Identification of recombination losses and charge collection efficiency in a perovskite solar cell by comparing impedance response to a drift-diffusion model.

Antonio Riquelme,Giles Richardson,Matthew J Wolf,Laurence J Bennett,Nicola E Courtier,Juan A Anta,Alison B Walker,Lidia Contreras-Bernal

doi:10.1039/d0nr03058a

Abstract

Interpreting the impedance response of perovskite solar cells (PSCs) is significantly more challenging than for most other photovoltaics. This is for a variety of reasons, of which the most significant are the mixed ionic-electronic conduction properties of metal halide perovskites and the difficulty in fabricating stable, and reproducible, devices. Experimental studies, conducted on a variety of PSCs, produce a variety of impedance spectra shapes. However, they all possess common features, the most noteworthy of which is that they have at least two features, at high and low frequency, with different characteristic responses to temperature, illumination and electrical bias. The impedance response has commonly been analyzed in terms of sophisticated equivalent circuits that can be hard to relate to the underlying physics and which complicates the extraction of efficiency-determining parameters. In this paper we show that, by a combination of experiment and drift-diffusion (DD) modelling of the ion and charge carrier transport and recombination within the cell, the main features of common impedance spectra are well reproduced by the DD simulation. Based on this comparison, we show that the high frequency response contains all the key information relating to the steady-state performance of a PSC, i.e. it is a signature of the recombination mechanisms and provides a measure of charge collection efficiency. Moreover, steady-state performance is significantly affected by the distribution of mobile ionic charge within the perovskite layer. Comparison between the electrical properties of different devices should therefore be made using high frequency impedance measurements performed in the steady-state voltage regime in which the cell is expected to operate.

Highlights

Perovskite solar cells (PSCs)[1,2] are one of the hottest research topics in contemporary photovoltaics due to the extremely rapid improvements in their performance, which has risen from a record photoconversion efficiency of 9.7% in 2012 3 to one of 25.2% in 2019.4 Metal halide perovskites are semiconductors with a direct band gap and strong absorption in the visible part of the spectrum.[5]
We show that our DD model can reproduce the main characteristics found in the impedance spectra, and provides a wide-ranging description of the Impedance spectroscopy (IS) features that determine the PSC photoconversion efficiency, noting that the main purpose of IS measurements is to detect and quantify loss mechanisms in a particular device
In particular we find that a DD model of the PSC can do so provided that it includes: (1) a three-layered planar configuration including n- and p-type selective contacts and a perovskite layer; (2) electrons in the electron transport layer (ETL); (3) electrons, holes and positively charged ionic vacancies in the perovskite active layer, (4) holes in the hole transport layer (HTL), (5) “rapid” diffusion coefficients for the electron/holes and “slow” diffusion coefficients for the ionic vacancies, respectively, (6) bulk electron–hole recombination in the perovskite layer mediated by a combination of bimolecular and electron-limited Shockley–Read–Hall kinetics and (7) band gaps, work functions, activation energies and dielectric constants in line with values reported in the literature

Summary

Introduction

Characteristics is that they have a mixed ionic-electronic conduction character, stemming from the high mobility of ionic vacancies in the perovskite structure.[6,7] This material property appears to be related to their exceptional optoelectronic properties, in particular, to their excellent photovoltaic charge separation behaviour,[8,9,10] but is associated with their chemical instability.[11]. Neukom et al.[35] were able to reproduce, within a single DD framework and with a single parameter set, a wide variety of experimental data for a specific PSC configuration (inverted), including small-perturbation frequency-modulated experiments such as IS at short-circuit None of these works focused on the high frequency impedance response and in particular how it can be used as a tool to deduce the physics associated with the all-important steady-state performance of the cell. Our model makes use of two simplifying assumptions, previously used in ref

30. The assumptions are

Results and discussion

Conclusions