NUMERICAL MODELING AND ANALYSIS OF HTM-FREE HETEROJUNCTION SOLAR CELL USING SCAPS-1D

In this research paper, a HTM-free perovskite solar cell (PSC) structure with Titanium (TiO 2 ), methyl ammonium lead triiodide (CH 3 NH 3 PbI 3 ) and platinum (pt) as electron transport material (ETM), photon harvester and metal back contact is proposed. Solar Cell Capacitance Simulator (SCAPS-1D) program was used to implement the model and simulation. Effect of parameters such as thickness of ETM, thickness of absorber, doping concentration of ETM & absorber and electron affinity (EA) of ETM were investigated systematically. From the obtained results, it was found that the parameters affect the performance of the solar cell. When the thickness of ETM was varied from 0.02 to 0.10 m  . The results show that photovoltaic parameters decrease with the thickness increase. When the thickness of the absorber was varied from 0.1 to 1.0 m  , the optimized value was found at thickness of 0.4 m  . When the doping concentration of absorber and EMT were varied from 10 10 –10 17 cm -3 and from 10 15 –10 20 cm -3 , the highest values of PCEs were obtained at 10 16 cm -3 and 10 20 cm -3 for Absorber and ETM. Also when the EA was varied in the range of 3.7 eV to 4.5 eV, the optimized value was at 3.7 eV. Upon optimization of the above mentioned parameters, power conversion efficiency (PCE) was found to be 25.75 %, short circuit current density (J sc ) 23.25 mAcm -2 , open circuit voltage (V oc ) 1.24 V and fill factor (FF) 89.50 %. The optimized result shows an improvement of ~1.95 times in PCE, ~1.06 times in J sc , ~1.44 times in V oc and ~1.28 times in FF as compared to the initial device with the following parameters, PCE=13.22 %, J sc =21.96 mAcm − 2 , V oc =0.86 V

Recently, organic-inorganic metal halide perovskite solar cells have taken the renewable energy community by storm and subsequently gained attention of several world's researchers due to its high performance and low cost. Several advantages of perovskite absorber have made it a choice of candidate for application photovoltaic structures, among which include, tuned band gap, small exciton energy, excellent bipolar carrier transport, long electron-hole diffusion, and amazingly high tolerance to defects [1][2][3][4][5][6][7]. The properties exhibited by this material in solar cells results to enhanced power conversion efficiency from 3.9 % [8] to over 25 % [9]. However, some draw backs such as instability, electron transport resistance between TiO2 and perovskite absorber and the use of costly hole transport material (HTM) (such as spiro-Omeotad) has prevented its outdoor usage. Therefore, the use of readily available and stable materials having high hole mobility with simple route of synthesis is desirable [10][11][12].
A report on perovskite solar cell without HTM was demonstrated firstly by Etgar et al. [13], where the perovskite absorber functions as a transporter of hole and harvester of photon energy simultaneously and results to a PCE of 5%. The results demonstrate simplicity and high reduction in cost of fabrication and a relatively improved stability as a result of HTM elimination. Also, in 2014, Li et al. [14] followed the same route by replacing metal back contact with laminated carbon nanotubes (CNTs) to collect holes from absorber and block electrons from ETM and results to a PCE of 6.87%. Eli et al. [9] developed a PSC without HTM with elcocarb as metal back contact to collect hole and realized a PCE of 3.80 %.
Recently a PCE of 10.95% with about 95% stability of their initial efficiency after being exposed to air (relative humidity of 25-35%) for 20 days, was demonstrated by Zhang et al [15]. In a simulation studies by Lin et al [16], a PCE of 15.02% was obtained with a structure without HTM and ZnO as ETM. Theoretical studies carried out by Wang and group [17] shows that the careful selection of thickness of the absorber and p-type doping were crucial to the PCE of the HTM-free PSCs.
The HTM-free PSC is a simple and promising way to realize good PCE but some factors deter its practical use. PSC devices mostly make use of costly gold (Au) as a back contact. The Au metal contact is not only expensive but the process of synthesis and development require the use of high-technology (such as high-vacuum evaporation technique). Platinum metal, with the work function of 5.93 [18], has been applied in perovskite solar cells with HTM and without HTM, but the PV parameters were of poor values for HTM (Jsc=7.17 mAcm -2 , Voc=0.69, FF=62.62 and PCE=3.08 %). And for the device without HTM, the PV parameters are Jsc=20.58 mAcm -2 , Voc=1.006, FF=71.07 and PCE=14.72 %.
To realize higher photovoltaic parameters and proper optimized architecture in HTM-free PSC, systematic understanding of the operational mechanism needs to be uncovered. Yet till now, numerical modeling and simulation of the HTM-free PSCs with platinum metal contact and TiO 2 ETM has seldom been reported. Platinum metal has a high work function and good chemical stability. In view of that, this research paper, attempts to present a detailed numerical modeling and simulation of HTM-free PSCs based on solar cell capacitance simulator (SCAPS) software to systematically study the influence of some material (such as thickness of ETM, thickness of absorber layer, doping concentration of

EEJP. 2 (2021)
Danladi Eli, Alhassan Shuaibu et al ETM, doping concentration of absorber layer, electron affinity of ETM etc), with the goal of uncovering the hidden mechanism for PCE improvement.

METHOD AND METHODOLOGY
The nature of the defect is set as Gaussian and defect density is set as 1×10 18 cm −3 [19,20]. Table 1 shows the defect parameters which are used in the simulation. Basic parameters for each material used in the simulation are summarized in Table 2. Thermal velocities of hole and electron are selected as 10 7 cms −1 [19][20][21][22]. The optical reflectance is considered to be zero at the surface and at each interface [19]. Parameters are optimized in the study by using control variable method. The initial total defect density of the absorber layer is assumed to be 2.5×10 13 cm −3 . Neutral Gaussian distribution defect is selected in the absorber layer and characteristic energy is set to be 0.1 eV [19]. One defect interface is inserted for carrier recombination. The interface defect layer (IDL) is introduced in the TiO 2 /CH 3 NH 3 PbI 3 interface to investigate the influence of interfacial recombination on the photovoltaic performance. The work functions of the front contact and back contact are 4.40 eV and 5.93 eV respectively [18,23]. A working temperature of 300K, solar spectrum AM1.5 and a Scanning voltage of 0-1.3 V were used for all simulations.

Structure of PSC in the simulation and Energy level diagram of HTM free PSC device
The modeled PSC and band structure of the HTM free perovskite solar cell obtained with simulated parameters in Tables 1 and 2   The value of ∆Ec prevents the flow of electron charge carrier from the electron transport layer to perovskite layer to the pt metal contact so as to avoid quenching in the perovskite layer. While the large value of ∆Ev denies the flow of holes to the platinum-back contact to prevent their recombination with the electrons in the perovskite layer. These values Numerical Modeling and Analysis of HTM-Free Heterojunction Solar Cell...

EEJP. 2 (2021)
can be seen important as they encouraged collection of charge carriers which results to higher photovoltaic performance in PSCs. As such, from the result of the band gap structure, TiO 2 and CH 3 NH 3 PbI 3 can form a p-n junction when combined together to be applied in photovoltaics.

Performance parameters from initial simulation
The J-V characteristics of the reference modeled PSC device under illumination and in the dark is shown in figure 2(a). Under illumination, a J sc of 21.96 mAcm −2 , Voc of 0.86 V, FF of 69.94 %, and PCE of 13.22 % are obtained. The V oc simulated in this studies agrees with V oc (0.85 V) in experimental work demonstrated by a group of researchers [9]. However, the values of PCE, FF and J sc are higher than ones obtained from experimental research work, this may be due to the lesser series resistance arising from pt metal contact or FTO and the optical reflectance is considered to be zero at the surface and at each interface during our simulation [25]. As illustrated in Figure 2(a), Under the dark condition, there is no current flowing thereby behaving as a diode resulting to an extreme minimum value of J sc when the voltage is less than the knee voltage (0.6 V) and decreases gradually when the voltage further increases [25]. This results to rectifying characteristics and this rectifying behavior is a feature of photovoltaic devices and is a consequence of the asymmetric junction needed to separate charges. In the quantum efficiency (QE) of the device shown in figure 2 (b) which is within the wavelength of 300 nm and 900 nm has maximum attained value of 90 % at 550 nm. Optical absorption edge is red shifted to 800 nm which corresponds to a band gap of 1.55 eV in CH 3 NH 3 PbI 3 . The QE sweeps across the whole visible spectrum which to an extent agrees with experimental work [9]. Jsc, Voc and FF decrease with the thickness of ETM. The slight decrease in the photovoltaic properties is due to fractional absorption of incident light by the TiO 2 layer and the bulk/surface recombination at the interface which result to lesser electron and hole pairs extraction [26]. The decrease in FF is connected to the increase in series resistance. Figure 3(b) displays the QE of the perovskite solar cell as a function of wavelength within the range of 300-900 nm with varied ETM layer thickness. It can be seen that the QE reaches a maximum value in the wavelength range of 380−570 nm and gradually decreases at longer wavelengths until 800 nm, which corresponds to its absorption spectrum demonstrated in figure 3(b). Table 3 shows the photovoltaic parameters obtained during the simulation.

Effect of thickness of absorber layer
The influence of thickness of absorber on the solar cell parameters; V OC , J SC , FF and PCE is shown in figure 4 (a). The J-V and QE of the varied absorber thickness is shown in Figure 4 (a) and (b).
As shown in Figure 4 (a), Jsc increases from 12.33 to 22.36 mAcm -2 with thickness increase from 0.1 to 0.6 which is due to the increase in carrier generation and dissociation, then starts decreasing from 0.7 to 1.0  m which is attributed to high recombination rate within the range of the thickness. FF decreases slightly with thickness increase in the perovskite layer. Table 4 shows the photovoltaic parameters obtained during the simulation. The PCE increase with increase in layer thickness from 0.1 to 0.4  m due to the production of new charge carriers. However, PCE decreases from thickness of 0.5 to 1.0  m due to lesser electron and hole pairs extraction rate that leads to recombination process [27].
Numerical Modeling and Analysis of HTM-Free Heterojunction Solar Cell... Figure 4 (c) exhibits the spectral response of the PSCs as a function of wavelength with varied CH 3 NH 3 PbI 3 layer thickness within range of 300 to 900 nm. The QE first increases rapidly with the CH 3 NH 3 PbI 3 thickness increasing from 0.1 to 0.4  m, and the QE increase slightly after the thickness is greater than 0.4  m, which shows that 0.4  m thickness of CH 3

Effect of doping concentration (NA) of absorber layer
Doping is the process of introducing impurities in absorber layer. The effect of doping concentration on the performance of perovskite solar cell is studied by choosing the values of N A in the range of 10 10 -10 17 cm -3 while keeping N D for ETM at 10 17 cm -3 . Table 5 gives the performance parameters of PSCs with various values of doping concentration. The highest value of PCE was obtained with doping concentration of 10 16 cm −3 which shows value of 14.89 %. The highest value of fill factor was also observed at the same N A . The PCE and the FF remained constant with increase in doping concentration from 10 10 -10 13 cm -3 and increases with doping concentration from 10 13 -10 16 cm -3 . Beyond the values, a decrease in PCE and FF was noticed. The obtained results show that charge carriers are transported and collected optimally at the same irradiance when N A of the CH 3 NH 3 PbI 3 is 1×10 16 cm −3 . The J sc and Voc remained constant with increasing N A from 10 10 -10 13 cm −3 , while beyond 10 13 cm −3 , J sc increases with increasing N A and V oc decreases with increasing N A . The ability of the photo-generated carriers is weak, resulting to a reduced Voc which caused a full depletion and strong electric field in the absorber layer.
The QE with respect to wavelength is as shown in Figure 5(b) with varied concentration from 10 10 -10 16 cm −3 . It can be seen that the QE rapidly increase within the wavelength range of 300 -380 nm and thereby maintained a constant until at 630 nm before it decreases rapidly to 800 nm from 10 10 -10 15 cm −3 . From 10 15 -10 13 cm −3 a gradual decrease from the cutout of 390 nm until 800 nm was observed. The result shows that the solar cell functions effectively within the visible region.

Effect of doping concentration (ND) of ETM
The effect of doping concentration on the performance of perovskite solar cells is examined by varying the values of N D in the range of 10 15 -10 20 cm -3 while keeping N A for absorber at 10 13 cm -3 . Figure 6 (a) and (b) shows the J-V curves of PSC with different values of doping concentration in ETM and QE with different values of doping concentration in ETM layer. When the doping concentration is varied from 10 15 -10 20 cm -3 , it was depicted from Table 6 that, the PCE increased from 12.52 % to 16.62 %. The increase in PCE is as a result of reduction in series resistance due to increase in optical conductivity of the ETM. Hence, the doping concentration N D is set at 10 20 cm -3 . Similarly, there was increase in other photovoltaic parameters (J sc , V oc and FF) with increase in doping concentration values. The optimized values of the performance parameters are PCE=16.62 %, Jsc=22.24 mAcm -2 , Voc=1.04 V and FF=72.18%.
The increase in the photovoltaic parameters with increasing N D could be explained as follows: Auger recombination rate decreases with doping density below 10 20 cm −3 . It is seen that the quenching losses decreases when N D is below 10 20 cm −3 . We therefore speculate that, optimum doping concentration of ETM enhances the Voc and Jsc which subsequently results to higher PCE.

Influence of electron affinity of ETM
The effect of electron affinity (EA) on the performance of perovskite solar cell is examined by varying the values of EA in the range of 3.7 eV to 4.5 eV. Figure 7 Table 7. It is now evident from our findings that proper selection ETM with good EA can reduce quenching losses in PSCs. Table 7 gives the performance parameters of PSCs with various values of EA. device that was not optimized. The behaviour of the JV curve in the dark and under illumination is as shown in figure  8(a). Experimental work of HTM free PSC published by other researchers are compared with the simulated results and summarized in Table 8(b).
In the experimental works, PCEs of 3.80 and 4.2 % were achieved for HTM free PSCs with TiO 2 as ETM. The photovoltaic parameters could be improved further to realize the high photovoltaic values achieved in the simulation. This could be realized by improving the film quality of both the absorber and ETM and also consider proper doping of the absorber and ETM in order to realize good electron density.
The energy diagram of the optimized device is as shown in figure 8(c). From the band structure, the conduction and valence band offset at the TiO 2 /CH 3 NH 3 PbI 3 interface were reduced to 0.06 eV and 1.98 eV, which can be considered beneficial for the flow of photo-excited charge carriers to the front electrode and back-metal contact in order to avoid their recombination and quenching losses. The quantum efficiency also shows stronger absorber in the visible region as compared to the device without optimization as shown in figure 8(b).

CONCLUSION
In this work, the HTM free PSC was investigated systematically using Solar Cell Capacitance Simulator (SCAPS-1D) program. The photovoltaic performance of the modeled device with various CH 3 NH 3 PbI 3 thicknesses, ETM thicknesses, ETM electron affinities, ETM doping concentrations and CH 3 NH 3 PbI 3 doping concentrations, has been analyzed. From the obtained results, it is found that the parameters affect the performance of the solar cell. The overall PCE, FF, Jsc, and Voc, of 25.75 %, 89.50 %, 23.25 mAcm -2 , and 1.24 V respectively were obtained by using all optimised parameters.