Graded 2D/3D Perovskite Hetero-Structured Films with Suppressed Interfacial Recombination for Efficient and Stable Solar Cells via DABr Treatment

Several strategies and approaches have been reported for improving the resilience and optoelectronic properties of perovskite films. However, fabricating a desirable and stable perovskite absorber layer is still a great challenge due to the optoelectronic and fabrication limitations of the materials. Here, we introduce diethylammonium bromide (DABr) as a post-treatment material for the pre-deposited methylammonium lead iodide (MAPbI3) film to fabricate a high-quality two-dimensional/three-dimensional (2D/3D) stacked hetero-structure perovskite film. The post-treatment method of DABr not only induces the small crystals of MAPbI3 perovskite secondary growth into a large crystal, but also forms a 2D capping layer on the surface of the 3D MAPbI3 film. Meanwhile, the grains and crystallization of 3D film with DABr post-treatment are significantly improved, and the surface defect density is remarkably reduced, which in turn effectively suppressed the charge recombination in the interface between the perovskite layer and the charge transport layer. The perovskite solar cell based on the DABr-treatment exhibited a significantly enhanced power conversion efficiency (PCE) of 19.10% with a notable improvement in the open circuit voltage (VOC) of 1.06 V and good stability, advocating the potential of this perovskite post-treatment approach.


Results and Discussions
Herein, we introduced a facile but novel halide perovskite material of DABr to treat pre-deposited MAPbI 3 films for achieving 2D/3D-stacked perovskite architecture. The developed technique involved the DABr post-treatment of one step MAI-based film and followed by brief annealing. DABr not only directly reacts with surplus of PbI 2 in the MAI based film, but also interacts with MAPbI 3 via organic-cation exchange to form 2Dcapped 3D MAPbI 3 hybrid film [25,37]. 2D/3D perovskite films exhibited excellent surface morphology with large grain size dimension, good crystallinity and notable uniformity. The 2D capping layer can effectively passivate the surface defects of 3D perovskite layer, leading to suppress the interfacial non-radiative recombination and boost the hole extraction, thus enhancing the V OC of the device [20]. Figure 1a schematically illustrates the stepwise experimental processes for preparing highly crystalline, large-grain DABr capped MAPbI 3 thin films via DABr post-treatment with altered DABr concentrations viz. 1.5, 2.5, and 3.5 mg·mL −1 in 2-isopropyl alcohol (IPA) solution. During the second stage of the spin-coating process, 100 µL of anhydrous chlorobenzene (CB) was dripped onto the center of the rotating film 10 s prior to the end program. As shown in Figure 1b, the surface morphology of the as-deposited MAPbI 3 perovskite film was optimized with a high-quality uniform top appearance by DABr treatment. It is inferred that the introduced DABr interrupts the centrifugal progress of MAPbI 3 and leads to a uniform expansion in all dimensions. The post-anneal for the MAPbI 3 film with DABr treatment causes the perovskite grains to grow significantly again, resulting in the increased crystallinity of the final perovskite films [25,34].
Molecules 2023, 28, x FOR PEER REVIEW 3 of 15 devices. As a result, the optimized devices achieved a high PCE of 19.10% and long-term stability.

Results and Discussions
Herein, we introduced a facile but novel halide perovskite material of DABr to treat pre-deposited MAPbI3 films for achieving 2D/3D-stacked perovskite architecture. The developed technique involved the DABr post-treatment of one step MAI-based film and followed by brief annealing. DABr not only directly reacts with surplus of PbI2 in the MAI based film, but also interacts with MAPbI3 via organic-cation exchange to form 2D-capped 3D MAPbI3 hybrid film [25,37]. 2D/3D perovskite films exhibited excellent surface morphology with large grain size dimension, good crystallinity and notable uniformity. The 2D capping layer can effectively passivate the surface defects of 3D perovskite layer, leading to suppress the interfacial non-radiative recombination and boost the hole extraction, thus enhancing the VOC of the device [20]. Figure 1a schematically illustrates the stepwise experimental processes for preparing highly crystalline, large-grain DABr capped MAPbI3 thin films via DABr post-treatment with altered DABr concentrations viz. 1.5, 2.5, and 3.5 mg·mL −1 in 2-isopropyl alcohol (IPA) solution. During the second stage of the spin-coating process, 100 µL of anhydrous chlorobenzene (CB) was dripped onto the center of the rotating film 10 s prior to the end program. As shown in Figure 1b, the surface morphology of the as-deposited MAPbI3 perovskite film was optimized with a high-quality uniform top appearance by DABr treatment. It is inferred that the introduced DABr interrupts the centrifugal progress of MAPbI3 and leads to a uniform expansion in all dimensions. The post-anneal for the MAPbI3 film with DABr treatment causes the perovskite grains to grow significantly again, resulting in the increased crystallinity of the final perovskite films [25,34].  The evolved changes in surface morphology of the prepared 2D-capped 3D perovskite films before and after the DABr surface treatment were explored through scanning electron microscopy (SEM) images, as shown in Figure 2a of the respective treated films is summarized in Figure S1. It was previously reported that the morphology of the perovskite film with a large grain size and a compact even surface is a critical factor to attain high photovoltaic performance [18]. The orthodox control film, as expected, showed irregular grain morphology with small average grain size distribution within the proximity of~200 nm (Figure 2a and Figure S1). Interestingly, when the deposited films were treated with varying concentrations of DABr, changes occurred in both the surface morphology and the grain size. The morphology of perovskite film treated with 1.5 mg·mL −1 DABr exhibited reasonable improvement in surface uniformity and compactness with a little surge in grain size, where the average grain size distribution is reported within the proximity of~450 nm (Figure 2b and Figure S1). The DABr-2.5 film showed large grain evolution and in uniformly packed symmetry, with grain size larger than approximately~600 nm, as witnessed in Figure 2c and Figure S1. However, the grain quality of the film deteriorated as the concentration of DABr further increased to 3.5 mg·mL −1 in solution (DABr-3.5), despite the fact that the grain size distribution remained approximately~620 nm (Figure 2d and Figure S1). Remarkably, the increased grain dimension of the perovskite film with DABr-2.5 could be caused by the newly formed layer that homogenously covered the control film, which is speculated to be the 2D perovskite capping layer. Therefore, the large grain 2D capping layer with fewer grain boundaries can suppress the defects of perovskite, resulting in inhibiting the nonradiative recombination and enhancing the device's performance [38][39][40]. The SEM images confirmed that surface treatment with a suitable small amount of DABr can promote a homogenous film crystallization, leading to good quality 2D-3D-stacked films with large grains and homogeneous surface morphology. The witnessed gradually and significantly suave film treated with DABr-2.5 confirms that 2.5 mg·mL −1 is the optimal concentration for DABr to improve the surface morphology of perovskite film. The evolved changes in surface morphology of the prepared 2D-capped 3D perovskite films before and after the DABr surface treatment were explored through scanning electron microscopy (SEM) images, as shown in Figure 2a-d. The calculated grain size distribution of the respective treated films is summarized in Figure S1. It was previously reported that the morphology of the perovskite film with a large grain size and a compact even surface is a critical factor to attain high photovoltaic performance [18]. The orthodox control film, as expected, showed irregular grain morphology with small average grain size distribution within the proximity of ~200 nm (Figures 2a and S1). Interestingly, when the deposited films were treated with varying concentrations of DABr, changes occurred in both the surface morphology and the grain size. The morphology of perovskite film treated with 1.5 mg·mL −1 DABr exhibited reasonable improvement in surface uniformity and compactness with a little surge in grain size, where the average grain size distribution is reported within the proximity of ~450 nm (Figures 2b and S1). The DABr-2.5 film showed large grain evolution and in uniformly packed symmetry, with grain size larger than approximately ~600 nm, as witnessed in Figures 2c and S1. However, the grain quality of the film deteriorated as the concentration of DABr further increased to 3.5 mg·mL −1 in solution (DABr-3.5), despite the fact that the grain size distribution remained approximately ~620 nm (Figures 2d and S1). Remarkably, the increased grain dimension of the perovskite film with DABr-2.5 could be caused by the newly formed layer that homogenously covered the control film, which is speculated to be the 2D perovskite capping layer. Therefore, the large grain 2D capping layer with fewer grain boundaries can suppress the defects of perovskite, resulting in inhibiting the non-radiative recombination and enhancing the device's performance [38][39][40]. The SEM images confirmed that surface treatment with a suitable small amount of DABr can promote a homogenous film crystallization, leading to good quality 2D-3D-stacked films with large grains and homogeneous surface morphology. The witnessed gradually and significantly suave film treated with DABr-2.5 confirms that 2.5 mg·mL −1 is the optimal concentration for DABr to improve the surface morphology of perovskite film.  To ponder into the nexus of DA and Br onto the pre-deposited film, 3D-MAPbI 3 films were treated with Br halide in combination with different DA alternate ammonium cations (BA and PEA). The SEM images of the resultant BABr-2.5 mg·mL −1 and PEABr-2.5 mg·mL −1 capped films are shown in Figure S2. PEABr-and BABr-treated films parade smaller grain compared with DABr-2.5 mg·mL −1 -treated film (Figure 2c). Consequently, it was seen that the surface morphology of the films post-treated with different ammonium salts was determined to the type of organic cation. The DABr performed impressively in this case. Figure S3 shows the molecular formulas and chemical structures of DABr, BABr and PEABr, depicting the position of the ammonium molecule in the chain network. To further explore that whether only Br that can effectively induce the emergence of 2D layer over MAPbI 3 film, the DA in combination with other halides are also applied to treat the surface of perovskite film. The SEM images of MAPbI 3 films treated with different DA halides are recorded, as shown in Figure S4. The grain size of all DA halides-treated films surged notably than that of the control film, which directs that all the halides perform better with DA cation in covering the 3D-MAPbI 3 films, but DABr stood exceptionally. With the same doping concentration for the DA cation, the grain size of the DABr-treated film is comparatively larger than that of the DAI and DACl-treated films with homogenous surface coverage. All the results indicate that both DA cation and Br anion are involved in the induced merger with MAPbI 3 grains during post-treatment. It was reported that this large homogeneous crystal grain formation of DABr-treated film could be attributed to the lower surface energy of DABr-treated perovskite [25,37,41]. The DABr-2.5 solution primarily dissolved in the small crystalline owing to the higher surface energy, whereas the second step involves the formation of Br-related large crystalline grain. The above results indicated that the 3D MAPbI 3-based perovskite layer treated with 2.5 mg·mL −1 DABr can induce secondary growth for the control film which incorporates with small size grains, thus turning them into larger grain size 2D-3D stacked films [34].
The crystalline phase and properties of the DABr-treated and control perovskite films were analyzed via an X-rays diffraction (XRD) pattern. Figure 3a shows the XRD patterns of 2D-3D based films with numerous DABr concentrations. As expected, MAPbI 3 -based film shows the prominent diffraction peaks of the pure phase tetragonal MAPbI 3 with moderate peak intensity [20,39,40]. For the DABr-treated films, the gradually strengthened pattern with the characteristic peak at ≈14.4 • shows a little shift towards the higher degree planes. As shown in Figure 3b, all three dominant peaks of the DABr-treated films are slightly shifted to larger diffraction angles, perhaps due to the incorporation of small radius Br ions into the base 3D MAPbI 3 layer [38][39][40][41]. As shown in Figure 3a, the peaks under 10 • are observed in the XRD patterns of DABr-treated perovskite films, which is the evidence for the appearance of 2D perovskite phase [35,39,40]. It is noticeable that the peak intensities of XRD pattern increase as the perovskite layer treats with DABr and the peak intensity rises up to the highest in DABr-2.5-treated perovskite film, which indicates the enhanced crystallinity of perovskite film, as shown in Figure 3a [42,43].
The influence of the generated 2D thin film on the optoelectronic properties of the perovskite films was explored using light absorption spectroscopy. The absorption profiles of the prepared films were investigated with the help of UV-vis absorption spectroscopy. We hypothesized that organic cations can minimize imperfection and trap states while also converting imperfections from 3D perovskite to 2D perovskite. Obtained spectra of the control and DABr-treated films, as shown in Figure 3c,d, reveal that the DABrtreated films exhibit comparatively lead in light absorption, with the DABr-2.5 being more robust in absorbing spectrum, followed by DABr-3.5 and DABr-1.5, respectively. The improved absorption characteristics of 2D-capping layer over 3D-MAPbI 3 perovskite films will exaggerate the short circuit current density (J SC ) of PSCs. As displayed in Figure 3d, the absorption edge of the DABr-treated film exhibits a slight blue shift compared with the control film, resulting from the generation of 2D perovskite and the introduction of Br into the film lattice structure [28]. It is well known that the bandgap of 2D perovskite is larger than that of 3D perovskite, thus the 2D/3D-stacked heterojunction will slightly enlarge the bandgap of perovskite film. In addition, the introduction of Br in perovskite film will make the valence band of the perovskite shift down, also leading to increase the perovskite bandgap. [20,21]. Furthermore, the light absorption in intensity of the treated film was improved compared to the non-treated film. This suggested the DABr post-treatment has a positive effect in light absorption ability of perovskite film. The high light absorption of the optimized DABr-treated perovskite films compared to the non-treated film can attribute not only to the increased crystallinity uniform perovskite with less defects. film can attribute not only to the increased crystallinity uniform perovskite with less defects.
In order to verify the hypothesis that the post-treatment of DABr perhaps leads to the variation in chemical bonding between I and Pb, X-ray photoelectron spectroscopy (XPS) was used to thoroughly investigate the effect of DABr on the chemical structure of control and DABr-2.5 perovskite films, as illustrated in Figures 3e,f and S5. The peaks of I 3d and Pb 4f shifted to larger energy binding as the perovskite treated DABr-2.5, demonstrating that there is an interaction between DABr, Pb 2+ , and I − [20,[30][31][32]. Based on the above-mentioned analyses, we came forward with a credible proposed notion regarding the intermediary transitional process for the synthesis of high-quality MAPbI3-base film with larger grain size and less defect crystallinity [40][41][42][43].  In order to verify the hypothesis that the post-treatment of DABr perhaps leads to the variation in chemical bonding between I and Pb, X-ray photoelectron spectroscopy (XPS) was used to thoroughly investigate the effect of DABr on the chemical structure of control and DABr-2.5 perovskite films, as illustrated in Figure 3e,f and Figure S5. The peaks of I 3d and Pb 4f shifted to larger energy binding as the perovskite treated DABr-2.5, demonstrating that there is an interaction between DABr, Pb 2+ , and I − [20,[30][31][32]. Based on the above-mentioned analyses, we came forward with a credible proposed notion regarding the intermediary transitional process for the synthesis of high-quality MAPbI 3 -base film with larger grain size and less defect crystallinity [40][41][42][43].
To further ponder into the trap density of the prepared films, photo-physical measurements were measured with photoluminescence (PL) and time-resolved photoluminescence (TRPL), as shown in Figure 4. Both samples were deposited on glass side substrate of FTO, so as to explore the charge injection between the HTL and active layer [23]. The observed decays were analyzed by a bi-exponential model and the lifetime parameters were extracted. The TRPL decay curve was fitted by a bi-exponential model to calculate the decay time (τ i ) of the control and DABr-2.5-treated films as Equation (1) and the average PL decay time (τ ave ) was calculated by Equation (2) [20,23,27].
where A i is the relative decay amplitude, τ i is the decay time, and f 0 is the constant. The related parameters and the calculated carrier lifetime from TRPL are summarized in Table S1. The decay curves were distributed into two decay parts, comprising a fast decay (τ 1 ) and slow decay (τ 2 ) [23,38]. The τ 1 was attributed to the non-radiative interfacial recombination, and the τ 2 was associated with the trap recombination in the perovskite bulk. The average carrier lifetime of the DABr-2.5-treated film was 145.36 ns, whereas the control layer exhibited a shorter average carrier lifetime of 63.34 ns. The extended carrier lifetime of the DABr-2.5-treated film (Figure 4a) indicates enhanced film luminescence due to the better carrier mobility, which is attributed to the reduced grain boundaries and trap sites as well [18,39,41]. Figure 4b exhibits the steady-state PL spectra of the pristine and DABr-2.5 films. In comparison with the control film, an elevated intensity peak with a justified blue shift in the PL spectrum of DABr-treated film is observed, supporting the fact that the treatment of DABr suppresses non-radiative recombination and widens the bandgap of perovskite film due to the incorporation of Br to the perovskite structure [20,26]. Such an improvement in the PL response of DABr-2.5 film could be a result of a well-engineered 2D-3D interface formed at the surface of 3D-MAPbI layers, leading to a decrease in statistics of the deep-level traps in bulk perovskite layers, which can greatly suppress the non-radiative recombination of charge carriers [36,40,44]. The corresponding film photoluminescence (PL) intensity mappings were performed to further investigate the effect of DABr treatment on the optical characteristic behaviors. The optical surface dynamics of the films are given in Figure 4c,d. Bright red color shows the PL intensity. PL intensity mapping provides a direct insight into the non-radiative exciton decay mechanism. The peak intensity of the control film is restricted by the bulk trap flux due to the poor crystal grain and poor carrier mobility [20]. However, the DABr-2.5 film shows a higher peak intensity, signifying a reduced non-radiative exciting decay. Figure 5a illustrates the device architecture of the 2D-capped 3D MAPbI 3 perovskite. 3D-MAPbI 3 film treated with DABr-2.5 is employed as an active layer to prepare PSC with a 2D-3D hetero-structure film. As exhibited in Figure 5a, the device structure is FTO/C-SnO 2 /perovskite/spiro-OMeTAD/Au. A cross-SEM image of the final prepared device is shown in Figure 5b, demonstrating that the underlying ETL is consistently covered by the unceasing and vertically growth 3D MAPbI 3 film with evenly distributed large-grains. To examine the photoelectric performance, the fabricated devices based on 3D-MAPbI 3 and 2D-capped 3D-MAPbI 3 films were measured under the forward and reverse scan directions, as presented in Figure 5c. The optimal PCE of the device based on 3D MAPbI 3 reached 16.98% PCE with J SC of 22.78 mA·cm −2 , V OC of 1.02 V and FF of 72.75% under reverse scan, while achieving a 13.55% PCE under forward scan (Figure 5c). However, the device based on 2D-capped 3D film showed a comparatively enhanced performance with the highest PCE value of 19.10%, J SC of 23.15 mA·cm −2 , V OC of 1.06 V, and FF of 77.85% under reverse scan, while achieving a proximate PCE of 18.97% under a forward scan (Figure 5c). Obviously, DABr can inhibit the hysteresis and internal losses of PSCs. Additionally, to investigate the influence of different halides on the device photoelectric performance, the devices based on 3D-MAPbI 3 with the post-treatment of different DA halides were fabricated. The J-V characteristic curves of the perovskite devices consisting of control, DABr-2.5, DAI-2.5-and DACl-2.5-treated films are shown in Figure S6 and the results are presented in Table S2. The device based on the 2D-3D film showed a significant improvement in the V OC and FF, indicating that the charge recombination is significantly restrained by the reduced defect density [36,43]. As expected and in the light of the above discussion, the treatment of DABr-2.5 not only outperformed the control device but also outperformed devices based on other DA halide-treated 2D-3D films.  Figure 5a illustrates the device architecture of the 2D-capped 3D MAPbI3 perovskite. 3D-MAPbI3 film treated with DABr-2.5 is employed as an active layer to prepare PSC with a 2D-3D hetero-structure film. As exhibited in Figure 5a, the device structure is FTO/C-SnO2/perovskite/spiro-OMeTAD/Au. A cross-SEM image of the final prepared device is shown in Figure 5b, demonstrating that the underlying ETL is consistently covered by the unceasing and vertically growth 3D MAPbI3 film with evenly distributed large-grains. To examine the photoelectric performance, the fabricated devices based on 3D-MAPbI3 and 2D-capped 3D-MAPbI3 films were measured under the forward and reverse scan directions, as presented in Figure 5c. The optimal PCE of the device based on 3D MAPbI3 reached 16.98% PCE with JSC of 22.78 mA·cm −2 , VOC of 1.02 V and FF of 72.75% under reverse scan, while achieving a 13.55% PCE under forward scan (Figure 5c). However, the device based on 2D-capped 3D film showed a comparatively enhanced performance with the highest PCE value of 19.10%, JSC of 23.15 mA·cm −2 , VOC of 1.06 V, and FF of 77.85% under reverse scan, while achieving a proximate PCE of 18.97% under a forward scan (Figure 5c). Obviously, DABr can inhibit the hysteresis and internal losses of PSCs. Additionally, to investigate the influence of different halides on the device photoelectric performance, the devices based on 3D-MAPbI3 with the post-treatment of different DA halides were fabricated. The J-V characteristic curves of the perovskite devices consisting of control, DABr-2.5, DAI-2.5-and DACl-2.5-treated films are shown in Figure S6 and the results are presented in Table S2. The device based on the 2D-3D film showed a significant improvement in the VOC and FF, indicating that the charge recombination is significantly restrained by the reduced defect density [36,43]. As expected and in the light of the above discussion, the treatment of DABr-2.5 not only outperformed the control device but also outperformed devices based on other DA halide-treated 2D-3D films.  Moreover, the statistical distribution of PCEs based on 30 individuals with control and DABr-2.5-treated perovskite films, as shown in Figure 5d, suggested the reliability and repeatability of the DABr-2.5 treatment method to produce 2D-capped 3D MAPbI 3 PSCs. The statistical constriction of the DABr-2.5-treated device exhibited much finer spreading in the boosted efficiency regions in comparison to control devices. Furthermore, the incident photon to current efficiency (IPCE) curves and the integrated short circuit current density values based on the control and DABr-treated devices are presented in Figure 5e. Results showed a slightly but an obviously amplified spectral conversion profile with augmented integrated short circuit current density for the DABr-2.5 device in comparison to the control film-based device. Marginally improved spectral response and sound rise in integrated J SC could be accredited to the morphological improvement of the stacked perovskite films caused by spectral sensitive 2D-capped layer homogenously encapsulating the vertically grown bottom 3D MAPbI 3 layer [25,28,36]. Figurer 5f shows the steady-state output photocurrent density and PCE curves of DABr-2.5-based devices at a bias of 0.90 V for 300 s. A steady-state output photo-current density of 21.05 mA cm −2 and the corresponding PCE of 18.25%. The above results indicated that DABr-based devices displayed a relatively high stable PCE curve and good reproducibility.
Moreover, the open-circuit voltage decay was measured in order to investigate the devices' transient process of charge recombination mechanism. As shown in Figure 6a, the DABr-2.5-treated device exhibited a slower V OC decay behavior in comparison to the control device. The estimated decay constant of the DABr-2.5-treated device is 1.75 ns, which is larger than the control device (0.96 ns). As can be seen from Figure 6b, the electron lifetimes (τ n ) can be calculated by using the following Equation (3) [18,21].
where k B represents the Boltzmann constant, T refers to the absolute temperature, and e is the elementary charge. Apparently, the DABr-2.5 device exhibits a longer τ n compared to the control device, indicating that the DABr-2.5 device has a lower charge recombination rate. τ n is the highest for each case at lower voltage region, particularly the device with DABr post-treatment, which is consistent with the V OC decay curve. The τ n of all the cases gradually decreases with the V OC increasing. However, the τ n of the DABr 2.5-based device remains higher than that of the control based, suggesting an improved film morphology [34,41]. Electrochemical impedance spectroscopy (EIS) studies were carried out to gain a better understanding of the charge dynamics and recombination mechanisms of both devices. As exhibited in Figure 6c, the charge recombination and electric properties of the devices were studied to explore the dynamics of charge carrier recombination at the interface between the perovskite and carrier transport layer. The equivalent circuit model illustrated in the inset of Figure 6c was applied to fit the data [45][46][47][48][49][50]. R s stands for the series resistance and R rec represents the recombination resistance. R s is almost the same for the control and DABr-2.5-treated device, whereas the DABr-2.5-treated device shows a higher R rec than that of the control device, indicating a lower recombination rate. This implies that DABr significantly suppresses recombination and enhances the carrier extraction for PSCs [20,[51][52][53][54][55]. These EIS results demonstrated that the post-treatment of the perovskite film with DABr-2.5 enhaced the contact at interface between perovskite and HTL, and declined charge carrier recombination. The enhanced R rec account is aligned with the observed incerased V OC and FF of the DABr-2.5 device [32,56]. Figure 6d shows the photo-graphs pictures of contact angles for control film and DABr-2.5-treated film. Obviously, the 2D film overlapping on the 3D film makes it more water-friendly and, by extension, more durable in relation to the amount of moisture in the air. The improvement in resistance against the humidity could probably help to improve the stability against the humidity [43,57]. The long-term stability has always been an urgent and difficult feature to achieve for PSCs. The stability of the aged devices based on 3D and 2D-3D films was tested under 30% RH at 25 • C. The PCE value of the device based on 3D film gradually reduced to nearly 20% after 700 h of aging, as exhibited in Figure 6e. In the meantime, the device based on 2D-3D film maintained 75% of the initial PCE value for 700 h. It is proved that the device based on 2D-3D film showed the resilient moisture stability, probably due to lack of any potential loophole in film morphology, which did not let any ambient species intercalate into the perovskite structure and decompose it. Consequently, it is concluded that the treatment of DABr for 3D perovskite film not only improve the device stability but can also refine the film crystal quality with large grains and fewer traps, which reduces the trap-assisted recombination and leads to a fine performance output.
Molecules 2023, 28, x FOR PEER REVIEW 10 of 15 DABr post-treatment, which is consistent with the VOC decay curve. The τn of all the cases gradually decreases with the VOC increasing. However, the τn of the DABr 2.5-based device remains higher than that of the control based, suggesting an improved film morphology [34,41]. Electrochemical impedance spectroscopy (EIS) studies were carried out to gain a better understanding of the charge dynamics and recombination mechanisms of both devices. As exhibited in Figure 6c, the charge recombination and electric properties of the devices were studied to explore the dynamics of charge carrier recombination at the interface between the perovskite and carrier transport layer. The equivalent circuit model illustrated in the inset of Figure 6c was applied to fit the data [45][46][47][48][49][50]. Rs stands for the series resistance and Rrec represents the recombination resistance. Rs is almost the same for the control and DABr-2.5-treated device, whereas the DABr-2.5-treated device shows a higher Rrec than that of the control device, indicating a lower recombination rate. This implies that DABr significantly suppresses recombination and enhances the carrier extraction for PSCs [20,[51][52][53][54][55]. These EIS results demonstrated that the post-treatment of the perovskite film with DABr-2.5 enhaced the contact at interface between perovskite and HTL, and declined charge carrier recombination. The enhanced Rrec account is aligned with the observed incerased VOC and FF of the DABr-2.5 device [32,56]. Figure 6d shows the photo-graphs pictures of contact angles for control film and DABr-2.5-treated film.

Device Fabrication
FTO glasses were etched by using zinc power and hydrochloric acid (HCl). The obtained glass substrates were twice ultra-sonically cleaned with deionized water (DIW), liquid detergent solution, IPA, ACN and then ethanol. The FTO substrates were dried at the temperature 150 • C. The FTO were sintered at 500 • C for 35 min, in air to removed residual organic matter. All the substrates were UV Ozone cleaned for 15 min subsequently. The SnO 2 precursor was spin-coated onto glass/FTO substrates at 5000 rpm for 30 s, and then baked on a hot plate in ambient air at 150 • C for 30 min. Then, films were cool down to room temperature. The perovskite MAPbI 3 precursor solution was prepared by dissolving lead iodide (1.2 mol/mL) and methylammonium iodide (1.2 mol/mL) in an anhydrous solvent system of DMSO:DMF (1:4 of volume ratio). The deposited precursor solution was spin-coated initially at 1100 rpm for 10 s and then at 4500 rpm for 30 s during the second step. During the second stage of the spin-coating process, 100 µL of anhydrous chlorobenzene (CB) was dripped onto the center of the rotating film in 10 s prior to the end program. The obtained film was annealed at 60 • C for 2 min and 100 • C for 20 min. 50 µL of DABr in 2-isopropanol (IPA) solution with different concentrations (1.5 mg·mL −1 , 2.5 mg·mL −1 and 3.5 mg·mL −1 ) was spin-coated on the surface of the prepared MAPbI 3 film for 3 s and spin-coated at 4200 rpm for 15 s. The DABr post-treated films were thermally annealed at 100 • C for 15 min. For hole transport material, 73 mg of spiro-OMeTAD, 1 mL of CB, 17.5 µL of lithium bis (trifluoromethanesulfonyl) imide (Li-TFSI) solution (520 mg of Li-TSFI in 1 mL of acetonitrile) and 28 µL of 4-tert-butylpyridine were mixed. The 25 µL spiro-OMeTAD solution was spin-coated on the top surface of perovskite film at 4500 rpm for 30 s. Lastly, the gold (Au) electrode layer was deposited on the top of perovskite film, with 60 nm thickness by thermal evaporation technique under vacuum, at a constant evaporation rate of 0.6 nm/s.

Characterizations
A scanning electron microscope (Hitachi, Tokyo, Japan SU8010) was utilized to study the morphology. X-ray photo-electron spectroscopy (XPS) spectra were tested by a thermo fisher scientific ESCALAB 250XI (Shanghai, China). The crystallinity of perovskite film was measure by X-ray diffraction Cu Kα beam radiation of 0.15406 nm (X Pert Pro, Almelo, The Netherlands). Using UV-Vis spectrophotometer (SOLID 3700, SHIMADZU, Tokyo, Japan), the absorption spectra were measured. The steady state PL spectra of the produced samples were looked at using Edinburgh PLS980 (Edinburgh Instruments Ltd, Edinburgh, UK). Steady PL was recorded with a laser confocal Raman spectrometer (Princeton Instruments, Acton Standard Series SP-2558, Edinburgh Instruments Ltd, Edinburgh, UK) and a 532 nm laser using a home-built confocal microscope on a 10 × 10 µm 2 sample area. TRPL measurement was recorded by the FLS980 steady-state/transient fluorescence spectrometer (Edinburgh Instruments Ltd, Edinburgh, UK). Samples were excited with a 488 nm pulsed diode laser with a repetition rate of 5 MHz and an excitation intensity of~14 nJ/cm 2 , and the pulsed source was at 460 nm wavelength. The J-V characteristic curves were measured by utilizing forward (−0.1 to 1.2 V) or reverse (1.2 to −0.1 V) scans under AM 1.5 G solar illumination with a power intensity of 100 mW/cm 2 . The solar simulator equipped with (Keithley 2400) source meter (94043A, Oriel Instruments, Franklin, MA, USA) was calibrated with silicon cell in a nitrogen atmosphere. The cut-off voltage is set as (1.5 V) or (−1.5 V) for reverse or forward scans. 100 mA was used for the cut off current. The scan speed was 93 mV/s with a scan delay of 0 s. The solar cells were masked with a black cap of 0.09 cm 2 area to evade the scattering light and to define the active region. The EIS data were fitted with help of ZSim-software version 3.20 with equivalent circuit. The EQE spectra was measured using 300 W Xenon Lamp with spectral resolution of 5 nm equipped with order sorting filter (Oriel Instruments, Franklin, MA, USA). A Zahner electrochemical workstation (Zahner Zennium, Kronach, Germany) was used for electrochemical impedance measurements (EIS) in the frequency range of 10 mHz to 1 MHz. Impedance data were analyzed using Z-view ZSim-software version 3.20 with equivalent circuit modeling software.

Conclusions
In summary, we developed a novel sequential deposition method for obtaining 2D-3D stacked perovskite film by the post-treatment of DABr. The introduction of DABr with an optimum proportion is benefit for obtaining a large grain and highly crystalline 2D-capped 3D-MAPbI 3 perovskite film, effectively suppressing the nonradiative recombination at the interface between the perovskite and the charge transport layers. This method took the perovskite post-treatment strategy to the next level and offered a simple way for the fabrication of hetero-structure perovskite film with some outstanding photovoltaic performance and resilient stability.