Professor Randi Azmi's Team Publishes in Science: Breakthrough Molecular Design Achieves 25.4% Efficiency in Stable Perovskite Solar Cells
Overview
A collaborative study about perovskite solar cells published in “Science” is led by Randi Azmi’s team at The Chinese University of Hong Kong, Shenzhen, together with multiple academic partners.
Multivalent, resonance- stabilized amidinium ligands enable stronger chemical coordination and reduced deprotonation compared with conventional monovalent ammonium ligands in low- dimensional perovskites. Based on this, Randi Azmi et al. introduce a controllable one- to two- dimensional (1D- to- 2D) structural transition strategy by systematically tuning ligand conformation, thereby modulating hydrogen bonding, π–π stacking, and basicity to elucidate the relationship between molecular structure, interfacial interactions, and resulting dimensionality. The 1D-amidinium perovskite structure, with its pronounced geometric anisotropy, impedes uniform surface coverage and defect passivation. In contrast, the 2D-amidinium perovskite forms a continuous, homogeneous interfacial layer, enabling more effective defect passivation and favorable energy-level alignment. With dimensionality control, inverted 3D/2D-amidinium perovskite solar cells deliver 25.4% power conversion efficiency (1.1 square centimeters, steady- state certified) and maintain >95% of their initial efficiency after 1100 hours of continuous 1-sun operation at 85°C.
The 4-cell minimodule
Image: Xiaoming Chang
Journal Introduction
Founded in 1880, Science is one of the world’s leading general-interest scientific journals. It publishes peer-reviewed original research that reports major advances with broad relevance across disciplines. Known for its rigorous review and selective editorial process, Science is widely recognized as a flagship venue for high-impact scientific findings. Articles published in Science are widely read and cited by researchers, and they often draw attention from industry and policymakers, helping inform research priorities and the translation of discoveries into technology.
?
Research Background
Although perovskite solar cells combine high efficiency with the potential for low-cost manufacturing, practical deployment remains limited by long-term thermal, light, and electrical stresses during operation. Interfacial degradation is a key bottleneck that prevents simultaneous gains in efficiency and lifetime, because the interfacial layer is often discontinuous or chemically labile, so early open-circuit voltage improvements cannot be sustained. A central challenge is therefore to build a stable and controllable interfacial passivation layer without adding process complexity, which is essential for scalable fabrication and real-world applications.
Introducing an ultrathin low-dimensional (LD, 1D/2D) perovskite capping layer to form a 3D/LD heterojunction has emerged as an effective route to improve stability by suppressing interfacial recombination, increasing open-circuit voltage, and hindering ion migration. However, conventional monovalent ammonium ligands bind weakly to the perovskite lattice and can undergo deprotonation, diffusion, or other side reactions under heat, light, or bias, leading to loss of coverage continuity and rapid performance decay.
Here we address these issues by linking ligand structure to both LD dimensionality and lateral uniformity. We design a series of ligands with varied headgroups (ammonium versus amidinium), spacers (phenyl versus pyridyl), and pyridinic-nitrogen positions, and propose a molecular design principle that combines multivalent anchoring with basicity-guided conformational control. This strategy enables a continuous and reconfigurable LD capping layer, delivering markedly improved operational stability under high-temperature continuous operation while maintaining high efficiency. Together, these results outline a scalable materials and processing pathway toward reliable inverted perovskite solar cells.
?
Research Content
1. From “trial-and-error” to “rule-based screening”: establishing a pKa+DFT dimensionality criterion
Introducing a low-dimensional (LD) surface layer on 3D perovskites can improve device performance, yet the dimensionality favored by different ligands (1D versus 2D) is often difficult to predict, leading to discontinuous coverage and irreproducible efficiency gains. To address this, the authors move from empirical trial-and-error to a rule-based screening framework. They combine two physicochemical descriptors—pKa?, which reflects the amidinium headgroup’s resistance to deprotonation and thus chemical stability, and pKa?, which captures the basicity of the pyridinic nitrogen and its propensity for coordination and hydrogen bonding—with DFT calculations comparing the reaction-energy difference between ligand-driven 1D and 2D structures (ΔE?D?1D). This yields a molecular-level criterion that predicts dimensionality preference (Fig. 1A). Device statistics support the criterion: ligands with ΔE?D?1D > 0 (e.g., m-APY, m-AMPY) more readily form 2D capping layers and show higher, tighter PCE distributions, whereas ligands with ΔE?D?1D < 0 (e.g., BZA, o-APY, p-APY) favor 1D phases and deliver only limited gains (Fig. 1B, E). Together, these results outline a practical pathway from criterion-guided ligand selection to controlled dimensionality and efficiency improvement.
Figure 1. Ligand screening for LD perovskite capping layers.
(A) pKa? of the amidinium headgroup and pKa? of the pyridinic nitrogen for each ligand.
(B) DFT-calculated reaction energies (ΔE) for forming 1D and 2D LD phases for each iodide ligand salt. ΔE?D?1D (dashed line) > 0 indicates a thermodynamic preference for the 2D phase, whereas ΔE?D?1D < 0 indicates that the 1D structure is more stable.
(C) Representative LD perovskite structures: (BZAM)?PbI? (2D), (BZA)?Pb?I? (1D), and (m-APY)?PbI? (2D).
(D) Quasi-Fermi level splitting (QFLS) of control (pristine 3D) devices and ligand-treated devices, including molecular passivation and 3D/LD heterojunction configurations.
(E) Power conversion efficiency (PCE) of devices with different passivation schemes (n = 9 independent cells per condition).
?
2. Microscopic validation: how a 2D preference yields a high-quality capping layer
To test whether the criterion translates into film-level coverage quality, the authors perform multi-scale characterization. GIWAXS indicates that 2D-m-APY–treated films exhibit more ordered, layered diffraction features, whereas 1D systems show more diffuse scattering signatures (Fig. 2A). SEM reveals that 1D ligands (e.g., BZA) readily form micron-scale rod-like crystallites on the surface, leading to non-uniform coverage, while 2D-m-APY produces a compact and uniform capping layer (Fig. 2B). AFM corroborates the smoother surface morphology of the 2D systems (Fig. 2C). Conductive AFM further maps this structural uniformity into electronic uniformity: films with a 2D capping layer display a more continuous current distribution relative to 1D counterparts and the control (Fig. 2D). Together, these results validate the dimensionality criterion at the microscopic level and explain why 2D phases more readily enable uniform coverage and lateral charge transport.
Figure 2. Thin-film characterization of 3D/LD perovskites.
(A) 2D grazing-incidence wide-angle X-ray scattering (GIWAXS) patterns of perovskite films.
(B) Top-view scanning electron microscopy (SEM) images of perovskite-film surfaces.
(C) Atomic force microscopy (AFM) height images of perovskite films; Max height (Max) and root-mean-square roughness (RMS) are indicated below each image.
(D) Conductive AFM (C-AFM) current maps of perovskite films.
?
3. Mechanistic insights: integrated verification of interfacial stability and energy-level alignment
Uniform coverage alone does not guarantee operational durability; the interface must also remain chemically stable and electronically favorable during processing and operation. The authors therefore use in situ PL to track passivation dynamics during spin-coating and annealing. Pyridinic-N–containing ligands (e.g., m-APY and m-AMPY) trigger an immediate PL enhancement, consistent with effective passivation of undercoordinated surface sites (Fig. 3A–H). Kinetic traces extracted from the in situ spectra show ligand-dependent binding dynamics, and m-APY maintains higher PL after annealing, consistent with more persistent passivation (Fig. 3G, H). XPS provides complementary chemical insight: m-AMPY exhibits post-annealing signatures consistent with deprotonation and subsequent reactions, suggesting a potential pathway for interfacial instability (Fig. 3I, J), whereas m-APY shows improved chemical robustness. UPS further indicates that 2D-preferring ligands induce larger shifts in work function and valence-band position, imparting a more pronounced n-type surface character that favors energy-level alignment and charge extraction in inverted devices (Fig. 3K, L).
Figure 3. Interfacial chemistry and passivation dynamics.
(A–F) In situ photoluminescence (PL) spectral sequences collected during spin-coating and subsequent annealing for films treated with different ligands, together with PL intensity maps.
(G, H) Time evolution of the PL peak intensity extracted from (A–F) during spin-coating (G) and annealing (H).
(I, J) X-ray photoelectron spectroscopy (XPS) of perovskite films: Pb 4f (I) and N 1s (J).
(K, L) Ultraviolet photoelectron spectroscopy (UPS) of perovskite films: secondary electron cutoff and valence-band spectra (K), and extracted work function (WF) and valence band maximum (VBM) (L).
?
4. Device performance and stability: from certified high efficiency to long-term operational validation
Implementing the optimized 2D-m-APY system in devices yields a PCE of 26.5% for 1.0 cm2 cells (Fig. 4A), with an independently certified stabilized efficiency of 25.4% on an active area of ~1.1 cm2 (NIM). The approach scales to 6.8 cm2 mini-modules while maintaining 24.2% efficiency with minimal fill-factor loss, consistent with good lateral uniformity (Fig. 4B). ToF-SIMS after thermal aging shows negligible diffusion for 2D-m-APY, whereas molecular migration is evident for 1D-BZA and 2D-m-AMPY (Fig. 4C–E). Under maximum power point tracking at 85 °C and continuous 1-sun illumination, encapsulated 2D-m-APY devices retain over 95% of their initial efficiency after more than 1100 h (Fig. 4F), demonstrating concurrent high efficiency and operational stability.
Figure 4. Photovoltaic performance and stability.
(A) Current density–voltage (J–V) curves of 2D-m-APY–based large-area devices (1.0 cm2; inset, device schematic).
(B) J–V curves of 2D-m-APY mini-modules (four sub-cells; 6.8 cm2; inset, module schematic).
(C) Time-of-flight secondary ion mass spectrometry (ToF-SIMS) depth profiles of BZA, m-APY, and m-AMPY in fresh and thermally aged perovskite films (85 °C, 12 h).
(D, E) Schematics comparing interfacial stability of 3D/LD perovskite heterojunctions with different ligands: fresh (D) and after thermal-stress aging (E).
(F) Operational stability of encapsulated 2D-m-APY devices under maximum power point (MPP) tracking at 85 °C under 1-sun illumination in ambient air.
?
Research Conclusions
This work reports an interfacial engineering strategy based on multivalent amidinium ligands that enables predictable control over the dimensionality and lateral uniformity of low-dimensional (LD) perovskite capping layers. By combining multidentate anchoring with basicity-guided control, the authors achieve ligand-directed formation of either one-dimensional (1D) or two-dimensional (2D) LD phases, overcoming the common failure modes of discontinuous coverage and chemically labile interfacial layers.
At the device level, the strategy delivers three advances. First, it enables high efficiency, with a certified stabilized PCE of 25.4% for ~1.1 cm2 single cells, and it scales to 6.8 cm2 mini-modules with a PCE of 24.2%, underscoring scalability and process compatibility. Second, it delivers robust high-temperature operational stability: encapsulated devices retain over 95% of their initial efficiency after >1100 h of maximum power point tracking at 85 °C under continuous 1-sun illumination. Third, it establishes a pKa?/pKa? + DFT criterion to predict dimensionality preference, and multi-scale characterization links the 2D capping layers to improved coverage uniformity, defect passivation, surface energetics, and chemical stability.
Together, these results provide a scalable interfacial route to unite efficiency and operational durability in inverted perovskite solar cells, while offering a predictive, tunable molecular-design framework for perovskite interface engineering.
?
Author Profiles
Prof. Randi Azmi received his B.Sc. in Condensed Matter Physics from Universitas Indonesia in 2014 and his Ph.D. in Applied Nanochemistry from Kookmin University (Korea) in 2020. Before joining The Chinese University of Hong Kong, Shenzhen, he worked as a Postdoctoral Researcher at King Abdullah University of Science and Technology (KAUST) in 2020 and was promoted to Research Scientist in 2024. Prof. Azmi has published/co-authored 45 papers in leading international journals (h-index = 30; >4,300 citations), including Science (3), Nature (2), Joule (3), Nature Communications (2), Advanced Materials (2), Advanced Energy Materials (7), and ACS Energy Letters (7), among others. In 2022, he received the ASEAN Young Scientist Award in recognition of his outstanding contributions to science, technology, and innovation. More recently, he was selected for the 2024 NSFC Overseas Excellent Young Scientists Fund and the 2024 Shenzhen Peacock Plan.
Liu Yanping is a Ph.D. student in Energy Engineering at the School of Science and Engineering. Under the supervision of Prof. Randi Azmi, she conducts research on heterojunction perovskite solar cells. She has published eight papers in journals including Nature Reviews Materials. Her research is closely aligned with clean energy technologies, with a primary focus on next-generation perovskite solar cells. She aims to enhance their power conversion efficiency through interfacial engineering, and her interests also extend to hydrogen production and emerging energy-storage technologies.
Rongbo Wang is a Ph.D. student in Energy and Engineering at the School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen. He received his B.Eng. degree from Xidian University in 2021 and his M.Sc. degree from Nankai University in 2024. He has published several papers in SCI Q1 journals, including four first-author (including co-first-author) articles in Journal of Energy Chemistry (2), Science China Chemistry (1), and Chemical Engineering Journal (1). He also holds one granted Chinese national invention patent and was awarded the National Scholarship for Graduate Students in 2023.
?
Research Team Introduction
Prof. Randi Azmi leads the HERO Laboratory (Heterojunction Materials Laboratory for Renewable Energy). The group conducts broad research on high-efficiency single-junction and multi-junction solar cells based on organic–inorganic hybrid heterojunction thin films. Its goal is to develop next-generation photovoltaic technologies that combine high power conversion efficiency with long-term stability. To achieve this, the team advances the field through a multidisciplinary approach, spanning the full pipeline from novel materials discovery to device fabrication, including process and materials optimization. The group also evaluates solar-cell performance under practical operating environments, with the ultimate aim of enabling industrial-scale deployment.
?
By The Randi Azmi Team
?
