Universidade de São Paulo

2014-12

Field-assisted sintering of undoped BaTiO3:

microstructure evolution and dielectric

permittivity Journal of the European Ceramic Society, London : Elsevier, v. 34, n. 15, p. 3655-3660, Dec. 2014http://www.producao.usp.br/handle/BDPI/50795

Downloaded from: Biblioteca Digital da Produção Intelectual - BDPI, Universidade de São Paulo

Biblioteca Digital da Produção Intelectual - BDPI

Departamento de Física e Ciências Materiais - IFSC/FCM Artigos e Materiais de Revistas Científicas - IFSC/FCM

Available online at www.sciencedirect.com

ScienceDirect

Journal of the European Ceramic Society 34 (2014) 3655–3660

Field-assisted sintering of undoped BaTiO3:Microstructure evolution and dielectric permittivity

Jean-Claude M’Peko a,b,∗, John. S.C. Francis a,c, Rishi Raj a

a Department of Mechanical Engineering, University of Colorado at Boulder, Boulder, CO 80309-0427, United Statesb Grupo Crescimento de Cristais e Materiais Cerâmicos, Instituto de Física de São Carlos, Universidade de São Paulo, Caixa Postal: 369,

CEP: 13560-970 São Carlos, SP, Brazilc President, FAST Ceramics, 3550 Frontier Ave, Unit A, Boulder, CO 80301, United States

Received 19 February 2014; received in revised form 23 April 2014; accepted 27 April 2014Available online 29 May 2014

Abstract

We report, for first time, how electric fields influence the sintering of undoped BaTiO3, a ferroelectric material, and how this process affects themicrostructure and the dielectric properties. Flash sintering is achieved at a furnace temperature of 688 ◦C under a field of 500 V cm−1, producingspecimens that are 94% dense. As a consequence, the grain size is much finer than in conventional sintering, which is shown to influence the Curietemperature and dielectric permittivity. Data obtained at different strengths of the electrical field, and current limits imposed on the specimen arepresented in the form of a “processing map” that separates the safe region, where sintering is uniform, from the fail region, where the current flowin the sample becomes localized. The map illustrates that ceramics can respond by different mechanisms, with the dominant mechanism changingwith the strength of the electrical parameters.© 2014 Elsevier Ltd. All rights reserved.

Keywords: BaTiO3; Flash sintering; Microstructure; Permittivity

1. Introduction

Barium titanate (BaTiO3), a ferroelectric material withperovskite structure, is used in high-permittivity capacitors,transducers, ferroelectric memories and positive temperaturecoefficient (PTC) resistors.1,2 Conventional sintering of bariumtitanate usually requires temperatures ≥1250–1300 ◦C.2,4 Liq-uid phase that may form during sintering can however lead toaccentuated abnormal grain growth.3,5

Hot-isostatic-pressing and spark plasma sintering areexpected to produce microstructures with smaller grainsizes.4,6–9 In this article we report the results from field-assistedsintering where the electrical field is applied directly to an

∗ Corresponding author at: Grupo Crescimento de Cristais e Materiais Cerâmi-cos, Instituto de Física de São Carlos, Universidade de São Paulo, Caixa Postal:369, CEP: 13560-970 São Carlos, SP, Brazil. Tel.: +55 1633739828;fax: +55 1633739824.

E-mail addresses: JeanClaude.MPeko@Colorado.Edu, peko@ifsc.usp.br,jcpeko@yahoo.com (J.-C. M’Peko).

otherwise freely sintering specimen placed in a conventionalfurnace. This process, also known as flash sintering, has beenshown to enhance the sintering kinetics of several materials.10–16

We find that this process leads to dense, fine-grained andhigh-permittivity ceramics of barium titanate. The furnace tem-peratures and process times are considerably below conventionalsintering.

2. Experimental procedure

BaTiO3 was synthesized by a hydrolysis method usingBa(OH)2 and Ti alkoxide, and calcined at 700 ◦C for 2 h, result-ing in a high-purity (>99.98%) powder with estimated specificsurface area of 16 m2/g, average particle size of 70 nm (withina narrow distribution of ±8 nm) and a Ba/Ti ratio of 1.002. Forsample preparation, the powders were cold-pressed at a uniaxialpressure of 280–300 MPa in a dogbone-shaped die, giving greenbodies with a gage section length of 20 mm, a rectangular crosssection of 3.5 mm × 1.6–2.0 mm, and green density ranging herefrom 60 to 64% of theoretical density (TD = 6.02 g/cm3). A

http://dx.doi.org/10.1016/j.jeurceramsoc.2014.04.0410955-2219/© 2014 Elsevier Ltd. All rights reserved.

3656 J.-C. M’Peko et al. / Journal of the European Ceramic Society 34 (2014) 3655–3660

drawing of the dogbone-like sample can be found in Ref. 14.Sintering was carried out in a molybdenum di-silicate furnace,in ambient air. Shrinkage was measured with a CCD camera thatrecorded photographs of the specimen at approximately 1 s inter-vals. The linear sintering strain was calculated from ∈ = ln(L/L0),where L is the time dependent length of the gauge section, and L0its initial value. Assuming shrinkage to be isotropic, the relativedensity of the specimens was then obtained from the expressionln(�/�0) = 3| ∈ |,17 where �0 is the green density at the start ofthe sintering experiment and � is the density during the sinteringprocess.

Electric field-assisted sintering experiments were carried outby following the procedure described in Ref. 14. The speci-men was suspended into the furnace with platinum wires whichalso served as the field and current carrying electrodes. Thefurnace temperature was raised with a constant heating rate of10 ◦C min−1, and the voltage was applied to the specimen whenthe furnace reached 500 ◦C. The current flowing through thespecimen increased abruptly at the onset of the flash, at whichpoint the power supply was switched (from voltage) to currentcontrol; this led to a fall in the voltage as the conductivity of thespecimen continued to increase, as discussed in earlier papers,see e.g., Ref. 15. The experiments were usually ended a few min-utes after the onset of the flash. In all these experiments (witha couple of exceptions described later) the current was set to alimit of 60 mA, which corresponds here to a current density ofabout 9.3 mA mm−2.

To get further information on conductivity behavior duringsintering, experiments were also carried out at isothermal fur-nace temperature. The specimens were heated up to a giventemperature in the 650–1100 ◦C range, followed by a step-wiseapplication of the electric field. A wide range of fields and cur-rent density limits were tested. The experiments were ended lessthan 1 min after the onset of the flash phenomenon.

Conventional sintering experiments were carried out at a con-stant heating rate (also 10 ◦C min−1), without the application ofan electric field (called the 0 V experiment). In this instance, thespecimen had to be heated to 1300 ◦C to achieve a minimum den-sity of 97%. For microstructure comparison with field-assistedsintering, conventional sintering at 1300 and 1350 ◦C was alsoconsidered.

The grain size of the specimens was measured over an area ofabout 3.0 mm × 1.5 mm. The sample was polished down to a fin-ish of 1 �m, followed by thermal etching at 1100 ◦C for 30 minin ambient air. The specimens were sputter coated with 2.5 nmof Au–Pd, and imaged in a field-emission scanning electronmicroscope (JEOL JSM-7401F). The micrographs were digi-tally analyzed and values of average grain size were estimatedusing the linear intercept method, as described elsewhere,18 inaccordance with ASTM standards.19 To assess the possibility ofgrain growth during thermal etching, as-sintered samples werealso fractured and the grain size estimated from SEM images ofthe fracture surfaces, and then compared with the values fromthe polished and etched cross-sections.

For electrical measurements, the specimens were cut intorectangular blocks and finished with a surface grinder usinga 300 grit diamond wheel, resulting in samples with area of

600 750 900 1050 1200 1350

-0.16

-0.12

-0.08

-0.04

True

Lin

ear S

train

( oC )

E (V cm-1) 0 150 200 250 300 375 425 500 1000

Furnace Temperature

0

Fig. 1. Temperature dependence of linear shrinkage strain for the conventionaland many electric field-assisted sintering experiments conducted on BaTiO3

specimens, with a current limit of 60 mA (about 9.3 mA mm−2).

9.7–12.3 mm2, and thickness of 1.0–1.5 mm. Platinum elec-trodes were painted onto both major surfaces, followed by dryingat 550 ◦C for about 10 min. An impedance analyzer (HP 4192A)was used for obtaining permittivity data from room temperatureto about 310 ◦C at 100 kHz.

3. Results and discussion

Fig. 1 shows the behavior of strain for the conventional andseveral electric field-assisted sintering experiments. The resultsare typical of flash sintering behavior where the electric fieldreduces considerably the (furnace) sintering temperature. Thepower dissipation and physical density from a few experimentsare depicted in Fig. 2. Note the peak in power dissipation fol-lowed by a decline in its magnitude, which is a characteristicof this phenomenon.15 There is however a subtle differencebetween the results at 150 V cm−1, where the onset of the flashregime occurs after the sample has sintered to a considerabledegree, and those above 250 V cm−1, where the onset of flashcoincides with sintering.

Such behavior has been reported before, and reflects twodifferent mechanisms of field-assisted sintering. The gradual

0.0

0.2

0.4

0.6

600 720 840 960 1080 1200 1320

60

70

80

90

100

V cm -1

0 V

500 375250

150

P (

W /

m m

3 )

( b )

ρ (

% T

D )

( a )

Current LImit = 9.3 mAmm–2

Furnace Temperature ( oC )

Fig. 2. Temperature dependence of (a) power dissipation and (b) density, allwith the current limit of 60 mA (about 9.3 mA mm−2).

J.-C. M’Peko et al. / Journal of the European Ceramic Society 34 (2014) 3655–3660 3657

Table 1Sintering conditions, density (ρ), in percentage of theoretical density (TD), andflash sintering temperature (TFLASH), where applying, of BaTiO3 specimensconventionally and flash sintered.

Sintering conditions ρ (% TD) TFLASHb (◦C)

1350 ◦C/1 ha 98.4 n/a1300 ◦C/5 mina 97.2 n/a150 V cm−1/60 mA/15 min 97.1 900200 V cm−1/60 mA/3 min 94.3 825250 V cm−1/60 mA/3 min 94.4 822300 V cm−1/60 mA/3 min 95.6 760375 V cm−1/60 mA/2 min 95.3 748425 V cm−1/60 mA/2 min 94.5 719500 V cm−1/60 mA/1 min 94.2 6881000 V cm−1/60 mA/1 min 91.6 612

a Conventional sintering.b Furnace temperature.

enhancement in sintering has been called FAST or Type A sin-tering, and the nearly stepwise sintering as flash or Type Bbehavior.10–13 The first has been explained in terms of reducedrate of grain growth under electric field. The underlying mecha-nism for flash sintering remains so far unclear. Both Joule heatingand electric field-induced generation of defects have been pro-posed as the underlying mechanism for flash sintering.15 Weshall comment further on this issue later in this article.

The experimental parameters and the results from the presentexperiments, shown in Fig. 1, are summarized in Table 1. Theyshow that high-density BaTiO3 electroceramics can be madein this way at furnace temperatures that are 500 ◦C, or more,below the temperature required in conventional sintering. This iscertainly a consequence of enhanced charge and mass transportprocesses under electric field action.

The conductivity data of various specimens calculatedfrom the instantaneous values of the current, with the powersupply set under voltage control, are shown in an Arrhe-nius plot in Fig. 3. The gradual increase in conductivity for100 V cm−1 and 150 V cm−1 is attributed to Joule heating ofthe specimens, but the nearly “vertical” rise in conductivity forE ≥ 375 V cm−1 results from a transition to flash sintering. Thecase of 250 V cm−1 represents the change from nominal to flashbehavior. The dotted line shows the Arrhenius fit to the nomi-nal data for conductivity, by considering a simple expression oftype: σ = σ0exp(− �H/kT), in which σ0 is the pre-exponentialfactor, �H the activation energy, k the Boltzmann’s constant,

0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.310-8

10-7

10-6

10-5

10-4

10-3

10-2

500375250150100

σ (

S c

m -

1 )

1 0 0 0 / T ( K - 1 )

Δ H = 1.3 e V

V cm - 1

Fig. 3. Temperature dependence (Arrhenius plot) of electrical conductivity. Thetemperatures given are those of the furnace, which coincide with the sampletemperatures before the onset of the flash.15 The activation energy (�H) for theconductivity in the Arrhenius regime is 1.3 eV.

while T stands for the absolute temperature. In the linear regime,shown by the Arrhenius fit, the furnace and sample temperaturescoincide.15 In the non-linear regime, after the initiation of theflash, the specimen temperature would be higher than the fur-nace temperature. The activation energy for the conductivity ismeasured to be 1.3 ± 0.1 eV, which is comparable to the valuesfor grain boundary conductivity in BaTiO3.20–22

The microstructure of the specimen conventionally sinteredat 1300 ◦C is contrasted in Fig. 4 to the specimen sintered underan electric field of 250 V cm−1. The former shows a bimodal-likemicrostructure with larger grains. The average grain size (D) isapproximately 5.2 �m. This trend to accentuated abnormal graingrowth is well known in BaTiO3-based ceramics, and arises fromdevelopment of a liquid phase-assisted sintering mechanism (viaeutectic reaction at ∼1320 ◦C) for specimens heat treated at tem-peratures approaching or greater than 1300 ◦C.3,5 The much finergrain size in the flash sintered specimens suggests the absenceof the liquid phase and, therefore, a specimen temperature ofless than 1300 ◦C.

The values of the grain size, D, for various specimens arelisted in Table 2. The grain size measured from polished cross-sections, which were thermally etched, and from the fracturesurfaces of as-sintered samples were identical, except for the500 V cm−1 case where the thermally etched samples gave aslightly larger grain size. This is presumably because the very

Table 2Sintering conditions, average grain size (D), and dielectric properties: low-temperature permittivity, ε′(35 ◦C), and loss tangent, tanδ(35 ◦C), Curie temperature, TC,maximum permittivity, ε′

m = ε′ (TC), and loss tangent, tanδm = tanδ(TC), of various BaTiO3 specimens, measured at 100 kHz.

Sintering conditions D (�m) ε′(35 ◦C) (×103) tanδ(35 ◦C) (×10−2) TC (◦C) ε′m (×103) tanδm (×10−2)

1350 ◦C/1 ha 15.0 1.93 3.2 132 9.05 5.81300 ◦C/5 mina 5.2 2.29 3.3 132 10.16 6.6150 V cm−1/60 mA/15 min 1.1 3.54 2.7 127 6.05 2.2250 V cm−1/60 mA/3 min 0.6 3.32 3.0 128 5.99 2.8375 V cm−1/60 mA/2 min 0.4 2.62 3.9 129 5.33 4.1500 V cm−1/60 mA/1 min 0.3–0.4b 2.15 2.6 128 3.62 2.4

a Occurrence of abnormal grain growth (conventional sintering).b In this instance the fracture surface yielded a smaller grain size than the polished and thermally etched sample.

3658 J.-C. M’Peko et al. / Journal of the European Ceramic Society 34 (2014) 3655–3660

Fig. 4. Representative micrographs of samples conventionally sintered at 1300 ◦C for 5 min, and a flash sintered specimen at a field of 250 V cm−1 with a currentlimit set at 9.3 mA mm−2.

small grain size of this sample led to some grain growth duringthermal etching. Note that the grain size of 300 nm given inTable 2 is only about 4.3 times the starting mean particle size(70 nm).

The values for the grain size measured in these flash sinteringexperiments are comparable to those obtained in hot-isostatic-pressing and spark plasma sintering experiments 4,6–9.

The uniformity of microstructure and densification in flashsintering is influenced by the current limit. In the present workwe find that there is a transition from uniform to non-uniformsintering as the current limit is increased beyond a certain point.

The micrographs shown in Fig. 5 are a representative exampleof microstructure appearance after occurrence of this transitionfrom “safe” to “fail” conditions, when the current limit was inthis case increased from 15.4 mA mm−2 to 77.2 mA mm−2, atan applied field of 150 V cm−1. At the higher limit, the currentthrough the specimen became localized, creating a tunneling-like physical damage (with melting trace) in the direction ofthe electric field. The grain size is highest within the localizedregion, where it is 85 �m in Fig. 5a.1. At the edge of the localizedregion it is 22 �m as shown in Fig. 5a.2, and reduces to thenominal value of 1.4 �m at a distance from the hole as given

Fig. 5. Representative micrographs of a “fail” specimen showing tunneling-like physical damage promoted by current localization. The sample was flashed at a fieldof 150 V cm−1 and current limit of 77.2 mA mm−2. The accompanying micrographs show details of the microstructure: (a.1) within the hole, and at the cross section(a.2) near the hole and (a.3) far from the hole.

J.-C. M’Peko et al. / Journal of the European Ceramic Society 34 (2014) 3655–3660 3659

0 200 400 600 800 1000

0

3

6

920

40

60

80

100

(650)(750)(1000)(700)

(1000)(900)

(1100) (700)(900)

J ( m

A /

m m

2 )

E ( V / ctm )

Flash Regime (SAFE)

FAIL

SAFE

IncreasingGr. Size In

crea

sing

Gr.

Siz

e &

Den

sity

FAIL (current localiza�on) Flash Temperature, oC

Fig. 6. Current density-electric field map applying in electric field-assisted sin-tering of BaTiO3.

in Fig. 5a.3. We suggest that grain coarsening within and nearthe hole should have been the result of liquid phase-assistedsintering3,5 due to a significant increase in the local temperature(most likely well above 1350–1400 ◦C). In contrast, at a distancefrom the hole, where a dense microstructure with grain sizesclose to 1 �m is observed, it may be said that the temperaturestayed significantly below the eutectic point (i.e., T < 1320 ◦C).The same information may also be inferred for those samplesdiscussed in Table 2 and showing average grain sizes ≤1 �m.

These observations are consistent with the idea that flash sin-tering is related not only to Joule heating but also to defectgeneration.15 In a recent study on flash processed yttria-dopedzirconia, the finding of a residual defect density has also beenattributed to the defect mechanism.23

The localization of current flow represents a transitionin mechanism from uniform sintering to highly non-uniformmicrostructure which we called above as the “safe” and “fail”conditions. This idea is developed into a processing map shownin Fig. 6. The map is drawn with the current density and theelectric field as the axes. These data were obtained from isother-mal furnace temperatures. The temperatures are marked withinparenthesis; they range from 650 to 1100 ◦C. The short-dashedline divides the “safe” and “fail” regimes.

Fig. 6 also shows the trends for the influence of the field andcurrent density on the grain size and densification. The arrowsnear the bottom right of the figure show that higher current leadsto higher densities. On the other hand the grain size is smaller athigher values of E and for lower values of J. Maps such as shownin Fig. 6 highlight that the mechanism of field-induced effectson sintering and microstructure evolution can change with theelectrical control parameters. It is important to exercise cautionwhen comparing results from experiments which may lie in dif-ferent regimes and, therefore, represent different mechanisms ofresponse to electrical parameters.

We now discuss the measurement and results concerning thedielectric properties. Both the real (ε′) and imaginary (ε′′) partsof permittivity (ε* = ε′ − jε′′) were measured, from which theloss tangent was also calculated as tanδ = ε′′/ε′. The data for ε′,

Fig. 7. Temperature dependence of dielectric permittivity (ε′), left-hand scale,and dielectric loss (tanδ), right-hand scale, measured at 100 kHz on the conven-tionally and flash sintered BaTiO3 ceramics listed in Table 2. The sample labelsare C1 (1350 ◦C/1 h), C2 (1300 ◦C/5 min), F1 (150 V cm−1), F2 (250 V cm−1),F3 (375 V cm−1) and F4 (500 V cm−1). The grain size dependence of permit-tivity (ε′) toward low temperature (at 35 ◦C) is highlighted in the inset.

as a function of temperature, obtained at 100 kHz, are shown onthe left in Fig. 7. The values for ε′ at 35 ◦C and at the Curie tem-peratures (TC) are listed in Table 2. TC decreases from 132 ◦C forconventionally sintered samples to 128 ± 1 ◦C for flash sinteredspecimens.

The above trends are well known for BaTiO3-based ceram-ics and arise from (i) a progressive change from tetragonal topseudo-cubic structure (volume effect-related gradual loss offerroelectricity) and (ii) the increase in grain-boundary density(low-permittivity phase-related surface effect) with decreasinggrain size.4,6–9,24 The inset in Fig. 7 shows the effect of grainsize on the permittivity at 35 ◦C. The effect separates into tworegimes, at first rising and then declining with grain size. Therising side of the curve, when D < 1 �m, is explained by the rea-sons just given. At larger grain size, where 1 < D < 10 �m, thedecline in permittivity is ascribed to a decrease in the mobilityof the ferroelectric 90◦ domain walls, whose width decreases asthe square root of D.4,7,8

The right-hand scale of Fig. 7 shows the loss tangent at100 kHz, as a function of temperature. These data are alsoincluded in Table 2. The third column gives the values for the losstangent at 35 ◦C, while the sixth column gives its maximum val-ues measured at the Curie temperature. Conventionally and flashsintered samples have comparable values for the loss tangent at35 ◦C (revealing less than 5%). However, at the Curie tempera-ture the flash sintered samples show a significantly lower loss,also as a consequence of the bulk- and surface-related grain sizeeffects mentioned above.

4. Conclusions

We have shown that electric field-assisted processing ofundoped barium titanate precludes liquid phase-enhanced sin-tering and abnormal grain growth that is often prevalent inconventional sintering. Nearly fully dense specimens with grainsizes as low as 300–400 nm can be produced. Higher appliedfields lead to smaller grain sizes, while higher settings of the

3660 J.-C. M’Peko et al. / Journal of the European Ceramic Society 34 (2014) 3655–3660

current limit increase the grain size. Very high current settingleads to a different mechanism of electrical behavior wherebythe current becomes localized into a narrow channel throughthe specimen. These results point toward the importance of rec-ognizing that ceramics can respond by different mechanismsdepending on the setting of the electrical parameters in theexperiment. One recalls the significance of deformation mecha-nism maps where the dominant mechanism can change with theapplied stress, the temperature and the grain size.

Finally, the ferroelectric properties (Curie temperature anddielectric permittivity) of these materials were found to dependon the grain size, as has been reported in the literature.

Acknowledgements

This work was supported by the Office of Naval Researchunder the guidance of Dr. Lawrence Kabacoff, under grant no.N00014-12-1-0710, plus FAPESP and CAPES funding agenciesfrom Brazil, under grants no. 2012/06448-0 and no. 9291/13-0,respectively.

References

1. Jaffe B, Cook WR, Jaffe H. Piezoelectric ceramics. New York: AcademicPress; 1971.

2. Hertling GH. Ferroelectric ceramics: history and technology. J Am CeramSoc 1999;82(4):797–818.

3. Lin TF, Hu CT, Lin IN. Influence of stoichiometry on the microstructureand positive temperature coefficient of resistivity of semiconducting bariumtitanate ceramics. J Am Ceram Soc 1990;73(3):531–6.

4. Arlt G, Hennings D, de With G. Dielectric properties of fine-grained bariumtitanate ceramics. J Appl Phys 1985;58(4):1619–25.

5. Cho YK, Kang SJL, Yoon DY. Dependence of grain growth and grain-boundary structure on the Ba/Ti ratio in BaTiO3. J Am Ceram Soc2004;87(1):119–24.

6. Frey MH, Xu Z, Han P, Payne DA. The role of interfaces on an apparentgrain size effect on the dielectric properties for ferroelectric barium titanateceramics. Ferroelectrics 1998;206–207(1–4):337–53.

7. Zhao Z, Buscaglia V, Viviani M, Buscaglia MT, Mitoseriu L, Testino A,et al. Grain-size effects on the ferroelectric behavior of dense nanocrystallineBaTiO3 ceramics. Phys Rev B 2004;70:024107.

8. Buscaglia MT, Viviani M, Buscaglia V, Mitoseriu L, Testino A, Nanni P,et al. High dielectric constant and frozen macroscopic polarization in densenanocrystalline BaTiO3 ceramics. Phys Rev B 2006;73:064114.

9. Takeuchi T, Tabuchi M, Kageyama H, Suyama Y. Preparation of denseBaTiO3 ceramics with submicrometer grains by spark plasma sintering.J Am Ceram Soc 1999;82(4):939–43.

10. Cologna M, Rashkova B, Raj R. Flash sintering of nanograin zirconia in<5 s at 850 ◦C. J Am Ceram Soc 2010;93(11):3556–9.

11. Cologna M, Prette ALG, Raj R. Flash-sintering of cubic yttria-stabilizedzirconia at 750 ◦C for possible use in SOFC manufacturing. J Am CeramSoc 2011;94(2):316–9.

12. Downs JA, Sglavo VM. Electric field assisted sintering of cubic zirconia at390 ◦C. J Am Ceram Soc 2013;96(5):1342–4.

13. Prette ALG, Cologna M, Sglavo VM, Raj R. Flash-sintering ofCo2MnO4 spinel for solid oxide fuel cell applications. J Power Sources2011;196:2061–5.

14. Cologna M, Francis JSC, Raj R. Field assisted and flash sintering of aluminaand its relationship to conductivity and MgO-doping. J Eur Ceram Soc2011;31:2827–37.

15. Raj R. Joule heating during flash-sintering. J Eur Ceram Soc2012;32:2293–301.

16. Zapata-Solvas E, Bonilla S, Wilshaw PR, Todd RI. Preliminary investigationof flash sintering of SiC. J Eur Ceram Soc 2013;33:2811–6.

17. Raj R. Separation of cavitation-strain and creep-strain during deformation.J Am Ceram Soc 1982;65(3):C46.

18. Mendelson MI. Average grain size in polycrystalline ceramics. J Am CeramSoc 1969;52(8):443–6.

19. American Society for Testing and Materials (ASTM). E1382: Standard testmethods for determining average grain size using semiautomatic and auto-matic image analysis. Annual Book of ASTM Standards, v. 03.01. WestConshohochen, PA: ASTM; 1991.

20. Hirose N, West AR. Impedance spectroscopy of undoped BaTiO3 ceramics.J Am Ceram Soc 1996;79(6):1633–41.

21. M’Peko JC, Ruiz-Salvador AR, Rodríguez-Fuentes G. Conductivity acti-vation energy and analysis of the sintering process of dielectric ceramics.Mater Lett 1998;36(5–6):290–3.

22. M’Peko JC. Dynamics of the electrical response of ceramic dielectricmaterials in the presence of interfacial blocking effects. J Mater Sci Lett2000;19(21):1925–7.

23. M’Peko JC, Francis JSC, Raj R. Impedance spectroscopy and dielectricproperties of flash versus conventionally sintered yttria-doped zirconiaelectroceramics viewed at the microstructural level. J Am Ceram Soc2013;96(12):3760–7.

24. Da Silva RS, Hernandes AC, M’Peko JC. Effect of pH-induced nanopowderdeagglomeration on sintering, microstructure and dielectric properties ofBa0.77Ca0.23TiO3 ceramics. Mater Res 2012;15(4):522–9.