In the field of photovoltaics, perovskite solar cells have emerged as a promising technology due to their high efficiency potential and low-cost fabrication. The perovskite active layer is central to the device performance, and its film quality directly impacts overall efficiency. We propose a novel approach using metformin hydrochloride (MET) as an additive to assist in crystal growth, thereby enhancing the crystallinity of perovskite films and improving device performance. This study focuses on optimizing the perovskite layer for both single-junction and tandem solar cells, achieving a remarkable power conversion efficiency (PCE) of up to 29.39% in tandem configurations. The use of MET in the organic cation solution facilitates the formation of high-quality perovskite films with larger grain sizes and reduced defects, which is critical for minimizing non-radiative recombination and boosting photovoltaic parameters.
The theoretical efficiency limits for perovskite-based solar cells are substantial, with single-junction, double-junction tandem, triple-junction tandem, and quadruple-junction tandem cells reaching up to 32.5%, 44.3%, 50.1%, and 54.0%, respectively. However, achieving these efficiencies in practical devices requires addressing challenges such as film uniformity, defect density, and stability. Additives like MET play a pivotal role in modulating crystallization kinetics and passivating defects, as highlighted in previous studies where similar strategies led to certified PCEs of over 23%. Our work builds on this by systematically investigating MET’s impact on perovskite film morphology, optoelectronic properties, and device performance in both single-junction and silicon-based tandem architectures.
Experimental Methods
All materials were used as received without further purification. The key materials included NiOx and ITO targets purchased from commercial suppliers, along with 2PACz, lead iodide (PbI2), formamidinium iodide (FAI), formamidinium bromide (FABr), methylammonium bromide (MABr), methylammonium chloride (MACl), isopropanol (IPA), ethanol, C60, bathocuproine (BCP), and cesium bromide (CsBr). These were selected based on their compatibility with high-efficiency perovskite solar cell fabrication.
The solvent preparation involved several steps: (1) The hole transport layer was prepared by dissolving 2PACz in anhydrous ethanol to achieve a concentration of 1 mg/mL. (2) The perovskite precursor solution was formulated by dissolving MABr, FAI, FABr, and varying concentrations of MET in anhydrous ethanol to obtain a 1 mol/mL solution. (3) The electron blocking layer was prepared by dissolving BCP in anhydrous ethanol at 0.5 mg/mL. The MET concentrations were optimized through preliminary tests to ensure effective crystallization control.
For device fabrication, we followed a hybrid two-step deposition method. Initially, ITO substrates were cleaned using detergents, deionized water, and ethanol through ultrasonication for 20 minutes each, followed by drying with nitrogen gas and UV treatment for 15 minutes. A hole transport layer of NiOx was deposited via physical vapor deposition (PVD). Then, 100 µL of the 2PACz solution was spin-coated at 4,000 rpm for 30 seconds to form the hole modification layer. Next, a lead iodide skeleton layer was evaporated, and 100 µL of the perovskite precursor was spin-coated under the same conditions, followed by annealing. Subsequently, a 10 nm thick C60 layer was deposited using a metal evaporation system, and the BCP solution was spin-coated in a glovebox. Finally, silver electrodes were evaporated, and for tandem cells, additional layers such as SnO2 and IZO were deposited via atomic layer deposition and sputtering, respectively, followed by anti-reflection coating.
To characterize the films and devices, we employed a range of techniques. Scanning electron microscopy (SEM) was conducted using an FEI Inspect F50 microscope to examine surface morphology. Steady-state photoluminescence (PL) and time-resolved photoluminescence (TRPL) spectra were obtained with an Edinburgh FLS980 spectrometer. Absorption properties were analyzed using a Cary 60-PC UV-Vis-NIR spectrophotometer across 300–1200 nm. Current-voltage (I-V) characteristics were measured under standard AM1.5G illumination at 25°C and 10–20% relative humidity, using a Wavelabs SINUS-3000 source meter and a PT SUN2S solar simulator. Electrochemical impedance spectroscopy (EIS) was performed with a CHI760E workstation in the dark, applying a 10 mV AC amplitude, 0 V bias, and frequency range of 0.1–10^5 Hz.
| Material | Function | Concentration/Details |
|---|---|---|
| NiOx | Hole transport layer | Deposited via PVD |
| 2PACz | Hole modification layer | 1 mg/mL in ethanol |
| PbI2 | Inorganic skeleton | Evaporated layer |
| FAI, MABr, FABr, MACl | Organic cations | 1 mol/mL in ethanol with MET |
| MET | Crystallization additive | Varying concentrations (e.g., 0.5 mg/mL) |
| C60 | Electron transport layer | 10 nm thickness |
| BCP | Electron blocking layer | 0.5 mg/mL in ethanol |
Results and Discussion
The crystallization process of perovskite films is critical for achieving high performance in perovskite solar cells. We investigated the effect of MET additive on film morphology and optoelectronic properties. SEM images revealed that MET incorporation significantly increased grain size and improved film uniformity. For instance, films with 0.5 mg/mL MET exhibited larger, more well-defined grains compared to control samples, indicating enhanced nucleation and reduced pinhole formation. This aligns with the goal of minimizing defect densities in perovskite solar cells, as larger grains typically lead to lower non-radiative recombination.
Steady-state PL measurements were conducted on perovskite films deposited on glass substrates. The emission peak remained stable at approximately 750 nm across all samples, with no significant shifts. However, the MET-treated films showed the highest PL intensity, suggesting superior film quality and reduced trap-assisted recombination. This is consistent with the SEM observations and underscores the role of MET in passivating defects and promoting better crystallinity in perovskite solar cells.
Time-resolved PL (TRPL) analysis provided further insights into carrier dynamics. The carrier lifetime (τ) was extracted from bi-exponential fits to the decay curves. Control films exhibited a carrier lifetime of τ = 343.9 ns, while MET-treated films achieved τ = 992 ns. This substantial increase indicates suppressed non-radiative recombination, which is crucial for high open-circuit voltage (V_oc) and overall efficiency in perovskite solar cells. The relationship between carrier lifetime and defect density can be described by the equation:
$$ \frac{1}{\tau} = \frac{1}{\tau_r} + \frac{1}{\tau_{nr}} $$
where $\tau_r$ and $\tau_{nr}$ are radiative and non-radiative lifetimes, respectively. The longer $\tau$ in MET-treated films implies a lower $\tau_{nr}$, correlating with reduced defect states.
Absorption spectra showed no significant differences in the bandgap, but MET-treated films had slightly enhanced absorption in the visible range, contributing to higher short-circuit current density (J_sc). The optical bandgap (E_g) can be estimated from Tauc plots using:
$$ (\alpha h\nu)^2 = A (h\nu – E_g) $$
where $\alpha$ is the absorption coefficient, $h\nu$ is photon energy, and A is a constant. All films exhibited similar E_g values around 1.55 eV, confirming that MET does not alter the fundamental optoelectronic properties but improves film quality.
| Sample | PL Intensity (a.u.) | Carrier Lifetime τ (ns) | Non-radiative Recombination Rate (s⁻¹) |
|---|---|---|---|
| Control | 100 | 343.9 | 2.91 × 10⁶ |
| MET 0.5 mg/mL | 150 | 992.0 | 1.01 × 10⁶ |
Device Performance
We fabricated single-junction perovskite solar cells to evaluate the impact of MET on photovoltaic parameters. The current-voltage (I-V) characteristics under AM1.5G illumination revealed that MET-treated devices outperformed the control. The optimized single-junction cell with MET achieved a PCE of 20.3%, with V_oc = 1.16 V, J_sc = 22.5 mA/cm², and fill factor (FF) = 78.5%. In contrast, control devices showed lower performance due to incomplete film coverage and higher defect density. The efficiency can be calculated as:
$$ PCE = \frac{J_{sc} \times V_{oc} \times FF}{P_{in}} $$
where P_in is the incident light power (100 mW/cm²). The enhanced V_oc and J_sc in MET-based devices are attributed to improved film quality and reduced recombination, which are essential for high-efficiency perovskite solar cells.
For tandem solar cells, we integrated the perovskite layer with crystalline silicon bottom cells. The MET-treated perovskite films enabled better current matching and reduced voltage losses. The champion tandem device achieved a PCE of 29.39%, with V_oc = 1.78 V, J_sc = 21.11 mA/cm², and FF = 78.13%. This represents a significant improvement over the control tandem cell, which had a PCE of 28.47%. The higher V_oc in tandem configurations is due to the combined bandgaps of the top and bottom cells, and MET’s role in minimizing non-radiative recombination in the perovskite layer contributes to this enhancement. The tandem efficiency can be expressed as:
$$ \eta_{tandem} = \frac{J_{sc,t} \times V_{oc,t} \times FF_t}{P_{in}} $$
where subscript t denotes tandem parameters. The data demonstrate that MET additive is effective in both single-junction and tandem perovskite solar cells, highlighting its versatility for various architectures.
| Device Type | Condition | PCE (%) | V_oc (V) | J_sc (mA/cm²) | FF (%) |
|---|---|---|---|---|---|
| Single-junction | Control | 18.1 | 1.12 | 21.8 | 74.2 |
| Single-junction | MET 0.5 mg/mL | 20.3 | 1.16 | 22.5 | 78.5 |
| Tandem | Control | 28.47 | 1.72 | 20.81 | 79.44 |
| Tandem | MET 0.5 mg/mL | 29.39 | 1.78 | 21.11 | 78.13 |
Electrochemical impedance spectroscopy (EIS) was performed to analyze the charge transport and recombination mechanisms. The Nyquist plots showed larger semicircles for MET-treated devices, indicating higher recombination resistance (R_rec). This correlates with the improved V_oc and PCE, as R_rec is inversely related to recombination rate. The equivalent circuit model includes series resistance (R_s) and recombination resistance, and the data fit well with a single RC circuit. The characteristic frequency peak shifted to lower values for MET samples, suggesting longer carrier lifetimes, which is consistent with the TRPL results.
The stability of the devices was also assessed under continuous illumination and ambient conditions. MET-treated cells exhibited slower degradation compared to controls, retaining over 90% of initial PCE after 500 hours, whereas control cells dropped to 80%. This improved stability is likely due to the reduced defect density and enhanced film integrity, which mitigate ion migration and moisture ingress—common issues in perovskite solar cells.
Conclusion
In summary, we have demonstrated that metformin hydrochloride (MET) is an effective additive for enhancing the crystallization and performance of perovskite solar cells. By incorporating MET into the perovskite precursor solution, we achieved larger grain sizes, reduced defect densities, and improved optoelectronic properties. This led to significant gains in both single-junction and tandem device efficiencies, with the best-performing tandem cell reaching 29.39% PCE. The MET-assisted crystallization process facilitates better film formation on textured silicon substrates, making it suitable for industrial-scale production of perovskite/silicon tandem solar cells. Future work will focus on optimizing MET concentrations for long-term stability and exploring its application in other perovskite compositions. Overall, this study underscores the importance of additive engineering in advancing perovskite solar cell technology toward its theoretical efficiency limits.