Perovskite Al-SrTiO3 semiconductor electrolyte with superionic conduction in ceramic fuel cells

Perovskite oxidedoping may modulatetheenergy bandgap strongly a ﬀ ecting the oxygenreduction activityand electrical properties with high promise for use as a low-temperature solid oxide fuel cell (LT-SOFC) electrolyte. Here, we show that a small amount of Al-doping into SrTiO 3 (cid:1) d may tune the energy band structure of SrTiO 3 (cid:1) d , triggering the electrochemical mechanism and fuel cell performance. The synthesized SrTiO 3 (cid:1) d and Al-SrTiO 3 (cid:1) d electrolytes are sandwiched between two symmetrical electrodes (Ni foam pasted NCAL). Ni-NCAL/SrTiO 3 (cid:1) d /NCAL-Ni and Ni-NCAL/Al-SrTiO 3 (cid:1) d /NCAL-Ni structures delivered a maximum power density of 0.52 W and 0.692 W and high ionic conductivity of 0.11 S cm (cid:1) 1 and 0.153 S cm (cid:1) 1 , respectively, under H 2 / Air atmosphere at low operational temperature of 520 (cid:3) C. Highways of ion transport along the surface and grain boundary (interface) are identi ﬁ ed as the main reason for good oxygen ion conduction. X-ray di ﬀ raction (XRD), scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HR-TEM), UV visible (UV), and X-ray photoelectron spectroscopy (XPS) were performed to investigate the structural, electrochemical, morphological, surface and interfacial properties of Al-SrTiO 3 (cid:1) d . The obtained results suggest that a certain amount of Al (20%) doping into SrTiO 3 (SrTi 0.8 Al 0.2 O 3 (cid:1) d ) improves the fuel cell performance and is a promising electrolyte candidate for LT-SOFC.


Introduction
Solid oxide fuel cells (SOFCs) convert chemical energy directly into electrical energy. The advantages of SOFCs include fuel exibility, high-energy conversion efficiency, and environmental friendliness. Key challenges for the commercialization of SOFC include lowering its operating temperature (700-1000 C) to improve its stability and reducing material costs while maintaining high performance at the same time. 1 Furthermore, other material aspects, such as bulk diffusion, are important in affecting the ionic transport in the SOFC electrolyte. Gadoliniadoped ceria (GDC) or Y 2 O 3 stabilized ZrO 2 (YSZ) are examples of electrolyte materials used in SOFC, but they mainly are affected by structural doping and the operational temperature of SOFC. [2][3][4][5] Various advanced thin-lm techniques have been employed to manufacture thin-lm YSZ-based electrolytes to reduce the operating temperature, but this has also encountered challenges with high cost, long duration of production, and scaling up. 6-10 Therefore, the search for alternative electrolyte materials with good ionic properties is of high importance and a challenge without any denitive explanations. [6][7][8][9][10][11][12] To tackle these issues, the interfacial diffusion phenomenon is a promising approach. [13][14][15][16] Mushtaq et al. reported a composite SrFe 0.75 Ti 0.25 O 3Àd /Sm 0.25 Ce 0.75 O 2 (SFT/SDC) heterostructure with high ionic conductivity of >0.1 S cm À1 achieving an impressive fuel cell performance of 920 mW cm À2 at 520 C. This composite combines a semiconductor (SFT) and ionic conductor (SDC) material to create interfacial mechanisms to obtain such high ionic conductivity. 13 A heterostructure composite of semiconductor Ba 0.5 Sr 0.5 Co 0.1 Fe 0.7 Zr 0.1 Y 0.1 O 3Àd (BSCFZY) with an ionic conductor Ca 0.04 Ce 0.80 Sm 0. 16 O 2Àd (SCDC) as a functional electrolyte has demonstrated enhanced ionic conductivity of 0.22 S cm À1 and high-power density of 900 mW cm À2 explained by the inuence of interfacial space charge regions and formation of built-in electric eld (BIEF) at the interface. 14 In addition, an attractive triple-conducting BaCo 0.4 Fe 0.4 Zr 0.1 Y 0.1 O 3Àd electrolyte has been reported, in which an effective p-n junction with BCFZY-ZnO was created to inhibit the electronic conduction and to enhance the charge transport at the interface. 15 A power density of 643 mW cm À2 was achieved at 550 C. Also, many single-phase semiconductors have been suggested as electrolyte membranes for advanced low-temperature solid oxide fuel cells (LT-SOFCs). For example, the semiconductor Li x Co 0.5 Al 0.5 O 2 has been used as an electrolyte, delivering a power density of 180 mW cm À2 at 525 C and protonic conductivity of 0.1 S cm À1 at 500 C. 16 Such high proton conductivity is affected by the incorporation of H as fuel into the fuel cell. 17 A similar example is perovskite SmNiO 3 (SNO), which when used as an electrolyte delivered a maximum power density of 225 mW cm À2 with high protonic conductivity at 500 C. 17 The incorporation of protons into the SNO perovskite electrolyte also causes Mott-transition transforming metallic conductor into a proton conductor, 17 H-SNO, which is mainly caused due to the coulombic interactions of electrons. 17 Such semiconductor electrolytes for the fuel cell can be categorized in the so-called single-layer fuel cell (SLFC) category. Also, perovskite materials such as SrTiO 3Àd have attracted extensive attention due to suitable energy bandgap, and physical, chemical, and electrical properties for the application of photocatalysts and fuel cells. Chen et al. have proposed SrTiO 3Àd as an electrolyte and delivered better fuel cell performance of 0.62 W cm À2 at 550 C. Also, several interesting single-layer semiconductors and their hybrid heterostructures have been reported to boost the ionic conductivity of the fuel cell at lower operating temperatures. [18][19][20][21][22][23][24][25][26][27] The above-stated literature claims that semiconductor electrolytes could be useful in fuel cell applications. Also, SrTiO 3 with a bandgap of 3.2 eV has proven to be worthy due to its enriched electrical, physical, and optical properties for various technologies, such as solar cells, batteries, photocatalysis, and fuel cell. According to previous studies, the low valence doping of different elements such as Al 3+ into Ti 4+ is more effective in boosting the electrical conductivity of SrTiO 3 . 27-29 Moreover, to the best of our knowledge, there is no experimental data have been reported to investigate the effect of Al doping on the energy bandgap and electrical properties for fuel cell application.
To close this knowledge gap, herein, we report a new promising perovskite material Al-SrTiO 3Àd (SrTi 0.8 Al 0.2 O 3Àd ). Unlike the SNO, the Al-SrTiO 3Àd (SrTi 0.8 Al 0.2 O 3Àd ) is almost an insulator in the as-prepared state but becomes a great ion conductor under the fuel cell conditions, resulting in excellent fuel cell performance of 0.692 W at 520 C. This greatly enhanced electrical property is attributed to a structural change via doping and a low grain boundary resistance mechanism related to the formation of a heterostructure semiconducting core with a superionic surface layer.

Synthesis and fabrication of cells
The solid-state method was used to synthesize Al-SrTiO 3Àd (SrTi 0.8 Al 0.2 O 3Àd ) electrolyte powders. The perovskite Al-SrTiO 3Àd was prepared by grinding precursors of SrCO 3 , TiO 2, and Al 2 O 3 oxides (Sigma Aldrich 99.9% purity). Initially, all oxides materials with appropriate amounts were mixed step by step, consequently grounded, and then ethanol was mixed to obtain a uniform and satisfactory solution. Thereaer, the sample was heated at 120 C to dry the electrolyte powder and then ground for 5 hours to obtain purely consistent material subsequently sintered at 1100 C for 6 hours. Besides, STO (SrTiO 3Àd ) was prepared by the sol-gel technique; the description can be found elsewhere. 28 Also, SDC and BZY were prepared through a sol-gel method, details synthesis can be found elsewhere. 29 Both pellets were sintered at 1500 C for 5 hours to obtain the ne structure of SDC and BZY, respectively. The whole synthesis procedure of Al-STO is presented in Fig. 1.

Material characterization
X-ray diffraction (XRD) was performed using an X-ray diffractometer, which contained the Cu (K-alpha) as a source of radiation to elucidate the phase of STO (SrTiO 3Àd ) and Al-doped STO (Al-SrTiO 3Àd ). Field emission-scanning electron microscopy (FE-SEM) was performed to investigate the morphologies of the materials. High resolution-transmission electron microscopy (HR-TEM, Tecnai, G2 F30) was performed at an accelerating voltage of 300 kV to investigate the morphology in detail. X-ray photoelectron spectroscopy was deployed to examine the apparent chemical properties of STO (SrTiO 3Àd ) and Al-doped STO (Al-SrTiO 3Àd ) materials either in terms of the oxidation state (O-S) or chemical species of surface atoms. UV-visible (UV-3600 spectrometer, MIOSTECHPTY Ltd.) and Ultraphotoelectron spectroscopy studies were performed to manipulate the energy bandgap and valence band maximum position.

Assembly and measurements of the pellet
Later, to fabricate the pellet of the desired electrodes, NCAL powder was purchased from the Tianjin Bamo Sci and Tec Joint-Stock Ltd., China. Also, Ni-foam was cropped and used to collect the current and support the cell. NCAL powder was appropriately mixed with a terpinol binder to obtain a ne and viscous slurry. Subsequently, the prepared slurry was brushed on Nifoam and then cooked at 80 degrees for 30 min for drying. Furthermore, Ni-foam pasted NCAL electrodes were censored into a pellet of 13 mm in diameter. Then, SrTiO 3Àd and Al-SrTiO 3Àd electrolyte powders were sandwiched between the Ni-NCAL electrodes and compressed under the 360 MPa to obtain a button-shaped pellet of 13 mm diameter and an active area of 0.64 cm 2 . The thickness of the prepared pellet was 1.5 mm, while the thickness of the electrolyte was 750 mm. Furthermore, two bilayer electrolyte cells were prepared with STO using SDC and BZY oxide ion and proton conductor in the following conguration of Ni-NCAL/SDC-STO/NCAL-Ni and Ni-NCAL/BZY-STO/NCAL-Ni to distinguish the charge carriers in the SrTiO 3Àd electrolyte. The prepared pellets were pressed at 360 MPa, with an active area of 0.64 cm 2 .
Hydrogen fuel (at 80-120 ml min À1 of ow rate) and air (at 150-200 ml min À1 ) were used to realize the functionality of the fuel cell. The current-voltage (I-V) and current-power (I-P) characteristic curves were plotted using the electronic load (IT8511, ITECH Electrical Co., Ltd) to analyze the fuel cell performance. Moreover, Electrochemical Impedance (EIS) spectra (Gammry-reference 3000, USA) were collected to determine the electrochemical properties under H 2 /air in the given range of frequency (from 0.1 Hz to 1 MHz) and amplitude of (10 mV). indicating SrTiO 3Àd , Al-SrTiO 3Àd belong to a pure cubic crystalline structure. Interestingly, this pattern is according to the previously reported literature. 28,29 Also, a pure cubic phase represents that Al is well doped into SrTiO 3Àd . The lattice parameters of SrTiO 3 and Al-SrTiO 3 were calculated using the Rietveld-renement to be (0.38) nm and (0.3902) nm, respectively. Also, it has been noticed that doping of Al into SrTiO 3Àd caused shiing of the peaks toward lower or higher angles, which mainly leads to an increase/decrease in the lattice constant. Also, it happened due to the difference in the ionic radius of the dopant Al (54 pm) and the host Ti (61 pm) elements, which strictly follow Vegards rule. 29 Also, it has been speculated that based on the ionic radius, Al 3+ is most likely to substitute Ti 4+ , as conrmed elsewhere. 29 Moreover, aer testing the scrapped power, the XRD pattern conrmed the formation of amorphous phase LiCO 3 , as displayed in the supplementary information (Fig. SI1). † Moreover, the morphologies of SrTiO 3Àd and Al-doped SrTiO 3Àd were investigated using the SEM (scanning electron microscopy) technique. The ImageJ soware was used to  This journal is © The Royal Society of Chemistry 2022

Structure and morphology
Sustainable Energy Fuels estimate the particle size from SEM images of the uniformly distributed particles. The particle sizes of SrTiO 3Àd and Al-SrTiO 3Àd lie in the range of 100-300 nm and 150-300 nm, respectively. SEM results conrmed the coherent distribution and modication due to the incorporation of Al into SrTiO 3Àd , as exhibited in Fig. 2(b-d). All particles are homogeneously distributed and well connected in the obtained structure of Al : STO. Also, the homogeneity and better connection among the particles make them more favorable for better and quick transportation of charges in the electrochemical reaction of a fuel cell. 30 Fig. 2(e) shows the EDS of Sr, Al, Ti, and O, which conrms that the particles are spread evenly even aer doping, revealing the uniformity, which affirms that Al was doped into SrTiO 3Àd (SrTi 0.8 Al 0.2 O 3Àd ). 28 Also, the TEM (transmission electron microscopy) technique was used to study the detailed morphology of SrTiO 3Àd and Aldoped SrTiO 3Àd, as depicted in Fig. 3(a-c). These TEM images conrmed that the Al-doping into SrTiO 3Àd is successful, which probably is benecial for ionic transport and fuel cell performances to be discussed in a later section. Fig. 3(d and g) shows the HR-TEM of Al-SrTiO 3Àd powder scrapped from the cell aer testing, leading to a clear view of the core-shell structure. Moreover, Fig. 3(e) is the illustration of the HR-TEM image of the Al-SrTiO 3Àd (SrTi 0.8 Al 0.2 O 3 ) lattice fringe (110) facet of 0.278 nm, which is less than the STO facet (0.275), conrming the replacement of Ti with Al leading to the expansion of the unit cell volume of the lattice Al-STO, as noticed in XRD. 28,29 Fig. 3(f) shows the SAEED pattern of the Al-SrTiO 3Àd lattice. Also, SEM and TEM images of SrTiO 3Àd and Al-SrTiO 3Àd are shown in the ESI Fig. SI2(a and b). †

XPS analysis
Furthermore, X-ray photoelectron spectra (XPS) were performed on STO and Al-SrTiO 3Àd to analyze surface properties and changes in electronic structures brought about by the Fig. 3 (a-c) HR-TEM images of Al-SrTiO 3Àd with different magnification scales, (d and g) are the HR-TEM view of the formation of core-shell structure after testing with different resolution scales of 10 nm and 20 nm, while (e and f) lattice fringes of Al-SrTiO 3Àd particles and SAEED pattern of Al-SrTiO 3Àd , respectively.

Sustainable Energy Fuels
This journal is © The Royal Society of Chemistry 2022 introduction of the Al lattice, as shown in Fig. SI3. † The XPS spectrum of SrTiO 3Àd with the Ti 2p core-level prepared using the sol-gel technique is illustrated in Fig. 4(a), which aligns well with the reported literature spectrum of SrTiO 3Àd . 31,32 The peaks of the Ti 4+ component located at 458.6 and 464.4 eV are allocated to Ti 2p 3/2 and Ti 2p 1/2, respectively. In contrast, the second doublet located at 457.2 and 463.1 eV gives evidence of the existing Ti 3+ species in SrTiO 3Àd , as shown in Fig. 4(b). 33 From relative peak ratios, mostly titanium is present (72.8%) as Ti 4+ . In comparison, Ti 3+ is available in a lower ratio of about 27.2%, which might appear on the surface in abundance because SrTiO 3Àd with a greater concentration of Ti 3+ is heavily colored. However, in this study, it appears almost white, and these results are well-matched with the reported data. 29 In contrast, a single spin-orbital doublet with the features of Ti 4+ was noticed in the spectrum of Al-doped SrTiO 3Àd , but there was no sign of Ti 3+ leading to the fact of Al contribution into the STO lattice prevented the involvement of Ti 3+ ions, as hypothesized in the literature. 34 In addition, Fig. 4(c and d) shows the Ospectra of SrTiO 3Àd and Al-SrTiO 3Àd , which suggested that the incorporation of Al 3+ ions leads to the modied O1s region of SrTiO 3Àd . The peak at 529.8 eV of the SrTiO 3Àd lattice is considered the prominent peak assigned to O 2À ions in the lattice, while the peak at 531.9 goes to hydroxyl-groups at the surface of particles allotted in previous reports. [35][36][37] Moreover, the peak at 528.6 eV is attributed to the lattice of O 2À ions, which are tentatively bonded with Ti 3+ ions in the lattice of SrTiO 3Àd . This clar-ication has been conrmed by the disappearance of the peak at 528.6 eV in the Al-SrTiO 3Àd lattice, which no longer contains any Ti 3+ ions. The peaks at 531.3 eV and 529.2 eV are allotted to the surface hydroxyl groups and lattice oxide O 2À ions, suggesting an increase in the surface hydroxyl groups in Al-SrTiO 3Àd . The peak at 531.8 eV in the Al-SrTiO 3Àd lattice is tentatively attributed to the oxide ions bonded to Al 3+ ions near the surface. 38 The peak at 533.1 eV belonged to SrCO 3 , which might be produced as a side product during the synthesis process. 39 Generally, a 0.5-0.6 shi in all peaks toward lower binding energy has been revealed aer a detailed inspection of Ti and O core-level spectra, and this change in B.E. might be attributed to the alternation in the charge distribution leading to enhancement of the oxygen vacancies, as a result, enhancing of the ionic conductivity. 40,41 Moreover, the peak area was expanded upon Fig. 4 (a-d) shows the XPS spectra of Ti 2p and oxygen O1s of SrTiO 3Àd and Al-SrTiO 3Àd while (e and f) display the XPS spectra of O1s and carbon before and after testing of Al-SrTiO 3Àd , (g) XPS spectra of Al-SrTiO 3Àd after testing.
This journal is © The Royal Society of Chemistry 2022 Sustainable Energy Fuels doping, resembling the creation of more O-vacancies and high ionic conduction channels. Fig. 4(e and f) reveals the XPS spectra of C1s and O1s of the prepared sample Al-SrTiO 3Àd powder before and aer the performance. The peak at 284.2 eV of C1s was assigned to C-H, while the peak at 288.9 eV was attributed to CO 3

2À
. XPS spectra of C1s give the clue of carbonate existence on the surface of Al-SrTiO 3Àd aer the fuel cell performance. Moreover, Fig. 4(g) shows the XPS spectra of Al-SrTiO 3Àd with the indication of the Li-1s peak, conrming the formation of Li 2 Co 3 aer fuel cell measurements. From the O1s spectra, the existence of the lattice oxygen hydroxyl groups and oxygen defects were noticed, as shown in Fig. 4(c and d). While the spectra aer the performance revealed the presence of oxygen with a high oxidation state in abundant O-vacancies, OH groups, and carbonates appeared on the surface of Al-SrTiO 3Àd leading to enhancing the ionic conductivity. 42

Fuel cell performance
To evaluate the fuel cell performance of SrTiO 3Àd and Al-SrTiO 3Àd electrolytes, the symmetrical electrodes (anode and cathode) NCAL were shaped into pellets in the following conguration of Ni-NCAL/SrTiO 3Àd and Al-SrTiO 3Àd /NCAL-Ni, as illustrated in Fig. 5(a and b). The electrochemical performance was evaluated in different gas H 2 /air environments at low operating temperatures of 420-520 C. The constructed device (Ni-NCAL/Al-SrTiO 3Àd /NCAL-Ni) conveyed the outstanding OCV 1.05, 1.03, and 1.04 V at 420, 470, and 520 C, respectively. Also, a peak power density (PPD) of 0.692 W was obtained utilizing the Al-SrTiO 3Àd electrolyte, superior to the pure SrTiO 3Àd (0.52 W) electrolyte at a low-temperature 520 C. It can be perceived that such high performance is typically attained in traditional SOFCs using thin-lm electrolytes. 43 The cell showed excellent performance, and some credit for the high performance goes to the composite Ni-NCAL electrodes. 44 Also, the obtained power density is much higher than that reported for titanate-based materials, such as STO, Fe-SrTiO 3, and Nb-SrTiO 3 , revealing the reliability of Al-SrTiO 3Àd as a competent electrolyte for LT-SOFCs. 17,28,30 The question is, how and why should Al-SrTiO 3Àd be used as an electrolyte with such high performance? Well, it has been reported that the Al-SrTiO 3Àd is an electronic conductor in a reducing atmosphere, 28 and the OCV agrees with the theoretical Nernst potential. 29,30 However, when H 2 and air are injected into both sides of the pellets, the OCV starts to increase, reaching 1.1 V. Such a high OCV indicates that neither the gas leakage nor the short-circuiting occurred in our device, guaranteeing a peak power density (PPD) of 0.692 W. The OCV Fig. 5 (a-c) Fuel cell performance of SrTiO 3Àd , Al-SrTiO 3Àd at different operating temperatures 420-520 C and repeated performance of Al-SrTiO 3Àd at 520 C, (d) reveals the cross-sectional view of the Al-SrTiO 3Àd electrolyte with surrounded symmetrical electrodes NCAL.

Sustainable Energy Fuels
This journal is © The Royal Society of Chemistry 2022 has shown the operating temperatures upgrading behavior; similarly, as temperature decreases, OCV increases, which follows the Nernst equation. 28,30 The higher power density and the OCV with the temperature variation avoid the shortcircuiting issue due to the formation of the interface junction of the core-shell heterostructure, doping, and the Schottky junction. 25,28,29 It has been speculated that the formation of the Schottky junction among the anode and electrolyte interface inhibits electronic conduction and enhances ionic transportation. 28 Fig. 5(c) exhibits the I-V/I-P characteristic curves of Ni-NCAL/Al-SrTiO 3Àd /NCAL-Ni in an H 2 /air environment delivering almost the same power density $0.692 W each time of repeating 4 to 5 times at 520 C. Fig. 5(d) displays the crosssectional view of the NCAL/Al-SrTiO 3Àd /NCAL cell surveyed via SEM (scanning electron microscopy), where Al-SrTiO 3Àd is well sandwiched and distinguished among the porous symmetrical electrodes (NCAL), which guarantees better performance. 45 NCAL has been recognized as an excellent electrode candidate either as a cathode or anode in search of better material with excellent ionic and electronic conductivity. Comparatively, NCAL has better catalytic activity and excellent redox (HOR and ORR) reactions in the anode and cathode zones, respectively. 45 The credit for the higher OCV goes to the good catalytic activity of electrodes NCAL (anode and cathode). Higher power and enhanced OCV evidenced the feasibility of Al-SrTiO 3Àd as a capable potential electrolyte in LT-SOFCs. All results imply that Al-SrTiO 3Àd is a better electrolyte, but still much more effort is needed, especially engineering ones (thin-lm techniques), and the heterostructure of Al-SrTiO 3Àd prepared using the thin-lm technique is in desperate demand to support the most advanced technology.

EIS analysis and ionic conductivity
Moreover, electrochemical analysis (EIS) was performed to further study the electrical properties of SrTiO 3Àd and Al-SrTiO 3Àd under the OCV conditions with an H 2 /air environment at different operating temperatures (420-520 C), as depicted in Fig. 6(a and b). On the real axis, rst, the intercept with a higher frequency corresponds to the ohmic resistance. It gives different values (0.32, 0.44, 0.64) for STO and (0.18, 0.20, 0.29) for Al-SrTiO 3Àd under different operating temperatures 520-420 C. NCAL has been recognized as a better electronic conductor, and when it combines or is pasted on Ni foam, it even delivers superior conductivity. 45 It means that our investigated cells' ohmic resistance should come from the SrTiO 3Àd /Al-SrTiO 3Àd electrolytes. Polarization resistance is dened as the difference between the high frequency and low-frequency intercept on the EIS curve's real axis, composed of several overlapping suppressed arcs. These suppressed arcs represent the physical and chemical processes corresponding to the HOR (hydrogenoxidation reaction) and ORR (oxygen-reduction reaction) at the anode and cathode sides. All the suppressed semi-circles were tted by applying the following circuit (LR o (QR 1 )(QR 2 )) in the ZSIMPWIN soware, where R 1 manifests the charge transfer resistance (grain-boundary resistance), R 2 corresponds to the mass transfer resistance, and R o represents the ohmic resistance. The detailed values of the tted circuit are shown in the supplementary information (Table SI 1 and 2). † The decline in the charge transfer resistance of the cell might be followed by the capacitance (C i ), which can be determined using the equa- R i corresponds to Q i, and n, represents the frequency power related to the Q value [0 < n < 1]. [43][44][45] The obtained capacitance and resistance values showed declining behavior, suggesting the enhanced charge transfer at the interface and the surface, leading to peak power density (PPD) and high ionic conductivity. The above results imply that doping causes to create more O-vacancies leading to higher ionic conductivity. Also, the presence of LiCO 3 was evidenced from the anode side of NCAL as Ni-NCAL was reduced to Ni, and Li-Co-Al and formed LiCO 3 by H 2 at the anode side. It might be possible to infer that the formed LiCO 3 plays an essential role in easing the conduction of oxygen ions (O 2À ). Overall, oxygen vacancies induced by doping and surface conduction of the core-shell interface play a critical role in boosting the ionic conductivity. 28 The mechanism of the core-shell heterojunction will be discussed in the later section.
The ionic conductivity of Al-SrTiO 3Àd is high enough for the reported heterostructure electrolyte materials in the fuel cell Fig. 6 (a and b) Electrochemical impedance spectroscopy curves of SrTiO 3Àd and Al-SrTiO 3Àd in H 2 /air environment at different operating temperatures 420-520 C.
This journal is © The Royal Society of Chemistry 2022 Sustainable Energy Fuels devices, especially at low operating temperatures. Therefore, to investigate the ionic conduction and charge transfer mechanism of Al-SrTiO 3Àd electrolyte, different gas atmospheres were provided to analyze the cell through changes in the impedance spectrum. At rst, the air-air atmosphere was inserted before the fuel cell performance to examine the ionic conductivity at 520 C. As a result, a giant arc appears, composed of many suppressed arcs with a total impedance approaching 500 U cm 2 , as depicted in Fig. 7(a). Such high resistance under the air-air atmosphere suggests low ionic conductivity in the current state.
In a while, one side of the air was replaced with H 2 and le for 10 minutes to ow appropriately. Aerward, it yielded lower ohmic resistance and polarization resistance at 520 C under H 2 /air environment, which is generously lower than the air/air atmosphere, as shown in Fig. 7(b). Later, H 2 in the cell was purged using N 2 and air for almost 1 hour, and then the air/air environment was repeated with EIS, which yielded virtually the same ohmic resistance as in H 2 /air environment; also, the total resistance did not repeat to the 500 ohm cm 2 leading to the fact that oxygen ions might be the charge carriers in Al-SrTiO 3Àd . 22,28 Such a high carrier concentration owes to the doping of Al into SrTiO 3Àd and the formation of the core-shell structure in a single-phase material for the fuel cell device, also, such results manifest that Al-SrTiO 3Àd holds a signicant capability to work as a potential electrolyte with better ionic conductivity for SOFCs. 28 In addition, two double-layer electrolytes were prepared to identify the charge carriers in Al-SrTiO 3Àd using the ion ltration experiment. SDC and BZY were synthesized using co-precipitation and sol-gel techniques as oxide ion and proton conductors, respectively. The prepared cells of the following conguration Ni-NCAL-SDC/Al-SrTiO 3Àd -NCAL/Ni & Ni-NCAL-BZY/Al-SrTiO 3Àd -NCAL/Ni were used to construct the I-V/I-P curves, as demonstrated Fig. 7(c). Ni-NCAL-SDC/Al-STO-NCAL/ Ni and Ni-NCAL-BZY/Al-STO-NCAL/Ni delivered the maximum power output of 180 mW cm À2 and 149 mW cm À2 with better OCVs of 0.97 and 1 V, respectively. The ionic conductivities of SDC and BZY were lower than that of Al-SrTiO 3Àd , as reported elsewhere. 27 The obtained power output of both double-layer electrolytes proved that Al-SrTiO 3Àd contains proton and oxide ion carriers.
Moreover, the ionic conductivity was elucidated using the I-V curve. The central region of the I-V curve gives the clue of ionic resistance of ohmic resistance, which was extracted to investigate the ionic conductivity of 0.153 S cm À1 at 520 C as shown in Fig. 7(d). 45 The obtained ionic conductivity of 0.153 S cm À1 is higher than STO, suggesting that doping and surface conduction are causing to enhance the ionic conduction. Moreover, the attained ionic conductivity is higher than the reported values for titanate-based materials (SrTiO 3 , Fe-SrTiO 3 ), which have increased the worth of Al-STO as a capable electrolyte for the fuel cell application. 17,28,30 The high ionic conductivity of 0.153 S cm À1 and the low activation energy of 0.31 eV indicate the capability of Al-SrTiO 3Àd as an electrolyte for SOFC devices at low operational temperatures. Such high ionic conduction is mainly a result of the formation of the core-shell heterojunction. 26,27 Fig. 7 (a and b) shows the EIS analysis under air-air and under H 2 -air, air-air environment at 520 C, while (c and d) exhibits the bilayer electrolyte performance of Al-STO with SDC and BZY at 520 C and ionic conductivity of STO and Al-STO at different operating temperatures 420-520 C.

Sustainable Energy Fuels
This journal is © The Royal Society of Chemistry 2022

Optical features and junction mechanism
To investigate the charge transport behavior in energy devices, exclusively in fuel cell devices, the energy band theory should be adopted to design the energy bands of materials from a different perspective. The energy bandgaps of SrTiO 3Àd , Al-SrTiO 3Àd , and Al-SrTiO 3Àd @carbonates aer the performance were investigated using the UV-analysis, as depicted in Fig. 8(a  and b). The absorption edge spectra for the SrTiO 3Àd electrolyte material were noticed at 240 nm. At the same time, Al-SrTiO 3Àd and Al-SrTiO 3Àd @carbonates or aer the performance seem to move toward a higher wavelength (red-shi), as hypothesized in previous reports. 28 According to the relationship between h (eV) and (ahy) 2 , Tauc plots were plotted and are displayed in Fig. 8(a  and b). The following relation was used to infer the energy bandgap. 29 The details of the above equation parameters are explained elsewhere. 29 The obtained energy bandgaps 3.2, 3.18, and 3.11 eV correspond to SrTiO 3Àd , Al-SrTiO 3Àd , and Al-SrTiO 3Àd @carbonates or aer the performance. The calculated energy bandgap is well matched to that published previously. 28,29 Aer incorporating Al in SrTiO 3Àd , the bandgap was reduced. Aerward, we noticed that aer the performance, the bandgap of Al-SrTiO 3Àd tended to decrease, which might have occurred due to the creation of intermediate states between the valence and conduction band. Moreover, reduced energy bandgap is mainly ascribed to creating O-vacancies in the lattice and at the surface and interface. This O-vacancies creation phenomenon leads to the establishment of the above-mentioned energy levels, which primarily affect the Fermi-level and energy bandgap. 43 Moreover, ultraphotoelectron spectroscopy (UPS) was used to determine the valence band values, which were 7.37 and 8.3 eV for Al-SrTiO 3Àd and Al-SrTiO 3Àd @carbonates, respectively, as displayed in (Fig. SI 4). † Fig. 8(c and d) shows the HR-TEM images identifying the core-shell heterojunction between the two phases and the energy band diagram constructed based on the inferred bandgap for Al-SrTiO 3Àd and Al-SrTiO 3Àd @carbonates. The band alignment between Al-SrTiO 3Àd or phase 1 and Al-SrTiO 3Àd @carbonates or phase 2 was established due to the reduction in the bandgap aer the performance in phase 2. The amorphous carbonate phase might accumulate on the surface of bulk Al-SrTiO 3Àd or phase 1 and permeate into the supercial Fig. 8 (a and b) Energy bandgaps of SrTiO 3Àd , Al-SrTiO 3Àd before and after the performance (c) s the HR-TEM image of core-shell, while (d) displays the junction mechanism of Al-SrTiO 3Àd and Al-SrTiO 3Àd @carbonates core-shell heterojunction.
This journal is © The Royal Society of Chemistry 2022 Sustainable Energy Fuels phase 1 to establish the core-shell heterostructure, as shown in Fig. 8(c). The core-shell heterojunction between phases 1 and phase 2 can be formed, leading to separating the electron-hole pairs and producing BIEF (build-in electric eld), as shown in Fig. 8(d). Reasonably, the junction between phase 1 and 2 leads to enhanced ionic conductivity, and BIEF blocks the electronic movement and facilitates fast charge transportation, as in the P-N junction. 14,15 It can be inferred that the core-shell heterojunction is the driving force to enhance the ionic conductivity and performance, and Al-SrTiO 3Àd holds the capability to be a capable electrolyte for advanced fuel cell technology. Fig. 9(a and b) shows SrTiO 3Àd and Al-SrTiO 3Àd structures in the b-c plane, signifying that Al is well tted into SrTiO 3Àd . Moreover, Fig. 9(c) exhibits the schematic diagram of a fuel cell composed of electrolyte Al-SrTiO 3Àd and symmetrical electrodes Ni-NCAL. Furthermore, as we know, the durability of lowtemperature SOFCs is crucially tricky in practical applications. However, the stability of the semiconductor fuel cells is crucial to promoting fuel cell technology. Therefore, to assess the stability of the semiconductor materials, Ni-NCAL/Al-SrTiO 3Àd / NCAL-Ni electrolyte voltage with constant current density was recorded as a function of time, and SOFC devices were set to operate at a low temperature of 520 C in different gas H 2 /air environments. The accomplished results of our intended fuel cell device stay stable for more than 28 h under 100 mA cm 2À at 520 C. Our proposed device has delivered an appropriate power density and stable voltage (0.84 V) at a low temperature of 520 C, as shown in Fig. 9(d). Still, we failed to obtain long-term stability, as such long-term stability (>100 h) needs better technical engineering and technology in terms of compatible electrodes, which can be pursued in the future.

Conclusion
A novel semiconductor membrane Al-SrTiO 3Àd was prepared as an electrolyte in layered (Ni-NCAL)/Al-SrTiO 3Àd (STO)/NCAL-Ni fuel cell at low-temperature operation (420-520 C). Al-SrTiO 3Àd powder was synthesized by the solid-state method and a symmetrical fuel cell was prepared by the co-pressing method using Al-SrTiO 3Àd as an electrolyte layer. The prepared device achieved a maximum power density of 0.692 W having high ionic conductivity of 0.153 S cm À1 at 520 C in H 2 /air. It was found that aer introducing H 2 into the cell, the Al-SrTiO 3Àd electrolyte experienced a signicant transition in the ionic conductivity, transforming from an insulator to a high ionic conductor. The experimental results of SDC/Al-SrTiO 3Àd and BZY/Al-SrTiO 3Àd double-layer electrolyte cells showed that the

Sustainable Energy Fuels
This journal is © The Royal Society of Chemistry 2022 Al-SrTiO 3Àd electrolyte is a mixed conductor of protons and oxygen ions. XRD, XPS, HRTEM, and UV-visible studies showed that a lithium carbonate layer was formed on the Al-STO surface aer the performance testing, which constitutes a composite electrolyte with a core-shell structure. The composite electrolyte Al-SrTiO 3Àd -Li 2 CO 3 has a high conductivity for oxygen, proton, and carbonate ions. The carbonate originates in NCAL from the atmospheric CO 2 and is transferred from the NCAL electrode to the electrolyte. The formation of the core-shell heterojunction inhibits the electronic conduction and seems to level up the charge carrier transport through the band alignment mechanism (BIEF) at the surface and interface. Profound studies on the core-shell junction would be useful for future studies on the subject.

Conflicts of interest
We do not have any competing nancial interests or personal relationships that may inuence the work reported here.