Fabrication and Characterization of Iron Oxide Nanoparticles Reinforced Vinyl-Ester Resin Nanocomposites

Available online at www.sciencedirect.com COMPOSITES SCIENCE AND TECHNOLOGY Composites Science and Technology 68 (2008) 1513–1520 www.elsevier.com/locate/compscitech

Fabrication and characterization of iron oxide nanoparticles reinforced vinyl-ester resin nanocomposites

Zhanhu Guo a,*, Kenny Lei a, Yutong Li b, Ho Wai Ng a, Sergy Prikhodko c, H. Thomas Hahn a

a Multifunctional Composites Lab, Mechanical and Aerospace Engineering Department, University of California, Los Angeles, CA 90095, United States b Western Digital Corporation, Magnetic Head Operation, Fremont, CA 94539, United States c Materials Science and Engineering Department, University of California Los Angeles, CA 90095, United States

Received 23 August 2007; received in revised form 13 October 2007; accepted 17 October 2007 Available online 26 October 2007

Abstract

Robust magnetic vinyl ester resin nanocomposites reinforced with iron oxide (Fe2O3) nanoparticles were fabricated. The particle functionalization with a bi-functional coupling agent methacryloxypropyl-trimethoxysilane (MPS) was observed to have a significant effect on the curing process and subsequent physical properties of the nanocomposites. Particle functionalization favors the composite fabrication with a lower curing temperature as compared to the as-received nanoparticles filled vinyl ester resin nanocomposites. Thermo- gravimetric analysis showed an increased thermo-stability in the functionalized nanoparticles filled vinyl ester resin nanocomposites as compared to the unmodified nanoparticle filled counterparts. The more uniform particle dispersion and the chemical bonding between nanoparticle and vinyl ester resin matrix were found to contribute to the increased thermal stability and enhanced tensile strength. The nanoparticles become magnetically harder after incorporation into the vinyl ester resin matrix. Published by Elsevier Ltd.

Keywords: A: Nanocomposites; A: Particle-reinforced composites; A: Polymer–matrix composites (PMCs); B: Mechanical properties; B: Magnetic properties

1. Introduction inorganic nanoparticles have attracted much interest due to their lightness, homogeneity, cost-effective processabil- Nanomaterials dramatically different from their bulk or ity, and tunable physical properties [9,11–13]. Vinyl-ester atomic counterparts have attracted much interest due to resin, as a structural polymer, was chosen as a polymer their unique physicochemical properties [1] for a wide matrix in current study due to the fact that the cured res- range of potential device applications such as UV lasers ins are thermosetting with a network structure possessing [2], solar cells [3,4], high-sensitivity chemical gas [5] or high resistance to the moisture and chemicals, and good volatile organic compound [6] sensors, and DNA mechanical properties. Thus the resultant composites have sequence [7] sensors. Nanostructural materials such as the potential applications in fabrication and building nanoparticles (NPs) or nanofibers have been used as fillers materials such as electrodeposition tanks, automotive in both the polymeric nanocomposites [8,9] to improve parts and marine vessels which require superior mechani- the mechanical, electric, electronic and optical properties, cal properties and/or high resistance to harsh environ- and the metallic nanocomposite [10] to control the elec- ments such as strong acid or base. Furthermore, the trodeposition. Polymer nanocomposites reinforced with functional groups of the polymer surrounding the nanoparticles enable these nanocomposites as good candidates for various applications such as site-specific molecule tar- * Corresponding author. Tel.: +1 310 206 8157. E-mail address: [email protected] (Z. Guo). geting in biomedical areas [14].

0266-3538/$ - see front matter Published by Elsevier Ltd. doi:10.1016/j.compscitech.2007.10.018 1514 Z. Guo et al. / Composites Science and Technology 68 (2008) 1513–1520

The existing challenges in the composite fabrication are catalyst or initiator, organic peroxide, liquid) was pur- to obtain uniform filler dispersion and to introduce strong chased from Akzo Nobel Chemicals. Cobalt naphthenate chemical bonding between the nanoparticles and the poly- (CoNap, OM Group, Inc.) was used as a catalyst promoter mer matrix, which are necessary to provide a high tensile (accelerator) to decompose the catalyst at room tempera- strength due to local stress within the nanocomposite. ture. Methacryloxypropyl-trimethoxysilane (MPS) and tet- The interfacial interactions between fillers and polymer rahydrofuran (THF, anhydrous) were purchased from matrix play a crucial role in determining the quality and Sigma–Aldrich Chemical Company. All the chemicals were properties of the nanocomposites [15]. The poor bonding used as-received without further treatment. linkage between the fillers and the polymer matrix such as the composites made by simple mixing [15] will intro- 2.2. Surface functionalization of iron oxide nanoparticles duce artificial defects, which consequently result in a dele- terious effect on the mechanical properties of the The nanoparticle functionalization follows procedures nanocomposites [16]. Introducing good linkages between similar to our earlier reported nanoparticle process [19– the fillers and the polymer matrix is still a challenge for spe- 21] and is described as follows. Fe2O3 nanoparticles cific composite fabrication. However, appropriate chemical (13.38 g, 83.8 mmol) were added into a mixture of 4 g engineering treatment of the nanofiller surface by introduc- MPS and 30 ml THF. The resulting colloidal suspension ing proper functional groups could improve both the was ultrasonically stirred (Branson 1510) for 1 h and pre- strength and toughness of the subsequent composites with cipitated by a permanent magnet at room temperature. improved compatibility between the nanofillers and the The precipitated nanoparticles were rinsed with THF to polymer matrix, and make the nanocomposites stable in remove excessive MPS for subsequent nanocomposite fab- harsh environments as well [17]. Thus, surface functionali- rication and the dried nanoparticles were used for further zation of nanoparticles with a surfactant or a coupling particle characterization. agent is important not only to stabilize the nanoparticles [18] during processing but also to render them compatible 2.3. Nanocomposite fabrication with the polymer matrix. In this paper, the effect of particle functionalization by a The as-received Fe2O3 nanoparticles or MPS functional- bi-functional methacryloxypropyl-trimethoxysilane (MPS) ized Fe2O3 nanoparticles were dispersed into 30 ml resin on on the vinyl ester resin curing process and the optimum con- a specific weight percentage basis. The dispersion was car- ditions for high-quality nanocomposite fabrication were ried out in an ice-water ultrasonic bath for about 1 h. The investigated. The functionalized iron oxide nanoparticles above particle-suspended solution was then ultrasonically reinforced vinyl ester nanocomposites showed enhanced stirred in an ice-water ultrasonic bath until the temperature mechanical properties under tensile study. The nanocom- was cooled down. Then, 2.0 wt% catalyst (initiator) was posites containing the functionalized iron oxide nanoparti- added into the nanoparticle/resin solution, which was stir- cles showed improved thermal stability as compared with red and degassed for 2 min. Next, 0.3 wt% promoter was the nanocomposites reinforced with the as-received iron added and mixed quickly. The mixed viscous solution oxide nanoparticles. The nanoparticles were found to be was poured into silicone rubber molds. The curing via magnetically harder (larger coercivity) after the nanoparti- free-radical bulk copolymerization or homopolymerization cles were dispersed in the vinyl ester resin matrix. initiated by the catalyst was done at 85 °C for 1 h under normal atmospheric conditions and cooled down to room 2. Experimental temperature naturally in the oven following procedures similar to those used for our reported alumina nanoparti- 2.1. Materials cles filled vinyl-ester resin nanocomposite fabrication [19]. A viscous liquid solution was still observed in the mold The polymeric matrix used was a vinyl ester resin, Der- after this curing process, indicating that nanofilling materi- akane momentum 411–350 (manufactured by the Dow als have a significant effect on the quality of fabricated Chemical Company), which is a mixture of 55 wt% vinyl nanocomposites. However, high-quality nanocomposites ester with an average molecule weight of 970 g/mol and were formed by room temperature curing for 24 h followed 45 wt% styrene monomers. Styrene with only one unsatu- by postcuring at 100 °C for 2 h. rated carbon–carbon double bond provides linear chain extension. Vinyl-ester monomers with two reactive vinyl 2.4. Characterization end groups enable the cross-linking for network formation. The liquid resin has a density of 1.045 g/cm3 and a viscosity A Fourier transform infrared (FT-IR) spectrometer was of 350 centipoises (cps) at room temperature. Iron oxide (c- used to test the physicochemical interaction between MPS Fe2O3, Nanophase Technologies) nanoparticles with an and Fe2O3 nanoparticles, and the change of the MPS func- average diameter of 23 nm and a specific surface area of tional groups after the nanoparticle modification. FT-IR 45 m2/g were functionalized and used as nanofillers for spectra were recorded in a FT-IR spectrometer (Jasco, the nanocomposite fabrication. Trigonox 239-A (curing FT-IR 420) in transmission mode under dried nitrogen Z. Guo et al. / Composites Science and Technology 68 (2008) 1513–1520 1515

flow (10 ccpm) condition. The liquid MPS dispersant was VSM Software was used to characterize the magnetism mixed with powder KBr, ground and compressed into a variation. pellet. Its spectrum was recorded as a reference to be compared with that of the MPS functionalized Fe2O3 nanopar- 3. Results and discussion ticles. The physicochemical interaction between nanoparticles and MPS was further characterized by Thermogravimetric analysis of the as-received iron oxide thermo-gravimetric analysis (TGA, Perkin–Elmer). TGA nanoparticles shows the existence of physicochemical mois- of the as-received and MPS-treated Fe2O3 nanoparticles ture either in the form of physical adsorption or in the was carried out from 25 to 600 °C with an argon flow rate chemically bonded hydroxyl groups in Fig. 1a, similar to of 50 ccpm and a heating rate of 10 °C/min. Thermal degradation of the nanocomposites with different nanoparticle loadings was studied by TGA with procedures similar to those used for the nanoparticles. The particle loading and functionalization effect on the curing process was investigated by the differential scanning calorimetry (DSC, Perkin–Elmer Instruments) under 10 ccpm nitrogen flow condition. The testing temperature was from 25 to 250 °C with a heating rate of 20 °C/min. The reaction enthalpy (J/g) and residual heat of reaction were measured from the area under the DSC peaks. Transmission electron microscope (TEM) was used to characterize the morphology (size and shape) of the as- received and MPS-treated nanoparticles. TEM was carried out in a JEM 100CX (JEOL) with an accelerating voltage of 100 keV. The samples were prepared by dispersing nanoparticles in ethanol, dropping some solution onto a holey carbon coated copper grid and drying the particles naturally in air. The dispersion quality of the nanoparticles in the vinyl-ester resin matrix was investigated by scanning electron microscopy (SEM) on the polished nanocomposite cross-sectional area. The SEM samples were carefully prepared as follows. The cured composite samples were polished with a 4000-grit sand paper and a following 50 nm alumina nanoparticle aqueous solution polishing to achieve a smooth surface, then washed with DI water, and followed by sputter coating a 3 nm gold. The mechanical properties were evaluated by tensile tests following the American Society for Testing and Mate- rials standard (ASTM, 2005, standard D 1708-02a). An Instron 4411 with Series IX software testing machine was used to measure the tensile strength and Young’s modulus. The dog-bone shaped specimens were prepared as described in the nanocomposite fabrication section. The specimen surfaces were smoothed with an abrasive sand paper (1000 grit) and the sanding strokes were made in the direction parallel to the long axis of the test specimen. A crosshead speed of 15 mm/min was used and strain (mm/ mm) was calculated by dividing the crosshead displacement (mm) by the gage length (mm). The magnetic properties were measured with a vibrating sample magnetometer (Model 1660) at room temperature. A 54.9 emu/g Ni standard sample was used to calibrate sample position and sensitivity. After calibration is done, Teflon tape was used to attach the sample onto the vibra- Fig. 1. (a) TGA of the as-received Fe2O3, pure MPS and MPS-treated Fe O nanoparticles; (b) FT-IR spectra of the as-received Fe O , pure tion holder and loaded into Electromagnet (model 3472- 2 3 2 3 MPS and MPS-treated Fe2O3 nanoparticles; TEM bright field micro- 70). Magnetic field was swept from À10,000 to 10,000 Oe structures of (c) as-received and (d) MPS-treated Fe2O3 nanoparticles; the with stepsize of 50 Oe. Digital Measurement System Easy inset shows the corresponding selected area electron diffraction. 1516 Z. Guo et al. / Composites Science and Technology 68 (2008) 1513–1520 the alumina, zinc oxide and copper oxide nanoparticles [19– 21]. Pure MPS is observed to evaporate at around 86 °C, dehydrate at temperature higher than 106 °C and com- pletely decompose at around 190 °C. However, bound onto the iron oxide nanoparticle surface, MPS exhibits a delayed complete decomposition temperature at about 345 °C. This observation indicates that there is a strong interaction between MPS and iron oxide nanoparticles, i.e., MPS is covalently bound onto the nanoparticle surface. In compar- ison with the as-received nanoparticles, MPS-treated iron oxide nanoparticles show an almost constant weight percentage at temperatures higher than 346 °C, indicating a complete loss of the hydroxyl groups (chemically adsorbed moisture) on the nanoparticle surface. Fig. 1b shows FT-IR spectra of as-received iron oxide nanoparticles, pure MPS, and MPS-functionalized iron oxide nanoparticles, respectively. The characteristic absorption peaks at 818 cmÀ1, À1 À1 1089 cm , and 1640 cm are due to –Si–OCH3, Si–O and C@C vibration of MPS, respectively. The peaks at 1720 cmÀ1 and 1167 cmÀ1 are due to the C@O and C–O vibration, respectively. MPS characteristic peaks, not observed in the non-modified iron oxide nanoparticles, were found in the MPS modified iron oxide nanoparticles. The disappearance of the peak at 818 cmÀ1 characteristic of – Si–OCH3 and the presence of other peaks characteristic of MPS in the MPS-treated-nanoparticles further indicate the successful functionalization of iron oxide nanoparticles with MPS. The characteristic stretching vibrations of iron oxide were observed in the range 450 and 800 cmÀ1 for the MPS-functionalized iron oxide nanoparticles. The peaks shifted as compared with those of the as-received nanoparticles, indicating a strong interaction between nanoparticles and MPS. Fig. 2. (a) TGA of liquid composite solution, room-temperature cured After the functionalization, the nanoparticles turned nanocomposites, and the nanocomposites with a 100 °C postcuring; (b) from dark brown to red, indicating the effect of the cou- DSC of pure resin and 15 wt% particle suspended liquid VE-AC pling agent. However, the crystal structure analysis did composite. not show any phase change as discussed in follows. Fig. 1c and d shows the bright-field TEM microstructures temperature for 24 h, and composites with a 2-h postcuring of the as-received and MPS-treated iron oxide nanoparti- at 100 °C, respectively. The particle loading in the compos- cles, respectively. The MPS treated nanoparticles are ites is 10 wt% and 25 wt%, respectively. Similar to our for- observed to be similar to the as-received nanoparticles. mer investigation on the pure liquid resin [21], the liquid The crystal phase of the as-received and functionalized nanocomposites suspended with as-received nanoparticles nanoparticles were investigated by selected area electron were observed to have solvent evaporation at temperatures diffraction (SAED) as shown in the inset of Fig. 1c and below 74 °C, heat-induced styrene evaporation at tempera- d, respectively. The crystalline structures of the as-received tures below 133 °C, and an initiating and full degradation and MPS-treated nanoparticle are similar with ring pat- temperature of 333 °C and 448 °C, respectively. The addi- terns. The calculated d-spacing of the diffraction rings tion of the as-received nanoparticles has little effect on (from the inside to the outward) is 1.63, 1.37, 1.11, 0.91 the thermal stability of the liquid vinyl-ester resin. The and 0.86 A˚ respectively, which can be assigned to observed plateau in the liquid composite at higher temper- {2,1,1}, {2,0,8}, {2,2,6}, {3,3,10} and {2,3,8} planes of atures was due to the released heat from the redox reaction iron oxide (Fe2O3, see the standard XRD file PDF#33- between Fe2O3 and coked carbon as observed in the CuO– 0664) rather than the other iron oxides such as FeO or vinyl ester nanocomposite [21]. However, dramatic differ- Fe3O4. This indicates that the particle color change after ence was observed in the liquid nanocomposites suspended functionalization was not due to the particle crystalline with MPS-functionalized nanoparticles, which is more sta- structure change. ble at temperatures lower than 350 °C. This indicates the Fig. 2a shows thermogravimetric curves of liquid com- functionalization process delays the evaporation of the sty- posite solution, cured pure resin, composites cured at room rene. After room temperature curing for 24 h, obvious Z. Guo et al. / Composites Science and Technology 68 (2008) 1513–1520 1517 weight loss of the styrene is still observed for the nanocom- increased after the incorporation of either as-received or posites filled with either the as-received or functionalized MPS-functionalized iron oxide nanoparticles as compared iron oxide nanoparticles, indicating the incomplete poly- to that of the pure resin. However, the initial curing tem- merization as expected. However, after postcuring at perature decreased after the incorporation of nanoparti- 100 °C for 2 h, the nanocomposites were observed to be cles. The functionalized nanoparticles lower the initial stable at temperatures lower than 313 °C and decompose curing temperature about 20 °C as compared to that of at temperatures higher than 320 °C. The resistance to ther- the as-received nanoparticles. The addition of as-received mal degradation was observed to be improved after the iron oxide nanoparticles has a much higher reaction heat nanoparticles were treated with MPS, indicating a strong 340.2 J/g (based on the pure resin) than that (280.0 J/g) interaction between the nanoparticles and the polymer of the pure resin. However, the functionalization lowers matrix. It was reported that metallic compounds could the reaction heat down to 253.6 J/g, which is much lower serve as catalysts to degrade the polymer matrix with an than that of the pure resin and will favor the engineering inferior thermal stability [22–25]. The presence of the as- application of the vinyl ester nanocomposite for marine received Fe2O3 nanoparticles in the nanocomposite may areas. facilitate the thermal degradation of the vinyl-ester resin Table 1 summarizes the DSC study on the curing of the by serving as a catalyst. However, the existence of MPS vinyl ester resin composites with different particle loadings. on the nanoparticle surface prevents the intimate contact The composites were prepared with different curing stages, of the iron oxide nanoparticles with the vinyl-ester resin i.e., curing temperatures. The initiating curing tempera- by passivating the particle surface and thus may improve tures were observed to increase after the room temperature the thermal stability of the nanocomposite. In addition, curing and to further increase after postcured at 100 °Cin the thermal stability was observed to be enhanced in the different particle-loading nanocomposite reinforced with 25 wt% loading as compared with the 10 wt%. A similar either the as-received or the MPS-treated nanoparticles. plateau arising from the released heat from exothermal The general trend is that the functionalization process low- reaction between iron oxide and carbon was also observed ers the initiating and peak curing temperatures. Similar to at high temperature, analogous to the TGA result of the that of the liquid composites, the reaction heat in the func- reduction of copper oxide with graphite by a mechanical tionalized nanoparticles reinforced nanocomposites at both alloying process [26]. the room temperature cured and postcured conditions is Park et al. [27] reported that the particle loadings have a much smaller than that of the as-received nanoparticles significant effect on the curing process for the particle size filled nanocomposites. The curing process behavior can in the nanosized scale. Here, the effect of the curing temper- be explained by considering the interaction between the ature and functionalization on the curing process was nanoparticles and the monomers. In the MPS functional- investigated using DSC. Fig. 2b shows the DSC curves of ized iron oxide nanoparticle suspended vinyl ester resin the pure resin, as-received and functionalized iron oxide liquid composites, the nanoparticles were coated with a nanoparticle suspended vinyl ester composites with a parti- layer of MPS, which prevents the direct contact between cle loading of 15 wt%. The peak curing temperature is the nanoparticles and vinyl ester monomers. Thus, the

Table 1 DSC Summary of the pure resin and nanocomposite at different process stage Condition Materials Particle loading Curing initiating Curing peak Complete heat of Degree of cure (wt%) temperature (°C) temperature (°C) reaction (J/g) (%) Liquid Pure resin 0 107 146 280.0 N/A As-received 10 92 156 303.4 N/A NPs 15 96 155 340.2 N/A 25 84 152 235.1 N/A Functionalized 10 48 131 227.6 N/A NPs 15 79 155 253.6 N/A 25 118 152 203.1 N/A RT cured As-received 10 105 148 168.7 44 NPs 15 102 148 181.1 47 25 102 150 215.0 8 Functionalized 10 45 131 76.2 66 NPs 15 95 148 170.5 33 25 121 148 158.2 22 Postcured As-received 10 139 171 30 90 100 °C NPs 15 118 159 26 93 25 162 189 17 93 Functionalized 10 N/A N/A 0 100 NPs 15 118 143 17 93 25 132 150 10 95 1518 Z. Guo et al. / Composites Science and Technology 68 (2008) 1513–1520 vinyl ester monomers in the composites will have greater the polymer matrix. However, in composites containing opportunity to interact with the catalyst and the polymer- the functionalized iron oxide nanoparticles shown in ization promoter. The extent of curing was observed to Fig. 3b, more uniformly dispersed nanoparticles are strongly depend on the particle loading. The higher the observed, indicating that nanoparticle surface functionali- particle loading, the lower the curing extent is at room tem- zation improved the nanoparticle dispersion in the polymer perature. In other words, the majority of the vinyl ester matrix. No observable void indicates a strong chemical monomers still keep liquid status even after 24 h curing interaction between the nanoparticles and polymer matrix. at room temperature. The vinyl ester resin in the compos- Thus, particle surface treatment causes the particle to dis- ites was determined to have more than 90% curing after perse more uniformly in the polymer matrix, and the a 2-h postcuring at 100 °C. formed chemical bonding between the particle and polymer In order to investigate the particle dispersion quality matrix favors a more compact solid structure without obvi- within the polymer matrix, the composites were polished ous gas gaps. to study their interior nanostructures. Fig. 3 shows the The functionalization effect on Young’s modulus (the SEM images of the cross-sectional area of the polished slope of the stress–strain curve in the low strain region, composite specimens with a 25 wt% particle loading. GPa) and tensile strength (the maximum stress in the Fig. 3a depicts the polished composites reinforced with stress–strain curve, MPa) were investigated by tensile tests. the as-received nanoparticles. The nanoparticles are Fig. 4 shows the typical tensile stress–strain curves of the observed to aggregate in the form of clusters at a micro- pure resin and vinyl ester resin nanocomposites with a metric scale with many voids (air gap) among the clusters, loading of 25 wt% as-received and MPS-functionalized indicating a poor adhesion between the nanoparticles and nanoparticles iron oxide nanoparticles. Young’s moduli were observed to increase significantly in the nanocomposites reinforced with both the as-received and MPS-treated iron oxide nanoparticles as compared with that of the cured pure resin. However, particle surface treatment has little effect on Young’s modulus. The tensile strength (65.5 MPa) of the nanocomposite reinforced with 25 wt% of the functionalized nanoparticles is much larger than that of the as-received nanoparticles filled nanocomposite (39.5 MPa), which is similar to our recent report on the alumina, copper oxide and zinc oxide nanoparticles reinforced vinyl ester nanocomposites [19–21] and attributed to the improved particle dispersion and enhanced interaction between the nanoparticles and polymer matrix. The micron-size clusters of agglomerated nanoparticles and the presence of voids in the as-received nanoparticles filled composites as shown in Fig. 3a introduce more concen-

Fig. 3. Micrographs of a cross-section of the iron oxide reinforced vinyl Fig. 4. Stress–strain curves of the cured pure resin and nanocomposites ester with a 25 wt% loading of (a) as-received and (b) MPS-functionalized reinforced with 25 wt% of the as-received and MPS-functionalized nanoparticles. nanoparticles. Z. Guo et al. / Composites Science and Technology 68 (2008) 1513–1520 1519 trated stresses on the interface, which result in a decrease in close contact of the pure iron oxide nanoparticles, and also the tensile strength. However, in the MPS-functionalized due to the polymer-particle interfacial effect [29]. iron oxide nanoparticles reinforced vinyl ester resin composites, well-dispersed nanoparticles as shown in Fig. 3b 4. Conclusion are strongly bound with the polymer matrix through the bridging effect of MPS between the nanoparticles and poly- The vinyl-ester resin thermosetting nanocomposites mer matrix. Therefore, the stress concentration is lower reinforced with iron oxide nanoparticles were prepared and the stresses can be more easily transferred from the and characterized. The optimum conditions for the iron matrix to the particles. The intimate contact between the oxide nanoparticles reinforced vinyl ester nanocomposites particles and the matrix also ensure a reduction of crack have been explored. Iron oxide nanoparticles functional- propagation. All the above features in the functionalized ized with a bi-functional coupling agent were found to nanoparticle filled composites definitely favor an increase favor the nanocomposite fabrication with a lower curing of the mechanical properties. temperature. The functionalization increases the adhesion Fig. 5 shows the magnetic hysteresis loops for the as- and the dispersion of the nanofiller into the matrix. The received iron oxide nanoparticles and the nanocomposite observed increased thermal stability and improved reinforced with 25 wt% of as-received and MPS-treated mechanical properties in the composites reinforced with iron oxide nanoparticles. The saturation magnetization of functionalized nanoparticles are closely attributed to the the as-received nanoparticles is about 68 emu/g, which is good nanoparticle dispersion in the polymer matrix and similar to that of the bulk iron oxide (74 emu/g). The par- to the introduced chemical bonding. The nanocomposites ticle loading estimated from the saturation magnetization become magnetically harder with a larger coercive force was to be 24.8 wt% and 22.5% for the MPS-treated and (coercivity) as compared to the lower coercivity of the bare as-received nanoparticles reinforced vinyl ester resin nano- superparamagnetic iron oxide nanoparticles, and are inde- composites, respectively. The coercivity (the magnitude of pendent of the nanoparticle functionalization. the external applied magnetic field necessary to return the material to a zero magnetization condition) was observed Acknowledgements to get enhanced from 17 to 212 Oe after the nanoparticles dispersed in the polymer matrix. In other words, the iron This paper is based on work partially supported by the oxide nanoparticles in the vinyl ester resin nanocomposites Air Force Office of Scientific Research through AFOSR become much harder after incorporation into a polymer Grant FA9550-05-1-0138 managed by Dr. B. Les Lee. Z. matrix. However, the nanoparticle surface functionaliza- Guo kindly acknowledges the support from the UC-Dis- tion has little effect on the magnetic properties, i.e., similar covery Grant ELE06-10268 and QuantumSphere research saturation magnetization and coercivity were observed in grant (QuantumSphere Inc.). the as-received and MPS-treated iron oxide nanoparticles reinforced vinyl ester resin nanocomposites. The big differ- References ence in the coercivity is due to the decreased interparticle dipolar interaction arising from the increased interparticle [1] Podlaha EJ, Li Y, Zhang J, Huang Q, Panda A, Lozano-Morales A, distance within the single-domain [28] as compared to the et al. Electrochemical deposition of nanostructured metals. In: Gogotsi Y, editor. Nanomaterials Handbook. CRC Press; 2006. p. 475–95. [2] Huang MH, Mao S, Feick H, Yan H, Wu Y, Kind H, et al. Room- temperature ultraviolet nanowire nanolasers. Science 2001;292:1897–9. [3] Bandara J, Weerasinghe HC. Enhancement of photovoltage of dye- sensitized solid-state solar cells by introducing high-band-gap oxide layers. Sol Energy Mater Solar Cells 2005;88:341–50. [4] Hosono E, Fujihara S, Honma I, Zhou H. The fabrication of an upright-standing zinc oxide nanosheet for use in dye-sensitized solar cells. Adv Mater 2005;17:2091–4. [5] Tang H, Yan M, Ma X, Zhang H, Wang M, Yang D. Gas sensing behavior of polyvinylpyrrolidone-modified ZnO nanoparticles for trimethylamine. Sens Actuat B 2006;113:324–8. [6] Huang GG, Wang CT, Tang HT, Huang YS, Yang J. ZnO nanoparticle-modified infrared internal reflection elements for selec- tive detection of volatile organic compounds. Anal Chem 2007;78:2397–404. [7] Kumar N, Dorfman A, Hahm JI. Ultasensitive DNA sequence detection using nanoscale ZnO sensor arrays. Nanotechnology 2006;17:2875–81. Fig. 5. Magnetic hysteresis loops of the as-received iron oxide nanopar- [8] Huang W, Han CD. Dispersion characteristics and rheology of ticles and nanocomposites with a 25 wt% loading of the as-received and organoclay nanocomposites based on a segmented main-chain liquid– MPS-treated nanoparticles. The inset shows the enlarged hysteresis loops crystalline polymer having pendent pyridyl group. Macromolecules at low applied magnetic fields. 2006;39:257–67. 1520 Z. Guo et al. / Composites Science and Technology 68 (2008) 1513–1520

[9] Chen Y, Sun L, Chiparus O, Negulescu I, Yachmenev V, Warnock [20] Guo Z, Wei S, Shedd B, Scaffaro R, Pereira T, Hahn HT. Particle M. Kenaf/ramie composite for automotive headliner. J Polym surface engineering effect on the mechanical, optical and photolumi- Environ 2005;13:107–14. nescent properties of ZnO/vinyl-ester resin nanocomposites. J Mater [10] Podlaha EJ, Landolt D. Pulse-reverse plating of nanocomposite thin Chem 2007;17:806–13. films. J Electrochem Soc 1997;144:L200–2. [21] Guo Z, Liang X, Pereira T, Scaffaro R, Hahn HT. CuO nanoparticle [11] Castro C, Ramos J, Millan A, Gonzalez-Calbet J, Palacio F. filled vinyl-ester resin nanocomposites: fabrication, characterization Production of magnetic nanoparticles in imine polymer matrixes. and property analysis. Compos Sci Technol 2007;67:2036–44. Chem Mater 2000;12:3681–8. [22] Rancourt JD, Taylor LT. Preparation and properties of surface- [12] Wetzel B, Haupert F, Zhang MQ. Epoxy nanocomposites with high conductive polyimide films via in situ codeposition of metal salts. mechanical and tribological performance. Comp Sci Technol Macromolecules 1987;20:790–5. 2003;63:2055–67. [23] Sawada T, Ando S. Synthesis, characterization, and optical properties [13] Guo Z, Park S, Wei S, Pereira T, Moldovan M, Karki AB, et al. of metal-containing fluorinated polyimide films. Chem Mater Flexible high-loading particle-reinforced polyurethane magnetic 1998;10:3368–78. nanocomposite fabrication through particle-surface-initiated poly- [24] Chiang PC, Whang WT. The synthesis and morphology characteristic

merization. Nanotechnology 2007;18:335704. study of BAO-ODPA polyimide/TiO2 nano hybrid films. Polymer [14] Kohli N, Worden RM, Lee I. Intact transfer of layered, bionano- 2003;44:2249–54. composite arrays by microcontact printing. Chem Commun [25] Hsu SC, Whang WT, Hung CH, Chiang PC, Hsiao YN. Effect of the 2005;3:316–8. polyimide structure and ZnO concentration on the morphology and [15] Judeinstein P, Sanchez C. Hybrid organic–inorganic materials: a land characteristics of polyimide/ZnO nanohybrid films. Macromol Chem of multidisciplinarity. J Mater Chem 1996;6:511–25. Phys 2005;206:291–8. [16] Zhang X, Simon LC. In situ polymerization of hybrid polyethylene– [26] Liu L, Zhang TJ, Cui K, Dong YD. Reduction of copper oxide with alumina nanocomposites. Macromol Mater Eng 2005;290:573–83. graphite by mechanical alloying. J Mater Res 1999;14:4062–9. [17] Gao SL, Mader E. Characterization of interphase nanoscale property [27] Park SS, Bernet N, Roche SDL, Hahn HT. Processing of iron variations in glass fibre reinforced polypropylene and epoxy resin oxide-epoxy vinyl ester nanocomposites. J Compos Mater composites. Compos A 2002;33:559–76. 2003;37:465–76. [18] Shenhar R, Norsten TB, Rotello VM. Polymer-mediated nanoparticle [28] Sorensen CM. Magnetic nanoscale materials in chemistry. In: assembly: structural control and applications. Adv Mater Klabunde KJ, editor. Nanoscale materials in chemistry. New York: 2005;17:657–69. Wiley-Interscience; 2001. p. 69–221. [19] Guo Z, Pereira T, Choi O, Wang Y, Hahn HT. Surface functionalized [29] Zhang D, Klabunde KJ, Sorensen CM, Hadjipanayis GC. Magne- alumina nanoparticle filled polymeric nanocomposites with enhanced tization temperature dependence in iron nanoparticles. Phys Rev B mechanical properties. J Mater Chem 2006;16:2800–8. 1998;58:14167–70.