Optimization on technical parameters for fabrication of SDC film by screen-printing used as electrolyte in IT-SOFC

Article abstract of DOI:10.1016/j.jpcs.2008.02.011

Sm-doped Ceria (SDC) electrolyte film was successfully fabricated on anode substrate of NiO-SDC by screen-printing. Some technical parameters for fabrication were investigated and optimized, including printing times, ink composition and sintering temperature. Scanning electron microscope (SEM) measurement was done to check the microstructures of SDC film and single cell. The parameters greatly affected the quality of SDC film and cell performance. The single cell with the optimum parameters exhibited an OCV of 0.82 V and a power density of 0.5 W/cm² at 600 °C.

Full text of DOI:10.1016/j.jpcs.2008.02.011

ARTICLE IN PRESS
Journal of Physics and Chemistry of Solids 69 (2008) 2019–2024
www.elsevier.com/locate/jpcs
Optimization on technical parameters for fabrication of SDC film by
screen-printing used as electrolyte in IT-SOFC
a
a,
Ã
a
a
a,b
a
Lijun Zhao , Xiqiang Huang , Ruibin Zhu , Zhe Lu , Weiwei Sun , Yaohui Zhang ,
a
Xiaodong Ge , Zhiguo Liu , Wenhui Su
a
a,c
a
Center for Condensed Matter Science and Technology, Harbin Institute of Technology, Harbin 150001, China
b
College of Natural Sciences, Jiamusi University, Jiamusi 154007, China
International Center for Material Physics, Academia, Shenyang 110015, China
c
Received 14 December 2007; received in revised form 14 February 2008; accepted 18 February 2008
Abstract
Sm-doped Ceria (SDC) electrolyte film was successfully fabricated on anode substrate of NiO-SDC by screen-printing. Some technical
parameters for fabrication were investigated and optimized, including printing times, ink composition and sintering temperature.
Scanning electron microscope (SEM) measurement was done to check the microstructures of SDC film and single cell. The parameters
greatly affected the quality of SDC film and cell performance. The single cell with the optimum parameters exhibited an OCV of 0.82 V
2
and a power density of 0.5 W/cm at 600 1C.
r 2008 Elsevier Ltd. All rights reserved.
Keywords: A. Thin films; D. Electrochemical properties; D. Microstructure
1
. Introduction
electrolyte film, including slurry coating [3], spary pyrolysis
10], pulsed laser deposition [10], electrophoretic deposition
[
Solid oxide fuel cell (SOFC) has attracted more and more
[5,11], dry pressing [6,12,13], tape-casting [2,14], chemical
vapor deposition (CVD) [15], DC magnetron sputtering
[16], sol–gel dip-drawing [17], spray coating [18,19], plaster
casting [20], slurry spin coating [21,22], screen-printing
[23–25], and so on. Compared with the other techniques,
screen-printing is low-cost, easy to realize, and suitable
for mass-production. In the screen-printing processing,
many technical parameters can greatly affect the quality
of electrolyte film and the fabrication reproducibility.
However, to our knowledge, there is hardly any report
on optimizing the parameters.
attention, due to its advantages: cleanliness, efficiency, low
pollution and fuel flexibility [1]. For long-life working and
encouraging its practical application, many research groups
focused their study on reducing the operation temperature
of SOFC [2–10]. Recently, CeO -based oxide, such as
2
Sm-doped Ceria (SDC) [2,4–6] and Gd-doped Ceria (CGO)
[
(
3,10], was frequently used as electrolyte for IT-SOFC
intermediate temperature SOFC), due to their high oxide-
ionic conductivity compared to Y O -stabilized ZrO (YSZ)
2
3
2
in the intermediate-temperature range. Furthermore, thermal
ꢁ
6
ꢁ1
expansion coefficient (12.5 ꢀ 10
K
) of doped Ceria is
much closer to that of Ni-based substrate, compared with
In this paper, effects of some technical parameters on
SDC electrolyte film and cell performance were investi-
gated and the parameters were optimized.
ꢁ
6
ꢁ1
YSZ (10.5 ꢀ 10
K ) [2].
For further improving cell performance, it is necessary to
reduce the thickness of electrolyte to lower ohmic loss.
Some research groups focused on techniques to fabricate
2. Experimental
Ã
NiO powder was prepared by a precipitation method.
Ba0.5Sr0.5Co0.8Fe0.2O3–d (BSCF) powder was synthesized
by a citric–nitrate process, and SDC (Sm0.2Ce0.8O1.9
Corresponding author. Tel.: +86 451 86418420;
fax: +86 451 86412828.
E-mail address: huangxq@hit.edu.cn (X. Huang).
)
0022-3697/$ - see front matter r 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jpcs.2008.02.011

ARTICLE IN PRESS
L. Zhao et al. / Journal of Physics and Chemistry of Solids 69 (2008) 2019–2024
2020
powder as well. Ammonia and Ni(NO ) ꢂ H O, Ba(NO ) ,
3
2
2
3 2
a
b
d
Sr(NO ) , Co(NO ) ꢂ 6H O, Fe(NO ) ꢂ 9H O, Ce(NO ) ꢂ
3
2
3 2
2
3 3
2
3 3
6
H O, and Sm O used as raw materials, were all analytical
2 2 3
reagent (A.R.). Please see Ref. [24] for detailed information
on the preparation process of NiO, BSCF and SDC
powder. The as-prepared NiO and SDC powder was mixed
in a weight ratio of 65:35 and ball-milled for 10 h with
ethanol as media. 15 wt% starch was added as pore former.
Then the mixed powder was pressed into pellets of 13 mm
in diameter and 0.5 mm in thickness. After being sintered at
1000 1C for 2 h, the pellets can be used as substrates for the
screen-printing processing.
c
Ink for screen-printing was composed of x wt% SDC
power and 100ꢁx wt% (x ¼ 30, 40, 50 and 60) ethyl
cellulose–terpineol vehicle. A terylene screen was used with
ꢁ
1
mesh count of 165 wires cm and wire-diameter of 30 mm.
The screen had an opening size of 30 mm and an open area
of 25%. One printing pass was done during one-time
printing process. Between two printing processes, the
printed film was dried in air at room temperature for
about 15 min. Different printing times (one-, three-, five
and seven-time) were employed to get green SDC films,
which were then sintered at different temperatures (1300,
e
g
f
1
5
350, 1400 and 1450 1C) for 4 h with a heating rate of
1C min . Then the BSCF powder was screen-printed
ꢁ
1
SDC
SDC
Anode
Anode
onto the surface of SDC films, and sintered at 970 1C for
3
h to obtain BSCF cathode.
The microstructures of the SDC films and single cells was
examined by a scanning electron microscope (SEM).
Solartron SI 1260 impedance/grain-phase analyzer in combi-
nation with SI 1287 electrochemical interface was employed
to measure the cell performance with a four-probe method.
Hydrogen and air was used as fuel and oxidant, respectively.
h
Anode
SDC
3
3
3
. Results and discussion
Anode
SDC
.1. Effect of printing times
.1.1. SEM
SEM result for the surface view and cross-sectional view
of SDC films with different printing times was shown in
Fig. 1. It was seen from the surface view that the number of
holes in SDC film decreased with the increase of printing
times. The SDC film fabricated by a seven-time printing
processing exhibited a densest structure.
Fig. 1. SEM result for surface views of SDC films with the printing times
of 1 (a), 3 (b), 5 (c), and 7 (d), and cross-sectional views of single cells with
the printing times of 1 (e), 3 (f), 5 (g), and 7 (h).
As shown in the cross-sectional images in Fig. 1,
thickness of SDC film increased with the printing times.
The linear dependence of the thickness of SDC film on the
printing times was plotted in Fig. 2. As shown in Fig. 2,
every two-time printing processing resulted in a 4.5 mm
increase of the thickness of SDC film. The thickness of
SDC film fabricated by a seven-time printing processing
was about 16.5 mm.
0.8 V, which was lower than the theoretical value due to the
low ionic transport number of SDC.
The dependence of OCV of the cells measured at 600 1C on
the printing times was plotted in Fig. 3. As shown in Fig. 3,
when the printing times was below 5, the OCV exhibited a
linear increase. When the printing times was beyond 5, the
OCV slightly increased to 0.82 V. This value was close to the
upper limit of OCV of single cell with SDC electrolyte.
I–V and I–P curves for the single cells measured at
600 1C were plotted in Fig. 4. As shown in Fig. 4, the
two single cells with SDC electrolyte film fabricated by
3
.1.2. Cell performance
As many researchers reported [2–4,6,26,27], single cell
with SDC electrolyte usually shows an OCV of only about

ARTICLE IN PRESS
L. Zhao et al. / Journal of Physics and Chemistry of Solids 69 (2008) 2019–2024
2021
18
16
14
12
10
8
6
4
2
0.9
0.6
.5
0.4
.3
1
3
5
7
-time
-time
-time
-time
0
0
0
0
.8
.7
.6
.5
0
0
0.4
0
0
0
0
.3
.2
.1
.0
0.2
0.1
0.0
0
.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Current density (A/cm2)
1
2
3
4
5
6
7
Fig. 4. I–V and I–P curves at 600 1C of single cells with the printing times
of 1, 3, 5 and 7.
Printing times
Fig. 2. Dependance of thickness of SDC films on printing times.
x wt% (x ¼ 30, 40, 50, and 60) SDC powder into ethyl
0
0
0
0
0
0
0
0
0
0
.9
.8
.7
.6
.5
.4
.3
.2
.1
.0
cellulose–terpineol vehicle.
3.2.1. SEM
Fig. 5 showed surface images of SDC films fabricated
using the ink containing x wt% (x ¼ 30, 40, 50 and 60)
SDC powder, and cross-sectional images of single cells. As
shown in the cross-sectional images in Fig. 5, all the SDC
films exhibited almost the same thickness of about 16.5 mm,
which indicated that screen-printing processing in this
paper was reproducible. As shown in Fig. 5(e)–(h), the
SDC film with x ¼ 40 was dense and some cracks occurred
on surfaces of the other three films. This result suggested
that the ink composition had a great effect on the quality of
the film. When the value of x was low, the ink contained
too much organic impregnant and the impregnant was
away from the green film by sintering at high temperatures,
which resulted in a large amount of pores in the film as
shown in Fig. 5(e). As the value of x was 50 or more, the
fluidity of ink was poor and the imprint, which was
produced by the terylene screen during the screen-printing
processing, was not able to disappear automatically.
Therefore, it was not surprising to find some cracks on
the films after sintering, as shown in Fig. 5(g) and (h).
1
3
5
7
Printing times
Fig. 3. Dependance of OCV (open-circuit voltage) of single cells on
printing times at 600 1C.
one- and three-time printing exhibited not-stable perfor-
mance and some data points were greatly away from the
trend with their OCVs very low. This result was attributed
to the gas cross-leakage of SDC electrolyte film. The cell
with SDC film fabricated by seven-time printing processing
3.2.2. Cell performance
2
exhibited the highest power density of about 0.5 W/cm .
To investigate the effect of ink composition on cell
performance, the dependence of OCV on the value of x,
and I–V and I–P curves of single cells with x ¼ 30, 40, 50,
and 60 was measured at 600 1C. The result was shown in
Figs. 6 and 7. As shown in Fig. 6, the cell with x ¼ 40
exhibited the highest OCV of 0.82 V. The other three cells
exhibited relatively low OCVs, which resulted from the gas
cross-leakage, due to the bad densification of SDC
electrolyte film. This result well-matched the analysis in
Section 3.2.1. As shown in Fig. 7, the cell with x ¼ 40
Although it is possible to slightly increase the OCV by
further increasing the printing times, the increase of
printing times results in a relatively high cost for
fabrication and a thicker electrolyte film, which is not
advantageous to cell performance. So it can be concluded
that the seven-time printing processing should be preferred
during SDC film fabrication by screen-printing.
3
.2. Effect of ink composition
2
provided the highest power density of about 0.5 W/cm and
2
a short-circuit current density of about 1.3 A/cm . Due to
In this section, the seven-time printing processing was
adopted. Ink for screen-printing was prepared by adding
their low OCVs, the other three cells with x ¼ 30, 50 and 60

ARTICLE IN PRESS
L. Zhao et al. / Journal of Physics and Chemistry of Solids 69 (2008) 2019–2024
2022
a
b
0
0
0
0
0
.8
.6
.4
.2
.0
SDC
Anode
SDC
Anode
c
d
30
40
50
60
SDC
Anode
SDC
Anode
x value
Fig. 6. OCV of single cells as a function of x value at 600 1C.
0
.9
.8
.7
.6
.5
.4
.3
.2
.1
.0
x = 30
x = 40
x = 50
x = 60
0.6
0.5
0
0
0
0
0
0
0
0
0
e
f
0
0
0
0
.4
.3
.2
.1
0.0
0
.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Current density (A/cm2)
g
h
Fig. 7. I–V and I–P curves at 600 1C of single cells with x ¼ 30, 40, 50,
and 60.
3.3.1. SEM
Fig. 8 showed SEM result for surface views of SDC films
sintered at 1300, 1350, 1400 and 1450 1C. As shown in
Fig. 8, the size of grain of in SDC films increased with the
sintering temperature. The two SDC films sintered at 1300
and 1350 1C exhibited porous structure, with the other two
ones sintered at 1400 and 1450 1C dense. So the lowest
sintering temperature, at which dense SDC films was
prepared, was 1400 1C.
Fig. 5. SEM result for cross-sectional views of the single cells with x ¼ 30
(
a), x ¼ 40 (b), x ¼ 50 (c), x ¼ 60 (d), and surface views of the SDC films
with x ¼ 30 (e), x ¼ 40 (f), x ¼ 50 (g), x ¼ 60 (h). The SDC films in the
single cells were fabricated by screen-printing processing, using the ink
containing x wt% SDC powder.
3.3.2. Cell performance
Fig. 9 showed OCVs of single cells with different
exhibited poor output performance. So the ink containing
40 wt% SDC powder should be preferred to fabricate SDC
film by screen-printing.
sintering temperatures of 1300, 1350, 1400 and 1450 1C
measured at 600 1C. With the increasing of sintering
temperature, the OCV increase greatly in the temperature
region below 1400 1C. The OCV of cell with the sintering
temperature of 1400 1C was 0.82 V. However, the cell with
the sintering temperature of 1450 1C exhibited an OCV of
about 0.8 V, which was lower than that of the cell with the
sintering temperature of 1400 1C. During the sintering
3
.3. Effect of sintering temperature
In this section, the ink containing 40 wt% SDC powder
was used to fabricate SDC films and a seven-time printing
processing was adopted.

ARTICLE IN PRESS
L. Zhao et al. / Journal of Physics and Chemistry of Solids 69 (2008) 2019–2024
2023
0
0
0
0
0
0
0
0
0
.9
.8
.7
.6
.5
.4
.3
.2
.1
a
b
1300°C
350°C
1400°C
450°C
0.6
0.5
1
1
0
0
0
0
.4
.3
.2
.1
c
d
0.0
0.0
0
.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
2
Current density (A/cm )
Fig. 10. I–V and I–P curves at 600 1C of single cells with different
sintering temperatures of 1300, 1350, 1400 and 1450 1C.
0.7
0.6
0.5
0.4
o
1
1
1
350 C
Fig. 8. SEM result for surface views of SDC films with different sintering
temperatures of 1300 1C (a), 1350 1C (b), 1400 1C (c) and 1450 1C (d).
o
400 C
o
450 C
0
0
0
0
0
.8
.6
.4
.2
0.3
0.2
0.1
0.0
High Frequency
Low Frequency
0.0
0.1
0.2
0.3
Z' (Ω cm2)
0.4
0.5
0.6
0.7
Fig. 11. Impedance spectra of single cells with different sintering
temperatures of 1300 1C (a), 1350 1C (b), 1400 1C (c) and 1450 1C (d)
under open-circuit condition at 600 1C.
.0
1
250
1300
1350
1400
1450
1500
Sintering Temperature (°C )
Fig. 9. OCV of single cells measured at 600 1C as a function of sintering
temperature.
density of the other three cells with the sintering
temperature of 1300, 1350 and 1450 1C. The cell with the
sintering temperature of 1450 1C exhibited a more serious
concentration polarization, which resulted from the over-
sintering of anode substrate during the preparation of SDC
film.
Fig. 11 showed the impedance spectra of the single cells
with different sintering temperatures of 1350, 1400 and
1450 1C, under open-circuit condition at 600 1C. The result
for the cell with the sintering temperature of 1300 1C was
not plotted here, because the data points were in a serious
disorder, which resulted from the not-dense electrolyte and
gas cross-leakage. As shown in Fig. 11, the intercept at
high frequency on the real axis was ohmic resistance (Ro),
and the difference between the two intercepts on the real
processing, there was stress between the electrolyte and
anode substrate. When the sintering temperature for the
fabrication of SDC film was increasing, the stress became
more and more. If the sintering temperature was too high,
the cell was cracked or broken, which resulted in a low
OCV.
Fig. 10 shows the output performance of single cells with
different sintering temperature of 1300, 1350, 1400 and
1
450 1C measured at 600 1C. The cell with the sintering
temperature of 1400 1C provided the highest power density
of 0.5 W/cm . Low OCVs badly affected the output power
2

ARTICLE IN PRESS
L. Zhao et al. / Journal of Physics and Chemistry of Solids 69 (2008) 2019–2024
2024
Table 1
Fitting result for Ohmic resistance (R
resistance (R) of the single cells with different sintering temperatures of
Acknowledgment
o
i
), interface resistance (R ) and total
The authors gratefully acknowledge the financial sup-
port from the Ministry of Science and Technology of China
863 under Contract no. 2007AA05Z139.
1350, 1400 and 1450 1C at 600 1C
Sintering temperature (1C)
1
350
1400
1450
References
2
R
R
o
i
(O cm )
0.22
0.42
0.64
0.08
0.14
0.22
0.064
0.146
0.21
2
(O cm )
[1] N.Q. Minh, J. Am. Ceram. Soc. 76 (3) (1993) 563–588.
[2] X. Zhang, M. Robertson, C. Deces-Petit, W. Qu, O. Kesler,
R. Maric, D. Ghosh, J. Power Sources 164 (2007) 668–677.
2
R ¼ R
o i
+R (O cm )
[3] Y. Liu, S. Hashimoto, H. Nishino, K. Takei, M. Mori, J. Power
Sources 164 (2007) 56–64.
axis was interface resistance (R ). The intercept at low
i
[4] S. Li, Z. Lu, N. Ai, K. Chen, W. Su, J. Power Sources 165 (2007)
97–101.
frequency on the real axis was the total resistance (R),
where R ¼ R +R . The detailed fitting result for R , R and
[5] M. Matsuda, T. Hosomi, K. Murata, T. Fukui, M. Miyake, J. Power
Sources 165 (2007) 102–107.
o
i
o
i
R was displayed in Table 1. The two cells with the sintering
temperature of 1400 and 1450 1C exhibited almost the same
[
6] C. Jiang, J. Ma, X. Liu, G. Meng, J. Power Sources 165 (2007)
1
34–137.
[7] A. Subramania, T. Saradha, S. Muzhumathi, J. Power Sources 165
(2007) 728–732.
interface resistance R , and the R of the cell with the
i
i
2
sintering temperature of 1350 1C was highest, 0.42 O cm .
With the increase of the sintering temperature, the total
resistance R decreased. The cell with the sintering tempera-
[8] D. Lee, J.-H. Han, Y. Chun, R.-H. Song, D.R. Shin, J. Power
Sources 166 (2007) 35–40.
[9] M. Kawano, H. Yoshida, K. Hashino, H. Ijichi, S. Suda,
K. Kawahara, T. Inagaki, Solid State Ion. 177 (2006) 3315–3321.
ture of 1450 1C exhibited the smallest ohmic resistance
2
(
R ¼ 0.064 O cm ), which resulted from the denser elec-
o
[10] J.L.M. Rupp, L.J. Gauckler, Solid State Ion. 177 (2006) 2513–2518.
[11] L. Jia, Z. Lu, X. Huang, Z. Liu, Z. Zhi, X. Sha, G. Li, W. Su, Ceram.
Int. 33 (2007) 631–635.
trolyte film sintered at 1450 1C. In Fig. 10, the cell with the
sintering temperature of 1450 1C provided a smaller
absolute value of slope than the cell with the sintering
temperature of 1400 1C in the region of low current density
in the I–V curve. This result was related to the smaller
[
[
12] C.R. Xia, M.L. Liu, Solid State Ion. 144 (2001) 249–255.
13] X. Xin, Z. Lu, Q. Zhu, X. Huang, W. Su, J. Mater. Chem. 17 (2007)
1627–1630.
14] J.V. Herle, R. Ihringer, R.V. Cavieres, L. Constantin, O. Bucheli,
J. Eur. Ceram. Soc. 21 (2001) 1855–1859.
[
ohmic resistance (R ) and total resistance (R) of the cell
o
[
[
15] K.W. Chour, J. Chen, R. Xu, Thin Solid Films 304 (1997) 106–112.
16] P.K. Srivastava, T. Quach, Y.Y. Duan, R. Donelson, S.P. Jiang,
F.T. Ciacchi, S.P.S. Badwal, Solid State Ion. 99 (1997) 311–319.
17] W.S. Jang, S.H. Hyun, J. Mater. Sci. 37 (2002) 2535–2541.
18] C.H. Wang, W.L. Worrel, S. Park, J.M. Vohs, R.J. Gorte,
J. Electrochem. Soc. 148 (8) (2001) A864–A868.
with the sintering temperature of 1450 1C as shown in
Table 1. However, the cell with the sintering temperature of
1
450 1C did not showed a highest power density, due to its
[
[
serious concentration polarization in the region of high
current density, as shown in Fig. 10. Therefore, the
optimum sintering temperature for screen-printing fabrica-
tion of SDC film was 1400 1C.
[
[
[
[
[
[
[
[
[
19] P. Charpentier, P. Fragnaud, D.M. Schleich, E. Gehain, Solid State
Ion. 135 (2000) 373–380.
20] J. Liu, W. Liu, Z. Lu, L. Pei, L. Jia, L.Y. He, W.H. Su, Solid State
Ion. 118 (1999) 67–72.
4
. Conclusions
21] K. Chen, Z. Lu, N. Ai, X. Huang, Y. Zhang, X. Ge, X. Xin, X. Chen,
W. Su, Solid State Ion. 177 (2007) 3455–3460.
22] J. Wang, Z. Lu, X. Huang, K. Chen, N. Ai, J. Hu, W. Su, J. Power
Sources 163 (2007) 957–959.
SDC electrolyte film was successfully fabricated by
screen-printing. Some technical parameters for the proces-
sing were optimized: printing times, ink composition and
sintering temperature. A seven or more times printing
processing was necessary to obtain dense SDC films. The
ink containing 40 wt% SDC powder was preferable for
screen-printing. The optimum sintering temperature for the
fabrication of SDC films was 1400 1C. The single cell with
23] Y. Zhang, X. Huang, Z. Lu, Z. Liu, X. Ge, J. Xu, X. Xin, X. Sha,
W. Su, J. Power Sources 160 (2006) 1065–1073.
24] Y. Zhang, X. Huang, Z. Lu, Z. Liu, X. Ge, J. Xu, X. Xin, X. Sha,
W. Su, J. Power Sources 160 (2006) 1217–1220.
25] X. Ge, X. Huang, Y. Zhang, Z. Lu, J. Xu, K. Chen, D. Dong, Z. Liu,
J. Miao, W. Su, J. Power Sources 159 (2006) 1048–1050.
26] I. Riess, M. Godickemeier, L.J. Gauckler, Solid State Ion. 90 (1996)
9
1–104.
the optimum parameters exhibited a power density of
2
27] M. Godickemeier, L.J. Gauckler, J. Electrochem. Soc. 145 (1998)
414–421.
0
.5 W/cm and an OCV of 0.82 V at 600 1C.