary chemical processes, it has a poor track record for
accurate quantification of elementary chemical processes. We will similarly de-
emphasize purely theoretical calculations of elementary chemical processes.
We will draw heavily on recent results of electronic structure calculations, which
are now capable of predicting energies of reactive intermediates and transition
states to within a few kcal/mole. Fitting into this category of calculations are
16,17
those carried out using the Gaussian-2 (G2) and G2Q methods , and those
carried out using various CASSCF/MRCI or CCSD(T) schemes with large basis
18
sets. These methods, when carefully used, are found to agree quite well with
one another; as a further test, results of the G2 method have been compared to
experimentally measured heats of atomization for 55 well-established species
containing first and second-row elements. The average absolute deviation in
16
heats of atomization was found to be 1.1 kcal/mole. Comparable results were
found in an examination of a number of well-characterized transition states
17
using a modification of the G2 method, dubbed G2Q. While it is arguable
whether this is truly "chemical accuracy", these results speak to the quality of
modern theoretical calculations, and the added insight these calculations can
bring into the often complex reaction mechanisms encountered in studying NHx
species.

1.2 Thermochemistry
Before we begin in earnest, it is important that we are all "playing by the same
rules," which in this case we take to mean the use of well defined, and hopefully
accurate, thermochemistry. Table 1.1 presents the experimentally known
thermochemistry for the H N O system. In general we follow the
x y z
19 20
recommendations of the most recent JANAF tables. Anderson has recently
re-evaluated data relating to the heats of formation of NH and NH ; we choose to
2

use his recommended value for NH. For NH we have chosen the value
2
72

recommended in the recent evaluation of RH bond strengths by Berkowitz,

21
Ellison and Gutman, which is in excellent agreement with Anderson's
recommended value. We have chosen to revise the JANAF heat of formation for
1
HNO to reflect a refined value of 16450 ± 10 crrr for the predissociation
22
threshold for the HNO (A)-state. This has the effect of lowering the heat of
formation of HNO by 1.6 kcal/mole. The heats of formation and classical well
depths, De(H-NO), for the excited states of HNO have been derived from that of
23 24 26
the ground state, making *se of the known spectroscopy of NO, and HNO, '
3 27
and calculation of the vibrational frequency for the vi mode in HNO (A) A".
The classical well depths for the three states of HNO have been included to
simplify comparison with theoretical results. The thermochemistry for the
majority of the H N O adduct species has not been experimentally determined,
x y z

although there is considerable spectroscopic data on H 2 N O , and limited

spectroscopic datn on other of the adducts.

Table 1.1: Experimental Thermochemistry for the Η, N, 0 system

species AHf(OK) De(H-NO)
ikcal/mole) (kcal/mole)
Η 51.634 ± 0.001
Ν 112.53 ± 0.02
0 58.98 ± 0.02
OH 9.2 ±0.3
NO 21.46 ± 0.04
NH 85.4 ± 0.3
NH 2 45.8 ± 0.3
N 0 2 20.4 ±0.1
H 0 2 -57.10 ±0.01
HNO (X^A* 26.06 ± 0.05 52.59 ± 0.03
HNO (A)3A" 44.0 ± 0.4 34.6 ± 0.4
HNO (A)1A" 63.67 ± 0.05 14.22 ± 0.03

2 THE Η, Ν, Ο SYSTEM
The reaction of NH with Ο accesses NHO/HON potential energy surfaces, and a
wealth of interesting chemistry emerges. If we look in detail we find that there
are three bimolecular reactions linking HNO containing species,
NH + 0*->H + NO (2.1)
NH + Ο <-> Ν + OH (2.2)
Η + NO Ν + OH (2.3)
plus four association/dissociation reactions,
Η + NO (+M) <-> HNO (+M) (2.4)
Η + NO (+M) <-> HON (+M) (2.5)
NH + Ο (+M) <-> HNO (+M) (2.6)
Ν + OH (+M) <-> HON (+M) (2.7)
 73

and one isomerization reaction,

HNO (+M) <-> HON (+M) (2.8).
Additionally, there are singlet, triplet and quintuplet surfaces which can be
accessed by these reactions. A rich reactive system, indeed.

2.1. The HNO Potential Energy Surface

The HNO potential energy surface has been of theoretical interest for a number of
28
years. Bruna and Marian and Bruha carried out MRD-CI calculations on the
lowest singlet and triplet states of HNO and NOH and the transition states for
isomerization. Later workers have refined and extended their calculations,
29 36
including other states and reaction pathways. " There have been several
37
surfaces generated including ones by Carter, Mills and Murrell, Dixon and
22 38
coworkers, and Colton and Schatz. Recently Guadagnini, Schatz and Walch
1 3 Χ
have produced a global potential energy surface for the lowest A', A" and Α"
27
states of HNO based on CASSCF/ICCI calculations with large basis sets. The
results of the ab initio calculations were scaled to reproduce the experimental
thermochemistry; the Ν + OH asymptote was adjusted by 4.4 kcal/mole, and the
f
Ο + NH asymptote by 2.6 kcal/mole. With this scaling they calculate D (H-NO) s
e

Figure 2.1: Potential energy surfaces for HNO species.

74

for the three lowest states of HNO of 50.1, 31.7 and 10.9 kcal/mole, respectively.
Comparing these D 's with the well depths tabulated in Table 1.1 suggests that
e

their surface is accurate to no better than perhaps 3.5 kcal/mole.

A schematic of the HNO potential energy surfaces is shown in Figure 2.1, with
correlations indicated, as well as energies of transition states linking various
27 34
reagents, with energies taken from Guadagnini, Schatz and Walch and Walch.

2.2. NH + Ο -> Products

We will consider the two reactions
NH + 0 - > H + NO Δ Η = -71.3 kcal/mole Γχη (2.1)
and
NH + 0 - > N + OH Δ Η = -22.7 kcaUmole Γχη (2.2)
3 3
together. The interaction of Ο P with NH Σ gives rise to 18 surfaces, 2 singlet,
2 triply degenerate triplets and 2 five-fold degenerate quintets. The two singlet
1 1
surfaces correlate with the A' and A" singlet states of HNO. The association
27
reaction proceeds without a barrier, forming highly excited HNO. The lifetime
_14 39
of this excited HNO adduct has been calculated to be of the order of 1 0 s ;
collisional stabilization to the intermediate should not be expected under
anything less than very extreme conditions. The spin-allowed, direct H-atom
transfer between Ν and Ο occurs on triplet and quintet surfaces. The process has
34
been examined theoretically by Walch. He carried out CASSCF/CCI
calculations, and located the transition states for the direct abstraction,
5 3
assuming a collinear geometry on the Π and Π surfaces. He found that the
classical barrier for the reaction on the triplet surface was 11.7 kcal/mole, while
the classical barrier on the quintet surface was 5.6 kcal/mole. These barriers
17
have also been calculated by Durant and Rohlfing using G2Q. They obtained
values of 11.9 kcal/mole on the triplet surface, and 5.4 kcal/mole on the quintet
5
surface, in excellent agreement with Walch's results. The linear Π configuration
is lower in energy than the nonlinear configurations, and thus represents the
unconstrained transition state on the quintet surface. However, as the linear
3
triplet state is bent, it splits into a A" surface, which is stabilized by the
3 3 f
deformation and correlates with the A" HNO surface, and a A surface, which is
34
destabilized by the deformation. Presumably, then, the reaction occurs mainly
on the A\ A" and nonlinear A" surfaces, since they connect reactants to
1 1 3

products without the presence of barriers higher than the reactant asymptote.
This picture of the reaction leads to the expectation of a fast radical-radical
reaction with little temperature dependence.
Experimental data on these reactions are sparse. Temps has measured a
-12 3
reaction rate coefficient of k(2.1+2.2) = 8 χ Ι Ο cm /molecule/s at room
40
temperature, using the discharge-flow technique with LMR detection.
41
Adamson and coworkers, using infrared kinetic spectroscopy, have measured an
= 1 1 3
overall rate of k(2.i+2.2) (6.5 ± 1.5) χ 10 cm /molecule/s. Hack, Wagner and
42
Zasypkin have used a quasi-static laser flash photolysis technique to measure
 75

an upper bound for reaction 2.2 of k2.2 ^ 1.7 χ 10* cm /molecule/s. The small
13 3

size of the branching fraction for OH production is what we would expect based on
the reaction of NH with Ο proceeding through a ΗΝΟ/ΗΟΝ adduct which then
dissociates, without a barrier, to products. The reactions are highly exothermic,
and we do not expect the back reactions, k.2.1 and k.2.2, to be important. Mertens
43
et a l . studied the reaction in shock-heated HNCO/N20/Ar mixtures in the
2730< Τ < 3380 Κ range. NH concentrations was monitored by cw, narrow-
linewidth laser absorption. By modeling a full reaction scheme they extracted a
= 10
temperature independent reaction rate coefficient of k(2.i+2.2) (1.5 ± 0.9) χ 10"
3
cm /molecule/s in the temperature range 2730 < Τ < 3380 Κ. They found that NH
concentrations in their HNCO/N20/Ar system were relatively insensitive to the
relative importance of reactions 2.1 and 2.2, hence they were unable to extract a
product branching fraction for the reaction. The roughly factor of 20 increase in
reaction rate between Temps' room temperature rate coefficient and Mertens et
al.'s 3000K rate coefficient is at odds with our expectations based on the
potential energy surface. Clearly more work is called for on this system.
The nascent product state distribution of NO from the Ο + NH —> Η + NO
39
reaction has been studied by Huang and Dagdigian. The reaction was studied
in a reaction cell at 60 mTorr. O-atoms were produced by microwave discharge of
O2 or 02/Ar mixtures, and NH was produced by 193 nm photolysis of NH3. They
obtained a nascent NO v=l rotational temperature of 1130 ± 50 K; their relative
NO vibrational populations are listed in Table 2.1. Huang and Dagdigian
calculated that, on average, 8% of the total reaction exothermicity appeared as
NO vibrational excitation, and 3% of the total reaction exothermicity appeared as
NO rotational excitation. This leaves, oil average, 92% of the reaction
exothermicity funrieled into product translational excitation. For a Η + HL —»
HH + L reaction, the reagent orbital angular momentum is funneled into Η Η
product rotational excitation, and the product rotational distribution is
determined by the velocity dependent opacity function for the reaction. The low
degree of rotational excitation suggests that reaction is favored for small impact
parameters and/or low collision velocities. The vibrational distribution is much
colder than predicted from a statistical prior distribution, but is well modeled by
calculation of the Franck Condon overlap between HNO V N O 0 and the free NO
=

molecule.
Table 2.1: Relative NO vibrational populations from the NH + Ο Η + NO
reaction.
ν relative population ν relative population
1 1 5 0.045 ± 0.017
2 0.43 ± 0.04 6 0.012 ± 0.005
3 0.20 ± 0.02 7 0.028 ± 0.019
4 0.17 ± 0.02 8 0.008 ± 0.007
76

2.3.N + OH<->NO + H
The reaction of Ν with OH is the third step in the Zeldovich mechanism for
44
conversion of N to NOx in combustion,
2

Ο + N <-> NO + Ν
2

Ν + 0 <-> NO + Ο
2

Ν + OH <-> Η + NO
This "thermal mechanism" accounts for a fraction of the NOx produced in
combustion of fuels which do not contain nitrogen. Reaction 2.3 is quite
exothermic, with
N + OH->H + NO Δ Η = -48.7 kcal/mole
Γχη (2.3)
4 2
The interaction of Ν S with OH Π will give rise to 16 electronic surfaces. If one
examines the potential energy surfaces in Figure 2.1 one sees that the relevant
3 5 5
state for reaction 2.3 is the triply-degenerate A" state of HON. The A' and A"
2
states of NOH are expected to be repulsive, and will not correlate with Η S + NO
2
Π products. Left out in the accounting (and in calculations of the potential
3
energy surface) is the A' state of HON. It is of higher energy than the states
which have been calculated, but may offer a direct path from reactants to
products.
The reaction rate coefficient of reaction 2.3 has been studied directly by Howard
45 46
and Smith, for 250 < Τ < 515 K, ' and by Brune, Schwab and Anderson at room
47
temperature. Howard and Smith used flash photolysis coupled with discharge
flow to measure the rate coefficient. They photolyzed H 0 in the presence of Ν 2

atoms created by a microwave discharge. The OH radical concentration profiles

were measured with lamp-induced fluorescence, while the N-atom concentrations
were derived using chemiluminescent titrations with NO. Analysis of their data
10 0 2 5 0 1 7
yielded a rate constant expression of k . = (2.21 ± 0.18) χ ΙΟ" Τ · ±
2 3
3
cm /molecule/s for the temperature range 250 < Τ < 515 Κ. Brune, Schwab and
Anderson used LMR, resonance absorption, and resonance fluorescence to
monitor species in their flow tube study of this reaction. Their measured rate at
11 3
room temperature, k .3 = (4.2 ± 0.8) χ 10" cm /molecule/s is in good agreement
2
11
with Howard and Smith's room temperature rate of k .3 = (4.7 ± 0.4) χ 10" 2
3 46 48
cm /molecule/s. Very recently Rowe, Sims and Smith have extended
measurement of this reaction rate coefficient down to 10 K, using laser-
photolysis/laser induced fluorescence technique in the discharge from a modified
Laval nozzle. They found that the rate continued to rise at lower temperatures.
The negative temperature coefficient for k .3 suggests that the reaction proceeds
2

without a barrier, passing through the HON well. The barrier heights calculated
27
by Guadagnini et al. suggest that isomerization of HON to HNO, followed by
dissociation to Η + NO, may compete with direct dissociation of HON to Η + NO.
At high temperatures the endothermic reverse reaction
H + NO->N + OH (-2.3)
begins to become important. This reaction serves to convert NO back into atomic
Ν in the context of the Zeldovich mechanism. This reaction was immensely
 77

popular in 1975 through 1977, with no fewer than 7 experimental studies of its
49 55
rate coefficient in the 2000 - 4000 Κ temperature range " . Results of these
determinations are plotted in Figure 2.2. Since the initial flurry of interest, the
rate coefficient has not been remeasured. The various experimental studies are
in fair agreement as to the temperature dependence of the reaction, but the rate
coefficients vary by roughly a factor of 10. The measured Arrhenius activation
energies are all slightly in excess of the reaction endothermicity. As important as
this reaction is, it is surprising that it has not been the subject of continued
experimental attention.
By use of the equilibrium constant one can predict k.2.3 based on measurement of
k2.3. It is interesting to use this to extrapolate Howard and Smith's data to
shock tube temperatures. Despite the length of the extrapolation, the predicted
rate falls within the range of the measured high temperature rates, as shown in
Figure 2.2.

2.4. Η + NO (+M) <-> HNO (+M)

From Figure 2.1 one sees that addition of Η-atom to NO correlates with the
lowest three states of HNO. The other three surfaces arising from combination of

h i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 • 1 • 1 • • ' ' 1 1 1 1 I l |

|-17 I. , , , I . . 1 • I • 1 1 1 I 1 1 • • 1 • • • • I • • • • 1 1 • • • 1 • • • •

0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6

1000/T (K" ) 1

Figure 2.2: Experimentally measured rate coefficients for Η + NO -> Products.

Ο - ref. 49, • - ref. 50, • - ref. 51, · - ref. 52, Δ - ref. 53, • - ref. 54, V - ref. 55, — rate
calculated from equilibrium constant and low temperature data of Howard and
46
Smith for Ν + OH -> Products. .
78

2 2
the S Η-atom with the Π NO correspond to a repulsive triplet state, with the
unpaired spins being those in the H-N (non) bond. CASSCF/CCI calculations by
33 27
Walch and Rohlfing as well as the calculations of Guadagini et al. predict that
the reaction on the !A' ground state surface occurs without an energetic barrier.
3
Walch and Rohfing found classical barriers of 4.1 kcal/mole on the A" surface
1 3
and 10.2 kcal/mole on the A" surface. The barrier on the A" surface has also
17
been calculated by Durant and Rohlfing, using the G2Q method. They find a
classical barrier height of 4.1 kcal/mole, in perfect agreement with the results of
Walch and Rohlfing. The surface of Guadagnini et al. has barriers of 4.6
kcal/mole and 11.1 kcal/mole for these transition states.
The association reaction of Η-atom with NO has been subject of experimental
56
study for decades. In 1929 Smallwood investigated the NO catalyzed
recombination of Η-atoms, which occurs when reaction 2.4 is followed by the fast
reaction
Η + HNO -> H + NO 2 (2.9).
57
In 1933 Harteck c jndensed product from the reaction of Η-atoms with NO, and
collected a solid of empirical formula HNO. The explosive nature of the adduct
was demonstrated by multiple detonations in the course of the experimental
58 59
study. In the '50s Cashion and Polanyi and Clement and Ramsay observed
chemiluminescence arising from the reaction. Transitions were found to originate
from both the electronically excited ^A" state and vibrationally excited levels of
1 I
the ground A state.
1 1
The presence of excited HNO A ^ products from the reaction of Η with NO is
interesting in light of theoretical predictions of a barrier for the adiabatic
1
reaction forming HNO A A" from Η and NO. A closer look at the spectroscopy of
HNO reveals the fact that the low lying states of HNO are mixed. Most apparent
is the rotational-level dependent predissociation in the HNO A^A" s t a t e . 26,60

22
This J-dependent predissociation has been analyzed by Dixon et al. and found
to arise from a electron orbital-rotational coupling with continuum levels of the
X A state. The variation in the predissociation threshold is linked to the
1

variation in the height of the J-dependent centrifugal barrier. Their detailed

analysis of this predissociation allows determination of Do(H-NO) = 16450 ± 10
1 61 62
cm" . Dixon and Rosser and Petersen also investigated the analogous
predissociation threshold in DNO, and found that it was shifted from the HNO
61
value by the difference in HNO and DNO zero point energies. This is the
behavior to be expected for a dissociation whose J = 0 surface is barrierless.
Additionally, the variation of the predissociation threshold with J is very
sensitive to the local curvature of the surface in the region around the centrifugal
barrier; existence of a J = 0 barrier would change the observed J-dependence of
the threshold, and is inconsistent with the experimental observations.
22
Dixon and coworkers also observed the onset of a rotation-independent
1
predissociation mechanism between 860 and 970 c m above the dissociation
3
limit. They ascribed it to spin-orbit coupling with continuum levels in the a A"
 79

3
state, and suggested that the barrier to HNO a A" formation was of this
33 17
magnitude. The barriers calculated by Walch and Durant and Rohlfing, and
27
Guadagnini et al. are somewhat higher than Dixon et al.'s range of 2.5 < Ε < 2.8
kcal/mole, but the difference is within expected error bounds.
There is also evidence for mixing of the states below the Η + NO dissociation
energy. There are numerous perturbations in the A^-A" state, and the lifetime of
the A A" state is found to be level-dependent, with efficient collisional mixing of
1

63 64
various levels. The results of the MRD-CI calculations by Bruna and the
22 27
potential surfaces constructed by Dixon et al. and Guadagnini et al. show that
3
the a A" HNO surface crosses the X^-A' HNO surface as the molecule is distorted
3
toward linearity or toward HON, with the A" state becoming the ground state
for those geometries.

10" ι ι I ιι

ο
Ο

-32
10 I • • • • I • • ' » ι ι I ι ι ι ι I ι ι ι ι I ι ι ι ι

1.5 2 2.5 3 3.5 4.5

1000/Τ (ΚΓ ) 1

Figure 2.3: Experimentally measured rate coefficients for Η + NO + H2 -> HNO +

H reaction. V - ref. 65, • - ref. 66, • - ref. 67, • - ref. 68, Ο - ref. 69, · - ref. 70, Δ
2

-ref. 71.
The HNO (A—>X) chemiluminescence, with its origin at -766 nm, has been used
by researchers to monitor Η-atom concentrations in the course of the Η + NO + Μ
-> HNO + Μ reaction. In the early '60s Clyne and Thrush carried out the first of
these studies, using the discharge-flow technique to measure the low pressure
recombination rate coefficient for the reaction with H2 as the third body. 65,72

They measured the reaction rate coefficient from 231 to 704 K, and found that it
exhibited the negative temperature dependence characteristic of recombination
80

2 3
reactions, with their data being well represented by k(H2) = 3.83 χ 1 0 '
-0 9 6 2
(Τ/298) · cm /molecule /s. The only subsequent temperature dependent studies
are due to Cvetanovic and coworkers and were only for H2 as the third body.
68,69

They used a modulated, Hg-sensitized technique to measure the rate coefficients.

In this technique Η-atoms were formed by exciting Hg-atoms using a modulated
mercury lamp, and then reacting the excited Hg-atoms with H2. The phase and
intensity of the HNO chemiluminescence was monitored with a lock-in amplifier.
This information was used to derive the kinetic parameters. Cvetanovic and
coworkers confirmed the decrease in reaction rate with increasing temperature,
although their measured rate coefficients are slightly larger than that measured
by Clyne and Thrush. The deuterium kinetic isotope effect has been measured by
73
Hartley and Thrush, who obtained a value of kn/kD = 0.9 ± 0.2 with Ar as the
71
third body, and by Ishikawa, Sugawara and Sato who obtained a value of kn/kD
= 1.0 ± 0.3 with H2 as the third body.
The relative efficiencies of H2, He, Ar and Ne as third bodies in the
65
recombination reaction was studied by Clyne and Thrush, and later by
67
Ahumada, Michael and Osborne who added Kr to the list of third bodies.
73
Hartley and Thrush extended the study of third body efficiencies to a number of
poly atomics. Unfortunately, all these workers restricted their work to room
temperature, and did not examine the differing temperature dependencies for the
relative third body efficiencies. A single study of the temperature dependence of
74
Ν2 was undertaken by Campell and Handy. They examined the limited
temperature range 360 < Τ < 425 Κ, and derived a relative efficiency
k2.4(N2):k2.4(Ar) = (1.64 ± 0.12):1.0. A collection of measured room temperature
third body efficiencies relative to Ar can be found in Table 2.2. Table 2.2 also
lists third body efficiencies corrected for relative collision velocities. With this
correction, data for the rare gases behaves as expected, with the heavier rare
gases being more efficient than the lighter ones. The same trend is seen in the
data for polyatomic colliders; the exception, H2O, probably owes its enhanced
efficiency to its dipole moment.

Table 2.2. Relative third body efficiencies for Η + NO + Μ -> HNO + M; Ar = 1.0
Clyne and Thrush Hartley and
65 ' —' ~ — -·~
ref. Ahumada, et. al.67
73
Thrush
Μ k/kAr velocity k/kAr velocity k/kAr velocity
corrected corrected corrected
H 2
1.7±0.3 0.19±0.04 1.9±0.3 0.20±0.03 1.60±0.09 0.18±0.01
He 0.8±0.2 0.15±0.04 1.10±0.04 0.23±0.01
Ne 0.8±0.2 0.6±0.1 0.50±0.03 0.35±0.02
Kr 1.40±0.09 1.81±0.11
C0 2 2.0±0.3 2.1±0.4
N 0 2 2.2±0.5 2.3+0.5
SF 6 3.6+0.6 5.2±0.9
H9O 6.1±1.4 4.0±0.9
 81

Clyne and Thrush also measured the rate coefficient for production of HNO (A ->
65
X) chemiluminescence. This was accomplished by first calibrating their
detection system for absolute chemiluminescence detection efficiency. NO
concentrations were obtained from flow rates, while Η-atom concentrations were
obtained by use of an isothermal calorimeter. The calorimetric determination of
atom concentrations involves measurement of the heat released when atoms
recombine on a suitable substrate, usually platinum foil or wire. These
measurements sufficed to determine the chemiluminescent reaction rate
coef^ient, independent of the overall reaction rate coefficient. They found that
the chemiluminescence rate coefficient was pressure independent, although it
was dependent on the identity of the bath gas. They also determined the
temperature dependence of the chemiluminescent channel, and found equally
good fits to an Arrhenius form, with activation energy of -1.4 ± 0.3 kcal/mole, and
n
a T form, with a n η of -2.8 ± 0.4.
This rather striking behavior was explained by Clyne and Thrush using a simple
65
reaction sequence:
Η + NO + Μ -> HNO (X) + Μ (2.11)
Η + NO + Μ -> HNO (A) + Μ (2.12)
HNO (A) -> HNO (X) + hv (2.13)
HNO (A) + Μ -> HNO (X) + Μ (2.14)
By applying the steady-state assumption to [HNO (A)] they obtained an
expression for the chemiluminescence intensity:
I = k .i [HNO (A)] = {k .i k2.i3[M]/(k .i3 + k . [M])}[H][NO]
2 3 2 2 2 2 14

In the limit of k . i 3 « k .i4[M], i.e. fluorescence quenching much faster than

2 2

radiative decay, this expression becomes independent of pressure, but is still

sensitive to the nature of the bath gas due to the Μ dependence of rates k . i and
2 2

k .14. The temperature dependence of the chemiluminescence also reflects

2

temperature dependencies for both k . i 2 and k2.i4» and is thus not constrained to
2

be equal to the temperature dependence of k2.1i· The radiative lifetime of HNO

75
(A) has been measured to be τ = 22 - 29 μβ, and while the quenching rate with
H2 has not been measured, we would expect quenching to be at least competetive
with radiative decay at Clyne and Thrush's pressures of 1-2 Torr.
Alternative mechanisms can be constructed in which the association of Η and NO
occurs only on the X*A' surface, with HNO (A) formed by reverse internal
conversion, and both X and A states undergoing collisional relaxation. For
example, consider the scheme:
Η + NO <-> HNO (X)t (2.15)
HNO (X)t <-> HNO (A)t (2.16)
HNO (X)t + Μ -> HNO (X) + Μ (2.17)
HNO (A)t + Μ -> HNO (A) + Μ (2.18)
HNO (A) + Μ -> HNO (X) + Μ (2.19)
HNO (A) -> HNO (X) + hv (2.20)
where f denotes vibrationally excited species. We can again solve for [HNO (A)]
82

by use of the steady-state assumption, and obtain, in the low-pressure limit

(which is appropriate for pressures below many atmospheres)
I = k .20[HNO (A)] = {K .l5K2.16k2.18k2.20[M]/(k .20 + k .i9[M])}[H][NO]
2 2 2 2

where the K's are equilibrium constants. Like Clyne and Thrush's equation, this
equation ceases to be pressure dependent when quenching becomes dominant,
retains the dependence on bath gas identity, and can have a temperature
dependence different from that of the overall recombination reaction.
There has been one measurement of the limiting high pressure rate constant for
the Η + NO -> HNO association, by Forte, who measured a rate of k«> = (8 ± 3) χ
10 3
10" cm /molecule/s at pressures above 400 atm in N 2 . 7 6

2.4. Η + NO (+M) <-> HON (+M)

Η-atom can also add to the O-atom of NO to form HON. On the A surface this 1 1

addition involves the formal promotion of an electron from Ο to N, and is

expected to proceed with a barrier; the reactions on the 1» A" surfaces also involve
3

33
breaking an NO π bond, and are also expected to proceed with barriers. This
27
analysis is borne out by the detailed calculations of Guadagnini et al., who find
X 3 1
barriers of 3.9, 13.5 and 25.6 kcal/mole for the A, A" and A" surfaces,
respectively. Pauzat et al. have also calculated the barriers on the ^A' and A"
77 3

surfaces, and calculate values of 6.7 and 13.0 kcal/mole, using the SOCI method.
Thus, one does not expect that direct formation of HON will be important in the
recombination of Η and NO. However, the lowest energy product channel in the Η
+ NO reaction is Ν + OH. This channel is obviously not directly accessible from
5 2
an HNO intermediate. In fact the ground state Ν S and OH Π products do not
correlate with the singlet HNO/HON surfaces. However, they do correlate with
3
the ground state A" state of HON. One therefore expects that the isomerization
process, reaction 2.8, may be important in the Η + NO —> Ν + OH reaction.
3
Alternatively, the reaction may proceed directly through the A surface, although
reaction on that surface presumably involves a barrier.

2.5. NH + Ο (+M) <-> HNO (+M) and Ν + OH (+M) <-> HON (+M)
The association reactions of NH with Ο and Ν with OH are very exothermic and
possess thermodynamically open product channels. We therefore expect that
they will have very short adduct lifetimes, and will not be stabilized under
39 1 4
normal conditions. For example, Huang calculated an RRK lifetime of 10 s
for the adduct arising from NH adding to O, which would require pressures of
4
~10 bar before stabilization would be observable. We therefore expect these
reactions to go on to yield bimolecular products, as discussed in 2.2 and 2.3.
 83

2.6. HNO (+M) <-> HON (+M)

The isomerization of HNO to HON is presumably a necessary step in reaction
2.3, assuming that the reaction goes via a stable intermediate. There is no
experimental data on this isomerization. The most recent theoretical
27
calculations suggest that all the barriers to isomerization from HNO lie above
the barriers to dissociation back to Η plus NO, while for HON the barrier to
isomerization on the 3A" surface is the only one which is below the barriers for
dissociation to Η plus NO. We thus expect that the major loss pathways for
HNO and HON will be dissociation rather than isomerization.

3. THE H NO SYSTEM
2

We next consider the reaction of NH with O, and, in general, reactions on the

2

H 2 N O potential energy surface.

3.1. The H2NO Potential Energy Surface

Properties of dihydroxy nitrosyl radical, H NO, and its isomers have been the
2
77,80 91
focus of several ab initio electronic structure studies. " The H NO radical is
2

the simplest member of a series of nitrosyl-radical spin labels, RiR NO. As 2

such, accurate prediction of its geometry and ESR spectrum is important for a
full understanding of the results of ESR investigations of more complex spin-
77,86,87
labeled s y s t e m s . A common thread through various studies has been the
concern about the planarity, or non-planarity, of H NO. Komaromi and
2

Table 3.1: Comparison of H?NO Geometries

NO r
ZHNH ZHNOH Method reference
78
0.99 > b
1.34 122.3 > 180.0 ) b
LMR
b
Davis et al.
79
1.010 ± 1.280 ± 122.7 ± 180.0 Microwave Mikami et al.
0.010 > 0.004
b
2.2 spectroscopy
80
1.014 1.265 119.5 180.0 CISD/ Soto et al.
cc-pVDZ
8 1
1.014 1.283 121.6 180.0 MRSDCI/ Cai
DZ + P
82
1.013 1.291 120.3 180.0 QCISD/ Barone et al.
D95(d,p)
83
1.037 1.298 114.0 180.0 CASSCF/VDZP Walch
1.015 1.268 119.4 180.0 CCD-ST/ Komaromi and
84
6-311+G(df,pd) Tronchet
85
1.017 1.273 119.4 180.0 G2 Yang et al.
a) Bond lengths in A; angles in degrees.
b) Fixed
84

84
Tronchet have calculated geometries for H 2 N O using upwards of 60
combinations of basis sets and methods, from HF/6-31G to CCD-ST/6-
80 82 83 85 89
311+G(df,pd). Their results are representative of other investigators, ' ' ' '
who find that the molecule has an almost flat bending potential out to about 30°
from planarity, at which point the potential energy begins rapidly rising. There is
consensus that the height of the inversion barrier is below the zero point energy
in the bend.
There have been two experimental investigations of the H 2 N O geometry, an LMR
78
study by Davis, et a l . , and a microwave study by Mikami, Saito and
79
Yamamoto. Neither study allowed for a complete determination of the H 2 N O
geometry, and the data were analyzed using input from early theoretical
calculations. Both experimental studies were able to establish that the H 2 N O
was planar. Their results are in reasonable agreement with recent theoretical
calculations, as shown in Table 3.1.
2
Various stationary points on the H 2 N O A" potential energy surface have been
80 92 83
characterized by Soto and Page, Page, Soto and McKee, Walch and Yang,
85
Koszykowski and Durant. Table 3.2 lists the energies, in kcal/mole relative to
the Η + HNO asymptote, found by the various groups. The agreement between
the various studies points out again the quality possible in modern electronic
structure calculations. The potential energy surface is sketched out in Figure 3.1.
It shows that the reaction of N H 2 with Ο is a highly exothermic, barrierless
recombination producing an energized H 2 N O species. This species can either
fragment to products, or isomerize to HNOH. HNOH exists in both cis and trans
conformations, which are linked by a low isomerization barrier. These species
can fragment to Η + HNO, or to NH + OH, or reisomerize to H 2 N O .

Table 3.2. Comparison of Calculated H 2 N O Energetics (kcal/mole, relative to

Η + HNO)
80 92 83 85
Soto Page Walch Yang Exper­
iment

NO + H2 -55.7 -- -- -57.1 -56.2

NH + OH -- 18.4 16.9
H NO (planar)
2 -58.8 -- -57.2 -59.8 --

--- -
(H NO -> H + NO) ts
2 2
-- 1.9
(H NO Η + HNO (X)) ts
2 2.7 2.3 1.2 --
(H NO -> trans-HNOH) ts
2

trans-HNOH -53.0 --- -9.2

-52.1
-8.3
-52.7 —

-
(trans-HNOH -> Η + HNO (X)) -- 9.0 8.3 7.6
ts
(trans-HNOH -> cis-HNOH) ts __ ._ -38.9 __

cis-HNOH -47.9 -- -46.9 -47.5 --

- -
(cis-HNOH -> Η + HNO (X)) ts 6.3 --
 85

93
Lifshitz, et al. used tandem mass spectroscopy to study H 2 N O and HNOH.
This work established the existence of stable H 2 N O and HNOH isomers, and
indicated that there was a substantial barrier to their interconversion, consistent
85
with the results of Yang et al.
81
Cai also calculated the energies of the first two excited doublet states of H 2 N O ,
which he placed 49.6 and 96.0 kcal/mole above the ground state. The lower of
these, a B 2 state (in C2v symmetry) correlates with reactants and is
2

energetically accessible to N H 2 + Ο reactants; its role in the overall reaction has

not been examined in detail.
The last two sets of surfaces correlating to N H 2 + Ο are two quartet surfaces,
which connect N H 2 + Ο with NH + OH via a direct Η-atom abstraction. The
transition state on the lower of these surfaces has been characterized by Yang et
85
al., they find a 6.7 kcal/mole barrier for the reaction.

A'
4

NH + 0
2

. ^ NH + OH

2 \
Η + HNO (X)
x
' A t *

f
1
1
ι A +'' '
f ' \

I /
I / cis-HNOH
trans-HNOH
H + N0
H NO
2
2

2 4
Figure 3.1: Potential energy surface for A" and A' states of H2NO. Energies
85
from G2/G2Q calculations of Yang, et al.

3.2. The N H 2 + Ο -> Products Reaction

There are four exothermic channels in the reaction of N H 2 with O:
NH + 0 ->H + HNO
2 A H = -48.7 kcal/mole (3.1a)
rxn

-> NH + OH A H = -10.2 kcaUmole (3.1b)

rxn

- ^ H + NO 2 A H = -83.3 kcal/mole (3.1c)

rxn

->H 0 + N 2 A H = -49.4 kca^mole (3. Id)

rxn
86

There are three reported measurements of the overall reaction rate, k3.i. Gehring
94
and coworkers used flash photolysis in a discharge-flow reactor to measure the
reaction rate coefficient, obtaining k3 χ = 3.5 χ 10" cm /molecule/s at 298 K. 12 3

95
Dransfeld and coworkers used the discharge-flow technique to study the
reaction, and obtained a much larger rate coefficient, k3.i = (8.8 ± 2.5) χ 1 0 -11

3
cm /molecule/s at 296 K. They also measured the channel-specific rates, finding
k 3 . i = (7.6 ± 2.0) χ 10" cm /molecule/s and k .ib = (1.1 ± 0.5) χ 1 0 "
a
11 3
3
11

cm /molecule/s. They were unable to detect H2O in their experiments, ruling out
3

channel 3. Id, and from the time profiles of H2 and NO products were also able to
41
rule out channel 3.1c. Recently, Adamson and coworkers used infrared kinetic
spectroscopy to study the reaction. Their measurement of the overall reaction
rate coefficient, k3 χ = (6.5 ± 1.3) χ 1 0 ' cm /molecule/s, is in good agreement
11 3

with Dransfeld, et al's rate. Adamson and coworkers established that <E>3.ib <
8%, somewhat smaller, but still consistent, with Dransfeld, et al's results.

Dagdigian and coworkers have carried out a study of the HNO product from the
NH + Ο reaction, and the OD product from the ND2 + Ο reaction. They found
2
96 97

that the products were much colder rotationally than predicted statistically.
They suggested that there are barriers present in both product channels.
85 92
However, both G2Q and MCSCF/CCI calculations do not find significant
barriers in these product channels. Recent work probing nascent rotational
populations of fragments from the barrierless decomposition of H 2 O 2 , 9 8

99
HN03, and C H 2 C O find that, for excess energies more than a few kcal/mole
1 0 0

above thresholds, the product rotational distributions are considerably colder

than phase space predictions. This suggests that rotationally cold fragments do
not require the presence of barriers in the dissociation channel.

3.3. Other Reactions

Of the other reactions in the H 2 N O system, only the reaction of Η with HNO,
Η + HNO -> products (3.2)
has been directly measured. Dodonov and coworkers used the discharge-flow
technique to obtain values of k3.2 = (1.2 ± 0.5) χ 10" cm /molecule/s and k3.210 3 101

1 0 3 102
= (1.3 ± 0.5) χ 1 0 cm /molecule/s at 295 K. This rate is consistent with work
by Washida and coworkers, who reported a lower limit of k3.2 > 1.6 χ 10~
103 12

3
cm /molecule/s at room temperature.

4. THE HNNO SYSTEM

The reactions of NH with NO and Η with N 2 O both access the HNNO potential
energy surface. The NH + NO —> N2 + OH channel has been postulated as a
104
source of OH in DeNOx systems. Additionally, it has been claimed that the
NH + NO —>· N 2 O + Η channel is a principle source of N 2 O in combustion. Of 7

course other flame studies have suggested that the reaction produces very little
 87

N2O. The Η + N 0 —» N + O H reaction is one of the reactions converting N 0 ,

1 0 5
2 2 2

an undesirable combustion product, to N . The potential energy surface of H N N O

2

is one of the better studied polyatomic potential energy surfaces, with a variety of
theoretical calculations which have been checked by comparison to experiments.

The reaction of N H with N O has two thermodynamically open paths leading to

ground state products:
N H + N O -> Η + N 0 AHrxn = -34.7 kcal/mole
2 (4.1a)
OH + N AHrxn = -97.6 kcaVmole
2 (4. lb)
The reaction of Η with N 0 has only one thermodynamically open path, although
2

hot-atom studies have observed the endothermic channels (4.2b) and (4.2c):
H + N 0 -> OH + N
2 AHrxn =-62.9 kcal/mole
2 (4.2a)
-> O H ( A ) + N AHrxn=29.7kcaymole
2 (4.2b)
-> N H + N O AHrxn = 34.7 kcaUmole (4.2c).

4.1 HNNO Potential Energy Surface

Various portions of the HNNO potential energy surface have been calculated by a
106 108 112 3 2
number of investigators. ' " The addition of NH Σ to NO Π gives rise to
2
12 electronic surfaces. In Ci symmetry these are 2 doubly-degenerate A surfaces
4
and 2 4-fold-degenerate A surfaces. The thermodynamically accessible products
Χ 2 Χ 2
are N Σ + OH Π and N 0 Σ + Η S . These products will not correlate to the
2 2

quartet states, and reaction to these products must take place on the doublet
109
surfaces or violate spin conservation. Fueno et a l . have characterized the cis
and trans HNNO isomers on both doublet surfaces and find them to be planar, of

NH + NO
ΝΝ0--Η

\
/ H--NN0 ^ '
/ N
\

-50
\ /* H + NoO \
\
CIS HNNO \
TRANS HNNO \
> \

w
-100 N + 0H
2

106.
Figure 4.1: Potential energy surface of HNNO 2A surface.
88

A' and A" symmetry. Further calculations of the stationary points on the HNNO
106 108
ground state surface predict that all of the stationary points are planar, ' "
2 2
110,112 j hi
n w the two A states become a ground A' surface and an excited
c n c a s e
2 2
A" surface. Walch has considered the details of the correlations of the A' and
2 2
A" surfaces, and has concluded that the A" surface is not reactive for NH + NO
reactants; the Η + N 0 asymptote does not correlate with it, and reaching the N
2 2
112
+ OH asymptote requires energetically unfavorable electron rearrangements.
2
As a result, the potential energy surface of interest is just that of the ground, A'
106
state. Results of the G2/G2Q calculation of the stationary points on the surface
are shown in Figure 4.1. Results of calculations for stationary points on the
2
HNNO X A' potential energy surface are collected in Tables 4.1 and 4.2.
a
Table 4.1: Energetics of stable HNNO species
cis-HNNO trans-HNNO N +OH 2 N 0+H 2 reference
106
-48.9 -56.0 -96.9 -34.5 Durant
112
-46.2 -51.5 -31.7 Walch
111
-51.9 -48.8 -99.1 -37.6 Miller
110 b
-30.9 -90.8 -28.9 Harrison '
-29.5 -91.6 -20.5 110
Harrison * c

-60.3 -60.5 -92.4 -30.9 Fueno109

-50.2 -47.3 -97.5 -31.4 Melius108

- -97.6 -34.7 experiment

a) all energies in kcal/mole, relative to NH + NO, 0 Κ
b) MP4/6-3lG(d)//MP2/6-3lG(d)
c) MP4/6-31 lG(d,p)//HF/6-31 lG(d,p)
3
Table 4.2: Energetics of HNNO transition-state species
cis trans N -OH H-N 0
2 cis-trans ts
2 reference
HNNO HNNO
106
2.9 0.1 -17.9 -25.3 -30.8 Durant
112
6.3 3.2 -15.4 -21.4 Walch
111
-22.4 -28.2 Miller
110 b
3.0 -6.9 Harrison '
110 c
3.6 -3.8 Harrison *
109
-9.0 -12.2 -28.7 Fueno
108
-17.3 -27.7 Melius
a) all energies in kcal/mole, relative to NH + NO, 0 Κ
b) MP4/6-3lG(d)//MP2/6-3lG(d)
c) MP4/6-31lG(d,p)//HF/6-31lG(d,p)
106
We turn our attention to a comparison of the recent calculations of Durant with
112
those of Walch. The relative energies of the various stationary points from the
two calculations are very close, with the only large error being in the location of
 89

the NH + NO asymptote. (Unfortunately, this is the reference point chosen by

both authors, so the derived energetics relative to NH + NO, shown in Tables 4.1
and 4.2, systematically disagree by 2.5 - 4.5 kcal/mole.) The NH + NO -> Η +
N 0 reaction represents a non-isodesmic process, meaning that there is a change
2

in the number and type of bonds in the process. Traditionally it is harder to

calculate accurate energies for non-isodesmic processes than for isodesmic
processes, although the results of the G2 method of Pople and coworkers seems to
be fairly accurate for both processes. The location of the N + OH and Η + N 0
2 2

asymptotes relative to the NH + NO asymptote can be calculated from

experimental heats of formation, and indicates that the G2 surface of Durant is
within 1 kcal/mole of the experimentally known energetics for asymptotic
reactants and products. If Walch's surface is moved up to reproduce the
experimentally known NH + NO —> N 0 + Η exothermicity then the two surfaces
2

agree to within 1 kcal/mole, except for the trans-HNNO isomer, where the error
increases to 1.5 kcal/mole. This suggests that these surfaces be considered as
accurate to within 1.5 kcal/mole, and the other surfaces can be judged relative to
them.
Experimental calibration of the HNNO potential energy surface has proceeded on
several fronts. Most interesting would be the comparison of the properties
calculated for the intermediate HNNO species with those experimentally
determined. Unfortunately, there is no experimental thermochemistry or
spectroscopic data for the intermediates. Kinetic information on the HNNO
system includes measurements of the reaction rate coefficients for the NH + NO
—> Products and the Η + N 0 —> Products reactions, as well as measurement of
2

the product branching fraction for the reaction of NH with NO. There are also a
number of dynamics studies, where nascent product state distributions have been
measured. These studies each contribute some knowledge concerning features of
the potential energy surface.

4.2. NH + NO -> Products

The reaction rate coefficient of NH with NO has been studied by a number of
42,104 113 116
workers at temperatures from a m b i e n t ' ' to shock tube and flame
43,105,117 121
temperatures, " with recent work by Wolfrum and coworkers covering
122
the range 293 < Τ < 1085 Κ. The directly measured rates are shown in Figure
4.2. There is general agreement as to the room temperature rate coefficient, with
-11 3
a weighted mean of k4.i = (5.2 ± 0.2) χ 1 0 cm /molecule/s (1 σ). The recent
43,119
shock tube measurements are in relatively good agreement that k4.i ~ 1 - 4 χ
11 3
10" cm /molecule/s for 2000 < Τ < 3500. The earlier study of Roose, Hanson and
118
Kruger relied on extensive modeling of shock heated NO/NH3/H /Ar mixtures;2

the disagreement between their results and the recent more direct
90

determinations reflects on the quality of their model. Other temperature

104
dependent studies include work of Harrison, Whyte and Phillips, who report a
temperature independent rate of k4.i = (5.8 ± 0.6) χ 1 0 cm /molecule/s for 269 < -11 3

122
Τ < 377 Κ, and the results of Wolfrum, et al., who report a three parameter fit
± 0 1
of k . i = 2.5 χ 10-10 (T/298)M-6° ° ) exp(-0.91 ± 0.05 kcal/mole/RT)
4
3
cm /molecule/s.

These results suggests that the reaction has a slight negative temperature
dependence, as we would expect for a radical-radical association reaction
occurring without a potential energy barrier. This result is consistent with the
106
entrance channel of the potential energy surface calculated by Durant, as well
112
as a recalibration of the calculations of Walch, as discussed above. The shock
43
tube study of Mertens, et al. is interesting, in that their results suggested that
the reaction has a slight positive temperature dependence in the 2000 < Τ <
3000K range. Their experiments, utilizing shock heated HNCO/NO/Ar mixtures
and cw, narrow-linewidth laser absorption monitoring of NH, are fairly direct, and
although the observed increase in rate coefficient with temperature is within
quoted uncertainties, it does appear real. One possible explanation of this is the
possible opening of an additional channel in the reaction, possibly the
endothermic production of N 2 H + Ο or the exothermic production of N + OH Α Σ. 2
2

100 1000 10000

Temperature (K)
F i g u r e 4.2: E x p e r i m e n t a l l y m e a s u r e d r a t e coefficients for NH + NO —> P r o d u c t s .
O - r e f . 113, V - r e f . 1 1 4 , B - r e f . 1 1 8 ^ - r e f . 104, · - ref. 4 3 , Δ - ref. 1 1 6 , D - r e f . 1 1 9 , X r e f .
123, • ref. 122, Τ ref. 42.
 91

More work is clearly indicated here.

The product branching fraction for the reaction of NH with NO has been measured
116 119 106
at room temperature by Matsui and coworkers Yokoyami, et al. , Durant ,
122 123
Wolfrum and coworkers and remeasured by Matsui and coworkers. Matsui
1
and coworkers used photolysis of HNCO to produce NH a A. This was reacted
1 3
with NO; a portion of the NH a A was quenched to NH Χ Σ, and a portion reacted,
3
producing N 0 + Η or N + OH products. The NH Χ Σ subsequently reacted with
2 2

the NO, producing N 0 + Η and N + OH products. In their original study they

2 2

observed OH profiles with growth rates different from the NH a^A decay rates,
3
and they assigned this OH to products of the reaction of NH Χ Σ. In their
reinvestigation they added variable amounts of Xe, which efficiently quenches
1
NHa A, and were able to more cleanly separate contributions due to reaction of
excited and ground state NH. They identified the growth rate in their observed
3
OH profiles as arising from vibrational relaxation of OH, not reaction of NH Χ Σ,
and concluded that the OH was produced by reactions of both NH electronic
states, revising the interpretation in their earlier work. After careful calibration
of their system they obtained a revised branching fraction of k 4 . i / k 4 . i = 0.80 ± a

0.18. Yokoyami, et al., as part of their shock tube study of this reaction, also
measured k^ia/k^i = 0.81 at 300 K. Durant used the discharge flow technique

0.8

0.6

0.4

0.2

0
0 500 1000 1500 2000 2500 3000 3500 4000
Τ (Κ)
Figure 4.3: Product branching fraction for NH + NO -> Η + N 0 reaction. • - ref. ,
2
1 2 4

• - ref. 119 (experiment), • - ref. 119 (theory), ref. Ο - ref. (experimental), ·

1 1 1 1 0 6

- ref. (theory), Δ - ref. 123, • ref. 122.

1 0 6
92

with mass spectrometric detection of reaction products to determine the product

branching fraction at room temperature. NH Χ Σ was formed by reaction of N H 3 3

1 2 5 , 1 2 6 1 5 1 8
with excess F . It was then reacted with excess N 0 , and the product
1 4 1 5 1 4 1 5 1 8
N N and N N 0 were measured mass spectrometrically. After correction
for N N produced by reaction of residual N H 2 from the NH source he obtained
1 4 1 5

values of k . i / k 4 . 1 = 0.8 ± 0.4 for NH reactant, and k . i / k 4 . l = 0.87 ± 0.17 for ND

4 a 4 a

reactant. Wolfrum and coworkers also used the laser photolysis/LIF technique,
x 3
photolyzing HNCO and using excess Xe to relax the NH a A to NH Χ Σ. They
measured a room temperature branching fraction of k4.i /k4.i = 0.85 ± 0.05. a

The product branching fraction for the reaction of NH with NO has also been
119
measured in shock tube experiments by Yokoyami, et a l . and by Mertens, et
43
al. Both groups pyrolyzed mixtures of HNCO/NO/Ar and monitored NH and OH
concentration profiles. Mertens, et al. operated in the temperature range from
2200 < Τ < 3350 Κ, and used cw narrow-linewidth laser absorption to monitor the
NH and OH. They obtained a branching fraction over the temperature range 2940
< Τ < 3040 Κ of k .ib/k4.i = 0.19 ± 0.10. Yokoyami, et al. used fluorescence from
4
2 3
thermally excited OH Α Σ and NH Α Π to monitor ground state OH and NH
populations. They derived a value for the other channel of k4.ib/k4.l 0.32 ± 0.07 =

at Τ = 3500 Κ. Thus one sees a gradual decrease in the importance of the Η +

N 2 O channel, which is what we would expect to see given the ordering of the
theoretically calculated barriers, i.e. E(NN-OH transition state) > E(H-NNO
transition state). As the temperature increases the importance of the barrier
heights for the two processes becomes less and less important, until at high
temperature the product branching fraction should just reflect the relative
Arrhenius Α-factors for the two channels.
119
Yokoyami, et a l . also observed an intense OH (A —> X) emission feature at
short time in their experiments, which they attributed to a third product channel:
2
NH + NO -> OH (Α Π) + N ΔΗ^η = -5.0 kcaUmole 2 (4. lc)
They calculated an upper bound of k4.i /k4.i < 0.01. This is the product channel
c
2 119
expected for reaction on the excited, A A" HNNO surface. A final product
channel which has been suggested at high temperature is the endothermic
formation of the transient species N 2 H : 1 2 7

NH + NO -> N H + 0 A H = 6.8kcaymole
2 (4. Id) rxn

The opening up of these product channels at high temperature could give rise to
the suggested positive temperature dependence of the rate coefficient observed by
111
Mertens, et a l . , and will also lead to a decrease in the branching fraction
k4.1a/k4.1-

There have been calculations of the product branching fractions by Miller, using
111 106
both a 1993 BAC/MP4 surface and using the G2 surface of Durant and by
119 109
Yokoyami, et a l . using the MRDCI surface of Fueno, et al. Miller calculated
 93

1 9 a t
k4.lb/(k4.la+k4.lb) = ° · 300 Κ, increasing to 0.30 at 3500 K. Yokoyami
obtained a value of k4.ib/(k4.la k4.1b) = 0.24 at 3500 K. These values are plotted,
+

together with the experimentally measured product branching fractions, in Figure

4.3. While the agreement between theory and experiment is not exact, it is
reasonably good considering the experimental uncertainties in the
measurements. The degree of agreement suggests that the theoretically
calculated potential energy surfaces do a reasonable job of capturing the
important features of the true surface, and that the measured product branching
fraction is not overly sensitive to barrier heights (certainly not relative to the
sensitivity of reaction rate coefficients on barrier heights!).

The measured product branching fractions confirm that the reaction of NH with
NO will be a source of N 0 . It also establishes that it is unlikely to be the source
2

of the OH product measured in several investigations of the NH + NO reaction. 2

ε
" φ

ο
α)

0.5 1.5
1000/T 1
(Κ )
Figure 4.4: Experimentally measured rate coefficients for Η + N 0 -> Products. 2
128 129 13 131 132 133 134
Ο - ref. , • - ref. , • - ref. °, · - ref. , Δ - ref. , • - ref. , V - ref. .

4.3. Η + N 0 -> Products

2

As mentioned above, the HNNO potential energy surface can also be accessed
from Η + N 0 reactants. This reaction has two channels: attack at the O-atom,
2

which results in direct abstraction of the O, forming N + OH products; and attack

2

at the terminal N-atom, which forms the HNNO intermediate before proceeding
94

on to products. The reaction has been studied by a number of groups during the
135,136
last 30 y e a r s , 128-134,137,138 j there is general agreement between the
a n (

studies above ~700 K, as is evident in Figure 4.4. If we fit all of the direct
determinations above 700K (with the exception of the somewhat low results of
131 -10
Glass and Q u y we obtain a consensus rate of k4. = 5.3 χ Ι Ο exp(-18.1 2
3
kcal/mole/RT) cm /molecule/s.
We can use this activation energy to calculate a barrier height for each channel,
139
assuming that it is the dominant channel, by first recognizing t h a t
E = E + RT + <ACp>T.
a 0

We further recognize that

+
<AC >T = [E(T)-E(0)] ransition state - [E(T)-E(0))] tants.
t reac

Calculation of the internal energy at temperature Τ for the reactants and

transition state is straightforward, given knowledge of the frequencies of the
140
species involved. The frequencies of the transition state for OH elimination
106 107
from HNNO and for direct attack on the O-atom have been calculated using
the G2Q method. The midpoint, in terms of 1/T, of the interval 700 < Τ < 2850 Κ
is 1100 Κ, and we present the calculation of <AC+>T for T=1100 Κ in Table 4.3.

From the experimental activation energy of 18.1 kcal/mole, the calculated values
+
of <AC >T, and knowing that RT = 2.2 kcal/mole at 1100 Κ we obtain values of E 0

= 16.5 kcal/mole for the barrier height for NN-OH reached via the HNNO
intermediate, and Eo = 16.2 kcal/mole for the barrier height for NN-OH reached
via direct attack on the O-atom. These values are in good agreement with those
calculated by the G2Q method, Eo = 16.6 kcal/mole for the barrier height for NN-
OH reached via the HNNO intermediate, and Eo = 16.9 kcal/mole for the barrier
height for NN-OH reached via direct attack on the O-atom. This suggests that
both channels contribute to the thermal reaction.

The only measurements of the rate coefficient for reaction 4.2 below 700 are due
134,138
to Fontijn and coworkers. They found that their Arrhenius plots were
strongly curved below Τ = 700 Κ, as shown in Figure 4.5. Their observed rates
were best fit with a sum of Arrhenius rates. Their resulting rate expressions
138
were:
14 10
k(H + N 0) = 4.2 χ 10- exp(-4.55/RT) + 3.7 χ 10' exp(-16.7/RT)
2
13 10
k(D + N 0) = 3.5 χ 10- exp(-7.15/RT) + 5.3 χ 10- exp(-18.2/RT)
2

The kinetic isotope effect also appears to be the result of a sum of processes,
decreasing sharply from a value of 2.2 at 450 Κ to ~ 1.3 at -700 K, and remaining
approximately constant to 1200 K. This was taken as a clear sign of a tunneling
dominated reaction at temperatures below 700 K. The system was successfully
modeled using a BAC-MP4 surface, assuming that the low temperature behavior
corresponded to addition to form HNNO, followed by tunneling to form N + 2
134,138
OH. Limitations in the accuracy of the BAC-MP4 energies for the various
 95

Table 4.3: Calculation of <AC+>T

species E-E 0 frequency E-E 0

(kcal/mole) (cm-l) (kcal/mole)

Η
translational 3.28
total 3.28
N 0
2

translational 3.28
rotational 2.19
vibrational 3.71
2633 0.25
1392 0.77
689 1.35
689 1.35
total 9.18
NN-OH transition state via HNNO
translational 3.28
rotational 3.28
vibrational 5.31
1765 0.56
1280 0.84
958 1.10
805 1.23
471 1.58
total 11.87
NN-OH transition state via direct attack
translational 3.28
rotational 3.28
vibrational 5.56
2146 0.39
1192 0.91
836 1.20
567 1.47
475 1.58
total 12.12
<AC+>T (H + N 0 -> NN-OH via HNNO) =
2
-0.6 kcal/mole
<AC+>T (H + N 02 NN-OH via direct attack) = -0.3 kcal/mole

barriers precluded definitive assignment of the microscopic mechanism for the

high temperature reaction, although the BAC-MP4 surface places the barrier to
the direct NNOH transition state 3 kcal/mole above that for the indirect NNOH
96

transition state, which lead to a prediction that the indirect channel dominated
134
reactivity for Τ < 2000 Κ. If one instead uses the energetics calculated using
1 0 6 , 1 0 7
G2Q one would expect that the activation energies of the two processes
would be equal, and the branching at high temperature into the two channels
would be solely determined by the ratio of the Arrhenius A-factors.

0.5 1 1.5 2 2.5 3

1000/T ( K )
1

138
Figure 4.5: Arrhenius plot of rates of: Η + N 0 -» Products,
2 D+ N 0
2

138
-> Products, and fit of data above 700 Κ (see text)

4.4. Η + N 0 Dynamics Studies

2

The reaction of translationally hot Η-atoms with N 0 has been the focus of
2

considerable interest by the dynamics community, due, in part, to the ease of LIF
detection of the OH product. However, the earliest of these studies, by Oldershaw
141
and P o r t e r predated the pre-eminence of laser methods, and used mass
spectrometry and gas chromatography to measure H , N and NO products from
2 2

the photolysis of HI/N 0 mixtures. By varying the photolysis wavelength they

2

were able to map out the reaction cross section as a function of translational
energy. By extrapolating these back to zero cross sections they were able to derive
values for the reaction thresholds. They obtained values of 10 ± 6 kcal/mole for
the threshold for reaction forming OH + N , 4.2a, and 38 ± 4 kcal/mole for reaction
2

forming NH + NO, 4.2c. These values compare favorably with later

 97

determinations of the barrier for reaction 4.2a (vide supra) and the experimental
endothermicity of reaction 4.2c.

Nascent OH rotational populations from the Η + N 0 -» OH + N reaction have

2 2

been measured by a number of workers. By varying the Η atom precursor and the
photolysis wavelength they have probed the reaction from its threshold up to
several eV above threshold. The measured rotational distributions are typically
made up of two Boltzmann distributions, which has been taken as evidence for
the two channels available for reaction. Measurement of the relative lambda-
doublet populations in the OH show a preponderance of the Π(Α') state. This is
the state which is expected to arise from dissociation from a planar complex, as is
the case for the reaction channel which proceeds through HNNO.

Additional information can be gained by carrying out the reaction with a HX-N 0 2

van der Waals cluster as the reactant. As in the case of the bulk reaction, the
process is started by photolysis of the HX molecule. The Η-atom approach to the
N 0 is now dependent on the (presently unknown) geometries of the van der
2
142
Waals clusters. Ohoyami et a l . carried out an early series of experiments,
using 266 nm photolysis of HI-N 0 clusters. They found OH rotational
2

populations which consisted of two Boltzmann distributions, with temperatures

of 120 Κ and 4000 K, in bulk samples and in beams with low stagnation
pressures. A third distribution, with a temperature of 1500 K, was found to grow
in at higher stagnation pressures. They found that the relative population of the
120 Κ component increased with increasing stagnation pressure, which they took
as evidence that it corresponded to a reaction channel associated with a more
stable van der Waals cluster geometry. There is general consensus that the more
stable geometry is the one with the Η-atom located on the Ο end of the N 0 2

molecule. The low rotational temperature is taken as evidence of the reactive

collisions only having small impact parameters for the O-atom attack. They also
observed the relative lambda-doublet population, Π(Α')/Π(Α"), as a function of
stagnation pressure, and saw that it declined from a value of ~2 in the bulk
sample to ~1 at stagnation pressures of 2.5 atm. They take this as evidence that
the reaction through the direct attack on the O-atom does not preferentially
produce one of the lambda doublets. This would be expected for a reaction which
is not constrained to take place in a plane; presumably the Η-atom can attack the
O-atom to leave the unpaired electron either in the plane or out of the plane of the
OH rotation. Finally, they observed the ratio of the Πχ/ to Π3/ spin states as
2
2
2
2

a function of rotational level. It was found to be near unity for most rotational
143
lines, consistent with results observed by others in bulk samples. However,
they find a preponderance of the Π3/ level for very low rotational quantum
2
2

numbers. They attribute this to angular momentum coupling between the

nascent OH and the I-atom. They theorize that this is more important in the case
98

of the direct O-atom attack because the I-atom would not have had time to move
away from the OH, whereas it would have had more opportunity to move away in
the case of the indirect reaction as it proceeded through the HNNO intermediate.
Wittig and coworkers have also carried out a number of studies on the Η + N 0 2
144 1 4 9
system, both in bulk samples and in van der Waals clusters. They find that
both OH and NH are formed in these reactions, with the branching ratio,
k4.2cZk4.2a approaching 0.5 for highly exothermic systems. 144
This ratio was
found to be environment sensitive, decreasing in clusters of HI-N2O relative to
bulk samples. This behavior is taken to indicate an entrance channel specificity,
i.e. that attack on the Ν end of N 2 O is more likely to lead to NH product, and that
the Η-atom produced by photolysis of HI-N2O is more likely to be directed to the
Ο end of N 2 O . This behavior is in keeping with our expectations based on our
1 4 9

knowledge of the HNNO potential energy surface, and suppositions about the
144
geometry of the HX-N2O van der Waals clusters. Hoffman, et a l . studied the
highly exothermic reaction initiated by ArF photolysis of HBr. In bulk samples
they observed intense OH A —> X chemiluminescence, which was absent in
reactions in the clusters. The chemiluminescence was also present, at much lower
levels, in reactions initiated by photolysis of HI at 240 < 1 < 260 nm, but has not
149
been observed in reactions with hot D-atoms. This is taken as evidence for the
non-adiabaticity of the reaction forming OH (A).

4.5. NH + NO Dynamics Studies

The OH product channel in the reaction of NH with NO has been also been
150
examined in experiments of Patel-Misra and Dagdigian. They studied the
reaction in a crossed beam apparatus, using ArF photolysis of N H 3 as their NH
source. They observed the nascent OH product from the reaction using pulsed
LIF, allowing them to measure the rovibronic distribution of the OH product.
They found OH rotational distributions which could be fit, at high J, to
temperatures of ~4400 K, with enhanced populations of low J levels. Their
143
observations are reminiscent of those of Hollingsworth, et a l . and Hoffmann, et
aj 144,145 p d t s from the Η + N 2 O reaction (vide supra). They also observed
o n r o u c

the preferential population of the Π(Α'), pointing to a planar transition state. One
point of divergence with the observations from the Η + N 2 O reaction is Patel-
2
Misra and Dagdigian's observation of preferential formation of the Π3/2 levels
for high J levels. The relative populations of the spin-orbit levels is determined
as the system is evolving from the polyatomic transition state, where the spin-
orbit splitting is quenched, to the diatomic OH product which is spin-orbit split.
However, what these observations are telling us is still uncertain.
 99

5. THE H N 0 SYSTEM
2 2

We will next turn our attention to the reaction of NH with NO, which accesses 2

the *A surface of H 2 N 2 O . There are several proposed channels for this reaction:
NH + NO -> N + H 0
2 2 2Δ Η = -124.4 ± 0.3 kcaVmole
Γχη (5.1a)
NH + NO -> N H + OH
2 2 Δ Η = 3 ± 2 kcal/mole
Γχη (5. lb)
NH + NO->N + H + OH Δ Η = -6.5 ± 0.4 kcal/mole
2 2 Γχη (5.1c)
NH + NO -> H + N 0
2 2 2Δ Η = -46.9 ± 0.3 kcaymole
Γχη (5.Id).
151
(We have used the Gaussian 2 heat of formation for N H of 61 ± 2 kcal/mole. ) 2

The reverse reactions -5.1a to -5.Id are mechanistically improbable, and will not
be considered further.
Recent years have seen the introduction of various DeNOx schemes which involve
injection of NHx containing species into exhaust gas from a combustion
1 4
process. " If the temperature is in the appropriate range there will be a
conversion of NO to N . The reaction chiefly responsible for this reduction is
2

reaction 5.1a. Modeling of the complex reaction system suggests that reactions
5.1b and/or 5.1c are also important as radical sources, allowing the DeNOx
chemistry to be self-sustaining. The obvious commercial impact of such processes
has resulted in a great deal of study of this reaction in the last twenty years.

5.1. The H NNO Potential Energy Surface

2

Portions of the potential energy surface for the NH + NO reaction have been
2

calculated by a number of workers in the last fifteen years. Calculated energies

for a number of stationary points important in reaction 5.1 are listed in Table 5.1.
152
Casewit and Goddard used Generalized Valence Bond (GVB) calculations to
evaluate energies of various isomers of H N 0 , including species such as
2 2
108
NN(H)OH and HNN(0)H. Melius and Binkley used BAC-MP4 to characterize
not only the isomers, but also the transition states linking them. Their
calculated transition state energies for the various 1,2 Η-atom transfers were all
energetically inaccessible to NH + NO reactants. As a result, NN(H)OH and
2

HNN(0)H species are not considered important intermediates in reaction 5.1.

Abou-Rachid and coworkers used SCF and CI calculations with the very small 4-
153,154
31G basis set to examine stationary points along a proposed reaction p a t h .
Harrison, Maclagan and Whyte carried out a more extensive study, performing
MP4SDQ/6-31G* calculations on the thermally accessible stationary points on
155
the NH + NO —» Products potential energy surface. However, neither their
2

study, nor the earlier work of Abou-Rachid and coworkers was of sufficiently high
level of theory to quantitatively capture features of the surface. Recently, Walch
has used CASSCF/ICCI with large basis sets to characterize a number of points
156
on the potential energy surface. His calculations agree well with the more
157
extensive study of Wolf and Durant, using the Gaussian-2 method, as proposed
17
by Pople, and extended for transition states by Durant and Rohlfing. The
convergence of the theoretical methods can be seen by examining the agreement
100

between these last two studies, as listed in Table 5.1, and by comparing the
overall reaction exothermicities calculated using G2 with the experimental
values. Overall the agreement is quite satisfactory, with errors generally <2
157
kcal/mole. The potential energy surface of Wolf and Durant is shown in Figure
5.1.
Table 5.1: Energetics of HaNNO Species
155
reference Melius 108
Harrison Yang 158

152 156
Casewit Abou- Walch experiment
153
Rachid
I -29.8 -48.1 -39.9 -37.0 -44.0 -45.8
Π -32.0 -47.5 -36.2 -45.7 -45.6
m -31.2 -45.5 -38.3 .34.4 -44.5 -45.1
IV -24.0 -41.1 -29.2 -39.2
V -32.4 -47.2 -36.1 -45.5 -45. 4
VI -19.7 0.5 -2.8 -14.4 -14.0
b
(-14.4 )
b
VII -26.2 -36.7
VIII 7.8 -6.8
b
(-7.2 )
b
Κ -25.9 -36.2
X -7.0 6.5 -7.4 -9.8
b
(-10.3 )
XI -25.2 -7.6 -21.0 -23.4
b
(-24.6 )
N -H ts
2 8.7 11.5 10.0
b
+ OH (9.9 )
NH+ 2 2.4 -3.5 6.3 4.5
OH
N +H+
2 -8.8 -14.7 -13.0 -4.7 -6.5
OH
N + 2 -134.2 -113.1 -113.2 -122.8 -124.4
H9O
a) All energies are in kcal/mole, relative to NH + NO, 0 K, including zero
2

point energies. Energies are evaluated at G2 level for equilibrium species and
G2Q level for transition states, unless otherwise noted.
b) Energy is evaluated at G2 level for transition state
The qualitative mechanism emerging from these calculations starts with the
barrierless association of NH with NO to form a highly energized H NNO
2 2

adduct. This adduct can then undergo a 1,3 Η-atom migration, yielding an
HNNOH species, again with considerable excess internal energy. This species,
which has a trans conformation about the NN bond and a cis conformation about
the NO bond, must isomerize to a cis conformation about the NN bond and a
trans conformation about the NO bond before it can undergo a second 1,3 H-atom
migration, yielding N + H 0 products. These isomerizations can take place in
2 2
 101

either order. Isomerization about the NO bond is effected by simple torsional

rotation, while isomerization about the NN bond occurs by passage through a
linear HNN transition state. The transition state for formation of Ν2 + H 2 O
products has a non-planar geometry, with a very flat bending potential. All of the
HNNOH species are connected to N 2 H + OH products by a simple bond fission,
which is slightly endothermic relative to the NH2 + NO reactants. The N 2 H
species, and the transition state linking it with N2 + H, have been characterized

ΝΗ + 0 Η '
2 ι

ιΝΗο + ΝΟ

Νο + Η + ΟΗ
φ VI
ο
Ε

1 "25

Ο )

φ
C
LU

-50
y-\ π
/"Ν

I
N + H 0L.
2 2

-125 1

Figure 5.1: Η2ΝΝΟ Potential Energy Surface

151,159 161
by a number of high level ab initio calculations. " The small barrier to
dissociation (5.5 - 8.5 kcal/mole) suggests that N2H will have a very short lifetime
with respect to dissociation. Indeed, calculations by Koizumi, Schatz and
Walch predict that only the lowest two N 2 H states will have lifetimes longer
162

than 1 ps. Lifetimes for the (000) and (001) states were calculated to be 3 ns and
0.2 ns, respectively. The N 2 H radical has been produced in a series of neutralized
ion beam experiments in which N H was formed by neutralization of a N 2 H
2
+

beam by metal vapor. They established an upper limit on the N 2 H lifetime in

163

their apparatus of <0.5 μβ, based on their inability to detect undissociated N 2 H

from the neutralization of the ion beam. We therefore expect that 5.1b will be
rapidly followed by 5.2,
N H -> N + Η
2 2 (5.2)
with 5.1c as the net reaction. Mechanistically, channel 5.1c will be entropically
disfavored, since it requires simultaneous fission of two bonds, and we expect
that most of the OH-formation will proceed through channel 5.1b.
102

5.2. The N H 2 + NO -> Products Reaction Rate Coefficient

There has been a flurry of experimental studies of the reaction rate coefficient in
5 94 164 166 157 167 176
the last 25 years, utilizing discharge-flow, ' ' - flash photolysis, ' -
177-179 118
pulsed r a d i o l y s i s , and shock tube techniques. The results of these
various studies are plotted in Figure 5.2. One of the more provocative features of
this plot is the factor of ~2 difference between the room temperature discharge
flow rate coefficients and rate coefficients measured using other techniques. This
discrepancy has been noted in the past, but never satisfactorily explained.
Although there are fewer studies, similar behavior is apparent in the reaction of
N H 2 with N O 2 , where the flow tube results of Hack, et al. are approximately a 5

173 179 182

factor of 2 slower than the results of other workers. ' '

- H . g L 1 1 1 I 1 1 1 I 1 1 1 I 1 1 1 I 1 1 • "I

2.2
2.6 2.8 3 2.4 3.2
logT(K)
Figure 5.2: Temperature Dependent Studies of the N H 2 + NO —> Products
Reaction. • - ref. 167, A - ref. 5, Τ - ref. 183, • - ref. 172, Δ - ref. 184, · - ref. 175, •
- ref. 176, Ο - ref. 157.
The wide variety of N H 2 sources and detection methods utilized in measuring the
reaction rate coefficient k s . i makes explaining the discrepancy in rate coefficients
difficult. Discharge-flow experiments have used reaction of OH-radicals with
1 6 6
NH3, 9
reaction of F-atoms with N H 3 , ' and SF -sensitized IRMPD of N H
4 5 1 6 5
6 2 4

as N H 2 sources, while other techniques have used flash lamp photolysis of

NH i67-i70,i72,i74
3 photolysis of N H 3 ' ' ' '
l a s e r or pulsed radiolysis of 1 5 7 1 7 1 1 7 3 1 7 5 1 7 6

ΝΗ3. 1 7 7 - 1 7 9
Detection schemes include mass spectrometry, ' resonance 5 94

172 157 164 166 168 171 173 175

fluorescence, laser-induced fluorescence, ' ' ' ' ' ' long-path laser
167 169 170 174 176
absorption, intracavity laser absorption ' ' and cavity-ring down.
 103

118
The one shock tube study, by Roose, Hanson and Kruger, consisted of shock
heating mixtures of NO, NH3, H and Ar. NH concentrations were measured by
2 2

observing emission from electronically excited NH at 538 nm. NH3, NH, N 0 2 2

and NO were also monitored by emission, and NO by IR laser absorption. Their

precision was limited by uncertainties in calibrating their detection systems; as a
result, their data is consistent with a temperature independent rate, although
they reported a positive temperature dependence for reaction ks.i. They also
observed N 0 products, although again their error bars were large enough to
2

preclude any definitive statements about the importance of the channel. Despite
its lack of precision, the work of Roose, et al. is consistent with a simple
extrapolation of lower temperature results.
A number of investigators have also examined the effect of pressure on the NH + 2

NO —> Products reaction. They all found that the reaction is pressure
5 9 4 1 5 7 1 6 5 1 6 7 1 6 9 1 7 3 1 7 4 1 7 7 1 8 5 1 8 6
independent for pressures below 2 t m . ' ' ' - ' ' ' . ' '
a
177
Particularly noteworthy are studies by Gordon and Mulac, who measured ks.i
in NH3 for pressures from 250 to 1520 Torr, Lesclaux, et al., who used N as 167
2
185
the bath gas in the pressure range 2 < Ρ < 700 Torr and Sarkisov, et a l . , who
also used N in the pressure range 1 < Ρ < 700 Torr. These measurements are in
2

keeping with our expectations for an association-type reaction with open product
channels below the energy of the reactants.

5.3. Products from the NH + NO -» Products Reaction

2

Experimental interest in the products of the reaction of NH with NO has a long 2

history. In 1939 Bamford photolyzed NH3 in the presence of NO at 373 Κ in a

187

series of stationary photolysis experiments . He identified H 0 and N as 2 2

photolysis products, and proposed a mechanism in which reaction 5.1a was the
188
loss channel for NH . Later work by Serewicz and Noyes in the vicinity of
2

room temperature also identified N 0 as a product of the photolysis; presumably

2

it was destroyed by mercury-sensitized photolysis in Bamford's work, since he

189
made extensive use of mercury seals in his vacuum apparatus. Srinivasan
used N H 3 to establish that the N 0 came solely from NO, while the N
1 5
2 2

incorporated one N-atom from ammonia and one from nitric oxide, thus ruling out
reaction 5.Id as a significant process. The N 0 was assumed to be formed by the
2

reaction of two HNO radicals, producing N 0 and H 0 . The final "classical"

2 2
190
photochemical study was that of Jayanty, Simonaltis and Heicklen whose
actinometric measurements established that the quantum yield for formation of
N as 1.0 ± 0.1 at 298 K, making N + H 0 the major channel in the reaction of
2 2 2

NH with NO at room temperature.

2

The reaction of NH with NO presumably proceeds through formation of a

2

vibrationally excited H NNO adduct. This adduct can be expected to have a very
2

short lifetime with respect to isomerization to HNNOH; RRKM calculations by

191
Phillips suggested that the lifetime was of the order of 10 ps. This result is
consistent with the lack of any observed pressure dependence of the reaction rate
104

94
coefficient. Gehring, et al. observed m/e = 45 and m/e = 46 peaks in the mass
spectrum of the products of the reaction of N H 2 with NO. Similarly, Farber and
192
H a r r i s observed m/e = 45 and m/e = 46 mass peaks from the products of the
reaction of N H 3 with NO over a vanadium oxide catalyst. These observations
would imply that the H 2 N N O adduct lifetime was orders of magnitude larger
than 10 ps. However, neither group performed isotopic substitution experiments
+
to confirm their assignments of m/e = 46 to H 2 N N O and m/e = 45 to HNNO .
+

193 1 +
Foner and Hudson observed ΝβΗβ"" and N4H4 in the products of electrodeless
discharge dissociated hydrazine which had been condensed in a liquid nitrogen
trap; presumably other hydrides of N 3 and N 4 are present in the gas phase
following condensation reactions of NHx species, and could have given rise to the
peaks observed by Gehring, et al. and Farber and Harris.
Table 5.2: Room Temperature Product Branching Fractions for N H 2 + NO ->
Products Reaction
Φ(Ν . + Η .0)
9 9
Φ(ΟΗ) ΦίΗ) Φ(ΗΝΟ) reference
190
1.0 ±0.1 Jayanty, et a l .
0.4 ±0.1 Silver and Kolb 165

(<0.33) (Silver and

194
Kolb )
171
>0.65 <0.05 Andresen, et a l .
172
<0.22 Stief, et a l .
195
0.13 ±0.02 Hall, et a l .
196
<0.13 Dolson
174
0.10 ±0.02 Bulatov, et a l .
175
0.10 ±0.025 Atakan, et a l .
<0.01 Unfried, et a l . 197

179
0.07 Pagsberg, et a l .
198
0.85 0.10 Stephens, et a l .
0.09 ± 0.02 Yang and
158
Durant
One might expect to be able to stabilize the H 2 N N O intermediate using matrix
199
isolation techniques, and Crowley and Sodeau observed IR absorption features
which they attributed to the H 2 N N O intermediate formed by photolysis of
N H 3 / N O or NH3/NOCI mixtures in Ar or N 2 matrices. The two sets of bands
observed are in reasonable agreement with predictions based on scaled HF/6-
200 1 5
31G* calculations, both in their location and in their response to N or D
201
isotopic substitution. Interestingly enough, their experiments suggest that
N 2 O + H2 are the primary products of the N H 2 + NO reaction at 4.2 K, with the
yield of H2O + N2 being considerably surpressed. The barrier for reaction 5.Id
108
has been calculated by Melius and Binkley to be 32 kcal/mole above the
reactants, making it thermally inaccessible. However, quantum mechanical
tunneling may be responsible for the reaction at 4.2 K.
 105

Because of its important role in various DeNOx processes, the product branching
fraction for the NH + NO reaction has been the focus of a number of direct
2

determinations. Table 5.2 summarizes room temperature measurements, while

temperature dependent product branching fractions are presented in Figure 5.3.

0.5 "Τ . ι ι ι

0.4

U
0.3
cd

0.2

Ο
0.1
EC
i
Ο
L -I I I I L_
0 •— -
200 400 600 800 1000 1200
Temperature

Figure 5.3: Temperature dependence of (ks.ib ks.lcVkB.l. · - ref. 175, • - ref.

+

174, • - ref. 198.

One of the interesting conclusions drawn from modeling DeNOx schemes is that
reactions 5.1b and/or 5.1c must have non-negligible rates at temperatures in
7
excess of 1000 K. This is a result of the requirement that the overall DeNOx
process be self-sustaining. Reaction 5.1a is chain terminating, removing one
radical center, NH . By contrast, reactions 5.1b and 5.1c are strongly chain
2

branching. They directly produce two radicals for every one consumed. Under
conditions encountered in most DeNOx schemes (excess 0 and H 0) the H-atom 2 2

reacts rapidly with 0 , producing another OH-radical and an O-atom. The O-

2

atom is rapidly converted to two more OH-radicals by reaction with H 0 , 2

resulting in production of four radical centers for every one consumed.

NH + NO -> N H + OH -> N + Η + OH
2 2 2

Η + 0 -> OH + Ο
2

Ο + H 0 -» 20H
2
106

net: NH + NO + 0 + H 0 -> N + 40H

2 2 2 2

In the absence of other important chain branching or chain terminating reactions

the models require (ks.ib + ks.icVks.i > 0.25 for the DeNOx process to be self-
sustaining. Of course, the models may well be wrong; extrapolation of lower
temperature data on the product branching fraction is presently too uncertain to
say definitively.

6. CONCLUSION
At this point we conclude our brief survey of NHx chemistry. It is hoped that this
examination of a facet of nitrogen chemistry has been useful. We have tried to
demonstrate the state of experimental studies of reaction rate coefficients and
products, and of theoretical studies of structures and energies of stable and
transient species encountered in these reactions. The combination of theoretical
and experimental studies has allowed us to learn things about these systems
that would not be possible without the integration of both approaches. It should
also be apparent that the richness of these systems leaves much still left to
learn. Questions range from questions about reaction rate coefficients and
products to questions about energy disposal in these reactions and the affect of
multiple electronic surfaces on the reaction mechanisms.

ACKNOWLEDGEMENTS
The author would like to acknowledge the support of the U.S. Department of
Energy, Office of Basic Energy Sciences, Division of Chemical Sciences.

REFERENCES
(1) Lyon, R. K. Int. J. Chem. Kinetics 1976, 8, 315.

(2) Muzio, L. J.; Arand, J. K.; Teixeira, D. P. Symp. (Int.) Combust. [Proc] 1977,
16, 1977.
(3) Perry, R. Α.; Siebers, D. L. Nature 1986, 324, 657.
(4) Lyon, R. K. Environ. Sci. Technol. 1987, 27, 231.
(5) Hack, W.; Schacke, H.; Schroter, M.; Wagner, H. G. Symp. (Int.) Combust.
[Proc] 1979,77, 505.
(6) Lesclaux, R. Reviews of Chemical Intermediates 1984, 5, 347.
(7) Miller, J. Α.; Bowman, C. T. Prog. Energy Combust. Sci. 1989, 75, 287.
(8) Hanson, R. K.; Salimian, S. In Combustion Chemistry-, J. W.C. Gardiner, Ed.;
Springer-Verlag: NY, 1984.
 107

(9) M a l l a r d , W . G.; Westley, R ; Herron, J. T . ; H a m p s o n , R. F . ; Frizzell, D . H .

N I S T Standard Reference Data Database 17: Chemical Kinetics: Gaithersburg, M D ,
1993.

(10) B a u l c h , D . L . ; C o b o s , C . J.; C o x , R. Α.; Esser, C ; Frank, P . ; Just, T . ; Kerr, J.

Α.; Pilling, M . J.; T r o e , J.; Walker, R. W . ; W a r n a t z , J. J. Phys. Chem. Ref. Data
1992, 2 7 , 4 1 1 .
(11) C o h e n , N . ; W e s t b e r g , K. R. J. Phys. Chem. Ref. Data 1991, 2 0 , 1211.

(12) T s a n g , W . ; Herron, J. T . J. Phys. Chem. Ref Data 1991,20, 6 0 9 .

(13) T s a n g , W . ; Herron, J. T . / . Phys. Chem. Ref Data 1992, 2 7 , 7 5 3 .

(14) Atkinson, R.; Baulch, D . L.; Cox, R. Α.; H a m p s o n , R. F., Jr.; Kerr, J. Α . ; T r o e ,
J. / . Phys. Chem. Ref Data 1992, 2 7 , 1125.

(15) D e M o r e , W . B . ; Sander, S. P . ; Golden, D . M . ; Molina, M . J.; H a m p s o n , R. F . ;

K u r y l o , M . J.; H o w a r d , C. J.; Ravishankara, A . R. In JPL Publication 90-1,
Pasadena, CA. 1992; p p 1.

(16) Curtiss, L . Α . ; Raghavachari, K.; Trucks, G . W . ; Pople, J. A . / . Chem. Phys.

1991,94, 7 2 2 1 .

(17) Durant, J. L.; Rohlfing, C. M . J. Chem. Phys. 1993, 98, 8 0 3 1 .

(18) Bauschlicher, C . W . ; Langhoff, S. R. Sci. 1991, 254, 3 9 4 .

(19) C h a s e , M . W . ; Davies, C. Α . ; D o w n e y , J. R.; Frurip, D . J.; M c D o n a l d , R . Α . ;

Syverud, A . N . J. Phys. Chem. Ref. Data 1985,14 (SuppL No. 1),

(20) A n d e r s o n , W . R. / . Phys. Chem. 1989, 93, 5 3 0 .

(21) B e r k o w i t z , J.; Ellison, G. B . ; G u t m a n , D . 7. Phys. Chem. 1994, 98, 2 7 4 4 .

(22) D i x o n , R. N . ; J o n e s , Κ. B . ; Noble, M . ; Carter, S. Molecular Physics 1981,42,

455.

(23) Huber, K. P . ; Herzberg, G . Molecular Spectra and Molecular Structure Constants

of Diatomic Molecules; Van Nostrand Reinhold C o m p a n y : N e w York, 1979.

(24) Ogilvie, J. F . Nature 1973, 243, 2 1 0 .

(25) Ellis, Η . B . ; Ellison, G. B . J. Chem. Phys. 1983, 78, 6 5 4 1 .

(26) Bancroft, J. L.; Hollas, J. M . ; R a m s a y , D . A . Canadian Journal of Physics 1962,

40, 3 2 2 .

(27) G u a d a g n i n i , R.; Schatz, G. C ; Walch, S. P . J. Chem. Phys. 1994, submitted,

(28) B r u n a , P . J.; Marian, C. M . Chem. Phys. Lett. 1979, 67, 109.

108

(29) Heiberg, Α.; Almlof, J. Chem. Phys. Lett. 1982, 85, 542.

(30) M e l i u s , C. F . ; Binkley, J. S. In The Chemistry of Combustion Processes', A. M .

D e a n ; M . - S . C h o u and D . Stern, Ed.; A C S S y m p o s i u m Series: 1 9 8 4 ; Vol. 2 4 9 .

(31) A d a m s , G. F . ; Bent, G. D . ; Barlett, R.; Purvis, G. D. In Potential Energy

Surfaces and Dynamics Calculations', D. G. Truhlar, Ed.; Plenum: N e w York, 1 9 8 1 ;
p p 133.

(32) N o m u r a , O.; Ikuta, S.; Igawa, A. In Applied Quantum Chemistry-, V. H. Smith Jr.;
H. F . Schaefer m and K. M o r o k u m a , Ed.; Reidel: Dordrecht, 1984; p p 2 4 3 .

(33) W a l c h , J. P.; Rohlfing, C. M . J. Chem. Phys. 1989, 91, 2939.

(34) W a l c h , S. P. J. Chem. Phys. 1990, 93, 8036.

(35) Pauzat, F.; Ellinger, Y.; Berthier, G.; G r i n , M . ; Viala, Y. Chem. Phys. 1993,
174, 7 1 .

(36) L e e , T. J. J. Chem. Phys. 1993, 9 9 , 9 7 8 3 .

(37) Carter, S.; Mills, I. M . ; Murrell, J. N . J. Chem. Soc. Faraday Trans. 2 1979,1,
148.

(38) Colton, M . C ; Schatz, G. C. J. Chem. Phys. 1985, 83, 4 8 2 3 .

(39) H u a n g , Y.-L.; D a g d i g i a n , P. J. J. Chem. Phys. 1992, 9 7 , 180.

(40) T e m p s , F . Ber. MPI Stromungsforschung 1983, 4, 140.

(41) A d a m s o n , J. D . ; Farhat, S. K.; Morter, C. L.; Glass, G. P.; Curl, R. F.; Phillips,
L. F . J. Phys. Chem. 1994, 98, 5665.

(42) Hack, W . ; W a g n e r , H. G.; Zasypkin, A. Ber. Bunsenges. Phys. Chem. 1994, 98,
156.

(43) Mertens, J. D . ; C h a n g , A . Y.; Hanson, R. K.; B o w m a n , C. T. Int. J. Chem. Kinet.

1991, 23, 173.

(44) Zeldovich, J. Acta Physiocochim. URSS. 1946, 21, 577.

(45) H o w a r d , M . J.; Smith, I. W . M . Chem. Phys. Lett. 1980, 69, 4 0 .

(46) H o w a r d , M . J.; Smith, I. W . M. J. Chem. Soc. Faraday Trans. 2 1981, 77, 9 9 7 .

(47) B r u n e , W . H.; S c h w a b , J. J.; Anderson, J. G. J. Phys. Chem. 1983, 87, 4 5 0 3 .

(48) R o w e , B . R.; S i m s , I. R.; Smith, I. W . M., private communication, 1994.

(49) Bradley, J. N . ; C r a g g s , P. Proceedings of the International Combustion

Symposium 1975, 7 5 , 8 3 3 .
 109

(50) D u x b u r y , J.; Pratt, Ν . H. Proceedings of the International Combustion

Symposium 1 9 7 5 , 7 5 , 8 4 3 .

(51) F l o w e r , W . L.; H a n s o n , R. K.; Kruger, C. H. Proceedings of the International

Combustion Symposium 1975, 7 5 , 8 2 3 .

(52) Koshi, M . ; A n d o , H.; Oya, M ; Asaba, T. Proceedings of the International

Combustion Symposium 1975, 7 5 , 9 0 9 .

(53) A n d o , H.; A s a b a , T. Int. J. Chem. Kinetics 1976, 8, 259.

(54) Flower, W . L.; H a n s o n , R. K.; Kruger, C. H. Combustion Science and

Technology 1977, 7 5 , 115.

(55) M c C u l l o u g h , R. W . ; Kruger, C. H.; Hanson, R. K. Combustion Science and

Technology 1977, 7 5 , 2 1 3 .

(56) S m a l l w o o d , Η. M . J. Am. Chem. Soc. 1929, 5 7 , 1985.

(57) Harteck, P. Ber. 1 9 3 3 , 66, 4 2 3 .

(58) C a s h i o n , J. K.; Polanyi, J. C. J. Chem. Phys. 1959, 30, 317.

(59) C l e m e n t , M . J. Y.; R a m s a y , D. A. Canadian Journal of Physics 1961, 39, 2 0 5 .

(60) Dalby, F . W . Canadian Journal of Physics 1958, 36, 1336.

(61) D i x o n , R. N . ; Rosser, C. A. Chem. Phys. Lett. 1 9 8 4 , 1 0 8 , 3 2 3 .

(62) Petersen, J. C. J. Mol Spect. 1 9 8 5 , 1 1 0 , 2 7 7 .

(63) D i x o n , R. N . ; N o b l e , M.; Taylor, C. Α.; D e l h o u m e , M . Discussions of the

Faraday Society 1 9 8 1 , 77, 1.

(64) B r u n a , P. J. Chem. Phys. 1 9 8 0 , 49, 39.

(65) C l y n e , Μ . Α. Α.; Thrush, B . A. Discussions of the Faraday Society 1962, 33,

139.

(66) Hikida, T.; E y r e , J. Α.; Dorfman, L. M . J. Chem. Phys. 1971, 54, 3422.

(67) A h u m a d a , J. J.; Michael, J. V.; Osborne, D . T. J. Chem. Phys. 1972, 5 7 , 3 7 3 6 .

(68) Atkinson, R.; Cvetanovic, R. J. Canadian Journal of Chemistry 1973, 5 7 , 370.

(69) Oka, K.; Singleton, D . L.; Cvetanovic, R. J. J. Chem. Phys. 1977, 67, 4 6 8 1 .

(70) O k a , K.; Singleton, D . L.; Cvetanovic, R. J. J. Chem. Phys. 1977, 66, 7 1 3 .

(71) Ishikawa, Y.; Sugawara, K.; Sato, S. Bull. Chem. Soc. Jpn. 1979, 5 2 , 3 5 0 3 .
110

(72) Clyne, Μ . Α . Α.; Thrush, Β . Α. Trans. Faraday Soc. 1962, 5 7 , 1305.

(73) Hartley, D . Β . ; Thrush, Β . A. Roy. Soc. London Proc. A 1967, 2 9 7 , 520.

(74) C a m p e l l , I. M ; H a n d y , B . J. J. Chem. Soc. Faraday Trans. 1975, 77, 2 0 9 7 .

(75) O b i , K.; M a t s u m i , Y.; Takeda, Y.; Mayama, S. Chem. Phys. Lett. 1983, 9 5 , 5 2 0 .

(76) Forte, Ε. N . J. Photochem. 1981, 7 7 , 13.

(77) Pauzat, F . ; Gritli, H.; Ellinger, Y.; Subra, R. J. Phys. Chem. 1984, 88, 4 5 8 1 .

(78) D a v i s , P . B . ; Dransfeld, P . ; T e m p s , F . ; Wagner, H. G. J. Chem. Phys. 1984, 81,

3763.

(79) M i k a m i , H . ; Saito, S.; Y a m a m o t o , S. J. Chem. Phys. 1991, 94, 3 4 1 5 .

(80) Soto, M . R.; P a g e , M . Chem. Phys. Lett. 1991,187, 3 3 5 .

(81) C a i , Z . L . Chem. Phys. 1993,169, 7 5 .

(82) B a r o n e , V . ; Grand, Α . ; Minichino, C.; Subra, R. J. Phys. Chem. 1993, 9 7 , 6 3 5 5 .

(83) W a l c h , S. P . J. Chem. Phys. 1993, 9 9 , 3804.

(84) K o m a r o m i , I.; Tronchet, J. M . J. Chem. Phys. Lett. 1993, 2 7 5 , 4 4 4 .

(85) Y a n g , D . L.; K o s z y k o w s k i , M . L.; Durant, J. L., Jr. / . Chem. Phys. 1994,101,

1361.

(86) Salotto, A . W . ; Burnelle, L. J. Chem. Phys. 1970, 53, 3 3 3 .

(87) Burton, B . ; Claxton, Τ. Α . ; Ellinger, Y. Computer Physics Communications 1979,

77, 27.

(88) B a r o n e , W . ; Cristinziano, P . L.; Lelj, F.; Pastore, Α.; Russo, N . J. Mol. Struct.
1982, 90, 59.

(89) Kysel, O.; M a c h , P . ; Haring, M . J. Mol. Struct. 1986,138, 299.

(90) Sana, M . ; Leroy, G.; Peeters, D . ; Younang, E. J. Mol. Struct. 1987, 757, 3 2 5 .

(91) S o , S. P . J. Phys. Chem. 1990, 94, 2344.

(92) P a g e , M ; Soto, M . R. J. Chem. Phys. 1993, 9 9 , 7709.

(93) Lifshitz, C . ; Ruttink, P . J. Α . ; Schaftenaar, G.; Terlouw, J. K. Rapid Commun.

Mass Spectrom. 1987, 7, 6 1 .

(94) Gehring, M . ; H o y e r m a n n , K.; Schacke, H.; Wolfrum, J. Symp. (Int.) Combust.,

[Proc] 1973,14, 99.
 Ill

(95) Dransfeld, P . ; H a c k , W . ; Kurzke, H . ; T e m p s , R ; Wagner, H. G. Symp. (Int.)

Combust. [Proc] 1985, 2 0 , 6 5 5 .

(96) Patel-Misra, D . ; D a g d i g i a n , P. J. Chem. Phys. Lett. 1991, 785,387.

(97) Patel-Misra, D . ; Sauder, D . ; Dagdigian, P. J. Chem. Phys. 1991, 9 5 , 9 5 5 .

(98) L u o , X . ; R i z z o , T . R. J. Chem. Phys. 1992, 96, 5129.

(99) Sinha, Α . ; V a n d e r W a l , R.; Crim, F . F . J. Chem. Phys. 1990, 9 2 , 4 0 1 .

(100) G a r c i a - M o r e n o , I.; L o v e j o y , E. R.; M o o r e , C. B . / . Chem. Phys. 1994,100,

8890.

(101) D o d o n o v , A . F . ; Z e l e n o v , V . V.; Tal'rose, V . L. Dokl. Phys. Chem. 1980, 2 5 2 ,

642.

(102) D o d o n o v , A . F . ; Z e l e n o v , V . V.; Strunin, V . P.; Tal'rose, V. L. Kinet. Catal

1981, 2 2 , 689.

(103) W a s h i d a , N . ; A k i m o t o , H . ; Okuda, M . / . Phys. Chem. 1978, 82, 2 2 9 3 .

(104) Harrison, J. Α . ; W h y t e , A . R.; Phillips, L. F . Chem. Phys. Lett. 1986, 7 2 9 , 3 4 6 .

(105) D e a n , A . M . ; C h o u , M . S . ; Stern, D . Int. J. Chem. Kinetics 1984,16, 6 3 3 .

(106) Durant, J. L . J. Phys. Chem. 1994, 98, 5 1 8 .

(107) Durant, J. L . unpublished results 1994.

(108) M e l i u s , C . F . ; B i n k l e y , J. S. Symp. (Int.) Combust. [Proc] 1984, 20, 5 7 5 .

(109) F u e n o , T . ; F u k u d a , M . ; Y o k o y a m a , K. Chem. Phys. 1988, 724, 2 6 5 .

(110) Harrison, J. Α . ; M a c l a g a n , R. G. A . R. J. Chem. Soc. Faraday Trans. 1990, 86,

3519.

(111) Miller, J. Α . ; M e l i u s , C . F . Symp. (Int.) Combust. [Proc] 1993, 24, 7 1 9 .

(112) W a l c h , S. P . / . Chem. Phys. 1993, 98, 1170.

(113) G o r d o n , S.; M u l a c , W . ; Nangia, P . J. Phys. Chem. 1971, 75, 2 0 8 7 .

(114) H a n s e n , I.; H o e i n g h a u s , K.; Zetzsch, C.; Stuhl, F . Chem. Phys. Lett. 1976,42,
370.

(115) C o x , J. W . ; N e l s o n , Η . H . ; M c D o n a l d , J. R. Chem. Phys. 1985, 96, 175.

(116) Y a m a s a k i , K.; O k a d a , S.; Koshi, M . ; Matsui, H . J. Chem. Phys. 1991, 9 5 , 5 0 8 7 .

(117) M o r l e y , C . Symp. (Int.) Combust. [Proc] 1981,18, 2 3 .

112

(118) R o o s e , Τ. R.; H a n s o n , R. K.; Kruger, C. H. Symp. (Int.) Combust., [Proc]

1981,18, 8 5 3 .

(119) Y o k o y a m a , K.; S a k a n e , Y.; Fueno, T. Bull. Chem. Soc Jpn. 1991, 64, 1738.

(120) V a n d o o r e n , J.; Sarkisov, Ο. M.; Balakhnin, V. P.; Van Tiggelen, P. J. Chem.

Phys. Lett. 1991,184, 2 9 4 .

(121) Sausa, R. C ; A n d e r s o n , W . R.; Dayton, D . C ; Faust, C. M.; H o w a r d , S. L.

Combust. Flame 1994, in press,

(122) Lillich, H.; Schuck, Α.; Volpp, H.; Wolfrum, J. Symp. (Int.) Combust. [Proc]
1994, 2 5 , in press.

(123) O k a d a , S.; Miyoshi, Α.; Matsui, H. J. Chem. Phys. 1994, submitted.

(124) M e r t e n s , J. D.; K o h s e - H o i n g h a u s , K.; Hanson, R. K.; B o w m a n , C. T. Int. J.

Chem. Kinet. 1991, 23, 6 5 5 .

(125) Walther, C. D.; W a g n e r , H. G. Ber. Bunsenges. Phys. Chem. 1983, 87, 4 0 3 .

(126) Hack, W . ; Kurzke, H.; Wagner, H. G. J. Chem. Soc. Faraday Trans. 2 1985, 81,
949.

(127) W a l c h , S. P.; D u c h o v i c , R. J.; Rohlfing, C. M . J. Chem. Phys. 1989, 90, 3230.

(128) Baldwin, R. R.; Gethin, Α.; Walker, R. W . Journal of the Chemical Society,
Faraday Transactions 1 1973, 69, 352.

(129) Albers, Ε. Α.; H o y e r m a n n , K.; Schacke, H.; Schmatjko, K. J.; Wagner, H . G.

Symp. (Int.) Combust. [Proc] 1975, 7 5 , 765.

(130) D e a n , A. M . ; Steiner, D . C ; W a n d e , Ε. E. Combust. Flame 1978, 32, 7 3 .

(131) Glass, G. P.; Q u y , R. B . J. Phys. Chem. 1979, 83, 30.

(132) D e a n , A. M . ; Johnson, R. L.; Steiner, D. C. Combust. Flame 1980, 37, 4 1 .

(133) Hidaka, Y.; T a k u m a , H.; Suga, M. Bull. Chem. Soc. Jpn. 1985, 58, 2 9 1 1 .

(134) Marshall, P.; Fontijn, Α.; Melius, C. F. J. Chem. Phys. 1987, 86, 5540.

(135) F e n i m o r e , C. P.; Jones, G. W . J. Phys. Chem. 1959, 63, 1154.

(136) D i x o n - L e w i s , G.; Sutton, Μ . M.; Williams, A. Symp. (Int.) Combust. [Proc]

1965,10, 4 9 5 .

(137) Balakhnine, V. P.; Vandooren, J.; Van Tiggelen, P. J. Combust. Flame 1977, 28,
165.

(138) Marshall, P.; K o , TV, Fontijn, A. J. Phys. Chem. 1989, 93, 1922.
 113

(139) B e n s o n , S. W . ; O'Neal, Η . E. Kinetic Data on Gas Phase Unimolecular

Reactions; National Bureau of Standards: Washington, 1970; Vol. 2 1 .

(140) M a y e r , J. E . ; Mayer, M . G. Statistical Mechanics; Wiley: N e w York, 1957.

(141) O l d e r s h a w , G . Α . ; Porter, D . A . Nature 1969, 223, 490.

(142) O h o y a m a , H . ; Takayangi, M . ; Nishiya, T.; Hanazaki, I. Chem. Phys. Lett. 1989,

162, 1.

(143) H o l l i n g s w o r t h , W . E.; Subbiah, J.; Flynn, G . W . ; Weston, R. E. J. Chem. Phys.

1985, 82, 2 2 9 5 .

(144) Hoffmann, G.; O h , D . ; l a m s , H.; Wittig, C . Chem. Phys. Lett. 1989, 7 5 5 , 3 5 6 .

(145) Hoffmann, G.; O h , D . ; Wittig, C . J. Chem. Soc. Faraday Trans. 2 1989, 85,
1141.

(146) Hoffmann, G . ; O h , D . ; Chen, Y . ; Engel, Υ . M . ; Wittig, C. Isr. J. Chem. 1990, 30,

115.

(147) S h i n , S. K.; C h e n , Y.; Nickolaisen, S.; Sharpe, S. W . ; Beaudet, R. Α . ; Wittig, C .

Adv. Photochem. 1991,16, 2 4 9 .

(148) Shin, S. K.; C h e n , Y.; O h , D . ; Wittig, C. Philos. Trans. R. Soc. London Ser. (A)
1990, 332, 361.

(149) B o e h m e r , E . ; Shin, S. K.; Chen, Y.; Wittig, C . J. Chem. Phys. 1992, 9 7 , 2 5 3 6 .

(150) Patel-Misra, D . ; Dagdigian, P. J. J. Phys. Chem. 1992, 96, 3232.

(151) P o p l e , J. Α . ; Curtiss, L. A . J. Chem. Phys. 1991, 9 5 , 4 3 8 5 .

(152) C a s e w i t , C . J.; Goddard, W . Α., ΙΠ J. Am. Chem. Soc. 1982,104, 3 2 8 0 .

(153) A b o u - R a c h i d , H . ; Pouchan, C ; Chaillet, M . Chem. Phys. 1984, 90, 2 4 3 .

(154) A b o u - R a c h i d , H . ; Pouchan, C. / . Mol. Struct. 1985, 727, 2 9 9 .

(155) Harrison, J. Α . ; Maclagan, G. A . R.; W h y t e , A . R. J. Phys. Chem. 1987, 9 7 ,

6683.

(156) W a l c h , S. P . J. Chem. Phys. 1993, 9 9 , 5 2 9 5 .

(157) Wolf, M . ; Durant, J. L. unpublished results 1994,

(158) Y a n g , D . L . ; Durant, J. L . unpublished results 1994,

(159) Curtiss, L . Α . ; D r a p c h o , D . L.; Pople, J. A. Chem. Phys. Lett. 1984,103, 4 3 7 .

(160) W a l c h , S. P . ; D u c h o v i c , R. J.; Rohlfing, C. M . J. Chem. Phys. 1989, 90, 3230.

114

(161) W a l c h , S. P . J. Chem. Phys. 1990, 93, 2 3 8 4 .

(162) K o i z u m i , H . ; Schatz, G . C ; W a l c h , S. P . J. Chem. Phys. 1991, 9 5 , 4 1 3 0 .

(163) Selgren, S. F . ; M c L o u g h l i n , P. W . ; Gellene, G. I. J. Chem. Phys. 1989, 90,

1624.

(164) Gersh, Μ . E . ; Silver, J. Α.; Zahniser, M . S.; Kolb, C. E.; B r o w n , R. G.;

G o z e w s k i , C. M . ; Kallelis, S.; W o r m h o u d t , J. C . Rev. Sci. Instrum. 1981, 5 2 , 1 2 1 3 .

(165) Silver, J. Α . ; Kolb, C . E . J. Phys. Chem. 1982, 86, 3240.

(166) Jeffries, J. B . ; M c C a u l l e y , J. Α.; Kaufman, F . Chem. Phys. Lett. 1984,106, 1 1 1 .

(167) Lesclaux, R.; K h e , P . V . ; Dezauzier, P.; Soulignac, J. C. Chem. Phys. Lett. 1975,
35, 4 9 3 .

(168) Hancock, G.; L a n g e , W . ; Lenzi, M ; W e l g e , Κ. H. Chem. Phys. Lett. 1975, 33,

168.

(169) Sarkisov, O . ML; Cheskis, S. G.; Sviridenkov, E. A. Bull. Acad. Sci. USSR, Div.
Chem. Sci. 1978, 2 7 , 2 6 1 2 .

(170) N a d t o c h e n k o , V . L.; Sarkisov, Ο. M . ; Cheskis, S. G. Fiz.-Khim. Protsessy

Gazov. Kondens. Fazakh 1979, 2 1 .

(171) Andresen, P . ; Jacobs, Α.; Kleinermanns, C ; Wolfrum, J. Symp. (Int.) Combust.

[Proc] 1982, 7 9 , 1 1 .

(172) Stief, L . J.; Brobst, W . D . ; Nava, D . R ; B o r k o w s k i , R. P.; Michael, J. V . J.

Chem. Soc. Faraday Trans. 2 1982, 78, 1391.

(173) W h y t e , A . R.; Phillips, L . F . Chem. Phys. Lett. 1983,102, 4 5 1 .

(174) Bulatov, V . P . ; Ioffe, Α . Α.; Lozovsky, V. Α . ; Sarkisov, Ο. M . Chem. Phys. Lett.

1989,161,141-146,

(175) Atakan, B . ; J a c o b s , Α.; W a h l , M . ; Weller, R.; Wolfrum, J. Chem. Phys. Lett.

1989, 7 5 5 , 6 0 9 .
(176) Diau, E . W . ; Y u , T.; W a g n e r , M . A. G.; Lin, M . C. J. Phys. Chem. 1994, in
press,

(177) G o r d o n , S.; M u l a c , w . ; Nangia, P. J. Phys. Chem. 1971, 7 5 , 2 0 8 7 .

(178) Gordon, S.; M u l a c , W . A . Int. J. Chem. Kinet. (Symp. 1) 1975, 2 8 9 .

(179) Pagsberg, P . ; Sztuba, B . ; Ratajczak, E.; Sillesen, A. Acta Chem. Scandinavica

1991, 45,329-334,

(180) Kurasawa, H . ; Lesclaux, R. Chem. Phys. Lett. 1979, 66, 6 0 2 .

 115

(181) X i a n g , T.-X.; T o r r e s , L. M . ; Guillory, W . A . J. Chem. Phys. 1985, 83, 1623.

(182) B u l a t o v , V. P.; Ioffe, Α. Α.; Lozovsky, V. Α.; Sarkisov, Ο. M . Chem. Phys. Lett.
1989, 759, 171.

(183) Silver, J. Α.; K o l b , C. E. Chem. Phys. Lett. 1980, 75, 191.

(184) B u l a t o v , V. P.; B u l o y a n , Α. Α.; Iogansen, Α. Α.; Sarkosov, Ο. M . ; Cheskis, C.

G. Sov. J. Chem. Phys. 1982, 8 7 8 .

(185) Sarkisov, Ο. M . ; Cheskis, S. G.; N a d t o c h e n k o , V. Α.; Sviridenkov, Ε. Α.; ;

V e d e n e e v , V. I. Arch. Combust. 1984, 4, 111.

(186) D i a u , E. W . - G . ; T s o , T.-L.; L e e , Y.-P. J. Phys. Chem. 1990, 94, 5 2 6 1 .

(187) B a m f o r d , C. H . Trans. Faraday Soc. 1939, 35, 568.

(188) S e r e w i c z , Α.; N o y e s , W . Α., Jr. J. Phys. Chem. 1959, 63, 8 4 3 .

(189) Srinivasan, R. J. Phys. Chem. 1960, 64, 6 7 9 .

(190) Jayanty, R. Κ. M ; Simonaltis, R.; Heicklen, J. J. Phys. Chem. 1976, 80, 4 3 3 .

(191) Phillips, L. F . Chem. Phys. Lett. 1 9 8 7 , 1 3 5 , 269.

(192) Farber, M . ; H a r r i s , S. P. J. Phys. Chem. 1 9 8 4 , 88, 680.

(193) F o n e r , S. N . ; H u d s o n , R. L. J. Chem. Phys. 1958, 2 9 , 4 4 2 .

(194) Silver, J. Α.; K o l b , C. E. J. Phys. Chem. 1987, 9 7 , 3 7 1 3 .

(195) Hall, J. L.; Zeitz, D . ; Stephens, J. W . ; Kasper, J. V. V.; Glass, G. P.; C u r l , R.

F.; Tittel, F . K. J. Phys. Chem. 1986, 90, 2 5 0 1 .

(196) D o l s o n , D . A . J. Phys. Chem. 1986, 90, 6 7 1 4 .

(197) Unfried, K. G.; G l a s s , G. P.; Curl, R. F. Chem. Phys. Lett. 1 9 9 0 , 1 7 3 , 3 3 7 .

(198) S t e p h e n s , J. W . ; Morter, C. L.; Farhat, S. K.; Glass, G. P.; Curl, R. F . / . Phys.

Chem. 1 9 9 3 , 9 7 , 8 9 4 4 .

(199) C r o w l e y , J. N . ; S o d e a u , J. R. J. Phys. Chem. 1990, 94, 8 1 0 3 .

(200) Harrison, J. Α.; M a c l a g a n , R. G. A. R.; W h y t e , A. R. Chem. Phys. Lett. 1986,

130, 9 8 .

(201) C r o w l e y , J. N . ; S o d e a u , J. R. J. Phys. Chem. 1987, 9 7 , 2024.

Research in Chemical Kinetics, Volume 3
R.G. Compton and G. Hancock (editors) 117
© 1995 Elsevier Science B.V. All rights reserved.

Recent Advances in the Kinetics of Radiolytic Processes

3 b
Simon M. Pimblott and Nicholas J. B. G r e e n

a
Radiation Laboratory, University of Notre Dame, Notre Dame, Indiana 4 6 5 5 6 , U.S.A.

d e p a r t m e n t of Chemistry, King's College London, Strand, London W C 2 R 2 L S , U.K.

Theoretical methods for elucidating the ultra-fast kinetics in the fast electron radiolysis of
liquids are described and discussed.

1 INTRODUCTION

As a high energy radiation particle, such as a fast electron, passes through a liquid it
leaves in its wake a track of events where energy has been transferred from the particle to the
liquid. Each energy loss event gives rise to a discrete ionization or excitation, or to a
correlated group of ionizations and excitations. The molecular ions, low energy electrons and
excited molecules undergo very rapid thermalization, fragmentation, and ion-molecule
reactions, and on a picosecond time scale these processes result in a spatially nonhomogeneous
cluster of highly reactive radicals and ions.3>4 The radiation-induced species diffuse rapidly and
undergo near diffusion-controlled reaction.5 Consequently, irradiation is followed by a period
of ultra-fast chemistry which reflects not only the reaction mechanisms, but also the initial
spatial distribution. The time scales of importance in radiolysis are summarized in Figure l β

The short-time chemical kinetics following irradiation contain the most direct
experimentally available information about the primary processes of radiolysis, and an accurate
theory is necessary to extract this information. Constructing a framework for modeling poses a
number of problems: (i) the determination of the relative disposition of the primary energy loss
events, (ii) the description of the physico-chemical processes which determine the number of
reactants in each radiation-induced cluster, (iii) the calculation of the spatial distribution of the
reactants within each cluster, and (iv) the diffusion-kinetic modeling of the chemical reactions.
Any complete analysis must contain an acceptable treatment of all these problems.

This article describes recent progress in the understanding of fast electron radiolysis.
T h e following section discusses the energy loss of a fast electron in water and the
consequences of the energy transfer. The third section of the review describes different
118

theoretical treatments of radiation-chemical kinetics. The development of the kinetic models

and their application to problems of particular interest are detailed, and studies addressing
some of the important unresolved questions are reviewed. Finally, the different models are
critiqued.

Physical
•
Physico-chemical
Chemical
Biochemical
Biological
C
ΌO Ό
C
O
Ο Ω. Φ φ
to Χ φ οΕ
(/) σ> ο
ο Χ ο ο ~

ο Ε c ε
E? - φ
cφ οCO
+

ο ο

LU χ
c ο ο
ο co ο

-16 -12 -8 -4 0

log(time/sec)

Figure 1. T i m e scales of the various processes of importance in radiolysis.

2 FAST ELECTRON RADIOLYSIS

2.1 Energy loss and spurs
T h e spatial distribution of the primary energy loss events along a radiation track is
determined by the properties of the particle and of the medium^. T h e effects of radiation
quality on chemistry or biology are usually parameterized in terms of the stopping p o w e r or
linear energy transfer (LET) of the radiation, S = - dEldx, which is a measure of the average
rate at which energy is lost per unit length of track. 8

Recently, accurate theoretical formulations for the stopping p o w e r , 5 , and for the
inelastic mean free path, Λ, have been d e v e l o p e d . ^ 10 T h e theory is based upon an
extrapolation of the dielectric response function of the medium in the energy-momentum
plane, ^ and relates the energy loss properties of the electron to the dipole oscillator strength
 119

distribution of the medium, /(ε). T h e resulting equations for the stopping p o w e r and for the
mean free path have the form*2» 13

ε
Ξ = 2χηΙ -/(ε)0(ε/Ε)άε (1)

7
Λ" = χη£~ f(E)L(E I Ε) I εάε (2)

4
where G and L are functions of ε/Ε, χ = ne 1 £ , with e and m being the charge and the rest
mass of an electron, η is the number density of molecules, and ε ^ and ε 0 are the maximum
and m i n i m u m permitted energy losses, respectively.

400

E(eV)

Figure 2. Effect of incident energy on the density-normalized stopping power of liquid water
for electrons.

T h e energy dependence of the density-normalized stopping p o w e r of an electron in

liquid water is shown in Figure 2, and the corresponding dependence of the density-normalized
inelastic mean free path is presented in Figure 3. T h e figures demonstrate that for fast electrons
the primary energy loss events are well-separated. Given a typical inter-cluster separation of
100 nm and a radical diffusion coefficient of 10" m V 8 1
this means that clusters of reactants
120

(commonly known as spurs) are effectively isolated from one another on a microsecond time
scale.

10' Η

1
10

Ε
ο
1 10°

gaseous water
ΙΟ" 1

I I I I 11 I I I

10' 10 2
10 3
10 4
10 5
10 6

E(eV)

Figure 3. Density-normalized inelastic mean free path of electrons in water.

Stopping p o w e r is an average quantity which does not make any allowance for the
spatial integrity of the energy loss events comprising a track. A variety of different methods
have been developed for modeling the structure of electron tracks. 13-19 y n e simplest of these
calculations give only the probability distribution of the energy losses 13-15 while more ?

complex calculations attempt to model the detailed spatial structure of the track. 16-19 j h e
accuracy of all the results is limited by the cross-sections employed. Accurate inelastic cross-
sections are available for many gases. 13,15,20 Some cross-sections have been inferred from
studies of condensed f i l m s ^ l , but information for condensed media is very limited. 15,22

Using the same approximation employed for the stopping p o w e r and for the mean free
path, the inelastic cross-section for events with energy less than ε can also be expressed in
terms of the dipole oscillator strength density, 13» 15

1
c(z, Ε) = (κ (Ε) + (χΙ E)\l f(y)M(y I Ε, ε / E)dy) I η (3)
 121

where Μ is a known function of 7, ε and E. Figure 4 compares the normalized cumulative

inelastic cross-sections, 15

Y = c(E,E)/a(E ,E),
max (4)

for 1 M e V electrons in gaseous and in liquid water. Energy losses are typically larger in liquid
water than in water vapor. Taken in conjunction with the differences shown in Figure 3, this
means that any attempt to model a track in liquid water using gas-phase c r o s s - s e c t i o n s ^ , 2 4
will give energy losses which are too small and too frequent, and therefore the model of the
track will not be accurate.

1.0

e(eV)

Figure 4. Normalized cumulative cross-section for inelastic collisions for a 1 M e V electron in

gaseous water ( — ) and in liquid water ( ).

Straightforward Monte Carlo simulation techniques employing electron energy-

dependent inelastic cross-sections, like those shown in Figure 4, have been used to model the
energy degradation of fast electrons and the probability distribution function of the energy
losses. 13,15,22 g y convention, primary energy loss events of energetic electrons are divided
into three classes : spurs, blobs and short tracks. 14 Spurs are events of less than 100 e V , blobs
are between 100 eV and 500 eV and short tracks are between 500 eV and 5 k e V . Energy loss
122

events greater than 5 keV are assumed to produce energetic electrons that can be treated in the
same way as the primary electron. This categorization of energy loss events is arbitrary and
was originally introduced to reflect the spatial nonhomogeneity of the clusters of reactants
produced. 14

E(eV)

Figure 5. Effect of incident electron energy on the mean energy loss in gaseous water (—)
and in liquid water ( ).

T h e calculated mean energy loss per event in water as a function of incident electron
energy is shown in Figure 5.15 The increase in the mean energy loss as the energy of the
primary electron decreases is simply a result of the convention that the final 5 keV track-end of
any track is regarded as a single entity. If the final track end is ignored, the mean energy loss is
almost independent of primary electron energy. The mean energy loss for a 1 M e V electron is
34 eV in the gas and 38 eV in the liquid. However, these means are influenced by a long tail
towards high energy, as shown in Figure 6.15 The energy loss distribution for liquid water is
clearly shifted to higher energies than that for gaseous water. T h e most probable energy loss in
the gas is 14 e V , while in the liquid it is 22 eV. For energies larger than 10 k e V , the incident
energy has very little effect on the energy loss distribution in both phases. 15 Clearly differences
between the two phases are significant and must be recognized in any acceptable analysis of
the radiation chemistry of aqueous systems or in radiation biology.
 123

0.06
1 MeV electrons

o.oo
ο 20 40 60 80 100
6(eV)

Figure 6. Probability density of energy losses along the track of a 1 M e V electron in gaseous
water ( ) and in liquid water ( ).

Similar calculations have been performed for ice, 15 gaseous h y d r o c a r b o n s ^ , ethers

and a l c o h o l s ^ and for dry D N A . 2 2 There are large differences between the predictions,
reflecting differences in the U V spectra of the media, and hence the dipole oscillator strength
distributions,/fe), used in the calculation of the cross-sections. Results obtained for the series
of gaseous alkanes C H n 2 n + 2 are all very similar, but they differ from those for gaseous water.
T h e largest differences seem to be between the gas phase and the condensed phases;
predictions for liquid water and ice and for dry D N A show some similarity, but are very
different from all the gas phase media studied, including gaseous water. T h e phase effects
observed are not simply due to density, as the energy loss spectrum shifts. Since the energy
loss properties of fast electrons depend upon the irradiated medium, it is important to use the
correct model system in any study.

M o r e sophisticated Monte Carlo calculations have been used to provide idealized track
structures in gaseous and liquid water, which means in the first instance the spatial positions of
the energy losses. 16-19 \ y h u the spectrum of energy losses in liquid water, and in numerous
e

gases, is n o w well-characterized, other important features which are necessary to generate a

track in 3 dimensions, such as the angle-dependence of differential cross-sections and the
behavior of low-energy electrons, are not well-understood. These calculations rely heavily
124

upon the cross-sections employed. While most of the necessary cross-sections are available for
gaseous water, the most studied medium, there is little information for condensed media. T h e
differential cross-sections necessary to describe many of the elastic and inelastic processes are
simply not available for condensed media and have to be extrapolated from the gas phase. 19

2.2 Ionization a n d excitation distribution

T h e modeling of the radiation chemical kinetics following irradiation requires a
description of the initial spatial distribution of the radiation-induced reactants.5,25-30
requires more information than simply the energy losses which comprise the track, since kinetic
calculations require a distribution describing the chemical contents of the spurs and their spatial
distributions.5»28,30 T n e step towards this end is to analyze the radiation physics of the
energy loss: at the simplest level, given a particular energy loss, is the event an ionization or an
excitation?

There is a great deal of knowledge about radiation physics in the gas phase, where the
structure in the ultra-violet optical spectrum gives cross-sections and thresholds for well-
defined excitations and ionizations. However, experiments on amorphous condensed matter
are much more difficult, and the ultra-violet spectrum is less well understood. For many years,
for example, there has been controversy about the assignment of the 22 eV maximum in the
U V spectrum of liquid water.31

No. of ionizations No. of ionizations

Figure 7. Effect of secondary electron energy on the distribution of the number of ionizations.
 125

Recently a new approximate, analytic theoretical approach has been developed for the
calculation of the entity-specific ionization and excitation probabilities for the fast electron
radiolysis of water.32 This method incorporates existing data on the ionization threshold and
the energy dependence of the ionization efficiency. When applied to water vapor these analytic
calculations are in close agreement with the predictions of M o n t e Carlo track structure
calculations. 17

Figure 7 shows the effect of secondary (progenitor) electron energy on the probability
distribution of the number of ionizations occurring in g a s e o u s ^ n d in liquid water. T h e
a

numbers of ionizations and excitations are not exactly proportional to the spur energy. For a
given energy loss the yield of (secondary) ionizations is considerably higher in the liquid than in
the gas. This difference primarily reflects the smaller amount of energy needed to create an
ionization in the liquid compared to the gas.

Figure 8. Distribution of numbers of ionizations and excitations for a 40 eV spur in liquid

water.

T h e joint distribution of the numbers of ionization and excitation events for a 4 0 eV

spur in liquid water is shown in Figure 8. Figure 9 shows the corresponding distribution per
event in a 1 M e V track.32 The distribution for the "typical" 40 eV energy loss event is very
different from that for the whole track. The most probable spurs in a 1 M e V track contain one
126

ionization, two ionizations or one ionization and one excitation, while a 40 eV spur usually
contains an excitation and one or two ionizations. The differences between Figures 8 and 9
show that it is necessary to incorporate an acceptable model for the whole track in a full kinetic
model of radiolysis. Since the chemistry of a spur depends strongly on the numbers of
ionizations and excitations it contains, the "typical" spur model is not likely to give accurate
insights into the relationship between track structure and chemistry.

excitations

Figure 9. Distribution of ionizations and excitations for the fast electron radiolysis of liquid
water.

The radiation physics of liquid hydrocarbons has not been studied in the same detail as
water, and the distributions of the numbers of ionizations and excitations in spurs are
frequently assumed to be the same as in water.33 As there are significant differences between
the energy loss distributions and the ionization thresholds in water and in hydrocarbons, this
assumption is doubtful.

Although the problem of excitation and ionization is a major gap in our knowledge at
present, the theoretical advances described here seem to provide a promising way forward. It is
likely that studies using synchrotron radiation, where the energy deposited in an event can be
 127

carefully controlled, will provide new experimental data, which will enable this approach to be
tested properly.

2.3 Thermalization a n d fragmentation

T h e excited states and ions produced in radiolysis only have very short lifetimes in
liquid water. They rapidly fragment or react with surrounding molecules. T h e electrons
produced in ionization events thermalize and are solvated very rapidly. Recent photoionization
experiments have given considerable insight into the formation of solvated electrons and O H
radicals.34-38 Additional information about these processes has been provided by low-energy
electron scattering experiments,^! and synchrotron studies. However, the current state of
knowledge is still very incomplete, even in a qualitative sense. It seems likely that not all the
important processes have been identified yet; for example, a resonance in the low-energy
electron energy loss spectrum of amorphous ice has recently been found, which coincides with
the direct production of molecular hydrogen.21 This could be dissociative electron attachment
to a water molecule, and has not been included in any published code for modeling radiation
chemistry to date.

At present, therefore, there is no detailed quantitative understanding of fragmentation

and thermalization processes of the molecular ions, low energy electrons or excited states
produced, either in liquid water or in other condensed media. As a result, it is necessary to
m a k e simple parameterizations to convert a given energy loss to the spatial distribution of
reactive species that is necessary to model chemistry. These parameters d o not usually have
any theoretical foundation, but are manipulated to give acceptable initial yields of the radiation-
induced s p e c i e s . 2 9 This j s0 n e of the major unsolved problems of radiation chemistry.

3 RADIATION C H E M I C A L KINETICS

In gaseous water, gaseous hydrocarbons, liquid water and solid D N A , well over 70
percent of the energy of a fast electron is lost in primary events smaller than 100 e V , 2 2 which
result in fewer than six or seven ionization/excitation events. ^2 T h e initial distribution of the
reactants in these small clusters is spatially nonhomogeneous and consequently the diffusion of
the reactants as well as the reaction rate parameters influences the fast chemistry following
radiolysis.

A number of different theoretical treatments have been developed for studying the
n o n h o m o g e n e o u s radiation chemical kinetics of solutions. These techniques can be classified
into t w o broad groups: deterministic^39-42 a n c j stochastic.26\28-30 The former use
conventional macroscopic treatments of concentration, diffusion and reaction to model the
chemistry of "typical, average" spurs. In contrast, stochastic analyses consider realistic clusters
128

of reactants and use probabilistic methods to model the kinetics. Each approach has
advantages and limitations. At present both treatments have a valuable role to play in the
modeling of short time radiation chemistry.

3.1 D e t e r m i n i s t i c k i n e t i c m o d e l s
Conventional deterministic treatments of spur kinetics model the fast chemistry using a
typical spur41>42 o r distribution of spurs.^ The spur is characterized by a space-dependent
concentration c for each species i and the spur chemistry is described by a set of coupled
t

diffusion-reaction equations. Each equation comprises terms which describe the diffusion of
species /, its reaction with other radiation-induced species and with homogeneously distributed
solutes (scavengers), and its production by reaction,

2
^- = D,V c,-lk,,c,c ) + lk „c c .
m m n (5)

In the equation D is the diffusion coefficient of species /. The initial spatial distributions of the
t

reactants are represented by spherical concentration profiles, which are usually assumed to be
Gaussian,

C ( r ) = 2 312 eXP( 2 2 σ ) ( 6 )
' (2πσ ) ~'' ' · '

where N[ is the number of particles of species / in the spur, although alternative distributions
have also been used.43 The parameter σ, is the standard deviation of the Gaussian distribution,
while N is calculated from the initial G-value, N = G^t IW0
i f spur eV (where ε is the energy
of the spur). T h e set of diffusion-reaction equations is solved by numerical m e t h o d s ^ >42 o r by
using the prescribed diffusion approximation,5>39 h i c h assumes that the concentration profile
w

remains Gaussian and analyzes the way in which the Gaussian spreads by diffusion and
reaction.

3.1.1 F A C S I M I L E m o d e l
Several different numerical treatments have been developed for solving the set of
coupled diffusion-reaction equations (5).41-43 Most recent studies have been performed using
a technique in which a "typical, average" spur is used as a representation of the electron track.
T h e spur is divided into concentric shells as shown in Figure 10.
 129

Figure 10. Division of the typical spur into concentric shells in the deterministic treatment of
spur kinetics using the F A C S I M I L E algorithm.

Each shell is sufficiently thin that the concentration of reactants within the shell can be
regarded as constant. Diffusion takes place between adjacent shells and reactions within a shell
are modeled using deterministic rates. The resulting set of coupled equations describing the
diffusion and the kinetics is solved using the F A C S I M I L E implementation of the Gear
a l g o r i t h m . ^ T h e parameters defining the spur for aqueous solutions have been optimized to
match measured experimental yields in simple scavenger s t u d i e s . ^ The optimum standard
deviations for the initial concentration profiles, σ-, are given in Table 1 and the energy of the

s
P >
u r
*W i s 6 2
· 5 e V
-

3.1.2 Prescribed diffusion

T h e prescribed diffusion approximation employed for solving the coupled diffusion-
reaction equations (5) assumes that reaction affects only the density of the reacting particles
and not the form of their nonhomogeneous concentration profile which is only modified by
diffusion. T h e prescribed diffusion approximation is suggested by the Green's function for the
diffusion equation being a Gaussian whose variance increases linearly with time so that

2 2
C (t) =o(t = 0) +2Dt. (7)

T h e prescribed diffusion methodology was originally used to study the kinetics of prototype
single species systems4>45,46 a n ( j n a s be e n extended to more general reaction schemes.5
130

Table 1. Diffusion coefficients and spur parameters for the radiation-induced species in
aqueous solutions.42

Reactant D χ 10 , 9
σ, nm G°, G 650
, a , ns
2^-1
m s molec./ molec./
lOOeV lOOeV

E
AQ"
4.5 2.23 4.8 2.55 0.91

9.0 0.85 4.8 - -

ι

OH 2.8 0.85 5.50 2.56 a

0.30 a

b
2.66 b
0.26

Η 7.0 0.85 0.42 - -

H 2 5.0 0.85 0.15 0.44 0.96

-
OH 5.0' - - - -

H 0 2 2 2.2 - 0 0.72 0.38

a b
deaerated solution, aerated solution

The assumption of a linear time dependence for σ is incorrect if the reaction kinetics
are nonlinear, as in radiolysis.^»47,48 Several attempts have been made to modify the
approximation to account for faster reaction close to the center of a spur, due to second order
kinetics.5.39 These modifications introduce a nonlinear time dependence for the width of the
concentration profile, i.e.

^ i =2D + / ( f ) . (8)
dt

T h e form of fit), depends on the approximation by which it is derived.

 131

3.2 Stochastic m e t h o d s
A variety of different stochastic treatments are being developed for modeling fast
radiation-chemical k i n e t i c s . ^ By far the most commonly used approach involves the Monte
Carlo r a n d o m flights simulation of the trajectories of the radiation-induced radicals and
ions.29,50-52 T h i s t V p e 0 f calculation is elaborate and computer intensive, but provides
detailed information about both the diffusion and the reaction processes. As the technique is
vulnerable to discretization (time step) errors, it is important to validate the simulation
methodology to demonstrate that the predictions are quantitatively reliable.49

Recently, a more efficient simulation method, the independent reaction times (IRT)
model, has been developed which requires only the generation of random reaction times for the
radiation-induced reactants.30,53-55 This technique is based on the independent-pairs
approximation that is implicit in the Smoluchowski-Noyes treatment of diffusion-limited
reactions.56 An elegant, analytic master equation (ME) formulation for spur kinetics, which
relies on the independent pairs approximation, has also been described.28,57-59

3.2.1 R a n d o m flights simulation

T h e random flights modeling of radiation chemical kinetics involves the simulation of
the diffusion and the reaction of the particles in an isolated track entity or a "representative
section of track". Several simulation methodologies at different levels of sophistication have
been developed, and have been described in detail.29,50-52 £ a c n simulation begins from a
realizable configuration, which may be generated at random from the assumed spatial profile of
a spur50-52 0 r from an idealized track s t r u c t u r e . ^ The diffusive trajectories of all the
surviving particles are simulated using a discretization of the appropriate stochastic differential
e q u a t i o n s . ^ Reaction is modeled by testing for overlap at the end of a time step or, as it is
quite possible for t w o particles to diffuse together and then separate during a time step, by
estimating the probability of encounter during a time step using a bridging process.60 Each
realization is continued until all reaction is complete or a predefined cut-off time is attained.
5 6
T h e kinetics are obtained by repeated simulation. Typically 10 to 1 0 realizations are required
to obtain statistical accuracy.

The methodology for random flights simulation has been described in

detail.51,52,54,61 j n t n e calculations reported here the diffusive j u m p of a particle / is given by
an Euler-type discretization of the stochastic differential equation62,63 £ a c n diffusive jump
has two c o m p o n e n t s ,

δη = fiDfitN (0,l)
3 + 2£-St (9)
132

where D is the diffusion coefficient of the particle, bt is the length of the time step and N is a
i 3

three dimensional normal random vector of unit variance.51>52 The fi^t term in the expression
(9) represents the motion caused by the buffeting of the solvent and the second reflects the
drift caused by any force F. Simulations have been reported, and the methodology validated,
for several types of idealized system.^ 1,52,54,59,61

For neutral reactants the force F= 0, and t h e r e f o r e ^

δ/; =j2D$tN (0,1)3 (10)

In this case, the simulation provides exact samples from the diffusive trajectories. When the
particles are charged, the strong Coulomb inter-ion forces modify the diffusion. The force on
ion / is then described by5 1 >52

(11)

where z- is the charge on ion /, and r~ is the inter-ion vector. T h e Onsager distance, r , is the
x c

distance at which the potential energy due to the Coulomb interaction of two unit charges has
magnitude k T. In the presence of an applied electric field, E, the expression for the force on
B

an ion is further modified, g i v i n g ^

F^-kJ^^ir^+zfE. (12)

where e is the electron charge.

In some simulations in the literature, the length of the time step, δί, is fixed.50 This
fixed time step treatment is computationally inefficient: when the particles are well-separated
the probability of encounter is small. Sophisticated methods have been developed which allow
more efficient computation by incorporating variable time steps.52,54 j n these treatments, the
time step, δ/, is determined by the proximity of the particles. The time step is selected by one
of t w o methods: (i) so that the pair with the shortest interparticle separation has an encounter
probability of less than 1 percent for the time step54 ? 0 r alternatively (ii) so that the
interparticle drift of no pair changes by more than 10 percent during the time step.52 in both
treatments, the minimum time step becomes increasingly small as two particles approach and a
minimum time step of at least 0.1 ps is usually employed to ensure that the simulation does not
become prohibitively slow.^4
 133

If a reaction is diffusion-controlled then reaction occurs with certainty on encounter. In

a simulation employing finite time steps, it is feasible for the simulated trajectory of a pair to
encounter during a time step and then "separate" before the end of the time step. T h e pan-
should react, but the reaction would not be registered in a simple model which only tests for
overlap at the end of the time step. The modeled chemistry then underestimates both the rate
and the a m o u n t of reaction. As long as the time step in which the encounter occurs is
sufficiently short that no third particle interferes, the problem of encounter during a time step
can be overcome using the conditional encounter probability for an interpolating "bridging
process".60 T w o forms of bridging process have been used : the Bessel bridge, which assumes
that the separation behaves as a Bessel process, and the Brownian bridge in which the
separation between the two particles is approximated by a Wiener process.

In the absence of any interparticle forces, the separation between two diffusing particles
is a Bessel process, and the probability of two particles with relative diffusion coefficient D'
encountering at a distance a while diffusing from a separation χ to a separation y in a time step
bt is

^ _ exp(-(x-a)(y-a)lD'ht)-exp(-xyID'ht)
B e s
(13)
~ l-exp(-xylD'ht)

The Bessel bridge is applicable to any size time step for a pair of uncharged reactants. T h e
Brownian bridge is based on the assumption that the radial drift of the pair does not change
significantly during the time step. This limitation is equivalent to method (ii) for determining
the length of the diffusive time step. Under these circumstances, the probability of encounter is

W Bro = exp(-(x - a)(y - a) I D'bt). (14)

T h e Brownian bridge can be applied to both charged and neutral pairs.

T h e geminate recombination of a pair of neutrals and of a pair of ions in solvents of

differing permittivity (water, alcohols, ethers and hydrocarbons) have been simulated using the
r a n d o m flights method described above, incorporating the Coulomb and applied forces. The
simulations match the corresponding exact solutions of the Debye-Smoluchowski equation
accurately,^ 1,52,54,59,61,64 a s demonstrated in Figures 11, 12 and 13. The time dependence
of the diffusion-controlled reaction of a geminate pair of neutrals is shown in Figure 11.
Included in the figure are the results of simulations ignoring the possibility of encounter and
separation during a time step, but using the same reaction distance. The consequences of this
134

error are very significant; the time dependence of the modeled kinetics is too slow and the
escape probability is overestimated. Clearly, neglecting the possibility of encounter during a
diffusive time step is equivalent to using a reaction distance smaller than the distance at which
encounter occurs.

0.2 h

1 1 1
0.0 — — — — — I
1 2 3 4 5
10° 10 10 10 10 10
time (ps)

Figure 11. Geminate recombination kinetics of a pair of uncharged radicals initially 1 nm apart
9 2 1
with relative diffusion coefficient 5 χ 10" m s" . The full line ( ) is the exact analytic
solution, the filled points ( · ) are the predictions of random flights simulation incorporating the
Brownian bridge and the open points ( O ) are the predictions of random flights simulation with
the possibility of encounter and separation during a time step neglected.

T h e simulated reaction kinetics of a molecular cation - molecular anion pair in a low

permittivity solvent like n-hexane are presented in Figure 12. R a n d o m flights simulation
accurately reproduces the numerical solution of the Debye-Smoluchowski equation. The
results of simulations examining the effect of weak applied electric fields on the yield of
geminate ion-pair escape are given in Figure 13.^9 The escape probability predicted by random
flights simulation is independent of the relative diffusion coefficient of the ion pair and is in
good agreement with that predicted by an analytic Onsager analysis.
 135

τ—ι ι ι 11111 1 I I 1—I I I Hll| I I I I I llll

10 1
10 2
10 3
10 4
10 5

time (ps)

Figure 12. Geminate recombination kinetics of an ion pair initially 6 nm apart with relative
9 2 1
diffusion coefficient 2.5 χ 10~ m s" in a n-hexane-like low permittivity solvent with lr l = 29 c

nm. T h e open points ( O ) are the numerical solution of the Debye-Smoluchowski equation, the
filled points ( · ) are the predictions of random flights simulation and the solid line ( ) is the
extrapolated solution from the grid constructed for the generation of reaction times in the I R T
model (see next section).

T h e random flights simulation technique has been extended to include systems where
separation and re-encounter occur, i.e. where reaction is only partially diffusion-controlled 61
or spin-controlled. When re-separation is possible following an encounter, it is necessary to
modify the simulated positions of the encountering particles, as they are not permitted to
penetrate one another. T h e technique for calculating the positions of the particles following a
reflecting encounter is described in detail in reference 6 1 , but in the simplest case, where the
two particles have the same diffusion coefficient, it is essentially as follows.

If the positions of the two particles are denoted X and X then the t w o vectors R = X
l 2 2

- Xj and S = X + Xj are
2 considered. Following a diffusive j u m p R and S become R+5R and
and the latter is unaffected by a reflective encounter. T h e new vector R+dR is resolved
into components, one parallel and one perpendicular to R. Of these only the parallel
component is altered by an encounter. T h e distance of closest approach, which is prescribed by
the nature of the collision, is calculated and this distance is used to modify the perpendicular
136

component of the vector R+ δ/? . A new vector is constructed to replace /7+δ/7 and the new
positions Χ + δ Χ and Χ + δ Χ evaluated from this new vector and from S + 6 S .
7 7 2 2

0.15 ι 1 « ι « 1 • 1 * 1

c/) 0.05 h
LU

0.00 1
' 1
' 1
' 1 1 1 1 1

0 2 4 6 8 10
Field strength (kV/cm)

Figure 13. Effect of applied field strength on the escape probability of an ion pair in n-hexane.
T h e initial relative distribution is a Gaussian of standard deviation 6.5 nm. T h e solid line is the
analytic Onsager solution and the points refer to the predictions of random flights simulation
for a molecular cation - molecular anion pair ( · ) and for a molecular cation-electron pair ( • ) .

N o additional approximations are introduced in the simulation of the kinetics of

multiple particle clusters, therefore the observed agreement for a geminate pair confirms
random flights simulation as an acceptable model with which to test more approximate
modeling techniques.

3.2.2 I n d e p e n d e n t r e a c t i o n t i m e s m o d e l
Full random flights treatment of diffusion and reaction provides a great deal more
information than is required for kinetic studies, which ultimately use only the reaction times of
the reacting radicals and ions. The IRT model is a Monte Carlo simulation technique which
employs the independent pairs approximation to allow the calculation of reaction times without
having to chart the diffusive trajectories of the reactants.53 As in the r a n d o m flights method,
the calculation starts from a set of coordinates generated from a suitable spatial distribution
and deals appropriately with any particles formed in a reactive configuration. T h e inter-particle
 137

separations between surviving particles are then used to generate random reaction times from
the reaction time probability distribution functions appropriate for isolated pairs with the same
separations. It is here that the independent pairs approximation is employed - each reaction
time is generated independently of any times already generated for other pairs. T h e minimum
of the resulting ensemble of times is taken to be the reaction time of the first pair. This pair is
allowed to react, and new reaction times are generated for any reactive products formed. The
minimum of the new ensemble of times is the next reaction time. The simulation proceeds in
this manner until the cut-off time is reached, or all the potentially reactive pairs have reacted.
Kinetics for a given spur are obtained by repeated simulation. T h e methodology of the I R T
model has been described in detail elsewhere.30>51-54,61

T h e success of the I R T model is largely dependent on using the correct probability

distribution functions for generating the reaction times. This reaction time distribution function
is determined by the nature of the pair and by its initial separation. For a diffusion-limited
reaction, the function is assumed to be the time-dependent reaction probability, W(r,a;t), for
the pair as if it were in isolation. The reaction probability is obtained by solution of the
backward, or adjoint, diffusion equation for the pair,65

2
dW ^,d W ^,(2 dU\dW
= D'— r + D' — (15)
at dr \r dr J dr

Here r and a are the initial separation and the encounter distance of the pair, respectively, U is
the potential energy of interaction between the pair in units of k T and D' is the relative
B

diffusion coefficient.

For neutral particles, in the first approximation, there is no potential energy of

interaction between the pair, i.e. U = 0, and exact analytic solution of the equation is possible
both when reaction is fully diffusion-controlled,^^

W(r,a;t) = —erfc jjr (16)

2
r \2(D't)" )

and when it is partially diffusion-controlled (approximated using the radiation boundary

0 7
condition) ^
138

(Π)
w
< r
' a ;
Ο
=
ζ» 0
*Serf W"«P( PY
c 2
+ Ύ )«#(β
2
+ Y))

ll2 l
where $ = (r-a)/(4D't) y b = D'!va and y = (D't)(l + b )la, with ν being the reaction
velocity defined by the radiation boundary condition.68

, 1 r -

0.001 ^

Ε - 0.01 ν
c
0.1 v. ^

- -

y
1 ι - _L ι

0 1 2 3
x (nm)

Figure 14. Comparison of effective distance scales for pure Coulomb potential (dotted line),
Debye-Hiickel screened Coulomb potential for solutions of ionic strength 0.001 M , 0.01 Μ
and 0.1 Μ (solid lines) and zero potential (dashed line).

A number of approximate analytic solutions are also available for ionic reactions in high
permittivity solvents.69-76 i t c a n ^ s h o w n that for the Coulomb and for the screened
Coulomb potential the approximations are equivalent to the use of an effective distance
s c a l i n g ^ , which represents the natural distance scale in the diffusion equation,77 and governs
both the ion-pair escape probability and the asymptotic approach to this escape probability.
The transformed distance scale is^7

x ={-j:exp(U(y))/y d y
eJf
2
y (18)
 139

where U(y) is the distance-dependent inter-particle potential in units of k T. B Figure 14

compares the distance scaling for a cation-anion pair in water. It shows the results for a pure
C o u l o m b potential (dotted line) and for the screened potential arising from Debye-Huckel
theory for solutions of ionic strength 0.001 M, 0.01 Μ and 0.1 M . Acceptable approximate
solutions for W(r,a;t) for an ion-pair in a high permittivity solvent, are obtained by substituting
a ana r ior
eff eff a and r in equations (16) and (17). 67

Reaction times are generated from the (approximate) reaction time distribution
functions by the inversion m e t h o d . ^

When the C o u l o m b force between an ion pair is strong, as in low permittivity solvents
such as hydrocarbons, there is no suitable simple approximation for W(r,a;t). T h e exact
Laplace transform of the geminate reaction probability has been obtained,79 however, the only
exact solutions available in the time domain are numerical, either by numerical inversion of the
Laplace transform*^ or by direct numerical solution of the backward equation.76
Consequently, a method has been developed for generating reaction times for the I R T model
by sampling and interpolation from a grid of numerical solutions. T h e technique has been
described in detail elsewhere.^l T h e important steps are the construction of the grid of
solutions and the interpolation between the grid points for the reaction time.

T h e grid of numerical solutions was constructed in a coordinate space obtained by

scaling the backward equation to a dimensionless formal using ρ = 2rlr c and τ = 4D'tlr^ and
then m a k i n g a transformation,

W(r,a;t) = W(a,r;o*) W*(r,a;t), (19)

0
to separate out the infinite time asymptote, W(r a;oo) t 9 which is known a n a l y t i c a l l y ^ 9

W(r,a;oo) = a lr .
eff eff (20)

T h e resulting transformed backward e q u a t i o n ^

dW* 2
d W* 2 , -i^dW
-z— = ^ - ^ +—(p-coth(p
2 2
Ί)-^— (21)
dt dp p 3p
140

was solved using the Crank-Nicolson method over the range -3 < ln(p) < 2 and -3 < Ιη(ζ) <
1.5, where ρ and ζ are the transformed inter-ion separation and reaction distance, respectively.
Outside the range of this grid the approximate solutions of the backward equation detailed in
references 76 a n d 82 ^ acceptable.

T h e random reaction time for an ion pair is generated by transforming the reaction
distance and interparticle separation for the pair. The three closest lattice points in the solution
grid are determined and a linear interpolation in the Ιη(ρ)-1η(ζ) plane gives the reaction time
for a given reaction probability. The geminate pair reaction probability obtained by the
sampling method is also included in Figure 12 and is in good agreement with the numerical
solution and with the random flights simulation.51>64

T h e random flights and IRT simulation techniques have been widely used to study
stochastic effects in idealized systems and to examine radiation chemical problems in water, in
intermediate permittivity solvents like alcohols and in hydrocarbons.64 Examples of the use of
these techniques are presented in Section 5. Some attempts have also been made to apply the
I R T method to complete radiation tracks or track sections, modeled by Monte Carlo
methods. 19,26,83 These methods are still in a developmental stage, but it seems likely that the
IRT method will be a very useful tool in analyzing realistic tracks in the near future.

3.2.3 M a s t e r e q u a t i o n a n a l y s i s
T h e random flights simulation and the I R T method both treat particles individually, so
that each realization of the simulation gives a typical life history of the system modeled. In a
given realization the number of particles of each species is an integer-valued function of time.
Many independent realizations are required to obtain statistically significant kinetics. T h e
master equation method attempts to avoid the need for many repetitive simulations by
analyzing the probability distribution of the spur contents. This can be done in the following
way. A state space is set up in which the states are defined by the number of particles of each
species present. Every time a reaction occurs some reactants are removed and the appropriate
products added; this corresponds to a transition from one state to another. T h e state space is
the collection of all possible such states, and a probability distribution for the spur contents
assigns a probability to each possible state.

In the master equation method all the possible transitions between states are given
model rates, which are calculated from the independent pairs approximation. This procedure
gives a set of coupled linear differential equations.28,49,57-59,84,85 £ a c n equation describes
the time dependence of the probability that the spur occupies a specific state , i.e. that it
contains the appropriate particles. For instance, the probability that a spur is in state j , P^
obeys an equation of the form
 141

^ = Σω,/,.,/;-Σω ,/: /> ; Μ (22)

Ul i k

where co^ and co^ are the numbers of independent routes leading from state / —> j and from j
—> k, respectively, in one transition. The first summation describes the formation of state j from
all other possible states / and the second summation, the decay of state j to the states L The
rate coefficients describing the transition from one state to another, the £, are time dependent.
For reaction between two spur particles, the rate coefficient is dependent upon the
nonhomogeneous distribution of the reactants and the independent pairs approximation is used
to express the rate coefficient in terms of the time-dependent pair survival probability, Ω, of
the reacting pair if they were in i s o l a t i o n , ^

e= ~ L * . (23)
α λ

The survival probability is given by

n(t) = \,g(r)(l-W(r,a;t))dr (24)

where g(T) is the probability density of the interparticle vector. If the reaction involves one of
the spur particles with a homogeneously distributed scavenger, the appropriate rate coefficient
is the time-dependent scavenging capacity,47 i . >e

t = k(t)[Sl (25)

where k(t) is the time-dependent second-order rate coefficient for the reacting species56,65
and [SJ is the concentration of the scavenger.

This type of approach can be applied to all systems, however, in practice it is useful for
only a limited number of schemes where the reaction mechanism and the nonhomogeneous
spatial distribution are simple.49 This is because the dimensionality of the state space increases
by one for every new chemical species added, and because no method has yet been developed
for dealing with reactive products in the master equation. Work on the latter problem is in
progress.
142

Table 2 : Reaction scheme for the radiolysis of water and aqueous solutions.91

Reaction 1
k(/ lO^M's' )

Rl e "+ e —> H + OH + OH"

2 0.55
aq ^aq

R2 —> Η 2.3

R3 e " + Η
aq
-> H + OH"
2 2.5

R4 e " + OH
aq —» OH' 3.0

R5 e " + H 0
aq 2 2
-» OH + OH" 1.1

R6 H a q
+
+ OH" -> H 0 2 14.3

R7 H + H -> H 2 0.78

R8 H + OH —> H 0 2 2.0

R9 Η + H 0 2 2
—» OH + H 0 2 0.009

RIO OH+OH -» H 0
2 2 0.55

Rll + 0
2
-» o" 2 1.9

R12 H + 0 2 -> H0 2 2.1

R13 H a q
+
+ 0 " 2 —> H0 2 3.8

4 RADIATION CHEMISTRY OF AQUEOUS SYSTEMS

On the picosecond time scale, fast electron radiolysis of water produces groups of very
reactive radicals and ions,6

H 0 +e"
2
+
>H;+OH + e;
radn
Η Θ
2 )Η Θ*
2 >H + O H
H 0* 2 >H + 0
2 >H + 20H
2
 143

T h e radiation-induced reactants undergo fast r e a c t i o n s * ^ , ^ w h i c h are close to diffusion

control,*^ and the observed chemistry therefore depends not only on the reaction mechanism,
but also on the spatial distribution of the reactants and the relative diffusion of the particles.^
T h e diffusion coefficients of the most important reactive species are given in Table 1 (earlier)
and a simple reaction scheme is given in Table 2.

T w o types of experiment are commonly used to probe the nonhomogeneous kinetics of

water : pulse radiolysis^^ and scavenger studies.90 In the former a short pulse of radiation is
used to p r o d u c e a non-equilibrium chemical system whose behavior is then monitored using an
optical, electrochemical or magnetic resonance technique. Scavenger experiments study the
influence of an added solute, known as a scavenger, on an observable yield, which may be the
product of the reaction of a scavenger with a radiation-induced radical or one of the molecular
products, H and H 0 . The effect of scavenger concentration on the yield is then examined.
2 2 2

- r γιτιπιι ι ι ι ι ι ι inii| ι ι ι iii

_ Hydrated electron

CD
-

..1 . . . .....1
» · ··· • • ·

10 6
10 7
10 8
10 9
10 10
10 11

1
k[S] (sec" )

Figure 15. Effect of hydrated electron scavenger on the yield of scavenged hydrated electrons.
Solid lines and filled points refer to the predictions of deterministic diffusion-kinetic
calculations. T h e open points are a compilation of experimental measurements42,92 a n c j t n e

dashed line is a best fit to this data.

4.1 S c a v e n g e r s t u d i e s
Deterministic calculations have been used to help elucidate the short time radiation
chemistry in the fast electron radiolysis of aqueous solutions and to quantify the effects of
144

4
s c a v e n g e r s . ^ Figure 15 shows the results of calculations examining the effects of a hydrated
electron scavenger on the scavenged yield of hydrated electron, G -(s). e T h e results are
presented as a function of the steady state scavenging capacity, s = k[S], and are compared
with a compilation of the available experimental data for a variety of different scavengers.42,92
Agreement between the modeled chemistry and the yields measured experimentally is good.

1 .0 I ' ' ' "'"I ' ' ι '""I ι » » »""| » » ι nni| » ι ι » nii|

Molecular hydrogen
0.8 -

0.6 h

CD

Q Q I ι ι ι 111 ill ι ι ι 11 m l ι ι ι mill ι ι ι mill ι ι iiinil I

Ίο 6
1 0 7
1 0 8
1 0 9
1 0 1 0
1 0 1 1

-1
k[S] (sec )

Figure 16. Effect of hydrated electron scavenger on the yield of molecular hydrogen. Key as in
Figure 15.

In the radiolysis of an aqueous solution, the yield of scavenged hydrated electrons

clearly depends on the scavenging capacity of the solution. This dependence has been
4
parameterized using a number of different empirical e q u a t i o n s . 2 , 9 3 y n e ^ e s t flt q4? t n e

experimental data in Figure 15 to the function

esc esc 12
G(s) =G +(G°- G ) (™^+™ ( 26)
2
I + (as)" +as/2

4
suggested by La Verne and P i m b l o t t ^ is also shown in the figure. T h e scavenged yield of
esc
hydrated electron increases from an escape yield, G y of 2.55 molecules / 100 e V t o a
(theoretical) upper limit, G°, of 4.8 molecules / lOOeV. The time constant obtained is 0.9 ns.
 145

Figure 16 examines the effects of hydrated electron scavenger on the yield of molecular
hydrogen, G (s).^94
H2 Again there is good agreement between calculation and experiment.
T h e yield of molecular hydrogen decreases as the scavenging capacity of the solution for the
hydrated electron increases. For calculation to match experiment a non-scavengeable, initial
yield of hydrogen of 0.15 molecules / lOOeV is necessary.42

Q I • n m u l « »nmtl « . .......I . .......I . .......I

4 5 6 7 8 9 10
10 10 10 10 10 10 10
1
k [S] (sec )
Figure 17. Effect of hydroxyl radical scavenger on the scavenged yield of hydroxyl radical in
aqueous solutions. Key as in Figure 15.

Similar calculations for hydroxyl radical scavengers are shown in Figures 17 and
18 4 2 , 9 4 T h e agreement with experiment is of the same order as found for the hydrated
electron and molecular hydrogen. Notice that there is no initial yield of hydrogen peroxide.
Best fit parameters for the four sets of data are given in Table 1 above. T h e time constants for
the molecular products were obtained from fitting equation (26) with s = k[S], however, for
future reference it should be remembered that scavenging either of the precursor reactants
prevents formation of the molecular product.

T h e calculations presented in Figure 15-18 suggest that the deterministic treatment

provides an acceptable way of parameterizing the chemistry in the fast electron radiolysis of
water and for investigating chemical problems of interest.
146

1.0 ηπη I r-Γτττττη ι-ι-rr

Hydrogen peroxide
0.8 h
ϊ> Δ
ο ^,Δ 1 f
o.6 γ-

0.4

0.2

0.0
10° 10° 10' 10° 10 10 ,υ
10'
-1
k [S] (sec )

Figure 18. Effect of hydroxyl radical scavenger on the yield of hydrogen peroxide in aqueous
solutions. Key as in Figure 15.

4.2 L a p l a c e t r a n s f o r m kinetics
While many experimental data about the influence of scavengers on the radiolysis of
water are available,42,92 h t e r e ar e V e r y few direct measurements of time-dependent yields,
G(t), 86,87 a n ( j s o a n alternative source of information is necessary. T h e dependence of the
scavenged yield of a geminate pair, Q(s), on the steady-state scavenging capacity, s = k[S],
and the decay kinetics, Q(t), in the absence of the scavenger, are linked mathematically by a
Laplace transform relation

&(s) = s£ Q(t)exp(-st)dt. (27)

Radiation-induced spurs frequently result from more than one ionization event, and
typically contain more than two reactant particles.32 Deterministic modeling studies have been
used to investigate whether a similar relationship can be used for the observed G-values, in
spite of the fact that the spur contains more than two particles and that the kinetics are
basically second-order,

G(s ) = sj~ G(t) exp(-st )dt (28)

 147

If this type of relation is accurate, it can be used to describe the radiolysis of water and to
predict the time-dependent G(t) from the scavenged yields, Gfaj.42,94

CD
3

Ίο 5
10 6
10 7
10 8
10 9
10 10
10 11

1
k - [S -] (sec" )
e e

Figure 19. T h e relation between the yield of scavenged hydrated electrons and the underlying
decay kinetics of the hydrated electron in the absence of scavenger. The solid points are the
scavenged yields predicted by deterministic calculations, c.f. Figure 15, and the solid line is the
Laplace transform of the modeled decay kinetics in the absence of scavenger.

In Figure 19 the method is tested by comparing the Laplace transform of the modeled
hydrated electron decay kinetics in the absence of scavenger with the scavenged yields
presented in Figure 15.42,92 y h Laplace relation is not exact because the kinetics are not first
e

order,6>47,48 however, calculations of the type reported here show it to be very accurate for
the chemistry of the hydrated electron (and for the hydroxyl radical).42

A Laplace transform relationship has also been found to link the yields of the molecular
products, H 2 and H 0 , in aqueous scavenger solutions and their formation kinetics in the
2 2

radiolysis of water in the absence of scavenger.42,94 The appropriate transform variable is

2k[S] rather than k[S], as either of two reacting particles may be scavenged. 9 4 a comparison
of the calculated scavenger concentration-dependent hydrogen yield and the Laplace transform
estimate is shown in Figure 20. The Laplace relations for the molecular products provide
access to their "experimental" formation kinetics from the experimental data summarized in
Figures 16 and 18.
148

0.5

0.4

0.3
CVJ
χ
Ο 0.2

0.1

0.0 il ι ι ι mill
,10
10 7
10 8
1 10 10|11
1
k - [S -] (sec" )
e e

Figure 20. T h e relation between the yield of molecular hydrogen and the underlying formation
kinetics of hydrogen in the absence of scavenger. The solid points are the yields predicted by
deterministic calculations, c.f. Figure 16, and the solid line is the Laplace transform of the
formation kinetics in the absence of scavenger.

T h e inverse transform of equation (26) is

esc
G (t) = G +2(G -G )F (j4tTa). esc
f (29)

where Fj is the auxiliary function for the Fresnel integrals.95 Simple substitution of the best fit
parameters given in Table 1 gives an accurate estimate of the desired decay/formation kinetics.
(The best fit parameters in Table 1 refer to the variation in the yields as a function of k[S]. As
the Laplace variable is 2k[S] for the molecular products, the α value in Table 1 has to be
divided by a factor of two before substitution in equation (29), i.e. replaced by a / 2 . ) This
information on the formation of the molecular products is not directly experimentally
accessible, and cannot be obtained in any other way.

4.3 C o o p e r a t i v e s c a v e n g i n g
Deterministic kinetic calculations show that the radiation chemistry of aqueous
solutions is dominated by the reaction of the hydrated electron with the hydroxyl radical,96\97
 149

- +OH 3χ.0"ΜΛ' ) Q H -
aq (R4)

T h e modeled yield of this reaction is about 1.2 molec. / lOOeV, which corresponds to the value
estimated from experimental yields,

0°OH ~ GOH ~2G H o « I-4 m o l e c . / 1 0 0 eV

E
2 2

when the a m o u n t of Η + O H (R8) reaction, 0.2 molecules / lOOeV, is subtracted.

1.5 ιιιιη ι ιιιιι»! ι IIIIIIIJ ι ιιιιιιη ι ιιιιιιΐ|

• ts ] = o
Saq

° [s l = o
OH
•° · ° ·ο • Ο
1.0 •o

CD
0.5

...J . . . ....J . . . ....j ...I

0.0
10 12
10° 10 b
10 7
10 8
10 y
10 10 1
10
1
s = k - [S -] + ke e O H [S ] OH (sec )
aq

Figure 2 1 . Effect of scavenging capacity on the yield of e " + O H reaction. T h e filled and a q

open points are for calculations with a hydroxyl radical scavenger and with a hydrated electron
scavenger respectively.

Scavenging either the hydrated electron or the hydroxyl radical will affect the
subsequent chemistry of the other, and the spur model has been used to examine the
cooperative effects of scavengers.96\97 Calculations shown in Figure 21 demonstrate that the
yield of reaction (R4) does not depend on whether the hydrated electron or the hydroxyl
150

radical is scavenged, but only on the total scavenging capacity of the solution for the hydrated
electron and the hydroxyl radical,

s = kJS _le + k [S ].
OH OH (29)
e r

As a result, the scavenged yield of the hydrated electron, and of the hydroxyl radical, depend
on both the scavenging capacity of hydrated electron scavenger and that of hydroxyl radical
scavenger.

aerated solution
[ H C 0 ] = 0.1 Μ
T> 5
2

Tfc 4
Ο
CO

l a
I
ο
CD

10
10 b
10' 10" 10 a
10 10 1

1
k - [S -] (sec )
e e

Figure 22. Cooperative effect of a hydrated electron scavenger on the scavenged yield of
hydroxyl radical from aerated 0.1 Μ formate solution. The open points are experimental
data^S and the solid line and filled points are theoretical predictions.

Figure 22 shows how the yield of C 0 due to O H radicals in the radiolysis of 0.1 Μ
2

formate solution is affected by the presence of a hydrated electron scavenger. T h e predictions

of the spur model97 are in good agreement with experimental m e a s u r e m e n t s ^ . T h e scavenged
yield increases from 3.3 m o l e c / 1 0 0 eV in aerated solution to an upper (theoretical) limit of 4.1
m o l e c / 1 0 0 eV. The difference represents the amount of reaction (R4) in aerated 0.1 Μ
formate solution. A surface plot calculated to show how the scavenging capacity of hydrated
electron scavenger and of hydroxyl radical scavenger influences the yield of the hydrated
 151

electron scavenged is presented in Figure 2 3 . Only the valley of the surface is accessible in
experimental studies.

8 log(k [S ])
OH OH

Figure 2 3 . Cooperative effects of scavengers on the scavenged yield of hydrated electron.

4.4 Nitrous oxide solution

Studies of radiation chemical kinetics have not been restricted to the primary radiation-
induced radicals (e ~, O H , and H) and the molecular products ( H and H 0 ) . Deterministic
aq 2 2 2

calculations have also been employed to describe the kinetics of the ionic species in water and
in nitrous oxide solution.99 Nitrous oxide is an efficient scavenger of the hydrated electron,91

9 I M S
e - + N 0
2 "°' "' "' )N +0- 2

l 8 x l 0 < M 8
0- + H 0 2 "' ' )OH + OH-

ι υ υ
While most radiation chemical studies have focused on the conversion of e ~ to Ο Η , the
aq

scavenging reaction also provides a convenient way of producing non-equilibrium

concentrations of OH".
152

time (sec)

Figure 24. Time dependences of the ionic species produced by the fast electron radiolysis of
water (open points) and saturated nitrous oxide solution (lines).

Figure 24 compares the predictions of calculations for the kinetics of the ionic species
in the radiolysis of water and of saturated nitrous oxide solution.99 T h e calculations show a
+
considerable difference in the time dependences of the yields of the cationic ( H a q ) and the
anionic (e " + O H " + O") species. The difference primarily reflects an increase in charge
aq

annihilation resulting because the reaction

Uxl0 M s
H ; + OH" " "' "' )H 0
2 (R6)

is considerably faster than reaction (R4). The difference in the yields is very apparent in the
predicted time dependence of the conductivity expected for the solutions, Figure 25.99 T h e dip
in the conductivity of the nitrous oxide solution at about 7 ns is due to the transient O" which
has a smaller equivalent conductance than either its parent, e ~ or its daughter, OH".
 153

0.3 ΙΙ| I I I I ΙΙΙΙ|

OOOOODD oocanffl

~ 0.1
ϋ ο Water
Ό ·Ν 0
2

C
Ο
ϋ
Ull I
0.0
10' 12
10 10 10" 9
10" 8
10~ 7
10"
6

time (sec)
Figure 2 5 . T i m e dependences of the transient conductivity produced by the fast electron
radiolysis of water (open points) and saturated nitrous oxide solution (line).

4.5 T e m p e r a t u r e d e p e n d e n c e of yields
Recent experiments, prompted by the need for information for nuclear plant design,
have probed the temperature dependence of radiation chemical kinetics.88,101-103 Analysis of
these experiments also provides information about diffusion and reaction processes influencing
the basic radiation chemical kinetics. The temperature dependence of the fast electron
radiolysis of aqueous solutions has been modeled using deterministic diffusion-kinetic
methods. fne D e s t fit of model to experiment occurs if the width of the spatial distribution
of the hydrated electron has an Arrhenius-like temperature dependence with an activation
energy similar to that for the movement of an electron between traps in water. T h e spatial
distribution of the other primary radiation-induced species appears to be unaffected by
temperature.

Figure 26 shows the influence of temperature on the radiation chemical yields of

hydrated electron, hydroxyl radical, molecular hydrogen and hydrogen peroxide in neutral
solution. T h e calculations shown in the figure predict an increase in the yields of the hydrated
electron and the hydroxyl radical and a decrease in the yields of molecular hydrogen and
hydrogen peroxide with increasing temperature. These results are in reasonable agreement with
the available experimental data. The g o o d match between calculation and experiment suggests
154

that the initial yields of the primary radiation-induced species are independent of temperature
and there is probably no need for a thermally activated route to molecular hydrogen
production.

1 1 — -

5h
•

e
aq
τ

CD

H 2 -
I AIL 1 ^Μ I

ι ι • Τ 1 •

300 400 500 600 300 400 500 600

temperature (K) temperature (K)

Figure 2 6 . Temperature dependences of radiation chemical yields. T h e solid lines are the
predictions of deterministic diffusion-kinetic calculations and the points are values obtained
experimentally. (Key as for Figures 5 and 6 of reference 104)

4.6 Critique of deterministic methods

The information presented demonstrates that the deterministic treatment of the
radiation chemical kinetics of water and of aqueous solutions provides a g o o d method for
analyzing experimental data. The deterministic approach is reasonably simple to understand
and to use and is based on familiar concepts from traditional macroscopic kinetics.
Deterministic methods are very flexible and can be used to investigate complex reaction
mechanisms in a simple manner. The studies described show that a single set of consistent
parameters gives wide agreement with an extensive set of experiments using scavengers. In
addition to being easily used, the deterministic models provide a useful method for testing the
approximations employed in the analysis of experimental data, c.f. the Laplace transform
relationship between the yields in scavenger experiments and the underlying decay and
formation kinetics.
 155

While deterministic methods provide a simple and useful model for predicting the
radiation chemical kinetics of aqueous systems, the track of a fast electron is condensed into an
averaged spur with several characteristic parameters. In consequence the treatment is
insensitive to the detailed structure of the track, and to the spur size distribution. T h e typical
average spur has no real physical significance: the optimized spur which reproduces the
experimentally observed chemistry of aqueous solutions in electron radiolysis has an energy of
62.5 e V , while the average energy loss of a fast electron in liquid water is slightly less than 40
e V . Furthermore, as the deterministic models of spur chemistry employ a typical average spur
with a nonhomogeneous concentration profile for each reactant, the calculations cannot
provide any insight into the effects of the spatial correlations of the reactants on the chemical
kinetics. This type of model is very useful for considering problems of a mechanistic nature,
but does not provide information about the details of the spatial distributions in a spur.

5 STOCHASTIC ANALYSES

Spurs produced in fast electron radiolysis are comprised of only small numbers of
particles.^2 Conventional models for the kinetics of these systems employ deterministic
methods to model reaction, which d o not consider the individual, and discrete, nature of the
particles.57 This fault is best demonstrated by considering a cluster of two (A/B) pairs. T h e
reaction distances for the three reactions A+A, A+B and B + B are the same, and the initial
coordinates of the four reactants are obtained from identical and independent distributions. As
there is one A-A distance, four A-B distances and one B-B distance, statistical considerations
require the ratio of the products, A : AB : B , to be 1:4:1. For a cluster of η pairs the same
2 2

2
arguments show that the product ratio is n(n-l)/2 : n : n(n-l)/2. In fact, this is the correct
statistical weighting for any η pair cluster, even when the particles are not identical. T h e ratio
prescribed by a deterministic treatment is always 1:2:1, irrespective of the number of pairs,
which is the large η limit of the correct result. It is apparent, even from this simple analysis,
that stochastic effects in nonhomogeneous kinetics can be substantial.

Stochastic techniques have been used to investigate a large number of problems of

interest in radiation chemical kinetics.

5.1 Prototype systems: single species spurs

If the reaction scheme is simple, the stochastic master equation treatment is much more
convenient for modeling kinetics than any other stochastic method, since it only requires a
standard differential equation solver, and is far less computer-intensive than either simulation
method. T h e most basic scheme possible is that for the prototype single species spur
156

R + R -» R . 2

T h e state of the spur is labeled by the number of particles that remain. If the particles are
identically distributed then the probability that a spur has exactly η particles is described by the
set of e q u a t i o n s ^ ,

dP
—± = / (n + 2)(n + 1 ) 1 P
2 R+R n+2 -/ n(n
2 -1)1 P
R+R n (30)
dt

where, under the independent pairs approximation, the time-dependent rate coefficient i+
R R

does not depend on n. Analytic solution is straightforward and the expectation value of the
number of particles surviving is given by^7

< n > = y 2 n
" ! Γ
< η
ο + 1
> ^ # ι - n + />/2;
0

n=„ m 2 n
T(n -n 0 + l)T((n +n
o + l)l2)

where n = 2 if n is even and n = 1 if n is odd. The predictions of the master equation model
m 0 m 0

are found to be in good agreement with the results given by random flights (and IRT)
simulation, which shows that the independent pairs approximation is acceptable, at least for
this simple system.57

T h e master equation technique and the IRT simulation method both employ the
independent pairs approximation. In the simulation model, the calculation considers a set of
coordinates for the reacting particles at the start of each realization. T h e resulting spatial
configuration maintains all the appropriate triangle inequalities. In contrast, the master
equation approach extends the independent pairs approximation to the initial configuration,
thus while the distribution of pair distances is correct, the triangle inequalities d o not hold.
Consequently, the predictions of the master equation and the I R T model differ slightly for
random spatial distributions, e.g. Gaussian spurs, and are only equivalent for fixed
configurations, e.g. four particles in a tetrahedral configuration.

5.2 Prototype systems: t w o species spurs

T h e study of the single species system serves only as a rudimentary test of the
independent pairs approximation; it is of very limited chemical value. T h e simplest system in
which the nature of reaction is significant is where the spurs are made up of pairs of t w o
species (A/B). T h e chemistry of idealized multi-pair clusters has been investigated in
depth.52,54,64 These studies focused primarily on the chemistry of (A/B) pairs in spurs where
 157

the A and Β concentration profiles are considered to be identical and independent.54,64 t w 0

different reaction schemes have been considered, (i) where the reaction distances, a , A+A aA+B

and a , B+B are the same, i.e.

A + A —> A 2

A + Β -> AB Scheme (I)

Β + Β -» B 2

and (ii) where 'like plus like' reaction does not occur,

A + Β -> AB Scheme (II)

i.e. a A+A = a B+B = 0.

Studies for both schemes have been made for neutral pairs using the two M o n t e Carlo
simulation models as well as the master equation method. The stochastic master equation for
the probability that a spur has m A particles and η Β particles is 54,64

dP
- i ^ = /2^ A+A ((m + 2)(m + l)P m+2 - ( -l)P )
n m m mn

dt
+e ((m
A+B + l)(n + l)P m+In+1 -mnP ) mn

+ y t ((n
2 B+B + 2)(n + l)P mn+2 -n(n- l)P ).
mA (32)

For scheme (I) when £ A + A = £ A + B = £ B + B y it is straightforward to obtain the yields of A,

B , A , A B and B
2 2 from the solution for the single species spur, equation (31). For spurs
initially with n pairs, i.e. 2n particles
0 0

<A>=<B>= / <N>=
2 1»ZL , + i
Γ(2η
° >+1
r«2n,-/ + / i / 2 V „ w

,5 2 T(2n -i 0 + l)T((2n +i 0 + l)l2)

where Ν is the total number of reacting particles. Since the ratio of the products A : A B : B 2 2

2
is n (n -l)/2
0 0 :n 0 : n (n -I)/2,
0 0

Η
<Λ >=<β >=
2 2 <· '> <η -/ <Ν»
0 2 (34)
2(2η 0 - 1)
158

and

< AB>= (n - /
0 2 < N » (35)
2n0 -1

A comparison of the predictions of random flights simulation and those of the I R T model and
the master equation method for scheme (I) shows no significant differences.54,64

Under scheme (II) only the recombination reaction takes place and £ A+A = tBjrB = 0. The
master equation for the probability that a spur has η pairs is reduced to^4,58

2 2
^ L = (n + l) tP -n tP n+t n (36)
dt

which is easily solved to give the expectation value of the number of surviving pairs,

2
<„>=£ W<«. + 1» V ( 3 7 )

n=,Y(n +n + l)T(n -n
0 o + l)

As for scheme (I), agreement between the IRT model, the master equation formulation, and
random flights simulation is good.^4

5.3 M o l e c u l a r p r o d u c t f o r m a t i o n
In a real system, different chemical species do not normally have the same diffusion
coefficients and reaction radii. However, it is possible to find systems in which the reaction
scheme is essentially equivalent to schemes (I) or (II). One such scheme is the formation of
molecular hydrogen and hydrogen peroxide in spurs initially comprised of hydrogen atom -
hydroxyl radical pairs.28 i n this simple model for the radiolysis of aqueous acid, the
scavenging of the hydrated electron by acid is assumed to be instantaneous and unscavengeable
sources of molecular hydrogen are neglected.30 The scheme incorporates three reactions

Η + Η -» H 2 <R7)
Η + OH -> H 0 2 (R8)
OH + OH -> H 0 2 2 (RIO)
 159

and the resulting master equation describing the probability that a spur has m hydrogen atoms
and η hydroxyl radicals is given by equation (32) (with m 0 = n ). T h e time dependent rate
0

coefficients, £ , H+U ?H+OW a n c

* ^oh+oh> ^ n o t
equal, since the reaction distances for the three
reactions and the relative diffusion coefficients are not equal, and the distributions of the
particles, governed by the parameters σ Η and σ Ο Η , are not the same. In consequence, it is
necessary to solve the master equation by numerical methods. Comparison of master equation
calculations, I R T simulations and random flights simulations shows good agreement.28

Because stochastic calculations are sensitive to the number of particles in the spur, it is
necessary to average the master equation results over a spur size distribution that is
appropriate for a radiation track. Calculations averaged in this way have been optimized with
respect to the unknown spur widths, σ Η and σ 0 Η , to match experimental escape yields of H,
OH, H 2 and H 0 .28
2 2 in these calculations, a single width parameter describes the
n o n h o m o g e n e o u s spatial profile of a reactant in all spurs, irrespective of the number of pairs.
For a spur of n pairs, the width of the profile is related to that in a single pair spur by the
0

4
prescription

<s (n ) = a (l)n " .

i o i 0
3
(38)

Calculations for several different distributions suggest unit radii in the ranges 0.4 n m < σ ( 1 ) < Η

0.8 n m and 0.2 nm < σ Ο Η ( 1 ) < 0.3 nm.52,64 These ra£ j u are considerably smaller than those
expected in reality, but it must be realized that the model system is a gross simplification of the
true chemistry, and consequently it is not surprising that the parameterization is not realistic.30
These preliminary calculations d o , however, emphasize the potential significance of statistical
effects in radiation chemical kinetics. A master equation treatment of the true short-time
radiation chemistry of water incorporating a full reaction scheme, c.f. Table 2, is too complex
to allow straightforward modeling.49 This is an area that needs further development.

5.4 Reaction in multi ion pair spurs

Approximate stochastic theories might be expected to break d o w n in systems where
some of the particles are charged, because they are based on the independent pairs
approximation. This means that although the reaction time of a pair is influenced properly by
the electrostatic interaction of that pair, the Coulomb forces arising from the presence of other
ions are neglected.

The accuracy of the independent pairs approximation in systems of multiple ion pairs
has also been investigated in depth using stochastic models.52,64 a s in the studies of neutral
+
systems, these calculations use two-species ( A / B ~ ) pairs in spurs. Studies for both schemes
160

(I) and (II) have been made for cation - anion pairs in a high permittivity solvent, such as
+
water. Calculations for two (A /B~) ion pairs reacting according to scheme (II) are shown in
Figure 27. T h e results of the IRT model are in good agreement with those of r a n d o m flights
simulation. Also included are the results for two neutral pairs with the same encounter
distance. T h e C o u l o m b force has a considerable effect on the modeled kinetics.
2
For scheme (I), the statistical weighting of the reactions is n (n -l)/2 0 0 :n 0 : n (n -l)/2,
0 0

however, even if a , A+A aA+B and a , B+B are the same, the ratios of the product yields are different
because the time-dependent rate coefficients i are affected by the C o u l o m b forces. It is
therefore not easy to estimate the correct product ratio analytically. The agreement between
the approximate I R T model and random flights simulation shown in Figure 27 and found in
reference 51 demonstrates that the independent pairs approximation is acceptable for
describing the reaction of ions in high permittivity systems.52,64

2.0

1.5

25 i.o
CD

0.5

0.0

10° 10 1
10 2
10 3
10 4
10 5

time (ps)
Figure 27. Comparison of random flights and I R T simulation for the kinetics of a t w o (A+/B")
ion pair spur (filled points and solid lines) and for the kinetics of a two neutral (A/B) pair spur
(open points and dashed lines). The points refer to the predictions of r a n d o m flights simulation
and the lines to I R T modeling. The same reaction distance was used for the A + + B" reaction
and for the A + Β reaction.

In addition to the simulations testing the accuracy of the independent pairs

approximation, investigations have also been reported to examine whether it is necessary to
 161

include charge effects explicitly in the model, or whether the effects can simply be
parameterized by using effective reaction distances.52 The reason for this is to check the long­
standing claim that charge effects are not important in spur kinetics, since they are already
accounted for in the experimental rate constants (through the effective reaction distance).^
Calculations using the same effective reaction distance, defined by equation (18), but for
neutral and for ionic pairs, show that the inclusion of charge has two observable consequences:
the product ratios are altered and the time scale of the reaction is speeded up.52

5.5 Ion recombination in low permittivity media

In low permittivity solvents only scheme (II) is feasible, because the strong Coulomb
forces prevent cation-cation and anion-anion encounters. In this instance, the agreement
between the predictions of the two simulation models depends upon the initial spatial
distribution of the ions.64 The I R T model accurately predicts the free ion yield from multiple
ion pair spurs, but noticeable errors are sometimes found in the time dependence of the
kinetics.51>64 Figure 28 shows the predicted ion recombination kinetics for a four ion pah-
spur.

T h e errors observed in the predictions of the I R T model are a reflection of the way in
which the Coulomb forces reinforce one another. The reaction time generated for a pair in the
I R T method only incorporates the force between the pair and there is a breakdown of the
independent pairs approximation when the strong inter-ion forces act in cooperation. This
breakdown appears under special circumstances, where the initial distribution is selected so
that the Coulomb forces interfere with one another; for spurs where the spatial distribution of
the reactants is spherically symmetrical the independent pairs approximation is found to hold
reasonably well.51,64

T h e conventional picture of multiple ion pair spurs in low permittivity solvents

considers the relatively slow-moving cations to be centrally located when compared to the
highly mobile anions, which are in a spherically symmetric distribution about the
cations.^06,107 T h e ion recombination kinetics in these spatially simple spurs are amenable to
study using the master equation method.58,108 The kinetics of a multiple ion pair spur are
described by equation (37). This solution, although explicit, is not entirely analytic as it
depends upon Ω , the time-dependent survival probability of an ion pair with the same
distribution of separations, which is not known analytically. 109

Comparison of the results of the master equation model with the simulation techniques
reveals g o o d agreement.58 The predictions for a four ion pair spur are included in Figure 28.
For t w o , three and four ion pair spurs the master equation model slightly overestimates the
amount of reaction (about 2%) and gives marginally faster reaction kinetics than both
162

simulation models While the two simulation models are in good agreement and predict the
same escape yields, the I R T treatment gives slightly slower reaction than r a n d o m flights
simulation. The difference between the I R T simulation and the master equation results from
the extension of the approximation to the initial configuration in the master equation analysis.

πη 1 r
"

v\
3 ο \
V °\
>

\°\
\.\
\\ e

\\

n = 4
0
\\
\ ~~-Γ^^^οηοοο 0 0 0 0 0 0 0

ι ι • imil -k. i u i ..1 . . ......1 .

11 10 9 8 7 6 5
10~ 10" 10" 10" 10~ 10~ 10~

time (ps)

Figure 28. Ion recombination kinetics of a four ion pair spur in which the initial positions of the
cations and anions are generated from Gaussians of standard deviation 2.5 n m and 6.0 nm,
respectively. Onsager distance = 29 nm. The open points are the results of r a n d o m flights
simulations, the solid line of IRT simulation and the dashed line of the master equation.

5.6 Effects of electric field on ion recombination kinetics

T h e r e is a large experimental literature probing the effect of an applied electric field on
the free ion yield following high energy electron radiolysis of hydrocarbons. 110-113 Recent
energy loss distribution calculations have suggested that a significant fraction of the spurs
contain multiple ion pairs. ^ T h e conventional model for ion recombination assumes that every
spur rapidly degrades to a single ion pair and the observed kinetics are those of the final pair.
T h e reaction probability of an ion pair in the presence of an electric field depends upon the
angle between the electric field and the inter-ion vector and upon the inter-ion
separation,! 14,115
 163

W (r, E
a;oo) = 7 - exp(-aEr I r ) £r(*
c + 7, r / r ) ( o £ r / r j * / fit / j
c
2
(39)
k=0

where T(/,JC) is the incomplete G a m m a f u n c t i o n 95

and α = $r (1 + COS&)
c with β = e/(2k T).
B

This equation reduces to* 15

E
W (r, a;<*>) = \ - exp(-r c /c)(l + aE)-.. (40)

for weak fields. If the inter-ion vector has a spherically symmetric distribution, g(r), the field
dependence of the free ion yield is given by

Ω (οο)
Ε
= j g(r)(l
r - W (X,t))dX
E
E
« a *°M(l + rfiE) (41)

and the angular dependence vanishes. Equation (41) predicts a linear increase in the free ion
yield with field strength and a slope-to-intercept ratio of r $. c

R a n d o m flights simulation and the master equation method have been used to probe the
influence of applied field strength on the ion recombination kinetics in multiple ion pair spurs in
low permittivity solvents.59 i n the master equation treatment the applied field is incorporated
in the time-dependent rate coefficient £, which is defined by

Ω £
dt

An analytic approximation for the field dependence of the free ion yield from multiple ion pah-
Ε
spurs is obtained by substituting Ω for Ω in equation (37). T h e predictions of the master
equation for the free ion yield from multiple ion pair spurs with realistic initial distributions are
c o m p a r e d with the results of random flights simulations in Figure 29.

T h e figure shows a linear increase in the predicted free ion yield with applied field
strength. Furthermore, the slope-to-intercept ratio for the effect of the field is numerically the
same as the analytic result for a single ion pair, that is rfi. T h e similarity between the multiple
and single ion pair cases is easily explained. Consider the application of a weak applied field. In
Ε
this limit, Ω is given by equation (41) and upon substitution into equation (37) it is apparent
5 9
that

< n
E
>=< n° > +>;β£ < (n° f >. (43)
164

2.0 ι • 1 1 1 1 ι » τ

0.0 1
' 1
' 1
» 1
' 1
' 1

0 2 4 6 8 10
Field strength (kV/cm)

Figure 29. Effect of applied field on free ion yield from multiple ion pair spurs. T h e lines are
the predictions of the Onsager analysis for a single pair. The solid and open points are the
results of random flights simulation and the master equation model, respectively, for spurs with
n ion pairs. T h e data for n = 2 and 3 are offset by 0.4 and 0.2 units, respectively.
Q Q

In theory, equation (43) suggests that the field dependence of the free ion yield should provide
information about the spur size distribution. Unfortunately this is not the case: for low
permittivity solvents the Fano factor,

2 2
F = (<n >-<n > )/<n> (44)

2
is close to 1 and <n > ~ <n>. In consequence, the slope-to-intercept ratio

E Ε
<n >/ <n° >=Ω /Ω° (45)

and there can be no clear (experimental) distinction between the field dependences of single
and multiple ion pair spurs.
 165

5.7 Reactions which are not diffusion-controlled

In all of the systems discussed so far the reactions have been treated as diffusion-
controlled. It has become clear recently from measurements of the temperature dependence of
the rate constants that many reactions in the radiation chemistry of water are not diffusion-
c o n t r o l l e d . ^ At r o o m temperature the reactions are all close to diffusion control, but at higher
temperatures, such as pertain in a water-cooled nuclear reactor, rates deviate considerably
from diffusion control.88

Elliot et al.88 have identified two different types of temperature dependence: one in
which the rate coefficient follows the Noyes expression describing the transition between
diffusion and activation control,56 and one in which the temperature dependence seems to
follow the diffusion coefficient, but the rate is reduced by a constant factor. M o s t reactions fall
in the former category, where the reaction can be described using the radiation boundary
condition. T h e H + H reaction, however, falls into the latter category, because the rate is
1
reduced by a spin statistical factor. ^

5.7.1 Partially diffusion-controlled reactions

In a deterministic theory there is no distinction between a fully and a partially diffusion
controlled reaction, since the only parameter affecting the reaction rate is the rate coefficient.
A recent paper has been published testing whether this is also the case for stochastic
theories.6I Spurs were simulated with the same effective reaction distance, but with different
true reaction distances. Once corrected for the obvious differences due to zero-time overlap
between particles, this study concluded that the importance of using the correct description
depends on the scale of the nonhomogeneous particle distribution. If it is narrow then the
differences are significant, but for broader distributions it is sufficient to use the effective
reaction distance and assume diffusion control, which is much faster to simulate. Investigations
of partially diffusion-controlled reactions between ions are in progress. Of course, if the
temperature dependences of the nonhomogeneous kinetics are of interest, it is necessary to
take into account the effect of the temperature-dependent "activation" step on the effective
reaction distance.

5.7.2 Spin-controlled reaction : the re-encounter approximation

T h e second reason for a reaction to be slower than diffusion control is that it may be
spin-selective. Many of the reactive particles in a spur are free radicals, and it is well known
1
from photochemistry that such reactions may be s p i n - c o n t r o l l e d . ^ This n o t o n i v affects the
1
kinetics, but it may also give rise to coherent spin effects such as C I D E P or C I D N P . ^

If reaction is spin-controlled, then the occurrence (or not) of reaction upon encounter is
1
determined by the spin state of the pair at the encounter time. ^ T h e spur kinetics and the
166

fractionation between products then depends not only on the initial spatial distribution of the
particles, but also on the initial spin state of the spur, the mechanisms by which this spin state
can evolve and the difference in reactivity between singlet and triplet. A preliminary
investigation of these effects using the master equation method has been reported. 84,85 i n m i s

simplified model, the encounter rate is assumed to be given by the diffusion-controlled

encounter rate, and to be unaffected by previous encounters. This approximation, applied after
an unreactive encounter, is referred to as the re-encounter approximation. Only singlet
encounters are assumed to be reactive. Four different limiting cases are identified, depending
on the relative time scales the spin relaxation and the spur kinetics. In the slow relaxation limit
spin correlation effects are found to have very large effects on the modeled chemistry.84,85

Recently, this master equation analysis has been tested by comparison with random
flights simulations. These proceed in the normal manner except that the spin wavefunction is
followed from encounter to encounter. The spin function is assumed not to evolve in time
except when an encounter occurs. When an encounter takes place, the singlet probability is
calculated from the spin wavefunction, and a random number is used to discover whether the
pair has reacted. If the pair reacts, then the singlet projection operator for the encounter pair is
applied to the wavefunction, which is then renormalized. This accounts for the fact that a
reaction of one pair can affect the singlet probabilities of other pairs, because the overall spin
state of the spur must be conserved. In the event of a triplet encounter, the singlet c o m p o n e n t
of the wavefunction is removed, and the wavefunction is renormalized. G o o d agreement is
found between the master equation results and the random flights simulations.

M o r e unpublished studies using the IRT model have focused on the re-encounter
approximation. Here, the first encounter time of each pair is generated using the equation
appropriate for diffusion-limited reaction. The spin state of the encountering pair is then used
to determine whether reaction occurs or a new re-encounter time is required. If the spin state
of the spur cannot evolve between encounters, the many rapid re-encounters that necessarily
follow the first encounter of a pair are unreactive, and need not be considered. T h e pair
remains unreactive until another pair encounters, at which point its spin state m a y change.

According to the re-encounter approximation, the rate of re-encounter following an

unreactive encounter is unaffected by the first encounter 119,120 T h i approximation allows
s

the generation of a re-encounter time from the original encounter time distribution function for
the pair, conditioned to be greater than the first encounter time. However, the re-encounter
approximation can be relaxed, as it is possible to generalize the I R T model to handle re-
encounters exactly within the independent pairs approximation. As the spin state of the pan-
that first encountered at tj remains unreactive until the encounter of a second (different) pair at
f , the original pair diffuses with a reflecting boundary for the time period At = t - tj. T h e
2 2
 167

cumulative probability distribution function for the separation, r, between a diffusing geminate
pair is then given by

2
F( r, At) = erf (ξ) - 2 ξ η βχρ(2ξτ\ + η )erfc{\ + η ; (46)

1/2 2
where £> = (r-a)/(4D't) and i\ = (D't)" /a. T h e function F(r,At) can be inverted
numerically to generate a random separation for the first encountering pair at the encounter
time of the second pair and this distance is then used to generate a re-encounter time for the
first pair.

It is found that the re-encounter approximation gives a small underestimate (of the
order of 5%) in the overall amount of reaction in the spur. This error helps to counteract a
similar error in the opposite direction incurred by the independent pairs approximation.
Consequently the preliminary comparison with random flights simulations suggest that the
master equation with the re-encounter approximation is slightly more accurate even than the
full I R T method, as a result of these self-canceling errors.

5.8 Spatial correlations and clustering

Conventional wisdom suggests that although stochastic methods may be necessary to
understand properly the kinetics of isolated spurs, if spurs are generated close to one another,
as in a short track or at the track end, deterministic methods should be accurate. This
supposition arises from the large number of possible combinations of particles, as the statistical
weights are close to the deterministic limit. A recent study has indicated that this may not be
the case, 121 because, although there are many possible combinations of particles, they are not
equally likely to lead to reaction. A track-end is laid down event by event with spatial
correlations resulting from the underlying physical processes. All the tests which imply that the
deterministic theory is acceptable for medium to large numbers of particles assume that the
spatial distributions of the particles of a given species are independent and identical, and
independent of all other species. 122 This [ clearly not true in a radiation track; the spatial
s

distribution of a particle depends on the event from which it originated, and a particle of one
species will be correlated with the other particles originating in the same event.

T h e effect of spatial correlations was investigated using a crude model track of

overlapping spurs made up of H/OH pairs, equally spaced along a line. T h e resulting track is
characterized by a spatial density for each species, which can be interpreted in two different
ways. T h e first way takes account of the correlations caused by particles being born in
correlated pairs. The second interpretation assumes that the track is characterized by a
concentration for each species. This classical interpretation of a concentration is clearly
168

unphysical in the case of a track. It results in random numbers of particles of each type being
associated with each spur, and in particular the numbers of Η atoms and O H radicals in a spur
are not necessarily the same. Deterministic calculations should agree well with the classical
interpretation of concentration. The spatial distribution of the track has considerable overlap
between adjacent spurs; however, the kinetics predicted by random flights simulation show that
there are important differences between the two interpretations, particularly in the product
yields. The ratio of H 2 or H2O2 yields to the water yield is much lower in the track with the
correlated distribution than in the purely random (classical) distribution. Similar results are also
found with more realistic track-ends.

From these results, it seems likely that deterministic models will lead to significant
errors in predicting the fractionation between the possible products of radical recombination,
even under conditions where spurs overlap significantly. Furthermore, the problem lies in the
very nature of a deterministic model, which applies classical kinetic rate laws to the spatial
concentration profiles. The fundamental assumption is that the track can be characterized by
the concentrations of each species. T h e macroscopic concept of concentration is inadequate to
define the kinetics of track-ends: two spatial distributions with the same concentrations give
different chemistry, because they have different pair correlations.

5.9 Critique of stochastic methods

Stochastic methods for modeling the diffusion-limited kinetics of spatially
nonhomogeneous systems are inherently probabilistic. These techniques require more detailed
information than conventional deterministic models about the initial spatial distribution of the
particles, about their diffusive motion and about the inter-particle reactions. However,
stochastic calculations provide a more acceptable approach for probing particular factors
influencing the nonhomogeneous kinetics; they have the correct statistical weighting for the
products of reaction and they can cope with spatial correlation effects.

T h e stochastic analyses are particularly good for investigating qualitative effects that
are difficult (or impossible) to study using a deterministic approach. Studies discussed in the
preceding sections considered the effects of charge, of partially diffusion-controlled and spin-
controlled reactions and the influence of clustering, i.e. the "pairwise" formation of reactants,
on the predicted chemistry. Calculations examining the dependence of spur chemistry on the
number of particles comprising the spur demonstrate that the radical decay kinetics show only
a small dependence on the size (content) of the spur, however, the formation of the
(molecular) products depends critically on the spur size. Stochastic methods are also useful for
testing the validity of the approximations used to model chemical kinetics. Specific examples
 169

discussed include the use of an effective distance scaling to account for the kinetic effects of
C o u l o m b inter-ion forces and tests of the accuracy of the re-encounter approximation.

All the calculations presented here involve idealized or simple reaction schemes and the
approximate stochastic methods are easily implemented for these schemes. H o w e v e r , this is
not true for more complex reaction mechanisms, as it is difficult to include secondary
reactions. Consequently, the methods, especially the master equation treatment, are not very
flexible and require further development to make the application to realistic schemes routine.

T h e comparison of the predictions of stochastic calculations with experimental

measurements is of limited value as the problem is over-parameterized. Stochastic calculations
require detailed spatial distributions for the reactants and suitable distributions are not
available. Furthermore, the random flights and IRT simulation techniques are computer-
intensive. T h u s far calculations have been limited to idealized track segments of high energy
electron tracks, to low energy track-ends and to spur size distributions ignoring the effects of
track-ends.

6 C O M M E N T S AND DISCUSSION

Stochastic and deterministic techniques for probing the short-time chemistry of

spatially nonhomogeneous systems found in high energy electron radiolysis have been
described in detail, and the use of these techniques to examine a variety of interesting problems
has been discussed. T h e two different approaches are based on very different assumptions. The
deterministic models consider a "typical, average" spur and describe the chemistry of this spur
using a set of coupled diffusion-reaction equations and spatially nonhomogeneous
concentration profiles for the reactants. The stochastic treatments recognize the individual
identities of the reacting particles and employ statistically correct methods to model the
kinetics of clusters with realizable initial configurations.

T h e results presented demonstrate that the deterministic treatment of the radiolysis of

aqueous solutions provides a suitable method for analyzing experimental data. The
deterministic approach is simple to understand and to use, and is sufficiently flexible that it can
be applied to complex reaction mechanisms in a facile manner. A single set of consistent
parameters gives wide agreement with an extensive set of experiments using scavengers. In
addition to being easily used, the deterministic models provide a useful method for testing the
some of the approximations employed in the analysis of experimental data, e.g. the Laplace
transform relationship between the yields in scavenger experiments and the underlying decay
and formation kinetics of the radiation-induced species. While the deterministic approach
provides a simple and useful model for predicting the radiation chemical kinetics of aqueous
systems, the track of a fast electron is condensed into an average spur with a few characteristic
170

parameters. Consequently, the modeled kinetics are insensitive to the detailed structure of the
track and to the spur size distribution.

Deterministic analyses of the nonhomogeneous kinetics of small clusters of reactants

are subject to a number of limitations. The most important of these is the incorrect
fractionation of the products. The predictions of deterministic calculations and stochastic
simulations for identical spurs have been compared. Studies of spurs made up of hydrogen
atom - hydroxyl radical pairs and of hydrated electron - hydrated proton - hydroxyl radical
groups show that for small spurs ( 1 , 2 or 3 ionization / excitation events) a deterministic
analysis gives very different product yields from those suggested by the statistically correct
stochastic methods.54,122 p o r larger spurs there is very little difference between the
predictions of the two different approaches unless spatial correlations are important.

In contrast to the conventional deterministic analyses of short time radiation chemical

kinetics, stochastic analyses are sensitive to the distribution of spurs, to their contents and to
the spatial distribution of the reactants. This sensitivity reflects the fact that the stochastic
models recognize the individual particles. Stochastic calculations highlight the chemical
consequences of track structure and of the detailed spatial correlations of the reactants. In
particular, they emphasize the effect of clustering on the ratios of the products of
nonhomogeneous reaction. As stochastic techniques consider the individual spur particles and
their specific nature they can be used to study phenomena averaged by the macroscopic rate
used in deterministic calculations. For instance they have been used to examine the effects of
spin correlations within a spur and the influence of an applied field upon the escape of ions
from multiple ion pair spurs.

T h e random flights and IRT simulation methods are (computationally) complex and the
methodologies require careful testing before application to and consideration of chemically
realistic systems. Furthermore, the formulation of a stochastic master equation for complex
reaction schemes is difficult as each possible state (set of contents) of the spur has to be
considered. Clearly while stochastic models have a number of limitations which must be
overcome in future studies, these methods provide many potentially important insights which
cannot be obtained from deterministic methods.

Acknowledgment
T h e authors wish to thank Dr. P. Clifford, Dr. A. Mozumder, Dr. J. A. LaVerne and
Prof. M. J. Pilling for numerous interesting and informative discussions. T h e research
described was supported by the Office of Basic Energy Sciences of the U . S . Department of
Energy. This is contribution SR-165 of the Notre Dame Radiation Laboratory.
 171

REFERENCES

(1) Wilson, C. T. R. Proc. Roy. Soc. 1 9 2 3 , 1 0 4 A , 1.

(2) Wilson, C. T. R. Proc. Roy. Soc. 1 9 2 3 , 1 0 4 A , 192.
(3) Kara-Michailova, E.; Lea, D. E. Proc. Camb. Phil. Soc. 1 9 4 0 , 3 6 , 1 0 1 .
(4) Samuel, A. H.; Magee, J. L. / . Chem. Phys. 1 9 5 3 , 2 5 , 1 2 9 .
(5) Schwarz, H. A. J. Phys. Chem. 1969, 73,1928.
(6) M o z u m d e r , Α.; Magee, J. L. Radiat. Phys. Chem. 1975, 7, 8 3 .
(7) Bethe, Η. Z. Phys. 1932, 76,293.
(8) Allen, A. O. The Radiation Chemistry of Water and Aqueous Solutions; Van Nostrand:
N e w York, 1 9 6 1 .
(9) Ashley, J. C. J. Electron Spectrosc. Relat. Phenom. 1 9 8 8 , 4 6 , 199.
(10) LaVerne, J. Α.; Mozumder, A. J. Phys. Chem. 1986, 90, 3242.
(11) Ashley, J. C ; Williams, M. W . R o m e Air Development Center, Griffiss Air Force Base,
N Y , 1983.
(12) Green, N . J. B.; LaVerne, J. Α.; Mozumder, A. Radiat. Phys. Chem. 1988, 32, 99.
(13) Pimblott, S. M.; LaVerne, J. A. J. Phys. Chem. 1991, 95, 3 9 0 7 .
(14) M o z u m d e r , Α.; Magee, J. L. J. Chem. Phys. 1966, 45, 3332.
(15) Pimblott, S. M.; LaVerne, J. Α.; Mozumder, Α.; Green, N . J. B . / . Phys. Chem. 1990,
94, 4 8 8 .
(16) Zaider, M.; Brenner, D. J.; Wilson, W. E. Radiat. Res. 1983, 9 5 , 2 3 1 .
(17) Paretzke, H. G.; Turner, J. E.; H a m m , R. N.; Wright, Η. Α.; Ritchie, R. H. J. Chem.
Phys. 1986, £ 4 , 3 1 8 2 .
(18) Paretzke, H. G.; Turner, J. E.; H a m m , R. N.; Ritchie, R. H.; Wright, H. A. Radiat. Res.
1991,127, 121.
(19) Hill, Μ. Α.; Smith, F. A. Radiat. Phys. Chem. 1994, 43, 2 6 5 .
(20) La Verne, J. Α.; Pimblott, S. M.; Mozumder, A. Radiat. Phys. Chem. 1991, 38, 7 5 .
(21) Sanche, L. In Excess Electrons in Dielectric Media; C. Ferradini and J.-P. Jay-Gerin,
Ed.; C R C Press: Boca Raton, 1991.
(22) LaVerne, J. Α.; Pimblott, S. M . Radiat. Res 1 9 9 3 , 1 3 5 , 16.
(23) Nikjoo, H.; Goodhead, D . T.; Charlton, D. E.; Paretzke, H. G. Physics in Medicine
and Biology 1 9 8 9 , 3 4 , 6 9 1 .
(24) G o o d h e a d , D . T.; Nikjoo, H. Int. J. Radiat. Biol. 1 9 8 9 , 5 5 , 5 1 3 .
(25) Turner, J. E.; Magee, J. L.; Wright, Η. Α.; Chatterjee, Α.; H a m m , R. N.; Ritchie, R. H.
Radiat. Res. 1983, 96, 4 3 7 .
(26) Zaider, M.; Brenner, D. J. Radiat. Res. 1 9 8 4 , 1 0 0 , 2 4 5 .
(27) Brenner, D . J.; Zaider, M . Radiat. Prot. Dosim. 1 9 8 5 , 1 3 , 111.
(28) Clifford, P.; Green, N. J. B.; Pilling, M. J.; Pimblott, S. M.; B u m s , W . G. Radiat. Phys.
Chem. 1 9 8 7 , 3 0 , 125.
(29) Turner, J. E.; H a m m , R. N.; Wright, Η. Α.; Ritchie, R. H.; M a g e e , J. L.; Chatterjee, Α.;
Bolch, W . E. Radiat. Phys. Chem. 1 9 8 8 , 3 2 , 5 0 3 .
(30) Pimblott, S. M.; Pilling, M. J.; Green, N. J. B. Radiat. Phys. Chem. 1991, 37, 311.
(31) LaVerne, J. Α.; Mozumder, A. Radiat. Res. 1 9 9 3 , 1 3 3 , 2 8 2 .
(32) Pimblott, S. M.; Mozumder, A. J. Phys. Chem. 1991, 9 5 , 7 2 9 1 .
(33) Baker, G. J.; Brocklehurst, B.; Hayes, M.; Hopkirk, Α.; Holland, D. M . P.; M u n r o , I.
H.; Shaw, D. A. Chem. Phys. Lett. 1 9 8 9 , 1 6 1 , 327.
172

(34) Long, F. Η.; Lu, Η.; Eisenthal, Κ. Β. Phys. Rev. Lett. 1990, 64, 1469.
(35) Long, F. H.; Lu, H.; Shi, X.; Eisenthal, Κ. B. Chem. Phys. Lett. 1991, 185, 4 7 .
(36) Pimblott, S. M. / . Phys. Chem. 1991, 95, 6946.
(37) Sander, M. U.; Luther, K.; Troe, J. J. Phys. Chem. 1993, 97, 11489.
(38) Gauduel, Y.; Pommeret, S.; Migus, Α.; Antonetti, A. Chem. Phys. 1990, 749, 1.
(39) Burns, W. G.; Curtis, A. R. J. Phys. Chem. 1972, 76, 3008.
(40) M o z u m d e r , Α.; Magee, J. L. Radiat. Res. 1966, 28, 2 1 5 .
(41) Burns, W . G.; Sims, Η. E.; Goodall, J. A. B. Radiat. Phys. Chem. 1984, 23, 143.
(42) LaVerne, J. Α.; Pimblott, S. M. / . Phys. Chem. 1991, 9 5 , 3196.
(43) T r u m b o r e , C. N.; Short, D. R.; Fanning Jr., J. E.; Olson, H. J. Phys. Chem. 1978, 82,
2762.
(44) Chance, Ε. M.; Curtis, A. R.; Jones, I. P.; Kirby, C. P. " F A C S I M I L E : A computer
program for flow and chemistry simulation, and general initial value problems," A E R E ,
Harwell, 1977.
(45) Lea, D. E. Actions of Radiation on Living Cells; C.U.P.: Cambridge, 1955.
(46) Lea, D. E. Proc. Camb. Phil. Soc. 1934, 30, 80.
(47) Green, N. J. B.; Pimblott, S. M. Molec. Phys. 1991, 74, 8 1 1 .
(48) Green, N. J. B.; Pimblott, S. M. Molec. Phys. 1991, 74, 7 9 5 .
(49) Green, N. J. B.; Pilling, M. J.; Pimblott, S. M. Radiat. Phys. Chem. 1989, 34, 105.
(50) Bartczak, W . M.; Hummel, A. J. Chem. Phys. 1987, 87, 5222.
(51) Green, N. J. B.; Pilling, M. J.; Pimblott, S. M.; Clifford, P. J. Phys. Chem. 1989, 93,
8025.
(52) Clifford, P.; Green, N. J. B.; Pilling, M. J.; Pimblott, S. M. J. Phys. Chem. 1987, 9 7 ,
4417.
(53) Clifford, P.; Green, N. J. B.; Pilling, M. J. J. Phys. Chem. 1982, 86, 1322.
(54) Clifford, P.; Green, N. J. B.; Oldfield, M. J.; Pilling, M . J.; Pimblott, S. M . J. Chem.
Soc, Faraday Trans. 1986, 82, 2 6 7 3 .
(55) Pimblott, S. M. J. Phys. Chem. 1992, 96, 4 4 8 5 .
(56) Noyes, R. M . Prog. React. Kinet. 1961, 7, 129.
(57) Clifford, P.; Green, N. J. B.; Pilling, M. J. J. Phys. Chem. 1982, 86, 1318.
(58) Green, N. J. B.; Pimblott, S. M. / . Phys. Chem. 1990, 94, 2 9 2 2 .
(59) Pimblott, S. M. J. Chem. Soc, Faraday Trans. 1993, 89, 3 5 3 3 .
(60) Green, N. J. B . Molec. Phys. 1987, 65, 1399.
(61) Pimblott, S. M.; Green, N. J. B. J. Phys. Chem. 1992, 96, 9 3 3 8 .
(62) Milshtein, G. N. Theory Prob. Appl. 1 9 7 8 , 2 3 , 396.
(63) Rumelin, W. SIAM J. Numer. Anal. 1982, 79, 604.
(64) Pimblott, S. M. D. Phil. Thesis, Oxford University, 1988.
(65) Rice, S. A. Diffusion-limited Reactions; Elsevier: Amsterdam, 1985; Vol. 2 5 .
(66) Chandrasekhar, S. Rev. Mod. Phys. 1943, 75, 59.
(67) Green, N. J. B.; Pimblott, S. M. / . Phys. Chem. 1989, 93, 5 4 6 2 .
(68) Collins, F. C ; Kimball, G. E. J. Colloid Sci. 1 9 4 9 , 4 , 4 2 5 .
(69) Flannery, M. R.; Mansky, E. J. Chem. Phys. 1 9 8 9 , 1 3 2 , 115.
(70) M a g e e , J. L ; Tayler, A. B . J. Chem. Phys. 1 9 7 2 , 5 6 , 3 0 6 1 .
(71) M o z u m d e r , A. J. Chem. Phys. 1974, 61,780.
(72) Pedersen, J. B.; Sibani, P. J. Chem. Phys. 1981, 75, 5 3 6 8 .
(73) Rice, S. Α.; Baird, J. K. J. Phys. Chem. 1978, 69, 1989.
 173

(74) Pedersen, J. Β.; Larsen, J. E. J. Chem. Phys. 1982, 77, 1061.

(75) Green, N . J. B . Chem. Phys. Lett. 1984,107, 4 8 5 .
(76) Clifford, P.; Green, N. J. B.; Pilling, M. J. J. Phys. Chem. 1984, 88, 4 1 7 1 .
(77) Ito, K.; M c K e a n , H. P. Diffusion Processes and their Sample Paths; 2nd ed.; Springer:
Berlin, 1974.
(78) Devroye, L. Non-uniform Random Variate Generation; Springer-Verlag: N e w York,
1986.
(79) H o n g , Κ. M.; Noolandi, J. J. Chem. Phys. 1978, 68, 5 1 6 3 .
(80) Noolandi, J. In Kinetics of Nonhomogeneous Processes. A Practical Guide for
Chemists, Biologists, Physicists and Material Scientists; G. R. Freeman, Ed.; Wiley-
Interscience: N e w York, 1987.
(81) Umberger, J.; LaMer, V. J. Am. Chem. Soc. 1945, 6 7 , 1099.
(82) Green, N. J. B.; Pilling, M . J.; Clifford, P. Molec. Phys. 1989, 67, 1085.
(83) Green, N. J. B.; Pilling, M. J.; Pimblott, S. M.; Clifford, P. J. Phys. Chem. 1990, 94,
251.
(84) Green, N . J. B.; Pimblott, S. M.; Brocklehurst, B. J. Chem. Soc, Faraday Trans.
1991, 87, 2 4 2 7 .
(85) Pimblott, S. M.; Green, N . J. B.; Brocklehurst, B . J. Chem. Soc, Faraday Trans.
1991, 87, 3 6 0 1 .
(86) Chernovitz, A. C ; Jonah, C. D. J. Phys. Chem. 1988, 92, 5946.
(87) Jonah, C. D.; Miller, J. R. J. Phys. Chem. 1977, 81, 1974.
(88) Elliot, A. J.; McCracken, D. R.; Buxton, G. V.; W o o d , N . D. / . Chem. Soc, Faraday
Trans. 1990, 86, 1539.
(89) Pulse Radiolysis; Tabata, Y., Ed.; C R C Press: Boca Raton, 1990.
(90) Buxton, G. V. In Radiation Chemistry. Principles and Applications; Farhataziz and M.
A. J. Rodgers, Ed.; V C H Publishers: New York, 1987.
(91) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B . J. Phys. Chem. Ref.
Data 1988, 7 7 , 5 1 3 .
(92) Pimblott, S. M.; La Verne, J. A. Radiat. Prot. Dosim. 1994, 52, 183.
(93) Balkas, Τ. I.; Fendler, J. H.; Schuler, R. H. J. Phys. Chem. 1970, 74, 4 4 9 7 .
(94) Pimblott, S. M.; La Verne, J. A. Radiat. Res. 1992, 729, 2 6 5 .
(95) Abramowitz, M.; Stegun, I. A. Handbook of Mathematical Functions; Dover: N e w
York, 1970.
(96) Pimblott, S. M.; LaVerne, J. A. J. Phys. Chem. 1992, 96, 8904.
(97) LaVerne, J. Α.; Pimblott, S. M. J. Chem. Soc, Faraday Trans. 1993, 89, 3 5 2 7 .
(98) Draganic, I. G.; Draganic, Z. D. J. Phys. Chem. 1973, 7 7 , 7 6 5 .
(99) Pimblott, S. M.; Schuler, R. H.; LaVerne, J. A. J. Phys. Chem. 1992, 96, 7 8 3 9 .
(100) Schuler, R. H.; Hartzell, A. L.; Behar, B. J. Phys. Chem. 1981, 85, 192.
(101) Elliot, A. J.; Chenier, M. P.; Ouellette, D. C J. Chem. Soc Faraday Trans. 1993, 89,
1193.
(102) Christensen, H.; Sehested, K. J. Phys. Chem. 1986, 90, 186.
(103) Sehested, K.; Christensen, H. Radiat. Phys. Chem. 1 9 9 0 , 3 6 , 4 9 9 .
(104) LaVerne, J. Α.; Pimblott, S. M. J. Phys. Chem. 1993, 97, 3 2 9 1 .
(105) McQuarrie, D. A. J. Appl. Prob. 1 9 6 7 , 4 , 143.
(106) M o z u m d e r , A. J. Chem. Phys. 1971, 5 5 , 3020.
(107) Dodelet, J.-P.; Freeman, G. R. Radiat. Phys. Chem. 1975, 7, 183.
174

(108) Green, Ν. J. Β.; Pimblott, S. Μ. Radiat. Phys. Chem. 1991, 37, 161.
(109) Delaire, J. Α.; Croc, E.; Cordier, P. J. Phys. Chem. 1981, 85, 1549.
(110) G e e , N.; Freeman, G. R. J. Chem. Phys. 1992, 96, 586.
(111) G e e , N.; Freeman, G. R. J. Chem. Phys. 1987, 86, 5716.
(112) G e e , N.; Senanayake, P. C ; Freeman, G. R. J. Chem. Phys. 1988, 89, 3710.
(113) Schmidt, W . F.; Allen, A. O. J. Chem. Phys. 1 9 7 0 , 5 2 , 2 3 4 5 .
(114) Pimblott, S. M.; Mozumder, Α.; Green, N. J. B. J. Chem. Phys. 1989, 90, 6 5 9 5 .
(115) Onsager, L. Phys. Rev. 1 9 3 8 , 5 4 , 554.
(116) Saltiel, J.; Atwater, B. W. Adv. Photochem. 1 9 8 8 , 1 4 , 1.
(117) Pedersen, J. B.; Freed, J. H. J. Chem. Phys. 1 9 7 3 , 5 8 , 2746.
(118) Brocklehurst, B . Int. Rev. Phys. Chem. 1 9 8 5 , 4 , 219.
(119) Brocklehurst, B . Radiat. Phys. Chem. 1989, 34, 5 1 3 .
(120) Brocklehurst, B . Faraday Discuss. Chem. Soc. 1984, 78, 3 0 3 .
(121) Green, N . J. B.; Harris, R. Chem. Phys. Lett. 1 9 9 2 , 1 9 8 , 8 1 .
(122) Pimblott, S. M.; LaVerne, J. A. Radiat. Res. 1990, 722, 12.
Research in Chemical Kinetics, Volume 3
R.G. Compton and G. Hancock (editors) 175
© 1995 Elsevier Science B.V. All rights reserved.

Effects of low frequency electromagnetic fields on crystal growth from

solutions

Dario T. Beruto and Marino Giordani

Interdepartmental Centre for Materials Engineering and Institute of Chemistry,

Faculty of Engineering, University of Genoa
Piazzale J. F. Kennedy, Fiera del Mare Pad. D, 16129 Genova, Italy

1. INTRODUCTION

The effects of electromagnetic fields on the behaviour and the properties of

matter have long been investigated since the corner-stone works by Faraday and
Maxwell. When the intensities and the frequencies of such fields are low (ELF-
fields), they often induce unique and puzzling effects. Galvani was the first to
notice that a weak electrical current causes changes in the functional properties
of muscles and in the precipitation of tartrate salts from a hydroalcoholic
solution. The evidence that weak fields can affect the behaviours of both living
[1] and non-living materials [2] is fascinating, and it is believed that this effect
occurs because ELF fields can modify mass-transport phenomena and the related
reactions of particles and/or ions, which play an important role in the overall
process.
It is surprising that sometimes ELF fields have noticeable effects in
comparison with the small amount of energy supplied by the applied fields. This
aspect has not been adequately explored yet. To investigate these problems, one
should try to know the nature and the role of the weak forces exerted by ELF
fields on matter, as well as the kind of cooperative steps that take part in the
general process. This is a difficult task and recently became the focus of
interdisciplinary research activities on biological systems [3-10]. By contrast, the
effects of these weak fields on the behaviour of inorganic materials have not been
extensively studied, although this research may provide very useful bases for
interpreting and predicting the properties of more complex systems, like
biological ones.
This paper tries to compensate for this back off of information by dealing with
the effects of low-intensity and electromagnetic low-frequency (ELF) fields on the
nucleation and growth processes of inorganic crystalline salts from solutions.
When ELF fields are applied to solutions and/or to suspensions and/or to
colloid systems, such media may become a space where a non-uniform electric
176

field is generated even if the associate magnetic field remains uniform. Under
the action of the non-uniform electric field charged particles and neutral ones can
move from lower-field regions to higher-field ones. This motion might give rise to
local differences in the spatial distribution of the particles in their ambient
medium. If aggregation phenomena and/or reactions between the components
occur, the compositions of the particle ensembles and the particle size might
change.
Applications of this theory to various branches of science are possible and of
great interest. An excellent book on dielectrophoresis (i. e., the motion induced by
non-uniform electric field in neutral matter), has been published [11]. The
phenomena described in this paper cannot be depicted as a dielectrophoresis
process; nevertheless some considerations about the relationships between the
particle size and the effects of non-uniform fields on particle dynamics can be
useful also for our systems.
According to the pioneering works by Wrede [12] and to those by the
Princenton and Cornel groups [13-23], it is possible to assume that the effects of
non-uniform fields are negligible at the molecular level and progressively
increase with the particle size.
When this general assumption is applied to the interactions between ELF-
fields and the ions, nuclei, clusters and fine particles suspended in the liquid
phase during the precipitation process, it is difficult to understand and explain
the effects of low-intensity exogenous fields, as their action is further reduced by
the medium size of the particle.
Accordingly, it becomes important both to define the variables that affect the
nucleation and the growth of the solid phase from the liquid one and to establish
which parameter may be significantly affected by the low-energy exchanges
between a system and ELF-fields. Furthermore, it is mandatory to design
experiments that may allow one to observe particle microstructures due to the
effect of an ELF-field and not to spurious contributions.
The purpose of the present article is to make a survey of these topics, with
special emphasis on our recent results in this field and on new data that are
interesting and useful in understanding the mechanism of mineral formation in
bone.

2. THEORY

The sequence of steps involved in the nucleation, crystallization and

precipitation processes of a solid product in a solution to which a precipitating
agent has been added is shown in Figure 1.
 177

SOLUTION

PRECIPITATING A G E N T

SUPERS^lTURATION

NUCLEATION

CRYSTAL G R O W T H

SEPARATION

SOLID P R O D U C T

Figure 1. Sequence of the steps involved in the nucleation, crystallization and

precipitation processes in a solution

In the medium formed by the initial solution and the added precipitating
agent, different kinds of cations and anions may interact both with water
molecules and with one another. If the solvent is water, the interactions between
charged atoms and polar molecules are caused by strong electrostatic forces of
attraction. These may lead to the formation of hydrated ions, whose nature and
structure are rather stable and responsible for all static and kinetic properties of
178

the solutions (viscosity, ionic mobility, diffusion, etc.) [24]. The nature of these
species is certainly affected by the nature of the solvent, which generally leads to
a solvation process. Once hydrated and/or solvated ions are formed, a set of
dipoles appear in the medium. The interactions of these species with an applied
electromagnetic field can be described, as a first approximation, by the
electromagnetic Lorenzt force. The dipolar part of this force is proportional to the
volume of the dipoles, and, according to previous findings [2], should not be
significant, as the volume of the dipoles is determined by the coordination
number of solvent molecules around the ions which should not to exceed the
molecular size. Furthermore, when an ELF field is applied, the value of such a
force is necessarily small, as the value of the applied field is small (B « 3 mT or
less). Accordingly, if one compares such a force with the strong forces responsible
for the formation of hydrated and/or solvated ions, it seems reasonable to
conclude that ELF fields cannot affect the nature and the structure of
hydrated I solvated ions.
In a static solution, the movement of solvated ions is described only by a
random walk process. For the sake of simplicity, let us assume that each solvated
ion is a sphere of radius r. Manipulations of the Fick diffusion equation [25] lead
to the well-known relationships:

D = kT/67rnr (1)

and

< 2 > = 2Dt

X (2)

where D is the species diffusion coefficient, k is the Boltzmann constant, Τ is the

2
absolute temperature, η is the viscosity coefficient, <x > is the average diffusion
distance of the species, and t is the time.
When an electromagnetic field is applied, the average diffusion distance
changes because of the Lorentz force, and, under some of assumptions, it has
been demonstrated by Kinouchi et al. [25] that an approximate solution, for t »
1, is equal to :
2
< 2>
X = 2[D/(l+b )]t (3)

where b is a parameter can be expressed as:

b= qB/(67rrtr) (4)

where q is the particle charge and Β is the applied field.

From equations (3) and (4) it is possible to derive the value of the field for a
known system that has suppressed the average diffusion distance. Estimated
values of Β that, for several substances, like calcium ions, lead to the suppression
 179

of the average diffusion distance indicate that the threshold field strength is of
the order of 10 T. Hence low-intensity ELF fields should not induce any
6

variation in the average diffusion path of the reactants. Accordingly, the average
probability of collision between cations and anions should not be affected by the
action of ELF fields

Figure 2. Ionic embryo model for ions in a solvent cavity.

The reaction mechanisms associated with the nucleation and the growth of the
solid product formed by the reactants in solution need therefore to be
180

investigated to understand the effects of ELF fields on crystal growth from

solutions.
These processes have been the subjects of many and important papers [26-28],
and it seems appropriate outline here the most important points of those theories
in order to decide on which reaction steps the ELF fields are acting.
Let S be a formed solid product and let us consider only homogeneous
nucleation processes. Before reaching a well-defined crystalline shape, the solid
product appears as a small nucleus. In order to grow, nuclei need to reach a
critical size. An experimental and observable nucleus of the solid product formed
from solution can be regarded as a droplet of critical size, made up of 50-250
atoms and/or ions. Let us concentrate on nuclei formed by ions, as they are closer
to the nuclei of many inorganic salts.
The building-blocks of these nuclei (i. e., embryos) can be described at least by
two models: ions in a cavity and diffused ions [29].
The formation of ions in a cavity requires strong interactions among ions so
that the internal energy lattice solvation may be overcome and the configuration
of the building block may become that of a small piece of crystal in a solvent
cavity (Figure 2). In this configuration, the ions have a geometry that is very
close to the final bulk one, but the molecular volume cannot be considered to be
equal to that of the final crystallite.

Figure 3. Model describing diffuse-ion embryos.

 181

In the formation of diffuse-ion embryos, the ionic forces between cations and
anions are not strong enough to overcome the solvation forces so that the
structure of solvated ions remains virtually unchanged in the building-block. As
a result, the embryo configuration can be depicted with a number of solvated ions
equal to the number required by the unit lattice salts spread over a diffuse
volume V (Figure 3).

These embryos have to be regarded as temporary groupings of ions analogous

to those considered by the theory of associated solutions, which by statistical
fluctuations get together according to the reaction :

n S ( l ) = Sn(s) (5)

where S(l) is a single embryo in the liquid phase (1) and S (s) is the solid nucleus
n

formed by η embryos.
How reaction (5) occurs is an intriguing problem. To comply with
thermodynamic principles, the free-energy change in such a reaction has to be
less than zero, but this requirement is not always met. The free-energy value is a
function of the number η of embryos and of the shape of the nucleus, as follows:

AG = n A G + ί σ b (6)

where AGb is the bulk free-energy change per embryo, η is the number of
embryos forming the droplet, f is a shape factor of the droplet, and σ is the
interfacial energy per unit surface area of the droplet. The first term in eq. (6) is
negative, as by convention, bond formation releases energy in the environment,
whereas the second term is positive, as energy needs to be done on the system in
order to form a solid/liquid interface. For a spherical nucleus of radius r formed
by η embryos (each with a volume equal to v), if the ions-in-a-cavity model is
applied, it is possible to write:

η = (4π/3) χ (r3/v) (7)

It should be noted that equation (7) is dependent on the shape of the nucleus,
and that it might become rather difficult to write it for a diffused-ion embryo
model.
By combining equations (6) and (7), one gets:

AG = η 2 7 3
Φ + nAG b (8)

where Φ is proportional to the interfacial free energy per embryo. The first term
of eq. (8) is always positive, whereas the second is always negative. For values of
η very close to zero, the change in free energy is positive, whereas it becomes
negative only for increasing values of n. Accordingly, reaction (5) is not
182

thermodynamically possible for all values of n, smaller than a critical one, which
can be derived from the condition:

QAG/3n) * = 0 n=n (9)

By solving equation (9), one obtains :

3 2 3
n* = h [ ( a v )/(AG ) ] b (10)

h being a numerical value equal to 32π/3 for a spherical nucleus.

The obtained result defines a known phenomenon concerning the precipitation
from a supersaturated solution. It is well known that a supersaturated solution
completely free from dust particles does not yield any precipitate as long as the
supersaturation remains below a critical limit. Above this limit value, a phase
change occurs spontaneously. This behaviour is related to the increased
solubility of very small crystals which is predicted by the Kelvin equation. In
order to deal with a solid phase as distinguished by the liquid one, i. e., to take
into account the solid-liquid interfacial energy contributions, one has to consider
solid phases formed by a number of embryos larger than n*.
A puzzling question is how embryos can reach this critical size, as the free
energy change for reaction (5) in that region is positive. To solve this problem,
one has to assume that, in this field, reactants are only in a liquid state, for a
solid state cannot exist. This leads to the approach derived from the molecular
association theory as applied to liquid states [30]. Embryos can form groups of
polymers by means of a set of molecular steps described by the general equation:

η SO) S (l)
n (11)

This reaction does not imply the formation of any interfaces and gives rise to
an entropy of mixing, which is completely neglected in reaction (5). The mixing
mechanism can provide the driving force that allows the reaction to occur.
Reaction (11) can proceed up to the critical grouping cluster component defined
by η = η*, which is entitle to give the phase transformation :

S *(l)
n <=> S *(s) n (12)

A detailed description of embryo dynamics was given by I. Prigogine et al. [33].

From these studies, it is possible to deduce that nucleation kinetics can be
expressed as:

J N = Ω (4 σ / k T ) ^ χ ρ [-(AG*/ kT)]
θ (13)

where Ω is a variable including frequency factors and AG* is the free-energy

change in eq. (5) when η is equal to n*. This equation shows that the free energy
 183

of formation of a critical nucleus plays a dominant role in determining the rate of

nucleation. This value is easily derived, for η = η*, from equations (10) and (8)
as:
3 2 2
AG* = h ( a v )/(AG ) b (14)

where the contributions of surface-term are neglected, as the nuclei have now
reached the critical size and, under this condition, the terms describing the bulk
free-energy change are dominant. The bulk free-energy change can be derived
from the well-known Gibbs-Kelvin equation as a function of the degree of
saturation, as follows:

AG = - ( m k T / v ° ) l n s
b (15)

where m is the number of ions in the neutral salt and v° is the molecular volume.
If v° is assumed to be almost equal to the molecular volume of an embryo, then
the nucleation rate can be written as:
2 3 4 3 2
J N = Ω (4 σ / kT)^ exp - [h (σ ν ) / (kT) (m In s*) ] (16)

This nucleation rate is very slow for values of s less than s*, and rapidly
increases when s approaches a critical value, which is obtained when nuclei of
critical radii are formed.
Any possible discussion and understanding of the effects of ELF fields on the
nucleation process should therefore start from an analysis of the different
variables that appear in equation (16). This equation is the result of a particular
and appropriate way of dealing with the complex mechanisms associated with
the dynamics of embryos up to the formation of nuclei of critical size that allow
the next crystal-growth steps. Furthermore, the same equation was tested
experimentally [31-33] and a close agreement between theoretical and
experimental results was obtained.
In principle, the variables that play an important role in eq. (16) and that
might be affected by ELF fields at constant temperature are:
1) the molecular volume of embryo and/or of the neutral salt;
2) the degree of supersaturation s*;
3) the interfacial energy of a nucleus of critical size σ.
Let us discuss these variables one by one.
As far as the influence of ELF fields on the molecular volume of an embryo is
concerned, it should be noted that the value of ν is dependent upon the force
balance that holds for an ion-embryo configuration with opposite charges. These
forces are very strong, as they are ionic in nature. It is hard to believe that low
electromagnetic Lorentz forces, like the one exerted by ELF fields, may affect the
structure of an ions-embryo configuration.
184

As far as the influence of ELF fields on the degree of supersaturation s is

concerned, the low electromagnetic forces should be able to affect the actual
concentration of ions in solution and/or the solubility product.
To change the actual concentration of the ions in solution the diffusion
patterns of the correspondent ions should be affected by ELF fields.
As we reported previously this possibility is far away to be satisfied by low
intensity ELF fields.
Consequently the only variable that in eq. (16) may be affected by an ELF field
is the interfacial energy between the solid nucleus of critical radius and the
liquid phase.
To explore how this can happen let us start with a discussion of the magnitude
of surface energy in ionic materials. The specific surface-energy values of these
salts are proportional to their Young modulus according to:

o sv =Ed°/7c 2
(17)

where Ε is the Young modulus, d° is the equilibrium separation distance among

atoms, and the symbol sv concerns the solid/vapour interface.
The specific solid/liquid energy interfaces, σ ι (si = solid/liquid), can be related to
8

the ones concerning σ through the adsorption of the liquid phase on the solid
8ν

nuclei by the equation:

σι δ = σ 8ν + Agads (18)

where A g is the free-energy change for the liquid-solid adsorption step

a d s

2
expressed per cm of nucleus surface.
For an exothermic adsorption step Ag ds is negative, therefore, o will be less
a s]

than the work a . sv

The heat of adsorption of liquid water on an ionic solid surface barely exceeds -
1
34.3 kJ mol" [34]. Neglecting the entropy terms and assuming that a water
2
molecule has a cross-section equal to about 0.1 nm [35], the change in free
2 2
energy per cm can be estimated to be of the order of 50 erg cm" . Accordingly,
the values of a j can be derived from the known values of o [36].
s sv

Table 1 presents the results. In the same table (last column), the
corresponding value of the free energy per nucleus of critical size is given,
assuming that 100 atoms form a critical nucleus according to a f.c.c. lattice. It is
worth noting that these values are large enough to overcome the thermal
disorder term kT which, at Τ = 300 Κ, is equal to about 26 meV, but that they do
not exceed a few hundred meV.
This means that solid/liquid interfaces cannot be destroyed by thermal
fluctuations, but that they might be affected by low ELF fields of the same
magnitude.
To evaluate the magnitude of the effect produced by an ELF field let us to
refer to a simple model like the one illustrated in Figure 4a, b.
 185

(b)

Figure 4. Model to illustrate how ELF field can strain the solid/liquid interface of
a nucleus of critical size.

Generally speaking, the surface of the solid/liquid interface nucleus can be

described as a surface characterized by a uniform charge density (Fig. 4a). When
186

an external electric field φ is applied, the charge density is modified by a

polarization effect (Fig. 4b). If the applied field has a low frequency, the new
configuration should be stable over a certain period of time. Because of this
charge redistribution, the surface is strained, so the final interfacial energy value
will be equal to the initial interfacial energy σ plus the strain energy due to the
δ1

deformation.

Table 1
Solid/vapour and solid/liquid specific surface free energy.

Experimental Calculated Calculated

Compound σ ι for nucleus of critical size
8
2 2
(erg cm" ) (erg cm" ) (meV)

CaC0 3
230 157 125
NaCl 300 227 180
LiF 340 267 212
BaF 2
280 207 165
CaF 2
450 377 300

This last term can be calculated as a vectorial product of the induced strain ε
and the corresponding stresses 7.
To calculate the stresses corresponding to the strains caused by the field, an
approach equal to the one recently developed by one of the authors [37] can be
used.

Let Δη be the charges per unit area that can be introduced or removed from
ς

the interface by the electric field φ. When the field does not affect the total
amount of charges but it simply redistributes them, Ariq might be regarded as the
difference between the maximum and the minimum charge densities at the
interface. In any case, the surface will be strained by a two-dimensional state of,
E , equal to :
q

e q = Δη Ω L ς 8 s (19)

where Ω is the area occupied by a unit charge and L is a two-dimensional shape

8 s

tensor accounting for the symmetry of the charge with respect to the surface
lattice. For now, the effect of orientating the charge distribution will be neglected
 187

and the state of strain £q will be considered isotropic. To keep the surface layer
joined to the underlying bulk, an elastic strain e j will oppose the inelastic one
e

and the resulting effective strain will be given by :

ε = ε ς + ει β (20)

Then, the strain energy interface, W , associated with the deformation will be
s

equal to:

W = 4 π r 7ε
s
2
(21)

where r is the radius of the nucleus of critical size and γ is the stress
corresponding to the inelastic strain caused by the electric field φ. The two-
dimensional state of stress γ is further associated with the state of elastic strain
ε by [38]:
β1

γ =[E d / l - < o ]
S s s £ e l (22)

where E and υ are, respectively, the Young and Poisson moduli of the solid-
s 8

liquid interface, and d is the thickness of the surface layer.

s

It can be shown [38] that, on account of the spherical symmetry and of the
well-known Laplace equation, the actual strain is a function of the charge-
induced inelastic strain according to:

ε = [ l / ( l + aXU q (23)

where a is a constant that is dependent upon microstructural properties of the

nuclei and that can be taken approximately equal to α « r/ds.
By combining equations (20)-(23), it is easy to obtain :

W = 4π r θe
s
2
q
2
(24)

where θ is a constant equal to :

2
θ = [α / (1 + α) ] [E d / (1 - υ )] s s 8 (25)

The values of the Young and Poisson surface moduli in eq. (25), at least as
orders of magnitude, can be taken equal to the ones tabulated for the bulk
phases. Experimental evidence [39] concerning the rheological behaviour of
dispersed aqueous systems allows us to establish that a reasonable value of d iss

equal to around 1-2 nanometers. Accordingly, for a critical nucleus of about 100
1
atoms, the value of the constant θ turns out to be of the order of 10" .
188

To be a plausible experimental variable, the value of the strain energy

associated with the effect of the electrical field has to exceed 20 meV, i. e., the
value of the term kT at 300 K. Application of equation (24) yields a value 50 meV
2
when the inelastic strain e is equal to 10" . This deformation should be produced
q

by the local electrical field Φ applied on the solid particles. Actually, the
evaluation of the local field in a solution with dispersed solid particles is a severe
problem. Studies are currently in progress to determine this value [40].
As a first attempt, let us consider a relationship like:

e =Ρφ
q (26)

where Ρ is a constant that takes into account the amount of surface charges that
can be polarized and φ is the applied local electrical field. If the value of φ is
_1
equal to 1 V m it is possible to demonstrate [41] that, when the surface
polarizable charges is equal to about same ten of electrons, then (eq. 26) the
2
correspondent value of e is equal to 10' and the value of the strain energy is
q

about 50 meV.
Hence it seems very reasonable to produce a strain energy of the same order of
magnitude as that of the solid/liquid interfacial energy under the effects of
1
electrical field as low as 1V m" .
Another suitable way of proving that a strain energy of the order of 50-100
meV can be produced by a low ELF field, lies in estimating the electromagnetic
energy that is ready to be used by a nucleus of critical size for straining its
surface/liquid interface.
By applying of the well known equation:
2
U = (1/μ°) (B )(V /2) tot (27)

where μ° is the magnetic permeability in a vacuum, Β is the applied magnetic

field, and V is the geometrical volume of the reactor to which the ELF field has
tot

been applied, it is possible to determine the total electromagnetic energy (U) that
is given to the system formed by dispersed solid and liquid phase.
If the number of critical nuclei in solution is known, one can to derive the
amount of electromagnetic energy per nucleus. When these calculations are made
15
for a value of Β equal to 3 mT and under our experimental conditions (i. e., 10
3
nuclei per cm of solution), an electromagnetic energy of the order of 100 meV per
nucleus can be derived.
Accordingly, the exchange of energy between ELF field and each nucleus is
greater enough to provide the energy requested to strain the nucleus/liquid
interface by an amount larger than the noises produced by the thermal
fluctuations.
This effect is not at all negligible, as it might ncrease the value of the
interfacial energy of a nucleus by a factor of two. Such an increase would involve
 189

several important consequences for both the nucleation process and the
subsequent growth step.
From eq. (14), it follows that doubling the interfacial energy value corresponds
to multiplying by a factor of 8 the apparent activation energy of the nucleation
process. This may have a considerable impact on the nucleation rate, as this rate
depends exponentially on the apparent activation energy according to eq. (16). As
a consequence the effect of ELF-field will be that of slowing down the nucleation
rate process and that of producing nuclei with a larger critical size because the
number of elementar embryos that need to be joined to reach the critical size will
increase according to eq. (10).
The subsequent crystal-growth step may be influenced by these effects. To
describe how this may occur, let us refer to a simple growth model, like the one
described in Figure 5.

Molecule

Figure 5. Model for crystal growth.

The molecules in solution will first adsorb on the critical nuclei and they will
proceed into the nucleus bulk phase to produce the growth process. This
sequence of steps can be described by the equations:
190

ki
A (sol) A (ads on S ) n+
(28)
k 2

k 3

A (ads on S +) n A (into the lattice of S +)

n
(29)

where kj is the kinetic constant and A denotes the molecules in solution.

Actually, the nature of these molecules may be the same as the one that formed
S or, more generally, it might be different. In the latter case, coprecipitation
n +

and/or other reactive processes can be described.

Under steady-state conditions one can write :

d[A(ads on S )]/dt = ki[A(sol)] - k [A(ads on S )] + k [A(ads on S )] = 0

n+ 2 n+ 3 n+ (30)

solving this equation allows us to express the growth rate as :

J G =[A(sol)] χ [ k i k / ( k + k )] 3 2 3 (31)

If k « k , then :
2 3

JG = [A(sol)] χ ki (32)

According to equation (32), the growth rate increases with the kinetic constant
of the adsorption step. Since the nuclei produced under ELF fields have a higher
interfacial energies, the adsorption step must be facilitated. Hence it may be
predicted that the grain growth rate of the solid product will be enhanced by the
effects of ELF fields. Actually, if the nature of the molecule A is different from
the nature of the solid nuclei S, equation (32) leads to predict that also the
kinetic rate between A and S will be increased. In the following, we shall test
these theoretical predictions.

3. EXPERIMENTAL

3.1. Techniques

The electromagnetic forces acting between solid nuclei of critical size formed in
a solution and an applied low-intensity ELF are the Lorentz forces and the
Maxwell stresses. If the solution flows through a circuit, a static magnetic field
can be used as a source of exogenous fields, whereas, if the solution is static (i.
e., the barycentric rate is zero), the source of ELF fields can be an alternating
electric current flowing in Helmoltz coils. In both cases, the resulting
 191

electromagnetic forces acting on the nuclei are the same; therefore, the observed
effects on the behaviour of solid nuclei should be equivalent.
In Figure 6 and in Figure 7, two examples of laboratory apparatus are given.
The apparatus shown in Figure 6 uses a Helmoltz-coils source, whereas, in the
laboratory apparatus in Figure 7, a static magnetic field device is used as an ELF
source.

R l J a c k e t e d beaker Cr Cryotermostat
Pa Plastic or glass stirrer TIC Temperature control
NC Motor control Ti Thermometer

Figure 6. Experimental apparatus with Helmoltz-coils source.

192

The treatment of a solutions (with and without the applied ELF field) must
follow a very precise protocol. It is very important that, during the treatment of
the same solution (with and without the applied ELF field), all the variables
should be the same so that a comparison of results may point out only the effect
of the exogenous field. This is not an easy task because the application of the
electromagnetic field might introduce some variables and/or change the values of
other variables.
For instance, in the case of an ELF field produced by Helmoltz coils,
extrathermal effects due to the electric current flowing in the coils should be
avoided. To this end, the same liquid solution should be poured into two vessels
(I and II in Fig. 6), placed midway between two dummy coils. On the active-coils
side (I), the alternating current generated by a power amplifier (driven by a
sinusoidal signal generator) flows to produce a preselected constant a. c.
magnetic field superimposed upon the earth's field. On the passive coils side (II),
wires of the same diameter and loop as utilized for the active coils must be used,
but the loop windings must be such as to generate an internal field Β equal to
zero. Accordingly, if there is any extrathermal effect due to the flow of the
electric current, this effect should be equal both for the experiments carried-out
under the action of the ELF and for those carried-out without applying any ELF.
When the liquid and/or the suspension circulates through a static magnetic
field device it should be stressed that the dynamics of the flowing system must be
the same in the ELF-treated and non-treated experiments. To this end the same
liquid solution must circulate through two devices of equal geometry (III and IV
in Fig. 7), and the static magnetic field must act only in one of these devices.
In addition, the pH value, the temperature, the concentration, the degree of
supersaturation, the digestion time, the stirring conditions etc, must be equal in
the ELF-treated and non-treated solutions/dispersion.
If further experimental operations, like the drying process, need to be
performed on the precipitate powders, before the final check on the
microstructural and chemical grain properties, the sequences and the conditions
to be used for the required steps must be equal for the ELF-treated and non-
treated samples.
 193

1-2 Solution reservoir III Device w i t h permanent magnet

3 Pump IV Device without permanent m a g n e t
4 Flowmeter 5-5* Sample for analysis

Figure 7. Experimental apparatus using static magnetic field.

194

3.2. Results

Direct experimental observations of the effects of ELF fields on the solid/liquid

interface energies of solid nuclei of critical size (formed during precipitation
processes from solutions) are very difficult because of the nuclei dimensions.
Nevertheless, any modification to the nature and to the microstructure of the
nuclei interfaces may strongly affect any subsequent reaction and/or process
where the nuclei are involved.
To test the theory defined in the Section 2 of this work, we shall refer to
experimental data that can be considered significant for three classes of
precipitation processes. The main differences among the three classes of
processes concern the steps that can be coupled with the growth mechanism.
In the first class of processes, the particle-growth step follows directly the
nucleation one, without any interactions with other solid-state mechanisms. For
the sake of clarity, let us call this process nucleation step and non-reactive
growth process.
The second class of processes, includes the process where the growth of a solid
product may proceed along two different lattice structures. Let us name this
process nucleation step and oriented growth process.
Finally, the third class of processes is characterized by a reaction between a
nucleus and other reactants; this reaction yields a solid product whose nature is
different from that of the nucleus. Let us call this process nucleation step and
reactive growth process.
In all three classes of processes, any modifications to the solid/liquid
interfacial energies of the nuclei produce significant and experimentally
observable variations in the nature and the microstructure of the final solid
product. This evidence will be used to test the theory previously defined.

3.2.1. Nucleation step and non-reactive growth process

As an example of this class of processes, let us consider the precipitation of
Barium Oxalate from aqueous and ethanol/water solutions [42, 43],
Aqueous solutions of Barium Nitrate containing 22 g Η were used, along with
Ammonium Oxalate 0.35 Μ as precipitating agent. The experimental apparatus
where the precipitation process was performed was the one described in the
previous section (Fig. 6). The precipitation experiments were carried out under
and without low-intensity ELF sinusoidal exposure, which yielded a magnetic
induction Β equal to 3 mT.
In Figure 8, the Barium concentration left in the aqueous solution vs. time is
plotted for two sets of experiments carried out under and without ELF exposure.
 195

Figure 8. Barium concentration left in aqueous solution vs. time.

196

Taking into account all sources of experimental errors, the experimental

results were such that we can state:
I. The effect of an applied ELF field does not change the total time required to
get a complete precipitation.
II. The barium concentration left in an aqueous solution significantly increases
when an ELF field is applied.
Both effects confirm the predictions of the theory on the effects of ELF fields
on the solid/liquid interfacial energies of the formed nuclei. Indeed, if this
parameter is increased by the exposure of the nuclei of critical size to an ELF
field, according to eq. (16) the number of nuclei of barium oxalate present in
solution at a certain time should decrease. Therefore, the corresponding barium-
ion concentration in solution should increase, hence, in accordance with eq. (32),
the rate of crystal growth should increase too. If so, as the total precipitation time
turned out to be equal for all experiments, the final microstructure of the
precipitate particles obtained under the ELF exposure should be larger. Figure
9a, b proves these considerations.
The average size of the precipitate particles obtained when the precipitation
process was performed under the ELF exposure is five times larger than the size
of the particles obtained when no ELF field was applied.
We regard these experimental results as reliable data that support our
theoretical predictions.
Ethanol/water mixture was used in proportion 1:3 (v/v) as solvent for the
_1
preparation of solutions of Barium Nitrate containing 22 g 1 , analogous
concentration to which used for the test carried out in aquesous solutions. As
precipitating agent Ammonium Oxalate 0.35 Μ was used. The precipitation
experiments were carried out under and without low-intensity ELF sinusoidal
exposure, which yielded a magnetic induction Β equal to 3 mT.
Figure 10 shows the precipitation data obtained without the ELF field vs time,
and those obtained by similar runs but in presence of a low-frequency, low-
intensity sinusoidal field.
 197

Figure 9. SEM micrographs of precipitate particles obtained in aqueous

solutions, with and without ELF field applied: a) Β = 0 mT, b) Β = 3 mT.
198

Figure 10. Barium concentration left in ethanol/water solution vs. time.

 199

The results obtained with these solutions are in agreement with the previous
one and reflect only the fact that the solubility product is less than the one in
pure water [43]. Accordingly the precipitation rate will be faster.

3.2.2. Nucleation step and oriented growth process.

Calcium hydrogencarbonate solutions were chosen to test this class of
reactions, as their behaviour is important for both inorganic and biological
systems. In these solutions, small nuclei of calcium carbonate can be present at
room temperature, for the standard Gibbs free energy associated with the
reaction *
2+
Ca (ao) + 2(HC0 )-(ao) = CaC0 (s) + H 0(1) + C0 (ao)
3 3 2 2 (33)

(ao = aqueous solution, s = solid phase, 1 = liquid phase) is negative and equal to
1
-24 kJ mol" . Furthermore, it has been proved that the growth of nuclei can be
favoured, by thermal heating in the temperature range 283-373 Κ and at a total
pressure of 0.1 MPa, in the directions of the solid phases corresponding to
aragonite and calcite. Evolution in the direction of the stabler solid phase
(calcite) becomes easier step if the nuclei growth reaction is a fast process [44].
As a consequence, it might be inferred that if the solid-liquid interfaces of the
nuclei are increased by exposing the calcium hydrogencarbonate solutions to a
low-frequency ELF field, their growth should proceed at a faster rate and the
final precipitate should be formed mainly by the calcitic phase. In a previous
research work, we explored this possibility and results turned out to be close
agreement with predictions [45].
Solutions with an equal chemical composition were made to circulate through
an apparatus like the one illustrated in Fig. 7. Treatments up to point 5 (Fig. 7)
were equal for all solutions; the passage through a magnetic device, which
induced on the charged surface of calcium carbonate nuclei an ELF field at a
frequency of 10-30 Hz, was utilized only for the solutions to be tested.
Subsequent thermal treatments to produce calcium carbonate precipitates were
all equal for treated and non-treated solutions. The X-ray diffractogram patterns
and the SEM micrographs of the precipitates concerning treated and non-treated
solutions are shown in Figure 11a, b and in Figure 12a, b. As it can be noticed,
the effects of the ELF fields lie in increasing the dimensions of the precipitate
grains and in orienting the growth process in the direction of the calcitic phase
formation. As has previously been observed [45], the effects of an ELF field are
dependent on the magnitude of the applied field in a non-linear way.
These experimental results further confirm the assumed kinetic mechanism
according to which ELF fields can be regarded as an external source that can
affect the surface properties of solid nuclei and all other processes that are
dependent on the existence of active critical nuclei.
200

A = Aragonite

C = Calcite
A

I I I I
40 35 30 25
2Θ (degrees)

Figure 11. X-ray difFractogram (λ = 1.7902 A) of the precipitate obtained by

evaporation, at a pressure of 0.1 Mpa and a temperature of 353 K, of aqueous
1
calcium hydrogencarbonate solutions, at a rate of 0.03 g s" . a) aqueous solution
before treatment with static magnetic field; b) aqueous solution after treatment
with static magnetic field.
 201

Figure 12. SEM micrographs of precipitate particles obtained under the same
conditions as in Fig. 11.
202

3.2.3. Nucleation step and reactive growth process

This is rather an attractive class of reactions, which include some important
phenomena that occur in the biomedical field as well as in the production of
ceramic powders to be used as raw materials for ceramic sensors [46]. In the
following we shall report new data on the formation of hydroxyapatite through
the precipitation process of calcium phosphate from aqueous solutions.
The effect of low-frequency ELF fields on the mechanisms of bone repair has
been focused in many important papers [7-9].
Experimental results strongly support the assumption that bone repair occurs
at a faster rate when ELF fields are applied; however, no definitive explanations
have been provide to understand the mechanisms of the complex interactions
between ELF fields and biological systems. A deeper insight into these
phenomena can be gained by studying the process step by step. Therefore, the
mineral formation of hydroxyapatite has been recognised to be of major
importance for the overall process.
It is very interesting to analyze 'in vitro' how such step may be affected by the
exposure of the precursory solutions to ELF fields.
From 225 ml of 0.04 Μ calcium chloride aqueous solutions, calcium phosphate
were precipitated by using Na HP0 as precipitating agent, at a concentration
2 4

0.036 M. The temperature and the pH value were kept constant, during the
process at 37° C and 7.40, respectively. As soon as nuclei and crystallites of
calcium phosphate were formed, a hydrolysis process occurred which transformed
the calcium phosphate into hydroxyapatite. More than likely, this step occurs
also 'in vivo'; hence the conclusions that can be drawn from 'in vitro' experiments
may provide useful information to understand the behaviour of a biological
system.
The experiments were carried out with and without application of a 3 mT ELF
field, for different precipitation times. The obtained precipitates were first
filtered, then dried, and finally examined with TG, XRD and N adsorption
2

techniques to derive information on the evolution of the solid-product

microstructure and of the chemical composition. A detailed description of these
experiments is given elsewhere [47]; here we report only the most important
results in order to discuss the implications of our theory for this set of reactions.
If ELF fields can modify the solid-liquid interfacial energies of calcium
phosphate nuclei, then not only the crystal growth step should proceed at a faster
rate but also the kinetics of the subsequent and/or the parallel calcium
phosphate hydrolysis step should be enhanced. Accordingly, experimental results
should point out the formation of larger precipitate crystallites with a chemical
composition richer in hydroxyapatite.
Nitrogen adsorption-desorption isotherms at 78 Κ were taken on precipitated
powders obtained at different digestion time with and without ELF fields
applied. From these data, it is possible to deduce information on the value of the
powder specific surface areas, on the total porosity, and on the pore size
distribution (see, for instance, [48-50]).
 203

Figures 13a, 14a and 15a illustrate the evolution of the adsorption-desorption
isotherms as a function of the digestion time.
Figures 13b, 14b and 15b give the asme informations at correspondent
digestion time for samples obtained under the influence of 3 mT electromagnetic
field.
Comparison among these data allows to reach a number of informations. First
the all will should be noted that all isotherms since to be long to a 4th type class
[35]. The applied ELF field does not seem to change this kind of behavior.
Newertheless, the values of the specific surface area, total porosity and
hysteresis shape, because the effect of the ELF field, seemto be significantly
different.
For our purpose we focus the attention on the specific surface area values and
not on the porosity value because this last parameter might be also be due to
intraparticles porosity, 1. e., to powder packing factor.
Accordingly in Table 2, the specific surface area data are summarized.

Table 2
Specific surface data.

ELF field exposure

B = OmT Β = 3 mT

2 1 2 1
Digestion time (min) SBET (m g" ) SBET (m g' )

15 131.07 107.80
60 110.62 71.99
120 116.43 74.18

Experimental error less than 10%

 204

(
rH
N
i
-
00
i
i
ΙΟ
-
Hi
<N
-
H
O
^
C
O
C
O
O

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
P/Po

Figure 13a. Nitrogen adsorption isotherms at 78 Κ for precipitates obtained without ELF field
for a digestion time of 15 min.
Figure 13b. Nitrogen adsorption isotherms at 78 Κ for precipitates obtained with ELF field
for a digestion time of 15 min.
205
 206

Ι
C
o
0
Ο

o
C
Ο
0
Ι

o
<
Ο
N
o
(
Ο
N
Ι

o
(2/guio) e u m j o A
r
Ο
-
o
|
Ο
r

uaSoj^x^
Ι

o
H
Ο
O
o
Ο
O
Figure 14.a. Nitrogen adsorption isotherms at 78 Κ for precipitates obtained without ELF field
for a digestion time of 60 min.
 io
CO
Ο
00

CO
Ο
io
Ο

Ο
Ο
CO

Ο
(S/grao) θ ί π η ρ Λ iiaSc-j^i^
Ο
Ο
ΙΟ

Ο
Ο
Ο
Ο

d
d
d

d
d
d

Ο
Ο

ί-
00

CO

CO
A /Χ \J
Figure 14.b. Nitrogen adsorption isotherms at 78 Κ for precipitates obtained with ELF field
for a digestion time of 60 min.
207
 208

i-i
00
ΙΟ
(Ν

Figure 15.a. Nitrogen adsorption isotherms at 78 Κ for precipitates obtained without ELF field
for a digestion time of 120 min.
Figure 15.b. Nitrogen adsorption isotherms at 78 Κ for precipitates obtained with ELF field
for a digestion time of 120 min.
209
210

The reported data clearly show that the value of the specific surface areas of
the products obtained under the action of a 3 mT ELF field are definitely smaller
than the corresponding values obtained for samples precipitated without
applying an ELF field. These results strongly suggest that the particles produced
under the action of an ELF field are larger than those obtained without applying
any ELF field.
In Table 3 the hydroxyapatite powder contents resulting from the TG and XRD
analysis are summarized.
The experimental error concerning XRD analysis may exceed the 5 %,
accordingly no definitive comparison can be done between the amount of
hydroxyapatite content in samples obtained with and without applied ELF field.
The experimental error related to the TG analysis is included within 1%,
accordingly from data of Table 3 (column first and second) it seems that the
amount of hydroxyapatite content obtained with applied ELF field is higher than
the one concerning the sample obtained without applied ELF field.

Table 3
Hydroxyapatite content.

Hydroxyapatite content

Calculated from TG data* Calculated from XRD data

Digestion time (min) Β = 0 mT Β = 3 mT Β = 0 mT Β = 3mT

15 45% 50 % 40 % 43%
60 75% 78 % 71 % 75%
120 76% 80 % 80 % 86%

*Experimental error less than 1%

Hence the data on the microstructure of the solid hydroxyapatite/calcium

phosphate crystallites and on their initial rate of conversion in to hydroxyapatite
due to the hydrolysis of the calcium phosphate, are all in agreement with the
theoretical predictions previously reported.
Extension of these results to the bone-repair process under the influence of a
low ELF field is still to be managed with care, as different interactions between
living cells and ELF field may be involved. Nevertheless, it worth noting that in
 211

agreement with the results above reported 'in vitro', also the grain concerning
thin film section of bone treated with an ELF field seems to have a greater size
than that concerning bone non-treated with electromagnetic field [51].
From a practical point of view the evidence that repaired bone can be
characterized by a larger size of hydroxyapatite grains not necessarily can
correspond to an increase in bone mechanical resistance. If so the biomedical
features concerning the application of an ELF field to the bone repair operations
needs to be evaluated with a global approach.

4. CONCLUSIONS

The accumulated experimental evidence for three different classes of

homogeneous nucleation and subsequent steps phenomena, is in close agreement
with theoretical predictions according to which application of low-intensity ELF
fields can modify the solid/liquid interfaces of initial solid nuclei of critical size.
Even though such a modification cannot be directly verified by experiments,
nevertheless, results seem to confirm this phenomenon.
The possibility that a modification to the initial surface properties of solid
nuclei may affect all other subsequent steps can be considered plausible only if
these steps are irreversible in nature (eqs. 28, 29, 31 and 32). This is consistent
with the evolution of the irreversible phenomena [52-54], in which boundary
conditions play a central role in determining the final states.
From this point of view, low-intensity ELF fields can be regarded as an
exogenous source of energy that, through modifications to the solid-liquid
interfacial energies of solid nuclei is able to produce non-equilibrium conditions
during the growth steps that lead to the formation of final solid products.

ACKNOWLEDGEMENT

Discussions with Dr. T. Aste and S. Mezzasalma and with Proff. B. Bianco, M.
Capurro and A. Chiabrera were very helpful in the development of this research
work.
This paper was supported by the Italian MURST 40% - (Project on Advanced
Materials), by the National Council of Research (CNR) (Technological
Committee) and by the P.F. MSTA (CNR).
212

REFERENCES

1. A. Chiabrera, C. Nicolini and H. P. Schwan Eds., Interactions between

Electromagnetic Fields and Cells, NATO ASI Series, Series A, Life Sciences,
Vol. 97, Plenum Press, New York, 1985.
2. S. Takashima, Electrical Properties of Biopolymers and Membranes, Adam
Hilger, Philadelphia, 1989.
3. B. Bianco and A. Chiabrera, Bioelectrochemistry and Bioenergetics, 28
(1992)355.
4. A. Chiabrera, B. Bianco, J. J. Kaufmann and A. A. Pilla, in Β Norden and C.
Ramel Eds, Interaction Mechanisms of Low Level Electromagnetic Fields in
Living Systems, Oxford Press, Oxford, 1992, 164.
5. A. Chiabrera, B. Bianco, J. J. Kaufmann and A. A. Pilla, in C. T. Brighton
and S. R. Pollack Eds., Electromagnetics in Medicine and Biology, San
Francisco Press, San Francisco, 1991, 21.
6. A. Chiabrera, B. Bianco, J. J. Kaufmann and A. A. Pilla, in C. T. Brighton
and S. R. Pollack Eds., Electromagnetics in Medicine and Biology, San
Francisco Press, San Francisco, 1991, 27.
7. C. A. Basset, Critical Reviews in Biomedical Engineering, 17(5) (1989) 451.
8. C. A. L. Bassett, M. Schink-Ascani and S. M. Lewis, Clinical Orthopaedics,
246(1989) 172.
9. W. J. W. Sharrard, J. of Bone and Joint Surgery, 72-B(3) (1990) 347.
10. M. Hinsenkamp and M. Rooze Eds., Current Concept in
Bioelectromagnetics, Bioelectrochemistry and Bioenergetics, 30 (1993).
11. H. A. Pohl, Dielectrophoresis, Cambridge University Press, Cambridge,
1978.
12. E. Wrede, Z. Physik, 44 (1927) 261.
13. H. A. Pohl, J.Appl. Phys., 29 (1958) 1182.
14. H. A. Pohl and J. P. Schwar, J.Appl. Phys., 30 (1959) 69.
15. H. A. Pohl, J.Electrochem. Soc, 107 (1960) 386.
16. H. A. Pohl and C. E. Plymale, J.Electrochem. Soc, 107 (1960) 390.
17. H. A. Pohl and J. P. Schwar, J.Electrochem. Soc, 107 (1960) 383.
18. H. A. Pohl, J.Appl. Phys., 32 (1961) 1784.
19. H. A. Pohl, J.Electrochem. Soc, 115 (1968) 115c
20. P. J. W. Debye, Phys. Rev., 91 (1953) 210.
21. P. Debye, P. P. Debye, Β. H. Eckstein, W. A. Barber and G. J. Arquette, J.
Chem. Phys., 22 (1954) 152.
22. P. Debye, P. P. Debye and Β. H. Eckstein, Phys. Rev., 94 (1954) 1412.
23. W. A. Barber, P. Debye and Β. H. Eckstein, Phys. Rev., 94 (1954) 1412.
24. O. Ya. Samoilov, Structure of Aqueous Electrolite Solutions and the
Hydration of Ions, Consultant Bureau, New York, 1965.
25. Y. Kinouchi, S. Tanimoto, T. Ushita, K. Sato, H. Yamaguchi and H.
Miyamoto, Bioelectromagnetics, 9 (1988) 159.
 213

26. W. Κ. Burton, Ν. Cabrera and F. C. Frank, Royal Soc. Phil. Trans., 243A
(1951) 299.
27. W. F. Dunning, in W. E. Garner Ed. Chemistry of the Solid State,
Butterworths Scientific Publications, London, 1955, Cap. 6.
28. Τ. N. Rhodin and D. Walton, in Metals Surfaces: Structure, Energetics and
Kinetics, American Society for Metals, Metals Park OH, 1962, Cap. 8.
29. A. G. Walton, The Formation and Properties of Precipitates, Interscience
Publishers, New York, 1967.
30. R. Defay, I. Prigogine, A. Bellemans and D. H. Everett, Surface Tension and
Adsorption, Longmans, London, 1966, Cap. XVIII.
31. D. Turnbull, J. Appl. Phys., 20 (1949) 817.
32. G. M. Pound and V. K. La Mer, J. Amer. Chem. Soc, 74 (1952) 2323.
33. W. Jacobi, Z. Naturforsch., 10 A (1955) 322.
34. P. F. Rossi and C. De Asmundis, Ann. Chim., 67 (1977) 65.
35. S. J. Gregg and K. S. W. Sing, Adsorption, Surface Area and Porosity,
Academic Press, London, 1982.
36. J. J. Gilman, J. Appl. Phys., 31 (1960) 2208.
37. D. Beruto and M. Capurro, J. of Materials Science, 28 (1993) 4693.
38. M. Capurro, personal communication.
39. D. Baldovino, S. Mezzasalma and D. Beruto, send for publication.
40. B. Bianco and A. Chiabrera, work in progress.
41. S. Mezzasalma, personal communication.
42. R. Berton, D. Beruto, B. Bianco, A. Chiabrera and M. Giordani,
Bioelectrochem. and Bioenerg., 30 1993 13.
43. R. Berton, D. Beruto, B. Bianco, A. Chiabrera and M. Giordani, in M. Blank
Ed. Electricity and Magnetism in Biology and Medicine, San Francisco Press,
S. Francisco, (1993) 902.
44. D. Beruto and M. Giordani, J. Cryst. Growth, 84, 679 (1987).
45. D. Beruto and M. Giordani, J. Chem. Soc. Faraday Trans., 89 (14), (1993),
2457.
46. D. Beruto, R. Botter, M. Giordani and G. Giannetti, in H. Hausner, G. L.
Messing and S. Hirano Eds. Ceramic Powder Processing Science, Deutsche
Keramische Gesellschaft, (1989) 77
47. D. Beruto and M. Giordani, send for publication.
48. D. Beruto, L. Barco, G. Belleri and A. W. Searcy, J. Amer. Cer. Soc, 64
(1981) 74 .
49. D. Beruto, L. Barco and A. W. Searcy, J. Amer. Cer. Soc, 67 (1984) 512.
50. D. Beruto, R. Botter and A. W. Searcy, J. Phys. Chem., 91 (1987) 3578 .
51. V. Cane, P. Botti, D. Farneti and S. Soana, J. of Orthopaedic Research, 9
(1991)908.
52. I. Prigogine, From Being to Becoming, W. H. Freeman and Company, San
Francisco, 1980.
53. P. Glansdorff and I. Prigogine, Structure, Stability et Fluctuations, Masson
et Cie, Paris, 1971.
54. S. Carr&, La formazione delle strutture, Bollati Boringhieri, Torino, 1989.
RESEARCH IN CHEMICAL KINETICS, VOLUME 5
R.G. Compton and G. Hancock (editors) 215
© 1995 Elsevier Science B.V. All rights reserved.

The kinetics of the hydrogen evolution reaction at nickel electrodes

J. T a m m a n d L. T a m m

Institute of Physical Chemistry, University of Tartu,

Tartu, EE2400, Estonia

1. I N T R O D U C T I O N

The study of cathodic h y d r o g e n evolution reaction has played a significant role

in t h e f o r m a t i o n o f m o d e r n e l e c t r o c h e m i s t r y as a b r a n c h o f s c i e n c e [ 1 - 1 5 ]
At t h e b e g i n n i n g o f t h i s c e n t u r y t h e d i s c h a r g e o f h y d r o g e n i o n s w a s c o n s i d e r e d
to be a simple reaction u n d e r s t a n d i n g of which w o u l d enable to build up the
general theoretical principles of electrochemical kinetics By the middle of the
c e n t u r y it w a s a l r e a d y c l e a r t h a t t h e h y d r o g e n e v o l u t i o n r e a c t i o n is a n y t h i n g b u t a
simple process [14]
T h e c a t h o d i c h y d r o g e n e v o l u t i o n is an e l e c t r o c a t a l y t i c p r o c e s s i n f l u e n c e d b o t h
by t h e p r o p e r t i e s o f t h e s u r f a c e o f t h e e l e c t r o d e a n d o f t h e s o l u t i o n . T h i s r e a c t i o n
m a y b e r e g a r d e d as i n v o l v i n g at l e a s t five c o n s e c u t i v e s t e p s as f o l l o w s :
1. T r a n s p o r t of the r e a c t a n t s to the surface.
2. A d s o r p t i o n of the reacting particles on the surface.
3 R e a c t i o n ( s ) on the surface
4 D e s o r p t i o n of the p r o d u c t s
5 T r a n s p o r t of the p r o d u c t s away from the surface.
Any of those steps could be rate-determining of the over-all process. If t h e
c o n c e n t r a t i o n o f r e a c t a n t s is h i g h e n o u g h o r s p e c i f i c m e a s u r e s a r e t a k e n t o
a c c e l e r a t e the t r a n s p o r t p r o c e s s e s (for instance using a r o t a t i n g disk e l e c t r o d e ,
flowing solution e t c [ 1 6 ] ) , t h e t r a n s p o r t o f r e a c t a n t s is fast e n o u g h n o t t o b e
taken into consideration. In t h i s c a s e t h e m e c h a n i s m o f t h e c h e m i c a l r e a c t i o n s
p r o c e e d i n g d i r e c t l y o n t h e s u r f a c e b e c o m e s a v a i l a b l e for s t u d y i n g .
T h e c a t h o d i c h y d r o g e n e v o l u t i o n r e a c t i o n ( H E R ) in a c i d i c s o l u t i o n s

+
2 H 0 3 + 2e" = H 2 + 2 H 0,
2

o r in a l k a l i n e s o l u t i o n s

2 H 0 + 2e" = H
2 2 + 2 OH"

is a t w o - e l e c t r o n p r o c e s s p r o c e e d i n g m o s t l i k e l y in t w o c o n s e c u t i v e s t e p s w i t h
a d s o r b e d a t o m i c h y d r o g e n a s a n i n t e r m e d i a t e T h e first s t e p o f t h e r e a c t i o n in
a c i d i c s o l u t i o n s is t h e d i s c h a r g e o f h y d r o g e n i o n s ( t h e V o l m e r ' s r e a c t i o n ) [ 4 , 9 ] :
216

f o l l o w e d by e l e c t r o c h e m i c a l d e s o r p t i o n ( t h e H e y r o v s k y ' s r e a c t i o n ) [ 3 ] :

H
1 0 +
a d s + H
ads + e = H
2.ads + H
2 ads'
0
<">

or r e c o m b i n a t i o n (the Tafel's reaction) [1]:

H
a d s
+ H
a * = H
2 . a d s d»)

of the originated atomic hydrogen.

In a l k a l i n e s o l u t i o n s t h e first s t e p is u s u a l l y t h e d i s c h a r g e o f p r o t o n s f r o m w a t e r
molecules

-
HO t + e = Η + OH , , (IV)
2 ads ads ads
f o l l o w e d by t h e e l e c t r o c h e m i c a l d e s o r p t i o n

H
A d s
+ H
a J s
+ e
- = H . „ + OH- ,,
2 a d i d (V)

or r e c o m b i n a t i o n ( r e a c t i o n III)
T h e i n d e x ads is a d d e d t o t h e s y m b o l s o f r e a c t a n t s in t h e r e a c t i o n e q u a t i o n s
a b o v e t o e m p h a s i s e t h e g r e a t e r or s m a l l e r i n f l u e n c e o f t h e s u r f a c e o n t h e e n e r g e t i c
p r o p e r t i e s o f t h e p a r t i c l e s t a k i n g p a r t in t h e r e a c t i o n
In t h e c a s e o f a l k a l i n e s o l u t i o n s a n o t h e r m e c h a n i s m is p o s s i b l e w i t h t h e
d i s c h a r g e o f alkali m e t a l c a t i o n s o n t h e e l e c t r o d e as a first s t e p f o l l o w e d by t h e
r e a c t i o n o f o r i g i n a t e d a c t i v e m e t a l a t o m s w i t h w a t e r m o l e c u l e s [ 1 7 , 1 8 ] , for
instance:

+
Na + e =Na, (VI)

+
2Na + 2 H 0 = H 2 2 + 2 Na + 2 OH' (VII)

T h e s t u d y o f c a t h o d i c h y d r o g e n e v o l u t i o n is s i g n i f i c a n t n o t o n l y f r o m t h e
theoretical point of view T h e r e a c t i o n is w i d e l y u s e d in t h e p r o d u c t i o n o f p u r e
h y d r o g e n by m e a n s o f w a t e r e l e c t r o l y s i s F u r t h e r m o r e , the anodic process of the
p r o d u c t i o n o f s e v e r a l s u b s t a n c e s ( s u c h as c h l o r i n e , m a n g a n e s e d i o x i d e , p e r o x i d e s
e t c ) is u s u a l l y a c c o m p a n i e d by t h e c a t h o d i c h y d r o g e n e v o l u t i o n r e a c t i o n
The regularities of cathodic hydrogen evolution have been investigated nearly
s i n c e t h e v e r y b e g i n n i n g o f t h i s c e n t u r y s t a r t i n g w i t h t h e h i s t o r i c a l w o r k o f Tafel
[ 1 ] C o m p a r a t i v e l y m u c h s u c c e s s from t h e t h e o r e t i c a l a s p e c t h a s b e e n a c h i e v e d in
t h e i n v e s t i g a t i o n s c a r r i e d o u t on m e r c u r y and o t h e r m e t a l s o r a l l o y s o f t h i s t y p e
[19,20] T h e p a r t i c u l a r a d v a n c e s in t h i s field a r e r e a s o n a b l e as t h e s u r f a c e o f
liquid m e t a l s is well r e p r o d u c i b l e , e n e r g e t i c a l l y h o m o g e n e o u s a n d c a n b e m o d e l l e d
m o r e a c c u r a t e l y t h a n t h e s u r f a c e o f solid e l e c t r o d e s
The utmost c o n s e q u e n c e has, certainly, the studying of electrochemical
p r o c e s s e s o n c a t a l y t i c a l l y a c t i v e m e t a l s o f g r e a t a p p l i c a t i o n , s u c h as p l a t i n u m -
 217

metals, iron, nickel etc. A t h o r o u g h r e v i e w by S. T r a s a t t i " E l e c t r o l y s i s o f

H y d r o g e n E v o l u t i o n : P r o g r e s s in C a t h o d e A c t i v a t i o n . " h a s b e e n p u b l i s h e d r e c e n t l y
[21]. The relationships between fundamental knowledge and practical
a c h i e v e m e n t s in t h i s field w i t h t h e t e c h n o l o g i c a l a s p e c t s b e i n g u n d e r d e e p e r
i n v e s t i g a t i o n a r e d i s c u s s e d in t h i s r e v i e w .
N i c k e l b e l o n g s t o e l e c t r o c a t a l y t i c a l m e t a l s a n d is t h e m o s t w i d e l y u s e d c a t h o d e
m a t e r i a l for w a t e r e l e c t r o l y s i s a n d o t h e r p r o c e s s e s o f p r a c t i c a l i n t e r e s t In t h i s
r e p o r t e m p h a s i s is t o b e p l a c e d p r i m a r i l y o n t h e p r o b l e m s , t h e s e t t l e m e n t o f w h i c h
is n e e d e d for t h e d e e p e r u n d e r s t a n d i n g o f t h e r e g u l a r i t i e s o f t h e c a t h o d i c h y d r o g e n
e v o l u t i o n r e a c t i o n from the t h e o r e t i c a l point of view T h e r e f o r e the results of the
investigations of the hydrogen evolution reaction obtained on pure nickel are of
main interest of present work. T h e r e v i e w d o e s n o t c l a i m t o b e an u l t i m a t e
t r e a t m e n t o f t h e p r o b l e m a s t h e i n t e r f a c e n i c k e l / s o l u t i o n is a v e r y c o m p l i c a t e d
s y s t e m a n d n o t e n o u g h in situ e x p e r i m e n t a l d a t a a r e a v a i l a b l e a b o u t t h e s t r u c t u r e
o f t h i s i n t e r f a c e . F o r t h a t r e a s o n it is n e c e s s a r y t o c o n f i n e w i t h s o m e simplified
model a p p r o a c h e s which reflect the p r o p e r t i e s of nickel e l e c t r o d e surface quite
approximately.

2. MAIN CHARACTERISTICS OF THE HYDROGEN EVOLUTION

REACTION

A p a r a m e t e r o f g r e a t i m p o r t a n c e in t h e d e s c r i p t i o n o f t h e p r o c e s s o f c a t h o d i c
h y d r o g e n e v o l u t i o n is h y d r o g e n o v e r p o t e n t i a l ( η ) s h o w i n g t h e shift o f t h e
e l e c t r o d e p o t e n t i a l u n d e r c a t h o d i c c u r r e n t in t h e r e s p e c t o f r e v e r s i b l e p o t e n t i a l in
the same solution E°:

η = E ° - Ε. (1)

In t h i s r e v i e w w e u s e t h e t r a d i t i o n a l a p p r o a c h : h y d r o g e n o v e r p o t e n t i a l is d e f i n e d
to have positive value
T h e first q u a n t i t a t i v e r e l a t i o n s h i p b e t w e e n t h e r a t e o f an e l e c t r o c h e m i c a l
r e a c t i o n i.e. c u r r e n t d e n s i t y a n d h y d r o g e n o v e r p o t e n t i a l is t h e Tafel e q u a t i o n [ 1 ] :

η = a + b In i . (2)

It is an e m p i r i c a l e x p r e s s i o n w i t h e x p e r i m e n t a l c o n s t a n t s a a n d b c a l l e d Tafel
constants. These constants are convenient to characterise the cathodic hydrogen
e v o l u t i o n p r o c e s s on a given metal
T h e metals are divided into t h r e e g r o u p s on the basis of the value of h y d r o g e n
overpotential:
a) t h e m e t a l s o f h i g h o v e r p o t e n t i a l s u c h a s m e r c u r y , tin, l e a d , b i s m u t h , e t c . for
w h i c h a is g r e a t e r t h a n 1 V ,
b) t h e m e t a l s o f l o w o v e r p o t e n t i a l , p r i m a r i l y t h e m e t a l s o f p l a t i n u m g r o u p w i t h a
l e s s t h a n 0.5 V ,
218

c) t h e m e t a l s o f a m e d i u m v a l u e o f h y d r o g e n o v e r p o t e n t i a l b e t w e e n t h e s e g r o u p s
mentioned above (0.5 V < a < 1 V) The typical metals of medium
overpotential are the ones of iron g r o u p .
Tafel s l o p e b r e p r e s e n t s t h e d e p e n d e n c e o f p o t e n t i a l o n t h e r a t e o f an
e l e c t r o c h e m i c a l r e a c t i o n ( c u r r e n t ) a n d c a n b e w r i t t e n in t h e f o r m

b = RT/ocF, (3)

w h e r e α is c h a r g e t r a n s f e r c o e f f i c i e n t . In t e r m s o f d e c i m a l l o g a r i t h m Tafel s l o p e b
will b e 0 , 1 1 8 V at α = 0,5 a n d Τ = 2 9 3 Κ
T h e net c u r r e n t d e n s i t y / c a n b e d e f i n e d as t h e d i f f e r e n c e o f f o r w a r d (/ ) a n d
b a c k w a r d (/' ) c o m p o n e n t s

i = T-T (4)

At t h e r e v e r s i b l e p o t e n t i a l w h e r e /" = 0 a n d η = 0 t h i s l e a d s t o t h e

T = r = i„, (5)

w h e r e / is t h e e x c h a n g e c u r r e n t d e n s i t y , w h i c h is a n o t h e r p a r a m e t e r o f t e n u s e d t o
characterise the hydrogen evolution process
B o t h c u r r e n t d e n s i t y / a n d t h e e x c h a n g e c u r r e n t d e n s i t y /' a r e d e t e r m i n e d , as a
(

r u l e , in r e s p e c t o f t h e a p p a r e n t g e o m e t r i c s u r f a c e a r e a , a l t h o u g h t h e r e a l a n d
a p p a r e n t s u r f a c e a r e a s m a y differ t o g r e a t e x t e n t T h e r e l a t i o n b e t w e e n t h e real
and a p p a r e n t a r e a s is c h a r a c t e r i s e d by t h e f a c t o r o f s u r f a c e r o u g h n e s s f which
m a y v a r y from a l m o s t 1 for t h e i d e a l l y s m o o t h s u r f a c e o f m o n o c r y s t a l s t o
4
r e m a r k a b l y high v a l u e s , for i n s t a n c e 1 0 in t h e c a s e o f R a n e y n i c k e l [ 2 1 ] .
C o n s e q u e n t l y t h e d i f f e r e n c e s in t h e v a l u e s o f s u r f a c e r o u g h n e s s m u s t b e t a k e n i n t o
a c c o u n t w h i l e c o m p a r i n g t h e real a c t i v i t i e s o f d i f f e r e n t s u r f a c e s A s a r u l e , b o t h / ()

a n d a a r e d e t e r m i n e d by e x t r a p o l a t i o n a n d t h e r e f o r e t h e a c c u r a t e e s t i m a t i o n o f
t h e i r v a l u e s will b e c o m p l i c a t e d if t h e p o l a r i s a t i o n c u r v e c o n s i s t s o f t w o o r m o r e
r e g i o n s w i t h d i f f e r e n t v a l u e s o f Tafel s l o p e b.
A p a r a m e t e r i n v o l v e d in t h e k i n e t i c r e l a t i o n s h i p s o f h y d r o g e n e v o l u t i o n is t h e
s u r f a c e c o v e r a g e o f t h e m e t a l by a d s o r b e d h y d r o g e n Θ Η

T h e r a t e o f t h e r e a c t i o n s o f ( I ) , ( I I ) a n d ( I I I ) if t h e d o u b l e l a y e r e f f e c t s
r e g a r d e d as c o n s t a n t ( a n d L a n g m u i r i s o t h e r m is a p p a r e n t l y h o l d for h y d r o g e n
adsorption ) may be described as [ 2 2 , 2 3 ] :

v
. = k
i C
H < 1
- e ) e x p ( c x FTi/RT)-k ^ e x p K
H t 1-cx^FTI/RT], (6)

v
2 =
k C 2 H ( ) .0 exp(a F^RT)-k
H 2 2 (l-0 )exp[-(l-a )F^RT]
H 2 , (7)

V
3 = K
3 E
« « 2 1
"M -^ (8)

w h e r e k is r a t e c o n s t a n t , α is t h e t r a n s f e r coefficient
 219

It is c l e a r t h a t Θ is s t r o n g l y d e p e n d e n t o n t h e c a t a l y t i c a c t i v i t y o f t h e m e t a l
Η

surface The metals of the platinum g r o u p , possessing maximum catalytic activity

in t h e p r o c e s s o f h y d r o g e n e v o l u t i o n , h a v e t h e s u r f a c e c o v e r a g e w i t h a d s o r b e d
hydrogen θ ~ 1 [20] T h e c a t a l y t i c a c t i v i t y o f n i c k e l is s o m e w h a t l o w e r b u t still
c o n s i d e r a b l y high a n d it is r e a s o n a b l e t o s u p p o s e t h a t Θ for n i c k e l c a n n o t b e t o o
Η

low either
T h e v a l u e o f Θ a n d its d e p e n d e n c e on p o t e n t i a l is o n e o f t h e m o s t i n d i c a t i v e
Η

p a r a m e t e r s for e s t a b l i s h i n g t h e m e c h a n i s m o f t h e h y d r o g e n e v o l u t i o n r e a c t i o n a n d
t h e r e f o r e p a r t i c u l a r a t t e n t i o n m u s t be p a i d t o t h e r e l i a b i l i t y a n d a c c u r a c y o f t h e
m e t h o d s for t h e d e t e r m i n a t i o n o f Θ { Γ

3. D E T E R M I N A T I O N OF SURFACE COVERAGE WITH ADSORBED

HYDROGEN

T h e e x p e r i m e n t a l m e t h o d s , m o r e o r l e s s a p p r o p r i a t e for t h e d e t e r m i n a t i o n o f t h e
d e g r e e of the c o v e r a g e of metal surface with the adsorbed intermediate
(hydrogen), are galvanostatic charging, potentiodynamic transients, overpotential
decay and e l e c t r o c h e m i c a l i m p e d a n c e
A t h o r o u g h a n a l y s i s o f t h e s e m e t h o d s is g i v e n in [ 2 2 ] t h u s o n l y a b r i e f a c c o u n t
o f t h e u n d e r l y i n g p r i n c i p l e s will b e g i v e n h e r e n e e d e d for t h e u n a m b i g u o u s
i n t e r p r e t a t i o n o f e x p e r i m e n t a l r e s u l t s o b t a i n e d on n i c k e l

(I) Galvanostatic charging curves [20,22,24,25].

T h e p r o c e d u r e i n v o l v e s c h a r g i n g u s i n g a r a t h e r small c o n s t a n t a n o d i c c u r r e n t
and determination of the electric charge required to ionise the h y d r o g e n adsorbed
on the surface of the e l e c t r o d e :

Η A - e" + H O = H O (VIII)
ads 2 3
If t h e r e is m o l e c u l a r h y d r o g e n in t h e s o l u t i o n t h e n a s i g n i f i c a n t r e - a d s o r p t i o n of
h y d r o g e n from the solution to the e l e c t r o d e surface may occur,

Η - 2 Η (IX)
2 ads v 7

a n d as a r e s u l t a h i g h a p p a r e n t s u r f a c e c o v e r a g e o f h y d r o g e n will b e o b t a i n e d . In
order to eliminate this s o u r c e of error the dissolved hydrogen has to be removed
from t h e s o l u t i o n by b u b b l i n g n i t r o g e n o r n o b l e g a s e s
This method, although, cannot be applied to electrodes polarised cathodically,
as t h e s o l u t i o n is s a t u r a t e d w i t h m o l e c u l a r h y d r o g e n in t h e c l o s e v i c i n i t y o f t h e
e l e c t r o d e a n d s m a l l h y d r o g e n b u b b l e s m a y c l i n g t o t h e e l e c t r o d e s u r f a c e so t h a t
re-adsorption of hydrogen cannot be avoided The influence of re-adsorption can
b e d i m i n i s h e d o r e v e n e l i m i n a t e d u s i n g t h e fast c h a r g i n g m e t h o d d e v e l o p e d in
2
[26,27] A very high current density, of the o r d e r of 1 A / c m , w a s used and the
4
p u l s e d u r a t i o n w a s r e d u c e d t o 10" s
220

T h e g a l v a n o s t a t i c c h a r g i n g m e t h o d c a n b e s u c c e s s f u l l y a p p l i e d o n l y if t h e r e is a
considerable difference between the potential c o r r e s p o n d i n g to the completion of
h y d r o g e n ionisation and that c o r r e s p o n d i n g to the c o m m e n c e m e n t of t h e
c h e m i s o r p t i o n of oxygen or oxide formation T h e r e f o r e t h i s m e t h o d is a p p l i c a b l e
s u c c e s s f u l l y o n l y for n o b l e m e t a l s . If t h e s u c c e e d i n g p r o c e s s o v e r l a p s t h e
h y d r o g e n i o n i s a t i o n as in t h e c a s e o f n i c k e l , t h e g a l v a n o s t a t i c c h a r g i n g m e t h o d
becomes inapplicable

(II) Double charging method [28,29]

In o r d e r t o o v e r c o m e t h e c o m p l i c a t i o n m e n t i o n e d a b o v e a p r o c e d u r e h a s b e e n
d e v e l o p e d , n a m e d by t h e a u t h o r s as " d o u b l e c h a r g i n g m e t h o d " During the
g a l v a n o s t a t i c t r a n s i e n t a n o d i c c u r r e n t d e n s i t y i is e x p e n d e d o n c h a r g i n g t h e
&

d o u b l e l a y e r a n d o n all a n o d i c F a r a d a i c p r o c e s s e s p r o c e e d i n g at p o t e n t i a l E,
consequently

i =C d AI
d

\ Wd t ) x
+ L,
v (9)

where C is t h e c a p a c i t y o f t h e d o u b l e l a y e r at a p o t e n t i a l E, i - t h e s u m o f
d ( y

c u r r e n t s o f all a n o d i c F a r a d a i c p r o c e s s e s p r o c e e d i n g at t h e p o t e n t i a l Ε The
c u r r e n t of F a r a d a i c p r o c e s s e s / can be divided into t w o p a r t s :

w h e r e / is t h e c u r r e n t d e n s i t y d u e t o t h e i o n i s a t i o n o f a d s o r b e d h y d r o g e n , / ' is an

the current density due to other anodic processes (primarily surface oxide
formation)
In a d d i t i o n t o t h e c o n v e n t i o n a l c h a r g i n g c u r v e a c o m p e n s a t i n g c u r v e is t a k e n
f r o m a p o t e n t i a l s u f f i c i e n t l y a n o d i c in r e s p e c t o f t h e h y d r o g e n r e v e r s i b l e p o t e n t i a l
in t h e s a m e s o l u t i o n a s s u m i n g t h a t t h e s u r f a c e c o v e r a g e by a d s o r b e d h y d r o g e n is
n e g l i g i b l e at t h i s p o t e n t i a l Only the Faradaic p r o c e s s of surface oxidation and
d o u b l e layer charging are a s s u m e d to p r o c e e d during a c o m p e n s a t i n g curve. Then

C
d . i — 1 = i
(- ) 1

d.l. . I a an v 7

V d t 72

By c o m b i n i n g t h e e q u a t i o n s ( 6 ) a n d ( 7 ) t h e f o l l o w i n g e q u a t i o n c a n b e d e r i v e d :

C .i.
d ί—]-ί—1 =0Η-Ϊ«Ι)Θ Η · (12)
_ldtJ 2 vdty,J
N e g l e c t i n g t h e effect of term /' ·θ it is p o s s i b l e t o c a l c u l a t e the amount of
adsorbed hydrogen
 221

T h e d o u b l e c h a r g i n g m e t h o d is a p p l i c a b l e t o e l e c t r o d e s u p o n w h i c h s i g n i f i c a n t
o x i d e f o r m a t i o n c o m m e n c e s at t h e p o t e n t i a l w h e r e t h e a d s o r b e d h y d r o g e n h a s b e e n
ionised to a great extent and a n o d i c dissolution of the metal o c c u r s to a negligible
extent [28].
T h e p r o b l e m s c o n c e r n i n g t h e a p p l y i n g o f t h i s m e t h o d for t h e d e t e r m i n a t i o n o f
a d s o r b e d h y d r o g e n o n n i c k e l will b e d i s c u s s e d l a t e r in m o r e d e t a i l .

( I I I ) P o t e n t i o d y n a m i c m e t h o d s [ 2 2 , 3 0 , 3 1]
T h e m e a s u r e m e n t s o f p o t e n t i o d y n a m i c t r a n s i e n t s is a v e r y w i d e l y u s e d m e t h o d
for t h e d e t e r m i n a t i o n o f t h e a m o u n t o f a d s o r b e d h y d r o g e n for t h e m e t a l s o f t h e
platinum g r o u p U n d e r the conditions w h e r e the p r o c e s s of h y d r o g e n ionisation
and t h e f o l l o w i n g o x i d a t i o n o f t h e s u r f a c e a r e n o t s e p a r a t e d c o n s i d e r a b l y t h e s a m e
d i f f i c u l t i e s a r i s e as w h i l e u s i n g g a l v a n o s t a t i c m e t h o d s .
In t h i s c a s e it is a l s o n e c e s s a r y t o m e a s u r e t w o c u r v e s , o n e o f t h e m s t a r t i n g
f r o m t h e p o t e n t i a l i n v e s t i g a t e d , t h e o t h e r as a c o m p e n s a t i n g c u r v e f r o m a m o r e
anodic potential. T h e a m o u n t of a d s o r b e d h y d r o g e n can then be calculated from
the difference of these curves.

(IV) Overpotential decay [32-40].

T h i s m e t h o d is b a s e d o n t h e a s s u m p t i o n t h a t t h e F a r a d a i c p r o c e s s t a k i n g p l a c e
d u r i n g t h e s t e a d y - s t a t e p o l a r i s a t i o n c o n t i n u e s after t h e i n t e r r u p t i o n o f t h e
polarisation causing self-discharge of the d o u b l e layer
T h e initial p a r t o f o v e r p o t e n t i a l d e c a y c u r v e s c a n b e v e r y s i m p l y u s e d t o
d e t e r m i n e t h e c a p a c i t a n c e o f t h e e l e c t r o d e at t h e initial o v e r p o t e n t i a l :

h
C = , (13)
(dr|/dt) =o t

w h e r e / is t h e c u r r e n t d e n s i t y j u s t b e f o r e t h e i n t e r r u p t i o n .
;

M o r e information can be obtained from o v e r p o t e n t i a l decay m e a s u r e m e n t s by

c a l c u l a t i n g e l e c t r o d e c a p a c i t a n c e a l o n g d e c a y c u r v e s in a w i d e r a n g e of
o v e r p o t e n t i a l s . In t h i s c a s e t h e e l e c t r o d e c a p a c i t a n c e is g i v e n by:

C = ΐ*— , (14)
drj/dt

w h e r e / is t h e s e l f - d i s c h a r g e c u r r e n t d e c r e a s i n g in t i m e . D i f f e r e n t m e t h o d s h a v e
been used to determine the value of / along the overpotential decay curve. The
s i m p l e s t w a y is t o c o m b i n e t h e e q u a t i o n ( 1 1 ) w i t h t h e T a f e l o n e , p r o v i d i n g t h e
r e l a t i o n b e t w e e n i a n d η is t h e s a m e as b e t w e e n / a n d η u n d e r p o l a r i s a t i o n [ 4 8 ] .
(

If t h e p o l a r i s a t i o n c u r v e is n o t l i n e a r , i.e. t h e T a f e l e q u a t i o n is n o t o p e r a t i v e ,
t h i s w a y o f c a l c u l a t i o n is n o t a p p l i c a b l e . T h e m o s t g e n e r a l w a y is t o d e t e r m i n e t h e
v a l u e s o f / at t h e c o r r e s p o n d i n g o v e r p o t e n t i a l o n t h e b a s i s o f e x p e r i m e n t a l η, log i
- c u r v e s [ 2 2 , 4 1 , 4 2 ] . T h e c o m p l i c a t i o n s m a y a r i s e in t h e a p p l i c a t i o n o f p o l a r i s a t i o n
c u r v e s if t h e c o n s i d e r a b l e h y s t e r e s i s is e v i d e n t b e t w e e n t h e c u r v e s m e a s u r e d in
222

opposite directions T h e p o s s i b i l i t i e s t o o v e r c o m e t h i s p r o b l e m will b e d i s c u s s e d

later.
The c a p a c i t a n c e to be determined consists of t w o c o m p o n e n t s - the d o u b l e layer
capacitance C and p s e u d o c a p a c i t a n c e c ^ :
d l

C = C + C (15)
d.l. ps x
'

The appearance of p s e u d o c a p a c i t a n c e is i n d u c e d by t h e v a r i a t i o n of Θ Η with

potential, thus

Η
<> = 1 Η * . (16)

w h e r e q is t h e c h a r g e o f a m o n o l a y e r o f a d s o r b e d h y d r o g e n t h e v a l u e o f w h i c h in
2
t h e c a s e o f n i c k e l is a p p r o x i m a t e l y 2 4 0 m C / c m in a v e r a g e [ 4 3 ] .
It is e v i d e n t t h a t o n l y t h e c h a n g e o f t h e s u r f a c e c o v e r a g e c a n b e d e t e r m i n e d by
the overpotential decay m e a s u r e m e n t s w h e r e a s the absolute value of Θ cannot be Η

d e t e r m i n e d in s u c h a w a y It m u s t b e e m p h a s i s e d t h e p s e u d o c a p a c i t a n c e is n o t
observed if θ is i n d e p e n d e n t of the overpotential in t h e range investigated,
a l t h o u g h the a b s o l u t e value of θ may be remarkably high.
T h e m e t h o d o f o v e r p o t e n t i a l d e c a y is m o r e p r e f e r a b l e for t h e d e t e r m i n a t i o n o f
p s e u d o c a p a c i t a n c e than the other ones described above, since the rate of the
c h a n g e o f p o t e n t i a l a n d t h e s e l f - d i s c h a r g e c u r r e n t d e n s i t y a r e d e t e r m i n e d by t h e
k i n e t i c p a r a m e t e r s o f t h e e l e c t r o c h e m i c a l r e a c t i o n s e q u e n c e i t s e l f i.e. t h e t r u e
a d s o r p t i o n p s e u d o c a p a c i t a n c e is o b s e r v e d e x p e r i m e n t a l l y [ 2 2 ]

(V) Impedance measurements [19,20,23,44-46].

T h e a c i m p e d a n c e m e a s u r e m e n t is o n e o f t h e m o s t w i d e l y u s e d m e t h o d for t h e
d e t e r m i n a t i o n of the c a p a c i t a n c e of the m e t a l - s o l u t i o n interface. Using suitable
e q u i v a l e n t - c i r c u i t s it is p o s s i b l e t o c a l c u l a t e t h e v a l u e s o f t h e C and the p s e u d o ­
c a p a c i t a n c e c a u s e d by t h e c h a n g e o f θ It m u s t b e n o t e d t h a t t h e v a l u e s o f
p s e u d o c a p a c i t a n c e m e a s u r e d by t h i s m e t h o d a r e r e m a r k a b l y d e p e n d e n t o n t h e a c
frequency The most reliable values of a d s o r p t i o n p s e u d o c a p a c i t a n c e could be
o b t a i n e d by c a r r y i n g o u t t h e m e a s u r e m e n t s at an e x t r e m e l y l o w f r e q u e n c y . If t h e
l o w - f r e q u e n c y m e a s u r e m e n t s a r e c a r r i e d o u t at c a t h o d i c p o t e n t i a l s t h e p r o c e e d i n g
o f F a r a d a i c p r o c e s s e s will i n t r o d u c e t h e a d d i t i o n a l c o m p l i c a t i o n s in t h e
interpretation of the results
It is o f i n t e r e s t t o c o n s i d e r t h e n a t u r e o f t h e d e g r e e o f c o v e r a g e d e t e r m i n e d by
different m e t h o d s H y d r o g e n a d s o r b e d on m e t a l s u r f a c e c a n b e d i v i d e d i n t o t w o
parts:

Θ„ = ν + Θ - ,
Η (17)

w h e r e Θ is t h e d e g r e e o f c o v e r a g e w i t h h y d r o g e n a d s o r b e d o n t h e s u r f a c e at t h e
r e v e r s i b l e h y d r o g e n p o t e n t i a l i.e. at η 0, is t h e d e g r e e o f c o v e r a g e w i t h
 223

cathodic hydrogen added to the θ ° during the cathodic process due to the
retarded p r o c e s s of the removal of adsorbed hydrogen. The amount of hydrogen
d e t e r m i n e d by b o t h g a l v a n o s t a t i c c h a r g i n g c u r v e s a n d p o t e n t i o d y n a m i c c y c l e s
c o r r e s p o n d s to the total value of adsorbed hydrogen θ providing the range of
p o t e n t i a l s i n v e s t i g a t e d is w i d e e n o u g h .
The overpotential decay method enables to calculate the part of adsorbed
h y d r o g e n i o n i s e d d u r i n g t h e o v e r p o t e n t i a l d e c a y i.e. o n l y t h e v a l u e o f θ , b u t c

o f t e n e v e n l e s s a s t h e o v e r p o t e n t i a l d e c a y c u r v e , as a r u l e , d o e s n o t r e a c h t h e
v a l u e η = 0. B y t h e i m p e d a n c e m e a s u r e m e n t s it is p o s s i b l e t o d e t e r m i n e o n l y t h e
p a r t o f t h e a d s o r b e d h y d r o g e n w h i c h f o l l o w s t h e slight v a r i a t i o n o f p o t e n t i a l in t h e
c o n d i t i o n s o f t h e a p p l i e d a l t e r n a t i n g c u r r e n t . It is e v i d e n t t h a t t h e l o w e r is t h e
f r e q u e n c y , t h e l a r g e r is t h e a m o u n t o f a d s o r b e d h y d r o g e n e n a b l e t o r e s p o n d t o t h e
c h a n g e o f p o t e n t i a l , i.e. t h e h i g h e r is t h e p s e u d o c a p a c i t a n c e . T h e r e f o r e t h e v a l u e s
o f Θ d e t e r m i n e d by d i f f e r e n t m e t h o d s n e e d n o t t o c o i n c i d e e n t i r e l y d u e t o t h e
Η

specific features of the m e t h o d s .

4. E X P E R I M E N T A L RESULTS

4 . 1 . I n f l u e n c e of t h e s u r f a c e s t a t e .
T h e e x p e r i m e n t a l r e s u l t s o f t h e h y d r o g e n o v e r p o t e n t i a l o n n i c k e l o b t a i n e d in
d i f f e r e n t l a b o r a t o r i e s by d i f f e r e n t r e s e a r c h e r s a r e c o n s i d e r a b l y v a r i a b l e as w e l l as
on most of other metals. Such a variation of results may be connected with the
low reproducibility of the experimental data. In fact t h e s i t u a t i o n is n o t so
h o p e l e s s if t o c o n s i d e r t h e r e s u l t s o f r e l i a b l e i n v e s t i g a t i o n s a n d n o t t o t a k e i n t o
a c c o u n t t h e o n e s o b t a i n e d in t h e c o n d i t i o n s o f o b v i o u s l y i m p u r e s o l u t i o n s o r
electrode material
The kinetic regularities of a reaction on a solution/electrode interface are
i n f l u e n c e d by a r e m a r k a b l y g r e a t e r n u m b e r o f f a c t o r s t h a n in t h e c a s e o f
h o m o g e n e o u s p r o c e s s e s in t h e b u l k o f s o l u t i o n T h e r e f o r e c o m p a r i n g different
results the m e t h o d s of preparation of electrodes must be taken into particular
account. It m u s t b e n o t e d t h a t t h e s t a n d a r d i s i n g o f e x p e r i m e n t a l c o n d i t i o n s as
w e l l as c o n t r o l l i n g o f o c c a s i o n a l v a r i a b l e s is c o n s i d e r a b l y c o m p l i c a t e d in t h e c a s e
of h e t e r o g e n e o u s processes.
T h e e s s e n t i a l r o l e o f t h e s u r f a c e s t a t e o f n i c k e l in t h e r a t e o f t h e h y d r o g e n
e v o l u t i o n r e a c t i o n b o t h in a c i d i c a n d a l k a l i n e s o l u t i o n s c a n b e c l e a r l y s e e n o n
F i g . 1 a n d 2.
In t h e s e r i e s o f m e a s u r e m e n t s p r e s e n t e d b e l o w all t h e p o s s i b l e f a c t o r s
influencing the h y d r o g e n e v o l u t i o n reaction (such as the c o n c e n t r a t i o n and
c o m p o s i t i o n of the solution, the conditions of polarisation, t e m p e r a t u r e , etc ) are
k e p t u n c h a n g e a b l e e x c e p t t h e m e t h o d for p r e p a r a t i o n o f t h e e l e c t r o d e s u r f a c e
W h e r e a s t h e f o l l o w i n g m e t h o d s for t h e p r e p a r a t i o n o f n i c k e l e l e c t r o d e s h a v e b e e n
u s e d [47-49]:
224

- log i (A/cm )
Fig. 1. P o l a r i s a t i o n c u r v e s of differently p r e p a r e d nickel e l e c t r o d e s in 0 25 Μ H S 0 2 4

solution:
1 - m e c h a n i c a l l y p o l i s h e d , 2 - h e a t e d in h y d r o g e n , 3 - c h e m i c a l l y p o l i s h e d

7 6 5 4 3
2
- log i ( A / c m )
F i g 2 P o l a r i s a t i o n c u r v e s for different n i c k e l e l e c t r o d e s in 0.5 Μ K O H s o l u t i o n
1 - c h e m i c a l l y p o l i s h e d , 2 - 4 - m e c h a n i c a l l y p o l i s h e d ; 2 - c o r r e c t e d by t h e
r o u g h n e s s f a c t o r , 4 - after a n o d i c a c t i v a t i o n
 225

a) m e c h a n i c a l p o l i s h i n g o f t h e s u r f a c e w i t h e m e r y p a p e r and t h e f o l l o w i n g
f i n i s h i n g w i t h fine q u a r z p o w d e r a n d r i n s i n g t h o r o u g h l y w i t h t r i d i s t i l l e d w a t e r
before every experiment
b ) c o n v e n t i o n a l m e c h a n i c a l p o l i s h i n g w a s f o l l o w e d by c h e m i c a l p o l i s h i n g in a
s o l u t i o n c o n t a i n i n g c o n c e n t r a t e d a c e t i c a c i d , a c e t i c a n h y d r i d e a n d n i t r i c a c i d , at
t e m p e r a t u r e 6 0 ° Γ d u r i n g 15 s.
c) after m e c h a n i c a l p o l i s h i n g o f e l e c t r o d e s h e a t i n g in d r y o x y g e n - f r e e h y d r o g e n at
n e a r l y 5 0 0 ° C for 2 h o u r s .
The data c h a r a c t e r i s i n g h y d r o g e n o v e r p o t e n t i a l on the described a b o v e t y p e s of
n i c k e l e l e c t r o d e s a r e p r e s e n t e d in T a b l e 1

Table 1
Characteristics of hydrogen evolution reaction on polycrystalline electrodes.
a, V b, V C, a, V l o
" S Jo 2
Electrode (A/cm ) μΡ^πι (A/cm ) 2 2

Apparent Corrected
Acidic solution
Mech. polished 0.595 0.122 4.9 80 0.671 5.5
Chem. polished 0.685 0.092 7.5 20 0.685 7.5
H e a t e d in h y d r o g e n 0.705 0.100 7.1 40 0.725 7.4
Alkaline solution
Mech polished 0.585 0.115 5.1 80 0.655 5.7
Chem. polished 0.800 0.110 7.3 20 0.800 7.3

It is n a t u r a l t o a s s u m e t h a t t h e s i g n i f i c a n t d e p e n d e n c e o f t h e h y d r o g e n
o v e r p o t e n t i a l o n t h e p r e p a r a t i o n m e t h o d o f t h e e l e c t r o d e m a y b e c a u s e d by t h e
d i f f e r e n c e in t h e a r e a s o f real s u r f a c e i.e. by t h e d i f f e r e n t s u r f a c e r o u g h n e s s . T h e
d e t e r m i n a t i o n o f t h e v a l u e o f t h e r e a l s u r f a c e a r e a o f p o l y c r y s t a l l i n e e l e c t r o d e s is
quite complicated On certain assumptions, the a p p r o x i m a t e determination can be
done using conventional electrochemical measurements For instance the
d e t e r m i n a t i o n o f t h e real s u r f a c e a r e a m a y b e b a s e d o n t h e p r o p o r t i o n a l r e l a t i o n
b e t w e e n that with the c a p a c i t a n c e of the double layer
O n e o f t h e m o s t c o n v e n i e n t m e t h o d s for t h e d e t e r m i n a t i o n o f t h e c a p a c i t a n c e o f
d o u b l e l a y e r u n d e r c a t h o d i c p o t e n t i a l s is t o a p p l y t h e initial p a r t s o f o v e r p o t e n t i a l
decay curves provided the capacitance obtained corresponds to that of the double
l a y e r T h i s a s s u m p t i o n is r e a s o n a b l e in t h e c a s e o f n i c k e l e l e c t r o d e s at sufficiently
h i g h o v e r p o t e n t i a l s , w h e r e t h e c a p a c i t a n c e is a p p r o x i m a t e l y i n d e p e n d e n t o f
overpotential (Fig. 3).
226

Ι 200
zL

I 100

cu
u

0.4 0.3 0.2 0.1

Overpotential, V
Fig 3. T h e d e p e n d e n c e o f t h e c a p a c i t a n c e o n t h e o v e r p o t e n t i a l for d i f f e r e n t l y
p r e p a r e d n i c k e l e l e c t r o d e s in 0 . 2 5 Μ H S 0 s o l u t i o n :
2 4

1 - m e c h a n i c a l l y p o l i s h e d , 2 - h e a t e d in h y d r o g e n , 3 - c h e m i c a l l y p o l i s h e d

T h e d o u b l e l a y e r c a p a c i t a n c e d a t a o b t a i n e d from t h e o v e r p o t e n t i a l d e c a y
m e a s u r e m e n t s a r e p r e s e n t e d in T a b l e 1. It c a n b e s e e n t h a t t h e l o w e s t v a l u e o f t h e
2
c a p a c i t a n c e 18 2 0 p F p e r c m o f t h e a p p a r e n t a r e a o f s u r f a c e is e x h i b i t e d by
c h e m i c a l l y p o l i s h e d n i c k e l e l e c t r o d e s . If t o c o m p a r e t h i s v a l u e w i t h t h a t o b t a i n e d
2
for l i q u i d m e r c u r y at n e g a t i v e s u r f a c e c h a r g e 16 . 18 μ Ρ / c m [44,45], the
c h e m i c a l l y p o l i s h e d n i c k e l s e e m s t o b e ideally s m o o t h , w h i c h is s o m e w h a t
questionable A s t h e real d o u b l e l a y e r c a p a c i t a n c e o f i d e a l l y s m o o t h n i c k e l is n o t
k n o w n e x a c t l y y e t , it is n o t p o s s i b l e t o d e t e r m i n e t h e a b s o l u t e v a l u e o f s u r f a c e
r o u g h n e s s of the nickel e l e c t r o d e s used
F o r t u n a t e l y , in t h i s a n a l y s i s w e a r e i n t e r e s t e d in t h e r e l a t i o n o f t h e real s u r f a c e
a r e a s o f d i f f e r e n t l y p r e p a r e d e l e c t r o d e s r a t h e r t h a n in t h e a b s o l u t e v a l u e s o f t h e m .
T h e r e f o r e it is r e a s o n a b l e t o t a k e t h e d o u b l e l a y e r c a p a c i t a n c e o f c h e m i c a l l y
p o l i s h e d e l e c t r o d e s h a v i n g t h e l o w e s t v a l u e o f t h a t o b t a i n e d for n i c k e l in t h i s w o r k
as a b a s e o f c o m p a r i s o n , t h a t m e a n s t h e f a c t o r o f s u r f a c e r o u g h n e s s o f t h e s e
e l e c t r o d e s is r e g a r d e d c o n v e n t i o n a l l y t o b e e q u a l t o 1 u n i t . This value of
c a p a c i t a n c e c a n b e u s e d for t h e c a l c u l a t i o n o f t h e r e l a t i v e f a c t o r s o f s u r f a c e
r o u g h n e s s o f all o t h e r t y p e s o f e l e c t r o d e s in r e s p e c t o f c h e m i c a l l y p o l i s h e d n i c k e l
which yields us the opportunity to correct the values of current densities and the
c o r r e s p o n d i n g v a l u e s o f a a n d i o n t h e b a s i s o f t h a t ( T a b l e 1).
o

A s c a n b e s e e n f r o m T a b l e 1 t h e c a p a c i t a n c e o f t h e e l e c t r o d e s h e a t e d in
h y d r o g e n differs from t h a t o f c h e m i c a l l y p o l i s h e d e l e c t r o d e s n e a r l y t w i c e a n d t h e
relative factor of surface r o u g h n e s s of these electrodes can be considered equal to
 227

2 The relation of the values of c a p a c i t a n c e of mechanically and chemically

p o l i s h e d n i c k e l is a p p r o x i m a t e l y 4 b o t h for a c i d i c a n d a l k a l i n e s o l u t i o n s a n d
t h e r e f o r e it c a n b e c o n c l u d e d t h a t t h e f a c t o r o f s u r f a c e r o u g h n e s s o f m e c h a n i c a l l y
p o l i s h e d e l e c t r o d e s is n e a r l y e q u a l t o 4 i n d e p e n d e n t l y o f t h e p H o f t h e s o l u t i o n .
T h i s r e s u l t i m p l i e s t h a t c o r r o s i o n o f n i c k e l in a c i d i c s o l u t i o n s u n d e r c a t h o d i c
p r o t e c t i o n u s e d in t h i s w o r k h a s n e g l i g i b l e effect o n t h e r o u g h n e s s o f t h e s u r f a c e .
T h e p o l a r i s a t i o n c u r v e s c o r r e s p o n d i n g t o t h e c o r r e c t e d by t h e " r e a l " c u r r e n t
d e n s i t i e s a r e s h o w n in F i g 2 a n d 4. It c a n b e s e e n t h a t t h e v a l u e s o f o v e r p o t e n t i a l
o n d i f f e r e n t e l e c t r o d e s h a v e c o m e c l o s e r t o e a c h o t h e r if t h e r e l a t i v e s u r f a c e
r o u g h n e s s is t a k e n i n t o c o n s i d e r a t i o n . T h i s a p p r o a c h is p a r t i c u l a r l y o b v i o u s in t h e
case of the polarisation curves of chemically polished e l e c t r o d e s and of e l e c t r o d e s
h e a t e d in h y d r o g e n , w h e r e t h e d i f f e r e n c e in o v e r p o t e n t i a l h a s a l m o s t d i s a p p e a r e d
( F i g 4 ) , b u t t h e o v e r p o t e n t i a l o f m e c h a n i c a l l y p o l i s h e d e l e c t r o d e s h a s still
r e m a i n e d r e m a r k a b l y l o w e r o f t h a t o n o t h e r e l e c t r o d e s in a c i d i c s o l u t i o n s a s w e l l
as in a l k a l i n e s o l u t i o n s ( F i g . 2 a n d 4 ) .
F r o m t h e s e o b s e r v a t i o n s it c a n b e c o n c l u d e d t h a t t h e l o w e r hydrogen
o v e r p o t e n t i a l , i . e . , t h e g r e a t e r a c t i v i t y in r e s p e c t o f t h e h y d r o g e n e v o l u t i o n
r e a c t i o n o n m e c h a n i c a l l y p o l i s h e d e l e c t r o d e s c a n n o t b e e x p l a i n e d o n l y by t h e
g r e a t e r real area of surface than that of o t h e r ones T h e catalytic activity of this
t y p e o f e l e c t r o d e s is o b v i o u s l y i n c r e a s e d in t h e r e s u l t o f m e c h a n i c a l a c t i v a t i n g o f
t h e s u r f a c e l a y e r o f m e t a l by m e a n s o f m e c h a n i c a l p o l i s h i n g , as t h i s p r o c e d u r e
i n t r o d u c e s e s s e n t i a l c h a n g e s in t h e s u r f a c e s t r u c t u r e - t h e g e n e r a t i n g o f a g r e a t
n u m b e r o f d e f e c t s a n d d i s l o c a t i o n s in it. C h e m i c a l p o l i s h i n g o f an e l e c t r o d e
r e m o v e s the surface layer d e f o r m e d d u r i n g the mechanical t r e a t m e n t by dissolving
it in t h e e l e c t r o l y t e s o l u t i o n . T h e r e n e w e d s u r f a c e o f m e t a l h a s l o w e r c a t a l y t i c
activity and consequently higher hydrogen overpotential. Such a remarkable
d e c r e a s e o f t h e m e t a l a c t i v i t y b e c o m e s e v i d e n t a l s o in t h e c a s e o f h e a t i n g t h e
e l e c t r o d e s in h y d r o g e n at t e m p e r a t u r e s h i g h e n o u g h t o r e c r y s t a l l i s e t h e s u r f a c e
layer.
In t h e c a s e o f a l k a l i n e s o l u t i o n s an a d d i t i o n a l m e t h o d o f t h e p r e - t r e a t m e n t o f
nickel surface can be applied - the anodic activation of a mechanically polished or
o t h e r w i s e p r e p a r e d e l e c t r o d e [ 4 8 , 5 1 ] . S u c h k i n d o f a n o d i c a c t i v a t i o n is c o m m o n
for t h e m e t a l s o f p l a t i n u m g r o u p [ 5 2 - 5 4 ] . During the anodic activation the
p o t e n t i a l o f n i c k e l e l e c t r o d e a t t a i n s t h e v a l u e o f a p p r o x i m a t e l y 1.5 V , i.e. p a s s e s
t h r o u g h the regions of the formation of Ni(II) and Ni(III) h y d r o x i d e s [ 5 5 ] . U n d e r
s u c h h i g h l y o x i d i s i n g c o n d i t i o n s m o s t o f t h e i m p u r i t i e s will b e d e s o r b e d f r o m t h e
s u r f a c e l a y e r a n d a f t e r t h e f o l l o w i n g c a t h o d i c r e d u c t i o n t h e s u r f a c e will b e l e s s
contaminated The lower overpotential of anodically activated nickel electrodes,
c o m p a r e d to the overpotential before activating (Fig 2), cannot be related to the
h i g h e r f a c t o r o f s u r f a c e r o u g h n e s s a s t h e c a p a c i t a n c e o f b o t h t y p e s o f e l e c t r o d e s is
228

I I I I I U
6 5 4 3 2
^ - log i (A/cm ) 2

Fig 4 P o l a r i s a t i o n c u r v e s o f n i c k e l e l e c t r o d e s in 0 2 5 Μ H S 0 2 4 solution
c o r r e c t e d by t h e s u r f a c e r o u g h n e s s f a c t o r :
1 - m e c h a n i c a l l y p o l i s h e d , 2 - h e a t e d in h y d r o g e n , 3 - c h e m i c a l l y p o l i s h e d

almost the same Q u e s t i o n s m a y a r i s e in r e s p e c t o f t h e c o m p l e t e n e s s o f t h e

following reduction during the following cathodic polarisation. Although the
i n c r e a s e o f t h e s u r f a c e r o u g h n e s s h a s n o t b e e n o b s e r v e d as a r e s u l t o f a n o d i c
a c t i v a t i o n , it c a n n o t b e s t a t e d firmly t h a t all t h e o x y g e n - c o m p o u n d s o r i g i n a t e d at
h i g h a n o d i c p o t e n t i a l s a r e e n t i r e l y r e d u c e d h a v i n g n o n e c a t a l y t i c effect o n t h e
HER
O n e of the factors which may influence the regularities of c a t h o d i c h y d r o g e n
e v o l u t i o n is t h e i n g r e s s o f h y d r o g e n i n t o m e t a l [ 5 6 ] The adsorbed hydrogen
a t o m s o r i g i n a t e d in t h e p r o c e s s o f t h e d i s c h a r g e o f p r o t o n s c a n b e t r a n s f e r r e d
a c r o s s t h e m e t a l s u r f a c e i n t o t h e m e t a l : H j —> H b
a( s a s

In t h i s m e c h a n i s m h y d r o g e n e n t e r i n g t h e m e t a l g o e s t h r o u g h t h e s a m e a d s o r b e d
i n t e r m e d i a t e as t h a t l e a d i n g t o h y d r o g e n e v o l u t i o n [ 5 7 ] A c c o r d i n g t o an
a l t e r n a t i v e m e c h a n i s m h y d r o g e n e n t e r s t h e m e t a l in t h e s a m e e l e m e n t a r y act as
t h a t in w h i c h it is d i s c h a r g e d [ 2 0 , 5 8 ] .

H +
i>- H ihs

In t h i s c a s e t h e i n t e r m e d i a t e s t a t e s for t h e e n t r y and h y d r o g e n e v o l u t i o n are

different T h e first m e c h a n i s m h a s b e e n t h e m o r e a c c e p t e d o n e [ 5 6 ] .
 229

A c c o r d i n g t o D o w d e n ' s t h e o r y t h e c a t a l y t i c a c t i v i t y o f d - m e t a l s is c o n n e c t e d
with the d-character of metals [59]. The l s - e l e c t r o n of adsorbed hydrogen atom
fills t h e 3 d - h o l e s o f n i c k e l a n d t h e c a t a l y t i c a c t i v i t y o f f o r m e d N i - H a d sis m u c h
l o w e r T h e s i m i l a r d e c r e a s e o f t h e c a t a l y t i c a c t i v i t y o f n i c k e l h a s b e e n o b s e r v e d in
t h e c a s e o f N i - C u a l l o y w h e r e 4 s - e l e c t r o n s o f c o p p e r fill t h e 3 d - h o l e s o f N i [ 6 0 ] .
It is r e a s o n a b l e t o c o n s i d e r t h a t t h e i n g r e s s o f h y d r o g e n will i n c r e a s e t h e h y d r o g e n
overpotential.
T h e i n g r e s s o f h y d r o g e n i n t o n i c k e l i n v e s t i g a t e d in t h e g r e a t e r p a r t o f w o r k s , is
i n d u c e d by t h e u s e o f v a r i o u s s t i m u l a t o r s , h a v i n g r e m a r k a b l e effect o n t h e k i n e t i c s
of hydrogen evolution and therefore these results cannot be considered directly.
N e v e r t h e l e s s , in s o m e r e p o r t s p u b l i s h e d in r e c e n t y e a r s t h e h y d r o g e n a t i o n o f
n i c k e l h a s b e e n r e a l i s e d w i t h o u t any s t i m u l a t o r s [ 6 1 , 6 2 ] . F o r i n s t a n c e in [ 6 2 ]
n i c k e l h a s b e e n h y d r o g e n a t e d g a l v a n o s t a t i c a l l y in H S 0 s o l u t i o n s w i t h c u r r e n t
2 4

d e n s i t y o f 5 χ 10 2 2
A/cm during 2 hours After interruption of the c a t h o d i c
polarisation the sample was dissolved anodically. It w a s f o u n d t h a t d u r i n g t h e
initial p e r i o d o f a n o d i c d i s s o l u t i o n h y d r o g e n is i o n i s e d p r e f e r a b l y w h i l e n i c k e l
d i s s o l u t i o n is i n h i b i t e d
It is c l e a r t h a t h y d r o g e n a b s o r b e d in n i c k e l h a s r e m a r k a b l e i n f l u e n c e o n t h e
energetic p r o p e r t i e s of nickel. Consequently, the possibility of h y d r o g e n a t i o n of
n i c k e l d u r i n g t h e m e a s u r e m e n t s in p u r e c o n d i t i o n s a n d t h e i n f l u e n c e o f it o n t h e
rate of h y d r o g e n evolution must be taken into particular account To elucidate the
possible influence of h y d r o g e n a t i o n on h y d r o g e n overpotential the results of the
H E R m e a s u r e m e n t s r e a l i s e d in d i f f e r e n t c o n d i t i o n s m u s t b e c o m p a r e d . T h e r e s u l t s
o b t a i n e d o n t h e e l e c t r o d e s h e a t e d in h y d r o g e n [ 6 3 - 6 6 ] a r e v a l u a b l e in t h i s r e s p e c t .
In [ 6 6 ] t h e r a p i d m e a s u r e m e n t s o f o v e r p o t e n t i a l w e r e c a r r i e d o u t in t h e d i r e c t i o n
o f i n c r e a s i n g c u r r e n t d e n s i t y o n l y u s i n g a fresh e l e c t r o d e for e v e r y r u n . C o n t r a r y
t o t h i s w o r k t h e o v e r p o t e n t i a l m e a s u r e m e n t s in [ 6 3 , 6 4 ] w e r e c a r r i e d o u t after t h e
p o t e n t i a l h a d r e a c h e d a s t e a d y s t a t e v a l u e . T h e r e f o r e , it is r e a s o n a b l e t o s u p p o s e
t h e o v e r p o t e n t i a l in t h e l a t t e r c a s e m a y h a v e a h i g h e r v a l u e as t h e r e s u l t o f t h e
h y d r o g e n a t i o n of the electrode. Unfortunately the data of [63,64] are not
p r e s e n t e d in d e t a i l e n o u g h t o m a k e r e l i a b l e c o n c l u s i o n s .
Special investigations have been carried out [47] to elucidate the d e p e n d e n c e of
o v e r p o t e n t i a l o n t h e p o l a r i s a t i o n c o n d i t i o n s o f e l e c t r o d e s p r e v i o u s l y h e a t e d in
h y d r o g e n . It h a s b e e n e s t a b l i s h e d t h a t it is p o s s i b l e t o o b t a i n a s e r i e s o f d i f f e r e n t
polarisation curves on the same e l e c t r o d e depending on the m a g n i t u d e of the
c u r r e n t d e n s i t y t o w h i c h t h e e l e c t r o d e h a s b e e n p o l a r i s e d after t h e i m m e r s i o n o f it
into the solution (Fig. 5). Increasing the polarising current density higher and
h i g h e r v a l u e s o f h y d r o g e n o v e r p o t e n t i a l a r e a c h i e v e d u n t i l t h e s t e a d y s t a t e is
realised T h e s e m e a s u r e m e n t s s h o w that different h y d r o g e n o v e r p o t e n t i a l values
c a n b e o b t a i n e d by t h e v a r i a t i o n o f p o l a r i s a t i o n c o n d i t i o n s o f t h e e l e c t r o d e , b e i n g
consistent with the supposition of the d e p e n d e n c e of the degree of h y d r o g e n a t i o n
on the applied current density As the h y d r o g e n overpotential on chemically
p o l i s h e d e l e c t r o d e s h a s a p p r o x i m a t e l y t h e s a m e v a l u e as in t h e c a s e o f t h e
e l e c t r o d e s h e a t e d in h y d r o g e n ( a f t e r c o r r e c t i n g t h e m o n t h e f a c t o r o f s u r f a c e
r o u g h n e s s ) it is r e a s o n a b l e t o s u p p o s e t h a t c h e m i c a l l y p o l i s h e d e l e c t r o d e s a r e a l s o
h y d r o g e n a t e d to similar extent
230

6 5 4 3
- log i (A/cm ) 2

Fig 5. P o l a r i s a t i o n c u r v e s o f n i c k e l e l e c t r o d e h e a t e d in h y d r o g e n in 0 . 2 5 Μ H S 0 2 4

solution, polarised to different c a t h o d i c current densities.

4.2. Single C r y s t a l Faces

The influence of v a r i o u s factors on the regularities of the p r o c e s s of h y d r o g e n
e v o l u t i o n o n n i c k e l c a n b e m o r e d e e p l y c o n s i d e r e d by t h e i n v e s t i g a t i o n s o f t h i s
p r o c e s s on different faces of m o n o c r y s t a l l i n e e l e c t r o d e s The influence of nickel
c r y s t a l l o g r a p h i c s t r u c t u r e o n t h e H E R h a s b e e n s t u d i e d o n l y in s o m e w o r k s [ 6 7 -
70] and the results obtained are considerably different. A s a m a t t e r o f fact t h e
d i f f e r e n c e in q u a l i t y o f n i c k e l s i n g l e c r y s t a l s m a y c a u s e r e m a r k a b l e v a r i a t i o n o f t h e
results obtained
In o u r w o r k t h e e l e c t r o d e s c u t from h i g h l y p u r e n i c k e l s i n g l e c r y s t a l ( a b o u t
9 9 . 9 9 9 % N i ) p r o d u c e d by c o n t a i n e r l e s s v a c u u m z o n e m e l t i n g w e r e a p p l i e d The
s i n g l e c r y s t a l s w e r e a l i g n e d via X - r a y d i f f r a c t i o n . T h e m i s a l i g n m e n t a n g e l s o f t h e
s t u d i e d s u r f a c e s o f d i f f e r e n t e l e c t r o d e s w e r e b e t w e e n 0.5 a n d 1.5° Prior to each
experiment the electrodes were polished chemically. The surface of prepared
electrodes were perfectly smooth when viewed with a magnification of 4 5 0 * . The
p o l a r i s a t i o n c u r v e s m e a s u r e d in 0 . 2 5 Μ Η S O , s o l u t i o n at t h e l o w - i n d e x f a c e s a r e
r
2 4
s h o w n in Fig 6 a n d t h e c h a r a c t e r i s t i c s o f t h e H E R a r e p r e s e n t e d in T a b l e 2 [ 7 0 ]
O n e c a n s e e t h a t e x c e p t for face ( 1 1 1 ) , h y d r o g e n o v e r p o t e n t i a l d o e s n o t d e p e n d
e s s e n t i a l l y on t h e c r y s t a l l o g r a p h i c s t r u c t u r e o f n i c k e l T h i s r e s u l t is c o n s i s t e n t
with that obtained on s o m e o t h e r metals [21]
 231

Table 2
C h a r a c t e r i s t i c s o f h y d r o s*en evolution reaction on single crystal nickel electrodes
Face a, V b, V -log i ( A / c m ) 0
2

(100) 0.76 0.113 6.7

(110) 0.635 0.102 6 8
(in) 0.83 0.117 6.5

- log i (A/cm ) 2

Fig. 6 P o l a r i s a t i o n c u r v e s in t h e s o l u t i o n o f 0 . 2 5 Μ H S 0 2 4 s o l u t i o n for single

crystal faces of nickel:
1 - f a c e ( 1 1 0 ) , 2 - face ( 1 0 0 ) , 3 - ( 1 1 1 ) .

T w o p o s s i b l e e x p l a n a t i o n s c a n b e o f f e r e d a b o u t t h e v e r y slight effect o f t h e
c r y s t a l l o g r a p h i c s t r u c t u r e of the metal on h y d r o g e n o v e r p o t e n t i a l [ 2 1 ] . T h e high
c o v e r a g e w i t h a d s o r b e d h y d r o g e n far from e q u i l i b r i u m m a y s m o o t h d o w n t h e
d i f f e r e n c e s in t h e h y d r o g e n a d s o r p t i o n e n e r g y f r o m f a c e t o f a c e . A n o t h e r p o s s i b l e
explanation can be that the competition of solvent adsorption with hydrogen
a d s o r p t i o n a l m o s t c o m p e n s a t e s t h e d i f f e r e n c e s t h a t w o u l d e x i s t for t h e b a r e m e t a l
surface [71]. Electroreflectance spectra suggest that water molecules are
chemisorbed on d-metals with partial injection of electrons into the d-band [71].
T h i s c i r c u m s t a n c e may* e s s e n t i a l l y effect t h e s t r e n g t h o f M e - H b o n d , a s t h e
a d s o r b e d h y d r o g e n a l s o i n j e c t s its e l e c t r o n i n t o t h e d - b a n d . It is e v i d e n t t h a t t h e
interaction of adsorbed h y d r o g e n and w a t e r molecules can not be r e d u c e d to the
232

c o m p e t i t i o n o f t h e m for a d s o r p t i o n s i t e s o n t h e s u r f a c e as t h e s i g n i f i c a n t c h a n g e o f
the energetic properties of the surface takes place.
It h a s b e e n e s t a b l i s h e d [ 7 0 ] t h a t t h e b e h a v i o u r o f n i c k e l face ( 1 1 1 ) in r e s p e c t o f
h y d r o g e n e v o l u t i o n s t a n d s o u t if t o c o m p a r e w i t h t h a t o f o t h e r f a c e s . It w a s f o u n d
t h a t η o n face ( 1 1 1 ) c o n t i n u e s t o i n c r e a s e after t h e i m m e r s i o n o f t h e e l e c t r o d e in
t h e s o l u t i o n for m u c h l o n g e r t i m e t h a n o n o t h e r s i n g l e c r y s t a l f a c e s
A c c o r d i n g t o [ 7 2 ] t h e a m o u n t o f a b s o r b e d h y d r o g e n in n i c k e l is d e p e n d e n t o n
t h e c r y s t a l l o g r a p h i c s t r u c t u r e o f m e t a l and is m o r e e s s e n t i a l for t h e c l o s e - p a c k e d
face ( 1 1 1 ) T h e r e f o r e it is r e a s o n a b l e t o c o n c l u d e t h a t t h e h i g h e r o v e r v o l t a g e o f
t h e H E R o n n i c k e l face ( 1 1 1 ) o b s e r v e d in [ 7 0 ] is c a u s e d by t h e i n g r e s s o f l a r g e r
a m o u n t s o f a t o m i c h y d r o g e n t h a n for o t h e r f a c e s

4.3. Low a n d high c u r r e n t densities.

The investigation of the regularities of the cathodic hydrogen evolution under
e x t r e m e c o n d i t i o n s p r o v i d e s w i t h a d d i t i o n a l i n f o r m a t i o n v a l u a b l e in r e s p e c t o f t h e
better u n d e r s t a n d i n g of the mechanism of overall p r o c e s s
T h e m e a s u r e m e n t s o f t h e H E R at e x t r e m e l y l o w c u r r e n t d e n s i t i e s in a c i d i c
s o l u t i o n s a r e n o t a v a i l a b l e as in t h e s e c o n d i t i o n s t h e v a l u e o f s t a t i o n a r y p o t e n t i a l
b e i n g set u p is d e t e r m i n e d by t h e r e l a t i o n b e t w e e n t h e r a t e o f c a t h o d i c h y d r o g e n
e v o l u t i o n and the rate of the dissolution of nickel, as b o t h p r o c e s s e s are
proceeding under these conditions with considerable rate .
In t h e c a s e o f a l k a l i n e s o l u t i o n s t h e r a t e o f d i s s o l u t i o n is e s s e n t i a l l y l o w e r a n d
the potential of the reversible h y d r o g e n e l e c t r o d e can be attained with great
precise [74] T h u s t h e p o l a r i s a t i o n m e a s u r e m e n t s r e f l e c t first o f all t h e
regularities of the p r o c e s s e s of cathodic evolution and anodic ionisation of
hydrogen. At sufficiently l o w o v e r p o t e n t i a l s w h e r e η « R T / F t h e r e l a t i o n s h i p
b e t w e e n η a n d / is l i n e a r ( F i g . 7) a n d t h e s l o p e o f t h i s s t r a i g h t line c a n b e u s e d for
the determination of the stoichiometric number ν [16,75,76]. The stoichiometric
n u m b e r s h o w s t h e n u m b e r o f a c t s o f t h e r a t e - d e t e r m i n i n g s t e p n e c e s s a r y for o n e
act o f t h e o v e r a l l r e a c t i o n a n d is g i v e n by

ν = 2ΐ„/(ώ/άη) η = ( ) (18)

A c c o r d i n g t o t h i s d e f i n i t i o n a v a l u e 2 for ν s h o w s t h a t t h e d i s c h a r g e s t e p is r a t e -
determining of the process [76]
If t h e v a l u e o f e x t r a p o l a t e d e x c h a n g e c u r r e n t d e n s i t y / o b t a i n e d by
e x t r a p o l a t i n g / f r o m r e m a r k a b l y l a r g e r o v e r p o t e n t i a l s t o η = 0, is u s e d for
c a l c u l a t i o n , t h e a p p a r e n t s t o i c h i o m e t r i c n u m b e r v' is t o b e o b t a i n e d [ 5 0 , 7 6 , 7 7 ] ;

v'=2ii /(di/dr|)
) T 1 = ( ) (18a)

T h e a p p a r e n t s t o i c h i o m e t r i c n u m b e r is a r b i t r a r y t o a c e r t a i n e x t e n t s i n c e it
d e p e n d s o n t h e s e c t i o n o f t h e p o l a r i s a t i o n c u r v e c h o s e n for t h e d e t e r m i n a t i o n o f
 233

F i g . 7. D e p e n d e n c e o f h y d r o g e n o v e r p o t e n t i a l o n c u r r e n t d e n s i t y at low
p o l a r i s a t i o n in 0.5 Μ K O H s o l u t i o n o n m e c h a n i c a l l y p o l i s h e d n i c k e l .

/ ' T h e p r o b l e m s c o n n e c t e d w i t h i n t e r p r e t a t i o n o f t h e v a l u e s o f ν a r e d i s c u s s e d in
d e t a i l in [ 5 0 ] .
T h e e x p e r i m e n t a l v a l u e s o f ν for n i c k e l d e t e r m i n e d in d i f f e r e n t w o r k s a r e v e r y
c l o s e t o 2 [ 5 0 , 6 6 , 7 3 ] , i m p l y i n g t h a t t h e p r o t o n d i s c h a r g e is t h e r a t e - d e t e r m i n i n g
step of hydrogen evolution N e v e r t h e l e s s , it m u s t b e p o i n t e d o u t t h a t t h e
d e t e r m i n e d p a r a m e t e r is a c t u a l l y t h e a p p a r e n t s t o i c h i o m e t r i c n u m b e r v\ t h u s t h e
u n a m b i g u o u s conclusions are not reasonable on the basis of that [50].
T h e r e s u l t s o f m e a s u r e m e n t s at h i g h c u r r e n t d e n s i t i e s a r e o f e s p e c i a l l y g r e a t
i n t e r e s t in t e r m s o f t h e e l u c i d a t i o n o f t h e m e c h a n i s m o f t h e h y d r o g e n e v o l u t i o n
reaction.
T h e e q u a t i o n ( 8 ) s h o w s t h a t t h e r a t e o f r e c o m b i n a t i o n o f h y d r o g e n a t o m s is
g o v e r n e d o n l y by t h e d e g r e e o f c o v e r a g e w i t h a d s o r b e d a t o m i c h y d r o g e n . In t h e
case of the retarded step of removal of adsorbed hydrogen Θ increases with Η

overpotential. If θ a t t a i n s t h e v a l u e o f 1, t h e s a t u r a t i o n c u r r e n t i h a s t o b e
§

observed and further increasing of the applied potential does not induce the
increase of the rate of h y d r o g e n evolution. Therefore the investigation of
h y d r o g e n e v o l u t i o n at s u f f i c i e n t l y h i g h c u r r e n t d e n s i t i e s will e n a b l e t o r e a l i s e if
t h e r e c o m b i n a t i o n s t e p m i g h t b e o p e r a t i v e in t h e o v e r a l l p r o c e s s .
A c c o r d i n g t o t h e t h e o r e t i c a l p r i n c i p l e s o f r e c o m b i n a t i o n r e a c t i o n w o r k e d o u t by
H o r i u t i [ 7 8 ] t h e s a t u r a t i o n c u r r e n t i for n i c k e l h a s t o b e e x h i b i t e d at 2 x l 0 A / c m
s
2 2

at 2 5 ° C . S e v e r a l a t t e m p t s h a v e b e e n m a d e for t h e d e t e r m i n a t i o n o f s a t u r a t i o n
current. In [ 7 9 ] t h e t e n d e n c y o f c u r r e n t t o a p p r o a c h a c o n s t a n t v a l u e n e a r
2
/ = 1 0 A / c m o n n i c k e l in 5 Μ HC1 s o l u t i o n h a v e b e e n o b s e r v e d , a l t h o u g h t h e
r e p r o d u c i b i l i t y o f t h a t h a s n o t b e e n t o o g o o d . In [ 8 0 , 8 1 ] t h e r a p i d i n c r e a s e o f η
234

2
with the increase of / beyond 1 A / c m with the saturation current around
2
100 A / c m h a s b e e n o b s e r v e d
It m u s t b e e m p h a s i s e d t h a t t h e r e a l i s a t i o n o f p o l a r i s a t i o n m e a s u r e m e n t s at s u c h
high c u r r e n t d e n s i t i e s is c o n n e c t e d w i t h g r e a t d i f f i c u l t i e s The main factors
c o m p l i c a t i n g t h e m e a s u r e m e n t s a r e t h e f o l l o w i n g : t h e shift o f t h e p o t e n t i a l d u e t o
the ohmic d r o p of potential b e t w e e n the e l e c t r o d e surface and the tip of Luggin's
capillary Δ Ε Γ t h e c o n c e n t r a t i o n o v e r p o t e n t i a l Δ Ε , t h e shift o f p o t e n t i a l Δ Ε
2

c a u s e d by t h e h e a t i n g u p o f t h e e l e c t r o d e a n d t h e s o l u t i o n in t h e v i c i n i t y o f t h e
e l e c t r o d e T h u s , in a d d i t i o n t o t h e d e s i r a b l e real v a l u e o f h y d r o g e n o v e r p o t e n t i a l
η , t h r e e m o r e t e r m s will b e i n v o l v e d in t h e v a l u e o f e x p e r i m e n t a l l y m e a s u r e d E:

Ε = η + ΔΕ + Δ Ε +ΔΕ (19)

T h e p r e d o m i n a t i n g o f t h e m is t h e o h m i c d r o p o f p o t e n t i a l Δ Ε as t h i s t e r m is
1

varying linearly with current contrary to the o v e r p o t e n t i a l of electrochemical

p r o c e s s d e p e n d i n g on the logarithm of current On the o t h e r hand the linear
d e p e n d e n c e b e t w e e n Δ Ε a n d / m a k e s it p o s s i b l e t o find t h e v a l u e o f AE at v e r y
] ]

h i g h c u r r e n t d e n s i t i e s by l i n e a r e x t r a p o l a t i o n T o diminish the influence of

concentration polarisation comparatively concentrated solutions are used, which
decreases also the ohmic drop of potential
T h e h e a t i n g o f t h e s o l u t i o n c a n b e d i m i n i s h e d t o a g r e a t e x t e n t e i t h e r by
s h o r t e n i n g o f t h e p o l a r i s a t i o n t i m e , for e x a m p l e u s i n g t h e g a l v a n o s t a t i c p u l s e
m e t h o d [ 8 0 ] o r by c a r r y i n g o u t t h e m e a s u r e m e n t s in f l o w i n g s o l u t i o n [ 7 9 8 2 , 8 3 ]
W h e n t h e t e r m s Δ Ε and Δ Ε d e v i a t e t h e p o l a r i s a t i o n c u r v e s t o w a r d s t h e h i g h e r
] 2

o v e r p o t e n t i a l s , Δ Ε ^ h a s an o p p o s i t e effect as t h e i n c r e a s e o f t e m p e r a t u r e
a c c e l e r a t e s the electrochemical reaction and increases the conductivity of the
s o l u t i o n Fig 8 p r e s e n t s t h e p o l a r i s a t i o n c u r v e s m e a s u r e d c u r r e n t d e n s i t i e s u p t o
2
100 A / c m [ 8 3 ] s h o w i n g t h e effect o f v a r i o u s m e t h o d s o n d i m i n i s h i n g t h e
deviation of the polarisation curve due to the ohmic d r o p of potential and heating
of the solution The flowing solution of velocity 5 m/s has been used and the
o h m i c d r o p o f p o t e n t i a l h a s b e e n d e t e r m i n e d by o s c i l l o g r a p h i c s w i t c h - i n a n d
s w i t c h - o f f m e a s u r e m e n t s for t h e c o r r e c t i o n o f t h e p o l a r i s a t i o n c u r v e s It h a s b e e n
established that the c h a n g e of o v e r p o t e n t i a l Δ Ε ] is e s s e n t i a l l y g r e a t e r if
d e t e r m i n e d by s w i t c h - o f f m e a s u r e m e n t s . A p o s s i b l e r e a s o n for it m a y be t h e
a d d i t i o n a l d r o p o f p o t e n t i a l , c a u s e d by small b u b b l e s o f h y d r o g e n a p p e a r i n g i n t o
the layer of solution b e t w e e n the e l e c t r o d e and the tip of Luggin's capillary The
r a t e o f t h e g e n e r a t i n g b u b b l e s is so high t h a t e v e n t h e s t r e a m o f t h e f l o w i n g
s o l u t i o n is n o t a b l e t o r e m o v e t h e m e n t i r e l y
C o r r e c t i o n s p r e s e n t e d a b o v e will m a k e it p o s s i b l e t o l i n e a r i s e t h e p o l a r i s a t i o n
2
c u r v e s a p p r o x i m a t e l y u p t o 10 A / c m At h i g h e r c u r r e n t d e n s i t i e s still e s s e n t i a l
d e v i a t i o n s c a n b e o b s e r v e d ( F i g 8) Remarkable fluctuations of overpotential
v a l u e s h a v e b e e n e l u c i d a t e d by o s c i l l o g r a p h i c m e a s u r e m e n t s , t h e a m p l i t u d e o f
f l u c t u a t i o n s b e i n g d e p e n d e n t on t h e r a t e o f t h e flow o f t h e s o l u t i o n If t h e
v e l o c i t y o f t h e flow w a s 5 m / s t h e f l u c t u a t i o n s o f Ε w e r e o b s e r v e d at c u r r e n t
 235

- 1 0 1 2

log i (A/cm ) 2

F i g . 8. P o l a r i s a t i o n c u r v e s for m e c h a n i c a l l y p o l i s h e d n i c k e l at h i g h c u r r e n t
d e n s i t i e s in 0.5 Μ K O H s o l u t i o n :
1 - u s u a l l y m e a s u r e d , 2 - 4 u s i n g flowing s o l u t i o n , 3,4 - c o r r e c t e d by t h e
o h m i c d r o p o f p o t e n t i a l , u s i n g s w i t c h - o f f c u r v e s , 4 - and a d d i t i o n a l l y u s i n g
2
linear e x t r a p o l a t i o n o f t h e o h m i c d r o p o f p o t e n t i a l at i > 10 A / c m .

2
d e n s i t i e s a b o v e 10 A / c m . At s u c h h i g h c u r r e n t d e n s i t i e s t h e f o r m a t i o n o f
m o l e c u l a r h y d r o g e n is s o r a p i d , t h a t a g a s e o u s film o f h y d r o g e n a p p e a r s o n t h e
s u r f a c e o f t h e e l e c t r o d e w h i c h c a n n o t b e r e m o v e d e v e n by t h e s t r e a m o f s o l u t i o n .
A s a r e s u l t an a d d i t i o n a l o h m i c r e s i s t a n c e b e c o m e s a p p a r e n t . T h e p o s s i b l e
e x p l a n a t i o n o f t h e f l u c t u a t i o n s o f o v e r p o t e n t i a l o b s e r v e d in [ 8 0 ] at h i g h c u r r e n t
d e n s i t i e s m a y b e t h e f o r m a t i o n o f a s u c h t h i n g a s e o u s film o n t h e e l e c t r o d e
s u r f a c e . A l t h o u g h t h e m e a s u r e m e n t s in t h i s w o r k h a v e b e e n c a r r i e d o u t by q u i t e
4 3
s h o r t g a l v a n o s t a t i c i m p u l s e s ( 1 0 ~ . . 1 0 ~ s), t h i s t i m e is l o n g e n o u g h t o p r o d u c e a
2
h y d r o g e n l a y e r o f t h i c k n e s s 2 0 . . . 100 μπι at c u r r e n t d e n s i t i e s 1 0 . . 8 0 A / c m
i n d u c i n g an a d d i t i o n a l o h m i c d r o p o f p o t e n t i a l ( m o r e o v e r , f l o w i n g s o l u t i o n w a s
n o t a p p l i e d in t h i s w o r k ) .
It h a s b e e n e s t a b l i s h e d in p r e v i o u s w o r k s [ 8 3 ] t h a t t h e d e v i a t i o n o f
2
o v e r p o t e n t i a l at c u r r e n t d e n s i t i e s o v e r 10 A / c m is p r o p o r t i o n a l w i t h c u r r e n t
density T h i s r e l a t i o n h a s b e e n u s e d for t h e c o r r e c t i o n o f p o l a r i s a t i o n c u r v e s at
v e r y h i g h c u r r e n t d e n s i t i e s , as c a n b e s e e n o n F i g . 8. O n t h e b a s i s o f t h e r e v i e w
p r e s e n t e d a b o v e it m a y b e c o n c l u d e d t h a t t h e a p p e a r a n c e o f s a t u r a t e d c u r r e n t o n
n i c k e l at h i g h o v e r p o t e n t i a l s h a s n o t b e e n e x p e r i m e n t a l l y p r o v e d yet
O n e p o s s i b i l i t y t o o v e r c o m e t h e c o m p l i c a t i o n s d i s c u s s e d a b o v e is t o d i m i n i s h
the surface area of e l e c t r o d e . N o w a d a y s different micro and ultra micro e l e c t r o d e s
236

4 2
a r e w i d e l y u s e d [ 8 4 ] , h a v i n g s u r f a c e a r e a s o f a b o u t 10" c m . I n t h i s c a s e t h e
2
e x p e r i m e n t a l r e a l i s a t i o n o f c u r r e n t d e n s i t i e s o f a b o u t 100 A / c m is n o t a p r o b l e m
H o w e v e r , t h e a p p l i c a t i o n o f v e r y small e l e c t r o d e s is c o n n e c t e d w i t h v e r y s t r i c t
r e q u i r e m e n t for t h e p u r i t y o f t h e s o l u t i o n , a s t h e r e l a t i o n b e t w e e n t h e v o l u m e o f
s o l u t i o n ( i . e . t h e e x t e n t o f i m p u r i t i e s ) a n d t h e s u r f a c e a r e a o f e l e c t r o d e is l a r g e r in
this case A n o t h e r c o m p l i c a t i n g f a c t o r o f m e a s u r e m e n t s at v e r y h i g h c u r r e n t
d e n s i t i e s is t h e i n g r e s s o f h y d r o g e n , s t i m u l a t e d by t h e i m p u r i t i e s a d s o r b e d o n t h e
electrode

4.4. Influence of the c o m p o s i t i o n of solution.

(I). General principles
H y d r o g e n e v o l u t i o n r e a c t i o n is a t y p i c a l h e t e r o g e n e o u s r e a c t i o n p r o c e e d i n g o n
the metal/solution interface the r a t e o f w h i c h is d e p e n d e n t o n t h e s u r f a c e
c o n c e n t r a t i o n o f r e a c t a n t s a n d t h e e x i s t e n c e o f free s i t e s o n t h e s u r f a c e
The adsorption of reactants, including the intermediate of the reaction atomic
h y d r o g e n , is o f g r e a t i m p o r t a n c e in t h e d e t e r m i n a t i o n o f t h e k i n e t i c r e l a t i o n s h i p s
of the reaction In a d d i t i o n t o t h e r e a c t a n t s t h e a d s o r p t i o n o f o t h e r p a r t i c l e s ,
e s p e c i a l l y a n i o n s , m a y h a v e a s i g n i f i c a n t effect o n t h e p r o c e s s The anions can be
divided into t w o groups:
1) i n d i f f e r e n t a n i o n s h a v i n g only e l e c t r o s t a t i c i n t e r a c t i o n w i t h t h e s u r f a c e
2) s p e c i f i c a l l y a d s o r b e d a n i o n s w i t h t h e i r a d s o r p t i o n b e i n g d i f f e r e n t f r o m t h e o n e
c a l c u l a t e d o n t h e b a s i s o f e l e c t r o s t a t i c i n t e r a c t i o n o n l y [ 2 3 ] a n d is m a i n l y o f
chemical nature
S u c h a d i v i s i o n is, o f c o u r s e , s o m e w h a t c o n v e n t i o n a l , a n d is d e p e n d e n t o n t h e
s e n s i t i v i t y o f t h e m e t h o d s u s e d for t h e d e t e r m i n a t i o n o f t h e s p e c i f i c c h a r a c t e r o f
adsorption F o r t h e m e t a l s b e h a v i n g in a c e r t a i n r a n g e o f p o t e n t i a l s a s "ideal
p o l a r i s e d e l e c t r o d e s " ( i . e . t h e c h a r g e is e x p e n d e d o n l y o n t h e c h a r g i n g o f t h e
d o u b l e l a y e r a n d n o e l e c t r o c h e m i c a l r e a c t i o n is o c c u r r i n g o n t h e s u r f a c e ) t h e
specific a d s o r p t i o n can be d e t e r m i n e d with satisfactory precision on the basis of
the p o t e n t i a l - d e p e n d e n c e of the adsorption of ions [45]. Nickel d o e s not behave
as an i d e a l c a p a c i t o r in at any r a n g e o f p o t e n t i a l s , t h e r e f o r e , s u c h k i n d o f d a t a a r e
not available For that reason the establishment of the specific c h a r a c t e r of the
a d s o r p t i o n o f a n i o n s o n n i c k e l is far from b e i n g an e a s y p r o b l e m . It is r e a s o n a b l e
2
to a s s u m e that o x o a n i o n s , such as SO , C I O / , etc., are a d s o r b e d on nickel
surface mainly non-specifically, but the a d s o r p t i o n of halide ions especially of Γ-
a n d B r " - i o n s is o f s p e c i f i c c h a r a c t e r [ 8 5 - 9 3 ] .
T h e s u r f a c e o f n i c k e l in a c i d i c s o l u t i o n s is p r i m a r i l y c o v e r e d by a d s o r b e d w a t e r
m o l e c u l e s a n d t o a s m a l l e r e x t e n t by h y d r o g e n i o n s a n d s p e c i f i c a l l y a d s o r b e d
anions The total potential difference b e t w e e n the metal and the solution Ε can be
r e g a r d e d as being formed of t w o parts: the difference b e t w e e n the bulk of the
solution and the plane of reacting particles ψ and the difference from that plane
}

to the metal surface Ε-ψ Γ

T h e w o r k r e q u i r e d t o t r a n s f e r an ion f r o m t h e b u l k o f s o l u t i o n t o t h e m e t a l
s u r f a c e c a n b e d e s c r i b e d as c o n s i s t i n g o f t w o c o n t r i b u t i o n s , o n e d e t e r m i n e d by t h e
p o t e n t i a l o f e l e c t r o d e , t h e o t h e r by t h e c h e m i c a l i n t e r a c t i o n , s i m i l a r t o t h a t o f
electrochemical potential Taking into consideration that the surface
 237

concentration of hydrogen ions is c o m p a r a t i v e l y small [94,95], the simplified

relation can be used:

c S
H o-= „ «
5
c
3
e x
P[(-8„ j 0
+ F
V,VRT] , (20)

where g is t h e s t a n d a r d f r e e e n e r g y o f t h e s p e c i f i c a d s o r p t i o n o f
H Q + hydrogen
3
ions, c * is t h e s u r f a c e c o n c e n t r a t i o n o f h y d r o g e n i o n s a n d c
H 0 + H Q + is t h e
3
c o n c e n t r a t i o n o f h y d r o g e n i o n s in t h e b u l k o f s o l u t i o n
In g e n e r a l p r o t o n s a r e t o b e r e d u c e d f r o m a p r o t o n - d o n o r H B :

HB + e"= Η + Β" , (Χ)

ads ads v 7

and c o n s e q u e n t l y

C S
H B = C
H B e X
P [ - g H B + Z F
V , ) / R T
] · 2 1
< >

w h e r e ζ is t h e c h a r g e o f t h e p r o t o n - d o n o r
It c a n b e s e e n f r o m t h e e x p r e s s i o n ( 2 1 ) t h a t t h e s u r f a c e c o n c e n t r a t i o n o f an
u n c h a r g e d p a r t i c l e b e i n g a p r o t o n - d o n o r is i n d e p e n d e n t o f t h e ψ - p o t e n t i a l .
If t h e L a n g m u i r i s o t h e r m is v a l i d t o d e s c r i b e t h e a d s o r p t i o n o f a t o m i c h y d r o g e n ,
t h e o v e r p o t e n t i a l g o v e r n e d by t h e s l o w d i s c h a r g e o f p r o t o n s c a n b e g i v e n in
g e n e r a l by t h e e x p r e s s i o n [ 9 6 ] :

( x F F a F F F ^ a y F l - 9 B a F
(22)

T h e e x p r e s s i o n m a y b e s i m p l i f i e d for f u r t h e r a n a l y s i s . A s s u m i n g t h e a d s o r p t i o n o f
r e a c t a n t s ( H B , Β , and H ) d o e s not d e p e n d on the c o n c e n t r a t i o n of solution and
p o t e n t i a l a n d 0 = c o n s t t h e e q u a t i o n ( 2 2 ) for a c i d i c s o l u t i o n s c a n b e p r e s e n t e d as
H

follows:

1-cc 1-ccRT. RT, .

η = ΰοη8ί + ψι l n c n . + — lni , n (23)
α α F 3 aF
a n d for a l k a l i n e s o l u t i o n s

RT RT
η = ΰοη5ί-ψ!- — lnc O H + — lni . (24)

It c a n b e s e e n t h a t t h e t w o t e r m s o f t h e e q u a t i o n ( 2 3 a n d 2 4 ) d e p e n d o n the
composition of solution. Unfortunately, precise data characterising the
potential are not available It is c o m m o n l y a s s u m e d t h a t ψ - p o t e n t i a l d o e s not
differ m u c h f r o m t h e \ | / - p o t e n t i a l ( t h e p o t e n t i a l o f t h e o u t e r p l a n e o f H e l m h o l z
0

layer) which may be determined from the double layer m e a s u r e m e n t s .

238

i|/ -potential can be expressed as follows [ 9 7 ] :

()

2RT ( σ ^
ψ = arcsh —, (25)
F UveRT/^cJ

w h e r e σ is t h e s u r f a c e c h a r g e a n d D is d i e l e c t r i c p e r m e a b i l i t y ,
At l o w s u r f a c e c h a r g e , i.e. n e a r t o t h e p o t e n t i a l o f z e r o c h a r g e , t h e ψ -potential
is g i v e n by

Ψο - r rw^r= ( 2 6
)

F o r s u f f i c i e n t l y h i g h s u r f a c e c h a r g e , i.e. far f r o m t h e p o t e n t i a l o f z e r o c h a r g e

2RT RT
\|/ =const +
() Ιησ lnc , if σ > 0 , (27)
F F
and

2RT RT
ψ = const
() 1ησ + — l n c , if σ < 0 . (28)
F F

Consequently, providing Ψ ] ^ Ψ 0 , ^ - p o t e n t i a l is p r o p o r t i o n a l t o l/-Jc near to

the potential of zero-charge and the Tj-potential exhibits a logarithmic
dependence on the c o n c e n t r a t i o n o f e l e c t r o l y t e far f r o m t h e p o t e n t i a l o f z e r o -
charge
A c c o r d i n g t o t h e r e l a t i o n s h i p s p r e s e n t e d a b o v e an a n a l y s i s c a n b e m a d e a b o u t
the influence of the composition of the solution on the hydrogen overpotential
under the conditions of absence of specific adsorption. The concentration-
/
d e p e n d e n c e o f 4 - p o t e n t i a l for t h r e e l i m i t i n g c o n d i t i o n s will b e c o n s i d e r e d :
1

1) the surface charge is approximately 0 (or, in general, Ψ^οοηβί if the

c o n c e n t r a t i o n o f t h e s o l u t i o n is h i g h e n o u g h ) ,
2) the range of high negative surface charge where the expression (28) is
operative,
3) t h e r a n g e of high positive surface c h a r g e , w h e r e the e x p r e s s i o n ( 2 7 ) can be
used
T h e m e a s u r e m e n t s o f h y d r o g e n o v e r p o t e n t i a l c a n b e c a r r i e d o u t e i t h e r in p u r e
a c i d o r a l k a l i s o l u t i o n o r in t h e s o l u t i o n s o f a c i d o r alkali at t h e p r e s e n c e o f t h e
e x c e s s o f salt. In t h e c a s e o f p u r e a l k a l i o r a c i d s o l u t i o n t h e c o n c e n t r a t i o n o f t h e
reactant and ^-potential are both determined by the concentration of the
electrolyte. In t h e s e c o n d c a s e T j - p o t e n t i a l is m a i n l y d e t e r m i n e d b y t h e e x c e s s o f
salt, c o n s e q u e n t l y V
F ~ c o n s t and only the c o n c e n t r a t i o n of the reactant
J depends
on t h e c o n c e n t r a t i o n of acid or alkali. Possible values of (dr|/dlnc)i= onst C for all
o c c a s i o n s d i s c u s s e d in c a s e o f α = 0 . 5 a r e p r e s e n t e d in T a b l e 3 .
 239

Table 3
V a l u e s o f ( d r j / d l n c ) i = c o n s t for v a r i o u s c o n d i t i o n s
solution σ=0 σ<0 σ>0
acid -RT/F 0 -2RT/F
a c i d + salt -RT/F -RT/F -RT/F
alkali -RT/F -2RT/F 0
a l k a l i + salt -RT/F -RT/F -RT/F

It c a n b e s e e n t h a t t h e o v e r p o t e n t i a l m a y v a r y w i t h t e n f o l d d i l u t i o n o f t h e
s o l u t i o n f r o m 0 t o 118 m V ( a t t = 2 5 ° C a n d oc=0.5) d e p e n d i n g o n t h e s u r f a c e
charge and the c o m p o s i t i o n of the solution. T h e t r e a t m e n t p r e s e n t e d a b o v e is
consistent with the mechanism of slow discharge (reaction I). The same relations
a r e v a l i d if t h e r a t e - d e t e r m i n i n g s t e p is t h e e l e c t r o c h e m i c a l d e s o r p t i o n ( r e a c t i o n
II). If t h e r e c o m b i n a t i o n o f h y d r o g e n a t o m s ( r e a c t i o n I I I ) is r a t e - d e t e r m i n i n g o f
the process, the composition of the solution has no direct influence on the rate of
the hydrogen evolution.
C o n s e q u e n t l y , the investigation of the d e p e n d e n c e of h y d r o g e n overpotential on
t h e c o m p o s i t i o n o f s o l u t i o n ( p r o v i d e d t h e s p e c i f i c a d s o r p t i o n o f a n i o n s is n o t -
p r o n o u n c e d ) enables to distinguish only the non-electrochemical recombination
r e a c t i o n a n d an e l e c t r o c h e m i c a l s t e p as r a t e - d e t e r m i n i n g o f t h e o v e r a l l r e a c t i o n ,
b u t n o t t o d i s t i n g u i s h p r o t o n d i s c h a r g e a n d e l e c t r o c h e m i c a l d e s o r p t i o n as t h e r a t e -
determining step.

(II) Experimental results.

The influence of the c o n c e n t r a t i o n of solution on h y d r o g e n overpotential on
n i c k e l h a s b e e n s t u d i e d in s e v e r a l w o r k s in a c i d i c as w e l l a s in a l k a l i n e s o l u t i o n s
[ 6 3 - 6 6 , 7 0 , 9 8 - 1 0 3 ] . It m u s t b e n o t e d , h o w e v e r , t h a t t h e r e s u l t s d o n o t c o i n c i d e
well. T h e r e m a r k a b l e v a r i a t i o n s of the results may be caused by s o m e u n n o t i c e d
variation of the conditions of the experiment when the c o n c e n t r a t i o n d e p e n d e n c e
o f o v e r p o t e n t i a l is s t u d i e d . O n e p a r a m e t e r w h i c h e a s i l y m a y c h a n g e d u r i n g t h e
m e a s u r e m e n t s is t h e s u r f a c e s t a t e o f t h e e l e c t r o d e . In o r d e r t o c o m p a r e t h e r e s u l t s
o b t a i n e d in d i f f e r e n t w o r k s o n e m u s t m a k e s u r e t h e c o n d i t i o n s o f c a r r y i n g o u t o f
the measurements are the same.
T h e investigation of the d e p e n d e n c e of overpotential on the c o n c e n t r a t i o n and
c o m p o s i t i o n o f t h e s o l u t i o n m a y b e c a r r i e d o u t d i f f e r e n t l y : o n o n e e l e c t r o d e in
s e v e r a l s o l u t i o n s o r o n d i f f e r e n t e l e c t r o d e s in d i f f e r e n t s o l u t i o n s . In t h e l a t t e r
c a s e it is n o t e x c l u d e d t h a t t h e s u r f a c e s t a t e m a y b e r e m a r k a b l y d i f f e r e n t in
different solutions, disturbing the d e p e n d e n c e of h y d r o g e n o v e r p o t e n t i a l on the
c o m p o s i t i o n of solution. T h e s e e r r o r s c a n b e d i m i n i s h e d by m e a s u r i n g
o v e r p o t e n t i a l o n t h e s a m e e l e c t r o d e in s e v e r a l s o l u t i o n s , w h i l e t h e c a t h o d i c
p o l a r i s a t i o n o f t h e e l e c t r o d e is n o t i n t e r r u p t e d a n d t h e i n f l o w o f air i n t o t h e cell is
excluded during the experiment [70,100,102]. The second technology described
g i v e s d o u b t l e s s m o r e r e l i a b l e r e s u l t s if t h e p o l a r i s a t i o n in t h e s o l u t i o n s o f d i f f e r e n t
c o n c e n t r a t i o n i n t r o d u c e s the variation of the surface state.
240

M o s t o f t h e m e a s u r e m e n t s in a c i d i c m e d i u m h a v e b e e n c a r r i e d out in the
s o l u t i o n s o f H S 0 a n d HC1 w i t h p H in 0 .3.5
2 4

It has been established that for mechanically polished electrodes

(dT)/dlog c ) i = c o n s t h a s a v a l u e in t h e r a n g e o f 5 8 . 1 0 m V , w h e r e a s t h e m o r e h i g h
v a l u e s h a v e b e e n o b t a i n e d for m o r e c o n c e n t r a t e d s o l u t i o n s ( F i g . 9 ) S u c h
d e p e n d e n c e h a s b e e n o b t a i n e d for H S 0 2 solutions measured on
4 different
e l e c t r o d e s as w e l l as o n t h e s a m e e l e c t r o d e , a n d for HC1 s o l u t i o n s if m e a s u r e d o n
the same electrode. F o r t h e n i c k e l e l e c t r o d e s h e a t e d in h y d r o g e n a n d t h e

chemically polished ones the value of ( d r | / dlog c ) ^ ^ ^ m

H S>^4
2 solutions is
c o n s i d e r a b l y s m a l l e r a n d in t h e l a t t e r c a s e d o e s n o t e x c e e d 10 m V .
In [ 6 3 - 6 6 ] t h e d e p e n d e n c e similar t o t h a t o n m e c h a n i c a l l y p o l i s h e d e l e c t r o d e s in
H S O , s o l u t i o n s h a s b e e n e x h i b i t e d a l s o in s o l u t i o n s o f HC1 for t h e e l e c t r o d e s
2 4
h e a t e d in h y d r o g e n ( m e a s u r e d o n d i f f e r e n t e l e c t r o d e s in d i f f e r e n t s o l u t i o n s ) . T h e
m o r e e s s e n t i a l c o n c e n t r a t i o n - d e p e n d e n c e o f o v e r p o t e n t i a l o b t a i n e d in [ 6 3 , 6 4 ] m a y
b e c a u s e d by t h e v a r i a t i o n o f t h e s u r f a c e s t a t e o f t h e e l e c t r o d e w i t h t h e
c o n c e n t r a t i o n o f t h e HC1 s o l u t i o n as p o i n t e d o u t in [ 1 0 4 , 1 0 5 ] .
The value of (dr|/dlog c ) i = c o n s tin t h e s o l u t i o n s o f H S 0 2 4 for s i n g l e c r y s t a l
e l e c t r o d e s h a s b e e n f o u n d t o b e d e p e n d e n t o n t h e s t r u c t u r e o f t h e f a c e [ 7 0 ] . In t h e
c a s e o f t h e f a c e ( 1 0 0 ) o v e r p o t e n t i a l is a l m o s t i n d e p e n d e n t o f t h e c o n c e n t r a t i o n o f
s o l u t i o n w h i l e η o n t h e f a c e ( 1 1 0 ) a n d o n face ( 1 1 1 ) i n c r e a s e s by 1 5 . . . 2 5 m V w i t h
t h e t e n f o l d d i l u t i o n o f acid s o l u t i o n w i t h o u t any c h a n g e in t h e s h a p e o f η , l o g /'-
curve [70],
O n l y few d a t a a r e a v a i l a b l e a b o u t t h e i n f l u e n c e o f t h e c o n c e n t r a t i o n o f
h y d r o g e n i o n s o n h y d r o g e n o v e r p o t e n t i a l in t h e s o l u t i o n s o f c o n s t a n t t o t a l
c o n c e n t r a t i o n o f e l e c t r o l y t e (in t h e s o l u t i o n s o f acid w i t h t h e e x c e s s o f s a l t )
[ 6 3 , 6 4 , 7 0 , 1 0 1 , 1 0 3 ] . T h e Tafel s l o p e b for t h e e l e c t r o d e s h e a t e d in h y d r o g e n is
f o u n d t o b e g r e a t e r if t h e e x c e s s o f salt is u s e d and t h e c o r r e s p o n d i n g p o l a r i s a t i o n
c u r v e i n t e r s e c t s w i t h t h e o n e m e a s u r e d in p u r e acid s o l u t i o n [ 6 3 ] . T h e s i m i l a r
effect h a s b e e n o b t a i n e d a l s o o n s i n g l e c r y s t a l e l e c t r o d e s [ 7 0 ] . The results
p r e s e n t e d a b o v e i m p l y t h a t t h e m e a s u r e m e n t s in a c i d i c s o l u t i o n s a r e c a r r i e d o u t in
the range of potentials near to the potential of zero charge [ 6 3 ,7 0 ] .
T h e p o l a r i s a t i o n m e a s u r e m e n t s in a l k a l i n e m e d i u m e x h i b i t slight d e p e n d e n c e o f
o v e r p o t e n t i a l o n t h e c o n c e n t r a t i o n o f alkali [ 6 3 - 6 6 , 9 8 ] o r n o d e p e n d e n c e at all
[ 7 3 , 9 9 ] ( F i g 10) The elucidation of c o n c e n t r a t i o n - d e p e n d e n c e of overpotential
in t h e i n t e r m e d i a t e r e g i o n o f p H is o f g r e a t i n t e r e s t , h o w e v e r , t h e m e a s u r e m e n t s in
c o m p a r a t i v e l y neutral solutions are c o n n e c t e d with serious difficulties.
A s h y d r o g e n i o n s a r e u s e d in t h e p r o c e s s o f c a t h o d i c h y d r o g e n e v o l u t i o n t h e
c o n c e n t r a t i o n p o l a r i s a t i o n m a y b e c o m e a p p a r e n t in d i l u t e d acid s o l u t i o n s shifting
t h e v a l u e o f p H o f t h e s o l u t i o n in t h e v i c i n i t y o f t h e e l e c t r o d e s u r f a c e ( p H ) . T h e s

i n c r e a s e o f p H is e s p e c i a l l y o b v i o u s in t h e c a s e o f d i l u t e d acid s o l u t i o n s w i t h t h e
s

e x c e s s o f salt [ 1 0 3 ] e s s e n t i a l l y d i s t u r b i n g t h e s h a p e o f p o l a r i s a t i o n c u r v e s
( F i g . 10).
 241

Fig 9. D e p e n d e n c e o f h y d r o g e n o v e r p o t e n t i a l on t h e c o n c e n t r a t i o n o f s o l u t i o n on
m e c h a n i c a l l y polished nickel e l e c t r o d e s :
1 - Η S O , solution, 2 - K O H solution.
2 4

4 3 2
- log i (A/cm ) 2

F i g . 10. P o l a r i s a t i o n c u r v e for m e c h a n i c a l l y polished n i c k e l in t h e s o l u t i o n of

0 . 0 0 5 Μ Η S O , + 0.5 Μ Κ S O .
2 4 2 4
242

T h e v a l u e o f p o t e n t i a l Δ Ε m e a s u r e d in t h e c o n d i t i o n s of concentration
polarisation can be r e g a r d e d as consisting of several t e r m s :

ΔΕ = η + Δ Ε ρ Η +ΔΕικ , (29)

w h e r e η is h y d r o g e n o v e r p o t e n t i a l , A E - t h e shift o f t h e p o t e n t i a l d u e t o t h e
p H

d i f f e r e n c e in p H b e t w e e n t h e s o l u t i o n s in t h e r e f e r e n c e e l e c t r o d e c o m p a r t m e n t a n d
in t h e c l o s e v i c i n i t y o f t h e s u r f a c e o f t h e e l e c t r o d e s t u d i e d . A E is t h e o h m i c
J R

d r o p of potential b e t w e e n the surface of the e l e c t r o d e and the tip of L u g g i n ' s

capillary
At sufficiently high c o n c e n t r a t i o n o f salt t h e d i f f e r e n c e of hydrogen
o v e r p o t e n t i a l f r o m Δ Ε is d e t e r m i n e d m a i n l y by Δ Ε . A p p l y i n g a p H p r o b e it is
ρ Η

p o s s i b l e t o e l u c i d a t e t h e v a l u e o f a p H at d i f f e r e n t c u r r e n t d e n s i t i e s . H o w e v e r ,
s

t h e a p p l i c a t i o n o f p H p r o b e is c o n n e c t e d w i t h s o m e c o m p l i c a t i o n s . T h e p r o b e m u s t
b e sufficiently s m a l l , o t h e r w i s e t h e v a l u e o f p H in a c e r t a i n d i s t a n c e o f t h e
e l e c t r o d e is m e a s u r e d r a t h e r t h a n t h e d e s i r a b l e v a l u e o f p H in t h e c l o s e v i c i n i t y o f
s

t h e e l e c t r o d e . B e s i d e s , it m u s t b e t a k e n i n t o c o n s i d e r a t i o n t h a t t h e c o m p a r a t i v e l y
t h i c k tip o f p H p r o b e will c a u s e s o m e s c r e e n i n g effect if t r a n s f e r r e d v e r y c l o s e t o
t h e e l e c t r o d e . A s a r e s u l t e s s e n t i a l d e v i a t i o n s f r o m t h e r e a l v a l u e o f p H will b e
s

obtained.
In d i l u t e d acid s o l u t i o n s w i t h t h e e x c e s s o f salt p H m a y a t t a i n e v e n t h e v a l u e
s

o f 1 0 - 1 1 , i.e. t h e H E R is p r o c e e d i n g r e a l l y in a l k a l i n e s o l u t i o n a n d t h e
corresponding term A E p H a c h i e v e s t h e v a l u e o f 4 0 0 . . . 4 5 0 m V . It c a n b e c o n c l u d e d
that the precise correction of polarisation curves on account of the change of p H s

is u n a c h i e v a b l e
The influence of c o n c e n t r a t i o n polarisation can be essentially diminished
c a r r y i n g o u t t h e m e a s u r e m e n t s after s w i t c h i n g o n t h e p o l a r i s i n g c u r r e n t in a t i m e
s h o r t e n o u g h for t h e p H in t h e c l o s e v i c i n i t y o f t h e e l e c t r o d e n o t t o b e c h a n g e d
r e m a r k a b l y y e t . T h e p o l a r i s a t i o n c u r v e built u p o n t h e b a s i s o f s w i t c h - o v e r c u r v e s
( f r o m a sufficiently l o w c a t h o d i c c u r r e n t d e n s i t y t o d i f f e r e n t h i g h e r c u r r e n t
d e n s i t i e s ) is p l o t t e d in F i g . 11 [ 1 0 2 , 1 0 3 ] . It c a n b e s e e n t h a t t h e p o l a r i s a t i o n
c u r v e in t h e r e g i o n o f h i g h e r c u r r e n t d e n s i t i e s is c o n t i n u e d w i t h a p p r o x i m a t e l y t h e
s a m e v a l u e o f Tafel s l o p e e x h i b i t e d in t h e r e g i o n o f l o w e r c u r r e n t d e n s i t i e s .
In p r i n c i p l e , t h e h y d r o g e n o v e r p o t e n t i a l m e a s u r e m e n t s a r e p o s s i b l e t o c a r r y o u t
a l s o in buffer s o l u t i o n s , b u t t h e r e s u l t s o b t a i n e d in t h e s e s o l u t i o n s c a n n o t b e e a s i l y
i n t e r p r e t e d A s a r u l e , t h e m e a s u r e m e n t s in buffer s o l u t i o n s i n t r o d u c e a d d i t i o n a l
i o n s a n d m o l e c u l e s in t h e s o l u t i o n w h i c h m a y a d s o r b o n t h e s u r f a c e o f t h e
e l e c t r o d e a n d t h u s t o i n f l u e n c e t h e r a t e o f h y d r o g e n e v o l u t i o n . B e s i d e s , it m u s t b e
t a k e n into c o n s i d e r a t i o n that not only w a t e r molecules or h y d r o x o n i u m ions, but
a l s o t h e m o l e c u l e s o f w e a k acid o r o t h e r h y d r o g e n a t e d p a r t i c l e s u s e d in b u f f e r
s o l u t i o n m a y b e h a v e as p r o t o n - d o n o r s .
 243

4 3
2
- log i ( A / c m )

F i g . 1 1 . P o l a r i s a t i o n c u r v e s for m e c h a n i c a l l y p o l i s h e d n i c k e l in 0 0 0 5 Μ H S 0
2 4 +
0.5 Μ Κ S O ,
2 4
1 - usually measured, 2- polarisation curve corrected on the basis of switch­
over curves.

I_J I I I
-0.2 -0.4 -0.6
P o t e n t i a l , V (n. h. e.)

Fig 12. C vs Ε c u r v e s for nickel single crystal face ( 1 0 0 ) in H C 1 0 s o l u t i o n s 4 of

different c o n c e n t r a t i o n :
1 - 0.25 M, 2 - 0.04 M, 3 - 0.014 M, 4 - 0.0065 M, 5 - 0.003 M.
244

Obviously the most reliable results of the influence of the concentration of

s o l u t i o n o n h y d r o g e n o v e r p o t e n t i a l h a v e b e e n o b t a i n e d in p u r e a c i d o r alkali
solutions.

4 . 5 . P o t e n t i a l of z e r o c h a r g e of n i c k e l .
The potential of zero charge Ε σ = 0 is o n e o f t h e m o s t i m p o r t a n t c h a r a c t e r i s t i c s o f
t h e m e t a l - s o l u t i o n i n t e r f a c e [ 1 0 6 ] . A d s o r p t i o n ability o f a l a r g e s c a l e o f
c o m p o u n d s i n c l u d i n g s u c h v e r y i m p o r t a n t o n e s as c o r r o s i o n i n h i b i t o r s is s t r o n g l y
dependent on the value of E n = 0 of the metal T h i s p a r a m e t e r is o f g r e a t
i m p o r t a n c e for t h e a n a l y s i s o f t h e i n f l u e n c e o f d i f f e r e n t f a c t o r s , e s p e c i a l l y t h e
c o m p o s i t i o n o f s o l u t i o n , on t h e k i n e t i c s o f t h e h y d r o g e n e v o l u t i o n r e a c t i o n .
D i f f e r e n t m e t h o d s h a v e b e e n u s e d for t h e d e t e r m i n a t i o n o f E „ . n The detail
a n a l y s i s o f t h e s u b j e c t h a s b e e n c a r r i e d o u t in [ 1 0 6 ] . T h e b e s t e x p e r i m e n t a l r e s u l t s
h a v e b e e n a c h i e v e d o n liquid m e t a l s a n d o n f a c e s o f s i n g l e c r y s t a l e l e c t r o d e s . T h e
precise establishment of E a = 0 by e l e c t r o c h e m i c a l m e t h o d s is a v a i l a b l e if t h e m e t a l
c a n b e r e g a r d e d a s an i d e a l p o l a r i s e d e l e c t r o d e in t h e c o r r e s p o n d i n g r e g i o n o f
potentials T h i s c o n d i t i o n is h a r d l y s a t i s f i e d for n i c k e l as t h e r e g i o n o f p o t e n t i a l s
w i t h o u t any F a r a d a i c p r o c e s s e s c a n n o t b e f o u n d o n t h i s m e t a l . Nevertheless
s e v e r a l a t t e m p t s h a v e b e e n m a d e t o d e t e r m i n e t h e p o t e n t i a l o f z e r o c h a r g e for
nickel [ 1 0 7 - 1 1 2 ] .
T h e f o l l o w i n g m e t h o d s h a v e b e e n u s e d for the determination of Ε σ = 0 of the
nickel/electrolyte interface: electrochemical impedance, galvanostatic pulses,
adsorption of organic c o m p o u n d s , scraping at c l o s e d c i r c u i t , s t a t i c f r i c t i o n
b e t w e e n the surface and the solution. Some o f t h e s e r e s u l t s a r e p r e s e n t e d in
T a b l e 4.

Table 4
Potential of zero charge of polycrystalline nickel.
pH Method Reference
0.94 -0.224 impedance [107]
1.4 -0.252 impedance [107]
2.14 -0.237 impedance [107]
2.97 -0.331 impedance [107]
3.5 -0.11 galvanostatic [111]
pulse
5.8 -0.37 galvanostatic [111]
pulse
3.0 -0.24 scraping [110]
5.6 -0.37 scraping [110]
10.3 -0.24 impedance [109]
13.0 -0.52 adsorption [109]
12.5 -0.30 friction [109]
 245

O n e c a n s e e t h a t , as a r u l e , t h e r e is e s s e n t i a l d e p e n d e n c e o f t h e v a l u e o f Ε σ = 0 on
t h e p H o f t h e s o l u t i o n . In t h e first a p p r o x i m a t i o n it is p o s s i b l e t o r e s u m e t h a t in
acidic solutions Ε σ = 0 for p o l y c r y s t a l l i n e n i c k e l is w i t h i n - 0 . 2 2 . . . - 0 , 3 7 V . T h e v a l u e
for a l k a l i n e s o l u t i o n s w i t h i n - 0 2 4 - 0 3 2 V c a n n o t b e a t t r i b u t e d t o p u r e n i c k e l
s u r f a c e as t h e s e p o t e n t i a l s ( 0 . 3 . . . 0 . 4 V in t h e r e s p e c t o f t h e p o t e n t i a l o f r e v e r s i b l e
h y d r o g e n e l e c t r o d e in t h e s a m e s o l u t i o n ) c o r r e s p o n d t o t h e r e g i o n w h e r e t h e
s u r f a c e o f n i c k e l is c o n s i d e r e d t o b e c o v e r e d w i t h N i ( O H ) 2

In t h e w o r k [112] the determination of Ε σ 0 on chemically polished nickel

single crystals has been carried out Electrochemical impedance has been used to
d e t e r m i n e t h e m i n i m u m o f c a p a c i t a n c e in 0 . 2 5 .0 0 0 2 5 Μ H C 1 0 solutions.
4

F i g . 12 s h o w s t h e C v s . Ε c u r v e s for f a c e ( 1 0 0 ) at d i f f e r e n t electrolyte
c o n c e n t r a t i o n s . It c a n b e s e e n t h a t t h e m i n i m u m in t h e c u r v e s b e c o m e s m o r e
p r o n o u n c e d u p o n the dilution of the solution. Similar results have been obtained
for f a c e ( 1 1 0 ) The capacitance has been observed to decrease also with the
d i l u t i o n o f s o l u t i o n if m e a s u r e d on face ( 1 1 1 ) or chemically polished
polycrystalline e l e c t r o d e s but no m a r k e d minimum can be seen on C vs. Ε curves.
A c c o r d i n g t o G r a h a m e ' s m o d e l if t h e s p e c i f i c a d s o r p t i o n o f i o n s is a b s e n t t h e
m e a s u r e d c a p a c i t a n c e c a n b e d e s c r i b e d as b e i n g c o n s i s t e n t o f t w o c a p a c i t o r s
c o n n e c t e d in s e r i e s : t h e c a p a c i t a n c e o f c o m p a c t l a y e r C a n d t h e d i f f u s e - l a y e r
H

capacitance, thus:

1/C=1/C +1/C n

At c o n s t a n t s u r f a c e c h a r g e a p l o t o f 1/C a g a i n s t 1 / C ( t h e P a r s o n s - Z o b e l c o ­
D

o r d i n a t e s ) [ 1 1 3 ] s h o u l d b e d e s c r i b e d by a s t r a i g h t line o f 1 u n i t s l o p e [ 1 1 3 ] .
F i g . 13 s h o w s s u c h p l o t s o f c o r r e s p o n d i n g c a p a c i t a n c e v a l u e s for t h e m i n i m u m o f
the C vs Ε curves [112]
It h a s b e e n o b t a i n e d t h a t for f a c e ( 1 0 0 ) a n d f a c e ( 1 1 0 ) t h e s e p l o t s a r e d e s c r i b e d
by s t r a i g h t l i n e s w i t h t h e s l o p e c l o s e t o 1 u n i t . It c a n b e c l a i m e d t h e r e f o r e t h a t t h e
m i n i m u m in t h e C v s . Ε c u r v e s is c a u s e d b y t h e d i f f u s e n e s s o f t h e e l e c t r i c d o u b l e
l a y e r a n d t h a t t h e p o t e n t i a l o f t h e m i n i m u m is t h e p o t e n t i a l o f z e r o free c h a r g e
with the values of Ε = - 0 . 3 9 ± 0 . 0 2 V for f a c e ( 1 0 0 ) a n d - 0 . 5 3 ± 0 . 0 2 V for f a c e
σ=0 v 7

(110) [112]
The different b e h a v i o u r of nickel face ( 1 1 1 ) can be related to m o r e essential
ingress of h y d r o g e n into this face s o m e w h a t disturbing the s t r u c t u r e of the
surface. As the surface of polycrystalline electrode consists of a large a m o u n t of
differently oriented single crystal faces the a p p e a r a n c e of a sharp minimum can
hardly be expected and the determination of the potential of zero charge on the
b a s i s o f i m p e d a n c e m e a s u r e m e n t s is c o m p l i c a t e d .
A m i n i m u m h a s a l s o b e e n e x h i b i t e d at f a c e ( 1 0 0 ) in t h e s o l u t i o n s o f H S 0 . 2 4

The minimum w a s also found to b e c o m e deeper with the increase of the dilution of
t h e s o l u t i o n b u t in c o n t r a s t t o t h a t in H C 1 0 s o l u t i o n s t h e p o t e n t i a l o f t h e
4
246

0.02 0.04 0.06 0.08

F i g . 13 P a r s o n s - Z o b e l p l o t for face ( 1 0 0 )

m i n i m u m s o m e w h a t d e p e n d s o n t h e acid c o n c e n t r a t i o n . T h e s e c o n d s i g n i f i c a n t
d i f f e r e n c e o b s e r v e d is t h e a p p e a r a n c e o f a h u m p ( m a x i m u m ) o n t h e C v s . Ε c u r v e s
in Η S O , s o l u t i o n s w h i c h h a s n o t b e e n s e e n in H O C ) solutions. These
2 4 4 . 2

o b s e r v a t i o n s indicate the n o t e w o r t h y specific a d s o r p t i o n of anions ( H S 0 " , S 0 4 4 ")

on nickel surface. T h i s m a y b e t h e m a j o r r e a s o n for t h e d e p e n d e n c e o f Ε σ = ( ) on the
p H o f t h e s o l u t i o n d e s c r i b e d in l i t e r a t u r e for s u l p h u r i c acid s o l u t i o n s [ 1 0 8 ] , The
results of the d e t e r m i n a t i o n of E^ on nickel confirm the a s s u m p t i o n [ 6 3 , 7 0 ] that
o

t h e p o t e n t i a l o f z e r o c h a r g e is w i t h i n t h e r e g i o n o f p o t e n t i a l s o f t h e h y d r o g e n
o v e r p o t e n t i a l m e a s u r e m e n t s in a c i d i c s o l u t i o n s . C o n s e q u e n t l y t h e v ^ - p o t e n t i a l for
n i c k e l is d e t e r m i n e d by t h e a d s o r p t i o n o f c o m p o n e n t s o f s o l u t i o n r a t h e r t h a n by
the diffuseness of the electric double layer
4 . 6 . C o v e r a g e of n i c k e l s u r f a c e w i t h a d s o r b e d h y d r o g e n .
The significance of the k n o w l e d g e of the d e g r e e of c o v e r a g e of nickel surface
with a t o m i c h y d r o g e n Θ especially the d e p e n d e n c e of that on potential has been
Η

emphasised above already. In c h a p t e r 3 a s h o r t r e v i e w o f t h e e x i s t i n g

e x p e r i m e n t a l t e c h n i q u e s w o r k e d o u t for t h e d e t e r m i n a t i o n o f a d s o r b e d h y d r o g e n
h a s b e e n g i v e n . T h e r e s u l t s o b t a i n e d by t h e s e t e c h n i q u e s for n i c k e l e l e c t r o d e s will
be discussed below:
 247

(I) Acidic solutions.

A s a r e m a r k a b l e c o r r o s i o n o f n i c k e l e l e c t r o d e s p r o c e e d s in a c i d i c s o l u t i o n s t h e
o p e n - c i r c u i t p o t e n t i a l c o r r e s p o n d s t o t h e c o r r o s i o n p o t e n t i a l o f n i c k e l in t h e
solution and the reversible h y d r o g e n potential cannot be established. Therefore
C
the m e a s u r e m e n t s of Θ ° and even of 9
Η H at l o w o v e r p o t e n t i a l s a r e n o t a v a i l a b l e .
T h u s a n o d i c c h a r g i n g c u r v e s a n d p o t e n t i o d y n a m i c c y c l i n g a r e n o t a p p l i c a b l e for
t h e d e t e r m i n a t i o n o f Θ in a c i d i c s o l u t i o n s . In t h i s c a s e it is p o s s i b l e t o d e t e r m i n e
Η

o n l y a p a r t o f Θ ° by e l e c t r o c h e m i c a l m e t h o d s .
Η

T h e m o s t c o n v e n i e n t m e t h o d s in t h e s e c o n d i t i o n s a r e t h e o v e r p o t e n t i a l d e c a y
and electrochemical impedance measurements [36,47,64,101]. At high
overpotentials the influence of corrosion is not pronounced and the
e l e c t r o c h e m i c a l b e h a v i o u r o f n i c k e l is d e t e r m i n e d by t h e r e g u l a r i t i e s o f t h e H E R
o n l y . A s c a n b e s e e n f r o m F i g . 3 t h e c a p a c i t a n c e o b t a i n e d by o v e r p o t e n t i a l d e c a y
m e a s u r e m e n t s h a s a n e a r l y c o n s t a n t v a l u e in t h e r e g i o n o f h i g h o v e r p o t e n t i a l s ,
w h i c h c a n b e r e g a r d e d as t h e c a p a c i t a n c e o f d o u b l e layer. A s n o r e m a r k a b l e
p s e u d o c a p a c i t a n c e h a s b e e n o b s e r v e d in t h e r e g i o n s t u d i e d it c a n b e c o n c l u d e d
t h a t t h e c h a n g e o f Θ ° at η > 0 . 1 5 V d o e s n o t e x c e e d 0 . 0 1 .
Η

(II) Alkaline solutions

As t h e r a t e o f c o r r o s i o n o f c o m p a c t n i c k e l is v e r y l o w in a l k a l i n e s o l u t i o n s
different m e t h o d s for t h e s t u d y o f a d s o r b e d h y d r o g e n a r e a p p l i c a b l e , in p r i n c i p l e .
T h e r e s u l t s o f v a r i o u s m e a s u r e m e n t s i n d i c a t e t h a t in t h e r e g i o n o f p o s i t i v e
p o t e n t i a l s n o t far f r o m t h e p o t e n t i a l o f r e v e r s i b l e h y d r o g e n e l e c t r o d e in t h e s a m e
solution remarkable oxidation of the surface of nickel p r o c e e d s (the reversible
p o t e n t i a l o f t h e r e a c t i o n Ni + 2 O H <=> N i ( O H ) + 2 e is 0.1 V [ 5 6 ] ) .
2 The
o x i d a t i o n p r o c e s s c a n b e o b s e r v e d as a m a x i m u m o f c u r r e n t o n p o t e n t i o d y n a m i c
i,E - c u r v e s o r p s e u d o c a p a c i t a n c e o n C , E - c u r v e s in t h e r e g i o n o f 0 . 2 . . . 0 . 5 V
( F i g . 14). A h u m p o b s e r v e d in s o m e c a s e s at t h e r e g i o n o f l o w e r p o s i t i v e
potentials can be attributed to the electrochemical formation of hydroxyl radicals
and/or ionisation of adsorbed or absorbed hydrogen [ 5 1 , 1 1 4 - 1 1 7 ] .
T h e s e possible processes cannot be distinguished easily and contrary
interpretations of the experimental results have been presented.
A n a p p l i c a b l e m e t h o d for t h e d e t e r m i n a t i o n o f Θ s e e m s t o b e t h e d o u b l e
Η

c h a r g i n g m e t h o d , u s e d in [ 2 9 ] t o d e t e r m i n e t h e d e g r e e o f c o v e r a g e w i t h h y d r o g e n
of nickel surface.
In o u r w o r k [ 1 1 8 ] an a d d i t i o n a l a t t e m p t h a s b e e n c a r r i e d o u t t o u s e t h e d o u b l e
c h a r g i n g m e t h o d for t h e d e t e r m i n a t i o n o f Θ in t h e c a s e o f n i c k e l in a l k a l i n e
Η

solution T w o series of measurements have been carried out. T h e first o n e

f o l l o w e d t h e t e c h n i q u e d e s c r i b e d in [ 2 9 ] , t h e n o r m a l c u r v e h a s b e e n m e a s u r e d
starting from different steady state o v e r p o t e n t i a l s , the c o m p e n s a t i o n c u r v e has
b e e n m e a s u r e d i m m e d i a t e l y a f t e r t h e n o r m a l c u r v e s t a r t i n g f r o m a small p o s i t i v e
p o t e n t i a l ( w i t h o u t p o l a r i s a t i o n at c a t h o d i c p o t e n t i a l s ) T h e t y p i c a l c h a r g i n g c u r v e s
o f t h i s s e r i e s h a v e p r e s e n t e d o n F i g . 15 ( c u r v e s 1 a n d 2 ) a n d t h e p l o t s o f ( i - i 0 ) H a n

v s . t i m e a r e s h o w n in F i g . 16. T h e a m o u n t o f c h a r g e d e t e r m i n e d o n t h e b a s i s o f
t h e s e c u r v e s h a s b e e n f o u n d t o b e e s s e n t i a l l y d e p e n d e n t o n t h e initial
o v e r p o t e n t i a l T h e s e r e s u l t s a r e in g o o d a g r e e m e n t w i t h t h o s e in [ 2 9 ] .
248

Potential, V
F i g . 14. D e p e n d e n c e o f c a p a c i t a n c e o n p o t e n t i a l for n i c k e l in 0.5 Μ K O H s o l u t i o n
at a n o d i c p o t e n t i a l s

Time, ms
Fig 15 C h a r g i n g c u r v e s for n i c k e l in 0.5 Μ K O H s o l u t i o n :
1,2 - t h e first s e r i e s , 3,4 - t h e s e c o n d s e r i e s , 2 , 4 - n o r m a l c u r v e s , 1,3 -
compensation curves
 249

1.0 2.0
Time, ms
F i g . 16. ( i - i 0 ) v s . t i m e p l o t s o f c h a r g i n g c u r v e s :
H a n

1 - Ε = - 0 . 3 V, 2 - Ε = - 0 . 2 V, 3 - Ε = - 0 . 1 V.

T h e s e c o n d s e r i e s h a s m e a s u r e d by a c h a n g e d t e c h n i q u e The electrode has

been polarised before every charging curve (both normal and c o m p e n s a t i o n
c u r v e s ) for 10 m i n u t e s at 0 3 5 V a n d after t h a t 5 min. at t h e s a m e initial p o t e n t i a l s
as in t h e first s e r i e s All t h e c u r v e s if m e a s u r e d by t h i s t e c h n i q u e h a d s i m i l a r
s h a p e s ( F i g 15, c u r v e s 3 a n d 4 ) .
C o m p a r i n g t h e r e s u l t s o f t h e s e t w o s e r i e s it c a n b e c l a i m e d t h a t in t h e c a s e o f
t h e first s e r i e s n o t all o f t h e o x i d e h a s b e e n r e d u c e d d u r i n g s h o r t c a t h o d i c
p o l a r i s a t i o n at l o w o v e r p o t e n t i a l s a n d t h u s e v e n n o r m a l c h a r g i n g c u r v e s h a v e b e e n
m e a s u r e d on partially oxidised surface, but the c o m p e n s a t i o n c u r v e s on essentially
o x i d i s e d s u r f a c e . C o n s e q u e n t l y , t h e l a r g e a m o u n t s o f c h a r g e o b t a i n e d in t h e first
series are consistent with the change of the amount of N i ( O H ) rather than of 2

a d s o r b e d h y d r o g e n on the nickel surface A l t h o u g h t h e t e c h n i q u e u s e d in t h e

second series enables to carry out the m e a s u r e m e n t s on reduced surface, the
d e t e r m i n a t i o n o f t h e v a l u e o f Θ by t h i s m e t h o d is c o m p l i c a t e d as t h e d i f f e r e n c e
Η

b e t w e e n n o r m a l a n d c o m p e n s a t i o n c u r v e s is v e r y small (in b o t h c a s e s t h e m a i n p a r t
o f c h a r g e is r e q u i r e d for t h e o x i d a t i o n o f t h e e l e c t r o d e ) . In v i e w o f t h e a n a l y s i s
p r e s e n t e d a b o v e it m u s t b e c o n c l u d e d t h a t t h e d o u b l e c h a r g i n g m e t h o d is n o t
c o n v e n i e n t t o a p p l y for t h e d e t e r m i n a t i o n o f Θ in t h e c a s e o f n i c k e l .
Η

T h e c h a n g e o f Θ ° c a n b e s u c c e s s f u l l y d e t e r m i n e d o n n i c k e l in a l k a l i n e s o l u t i o n s
Η

using overpotential decay measurements [36,40,41,42,49,119]. Remarkable

p s e u d o c a p a c i t a n c e h a s b e e n o b s e r v e d in t h e r a n g e o f c o m p a r a t i v e l y low
o v e r p o t e n t i a l s d e p e n d e n t o n t h e t y p e o f n i c k e l e l e c t r o d e u s e d . In s o m e c a s e s a
250

m a x i m u m over 0 05 V has been noticed on C vs. η c u r v e s [ 4 1 , 1 1 9 ] . At l o w

o v e r p o t e n t i a l s t h e v a l u e s o f dT|/dt a n d t h e F a r a d a i c c u r r e n t i a r e q u i t e small a n d
t

t h e e r r o r o f t h e i r d e t e r m i n a t i o n is c o n s i d e r a b l y l a r g e if t o c o m p a r e w i t h t h e c a s e
of higher overpotentials T h e r e f o r e t h e v a l u e o f p s e u d o c a p a c i t a n c e in t h i s r e g i o n
o f o v e r p o t e n t i a l s c a n b e o b t a i n e d by o v e r p o t e n t i a l d e c a y c u r v e s w i t h g r e a t e r r o r
[ 1 2 0 ] In o r d e r t o o v e r c o m e t h i s s o u r c e o f e r r o r t h e s w i t c h - o v e r c u r v e s h a v e b e e n
used [42] T h e initial c u r r e n t d e n s i t y w a s s w i t c h e d off a n d a q u i t e small a n o d i c
current density i w a s switched on.
a

In t h i s c a s e t h e p s e u d o c a p a c i t a n c e c a n b e c a l c u l a t e d by t h e r e l a t i o n :

C = -ii±iiL. (30)
dr|/dt

T h e C v s η c u r v e s d e t e r m i n e d by o v e r p o t e n t i a l d e c a y a n d o v e r - s w i t c h i n g
t r a n s i e n t s a r e s h o w n in F i g . 17 a n d 18
It c a n b e s e e n t h a t if i is n o t t o o l a r g e C v s η c u r v e s c a l c u l a t e d f r o m d e c a y
a

a n d o v e r - s w i t c h i n g c u r v e s a r e c l o s e a n d o n l y at η < 0.1 t h e d i f f e r e n c e b e c o m e s
remarkable It is e v i d e n t t h a t at l o w o v e r p o t e n t i a l s m o r e r e l i a b l e r e s u l t s c a n b e
obtained on the basis of switch-over curves. The results obtained and the
theoretical values of Θ ° calculated according to the mechanism of slow
Η

r e c o m b i n a t i o n [ 7 8 ] a r e p r e s e n t e d in T a b l e 5.

Table 5
Surface c o v e r a g e of nickel electrodes with adsorbed hydrogen
Type of electrode at η , V
0.15 0.20 0.25 0.30
mechanically 0.058 0.075 0.092
—
polished
chemically 0.025 0.033 0.043 0.05
polished
theoretically 0.13 0.17 0.23 0.26
calculated value

It c a n b e s e e n t h a t Θ ° for m e c h a n i c a l l y p o l i s h e d n i c k e l is a p p r o x i m a t e l y t w o t i m e s
Η

l a r g e r t h a n for c h e m i c a l l y p o l i s h e d n i c k e l , b u t t h e v a l u e s o f Θ ° in b o t h c a s e s a r e
Η

e s s e n t i a l l y l o w e r t h a n t h a t p r e d i c t e d by t h e r e c o m b i n a t i o n t h e o r y o f t h e H E R .
 251

0.15 0.10 0.05 0

Overpotential, V
F i g . 17. C v s . η c u r v e s for m e c h a n i c a l l y p o l i s h e d n i c k e l in 0.5 Μ K O H .
2 2
1 - i = 0, 2 - i = 0 . 0 5 m A / c m , 3 - i = 0 . 2 m A / c m
a a a

Overpotential, V
F i g . 18. C v s . η c u r v e s for c h e m i c a l l y p o l i s h e d n i c k e l in 0.5 Μ K O H :
1 - o b t a i n e d b y o v e r p o t e n t i a l d e c a y , 2 - o b t a i n e d by a s w i t c h - o v e r c u r v e .
252

4.7. Influence of halide ions.

Halide ions belong to surface active ions having essential influence on the
r e g u l a r i t i e s o f t h e h y d r o g e n e v o l u t i o n r e a c t i o n It h a s b e e n e s t a b l i s h e d t h a t , as a
rule, halide ions d e c r e a s e the value of overpotential (at constant current density)
on the metals of high o v e r p o t e n t i a l and increase the value of η on the metals of
medium overpotential [19,93].
A c c o r d i n g t o t h e c o n v e n t i o n a l m o d e l t h e effect o f h a l i d e i o n s o n t h e v a l u e o f
h y d r o g e n overpotential can be divided into t w o parts:

Δη = A y + ATi ads (31)

T h e first t e r m o f e q u a t i o n ( 3 1 ) Δ η ψ ι reflects the influence of the adsorption of

halide ions on the overpotential d u e to the change of Tj-potential, the second
t e r m Δ η ^ is t h e shift o f o v e r p o t e n t i a l d u e t o t h e c h a n g e o f t h e v a l u e o f t h e
e n e r g y o f t h e m e t a l - h y d r o g e n b o n d . T h i s i n t e r p r e t a t i o n is a little simplified as t h e
t e r m Δ η ^ is r e l a t e d t o t h e i n f l u e n c e o f s e v e r a l f a c t o r s [ 9 3 ] :

Αη^=^ψ\ [Η )η Λ + ^ + ψ + ^ ^ (32)

T h u s t h e a d s o r p t i o n o f h a l i d e i o n s m a y c h a n g e t h e e n e r g i e s o f a d s o r p t i o n o f all t h e
reactants taking part of the HER.
The t w o terms of the expression (31) have contrary influence on h y d r o g e n
o v e r p o t e n t i a l T h e first t e r m h a s a n e g a t i v e v a l u e a n d d e c r e a s e s t h e o v e r p o t e n t i a l ,
t h e s e c o n d o n e h a s a p o s i t i v e v a l u e a n d i n c r e a s e s t h e h y d r o g e n o v e r p o t e n t i a l . It
h a s b e e n s h o w n t h a t in t h e c a s e o f n i c k e l t h e s e c o n d t e r m is t h e d o m i n a t i n g o n e
T h e effect o f h a l i d e i o n s o n h y d r o g e n o v e r p o t e n t i a l o n n i c k e l d e p e n d s o n t h e
nature and c o n c e n t r a t i o n of halide ions and on the potential of the e l e c t r o d e The
shift o f o v e r p o t e n t i a l for m e c h a n i c a l l y p o l i s h e d n i c k e l c a u s e d by t h e i n f l u e n c e o f
d i f f e r e n t h a l i d e i o n s o f different c o n c e n t r a t i o n at c o n s t a n t c u r r e n t d e n s i t y is s h o w n
in F i g . 19. T h e p o l a r i s a t i o n c u r v e s for n i c k e l m e a s u r e d in t h e p r e s e n c e o f i o d i d e
i o n s in t h e s o l u t i o n a r e p r e s e n t e d in F i g . 2 0 . It h a s b e e n e s t a b l i s h e d t h a t , in
g e n e r a l , Tafel s l o p e h d e c r e a s e s w i t h t h e i n c r e a s e o f t h e c o n c e n t r a t i o n o f h a l i d e
ions T h i s t e n d s t o b e e s s e n t i a l l y r e m a r k a b l e in t h e p r e s e n c e o f b r o m i d e i o n s in
t h e s o l u t i o n at l o w o v e r p o t e n t i a l s ( i . e . t h e r e g i o n o f p o t e n t i a l s c o r r e s p o n d i n g t o
small n e g a t i v e o r e v e n p o s i t i v e c h a r g e o f n i c k e l s u r f a c e ) w h e r e t h e a d s o r p t i o n o f
h a l i d e i o n s is s t r o n g e r a n d c o n s e q u e n t l y its effect o n o v e r p o t e n t i a l is m o r e
e s s e n t i a l ( F i g . 2 1 , c u r v e 3)
In o r d e r t o g e t m o r e d e t a i l e d i n f o r m a t i o n a b o u t t h e effect o f c h e m i s o r p t i o n o f
h a l i d e i o n s on t h e h y d r o g e n o v e r p o t e n t i a l t h e c h a n g e o f t h e o v e r p o t e n t i a l after
switching over the polarising current has been investigated. The electrode was
p o l a r i s e d at h i g h c u r r e n t d e n s i t y i r T h e n it w a s s w i t c h e d o v e r t o a l o w e r
 253

5 4 3 2 1
3
- log c H a l . (mol/dm )

F i g . 19 I n f l u e n c e o f t h e p r e s e n c e o f h a l i d e i o n s o f d i f f e r e n t c o n c e n t r a t i o n o n
h y d r o g e n o v e r p o t e n t i a l o n m e c h a n i c a l l y p o l i s h e d n i c k e l in 0 . 0 5 Μ H S 0
2 4

solution:
1 - C I - i o n s , 2 - Br" - i o n s , 3 - Γ - i o n s .

4 3
- log i ( A / c m )

F i g . 2 0 . P o l a r i s a t i o n c u r v e s for m e c h a n i c a l l y p o l i s h e d n i c k e l in 0 . 0 5 Μ H S0
2 4

s o l u t i o n s c o n t a i n i n g KI o f d i f f e r e n t c o n c e n t r a t i o n :
5 4 3 2 1
1 - 0, 2 - 1 0 - M , 3 - 1 0 " M , 4 - 1 0 " M , 5 - 1 0 - M , 6 - 1 0 ' M .
254

F i g . 2 1 . P o l a r i s a t i o n c u r v e s for m e c h a n i c a l l y p o l i s h e d n i c k e l m e a s u r e d in:
1 - 0 . 0 5 Μ H S 0 , 2,3 - 0 . 0 5 Μ H S 0 + 0.1 Μ K B r , 2 - b u i l t u p o n t h e
2 4 2 4

basis of switch-over curves, 3 - usually measured.

6 8

Time, ms

F i g . 2 2 M i n i m a o f s w i t c h - o v e r c u r v e s in 0 05 Μ H S 0 2 4 solution:
1 2 3
1 - 1 0 ' Μ KC1, 2 - 1 0 ' Μ K B r , 3 - 10" Μ Κ Ι .
 255

v a l u e i a n d t h e c o r r e s p o n d i n g o v e r p o t e n t i a l - t i m e t r a n s i e n t h a s b e e n r e g i s t e r e d by
2

oscillograph It h a s b e e n e s t a b l i s h e d t h a t if t h e r a t i o o f i : i is w i t h i n t h e r a n g e
{ 2

of 5 - 1000, the switch-over curves exhibit a clear minimum (Fig. 2 2 ) , the depth of
which d e p e n d s both on the value of i : i and on the nature and c o n c e n t r a t i o n of
} 2

h a l i d e i o n s in t h e s o l u t i o n T h e s w i t c h - o v e r c u r v e s m e a s u r e d in t h e a b s e n c e o f
h a l i d e i o n s in t h e s o l u t i o n h a v e a s m o o t h s h a p e - t h e o v e r p o t e n t i a l falls
m o n o t o n o u s l y with time according to the discharge of the double layer of the
e l e c t r o d e by t h e r e d u c t i o n o f h y d r o g e n i o n s Therefore the appearance of a
minimum on o v e r p o t e n t i a l - t i m e curves can be related to the change of adsorption
of halide ions on the electrode surface T h e c h a n g e o f t h e p o t e n t i a l in t h e
direction of less negative potentials during s w i t c h - o v e r curves favours the
adsorption of halide ions on nickel surface and consequently increases h y d r o g e n
overpotential
The results of the switch-over m e a s u r e m e n t s show clearly that the adsorption of
h a l i d e i o n s is n o t a v e r y q u i c k p r o c e s s [ 9 0 ] It is p o s s i b l e t o d i v i d e s w i t c h - o v e r
curves into t w o branches In t h e first o n e t h e o v e r p o t e n t i a l d e c r e a s e s r a p i d l y d u e
t o t h e d i s c h a r g e o f d o u b l e l a y e r a n d t h e n b e g i n s t o i n c r e a s e s l o w l y in t h e s e c o n d
branch At t h e m i n i m u m p o i n t t h e d o u b l e l a y e r is a l m o s t d i s c h a r g e d a l r e a d y b u t
the adsorption of halide ions has not changed remarkably yet. Using the values of
o v e r p o t e n t i a l at m i n i m u m p o i n t s o f t h e s w i t c h - o v e r c u r v e s t o d i f f e r e n t c u r r e n t
d e n s i t i e s it is p o s s i b l e t o build u p t h e p o l a r i s a t i o n c u r v e s c o r r e s p o n d i n g t o t h e
nearly constant d e g r e e of the surface c o v e r a g e with chemisorbed halide ions.
A p o l a r i s a t i o n c u r v e built u p by t h e d e s c r i b e d a b o v e m e t h o d s t a r t i n g f r o m t h e
initial o v e r p o t e n t i a l 0 . 3 V for n i c k e l e l e c t r o d e s in t h e p r e s e n c e o f b r o m i d e i o n s in
t h e s o l u t i o n is p r e s e n t e d in F i g 21 As one can see such a non-stationary
p o l a r i s a t i o n c u r v e is shifted t o w a r d s l o w e r o v e r p o t e n t i a l s in c o m p a r i s o n w i t h t h e
o n e m e a s u r e d by t h e u s u a l m e t h o d in s t e a d y - s t a t e c o n d i t i o n s a n d h a s a r e m a r k a b l y
h i g h e r Tafel s l o p e b e i n g similar t o t h a t o f t h e c u r v e m e a s u r e d in t h e a b s e n c e o f
b r o m i d e i o n s T h e d e c r e a s e o f t h e Tafel s l o p e o f n o n - s t a t i o n a r y p o l a r i s a t i o n c u r v e
in t h e r e g i o n o f l o w e r o v e r p o t e n t i a l s c a n b e e x p l a i n e d a s s u m i n g t h a t in t h i s c a s e
the adsorption of halide ions has already somewhat changed before the minimum
point on the η vs time curve. Consequently the degree of the surface c o v e r a g e
with strongly a d s o r b e d halide ions d o e s not c o r r e s p o n d any longer t o t h e constant
initial v a l u e b u t h a s s o m e w h a t i n c r e a s e d .
M i n i m a o f η v s . t c u r v e s in t h e p r e s e n c e o f c h l o r i d e i o n s in t h e s o l u t i o n a r e
c o n s i d e r a b l y l e s s d e e p t h a n in c a s e o f b r o m i d e a n d e s p e c i a l l y i o d i d e i o n s . T h e
u p p e r p a r t o f t h e n o n - s t a t i o n a r y p o l a r i s a t i o n c u r v e h a s e v e n a little h i g h e r v a l u e o f
T a f e l s l o p e t h a n t h a t in t h e a b s e n c e o f h a l i d e i o n s in t h e s o l u t i o n . This
p h e n o m e n o n s u p p o r t s t h e s u p p o s i t i o n o f t h e r e m a r k a b l y m o r e p r o n o u n c e d effect
of the change of Y ^ p o t e n t i a l on the h y d r o g e n overpotential on nickel d u e t o the
c o m p a r a t i v e l y w e a k a d s o r p t i o n of chloride ions differently from t h e s t r o n g
adsorption of b r o m i d e and iodide ions on nickel .
Very interesting results have been obtained from o v e r p o t e n t i a l decay and
s w i t c h - o v e r m e a s u r e m e n t s for s o l u t i o n s c o n t a i n i n g b r o m i d e o r i o d i d e - i o n s . T h e
essential pseudocapacitance has been established on C vs. η - curves (Fig. 23)
[91], being dependent on the nature and concentration of halide ions and on the
256

initial o v e r p o t e n t i a l . T h e v a l u e s o f c a p a c i t a n c e o f t h e c u r v e s h a v e b e e n c a l c u l a t e d
by e q u a t i o n s ( 1 4 ) a n d ( 3 0 ) u s i n g t h e c u r r e n t d e n s i t i e s c o r r e s p o n d i n g t o n o n -
stationary polarisation c u r v e s built up on the basis of s w i t c h - o v e r c u r v e s . Such a
p s e u d o c a p a c i t a n c e m a y b e c a u s e d e i t h e r by t h e e s s e n t i a l c h a r g e t r a n s f e r d u r i n g t h e
c h e m i s o r p t i o n o f h a l i d e i o n s o r by t h e i o n i s a t i o n o f s o m e a m o u n t o f t h e a d s o r b e d
o n n i c k e l a t o m i c h y d r o g e n d i s p l a c e d by h a l i d e - i o n s . It is r e a s o n a b l e t o a s s u m e
t h a t b o t h o f t h e p r o c e s s e s t a k e p l a c e at t h e s a m e t i m e . T h e m a x i m u m v a l u e o f
total c h a r g e calculated from C vs. η - curves c o r r e s p o n d s to 0.15 m o n o l a y e r of
a t o m i c h y d r o g e n ( a s s u m i n g t h a t all t h e c h a r g e c o r r e s p o n d s t o t h e i o n i s a t i o n o f
atomic hydrogen).
It is n o t v e r y e a s y t o d i s t i n g u i s h t h e s e p r o c e s s e s by e l e c t r o c h e m i c a l m e t h o d s .
H o w e v e r , t h e c h a r g e t r a n s f e r s e e m s t o b e t h e d o m i n a t i n g p r o c e s s at h i g h e r
o v e r p o t e n t i a l s a n d t h e c o n t r i b u t i o n o f a d s o r b e d h y d r o g e n m a y b e m o r e e s s e n t i a l at
l o w e r o v e r p o t e n t i a l s . It is r e a s o n a b l e t o c o n c l u d e t h a t at l e a s t h a l f o f t h e t o t a l
c h a r g e is c o n n e c t e d w i t h t h e i o n i s a t i o n o f a d s o r b e d a t o m i c h y d r o g e n a n d
c o n s e q u e n t l y 0 . 0 7 < Δ Θ < 0 . 1 5 . T h i s a d s o r b e d h y d r o g e n m a y c o r r e s p o n d t o Θ ° at
Η Η

0
l o w o v e r p o t e n t i a l s b u t t h e m a i n p a r t o f it c o r r e s p o n d s e v i d e n t l y t o θ

5
^ 1000 h

c 1
c3

& 500 h

0.4 0.3 0.2 0.1

Overpotential, V
F i g . 2 3 C v s . η c u r v e s for m e c h a n i c a l l y p o l i s h e d n i c k e l in 0 . 2 5 Μ H S 0 2 4 solution
in t h e p r e s e n c e o f Γ - i o n s o f d i f f e r e n t c o n c e n t r a t i o n :
5 4 3 2
1-0, 2 - 10" M, 3 - 10- M, 4 - 10" M, 5 - 2·10- Μ.
 257

5. C O N C L U S I O N .

The most important data concerning h y d r o g e n evolution reaction on nickel have

b e e n p r e s e n t e d in t h i s r e v i e w . O n t h e b a s i s o f t h e s e d a t a it is p o s s i b l e t o i n d u c e a
conclusion of the mechanism of this process. Nevertheless, the experimental
r e s u l t s d o n o t f o r m a w h o l e , a s s o m e v e r y i m p o r t a n t d a t a a r e still n o t a v a i l a b l e y e t
and therefore the a m b i g u o u s conclusion cannot be done. T h e results p r e s e n t e d
a b o v e s u g g e s t that the r e c o m b i n a t i o n of a d s o r b e d a t o m i c h y d r o g e n (the Tafel
r e a c t i o n ) c a n n o t b e t h e r a t e - d e t e r m i n i n g s t e p o f t h e o v e r a l l p r o c e s s , at l e a s t at
higher current densities. The strongest s u p p o r t s to this statement are the quite
ς
low values of θ Η a n d its w e a k d e p e n d e n c e o n o v e r p o t e n t i a l at m o d e r a t e a n d h i g h
current densities.
In t h e first a p p r o x i m a t i o n it is r e a s o n a b l e t o s u p p o s e t h a t t h e h y d r o g e n
evolution reaction proceeds according to the Volmer-Heyrovsky mechanism. In
order to determine the rate-determining step to the Volmer-Heyrovsky mechanism
t h e v a l u e o f l i m i t i n g c o v e r a g e θ „ ,. is n e e d e d [ 1191:

θ Η = ( 3 3 )
* " ϊ ^ '
w h e r e k is t h e r a t e c o n s t a n t o f d i s c h a r g e r e a c t i o n a n d k - t h e r a t e c o n s t a n t o f
] 2

electrochemical desorption reaction. F u r t h e r t h e a n a l y s i s will b e c a r r i e d o u t

s e p a r a t e l y for a c i d i c a n d a l k a l i n e s o l u t i o n s .
A s it h a s b e e n s h o w n a b o v e , for a c i d i c s o l u t i o n s it is i m p o s s i b l e t o d e t e r m i n e
the value of stoichiometric number and the c o v e r a g e of the surface of nickel
electrodes with adsorbed a t o m i c h y d r o g e n at t h e r e v e r s i b l e h y d r o g e n p o t e n t i a l
a n d e v e n at l o w c a t h o d i c p o t e n t i a l s d u e t o t h e r e m a r k a b l e r a t e o f t h e c o r r o s i o n
p r o c e s s o f n i c k e l at t h e s e p o t e n t i a l s in t h i s m e d i a . T h u s t h e s i g n i f i c a n t p a r a m e t e r s
o f m e c h a n i s m Θ a n d its v a r i a t i o n at l o w o v e r p o t e n t i a l s a r e n o t k n o w n for a c i d i c
Η

media, therefore we have no direct criteria to distinguish the slow discharge (the
V o l m e r ' s m e c h a n i s m ) and the slow electrochemical d e s o r p t i o n (the H e y r o v s k y ' s
mechanism).
S o m e a d d i t i o n a l i n f o r m a t i o n is a v a i l a b l e f r o m t h e s t u d i e s o f t h e i n f l u e n c e o f
halide ions on hydrogen overpotential. It h a s b e e n e s t a b l i s h e d t h a t h a l i d e i o n s
i n c r e a s e t h e h y d r o g e n o v e r p o t e n t i a l r e m a r k a b l y . It is r e a s o n a b l e t o s u p p o s e t h a t
the term Δ η in E q . ( 3 1 ) is m a i n l y d e t e r m i n e d b y t h e d e c r e a s e o f t h e e n e r g y o f
adsorption of the hydrogen g . Consequently, the increase of overpotential due to
H

the adsorption of halide ions suggests that the rate-determining step has to be the
discharge. Contrary to this, the decrease of g H has to accelerate the
electrochemical desorption of adsorbed hydrogen and therefore to d e c r e a s e
h y d r o g e n o v e r p o t e n t i a l if t h i s s t e p is l i m i t i n g t h e o v e r a l l p r o c e s s . N e v e r t h e l e s s , it
must be noted that the rate of the electrochemical desorption cannot be much
higher than that of the discharge reaction as the c o v e r a g e of nickel surface with
a d s o r b e d h y d r o g e n is n o t e x t r e m e l y l o w t h a t is c o n f i r m e d b y t h e i o n i s a t i o n o f
remarkable amount of hydrogen due to the adsorption of halide ions.
258

M o r e e x p e r i m e n t a l d a t a a r e a v a i l a b l e for a l k a l i n e m e d i a . T h e m o s t s i g n i f i c a n t
o f t h e m a r e t h e d a t a o f Θ ° at c a t h o d i c p o t e n t i a l s .
Η It h a s b e e n s h o w n a b o v e t h a t
Θ ° d o e s n o t e x c e e d 0.1 p a r t o f m o n o l a y e r M u c h h i g h e r v a l u e s o f Θ ° o b t a i n e d in
Η Η

[41] are probably c o n n e c t e d with the specific p r o p e r t i e s of the nickel e l e c t r o d e s

u s e d in t h i s w o r k . T h e c a t a l y t i c a c t i v i t y o f t h e n i c k e l film e v a p o r a t e d o n g l a s s
u s e d in [ 4 1 ] s e e m s t o b e m u c h h i g h e r t h a n t h e a c t i v i t y o f t h e s u r f a c e o f u s u a l
c o m p a c t nickel
c
A s t h e e x a c t v a l u e o f 9 „ is still n o t a v a i l a b l e w e c a n o n l y c o n s i d e r t h a t θ „ is
J
Η H,lim

within the limits 0.05 < Θ Η < 0.95 and c o n s e q u e n t l y the ratio of c o n s t a n t s k k r 2

varies a p p r o x i m a t e l y from 0 05 to 20. The same values of the ratio of c o n s t a n t s of

discharge and the electrochemical desorption have been obtained from the analysis
of g a l v a n o s t a t i c switch-in c u r v e s [121] and a similar ratio has been calculated on
the basis of overpotential decay transients [119,120].
In t h e c a s e o f s o m e t y p e s o f n i c k e l e l e c t r o d e s e s p e c i a l l y for a n o d i c a l l y a c t i v a t e d
o n e s t w o s l o p e s o n Tafel p l o t s h a v e b e e n o b s e r v e d [ 4 9 , 1 2 0 , 1 2 2 ] . T h e a p p e a r a n c e
o f t w o d i f f e r e n t r e g i o n s o n p o l a r i s a t i o n c u r v e s h a s b e e n e x p l a i n e d by t h e i n f l u e n c e
o f a n o t h e r e l e c t r o d e p r o c e s s at l o w e r o v e r p o t e n t i a l s e.g. o f e l e c t r o d e c o r r o s i o n .
S u c h an e s s e n t i a l i n f l u e n c e o f c o r r o s i o n o n t h e r e g u l a r i t i e s o f h y d r o g e n e v o l u t i o n
in a l k a l i n e m e d i a is n o t p r o b a b l e . T h e m o d e l c a l c u l a t i o n s o f p o l a r i s a t i o n c u r v e s
for v a r i o u s c o m b i n a t i o n s o f t h e v a l u e s o f t h e r a t e c o n s t a n t s o f d i f f e r e n t s t e p s h a v e
been carried out [123] It h a s b e e n e s t a b l i s h e d t h a t t h e r e c a n n o t b e f o u n d a
c o m b i n a t i o n of rate c o n s t a n t s and Θ values to derive a polarisation c u r v e with
Η

t w o r e m a r k a b l y d i f f e r e n t s l o p e s h a v i n g t h e h i g h e r s l o p e at l o w e r o v e r p o t e n t i a l s .
Such a shape of polarisation curves obtained experimentally can be explained
s u p p o s i n g that t h e r e are places of t w o different type on the surface of nickel
electrode It is p o s s i b l e t h a t o n t h e l a r g e r p a r t o f t h e s u r f a c e t h e V o l m e r -
H e y r o v s k y m e c h a n i s m is o p e r a t i v e b u t o n t h e s m a l l e r p a r t t h e r a t e - d e t e r m i n i n g
s t e p is t h e r e c o m b i n a t i o n o f h y d r o g e n a t o m s ( t h e T a f e l ' s r e a c t i o n ) [ 1 4 ] . A l m o s t
t h e s a m e r e s u l t s c a n b e o b t a i n e d if t o s u p p o s e t h a t t h e p r o c e s s is o c c u r r i n g by t h e
V o l m e r - H e y r o v s k y m e c h a n i s m but with essentially different transfer coefficients
on different p a r t s of surface [ 1 2 2 ] .
T h e a p p r o a c h to the m e c h a n i s m of the h y d r o g e n e v o l u t i o n r e a c t i o n p r e s e n t e d
a b o v e h a s b e e n d e v e l o p e d , in p r i n c i p l e , for m e t a l s o f high o v e r p o t e n t i a l . In t h e
c a s e o f m e t a l s o f m e d i u m o v e r p o t e n t i a l s u c h as n i c k e l it is v e r y p r o b a b l e t h a t
s i g n i f i c a n t i n t e r a c t i o n s b e t w e e n t h e m e t a l s u r f a c e a n d as i o n s as w a t e r m o l e c u l e s
t a k e p l a c e , e s p e c i a l l y if t h e m e t a l s u r f a c e h a s p o s i t i v e c h a r g e It is r e l i a b l e t h a t
s u c h i n t e r a c t i o n s h a v e an e s s e n t i a l effect o n t h e d i s t r i b u t i o n o f p o t e n t i a l o n t h e
m e t a l / s o l u t i o n i n t e r f a c e c a u s i n g s i g n i f i c a n t d i f f e r e n c e s in t h e r e g u l a r i t i e s o f
hydrogen evolution reaction Additional experimental and theoretical studies
s h o u l d b e c a r r i e d o u t b e f o r e any n o v e l a p p r o a c h c a n b e d e v e l o p e d .
 259

REFERENCES

1 J Tafel, Ζ Phys. Chem., 50 ( 1 9 0 5 ) 6 4 1 .

2 J A V B u t l e r , T r a n s . F a r a d a y S o c . , 19 ( 1 9 2 4 ) 7 2 9 .
3 . J. H e y r o v s k y , R e c . T r a v . C h i m . , 4 6 ( 1 9 2 7 ) 5 8 2 .
4. T. E r d e y - G r u z a n d M . V o l m e r , Z . P h y s . C h e m . , A 150 ( 1 9 3 0 ) 2 0 3 .
5. N . K o b o z e v a n d N . N e k r a s o v , Ζ E l e k t r o c h e m . , 3 6 ( 1 9 3 0 ) 5 2 8 .
6. R. G u r n e y , P r o c . R o y . S o c , A 134 ( 1 9 3 1 ) 137
7. A. F r u m k i n , Z . P h y s . C h e m . , A 160 ( 1 9 3 2 ) 1 16.
8. J . A . V . B u t l e r , T r a n s . F a r a d a y S o c , 2 8 ( 1 9 3 2 ) 3 7 9 .
9. T. E r d e y - G r u z a n d M . V o l m e r , Z . P h y s C h e m . , A 162 ( 1 9 3 2 ) 5 3 .
10. A. F r u m k i n , Z . P h y s . C h e m . , A 164 ( 1 9 3 3 ) 1 2 1 .
11 J H o r i u t i a n d M . P o l a n y i , A c t a P h y s i c o c h i m . U R S S , 2 ( 1 9 3 5 ) 5 0 5 .
12. J . A . V . B u t l e r , P r o c . R o y S o c , A 157 ( 1 9 3 6 ) 4 2 3 .
13 G. O k a m o t o , J H o r i u t i a n d K. H i r o t a , Sci. P a p . I n s t . P h y s . a n d C h e m . R e s . ,
29 (1936) 223
14 A. F r u m k i n , D i s c u s s . F a r a d a y S o c , N o . 1 ( 1 9 4 7 ) 5 7 .
15. J . O ' M B o c k r i s , in M o d e r n A s p e c t s o f E l e c t r o c h e m . , V o l . 1, E d s . J . O ' M .
Bockris and B E . Conway, Butterforths, L o n d o n , 1954, p. 2 0 1 .
16 K. V e t t e r , E l e k t r o c h e m i s c h e K i n e t i c , S p r i n g e r , B e r l i n , 1 9 6 1 .
17. J . O ' M B o c k r i s a n d E C . P o t t e r , J. E l e c t r o c h e m . S o c , 9 9 ( 1 9 5 2 ) 1 6 9 .
18 J . O ' M . B o c k r i s a n d R . G . H . W a t s o n , J C h e m . P h y s . , 4 6 ( 1 9 5 2 ) C 7 0 .
19. A . N F r u m k i n , in A d v a n c e s in E l e c t r o c h e m . a n d E l e c t r o c h e m . E n g . , V o l . 1,
Ed. P. Delahay, Interscience, N Y , L o n d o n , 1 9 6 1 , p. 65.
2 0 A . N F r u m k i n , in A d v a n c e s in E l e c t r o c h e m . a n d E l e c t r o c h e m . E n g . , V o l 3 ,
Ed. P. Delahay, Interscience, N Y , L o n d o n , 1963, p. 287
21 S. T r a s a t t i , in A d v a n c e s in E l e c t r o c h e m . S c i e n c e a n d E n g , V o l . 2 , E d s . H.
G e r i s c h e r a n d C W . T o b i a s , V C H , W e i n h e i m , 1 9 9 2 , p . 2.
2 2 . E . G i l e a d i a n d B E . C o n w a y , in M o d e r n A s p e c t s o f E l e c t r o c h e m i s t r y , V o l . 3 ,
E d s . J O'M B o c k r i s and B E . C o n w a y , B u t t e r w o r t h s , L o n d o n , 1964, p. 347.
23 B E . C o n w a y , T h e o r y and Principles of E l e c t r o d e P r o c e s s e s , R o n a l d Press,
NY, 1965.
24 A Slygin and A Frumkin, Acta Physicochim. U . R . S . S . , 3 ( 1 9 3 5 ) 79
2 5 A . N F r u m k i n , O A . P e t r i a n d R . V M a r v e t , J. E l e t r o a n a l C h e m . , 1 2 ( 1 9 6 6 )
504.
2 6 . J D P e a r s o n a n d J . A . V . B u t l e r , T r a n s . F a r a d a y S o c , 3 4 ( 1 9 3 8 ) 1 163
2 7 . Μ B r e i t e r , C A . K n o r r a n d W V o l k l , Z . E l e k t r o c h e m . , 59 ( 1 9 5 5 ) 6 8 1 .
2 8 Μ Α V D e v a n a t h a n , J . O ' M B o c k r i s a n d W . M e h l , J. E l e c t r o a n a l . C h e m . ,
1 ( 1 9 5 9 / 6 0 ) 143.
2 9 . Μ Α V D e v a n a t h a n a n d Μ S e l v a r a t n a m , T r a n s . F a r a d a y S o c , 56 ( 1 9 6 0 ) 1 8 2 0 .
3 0 H. G e r i s c h e r a n d W . M e h l , Z . E l e k t r o c h e m . , 59 ( 1 9 5 5 ) 1 0 4 9
3 1 . M . W B r e i t e r , J E l e c t r o a n a l . C h e m . , 19 ( 1 9 6 8 ) 131
3 2 . G A. A r m s t r o n g a n d L A V B u t l e r , T r a n s . F a r a d a y S o c , 2 9 ( 1 9 3 3 ) 1261
33 A . N . F r u m k i n , A c t a P h y s i c o c h i m . U R S S , 18 ( 1 9 4 3 ) 2 3 .
3 4 A . N F r u m k i n a n d Ν A l a d z h a l o v a , A c t a P h y s i c o c h i m U R S S , 19 ( 1 9 4 4 ) 1.
35 M B M o r l e y a n d F E W W e t m o r e , C a n . J. C h e m . , 3 4 ( 1 9 5 6 ) 3 5 9 .
260

36. V Ε P a s t a n d Ζ A I o f a , D o k l . A N S S S R , 106 ( 1 9 5 6 ) 1 0 5 2 .
37. B E C o n w a y a n d P L . B o u r g a u l t , T r a n s . F a r a d a y S o c , 58 ( 1 9 6 2 ) 5 9 3 .
38. U . V . P a l m a n d V . E . P a s t , D o k l . A H S S S R , 146 ( 1 9 6 2 ) 1 3 7 4 .
39. B E . C o n w a y a n d P L . B o u r g a u l t , C a n . J. C h e m . , 3 7 ( 1 9 5 9 ) 2 9 2 .
40. V . E . P a s t a n d Z . A . I o f a , Z h . F i z . K h i m . , 33 ( 1 9 5 9 ) 1 2 3 0 .
41. A. M a t s u d a a n d T. O h m o r i , J. R e s . I n s t . C a t a l y s i s , H o k k a i d o U n i v 10 ( 1 9 6 2 )
215.
42 L. T o h v e r , J. T a m m a n d V . P a s t , A c t a C o m m . U n i v . Tartuensis,
N o . 2 8 9 ( 1 9 7 1 ) 3.
43 R H . Burshtein, A G . Pchenitchnikov, V.D Kovalevskaja and M E . Belyaeva,
Elektrokhimiya 6 (1970) 1756.
44 D C . G r a h a m e , C h e m . R e v . , 41 ( 1 9 4 7 ) 4 4 1 .
45. Ρ Delahay, D o u b l e Layer and E l e c t r o d e Kinetics, Wiley, N Y , 1965.
46 A. H a m e l i n , in M o d e r n A s p e c t s o f E l e c t o c h e m i s t r y , V o l . 16, E d s . B E .
C o n w a y and J.O'M. B o c k r i s , P l e n u m P r e s s , N Y . 1985, p. 1
47. L. T a m m , J. T a m m a n d V P a s t , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 3 3 2 ( 1 9 7 4 )
3.
48. J. T a m m a n d V P a s t , in D o u b l e L a y e r a n d A d s o r p t i o n o n S o l i d E l e c t r o d e s ,
V o l 1, T a r t u U n i v e r s i t y , 1 9 6 8 , p . 1 5 8 .
49 L T o h v e r a n d J. Τ a m m , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 2 6 5 ( 1 9 7 0 ) 14.
50. A C . M a k r i d e s , J. E l e c t r o c h e m . S o c , 109 ( 1 9 6 2 ) 2 5 6 .
51 J . L W e i n i n g e r a n d M . W . B r e i t e r , J. E l e c t r o c h e m . S o c , 111 ( 1 9 6 4 ) 7 0 7 .
52 G A r m s t r o n g , F H i m s w o r t h a n d J . A . V . B u t l e r , P r o c . R o y . S o c , A 143
(1933) 89.
53. F. Will a n d C. K n o r r , Ζ E l e k t r o c h e m . , 6 4 ( 1 9 6 0 ) 2 5 8 , 2 7 0
54. Τ B i e g l e r , J. E l e c t r o c h e m . S o c , 116 ( 1 9 6 9 ) 1 1 3 1 . _ ^
55. B . G . P o u n d , in M o d e r n A s p e c t s o f E l e c t r o c h e m . , V o l . 2 5 , E d s . J . O ' M . B o c k r i s
and B E C o n w a y , 1993, p. 63
56 M . P o u r b a i x , A t l a s o f E l e c t r o c h e m . E q u i l i b r i a in A q u e o u s Solutions,
Pergamon, Oxford, 1966.
57. J . O ' M . B o c k r i s , J. M c B r e e n a n d L. N a n i s , J. E l e c t r o c h e m . S o c , 1 12 ( 1 9 6 5 )
1025.
58 I A. B a g o t s k a y a , Z h . F i z . K h i m . , 3 6 ( 1 9 6 2 ) 2 6 6 7 .
59 P A D o w d e n a n d P . W . R e y n o l d s , D i s c u s s . F a r a d a y S o c , 8 ( 1 9 5 0 ) 184.
60 B E . C o n w a y , E M B e a t t y and P A D D e M a i n e , E l e c t r o c h i m . Acta, 7 ( 1 9 6 2 )
39.
61. Y a . B S k u r a t n i k , Α . Ε K o z a c h i n s k i i , A . P . P c h e l n i k o v a n d V . V . L o s e v , J.
E l e c t r o a n a l . C h e m . , 3 6 6 ( 1 9 9 4 ) 31 1.
62 A Ε Kozachinskii, A.P. Pchelnikov, Y a . B . Skuratnik and V.V. L o s e v ,
Elektrokhimiya, 29 (1993) 508.
63. Ρ L u k o v t s e v , S. L e v i n a a n d A. F r u m k i n , Z h . Fiz. K h i m . , 13 ( 1 9 3 9 ) 9 1 6 .
64 A L e g r a n a n d S L e v i n a , Z h . F i z . K h i m . , 14 ( 1 9 4 0 ) 2 1 1 .
65 Ρ L u k o v t s e v a n d S L e v i n a , Z h . Fiz. K h i m , 21 ( 1 9 4 7 ) 5 9 9 .
66 J. O ' M . B o c k r i s a n d E C . P o t t e r , J. C h e m . P h y s . , 2 0 ( 1 9 5 8 ) 6 1 4 .
67 R. P i o n t e l l i , L. P e r a l d o B i c e l l i a n d A. L a . V e c c h i a , A t t i A c a d . N a z . L i n c e i
R e n d , CI Sci. F i s . M a t N a t . , 2 7 ( 1 9 5 9 ) 3 1 2 .
 261

6 8 . J M i e l u c h , B u l l . A c a d P o l S c i . , S e r Sci. C h i m . , 17 ( 1 9 6 9 ) 4 3
69 V A L a v r e n k o , L.N. Y a g u p o l s k a y a and V I . Tikush, Elektrokhimiya, 6 ( 1 9 7 0 )
887.
70 J Arold and J T a m m , E l e k t r o k h i m i y a , 25 ( 1 9 8 9 ) 4 1 8
7 1 . S. T r a s a t t i , E l e c t r o c h i m . A c t a , 2 8 ( 1 9 8 3 ) 1 0 8 3 .
72 J P i e l a s z e k , B u l l . A c a d . P o l . Sci , S e r Sci C h i m . , 2 0 ( 1 9 7 2 ) 4 8 7 , 61 1
73 L. T a m m , J T a m m a n d V P a s t , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 3 7 8 ( 1 9 7 6 )
3.
7 4 J. H o r i u t i , J. R e s . I n s t . C a t a l y s i s , H o k k a i d o U n i v . , 1 ( 1 9 4 8 ) 8.
7 5 . R. P a r s o n s , T r a n s F a r a d a y S o c , 4 7 ( 1 9 5 1 ) 1 3 3 2
7 6 K . J V e t t e r , Z . E l e k t r o c h e m . , 59 ( 1 9 5 5 ) 4 3 5
7 7 . A . N F r u m k i n , D o k l . A N S S S R , 119 ( 1 9 5 8 ) 3 1 8
78 J H o r i u t i , J R e s I n s t C a t a l y s i s , H o k k a i d o U n i v . , 4 ( 1 9 5 5 ) 5 5 .
7 9 J O ' M B o c k r i s a n d A . M . A z z a m , T r a n s . F a r a d a y S o c , 4 8 ( 1 9 5 2 ) 145
8 0 . H. K i t a a n d Τ Y a m a z a k i , J R e s I n s t . C a t a l y s i s , H o k k a i d o U n i v . , 11 ( 1 9 6 3 )
10.
81 Η K i t a a n d Ο N o m u r a , J. R e s . I n s t . C a t a l y s i s , H o k k a i d o U n i v . , 12 ( 1 9 6 4 )
107.
82 B . N K a b a n o v , A c t a Physicochim U . R . S . S . , 4 ( 1 9 3 6 ) 846.
8 3 . Τ Silk, J. T a m m a n d V. P a s t , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 3 3 2 ( 1 9 7 4 )
18.
8 4 . R . M . W i g h t m a n a n d D O Wipf, in E l e c t r o a n a l . C h e m . , V o l . 15, E d . A . J . B a r d ,
Dekker, NY, 1989, ρ 267
85. E.I. M i c h a i l o v a and Z.A. Iofa, E l e k t r o k h i m i y a , 1 ( 1 9 6 5 ) 107.
86 A T Skljarov and Y a M . Kolotyrkin, E l e k t r o k h i m i y a , 3 ( 1 9 6 7 ) 1462.
87. N . D T o m a s h o v , N M . S t r u k o v and L P . Vershina, Elektrokhimiya, 5 ( 1 9 6 9 )
26.
8 8 . S. R a s h k o v , J H r i s t o v a a n d N . P a n g a r o v , C o m p t . R e n d . A c a d . B u l g . S c i . , 2 3
(1970)387.
89 E.I. M i k h a i l o v a and Z.A. Iofa, E l e k t r o k h i m i y a , 6 ( 1 9 7 0 ) 231
9 0 . L. T a m m , J. T a m m a n d V. P a s t , E l e c t r o k h i m i y a , 9 ( 1 9 7 3 ) 1 3 8 2 .
9 1 . L. T a m m , J. T a m m a n d V. P a s t , E l e k t r o k h i m i y a , 1 1 ( 1 9 7 5 ) 1 5 8 1 .
9 2 . L. T a m m , J T a m m a n d V. P a s t , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 3 7 8 ( 1 9 7 6 )
17.
9 3 . J. T a m m , L. T a m m , P . V a r e s a n d J. A r o l d , A c t a C o m m . U n i v T a r t u e n s i s , N o .
905 ( 1 9 9 0 ) 107.
94. L.I. Krishtalik, Elektrokhimiya, 6 ( 1 9 7 0 ) 1 165.
9 5 . L . I . K r i s h t a l i k , J. E l e c t r o a n a l . C h e m . , 35 ( 1 9 7 2 ) 1 5 7 .
9 6 . J. T a m m , L. T a m m a n d P . V a r e s , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 7 5 7
(1986) 34.
97 A.N. Frumkin, V.S B a g o t s k i i , Z.A. Iofa and B.N. K a b a n o v , Kinetika
Electrodnyh Processov, M o s c o w Univ. Press, M o s c o w , 1952.
9 8 . I . A . A m m a r a n d S.A. A w a d , J. P h y s . C h e m . , 6 0 ( 1 9 5 6 ) 8 3 7 .
9 9 . A. M a t s u d a a n d Τ O h m o r i , J. R e s . I n s t . C a t a l y s i s , H o k k a i d o U n i v . , 10 ( 1 9 6 2 )
203.
1 0 0 . L. T a m m , J. T a m m a n d V . P a s t , E l e k t r o k h i m i y a , 10 ( 1 9 7 4 ) 8 3 .
262

101 L T a m m , J. T a m m and V P a s t , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 3 3 2
( 1 9 7 4 ) 12
102 J T a m m a n d L. T a m m , E l e k t r o k h i m i y a , 12 ( 1 9 7 6 ) 9 5 5 .
103 J T a m m , L T a m m a n d A. P i k a t , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 3 7 8
(1976) 11.
104 L.I A n t r o p o v , Z h Fiz. K h i m . , 2 6 ( 1 9 5 2 ) 1 6 8 8 .
105 L . I . A n t r o p o v , Z h Fiz. K h i m . , 2 8 ( 1 9 5 4 ) 1 3 3 6
106 A N . F r u m k i n , P o t e n t i a l o f Z e r o C h a r g e , 2 n d ed. N a u k a , M o s c o w , 1 9 8 2 .
107 V L K h e i f e t s a n d L S R e i s h a k r i t , U c h Z a p . L e n i n g r G o s U n i v . , S e r .
K h i m . , N o . 13 ( 1 9 5 3 ) 1 7 3 .
108 V L K h e i f e t s a n d Β S K r a s i k o v , Z h . F i z K h i m , 3 1 ( 1 9 5 7 ) 1 9 9 2 .
109 J . O ' M B o c k r i s , S.D. A r g a d e a n d Ε G i l e a d i , E l e c t r o c h i m . A c t a , 14 ( 1 9 6 9 )
110 E M L a z a r o v a , E l e k t r o k h i m i y a , 14 ( 1 9 7 8 ) 1 3 0 0 .
H I T . O h m o r i , J. E l e c t r o a n a l . C h e m . , 1 5 7 ( 1 9 8 3 ) 159.
112 J A r o l d a n d J T a m m , E l e k t r o k h i m i y a , 2 5 ( 1 9 8 9 ) 141 1.
113 R P a r s o n s a n d F . G . R Z o b e l , J. E l e c t r o a n a l . C h e m . , 9 ( 1 9 6 5 ) 3 3 3 .
114 G u L i n - i n , Ν A S h u m i l o v a a n d V S . B a g o t s k i i , E l e k t r o k h i m i y a , 3 ( 1 9 6 7 ) .
115 Ν A S h u m i l o v a a n d V S B a g o t s k i i , E l e c t r o c h i m . A c t a , 13 ( 1 9 6 8 ) 2 8 6 .
116 V . R L o o d m a a , M E . H a g a and V . E . P a s t , Isv. V u z o v , K h i m i K h i m .
Tekhnol., 9 (1966) 794.
117 L A . B u r k a l t s e v a a n d A G P s h e n i c h n i k o v , E l e k t r o k h i m i y a , 12 ( 1 9 7 6 ) 4 2 .
118 V P a s t , J T a m m a n d T. T e n n o , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 2 1 9
( 1 9 6 8 ) 30
119 B . C o n w a y a n d L. B a i , C h e m . S o c , F a r a d a y T r a n s I, 81 ( 1 9 8 5 ) 1 8 4 1 .
120 A L a s i a a n d A R a m i , J E l e c t r o a n a l . C h e m . , 2 9 4 ( 1 9 9 0 ) 1 2 3 .
121 T. Silk, J T a m m a n d V P a s t , A c t a C o m m . U n i v T a r t u e n s i s , N o . 3 7 8 ( 1 9 7 6 )
121.
1 2 2 . J. T a m m a n d V. P a s t , A c t a C o m m . U n i v . T a r t u e n s i s , N o . 2 3 5 ( 1 9 6 9 ) 2 0 .
Research in Chemical Kinetics, Volume 3
R.G. Compton and G. Hancock (editors) 263
© 1995 Elsevier Science B.V. All rights reserved.

Kinetics & mechanisms of silylene reactions : A prototype for

gas-phase acid/base chemistry
8 b
R. Becerra and R. Walsh
a
lnstituto Quimica Fisica "Rocasolano", C.S.I.C., C/Serrano, 119, 28006,
Madrid, Spain

department of Chemistry, University of Reading, Whiteknights, P.O.Box 224,

Reading RG6 2AD., U.K

Contents
1. BACKGROUND
1.1 Introduction
1.2 Silylene structure, bonding and stability
1.3 Nature of silylene reaction processes
1.4 Comparison of silylenes with carbenes
1.5 Advantages of the gas-phase

2. REQUIREMENTS FOR DIRECT STUDY

2.1 General
2.2 Silylene photochemical sources
2.3 Detection and monitoring
2.4 Data processing

3. THE INSERTION PROCESS

3.1 SiH insertion reactions
2

(a) SiH + H /D /HD (and SiD 4- H /D )

2 2 2 2 2 2

(b) SiH + SiH 2 4

(c) SiH + Si H2 2 6

(d) SiH + Si H2 3 8

(e) SiH + Me SiH . (n = 1-4)

2 n 4 n

(f) SiH + alkanes

2

3.2 Insertion reactions of substituted silylenes

(a) SiMe + Me SiH . (n = 0-4)
2 n 4 n

(b) MeSiH + Me SiH . (n = 0-3) n 4 n

(c) PhSiH + Me SiH . (n = 0-3) n 4 n

(d) CISiH, SiCI , SiF 2 2

264

3.3 General discussion of the Si-Η insertion process

(a) Comparison of substituent effects
(b) Nature of the Si-Η insertion process
(c) Intermediate complex mechanism

4. THE J7-TYPE ADDITION PROCESS

4.1 SiH addition reactions
2

(a) SiH + C H / C D
2 2 4 2 4

(b) SiH + C H 2 3 6

(c) SiH + C H 2 4 6

(d) SiH + C H / C D
2 2 2 2 2

4.2 SiMe addition reactions

2

4.3 SiCI addition reactions

2

4.4 General discussion of the tf-type addition process

(a) Comparisons of silylene reactivity
(b) Energy considerations

5. REACTIONS WITH O-DONOR MOLECULES

5.1 SiMe + water, alcohols, ethers
2

5.2 SiH + M e 0
2 2

5.3 Reactivity of complexed SiMe 2

6. REACTIONS WITH SMALL INORGANIC MOLECULES

7. GENERAL COMMENTS AND CONCLUSIONS

7.1 The future for silylene kinetic studies
7.2 The general prospect

1. BACKGROUND

1.1 Introduction
Gas-phase chemistry is understandably dominated by studies of
hydrocarbon reactions. This situation has come about because of the
importance of fossil fuels in the world economy. Gas-phase hydrocarbon
reactions, in the main, fall into the categories of pyrolysis, or combustion
including atmospheric oxidation (so-called 'slow-combustion'). The mechanisms
of such processes almost inevitably involve free-radical intermediates and so
gas-phase chemistry has grown up largely illustrated by examples of free-radical
reactions.
By contrast, other areas of chemistry are poorly represented. One such is
that of silane and organosilicon chemistry. Organosilicon chemistry has a
considerable importance in polymer production, the best-known examples of
which are the silicones [1]. Silane chemistry is important to the vapour
deposition processes (CVD) which are used to produce silicon-containing
electronic materials [1]. The characteristic mechanisms of silane and
 265

organosilicon compound breakdown are, in general, not free radical processes,

but rather they involve the intermediacy of silylenes. Silylene reactions
represent a different type of chemistry than free radical processes and therefore
this gives silane and organosilicon chemistry a distinctive character. It is true
that silylenes are the silicon analogues of carbenes and there is a considerably
developed chemistry of carbenes. However, as we shall see, studies of carbene
reactions are hampered by certain difficulties which do not affect silylene
studies. Thus in a variety of ways progress has been greater in understanding
the behaviour of silylenes than that of carbenes. A caveat should be entered.
All silylenes thus far studied possess singlet ground states, whereas many
carbenes exist in triplet ground states. It is the chemistry and reactivity of the
singlet states of silylenes which has developed. There is as yet no experimental
knowledge of the chemical behaviour of triplet (excited) states of silylenes.
Our knowledge of silylenes has grown steadily since the first review of their
occurrence and behaviour by Atwell and Weyenberg in 1969 [2,3]. The
accumulating knowledge of their chemical behaviour has been particularly well
described in a series of notable reviews by Gaspar [4,5], as well as in articles
by others [6-10]. Most recently Strausz and his colleagues [11] have
documented the increasing data base of experimental rate constants for silylene
reactions. It is not the purpose of the present article to repeat these surveys
in full detail. Rather our objective is to describe the insight gained into the
mechanisms of prototype silylene processes as a result of the application of
modern, time-resolved, techniques to measurements of their rate constants.
The first experiment of this kind was performed by Inoue and Suzuki [12] in
1985, who used laser induced fluorescence to detect and monitor SiH [13,14].
2

Since 1985 a steadily accumulating number of measurements of absolute rate

constants have been published. The majority, but not all, of this work has been
undertaken in the laboratories of Gaspar, Strausz, Jasinski and in our own.
Prior to these investigations, information about the formation, intermediacy and
reactivity of silylenes was inferred indirectly from final product distributions in
numerous extensive experiments of a more classical kind using thermal or
photochemical decomposition of appropriate organosilicon precursors to
generate the silylenes. In these earlier studies direct measurements of rate
constants of the intermediate silylenes were not possible, although relative rate
constants could be obtained from carefully designed competitive studies, or
inferred from kinetic modelling. Thus a picture emerged of silylenes as fairly
reactive intermediates (not as reactive as carbenes) participating in reactions of
certain types (insertions, additions) with certain bond types (but not others).
The information was substantial but the picture lacked sharp detail. In a
notable statement in their 1987 article [15] Gaspar and colleagues provided a
focus for progress by framing the questions - What are the transition-state
structures? What is the balance between electrophilic and nucleophilic
character of the attacking silylene species in the transition state? What role do
steric and electronic effects play in silylene reactions? To these we may add
the more basic question - How do the structures and energies of silylenes affect
their reactivity? This article attempts to show how the absolute rate studies
have begun to throw some light on these matters. In order to set the scene we
266

describe in this section some basic knowledge about the structure and
energetics of silylenes as well as of their reaction types. Also included here is
some information provided by theory which has progressed in tandem with
experimental measurements. A very useful account of recent theoretical
calculations on silylenes has been given by Apeloig [16].

1.2 Silylene structure, bonding and stability

Ί
Symmetrical silylenes, such as SiH , have C structures with Α ground
2 2 v Ί

states and B^ excited states [16]. An orbital occupancy diagram is shown in

3

Figure 1.

Ί 3
Α, state Β Ί state

Figure 1. Orbital diagram for the lowest states of silylene

Experimental studies, based on analysis of the visible vibronic absorption

spectrum [17], have given values for the Si-Η bond length of 1.516A and HSiH
bond angle of 9 2 . 8 ° . Theoretical studies [16] are in excellent agreement.
A comparison with the angles of methylene, C H , is shown in Table 1, 2

which indicates that in general the bond angles are ca 10° less in SiH than 2

CH . 2

Table 1
Bond angle values for lower electronic states of methylene and silylene
Species HCH Species HSiH
1
CH 2 ( A,) 102.4 a
SiH ('Α,)
2 92.8 a

C H ^B,)
2 130 s 3
SiH ( B,)
2 118.5 b

a. Experimental values, b. Theoretical value.

 267

This is consistent with the general observation that silicon prefers to have
non-bonding electrons in atomic orbitals with a high percentage of s-character.
Luke et al [18] have calculated that in the Α state the lone pair is in a hybrid
λ
λ

orbital with 8 9 % s-character in SiH compared with 6 0 % s-character in C H .

2 2

This reluctance of s orbitals to become involved in the bonding in silylene,

reflects the decreased screening of valence s orbitals (compared with ρ orbitals)
from the atomic nucleus in higher row elements of the periodic table. It is
connected with the inert pair effect and general increasing prevalence of lower
valence states in higher rows of the periodic table. This phenomenon has been
quantified by Walsh for silylenes in the concept of Divalent State Stabilisation
Energy or DSSE [19-21]. This is defined, for the silylene SiX , as follows: 2

DSSE = 0 ( X S i - X ) - D(X Si-X)

3 2

where D(X Si-X) and D(X Si-X) correspond to the first and second dissociation
3 2

energies of the silane SiX . This equation reflects the fact that whereas
4

breaking the first bond effectively produces little or no rehybridisation (the Si

3
atomic orbitals being essentially sp in both SiX and SiX ) breaking the second
4 3

bond causes the rehybridisation of the Si atomic orbitals to become mainly ρ in

the Si-X bonds, and s in the lone pair. Walsh [21] has estimated values for
DSSE for a number of silylenes and these are shown in Table 2.

Table 2
Thermodynamic data for silylenes [21]
Species AH ° I kJ mol
f
1
DSSE I kJ mol 1

SiH 2 273 ± 2 93 ± 3
a
SiMe 2 135 ± 8 121 ± 1 2
SiF 2 -590 ± 8 205 ± 42

SiCI 2
-169 ± 8 159 ± 17
SiBr 2
-46 ± 8 142 ± 50
Sil 2
-92 ± 8 134 ± 54

a. Revised in this review.

The numbers on which these values depend are taken from the most recent
figures for bond dissociation energies [21,22] in combination with enthalpies
of formation, AH °, of the silylenes themselves [21,23]. Although the errors are
f

in some cases still large, it is nevertheless clear that silylenes are stabilised by
electronegative substituents. The small but significant difference for SiMe and 2

SiH shows that, as expected, methyl groups act as weakly electronegative

2

substituents. The expectation in this case derives from Pauling

268

electronegativity values (2.7 for C vs 2.2 for Η relative to 1.8 for Si). The
usefulness and significance of the idea of DSSE has been expanded and
extended to other elements by Grev [24].
In a theoretical examination of substituent effects in silylenes, Luke et al [18],
have reached similar conclusions for the monosubstituted silylenes RSiH, where
R represent groups -Li, -BeH, -BH , - C H , - N H , -OH and -F. These results show
2 3 2

that the most effective stabilisation is achieved with /7-donors (eg -NH ). This
2

suggests the importance of change transfer to the empty 3p orbital and

involvement of resonance structures shown in figure 2.

Figure 2. Contributing structures for silylenes with lone-pair carrying

substituents

It is important to stress the point that stabilisation of silylenes, as discussed

here, is essentially thermodynamic in nature, and does not necessarily have any
bearing on the kinetics and general reactivity of the silylenes (as we shall see).

1.3 Nature of silylene reaction processes

It is clear from Figure 1 that (^(silylenes have two potential sites of
reactivity, viz. an empty ρ orbital and a filled lone-pair orbital. All their reactions
involve the use of both of these orbitals. Since these reactions all involve
acceptance of a lone pair (by the Si 3p orbital) and the donation of a long pair
(the Si s pair) to the creation of a new bond (with a substrate) these processes
2

may be considered examples of acid/base behaviour. More commonly the

terminology "electrophilic" and "nucleophilic" has been used by theoreticians
who have noted on the basis of ab initio calculations [25-27] that these
processes appear to occur in stages. This may be illustrated for one of the most
widely studied processes, the Si-Η insertion reaction shown in Figure 3.
These ideas are very useful since they offer experimentalists a conceptual
framework within which to probe the nature of the processes. If the stages of
interaction do not occur simultaneously the question arises of whether other
intermediates with only one lone pair interaction, may be involved. Thus
questions which naturally arise are.. Does the "electrophilic" stage precede the
"nucleophilic" stage? .. or vice versa? .. Or, indeed, do they occur
 269

synchronously? In fact all of the major types of silylene process viz. σ-bond
insertion, #7-bond addition and reaction with η-type donors, can be considered
in this manner. These ideas are further illustrated during the discussion of
specific reaction examples.

1.4 Comparison of silylenes with carbenes

Although a detailed discussion of carbene 0 Α λ ) reactions is beyond the
scope of this article, in principle they may be equally well discussed within the
same framework of approach adopted here. However, despite intensive
investigation [28] progress has been hampered by certain practical difficulties

Figure 3. Interactions involved during the Si-Η insertion process of silylenes

with their study. There are two main reasons for this. The first arises because,
3 1
for many carbenes, the ground state is B., and even if Α can be prepared free
Ί

from B
3
V
1
it is extremely difficult in kinetic studies of the A , state with reactive
substrate molecules to separate the component of removal of by reaction
from that due to deactivation (collisionally induced intersystem crossing)
[29,30]. The second reason for difficulty arises with substituted alkyl carbenes.
Alkyl substituents (such as methyl) are useful probes as modifiers of reactivity
but bimolecular studies of alkylcarbenes are made difficult (if not impossible) by
the very rapid intramolecular 1,2 Η-shift process leading to alkenes [31], for
example, dimethylcarbene to propene:
270

CMe 2 -* MeCH = C H 2

The analogous rearrangement for dimethylsilylene, viz:

SiMe 2 MeSiH = C H 2

has a substantial activation energy barrier [32,33] and is therefore very slow,
thus permitting the study of SiMe kinetics, untroubled by intramolecular
2

reaction. This does not mean that a great deal of understanding of carbene
reactions has not been achieved [34,35] but rather the potentially fruitful route
of direct measurement of absolute rate constants for a variety of carbenes is
considerably hindered.
A specific point that is worth mentioning is that of energy release. This
may be illustrated by the comparison of the prototype reactions of C H and 2

SiH as shown in Table 3.

2

Table 3
Energy release for prototype methylene and silylene reactions
-AH°
1
Reaction / kJ mol' Ref.

^CH 2 C H -* C Hg 4 2 439 [36]

^CH + C H ~* ο-ΟβΗρ
2 2 4 429 [36]
Ί
ΟΗ 2 + H 0 ^2 CH3OH 390 [36]

*S\H 2 + SiH -* Si H 4 2 6
227 [21]
^SiH + C H -* c-SiC Hg
2 2 4 2 201 [37]
1
SiH 2 + H 0 -* H3S1OH
2 311 estd.

All reactions are highly exothermic, since each process involves the creation
of two bonds at the expense of one. However the C H reactions produce ca 2

double the energy release of the SiH reactions. This is because in the SiH 2 2

case the reactions involve giving up the DSSE of silylene. The practical
outcome is that the products of an SiH reaction are initially much less 2

vibrational^ excited than those of a C H reaction. 2

One of the well known phenomena of C H reactions is that of "chemical 2

activation" in which initially formed products rearrange further to secondary

products [38]. This characteristic is less evident in SiH reactions although, as 2

will be seen, the consequences of energy release manifest themselves in the

different general phenomenon of the third-body stabilisation requirement for
association processes. This is discussed in more detail in specific examples.
 271

1.5 Advantages of the gas-phase

One advantage of the gas-phase is very straightforward VIZ. that the
reactions can be studied free of the complicating effects of solvent. It is clear
that species with the ambiphilic character of silylenes will be particularly prone
to interaction with solvents. This may be illustrated by time-resolved studies
of reactions of SiMe with alcohols and ethers carried out by Levin ET AL [39].
2

The reactivity of SiMe is modified in the presence of these O-donor species,

2

as discussed in section 5, almost certainly due to complexation with the

solvent. However in non-polar solvents, free of impurities, there is no reason
why solution studies should not yield essentially similar results to the gas-
phase. If differences occur they should be instructive. Moreover there are many
silylenes which are difficult to generate in the gas-phase because they have
involatile precursors. In these cases solution studies are the only possibility.
Another advantage of the gas-phase is that it allows us to probe the
consequences of energy release. The pressure dependences associated with
third-body assisted association reactions, permit us to gain information about
the thermal stability of the products (as we shall see). This is not possible in
solution.
Of course most real-life chemistry involves the use of solvents and therefore
solution reactions unquestionably need detailed kinetic investigation in their
own right. But the gas-phase serves as the vital reference point of the solvent-
free system. Thus this review is for the most part limited to gas-phase data and
solution studies are only discussed where they throw extra light on the
mechanism in question.

2. REQUIREMENTS FOR DIRECT STUDY

2.1 General
Silylenes are among the most reactive transient species known and
therefore their study via direct monitoring requires appropriate fast response
techniques [40]. The most widely used method is that of laser flash photolysis
with laser absorption detection used in our laboratories [41] and those of
Jasinski [42] (so-called LRAFKS technique). Typically, the most reactive
silylenes, like SiH itself, when created in a transient high concentration pulse,
2

decay away via reaction on microsecond timescales. This imposes two

requirements on an experimental monitoring system (i) a rapid silylene creation
source and (ii) a fast response detection system. The first is generally met by
use of excimer (sometimes more correctly called exciplex) lasers with typically
intense UV pulses (of energy CA 100 mJ/pulse and with CA 10 nsec linewidths)
in combination with a suitable UV absorbing molecular silylene precursor.
Conventional UV flash lamp sources can also be used for silylenes such as
SiCI , which are less reactive [43]. Other methods for creating silylenes include
2

the use of infrared multiphoton decomposition (IRMPD) [44], although this has
not been much used in direct kinetic studies. The second requirement is met
by use of optical techniques in combination with fast response electronic
recording. Optical absorption can be used where the combination of (a) the
272

silylene extinction coefficient (or oscillator strength), (b) the transient

concentration and (c) the absorbing path length are sufficient to obtain good
signals. This has been used for the silylenes, SiH , SiD , MeSiH, SiMe , PhSiH,
2 2 2

SiCI , SiBr . The alternative is to use optical emission as in laser induced

2 2

fluorescence (LIF). This is in principle more sensitive but is limited to those

silylenes which emit (SiH , SiHCI, SiF ). These points are expanded in the
2 2

following sections. They are illustrated mainly by our own techniques and
experience but the general principles apply to all such studies.

2.2 Silylene photochemical sources

The UV photochemistry of silicon containing molecules is not a well
developed subject. However, it is known that both aromatic silanes and non-
aromatic disilanes have pronounced UV absorptions. It turns out that these
molecules do provide most of the known photochemical sources of silylenes.
A list of these precursor species is shown in Table 4.

Table 4
Silylene photochemical sources
Silylene Precursor λ 1 nm Ref.
SiH 2 PhSiH 3 193 [12]
[42]
[45]
SiH 2 SiH l 3 248 [42]
SiH 2 Si H 2 6 193 [46]
Si H 3 8

SiD 2 PhSiD 3 193 [47]

SiMe 2 PMDS a
193 [48]
MeSiH DMDS b
193 [49]
PhSiH PhSiH 3 193 [50]
CISiH CISiH 3 193 [51]
CI SiH
2 2 [52]
SiCI 2 Si CI2 6 UV C
[43]
SiBr 2 SiBr 4 UV C
[53]

a. Pentamethyldisilane. b. 1,2-Dimethyldisilane. c. Broad band.

There is almost certainly scope for the development of further molecular

sources. One of the problems with existing sources is the occurrence of side
reactions. With the more hydrogenated disilanes, such as MeSiH SiH Me dust 2 2
 273

is formed simultaneously with MeSiH [49]. This can be minimised with

sufficiently low precursor pressures. With PhSiH both SiH and PhSiH are
3 2

formed simultaneously [54] but since these absorb in different places, they do
not interfere with one another. A good quantum yield of silylene is important
since it is advantageous to keep the precursor pressure as low as possible in
order to minimise kinetic losses of silylene through reaction with the precursor
itself.
A particular problem with current sources is the use of 193nm radiation.
Many molecules of potential interest as reactive substrates for the silylenes
generated, themselves absorb and undergo photochemistry at this wavelength.
This makes them difficult or impossible to study at present.

2.3 Detection and monitoring

Silylenes possess strong absorptions in the visible and UV regions arising
Ί 1
from electronic transitions of the type Β « - A . The spectroscopy of silylenes
Ί 1

has been studied by both gaseous flash photolysis methods and in low
temperature matrices. The first silylene spectrum to be characterised was that
of SiH itself observed in the flash photolysis of PhSiH [55]. Subsequently,
2 3

much improved spectra were obtained by Dubois, Herzberg and Verma [56] in
the 480-650 nm region, from silylene produced by the flash discharge of silane
in excess hydrogen. The spectrum consisted of seven bands, with the
strongest bands at 610.0, 579.6 and 552.6 nm. A detailed rotational analysis
of these bands was carried out by Dubois [17]. Jasinski and Chu [57] have
recorded the most highly resolved spectrum to date using laser absorption flash

I
4 3 2 3 ,|
7 5
ζ I
I I

j ii I
CO 4,
cr
< 2j
<
ζ
Z.J-i
ζ
ο
8
I
CO
<

17252.00 17268.00
VACUUM WAVENUMBER/cm" 1

Figure 4. A portion of the visible rovibronic absorption spectrum of SiH [57]. 2

274

kinetic spectroscopy. They scanned the spectral region from 17242 to 17352
1 Ί
cm" , around the origin of the (0,2,0) «- (0,0,0) vibronic band of the ^ «- Α.,
transition. Their spectrum consisted of approximately 100 single rotational
lines showing that it was considerably richer than previously believed. About
half the lines were readily assigned using known SiH transitions [17]. A
2

portion of the spectrum is shown in Figure 4 including the intense Q , 5 line R

0

which has been used by Jasinski [42] and ourselves [45] to monitor SiH . 2

The narrow line widths of ca 0.05 cm' (FWHM) of this highly resolved
1

(small molecule) spectrum means that a narrow line optical probe is required for
its monitoring. This is provided by a single-mode (ring) dye laser which can be
1
tuned to the appropriate wavenumber (17259.50 cm' ). For SiD an assigned 2

high resolution rovibronic spectrum is not available but from a point-by-point

scan. Mason, Frey and Walsh [47] have obtained a spectrum suitable for kinetic
studies using laser flash photolysis of PhSiD . 3

Other silylenes have broader band spectra many of which have been
recorded by matrix isolation and documented by West et al [58]. An example
of a spectrum of this type is that for MeSiH shown in Figure 5. The gaseous
spectrum is typically slightly red-shifted relative to that of the matrix. Table 5
shows a list of absorbing regions for silylenes which have been investigated in
the gas phase. There may be an error in the published PhSiH spectrum [50].
This is discussed in section 3. These species place less demand on the optical
probe system, and an argon ion laser has proven to be a convenient absorption

υ
I I 1 1 1 1 1
420 440 460 480 500 520 540
wavelength / nm

Figure 5. The visible absorption spectrum of MeSiH

monitor. Several silylenes (MeSiH, SiMe , PhSiH) possess absorptions

2

conveniently within the range (450-515 nm) of the several lines available for
 275

study. Another convenient feature of many of the silylene absorption spectra

is their occurrence in the visible region, which is uncluttered by absorptions of
many of the desirable reactant molecules. The use of lasers as detectors offers
the benefit of increasing the absorption path length by means of multipassing
the beam (using Whites optics) which increases path lengths in practice to well
over 1 metre [48]. An important experimental check, is to verify that the
kinetics are independent of the absorption wavelength.

Table 5
Approximate absorption (monitoring) wavelengths for silylenes ( B <- A, band)
1
1
1

Silylene λ 1 nm Ref
SiH 2 480-650 [42]
(579.39) [45]
SiD 2 480-650 [47]
(575.14)
SiMe 2 420-500 [48]
MeSiH 430-520 [49]
PhSiH 460-520 [50]
SiCI 2 308-328 [43]
SiBr 2 340-400 [53]

A schematic diagram of the laser flash photolysis apparatus and

experimental setup for such experiments in our laboratories is shown in Figure
6. With modern fast response photodiodes and electronic signal detection,
amplification and storage are relatively straightforward. RF noise from the
excimer laser pulse can be a problem but with suitable screening of the
electronics can be kept to a minimum. The raw experimental result consists of
the time evolution of the transient silylene absorption over a suitable time scale
(usually microseconds). Signal-to-noise is improved by the accumulation of the
traces from several photolysis runs (excimer shots) using a transient digitizer.
It is important to check that such experimental repetition does not seriously
perturb the gaseous reaction mixtures during a series of photolysis shots. To
avoid reactant depletion or product interference gaseous flowing systems can
be employed. However if materials are costly or of limited availability a static
reaction vessel can be used provided the irradiated volume is a relatively small
proportion of the total ( < 5%). In this way, with shot repetition rates of ca
1Hz, there is plenty of time for replenishment of the reaction zone by diffusion
between photolysis laser pulses.
Under typical operating conditions, excimer laser pulses of ca 100 mJ can
create transient concentrations in the reaction zone of ca 0.1-1 mTorr ( 1 0 1 3

molecule cm" ) which give detectable absorbances of ca 0.05. Precise

3
276

Argon Ion Ring Dye Laser

Laser

Chopper
Ring
Dye Οψ9Γ j^M Iodine j ^
Laser er
Polarizer
Control

Wavemeter

Lock In

PC Amp.

V
l > ] Polarizer

Photodiode

Figure 6. Schematic diagram of the laser flash photolysis apparatus used to

study kinetics of SiH reactions [45]
2
 277

knowledge of the silylene concentrations are not usually required since the
decay kinetics are almost invariably quasi-first order.
Gaseous reaction mixtures usually consist of a few mTorr of precursor
(sufficient to give a silylene signal), an appropriate pressure of reactive
substrate (which can be from a few mTorr up to several Torr, depending on
reactivity) and an inert diluent gas. The diluent serves two purposes. The first
function is to thermalize the products of laser photolysis so that the silylene is
present with a (relaxed) Boltzmann energy distribution. Usually 5 Torr (and
often less) are sufficient for this purpose. Polyatomic bath gases, such as SF 6

which does not interfere with the chemistry, are best for this. Secondly, the
inert diluent serves as a pressure control for reaction when the overall kinetics
are pressure dependent. In practice the maximum pressure has turned out to be
usually ca 100 Torr for many silylene studies. Signals appear to be significantly
quenched at higher pressures (although the reasons are not understood). At
higher pressures care has to be taken to ensure complete mixing of reagents.

2.4 Data processing

The extremely high reactivities of silylenes mean that even in the presence
of only a few mTorr of substrates, silylene decay is dominated by bimolecular
reaction with the substrate, viz:

SiX + substrate -» product

2

Since under operating conditions, substrate concentrations are always greater

than those of the silylene, SiX , this means that the SiX decay kinetics are
2 2

quasi-first order with a rate constant, k , linearly dependent on [substrate].

obs

The true second order rate constant is obtainable from studies of this
dependence. This is standard kinetic treatment, easily programable with
microcomputers with standard error packages.
Occasionally (for example in the reactions of SiMe with ethers [59]) more
2

complex kinetics arise and this can be handled through alternative computer
fitting routines or numerical integration packages, for a hypothesised
mechanistic model.
For further experimental details of silylene studies, readers are referred to
individual papers.

3. THE INSERTION PROCESS

It is often stated that silylenes characteristically insert into S i - Η , Si-OR and

O - H bonds. While this is formally correct, in fact it is believed that the
reactions with Si-OR and O-H bonds proceed via initial silylene co-ordination
to oxygen. Therefore these latter reactions are treated separately (in section
5). This section, therefore, is devoted largely to the Si-Η insertion process,
although the important reaction with H and its isotopic variants is also
2

included.
278

3.1 S i H insertion reactions

2

(a) SiH + Ho/Do/HD (and SiD + H /D )

2 2 2 2

The first absolute rate constant for the reaction of SiH with H was 2 2

reported by Inoue and Suzuki [12] using the laser-induced fluorescence method
(LIF) to monitor the concentration of SiH . At the total pressure of 1.8 Torr 2
13 3 1 1
(He) they found a rate constant of ( 1 . 0 ± 0 . 4 ) χ 10 cm molec" s" . Nearly at
the same time Jasinski [60] measured the rate constant for SiH with D using 2 2

the laser resonance absorption flash kinetic spectroscopy (LRAFKS) technique.

The rate constant obtained at 2-3 Torr total pressure (He) was (2.7 ± 0 . 7 ) χ
10 c m molec'V . This value is much higher than that obtained by Inoue and
1 2 3 1

Suzuki for reaction with H . 2

Jasinski reasoned that his rate constant more nearly represented the true
rate of insertion because only one of the six possible channels for
decomposition of the vibrational^ highly excited adduct, S i H D \ leads back 2 2

to SiH , whereas S i H \ formed from SiH + H will revert to reactants unless

2 4 2 2

collisionally stabilised thus making it a much more pressure dependent reaction.

The immediate result of these measurements, however, was to dispel the
view that the reaction of SiH + H had a small but positive activation energy
2 2
1 1
variously estimated at 23.1 kJ mol" [61] and 43 kJ mol" [62]. These values
were derived from relative rate measurements which were based (in hindsight)
on unreliable assumptions. If correct the room temperature rate constant for
SiH -f H would have had to be at least four orders of magnitude lower than
2 2

observed. Interestingly at this time theoretical calculations showed a drop in

barrier from ca 30 kJ mol" [27,63-65] to 7 kJ mol" [66]. The magnitude of
1 1

the barrier was measured, using the same technique as Jasinski, by Baggott et
al [67] who obtained a temperature independent rate constant of (1.8 ± 0 . 2 ) χ
101 2
c m molec'V for SiH + D over the range 268-330 Κ (at 5 Torr total
3 1
2 2

pressure in SF ). This corresponds to zero activation energy and the authors

6
1
estimated a maximum possible value of 1.3 kJ mol" . More recently Al-Rubaiey
and Walsh [68] have extended the temperature range for this reaction up to
1
553 Κ and now found a negative activation energy of -2.03 kJ mol" .
Jasinski and Chu [57] measured absolute rate constants for the SiH + H 2 2

reaction in the 1-100 Torr pressure range (He). The pressure dependence of the
rate constants for this reaction was reproduced by RRKM calculations and the
high pressure limiting value derived of k°° = 3.2 χ 10 cm" molec" s" , in good 1 2 3 1 1

agreement with the value obtained from the SiH + D experiments. The rate 2 2
13 3 1 1
constant measured at 2 Torr total pressure was 2.7x10" cm molec" s" , in
rough agreement with the value obtained at 1.8 Torr by Inoue and Suzuki [12].
These studies also confirmed that the reactions of SiH with HD and D show 2 2

much less pressure dependence and the isotope effect appears to be small [57].
Table 6 summarises these results and also includes a recent value for the
rate constant for SiD + H , which indicates that the isotope effects, although
2 2

small are certainly not zero [47]. Just as for the SiH + D system, this reaction 2 2

is almost pressure independent and should represent the true rate of insertion
(Jasinski's argument).
The study of SiD + H has been extended up to higher temperatures [69].
2 2

The Arrhenius parameters for these two processes are shown in Table 7 along
 279

with those obtained in a comprehensive (unpublished) study of the (pressure

dependent) SiH + H and SiD + D reactions [68].
2 2 2 2

These most recent investigations [68,69] extend the previous studies, but
are consistent with them. They include transition state theory calculations of
isotope effects which are beyond the scope of this review, but which provide
a distinction between two published transition state structures for the SiH + 2

H reaction. These are shown in the Figure 7.

2

Table 6
Rate constants for SiH + H and isotopic variants at room temperature
2 2

Reaction /r/10 12 3
cm molec"V 1
Ref.
SiH + H
2 2 3.2 a
[57]
SiH + D
2 2 2.6 ± 0.7 [60]
1.9 ± 0.7 [67]
SiH 2 + HD 2.0 ± 0.4 [57]
SiD + H 2 2 3.8 ± 0.1 [47]
a. Infinite pressure value.

Table 7
Arrhenius parameters for SiH + H and isotopic variants
2 2

Τ range/ Κ
1
Reaction 3
log(y4/cm molec" s" ) 1 1
£ /kJ m o l
a

SiH 2 + D 2 300-553 -12.07 -2.03

SiD + H
2 2 294-528 -11.87 -2.54

a
SiH 2 + H 2 300-513 -12.01 3
-2.34
b
-11.75 b
-1.84

8
SiD + D
2 2 298-498 -12.35 3
-3.14
b
-12.09 b
-3.52

a. Infinite pressure values obtained from extrapolated rate constants.

b. Values obtained via transition state theory of isotope effects.

The isotope effects support the ab initio structure (C symmetry) of Gordon s

et al [66] rather than the semi empirical one ( C symmetry) of Roenijk et al 2 v

[70]. However RRKM calculations based on the ab initio structure do not

reproduce the measured pressure dependence of the rate constants very well
[68]. Nevertheless the ab initio structure seems more likely since it represents
280

a geometry of approach of H to SiH in which the electrophilic interaction

2 2

between the H bonding electrons and empty ρ orbital of SiH is favoured thus
2 2

strongly suggesting an electrophilically-led reaction.

Η
Η Η
Si ::'
Η Η
Η

C 2 v structure [70] C s structure [66]

Figure 7. Transition state structures for reaction of SiH + H 2 2

(b) SiH + SiH^ 2

Insertion of SiH into the Si-Η bond is one of the key processes in chemical
2

vapour deposition. Insertion into the Si-Η bonds of SiH is the most 4

fundamental of these processes. Once again direct measurements have

supplanted the older, much lower estimates of the rate constant [61,62]. Inoue
and Suzuki [12] measured the absolute rate constant, using the LIF technique,
for this reaction at 1 Torr total pressure (He). At 298 Κ they found a value of
(1.1 ± 0 . 2 ) χ 1 0 " c m molec'V . Jasinski and Chu [57], using the LRAFKS
10 3 1

1 1 3
technique, determined the absolute rate constant to be (6.7 ± 0 . 7 ) χ 10 c m
1
molec'V at 1 Torr (He) in reasonable agreement. These authors also studied
the effect of pressure up to 95 Torr and performed RRKM calculations to
reproduce the observed dependence. This yielded the high pressure limiting rate
constant, k°° = 2.2 χ 1 0 " c m molec'V . 10 3 1

Using the same technique Baggott et al [71] measured absolute rate

constants for this reaction and obtained values ca 1.7 times higher over the
same pressure range (He bath gas), thus leading to a value for k°° = 4 . 0 χ 10 1 0

c m molec'V . The cause of this discrepancy is not obvious.

3 1

Another study by Dietrich et al [72] using IRMPD to generate SiH from 2

SiH yields pressure dependent rate constants leading to k°° = (1.1 ± 0 . 2 ) χ

4

1 0 " c m molec V . However these experiments were carried out under solid
10 3 1
 281

deposition conditions and it is not clear that this corresponds to comparable

(298 K) temperatures. A more complete temperature and pressure study has
been carried out by Becerra et al [73], yielding the rate constants a selection of
which are shown in Table 8.

Table 8
Rate constants for SiH + SiH at two pressures and six temperatures
2 4

k I 10 1 0 3
cm molec'V 1

T/K Ρ = 10 Torr Ρ = oo a

296 4.1 ± 0.3 4.6 ± 0.3

363 3.2 ± 0.3 3.9 ± 0.3
432 2.3 ± 0.3 3.2 ± 0.3
488 1.6 ± 0.3 2.6 ± 0.3
578 1.2 ± 0.4 2.6 ± 0.4
658 0.8 ± 0.4 2.2 ± 0.4

a. Values obtained by extrapolation

-9.3 Η

log([SF ]/Torr)
e

Figure 8. Pressure and temperature dependence of rate constants for SiH 2 +

SiH 4
282

In this table, k°° values were obtained by fitting RRKM calculated fall-off
curves as shown in Figure 8. The modelling of these data used a variational
transition state. The 298 Κ value of k°° confirms the earlier measurements in
our laboratory and points to the high collisional efficiency of this reaction.
Arrhenius treatment of the temperature dependence of k°° leads to A°° = 1.3
1 0 3 1 1 1
χ 10 c m molec" s and E = -3.0 kJ mol" although the plot is slightly
a

curved. The negative activation energy is consistent with the idea of a weak
intermediate complex on the approach potential for this reaction. Ab initio
1
calculations [73] support the presence of such a complex at ca 52 kJ mol"
below the reaction threshold although the barrier to rearrangement of this
1
complex to S i H is only 6.5 kJ mol' . The structure of the complex is shown
2 6

in Figure 9.

1.525/*/ Χ 2.046/*

Figure 9. Structure of the intermediate complex in the reaction of SiH + SiH 2 4

The geometry of the complex suggests substantial transfer of Η from SiH 4

to SiH again supporting the idea of an electrophilically led insertion process.

2

Although thermochemical considerations are not the focus of this review,

the rate constants of this reaction system and its reverse [74] are the basis of
1
the currently most precise value for LH °(S\H ) = 273 ± 2 kJ mol" [75].
f 2

(c) SiH + S i H 2 2 6

Once again direct measurements have supplanted older values [61,62].

Absolute rate constants have been obtained in the same laboratories as for SiH 2

+ SiH [12,57,71,75]. Values are shown in Table 9.

4
10 3 1 1
The value of Dietrich et al [72] ( 2 . 7 ± 0 . 4 ) χ 1 0 " c m molec" s" ) was
obtained at an uncertain temperature and is not included. The main discrepancy
is between Jasinski and Chu [57] and the most recent values found in our
laboratories [75]. Additionally we have found a pressure independent reaction
between 3 and 30 Torr in SF and only a slight dependence on bath gas (Ar,
6

C H , SF ). Inoue and Suzuki's value [12] is in line with this. Jasinski and Chu
3 8 6

measured a rate constant variation of more than a factor of 2 between 1 and

9.5 Torr (He). Our feeling is that there is some error here because the
 283

energised S i H * product has an alternative pathway to decomposition which

3 8

is certainly predominant at higher temperatures [76] VIZ:

SiH + S i H
2 2 6 τ± S i H 3 8
#
SiH SiH + SiH
3 4

This means that pressure dependence in this system should be relatively

unimportant.

Table 9
Comparison of room temperature rate constants for SiH + S i H 2 2 6

Ρ 1 Torr
1 0 3 1
Ar/10 cm molec'V Reference
5.7 ± 0.2 1 ( He ) [12]
1.5 ± 0.2 1 ( He ) [57]
2.8 ± 0.3 5 ( He ) [57]
3.4 ± 0.3 9.5 ( He) [57]
4.6 ± 0.7 5 ( Ar ) [71]
5.8 ± 0.6 10 ( Ar ) [75]
6.6 ± 0.3 10 ( SF ) 6 [75]
6.4 ± 0.2 10 ( C H ) 3 8 [75]

The temperature dependence (295-590 K) of this reaction, studied by

Becerra ETAI [75], gives the average Arrhenius parameters A = 3.1 χ 1 0 " c m 1 0 3

molec" s"\ E = -1.9 kJ mol" , although the Arrhenius plot is slightly curved.
1
A
1

The negative temperature dependence and high collisional efficiency of this

reaction suggest that similar considerations determine its mechanism as for SiH 2

+ SiH . 4

(d) SiH + S i H
2 3 8

No published kinetic data exists for this reaction. Its kinetics have been
investigated by Becerra er AL [75], in an unpublished study, for a temperature
range of 297-595 Κ in C H and Ar buffer gases, and at six temperatures over
3 8

the range 295 to 578 Κ in SF . S i H was in this case both the precursor of
6 3 8

SiH and the substrate for the reaction. The reaction was tested for pressure
2

dependence over the range 1 - 30 Torr. None was found.

10 3 1
Rate constants at room temperature were (7.8 ± 0.9) χ 1 0 " c m molec'V
1 0 3 1 1
in SF and ( 8 . 0 ± 0 . 2 ) χ 10
6 c m molec" s" in C H . In Ar, values were 3 8

somewhat lower (between 50 and 8 0 % of those in SF ) over the temperature 6

range. We suspect an undetected experimental error in the Ar experiments.

The increased molecular complexity of this reaction (relative to SiH + SiH and 2 4

SiH + Si H ) combined with the probability of alternative decomposition

2 2 6
284

pathways from the energised product (cf SiH + Si H ) explain the lack of a
2 2 6

pressure dependence in this reaction. The temperature dependence gives rise

10 3 1 1
to the average Arrhenius parameters, A = 3.7 χ 1 0 " c m molec" s" and E a
1
= -2.0 kJ mol" , although the Arrhenius plot is slightly curved. These curved
Arrhenius plots for the reactions of SiH with SiH , S i H and S i H are shown
2 4 2 6 3 8

in Figure 10. The magnitude of the rate constants and the almost statistical
relationship between them (Table 10) attests to the high collisional efficiency
of these reactions, and the looseness of the activated complexes.

Figure 10. Arrhenius plot for reactions of SiH with silanes.

2
 285

Table 10
Comparison of rate constants for SiH insertion (per Si-Η bond) 2

A710 1 0
cm molec"V
3 1

Substrate
Τ « 298 Κ Τ « 588 Κ

SiH 4 1.15 0.59

Si H 2 6 1.08 0.72
Si H 3 8 0.99 0.72

(e) SiH + Me SiH>,

2 n n (n = 1-4)

Baggott et al [71] measured the first rate constants for reaction of SiH 2

with the methylsilanes (MeSiH , Me SiH , Me SiH and Me Si). The study was
3 2 2 3 4

carried out at room temperature and the results are shown in Table 1 1 .

Table 11
Rate constants for SiH reactions with the methylsilanes
2

Species */10 10
cm molecV
3 1

MeSiH 3 3.7 ± 0.2

Me SiH
2 2 3.3 + 0.3
Me SiH3 2.5 ± 0.1
Me Si 4 < 0.0027

No pressure dependence was found for SiH + MeSiH and it was assumed 2 3

therefore that the other reactions were pressure independent.

Carpenter and Walsh [77] have extended these studies for MeSiH M e S i H 3 2 2

and Me SiH over a temperature range of 295-625 Κ and a pressure range of 3

3

to 100 Torr (SF ). No pressure dependencies were found except for SiH +
6 2

MeSiH at 600 K. The measured Arrhenius parameters can be seen in Table 12.
3

The mechanism of these reactions is discussed later. The study of SiH + 2

Me Si has been repeated by Becerra and Walsh [78] giving an (apparent) rate
4
13 3 1 1
constant of ( 1 . 7 ± 0 . 2 ) χ 10~ c m molec s" . The rate constant value was
shown to reduce with further purification of the already 9 9 . 9 % pure Me Si. 4

However even 0.1 % of a remaining reactive impurity would still account for this
rate constant. There are reasons for believing that the reaction of SiH with 2

Me Si is very much slower (see below).

4
286

Table 12
Arrhenius parameters for SiH reactions with the methylsilanes
2

Species log(/4/cm molec V )

3 1
£ /kJ mol"
a
1

MeSiH 3 -9.82 -2.39

Me SiH 2 2 -10.00 -3.12
Me SiH 3 -10.14 -3.20

(f) SiH + alkanes

2

Inoue and Suzuki [12] reported the absolute rate constant, k = ( 1 . 0 ± 0 . 3 )

1 2 3 1
χ 10 c m molec'V for the reaction of SiH with C H measured by the LIF 2 4

technique at 1 Torr total pressure. Chu et al [42], using the LRAFKS technique,
determined upper limits for the rate constants of the reactions with C H and 4

C H at 5 Torr total pressure. They found substantially lower values of

2 6
14 14 3 1
( 2 . 5 ± 0 . 5 ) χ 1 0 " and ( 1 . 2 ± 0 . 5 ) χ 10~ c m molec'V respectively. The
authors explained that a 0.01 % impurity in the methane or ethane, reacting on
every collision with SiH , could account for all of the observed silylene removal.
2

This interpretation is also consistent with the observation that ethane appears
to react slightly slower even though more CH bonds are available. Eley et al
[79] found no observable reaction of silylene with methane in their competitive
rate study. These results are consistent with previous indirect evidence [80]
and ab initio calculations [27] all of which conclude that there is a substantial
energy barrier to silylene insertion into C-H bonds in saturated hydrocarbons.
In order to test the reactivity with strained carbon-carbon bonds Chu et al [42]
examined the reaction of silylene with cyclopropane. The attempt was
unsuccessful in the sense that all observed reactivity could be accounted for by
the « 1 % propylene impurity in the commercially available sample of
cyclopropane.

3.2 Insertion reactions of substituted silylenes

This section begins with dimethylsilylene, SiMe . Time-resolved 2

experimental studies on SiMe in the gas-phase began in 1987 with the work
2

of Baggott et al [48]. At the same time Gaspar et al [15] published their

account of solution studies of the direct kinetics of silylenes, including SiMe . 2

Prior to this there had been a good deal of controversy about the absorption
spectrum of SiMe a summary of which is given in Apeloig's article [16]. This
2

was resolved with agreement that the absorption spectrum of SiMe possesses 2

a broad maximum at 4 5 0 nm [81]. The gas-phase spectrum is similar [48] but

with a small red shift. The solution spectrum obtained by Gaspar e f a / [ 1 5 ] was
also similar but it appears probable that it was (at least in part) due to another
species since subsequent experiments by Shizuka et al [82] in solution using a
different source of SiMe give significantly different values for some of the rate
2

constants. These later values were more consistent with those from the gas-
phase, described below. As before absolute measurements replaced earlier
estimates [83], which were much closer for SiMe than for SiH . 2 2
 287

Μ SiMe + Μθ^ίΗ,,
2 (η = 0-4) n

The results obtained by Baggott et al [48] on this reaction system are

shown in Table 13. These were landmark results; the first gas-phase directly
measured rate constants for an alkyl-substituted silylene.

Table 13
Rate constants for SiMe reactions with silane and the methylsilanes
2

1
Substrate k/10 1 3 3
cm molec-V
SiH 4 2.0 ± 0.3
MeSiH 3 19 ± 2
Me SiH 2 2 55 ± 5
Me SiH 3 45 ± 5
Me Si 4 < 0.5

They showed (i) the significant activating effect of methyl substituents in the
substrate silane, and (ii) the significant reduction in silylene reactivity, resulting
from methyl substitution (by comparison of the rate constants with those for
SiH [12,57], subsequently confirmed in similar experiments [71,73]). They
2

reconfirmed the extremely low reactivity of silylenes for both C-H and Si-C
bonds, from the essential lack of reaction with Me Si. Product studies 4

confirmed the reactions were Si-Η bond insertions. Amongst alternative

explanations for the methyl substituent effect was that of changes in Si-H
bond polarity with increasing methyl substitution. This is discussed later.
Further studies by Baggott et al [84] over the temperature range 295-608 Κ
established the Arrhenius parameters shown in Table 14.

Table 14
Arrhenius parameters for SiMe insertion reactions 2

1
Substrate
3 1
logW/cm molec" s" ) 1
£ /kJ mol'
a

SiH 4 -12.54 ± 0.16 + 1.25 ± 1.23

MeSiH 3 -12.90 ± 0.09 -6.73 ± 0.68
Me SiH
2 2 -13.15 ± 0.05 -10.55 ± 0.38
Me SiH3 -13.41 ± 0.14 -11.20 ± 1.03

These results showed that the methyl substituent effects in the substrate,
persist from room temperature to higher temperatures leading to increasingly
negative activation energies with increased methyl substitution. These studies
288

reinforced an emerging characteristic of silylene kinetics, viz. that of decreasing

rates with increasing temperature. This is seen very clearly in the graphical
representation of these results in Figure 1 1 . The magnitudes of the rate
constants show that these SiMe reactions (unlike those of SiH ) do not occur
2 2

close to the collision rate. The explanation for this and the mechanism of these
reactions is discussed later. The kinetic results for SiMe insertion may be
2

combined with those for reverse decomposition of the (product) disilanes to

give AH ° (SiMe ). This is another story of twists and turns, with the key items
f 2

of uncertainty having been, until recently, the enthalpies of formation of the

methylated disilanes. This appears to have been settled by a recent solution
calorimetric study by Pilcher et al [23] leading to LH °(SMe )
f 2 = 140 ± 6 kJ
1
mol" in reasonable agreement with other values [85]. This corresponds to a
1
DSSE value of 121 ± 12 kJ mol" (see Table 2). Readers are referred to original
papers [39,82] for discussion of the solution phase studies of SiMe . 2

-11.0 Ί

-if.u 1 , 1 , 1
250 350 450 550 650
T/K

Figure 1 1 . Temperature dependence of rate constants (per Si-Η bond) for

reactions of SiMe with methylsilanes.
2

(b) MeSiH + Me^iH^ (n = 0-3)

n

The first, and at the present time only, direct kinetic study of MeSiH has
been carried out by Becerra et al [49]. Following after earlier studies with SiH 2
 289

[71] and SiMe [48] this was limited to SiH , MeSiH , M e S i H and Me SiH,
2 4 3 2 2 3

since Me Si appeared unreactive to the other silylenes. Products were

4

confirmed by gc analysis, and reactions checked for the absence of total

pressure effects (in both Ar and SF ). These studies were more difficult than
6

those of SiH and SiMe , because of the combination of low sensitivity of

2 2

detection for MeSiH, combined with its high reactivity. This gives rise to more
scatter in the individual rate constants and consequently a more erratic
temperature dependence, as shown in the Arrhenius parameters of Table 15.
Because of the scatter the authors also give adjusted E values based on an a
1 2 4 3 1 1
assumed A factor (per Si-Η bond) of 10 c m molec" s" . These values are
shown in parentheses.

Table 15
Arrhenius parameters for MeSiH insertion reactions
3 1 1
Substrate logW/cm molec" s" ) ^ / k J mol
SiH 4 -11.4 ± 0.2 (-11.8) -7.5 ± 1.7 (-10.9)
MeSiH 3 -12.1 ± 0.3 (-11.9) -14.5 ± 2.7 (-12.5)
Me SiH
2 2 -11.8 ± 0.2 (-12.1) -11.5 ± 2.1 (-14.3)
Me SiH 3 -12.8 ± 0.1 (-12.4) -18.4 ± 0.9 (-14.9)

The trends with substrate methyl substitution are, however, the same as
for the insertion reactions of SiMe [84]. Negative activation energies are again
2

found in these reactions. The rate constants are much closer in magnitude to
those of SiH than to those of SiMe . The mechanism of these reactions is
2 2

discussed later.
The kinetic results for MeSiH insertion have again been combined with
those of the reverse (decomposition) reactions by Becerra et al [49] to give
1
A^°(MeSiH) = 2 0 2 ± 6 kJ mol" in agreement with other values [85].

(c) PhSiH + Me SiH (n=0-3) n 1 n

The first, and again only, study of PhSiH has been carried out by Blitz et al
[50]. The general findings are very similar to those for MeSiH. PhSiH is close
in reactivity to MeSiH. It is worth mentioning that the original spectrum
attributed to PhSiH between 540 and 600 nm is probably due to SiH (the 2

precursor or source, PhSiH , is common to both). However since PhSiH was

3

monitored at 515 nm (where SiH does not absorb) for the time-resolved
2

studies the reported kinetics are not affected. The results in terms of Arrhenius
parameters are given in Table 16.
Because of the scatter the authors also give adjusted E values based on an a
1 2 0 3 1 1
assumed A factor (per Si-Η bond) of 1 0 " c m molec" s" . These values are
shown in parentheses. The trends are similar to those for MeSiH [49]. The
mechanism of these reactions is discussed later.
290

Table 16
Arrhenius parameters for PhSiH insertion reactions
1
Substrate
3
logW/cm molec" s" ) 1 1
E /kJ m o l
a

SiH 4
-11.54 ± 0.50 (-11.4) -7.1 ± 4.1 (-6.0)
MeSiH 3
-11.60 ± 0.69 (-11.5) -12.5 ± 5.7 (-12.0)
Me SiH
2 2
-11.31 ± 0.16 (-11.7) -9.3 ± 1.2 (-12.6)
Me SiH3
-12.14 ± 0.28 (-12.0) -15.6 ± 2.3 (-15.0)

(d) CISiH. SiCU. SiF 2

Ho et al [52], using LIF to monitor CISiH, have obtained the rate constants
listed in the Table 17 at total pressures of 500 Torr (N ). 2

Table 17
Rate constants for CISiH insertion reactions
13 3 1
Substrate */10" cm molec"V
H 2
< .003
SiH 4
0.053 ± 0.001
CI SiH
2 2
4.1 ± 0.3

These values are considerably less than those of MeSiH of PhSiH thus
indicating the substantial deactivating effect of the chlorine substituent in the
silylene. It is worth noting, however, that like methyl-substitution, chlorine-
substitution in the silane is considerably activating.
There are no absolute rate constants for insertion reactions of SiCI 2

although Safarik et al [86] did explore its reaction with i - C H , for which 4 10

reaction is either very slow or non-existent. No Si-Η insertion process has been
studied.
Freedman et al [87] found no reaction of SiF with H even up to 1400 K, 2 2
15 3 1 1
setting an upper limit of 10" c m molec" s" for the rate constant even at the
highest temperature. This is consistent with ab initio calculations of Sosa and
Schlegel [88], which predict a large increase in activation barrier from the
reaction of SiH + H to that of SiF + H . 2 2 2 2

Relative rate studies [89] for silylene and halogenated silylenes have
established the following relative order of reactivity for insertion into SiH bonds:
SiH > CISiH > FSiH > > SiCI , SiF .
2 2 2

3.3 General discussion of the Si-Η insertion process

Although direct silylene kinetics is still a relatively new field, the published
studies on the Si-Η insertion process have already supplied a considerable
 291

amount of detail. Substituent effects have been probed and activation energies
obtained. These have in turn provided insight into the nature of the process and
its detailed mechanism which are discussed in more detail here.

(a) Comparison of substituent effects

The differing reactivities of SiH , SiMe , MeSiH and PhSiH are brought out
2 2

clearly by the comparisons in Tables 18 and 19. The two temperatures

selected, 298 Κ and 600 K, represent approximately the two extremes of the
experimental range of study.

Table 18
1 3 1 1
Comparison of insertion rate constants (10" ° cm molec" s" ) at 298 Κ
silylene
silane SiH 2 SiMe 2 MeSiH PhSiH

SiH 4 4.0 0.0020 0.81 0.51

MeSiH 3 3.7 0.019 2.55 3.9
Me SiH
2 2 3.3 0.055 1.69 2.1
Me SiH3 2.5 0.045 2.78 3.9

Table 19
1
Comparison of insertion rate constants (10" ° cm molec" s" ) at 600 Κ
1 3 1

silylene
silane SiH 2 SiMe 2 MeSiH PhSiH

SiH 4 2.2 0.0024 0.179 0.12

MeSiH 3 2.4 0.0046 0.133 0.31
Me SiH
2 2 1.9 0.0060 0.164 0.32
Me SiH3 1.4 0.0045 0.060 0.165

The tables bring out a number of factors. The silylene reactivity sequence
is: SiH > PhSiH ~ MeSiH > SiMe . However replacing Me-for-H in the
2 2

silylene does not produce a constant effect. The differences in reactivity

between SiH and MeSiH are much less than those between MeSiH and SiMe ,
2 2

although the relative rate factors get closer as temperature increases, because
of the sharper drop in values for MeSiH than for SiMe . The PhSiH is slightly 2

more reactive than MeSiH except with SiH itself. The substrate reactivity
4

sequence is less obvious, but on a per Si-Η basis and allowing for a slight
scatter, is: Me SiH > M e S i H > MeSiH > SiH for all silylenes at both
3 2 2 3 4
292

temperatures. Table 10 shows that on a per Si-Η basis SiH , S i H and S i H 4 2 6 3 8

have almost identical reactivities for SiH . This, incidentally, tends to argue
2

against any pathway involving Si-Si insertion, which has been ruled out in other
cases [90].

(b) Nature of the Si-Η insertion process

This process, as pointed out in the introduction, can be discussed in terms
of the coupling of electrophilic and nucleophilic interactions involved [25-27].
In the general reaction:

SiX + HSiY
2 3 HX SiSiY2 3

three possibilities have been considered [84], viz.

(i) a fully synchronous process in which both electrophilic, e, and nucleophilic,
n, stages occur simultaneously;
(ii) a process led by the electrophilic, e, interaction viz.

X X
Η

OSiO Η- -Si V / 'Si- -Si

Si - - H - Si
i
Π γ
Υ

(iii) a process led by the nucleophilic, n, interaction viz.

Γ\ Η Η Η Γ
Χ \I ^ ^ S .Υ Χ •·..)κΛ ( Υ \ /
X ^ S ' ® S«C 3 ^ S C — - S i — S ^ Y

ϋ Χ ^ ϋ /
χ
x
Γ / ^ Y γ
7 \
Γ Χ
Γ

In processes (ii) and (iii) distinct but different intermediate complexes are
involved. The clue about which is the best description comes from calculations
by Gordon [91] referred to in our recent discussion of this question [49]. The
geometry of the intermediate complex in the SiH + SiH reaction (Figure 9) 2 4

shows almost complete transfer of Η from SiH to SiH , thus favouring (ii) with
4 2

the electrophilic interaction leading. Figure 9 suggests some degree of Si-Si

bond formation. However the looseness of structure (high entropy requirement
 293

of the transition state) argues against the completely synchronous process.

This is supported by more recent theoretical calculations by Sakai and
Nakamura [92] suggesting a hydride anion transfer almost corresponding to ion
pair formation. This is not dissimilar to Hoffmann's finding [93], that in the C H2

+ C H reaction the initial approach is abstraction-like. This harks back to an old

4

suggestion of Benson [94] about radical pair formation in the C H + C H 2 4

reaction and the involvement of ionic states. The Η-bridged structure for the
process (ii) is supported by the calculations of Trinquier [95] who has shown
that the structures of the type H M . . H . . M H become increasingly stable as Μ
2 3

changes from C to Pb down group 14 of the periodic table. The methyl group
substituent effects can now be understood in terms of process (ii) although on
their own, the effects do not provide an unambiguous interpretation.
Since carbon is more electronegative than silicon it is plausible that methyl
substituents can function as weakly electron-withdrawing groups. They tend
to draw off negative charge inductively from silicon centres. Silylenes are so
electrophilic that we must presume they have little difficulty in accepting an
electron pair regardless of substituent (X = Η or Me). The donor character of
methylsilanes (Y = Me) is also probably not affected significantly by methyl
groups (Si-Η bond dissociation energies are virtually constant in the
methylsilanes and independent of methyl group substitution [96,97]). It is the
second, nucleophilic, stage where the methyl groups exert their influence. If
the methyl groups are present in the silylene (X = Me) then it will be more
reluctant to donate its lone pair. This suggests the second step should be
slower for SiMe than SiH . If the methyl groups are present on the silane (Y
2 2

= Me) then they will help draw off charge and facilitate acceptance of the lone
pair. Thus if the first (e) stage is rate determining, the Υ Ξ Me substituent
effect will be negligible. This closely corresponds to the SiH insertion. If, on
2

the other hand, the second (n) stage is rate determining then the Y = Me
substituent effects will become significant. This corresponds to the SiMe 2

insertion. With MeSiH the situation is in between, suggesting a possible switch

with the e stage rate determining at lower temperatures and η stage at higher
temperatures.
The effects of Ph-for-Me substitution on the silylene are very small and so
are compatible with this explanation of the insertion process. The phenyl
group, in spite of its increased size and different electronic character from
methyl, clearly plays no special role. There appears to be no special interaction
(#T-type or otherwise) with the empty or filled silylene orbitals. This is explained
by the poor overlap between Si(3p) orbitals and C(2p) orbitals and is supported
by the observation of an insignificant energetic interaction in the phenylsilyl
radical [98].
The effects of CI-for-Me substitution on the silylene are large. CI may
deactivate the silylene in two ways. Back donation of a chlorine atom lone pair
of electrons into the Si empty ρ orbital (known to stabilise the silylene [16], as
shown in Figure 2) will reduce the electrophilic character and may slow the e
stage. The electronegativity of CI will cause orbital contraction of the silylene,
making it a less effective donor and thus slowing the η stage.
294

(c) Intermediate complex mechanism

The negative activation energies of these reactions, in some cases quite
large, and the probable two stage character of this reaction, discussed above,
have led us to consider a general kinetic treatment of a two stage process
involving an intermediate complex. This has been outlined for both SiMe [84] 2

and MeSiH [49]. Part of this treatment for the latter case is reproduced here.
The insertion process is written as follows:

1 2
MeSiH + H-SiR «=* MeSiH—HSiR MeH SiSiR
3 3 2 3

-1

where R = Η and/or Me.

Step (1) may be viewed as a loose association of the methylsilylene with
the substrate to form the Η-bridged complex. Step (2) involves the
reorganization of the intermediate complex, requiring restriction of its motion
via a much tighter transition state in which the two Si atoms approach whilst
the Η still bridges before completing its migration to give the product disilane.
The observation of single-exponential decays supports the assumption of a
steady-state population of intermediate complexes. The phenomenological
second-order rate constant, k, is given by:

k = k^k l(k 2 A + k ) = * / ( 1 + *-i/*2>

2 Ί

Since step (1) is a loose association process the authors assume that it will
have at most a weak temperature dependence, i.e. £ is approximately zero. 7

This suggests the major temperature dependence is associated with kAlk2. For
MeSiH the observed rate constants, k, are sufficiently high, that they approach
the collisional rate especially at the lower temperatures. This implies that at
sufficiently low temperatures k « k and k lk is small. Then as the
y A 2

temperature increases k /k becomes more significant. The mechanism

A 2

switches from one in which step (1) is mainly rate determining to one in which
step (2) becomes rate determining. The same analysis of SiMe data [84] 2

suggested step (2) was rate determining over the whole temperature range of
study (300-600 K).
In order to carry out the analysis the authors made the assumption that
k = L χ 10 c m molec' s" (where L is the path degeneracy). The values
}
1 0 3 1 1

for k lk were then calculated from the observed values for k. For the MeSiH
A 2

insertion reactions the following general expression was found:

k /k
A 2 « 10 2 6
exp((£ - £,)//? 7)
2

The derived values for E - E are shown in Table 20 as well as those derived
2 A

for SiMe . 2
 295

Table 20
Activation energy differences for rearrangement (£ ) and redissociation (£ ) of
2 r

intermediate complexes in the Si-Η insertion reactions of MeSiH and SiMe . 2

1
(E - £.,) / kJ m o l
2

Substrate MeSiH SiMe 2

SiH 4 -12.6 -0.3

MeSiH 3 -14.7 -6.0
Me SiH 2 2 -16.7 -9.4
Me SiH 3 -17.3 -10.5

These values represent the lowering of the secondary barrier (step (2))
compared with that for redissociation of the intermediate complex (step (-1)).
2 6
This can be seen in Figure 12. The A factor ratio, A_ IA = 1 0 , gives 1 2

support to the view that the first transition state (TS1) is significantly looser in
structure than the second (TS2). Figure 12 shows two representative
(qualitative) potential energy surfaces with an indication of the positions of the
two transition states involved in the process.
The differences between the Si-Η insertion reactions of MeSiH and SiMe 2

over the temperature range 300 to 600 Κ may be rationalised. For both
silylenes their kinetic behaviour is determined by the relative tendencies of the
intermediate complexes to redissociate (k_ ) or rearrange (k ). }For all the 2

insertion reactions of SiMe redissociation of the complexes is more favourable

2

than rearrangement, because the large entropic advantages outweigh the small
energetic disadvantages. For the insertion reactions of MeSiH, the
redissociation and rearrangement of the complexes are more closely competitive
with one another. This is because the entropic advantage of redissociation is
partially offset by the energetic disadvantage. The effect of energy is to favour
rearrangement at lower temperatures. It also favours rearrangement of the
complexes formed with the higher methyl-substituted silanes.
An examination of the figures of Table 20 shows that for a given substrate
molecule \E - £ | is larger for reactions of MeSiH than for those of SiMe .
2 ; 2

The values for | £ - £ | for reactions of SiH cannot be derived from

2 ; 2

experiment because the reactions are so fast that their rate constants only
show slight decreases over the temperature range 300-600 K. Effectively this
means that step (1) is rate controlling under all conditions. This must be
because £ is small compared with £ . In other words, even though they
2 ;

cannot be measured, | £ - £ | values are large for SiH , consistent with trends
2 ; 2

of Table 20. This is supported by the ab initio calculations for SiH + SiH [91]. 2 4

The increases of | £ - £ | with methyl substitution on the silane, could reflect

2 ;

either an increasing stabilisation of the intermediate complex (increasing £ ) or ;

Figure 12. Schematic potential energy surface for the Si-Η insertion process of
MeSiH. Solid line applies to the more favourable reaction (with Me SiH); 3

dashed line applies to less favourable reaction (with SiH ).

4

a decreasing stabilisation of the transition state, TS2, for rearrangement

(decreasing E ). It is not obvious which of these is more likely. However the
2

well depths associated with these complexes cannot be too great or non-
stationary state kinetic effects would be observed. An estimate of < 41 kJ
1
mol' has been made [84] for the well depth of the complex formed from SiMe 2

+ HSiMe .3

Baggott et al [84] have noted that other theories such as that developed by
Houk et al [99] based on the idea of entropy control are not inconsistent with
the kinetic data on silylene insertion reactions. The secondary barrier (TS2) in
the intermediate complex model represents an entropy bottleneck. Baggott et
 297

al have discussed in detail possible differences between these theories and tests
to distinguish them.

4. THE 77-TYPE ADDITION PROCESS

After the Si-Η bond insertion reaction this is probably the most widely
studied process. One problem of this reaction is that the initial products of
addition are the strained ring siliranes (addition to alkenes) [100] and silirenes
(addition to alkynes) [101] which are relatively unstable particularly in the gas
phase. This instability is a combination of handling (analytical) difficulty and
thermal instability. Although this may seem a disadvantage, study of the
reactions by direct monitoring of the silylene decay not only avoids this
difficulty but in some cases provides information on the thermal stability of the
products. This section covers the addition reaction to alkenes, alkynes and
dienes, including, in the cases of ethene and ethyne, their isotopic variants
C D and C D .
2 4 2 2

4 . 1 . SiH addition reactions

2

(a) SiH + C o h y C f o
2

Inoue and Suzuki [12] measured the first absolute rate constant for SiH 2

with C H at room temperature, using the LIF technique. They found a value
2 4
11 3 1 1
of k = (9.7 ± 1.2) χ 10~ c m molec s at 1 Torr total pressure in He. Using
the LRAFKS technique, Chu et al [42] obtained a value of ( 2 . 7 ± 0 . 3 ) χ 10 1 1

3 1 1
c m molec" s" under the same conditions, and additionally observed a pressure
dependence. Al Rubaiey et al [102] have repeated these measurements in Ar
and SF bath gases. The values are compared in Table 2 1 .
6

Table 21
Rate constants for SiH + C H in the presence of different bath gases
2 2 4

k I 10" c m m o i e c - V
10 3 1

P / Torr He Ar° SF 6
C

1 0.97 a
0.39 0.67
b
0.27
5 0.55 b
0.79 1.30
10 0.80 b
1.30 1.58

a. Ref. [12] b. Ref. [42] c. Ref. [102]

Al-Rubaiey ef a/[102] have extended their measurements over the pressure

range 1-50 Torr and shown that, apart from the Inoue and Suzuki value, the
results can be explained in terms of a third body assisted association process
with differing collisional efficiencies for He, Ar and SF , viz: 6
298

CH * 2

SiH + H C = C H
2 2 2 τ± H SiC"
2 I
CH 2

i +M

^CH 2

H SiC
2 I
XH 2

Additionally Al Rubaiey et al [102] have shown that the pressure dependence

can be modelled via an RRKM calculation. To obtain a good fit to the data they
required a value for the activation energy for silirane decomposition of ca 197
1
kJ mol" consistent with theoretical calculations [103,104], but in disagreement
1
with a measured value of 130 kJ mol" for the analogous hexamethylsilirane
decomposition [105] (the only silirane decomposition for which at present there
is an experimental activation energy). In a more extensive study Al-Rubaiey and
Walsh [37] have studied the reaction over the pressure range 1-100 Torr (in
SF ) and at five temperatures in the range 298-595 K. Some of the rate
6

constants are shown in Table 22.

The infinite pressure values were found by extrapolation of the pressure
dependent curves with the assistence of RRKM modelling. The temperature
3
dependent rate constants correspond to Arrhenius parameters of logW^/cm
1 1 1
molec" s" ) = -9.97 and E °° = -2.9 kJ mol' . These indicate a fast
a

collisionally-controlled association process.

Table 22
Rate constants for SiH -I- C H at two pressures and five temperatures
2 2 4

k I 10 1 0 3
cm molec" s" 1 1

T/K Ρ= 10 Torr Ρ = oo a

298 1.58 ± 0.03 3.5 ± 1.2

355 1.15 ± 0.02 3.0 ± 1.2
415 0.731 ± 0.017 2.5 ± 1.0
515 0.357 ± 0.005 2.2 ± 0.9
595 0.148 ± 0.007 1.9 ± 0.9

a. Values obtained by extrapolation.

 299

In a disagreement over product formation Al-Rubaiey and Walsh [37]

searched, using GC analysis, but found no evidence for vinylsilane formation in
this reaction, as claimed by Fisher and Lampe [106]. Other analytical evidence
[62] is against vinylsilane formation and Al-Rubaiey and Walsh have suggested
that Fisher and Lampe may have seen the mass spectrum of silirane. The latter
have recently measured relative rate constants for reaction of SiH with C H
2 2 4

and with SiH . It seems [37] as if these are not in good agreement with
4

absolute values. This may be the result of non-thermalised SiH produced by

2

the IR multiphoton method. Ab initio theoretical calculations by Anwari and

Gordon [107] of the reaction of SiH with C H indicate a reaction with no
2 2 4

energy barrier, consistent with the high rate constants. In similar fashion to the
insertion reaction (see sections 1 and 3) the reaction may be described as
occuring in two stages viz. an initial π attack ("electrophilic stage" - donation
of C = C /7 electrons into the Si 3p orbital), follow by σ attack ("nucleophilic
stage" - donation of the silicon lone pair electrons into the C = C antibonding π*
orbital). Figure 13 shows these processes and gives an approximate indication
of the geometry of the transition state.

Figure 13. Orbital interactions involved during the process of ,τ-addition of SiH 2

to C H
2 4

An unpublished study of the reaction of SiH + C D by Al-Rubaiey and

2 2 4

Walsh [108] shows that there is a significant and pressure dependent isotope
effect which is more important at low than high pressures as shown in Figure
14.
300

Figure 14. Comparison of pressure dependent rate constants for reactions of

SiH with C H and C D at 298 K.
2 2 4 2 4

This has been investigated at five temperatures in the range 291-595 K.

The high pressure limiting rate constants are, within experimental error, the
same as those for SiH + C H , but RRKM fitting of the fall-off curves on the
2 2 4

assumption of involvement of 2,2,3,3-tetradeuterosilirane alone did not fit the

data. At low pressures the reaction was too fast. This suggests the
involvement of a rapid isotopic scrambling mechanism as occurs for the SiH 2

+ C D reaction [109,110] (see subsection (d)). A mechanism for this, based

2 2

on higher temperature studies [62,111] plausibly involves the intermediacy of

ethylsilylene-d , viz:4

H CD * N 2

SiH + C D 2 τ±2 Si
4 | χ

W' CD \ ^ 2

HSiCD CD H
2 2

SiHD + C H D 2 3 «- Si I
y X
D CHD \

DSiCHDCD H 2

D x C H D * ^ ^
SiD + C H D
2 2 2 2 «- Si^ I
CHD
 301

The propensity for such rapid processes has been documented by Davidson
[112] and the energetic considerations are discussed later.

(b) SiH + C H 2 3 6

The only published rate constant for this reaction is a value of ( 1 . 2 ± 0 . 1 )

χ 10 c m molec s" obtained by Chu et al [42] at room temperature and 5
1 0 3 1 1

Torr (He). In an unpublished study in our laboratories, Al-Rubaiey [113] has

found the reaction to be weakly pressure dependent (1-100 Torr, SF ) in the 6

temperature range 294-520 K. At 5 Torr (SF ) a rate constant of ca 2.0 χ 10 6

1 0

3 1 1
c m molec" s" can be interpolated in reasonable consistency with value in He
[42], allowing for different collision efficiencies. The high pressure limiting rate
constants give the Arrhenius parameters \og(A /cm molec" s" ) = -10.38 and 00 3 1 1

Ea°° = -5.1 kJ mol" . These rate constants are ca 20% smaller than those for
1

SiH + C H , although because of the extrapolation they could be equal (within

2 2 4

error limits). Thus the methyl substituent effect on this reaction is slightly
negative. This contrasts with the analogous results for SiMe [114] (discussed 2

in section 4.2). RRKM modelling of the pressure dependence can be made to

fit provided the activation energy for decomposition of the product, 2-
methylsilirane (back to SiH + C H ) is ca 160 kJ mol" . This is significantly 2 3 6
1

less than the activation energy for decomposition of silirane itself (obtained
from the SiH + C H system). Unless there is an undiscovered source of error
2 2 4

in this study, this means that there is a surprising methyl substituent effect on
silirane ring stability (and presumably strain energy). This finding, however, is
consistent with the difference already noted, with hexamethylsilirane [105].

(c) SiH + C , H 2 6

The only published rate constant for this reaction has a value of ( 1 . 9 ± 0 . 2 )
χ 10" c m molec s "\ obtained by Chu et al [42] at room temperature and
10 3 1

5 Torr (He). This is close to the collision number and consistent with values for
the reactions of SiH with C H and C H . Pressure and temperature variation
2 2 4 3 6

studies have not yet been carried out. This experimental value is ca 1 0 greater 5

than that proposed by Rogers et al [115] derived from relative rate studies. It
is also not very consistent with relative rate studies of Gaspar et al [116], who
found that SiH apparently reacts 1 2 ± 4 times faster with SiH than C H (at
2 4 4 6

298 K) from chemical product analysis (and 9 ± 1 from nuclear recoil

measurements [117]). It is possible that the high values for these ratios arise
because of incomplete product recovery from the SiH + C H reaction (for 2 4 6

which only 4-silacyclopentene, the rearranged product of 2-vinylsilirane, has

ever been detected).

(d) SiH + C H / C D 2 2 2 2 2

Chu et al [42] obtained a value of (9.8 ±1.2) χ 10 c m molec" s" for 1 1 3 1 1

SiH + C H at 298 Κ and 5 Torr (He). Becerra et al [109,110] have made a

2 2 2

comprehensive study of this reaction as well as its isotopic variant, SiH + 2

C D , over the pressure range 1-100 Torr (SF ) and temperature range 291-
2 2 6

613 K. A selection of rate constants is shown in Table 23.

302

Table 23
1 3 1 1
Rate constants (10" ° cm molec" s" ) for SiH 2 + C H /C D
2 2 2 2 at 10 Torr total
pressure (SF ) 6

77K C H 2 2 T/K C D 2 2

291 3.21 ± 0.33 291 3.74 ± 0.16

346 2.56 ± 0.04 346 2.94 ± 0.07
399 1.99 ± 0.05 395 2.63 ± 0.06
483 1.26 ± 0.04 481 2.02 ± 0.08
613 0.61 ± 0.05 613 1.44 ± 0.06

The room temperature value for SiH + C H is consistent with that of Chu et
2 2 2

al [42] taking into account pressure and bath gas differences. Just as for the
SiH + C H reaction, the rate constants show a negative temperature
2 2 4

dependence and an isotope effect favouring SiH + C D . The rate constants 2 2 2

also show a pressure dependence as illustrated for SiH + C H in Figure 15. 2 2 2

log([SF ]/Torr)
e

Figure 15. Pressure and temperature dependence of rate constants for SiH + 2

C H
2 2 <
 303

Extrapolation to infinite pressure gives rate constants corresponding to the

3 1 1
Arrhenius parameters of log(/4°7cm molec" s" ) = -9.99 and E °° = -3.3 kJ a
1
mol" . The curves for SiH + C D converge to the same limit within
2 2 2

experimental error. These are again consistent with a fast collision-controlled

association process, which is third-body-assisted at lower pressures. These
studies bring together a total of four direct investigations of SiH with PhSiH 2 3

[71], Me SiH [71], C H [37] and C H [110] which have been shown [37] to
3 2 4 2 2

be consistent with relative rate measurements carried out in the Reading

laboratories by Eley et al [79] prior to the first absolute rate measurements.
Other estimates of the rate constants for SiH + C H based on relative rate 2 2 2

constants at higher temperatures [62] are significantly in error.

The direct experimental results are supported by the ab initio calculation of
Boatz et al [118] which indicates a barrierless reaction. The mechanism of
addition is similar to that for SiH + C H and is further discussed in section
2 2 4

4.2.
RRKM theoretical attempts to fit the pressure dependence of SiH + C H , 2 2 2

in contrast to the SiH 4- C H reaction, were only successful with an

2 2 4

unrealistically high activation energy for decomposition of the supposed silirene

product [109]. This led to the suggestion that another product viz.
ethynylsilane (silylacetylene) was probably formed according the scheme:

CH*
SiH 2 + HC = CH H Si ^ IIII
2
HoSiC = CH
CH

φ +M

.CH
H Si^
2

CH

Subsequent experiments [119] have confirmed the formation of

ethynylsilane. Modelling of this system [110] gave a value for the activation
1
energy for the isomerisation of silirene to ethynylsilane of ca 156 kJ mol" ,
assuming a fairly tight transition state. The isotope effect points to a similar
isotopic scrambling mechanism for SiH + C D as observed for SiH + C D . 2 2 2 2 2 4

Becerra and Walsh [110] have proposed a mechanism involving vinylsilylene-d 2

viz:
304

Η CD*
SiH + C D
2 2 2

HSiCD = CDH

SiHD + C HD 2

DSiCH = CDH

SiD + C H
2 2 2

This reaction has been incorporated into the model to fit the pressure
dependence of SiH + C D , from which an approximate fit gives an activation
2 2 2

energy of ca 151 kJ mol" for vinylsilylene-d formation from silirene-d . This

1
2 2

complex mechanism is very similar to that for the SiH + C H reaction system
2 2 2

suggested by Ring, O'Neal and coworkers [111,115] to operate at the higher

temperatures of pyrolysis studies. It appears to occur as low as room
temperature.

4.2 SiMe addition reactions

2

Room temperature absolute rate constants for addition of SiMe to alkenes, 2

dienes and alkynes have been obtained by Baggott et al [114]. These are
shown in Table 24.
These rate constants were not thought to be pressure dependent although
the test was only for SiMe + C H and for a limited pressure range of 5-20
2 2 4

Torr (SF ). There are no other gas phase data for comparison but rate
6

constants of similar magnitude to those of Table 24 for other alkenes, dienes

and alkynes have been obtained by Levin et al [39] in cyclohexane solution
(using dodecamethylcyclohexasilaneas precursor). The solution data of Gaspar
et al [15] (already mentioned) give apparent rate constants for reaction of SiMe 2

with dienes in cyclohexane (using (PhMe Si) SiMe as precursor) which are at
2 2 2

least an order of magnitude lower than those of Levin et al [39] and probably
should be attributed (as discussed) to another intermediate.
In none of the time-resolved studies did the authors report product detection
but since these should have been siliranes or silirenes this is not too surprising.
Baggott et al [114] failed to find identifiable products using gas
chromatography. Even in the case of SiMe with buta-1,3-diene, the known
2

high temperature product 3,3-dimethyl-3-silacyclopentene was not found. They

suggested that the initial vinylsilirane product does not rearrange at room
temperature. The mechanism of reaction of SiMe with dienes has been 2

discussed in several papers [112,120,121]]. Baggott et al [114] have shown

 305

that the rate constants of Table 24 correlate with substrate ionisation energies,
as shown in Figure 16.

Table 24
Rate constants for addition reactions of SiMe at 298 Κ [114] 2

Reactant k I 10" 11 3
cm molec V 1

CH2 CH2 = =
2.21 ± 0.12
CH CH=CH2 3 3.72 ± 0.27
(C 143)20=0 Η 2 6.82 ± 0.18
t-CH CH=CHCH 3 3 4.43 ± 0.19
CH CH=C(CH )
3 3 2 5.46 ± 0.22
(CH ) C=C(CH )
3 2 3 2 4.89 ± 0.22
CICH=CH 2 1.25 ± 0.04
FCH=CH 2 0.88 ± 0.05
t-FCH=CHF 0.072 ± 0.009
CH =CHCH=CH
2 2
7.45 ± 0.45
CH = CH 4.63 ± 0.22
CH^CCH 3 9.37 ± 0.39
CH C = C C H 3 3 17.0 ± 0.9

-9.5 Ί

w -10.0 A
=<

8.0 11.0

IP/eV

Figure 16. Correlations between SiMe addition rate constants and ionisation
2

energies of alkenes/alkynes.
306

The data for alkenes and alkynes fall on different but parallel lines. The
more substituted alkenes fall below the appropriate correlation line and the
authors suggested that the explanation for this might be a steric effect
associated with crowding between the methyl groups in SiMe and those of the
2

alkene. Figure 17 shows, therefore, the preferred approach geometry in the

SiMe + i-C H reaction. This also provides indirect evidence for the non-least
2 4 8

motion approach pathway indicated by theoretical calculations [107].

Figure 17. Orbital interactions and preferred geometry of approach during the
process of /7-addition of SiMe to isobutene.
2

Temperature variation studies have been undertaken on a selected number

of alkenes/alkynes by Blitz [122]. These results show that rate constants
decrease sharply (by factors of between 3.5 and 15 over the temperature range
298 to ca 580 K) for the five compounds studied ( C H , C H , i-C H , trans-
2 4 3 6 4 8

C H F and C H C ^ C C H ) . Arrhenius plots of the data are slightly curved.

2 2 2 3 3

These results are not analysed in detail here but appear to suggest the possible
involvement of intermediate complexes in this reaction, although the only
theoretical evidence for this seems to be in case of alkynes [118].
 307

4.3 SiCI addition reactions

2

After the discovery of the UV absorption spectrum of SiCI [43], Safarik et 2

al [86] measured the absolute rate constants for the reactions of SiCI with a 2

series of unsaturated hydrocarbons using the flash photolysis-kinetic absorption

spectroscopy technique. The results are shown in Table 25.
From this table we can see the rate constant values increasing with the
substitution on both double and triple bonds. This suggests again that the
electrophilic interaction is dominant in influencing the rate of reaction even
though SiCI is a less reactive silylene than either SiH or SiMe . No products
2 2 2

were positively identified although mass spectral evidence [43] pointed to

formation of a compound thought to have arisen by rearrangement of the
initially formed dichlorosilirane (in the reaction of SiCI -I- i-C H ). 2 4 8

Table 25
Rate constants for SiCI addition reactions at 298 Κ [86]
2

Reactant k I 10" 13 3
cm molec" s 1

C H
2 2 0.71 ± 0.18
C H
2 4 1.29 ± 0.33
C H
3 6 3.82 ± 0.10
t-C H 4 8 5.15 ± 0.13
1-C H 4 6 13.1 ± 0.31

4.4 General discussion of the rr-type addition process

The kinetic (and theoretical) data clearly support a process involving an
initial electrophilic interaction, as already indicated. The question of whether
intermediate complexes are involved is an open one and needs further
investigation. However, the mechanism of addition can be usefully illuminated
first by comparisons of silylene reactivity and secondly by discussion of the
energetics of the prototype silylene addition reactions.

(a) Comparisons of silylene reactivity

The growing data base of absolute rate constants collected in this review
enable us to compare silylene reactivities both with one another and with
methylene. The values of rate constants for their reactions with selected
unsaturated hydrocarbons are shown in Table 26.
1
It should be noted that the data for C H ( A ) have been measured and
2 1

corrected for non-reactive collisionally-induced intersystem crossing in the same

studies and also that the SiH data represent the limiting high pressure values.
2

So far as is known the reactions of C H , SiMe and SiCI are pressure

2 2 2

independent processes. The rate constants for C H and SiH are close to the
2 2
308

Table 26
Comparison of rate constants for addition reactions of methylene and several
silylenes OA, states)
1 2
*/ ισ 3
cm molec~ s" 1 1

Species 1
CH 2 SiH 2 SiMe 2 SiCI 2

C2H2 280 a
400 47 0.071
C H
2 4 190 b
350 22 0.130

C H
3 6
250 b
340 39 0.38
i-C H - -
b
4 8 250 78
1,3-C H -
b
4 6 260 190 75

a. Ref. [123] b. Ref. [35]

collisional maximum. SiH is about twice as reactive as C H in accordance with

2 2

the simple idea of an electrophilic reaction in which the initial interaction is

dominated by the size of the receptor orbital (3p for SiH , 2p for C H ) . Me and 2 2

CI substitution on the silylene deactivate it, just as was seen for the insertion
reaction. The reasons for this, viz., orbital contraction by electronegative
substituents, and back donation from the CI lone pair into the empty Si 3p
orbital are almost certainly the same.

(b) Energy considerations

The mechanisms of the prototype 77-addition processes are examined here
in terms of energy surfaces calculated by Al-Rubaiey and Walsh [37] (SiH + 2

C H ) and Becerra and Walsh [110] (SiH + C H ) . These are shown in Figures
2 4 2 2 2

18 and 19.
Figure 18 shows that silirane formed from SiH + C H can either revert to 2 2 4

SiH + C H or rearrange to vinylsilane (and other isomers) with approximately

2 2 4

equal energies. Since Al-Rubaiey and Walsh [37] found no evidence for
isomerisation but only for reversion (inferred from the pressure dependence of
the kinetics), they suggested that A factors (or relative looseness of ^transition
states) was the explanation for this. The energy of ethylsilylene shows that it
is a plausible species to explain the isotopic scrambling mechanism suggested
[108] i n t h e S i H + C D studies. An estimate has been made of the activation
2 2 4

barrier to its formation from silirane in higher temperature studies [124]. The
value is dependent on the ring strain in silirane but in any case is fairly low.
What is clear is that, because ethylsilylene is endothermic relative to silirane,
if ethylsilylene can be readily formed it will rapidly revert to silirane, thus
accounting for the proposed label scrambling, in the isotopically substituted
silirane species.
 309

SiH .C H 2 2 4

SiH

Χ
<

Figure 18. Enthalpy surface for the reaction of SiH + C H 2 2 4

SiH 2 • C H 2 2

500

TS3
TS4

400

Ο SiH 3

Ε
"3

X 300
<3

SiH 2

: —SiH,

200 \—

Figure 19. Enthalpy surface for the reaction of SiH + C H 2 2 2

310

Examination of Figure 19 shows that, in the SiH + C H reaction, silirene

2 2 2

(in contrast to silirane) can isomerise more easily than revert to SiH + C H . 2 2 2

This can be explained by the low energy accessibility of the silylvinylidene

intermediate leading to ethynylsilane (as found by Becerra and Walsh [119]), so
that even if the transition state for its formation is tight, the energetic
advantage is sufficient to offset the benefit of the loose transition state for
reversion of silirene to SiH + C H . 2 2 2

The energies of vinylsilylene and the transition state for its formation show
that they can explain the isotopic scrambling mechanism suggested [109,110]
in the SiH 4- C D studies. Because vinylsilylene is endothermic relative to
2 2 2

silirene, when vinylsilylene is formed it will rapidly revert to silirene, again

explaining the proposed label scrambling mechanism in the isotopic silirene
species.
As explained in section 1, SiH reactions are considerably less exothermic
2

than those of C H . It is the lowering of energy associated with silylenes (DSSE)

2

which significantly determines the energy surface diagram and gives rise to
their special chemistry. The facility of the three membered rings to open to
silylenes (or decompose to give SiH ) shows that ring opening to biradicals (as
2

for cyclopropane and cyclobutane [125]) is by no means the only or obvious

fate in the thermolysis of small strained rings.

5. REACTIONS WITH O-DONOR MOLECULES

That there is a high propensity for reaction of silylenes with lone-pair donor
molecules is most vividly illustrated by the matrix isolation studies of the groups
of West [126,127] and Ando [128,129] which have revealed the existence of
molecular complexes, stabilised at low temperatures, between silylenes and a
variety of Ο-, N- and S- containing molecules. The formation of such species
is strongly supported by theoretical calculations [130] which show the
existence of energy minima on the potential surfaces for reactions of SiH with 2

N H , H 0 , HF, PH , H S and HCI. This is illustrated in Figure 20 for the

3 2 3 2

reaction of SiH with H 0 .2 2

The H S i - O H complex has a significant barrier to rearrangement (via 1,2 H-

2 2

migration) to the final product H SiOH. Similar surfaces, but with differing well
3

depths and rearrangement barriers for the donor-acceptor complexes, were

found for the other reactions. Prior to the matrix-isolation and theoretical work,
Weber and co-workers had already proposed the idea of the formation of an
initial zwitterionic complex in the reactions SiMe with alcohols and ethers
2

[131-133] based on product studies and isotope effect measurements in

solution. There is very little direct kinetic data on these reactions and in
virtually all that exists is on reactions of SiMe with alcohols and ethers, either
2

in solution [39,82] or the gas phase [59,134]. The kinetics are considered
within the mechanistic framework implied by involvement of the donor-acceptor
(zwitterionic) intermediate complexes.
 311

H Si
2 +

H SiOH
3

Figure 20. Potential energy surface for reaction of SiH with H 0 2 2

5 . 1 . SiMe + water, alcohols, ethers

2

Gas-phase data at 295K have been obtained by Baggott et al [59,134].

The rate constants obtained are shown in Table 27. This is the only gas-phase

Table 27
Gas-phase rate constants for reactions of SiMe with water, alcohols and ethers
2

at 298 Κ
Reactant A710 1 2 3
cm molecV 1
Ρ 1 Torr (Ar)
H 0
2 0.12 ± 0.05 10
D 0
2 0.077 ± 0.005 10
MeOH 1.18 ± 0.09 5
MeOD 1.03 ± 0.05 5
a
Me 0 2 2.41 5-15
Oxirane 3.29 ± 0.06 1
3.59 ± 0.08 2
3.80 ± 0.14 3.5
4.16 ± 0.09 5
4.60 ± 0.09 10
Oxetane 33.4 ± 1.17 5
3,3-Dimethyloxetane 46.6 ± 2.4 5
a
Tetrahydrofuran (THF) 3.92 2,5
a. Obtained from modelling biexponential decays.
312

data. The solution phase data is discussed later. Two rate constants in Table
27 were obtained by modelling bi-exponential decay traces. These are the only
two cases thus far uncovered of such kinetic behaviour for silylene reactions.
An example of these decay traces is shown in Figure 2 1 .

10 20 30 40

TIME///S

Figure 21 SiMe time decay profiles for reaction with (a) M e 0 (3.5 Torr) and
2 2

(b) THF (1.6 Torr). Solid lines correspond to a four step mechanism in each
case

Baggott, Blitz and Lightfoot [59] considered a number of mechanisms to

account for this behaviour for the reaction with M e 0 . All of them involved
2

reversible formation of the donor-acceptor complex, viz:

Me Si + OMe
2 2 τ± Me Si—OMe
2 2

The preferred fit mechanism then involved a scheme in which uncomplexed

(i.e. residual) SiMe decayed slowly by reaction with precursor augmented by
2

reaction with photolysis products of M e 0 . This and the study of SiMe with
2 2

TMF thus provide kinetic confirmation of the involvement of donor-acceptor

complexes. From the kinetic data the authors estimated potential well depths
(AH °/kJ mol" ) of -37 ± 3 and -45 ± 3 for the two complexes using a
f
1

thermodynamic estimate of the entropy change involved in their formation.

This is broadly consistent with the theoretical estimate of the stability of H S i - 2

O H [130,135]. Even more convincing was the study, in cyclohexane solution,

2

carried out by Levin et al [39]. They observed directly, via its absorption at 310
nm, the formation of the Me Si-THF complex concomitent with SiMe decay
2 2

in the presence of THF. The growth and decay constants of the two species
 313

were the same. However there are two differences with the gas phase. First
biexponential behaviour was not observed. This is explained by the use of an
8-fold higher concentration of THF by Levin et al [39] which drives complex
formation nearly to completion. The second difference is that the rate constant
1 1 3 1 1
for SiMe + THF is 2.2 χ 10 c m molec" s" . This is a factor of 5.5 greater
2

than that in the gas-phase. This is a large factor for a non-polar solvent like
cyclohexane and there may be another explanation (see later).
The rate constants of Table 27 were correlated by Baggott et al [134] with
the ionisation energies of the substrate molecules (both for the "in plane" (a ) ;

and "out of plane" (b ) orbitals). The correlation was reasonable for H 0 ,

7 2

methanol and unstrained ethers but poor for the strained rings (particularly
oxetane). Thus this provided some support for the rate influencing effect of the
electrophilic interaction. The following mechanism was proposed:

0
""· Y
η Y

This is simply the extension of the complex forming reaction to allow for its
rearrangement to final product (via Y-group migration) and corresponds to the
two stage, "electrophilic" followed by "nucleophilic", process proposed for the
Si-Η insertion reaction.
However the SiMe + MeOH/MeOD results pose some difficulties. The
2

isotope effect in the gas-phase is ca 1.15, in contrast to a value of 2.14 for

SiMe + EtOH/EtOD in cyclohexane [133]. The latter value led Steele & Weber
2

to suggest the second step in the process was rate determining, whilst the gas-
phase data suggest a less clear cut situation. There are several values for
absolute rate constants for reaction with alcohols (in non polar solvents) which
are shown in Table 28. The earlier questionable [39] values [15] are not
included.
The data show no great effect of alkyl groups on the rate constants, but the
value for MeOH exceeds the gas-phase value by more than a factor of 10.
Unless there is an unexpected solvent effect there may be undetected pressure
dependences of the gas-phase rate constants for some of the reactions
(particularly the small molecules) in Table 27. At the present time until this
question is resolved, there will remain a question mark about which step is rate
determining in the SiMe + alcohol systems. The high values for the rate
2

constants of Table 28 tend to support the first step (zwitterion formation).

Further discussions of the mechanism of the SiMe + MeOH system may be
2

found in the original paper [134].

314

Table 28
Rate constants for reactions of SiMe with alcohols in solution at 298 Κ 2

11 3 1
Reactant k/10" cm molec" s" 1
Solvent Reference
MeOH 1.5 methylcyclohexane [82]
EtOH 1.5 cyclohexane [39]
Me CHOH
2 1.7 cyclohexane [39]
Me COH3 1.7 cyclohexane [39]

For SiMe + H 0 , the most recent theoretical calculation indicates a

2 2
1
secondary barrier of ca 26 kJ mol" , which would certainly make the second
step rate determining. The measured gas-phase rate constant however would
1
only accommodate a maximum (overall) barrier of 17 kJ mol" [134]. The
uncertainties of the data on SiMe + H 0 / D 0 unfortunately do not allow any
2 2 2

conclusions based on the isotope effect.

More recently Blitz [122] has studied the temperature dependence of the
SiMe + MeOH reaction at pressures of 10 Torr (Ar) over a temperature range
2

295-421 K. The rate constants decreased strongly with temperature

3 1 1
corresponding to Arrhenius parameters of logW/cm molec" s" ) = -14.7 and
1
E = -15.7 kJ mol" . A brief check apparently showed the reaction to be
a

independent of pressure. At temperatures above 421 Κ the SiMe decay plots 2

were not single exponential. These data are broadly consistent with the
complex mechanism, although the question of the rate determining step (at
295 K) remains.
In the ether systems at 295 Κ [134] a pressure dependence was not
observed for SiMe + M e 0 , but was for SiMe + oxirane. However in the
2 2 2

former case the pressure range was limited to 5-15 Torr (Ar). Higher rate
constants (at infinite pressure) for these two reactions would certainly be more
consistent with those SiMe + oxetane and SiMe + 3,3-dimethyloxetane.
2 2

The proposed mechanism for the oxirane reaction was:

CH 2 0λ\ 2

Me Si + 0
2 χ I ^ Me Si—O^J -* M e S i = 0 + C H
2 2 2 4

CH 2 CH 2

Product C H was detected, although Me Si = 0 (or its oligomers) were not.

2 4 2

Since the overall reaction is not an association process the pressure dependence
was argued to be related to the formation of the intermediate complex. This
has been further supported by the unpublished results of Blitz [122] which
show that the pressure dependence persists up to 4 3 0 Κ but diminishes in
magnitude. The rate constants also decrease (at a given pressure). At 5 Torr
3 1 1
(Ar) the Arrhenius parameters are logW/cm molec" s" ) = -12.84 and E = g
1
-8.2 kJ mol" . The pressure dependence remains to be modelled but its
temperature dependence is contrary to expectations for a third body assisted
 315

association process. If confirmed this would be the first evidence for pressure
stabilisation of a donor-acceptor complex.

5.2 SiH + M e 0
2 2

SiH kinetic studies with either H 0 or MeOH have yet to be undertaken.

2 2

The reaction with M e 0 is the only reaction of SiH with an O-donor for which
2 2

there are absolute rate constants. King, Lawrance and Staker [136] have
12 3 1 1
obtained values of between 6.5 χ 10" c m molec" s" at 3 Torr and 1.25 χ
11 3 1 1
10" c m molec" s at 11 Torr (Ar), both at 298 K. Additionally they have
studied the temperature dependence from 298-434 Κ from which they have
3 1 1
obtained the Arrhenius parameters (at Ar, 5 Torr) of log(>4/cm molec" s" ) = -
1
13.71 and E = -15.3 kJ mol" . These however do not represent the limiting
a

high pressure values since the rate constants are clearly pressure dependent.
Becerra, Carpenter and Walsh [137] have confirmed the pressure
dependence of the reaction in SF and obtained values of 1.35, 1.72, 4.80 and
6
11 3 1 1
7.54 χ 10" c m molec" s" at pressures of 3, 5, 20 and 50 Torr. The results
are clearly consistent with a third body assisted association reaction with SF 6

a more efficient collider than Ar. The surprise is that this reaction shows no
evidence (biexponential decays) of reaching equilibrium as observed for SiMe 2

+ M e 0 . An independent GC search failed to find evidence for MeSiH OMe

2 2

[138]. This appears to suggest that the reaction proceeds irreversibly to the
zwitterionic complex as shown below, and that there is no further reaction.

H Si + O M e
2 2 H Si---OMe
2 2

The implication is that the pressure dependence reflects this association and
that the complex is more strongly bound than M e S i - O M e . It is not clear why 2 2

this should be so. RRKM calculations (not yet undertaken) should give
information about the binding energy.

5.3 Reactivity of complexed SiMe 2

Studies of SiMe in the presence of several complexing agents in

2

cyclohexane solution have been carried out by Levin et al [39]. Although not
a gas-phase experiment this study offers the first direct kinetic information
about "solvated" silylenes ("silylenoids" would be another name) and we
thought this important enough for inclusion. Table 29 shows the data for a
number of reactions of the SiMe -THF complex and also for reactions of 2

several other SiMe complexes with C H O H . These were obtained by direct

2 2 5

monitoring of the absorptions of the complexes.

These results clearly show the reduced reactivity of the complexed SiMe 2

relative to the free silylene in the reactions with ethanol. Comparison with data
in sections 3 and 4 also shows the reduced reactivity of the SiMe -THF 2

complex relative to free SiMe in the Si-Η insertion process and the alkene
2

addition reactions. The authors did not attempt to correlate the reactivities of
different complexes with any solvent or other parameter of the complexing
agent. However they did apply some checks (a) to show product ratios were
consistent with the kinetics and (b) to rule out the possibility that these rates
316

Table 29
Rate constants for SiMe complexes in solution [39]
2

Complexing agent Reagent /r/cm molec" s"

3 1 1

None C H OH
2 5 1.5 χ 1 0 - 1 1

THF C H OH 2 5 3.2 χ - 1 3
1 Q

(CH ) COH 3 3 7.3 χ 1 0 - 1 4

4.0 χ 10-
=
CH2 CH(CH ) CH 2 3 3
1 5

(CH ) SiCH=CH
2 3 2 4.3 χ 1 0 1 5

(CH CH CH ) SiH
3 2 2 3 1.4 χ - 1 5
1 Q

Et 0 2 C H OH
2 5 1.6 χ 1 0 - 1 2

Et N 3 C H OH
2 5 5.5 χ 10- 1 4

1 3
CH CN
3 C H OH
2 5 1.8 χ 1 0

could be attributable to a low concentration of uncomplexed SiMe in 2

equilibrium with the complexes themselves. Presumably then the measured rate
constants refer to nucleophilic displacement reactions of the complexing agent
from the complex by the reactive substrate, viz:

Me Si-complex + substrate -> Me Si-substrate (product) + complex

2 2

These are therefore examples of 5^2 reactions on silicon.

6 . REACTIONS WITH SMALL INORGANIC MOLECULES

Silylenes react with many small inorganic molecules. These reactions belie
their apparent simplicity in the potential variety of secondary chemistry which
can arise. Although very little of the chemistry is known with great confidence
there has been considerable discussion and conjecture. Much of this has
already been discussed in the excellent review of Safarik et al [ 1 1 ] . Since
relatively little has been published since then we decided against a detailed
repetition of the discussion of these reactions. For completeness however we
include a tabulation of rate constant data. To provide a contrast with the earlier
review this is done by substrate molecule. The data are shown in Table 30.
A number of these reactions are pressure dependent which makes
comparison of their rate constants a tricky proposition. Where comparisons are
possible they generally conform to the silylene reactivity sequence SiH > 2

SiMe > SiCI > SiBr . Shortage of data makes the placement of SiF in this
2 2 2 2

sequence impossible. It is generally agreed that SiF is fairly unreactive [ 5 , 7 ] . 2

 317

Exceptions to the general sequence, however, exist since with CO, SiH and 2

SiMe do not react under experimental conditions whereas SiCI does. This
2 2

Table 30
Rate constants for silylene reactions with small inorganic molecules at 298 Κ
Substrate Silylene k 1 cm molec"V
3 1
Ρ 1 Torr Ref.
0 2 SiH 2 ( 7 . 5 ± 0 . 8 ) χ 10 1 2
1 ( He ) [139]
( 1 . 4 ± 0 . 2 ) χ 10" 11
9.5 ( He ) [139]
SiMe (2.5±0.2) χ 1 0 1 3
2

1 2
5 (Ar) [134]
( 4 . 2 ± 0 . 5 ) χ 10 ( Solution ) [39]
SiF 2
< 2 χ 10 1 7

2 (Ar) [87]
SiCI 2
( 5 . 6 5 ± 0 . 1 2 ) χ 10 1 2

100 ( A r ) [140]
SiBr 2
(9.3±0.07) χ 1 0 1 3

40 ( Ar ) [141]
11
NO SiH 2 ( 1 . 5 ± 0 . 2 ) χ 10" 1 ( He ) [139]
1 1
(2.1 ± 0 . 2 ) χ 10 9.5 ( He ) [139]
SiCI 2
( 2 . 6 6 ± 0 . 0 2 ) χ 10" 12
100 ( Ar ) [142]
SiBr 2
( 4 . 6 5 ± 0 . 0 7 ) χ 10 1 3
40 ( Ar ) [141]
CO SiH 2 < 1 χ 10 1 3
5 ( He ) [139]

SiMe 2 < 5 χ 10 1 4
5 (Ar) [134]

SiCI 2 1.05 χ 10 1 2
100 ( Ar ) [143]
N 0 2
SiH 2 ( 1 . 9 0 ± 0 . 0 9 ) χ 10 1 2
Ρ indep. [144]

SiCI 2 ( 9 . 4 7 ± 0 . 0 5 ) χ 10" 13
100 ( Ar ) [143]
HCI SiH 2 ( 4 . 3 ± 0 . 6 ) χ 10" 12
1 ( He ) [139]
( 7 . 5 ± 1 . 0 ) χ 10" 12
9.5 ( He ) [139]
Cl 2 SiH 2 ( 1 . 4 ± 0 . 2 ) χ 10 1 0
5 ( He ) [139]

SiF 2 (5.1 ± 0 . 6 ) χ 10 1 3
2.5 ( Ar ) [145]

F 2
SiF 2 ( 4 . 7 ± 0 . 3 ) χ 10 1 3
< 1 0 ( Ar ) [145]
318

may be a question of third body stabilisation (i.e. a pressure effect), which was
not sufficient under experimental conditions for SiH and SiMe since stable2 2

adducts have been calculated to exist [146] and indeed observed in low
temperature matrices [147,148]. It is probable that, as with the reactions
already discussed, the electrophilic character of the silylene determines the
initial interaction with these substrates since they are all capable of lone pair
donation.
The only measured temperature dependences amongst these reactions are
those for SiH + N 0 [144], SiF + Cl [145] and SiF + F [145]. SiH +
2 2 2 2 2 2 2

N 0 has a negative temperature dependence (295-747 K) with E = -2.0 ± 0 . 3

2 a
1
kJ mol" whilst SiF -I- F has a positive temperature dependence (282-617 K)
2 2
1
with E = + 7 . 7 ± 0 . 8 kJ mol" . For SiF + Cl , rate constants increase with
a 2 2

temperature but the Arrhenius plot is curved above 700 K. Between 281-700
K, E = 3.3 kJ mol \
a

The relative rate studies of the group of Lampe, using IRMPD generation of
SiH , of the reactions of SiH + HCI [149], and SiH + NO [150], in each case
2 2 2

relative to SiH + SiH , do not seem to match the absolute values very closely.
2 4

7. GENERAL COMMENTS AND CONCLUSIONS

In conclusion to their survey of silylene rate constants in 1990, Safarik et

al [11 ] highlighted the fact that the absolute rate constant measurements which
only began in 1985, had conclusively demonstrated that silylene reactions were
much faster than was estimated earlier from indirect or relative methods. More
cautiously they stated that only a few broad correlations could be made of
silylene reactivity and much more work was needed to derive a clear
appreciation of structure-reactivity relationships. We may now be a little more
confident. Significant patterns have begun to emerge. Four years ago, at the
time of the earlier review, almost no temperature dependence studies had been
carried out. During the intervening period a significant number have been
undertaken and published. One finding, that of negative activation energies,
seems to be almost universal. The magnitudes of some of these negative
activation energies strongly support mechanisms involving intermediate
complexes. In the case of reaction with O-donors, such complexes have been
directly observed and their kinetics monitored. Equilibrium between the free
silylene (in the case of SiMe ) and the donor adducts have been established in
2

two examples and the stabilities of the complexes (in terms of their binding
energies) have been measured. For the Si-Η insertion reaction the
comprehensive investigation of methyl substituent effects has revealed a wealth
of detail, not all of it yet fully understood, but consistent with a very polar, H-
bonded, intermediate complex. By and large, theoretical calculations of
potential energy surfaces have confirmed (or indeed anticipated) this picture.
The simple idea of electron pair donor-acceptor interactions seems to
underpin the mechanisms of all these processes. It is remarkable that for all
three major classes of reaction, the Si-Η bond insertion, the ,τ-type addition and
the η-donor addition, they can be described as two stage processes involving
 319

an initial electrophilic stage followed by a subsequent nucleophilic stage. The

coupling of these stages may be very close, as in the case of the SiH + SiH 2 4

reaction, with a barely identifiable intermediate. Alternatively the stages may

be well separated, as in the Me Si + O M e reaction where the intermediate,
2 2

M e S i - O M e is sufficiently stable that there is no actual evidence for the

2 2

second stage, viz. its unimolecular rearrangement to Me SiOMe, on the 3

timescale of the experiments. The contrasting behaviour of the different

silylenes has begun to take shape. The prototype, SiH , emerges as one of the
2

most reactive species known, with rate constants at the collisional limit in
several of its reactions. Substitution of Η for other groups reduces the
reactivity, the effect depending on the substituent. Methyl groups deactivate
to a variable extent, depending on the reaction and whether one or both
hydrogens in SiH are replaced. This may be understood, generally, in terms
2

of an electron withdrawal (inductive) effect causing shrinkage of the electron-

pair-seeking Si(3p) orbital on the silylene. Methyl groups, being marginally
electronegative relative to silicon behave differently than in purely organic
systems. Substitution of Η by halogens causes greater deactivation still,
explained by the back donation (77-type interaction) of lone pair electron density
into the empty Si(3p) orbital thus reducing its electrophilicity.
The movement of electron pairs and therefore centres of charge density in
silylene reactions causes charge separation in the intermediates. It is therefore
to be expected that silylene reactions will be particularly sensitive to
polarisation effects. This can be vividly illustrated by the contrast in behaviour
in the reaction of silylenes with Si-Η and C - H bonds. In the former case the
i+
polarity of the bond is S i - H * " and the electrophilic silylene has no difficulty
approaching the negatively charged hydrogen in the initial interaction stage. By
4-
contrast carbon-hydrogen bonds are polarised C^'-H* and the silylene access
to the desired electron pair is hampered by the positively charged hydrogen.
Reactions of SiH with C-H bonds are not observed, whereas with Si-Η bonds
2

they can occur on every collision! It is worth noting that C - H bonds are not
totally immune from reaction with silylenes since, when they are built into the
same molecule, reaction can occur. The rearrangement of ethylsilylene to
silirane is an illustration of this. There are many other higher temperature
examples of these intramomecular C-H insertion processes [5]. The process,
however, looks as though it requires considerable activation. A lot of effort has
been made, particularly by Davidson and coworkers [112,151] to estimate the
activation energies involved.
We may address one final question concerning the behaviour of silylenes.
How does their stability affect their reactivity? Here we face a paradox. It
almost seems that the answer is not at all! There is no doubt that, in terms of
thermodynamics and chemical bonding, silylenes are stabilised, the magnitude
of this having been formalised in the DSSE scale. Yet silylenes can react on
every collision. SiH adds to C H faster than C H (which is not stabilised). It
2 2 4 2

is true that the more stabilised silylenes, such as SiCI , are less reactive than
2

the prototype SiH , but nevertheless the reactivity is remarkably high. Another
2

interesting comparison would be SiCI with CCI but direct data are not
2 2

available for the latter. Where the stabilisation energy effects impose
320

themselves on the reactivity pattern is in the diminished energy release of

silylene reactions (in comparison, say, to carbene reactions). This was
mentioned at the beginning of this article. Its detailed manifestation emerges
in the secondary behaviour of the initially-formed silylene reaction products.
Products such as S i H from SiH + SiH , or silirane from SiH + C H only have
2 6 2 4 2 2 4

enough energy to revert (partially) to reactants, as shown by the third body

stabilisation requirement. In the analogous C H reactions, with their greater
2

energy release, extensive further reaction can occur through the effects of
chemical activation. Other consequences of silylene stabilisation are the
involvement of substituted silylenes such as ethyl- and vinyl-silylene in the
rearrangements of silirane and silirene, respectively, as revealed by the isotopic
scrambling studies. Thus the connection between reactivity and stability for
silylenes is a rather loose one. Silylene reactions offer yet one more example
to chemists, and indeed an excellent one, that thermodynamic and kinetic
stability are two quite different matters.

7.1 The future for silylene kinetic studies

Although the pattern of kinetic and mechanistic behaviour of silylenes has
begun to emerge a lot remains to be done. Identifying, characterising and
learning about the intermediate complexes in silylene reactions needs a lot more
work. Whether they even occur in the /r-type addition reaction is an open
question. Certainly the separation of the electrophilic and nucleophilic stages
of these processes and their individual kinetic characterisation is a challenging
prospect. Combined pressure and temperature variation studies remain to be
carried out for many of the simplest silylene reactions even within the currently
easily accessible range. Extension of studies outside of the current range,
particularly to sub-ambient temperatures would be very useful. Product
characterisation remains a problem in many reactions. Development of methods
to handle and study some of the specially sensitive compounds involved such
as siliranes and silirenes would be very useful. The powerful methods of
reaction dynamics (study of state-selected species, beam orientation studies,
examination of the details of energy partitioning amongst product quantum
states) have not yet been applied at all in this area. Silylene reactions and
behaviour could surely benefit enormously from use of these techniques.

7.2. The general prospect

Silylenes are both electron-pair acceptors and donors at the same time.
They are both Lewis acids and Lewis bases. They are ambiphilic. Their
reactions represent a class of acid/base processes in the gas-phase now
sufficiently broad to provide models of behaviour. There are few other gas-
phase examples, although solution chemistry abounds with them. Free of the
complications of solvent effects, gas-phase silylene chemistry thus offers a
blueprint which may be extended to other areas. Two which spring to mind are
boron hydride chemistry, and carbenium ion-molecule reactions. An
examination of these fields from a different perspective offers a stimulating
prospect.
 321

ACKNOWLEDGEMENT

The authors would like to express their thanks to many colleagues at the
University of Reading who have contributed to the work described in this
article, as well as to members of the wider organosilicon chemistry community
who have provided stimulation, encouragement, advice and much helpful
discussion on this subject, over the years.

REFERENCES

1. E.G. Rochow, Silicon and Silicones, Springer-Verlag, Berlin, 1387.

2. W.H. Atwell and D.R. Weyenberg, Angew. Chem. Int. Ed. Engl., 8 (1969)
469.
3. Silylenes were discussed earlier than 1969, but as part of an overall
review of Carbene Analogues by O.M. Nefedov and M.N. Manakov,
Angew. Chem. Int. Ed. Engl., 5 (1966) 1 0 2 1 .
4. P.P. Gaspar and B.J. Herold, Chap. 13 in Carbene Chemistry, W. Kirmse
(ed.). Academic Press, New York, 1 9 7 1 , p. 504.
5. P.P. Gaspar in Reactive Intermediates, M. Jones Jr. and R.A. Moss (eds.),
Wiley, New York; 1 (1978) 229; 2 (1981) 335; 3 (1985) 333.
6. E.A. Chernyshev, D.G. Komalenkova and S.A. Bashkirova, Russ. Chem.
Rev., 45 (1976) 913.
7. Y-N. Tang, Chap. 4 in Reactive Intermediates, R.A. Abramovitch (ed.).
Plenum, 1982, Vol. 2, p. 297.
8. C-S. Liu and T-L. Hwang, Chap. 1 in Adv. in Inorganic Chemistry and
Radiochemistry, H.J. Emeleus and A.G. Sharpe (eds.). Academic Press,
London, 1985, Vol. 29, p. 1.
9. I.M.T. Davidson, Ann. Rep. Prog. Chem., Sect. C, 81 (1985) 4 7 .
10. M.J. Almond, Short Lived Molecules, Ellis Horwood, London, 1990,
Chap. 5, p. 93.
11. I. Safarik, V. Sandhu, E.M. Lown, O.P. Strausz and T.N. Bell, Res. Chem.
Intermediates, 14 (1990) 105.
12. G. Inoue and M. Suzuki, Chem. Phys. Lett., 122 (1985) 3 6 1 .
13. In fact the first claimed absolute rate study of a silylene was by Gaspar
and colleagues [14], who reported time resolved kinetics in solution,
attributed to PhSiMe. It is now thought, from the reported rate constant
values, that the transient involved was probably not a silylene.
14. P.P. Gaspar, B-H. Boo, S. Chari, A.K. Ghosh, D. Holten and S. Konieczny,
Chem. Phys. Lett., 105 (1984) 153.
15. P.P. Gaspar, D. Holten, S. Konieczny and J.Y. Corey, Acc. Chem. Res.,
20 (1987) 329.
16. Y. Apeloig, Chap. 2 in The Chemistry of Organic Silicon Compounds (Part
1), S. Patai and Z. Rappoport (eds.), Wiley, New York, 1989, p. 57.
17. I. Dubois, Can. J. Phys., 46 (1968) 2485.
18. B.T. Luke, J.A. Pople, M-B. Krogh-Jespersen, Y. Apeloig, M. Kami, J.
Chandrasekharand P.v.R. Schleyer, J. Am. Chem. S o c , 108 (1986) 270.
322

19. R. Walsh, J. Chem. Soc, Faraday Trans. 1, 79 (1983) 2233.

20. R. Walsh, Pure & Appl. Chem., 59 (1987) 69.
21. R. Walsh, Chap. 11 in Energetics of Organometallic Species. J.A.
Martinho-Simoes (ed.), NATO-ASI Series C, Vol. 367, Kluwer, Dordrecht,
1992, p. 1 7 1 .
22. W . J . Bullock, R. Walsh and K.D. King, J. Phys. Chem., 98 (1994) 2595.
23. G. Pilcher, M.P.L. Leitao, Y. Meng-Yan and R. Walsh, J. Chem. S o c ,
Faraday Trans., 87 (1991) 8 4 1 .
24. R. Grev, Adv. Organometallic Chem., 33 (1991) 125.
25. H. Kollmar, Tetrahedron, 28 (1972) 5893.
26. H. Kollmar and V. Staemmler, Theor. Chem. Acta, 51 (1979) 207.
27. M.S. Gordon and D.R. Gano, J. Am. Chem. S o c , 106 (1984) 5 4 2 1 .
28. Kinetics and Spectroscopy of Carbenes and Biradicals, M.S. Platz (ed.).
Plenum, New York, 1990.
29. M.N.R. Ashfold, M.A.Fullstone, G. Hancock and G.W. Ketley, Chem.
Phys., 55 (1981) 245.
30. A.O. Langford, H. Petek and C.B. Moore, J. Chem. Phys., 78 (1983)
6650.
31. H.M. Frey and I.D.R. Stevens, J. Chem. S o c , (1962) 3514.
32. H.F. Schaefer III, Acc. Chem. Res., 15 (1982) 283.
33. R. Walsh, J. Chem. S o c , Chem. Comm., (1982) 1415.
34. T. Bohland, F. Temps and H.Gg. Wagner, Ber. Bunsenges. Phys. Chem.,
89 (1985) 1013.
35. W. Hack, M. Koch, R. Wagener and H.Gg. Wagner, Ber. Bunsenges.
Phys. Chem., 93 (1989) 165.
36. JANAF Thermochemical Tables, 2nd edn., D.R. Stull and H. Prophet
(eds.), NSRDS-NBS 37, Washington D.C., 1 9 7 1 .
37. N. Al-Rubaiey and R.Walsh, J. Phys. Chem., 98 (1994) 5303.
38. B.S. Rabinovitch and M.C. Flowers, Quart. Rev., 18 (1964) 122.
39. G. Levin, P.K. Das, C. Bilgrien and C.L. Lee, Organometallics, 8 (1989)
1206.
40. Reactive Intermediates in the Gas Phase. Generation and monitoring.
D.W. Setser (ed.). Academic Press, New York, 1979.
41. J.E. Baggott, H.M. Frey, P.D. Lightfoot and R. Walsh, J. Phys. Chem., 91
(1987) 3386.
42. J.O. Chu, D.B. Beach and J.M. Jasinski, J. Phys. Chem., 91 (1987)
5340.
43. B.P. Ruzsicska, A. Jodhan, I. Safarik, O.P. Strausz and T.N. Bell, Chem.
Phys. Lett., 113 (1985) 67.
44. J.W. Thoman Jr. and J.I. Steinfeld, Chem. Phys. Lett., 124 (1986) 35.
45. J.E. Baggott, H.M. Frey, K.D. King, P.D. Lightfoot, R. Walsh and I.M.
Watts, J. Phys. Chem., 92 (1988) 4025.
46. R. Becerra, H.M. Frey, B.P. Mason and R. Walsh, unpublished results.
47. B.P. Mason, H.M. Frey and R. Walsh, J. Chem. S o c , Faraday Trans., 89
(1993) 4405.
48. J.E. Baggott, M.A. Blitz, H.M. Frey, P.D. Lightfoot and R. Walsh, Chem.
Phys. Letters, 135 (1987) 39.
 323

49. R. Becerra, Η.Μ. Frey, Β.P. Mason and R. Walsh, J. Chem. S o c , Faraday
Trans., 89 (1993) 4 1 1 .
50. M.A. Blitz, H.M. Frey, F.D. Tabbutt and R. Walsh, J. Phys. Chem., 94
(1990) 3294.
51. G. Herzberg and R.D. Verma, Can. J. Phys., 42 (1964) 395.
52. P. Ho, W.G. Breiland and R.W. Carr, Chem. Phys. Lett., 132 (1986) 4 2 2 .
53. B.P. Ruzsicska, A. Jodhan, I. Safarik, O.P. Strausz and T.N. Bell, Chem.
Phys. Lett., 139 (1987) 72.
54. J.E. Baggott, H.M. Frey, P.D. Lightfoot and R. Walsh, Chem. Phys. Lett.,
125 (1986) 22.
55. G. Herzberg, Molecular Spectra and Molecular Structure, Van Nostrand
Reinhold, New York, 1966.
56. I. Dubois, G. Herzberg and R.D. Verma, J. Chem. Phys., 47 (1967) 4 2 6 2 .
57. J.M. Jasinski and J.O. Chu, J. Chem. Phys., 88 (1988) 1678.
58. M.J. Michalczyk, M.J. Fink, D.J. De Young, C.W. Carlson, K.L. Welsh
and R. West, Silicon, Germanium, Tin and Lead Compounds, 9 (1986)
75.
59. J.E. Baggott, M.A. Blitz and P.D. Lightfoot, Chem. Phys. Lett., 154
(1989) 330.
60. J.M. Jasinski, J. Phys. Chem., 90 (1986) 555.
61. P. John and J.H. Purnell, J. Chem. S o c , Faraday Trans., 69 (1973)
1455.
62. D.S. Rogers, K.L. Walker, M.A. Ring and H.E. O'Neal, Organometallics,
6 (1987) 2313.
63. M.S. Gordon, J. Chem. S o c , Chem. Comm., (1981) 890.
64. R.S. Grev and H.F. Schaefer III, J. Chem. S o c , Chem. Comm., (1983)
785.
65. A. Sax and G. Olbrich, J. Am. Chem. S o c , 107 (1985) 4 8 6 8 .
66. M.S. Gordon, D.R. Gano, J.S. Binkley and M.J. Frisch, J. Am. Chem.
Soc, 108 (1986) 2 1 9 1 .
67. J.E. Baggott, H.M. Frey, K.D. King, P.D. Lightfoot, R. Walsh and I.M.
Watts, J. Phys. Chem., 92 (1988) 4025.
68. N. Al-Rubaiey and R. Walsh, unpublished results.
69. B.P. Mason, H.M. Frey and R. Walsh, unpublished results.
70. K.F. Roenigk, K.F. Jensen and R.W. Carr, J. Phys. Chem., 91 (1987)
5726.
71. J.E. Baggott, H.M. Frey, P.D. Lightfoot, R. Walsh and I.M. Watts, J.
Chem. S o c , Faraday Trans., 86 (1990) 27.
72. T.R. Dietrich, S. Chiussi, M. Marek, A. Roth and F.J. Comes, J. Phys.
Chem., 95 (1991) 9302.
73. R. Becerra, H.M. Frey, B.P. Mason, R. Walsh and M.S. Gordon, J. Am.
Chem. S o c , 114 (1992) 2 7 5 1 .
74. J.G. Martin, M.A. Ring and H.E. O'Neal, Int. J. Chem. Kinet., 19 (1987)
715.
75. R. Becerra, H.M. Frey, B.P. Mason, R. Walsh, unpublished results.
76. A.J. Vanderwielen, M.A. Ring and H.E. O'Neal, J. Am. Chem. S o c , 97
(1975) 993.
324

77. I.C. Carpenter and R. Walsh, unpublished results.

78. R. Becerra and R. Walsh, unpublished results.
79. C D . Eley, M.C.A. Rowe and R. Walsh, Chem. Phys. Lett., 126 (1986)
153.
80. B.A. Sawrey, H.E. O'Neal, M.A. Ring and D. Coffey Jr., Int. J. Chem.
Kinet., 16 (1984) 3 1 .
81. G. Raabe, H. Vancik, R. West and J. Michl, J. Am. Chem. S o c , 108
(1986) 6 7 1 .
82. H. Shizuka, H. Tanaka, K. Tonokura, K. Murata, H. Hiratsuka, H. Ohshita
and M. Ishikawa, Chem. Phys. Lett., 143 (1988) 225.
83. I.M.T. Davidson and N.A. Ostah, J. Organometal. Chem., 206 (1981)
149.
84. J.E. Baggott, M.A. Blitz, H.M. Frey and R. Walsh, J. Am. Chem. S o c ,
112 (1990) 8337.
85. H.E. O'Neal, M.A. Ring, W.H. Richardson and G.F. Licciardi,
Organometallics, 8 (1989) 1968.
86. I. Safarik, B.P. Ruzsicska, O.P. Strausz and T.N. Bell, Chem. Phys. Lett.,
113 (1985) 7 1 .
87. A. Freedman, K.E. McCurdy, J. Wormhoudt and P.P. Gaspar, Chem.
Phys. Lett., 142 (1987) 255.
88. C. Sosa and H.B. Schlegel. J. Am. Chem. S o c , 106 (1984) 5847.
89. R.L. Jenkins, A.J. Vanderwielen, S.P. Ruis, S.R. Gird and M.A. Ring,
Inorg. Chem., 12 (1973) 2968.
90. M. Ishikawa, T. Takaoka, and M. Kumada, J. Organometal. Chem.,
42 (1972) 333.
91. M.S. Gordon, private communication.
92. S. Sakai and M. Nakamura, J. Phys. Chem., 97 (1993) 4 9 6 0 .
93. R.C. Dobson, D.M. Hayes and R. Hoffmann, J. Am. Chem. S o c , 93
(1971) 6188.
94. S. W. Benson, Adv. Photochem., 2 (1964) 1.
95. G. Trinquier, J. Chem. S o c , Faraday Trans., 89 (1993) 775.
96. R. Walsh, A c c Chem. Res., 14 (1981) 24.
97. R. Walsh, Chap. 5 in The Chemistry of Organic Silicon Compounds (Part
1), S. Patai and Z. Rappoport (eds.), Wiley, New York, 1989, p. 3 7 1 .
98. M. Barber, A.M. Doncaster and R. Walsh, Int. J. Chem. Kinet., 14 (1982)
669.
99. K.N. Houk, N.G. Rondan and J. Mareda, Tetrahedron, 14 (1985) 1555.
100. V.J. Tortorelli, M. Jones Jr., S. Wu and Z. Li, Organometallics, 2 (1983)
759.
101. R.T. Conlin and P.P. Gaspar, J. Am. Chem. S o c , 98 (1976) 3715.
102. N. Al-Rubaiey, H.M. Frey, B.P. Mason, C. McMahon and R. Walsh, Chem.
Phys. Lett., 204 (1993) 3 0 1 .
103. J.A. Boatz and M.S. Gordon, J. Phys. Chem., 93 (1989) 3025.
104. D.A. Horner, R.S. Grev and H.F. Schaefer III, J. Am. Chem. S o c , 114
(1992) 2093.
105. D.H. Berry, private communication.
 325

106. J.R. Fisher and F.W. Lampe. J. Photochem. Photobiol. A, Chem., 58

(1991) 173.
107. F. Anwari and M.S. Gordon, Israeli J. Chem., 23 (1983) 129.
108. N. Al-Rubaiey and R. Walsh, unpublished results.
109. R. Becerra, H.M. Frey, B.P. Mason and R. Walsh, J. Chem. S o c , Chem.
Comm., (1993) 1050.
110. R. Becerra and R. Walsh, Int. J. Chem. Kinet., 26 (1994) 45.
1 1 1 . J.W. Erwin, M.A. Ring and H.E. O'Neal, Int. J. Chem. Kint., 17 (1985)
1067
112. I.M.T. Davidson, J. Organometal. Chem., 437 (1992) 1.
113. N. Al-Rubaiey and R. Walsh, unpublished results.
114. J.E. Baggott, M.A. Blitz, H.M. Frey, P.D. Lightfoot and R. Walsh, J.
Chem. S o c , Faraday Trans., 84 (1988) 515.
115. D.S. Rogers, H.E. O'Neal and M.A. Ring, Organomet., 5 (1986) 1467.
116. P.P. Gaspar, S. Konieczny and S.H. Mo, J. Am. Chem. Soc, 106 (1984)
424.
117. R-J. Hwang and P.P. Gaspar, J. Am. Chem. Soc, 100 (1978) 6626.
118. J.A. Boatz, M.S. Gordon and L.R. Sita, J. Phys. Chem., 94 (1990) 5488.
119. R. Becerra and R. Walsh, unpublished results.
120. D. Lei, R-J. Hwang and P.P. Gaspar, J. Organometal. Chem., 271 (1984)
1.
1 2 1 . M.P. Clarke and I.M.T. Davidson, J. Chem. S o c , Chem. Comm., 3
(1988) 2 4 1 .
122. M.A. Blitz, Ph.D. Thesis, University of Reading, 1990.
123. W. Hack, M. Koch, H.Gg. Wagner and A. Wilms, Ber. Bunsenges. Phys.
Chem., 92 (1988) 674.
124. A.P. Dickenson, H.E. O'Neal and M.A. Ring, Organometallics, 10 (1991)
3513.
125. H.M. Frey and R. Walsh, Chem. Rev., 69 (1969) 103.
126. G.R. Gillette, G.H. Noren and R. West, Organometallics, 6 (1987) 2617.
127. G.R. Gillette, G.H. Noren and R. West, Organometallics, 8 (1989) 487.
128. W. Ando, K. Hagiwari and A. Sekiguchi, Organometallics, 6 (1987) 2270.

129. W. Ando, A. Sekiguchi, K. Hagiwari, A. Sakakibari and H. Yoshida,

Organometallics, 7 (1988) 558.
130. K. Raghavachari, J. Chandrasekhar, M.S. Gordon and K.J. Dykema, J.
Am. Chem. S o c , 106 (1984) 5853.
1 3 1 . D. Tzeng and W.P. Weber, J. Am. Chem. S o c , 102 (1980) 1 4 5 1 .
132. T.Y. Gu and W.P. Weber, J. Am. Chem. S o c , 102 (1980) 1 6 4 1 .
133. K.P. Steele and W.P. Weber, Inorg. Chem., 20 (1981) 1302.
134. J.E. Baggott, M.A. Blitz, H.M. Frey, P.D. Lightfoot and R. Walsh, Int. J.
Chem. Kinet., 24 (1992) 127.
135. S. Su and M.S. Gordon, Chem. Phys. Lett., 204 (1993) 306.
136. K.D. King, W.D. Lawrance and W.S. Staker, private communication.
137. R. Becerra, I.M. Carpenter and R. Walsh, unpublished results.
138. N. Al-Rubaiey and R. Walsh, unpublished results.
326

139. J.O. Chu, D.B. Beach, R.D. Estes and J.M. Jasinski, Chem. Phys. Lett.,
143 (1988) 135.
140. V. Sandhu, A. Jodhan, I. Safarik, O.P. Strausz and T.N. Bell, Chem. Phys.
Lett., 135 (1987) 260.
1 4 1 . V. Sandhu, O.P. Strausz and T.N. Bell, Res. Chem. Intermediates, 13
(1990) 43.
142. V. Sandhu, I. Safarik, O.P. Strausz and T.N. Bell, Res. Chem.
Intermediates, 11 (1989) 19.
143. V. Sandhu, O.P. Strausz and T.N. Bell, Res. Chem. Intermediates, 11
(1989) 235.
144. R. Becerra, H.M. Frey, B.P. Mason and R. Walsh, Chem. Phys. Lett., 185
(1991) 415.
145. A.C. Stanton, A. Freedman, J. Wormhoudtand P.P. Gaspar, Chem. Phys.
Lett., 122 (1985) 190.
146. T.P. Hamilton and H.F. Schaefer III, J. Chem. Phys., 90 (1989) 1 0 3 1 .
147. C.A. Arrington, J.T. Petty, S.E. Payne and W.C.K. Haskins, J. Am. Chem.
S o c , 110 (1988) 6240.
148. M-A. Pearsall and R. West, J. Am. Chem. S o c , 110 (1988) 7229.
149. C.B. Moore, J. Biedrzycki and F.W. Lampe, J. Amer. Chem. S o c , 106
(1984) 7 7 6 1 .
150. T. Dohmaru and F.W. Lampe, J. of Photochem., 41 (1988) 275.
1 5 1 . I.M.T. Davidson and R.J. Scampton, J. Organometal. Chem., 271 (1984)
249.