Content uploaded by Riina Aav
Author content
All content in this area was uploaded by Riina Aav
Content may be subject to copyright.
Organic acid catalyzed synthesis of 5-methylresorcinol based
organic aerogels in acetonitrile
Anna-Liisa Peikolainen
•
Olga Volobujeva
•
Riina Aav
•
Mai Uibu
•
Mihkel Koel
Ó Springer Science+Business Media, LLC 2011
Abstract A method of preparing 5-methylresorcinol and
formaldehyde based organic aerogels in non-aqueous
media with a benzoic acid derivative as a catalyst is being
proposed in this paper. Here acetonitrile is used as a sol-
vent that allows direct drying with carbon dioxide over the
supercritical state without the need for a solvent exchange.
The acidic properties of 2,6-dihydroxy-4-methyl benzoic
acid promote the reaction of sol–gel polymerization, and at
the same time it takes part in the reaction as a co-monomer
and influences the nanostructure of the material. The
evolution of the polymer was monitored using nuclear
magnetic resonance spectroscopy and the structure of the
resulting organic aerogels depending on the molar ratio of
5-methylresorcinol to 2,6-dihydroxy-4-methyl benzoic
acid was studied by nitrogen adsorption–desorption mea-
surements, scanning electron microscopy and infrared
spectrometry.
Keywords Aerogel Organic aerogel
5-methylresorcinol 2,6-dihydroxy-4-methyl benzoic acid
Supercritical drying Oil-shale
1 Introduction
Organic aerogels are precursors for the preparation of
carbon aerogels, which are highly nanoporous materials
with large specific surface areas. Their porous structure of
interconnected pores allows them to be used as adsorptive
materials [1–3]; when doped with precious metal, carbon
aerogels become catalyst carriers [4–6] and can be used as
electrodes for electrical double-layer capacitors [7, 8] and
fuel cells [9, 10]. Carbon aerogels are electrically con-
ductive and can be used as electrodes for electrical actua-
tors [11].
Well known is an organic aerogel preparation method
described by Pekala [12], by which aerogels were prepared
from resorcinol–formaldehyde gels that were formed in
basic aqueous solution and dried by supercritical CO
2
extraction. Before the extraction water, which is poorly
miscible with CO
2
, was exchanged for acetone prior to
drying. There are a few examples where, in order to sim-
plify the procedure, CO
2
miscible alcohols have been used
as solvents for gel formation [13, 14].
Mulik et al. have synthesized aerogels from resorcinol–
formaldehyde under acid conditions (HCl) in acetonitrile. In
their experiments, acetonitrile was exchanged for acetone
prior to supercritical CO
2
drying. [15] Recently, the same
group replaced HCl with hydrated metal ions as Brønsted
acids for the catalysis of the resorcinol–formaldehyde
gelation in acetonitrile/ethanol, resulting in interpenetrating
networks of metal oxides and resorcinol–formaldehyde
[16]. The mixed-gels were again solvent-exchanged with
acetone, however, this work is conceptually analogous to
the incorporation of the catalyst to the gel reported here. In
accord with the study of A.W. Francis on phase equilibria of
ternary systems with liquid carbon dioxide, acetonitrile and
water [17], it is possible directly to apply CO
2
extraction to
A.-L. Peikolainen (&) R. Aav M. Koel
Department of Chemistry, Tallinn University of Technology,
Akadeemia tee 15, 12618 Tallinn, Estonia
e-mail: annnaliiisa@gmail.com
O. Volobujeva
Department of Materials Science, Tallinn University
of Technology, Ehitajate tee 5, 19086 Tallinn, Estonia
M. Uibu
Laboratory of Inorganic Materials, Tallinn University
of Technology, Ehitajate tee 5, 19086 Tallinn, Estonia
123
J Porous Mater
DOI 10.1007/s10934-011-9459-8
the gel that contains acetonitrile. This has been done by
Willey et al. [18], who, in order to dry the humic acid gel,
replaced the original solvent (water) in the gel by acetoni-
trile prior to supercritical CO
2
drying.
In our previous studies we have been concentrated on
using 5-methylresorcinol (5-MR) for aerogel preparation
[5, 19] where we have been successful in synthesizing the
gels in water and in methanol in the presence of inorganic
base catalysts. The scope of this work was to develop a
simple and fast procedure for producing aerogels from
5-MR in acetonitrile (ACN)—a solvent that does not
require solvent exchange prior to drying. In order to pre-
vent inorganic impurities an organic catalyst was desired.
Acetic acid [20], oxalic acid and para-toluenesulfonic acid
[21] have been used in aqueous systems as catalysts for
aerogel preparation earlier. Here we propose a different
efficient acidic organic catalyst, 2,6-dihydroxy-4-methyl
benzoic acid (dHMBA). dHMBA has a very similar
molecular structure to the monomer, 5-MR, and for this
reason we expected that that derivative of benzoic acid,
besides being a catalyst, would also take part in the sol–gel
polycondensation as a co-monomer. Therefore, dHMBA
would get incorporated in the aerogel backbone and it
would not have to be removed after gelation, thus elimi-
nating any need for time-consuming post-gelation washes.
To confirm this hypothesis, NMR spectra were gradually
recorded during the gelation process. The amount of cat-
alyst in the sol has been found to be very important in
tailoring the structure of the resulting gel: an increase in the
amount of the catalyst in the sol leads to smaller gel par-
ticles, smaller pores and a larger specific surface area [22–
25]. To test the influence of the 5-MR/dHMBA ratio on gel
formation and on the morphology of the resulting aerogel
material, a series of organic aerogels from 5-MR and
dHMBA were synthesized.
Also benzoic acid and alternative hydroxy derivatives of
benzoic acid such as 4-hydroxybenzoic acid, 2,5-dihy-
droxybenzoic acid and 2,6-dihydroxybenzoic acid were
tested as potential catalysts for gel preparation. A
homogenous gel was obtained when dHMBA or 2,6-
dihydroxybenzoic acid was used. Our further study was
focused on the use of dHMBA because it led to substan-
tially faster gelation compared to 2,6-dihydroxybenzoic
acid.
2 Experiment
2.1 Materials
5-Methylresorcinol (99.6%) and 2,6-dihydroxy-4-methyl
benzoic acid ([99%) are by-products of the local oil shale
processing industry and were obtained from Carboshale
OU
¨
, Estonia. Formaldehyde, 37 wt% solution in water
(stabilized with 10–15% of methanol), was purchased from
Sigma–Aldrich. Acetonitrile, HPLC grade, came from
Rathburn Chemicals Ltd., Germany.
The supercritical extraction system with a double clamp
autoclave, 100 ml in volume, was constructed by NWA
Analytische Meßgera
¨
te GmbH, Germany. CO
2
(99.8%)
was obtained from AS Eesti AGA.
2.2 Analysis techniques
Infrared (IR) measurements were performed with a Spec-
trum BX FT-IR System (Perkin Elmer) in the 4,000 to
400 cm
-1
range by the KBr disk method. Nuclear magnetic
resonance (NMR) spectra were recorded on a Bruker 800
Avance III NMR spectrometer. Solvent peaks of CD
3
CN in
13
C (118.69 ppm) and
1
H (1.94 ppm) NMR were used as
chemical shift references. The morphology of the materials
was determined using a Zeiss ULTRA 55 high resolution
scanning electron microscope (SEM), and nitrogen
adsorption analyses were performed using a Sorptometer
KELVIN 1042 built by Costech International. Helium was
used as a carrier gas, nitrogen as the adsorptive gas. The
specific surface area (S
BET
) was calculated according to the
Brunauer-Emmet-Teller theory. Shrinkage was calculated
as a decrease in the cross-sectional area of the cylindrical
gel monolith during drying (Eq. 1).
Shrinkage; % ¼
A
gel
A
OA
A
gel
100 ð1Þ
A
gel
and A
OA
are the cross-sectional areas of the gel and
the corresponding aerogel, respectively.
2.3 Aerogel preparation
Appropriate amounts of the aromatic precursors 5-MR and
dHMBA were dissolved in acetonitrile, and a formalde-
hyde solution (in water) was added subsequently. The
molar ratio of aromatic monomers to formaldehyde (FA)
was fixed at 0.5 and the molar ratio of the solvent (H
2
Oin
FA solution and ACN) to the aromatic monomers was kept
at 50. The molar ratio of 5-MR to dHMBA was varied from
90 and 10 mol% (indicated as 90/10) to 25 and 75 mol%
(indicated as 25/75) of the total molar amount of aromatic
monomers in the sol, respectively. Sol–gel polycondensa-
tion occurred at 25 °C within 3 h, depending on the com-
position of the sol. Gel formation was affirmed by tilting
the sol-filled test tube 45° while observing the movement
of the surface of the sol.
After aging the gel a minimum of 24 h, it was removed
from the test tube and placed directly into the autoclave,
J Porous Mater
123
where the drying of the material resulted in an organic
aerogel. An optimized three-step process was followed for
drying. First, the gel was introduced to liquid CO
2
at a
pressure of 200 bars at 25 °C for 20 min to fill the pores of
the gel with liquid CO
2
and mix it with ACN. The exit
valve of the autoclave was then opened and the internal
pressure reduced to 100 bars, allowing the liquid CO
2
to
flow through the gel at a constant 100 bars at 25 °C for 2 h
to replace the mixture of CO
2
and ACN with CO
2
. After
replacement, the temperature inside the autoclave was
raised to 45 °C and supercritical CO
2
extraction (SCE) was
carried out for 2 h. The extraction was completed by de-
pressurising the autoclave to atmospheric pressure, and
then lowering the temperature in the autoclave to ambient
temperature.
3 Results and discussion
The correlation between the ratio of 5-MR to dHMBA and
the morphology of the resulting aerogel is complicated due
to the dual role of dHMBA in the system. A detailed
mechanism of electrophilic aromatic substitution on res-
orcinol is being proposed by Mulik et al. [15]. Under acidic
conditions, the mechanism of polymer formation starts
with protonation of formaldehyde followed by nucleophilic
attack by the p-system of aromatic monomer. Reactive
positions of the aromatic ring are the 2-, 4- and 6- positions
in the case of the 5-MR ring, and the 3- and 5- ring posi-
tions in the case of dHMBA due to the electronic properties
of the substituents. This substitution leads to the formation
of hydroxymethylated monomers. Further condensation of
the hydroxymethyl group with an unsubstituted site in the
aromatic ring results in CH
2
bridge formation between the
aromatic monomers [15].
The dual behaviour of dHMBA can be seen from the
effect of the 5-MR/dHMBA ratio on the gel formation time
(Fig. 1).
At lower dHMBA concentrations (10–25%) the speed of
polycondensation increased as the amount of catalyst
(dHMBA) was increased. At the same time, the increased
amount of co-monomer (dHMBA) with two available
reaction sites slows the formation of a solid cross-linked
network. This effect could be noticed when the amount of
dHMBA was increased from 25 to 40%. From 40% of
dHMBA the rate of gelation again increased which may be
related to the formation of a different type of polymer
structure (shown below). The gel formation time as low as
97 min was achieved when the amount of dHMBA
exceeded the amount of 5-MR three times.
The function of dHMBA as a monomer was tested by
preparing a sol from dHMBA with FA in ACN without
5-MR in the mixture. A transparent yellow sol turned
opaque in 1 h, but did not lead to the further formation of a
solid gel network. This is probably because of the bifunc-
tionality of dHMBA: it is able to form a linear polymer but
not a three-dimensional network. The sample was dried
under ambient conditions and the remaining solid was then
analysed by IR spectrometry. In Fig. 2, the IR spectrum of
the sample is compared to the spectrum of the gel from
5-MR and FA in water prepared according to a formula
described by Pe
´
rez-Caballero et al. [26].
At wavenumbers 1470 and 2930 cm
-1
, the absorption of
C–H bonds from methylene bridges can be recognized, and
the absorption at 1110 cm
-1
corresponds probably to C–O
bending vibrations of formaldehyde condensation products
such as polymethylene glycol, hemiformal etc. [27], how-
ever, in the case of base-catalyzed 5-MR-FA aerogel the
absorption could as well correspond to C–O from the
methylene ether bridge linking the aromatic rings[28].
The absorption bands at 1680 cm
-1
and 1200 cm
-1
indi-
cate the C=O and C–O bonds from carboxyl group of
dHMBA, which are also present in the spectrum of pure
dHMBA (Fig. 3).
The evolution of gel formation was monitored using
NMR spectroscopy to confirm that dHMBA is a
Fig. 1 Gel formation time at different %5-MR/%dHMBA ratios at
25 ± 1 °C (the exact value of %dHMBA is depicted next to each data
point)
Fig. 2 IR spectra of 5-MR-FA and dHMBA-FA polymers
J Porous Mater
123
bifunctional molecule and that polymerization does not
occur via the carboxyl group. A sample with a 5-MR/
dHMBA molar% ratio of 50/50 was used. (Fig. 4) The
spectra indicated as 0 min (
13
C,
1
H) were measured before
adding FA solution to the sample, and the subsequent
spectra were recorded during 142 (
13
C NMR) and 120
(
1
H NMR) minutes after the addition of FA. The poly-
merization proceeded during the NMR measurements. The
formation of new compounds can be seen after the addition
of FA and the disappearance of all aromatic signals within
2 h, only intermediates of the sol formation could be
recorded in solution phase. The new signals on the spectra
between 83–95 and 50–57 ppm are from FA solution and
can be attributed to condensation products of FA and
methanol. On the
13
C NMR spectra, the new signals of
esters were expected to appear in the area of lower fre-
quency than a signal of carboxylic carbon in parent acid
(\172 ppm).
On the enlargement of the spectral fragment recorded
after 35 min can be seen that the only new signals corre-
sponding to the carboxylic carbons were resonating at
higher frequency (173 ppm) than its parent acid. These
signals can be attributed to newly formed carboxylic acids,
which are intermediates of polymerization and the ester
formation was not observed.
The same can be deduced from the
1
H NMR spectra
(Fig. 5). All monomer peaks disappear within 2 h. It can
also be followed that the acidity of the solution phase
decreases during the reaction, as resonating frequency of
protic hydrogens (a broad singlet) decreases from 4.3
(6 min) to 3.4 ppm (120 min). On the spectrum recorded
before adding an aqueous solution of FA, the signals of
hydrogens H3
0
and H5
0
are overlapping with a signal from
protic hydrogens (6.4 ppm). The signals at 4.6–4.9 and 3.3
and 3.4 correspond to hydrogens from FA and methanol
polycondensates.
These results are in a good agreement with the gel
formation time data. Also, faster consumption of 5-MR
than dHMBA can be followed on
13
C and
1
H spectra,
which is due to the deactivating influence of carboxyl
group of dHMBA in aromatic substitution.
Measurements of nitrogen sorption on organic aerogels
showed that an increase in the amount of catalyst increases
the specific surface area of the material and the total vol-
ume of the pores only in the limited range of the dHMBA
concentration. The largest specific surface area and the
largest total pore volume, 600 m
2
/g and 800 mm
3
/g,
respectively, were achieved at 5-MR/dHMBA ratio of
75/25 as can clearly be seen in Fig. 6. There were no
micropores (pores with diameter less than 2 nm) in 5-MR-
dHMBA-FA organic aerogels.
On SEM micrographs (Fig. 7), it can be seen that the
morphology of the materials at ratios 75/25 and 25/75 are
completely different. When the amount of 5-MR prevails,
the structure of the aerogel is composed of uniform
spherical clusters (diameters 24–25 nm), which is similar
to that of 5-MR-FA aerogels prepared in basic aqueous
Fig. 3 IR spectrum of raw dHMBA
Fig. 4 a
13
C NMR spectra of 5-MR-dHMBA-FA sol in CD
3
CN recorded during142 min after the addition of FA, b enlargement of the spectra
recorded 35 min after the addition of FA; 200 scans per spectrum (201 MHz)
J Porous Mater
123
solvent [5]. Increasing the percentage of dHMBA led to the
presence of large macropores due to the formation of
strings of clusters. The strings of clusters with a broad size
distribution can already be observed at a 5-MR/dHMBA
ratio of 60/40 which correlates well with the results of gel
formation time measurements. At 25/75, the cluster size
distribution is again quite uniform, with diameters of
approximately 15 nm. The morphology of aerogels 50/50
and 25/75 appears to be similar. The strings of clusters are
longer with a higher amount of dHMBA, and the mac-
ropores are larger. Although the yellow colour of the sols
intensified when the percentage of dHMBA was increased,
all the gels in the examined range were transparent and
equally bright orange in colour. After drying, the trans-
parency remained to some extent, while aerogels with a
higher percentage of 5-MR were darker than those with a
higher percentage of dHMBA, the colour ranging from
dark red to bright orange, respectively.
The densities of the materials correspond with the
results of SEM analysis: by increasing the percentage of
dHMBA in the sol, increasingly large macropores form in
the gel and a material with a lower density is achieved
(Fig. 8). Increasing macroporosity facilitates easy removal
of the solvent from the gel leading to smaller shrinkage of
the material during drying. However, from 5-MR/dHMBA
ratio of 30/70 the density and shrinkage started to increase
which is most likely due to increasingly tenuous network
which is not supporting the overall shape of the monolith
sufficiently. Aerogels become brittle. Along with the vol-
ume decrease during drying, the density of the aerogel
increases.
Fig. 5
1
H NMR spectra of
5-MR-dHMBA-FA sol in
CD
3
CN recorded during
120 min after the addition of
FA; 4 scans per spectrum
(800 MHz)
Fig. 6 Specific surface area (S
BET
) and total volume of pores (V
tot
)
of aerogels with different ratios of %5-MR/%dHMBA (the exact
value of %dHMBA is depicted next to the data point)
Fig. 7 SEM micrographs of aerogels prepared at different 5-MR/dHMBA ratios
J Porous Mater
123
4 Conclusions
The 5-MR-dHMBA-FA system was found to be suitable
for forming gels. The use of acidic organic co-monomer as
a catalyst allowed producing organic aerogels free from
inorganic impurities. Acetonitrile as a solvent enabled the
material to be dried with supercritical carbon dioxide
without the need of a solvent exchange prior to drying.
This modification saves time and the consumption of
organic solvents is substantially reduced. Because dHMBA
has a dual role in polymerization, the ability to tune the
specific surface area and porosity of an aerogel by varying
the concentration of dHMBA in the sol is limited; however,
it is important to study how this material behaves during
pyrolysis, by which the organic aerogel will be turned into
a usable carbon aerogel.
Acknowledgments The authors would like to thank Tiiu Kailas for
the IR spectroscopy measurements, Kristiina Kreek for preparing
many samples for the study, To
˜
nis Pehk and Marina Kudrjas
ˇ
ova for
the NMR analysis, and Rein Kuusik for fruitful discussions. The
financial support of ETF grant 7303 is gratefully acknowledged.
References
1. A.K. Meena, G.K. Mishra, P.K. Rai, C. Rajagopal, P.N. Nagar, J.
Hazard. Mater. B122, 161 (2005)
2. F.J. Maldonado-Ho
´
dar, C. Moreno-Castilla, F. Carrasco-Marı
´
n,
A.F. Pe
´
rez-Cadenas, J. Hazard. Mater. 148, 548 (2007)
3. D. Wu, Z. Sun, R. Fu, J. Appl. Polym. Sci. 99, 2263 (2006)
4. S. Martı
´
nez, L. Martı
´
n, E. Molins, M. Moreno-Man
˜
as, A. Roig,
Vallribera Monatshefte fu
¨
r Chemie 137, 627 (2006)
5. F. Pe
´
rez-Caballero, A.-L. Peikolainen, M. Uibu, M. Herbert, A.
Galindo, F. Montilla, M. Koel, Oil Shale 26, 28 (2009)
6. C. Moreno-Castilla, F.J. Maldonado-Ho
´
dar, Carbon 43, 455
(2005)
7. L. Zhang, H. Liu, M. Wang, W. Liu, Rare Metals 25, 51 (2006)
8. S.J. Kim, S.W. Hwang, S.H. Hyun, J. Mater. Sci. 40, 725 (2005)
9. N. Job, S. Berthon-Fabry, M. Chatenet, J. Marie, M. Brigaudet,
J.-P. Pirard, Top. Catal. 52, 2117 (2009)
10. J. Marie, S. Berthon-Fabry, M. Chatenet, E. Chainet, R. Pirard, N.
Cornet, P. Achard, J. Appl. Electrochem. 37, 147 (2007)
11. J.Torop, J. Leis, M Arulepp, U. Juhanson, A. Aabloo. Estonian
Patent EE200800042 (2010)
12. R.W. Pekala. US Patent 4873218 (1989)
13. K. Barral, J. Non-Cryst. Solids 225, 46 (1998)
14. G.Qin, S. Guo, Carbon 39, 1929 (2001)
15. S. Mulik, C. Sotiriou-Leventis, N. Leventis, Chem. Mater. 19,
6138 (2007)
16. N. Leventis, N. Chandrasekaran, A.G. Sadekar, S. Mulik, C.
Sotiriou-Leventis, J. Mater. Chem. 20, 7456 (2010)
17. A.W. Francis, J. Phys. Chem. 58, 1099 (1954)
18. R.J. Willey, A. Radwan, M.E. Vozzella, A. Fataftah, G. Davies,
E.A. Ghabbour, J. Non-Cryst. Solids 225, 30 (1998)
19. F. Pe
´
rez-Caballero, A.-L. Peikolainen, M. Koel, M. Herbert, A.
Galindo, F. Montilla, Open Petrol. Eng. J. 1, 42 (2008)
20. R. Brandt, R. Petricevic, H. Pro
¨
bstle, J. Fricke, J. Porous Mat. 10,
171 (2003)
21. D. Faire
´
n-Jime
´
nez, F. Carrasco-Marı
´
n, C. Moreno-Castilla, Car-
bon 44, 2301 (2006)
22. R.W. Pekala, D.W. Schaefer, Macromolecules
26, 5487 (1993)
23. H. Tamon, H. Ishizaka, M. Mikami, M. Okazaki, Carbon 35, 791
(1997)
24. H. Tamon, H. Ishizaka, T. Araki, M. Okazaki, Carbon 36, 1257
(1998)
25. J. Shen, J. Hou, Y. Guo, H. Xue, G. Wu, B. Zhou, J. Sol-Gel Sci.
Techn. 36, 131 (2005)
26. F. Pe
´
rez-Caballero et al., Micropor. Mesopor. Mater. 108, 230
(2008)
27. I. Poljans
ˇ
ek, U. S
ˇ
ebenik, M. Krajnc, J. Appl. Polym. Sci. 99, 2016
(2006)
28. K. Siimer, T. Kaljuvee, P. Christjanson, T. Pehk, I. Saks, J.
Therm. Anal. Calorim. 91, 365 (2008)
Fig. 8 Shrinkage during supercritical CO
2
drying and the densities of
aerogels at different %5-MR/%dHMBA ratios
J Porous Mater
123
