- Email: [email protected]
acrylic resins mixtures
Polymer 53 (2012) 6039e6044
Contents lists available at SciVerse ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
In-situ graphene oxide reduction during UV-photopolymerization of graphene oxide/acrylic resins mixtures P. Fabbri a, e, *, L. Valentini b, e, S. Bittolo Bon b, e, D. Foix c, e, L. Pasquali a, d, M. Montecchi a, d, M. Sangermano c, e a
Department of Materials and Environmental Engineering, University of Modena and Reggio Emilia, Strada Vignolese 905/A, 41125 Modena, Italy Department of Civil and Environmental Engineering, University of Perugia, Strada di Pentima 4, 05100 Terni, Italy Department of Applied Science and Technology, Polytechnic of Turin, C.so Duca Degli Abruzzi 24, 10129 Torino, Italy d CNR e Istituto Officina dei Materiali, s.s. 14 Km 163, 5I-34149 Trieste, Italy e INSTM Local Units e Italian Consortium for Materials Science and Technology, Via G. Giusti 9, 20155 Firenze, Italy b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 July 2012 Received in revised form 11 October 2012 Accepted 23 October 2012 Available online 1 November 2012
The preparation of electrically conductive acrylic resins containing reduced graphene oxide (rGO) by photopolymerization is presented. The synthesis consists of a single-step procedure starting from a homogeneous water dispersion of GO, which undergoes reduction induced by the UV radiation during the photopolymerization of an acrylic resin. The role played by the amount of radical photoinitiator added to the resin has been evaluated in relation to the in-situ reduction of GO, that was monitored by Xray photoelectron spectroscopy. Results show that the UV-induced photopolymerization of acrylic resins with added GO gives rise to conductive acrylic composites thanks to the simultaneous reduction of GO to rGO and crosslinking of the resin. On this basis UV-induced photopolymerization is proposed as a sustainable strategy for the production of conductive graphene/polymer composites. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Graphene oxide UV-photopolymerization Acrylic resin
1. Introduction In the last few years the range of carbon-based polymer composites has been mainly dominated by carbon nanotubes (CNTs) due to the development of various functionalization techniques and polymer-grafting procedures that allowed the efficient dispersion of CNTs bundles in polymer matrices [1e3]. In order to find convenient solutions to the high costs and technical difficulties associated to the use of CNTs, the scientific interest recently moved towards different forms of graphitic carbon as fillers to prepare polymer composites, and graphene has rapidly become the most intensively investigated allotropic form of carbon due to its extraordinary electronic transport properties [4] associated with exceptional thermo-mechanical features, which can be successfully provided to polymer composites [5,6]. Numerous strategies for the preparation of homogeneously dispersed graphene-polymer composites have already been presented for a variety of advanced polymer matrices [7,8], but the literature unanimously agrees that
* Corresponding author. Department of Materials and Environmental Engineering, University of Modena and Reggio Emilia, Strada Vignolese 905/A, 41125 Modena, Italy. Tel.: þ39 059 2056220. E-mail address: [email protected] (P. Fabbri). 0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.polymer.2012.10.045
a significant amount of research efforts still has to be dedicated to the development of easy and cost-effective methodologies towards the large-scale application of graphene in functional polymercomposite materials for technological uses, as biosensors, energy storage, electronic devices. Unfortunately graphene sheets, similar to CNTs, present strong pep stacking between sp2 carbon layers and poor surface interactions with the polymer matrices, resulting in extended fillerefiller aggregation in composites. This major drawback to the use of graphene as functional filler for high performance composites has been addressed by developing strategies involving the oxidation of graphene to GO and its exfoliation by chemical process which generates epoxy, alcohol and carboxylic acid groups on its surface [9], followed by partial reduction to achieve rGO which is easily processable in both aqueous and organic media as single layer sheets [10,11]. The presence of hydrophilic organic functional groups facilitate exfoliation and dispersion of GO in aqueous solution, but the material shows insulating properties because of the numerous defects which disrupt the sp2 structure. To improve electrical performance subsequent reduction processes are necessary to restore the planar sp2 structure. Several types of reduction agents have been reported in literature such as hydrazine [12] or strong alkaline media [13]. Besides the reduction with chemical agents, electrochemical [14], photochemical [15] and thermal [16]
6040
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
reduction methods have been developed as well, and were demonstrated to be more sustainable reduction strategies with respect to the chemical approach. In this context photoreduction induced by interactions with radiation recently attracted the greater attention due to the observation that irradiation of GO in solid state or in solution by sunlight, UV or excimer laser radiation reduces it to graphene with a strongly reduced amount of oxygen functionalities onto the surface [17,18]; detailed investigation into the GO photoreduction process has just begun [19] and can be considered at an early state. It has been demonstrated that irradiation by excimer laser or by UV light with high energy per unit pulse can break the bonds between the graphene sheet and the oxygen functionalities in GO thus inducing a reduction of the oxygen related functionalities; the threshold of excitation light resulting in photolysis products corresponds to an energy of 3.2 eV and the conductivity of the GO layers is enhanced by exposure to UV light [20]. rGO produced by interactions with radiation has advantages over other techniques, mainly related to the lower level of impurities retained. We recently demonstrated that the mentioned strategy of in-situ reduction of GO to rGO by UV exposure can be efficiently exploited during the UV-induced photopolymerization of acrylic resins [21]. The observed electrical conductivity of the hybrid films obtained by photopolymerization of GO-acrylic resins mixtures is not consistent with the insulating behaviour of both GO flakes and acrylic resin. In this paper the reduction of GO oxygen functionalities to the graphene basal plane operated by the radicals generated during the photopolymerization of an acrylic resin has been demonstrated and quantitatively evaluated by spectral investigations performed by XPS spectroscopy. Finally, the properties of the cured rGO/acrylic resin mixtures have been extensively investigated by electrical characterization (IeV curves), morphological analysis (AFM) and thermal studies (DSC). 2. Experimental 2.1. Materials Single-layered GO made by modified Hummers method (Cheaptubes USA, average thickness 0.7e1.2 nm). The acrylic resin polyethyleneglycol diacrylate (PEGDA, Ebecryl11Ò, Mw z 740 g/ mol, density ¼ 1.12 g/cm3, CYTEC). The radical photoinitiator 4-(2hydroxyethoxy)phenyl-(2-hydroxy-2-propyl)ketone (PI, IrgacureÒ 2959, BASF). 2.2. Experiments of UV-induced photoreduction of GO GO was dispersed in water (1 mg/ml) by Ultraturrax mixing until a homogeneous dispersion was obtained (30,000 rpm for 5 min). PI was added in amounts ranging from 0 to 3% wt/vol and the solutions were UV-irradiated with a medium pressure mercury lamp under nitrogen with a light intensity on the surface of the sample of 30 mW/cm2 for different time up to 5 min. The irradiated GO was separated from the solutions by evaporation of water to be ready for further characterization.
2.3. Preparation of graphene-acrylic composites GO was dispersed in water (1 mg/ml) by Ultraturrax mixing (30,000 rpm for 5 min) and the water dispersion was mixed with PEGDA in order to add an actual content of GO in the range from 1 to 4% by weight with respect to the resin. The formulations were added to 4 %wt of PI, coated on glass slides or silicon wafers and UVcured, under nitrogen, with a light intensity on the surface of the sample of 30 mW/cm2 for different time up to 10 min. The cured GO/PEGDA composites were ready for further characterization without need of further actions. 2.4. Characterization Acrylic double bond conversion, as a function of irradiation time, was evaluated by means of FT-IR spectroscopy, employing a Thermo-Nicolet 5700 instrument. For this purpose, the GO/ PEGDA formulations were coated on silicon wafers (Olivetti, Italy) by using a wire wound applicator with a controlled thickness of 25 microns for the deposited wet film, and simultaneously exposed to the UV beam, which induces polymerization, and to the IR beam, which makes an in-situ evaluation of the extent of reaction. Acrylic double bonds conversion was monitored through the decrease in the absorbance peak centred at 1630 cm1, normalized with respect to the carbonyl peak centred at 1700 cm1. A medium pressure mercury lamp (Hamamatsu) equipped with an optical guide was used to induce photopolymerization. The irradiation was performed under nitrogen atmosphere. The gel content was determined on the cured composites by measuring the weight loss of a one-piece specimen (200 mg) after 24 h extraction in chloroform (10 ml) at room temperature (25 C). Differential scanning calorimetry (DSC) was performed under nitrogen flux, in the range between 20 C and 100 C, with a DSCQ 1000 of TA Instruments equipped with a low temperature probe. The electrical characterization was performed using a 2-probe method by computer-controlled Keithley 4200 Source Measure Unit. GO dispersions were drop cast on Al-electrodes (1 mm 10 mm) spaced 1 mm having an average thickness of w70 nm thermally evaporated on glass substrate. The current/ voltage characteristic was recorded through Al-electrodes thermally evaporated on the GO/PEGDA coated quartz substrates. A voltage sweep was sourced between the electrodes and the output current flowing across the sample was measured in the direction parallel to the surface of the sample; therefore the measured value of current flow takes into account both surface and volume resistivity contributions [22]. The morphology of the prepared samples was investigated by atomic force microscopy (AFM) and field emission scanning electron microscopy (FESEM, Zeiss Supra 25). AFM images were obtained operating in phase contrast tapping mode with a scanning probe microscope (Nanosurf easyScan DFM). Height and phase images were obtained under ambient conditions with a typical scan speed of 0.5e1 line/s, using a scan head with a maximum range of 40 mm 40 mm. The films surface was observed by FESEM operating at 5.0 kV accelerating voltage without deposition of any conductive coating; images were constructed from the detector of secondary electrons.
Fig. 1. a-Cleavage process of the photoinitiator aryl-alkyl ketone.
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
X-ray photoelectron spectroscopy (XPS) was taken on GO and rGO dried samples, deposited onto Si wafers (covered with their native oxide). Spectra were acquired with non-monochromatic MgKa photons (hn ¼ 1253.6 eV) from a Vacuum Generators XR3 dual anode source operated at 15 kV, 16 mA. Photoemission data were collected with a double pass Perkin Elmer PHI 15-255G cylindricalmirror electron analyzer. The analyser resolution was set to 0.5 eV. The spectra are reported as a function of the electron binding energy (BE), referenced to the Au 4f7/2 signal (84 eV) of a calibration sample. A background subtraction was applied to the spectra and replicas due to Mg-Ka satellites were removed [23].
6041
Table 1 Relative percentage of carbon components for rGO samples with different amount of PI before and after UV irradiation. Sample
P1 (%)
P2 (%)
P3 (%)
P4 (%)
PI0 t0 PI0 t50 PI1% t0 PI1% t50 PI3% t0 PI3% t50
53 53 55 68 43 70
32 32 27 18 33 20
15 15 18 10 21 6
e e e 4 3 4
3. Results and discussion 3.1. UV-induced photoreduction of GO Reduction of GO by UV light is usually carried out in the liquidphase using different types of reducing agents such as hydrazine [9], hydrogen [24] and metal oxides [25,26]. The effect of UV irradiation in the presence of an electron-donor species was shown to proceed through photochemical transformations involving
Fig. 2. Photoemission spectra from the C1s level taken on samples with different amount of PI (0%, 1%, 3%wt), before (t0) and after (t50 ) irradiation with UV light for 5 min.
Fig. 3. IeV curves for the films prepared with GO colloidal suspensions in water (1 mg/ ml) containing (a) 0, (b) 1 or (c) 3 %wt of PI before (t0) and after UV irradiation for 5 min (t5).
6042
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
reduction of GO to graphene by cleavage of the bonds between the graphene sheets and oxygen functionalities, mediated by UV light and radicals. The high electrical resistivity of GO is due to the existence of oxygen-containing groups, which could introduce defects to graphene. In contrast, deoxygenation could recover the GO conductivity to some extent. In this view, the presence of the radical photoinitiator for the curing of acrylic resins should play a significative role in the UV-induced photoreduction process of GO, due to the formation of radicals. In order to evaluate the effect of different contents of PI during the UV irradiation of stable GO colloidal suspensions in water, suspensions were added to increasing amounts of PI (namely 0, 1, 2 or 3 %wt) and irradiated for different time up to 5 min. The adopted photoinitiator belongs to the family of aryl-alkyl ketones, commonly used to promote the UV-curing of acrylic resins due to their high efficiency in radicals production upon light absorption with a maximum in the range 280e340 nm. PI works mostly through an a-cleavage process (Norrish type I) and exhibit a high intrinsic reactivity for initiating the polymerization of acrylic resins through radicals forming (Fig. 1). This has to be taken into consideration when UV irradiation of GO water dispersions occurs in the presence of this radical photoinitiator, i.e. during the UV-curing of GO/PEGDA mixtures, because its radicals should play a fundamental role in the photoinduced reduction of GO. The effects induced on GO by the UV treatment applied in the presence of amounts up to 3 %wt of radical PI in water solution were evaluated by XPS. Fig. 2 reports the C 1s levels of GO samples analysed before (t0) and after 5 min (t50 ) of UV irradiation applied in the presence of different amounts of PI (PI0%, PI1%, PI3% respectively for 0, 1 and 3 %wt of PI in water solution), without addition of PEGDA. The spectra were fitted with Voigt peaks, representing the contributions from different functional groups. Before irradiation (t0) the spectra present three main structures at 284.9 eV (P1), at 287.1 eV (P2) and at 288.5 eV (P3). A minor structure is observed at about 293 eV (P4) in the sample with 3%wt of PI. Structure P1 is mainly associated to carbons in sp2 configuration in C rings (CeC; contributions from CeH and CH3 groups may also be included), while P2, P3 and P4 structures are typical for oxidized C species on GO. According to the recent experimental-theoretical investigation by W. Zhang et al. [27], different types of functional species contribute to P2, being not possible to resolve them separately at the given experimental resolution: these include hydroxyl (CeOH) and carbonyl (C]O) groups, carbon atoms bonded to epoxy groups, oxidised edge groups. Feature P3 is assigned to carboxyl (COOH)
groups, epoxy pairs, epoxy-CeOH moieties. Feature P4 is finally assigned to traces of COOO species. By comparison of spectra recorded for samples PI0 t0 and PI0 t50 , corresponding to GO before and after UV irradiation in the complete absence of PI, it is possible to note that almost no changes occurred. On the contrary, spectra of GO irradiated in the presence of 1 %wt or 3%wt of PI show pronounced variations with respect to pristine GO. In fact, when irradiation occurs in the presence of PI, oxidised C species drastically decrease, indicating a considerable reduction of GO. Carbon atoms in the sp2 configuration give rise to the most prominent feature in GO subjected to UV irradiation in the presence of PI, and this prefigures a significative increase in its electrical conductivity. The effect is evident for samples PI1%, but it is further pronounced for PI3% sample. The relative amounts of the different carbon C1s components for the different samples before and after UV-photoreduction are reported in Table 1. After irradiation the C sp2 component (P1) approximately increases from 50% to 70% of the whole signal. In parallel, the oxygenated carbon components reduce to about one half of their amount before irradiation. The spectra clearly indicate the high efficiency of the UV-induced photoreduction of GO in the presence of PI. The changes in the chemical surface composition observed by XPS spectroscopy, and the increase in sp2 carbons following UV irradiation in the presence of PI, find a direct correlation in the increase in electrical conductivity observed for the rGO. Fig. 3 reports a significative relative change of the electric current generated after 5 min of UV irradiation for the sample PI3% t50 (Fig. 3c) compared to the almost unvaried electrical behaviour of samples cured with lower amounts of PI, namely PI1% t50 (Fig. 3b). Accordingly with L. Valentini et al. [28] the slight increase of the conductivity for the PI0% t50 sample (Fig. 3a) has to be ascribed to the decreasing of the electrical resistivity of UV irradiated graphene oxide. The FESEM characterization performed on films prepared with GO colloidal water suspensions containing 3 %wt of PI before (Fig. 4a) and after UV irradiation (Fig. 4b) demonstrates that, upon UV exposure, the graphene sheets roughness increases. In particular an area roughness change from 50 nm to 170 nm was estimated by AFM analysis for the not irradiated and irradiated samples, respectively. Corrugation effects have been previously observed for chemically converted graphene produced by reduction of graphene oxide [29]. Simulations suggest that different chemical addends and their arrangements result in significant
Fig. 4. FESEM images for the films prepared with GO colloidal suspensions in water (1 mg/ml) containing 3 %wt of PI before (a) and after UV irradiation for 5 min (b).
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
6043
Table 2 Properties of UV-cured graphene-acrylic composites. Sample PEGDA PEGDAþ1 PEGDAþ2 PEGDAþ3 PEGDAþ4 a b
Fig. 5. FT-IR spectra of PEGDA before and after 1 min of UV irradiation.
wrinkling and even bending of graphene sheets [30]. The removal of the oxygen-containing functional groups from the basal plane of graphene flakes upon UV irradiation therefore contributes to the change in its corrugation [31]. 3.2. UV-photocuring of graphene-acrylic composites On the basis of the results observed for the efficient UVphotoreduction of GO in the presence of a radical photoinitiator, a convenient one-step procedure for the fabrication of homogeneously dispersed, electrically conductive graphene-acrylic composites obtained by UV polymerization of PEGDA and simultaneous reduction of GO to rGO can be proposed. GO easily forms stable colloidal suspensions in water that can be efficiently exfoliated to produce single-layered GO by ultrasonic treatment. Since the direct dispersion of GO into an acrylic resin would induce immediate aggregation, addition of exfoliated aqueous GO dispersions to uncured PEGDA was pursued. An effective mixing of the monomer with the GO water dispersion was allowed due to the pronounced hydrophilic character of the
Fig. 6. FT-IR spectra of GO/PEGDA mixtures (4 %wt GO) before and after 1 min of UV irradiation.
%wt %wt %wt %wt
GO GO GO GO
Conv. FT-IR (%)a
Gel content (%)
Tg ( C)b
98 98 97 98 99
100 99 99 98 100
60 57 35 30 28
Determined after 1 min of irradiation by FT-IR analysis. Determined on UV-cured films by means of DSC analysis.
monomer, and its complete miscibility with water in the concentration range used for preparing the samples. In order to study the effect of the presence of GO water dispersion on the photopolymerization mechanism of PEGDA by means of FT-IR analysis, samples containing an actual content up to 4 %wt of GO with respect to the resin were prepared. GO/PEGDA mixtures were added to 4 %wt of PI as catalyst for PEGDA curing; this amount of PI, slightly higher than that (up to 3 %wt) used in the experiments of photoreduction of GO, was required to ensure the generation of radicals in proper amount for the double action of reduction of GO and curing of the diacrylate monomer. Interestingly, FT-IR monitoring performed in-situ during UVcuring demonstrated that the addition of GO water dispersion does not significantly influence the final acrylic double bond conversion of PEGDA. In Fig. 5 the FT-IR spectra before and after 1 min of UV irradiation of PEGDA are reported; the almost complete disappearance of the peak centred at 1630 cm1, is a clear indication of an almost complete conversion of the acrylic double bonds. In Fig. 6 analogous FT-IR spectra, before and after 1 min of UV irradiation, are reported for the GO/PEGDA mixture containing 4 % wt of GO. Even in this case an almost complete conversion of acrylic double bond is achieved within 1 min of UV irradiation. These data clearly show that the presence of GO water dispersion and GO reduction to rGO from the photogenerated radicals do not interfere with the curing process of PEGDA. This is an interesting achievement since, notwithstanding the possible shielding effect of the carbon nanoplatelets and the consumption of radicals for the reduction of GO, the photocuring kinetic and final acrylic double bond conversion are not affected, and it is possible to achieve fully cured rGO-PEGDA composites in a one-step procedure. GO/PEGDA mixtures were irradiated up to 10 min to reach a high crosslinking density, for ensuring elevated thermo-mechanical
Fig. 7. IeV curves of PEGDA containing 2, 3 and 4 %wt rGO after 10 min of UV irradiation.
6044
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
after 10 min of UV irradiation; the presence of partially exfoliated graphene platelets into the polymer matrix can be distinguished. 4. Conclusions An advantageous strategy for the preparation of electrically conductive polymer composites based on acrylic resins and graphene oxide has been presented. Investigations performed by X-rays photoelectron spectroscopy demonstrated that the in-situ reduction of GO is promoted by the radicals deriving from the cleavage of the photoinitiator used for the crosslinking of PEGDA resin; after 5 min of UV irradiation the C sp2 component of GO approximately increases from 50% to 70% of the whole C signal. In parallel, the oxygenated carbon components reduce to about one half of their amount before irradiation. In-situ monitoring of PEGDA photopolymerization by FTIR spectroscopy show that the presence of GO water dispersion and GO reduction to rGO from the photogenerated radicals do not interfere with the curing process of PEGDA; complete conversion of acrylic double bonds was reached for all rGO/PEGDA mixtures. IeV measurements demonstrated that the electrical percolation threshold for rGO/PEGDA composites is located between 3% and 4% weight of rGO. References Fig. 8. AFM picture (35 35 mm) of rGO/PEGDA (4 %wt rGO) after 10 min of UV irradiation. The arrows indicate the partially exfoliated graphene sheets.
properties of the acrylic composites. High gel contents (above 98%, see Table 2) were measured for all the cured films, as an indication of the formation of a tight crosslinked network and the absence of extractable monomers or oligomers in the cured system, in accordance with FT-IR monitoring. Differential scanning calorimetry (DSC) was performed on cured films in order to evaluate the glass transition temperature (Tg, Table 2) of the rGO/PEGDA composites with respect to pristine PEGDA, that shows its Tg at 60 C. An increase of Tg by increasing GO content into the UV-curable formulations results evident; this observation is consistent with a strong interfacial interaction between the polymer matrix and the rGO sheets, which hinder the mobility of the PEGDA network with a consequent increase of Tg. Finally, the electrical properties were measured on cured rGO/ PEGDA composites as a function of irradiation time. IeV curves reported in Fig. 7 show that UV-cured rGO/PEGDA composites are electrically conductive and the electrical percolation threshold is found between 3% and 4% by weight of rGO. To explain electrical behaviour of rGO/PEGDA composites two aspects have to be taken into consideration: the contribution of water to the surface conductivity when GO/water dispersion is added to the resin and UV induced reduction of GO. In previous investigations we demonstrated that the addition of water does not alter the conductivity of the neat polymer, since UV-photocuring of water/ PEGDA mixtures gives rise to insulating materials [21]; on the other hand, XPS investigations demonstrated that UV irradiation changes the GO conductivity rendering it conductive due to reduction of oxygen-containing functional groups on its surface. rGO distribution in the composite materials was investigated by phase contrast tapping mode AFM. This non-invasive technique has been used to qualitatively map the films surfaces evidencing the contrast between the hard graphene sheets and the soft polymer matrix. Fig. 8 shows the AFM picture of PEGDA containing 4 %wt GO
[1] Vaisman L, Marom G, Wagner HD. Adv Funct Mater 2006;16(3):357e63. [2] Liu YL, Chen WH. Macromolecules 2007;40(25):8881e6. [3] Liu YT, Zhao W, Huang ZY, Gao YF, Xie XM, Wang XH, et al. Carbon 2006; 44(8):1613e6. [4] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, et al. Science 2004;306:666e9. [5] Stankovich S, Dikin DA, Dommett GHB, Kohlhaas KM, Zimney EJ, Stach EA, et al. Nature 2006;442:282e6. [6] Kim H, Abdala AA, Macosko CW. Macromolecules 2010;43(16):6515e30. [7] Xu Z, Gao C. Macromolecules 2010;43(16):6716e23. [8] Fabbri P, Bassoli E, Bittolo Bon S, Valentini L. Polymer 2012;53(4):897e902. [9] Gao W, Alemany LB, Ci L, Ajayan PM. Nat Chem 2009;1(5):403e8. [10] Li D, Mueller MB, Gilje S, Kaner RB, Wallace GG. Nat Nanotechnol 2008;3(2): 101e5. [11] Paredes JI, Villar-Rodil S, Fernandez-Merino MJ, Guardia L, Martinez-Alonso A, Tascon JMD. J Mater Chem 2011;21(2):298e306. [12] Dreyer DR, Park S, Bielawski CW, Ruoff RS. Chem Soc Rev 2010;39:228e40. [13] Fan X, Peng W, Li Y, Li X, Wang S, Zhang G, et al. Adv Mater 2008;20(23): 4490e3. [14] Guo HL, Wang XF, Qian QY, Wang FB, Xia XH. ACS Nano 2009;3(9):2653e9. [15] Williams G, Seger B, Kamat PV. ACS Nano 2008;2(7):1487e91. [16] Lin Z, Yao Y, Li Z, Liu Y, Li Z, Wong CP. J Phys Chem C 2010;114(35):14819e25. [17] Kumar P, Das B, Chitara B, Subrahmanyam KS, Gopalakrishnan K, Krupanidhi SB, et al. Macromol Chem Phys http://dx.doi.org/10.1002/macp.201100451. [18] Guardia L, Villar-Rodil S, Paredes JI, Rozada R, Martinez-Alonso A, Tascon JMD. Carbon 2012;50(3):1014e24. [19] Plotnikov VG, Smirnov VA, Alfimov MV, Shulga YM. High Energ Chem 2011; 45(5):411e5. [20] Shulga YM, Martynenko VM, Muradyan VE, Baskakov SA, Smirnov VA, Gutsev GL. Chem Phys Lett 2010;498:287e91. [21] Sangermano M, Marchi S, Valentini L, Bittolo Bon S, Fabbri P. Macromol Mater Eng 2011;296:401e7. [22] Stauffer D, Aharony A. Introduction to percolation theory. London: Taylor & Francis; 1992. [23] Tougaard S. Surf Sci 1989;216:343e60. [24] Matsumoto Y, Koinuma M, Kim SY, Watanabe Y, Taniguchi T, Hatakeyama K, et al. ACS Appl Mater Inter 2010;2(12):3461e6. [25] Akhavan O. Carbon 2011;49(1):11e8. [26] Liu X, Pan L, Zhao Q, Lv T, Zhu G, Chen T, et al. Chem Eng J 2012;183:238e43. [27] Zhang W, Carravetta V, Li Z, Luo Y, Yang J. J Chem Phys 2009;131:244505/1e 244505/6. [28] Bittolo Bon S, Piccinini M, Mariani A, Kenny JM, Valentini L. Diam Relat Mater 2011;20:871e4. [29] Wang S, Zhang Y, Abidi N, Cabrales L. Langmuir 2009;25:11078e81. [30] Schniepp HC, Li JL, McAllister MJ, Sai H, Herrera-Alonso M, Adamson DH, et al. J Phys Chem B 2006;110:8535e9. [31] Paci JT, Belytschko T, Schatz GC. J Phys Chem C 2007;111:18099e111.
Contents lists available at SciVerse ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
In-situ graphene oxide reduction during UV-photopolymerization of graphene oxide/acrylic resins mixtures P. Fabbri a, e, *, L. Valentini b, e, S. Bittolo Bon b, e, D. Foix c, e, L. Pasquali a, d, M. Montecchi a, d, M. Sangermano c, e a
Department of Materials and Environmental Engineering, University of Modena and Reggio Emilia, Strada Vignolese 905/A, 41125 Modena, Italy Department of Civil and Environmental Engineering, University of Perugia, Strada di Pentima 4, 05100 Terni, Italy Department of Applied Science and Technology, Polytechnic of Turin, C.so Duca Degli Abruzzi 24, 10129 Torino, Italy d CNR e Istituto Officina dei Materiali, s.s. 14 Km 163, 5I-34149 Trieste, Italy e INSTM Local Units e Italian Consortium for Materials Science and Technology, Via G. Giusti 9, 20155 Firenze, Italy b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 July 2012 Received in revised form 11 October 2012 Accepted 23 October 2012 Available online 1 November 2012
The preparation of electrically conductive acrylic resins containing reduced graphene oxide (rGO) by photopolymerization is presented. The synthesis consists of a single-step procedure starting from a homogeneous water dispersion of GO, which undergoes reduction induced by the UV radiation during the photopolymerization of an acrylic resin. The role played by the amount of radical photoinitiator added to the resin has been evaluated in relation to the in-situ reduction of GO, that was monitored by Xray photoelectron spectroscopy. Results show that the UV-induced photopolymerization of acrylic resins with added GO gives rise to conductive acrylic composites thanks to the simultaneous reduction of GO to rGO and crosslinking of the resin. On this basis UV-induced photopolymerization is proposed as a sustainable strategy for the production of conductive graphene/polymer composites. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Graphene oxide UV-photopolymerization Acrylic resin
1. Introduction In the last few years the range of carbon-based polymer composites has been mainly dominated by carbon nanotubes (CNTs) due to the development of various functionalization techniques and polymer-grafting procedures that allowed the efficient dispersion of CNTs bundles in polymer matrices [1e3]. In order to find convenient solutions to the high costs and technical difficulties associated to the use of CNTs, the scientific interest recently moved towards different forms of graphitic carbon as fillers to prepare polymer composites, and graphene has rapidly become the most intensively investigated allotropic form of carbon due to its extraordinary electronic transport properties [4] associated with exceptional thermo-mechanical features, which can be successfully provided to polymer composites [5,6]. Numerous strategies for the preparation of homogeneously dispersed graphene-polymer composites have already been presented for a variety of advanced polymer matrices [7,8], but the literature unanimously agrees that
* Corresponding author. Department of Materials and Environmental Engineering, University of Modena and Reggio Emilia, Strada Vignolese 905/A, 41125 Modena, Italy. Tel.: þ39 059 2056220. E-mail address: [email protected] (P. Fabbri). 0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.polymer.2012.10.045
a significant amount of research efforts still has to be dedicated to the development of easy and cost-effective methodologies towards the large-scale application of graphene in functional polymercomposite materials for technological uses, as biosensors, energy storage, electronic devices. Unfortunately graphene sheets, similar to CNTs, present strong pep stacking between sp2 carbon layers and poor surface interactions with the polymer matrices, resulting in extended fillerefiller aggregation in composites. This major drawback to the use of graphene as functional filler for high performance composites has been addressed by developing strategies involving the oxidation of graphene to GO and its exfoliation by chemical process which generates epoxy, alcohol and carboxylic acid groups on its surface [9], followed by partial reduction to achieve rGO which is easily processable in both aqueous and organic media as single layer sheets [10,11]. The presence of hydrophilic organic functional groups facilitate exfoliation and dispersion of GO in aqueous solution, but the material shows insulating properties because of the numerous defects which disrupt the sp2 structure. To improve electrical performance subsequent reduction processes are necessary to restore the planar sp2 structure. Several types of reduction agents have been reported in literature such as hydrazine [12] or strong alkaline media [13]. Besides the reduction with chemical agents, electrochemical [14], photochemical [15] and thermal [16]
6040
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
reduction methods have been developed as well, and were demonstrated to be more sustainable reduction strategies with respect to the chemical approach. In this context photoreduction induced by interactions with radiation recently attracted the greater attention due to the observation that irradiation of GO in solid state or in solution by sunlight, UV or excimer laser radiation reduces it to graphene with a strongly reduced amount of oxygen functionalities onto the surface [17,18]; detailed investigation into the GO photoreduction process has just begun [19] and can be considered at an early state. It has been demonstrated that irradiation by excimer laser or by UV light with high energy per unit pulse can break the bonds between the graphene sheet and the oxygen functionalities in GO thus inducing a reduction of the oxygen related functionalities; the threshold of excitation light resulting in photolysis products corresponds to an energy of 3.2 eV and the conductivity of the GO layers is enhanced by exposure to UV light [20]. rGO produced by interactions with radiation has advantages over other techniques, mainly related to the lower level of impurities retained. We recently demonstrated that the mentioned strategy of in-situ reduction of GO to rGO by UV exposure can be efficiently exploited during the UV-induced photopolymerization of acrylic resins [21]. The observed electrical conductivity of the hybrid films obtained by photopolymerization of GO-acrylic resins mixtures is not consistent with the insulating behaviour of both GO flakes and acrylic resin. In this paper the reduction of GO oxygen functionalities to the graphene basal plane operated by the radicals generated during the photopolymerization of an acrylic resin has been demonstrated and quantitatively evaluated by spectral investigations performed by XPS spectroscopy. Finally, the properties of the cured rGO/acrylic resin mixtures have been extensively investigated by electrical characterization (IeV curves), morphological analysis (AFM) and thermal studies (DSC). 2. Experimental 2.1. Materials Single-layered GO made by modified Hummers method (Cheaptubes USA, average thickness 0.7e1.2 nm). The acrylic resin polyethyleneglycol diacrylate (PEGDA, Ebecryl11Ò, Mw z 740 g/ mol, density ¼ 1.12 g/cm3, CYTEC). The radical photoinitiator 4-(2hydroxyethoxy)phenyl-(2-hydroxy-2-propyl)ketone (PI, IrgacureÒ 2959, BASF). 2.2. Experiments of UV-induced photoreduction of GO GO was dispersed in water (1 mg/ml) by Ultraturrax mixing until a homogeneous dispersion was obtained (30,000 rpm for 5 min). PI was added in amounts ranging from 0 to 3% wt/vol and the solutions were UV-irradiated with a medium pressure mercury lamp under nitrogen with a light intensity on the surface of the sample of 30 mW/cm2 for different time up to 5 min. The irradiated GO was separated from the solutions by evaporation of water to be ready for further characterization.
2.3. Preparation of graphene-acrylic composites GO was dispersed in water (1 mg/ml) by Ultraturrax mixing (30,000 rpm for 5 min) and the water dispersion was mixed with PEGDA in order to add an actual content of GO in the range from 1 to 4% by weight with respect to the resin. The formulations were added to 4 %wt of PI, coated on glass slides or silicon wafers and UVcured, under nitrogen, with a light intensity on the surface of the sample of 30 mW/cm2 for different time up to 10 min. The cured GO/PEGDA composites were ready for further characterization without need of further actions. 2.4. Characterization Acrylic double bond conversion, as a function of irradiation time, was evaluated by means of FT-IR spectroscopy, employing a Thermo-Nicolet 5700 instrument. For this purpose, the GO/ PEGDA formulations were coated on silicon wafers (Olivetti, Italy) by using a wire wound applicator with a controlled thickness of 25 microns for the deposited wet film, and simultaneously exposed to the UV beam, which induces polymerization, and to the IR beam, which makes an in-situ evaluation of the extent of reaction. Acrylic double bonds conversion was monitored through the decrease in the absorbance peak centred at 1630 cm1, normalized with respect to the carbonyl peak centred at 1700 cm1. A medium pressure mercury lamp (Hamamatsu) equipped with an optical guide was used to induce photopolymerization. The irradiation was performed under nitrogen atmosphere. The gel content was determined on the cured composites by measuring the weight loss of a one-piece specimen (200 mg) after 24 h extraction in chloroform (10 ml) at room temperature (25 C). Differential scanning calorimetry (DSC) was performed under nitrogen flux, in the range between 20 C and 100 C, with a DSCQ 1000 of TA Instruments equipped with a low temperature probe. The electrical characterization was performed using a 2-probe method by computer-controlled Keithley 4200 Source Measure Unit. GO dispersions were drop cast on Al-electrodes (1 mm 10 mm) spaced 1 mm having an average thickness of w70 nm thermally evaporated on glass substrate. The current/ voltage characteristic was recorded through Al-electrodes thermally evaporated on the GO/PEGDA coated quartz substrates. A voltage sweep was sourced between the electrodes and the output current flowing across the sample was measured in the direction parallel to the surface of the sample; therefore the measured value of current flow takes into account both surface and volume resistivity contributions [22]. The morphology of the prepared samples was investigated by atomic force microscopy (AFM) and field emission scanning electron microscopy (FESEM, Zeiss Supra 25). AFM images were obtained operating in phase contrast tapping mode with a scanning probe microscope (Nanosurf easyScan DFM). Height and phase images were obtained under ambient conditions with a typical scan speed of 0.5e1 line/s, using a scan head with a maximum range of 40 mm 40 mm. The films surface was observed by FESEM operating at 5.0 kV accelerating voltage without deposition of any conductive coating; images were constructed from the detector of secondary electrons.
Fig. 1. a-Cleavage process of the photoinitiator aryl-alkyl ketone.
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
X-ray photoelectron spectroscopy (XPS) was taken on GO and rGO dried samples, deposited onto Si wafers (covered with their native oxide). Spectra were acquired with non-monochromatic MgKa photons (hn ¼ 1253.6 eV) from a Vacuum Generators XR3 dual anode source operated at 15 kV, 16 mA. Photoemission data were collected with a double pass Perkin Elmer PHI 15-255G cylindricalmirror electron analyzer. The analyser resolution was set to 0.5 eV. The spectra are reported as a function of the electron binding energy (BE), referenced to the Au 4f7/2 signal (84 eV) of a calibration sample. A background subtraction was applied to the spectra and replicas due to Mg-Ka satellites were removed [23].
6041
Table 1 Relative percentage of carbon components for rGO samples with different amount of PI before and after UV irradiation. Sample
P1 (%)
P2 (%)
P3 (%)
P4 (%)
PI0 t0 PI0 t50 PI1% t0 PI1% t50 PI3% t0 PI3% t50
53 53 55 68 43 70
32 32 27 18 33 20
15 15 18 10 21 6
e e e 4 3 4
3. Results and discussion 3.1. UV-induced photoreduction of GO Reduction of GO by UV light is usually carried out in the liquidphase using different types of reducing agents such as hydrazine [9], hydrogen [24] and metal oxides [25,26]. The effect of UV irradiation in the presence of an electron-donor species was shown to proceed through photochemical transformations involving
Fig. 2. Photoemission spectra from the C1s level taken on samples with different amount of PI (0%, 1%, 3%wt), before (t0) and after (t50 ) irradiation with UV light for 5 min.
Fig. 3. IeV curves for the films prepared with GO colloidal suspensions in water (1 mg/ ml) containing (a) 0, (b) 1 or (c) 3 %wt of PI before (t0) and after UV irradiation for 5 min (t5).
6042
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
reduction of GO to graphene by cleavage of the bonds between the graphene sheets and oxygen functionalities, mediated by UV light and radicals. The high electrical resistivity of GO is due to the existence of oxygen-containing groups, which could introduce defects to graphene. In contrast, deoxygenation could recover the GO conductivity to some extent. In this view, the presence of the radical photoinitiator for the curing of acrylic resins should play a significative role in the UV-induced photoreduction process of GO, due to the formation of radicals. In order to evaluate the effect of different contents of PI during the UV irradiation of stable GO colloidal suspensions in water, suspensions were added to increasing amounts of PI (namely 0, 1, 2 or 3 %wt) and irradiated for different time up to 5 min. The adopted photoinitiator belongs to the family of aryl-alkyl ketones, commonly used to promote the UV-curing of acrylic resins due to their high efficiency in radicals production upon light absorption with a maximum in the range 280e340 nm. PI works mostly through an a-cleavage process (Norrish type I) and exhibit a high intrinsic reactivity for initiating the polymerization of acrylic resins through radicals forming (Fig. 1). This has to be taken into consideration when UV irradiation of GO water dispersions occurs in the presence of this radical photoinitiator, i.e. during the UV-curing of GO/PEGDA mixtures, because its radicals should play a fundamental role in the photoinduced reduction of GO. The effects induced on GO by the UV treatment applied in the presence of amounts up to 3 %wt of radical PI in water solution were evaluated by XPS. Fig. 2 reports the C 1s levels of GO samples analysed before (t0) and after 5 min (t50 ) of UV irradiation applied in the presence of different amounts of PI (PI0%, PI1%, PI3% respectively for 0, 1 and 3 %wt of PI in water solution), without addition of PEGDA. The spectra were fitted with Voigt peaks, representing the contributions from different functional groups. Before irradiation (t0) the spectra present three main structures at 284.9 eV (P1), at 287.1 eV (P2) and at 288.5 eV (P3). A minor structure is observed at about 293 eV (P4) in the sample with 3%wt of PI. Structure P1 is mainly associated to carbons in sp2 configuration in C rings (CeC; contributions from CeH and CH3 groups may also be included), while P2, P3 and P4 structures are typical for oxidized C species on GO. According to the recent experimental-theoretical investigation by W. Zhang et al. [27], different types of functional species contribute to P2, being not possible to resolve them separately at the given experimental resolution: these include hydroxyl (CeOH) and carbonyl (C]O) groups, carbon atoms bonded to epoxy groups, oxidised edge groups. Feature P3 is assigned to carboxyl (COOH)
groups, epoxy pairs, epoxy-CeOH moieties. Feature P4 is finally assigned to traces of COOO species. By comparison of spectra recorded for samples PI0 t0 and PI0 t50 , corresponding to GO before and after UV irradiation in the complete absence of PI, it is possible to note that almost no changes occurred. On the contrary, spectra of GO irradiated in the presence of 1 %wt or 3%wt of PI show pronounced variations with respect to pristine GO. In fact, when irradiation occurs in the presence of PI, oxidised C species drastically decrease, indicating a considerable reduction of GO. Carbon atoms in the sp2 configuration give rise to the most prominent feature in GO subjected to UV irradiation in the presence of PI, and this prefigures a significative increase in its electrical conductivity. The effect is evident for samples PI1%, but it is further pronounced for PI3% sample. The relative amounts of the different carbon C1s components for the different samples before and after UV-photoreduction are reported in Table 1. After irradiation the C sp2 component (P1) approximately increases from 50% to 70% of the whole signal. In parallel, the oxygenated carbon components reduce to about one half of their amount before irradiation. The spectra clearly indicate the high efficiency of the UV-induced photoreduction of GO in the presence of PI. The changes in the chemical surface composition observed by XPS spectroscopy, and the increase in sp2 carbons following UV irradiation in the presence of PI, find a direct correlation in the increase in electrical conductivity observed for the rGO. Fig. 3 reports a significative relative change of the electric current generated after 5 min of UV irradiation for the sample PI3% t50 (Fig. 3c) compared to the almost unvaried electrical behaviour of samples cured with lower amounts of PI, namely PI1% t50 (Fig. 3b). Accordingly with L. Valentini et al. [28] the slight increase of the conductivity for the PI0% t50 sample (Fig. 3a) has to be ascribed to the decreasing of the electrical resistivity of UV irradiated graphene oxide. The FESEM characterization performed on films prepared with GO colloidal water suspensions containing 3 %wt of PI before (Fig. 4a) and after UV irradiation (Fig. 4b) demonstrates that, upon UV exposure, the graphene sheets roughness increases. In particular an area roughness change from 50 nm to 170 nm was estimated by AFM analysis for the not irradiated and irradiated samples, respectively. Corrugation effects have been previously observed for chemically converted graphene produced by reduction of graphene oxide [29]. Simulations suggest that different chemical addends and their arrangements result in significant
Fig. 4. FESEM images for the films prepared with GO colloidal suspensions in water (1 mg/ml) containing 3 %wt of PI before (a) and after UV irradiation for 5 min (b).
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
6043
Table 2 Properties of UV-cured graphene-acrylic composites. Sample PEGDA PEGDAþ1 PEGDAþ2 PEGDAþ3 PEGDAþ4 a b
Fig. 5. FT-IR spectra of PEGDA before and after 1 min of UV irradiation.
wrinkling and even bending of graphene sheets [30]. The removal of the oxygen-containing functional groups from the basal plane of graphene flakes upon UV irradiation therefore contributes to the change in its corrugation [31]. 3.2. UV-photocuring of graphene-acrylic composites On the basis of the results observed for the efficient UVphotoreduction of GO in the presence of a radical photoinitiator, a convenient one-step procedure for the fabrication of homogeneously dispersed, electrically conductive graphene-acrylic composites obtained by UV polymerization of PEGDA and simultaneous reduction of GO to rGO can be proposed. GO easily forms stable colloidal suspensions in water that can be efficiently exfoliated to produce single-layered GO by ultrasonic treatment. Since the direct dispersion of GO into an acrylic resin would induce immediate aggregation, addition of exfoliated aqueous GO dispersions to uncured PEGDA was pursued. An effective mixing of the monomer with the GO water dispersion was allowed due to the pronounced hydrophilic character of the
Fig. 6. FT-IR spectra of GO/PEGDA mixtures (4 %wt GO) before and after 1 min of UV irradiation.
%wt %wt %wt %wt
GO GO GO GO
Conv. FT-IR (%)a
Gel content (%)
Tg ( C)b
98 98 97 98 99
100 99 99 98 100
60 57 35 30 28
Determined after 1 min of irradiation by FT-IR analysis. Determined on UV-cured films by means of DSC analysis.
monomer, and its complete miscibility with water in the concentration range used for preparing the samples. In order to study the effect of the presence of GO water dispersion on the photopolymerization mechanism of PEGDA by means of FT-IR analysis, samples containing an actual content up to 4 %wt of GO with respect to the resin were prepared. GO/PEGDA mixtures were added to 4 %wt of PI as catalyst for PEGDA curing; this amount of PI, slightly higher than that (up to 3 %wt) used in the experiments of photoreduction of GO, was required to ensure the generation of radicals in proper amount for the double action of reduction of GO and curing of the diacrylate monomer. Interestingly, FT-IR monitoring performed in-situ during UVcuring demonstrated that the addition of GO water dispersion does not significantly influence the final acrylic double bond conversion of PEGDA. In Fig. 5 the FT-IR spectra before and after 1 min of UV irradiation of PEGDA are reported; the almost complete disappearance of the peak centred at 1630 cm1, is a clear indication of an almost complete conversion of the acrylic double bonds. In Fig. 6 analogous FT-IR spectra, before and after 1 min of UV irradiation, are reported for the GO/PEGDA mixture containing 4 % wt of GO. Even in this case an almost complete conversion of acrylic double bond is achieved within 1 min of UV irradiation. These data clearly show that the presence of GO water dispersion and GO reduction to rGO from the photogenerated radicals do not interfere with the curing process of PEGDA. This is an interesting achievement since, notwithstanding the possible shielding effect of the carbon nanoplatelets and the consumption of radicals for the reduction of GO, the photocuring kinetic and final acrylic double bond conversion are not affected, and it is possible to achieve fully cured rGO-PEGDA composites in a one-step procedure. GO/PEGDA mixtures were irradiated up to 10 min to reach a high crosslinking density, for ensuring elevated thermo-mechanical
Fig. 7. IeV curves of PEGDA containing 2, 3 and 4 %wt rGO after 10 min of UV irradiation.
6044
P. Fabbri et al. / Polymer 53 (2012) 6039e6044
after 10 min of UV irradiation; the presence of partially exfoliated graphene platelets into the polymer matrix can be distinguished. 4. Conclusions An advantageous strategy for the preparation of electrically conductive polymer composites based on acrylic resins and graphene oxide has been presented. Investigations performed by X-rays photoelectron spectroscopy demonstrated that the in-situ reduction of GO is promoted by the radicals deriving from the cleavage of the photoinitiator used for the crosslinking of PEGDA resin; after 5 min of UV irradiation the C sp2 component of GO approximately increases from 50% to 70% of the whole C signal. In parallel, the oxygenated carbon components reduce to about one half of their amount before irradiation. In-situ monitoring of PEGDA photopolymerization by FTIR spectroscopy show that the presence of GO water dispersion and GO reduction to rGO from the photogenerated radicals do not interfere with the curing process of PEGDA; complete conversion of acrylic double bonds was reached for all rGO/PEGDA mixtures. IeV measurements demonstrated that the electrical percolation threshold for rGO/PEGDA composites is located between 3% and 4% weight of rGO. References Fig. 8. AFM picture (35 35 mm) of rGO/PEGDA (4 %wt rGO) after 10 min of UV irradiation. The arrows indicate the partially exfoliated graphene sheets.
properties of the acrylic composites. High gel contents (above 98%, see Table 2) were measured for all the cured films, as an indication of the formation of a tight crosslinked network and the absence of extractable monomers or oligomers in the cured system, in accordance with FT-IR monitoring. Differential scanning calorimetry (DSC) was performed on cured films in order to evaluate the glass transition temperature (Tg, Table 2) of the rGO/PEGDA composites with respect to pristine PEGDA, that shows its Tg at 60 C. An increase of Tg by increasing GO content into the UV-curable formulations results evident; this observation is consistent with a strong interfacial interaction between the polymer matrix and the rGO sheets, which hinder the mobility of the PEGDA network with a consequent increase of Tg. Finally, the electrical properties were measured on cured rGO/ PEGDA composites as a function of irradiation time. IeV curves reported in Fig. 7 show that UV-cured rGO/PEGDA composites are electrically conductive and the electrical percolation threshold is found between 3% and 4% by weight of rGO. To explain electrical behaviour of rGO/PEGDA composites two aspects have to be taken into consideration: the contribution of water to the surface conductivity when GO/water dispersion is added to the resin and UV induced reduction of GO. In previous investigations we demonstrated that the addition of water does not alter the conductivity of the neat polymer, since UV-photocuring of water/ PEGDA mixtures gives rise to insulating materials [21]; on the other hand, XPS investigations demonstrated that UV irradiation changes the GO conductivity rendering it conductive due to reduction of oxygen-containing functional groups on its surface. rGO distribution in the composite materials was investigated by phase contrast tapping mode AFM. This non-invasive technique has been used to qualitatively map the films surfaces evidencing the contrast between the hard graphene sheets and the soft polymer matrix. Fig. 8 shows the AFM picture of PEGDA containing 4 %wt GO
[1] Vaisman L, Marom G, Wagner HD. Adv Funct Mater 2006;16(3):357e63. [2] Liu YL, Chen WH. Macromolecules 2007;40(25):8881e6. [3] Liu YT, Zhao W, Huang ZY, Gao YF, Xie XM, Wang XH, et al. Carbon 2006; 44(8):1613e6. [4] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, et al. Science 2004;306:666e9. [5] Stankovich S, Dikin DA, Dommett GHB, Kohlhaas KM, Zimney EJ, Stach EA, et al. Nature 2006;442:282e6. [6] Kim H, Abdala AA, Macosko CW. Macromolecules 2010;43(16):6515e30. [7] Xu Z, Gao C. Macromolecules 2010;43(16):6716e23. [8] Fabbri P, Bassoli E, Bittolo Bon S, Valentini L. Polymer 2012;53(4):897e902. [9] Gao W, Alemany LB, Ci L, Ajayan PM. Nat Chem 2009;1(5):403e8. [10] Li D, Mueller MB, Gilje S, Kaner RB, Wallace GG. Nat Nanotechnol 2008;3(2): 101e5. [11] Paredes JI, Villar-Rodil S, Fernandez-Merino MJ, Guardia L, Martinez-Alonso A, Tascon JMD. J Mater Chem 2011;21(2):298e306. [12] Dreyer DR, Park S, Bielawski CW, Ruoff RS. Chem Soc Rev 2010;39:228e40. [13] Fan X, Peng W, Li Y, Li X, Wang S, Zhang G, et al. Adv Mater 2008;20(23): 4490e3. [14] Guo HL, Wang XF, Qian QY, Wang FB, Xia XH. ACS Nano 2009;3(9):2653e9. [15] Williams G, Seger B, Kamat PV. ACS Nano 2008;2(7):1487e91. [16] Lin Z, Yao Y, Li Z, Liu Y, Li Z, Wong CP. J Phys Chem C 2010;114(35):14819e25. [17] Kumar P, Das B, Chitara B, Subrahmanyam KS, Gopalakrishnan K, Krupanidhi SB, et al. Macromol Chem Phys http://dx.doi.org/10.1002/macp.201100451. [18] Guardia L, Villar-Rodil S, Paredes JI, Rozada R, Martinez-Alonso A, Tascon JMD. Carbon 2012;50(3):1014e24. [19] Plotnikov VG, Smirnov VA, Alfimov MV, Shulga YM. High Energ Chem 2011; 45(5):411e5. [20] Shulga YM, Martynenko VM, Muradyan VE, Baskakov SA, Smirnov VA, Gutsev GL. Chem Phys Lett 2010;498:287e91. [21] Sangermano M, Marchi S, Valentini L, Bittolo Bon S, Fabbri P. Macromol Mater Eng 2011;296:401e7. [22] Stauffer D, Aharony A. Introduction to percolation theory. London: Taylor & Francis; 1992. [23] Tougaard S. Surf Sci 1989;216:343e60. [24] Matsumoto Y, Koinuma M, Kim SY, Watanabe Y, Taniguchi T, Hatakeyama K, et al. ACS Appl Mater Inter 2010;2(12):3461e6. [25] Akhavan O. Carbon 2011;49(1):11e8. [26] Liu X, Pan L, Zhao Q, Lv T, Zhu G, Chen T, et al. Chem Eng J 2012;183:238e43. [27] Zhang W, Carravetta V, Li Z, Luo Y, Yang J. J Chem Phys 2009;131:244505/1e 244505/6. [28] Bittolo Bon S, Piccinini M, Mariani A, Kenny JM, Valentini L. Diam Relat Mater 2011;20:871e4. [29] Wang S, Zhang Y, Abidi N, Cabrales L. Langmuir 2009;25:11078e81. [30] Schniepp HC, Li JL, McAllister MJ, Sai H, Herrera-Alonso M, Adamson DH, et al. J Phys Chem B 2006;110:8535e9. [31] Paci JT, Belytschko T, Schatz GC. J Phys Chem C 2007;111:18099e111.