Advanced low-temperature solid oxide fuel cells based on a built-in electric field

Solid oxide fuel cells (SOFCs) show considerable promise for meeting the current ever-increasing energy demand and environmental sustainability requirements as a result of their high efficiency and low environmental impact. To enable high ionic conductivity, SOFCs are often required to operate at high temperature, which in turn results in high costs^[1]. Therefore, lowering the operational temperatures has become a major priority in SOFC research and development^[2]. According to the traditional concepts of SOFCs, single semiconductor materials are usually considered as electrolyte membrane due to their higher ionic conductivity, with heterostructures constructed from different semiconductor materials having never been considered. Recently, Meng et al.^[3] made an important breakthrough in low-temperature SOFCs by introducing semiconductor heterojunction membranes to function alternatively as electrolytes with better performance. This novel fuel cell design is known as a semiconductor-ionic membrane fuel cell (SIMFC)^[3-5]. Zhang et al.^[6], Nie et al.^[7], Deng et al.^[8], Mushtaq et al.^[9], and Afzal^[10] used semiconductor materials, including Ni_0.8Co_0.15Al_0.05LiO_2-δ^[6], La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ^[7], Sr₂Fe_1.5Mo_0.5O_6-δ^[8], SrFe_0.75Ti_0.25O_3-δ^[9], Ba_0.5Sr_0.5Co_0.8Fe_0.2O_3-δ^[10], ZnO^[11], CeO_2-δ^[12], SrTiO₃^[13] and Ba_0.5Sr_0.5Co_0.1Fe_0.7Zr_0.1Y_0.1O_3-δ (BSCFZY)^[14], to construct a semiconductor membrane (SM) with enhanced ion conduction while electron conduction is blocked via the principles of a semiconductor heterojunction^[15-17]. The SM can be a composite made by a semiconductor with an ionic conductor material, e.g., Sm-doped ceria, or a semiconductor alone, e.g., CeO_2-δ^[12], SrTiO₃^[13] or BSCFZY^[14], while the traditional SOFC electrolyte component is made using a pure ionic conductor.

Recently, Wang et al.^[18] reported a 3C-SiC, which was tuned for protonic conducting properties via the construction of an n-p heterostructure composite with Na_0.6CoO₂, exhibiting an ionic conductivity of 0.12 S cm^-1 at 550 °C. Lu et al.^[19] reviewed recent progress in lowering the temperature of SOFCs by using semiconductor-ionic conductor nanomaterials. The development in the application of nanostructured pure ionic conductors, semiconductors and their nanocomposites as membranes is highlighted in this review. Xu et al.^[20] reported a SIMFC using a composite of Ba-Co-Ce-Y-O and CeO₂, reaching a remarkable peak power density of 1140 mW·cm^-2 at 550 °C. Zhu et al.^[21] produced a nanoscale perspective of solid oxide and semiconductor membrane fuel cells from materials to technology. They discussed the nanoscale electrochemical phenomena of SIMFCs. Different from the traditional concept where semiconductor materials are widely used in photoelectric conversion and photocatalysis^[22], they also applied them to replace the electrolytes in fuel cells.

It is well known that semiconductor materials have already been successful in photovoltaic cells based on a built-in electric field (BIEF)^[23]. Generally, when p- and n-type semiconductors are contacted, the redistribution of charges at the interface constitutes a space-charge region with the BIEF pointing from the n- to p-type region^[24]. The BIEF is also applied in lithium-ion batteries. Qiao et al.^[25] demonstrated a BIEF to reduce the space charge layer formation and boost lithium-ion transport in all-solid-state lithium-ion batteries by an in-situ differential phase contrast scanning transmission electron microscopy technique and finite element method simulations.

Most importantly, Zhang et al.^[6], Nie et al.^[7], Deng et al.^[8], Mushtaq et al.^[9], and Afzal^[10] successfully applied semiconductor materials in SOFCs by compositing semiconductor and ionic conductor materials to construct SIMFCs with BIEF effects. To understand the principle of SIMFCs, a physical mode based on a BIEF and alignment of the energy band, similar to a perovskite solar cell (PSC), was proposed by Zhu et al.^[16]. Generally, the perovskite absorbers of a PSC undergo photoexcitation and charge separation under light illumination. The holes move to the metal contact through the hole-transporter materials while electrons are collected by the electron transport layer and move to the fluorine-doped tin dioxide conductive substrate to generate the electricity^[26,27]. The BIEF in a PSC can effectively prohibit the electrons passing through the device itself. Taking advantage of this concept, Zhu et al.^[16] designed a novel fuel cell with a nanocomposite functional layer, where the short circuit issue can be eliminated by a heterojunction structure instead of using the ionic electrolyte layer in SOFCs. The mechanisms of SIMFCs can be explained using the principle of a PSC, as shown in Figure 1A. The charge separation is caused by the well-aligned band positions between the perovskite and electron/hole conducting layers. Inspired by this idea, Zhu et al.^[16] constructed an n-type La_0.2Sr_0.25Ca_0.45TiO₃ and p-type La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) junction, where the electrons cannot pass through the junction. Furthermore, the BIEF in the SIMFC membrane can drive H⁺ or O^2- across the junction, as illustrated in Figure 1B.

Figure 1. Energy level diagrams of (A) a PSC and (B) the fuel-to-electricity conversion device inspired by the PSC structure^[16]. LSCT: La_0.2Sr_0.25Ca_0.45TiO₃; LSCF: La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ; SCDC: Ce_0.8Sm_0.05Ca_0.15O_2-δ; NCAL: Ni_0.8Co_0.15Al_0.05LiO_2-δ; FTO: fluorine-doped tin oxide; HTM: hole transport materials.

Interestingly, a Schottky junction (SJ) was also found in this all new device, where Ni_0.8Co_0.15Al_0.05LiO_2-δ (NCAL) is reduced into a Ni-Co alloy at the anode with H₂, as indicated in Figure 1B. After the SJ is formed between the anode and electrolyte, it can also inhibit electrons passing through the electrode and SM while simultaneously enhancing the transportation of H⁺ or O^2- due to the BIEF formation at the interface. The used materials for PSC and SIMFC devices are listed in Table 1. It is true that the junction plays a vital role in blocking electrons crossing over the internal device to avoid short circuiting and also in promoting the ionic transport process.

According to our previous research, the BIEF can be formed by a Schottky heterojunction^[24], intrinsic-negative (i-n) heterojunction^[12] and p-n heterojunction^[25]. In a Schottky heterojunction, the BIEF can be built up simply at the interface of the metal (electrode) and semiconductor (electrolyte) regions. Yun et al.^[28] constructed SJ fuel cells using a p-type semiconductor material, namely, a LiNi_0.85Co_0.15O_2-δ (LCN) composite with Ce_0.8Sm_0.2O_1.9-Na₂CO₃, which exhibited a high power output of 1000 mW cm^-2 at 550 °C. A thin Ni-metal layer originating from reduction of the semiconducting oxide LCN is formed at the H₂ side. Therefore, a Ni-metal/p-type LCN-semiconductor SJ is formed that can accelerate ion transport capacity while inhibiting electrons from passing through the junction formed at the anode/semiconductor membrane interface, as shown in Figure 2A.

Figure 2. (A) Schottky junction device and energy band diagram for a metal/p-semiconducting oxide interface^[28]. (B) Charge separation at the interface of a CeO_2-δ/CeO₂ particle and proton transport to the near-surface layers of the particle^[30].

In order to prove the as-reported SJ, Zhu et al.^[29] tested the response current as a function of bias voltage for a half cell with the structure of NCAL-Ni/7LCP-3ZnO at 550 °C with air and a H₂ flow to the NCAL electrode side. The results indicated that an apparent rectification response emerged in the I-V characteristic analogous to the reported I-V response of a Ni/ZnO-polar contact, illustrating a Schottky contact between the reduced anode and membrane layer. Furthermore, the characterization of semiconductor properties and band structures is very important for SIMFCs based on semiconductor-ionic membranes with BIEF effects, which are very different from conventional SOFCs based on ionic conducting electrolytes. Therefore, various new characterization techniques from semiconductor aspects, like ultraviolet-visible spectroscopy^[30], ultraviolet photoelectron spectroscopy^[30], Hall coefficient measurements^[30], density functional theory calculations^[31] and so on, have been introduced to determine the band structures and prove the BIEF effects on SIMFCs. The different characteristics of SIMFCs compared to SOFCs are obvious because of the use semiconductor-based membranes vs. conventional electrolytes.

In an i-n heterojunction, an i-n type interface contact is constructed, e.g., a CeO₂/CeO_2-δ core-shell heterostructure, where CeO₂ is an intrinsic i-type semiconductor and CeO_2-δ is an n-type semiconductor. A charge separation occurs at the CeO_2-δ/CeO₂ interface. The electrons can transfer from the shell to the core while it is forbidden to pass through from the core to the shell. After the CeO₂/CeO_2-δ core-shell heterostructure forms, an electron depletion region is built on the CeO_2-δ side at the interface and an electron accumulation region is simultaneously formed on the CeO₂ side of the interface. On this basis, the local charge distribution and the electric field or the BIEF are formed among particle surfaces, which stop protons from migrating deep inside the shell to pass through the interface and suppress the bulk infiltration of the surface protons, as shown in Figure 2B. Benefiting from the BIEF, a “proton shuttle” is constructed in the continuous highly conducting regions formed in the ceria-semiconductor membrane of the SOFC.

To further understand the new mechanism in SIMFCs, Xia et al.^[30] constructed a BaCo_0.4Fe_0.4Zr_0.1Y_0.1O_3-δ (BCFZY)-ZnO p-n heterostructure for low-temperature SOFCs. They found that the as-prepared heterostructure exhibits a hybrid H⁺/O^2- conducting capability during fuel cell operation. When two semiconductors with different energy band levels are combined, conduction band offset (ΔEc) and valence band offset (ΔEv) will be induced, thus forming potential barriers to form the BIEF. To explain its mechanism, a BIEF in the as-prepared BCFZY-ZnO p-n heterostructure was introduced, where the junction prevented the electron passing through and the ionic conductivity can be enhanced by the BIEF, as illustrated by Figure 3.

Figure 3. Schematic diagram of a typical p-n heterojunction formed at the heterophasic interface of a p-type (BCFZY)-n-type (ZnO) semiconductor membrane and the corresponding energy band alignment mechanism proposed for interpreting the charge separation and ionic transportation process^[30]. BCFZY: BaCo_0.4Fe_0.4Zr_0.1Y_0.1O_3-δ; NCAL: Ni_0.8Co_0.15Al_0.05LiO_2-δ; CB: conduction band; VB: valence band.

Cai et al.^[32] explained the SIMFC device from the conventional three layers of anode/electrolyte/cathode to a “three in one” membrane, as shown in Figure 4. A semiconductor-ionic membrane based on NiO-yttrium-stabilized zirconia (YSZ)-LSCF composites was used to construct a SIMFC device. Under H₂/air conditions, the SJ BIEF is formed that can avoid the electronic short-circuit problem. As a result, A high power density of 680 mW·cm^-2 at 600 °C was achieved with an open circuit voltage of 1.11 V.

Figure 4. From conventional three layered fuel cells to a “three in one” membrane with a BIEF. BIEF: Built-in electric field.

Encouraged by this new concept, Cai et al.^[33] further constructed a bulk heterostructure nanocomposite electrolyte of Ce_0.8Sm_0.2O_2-δ-SrTiO₃ for low-temperature SOFCs, exhibiting a peak power density of 892 mW·cm^-2 with an open circuit voltage of 1.1 V at 550 °C. Generally, SIMFCs are often composted by a semiconductor and ionic conductor, where the heterostructure plays a key role in achieving the high performance. To obtain a deep scientific understanding of SIMFCs, Zhang et al.^[34] reviewed superionic conductivity in ceria-based heterostructure composites for low-temperature SOFCs. Hu et al.^[35] discussed recent research and development in junctions and energy bands on novel semiconductor-based fuel cells.

Overall, significant progress has been achieved in the field of SIMFCs due to the innovation of semiconductor materials as membranes and advances in forming the BIEF. A high power density of ~1000 mW·cm^-2 at 550 °C has been achieved from reported SIMFCs with BIEF effects for low-temperature SOFCs. Previous studies on SIMFCs are highlighted in Figure 5.

Figure 5. Summary of previous work on SIMFCs. LCCO-ZnO: La/Pr co-doped CeO₂-ZnO; LSCF-SCDC: La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ-Sm and Ca co-doped CeO₂; MLCO-SDC: Mg-doped LiCoO₂-Sm doped CeO₂; CeO₂/CeO_2-δ: the core hall structure of the CeO₂/CeO_2-δ; BCFZY-ZnO: BaCo_0.4Fe_0.4Zr_0.1Y_0.1O_3-δ-ZnO; LNO-SDC: LaNiO₃-Sm doped CeO₂; NSTO: Nb-doped SrTiO_3-δ; CZO-SDC: Co_0.2Zn_0.8O-Sm-doped CeO₂; BCFCT-YSZ: BaCo_0.2Fe_0.1Ce_0.2Tm_0.1-Zr_0.3Y_0.1O_3-δ; BSFS-SDC: BaSrFeSbO₃-Sm doped CeO₂.

As shown in Table 1, a V_oc of > 1.0 V was individually realized in different heterostructures, suggesting that the performance of devices can be fully improved by the BIEF produced from semiconductor heterostructures. To achieve this goal, the development of high-performance SIMFCs is of significant importance. It is encouraging that the long-term stability of SIMFCs can reach over 100 h, according to recent reports [Table 2]. We believe that the as-reported SIMFCs will provide a new route for SOFC research and development towards commercialization. Compared with traditional SOFCs, SIMFCs exhibit various advantages, e.g., high ionic conductivity resulting in higher performances, including current and power outputs at low temperature, simple structures (three in one) and so on. However, in future research and development, some critical perspectives are suggested:

(1) Long-term stability is currently absent from engineering efforts with regards to commercialization. More effort should be contributed to the engineering and scaling-up of SIMFCs;

(2) In order to develop long-term SIMFC durability, the development of compatible electrode materials has made good progress;

(3) Relevant theoretical models and calculations should be employed to guide further research and development. In particular, the physical properties and effect of the BEIF formed from various heterojunctions, e.g., bulk and planar p-n, Schottky, n-i (intrinsic or insulating) junctions;

(4) Some new technologies and technical processes combined with SOFCs and protonic ceramic fuel cells should be introduced to develop durable SIMFCs.

The first demonstration of SOFC technology was made in the 1930s by Baur and Preis^[36] and used zirconia stabilized with 15 wt.% of yttria (the so-called Nernst Mas) as the electrolyte, iron or carbon as the anode and magnetite (Fe₃O₄) as the cathode. Long durability could not be achieved until the compatible electrodes, NiO-YSZ cermet and especially perovskite oxide cathode materials, were discovered and technically developed to incorporate with the YSZ electrolyte. This took several decades. Compatible electrodes for semiconductor-ionic material membranes have yet to be employed into SIMFCs. Nevertheless, significant progress has been made in this area.

It is also noteworthy that SIMFCs are built not only on electrochemistry but also semiconductor physics^[37]. We expect that this report can arouse significant attention from related research fields and disciplines to overcome the bottleneck of SOFC commercialization.