Inkjet printing of gadolinium-doped ceria electrolyte on NiO-YSZ
substrates for solid oxide fuel cell applications
Chingfu Wang
•
Rumen I. Tomov
•
R. Vasant Kumar
•
Bartek A. Glowacki
Received: 18 January 2011 / Accepted: 24 May 2011 / Published online: 4 June 2011
Ó Springer Science+Business Media, LLC 2011
Abstract A sol–gel-based precursor solution of gadolin-
ium-doped ceria (CGO) was developed for deposition by
inkjet printing. A stable precursor was synthesised by
dissolving stoichiometric amounts of cerium (III) acetate
hydrate and gadolinium (III) acetate hydrate in propionic
acid, and diluted to 0.75 M with 1-propanol. The sintering
behaviour of the printed precursor was investigated. Since
the commonly used method of dilatometry is only appli-
cable to bulk samples, an alternative approach using Dif-
ferential scanning calorimetry (DSC) has been explored.
The sintering temperature of the printed precursor was
estimated by subtracting contributions from energy stored
due to heat capacity and activation energy of ionic mobility
from the DSC heat flow signal. Based on this modelling it
was found that the optimum sintering temperature of the
acetate-based CGO precursor was 1100 ± 55 °C, a result
independently confirmed by SEM imaging of printed pre-
cursor coating on NiO-YSZ cermet. A gadolinium-doped
ceria (CGO) thin film was then directly deposited on a
porous NiO-YSZ cermet anode composite by inkjet print-
ing. After co-sintering, it was shown that crack-free and
continuous coating thinner than 10 lm of CGO can be
readily produced. These results suggest that the inkjet
printing technique can be successfully implemented to
fabricate a thin film of dense electrolyte ([98%) for solid
oxide fuel cell (SOFC) applications.
Introduction
Gadolinium-doped ceria (CGO) is a promising electrolyte
material for intermediate temperature (below 600 °C) solid
oxide fuel cells (IT-SOFC) due to its high ionic conduc-
tivity and adequate chemical stability [1]. The electrolyte
layer must be gas-tight in order to prevent mixing of the
fuel and the oxidant, and sufficiently thin to minimise
ohmic resistance. The ideal electrolyte layer has to be
dense without any porosity, so that the conducting cross-
section is maximised. In SOFC development a serious
effort has been devoted to creating a dense, thin electrolyte.
There are two main categories of processing routes which
are sufficiently cost-effective to be considered, one being
the conventional ceramic powder route and the other being
the chemical route.
The former technique employs sub-micron ceramic
particles prepared as a stable colloidal suspension in the
form of a slurry or paste. The suspension can then be
deposited on the substrate with commonly used deposition
methods such as screen printing [2, 3] or tape casting [4].
However, the conventional technique requires a high sin-
tering temperature compared with chemical route (above
1300 °C, due to the relatively coarse particles size ranging
from 0.5 to 5 lm, often further accentuated by agglomer-
ation), and the reliable deposition of coatings with thick-
nesses less than 10 lm by these methods presents a
challenge. Relatively high sintering temperature is partic-
ularly problematic since the anode layer will undergo sin-
tering as well, resulting in a reduction in its porosity
leading to high concentration polarisation losses during
fuel cell operation.
Hence, an increasing trend is towards development of
chemical techniques based on sol–gel route. A chemical sol
precursor can be made by dissolving metal salts in suitable
C. Wang (&) Á R. I. Tomov Á R. Vasant Kumar Á
B. A. Glowacki
Material Chemistry and Applied Superconductivity Group,
Department of Materials Science and Metallurgy, University
of Cambridge, Pembroke Street, Cambridge CB2 3QZ, UK
e-mail: cw427@cam.ac.uk
123
J Mater Sci (2011) 46:6889–6896
DOI 10.1007/s10853-011-5653-y