# Investigation of the Spatially Dependent Charge Collection Probability in CuInS$_2$/ZnO Colloidal Nanocrystal Solar Cells

## TL;DR

This study investigates how the spatial location of charge generation within CuInS$_2$/ZnO nanocrystal solar cells affects charge collection efficiency, revealing a narrow active zone near the interface and providing a transferable optical modeling approach.

## Contribution

It introduces an optical modeling method to analyze charge collection spatial dependency without electrical parameters, applied to CuInS$_2$/ZnO solar cells.

## Key findings

- Charge collection mainly occurs within 40 nm of the interface.
- A significant 'dead zone' exists where charge is not collected.
- The optical model accurately predicts charge collection profiles.

## Abstract

Solar cells with a heterojunction between colloidal CuInS$_2$ and ZnO nanocrystals are an innovative concept in solution-processed photovoltaics, but the conversion efficiency cannot compete yet with devices employing lead chalcogenide quantum dots. Here, we present a detailed study on the charge collection in CuInS$_2$/ZnO solar cells. An inverted device architecture was utilized, in which the ZnO played an additional role as optical spacer layer. Variations of the ZnO thickness were exploited to create different charge generation profiles within the light-harvesting CuInS$_2$ layer, which strongly affected both the external and internal quantum efficiency. By the reconstruction of these experimental findings with the help of a purely optical model, we were able to draw conclusions on the spatial dependency of the charge collection probability. We provide evidence that only carriers generated within a narrow zone of circa 40 nm near the CuInS$_2$/ZnO interface contribute to the external photocurrent. The remaining part of the absorber can be considered as "dead zone" for charge collection, which reasonably explains the limited device performance and indicates a direction for future research. From the methodical point of view, the optical modeling approach developed in the present work has the advantage that no electrical input parameters are required and is believed to be easily transferable to other material systems.