Wittichenite semiconductor of Cu3BiS3 films for efficient hydrogen evolution from solar driven photoelectrochemical water splitting

A highly efficient, low-cost and environmentally friendly photocathode with long-term stability is the goal of practical solar hydrogen evolution applications. Here, we found that the Cu3BiS3 film-based photocathode meets the abovementioned requirements. The Cu3BiS3-based photocathode presents a remarkable onset potential over 0.9 VRHE with excellent photoelectrochemical current densities (~7 mA/cm2 under 0 VRHE) and appreciable 10-hour long-term stability in neutral water solutions. This high onset potential of the Cu3BiS3-based photocathode directly results in a good unbiased operating photocurrent of ~1.6 mA/cm2 assisted by the BiVO4 photoanode. A tandem device of Cu3BiS3-BiVO4 with an unbiased solar-to-hydrogen conversion efficiency of 2.04% is presented. This tandem device also presents high stability over 20 hours. Ultimately, a 5 × 5 cm2 large Cu3BiS3-BiVO4 tandem device module is fabricated for standalone overall solar water splitting with a long-term stability of 60 hours.

The authors have carried out an impressive, intensive study on optimised Cu3BiS3 photocathodes (Voc = 0.9 V vs RHE and J0 = 7mA/cm^2) and their combination with BiVO4 to demonstrate unbiased solar-water splitting with a 2% solar-to-hydrogen efficiency. This device was also fabricated on a 5x5cm scale and photocurrent losses of 30% were recorded over long-term studies of 60h. These is a very interesting result, however, as the authors stated correctly as well, Cu3BiS2 photocathodes are not a new invention e.g., J. Yin and J. Jia (CrystEngComm, 2014(CrystEngComm, , 16, 2795 used TiO2/Cu3BiS3 photocathodes already in 2014 and investigated their photoelectrochemical properties. They photoelectrochemical characteristics were not as good as the ones discussed here (Voc = 0.48V and J0 = 6mA), but still close. It is therefore not believed that this manuscript would be of interest to readers of Nat. Commun. There are, moreover, a few aspects in the manuscript which are confusing: (1) The results part begins with a figures from the SI -they should start with discussing figures in the main article.
(2) The performance improvement of the Pt-TiO2/CdS/Cu3BiS3 photocathode with respect to the original Pt-Cu3BiS3 photocathode seems largely due to the introduction of the protective TiO2 layer as the authors also state. This suggests that the TiO2 layer not only has a stabilising effect, but it could potentially also function as an anti-reflection layer and therefore improves the light absorption properties of the device -which is also not novel (see Cheng et al. ACS Energy Lett. 2018, 3, 8, 1795-1800. It is suggested that the authors undertake further absorption/reflection studies of their Pt-TiO2/CdS/Cu3BiS3 device with and without TiO2 layer. (3) Experimental details are missing, e.g., the acid with which the pH of the buffer solution was adjusted, the photoelectrochemical set-up (light intensity, cell windows etc.), if a two-compartment cell was used for the long-term measurements etc.

Reviewer #2 (Remarks to the Author):
This work reports a Cu3BiS3-based photocathode and Cu3BiS3-BiVO4 tandem cell for unassisted solar water splitting with remarkable solar-to-hydrogen efficiency (2.04%) and stability (20 h). This manuscript has novelty in that it has created a tandem device with a new photocathode material Cu3BiS3. The authors achieved a photocurrent density (7 mA/cm2 at 0 VRHE) 70 times higher than the previously reported peak value (0.1 mA/cm2 at 0 VRHE) for the Cu3BiS3 photocathode by applying CdS/TiO2/Pt. Considering their originality, this reviewer recommends accepting this paper in this journal after only after properly addressing the following concerns by revision, especially Comments 1-2..
1. There appears to be an error in the Mott-Schottky curve for Cu3BiS3 in Figure 2b. Since Cu3BiS3 is a p-type material, the Mott-Schottky plot should have a negative slope and the flatband potential should be located near the onset potential. However, all curves in Figure 2b have a positive slope with a very negative flat band potential. Why? 2. The authors mentioned there is no obvious secondary phase or impurity phase in Cu3BiS3 photocathodes. However, as shown in XRD in Figure 1a and Figure S1, all four Cu3BiS3 samples have some impurity such as Copper sulfide or Molybdenum sulfide. Even the optimized sample (380 ℃ annealed) has an impurity peak (Mo2-xS). Synthesis should be improved to prepare a pure phase to clearly claim the unprecedented PEC performance of your Cu3BiS3-based photocathode. Otherwise, the role of the impurity phase should be clarified with additional material characterization.
5. There is a typo. (5*5 cm → 5*5 cm2) Reviewer #3 (Remarks to the Author): This paper concerns development of a water-splitting device based on a Cu3BiS3 photocathode and BiVO4 photoanode. The core claim is that Cu3BiS3 will be an industrially competitive solution for high performing, stable devices for solar hydrogen production in the near future.
From the scientific point of view the manuscript is well written and the investigations are detailed and appropriate. Specific comments are given below. However I do feel that the authors have overstated the impact of their work in relation to the state of the art for solar water spltting.
In the introduction, the performance of the Cu3BiS3 device is related to similar device types relying on various photoanode/photocathodes (given in Table 1). In this context, 2% efficiency and 60 hours of stability are reasonable results. But in the context of the state of the art of solar-driven water splitting as a whole, where the theoretical limit of efficiency is around 25% or more, the results are rather modest, and the major challenges in reaching this level are not illuminated. A more balanced introduction is needed so that the reader can put the claims of the authors into proper context. See e.g. * Energy Environ. Sci., 2015,8, 2811-2824* Energy Environ. Sci., 2018,11, 1977-1979* Chem. Soc. Rev., 2019,48, 1908-1971 Throughout the paper, the authors overstate the impact of their results and the status of the field as a whole. e.g. Page 5: "the Cu3BiS3-based photocathode will attract much attention worldwide and deserves to be widely investigated for large-scale industrial production in the near future." In fact, the performance of the given device, being far below simple solar-driven water electrolysis, could not possibly make the technology cost effective without major improvements. Similarly, repeated claims of "highly industrialized", "large scale" (meaning 5x5cm), "further industrialization" are all spurious, since there is no industry around this type of device. This type of language gives a misleading impression to non-specialists, and for specialists it appears that the authors are engaging in false advertising. Therefore it is advisable to moderate the language carefully.
Other comments * Why were the stability test stopped after short times (20h, 60h) when the devices were still performing quite well? Could data for longer tests be included? * For the TiO2-protected photocathodes, can the authors comment on the recoverability of performance if a fresh electrolyte is added? What is the cause of current reduction, given that no corrosion was detected?
I am fully satisfied with the revision with response to my comments, and now recommend its acceptance for puiblication in the current form.