View Article Online / Journal Homepage / Table of Contents for this issue

PAPER www.rsc.org/materials | Journal of Materials Chemistry

Macroporous elastic polyacrylamide gels prepared at subzero temperatures:

control of porous structure
Fatima Plieva,ab Xiao Huiting,a Igor Yu. Galaev,a Björn Bergenståhlc and Bo Mattiasson*a
Received 23rd May 2006, Accepted 17th August 2006
First published as an Advance Article on the web 31st August 2006
DOI: 10.1039/b606734d
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM.

Macroporous polyacrylamide gels (MPAAGs) with unique elastic morphology and open porous
structure are prepared at subzero temperatures. The porous structure of MPAAGs consisting of
large, 1–100 mm-sized interconnected pores, is controlled by the freezing temperature, the content
of the initiator system (ammonium persulfate and N,N,N9,N9-tetramethyl-ethylenediamine) in the
initial reaction mixture and the solvent used. The initiator content, through its effects on the
kinetics of polymerization, is an efficient tool of control allowing for the preparation of well
defined macroporous elastic structures with either open interconnected pores or closed pores. In
the semi-frozen reaction mixture, the dissolved monomers and initiators are concentrated in the
unfrozen liquid microphase where the polymerization reaction proceeds. The final freezing
temperature, Tf (defined as temperature fixed in a low temperature thermostat), as well as the
solvent used affect the porous structure through their effect on the formation of the unfrozen
liquid microphase.

Introduction cost-effective continuous chromatographic pAAm bed with

improved flow path properties.16
Macroporous polymeric materials are of significant, both Recently we have introduced a monolithic polyacrylamide
fundamental and technological, interest.1 In particular, gel (MPAAG) for bioseparation with fundamentally new
porous hydrogels are prepared using several techniques, such properties. The MPAAGs synthesized in semi-frozen aqueous
as freeze-drying,2 porogenation,3 microemulsion formation,4 media are highly elastic materials with large, 1–100 mm-sized
gas blowing technique5 and phase separation.6,7 Macroporous interconnected pores.17 One of the unique properties of
hydrophilic materials with open porous structure are gaining MPAAGs is their high mechanical stability caused by the fact
increasing interest in biotechnology and biomedicine.8,9 that pore walls in MPAAG are formed from a highly
Macroporous hydrogels with interconnected pores and concentrated polymer phase.18 Due to the large interconnected
tissue-like elasticity are attractive scaffolds for drug delivery pores, the water remains inside due to capillary forces thus
and tissue engineering.10–12 Typically, porosity of a cell making the MPAAG monoliths ‘‘drainage-protected’’. Due to
scaffold should be at least 90%, to provide a high surface the elasticity and sponge-like morphology, MPAAGs can
area for maximizing cell seeding and attachment and mass withstand large deformations and can be easily compressed up
transport of nutrients and oxygen.13 It is important to to 80–90% without being mechanically damaged whereas
retain proper control over the porosity of the macroporous traditional polyacrylamide gels are brittle and easily destroyed
materials. when deformed. Recently we showed that affinity bound
Polyacrylamide (pAAm) gels are well studied gels having bioparticles can be efficiently detached from MPAAGs by
enormous importance in biochemical separations, with parti- mechanical compression.19
cular emphasis on protein analysis and characterization, The challenge is to control the porous structure as well as
since the pore size in pAAm gels is of the right magnitude overall properties of the gels to allow for a rational design of
for efficient sieving of proteins via electrophoretic techni- MPAAGs for particular applications. Our previous results
ques.14 Different approaches were used for the synthesis of indicated that the porous structure was efficiently controlled
pAAm gels with novel properties and porous structure. by varying the monomer concentration and the nature of
Righetti et al. synthesized macroporous acrylamide hydrogels the cross-linker, which in turn changed the size of the
adding a hydrophilic polymer (polyethylene glycol) into interconnected pores and the thickness of the pore walls in
the monomer mixtures.15 Hjerten prepared a compressed MPAAGs.20 In this study we aimed to show detailed control
of the pore morphology in MPAAGs by varying the final
freezing temperature Tf (or temperature set in the low
a
Department of Biotechnology, Center for Chemistry and Chemical temperature thermostat LAUDA RK20KP), the content of
Engineering, Lund University, P. O. Box 124, SE-22100 Lund, Sweden. the initiating system (ammonium persulfate/N,N,N9,N9-tetra-
E-mail: bo.mattiasson@biotek.lu.se; Fax: +46-46-2224713; methyl-ethylenediamine, APS/TEMED) in the initial reaction
Tel: +46-46-2228264 mixture and the solvent used. Changes of these parameters
b
Protista Biotechnology AB, SE-22370 Lund, Sweden
c
Department of Food Technology, Center for Chemistry and Chemical resulted in marked changes of the elasticity and pore size of the
Engineering, Lund University, P. O. Box 124, SE-22100 Lund, Sweden MPAAG monoliths.

This journal is ß The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 4065–4073 | 4065
 View Article Online

Experimental was determined (md). The gel fraction yield was defined as
(md/mt) 6 100%, where mt is the total mass of the monomers
Materials in the feed mixture.
Acrylamide (AAm, more than 99.9% purity, electrophoresis Water flow resistance was determined as follows: the
reagent), N,N,N9,N9-tetramethyl-ethylenediamine (TEMED), MPAAG monoliths of the same geometrical size (inner
ammonium persulfate (APS), N,N9-methylenebis(acrylamide) diameter and height of the monolithic cryogel rod in the
(MBAAm), allyl glycidyl ether (AGE, 99%) were from Aldrich swollen state were 1 and 4 cm, respectively) were inserted
(Aldrich, Steinheim, FRG). Formamide was from Sigma into a glass column (i.d. 1 cm). The water flow path was
(St. Louis, USA). Dioxane and sodium chloride were from determined as a flow rate through the MPAAG monoliths
at the constant hydrostatic pressure equal to 1 m of water
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM.

Merck (Darmstadt, Germany).

column. All measurements were done in triplicate and the
Preparation of MPAAGs average values are presented.
For the compression tests the MPAAGs (prepared inside a
The MPAAGs with functional epoxy groups (prepared glass column with diameter 13 mm) were cut when frozen
according to the mode ‘‘freezing before gelation’’) were into 10 mm height cylindrical probes. Compression tests were
produced as follows: monomers (3.67 g of AAm, 0.60 g of performed with a TA-XT2 instrument (Stable Micro Systems,
MBAAm and 1.0 ml of AGE) were dissolved in deionized
Godalming, Surrey, UK) at 22 uC, using a 36 mm diameter
water (final concentration 10%). The co-monomer AGE was
plunger. All compressions tests were performed in deionized
included in the reaction mixture to incorporate the functional
water. Cryogel probes, soaked in deionized water, were
epoxy groups onto the surface of the MPAAGs. The molar
compressed up to a deformation level of 40% at a crosshead
partition of the functional co-monomer, AGE, was kept at the
speed of 1 mm s21. Four probes for each sample were
same level all the time (AGE/AAm 0.14 mol/mol). Free radical
measured. The force–displacement curves were used to
polymerization was initiated by TEMED (1.2%) and APS
calculate the compressive strength (E) (at a strain of 40%) as
(1.2%). The reaction mixture was poured into glass columns
follows: E = F/pR2 6 (Dho/h)21, where F is the force applied to
(i.d. 10 mm) and was frozen at the final temperature (Tf) of
the cryogels with a cross-sectional area pR2, h is the height of
212, 220 or 230 uC for 16 h. After washing with water the
the sample at compression and Dho is the change in height of a
MPAAGs were dried in an oven at 60 uC overnight and stored
sample during compression (displacement of 40%).
in the dry state. The dried polymer samples were re-swollen in
For the SEM studies, the samples were fixed in 2.5%
deionized water before use. For preparation of conventional
glutaraldehyde in 0.1 M sodium phosphate buffer overnight.
acrylamide gel (PAAG), the reaction mixture was kept at
Then the samples were dehydrated in ethanol (0%–50%–75%–
ambient temperature (about 22 uC) for 16 h followed by
99.5%) and critical point dried. The dried sample was coated
intensive washing in water.
with gold–palladium (40 : 60) and examined using a JEOL
The MPAAGs in the presence of 5% initiating system
JSM-5600 LV scanning electron microscope.
(according to the mode ‘‘freezing after gelation’’) were
prepared as described above at a final freezing temperature
Tf of 212 uC. Results and discussion
The MPAAGs (as well as conventional gels at 22 uC) in The porous structure defines the particular application of
formamide (95%) and dioxane (95%) media were prepared as MPAAGs. However, the study of pores in wet hydrogels
described above except that the amount of initiating pair presents a significant challenge, as there is no general method
(APS/TEMED) was kept at 0.5 and 1.4% for the formamide to determine pore size and structure in soft and highly
and dioxane media, respectively (to keep the mode ‘‘freezing hydrated systems, in contrast to dry rigid materials.
before gelation’’).
The porosity and pore size distribution of a porous rigid
The temperature profile (freezing curve) was recorded by
material in the dry state are traditionally determined by
placing a thermocouple (T-type) directly into the water or
mercury intrusion porosimetry.21–23 It was not possible to use
reaction mixture, 5 ml (cooled in an ice bath for 30 min to
this method for dried MPAAGs due to the soft nature of these
6 uC) in the glass column (i.d. 10 mm). The thermocouple was
materials.
placed at the center of the glass column at half of the height
Environmental Scanning Electron Microscopy (ESEM)
of the reaction mixture. The sample was placed in a chamber
allows for the examination of hydrated samples in their
of the low temperature thermostat LAUDA-RK20KP cooled
natural state.24–27 We have previously used ESEM to analyze
to the fixed final temperature Tf. The temperature of the
the porous structure of MPAAGs prepared from reaction
sample was measured using a digital thermometer Dual LogR.
mixtures with different monomer concentrations.20 One of
Temperature changes in the sample depending on the cooling
the important advantages of the ESEM technology is the
time were recorded. Average values of three measurements are
possibility of monitoring changes in the structure of the
presented.
material whilst allowing the sample to dehydrate slowly.
The ESEM images of MPAAGs at different degrees of
Characterization of MPAAG
dehydration showed a real view of a fully hydrated sample
The gelation yield was determined as follows: the swollen (at low degrees of dehydration, when all pores were filled with
cryogel sample (1 ml) was put in an oven at 60 uC for drying. water) and the surface details, i.e. the macroporous structure
After drying till constant weight, the mass of the dried sample with interconnected pores of hundreds of micrometers in size

4066 | J. Mater. Chem., 2006, 16, 4065–4073 This journal is ß The Royal Society of Chemistry 2006
 View Article Online

No re-swelling within 1 h

No re-swelling within 1 h
Gelation yield was determined as described in the Experimental section. b The water flow resistance was measured as described in the Experimental section. c The properties of a traditional
polyacrylamide gel prepared at room temperature and initiator concentration 1.2% are presented for comparison. d Compressive strength at failure at 24 and 30% compression for 1.2-PAAG
at high degrees of dehydration.20 However, ESEM is not
suitable to reveal the fine structure at the micrometer scale.

Re-swelling time of
Scanning Electron Microscopy (SEM) is suitable to charac-
terize the fine structures of porous materials but only in the

dried sample/s
dried state. For many soft and highly hydrated materials
this technique can hardly be used due to considerable
changes in pore structure when drying prior to sample

5–6
preparation for SEM. However, in the case of macroporous
cryogels (particularly, in the case of MPAAGs), SEM is the

Practically no liquid flow through the sample

Practically no liquid flow through the sample

method of choice for the analysis of porous structure due to
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM.

the excellent mechanical stability of the MPAAGs. The pore

walls in elastic and spongy MPAAGs are formed from a
highly concentrated polymer network due to the pronounced
concentration of the dissolved reagents in small non-frozen
regions where the polymerization reaction proceeded.17,18,28

Water flow path/cm h21 b

Not surprisingly, the porous structures revealed by both
ESEM and SEM at the 1–100 mm scale were essentially
the same.17,20,29
The deterioration of the porous structure of the MPAAGs
due to the processing of MPAAG samples for SEM analysis

Table 1 Properties of the MPAAGs prepared from reaction mixtures with different contents of initiating system (APS/TEMED)
(i.e. drying of the sample via gradual replacement of water

310
with ethanol, followed by critical point drying18) was non-
significant and the fine porous structure was essentially

Compressive
strength/kPa
preserved due to the mechanical stability of the pore walls in
MPAAGs. Thus, we believe that the fine porous structure of

43.9d

214.3d
28.3
the MPAAGs (roughness, tortuosity, micro/macro-porosity)
visualized by the SEM technique reflected the real structure of
the material, rather than appeared as the result of drying
Gel fraction
artifacts.20,29,30 yielda (%)

The effect of initiator content on the MPAAG porous structure 97

73
When preparing polyacrylamide gels at subzero temperatures, 99
Elastic, non-spongy
Elastic and spongy

one could discriminate two possible situations: i) freezing of

(control gel) and 5-MPAAG, respectively. For details see the Experimental section.
the solution before the gelation takes place and ii) freezing
Appearance

of the already gelated mixture. The polymerization rate is

Brittle gel

controlled by the amount of initiating system in the

reaction mixture. In the free radical polymerization reactions,
the concentrations of the initiator ammonium persulfate
(APS) and activator, N,N,N9,N9-tetramethyl-ethylenediamine
prepared MPAAG
Notation for the

(TEMED) have a great influence on the polymerization rate

as well as on the molecular weight of the resulting polymers.
1.2-MPAAG
5-MPAAG
1.2-PAAG

The polymerization reaction starts with the reaction between

APS and TEMED to form free radicals to initiate the
polymerization reaction of the monomers (AAm, AGE and
MBAAm) as well as the cross-linking of polymer chains with
Gelation at room temperature

the cross-linker MBAAm. The MPAAGs were prepared at

212 uC in the presence of different initiator concentrations
Freezing before gelation

(1.2 and 5%) in the reaction feed. The choice of initiator

Freezing after gelation

concentrations was motivated by the attempts to create

Freezing/gelation

conditions corresponding to the different cryogelation situa-

tions. As expected, the prepared MPAAGs had different
morphologies and porous structures (Table 1).
The 1.2-MPAAGs (‘‘freezing before gelation’’) had a spongy
and elastic structure and re-swell after contact with water
within seconds. One could conclude that cryopolymerization
conc. (%)

proceeded in the non-frozen microphase remaining after

Initiator

freezing of the reaction feed and resulted in the formation of

1.2c
1.2

a macroporous structure with interconnected pores. It should

5
a

This journal is ß The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 4065–4073 | 4067
 View Article Online

MPAAGs prepared in the presence of 1.2% initiator system

with an overcooling temperature of 211 uC). As ice crystals
were formed within the already formed gel (where it was not
possible for ice crystals to merge with each other), a porous
structure with large (mm-sized) closed (not interconnected)
macropores was formed resulting in a higher flow resistance of
5-MPAAG as compared to 1.2-MPAAG (Table 1).
Polyacrylamide gels prepared at room temperature with a
cross-linking ratio (MBAAm/(AAm + MBAAm) 6 100%)
above 5–6% are opaque with a concomitant increase in pore
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM.

size.37 It is generally accepted that in water, the relative

reactivity of MBAAm is higher than that of AAm, which
Fig. 1 Thermograms for freezing aqueous solutions of the monomers causes the formation of MBAAm clusters within the gel struc-
in the presence of 1.2% (solid line) and 5% (dashed line) initiating ture.38 Alternatively, the high hydrophobicity of MBAAm
systems (APS/TEMED). For details see the Experimental section. resulted in phase separation of polymer segments enriched
with MBAAm.39 As the MBAAm content in the reaction feed
be noted that such structures were successfully used as in our case was high (10%), the 1.2-PAAG produced at
chromatographic monolithic sorbents for chromatography room temperature was microporous (Fig. 2) with high flow
of particulate containing liquids18,31 or microbial cells17,18,32 resistance and poor ability to reswell after drying. The gel 1.2-
or as cell culture reactors.33 MPAAG on the other hand had interconnected macropores of
As the concentration of the APS/TEMED pair was up to 100 mm in size surrounded with smooth and dense pore
increased from 1.2 to 5%, the rigidity of the material was walls (Fig. 3 a,b).
increased (Table 1). The freezing thermograms obtained for When the gel had been already formed before ice crystal-
the reaction mixture in the presence of 1.2 and 5% initiator, lization, ice crystals could only move apart the existing gel
respectively, showed how the temperature changed during the structure resulting in the formation of large, but not connected
cooling of the reaction mixture until reaching the freezing pores in 5-MPAAG (Fig. 3 c, d). This structure explains the
temperature, i.e. the temperature fixed in the low temperature properties of 5-MPAAG. Dense pore walls endow 5-MPAAG
thermostat. Analysis of the freezing thermograms is highly with elasticity in contrast to polyacrylamide gels produced at
instrumental for the determination of phase and structural room temperature. However, the closed pores explain the high
changes during cooling and crystallization within the sys- flow resistance of 5-MPAAG when compared to that of
tem.34,35 Overcooling to 211 uC for 2 min was observed for the 1.2-MPAAG with an interconnected pore structure (Table 1).
reaction mixture frozen at 212 uC in the presence of 1.2% The development of a closed pore structure when freezing
initiating system (Fig. 1, solid line). Overcooling (or super- takes place in the presence of an already formed polymer
cooling) is defined as cooling below the initial freezing point network could be explained as follows. When water freezes in a
of the water without forming ice crystals. This is a non- porous medium, the freezing point of water in the capillary is
equilibrium, metastable state of water. The nucleation lowered as the diameter of the capillary decreases. The
temperature of water is affected by both the cooling rate and depression of the freezing temperature is as large as 20 uC in
the addition of a nucleation agent.36 Once the critical mass of pores of 2 nm diameter40 meaning that there is liquid water in
nuclei was reached, the system nucleated at a temperature of small pores at 212 uC, i.e. at the temperature used for the
about 22.5 uC, 2.6 min after the reaction mixture was placed MPAAG production. In the polymer network of the ordinary
in the thermostat (Fig. 1, solid line). After crystallization was polyacrylamide gel one could expect pores of a few nanometers
completed, the temperature dropped slowly till the fixed Tf
value, 212 uC. In the presence of 5% initiating system, the
picture is completely different. The overcooling is much less,
to 25 uC, and the system started to nucleate much faster,
1.5 minutes after the reaction mixture was placed in the
thermostat, followed by a freezing plateau at 20.9 uC. Ice
crystallization in this case took place at the temperature
close to 0 uC which is typical for plain water, whereas in the
presence of 1.2% initiating system the crystallization proceeded
at 22.5 uC probably due to the presence of non-reacted
monomers which osmotically depressed the crystallization
temperature (Fig. 1, dashed line). This explanation is
consistent with the assumption of ‘‘freezing before gelation’’
in the presence of 1.2% initiator and ‘‘freezing after gelation’’
in the presence of 5% initiator. Moreover, the junctions of the
already formed gel network performed in the latter case as ‘‘ice
nucleators’’, facilitating the ice crystal formation and resulting Fig. 2 SEM of PAAG prepared in aqueous medium at room
in raising the overcooling temperature to 25 uC (compared to temperature in the presence of 1.2% initiating system (APS/TEMED).

4068 | J. Mater. Chem., 2006, 16, 4065–4073 This journal is ß The Royal Society of Chemistry 2006
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM. View Article Online

Fig. 3 SEMs of MPAAG prepared in aqueous media at 212 uC with different contents of initiating system in the initial reaction mixture: (a,b)
1.2-MPAAG (freezing before gelation). (c,d) 5-MPAAG (freezing after gelation).

in size as protein molecules with sizes of a few nanometers The load–displacement curves obtained for 1.2-MPAAG
penetrate through the gel, e.g. during electrophoresis. showed typical behavior of highly elastic materials (Fig. 4).
However, there is a distribution of pore sizes. Hence, the 1.2-MPAAG showed no breaking point against compression
water starts freezing in larger voids whereas it is transported least at a strain of 90% (data not shown). These results indicate
through smaller pores to join the ice crystal as long as the that 1.2-MPAAG (i.e. prepared according to ‘‘freezing before
water remains above its freezing temperature in the smaller gelation’’) have extensive flexibility and shape-recovery pro-
pores. That process results in the formation of ‘‘ice lenses’’ perties against compression. The compressive strength for
surrounded by relatively desiccated areas. To some extent the 5-MPAAG was higher than that of 1.2-MPAAG (Table 1),
pore formation in the cryogels in this case is similar to ‘‘frost however failure occurred at compression .30% for 5-MPAAG
heave’’.41 Melting of ice lenses resulted finally in the formation due to the increased rigidity (compared to 1.2-MPAAG) of
of the closed pore structure of MPAAG. these gels and the formation of a porous structure with closed
macropores.

The effect of freezing temperature on the MPAAG porous

structure
The freezing rate is one of the crucial parameters during the
preparation of the MPAAGs that defines their porous
structure via affecting the size of the ice crystals formed.42,43
The lower the freezing rate (or the higher the fixed freezing
temperature), the larger the size of growing ice crystals and, as
a result, cryogels with larger pore sizes are prepared.30
However, at relatively high freezing temperatures there is a
risk that the solution to be frozen will remain unfrozen in a
supercooled state. So, the temperature should be low enough
to ensure freezing of the reaction mixture. At the same time, a
Fig. 4 Load–displacement curves for 1.2-MPAAG (solid line) and too low freezing temperature results in the preparation of
5-MPAAG (dashed line). For details see the Experimental section macroporous gels with increased hydrodynamic resistance.42,43

This journal is ß The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 4065–4073 | 4069
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM. View Article Online

Fig. 5 Thermograms for freezing aqueous solutions of monomers

at 212, 220 and 230 uC in the presence of 1.2% initiating system
(APS/TEMED).

The higher the rate of freezing, the more nucleation is

promoted, and the higher the number of small ice crystals
formed.30,44,45 Three Tf temperatures (230, 220 and 212 uC)
were tested to study the influence of the freezing temperature
on the porous properties of MPAAGs. As discussed above,
freezing at a relatively high Tf (212 uC) resulted in a
pronounced overcooling at approx. 210 uC followed by a
crystallization period at about 22 uC (Fig. 5). A freeze–
concentration process occurred as water was frozen out from
the solution. The increase in viscosity of the unfrozen liquid
phase hindered further crystallization, which was over after
about 8 min (Fig. 5). At lower temperatures (Tf 220 and
Fig. 6 SEMs of MPAAG prepared in aqueous media at 230 uC in
230 uC, respectively), the freezing thermograms revealed
the presence of 1.2% initiating system (APS/TEMED).
practically no overcooling with a small crystallization plateau
near 0 uC (Fig. 5). A small peak observed at about 3.3 min
for the thermogram obtained at Tf 2 30 uC (a temperature macroporous structure formed after polymerization at
230 uC (Fig. 6).
which is below the eutectic point of about 220 uC for the
water–acrylamide system46,47) indicated probably the complete
crystallization of the liquid microphase that is not frozen at The solvent effect on the MPAAG porous structure
higher temperatures (Fig. 5). Different solvents have been employed as porogenic agents
As fast freezing promotes the formation of ice crystals with when preparing monolithic polymeric supports.48–50 Finding
smaller sizes compared to slow freezing at higher temperatures, an appropriate porogen is considered to be a key step in the
the flow resistance of MPAAG-20 and MPAAG-30 was higher production of polymer matrices with good performance in
than that of MPAAG-12 (Table 2). The gel fraction yield for chromatographic applications.51
the MPAAG-30 was also lower than that for MPAAG-12 In case of monolithic gels prepared at subzero temperatures,
probably due to the quenching of the polymerization process the crystals of frozen solvent perform as porogens, thus the
when the system was completely frozen (no liquid phase left) presence of solvents other than water has an effect on the
at a temperature below the eutectic point. As a result, the porous structure both via affecting the solubility of polymer
compressive strength for MPAAG-30 was lower compared to and monomer and via affecting the size and shape of solvent
MPAAG-12 and MPAAG-20 (Table 2). In agreement with the crystals formed in the semifrozen system and the volume of
high flow resistance of MPAAG-30, SEM revealed no the nonfrozen region (‘‘liquid microphase’’). The choice of

Table 2 Properties of the MPAAGs prepared at different temperatures from the reaction mixtures with 1.2% content of initiating system
(APS/TEMED)
Temperature/uC Notation for prepared MPAAG Gel fraction yield (%)a Compressive strength/kPab Water flow path/cm h21 c

212 MPAAG-12 79 ¡ 4 28.3 310

220 MPAAG-20 83 ¡ 3 29.2 215
230 MPAAG-30 68 ¡ 4 21.0 30
a
Gelation yield was determined as described in the Experimental section. b Compressive strength was determined as described in the
Experimental section. c Flow resistance was measured as described in the Experimental section.

4070 | J. Mater. Chem., 2006, 16, 4065–4073 This journal is ß The Royal Society of Chemistry 2006
 View Article Online

Table 3 Properties of MPAAGs prepared in water and water–organic media at 220 uC

Notation for Gel fraction Compressive Water flow
Solvent prepared MPAAG Appearance yield (%)a strength/kPa path/cm h21 b

Water (100%) W-MPAAG Elastic and spongy 75 28.3 340

Formamide–water (95 : 5) F-MPAAG Elastic and spongy, 84 3.6c 114
oriented porosity
Dioxane–water 95 : 5 D-MPAAG Elastic, non-spongy 82 6.5c 60
a
The MPAAGs were prepared according to the ‘‘freezing before gelation’’ approach (in the presence of 1.2, 0.5 and 1.4% APS/TEMED
system for W-MPAAG, F-MPAAG and D-MPAAG, respectively). Gelation yield was determined as described in the Experimental section.
b
Flow resistance was measured as described in the Experimental section. c Compressive strength at failure (18% compression for F-MPAAG
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM.

and 22% compression for D-MPAAG). For details see the Experimental section.

solvents for preparation of cryogels is mainly specified by the for PAAG synthesized in formamide (Fig. 8a). In contrast, the
melting point of the solvents, i.e. the solvents should be PAAG synthesized in 95% 1,4-dioxane showed microporous
freezable at moderately low temperatures (not lower than structure due to the segregation of the polymeric phase during
220 uC). The solvents can be divided into two categories: the polymerization process (Fig. 8c). The SEM micrograph of
solvents which are miscible with the resultant polymeric F-MPAAG prepared at 220 uC showed an oriented porous
network and solvents in which phase separation is initiated structure with long pores (Fig. 8b) as a result of the formation
during the polymerization process. Polymerization induced of needle like formamide crystals. Due to the solvent induced
phase separation (PIPS) is a process in which an initially phase separation process, a bimodal porous structure (com-
homogeneous solution of monomer and solvent becomes posed of large macropores with microporous pore walls) was
phase separated during the course of its polymerization.39 visualized for D-MPAAG (Fig. 8d). The high water flow
Righetti et al. synthesized macroporous acrylamide hydrogels resistance for D-MPAAG confirmed the preparation of a
by adding a hydrophilic polymer (polyethylene glycol) into porous structure with closed macropores (Table 3). The
the monomer mixtures.15 It was postulated that the competi- compressive strengths of F-MPAAG and D-MPAAG were
tion between the gelation process and the phase separation less than that of W-MPAAG (Table 3) due to the different
process is responsible for the formation of pores that are much porous structures of these gels.
larger than those obtained in the absence of added hydrophilic
polymer.15,52 Conclusion
The organic solvents formamide and 1,4-dioxane with
melting temperatures of 2 and 11 uC, respectively, and Pore size, shape and morphology in the macroporous poly-
different polarities (in the order: formamide . water . 1,4- acrylamide gels produced by radical polymerization at subzero
dioxane) were used for preparation of MPAAGs in water– temperatures can be controlled in a rational way. Depending
organic media at 220 uC (Table 3). The amount of initiating on the conditions used (the content of the initiator, freezing
system (APS/TEMED) in the reaction mixtures was kept in the temperature and the solvent used) a broad variety of porous
ratio to maintain the mode ‘‘freezing before gelation’’. Due to structures ranging from closed macropores with microporous
the association of polar molecules of formamide with the walls to open interconnected macropore systems with
monomer molecules (AAm and MBAAM), the formamide nonporous walls and uniform (in water) or oriented (in
solvent belongs to the first category (i.e. solvent miscible formamide) porosity have been produced. Presumably, the
with resultant polymeric network) and 1,4-dioxane, as
expected, is able to induce phase separation during the
polymerization process.
The freezing thermogram in formamide medium showed
overcooling to 214 uC for 2 min followed by a crystallization
plateau at around 25 uC (Fig. 7, dashed line). The freezing
thermogram in 90% dioxane is characterized by the plateau of
the solvent crystallization around 2.7 uC despite the fact that
pure dioxane freezes at 11 uC (Fig. 7, dotted line). In both
systems (formamide and 1,4-dioxane media) as well as in water
(Fig. 7, solid line, presented for comparison) the reaction
mixtures were frozen within 6 min (Fig. 7).
The SEM analysis of conventional PAAG synthesized in
95% formamide indicated a nonporous structure (Fig. 8a). The
association of the monomers (AAm and MBAAM) and the
Fig. 7 Thermograms for freezing of monomer solutions in 95%
polymer network with formamide results in better solvation of dioxane in the presence of 1.4% initiating system (APS/TEMED)
hydrophobic MBAAM in the formamide medium compared (dotted line) and 95% formamide in the presence of 0.5% initiating
to water and the probability for the formation of highly system (APS/TEMED) (dashed line) at 220 uC. The freezing
concentrated MBAAM regions within the network (that were thermogram of the monomer solution in water in the presence of
observed for PAAG prepared in water, Fig. 2) has disappeared 1.2% initiating system is presented for comparison (solid line).

This journal is ß The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 4065–4073 | 4071
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM. View Article Online

Fig. 8 SEMs of conventional PAAG prepared at 22 uC (a) and F-MPAAG prepared at 220 uC (b) in 95% formamide in the presence of 0.5%
initiating system (APS/TEMED); and conventional PAAG prepared at 22 uC (c) and D-MPAAG prepared at 220 uC (d) in 95% dioxane in the
presence of 1.4% initiating system.

bimodal pore distribution is the result of the combination of 7 Y. S. Nam and T. G. Park, J. Biomed. Mater. Res., 1999, 47,
8–17.
two processes, the formation of macropores in the non-frozen
8 J. Chen, S. Jo and K. Park, Carbohydr. Polym., 1995, 28, 69–76.
liquid phase (when solvent crystals perform as porogens) and 9 N. A. Peppas, Y. Huang, M. Torres-Lugo, J. H. Ward and
the phase separation of the polymer synthesized within the J. Zhang, Annu. Rev. Biomed. Eng., 2000, 02, 9–29.
non-frozen liquid phase. 10 M. H. Sheridan, L. D. Shea, M. C. Peters and D. J. Mooney,
J. Controlled Release, 2000, 64, 91–102.
11 Q. Hou, D. W. Grijpma and J. Feijen, J. Biomed. Mater. Res.,
Acknowledgements Part B: Appl. Biomater., 2003, 67B, 732–740.
12 A. N. Stachowiak, A. Bershteyn, E. Tzatzlos and D. J. Irvine, Adv.
This work was financially supported by Protista International Mater., 2005, 17, 399–403.
AB (Bjuv, Sweden), The Swedish Foundation for Strategic 13 Q. Cai, J. yang, J. Bei and S. Wang, Biomaterials, 2002, 23,
4483–4492.
Research, The Swedish Foundation for International 14 P. G. Righetti and C. Gelfi, J. Chromatogr., B, 1996, 699, 63–75.
Cooperation in Research and Higher Education (STINT, 15 D. Asnaghi, M. Giglio, A. Bossi and P. G. Righetti, J. Chem.
IG2003-2089), The Swedish Institute (Visby Program, projects Phys., 1995, 102, 9736–9742.
2886/2002 and 01211/2004), Royal Swedish Academy of 16 S. Hjerten, Y. Y.-M. Li, J.-L. Liao, J. Mohammad, K. Nakazato
and G. Pettersson, Nature, 1992, 356, 810–811.
Sciences.
17 F. M. Plieva, J. Andersson, I. Y. Galaev and B. Mattiasson, J. Sep.
Sci., 2004, 27, 828–836.
References 18 P. Arvidsson, F. M. Plieva, V. I. Lozinsky, I. Y. Galaev and
B. Mattiasson, J. Chromatogr., A, 2003, 986, 275–290.
1 H.-P. Hentze and M. Antonietti, Rev. Mol. Biotechnol., 2002, 90, 19 M. B. Dainiak, A. Kumar, I. Y. Galaev and B. Mattiasson, Proc.
27–53. Natl. Acad. Sci. U. S. A., 2006, 103, 849–854.
2 V. R. Patel and M. M. Amiji, Pharmacol. Res., 1996, 13, 588–593. 20 F. M. Plieva, M. Karlsson, M.-R. Aguilar, D. Gomez,
3 C. D. Wood and A. I. Cooper, Macromolecules, 2001, 34, 5–8. S. Mikhalovsky and I. Y. Galaev, Soft Matter, 2005, 1, 303–309.
4 D. J. Bennett, R. P. Burford, T. P. Davis and H. J. Tilley, Polym. 21 D. Josic, A. Buchacher and A. Jungbauer, J. Chromatogr., B, 2001,
Int., 1995, 36, 219–226. 752, 191–205.
5 K. Kabiri, H. Omidian and M. J. Zohuriaan-Mehr, Polym. Int., 22 C. Viklund, F. Svec, K. Irgum and J. M. J. Frechet, Chem. Mater.,
2003, 52, 1158–1164. 1996, 8, 744–750.
6 S. Hjerten, J.-L. Liao and R. Zhang, J. Chromatogr., 1989, 473, 23 H. Zou, X. Huang, M. Ye and Q. Luo, J. Chromatogr., A, 2002,
273–275. 954, 5–32.

4072 | J. Mater. Chem., 2006, 16, 4065–4073 This journal is ß The Royal Society of Chemistry 2006
 View Article Online

24 M. Manero, F. J. Gil, E. Padros and J. A. Planell, Microsc. Res. 39 A. Y. Kwok, G. G. Qiao and D. H. Solomon, Polymer, 2003, 44,
Technol., 2003, 61, 469–480. 6195–6203.
25 Q. F. Wei, X. Q. Wang, R. R. Mather and A. F. Fotheringham, 40 G. K. Rennie and J. Clifford, J. Chem. Soc., Faraday Trans. 1,
Fibres Text. East. Eur., 2004, 12, 79–83. 1977, 73, 680–689.
26 A. M. Donald, C. He, C. P. Royall, M. Sferrazza, N. A. 41 K. Watanabe, Y. Muto and M. Mizoguchi, Cryst. Growth Des.,
Stelmashenko and B. L. Thiel, Colloids Surf., 2000, 174, 37–53. 2001, 1, 207–211.
27 R. Rizzieri, F. S. Baker and A. M. Donald, Polymer, 2003, 44, 42 V. I. Lozinsky, E. S. Vainerman, S. A. Ivanova, E. F. Titova,
5927–5935. M. I. Shtilman, E. M. Belavtseva and S. V. Rogozhin, Acta Polym.,
28 V. I. Lozinsky, Russ. Chem. Rev., 2002, 71, 489–511. 1986, 37, 142–146.
29 F. M. Plieva, M. Karlsson, M.-R. Aguilar, D. Gomez, 43 F. M. Plieva, I. N. Savina, S. Deraz, J. Andersson, I. Y. Galaev
S. Mikhalovsky, I. Y. Galaev and B. Mattiasson, J. Appl. Polym. and B. Mattiasson, J. Chromatogr., B, 2004, 807, 129–137.
Sci., 2006, 100, 1057–1066. 44 S. Ueno, G.-S. Do, Y. Sagara, K. Kudoh and T. Higuchi, Int. J.
Published on 31 August 2006. Downloaded by Purdue University on 10/22/2024 7:01:11 PM.

30 V. I. Lozinsky, F. M. Plieva, I. Y. Galaev and B. Mattiasson, Refrig., 2004, 27, 302–308.

Bioseparation, 2002, 10, 163–188. 45 R. Lawrence, F. Consolacion and P. Jelen, Food Technol., 1986, 60,
31 A. Hanora, I. Savina, F. M. Plieva, V. A. Izumrudov, 77–83.
B. Mattiasson and I. Y. Galaev, J. Biotechnol., 2006, 123, 343–355. 46 V. I. Lozinsky, S. A. Morozova, E. S. Vainerman, E. F. Titova,
32 M. B. Dainiak, F. M. Plieva, I. Y. Galaev, R. Hatti-Kaul and M. I. Shtilman, E. M. Belavtseva and S. V. Rogozhin, Acta Polym.,
B. Mattiasson, Biotechnol. Prog., 2005, 21, 644–649. 1989, 40, 8–15.
33 A. Kumar, V. Bansal, K. S. Nandakumar, I. Y. Galaev, 47 V. I. Lozinsky, E. S. Vainerman, G. F. Korotaeva and
P. K. Roachoudhry, R. Holmdahl and B. Mattiasson, S. V. Rogozhin, Colloid Polym. Sci., 1984, 262, 617–622.
Biotechnol. Bioeng., 2006, 93, 636–646. 48 F. Svec and J. M. J. Frechet, Ind. Eng. Chem. Res., 1999, 38,
34 M. S. Rahman, N. Guizani, M. Al-Khaseibi, S. A. Al-Hinai, 34–48.
S. S. Al-Maskri and K. Al-Hamhami, Food Hydrocolloids, 2002, 49 E. C. Peters, M. Petro, F. Svec and J. M. J. Frechet, Anal. Chem.,
16, 653–659. 1998, 70, 2288–2295.
35 J. Liu and Y.-X. Zhou, Anal. Bioanal. Chem., 2003, 377, 173–181. 50 A. Strancar, A. Podgornik and M. Barut, Bioforum Int., 2002, 3,
36 S.-L. Chen and T.-S. Lee, Int. J. Heat Mass Transfer, 1998, 41, 152–153.
769–783. 51 G. S. Chirica and V. T. Remcho, J. Chromatogr., A, 2001, 924,
37 A. Palm and M. V. Novotny, Anal. Chem., 1997, 69, 4499–4507. 223–232.
38 D. Asnaghi, M. Giglio, A. Bossi and D. G. Righetti, Macro- 52 D. Asnaghi, M. Giglio, A. Bossi and P. G. Righetti, J. Mol. Struct.,
molecules, 1997, 30, 6194–6198. 1996, 383, 37–42.

This journal is ß The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 4065–4073 | 4073