Crosslinked poly(ethylene oxide) containing siloxanes fabricated through thiol-ene photochemistry

Crosslinked Poly(ethylene oxide) Containing Siloxanes Fabricated

Through Thiol-Ene Photochemistry

Victor A. Kusuma,1 Elliot A. Roth,1 William P. Clafshenkel,2 Steven S. Klara,3 Xu Zhou,1

Surendar R. Venna,1 Erik Albenze,1 David R. Luebke,1 Meagan S. Mauter,3

Richard R. Koepsel,2 Alan J. Russell,2 David Hopkinson,1 Hunaid B. Nulwala1,4

1U.S. Department of Energy, National Energy Technology Laboratory, 626 Cochrans Mill Rd., Pittsburgh, Pennsylvania 152362Institute for Complex Engineered Systems, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 152133Department of Chemical Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, Pennsylvania 152134Department of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213

Correspondence to: H. Nulwala (E-mail: [email protected])

Received 23 December 2014; accepted 21 January 2015; published online 00 Month 2015

DOI: 10.1002/pola.27594

ABSTRACT: Homogenous amphiphilic crosslinked polymer films

comprising of poly(ethylene oxide) and polysiloxane were syn-

thesized utilizing thiol-ene “click” photochemistry. A system-

atic variation in polymer composition was Carried out to

obtain high quality films with varied amount of siloxane and

poly(ethylene oxide). These films showed improved gas sepa-

ration performance with high gas permeabilities with good

CO2/N2 selectivity. Furthermore, the resulting films were also

tested for its biocompatibility, as a carrier media which allow

human adult mesenchymal stem cells to retain their capacity

for osteoblastic differentiation after transplantation. The

obtained crosslinked films were characterized using differential

scanning calorimetry, dynamic mechanical analysis, thermog-

ravimetric analysis, FTIR, Raman-IR, and small angle X-ray

scattering. The synthesis ease and commercial availability of

the starting materials suggests that these new crosslinked

polymer networks could find applications in wide range of

applications. VC 2015 Wiley Periodicals, Inc. J. Polym. Sci., Part

A: Polym. Chem. 2015, 00, 000–000

KEYWORDS: biocompatible; biocompatibility; crosslinking; dif-

ferential scanning calorimetry (DSC); films; gas permeation;

gas separation membrane; membranes; photochemistry; poly-

ethylene glycol; poly(ethylene oxide); polysiloxanes; siloxane;

thiol-ene click

INTRODUCTION Development of new polymer syntheticmethods that are robust, facile, and proceed with high fidel-ity is desired in the area of functional polymeric materials.1–8

A variety of synthesis routes have been developed and ofparticular interest are the chemistries which envelope theconcepts of “click” chemistry and many new materials havebeen developed.9–17 Among these, thiol-based synthesisstrategies have been very successful due to their insensitivitytowards water and oxygen, quantitative yields, and mild con-ditions.18–23 Thiol-ene reaction, due to its ease, offers a sim-ple pathway to access functional materials and propertieswhich were not generally accessible.

Poly(ethylene oxide) (PEO) is an interesting polymer and hasfound use in a diverse range of applications: from drug deliv-ery,24 biocompatible devices,25 industrial gas separation,26

solid polymer electrolyte,27 to gas chromatography.28 Most ofthe desirable properties stem from the fact that it is polarand hydrophilic yet compatible with a wide range of sol-vents. This allows further modifications with other functional

groups to confer additional properties, such as amphiphilic-ity, allowing it to be used in other applications, like as a sur-factant in industrial processes or nanotechnology.29 Siloxane-based functional groups are often added to PEO to conferamphiphilicity, which allows for a greater possible range ofapplications.30,31 However, adding polysiloxanes alone isoften difficult precisely because they are inherently hydro-phobic, and thus a robust chemistry is still required.

Thiol functionalized molecules can be effective hydrogendonors in free-radical polymerization reactions, and as suchhave attracted much interest for use in various polymeriza-tion reactions.3,4 The thiol-ene click chemistry describedherein offers a way of integrating a polysiloxane into a cross-linked PEO hydrogel in a homogeneous manner, as has beenobserved in other inherently incompatible systems.3 Severalthiol-ene crosslinked systems consisting of functionalizedoligomers of PEO and/or siloxanes have recently beenreported.16,32–35 In this study, a commercially available thiol-functionalized polysiloxane was crosslinked with

VC 2015 Wiley Periodicals, Inc.

WWW.MATERIALSVIEWS.COM JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000 1

JOURNAL OFPOLYMER SCIENCE WWW.POLYMERCHEMISTRY.ORG ARTICLE

poly(ethylene glycol) diacrylate (PEGDA), a common PEO-based acrylate crosslinker which has been studied exten-sively for many applications.36–39 A proposed reaction mech-anism for thiol-acrylate polymerization is illustrated inScheme 1. Unlike other thiol-ene reactions, the acrylate-acrylate homopolymerization reaction rate is comparablewith the thiol-acrylate radical addition and chain transferrates.4 Therefore, the combination of thiol and acrylatemonomers allows the relative monomer composition to betuned while still obtaining a highly crosslinked network,making a wider range of polymer properties accessible.

Two possible applications for the polymers explored in thisstudy are as cell growth media and as gas separation mem-branes. Human adult mesenchymal stem cells (hAMSCs) via-bility and differentiation studies were performed to evaluatetheir potential biocompatibility. For gas separation, we eval-uated their performance as CO2-selective membranes, whichcan be used, for example, for CO2 separation from industrialflue gas, by measuring their pure gas CO2, N2, and CH4 per-meabilities at 40 �C. While crosslinked PEO has very goodCO2/N2 selectivity and moderate CO2 permeability, the addi-tion of siloxane functional group can potentially increase the

CO2 permeability further. In addition, infrared and Ramanspectroscopy were performed to confirm the reaction, whiledynamic mechanical analysis (DMA), differential scanningcalorimetry (DSC) and small angle X-ray scattering (SAXS)characterized the structure and thermomechanical propertiesof the copolymers.

EXPERIMENTAL

MaterialsAll chemicals used in this work were commercially obtainedat highest available purity and were used as received. PEGDA(Mn5 700), 2,2-dimethoxy-2-phenylacetophenone (DMPA),pentaerythritol tetrakis(3-mercaptopropionate) (tetrathiol),acetone and toluene were obtained from Sigma-Aldrich.(Mercaptopropyl) methyl siloxane homopolymer, 75–150 cSt(thiosiloxane) was obtained from Gelest, Inc. (Morrisville,PA) (Gelest cat number SMS-992). This particular thiol func-tionalized siloxane oligomer has MW 5 4000–7000. Thechemical structures are given in Scheme 2. All gases used inthis work were ultra high purity or research grade (CO2 andN2: 99.999%; CH4: 99.97%) and were obtained from ButlerGas Company (McKees Rocks, PA).

Film FormationMonomer solutions were made by directly mixing PEGDA andDMPA with the thiosiloxane at desired compositions. The typi-cal batch size was 1 g (total monomer basis). DMPA wasadded at 1 wt % of total monomers. Magnetic mixing withaid of a stirring magnetic bar was utilized for at least 30 minuntil the DMPA was completely dissolved. For toluene solu-tions, 60 wt % toluene (i.e. 1.5 g toluene per 1 g total mono-mer) was added after all other components were added. Thecomposition of PEGDA to thiosiloxane was based on functionalgroup ratio, i.e. 20% SH corresponds to 1 thiol per 4 acrylategroup. On weight basis, the compositions are given in Table 1.

The monomer solution will polymerize slowly even without UVinitiation. However, the reaction rate appeared to be lower forthe thiosiloxane than the tetrathiol, which is a multifunctionalthiol more commonly used for copolymerization in thiol-enepolymer studies.4,36 To illustrate, the thiosiloxane monomersolutions may be kept under refrigeration for �2 months with-out significant gelation (i.e. no visible solid or semisolid portionwhen tipping the vial), whereas the tetrathiol solutions (partic-ularly 33% SH or higher) may gel overnight in the refrigerator.For this study, the solutions were cast typically within 2 h frominitial mixing to minimize undesired prepolymerization.

Free-standing films were cast in a similar manner asdescribed in previous publications.40,41 The monomer

SCHEME 1 Free radical polymerization mechanism of poly(eth-

ylene glycol) diacrylate and a multifunctional thiol (modeled

after Hoyle and Bowman4).

SCHEME 2 Chemical structures of a) PEGDA, b) thiosiloxane,

and c) tetrathiol.

TABLE 1 Monomer Solution Composition (Solvent-Free Basis)

SH% Content PEGDA (g) Thiosiloxane (g) DMPA (mg)

20 0.912 0.088 10

33 0.839 0.161 10

50 0.723 0.277 10

ARTICLE WWW.POLYMERCHEMISTRY.ORGJOURNAL OF

POLYMER SCIENCE

2 JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000

solution was pipetted into a gap between two quartz platesseparated by a pair of steel shims (nominal thicknesses 102mm or 203 mm). The UV source was a Blak-Ray B-100APhigh intensity lamp (from UVP, LLC—Upland, CA), with nomi-nal intensity of 8.9 mW/cm2 at 25 cm. The quartz plateassembly was placed inside the exposure box provided withthe lamp for 90 s. This exposure time was enough to ensurecomplete acrylate conversion, as confirmed by the FTIR (dis-cussion follows). While we did not have the means to moni-tor reaction kinetics at this time scale, Kwisnek et al.reported that complete conversion of PEGDA/tetrathiolunder oxygen deficient conditions can be achieved within 5 sunder 20 mW/cm2 intensity.36 After 90 s, the films wereremoved by separating the quartz plates carefully. When tol-uene was used as solvent, it was removed by evaporationunder ambient conditions in a laboratory fume hood for �15min. All films were post-treated in acetone to remove anyremaining solvent or sol (i.e. low molecular weight oligomersor unreacted monomers); however, mass loss post treatmentwas insignificant, suggesting essentially negligible extractablecomponent.

SpectroscopyThe instrument used for FTIR was a Nicolet IR 100 withATR module from Thermo Scientific. Liquid monomer orsolid polymers were placed directly on the ATR crystal.Sixty-four scans with resolution of 2 cm21 were used toobtain the final spectra.

Raman spectroscopy was performed on a Horiba ScientificLabRAM HR spectrometer having a Spectra Physics 532 nmNd:YAG laser with an operating power of 0.2 W. Ten scanswith <5 s exposure time were taken to obtain the finalspectra.

Small Angle X-Ray ScatteringAll SAXS measurements were performed on a Rigaku SAXSinstrument, model MSA BS0187 at a wavelength of0.15405 nm. SAXS measurements were recorded with anelectronic detector at a distance of 1786.27 mm from thesource. Thin films of each sample were irradiated for 30 minunder vacuum conditions. All analysis of scattering patternswas completed with SAXS GUI software.

Thermal AnalysisThermogravimetric analysis (TGA) was performed on a TAInstruments Q500 TGA at a heating rate of 15 �C/minbetween 125 �C and 1500 �C under nitrogen. DSC was per-formed on a TA Instruments Q2000 DSC under nitrogen.With sample size between 5 to 10 mg placed inside TzeroTM

aluminum pans, the samples were repeatedly heated to1150 �C at 15 �C/min and cooled to 2150 �C at 210 �C/min. Glass transition temperature was defined as the inflec-tion point of the transition on the third heating cycle. Melt-ing point was defined at the peak on the third heating cycle.

DMA was performed in a TA Instruments Q800 DMA with aliquid nitrogen cooling accessory. The tested film was cutinto �5.3 mm wide by �20 mm long rectangular strips and

mounted on the instrument’s tension clamp accessory. Beforethe experiment, the films were heated to 1105 �C and thencooled at 210 �C/min to 290 �C. Temperature sweepexperiment was performed under dynamic oscillatory fre-quency of 1 Hz and nitrogen purge at a heating rate of 13�C/min from 290 �C to 150 �C. The primary transition tem-perature (Ta) of a polymer was defined as the peak maxi-mum in the loss tangent (tan d) curve.

hAMSC Viability and Differentiation StudyFor viability and differentiation studies, multipotent hAMSCs(Lonza, Walkersville, MD) were cultured on polymer films atan initial seeding density of 2.5 3 104 cells/mL or 5.0 3

104 cells/mL, respectively, in 96-well plates. Polymer filmswere cut to fit the bottom of each well using a 5-mm biopsypunch, and were inserted before seeding. All cells were incu-bated under 90% humidity at 37 �C and 5% CO2. For theassessment of viability, cells were cultured in completegrowth medium (Lonza, Walkersville, MD) for 1, 3, and 7days. Medium was changed every second day. Viability wasassessed using the Cell Titer GloVR Luminescent Cell Viabilityassay (Promega, Madison, WI). Briefly, 100 mL of Cell TiterGloVR reagent was added to 100 mL of cell culture mediumcontaining cells. The contents were mixed for 2 min on anorbital shaker to facilitate lysis of the cells, and then left atroom temperature for 10 min to allow the luminescent sig-nal to stabilize. Luminescence was recorded using a SynergyH1 Hybrid Multi-Mode microplate reader (BioTek, Winooski,VT) with a 1 s integration time.

For differentiation experiments, cells were left to adhere over-night in complete growth medium and then were cultured for 7days in osteogenic medium (i.e., fresh growth medium supple-mented with 0.1 mM dexamethasone, 10 mM b-glycerophosphate, and 0.3 mM ascorbic acid; Sigma-Aldrich).Differentiation of the hAMSCs was qualitatively assessed bystaining for alkaline phosphatase (ALP) activity, a knownmarker of osteoblast differentiation (Sigma-Aldrich). Upon reac-tion with the kit substrate and diazonium salt, the phosphataseactivity of hAMSCs produces an insoluble, visible pigment.Images of hAMSCs were captured using an inverted Leica DMIRB microscope (Leica Microsystems GmbH, Wetzlar, Germany).

Gas Permeation MeasurementPure gas permeability was measured using the constant vol-ume (isochoric) method commonly used by many researchgroups.42,43 Briefly, the method measures the gas fluxthrough the membrane into a vacuum reservoir of knownvolume by monitoring the rate of pressure increase in thereservoir. The instrument was constructed using off-the-shelfpositive pressure fittings (from Swagelok, supplied by Pitts-burgh Valve & Fitting Co.) and vacuum fittings (from Kurt J.Lesker Co., Jefferson Hills, PA). The cell uses a 2–3/4” CFmodule design similar to the one reported by Perez et al.,42

except that we used a VitonVR O-ring to seal the membraneto the cell. Temperature control was provided by a customTenney T10 environmental chamber (Thermal Product Solu-tions—White Deer, PA). A newly loaded membrane was



evacuated overnight to remove all absorbed gases beforebeginning an experiment. Typical upstream pressures duringan experiment were 0.10 MPa, 0.15 MPa, and 0.30 MPa;downstream pressure was maintained at low vacuum (below1.3 kPa). A constant temperature of 406 1 �C was main-tained by the environmental chamber. Gas permeability val-ues reported in this study represent the mean from at leastthree measurements per data point.

RESULTS AND DISCUSSION

Reaction ConversionPhotopolymerization is a well-established technique that canrapidly transform solvent-free liquid monomer into solid poly-mer films at ambient temperature.44 With acrylate monomers,a near-complete conversion can be achieved in matter of sec-onds—too fast for the kinetics to be elucidated without speci-alized equipment. Therefore, our present analysis focusedexclusively on thiol and acrylate conversion on the final prod-uct, in a similar manner as previous studies.37,41,45

FTIR analysis of the monomer mixtures showed the charac-teristic peaks associated with the CH2@CHA present in the

acrylate functional group of PEGDA: at 812 cm21, 980 cm21,1190 cm21 and 1410 cm21.37 After 90 s UV exposure, allthese peaks completely disappeared [Fig. 1(a)], indicatingcomplete acrylate conversion was likely achieved in thisreaction. This also shows that the reaction was not limitedby monomer mass transfer, regardless of whether the reac-tion was performed in toluene or solvent-free conditions.Further, FTIR analysis also confirmed that the solventremoval method performed on the solvent-cast films waseffective, as evidenced by complete absence of peaks associ-ated with the toluene in the final spectra.

Unlike the acrylates, the thiol and thioether functionalgroups associated with the thiosiloxane or tetrathiol did notappear in the FTIR spectra. Instead, Raman spectroscopywas utilized to evaluate the extent of thiol (i.e. R-S-H) con-version to thioether (i.e. R-S-R) groups as a result of thiol-acrylate reaction.32 Due to the concurrent thiol-acrylate andacrylate-acrylate reactions (Scheme 1), Kwisnek et al. sug-gested that a ratio of 1 thiol to 4 acrylate (i.e. 20% SH) wasoptimal for a PEGDA1 tetrathiol monomer mixture to reactwhile leaving little excess (unreacted) thiol in the final prod-uct.36 The Raman spectra of the thiosiloxane copolymer films

FIGURE 1 a) FTIR and b) Raman spectra of various PEGDA/thiosiloxane copolymers polymerized with toluene co-solvent. Peaks of

interest, as described in text, are marked. Baselines are vertically shifted for clarity.


POLYMER SCIENCE


[Fig. 1(b)] also suggested the same optimal ratio. The S-Hstretching vibration (around 2575 cm21) characteristic ofthiol group was readily observed in the 50% SH film. Thispeak was weaker in the 33% SH film and was virtuallyundetected in the 20% SH film and the PEGDA film. In con-trast, the CAS stretching vibration peaks (around 650 cm21)characteristic of both thiol and thioether (the reaction prod-uct) increased in intensity with %SH content. These signifiedthat around 20% SH, almost all the original thiols were con-

verted into thioethers; further increases in initial thiol con-centration simply introduced excess thiols into the polymer.Another significant emerging peak around 510 cm21 wasattributed to the Si-O stretching vibration due to increasingsiloxane content.

Polymer Network StructureOne of the main difficulties in combining PEO and polysilox-anes is overcoming the miscibility gap between the two com-ponents. In applications such as contact lenses, where phaseseparation can result in opacity and is thus generally unde-sirable, careful phase engineering through block copolymer-ization of smaller individual segments has been used.46 Inother applications such as gas separation membranes, whereboth phases have specific advantages in terms of permeabil-ity and selectivity, the desired outcome is property averaging,which required minimal phase separation.31

In contrast to PEGDA and tetrathiol, which are fully miscible,PEGDA and thiosiloxane are only partially miscible at thecompositions studied here. A translucent solution wasobtained when the two monomers were mixed without a sol-vent at room temperature (23 �C). As PEGDA was completelyimmiscible with silicone oil (i.e. polydimethylsiloxane), thethiosiloxane clearly had a better interaction with the polarPEGDA, most likely as a result of the more polar thiol sidegroup. Polymerizing this solvent-free monomer mixture gavetranslucent films with good mechanical properties [Fig. 2(a)].However, completely transparent solutions were obtained bymixing in a common solvent, a strategy that was also usedsuccessfully in a previous study to improve compatibility.40

Aside from toluene, other effective co-solvents identifiedincluded chloroform and N,N-dimethyl formamide (DMF).Once the polymerization reaction was complete, the filmremained optically transparent even after the solvent wasremoved [Fig. 2(b)]. The transparency, combined with resultsfrom the SAXS studies (in a later section), confirmed that thefilms were fairly homogenous.

Crosslinked PEO made from solvent-free polymerization ofPEGDA is a rubbery, completely amorphous material.37 As

FIGURE 2 Physical appearances of 20% SH PEGDA/thiosilox-

ane films cast a) solvent-free, and b) with 60 wt % toluene co-

solvent. [Color figure can be viewed in the online issue, which

is available at wileyonlinelibrary.com.]

FIGURE 3 Second scan DSC heating curves of crosslinked

PEGDA/thiosiloxane films, vertically shifted for clarity.

TABLE 2 Transition and Degradation Temperatures of PEGDA/

Thiosiloxane Copolymers

SH%

Content Solvent

Tg

(�C)aTm

(�C)aTa

(�C)bT95

(�C)cT90

(�C)c

0 Neat 236 – 232 308 342

0 Toluene 240 27 234 349 364

20 Neat 239 – 234 N/A N/A

20 Toluene 243 – 238 333 345

33 Toluene 246 – 240 344 355

50 Toluene 250 – 245 335 347

a DSC results.b DMA results.c Temperature at which 95% and 90% of polymer mass at 120 �C

remained in the TGA.



http://wileyonlinelibrary.com

shown in Figure 3 and Table 2, the glass transition tempera-ture (Tg) of this material as measured by the DSC was 236�C. Adding toluene into the prepolymer PEGDA resulted in apolymeric product with decreased crosslink density andslightly decreased glass transition temperature, as had beenobserved previously40 and confirmed by the DMA results inFigure 4. The toluene content in the solvent-cast PEGDA filmwas higher (i.e. 60 wt % toluene) than in the previous study,and we observed by DSC that the film exhibited not only aglass transition but also characteristic cold crystallization fol-lowed by melting as the film was heated further. This sug-gested that at this low level of crosslinking, the PEO chainswere more likely to form linear chains, which would startpacking into crystalline structures.37

The addition of thiosiloxane into the solvent-cast networkresulted in a steady decrease of Tg with increasing siloxanecontent, as determined by DSC (Fig. 3). This decrease wascaused by an increasing overall molecular mobility of thecopolymer networks with the introduction of the highly flexi-ble siloxane chains. In addition, there was no emerging sec-ond glass transition temperature between 293 �C (the Tg ofthe thiosiloxane) and 236 �C (Tg of crosslinked PEGDA).These phenomena were also observed in the DMA (Fig. 4),further supporting our view that the networks remainedhomogeneous even with addition of thiosiloxane moieties. Infact, the tan d plot showed a narrowing transition withincreasing thiosiloxane content, indicative of increasing net-work homogeneity. We attribute this to the nature of thiol-ene polymerization reaction, which has been observedimproving network homogeneity in other acrylate systemsby delaying the onset of gel formation, thus increasing mono-mer diffusivity during polymerization.3

Without the addition of toluene as a compatibilizer, weexpected the solvent-free 20 SH% PEGDA/thiosiloxane filmto exhibit some phase separation. However, we did notobserve a second glass transition in the DSC, which would

indicate significant phase separation. A slight reduction ofthe primary Tg to 239 �C was observed, which was slightlyhigher than the solvent-cast version (i.e. 243 �C). The stor-age modulus in the DMA temperature sweep also showed asingle glass to rubber relaxation.

The rubbery plateau moduli for the PEGDA/thiosiloxane,which were indicative of the crosslink density,47 were rela-tively independent of the siloxane content within the compo-sition range in this study. The primary determinant of therubbery plateau moduli appeared to be the amount of co-solvent in the monomer mixture. The compatibilizer in this

FIGURE 4 a) Storage moduli and b) tan d curves of crosslinked PEGDA/thiosiloxane films as obtained by the DMA.

FIGURE 5 Small angle X-ray scattering results for selected

PEGDA/thiosiloxane crosslinked polymer films. Baselines are

vertically shifted for clarity.


POLYMER SCIENCE


case acted as a diluent, which promoted lower crosslink den-sity: a similar phenomenon was also observed previ-ously.40,47 Physically, the dilution resulted in slightly weakerpolymer films compared to the solvent-free analogues. Thus,it was important that the amount of diluent for improvinghomogeneity should be kept to a minimum. Upon furtherinvestigation, approximately 50 wt % toluene appeared to bethe minimum amount that yielded optically clear films.

The thermal stability of the crosslinked PEGDA/thiosiloxanefilms was excellent. TGA curves showed virtually no polymermass loss below 280 �C. The polymers exhibited a singledecomposition event with onset temperatures around 310�C.

A further evidence of the phase homogeneity in these poly-mers was found in the SAXS results (Fig. 5). There were nosignificant peaks observed within the measurement range,signifying no features greater than 3 nm in dimension wasdetected for all the films tested. This was true even for thefilms cast solvent-free.

Finally, we found evidence that crosslinked polymers con-taining thiol-functionalized monomers stored under ambientconditions will slowly oxidize over time. This behavior wasparticularly prevalent in PEGDA/tetrathiol films, with filmssoftening or even turning into a viscous liquid within a year.The degradation kinetics appeared to be slower for thePEGDA/thiosiloxane films observed over a similar timeframe. This long-term behavior must be taken into accountwhen considering these materials for a particular application.In this study, we have focused only on short-term behavior,with all measurement performed within a month of filmcasting.

BiocompatibilityRemodeling processes in the adult human that regenerateand repair damaged tissue require a constant supply of new

cells that are capable of proliferating and differentiating intomany lineages. The multipotent nature of mesenchymal stemcells (hAMSCs) allows them to differentiate into several line-ages such as fibroblast, osteoblasts, chondrocytes, tenocytes,myoblasts, and adipocytes.48 The chemical cues that drivehAMSCs toward an osteoblast phenotype can be replicated invitro by supplementing complete growth medium with dexa-methasone, b-glycerophosphate, and ascorbic acid.49 Mesen-chymal stem cells are widely distributed in the body andhave well-documented reparative effects in preclinical andclinical models.50 The combination of multipotent stem cellswith biomaterial scaffolds is a promising strategy for thelocal repair and regeneration of damaged tissues. However,this requires a biocompatible interaction between the cellsand biomaterial to ensure that key processes integral torepair, such as recruitment and terminal differentiation, areunaffected. Thus, the composition of the biomaterial plays asignificant role in directing the fate of endogenous or seededprogenitor cells. Because the PEGDA/thiosiloxane films con-sisted of ethylene oxide and siloxane repeat units, whichcould be found in many biocompatible materials, we wereinterested to assess their biocompatibility using hAMSCs asmodel stem cells.

Viability of hAMSCs was dependent on polymer film compo-sition (Fig. 6). In this study, we analyzed the three toluenecast PEGDA/thiosiloxane films in addition to the solvent-freecrosslinked PEGDA and the PEGDA/tetrathiol series. Duringinitial screening, we found that the hAMSCs were not viableon the crosslinked PEGDA and all the PEGDA/tetrathiol films.The 20% SH PEGDA/thiosiloxane film demonstrated similarviability to hAMSCs grown on TC-treated polystyrene dish(control) on day 1. Although viability declined for subse-quent days, it appeared to stabilize between days 3 and 7.The 33% SH film demonstrated the greatest viability andwas consistently greater than or equal to control hAMSCs

FIGURE 6 Viability of cultured hAMSCs on PEGDA/thiosiloxane

polymer film surfaces over a 7-day period. Results are normal-

ized as a percentage of viability achieved in hAMSCs cultured

on TC-treated polystyrene dish.

FIGURE 7 Osteoblastic differentiation of hAMSCs cultured on

PEGDA/thiosiloxane polymer surfaces for 7 days under osteo-

genic conditions. ALP-positive cells are stained reddish purple.

[Color figure can be viewed in the online issue, which is avail-

able at wileyonlinelibrary.com.]



http://wileyonlinelibrary.com

over the 7 day period. In contrast, the 50% SH film demon-strates very low viability, which was likely correlated with alack of robust attachment of hAMSCs to the polymer surface.Thus, it appeared that there was an optimum amount of thi-osiloxane that ensured maximum hAMSC viability.

The proliferation results correlated well with the differentia-tion potential of hAMSCs cultured on the polymer film sur-face (Fig. 7).

Interestingly, however, the capacity of hAMSCs to differenti-ate on the polymer films tested did not seem to depend onfilm composition. Nonetheless, those films with the bestattachment and viability (i.e. the 20% SH and 33% SHPEGDA/thiosiloxane films) consequently demonstrated ALPactivity comparable to hAMSCs induced to differentiate onTC-treated polystyrene dish. Despite positive staining forALP activity in hAMSCs grown on the 50% SH film, low via-bility of the cell line on this film likely makes it the leastsuitable as a carrier of hAMSC for transplantation. Overall,results from these preliminary assessments indicated thatmultipotent hAMSCs were viable and retained their capacityfor osteoblastic differentiation on certain types of polymerfilms, such as the 33% SH film. They would likely be thebest candidates for use as carriers for stem cell transplanta-tion at injury sites. Chemical signals from the surroundingenvironment would then support the differentiation of thegrafted stem cells into mature cell types, facilitating repara-tive processes.

Gas Transport PropertiesAdding siloxanes is a common strategy for increasing gaspermeability in polymers for gas separation and for soft con-tact lens applications.31,40,46,51 Owing to the highly flexible,non-polar nature of the siloxane chains, polymers containingsiloxanes (such as poly(dimethylsiloxane), PDMS) haveamong the highest gas permeability known.52 However, thegas selectivity (i.e. the ratio of permeation rate) between dif-ferent gases tends to be less pronounced than glassy poly-mers because PDMS is less size selective.53 The gasselectivity instead was dictated more by the difference in gassolubility. For applications that require CO2 separation fromlight gases such as N2 or CH4, PDMS is not a good candidatebecause the CO2/N2 and CO2/CH4 selectivity are unaccept-ably low.53 In contrast, while crosslinked PEO networks tendto have more modest gas permeabilities, they also have verygood CO2/N2 selectivity due to the favorable interactionbetween the quadrupolar CO2 molecules with the polar eth-ylene oxide moieties.37 To illustrate this point, at 35 �CPDMS has CO2 permeability of �3800 barrer (1barrer5 10210 cm3(STP).cm/(cm2 s cmHg)5 7.5 3 10218

m3(STP).m/(m2 s Pa)) with CO2/N2 selectivity of 9.5;53 incontrast, crosslinked PEGDA has CO2 permeability of �110barrer with CO2/N2 selectivity of 50.37 Combining PEO andsiloxane moieties in a homogeneous crosslinked network,therefore, may yield a copolymer with intermediate proper-ties: specifically, films with higher permeability than cross-linked PEGDA which retains much of the CO2/N2 selectivity.

The gas permeability results of crosslinked PEGDA/thiosilox-ane copolymers are shown in Figure 8(a), showing increasingpermeability with siloxane content. The gas transport proper-ties of PEGDA/tetrathiol has been explored in a previousstudy;36 therefore, they were also studied for comparison. Asobserved previously, slight increase in gas permeability of thecrosslinked PEGDA could be attributed to the reduction incrosslink density due to toluene dilution.40 Dilution with 60wt % toluene increased the CO2 permeability of crosslinked

FIGURE 8 a) Pure gas permeability and b) ideal gas selectivity

with different multifunctional thiol co-polymer content, taken at

40 0�C and upstream pressure of 0.15 MPa. When error bars fall

within the markers, they are omitted for clarity.


POLYMER SCIENCE


PEGDA from 90 barrer to 150 barrer. Adding 20% SH thiosi-loxane yielded polymer with CO2 permeability of 160 barrer,and further increase to 260 barrer was obtained with additionof 50% SH thiosiloxane. Only slight increase was obtained forthe PEGDA/tetrathiol copolymers compared to the solvent-free crosslinked PEGDA, for example, 110 barrer for the 50%SH tetrathiol, confirming the result reported in a previousstudy.36 Despite the polar nature of the thioether group, Kwis-nek et al. did not observe that they significantly influence thegas transport properties of the copolymers.54 Instead, the gaspermeability increase observed presently was attributed tothe siloxane chains, which are very flexible and permits highergas diffusivity.

The addition of siloxane chains, which are less polar thanthe PEO moieties, not surprisingly reduced the CO2/N2 andCO2/CH4 selectivity, as shown in Figure 8(b). However,because the amount of thiosiloxane was relatively small evenin the 50% SH polymer (cf. Table 1), the selectivity reductionwas relatively modest. For example, the 50% SH polymerhas CO2/N2 selectivity of 43, which is a slight reductionfrom crosslinked PEGDA (i.e. 55) but compensated by a 75%increase in CO2 permeability. The results demonstrated thatwe have achieved our objective in terms of gas transportproperties, which was property averaging between the PEOand siloxane moieties.

CONCLUSIONS

A crosslinked polymer network consisting of PEO and polysi-loxane moieties have been fabricated using PEGDA and a mer-captopropyl functionalized polysiloxane. The polymerizationutilized thiol-acrylate photochemistry, which resulted in a rela-tively homogeneous network with essentially complete conver-sion after a relatively short exposure time. These crosslinkedpolymers exhibit properties, such as hydrophobicity and gastransport, which are intermediate of their components and canbe easily tuned by changing their relative composition. In allcases a modest selectivity reduction was observed such as the50% SH polymer has CO2/N2 selectivity of 43, which is slightlyreduced from crosslinked PEGDA (i.e. 55). However, this reduc-tion in selectivity was easily compensated by a marked improve-ment CO2 permeability (75% increase in permeability overPEGDA). Preliminary assessment also revealed that these filmscould be tuned for biocompatibility, with some films showingpotential as stem cell carrier materials in transplantation. Weanticipate that the ease of use and the robustness of the chemis-try will make these materials useful in a variety of applications.In addition, the commercial availability of the precursors shouldalso encourage engineers and chemist to use these materials intheir research. Furthermore, the availability of excess thiol inthe obtained films can enable post-modification and tune theproperties for specific applications.

ACKNOWLEDGMENTS

This technical effort was performed in support of the U.S.Department of Energy’s National Energy Technology Labora-tory’s ongoing research on CO2 capture under the contract DE-

FE0004000. The authors thank Alex Hallenbeck and John Kitch-in’s research group for the Raman spectroscopy and AdefemiEgbebi for running the TGA experiments.

REFERENCES AND NOTES

1 C. J. Hawker, K. L. Wooley, Science 2005, 309, 1200–1205.

2 J. M. Spruell, M. Wolffs, F. A. Leibfarth, B. C. Stahl, J. Heo, L.

A. Connal, J. Hu, C. J. Hawker, J. Am. Chem. Soc. 2011, 133,

16698–16706.

3 C. E. Hoyle, A. B. Lowe, C. N. Bowman, Chem. Soc. Rev.

2010, 39, 1355–1387.

4 C. E. Hoyle, C. N. Bowman, Angew. Chem. Int. Ed. Engl.

2010, 49, 1540–1573.

5 C. J. Hawker, V. V Fokin, M. G. Finn, K. B. Sharpless, Aust. J.

Chem. 2007, 60, 381–387.

6 L. Ackermann, H. K. Potukuchi, Org. Biomol. Chem. 2010, 8,

4503–4513.

7 J.-F. Lutz, Angew. Chem. Int. Ed. Engl. 2007, 46, 1018–1025.

8 A. Saha, S. De, M. C. Stuparu, A. Khan, J. Am. Chem. Soc.

2012, 134, 17291–17297.

9 H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. Int.

Ed. Engl. 2001, 40, 2004–2021.

10 M. Lartey, M. Gillissen, B. J. Adzima, K. Takizawa, D. R.

Luebke, H. B. Nulwala, J. Polym. Sci. Part A: Polym. Chem.

2013, 51, 3359–3364.

11 K. Takizawa, H. Nulwala, R. J. Thibault, P. Lowenhielm, K.

Yoshinaga, K. L. Wooley, C. J. Hawker, J. Polym. Sci. Part A:

Polym. Chem. 2008, 46, 2897–2912.

12 C. Besset, S. Binauld, M. Ibert, P. Fuertes, J.-P. Pascault, E.

Fleury, J. Bernard, E. Drockenmuller, Macromolecules 2010, 43,

17–19.

13 N. Cengiz, J. Rao, A. Sanyal, A. Khan, Chem. Commun.

(Camb). 2013, 49, 11191–11193.

14 B. P. Mudraboyina, M. M. Obadia, I. Allaoua, R. Sood, A.

Serghei, E. Drockenmuller, Chem. Mater. 2014, 26, 1720–1726.

15 P. Dimitrov-Raytchev, S. Beghdadi, A. Serghei, E.

Drockenmuller, J. Polym. Sci., Part A: Polym. Chem. 2013, 51,

34–38.

16 T. Yang, M. Malkoch, A. Hult, J. Polym. Sci., Part A: Polym.

Chem. 2013, 51, 363–371.

17 T. Yang, M. Malkoch, A. Hult, J. Polym. Sci., Part A: Polym.

Chem. 2013, 51, 1378–1386.

18 A. B. Lowe, Polym. Chem. 2014, 5, 4820–4870.

19 K. L. Killops, L. M. Campos, C. J. Hawker, J. Am. Chem.

Soc. 2008, 130, 5062–5064.

20 L. M. Campos, T. T. Truong, D. E. Shim, M. D. Dimitriou, D.

Shir, I. Meinel, J. A. Gerbec, H. T. Hahn, J. A. Rogers, C. J.

Hawker, Chem. Mater. 2009, 21, 5319–5326.

21 S. Binder, I. Gadwal, A. Bielmann, A. Khan, J. Polym. Sci.,

Part A: Polym. Chem. 2014, 52, 2040–2046.

22 I. Gadwal, S. Binder, M. C. Stuparu, A. Khan, Macromole-

cules 2014, 47, 5070–5080.

23 S. De, A. Khan, Chem. Commun. (Camb). 2012, 48, 3130–

3132.

24 R. A. Scott, N. A. Peppas, Biomaterials 1999, 20, 1371–1380.

25 D. L. Elbert, J. A. Hubbell, Annu. Rev. Mater. Sci. 1996, 26,

365–394.

26 S. L. Liu, L. Shao, M. L. Chua, C. H. Lau, H. Wang, S. Quan,

Prog. Polym. Sci. 2013, 38, 1089–1120.



27 Y. Kang, K. Cheong, K. A. Noh, C. Lee, D. Y. Seun, J. Power

Sources 2003, 119–121, 432–437.

28 J. de Zeeuw, J. Luong, Trends Anal. Chem. 2002, 21, 594–607.

29 A. Pitto-Barry, N. P. E. Barry, Polym. Chem. 2014, 5, 3291–

3297.

30 D.-J. Guo, H.-M. Han, S.-J. Xiao, Z.-D. Dai, Colloids Surf. A:

Physicochem. Eng. Asp. 2007, 308, 129–135.

31 S. R. Reijerkerk, M. H. Knoef, K. Nijmeijer, M. Wessling, J.

Membr. Sci. 2010, 352, 126–135.

32 S. Mongkhontreerat, K. €Oberg, L. Erixon, P. L€owenhielm, A.

Hult, M. Malkoch, J. Mater. Chem. A 2013, 1, 13732–13737.

33 L. M. Campos, I. Meinel, R. G. Guino, M. Schierhorn, N. Gupta,

G. D. Stucky, C. J. Hawker, Adv. Mater. 2008, 20, 3728–3733.

34 N. Gupta, B. F. Lin, L. M. Campos, M. D. Dimitriou, S. T.

Hikita, N. D. Treat, M. V Tirrell, D. O. Clegg, E. J. Kramer, C. J.

Hawker, Nat. Chem. 2010, 2, 138–145.

35 J. N. Hunt, K. E. Feldman, N. a Lynd, J. Deek, L. M. Campos,

J. M. Spruell, B. M. Hernandez, E. J. Kramer, C. J. Hawker,

Adv. Mater. 2011, 23, 2327–2331.

36 L. Kwisnek, S. Heinz, J. S. Wiggins, S. Nazarenko, J. Membr.

Sci. 2011, 369, 429–436.

37 H. Lin, T. Kai, B. D. Freeman, S. Kalakkunnath, D. S. Kalika,

Macromolecules 2005, 38, 8381–8393.

38 H. Ju, B. D. McCloskey, A. C. Sagle, V. A. Kusuma, B. D.

Freeman, J. Membr. Sci. 2009, 330, 180–188.

39 M. B. Mellott, K. Searcy, M. V Pishko, Biomaterials 2001, 22,

929–941.

40 V. A. Kusuma, B. D. Freeman, S. L. Smith, A. L. Heilman, D.

S. Kalika, J. Membr. Sci. 2010, 359, 25–36.

41 V. A. Kusuma, B. D. Freeman, M. A. Borns, D. S. Kalika, J.

Membr. Sci. 2009, 327, 195–207.

42 E. V Perez, K. J. Balkus, J. P. Ferraris, I. H. Musselman, Rev.

Sci. Instrum. 2013, 84, 065107.

43 H. Lin, B. D. Freeman, In Springer-Handbook of Materials

Measurement Methods; H. Czichos, L. E. Smith, T. Saito, Eds.;

Springer: Berlin, 2006; pp 371–387.

44 C. Decker, Macromol. Rapid Commun. 2002, 23, 1067–1093.

45 V. A. Kusuma, S. Matteucci, B. D. Freeman, M. K. Danquah,

D. S. Kalika, J. Membr. Sci. 2009, 341, 84–95.

46 P. C. Nicolson, J. Vogt, Biomaterials 2001, 22, 3273–

3283.

47 S. Kalakkunnath, D. S. Kalika, H. Lin, B. D. Freeman, J.

Polym. Sci. Part B: Polym. Phys. 2006, 44, 2058–2070.

48 K. H. Kraus, C. Kirker-Head, Vet. Surg. 2006, 35, 232–242.

49 J. A. Roth, B.-G. Kim, W.-L. Lin, M.-I. Cho, J. Biol. Chem.

1999, 274, 22041–22047.

50 X. Wei, X. Yang, Z. Han, F. Qu, L. Shao, Y. Shi, Acta Phar-

macol. Sin. 2013, 34, 747–754.

51 P. A. Gurr, J. M. P. Scofield, J. Kim, Q. Fu, S. E. Kentish, G.

G. Qiao, J. Polym. Sci. Part A: Polym. Chem. 2014, 52, 3372–

3382.

52 S. A. Stern, V. M. Shah, B. J. Hardy, J. Polym. Sci., Part B:

Polym. Phys. 1987, 25, 1263–1298.

53 T. C. Merkel, V. I. Bondar, K. Nagai, B. D. Freeman, I.

Pinnau, J. Polym. Sci., Part B: Polym. Phys. 2000, 38, 415–

434.

54 L. Kwisnek, J. Goetz, K. P. Meyers, S. R. Heinz, J. S.

Wiggins, S. Nazarenko, Macromolecules 2014, 47, 3243–3253.


POLYMER SCIENCE


Crosslinked poly(ethylene oxide) containing siloxanes fabricated through thiol-ene photochemistry

Documents