Synthesis, crystal structure, optical properties, thin-film deposition, and SCAPS-1D simulation of a new lead-free 2D hybrid perovskite (C₆H₇NBr)₂[CuCl₄] for potential photovoltaic applications

Abstract

A novel lead-free two-dimensional (2D) hybrid perovskite, (C₆H₇NBr)₂[CuCl₄], was synthesized via slow evaporation at room temperature and successfully deposited as relatively homogeneous thin films using a simple one-step spin-coating method. It was characterized by single-crystal X-ray diffraction, spectroscopic techniques, and thermal analysis. The compound crystallizes in the orthorhombic non-centrosymmetric space group Pca2₁, forming layered CuCl₆ octahedra separated by 4-bromoanilinium cations. It exhibits a direct optical bandgap of 2.30 eV. Hirshfeld surface analysis further provided detailed insight into crystal packing and intermolecular interactions. The photovoltaic potential of this new perovskite as a light absorber layer was theoretically evaluated using SCAPS-1D simulations in the device configuration FTO/TiO₂/(C₆H₇NBr)₂[CuCl₄]/Spiro-OMeTAD/Au. The optimized device parameters yielded a fill factor (FF) of 71.50%, an open-circuit voltage (V_oc) of 1.2 V, and a power conversion efficiency (PCE) approaching 10%. These findings demonstrate the strong potential of this environmentally friendly copper-based perovskite for promising lead-free photovoltaic applications, contributing to the advancement of sustainable and scalable solar energy technologies.

A lead-free 2D copper hybrid perovskite (C₆H₇NBr)₂[CuCl₄] with a 2.30 eV bandgap was synthesized and deposited as thin films. SCAPS-1D simulations predict PCE ∼9.43% in an FTO/TiO₂/perovskite/Spiro-OMeTAD/Au device.

Introduction

Hybrid organic–inorganic perovskites (HOIPs) have emerged as a highly versatile class of functional materials, owing to their exceptional structural flexibility and outstanding optoelectronic properties.^1–4 These materials synergistically combine the mechanical softness and dielectric tunability of organic cations with the electronic functionality of inorganic frameworks, enabling efficient exciton confinement, strong light–matter interactions, and highly tunable bandgaps.^5–7

As a result, HOIPs are promising candidates for next-generation optoelectronic devices, such as solar cells (PSC), light-emitting diodes (LEDs), and photodetectors.^8–10

A key subclass is the Ruddlesden-Popper (RP) family of layered (2D) perovskites, with the general formula (R–NH₃)₂B_nX_3n+1, where R–NH₃⁺ is a monovalent organic cation, B is a divalent metal cation, and X is a halide anion.^11,12 In these structures, corner-sharing BX₆ octahedra form inorganic slabs of thickness n, separated by organic bilayers. This architecture allows precise tuning of excitonic binding energies, band structure, and environmental stability-often superior to that of conventional 3D perovskites.^11,12 One of the key reasons for the impressive performance of copper-based 2D halide perovskites is the earth abundance, low toxicity, and rich coordination chemistry of Cu²⁺.^13,14 Moreover, the 3d⁹ electronic configuration of Cu²⁺ induces pronounced Jahn–Teller distortions, leading to anisotropic coordination geometries that strongly influence electronic coupling, optical d–d transitions, and ligand-to-metal charge-transfer (LMCT) processes within the inorganic layers. When combined with carefully selected organic spacers, these distortions give rise to distinctive structural motifs and enhanced optoelectronic properties.^15–18

The organic cation plays a decisive role in directing both the structural organization and functional properties of hybrid halide materials. Aromatic spacers such as 4-bromoanilinium promote dense π–π stacking and halogen-assisted N–H⋯X hydrogen bonding, thereby reinforcing interlayer cohesion at the organic–inorganic interface. In addition, the presence of the heavy bromine substituent modulates exciton dynamics, bandgap energy, photoluminescence quantum yield, and charge-transport behavior, making 4-bromoanilinium a highly effective spacer for fine-tuning optoelectronic performance.^19,20

The importance of such tailored organic–inorganic interactions is further illustrated in related hybrid systems. For example, the one-dimensional cadmium-based compound (C₆H₇NBr)₃[CdBr₅] exhibits an indirect bandgap of 3.93 eV, with valence-band contributions from Br-4p and Cd-4d orbitals and a conduction band dominated by Cd-5s states. Upon UV excitation, it displays broadband white-light emission with a high color rendering index (CRI) of 98 and CIE coordinates of (0.30, 0.32), originating from self-trapped excitons in distorted CdBr₆ octahedra combined with intrinsic fluorescence of the 4-bromoanilinium cation.²⁰ Collectively, these observations underscore how halogenated aromatic spacers, through cooperative structural distortions and interfacial interactions, can unlock advanced functionalities for solid-state lighting and broader optoelectronic applications.

Since the early 2000s, hybrid perovskites have undergone remarkable development for solar energy conversion, particularly in diammonium-templated architectures that enhance structural stability and optoelectronic performance. Notably, 2D/3D hybrid perovskites incorporating octane-1,8-diaminium (ODA) as a spacer achieved an initial power conversion efficiency (PCE) of 18.50% (V_oc = 1.16 V), which increased to a champion value of 22.19% (V_oc = 1.26 V) after surface optimization establishing a benchmark for wide-bandgap perovskite solar cells.⁸ Similarly, Pb–Sn mixed perovskites employing propanediamine (PDA) cations reached a PCE of 20.03% with a narrow bandgap of 1.26 eV and strong absorption, integrated between efficient electron- and hole-transport layers.²¹

In response to concerns over toxicity and long-term stability, copper-based systems have emerged as promising lead-free alternatives.^22–24 For example, studies on (C₆H₁₀N₂)[CuCl₄] reported a bandgap of ∼2.11 eV and a simulated PCE of 10.22% (FF = 81.34%, J_sc = 7.21 mA cm⁻², V_oc = 1.73 V) in an ITO/PTAA/(C₆H₁₀N₂)[CuCl₄]/C₆₀/Au configuration under AM1.5 G illumination at 300 K using drift-diffusion modeling in TiberCAD.²⁵ Further advancing this strategy, three Cu-based hybrid perovskites with the formula (CH₃NH₃)₂CuX (X = Cl₄, Cl₂I₂, Cl₂Br₂) were synthesized to replace toxic Pb with earth-abundant Cu. These materials exhibited tunable absorption across 300–900 nm, suitable bandgaps, and green photoluminescence associated with Cu²⁺ species. Devices fabricated in a mesoporous architecture (glass/FTO/TiO₂/perovskite/Spiro-OMeTAD/Au) achieved experimental PCEs up to 2.41% for (CH₃NH₃)₂CuCl₄, while theoretical simulations predicted efficiencies approaching ∼21.76%, underscoring the potential of copper as a sustainable and environmentally benign substitute for lead in next-generation perovskite solar cells.²⁶

Building on these advances, the present study introduces the novel copper-based 2D hybrid perovskite (C₆H₇NBr)₂[CuCl₄], synthesized via slow evaporation and spin-coating methods and characterized comprehensively through experimental and theoretical approaches. This work elucidates the structure–property relationships in copper-based perovskite, underscoring its potential as a sustainable, lead-free 2D perovskite platform for advanced optoelectronic and photovoltaic applications.

Experimental section

Formation of the single crystals

All chemical reagents were purchased from Sigma-Aldrich and used without further purification. The bromoanilinium copper perovskite, was synthesized using a slow evaporation method. The stoichiometric ratio of the precursors was fixed at a 2 : 1 molar ratio of 4-bromoaniline to CuCl₂, consistent with the compound's chemical formula. In a typical synthesis, CuCl₂ (1 mmol, 0.17 g) and 4-bromoaniline (3 mmol, 0.55 g) were dissolved in a mixed solvent of ethanol and concentrated hydrochloric acid (10 : 1 v/v) under continuous magnetic stirring at 60 °C for 20 minutes, producing a homogeneous green solution. The presence of HCl promotes complete protonation of the amine group, generating the bromoanilinium cation and stabilizing the [CuCl₄]²⁻ complex anion. The resulting clear solution was left to slowly evaporate at room temperature for approximately two weeks, yielding a new yellow platelet crystal of (C₆H₇NBr)₂[CuCl₄]. These crystals were suitable for single-crystal X-ray diffraction analysis.

X-ray crystallographic study

A yellow plate-shaped single crystal of the title compound formulated as (C₆H₇NBr)₂[CuCl₄], with approximate dimensions of 0.18 × 0.13 × 0.06 mm³, was selected and mounted for single-crystal X-ray diffraction analysis. Intensity data were collected at 100 K on a Rigaku XtaLAB Synergy diffractometer equipped with a HyPix detector, using Cu Kα radiation (λ = 1.54184 Å). A total of 181 440 reflections were measured over a θ range of 2.6–66.5°, yielding 21 589 independent reflections (R_int = 0.078) with 100% completeness to θ = 66.5°. Absorption corrections were applied numerically based on Gaussian integration over a multifaceted crystal model and supplemented by empirical scaling using spherical harmonics (T_max/T_min = 0.866/0.326; µ = 12.23 mm⁻¹). The structure was solved in the orthorhombic chiral space group Pca2₁ using intrinsic phasing (dual-space) as implemented in SHELXT-2018/2 (Sheldrick, 2015) and refined against F² by full-matrix least-squares methods using SHELXL-2018/3 (Sheldrick, 2015) within the Olex2-1.5 graphical interface. All non-hydrogen atoms were refined anisotropically, while hydrogen atoms bonded to carbon and nitrogen were placed in geometrically calculated positions (C–H = 0.97 Å; N–H = 0.89 Å) and refined using a riding model. The refinement, based on 21 589 reflections, 1153 parameters, and 951 restraints, converged to R₁ = 0.057 and wR₂ = 0.132 for 20 310 reflections with I > 2σ(I), and R₁ = 0.065 and wR₂ = 0.140 for all data, with a goodness-of-fit S = 1.08. The absolute structure was refined as a two-component inversion twin, yielding a Flack parameter of 0.49(4), indicating inversion twinning with approximately equal domain fractions. As a result, the absolute structure cannot be unambiguously determined, although the non-centrosymmetric space group Pca2₁ is correct. The crystallographic data have been deposited with the Cambridge Crystallographic Data Centre (CCDC) under deposition number 2499237. Full details of data collection and structure refinement are provided in the associated CIF file.

Spectroscopic measurements

The IR spectrum of the title compound was recorded on a PerkinElmer spectrophotometer in the range of 400–4000 cm⁻¹. The UV-vis diffuse reflectance spectrum was measured at room temperature using a Varian Cary 5000 UV-vis-NIR spectrophotometer over the wavelength range of 200–1000 nm.

Thermal analysis

The thermal properties of the sample, were characterized using a PerkinElmer Pyris Differential Scanning Calorimeter (DSC). A precise sample mass of 4.9 mg was subjected to analysis under a nitrogen purge gas flowing at 20.0 mL min⁻¹. To study its phase transitions, the sample underwent a controlled thermal cycle. It was first heated from −50.0 °C to 190.0 °C at a constant rate of 5.0 °C min⁻¹. Subsequently, it was cooled from 190.0 °C back to −50.0 °C at the same rate.

Results and discussion

Decoding the crystal structure: from layers to interactions

Single-crystal X-ray diffraction (SC-XRD) analysis reveals that 4-bromoanilinium tetrachlorocuprate crystallizes in the orthorhombic system, adopting the non-centrosymmetric space group Pca2₁, with unit cell parameters a = 34.4338(2) Å, b = 49.4356(3) Å, c = 7.4578(1) Å, and a unit cell volume of 12 695.08(11) ų. The structure contains four formula units per unit cell (Z = 4), and the final refinement converged to R₁ = 0.0572 and wR₂ = 0.1308. These parameters were determined through X-ray single crystal diffraction analysis, meticulously documented in CIF file (SI) which encompasses comprehensive crystallographic data, data collection details, and refinement information. The formula unit comprises two crystallographically independent 4-bromoanilinium cations (C₆H₇NBr)⁺ and one discrete [CuCl₄]²⁻ anion. To maintain charge neutrality, two organic cations associate with each divalent anion, yielding the neutral stoichiometric formula (C₆H₇NBr)₂[CuCl₄], as illustrated in Fig. 1a. Notably, all atoms in the structure occupy the Wyckoff 4a general positions, allowing the space group symmetry operations to generate the full crystal lattice while accommodating the asymmetric geometries of the organic cations and inorganic anions. Importantly, no inversion center is present within the orthorhombic framework, reflecting the non-centrosymmetric nature of the Pca2₁ space group. This symmetry enables a polar arrangement of both the inorganic and organic components.

Fig. 1. (a) Formula unit of 4-bromoanilinium tetrachlorocuprate perovskite and (b) crystal packing framework of the compound viewed along the ab and bc planes.

To unravel the crystal architecture of 4-bromoanilinium tetrachlorocuprate, it is important to note that its structure, featuring alternating zigzag organic and inorganic layers, is derived from the Ruddlesden-Popper (RP) A₂BX₄ parent perovskite framework^11,12 (Fig. 1b). In the 4-bromoanilinium cation (C₆H₇NBr)⁺, the bromo substituent occupies the para position relative to the protonated –NH₃⁺ group on the aromatic ring, placing the Br atom directly opposite the amino group group with a C–C–C–Br torsion angle of almost 176° indicates that the bromine atom is almost coplanar with the benzene ring. The value being close to 180° means the C–Br bond lies nearly in the plane of the aromatic system, showing only a very small deviation (4°) from perfect planarity. This arrangement allows the cation to adopt a planar geometry that facilitates efficient packing between the inorganic layers. In unison, these interactions reinforce the structural integrity of the new layered A₂BX₄ framework and promote a polar organization of the hybrid perovskite.

Complementing this organic scaffold, the inorganic sublattice consists of isolated [CuCl₄]²⁻ units, where each copper(ii) ion is coordinated by four chloride ligands in an almost square-planar geometry, oriented nearly parallel to the a-axis (Fig. 1b). The Cu–Cl bond lengths range from 2.289 to 2.294 Å, and the Cl–Cu–Cl bond angles are close to 90°, with values of 89.704°, 89.755°, 90.146°, and 90.395°, consistent with a nearly regular planar coordination environment. Minor deviations from ideal values arise from crystal packing effects and electrostatic interactions with the surrounding organic cations. These structural features, combined with the organic–inorganic interactions, collectively define the stability, polarity, and layered nature of the Ruddlesden-Popper (RP) hybrid perovskite.

Beyond the apparent square-planar motif, each copper center adopts a (4 + 2) distorted octahedral [CuCl₆]⁴⁻ geometry (Fig. 2). The four equatorial Cu–Cl bonds are short and nearly equivalent (2.287–2.296 Å), whereas the two axial bonds are markedly elongated (2.873 and 2.904 Å), reflecting the Jahn–Teller distortion characteristic of d⁹ Cu²⁺ ions. The Cl–Cu–Cl bond angles deviate from ideal octahedral values, ranging from 89.704° to 179.232°, consistent with literature reports for related tetrachlorocuprate(II) anions.^27–29

Fig. 2. Crystal packing of 4-bromoanilinium tetrachlorocuprate (II), illustrating the layered structure formed by corrugated inorganic CuCl₆ sheets and organic cations (H atoms omitted for clarity).

This axial elongation arises from the Jahn–Teller effect,^15–18 further modulated by crystal packing and electrostatic interactions between the anionic [CuCl₆]⁴⁻ layers and the protonated 4-bromoanilinium cations. The resulting coordination environment couple's electronic asymmetry with supramolecular stabilization, giving rise to the corrugated two-dimensional inorganic framework.^27–29 To quantitatively assess the degree of geometric distortion within the inorganic framework, a bond-length distortion parameter (Δ) was introduced. This parameter measures the deviation of individual Cu–Cl bond lengths within the [CuCl₆]²⁻ octahedron from their average value, thereby providing a numerical description of the Jahn–Teller distortion characteristic of Cu²⁺ systems. The distortion parameter Δ is calculated according to the relation:^27–29

1

where N = 6 represents the total number of bonds, di denotes each individual Cu–Cl bond distance, and d_m is the average Cu–Cl bond distance, the calculated bond distortion parameter Δ for 4-bromoanilinium tetrachlorocuprate is 0.0136. This significant, non-zero value quantitatively confirms that the CuCl₆ octahedron is highly distorted. This elongation reflects a pronounced Jahn–Teller effect, characteristic of copper(ii) complexes.^27–29

In this structure, the inorganic anion contains a single crystallographic site for the copper atom (Wyckoff site 4a), coordinated by four chlorine atoms in a square-planar geometry. These units are further connected through octahedral linkages, forming a periodic network that builds the A₂BX₄ framework. However, the expansion of the coordination environment from four-to six-fold is clearly illustrated from Fig. 1b and 2, where elongated CuCl₆ octahedra connect at their corners, forming a structure reminiscent of stacked layers. Within these mineral-like layers, each copper ion is surrounded by six chlorine atoms: four equatorial chlorides forming the square plane and two axial chlorides contributed by adjacent layers. The equatorial chlorines lie in close proximity to the copper centers, establishing strong in-plane interactions, while the axial chlorines are positioned above and below the CuCl₄ planes of each individual formula unit, and are not shared with neighboring copper ions.

The octahedral sheets exhibit two distinct types of Cu–Cl bonds: four equatorial (lateral) bonds, measuring 2.289–2.294 Å (Cu1–Cl1, Cl2, Cl3, Cl4), with Cl1 and Cl2 acting as µ₂-bridging chlorides, and two elongated axial bonds of 2.900 and 2.920 Å connecting to the terminal chlorides (Cu-axial Cl1 and Cu-axial Cl2), oriented nearly perpendicular to the plane (Fig. 3). The inorganic layer shows that each [CuCl₆]⁴⁻ octahedron is linked to four adjacent octahedra via corner-sharing µ₂–Cl⁻ bridges at the basal positions, with each bridging ligand coordinating two Cu²⁺ centers. This connectivity generates a corrugated two-dimensional network, as evidenced by the near-linear bond angles of 162.763° and 161.735° (Cu1–µ₂Cl2–Cu2 and Cu3–µ₂Cl2–Cu2) and the trans configuration of the axial Cl⁻ ligands (µCl1–Cu1–µCl2; 179.232°), confirming the corrugated 2D topology of this n = 1 Ruddlesden-Popper perovskite (Fig. 3).

Fig. 3. (a) Two-dimensional infinite inorganic chain, and (b) geometric description of the inorganic framework, showing a distorted CuCl₆ octahedron due to the Jahn–Teller effect.

The layered network is oriented perpendicular to the a-axis (Fig. 3), with a repeat sequence of Cu–Cu units in the bc plane every a/2 = 17.253 Å, contributing to the orthorhombic lattice. Each a/2 segment along a (corresponding to the interlayer Cu–Cu distances) contains two 4-bromoanilinium cations within the ab plane (Fig. 3). The bromine atoms point in opposite directions, with a Br⋯Br separation of 16.540 Å, while the nitrogen atoms are located in close proximity, separated by 4.506 Å. The molecular diagram of the organic component (Fig. 3) reveals a distinct layered arrangement parallel to the bc plane and stacked along a. Along the b and c directions, the cations form parallel rows, with successive molecules tilted in alternating orientations, generating a regular zigzag layered pattern. Alternation occurs between type 1 (b-direction) and type 2 (c-direction) cation layers, with a Br⋯Br distance of 5.625 Å between adjacent layers and bromine atoms oriented antiparallel. This arrangement enhances structural stability and promotes efficient crystal packing.

The present compound adopts the typical layered Ruddlesden-Popper (RP) perovskite structure, in which the inorganic framework consists of isolated [CuCl₄]²⁻ units forming corrugated two-dimensional sheets (Fig. 3). The organic cations, (C₆H₇NBr)⁺ (bromoanilinium), are intercalated between these layers, balancing the charge and forming hydrogen-bonded bridges with the chloride atoms. The NH₃⁺ termini of the bromoanilinium cations act as hydrogen-bond donors toward neighboring [CuCl₄]²⁻ anions, generating a dense and asymmetric network of charge-assisted N–H⋯Cl interactions. Most of these hydrogen bonds are nearly linear (two-center), with N–H⋯Cl angles ranging from 147° to 174° and H⋯Cl distances of 2.290–2.470 Å, indicative of strong, directional bonding that stabilizes the RP lattice.³⁰ In addition, several ammonium groups participate in bifurcated (three-center) interactions. These weaker, more delocalized contacts exhibit longer N⋯Cl distances (3.204–3.429 Å) and more bent angles (120–161°), enhancing the overall connectivity of the 2D network. Based on hydrogen-bond topology, chloride ions can be classified as terminal (accepting one H-bond), bridging (accepting two H-bonds forming chain linkages), or multi-bridging (accepting three H-bonds, acting as hubs for both linear and bifurcated interactions). Altogether, the linear N–H⋯Cl bonds tend to orient out-of-plane due to symmetry operations, whereas the bifurcated ones are locally in-plane, collectively reinforcing packing efficiency and the layered or helical nature of the ionic framework.

Besides hydrogen bonding, the aromatic bromophenyl rings adopt a nearly parallel orientation, indicative of weak packing interactions. The large centroid–centroid separations of 6.630, 6.726, and 6.790 Å exclude the presence of true π–π stacking,^31,32 and no significant intermolecular C–H⋯Br interactions are observed. Within each bromoanilinium cation, a H⋯Br contact of ∼2.9 Å represents a weak intramolecular interaction that helps stabilize the cation's conformation, indicating that van der Waals forces predominate in the crystal packing. Shorter centroid⋯H distances (centroid⋯H1 and centroid⋯H2), where H1 and H2 are the closest hydrogens to the centroid, confirm the presence of edge-to-face (C–H⋯π) T-shaped interactions (Fig. S1, SI). For H⋯π contacts parallel to the reversed T-shaped 2BA⁺ cation, the distances range from 2.725 to 3.654 Å, while those parallel to the standard T-shaped arrangement range from 2.761 to 3.698 Å (Fig. 1, SI). Together with C–H⋯Cl contacts (2.976–3.079 Å), these interactions contribute to the overall stability of the supramolecular framework. The combination of N–H⋯Cl hydrogen bonds, edge-to-face (C–H⋯π) interactions, and occasional C–H⋯Cl contacts generates a robust 2D supramolecular architecture, stabilizing the layered structure and subtly influencing electronic properties such as the bandgap, a hallmark of copper halide hybrid perovskites. These interactions bias the orientation of the aromatic rings, often producing a tilted geometry relative to the inorganic plane, thereby modulating interlayer registry and optimizing hydrogen-bond geometries. The bromine atoms on the aromatic rings primarily contribute electron density that enhances van der Waals cohesion between the organic cations and the inorganic sheets.

Infrared analysis

Fig. S2 (SI) presents the infrared spectrum of the studied 2D hybrid perovskite (C₆H₇NBr)₂[CuCl₄], recorded at room temperature over the range 400–4000 cm⁻¹, where the cation is protonated 4-bromoanilinium ((4-BrC₆H₄NH₃)⁺). The spectrum exhibits characteristic absorption bands from both organic and inorganic components, including a broad N–H stretching envelope (3010–2580 cm⁻¹) indicative of strongly hydrogen-bonded –NH₃⁺ groups involved in charge-assisted N–H⋯Cl interactions with the [CuCl₄]²⁻ anion. Prominent bands at 1558 and 1476 cm⁻¹ are assigned to NH₃⁺ deformation and aromatic C C/C–N stretching modes, while a series of absorptions between 1298–954 cm⁻¹ arise from in-plane ring deformations. The sharp peak at 800 cm⁻¹ is diagnostic of C–H out-of-plane bending in para-disubstituted benzene rings, confirming bromine at the 4-position. Low-frequency bands at 694 and 619 cm⁻¹ correspond to aryl C–Br stretching. Below 500 cm⁻¹, intense and structured absorptions reflect the expected Cu–Cl stretching modes of the [CuCl₄]²⁻ anion and associated inorganic lattice vibrations. These spectroscopic features unequivocally confirm the coexistence of the protonated 4-bromoanilinium organic sublattice and the tetrachlorocuprate inorganic framework, fully consistent with the layered 2D structure established by X-ray diffraction. Furthermore, the observed vibrational characters are in good agreement with reports on structurally analogous copper-based hybrid perovskites.^19,20

Semiconductive nature

To gain deeper insight into the electronic transitions, the diffuse reflectance spectrum was analyzed in the UV-vis-NIR region. Fig. 4 presents the raw reflectance and Kubelka–Munk data along with their first derivatives, highlighting the main absorption features. A multi-peak Gaussian deconvolution was carried out to resolve the individual electronic transitions contributing to the absorption spectrum of (C₆H₇NBr)₂[CuCl₄] (Fig. 5). The spectral region below 600 nm is dominated by a broad composite band, which can be deconvoluted into three overlapping peaks centered at approximately 264, 376, and 448 nm. These bands are attributed to ligand-to-metal charge transfer (LMCT) transitions from Cl⁻ to Cu²⁺.^15–18 On the basis of the electronic structure of elongated CuCl₆ octahedra with D_4h symmetry,^13,33 the corresponding LMCT transitions are tentatively assigned and summarized in Table 1, with reference to similar copper-based perovskite structure.¹⁷ In general, for the family of (R–NH₃)₂[CuX₄] hybrid perovskites (X = Cl, Br, or I), the valence band maximum mainly originates from halogen X(3p) orbitals, while the conduction band minimum is dominated by Cu(3d) states, which supports the LMCT character of the observed high-energy transitions.³⁴ At lower energies, a broad absorption band centered at 847 nm is observed and is assigned to d–d transitions of Cu²⁺ ions in a tetragonally distorted octahedral environment.³⁵ Compared to the reference compound (CH₃NH₃)₂[CuCl₄],³⁶ where this band appears at 769 nm, the red-shift indicates a more pronounced axial elongation of the CuCl₆ octahedra in the title compound. This distortion arises from the combined effects of the crystal field and the Jahn–Teller effect,^15–18 which remove the degeneracy of the Cu²⁺ d orbitals and split them into four energy levels: e_g, b_2g, a_1g, and b_1g.¹⁷ Furthermore, the absorption features within the visible region account for the yellow coloration of the complex and provide valuable insight into its electronic structure, highlighting its potential relevance for optoelectronic applications.^25–29

Fig. 4. (a) Solid-state reflectance spectrum with its derivative, and (b) Kubelka–Munk function F(R) plotted as a function of wavelength with its derivative.

Fig. 5. A multi-peak Gaussian fitting of the Kubelka–Munk absorption F(R).

Table 1. Assignment of the observed electronic transitions compared to those of (C₅H₁₄N₂)[CuCl₄].¹⁷.

Assignment	Observed electronic transitions (nm)	Transitions in (C5H14N2)[CuCl4] (nm)
a2u(σ + π) → b1g(x2 − y2)	—	—
eu(σ + π) → b1g(x2 − y2)	264	249
eu(π + σ) → b1g(x2 − y2)	—	288
b2u(π) → b1g(x2 − y2)	376	349
eu(π + σ) → b1g(x2 − y2)	—	406
a2u(π + σ) → b1g(x2 − y2)	470	448
b2g(xy) → b1g(x2 − y2)	847	830

To quantify the optical band-gap energy, the reflectance data were transformed using the Kubelka–Munk function (eqn (2)). In this method, F(R) versus hν is generated, where F(R) is proportional to the absorption coefficient α, and hν is the photon energy (eqn (3)). This approach permits the estimation of the direct band-gap by applying the Kubelka–Munk formalism to the reflectance data and performing a linear extrapolation of the appropriate region, following established procedures:³⁷

F(R) = (1 − R)²/(2R) 2

where F(R) is proportional to the absorption coefficient α and R is the reflectance. Given that;³⁸

[F_KM(R) × hν] = A(hυ − E_g)^m 3

Here, hν is the photon energy, A is a proportionality constant, and the exponent m defines the nature of the electronic transition (m = 1/2 for direct allowed and m = 2 for indirect allowed transitions).

Elucidating the character of the fundamental band gap whether direct or indirect entails a precise analysis of the absorption edge. This is accomplished by examining the trajectory of the absorption coefficient as a function of photon energy (hν), a dependency articulated by the Tauc equation (eqn (4)), which reveals the underlying physics of the electronic transition.²⁹

(αhν)^1/m = A₁(hν − E_g) 4

To adapt the Tauc model for our diffuse reflectance data, we leverage the proportionality between F(R) and α, recasting eqn (5) in terms of the directly measurable F(R):²⁹

(F(R) × hν)^1/m = A₂(hν − E_g) 5

The absorption edge was magnified using a derivative method to enhance the accuracy of the band-gap determination. The energy corresponding to the maximum in the d[ln(F(R)hν)]/d(hν) plot, as governed by eqn (6) and (7), was assigned as the E_g value:²⁹

ln[F(R) × hν] = ln A + m ln(hν − E_g) 6

7

The precise determination of the optical band gap, including both its energy and fundamental character, was achieved through a rigorous two-step analytical methodology. The initial phase of the investigation, detailed in the derivative spectrum of the absorption profile (Fig. 6a), focused on identifying the exact transition onset. A well-defined discontinuity, manifesting as a distinct peak at 2.33 eV, is clearly observed. This feature is a marking the material's fundamental band-gap energy (E_g).

Fig. 6. (a) d[ln(F(R))]/d(hv) vs. hν and (b) [ln(F(R)hν)] vs. [ln(hν − E_g)].

Subsequently, to definitively classify the nature of this electronic transition, a more nuanced analysis was performed. This involved constructing the plot of ln[F(R) × hν] versus ln(hν − E_g) and meticulously examining the slope of its linear region (Fig. 6b). The value of this slope is a direct indicator of the exponent m in the Tauc relation, which governs the transition's probability. The experimentally determined slope was found to be 0.476. This value is in remarkable agreement with the theoretical exponent of 1/2, which is the definitive hallmark of a direct allowed electronic transition. The convergence of these two independent analytical approaches the identification of the transition energy and the verification of its direct nature provides compelling and conclusive evidence that the newly elaborated crystal is unequivocally a direct band-gap semiconductor.

The Tauc plot derived from this analysis are presented in Fig. 7a. A linear extrapolation of this plot yielded a direct bandgap of 2.30 eV, confirmed by a linear fit performed in the absorption edge region of the [F(R) × hν]² plot as a function of photon energy (hν) (Fig. 7b). The fitting yielded a high correlation coefficient (R² = 0.9866), with a slope of 237.946 and an intercept of −545.129. Extrapolation of the linear region gave an optical band-gap energy of 2.29 eV. This value is consistent with those reported for other hybrid organic–inorganic copper-based compounds, confirming the semiconducting nature of the synthesized material.^25–29

Fig. 7. (a) Direct bandgap (E_g) analysis and (b) Linear fit performed in the absorption edge region of the [F(R) × hν]² plot as a function of photon energy (hν).

The degree of disorder in the material can be effectively assessed through the Urbach energy (E_u), which quantifies the energetic width of the exponential band-tail states at the band edges arising from structural and thermal disorder.³⁹ For the two-dimensional hybrid perovskite (C₆H₇NBr)₂[CuCl₄], the Urbach energy (E_u) is estimated to be approximately 0.57 eV (680 meV), as determined from standard Urbach tail analysis of the diffuse reflectance data.⁴⁰ This value is comparable to that reported for the related compound (DAP-H₂)[CuBr₄] (DAP = hexahydrodiazepine), where E_u = 0.62 eV.⁴¹ This value is obtained by fitting the linear region of the absorption tail below the direct optical bandgap of 2.30 eV, yielding a slope of 1.73 in the ln(α) versus photon energy (hν) plot, which corresponds to E_u ≈ 1/1.73. This relatively large Urbach energy reflects a significant degree of energetic and structural disorder. Such behavior is consistent with the intrinsic characteristics of the material, notably the pronounced Jahn–Teller distortion of the CuCl₆ octahedra (Δ = 0.0136) and the corrugated layered Ruddlesden-Popper-like architecture.

The threshold wavelength λ_i = 588 nm (photon energy ≈ 2.11 eV, calculated as E = 1240/λ ≈ 2.11 eV) was determined by linear extrapolation in the wavelength–domain plot of Fig. 8, where (α/λ)² is plotted versus 1/λ. The linear fit intersects the horizontal axis ((α/λ)² = 0) at 1/λ_i, yielding λ_i = 588 nm. This wavelength-scale adaptation of Tauc-like analysis often produces a near-linear region near the absorption onset in low-dimensional or disordered materials such as 2D perovskites, allowing straightforward determination of the effective threshold wavelength for significant photon absorption. Below λ_i (higher energy), absorption is substantial; above λ_i (lower energy), it is weak and dominated by sub-gap tail states. The value of 588 nm (∼2.11 eV) is reasonably consistent with the Tauc-derived direct bandgap E_g = 2.30 eV (∼539 nm). The ∼0.19 eV red-shift is expected, as this method captures the practical onset of strong absorption influenced by the Urbach tail and structural disorder (e.g., Jahn–Teller distortion in CuCl₆ octahedra), while the Tauc E_g represents the extrapolated ideal band-to-band transition energy. Complementary methods (derivative curves (Fig. 5), Tauc 2.30 eV (Fig. 7), this extrapolation ∼2.11 eV (Fig. 8b)) show good overall agreement, confirming reliable optical characterization.

Fig. 8. Evaluation of structural disorder and band-edge parameters; (a) Urbach tail energy E_u = 0.57 eV derived from ln(α) vs. hν and (b) optical threshold wavelength λ_i = 588 nm determined from (α/λ)²vs. 1/λ plot.

Solvent-dependent spin-coating of (C₆H₇NBr)₂[CuCl₄] thin films

Thin films of (C₆H₇NBr)₂[CuCl₄] were successfully prepared via a one-step spin-coating method¹⁸ using methanol (MeOH), N,N-dimethylformamide (DMF), and dimethyl sulfoxide (DMSO) as solvents. The complex was completely soluble in all three solvents, producing uniformly yellow-colored films upon deposition (Table 1). A fixed solution concentration of 30 mg in 50 µL was used for each solvent, and the films were deposited at spin speeds ranging from 1000 to 6000 rpm, depending on the solvent (Table 2). After deposition, thermal annealing was performed to improve film quality and remove residual solvent: MeOH-based films were annealed at 60 °C for 10 min, while DMF and DMSO films were treated at 80 °C and 100 °C, respectively, for 30 min. These temperatures were selected according to the solvents' boiling points (64.7 °C for MeOH, 153 °C for DMF, and 189 °C for DMSO), ensuring complete solvent evaporation and optimal film formation.

Table 2. Solvent properties and film color.

Solvent	Boiling point	Solubility status	Color
Methanol	64.7	Soluble	Yellow
N,N-dimethyl formamide	153	Soluble	Yellow
Dimethyl sulfoxide	189	Soluble	Yellow

As shown in Fig. 9a, all deposited films appear yellow and mostly uniform, confirming good crystallization. Slight surface imperfections are visible in some DMF-prepared films, likely due to slower solvent evaporation or minor inhomogeneities in the precursor layer. Solvent volatility and spin speed strongly influenced film morphology (Table 3): methanol, due to its high volatility, required moderate to high spin speeds (3000–4000 rpm) to prevent cracking or incomplete coverage. DMF provided an optimal balance between evaporation rate and film smoothness, yielding uniform films at 2500–3000 rpm. DMSO, being the least volatile solvent, produced thicker and smoother films but required careful annealing at 100 °C to remove residual solvent. The solvent plays a crucial role in determining the nucleation, growth, and final morphology of the (C₆H₇NBr)₂[CuCl₄] thin films. Optical microscopy images (Fig. 9b) show clear differences depending on the solvent used. The DMF-based film exhibits sparse coverage with isolated nuclei and defects, indicating low nucleation and limited crystal coalescence. The MeOH-based film shows higher nucleation density with aggregated crystalline domains that are partially connected. In contrast, the DMSO-based film presents a more uniform and compact morphology with fine, densely distributed grains, resulting in improved surface coverage despite remaining polycrystalline. Overall, DMF leads to poor film continuity, MeOH produces partially connected island structures, and DMSO yields the most homogeneous and well-covered morphology due to higher nucleation density and more controlled crystal growth.

Fig. 9. (a) Single crystals and optical appearance of (C₆H₇NBr)₂[CuCl₄] thin-film perovskite samples prepared using different solvents and spin-coating speeds (2000–4000 rpm for 30 s) in DMF and methanol. The top row corresponds to films prepared from DMF solution, the middle row corresponds to films prepared from MeOH solution, while the bottom row corresponds to films prepared from DMSO. (b) Thin films prepared from DMSO at 3000 and 1000 rpm for 30 s. (c) Optical microscope images of the prepared films from left to right: MeOH, DMF, and DMSO.

Table 3. Solution concentration and spin-coating parameters.

Solvent	Concentration	Spin-coating
Methanol	30 mg/50 µL	2000 rpm/30 s; 3000 rpm; 4000 rpm and 6000 rpm
N,N-dimethyl formamide	30 mg/50 µL	2000 rpm/30 s; 2500 rpm; 3000 rpm and 4000 rpm
Dimethyl sulfoxide	30 mg/50 µL	1000 rpm/30 s; 2000 rpm; 3000 rpm

Optical absorption behavior and solvent-dependent film quality

The optical properties of the (C₆H₇NBr)₂[CuCl₄] thin films were investigated using UV-vis absorption spectroscopy, as shown in Fig. 10. Panel (a) compares films prepared from three different solvents; methanol (MetOH), dimethylformamide (DMF), and dimethyl sulfoxide (DMSO). For the UV-vis spectra, all samples exhibit two characteristic absorption peaks at approximately 330 nm and 410 nm, corresponding to electronic transitions within the Cu²⁺-based layered perovskite structure. The film synthesized using methanol shows the highest absorption intensity (≈1.25), indicating enhanced film compactness, superior crystallinity, or greater thickness compared to the others. The DMF-prepared film presents moderate absorbance (∼0.7), suggesting less uniform coverage or partial transparency, while the DMSO-based film exhibits very weak absorption (<0.2), implying poor film formation or incomplete crystallization. These results demonstrate that solvent selection plays a crucial role in determining film quality and optical performance, with methanol providing the most homogeneous and optically active films.

Fig. 10. (a) Comparison of thin films synthesized from methanol (MetOH), DMF, and DMSO solvents, showing the influence of solvent choice on optical absorption intensity and film quality. (b) Absorption spectra of a methanol-based film recorded immediately after preparation and after five days of ambient exposure, demonstrating excellent optical stability.

Panel (b) illustrates the absorption spectra of a methanol-based film measured immediately after preparation and after five days of ambient exposure. Both spectra retain identical features, with only a slight increase in absorbance over time, indicating the absence of degradation or structural transformation. This minor variation may be attributed to slow solvent evaporation or further crystallization during storage. Overall, the (C₆H₇NBr)₂[CuCl₄] thin films exhibit excellent optical stability and reproducibility, confirming their robustness and suitability for potential optoelectronic applications.

In summary, the morphological and optical analysis of the (C₆H₇NBr)₂[CuCl₄] thin films reveals that solvent selection plays a pivotal role in determining both surface coverage and optical performance. While DMSO promotes the most uniform surface morphology due to its controlled evaporation kinetics, it yields films of inferior optical density, likely as a consequence of incomplete crystallization. Methanol (MeOH), on the other hand, emerges as the optimal solvent, producing films with the highest optical quality and crystalline density.

Solar cell simulation methodology

In this work, we initially attempted to fabricate a complete experimental solar-cell device; however, a major material incompatibility issue was encountered during the deposition of the hole transport layer (HTL). In particular, the conventional dopants used for Spiro-OMeTAD (LiTFSI and tBP) caused dissolution of the underlying copper-based perovskite layer during the spin-coating process. Since undoped Spiro-OMeTAD does not provide sufficient hole mobility or efficient charge transport, a functional experimental device could not be successfully completed using the current fabrication approach. This solvent/dopant incompatibility represents an important observation for copper-based perovskite systems and suggests that future studies should explore alternative HTL materials, protective interlayers, or modified deposition strategies to achieve stable device fabrication. Therefore, to avoid any uncertainties and to further evaluate the photovoltaic potential of the material, we proceeded with numerical solar cell simulations.

The numerical simulations of the perovskite solar cell were performed using the one-dimensional Solar Cell Capacitance Simulator (SCAPS-1D),⁴² an open-source drift-diffusion software developed at Ghent University Electronics and Information Systems (ELIS) Department. The simulator self-consistently solves Poisson's equation together with the electron and hole continuity equations under steady-state conditions,⁴³ enabling accurate modeling of charge-carrier generation, transport, and recombination processes in thin-film photovoltaic devices.

Poisson's equation is expressed as follows:

8

where p and n represent the free hole and electron densities, respectively, while N_D and N_A denote the concentrations of ionized shallow donors and shallow acceptors. The terms n_t (x) and p_t (x) indicate the densities of trapped electrons and trapped holes, respectively, and ε is the dielectric constant of the material. The symbol ψ represents the electrostatic potential.

Continuity equation for electrons:

9

In this expression, J_n stands for the electron current density, R_n is the total recombination rate for electrons, and G_n corresponds to the generation rate of electrons.

Electron current density:

10

where q is the elementary charge, µ_n is the hole mobility, E the electric field, D_n the hole diffusion coefficient, and n represents the number density of electron.

The continuity equation for hole:

11

where J_p denotes as hole current density, R_p is the total recombination rate for hole, and G_p is the generation rate for hole.

12

where q is the elementary charge, µ_p is the hole mobility, E the electric field, D_p the hole diffusion coefficient, and p the free hole density.

In steady-state conditions , SCAPS-1D iteratively solves this coupled system of eqn (8)–(12) self-consistently across the device structure.

The investigated device adopts a conventional n–i–p architecture,⁴⁴ illuminated from the transparent front contact side under standard test conditions (AM1.5 G spectrum, 100 mW cm⁻², 300 K). The full layer sequence consists of glass, FTO (fluorine-doped tin oxide) as the front transparent electrode, a compact TiO₂ electron transport layer (ETL), the perovskite absorber (C₆H₇NBr)₂[CuCl₄], Spiro-OMeTAD as the hole transport layer (HTL), and Au as the back metal contact electrode (Fig. 11). Key material parameters, including bandgaps, electron affinities, doping concentrations, carrier mobilities, defect densities, were selected based on well-established literature values for high-efficiency perovskite devices^25,44 and are summarized in Table 4.

Fig. 11. Design configuration of the (C₆H₇NBr)₂[CuCl₄] PSC.

Table 4. Baseline electrical and optical parameters used in SCAPS-1D simulations.^25,44.

Parameter	FTO	ETL (TiO2)	(C6H7NBr)2[CuCl4]	HTL (Spiro-OMeTAD)
Thickness (nm)	200	80	320	120
Bandgap Eg (eV)	3.3	3.3	2.30	3.6
Acceptor density NA(cm−3)	0	0	10	1018
Donor density ND (cm−3)	1019	1020	1.1 × 1010	0
Hole mobility µh (cm2 V−1 s−1)	101	107	101	107
Electron mobility µe (cm2 V−1 s−1)	2 × 101	102	101	107
Effective density of states VB Nv (cm−3)	1019	2.2 × 1019	1018	1019
Effective density of states CB Nc (cm−3)	1019	2.2 × 1019	1018	2.8 × 1018
Electron affinity (eV)	4	4.4	4.1	1.46
Dielectric permittivity	9	11	13.6	12
Defects Nt (cm−3)	1010	1015	1014	1016

Fig. 12a illustrates the band diagram of a lead-free two-dimensional hybrid copper(ii)-based perovskite solar cell with the configuration FTO/TiO₂/(C₆H₇NBr)₂[CuCl₄]/Spiro-OMeTAD/Au. The energy band diagram clearly illustrates the overall device architecture and charge transport mechanisms across the different layers. On the left side (x ≈ 0–120 nm), the TiO₂ electron transport layer (ETL) is observed, characterized by a high conduction band edge (E_c ≈ 3.57 eV) and a valence band edge (E_v) close to 0 eV. The middle region (x ≈ 120–520 nm) corresponds to the perovskite absorber layer, which exhibits a lower band gap, while the right side (x > 520 nm) represents the Spiro-OMeTAD hole transport layer (HTL) followed by the Au contact. This configuration is fully consistent with the intended device structure. At the TiO₂/perovskite interface (around x ≈ 120 nm), a pronounced downward conduction band step is observed, where E_c decreases sharply from approximately 3.57 eV in TiO₂ to about 1.2 eV in the perovskite, producing a large conduction band offset (CBO ≈ 2.37 eV). This energetic alignment promotes efficient electron extraction toward the ETL. Simultaneously, the significant valence band offset (VBO ≈ 1.05, −0.05 eV to −1.12 eV eV) creates an effective barrier that suppresses hole leakage into the ETL. The quasi-Fermi levels further confirm this behavior, with the electron quasi-Fermi level (F_n) remaining high and the hole quasi-Fermi level (F_p) positioned lower, indicating strong carrier separation and non-equilibrium conditions under operation. Within the perovskite absorber layer, both E_c and E_v show a gradual slope, reflecting the presence of a built-in electric field across the depletion region. At the perovskite/Spiro-OMeTAD interface (480–520 nm), the conduction band E_c exhibits a small positive cliff of 0.3 eV (from 1.47 eV before the notch to 1.17 eV after), which acts as a localized barrier that limits electron backflow into the HTL. In contrast, the valence band shifts downward, facilitating efficient hole extraction into the Spiro-OMeTAD layer (VBO from −0.85 eV to −2.12 eV across the narrow spike region). The quasi-Fermi levels remain well separated across this region, maintaining favorable alignment for charge collection (F_n high near 3.1 eV on the left, F_p near 0 eV initially). Overall, the significant splitting between F_n and F_p across the absorber layer reflects strong quasi-Fermi level separation, which is directly associated with a high open-circuit voltage potential. This confirms efficient carrier injection and extraction mechanisms within the device under illumination.^45,46

Fig. 12. (a) Calculated energy band structure and (b) simulated J–V curve of the device based on (C₆H₇NBr)₂[CuCl₄] absorber.

The numerical simulations began with equilibrium calculations in the dark at zero bias to ensure numerical stability and convergence. Illuminated current–voltage (J–V) characteristics were then obtained by sweeping the voltage from 0 V to 1.5 V under standard AM1.5G illumination (100 mW cm⁻², 300 K), employing fine voltage steps of 0.150 V near the maximum power point to precisely capture the photovoltaic response. The initial simulation yields the main photovoltaic performance parameters, namely the short-circuit current density (J_sc), open-circuit voltage (V_oc), fill factor (FF), and power conversion efficiency (PCE, denoted η). The corresponding results are PCE = 9.38%, J_sc = 10.99 mA cm⁻², V_oc = 1.53 V, and FF = 55.43%. Here, we observe that a simulated device exhibiting a high V_oc (1.53 V) together with a low FF (55.43%) is inconsistent, since high V_oc devices typically also present higher fill factors. To resolve this discrepancy, we refined the interface defect model by introducing more realistic parameters, including neutral defect states with reduced capture cross-sections (10⁻¹⁹ cm⁻²), a controlled energetic distribution, and an optimized defect density (Table 5). In addition, series and shunt resistances were incorporated into the SCAPS model to better represent real device behavior. A finite series resistance of R_s = 0.5 Ω cm⁻² was introduced to account for resistive losses associated with charge transport through the layers and contacts, while a very high shunt resistance (R_sh ≈ 1030 Ω cm⁻²) was used to minimize leakage currents.

Table 5. Interface defect parameters.

Parameter	Value
Interface	HTL/Perovskite; Perovskite/ETL; ETL/FTO
Defect type	Neutral
Capture cross-section (electrons/holes) (cm2)	10−19
Energetic distribution	Single
Energy level (Et)	0.6
Total density (cm−2)	1.0 × 108

Following these modifications, the simulated device shows improved and more realistic performance (Table 6). The results confirm that the previously overestimated V_oc (1.53 V) was mainly due to an incomplete description of recombination processes, as well as the influence of series resistance and shunt effects in the device. Overall, the final simulation provides a better balance between V_oc and FF, highlighting the importance of interface engineering and defect control in achieving promising lead-free perovskite solar cells. The final simulated photovoltaic parameters, extracted from the J–V curve (Fig. 12b), are: PCE = 9.43%, FF = 71.50%, J_sc = 10.99 mA cm⁻², and V_oc = 1.2 V.

Table 6. Photovoltaic performance parameters (I–V characteristics) of perovskite solar cells incorporating as absorber layer.

Perovskite absorber	J sc (mA cm−2)	V oc (V)	FF (%)	PCE(η) (%)	Ref.
(C6H7NBr)2[CuCl4]	10.99	1.2	71.50	9.43	This work
(C6H10N2)[CuCl4]	7.21	1.73	81.34	10.22	Simulated25
(CH3NH3)2[CuCl2Br2]	21.78	0.845	81.55	15.03	Simulated26
(CH3NH3)2[CuCl4]	8.12	0.560	52.99	2.41	Experimental26
MAPbI3	10.001442	1.0393	86.19	8.96	Simulated44
MAPbI3	10.57	1.03	82	8.94	Experimental47
MAPbBr3	8.7	1.581	67.4	9.27	Experimental48

The performance of perovskite solar cells (PSCs) is strongly influenced by the careful tuning of various layer parameters, including material thickness, band gap energy, defect density, donor and acceptor concentrations, operating temperature, and other factors.^44,49 In this study, we examine the impact of the band gap of the ETL, HTL, and absorber layer across different sections of the device.

Optimization of the electron transport layer (ETL)

In perovskite solar cells, the electron transport layer (ETL) plays an essential role by efficiently collecting electrons generated in the perovskite absorber and transporting them to the front electrode.⁵⁰

The SCAPS-1D simulation results show how the electron transport layer (TiO₂) settings affect the solar cell performance of a lead-free copper-based perovskite. For the band gap range (2.9–3.4 eV) (Fig. 13), PCE and V_oc show a slight decreasing trend, with their maximum values observed near 2.9 eV (PCE ≈ 9.4235% and V_oc ≈ 1.1998 V). The J_sc remains nearly constant throughout the investigated range (∼10.9980–10.9962 mA cm⁻²), indicating that the photocurrent is only weakly affected by the ETL band gap variation. Similarly, the FF remains almost constant at approximately 71.41% in the 2.9–3.3 eV range, with a slight increase to about 71.42% at 3.4 eV. These results indicate that the ETL band gap has only a limited influence on the overall device performance within the investigated interval. The highest PCE (∼9.4235%) is achieved around 2.9 eV, close to the typical TiO₂ band gap values.⁵¹

Fig. 13. Impact of band gap optimization on (a) PCE and V_oc (b) J_sc and FF.

Optimization of the hole transport layer (HTL)

The hole transport layer (HTL) is responsible for selectively extracting photogenerated holes from the perovskite absorber and providing an effective pathway for their transport to the back electrode.⁵²

The study examines how changes in the band gap of the hole transport layer (Spiro-OMeTAD) affect the performance of copper-based perovskite solar cells (Fig. 14). In the band gap range of ∼3.56–3.66 eV (wider than the typical ∼3.0 eV for Spiro-OMeTAD⁵³), the PCE decreases steadily from ∼9.8% to ∼7.8% as the band gap increases. V_oc increases slightly from ∼1.18 V to ∼1.25 V, while J_sc drops from ∼11.1 to ∼9.8 mA cm⁻² due to reduced absorption or extraction, and FF falls from ∼75% to ∼63% because of higher transport barriers. Overall, the simulations suggest that a band gap around ∼3.56 eV (or as low as possible in this range) gives the best balance between FF and PCE, supporting efficient hole extraction and higher J_sc, while lower band gaps improve selectivity and overall efficiency.

Fig. 14. Impact of band gap optimization on (a) PCE and V_oc (b) J_sc and FF.

Optimization of the perovskite absorber layer

The perovskite absorber layer is the central component of PSCs, where it absorbs incident solar photons and generates free charge carriers (electrons and holes).⁵⁴

The SCAPS-1D simulation results (Fig. 15) show how the parameters of the perovskite absorber layer affect the performance of the copper-based solar cell (C₆H₇NBr)₂[CuCl₄], which has an experimental band gap of 2.30 eV. In the band gap range of ∼1.6–2.8 eV (typical for Cu-based hybrids²⁵), the PCE reaches a maximum around 1.9 eV (∼9.42%). V_oc increases from ∼1.9977 V and then levels off around 1.9980 V, reflecting reduced non-radiative losses. The J_sc remains nearly constant (∼10.9973 mA cm⁻²) up to a band gap value of 2.30 eV, then gradually decreases to 10.9958 mA cm⁻² at 2.7 eV, while the FF peaks at approximately 2.7 eV (∼71.423%), it remains nearly constant at about 71.413% within the band gap range of 1.9–2.3 eV. This indicates that the optimal band gap (∼1.9 eV) ensures good alignment with transport layers like TiO₂ and Spiro-OMeTAD, reduces interfacial barriers and recombination, and maximizes light absorption across the solar spectrum. Although the synthesized absorber (C₆H₇NBr)₂[CuCl₄] exhibits an experimental band gap of 2.30 eV, the SCAPS-1D simulations explored a realistic range of band-gap values in order to understand how variations in the absorber energy gap affect the interplay between J_sc, V_oc, FF, and overall power conversion efficiency (PCE). Such parametric optimization is a common practice in simulation studies and does not imply that the experimental material intrinsically possesses the optimized band-gap value. Nevertheless, from an experimental perspective, partial substitution of Cl⁻ by Br⁻ or I⁻ could reduce the band gap through stronger orbital overlap and an upward shift of the valence-band maximum.²⁶ This could extend light absorption toward longer wavelengths and improve photocurrent generation.²⁶ A similar trend has been reported for several copper-based mixed-halide perovskites, such as (CH₃NH₃)₂CuCl₄, (CH₃NH₃)₂CuCl₂Br₂, and (CH₃NH₃)₂CuCl₂I₂, where partial halide substitution significantly affected the optical and photovoltaic properties. The corresponding band-gap energies were reported to be 2.36, 1.99, and 1.04 eV, respectively, demonstrating a progressive band-gap reduction with halide modification. In addition, the powders exhibited a distinct color transition from yellow to dark brown as the Cl/Br/I ratio changed, further confirming the strong dependence of the electronic structure and optical absorption on the halide composition.²⁶

Fig. 15. Impact of band gap optimization on (a) PCE and V_oc (b) J_sc and FF.

Conclusion

The comprehensive structural, spectroscopic, optical and thermal characterization of (C₆H₇NBr)₂[CuCl₄] demonstrates its formation as a robust 2D Ruddlesden-Popper (RP) hybrid perovskite with n = 1. The crystal architecture features discrete, corner-sharing [CuCl₆]⁴⁻ octahedra forming corrugated inorganic layers, alternated with planar 4-bromoanilinium cations that orchestrate a dense network of directional N–H⋯Cl hydrogen bonds, near edge-to-face (C–H⋯π) interactions, and C–H⋯Cl contacts. This highlights the subtle interplay between the organic and inorganic motifs in defining the material's geometry and cohesion. The copper-chloride octahedra exhibit pronounced Jahn–Teller distortions, quantified by a bond-length distortion parameter Δ = 0.0136, reflecting the intrinsic electronic asymmetry of Cu²⁺ centers and its influence on layer corrugation. Optical analyses reveal a direct bandgap of 2.30 eV, corroborated by both Tauc and derivative methods, indicating semiconducting behavior suitable for optoelectronic applications. Thermal studies confirm the material's stability up to ∼150–160 °C, beyond which partial decomposition of the organic scaffold occurs consistent with the behavior of related Cu(ii)-based perovskites. Jointly, these findings underscore the synergistic design of organic–inorganic hybrid lattices, where the delicate balance of hydrogen bonding, near π-stacking, and halogen-mediated contacts governs both structural robustness and electronic functionality. The polar, layered framework of (C₆H₇NBr)₂[CuCl₄] not only enriches the chemistry of copper-based hybrid perovskites but also establishes a versatile platform for exploring next-generation optoelectronic devices, where tunable structural motifs directly modulate optical and electronic properties. In addition, the photovoltaic behavior of this material as a light-absorbing layer was explored through SCAPS-1D numerical predictive tool to investigate the device architecture FTO/TiO₂/(C₆H₇NBr)₂[CuCl₄]/Spiro-OMeTAD/Au, focusing on the band gap optimization of the absorber layer, hole transport layer, and electron transport layer. Modeling results identified optimal parameters for the active material (C₆H₇NBr)₂[CuCl₄] at a band gap of 1.9 eV; for the Spiro-OMeTAD 3.56 eV; for the TiO₂ 2.9 eV. with key parameters at 300 K yielding J_sc = 10.99 mA cm⁻², V_oc = 1.2 V, FF = 71.50%, and PCE(η) = 9.43%. These findings highlight the potential for designing environmentally friendly, lead-free perovskite solar cells with improved performance through careful optimization of material properties and device parameters. However, it is important to emphasize that the presented results are based solely on numerical simulations. Therefore, no experimental device fabrication has been carried out in this study. The experimental realization, fabrication, and subsequent validation of these devices will be the focus of future work to confirm and complement the simulated performance trends.

Author contributions

Wissem Hallab: software, formal analysis, writing – original draft. K. S. Al-Namshah: investigation, validation. Mahdi Gassara: formal analysis, investigation. Ahlem Guesmi: investigation, formal analysis. Noureddine Mhadhbi: methodology, investigation. Naoufel Ben Hamadi: formal analysis, investigation. Marcus Korb: data curation, investigation. Houcine Naïli: validation, supervision, review & editing.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding

This work was supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) (grant number IMSIU-DDRSP2603).

Data availability

CCDC 2499237 contains the supplementary crystallographic data for this paper.⁵⁵

The data that supported the findings of this study were available upon reasonable request.

Supplementary information (SI): additional structural, spectroscopic, thermal, and Hirshfeld surface analysis data for (C₆H₇NBr)₂[CuCl₄]. See DOI: https://doi.org/10.1039/d6ra03133a.