The rapid advancement of perovskite solar cells has positioned them as a leading technology in photovoltaics, with power conversion efficiencies surpassing 27%. This progress is largely attributed to a deeper understanding of material properties and device physics, particularly the role of interfaces in charge dynamics. However, the electron transport layer/perovskite active layer (ETL/PVK) heterojunction in n-i-p structured perovskite solar cells not only facilitates charge separation but also fundamentally dictates the crystallization behavior of the perovskite film. The ETL surface acts as a template for nucleation and growth, encoding the final microstructure—grain size, orientation, and boundary properties—through its physicochemical characteristics. This review systematically examines how ETL interface engineering influences perovskite crystallization, drawing on thermodynamic and kinetic principles to elucidate the mechanisms governing film quality and device performance.
The crystallization of perovskite films begins with nucleation from a precursor solution on the ETL surface. According to classical nucleation theory, the energy barrier for nucleation determines the density and mechanism of crystal formation. The surface energy and wettability of the ETL are critical in modulating this barrier. High surface energy substrates, such as hydroxyl-rich TiO₂, reduce the nucleation barrier, leading to increased nucleation density and finer grains. Conversely, low surface energy surfaces, like those modified with self-assembled monolayers (SAMs), suppress nucleation, resulting in larger grains but potentially poorer coverage. The contact angle of the precursor solution on the ETL serves as a quantitative measure of wettability. Optimal wetting (contact angles between 10° and 30°) ensures uniform spreading and controlled solvent evaporation, while extreme values cause film discontinuities or dewetting. The relationship between surface energy (γ), contact angle (θ), and nucleation barrier (ΔG*) can be expressed as:
$$ \Delta G^* = \frac{16\pi \gamma^3}{3(\Delta G_v)^2} f(\theta) $$
where ΔGv is the volumetric free energy change and f(θ) is a function of the contact angle, emphasizing the role of interface properties in crystallization kinetics.
| ETL Type/Modification | Surface Energy (mJ/m²) | Contact Angle (°) | Nucleation Density (cm⁻²) | Average Grain Size (nm) |
|---|---|---|---|---|
| Mesoporous TiO₂ | 75–85 | 0–5 | 10¹⁰–10¹¹ | 100–300 |
| SnO₂ Thin Film | 60–70 | 15–25 | 10⁹–10¹⁰ | 300–600 |
| PCBM-Modified SnO₂ | 40–50 | 30–40 | 10⁸–10⁹ | 600–1000 |
| SAMs-Modified SnO₂ | 30–45 | 25–35 | 10⁷–10⁸ | 1000–2000 |
Surface functional groups on the ETL, such as -OH, -COOH, and -NH₂, engage in coordination chemistry with perovskite precursors (e.g., PbI₂), guiding crystal growth direction and rate. Carboxyl groups form bidentate complexes with Pb²⁺, while amino groups exhibit dual functionality by coordinating with Pb²⁺ via lone pairs and forming electrostatic interactions with I⁻ through protonated -NH₃⁺. These interactions lower the activation energy for nucleation and promote oriented growth. For instance, SAMs with phosphonic acid groups (-PO₃H₂) form tridentate coordinations with metal oxide surfaces, inducing highly ordered (100)-oriented perovskite films. The binding energy (Ebind) between functional groups and perovskite precursors can be modeled as:
$$ E_{\text{bind}} = -\frac{k q_1 q_2}{r} $$
where k is a constant, q₁ and q₂ are charges, and r is the interaction distance, highlighting the electrostatic nature of these interactions.
The crystallization pathway from solution to solid is intricately guided by the ETL interface. Near the surface, precursor ions (Pb²⁺, I⁻, MA⁺/FA⁺) exhibit altered solvation structures and concentrations, pre-organizing into intermediate phases like MA₂Pb₃I₈(DMSO)₂ on TiO₂. This pre-organization reduces the free energy for nucleation and directs phase transformation. In contrast, inert SAM-modified surfaces suppress the formation of non-perovskite phases (e.g., δ-FAPbI₃), enabling direct crystallization of the photoactive α-phase. The thermodynamic driving force for phase transformation is given by:
$$ \Delta G_{\text{trans}} = \Delta H – T \Delta S $$
where ΔH and ΔS are the enthalpy and entropy changes, respectively, and T is temperature. Interface-induced strain can stabilize metastable phases, as compressive stress promotes δ-phase formation while tensile stress favors the α-phase.
Grain orientation in perovskite films is induced by lattice matching and surface functional groups. Epitaxial or quasi-epitaxial growth occurs when the lattice mismatch between ETL and perovskite is small (<5%), as seen on LaNiO₃ substrates, where (100)-oriented MAPbI₃ grows due to a 3% mismatch. Functional groups selectively adsorb to specific crystal faces, altering surface energies and promoting preferred orientations. For example, carboxyl-terminated SAMs reduce the surface energy of (110) faces, favoring (100) growth. The interfacial stress arising from thermal expansion mismatch also influences orientation; substrates with lower coefficients of thermal expansion (e.g., TiO₂) introduce tensile stress upon cooling, promoting (111) orientation. The stress (σ) at the interface can be estimated as:
$$ \sigma = E \cdot \alpha \cdot \Delta T $$
where E is Young’s modulus, α is the thermal expansion coefficient difference, and ΔT is the temperature change.
| ETL Type | Primary Induced Orientation | Degree of Orientation | Carrier Mobility (cm²/V·s) | Ion Migration Activation Energy (eV) |
|---|---|---|---|---|
| Mesoporous TiO₂ | (110) | Medium | 10–20 | 0.40–0.50 |
| SnO₂ Nanocrystals | (100) | High | 20–30 | 0.50–0.60 |
| NiOₓ Nanoparticles | (111) | Medium | 15–25 | 0.45–0.55 |
| 4PACz-SAMs | (100) | Very High | 30–40 | 0.60–0.70 |
Asymmetric influences of bottom (ETL/PVK) and top (PVK/HTL) interfaces further underscore the dominance of ETL in crystallization. The bottom interface governs the entire process from nucleation to phase transformation, while the top interface only affects later stages of growth and termination. This temporal asymmetry means that ETL properties fundamentally establish film quality, regardless of top interface modifications.
The crystallization quality directly impacts device performance in perovskite solar cells. Grain size inversely correlates with defect density; larger grains reduce grain boundary areas, which are hotspots for non-radiative recombination. Increasing average grain size from 0.2 μm to 1 μm can lower bulk defect density by 1–2 orders of magnitude, enhancing open-circuit voltage (VOC). Grain boundaries also exhibit type-dependent properties: low-angle boundaries in (100)-oriented films have lower defect densities than random high-angle boundaries. Crystallographic orientation introduces anisotropy in charge transport and ion migration. The (100) direction shows the highest carrier mobility, improving fill factor (FF) and short-circuit current density (JSC), while (111)-oriented films exhibit higher ion migration activation energies, reducing hysteresis and improving stability. The carrier mobility (μ) along different crystallographic directions can be described by:
$$ \mu = \frac{q \tau}{m^*} $$
where q is charge, τ is scattering time, and m* is effective mass, which varies with orientation due to band structure differences.
Interface stress profoundly affects long-term stability. Thermal mismatch stress, as between TiO₂ (α ≈ 8×10⁻⁶ K⁻¹) and MAPbI₃ (α ≈ 50×10⁻⁶ K⁻¹), induces tensile stress during cooling, promoting crack formation. Lattice mismatch stress in quasi-epitaxial systems creates periodic strain fields that may accelerate degradation. Importantly, tensile stress stabilizes the photoactive α-phase of FAPbI₃, delaying α→δ phase transformation, while compressive stress favors δ-phase formation. The stress-phase stability relationship is critical for designing durable perovskite solar cells.
Advanced interface engineering strategies employ multiscale designs to optimize crystallization. Molecular-scale control using SAMs with tailored functional groups (e.g., bifunctional molecules with -PO₃H₂ and -COO⁻) simultaneously optimizes electronic properties and crystallization guidance. Nanoscale patterning via nanoimprinting or block copolymer self-assembly creates periodic templates for spatially controlled nucleation. Gradient interface designs introduce compositional or band alignment transitions to mitigate lattice and thermal mismatch. These approaches leverage the fundamental understanding of ETL/perovskite interactions to enhance film quality and device performance.
In situ characterization and theoretical modeling are invaluable for elucidating crystallization mechanisms. Techniques like grazing-incidence X-ray diffraction (GI-XRD) and time-of-flight secondary ion mass spectrometry (ToF-SIMS) provide real-time insights into structural evolution at the interface. Multiscale simulations, from density functional theory (DFT) to phase-field models, reveal electronic structures and predict crystallization pathways. Machine learning-assisted molecular dynamics can handle the complex spatiotemporal scales of perovskite crystallization, enabling predictive design of interface materials.
Future challenges and opportunities lie in high-throughput experimentation coupled with data science to identify optimal interface materials. Dynamic interfaces that adapt to environmental changes (e.g., light, temperature) could guide crystallization initially and passivate defects during operation. Biomimetic designs, inspired by natural biomineralization, may offer strain-tolerant structures. Commercialization requires scaling interface engineering to large-area, low-cost manufacturing processes, ensuring compatibility with encapsulation techniques. The integration of interface design with packaging will be crucial for long-term stability of perovskite solar cells.
In conclusion, ETL interface engineering is pivotal for controlling perovskite crystallization, influencing nucleation, growth, orientation, and stability. By manipulating surface energy, functional groups, and stress states, high-quality films with optimal microstructures can be achieved, advancing the performance and durability of perovskite solar cells. Continued research into multiscale interface strategies and advanced characterization will drive further innovations in this rapidly evolving field.