Comprehensive Analysis of Fullerene- and Non-Fullerene-Based Bulk Heterojunction Solar Cells Using Numerical Simulation

Abstract

In recent years, two-dimensional (2D) materials have been widely used for various applications due to their low cost, high charge carrier mobility, and tunable electronic structure. Here, in this study, we present the application of molybdenum disulfide (MoS₂) used as a hole transport layer (HTL) material for fullerene (FA) and non-fullerene (NFA)-based organic photovoltaic (OPV) devices. A numerical simulation is carried out for these types of solar cells, and the SCAPS-1D software tool is used. Our study is specifically focused on the impact of thickness, the optimization of interface engineering, and the effect of high-temperature analysis to improve the output characteristics. The influence of interface defects between the HTL/active layer and the active layer/ETL (electron transport layer) is also contemplated. After optimization, the obtained power conversion efficiency (PCE) of these NFA- and FA-based devices is reported as 16.38% and 9.36%, respectively. A reflection coating study is also carried out to improve the power conversion efficiency of these devices. Here, the presented results demonstrate that molybdenum disulfide (MoS₂) as a 2D material can be successfully used as an HTL material for high-efficiency OPV devices, both for fullerene (FA)- and non-fullerene (NFA)-based solar cells.

1. Introduction

Due to the increasing demand for energy, renewable energy resources are being developed, which are essential [1,2]. However, a large amount of energy is still obtained from fossil fuels. To find alternatives to non-renewable energy resources, very efficient “green energy” is needed. The fastest growing field of renewable energy is solar energy, which can be utilized into two ways, i.e., directly or indirectly. In the direct way, solar energy is directly converted into electrical energy through PV, while in the indirect way, firstly, the solar energy is converted into heat energy, and then the heat energy is converted into electrical energy [3,4]. Solar cells can be divided into three generations. Third-generation solar cells include dye-sensitized solar cells (DSSCs), organic solar cells (OSCs), and perovskite solar cells (PSCs) [5,6,7].

Nowadays, researchers are mainly focused on OSCs due to reasons such as their low economic cost and rapid increase in power conversion efficiency (PCE) [8,9]. As OSCs of various generations, among which bulk heterojunction (BHJ) OSCs are widely used due to their architecture,which can be modified to obtain efficient and stable PV devices. In BHJ OSCs, the large interfacial area is employed between the donor and acceptor materials, in which more charge collection and separation occur [10,11,12].

Acceptor materials can be divided into two categories, i.e., fullerene acceptors (FAs) and non-fullerene acceptors (NFAs). Currently, NFAs are widely used; however, the benefit of FAs cannot be ignored because FAs also have good properties such as high electron affinity and mobility at room temperature, i.e., 1 cm² V⁻¹ s⁻¹. However, they show rather weak absorption and suffer from a high preparation cost and poor stability [12,13,14]. The main advantages of NFAs compared to FAs is that NFAs have a low production cost, as well as low losses in voltage, and that the energy levels of NFAs can easily be tuned [12,13,14,15].

Numerous theoretical and simulation studies have been carried out on polymer-based solar cells. In [16], the authors reported that using PCBM/P3HT as an absorber can achieve an efficiency of 2.8%. To determine the optimum thickness of the active layer, a simulation study is carried out on PTB7:Th/PC71BM [17], where it is demonstrated that 8.15% can achieved for 270 nm thickness. A simulation study in [18] performed on the PTB7:PCBM active layer by using PEDOT:PSS as a hole transport layer (HTL) and PFN-Br as an electron transport layer (ETL) achieved an efficiency of 8% by optimization techniques.

Also, different inorganic materials have been used as HTLs in OSCs. Chahrazed et al. [19] performed their analysis on PTB7:PCBM by using different inorganic hole transport materials, i.e., MoO₃, NiO, V₂O₅, and Cu₂O. They examined the effect of absorbers and HTM on the device’s performance, and the best efficiency is reported for Cu₂O, i.e., 5.44% [19]. Elham et al. [20] carried out a simulation study on P3HT:PCBM by using tungsten oxide (WO₃) as an HTL, and in this study, PEDOT:PSS was replaced by WO₃. It was analyzed that the PCE was enhanced from 5.21% to 8.67% by using silver as a back contact along with WO₃, while in the case of a gold (Au) back contact, the PCE was enhanced from 4.83% to 8.4%.

Interfacial thin layers have been investigated to improve the efficiency of OSCs. In [21], the authors carried out a numerical investigation of P3HT:PCBM by using ZnO as an interfacial layer, while PEDOT:PSS was employed as an HTL material. In this article, they varied the thickness of the ZnO layer, and it was concluded that 20 nm was the optimized thickness of the interface layers, and better efficiency was achieved, i.e., 4.88%.

Different materials have been considered as NFAs in the structure of BHJ organic solar cells. Zhan et al. invented one of the most important NFAs, i.e., the (3,9-bis(2-methylene-(3-(1,1dicy-anomethylene)-indanone)-5,5,11,11-tetraki (4-hexylphenyl)-dithieno [2,3-d:2,3-d]-s-indaceno [1,2-b:5,6-b]dithiophene (ITIC) small molecule. It turns out that ITIC is an excellent acceptor due to its superior characteristics. Several studies employed ITIC as an acceptor and PBDB-T as a donor [22]. Nathya and Sudheer [23] performed a numerical simulation of NFA OSCs by using various HTLs, i.e., CuI, MEH-PPV, NiO, and PEDOT:PSS. Abdelaziz et al. [24] reported a study on PBDB-T:ITIC by using PEDOT:PSS as an HTL material and also optimized the active layer thickness, defect density, interface defect density, and temperature effect, while the efficiency was reported as 14.25%. Ayushi Katariy et al. [25] presented a study on optimization of an NFA-based OSC by using various HTM layers, i.e., CuI, CuSCN, and Cu₂O, and different ETMs, i.e. Tio₂, SnO₂, and C60. From the simulation, it was observed that the best combination of ETL/HTL along with the active layer (PBDB-T:ITIC) was SnO₂ (ETL) and Cu₂O (HTL), and the PCE was reported as 27.92%. Eriwidianto et al. [26] performed a comprehensive analysis of the absorber layer (PBDB-T:ITIC) by using novel HTL materials by replacing PEDOT:PSS with 2D materials. From the analysis, the highest efficiency was reported for WS₂, i.e., 23.55%. Nowsherwan et al. [27] performed an analysis on the optimization of PBDB-T:ITIC-based OSCs by employing graphene oxide (GO) as the HTL, and the efficiency was reported as 17.36%.

Next, an investigation on inverted-structure PBDB:T/PZT-based polymer solar cells was carried out by the authors of [28], and the obtained efficiency was reported as 19.92%. From the analysis, it was observed that the inverted structure had better performance than the conventional structure because in these inverted structures, the light enters from the ETL side, which is thinner as compared to HTLs [28].

Two-dimensional materials are explored for various different applications, such as photocatalysis and solar cells [26,29,30]. Two-dimensional materials have attracted attention from the scientific community due to their characteristics, including attractive electrical, optical, and mechanical properties [31,32,33]. Graphene oxide (GO) is one of the most important graphene derivatives and it can be prepared by introducing functional groups to the graphene surface. By adding these functional groups, the graphene optical and electronic properties can be changed [34,35,36,37,38,39,40]. Besides graphene and graphene dioxide, other 2D materials, i.e., transition metal dichalcogenides (TMDSs), are also a subject of interest [39,40,41]. The parameters of the CTLs (charge transport layers), such as their energy gaps (E_g) and work functions (Wf), can be controlled by using various techniques, such as doping, atomic scale thickness, and passivation techniques [41]. Previously, low-dimension materials such as MoS₂ and WS₂ were used in OPV structures due to their excellent carrier mobility, high optical transparency, and suitable energy levels [41,42,43]. These 2D materials not only improve photon absorption but also charge transport, and they decrease carrier recombination. As a result of these processes, PV parameters such as the current density (J_sc) and power conversion efficiency (PCE) can be improved.

The main focus of our work is computational studies of the structures of P3HT:PCBM and PBDB-T:ITIC-OE, using a 2D HTL material, along with PFN-Br as the ETL material. Numerical modelling is an effective method to understand the physics of device effects and to improve the efficiency of devices for future photovoltaic applications. Previously, PEDOT:PSS has been widely used as an HTL material in OSCs due to features such as its high work function and conductivity. But the main problem is that the PEDOT:PSS nature is acidic and it degrades metal electrodes. To overcome this issue, other materials such as V₂O₅, WO₃ and NiO₂ have been used, but these materials require high annealing temperatures. To extract charges from the active layer and move towards electrodes efficiently, efficient HTL materials are needed.

In the literature, a similar study was performed on PBDB-T:ITIC [26] by using 2D HTL materials, but in our analysis, presented herein, we used a modified active layer of ITIC-OE, which has a high dielectric constant, compared to ITIC. In our study, P3HT:PCBM (fullerene-based active layer) and PBDB-T:ITIC-OE (non-fullerene-based modified active layer) are used to compare the effects of different types of acceptor on the PV parameters of BHJ organic solar cells. Moreover, our analysis is carried out by using molybdenum disulfide as a 2D-HTM along with PFN-Br as the ETL material.

In this study, we present, step by step, the influence of different factors on the PV parameters of fullerene- and non-fullerene-based solar cells. This comprehensive analysis is focused on the absorber layer thickness, the density of defects, the interface defect density, and the effect of temperature on PV parameters. At the end, it is shown that the PCE is enhanced by employing back-side reflection coating.

2. Materials and Methods

2.1. Numerical Approach and Proposed Structure

Numerical simulation can reduce time and expense, particularly in the case of errors and uncertainty that can occur during an experiment. The SCAPS-1D software has been used for the comprehensive analysis of fullerene- and non-fullerene-based acceptors by using 2D-HTL materials. The SCAPS-1D (SCAPS 3.3.11) software was invented by Ghent University in Belgium [44]. Defects can be introduced in the bulk of a material or at the interfaces, i.e., HTL/active layer or active layer/ETL defects. Common measurements such as IV, IQE, CV and CF can be simulated by using the three main equations [45] of the Poisson Equation (1), the transport Equation (2), and the continuity Equation (3).

Poisson equation:

$${\displaystyle \frac{{\partial }^2\phi }{{\partial }^{2}x}}=-{\displaystyle \frac{\partial E}{\partial x }}=-{\displaystyle \frac{\rho }{{\epsilon }_s}}=-{\displaystyle \frac{q}{{\epsilon }_s}} [p–n+ N_D\left(x\right)-N_A\left(x\right) \pm N_{def}\left(x\right)]$$

(1)

Transport equation:

$${\mathrm{J}}_{\mathrm{n},\mathrm{p}}={\mathrm{nq}\mathsf{\mu }}_{\mathrm{n} }\mathrm{E} +{ \mathrm{qD}}_{\mathrm{n}}{\displaystyle \frac{\partial \mathrm{n} }{\partial \mathrm{x} }}+{ \mathrm{pq}\mathsf{\mu }}_{\mathrm{p}}\mathrm{E} +{ \mathrm{qD}}_{\mathrm{p}}{\displaystyle \frac{\partial \mathrm{p}}{\partial \mathrm{x} }}$$

(2)

Continuity equation:

$${\displaystyle \frac{\partial \mathrm{n},\mathrm{p} }{\partial \mathrm{t} }}={\displaystyle \frac{1{ \mathrm{J}}_{\mathrm{n}} }{\mathrm{q} \partial \mathrm{x} }}+\left({\mathrm{G}}_{\mathrm{n}}-{ \mathrm{R}}_{\mathrm{n}}\right)+{\displaystyle \frac{1{ \mathrm{J}}_{\mathrm{p}} }{\mathrm{q} \partial \mathrm{x} }}+\left({\mathrm{G}}_{\mathrm{p}}-{ \mathrm{R}}_{\mathrm{p}}\right)$$

(3)

In the above equations, φ, ε, q, p, n, ND, N_A, and N_def represent electrostatic potential, dielectric permittivity, electron charge, hole and electron concentration, donor density, acceptor density and defect density, respectively. J_n, μ_n, μ_h, D_N, D_P, and E represent the current density for holes and electrons, electron mobility, hole mobility, the diffusion coefficient for electrons and holes, and electric field, respectively. G_n, G_P, R_n, and R_p represent the generation and recombination rates for holes and electrons, respectively. The configurations of simulated cells are shown below in Figure 1.

2.2. Parameters of Simulated Devices

The general structure of the solar cell used for the simulation is as follows: front contact/ETL/absorber/HTL/back contact. In this simulation, two different active layers were used; one active layer is based on a polymer and non-fullerene-based acceptor (PBDB-T:ITIC-OE), while the other one consists of a polymer and fullerene-based acceptor (P3HT:PCBM), along with MoS₂ as the HTL material and PFN-Br as the ETL. Herein, PCBM is used as the fullerene acceptor while ITIC-OE is employed as the non-fullerene-based acceptor in the active layer of BHJ solar cells. Fluorine doped tin oxide (FTO) is used as front contact electrode while Platinum (Pt) i s employed as back contact electrode . The work function of front and back electrode is used as 4.08 eV and 5.64 eV respectively . The main physical and technological parameters, along with symbols for the active layers, HTL, and ETL are tabulated in

Table 1. Physical parameters used in these simulations for the ETL, HTL, and non-fullerene- and fullerene-based active layers.

Parameters *	MoS2 [26]	PBDB-T:ITIC-OE [23,45,46,47]	PFN-Br [48]	P3HT:PCBM [20]
Eg [eV]	1.23	1.2	2.8	1.85
εr	4	6.1	5	3.5
Ea [eV]	4.2	4.030	4	3.95
µn [cm2V−1s−1]	100	1.2 × 10−5	1 × 10−4	2 × 10−3
µp [cm2V−1s−1]	150	3.5 × 10−4	2 × 10−6	2 × 10−3
ND [cm−3]	0	0	9 × 1018	0
NA [cm−3]	2 × 1018	0	0	0
Nt [cm−3]	1015	1012	109	1014
NC[cm−3]	2.2 × 1018	1 × 1019	1 × 1019	1 × 1019
Nv [cm−3]	1.8 × 1018	1 × 1019	1 × 1019	1 × 1019

* E_g—_energy gap; ε_r—relative dielectric constant permittivity; E_a—electron affinity; µ_n—electron mobility; µ_p—hole mobility; N_D—donor density; N_A—acceptor density; N_t—defect density; N_C—conduction band effective density of state; N_v—valence band effective density of state.

.

All values of parameters which are required for the simulation are collected from the literature. This simulation is performed under the standard test conditions (STCs), which include air mass of 1.5 G, temperature of 300 K, and irradiance of 1000 W/m². For each layer, the optical absorption coefficient α(hv) is set in the SCAPS model, which uses the square root model [26].

3. Results and Discussion

3.1. Influence of Thickness on Fullerene- and Non-Fullerene-Based Active Layers

The thickness of the active layer plays a very important role to enhance the performance of a device and to improve its current density. Ideal thickness is required to obtain reasonable output from OPV devices. A too-thick layer can cause recombination before reaching the electrode, while a too-thin layer diminishes photon absorption and decreases the photocurrent value. From previous research papers [20,23,26,45,46], it was determined that the active layer (BHJ) thickness can affect the output characteristics, such as Jsc (short-circuit current), Voc (open-circuit voltage), PCE, and FF (fill factor). Here, in this study, the active layer thickness of the fullerene-based acceptor varied from 20 nm to 300 nm, while for the non-fullerene-based acceptor, it changed from 20 nm to 380 nm. It is observed that, in the case of the fullerene-based BHJ layer above 290 nm thickness, the PCE decreases, while for the non-fullerene-based BHJ layer, the PCE decreases above 370 nm. For comparative purposes, the thickness of both active layers is set as 140 nm, because above 140 nm, the PCE is not significantly improved in the case of non-fullerene-based acceptors. The HTL and ETL thicknesses are set as 10 nm and 5 nm, respectively.

It is analyzed that Voc, Jsc and PCE improved for both fullerene- and non-fullerene- based organic solar cells with the increase in thickness, which is illustrated in Figure 2a,b,d. This means that with the increase in thickness, both light absorption and the generation of charge carriers increase, which can improve the performance of solar cells. The FF decreases for both fullerene- and non-fullerene-based solar cells when the active layer thickness increases from 40 to 140 nm and 60 to 140 nm, respectively, which is illustrated in Figure 2c. The decrease in FF might be due to the increase in series resistance. Moreover, the value of Voc slightly increases, which is affected by Jo (the reverse saturation current density). Mathematically, it is written in [45] as the following dependences (4):

$$\mathrm{FF}={\displaystyle \frac{V_{MP }{J}_{MP}}{V_{OC }{J}_{SC} }};{ \mathrm{V} }_{\mathrm{OC}}={\displaystyle \frac{\mathrm{nKT}}{\mathrm{q}}} \left[{\mathrm{I}}_{\mathrm{n}}\left({\displaystyle \frac{\mathrm{Jsc}}{{\mathrm{J}}_{\mathrm{O}}}}+1\right)\right]$$

(4)

where (V_MP, J_MP) represent the voltage and current density at maximum power, while ${\displaystyle \frac{{\mathrm{K}}_{\mathrm{B}}\mathrm{T}}{\mathrm{q}}}$ represents thermal voltage.

From the above equation, it is clear that V_oc is limited by Jo.

3.2. Optimization of Defect Density through Interface Engineering (Fullerene- and Non-Fullerene-Based Active Layers)

The active layer is the region in solar cells where photovoltaic processes take place, which include the generation, recombination, and transportation of charge carriers. In the previous literature, it is reported that morphological properties have a major impact on the photo-generated diffusion length [23]. Defect density indicates the morphology and quality of active layers and can affect the performance of organic-based solar cells. Charge carriers trap these defects and can cause non-radiative (electron and hole pairs) recombination, which reduces the diffusion length, according to the Shockley–Read–Hall recombination (SRH) [23,26,45] presented below (5):

$$\mathrm{R} ={\displaystyle \frac{\mathrm{np}-n{i}^2}{\mathsf{\tau }\mathrm{p} \left(\mathrm{n} + \mathrm{nt}\right)+ \mathsf{\tau }\mathrm{n}\left(\mathrm{P} + \mathrm{Pt}\right)}}$$

(5)

In the above equation, P and n represent the concentration of electrons and holes; $\mathsf{\tau }\mathrm{p}$ and $\mathsf{\tau }\mathrm{n}$ represent the lifetime of electrons and holes, respectively. The lifetime of charge carriers and diffusion length are represented by Equations (6) and (7) respectively.

$$\mathsf{\tau }={\displaystyle \frac{1}{ \mathsf{\sigma }\ast \mathrm{Nt}\ast \mathrm{Vth}}}$$

(6)

$${\mathrm{L}}_{\mathrm{n},\mathrm{p}}=\sqrt{{\mathrm{D}}_{\mathrm{p},\mathrm{n}}{ \mathsf{\tau }}_{\mathrm{n},\mathrm{p}}}$$

(7)

$${\mathrm{D}}_{\mathrm{p},\mathrm{n}}=\left[\left({\displaystyle \frac{{\mathrm{K}}_{\mathrm{B}}\mathrm{T}}{\mathrm{q}}}\right){\mathsf{\mu }}_{\mathrm{n},\mathrm{p}}\right]$$

(8)

In Equations (6)–(8), σ, N_t, V_th, and D_p,n represent the cross-section area, defect density, and diffusion coefficients for electrons and holes, respectively.

Defect density is an undesirable phenomenon in any material. Defects can be seen in the bulk or on the surface and may be in the form of Schottky points or interfacial and Frenkel defects. The presence of defects reduces the diffusion length of charge carriers and enhances electron–hole recombination. In this study, the defect densities for both layers are changed from 10¹¹ cm⁻³ to 10¹⁴ cm⁻³. The optimized defect density values in the literature for non-fullerene acceptors are reported around 10¹² or 10¹³ cm⁻³, while for fullerene- based acceptors, the values are 10¹⁴ cm⁻³ or 10¹⁵ cm⁻³. Below this value, the output characteristics for both active layers are enhanced, but it is very difficult to design such a type of solar cell with less defects. As defect densities increase from 10¹¹ to 10¹⁴ cm⁻³, the PCE and Jsc values decrease, which is depicted in Figure 3a,b.

3.3. Impact of Carrier Diffusion Length on Solar Cell Performance

Carrier diffusion length is an important factor in solar cell performance that depends on the defect density. In this simulation study, the diffusion lengths for both solar cells are calculated, which is tabulated in

Table 2. Defect density variation along with diffusion length and PCE for non-fullerene-based solar cells.

Nt (1/cm3)	1011	1012	1013	1014
Ln (nm)	180	56	18	5.6
FF (%)	64.02	58.87	46.70	41
PCE (%)	16.49	14.66	10.11	6.69

and

Table 3. Defect density variation along with the diffusion length and PCE for fullerene-based solar cells.

Nt (1/cm3)	1011	1012	1013	1014
Ln (nm)	2300	720	230	72
FF (%)	79.65	79.58	78.87	73.36
PCE (%)	8.77	8.76	8.67	7.90

. It is analyzed that the diffusion length decreases with the increase in defect density, which affects the photovoltaic characteristics. A strong impact of the diffusion length on the FF value is observed, which decreases with the decrease in diffusion length. The lowest fill factor is observed at diffusion lengths of 5.6 nm and 72 nm for NFA- and FA-based active layers, with the maximum at 180 nm and 2300 nm, respectively. It is also observed that PCE decreases with the decrease in the diffusion length of charge carriers. So, the optimized diffusion length for fullerene-based solar cells is 72 nm, and that for non-fullerene-based solar cells is 56 nm. The charge carrier diffusion length depends on the diffusion of electrons and holes and the lifetimes of charge carriers. Mathematically, from Equation (8), diffusion depends on the electron mobility. Here, in this study, the electron mobility of fullerene-based blends is higher than for non-fullerene-based blends, so the diffusion length of fullerene-based blends is higher than non-fullerene-based blends.

3.4. Effect of Density of Defects on the Active Layer Interfaces

There are two types of interfaces: (a) HTL/active layer and (b) active layer/ETL. In this simulation study, the defect densities of both interfaces are varied from 10¹⁰ cm⁻² to 10¹⁴ cm⁻² for fullerene- and non-fullerene-based acceptors, which is shown in Figure 4a–d. From the analysis, it is observed that by changing the interface density of HTL/NFA, the Jsc decreases from 22.67 mA/cm² to 22.56 mA/cm², while the PCE decreases from 14.66% to 14.47%. However, for the NFA/ETL interface, Jsc decreases from 22.67 mA/cm² to 22.48 mA/cm², while PCE goes down from 14.66% to 14.04%. The interface defect density of fullerene-based acceptors is also analyzed. It is concluded that HTL/FA has no impact on the performance of solar cells, while FA/ETL has a little impact on the performance of the device. The reason is that the structure is inverted, and the light is illuminated from the ETL side. For practical concerns, the interface defect density value for both layers is set as 10¹⁰ cm⁻².

3.5. Effect of Doping on Charge Transport Layers

The main purpose of doping is to improve the charges’ movement towards their respective electrodes. The main criteria to be fulfilled by the doping carrier are minimum resistivity and maximum conductivity. By increasing the doping concentration, the electrical field at the absorbing interface is improved, which boosts the charge movement and enhances the output characteristics of solar cells [49]. The doping values for both charge transport layers (CTLs), i.e., for the HTL and ETL, are taken from the literature, and are 2 × 10¹⁸ cm⁻³ and 9 × 10¹⁸ cm⁻³, respectively. In order to observe the effect of doping, these doping values are changed from 2 × 10¹⁸ cm⁻³ to 2 × 10¹⁵ cm⁻³ in the case of hole transport layers, while in the case of electron transport layers, the doping value is changed from 9 × 10¹⁸ cm⁻³ to 9 × 10¹⁵ cm⁻³, which is shown in Figure 5a–d. It is observed that for non-fullerene-based acceptors, by varying the doping density of the HTL, the Jsc is increased from 22.46 mA/cm² to 22.67 mA/cm² and the PCE is enhanced from 14.25% to 14.66%, while for fullerene-based acceptors, the Jsc is not affected, and simultaneously, the PCE value is improved from 7.50% to 7.90%. The doping effect on the ETL for fullerene- and non-fullerene-based acceptors is also studied. It is analyzed that by varying the doping value of the ETL, the Jsc and PCE for non-fullerene-based acceptors is increased from 22.29 mA/cm² to 22.67 mA/cm², while the PCE is enhanced from 13.97% to 14.66%. It turns out that for the fullerene-based acceptor, Jsc is not effected, while the PCE is increased from 7.76% to 7.90%. From the whole analysis, it is concluded that by introducing suitable doping to the HTL and ETL, the output characteristics, i.e., the PCE, of fullerene- and non-fullerene- based solar cells are enhanced.

3.6. High-Temperature Analysis

Simulations are carried for solar cells working at room temperature. However, further investigations are required as solar panels operate at higher temperatures (up to 400 K) when installed outside. Thus, it is necessary to examine the operating temperature and minimize the sensitivity. Figure 6a,b represent the impact of temperature on output characteristics, such as the Voc and PCE, when the temperature is varied from 300 K to 400 K. It is seen that both parameters decrease with increasing temperature. Equation (4) shows how the performance indicators depends on temperature.

The reverse saturation current also depends on the temperature. From Equation (4), it is seen that as the working temperature increases, the saturation current also rises; as a result, the Voc is affected. This phenomenon is related to the intrinsic carrier concentration (ni), which is further related to the band gap (E_g), as it is seen in the mathematical equation [45]. However, in our simulation, the saturation current slightly decreased because of high-temperature effects on electron and hole mobility, leading to a decrease in the PCE.

$$n{i}^2=K1e^{-E_g/KT}$$

(9)

In Equation (9), K1 is constant.

The band gap (E_g) is inversely proportional to temperature (T) by the following mathematical relationship [45]:

$$E_g \left(T\right)=E_g \left(0\right)-{\displaystyle \frac{\alpha T^2}{T+\beta }}$$

(10)

Eg (0), $\alpha$, $\beta$ in Equation (10) refer to the band gap at zero Kelvin and material constants, respectively.

As the temperature rises, the electron–hole pairs’ excitation increases, due to which the band gap become unstable; as a result, the power conversion efficiency (PCE) decreases.

3.7. Efficiency Enhancement through Reflection Coating

Reflection coating is a technique to enhance the PCE of solar cell devices. By employing reflection coating on the back side of a solar cell, the absorption capacity is enhanced, which increases the PCE of the device [45,46,50]. As we know, photons of high wavelength are absorbed by the absorber; sometimes they pass without absorbing. So, in this simulation study, back-side reflection coating is applied on the back contact, and its value is changed from 0% to 50%. It is observed that the Jsc and PCE of both absorber layers (fullerene- and non-fullerene-based) are enhanced, which is shown in Figure 7a,b. The PCE of the non-fullerene-based acceptor device is improved by 1.72%, while that of the fullerene-based device is enhanced by 1.4%.

3.8. J-V Characteristics of Non-Fullerene- and Fullerene-Based Solar Cells

The final results of the simulations are shown in

Table 4. Photovoltaic parameters of fullerene- and non-fullerene-based solar cells.

Non Fullerene-Based Solar Cell	Voc [V]	Jsc [mA cm−2]	FF [%]	PCE [%]
Before reflection	1.09	22.67	58.87	14.66
After reflection	1.10	25.66	57.83	16.38
Fullerene-based solar cell:
Before reflection	1.28	8.40	73.36	7.90
After reflection	1.28	10.01	72.72	9.36

, while the J-V curves of both solar cells are shown in Figure 8a,b.

It is analyzed that after using the reflection coating, the PCE and JSC of non-fullerene based solar cells are improved from 14.66% to 16.38% and from 22.67 mA/cm² to 25.66 mA/cm², while in the case of fullerene-based solar cells, these parameters are improved from 7.90% to 9.36% and from 8.40 mA/cm² to 10.01 mA/cm², respectively. The increase in Jsc and PCE is caused by the generation of charge carriers and better light absorption.

3.9. Comparison of PV Parameters of FA- and NFA-Based Devices (from Simulation and Experimental Studies)

A comparative analysis was carried and the output characteristics from the simulation and experimental results are shown in

Table 5. Comparative analysis with previous work.

HTL/Active Layer	Voc	Jsc	FF	PCE	References
MoS2/PBDB-T:ITIC-OE	1.10	25.66	57.83	16.38	Simulation [This work]
MoS2/P3HT:PCBM	1.28	10.01	72.72	9.36	Simulation [This work]
PEDOT:PSS/PBDB-T:ITIC-OE	0.85	14.8	67	8.5	Experiment [47]
PEDOT:PSS/PBDB-T:ITIC-OE	0.84	14.61	66.8	8.3	Simulation [23]
MoS2/PBDB-T:ITIC	1.04	25.21	76.57	20.05	Simulation [26]
PEDOT:PSS/P3HT:PCBM	0.62	13.44	69.23	5.21	Simulation [20]
WO3/P3HT:PCBM	0,61	15.65	72.04	8.67	Simulation [20]

. It is analyzed that the proposed structures show promising results.

4. Conclusions

In this simulation study, we used the SCAPS-1D software to simulate fullerene- and non-fullerene-based solar cells, using MoS₂ as a 2D-HTL material. The simulation results demonstrate that by increasing the thickness of the active layer, the photovoltaic performance of the proposed devices is improved. It is analyzed that by increasing the thickness of NFA-based acceptors, the Jsc and PCE significantly improve, while in the case of FA-based acceptors, the Jsc and PCE are slightly improved compared to non-fullerene-based active layers. The reason may be connected with the band gap of the active layer, because the band gap of the FA-based active layer is greater than in the case of non-fullerene-based ones, which limits the Jsc and PCE of solar cells. The fill factor is decreased with the increase in thickness due to the uneven charge extraction of electrons and holes.

Defect concentration affects the stability and performance of solar cells due to recombination. By varying the defect density, it was demonstrated that defects shorten the diffusion length of charge carriers and increase the recombination in the active layer, which degrades the PCE of devices. The optimized defect density for NFA- and FA-based devices is 1 × 10¹² cm⁻³ and 1 × 10¹⁴ cm⁻³, respectively. Below this value, the performance of devices improved, but in practice, it is very difficult to design such low-defect-density devices. From the above simulation study, it is analyzed that solar cells are temperature-sensitive, and by increasing the temperature, the performance of both types of solar cells degrades, which is reflected as a smaller efficiency (PCE) value at higher temperatures. Doping is a technique to enhance photo-generated carriers. It is concluded that by increasing the dopant concentration in the HTL and ETL, the performance of solar cells improves. For the practical concerns, the doping values for the HTL and ETL are 2 × 10¹⁸ cm⁻³ and 9 × 10¹⁸ cm⁻³, respectively.

After optimizing the active layer thickness, defect density, interface defects, and temperature, then the reflection coating is applied on the back side of the solar cell. It is analyzed that the Jsc and PCE of these proposed solar cells are improved due to the generation of charge carriers, which improve the PCEs of both types of devices. Our simulation shows promising results, and the efficiency (PCE) value of NFA- and FA-based solar cells achieved 16.38% and 9.36%, respectively. So, overall, it is also concluded that two-dimension materials (like MoS₂) can be successfully used for the performance enhancement of organic BHJ solar cells and can be implemented for real-time photovoltaic applications, after investigations.