2016 Volume 63 Issue 7 Pages 663-667
Sm-doped CeO2 ceramics were prepared by conventional solid state method. The electrical conductivity of the samples was measured under conventional heating and 24 GHz millimeter-wave (MMW) heating, and the results were compared. During MMW heating, 5 mol% Sm-doped CeO2 pellets with different thickness were used as susceptors, meanwhile Al2O3 fiber board was used as thermal insulator and its design varies by the number of open channels. The aforementioned susceptors and thermal insulators dependency of conductivity values during MMW heating were studied. Conductivity of samples under MMW heating was found to be higher than under conventional heating. Results showed that the different susceptor thickness and thermal insulator design could result in different conductivity values. These results were attributed to the phenomenon of heat dissipation from surface and the amount of direct radiation reached the sample. Specific susceptor thickness and specific insulator design which leads to highest conductivity were identified. By combining these two effects, the largest enhancement of conductivity of ceria based ceramics was obtained.
Microwave heating has become an alternative heating source as it offers many advantages such as rapid, volumetric and selective heating, which result in reduction of time, energy and cost, improvement of heating uniformity and product performance1–4). Due to these benefits, interest in microwave studies increase continuously in many areas including ceramics processing. Microwave has been used as heating source to replace conventional heating in ceramics sintering, drying, joining, fabrication, creep deformation, and so on, and significant improvement in ceramics properties such as in microstructure, density and strength have been reported in literatures3,4). Kishimoto et al. have compared electrical conductivity of stabilized zirconia under conventional and millimeter-wave heating, and result shows that the conductivity value is higher for the latter5).
During microwave processing, temperature control is vital aspect in order to optimize microwave effect. The use of thermal insulator and susceptor become essential in any microwave processing. Commonly the role of susceptor is as an external heat generator especially when using microwave with lower frequency. Alumina fiber board is a typical thermal insulator used in microwave processing to control the heat loss from the load. Few reports discussing the importance of thermal insulator and susceptor in terms of materials and configuration can be found in the literature6,7). Thus, careful consideration on susceptor and thermal insulator should be given in microwave processing.
Among the fuel cells, solid oxide fuel cell, SOFC holds the greatest potential as power generator due the highest efficiency. SOFC which employs well-known stabilized zirconia as electrolyte has standard operating temperature of 1073.15~1273.15 K. High operation temperature leads to cell mechanical and chemical properties issues8–11). Thus, recent works on SOFC are mainly focus on lowering the operation temperature to intermediate temperature, IT. A promising candidate for IT-SOFC electrolyte is CeO2 which has fluorite structure. It is reported that doped CeO2 exhibits higher conductivity than YSZ12,13). Unfortunately, there is contribution from electron to the conductivity of doped CeO2 which can lower the cell performance. However, from a viewpoint of higher conductivity at lower temperature, doped CeO2 is preferable. The aim of this work was to find the optimum thermal environment for in-situ electrical conductivity measurement of samarium-doped ceria (abbreviated to SDC) under 24 GHz MMW heating.
Ce1−xSmxO2−x/2 (x = 0.10, 0.15, 0.20, 0.25; abbreviated to 10SDC, 15SDC, 20SDC, 25SDC respectively) were prepared via conventional solid state method by using cerium nitrate hexahydrate (Ce(NO3)3·6H2O) (Nacalai Tesque, Inc., Japan) and samarium nitrate hexahydrate, 99.9 % (Sm(NO3)3·6H2O) (Kishida Chemical Co. Ltd., Japan) as raw materials. Required precursor powders were mixed and stirred with rotary speed of 360 rpm in ethanol for 1 hour at room temperature. Obtained slurry was heated under sodium lamp to remove the liquid phase, and then the powder was calcined at 1273.15 K for 2 h in argon atmosphere. Next, powder was ball-milled in an ethanol dispersion using planetary type mill (Pulverisette, Fritsch, Germany) for 1 h operated at 200 rpm. Obtained powder was pelletized under uniaxial press at 15 MPa, followed by cold-isostatic press at 125 MPa. Green pellets were sintered at 1823.15 K for 5 h in air with heating and cooling rate of 5 K/min. The densities of samples were determined by Archimedes method. Powder X-ray Diffraction with CuKα was used to identify the crystal phases.
2.2 Impedance measurementComplex impedance measurement of SDCs was measured using two-probe method. Pt paste was applied on both end surfaces of 4.0 × 4.0 × 10.0 mm3 rectangular sample and heated at 1373.15 K for 10 min. Measurement was performed from 673.15~1273.15 K in a frequency range from 0.1 Hz~1 MHz with an applied voltage of 300 mV using impedance analyzer (Modulab Xm, Solatron, UK). For conventional heating, sample was heated using conventional electric tube furnace. Meanwhile for measurement under MMW heating, permanent magnet gyrotron system with frequency of 24 GHz and power of 3 kW was used. Sample arrangement is based on previously developed sample set up as illustrated in Fig. 15). An additional sample which has same composition and dimensions with conductivity measuring sample was used to monitor the temperature. Thermocouple was inserted carefully and positioned inside the additional sample. Samples were sandwiched between susceptors to assist heating and prevent heat radiation. 5SDC was chosen as susceptor as it possesses lower electrical conductivity than samples and has similar MMW absorption properties with samples. The space between the two samples and the susceptors was filled with boron nitride, BN powder served as the electrical insulator between sample and susceptor, and to provide effective thermal transport.
Sample setup for conductivity measurement under millimeter-wave irradiation. a) Top view of fiber board with open channel, b) Front view of sample and susceptor setup.
Firstly, to investigate the optimal heating environment for effective MMW heating, the influence of susceptor thickness and fiber board configuration on the conductivity has been studied. Note that, only configuration of middle fiber board was changed, while other fiber boards remained unchanged. Measurement was performed on 20SDC under nine different environments as tabulated in Table 1. After the optimum environment has been verified, conductivity of SDC samples with different dopant concentration was measured.
| Environment | Upper susceptor thickness, mm | Lower susceptor thickness, mm | Fiber board size* | Open channel on fiber board |
|---|---|---|---|---|
| 1 | 1.0 | 3.5 | Medium | nil |
| 2 | 2.0 | 3.5 | Medium | nil |
| 3 | 3.0 | 3.5 | Medium | nil |
| 4 | 2.0 | 3.0 | Medium | nil |
| 5 | 2.0 | 3.8 | Medium | nil |
| 6 | 2.0 | 3.5 | Small | nil |
| 7 | 2.0 | 3.5 | Large | nil |
| 8 | 2.0 | 3.5 | Medium | 1 |
| 9 | 2.0 | 3.5 | Medium | 2 |
All prepared SDC samples have relative density more than 94 % to the theoretical value and exhibit single phase cubic fluorite structure which is in agreement with JCPDS#34-0394. The results of conductivity of 20SDC under nine different thermal environments are shown in Fig. 2. For comparison, conductivity measured under conventional heating is also plotted here. All measurements under MMW irradiation exhibit higher conductivity value than under conventional heating with noticeable different enhancements. Enhancements of conductivity differ in wide range which is 3 to 9 times showing that the effect of MMW heating is greatly dependence on sample’s surrounding ambience.
Arrhenius plots of 20SDC under different thermal environments.
To give clear view on effect of each environment, results of respective parameters dependence of conductivity in lower temperature range are shown separately in Fig. 3a), 3b), 3c) and 3d). Dashed lines indicate the value of conductivity obtained from conventionally heated sample at each temperature. Firstly, as shown in Fig. 3a) we studied the effect of upper susceptor thickness of 1.0 mm, 2.0 mm and 3.0 mm which corresponding to Environment 1, 2 and 3, respectively. Fig. 3b) shows the result of conductivity dependence on lower susceptor thickness of 3.0 mm, 3.5 mm and 3.8 mm, which corresponding to Environment 4, 5, and 6, respectively. Significant different in conductivity was observed when using various susceptors thickness. In these figures, it is clear that by using upper susceptor with thickness of 2.0 mm, highest conductivity can be obtained, while for lower susceptor, optimum thickness is 3.5 mm. Measurement with thinner or thicker susceptor resulted in lower conductivity. This result indicates that susceptor with specific thickness can maximize the prevention of heat radiation from sample surface thus minimize thermal gradient between surface and inner part of sample, as well as maximize self-heating in sample initiated by MMW irradiation, which consequently leads to effective MMW heating.
Electrical conductivity of 20SDC under different thermal environments: a) various upper susceptor thicknesses, b) various lower susceptor thicknesses, c) various fiber board sizes, d) various fiber board configurations, respectively.
In the case of fiber board thermal insulator, to ensure penetration of MMW energy with negligible decreasing intensity it must be transparent to MMW irradiation. Measurements under Environment 2, 6, and 7 (Fig. 3c)) show no significant change was observed on conductivity value regardless of fiber board size. It can be said that alumina fiber board that had been used in this work is porous enough to allow penetration of MMW energy to reach sample and it also free from impurity that might absorb MMW energy, which can reduce percentage of energy to be absorbed by sample. In fact, prior to measurement, the fiber board was heat-treated at 1273.15 K for 1 h to release unnecessary organic impurities.
Fiber board’s configuration (Environment 2, 8 and 9) plays vital influence on the heating effectiveness. An effective microwave heating requires the configuration of fiber board to maximize energy input from microwave source to the sample and minimize energy output from the sample due to radiation loss. Results from this effect can be seen from Fig. 3d) which show remarkable difference in enhancement especially when fiber board with two open channels was used conductivity dropped drastically when comparing with fiber board without open channel. Fiber board with open channel is thought to increase the MMW energy dose directly onto sample. However this configuration resulted in larger heat loss from the sample. This phenomenon resulted in ineffective millimeter wave heating.
It can be concluded that, combining the optimum susceptor and fiber board which leads to high efficient MMW heating can results in high conductivity. By analyzing results obtained here, we conclude that environment 2 resulted in the highest conductivity value, thus this environment was chosen for further conductivity measurement under MMW irradiation for SDC.
Fig. 4 compares the conductivity of 10SDC, 15SDC, 20SDC and 25SDC measured under conventional and MMW heating. Under conventional heating, conductivity can be ranked as follows; 20SDC > 25SDC > 15SDC > 10SDC. Comparable results were previously reported by Yahiro et al.14). Under MMW heating, conductivity of each sample is higher that under conventional heating. At 673.15 K conductivity of 20SDC is 9 times larger when heating under MMW heating resulted in highest value among the SDCs. Enhancement of 15SDC and 25SDC is 10 times and 3 times, respectively. 15SDC shows larger enhancement and it surpasses the conductivity value of 25SDC under MMW heating. MMW provides additional energy to facilitate ion conduction in crystal structure, hence the conductivity enhancement. In SDC, because conductivity depends strongly on the dopant concentration, lower conductivity of 25SDC is believed due to association of oxygen ion vacancy and dopant cation. When heating under MMW irradiation, for example at 673.15 K for 20SDC, similar value can be obtained only if it is measured at minimum 823.15 K under conventional heating, showing that processing temperature can be lowered up to 150.00 K. By exploiting the benefits of MMW which is favorable in lowering the operation temperature and time, as well as selective heating, MMW reassures that energy and cost can be reduced.
Arrhenius plots of SDCs under conventional and millimeter-wave irradiation heating.
Activation energy, Ea was calculated from the slope of Arrhenius plot in Fig. 4 for each sample and the result is plotted in Fig. 5. Generally, Ea is the summary of migration energy, Em and cation-vacancy association energy, E0. At low temperature region, both Em and E0 governs the conductivity mechanism while at high temperature region, only Em contributes to the conduction as oxygen vacancies are freed from dopant cation. Thus E0 can be calculated by subtract the Em from Ea which is determined from slope from lower temperature region15,16). The Arrhenius plots bend at around 873.15 K which is similar to reported data by Yifeng et al.17). It can be seen from Fig. 5 that both Em and Ea decreased when samples were heated under MMW. Nevertheless, reduction in E0 is larger than that in Em showing that conductivity enhancement is dominated by E0 rather than Em. For instance, in the case of 20SDC which shows largest enhancement, ratio of EMMW to the EConv is 0.80 and 0.26 for Em and E0, respectively. Reduction in apparent activation energy especially E0 is interpreted as an influence from MMW irradiation known as non-thermal effect. Non-thermal effect explains that when molecules absorb energy from MMW, the energy will not immediately convert to heat but instead induce diffusion process, which lead to increase conductivity at lower temperature18,19). Such effect allows electromagnetic field to induced additional driving force which directly exerts the charge carrier known as ponderomotive force, PMF which assist ion-vacancy dissociation. From results obtained in this work, effect of MMW is remarkable as more than two-thirds of E0 is supplied by MMW energy. It is reported by Kishimoto et al. that in stabilized ZrO2 fluorite structure, instead of resulted in lower E0, conductivity measurement under MMW irradiation lead to lowering Em5). Conductivity of 20SDC in this work is found to outperform the conductivity of reported 8YSZ under MMW heating especially at lower temperature region. A comprehensive investigation is required to find the factors which may cause such result such as dielectric property and absorption property of host ion for better understanding.
Activation energy of SDCs under conventional and millimeter-wave irradiation heating. The numbers on the graph are the value of EMMW/EConv.
Thermal environment for conductivity measurement under MMW heating was optimized by changing the susceptor thickness and alumina fiber board configuration. A significant difference results obtained by changing thermal environment shows that such preliminary study is important in MMW process to before continue with further experiment.
All samples shows higher conductivity when heated under MMW irradiation compared to under conventional heating. Conductivity enhancement in SDCs is attributed by reduction in apparent association energy of vacancy-cation rather than migration energy due to MMW non-thermal effect.
This work was partially supported by JSPS KAKENHI Grant-in-Aid for Exploratory Research, Grant Number 2663031806.
