Full Paper
UV-Cured Epoxy–Zno Composites: Preparation and Characterization Marco Sangermano,* Federica Sordo, Mario Giovine, Galder Kortaberria
Epoxy–ZnO composites are prepared via UV-induced cationic polymerization. Complete unsoluble materials are achieved after 10 min of irradiation. Thermal analyses shows an increase of Tg in the presence of the ZnO attributable to a strong hindrance effect of the polymer chain mobility. Electrical characterization of the cured materials shows the decrease of the real part of permittivity with frequency. Both permittivity and AC conductivity values increase with nanofiller content, especially for the highest contents. This is described as typical behavior of nanocomposites made from polymers and fillers with higher permittivity and conductivity values, in accordance with previous data reported in literature for similar composites.
1. Introduction In the last few years ceramic–polymer composites were deeply investigated for their enhanced mechanical, thermal and functional properties.[1] In particular, the hybrid systems demonstrated to present the advantages of the two phases, with a possible improvement in their properties due to the interfacial phenomena.[2,3] In the case of ferroelectric composites, it is possible to couple the high dielectric and piezoelectric features of ceramics with the
M. Sangermano, F. Sordo, M. Giovine Local Unit INSTM, Politecnico di Torino, Dipartimento di Scienza Applicata e tecnologia, C.so Duca degli Abruzzi 24, 10129 Torino, Italy E-mail:
[email protected] F. Sordo Laboratoire de Technologie des Composites et Polyme`res (LTC), Ecole Polytechnique Fe´de´rale de Lausanne (EPFL), Station 12, CH-1015 Lausanne, Switzerland G. Kortaberria Materials and Technologies Group, Departamento de Ingenier´a Qu´mica y Medio Ambiente, Escuela Universitaria Polite´cnica, Universidad del Pais Vasco/Euskal Herriko Unibertsitatea, Plaza Europa 1, 20018 San Sebastian, Spain
1304
Macromol. Mater. Eng. 2013, 298, 1304–1308 ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
good mechanical properties and the easy processability of polymers.[4,5] These materials were exploited in various electronic applications, such as ultracapacitors for energy storage, transducers, piezosensors, and hydrophones.[6–8] We have recently investigated the preparation of BaTiO3 acrylic [9] and epoxy coatings;[10] these composites showed a homogeneous distribution of the ceramic particles within the organic matrix and electrical characterization showed a linear increase of dielectric constant by increasing the filler content, while low dielectric loss values were reached. Among ceramic fillers zinc oxide posses many interesting properties such as dielectric, piezoelectric, pyroelectric, optical and electrical.[11,12] ZnO is a II–IV compound semiconductor with a wide direct-band gap of 3.3 eV [13] that has been found to be a good candidate in some application fields.[14,15] There are not many works in the literature regarding the dielectric properties of epoxy/ZnO nanocomposites. Singha and Thomas [16] investigated epoxy/ZnO nanocomposites with a maximum filler amount of 5 wt% at microwave frequencies, finding that for low filler contents permittivity values decreased with respect to those of the unfilled matrix, explaining the phenomena in terms of surface interactions between the filler and the matrix that could constrain polymer chain mobility. Fothergill et al [17]
wileyonlinelibrary.com
DOI: 10.1002/mame.201200444
UV-Cured Epoxy–Zno Composites Preparation and Characterization www.mme-journal.de
reached to the same conclusion for epoxy nanocomposites with 10 wt% ZnO at lower frequencies, underlaying that the interaction zone surrounding the nanoparticles had a deep effect on the dielectric behavior of the nanocomposite. In our case, photocurable epoxy formulation with a considerably higher ZnO filler amount (from 50 to 75 wt%) has been used for nanocomposite preparation, while the frequency range in which their dielectric properties have been analyzed has been extended from 0.1 Hz up to 1 MHz. Epoxy/ZnO composites were prepared through UVcuring process. In fact UV-curing is a simple technique and usually does not require solvents or temperature treatments. UV-cured polymers are very promising for electronic applications because they are able to coat heatsensitive devices and circuits without damaging them.[18] Finally, it is important to underline that the UV-induced polymerization does not require solvents or temperature treatments and then it can be considered environmentally friendly.[19] The UV-cured materials were fully investigated by thermal, morphological and electrical characterizations.
2. Experimental Section 2.1. Material The epoxy resin 1,6-hexanediol diglycidyl ether (HDGE) was purchased by EMS (Switzerland, RV1812) and used as a reference UV-curable resin. The cationic photoinitiator Iodonium, (4methylphenyl)[4-(2-methylpropyl)phenyl]-exafluorophosphate(1), (Irgacure 250) was gently given from BASF. The ceramic filler zinc oxide was purchased from Aldrich, with an average size between 200 and 500 nm.
2.2. Sample Preparation A mixture of HDGE and zinc oxide at different weight percentages (50, 60, 70, and 75 wt%) were prepared. Homogeneous formulations and a good zinc oxide powder’s dispersion had been obtained by using ultra-turrax for about 10 min at the maximum speed (30 000 rpm). Then 4 wt% of photoinitiator (Irgacure 250) was added and the formulations were subjected to an ultrasonic bath for about 20 min. The formulations have been coated between 2 PP substrates and cured under UV-lights for about 5 min (UV intensity ¼ 40 mW cm2, as measured with EIT instrument). The investigated formulations are reported in Table 1.
2.3. Characterization The kinetic of the photo-polymerisation was investigated by realtime FT-IR spectroscopy by using a Thermo-Nicolet 5700 instrument. The specimens were exposed simultaneously to the UV beam, which induces the photo-polymerisation, and to the IR beam, which analyses in real-time the extent of the reaction. Because the IR absorbance is proportional to the monomer concentration,
www.MaterialsViews.com
Table 1. Composition of the investigated formulations.
HDGE [wt%]
ZnO [wt%]
HDGE
100
0
HDGE-50 wt% ZnO
50
50
HDGE-60 wt% ZnO
40
60
HDGE-70 wt% ZnO
30
70
HDGE-75 wt% ZnO
25
75
conversion versus irradiation time profiles can be obtained. Epoxy groups conversion was followed by monitoring the decrease in the absorbance of the epoxy groups centred at 750 cm1. Differential scanning calorimetry (DSC) was carried out on a Mettler DSC 30 (Switzerland) apparatus equipped with a lower temperature refrigerating system, at a scanning rate of 20 8C min1 from 60 8C to 100 8C. The glass transition temperature (Tg) was assumed as the mean value of the energy jump of the thermogram (average value between the onset and the endpoint of the glass-transition range). Gel content was determined on the cured films by measuring the weight loss after 24 h extraction with chloroform at room temperature, according to the ASTM D2765-84 technical standard. Composites morphology was investigated by field emission scanning electron microscopy (FESEM, Carl Zeiss-Supra 40) using an over chromium-coated (5 nm) on the fractured surfaces. Electrical characterization was carried out by a Novocontrol Alpha high resolution analyzer over a frequency range between 100 Hz and 1 MHz, by applying a voltage of 1 V at room temperature. Lower frequencies were not measured in order to avoid interfacial and electrode blocking effects whose appearance led to a relevant increase of the dielectric constant values. The instrument was interfaced to a computer and equipped with a Novocontrol cryogenic system for temperature control. Circular UV-cured films were placed between the gold platted electrodes in a sandwich configuration.
3. Result and Discussion The effect of the presence of ZnO filler on cationic UV-curing process of the HDGE epoxy resin was investigated recording the RT-FTIR analyses for the pristine epoxy resin and for epoxy formulations containing increasing content of ZnO in the range between 50 and 75 wt%. The aim of adding such a large amount of filler is related to reach percolation and modify the electrical properties of the crosslinked epoxy network. Since the amount of the ceramic filler is very high and therefore the influence on the photocuring process could be critical. In Table 2 the final epoxy group conversion are reported for the investigated formulations UV-irradiated for 5 or 10 min.
Macromol. Mater. Eng. 2013, 298, 1304–1308 ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1305
M. Sangermano et al. www.mme-journal.de
Table 2. Total conversion values of epoxy group after 5 and 10 min of UV irradiation, and gel content values.
HDGE
% conversion [5 min]
% conversion [10 min]
% Gel
Tg [-C]
100
100
99
15
HDGE-50 wt% ZnO
96
100
97
28
HDGE-60 wt% ZnO
100
100
97
43
HDGE-70 wt% ZnO
91
97
98
70
HDGE-75 wt% ZnO
83
91
97
150
It is evident that when ZnO content is below 70% it is enough to irradiate the formulation for 5 min in order to reach a high epoxy group conversion and fully cured films. By increasing the ZnO content above 70 wt% a longer irradiation time is needed in order to increase the epoxy group conversion and reaching to fully cured materials. These data show that even at high ZnO the photocuring process is highly efficient in epoxy crosslinks. The gel content values (reported in Table 2) is in accordance with the RT-FTIR data, showing values always above 97%, indicating the almost absence of extractable monomers or oligomers. The thermal properties of the UV-cured films were evaluated via DSC analyses. The Tg values are reported in Table 2 and it is possible to observe an important increase of Tg values by increasing the ZnO content. The pristine UVcured epoxy film is very flexible with a Tg value of about 15 8C. When 50 wt% of ZnO is added to the epoxy photocurable resin the Tg of the crosslinked network increased to 28 8C and reached a value of 150 8C when 75 wt% of the ceramic filler was dispersed in the polymer precursor. This behavior was expected and can be attributed to a strong hindering effect of the rigid ceramic filler on the polymeric chain mobility, with a consequent increase of the Tg values. A concomitant increase on surface hardness and increase scratch resistance can be expected but not tested on these materials.
In Figure 1 it is reported the FE-SEM images of the fracture surfaces of the UV-cured epoxy samples containing respectively 50 and 70 wt% of ZnO. We can observe that the microscopic fillers are strongly interacting with the polymer matrix and homogeneously distributed. Electrical characterization was performed by means of dielectric spectroscopy on UV-cured composites. Figure 2 shows the evolution of the real part of permittivity (dielectric constant, e0 ) and AC conductivity (sAC) with frequency for all the systems analyzed at 25 8C. A decrease in permittivity with frequency can be seen for all the systems analyzed. In a typical epoxy system, the epoxy component of permittivity is governed by the number of orientable dipoles present in the system and their ability to orient under the applied field. At the lower range of frequencies, all the dipolar groups in the epoxy chains can orient themselves resulting in higher permittivity values at these frequencies. As the frequency increases, the bigger dipolar groups find it difficult to orient, so the contribution of these groups to the permittivity is reduced, resulting in a continuously decreasing permittivity of the epoxy system. In the same way, the inherent permittivity of ZnO nanoparticles also decreases when frequency increases.[20] Since the permittivities of both epoxy and ZnO nanoparticles will decrease with increasing frequency, the
Figure 1. FE-SEM images of the fracture surfaces of the UV-cured epoxy samples containing respectively 50 and 70 wt% of ZnO.
1306
Macromol. Mater. Eng. 2013, 298, 1304–1308 ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
UV-Cured Epoxy–Zno Composites Preparation and Characterization www.mme-journal.de
a)
b) 10000
1E-5
HDGE 50 wt% ZnO 70 wt% ZnO 75 wt% ZnO
1E-6
1E-7
σAC (S/cm)
ε'
1000
HDGE 50 wt% ZnO 70 wt% ZnO 75 wt% ZnO
100
1E-8
1E-9
10
1E-10 1 0
2
4
6
0
2
log f (Hz)
4
6
log f (Hz)
Figure 2. Evolution of (a) dielectric constant and (b) conductivity with frequency at room temperature for all the composites analyzed.
seen in Figure 2b. This is also the typical behavior of adding semi-conductive fillers to an insulating polymer, increasing the former the conductivity of the nanocomposite. Figure 3 shows the evolution of both e0 and sAC values with filler amount at two representative frequencies, 1 kHz and 1 MHz. Both values increased with filler amount, being the increase more evident at the lower frequency for permittivity and at the higher frequency for conductivity. With respect to the initial permittivity value of epoxy matrix (9 at 1 kHz and 7.2 at 1 MHz) it increased an order of magnitude (122 at 1 kHz and 65 at 1 MHz) for the highest ZnO content. The increase is more pronounced for 70 and 75 wt% ZnO nanocomposites, whereas 50 wt% ZnO nanocomposite shows permittivity values higher but quite close (14 and 7.5, respectively) to those of neat epoxy. Regarding conductivity values, differences are more evident at lower frequencies. At 1 kHz, conductivity values increased one order of magnitude for the nanocomposite with 50 wt% and two orders of magnitude for those with 70 and 75 wt% (3.9 1010, 2.8 109, 3.9 108, and 4.9 108 S cm, respectively). At 1 MHz
permittivity of the nanocomposite is observed to decrease, as can be seen in the figure. As it can be seen, e0 values increased with filler amount when compared with those of pure epoxy matrix, especially at lower frequencies. The addition of fillers with a higher permittivity to a polymer causes an increase in the permittivity of the composite.[16] As was commented above, some authors have found a decrease of the permittivity with the incorporation of ZnO nanoparticles, especially at high frequencies and for lower amounts of filler of around 0.1 and 0.5 wt%, attributing it to interactions between the filler and matrix which hindered the mobility of epoxy chains. For higher filler amounts, as it is in our case, those authors also found an increase of permittivity with filler amount, which has been described as the typical behavior. Indeed, the addition of the filler is done for increasing permittivity for electronic applications. Conductivity values increased with frequency, in an opposite way than permittivity. On the other hand, AC conductivity values increased with filler amount in all the frequency range but especially at lower ones, as it can be
a) 140
b)
1 kHz 1 MHz
120
1E-6
100
1E-7
σAC (S/cm)
ε'
80 60 40
1 kHz 1 MHz
1E-8
1E-9 20
1E-10
0 -10
0
10
20
30
40
50
60
70
80
-10
0
10
20
wt% ZnO
30
40
50
60
70
80
wt% ZnO
Figure 3. Evolution of (a) dielectric constant and (b) conductivity with filler amount at two representative frequencies of 1 kHz and 1 MHz.
www.MaterialsViews.com
Macromol. Mater. Eng. 2013, 298, 1304–1308 ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1307
M. Sangermano et al. www.mme-journal.de
differences were not so high, increasing one order of magnitude for all the nanocomposites when compared with that pure epoxy matrix (3.9 E 7, 1.4 106, 2.1 106, and 3.3 106 S cm, respectively).
Keywords: electrical properties; epoxy composites; UV-curing; ZnO
[1] [2] [3] [4]
4. Conclusion Epoxy–ZnO composites were prepared via UV-induced cationic polymerization. The UV-curing process was investigated showing a detrimental effect induced by the presence of the filler, with a decrease of the epoxy group conversion. This detrimental effect can be overcome by irradiating the samples for longer time. Complete unsoluble materials are achieved after 10 min of irradiation. Thermal analyses showed an increase of Tg in the presence of the ZnO attributable to a strong hindrance effect of the polymer chain mobility. Electrical characterization of the cured materials showed the decrease of the real part of permittivity with frequency, due to the decrease of both epoxy and nanoparticles dielectric constant, which made the overall constant of the composites to decrease. Both permittivity and AC conductivity values increased with nanofiller content, especially for the highest contents. This has been described as the typical behavior of nanocomposites made from polymers and fillers with higher permittivity and conductivity values, in accordance with previous data reported in literature for similar composites.
[5] [6] [7] [8] [9] [10]
[11] [12] [13] [14] [15] [16] [17] [18] [19]
Received: December 7, 2012; Accepted: February 6, 2013; Published online: May 10, 2013; DOI: 10.1002/mame.201200444
1308
[20]
T. R. Cooke, J. Am. Ceram. Soc. 1991, 74, 2959. C. Sanchez, F. Ribot, B. Lebeau, J. Mater. Chem. 1999, 9, 35. C. Sanchez, F. Ribot, New J. Chem. 1994, 18, 1007. S. F. Wang, Y. R. Wang, K. C. Cheng, Y. P. Hsaio, Ceram. Int. 2009, 35, 265. X. Yang, Z. Yang, C. Mao, J. Du, J. Rare Metals. 2006, 25, 250. S. K. Bhattacharya, R. R. Tummala, J. Mater. Sci. Mater. Electron. 2000, 11, 253. Z. M. Dang, Y. F. Yu, H. P. Xu, J. Bai, J. Compos. Sci. Technol. 2008, 68, 171. L. Ramajo, M. S. Castro, M. M. Reboredo, Composites, Part A 2007, 38, 1852. M. Lombardi, A. Guerriero, G. Kortaberria, I. Mondragon, M. Sangermano, L. Montanaro, Polym. Compos. 2011, 32, 1304. M. Sangermano, M. Lombardi, A. Guerriero, G. Kortaberria, I. Mondragon, F. Pirri, L. Montanaro, Macromol. Mater. Eng., DOI: 10.1002/mame.201200130. S. Ostrovsky, G. Kazimirsky, A. Gedanken, C. Brodie, Nano Res. 2009, 2, 882. Z. Fan, J. G. Lu, J. Nanosci. Nanotechnol. 2005, 5, 1561. V. Srikant, D. R. Clarke, J. Appl. Phys. 1998, 83, 5447. S. C. Tjong, G. D. Liang, Mater. Chem. Phys. 2006, 100, 1. J. I. Hong, P. Winberg, L. Schadler, R. W. Siegel, Mater. Lett. 2005, 59, 473. S. Singha, M. J. Thomas, IEEE Trans. Dielectr. Electr. Insul. 2009, 16, 531. J. C. Fothergill, J. K. Nelson, M. Fu, IEEE Conf. Electr. Insul. Dielectr. Phenomena 2004, 16, 406. Radiation Curing in Polymer Science and Technology, Vol. I–IV (Eds: J. P. Fouassier, J. F. Rabek), Elsevier, London 1993. J. P. Fouassier, J. C. Rabek, Radiation Curing in Polymer Science and Technology, Elsevier, London 1993. L. M. Levinson, H. R. Philipp, J. Appl. Phys. 1976, 47, 1117.
Macromol. Mater. Eng. 2013, 298, 1304–1308 ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.MaterialsViews.com
