Review Reports

Reviewer’s comments

Author’s response

Reviewer 1

The study, titled "Intrinsic photoconductivity spectral dependence as a tool for prediction of open-circuit voltage in organic solar cells," by Raitis Grzibovskis, Andis Polaks, and Aivars Vembris from the University of Latvia, delves into predicting parameters of organic photovoltaic (OPV) cells. The paper emphasizes the advantages of organic materials in solar cell development, such as cost-effectiveness and adaptability. Traditionally, the open-circuit voltage (Uoc) is estimated using the ionization energy of electron donor and acceptor molecules. This research introduces a novel approach using photoconductivity measurements on donor: acceptor films. Results indicate that this method offers more accurate Uoc predictions than conventional methods. The study utilized various organic materials and employed photoelectron emission spectroscopy for measurements. Generally, we think this manuscript is of good quality and can be accepted by MDPI Energies after the authors address the following questions:

1.     In this study, the authors used materials like P3HT, PCDTBT, PBDB-T, PCBM, and Y-series acceptors. How were these particular organic electron donor and acceptor materials selected over others available in the field?

The materials were chosen by three principles:

1) Various material groups/ classes to avoid showing the effect only in one small class of materials;

2) Their energy level values covered a wide range of possible U_oc values from 0.45 up to 1.20 V;

3) PBDB-T series and Y- series allowed us to evaluate whether some material group creates systematic errors in this research.

2.     Authors used the same device fabrication conditions (solution concentrations, D:A ratios, and spin-coating process) to make the films for their photoconductivity measurement in this study. However, it is well known that the fabrication conditions have a significant impact on active layer film morphology (Energy & Environmental Science 16 (3), 1234-1250), which can therefore impact the Uoc value of the device. We suggest that the authors should fabricate devices with different fabrication conditions, such as different D:A ratio and different film thicknesses. And then apply their photoconductivity measurement techniques on these devices to explore the impact of fabrication conditions on the accuracy and sensitivity of the Uoc prediction.

An additional series of samples was made using a PBDB-T-2Cl:Y5 material combination. The donor:acceptor mass ratio was varied (1.2:1, 1:1, 1:1.2), as well as samples made from different solutions (the chloroform was added to the chlorobenzene). Here we observed two effects:

1) mass ratio changes gave insignificant changes in U_oc and E_CT. The changes were around 0.02 V which can be considered to be within the margin of error;

2) adding chloroform to the solution lowered the U_oc by 0.05 V. The E_CT value obtained from photoconductivity measurements was lower by 0.03 eV compared to the sample without chloroform. It shows that we can detect changes in the photoconductivity spectrum if the quality of the active layer is significantly altered.

3.     More and more research has shown that the small modification of the molecular structure significantly changes the Uoc and current density of the OPV devices (JACS 143 (16), 6123-6139. JOULE 10.1016/j.joule.2023.07.005). These changes in device parameters may not be explained directly by the photoconductivity difference of the materials. We would like to author to include some discussion on this topic in this manuscript as it can benefit the readers from OPV synthetic background.

There are changes in the energy level values when PBDB-T, as well as Y5 molecules, are modified by chlorine or fluorine atoms- their energy levels are “deeper”. A deeper ionization energy level value for electron donor material means higher U_oc (expected as well as obtained). Deeper energy levels for electron acceptor material mean lower U_oc as the difference between the ionization energy level of the donor (I_d) and the electron affinity of the acceptor (EA_a) decreases. In this work, no systematic errors (always underestimating or always overestimating) were observed for any material class or any material modification.

The reasons behind the changes in charge carrier transport (obtained current), changes in the active layer morphology due to the “softer” or “stiffer” bonds, or similar effects due to the added chlorine/ fluorine atoms were out of the scope of this manuscript.

Reviewer’s comments

Author’s response

Reviewer 3

This work demonstrates that the photoconductivity measurement enables better U_oc estimation of OPV devices. The charge transfer energy (E_CT) was obtained by measuring the photoconductivity spectrum of the D:A film. This approach provides less scattered results with a higher correlation coefficient compared to the U_oc estimation using energy level values. This job is with good novelty and should have a broad readership. I suggest that this paper could be published after being revised according to the following suggestions:

1.      The value was obtained as threshold energy in the photoconductivity spectrum. In Fig. 2b, is the method of determining gap energy by the intersection of two red lines reasonable? Can the authors give more sufficient evidence?

In Fig.2b we have shown the example of gap energy (E_g) determination. According to the power law β(hν)∝(hν-E_g )^y, in the perfect case, there should be no photocurrent when hν< E_g.  The small photoconductivity signal below this gap (in this case- in the spectral region between 1.3 and 1.5) is considered to be created by the impurities in the material (PBDB-T-2Cl) as the material was used as-received (without any purification). When the photon energy is higher than the E_g, the photoconductivity efficiency rapidly increases. In this case, we cannot simply extrapolate this rapid increase just to the value of β(hν)^2/5=0, but we need to take into account the small photocurrent below Eg. That is why we use the cross point between two lines.

2.      Figure 2 shows the spectrum of photoelectron emission and photoconductivity. The meaning of the vertical coordinates (Y^2/5 and β^2/5) of these two graphs is not clearly indicated.

The explanation of Y and β coordinates and why such power laws are used has been added to the Results section in the manuscript.