Materials Today Communications 4 (2015) 93–100

Contents lists available at ScienceDirect

Materials Today Communications

journal homepage: www.elsevier.com/locate/mtcomm

Swelling behavior and network characterization of hydrogels from

linear polyacrylamide crosslinked with glutaraldehyde
Ivan Dmitriev ∗ , Ivan Kuryndin, Natalya Bobrova, Michael Smirnov
Institute of Macromolecular Compounds, Russian Academy of Sciences, Bolshoi pr. 31, Saint Petersburg 199004, Russia

a r t i c l e i n f o a b s t r a c t

Article history: Polyacrylamide hydrogels have been prepared by crosslinking reaction between linear polyacrylamide
Received 15 May 2015 (PAAm) and glutaraldehyde (GA) in acidic medium. The equilibrium swelling degree, modulus of elastic-
Accepted 1 June 2015 ity and structure parameters of the network have been investigated as functions of the initial polymer
Available online 10 June 2015
concentration and the crosslinker content. Stability of GA crosslinks between PAAm chains has been
demonstrated under the acid and basic conditions. It has been suggested that the crosslinking reaction
can be presented as two stage process, which distinguishes different accessibility of aldehyde and amide
groups for cross-linking agent in conditions of macromolecule distancing. The method provides forma-
tion of PAAm hydrogels with the values of the swelling degree and mechanical characteristics, which are
close to the hydrogels prepared by traditional method of free radical crosslinking copolymerization.
© 2015 Published by Elsevier Ltd.

1. Introduction polymers being electroactive can easily react with ox-red initiator
systems which are usually used for radical polymerization.
Hydrogels on the base of crosslinked polymers are perspec- The limitations of the discussed procedures can be overcome
tive stimuli responsible materials since they are sensitive to the using the method which is the main pathway for preparation of
electric field [1], temperature [2], light [3] and chemical compo- hydrogels from natural polymers such as chitosan [12] or cellu-
sition of the environment [4]. Encapsulation of active particles lose [13]. The solution containing both linear polymer and filler
or molecules, for example, conducting polymers [5,6,7] inside the can be chemically crosslinked by the reaction of functional groups
hydrogel is one of the most promising ways to impart them new of the polymer and crosslinker agent. Despite long-standing his-
«smart» properties. This work was motivated by practical task to get tory of investigations in the case of polyacrylamide this approach
polyacrilamide/conducting polymers hydrogels. Crosslinked poly- has not developed yet. In this work, the crosslinking of linear PAAm
acrylamide (PAAm) is accounted to be very attractive system for in aqueous solution with glutaraldehyde (GA) was investigated
composites preparation [8,9,10]. However, different approaches with classical physical characterization. GA easily reacts with amide
have been used to obtain hydrogel-based composites with uni- groups and also it is cheap and nontoxic substance. It was shown
form distribution of an active component [11]. One of them is by other authors that chemical crosslinking reaction between GA
sorption of the filler particles into the pre-synthesized hydrogel and amine or hydroxyl groups is very useful in constructing func-
sample with or without subsequent chemical reactions in situ. Main tional polymeric materials from proteins, aminopolysaccharides
disadvantage of this method is slow diffusion of reagents which [12], and other natural [14] and synthetic polymers [15]. In spite of
causes non-homogeneous distribution of filler inside the gel espe- the poor basicity of amide group in comparison with amine one the
cially in the case of chemical reactions performed in situ. Another GA was shown to be effective crosslinker for PAAm in acidic solu-
method is the three-dimensional radical copolymerization of acry- tions [16]. The mechanism of the reaction of GA with amino groups
lamide and bifunctional crosslinker in the solution or dispersion was shown to be quite complex [12,17–20]. It involves oligomer-
of the active component. This process can be realized only if the ization of GA molecules by the aldol condensation which is induced
filler does not react with both acrylamide molecules and initiator by imine of Schiff base formed between one GA molecule and one
of acrylamide polymerization. For example, particles of conducting amino group of the polymer [20]. The final crosslinked structure is
a linear oligomer of GA with several linkages branching off from it.
We can expect the same mechanism for the crosslinking of PAAm
which contains amide groups (Fig. 1). However, the influence of
ratio between quantity of GA and amide groups (␯(GA):␯(AAm)) on
∗ Corresponding author. the properties of hydrogel is not clear and will be investigated in
E-mail address: [email protected] (I. Dmitriev).

http://dx.doi.org/10.1016/j.mtcomm.2015.06.005
2352-4928/© 2015 Published by Elsevier Ltd.
94 I. Dmitriev et al. / Materials Today Communications 4 (2015) 93–100

CONH2

CONH2
O NH2
CONH2

OHC (CH2)3 CHO

H+

H 2O

CONH2
CONH2

H+ CONH2
CONH2 O N
O N CONH2
CONH2 O
OHC (CH2)3 CHO H2O
H
O H
O H

H2NOC

CONH2 CONH2 CONH2

CONH2
O
N
CONH2 H
O N

O O
H

H O
N
O
CONH2
O N CONH2 CONH2

CONH2 CONH2 CONH2 CONH2

H2NOC

Fig. 1. Crosslinking of PAAm by GA.

this work in wide range of value: from 1:2 to 1:200. Authors believe following procedure. Solution of 18.76 g (0.264 mol) of acrylamide
that the novelty of this work also lies in determination of opti- in 500 ml of distilled water was degassed under vacuum. Then
mal ratio between swelling properties and mechanical properties 0.456 g (2 mmol) of (NH4 )2 S2 O8 and 0.34 ml (2 mmol) of N,N,N ,N -
of PAAm/GA hydrogels. tetramethylethylenediamine was added and the reaction mixture
was stirred at 40 ◦ C during 2 h under flow of nitrogen. PAAm was
2. Experimental part precipitated in 2.5 l of methanol and collected and dried under vac-
uum. As it was found by the measurements of the characteristic
2.1. Hydrogel preparation viscosity at different concentrations, the molecular weight of PAAm
(M␩ ) is 264,000 and the overlapping polymer volume fraction (ϕ*)
PAAm was synthesized by radical polymerization of acry- is 0.005.
lamide (Aldrich) in 3.75% water solution of monomer by the
 I. Dmitriev et al. / Materials Today Communications 4 (2015) 93–100 95

It was shown by the primary tests that a minimum mass concen-

tration of PAAm (␻) required to form a gel is 5%, which corresponds
to the polymer volume fraction ϕ = 0.04. Hydrogels were prepared
by crosslinking of PAAm in water solution with GA (25% solution
purchased from Vekton, Russia). Mass content of polymer ␻(PAAm)
was varied in the range of 5–20% and amount of crosslinker solution
was determined according to desired molar ratio between aldehyde
and amide groups: ␯(GA):␯(AAm) = 1:200; 1:100; 1:50; 1:10; 1:2.
Then concentrated hydrochloric acid 0.2 ml (0.238 g) was added to
the mixture which was placed in the test tube and was kept at 40 ◦ C
for 5–72 h.
The following variables were used for defining the composition
of reaction mixture during gel preparation:
- volume fraction of the linear PAAm:
Q0 2 −1
ϕ20 = (1 + ) , (1)
1 Fig. 2. Stress-strain data for PAAm hydrogels just after their preparation at f = 1 (1),
0.2 (2), 0.04 (3), 0.02 (4), 0.01 (5).
- relation between quantity of aldehyde and amide groups:
2v(GA) weight loss during the measurements, due to water evaporation,
f = , (2)
v(AAm) was negligible.
where Q0 = 1/␻ − 1 is the swelling degree of hydrogel just after The elastic modulus was determined from the slope of linear
preparation which depends only on set polymer concentration; 1 dependence:
and 2 are the densities of solvent (water) and PAAm, which were −2
␴ = G(␭ − ␭ ) (4)
taken as 1 and 1.302 g/ml, respectively.
Because of equivalent influence of the values ϕ20 and f on the where ␴ is the force acting per unit cross-sectional area of the unde-
equilibrium swelling degree and crosslinking density of hydrogels formed gel specimen and ␭ is the deformation ratio (deformed
[21] two sets of the variable parameters were chosen: length/initial length). The elastic modulus measurements were
restricted to the region of low strain within which the variation G
1. in the range of ϕ20 = 0.04/0.16 (␻ = 5/20%) at f = 0.2; and (␭ − ␭−2 ) was accurately linear. Typical stress–strain data are
2. at ϕ20 = 0.08 (␻ = 10%) in the range of f = 0.01/1. shown in Fig. 2.

2.2. Swelling measurements 3. Results and discussion

Swelling degree (Q) was measured on disk-shaped hydrogels 3.1. Formation and equilibrium swelling of hydrogels
12 mm in diameter and 5 mm height which were placed in distilled
water or solutions of HCl or NaOH with different pH at room tem- It follows from the supposed mechanism of crosslinking (Fig. 1)
perature for 10–20 days. The variation of mass of a swelling sample that the reaction between GA and PAAm leads to the addition
(ms ) with time was determined and Q was calculated by: of pendant N CH (CH2 )3 CHO groups to the polymer chain.
According to literature [20] the next step is the oligomerisation
ms − md
Q = (3) of GA activated by the first linked molecule forming grafting car-
md
bon chain with several aldehyde groups connected to it. After
where md is the mass of the sample after complete drying in vac- 100–150 min of the reaction course the viscosity of mixture starts
uum. The equilibrium swelling capacities of the gels were measured increasing because of crosslinks formation. Then solution loses its
at different stages of the experimental work and in the text they fluidity due to continuous polymer network appearance. After that
are marked as follows: Q – swelling degree after synthesis; Q – the hydrogel sample can be pulled out from the reaction vessel and
reswelling degree after drying of the sample. its swelling degree (Q) can be measured in dependence of their
The interpretations of the swelling experiments were made in preparation time. The results are shown in Fig. 3a. At the begin-
terms of the volume fraction of crosslinked polymer after hydro- ning of curves in Fig. 3a value of Q rapidly decreases in result of
gel preparation ϕ20 and at equilibrium swelling of the hydrodel ϕ2 , further crosslinking and then the drop of Q practically stops. The
which was calculated according to Eq. (1) replacing Q0 with Q. final values of Q are approximately 12, 40 and 40 g/g for samples
with f = 0.04, 0.02, 0.01, respectively.
2.3. Mechanical measurements Prepared samples were carefully washed with large amount of
water. The absence of residual free GA in hydrogel was proved by
Uniaxial compression experiments were performed on the gels the test with Fehling’s solution. Then the samples were dried in
as prepared and after their equilibrium swelling in water. The vacuum until fixed weight was reached and immersed in water
stress-strain curves were measured using R-5 testing machine again. As it is shown in Fig. 3b reswelling ability (Q ) for the net-
(Tochpribor, Ivanovo, Russia). All the mechanical measurements works depends on the crosslinking duration. The finite value of Q
were carried out at room temperature on the disk samples with (approximately 10 g/g) was lower than Q for the most of the pre-
the same shape as for the swelling experiments. The speed of the pared hydrogels. So the network density became higher after the
parallel crossheads displacement was 1 mm/min. Reversibility of drying of the samples. It is obvious that when sample loses retained
the stress–strain curves was tested by recording the cycles of the water the PAAm chains becomes closer to each other making easier
consecutive load-reload acts up to about 20% compression. The the formation of additional crosslinks. Thus, one needs to assume
loading-reloading processes yielded almost identical stress–strain the existence of aldehyde groups which are connected to the poly-
relationships. From 4 to 5 repeated measurements, the standard mer chain but not involved yet at the crosslinks in the hydrogel just
deviations in the modulus value were less than 5%. The sample after preparation.
96 I. Dmitriev et al. / Materials Today Communications 4 (2015) 93–100

Q, g/g Q', g/g

3
160 20 3
b
2 2
120 a 15

80 1
10

40 1 5

0 24 48 72 t, h 0 24 48 72
t, h

Fig. 3. Dependences of swelling degree of PAAm hydrogels in water as prepared (a) and reswelling after drying (b) on duration of crosslinking reaction at ϕ20 = 0.08, f = 0.04
(1), 0.02 (2), 0.01 (3).

Table 1 Fig. 3 makes it possible to conclude that the fraction of the

Dependence of the swelling degree on pH for the hydrogel, prepared at ϕ20 = 0.08,
unreacted aldehyde groups in the equilibrium swollen samples
f = 0.04, during 1 day.
decreases with increasing of the crosslinker concentration. At
pH 1.0 2.1 3.0 7.0 9.8 12.1 13.6 f ≥ 0.04, the effect of «conservation» of the unreacted groups disap-
Q, g/g 10.3 11.2 11.5 11.4 76 110 50
pears and we have expectable situation when Q ≈ Q. The influence
of the initial polymer concentration (ϕ20 ) and parameter f on the
Table 2 values Q and Q is demonstrated in Fig. 4a and b. It is shown that
Elastic modulus (G), stress at break (␴b ) and break compression (b ) for the samples the increase of the values ϕ20 or/and f leads to the decrease of both
prepared at ϕ20 = 0.08, f = 0.04 during various time (t). Q and Q . When ϕ20 and f becomes higher than 0.08 and 0.2, respec-
t, h G, kPa ␴b , kPa b , % tively, the equilibrium swelling degree drops even below than the
Q0 . Actually the condition Q < Q0 supposes contraction of the gel
12 7 184 80
24 7 135 81
phase during the formation of the polymer network and this phe-
36 11 94 86 nomenon is known as macrosyneresis [22]. It occurs in the situation
when contraction forces which arise from formation of cross-links
exceed the expansion forces of the gel swelling. According to Fig. 4a,
The dependence of swelling degree on pH of water solution for decreasing of the ϕ20 allows to avoid the phase separation and reach
hydrogels prepared at ␻(PAAm) = 10% and ␯(GA):␯(AAm) = 1:100 the condition Q > Q0 (in the region of small values of ϕ20 ). It should
during 1 day was determined (Table 1). The swelling degree is be assumed that a critical value of crosslinking density which is
approximately the same for the samples in the acid pH range. responsible for the gel contraction exists. If the crosslinking den-
This observation points out that the crosslinks formed from GA sity is higher than the critical value, a pure solvent phase is pushed
molecules are chemically stable in acidic conditions. In basic con- out from the network and the gel phase shrinks [22].
ditions dependence of the Q from pH shows the strong peak with Thus, the cross-linking reaction of PAAm/GA gels demonstrates
maximum near pH 12. Hydrogels kept their integrity even at max- two features: storing of the unreacted aldehyde groups at low val-
imum swelling degree. So the crosslinks are stable at the basic ues of f and ϕ20 and the phase separation at high values of f and ϕ20 .
conditions also. The reason for increasing of Q at pH from 7 to 12 is From practical point of view, it was interesting to find crosslinking
the hydrolysis of the amide groups of PAAm with formation of the parameters which allows to obtain PAAm network sample with
carboxylate groups. This reaction converts the hydrogel from non- Q > Q0 , Q = Q and maximum possible value of Q. The suitable con-
ionic to polyelectrolyte one because of osmotic pressure increase ditions for these requirements have been found as f = 0.04 and
in the samples volume. Decreasing of Q at pH > 12 takes place for ϕ20 = 0.08 (␻ = 10%) at Q = 11 g/g.
the reason that a concentration of alkali becomes high enough to
screen electric charges on the polyelectrolyte chain and to level
osmotic pressure of dissociated COONa groups. So both the factors
providing growth of the swelling degree are suppressed. 3.2. Analysis of the PAAm/GA hydrogels structure on the base of
The effect of the polymerization time on the structure of the mechanical measurements
samples was studied by measurements of their mechanical char-
acteristics during compression. It is seen in Table 2 that elastic In this subsection, we will discuss the effects of ϕ20 and f upon the
modulus is higher for the sample crosslinked during 36 hour than equilibrium swelling behavior, elasticity and some related struc-
for ones prepared in the shorter reaction period. At the same time, ture parameters of the hydrogels. In Fig. 5, the modulus of elasticity
the values of relative deformation and stress at break become lower of the as prepared hydrogels G0 and the hydrogels after equilibrium
with increasing of the reaction duration. swelling G are shown as functions of the initial polymer (PAAm)
According to the presented data the crosslinking reaction of volume fraction and the cross-linker (GA) concentration. It is seen,
PAAm with GA can described as two stage process: that the crosslinking parameters ϕ20 and f have a strong influence
Stage I: Covalent binding of GA molecules to the PAAm and on the mechanical response of the hydrogels. It is obvious, that the
oligomerization of dialdehyde. As a result PAAm with grafted oligo- increasing in G0 , G (Fig. 5) and decreasing in Q, Q , Q0 (Fig. 3) must
GA side chains is formed. Reaction of the most available amide be a direct consequence of the crosslinking density growth.
groups with GA oligomers with formation of weakly crosslinked The effective network chain density e can be calculated from
polymer network. the compression modulus G using the following equation [23]:
Stage II: Formation of additional junctions by the reaction of
the less available aldehyde and amid groups. This stage can be 2/3
accelerated under the contraction (deswelling) of the hydrogel. G = Ave RT (ϕ20 ) (ϕ2 )2/3 , (5)
 I. Dmitriev et al. / Materials Today Communications 4 (2015) 93–100 97

Fig. 4. (a) Dependences of Q (1), Q (2) and Q0 of the linear PAAm (3) on the volume fracture of crosslinked polymer after the gel preparation at f = 0.2. (b). Dependences of
equilibrium swelling degree of PAAm hydrogels in water as prepared on relation between quantity of aldehyde and amide groups.

Fig. 5. (a) Dependences of the elasticity modulus of as prepared PAAm hydrogels G0 (1) and after their equilibrium swelling G (2) on the volume fraction of crosslinked
polymer (f = 0.2). (b) Dependences of the elasticity modulus of as prepared PAAm hydrogels G0 (1) and after their equilibrium swelling G (2) on the crosslinker concentration
f ϕ20 = 0.08.

where R is the gas constant, T is absolute temperature, A is the

coefficient, which equals to 1 for an affine network and 1–2/␺ for a
phantom network, and ␺ is the functionality of the crosslinks.
For the hydrogels just after preparation, the modulus G0 can be
expressed as follows:

G0 = Ave RT ␸02 . (6)

The theoretical basis of Eqs. (5) and (6) assumes a network

of Gaussian chains. Applicability of these relations to PAAm/GA
hydrogels was examined by the approach introduced by authors
[24,25]. From Eqs. (5) and (6) the ratio of the elastic modulus of
the gel at a given swelling degree to that of the same gel after its
preparation can be given by:
 1/3
G ϕ2
= . (7) Fig. 6. The reduced modulus G/G0 of the hydrogels as a function of the normalized
G0 ϕ20
polymer concentration ␸2 /␸02 .

All the points from Figs. 4 and 5 are represented in accordance

with Eq. (7) in logarithmical coordinates and collected in Fig. 6. The
slope of the dependence is −0.327 that is close to the theoretical
value of −1/3. This proves that PAAm/GA hydrogels behave as a swelling degree (Fig. 3) we provided the calculations both at A = 1
network of Gaussian chains and we can use Eqs. (5) and (6) for (affine network) and at A = 0.5 (phantom network with function-
calculation of ␯e ality ␺ = 4). It should be noted also that we can only hypothesize
Eqs. (5) and (6) can be based on two different assumptions: about actual chemical structure of a crosslink point and about its
that the crosslinks behave either according to a fully mobile phan- effective functionality. Considering the case with ␺ = 4 we imply
tom model (when A = 0.5 for a crosslinking agent with functionality that a single GA molecule makes a singular crosslink but it is hardly
␺ = 4) or according to an immobile affine model (when A = 1). It acceptable taking into account the more complicated structure of
should be expected that not all the crosslinks deform affinely in crosslinking unit proposed above (Fig. 1). It should be expected that
real networks but steric restrictions are strong enough to prevent the more probable case is ␺ > 4 when the effective value of A is
full mobility. Thus, the character of deformation of real network in the range 0.5 < A < 1. Though we know nothing about a singular
is between these extreme situations. Since, the gels prepared in crosslink structure, we suppose that the calculation at ␺ = 4 can be
this study are characterized by a wide range of the values of the useful to analyze the crosslinking efficiency.
98 I. Dmitriev et al. / Materials Today Communications 4 (2015) 93–100

Fig. 7. Dependences of theoretical ␯t (1) and effective ␯e (2, 3) crosslinking density for phantom (2) and affine (3) models of PAAm hydrogels on the polymer volume fraction
after the gel preparation at f=0.2 (a) and on the crosslinker concentration at ϕ20 = 0.08 (b).

Table 3
Parameters ␣ and ␤ from Eq. (10) (correlation factor R2 = 0.993).

Network ␣ (mol/m3 ) ␤

Affine 21.4 0.25

Phantom 42.8 0.50

where ␣ is the portion of effective crosslinks which exists in linear

polymer solution even in the absence of any chemical crosslinker
and ␤ is a parameter of crosslinking efficiency which equals e /t
when ␣ = 0.
Table 3 shows ␣ and ␤ values calculated for both phantom
and affine network. The magnitude ␣ denotes physical, noncova-
Fig. 8. Dependence of the effective crosslinking density on the theoretical one cal- lent crosslinks (e.g., entanglements) contributing to the measured
culated from phantom (1) and affine (2) network for PAAm hydrogels prepared at crosslinks number. It can be definitely concluded that the non-
ϕ20 = 0.08 and at different f. chemical crosslinks remarkably influence on the effective crosslink
density of the PAAm hydrogels. According to the polymer solu-
The theoretical crosslinking density (t ) was calculated assum- tion entanglements conception [27], it should be suggested that
ing quantitative reaction and an ideal network formation from the these physical crosslinks are provided by chains in a semi-diluted
relationships [26]: entangled solution when the polymer concentration C20 , which
␺ corresponds to the volume polymer fraction ϕ20 , exceeds the entan-
Vt = C(GA) (8) gled polymer solution concentration Ce . The value Ce separates the
2
semi-delute unentangled and semi-delute entangled regimes of
fC(AAm) f ␻(PAAm) dissolution. A rough estimate of the entanglement concentration
C(GA) = = (9)
2 2Mr(AAm)100 can be given by [28].
where C(GA) and C(AAm) are the concentrations of GA and AAm 2 Me
correspondingly, Mr(AAm) is molar mass of AAm and  is density Ce ≈ (11)
M
of solution which was taken as density of water.
The dependencies of e and t on ϕ20 and f calculated from Eqs. where Me is the entanglement molecular weight of the polymer
(5) and (8) are given in Fig. 7. It is seen, that an increase in t leads to melt (for PAAm Me = 9100 from rheological data [29]). For the PAAm
an increase in e . The effective crosslink density e is very sensitive with M ≈ 264000 Eq. (11) gives Ce = 4.5 g/dl, which corresponds to
to the parameter ϕ20 (Fig. 7a) that may be explained by the decreas- ϕ2e = 0.035. This value is below the concentration range for PAAm
ing of the number of GA molecules wasted on reactions among gels preparation and the condition ϕ20 ϕ2e can be the reason for the
themselves making elastically non-effective groups at low values unusual experimental result in Fig. 7b when e > t for the hydrogels
of ϕ20 [25]. At the same time, e and t dependencies on f (Fig. 7b) prepared at ϕ20 = 0.08 and f < 0.2. The more detailed analysis needs
have a different character: at low values of f (f ≤ 0.04) the unusual to be supported by additional experiments and is beyond the scope
situation when e > ␯t is observed. of the present paper.
It is obvious that the real number of crosslinks differs from the The parameter ␤ is a measure of cross-linking efficiency and its
number of crosslinks that theoretically corresponds to the amount value is ≤1 [25]. Fig. 9 shows that ␤ reaches the maximum value
of crosslinking agent actually added to the initial solution of the at ϕ20 = 0.08 and doesn’t increase at further growth of ϕ20 . Increas-
linear polymer. Equality in the effective and theoretical crosslink- ing of the crosslinker concentration f does not allow to raise the
ing densities is statistically improbable. To clear up the role of the crosslinking efficiency due to the formation of elastically ineffective
crosslinker concentration f in the PAAm hydrogels properties, it is dangling ends and loops in the hydrogel [22,30]. At the same time, ␤
interesting to analyze the relationship between e on t . is almost independent from value of parameter f (Fig. 8) and equals
The dependences of e on t both for the phantom and for the 0.25 and 0.50 for the calculations based on the affine and phantom
affine network are shown in Fig. 8. They have linear modes and can models correspondingly (Table 3). In other words, 50–75% of the
be represented according to the equation [26]: crosslinker molecules approximately waste on elastically uneffec-
tive structures. It should be mentioned, for comparison’s sake, that
ve = ␣ + ␤vt (10) the equilibrium swelling degree Q and the derived network param-
 I. Dmitriev et al. / Materials Today Communications 4 (2015) 93–100 99

The experimental points in Fig. 10 can be represented by the

least squares as the straight line and an expression [25,26]:

 = 1 + 2 ϕ2 (13)

where ␹1 = 0.465 ± 0.002 and ␹2 = 0.497 ± 0.017. The component

␹1 corresponds to uncrosslinked linear PAAm at infinite dilution. It
should be noted, that the conditions ␹1 ≤ 1/2 and ␹2 > 1/3 are the
requirement for the collapse of nonionic gel [31]. Similar features
were also observed on the other hydrogels of PAAm family [25].

4. Conclusion

It has been shown, that PAAm hydrogels can be prepared

Fig. 9. The crosslinking efficiency ␤ shown as a function of the polymer network by crosslinking reaction between linear PAAm and GA in acidic
concentration ␸02 calculated from phantom (1) and affine (2) network for PAAm
medium. The equilibrium swelling degree, modulus of elasticity
hydrogels prepared at f = 0.2.
and structure parameters of the network derived from the rubber
elasticity theory were investigated. The swelling capacity of the gels
increases with decreasing concentration of GA or PAAm. The sta-
bility of GA cross-links between PAAm chains was demonstrated
under the acid and basic conditions. The effect of «conservation»
of unreacted side chains in the first act of equilibrium swelling
has been found at low values of the polymer volume fracture in
the solution. It has been suggested that the crosslinking reaction
can be presented as two stage process, which distinguishes differ-
ent accessibility of aldehyde and amide groups for cross-linking
agent in conditions of macromolecule distancing. The effect of
macrosyneresis was observed in highly crosslinked gels. At the
same time, this method provides formation of PAAm hydrogels
with the values of the swelling degree and mechanical charac-
teristics, which are close to the hydrogels prepared by traditional
method of free radical crosslinking copolymerization. The derived
Fig. 10. Dependence of the ␹ parameter on the volume fraction of polymer in the effective crosslinking density was about 25–50% of the theoreti-
PAAm gel.
cal (stoichiometric) value. It was found that the minimal polymer
concentration for the gel formation is near to the entanglement
concentration for PAAm.
eters e and ␤ of present gels are close to the same characteristics
for hydrogels prepared by the free radical cross-linking copolymer- Acknowledgement
ization of AAm with N,N -methylene-bis-acrylamide in aqueous
solution (at the similar values of monomer/crosslinker ratio and This work was supported by the Russian Foundation of Basic
volume fraction of monomer) [25]. The parameter ␤ is supposed to Research (Grant № 13-03-00219).
be increased by raising of the ϕ20 , however, it is not so interesting
from a practical point of view because of the macrosyneresys and
References
decreasing of Q value (Fig. 4a).
Values of the polymer–water interaction parameter ␹ were cal- [1] H.L. Lim, J.C. Chuang, T. Tran, A. Aung, G. Arya, S. Varghese, Adv. Funct. Mater.
culated from the Flory–Rehner expression for both an affine (A = 1) 21 (2011) 55.
network and a phantom tetrafunctional one (A = 0.5) [23]: [2] I. Ankareddi, C.S. Brazel, J. Appl. Polymer Sci. 120 (2011) 1597.
[3] Y. Yu, T. Ikeda, Macromol. Chem. Phys. 206 (2005) 1705.
[4] M.S.M. Eldin, H.M. El-Sherif, E.A. Soliman, A.A. Elzatahry, A.M. Omer, J. Appl.
2/3
Ave V1 (ϕ20 ) (ϕ2 )1/3 + ln(1 − ϕ2 ) + ϕ2 Polym. Sci. 122 (2011) 469.
=− (12) [5] M.A. Smirnov, N.V. Bobrova, I. Yu. Dmitriev, V. Bukolšek, G.K. Elyashevich,
ϕ22 Polym. Sci. Ser. A 53 (2011) 67.
[6] R. Kishi, K. Hiroki, T. Tominaga, K.-I. Sano, H. Okuzaki, J.G. Martinez, T.F. Otero,
Y. Osada, J. Polym. Sci. Polym. Phys. Ed. 50 (2012) 790.
where V1 is the molar volume of water (18.05 cm3 /mol). The results [7] G.K. Elyashevich, M.A. Smirnov, N.V. Bobrova, I. Yu. Dmitriev, V. Bukošek,
Smart Nanocompos. 3 (2012) 156.
of ␹ calculation plotted against the equilibrium volume fraction
[8] P.V. Vlasov, M.A. Smirnov, I. Yu. Dmitriev, N.N. Saprykina, G.K. Elyashevich,
ϕ2 are shown in Fig. 10. It is seen, that the value of ␹ increases Russ. J. Appl. Chem. 87 (2014) 490.
with ϕ2 increasing and ␹ has strong concentration dependence. The [9] Q. Tang, J. Wu, H. Sun, S. Fan, D. Hu, Carbohydr. Polym. 74 (2008) 215.
increase in ␹ means a decrease in the thermodynamic goodness of [10] E. Alveroglu, H. Sozeri, U. Kurtan, M. Senel, A. Baykal, J. Mol. Struct. 1036
(2013) 386.
solvent, which corresponds to an increase in effective crosslinking [11] P. Thoniyot, M.J. Tan, A.K. Anis, D.J. Young, X. Loh, J. Adv. Sci. 2 (2015) 1400010.
density. At the same time, GA cross-links consists from hydropho- [12] V. Crescenzi, A. Francescangeli, A. Taglienti, D. Capitani, L. Mannina,
bic chains (Fig. 1) and their number also decreases the compatibility Biomacromolecules 4 (2003) 1045.
[13] H. Kono, S. Fujita, Carbohydr. Polym. 87 (2012) 2582.
of water and the hydrogel. Value of ␹ = 0.499 at ϕ2 = 0.056 cor- [14] A. Bigi, G. Cojazzi, S. Panzavolta, K. Rubini, N. Roveri, Biomaterials 22 (2001)
responds to the hydrogel with e = 107 mol/m3 (for the model of 763.
phantom network) obtained at f = 0.04, ϕ20 = 0.08. Thus, water is [15] L. Zhang, P. Yu, Y. Luo, Sep. Purif. Technol. 52 (2006) 77.
[16] P. Lu, Y.-L. Hsieh, Polymer 50 (2009) 3670.
a thermodynamically good medium (␹ < 0.5) for PAAm/GA hydro- [17] P.M. Hardy, A.C. Nicholls, H.N. Rydon, J. Chem.Soc. Perkin Trans. 1 9 (1976)
gels prepared within the crosslinker content f < 0.04 and the initial 913.
polymer concentration ϕ20 0.08. [18] V.E. Fadouloglou, M. Kokkinidis, N.M. Glykos, Anal. Biochem. 373 (2008) 404.
100 I. Dmitriev et al. / Materials Today Communications 4 (2015) 93–100

[19] I. Migneault, C. Dartiguenave, M.J. Bertrand, K.C. Waldron, Biotechniques 37 [24] N. Orakdogen, O. Okay, Eur. Polym. J. 42 (2006) 955.
(2004) 790. [25] N. Gundogan, O. Okay, W. Opperman, Macromol. Chem. Phys. 205 (2004) 814.
[20] J. Kawahara, K. Ishikawa, T. Uchimaru, H. Takaya, et al., Chemical [26] W. Xue, M.B. Huglin, B. Liao, Eur. Polym. J. 43 (2007) 4355.
Cross-linking by Glutaraldehyde Between Amino Groups: Its Mechanism and [27] M. Rubinstein, R.H. Colby, Polymer Physics, Oxford University Press, 2003.
Effects. Polymer Modidfaction, in: G. Swift (Ed.), Plenum press, New-York, [28] E.D.V. Meerwall, E.J. Amis, J.D. Ferry, Macromolecules 18 (1985) 260.
1997, p. 119. [29] Y.H. Zang, P.J. Carreau, J Appl. Polym. Sci. 42 (1991) 1965.
[21] L. Bromberg, A. Yu. Grosberg, E.S. Matsuo, Y. Suzuki, T. Tanaka, J. Chem. Phys. [30] J.E. Mark, Adv. Polym. Sci. 6 (1982) 1.
106 (1997) 2906. [31] B. Erman, P.J. Flory, Macromolecules 19 (1986) 2342.
[22] K. Dusek, W. Prins, Adv. Polym. Sci. 6 (1969) 1.
[23] P.J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca,
NY, 1953.