Performance evaluation of an all inorganic CsGeI₃ based perovskite solar cell by numerical simulation

Highlights

• SCAPS-1D simulation of lead-free perovskite CsGeI₃-based solar cell was performed.

• TiO₂ and CuI were found to be the best material for ETL and HTL, respectively, to achieve optimal efficiency.

• The optimized efficiency has been calculated to be 10.8%.

Abstract

The rapid advancement of organo-metal halide perovskite solar cells has led to a certified power conversion efficiency (PCE) of >25%. However, their ability to be implemented and commercialized on a large scale is currently limited due to the presence of toxic elements and the costly hole transport material (HTM) in the highly efficient cells. In the present work, a theoretical study on lead-free, environmental-friendly CsGeI₃ based solar cell (SC) is demonstrated. Factors influencing the device parameters, such as thickness, doping, and defect concentration, have been thoroughly investigated to better understand material properties. With optimal material properties, the simulation results demonstrated 10.8% optimized PCE, J_sc of 22.08 mAcm⁻², V_oc of 0.667 V, and FF of 73.49%. These findings open up a new path for CsGeI₃ as light-absorbing material to achieve clean, renewable energy.

Introduction

Perovskite solar cells (PSCs) have established themselves as viable candidates in the solar industry due to their easy preparation process and low cost compared to standard crystalline Si SC [1]. Since the first work on PSC in 2009, PCE is drastically enhanced from 3.8% [2] to more than 25% within a decade [3]. The high PCE is due to several distinguished properties of perovskites like low exciton binding energy, improved absorption properties, and considerable diffusion length of carriers [4,5]. However, instability causes by the volatile organic components in PSC acts as a barrier for its development [6,7]. Recently, inorganic halide perovskites (IHPs) have gained momentum due to their inherent stability and outstanding photovoltaic performance [8,9]. Among these IHPs, cesium lead halide based SCs exhibited high efficiency and stability [10,11]. However, toxicity arises from the heavy toxic element like Pb, which is hazardous to human health and the environment, hinders its commercialization [6,12,13]. Furthermore, the highly efficient CsPbI₃ exhibits a non-perovskite yellow δ-phase at room temperature [14,15].

To address the issues of toxicity, lead-free perovskites have been explored by substituting Pb with similar kinds of divalent elements such as Sn [16], Ge [17], Ti [18], Bi [19], etc. In addition, several lead-free double perovskites have been explored [20,21]. Among these alternatives, Sn-based PSCs have demonstrated the highest efficiency, thereby having sparked so much interest in PSC. Furthermore, it was reported that CsSnI₃ exhibited large hole mobility of 585 cm² V⁻¹s⁻¹ compared to its Pb counterpart [22]. Very recently, the best efficiency of 10.1% was achieved from CsSnI₃ PSCs [22]. However, the oxidation of Sn element from Sn²⁺ to Sn⁴⁺ under ambient atmosphere induces instability in cesium tin halide perovskites, resulting in degradation of PSCs [23]. In recent years, several theoretical and experimental works have been performed on lead-free Ge-based perovskites for photovoltaic applications. By alloying Sn and Ge, Chen et al. synthesized CsSn_0.5Ge_0.5I₃-based PSCs that showed improved stability and air tolerance at ambient conditions. Because of the highly oxidized characteristics of Ge (II), a homogeneous native oxide surface was developed on CsSn_0.5Ge_0.5I₃, demonstrating improved stability, more than that of CH₃NH₃PbI₃ perovskite [24]. DFT-based results showed that CsGeI₃ possesses a direct band gap of 1.63 eV, through strain engineering, its properties such as band gap could be tuned for suitable SC applications [25]. Computational results revealed that CsGeI₃ possesses a smaller effective mass for the hole and can be employed as HTM. By the comprehensive theoretical and experimental study of CsGeI₃, Krishnamoorthy et al. suggested that Ge is a possible alternative for Pb in halide perovskites.

Furthermore, it was reported that CsGeI₃ exhibits a stable perovskite structure up to 350 ^°C and delivered a PCE of 0.11% [17]. In another work, 3.2% PCE, J_sc of 10.5 mAcm⁻² and V_oc of 0.57 V, was reported on CsGeI₃-based PSCs. Recently, L.J. Chen synthesized CsGeI₃ perovskite quantum rods using a solvothermal method and obtained 4.9% PCE by tuning the configuration of quantum rods [26]. In addition, a thermogravimetric study on CsGeI₃ perovskite revealed that Ge-based perovskites are more thermally stable under device operating conditions [27]. These findings demonstrate Ge-based inorganic perovskites to be a promising candidate for lead-free PSCs.

Like perovskite materials, organic charge transport materials (CTMs), for example, spiro-OMeTAD, PEDOT:PSS and PCBM become unstable under illumination and ambient conditions [[28], [29], [30], [31]]. Furthermore, several organic CTMs react with perovskite, exhibit hostile behaviour because of their hygroscopic nature, and are expensive to prepare [32]. To overcome these drawbacks, inorganic CTMs, such as NiO [33], TiO₂ [34,35], SnO₂ [36,37] have been explored, which offer better transparency in UV, visible and infrared spectrums, a wide band gap, high carrier mobility, improved thermodynamic and chemical stability, and a simple synthesis method [38].

Here, we have presented a simulation study on lead-free CsGeI₃ PSCs and show its potential application in photovoltaics. We have proposed an all-inorganic PSC by using inorganic CTMs, e.g., TiO₂, SnO₂, ZnO, CdS, WS₂, CuI, Cu₂O, CuSCN and NiO. By optimizing various material properties like thickness, doping concentration, defect density, and studying the resistances effect and working temperature, we achieved an optimal efficiency of 10.8%. In addition, we have investigated the impact of various materials as a metal electrode on device performance. Furthermore, the presented simulation findings could aid in the design and fabrication of future lead-free Ge-based PSCs.

The numerical simulation has been conducted by employing 1D-Solar Cell Capacitance Simulator (1D-SCAPS) software [39], under AM 1.5 G. Solving Poisson's equation (1), and continuity equation of electron (2) and hole (3), SCAPS can evaluate several electrical and optical properties like current-voltage characteristics, energy bands and quantum efficiency.

$$\frac{\partial }{\partial x}\left(-\epsilon \left(x\right)\frac{\partial V}{\partial x}\right)=q\left[p\left(x\right)-n\left(x\right)+N_D^{+}\left(x\right)-N_A^{-}\left(x\right)+p_t\left(x\right)-n_t\left(x\right)\right]$$

$$\frac{\partial n}{\partial t}=\frac{1}{q}\frac{\partial J_n}{\partial x}+G_n-R_n$$

$$\frac{\partial p}{\partial t}=-\frac{1}{q}\frac{\partial J_p}{\partial x}+G_p-R_p$$

Here $q,\epsilon ,V,p\left(x\right),n\left(x\right),N_D^{+}\left(x\right),N_A^{-}\left(x\right),p_t\left(x\right)and{n}_t\left(x\right)$ represent electronic charge, dielectric permittivity, electric potential, free hole concentration, free electron concentration, ionized donor concentration, ionized acceptor concentration, trap density of holes and trap density of electrons, respectively. Furthermore, J_n, J_p, G_n, G_p, R_n and R_p denote current density, generation and recombination rate for electrons and holes, respectively.

The structure of the PSC employed in the initial simulation process is illustrated in Fig. 1(a), which comprises of FTO-coated glass, TiO₂ for electron transport layer (ETL), CsGeI₃ for light-absorbing material, spiro-OMeTAD layer for hole transport layer (HTL), and Ag for the electrode. The basic material properties extracted from various theoretical and experimental results are listed in Table 1. For all the materials, the thermal velocity of electrons and holes are taken as 1 × 10⁷ m/s, while the absorption coefficient α has been calculated using the formula $\alpha =A_{\alpha }{(h\nu -E_g)}^2$ with the value of $A_{\alpha }$ as 10⁵ [40]. We have chosen neutral single distribution with 0.1 eV characteristic energy to introduce the defects in each device layer. Moreover, two interface layers, TiO₂/CsGeI₃ and CsGeI₃/spiro-OMeTAD are included in the device structure to investigate the impact of interface recombination, thereby simulating a more realistic situation. The physical parameters of interface layers are described in Table S1. The total defect densities for both the interfaces are taken to be 1 × 10¹⁶ cm⁻³. The work functions of back and front contact are taken as 4.57 and 4.4 eV, respectively.

Using the initial parameters summarized in Table 1 and Table S1, we plotted the energy band diagram, current-voltage characteristics, and QE as depicted in Figs. 1(b) and figure 2 (a) and (b), respectively. The initial simulated device achieved 4.99% PCE, J_sc of 17.87 mAcm⁻², V_oc of 0.47 V, and FF of 59.68%, almost close to reported experimental findings as illustrated in Table 2. From Fig. 2, a good match between experimental results and our simulation was noticed, demonstrating validation of this work. However, the simulated FF was relatively higher than the experimental value, as we have neglected the series and shunt resistance in the simulation process [41].