Enhanced optoelectrical dynamics in ambient-processed CsFAMAPbX3 perovskite solar cells by lithium organic salt doping

Perovskite solar cells will be a model for future solar energy systems, thanks to the remarkable properties of their perovskite-based materials [1, 2]. Since their introduction, these materials have driven a rapid increase in power conversion efficiency, climbing to over 26% in less than two decades. However, a significant challenge remains, i.e., the stability as well as the ambient processability of perovskite solar cells [3, 4]. One of the key issues is that perovskite materials are highly sensitive to environmental factors, particularly oxygen and water vapor [5‐7]. This causes that perovskite solar cells can only be fabricated in tightly controlled atmospheres where the levels of oxygen and moisture are kept extremely low. This sensitivity is a major barrier not only for large-scale production but also for their durability and performance in real-world environments, where exposure to air and humidity is inevitable. Addressing these air-processability issues is critical for industrialization and widespread use of this system in renewable energy generation.

To overcome this problem, various strategies have been introduced, particularly composition engineering [8‐10] and interface engineering [11]. In previous study, we have introduced compositional engineering to increase the stability of the perovskite system against humidity and their processability under ambient condition by adding hexamethylene tetramine (hexamine) molecules into the perovskite lattice [6, 12]. As a result, besides their ambient processability, the perovskite layer can be stable under relative humidity up to 90%. The van der Waals effect in the perovskite lattice is the key process to increase this stability. This success allows the preparation of perovskite solar cells in ambient conditions with humidity levels as high as 40%. However, despite the increase in the stability along with ambient processability of the perovskite film by using this technique, the optoelectrical properties of the perovskite material were found to be lower compared to the perovskite system fabricated in a controlled environment. Therefore, a systematic approach is needed to improve the optoelectrical properties of humid-stable perovskite solar cells to make them comparable to those prepared in a controlled environment.

Doping or additive engineering is considered a sophisticated approach to improve the optoelectrical properties of perovskite layers [13, 14]. It has been witnessed that doping may enhance the stability and reduce defects, tune the bandgap and charge carrier dynamics, tailor the optical properties, and improve the interface and film quality [15‐18]. In this context, we propose lithium bis(trifluoromethanesulfonyl)imide, commonly known as Li-TFSI, as a potential dopant to improve the optoelectrical properties of ambient-processed CsFAMAPbX₃ perovskites. Li-TFSI has been shown to improve the conductivity and charge carrier lifetime of the spiro-MeOTAD hole transport layer, leading to enhanced performance of perovskite solar cells [19, 20]. We hypothesize that by incorporating Li-TFSI into the perovskite layer, it is expected to increase the carrier concentration and reduce charge recombination, ultimately leading to higher power conversion efficiency [21, 22]. It has been widely reported that the Li-TFSI doping into perovskite increases carrier concentration via: Extrinsic doping by direct Li + incorporation, facilitated by the nature of small Li⁺ ions that can enter the perovskite lattice and bind to the lattice via interstitial sites or vacancies filing. When Li⁺ occupies interstitial positions or bonds to lattice vacancy sites, it can donate charge and change the Fermi level, effectively increasing free carrier concentration and changing conductivity or band filling. This has been observed in studies on FAPbI₃ perovskite system [23]. This phenomenon has also been observed in the hole transport layer (HTL) system that is doped by the Li-TFSI where HTL conductivity enhanced and the hole extraction improved, an indication of an increased in the carrier availability for device operation [24]. Li-TFSI doping into perovskite is also known to reduce the carrier recombination in the device. Coordination of Li + ions to under-coordinated lead or halide vacancy sites, reducing charged defect density in bulk or at grain boundaries and decreasing non-radiative recombination channels [23]. Similar effect of Li-TFSI doping to other system, such as TiO2, has also been reported [24, 25]. In other side, the TFSI anion can interact with surface halide vacancies and passivate them, which lowers surface recombination velocity. This is often complementary to the Li⁺ action [26]. It has also been reported that low concentration of Li-TFSI doping into perovskite influences crystallization, producing larger grains and fewer grain boundaries, or more favorable facet formation, which reduces recombination-prone grain-boundary defects [27]. Furthermore, Li-TFSI has been shown to improve the stability of perovskite solar cells under humid conditions by forming a protective layer on the surface of the perovskite film. This layer acts as a barrier against moisture intrusion, thereby preventing degradation and maintaining device performance over time.

We discovered that incorporating Li-TFSI salt into the PbI₂ layer during the growth of the CsFAMAPbX₃ perovskite layer, using the two-step process technique in an ambient atmosphere, considerably enhances both the bulk structure and morphology of the resulting perovskite film. This improvement contributes to a more uniform and high-quality film, which is essential for optimizing solar cell performance. The optimum Li + doping concentration is 3 mg/mL. Optical properties analysis using UV–Vis and external quantum efficiency (EQE) methods reveals that the presence of Li-TFSI salt leads to a reduction in the optical band edge by approximately 0.01 eV, indicating enhanced light absorption capabilities. Besides, perovskite solar cells fabricated with Li-TFSI-doped CsFAMAPbX₃ demonstrate impressive power conversion efficiencies reaching as high as 21.24%. These cells exhibit a short-circuit current density (J_sc) of 25.2 mA/cm², an open-circuit voltage (V_oc) of 1.11 V, and a fill factor (FF) of 75.68%, all considerably outpacing devices made from pristine CsFAMAPbX₃ perovskite, which achieve maximum efficiencies of only 17.37%, with J_sc, V_oc, and FF values of 23.1 mA/cm², 1.09 V, and 69.09%, respectively. The champion device can retain 91.8% of its initial efficiency when being aged for 27 days in an N₂ atmosphere. These findings suggest that the incorporation of the organic salt Li-TFSI has substantial potential for advancing the performance of perovskite solar cells, particularly when fabricated in ambient environments.

We have fabricated a PSC using ambient-processed CsFAMAPbX₃ perovskite doped with Li-TFSI to evaluate doping impact on photovoltaic performance. The PSC device structure is shown in Fig. 1a. We found that the addition of Li-TFSI into the underlying PbI₂ layer during perovskite layer preparation via the two-step method effectively enhanced the power conversion efficiency. In the typical process, the power conversion efficiency (PCE) of the PSC increases with the introduction of Li-TFSI doping and optimum at doping level of 3 mg/mL. The performance is then gradually decreases when the doping concentration further increases above 3 mg/mL, i.e., 4 and 5 mg/mL. At the optimized condition, the PCE reach the level of 21.24% with corresponding short-circuit current (J_sc), open-circuit voltage (V_oc), and fill factor (FF) of 25.2 mA/cm², 1.11 V, and 75.68%, respectively (Fig. 1b). The average PCE across multiple devices at this optimized condition was 19.97 ± 1.1%, with average values of J_sc, V_oc, and FF at 24.8 ± 0.62 mA/cm², 1.12 ± 0.03 V, and 71.1 ± 2.94%, respectively (Fig. 1c–f).

Full size image

This performance is significantly higher than that of devices fabricated with others Li⁺ doping concentrations. For example, the champion PSC devices fabricated using undoped CsFAMAPbX₃ (pristine) demonstrated a PCE only as high as of 17.37% with the average PCE of 15 ± 1.18%. Meanwhile, the PCE for devices prepared using 2 mg/mL, 4 mg/mL, and 5 mg/mL of Li⁺ doping is only as high as 20.65, 18.73, and 18.38%, respectively. Their average PCEs are 17.3 ± 2.60, 17.6 ± 0.81, and 17.4 ± 1.04%, respectively. The detailed comparison of the photovoltaic parameter of the devices, including J_sc, V_oc, and FF, is listed in Table 1.

We carried out external quantum efficiency analysis on the device with optimized condition (doped with 3 mg/mL of Li⁺) by using a device with more or less similar performance with the champion device and compared it with the pristine device. The result is shown in Fig. 2a. As the figure reveals, PSC device using the optimized CsFAMAPbX₃ perovskite exhibits a wider spectral response compared to pristine PSC device. For example, the EQE absorption edge for the optimized doped perovskite is 1.495 eV. Meanwhile, it is 1.512 eV for the pristine device (Fig. 2b). This agrees well with the optical band gap analysis using Tauc method (Fig. 2c and Table 1 for detail) that reveals that the doping process do shifted the energy gap to the right, reflecting the increase of absorption windows of the CsFAMAPbX₃ perovskite when being doped with Li-TFSI. This certainly yields enhanced photovoltaic response, the fact that agree with the J–V results presented in Fig. 1a. Energy gap red-shift calculations from both Tauc and EQE analysis are 0.01 and 0.017 eV, respectively. The shift seems to be low in value, but this reflects effective modification of perovskite optoelectrical properties when being doped with Li-TFSI. Figure 2a also shows the integrated current density obtained from the EQE measurement (J_EQE). The J_EQE for Li-TFSI 3 mg doped PSC device is 24.17 mA/cm². Meanwhile, the J_EQE for the reference PSC device is 23.56 mA/cm². The result shows that the total current density obtained using J–V and EQE analysis is highly consistent to each other.

Full size image

The exact mechanism for the improvement of the PSC performance when perovskite materials was doped with Li-TFSI is not yet understood. However, we hypothesized that the following process might take place. (i) Enhanced conductivity. Li-TFSI is known to greatly improve both the ionic and electronic conductivity in the perovskite layer [20]. In current study, this phenomenon could also be occurred. This enhancement assists more efficient charge transport within the material, reduces recombination losses during the charge collection process, and eventually leads to improved overall device efficiency. Electrochemical impedance spectroscopy analysis result in the dark (Fig. 3a), which reflects an overall transport property in the device (from charge transfer resistance (R_CT) and recombination resistance (R_REC) value) verifies this phenomenon. The EIS analysis in the dark allows the isolation of intrinsic material properties and phenomena from complicating effects of light-induced charge carriers during the carrier dynamic analysis in the device. Meanwhile, under illumination, perovskites can generate excess charge carriers that lead to non-linear behavior, masking the fundamental electronic and ionic processes within the material. Thus, from the EIS result in the dark, the intrinsic impedance characteristics of the device, charge transport mechanisms, and interface dynamics can be clearly interpreted. As Fig. 3a reveals, the Nyquist plot of the entire sample consists of one semi-circle at the high-frequency region (0–200 kΩ (Z_re)). By fitting the Nyquist plots using an equivalent circuit model (Fig. 3a), consisting of three resistive elements—series resistance (R_s), charge transfer resistance (R_CT), and recombination resistance (R_REC)—and two capacitive elements—interfacial capacitance (CP1) and recombination pathway capacitance (CP2) and Warburg element representing ionic conduction in the device—model that has been widely adopted to describe PSCs that capture both bulk transport and interfacial phenomena, we found the following (see Table S1 for detail): (i) The series resistance (R_s) values decreased with the increase of additive concentration and reached the lowest in the champion device. Further increasing of additive concentration slightly increased the R_s value, indicating effecting modification of the bulk resistance of the perovskite absorber layer. In the typical process, the R_s value for the pristine sample is 15.25 Ω. The R_s value decreased to 5.25 Ω in the champion device. The change in the bulk resistance of the perovskite absorber and the transparent conducting oxide upon additive incorporation has been reported elsewhere [36]. (ii) Charge transfer resistance (R_CT) and interfacial capacitance (CP1). A significant reduction in R_CT was observed with increasing Li-TFSI concentration, reaching a minimum in the champion device (from 535.1 Ω in the pristine sample to 451 Ω in the champion device). This trend reflects enhanced interfacial charge transfer kinetics, likely due to trap passivation and improved hole transport facilitated by Li-TFSI doping. The concurrent decrease in CP1, i.e., from 3.36 × 10^–8 F in the pristine sample to 2.24 × 10^–8 F in the champion device, supports this interpretation, as lower interfacial capacitance indicates reduced charge accumulation at defect sites, thereby suppressing interfacial recombination. Previous studies have shown that Li-TFSI, often in complexation with 4-tert-butylpyridine (tBP), improves hole transport layer conductivity and reduces trap-assisted recombination [20]. (iii) Recombination resistance (R_REC) and recombination pathway capacitance (CP2). R_REC increased steadily with higher Li-TFSI concentrations, peaking in the champion device. In the typical process, the R_REC valued increased from 4.63 × 10⁴ in pristine sample to 6.84 × 10⁴ in the champion device. Elevated R_REC values signify effective suppression of non-radiative recombination, consistent with improved crystallinity and reduced defect density in the perovskite films. The increase in CP2 corroborates this phenomenon, as greater charge storage along recombination pathways reflects longer carrier lifetimes and reduced recombination probability (see Table S1, Supporting Information). (iv) The ionic conduction in the device effectively decreases with the incorporation of additive and minimum at the champion device, indicating efficient suppression of ionic migration in the device. For example, the ionic conductivity in pristine device is 3.65 × 10^–6 S. It reduced to 2.43 × 10^–6 S in the champion device. Ionic migration is among the culprit for the crystalline phase instability in the perovskite solar cells. These results highlight the role of Li-TFSI in defect passivation and crystal growth regulation, in agreement with recent reports on additive-assisted perovskite film engineering.

Full size image

Carrier lifetime evaluation using Bode phase analysis further corroborated with above discussed findings. Bode phase analysis (Fig. 3b) revealed a substantial enhancement in carrier recombination lifetime upon Li-TFSI incorporation. The pristine device exhibited a lifetime of 159 μs, which increased to 398.4 μs in the champion device. This improvement is consistent with the suppression of trap-assisted recombination and the enhancement of optoelectronic quality of the perovskite absorber. Such lifetime extension has been widely reported as a hallmark of improved film morphology and reduced defect density in Li-TFSI-doped PSCs. The combined EIS and Bode analyses confirm that Li-TFSI additives enhance carrier dynamic in the device that are attributed to defect passivation and enhanced crystallinity.

(ii) Improved perovskite phase’s crystallinity [7, 37, 38]. The incorporation of Li-TFSI has been shown to promote better crystallinity of the CsFAMAPbX₃ perovskite layer, as reflected by the significant improvement in the perovskite’s phase X-ray diffraction peak intensity, i.e., 2θ of 13.9, 19.9, 25, 28, 32, 35, 41, 43.9, and 49°, and the reduction of PbI₂ phase peak intensity (i.e., 2θ ~ 12.3°) when Li-TFSI added into the perovskite lattice (Fig. 3c). In normal case, the increase of peak intensity is related to the fine structure and high crystallite size of the material. Higher crystallinity often results in larger grain sizes, which can reduce the presence of grain-boundary-related defects. According to the zoomed-in on one of the characteristic peaks of perovskite, i.e., (001) plane (Fig. 3d), the doping has significantly shifted the peaks to the right. This reflects that the doping caused stress to the lattice, which in turn modify the entire properties of the crystal. However, when concentration of Li-TFSI used is relatively high, i.e., 5 mg/mL, PbI₂ phase peak intensity becomes increase. We assumed that this is due to ineffective Li⁺ ion incorporation into the lattice, promoting Li⁺ ion segregation. This encourages the formation of PbI₂ phase as a secondary phase due to insufficient reactivity or incorporation into the lattice, the result of Li⁺ ion segregation.

FESEM analysis results shown in Fig. 4 more or less support this fact where the addition of Li-TFSI dopant produces improved grain crystallinity. The pristine sample (Fig. 4a-a1) clearly shows that the grain size does not exhibit faceting, reflecting less shape-crystallinity. Under the doping condition, the perovskite crystal grain indicates a clear faceting (see Fig. 4b to 4e), a parameter that shows enhanced crystallinity properties that is beneficial for interfacial carrier dynamic. Nevertheless, grain size does not change so significant upon being doped with Li-TFSI. In addition, however, at higher Li-TFSI doping concentrations, particularly at 4 and 5 mg/mL, noticeable pin-hole formation is observed within the perovskite layer. This phenomenon can be attributed to the incorporation of Li⁺ ions into the crystal lattice, which induces lattice segregation and increases surface tension during film formation. The elevated surface tension destabilizes the film morphology, making pin-hole formation inevitable, thereby deteriorating film quality. Such defects act as recombination centers and hinder charge transport, ultimately reducing device efficiency. This could also be the reason for the decrease of the performance when Li⁺ concentration further increases above 3 mg/mL. To achieve improved morphology and optimal performance, a medium doping concentration, especially around 3 mg/mL, offers a more balanced condition that minimizes pinholes while maintaining favorable electronic properties.

Full size image

In addition, the refinement of the phase crystallinity can also enhance the interface quality between the perovskite layer and other layers, such as the electron transport layer. This improvement leads to enhanced charge extraction and a reduction in interfacial recombination, which can be a major barrier to device efficiency. EIS analysis result in Fig. 3a verified these processes where the interfacial charge transfer dynamic, reflected by the interfacial charge transfer resistance value, increases when the perovskite is doped with the Li-TFSI. (iii). Defect passivation [1, 8, 15]. The addition of Li-TFSI into the perovskite might also passivate the defects on the surface and the grain boundary of the materials. Dark current analysis results as presented by semi-log plot of J–V in the dark (Fig. 5a) confirm these phenomena where the addition of Li-TFSI into the perovskite materials significantly reduce the leakage current in the device (see the negative bias of the device), a reflection of the defect properties in the active materials. This defects density can be analyzed by interpolating a double-log J–V curve from the J–V response in the dark (Fig. 5a). The result is shown in Fig. 5b. The trap density (N) can be estimated using the relation \(N=\frac{3\varepsilon {\varepsilon }_{0}{V}_{TFL}}{e{L}^{2}}\), where L is the perovskite thickness (~ 390 nm), ε is the relative dielectric constant of the perovskite (ε ~ 6.6), and ε₀ is the vacuum permittivity. The trap-filling voltage (V_TFL) is determined from the intersection of the linear region corresponding to the trap-filling transition with the voltage axis (see Fig. 5b). It is obtained that the defects density in the champion Li-TFSI-doped perovskite materials device is as low as 4.32 × 10²¹ cm⁻³. Meanwhile, it is as high as 4.70 × 10²¹ cm⁻³ in the undoped perovskite-based device. These defects are often sources of non-radiative recombination, which affects the overall efficiency of the solar cell. By reducing these defects, Li-TFSI can greatly contribute to higher open-circuit voltage (V_oc) and fill factor (FF), resulting in more efficient photovoltaic devices, which is in agreement with the photovoltaic properties results shown in Fig. 1. Nevertheless, the optimum device is not the system with the lowest defect density (see Table 1). However, because of its superiority in term of interfacial charge transfer dynamic, carrier mobility (discussed below), and crystallinity properties, as discussed earlier, its photovoltaics performance is higher than other Li⁺-doped devices. We then evaluated the space charge limited current (SCLC) mobility of the device using the double-log J–V and Child equation (\(J=\frac{9\varepsilon {\varepsilon }_{0}\mu {V}^{2}}{8{L}^{3}}\), where μ is SCLC carrier mobility) to evaluate the extent effect of Li-TFSI doping on the transport properties of the materials. It is found that the SCLC carrier mobility improves significantly with the incorporation of Li-TFSI into the perovskite lattice. The carrier mobility in the optimum sample is as high as 4.51 × 10^–5 cm²/Vs, which is 100 times higher than the pristine sample (4.24 × 10^–7 cm²/Vs). The carrier mobility of the optimum sample is also significantly higher than other Li⁺-doped device system (Table 1). The refinement of the phase crystallinity upon doping is further verified by the photoluminescence spectroscopy analysis results. As Fig. 5c reveals, there is a significant shifting in the absorption edge and the photoluminescence peak (See also Fig. 2b and c), reflecting phase crystallinity improvement. As has been discussed earlier, the absorption band edge shift to the lower optical energy can be up to 0.017 eV in the optimum device compared with the pristine system, broadening their spectral sensitivity. We then carried out transient photoluminescence (TRPL) analysis to find the effect of doping on the carrier lifetime. The result is presented in Fig. 5d (the TRPL spectrum of each sample is stacked on top of each other for clarity). The carrier lifetime analysis obtained from the transient photoluminescence analysis result found that the excitonic lifetime increases with the increasing of Li⁺ doping concentration and maximum at the optimized device. For example, it increases from 22.36 ns in the pristine sample to 223.70 ns in the optimum Li⁺-doped device system. (see Fig. 5d and Table 1). This result corroborates with other analysis results, confirming positive impact of Li-TFSI doping on the optoelectrical properties of the CsFAMAPbX₃ perovskite system. The cumulative effects of the enhanced crystallinity, improved charge transport, reduced defect levels, and increased structural stability can collectively lead to remarkable improvements in photovoltaic performance.

Full size image

The doping effect on the performance of the perovskite solar cells obtained in this study is quite significant that record a 22.3% improvement from the pristine device, reflecting fine synergetic role of Li-TFSI in improvement of structural growth and the optoelectrical performance. The improvement of the performance upon doping surpasses the improvement of the performance using recent method, particularly molecular additive engineering method. Table 2 shows comparison of the present enhancement in the performance upon doping with the Li-TFSI with the recent method. As the table shows, the current method can enhance the power conversion efficiency that exceeds the available sophisticated method.

The exact mechanism of Li-TFSI doping process into the CsFAMAPbX₃ perovskite has not yet been understood. Nevertheless, the following process can be considered. In the solution, the Li-TFSI may dissociate into Li⁺ and TFSI⁻ ion. Li⁺ ion then may fill perovskite A-site along with the methylammonium ion. Meanwhile, the TFSI⁻ ion might coordinate with the iodide ion vacancy site at the grain surface, passivating the defects. Both processes will improve the optoelectrical properties of the CsFAMAPbX₃ perovskite. This process has been reported in the recent literature [46]. It is reported that doping Li-TFSI onto the MAPbI₃ perovskite results in, while Li⁺ ion occupies the A-site of perovskite, TFSI⁻ ion coordinate with the uncoordinated Pb²⁺ ions, as reflected by the red-shift in the XPS binding energy of Pb 4f. This certainly indicates that the TFSI⁻ may passivate the iodide vacancy at the perovskite grain’s surface. While this result observed in the MAPbI₃ perovskite, similar process would be occurred in the CsFAMAPbX₃ perovskite system. In addition, when Li-TFSI is introduced into the CsFAMAPbX₃ perovskite, the Li⁺ ion in the Li-TFSI primarily acts as a p-type dopant in the material, as observed in the doping process of hole transport materials spiro-OMeTAD. In the perovskite lattice, similar process might occur where Li-TFSI introduces lithium ions (Li⁺). These ions enhance their electrical conductivity by promoting the formation of holes (positive charge carriers) [19]. At the same time, TFSI⁻ anions effectively interact with the perovskite to enhance hole mobility within the material. Earlier study indicates that Li-TFSI doping can lead to a major increase in hole formation, as evidenced by a substantial enhancement in electron spin resonance signals, which reveals a marked increase in spin density [21].

We further evaluated the stability of Li-TFSI-doped CsFAMAPbX₃ perovskite solar cells by monitoring their photovoltaic performance over a 27-day storage period under an N₂ atmosphere (Fig. 6). The devices retained 91.8% of their initial efficiency after 27 days. Although this value is lower than the stability benchmarks reported for state-of-the-art perovskite solar cells, which can preserve > 95% of their initial efficiency under maximum power point tracking (MPPT) for 1000 h, the present results remain encouraging. It should be noted that fluctuations in measurement conditions—such as contact instability and electrical noise—can influence performance tracking. Therefore, the observed results highlight the excellent intrinsic stability resulted from Li⁺ incorporation. Ongoing work aims to obtain full MPPT stability profiles under continuous operational conditions to more precisely elucidate the role of Li⁺ doping in enhancing device durability. For the industrialization and commercialization prospect of the present results, scalability of the perovskite solar cells device is a challenging aspect and needs a further study. We are in the process of scaling up of the device and analyze suitable method and how preparation approach influences the perovskite solar cells performance. Finally, how the doping approach is economically beneficial over the other systems will also be evaluated. The result of the analysis will be reported in different communication.

Full size image

The impact of Li-TFSI addition into the ambient-processed CsFAMAPbX₃ perovskite structure and optoelectrical properties has been investigated. In the typical process, we discovered that the introduction of Li-TFSI, which dissociates into Li¹⁺ and TFSI⁻ ions, considerably enhances the bulk structure and morphology as well as optoelectrical properties of the resulting perovskite film. Morphological analysis found that the crystallinity of the grain and the size of the grain increases with the addition of Li-TFSI dopant. The result shows that the perovskite grain becomes faceted and has smooth topology, the feature beneficial for enhanced interfacial coupling and charge transfer dynamic. It is also discovered that the doping process substantially reduces the optical band edge of the perovskite material by approximately 0.01 eV, suggesting enhanced spectral response. The defect density in the perovskite material is also dramatically reduced upon being doped with Li-TFSI as verified by the dark current and photoluminescence analysis results. Perovskite solar cells fabricated with Li-TFSI-doped CsFAMAPbX₃ produces impressive power conversion efficiencies reaching as high as 21.24%. These cells demonstrate impressive performance metrics, exhibiting a short-circuit current density (J_sc) of 25.2 mA/cm², an open-circuit voltage (V_oc) of 1.11 V, and a fill factor (FF) of 75.68%. Such values greatly surpass those achieved by pristine devices, which only reach maximum efficiencies of about 17.37%, accompanied by J_sc, V_oc, and FF readings of 23.1 mA/cm², 1.09 V, and 69.09%, respectively. The champion device retains 91.8% of its initial efficiency after storage for 27 days in N₂ atmosphere. These findings strongly suggest that the incorporation of the organic salt Li-TFSI not only enhances the efficiency of perovskite solar cells but also demonstrates substantial potential for advancing the performance of ambient-processed CsFAMAPbX₃ perovskite solar cells. Nevertheless, current advances in the additive engineering have come up with many potential dopants for improving the perovskite solar cells performance, particularly the molecules that suppress a rapid crystallization of perovskite and facilitate complete perovskite phase formation. They include oxydibenzenesulfonyl chloride (OBSC), poly(vinyl acetate) (PVAc), and methyl carbamimidothioate hydroiodide (MCH). These molecules have similar function as Li-TFSI and can bind to PbI₂ in the solution to slow down the crystallization process for highly pure perovskite phase formation. An optimized device fabrication using the new additive molecules may lead to the high-performance perovskite solar cells production.