at SciVerse ScienceDirect

Polymer 54 (2013) 542e547

Contents lists available

Polymer

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

Increasing the oxidative stability of poly(dicyclopentadiene) aerogels byhydrogenation

Jeremy M. Lenhardt a,*, Sung Ho Kim a, Art J. Nelson a, Pooja Singhal a, Theodore F. Baumann a,Joe H. Satcher Jr. a,b

aChemical Sciences Division, Lawrence Livermore National Laboratory, Livermore, CA 94550, USAbNanoscale Synthesis and Characterization Laboratory, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA

a r t i c l e i n f o

Article history:Received 1 August 2012Received in revised form27 November 2012Accepted 1 December 2012Available online 6 December 2012

Keywords:Poly(dicyclopentadiene)AerogelsHydrogenation

* Corresponding author.E-mail address: lenhardt2@llnl.gov (J.M. Lenhardt

0032-3861/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.polymer.2012.12.002

a b s t r a c t

Ring opening metathesis polymerization (ROMP) of cycloolefins is a promising new route for thepreparation of polymeric aerogels. The resulting unsaturation in the polymer backbone, however, makesthese particular systems susceptible to oxidative degradation under ambient conditions. One method toincrease the oxidative stability of these aerogels is to hydrogenate the material. In the present study,hydrogenation of poly(dicyclopentadiene) gels was achieved through thermolysis of para-toluenesulfonyl hydrazide in the presence of tripropylamine followed by solvent exchange and super-critical drying to form the hydrogenated aerogel (H-pDCPD). Aerogels were prepared with varyingdensities and were characterized by FTIR-ATR, elemental analysis, BET, SEM, XPS, DSC and TGA. Theoxidative stability of H-pDCPD aerogels over pDCPD was investigated through thermolysis in the pres-ence of atmospheric oxygen. We report herein the synthesis and characterization of this new material.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

There is increasing interest in the study of aerogels, a class ofmaterials exhibiting low density, high porosity, high surface areaand low thermal conductivity [1e5]. Aerogels are typicallysynthesized by taking a wet gel precursor and subjecting thematerial to supercritical drying, this process negates the effects ofcapillary stresses that otherwise lead to gel destruction orshrinkage upon solvent removal. The result is a solid networkwherein the liquid has been replaced with the ambient atmo-sphere, giving a low-density scaffold. Depending on their chemicalcomposition, aerogels have found some applications as thermal [6]and acoustic insulators [7], supercapacitors [8], and potentialchemical/energy storage platforms [9]. The uses of aerogels are asvaried as their composition; the first aerogels being composed ofsilica [10] followed soon after by the preparation of both transitionmetal [11] and lanthanide oxides [12]. Though ‘classic’ aerogels areoxide-based; researchers have begun to successfully prepareorganic aerogels such as resorcinol-formaldehyde [1], cellulose[13], polyurethane [14], and, of interest to us, poly(-dicyclopentadiene) [15e17].

).

All rights reserved.

Materials based on the polymerization of dicyclopentadiene(DCPD) to poly(dicyclopentadiene) (pDCPD) are gaining interestlikely due to the ease of synthesis, cheap material cost, tolerance ofpolymerization conditions to oxygen and water, and the potentialfor polymer modification through the reaction of pDCPD alkenesalong the polymer backbone [18]. A combination of additionalphysical characteristics such as good physical and electrical prop-erties have provided the impetus for its’ use as a replacementmaterial for reaction injection molding parts [19]. The popularity ofpDCPD and continued interest in aerogels has naturally led to therecent preparation pDCPD aerogels [15e17].

These pDCPD aerogels are easily prepared first by ROMP of DCPDthen supercritical CO2 drying of wet gels to give aerogels withnominal densities ranging from ca. 30e300 mg mL�1. The aerogelsare typically fibrous in nature with large surface areas(>200 m2 g�1) and low thermal conductivity (ca. 20 mW m�1 K�1)creating the possibility of its use as a novel insulation material.

The drawback of pDCPD materials is that the high density ofalkenes along the polymer backbone makes this material suscep-tible to oxidative damage [20]. In many cases, the preparation ofpDCPD polymers is coincident with the formation of small amountsof oxidation products along the polymer. In the case of solid pDCPDfilms, the oxidation process forms an oxide film that preventsfurther oxidation of the inner material [21]; this is likely not thecase for an aerogel wherein the material porosity is larger than 90%

J.M. Lenhardt et al. / Polymer 54 (2013) 542e547 543

and an outer, oxide film will not inhibit oxidation of the innerdomains. To protect these aerogels from oxidative stress, we reportherein that the hydrogenation of pDCPD gels followed by super-critical drying to form hydrogenated H-pDCPD aerogels as a viableroute to such an oxidatively stable aerogel material.

2. Methods

2.1. Preparation of poly(dicyclopentadiene) (pDCPD) gel

In a sample pDCPD gel preparation (30 mg mL�1), 0.36 g dicy-clopentadiene (2.7 � 10�3 mol) was dissolved in 12 mL 1,2-dichlorobenzene and to this solution was added 90 mL of an8 mg mL�1 catalyst solution (0.7 mg Grubbs G1 catalyst, 0.2 wt %)and allowed to gel for 16 h. After 16 h the gel was removed andplaced into a solution of ca. 50 mL 1,2-dichlorobenzene with 1.5 mLethyl vinyl ether and deoxygenated with bubbling N2. Gels werestored in this solution prior to hydrogenation and/or solventexchange prior to supercritical drying. Prior to supercritical drying,pDCPD gels were solvent exchanged in acetone for 3 days with dailychanging of the solvent.

2.2. Hydrogenation of poly(dicyclopentadiene) (H-pDCPD) gel

In a sample hydrogenation reaction, a 30 mg mL�1 pDCPD gelfrom a 1 cm3 glass mold (30 mg pDCPD, 0.46 mmol alkenes) wasremoved from the 1,2-dichlorobenzene/ethyl vinyl ether storagesolution added to 6 mL 1,2-dichlorobenzene. To the solution wasadded 0.68 g (3.7 mmol, 8 eq. vs alkenes) para-toluenesulfonylhydrazide and 2.8 mL tripropylamine (15 mmol, 33 eq. vs. alkenes)and the solution was deoxygenated with bubbling N2. The solutionwas placed into an oil bath at 75 �C under N2 and allowed to reactfor 3 days. After reaction, the gels were immediately placed intoacetone for solvent exchange for 3 days with daily changing of thesolvent.

2.3. Supercritical drying

Supercritical drying was conducted in the same manner for thepreparation of pDCPD and H-pDCPD aerogels. In a Polaron super-critical dryer set to 10 �C was added acetone then the pDCPD orH-pDCPD gels. The Polaron was charged with 800 psi CO2 and theacetone was exchanged with liquid CO2 for 2 days. After 2 days thetemperature was raised to 40 �C and the pressure was maintainedat ca. 1300 psi by slow leaching of the heated CO2. Once thetemperature and pressure stabilized, the supercritical conditionwas maintained for 3 h followed by slow removal of the

Fig. 1. The preparation of poly(dicyclopentadiene) (pDCPD) gels through ring opening metausing para-toluenesulfonyl hydrazide (TSH) results in formation of hydrogenated poly(dicyformation of the pDCPD and H-pDCPD aerogels. Chemical structures are shown to indicatepolymer of which occur in a random fashion.

supercritical fluid by outgassing. The supercritically dried sampleswere stored in vacuum prior to analysis.

2.4. Sample oxidation

Either pDCPD or H-pDCPD aerogels were placed into a YamatoDX 300 drying ovenwith Yamato Hi-Tech temperature controller atthe required temperature. The ovenwas open to the atmosphere atthe top of the oven to allow for ambient oxygen to be present.

2.5. Characterization

IR spectra were recorded on a Perkin Elmer Spectrum GX FTIR-ATR. Spectra are averaged over ca. 15 scans with a resolution of4 cm�1. SEM images were collected on a Jeol JSM-7401F scanningelectron microscope at an acceleration voltage of 2 kV in a lowersecondary electron image (LEI) mode. Surface area determinationand pore volume and size analysis were performed by Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methodsusing an ASAP 2000 surface area analyzer (Micromeritics Instru-ment Corporation). Samples of approximately 0.1 g were placedunder vacuum (10-5 Torr) at room temperature for at least 24 h toremove all adsorbed species. Nitrogen adsorption data were thentaken at five relative pressures from 0.05 to 0.20 at 77 K to calculatethe surface area by BET theory. For BJH analyses, average pore sizeand pore volume were calculated using data points from thedesorption branch of the isotherm. Galbraith Laboratories, Knox-ville, TN conducted the elemental analyses. DSC measurementswere conducted on a Pyris Diamond DSC (Perkin Elmer) on 3e5 mgsamples. The samples were first cooled to �40 �C (20 �C min�1),heated to 300 �C, cooled to �40 �C and heated a second time to300 �C. The reported Tg values are from the half height of transitionduring the second heating cycle. TGA measurements were con-ducted on a Shimadzu TGA-50 thermogravimetric analyzer undera stream of air (10 mL min�1) from 25 �C to 600 �C at a ramptemperature of 20 �C min�1. The X-ray photoelectron spectroscopy(XPS) analysis was carried out using a focused monochromatic AlKa X-ray (1486.7 eV) source for excitation and a spherical sectionanalyzer. A 200 mm diameter X-ray beamwas used for analysis. TheX-ray beam is incident normal to the sample and the X-ray detectoris at 45� away from the normal. The pass energy was 23.5 eV givingan energy resolution of 0.3 eV that when combined with the0.85 eV full width at half maximum (FWHM) Al Ka line width givesa resolvable XPS peak width of 1.2 eV FWHM. The collected datawere referenced to an energy scale with binding energies for Cu2p3/2 at 932.72 � 0.05 eV and Au 4f7/2 at 84.01 � 0.05 eV. Lowenergy electrons were used for specimen neutralization.

thesis polymerization (ROMP) of dicyclopentadiene (DCPD). Hydrogenation of the gelclopentadiene) (H-pDCPD). Supercritical drying (SCD) of these wet gels leads to thethe major species in the ROMP reaction; that of linear and cross-linked sections of the

Fig. 2. Hydrogenation of pDCPD gels was monitored by FTIR-ATR by following resonances at 3050 cm�1, 972 cm�1 and 708 cm�1. Spectra are normalized to the resonance at ca.1432 cm�1. Above, hydrogenation was conducted at 75 �C in the presence of 8 eq. TSH and 32 eq. TPA in 1,2-dichlorobenzene on 30 mg mL�1 pDCPD gels for the annotated numberof days. Attributed resonances are noted above the IR spectra. A plot of the relative IR intensity (right) versus reaction time shows consumption of the IR resonances after ca. 3 daysof reaction. The observed ‘increase’ in intensity for the 3050 cm�1 absorbance is the result of incomplete removal of TSH reactive byproducts from the hydrogenation reaction. Fitprovided to guide the eye.

Table 1Measured densities (d), pore volume (PV), pore diameter (PD) and aerogel surfacearea (SA) for pDCPD and H-pDCPD aerogels. Numbers beneath Sample ID indicatethe mixed density (mg mL�1) of the wet gel prior to hydrogenation and supercriticaldrying.

Sample ID d/mg mL�1 PV/mL g�1 PD/nm SA/m2 g�1

25-pDCPD 30 0.87 13.9 303H-pDCPD 48 0.79 25.5 19230-pDCPD 42 1.02 15.8 303H-pDCPD 71 0.93 18.5 22640-pDCPD 78 1.19 18.5 282H-pDCPD 128 0.84 21.3 16750-pDCPD 109 1.49 22.2 284H-pDCPD 166 1.15 25.2 176

J.M. Lenhardt et al. / Polymer 54 (2013) 542e547544

3. Results and discussion

3.1. pDCPD hydrogenation

To hydrogenate pDCPD gels to H-pDCPD, we used para-tolue-nesulfonyl hydrazide (TSH), awell-established hydrogenation agentthat has found wide utility in polymer hydrogenation reactions[22]. Since we conducted hydrogenation reactions on solid pDCPDgels, we found that the reaction required both longer reactionstimes and lower temperatures to (1) decrease the rate of N2 gasevolution that leads to trapped bubbles in the pDCPD gel and/ordegradation of thematerial and (2) allow the hydrazide to percolatethe gel network to achieve full hydrogenation. Additionally,hydrogenation reactions were conducted in the presence of tri-propylamine (TPA) in order to inhibit the reaction of pDCPD alkeneswith the formed sulfinic acid [22]. For clarity of terminology, wereport through the text themixed concentration of the prepared gelin mg mL�1 followed by the gel name, either pDCPD or thehydrogenated H-pDCPD. For example, a 30 mg mL�1 pDCPD gel isnamed 30-pDCPD and its hydrogenated counterpart as 30-H-pDCPD. The naming does not reflect the measured density ofthe formed aerogel after supercritical drying (Fig. 1).

We first determined the reaction temperature and durationrequired for hydrogenation of the pDCPD gels. Here, 30 mg mL�1

pDCPD [30-pDCPD] gels were subjected to hydrogenation for

Fig. 3. Elemental analysis of pDCPD and H-pDCPD aerogels of varying initial densities (as mixpercentages indicate the degree of hydrogenation as calculated from the CHx balance.

varying times in the presence of 8 equivalents TSH and 32 eq. TPA at75 �C for up to 7 days of reaction in 1,2-dichlorobenzene. The extentof reaction was followed using FTIR-ATR on supercritically driedaerogels by monitoring the intensity of IR absorbances at3050 cm�1 (sp2 ¼ CeH), 972 cm�1 (trans double bond, pDCPDmainchain) and 708 cm�1 (cis cyclic double bond, pDCPD unopenedcyclopentene) [23]. An analysis of the relative IR intensity ofhydrogenated H-pDCPD aerogels (Fig. 2) showed consumption ofalkene absorbance in the IR spectra after ca. 3 days of reactionunder these conditions, forming hydrogenated 30-H-pDCPD. Theeffect of TSH concentrationwas also investigated. Hydrogenation of

ed, not directly measured here), here 100% hydrogenation leads to a CH1.6 material. The

Fig. 4. SEM images from 25-pDCPD (left) and 25-H-pDCPD (right) show that the structure of the formed aerogel is unaffected by hydrogenation of the polymer backbone.

J.M. Lenhardt et al. / Polymer 54 (2013) 542e547 545

30-pDCPD gels was conducted for 3 days at 75 �C wherein thenumber of equivalents (versus alkenes) of TSHwas increased from 1eq. to 10 eq. An analysis of the relative IR intensities of supercriti-cally dried aerogels versus TSH showed a complete reaction whengreater than 6 eq. of TSHwas used in the reaction (see SI, Fig. S1). Toensure complete hydrogenation under these conditions where thereaction temperature is 75 �C, we therefore define our standardhydrogenation condition in which 8 eq. TSH and 32 eq. TPA areallowed to hydrogenate the pDCPD gels for a period of 3 days.

To verify the formation of H-pDCPD aerogels, we next hydro-genated a series of pDCPD gels of varying concentration (25, 30, 40and 50 mg mL�1), supercritically dried this series of H-pDCPDaerogels, and both FTIR-ATR spectra and elemental analysis (Fig. 3)were used to establish the aerogels as H-pDCPD. In all aerogels,FTIR-ATR spectra showed the disappearance of IR absorbances at3050 cm�1, 972 cm�1 and 708 cm�1. Elemental analysis corrobo-rated the initial pDCPD aerogels (pDCPD ¼ C10H12; or CH1.2) ashaving a CH balance of CH1.2. After hydrogenation (H-pDCPD ¼ C10H16; or CH1.6) the best results were obtained for 25-H-pDCPD at CH1.58. The CHx balances for H-pDCPD aerogels decreasedsomewhat with increasing concentration to CH1.58, CH1.52 andCH1.50 for aerogel concentrations of 30, 40 and 50 mg mL�1

respectively indicating an extent of hydrogenation of 97%, 95%, 80%and 74% for aerogels of increasing concentration (Fig. 3).

3.2. Aerogel characterization

The processing of wet gels into aerogels is commonly followedby shrinkage of the wet gel to an aerogel of increased density [16].

Fig. 5. Images (left, numbers indicate mixed concentration/mg mL�1) of pDCPD aer-ogels before and after thermolysis for 2 h at 135 �C in air; FTIR-ATR spectra (right)show formation of large absorbances at 3400 cm�1 and 1700 cm�1. Spectra are overlaidfrom before (- - -) and after ( ) thermolysis.

We therefore monitored the densities of pDCPD and H-pDCPDaerogels, and, in addition, the morphological characteristics of theas-formed aerogels. The pDCPD gels were first prepared in eithercylindrical or square molds, then supercritically dried (SCD) withCO2. After SCD, the samples were weighed and measured todetermine the aerogel density (Table 1). The resulting pDCPD aer-ogels had densities ranging from 30 mg mL�1 (25-pDCPD) to109 mg mL�1 (50-pDCPD), or shrinkage factors (final density O

initial density) that increased from 1.2 to 2.2 (Table 1). Bycomparison, the densities of H-pDCPD aerogels ranged from48 mg mL�1 (25-H-pDCPD) to 166 mg mL�1 (50-H-pDCPD), orshrinkage factors from 1.9 to 3.3. Surface area determination andpore volume and size analysis were next performed by Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods(Table 1). Here, pore volumes of H-pDCPD aerogels were ca. 20%smaller, pore diameters ca. 15% larger, and surface areas ca. 35%smaller than pDCPD aerogels. Additionally, the morphologicalcharacteristics imaged using SEM (Fig. 4), showed that the fibrillarnetwork characteristic of pDCPD aerogels was conserved afterhydrogenation.

3.3. Aerogel thermolysis

The oxidative stability of H-pDCPD aerogels versus pDCPD wasconfirmed by thermal treatment of these materials in an oven in air(ambient environment). Samples were heated to 135 �C for 2 h andthe FTIR-ATR spectra were recorded to check for the formation of

Fig. 6. Thermal treatment of H-pDCPD aerogels shows the lack of formation ofabsorbances at 3400 cm�1 and 1700 cm�1 indicating that these aerogels are increasingstable to oxidation versus pDCPD aerogels. Left, FTIR-ATR spectra before (- - -) and after( ) thermolysis and images (right) of the aerogels before and after thermolysis (initialon left, thermalized on right). Numbers indicate the mixed aerogel concentration inmg mL�1.

Fig. 7. X-ray photoelectron spectroscopy (XPS) survey spectra (top) of 25-pDCPD (left) and 25-H-pDCPD (right) before and after (D) thermolysis for 2 h at 135 �C in air. *Observedshoulders on the oxygen and carbon 1s peaks in the survey spectrum are absent in high resolution scans are so are attributed to sample charging during acquisition. High resolutionscans (bottom, from which quantifiable data are obtained) of the oxygen 1s and carbon 1s regions of the spectrum for both 25-pDCPD and 25-H-pDCPD before (- - -) and afterthermolysis for 2 h at 135 �C.

J.M. Lenhardt et al. / Polymer 54 (2013) 542e547546

carbonyl (ca. 1700 cm�1) and/or hydroxyl (ca. 3400 cm�1) absor-bances. The pDCPD aerogels heated to this temperature and timeunderwent discoloration from white to brown/yellow coincidentwith formation of strong FTIR absorbances at both 3400 cm�1 and1700 cm�1 (Fig. 5). In contrast, H-pDCPD aerogels (Fig. 6) remainednearly white in color (slightly colored, 40- and 50-H-pDCPD likelydue to incomplete hydrogenation) and no new FTIR-ATR absor-bances were observed to indicate formation of any oxygen-containing moieties.

One thermalized sample set consisting of 25-pDCPD and 25-H-pDCPD was further analyzed using X-ray photoelectron spec-troscopy (XPS) to analyze the aerogels for oxygen content (Fig. 7).The XPS spectrum of 25-pDCPD showed 79.2% carbon and 16.3%oxygen with slight contributions from fluorine (4.3%) and chlorine(<1%) (unknown contaminants).

Thermolysis of this aerogel (135 �C, 2 h in air) was followed byan increase in the total oxygen content in the aerogel to 20.5%. TheXPS spectrum of 25-H-pDCPD aerogel showed virtually no increase

Fig. 8. DSC (top) traces of 25-pDCPD and 25-H-pDCPD aerogels show a reduction in Tg for thnot undergo significant weight loss until ca. 460 �C.

in oxygen content after thermolysis (Fig. 7). The hydrogenatedaerogel initially consisted of 93.9% carbon, 4.6% oxygen and smallcontributions (ca. 1.5%) from nitrogen, chloride, sodium and zinc(unknown contaminants). Thermal treatment (135 �C, 2 h) led to anaerogel composition of 95.3% carbon and 4.7% oxygen.

While thermolysis of H-pDCPD aerogels confirmed theiroxidative stability, one interesting effect was a marked shrinkage ofthese aerogels as a result of the applied heat. For example, themeasured density of 25-H-pDCPD was found to be 710 mg mL�1

after thermal treatment at 135 �C for 2 h, a 14-fold increase inaerogel density. By inspection, pDCPD aerogels did not suffer suchan extreme shrinkage, but the aerogel dimensions became toodistorted after heating to reliably calculate a final density in thesesamples.

The increase in density upon heating is a direct result of thedecrease in aerogel Tg upon hydrogenation (Fig. 8). Weighing thesamples before and after thermolysis shows no change inmass, andTGA measurements (Fig. 8) show that both pDCPD and H-pDCPD

e hydrogenated aerogel. TGA measurements show both 25-pDCPD and 25-H-pDCPD do

Fig. 9. Evolution of the densities of 25-pDCPD and 25-H-pDCPD during stepwisethermolysis to a final temperature of 135 �C. 25-H-pDCPD shows an abrupt shrinkagewhen the thermolytic temperature goes above the Tg.

J.M. Lenhardt et al. / Polymer 54 (2013) 542e547 547

aerogels are stable to thermal weight loss to >400 �C. DSCmeasurements, on the other hand, show that the Tg of H-pDCPDaerogels decreases from 168 � 4 �C (pDCPD) to 111 � 3 �C.

To probe this effect, 25-pDCPD and 25-H-pDCPD aerogels weresubjected to stepwise thermolysis with density determination aftereach thermal treatment (Fig. 9). Heating these samples to 80 �C for16 h led to slight densification to 40 mg mL�1 and 52 mg mL�1

respectively and additional heating to 100 �C for 8 h was followedby additional shrinkage to 46 mg mL�1 and 89 mg mL�1. Complete‘shrinkage’ of the 25-H-pDCPD aerogel was realized after an addi-tional heating to 135 �C for 4.5 h. Here, the final density of 25-pDCPD aerogel was measured as 45 mg mL�1 while the density of25-H-pDCPD increased dramatically to 765 mg mL�1. Again, thepDCPD aerogel discolored and its FTIR spectrum showed strongabsorbances at 3400 cm�1 and 1700 cm�1 while the H-pDCPDaerogel remained white in color and its FTIR spectrum remainedvirtually unchanged.

4. Conclusion

The hydrogenation of pDCPD gels to form H-pDCPD aerogelstherefore significantly increases the oxidative stability of thissystem while suffering from a decrease in the materials’ operatingtemperature. Characterization of these new aerogels’ propertiesshows that while the fibrillar morphology remains constant, somedeviation is observed in the measured pore volume, pore size andaerogel surface area that are consistent with the increase ofshrinkage of H-pDCPD during aerogel preparation. The inherently

low cross-link density of pDCPD aerogelsmay be responsible for theobservance of Tgs in these materials at 168 �C (pDCPD) and 111 �C(H-pDCPD) and the ability of these materials to shrink at temper-atures above Tg. The ease of preparation of these H-pDCPD aerogelsand their increased oxidative stability creates the need forsynthesizing pDCPD gels of increasing cross-link density prior tohydrogenation. We believe these materials may be less susceptibleto shrinkage during supercritical drying and may increase theoperating temperature for the hydrogenated system.

Acknowledgment

This work was performed under the auspices of the U.S.Department of Energy by Lawrence Livermore National Laboratoryunder Contract DE-AC52-07NA27344. IM Release # LLNL-JRNL-564354.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.polymer.2012.12.002.

References

[1] Pekala RW. J Mater Sci 1989;24(9):3221e7.[2] Gesser HD, Goswami PC. Chem Rev 1989;89(4):765e88.[3] Rolison DR, Long JW, Lytle JC, Fischer AE, Rhodes CP, McEvoy TM, et al. Chem

Soc Rev 2009;38(1):226e52.[4] Pierre AC, Pajonk GM. Chem Rev 2002;102(11):4243e66.[5] Hüsing N, Schubert U. Angew Chem Int Ed 1998;37(1e2):22e45.[6] Yoldas BE, Annen MJ, Bostaph J. Chem Mater 2000;12(8):2475e84.[7] Schmidt M, Schwertfeger FJ. Non-Cryst Solids 1998;225(0):364e8.[8] Saliger R, Fischer U, Herta C, Fricke JJ. Non-Cryst Solids 1998;225(0):81e5.[9] Kabbour H, Baumann TF, Satcher JH, Saulnier A, Ahn CC. Chem Mater 2006;

18(26):6085e7.[10] Kistler SS. Nature 1931;127:741.[11] Gash AE, Tillotson TM, Satcher Jr JH, Hrubesh LW, Simpson RL. J Non-Cryst

Solids 2001;285(1e3):22e8.[12] Tillotson TM, Sunderland WE, Thomas IM, Hrubesh LW. J Sol-Gel Sci Technol

1994;1(3):241e9.[13] Jin H, Nishiyama Y, Wada M, Kuga S. Colloids Surf A Physiochem Eng Aspects

2004;240(1e3):63e7.[14] Rigacci A, Marechal JC, Repoux M, Moreno M, Achard PJ. Non-Cryst Solids

2004;350(0):372e8.[15] Lee J, Gould G. J Sol-Gel Sci Technol 2007;44(1):29e40.[16] Dawedeit C, Kim SH, Braun T, Worsley MA, Letts SA, Wu KJ, et al. Soft Matter

2012;8(13):3518e21.[17] Kim SH, Worsley M, Valdez C, Shin SJ, Dawedeit C, Braun T, et al. RSC Adv

2012;2(23):8672e80.[18] Perring M, Long TR, Bowden NB. J Mater Sci 2010;20(39):8679e85.[19] Yao Z, Zhou L-w, Dai B-b, Cao KJ. Appl Polym Sci 2012;125(4):2489e93.[20] Hayano S, Takeyama Y, Tsunogae Y, Igarashi I. Macromolecules 2006;39(14):

4663e70.[21] Yang Y-S, Lafontaine E, Mortaigne BJ. Appl Polym Sci 1996;60(13):2419e35.[22] Hahn SF. J Polym Sci A Polym Chem 1992;30(3):397e408.[23] Abadie MJ, Dimonie M, Couve C, Dragutan V. Eur Polym J 2000;36(6):1213e9.