Cite this: RSC Advances, 2013, 3,3599

Effect of Zr substitution for Ce inBaCe0.8Gd0.15Pr0.05O32d on the chemical stability in CO2

and water, and electrical conductivity3

Received 10th September 2012,Accepted 25th December 2012

DOI: 10.1039/c2ra22097k

www.rsc.org/advances

Sukhdeep Gill,a Ramaiyan Kannan,a Nicola Maffeib

and Venkataraman Thangadurai*a

In this paper, for the first time, we report the chemical stability of a highly proton conducting Gd+Pr-

codoped BaCe0.82xZrxGd0.15Pr0.05O32d (BCZGP) (0.01 , x , 0.3) as a function of Zr-doping in H2O vapour,

30 ppm H2S in H2, and pure CO2 along with its electrical conductivity in air, N2 + 3% H2O, H2 + 3% H2O and

N2 + D2O. All prepared BCZGP compositions retain the original cubic perovskite-type structure in 30 ppm

H2S in H2 at 600 uC. BCZGP with x = 0.3 shows significant stability under pure CO2 at 400 uC, while upon

exposure to H2O vapor all compositions form Ba(OH)2?xH2O. The maximum electrical conductivity

obtained with higher Zr-doping in BCZGP (x = 0.3) is 7.6 6 1023 S cm21 which is about 30% of that of the

parent compound BaCe0.8Gd0.15Pr0.05O32d. Current work clearly shows that Zr-doping at x = 0.3 increases

the stability of BCZGP under 30 ppm H2S and pure CO2 at intermediate temperatures (T ¡ 400 uC), and

retains good proton conductivity in H2 containing atmosphere.

Introduction

Solid oxide fuel cells (SOFCs) have recently been attractingtremendous research attention for future energy needsespecially in stationary power applications due to its highefficiency, fuel flexibility, and combined heat and power co-generation. State-of-the art Y-doped ZrO2 (YSZ) electrolytebased SOFCs suffer several material issues such as disintegra-tion of the electrode and electrolyte components due to thehigh operating temperatures (¢800 uC). Hence, the search fornew solid electrolytes exhibiting fast proton and oxide ionconduction that can operate in the intermediate temperature(IT) range (350–700 uC) has been carried out for the past fewdecades.1–18 A variety of proton and oxide ion conductors werereported with different inorganic crystal structures thatinclude perovskites, perovskite-related layered perovskites,brownmillerites, pyrochlores, apatites, and fluorites.19–21

Among the proton conductors, perovskite-type acceptor-dopedBaCeO3 (BC) showed the highest proton conductivity underwater vapor compared to the corresponding alkaline-earth andZr analogues.22–24

Following their possible role in IT-SOFCs, significant efforthas been paid to understanding their crystal structures,chemical stability under different atmospheres (e.g., CO2,water vapor), and physical-chemical properties. For example,Knight found the cubic perovskite-type structure (space group:Pm3m) was formed at temperatures above 1173 K, while at lowtemperatures the orthorhombic phase (space group: Imma)was formed.25–27 Proton conductivity in BCs is achieved by thesubstitution of aliovalent metal ions for Ce which createsoxygen vacancies that could lead to the formation of hydroxylprotons under moisture at elevated temperatures.28,29 Theseprotons migrate from one lattice oxygen to another, therebycreating significant proton conduction. The key setback withacceptor-doped BCs is their chemical stability in the operatingenvironment of SOFCs, i.e., under CO2 (by-products whenhydrocarbon used as fuels) and H2O vapor (by-products whenH2 or hydrocarbons is used as fuels) where they form BaCO3

and Ba(OH)2, respectively.30,31 Subsequently, attempts havebeen made to improve the chemical stability by doping highelectronegativity elements at the B site.32,33 For example, solidsolutions of BaCeO3 and BaZrO3 have been prepared, wherethe incorporation of Zr (electronegativity for Zr is 1.33 against1.12 for Ce) improves the chemical stability under CO2, whileadversely affecting the proton conductivity.34–36

Y and Y+Yb-codoped Ba(CeZr)O3 have been shown to bechemically stable and remain as promising candidates for ITproton conductors, although a drop in electrical conductivitywas observed with increasing Zr content.37,38 Pr-dopedBa(Pr0.6Gd0.4)O32d (BPG) and BaCe0.8Y0.15Pr0.05O32d (BCYP)

aUniversity of Calgary, Department of Chemistry, 2500 University Dr, NW, Calgary,

Alberta, T2N 1N4, Canada. E-mail: vthangad@ucalgary.cabCanmet ENERGY, Transportation Energy Technology, Natural Resources Canada,

Ottawa, K1A 1M1, Canada

3 Electronic supplementary information (ESI) available: PXRD after chemicalstability measurements at 800 uC and images of the samples before and aftervarious stability measurements. See DOI: 10.1039/c2ra22097k

RSC Advances

PAPER

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3599–3605 | 3599

Dow

nloa

ded

by P

urdu

e U

nive

rsity

on

11 M

arch

201

3Pu

blis

hed

on 0

4 Fe

brua

ry 2

013

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2R

A22

097K

View Article OnlineView Journal | View Issue

have been shown to have very high proton conductivity atintermediate temperatures under humid environments.27,39–41

Very recently, we demonstrated that Gd+Pr co-doped BCs(BCGP) could achieve electrical conductivities comparable tothose of Y-doped BCs, but their stability remained poor underCO2, as with the majority of doped BCs.42,43 Moreover, Gd+Prhave higher electronegativity and ionic radius characteristicsthan those of Y+Yb (electronegativity of Gd : Pr is 1.2 : 1.13against the 1.22 : 1.1 of Y : Yb, and ionic radius in theoctahedral coordination 93.8 : 99 pm versus 90 : 86 pmrespectively). Since Zr is known to increase the stability as inthe case of Y doping where only after doping with Zr theyachieved significant stability, it is imperative that we continueour pursuit of a chemically stable solid electrolyte for SOFCswithout surrendering much proton conductivity by system-atically introducing Zr in BCGP. To our knowledge, there is noreport available on improving the stability of BCGP by Zrdoping. In this work, we systematically investigate thechemical stability of Zr-doped BCZGP under CO2, 30 ppmH2S in H2, and H2O vapor and their electrical properties in air,N2 + 3% H2O, H2 + 3% H2O and N2 + 3% D2O and established avery important fundamental relationship of chemical compo-sition–stability–proton conductivity in the Zr-doped BCZGPsystem.

Experimental

Sample preparation

Perovskites of the nominal chemical formula BaCe0.82xZrx

Gd0.15Pr0.05O32d (BCZGP) (x = 0.01, 0.05, 0.1, 0.2, 0.3) wereprepared by a solid state (ceramic) method using high purity(.99.9%) metal oxide and carbonate precursors such asBaCO3, CeO2, ZrO2, Gd2O3, and Pr6O11. All these startingmaterials were mixed according to the required weight ratiosand ball milled in 2-propanol for 6 h using zirconia balls.These materials were calcined at 1350 uC for 24 h and re-ball-milled for 4 h at 200 rpm and were then made into a pelletusing an isotactic press and sintered at 1400 uC (CarboliteRHF16/3, England) for 24 h.44,45 The samples were character-ized through powder X-ray diffraction (PXRD using a BrukerD8 powder X-ray diffractometer; Cu-Ka, 40 kV, 40 mA) at roomtemperature.

Chemical stability

For chemical stability measurements in different atmospheresthe pellets were ground into a fine powder before actualmeasurement. The apparatus used for chemical stability underH2O vapor, pure CO2 and 30 ppm H2S in H2 is shown in Fig. 1.For stability under H2O vapor, a small pellet was powderedand placed above water as shown in Fig. 1a in a round bottomflask fitted with a condenser. A hot plate was used to boil thewater and the experiment was continued for 24 h after whichthe samples were collected and dried in an oven at 100 uCovernight. For stability under different gas atmospheres, thepowdered samples were placed on an alumina boat and placedinside a quartz tube which was inserted inside a tubular

furnace (Fig. 1b). The gas from a bottle was passed over thesample through a gas inlet and outlet for 24 h at differenttemperatures and then the sample was collected for furthercharacterization. PXRD and thermogravimetric (TGA) analysis(Mettler Toledo, TGA/DSC/HT1600) were performed on thesesamples before and after the stability tests.

Electrical conductivity

Conductivity measurements were carried out using a Solartron1260 instrument in the frequency range of 0.1 Hz to 1 MHzwith an AC amplitude of 100 mV on the sintered pellets(density higher than 90%, y1 cm in diameter and 2 mmthickness). These pellets were polished and coated with Ptpaste (LP A88-11S ink, Heraeus Inc., Germany) and dried at800 uC for 2 h in air to remove the organic binders. Pt wireswere used as current collectors by attaching them to thesurface of the pellet using a spring-loaded contact. Theelectrical conductivity measurements were performed in air,N2 + 3% H2O, H2 + 3% H2O and N2 + D2O atmospheres. Thetemperature was controlled using a Barnstead tubular furnace(model 21100) and held constant for a minimum of 2 h to amaximum of 12 h before each measurement. The conductivitymeasurements for each sample were collected in the range of400–700 uC. The AC impedance plots were analysed usingZ-view software.

Results and discussion

Phase characterization

Powder X-ray diffraction patterns (PXRD) of BCZGP samplescalcined at 1400 uC with different Zr-doping levels (x = 0.01–0.3) reveal the formation of single-phase cubic perovskitestructure, as shown in Fig. 2a. Fig. 2b shows the diffractionpeak shifting for the (110) peak from left to right withincreasing Zr content and Fig. 2c demonstrates the decrease inlattice constant as a function of Zr in BCZGP. The decrease incell constant with increasing Zr content is consistent with thesmaller Shannon’s ionic radius of zirconium(IV) ions (72 pmfor octahedral coordination) than those of cerium(IV) ions (87pm).46 A maximum 2h shift of 0.5u is observed at higher Zrloading (x = 0.3). Table 1 lists the observed and calculatedPXRD data and the lattice constants for all prepared BCZGP.

Fig. 1 Schematic representation of the experimental setup used for (a) H2Ovapor and (b) chemical stability measurements under test gas conditions.

3600 | RSC Adv., 2013, 3, 3599–3605 This journal is � The Royal Society of Chemistry 2013

Paper RSC Advances

Dow

nloa

ded

by P

urdu

e U

nive

rsity

on

11 M

arch

201

3Pu

blis

hed

on 0

4 Fe

brua

ry 2

013

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2R

A22

097K

View Article Online

Chemical stability

Fig. 3a and 3b show the observed PXRD patterns of BCZGPsamples after exposure to pure CO2 at 400 uC and 600 uC for 24h, respectively. At 400 uC, the incorporation of higher Zr (x =0.3) increases the stability of the compound as confirmed bythe absence of any additional peaks. However, with Zr dopinglower than x = 0.3 peaks corresponding to BaCO3 are observed.Nevertheless, at 600 uC all the BCZGP compositions showBaCO3 peaks and only BCZGP x = 0.3 could retain to someextent the perovskite structure.14,21 The BaCO3 formation isproved by peaks marked with ‘#’ at 2h values around 25 uC andsimilar peaks were present in the PXRD patterns of BCGPsamples studied by Dauter et al.42 and Bhella et al.47 CO2

stability measurements carried out at 800 uC on BCZGPsamples also show similar trends in the PXRD patterns (Fig.S1,3 see ESI).

To understand the level of Ba to BaCO3 conversion and theeffect of dopant atoms on improving the stability, wemeasured the weight increment after exposing the samplesto 100% CO2 at two temperatures (400 and 600 uC) and the

results are shown in Fig. 4. The weight of the sample increaseswith temperature due to the formation of BaCO3 and decreaseswith increase in Zr content which suggests that while Zrdoping is helping in improving the chemical stability underCO2 environment it is not high enough to make BCZGPsamples completely stable against CO2 especially at elevatedtemperatures. Apart from a few reports, inBaCe0.82xZrxY0.2O32d based compounds only at x ¢ 0.8,complete stability under pure CO2 at elevated temperaturesis achieved while significant stability is attained at x ¢

0.5.14,37,38 However, we doped to a maximum of x = 0.3 mainlybecause of the detrimental effect of Zr on the electricalconductivity.14

Fig. 5a shows the PXRD patterns of the BCZGP (x = 0.05–0.3)after the chemical stability test in 30 ppm H2S in H2 at 600 uCfor 20 h; they resemble the PXRD patterns observed for as-sintered samples in Fig. 2. Thus, the BCZGP samples remainchemically stable under 30 ppm H2S environment at 600 uC,although some unidentified impurity peaks are seen withBCZGP x ¢ 0.2. Similarly, the lattice parameters have alsochanged slightly as observed by the peak shifting to the rightin comparison to the PXRD patterns seen in Fig. 2. This couldbe attributed to the reduction of samples in H2 environmentthat could result in oxide ions leaving the lattice structure (a =

Fig. 2 (a) PXRD patterns of BaCe0.82xZrxGd0.15Pr0.05O32d (x = 0.01, 0.05, 0.1, 0.2and 0.3) showing the formation of single-phase cubic perovskite structure, (b)magnified view of the (110) peak clearly indicating the shift in peak positiontowards higher 2h values and (c) variation of lattice constant as a function of Zrconcentration.

Table 1 Indexed PXRD data of perovskite-type BaCe0.82xZrxGd0.15Pr0.05O32d (x = 0.01, 0.05, 0.1, 0.2 and 0.3)

h k l

0.01 0.05 0.1 0.2 0.3

d(obs) (Å) d(cal) (Å) I(obs) (%) d(obs) (Å) d(cal) (Å) I(obs) (%) d(obs) (Å) d(cal) (Å) I(obs) (%) d(obs) (Å) d(cal) (Å) I(obs) (%) d(obs) (Å) d(cal) (Å) I(obs) (%)

1 1 0 3.095 3.095 100 3.086 3.086 100 3.085 3.085 100 3.069 3.069 100 3.046 3.046 1002 0 0 2.196 2.189 15 2.191 1.182 16 2.188 2.181 17 2.172 2.170 17 2.159 2.154 222 1 1 1.790 1.787 22 1.787 1.782 26 1.786 1.781 25 1.775 1.772 25 1.766 1.759 272 2 0 1.551 1.548 10 1.548 1.543 10 1.546 1.542 9 1.536 1.535 8 1.531 1.523 113 1 0 1.390 1.384 5 1.388 1.380 6 1.383 1.380 7 1.377 1.373 6 1.371 1.362 92 2 2 1.267 1.263 2 1.264 1.260 2 1.264 1.259 2 1.257 1.253 2 1.251 1.243 3

a = 4.387(1) Å a = 4.377(4) Å a = 4.375(2) Å a = 4.348(1) Å a = 4.332(2) Å

Fig. 3 PXRD patterns obtained for BaCe0.82xZrxGd0.15Pr0.05O32d (BCZGP) sam-ples after chemical stability measurements under (a) pure CO2 at 400 uC for 24 hand (b) pure CO2 at 600 uC for 24 h (# indicates peaks corresponding to BaCO3

formation (JCPDS card number: 5-378)).

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3599–3605 | 3601

RSC Advances Paper

Dow

nloa

ded

by P

urdu

e U

nive

rsity

on

11 M

arch

201

3Pu

blis

hed

on 0

4 Fe

brua

ry 2

013

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2R

A22

097K

View Article Online

4.332(2) Å and 4.320(2) Å for BCZGP x = 0.3 before and afterH2S treatment respectively).48 However, upon exposure to H2Ovapor for 24 h at y90 uC, all the investigated BCZGP samplesbecame amorphous and peaks corresponding to CeO2 couldbe seen in their PXRD patterns (Fig. 5b). This is furtherassociated with a change in appearance as these dark greypowders turned yellow (Fig. S2,3 see ESI). It is worth notinghere that after exposure to pure CO2 at 600 uC these powdersalso turned yellow (Fig. S23). Although SOFCs operate at highertemperatures, we carried out water vapor stability measure-ments at low temperatures specifically because the free energychange for the Ba(OH)2 formation reaction from BCs reachesequilibrium conditions at 403 uC and above this temperatureBCs remain stable.49,50 For example, incorporation of 20 mol%Gd for Ce has been reported to increase the stability under

water vapor at higher temperatures, but at low temperature itfails to provide chemical stability.30 In a similar study on thestability of BCs, Virkar suggested a bulk-type decompositionmechanism where the diffusion of water in the BC lattice isfollowed by the conversion of Ba to Ba(OH)2.30,31,51 In thepresent study as well, the incorporation of Zr seems to be notoffering any long-term stability since all the samples havedecomposed upon exposure to H2O vapor.

Fig. 6 shows the TGA curves obtained in air for the BCZGP asprepared and H2O vapor treated samples, revealing y7–8%weight loss in the latter due to the removal of water below 200uC and an overall weight loss of 12.5 wt.% by 600 uC. This is inaccordance with the complete conversion of Ba toBa(OH)2?xH2O (in the hydrated form) since a theoreticalweight loss of only 6% is expected if only Ba(OH)2 is formedwithout hydration during water treatment. Barium hydroxideis known to exist in three forms, Ba(OH)2?xH2O (x = 0, 1, 8),and the observed weight loss of up to 12.5% suggests thepresence of water of hydration in the water vapor treatedsamples. TGA curves of as-prepared samples show a weightloss less than 0.5% clearly indicating that the composition ofthe sample is significantly affected after exposure to vapor.Thus, the incorporation of Zr has not increased the stability ofthe Gd+Pr-doped BCs. In a similar study on BaCe0.92x

ZrxY0.1O2.95 based systems, stability under boiling waterconditions could be achieved only after significant replace-ment of Ce with Zr (x ¢ 0.5).21 Thus, the observed behaviourin this case is in accordance with reports and a small amountof doping of Zr (x ¡ 0.3) will not induce any significantstability upon exposure to H2O vapor at temperatures below100 uC.

Electrical conductivity

Fig. 7a and 7b show the equivalent circuit fitted impedanceplots obtained for BCZGP (x = 0.1) at 500 uC and 700 uC,respectively, in various atmospheres. Similar AC impedanceplots were observed for doped BCs.34,42 The validity of ourfitting is further confirmed by Kramers–Kronig (KK) transfor-

Fig. 4 Weight gain in BaCe0.82xZrxGd0.15Pr0.05O32d (BCZGP) samples as afunction of Zr concentration after chemical stability measurements under pureCO2 at 400 and 600 uC. Here x values are 0.01, 0.05, 0.1, 0.2 and 0.3.

Fig. 5 PXRD patterns obtained with various BCZGP compositions after exposureto (a) 30 ppm H2S in pure H2 for 20 h at a flow rate of 20 ml per minute and (b)H2O vapor for 24 h at y90 uC.

Fig. 6 TGA curves obtained for H2O vapor treated (solid lines) and as-prepared(dotted lines) samples of BaCe0.82xZrxGd0.15Pr0.05O32d (BCZGP) (x = 0.01, 0.05,0.1, 0.2 and 0.3).

3602 | RSC Adv., 2013, 3, 3599–3605 This journal is � The Royal Society of Chemistry 2013

Paper RSC Advances

Dow

nloa

ded

by P

urdu

e U

nive

rsity

on

11 M

arch

201

3Pu

blis

hed

on 0

4 Fe

brua

ry 2

013

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2R

A22

097K

View Article Online

mation which matches well with the fitted results and the ratioZFit 2 ZKK/ZKK remained close to zero suggesting the goodnessof the fit (Fig. 8). Similar studies have been carried out on solidstate electrolytes where a residual ratio close to zero indicatesthe accuracy of the fit.52–54 Also the chi-squared valuesobtained with our fitting are of the order of 1024 furtherconfirming the correctness of the fit. The results obtained at500 uC through the fitting are given in Table 2. The presence ofbulk and grain-boundary resistance in the prepared material isindicated by the observation of more than one semicircle in airand other environments. The non-blocking nature of theelectrode/electrolyte interface is confirmed by the low-fre-quency intercept on the real axis. In 3% H2O + H2 conditions,only one semicircle is observed at temperatures ¢700 uCsuggesting the possible elimination of grain-boundary effectsin the observed Nyquist plot. However, at higher Zr content(¢0.1) two semicircles were observed indicating the role of Zrin reducing the sinterability and in turn the electricalconductivity. Among the different operating conditions, alarge decrease in conductivity is observed in N2 + D2O

compared to that of N2 + 3% H2O, proving proton transportin the material.

Further, in an attempt to identify the bulk and electrode-sample interface effects, we calculated the capacitance fromthe fitting parameters. The data was fitted with circuitsconsisting of R and CPE in parallel combination and thecircuits used for fitting can be found elsewhere.55 The actualcapacitance can be calculated using the expression:55–57

C~R

1{n

n

� �Q

1

n

� �(1)

In air, the capacitance observed with the high-frequency partsemicircle was in the range of 1028–10210 F, while that of thesemicircle observed in the intermediate-frequency range wason the order of 1026 to 1028 F (Table 2). These values are inaccordance with the reported range for bulk and sample/electrode interfaces in ceramic materials. The capacitancecorresponding to the low frequency semicircle falls in therange of electrochemical reactions.58 The reported electricalconductivity in the present work can be considered as the totalconductivity since the low-frequency intercept or minimum tothe real axis was used to determine the electrical conductivityas it was not possible to separate the bulk and grain-boundaryconductivity over the investigated temperature regime.

Fig. 7 Typical equivalent circuit fitted AC impedance Nyquist plots obtainedwith BCZGP x = 0.1 under various atmospheres (a) at 500 uC, (b) at 700 uC.

Fig. 8 Compliance of the Kramers–Kronig transformed data with the Nyquistdata obtained at (a–b) 500 uC and (c–d) 700 uC.

Table 2 Fitting parameters of impedance plots obtained with BaCe0.7Zr0.1Gd0.15Pr0.05O32d (BCZGP at x = 0.1) at 500 uC under different environments

Compound Conditions R1 (V) Q1 (F) n1 C1 (F) R2 (V) Q2 (F) n2 C2 (F) R3 (V) Q3 (F) n3 C3 (F)

BCZGP (x = 0.1) Air 516 9.9 6 1029 0.80 4.4 6 10210 472 1.4 6 1025 0.54 2.0 6 1027 1446 1.2 6 1023 0.24 6.5 6 1023

N2+3%H2O 621 1.5 6 1029 0.93 4.9 6 10210 4001 3.8 6 1026 0.58 1.8 6 1027 15 625 1.1 6 1024 0.26 4.9 6 1024

N2+D2O 2099 3.2 6 1029 0.79 1.4 6 10210 1778 1.4 6 1026 0.95 1.1 6 1026 38 215 8.9 6 1026 0.38 1.6 6 1026

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3599–3605 | 3603

RSC Advances Paper

Dow

nloa

ded

by P

urdu

e U

nive

rsity

on

11 M

arch

201

3Pu

blis

hed

on 0

4 Fe

brua

ry 2

013

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2R

A22

097K

View Article Online

Typical Arrhenius plots obtained for BCZGP (x = 0.3) atdifferent environments are shown in Fig. 9a, while acomparison of the total electrical conductivity obtained withBCZGP x = 0.3 and some of the Y and Pr doped BCs reported inthe literature is given in Fig. 9b. The activation energiesobtained from the Arrhenius plots are presented in Table 3.Clearly under H2 + 3% H2O environment, the conductivity ofthe material has increased significantly while in N2 + 3% H2O,the total conductivity is comparable to that in air. However,under N2 + D2O conditions the conductivity drastically reducedsuggesting the role of protons in providing good conductivityin BC based perovskites.36,59,60 The decrease in electrical

conductivity in D2O medium can be attributed to an isotopeeffect, i.e.

D2OzOxozV..

o '2OD.o (2)

where, Oxo, VNN

o, and ODNo represent the oxygen lattice site,

oxygen vacancy site, and oxygen vacancy filled with OD group,respectively.

The increased electrical conductivity in H2 + 3% H2O couldbe attributed to the incorporation of a significant amount ofprotons into the ceramic matrix that helps in achievingpredominantly protonic conduction while in other atmo-spheres the conductivity is due to mixed ionic (protons, oxideions, and holes) conduction. This is further supported bythe reduced activation energy in H2 + 3% H2O atmospheresindicative of an easy charge transport. A maximum conductiv-ity of 7.8 6 1023 S cm21 at 700 uC is observed for BCZGP x =0.3 in H2 + 3% H2O compared to that of 2.6 6 1022 S cm21 forthe Zr free parent compound, BCGP. Comparison of con-ductivity values reported in the literature for Y doped BCs withBCZGP x = 0.3 suggests that Pr doping provides the best protonconductivity for BCs, and although Zr doping in our case hasreduced the conductivity values, it is still comparable to that ofY and Zr co-doped BCs.

Conclusions

In summary, we found that the incorporation of Zr in BCZGPhas helped in achieving better stability in H2S at 600 uC and inpure CO2 at 400 uC. However, in H2O vapor and in CO2

atmosphere at elevated temperatures the Zr doping does notimprove the stability as confirmed by the formation ofamorphous Ba(OH)2 and BaCO3 peaks in PXRD. Further, theconductivity of the BCZGP samples doped with Zr content x =0.3 was 7.6 6 1023 S cm21 at 700 uC, 30% of the non-dopedBCGP’s conductivity of 2.6 6 1022 S cm21. Thus, it seemsdespite increased electronegativity the higher ionic radii of Gd+ Pr seem to have reduced the chemical stability and theincorporation of Zr in the BCGP ceramic matrix will not besufficient to increase the chemical stability. Hence thesolution to this problem must be sought after various othercombinations with or without Zr.

Fig. 9 (a) Arrhenius plots obtained under different environments with BaCe0.5Zr0.3

Gd0.15Pr0.05O32d (BCZGP x = 0.3); (b) comparison of BCZGP x = 0.3 (present work)conductivity with literature reports for BaCe0.7Zr0.1Y0.2O32d (BCZY7),38 BaCe0.8Y0.15

P0.05O32d (BCYP), BaCe0.4Zr0.4Y0.15Pr0.05O32d (BCZYP), and BaCe0.4Zr0.4Y0.2O32d (BCZY),40

BaCe0.5Gd0.15Pr0.05O32d (BCGP),42 BaCe0.75Y0.25O32d (BCY25),61 and BaCe0.9Y0.1O32d

(BCY10).62

Table 3 Comparison of proton conductivity s (S cm21) and activation energy Ea (eV) under different environments for BaCe0.82xZrxGd0.15Pr0.05O32d (x = 0.01, 0.05,0.1, 0.2 and 0.3)

x

Air N2 + 3% H2O H2 + 3% H2O N2 + D2O

s700 uC Ea s700 uC Ea s700 uC Ea s700 uC Ea

0.01 2.5 6 1023 0.76 8.8 6 1024 0.68 8.3 6 1023 0.45 9.4 6 1025 0.900.05 7.8 6 1023 0.59 7.8 6 1023 0.59 8.2 6 1023 0.71 3.3 6 1024 0.970.1 3.3 6 1023 0.57 9.3 6 1024 0.76 2.2 6 1023 0.44 5.2 6 1024 0.970.2 6.6 6 1023 0.62 2.5 6 1023 0.61 3.2 6 1023 0.43 5.7 6 1024 0.740.3 4.3 6 1023 0.57 1.1 6 1023 0.6 7.6 6 1023 0.71 4.2 6 1024 0.83

3604 | RSC Adv., 2013, 3, 3599–3605 This journal is � The Royal Society of Chemistry 2013

Paper RSC Advances

Dow

nloa

ded

by P

urdu

e U

nive

rsity

on

11 M

arch

201

3Pu

blis

hed

on 0

4 Fe

brua

ry 2

013

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2R

A22

097K

View Article Online

Acknowledgements

One of us (V.T.) thanks Natural Resources Canada (NRCan) forproviding financial support and the Natural Science andEngineering Research Council (NSERC) of Canada and theCanada Foundation for Innovation (CFI) for their support.

Notes and references

1 E. D. Wachsman and K. T. Lee, Science, 2011, 334, 935–939.2 Z. Shao and S. M. Haile, Nature, 2004, 431, 170–173.3 N. Maffei, L. Pelletier, J. P. Charland and A. McFarlan, J.

Power Sources, 2006, 162, 165–167.4 Y. -H. Huang, R. I. Dass, Z. -L. Xing and J. B. Goodenough,

Science, 2006, 312, 254–257.5 N. P. Brandon, S. Skinner and B. C. H. Steele, Annu. Rev.

Mater. Res., 2003, 33, 183–213.6 L. Carrette, K. A. Friedrich and U. Stimming, Fuel Cells,

2001, 1, 5–39.7 A. Lashtabeg and S. J. Skinner, J. Mater. Chem., 2006, 16,

3161–3170.8 V. Thangadurai and W. Weppner, Ionics, 2006, 12, 81–92.9 J. C. Boivin and G. Mairesse, Chem. Mater., 1998, 10, 2870–2888.

10 K. R. Kendall, C. Navas, J. K. Thomas and H. -C. Z. Loye,Solid State Ionics, 1995, 82, 215–223.

11 J. A. Kilner, Solid State Ionics, 2000, 129, 13–23.12 H. Inaba and H. Tagawa, Solid State Ionics, 1996, 83, 1–16.13 S. Tominaka, N. Akiyama, F. Croce, T. Momma, B. Scrosati

and T. Osaka, J. Power Sources, 2008, 185, 656–663.14 E. Fabbri, A. D’Epifanio, E. Di Bartolomeo, S. Licoccia and

E. Traversa, Solid State Ionics, 2008, 179, 558–564.15 E. C. Subbarao and H. S. Maiti, Solid State Ionics, 1984, 11,

317–338.16 M. Mogensen, N. M. Sammes and G. A. Tompsett, Solid

State Ionics, 2000, 129, 63–94.17 A. J. Jacobson, Chem. Mater., 2009, 22, 660–674.18 E. Fabbri, D. Pergolesi and E. Traversa, Chem. Soc. Rev.,

2010, 39, 4355–4369.19 D. J. L. Brett, A. Atkinson, N. P. Brandon and S. J. Skinner,

Chem. Soc. Rev., 2008, 37, 1568–1578.20 E. Kendrick, M. S. Islam and P. R. Slater, J. Mater. Chem.,

2007, 17, 3104–3111.21 Z. Zhimin, Solid State Ionics, 2007, 178, 213–220.22 H. Iwahara, T. Mori and T. Hibino, Solid State Ionics, 1995,

79, 177–182.23 H. Iwahara, T. Esaka, H. Uchida and N. Maeda, Solid State

Ionics, 1981, 3–4, 359–363.24 D. Hirabayashi, A. Tomita, S. Teranishi, T. Hibino and

M. Sano, Solid State Ionics, 2005, 176, 881–887.25 K. S. Knight, Solid State Ionics, 1994, 74, 109–117.26 K. S. Knight, Solid State Ionics, 2001, 145, 275–294.27 E. Gorbova, V. Maragou, D. Medvedev, A. Demin and

P. Tsiakaras, J. Power Sources, 2008, 181, 207–213.28 K. D. Kreuer, Annu. Rev. Mater. Res., 2003, 33, 333–359.29 G. J. Kearley, R. P. White, C. Forano and R. C. T. Slade,

Spectrochim. Acta, Part A, 1990, 46, 419–424.30 S. V. Bhide and A. V. Virkar, J. Electrochem. Soc., 1999, 146,

2038–2044.31 C. W. Tanner and A. V. Virkar, J. Electrochem. Soc., 1996,

143, 1386–1389.

32 H. Matsumoto, Y. Kawasaki, N. Ito, M. Enoki andT. Ishihara, Electrochem. Solid-State Lett., 2007, 10, B77.

33 H. Iwahara T. Hibino and M. Yamada, in Solid Oxide FuelCells, ed. S. C. Singhal and H. Iwahara, PV 93–4, p. 137, TheElectrochemical Society Proceedings Series, Pennington,NJ, 1993.

34 K. Katahira, Y. Kohchi, T. Shimura and H. Iwahara, SolidState Ionics, 2000, 138, 91–98.

35 T. Hibino, K. Mizutani, T. Yajima and H. Iwahara, SolidState Ionics, 1992, 57, 303–306.

36 K. H. Ryu and S. M. Haile, Solid State Ionics, 1991, 125, 355–367.37 L. Yang, S. Wang, K. Blinn, M. Liu, Z. Liu, Z. Cheng and

M. Liu, Science, 2009, 326, 126–129.38 C. Zuo, S. Zha, M. Liu, M. Hatano and M. Uchiyama, Adv. Mater.,

2006, 18, 3318–3320.39 E. Fabbri, L. Bi, D. Pergolesi and E. Traversa, Energy Environ. Sci.,

2011, 4, 4984–4993.40 J. Melnik, J. Luo, K. T. Chuang and A. R. Sanger, The Open

Fuels & Energy Science Journal, 2008, 1, 7–10.41 T. Fukui, S. Ohara and S. Kawatsu, Solid State Ionics, 1999,

116, 331–337.42 J. Dauter, N. Maffei, S. S. Bhella and V. Thangadurai, J.

Electrochem. Soc., 2010, 157, B1413–B1418.43 S. S. Bhella, T. Furstenhaupt, R. Paul and V. Thangadurai,

Inorg. Chem., 2011, 50, 6493–6499.44 N. Maffei, L. Pelletier and A. McFarlan, J. Power Sources,

2008, 175, 221–225.45 L. Pelletier, A. McFarlan and N. Maffei, J. Power Sources,

2005, 145, 262–265.46 R. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr.,

Theor. Gen. Crystallogr., 1976, 32, 751–767.47 S. S. Bhella and V. Thangadurai, J. Power Sources, 2009, 186,

311–319.48 W. -Y. Tan, Q. Zhong, M. -S. Miao and H. -X. Qu, Ionics,

2009, 15, 385–388.49 Thermodynamic database MALT group, Thermodynamic

database MALT for Windows, Kagaku Gijutsu-Sha, 2005.50 P. Sawant, S. Varma, B. N. Wani and S. R. Bharadwaj, Int. J.

Hydrogen Energy, 2012, 37, 3848–3856.51 S. Gopalan and A. V. Virkar, J. Electrochem. Soc., 1993, 140,

1060–1065.52 P. Agarwal, O. C. Moghissi, M. E. Orazem and L. H. Garcia-

Rubio, Corrosion, 1993, 49:4, 278–289.53 B. A. Boukamp, Solid State Ionics, 2004, 169, 65–73.54 N. Hildenbrand, B. A. Boukamp, P. Nammensma and D. H.

A. Blank, Solid State Ionics, 2011, 192, 12–15.55 Q. Li and V. Thangadurai, J. Mater. Chem., 2010, 20, 7970–7983.56 P. Hjalmarsson, M. Søgaard and M. Mogensen, Solid State

Ionics, 2009, 180, 1395–1405.57 E. Chinarro, J. R. Jurado, F. M. Figueiredo and J. R. Frade,

Solid State Ionics, 2003, 160, 161–168.58 J. T. S. Irvine, D. C. Sinclair and A. R. West, Adv. Mater.,

1990, 2, 132–138.59 W. B. Wang, J. W. Liu, Y. D. Li, H. T. Wang, F. Zhang and G.

L. Ma, Solid State Ionics, 2010, 181, 667–671.60 K. Kunstler, H. J. Lang, A. Maiwald and G. Tomandl, Solid

State Ionics, 1998, 107, 221–229.61 A. Tomita, T. Hibino, M. Suzuki and M. Sano, J. Mater. Sci.,

2004, 39, 2493–2497.62 W. Suksamai and I. S. Metcalfe, Solid State Ionics, 2007,

178, 627–634.

This journal is � The Royal Society of Chemistry 2013 RSC Adv., 2013, 3, 3599–3605 | 3605

RSC Advances Paper

Dow

nloa

ded

by P

urdu

e U

nive

rsity

on

11 M

arch

201

3Pu

blis

hed

on 0

4 Fe

brua

ry 2

013

on h

ttp://

pubs

.rsc

.org

| do

i:10.

1039

/C2R

A22

097K

View Article Online