Journal of Non-Crystalline Solids 356 (2010) 763–767
Contents lists available at ScienceDirect
Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/locate/jnoncrysol
Dielectric spectra of natural cork and derivatives M.C. Lança a,*, M. Brandt b, E.R. Neagu a,c, C.J. Dias a, J.N. Marat-Mendes a a
Departamento de Ciência dos Materiais (I3N/CENIMAT), Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal Hochshule (FH), Geusaer Straße, D-06217 Merseburg, Germany c Department of Physics, Technical University of Iasi, B-dul D. Mangeron 67, Iasi 700050, Romania b
a r t i c l e
i n f o
Article history: Available online 3 February 2010 Keyword: Dielectric properties, relaxation, electric modulus
a b s t r a c t Cork is a cellular biomaterial that has unique characteristics that make it suitable for many types of applications. Since it is also an electrical insulator, the study of its electrical and dielectric properties can lead to new interesting applications. The moisture present in cork and derivatives has a very important role on the dielectric properties. In this work a composite made of both recycled cork and TetraPakÒ used containers was studied and compared with other cork products. The dielectric relaxation spectra of natural cork (as received), commercial cork agglomerate and of a composite cork/TetraPakÒ was investigated in the temperature range of 50 to 120 °C and in the frequency range of 101 Hz–2 MHz. For some samples of the composite a small amount of paraffin was added. The highest values for the imaginary part of the dielectric permittivity were found for the commercial material and the composite without paraffin. The lowest was found for the cork/TetraPakÒ/paraffin composite. The influence of humidity content was investigated for the composite with wax. Natural cork shows a peak around 80 °C (not seen in the derivative materials). The commercial agglomerate and the cork/TetraPakÒ/paraffin composite show a peak around 40–50 °C. In the composite this peak becomes smaller as humidity is removed. Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction Cork is a natural cellular material which has been used for centuries. Its unique mechanical, acoustical and thermal properties make cork and its derivative products ideal for many different types of applications [1,2]. Cork is a good electrical insulator and because of the cells filled with gas it might be possible to use it as a porous dielectric, which can be electrically charged and it is able to retain this charge. So it could be of great interest to study the dielectric properties of these materials with the aim of finding possible new applications. For instance, to determine if it has piezoelectric properties that would make it suitable for use as smart sensors. Previous studies have already shown that the sample humidity greatly influences the electrical properties of cork products and consequently charge storage [3–5]. In this work samples from natural cork, commercial cork agglomerates and a composite of cork/TetraPakÒ or cork/TetraPakÒ/paraffin were studied. The composite is a recycled material of both cork and TetraPakÒ. This cork derivative can be used as cheaper and environmental-friendly replacement of cork and related materials. In order to understand the influence of water content samples could be or not dried prior to preparation and/or measurement. Measurements were performed using a broadband
* Corresponding author. Tel.: +351 21 294 8564; fax: +351 21 295 7810. E-mail address:
[email protected] (M.C. Lança). 0022-3093/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2009.09.042
impedance analyzer (101 Hz–2 MHz) and the temperature range used in measurements was 50 to 120 °C. The results for the different types of cork samples were compared and also with previous published results for natural cork [3]. The influence of drying was investigated in order to improve the ability of the materials to store space charge and consequently to optimize preparation. 2. Experimental details 2.1. Sample preparation Thin samples of natural cork of 75 lm thickness were supplied by industry and measured as received. The cork agglomerate is a commercial one obtained from cork waste from the production of cork stoppers and from lower quality cork that was not suitable for stoppers. Usually an adhesive is added to improve adhesion of the grains. The sample thickness is 4 mm. The cork/TetraPakÒ (50/50 wt%) composite is obtained from cork grains and grinded waste of TetraPakÒ containers. The materials are mixed and hot pressed. In an attempt to better control the water content of the composite a small amount of commercial paraffin, cork/TetraPakÒ/paraffin (48/48/4 wt%) composite, was also added to the mixture. For the samples of the composite with paraffin the grain mixtures were dried in an oven at 50 °C during
764
M.C. Lança et al. / Journal of Non-Crystalline Solids 356 (2010) 763–767
1 week prior to pressing. For the composites disks of around 1.5 mm thickness were produce to be used in the dielectric relaxation spectroscopy (DRS) measurements. Some of the samples were further dried after preparation in an oven in air at 50 or 70 °C or in a chamber at primary vacuum (25 °C and 4 mbar). DRS measurements were performed using a NovoControl-GmbH Alpha-N broadband impedance analyzer (frequency range was 101 Hz–2 MHz) and the temperature was controlled by a NovoControl-GmbH Quatro Criosystem BDS-1100 (temperature range used in measurements was 50 to 120 °C) to an accuracy of 0.1 °C. 3. Results and discussion 3.1. Comparison between natural cork, cork agglomerate and cork composite In Fig. 1 is shown the imaginary part of the dielectric permittivity (e00 ) of the different types of cork samples measured. These results were obtained with samples not dried. Natural cork (a) and cork agglomerate (b) were analyzed as received. In the preparation of the samples of cork/TetraPakÒ (c) the powder was not dried before hot pressing contrary to the samples of cork/TetraPakÒ/paraffin for which the powder was previously dried in oven before pressing. The lowest values of e00 are observed for natural cork (Fig. 1(a)) and for the cork/TetraPakÒ/paraffin composite (Fig. 1(d)). One peak can be seen on both spectra, but their maxima are at different temperatures: around 80 °C for natural cork and around 40–50 °C for
the composite. The peak in natural cork was studied by MaratMendes and Neagu [3] and it decreases as the sample is dried. Additionally, natural cork is the only material that does not show a rise in e00 for higher temperatures and lower frequencies. However the cork samples were much thinner than the ones for the other materials. The peak observed in the composites appears also in the other two materials (Fig. 1(b) and (c)) at a similar temperature. As can be seen in Fig. 1 the higher values for e00 were obtained for the commercial agglomerate and the cork/TetraPakÒ composite. However the two spectra are different, in Fig. 1(b) is visible the peak around 40–50 °C while in Fig. 1(c) only the increase at higher temperatures is observed. The spectra in Fig. 1(c) and (d) have similar features (peak at around 40–50 °C and an increase for higher temperature and lower frequency) in spite of a difference of nearly two orders of magnitude. It is possible that the adhesive and the paraffin give rise to peaks with similar temperature maxima (because both peaks are related to the presence of water as it will be discussed further ahead). Since the cork/TetraPakÒ/paraffin composite has the lowest e00 , it seems to be the better insulator and probably the more able to store charge. 3.2. Influence of the water content on the DRS of cork/TetraPakÒ/ paraffin composite As a consequence of the results shown in Fig. 1 further measurements were made on the cork/TetraPakÒ/paraffin composite with different water content.
Fig. 1. DRS for natural cork (a), commercial cork agglomerate (b), cork/TetraPakÒ composite (c) and cork/TetraPakÒ/paraffin composite (d).
M.C. Lança et al. / Journal of Non-Crystalline Solids 356 (2010) 763–767
The spectra seen in Fig. 2 are: (a) for a sample not dried after pressing – run 1; (b) run 2 – for the same sample after run 1 (since the final temperature of run 1 was 120 °C the sample went through a drying process) and (c) a sample dried for 5 days in primary vacuum (4 mbar). It is clearly observed that the peak around 40 °C present in Fig. 2(a) has a reduced intensity in Fig. 2(b) and it is only slightly visible in Fig. 2(c). Since the water content decreases from (a) to (c), it is likely that the peak can be attributed to water in the composite. These results agree well with the ones obtained for this material by isothermal charging/discharging currents [6]. Yet it is not possible to know if it is related to bond or free water.
765
The increase observed for higher temperatures remains almost unchanged. So it is reasonable to conclude that is related to intrinsic properties of the material and not to the humidity. Therefore the water present in the composite masks the intrinsic properties but also makes the material more conductive and less able to store electrical charges for longer times. 3.3. Data analysis To attempt to identify the physical mechanisms responsible for the different features in the DRS spectra different formalisms for dielectric data analysis were used for the cork/TetraPakÒ/paraffin composite. Since the intrinsic properties are observed more clearly in the dried materials more attention has been made to the samples of the composite dried in vacuum. 3.3.1. Permittivity formalism In Figs. 1 and 2 the graphs of the imaginary part of the permittivity together with the plots of real part of the permittivity, e0 (not shown here) were analyzed (Eq. (1)) so that possible relaxation mechanisms (dipolar or from trapped space charge) could be detected (shown as peaks in e00 graphs and a step in e0 ).
e ðf Þ ¼ e0 ðf Þ ie00 ðf Þ:
ð1Þ
As stated before a peak around 40 °C related to water content was found. It is known that dc electrical conductivity gives rise to an increase in e0 with decreasing frequency (e0 f1) and to a plateau in e00 (e00 = ro/2pfeo) [7]. For the conductivity (Eq. (2)): the real part will be constant and the imaginary part will be directly proportional to the frequency.
r ¼ ixe eo :
ð2Þ
So if there is pure electronic polarization, the plot of the real part of conductivity vs frequency will show a plateau for lower frequencies. In Fig. 3 it can be observed that r0 is not constant with frequency for the temperature between 0 and 90 °C. This result is also valid for all the temperature range of the measurements. So we can say that the material presents no dc conductivity for the given experimental conditions and measurement limits (no onset of a plateau is observed for the lowest frequencies for any temperature in the measurements). 3.3.2. Impedance formalism The impedance formalism is obtained by transformation of the complex permittivity:
Fig. 2. DRS spectra of cork/TetraPakÒ/paraffin composite with different water content: (a) not dried after pressing, (b) dried from the previous DRS measurement (that reached 120 °C) and (c) dried 5 days in vacuum.
Fig. 3. Real part of the conductivity plot of the cork/TetraPakÒ/paraffin dried for 5 days in vacuum.
766
Z ðf Þ ¼
M.C. Lança et al. / Journal of Non-Crystalline Solids 356 (2010) 763–767
1 ; i2pfC o e ðf Þ
ð3Þ
were Z* is the complex impedance and Co the geometrical capacitance when the sample is replaced by vacuum. Nyquist plots (imaginary impedance vs real impedance) are used to better distinguish between MWS and electrode phenomena [8,9]. The phenomena are observed as depressed semicircles. The electrode polarization gives rise to higher capacitance and larger relaxation times than bulk polarization. This will results in two separated arcs on a Nyquist plot for each of the two different processes [9]. Fig. 4 shows the Z00 (Z0 ) plot for higher temperatures. No semicircle is observed on the graph. This means that only one relaxation mechanism is involved and it is only partially seen. 3.3.3. Modulus formalism The complex modulus is the inverse of the complex permittivity, M* = 1/e*. In this formalism the influence of electrode polarization is suppressed. Conductivity effects will appear as peaks in the log M00 vs log f plot at low frequency [7]. The analysis of the composite dried in vacuum for DRS measurements made at higher temperatures can reveal these peaks. In Fig. 5 is shown the Arrhenius plot for the frequency maxima of the peaks found in M00 graph for isothermal DRS measurements made at 90, 100 and 110 °C. For lower temperatures the maximum was below the frequency measurement range. The value obtained for the activation energy is low, Ea = 0.3 eV according to Eq. (4),
f ðTÞ ¼ fo eEa =kB T ;
ð4Þ
kB is the Boltzmann constant and fo the relaxation rate in the high temperature limit. From the frequency of the maxima in the M00 graph (xm) and from the permittivity data, the permittivity in the high frequency limit (e1) can be estimated and a value for the conductivity ro can be calculated,
ro ¼ eo e1 xm :
ð5Þ
From Fig. 6(a), a drop of one order of magnitude in conductivity can be observed between samples not dried and dried due to a first DRS run up to 110 °C. The plot was made for lower temperature because no peaks were observed for the non-dried sample in the imaginary modulus graph at higher temperatures. In Fig. 6(b) similar values are obtained for the vacuum dried and DRS dried samples. This second plot was made only at high temperature since, no other maximum were observed for the vacuum dried sample. 3.3.4. Peak analysis van Turnhout and Wübbenhorst [10] proposed a method that allows a better analyzes of the relaxation peaks in the data because conductivity effects are not seen. A plot is made of the derivative of the real part of the permittivity, which is related to the imaginary part by
oe0 2 ¼ e00 ; oðlog xÞ
ð6Þ
Fig. 4. Nyquist plot of the impedance of the cork/TetraPakÒ/paraffin dried for 5 days in vacuum.
Fig. 5. Arrhenius plot of the maximum of frequency of the peaks found in the imaginary modulus vs frequency graph of the cork/TetraPakÒ/paraffin dried for 5 days in vacuum.
Fig. 6. Arrhenius plot of the conductivity calculated according to Eq. (5) of the cork/ TetraPakÒ/paraffin dried during 5 days in vacuum: effect of water content in DRS spectra for (a) low and (b) high temperature measurements.
M.C. Lança et al. / Journal of Non-Crystalline Solids 356 (2010) 763–767
767
intensity of the peak around 40–50 °C with decreasing in humidity shows the importance of controlling the amount of water in the material. It is probable that the paraffin will make a (thin) film covering the grains of cork and TetraPakÒ. Since paraffin is hydrophobic it will result in a better control of water absorption and desorption by the grains of the composite. As well it would be interesting to see, in further work, if water re-absorption has been reduced (or at least slowed down) by the presence of paraffin. This might increase the ability of the composite of cork/TetraPakÒ/paraffin to store and retain electric charge and thus to be used as a sensor. Acknowledgements
Fig. 7. Arrhenius plot of the minimum of frequency of the peaks found in the derivative of the real part of permittivity vs frequency graph of the cork/TetraPakÒ/ paraffin dried during 5 days in vacuum.
x = 2pf. In the plot of this derivative vs log f the peaks in e00 appear as minima and it is possible a better separation between overlapping peaks, besides the removal of the conductivity effects that can also mask peaks. From this kind of plots for low temperature a peak was observed and an Arrhenius plot of the frequency minima was made (Fig. 7). The activation energy estimated from an Arrhenius dependence (Eq. (4)) is of 0.3 eV. 4. Conclusions The new composite of cork/TetraPakÒ with the addition of paraffin appears to have improved electrical insulation characteristics since it has the lowest values for the permittivity. The decrease in
The authors are indebted to Eng Luis Gil and Eng. Paulo Cortiço of INETI/Portugal for helping in producing the composites. References [1] L.J. Gibson, M.F. Ashby, Cellular Solids-Structure and Properties, Cambridge University, Cambridge, 1997. [2] S.P. Silva, M.A. Sabino, E.M. Fernandes, V.M. Correlo, L.F. Boesel, R.L. Reis, Int. Mater. Rev. 50 (2005) 345. [3] J.N. Marat-Mendes, E.R. Neagu, Ferroelectrics 294 (2003) 123. [4] M.C. Lança, E.R. Neagu, P. Silva, L. Gil, J.N. Marat-Mendes, Mater. Sci. Forum 514–516 (2006) 940. [5] M. Carmo Lança, W. Wirges, E.R. Neagu, R. Gerhard-Multhaupt, J. MaratMendes, J. Non-Cryst. Solids 353 (2007) 4501. [6] M.C. Lança, S. Peuckert, E.R. Neagu, L. Gil, P.C. Silva, J. Marat-Mendes, Mater. Sci. Forum 587&588 (2008) 613. [7] F. Kremer, A. Schönals, Broadband Dielectric Spectroscopy, Springer, Berlin, 2002. [8] A.K. Jonscher, Dielectric Relaxation in Solids, Chelsea Dielectrics, London, 1983. [9] E. Neagu, P. Pissis, L. Apekis, J. Appl. Phys. 87 (2000) 2914. [10] J. van Turnhout, M. Wübbenhorst, Dielectr Newsl, Issue November, NovoControl, 2000, p. 1.
