Unveiling the Potential of Novel Ternary Chalcogenide SrHfSe₃ for Eco-Friendly, Self-Powered, Near-Infrared Photodetectors: A SCAPS-1D Simulation Study

Abstract

Ternary chalcogenide-based sulfide materials with distorted morphologies such as BaZrS₃, CaZrS₃, and SrZrS₃, have recently gained much attention in optoelectronics and photovoltaics due to their high structural and thermal stability and compatibility with low-cost, earth-abundant synthesis routes. However, their relatively large bandgaps often limit their suitability for near-infrared (NIR) photodetectors. Here, we conducted a comprehensive investigation of SrHfSe₃, a ternary chalcogenide with an orthorhombic crystal structure and distinctive needle-like morphology, as a promising candidate for NIR photodetection. SrHfSe₃ exhibits a direct bandgap of 1.02 eV, placing it well within the NIR range. Its robust structure, high temperature stability, phase stability and natural abundance make it a compelling material for next-generation, self-powered NIR photodetectors. An in-depth analysis of the SrHfSe₃-based photodetector was performed using SCAPS-1D simulations, focusing on key performance metrics such as J–V behavior, photoresponsivity, and specific detectivity. Device optimization was achieved by thoroughly altering each layer thickness, doping concentrations, and defect densities. Additionally, the influence of interface defects, absorber bandgap, and operating temperature was assessed to enhance the photoresponse. Under optimal conditions, the device achieved a short-circuit current density (Jsc) of 45.88 mA/cm², an open-circuit voltage (Voc) of 0.7152 V, a peak photoresponsivity of 0.85 AW⁻¹, and a detectivity of 2.26 × 10¹⁴ Jones at 1100 nm. A broad spectral response spanning 700–1200 nm confirms its efficacy in the NIR region. These results position SrHfSe₃ as a strong contender for future NIR photodetectors and provide a foundation for experimental validation in advanced optoelectronic applications.

1. Introduction

Photodetectors are optoelectronic gadgets that convert optical signals into their equivalent electrical signals [1]. The process begins when the absorber material interacts with incoming photons, leading to the generation of electron–hole pairs. These charge carriers are then driven by an internal electric field, resulting in the production of a photocurrent or electrical signal [2]. They are currently becoming indispensable devices in our daily life due to their various applications from optical communication to environmental sensing, missile warning, aviation, target detection, biomedical imaging, and security [3,4,5,6].

To evaluate the photodetector performance, various performance metrics are required to fully determine the device’s capability. The detector responsivity quantifies how the device can convert the input light intensity into an electrical signal. It is primarily governed by the material’s charge transport properties, carrier mobility, and optical absorption [7]. The dark current, which is the response of the photodetector while there is no illumination, represents the saturation carrier flow. Elevated dark current levels may obscure weak photocurrent signals and degrade the signal-to-noise ratio. The detectivity, however, is the measure of the photodetector’s sensitivity to the lowest signal, and is one of the key factors, which is governed by the dark current of the photodetector [8]. The ability of the photodetector to respond selectively and efficiently across different spectral wavelengths is also a pivotal characteristics that determine its suitability for applications such as UV, visible, or infrared sensing [9]. Apart from these quantifiable characteristics, the device’s ability to operate in self-powered mode is vital for applications requiring low energy consumption and battery-free sensing [10].

Most of the photodetectors available on the market rely on conventional materials such as InGaAs, and HgCdTe, which provide high detectivity and fast response times [11,12]. However, InGaAs and HgCdTe-based photodetectors face several challenges, including high manufacturing costs, reliance on cryogenic temperature control, and the presence of toxic or heavy-metal elements [13,14]. Due to its abundance, non-toxicity, and well-matured manufacturing technology, silicon has become a standard semiconductor material for commercialized photodetectors. However, it has a bandgap of 1.12 eV, which limits its light absorption to wavelengths below 1100 nm. This impedes its effectiveness for the near-infrared region, such as optical communication applications [15]. Although the Ge has a narrow bandgap of 0.66 eV [16], allowing for detecting the near-infrared wavelength, its indirect bandgap nature requires a thicker Ge layer to absorb sufficient light across the spectrum [17]. In addition, it suffers from thermal instability and a large dark current density of approximately 1.12 mA/cm² [18]. Many current NIR photodetectors rely on narrow bandgap materials such as PbS quantum dots (QDs) and InSb due to their high optical absorption and sensitivity [19,20]. However, these materials face significant challenges; for instance, PbS is inherently toxic, while InSb is expensive. These drawbacks hinder their integration into large-scale commercial devices [21].

Graphene has emerged as a layered material with remarkable properties, including ultra-high carrier mobility and broadband optical absorption [22]. Despite these encouraging features, it has a low photoresponse and a high dark current due to its relatively modest absorption rate of only ~2.3% [23]. Even though black phosphorus (BP) is a layered material with a tunable bandgap ranging from 0.3 to 2 eV [24], allowing for spectral tailoring from the visible to the near-infrared spectral wavelength. However, it suffers from environmental instability, making it challenging to design durable photodetectors. Likewise, engineering heterostructures based on BP is costly due to the complexity and the need for intricate transfer processes [25,26]. Moreover, tellurium (Te) with a layered structure has attracted considerable attention due to its tunable bandgap (0.3–1.0 eV), enabling broadband photodetection in the mid- to near-infrared range. This material exhibits high inherent carrier mobility (∼10⁵ cm²/V·s) and strong optical absorption, good air stability, and excellent electrical conductivity, making it a promising candidate for NIR photodetector applications [27,28,29,30]. However, the limited natural abundance of tellurium, along with complex manufacturing procedures, increases production costs and limits their suitability for large-scale deployment [31]. Meanwhile, organic and metal halide perovskites exhibit intriguing optical and electronic properties. They possess tunable band gaps and can be fabricated at low temperatures using facile solution-processing methods. However, they suffer from toxicity concerns, susceptibility to environmental degradation, and poor thermal stability, making them unfavorable for long-term commercial applications [32,33]. Consequently, there is a pressing need to address these challenges by exploring alternative materials that offer low toxicity, defect tolerance, cost-effectiveness, high stability, strong optical absorption, and sensitivity in the near-infrared region, which is pivotal for advancing the field.

Among these promising materials are ternary chalcogenides with the general formula MNY₃, where M signifies group II cations, N denotes group IV transition metals, and Y corresponds to chalcogen anions. These materials hold potential due to their high stability, earth abundance, and eco-friendly characteristics. In addition, the band gaps of this material’s family range from 0.3 to 2.3 eV, and they exhibit high optical absorption, making them ideal semiconductors for various optoelectronic devices [34,35]. These semiconductors exhibit two distinct structural phases: the β phase adopts a distorted perovskite (DP) structure analogous to GdFeO₃, while the α phase resembles the NH₄CdCl₃-type structure, commonly referred to as the needle-like (NL) phase. [36]. Several research studies have been conducted on the development of photodetectors using chalcogenide perovskites with a distorted (β) phase. For instance, BaZrS₃ photodiodes have been reported to exhibit a responsivity of 46.5 mA/W at 5 V at incident light of 532 nm [32]. Similarly, Yurun et al. fabricated a photoconductor based on distorted perovskite SrZrS₃ with a high bandgap of 2.9 eV, exhibiting excellent responsivity of approximately 8 A/W at a wavelength of 405 nm [37]. Yang et al. designed a HfSnS₃ nanowire-based photoconductor demonstrating wideband photoresponse from UV to NIR, with high responsivity of 11.5 A/W and detectivity of 8.2 × 10¹¹ Jones. [38]. However, nearly all these devices are based on distorted ternary chalcogenide perovskites containing sulfur, which have larger band gaps, limiting their spectral response mainly to the visible regime. Even though HfSnS₃ demonstrated photoresponse up to 802 nm, it cannot detect light beyond 1000 nm, making it unsuitable for near-infrared (NIR) detection. Therefore, selenide-based ternary chalcogenides with needle-like (α) phase have shown a high promise with a narrow bandgap [39,40]. In addition, these materials-based selenides exhibit narrow bandgaps and higher carrier mobilities as high as 100 cm² V⁻¹ s⁻¹ compared with sulfur-based counterparts, making them a strong candidate for NIR photodetection [34].

Among these ternary chalcogenides with needle-like phases, SrHfSe₃, identified as a potential semiconductor, possesses a narrow direct bandgap of 1.02 eV, as confirmed by diffuse reflectance spectroscopy for SrHfSe₃ that has been synthesized using a high-temperature solid-state reaction [39]. In addition, DFT calculations using the HSE06 hybrid approximation function further support this finding, consistently predicting a direct optical bandgap close to 1.0 eV [40]. This agreement between theoretical and experimental results makes it highly suitable for near-infrared photodetectors. It also possesses a high Spectroscopic Limited Maximum Efficiency (SMSE) of 26.64%, small effective masses ($m_e^{*}$ < 1.0 and $m_{h }^{*}$ < 1.0), high static dielectric constant, and strong light absorption in the order of 10⁵ cm⁻¹ [40]. In addition, SrHfSe₃ exhibits a low thermal conductivity of 0.9 W/m·K at ambient conditions, which further decreases to 0.77 W/m·K upon 1 mol % Sb-doping, an advantageous feature for suppressing thermal noise in photodetectors [34,39]. Apart from the electronic, optical, and transport properties, the SrHfSe₃ exhibits excellent structural and phase stability, along with strong defect tolerance and longer non-radiative carrier lifetimes (1.18 ns), indicating it has exciting properties [40]. Like other ternary chalcogenides with a needle-like phase, SrHfSe₃ crystallizes in an NH₄CdCl₃-type orthorhombic structure (Pnma, No. 62), characterized by edge-sharing HfSe₆ octahedra that form quasi-one-dimensional chains [34]. Moreover, thermodynamic stability, indicated by a formation energy (E_f < 0 eV per atom) and an energy above the hull (∆E_h < 0.1 eV per atom), suggests that this compound is stable and does not require special synthesis methods [40]. Owing to its intriguing properties, SrHfSe₃ can be utilized in a wide range of applications, including photodetectors, sensors, light-emitting diodes (LEDs), thermoelectric devices, and optical communication systems [40]. Despite its promising characteristics, no theoretical or experimental studies have yet been reported in the literature on a photodetector based on the SrHfSe₃ compound.

Zinc selenide (ZnSe) is a binary chalcogenide semiconductor belonging to an II–VI family with a wide band gap of 2.7 eV at room temperature, employed in this work as a window layer. It owns high electron and hole mobilities, rendering its suitability as a window layer in heterojunction devices, where it promotes efficient carrier transport while minimizing recombination [41]. ZnSe’s well-aligned conduction band facilitates efficient electron transportation while effectively blocking holes, thereby suppressing interfacial recombination and minimizing leakage current, which collectively enhances the device’s overall performance. With its high chemical stability and resistance to humidity, ZnSe is well-suited for industrial-scale production [42]. It exhibits significant resistance to intense UV and X-ray radiation, and a high breakdown electric field strength of around 1 MV/cm. ZnSe can be fabricated using various methods, such as thermal evaporation, metal–organic vapor-phase epitaxy, molecular beam epitaxy, plasma-enhanced chemical vapor deposition, sputtering, and electrochemical deposition [41,43]. These fabrication routes, combined with its favorable properties, position ZnSe as a promising window or buffer layer material for photodetector applications.

AgCuS is employed as the back surface field (BSF) layer on the other side of the photodetector to enhance hole extraction and reduce non-radiative recombination. It consists of silver, copper, and sulfur, which form a relatively non-toxic composition using earth-abundant elements, making AgCuS well aligned with the goals of cost-effective and sustainable photodetector technologies. With a direct bandgap of around 1.25 eV, strong light absorption, and a notable thermopower of ~665 μV/K at room temperature, AgCuS has recently gained interest in optoelectronics. Although it was originally explored for its thermal phase transitions and ionic conductivity, its electronic and optical properties now position it as a valuable candidate for device applications [31]. AgCuS can be synthesized using hydrothermal methods and solid-state reaction [44,45]. It has a favorable valence band alignment, which facilitates efficient and strong hole extraction while blocking minority carriers (electrons) when used as a back surface field (BSF) layer, thereby enhancing the Voc and Jsc as demonstrated by previous studies [31]. These features, along with suitable lattice energy and high carrier mobility, make it a promising candidate for use as a back surface field (BSF) layer in photodetectors [1].

Motivated by these intriguing properties of the aforementioned materials, we aim to design and simulate a near-infrared (NIR) photodetector based on a ZnSe/SrHfSe₃/AgCuS heterostructure using the SCAPS-1D simulation tool. In this design, ZnSe acts as the window layer, SrHfSe₃ functions as the primary light-absorbing material, and AgCuS works as the back surface field (BSF) layer. A comprehensive analysis was conducted to evaluate the photoresponse characteristics of the SrHfSe₃-based photodetector, focusing on key performance indicators including J–V behavior, photoresponsivity, and specific detectivity. Critical parameter tuning was performed by varying the thickness, doping concentration, and defect density of each layer of the device to attain the optimal condition. The simulation results highlight the promise of SrHfSe₃ as a stable, earth-abundant, and high-efficiency semiconductor for NIR photodetection, offering a solid theoretical basis for future experimental validation and device development.

2. Materials and Methods

Proposed Design and Band Alignment

This section illustrates the simulation method and the band alignment of the proposed SrHfSe₃-based photodetector. Several simulation software packages have been commonly used in the literature to analyze photodetector characteristics, such as SCAPS-1D (3.3.12), Lumerical CHARGE, COMSOL Multiphysics, and Silvaco ATLAS [46,47,48,49]. Among these software packages, SCAPS-1D, developed by Professor M. Burgelman at the University of Gent, has recently attracted significant interest for simulating solar cells and various types of photodetectors, such as Schottky, pn, and pin structures [31,50,51]. SCAPS-1D stands out for its user-friendly interface and free availability, making it a prevalent choice for academic and industrial research. This software allows one to simulate devices consisting of up to seven layers, providing a flexible platform for comprehensive analysis. In addition, it has demonstrated excellent agreement with experimental results across a range of photovoltaic [52,53]. Using SCAPS-1D software, one can optimize many critical parameters for each photodetector layer, such as thickness, doping level, bulk and interface defects, bandgap, metal contacts, and radiative recombination coefficients [51,54]. SCAPS-1D also provides various output parameters essential for device analysis, such as band alignment, photocurrent, dark current, external quantum efficiency (EQE), electric field, temperature-dependent behavior, device capacitance, and generation and recombination rates [55]. The software solves three common semiconductor equations (Poisson’s equation and the electron–hole continuity equation) under 1D conditions [21].

$${\displaystyle \frac{d^2\mathsf{\Psi }}{d{x}^2}}={\displaystyle \frac{q}{{\epsilon }_s}}\left[p\left(x\right)-n\left(x\right)+N_d^{+}-N_a^{-}+p_t\left(x\right)-n_t\left(x\right)\right]$$

(1)

$${\displaystyle \frac{{𝜕}_n\left(x,t\right)}{{𝜕}_t}}={\displaystyle \frac{I}{q}}{\displaystyle \frac{𝜕J_n}{𝜕x}}+G_n\left(x\right)-R_n\left(x\right)$$

(2)

$${\displaystyle \frac{{𝜕}_p\left(x,t\right)}{𝜕t}}={\displaystyle \frac{I}{q}}{\displaystyle \frac{𝜕J_p}{𝜕x}}+G_p\left(x\right)-R_p\left(x\right)$$

(3)

where $\mathsf{\Psi }$ is the electrostatic potential, $q$ symbolizes the electron charge, ${\epsilon }_s$ represents the material permittivity. The $N_d^{+}$ and $N_a^{-}$ denote the donor and acceptor concentrations, in turn. The $p\left(x\right)$ stands for hole concentration while $n\left(x\right)$ represents the electron concentration. Moreover, $p_t\left(x\right)$and $n_t\left(x\right)$ describe the spatial distributions of holes and electrons individually. The $J_n$ and $J_p$ stand for current density for electrons and holes, respectively. The $G_n\left(x\right)$ and $G_p\left(x\right)$ indicate the carrier generation rate for electrons and holes, respectively. The recombination rates for electrons and holes are denoted by $R_p\left(x\right)$ and $R_p\left(x\right)$ as illustrated in Equations (2) and (3), individually.

Figure 1a shows the schematic of the proposed photodetector, which consists of three-layer stacks: ZnSe/SrHfSe₃/AgCuS. The ZnSe layer is highly transparent across a wide spectral range and acts as a window and hole-blocking layer [56]. The SrHfSe₃ layer is the main absorber, having a narrow bandgap, exhibiting strong absorption in the ultraviolet and visible regions. The AgCuS layer functions as a BSF and electron blocking layer, facilitating efficient charge separation. The light enters through the top Ti electrode, passes to the ZnSe layer, and reaches the SrHfSe₃ absorber, where photogenerated carriers are generated and separated by the internal built-in potential. These carriers are further improved by the AgCuS layer, which enhances collection efficiency. The enhanced photogenerated carriers then drift toward the Cu contact, which serves as the anode for the photodetector, completing the device circuit and enabling efficient charge extraction. Figure 1b presents the corresponding device structure as modeled in SCAPS-1D simulations. The device is illuminated from the right side, allowing AM1.5G light to enter through the ZnSe layer.Due to limited experimental data on SrHfSe₃, the electron affinity is anonymously selected to be 3.9 eV, based on analogous values reported for compounds within the same materials family. The ternary chalcogenides with a formula ABSe₃ exhibit an electron affinity in the range of 3.2–4.8 eV [57,58,59].Engineering the band energy alignment between each layer of the photodetector is crucial for understanding charge carrier recombination at heterojunction interfaces and enabling efficient generation and extraction of photogenerated carriers. The band alignment between each layer of the photodetector under illumination is delineated in Figure 1c, which shows the conduction band edge (Ec, black line), valence band edge (Ev, green line), along with the electron quasi-Fermi level (Fn, red line) and the hole quasi-Fermi level (Fp, blue line). The conduction band minima of ZnSe and SrHfSe₃ are located at −4.09 eV and −3.90 eV, resulting in a conduction band offset (CBO) of −0.19 eV. As shown in Figure 1c, a conduction band cliff shape is formed at the interface, which can assist in electron flow toward the ZnSe layer. In addition, based on the bandgaps of ZnSe (2.7 eV) and SrHfSe₃ (1.02 eV), the valence band offset (VBO) at that junction is around 1.87 eV, which effectively blocks hole leakage into the ETL, forming an effective pn heterojunction at the ZnSe/SrHfSe₃ interface.

In contrast, AgCuS, featuring a bandgap of 1.25 eV and an electron affinity of 3.35 eV, serves as the hole transport layer. Its valence band maximum lies at −4.60 eV. Compared to SrHfSe₃, this creates a valence band offset (VBO) of about 0.32 eV, which facilitates efficient hole extraction into the HTL while maintaining a slight barrier to suppress carrier back-injection. Overall, this band alignment promotes directional carrier transport and lowers recombination, which enhances the stability and efficiency of photodetector performance. Equations (4) and (5) were used to determine the CBO and the VBO at the ZnSE/SrHfSe₃ and SrHfSe₃/AgCuS interfaces, respectively [60].

$$CBO=x_{absorber }-x_{g,ETL}=-0.19 eV$$

(4)

$$VBO={(x}_{g,HTL}-x_{HTL)-(}{x}_{absorber}-x_{g,absorber})=0.32 eV$$

(5)

where the $x_{ABS}$ refers to the electron affinity of the absorber, $x_{HTL}$ denotes the electron affinity for the HTL, $x_{g,HTL}$represents the bandgap of the HTL, and the $x_{g,ABS}$is the bandgap of the absorber.

Table 1. Summary for Ec (eV) and Ev (eV) for each layer of the device.

Layer	χ (eV)	Eg (eV)	Ec (eV)	Ev (eV)
ZnSe	4.09	2.70	–4.09	– 6.79
SrHfSe3	3.90	1.02	–3.90	–4.92
AgCuS	3.90	1.25	–3.35	– 4.60

summarizes the Ec (eV) and Ev (eV) for each layer of the photodetector.

Simulating the photodetector’s performance requires knowledge of the absorption coefficient for each layer in the device. In this study, the SCAPS-1D simulation software applies Equation (6) to calculate the absorption coefficient (α) as a function of photon energy $hv$ and bandgap $E_g$ parameters [61].

$$\alpha (\lambda )=\left(A+{\displaystyle \frac{B}{hv}}\right)+\sqrt{hv-E_g}$$

(6)

where the model parameters A and B denote the device dispersion. As shown in Figure 1d, the SrHfSe₃ layer displays a stronger optical absorption compared to ZnSe and AgCuS, positioning it as the primary absorber in the photodetector structure. The absorption profile of SrHfSe₃ reaches values of around 2.25 × 10⁵ cm¹, spanning the ultraviolet region and gradually decreasing with increasing wavelength. The absorption edge approaches 1225 nm, corresponding to the material’s direct bandgap of about 1.02 eV. This strong and broad absorption behavior confirms SrHfSe₃’s suitability for broadband and near-infrared photodetection applications. Hence, the proposed device is expected to display high photoresponsivity and EQE across a broad spectral range of 400–1200 nm, which corresponds to the high optical absorption characteristics of SrHfSe₃.

To calculate the effective density of states in the conduction band ${ N}_C$ and valence bands $N_v$of SrHfSe₃ material, the standard formulas were used as shown in Equations (7) and (8) [62].

$${ N}_C={2\left({\displaystyle \frac{ m_e^{*}T{k}_B}{2\pi {\mathit{ħ}}^2}}\right)}^{3/2}$$

(7)

$$N_v={2\left({\displaystyle \frac{ m_h^{*}T{k}_B}{2\pi {\mathit{ħ}}^2}}\right)}^{3/2}$$

(8)

where $m_e^{*}$ and $m_h^{*}$correspond to the effective masses of electrons (${0.401 m}_e^{*})$and holes (${0.536 m}_h^{*})$, respectively [40]. The $T$reflects the absolute temperature, $k_B$ signifies Boltzmann’s constant, and ℏ is the reduced Planck’s constant, which is given by $\mathrm{\hslash }={\displaystyle \frac{ h}{2\pi }}$, where $h$ is the Planck’s constant (6.634 × 10⁻³⁴ J-s).

To evaluate the photodetector’s performance, the responsivity (A/W) is a crucial parameter that is calculated using the following Equation (9) [63].

$$R={\displaystyle \frac{ \eta *\lambda \left(nm\right)}{1240 \left(nm\right)}}={\displaystyle \frac{ J_{ph}}{P_o}}$$

(9)

where η is quantum efficiency, $\lambda$ represents the incident wavelength, $J_{ph}$ stands for the photocurrent density (A/cm²), and $P_o$ denotes the incident light power density (mW/cm²).

Another important figure of merit for evaluating photodetectors’ performance is the specific detectivity (D*), measured in Jones. It can be calculated using Equation (10) [64].

$$D^{*}={\displaystyle \frac{R}{\sqrt{2q{J}_o}}}$$

(10)

where the q is the elementary charge, and J₀ is the reverse saturation current density.

The reverse saturation current density of the photodetectors operating in self-powered mode is determined using the following Formula (11) [63].

$$J_o={\displaystyle \frac{J_{sc}}{e^{\frac{V_{oc}}{V_{th}}}-1}}$$

(11)

where the Vth is thermal voltage, typically having a value of 25.85 mV under ambient conditions (300 K) [65,66]. Both J_SC and V_OC are obtained from SCAPS-1D software [67].

In this calculation, the thermal velocities of both electrons and holes were assumed to be 1.00 × 10⁷ cm/s. A standard light intensity of 100 mW/cm² was used for the simulations. The analysis of the SrHfSe₃ photodetector was conducted at 1100 nm, as this wavelength closely corresponds to the material’s bandgap energy of 1.02 eV, enabling efficient photon absorption and carrier generation. Subsequently, the device’s spectral photoresponse was examined over a broader wavelength range to evaluate its performance under varying broad spectral wavelength. The temperature was initially set to 300 K and later varied to assess its influence on the photodetector’s behavior. All other parameters related to the absorber, electron transport layer (ETL), and hole transport layer (HTL) are summarized in

Table 2. Input parameters used for simulation.

Structure	ZnSe [41]	SrHfSe3 [34,40]	AgCuS [31]
Thickness (µm)	0.05	1	0.2
Bandgap (eV)	2.7	1.02	1.25
Electron Affinity (eV)	4.09	3.9	3.350
Dielectric permittivity	10.000	7.45	10.000
Effective DOS at CB (cm−3)	1.50 × 1018	5.60 × 1018	1.99 × 1019
Effective DOS at VB (cm−3)	1.800× 1019	5.40 × 1018	1.72 × 1019
Thermal velocity of electron (cm−1)	1.00 × 107	1.00 × 107	1.00 × 107
Thermal velocity of holes (cm−1)	1.00 × 107	1.00 × 107	1.00 × 107
Electron mobility (cm2 cm−1s−1)	5.00 × 101	3.647 ×101	1.00 × 102
Hole mobility (cm2 cm−1s−1)	6.094 × 101	6.094× 101	6.60 × 101
Bulk defect density (cm−3)	1.00 × 1015	1.00 × 1014	1.00 × 1015
Shallow uniform acceptor density NA (cm−3)	0	1.00 × 1017	1.00 × 1019
Shallow uniform doner density Nd (cm−3)	1.00 × 1018	0	0

. The front and back contact parameters used in the simulation are listed in

Table 3. The work functions and surface recombination velocities for the contacts used in the simulation.

Contacts	Unit	Back Contact Parameters	Front Contact Parameters
Metal work function	eV	4.7 [31]	3.84 [55]
Surface recombination velocity of holes	cm/s	1.00 × 107	1.00 × 107
Surface recombination velocity of holes	cm/s	1.00 × 107	1.0 × 107

. It is important to note that, to accurately reproduce the results, the voltage step size in SCAPS-1D simulations was set to 0.0100 V.

3. Results and Discussions

3.1. The Impact of Absorber Layer SrHfSe₃ on the Photodetector Performance

Figure 1a illustrates how varying the thickness of the SrHfSe₃ absorber layer from 0.4 µm to 1.6 µm impacts key photodetector parameters, including Jsc, Voc, responsivity (R), and detectivity (D*). During this thickness-dependent analysis, both the doping density and defect level were fixed at 10¹⁸ cm⁻³ and 10¹⁴ cm⁻³, respectively. As depicted in Figure 2a, Jsc undergoes a noteworthy increase, rising from 43.07 mA/cm² to 46.38 mA/cm² as the absorber thickness increases. This enhancement is attributed to the increased light absorption in thicker layers, leading to greater photogeneration rates and, consequently, higher photocurrents [21]. Conversely, Voc steadily declines as thickness increases, dropping from 0.733 V to 0.704 V, likely due to enhanced bulk recombination in thicker layers, which lowers the carrier lifetime, as carriers tend to recombine before being extracted by the electrodes. Such behavior has also been reported in the literature [31].

The variation in the responsivity and the detectivity as the SrHfSe₃ absorber thickness switches from 0.4 µm to 1.6 µm is delineated in Figure 1b. The amount of responsivity progressively increases from 0.644 A/W at 0.4 µm and reaches its upper limit at 0.879 A/W at 1.6 µm. This steady enhancement with the increase in the thin film width is ascribed to improved QE and enhanced photon-to-current conversion in thicker absorbers. This behavior of R is consistent with findings reported in the literature [42]. At the wavelength of 1100 nm, proximate to the bandgap edge of SrHfSe₃, the responsivity attains its highest point of 0.85 A/W at an absorber thickness of 1 μm.

On the other hand, the detectivity follows a reverse behavior, peaking at its highest limit at 2.53 × 10¹⁴ Jones at 0.4 µm device’s thickness and then falling with increasing film thickness, eventually dropping to 1.9 × 10¹⁴ Jones at 1.6 µm as depicted in Figure 2b. This behavior can be ascribed to the simultaneous enhancement of QE with increasing absorber thickness, and the accompanying decline in photo-carrier collection efficiency as the layer becomes thicker [68]. In addition, the increase in recombination rate serves as the underlying reason for this effect, as demonstrated by Equation (10) [67], which shows that detectivity decreases with rising dark current. Under an incident wavelength of 1100 nm, a peak detectivity of 2.26 × 10¹⁴ Jones was attained when the absorber thickness was at its optimal value of 1 µm. Based on this analysis, it is important to highlight the trade-off between responsivity and detectivity as the absorber thickness varies. From an optimization standpoint, a SrHfSe₃ thickness of 1.0 µm offers favorable device performance, with the photodetector achieving optimal characteristics, including a Jsc of 45.88 mA/cm², a Voc of 0.7152 V, a responsivity of 0.85 A/W, and a detectivity of 2.26 × 10¹⁴ Jones making it a practical candidate for NIR photodetector applications.

Apart from the impact of the absorber thickness, the doping concentration of the absorber layer is a critical parameter that governs the electrostatic potential, carrier lifetime, and recombination dynamics within the photodetector. Therefore, its systematic analysis is necessary to optimize device performance and ensure efficient charge carrier transport with minimal recombination losses. In this investigation, both the thickness and the defect density of SrHfSe₃ were fixed at 1 μm and 1 × 10¹⁴ cm⁻³, respectively. The doping density of SrHfSe₃ was varied across a range from 1 × 10¹⁵ to 1 × 10²⁰ cm⁻³ to clarify its impact on photodetector performance. Figure 2c reveals that at lower doping levels, from 1 × 10¹⁵ to 1 × 10¹⁷ cm⁻³, Jsc slightly decreases ranging from 45.92 to 45.88 mA/cm². As the doping concentration increases to 1 × 10¹⁸ and 1 × 10¹⁹ cm⁻³, Jsc gradually decreases to 45.736 and 45.466 mA/cm², respectively. Subsequently, Jsc noticeably drops until it reaches 45.317892 mA/cm² at 1 × 10²⁰ cm⁻³. This occurs because higher carrier concentrations within the absorber layer lead to increased carrier recombination rates [69]. In contrast, Voc shows a progressive improvement across the doping range. Specifically, it rises from 0.6639 to 0.7152 V as the doping density increases from 1 × 10¹⁵ to 1 × 10¹⁷ cm⁻³. With further increases in doping to 1 × 10¹⁸ and 1 × 10¹⁹ cm⁻³, Voc continues to improve, reaching 0.7603 and 0.7951 V, respectively. At 1 × 10²⁰ cm⁻³, Voc significantly increases to 0.8417 V. The primary reason for the improvement in Voc is the reduction in the diode ideality factor and the increase in the built-in potential with an increase the doping density of the absorber layer, resulting in the increase in the Voc [70]. In addition, the increase in Voc may also be attributed to the decrease in the reverse saturation current, which results from the higher acceptor dopant concentration [31].

As shown in Figure 2d, the responsivity remains nearly constant at approximately 0.85 A/W for doping concentrations up to 10¹⁷ cm⁻³. However, it gradually decreases to around 0.83 A/W as the absorber doping level increases to 10²⁰ cm⁻³. This decline is attributed to enhanced carrier recombination within the increasing the absorber layer doping level [64]. The maximum responsivity of 0.85 A/W is observed at the doping concentration of 10¹⁸ cm⁻³ at a wavelength of 1100 nm. In contrast, detectivity shows a consistent upward pattern with increasing doping, improving significantly from 0.84 × 10¹⁴ to 25.65 × 10¹⁴ Jones across the studied doping range. This enhancement is primarily due to the suppression of dark current and improved noise-limited sensitivity at higher doping levels. The increase in detectivity is closely related to the reduction in dark current, which is inversely proportional to Voc, as indicated by Equation (11). As Voc increases with doping, as previously observed, detectivity also improves [68]. This behavior is consistent with prior reports on chalcogenide-based photodetectors, such as TiS₃ devices [64]. While extremely high doping levels can enhance both responsivity and detectivity, they present fabrication challenges and introduce trade-offs that can degrade overall device performance. Therefore, a doping concentration of 1 × 10¹⁷ cm⁻³ is identified as an optimal and practically feasible choice. At this level, the device demonstrates strong performance, including a Jsc of 45.88 mA/cm², a Voc of 0.7152 V, a responsivity of 0.85 A/W, and a detectivity of 2.26 × 10¹⁴ Jones, without compromising manufacturability or long-term reliability.

Figure 2e illustrates the variation in Jsc and Voc of the SrHfSe₃-based photodetector as the trap density of SrHfSe₃ increases from 1 × 10¹² to 1 × 10¹⁷ cm⁻³, while maintaining a constant device thickness and doping level of 1 μm and 1 × 10¹⁷ cm⁻³, respectively. Both Jsc and Voc exhibit a consistent descending behavior with increasing trap concentration. At lower defect levels from 10¹² to 10¹³ cm⁻³, Jsc remains nearly unchanged, decreasing only slightly from 45.96 to 45.95 mA/cm², while Voc shows a minor reduction from 0.786 to 0.764 V. As the defect density increases to 10¹⁴ cm⁻³, Jsc declines modestly to 45.88 mA/cm², accompanied by a more notable drop in Voc to 0.715 V. With further increases to 10¹⁵ and 10¹⁶ cm⁻³, Jsc decreases to 45.23 mA/cm² and 41.83 mA/cm², respectively, while Voc falls to 0.656 V and 0.595 V. At the highest defect density of 10¹⁷ cm⁻³, the device shows a sharp deterioration, with Jsc and Voc dropping to 36.99 mA/cm² and 0.524 V, respectively.

Figure 2f displays the corresponding degradation in photoresponsivity and specific detectivity as the trap density increases across the same range. The device initially exhibits excellent performance at 1 × 10¹² cm⁻³, with a responsivity of 0.85 A/W and detectivity of 9.0 × 10¹⁴ Jones, indicating efficient carrier collection and low noise levels. However, as the trap density rises, these values progressively decline. At 1 × 10¹⁷ cm⁻³, the responsivity reduces significantly to 0.53 A/W, while detectivity drops sharply to 0.0389 × 10¹⁴ Jones. This performance degradation across all parameters including Jsc, Voc, responsivity, and detectivity, is primarily attributed to enhanced trap-assisted recombination and increased dark current at higher defect densities, which was also observed in the previous studies [68,71]. To ensure high-efficiency operation, it is therefore essential to limit the trap density to the ideal values. A defect concentration of 1 × 10¹⁴ cm⁻³ is selected as the optimal trade-off point, where the device retains strong performance metrics, including Jsc = 45.88 mA/cm², Voc = 0.7152 V, R = 0.85 A/W, and D = 2.26 × 10¹⁴ Jones.

3.2. The Impact of Window Layer (ZnSe) on the Photodetector Performance

The effects of window layer ZnSe thickness on the photodetector performance were evaluated by varying its thickness from 0.05 µm to 0.25 µm, while fixing the doping and defect densities at 10¹⁸ cm⁻³ and 10¹⁵ cm⁻³, respectively. As shown in Figure 3a, both Jsc and Voc remain steady at 45.88 mA/cm² and 0.7152 V, respectively, as the ZnSe thickness varies. This means that the light transmission and carrier collection are not impeded by the thickness. This behavior occurs due to the saturation in the optical absorption, as the diffusion length exceeds the layer thickness, leading to a reduction in recombination losses [64]. In the meantime, the fluctuation of the ZnSe thickness yielded no discernible changes in the responsivity and detectivity, which remained constant at 0.85 A/W and 2.26 × 10¹⁴ Jones, respectively, as shown in Figure 3b. This validates the performance metrics’ stability under different ZnSe thicknesses, which is in good agreement with the results reported in the literature [68]. Considering the performance uniformity under the ZnSe thickness variation, the thickness of 0.05 µm is chosen for further optimization, since it offers low cost, and enhanced transparency, all while maintaining high photodetector efficiency.

Figure 3c,d illustrate the influence of varying the ZnSe window layer doping density on the photodetector characteristics over the range from 10¹⁵ to 10²⁰ cm⁻³. Figure 3a reveals that Jsc slightly increases with higher doping, rising from 45.88 to 45.89 mA/cm², likely due to improved field-assisted carrier extraction, while Voc remains stable at 0.71 V as doping increases. Like the J-V characteristics, as shown in Figure 3d, the responsivity (red curve) remains unchanged at 0.85 A/W at 1100 nm incident light wavelength, with increasing doping levels. In contrast, the detectivity witnesses a slight reduction from 2.27 to 2.22 × 10¹⁴ Jones, likely due to increased dark current at higher doping concentrations. However, it remains relatively constant for doping concentrations below 10¹⁶ cm⁻³ and above 10¹⁸ cm⁻³. A moderate ZnSe doping density around 10¹⁸ cm⁻³ appears to strike a favorable balance between carrier collection and noise suppression, making it optimal for photodetector performance. At this doping level, the detectivity reaches 2.26 × 10¹⁴ Jones at a wavelength of 1100 nm.

The impact of defect density in the ZnSe window layer on the performance of the SrHfSe₃-based photodetector is investigated by varying the defect density from 10¹² to 10¹⁷ cm⁻³ while the thickness and the doping of ZnSe are fixed at the optimum level of 0.05 µm and 10¹⁸ cm⁻³, respectively. As shown in Figure 3e, Jsc remained independent of the defect density of the ZnSe, holding a value at 45.88 mA/cm². Equally, the Voc maintains a stable value at 0.7152 V regardless of the defect concentration, reflecting stable junction quality and efficient carrier separation. Similarly, the responsivity and detectivity, as shown in Figure 3f, stay entirely unaffected by changes in the ZnSe defect density. The responsivity is constant at 0.85 A/W, while the specific detectivity holds stable at 2.26 × 10¹⁴ Jones. A ZnSe defect density of 10¹⁵ cm⁻³ appears to be an optimal value for achieving a favorable balance between carrier collection and recombination.

3.3. The Impact of the BSF Layer (AgCuS) on the Photodetector Performance

This section describes the influence of the BSF layer on the overall photodetector’s performance. Figure 4a,b illustrate the influence of AgCuS BSF layer thickness on photodetector characteristics as it varies from 0.05 µm to 0.30 µm, while other parameters are kept constant. As shown in Figure 4a, the J–V characteristics are nearly identical across the AgCuS thickness range, and the Voc remains constant at 0.7152 V. However, the Jsc exhibits a minor but consistent increase from 45.877 to 45.889 mA/cm². This occurs since the carrier diffusion length exceeds the BSF thickness, preventing significant recombination losses [41]. Figure 4b further confirms a stable pattern in both responsivity at 0.85 A/W and detectivity at 2.6 × 10¹⁴ Jones, despite fluctuations in BSF thickness. The reason for such an unaffected performance is that the optical absorption reaches its saturation level, showing minimal improvement with further increases [68]. As the increase in Jsc becomes marginal beyond 0.20 µm and no further gains are observed in R and D, a BSF thickness of 0.20 µm is selected as the optimal value. It offers a balanced configuration with slight performance improvement without introducing unnecessary material usage or fabrication complexity.

Figure 4c,d assess the impact of doping concentration in the AgCuS BSF layer on the photodetector’s performance, with the BSF thickness fixed at 0.2 µm and the defect density maintained at 10¹⁵ cm⁻³ at 1100 nm light wavelength. As the doping density increases from 10¹⁵ to 10²⁰ cm⁻³, Jsc rises from 45.674 to 45.911 mA/cm², attributed to a stronger built-in electric field that enhances charge carrier separation and suppresses recombination at the back contact. Similarly, a slight improvement in Voc, from 0.7136 V to 0.7194 V, is observed. Likewise, the responsivity experiences a slight increase from 0.84 to 0.85 A/W, while the detectivity shows a more pronounced enhancement from 2.19 × 10¹⁴ to 2.45 × 10¹⁴ Jones at 10²⁰ cm⁻³. At 1100 nm and a doping level of 10¹⁹ cm⁻³, an optimum detectivity of 2.26 × 10¹⁴ Jones, where the device can work without introducing resistive or recombination losses, and it can be practically achievable using standard doping methods. This pattern of responsivity and detectivity variation with increasing BSF layer doping concentration agrees with previously reported work [64]. At this doping level, the device exhibits Jsc, Voc, responsivity, and detectivity values of 45.88 mA/cm², 0.7152 V, 0.85 A/W, and 2.26 × 10¹⁴ Jones, respectively.

Figure 4e,f shows the influence of defect density in the AgCuS BSF layer on the photodetector’s performance, keeping the BSF thickness at 0.2 µm and the doping concentration at 10¹⁹ cm⁻³. As shown in Figure 4e, the J-V characteristics remain almost perpetual as the defect density switches from 10¹² to 10¹⁶ cm⁻³. The Voc holds a consistent value of 0.7152 V, and the Jsc holds steady at around 45.88 mA/cm², indicating minimal impact of defect-related recombination within this range. Similarly, Figure 4f shows that both responsivity 0.85 A/W and detectivity 2.26 × 10¹⁴ Jones stay stable across all defect levels, further suggesting that carrier transport and collection in the AgCuS BSF layer are largely unaffected by moderate defect fluctuations. These findings are in strong agreement with prior reports [68]. As the device exhibits similar tolerance and performance robustness across a wide range of defect densities in the AgCuS BSF layer, a defect density of 10¹⁵ cm⁻³ can be considered a practical and fabrication-friendly choice. This level does not compromise efficiency and provides a solid foundation for further optimization.

3.4. The Impact of the Interfacial Defects on the Photodetector Performance

The density of interfacial defects plays a crucial role in determining the performance of heterojunction photodetectors, especially when there is a mismatch in lattice structure or energy levels between the absorber and contact layers. Therefore, a reasonable number of interface defects needs to be considered and analyzed as it impacts photodetector performance. Here, we explored how varying the interfacial defect density from 1 × 10⁸ to 1 × 10¹⁶ cm⁻³ affects the behavior of SrHfSe₃-based photodetectors at the two different photodetector interfaces, including ZnSe/SrHfSe₃ and SrHfSe₃/AgCuS.

Figure 5a,b show the effect of variation in interfacial defect density at the ZnSe/SrHfSe₃ junction on the photodetector performance. As shown in Figure 5a, the Jsc remains stable at 45.88 mA/cm² up to 10¹³ cm⁻³, while Voc decreases significantly from 0.7152 V to 0.4726 V as the defect density increases from 10⁸ to 10¹⁶ cm⁻³. This behavior results from enhanced carrier recombination and increased dark current due to the higher density of interfacial defects, both of which contribute to the reduction in Voc [51]. In terms of responsivity, its value remains constant at 0.85 A/W at 1100 nm across all defect levels. Unlike the responsivity, the detectivity holds steady at 2.26 × 10¹⁴ Jones up to 10¹⁰ cm⁻³ and then begins to noticeably degrade beyond 10¹² cm⁻³, dropping to approximately 0.0207 × 10¹⁴ Jones at 10¹⁶ cm⁻³. This reduction is primarily attributed to the increase in dark current at higher interfacial defect densities. Thus, maintaining an interfacial defect density below 10¹¹ cm⁻³ is critical for preserving low-noise and high-sensitivity performance at the front interface. Based on these findings, an interfacial defect density of 10¹⁰ cm⁻³ is selected as the optimal value for the ZnSe/SrHfSe₃ interface for further optimization.

Figure 5c,d display the behavior of the photodetector performance with changes in the SrHfSe₃/AgCuS interface, as the defect density increases from 10⁸ to 10¹⁶ cm⁻³. As illustrated in Figure 5c, the Jsc decreases from 45.91 mA/cm² at 10⁸ cm⁻³ to 41.24 mA/cm² at 10¹⁶ cm⁻³, reflecting a gradual reduction in carrier extraction efficiency. Similarly, the Voc drops from 0.721 V to 0.5969 V. The fall in both Jsc and Voc is attributed to the increasing defect density, which enhances charge carrier recombination at the SrHfSe₃/AgCuS interface [72]. Unlike the ZnSe/SrHfSe₃ interface, both responsivity and detectivity decrease adversely with increasing SrHfSe₃/AgCuS interface defect densities. The responsivity remains nearly constant at 0.85 A/W up to 10¹⁰ cm⁻³ at 1100 nm, as depicted in Figure 5d. However, beyond 10¹¹ cm⁻³, it significantly drops to 0.65 A/W at 10¹⁶ cm⁻³. Meanwhile, the detectivity remains stable at 2.26 × 10¹⁴ Jones up to 10¹⁰ cm⁻³, then exhibits a sharp decline, reaching 0.187 × 10¹⁴ Jones at 10¹⁶ cm⁻³. Compared with the ZnSe/SrHfSe₃ interface, the SrHfSe₃/AgCuS is more vulnerable to interfacial degradation, and to ensure high photodetector efficiency and low-noise operation, the SrHfSe₃/AgCuS interface defect density must be maintained below 10¹¹ cm⁻³. Consequently, a defect density of 10¹⁰ cm⁻³ is selected for SrHfSe₃/AgCuS interface for further calculations.

3.5. The Impact of the Working Temperature on the Photodetector Performance

So far, all simulated calculations have been conducted at room temperature (300 K). However, to ensure stable and reliable photodetector performance, particularly for near-infrared applications operating in high-temperature environments, robust thermal management is critically important. Here, the simulated characteristics of the SrHfSe₃-based photodetector across a temperature range of 300 K to 500 K are visualized in Figure 6. As shown in Figure 6a, the Jsc remains nearly constant, increasing only slightly from 45.88 to 45.96 mA/cm², while the Voc gradually decreasing with rising temperature, dropping from 0.7152 V at 300 K to 0.4407 V at 500 K. This degradation is primarily caused by the amelioration in intrinsic carrier concentration and enhanced nonradiative recombination at higher temperatures, both of which contribute to the rise in the reverse saturation current and the subsequent drop in Voc [73].

Figure 6b shows the impact of altering the working temperature within a range from 300 K to 500 K on both responsivity and detectivity. The responsivity remains remarkably stable, obeying the Jsc pattern, increasing only slightly from 0.850 A/W at 300 K to 0.852 A/W at 500 K at 1100 nm. This signifies that the material maintains efficient photocarrier generation across variable temperatures. However, the D* exhibits a steep decline, dropping from its peak value of 2.26 × 10¹⁴ Jones at 300 K to just 1.12 × 10¹² Jones at 500 K. The rationale behind the sharp reduction in detectivity is mainly due to the elevated dark current at higher temperatures [71]. This increase in dark current may result from temperature-induced non-uniformities in device operation. Elevated temperatures can narrow the semiconductor bandgap, enhancing thermal carrier generation and thus increasing the dark current. This rise in dark current leads to a reduction in Voc, which is logarithmically and inversely related to dark current, as demonstrated by Equation (11). Working temperature variations may further cause spatial fluctuations in Voc, contributing to performance instability [31]. This behavior is in agreement with previous reports, which also observed stable responsivity and a decrease in detectivity with increasing operating temperature [51].

Figure 6c shows the impact of device working temperature on the dark current density. As the temperature increases from 300 K to 500 K, the dark current density undergoes a significant increase from 4.43 × 10⁻¹¹ to 1.81 × 10⁻⁶ mA/cm². This severe increase suggests that more charge carriers are being thermally generated at higher temperatures, leading to increased leakage current in the device [41,62]. Figure 6d displays the on/off ratio as a function of the working temperature variation. A dramatic drop in the on/off ratio is observed, where the ratio drops from 1.04 × 10¹² at 300 K to 2.55 × 10⁷ at 500 K. This decline is mainly due to the increase in dark current, which reduces the contrast between the device’s light and dark states. As a result, the photodetector’s performance becomes less efficient at higher temperatures, indicating that its operation is strongly affected by thermal effects.

3.6. The Impact of the SrHfSe₃ Bandgap Tuning on the Photodetector Performance

To gain more qualitative insights, we investigate how Sb-doping affects the band gap of the SrHfSe₃ layer and, in turn, the device’s electrical characteristics. Pristine SrHfSe₃ has a direct band gap of approximately 1.00–1.02 eV. When doped with Sb, the band gap increases to 1.06 eV at 0.5 mol % and to 1.15 eV at 1.0 mol %, as confirmed by both density functional theory (DFT) calculations and experimental data [39]. The direct band gap makes SrHfSe₃ suitable for near-infrared photodetectors, and Sb doping enhances its response across a broader wavelength range. This spectral tunability highlights Sb-doping as a promising strategy for developing photodetectors with broad spectral sensitivity.

The photodetector characteristics with the alteration of SrHfSe₃ bandgap from 1.02 eV to 1.15 eV are shown in Figure 7. As shown in Figure 7a, the J–V characteristics reflect a clear dependence of Jsc and Voc on bandgap energy. At 1.02 eV, the device achieves the highest Jsc of around 45.88 mA/cm² due to stronger near-infrared (NIR) light absorption. However, the Voc is lower in this case, likely due to increased carrier recombination or a weaker built-in electric field. Increasing the bandgap to 1.06 eV leads to a modest drop in Jsc (44.39 mA/cm²), but Voc significantly improves due to enhanced carrier separation and suppressed recombination. At 1.15 eV, photocurrent drops further to 39.92 mA/cm² despite a higher Voc, indicating reduced photon absorption as the cutoff shifts away from the 1100 nm operating wavelength.

Figure 7b presents the impact of bandgap variation on responsivity and specific detectivity. The highest responsivity of 0.85 A/W occurs at 1.02 eV, in line with the stronger photocurrent. At 1.06 eV, responsivity slightly decreases to 0.81 A/W, but detectivity peaks at 3.97 × 10¹⁴ Jones due to an optimal balance between signal and dark noise. The noticeable drop in responsivity and detectivity at a bandgap of 1.15 eV is closely linked to the device’s quantum efficiency at 1100 nm. As shown in Figure 7a, the EQE plot, the 1.15 eV structure exhibits a much earlier cutoff compared to the 1.02 eV and 1.06 eV cases, which maintain high quantum efficiency near this wavelength. Since fewer photons are absorbed at 1100 nm when the bandgap is too large, the resulting photocurrent drops significantly. This loss in signal directly reduces responsivity and ultimately degrades detectivity. In contrast, the 1.06 eV configuration sustains strong EQE at 1100 nm, explaining its peak detectivity. However, as depicted in Figure 7c, the EQE remains relatively stable across the wavelength range, showing only minor fluctuations at longer wavelengths [74].

Figure 7d illustrates the impact of bandgap energy on the dark current density and on/off ratio of the proposed photodetector. As the bandgap increases from 1.02 eV to 1.15 eV, a clear inverse relationship is observed between dark current and the on/off ratio. At 1.02 eV, the device exhibits a relatively high dark current density of 4.43 × 10⁻¹¹ mA/cm², accompanied by a modest on/off ratio of 1.04 × 10¹². This elevated dark current can be attributed to the narrower energy gap, which facilitates thermal generation of carriers and leakage pathways under dark conditions. As the bandgap widens to 1.06 eV, the dark current significantly decreases to 1.72 × 10⁻¹¹ mA/cm², while the on/off ratio improves to 2.57 × 10¹², indicating enhanced discrimination between photo and dark currents. Notably, at 1.15 eV, the dark current reaches its lowest value of 2.71 × 10⁻¹² mA/cm², while the on/off ratio peaks at 1.47 × 10¹³. This substantial improvement reflects the suppressive effect of a wider bandgap on thermally induced carrier excitation, thereby enhancing signal clarity and noise immunity. Overall, these results highlight that increasing the bandgap of the absorber layer effectively suppresses dark current while significantly enhancing the photodetector’s switching performance, making it highly suitable for high-sensitivity and low-noise optoelectronic applications.

Overall, the results highlight the importance of bandgap engineering in optimizing device performance. From an application perspective, a bandgap of around 1.02 eV is best suited for SrHfSe₃ photodetectors targeting near-infrared operation, while 1.06 eV offers optimal performance for self-powered or low-light conditions. Therefore, a 1.02 eV bandgap is selected for further calculations.

3.7. The Impact of the Change in the Input Power on the Phototodetector Performance

So far, all simulations were primarily carried out under a standard illumination intensity of 100 mW/cm². However, to rigorously assess the light-intensity-dependent behavior of the device and highlight the critical limitations of the SrHfSe₃-based photodetector, the variation in Jsc, Voc, responsivity, and detectivity with increasing illumination was further explored, as illustrated in Figure 6. As displayed in Figure 6a, it is revealed that as the light intensity rises from 5 mW/cm² to 100 mW/cm², Jsc exhibits a substantial rise from 2.29 mA/cm² to 45.89 mA/cm². This enhancement is primarily due to the elevated photon flux, which increases the generation rate of charge carriers. The relationship between Jsc and incident power (P) follows a power-law behavior, given by Jsc ∝ P^α, where α typically ranges from 0.9 to 1.0 in ideal photodiodes, indicating a near-linear dependence [75]. Similarly, Voc evolves from 0.6275 V to 0.7152 V as the illumination intensity increases. This improvement is attributed to the enhancement of the built-in electric field and the suppression of carrier recombination, leading to more efficient charge separation. The overall behavior is consistent with classical photodiode operation, where Jsc scales nearly linearly while Voc exhibits a logarithmic dependence on input power, in agreement with a previously reported simulation-based study [64].

Figure 8b illustrates the behaviors of responsivity (R) and detectivity (D*) with illumination power. The responsivity remains almost constant at approximately 0.85 A/W across the optical power range. On the other hand, detectivity gradually optimized from 1.86 × 10¹⁴ to 2.26 × 10¹⁴ Jones as the input power increases. This likely results from a reduced signal-to-noise ratio at higher optical power, combined with saturation effects and dynamic range limitations of the photodetector. These findings align well with the previously reported work [64].

3.8. The Impact of the Series, Shunt Resistance, and Electron Affinity on the SrHfSe₃ Photodetector Performance

The influence of series and shunt resistances on Voc and Jsc is depicted in Figure 9a,b, respectively. The series resistance originates from the body and the contact, while the shunt resistance typically stems from fabrication flaws and reverse leakage current [76]. Consequently, controlling both parameters, especially the series resistance, is crucial for enhancing photodetector performance. Figure 9a shows the fluctuation of Voc and Jsc with increasing series resistance from 0 to 20 Ω. As the resistance increases, Voc remains nearly constant at nearly 0.71 V. On the contrary, Jsc stays perpetual at 45.88 mA/cm² up to 4 Ω. Subsequently, it drops sharply at 16 Ω to 40.71 mA/cm² and further decreases to 33.80 mA/cm² at 20 Ω. This reduction in Jsc is likely due to increased resistive losses, which impede carrier collection and thereby suppress the short-circuit current. This can be further explained by the following Equation (12) [62]:

$$J_{\mathrm{p}}=J_L+J_0\left(e^{{\displaystyle \frac{V_{oc}q}{nkT}}}-1\right)-{\displaystyle \frac{V_{oc}+J_p{R}_S}{R_{Sh}}}$$

(12)

where the $J_{\mathrm{p}}$, $R_{Sh}$, $R_S$, $J_L$, and $J_0$ represent the photocurrent, shunt resistance, series resistance, light-generated current, and reverse saturation current, respectively. As can be observed from the equation, an increase in series resistance leads to a reduction in the photocurrent.

Apart from the series resistance, the impact of shunt resistance on Voc and Jsc is also studied, as shown in Figure 9b. It is observed that both Voc and Jsc exhibit stable trends as the shunt resistance increases from 1 to 20 kΩ, maintaining values around 0.7152 V and 45.88 mA/cm², respectively. Figure 9c,d display the Voc and Jsc dependent on the electron affinity of the SrHfSe₃-based photodetector, respectively. As the electron affinity increases from 3.9 eV to 4.4 eV, the Voc gradually drops from 0.7152 V to 0.6035 V. The reason for this decline may likely result from less favorable band alignments between the photodetector layers that reduce the built-in potential and increase recombination losses. Concurrently, the Jsc shows a slight variation with the increase in electron affinity. It marginally decreases from 44.89 to 44.4 mA/cm² as the electron affinity increases from 3.9 to 4.0 eV, and then rises, reaching a peak value of 45.90 mA/cm² at 4.3 eV. The value of 3.9 eV appears to be the optimal point at which the detector exhibits both high Voc and Jsc, and can thus be used for further calculations.

3.9. The Built-In Potential of the Photodetector

The built-in potential (φ_bi) across the device heterojunctions plays a critical role in enhancing the internal electric field and promoting efficient charge separation and extraction. It can be estimated from the intercept of the Mott–Schottky (1/C)²–V plot as illustrated by Equation (13) [55].

$${\displaystyle \frac{1}{C^2}}={\displaystyle \frac{c\left(\frac{KT}{q}-{\varnothing }_{bi}-V\right)}{q{A}^2{\epsilon }_0{\epsilon }_{{SrHfSe}_3}{N}_D}}$$

(13)

Here, ${\epsilon }_{{\mathrm{S}\mathrm{r}\mathrm{H}\mathrm{f}\mathrm{S}\mathrm{e}}_3}$ and ε₀ denote the permittivity of SrHfSe₃ and vacuum, respectively. The C denotes the capacitance at the junction, and V is the applied voltage. The $N_D$ represents the donor concentration, q is the elementary charge (1.6 × 10⁻¹⁹ C), and A corresponds to the diode’s active area.

Figure 10a,b presents the plots of ${\displaystyle \frac{1}{C^2}}$ versus voltage for the heterojunctions ZnSe/SrHfSe₃ and SrHfSe₃/AgCuS, respectively. The red curves in both figures represent simulation results obtained using the SCAPS-1D (3.3.12) software, while the blue curves correspond to the fitted data derived using OriginPro 2025 software to extract the built-in potential. To determine the effective built-in potential of the entire system, the built-in voltages from each junction were individually extracted from the interceptions of the linear portions of the fitted curves on the voltage axis. These values are found to be 0.82 V for the ZnSe/SrHfSe₃ junction and 0.81 V for the SrHfSe₃/AgCuS junction. Summing these values yield a total built-in potential of approximately 1.63 V for the complete device architecture.

The relatively high total potential can be attributed to the well-aligned energy bands at the ZnSe/SrHfSe₃ and SrHfSe₃/AgCuS interfaces, which facilitate efficient charge separation and transport across the heterojunctions.

3.10. The Optimized Design with and Without the BSF Layer

Figure 11a shows the impact of adding a BSF layer between SrHfSe₃ and the back contact Cu on the photodetector performance. It is evident that the installation of an AgCuS (BSF) layer into the photodetector structure substantially boosts the overall performance. As shown in Figure 11a, the device yields a Jsc of 41.31 mA/cm² and a Voc of 0.5975 V without the AgCuS layer. Upon addition of the AgCuS BSF, the Voc grows to 0.7152 V, and Jsc jumps to 45.88 mA/cm². With respect to the EQE, adding the BSF layer into the photodetector design enhanced the EQE in the 300 to 1200 nm range, retaining almost 100% up to 830 nm as depicted in Figure 11b. Beyond this wavelength, it gradually declined, approaching nearly zero over 1210 nm.

The wavelength-dependent spectral responsivity and detectivity are depicted individually in Figure 11a,d. As shown in Figure 11b, the spectral responsivity rises with the inclusion of the BSF layer, attaining its highest value of around 0.85 A/W at 1100 nm. Conversely, without a BSF layer, the summit responsivity is shifted to around 1040 nm, where it attains a lower maximum of 0.675 A/W.After this peak, the response drops with growing wavelength for both configurations with and without the BSF layer, ultimately approaching zero beyond 1210 nm.

In addition, as shown in Figure 11d, the integration of the BSF layer leads to an increase in the detectivity, reaching a dominant peak value of 2.26 × 10¹⁴ Jones at 1100 nm. This significantly surpasses the detectivity of the configuration without the BSF layer, which peaks at around 1.94 × 10¹³ Jones at 1040 nm. The underlying physics behind this enhancement is that adding the AgCuS BSF layer improves detector performance from 400 to 1100 nm by enhancing light absorption and increasing Jsc, which is directly linked to responsivity. Adding a Back Surface Field (BSF) layer forms a dual junction, typically an n/p or p/p⁺ structure, that enhances the built-in potential within the device. This improved junction configuration helps reduce dark current collection, thereby increasing the Voc. The increase in Voc is associated with a reduction in dark current, as shown in Equation (14) [31].

$$V_o={\displaystyle \frac{k_{B}T}{q}}ln{\displaystyle \frac{J_{sc}}{J_0}}$$

(14)

In addition, the BSF layer reduces recombination losses by reflecting incident photons, which further improves Voc and results in better detectivity, as demonstrated in Equation (10) [67,68].

As shown in

Table 4. Comparison of photodetector performance for the structure with and without the BSF layer.

Structure	Responsivity (A/W)	Detectivity (Jones)	Wavelength (nm)
ZnSe/SrHfSe3	0.675	1.94 × 1013	1040
ZnSe/SrHfSe3/ AgCuS	0.85	2.26 × 1014	1100

, the incorporation of the AgCuS layer significantly enhances the overall efficiency of the device. Notably, both the spectral responsivity and detectivity reach their peak values with the inclusion of the BSF layer, observed at different wavelengths of 1100 nm and 1040 nm, respectively.

3.11. Performance Comparison

Table 5 presents a comprehensive comparison of the photoresponse characteristics reported in various experimental and simulation-based photodetector studies, including the results from our current work. Notably, various reported work was conducted using the SCAPS-1D simulation software, which further validates its reliability and versatility in modeling diverse photodetector structures. This comparison highlights SCAPS-1D as an efficient and accessible tool for simulating a wide range of device architectures, including Schottky, pn, and pin configurations, such as the one implemented in our design. It is worth mentioning that the reported photoresponsivity and photosensitivity values correspond to their peak values, which are wavelength-dependent and influenced by various device factors.

The table also includes experimental devices such as the Si:S photodetector, which demonstrates a trade-off between scalability and performance. Fabricated via ion implantation and annealing, the device offers CMOS compatibility but exhibits limited responsivity (0.1073 A/W). The table also includes compounds containing elements such as Cd and Pb, which are well known for their toxicity and raise significant environmental and health concerns. Regardless of their performance, materials containing Ge, In, and Te are associated with high costs due to their limited availability in the Earth’s crust. The responsivity reported in these works is relatively low compared to our device performance, except in the cases of BeSiP₂ and TiSe₂, MoS_2, Ag₃CuS_2, and Gr/GaAs photodetectors. In addition, the performance of these devices is limited to wavelengths around 1000 nm and shows poor response beyond 1100 nm due to their relatively wide bandgap.

Apart from exhibiting high performance, the photodetector is primarily composed of SrHfSe₃ as the main absorber material. This compound consists of strontium (Sr), hafnium (Hf), and selenium (Se), elements that are relatively earth-abundant and less toxic compared to those used in conventional semiconductors. Their availability and environmental compatibility make SrHfSe₃ a promising choice for sustainable optoelectronic applications [39,77,78].

Table 5. Comparison of the proposed photodetector with the literature work.

Structure	Types of Work	λ(nm)	Software Used /Method	Responsivity (A/W)	Detectivity (Jones)	Reference
Si:S	Experimental	1310	Ion implantation + rapid thermal annealing	0.1073	-	[79]
p-WSe2/n-Ge	Experimental	1550	Mechanical exfoliation + Ion implantation	1.3	2.5 × 1010	[80]
InGaAs/InAs	Experimental	1550 2000	CVD	0.6 at 1550 nm 15 at 2000 nm	2.4 × 1014 3.8 × 1010	[11]
p-n Ge	Simulation	1550	Lumerical Change FDTD	0.43	-	[47]
Graphene/GaAs	Simulation	725	COMSOL Multiphysics	0.514	1.16 × 1011	[81]
InGaAs/InAs/InSb/InP HEMT	Simulation	900–1700	TCAD Silvaco	15.75	4.0384 × 1010	[49]
p-MoS2	Simulation	700	SCAPS-1D	0.37	3.27 × 1014	[82]
n-CdS/p-Cu2ZnGeSe4/p+-ZnTe	Simulation	780	SCAPS-1D	0.58	8.28 × 1017	[46]
PbS/TiS3	Simulation	780	SCAPD 1D	0.36	3.9 × 1013	[4]
n-ZnSe/p-TiSe2/p+-WSe2	Simulation	920	SCAPS 1D	0.670	12.90 × 1014	[41]
n-In2S3/p-BeSiP2/p+-MoS2	Simulation	860	SCAPS 1D	0.64	3.63 × 1016	[67]
n-WS2/p-Ag3CuS2/p+-BaSi2	Simulation	1065	SCPDS 1D	0.790	4.73 × 1014	[63]
n-ZnSe/p-SrHfSe3/p+-AgCuS	Simulation	1100	SCAPS-1D	0.850	2.26 × 1014	This work

Table 5. Comparison of the proposed photodetector with the literature work.

Structure	Types of Work	λ(nm)	Software Used /Method	Responsivity (A/W)	Detectivity (Jones)	Reference
Si:S	Experimental	1310	Ion implantation + rapid thermal annealing	0.1073	-	[79]
p-WSe2/n-Ge	Experimental	1550	Mechanical exfoliation + Ion implantation	1.3	2.5 × 1010	[80]
InGaAs/InAs	Experimental	1550 2000	CVD	0.6 at 1550 nm 15 at 2000 nm	2.4 × 1014 3.8 × 1010	[11]
p-n Ge	Simulation	1550	Lumerical Change FDTD	0.43	-	[47]
Graphene/GaAs	Simulation	725	COMSOL Multiphysics	0.514	1.16 × 1011	[81]
InGaAs/InAs/InSb/InP HEMT	Simulation	900–1700	TCAD Silvaco	15.75	4.0384 × 1010	[49]
p-MoS2	Simulation	700	SCAPS-1D	0.37	3.27 × 1014	[82]
n-CdS/p-Cu2ZnGeSe4/p+-ZnTe	Simulation	780	SCAPS-1D	0.58	8.28 × 1017	[46]
PbS/TiS3	Simulation	780	SCAPD 1D	0.36	3.9 × 1013	[4]
n-ZnSe/p-TiSe2/p+-WSe2	Simulation	920	SCAPS 1D	0.670	12.90 × 1014	[41]
n-In2S3/p-BeSiP2/p+-MoS2	Simulation	860	SCAPS 1D	0.64	3.63 × 1016	[67]
n-WS2/p-Ag3CuS2/p+-BaSi2	Simulation	1065	SCPDS 1D	0.790	4.73 × 1014	[63]
n-ZnSe/p-SrHfSe3/p+-AgCuS	Simulation	1100	SCAPS-1D	0.850	2.26 × 1014	This work

4. Conclusions

This study has designed and simulated a photodetector based on n-ZnSe/p-SrHfSe₃/p⁺-AgCuS using SCAPS-1D. Various parameters were thoroughly evaluated and optimized, including the energy bands, J–V curves, photoresponsivity, and specific detectivity. To optimize the photodetector performance, many factors such as the thickness, defects, and the doping density for each layer of the device were precisely optimized. In addition, the interface defects, bandgap of the main absorber, working temperature, and input light intensity were also evaluated to optimize the proposed design further. The optimized parameters consist of a 1 µm thick absorber layer, a 0.05 µm window layer, and a 0.200 µm thick BSF layer. The doping densities for the absorber, window layer, and BSF layer are found to be 10¹⁷ cm⁻³, 10¹⁸ cm⁻³, and 10¹⁹ cm⁻³, respectively. The optimized device displayed a photocurrent of 45.88 mA/cm² and a voltage of 0.7512 V at room temperature (300 K). Under a light wavelength of 1100 nm, the simulated photodetector shows off a photoresponsivity of 0.85 A/W and a detectivity of 2.26 × 10¹⁴ Jones. The optimized interface defect level was found to be 10¹⁰ cm⁻² for both n-n-ZnSe/p-SrHfSe₃ and n-ZnSe/p-SrHfSe₃/p⁺-AgCuS interfaces. In addition, the device displayed a better spectral photoresponse in the 700–1200 nm range, with the highest photoresponse attained at 1100 nm, indicating suitability for near-infrared detection. Finally, compared to previous strategies reported in the literature, this device demonstrates significant promise for research areas that require high-performance and stable photodetectors. Its high responsivity and detectivity significantly reinforce its potential for next-generation photodetector applications.