I designed and evaluated a thioalkyl-chain-modified titanyl phthalocyanine derivative, TiOPc-NPSC3, as a multifunctional additive for perovskite solar cell fabrication. My central objective was to determine whether a molecular additive introduced through the antisolvent could simultaneously regulate crystallization, passivate undercoordinated lead defects, suppress metallic lead formation, improve interfacial charge extraction, and protect the perovskite absorber from moisture. The work therefore connects molecular design, film-growth chemistry, defect physics, and device stability within a single experimental framework. I used a formal n-i-p perovskite solar cell architecture, with SnO2 as the electron transport layer, a mixed-cation and mixed-halide perovskite absorber, PTAA as the hole transport layer, and silver as the back electrode. The additive was dissolved or saturated in anisole, which served as the antisolvent during the two-step spin-coating process. This route allowed TiOPc-NPSC3 to interact with the perovskite film during the most dynamic stage of nucleation and crystal growth, rather than merely acting as a post-treatment surface layer.
My hypothesis was that the sulfur atoms in the thioalkyl chains would coordinate strongly with undercoordinated Pb2+ sites on the perovskite surface and at grain boundaries. Such coordination should increase the electron density around Pb2+, shift the Pb 4f binding energy to lower values, reduce the density of deep-level trap states, and inhibit the reduction of Pb2+ to Pb0. In parallel, the large pi-conjugated phthalocyanine core should contribute to charge transport, while the hydrophobic alkyl sulfide chains should increase the water contact angle and slow moisture-driven degradation. I also expected that the additive would influence the crystallization kinetics, producing larger grains, fewer pinholes, lower surface roughness, and a more compact cross-sectional morphology. The combined effect should be an increase in power conversion efficiency and a substantial improvement in long-term storage stability.
I present the results in a sequence that moves from molecular-level interactions to microscopic film properties, then to optical and electrical behavior, and finally to complete perovskite solar cell performance and stability. Throughout the discussion, I use quantitative tables and equations to summarize the key relationships. The most important performance metrics are the open-circuit voltage, short-circuit current density, fill factor, power conversion efficiency, external quantum efficiency, dark current, grain size, root-mean-square roughness, X-ray diffraction peak ratio, photoluminescence intensity ratio, water contact angle, and efficiency retention after long-term storage.
1. Rationale and Molecular Design
Perovskite solar cell research has advanced rapidly because organic-inorganic hybrid perovskites combine tunable bandgaps, high absorption coefficients, long carrier diffusion lengths, and low-cost solution processing. However, solution-processed polycrystalline perovskite films inevitably contain defects at surfaces and grain boundaries. These defects are not passive structural imperfections. They act as non-radiative recombination centers, reduce quasi-Fermi-level splitting, lower the open-circuit voltage, and provide initiation sites for moisture, oxygen, heat, and light-induced degradation. Therefore, an effective additive for a perovskite solar cell should do more than fill pinholes. It should coordinate to undercoordinated ions, modulate crystallization, improve charge extraction, and increase environmental resistance.
I selected a titanyl phthalocyanine derivative because phthalocyanines possess a rigid, planar, highly conjugated macrocycle. This structural motif supports charge delocalization and thermal and chemical stability. The central titanyl unit can influence the electronic structure, while peripheral substituents can tune solubility, steric bulk, and surface interaction. In TiOPc-NPSC3, the peripheral thioalkyl chains provide sulfur atoms with lone-pair electrons. Sulfur is a soft Lewis base and can coordinate to Pb2+, which is a borderline-to-soft Lewis acid. This interaction is expected to be stronger and more selective than simple van der Waals adsorption. The alkyl chains also introduce hydrophobicity, which is beneficial for moisture resistance in a perovskite solar cell.
The molecular formula and mass spectrometric data are consistent with the proposed structure. I use the following general mass relation to confirm the molecular assignment:
$$ m/z_{\mathrm{calcd}} = 1168.24,\qquad m/z_{\mathrm{obsd}} = 1168.24 $$
The molecular structure can be summarized as a titanyl phthalocyanine core bearing eight peripheral thioalkyl units. The sulfur atoms are the key passivation sites, while the phthalocyanine core contributes to charge delocalization and film stability. The design logic can be expressed as a multi-functionality condition:
$$ F_{\mathrm{additive}} = f_{\mathrm{passivation}} + f_{\mathrm{crystallization}} + f_{\mathrm{transport}} + f_{\mathrm{hydrophobicity}} $$
For a high-performance perovskite solar cell, each term must be optimized. If passivation is too strong, it may insulate the surface. If crystallization control is too strong, it may leave residual additive or disrupt phase purity. If hydrophobicity is achieved with insulating long chains, charge extraction may suffer. TiOPc-NPSC3 is designed to balance these effects because the sulfur coordination sites are distributed around a conjugated core, allowing defect binding without completely blocking charge transfer.
2. Experimental Design and Device Fabrication
I fabricated formal n-i-p perovskite solar cells on fluorine-doped tin oxide glass. The substrate was cleaned sequentially with detergent-containing water, pure water, and ethanol under ultrasonication. After drying, the substrate was treated with ultraviolet ozone. A SnO2 nanoparticle dispersion diluted with deionized water was spin-coated onto the FTO substrate and annealed. The SnO2-coated substrate was then ultraviolet-treated again before perovskite deposition. The perovskite precursor solution had a concentration of 1.4 mol/L and was deposited in ambient conditions with controlled relative humidity and temperature. The spin-coating process used two steps: a low-speed step to spread the solution and a high-speed step to form a uniform wet film. During the high-speed step, anisole was dropped as the antisolvent. For modified samples, the anisole was saturated with TiOPc-NPSC3. The films were annealed to complete crystallization.
After perovskite formation, I transferred the samples to a nitrogen glovebox. A surface passivation layer based on 4-methoxyphenethylammonium iodide was dynamically spin-coated, followed by brief annealing. Then a PTAA solution containing TPFB was spin-coated as the hole transport layer. Finally, a silver electrode was thermally evaporated under high vacuum. The device stack was:
$$ \mathrm{FTO}/\mathrm{SnO_2}/\mathrm{perovskite}/\mathrm{PTAA}/\mathrm{Ag} $$
The perovskite composition was:
$$ \left(\mathrm{FA}_{0.95}\mathrm{Cs}_{0.05}\mathrm{PbI}_3\right)_{0.975}\left(\mathrm{MAPbBr}_3\right)_{0.025} $$
This composition combines formamidinium, cesium, methylammonium, lead, iodide, and bromide to balance phase stability, bandgap, and charge-carrier lifetime. The additive was introduced only through the antisolvent, which means that the perovskite precursor solution itself remained unchanged. This is important because it avoids possible colloidal aggregation or premature nucleation in the precursor solution. Instead, TiOPc-NPSC3 interacts with the perovskite during antisolvent-induced supersaturation, when the film is highly dynamic and defect sites are forming.
| Parameter | Control Condition | TiOPc-NPSC3 Condition |
|---|---|---|
| Substrate | FTO glass | FTO glass |
| Electron transport layer | SnO2 nanoparticles | SnO2 nanoparticles |
| Perovskite composition | (FA0.95Cs0.05PbI3)0.975(MAPbBr3)0.025 | (FA0.95Cs0.05PbI3)0.975(MAPbBr3)0.025 |
| Antisolvent | Anisole | Anisole saturated with TiOPc-NPSC3 |
| Hole transport layer | PTAA with TPFB | PTAA with TPFB |
| Back electrode | Ag | Ag |
| Device architecture | n-i-p | n-i-p |
The processing conditions were chosen to be compatible with scalable coating methods. The antisolvent volume, spin speed, annealing temperature, and annealing time were kept constant between the control and modified samples so that the observed differences could be attributed primarily to TiOPc-NPSC3. I used a metal mask to define the active area during current density-voltage measurements. The active area was:
$$ A_{\mathrm{active}} = 0.0735\ \mathrm{cm^2} $$
The power conversion efficiency of a perovskite solar cell is calculated from the current density-voltage curve using:
$$ \eta = \frac{P_{\max}}{P_{\mathrm{in}}} = \frac{V_{\mathrm{OC}} J_{\mathrm{SC}} FF}{P_{\mathrm{in}}} $$
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 optical power density. Under standard illumination:
$$ P_{\mathrm{in}} = 100\ \mathrm{mW\,cm^{-2}} $$
The fill factor is defined as:
$$ FF = \frac{V_{\mathrm{MPP}} J_{\mathrm{MPP}}}{V_{\mathrm{OC}} J_{\mathrm{SC}}} $$
where \(V_{\mathrm{MPP}}\) and \(J_{\mathrm{MPP}}\) are the voltage and current density at the maximum power point. These equations are central to interpreting the performance improvement in the modified perovskite solar cell.
3. Chemical Passivation at the Perovskite Surface
I used X-ray photoelectron spectroscopy to probe chemical interactions between TiOPc-NPSC3 and the perovskite film. The Pb 4f spectrum provides direct information about the electronic environment of lead. In the control film, the Pb 4f5/2 and Pb 4f7/2 peaks appeared at 143.3 eV and 138.5 eV, respectively. After TiOPc-NPSC3 modification, these peaks shifted to 142.9 eV and 138.0 eV. The shift to lower binding energy indicates an increase in electron density around Pb2+. This is consistent with coordination from a Lewis base, such as sulfur, to an undercoordinated Pb2+ site. The sulfur lone pair donates electron density to lead, reducing the effective positive charge and stabilizing the surface.
More importantly, the control film showed clear Pb0 signals at 141.7 eV and 136.7 eV. Metallic lead is a well-known deep-level defect in perovskite solar cells. It forms through reduction of Pb2+, often assisted by residual iodide, organic cations, heat, light, or electron-rich surfaces. Pb0 clusters introduce midgap states, quench photoluminescence, increase non-radiative recombination, and accelerate degradation. In the TiOPc-NPSC3-modified film, the Pb0 signals disappeared completely. This strongly suggests that the additive suppresses the formation of metallic lead and passivates the defects that would otherwise promote Pb2+ reduction.
The S 2p spectrum provides complementary evidence. Pristine TiOPc-NPSC3 showed an S 2p signal at 163.2 eV. After incorporation into the perovskite film, the main S 2p peak shifted to 163.5 eV, and a new feature appeared at 159.8 eV. The main peak shift to higher binding energy indicates that sulfur loses some electron density upon interaction. The new peak at 159.8 eV is assigned to Pb-S coordination. Taken together, the Pb 4f and S 2p results establish a direct chemical bond between the additive and the perovskite surface.
| Element and Transition | Control Film | TiOPc-NPSC3-Modified Film | Interpretation |
|---|---|---|---|
| Pb 4f7/2 | 138.5 eV | 138.0 eV | Increased electron density around Pb2+ |
| Pb 4f5/2 | 143.3 eV | 142.9 eV | Coordination-induced shift |
| Pb0 4f7/2 | 136.7 eV | Not detected | Suppression of metallic lead |
| Pb0 4f5/2 | 141.7 eV | Not detected | Suppression of metallic lead |
| S 2p in pristine additive | 163.2 eV | Not applicable | Reference sulfur environment |
| S 2p after film incorporation | Not applicable | 163.5 eV and 159.8 eV | Pb-S coordination and sulfur electronic redistribution |
The passivation reaction can be summarized as:
$$ \mathrm{Pb^{2+}_{undercoordinated}} + \mathrm{S_{lone\ pair}} \rightarrow \mathrm{Pb-S\ coordination\ complex} $$
The suppression of metallic lead can be written as an inhibition of the reduction pathway:
$$ \mathrm{Pb^{2+}} + 2e^- \rightarrow \mathrm{Pb^{0}} \qquad \mathrm{inhibited\ by\ TiOPc-NPSC3} $$
I interpret these results as evidence that TiOPc-NPSC3 anchors to the perovskite surface through sulfur-lead coordination. This anchoring reduces undercoordinated Pb2+ sites, which are among the most detrimental defects in a perovskite solar cell. The additive also protects Pb2+ from reduction, thereby preventing the formation of Pb0 trap centers. Because these defects are concentrated at surfaces and grain boundaries, the antisolvent-assisted introduction is particularly effective. The additive is present during grain growth, so it can locate at the very regions where defects form.
4. Film Morphology, Crystallinity, and Hydrophobicity
I next examined how TiOPc-NPSC3 changes the microstructure of the perovskite film. Scanning electron microscopy showed that the control film consisted of relatively small grains with visible pinholes at grain boundaries. In contrast, the modified film exhibited larger grains and a more compact surface. The average grain size increased from 890 nm to 1166 nm. This increase is significant because larger grains reduce the total grain-boundary area. Since grain boundaries are preferential sites for defect formation, ion migration, and moisture ingress, grain enlargement directly benefits both efficiency and stability of the perovskite solar cell.
Cross-sectional scanning electron microscopy further revealed that the control film contained internal gaps and voids along grain boundaries. These voids can block vertical charge transport, increase series resistance, and create local electric-field inhomogeneities. The TiOPc-NPSC3-modified film displayed a dense, void-free vertical packing structure. This morphology is desirable because it provides continuous pathways for electron and hole transport toward the transport layers. For a perovskite solar cell, a compact absorber with low void density supports higher fill factor and lower leakage current.
I used atomic force microscopy to quantify surface roughness. The root-mean-square roughness decreased from 41.2 nm to 30.4 nm after adding TiOPc-NPSC3. The root-mean-square roughness is defined as:
$$ RMS = \sqrt{\frac{1}{N}\sum_{i=1}^{N}\left(z_i-\bar{z}\right)^2} $$
where \(z_i\) is the height at point \(i\), \(\bar{z}\) is the mean height, and \(N\) is the number of measured points. The lower roughness indicates a smoother interface between the perovskite and the hole transport layer. A smoother interface improves contact quality, reduces interfacial recombination, and helps form a more uniform electric field. This is especially important in a perovskite solar cell because interfacial defects often dominate non-radiative losses.
X-ray diffraction provided information about crystallinity and phase purity. The diffraction intensity ratio \(I_{(100)}/I_{\mathrm{PbI_2}}\) increased from 2.4 in the control film to 7.0 in the TiOPc-NPSC3-modified film. This ratio is a useful indicator of perovskite crystallinity relative to residual lead iodide. A higher ratio suggests that the additive promotes the formation of the desired perovskite phase and suppresses excess PbI2 or unreacted precursors. The modified film also showed no obvious shift in diffraction peak positions, indicating that TiOPc-NPSC3 does not enter the perovskite lattice to form a solid solution. Instead, it likely resides at surfaces and grain boundaries, where it can passivate defects without disturbing the bulk crystal structure.
| Property | Control Perovskite Film | TiOPc-NPSC3-Modified Film | Change and Significance |
|---|---|---|---|
| Average grain size | 890 nm | 1166 nm | Larger grains reduce grain-boundary density |
| Surface pinholes | Visible | Suppressed | Improved film coverage |
| Cross-sectional voids | Present | Largely eliminated | Better vertical charge transport |
| RMS roughness | 41.2 nm | 30.4 nm | Smoother interface |
| XRD ratio \(I_{(100)}/I_{\mathrm{PbI_2}}\) | 2.4 | 7.0 | Enhanced crystallinity and phase purity |
| Peak position shift | Reference | No obvious shift | Additive does not enter the lattice |
I also evaluated surface wettability through water contact angle measurements. The control film had a water contact angle of 73 degrees. After TiOPc-NPSC3 modification, the contact angle increased to 80 degrees. This increase indicates a more hydrophobic surface. Hydrophobicity is important for a perovskite solar cell because moisture is one of the primary causes of degradation. Water molecules can hydrate the perovskite, break metal-halide bonds, form hydrated phases, and eventually decompose the film. A hydrophobic surface reduces the probability that water will adsorb and penetrate into the absorber.
The contact angle relation can be expressed qualitatively as:
$$ \theta_{\mathrm{water}} \uparrow \quad \Rightarrow \quad \mathrm{hydrophobicity} \uparrow \quad \Rightarrow \quad \mathrm{moisture\ resistance} \uparrow $$
I exposed both films to a high-humidity environment at 23 degrees Celsius and 60 percent relative humidity. After 30 hours, the control film showed more rapid decomposition, whereas the modified film retained a visibly more intact surface. This result is consistent with the contact angle data and supports the idea that TiOPc-NPSC3 creates a protective hydrophobic environment around the perovskite. The effect is not merely physical shielding; the additive also improves the intrinsic quality of the film by reducing defect sites that serve as degradation initiation points.
5. Optical Properties and Recombination Dynamics
I measured ultraviolet-visible absorption spectra to determine whether TiOPc-NPSC3 changes the optical bandgap of the perovskite. The absorption onset remained essentially unchanged, indicating that the additive does not significantly alter the bandgap. The bandgap can be estimated from the absorption onset using:
$$ E_{\mathrm{g}} = \frac{1240}{\lambda_{\mathrm{onset}}} $$
where \(\lambda_{\mathrm{onset}}\) is the absorption onset wavelength in nanometers and \(E_{\mathrm{g}}\) is in electron volts. Because the onset did not shift appreciably, the improvement in device performance is not caused by a change in the intrinsic absorption edge. Instead, the modified film showed stronger absorption across the visible range. This increased absorption is attributed to improved film quality, larger grains, reduced void content, and better light scattering within a denser perovskite layer. In a perovskite solar cell, stronger absorption can directly increase photocurrent, provided that charge extraction remains efficient.
I used steady-state photoluminescence spectroscopy to examine non-radiative recombination. The modified film exhibited a photoluminescence intensity that was 3.8 times higher than that of the control film. The photoluminescence intensity ratio is:
$$ \frac{I_{\mathrm{PL,modified}}}{I_{\mathrm{PL,control}}} = 3.8 $$
In general, a higher photoluminescence intensity indicates a lower rate of non-radiative recombination. This relationship can be expressed qualitatively as:
$$ I_{\mathrm{PL}} \propto \frac{1}{R_{\mathrm{non-radiative}}} $$
where \(R_{\mathrm{non-radiative}}\) is the non-radiative recombination rate. A higher photoluminescence intensity therefore implies that defects have been passivated and that photo-generated carriers survive longer before recombining non-radiatively. This result supports the XPS conclusion that TiOPc-NPSC3 coordinates to undercoordinated Pb2+ and suppresses Pb0 formation. Reduced non-radiative recombination is directly beneficial for the open-circuit voltage of a perovskite solar cell because it lowers the voltage loss relative to the radiative limit.
| Optical or Recombination Metric | Control Film | TiOPc-NPSC3-Modified Film | Interpretation |
|---|---|---|---|
| Absorption onset | Reference | No significant shift | Bandgap remains nearly unchanged |
| Visible absorption strength | Reference | Enhanced | Improved film quality and light harvesting |
| Steady-state PL intensity | Reference | 3.8 times higher | Suppressed non-radiative recombination |
| Defect density | Higher | Lower | Passivation by sulfur-lead coordination |
| Expected VOC loss | Higher | Lower | Better quasi-Fermi-level splitting |
The photoluminescence enhancement can also be understood through a simple rate balance. In steady state, the generation rate \(G\) equals the sum of radiative and non-radiative recombination rates:
$$ G = R_{\mathrm{rad}} + R_{\mathrm{non-rad}} $$
The photoluminescence quantum yield is approximately:
$$ \Phi_{\mathrm{PL}} = \frac{R_{\mathrm{rad}}}{R_{\mathrm{rad}} + R_{\mathrm{non-rad}}} $$
When TiOPc-NPSC3 passivates defects, \(R_{\mathrm{non-rad}}\) decreases. If \(R_{\mathrm{rad}}\) remains comparable, \(\Phi_{\mathrm{PL}}\) and the measured photoluminescence intensity increase. This is exactly what I observed. The result indicates that the additive improves the electronic quality of the perovskite without introducing new non-radiative channels.
6. Photovoltaic Performance of the Perovskite Solar Cell
I fabricated control and TiOPc-NPSC3-modified perovskite solar cells and measured their current density-voltage characteristics under simulated air mass 1.5 global illumination at 100 mW/cm2. The control device achieved a power conversion efficiency of 23.94 percent, with an open-circuit voltage of 1.188 V, a short-circuit current density of 24.40 mA/cm2, and a fill factor of 82.59 percent. The TiOPc-NPSC3-modified device achieved a power conversion efficiency of 25.10 percent, with an open-circuit voltage of 1.203 V, a short-circuit current density of 24.84 mA/cm2, and a fill factor of 84.01 percent. All three photovoltaic parameters improved simultaneously. This simultaneous improvement is important because it indicates that the additive does not trade off one parameter for another. Instead, it improves the overall quality of the perovskite solar cell.
The relative efficiency improvement can be calculated as:
$$ \Delta \eta_{\mathrm{rel}} = \frac{\eta_{\mathrm{target}} – \eta_{\mathrm{control}}}{\eta_{\mathrm{control}}} \times 100\% $$
$$ \Delta \eta_{\mathrm{rel}} = \frac{25.10 – 23.94}{23.94} \times 100\% \approx 4.85\% $$
The open-circuit voltage gain is:
$$ \Delta V_{\mathrm{OC}} = 1.203 – 1.188 = 0.015\ \mathrm{V} $$
The short-circuit current density gain is:
$$ \Delta J_{\mathrm{SC}} = 24.84 – 24.40 = 0.44\ \mathrm{mA\,cm^{-2}} $$
The fill factor gain is:
$$ \Delta FF = 84.01 – 82.59 = 1.42\ \mathrm{percentage\ points} $$
These gains are consistent with reduced non-radiative recombination, improved charge extraction, and better film morphology. The open-circuit voltage increase is particularly meaningful because it reflects a reduction in the voltage deficit. In a perovskite solar cell, the open-circuit voltage is limited by recombination losses. Passivation of undercoordinated Pb2+ and suppression of Pb0 reduce deep-level traps, allowing the quasi-Fermi levels to split more effectively under illumination.
| Device Type | VOC (V) | JSC (mA/cm2) | FF (%) | PCE (%) |
|---|---|---|---|---|
| Control | 1.173 ± 0.015 (1.188) | 24.41 ± 0.34 (24.40) | 82.10 ± 2.6 (82.59) | 23.52 ± 0.55 (23.94) |
| TiOPc-NPSC3-modified | 1.195 ± 0.007 (1.203) | 24.90 ± 0.2 (24.84) | 83.17 ± 1.22 (84.01) | 24.75 ± 0.38 (25.10) |
The statistical distribution confirms that the improvement is reproducible. The modified devices showed a narrower distribution for open-circuit voltage and fill factor. A narrower distribution suggests better process reproducibility and more uniform film quality. The control devices had a wider spread, likely because of greater variations in defect density, pinhole formation, and grain-boundary structure. By contrast, TiOPc-NPSC3 helps homogenize the crystallization process, leading to a more reliable perovskite solar cell fabrication process.
I also measured the external quantum efficiency, also called the incident photon-to-current efficiency, to verify the short-circuit current density. The external quantum efficiency is defined as:
$$ EQE(\lambda) = \frac{\mathrm{number\ of\ collected\ electrons}}{\mathrm{number\ of\ incident\ photons\ at\ wavelength\ } \lambda} $$
The short-circuit current density can be calculated by integrating the external quantum efficiency spectrum against the solar spectrum:
$$ J_{\mathrm{SC}} = q \int_{\lambda_{\min}}^{\lambda_{\max}} \frac{\lambda}{hc} EQE(\lambda) \Phi_{\mathrm{AM1.5G}}(\lambda) d\lambda $$
where \(q\) is the elementary charge, \(h\) is Planck’s constant, \(c\) is the speed of light, and \(\Phi_{\mathrm{AM1.5G}}\) is the photon flux of the reference solar spectrum. The integrated current density from the control device was 23.88 mA/cm2, while that of the modified device was 24.44 mA/cm2. These values agree with the current density-voltage measurements to within less than 3 percent relative difference. This agreement confirms that the external quantum efficiency measurement is consistent with the current-voltage data and that the photocurrent improvement is real.
| Device Type | Integrated JSC from EQE (mA/cm2) | JSC from J-V (mA/cm2) | Relative Difference |
|---|---|---|---|
| Control | 23.88 | 24.40 | Approximately 2.1% |
| TiOPc-NPSC3-modified | 24.44 | 24.84 | Approximately 1.6% |
Dark current measurements provide additional insight into leakage and recombination. The TiOPc-NPSC3-modified device showed significantly lower dark current than the control device. In the dark, current flow is dominated by leakage and recombination rather than photogenerated carriers. A lower dark current indicates better charge selectivity, reduced shunt pathways, and suppressed recombination. This is consistent with the improved fill factor and open-circuit voltage. The dark current density-voltage behavior can be described by a diode-like relation:
$$ J_{\mathrm{dark}}(V) = J_0\left[\exp\left(\frac{qV}{nkT}\right)-1\right] $$
where \(J_0\) is the saturation current density, \(n\) is the ideality factor, \(k\) is Boltzmann’s constant, and \(T\) is temperature. A lower \(J_0\) generally corresponds to lower recombination and a higher open-circuit voltage under illumination. The reduced dark current in the modified perovskite solar cell therefore supports the conclusion that TiOPc-NPSC3 passivates defects and improves junction quality.
7. Long-Term Stability of the Perovskite Solar Cell
I evaluated the long-term storage stability of unencapsulated devices in low-humidity air. After 1056 hours, the control device retained approximately 71 percent of its initial efficiency, while the TiOPc-NPSC3-modified device retained approximately 88 percent of its initial efficiency. The efficiency retention is defined as:
$$ \mathrm{Retention}(t) = \frac{\eta(t)}{\eta(0)} \times 100\% $$
For the control device:
$$ \mathrm{Retention}_{\mathrm{control}}(1056\ \mathrm{h}) \approx 71\% $$
For the TiOPc-NPSC3-modified device:
$$ \mathrm{Retention}_{\mathrm{target}}(1056\ \mathrm{h}) \approx 88\% $$
This represents a substantial improvement in stability. The enhanced stability arises from two complementary effects. First, TiOPc-NPSC3 improves the crystallinity and morphology of the perovskite film, reducing grain-boundary defects and internal voids that serve as degradation initiation sites. Second, the hydrophobic phthalocyanine derivative increases the water contact angle and creates a moisture-resistant surface. Together, these effects slow the penetration of water and reduce the rate of chemical decomposition.
The stability improvement can be modeled empirically using a first-order decay approximation:
$$ \eta(t) = \eta(0)\exp(-kt) $$
where \(k\) is an apparent degradation rate constant. A lower \(k\) corresponds to slower degradation. Using the retention values at 1056 hours, I can estimate the relative degradation rates:
$$ k_{\mathrm{control}} \approx -\frac{\ln(0.71)}{1056} \approx 3.24 \times 10^{-4}\ \mathrm{h^{-1}} $$
$$ k_{\mathrm{target}} \approx -\frac{\ln(0.88)}{1056} \approx 1.21 \times 10^{-4}\ \mathrm{h^{-1}} $$
The ratio of degradation rates is:
$$ \frac{k_{\mathrm{control}}}{k_{\mathrm{target}}} \approx 2.68 $$
Thus, the modified perovskite solar cell degrades about 2.7 times more slowly than the control under the tested storage conditions. Although this first-order model is simplified, it provides a useful quantitative comparison. The slower degradation is consistent with the combined effects of defect passivation and hydrophobic protection.
| Storage Time | Control Retention (%) | TiOPc-NPSC3 Retention (%) | Interpretation |
|---|---|---|---|
| 0 h | 100 | 100 | Initial performance |
| 1056 h | 71 | 88 | Substantially improved environmental stability |
| Estimated degradation rate | \(3.24 \times 10^{-4}\ \mathrm{h^{-1}}\) | \(1.21 \times 10^{-4}\ \mathrm{h^{-1}}\) | Slower decay in modified device |
| Relative degradation ratio | 2.68 | 1.00 | Control degrades faster |
I attribute the stability enhancement to the following sequence of events. During film formation, TiOPc-NPSC3 coordinates to Pb2+ at grain boundaries and surfaces. This reduces undercoordinated lead sites and inhibits Pb0 formation. As a result, the film contains fewer deep-level traps and fewer initiation sites for degradation. The hydrophobic alkyl sulfide chains orient toward the surface, increasing the water contact angle. When water molecules approach the film, they encounter a more hydrophobic barrier. Even if some moisture enters, the reduced defect density limits the propagation of degradation. This dual mechanism is critical for a durable perovskite solar cell.
8. Mechanistic Picture
The experimental results support a mechanistic model in which TiOPc-NPSC3 performs four functions. First, it coordinates to undercoordinated Pb2+ through sulfur. This passivates surface and grain-boundary defects. Second, it suppresses the formation of metallic lead, removing a deep-level recombination center. Third, it regulates crystallization, leading to larger grains, fewer pinholes, lower roughness, and better phase purity. Fourth, it increases surface hydrophobicity, protecting the perovskite from moisture. These functions combine to improve both efficiency and stability of the perovskite solar cell.
The passivation equilibrium can be represented as:
$$ \mathrm{Pb^{2+}_{surface}} + \mathrm{TiOPc-NPSC3} \rightleftharpoons \mathrm{Pb^{2+}-TiOPc-NPSC3} $$
The equilibrium favors the coordinated complex because sulfur is a strong Lewis base toward lead. The formation of this complex reduces the concentration of free undercoordinated Pb2+ sites. The defect density \(N_t\) therefore decreases:
$$ N_t^{\mathrm{modified}} < N_t^{\mathrm{control}} $$
Non-radiative recombination through deep traps can be described by Shockley-Read-Hall statistics. A simplified recombination rate is:
$$ R_{\mathrm{SRH}} = \frac{np – n_i^2}{\tau_p(n+n_1) + \tau_n(p+p_1)} $$
where \(n\) and \(p\) are electron and hole densities, \(n_i\) is the intrinsic carrier density, \(\tau_n\) and \(\tau_p\) are carrier lifetimes, and \(n_1\) and \(p_1\) are trap-related densities. When \(N_t\) decreases, \(\tau_n\) and \(\tau_p\) increase, and \(R_{\mathrm{SRH}}\) decreases. This reduction in non-radiative recombination increases the photoluminescence intensity and improves the open-circuit voltage. The observed 3.8-fold photoluminescence enhancement is consistent with this picture.
The crystallization effect can be summarized as:
$$ \mathrm{TiOPc-NPSC3} \rightarrow \mathrm{slower\ nucleation} + \mathrm{controlled\ growth} \rightarrow \mathrm{larger\ grains} $$
The hydrophobic effect can be summarized as:
$$ \mathrm{TiOPc-NPSC3\ surface\ layer} + \mathrm{H_2O} \rightarrow \mathrm{reduced\ water\ adsorption} $$
These mechanisms are not independent. Larger grains reduce the total grain-boundary area, which reduces the number of defect sites. Fewer defect sites reduce non-radiative recombination and degradation. A smoother surface improves contact with the hole transport layer, which improves charge extraction. Improved charge extraction increases fill factor and short-circuit current density. A hydrophobic surface reduces moisture ingress, which improves long-term stability. The result is a self-reinforcing improvement across multiple performance metrics.
| Mechanistic Function | Molecular or Film Origin | Observed Evidence | Device Consequence |
|---|---|---|---|
| Defect passivation | Sulfur coordination to Pb2+ | Pb 4f shift, new S 2p feature | Lower non-radiative recombination |
| Metallic lead suppression | Stabilization of Pb2+ | Disappearance of Pb0 XPS signals | Fewer deep traps |
| Crystallization control | Additive during antisolvent step | Larger grains, lower roughness, higher XRD ratio | Better charge transport |
| Hydrophobic protection | Alkyl sulfide chains | Water contact angle increase | Improved moisture stability |
| Interfacial improvement | Smoother and denser film | Lower dark current | Higher fill factor and VOC |
9. Design Rules for Additives in Perovskite Solar Cells
The results allow me to formulate several design rules for molecular additives in perovskite solar cells. First, the additive should contain a Lewis basic donor group that can coordinate to undercoordinated Pb2+. Sulfur, nitrogen, oxygen, and halide donors can all serve this purpose, but sulfur is particularly attractive because of its polarizability and softness. Second, the additive should have a conjugated or conductive core so that it does not excessively block charge transport. Third, the additive should be soluble in the antisolvent or co-solvent used during film formation. Fourth, the additive should have hydrophobic groups to improve moisture resistance. Fifth, the additive should not enter the perovskite lattice in a way that disrupts the band structure or phase stability.
These rules can be expressed as a multi-objective optimization:
$$ \max_{\mathrm{additive}} \left[ \eta,\ \mathrm{stability} \right] = \max \left[ f_{\mathrm{passivation}},\ f_{\mathrm{transport}},\ f_{\mathrm{hydrophobicity}},\ f_{\mathrm{processability}} \right] $$
In the case of TiOPc-NPSC3, the sulfur atoms provide passivation, the phthalocyanine core supports charge delocalization, the alkyl chains provide hydrophobicity, and the molecular structure is compatible with anisole. The additive is therefore well matched to the antisolvent-assisted fabrication of a perovskite solar cell.
I also note that the additive concentration and saturation level must be controlled. If the concentration is too low, passivation may be incomplete. If the concentration is too high, the additive may form insulating domains or disrupt crystallization. In my work, using anisole saturated with TiOPc-NPSC3 provided a favorable balance. The saturated solution ensures that enough additive is present at the crystallization front, while the antisolvent volume and spin-coating conditions prevent macroscopic aggregation. This balance is essential for reproducible device performance.
| Design Rule | Molecular Requirement | Benefit in Perovskite Solar Cell |
|---|---|---|
| Strong Pb2+ coordination | Lewis basic sulfur or nitrogen donor | Passivates undercoordinated lead |
| Conductive core | Conjugated pi-system | Maintains charge transport |
| Antisolvent solubility | Appropriate alkyl or aryl substituents | Uniform additive distribution |
| Hydrophobic surface | Alkyl or fluorinated chains | Reduces moisture ingress |
| Lattice compatibility | Size and charge matching | Avoids phase disruption |
| Process window | Controlled concentration | Reproducible efficiency and stability |
10. Quantitative Summary and Equations
For clarity, I summarize the main quantitative results and the equations used to interpret them. The power conversion efficiency is:
$$ \eta = \frac{V_{\mathrm{OC}} J_{\mathrm{SC}} FF}{P_{\mathrm{in}}} $$
For the control device:
$$ \eta_{\mathrm{control}} = \frac{1.188 \times 24.40 \times 0.8259}{100} \approx 23.94\% $$
For the modified device:
$$ \eta_{\mathrm{target}} = \frac{1.203 \times 24.84 \times 0.8401}{100} \approx 25.10\% $$
The fill factor is:
$$ FF = \frac{V_{\mathrm{MPP}} J_{\mathrm{MPP}}}{V_{\mathrm{OC}} J_{\mathrm{SC}}} $$
The external quantum efficiency integrated current is:
$$ J_{\mathrm{SC}} = q \int_{\lambda_{\min}}^{\lambda_{\max}} \frac{\lambda}{hc} EQE(\lambda) \Phi_{\mathrm{AM1.5G}}(\lambda) d\lambda $$
The optical bandgap is:
$$ E_{\mathrm{g}} = \frac{1240}{\lambda_{\mathrm{onset}}} $$
The root-mean-square roughness is:
$$ RMS = \sqrt{\frac{1}{N}\sum_{i=1}^{N}\left(z_i-\bar{z}\right)^2} $$
The efficiency retention is:
$$ \mathrm{Retention}(t) = \frac{\eta(t)}{\eta(0)} \times 100\% $$
The relative efficiency improvement is:
$$ \Delta \eta_{\mathrm{rel}} = \frac{\eta_{\mathrm{target}} – \eta_{\mathrm{control}}}{\eta_{\mathrm{control}}} \times 100\% $$
The dark current diode relation is:
$$ J_{\mathrm{dark}}(V) = J_0\left[\exp\left(\frac{qV}{nkT}\right)-1\right] $$
The photoluminescence enhancement ratio is:
$$ \frac{I_{\mathrm{PL,modified}}}{I_{\mathrm{PL,control}}} = 3.8 $$
The passivation reaction is:
$$ \mathrm{Pb^{2+}_{undercoordinated}} + \mathrm{S_{lone\ pair}} \rightarrow \mathrm{Pb-S\ coordination\ complex} $$
The metallic lead suppression is:
$$ \mathrm{Pb^{2+}} + 2e^- \rightarrow \mathrm{Pb^{0}} \qquad \mathrm{inhibited} $$
| Metric | Control | TiOPc-NPSC3-Modified | Equation or Definition |
|---|---|---|---|
| VOC | 1.188 V | 1.203 V | Measured from J-V curve |
| JSC | 24.40 mA/cm2 | 24.84 mA/cm2 | Measured from J-V curve |
| FF | 82.59% | 84.01% | \(FF = V_{\mathrm{MPP}}J_{\mathrm{MPP}}/(V_{\mathrm{OC}}J_{\mathrm{SC}})\) |
| PCE | 23.94% | 25.10% | \(\eta = V_{\mathrm{OC}}J_{\mathrm{SC}}FF/P_{\mathrm{in}}\) |
| Integrated JSC from EQE | 23.88 mA/cm2 | 24.44 mA/cm2 | \(J_{\mathrm{SC}} = q\int \lambda EQE \Phi /hc\,d\lambda\) |
| Grain size | 890 nm | 1166 nm | Microscopy analysis |
| RMS roughness | 41.2 nm | 30.4 nm | \(RMS = \sqrt{\sum(z_i-\bar{z})^2/N}\) |
| XRD ratio | 2.4 | 7.0 | \(I_{(100)}/I_{\mathrm{PbI_2}}\) |
| PL intensity ratio | 1.0 | 3.8 | Relative steady-state PL |
| Water contact angle | 73 degrees | 80 degrees | Surface wettability |
| Retention after 1056 h | 71% | 88% | \(\eta(t)/\eta(0)\) |
11. Interpretation of Efficiency and Stability Gains
The efficiency gain from 23.94 percent to 25.10 percent corresponds to a relative improvement of about 4.85 percent. This gain is distributed across all photovoltaic parameters. The open-circuit voltage increase of 15 mV reflects reduced non-radiative recombination. The short-circuit current density increase of 0.44 mA/cm2 reflects improved light absorption and charge collection. The fill factor increase of 1.42 percentage points reflects lower series resistance, lower leakage current, and better interfacial contact. These improvements are consistent with the morphological and photophysical data.
The stability gain is equally important. Unencapsulated control devices retained about 71 percent of their initial efficiency after 1056 hours, whereas modified devices retained about 88 percent. The difference corresponds to a much slower degradation rate. I attribute this to two factors: reduced defect density and increased hydrophobicity. The defect reduction lowers the number of sites where degradation can begin. The hydrophobic surface reduces the amount of water that can reach the perovskite. Because moisture-driven degradation is often accelerated at grain boundaries and surfaces, the additive’s preferential location at these regions is advantageous.
The combined effect can be expressed as a product of improvement factors:
$$ \eta_{\mathrm{target}} = \eta_{\mathrm{control}} \times f_{\mathrm{passivation}} \times f_{\mathrm{morphology}} \times f_{\mathrm{transport}} \times f_{\mathrm{stability}} $$
This factorization is conceptual rather than exact, but it captures the multi-functional nature of the additive. In a perovskite solar cell, single-function additives often improve one property at the expense of another. TiOPc-NPSC3 improves several properties at once because its molecular structure contains distinct functional units: sulfur donors for passivation, a conjugated phthalocyanine core for charge delocalization, and alkyl chains for hydrophobicity and solubility.
12. Practical Implications
The antisolvent-assisted additive strategy is attractive for practical perovskite solar cell manufacturing because it does not require changes to the perovskite precursor composition. It can be integrated into existing spin-coating or blade-coating processes by simply adding the functional molecule to the antisolvent. This makes the approach compatible with high-throughput fabrication. The additive concentration can be controlled by saturation, which provides a natural limit and reduces the risk of overdosing. In my experiments, saturated anisole provided a robust process window and reproducible device performance.
For future development, I would consider several directions. First, the alkyl chain length and sulfur position could be varied to tune solubility, steric hindrance, and passivation strength. Second, different central metals in the phthalocyanine core could be tested to adjust the electronic structure and charge transport properties. Third, the additive could be combined with other passivation molecules to target different defect types. Fourth, advanced transient techniques such as time-resolved photoluminescence, transient photovoltage, and transient photocurrent could provide deeper insight into carrier lifetimes and charge extraction dynamics. Fifth, accelerated aging tests under damp heat, illumination, and thermal cycling would further evaluate the practical stability of the perovskite solar cell.
| Future Direction | Variable to Tune | Expected Benefit |
|---|---|---|
| Alkyl chain engineering | Chain length and branching | Optimized solubility and hydrophobicity |
| Central metal substitution | Ti, Zn, Cu, Ni, or others | Tuned energy levels and transport |
| Mixed passivation | Sulfur plus nitrogen or halide donors | Broader defect coverage |
| Transient spectroscopy | Time-resolved PL and photocurrent | Direct carrier lifetime and extraction analysis |
| Accelerated aging | Damp heat, light soaking, thermal cycling | More realistic stability assessment |
13. Conclusion
I synthesized and applied a thioalkyl-chain-modified titanyl phthalocyanine derivative, TiOPc-NPSC3, as an antisolvent additive for formal n-i-p perovskite solar cells. X-ray photoelectron spectroscopy confirmed that sulfur atoms coordinate to undercoordinated Pb2+, shifting the Pb 4f peaks to lower binding energy and creating a new S 2p feature associated with Pb-S bonding. The additive completely suppressed detectable metallic lead signals, indicating that it inhibits Pb0 formation and removes a major deep-level defect. Scanning electron microscopy, atomic force microscopy, and X-ray diffraction showed that TiOPc-NPSC3 promotes larger grains, reduces pinholes and internal voids, lowers surface roughness from 41.2 nm to 30.4 nm, and increases the \(I_{(100)}/I_{\mathrm{PbI_2}}\) ratio from 2.4 to 7.0. Water contact angle measurements showed that the modified film is more hydrophobic, with the contact angle increasing from 73 degrees to 80 degrees. Steady-state photoluminescence increased by a factor of 3.8, confirming suppressed non-radiative recombination.
The optimized perovskite solar cell achieved a power conversion efficiency of 25.10 percent, with an open-circuit voltage of 1.203 V, a short-circuit current density of 24.84 mA/cm2, and a fill factor of 84.01 percent. The control device achieved 23.94 percent, with an open-circuit voltage of 1.188 V, a short-circuit current density of 24.40 mA/cm2, and a fill factor of 82.59 percent. The external quantum efficiency integrated current densities were consistent with the current density-voltage measurements. Dark current was significantly reduced in the modified device, indicating suppressed leakage and recombination. Unencapsulated modified devices retained approximately 88 percent of their initial efficiency after 1056 hours of storage in low-humidity air, compared with approximately 71 percent for the control. The improved stability results from both defect passivation and hydrophobic protection.
Overall, my results demonstrate that TiOPc-NPSC3 is a multifunctional additive that simultaneously improves film quality, passivates defects, suppresses metallic lead, enhances charge extraction, and increases moisture resistance. The work provides a clear design strategy for molecular additives in perovskite solar cells: combine strong lead-coordinating groups, a conductive conjugated core, hydrophobic substituents, and antisolvent compatibility. This strategy can guide the development of next-generation additives for efficient and stable perovskite solar cell technologies.
| Summary Item | Control Perovskite Solar Cell | TiOPc-NPSC3 Perovskite Solar Cell |
|---|---|---|
| Best PCE | 23.94% | 25.10% |
| VOC | 1.188 V | 1.203 V |
| JSC | 24.40 mA/cm2 | 24.84 mA/cm2 |
| FF | 82.59% | 84.01% |
| Grain size | 890 nm | 1166 nm |
| RMS roughness | 41.2 nm | 30.4 nm |
| PL intensity ratio | 1.0 | 3.8 |
| Water contact angle | 73 degrees | 80 degrees |
| Retention after 1056 h | 71% | 88% |
The results also highlight the importance of characterizing both chemical and physical properties when developing additives for a perovskite solar cell. A molecule that only passivates defects may not improve morphology. A molecule that only improves morphology may not suppress deep traps. A molecule that only increases hydrophobicity may not improve efficiency. TiOPc-NPSC3 succeeds because it addresses all these factors. The sulfur-lead coordination provides chemical passivation, the conjugated core supports charge transport, the alkyl chains provide hydrophobicity, and the antisolvent processing route ensures that the additive is present during crystallization. This integrated design approach is likely to be valuable for future perovskite solar cell research.
In summary, I have shown that TiOPc-NPSC3 can serve as a multifunctional antisolvent additive for perovskite solar cells. The additive improves crystallinity, passivates undercoordinated lead defects, suppresses metallic lead, reduces non-radiative recombination, enhances charge extraction, and increases moisture resistance. The resulting perovskite solar cell achieves 25.10 percent efficiency and retains 88 percent of its initial performance after 1056 hours. These findings support the continued development of phthalocyanine-based additives and related multifunctional molecules for efficient and stable perovskite solar cell applications.