Controlling the Concentration of Copper Sulfide Doped with Silver Metal Nanoparticles as a Mechanism to Improve Photon Harvesting in Polymer Solar Cells

Abstract

The development of thin-film organic solar cells (TFOSCs) is pivotal for advancing sustainable energy technologies because of their potential for low-cost, lightweight, and flexible photovoltaic applications. In this study, silver-doped copper sulfide (CuS/Ag) metal nanoparticles (MNPs) were successfully synthesized via a wet chemical method. These CuS/Ag MNPs were incorporated at varying concentrations into a poly(3-hexylthiophene) (P3HT) and [6,6]-phenyl-C61-butyric acid methyl ester (PC61BM) blend, serving as the active layer to enhance the photovoltaic performance of the TFOSCs. The fabricated TFOSC devices were systematically evaluated based on the optical, electrical, and morphological characteristics of the active layer. By varying the concentration of CuS/Ag MNPs, the influence of nanoparticle doping on photocurrent generation was investigated. The device incorporating 1% CuS/Ag MNPs exhibited the highest power conversion efficiency (PCE) of 5.28%, significantly outperforming the pristine reference device, which achieved a PCE of 2.53%. This enhancement is attributed to the localized surface plasmon resonance (LSPR), which augments charge transport and increases optical absorption. The CuS/Ag MNPs were characterized using ultraviolet–visible (UV-Vis) absorption spectroscopy, transmission electron microscopy (TEM), scanning electron microscopy (SEM), and energy-dispersive dispersion (EDX) analysis. These findings underscore the potential of CuS/Ag MNPs in revolutionizing TFOSCs, paving the way for more efficient and sustainable solar energy solutions.

1. Introduction

Thin-film organic solar cells (TFOSCs) are emerging as a promising next-generation sustainable energy source, offering a viable alternative to conventional inorganic photovoltaic technologies. They have attracted considerable interest in both academic and industrial sectors due to their key advantages, including solution processability, low production cost, lightweight design, mechanical flexibility, and suitability for portable and flexible solar energy applications [1,2,3,4,5,6]. Their compatibility with roll-to-roll manufacturing processes makes TFOSCs particularly attractive for large-scale production. As global energy demands continue to rise, TFOSC technology has gained significant attention for its potential to provide scalable, cost-effective, and sustainable energy solutions, positioning it as a promising candidate to address future energy challenges. In recent developments, TFOSCs utilizing non-fullerene acceptors (NFAs) have surpassed 19% power conversion efficiency (PCE), which is a significant milestone in the field of organic photovoltaics [7,8,9]. This significant achievement highlights the continuous improvements in material design, processing techniques, and device architectures within the field. Nevertheless, the conductivity of polymer solar cells can be enhanced to levels comparable to metals through doping, owing to their tunable energy bandgap. Recent advancements in TFOSCs incorporating non-fullerene acceptors (NFAs) have achieved power conversion efficiencies (PCEs) exceeding 19%, marking a significant milestone in the field of organic photovoltaics [7,8,9]. This achievement underscores ongoing progress in material design, processing strategies, and device architecture optimization. However, it can be significantly improved through doping, facilitated by the tunable energy bandgap of organic semiconductors. Such meaningful properties of polymers have not only opened significant potential for their applications in photonic devices but also in electronic devices. For instance, conjugated polymers can achieve hole mobilities up to ${10}^{-2} {\mathrm{c}\mathrm{m}}^2/\mathrm{V}\mathrm{s},$ while electron mobilities are typically lower, around ${10}^{-4} {\mathrm{c}\mathrm{m}}^2/\mathrm{V}\mathrm{s}.$ This disparity leads to imbalanced charge transport, which is a critical challenge in optimizing device performance. This imbalance can be mitigated by incorporating fullerene polymers, which act as electron acceptors. The most efficient TFOSCs are based on a bulk heterojunction (BHJ) architecture, where a blend of donor (D) and acceptor (A) materials self-assembles into interpenetrating yet phase-separated nanoscale domains, thereby facilitating efficient exciton dissociation and charge transport, which are critical for high device performance [10,11,12]. The mixture of poly(3-hexylthiophene) (P3HT) and the fullerene derivative [6,6]-phenyl-C61-butyric acid methyl ester (PC61BM) is among the most widely studied BHJ active layer materials in TFOSCs [13,14,15,16]. These polymers have been widely employed as reference materials in TFOSCs research due to their commercial availability and consistent performance in device fabrication.

Furthermore, the PCE of TFOSCs is primarily governed not only by the efficiency of light absorption but also by the effective collection of charge carriers at the electrodes [2]. Due to the low optical absorption band width and short lifetime of the excitons, the optimum thickness of the active layer in TFOSCs is typically limited to around 200 nm, which further constrains the device’s ability to absorb more incident photons [17,18,19]. For these reasons, increasing the optical path length of light within the device is vital, as it enhances light absorption and contributes to a higher PCE [19]. Effective light trapping can be achieved using structures that are comparable to or smaller in size than the wavelength of the light, combined with materials that interact strongly with light, such as semiconductors and metal nanoparticles (MNPs) [20]. The significant UV–Visible absorption demonstrated by MNPs is attributed to the excitation of localized surface plasmon resonances (LSPRs), which arise from the collective oscillation of conduction electrons at the nanoparticle surface [20]. It has been reported that when the frequency of plasmon oscillation matches that of the incident electromagnetic (EM) radiation, resonance occurs, resulting in enhanced photon absorption facilitated by metal nanoparticles (MNPs) [21]. However, the resonance frequency is influenced by numerous factors, such as the shape, size, nature, and composition of MNPs, as well as the dielectric properties of the surrounding medium [20]. Therefore, incorporating MNPs into the active layer of TFOSCs offers a promising strategy to enhance light absorption, thereby improving photon-induced charge carrier generation through mechanisms such as light scattering and LSPR [22,23,24,25,26,27,28,29]. LSPR is the interaction between the surface electron plasma of metals and the incident EM field, leading to improved optical absorption, exciton dissociation, charge transport, and enhanced device parameters in the photoactive layer.

Interestingly, silver (Ag) is the most prominent MNP used in TFOSC research due to its strong ability to enhance both LSPR and light scattering within the dielectric medium [27]. It exhibits high light scattering efficiency along with a short-range LSPR effect [27]. On the other hand, copper sulfide (CuS), which acts as an ionic compound, has the potential to enhance the conductivity of the polymer medium. It has been reported to improve the performance of TFOSCs via its plasmonic resonance effects when incorporated as MNPs into the hole transport layer (HTL) [18]. In a recent investigation, the highest reported power conversion efficiency (PCE) was 4.51% [18]. In this study, newly silver-doped copper sulfide (CuS/Ag) MNPs were effectively synthesized via a wet chemical processing technique and incorporated at various concentrations into a P3HT/PC61BM blend active layer to enhance photon harvesting. As a result, the highest PCE of 5.28% was achieved at a doping level of 1% by weight, compared to 2.53% for the pristine reference cell. All devices were fabricated and tested under ambient laboratory conditions.

2. Materials and Methods

2.1. Materials

The chemical materials used to synthesize silver-doped copper sulfide (CnS/Ag) metal nanoparticles (MNPs) were purchased from commercial suppliers and used directly, bypassing any further purification processes. Specifically, copper nitrate (Cu(NO₃)₂·3H₂O, >99.5%), silver nitrate (Ag(NO₃)₂, >99.5%), sodium borohydride (NaBH₄, >99.98%), and thiourea (CH₄N₂S) were procured from Merck, Darmstadt, Germany. Moreover, the chemical reagents required for the preparation of solar cells, such as poly(3-hexylthiophene) (P3HT, 95%), [6,6]-phenyl-C61-butyric acid methyl ester (PCBM, 95%), and Poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PEDOT/PSS), were also bought from Merck, Germany.

2.2. Synthesis of CuS/Ag MNPs

The synthesis of CuS/Ag MNPs was carried out using a wet chemical processing method. In this new study, the process started with the preparation of solutions using 0.2 M Cu(NO₃)₂, 0.2 M CH₄N₂S, 0.1 M Ag(NO₃)₂, and 0.3 M NaBH₄. Each compound was dissolved in 50 mL of deionized water in four clean, distinct flasks to form aqueous solutions. The Cu(NO₃)₂, CH₄N₂S, and Ag(NO₃)₂ solutions were sequentially mixed in a 500 mL beaker. The Cu(NO₃)₂ solution was first combined with CH₄N₂S, followed by the dropwise addition of Ag(NO₃)₂ while continuously stirring the mixture to ensure uniformity. This was then followed by the addition of the fourth solution, NaBH₄, which functions as a reducing agent. The mixture solution was continuously stirred for 3 h on a hot plate with a magnetic stirrer, maintaining an average temperature of 40 °C for better miscibility. However, the resulting mixture was repeatedly cleaned and filtered several times by means of deionized water to effectively remove sodium ions. It was then dried in a vacuum oven at 70 °C for 2 h, thereby yielding CuS/Ag MNPs (see Figure 1). Furthermore, the synthesized CuS/Ag MNPs were characterized for their structural, optical, and morphological properties using several spectroscopic and microscopic techniques, including energy-dispersive X-ray (EDX), scanning electron microscopy (SEM), and transmission electron microscopy (TEM).

2.3. Device Fabrication of CuS/Ag MNPs

The TFOSCs, which consisted of various material layers, were fabricated in the following sequence: glass/ITO/PEDOT/PSS/P3HT+PC61BM+CuS+Ag MNPs/LiF/Al. Figure 2 presents a schematic diagram illustrating the many layers that constitute the solar cell. The solar cell fabrication started with the etching of unpatterned indium tin oxide (ITO)-coated glass substrates using a warm acid solution, which was composed of hydrochloric acid (HCl), hydrogen oxide (H₂O), and nitric acid (HNO₃) in a volumetric ratio of 48:48:4. This process selectively removed portions of the ITO layer.

The substrates were cleaned sequentially in an ultrasonication bath using detergent, deionized water, acetone, and isopropanol, each for 10 min. They were dehydrated in an oven at 100 °C for 20 min. A thin layer of PEDOT/PSS was spin-coated onto the partially etched glass substrates at 3500 revolutions per minute (rpm) for 60 s. The coated films were subsequently dried in an oven at 100 °C for 20 min. The active layer solution was prepared by dissolving a 1:1 weight ratio blend of P3HT and PC₆₁BM in chloroform solvent at a 20 mg/mL concentration. Moreover, CuS/Ag MNPs were combined into the mixture solution at the concentrations of 1%, 3%, and 5% by weight, with each concentration prepared in a separate solution. Each solution was stirred using a magnetic stirrer on a hot plate at an average temperature of 40 °C for 3 h to ensure adequate miscibility. The photoactive films were spin-coated at 1200 rpm for 40 s and subsequently dried in a furnace under a nitrogen atmosphere at 100 °C for 5 min. Finally, the samples were moved into a vacuum chamber to deposit an ETL (LiF) and an Al electrode. The two were deposited at a thickness of 0.6 and 80 nm, respectively. The current density–voltage (J-V) data were obtained using a computer-controlled Keithley HP2420 source meter in conjunction with a solar simulator (model SS50AAA) operating under AM1.5 illumination at 100 mW/cm². The optical properties of the photoactive films were studied using UV-Vis spectroscopy.

3. Results and Discussion

3.1. Characterizations of CuS/Ag MNPs

The surface morphology and structural characteristics of the synthesized CuS/Ag MNP powders were examined using TEM and SEM, as shown in Figure 3. The elemental composition of CuS/Ag MNPs was investigated by EDX analysis (see Figure 3). The images are a good source of information about the particle size, shape, and distribution of CuS/Ag MNPs, which can influence the optical and electrical properties. The TEM images, as provided in Figure 3a,b, show the formation of disc and spherical-like structures of various sizes of CuS/Ag MNPs. The size of the MNPs measured from TEM ranges from 10.16 nm to 25.21 nm. The SEM image in Figure 3c clearly shows the uniform distribution of flower-like structures, which are consistent with the images taken with TEM. The EDX spectrum revealed characteristic peaks corresponding to copper, sulfur, and silver, along with slight traces of carbon and oxygen, as shown in Figure 3d. The presence of carbon probably indicates partial oxidation of an element exposed to the ambient atmosphere. In general, the presence of CuS/Ag MNPs in the photoactive medium not only enhances photon harvesting but is also expected to display LSPR absorption.

3.2. Optical Absorption of CuS/Ag MNPs

The optical absorption spectra were measured from CuS/Ag MNPs powder suspension in deionized water using a UV-Vis spectrometer, as displayed in Figure 4a. The CuS/Ag MNPs exhibit an extensive absorption spectrum ranging from 370 to 700 nm, with a prominent peak centered at 400 nm and a tail extending into the NIR region. The wide absorption band in the IR region is attributed to the LSPR, which depends on the size, nature, and shape of the MNPs [30,31]. The optical bandgap of the CuS/Ag MNPs was estimated from the onset of their optical absorption spectrum in deionized water, as shown in Figure 4b. The most prominent technique for determining the energy bandgap is Tauc’s plot, and it can be calculated by the following equation:

$$\alpha hv=\beta {(hv-E_g)}^n$$

(1)

where α is the absorption coefficient, h is Planck’s constant, v represents the photon frequency, β is a material-dependent constant, $E_g$ is the optical energy bandgap, and n is an exponent that depends on the electronic transition. Nevertheless, the value of n is 2 for direct allowed transitions and 0.5 for indirect allowed transitions, as reported in the literature [32,33].

The value of $E_g$ was determined from the intercept of the tangent line to the curve at (αhv)² = 0 in the plot of (αhv) versus hv, and the direct energy bandgap of CuS/Ag MNPs was found to be 2.03 eV (see Figure 4b). Figure 4c shows the absorption spectra of the films with and without CuS/Ag MNPs, at doping levels of 1%, 3%, and 5%. The pristine film exhibited the characteristic features of the P3HT/PC₆₁BM mixture, with a primary peak centered at 519 nm and accompanied by well-defined vibronic shoulders. Upon doping with CuS/Ag MNPs, the films displayed enhanced absorption intensity compared to the reference film. This enhancement is concentration-dependent, with higher nanoparticle loading generally leading to stronger absorption, indicating improved light-harvesting capability due to the plasmonic effects and possible scattering contributions of the embedded nanoparticles. However, the absorption films incorporating CuS/Ag MNPs at various concentrations revealed two additional, well-defined absorption features located at approximately 390 nm and 850 nm. Such an absorption feature at 390 nm is associated with electron band-to-band transitions within the MNPs, while the broader peak around 850 nm arises because of the presence of LSPR and light scattering effects induced by the MNPs, contributing to enhanced optical absorption in the NIR region. The intensity of optical absorption varied with the concentration of MNPs in the medium. A maximum absorption peak was found at 1% CuS/Ag MNPs in the active layer. However, when the concentration was enhanced to 5%, the absorption intensity decreased. This reduction is attributed to the formation of a high concentration of defects in the active layer, which promotes charge carrier recombination. The increased non-radiative recombination reduces the effective absorption, thereby lowering the intensity of the absorption peak. Moreover, slight variations in the film’s energy bandgap were observed, as shown in Figure 4d and

Table 1. Device parameters of TFOSCs with and without CuS/Ag MNPs in the active layer.

CuS/Ag (wt%)	Eg	Eloss (eV)	Voc (V)	Jsc (mA/cm2)	FF (%)	PCE (%)	Rs (Ωcm2)
Pristine	1.75	1.20	0.55	10.31	39.43	2.53	827
1%	1.61	1.06	0.55	16.30	56.36	5.28	337
3%	1.66	1.11	0.55	15.69	52.93	4.82	350
5%	1.69	1.14	0.55	14.70	50.42	4.02	478

. These bandgap values are duly assigned to reduced charge recombination, resulting from the interaction between CuS/Ag MNPs and the polymer cells. The better concentration of CuS/Ag MNPs for effectively minimizing charge recombination was found at the doping levels of 1%, 3%, and 5% by weight. As a result, the photon energy loss was analyzed based on the variation in the energy bandgap of the films, which can be defined mathematically as $E_{loss}=$ $E_g$ − q$V_{oc}$, where q$V_{oc}$ is the energy derived from the cell. The magnitude of $E_{loss}$ increased from 1.06 eV to 1.20 eV across the solar cells under investigation and appeared to be dependent on the concentration of CuS/Ag MNPs. Notably, the highest $E_{loss}$ was recorded for the reference device without CuS/Ag MNPs, suggesting that the incorporation of CuS/Ag MNPs effectively reduces $E_{loss}$ by suppressing charge recombination. Finally, the best device performance in this investigation was achieved at a 1% doping level, where reduced series resistance ($R_s$) and enhanced photon harvesting contributed significantly to the improved efficiency of the TFOSCs.

3.3. J-V Characteristic of CuS/Ag MNPs

The characterizations of the fabricated devices were duly conducted using J-V performance analysis with varying concentrations of CuS/Ag MNPs in the active layer, as shown in Figure 5.

The concentrations of CuS/Ag MNPs incorporated into the active layer were varied at doping levels of 0% (pristine), 1%, 3%, and 5%. The pristine device, without the inclusion of CuS/Ag MNPs, served as a reference for comparison. The solar cells incorporating CuS/Ag MNPs generally exhibited a significant enhancement in photocurrent generation. This cooperative effect of CuS/Ag MNPs led to effective light trapping, resulting in an increase in short-circuit current density ($J_{sc}$) from 10.31 mA/cm² for the pristine (0%) device to 16.30 mA/cm² (1%), 15.69 mA/cm² (3%), and 14.70 mA/cm² (5%). As a result, the PCE of the devices increased from 2.53% for the pristine (undoped) active layer to a maximum of 5.28% at a CuS/Ag MNP doping concentration of 1%. It is important to note that all key device parameters obtained from the newly fabricated solar cells, including the open-circuit voltage ($V_{oc}$), $J_{sc}$, fill factor (FF), and PCE showed notable improvements, as summarized in Table 1.

Furthermore, the interaction between photogenerated excitons and the localized electromagnetic (EM) field induced by the MNPs enhances exciton diffusion and charge dissociation, ultimately leading to improved photocurrent generation [28,29]. The observed improvement in device performance is primarily attributed to the enhanced photocurrent collection enabled by the incorporation of CuS/Ag MNPs. In CuS/Ag-doped solar cells, the increase in $J_{sc}$ is mainly due to more efficient light harvesting and suppressed charge recombination. These enhancements are also associated with a reduction in $R_s$, which facilitates more efficient charge extraction and minimizes recombination losses compared to the pristine device.

3.4. Charge Transport of CuS/Ag MNPs

The charge transport properties of the photoactive films incorporated with CuS/Ag MNPs were explored through space-charge-limited current (SCLC), which was conducted under dark conditions. The SCLC measurements were conducted under conditions where all traps in the medium were filled and the density of photogenerated charge carriers was low, ensuring accurate assessment of the absorber layer’s transport properties (see Figure 6a). Figure 6b presents the SCLC data obtained from the forward bias region of the dark current, starting from the injection-limited region at approximately 1.5 V and extending to the point where the current is saturated. The charge carrier mobilities of the solar cells were subsequently calculated from the current density characteristics using Mott–Gurney’s law, as given in Equation (2).

$$J={\displaystyle \frac{9}{8}}\epsilon {\epsilon }_o{\mu }_o{\displaystyle \frac{V^2}{L^3}}$$

(2)

where ε is the relative dielectric permittivity of the solar cell device fabrication layer, ε₀ is the permittivity of free space, ε₀ = 8.85 × 10⁻¹² F/cm, µ₀ is the zero-field mobility, L denotes the thickness of the solar absorber up to 100 nm, and V is the applied voltage across the device, which can potentially be corrected by accounting for the built-in potential voltage ($V_{bi}$), as suggested by recent studies [34].

$$V=V_{ap}-V_{bi}$$

(3)

where $V_{ap}$ represents the applied bias voltage and µ represents mobility, which is primarily reliant on the electric field and can be calculated via the Poole–Frenkel (PF) equation, as shown below:

$$\mu ={\mu }_{o}exp\left(\gamma \sqrt{E}\right)$$

(4)

where µ_o is the zero-field mobility and γ is the field activation factor. By combining Equations (2) and (4), the field-dependent SCLC is examined using Mott–Gurney’s law [18], as provided below:

$$J_{SCLC}={\displaystyle \frac{9}{8}}\epsilon {\epsilon }_o{\mu }_o{\displaystyle \frac{V^2}{L^3}}exp\left(0.89\gamma \sqrt{{\displaystyle \frac{V}{L}}}\right)$$

(5)

However, the field activation factor, $\gamma =B\left({\displaystyle \frac{1}{k_{B}T}}-{\displaystyle \frac{1}{k_B{T}_0}}\right)$, is temperature-dependent, where T is the measurement temperature. T₀ and B are characteristic of the material and mostly determined based on its intrinsic properties [35]. The negative sign found at $\gamma$ indicates that the mobility diminishes at high applied electric fields in TFOSCs (see

Table 2. The charge transport parameters of TFOSCs.

CuS/Ag (wt%)	μo (cm2S−1V−1)	γ (cmV−1)
Pristine	2.6413 × 10−4	−1.7798 × 10−4
1%	1.1272 × 10−3	−1.2679 × 10−4
3%	1.5912 × 10−3	−1.4261 × 10−4
5%	1.7259 × 10−3	−1.5789 × 10−4

). The measured SCLC data show excellent agreement with the predictions of Equation (5), as illustrated in Figure 6a,b. As a result, the ${\mu }_o$ values of the CuS/Ag MNPs at 1%, 3%, and 5% exceed those of the reference cell by one order of magnitude (see Table 2). This clearly indicates that doping the active layer with MNPs at different concentrations enhances charge transport processes because of the existence of the LSPR effect. This provides clear evidence of photocurrent enhancement resulting from the suppression of charge recombination.

4. Conclusions

In conclusion, CuS/Ag MNPs were effectively synthesized by the wet chemical processing method and incorporated into the active layer of TFOSCs. The inclusion of CuS/Ag MNPs significantly influenced both the optical and electrical properties of the active layer. Their cooperative effects enhanced light trapping and photon harvesting, resulting in a substantial increase in $J_{sc}$ from 10.31 mA/cm² for the pristine device to 16.30 mA/cm² at an optimal doping concentration of 1% under ambient laboratory conditions. Consequently, the PCE improved from 2.53% in the undoped device to 5.28% with the incorporation of 1% CuS/Ag MNPs. Such a performance enhancement observed with CuS/Ag MNPs is mostly attributed to a reduction in the series resistance ($R_s$), which facilitates more efficient charge extraction and reduces recombination losses compared to the pristine reference device.