In my work, I focused on the buried interface of an inverted perovskite solar cell because this interface between the hole transport layer and the perovskite layer controls not only charge transfer and extraction but also the crystal quality of the perovskite film. I synthesized a polyurethane containing boronic ester groups and used it as a buried-interface passivation material. The resulting inverted perovskite solar cell reached a power conversion efficiency of 22.68%. I used Kelvin probe force microscopy, Fourier transform infrared spectroscopy, X-ray photoelectron spectroscopy, scanning electron microscopy, X-ray diffraction, steady-state photoluminescence, current-density-voltage measurements, dark current-density-voltage measurements, space-charge-limited current analysis, and long-term stability tests to understand the role of the polyurethane interlayer. The central conclusion of my study is that the polyurethane layer provides multiple chemical interactions with the perovskite, reduces interfacial defects, improves energy-level alignment, enlarges perovskite grains, suppresses nonradiative recombination, and increases the operational stability of the perovskite solar cell.
Background and Motivation
Metal halide perovskite solar cell technology has attracted intense attention because of low production cost and rapidly increasing efficiency. Since the first reported efficiency of 3.8%, single-junction perovskite solar cell efficiency has risen to 27.0%, making it comparable to conventional silicon-based cells. For further improvement of the perovskite solar cell, selecting suitable chemical substances as passivation materials is a key strategy. Polymeric materials are especially attractive because of their diverse properties and tunable physical and chemical characteristics. Among them, polyurethane has been used in perovskite solar cell modification to give self-healing ability, optimize surface morphology, and passivate grain boundaries. However, polyurethane as a buried-interface passivation material for an inverted perovskite solar cell has been studied relatively little. I therefore designed a boronic ester containing polyurethane and applied it at the hole transport layer/perovskite buried interface.
I chose an inverted device architecture because it offers low-temperature processing and good compatibility with organic hole transport materials. The general device structure I constructed is:
$$ \text{ITO}/\text{MeO-2PACz}/\text{PU}/\text{perovskite}/\text{PEAI}/\text{PC}_{61}\text{BM}/\text{BCP}/\text{Ag} $$
Here, ITO is the transparent bottom electrode, MeO-2PACz is the self-assembled hole transport monolayer, PU is the polyurethane buried-interface passivation layer, PEAI is the perovskite surface passivator, PC61BM is the electron transport layer, BCP is the hole-blocking layer, and Ag is the metal top electrode. The perovskite solar cell therefore contains a carefully engineered buried interface that can influence both charge extraction and film formation.
Materials and Reagents
I used the following materials for polyurethane synthesis and perovskite solar cell fabrication. I list them in a general way together with their roles in my experiments.
| Material | Role in My Experiment | Relevant Property |
|---|---|---|
| Isophorone diisocyanate | Diisocyanate monomer for polyurethane synthesis | Provides NCO groups for step-growth polymerization |
| Polycaprolactone diol | Polyol soft segment | Controls chain flexibility and film-forming behavior |
| Boronic ester containing diol | Functional monomer | Introduces boron-oxygen functional groups for passivation |
| N,N-dimethyl-N,N’-bis(2-hydroxypropyl)-1,3-propanediamine | Chain extender | Introduces tertiary amine and hydroxyl groups |
| Dibutyltin dilaurate | Catalyst | Accelerates urethane formation |
| N,N-dimethylacetamide | Reaction solvent | Controls viscosity during polymerization |
| MeO-2PACz | Hole transport self-assembled monolayer | Improves hole extraction from perovskite |
| FAI | Perovskite precursor | Formamidinium source |
| PbI2 | Perovskite precursor | Lead iodide source |
| CsI | Perovskite precursor | Cesium source for compositional tuning |
| MACl | Perovskite additive | Improves crystallization |
| PEAI | Surface passivator | Reduces surface defects |
| PC61BM | Electron transport material | Accepts electrons from perovskite |
| BCP | Hole-blocking layer | Suppresses hole leakage at the cathode |
| DMF and DMSO | Perovskite solvents | Control precursor solubility and crystallization |
| Chlorobenzene | Antisolvent | Triggers rapid perovskite nucleation |
| Ethanol and isopropanol | Processing solvents | Disperse hole transport and passivation materials |
Polyurethane Synthesis
I synthesized the boronic ester containing polyurethane through a step-growth polymerization route. First, I mixed isophorone diisocyanate and polycaprolactone diol at 80 °C under mechanical stirring for prepolymerization. I added dibutyltin dilaurate as a catalyst and continued the reaction for 3 h. Then I added the boronic ester containing diol and reacted for 2 h. Finally, I added the chain extender and continued the reaction for another 2 h. During the reaction, I added anhydrous N,N-dimethylacetamide as needed according to viscosity changes. The final product was a polyurethane containing boronic ester groups.
The main reaction steps can be summarized as follows:
$$ \text{IPDI} + \text{PCL} \xrightarrow{\text{catalyst}, 80^\circ\text{C}} \text{NCO-terminated prepolymer} $$
$$ \text{prepolymer} + \text{HDB} \xrightarrow{80^\circ\text{C}, 2\text{h}} \text{boronic ester incorporated chain} $$
$$ \text{chain} + \text{DPA} \xrightarrow{80^\circ\text{C}, 2\text{h}} \text{polyurethane} $$
I summarize the nominal feed amounts and molar quantities in the table below.
| Component | Mass Used | Molar Amount | Function |
|---|---|---|---|
| Isophorone diisocyanate | 3 g | 13.5 mmol | Diisocyanate monomer |
| Polycaprolactone diol | 4.5 g | 4.5 mmol | Polyol soft segment |
| Boronic ester containing diol | 0.91 g | 2.5 mmol | Boronic ester functional monomer |
| Chain extender | 1.09 g | 5.0 mmol | Chain extension and functional group introduction |
The polyurethane chemical structure contains urethane linkages, polycaprolactone segments, tertiary amine groups, hydroxyl groups, and boronic ester groups. The boronic ester group is especially important because it can interact with lead-containing defects at the perovskite surface. The urethane carbonyl group can coordinate to undercoordinated lead ions. The amine and hydroxyl groups can form hydrogen bonds with halide ions and organic cations.
Perovskite Solar Cell Fabrication
I fabricated the inverted perovskite solar cell step by step. First, I cleaned ITO substrates by ultrasonication in pure water and ethanol for 20 min each and dried them with a gas gun. I treated the cleaned ITO with ultraviolet-ozone for 20 min. I then spin-coated a 0.5 mg/mL MeO-2PACz ethanol solution onto ITO at 3000 rpm and annealed it at 100 °C for 10 min. Next, I spin-coated a polyurethane solution in DMF onto the ITO/MeO-2PACz substrate at 3000 rpm and annealed it at 100 °C for 10 min. The polyurethane concentration was varied to find the best buried-interface condition.
For the perovskite precursor, I dissolved PbBr2, MABr, CsI, MACl, FAI, and PbI2 in a DMF:DMSO = 4:1 mixed solvent. I deposited the precursor on the polyurethane layer by a two-step spin-coating method. The first step was 1000 rpm for 5 s, and the second step was 5000 rpm for 40 s. Eight seconds before the end of the second step, I dropped chlorobenzene as an antisolvent. I annealed the film at 110 °C for 20 min. I then spin-coated a 3 mg/mL PEAI isopropanol solution onto the perovskite layer at 4000 rpm. After that, I spin-coated a 20 mg/mL PC61BM chlorobenzene solution at 2000 rpm, followed by a 0.5 mg/mL BCP isopropanol solution at 5000 rpm. Finally, I thermally evaporated about 70 nm of Ag as the top electrode.
| Step | Material | Concentration | Spin-Coating Condition | Annealing |
|---|---|---|---|---|
| Substrate cleaning | Water and ethanol | Not applicable | Ultrasonication for 20 min each | Not applicable |
| UV-ozone treatment | Not applicable | Not applicable | 20 min | Not applicable |
| Hole transport layer | MeO-2PACz in ethanol | 0.5 mg/mL | 3000 rpm | 100 °C for 10 min |
| Buried-interface layer | Polyurethane in DMF | 0.05–0.2 mg/mL | 3000 rpm | 100 °C for 10 min |
| Perovskite layer | Mixed precursor | As described | 1000 rpm for 5 s, then 5000 rpm for 40 s | 110 °C for 20 min |
| Surface passivation | PEAI in isopropanol | 3 mg/mL | 4000 rpm | Not applicable |
| Electron transport layer | PC61BM in chlorobenzene | 20 mg/mL | 2000 rpm | Not applicable |
| Hole-blocking layer | BCP in isopropanol | 0.5 mg/mL | 5000 rpm | Not applicable |
| Top electrode | Ag | Not applicable | Thermal evaporation | About 70 nm |
Characterization Methods
I characterized the photovoltaic performance of the perovskite solar cell using a solar simulator and a digital source meter. I measured current density-voltage curves under simulated sunlight. I used Kelvin probe force microscopy to measure the surface contact potential difference of the hole transport layer before and after polyurethane deposition. I used Fourier transform infrared spectroscopy and X-ray photoelectron spectroscopy to analyze functional group shifts and elemental compositions. I used scanning electron microscopy to examine perovskite film surface morphology. I used X-ray diffraction to study crystal structure. I used steady-state photoluminescence to evaluate radiative recombination and nonradiative recombination. I also used dark current-density-voltage curves and single-carrier devices to estimate defect densities.
| Technique | Information Obtained | Role in My Analysis |
|---|---|---|
| Solar simulator and source meter | Current density-voltage curves | Determined VOC, JSC, FF, and PCE |
| Kelvin probe force microscopy | Contact potential difference | Evaluated surface potential and energy-level alignment |
| Fourier transform infrared spectroscopy | Vibrational modes | Identified B-O, C=O, N-H, and C-O interactions |
| X-ray photoelectron spectroscopy | Core-level binding energies | Probed chemical environment of C, Pb, and I |
| Scanning electron microscopy | Surface morphology | Compared grain size and film uniformity |
| X-ray diffraction | Crystal structure | Tracked perovskite phase and PbI2 residue |
| Steady-state photoluminescence | Emission intensity | Assessed nonradiative recombination |
| Dark current-density-voltage measurement | Leakage current | Estimated defect-related conduction |
| Space-charge-limited current | Trap-filled limit voltage | Quantified relative defect density |
Photovoltaic Performance of the Perovskite Solar Cell
I first optimized the polyurethane concentration because the buried-interface layer must be thin enough to allow charge transport but thick enough to passivate defects. I tested control devices and devices with 0.05, 0.1, and 0.2 mg/mL polyurethane. The best performance was obtained with 0.1 mg/mL polyurethane. The power conversion efficiency of the perovskite solar cell was calculated using:
$$ \eta = \frac{V_{\mathrm{OC}} J_{\mathrm{SC}} FF}{P_{\mathrm{in}}} $$
where \(V_{\mathrm{OC}}\) is the open-circuit voltage, \(J_{\mathrm{SC}}\) is the short-circuit current density, \(FF\) is the fill factor, and \(P_{\mathrm{in}}\) is the incident power density. In my measurements, \(P_{\mathrm{in}}\) was 100 mW cm\(^{-2}\). I summarize the best photovoltaic parameters in the table below.
| Polyurethane Concentration (mg/mL) | VOC (V) | JSC (mA/cm2) | FF (%) | PCE (%) |
|---|---|---|---|---|
| Control | 1.104 | 23.04 | 83.90 | 21.33 |
| 0.05 | 1.119 | 23.31 | 85.09 | 22.19 |
| 0.1 | 1.132 | 23.38 | 85.71 | 22.68 |
| 0.2 | 1.135 | 23.18 | 84.78 | 22.30 |
The control perovskite solar cell had a PCE of 21.33%, with VOC = 1.104 V, JSC = 23.04 mA cm\(^{-2}\), and FF = 83.90%. After introducing 0.1 mg/mL polyurethane, the perovskite solar cell reached a PCE of 22.68%, with VOC = 1.132 V, JSC = 23.38 mA cm\(^{-2}\), and FF = 85.71%. The improvement in VOC indicates reduced nonradiative recombination and better energy-level alignment. The increase in JSC suggests improved charge extraction and larger grain size. The increase in FF suggests lower series resistance and reduced defect-related recombination. At 0.2 mg/mL, the performance decreased slightly, probably because an overly thick polyurethane layer increased the tunneling or transport barrier at the buried interface. Therefore, 0.1 mg/mL was the optimal concentration for the buried-interface passivation of the inverted perovskite solar cell.
I also measured dark current-density-voltage curves. A lower dark current density at a given reverse or forward bias indicates suppressed leakage and fewer defect-related conduction pathways. In my optimized device, the polyurethane-modified perovskite solar cell showed reduced dark current density relative to the control, which supports the conclusion that the polyurethane interlayer reduces interfacial defects. I used single-hole devices to quantify this effect more directly. The trap-filled limit voltage \(V_{\mathrm{TFL}}\) is related to the defect density \(N_t\) by:
$$ V_{\mathrm{TFL}} = \frac{q N_t L^2}{2 \epsilon \epsilon_0} $$
so that
$$ N_t = \frac{2 \epsilon \epsilon_0 V_{\mathrm{TFL}}}{q L^2} $$
where \(q\) is the elementary charge, \(L\) is the active layer thickness, \(\epsilon\) is the relative dielectric constant, and \(\epsilon_0\) is the vacuum permittivity. Because the thickness and dielectric constant are approximately the same for the two devices, the relative defect density can be estimated from the ratio of \(V_{\mathrm{TFL}}\) values.
| Device | VTFL (V) | Relative Nt (a.u.) | Interpretation |
|---|---|---|---|
| Control | 1.17 | 1.00 | Reference defect density |
| Polyurethane-modified | 1.00 | 0.855 | About 14.5% reduction in relative defect density |
$$ \frac{N_{t,\mathrm{PU}}}{N_{t,\mathrm{control}}} = \frac{1.00}{1.17} = 0.855 $$
$$ \text{Relative reduction} = 1 – 0.855 = 0.145 = 14.5\% $$
The reduced \(V_{\mathrm{TFL}}\) and estimated defect density indicate that the polyurethane layer decreases trap states in the perovskite solar cell. This is consistent with the improved VOC and FF. I also tested long-term stability. I stored unencapsulated devices in a nitrogen glovebox at 25 °C for 43 days. The control perovskite solar cell retained 72% of its initial efficiency, while the polyurethane-modified perovskite solar cell retained 94%. This improvement suggests that the buried-interface polyurethane layer suppresses defect formation and protects the perovskite from degradation.
| Storage Time (Days) | Control Retention (%) | Polyurethane-Modified Retention (%) |
|---|---|---|
| 0 | 100 | 100 |
| 43 | 72 | 94 |
Kelvin Probe Force Microscopy and Energy-Level Alignment
I used Kelvin probe force microscopy to examine the surface potential of the hole transport layer before and after polyurethane modification. The contact potential difference is related to the work function difference between the tip and the sample:
$$ V_{\mathrm{CPD}} = \frac{\Phi_{\mathrm{tip}} – \Phi_{\mathrm{sample}}}{e} $$
where \(\Phi_{\mathrm{tip}}\) is the tip work function, \(\Phi_{\mathrm{sample}}\) is the sample work function, and \(e\) is the elementary charge. The average contact potential difference decreased from 72.7 mV for the ITO/MeO-2PACz substrate to 54.8 mV for the ITO/MeO-2PACz/polyurethane substrate. The change is:
$$ \Delta V_{\mathrm{CPD}} = 54.8 – 72.7 = -17.9 \ \mathrm{mV} $$
A decrease in contact potential difference corresponds to an increase in sample work function and a downward shift of the Fermi level. The estimated work function change is:
$$ \Delta \Phi_{\mathrm{sample}} = -e \Delta V_{\mathrm{CPD}} = +17.9 \ \mathrm{meV} $$
This change is beneficial for energy-level matching between the hole transport layer and the perovskite layer. Better energy-level alignment can reduce the hole extraction barrier and improve hole transport in the perovskite solar cell. The Kelvin probe force microscopy result therefore supports the photovoltaic improvement observed after polyurethane modification.
| Substrate | Average CPD (mV) | Relative Change | Expected Effect |
|---|---|---|---|
| ITO/MeO-2PACz | 72.7 | Reference | Reference energy alignment |
| ITO/MeO-2PACz/PU | 54.8 | −17.9 mV | Downward Fermi-level shift and improved hole extraction |
Morphology and Crystallization of the Perovskite Solar Cell
I examined the surface morphology of perovskite films deposited on ITO/MeO-2PACz and ITO/MeO-2PACz/polyurethane substrates. The perovskite film on the polyurethane-modified substrate showed larger grain size. The larger grain size can be attributed to strong interaction between polyurethane and Pb2+, which slows the crystallization rate of the perovskite film. A slower crystallization rate often allows grains to grow more uniformly and reduces grain boundary density. In a perovskite solar cell, fewer grain boundaries generally mean fewer defects and less nonradiative recombination.
I also used X-ray diffraction to study the crystallization kinetics. The perovskite film on the polyurethane-modified substrate showed a stronger characteristic peak from the (001) plane. At the same time, the PbI2 peak intensity decreased significantly. The PbI2 peak intensity is related to the degree of phase conversion. A weaker PbI2 peak indicates more complete conversion from precursor phases to the perovskite phase. Therefore, polyurethane modification improves the crystallization and phase purity of the perovskite film in the perovskite solar cell.
I estimated the crystallite size using the Scherrer equation:
$$ D = \frac{K \lambda}{\beta \cos \theta} $$
where \(D\) is the crystallite size, \(K\) is the shape factor, \(\lambda\) is the X-ray wavelength, \(\beta\) is the full width at half maximum, and \(\theta\) is the Bragg angle. Although I did not rely on a single numerical value for the whole film, the trend of sharper and stronger diffraction peaks is consistent with larger crystalline domains and improved structural order.
| Characterization | Control Perovskite Film | Polyurethane-Modified Perovskite Film | Interpretation |
|---|---|---|---|
| SEM grain size | Smaller grains | Larger grains | Slower crystallization and better film growth |
| XRD (001) intensity | Reference | Stronger | Improved preferred orientation and crystallinity |
| XRD PbI2 intensity | Higher | Lower | More complete phase conversion |
| Steady-state PL intensity | Lower | Higher | Reduced nonradiative recombination |
Steady-state photoluminescence spectra showed that the polyurethane-modified perovskite film had stronger PL intensity. A stronger PL intensity usually means that nonradiative recombination is reduced. This is consistent with defect passivation at the buried interface. In the perovskite solar cell, reduced nonradiative recombination directly contributes to higher VOC and improved overall efficiency.
X-ray Photoelectron Spectroscopy Analysis
I used X-ray photoelectron spectroscopy to analyze chemical interactions at the buried interface. I examined C 1s, Pb 4f, and I 3d core levels. In the C 1s spectrum, the control perovskite film showed three peaks assigned to C-C bonds, the N+=CH-N bond in FA+, and the CH3-N bond in MA+. After polyurethane modification, a new peak appeared, which I assigned to the C=O bond of polyurethane. In addition, the N+=CH-N and CH3-N peaks shifted in binding energy. This shift indicates that FA+ and MA+ interact with the C-O groups of the urethane linkage through hydrogen bonding.
In the Pb 4f spectrum, the two characteristic peaks shifted to lower binding energy after polyurethane modification. This shift is due to electron transfer from the lone pair of the C=O group to Pb2+. Such coordination reduces the electron deficiency of undercoordinated Pb2+ and passivates lead-related defects. In the I 3d spectrum, the two characteristic peaks also shifted after polyurethane modification. This shift can be attributed to hydrogen bonding between the N-H group of polyurethane and I−. This interaction helps stabilize halide ions at the buried interface.
| Core Level | Species | Control Film | Polyurethane-Modified Film | Proposed Interaction |
|---|---|---|---|---|
| C 1s | C-C | Reference peak | Reference peak | No specific interaction |
| C 1s | N+=CH-N in FA+ | Observed | Binding energy shift | Hydrogen bonding with C-O |
| C 1s | CH3-N in MA+ | Observed | Binding energy shift | Hydrogen bonding with C-O |
| C 1s | C=O from polyurethane | Absent | New peak | Urethane carbonyl present at interface |
| Pb 4f | Pb2+ | Reference peaks | Shift to lower binding energy | Electron donation from C=O to Pb2+ |
| I 3d | I− | Reference peaks | Binding energy shift | Hydrogen bonding with N-H |
The X-ray photoelectron spectroscopy results provide direct evidence that polyurethane does not simply act as a physical spacer. It chemically interacts with the perovskite bottom surface. These interactions are crucial for passivating defects and stabilizing the buried interface of the perovskite solar cell.
Fourier Transform Infrared Spectroscopy Analysis
I further used Fourier transform infrared spectroscopy to identify the functional groups responsible for chemical passivation. I compared neat polyurethane, a polyurethane/PbI2 mixture, and a film obtained by annealing a polyurethane solution in DMF. In neat polyurethane, the B-O vibration was observed. In the polyurethane/PbI2 mixture, the B-O vibration disappeared. To exclude the possibility that the B-O vibration disappeared because of DMF dissolution and annealing, I tested the polyurethane film prepared from DMF solution after annealing. The B-O vibration remained in that film. Therefore, the disappearance of the B-O vibration in the polyurethane/PbI2 mixture is due to interaction with PbI2, not due to solvent or thermal treatment.
| Sample | B-O Vibration | Interpretation |
|---|---|---|
| Neat polyurethane | Observed | Boronic ester group is present |
| Polyurethane and PbI2 mixture | Disappeared | B-O group interacts with PbI2 |
| Polyurethane film from DMF after annealing | Observed | Solvent and annealing do not remove B-O |
This result indicates that the boronic ester group participates in chemical passivation at the buried interface. The B-O group can coordinate to Pb2+ and thereby reduce lead-related defects. Because the buried interface is often difficult to access after device fabrication, this FTIR comparison provides a useful model for understanding the interfacial chemistry of the perovskite solar cell.
Chemical Passivation Mechanism
On the basis of X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy, I propose that polyurethane passivates the perovskite bottom surface through multiple chemical interactions. These interactions include Lewis acid-base coordination, hydrogen bonding, and electron donation. The main reactions can be written as:
$$ \mathrm{PU{-}C{=}O} + \mathrm{Pb^{2+}} \rightleftharpoons [\mathrm{PU{-}C{=}O}\rightarrow \mathrm{Pb^{2+}}] $$
$$ \mathrm{PU{-}B{-}O} + \mathrm{Pb^{2+}} \rightleftharpoons [\mathrm{PU{-}B{-}O}\rightarrow \mathrm{Pb^{2+}}] $$
$$ \mathrm{PU{-}NH} + \mathrm{I^-} \rightleftharpoons [\mathrm{PU{-}NH}\cdots \mathrm{I^-}] $$
$$ \mathrm{PU{-}C{-}O} + \mathrm{FA^+}/\mathrm{MA^+} \rightleftharpoons [\mathrm{PU{-}C{-}O}\cdots \mathrm{H{-}N}] $$
These reactions show that the polyurethane layer can target several types of defects at the buried interface. Undercoordinated Pb2+ is passivated by C=O and B-O groups. Halide vacancies are passivated by N-H hydrogen bonding. Organic cation vacancies or unsatisfied organic cations are stabilized by C-O hydrogen bonding. The combination of these interactions reduces trap density and suppresses nonradiative recombination in the perovskite solar cell.
| Polyurethane Functional Group | Target Defect or Species | Interaction Type | Effect on Perovskite Solar Cell |
|---|---|---|---|
| C=O | Undercoordinated Pb2+ | Lewis acid-base coordination | Reduces lead-related traps and nonradiative recombination |
| B-O | Pb2+ and PbI2-related defects | Coordination | Passivates the buried interface and improves phase stability |
| N-H | I− and halide vacancies | Hydrogen bonding | Stabilizes halide ions and reduces ionic defects |
| C-O | FA+ and MA+ | Hydrogen bonding | Improves organic cation anchoring and film stability |
The chemical passivation mechanism explains several experimental observations. First, the reduced dark current and lower \(V_{\mathrm{TFL}}\) indicate fewer defect-related conduction pathways. Second, the enhanced PL intensity indicates suppressed nonradiative recombination. Third, the improved VOC and FF indicate better interfacial charge behavior. Fourth, the improved stability indicates that the polyurethane layer stabilizes the perovskite bottom surface. Therefore, the buried-interface polyurethane layer plays a multifunctional role in the inverted perovskite solar cell.
Interfacial Charge Transfer and Device Physics
I further considered the device physics of the perovskite solar cell. The short-circuit current density can be expressed as:
$$ J_{\mathrm{SC}} = q \int \Phi_{\mathrm{AM1.5G}}(\lambda) EQE(\lambda) d\lambda $$
where \(\Phi_{\mathrm{AM1.5G}}\) is the photon flux of the AM1.5G spectrum and \(EQE\) is the external quantum efficiency. Improved charge extraction at the buried interface can increase EQE and therefore \(J_{\mathrm{SC}}\). The open-circuit voltage can be related to recombination by:
$$ V_{\mathrm{OC}} = \frac{n k_{\mathrm{B}} T}{q} \ln\left(\frac{J_{\mathrm{ph}}}{J_0} + 1\right) $$
where \(n\) is the ideality factor, \(k_{\mathrm{B}}\) is the Boltzmann constant, \(T\) is temperature, \(J_{\mathrm{ph}}\) is the photogenerated current density, and \(J_0\) is the saturation current density. Defect passivation reduces \(J_0\), which increases \(V_{\mathrm{OC}}\). The fill factor is influenced by series resistance and shunt resistance:
$$ FF = \frac{V_{\mathrm{MPP}} J_{\mathrm{MPP}}}{V_{\mathrm{OC}} J_{\mathrm{SC}}} $$
where \(V_{\mathrm{MPP}}\) and \(J_{\mathrm{MPP}}\) are the voltage and current density at the maximum power point. Reduced defect density lowers recombination losses and improves FF. The observed increases in VOC, JSC, and FF after polyurethane modification are therefore consistent with reduced interfacial recombination and improved charge transfer in the perovskite solar cell.
| Photovoltaic Parameter | Physical Meaning | Observed Change after PU | Proposed Reason |
|---|---|---|---|
| VOC | Open-circuit voltage | Increased from 1.104 V to 1.132 V | Reduced nonradiative recombination and better energy alignment |
| JSC | Short-circuit current density | Increased from 23.04 to 23.38 mA/cm2 | Improved charge extraction and larger grains |
| FF | Fill factor | Increased from 83.90% to 85.71% | Lower series resistance and fewer trap-assisted losses |
| PCE | Power conversion efficiency | Increased from 21.33% to 22.68% | Combined improvement in VOC, JSC, and FF |
Stability of the Perovskite Solar Cell
Stability is a critical challenge for the perovskite solar cell. In my study, the polyurethane-modified device retained 94% of its initial efficiency after 43 days in a nitrogen glovebox, whereas the control device retained only 72%. This improvement can be explained by the chemical passivation of the buried interface. By coordinating to Pb2+, hydrogen bonding to I−, and anchoring organic cations, polyurethane reduces the number of undercoordinated ions and vacancy defects. These defects are often initiation sites for degradation. In addition, the polyurethane interlayer may prevent direct contact between the perovskite and the hole transport layer from causing unwanted interfacial reactions. The improved stability suggests that the buried-interface polyurethane layer is beneficial not only for efficiency but also for the long-term operation of the perovskite solar cell.
| Stability Metric | Control Device | Polyurethane-Modified Device | Improvement |
|---|---|---|---|
| Initial PCE | 21.33% | 22.68% | +1.35 percentage points |
| PCE after 43 days | 72% of initial | 94% of initial | +22 percentage points retention |
| Main degradation factor | Interfacial defects and recombination | Suppressed by multiple chemical interactions | More stable buried interface |
Comparison of Control and Polyurethane-Modified Perovskite Solar Cell
I summarize the overall comparison between the control and polyurethane-modified perovskite solar cell in the following table. The table combines photovoltaic performance, defect density, surface potential, morphology, crystallization, recombination, and stability.
| Property | Control Perovskite Solar Cell | Polyurethane-Modified Perovskite Solar Cell | Benefit of Polyurethane |
|---|---|---|---|
| VOC | 1.104 V | 1.132 V | Higher voltage due to reduced recombination |
| JSC | 23.04 mA/cm2 | 23.38 mA/cm2 | Better extraction and larger grains |
| FF | 83.90% | 85.71% | Lower series resistance and fewer traps |
| PCE | 21.33% | 22.68% | Overall performance enhancement |
| VTFL | 1.17 V | 1.00 V | Lower relative defect density |
| Surface CPD | 72.7 mV | 54.8 mV | Better energy-level alignment |
| Grain size | Smaller | Larger | Reduced grain boundary density |
| XRD PbI2 peak | Stronger | Weaker | More complete phase conversion |
| PL intensity | Lower | Higher | Reduced nonradiative recombination |
| Stability after 43 days | 72% retention | 94% retention | Improved long-term stability |
Interpretation of the Buried Interface
The buried interface is one of the most important regions in an inverted perovskite solar cell. During perovskite film formation, the bottom surface is in contact with the hole transport layer. If this interface contains defects, it can cause charge accumulation, nonradiative recombination, and poor adhesion. In my approach, the polyurethane layer modifies this interface before perovskite deposition. The polyurethane functional groups remain at the bottom surface and interact with the perovskite precursors and final perovskite film. This pre-placed interlayer strategy is different from post-treatment of the top surface. It allows chemical passivation exactly where the perovskite nucleates and grows. As a result, the polyurethane layer influences both crystallization and defect chemistry.
I also note that the polyurethane layer must be carefully controlled. If it is too thin, the passivation effect may be insufficient. If it is too thick, it may block charge transport or increase series resistance. My concentration study showed that 0.1 mg/mL is the best compromise. At this concentration, the polyurethane layer provides enough functional groups for passivation while maintaining efficient hole extraction in the perovskite solar cell.
Role of Boronic Ester Groups
The boronic ester group is a distinctive feature of my polyurethane design. Boronic ester groups contain boron-oxygen bonds that can interact with lead ions and lead-containing defects. The FTIR results showed that the B-O vibration disappeared in the polyurethane/PbI2 mixture, indicating a direct interaction with PbI2. This interaction is important because PbI2 residues and undercoordinated Pb2+ are common defects at the buried interface. By coordinating to these species, the boronic ester group can reduce defect density and improve phase stability. In the perovskite solar cell, this contributes to higher efficiency and better stability.
The boron-oxygen interaction can be represented as:
$$ \mathrm{B{-}O} + \mathrm{Pb^{2+}} \rightarrow \mathrm{B{-}O}\rightarrow \mathrm{Pb^{2+}} $$
This coordination is part of the broader chemical passivation mechanism. It complements the C=O coordination, N-H hydrogen bonding, and C-O hydrogen bonding. The multiple functional groups in polyurethane make it a versatile passivation material for the perovskite solar cell.
Role of Urethane and Amine Groups
The urethane linkage provides C=O and N-H groups. The C=O group acts as a Lewis base and donates electron density to Pb2+. The N-H group acts as a hydrogen-bond donor to I−. The tertiary amine and hydroxyl groups from the chain extender provide additional lone pairs and hydrogen-bonding sites. These groups are distributed along the polymer chain, which allows multidentate interactions with the perovskite surface. Multidentate passivation can be more stable than monodentate passivation because multiple bonds cooperate to anchor the polymer at the interface. This may explain why the polyurethane-modified perovskite solar cell showed improved stability as well as improved efficiency.
I summarize the proposed multidentate passivation as follows:
$$ \mathrm{PU} \leftrightarrow \mathrm{Pb^{2+}} \quad \text{through C=O and B-O} $$
$$ \mathrm{PU} \leftrightarrow \mathrm{I^-} \quad \text{through N-H} $$
$$ \mathrm{PU} \leftrightarrow \mathrm{FA^+}/\mathrm{MA^+} \quad \text{through C-O} $$
These interactions reduce ionic defects, improve interfacial contact, and stabilize the perovskite lattice at the buried interface. The perovskite solar cell therefore benefits from both electronic passivation and structural stabilization.
Implications for Future Perovskite Solar Cell Design
My results suggest several design principles for future perovskite solar cell development. First, buried-interface engineering should be considered as important as top-surface passivation. Second, polymers with multiple functional groups can provide simultaneous passivation of different defect types. Third, boronic ester groups are useful for lead-related defect passivation. Fourth, the interlayer thickness must be optimized to balance passivation and charge transport. Fifth, chemical interactions at the buried interface can improve both efficiency and stability.
In future work, I would like to combine polyurethane buried-interface passivation with other strategies, such as additive engineering, compositional tuning, and encapsulation. I would also like to study the spatial distribution of polyurethane functional groups using advanced depth-profiling techniques. In addition, I would like to test the polyurethane interlayer in larger-area perovskite solar cell devices and in tandem configurations. These studies could further clarify the role of polymer chemistry in perovskite solar cell stability and performance.
| Design Principle | Implementation in My Work | Expected Benefit |
|---|---|---|
| Buried-interface passivation | Polyurethane deposited on hole transport layer | Defect reduction at the bottom perovskite surface |
| Multidentate chemistry | C=O, B-O, N-H, C-O groups | Simultaneous passivation of Pb2+, I−, FA+, and MA+ defects |
| Thickness optimization | 0.1 mg/mL polyurethane | Balanced passivation and charge transport |
| Energy-level tuning | Surface CPD decreased by 17.9 mV | Improved hole extraction |
| Crystallization control | Larger grains and weaker PbI2 peak | Higher quality perovskite film |
| Stability enhancement | 94% retention after 43 days | Longer device lifetime |
Summary of Experimental Results
In summary, I synthesized a boronic ester containing polyurethane and used it as a buried-interface passivation material for an inverted perovskite solar cell. The best device achieved a power conversion efficiency of 22.68%, with VOC = 1.132 V, JSC = 23.38 mA cm\(^{-2}\), and FF = 85.71%. The polyurethane layer reduced the surface contact potential difference of the hole transport layer from 72.7 mV to 54.8 mV, which improved energy-level alignment. It increased perovskite grain size, enhanced the (001) XRD peak, reduced the PbI2 peak, and increased PL intensity. X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy revealed multiple chemical interactions between polyurethane and the perovskite, including C=O coordination to Pb2+, B-O interaction with PbI2, N-H hydrogen bonding to I−, and C-O hydrogen bonding to FA+ and MA+. These interactions passivated the buried interface, reduced defect density, suppressed nonradiative recombination, and improved stability. The polyurethane-modified perovskite solar cell retained 94% of its initial efficiency after 43 days, compared with 72% for the control.
Overall, my study demonstrates that polyurethane with boronic ester groups is a promising buried-interface passivation material for the perovskite solar cell. It provides chemical passivation, morphology control, energy-level alignment, and stability enhancement. I believe that this strategy can contribute to the further development of efficient and stable perovskite solar cell technology.
$$ \eta_{\mathrm{best}} = 22.68\% $$
$$ V_{\mathrm{OC,best}} = 1.132 \ \mathrm{V} $$
$$ J_{\mathrm{SC,best}} = 23.38 \ \mathrm{mA\,cm^{-2}} $$
$$ FF_{\mathrm{best}} = 85.71\% $$
$$ \text{Retention after 43 days} = 94\% $$