Photophysical Study on the Effect of the External Potential on NiO-Based Photocathodes

In the present study, we investigate the effects of the applied external potential on a dye-sensitized NiO photocathode by time-resolved photoluminescence and femtosecond transient absorption spectroscopy under operating conditions. Instead of the anticipated acceleration of photoinduced hole injection from dye into NiO at a more negative applied potential, we observe that both hole injection and charge recombination are slowed down. We cautiously assign this effect to a variation in OH– ion concentration in the inner Helmholtz plane of the electrochemical double layer with applied potential, warranting further investigation for the realization of efficient solar fuel devices.

washed by Milli-Q water.Finally, the NiO films were obtained by annealing in air at 450 °C for 1 hour.

Characterization
All the in-situ experiments were carried out in a three-electrode quartz cell (10x10 mm, Hellma, 101 -Macro cells, Figure S1) with the NiO/P1 film as the working electrode, a Ag/AgCl reference electrode and a Pt wire as the counter electrode.The electrolyte used was a 0.1 M phosphate buffer solution (PBS) with a pH value around 7. Before each experiment, the electrolyte was degassed by N 2 for more than 30 min.The UV-Vis absorbance spectra of the films were recorded using a ThermoSci EVO600 spectrometer.The valence states of the NiO under different applied potentials were analyzed immediately after applying the potential for 5 min.by X-ray Photoelectron Spectroscopy (PHI Quantera SXM).The 5 min.treatment should be sufficient to achieve a homogeneous oxidation profile across the entire NiO layer.Electrochemically oxidized Ni 3+ can be stored even without electrolyte due to its capacitive properties 1 and should therefore be detectable by ex-situ XPS.The nanomorphology of the layers was studied by a Zeiss MERLIN HR-SEM.

Time-resolved photoluminescence spectroscopy
The setup used for time-resolved photoluminescence experiments was described in detail in our previous work. 2Briefly, a Fianium laser (FP-532-1-s, center wavelength 532 nm, pulse duration of 300 fs, 80.37 MHz repetition rate) was used as the light source.For experiments with UV excitation (267 nm), the Fianium output was focused into a 3 mm β-BaB 2 O 4 crystal (Newlight Photonics).The residual 532 nm output was removed by using three dichroic mirrors (Thorlabs, MBI-K04) and a FGUV11-UV filter (Thorlabs).The λ exc.= 267 nm and λ exc.= 532 nm experiments were performed using a power of 27 μW and 8.2 μW, respectively.The sample was kept in a quartz cuvette (Hellma, 10 mm optical path length) as the working electrode with a 0.1 M phosphate buffer solution (PBS, pH=7) as the electrolyte, a Pt counter electrode and an Ag/AgCl reference electrode.The applied potential was controlled by a Emstat3 potentiostat (PalmSens).The spectral calibration was checked and adapted if necessary using a Hg/Ar calibration lamp (Oriel, LSP035).The spectral sensitivity of the photoluminescence spectra was corrected by the equations below, which were determined by measuring the spectrum of a black body lamp (Ocean Optics, HL-2000) with its calibrated spectrum:

Femtosecond transient absorption spectroscopy (fs TA)
The setup used for femtosecond transient absorption experiments was described in detail in our previous work. 2,3To avoid a potential (verified to be minor) role of sample variation, all comparative experiments were performed on the same NiO/P1 sample.

Supplementary results
The UV-Vis absorbance spectra of P1 in ethanol, deposited onto NiO or ZrO 2 are shown in Fig. S2.Due to electronic coupling between dye and semiconductor, the UV-vis spectra of the dye on the metal oxide show a broadening and red-shift in absorbance compared to the dye in solution, similar as in the literature 4 .A difference in electronic coupling between the P1 dye and ZrO 2 and NiO can also explain the small difference between the spectra of P1 on ZrO 2 and NiO.        Figure S8 shows the photoluminescence decays of NiO films in PBS following excitation at 267 nm at various applied potentials.The spectra at +0.8 V and +0.4 V show a higher photoluminescence intensity, which can be explained by less band bending (see schematic diagram in Figure S9).According to the dead layer model, the photoluminescence intensity relates to band bending. 8,9With more band bending, the dead layer is thicker, which results in a lower photoluminescence intensity originating from the part of the layer without band bending.NiO is a material with an indirect band gap and a defect-rich surface, resulting in a very weak photoluminescence signal.In PBS, the additional band bending caused by the NiO/electrolyte interface leads to quenching of most of the PL signal, both at 0 V and at negative potentials.S1.

Figure S1 .
Figure S1.The three electrode cell used in this work.

Figure
Figure S2.UV-Vis absorbance spectra of the P1 dye in ethanol, on NiO and on ZrO 2 , corrected for the signal of the substrate, if used.

FigureFigure
Figure S3.Surface (a) and cross-sectional (b) SEM images of NiO on FTO.

Figure
FigureS3shows the surface and cross-sectional SEM images of NiO on FTO, and FigureS4the XRD

Figure S6 .
Figure S6.Ni 2p XPS spectra of the NiO film immediately after applying the indicated external potential

Figure
Figure S6 shows the Ni 2p XPS spectra of the NiO film immediately after applying different external

Figure
FigureS10shows the PL decay profile of ZrO 2 /P1 in PBS electrolyte under various external potentials following excitation at 532 nm.Recent work by Tian et al.10 on ZrO 2 sensitized with the PB6 dye shows that potential induced changes in UV-VIS spectra are predominantly due to the dye.It is obvious that

Figure S11 .
Figure S11.Transient absorption kinetic traces at 570 nm after excitation at 500 nm of same NiO/P1

Figure S12 .
Figure S12.Photophysical model used to describe the femtosecond transient absorption data, IRT =

Figure S13 .
Figure S13.Species associated spectra obtained from target analysis of the TA data of NiO/P1 in PBS

Table S1 .
Time constants from target analysis of NiO/P1 in PBS electrolyte (pH=7) under various external potentials vs. Ag/AgCl.IRT = instrumental response time (100-150 fs).The transient absorption data and role of the applied bias potential are well described by the photophysical model shown in Figure S12, as apparent from the fits included as solid lines in Figures 3