PDF Publication Title:
Text from PDF Page: 004
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-20287-w process. The GaAs photoelectrode was sealed onto a custom-made graphite plate, which has an open window of 9 mm × 9 mm at the center (see photograph in Supplementary Fig. 3b, and fabrication details in the Methods section) and assembled into the SFB device to expose the carrier transfer surface of GaAs. Such back-illumination design is uncommon among previously reported III-V semi- conductor-based PEC devices40–44 and provides more freedom to utilize thick and/or opaque protection layers without affecting light- harvesting capability of the photoelectrode. The stability of the SFB photoelectrode in direct contact with aqueous electrolyte is also critical. Although the chemical and PEC stability of GaAs substrates is better in comparison with that of Ge used in the previous III–V tandem solar cells15, it is still far from desired for practical long term operation. The stability of GaAs in aqueous electrolytes with different pH also varies significantly with the poorest stability in solutions with extreme pH values (higher than pH 12 or lower than pH 3) and the best stability in neutral solutions45. Previous reports have shown that the introducing of a TiO2 surface protection layer on III–V semiconductors can effectively protect the photoelectrodes from photocorrosion32,40,42,46. Therefore, we further deposited a TiO2 thin film by atomic layer deposition (ALD) on the electrolyte contacting surface side of GaAs cells (Fig. 1a) to serve as the protection layer and enable stable long term operation. A thin layer of Ti (5 nm) was also deposited before the TiO2 layer of 80 nm to promote adhesion and another thin layer of Pt (10 nm) was deposited on TiO2 to enhance charge extraction at the photoelectrode/electrolyte interface. Potential match modeling of SFB and performance estimate. In order to achieve an efficient and stable SFB, we need to match the E0cell of the SFB with the photovoltage of the photoelectrodes14 and satisfy the essential requirements for robust and noncorrosive electrolytes. To find the best-matched E0cell for the SJ-GaAs solar cell, we used a numerical modeling method to simulate the rela- tionship between SOEE and E0cell using the J–V curve of the solid- state GaAs solar cell (shown in Fig. 1b, see “Methods” for simula- tion details). As a result, the SOEE-E0cell simulation predicts an optimal E0cell of 0.74 V with a maximum SOEE (SOEEmax) of 16.5% (Fig. 1d). To find stable redox couples with good potential matches with the SJ-GaAs cell, we turn to the neutral pH BTMAP-Vi and BTMAP-Fc redox couples. Enabled by the strong electrostatic repulsion from the positively charged BTMAP side chains and the relatively large molecular size of the two BTMAP redox couples, the RFB built with these redox couples is one of the most stable neutral pH aqueous organic RFBs reported so far47. More importantly, the E0cell matches the optimized voltage predicted above well for the SJ- GaAs cell. The electrochemical properties of the BTMAP-Vi and BTMAP-Fc redox couples were characterized by three-electrode Excellent stability was demonstrated for a neutral pH RFB built with 0.20M BTMAP-Vi/BTMAP-Fc redox couples and cycled galvanostatically at different current densities of 5–100 mA cm−2 for 5 cycles each (Supplementary Fig. 4). Figure 2b displays the Coulombic efficiency (CE, green dots), voltage efficiency (VE, red dots) and energy efficiency (EE, blue circles) according to the RFB galvanostatic cycling results. The equations for calculating the CE, VE, and EE are in the Methods. The RFB maintained a nearly constant CE > 99% over 29 h. The cell potential-capacity profile during the galvanostatic cycling test of the RFB with 0.20M BTMAP-Vi/BTMAP-Fc redox couples revealed an average galvanostatic–potentiostatic charge/discharge capacity of 2.41 Ah L−1 (energy density of 2.05 Wh L−1) that could be obtained even at a high current density of 50 mA cm−2 (Supplementary Fig. 5). This is very close to the theoretical capacity of 2.68 Ah L−1. Due to the excellent cycling stability and good voltage match for the SJ- GaAs photoelectrode, we built up an integrated SJ-GaAs SFB device with these neutral pH BTMAP-Vi/BTMAP-Fc redox cou- ples and evaluated its overall charge/discharge performances. Optimization of SFB operating conditions. Because SJ GaAs solar cells yield a high photocurrent density of about 30 mA cm−2, we first need to optimize the electrolyte concentration and flow rate in the integrated SFB device to prevent the accumulation of the photoexcited carriers at the photoelectrode/electrolyte inter- face. We carried out linear sweep voltammetry (LSV) measure- ments of the integrated SFB device with different redox couple concentrations of 0.1 and 0.2 M, and various flow rates from 20 to 120 mL min−1 (measured under solar cell mode, as shown in Supplementary Fig. 6). We observed improvements of the Jsc, Voc, and FF with the higher redox couple concentration and flow rates, which indicate more facile electrode kinetics and charge carrier transport48. These measurements showed that a concentration of 0.1 M results in much lower Jsc but a concentration of 0.2 M and an optimized flow rate of 60 mL min−1 can support sufficient electrochemical mass transport between the SJ-GaAs photoanode and the inert carbon felt. In preliminary SFB cycling tests using the electrolyte flow rates of 40, 60, and 80 mL min−1 (10 cycles for each condition as displayed in Supplementary Fig. 7), the overall SFB performance shows a slightly improved with the flow rate higher than 60 mL min−1, and the highest initially SOEE of 15.1% can be achieved with the flow rate of 80 mL min−1. However, it is at the cost of extremely fast performance decay in only three cycles (Supplementary Fig. 7c, d) due to serious surface destruction. Post-mortem optical imaging (Supplementary Fig. 3c–e) and X- ray photoelectron spectroscopy surface analysis (Supplementary Fig. 8) were further performed to understand the failure mechanism of the surface protection layer of Ti/TiO2/Pt. They suggested that the TiO2 protection layer (together with the Pt layer on top) disappeared after cycling for the case of the 80 mL min−1 flow rate, but remained mostly intact for the cases of lower flow rates. The disappeared Pt XPS signal and the newly emerged Ga and As signals after cycling at the 80 mL min−1 flow rate indicated that the Ti/TiO2/Pt protection layer was peeled off during the SFB cycling test to expose the vulnerable GaAs substrate to the elec- trolyte. Therefore, the observed SFB device performance decay was likely caused by such mechanical damage under the high flow rate. Therefore, in the remaining SFB measurements of this work, the electrolyte flow rate was set to 60 mL min−1 to balance the effi- ciency and stability concerns. Figure 3a shows the LSV behavior of the integrated SFB measured at 50% SOC under solar cell mode with the optimized electrolyte concentration (0.2 M) and flow rate (60 mL min−1) and one Sun of AM 1.5 G illumination. The SJ-GaAs photoanode in the integrated SFB device exhibited a Jsc of 28.6 mA cm−2 (the area of cyclic voltammetry (CV) measurement. As shown in Fig. 2a, the formal potential of BTMAP-Vi E0 and BTMAP-Fc E0 are Vi Fc −0.353 and 0.382 V vs. SHE, respectively. Note that the SJ-GaAs cell acts as a photoanode in this work, while a graphite plate is at the cathode side for charging BTMAP-Vi. Although the formal potential of BTMAP-Vi is −0.353 V vs. SHE, the overpotential of the graphite plate is too high for the hydrogen generation reaction to occur. Hence, an E0cell of 0.735 V for the SFB could be estimated. The charge/discharge reactions of these two redox couples are described below: Vi4þ þ e ! Vi3þ ðchargeÞand Vi3þ e ! Vi4þ ðdischargeÞ; E0Vi 1⁄4 0:353 V : ð2Þ FcII e !FcIIIðchargeÞandFcIII þe !FcIIðdischargeÞ;E0Fc 1⁄40:382V: ð3Þ 4 NATURE COMMUNICATIONS | (2021)12:156 | https://doi.org/10.1038/s41467-020-20287-w | www.nature.com/naturecommunicationsPDF Image | flow battery enabled single-junction GaAs photoelectrode
PDF Search Title:
flow battery enabled single-junction GaAs photoelectrodeOriginal File Name Searched:
s41467-020-20287-w.pdfDIY PDF Search: Google It | Yahoo | Bing
Salgenx Redox Flow Battery Technology: Salt water flow battery technology with low cost and great energy density that can be used for power storage and thermal storage. Let us de-risk your production using our license. Our aqueous flow battery is less cost than Tesla Megapack and available faster. Redox flow battery. No membrane needed like with Vanadium, or Bromine. Salgenx flow battery
CONTACT TEL: 608-238-6001 Email: greg@salgenx.com (Standard Web Page)