Experimental investigation of additive free-low-cost vinyl triarylamines based hole transport material for FAPbI3-based perovskite solar cells to enhance efficiency and stability

Perovskite-based solar cells have drawn a lot of attention recently because they possess many desirable qualities, including strong photon absorption, large carrier lifetime, ambipolar transmission, and low exciton binding energy. With continual optimization of each functional layer, particularly the active layer and hole transporting layer, the power conversion efficiency (PCE) of perovskite materials has reached over 25%. Spiro-OMeTAD is a widely utilized hole transport material (HTM) for efficient solar cell operation. To improve conductivity, this material is often doped with additives such as 4-tert-butylpyridine (TBP) or bis(trifluoromethane)sulfonimide lithium salt (Li-TFSI). Unfortunately, these additives can weaken the perovskite layer and reduce device stability. In this work, we enhanced the efficiency as well as stability of formamidinium-based perovskite using additive-free, cost-effective HTM based on vinyl triarylamines developed by the Tokyo chemical industry. We have deposited vinyl triarylamines-based HTM on both FAPbI3 and MAPbI3 perovskite. To compare the results, we have deposited traditional additive-based as well as additive free Spiro-OMeTAD on FAPbI3 perovskite. Results are encouraging as the FAPbI3-based device showed a decent power conversion efficiency of 16.86%, which is higher than when the same HTM is deposited on the MAPbI3-based device and comparable with doped Spiro-OMeTAD and much higher than undoped Spiro-OMeTAD based HTM deposited on FAPbI3 perovskite. Enhancement in device performance is attributed to better hole mobility and favourable energy band positioning of vinyl triarylamines based hole transport layer w.r.t FAPbI3 perovskite. The PCE of a FAPbI3-based device using the suggested HTM (SHTM) suffers only a 12% decrease while following the maximum power point for 1800 h in ambient air.


Introduction
Perovskite solar cells have garnered a substantial amount of scientific attention since their power conversion efficiencies are already superior to a number of known thin-film photovoltaic technologies. Common perovskite materials used in solar cells include methylammonium (MA) lead iodide (MAPbI 3 ), formamidinium (FA) lead iodide (FAPbI 3 ), and cesium lead triiodide (CsPbI 3 ). FAPbI 3 has the greatest structural symmetry, is more thermally stable than MAPbI 3 , and has a bandgap closest to the Shockley-Queisser limit (1.34 eV) [1][2][3]. In light of this, PSCs based on FA have rapidly risen in prominence as a promising new field of study for the perovskite research fraternity. Despite the fact that FAPbI 3 has great potential to simultaneously attain high power conversion efficiency and strong stability, it is nevertheless struggling with its own set of issues. The lack of phase stability is by far the most problematic of these issues [4,5]. The hexagonal yellow-colored δ phase of FAPbI 3 is thermodynamically stable at room temperature. It is not possible to utilize it directly as a light-absorbing material because it has a broad bandgap, which is around 2.4 eV. For solar cell applications, a low bandgap black cubic phase (α-FAPbI 3 ) is required, but it is obtained at a high temperature (above 400 K) [6,7]. This has drawbacks in its production because of the high formation temperature. When exposed to open air or in polar solvents, α-FAPbI 3 will revert back to the inactive δ-FAPbI 3 phase even after being produced at high temperatures [8,9]. Because of this poor transformation process, the perovskite and its related device deteriorate irreversibly due to changes in composition and the creation of defects. To slow down the phase change in FAPbI 3 and enhance the film's shape, several research groups now use a technique based on a combination of cations and halide ions. In addition, a device containing MA runs the risk of experiencing long-term thermal instability; an I/Br mixture-based PSC, when exposed to light for an extended period of time, will experience severe ion migration; and perovskites based on FA and CS are difficult to achieve high power conversion efficiency and are also susceptible to phase change [10][11][12][13][14]. As a result, the development of pure FAPbI 3 has emerged as a significant approach toward the achievement of high-efficiency and stable solar cells. Other important factors that affect how well the device works as a whole are choosing the right electron transport and hole transport materials.
PCS is designed to be in the form of a sandwich structure, which is comprised of an active layer that is responsible for the absorption of light, an n-type electron transport layer (ETL), and a p-type hole transport layer (HTL). These transporting layers are responsible for transporting a hole and an electron to their corresponding electrodes. As a result, it is essential to make use of both the ETL and HTL in order to separate the light-generated carriers in the active layer in a timely manner. The continual use of conduction-improving chemicals in the hole transporting material is an important aspect of efficient and stable perovskite solar applications. Lithium, antimony-based compound (LiTFSI), and tert-butylpyridine (TBP) are frequently added to the widely used hole-transporting materials Spiro-OMeTAD and poly(triarylamine) (PTAA) [15][16][17]. Because of its hygroscopic character, the dopant LiTFSI, which is extensively used in Spiro-OMeTAD, may help enhance the infiltration of moisture from the surrounding environment into the perovskite layer [18][19][20][21][22]. It has also been demonstrated that the chemical interactions between Spiro-OMeTAD + and Ithat take place at the interface between perovskite and Spiro-OMeTAD may trigger deterioration [23,24]. It has been shown that the contact between Spiro-OMeTAD and the top electrode (Ag or Au) is also responsible for the deterioration of PSCs in a number of different experiments [25]. In industrial applications where the device's stability is critical, additive-free hole transporting materials are therefore essential. In addition, traditional HTMs have a price tag that is too high for the mass manufacture of low-cost PSCs. To make PSCs on a commercial scale, it is very important to come up with HTMs that don't need any additives and are cheap.

Related work
Several research groups have reported the synthesis of low-cost HTMs on FAPbI 3 perovskite, but these often required additional, time-consuming purification processes unsuitable for large-scale production, such as column chromatography, before the isolated product could be used. Boix and his group first fabricated a FAPbI 3 -based perovskite solar cell for using Spiro-OMeTAD as an HTL and calculated key parameters like band gap, efficiency, and absorption [9]. The power conversion efficiency of the top-performing mesoporous n-i-pbased device exceeded 4%. The morphological instability of the photosensitive black phase of FAPbI 3 was another issue that was brought up by this group. Chang and colleagues demonstrated a descent efficiency of 16.83% using PTAA as an HTL in a FAPbI 3 -based device [26]. They also studied the effects of the CsCl additive on the morphological and optoelectronic properties of FAPbI 3 -based perovskite films. The major limitation of this study was the poor stability of the pure FAPbI 3 -based device. They improved the stability by using a CsCl additive in the FAPbI 3 precursor solution, as the final device retained 55% PCE after 500 h. In the year 2020, Hidetaka Nishimura and his co-workers from Tokyo's chemical industry synthesized three additive-free vinyl triarylamines-based hole transport materials and deposited these HTMs on MAPbI 3 perovskite [27]. They named these novel materials TOP-HTM1, TOP-HTM2, and TOP-HTM3. These HTMs are available for commercial use. This group shows a descent efficiency of 16.6% for additive-free TOP-HTM3 material [27]. Our findings suggest that most of the research groups enhanced the efficiency and stability by controlling the morphology of FAPbI 3 film by using different techniques like defect passivation, dimensional regulation, strain engineering, etc [28][29][30][31]. We have also summarized a few works in table 1.
In this study, we used a low-cost additive-free hole transporting material (E, E, E)-1,2,4,5-Tetrakis [4-[bis(4methoxyphenyl)amino] styryl]benzene] (SHTM) developed by Tokyo Chemical Industry for FAPbI 3 -based perovskite. Trans-vinylene units are used to insert dimethoxytriphenylamine units at the 1,2,4,5-positions of a benzene core to create a X-shaped compound in this suggested hole-transporting material (SHTM). Perovskite solar cells have been created using a conventional device architecture of glass/fluorine-doped tin oxide (FTO)/(h-TiO 2 )/(mp-TiO 2 )/FAPbI 3 or MAPbI 3 /SHTM/Au. PSC devices were also manufactured and analyzed using Spiro-OMeTAD as the HTM so that we could examine how well our novel HTMs performed in comparison to the performance of traditional Spiro-OMeTAD based HTM. Under identical device fabrication circumstances, it was discovered that Spiro-OMeTAD-based PCS provided a PCE of 15.11%. The SHTM and Spiro-OMeTAD are both good in terms of power conversion efficiency. However, the SHTM-based device is much more stable than the traditional HTM based on Spiro-OMeTAD.

Device fabrication procedure
To begin, an F-doped SnO 2 (FTO, Pilkington, TEC8) substrate measuring 20 mm by 30 mm (20 × 30) was etched using zinc powder and 2 M of hydrochloric acid. After that, an ultrasonic device was used to clean the substrate in stages using soap, distilled water, ethanol, acetone, and IPA for a total of ten minutes. The holeblocking TiO 2 (h-TiO 2 ) layer was successfully fabricated after the h-TiO 2 precursor was dropped on the FTO and spin-coated at 4000 rpm for 30 seconds. This was accomplished by annealing the layer at 450°C for an hour. After the h-TiO 2 layers had reached room temperature (RT), 80 ml of the mp-TiO 2 precursor was spin-coated onto the h-TiO 2 layer at 4500 rpm for 30 seconds. This was followed by an annealing process at 500°C for one hour in order to generate the mp-TiO 2 scaffold. The 3D FAPbI 3 perovskite layer was spin-coated over the mp-TiO 2 substrate at 1000 rpm for 4s and 6000 rpm for 25 s, then 200 μl of CB anti-solvent was immediately dripped onto the intermediate perovskite layer after 8 s in the second step of the spin-coat program. The FAPbI 3 films were then heated for 10 min at 155°C. HTM was deposited on perovskite at 4000 rpm for 30 seconds before being annealed at 70°C for 30 min In order to finish the construction of the PSC, a layer of gold electrodes 80 nm thick was thermally evaporated on top of the HTM at a rate of 1 A°/s while the area was subjected to a high degree of vacuum. Following completion of the process, the solar cell was stored in a dry environment for a period of twelve hours in order to encourage the oxidation of HTL. For depositing MAPbI 3 and Spiro-OMeTAD, we have adopted a similar process as reported in our recent work [35]. The stepwise process of synthesizing different layers is shown in figure 1.

Measurements
Recording the absorbance spectra of FA and MA-based perovskite layers was accomplished with the assistance of a Shimadzu UV-2700600i UV-Visible spectrophotometer [36]. For the purpose of recording photoluminescence (PL) responses, an FL-Ar-2015 spectrometer was used, and the exciting wavelength was set at 450 nm. A Mira3-XMU, TESCAN FE-SEM device was used to evaluate the surface morphological characteristics of FAPbI 3 layer. The photoactive alpha phase of the FAPbI 3 was identified by XRD measurement using a D2 phaser-based Brucker XRD spectrometer [37]. With the use of a digital 2400 Keithley source meter, the current density-voltage (J-V) properties of solar cells were measured and recorded when the cells were exposed to an AM 1.5 G simulation of Sunshine. Through the use of a Q Test Station 2000ADI system, IPCE spectra of fabricated solar cells were analyzed. A Solartron 1260 potentiostat/galvanostat was employed to record electrochemical impedance spectroscopy (EIS) of perovskite solar cells at a voltage bias of 0.8 V in the dark, in the frequency range of 0.1 Hz to 1 MHz.

Result and discussion
The proposed FAPbI 3 /MAPbI 3 -based PSC devices with new, additive-free vinyl triarylamines-based HTM (SHTM) and conventional Spiro-OMeTAD HTM are shown schematically in figure 2. In order to do a comparative analysis of various solar parameters, we have fabricated and characterized four different device configurations, whose layered arrangement is given below.
Step wise procedure to fabricate the proposed perovskite solar cells.
Configuration 4 has been already reported by a team led by Hidetaka Nishimura from Tokyo Chemical Industry, but we have again fabricated it as we want to do the comparative analysis of the above device configurations fabricated in similar environmental conditions. After the device was built, we used different characterization methods to explore its structural, optical, and electrical properties as well as its morphological features.

UV-Visible absorption spectroscopy and steady-state PL measurement
The UV-vis absorption profile of the SHTM was determined by measuring it in a solution of 1,1,2,2tetrachloroethane (TeCA) and in a film spin-coated on a quartz substrate, both of which are necessary features of the synthesized molecules as HTMs ( figure 3). The SHTM molecule displays strong UV and short-wavelength visible absorption in TeCA solution, which is connected to a π-π * transition. The maximum absorption peak of the SHTM is achieved at 415 nm, which is shifted to a longer wavelength compared to the traditional Spiro-OMeTAD (374 nm) [38]. Therefore, the primary absorption peak of the SHTM is sufficiently separated from the wavelength at which perovskite has the highest possible level of absorption. This suggests that the additive-free  HTM is transparent and may be acceptable for use as an HTM. The dotted lines in figure 3 depict the spin-coated film. The dotted plot shows that the absorption edge of the suggested HTM was pushed to longer wavelengths. This suggests that there was some level of intermolecular interaction. We have not performed the other necessary analyses (thermo gravimetric analysis (TGA), cyclic voltammetry, etc) as our experimental results also gave almost similar results as calculated by researchers from Tokyo Chemical Industry [27]. Figure S1 displays the UV-visible spectra (left side) and the matching PL intensity (right side) ( Figure S2) for the FA and MA-based perovskite films. These two perovskite materials were spin-coated on the same substrate used to manufacture the devices. The MA-based perovskite film showed the absorption onset at 780.4 nm and the FA-based perovskite showed the absorption onset at 810 nm, which is well in line with the reported onset for MAPbI 3 and FAPbI 3 -based perovskite films, respectively.
The absorption spectra cover the whole visible spectrum for both types of perovskite, as indicated in the UVvis plot. Photoluminescence measurements for both types of perovskite were collected on an FL-Ar-2015 spectrometer that was outfitted with a microsecond xenon flash lamp at an excitation wavelength of 450 nm. With a peak wavelength of 810 nm and 780 nm for the FA-and MA-based perovskite, respectively, strong PL intensity and well-defined spectra were observed. These wavelengths correspond to the absorption onset wavelength determined from the UV-vis spectrum of the FA-and MA-based perovskite films. The PL intensity of FA based perovskite is higher as compared to MA based perovskite as shown in figure S2. Due to the high intensity of the PL signal, it can be deduced that the amount of non-radiative recombination that takes place in the FA-based perovskite is much lower than that which takes place in the MA-based perovskite. PL intensity also gives an indication of the crystallinity of synthesized material. Higher the intensity higher the material crystallization and vice versa. The associated Tauc curves that were used to quantify the band-gap energy (E g ) of MAPbI 3 and FAPbI 3 layers that were produced on the same substrate are shown in figures 4(a)-(b).
The optical bandgap energy calculated from the Tauc plot for MA and FA-based perovskites was found to be 1.58 eV and 1.53 eV, respectively, which is in line with the value obtained from UV-vis spectroscopy (E g = 1240/ λ). The bandgap that was obtained for both kinds of perovskite is greater than the value that was stated (1.55 eV for MAPbI 3 and 1.48 eV for FAPbI 3 ) [9,39]. This energy gap, on the other hand, for halide perovskite falls somewhere in the range of 1.5-2.3 eV, depending on the amount of halide present and the size of the crystallite.

Photoluminescence (PL) quenching analysis
Steady-state PL quenching was investigated for both pure perovskite and perovskites/HTMs layers to understand more about the interfacial charge separation. The PL peak of the pristine MA and FA-based perovskite films was observed at 780 and 810 nm, respectively. The intensity of the PL peak was decreased by 32% when SHTMs were spin-coated on MA-based perovskite layers ( figure 5(a)). Whereas the intensity of the PL peak was decreased to 10.8% and 12% when proposed SHTMs and traditional Spiro-OMeTAD were spincoated on FA-based perovskite layers ( figure 5(b)). High PL quenching in SHTM and traditional Spiro-OMeTAD clearly indicate effective hole extraction from FA-based perovskite. PL quenching in MA-based perovskite is lower as compared to FA-based perovskite; this will lead to improved J sc and fill factor in FA-based perovskite.

XRD analysis of FA and MA based perovskite layer
For the purpose of investigating the crystallinity of MA and FA-based perovskites, the x-ray diffraction (XRD) patterns of the samples were acquired. Figure S3a shows the XRD pattern of MA-based perovskite layers. There are peaks at around 14.02, 24.39, and 28.26, corresponding to the crystal planes (110), (202), and (220), respectively. This is in reference to the tetragonal phase of MA-based perovskite [40,41]. The additional peak marked by star ( * ) in figure S3a corresponds to a peak position of PbI 2 . The PbI 2 -rich area in the perovskite layer should have centres for charge trapping in the bulk of the perovskite layer. This would result in an increase in the recombination of generated charge carriers and a reduction in the amount of charge that may be transferred across interfaces in a PSC device. The MA-based perovskite layer that was made has the desired (110)-oriented crystal structure, which makes it good for use in solar applications. Figure S3b displays the x-ray diffraction patterns obtained from the FA-based perovskite layers. XRD patterns show two primary peaks at 2θ = 14.06°a nd 28.20°, which correspond to the (001) and (002) plane orientations of the α-FAPbI 3 phase [42]. When compared to the MA-based perovskite, the FA-based perovskite exhibits a change in the position of the diffraction peaks toward lower degrees. This is because the more massive FA cation has replaced the more diminutive MA cation in the structure. In comparison to MAPbI 3 perovskite, the peaks in the XRD plots of the FAPbI 3 perovskites were more intense and crisp, indicating that the films were more successfully crystallized. We have also calculated the full width at half maximum (FWHM) of (110) and (001) XRD peaks of MA and FAbased perovskite. Fitted values of 0.2022 and 0.1038 were obtained for the MA and FA-based perovskite, respectively. It indicates that FAPbI 3 has a larger grain size compared to MAPbI 3 . Figures 6(a) and (b) show the FESEM images of the MAPbI3 and FAPbI3 layers, which can be used to look at their micromorphology. From the FSEM image, it is concluded that FAPbI 3 has a large and uniform grain size as compared to MAPbI 3 perovskite. The FESEM images of both types of perovskite (MA and FA) show how the morphology of the films affects the movement of charge carriers. Larger grains in FAPbI 3 perovskite (figure 6(d)) show a preferential growth direction, whereas smaller grains developed in MAPbI 3 are more haphazardly relative to the target surface of the substrate ( figure 6(c)). Charges appear to be spatially restricted inside a grain for the smaller grains, and grain borders may operate as an obstacle for charge transmission. In FAPbI 3 , the perovskite grains get bigger and more aligned, grain-to-grain connections get better, and charges can move across the boundaries of aligned grains. This makes the separation of charges at the interface more even.

J-V Analysis of different configurations
We have used reverse J-V sweep under the AM 1.5 Sunlight simulator to explore the solar characteristics of our three different device setups. The comparison of J-V graphs for different configurations is shown in figure 7(a), and table 2 shows the compiled cell parameters. We also constructed and examined the PSCs utilizing the traditional doped Spiro-OMeTAD and undoped Spiro-OMeTAD HTM for FA-based perovskite in order to compare the performance of the proposed additive-free suggested HTM (SHTM). In total, we have fabricated four cells, one for MA-based perovskite using SHTM and three FA-based perovskites using SHTM and doped Spiro-OMeTAD and undoped Spiro-OMeTAD HTM, respectively. Spiro-OMeTAD was the sole component that was dissolved in a chlorobenzene solution that also included the dopants tBP and Li-TFSI before being placed onto the perovskite layer. In contrast, there were no doping agents used in the development of the proposed SHTM. A somewhat inferior performance is shown when the proposed dopant-free SHTM is used as an HTM on MA-based perovskite devices, with a fill factor of just 71.1% and a PCE of less than 16%. This result is almost in agreement with the previously reported value by Nishimura and his co-worker [27].  Inferior photovoltaic performance could be due to poor PL quenching of SHTM with MA-based perovskite. Low PL quenching indicates poor hole extraction from the perovskite layer, which in turn reduces the overall performance parameters. The lower value of the fill factor could be due to hole trapping in iodide-rich regions formed due to phase segregation acting as defect states. The proposed dopant-free SHTM deposited on FAPbI 3 perovskite exhibits performance that is on par with that of the reference cell based on doped and undoped Spiro-OMeTAD HTM, which displays the greatest performance with a fill factor of 77.7%, V oc of 1.09 V, J sc of 20.13 mA cm −2 and PCE of 16.86%. These findings are in excellent accord with our PL quenching results and suggest that hole extraction and hole injection from the valence band of FAPbI 3 perovskite into the HOMO of SHTM are considerably more efficient than those of doped and undoped Spiro-OMeTAD. The fabricated device based on undoped Spiro-OMeTAD exhibited a substantial reduction in the FF, resulting in a much lower PCE of just 10.44% under the same fabrication circumstances, which may be driven in large part by the reduced hole mobility of undoped Spiro-OMeTAD compared with that of doped one. Figure 7(b) shows the IPCE spectrum of the perovskite devices (MAPbI 3 and FAPbI 3 ), including all HTMs for different device configurations. All devices exhibit good performance over 70% from 350 to 750 nm, which is intriguing and implies that our proposed additive-free SHTM may be a competitive alternative to additive-based Spiro-OMeTAD in high-performance PSCs. The measured J sc agrees with the fact that the IPCE of doped Spiro-OMeTAD is only a little bit lower than the SHTM.

Interfacial charge transfer analysis
To understand the charge transfer process at the HTL/perovskite interface, electrochemical impedance spectroscopy (EIS) was used to characterize perovskite solar cells with different HTLs. EIS can be used to measure the charge transfer resistance (R CT ) of the interface, which is an important parameter for understanding the charge transport properties of the solar cell. . EIS can be used to measure the impedance of a sample as a function of frequency, and the R CT can be calculated from the impedance data. The R CT can be measured by fitting a two-resistor model to the impedance data. By measuring the R CT , we can gain insight into the charge transport properties of the perovskite/HTM interface. Figure 8 displays Nyquist plots of the fabricated cells that were measured in the dark. These plots reveal that the cells have semicircles, which are correlated to the resistance of charge transfer at the perovskite/HTM interface. As illustrated in the inset of figure 8, the charge transfers resistance (R CT ) values were fitted using a conventional equivalent circuit, and the resulting findings are presented in table 3. The findings are consistent with the performance of the device as determined by its J-V characteristics tabulated in table 2. The FAPbI 3 -based perovskite solar cell fabricated using the SHTM shows the lowest R CT , indicating the most efficient charge transfer at the perovskite/HTM interface. This finding is a complete agreement with the efficiency obtained from SHTM-based cells.

Stability analysis of different configurations
One of the most significant obstacles to overcome on the path to commercialization is the stability of PSCs. As a result, we performed a stability study and evaluated the cell's performance in two environments with relative humidity (RH) levels ranging from 40%-50% ( figure 9(b)) and less than 20% ( figure 9(a)).
Experimental results showed that dopant-free SHTM-based devices (both MAPbI 3 and FAPbI 3 ) are more stable than dopant-based traditional Spiro-OMeTAD-based devices. The FAPbI 3 device based on the SHTM as a hole transport layer retains 88% of the initial efficiency, while the Spiro-OMeTAD-based device deteriorates more quickly and only retains 75% of the initial performance after 1800 h when the relative humidity level is kept below 20%. The superior stability of the device using the SHTM could be due to the relatively high  decomposition temperature (400°C) and the absence of additives, which might cause less deterioration of holetransporting and perovskite layers as compared to the additive-based Spiro-OMeTAD-based device. The Au electrode diffusion into the Spiro-OMeTAD layer reduces glass transition temperature at high-temperature operation, which is one of the reasons for the poor device stability of Spiro-OMeTAD-based devices. Similar stability trends were observed when stability tests were conducted at higher RH values (40%-50%). A device composed of FTO/h-TiO 2 /mp-TiO 2 /FAPbI 3 /SHTM/Au retains 83% of the initial performance after 1800 h, which is much greater than doped Spiro-OMeTAD-based device which retain only 69% of the initial performance. It is very important to test the stability of proposed configurations under illumination in ambient conditions since solar cells must work under illumination, and perovskite materials and devices can break down in different ways in these conditions. Surprisingly, after 400 h of continuous illumination at ambient temperature, the suggested devices without any encapsulation only exhibit 10% PCE degradation, whereas devices based on doped Spiro-OMeTAD exhibit 31% PCE degradation ( figure 9(c)). The astounding performance of the proposed device is primarily because of its superior structural qualities, which include compact shape, high PL quenching, decreased grain boundaries, and greater inherent stability of the FAPbI 3 -based Perovskite.

Conclusion
In conclusion, mesoporous n-i-p perovskite solar cells were fabricated using an inexpensive, additive-free hole transport material based on vinyl triarylamines, and various solar parameters were calculated. The SHTM showed promising results in FAPbI 3 -based devices, which achieved a high power conversion efficiency (PCE) of 16.86%, surpassing the PCE of 14.31% obtained in MAPbI 3 -based devices. Furthermore, the performance of our SHTM-based devices was compared with that of commonly used doped Spiro-OMeTAD and rarely used undoped Spiro-OMeTAD HTM-based FAPbI 3 solar cells. Our experimental results indicated that our SHTM based device yielded a higher PCE than undoped Spiro-OMeTAD, and it also showed a significant improvement in PCE over doped Spiro-OMeTAD as an HTM. Additionally, the stability of the SHTM-based device was found to be superior to that of the doped Spiro-OMeTAD-based devices, and comparable to that of the undoped Spiro-OMeTAD-based devices. The FAPbI 3 -based devices with SHTM as the hole transport layer retained 88% of the initial efficiency, while the same perovskite-based device utilizing an additive-containing Spiro-OMeTAD as the HTM showed a faster performance deterioration, retaining only 75% of the initial efficiency after 1800 h when the relative humidity level was maintained below 20%. The stability of the SHTM-based devices was also evaluated under continuous illumination, and the results showed only a 10% decrease in PCE after 400 h of continuous illumination at ambient temperature without encapsulation. These findings highlight the potential of SHTM in the development of highly efficient and stable perovskite solar cells for commercial applications.

Data availability statement
The data that support the findings of this study are available upon reasonable request from the authors.