Supporting Information Surface modification of the CoOx loaded ...
Supporting Information Surface modification of the CoOx loaded BiVO4 photoanodes with ultrathin p-type NiO layers for the improved solar water oxidation Miao Zhong†‡, Takashi Hisatomi†‡, Yongbo Kuang†‡, Jiao Zhao†‡, Min Liu†‡, Akihide Iwase§, Qingxin Jia‡§, Hiroshi Nishiyama†‡, Tsutomu Minegishi†‡, Mamiko Nakabayashi¶, Naoya Shibata¶, Ryo Niishiro‡#, Chisato Katayama‡⊥, Hidetaka Shibano‡, Masao Katayama†‡, Akihiko Kudo‡§, Taro Yamada†‡ and Kazunari Domen*†‡ †
Department of Chemical System Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo
113-8656, Japan ‡
Japan Technological Research Association of Artificial Photosynthetic Chemical Process (ARPChem),
5-1-5 Kashiwanoha, Kashiwa-shi, 277-8589 Chiba, Japan §
Department of Applied Chemistry, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku-ku, Tokyo
162-8601, Japan ¶
Institute of Engineering Innovation, The University of Tokyo, 2-11-16, Yayoi, Bunkyo-ku, Tokyo 113-8656,
Japan. #
Mitsui Chemicals, Inc., 580-32 Nagaura, Sodegaura, 299-0265 Chiba, Japan
⊥
Fujifilm Corporation, 577, Ushijima, Kaisei-Machi, Ashigarakami-gun, 258-8577 Kanagawa, Japan
S1
Fabrication of the BiVO4 photoanodes 1.
Synthesis of the BiVO4 particles The BiVO4 particles were synthesized by the microwave-assisted liquid-solid state reactions as discussed in details in the previous work1. First, the K3V5O14 precursor was synthesized by the solid-state reaction at 723 K for 5 h using K2CO3 (Kanto Chemical, 99.5%) and V2O5 (Wako Pure Chemical Industries 99.9%). Then, the precursors of Bi(NO3)3·5H2O (Kanto Chemical; 99.9%) and K3V5O14 were stirred in pure water under the microwave (EYELA, MWO-1000S) irradiation (100 W, 2450 MHz) for 4 h. After cooling down to the room temperature, the yellow powder was obtained.
2.
Loading of the CoOx catalysts on the BiVO4 particles For the CoOx loading, BiVO4 particles were impregnated in an aqueous solution containing a calculated amount of cobalt nitrate2. The solution was then dried on a heated water bath. The as-impregnated BiVO4 particles was put in an alumina tube and heated in air at 573 K for 1 h.
3.
Fabrication of the CoOx/BiVO4 by the particle transfer method The CoOx/BiVO4 photoanodes were prepared by the particle transfer method3. A schematic to illustrate the electrode fabrication process is shown in Fig. S1. In brief, ~ 10 mg prepared CoOx/BiVO4 particles were suspended in a 450 μl 2-propanol solution. The particle suspension solution was sonicated for 5 minutes to obtain a uniform suspension solution. Then, the uniform suspension solution was dropped casting on a 1 cm × 3 cm glass substrate for three times. In each time, 150 μl suspension solution was dropped on the glass substrate and the dropped glass substrate was fully dried in air before carrying out the next drop casting process. After three times, a thin layer of Ti (2-5 μm) was sputtered on the CoOx/BiVO4 particles to form the electrical contact layers. A different glass substrate with an adhesive carbon tape was used to attach the sputtered metal films. After peeling off the metal films with particles, the transferred electrode was sonicated for 10 s in water to remove the excessive particles on the surface. In this way, the CoOx/BiVO4/Ti electrode was prepared. The thickness of the mono-layer CoOx/BiVO4 is estimated to be about 500 nm which is already thick enough for efficient solar light absorption.
4.
Deposition of NiO on the CoOx/BiVO4 photoanode using atomic layer deposition (ALD) ALD is designed for the conformal deposition of thin films with controlled thickness. Different from the chemical vapor deposition (CVD), the precursors were kept separated throughout the ALD process using the purge and the pulse sequences. At the certain growth temperature, the growth rate is controlled. The temperature in our ALD reaction chamber was fixed at 260 °C to avoid unnecessary chemical
S2
decomposition of the Ni(thd)2 precursor and prevent the CVD-like growth. The Ni(thd)2 precursor was heated to 165 °C to ensure a sufficient vapor pressure for the ALD process. The temperature of the water precursor was set to 15 °C to maintain its vapor pressure for realizing a controllable ALD growth. In a completed ALD cycle, water was first pulsed into reaction chamber for 6 s and then purged for 8 s to get rid of the non-adsorbed water molecules. The N2 gas in a 500 sccm flow rate was pulsed into the Ni(thd)2 precursor chamber for 2.5 s to increase the volumetric vapor feed in the coming pulsing Ni(thd)2 precursor sequence. Then, the Ni(thd)2 with N2 as a carrier gas was pulsed into the reaction chamber for 3 s, followed by an 8-second purge process. By repeating the above sequences, the deposition of NiO with different thickness can be realized.
Figure S1. Schematic of the CoOx/BiVO4 electrode fabrication process by the particle-transfer method.
Electrochemical characterization of the ALD NiO/FTO and FTO substrates Before the ALD of NiO on the particle-transferred BiVO4 photoanodes, the NiO was first deposited on FTO substrates and on flat silicon wafers for investigations. The dark CV scans of the NiO-deposited FTO electrode exhibited prominent Ni2+/Ni3+ redox peaks over the FTO electrode background in Figure S2a and S2b, indicating that NiO was successfully deposited.
Estimation of the growth rate of the ALD NiO The ALD NiO growth rate is roughly estimated to be ~ 0.3 Å per cycle by the cross-sectional SEM images of
S3
the NiO layer deposited on the silicon substrates using different ALD cycles as shown in Figure S2c.
(a)
(c)
(b)
Figure S2. (a-b) CV scans for the electrodes of 200-cycle ALD NiO on FTO (a) and pure FTO (b) in 0.5 M Na2SO4 solution at pH 10. (c) The ALD NiO growth as a function of ALD cycles.
S4
Improved NiO layer conformity on BiVO4 particles using ALD Compared to the chemical impregnation of NiO on BiVO4 particles using the solution method, the ALD NiO shows largely improved conformity (Figure S3). Figure S3 shows the SEM images of the NiO deposited BiVO4 particles using the solution method and the ALD. In the solution method, a certain amount of Ni(NO3)2 (a calculated amount of 0.2, 0.5 and 1.0 % of the final product NiO with respect to the BiVO 4 in weight) were dissolved into deionized water. BiVO4 powder was immersed into the prepared Ni(NO3)2 aqueous solution with gentle stirring to load Ni(NO3)2 on BiVO4 uniformly. After drying the aqueous solution, the Ni(NO3)2-loaded BiVO4 particles were heated at 573 K in air for 2 h for the formation of NiO. The ALD deposition of NiO on BiVO4 particles were performed following the ALD process described above. From the SEM images, it is clearly observed that the ALD realizes a conformal deposition of NiO nano-particles wrapping the BiVO4 over a large area.
Figure S3. SEM images of the NiO-deposited BiVO4. (a-c) NiO loaded BiVO4 particles using the solution method. (d) NiO deposited on BiVO4 particles by ALD.
S5
Mott-Schottky analyses of the ALD NiO deposited FTO electrode The p-type character of the as-deposited NiO layer is an important character for the charge separation and it is observed by the Mott-Schottky analysis in Figure S4. The Mott-Schottky analyses of the NiO deposited FTO (200 ALD cycles) and bare FTO were measured in 0.5 M Na2SO4 solution at pH 6.6 in dark at the frequency of 10 kHz.
Figure S4. The Mott-Schottky plots of the bare FTO (blue) and 200-cycle ALD NiO on FTO (red) in 0.5 M Na2SO4 at pH 6.6 in dark with 10 kHz frequency.
S6
Lsv forward and backward scans of the NiO/CoO x/BiVO4 and CoOx/BiVO4 photoanodes under chopped AM 1.5G illumination.
Figure S5. Lsv forward and backward scans. The NiO/CoOx/BiVO4 (a) and CoOx/BiVO4 (b) photoanodes measured in 0.1 M KPi at pH 7 under chopped AM 1.5G illumination.
S7
IPCE analyses of the NiO/CoOx/BiVO4 and the CoOx/BiVO4 photoanodes The wavelength dependence of IPCE was measured at different applied potentials under monochromatic irradiation from a Xe lamp (Asahi Spectra, MAX-302) equipped with bandpass filters (central wavelengths: 400-540 nm, every 20 nm; full width at half maximum: 10 nm). The irradiance spectra of the light incident on the electrode surface were measured with a spectroradiometer (EKO Instruments, LS-100). The IPCE at each wavelength was calculated via the equation: IPCE
( J light J dark（ ) mA×cm-2） 1240 （V×nm） , -2 Pmono （mW×cm ）（ nm）
To accurately measure the IPCE data, each IPCE curve plotted in Figure S6 is calculated using the steady-state photocurrent density at a fixed potential with different incident wavelengths. The onsets in the IPCE curves for both of the CoOx/BiVO4 and NiO/CoOx/BiVO4 photoanodes were consistent with the light absorption spectra of BiVO4, indicating that the anodic photocurrent were attributed to the light absorption in BiVO4.
Figure S6. IPCE spectra at the different applied potentials in 0.1M KPi solution at pH 7. (a) The NiO/CoOx/BiVO4 (~ 6 nm NiO, ALD 200 cycles, CoOx 1wt%) photoanode. (b) The CoOx/BiVO4 (CoOx 1wt%) photoanode.
S8
The enlarged SEM images of the CoOx/BiVO4 (CoOx 1wt%) and NiO/CoOx/BiVO4 (~ 6 nm NiO by ALD 200 cycles, CoOx 1wt%) photoanodes before and after the PEC measurements.
Figure S7. (a-d) The SEM images of the CoOx/BiVO4 (CoOx 1 wt%) photoanode before (a) and after the PEC measurements (b), and, the NiO/CoOx/BiVO4 (~ 6 nm NiO by 200 ALD cycles, CoOx 1 wt%) photoanode before (c) and after the PEC measurements (d).
S9
PEC characterization of the in situ formed ultrathin CoOx layer on the CoOx/BiVO4 phtoanode.
Figure S8. PEC measurements of the in-situ formed ultrathin CoOx. The lsv scans of the CoOx/BiVO4 (CoOx 1wt%) photoanode in 0.1 M pure KPi solutions at pH 7 under chopped AM 1.5G illumination after stability measurements for different periods. Each amperometric stability measurement was performed in 0.1 M pure KPi solution (pH 7) at 1.0 VRHE for 10 min. and then followed by the lsv scans at a scan rate of 10 mV s-1. The photocurrent density decreases at the low applied potential region in the lsv scans after the PEC stability test in pure KPi solutions, indicating that the CoOx layer on the surface gradually dissolves in the KPi solution.
S10
Linear sweep voltammetry (LSV) scans of the NiO/CoOx/FTO, the NiO/FTO and the CoOx/FTO electrodes in dark electrolysis.
Figure S9. Dark LSV scans. LSV scans of the NiO/CoOx/FTO (red), the NiO/FTO (green) and the CoOx/FTO (black) electrodes in 0.1 M NaOH at pH 13 in dark.
S11
XRD analysis of the synthesized NiOOH The optical images of the NiO (Wako), the Ni(OH)2 (aldrich) and the synthesized NiOOH are shown in Figure S10. NiOOH are prepared by the reaction of Ni(OH)2 with sodium hypochlorite (NaClO). In brief, Ni(OH)2 powder was added into the concentrated NaClO aquesous solution for the oxidizing reaction with stirring. The obtained black suspension was filtered out and rinsed by NaOH and pure water. The suspension was then dried at 60 °C overnight. The XRD analyses for the Ni(OH)2 and NiOOH are shown in Figure S10. It is clearly evidenced the NiOOH (003) and NiOOH (101) peaks from the XRD θ/2θ scan of the NiOOH sample. In addition, the Ni(OH)2 diffraction peaks could also be observed in the synthesized NiOOH sample. It is therefore inferred that the final product contains the amorphous/crystalline NiOOH with unreacted Ni(OH)2.
Figure S10. Optical images of the NiO, Ni(OH)2 and NiOOH samples (above) and XRD analyses for the Ni(OH)2 and NiOOH.
S12
The PEC water oxidation and sulfite oxidation performances of the bare BiVO 4, the CoOx/BiVO4, the NiO/BiVO4, and the NiO/CoOx/BiVO4 photoanodes.
Figure S11. Chopped LSV scans representing the PEC water oxidation and sulfite oxidation performances with different photoanodes under AM 1.5G illumination. (a) The bare BiVO4 photoanode. (b) The CoOx/BiVO4 (CoOx 1wt%) photoanode. (c) The NiO/BiVO4 (~ 6 nm NiO with ALD 200 cycles) photoanode. (d) The NiO/CoOx/BiVO4 (~ 6 nm NiO with ALD 200 cycles, CoOx 1 wt%) photoanode.
S13
The enlarged SEM images of the different ALD cycles (100-300) NiO/CoOx/BiVO4 photoanodes after the PEC measurements.
Figure S12. SEM characterizations of the NiO/CoOx/BiVO4 photoanodes after the PEC measurement, ALD NiO (a) 100 cycles, (b) 200 cycle and (c) 300 cycles.
S14
Electro-impedance study of the NiO/CoOx/BiVO4 photoanodes with different ALD cycles. As shown in the electrochemical impedance spectroscopy analyses, the resistance of the NiOOH layer increases with the increase of the ALD cycles, indicating that thicker NiOOH were formed on the surfaces of the NiO/CoOx/BiVO4 photoanodes. Therefore, a portion of the applied potential will drop across this NiOOH film when the current passes. This would ultimately lead to a lower apparent catalytic activity relative to a more-conductive film. As a result, the large increase in the resistance of the NiOOH layer significantly decreases the PEC performances. The equivalent circuit with the fitted results are shown in Fig S13. We suggest that the two semi-circles obtained in the EIS analyses are attributed to the semiconductor solid/solid junction of the NiO/BiVO4 or the CoOx/BiVO4 and the interface junction of the NiO/NiOOH or the NiOOH/electrolyte. Note that the length per unit in the Zre and Zim axes in Fig. S13 (a-d) is different.
Figure S13. Electrochemical impedance spectroscopy analyses. The nyquist plots of the NiO/CoOx/BiVO4 photoanodes with different ALD cycles measured under AM 1.5G illumination in 0.1 M pH 7 KPi solution at open-circuit conditions.
S15
Band diagrams and the measured OCV values of the bare BiVO4, the CoOx/BiVO4 and the NiO/CoOx/BiVO4 photoanodes in dark and under AM 1.5G illumination.
Figure S14. The band bending diagrams with the corresponding measured OCV values of the bare BiVO4, the CoOx/BiVO4 and the ALD different-cycle NiO/CoOx/BiVO4 photoanodes in dark and under AM 1.5G illumination. The OCV measurement was performed in 0.1 M KPi solution at pH 7 with Ar bubbling under Open-circuit condition with the bare BiVO4, the CoOx/BiVO4 and the NiO/CoOx/BiVO4 photoanodes as the working electrodes and the Ag/AgCl as the reference electrode.
S16
Passivation of the BiVO4 surface states with ALD Al2O3. As shown in Figure S15, a fast OCP-decay is obtained after the ALD Al2O3 on the CoOx/BiVO4 anodes. It is evidence of the surface passivation effect for the reduced surface trapped carriers. However, it is found that the OCP values decrease in the Al2O3/CoOx/BiVO4 photoanodes compared to that in the CoOx/BiVO4 photoanodes. This is because the Al2O3 is an insulator and no p-n junction is formed. Further, the OER activity of Al2O3 is negligible compared to the NiOOH. As a result, the conformal deposition of the CoOx/BiVO4 photoanode with ion-impermeable Al2O3 decreases the PEC performances.
Figure S15. The OCP values under AM 1.5G illumination and in dark for the CoO x-BiVO4 photoanode and Al2O3/CoOx/BiVO4 photoanodes.
S17
The morphologies of the different amounts of CoOx loaded BiVO4.
Figure S16. SEM images of the BiVO4 particles with the different loaded amounts of CoOx.
S18
PEC performances of the NiO/CoOx/BiVO4 photoanodes with the different loaded amounts of CoOx but the same 200-cycle ALD NiO top layer.
Figure S17. The PEC performances of the NiO/CoOx/BiVO4 photoanode with the different amounts of CoOx and with the same 200-cycle ALD NiO.
S19
References 1.
Soma, K.; Iwase, A.; Kudo, A. Catal. Lett. 2014, 144, 1962-1967.
2.
Zhang, F.; Yamakata, A.; Maeda, K.; Moriya, Y.; Takata, T.; Kubota, J.; Teshima, K.; Oishi, S.; Domen, K. J. Am. Chem. Soc. 2012, 134, 8348−8351.
3.
Minegishi, T.; Nishimura, N.; Kubota, J.; Domen, K. Chem. Sci. 2013, 4, 1120-1124.
S20
Department of Chemical System Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo
113-8656, Japan ‡
Japan Technological Research Association of Artificial Photosynthetic Chemical Process (ARPChem),
5-1-5 Kashiwanoha, Kashiwa-shi, 277-8589 Chiba, Japan §
Department of Applied Chemistry, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku-ku, Tokyo
162-8601, Japan ¶
Institute of Engineering Innovation, The University of Tokyo, 2-11-16, Yayoi, Bunkyo-ku, Tokyo 113-8656,
Japan. #
Mitsui Chemicals, Inc., 580-32 Nagaura, Sodegaura, 299-0265 Chiba, Japan
⊥
Fujifilm Corporation, 577, Ushijima, Kaisei-Machi, Ashigarakami-gun, 258-8577 Kanagawa, Japan
S1
Fabrication of the BiVO4 photoanodes 1.
Synthesis of the BiVO4 particles The BiVO4 particles were synthesized by the microwave-assisted liquid-solid state reactions as discussed in details in the previous work1. First, the K3V5O14 precursor was synthesized by the solid-state reaction at 723 K for 5 h using K2CO3 (Kanto Chemical, 99.5%) and V2O5 (Wako Pure Chemical Industries 99.9%). Then, the precursors of Bi(NO3)3·5H2O (Kanto Chemical; 99.9%) and K3V5O14 were stirred in pure water under the microwave (EYELA, MWO-1000S) irradiation (100 W, 2450 MHz) for 4 h. After cooling down to the room temperature, the yellow powder was obtained.
2.
Loading of the CoOx catalysts on the BiVO4 particles For the CoOx loading, BiVO4 particles were impregnated in an aqueous solution containing a calculated amount of cobalt nitrate2. The solution was then dried on a heated water bath. The as-impregnated BiVO4 particles was put in an alumina tube and heated in air at 573 K for 1 h.
3.
Fabrication of the CoOx/BiVO4 by the particle transfer method The CoOx/BiVO4 photoanodes were prepared by the particle transfer method3. A schematic to illustrate the electrode fabrication process is shown in Fig. S1. In brief, ~ 10 mg prepared CoOx/BiVO4 particles were suspended in a 450 μl 2-propanol solution. The particle suspension solution was sonicated for 5 minutes to obtain a uniform suspension solution. Then, the uniform suspension solution was dropped casting on a 1 cm × 3 cm glass substrate for three times. In each time, 150 μl suspension solution was dropped on the glass substrate and the dropped glass substrate was fully dried in air before carrying out the next drop casting process. After three times, a thin layer of Ti (2-5 μm) was sputtered on the CoOx/BiVO4 particles to form the electrical contact layers. A different glass substrate with an adhesive carbon tape was used to attach the sputtered metal films. After peeling off the metal films with particles, the transferred electrode was sonicated for 10 s in water to remove the excessive particles on the surface. In this way, the CoOx/BiVO4/Ti electrode was prepared. The thickness of the mono-layer CoOx/BiVO4 is estimated to be about 500 nm which is already thick enough for efficient solar light absorption.
4.
Deposition of NiO on the CoOx/BiVO4 photoanode using atomic layer deposition (ALD) ALD is designed for the conformal deposition of thin films with controlled thickness. Different from the chemical vapor deposition (CVD), the precursors were kept separated throughout the ALD process using the purge and the pulse sequences. At the certain growth temperature, the growth rate is controlled. The temperature in our ALD reaction chamber was fixed at 260 °C to avoid unnecessary chemical
S2
decomposition of the Ni(thd)2 precursor and prevent the CVD-like growth. The Ni(thd)2 precursor was heated to 165 °C to ensure a sufficient vapor pressure for the ALD process. The temperature of the water precursor was set to 15 °C to maintain its vapor pressure for realizing a controllable ALD growth. In a completed ALD cycle, water was first pulsed into reaction chamber for 6 s and then purged for 8 s to get rid of the non-adsorbed water molecules. The N2 gas in a 500 sccm flow rate was pulsed into the Ni(thd)2 precursor chamber for 2.5 s to increase the volumetric vapor feed in the coming pulsing Ni(thd)2 precursor sequence. Then, the Ni(thd)2 with N2 as a carrier gas was pulsed into the reaction chamber for 3 s, followed by an 8-second purge process. By repeating the above sequences, the deposition of NiO with different thickness can be realized.
Figure S1. Schematic of the CoOx/BiVO4 electrode fabrication process by the particle-transfer method.
Electrochemical characterization of the ALD NiO/FTO and FTO substrates Before the ALD of NiO on the particle-transferred BiVO4 photoanodes, the NiO was first deposited on FTO substrates and on flat silicon wafers for investigations. The dark CV scans of the NiO-deposited FTO electrode exhibited prominent Ni2+/Ni3+ redox peaks over the FTO electrode background in Figure S2a and S2b, indicating that NiO was successfully deposited.
Estimation of the growth rate of the ALD NiO The ALD NiO growth rate is roughly estimated to be ~ 0.3 Å per cycle by the cross-sectional SEM images of
S3
the NiO layer deposited on the silicon substrates using different ALD cycles as shown in Figure S2c.
(a)
(c)
(b)
Figure S2. (a-b) CV scans for the electrodes of 200-cycle ALD NiO on FTO (a) and pure FTO (b) in 0.5 M Na2SO4 solution at pH 10. (c) The ALD NiO growth as a function of ALD cycles.
S4
Improved NiO layer conformity on BiVO4 particles using ALD Compared to the chemical impregnation of NiO on BiVO4 particles using the solution method, the ALD NiO shows largely improved conformity (Figure S3). Figure S3 shows the SEM images of the NiO deposited BiVO4 particles using the solution method and the ALD. In the solution method, a certain amount of Ni(NO3)2 (a calculated amount of 0.2, 0.5 and 1.0 % of the final product NiO with respect to the BiVO 4 in weight) were dissolved into deionized water. BiVO4 powder was immersed into the prepared Ni(NO3)2 aqueous solution with gentle stirring to load Ni(NO3)2 on BiVO4 uniformly. After drying the aqueous solution, the Ni(NO3)2-loaded BiVO4 particles were heated at 573 K in air for 2 h for the formation of NiO. The ALD deposition of NiO on BiVO4 particles were performed following the ALD process described above. From the SEM images, it is clearly observed that the ALD realizes a conformal deposition of NiO nano-particles wrapping the BiVO4 over a large area.
Figure S3. SEM images of the NiO-deposited BiVO4. (a-c) NiO loaded BiVO4 particles using the solution method. (d) NiO deposited on BiVO4 particles by ALD.
S5
Mott-Schottky analyses of the ALD NiO deposited FTO electrode The p-type character of the as-deposited NiO layer is an important character for the charge separation and it is observed by the Mott-Schottky analysis in Figure S4. The Mott-Schottky analyses of the NiO deposited FTO (200 ALD cycles) and bare FTO were measured in 0.5 M Na2SO4 solution at pH 6.6 in dark at the frequency of 10 kHz.
Figure S4. The Mott-Schottky plots of the bare FTO (blue) and 200-cycle ALD NiO on FTO (red) in 0.5 M Na2SO4 at pH 6.6 in dark with 10 kHz frequency.
S6
Lsv forward and backward scans of the NiO/CoO x/BiVO4 and CoOx/BiVO4 photoanodes under chopped AM 1.5G illumination.
Figure S5. Lsv forward and backward scans. The NiO/CoOx/BiVO4 (a) and CoOx/BiVO4 (b) photoanodes measured in 0.1 M KPi at pH 7 under chopped AM 1.5G illumination.
S7
IPCE analyses of the NiO/CoOx/BiVO4 and the CoOx/BiVO4 photoanodes The wavelength dependence of IPCE was measured at different applied potentials under monochromatic irradiation from a Xe lamp (Asahi Spectra, MAX-302) equipped with bandpass filters (central wavelengths: 400-540 nm, every 20 nm; full width at half maximum: 10 nm). The irradiance spectra of the light incident on the electrode surface were measured with a spectroradiometer (EKO Instruments, LS-100). The IPCE at each wavelength was calculated via the equation: IPCE
( J light J dark（ ) mA×cm-2） 1240 （V×nm） , -2 Pmono （mW×cm ）（ nm）
To accurately measure the IPCE data, each IPCE curve plotted in Figure S6 is calculated using the steady-state photocurrent density at a fixed potential with different incident wavelengths. The onsets in the IPCE curves for both of the CoOx/BiVO4 and NiO/CoOx/BiVO4 photoanodes were consistent with the light absorption spectra of BiVO4, indicating that the anodic photocurrent were attributed to the light absorption in BiVO4.
Figure S6. IPCE spectra at the different applied potentials in 0.1M KPi solution at pH 7. (a) The NiO/CoOx/BiVO4 (~ 6 nm NiO, ALD 200 cycles, CoOx 1wt%) photoanode. (b) The CoOx/BiVO4 (CoOx 1wt%) photoanode.
S8
The enlarged SEM images of the CoOx/BiVO4 (CoOx 1wt%) and NiO/CoOx/BiVO4 (~ 6 nm NiO by ALD 200 cycles, CoOx 1wt%) photoanodes before and after the PEC measurements.
Figure S7. (a-d) The SEM images of the CoOx/BiVO4 (CoOx 1 wt%) photoanode before (a) and after the PEC measurements (b), and, the NiO/CoOx/BiVO4 (~ 6 nm NiO by 200 ALD cycles, CoOx 1 wt%) photoanode before (c) and after the PEC measurements (d).
S9
PEC characterization of the in situ formed ultrathin CoOx layer on the CoOx/BiVO4 phtoanode.
Figure S8. PEC measurements of the in-situ formed ultrathin CoOx. The lsv scans of the CoOx/BiVO4 (CoOx 1wt%) photoanode in 0.1 M pure KPi solutions at pH 7 under chopped AM 1.5G illumination after stability measurements for different periods. Each amperometric stability measurement was performed in 0.1 M pure KPi solution (pH 7) at 1.0 VRHE for 10 min. and then followed by the lsv scans at a scan rate of 10 mV s-1. The photocurrent density decreases at the low applied potential region in the lsv scans after the PEC stability test in pure KPi solutions, indicating that the CoOx layer on the surface gradually dissolves in the KPi solution.
S10
Linear sweep voltammetry (LSV) scans of the NiO/CoOx/FTO, the NiO/FTO and the CoOx/FTO electrodes in dark electrolysis.
Figure S9. Dark LSV scans. LSV scans of the NiO/CoOx/FTO (red), the NiO/FTO (green) and the CoOx/FTO (black) electrodes in 0.1 M NaOH at pH 13 in dark.
S11
XRD analysis of the synthesized NiOOH The optical images of the NiO (Wako), the Ni(OH)2 (aldrich) and the synthesized NiOOH are shown in Figure S10. NiOOH are prepared by the reaction of Ni(OH)2 with sodium hypochlorite (NaClO). In brief, Ni(OH)2 powder was added into the concentrated NaClO aquesous solution for the oxidizing reaction with stirring. The obtained black suspension was filtered out and rinsed by NaOH and pure water. The suspension was then dried at 60 °C overnight. The XRD analyses for the Ni(OH)2 and NiOOH are shown in Figure S10. It is clearly evidenced the NiOOH (003) and NiOOH (101) peaks from the XRD θ/2θ scan of the NiOOH sample. In addition, the Ni(OH)2 diffraction peaks could also be observed in the synthesized NiOOH sample. It is therefore inferred that the final product contains the amorphous/crystalline NiOOH with unreacted Ni(OH)2.
Figure S10. Optical images of the NiO, Ni(OH)2 and NiOOH samples (above) and XRD analyses for the Ni(OH)2 and NiOOH.
S12
The PEC water oxidation and sulfite oxidation performances of the bare BiVO 4, the CoOx/BiVO4, the NiO/BiVO4, and the NiO/CoOx/BiVO4 photoanodes.
Figure S11. Chopped LSV scans representing the PEC water oxidation and sulfite oxidation performances with different photoanodes under AM 1.5G illumination. (a) The bare BiVO4 photoanode. (b) The CoOx/BiVO4 (CoOx 1wt%) photoanode. (c) The NiO/BiVO4 (~ 6 nm NiO with ALD 200 cycles) photoanode. (d) The NiO/CoOx/BiVO4 (~ 6 nm NiO with ALD 200 cycles, CoOx 1 wt%) photoanode.
S13
The enlarged SEM images of the different ALD cycles (100-300) NiO/CoOx/BiVO4 photoanodes after the PEC measurements.
Figure S12. SEM characterizations of the NiO/CoOx/BiVO4 photoanodes after the PEC measurement, ALD NiO (a) 100 cycles, (b) 200 cycle and (c) 300 cycles.
S14
Electro-impedance study of the NiO/CoOx/BiVO4 photoanodes with different ALD cycles. As shown in the electrochemical impedance spectroscopy analyses, the resistance of the NiOOH layer increases with the increase of the ALD cycles, indicating that thicker NiOOH were formed on the surfaces of the NiO/CoOx/BiVO4 photoanodes. Therefore, a portion of the applied potential will drop across this NiOOH film when the current passes. This would ultimately lead to a lower apparent catalytic activity relative to a more-conductive film. As a result, the large increase in the resistance of the NiOOH layer significantly decreases the PEC performances. The equivalent circuit with the fitted results are shown in Fig S13. We suggest that the two semi-circles obtained in the EIS analyses are attributed to the semiconductor solid/solid junction of the NiO/BiVO4 or the CoOx/BiVO4 and the interface junction of the NiO/NiOOH or the NiOOH/electrolyte. Note that the length per unit in the Zre and Zim axes in Fig. S13 (a-d) is different.
Figure S13. Electrochemical impedance spectroscopy analyses. The nyquist plots of the NiO/CoOx/BiVO4 photoanodes with different ALD cycles measured under AM 1.5G illumination in 0.1 M pH 7 KPi solution at open-circuit conditions.
S15
Band diagrams and the measured OCV values of the bare BiVO4, the CoOx/BiVO4 and the NiO/CoOx/BiVO4 photoanodes in dark and under AM 1.5G illumination.
Figure S14. The band bending diagrams with the corresponding measured OCV values of the bare BiVO4, the CoOx/BiVO4 and the ALD different-cycle NiO/CoOx/BiVO4 photoanodes in dark and under AM 1.5G illumination. The OCV measurement was performed in 0.1 M KPi solution at pH 7 with Ar bubbling under Open-circuit condition with the bare BiVO4, the CoOx/BiVO4 and the NiO/CoOx/BiVO4 photoanodes as the working electrodes and the Ag/AgCl as the reference electrode.
S16
Passivation of the BiVO4 surface states with ALD Al2O3. As shown in Figure S15, a fast OCP-decay is obtained after the ALD Al2O3 on the CoOx/BiVO4 anodes. It is evidence of the surface passivation effect for the reduced surface trapped carriers. However, it is found that the OCP values decrease in the Al2O3/CoOx/BiVO4 photoanodes compared to that in the CoOx/BiVO4 photoanodes. This is because the Al2O3 is an insulator and no p-n junction is formed. Further, the OER activity of Al2O3 is negligible compared to the NiOOH. As a result, the conformal deposition of the CoOx/BiVO4 photoanode with ion-impermeable Al2O3 decreases the PEC performances.
Figure S15. The OCP values under AM 1.5G illumination and in dark for the CoO x-BiVO4 photoanode and Al2O3/CoOx/BiVO4 photoanodes.
S17
The morphologies of the different amounts of CoOx loaded BiVO4.
Figure S16. SEM images of the BiVO4 particles with the different loaded amounts of CoOx.
S18
PEC performances of the NiO/CoOx/BiVO4 photoanodes with the different loaded amounts of CoOx but the same 200-cycle ALD NiO top layer.
Figure S17. The PEC performances of the NiO/CoOx/BiVO4 photoanode with the different amounts of CoOx and with the same 200-cycle ALD NiO.
S19
References 1.
Soma, K.; Iwase, A.; Kudo, A. Catal. Lett. 2014, 144, 1962-1967.
2.
Zhang, F.; Yamakata, A.; Maeda, K.; Moriya, Y.; Takata, T.; Kubota, J.; Teshima, K.; Oishi, S.; Domen, K. J. Am. Chem. Soc. 2012, 134, 8348−8351.
3.
Minegishi, T.; Nishimura, N.; Kubota, J.; Domen, K. Chem. Sci. 2013, 4, 1120-1124.
S20
