MICROSTRUCTURE AND ELECTRICAL PROPERTIES

Transcription

1 MICROSTRUCTURE AND ELECTRICAL PROPERTIES OF MICROWAVE-SINTERED PTC THERMISTORS Ming Fu 1, Dinesh Agrawal 2 and Yi Fang 2 1 Department of Electronic Science and Technology, Huazhong University of Science and Technology, China 2 Materials Research Institute, Pennsylvania State University, University Park, PA, USA BaTiO 3 based positive temperature coefficient (PTC) thermistors have been prepared by microwave sintering. The sintering process was carried out in a 2.45 GHz multimode microwave tube furnace without any susceptor. The heating behavior of the PTC material, infl uence of processing conditions on microstructure, and electrical properties of the sintered samples were investigated. The sintering temperature range varied from 1275 C to 1350 C. The optimum density and microstructure were achieved between 1300 C to 1325 C. Samples were also microwave sintered at 1300 C for various soaking times from 10 min to 90 min. It was found that soaking between 30 min to 60 min generated a uniform microstructure. Cooling rate varied from 7.5 C/min to 2 C/min. The study showed that the cooling rate has an infl uence on PTC effect but not on microstructure. Slower cooling rate is essential for achieving better PTC effect. Compared to conventional method, the microwave-processed samples exhibited higher density and better PTC effect. The relative density of microwave sample was 93.5%, compared to 90.2% of the conventional sample. The R 25, α, β, T T, and R of the microwave sintered samples are 17.6 Ω, c max 15.8, , 88.0 C, and 11.5 MΩ, respectively, compared to 12.1 Ω, 14.6, , 88.1 C, and 1.3 MΩ, respectively, for the conventional sample. Submission Date: October 14, 2006 Acceptance Date: February 23, 2007 Publication Date: March 5, 2007 INTRODUCTION BaTiO 3 based positive temperature coefficient (PTC) thermistors are widely used as over-current protectors, motor starters for refrigerator and air conditioners, temperature sensors, and so on. This kind of device is fabricated with BaTiO 3 and SrTiO 3 as raw materials with some dopants such as Y 2 O 3, SiO 2, Mn(NO 3 ) 2, etc. The PTC effect is significantly affected by processing conditions such as sintering temperature, soaking time, Keywords: thermistors, microwave sintering, microstructure, electric properties, BaTiO 3 cooling rate and other factors. The electrical characteristics of PTC resistors, such as voltage with-standing ability, large current impact property, and PTC aging behavior, are all affected by the microstructure of PTC resistors. Conventionally, PTC thermistors are sintered in a resistant furnace or tunnel kiln at C for minutes. It is usually a high-energy, intensive process. Microwave energy has been widely used to process ceramics and other materials [Agrawal, 1998; Agrawal, 2005]. Compared with conventional processes, a microwave process has some potential advantages such as high heating rates, shorter sintering time, lower sintering International Microwave Power Institute 133

2 temperature, fine microstructures, and energy savings [Agrawal et al., 2005; Agrawal, Cheng and Roy, 2004]. Microwave sintering of some BaTiO 3 based materials has been reported. For example, microwave sintering of BaTiO 3 - based multilayer ceramic capacitors (MLCC) was reported by Fang et al. [2005]. Microwave processing of lead-free BaTiO 3 piezoelectric ceramics was investigated by Takahashi et al. [2006]. Dielectric properties of microwave sintered rare-earth doped barium titanate cramics was reported by Mahboob et al. [2006] and microwave calcination of barium strontium titanate dielectric materials was studied by Selmi et al.[1992]. In this paper, the feasibility of microwave processing of PTC thermistors was investigated. The influence of processing conditions on microstructures and electrical properties of PTC thermistors was studied in detail. EXPERIMENTAL A 2kW, 2.45GHz industrial tube microwave furnace was used to sinter the PTC samples. The green PTC specimens mainly consisting of BaTiO 3 and SrTiO 3, and were circular pellets, 10.1 mm in diameter and 2.85 mm in thickness. For each run, five stacked pellets were used. All samples were sintered in ambient atmosphere without any microwave susceptor. Temperature was measured with an infrared pyrometer (Raytek model Rayma2SCCF) with a threshold at 400 C. In order to test the heating behavior of the PTC materials, the output power was set at a constant 2 kw during the entire microwave heating process. The first group of samples were sintered at various temperatures from 1275 C to 1350 C, with a 60-minute soaking time and a cooling rate of 7.5 C/min. The cooling rate was controlled from sintering temperature to 800 C. Below 800 C, samples were allowed to cool naturally to room temperature. The second of group samples were sintered for different soaking times ranging from 10 min to 90 min, while the sintering temperature was 1300 C and cooling rate was 7.5 C/min. The third group of samples were sintered at different cooling rate from 7.5 C/min to 2 C/min, while sintering temperature was 1300 C and soaking time was 10 min. For comparison, some samples were also sintered by conventional method at 1300 C/60min with the cooling rate at 7.5 C/min. Scanning electron microscopy (SEM) was used to examine the microstructure of the sintered samples. Powder X-ray diffractometry (XRD) was used to identify phase composition of the samples sintered both by microwave and conventional methods. Samples were screenprinted with two layer electrodes: the bottom layer was an ohmic silver-zinc electrode and the top layer was a silver electrode. The R-T curves of PTC elements were measured with an automated PTC resistance-temperature measuring system. RESULTS AND DISCUSSION Microwave heating behavior of PTC PTC samples were heated by microwave energy at a constant power output from room temperature to the sintering temperature, without any pre-heating microwave susceptor, thus it was purely microwave heating. Figure 1 exhibits a typical microwave heating profile of the PTC samples. Due to the threshold of the IR pyrometer, no data was recorded below 400 C. The heating curve shows that the PTC material absorbed microwave energy very well. The changing slope indicates that heating rate rose with temperature. The heating rate was 30 C/min at 400 C, and it rose to 56 C/min at 1200 C. The thermal run-away was not encountered during the microwave heating process. 134 Journal of Microwave Power & Electromagnetic Energy Vol. 40, No. 3, 2007

3 during heating, the εʹ and tan δ of PTC materials increase rapidly with temperature, and this accelerates the absorption of microwaves [Agrawal et al., 2004]. The great changes of dielectric constant and loss tangent explain why the heating rate of the PTC samples rose with temperature. Influence of sintering temperature on microstructure Figure 1. Heating curve of PTC samples during microwave heating The power adsorbed per unit volume in a dielectrically lossy material, P (W/m 3 ), is given by [Sutton, 1989] P = σ E 2 = 2πfπ ε 0 ε tan δ E 2 (1) where E (V/m) is the magnitude of internal field, ε 0 is the permittivity of free space, εʹ is the relative dielectric constant, f is the frequency (GHz), and σ is the total effective conductivity (S/m) caused by conduction and displacement currents. Equation (1) shows that the power absorbed varies linearly with frequency, relative dielectric constant, and tan δ, and varies with the square of electric field. PTC materials mainly consist of BaTiO 3 and SrTiO 3. The εʹ and tan δ of PTC green pellets were measured to be and 0.012, respectively, at 10 khz. The εʹ and tan δ of PTC ceramics (as-sintered by microwave energy) were measured to be and 261.6, respectively, at 10 khz. Therefore, The SEM micrographs of samples sintered at different temperatures are shown in Figure 2. Sintering temperature varied from 1275 C to 1350 C, while the soaking time was kept unchanged at 60 min and the cooling rate at 7.5 C/ min. Clearly, the sample sintered at 1275 C has more pores than other samples. The microstructures of samples sintered at 1275 C and 1350 C are not as uniform as the others. Samples sintered at 1300 C and 1325 C are very dense and uniform. The density of samples (a), (b), (c), and (d) are 5.07, 5.32, 5.36, and 5.18g/cm 3, respectively. Therefore, 1275 C is too low to prepare high density PTC ceramics, while 1350 C is too high, resulting in a non-uniform microstructure. The optimal temperature range for microwave sintering the current PTC materials is between 1300 C and 1325 C. Effects of soaking time on microstructure Samples were processed at various soaking times from 10 min to 90 min, while the sintering temperature was kept constant at 1300 C (a) (b) (c) (d) Figure 2. SEM micrographs of samples sintered at different temperatures (a) 1275 C, (b) 1300 C, (c) 1325 C, (d) 1350 C. International Microwave Power Institute 135

4 (a) (b) (c) (d) Figure 3: SEM micrographs of PTC samples microwave-sintered for (a) 10min, (b) 30min, (c) 60min, and (d) 90min. (a) (b) (c) (d) Figure 4. SEM micrographs of samples microwave processed at (a) 7.5 C/min, (b) 3.75 C/min, (c) 2.5 C/min, and (d) 2 C/min. with the cooling rate at 7.5 C/min. The SEM micrographs of samples sintered at different soaking times are shown in Figure 3. The results indicate that the sample sintered for 10 min still had some porosity and that sintered for 90 min had a secondary grain growth. Samples sintered at 30 min and 60 min were uniform and highly dense. The densities of samples (a), (b) (c) and (d) were 5.10, 5.27, 5.32, and 5.43g/cm 3, respectively. Therefore, the optimal soaking time for microwave processing PTC is 30 to 60 min. Effects of cooling rate on microstructures and PTC effect The SEM micrographs of samples processed at different cooling rates are shown in Figure 4. Samples sintered for 10 min at 1300 C were cooled at 7.5 C/min, 3.75 C/min, 2.5 C/min, and 2 C/min, respectively. SEM shows that cooling rate did not significantly affect microstructure, but the porosity reduced with the increase of cooling time. The sintered densities of samples (a), (b), (c) and (d) are 5.10, 5.25, 5.30, and 5.34 g/cm 3, respectively. The PTC characteristics of samples sintered at different cooling rates are shown in Figure 5. The main parameters of R-T curve is R 25, T c, α T, and resistance jump orders (β). The resistance temperature coefficient (α T ) calculated with equation 2 [Zhou, 1990]: lg R = lg R p b α T Tp Tb (2) where T p = T c +25, T b = T c +15, R p is the resistance at T p, and R b is the resistance at T b. Curie temperature (T c ) is the temperature when R equals 2 times the room temperature resistance (R 25 ). β is the resistance rise ratio defined by β = R Rmax min (3) where R max is the maximum of sample resistance and R min is the minimum of sample resistance. The electrical properties of sample (a), (b), (c) and (d) are shown in Table 1. The results show that with the increase of cooling time, PTC effect is enhanced characterized by α T and β rise with the cooling time. According to Heywang [1961], the resistivity 136 Journal of Microwave Power & Electromagnetic Energy Vol. 40, No. 3, 2007

5 Table 1: Electrical properties of samples at different cooling rates. Specimen sample a sample b sample c sample d Electrical properties (7.5 C/min) (3.75 C/min) (2.5 C/min) (2.0 C/min) R 25 (Ω) α T T c ( C) β ρ at the cubic phase is given by and ϕ ρ = ρ 0 exp kt (4) 2 2 ϕ = e n s 8 ε0εnd (5) where n s is the concentration of trapped electrons at the grain boundary acceptor-type states (cm -2 - ), N the donor concentration (cm d 3 ), e the electron charge, and ε the dielectric constant. Since the dielectric constant at the cubic phase obeys Curie-Weiss law ε=c/(t- T 0 ), Eqs. (4) and (5) explain the PTC character of samples, i.e., resistivity increases several orders with rising of temperature. The slower cooling rate accelerates a higher barrier φ, thus the PTC effect is enhanced. Therefore, a slower cooling rate should to be chosen when PTC is sintered by microwave energy, as long as the R 25 of samples matches the requirements of the specified applications. Comparison of microwave and conventionally processed samples SEM shows that the conventional sample has more porosity than the microwave sample. The relative density (with respect to the theoretical density is 5.7g/cm 3 for PTC ceramics) of the microwave sample was 93.5%, compared to 90.2% of the conventional sample. PTC Effect The R-T curves of both the microwave and conventionally processed samples are shown in Figure 7. The R-T characteristics show that Figure 5: R-T curves of samples at different cooling rates: (a) 7.5 C/min; (b) 3.75 C/min; (c) 2.5 C/min; and (d) 2 C/min. Microstructure The SEM micrographs of samples sintered by conventional and microwave methods are shown in Figure 6. Both samples were sintered at 1300 C in 60 min, and cooled at 7.5 C/min. a) b) Figure 6: SEM micrographs of samples sintered by (a) conventional and (b) microwave process. International Microwave Power Institute 137

6 Figure 7: R-T curves of microwave and conventional samples. Figure 8: XRD patterns of microwave and conventional samples. R 25, α T,β, T c, and R max of the microwave sample are 17.6 Ω, 15.8, , 88.0 C, and 11.5 MΩ, respectively, compared to 12.1 Ω, 14.6, , 88.1 C, and 1.3 MΩ, respectively, for the conventional sample. The experimental results show that under the same processing conditions, the microwave samples have better PTC effect than that of conventional ones. This is mainly attributed to the better grain boundary structure of the microwave sample as shown in Figure 6(b) [Hrovat et al., 2007]. Phase identification The XRD results are shown in Figure 8. XRD patterns show that the samples sintered by both microwave and conventional methods have the same major phase composition, Ba 0.70 Sr 0.30 TiO 3, indicating that microwave processing improved microstructure and PTC property, but did not change the composition of the sintered samples. CONCLUSIONS PTC materials can be easily heated by microwave energy at 2.45 GHz with out using any susceptor. Compared to conventional process, microwave sintering achieved better microstructure and PTC effect as characterized by higher temperature coefficient and resistance rise ratio. Both the microwave and conventional samples have the same main phase. The optimal processing parameters for microwave sintering PTC were found to be in the temperature range of 1300 C-1325 C and soaking time minutes, with a cooling rate as low as possible. ACKNOWLEDGEMENTS The authors acknowledge Shenzhen Ampron Sensitive Components Co. Ltd. for providing PTC green pellets, and Wuhan Supernano Optoelec Technology Co. Ltd. for providing silver electrode pastes for this study. The partial financial support by the Microwave Processing and Engineering Center, Penn State University is gratefully acknowledged. REFERENCES Agrawal,D.K. (1998). Microwave Processing of Ceramics: A Review., Current Opinion in Solid State & Mat Sci, 3(5), pp Agrawal, D., J. Cheng and R. Roy (2004). Microwave Sintering,Brazing and Melting of MetallicMaterials, 138 Journal of Microwave Power & Electromagnetic Energy Vol. 40, No. 3, 2007

7 Proc. Intl. Symp. On Heating by Electromagnetic Sources, pp.717. Agrawal, D., J. Cheng, M. Jain, G. Skandan, R. Dowding, K. Cho, B. Klotz and D. Kapoor (2004). Microwave Sintering of Tungsten and Alloys, 8th Intl Conf. Sci. Hard Mats., pp Agrawal,D. (2005). "Microwave Sintering Developments Spur Emergence of New Materials & Technologies," Industrial Heating, 72(6), pp Agrawal,D., Cheng, J., Fang, Y. and Roy, R. (2005). Microwave Processing of Ceramics, Composites and Metallic Materials, Am. Cer. Soc. Publ., pp Fang, Y., Lanagan, M.T., Agrawal, D.K., Yang, G.Y., Randal, C.A., Shrout, T.R., Henderson, A., Ramdall, M., and Tajuddin, A. (2005). An Investigation Demonstrating the Feasibility of Microwave Sintering of Base-Metal-Electrode Multilayer Capacitors, J. Electroceramics, 15, pp Heywang, W. (1961). "Barium Titanate as a Seimiconductor with Blocking Layers." Solid State Electron., 3(1), pp Hrovat, M., Belavi, D., Kita, J., Holc J., Cilenšek, J., Golonka L. and Dziedzic, A. (2007). Thick- Film PTC Thermistors and LTCC Structures: The Dependence of the Electrical and Microstructural Characteristics on the Firing Temperature, J. European Ceramic Society, 27(5), pp Jida, S., Suemasu, T. and Miki, T. (1999). Effect of Microwave Heating on BaTiO 3 : Nb Ceramics with Positive Temperature Coefficient of Resistivity, J. Appl Phy, 86(4), pp Mahboob, S., Dutta, A.B., Prakash, C., Swaminathan, G., Suryanaryana, S.V., Prasad, G., and Kumar, G.S. (2006). Dielectric Behavior of Microwave Sintered Rare-Earth Doped BaTiO3 Ceramics, Materials Science and Engineering: B, 134, pp Selmi, F., Guerin, F., Yu, X.D., Varadan, V.K, Varadan, V.V., and Komarneni, S. (1992). Microwave Calcination and Sintering of Barium Strontium Titanate, Materials Letters, 12(6), pp Sutton, H. W. (1989). Microwave Processing of Ceramic Materials., Ceramic Bulletin, 68(2), pp Takahashi, H., Numamoto, Y., Tani, J., Matsuta, K., Qiu J., and Tsurekawa, S. (2006). Lead-Free Barium Titanate Ceramics with Large Piezoelectric Constant Fabricated by Microwave Sintering, Jpn J. Appl. Phy., 45(1), pp. L30-L32. Zhou, D. X. and S. P. Gong (1990). PTC Materials and Applications. Huazhong University of Science and Technology, Wuhan. International Microwave Power Institute 139