Strengthening Network of Polyacrylic Acid/Silica

Nanocomposite Hydrogels

Zilu Liu,1 Juan Du,1 Yun Tan,1 Liqin Cao,1 Shimei Xu ,2 Jianbin Huang1,3
1
Key Laboratory of Oil and Gas Fine Chemicals, Ministry of Education, Xinjiang University,
Urumqi 830046, China

2
College of Chemistry, Sichuan University, Chengdu 610064, China

3
Beijing National Laboratory for Molecular Sciences (BNLMS), College of Chemistry and Molecular
Engineering, Peking University, Beijing 100871, People’s Republic of China

The polyacrylic acid (PAA)/silica nanocomposite hydro- distributing load to other polymer chains when one polymer
gels with high mechanical property were prepared by chain fails under large deformation [8]. Although the clay-
combining sol-gel process and in situ free radical poly-
merization using 3-methacryloxypropyltrimethoxysilane
based hydrogels exhibit high stretchability, low modulus
(MPTMS) as silane and acrylic acid (AA) as monomers. might be one potential drawback in some application where
The structure-property relationship of nanocomposite slight deformation is required. In contrast, silica-based
hydrogels was characterized by FT-IR, DMA, TGA, XRD, hydrogels exhibit high modulus by strong adsorption of
TEM, and static mechanical testing. The dependence of polymer chains onto silica particles [5, 9]. However, in
mechanical property on the network parameter was
investigated. The elastic modulus and compression
some systems where the polymer chains do not interact
fracture stress of PAA/silica nanocomposite hydrogels with the silica particles, large increases in strength, and
reached 1.27 and 1.35 MPa at molar ratio of MPTMS to modulus are not observed even after silica are introduced
AA of 0.034. The enhancement of mechanical property [10, 11].
was attributed to the multiple cross-linking networks The mechanical property of silica-based nanocomposite
between MPTMS and AA by covalent bonding and hydro-
gen bonding. POLYM. COMPOS., 00:000–000, 2017. V C 2017
hydrogels was heavily dependent of the compatibility and
Society of Plastics Engineers interaction between nanoparticles and polymer chains. In
the fabrication of the nanocomposite hydrogels, sequential
sol-gel process and free radical polymerization process is
widely used. As a result, weak mechanical properties might
INTRODUCTION be observed due to poor compatibility and heterogeneous
Nanocomposite hydrogels have attracted much attention distribution [12, 13]. The poly(N,N-dimethylacrylamide)/
due to the improved mechanical properties when compared silica nanohybrid hydrogel with nominal stress of only 60
with their unmodified counterparts [1–3]. Recent develop- KPa was synthesized by free radical polymerization of N,N-
ment on the polymer hydrogel-particle system reveals that dimethylacrylamidein silica nanoparticles suspension [14].
physical interaction between polymer and particles (i.e., When acrylamide was used instead of N,N-dimethylacryla-
clay or silica) contributes to the high mechanical stress of mide, the resultant poly(acrylamide)/silica nanocomposite
the final hydrogels [4–7]. In the case of clay-based nano- hydrogels showed a stress of 250 KPa due to improved
composite hydrogels, clay functions as multifunctional interaction between silica and polymers [15]. In above
crosslinker and avoids severe stress concentration by cases, silica nanoparticles were usually used as reinforcing
agents by acting like weak physical crosslinks within a
Correspondence to: S. Xu; e-mail: xushmei@hotmail.com (or) L. Cao;
strong chemical network. Poor compatibility between nano-
e-mail: cao_lq@163.com particles and polymer chains was one of main reasons for
Contract grant sponsor: NSFC-Xinjiang Joint Fund for Local Outstanding the unsatisfactory mechanical properties. To improve the
Youth; contract grant number: U1403392. compatibility, the polyacrylic acid (PAA)/silica nanocom-
Additional Supporting Information may be found in the online version posite hydrogel was prepared by in situ graft polymeriza-
of this article.
DOI 10.1002/pc.24440
tion of acrylic acid (AA) from the surface of modified silica
Published online in Wiley Online Library (wileyonlinelibrary.com). nanoparticles [16]. The hydrogel exhibited tensile strength
C 2017 Society of Plastics Engineers
V of 313 KPa. In this case, silica nanoparticles were modified

POLYMER COMPOSITES—2017
by 3-methacryloxyltrimethoxysilane to improve the com- H-600, Hitachi) with an accelerating voltage of 100 KV.
patibility between silica and polymer by covalent bonding, The infrared spectra of the PAA/SiO2 dispersions and the
but this process led to a heterogeneous distribution of silica hydrogel were performed by Fourier transform infrared
nanoparticles. Instead the poly [N-isopropylacrylamide-co- spectroscopy (EQINOX55, Bruker) in the range of 400–
(3-methacryloxypropyltrimethoxysilane)] (pNS)/silica 4000 cm21 with 2 cm21 spectral resolution. The compres-
hybrid hydrogels were synthesized by on-site gelation of sive stress-strain measurements were performed on cylin-
pNS in suspension of silica nanoparticles, and the hydrogel drical hydrogel samples (about 12.5 mm diameter and
showed a compression stress of 490 kPa [17]. However, so 13.5. mm length) with a static mechanical tester (H5KT,
far mechanical property of the silica-based hydrogels was Tinius Olsen) at room temperature. The crosshead speed
still beyond satisfaction. was 10 mm/min. The thermomechanical properties of the
In attempt to further enhance mechanical property, we pre- nanocomposite hydrogels were measured on a TA DMA
pared the PAA/silica nanocomposite hydrogels with compres- Q800 apparatus. Dynamic frequency spectra of samples
sion fracture stress of 1.35 MPa by one-pot process involving were obtained in compress mode at 308C, at frequency
simultaneous sol-gel reaction and in situ free radical polymeri- ranging from 0.1 to 10 Hz. To ensure linear viscoelastic
zation. The polymer chains were synthesized by copolymeri- response the amplitude of the deformation was set at 30
zation of AA with 3-methacryloxypropyltrimethoxysilane lm. Dynamic temperature spectra of the samples also
(MPTMS), while the silica nanoparticles were produced by were obtained in compress mode at a vibration frequency
sol-gel reaction of MPTMS in the catalysis of AA, which fixed of 1 Hz, at temperatures ranging from 25 to 1158C, at a
the polymer chains as cross-linking points. The diameter of sil- rate of 38C/min. The amplitude of the deformation was
ica nanoparticles was easy to be regulated by controlling the set at 30 lm. X-ray diffraction (XRD) was carried out
ratio of MPTMS to AA. Furthermore, it was worthy to men- using a Bruker D8 Advance X-ray diffractometer (40 KV,
tion that there existed an extra chemical network caused by 40 mA) and a curved graphite crystal monochromator Cu
esterification between AA and MPTMS. The network parame- Ka in the range of 5–758 with the scanning step of 0.028
ters and strengthening mechanism were discussed. and each step stayed for 0.1 s.

EXPERIMENTAL RESULTS AND DISCUSSION

Materials Cross-Linking Network of PAA/Silica Nanocomposite

Hydrogels
AA and MPTMS was purchased from J&K Scientific
Ltd and Aladdin, respectively. Sodium carbonate anhy- During the preparation process of PAA/silica nano-
drous (Na2CO3) was produced by Tianjin Fuchen Chemi- composite hydrogels, copolymerization of MPTMS and
cal Reagent Factory. K2S2O8 (KPS) was received from AA was carried out after initiated by KPS while sol-gel
Hengxing Chemical Reagent Factory. Sodium hydroxide reaction of MPTMS were conducted to produce silica
(NaOH) was purchased from Tianjin Zhiyuan Chemical nanoparticles in the catalysis of AA (Fig. 1). In this case,
Reagent. Deionized water was used in whole experiment. silica functioned as a multifunctional crosslinker and
All reagents were used without further purification. interacted with PAA by covalent bonds.
Characteristic bands of PAA/silica hydrogels were
observed at 1246 cm21 (SiACH2A), 1115 cm21 (SiAOASi),
Synthesis of the Nanocomposite Hydrogels
1705 cm21 (ACOOH), and 2943 cm21 (ACH3) [18–20], sug-
AA (6.8 g) was first mixed with water (5 g), under stir- gesting a successful condensation of MPTMS and copolymeri-
ring at room temperature for 15 min, followed by the addi- zation of AA with MPTMS (Fig. 2a). The polymerization was
tion of Na2CO3 to adjust pH to 3–4. Afterwards MPTMS further confirmed since C@C peak was not observed at 1600–
(the molar ratio of MPTMS to AA was 0.013, 0.034, and 1700 cm21. The amount of silica bridge N was evaluated by
0.056, respectively) was added into the above mixture and the ratio of peak intensity of SiAOASi (I1) to ACOOH (I2)
stirred for 30 min for hydrolysis. Initiator KPS solution under per mole MPTMS, since the number of ACOOH was
(2 wt%, 1 mL) was then added under stirring for 15 min. fixed (Table 1).
The polymerization was carried out at 508C for 48 h to obtain The calculation result showed that with the increasing
PAA/silica nanocomposite hydrogels. The resultant hydro- of ratio of MPTMS to AA from 0.013 to 0.056, the
gels were denoted as (PAA/silica)X, where x was the molar amount of silica bridges decreased instead from 49.15 to
ratio of MPTMS to AA. The synthesis process of PAA was 12.39. Besides, a weak shoulder was observed next to the
same as the above process except free of MPTMS. carboxyl characteristic peak at 1705 cm21. It suggested
that esterification might happen between AA and MPTMS
(Fig. 1d) in some degree besides of condensation between
Characterization
MPTMS (Fig. 1c). To confirm the presence of the ester
The SiO2 morphology in PAA/SiO2 dispersion was groups, the hydrogel (PAA/silica)0.034 was soaked in
investigated by transmission electron micrograph (TEM, dilute NaOH solution (2 mol/L) to neutralize carboxylic

2 POLYMER COMPOSITES—2017 DOI 10.1002/pc

 FIG. 1. Formation of hydrogel network and photographs of nanocomposite hydrogel. [Color figure can be
viewed at wileyonlinelibrary.com]

groups of PAA. As expected (Fig. 2b), the characteristic was evaluated by the ratio of peak intensity of ester
peak of carboxylic groups moved toward low wavenum- groups to ACH3 (Fig. 2b), since the number of ACH3
ber and was observed at 1570 cm21 after neutralization, was fixed. The ratio of peak intensity of ester groups to
so the characteristic peaks of the resultant carboxylate ACH3 when soaked in NaOH solution for 0.5 h was 1.77.
and ester groups could be recognized clearly when soaked However, further to extend the neutralization time to
in NaOH solution for 0.5 h. The amount of ester groups 24 h, the ester peak became weaker (the ratio of peak

FIG. 2. FT-IR spectra of PAA/silica hydrogels (a), the (PAA/silica)0.034 hydrogel soaked in NaOH for dif-
ferent time and the aqueous dispersion of MPTMS and AA completely neutralized with NaOH before; (b).
[Color figure can be viewed at wileyonlinelibrary.com]

DOI 10.1002/pc POLYMER COMPOSITES—2017 3

 TABLE 1. The amount of silica bridge of PAA/silica hydrogels. solution was much higher than the one in water due to
increased repulsion among PAA chains after neutraliza-
Sample I1 I2 I1/I2 N
tion. And the hydrogel became soluble and viscous after
(PAA/silica)0.013 34.682 54.256 0.639 49.15 soaking in NaOH for 4 days. It could be explained as that
(PAA/silica)0.034 19.027 34.300 0.555 16.32 the ester groups were hydrolyzed in the alkaline solution
(PAA/silica)0.056 45.988 68.692 0.669 12.39 and the chemical crosslinking network was broken. In
comparison, PAA hydrogen became soluble and viscous
after soaking in water for 2 days due to broken intramo-
intensity was 1.42) instead due to hydrolysis in alkaline lecular hydrogen bond by water. This result confirmed
solution. In order to ascertain that the ester groups were that there existed chemical network in the hydrogel. In
not from MPTMS but the esterification between AA and fact, there were two kinds of covalent cross-linking
MPTMS, an aqueous dispersion of AA/MPTMS mixture network in PAA/silica hydrogels: one was from radical
was prepared by using completely neutralized AA to pre- polymerization of AA and MPTMS (Fig. 1c), and the
vent the esterification (Fig. 2b). The characteristic peak other was from esterification between carboxylic acid
of ester groups in MPTMS was quite weak (the ratio of of AA and hydroxyl groups of hydrolyzed MPTMS
peak intensity was 1.33), suggesting that the ester groups (Fig. 1d).
in the hydrogel were mainly from the esterification
between AA and MPTMS.
To confirm the role of chemical network from the Network Density of PAA/Silica Nanocomposite Hydrogels
ester groups in nanocomposite hydrogels, hydrogel (PAA/ Storage modulus G0 was always higher than loss mod-
silica)0.034 was soaked to observe the swelling in water ulus G00 and both of them showed weak dependence to
and NaOH solution respectively (Fig. 3). It was found the frequency x (Fig. 4a). It suggested solid-like behavior
that the swelling of the hydrogel increased with the and stable cross-linking network of the nanocomposite
increasing time, but the equilibrium swelling in NaOH hydrogels. However, the change of G0 was not

FIG. 3. Photographs of swelling of hydrogel (PAA/silica)0.034 in water (a–c) and NaOH (d–f), and PAA
hydrogel in water (g–i) for different time: from left to right, the time was 0, 24, and 48 h, respectively.
[Color figure can be viewed at wileyonlinelibrary.com]

4 POLYMER COMPOSITES—2017 DOI 10.1002/pc

 FIG. 4. Dynamic thermomechanical analysis of nanocomposite hydrogels (a) Storage modulus G0 (solid)
and loss modulus G00 (hollow) vs. frequency of PAA/silica hydrogels: (䊏, w) (PAA/silica)0.013, (䉱, D)(PAA/
•
silica)0.034, and ( , 䊊) (PAA/silica)0.056; (b) Storage modulus G0 and loss modulus G00 of (PAA/silica)0.034
vs. temperature. [Color figure can be viewed at wileyonlinelibrary.com]

proportional to the ratio of MPTMS to AA. Considering energy required for hydrogen bonds dissociation, and Xc
G0 was associated with the network chain density of was the physical cross-linking density of hydrogel net-
hydrogels, which was used to evaluate the interaction in work. Since the effective network chain density was cal-
the hydrogels. The effective network chain density N in culated as 13.66 mol/m3 in (PAA/silica)0.034 hydrogel, it
hydrogels was related to the equilibrium modulus Ge could be concluded that the covalent cross-linking acted a
[21]: predominant role in the hydrogel network.

Ge 5NRT
Mechanical Strength of PAA/Silica Nanocomposite
here, Ge was the equilibrium modulus at G0 vs. frequency Hydrogels
curves where the plateau appeared while G00 was much
lower than G0 . R was the gas constant and T was absolute The compression fracture stress of PAA/silica hydro-
temperature. The result showed that N increased from gels were 0.93, 1.35, and 0.83 MPa respectively with the
6.42 to 13.66 mol/m3 and then decreased to 9.42 mol/m3 increasing of ratio of MPTMS to AA from 0.013 to 0.056
with the ratio of MPTMS to AA rising from 0.013 to (Fig. 5a) and the compression fracture deformation
0.056. It suggested that the increase of MPTMS did not decreased from 64.51 to 35.46% accordingly. Compared
lead to monotonous increase in the effective network with pure PAA hydrogel (fracture stress: 2.83 MPa; frac-
chain density. More efficient cross-linking was found in ture strain: 76.04%), PAA/silica hydrogels showed a
hydrogel (PAA/silica)0.034. lower mechanical stress but higher elastic modulus. More-
To further confirm the presence of physical bonds in over, the silica-based hydrogels remained solid-like
the PAA/silica hydrogel, the temperature sweep was per- behavior but PAA hydrogels became sol state when
formed at 1 Hz (Fig. 4b). Both storage modulus G0 and immersed in water. This difference would make the PAA/
loss modulus G00 increased as a whole due to the evapora- silica hydrogels applicable in practice.
tion of water. But in the range of 60–808C, G0 showed an The experimental elastic modulus E of silica nanocom-
abnormal decrease. This abnormal trend could ascribe to posite hydrogels was received from the liner elastic por-
the break of hydrogen bonds among SiAOH and AC@O tion of stress-strain curves (Fig. 5a) and was given by [8]:
groups with the increasing of temperature. The activation
energy Ea and the physical cross-linking density Xc of E5rnominal =ð12kÞ
hydrogel network were calculated as 5.38 kJ/mol and k5H=H0
2.90 mol/m3, respectively, according to the Arrhenius
equation [22]: where E was elastic modulus, rnominal was pressure, H
. was the deformed length and H0 was the original length.
0 
lnG 5lnAR1Ea RT The theoretical elastic modulus could be predicted by
T Guth-Gold equation [8]:
  
Xc 5Aexp Ea RT  
E5E0 112:5U114:1U2
where G0 was the storage modulus, T was temperature, A
was a constant, R was gas constant, Ea was the activation

DOI 10.1002/pc POLYMER COMPOSITES—2017 5

 FIG. 5. Compressive properties of the PAA/silica nanocomposite hydrogels and PAA hydrogel: (a) the
stress-strain curves; (b) the elastic modulus E from compressive testing vs. volume fractions Usilica of silica
at the testing conditions. The solid line was the predicted value of E calculated by Guth-Gold model. [Color
figure can be viewed at wileyonlinelibrary.com]

FIG. 6. TEM images of silica in hydrogel (a) (PAA/silica)0.013; (b) (PAA/silica)0.034; and (c) (PAA/silica)0.056.

where U was the silica volume fraction and E0 was the weak interaction. The diameter r of silica nanoparticles
elastic modulus of polymer chains. This equation was from hydrogel (PAA/silica)0.013, (PAA/silica)0.034, and
applied to evaluate the elastic modulus of network with (PAA/silica)0.056 with 516, 29, and 43 nm (Fig. 6) and the
specific surface d was calculated from d 5 6000/r. The den-
sity q could be calculated from q 5 3/(dR)[8], where R was
gas constant. The amount of silica m in each hydrogel was
obtained using TGA (Supporting Information Fig. S1).
Then, the silica volume fraction U was calculated from
U 5 m/(qV)[8], where V was the volume of hydrogel. The
experimental and calculated values of elastic modulus were
showed in Fig. 5b. The experimental value deviated from
Guth-Gold model obviously, and it strongly suggested that
the introduction of silica nanoparticles increased the inter-
action and the elastic modulus was proportionally enhanced
to the mole ratio of MPTMS to AA. TEM observation also
showed that the dispersity of nanoparticles was improved
with the increasing of the mole ratio of MPTMS to AA.
Besides that, the morphology of (PAA/silica)0.034 (Fig. 6b)
was different from (PAA/silica)0.013 and (PAA/silica)0.056
FIG. 7. XRD patterns of PAA/silica nanocomposite hydrogels. [Color where SiO2 were observed as single particle. In contrast,
figure can be viewed at wileyonlinelibrary.com] there are some SiO2 particle clusters observed in hydrogel

6 POLYMER COMPOSITES—2017 DOI 10.1002/pc

 FIG. 8. Schematic illustration of structure of silica/polymer nanoparticle hydrogels. The silica bridges
between silica nanoparticle and ester groups (blue part) constituted an elastic network, whereas the hydrogen
bond (dashed line) resulted in viscoelastic properties. [Color figure can be viewed at wileyonlinelibrary.com]

(PAA/silica)0.034 instead from the enlarged image (Fig. 6b) appear in a heterogeneous distribution system. For a well
and it was possibly caused by the small particle size of SiO2 distributed system, the mechanical property was strongly
(29 nm). related to the interaction [13, 25, 26]. In this work, PAA/
The size of silica nanoparticles was observed as 516, silica was prepared without additional chemical cross-
29, and 43 nm at the mole ratio of MPTMS to AA was linker. However, two chemical networks formed through
0.013, 0.034, and 0.056, respectively. In fact, it was rela- copolymerization of AA and MPTMS, as well as esterifica-
tive content of MPTMS to water which acted an impor- tion between AA and MPTMS. Besides, self-crosslinking
tant role on the silica size [23]. With the increasing of of PAA or hydrogen-bonding between PAA and silica led
molar ratio of MPTMS to AA from 0.013 to 0.034, the to physical network (Fig. 8). Different to previous system in
diameter of silica nanoparticles decreased by more than which the silica nanoparticles were used as reinforcing
90%. In low molar ratio of MPTMS to AA, the amount agents by acting like weak physical cross-links within a
of silica bridges was high and a lot of small subparticles strong chemical network, our work presented the formation
were prepared in a short period. But, the hydrogen bond of silica-based chemical network mixed with small portion
of silica subparticles under a low molar ratio was stronger of physical crosslinks. The strong interface between silica
than that of a higher molar ratio due to more water con- particles and polymer matrix greatly affected mechanical
tent in the former case. Therefore, subparticles in the strength of the hydrogel.
lower molar ratio were inclined to agglomerate with each At the low molar ratio of MPTMS to AA, the interaction
other and grew into large particles [24]. However, when between nanoparticles and network chains improved the
the molar ratio of MPTMS to AA was higher than the elasticity of hydrogel network (Fig. 8). The increasing of
0.034, the silica subnanoparticles would agglomerate molar ratio in some degree enhanced the effective network
again due to the surface effect and the thus the silica size chain density, which was the most vital factor to decide the
increased instead. mechanical property of nanocomposite hydrogels. How-
The amorphous peaks of organic polymers were ever, further to increase the molar ratio, the diameter of sil-
observed at 2h 5 37.768, while the diffraction peaks of sil- ica nanoparticles increased slightly due to surface effect
ica nanoparticles were 17.778, 18.028, and 18.978, respec- while the effective network chain density decreased. Vol-
tively, with the increasing of mole ratio of MPTMS to AA ume exclusion effect of silica nanoparticles limited free
from 0.013 to 0.056 (Fig. 7). According to Bragg’s equa- mobility of polymer chains, and led to the loss of elasticity
tion, nk 5 2dsinh, where k was the wavelength of the X-ray of nanocomposite hydrogel. In contrast, the hydrogen-
of Cu Ka radiation, and d was the d-spacing. The d-spacing bonding and close entanglement of long polymer chains
was 0.50, 0.49, and 0.47 nm accordingly. It suggested the was the main interaction in PAA hydrogel and the deforma-
network chains were immobilized around the silica nano-
tion and free mobility were not limited due to the weak
particles because of the increased interaction.
interaction. And it led to the poor elasticity of the hydrogel.

Strengthening Mechanism of PAA/Silica CONCLUSIONS

Nanocomposite Hydrogels
The PAA/silica nanocomposite hydrogels had been pre-
In principle, mechanical property of nanocomposite pared by simultaneous sol-gel reaction and in situ free radi-
hydrogels was dependent of the distribution level of nano- cal polymerization. At molar ratio of MPTMS to AA of
particles, since the stress concentration was inclined to 0.034, the nanocomposite hydrogel showed high

DOI 10.1002/pc POLYMER COMPOSITES—2017 7

compression stress of 1.35 MPa and elastic modulus of 1.27 11. W.-C. Lin, A. Marcellan, D. Hourdet, and C. Creton, Soft
MPa. The chemical bonding between AA and MPTMS Matter, 7, 6578 (2011).
(i.e., copolymerization of double bonds, and esterification) 12. L. Carlsson, S. Rose, D. Hourdet, and A. Marcellan, Soft
led to the formation of chemical network and improve the Matter, 6, 3619 (2010).
compatibility between silica and polymers. The chemical 13. J. Yang, C.-R. Han, J.-F. Duan, F. Xu, and R.-C. Sun, J.
network was greatly contributed to high modulus of the Phys. Chem. C, 117, 8223 (2013).
nanocomposite hydrogels. The structure-property relation- 14. S. Rose, A. Dizeux, T. Narita, D. Hourdet, and A.
ships developed in this work would greatly contribute to the Marcellan, Macromolecules, 46, 4095 (2013).
understanding of nanocomposite hydrogels. 15. J. Yang, L.-H. Deng, C.-R. Han, J.-F. Duan, M.-G. Ma, X.-
M. Zhang, F. Xu, and R.-C. Sun, Soft Matter, 9, 1220 (2013).
REFERENCES AND CITED WORK 16. F.-K. Shi, X.-P. Wang, R.-H. Guo, M. Zhong, and X.-M.
Xie, J. Mater. Chem. B, 3, 1187 (2015).
1. X. Nie, A. Adalati, J. Du, H. Liu, S. Xu, and J. Wang, 17. M. Ashraful Alam, M. Takafuji, and H. Ihara, J. Colloid
Appl. Clay Sci., 97–98, 132 (2014). Interface Sci., 405, 109 (2013).
2. J. Du, R. Wu, H. Liu, X. Nie, H. Li, S. Xu, and J. Wang, 18. M. H. Huang, H. M. Soyez, B. S. Dunn, J. I. Zink, Chem.
Polym. Compos., 36, 538 (2015). Mater., 12, 231–235 (2000).
3. P. Chen, S. Xu, R. Wu, J. Wang, R. Gu, and J. Du, Appl. 19. N. Fumito, J.M. Mckiernan, D. Bruce, J.I. Zink, B.C.
Clay Sci., 72, 196 (2013). Jeffrey, and A.J. Hurd, J Am Ceram. Soc., 78, 1640 (1995).
4. M. Zhong, F.-K. Shi, Y.-T. Liu, X.-Y. Liu, and X.-M. Xie, 20. C.J. Brinker, G.C. Frye, A.J. Hurd, and C.S. Ashley, Thin
Chin. Chem. Lett., 27, 312 (2016). Solid Films, 201, 97 (1991).
5. L. Petit, L. Carlsson, S. Rose, A. Marcellan, T. Narita, and 21. L. Xiong, X. Hu, X. Liu, and Z. Tong, Polymer, 49, 5064
D. Hourdet, Macromol. Symp., 337, 58 (2014). (2008).
6. A. Giannakas, M. Vlacha, C. Salmas, A. Leontiou, P. 22. Y. Wang, and D. Chen, J. Colloid Interface Sci., 372, 245
Katapodis, H. Stamatis, N.M. Barkoula, and A. Ladavos, (2012).
Carbohydr. Polym., 140, 408 (2016). 23. L.I.U. Jian-hong, Y. jie, H. min, and L. Sheng-Jun, Chin. J.
7. C. Teng, J. Qiao, J. Wang, L. Jiang, and Y. Zhu, ACS Colloid Polym., 28, 19 (2010).
Nano, 10, 413 (2016). 24. K.D. Kim, and H.T. Kim, J. Sol-Gel Sci. Technol., 25, 183
8. W.-C. Lin, W. Fan, A. Marcellan, D. Hourdet, and C. (2002).
Creton, Macromolecules, 43, 2554 (2010). 25. J.C. Tan, and A.K. Cheetham, Chem. Soc. Rev., 40, 1059
9. J. Wei, J. Wang, S. Su, S. Wang, and J. Qiu, J. Mater. (2011).
Chem. B, 3, 5284 (2015). 26. S. Pavlidou, and C.D. Papaspyrides, Prog. Polym. Sci., 33,
10. J. Yang, X.-P. Wang, and X.-M. Xie, Soft Matter, 8, 1058 (2012). 1119 (2008).

8 POLYMER COMPOSITES—2017 DOI 10.1002/pc