X-ray diffraction and differential scanning calorimetry studies of a BaTi[O.sub.3]/polyvinylidene fluoride composites.
Electroactive composites, consisting of polymer matrix loaded with electroactive ceramics, are still of interest, because their properties can be tailored to the requirements of smart structures, sensors, and actuators (1). These composites combine the advantages of ceramics and polymers, presenting a novel type of materials that is easy to process and with very attractive physical properties for industrial applications. One of these polymer-ceramics composites is the barium titanate/polyvinylidene fluoride [(BT).sub.x] [(PVDF).sub.100-x], series. These [(BT).sub.x], [(PVDF).sub.100-x] series were successfully used as a dielectric layer inserted between rear electrode and phosphor layer in the high-brightness inorganic EL device (2).
PVDF is a semicrystalline polymer in which five crystallographic forms are observed with different conformations (3, 4). These crystalline forms can transform to each other under specific conditions. The most common polymorph of PVDF--the [alpha]-form--does not show a net lattice polarization, whereas the [beta]-form shows a large spontaneous lattice polarization, making it technologically most desirable (5-7). This polymer (PVDF) has already been commercialized in transducer, pyroelectric, piezoelectric, and some actuator applications (8).
BT, with its lead-free feature, is widely used in the electronic industry, and its particles larger than 0.5 [micro]m usually show a tetragonal-to-cubic phase transition at a Curie temperature and is known to change with particle size (9). It exhibits piezoelectric characteristics below its Curie point (because the center of symmetry is lost owing to the transition from cubic to a tetragonal structure). The best dielectric properties for BT are obtained with tetragonal phase (10). One of its industrial applications is the multilayer ceramic capacitors (MLCCs). This MLCC industry prefers BaTi[O.sub.3] powders with a tetragonality higher than 1.008 (11). Tetragonality, the lattice distortion (=c/a), or the degree of ferroelectricity, is defined as the relative ratio of lattice parameter of c-axis to a-axis as measured by X-ray diffraction (XRD).
The polymer-ceramic [(BT).sub.x] [(PVDF).sub.100-x], composites have been studied for various electronic/microelectronic applications. Their structural, microstructural, dielectric, hysteresis, resistivity, and pyroelectric properties were studied (1), (12-14). It was concluded that interactions between PVDF and BaTi[O.sub.3] cannot be neglected, because they affect the melting point and heat of fusion of PVDF in BaTi[O.sub.3]/ PVDF composites as the weight fraction of BaTi[O.sub.3] is varied (15). It was proved that the composites possessed superior properties compared to their individual components (1), (16). The sample preparation technique was found to play a crucial role in the properties of these composites (5), (13), (16).
The activation energy of the segmental motion of the polymer matrix was found to increase with increasing contents of the filler (1). It was also found that the conduction process going on in these composites is a thermally activated process (12).
It appears that almost no work has been reported on the variation of tetragonality of BT in any polymer/ceramic composite as a function of BT content. Also, all the studied composites here show [alpha]-phase of PVDF in it. Till now, this [alpha]-phase is not so important as the [beta]-phase, and so we studied it for scientific interest and to open the door for more future studies.
In this work, the ratio of the constituents of this composite is altered, and the structural and thermal changes are studied using XRD and DSC techniques paying attention to the tetragonality change during composition change. Also, we study the effect of annealing and stretching on the structure of [(BT).sub.x] [(PVDF).sub.100-x] composites.
The BT and PVDF used in this work were obtained from (Fluky Chemika Co., and Aldrich (Switzerland), respectively. The samples were prepared by taking different weight percentages (0, 0.5, 1, 2, 3, 5, 7, 9, 10, 15, 20, and 30%) of BT powder and mixing them thoroughly in PVDF powders, using a stirrer for 10 min at room temperature (25[degrees]C). A droplet of acetone was used as a thinner in the mixing process to facilitate a good mixing. The mixture was fused at 200[degrees]C for 10 min to get a solid paste. Then, the solid paste of composites was pressed between two thin aluminium foils at 201[degrees]C [+ or -] 2[degrees]C for 2 min at a pressure of 6 X [10.sup.3] kN/[m.sup.2] to have the films of PVDF/BT. After this step, the composite samples were slowly cooled down to room temperature, and the foils were removed. Foils were cut into pieces of 60 mm long and 5 mm in width. The thickness of the samples lies between 100 and 140 [micro]m. This method of simple mixing of polymer with BT powder was used by many authors for preparing similar samples (1), (13), (17).
X-Ray and DSC Measurements. The samples were characterized using XRD powder analysis. A Siemens X-ray diffractometer with Cu [K.sub.[aslpha]] radiation ([lambda] = 1.5406 A) was used. XRD measurements were performed at room temperature in the range of 20 = 10[degrees] - 80[degrees] with very slow rate of 0.02[degrees] per step.
DSC thermograms for all samples were achieved using a Pyris 1 DSC outbalance (From Perkin Elmer) at a rate of 0.5 [degrees] C/min. The heating and cooling range are -100-200[degrees]C.
Stretching Measurements. A manually calibrated apparatus was used to study the effect of stretching on the physical properties of the studied composites. The sample films were uniaxially stretched at room temperature for 10, 20, 30, and 40% of its original length.
Annealing Measurements. The sample was annealed to 180[degrees]C for 3 h at a rate of 2 [degrees] [min.sup.-1] and cooled at a 1 [degrees] [min.sup.-1] to RT(25[degrees]C). The annealing was carried out to delete the thermomechanical history of the sample before any additional treatment. A part of this history is recorded in the crystalline phase and can be deleted only by annealing at a temperature above the melting point (~177[degrees]C).
RESULTS AND DISCUSSION
Figure 1 shows the X-ray diffraction (XRD) patterns of [(BT).sub.x] [(PVDF).sub.100-x] composite samples for x = 0, 0.5, 7, 10, 30, and 100. XRD pattern of pure PVDF (Fig. la) shows a sharp reflection peak at 20 value of 20[degrees] corresponding to (110) diffraction planes. Also, the calculated lattice parameters are a = 5 A, b = 9.5 A, and c = 9.6 A. This is a typical [alpha]-form PVDF peak. The other prominent peaks are at 2[theta] = 17.75[degrees], 18.41[degrees], 26.56[degrees], 33.11[degrees], 35.750[degrees], and 38.62[degrees], indicating that PVDF is mainly in its [alpha]-form (14). We had not observed any peak at 2[theta] = 20.57[degrees] or at 36.3[degrees], which are the peaks for the [beta]-form of PVDF (18). It is also clear in Fig. 1 that while the amount of BT increases in the composite, the intensity of the BT peaks increases quickly, and the intensity of PVDF peaks tends to diminish. The clear and sharp crystalline diffraction peaks in pure PVDF become less prominent in the presence of BT. The decrease in intensity and gradual broadening of the diffraction peaks with BT content suggest a decrease in both crystalline size and the degree of crystallinity of PVDF. This is because the BT particles dispersed in the PVDF might have an important effect on the crystallization of the PVDF polymer (19). That is to say, the amorphous region of the PVDF polymer is enlarged in the expense of the crystallized region. So, one can say that the hindrance to the PVDF crystallization becomes more and more serious with the increase of BT ratio in the composite. This behavior could be attributed to the interaction between [Ba.sup.2+] ions and fluorine in the polymer. PVDF is not an ionomer; the highly depolarized CF bond is capable of forming a weak acid-based complexation with barium and thus retards the order of crystalline PVDF. This crystallization collapse could also be related to peaks intensity simply and not to the absence of polymer, that is, due to BT be much more crystalline than PVDF.
The XRD spectrum of BT (Fig. 10 shows 10 peaks in the 10[degrees]-80[degrees] range. These peaks appear in the XRD patterns of all composite samples (except pure PVDF). In contrast to the earlier similar published data (14), we observe that there is no change in the position of the reflection peaks due to BT, that is, all the BT characteristic XRD peaks appear in the composite samples in the same position of pure BT whatever the BT content. The peaks appear at 2[theta] values of 22.21[degrees], 31.54[degrees], 38.89[degrees], 45.330[degrees], 50.94[degrees], and 56.23[degrees]. This behavior (the absence of any change in the reflection peaks of BT) could suggest that there is no change in the morphology of BT, although it was coated by PVDF particles.
Effect of Annealing and Stretching on the Structure of [(BT).sub.x] [(PVDF).sub.100-x], Composite Samples
Figure 2 shows the effect of annealing and stretching on the XRD patterns of [(BT).sub.7] [(PVDF).sub.93] composite sample as a representative one.
The crystalline structure of the polymer depends strongly on the conditions under which it is formed. In case of PVDF polymer, a high or moderate cooling rate generally produces the [alpha]-conformation directly from the melt (20). The latter can be transformed to the [beta]-conformation by stretching the samples. Salimi and Yousefi (21) concluded that the maximum [beta]-phase content was obtained at 90[degrees]C and at a stretching ratio of about 4.5-5. They also concluded that the stretching ratio affects [beta]-phase content more than the stretching temperature. We study the effect of stretching on the [(BT).sub.x] [(PVDF).sub.100-x] composite samples at room temperature with small stretching ratios to follow up the changes in the crystalline structure in this range. Figure 2 shows the XRD patterns for the as-received, annealed, and stretched samples (the sample was stretched to 1.4 of its original length). The figure intimates that only the [alpha]-phase exists in the three spectra, and there are no evidences for the existence of any other phases. Once more, all the composite characteristic XRD peaks appear in the same position in the three samples shown. The composite samples seem not to change under annealing, but the BT tetragonality changes will be discussed in the next paragraph.
Effect of PVDF on the Tetragonality of BT in [(BT).sub.x] [(PVDF).sub.100x] Composites
Figures 1 and 2 show a noticeable splitting of the XRD peaks of BT in the [(BT).sub.x] [(PVDF).sub.100-x] composite samples at 2[theta] = 45[degrees]. These peaks [corresponding to (200) diffraction plane] are used to distinguish a cubic structure from a tetragonal structure (16). At this angle, each of these peaks seems not to be single. Theoretically, 100% tetragonal BT has two separate peaks between 2[theta] = 44[degrees] and 47[degrees]. Complete cubic barium titanate shows just one peak. A mixture of tetragonal and cubic BT will show all intermediate forms between one and two peaks. In addition to the ratio c/a, tetragonality can also be expressed by comparing the relative height of the (002) and (200) peaks (22).
Nanosized BT powder with a high tetragonality is essential in the manufacturing of MLCCs. Tetragonality diminishes with a decrease in particle size and disappears below a critical particle size. Many researchers studied the BT critical particle size at which the tetragonality disappears. All the results show differences in this critical particle size (23-25). We used the following Scherrer formula to estimate the grain size for the used BT (from the XRD pattern of Fig. 1) (26):
[d.sub.XRD] = k[lambda]/[beta]([theta]) cos [theta] (1)
where [lambda] is the X-ray wavelength, [beta] the full width at half maximum (FWHM) of the diffraction line, and [theta] the diffraction angle with the constant K = 1. We fit the most intense (101) peak as Gaussian using [Microcal (TM) origin, ver.7/Microcal Software, Northampton, USA] program. The average particle size values obtained are ~110 nm.
Figure 3 shows the XRD patterns of [(BT).sub.x] [(PVDF).sub.100-x] composite samples for 2[theta] = 44[degrees]-46[degrees]. The figure shows two reflection peaks: (002) and (200), The peaks are enlarged to compare the BT tetragonality. Curves of Fig. 3 were fitted to two Gaussian curves that corresponds to BT (200) and (002) peaks using the above-mentioned Microcal (TM) program. The measured BT tetragonalities (c/a) were plotted as a function of BT content in Fig. 4. The figure intimates that tetragonality increases with the increase in BT content up to 30%, where it shows a saturation value of about 1.005 for BT ratio equal to or larger than 30%. This maximum value is low compared to 1.008, which is the preferable value for MLCC industry (11).
Li and Shih (27) concluded that clustering has an important effect on the tetragonal crystal structure of BaTi[O.sub.3] particles. A particle in a cluster may behave differently from an individual particle, because it is in contact with other particles. According to this view, one can say that the inclusion of PVDF hinders the particle growth by increasing the distance among particles in the [(BT).sub.x] [(PVDF).sub.100-x] composites leading to less strain resulting in a decrease in tetragonality. In case of the composite sample that contains 0.5% BT, the nanocrystals of BT are completely surrounded by PVDF polymer, and so the microstrain is minimum and the distortion from the cubic structure gets its minimum value as our results found.
Also, it is worth noting in Fig. 3 that the tetragonality for the as-received sample contains 7% BT is 1.0029, which is lower than the value of the stretched sample (1.0048). So, it seems that stretching enhances tetragonality.
Differential Scanning Calorimetry
Figure 5 shows the DSC heating curves of [(BT).sub.x] [(PVDF).sub.100-x] composites. It intimates dual melting endotherms for each thermogram. This behavior (reorganization) is observed earlier for some polymers, when the scanning rate is [less than or equal to]5[degrees]C/min, which is our case (28). Reorganization is a process when improvements or perfections of the initial metastable crystallites occur (29). Reorganization is usually observed on heating as a double-melting peak, where the lower temperature peak represents the melting of the original metastable crystallites formed at the crystallization temperature from the amorphous phase, while the upper temperature peak corresponds to the melting of the crystallites recrystallized and perfected during the heating process of the DSC experiment itself. The faster the heating rate, the more pronounced the lower temperature peak becomes, because less time is available for the metastable crystal to perfect itself. To test this postulation, we quench the composite sample [(BT).sub.7] [(PVDF).sub.93] in liquid nitrogen from room temperature. This could cause the damage of most of the existing crystallization phases and may be due to the difference of expansion coefficients of its constitutes. DSC traces of this sample (included in Fig. 5) show more than three melting peaks, indicating that reorganization process of more than two phases formed during the heating process of DSC measurements. That is to say, more than three kinds of distribution of crystal perfection may occur when the sample is nitrogen-quenched. None of these peaks lie in the range of melting temperature of [beta]-conformation (5). In the DSC trace of the as-received PVDF sample, the small low-temperature broad peak lies at 165[degrees]C, and the sharp peak lies at 176.9[degrees]C. The narrowing of the melting peak implies recrystallization of imperfect crystallites. The melting peak at 176.9[degrees]C is proved to be the melting temperature of the [alpha]-conformation (5), and so our sample contains [alpha]-conformation only, and we have not seen any melting peak in the range of the melting temperature of [beta]-conformation, which varies between 190 and 300[degrees]C (5). The melting temperature values ([T.sub.m]) of the samples decrease with increasing BT content. The DSC measurements (Table 1) show that the melting temperature values [T.sub.m] of the samples decrease with BT content. It seems that, on increasing the BT content, the crystallization process is more hindered, leading to smaller and smaller crystalline regions richer and richer in defects. This decrease in crystalline regions and the enrichment of crystal defects with BT content of the samples are reflected in the decrease of [T.sub.m] of the samples and are in good agreement with the XRD results.
It was also found (Table 1) that the heat of fusion of PVDF in the composite decreases with BT content. The decrease in the heat of fusion indicates a lower extent of crystallinity.
TABLE L Compositions and thermal properties of PVDF and composites. BT PVDF [T.sub.m] [T.sub.c] [DELTA][H.sub.m](J/g) (%) (%) 0 100 176.89 152.45 32.22 0.5 99.5 175.51 151.94 24.5 10 90 172.96 151.92 11.77 30 70 171.82 150.15 2.59 BT [X.sub.c] (%) 0 31 0.5 24 10 12.7 30 3.6
The investigation of the effect of BT on the glass transition temperature of the composites [T.sub.g] shows that [T.sub.g] is slightly increased with BT content, for example, [T.sub.g] for pure PVDF is - 38[degrees]C while its value for [(BT).sub.20][(PVDF).sub.80] is -35[degrees]C. This increase in [T.sub.g] could be attributed to the arrest of the segmental motion of BT/ PVDF interface with its neighborhood (30). This is consistent with the conclusion of Kulek et al. (1) that the activation energy of the segmental motion of the polymer matrix is increased with BT content.
A similar trend is also observed in the crystallization behavior of the composite Fig. 6. The results show that BT serves as an antinucleation agent for PVDF, enabling PVDF to crystallize at a slightly lower temperature upon cooling. The melting temperature [T.sub.m], crystallization temperature [T.sub.c], and heat of fusion [DELTA][H.sub.m] of the composite samples are summarized in Table 1, and [DELTA][H.sub.m], was calculated from the areas of the melting peaks in Fig. 5. The degree of crystallinity ([X.sub.c]) was calculated by
[X.sub.c] = [([DELTA][H.sub.m])/([DELTA][H.sub.0] x [X.sub.m])] x 100% (2)
where [DELTA][H.sub.0] is the heat of fusion of 100% crystalline PVDF (102.5 J/g) (31) and [X.sub.m] is the weight fraction of PVDF in the composite.
This result is consistent with the results of Fang et al. (32) who demonstrated that the addition of a small amount of BT powders into the poly(vinyledene fluoride-trifluoroethylene) matrix inhibits the growth of the crystallite size and causes reduction in the crystalline content and a loosely packed molecular chain structure.
Both the X-ray diffractometry and DSC results indicate that the PVDF remained in the [alpha]-phase in the [(BT).sub.x] [(PVDF).sub.1OO-x] composite, regardless the BT ratio. The hindrance to the PVDF crystallization becomes more and more serious with the increase of BT ratio in the composite.
Tetragonality of BT nanoparticles in [(BT).sub.x] [(PVDF).sub.1OO-x] composite increases with the increase of BT ratio up to 30%, where it shows a saturation value.
Stretching the ([(BT).sub.x] [(PVDF).sub.1OO-x] composite samples enhances tetragonality.
Double-melting endotherms and crystallization exotherms are seen in the melting and crystallization behavior of the composite samples due to reorganization process.
The crystallization of pure PVDF was diminished by the inclusion of BT, and the melting temperature of the composite decreases with the increase of BT ratio in the composite. It seems that BT acts as an antinucleating agent. The decrease of the heat of fusion of PVDF in the composite with the increase of BT ratio could be attributed to the changes taken place in the crystallinity degree of PVDF in the composite.
[T.sub.g] of the [(BT).sub.x] [(PVDF).sub.1OO-x] composite samples is slightly increased with the BT content.
(1.) J. Kulek, I. Szatraniak, B. Hilczer, and M. Polomska, .1. Non-Cryst. Solids, 353, 4448 (2007).
(2.) Y.H. Son, S.J. Kim, Y.M. Kim, G.E. Jang, M.S. Yoon, S.C. Ur, S.L. Ryu, and S.Y. Kweon, Integr. Ferroelectr., 98, 183 (2008).
(3.) M.G. Broadhurst, G.T. Davis, J.E. McKinney, and R.E. Collins, J. Appl. Phys., 49, 4992 (1978).
(4.) A.J. Lovinger, Macromolecules, 15, 40 (1982).
(5.) B.E. Elrnohajir and N. Heymans, Polymer. 42, 5661 (2001).
(6.) R. Gregorio. Jr. and E.M. Ueno, J. Mater. Sci., 34, 4489 (1999).
(7.) B.J. Jungnickel, "Poly(vinylidene fluoride) (overview)" in Polymeric Materials Ham/hook, J.C. Salamone, Ed., CRC Press, New York, 7115 (1999).
(8.) Z.M. Dang, L. Wang, Y. Yin, Q. Zhang, and Q.Q. Lei, Mr. Mater., 19, 852 (2007).
(9.) W. Kanzig, Ferroelectrics, 74, 285 (1987).
(10.) H. Xu and L. Gao. J. Am. Ceram. Soc.. 86, 203 (2003).
(11.) S.W. Kwon and D.H. Yoon, Ceram. hit., 33, 1357 (2007).
(12.) K. Prasad, A. Prasad. K.P. Chandra, and A.R. Kulkarni, Integr. Ferroelectr., 117, 55 (2010).
(13.) R. Sekar, A.K. Tripathi, and R.K.C. Pillai, Mater. Sri. Eng., B5, 33 (1989).
(14.) C. Muralidhar and P.K.C. Pillai, .1. Mater. Sri., 23, 410 (1988).
(15.) C. Muralidhar and P.K.C. Pillai, Mater. Res. Bull.. 23, 323 (1988).
(16.) Z.M. Dang, H.Y. Wang, Y.H. Zhang, and J.Q. Qi, Macromol. Rapid Commun., 26, 1185 (2005).
(17.) H.1. Flsiang, K.Y. Lin, F.S. Yen, and C.Y. Hwang, J. Mater. Sri., 36, 3809 (2001).
(18.) B. Mohammadi, A.A. Yousefi, and S.M. Bellah, Polym. Test., 26, 42 (2007).
(19.) C.W. Nan, Plays. Rev. B, 63, 176201 (2001).
(20.) R. Gregorio Jr. and M. Cestari, J. Po/ym. Sci., Part B: Polym. Phys., 32, 859 (1994).
(21.) A. Salimi and A.A. Yousefi, Polym. Test., 22, 699 (2003).
(22.) F. Baeten, B. Derks, W. Coppens, and E. van Kleef, .1. Eur. Ceram. Soc., 26. 589 (2006).
(23.) Z.M. Dang, H.P. XII, and H.Y. Wang, Appl. Pins. Lett.. 90, 012901 (2007).
(24.) T. Ahmd, G. Kavitha, C. Narayana, and A.K. Ganguli, .1. Mater. Res., 20, 1415 (2005).
(25.) T. Ohno, D. Suzuki, H. Suzuki, and T. Ida, J. Soc. Powder Technol., 41, 86 (2004) in Japanese, translated by Kona Edit. Commit., Kona, 22, 195 (2004).
(26.) A.J.C. Wilson, Proc. Phys. Soc. Lend., 80, 286 (1962).
(27.) X. Li and W.H. Shih, J. Am. Ceram. Soc., 80, 2844 (1997).
(28.) D.J. Lin, C.L. Chang, F.M. Huang, and L.P. Cheng. Polymer, 44. 413 (2003).
(29.) B. Wunderlich, Thermal Characterization of Polymeric Materials, 2nd ed., E.A. Turi, Ed., Academic Press, New York. 205 (1997).
(30.) Y.H. Yu, C.Y. Lin, J.M. Yeh, and W.H. Lin, Polymer, 44, 3553 (2003).
(31.) W.T. Mead, A.E. Zacharidas, T. Shimada, and R.S. Porter, Macromolecules, 12, 473 (1979).
(32.) F. Fang, W. Yang, M.Z. Zhang, and Z. Wang, Compos. Sci. Technol., 69, 602 (2009).
Correspondence to; M.Y.F. Elzayat; e-mail: firstname.lastname@example.org Somya El Sayed is currently at Department of Physics, Faculty of Education, Al Tail" University, Al Khurmah, Kingdom of Saudi Arabia.
M.Y.F. Elzayat, (1) S. EI-Sayed, (1) H.M. Osman, (2) M. Amine (2)
(1) Department of Physics, Faculty of Science, Fayoum University, Fayoum, Egypt
(2) Department of Physics, Faculty of Science, Cairo University, Giza, Egypt
|Printer friendly Cite/link Email Feedback|
|Author:||Elzayat, M.Y.F.; El-Sayed, S.; Osman, H.M.; Amin, M.|
|Publication:||Polymer Engineering and Science|
|Date:||Sep 1, 2012|
|Previous Article:||Effect of injection speed on phase morphology, crystallization behavior, and mechanical properties of polyoxymethylene/poly(ethylene oxide)...|
|Next Article:||Processing and mechanical properties of thermoplastic composites based on cellulose fibers and ethylene-acrylic acid copolymer.|