Exploring resin viscosity effects in solventless processing of nano-SiO2/epoxy polymer hybrids

Cite this: RSC Advances, 2013, 3,3885

Exploring resin viscosity effects in solventlessprocessing of nano-SiO2/epoxy polymer hybrids

Received 8th June 2012,Accepted 9th January 2013

DOI: 10.1039/c3ra21150a

www.rsc.org/advances

Adeel Afzal,*abc Humaira Masood Siddiqi,*a Shaukat Saeedd and Zahoor Ahmade

Bisphenol A diglycidylether (DGEBA) based low viscosity, liquid epoxy resins are widely used as basis for

advanced polymers and nanocomposites, adhesives, protective coatings and encapsulation. We present

‘‘green’’ synthesis of nano-SiO2/epoxy polymer hybrids by a two-step chronological polymerization of

inorganic and organic monomers without the use of diluents (solvents). Two types of liquid epoxy resins,

D.E.R. 330 and D.E.R. 332, are used to demonstrate the influence of resin viscosity on microstructure,

tensile strength and thermal stability of resulting hybrids. Obviously, differences in viscosity of two epoxy

resins originate from variations in respective chain lengths, i.e. molar mass, which affect the overall

crosslink density and properties of hybrids. In addition, grafting of nano-SiO2 phases with organosilane is

performed to achieve inorganic–organic (IO) phase interlinking and to investigate its consequences. Nano-

SiO2/epoxy hybrids are characterized by FTIR spectroscopy and XPS. AFM is used to study microstructure

and surface properties of hybrids. AFM images show good distribution of nano-SiO2 phases within epoxy

polymer. It is observed that the size of nano-SiO2 grows significantly, if resin viscosity is increased or if

covalent IO phase interlinks are not present. Tensile measurements show considerable improvement in

strength and modulus of nano-SiO2/epoxy polymer hybrids as compared to neat epoxy polymers. DSC and

TGA also demonstrate an increase in glass transition temperature (Tg) and thermal stability. We observe

that viscosity effects are evenly pronounced in solventless processing of nano-SiO2/epoxy polymer hybrids,

and small changes in resin viscosity influence the miscibility of IO phases, the dispersion of SiO2 and the

performance of resulting hybrids.

1 Introduction

Inorganic–organic (IO) hybrids consisting of inorganic nano-phases, e.g. SiO2 nanoparticles, dispersed into epoxy polymermatrix have been studied extensively due to their desired andsometimes exceptional end properties resulting from thecombination of flexibility and toughness of epoxy polymersas well as hardness and heat resistance of SiO2 nanoparti-cles.1–10 Among different processing approaches, the sol–gelprocess10,11 is known to be a gentle technique that allowssynthesis of nano-SiO2 phases at low temperature. In addition,size, shape, nature and density of nano-SiO2 can be controlledby optimizing sol–gel reaction parameters such as theconcentration of water and Si(OR)4, acidic and-or basiccatalysts, solvents, temperature etc.

Previously, sol–gel SiO2/epoxy polymer hybrids have beenprepared using different routes. For instance, Matejka and co-workers12 reported the synthesis of such hybrids using one-step process and two-step simultaneous or sequential proce-dures. In one-step process, inorganic and organic componentsare mixed in a solvent and reacted altogether. On the contrary,two-step simultaneous and sequential processes involve eitherpre-hydrolysis of alkoxysilane before mixing it with organicmonomers to start simultaneous polymerization or reaction oforganic monomers at first to form polymer, which issubsequently swollen in excess of sol–gel solution.

Principally, these methods are based on the use of diluentsto ease processing of SiO2/epoxy polymer hybrids. The mostcommonly used diluents (solvents) e.g., ethanol, tetrahydro-furan (THF) etc., also bring about some undesirable character-istics to hybrid materials such as extensive shrinkage, higherporosity and lowering of glass transition (Tg) temperature.13 Itis therefore a recurring challenge to develop a solventless sol–gel processing strategy to prepare high performance SiO2/epoxy hybrids.

Phonthamachai et al.14 used a one-pot solventless proce-dure to synthesize SiO2/epoxy hybrids, which involved simul-taneous polymerization of inorganic and organic monomers.Benes et al.3,15 described a solventless, two-step procedure to

aDepartment of Chemistry, Quaid-i-Azam University, Islamabad, 45320, Pakistan.

E-mail: [email protected]; Fax: +92 51 9064 2241; Tel: +92 51 9064 2148bInterdisciplinary Research Centre in Biomedical Materials, COMSATS Institute of

Information Technology, Defence Road, Off. Raiwind Road, Lahore, 54000, Pakistan.

E-mail: [email protected] Colleges at Hafr Al-Batin, King Fahd University of Petroleum and

Minerals, P.O. Box 1803, Hafr Al-Batin, 31991, Saudi Arabia.dDepartment of Metallurgy and Materials Engineering, Pakistan Institute of

Engineering and Applied Sciences, Islamabad, 45650, Pakistan.eDepartment of Chemistry, Kuwait University, Safat, 13060, Kuwait.

RSC Advances

PAPER

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3885–3892 | 3885

prepare epoxy-silica hybrids, which involved synthesis ofprehydrolysed-and-condensed inorganic clusters first andtheir subsequent addition to a mixture of epoxy resin anddiamine monomers to yield SiO2/epoxy hybrids. Recently, wereported a different approach for ‘‘green’’ (or solventless)processing of nano-SiO2/epoxy hybrids based on two-stagechronological polymerization of inorganic and organic pre-cursors, where hydrolysis-condensation of alkoxysilanes wascarried out into liquid epoxy resin, and the mixture wassubsequently cured with a diamine hardener.16

Importantly, these studies3,14–16 demonstrated that thermaland mechanical properties of SiO2/epoxy hybrids were sig-nificantly improved using a solventless approach. Additionally,it is a fact that low viscosity, liquid epoxy resin D.E.R. 332 wasemployed in these experiments to substantiate homogeneousdispersion of nano-SiO2 phases in epoxy polymers. The mainaim of this study was to explore the effects of resin viscosityand mass on microstructure, thermal and mechanical proper-ties of nano-SiO2/epoxy hybrids prepared via solventless, two-stage chronological polymerization of inorganic and organicreactants. We used liquid epoxy resins with slightly differentviscosities to study structure-property- processing relation-ships. Moreover, nano-SiO2 phases were grafted with anorganosilane coupling agent to manipulate interfacial strengthand study its effects on the performance of these hybrids.

2 Experimental

2.1 Materials

Tetraethylorthosilicate (TEOS) and (3-glycidoxypropyl)tri-meth-oxysilane (GPTS) were obtained from Aldrich in the highestavailable purity. Two low viscosity liquid epoxy resins of thetype bisphenol A diglycidylether, D.E.R.TM 330 and D.E.R.TM

332 were obtained from Dow Chemicals. Table 1 providescharacteristic mass, viscosity and epoxide content in the twoepoxy resins. Jeffamine D400 with a molar mass of 428 g mol21

was received from Huntsman Corporation. All chemicals wereused as received without further purification.

2.2 Types of nano-SiO2/epoxy polymer hybrids

Two neat epoxy-amine polymers and different types of nano-SiO2/epoxy polymer hybrids were designed with either ofD.E.R. 330 and D.E.R. 332 epoxy resins, i.e. in total six sampleswere prepared, as described below;

2.2.1 JE neat polymers. These are reference polymersobtained by curing either of two epoxy resins with Jeffamine.

2.2.2 T/JE hybrids. These are simple un-grafted nano-SiO2/epoxy polymer hybrids in which nano-SiO2 phases werederived exclusively from TEOS.

2.2.3 GT/JE hybrids. These are organosilane grafted nano-SiO2/epoxy polymer hybrids in which nano-SiO2 phases werederived from a combination of GPTS and TEOS, mixed inmolar ratio of 1 : 16.4.

The amount of SiO2 in T/JE and GT/JE type hybrids was fixedat 12.5 phr, i.e. parts per hundred grams of the matrix. Table 2illustrates the sample designation and the exact compositionof JE neat polymers and different types (T/JE or GT/JE) of nano-SiO2/epoxy polymer hybrids.

The reactants were mixed in molar ratios according to well-recognized stoichiometry.8,17 The relative amounts of epoxyresin and diamine were adjusted on addition of GPTS, asdescribed elsewhere.16

2.3 Synthesis

Neat epoxy-amine polymers (JE330 and JE332) were synthe-sized by mixing stoichiometric amounts of Jeffamine D400with respective epoxy resin in a Teflon beaker. The mixture wasstirred vigorously for 1 h at 25 uC and subsequently casted inopen Teflon moulds with dimensions (l 6 w 6 h = 50 6 10 60.2 mm). Optically transparent thin polymer films wereobtained by curing the mixture at 100 uC for 5 h, and at 110uC for 1 h.

T/JE type nano-SiO2/epoxy hybrids were prepared as thin, freestanding polymer films. Calculated amounts of TEOS weremixed into respective epoxy resins under absolute dry atmo-sphere inside a glove box, and stirred vigorously for 2 h at 25 uCto obtain a homogeneous mixture. The sol–gel reaction wasinitiated by the addition of stoichiometric amount of water,already at pH = 2. The mixture was stirred for 1 h at 25 uC and 4h at 60 uC to complete the sol–gel process. Jeffamine was addedto the mixture, and it was further stirred for 1 h at 25 uC.Transparent thin films were obtained by casting and curing themixtures according to the abovementioned procedure.

GT/JE type nano-SiO2/epoxy hybrids vary from T/JE hybrids inthe presence of GPTS as the interphase linker. Thus, as the firststep, calculated amount of epoxy resin was mixed with GPTS inabsolute dry atmosphere, and stirred vigorously for 2 h at 25 uC.Later, TEOS was added to the mixture and the aforementionedprocedure was followed to synthesize GT/JE hybrids.

2.4 Methods

Fourier transform infrared (FTIR) spectroscopy was used tocharacterize various kinds of nano-SiO2/epoxy polymer

Table 1 Characteristics of different epoxy resins used in this study

D.E.R.TM 330 D.E.R.TM 332

Epoxide equivalent weight (g eq.21) 176–185 171–175Epoxide percentage (%) 23.2–24.4 24.6–25.1Epoxide group content (mmol kg21) 5400–5680 5710–5850Viscosity @ 25 uC (mPa.s) 7000–10 000 4000–6000

Table 2 Sample designation and composition of synthesized polymers andhybrids

Sample Epoxy (g) Amine (g) TEOS (g) GPTS (g) H2O (g)

JE330 6.34 3.66 — — —JE332 6.21 3.79 — — —T/JE330 5.55 3.20 4.354 — 1.125T/JE332 5.43 3.32 4.354 — 1.125GT/JE330 5.33 3.20 4.132 0.285 1.117GT/JE332 5.22 3.32 4.132 0.285 1.117

3886 | RSC Adv., 2013, 3, 3885–3892 This journal is � The Royal Society of Chemistry 2013

Paper RSC Advances

hybrids. The PerkinElmer system 2000 series FTIR spectro-meter was used to examine thin films and infrared spectrawere recorded in the range of 4000–600 cm21.

Surface elemental characterization of nano-SiO2/epoxy poly-mer hybrids was performed with the help of X-RayPhotoelectron Thermo VG Theta Probe spectrometer,equipped with a micro-spot monochromatized Al Ka source.Both survey scans and high-resolution elemental spectra wereacquired in fixed analyser transmission mode with pass energyof 150 eV and 100 eV, respectively.

The surface morphology of 150–200 mm thick free-standingfilms was studied by atomic force microscopy (AFM). The AFMimages were recorded on a Multimode Digital InstrumentsNanoscopeTM dimension IIIa AFM, equipped with PPP-NCLRtapping mode probe, NanosensorsTM Switzerland, underambient conditions using contact mode. The scanning probewas equipped with a triangular cantilever having a pyramidalsilicon nitride tip. The micrographs were analysed on WSxM3.0 Beta V.8.1 software, by Nanotech Electronica S.L.

The mechanical tests were performed with thin films ofthickness 150–200 mm cut into small pieces of 15.0 5.0 mm,and vacuum dried at 70 uC for 5 h. The stress-strain curveswere obtained with Testometric Universal Testing MachineM350–500, using a cross head speed of 5 mm min21. For highaccuracy and precision, a sensitive load cell of 100 Kgf capacitywith 1.0 mg load sensitivity and a minimum of 0.01 mm crosshead speed, was used. The pneumatic griping system consist-ing of nitrile rubber lined faces, especially designed for thinfilms’ firm griping, was used to avoid any slippage during thetest. The measurements were performed at 25 uC and theaverage values obtained from at least six replicas werereported.

Differential scanning calorimetry (DSC) was performed withthe Perkin–Elmer DSC 7 instrument by heating approximately10 mg of sample in a sealed aluminium pan under inertatmosphere. The instrument was ramped at 5 uC min21 from220 to 310 uC. Normally, a second DSC scan from 0 to 120 uCwas also performed to report precise values of Tg.

Decomposition profile of nano-SiO2/epoxy polymer hybridsas well as neat epoxy polymers was obtained via thermogravi-metric analysis (TGA). Perkin–Elmer TGA 7 Analyser was usedfor this purpose. Approximately, 10 mg of samples were heatedto 800 uC at a rate of 20 uC min21 under inert atmosphere, andweight loss was monitored as a function of temperature.

3 Results

3.1 Characterization

The characteristic FTIR spectra of T/JE330 and GT/JE330hybrids are shown in Fig. 1, which provide imperative evidencefor the hydrolysis-condensation of TEOS and GPTS as well asthe crosslinking of epoxy and amine monomers. The char-acteristic Si–O–Si and Si–OH stretching frequencies are usuallyobserved in the range of 1095–1075 cm21 and around 3400cm21, respectively.18 The absorptions at 1085 and y3400cm21 thus confirm the formation of SiO2 phases. Thecrosslinking of epoxide and amino groups is verified by

weakening and finally disappearance of the intrinsic C–O–Cstretching of epoxy resins at 925 and 865 cm21, respec-tively.16,19 The presence of sharp tert-amine absorption at 1184cm21 further confirms the crosslinking of epoxy groups.19

Furthermore, the characteristic absorption of epoxide at 925cm21 is examined during the cure to estimate the extent ofconversion of epoxy resins. FTIR results show that intensity ofcyclic C–O–C stretch diminishes as the cure proceeds, seeFig. 1 (inset). Apparently, the intensity of cyclic C–O–Cstretching is negligible at the end of the curing cycle, whichsignifies the conclusion of epoxide ring opening reaction withdiamine.

The surface chemical characterization of nano-SiO2/epoxyhybrids is performed by means of XPS to determine thesurface elemental composition and chemical states. Theresults of surface elemental analysis of different types ofnano-SiO2/epoxy polymer hybrids are reported in Table 3.Carbon, nitrogen, oxygen and silicone were detected by XPS onthe surface of all samples. In order to precisely determine thecontents of functional groups of Si and C atoms, deconvolu-tion was carried out using Avantage data system provided byThermo Scientific. As an example, the survey scan and thehigh resolution atomic spectra of a T/JE hybrid are representedin Fig. 2.

The major difference between T/JE and GT/JE hybrids wasobserved in the respective high resolution spectra of silicone(Si 2p), shown in Fig. 2. Si 2p spectrum of GT/JE hybridexhibits two distinct peaks of 1.37 eV FWHM at 102.1 and103.4 eV. These peaks are attributed to CH2–SiO3 (derived fromGPTS) and SiO4 (derived from TEOS), respectively.20 In case ofT/JE hybrid, Si 2p spectrum shows a featureless peak of 1.36 eVFWHM at 103.3 eV with a low BE shoulder (or tail). Thesedifferences in Si 2p spectra of T/JE and GT/JE hybrids areevident of the presence of GPTS in GT/JE hybrid. The presenceof single featureless Si 2p peak at 103.4 ¡ 0.1 eV also confirmsthe formation of SiO2 phases within epoxy polymers.20,21

Fig. 1 FTIR spectra of nano-SiO2/epoxy polymer hybrids. The inset showscharacteristic IR absorption of cyclic C–O–C stretching during the course ofcuring: a decrease in intensity is obvious as reaction proceeds.

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3885–3892 | 3887

RSC Advances Paper

Additionally, O 1s spectrum provides further evidence of theformation of SiO2 phases, as it exhibits a peak of 1.33 eVFWHM at 532.3 eV. According to the pertinent literaturesurvey, O 1s peak attributed to O–Si linkages in SiO2 appears at532.4 eV.21

The deconvolution of O 1s spectrum gives a second peak at533.8 eV, which is attributed to O–C and O–H linkages in thepolymer backbone. Similarly, C 1s spectra were deconvolutedinto three distinct components. The first C 1s peak, located at284.8 eV, corresponds to C–C or C–H group, while the secondone, located at 286.3 eV, is assigned to C-atoms in C–O–C andC–O–H groups. C 1s peak at higher BE corresponds to CLOlinks present as oxidized components of the polymer or in theform of adsorbed CO2. C–N groups were not detected probablydue to very low atomic concentration of N-atoms on the

surface of hybrids. However, the high resolution N 1sspectrum is promising for XPS characterization of nano-SiO2/epoxy polymer hybrids. It shows a solitary N 1s peak at 398.6 ¡

0.2 eV, which is attributed to tert-amines.22 Hence, thepresence of single N 1s peak substantiates the crosslinkingof epoxy and amine monomers.

3.2 Surface morphology

Topography and surface properties of thin nano-SiO2/epoxyhybrid films are profoundly influenced by size and distribu-tion of SiO2 phases. AFM was used to analyse surfacemorphology of neat polymers and different types of nano-SiO2/epoxy hybrids. The surface properties are recorded inTable 4. The morphology of JE330 and JE332 polymers isconsistently uniform with RMS roughness (Rq) of 1.3 and 1.1

Table 3 XPS surface elemental analysis and chemical speciation of various types of nano-SiO2/epoxy polymer hybrids

Region Binding energy (eV) Attribution

Relative abundance (At. %)

T/JE330 GT/JE330 T/JE332 GT/JE332

C 1s 284.8 ¡ 0.1 C–C a 64.6 ¡ 0.5 63.9 ¡ 0.5 62.1 ¡ 0.5 59.7 ¡ 0.5286.3 ¡ 0.1 C–O–R b 7.5 ¡ 0.5 8.3 ¡ 0.5 9.1 ¡ 0.5 10.7 ¡ 0.5288.8 ¡ 0.1 CLO c 0.2 ¡ 0.1 0.3 ¡ 0.1 0.3 ¡ 0.1 0.2 ¡ 0.1

N 1s 398.6 ¡ 0.2 N–(R)3 (tert-amine) 0.6 ¡ 0.2 0.7 ¡ 0.2 1.2 ¡ 0.2 1.0 ¡ 0.2O 1s 532.3 ¡ 0.1 O–Si (SiO2) 13.4 ¡ 0.5 12.2 ¡ 0.5 11.5 ¡ 0.5 12.1 ¡ 0.5

533.8 ¡ 0.1 R–O–C b 6.9 ¡ 0.5 8.5 ¡ 0.5 9.9 ¡ 0.5 10.5 ¡ 0.5Si 2p 102.1 ¡ 0.1 (CH2)–Si–(O)3

d — 0.9 ¡ 0.2 — 1.1 ¡ 0.2103.4 ¡ 0.1 Si–(O)4 (SiO2) 6.8 ¡ 0.5 5.2 ¡ 0.5 5.9 ¡ 0.5 4.7 ¡ 0.5

a Aliphatic and aromatic carbons. b Alcohol and ether groups (R = hydrogen or alkyl group). c Carboxylic acids and-or adsorbed carbondioxide. d Silica bonded to alkyl groups (originating from GPTS).

Fig. 2 The survey scan and the high resolution XPS spectra of T/JE type nano-SiO2/epoxy polymer hybrids. High resolution Si 2p spectrum of a GT/JE hybrid is alsoshown.

3888 | RSC Adv., 2013, 3, 3885–3892 This journal is � The Royal Society of Chemistry 2013

Paper RSC Advances

nm, respectively. The surface properties of both neat polymersare therefore comparable, and no significant distinction inmorphology is observed on increasing resin viscosity.

The existence of nano-SiO2 phases into epoxy polymers ismanifested through AFM images of nano-SiO2/epoxy hybrids,as shown in Fig. 3. The average domain size for silica, asmeasured from AFM phase images (not shown), is in the rangeof 50–100 nm for different types of hybrids. The surfacemorphology of T/JE332 (Fig. 3a) and T/JE330 (Fig. 3b) hybridsis specified by several height points (or peaks) distributedalongside xy-plane. These peaks are recognized as protractednano-SiO2 phases. It is evident that dispersion of SiO2 ishomogeneous and cluster formation is rare. However, the sizeand distribution of nano-SiO2 is influenced by resin viscosity.The respective histograms reveal that lower the resin viscosity,the smaller the particle size and better the dispersion of SiO2,(as discussed later in Section 4).

In the case of T/JE330 hybrid, the peaks are distributed overa wider size range with majority of them lying within 26–46nm. On the other hand, smaller size and relatively uniformdistribution are obvious in the case of T/JE332 hybrid withmain-stream height data found in the range of 10–30 nm. It isdue to the greater viscosity of DER 330 epoxy that distributionof TEOS monomers is hindered, consequently formingrelatively dense nano-SiO2 phases. In GT/JE hybrids, theinclusion of GPTS is expected to form covalent IO interlinks,thereby reinforcing interphase interactions. These IO phase-interlinks consequently influence size and overall distributionand size of nano-SiO2 phases, as shown in Fig. 4. In GT/JE332hybrid (Fig. 4a), for instance, the dispersion of SiO2 isparticularly constant with significantly smaller peak heights,i.e. in the range of 6–16 nm.

Similarly, in the case of GT/JE330 hybrid (Fig. 4b), smallsized nano-SiO2 phases are formed with majority of peakslying in the range of 8–24 nm. Interestingly, the stronger

Table 4 Surface properties of neat epoxy polymers and different types of nano-SiO2/epoxy hybrids

Surface properties JE330 JE332 T/JE330 T/JE332 GT/JE330 GT/JE332

RMS roughness (Rq) (nm) 1.3 ¡ 0.1 1.1 ¡ 0.1 7.2 ¡ 0.4 6.4 ¡ 0.3 5.2 ¡ 0.3 4.6 ¡ 0.2Average height (nm) 10 6 30 37 19 17Maximum height (nm) 22 17 82 74 55 51

Fig. 3 AFM images of the T/JE type nano-SiO2/epoxy hybrids. Fig. 4 AFM images of the GT/JE type nano-SiO2/epoxy hybrids.

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3885–3892 | 3889

RSC Advances Paper

interfacial interactions overcome the subdued effects of resinviscosity in T/JE330 hybrids to some extent, and the dispersionof SiO2 is appreciably superior in GT/JE330 hybrid.

3.3 Mechanical properties

Mechanical properties of neat polymers and different types ofnano-SiO2/epoxy hybrids are determined from the respectivetensile measurements, according to the ASTM D882 teststandard specifications for thin plastic sheeting.23 Typicalstress-strain curves of various nano-SiO2/epoxy hybrid filmsare shown in Fig. 5, and the respective mechanical propertiesare shown in Table 5. It is evident that yield strength andtensile modulus of nano-SiO2/epoxy hybrids are significantlyhigher as compared to neat polymers. Resin mass (andviscosity) effects are noticeable and clearly demonstrated.Tensile modulus and yield strength tend to decrease withincrease in resin viscosity and equivalent resin mass owing tothe decrease in crosslinking density. The ultimate fracturestrength, however, increases with the increase in resin mass,which is attributed to the inherent ductility of the resin.24

A steady increase in mechanical properties of nano-SiO2/epoxy hybrids is observed in the presence of GPTS. IO phase

interlinks formed by GPTS help in transferring the stressefficiently to polymer chains, thereby enhancing the mechan-ical strength of GT/JE type hybrids.25 In case of T/JE hybrids,however, weaker H-bonds at IO interface are incapable of theeffective dispersal of stress. Consequently, T/JE hybridspossess much lower strength and modulus.

3.4 Glass transition temperature (Tg)

DSC scans of nano-SiO2/epoxy polymer hybrids demonstrate asignificant increase in their glass transition temperature (Tg)in comparison to that of neat epoxy polymers, see Table 6.Additionally, resin viscosity and IO phase interlinks have amarked influence on the Tg of nano-SiO2/epoxy hybrids. It isobserved that Tg decreases with the increase in resin mass(and viscosity) owing to the decrease in crosslinking density ofresulting polymers.26,27 For instance, JE332 neat polymer hasapproximately 2 uC higher Tg as compared to JE330 neatpolymer, and this tendency is found preserved in therespective nano-SiO2/epoxy hybrids.

In addition, the microstructure and the interphase interac-tions have more pronounced influence on Tg of nano-SiO2/epoxy hybrids. T/JE type hybrids, whether they are based onD.E.R. 330 or D.E.R. 332 epoxy resin, Tg are much lower thanGT/JE hybrids, which is attributed to the strong inhibition ofmatrix chains’ mobility by covalent bonding between nano-SiO2 and epoxy phases.16,27

3.5 Thermal stability

TGA traces of neat epoxy polymers and nano-SiO2/epoxyhybrids give a measure of their relative thermal stability.Fig. 6 shows typical TGA thermograms of neat polymers andnano-SiO2/epoxy hybrids. Interestingly, TGA results are com-parable for different types of hybrids especially at the onset ofthermal degradation, and no particular trends are observed inthe degradation profile of various hybrids. Nonetheless,thermal stability of epoxy polymers is considerably enhancedby the addition of nano-SiO2 phases.

The degradation temperatures Td10 (at 10% weight loss),Td50 (at 50% weight loss), and Tmax (temperature at which rateof decomposition in maximum) demonstrate that rate ofthermal oxidative degradation of nano-SiO2/epoxy hybrids issignificantly lowered. It is due to the fact that nano-SiO2

Fig. 5 Typical stress-strain responses of nano-SiO2/epoxy hybrids.

Table 6 Thermal properties of neat epoxy polymers and different types of nano-SiO2/epoxy hybrids

Thermal properties JE330 JE332 T/JE330 T/JE332 GT/JE330 GT/JE332

Tg (uC) 42.5 ¡ 0.5 44.6 ¡ 0.5 46.0 ¡ 0.5 47.2 ¡ 0.5 51.1 ¡ 0.5 53.5 ¡ 0.5Td10 (uC) 316 323 342 349 348 356Td50 (uC) 351 359 397 404 426 442Tmax (uC) 365 376 392, 565 396, 614 420, 636 408, 602

Table 5 Mechanical properties of neat epoxy polymers and different types of nano-SiO2/epoxy hybrids

Mechanical properties JE330 JE332 T/JE330 T/JE332 GT/JE330 GT/JE332

Yield strength (MPa) 18.7 19.9 27.9 28.3 29.8 33.1Fracture strength (MPa) 16.4 14.8 21.5 17.9 25.1 22.3Tensile modulus (GPa) 0.64 ¡ 0.05 0.77 ¡ 0.05 1.00 ¡ 0.10 1.16 ¡ 0.06 1.20 ¡ 0.05 1.28 ¡ 0.05

3890 | RSC Adv., 2013, 3, 3885–3892 This journal is � The Royal Society of Chemistry 2013

Paper RSC Advances

phases inhibit degradation and bring stability to hybridmaterials at elevated temperatures.28,29 Resin mass andviscosity effects on thermal degradation behaviour of neatepoxy polymers as well as nano-SiO2/epoxy hybrids arenegligible. IO interphase crosslinking, however, is responsibleto enhance relative thermal stability of GT/JE hybrids at hightemperatures as compared to T/JE hybrids.

4 Discussion

The structure-property-processing relationships in differenttypes of nano-SiO2/epoxy polymer hybrids are studied withspecific emphasis on exploring resin viscosity and mass effectson microstructure, surface and physical properties of these IOhybrids. Albeit the differences between resin viscosity andmass are marginal, they profoundly influence the microstruc-ture and the performance of nano-SiO2/epoxy hybrids preparedvia solventless, two-stage chronological polymerization ofinorganic and organic precursors.

It is well-known that thermal and mechanical properties ofnano-SiO2/epoxy hybrids are primarily determined by crosslinkdensity and interfacial strength.30,31 The following paragraphsdiscuss how slight changes in resin viscosity and masstransform crosslink density and interfacial strength tocritically influence the ultimate properties of these hybrids.

4.1 Resin viscosity vs. microstructure and performance ofnano-SiO2/epoxy polymer hybrids

Acid catalysed hydrolysis-condensation of alkoxysilanes yieldsextended coil-like, ramified nano-SiO2 structures.12,32 Surfacemorphology of various nano-SiO2/epoxy hybrids essentiallysupports this phenomenon. AFM images in Fig. 3 and 4 shownano-SiO2 phases consistently dispersed into epoxy polymer.However, size and dispersion of SiO2 are evidently affected byresin viscosity. In fact, in solventless processing of thesehybrids, surface properties and interface strength are manipu-lated greatly by resin viscosity.

In the case of low viscosity epoxy resin, the scattering of tinySiO2 phases (or particles) is enhanced, and consequentlyparticle growth is restricted. On the other hand, slight increase

in resin viscosity limits particle dispersion. We thereforeobserved larger SiO2 particles or clusters in case of T/JE330hybrid (Fig. 3b). Eventually, smaller nano-SiO2 phases offergreater surface area for interphase interactions, therebyimproving interfacial strength. Addition of GPTS furtherreduces the size of nano-SiO2 phases partly by steric cappingof the relatively smaller SiO2 particles and partly by improvingtheir adhesion with the matrix. Hence, the interactionsbetween IO phases are considerably strengthened in GT/JEhybrids.

Surface characterization of nano-SiO2/epoxy hybrids pro-vides substantial evidence of the influence of resin viscosity onsize and dispersion of nano-SiO2 phases within epoxypolymers, interface strength, and performance of varioushybrids. Thus, mechanical and thermal properties of D.E.R.332-based nano-SiO2/epoxy hybrids are superior to D.E.R. 330-based hybrids due to slightly lower viscosity of D.E.R. 332 andhigher interface strength of GT/JE332 and T/JE332 hybrids ascompared to GT/JE330 and T/JE330 hybrids.

4.2 Resin mass vs. microstructure and performance of nano-SiO2/epoxy polymer hybrids

Mechanical and thermal characterization results demonstratethat nano-SiO2/epoxy hybrids based on low molecular weightepoxy resin are better. This may be explained on the basis ofrubber elasticity theory, which states that the crosslink densityof epoxy networks is inversely related to the molecular weightbetween crosslinks, i.e. mass of thermosetting resin.28

In addition to the changes in crosslink density, it isimportant to understand the structural changes and chemistryof nano-SiO2/epoxy hybrids associated with the changes inresin mass. For instance, the increase in resin mass means:the concentration of amine groups is decreased in the hybrid,while ether content is increased (see Table 2 and XPS results).Furthermore, an epoxide group produces one hydroxyl groupupon ring opening. Consequently, the number of hydroxylgroups per unit volume is also decreased with the increase inresin mass. It is a fact that nano-SiO2/epoxy hybrids based onD.E.R. 332 epoxy resin contain greater number of hydroxyl andamine groups, while lesser ether groups per unit volume of thecured structure as compared to the D.E.R. 330 analogues.

Unfortunately, it is difficult to precisely evaluate the effectsof this chemical nature on crosslink density and properties ofhybrids. Nonetheless, the interface is believed to be stronger inthe presence of larger number of amine and hydroxylfunctional groups, since these groups are known to strengthenphysical interactions such as hydrogen bonds between IOphases.33 In addition, the presence of GPTS in GT/JE typehybrids ensures chemical crosslinking of nano-SiO2 phaseswith epoxy polymers. On the basis of these facts, the interfacestrength and the crosslink density in nano-SiO2/epoxy hybridsare expected to decrease in the following order: GT/JE332 .

GT/JE330 . T/JE332 . T/JE330. To summarize, it is obviousthat the performance of various nano-SiO2/epoxy hybrids alsodeclines in the same order.

Fig. 6 Typical TGA curves of neat epoxy polymers and nano-SiO2/epoxy hybrids.

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3885–3892 | 3891

RSC Advances Paper

5 Conclusions

Solventless processing and analytical characterization of sur-face, mechanical and thermal properties of nano-SiO2/epoxypolymer hybrids are described with particular emphasis onstudying the influence of resin viscosity (and mass) oninterfacial strength, crosslink density, and performance ofthese hybrids. Tensile strength, modulus, Tg and thermalstability of nano-SiO2/epoxy hybrids are substantially improvedas compared to neat epoxy polymers. Neat polymers as well asnano-SiO2/epoxy hybrids perform best, if low molecular weightand low viscosity epoxy resin are utilized.

It is demonstrated that the ultimate properties of hybridmaterials are effectively controlled by the nature, size anddispersion of nano-SiO2 phases, the microstructure and degreeof crosslinking in nano-SiO2/epoxy hybrids, and the intensityof IO interphase interactions. In conclusion, it is substantiatedthat all these performance determining factors for nano-SiO2/epoxy hybrids are radically affected by resin viscosity, andfurther increases in resin viscosity can have detrimental effectson properties of similar hybrids prepared via solventlessapproach.

Acknowledgements

This work is supported by the University Research Fund (URF)of Quaid-i-Azam University, Islamabad, and by the NationalResearch Program for Universities (NRPU) of HigherEducation Commission (Grant No. 1308). AA thanks Prof. F.L. Dickert (Univ. Wien, Austria) for providing AFM facility, andProf. L. Sabbatini, Prof. L. Torsi, Dr N. Cioffi and Dr N.Ditaranto (Univ. di Bari, Italy) for providing XPS characteriza-tion facilities, basic training, and helpful discussions.

References1 R. K. Donato, K. Z. Donato, H. S. Schrekker and L. Matejka,

J. Mater. Chem., 2012, 22, 9939.2 M. Lettieri, F. Lionetto, M. Frigione, L. Prezzi and

L. Mascia, Polym. Eng. Sci., 2011, 51, 358.3 H. Benes, J. Galy, J.-F. Gerard, J. Plestil and L. Valette, J. Sol-

Gel Sci. Technol., 2011, 59, 598.4 G. Ragosta, M. Abbate, P. Musto, G. Scarinzi and L. Mascia,

Polymer, 2005, 46, 10506.5 N. Suzuki, M. B. Zakaria, Y.-D. Chiang, K. C.-W. Wu and

Y. Yamauchi, Phys. Chem. Chem. Phys., 2012, 14, 7427.6 Y. Deng, J. Bernard, P. Alcouffe, J. Galy, L. Dai and J.-

F. Gerard, J. Polym. Sci., Part A: Polym. Chem., 2011, 49,4343.

7 M. Lettieri and M. Frigione, Constr. Build. Mater., 2012, 30,753.

8 T. Nazir, A. Afzal, H. M. Siddiqi, Z. Ahmad and M. Dumon,Prog. Org. Coat., 2010, 69, 100.

9 S. R. Davis, A. R. Brough and A. Atkinson, J. Non-Cryst.Solids, 2003, 315, 197.

10 F. Bondioli, M. E. Darecchio, A. S. Luyt and M. Messori, J.Appl. Polym. Sci., 2011, 122, 1792.

11 M. Pagliaro, R. Ciriminna and G. Palmisano, J. Mater.Chem., 2009, 19, 3116.

12 L. Matejka, O. Dukh and J. Kolarik, Polymer, 2000, 41, 1449.13 S. R. Lu, H. L. Zhang, C. X. Zhao and X. Y. Wang, J. Mater.

Sci., 2005, 40, 1079.14 N. Phonthamachai, H. Chia, X. Li, F. Wang, W. W. Tjiu and

C. He, Polymer, 2010, 51, 5377.15 H. Benes, J. Galy, J.-F. Gerard, J. Plestil and L. Valette, J.

Appl. Polym. Sci., 2012, 125, 1000.16 A. Afzal and H. M. Siddiqi, Polymer, 2011, 52, 1345.17 M. Ochi, R. Takahashi and A. Terauchi, Polymer, 2001, 42,

5151.18 K. H. Wu, C. M. Chao, C. J. Yang and T. C. Chang, Polym.

Degrad. Stab., 2006, 91, 2917.19 M. Ochi and R. Takahashi, J. Polym. Sci., Part B: Polym.

Phys., 2001, 39, 1071.20 K. Sever, Y. Seki, H. A. Gulec, M. Sarikanat, M. Mutlu and

I. Tavman, Appl. Surf. Sci., 2009, 255, 8450.21 Y. S. Lin, Y. H. Liao and C. H. Hu, J. Non-Cryst. Solids, 2009,

355, 182.22 C. D. Wagner, A. V. Naumkin, A. Kraut-Vass, J. W. Allison,

C. J. Powell and J. R. Rumble, NIST X-Ray PhotoelectronSpectroscopy Database, NIST Standard Reference Database20, Version 3.5.

23 ASTM D882-10, Annual Book of ASTM Standards, Am. Soc.Test. Mater., 2010, 08.01 (PA).

24 T. K. Chen and H. J. Shy, Polymer, 1992, 33, 1656.25 A. Afzal, H. M. Siddiqi, S. Saeed and Z. Ahmad, Mater.

Express, 2011, 1, 299.26 P. Rosso and L. Ye, Macromol. Rapid Commun., 2007, 28,

121.27 H. Zhang, Z. Zhang, K. Friedrich and C. Eger, Acta Mater.,

2006, 54, 1833.28 J. S. Hwang, M. J. Yim and K. W. Paik, J. Electron. Mater.,

2006, 35, 1722.29 A. Afzal, H. M. Siddiqi, N. Iqbal and Z. Ahmad, J. Therm.

Anal. Calorim., 2013, 111, 247–252.30 L. Prezzi and L. Mascia, Adv. Polym. Technol., 2005, 24, 91.31 T. Nazir, A. Afzal, H. M. Siddiqi, S. Saeed and M. Dumon,

Polym. Bull., 2011, 67, 1539.32 L. Matejka, O. Dukh, D. Hlavata, B. Meissner and J. Brus,

Macromolecules, 2001, 34, 6904.33 J. Lee and A. F. Yee, Polymer, 2000, 41, 8375.

3892 | RSC Adv., 2013, 3, 3885–3892 This journal is � The Royal Society of Chemistry 2013

Paper RSC Advances