RESEARCH ARTICLE | DECEMBER 22 2004
The Infrared and Raman Spectra of Condensed Nitric Oxide
A. Lee Smith; William E. Keller; Herrick L. Johnston
J. Chem. Phys. 19, 189–192 (1951)
https://doi.org/10.1063/1.1748157
CrossMark
View Export
Online Citation
Articles You May Be Interested In
The vibrational spectra of molecular ions isolated in solid neon. IV. NO+, NO−, ONNO+, and ONNO−
J. Chem. Phys. (December 1990)
Downloaded from http://pubs.aip.org/aip/jcp/article-pdf/19/2/189/8111652/189_1_online.pdf
The 16 valence electronic states of nitric oxide dimer ( NO ) 2
J. Chem. Phys. (August 1998)
Ring structure of the NO dimer radical cation: A possible new assignment of the mysterious IR absorption
at 1424 cm−1
J. Chem. Phys. (November 2002)
SPECTRA OF CONDENSED NO 189
Table II indicates that further reactions of the chloro- the above mechanism. The decrease in yield of methane
methyl radical take place. The abstraction of a hydro- on adding carbon dioxide may be expected to be due to
gen atom from the chloromethyl radical by either a the quenching reaction
methyl radical or a chlorine atom, and subsequent
combination of the CHCI radical with methylene is a Hg(63Pl)+ CO 2= Hg(61So)+ CO 2 (8)
possible source of vinyl chloride at low pressures. Such in competition with reaction (1). From reactions (1),
reactions would not alter the quantum yields of methane (2), and (8) by the usual stationary state treatment we
and hydrogen chloride. Vinyl chloride may also be obtain
formed at low pressures by reactions subsequent on
quenching by ethylene. Further experiments would be
(1)
required to test the verity of these suggestions, but it
seems fairly certain that reactions (1) to (7) describe
the main course of the decomposition at high pressures. where RCH4 represents the rate of methane production,
According to the above mechanism, 1/JC2H4C12+I/JC2H4 kl and ks are the velocity constants for reactions (1)
should be unity at high pressures. Owing to a fairly and (8), and the terms in parentheses denote concentra-
wide scatter in the experimental results for ethylene, it tions. A linear relationship should therefore obtain
is not possible to estimate the quantum yield of this between 1/RCH4 and (C0 2) / (CHaCl). This plot is shown
product exactly at high pressures. Taking, however, in Fig. 6. From the slope of the line and the value of the
Downloaded from http://pubs.aip.org/aip/jcp/article-pdf/19/2/189/8111652/189_1_online.pdf
a mean value of approximately 0.25 for this quantity, intercept at (C0 2)/(CH aCI)=0, we obtain a value of
we find a value of 0.92 for the sum of the above quan- 1/2.19 for the ratio ks/k 1•
tum yields at 600-mm pressure. Further support for Using this result, we may obtain an approximate
the above mechanism comes from the low experimental value for the quenching cross section of methyl chloride.
activation energies found for methane and hydrogen This works out to be S.7XIQ-16 cm 2, which seems a
chloride. The temperature coefficients for the rates of reasonable value.
production of these two products should be zero by the Acknowledgment.-We are indebted to Dr. F. P.
above mechanism, and actually are very low. Lossing and Miss Frances Gauthier for the mass spec-
The results in Table V, showing the effect of added trometer analyses and to Dr. R. N. Jones for the infra-
carbon dioxide on the reaction, also lend support to red spectra.
THE JOURNAL OF CHEMICAL PHYSICS VOL U MEl 9. N U M B (R 2 FEBRUARY. 1951
The Infrared and Raman Spectra of Condensed Nitric Oxide*
A. LEE SMITH, WILLIAM E. KELLER, AND HERRICK L. JOHNSTON
Cryogenic Laboratory and Department of Chemistry, The Ohio State University, Columbtts, Ohio
(Received October 2, 1950)
The infrared and Raman spectra of liquid and solid nitric oxide have been studied in an attempt to prove
the existence of the (NO). molecule. The liquid has four strong Raman lines at 1861, 262, 196, and 167 em-I;
two weaker Raman lines at 1760 and 487 em-I; and two strong infrared bands at 1863 and 1770 em-I. Several
weaker infrared bands can be satisfactorily assigned as combinations by using the observed strong infrared
and Raman frequencies. The spectra were found to be due exclusively to a dimer which probably exists as a
bent ONNO molecule.
INTRODUCTION 112.8°K show excellent agreement with theory, indi-
cating that dimerization of the gas is negligible at
N ITRIC oxide is unique among stable diatomic
molecules in that it possesses an odd electron. temperatures below the boiling point (121.36°K). This
conclusion is substantiated by the findings of Johnston
This electron indicates that a moderately stable dimer
might form. According to Pauling, l however, a three- and Weimer,3 who showed from data of state that
electron bond stabilizes the NO molecule to such an association at the boiling point is less than 0.5 percent.
extent that the heat of dimerization becomes negative. It is fairly well established, however, that nitric oxide
Data on the magnetic susceptibility of the gas2 to molecules in the liquid state are polymerized. This is
* This work was supported in part by the ONR under contract inferred from the unusually high and strongly tempera-
with the Ohio State University Research Foundation. ture-dependent heat capacity of the liquid as well as
1 Pauling, Nature of the Chemical Bond (Cornell University
Press, Ithaca, N. Y., 1945), p. 266.
2 Wiersma, De Haas, and Capel, Leid. Comm. 212b. 3 Johnston and Weimer, J. Am. Chern. Soc. 56, 625 (1934).
190 SMITH, KELLER, AND JOHNSTON
from its high entropy of vaporization,4 from its mag- EXPERIMENTAL
netic rotatory power,!; and from its magnetic suscep- Nitric oxide was prepared according to the method of
tibility.6 The degree of association, calculated from this Johnston and Giauque, and our product should contain
magnetic data, is found to be of the order of 95 percent. less than 0.001 M percent impurity. Infrared spectra
Polymerization is further evidenced by data of state on were obtained with a model 12-B Perkin-Elmer spec-
compressed gases 7• 8 and by the appearance of an trometer, using a variable thickness low temperature
anomalous frequency (260 cm-I ) in the Raman spec- transmission cell. IS Sample thicknesses ranged from
trum of the liquid. 9 0.05 to 1.5 mm. The entire region from 400 to 6000 cm- I
The absorption spectrum of liquid nitric oxide has was investigated, using LiF, NaCI, and KBr prisms.
been studied in the visible and ultraviolet regions of Careful corrections were made for water vapor absorp-
the spectrum by Vodarlo and by Bernstein and Herz- tion.
berg,!1 who found strong continuous absorption from Raman spectra were obtained with a Lane-Wells
4000A to shorter wavelengths and from 5600A to longer spectrograph and power supply, using a specill-l low
wavelengths, with maximum transmission near 46ooA. temperature cryostatl 4 in which the sample could be
The spectrum was not temperature dependent. It was maintained at any temperature between 20° and 300oK.
suggested that the continuous absorption is due to the The Hg 4358 line was used for excitation, since liquid
molecule (NO)2' nitric oxide is transparent in the region 4000 to 5600A.
Downloaded from http://pubs.aip.org/aip/jcp/article-pdf/19/2/189/8111652/189_1_online.pdf
Rice,!2 on the basis of a thermodynamic treatment, The method of polarized incident light was employed
concludes that the liquid is almost completely asso- for depolarization measurements.
ciated, and calculates the heat of dissociation to be 3910
and 2870 cal per M of (NOh in the liquid and gas, RESULTS
respectively. This result agrees quite well with a heat
Gaseous nitric oxide was found to give the now well-
of dissociation of 4000 cal calculated from other
known infrared vibration-rotation spectrum consisting
sources. s It is interesting to note that Johnston and
of P, Q, and R branches, with the rotational lines nicely
Giauque,4 in their paper on the heat capacity of solid
resolved and the band center at 1876 cm-l (Fig. 1)
and liquid nitric oxide, report a discrepancy of ! Rln2
in agreement with the data of Nielsen and Gordy.lii
between the observed and calculated entropy. They Spectra of the liquid and solid in this region are shown
suggest that this difference could be accounted for if in Fig. 2. It is seen that a considerable change takes
the solid were a mixture of two polymerized isomers in place in the spectrum upon passing to the liquid, with
equal amount. the 1876-cm-1 band being replaced by two intense bands
In the present investigation, we have observed the at 1770 and 1863 cm-l. Several weaker bands can also
infrared and Raman spectra of the liquid and solid in be seen. The solid shows even more complication, and
an attempt to find direct evidence for the dimer and to while the 1862-cm-1 absorption has sharpened, the
deduce its structure. 1770-cm-1 band has apparently split and some new
bands have appeared. The first vibrational overtones,
shown in Fig. 3, consist of a very broad general absorp-
tion with three sharp bands at the center, in both the
liquid and the solid.
A microphotometer tracing of the Raman spectrum
FIG. 1. Infrared spectrum of NO gas.
4 Johnston and Giauque, J. Am. Chern. Soc. 51,3194 (1929).
• Bizette and Tsai, Compt. rend. 204, 1638 (1937).
6 Bizette and Tsai, Compt. rend. 206, 1288 (1936).
7 Duclaux, J. Phys. Radium [vii] 8,277 (1937).
8 Eucken and d'Or, Nachr. Gesell. Wiss. Gottingen, Math-
physik. Kl. (1932), p. 107. FIG. 2. Infrared absorption spectra of liquid and solid nitric oxide.
9 Vodar, Jardillier, and Mayence, Compt. rend. 222, 1343
(1946). 13 H'Olden, Taylor, and Johnston, J. Opt. Soc. Am. 40,757 (1950).
10 Vodar, Compt. rend. 204, 1467 (1937). 14 Taylor, Smith, and Johnston, J. Opt. Soc. Am. (to be pub-
11 Bernstein and Herzberg, J. Chern. Phys. 15, 77 (1947). lished).
12 Rice, J. Chern. Phys. 4, 367 (1936). 16 Nielsen and Gordy, Phys. Rev. 56, 781 (1939).
SPECTRA OF CONDENSED NO 191
FIG. 3. Vibrational overtones in liquid and solid nitric oxide. FIG. 4. Microphotometer tracing of the Raman spectrum of
of liquid nitric oxide.
of the liquid is shown in Fig. 4. Four lines can be seen
definitely, the most intense at 1861 cm-I, and three impurity. Second, the polymer could contain two, three,
somewhat weaker lines at low frequencies. With a four, or more NO molecules. The spectra, however, are
Downloaded from http://pubs.aip.org/aip/jcp/article-pdf/19/2/189/8111652/189_1_online.pdf
64-hr exposure, very weak lines seem to be present at believed to be exclusively those of the diiner since the
about 486 and 1760 cm-I. The solid showed no trace frequencies do not coincide with those of any logical
of a Raman spectrum even with exposures sufficiently foreign molecules nor with that of the gas. Furthermore,
long to record the four most intense lines of the liquid. the preparation and purification procedures are known
Table I shows a summary of the observed frequen- to be effective. The low value of the magnetic suscep-
cies. The strong bands at 1770 and 1863 em-I can be tibility of the liquid precludes the existence of a mono-
assigned immediately as fundamentals, and the bands mer or trimer, and it would be unusual if a tetramer
around 3600 em-I are evidently overtones or combina- were to have only five observable fundamental fre-
tions of the two strong fundamentals. It is important quencies. We cannot eliminate the possibility of the
to note that the entire infrared spectrum of the liquid dimer existing with two configurations, but if this is the
can be accounted for in terms of the two strong infrared case, it seems unlikely that the structures would be
frequencies and the frequencies observed in the Raman radically different. There is no marked temperature
spectrum. dependence of the band intensities in the liquid (or in the
solid above 40 0 K), indicating that the liquid does not
DISCUSSION AND INTERPRETATION dissociate appreciably even near its boiling point.
Although various explanations could be considered In order to show the effect of temperature and of
for the observed spectra of the condensed nitric oxide, concentration on the degree of dimerization, infrared
only two are logically possible. First, more than one spectra of solutions of nitric oxide in liquid nitrogen
species of the molecule is present; the extraneous ma- and in liquid krypton were studied. Krypton was chosen
terial could be a monomer, a polymer with two or more because its liquid range (116-121°K) coincides in part
configurations, another oxide of nitrogen, or some other with that of nitric oxide. The pure rare gas was obtained
TABLE 1. Frequencies in cm-1 of infrared bands and Raman lines in condensed nitric oxide.
Band Liquid Solid Band Liquid Solid
No. I.R. Raman I.R. Assignment No. I.R. Raman I.R. Assignment
1 ~5390*V.W. ~5390*V.W. 2F1+F2 13 (1783) 1795 S.
2 ~5200*V.W. ~5200*V.W. 2F2+F 1 14 1770 V.S. 1760 V.W.(?) 1760 S. F2
and/or
3F2
3 3708 M. 3712 M. 2Fl 15 1745 V.W.(?)
4 3600 S. 3596 M. F1+F2 16 1715 M.
5 3515 M. 3507 M. 2F2 17 1600 M. 1586 V.W. F1-F,
and/or
F 2-F.
6 2124 W. 2130 W. F,+F, 18 1545 V.W.(?)
7 2032 M. 2038 W. F,+F. 19 1506 W. F 2-F,
and/rj/
F2+ 3 20 519 V.W. 2F3
8 1965 M. 1962 M. F2+F. 21 487 V.W.(?) 486 V.W.
9 1918 M. F2+F. 22 262 S. (P) F.
10 1863 V.S. 1861 S. (P) 1865 V.S. F, 23 196W. F.
11 1850 W. 24 167 M. F.
12 1822 S.
* Calibration uncertain to about ±30 em-l ,
192 SMITH, KELLER, AND JOHNSTON
from the Matheson Company (Joliet, Illinois) and used all of which occur with appreciable intensity, lend
without further purification. The mole fraction of NO support to this conclusion.
in krypton was 0.017, and in nitrogen it was slightly 2. The frequencies of the stretching modes are
larger. The results are shown in Fig. 5. It can be seen ~ufficiently high to indicate the presence of at least a
that at a temperature of 117°K, the diluted nitric oxide double bond between the 0 and the N. This require-
is partially dissociated into ·the monomer which is ment, plus the difficulty of writing electronic configura-
vibrating with a frequency of 1872 cm-I . At the tem- tions, eliminates all NOON structures as well as the
perature of liquid nitrogen, almost complete association triangular and square models.
3. The NO NO type structures are somewhat ques-
tionable for reasons given in the preceding paragraph.
Furthermore, nitrogen and oxygen are enough alike so
that one would expect the in-phase and out-of-phase
stretchings to have very different intensities in the
infrared, but this is not observed. It is true, however,
that while these considerations weaken the case for the
NONO linkage, they do not exclude it. This configura-
tion, while a possibility, is somewhat less probable than
Downloaded from http://pubs.aip.org/aip/jcp/article-pdf/19/2/189/8111652/189_1_online.pdf
those considered subsequently.
Three possibilities remain, all of the bent ONNO
type. The models in which the ONN bond angles are
T, 117"K (KRYPTON SO~VENT) different (point groups C. and CI) appear to be highly
~O~--~------------------------~~~
unlikely, so we are left with the two bent structures
belonging to point groups C2 and C 2v , which is what we
expect from valence considerations anyway. It is diffi-
80 cut to choose between these two structures.
Frequency assignments for the liquid can be partially
10
made on the basis of available evidence, but the fact
that one fundamental has not been found makes a com-
plete assignment impossible. Since the 1863-cm-1 line is
40
polarized, it must be of species A and thus arises from
the symmetrical NO stretching. This leaves v= 1770
20 cm-I as caused by the asymmetric stretch.'6
T' 7'". (NITROGEN SO~VENT)
Not much can be said concerning the low vibrational
frequencies (in view of their proximity) other than the
0~1~70~O--~~~~~'~80~O~~~~~~"~O~O~~~
fact that the polarized 262-cm-' line arises from a
FREQUENCY (em"')
symmetrical motion. The 487-cm-' frequency is diffi-
FIG. 5. Spectra of nitric oxide solutions. cult to explain; it may be the sixth fundamental but it
apparently does not combine with the other funda-
has occurred because of the change of the equilibrium mentals, and yet it cannot be satisfactorily assigned as a
constant with temperature. combination or overtone.
If bonding considerations are disregarded for the A possibility which has not been dealt with is that
time being, some twenty-odd possible structures for the low-lying electronic levels may be present; the data
dimer can be drawn. The following observations allow are insufficient to settle this point.
us to eliminate many of these models:
16 This band behaves in an interesting fashion in the diluted
1. The coincidence of the same strong infrared and liquid. It apparently is split into two components, one at 1770
Raman band at 1860 cm-I makes a structure with a cm-I and the other at about 1785 em-I. It is unlikely that these
center of symmetry unlikely. The appearance of the are two different fundamentals, and the splitting may be due to
the existence of two kinds of dimer molecules. It appears that a
1770 and 1860 overtones and of the combination of two further splitting occurs in the solid, giving rise to the bands· at
Raman polarized bands (1865+ 262) in the infrared, 1760 and 1795 em-I.
