FUELCELLS
Proton transport enabled by a field-induced metallic
statein a semiconductor heterostructure
Y. Wu^1 , B. Zhu1,2†, M. Huang^3 *, L. Liu^1 , Q. Shi^1 , M. Akbar^3 , C. Chen^4 , J. Wei^5 , J. F. Li^5 , L. R. Zheng^6 ,
J. S. Kim^7 , H. B. Song^1 †
Tuning a semiconductor to function as a fast proton conductor is an emerging strategy in the
rapidly developing field of proton ceramic fuel cells (PCFCs). The key challenge for PCFC researchers
is to formulate the proton-conducting electrolyte with conductivity above 0.1 siemens per centimeter
at low temperatures (300 to 600°C). Here we present a methodology to design an enhanced
proton conductor by means of a NaxCoO 2 /CeO 2 semiconductor heterostructure, in which a
field-induced metallic state at the interface accelerates proton transport. We developed a PCFC
with an ionic conductivity of 0.30 siemens per centimeter and a power output of 1 watt per
square centimeter at 520°C. Through our semiconductor heterostructure approach, our results
provide insight into the proton transport mechanism, which may also improve ionic transport in
other energy applications.
T
he key challenge for research with new-
generation proton ceramic fuel cells
(PCFCs) is to fabricate electrolytes with
high proton conductivity ( 1 – 3 ). Good-
enough proposed the development of
oxide ion conductors based on the structural
design for solid oxide fuel cells (SOFCs) ( 4 ).
In this approach, oxide ion conductors are
designed by doping such that host cations are
replaced with those of lower valence to cre-
ate oxygen vacancies for O^2 −conductivity (e.g.,
Y3+or Sm3+replaces Zr4+or Ce4+in zirconia or
ceria fluorite structures). However, the struc-
tural doping has not yet brought out alternatives
to the conventional yttrium-stabilized zirconia
electrolyte. The report of a functional triple-
charge–conducting BaCo0.4Fe0.4Zr0.1Y0.1O 3 −d
cathode has prompted exploration of a strat-
egy to make PCFCs competitive with SOFCs
( 5 ), but the conductivities of proton ceramic
electrolytes ( 6 – 8 ) are still far below the de-
sired value of 0.1 S cm−^1. Our target is to de-
velop functional conductors with proton
conductivity of >0.1 S cm−^1 at 500°C to meet
the need for advanced PCFCs. Interfacial ionic
transport properties in semiconductor hetero-
structures and local electric field (LEF) ef-
fects have notable advantages in terms of
accessibility and tunability for creating ex-
tended functionalities ( 9 , 10 ). Here, we intro-
duce p-type NaxCoO 2 (NCO) materials (table
S1) to construct a semiconductor heterostruc-
ture system with n-type CeO 2 (Fig. 1A). TheLEF is built to form proton transport chan-
nelsattheNCOsurfacetoeffectivelymodulate
charge transport properties and confine fast
ionic mobility to the NCO/CeO 2 heterostruc-
ture interfaces. This methodology is in marked
contrast to the traditional approach of doping
crystal structures.
To elucidate the origin of the emergent elec-
tronic state in the NCO/CeO 2 heterostructure,
we performed first-principles calculations based
on density functional theory for three systems:
bulk NCO and CeO 2 (fig. S1), surfaces of the
NCO and CeO 2 , and an interface of the NCO/
CeO 2 (Fig. 1B). The partial density of states (fig.
S2)revealedthattheinterfaceacrossaCoO 2
layer of NCO and a Ce–O layer of CeO 2 causes
band bending, with the d-orbital electrons dip-
ping below the Fermi level, thereby forming
the metallic state. Further, the ionic migration
energy (e) enhances our understanding of the
optimal paths of proton transport. Although
protons may be transported through the lat-
tice of CeO 2 or layered-structure NCO,ei(i=
1and2forCeO 2 and NCO, respectively) is high
because of the structural bonding energy. By
contrast, the H···NCO bonding interaction184 10 JULY 2020•VOL 369 ISSUE 6500 sciencemag.org SCIENCE
(^1) Engineering Research Center of Nano-Geo Materials of
Ministry of Education, Faculty of Materials Science and
Chemistry, China University of Geosciences, Wuhan, 430074,
China.^2 Energy Storage Joint Research Center, Southeast
University School of Energy and Environment, Southeast
University, Nanjing, 210096, China.^3 Key Laboratory of Ferro
and Piezoelectric Materials and Devices of Hubei Province,
Faculty of Physics and Electronic Sciences, Hubei University,
Wuhan, 430062, China.^4 Huazhong University of Science and
Technology, Wuhan, 430074, China.^5 College of Chemistry
and Chemical Engineering, Xiamen University, Xiamen,
361005, China.^6 Institute of High Energy Physics, Chinese
Academy of Sciences, Beijing, 100049, China.^7 Department
of Aeronautical and Automotive Engineering, Loughborough
University, Loughborough LE11 3TU, UK.
*These authors contributed equally to this work.
†Corresponding author. Email: [email protected] (B.Z.); songhb@
cug.edu.cn (H.B.S.)
Fig. 1. Design of the NCO/CeO 2 heterostructure functionalities for fast proton migration.(A) Schematic
of the NCO/CeO 2 fuel cell and its operation mechanism (a LEF built at the NCO/CeO 2 interface from the CeO 2 side
to the NCO side confines protons along the NCO surface). (B) Total density of states of the surfaces of CeO 2 and
NCO and the NaxCO 2 /CeO 2 interface (x=0.60).(C) CeO 2 lattice with (111) plane, interface of the NCO/CeO 2
heterostructure, and NCO layer with (001) plane, with proton migration activation energiese 1 ,e 2 , ande 3 , respectively.
RESEARCH | REPORTS