Investigating the effects of modified anchoring groups on the surface structure of perovskites for enhanced stability†

Qingxia Zhao, Xiufang Hou*, Wenjing Xie, Leixing Luo, Shanshan Xu, Xiaoming Song and Qingbo Wei
Shaanxi Key Laboratory of Chemical Reaction Engineering, College of Chemistry and Chemical Engineering, Yan'an University, Yan'an, Shaanxi 716000, China. E-mail: houxiufang@yau.edu.cn

First published on 28th July 2025

---

---

1. Introduction

Perovskites, recognized as “dazzling star materials” in third-generation photovoltaic technology, exhibit a wide range of functional properties attributed to their diverse composition and structure.¹ Methylammonium lead halide perovskite (MAPbI₃) serves as a classic material for a perovskite absorber layer. Recently, several novel photovoltaic materials have emerged. For instance, chalcogenide perovskites possess admirable structural, electronic and optical properties, rendering them highly promising for solar energy conversion.^2,3 Additionally, antiperovskite nitrides (X₃NA; X = Mg, Sr, Ca, and Ba; and A = Sb and As) are identified as potential candidates for solar light harvesting applications.⁴ Furthermore, vacancy-ordered double perovskites Cs₂BI₆ (B = Pt, Pd, Te, and Sn) represent environmentally friendly and stable alternatives to lead halide perovskites.⁵ These materials hold significant potential to address the stability and toxicity challenges associated with conventional perovskite solar cells.

The photoelectric conversion efficiency (PCE) of perovskite solar cells (PSCs) has increased from 3.8% in 2009 to 26.7% today.^6,7 This substantial increase in PCE is attributed to the excellent optoelectronic properties of perovskites, including their bipolar transport,⁸ adjustable bandgap,⁹ high absorption coefficient,¹⁰ large carrier diffusion length,¹¹ and low exciton binding energy.¹² Despite these advantages, the instability of perovskite materials still represents a major barrier to the commercialization of PSC devices. This is because perovskites degrade rapidly when they are exposed to light,¹³ oxygen,¹⁴ moisture,¹⁵ or heat.¹⁶ For instance, moisture has a major influence on the PSC performance of MAPbI₃.¹⁷ In addition, the instability of PSCs typically arises from the inherent soft structural characteristics of perovskites, which lead to the volatilization and degradation of methylamine (MA), poorly coordinated ions (Pb²⁺ and I⁻), and the formation of defect states.^6,18 Uncoordinated Pb ions are regarded as a primary source of defects because of their low formation energy and potential to aggregate on the surface of perovskites.¹⁹ To effectively address the stability problems, various methods have been developed, including component regulation,²⁰ dimensionality reduction,²¹ and surface passivation.²² Surface passivation is an effective method for preparing high-performance PSCs under ambient conditions. Chemical substances are carefully selected for use as surface passivators to enhance the photovoltaic performance of perovskite-based devices. Lewis bases containing electron-donating functional groups—such as pyridine, hydroxyl, cyano, thiol, and carboxyl groups²³—have the capacity to simultaneously enhance both photovoltaic performance and stability.²⁴ This enhancement is attributed to the ability of Lewis bases to donate electrons to the uncoordinated Pb²⁺ ions on the surface of perovskites, thereby suppressing nonradiative interface recombination.

Various small organic molecules have been used to passivate the surface and defects of perovskite materials.^25,26 Xu et al.²⁷ compared the passivation effects of pyridine and its derivatives, which contain different functional groups (NH₂, CHO, and COOH). They discovered that Py-NH₂ had the highest passivation potential, increasing the PCE of C-PSCs from 11.6% to 14.8% while resulting in acceptable long-term stability. Wang et al.²⁸ analyzed theophylline, caffeine, and theobromine molecules with similar structures. They reported that the use of theophylline as a passivation agent facilitated multisite interactions and induced synergistic passivation effects because of its optimal binding configuration. Chen et al.²⁹ employed experimental and theoretical methods to explore the passivation effects of poly(vinyl alcohol) (–OH), polymethyl acrylate (–CO), and poly(acrylic acid) (–COOH). Their results indicated that poly(acrylic acid) (–COOH) not only formed complexes with uncoordinated Pb²⁺ ions but also selectively interacted with I⁻ and methylamine ions through hydrogen bonding, leading to enhanced photoelectric performance and moisture resistance stability in PSCs.

Matching the size of the passivation molecule (PM) with the perovskite lattice is an effective approach for enhancing the performance of PSCs. Yang et al.³⁰ designed a series of 3-alkylthiophene-based molecules for enhancing the stability and performance of PSCs by increasing the length of the hydrophobic alkyl chains. Huang et al.³¹ synthesized various alkylamine molecules by extending the exposed bilateral alkyl chain, which in turn improved the efficiency and operational stability of perovskites. Zhang et al.³² theoretically designed a series of PMs derived from 2-mercaptopyridine by extending the alkyl chain between the N and S atoms. They reported that compared with 2-mercaptopyridine, the newly developed PM, namely 2-(2-pyridyl)ethanethiol, exhibited considerably higher anchoring capability, thereby greatly enhancing the stability of PSCs in humid environments.

Although many PMs have been reported from both experimental and theoretical research,^33–35 variations in experimental conditions and theoretical perspectives have hindered the establishment of rules for selecting these molecules and developing effective passivation strategies. First-principles calculations provide insights into the static properties of materials, whereas molecular dynamics simulations provide insights into their dynamic properties. This “static–dynamic”³⁶ evaluation strategy can facilitate a comprehensive understanding and prediction of material properties from a microscopic perspective, thereby enabling the development of effective passivation strategies.

2,6-Di-tert-butyltetramethylphenol (BHT–OH–CH₃) is a chemical compound that has attracted major attention.^37–39 This compound's antioxidant properties contribute to the crystalline quality of perovskite films. Its tert-butyl groups resist water molecules in the environment, and its hydrophilic hydroxyl groups serve to passivate surface defects in perovskites. In this study, we examined the passivation mechanism of BHT–OH–CH₃ by conducting a “static–dynamic” strategy. We systematically analyzed the electronic structures of various passivators, the interfacial interactions between each passivator and MAPbI₃, and the water resistance of these passivators under humid conditions. We attempted to enhance the performance of perovskites by modifying their anchoring groups and to mitigate the influence of water molecules on the perovskite surface by extending their alkyl chains based on BHT molecules.

2. Computational details

Structural optimization and frequency calculations were conducted for individual passivated molecules through density functional theory based on B3LYP/6-31G(d,p)^40–42 using the Gaussian 09 software package.⁴³ A PM@MAPbI₃ adsorption model was established by positioning a PM on the surface of perovskites. The structural and electronic properties of this system were determined using the Vienna Ab initio Simulation Package together with the generalized gradient approximation and Perdew–Burke–Ernzerhof exchange–correlation functional.^44–46 The PBE functional was chosen without considering the spin–orbit coupling (SOC) effect because it eliminates accidental errors and has been shown to predict the band gaps of halide perovskite materials.^47,48 van der Waals interactions were incorporated using the supplemented DFT-D3 correction method.⁴⁹ Density functional theory calculations, along with X-ray diffraction studies, have confirmed that the MAPbI₃ (001) crystal face is the most stable configuration of perovskites.⁵⁰ This study focused on the PbI₂-termination of the perovskite surface, primarily because Lewis base additives have strong anchoring ability for uncoordinated lead.⁵¹ Consequently, the perovskite model consisted of three layers and had a 144-atom MAPbI₃ (2 × 2) (001) surface. During optimization, the upper layer was relaxed, but the lower two layers were fixed. A vacuum buffer with a thickness of 20 Å was introduced along the z-direction to prevent any interference between periodic crystal planes. The simulation cell had dimensions of 12.58 Å × 12.54 Å × 36.88 Å. A Γ-centered Monkhorst–Pack k-point mesh of size 3 × 3 × 1 was utilized. A plane wave kinetic energy cutoff of 400 eV was employed. In terms of structural optimization, the convergence threshold applied was 10⁻⁴ eV, and all forces had to be smaller than 0.02 eV Å⁻¹. A humid environment was simulated by including four water molecules at equal heights within the cell. The moisture-related stability of the H₂O/PM@MAPbI₃ system was evaluated through ab initio molecular dynamics (AIMD) simulations conducted using CP2K software; the NVT ensemble and Nosé–Hoover thermostat were employed for these simulations.^52–54 The target temperature was 300 K, and each trajectory was simulated for 20 ps, with the atomic time step being 1 fs. Snapshots of the AIMD trajectories were captured using Visual Molecular Dynamics software.⁵⁵

3. Results and discussion

3.1 Geometric structures and properties of PMs

Fig. 1 illustrates the molecular structure of BHT–OH–CH₃ and its derivatives. To increase the strength of the interaction between the PMs and the surface of the perovskite, we substituted the OH group of BHT–OH–CH₃ with various anchoring groups: SH, NH₂, CN, CHO, or COOH. Additionally, to enhance the coverage of PMs on the surface of the perovskite, we increased the length of the tert-butyl alkyl chain and modified the structures of BHT–OH–CH₃ and BHT–COOH–CH₃ to BHT–OH–C₂H₅ and BHT–COOH–C₂H₅, respectively.

The coverage of PMs on the perovskite surface has a crucial effect on the performance and stability of PSCs.⁵⁶ Therefore, determining the size of the area of contact between PMs and PbI₂-terminated surfaces is essential.⁵⁷ In this study, we used the Multiwfn program^58,59 to calculate the molecular sizes of the PMs and the PbI₂-terminated perovskite (Fig. 2) and to evaluate the coverage of the PMs on the surface of the perovskite. BHT–OH–CH₃ was found to have an equilibrium length of 11.85 Å and width of 8.96 Å, whereas the new PMs (BHT–SH–CH₃, BHT–NH₂–CH₃, BHT–CN–CH₃, and BHT–CHO–CH₃) had greater lengths and widths. Specifically, BHT–COOH–CH₃ had a length of 11.79 Å and a width of 10.22 Å, and surface PbI₂ had a length of 12.58 Å and a width of 9.41 Å. These results indicated that even when the anchoring group was modified, the PMs could not completely protect the perovskite surface. To address this limitation and further increase the durability of the perovskite, we extended the hydrophobic tert-butyl chain in BHT–OH–CH₃ and BHT–COOH–CH₃. This extension significantly increased the volumes of BHT–OH–C₂H₅ and BHT–COOH–C₂H₅, enabling their strong attachment to the surface of PbI₂, particularly for BHT–COOH–C₂H₅. This modification facilitated the formation of a robust, water-resistant PM barrier on the surface of the perovskite.

Electron density is a crucial factor determining the activity of electrophilic and nucleophilic sites within a molecule.⁵⁹ We used the molecular surface electrostatic potential (ESP) to predict potential reaction sites on various PMs and the surface of the perovskite. In Fig. 3, the red region represents a positive charge at the PbI₂-terminated surface, predominantly concentrated around four adjacent Pb atoms. By contrast, the light-blue region represents a negative charge localized on the I atoms on the surface of the perovskite. We discovered that for all PMs, the ESP was negative primarily near the heteroatoms. Except for BHT–SH–CH₃, the extremely negative ESP values of BHT–NH₂–CH₃, BHT–CN–CH₃, BHT–CHO–CH₃, and BHT–COOH–CH₃ were all lower than those of BHT–OH–CH₃. This finding was attributed to the lower electronegativity of S compared with N and O, which diminished the electron-withdrawing capability of the –SH group. The positive charges of these PMs were predominantly distributed around the hydrogen atoms of the anchoring and alkyl groups. A comparison of BHT–COOH–C₂H₅ and BHT–COOH–CH₃ revealed that a change in the length of the alkyl chain had a minimal effect on the ESP distribution. Taken together, these findings suggest that BHT–COOH–CH₃ and BHT–COOH–C₂H₅ exhibit stronger interactions with lead and iodine on the surface of the perovskite than does BHT–OH–CH₃.

3.2 Structure and adsorption energy of PM@MAPbI₃

Examining the interface characteristics of PM@MAPbI₃ systems is essential because these characteristics not only reflect a realistic environment but also elucidate the influence of PMs on the performance of perovskite materials.⁶⁰ Passivators should exhibit strong binding affinity with the perovskite surfaces to ensure long-term and effective protection. In this study, given the ESP distributions and molecular structural characteristics of the PMs, the passivators were oriented vertically on the surface of perovskites. This orientation enabled heteroatoms such as O, N, and S to coordinate with the Pb²⁺ ions of perovskites, while the H atom of the anchoring group was directed toward the electronegative I atom. The complete optimized configuration and detailed interface configuration of PM@MAPbI₃ are illustrated in Fig. S1 (ESI†) and Fig. 4, respectively. The Pb–O and H⋯I distances were measured as 2.93 and 3.27 Å, respectively, for BHT–OH–CH₃@MAPbI₃. The H⋯I bond length (3.15 Å) in BHT–SH–CH₃@MAPbI₃ and BHT–NH₂–CH₃@MAPbI₃ was slightly shorter than that in BHT–OH–CH₃@MAPbI₃. Additionally, the Pb–O and H⋯I bond lengths were further decreased in BHT–CHO–CH₃@MAPbI₃, BHT–COOH–CH₃@MAPbI₃, and BHT–COOH–C₂H₅@MAPbI₃. Notably, the Pb–N bond length was only 2.58 Å in BHT–CN–CH₃@MAPbI₃, attributable to the strong electron-withdrawing capability of the –CN group.

Stable adsorption of passivated molecules can considerably enhance the stability of the perovskite against humidity.⁶¹ Adsorption energy (E_ads) offers a means for quantifying the strength of the interaction between a passivated molecule and perovskite.²⁴ This energy is calculated as follows: E_ads = E_PM@MAPbI3 − E_MAPbI3 − E_PM, where E_PM@MAPbI3 and E_MAPbI3 are the total energy of MAPbI₃ slabs with and without passivated molecules, respectively, and E_PM is the energy of individual passivated molecules in a vacuum. A more negative E_ads value indicates greater adsorption strength. In this study, we discovered that BHT–OH–CH₃ exhibited a weak interaction with the surface of MAPbI₃, accounting for the unstable passivation effect observed in dynamics simulations. Therefore, to enhance the direct interaction between the PM and perovskite, we modified the anchoring groups and designed five new passivators. These new passivators had stronger interactions with the perovskite, leading to a stable passivation effect. Comparison of the adsorption energies of BHT–COOH–C₂H₅@MAPbI₃ (−1.80 eV) and BHT–COOH–CH₃@MAPbI₃ (−1.35 eV) on the surface of the perovskite revealed that extending the tert-butyl alkyl chain increased the strength of the adsorption. This effect was attributable to the increased number of H⋯I interaction, as illustrated by the dashed blue line in Fig. 4. A similar trend was discovered in the cases of BHT–OH–C₂H₅@MAPbI₃ (−1.24 eV) and BHT–OH–CH₃@MAPbI₃ (−0.78 eV). Of the eight passivators, BHT–COOH–C₂H₅ has the strongest adsorption on the surface of the perovskite because of the synergistic effect between the carboxyl anchoring groups and extended alkyl chains.

3.3 Electronic properties of PM@MAPbI₃

Fig. 5 illustrates the charge density differences and Bader charges of the PMs@MAPbI₃ systems, and it provides a visual representation of the interactions and electron transport characteristics of the PMs and the surface of the perovskite. The interactions primarily occur at the interface of the passivation systems. As shown in Fig. 5(a), the main interaction was observed between Pb and O in the hydroxyl group, with electron flow directed from O to Pb. The second interaction was that between I and H in the hydroxyl group, with the electron flow directed from I to H. Additionally, tert-butyl H had a minimal effect on the surface iodine atoms of the perovskite. In the case of BHT–CN–CH₃@MAPbI₃ [Fig. 5(d)], the main interaction region involved Pb and N in the cyano group; this was the strongest interaction across all systems and was attributable to the negative ESP of N in the cyano group (Fig. 3) and the short distance between N and Pb [Fig. 4(d)]. In BHT–CHO–CH₃@MAPbI₃, an interaction was discovered between O in the aldehyde group and Pb in addition to that between H in the aldehyde group and I. Notably, the interaction with I was stronger for the H in the carboxyl group in BHT–COOH–CH₃@MAPbI₃ [Fig. 4(f)] than for the H in the aldehyde group in BHT–CHO–CH₃@MAPbI₃ [Fig. 4(e)]. In addition, the extended alkyl chain in BHT–COOH–C₂H₅@MAPbI₃ increased the number of interactions between H and I. Therefore, of all systems, BHT–COOH–C₂H₅@MAPbI₃ interacted most strongly with the surface of the perovskite. Bader charge analysis revealed that electrons were transferred from the PMs to perovskite in all systems except BHT–OH–C₂H₅@MAPbI₃ in Fig. 5(g). Of all systems, BHT–CN–CH₃@MAPbI₃ exhibited the highest charge transfer, amounting to 0.1154 e⁻. This is because there is only charge transfer from heteroatom N to Pb in BHT–CN–CH₃@MAPbI₃, while for other systems, in addition to the charge transfer from the heteroatom (O or S) to Pb, there is also charge transfer from I to H. The directions of these two charge transfers are opposite, so the net charge transfer in other systems is less than that in BHT–CN–CH₃@MAPbI₃. According to the plane average charge difference in the z-direction (Fig. S2, ESI†), the redistribution of interface charge in PM@MAPbI₃ resulted in the formation of a built-in electric field. The new passivation system generated a stronger built-in electric field at the interface of the perovskite compared with that in BHT–OH–CH₃@MAPbI₃, and this stronger field is advantageous for carrier separation.⁶² The findings were consistent with the influence of various PMs on the charge density of the vertical Pb–I skeleton of the perovskite (Fig. 5).

The band edge of the perovskite considerably influences carrier dynamics. A higher density of states at the band edge is associated with faster carrier relaxation to the valence band (VB) and conduction band (CB), facilitating transfer to adjacent layers.⁶³ The projected density of states of bare MAPbI₃ was contributed by I, Pb, and MA (Fig. S3, ESI†). The VB maximum (VBM) and CB minimum (CBM) edges are predominantly composed of I and Pb orbitals, respectively. Moreover, MA made a minimal contribution to the band edge, and its contribution was primarily located deep within the VB, consistent with the findings from the literature survey.⁶⁴ For the passivation systems, we examine three components: the total density of states (TDOS) and the projected density of states (PDOS) of both MAPbI₃ and the PMs, as shown Fig. 6. We discovered that the PMs did not generate trap states and only caused a slight disturbance of the density of states of the perovskite. Consequently, the electronic properties of the perovskite were largely unchanged. To clearly illustrate the contribution of the PM, the contribution value was amplified by a factor of 10. As shown in Fig. 6(a)–(c), BHT–OH–CH₃, BHT–SH–CH₃, and BHT–NH₂–CH₃ heavily contributed to the VB edge, promoting the transfer of holes from the perovskite layer to the hole transport layer. By contrast, BHT–CN–CH₃ and BHT–CHO–CH₃ heavily contributed to the CB edge, promoting the transfer of electrons from the perovskite layer to the electron transport layer. Additionally, BHT–COOH–CH₃ and BHT–COOH–C₂H₅ contributed to the electron density and hole density within the VB and CB. Taken together, these findings indicate that the influence of the PMs on band edges is predominantly determined by the nature of the anchoring group.

Fig. 7 presents the calculated band structures of the passivated perovskite. The bandgap values for the PM@MAPbI₃ systems were determined to be 1.48 eV for BHT–OH–CH₃, 1.55 eV for BHT–SH–CH₃, 1.53 eV for BHT–NH₂–CH₃, 1.57 eV for BHT–CN–CH₃, 1.55 eV for BHT–CN–CH₃, and 1.54 eV for BHT–COOH–CH₃. Compared with the pristine structure (1.48 eV) shown in Fig. S3(b) (ESI†), all adsorption systems had a slightly larger bandgap with the exception of BHT–OH–CH₃@MAPbI₃. Typically, a greater bandgap indicates that the perovskite can absorb higher-energy photons, which is essential for enhancing the PCE of PSCs.⁶⁵ According to the CBM and VBM of the perovskite comprise Pb and I atoms in antibonding states,⁶⁶ we examined the first and second perpendicular lengths of the Pb–I bond at the adsorption sites of the perovskite before and after passivation, as well as the Pb–I–Pb angles at the equatorial and apical positions of the topmost layer of the perovskite (Fig. S4 and Table S1, ESI†). These phenomena collectively resulted in a smaller orbital overlap between Pb and I, leading to a lower VBM and CBM, with the VBM being more sensitive. Consequently, the bandgap of the novel systems was slightly increased.⁶³ Furthermore, the straight band introduced by BHT–OH–CH₃, BHT–NH₂–CH₃, and BHT–OH–C₂H₅ at −0.52, −0.30, and −0.30 eV, respectively, at the VB edge of the perovskite functioned as a carrier recombination center, indicating that these three passivators were suboptimal. The contribution of other PMs to the CB and VB resulted in a larger bandgap and the absence of trap states, which in turn inhibited electron–hole recombination in the perovskite and led to a higher open-circuit voltage.^67,68

Fig. S5 (ESI†) presents the charge distribution diagrams of the VBM − 1, VBM, CBM, and CBM + 1 for perovskite before and after passivation. In the pristine perovskite, the charge of the VBM − 1 and VBM was primarily localized on the top and bottom layers of the perovskite, specifically on the Pb and I atoms. In addition, the charge of the CBM was distributed across the Pb atoms on the upper and middle layers of the perovskite. However, each PM with a distinct anchoring group had a unique effect on the electron density distribution at the surface of the perovskite. In BHT–OH–CH₃@MAPbI₃, BHT–NH₂–CH₃@MAPbI₃, and BHT–OH–C₂H₅@MAPbI₃, the charge of the VBM − 1 was primarily located on the passivated molecules, accounting for the straight-line band structure observed in Fig. 7. With the exception of BHT–OH–CH₃, in the new passivation systems, the charge of the CBM was predominantly distributed on the Pb interlayer of the perovskite, whereas the charge distributions of the VBM and CBM + 1 closely resembled those of the original perovskite. These results indicated that the new PMs diminished the overlap between the VBM and CBM orbitals in the perovskite, thereby reducing nonradiative recombination and enhancing the photoelectric performance of the perovskite.⁶⁹

3.4 Molecular dynamics of PM@MAPbI₃ with water

AIMD simulations were conducted to evaluate the stability of MAPbI₃ when protected by various anchor group passivators and subjected to humid conditions. According to the literature, each passivation system comprises two PMs and four water molecules to prevent excess water molecules from forming a hydrogen bond network.^24,32,70 Fig. S6 displays selected snapshots from a 20 ps simulation of H₂O/PM@MAPbI₃. Fig. 8 depicts the variation in height of the oxygen atoms in the water molecules along the AIMD trajectory, with the heights of the passivated molecules indicated by dashed black lines. In BHT–OH–CH₃@MAPbI₃, at 5 ps, the water molecules began entering the perovskite from both sides of the passivated molecules. At 15 ps, one water molecule began to penetrate the interior of the perovskite, and the penetration was complete by 20 ps, as shown in Fig. S6(a) (ESI†). As depicted in Fig. 8(b)–(f), at least one water molecule moved beneath the passivation layer in almost all passivation systems, potentially leading to infiltration within the perovskite structure. This phenomenon was attributed to the incomplete PM coverage of the perovskite surface, with the edges of the perovskite being exposed. These findings are consistent with the projected PM coverage calculations (Fig. 2). To address this problem, we leveraged the advantages of alkyl chains, which are resistant to water^30,31 and increased molecular size, to design new molecules (BHT–OH–C₂H₅ and BHT–COOH–C₂H₅) using BHT–OH–CH₃ and BHT–COOH–CH₃. In this approach, the surface passivation achieved with BHT–OH–C₂H₅ and BHT–COOH–C₂H₅ blocked the intrusion of water molecules. This blocking effect was attributed to the increased steric hindrance and hydrophobicity resulting from the extended alkyl chains of the PMs. Taken together, the findings indicate that optimizing the coverage of PMs on the surface of perovskite is a feasible strategy for enhancing the stability of perovskite materials in humid environments.

To determine the differences between the various passivation systems, we examined the evolution of the distances between the heteroatoms (O, S, and N) in passivated molecules and the surface Pb atoms (Fig. S7, ESI†). We discovered that in BHT–OH–CH₃@MAPbI₃, water molecules entered the vicinity of the PMs at 5 ps, causing the passivator to drift. This drift resulted in a dramatic change in the O–Pb bond length after 8.94 ps, ultimately leading to a loss of the passivation effect. This phenomenon was due to the weak adsorption of BHT–OH–CH₃ on the surface of the perovskite. A similar phenomenon was discovered in BHT–NH₂–CH₃@MAPbI₃ and BHT–OH–C₂H₅@MAPbI₃. By contrast, in H₂O/BHT–CN–CH₃@MAPbI₃, H₂O/BHT–CHO–CH₃@MAPbI₃, and H₂O/BHT–COOH–CH₃@MAPbI₃, even when water molecules moved from the exposed sides to the lower layer of the passivated molecules, the O(N)–Pb bond lengths did not change considerably. This stability was primarily attributable to the ability of the PMs to stably adsorb on the perovskite surface. In the case of H₂O/BHT–COOH–C₂H₅@MAPbI₃, BHT–COOH–C₂H₅ drifted only slightly because of the PM's strong adsorption and high coverage on the perovskite surface. According to our AIMD simulations, in order for the passivation system to be effective, a variety of factors need to be considered such as strong anchoring groups to stabilize the PM on the perovskite surface, the coverage of the PM on the perovskite surface, and multiple interactions between the passivation molecules and the perovskite surface.

Lattice disorder increases the overlap of electron–hole wave functions, which in turn affects nonadiabatic electron–phonon coupling and has a detrimental effect on the efficiency and stability of PSCs.^71,72 Atomic fluctuations can be used to evaluate the stability of materials in the degradation studies of systems exposed to water, light, and oxygen.⁷³ When atomic fluctuations increase, the framework of the perovskite starts to deform considerably, thereby accelerating the process of degradation. In this study, we quantified the atomic fluctuations of H₂O/PM@MAPbI₃ during the simulation period (Fig. 9) and calculated the average standard deviation (σ) of the positions of MA, I, and Pb as follows:^74,75

where r_i,j represents the position of MA, I, or Pb in particle i at step j; r_i represents the average position in the AIMD trajectory of particle i after m steps; and n is the total number of MA, I, or Pb particles. Overall, among all systems, MA exhibited the least volatility, whereas I exhibited the most volatility. Compared with BHT–OH–CH₃, the newly designed PMs had lower Pb volatility, primarily because of the strong interaction between heteroatoms in the new anchoring groups and Pb on the perovskite surface. In BHT–NH₂–CH₃, BHT–CN–CH₃, and BHT–COOH–CH₃, the I volatility was abnormally high because of the limited coverage of PMs on the perovskite surface. Water molecules moved down both sides of the PMs and approached the perovskite surface. This resulted in major fluctuations in the position of I within the passivation system, which in turn led to the detachment of I from the surface. By contrast, in BHT–OH–C₂H₅ and COOH–C₂H₅, the longer alkyl chain resulted in the volatility of I and Pb being markedly reduced. The heteroatom of the anchoring group coordinated with Pb, whereas H formed hydrogen bonds with iodine, thereby stabilizing Pb and I, inhibiting the formation of iodine vacancy defects and mitigating the relaxation effect.⁶³ Taken together, these findings indicate that extending the alkyl chain of a PM enhances the structural integrity of the perovskite and increases its stability.

4. Conclusions

In this study, we systematically examined eight PMs with similar structures but different functional groups and branch chain lengths to determine their interactions with the surface of the perovskite and their resistance to water molecules. Using the “static–dynamic” strategy, we discovered that various functional groups can modulate the interaction energy between PMs and the perovskite surface, while modifying the PM chain length can influence the PM coverage of the perovskite surface. Of the eight passivators examined in this study, BHT–COOH–C₂H₅ showed the most effective passivation on the surface of the perovskite. This is attributed to the fact that BHT–COOH–C₂H₅ exhibited the strongest interaction and highest coverage on the surface of the perovskite, and multiple interactions between the passivation molecule and the perovskite surface, giving it the potential to enhance the PCE and stability of perovskite cell devices. Overall, appropriate fine-tuning of the passivator structure may help improve the stability of PSCs and accelerate their commercialization.

Author contributions

Qingxia Zhao: formal analysis, investigation, and writing – original draft. Xiufang Hou: methodology and writing – review and editing. Wenjing Xie: methodology and supervision. Leixing Luo: supervision and validation. Shanshan Xu: formal analysis and investigation. Xiaoming Song: software and supervision. Qingbo Wei: resources and conceptualization.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the ESI.†

Acknowledgements

This work is financially supported by the National Natural Science Foundation of China (22169022) and the Training Program of Innovation and Entrepreneurship for Undergraduates (D2024242).

References

1. A. E. Maughan, A. M. Ganose, M. M. Bordelon, E. M. Miller, D. O. Scanlon and J. R. Neilson, J. Am. Chem. Soc., 2016, 138, 8453–8464 CrossRef CAS PubMed .
2. R. Raj, R. Singh and M. Guin, ChemistrySelect, 2023, 8(45), e202303550 CrossRef CAS .
3. G. G. Njema and J. K. Kibet, Next Nanotechnol., 2025, 7, 100102 CrossRef .
4. S. Monga, M. Jain, C. Draxl and S. Bhattacharya, Phys. Rev. Mater., 2024, 8, 105403 CrossRef CAS .
5. B. Preeti, J. Manjari, S. Sajjan and B. Saswata, J. Phys. Chem. Lett., 2022, 13, 11655–11662 CrossRef PubMed .
6. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051 CrossRef CAS PubMed .
7. National Renewable Energy Laboratory, Best Research-Cell Efficiencies Chart, https://www.nrel.gov/pv/cell-efficiency.html.
8. G. Giorgi, J.-I. Fujisawa, H. Segawa and K. Yamashita, J. Phys. Chem. Lett., 2013, 4, 4213–4216 CrossRef CAS PubMed .
9. Z. Hu, Z. Lin, J. Su, J. Zhang, J. Chang and Y. Hao, Sol. RRL, 2019, 3, 1900304 CrossRef .
10. X. Zhao, P. Zhang, T. Liu, B. Tian, Y. Jiang, J. Zhang, Y. Tang, B. Li, M. Xue, W. Zhang, Z. Zhang and W. Guo, Science, 2024, 385, 433–438 CrossRef CAS PubMed .
11. Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao and J. Huang, Science, 2015, 347, 967–970 CrossRef CAS PubMed .
12. M. Abbas, B. Cai, J. Hu, F. Guo, Y. Mai and X.-C. Yuan, ACS Appl. Mater. Interfaces, 2021, 13, 46566–46576 CrossRef CAS PubMed .
13. N. Aristidou, I. Sanchez-Molina, T. Chotchuangchutchaval, M. Brown, L. Martinez, T. Rath and S. A. Haque, Angew. Chem., Int. Ed., 2015, 127, 8326–8330 CrossRef .
14. L. Zhang and P. H.-L. Sit, J. Mater. Chem. A, 2017, 5, 9042–9049 RSC .
15. E. Mosconi, J. M. Azpiroz and F. De Angelis, Chem. Mater., 2015, 27, 4885–4892 CrossRef CAS .
16. M. T. Mbumba, D. M. Malouangou, J. M. Tsiba, M. W. Akram, L. Bai, Y. Yang and M. Guli, J. Mater. Chem. C, 2021, 9, 14047–14064 RSC .
17. S. Yang, Y. Wang, P. Liu, Y.-B. Cheng, H. J. Zhao and H. G. Yang, Nat. Energy, 2016, 1, 15016 CrossRef CAS .
18. Y. Zou, R. Guo, A. Buyruk, W. Chen, T. Xiao, S. Yin, X. Jiang, L. P. Kreuzer, C. Mu and T. Ameri, ACS Appl. Mater. Interfaces, 2020, 12, 52643–52651 CrossRef CAS PubMed .
19. X. Zheng, B. Chen, J. Dai, Y. Fang, Y. Bai, Y. Lin, H. Wei, X. C. Zeng and J. Huang, Nat. Energy, 2017, 2, 17102 CrossRef CAS .
20. J. K. Nam, S. U. Chai, W. Cha, Y. J. Choi, W. Kim, M. S. Jung, J. Kwon, D. Kim and J. H. Park, Nano Lett., 2017, 17, 2028–2033 CrossRef CAS PubMed .
21. P. Gao, A. R. Bin Mohd Yusoff and M. K. Nazeeruddin, Nat. Commun., 2018, 9, 5028 CrossRef PubMed .
22. Q. Jiang, Y. Zhao, X. Zhang, X. Yang, Y. Chen, Z. Chu, Q. Ye, X. Li, Z. Yin and J. You, Nat. Photonics, 2019, 13, 460–466 CrossRef CAS .
23. F. Zhang and K. Zhu, Adv. Energy Mater., 2020, 10, 1902579 CrossRef CAS .
24. W. Zhang, Q.-S. Li and Z.-S. Li, Appl. Surf. Sci., 2021, 563, 150267 CrossRef CAS .
25. N. K. Noel, A. Abate, S. D. Stranks, E. S. Parrott, V. M. Burlakov, A. Goriely and H. J. Snaith, ACS Nano, 2014, 8, 9815–9821 CrossRef CAS PubMed .
26. L. Zhang, X. Liu, J. Li and S. McKechnie, Sol. Energy Mater. Sol. Cells, 2018, 175, 1–19 CrossRef CAS .
27. K. Zou, Q. Li, J. Fan, H. Tang, L. Chen, S. Tao, T. Xu and W. Huang, ACS Mater. Lett., 2022, 4, 1101–1111 CrossRef CAS .
28. R. Wang, J. Xue, K.-L. Wang, Z.-K. Wang, Y. Luo, D. Fenning, G. Xu, S. Nuryyeva, T. Huang and Y. Zhao, Science, 2019, 366, 1509–1513 CrossRef CAS PubMed .
29. X. Li, W. Sheng, X. Duan, Z. Lin, J. Yang, L. Tan and Y. Chen, ACS Appl. Mater. Interfaces, 2021, 14, 34161–34170 CrossRef PubMed .
30. T. Y. Wen, S. Yang, P. F. Liu, L. J. Tang, H. W. Qiao, X. Chen, X. H. Yang, Y. Hou and H. G. Yang, Adv. Energy Mater., 2018, 8, 1703143 CrossRef .
31. W.-Q. Wu, Z. Yang, P. N. Rudd, Y. Shao, X. Dai, H. Wei, J. Zhao, Y. Fang, Q. Wang and Y. Liu, Sci. Adv., 2019, 5, eaav8925 CrossRef CAS PubMed .
32. W. Zhang, Q. S. Li and Z. S. Li, Adv. Mater. Interfaces, 2022, 9, 2101881 CrossRef CAS .
33. J. Xie, K. Yan, H. Zhu, G. Li, H. Wang, H. Zhu, P. Hang, S. Zhao, W. Guo and D. Ye, Sci. Bull., 2020, 65, 1726–1734 CrossRef CAS PubMed .
34. M. Wang, H. Sun, M. Wang, L. Meng and L. Li, Adv. Mater., 2024, 36, 2306415 CrossRef CAS PubMed .
35. C. Li, X. Wang, E. Bi, F. Jiang, S. M. Park, Y. Li, L. Chen, Z. Wang, L. Zeng and H. Chen, Science, 2023, 379, 690–694 CrossRef CAS PubMed .
36. R. Guan, A. Shi, X. Zhang, B. Wang, Y. Li and X. Niu, ACS Catal., 2024, 14, 18580–18589 CrossRef CAS .
37. Y. Cheng, Q. Wei, N. Wang, Z. Ye, Y. Zhao, Q. Wang, D. Chu, L. Zan, F. Fu and Y. Liu, Chin. Chem. Lett., 2023, 34, 107933 CrossRef CAS .
38. L. Jin, Y. Huang, Z. Ma, H. Mao, Y. Li, H. Yu and C. Peng, NANO, 2019, 14, 1950022 CrossRef CAS .
39. S. Kumar, Y. Choi, S.-H. Kang, N. K. Oh, J. Lee, J. Seo, M. Jeong, H. W. Kwon, S. I. Seok and C. Yang, ACS Appl. Mater. Interfaces, 2019, 11, 38828–38837 CrossRef CAS PubMed .
40. K. Raghavachari, Theor. Chem. Acc., 2000, 103, 361–363 Search PubMed .
41. C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789 CrossRef CAS PubMed .
42. P. Hobza, J. Šponer and T. Reschel, J. Comput. Chem., 1995, 16, 1315–1325 CrossRef CAS .
43. G. W. T. M. J. Frisch, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman,G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,H. Nakai, T. Vreven, J. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken,C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision E.01, Gaussian Inc., Wallingford, CT, 2013 Search PubMed .
44. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169 CrossRef CAS PubMed .
45. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15–50 CrossRef CAS .
46. G. Kresse, J. Non-Cryst. Solids, 1995, 192, 222–229 CrossRef .
47. W.-J. Yin, T. Shi and Y. Yan, J. Phys. Chem. C, 2015, 119, 5253–5264 CrossRef CAS .
48. L. Zhang, F. Yu, L. Chen and J. Li, Appl. Surf. Sci., 2018, 443, 176–183 CrossRef CAS .
49. S. Grimme, J. Comput. Chem., 2006, 27, 1787–1799 CrossRef CAS PubMed .
50. L. Zhang, L. Xu, J. Su, D. Lu and J. Li, J. Phys. Chem. C, 2017, 121, 24612–24617 CrossRef CAS .
51. J. Jiang, W. Lin, E. Liu, J. Sha and L. Ma, Appl. Surf. Sci., 2022, 597, 153716 CrossRef CAS .
52. J. Hutter, M. Iannuzzi, F. Schiffmann and J. VandeVondele, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2014, 4, 15–25 CAS .
53. J. VandeVondele, M. Krack, F. Mohamed, M. Parrinello, T. Chassaing and J. Hutter, Comput. Phys. Commun., 2005, 167, 103–128 CrossRef CAS .
54. B. G. Lippert, J. Hutter and M. Parrinello, Mol. Phys., 2010, 92, 477–488 CrossRef .
55. W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics, 1996, 14, 33–38 CrossRef CAS PubMed .
56. S. M. Park, A. Abtahi, A. M. Boehm and K. R. Graham, ACS Energy Lett., 2020, 5, 799–806 CrossRef CAS .
57. H. Chen, C. Liu, J. Xu, A. Maxwell, W. Zhou, Y. Yang, Q. Zhou, A. S. Bati, H. Wan and Z. Wang, Science, 2024, 384, 189–193 CrossRef CAS PubMed .
58. H. Zhang, A. C. Reber, L. Geng, D. Rabayda, H. Wu, Z. Luo, J. Yao and S. N. Khanna, CCS Chem., 2019, 1, 571–581 CrossRef CAS .
59. T. Lu and F. Chen, J. Comput. Chem., 2012, 33, 580–592 CrossRef CAS PubMed .
60. X. Hou, W. Zhang and Q.-S. Li, Phys. Chem. Chem. Phys., 2023, 25, 13383–13392 RSC .
61. H. Zhang, Y. Wu, C. Shen, E. Li, C. Yan, W. Zhang, H. Tian, L. Han and W. H. Zhu, Adv. Energy Mater., 2019, 9, 1803573 CrossRef .
62. S. Wang, A. Wang, X. Deng, L. Xie, A. Xiao, C. Li, Y. Xiang, T. Li, L. Ding and F. Hao, J. Mater. Chem. A, 2020, 8, 12201–12225 RSC .
63. J. Jiang, W. Lin, E. Liu, J. Sha and L. Ma, Appl. Surf. Sci., 2022, 597, 153716 CrossRef CAS .
64. R. Long, W. Fang and O. V. Prezhdo, J. Phys. Chem. Lett., 2016, 7, 3215–3222 CrossRef CAS PubMed .
65. T. Nie, Z. Fang, X. Ren, Y. Duan and S. Liu, Nano-Micro Lett., 2023, 15, 70 CrossRef CAS PubMed .
66. S. Tao, I. Schmidt, G. Brocks, J. Jiang, I. Tranca, K. Meerholz and S. Olthof, Nat. Commun., 2019, 10, 2560 CrossRef PubMed .
67. V. Carla and G. Feliciano, J. Phys. Chem. C, 2015, 119, 25209–25219 CrossRef .
68. R. Long, J. Liu and O. V. Prezhdo, J. Am. Chem. Soc., 2016, 138, 3884–3890 CrossRef CAS PubMed .
69. N. Chen, W. Zhang and Q.-S. Li, Nanoscale, 2023, 15, 10730–10739 RSC .
70. W. Zhang, Q.-S. Li and Z.-S. Li, J. Phys. Chem. Lett., 2022, 13, 6686–6693 CrossRef CAS PubMed .
71. J. He, W.-H. Fang, R. Long and O. V. Prezhdo, Nano Energy, 2021, 79, 105491 CrossRef CAS .
72. Z. Zhang and R. Long, J. Phys. Chem. Lett., 2019, 10, 3433–3439 CrossRef CAS PubMed .
73. J. He, W.-H. Fang, R. Long and O. V. Prezhdo, J. Am. Chem. Soc., 2020, 142, 14664–14673 CrossRef CAS PubMed .
74. Z. Guo, J. Wang and W.-J. Yin, Energy Environ. Sci., 2022, 15, 660–671 RSC .
75. L. Qiao, X. Sun and R. Long, J. Phys. Chem. Lett., 2019, 10, 672–678 CrossRef CAS PubMed .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5nj00849b

This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2025