Self‐Limiting Ultrathin Heterocycle‐Bridged Interface Enabling Efficient p–i–n PbS Quantum Dot Solar Cells

ABSTRACT

Hole‐transporting self‐assembled monolayers (SAMs) have driven rapid efficiency gains in solution‐processed p–i–n solar cells over the past few years. By contrast, while efficient electron extraction is equally crucial for device operation, electron‐transporting SAMs remain scarcely explored, thereby constraining further improvements in device performance. Here, we present an electron‐extraction‐modulating self‐limiting ultrathin molecular layer strategy for PbS quantum dot (QD) solar cells. The designed thiophene‐based molecules feature thiol anchors that coordinate with surface Pb^{2 +} ions to suppress nonradiative recombination, while their electron‐rich aromatic rings engage in π–π interactions with the top C₆₀ layer to enhance electron extraction. Critically, their high volatility enables the removal of excess molecules during annealing, yielding a compact, ultrathin molecular modification at the interface. This interface engineering delivers a power conversion efficiency of 14.47% in p–i–n PbS QD solar cells‐the highest value reported to date. These results provide a robust and scalable pathway for applying self‐limiting ultrathin molecular layer as a powerful paradigm for interfacial design.

A self‐limiting thermal evaporation strategy is developed to construct a heterocycle‐bridged interface using volatile functional molecules for lead sulfide quantum dot solar cells. This approach successfully removes weakly physisorbed excess molecules, creating a nanoscale ultrathin layer that enhances charge extraction. Consequently, an outstanding power conversion efficiency of 14.47% is achieved for p–i–n photovoltaic device.

1. Introduction

Colloidal lead sulfide (PbS) quantum dots (QDs) have emerged as a promising candidate material for next‐generation photovoltaics, owing to their tunable bandgap [1, 2, 3] solution processability [4, 5, 6, 7, 8], and potentially low manufacturing costs [9]. In particular, PbS QDs possess a unique photon‐harvesting capability in the 1100–2500 nm range [10, 11, 12], beyond the absorption limit of mainstream photovoltaic materials such as perovskites, organics and crystalline silicon which allows them to complement wide‐bandgap absorbers and maximize solar spectrum utilization [13, 14, 15]. This spectral advantage makes PbS QDs an attractive narrow‐bandgap candidate for tandem solar cells.

Recent advances in p‐i‐n structured perovskite and organic solar cells have markedly accelerated the development of high‐efficiency tandem architectures [16, 17, 18, 19, 20, 21], owing to the superior process compatibility and reduced parasitic absorption of the p‐i‐n configuration compared with the conventional n‐i‐p design [22, 23, 24]. To date, n‐i‐p architectures have achieved a laboratory record power conversion efficiency (PCE) of 15.45% [25] and a state‐of‐the‐art certified PCE of 13.40% [9]. However, the efficiencies of p‐i‐n structured PbS QDSCs have largely remained below 10% [26, 27, 28, 29, 30, 31, 32, 33]. Very recently, our group realized a notable breakthrough by introducing a novel hole transport layer (HTL), which enhanced the certified PCE of p‐i‐n PbS QD solar cells to 13.60% [34]. This milestone, achieved for the first time, surpasses the performance of the widely studied n‐i‐p architecture, thereby opening new avenues for PbS QD photovoltaics and their integration into tandem devices.

The recent rapid progress of p‐i‐n perovskite and PbS QD solar cells has largely benefited from the discovery of a family of carbazole‐based molecules with phosphonic acid groups (2PACz, MeO‐2PACz, and Me‐4PACz) that serve as SAM hole‐transporting materials [24, 35, 36, 37, 38, 39]. The phosphonic acid moiety can tightly anchor to oxide substrates (ITO or NiO) through a dehydration reaction, enabling an oriented assembly at the interface [40, 41]. Their nearly monomolecular‐layer characteristics facilitate the formation of an oriented interfacial dipole, thereby promoting efficient hole transport [42, 43, 44]. In contrast to the remarkable advances in hole‐transporting SAM molecules, highly efficient electron‐transporting SAMs remain largely unexplored, which may represent a critical avenue for further efficiency improvements.

Herein, we introduce a series of electron‐extraction‐modulating self‐limiting ultrathin molecular featuring thiophene backbones and thiol anchoring groups (–SH) at the PbS/C₆₀ interface. The –SH group strongly coordinates with unsaturated Pb^{2 +} ions on the PbS QD surface, thereby suppressing nonradiative recombination. Meanwhile, the electron‐rich heteroaromatic ring engages in π–π stacking with C₆₀, enhancing interfacial contact and facilitating efficient electron extraction. More importantly, owing to their high volatility, excess molecules that do not bind to the QD surface are readily removed during annealing. Benefiting from the stark contrast in thermal stability between weakly physisorbed and strongly chemisorbed molecular states, a nanoscale ultrathin molecular layer can be spontaneously established during mild annealing. Through this self‐limiting strategy, we achieved a PCE of 14.47% in PbS QD solar cells, representing the highest efficiency reported thus far for p‐i‐n devices. This advance not only sets a new performance benchmark for p‐i‐n PbS QD solar cells but also offers a viable pathway toward scalable and reliable device fabrication, benefiting from the unique self‐limiting ultrathin molecular layer characteristics.

2. Results and Discussion

2.1. Interface Engineering Mechanism Analysis

We employed π‐conjugated heteroaromatic molecules containing thiol groups to construct an effective interfacial modification layer. To finely tune the molecular properties, we designed derivatives based on the mercaptothiophene (MT) framework by introducing one and two nitrogen atoms into the thiophene five‐membered ring, thereby forming mercaptothiazole (MTZ) and mercaptothiadiazole (MTD), respectively. The electrostatic potentials (ESP) of the dipolar molecules were first calculated to gain the charge distribution within these molecules. As shown in Figure 1a, ESP analysis indicates that all three molecules possess high polarity and localized negative regions near the thiol and adjacent heteroatoms, suggesting strong electron‐donating characteristics and favorable interaction with Pb^{2 +} via Lewis acid‐base coordination. The density functional theory (DFT) calculations were further performed to evaluate adsorption energies of molecules on the PbS surface (Figures 1b and S1). The adsorption strength follows the trend MTD > MTZ > MT, indicating that additional nitrogen atoms contribute to enhanced coordination with surface Pb^{2 +} ions and a more stable interfacial dipole configuration.

FIGURE 1.

(a) Electrostatic potential distributions of MT (C₄S₂H₄), MTZ (C₃NS₂H₃), and MTD (C₂N₂S₂H₂) molecules. The atomic colors are as follows: S (yellow), N (blue), C (gray), H (white). (b) Calculated binding energies of MT, MTZ, and MTD molecules adsorbed on the PbS QD active layer. (c) Charge density difference of C₆₀ anchoring onto PbS (111) surfaces passivated by MT, MTZ, and MTD molecules, as well as the pristine PbS (111) surface (Control). Yellow and cyan represent charge accumulation (negative charge) and depletion (positive charge), respectively.

We further study the effect of these dipolar molecules on interfacial charge transport. Charge density difference plots at the PbS/C₆₀ interface (Figure 1c) reveal the influence of molecular dipoles on interfacial charge redistribution. Compared to the unmodified interface, all dipole‐treated systems exhibit significantly increased charge overlap, indicative of improved electronic coupling. Notably, the MTD‐modified interface shows the strongest interfacial interaction, consistent with its highest adsorption energy and optimized molecular orientation. The interfacial electronic coupling was further quantified through charge density difference calculations (Figure S2). The net electron transfer (q) at the interface increases from 0.225 e for the control sample to 0.927, 0.822, and 1.113 e for the MT, MTZ, and MTD modified systems, respectively. This substantial enhancement in q values confirms that the introduction of specific dipolar molecules effectively modulates the interfacial electronic environment, facilitating more efficient charge density redistribution and coupling between the PbS QDs and the C₆₀ layer.

These results collectively demonstrate that the introduction of dipolar molecules facilitates interfacial charge extraction, while the incorporation of nitrogen atoms into the thiophene framework strengthens interfacial binding and further enhances charge transfer. Such dipole‐assisted interface engineering provides a promising strategy to reinforce the built‐in electric field and thereby improve the overall performance of PbS QD solar cells.

2.2. Device Performance Analysis of Dipolar Heteroaromatic Molecule Modification

To evaluate the performance modulation of three dipole molecules (MT, MTZ, and MTD) in practical devices, we fabricated PbS QD solar cells with a p‐i‐n device architecture as illustrated in Figure 2a. The control device delivered a PCE of 12.30%, consistent with our previous reports using the same device architecture [34]. Incorporation of dipolar molecules consistently resulted in pronounced improvements in photovoltaic performance. Specifically, MT‐ and MTZ‐modified devices achieved champion PCEs of 13.27% and 13.70%, respectively (Table 1). In contrast as shown in Figure 2b, the MTD‐modified devices exhibited the most significant performance enhancement, with a champion PCE of 14.47%, which is the highest reported efficiency for PbS QD solar cells with a p‐i‐n structure (Figure 2c). The key performance metrics for the MTD device include a V _oc of 0.69 V, J _sc of 30.97 mA cm^{− 2}, and FF of 67.70%, all superior to the control values (V _oc: 0.65 V, J _sc: 29.53 mA cm^{− 2}, FF: 64.08%, Figures S3 and S4). The average PCE was also improved from 12.11% (control) to 14.21%, with the device distribution concentrated above 14%, demonstrating excellent uniformity and reproducibility. Beyond initial performance, the operational stability of the MTD‐modified devices was evaluated under continuous 1‐sun illumination (Figure S5). The MTD‐modified device retained over 98% of its initial PCE after 24 h of continuous operation.

FIGURE 2.

(a) Schematic illustration of the p‐i‐n device architecture. (b) J–V curves of the best‐performing devices for the control and MTD‐treated samples, with the inset showing the corresponding PCE distributions. (c) Summarized PCE of reported p‐i‐n PbS CQD solar cells. (d) Statistical PCE distribution of devices fabricated with various MTD concentrations. (e) Schematic representation of molecular distribution on the MTD‐treated PbS surface after annealing. (f) Optical images of the MTD‐treated PbS films during annealing at 55 °C for different durations (30‐150 s).

TABLE 1.

Photovoltaic parameters of devices with different molecules.

		V OC (V)	J SC (mA cm−2)	FF (%)	PCE (%)
PbS	Champion	0.65	29.53	64.08	12.30
	Average	0.64 ± 0.02	29.01 ± 0.54	63.68 ± 0.80	12.11 ± 0.14
PbS/MT	Champion	0.66	30.34	66.16	13.27
	Average	0.65 ± 0.01	29.28 ± 0.53	65.34 ± 0.77	13.07 ± 0.17
PbS/MTZ	Champion	0.67	30.61	66.79	13.70
	Average	0.66 ± 0.02	29.68 ± 0.48	65.58 ± 0.79	13.62 ± 0.12
PbS/MTD	Champion	0.69	30.97	67.70	14.47
	Average	0.68 ± 0.01	30.57±0.46	67.68 ± 0.62	14.21 ± 0.15

For interfacial dipole modification, the concentration of the interfacial layer is critical, with the ideal scenario being the formation of a self‐limiting ultrathin molecular layer. In many cases, this process is highly sensitive to fabrication parameters: excessive concentrations may introduce transport barriers, whereas insufficient concentrations result in incomplete coverage. Such a narrow optimization window not only restricts device performance but also compromises reproducibility. We systematically investigated the effect of MTD deposition concentration on device performance. As shown in Figure 2d, the efficiency increased steadily with increasing MTD concentration, reaching an optimum at 1 mg mL^{− 1}, which likely corresponds to the formation of a compact self‐limiting ultrathin molecular layer. Remarkably, further increases in concentration did not degrade performance; instead, the efficiency plateaued up to 15 mg mL^{− 1}. This saturation behavior suggests a self‐limiting adsorption mechanism during device fabrication.

We hypothesize that MTD molecules at the interface form chemical bonds with PbS through thiol groups, establishing the self‐limiting ultrathin molecular layer, while excess unbound MTD molecules evaporate during the annealing process (at 55°C), as illustrated in Figure 2e. Consequently, the effective dipole layer remains confined to a nanoscale ultrathin thickness, while surface coverage saturates with increasing concentration. This mechanism is further supported by visual observations during the annealing process (Figure 2f). Immediately after spin‐coating, when the precursor concentration reached 10 mg mL⁻¹, the MTD‐modified PbS QD films exhibited a characteristic snowflake‐like matte appearance, likely due to the presence of excess unbound molecular residues. As annealing progressed from 30 to 150 s, these features gradually disappeared, and the film surface evolved into a smooth, uniform PbS QD layer with a glossy finish. Figure S6 further substantiates that MTD is removed by volatilization under low‐temperature annealing. This visual transition reflects the evaporation of unbound molecules and the formation of a thin and compact interfacial dipole layer. To quantify the thickness of this interfacial layer, atomic force microscopy (AFM) step‐height measurements were performed by creating a step edge between the bare PbS and MTD‐modified regions (Figure S7). The measured height difference is approximately 2 nm. This nanoscale dimension confirms the formation of an ultrathin molecular layer on the QD surface. The thermally driven reorganization and selective retention of chemically anchored MTD molecules confirm the self‐limiting nature of the thickness. This behavior not only facilitates reproducible film quality but also supports potential for large‐area device fabrication.

2.3. Interfacial Chemical Binding Analysis

To elucidate the chemical interactions of dipolar molecules on PbS QD films, we conducted systematic characterizations by x‐ray photoelectron spectroscopy (XPS) and Fourier‐transform infrared spectroscopy (FTIR). As shown in high‐resolution XPS spectra of the N 1s region (Figure 3a), no nitrogen signals were observed in the pristine PbS control and MT samples, whereas significant N 1s peaks appeared after MTZ and MTD modification, indicating the effective binding of dipolar molecules on the PbS film surface. Further analysis of the Pb 4f spectra (Figure 3b) showed characteristic Pb peaks at 137.7 eV (Pb 4f _7/2) and 142.6 eV (Pb 4f _5/2) in the control sample. Upon introduction of dipolar molecules, a clear shift toward lower binding energies occurred, specifically by 0.14 eV (MT), 0.15 eV (MTZ), and 0.20 eV (MTD). This binding‐energy shift indicates increased electron density around Pb^{2 +} ions, demonstrating significant electron coupling between molecules and surface Pb atoms. Notably, the concurrent anti‐directional shift of the S 2p peaks (Figure S8) suggests that the electronic binding energy changes are dominated by localized charge redistribution. Furthermore, the slight broadening of the Pb 4f peak (FWHM from 1.04 to 1.11 eV) reflects newly coordinated surface Pb atoms. To evaluate the thermal stability of the MTD molecular layer, XPS measurements were performed under continuous annealing at 55°C (Figure S9). The N/Pb atomic ratio exhibits a distinct two‐stage evolution: a rapid initial decline within the first 5 min, attributed to the desorption of loosely bound physisorbed residues followed by a stable plateau that persists up to 60 min (Table S1). This constant ratio provides direct evidence of a robust, chemically anchored ultrathin molecular layer that remains intact under prolonged thermal stress. To further investigate the chemical adsorption of dipolar molecules on PbS surfaces, FTIR spectra were measured (Figures 3c,d). MTD is characterized by distinct vibrational bands at 2808 and 1046 cm⁻¹, ascribed to the ‐SH and N–N stretching modes, respectively. Additionally, a specific peak at 1105 cm⁻¹, corresponding to the C═S bond, confirms the presence of the MTD tautomeric (thione) structure [45]. The MTD molecules exist in a dynamic tautomeric equilibrium as illustrated below:

FIGURE 3.

High‐resolution XPS spectra of N 1s (a) and Pb 4f (b) for the control PbS film and PbS films treated with MT, MTZ, and MTD molecules. FTIR spectra tracking the chemical interaction in PbS/MTD films: (c) The disappearance of the S–H stretching vibration and (d) the evolution of C═S and N–N stretching vibrations for PbS, PbS/MTD with varying annealing times (0, 5, 10 min), and MTD.

Theoretical and experimental insights suggest that the thiol form possesses a superior binding affinity for the PbS surface, enabling it to anchor firmly via the formation of stable Pb–S covalent bonds. Conversely, the thione form interacts weakly with the quantum dots. Upon prolonging the annealing time of MTD on the PbS surface, the ‐SH signal progressively diminishes and ultimately vanishes, while the N–N peak shifts to 1065 cm⁻¹. These spectral evolutions indicate the effective coordination of MTD with QDs, resulting in the formation of stable Pb‐S covalent bonds. Furthermore, the concurrent attenuation of the C═S peak intensity suggests a reduction in the tautomeric isomer content. This can be attributed to the fact that during the thermal annealing process, the unreacted thione species either volatilize due to the lack of chemical fixation or convert into the thiol form to participate in surface coordination. Collectively, these observations demonstrate that the MTD molecules undergo a significant structural transformation during their interaction with the lead sulfide surface (Figure S10). Coupled with the electrostatic potential (ESP) distributions obtained from first‐principles calculations (Figure 1a), these experimental results strongly support the notion that the polar nitrogen‐containing aromatic structure in MTD molecules facilitates electron localization, thereby enabling strong and stable adsorption onto PbS surfaces.

2.4. Surface Morphology and Energy Level Modulation Analysis

To further elucidate the effects of dipolar molecules on the surface morphology and electronic structure of PbS QD films, we first characterized the surface topography of films treated with different molecules using atomic force microscopy (Figure S11). The root‐mean‐square (RMS) roughness (Ra) of the pristine PbS film is 1.69 nm. Following dipolar molecule modification, the film surface roughness was slightly reduced. The influence of the interfacial dipole layer on the growth of the subsequent C₆₀ electron transport layer was further evaluated via AFM (Figure S12). The C₆₀ film deposited on the pristine PbS surface exhibits a Ra of 7.42 nm. Upon modification with dipolar molecules, the Ra of the C₆₀ layer is slightly reduced to 7.10 nm (MT), 6.93 nm (MTZ), and 6.68 nm (MTD), respectively. This morphological refinement indicates that the incorporation of dipolar molecules also promotes a relatively more uniform and compact C₆₀ coverage. Subsequently, Kelvin probe force microscopy (KPFM) was employed to investigate the surface potential distribution and fluctuations on PbS films modified with different dipolar molecules (Figure 4a). Compared to the control sample (−266 ± 7.64 mV), the samples covered by dipolar molecules exhibited higher surface potentials. Notably, the surface potential of the MTD‐treated sample increased most significantly (−165 ± 4.27 mV). This enhancement indicates that dipolar molecules introduce an effective electrostatic field on the PbS surface, which may promote electron transport to ETL [46]. The statistical distributions of surface potentials for various samples are presented in Figure 4b. The control PbS QD film exhibited a broad distribution with a full width at half maximum (FWHM) of 15.27 mV. In contrast, the dipole‐modified samples showed narrower potential distributions, with FWHM values decreasing to 11.61 mV (MT), 10.68 mV (MTZ), and 8.54 mV (MTD), respectively. These results clearly indicate that dipolar molecule modifications significantly reduce the surface potential heterogeneity, resulting in a more uniform surface electric field environment (Figure S13). This electronic uniformity is further corroborated by FTIR mapping performed (Figure S14). By integrating the characteristic vibration peaks of MTD, the mapping image reveals uniform signal intensity across the entire scanned region, whereas the control film shows no detectable signal. This result demonstrates that MTD molecules are homogeneously distributed on the PbS surface without obvious aggregation. Such uniformity likely arises from the stable adsorption of dipolar molecules on the film surface [47].

FIGURE 4.

(a) KPFM images of the control PbS film and PbS films treated with MT, MTZ, and MTD molecules, respectively; the scan area is 2 × 2 µm². (b) Histograms of the surface potential distributions extracted from the corresponding KPFM images. (c) Secondary electron cutoff spectra of the control and molecule‐treated PbS films. (d) Energy level diagrams derived from the UPS results and material band gaps.

To further investigate the impact of dipolar molecules on the band structure, ultraviolet photoelectron spectroscopy (UPS) measurements were performed (Figure S15). The secondary electron cutoff spectra are shown in Figure 4c. The optical bandgap of the PbS QDs was estimated to be 1.2 eV, based on the EQE spectrum up to the edge. The resulting energy level alignment is depicted in Figure 4d and further quantified in Figure S16. For the control sample, the estimated energetic offset between the PbS CBM and the C₆₀ LUMO is 0.85 eV. In contrast, the MTD‐modified films show a shifted PbS CBM, corresponding to a reduced estimated offset of 0.46 eV. These results suggest that MTD modification leads to a more favorable interfacial energy‐level alignment for electron extraction. Such an improved energetic landscape may help reduce interfacial energy loss, mitigate charge accumulation, and suppress recombination losses at the PbS/C₆₀ interface. Furthermore, the more continuous energy‐level profile is consistent with the reduced contact resistance and improved carrier transport kinetics observed in the MTD‐modified devices [48, 49].

2.5. Carrier Mobility and Recombination Dynamics Analysis

We further employed optical and electrical characterizations to investigate the influence of different dipolar molecules on carrier dynamics and device physics. Figure 5a shows the photoluminescence (PL) spectra of FTO/HTL/PbS/dipolar molecule structures without the introduction of the C₆₀ ETL. Compared with the control sample, all dipolar‐modified samples exhibited significantly enhanced PL intensities. This behavior indicates that dipolar molecules also effectively passivated trap states on the PbS QD surface, PL measurements were also performed on bare PbS films without the HTL (Figure S17, Supporting Information), which yielded consistent results. Notably, the MTD‐modified film shows the highest PL, which is consistent with its highest adsorption energy on PbS. Time‐resolved photoluminescence (TRPL) measurements (Figure S18, Supporting Information) show that the average carrier lifetime (t_avg) increases from 23.07 ns (control) to 43.01 ns (MTD). This significant extension of t_avg confirms suppressed non‐radiative recombination and reduced trap‐assisted losses, consistent with the test results of trap density of states (tDOS, Figure S19, Supporting Information), corroborating the improved carrier stability and device performance. Upon introducing the C₆₀ ETL (Figure 5b), the PL intensities of all samples were quenched to varying extents, indicating that the C₆₀ layer facilitated exciton dissociation. Although the PL intensity decreases overall after the introduction of C₆₀, the residual PL signals of the dipolar molecule‐modified samples remain significantly higher than that of the control. To more accurately evaluate the interfacial charge extraction behavior, we quantified the PL quenching ratio [$(P{L}_{\mathrm{w}/\mathrm{o}{\mathrm{C}}_{60}}-P{L}_{{\mathrm{C}}_{60}})/P{L}_{\mathrm{w}/\mathrm{o}{\mathrm{C}}_{60}}$], which provides a comparative measure of the relative extent of photogenerated carrier extraction across the PbS/C₆₀ interface. As shown in Figure 5c (and detailed in Figure S20), the MTD‐modified films exhibited the largest quenching ratio, suggesting more favorable interfacial charge extraction at the PbS/C₆₀ interface. This result is consistent with the improved energy‐level alignment and pronounced interfacial charge redistribution discussed in previous sections, supporting the dual role of MTD in surface passivation and interfacial charge‐extraction modulation.

FIGURE 5.

Steady‐state PL spectra of PbS films without (a) and with (b) C₆₀ coating. The insets illustrate the sample structures and laser illumination directions. (c) Integrated PL intensities calculated from the spectra in (a and b). TPC (d) and TPV (e) decay curves of the control and molecule‐treated PbS devices. (f) Transfer characteristics of bottom‐gate, bottom‐contact FETs based on pristine PbS and MTD‐modified PbS QD films.

To further investigate charge transfer and recombination processes, transient photocurrent (TPC) and transient photovoltage (TPV) measurements were conducted (Figure 5d,e). In the TPC curves, the charge decay time of the control device was 14.19 µs, whereas the MT‐, MTZ‐, and MTD‐modified devices exhibited significantly shorter decay times of 9.57, 6.52, and 5.22 µs, respectively (Table S2), indicating improved charge extraction efficiency, with MTD showing the best performance, which minimizes interfacial charge accumulation and facilitates higher J _sc and FF. The TPV measurements further revealed the changes in charge recombination dynamics. The control device exhibited a photovoltage decay time of 16.27 µs (Table S3), while dipolar‐modified devices showed extended decay times of 21.36 µs (MT), 30.11 µs (MTZ), and 42.07 µs (MTD). This trend clearly indicates that dipolar molecules prolong carrier lifetimes and suppress interfacial recombination, with MTD‐modified devices exhibiting the most pronounced carrier stability improvement. This suppression of carrier loss, which correlates with the reduced tDOS, is the primary driver for the observed enhancement in V _oc [50].

To further elucidate the interfacial electronic effects induced by MTD, we fabricated bottom‐gate bottom‐contact field‐effect transistors (FETs) using pristine PbS and MTD‐modified PbS films as active layers. As shown in Figure 5f, pristine PbS exhibits obvious n‐type behavior with dominant electron transport characteristics. In contrast, the PbS/MTD device displays ambipolar behavior, with the emergence of hole conduction in the negative gate voltage region. These results provide further evidence that the interfacial dipole layer not only optimizes energy‐level alignment but also strengthens electronic coupling and directional charge extraction at the PbS/C₆₀ interface [51, 52]. Furthermore, MTD‐modified devices exhibit significantly suppressed hysteresis in their transfer characteristics, indicating improved electronic quality and reduced charge trapping. This is consistent with the effective surface passivation and optimized interfacial dynamics, which ensure more stable and reversible charge transport. To validate these findings, electrochemical impedance spectroscopy (EIS) was performed to measure series resistance (R _s) and charge transfer resistance (R _tr), and the Nyquist plots are shown in Figure S21. As summarized in Table S4, MTD‐modified devices exhibited significantly reduced R _s values and increased R _tr values, reflecting reduced series resistance, and enhanced interfacial charge transfer capability, thereby promoting efficient carrier transport. To further quantify the internal device physics, capacitance‐voltage (C–V) analysis was performed (Figure S22). As summarized in Table S5, MTD modification leads to an increased built‐in potential (V _bi) from 0.70 to 0.81 V and an expanded depletion width (W) from 153 to 163 nm. This strengthened internal electric field and broadened depletion zone collectively facilitate exciton dissociation and enhance carrier collection efficiency [53], further corroborating the optimized interfacial carrier dynamics induced by the MTD dipole layer.

3. Conclusion

In this study, we have demonstrated an electron‐extraction‐modulating self‐limiting ultrathin molecular layer strategy that enables record PCE in p‐i‐n PbS QD solar cells. The designed dipolar heteroaromatic molecules effectively suppress nonradiative recombination through strong Pb‐S coordination while enhancing electron extraction via π–π interactions with C₆₀. Their high volatility further allows the removal of excess molecules during annealing, yielding compact and ultrathin interfacial layers. This synergistic interfacial engineering approach has elevated the PCE to 14.47%, the highest reported to date for p‐i‐n PbS QD devices. Our work demonstrates that electron‐extraction‐modulating self‐limiting ultrathin molecular layers are equally important as their hole‐transporting counterparts for advancing device performance. Furthermore, the demonstrated high‐efficiency p‐i‐n PbS QD subcells, with their extended near‐infrared absorption, offer strong potential for single integration into future monolithic tandem solar cells.

4. Experimental Section

4.1. Chemicals

Lead iodide (PbI₂, 99+%), N,N′‐Diphenylthiourea (DPhTA, 98%), N,N‐dimethylformamide (DMF, 99.8% anhydrous), 3‐Mercaptopropionic acid (MPA, 99%), Butylamine (BA, >99%), Toluene (TOL, 95%), Hexane (Hex, 95%), Isopropanol (IPA, 95%), Acetone (95%), MeO‐2PACz (99%, Sigma), Acetonitrile (99.9%), 2‐Methoxyethanol (95%), Acetonitrile (99.9%), Nickeltrioxide (NiO, 99%), Hexafluoro‐2‐propanol (6F‐IPA, 99.5%), 2‐Mercapto‐1,3,4‐thiadiazole (MTD, 98%), 2‐Mercaptothiazole (MTZ, 98%), 2‐Mercaptothiophene (MT 97%, Innochem), The glue dispenser/spin coater (Brand‐REESEEN, PvS‐mini9) used in the lab are from Jiangyin J. Wanjia Technology Co., Ltd.

4.2. Synthesis and Purification of PbS CQD Ink

The preparation of lead sulfide CQD ink follows our previous research method [54]. The specific steps are as follows: First, in a nitrogen atmosphere, 6 mmol of PbI₂ is dissolved in 8  mL of DMF and continuously stirred (operated at room temperature). After the PbI₂ is fully dissolved, it is transferred to ice water and stirred for another 10 min. Then, 16 µL of MPA and 1 mL of BA are mixed and injected into the iodide lead solution. One minute later, 1 mmol of DPhTA dissolved in 1 mL of DMF is injected into the solution, and the solution immediately turns black. After 10 min, toluene is added as a reverse solvent, and the mixture is centrifuged at 8000 rpm for 5 min. Finally, CQD solid precipitate is obtained. The structural and optical properties of the as‐synthesized PbS QDs were further characterized (Figure S23).

4.3. Device Fabrication

FTO substrates were sequentially cleaned by acetone, detergent, deionized water, acetone, and isopropanol under sonication, followed by UV‐ozone treatment for 30 min. A NiOx layer was first deposited by spin‐coating a 10 mg mL^{− 1} NiOx precursor solution in air at 3000 rpm for 30 s, followed by annealing at 150 °C for 10 min. The substrates were then transferred into a nitrogen‐filled glovebox, where a self‐assembled monolayer (MeO‐2PACz, 1 mg mL⁻¹ in dimethoxyethanol) was spin‐coated at 3000 rpm for 30 s and annealed at 100 °C for 10 min. Subsequently, a PbS QD film (∼1200 mg mL^{− 1} in DMF) was spin‐coated at 2000 rpm for 40 s and annealed at 70 °C for 10 min. A dipolar layer was then formed by spin‐coating an MTD solution (1 mg mL⁻¹ in 6F‐IPA) at 3000 rpm for 30 s in a nitrogen‐filled glovebox (H₂O  <  0.01 ppm, O₂  <  0.01 ppm), followed by annealing at 55 °C for 5 min. Finally, 30 nm C₆₀, 10 nm BCP, and 120 nm Ag were sequentially deposited via thermal evaporation under high vacuum to complete the device fabrication.

4.4. Computational Methods

The DS‐PAW package was used to perform the density functional theory (DFT) calculations [55], using the Perdew‐Burke‐Ernzerhof (PBE) functional within the generalized gradient approximation (GGA) [56]. The project‐augmented wave (PAW) method was used to represent the core‐valence electron interaction [57]. The DFT‐D3 method was used to describe the dispersion interactions [58]. The valence electrons were expanded in a plane‐wave basis set with a cutoff energy of 450 Ev. The (111) surface of PbS was selected to construct model with a four‐layer slab model with a vacuum of 20 Å by the Device Studio software [5], corresponding a k‐points mesh of 1×1×1. During the structural optimization, the bottom two layers were fixed, and the top two layers and the adsorbates on surface were completely relaxed. The geometry optimization is stopped until the maximal force on each relaxed atom was less than 0.05 Ev/Å. To evaluate the intrinsic chemical bonding strength, the adsorption energies were calculated based on the interaction between a single deprotonated molecule and the PbS surface. In this model, the thiol group is stripped of its proton, and the sulfur atom is directly coordinated to a surface Pb atom‐a configuration that aligns with the chemical environment substantiated by our FTIR analysis. The adsorption energy (E _ads) of molecule was calculated by the equation:E _ads = E _x/surf − E _surf − E _x, where E _x/surf, E _surf, and E _x are the surface with adsorbate, the clean surface and the total energy of gaseous molecule, respectively. The more negative E _ads means the stronger adsorption strength.

4.5. Measurement and Characterization

The surface morphologies and electronic potentials of a series of films were characterized by a Bruker Dimensional Icon AFM in tapping mode and in peakforce KPFM mode, in which the tapping mode used silicon cantilevers with a spring constant of 40 N m⁻¹ and the PF‐KPFM mode used OSCM‐PT probes coated with Pt. UV‐vis NIR spectra were recorded on a Perkin Elmer model Lambda 750. Current density‐voltage characteristics of the devices were measured under ambient air using a programmable Keithley 2401 source meter under a simulated AM 1.5G solar irradiation of 100 mW cm⁻². (Newport, Class AAA solar simulator, 94023A‐U) The light intensity is calibrated by a certified Oriel Reference Cell (91150 V) and verified with an NREL‐calibrated Hamamatsu S1787‐04 diode. The ITO/electrode overlap area is 0.0725 cm². The devices are measured with shadow mask, determining the active area of 0.04 cm². Voltage swept from −0.8 to 0.8 V with a speed of 0.01 V per point and a dwell time of 10 ms. The UPS/XPS tests were conducted by spin‐coating films on the FTO substrate (with the same details as in device fabrication). FTIR mapping was performed in reflection mode on gold‐coated silicon (Au/Si) substrates to enhance the signal sensitivity of the thin films. The PbS QD films were spin‐coated onto the Au/Si surface, and mapping data were acquired over a 40 × 40 µm area. The spatial distribution of MTD molecules was visualized by integrating the characteristic vibrational peak area of MTD (located at 1050‐1200 cm⁻¹). TPC and TPV were obtained under light through a waveform generator (Keysight 33500B) and digital storage oscilloscopes (Keysight DSOX1204A). EIS measurements were carried out through a Zahner IM6 electrochemical workstation.

4.6. FET Fabrication and Measurements

The PbS CQD FETs were fabricated on top of highly doped Si substrates covered with a 230 nm thermally grown SiO₂ dielectric. Prepatterned interdigitated electrodes consisting of 10 nm ITO and 35 nm of Au are serve as source and drain electrodes with a channel width of 10 mm and length of 20 µm. The substrates were cleaned by sonication in acetone and isopropanol, dried in the oven, and treated with O₂ plasma prior the film deposition. The CQD inks in DMF (100 mg mL⁻¹) were deposited onto the substrates by spin‐coating at 2000 rpm for 40s. The MTD molecules in 6F‐IPA (2 mg mL⁻¹) were deposited by spin‐coating at 3000 rpm for 30s. FETs were annealed for 10 min at 100°C. All transistor measurements were performed with an Agilent E5262B semiconductor parameter analyzer. All fabrication and measurement steps were performed in N₂ filled gloveboxes with O₂ and H₂O concentrations below 0.1 ppm. The extraction of electron and hole linear mobility could be referred to in the previous report [59].

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adma74170‐sup‐0001‐SuppMat.pdf.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.