Enhanced Efficiency of Polycrystalline Silicon Solar Cells Using ZnO-Based Nanostructured Layers

Abstract

In the context of the global energy transition, enhancing the efficiency of polycrystalline silicon-based solar cells remains a critical research priority. This study investigates the integration of ZnO-based nanostructured layers. ZnO and Al-doped ZnO nanoparticles, synthesized via hydrothermal methods and concentrated solar power (CSP) vapor condensation, exhibiting diverse morphologies—nanorods, spheres, and whisker structures—were deposited onto commercial solar cells using the spin coating technique. Structural, morphological, and spectroscopic analyses confirmed the formation of crystalline layers with high active surface areas and controlled morphology. Photovoltaic performance was assessed using a dedicated hardware–software system under real sunlight conditions. The results demonstrate a significant increase in energy efficiency, reaching up to 10.97%, compared with 1.51% for polycrystalline silicon cells without any supplementary layers. This improvement is attributed to enhanced light absorption, reduced carrier recombination, and more efficient charge transport, driven by nanoscale design and doping. This study underscores the importance of sustainable synthesis and morphological control in the development of high-performance and cost-effective solar technologies.

1. Introduction

In an era where energy sustainability and environmental protection are global priorities, research in the field of advanced materials plays a crucial role in the development of efficient and environmentally friendly solutions. Metal oxide-based materials have gained special attention due to their unique properties, which include high optical transparency, chemical and electrical stability, solar energy conversion capacity, and efficiency in carrying electrical charges [1,2,3,4].

These features make them suitable for a wide range of applications, from state-of-the-art solar cells and optoelectronic devices to photocatalysis processes used in environmental remediation.

A major problem with current renewable energy technologies is conversion efficiency and production cost. Traditional solar cells, especially silicon-based ones, dominate the PV market, but high manufacturing costs and limited energy conversion efficiency have led to the exploration of more affordable and high-performance alternatives. In this context, metal oxides have been intensively investigated as candidate materials for thin films in advanced photovoltaic architectures, due to their advantages in terms of resource availability, stability in diverse environments, and compatibility with large-scale manufacturing methods [5,6].

Since the early investigations into solar energy conversion by Antoine-César Becquerel in 1839, followed by Charles Fritts (who designed the earliest physical representation of a solar cell) in 1883, and continuing to the present day, the optimization of operational conditions and the improvement of energy efficiency have remained key objectives in the field of scientific research.

Even though coal remained an important source of fuel for electricity production globally in 2023, its share has gradually started to decline in favor of renewable energy sources [7]. At the same time, global electricity production continued to grow, driven by the post-pandemic economic recovery, the accelerated electrification of the transport sector, and the expansion of digital and industrial infrastructure. In this context, wind and solar energy have strengthened their position as viable and sustainable alternatives. In 2023, wind energy came to contribute about 7.8% to global electricity production, while solar energy reached a share of 5.5%, which means that, together, these two clean sources generated more than 13% of the world’s electricity [7]. This is a notable development, especially compared with previous years when their contribution was significantly more modest. For the first time in history, in 2023, renewables (including hydropower, solar, wind, biomass, and geothermal) exceeded the threshold of 30% of total global electricity production. This achievement marks a turning point in the global energy transition and underlines the growing commitment of states to achieve decarbonization goals. Solar energy, in particular, has established itself as the third most important source of renewable energy, after hydropower and wind energy, due to the continuous decrease in the costs of photovoltaic technology and the ease of implementation both in residential and industrial-scale projects.

According to the same source, analysis of the dynamics of solar energy production shows that more than 13% of all renewable energy produced worldwide comes from the use of photovoltaic cells, a percentage that is constantly increasing. This trend reflects not only technological advances but also the active involvement of the private sector and individual consumers in the adoption of green solutions.

Of the three generations of solar cells, crystalline silicon solar cells represent the oldest and most advanced photovoltaic technology. In contrast, thin-film solar cells are 350 times thinner and more flexible, being mainly intended for industrial applications for large energy capacities. The essential factor that differentiates them is the type of semiconductor material used to generate the photovoltaic effect. In the case of thin-film solar cells, the semiconductor layer is sandwiched between transparent conductive layers, which serve both as electrical contacts and as pathways for light transmission [8]. Although silicon is used in some cases, it is not in the form of a solid wafer but in a non-crystalline variant [9].

Thin-film solar cells are fabricated through the successive vacuum deposition of thin layers—typically ranging from 1 to 10 μm in thickness—onto various substrates, including polymers, glass, and metals. Etching techniques are subsequently employed to create integrated modules across large surface areas [10].

Multi-junction solar cells are made by advanced technologies, with two or more overlapping junctions. Their commercial viability depends on slow layer-deposition processes, which are used in concentrated photovoltaic systems or in space applications.

Silicon, under various structural morphologies, is the predominant material used in the manufacture of solar cells. In 2020, silicon wafer-based technology accounted for about 95% of total production. The importance of this technology derives from the fact that the manufacturing processes take place at low temperatures and use very thin silicon wafers, about 100 μm thick. Due to the increase in efficiency and the use of thinner wafers, the amount of silicon needed to manufacture solar cells has decreased significantly, reaching about 3 g/Wp [8,10,11]. Given that silicon is predominantly derived from abundant raw materials such as sand or quartz, it is considered fully recyclable.

Among metal oxide-based materials, ZnO- and Cu₂O-based thin-film heterojunctions are considered promising solutions for improving solar cell performance. These structures allow the optimization of the sunlight absorption process and increase the efficiency of energy conversion through a better separation of the carriers from the charge. Cu₂O, for example, is an oxide-type p-semiconductor with a bandgap of about 2.1 eV, making it an ideal material for sunlight absorption layers. In combination with material n of type ZnO, a heterojunction is formed that can be used for the generation of electricity by photovoltaic effect.

Another advantage of metal oxides in the photovoltaic field is their versatility. Titanium (TiO₂), zirconium (ZrO₂), and tin (SnO₂) oxides are used to improve the efficiency of solar cells due to their high transparency and compatibility with various solar cell architectures, including perovskites. In addition, new approaches to incorporating plasmonic nanoparticles into perovskite layers have demonstrated significant improvements in light absorption, leading to conversion efficiency increases of up to 12%.

An essential aspect in optimizing the performance of metal oxides is the technology used to deposit them on various substrates. Modern deposition methods, such as chemical vapor deposition (CVD), physical vapor deposition (PVD), electrochemical deposition, and plasma-assisted techniques, allow precise control of the thickness, composition, and morphology of thin films, directly influencing the efficiency of photovoltaic devices and those used in environmental remediation [12].

For example, recent studies on optimizing the size and shape of ZnO nanoparticles have demonstrated that well-controlled structures can significantly improve the performance of OLED (organic light-emitting diode) devices and solar cells by improving load carrying and reducing optical losses [13]. These findings highlight the importance of nanoscale material engineering in the development of advanced solutions for energy and the environment.

Optimizing solar cell architectures through the use of advanced nanomaterials and specialized optical layers is a key direction to improve the performance of these technologies. In this context, the present study aims to analyze the impact of the use of ZnO nanoparticles and anti-reflective coatings on optoelectronic conversion in emerging solar cells, contributing to the development of efficient and sustainable solutions for solar energy capture.

Recent studies have highlighted the incorporation of metal oxide thin films—such as ZnO, SiO₂, TiO₂, NiO, Al₂O₃, and MgO—in solar cell fabrication, owing to their favorable electrical and optical properties [14,15,16,17,18]. These films have p- or n-type conductivity and a bandgap range of 3 to 7 eV. Deposition techniques for metal oxide layers have become increasingly diverse, encompassing methods such as sol–gel processing, spin coating, CVD (chemical vapor deposition), PVD (physical vapor deposition), RF (sputtering), and printing. Among these materials, magnesium oxide (MgO) has emerged as a particularly promising candidate, primarily due to its high optical transmittance—approximately 91.48% in the visible spectrum [10]. Recent trends emphasize the use of MgO as an intermediate layer in various types of solar cells, including perovskite-based, dye-sensitized, and polymer solar cells [14,15,16].

The use of a solar reactor in the process of vapor condensation of metal oxide powders represents an advanced and sustainable approach in the synthesis of nanomaterials, with the main objective of capitalizing on CSP to induce the thermal evaporation of metallic or oxide precursors, followed by vapor condensation in the form of nanoparticles. This technology fits into the paradigm of ecological production of nanomaterials, considerably reducing the carbon footprint associated with conventional synthesis methods, which involve high energy consumption and the generation of polluting emissions [10,19,20,21,22]. In this context, in the solar reactor, the solar radiation is focused and converted into intense thermal energy, capable of reaching temperatures above 2000 °C at the focal point, without the input of fossil fuels.

The purpose of using this type of reactor in the vapor condensation process is twofold. On the one hand, it allows a rapid and efficient evaporation of the target materials in a controlled atmosphere (inert or slightly reducing), favoring the formation of a cloud of metallic or oxide vapors. On the other hand, it facilitates nucleation and particle growth at the nanoscale through the fine control of cooling and condensation parameters. The key advantage of this method is the ability to produce powders with spherical morphology, homogeneous dimensions, and high crystallinity, without organic contaminants or synthesis residues. Moreover, the high cooling rate inherent in the process allows metastable phases or amorphous structures to be obtained, which are difficult to access by traditional thermal methods.

In particular, for metal oxides, solar vapor condensation provides a unique platform for controlling the stoichiometry and oxidation state, due to the adjustable atmosphere in the reactor and the steep thermal gradient between the vaporization and condensation zones [19,23]. Thus, mixed oxides, dopant compounds, or systems with a core–shell structure can be synthesized, which is essential in applications such as photocatalysis, gas sensors, magnetic materials, or energy storage. In addition, the integration of direct solar energy in this process strengthens the position of the solar reactor as a strategic technology within the advanced synthesis of oxide nanomaterials.

The spin coating method is a versatile and widely used technique for the uniform deposition of thin films on flat substrates, having significant applications in the manufacture and optimization of solar cells [24,25,26,27,28]. In the context of photovoltaic devices, this method is frequently used for the application of functional additional layers, especially those based on metal oxides, which play a critical role in improving the performance of solar cells by optimizing the collection and separation of charges, reducing recombination, and improving the chemical and structural stability of the device [27,28].

Spin coating allows precise control of the thickness of the deposited layer by adjusting process parameters such as rotational speed, viscosity of the precursor solution, and rotation time. This precision is essential in interface engineering, where the thickness and homogeneity of the metal oxide layer—such as TiO₂, ZnO, SnO₂, or NiO—directly influence the electronic behavior of heterojunctions and the optoelectronic properties of the cell [24]. Furthermore, the technique is compatible with a wide range of soluble materials, including sol–gel precursors, colloidal suspensions, or nanoparticle solutions, making it ideal for the integration of nanostructured materials with advanced functional properties. A notable advantage of the method lies in its operational simplicity and low cost, which makes it recommended not only for fundamental research but also for scalable manufacturing processes, especially in emerging solar cell architectures.

2. Materials and Methods

2.1. Elaboration and Characterization

Zinc nitrate hexahydrate, Zn(NO₃)₂·6H₂O, KOH, and AlCl₃ analytical purity (Sigma Aldrich purchase) were used as the precursors for the hydrothermal synthesis of pure and 10% at. Al-doped ZnO powders. The precursor concentration of 1 M at pressure 4.5 bar was heated at 200 °C for 40 min in a Cortest autoclave, capacity 2 L. After the hydrothermal process, the powders were washed with ethanol and deionized water to prevent agglomeration. The hydrothermal powders were used as precursors in the solar concentrated reactor to obtain nanostructured particles and different morphologies.

The entire experimental assembly used to make the metal/nano/oxide powders was a semi-automated solar reactor (Figure 1), designed to operate under strict conditions involving environmental parameters and control and drive operations. Its operation presupposes, first of all, the existence of a minimum intensity of solar radiation, an essential condition for initiating the experiment. This value, correlated with the position and solar trajectory estimated through GPS data and solar maps, allows the automated tracking system (flat mirror redirecting sunlight) to orient the concentrating plane mirrors in such a way to maximize the capture of the solar flux at the laboratory level. After focusing the radiation, the micrometer oxide sample is precisely positioned in the center of the reactor focus using a three-axis positioning system to ensure optimal exposure to the concentrated energy. Subsequently, the internal pressure in the glass balloon is set and maintained at a reference value with the help of the vacuum pump with manual control, which ensures a controlled atmosphere inside the glass flask. In parallel, the dual cooling system is activated, designed to dissipate heat at the level of the sample holder and the collector placed before the ceramic filter, thus protecting the sensitive components of the installation. At the end of the experiment, the depressurization of the reactor is carried out in a controlled manner, followed by the displacement of the assembly in a neutral area, without air currents that could disperse the nanopowders obtained. In this step, the system is unpacked and the experimental product (nanopowders) is collected from the filter, the collector, and the inner surface of the heat-resistant glass flask. Shutting down the cooling system is not mandatory due to its high autonomy in terms of moving the assembly, but, if it is deactivated, it is necessary to reactivate it before starting the next experiment.

Table 1. Synthesis parameters of nanopowders.

Sample	Solar Flux	Pressure
Np_CSP/ZnO	750	20 torr
Np_CSP/10AlZnO	853	20 torr

presents the parameters of CSP synthesis starting with the hydrothermal powders.

All powders prepared by hydrothermal synthesis and using CSP were used as precursors for spin coating deposition.

For the experimental studies, commercial polycrystalline silicon solar cells with dimensions of 5.2 × 1.9 cm were used, which were subjected to a rigorous cleaning process in an ultrasonic bath, using ethanol and deionized water for 480 s for every cleaning solution in order to remove surface impurities. Zinc oxides were dispersed in ethylene glycol using the ultrasonic bath for 10 min at a frequency of 42 kHz.

In the initial deposition stage, 100 μL of the precursor solution was applied using the dynamic spin coating technique, with the substrate rotated at 25 rps using SCV-10 equipment, as presented in Figure 2. In the subsequent step, 16 aliquots of 50 μL were added at 30 s intervals while maintaining a rotation speed of 33 rps. In the third step, the coated samples were subjected to a post-deposition heat treatment, consisting of keeping them on a hot plate, MS-H280-Pro, at temperatures between 180 °C for a duration of 20 min in order to consolidate the layer and optimize the morphology.

The description of the samples is presented in

Table 2. Sample description.

Sample Code	Layer Description	Precursor Synthesis
PSi	Si polycrystalline solar cell, without additional layers added	-
ZnO	Si polycrystalline solar cell with ZnO layer	Hydrothermal synthesis of ZnO
10AlZnO	Si polycrystalline solar cell with Al-doped ZnO layer	Hydrothermal synthesis of Al-doped ZnO
CSP/ZnO	Si polycrystalline solar cell with ZnO layer	CSP synthesis of ZnO
CSP/10AlZnO	Si polycrystalline solar cell with Al-doped ZnO layer	CSP synthesis of Al-doped ZnO

.

The methods used for qualitative and morphological characterization of the obtained powders were ATR-FTIR (attenuated total reflectance–Fourier-transform infrared spectroscopy) and SEM (scanning electron microscopy).

X-ray diffraction analysis (XRD) was the method used to structurally characterize the elaborated powder by hydrothermal and CSP processes. Analysis was performed in a Project SFERA framework, using X-ray diffraction (Philips PW 1710), with CuKa radiation. The angular domain was between 15 and 80°. The JCPDS 36-1451 standard [6], characterized by the P6₃mc space group and group parameters of a ≈ 3.249 Å and c ≈ 5.206 Å, was used for the identification of the crystalline phase. The Scherrer equation was used to determine the grain size.

For the examination of particle morphology, scanning electron microscopy with secondary electron detectors (SE (secondary electron)), a Leo (Zeiss) 1530 microscope was used. The particles were spread on carbon type and analyzed at variable pressure, 20 kV acceleration voltage, and 15 µA emission current.

The equipment used to determine the purity of all samples by ATR-FTIR spectra was Bruker Tensor 27, in the range 4.000–350 cm⁻¹, with a resolution of 4 cm⁻¹.

2.2. Solar Cell Hardware and Software Test System

With the primary objective of optimizing energy efficiency through the integration of solar cells coated with advanced optoelectronic layers, a hardware–software system was designed and implemented to enable real-time monitoring and precise control of the cells’ operating parameters [29]. The system facilitates continuous acquisition of key metrics, including current, voltage, and incident solar irradiance, thereby allowing comprehensive performance evaluation under dynamic environmental conditions [30,31]. For this purpose, an integrated acquisition and control system was developed, supported by a software application capable of correlating environmental parameters—such as temperature and illumination intensity—with the characteristic electrical parameters of the solar cells, including open-circuit voltage, load voltage, and short-circuit current.

The general structure of the testing system used to evaluate the efficiency of solar cells is presented in Figure 3. It integrated several essential functional blocks: the solar cell testing block, composed of commercial UFY002914-type cells; the controlled electronic load block, consisting of four programmable current sources that enabled current regulation through each cell; the acquisition and control block, responsible for bidirectional communication with the dedicated software and for interfacing with the electronic components; a stabilized power supply, which provided the DC voltage required for system operation; and a Volcraft PL-110SM pyranometer, used to monitor the level of incident illumination on the cells during testing.

The acquisition of experimental data was performed via the NI USB-6211 interface, which, according to the manufacturer’s technical specifications, features 16 analog input channels (16-bit resolution, 250 kS/s sampling rate), 2 analog output channels, and 8 digital bidirectional channels. In the configuration used for the conducted tests, eight analog input channels were employed to monitor the voltages of the four solar cells (both in open-circuit and under-load conditions), along with one analog output channel dedicated to electronic load control. Figure 4 illustrates the testing steps of polycrystalline solar cells before and after the deposition of additional layers using a software control system. The interface was connected to a software application developed in Python 3.9, which was responsible for full system management (load control, data acquisition, processing, and saving in a CSV file).

Additionally, the application enabled the automatic generation of comparative plots for analyzing the electrical performance of two or more solar cells, using data vectors structured in a format optimized for interpretation.

3. Results and Discussion

3.1. Qualitative Analysis by Spectroscopy

Fourier-transform infrared (FTIR) spectroscopy was employed as a qualitative analytical method specific to nanostructured samples. Accordingly, the powders synthesized in the solar reactor were analyzed, and the corresponding spectra are presented in Figure 5.

The black curve exhibits sharp and well-defined peaks corresponding to Zn–O bond vibrations and the characteristic crystal lattice of pure ZnO, which serves as a reference for ZnO identification.

Figure 5 illustrates the following characteristic spectral regions:

• The region between 350 and 400 cm⁻¹ is specific to the vibrational modes of the ZnO crystal lattice. Changes observed in this region can be attributed to Al doping, which may alter the crystalline structure. The red spectrum displays features similar to the green one, confirming the presence of Al-doped ZnO.
• The minor absorption at ~2360 cm⁻¹ is due to carbon dioxide (CO₂) in the environment.

By comparing the green and red curves with the black curve as reference, noticeable variations in the 480–500 cm⁻¹ region can be observed, confirming the influence of Al doping and the synthesis method on the structural properties of the ZnO-based powders.

3.2. Structural Analysis

X-ray measurements were performed in the “as-elaborated” powders, as presented in Figure 6. All diffractograms of pure ZnO and Al-doped powders show only the diffraction peaks corresponding to ZnO, hexagonal crystalline lenses typical of the wurtzite structure, with well-defined reflections. The analysis of the XRD diffractograms of the powders synthesized by hydrothermal and CSP method confirm the formation of the typical ZnO hexagonal phase in both samples, without the appearance of secondary phases, as presented in Figure 6. The main reflections occur around 2θ values of 31.7°, 34.4°, 36.2°, 47.5°, 56.6°, 62.8°, 66.3°, 68.0°, 69.2°, 72.6°, and 76.8°, being assigned to the crystallographic planes (100), (100), (002), (101), (102), (103), (200), (112), (201), (004), and (202), respectively. However, the intensity and clarity of the peaks differ between methods. The SCP samples show more intense and better defined reflections, suggesting a higher crystallinity than the hydrothermal samples. In the case of the 10% at. Al-doped ZnO sample (Figure 6a), hydrothermally elaborated, a decrease in intensity and a widening of the diffraction tips are observed, which suggests a slight decrease in crystalline order and a reduction in the size of crystallites. The absence of additional reflections indicates that aluminum doping does not lead to the formation of XRD-detectable secondary phases, and Al³⁺ is possible to be incorporated into the ZnO lattice by the isomorphic substitution of Zn²⁺ ions.

In the case of the 10% at. Al-doped ZnO, elaborated by CSP (Figure 6b), a slight widening and decrease in the intensity of the peaks is observed, which indicates a possible decrease in the size of the crystallites and a distortion of the lattice caused by the substitution of Zn²⁺ with Al³⁺. The results highlight the efficiency of the CSP method in obtaining ZnO and Al-doped ZnO powders with good crystallinity and small grain size.

The graphical representation of the grain size determined by the Scherrer equation highlights significant differences between ZnO samples and those doped with 10% at. Al. Figure 7a (hydrothermal ZnO powder) shows a relatively uniform distribution of crystallite size (20–35 nm), while Al doping leads to a much wider variation, with values between 30 and 75 nm. This dispersion indicates an anisotropic growth behavior of crystallites in the presence of aluminum, possibly due to crystal cell stresses and changes in the energy surface of growth planes. The maximum size is observed on plane (004), suggesting a preferential orientation in this direction.

Figure 7b (grain size variation of the powders obtained by the CSP method) shows a significant increase in crystallite size in the case of the 10% Al-doped sample compared with pure ZnO. For ZnO, the values are between 22 and 33 nm, with a relatively uniform distribution, specific to homogeneous crystallization. In contrast, Al doping causes an anisotropic increase in crystallites, with sizes reaching up to 45 nm, especially on planes (002) and (200), indicating a preferential orientation of growth in these directions. These results highlight the combined influence of doping and intense heat energy from the CSP process on the mechanisms of nucleation and crystal growth.

3.3. Morphological Characterization

Analysis of SEM micrographs of all samples elaborated by hydrothermal and CSP processes aims to determine particle size and morphology. Particle size distribution histograms are performed for particles with very clear and fully visible outlines.

The morphology of the pure ZnO particles, shown in Figure 8a, is characteristic of hydrothermal synthesis, exhibiting anisotropic growth driven by the wurtzite crystal structure, resulting in a rod-like shape. This morphology indicates a high surface-to-volume ratio and greater exposure of the active crystal facets.

The elongated morphology suggests unidirectional growth, with particle lengths ranging from approximately 115 to 250 nm. The particle size distribution histogram in Figure 8b reveals a monomodal slightly asymmetric distribution, attributed to nanoparticle agglomeration. Such agglomeration suggests a tendency for nanorod self-assembly, which may be influenced by intermolecular forces and surface energy.

The maximum peak of the histogram lies in the 200–220 nm range, with an average particle size of approximately 210 ± 40 nm.

The morphology of the Al-doped ZnO particles, shown in Figure 9a, consists of spherical agglomerates resulting from the doping process, which alters particle growth dynamics. A pronounced tendency toward agglomeration is observed, likely due to electrostatic interactions.

The particle size distribution histogram presented in Figure 9b, with a single dominant peak, indicates homogeneous particle sizes and improved crystallization control.

The average particle size is significantly smaller compared with pure ZnO, approximately 50 ± 15 nm.

Al doping has notably influenced the crystallization process by limiting axial particle growth and promoting multiple nucleation events.

The SEM image of ZnO particles synthesized using CSP, from Figure 10, reveals a distinct morphology characterized by the formation of whisker-like structures with tetrapod-like orientation, interconnected within a three-dimensional network. This complex architecture suggests accelerated directional growth, driven by the thermal energy generated through exposure to CSP. The nanorods exhibit lengths ranging from 200 to 600 nm, with diameters between approximately 100 and 300 nm, consistent with the particle size distribution histogram, which indicates an average size of about 300 ± 100 nm.

In contrast with particles obtained via hydrothermal synthesis, ZnO synthesized under CSP demonstrates enhanced directional growth and highly favorable morphology for applications due to the increased active surface area and the potential for efficient charge transport.

The Al-doped ZnO particles synthesized using CSP exhibit a dual morphology, consisting of both whisker-like structures and spherical particles. The SEM image shown in Figure 11a highlights the formation of well-defined whiskers with lengths up to 1.4 µm, resulting from the directional crystal growth induced by high temperatures and thermal flux generated by CSP. Simultaneously, homogeneously distributed spherical particles are also observed, with an average size of approximately 55 ± 20 nm.

This dual morphology may result from a diffusion-controlled growth mechanism, where the growth rate is limited by material transport. The particle size distribution histograms in Figure 11b,c confirm this morphological segregation. The whiskers exhibit a broad distribution with a maximum in the 500–600 nm range, while the spherical particles are concentrated within the 40–100 nm interval. This mixed morphology, achieved in a single-step process, offers potential synergistic advantages by combining the high surface area of the spheres with the directional transport properties of the whiskers.

A comparative analysis of pure and Al-doped ZnO synthesized via hydrothermal methods with that produced using CSP reveals the significant influence of aluminum doping and solar energy input on both particle morphology and size. While hydrothermal synthesis enables the formation of well-defined particles, Al doping leads to the development of nanoscale rod-like morphologies. The use of solar energy, in turn, promotes the formation of whisker-type structures, characterized by accelerated crystal growth and increased morphological diversity—features that are particularly relevant for advanced applications in functional materials.

3.4. Electrical Parameter Acquisition and Testing of Solar Cells

In the experiment, four commercial solar cells without additional deposited layers are connected to the controlled electronic load block of the control and acquisition system, each cell being associated with a variable electronic load, adjustable between 0 mA and 500 mA in 5 mA increments. For each step, 100 voltage values are acquired, and their average is computed. Subsequently, measurements are performed on the three blank cells, and the voltage corresponding to each load current value is acquired using the data acquisition board. This process is repeated five more times to validate the experimental results (see Figure 12a).

Comparison of the acquisitions values for each cell reveal possible voltage variations. If, for the same current value and the same illumination level, the variations exceed the 10% threshold, the measurements are repeated to ensure the reproducibility of the data. Subsequently, the measured values are averaged, generating the characteristic curves of blank cells, without additional layers (Figure 12b). After this step, additional layers are deposited on the four cells, and the initial steps of the experiment are repeated for these modified cells in order to compare the electrical characteristics. The obtained data are saved in a CSV file, and the characteristic curves (Figure 13) are plotted for final analysis. The initial measurements (on cells without additional layers deposited) are made on 1 July 2024, between 12:00 and 13:00 (Romanian time), under a light intensity of 852 W/m², while the measurements for cells with additional layers are made on 8 July 2024, in the same time interval, under a light intensity of 834 W/m².

3.5. Influence of Additional Layers: Statistical Analysis

The present analysis aims to highlight the influence of the physicochemical parameters of the additional layers applied on polycrystalline silicon (PSi) solar cells on the electrical performance obtained. The types of synthesis of zinc oxides and the sizes and shapes of the particles used are correlated with the electrical power, energy efficiency, and relative increase compared with the reference cells (PSi without additional layer).

According to Table 2, four types of solar cells are tested and analyzed.

A comparison between the current–voltage (I–V) and power–voltage (P–V) characteristics obtained by measurements of commercial polycrystalline silicon solar cells without additional layers and of the same cells after the deposition of additional layers is shown in Figure 14. In Figure 14a, the comparison is made in relation to the average performance of untreated commercial cells and, in Figure 14b, the reference is a cell with a layer of commercial ZnO.

3.6. Effect of Layer Synthesis on Cell Performance

Analysis of the obtained power values (see Table 2 and Figure 15) reveals a progressive and significant improvement in the performance of solar cells following the application of additional layers and the change in synthesis technology. The reference PSi cell, without any additional layer, records an average power of 12.726 mW. The addition of a layer of ZnO, hydrothermally elaborated, containing nanometric-sized rod-like particles, increases the power to 20.275 mW, marking an absolute gain of 7.549 mW and a relative increase of 59.3%. Replacing the ZnO with Al-doped ZnO also synthesized via a hydrothermal method (10AlZnO) significantly raises the power to 88.421 mW, representing a difference of 68.146 mW compared with ZnO and an increase of over 336%. Applying the CSP method to pure ZnO (CSP/ZnO) leads to a power output of 91.484 mW, with an additional gain of 3.063 mW compared with 10AlZnO and an increase of 615% compared with PSi, clearly demonstrating the advantages of solar synthesis. Finally, the best-performing layer, CSP/10AlZnO, delivers an average power of 92.322 mW, which is only 0.838 mW higher than CSP/ZnO, but offers a cumulative advantage of 625.5% over the baseline PSi cell. Compared with plain ZnO, this final layer yields an increase of 72.047 mW, corresponding to a 355.2% improvement, emphasizing the synergistic effect of both doping and CSP processing. Thus, each stage of structural and morphological modification of the additional layer contributes quantitatively and significantly to the enhancement of solar cell performance.

The average energy efficiency values indicate a clear evolution depending on the nature and structure of the applied additional layer (see

Table 3. Energy influence of additional layers deposited on commercial Si polycrystalline solar cells.

Sample Code	Power [mW]	Power Difference [mW]	Power Difference [%]	Energy Efficiency [%]
PSi	12.726	-	-	1.511
ZnO	20.275	7.549	59.323	2.408
10AlZnO	88.421	75.695	594.829	10.504
CSP/ZnO	91.484	78.758	618.900	10.867
CSP/10AlZnO	92.322	79.597	625.488	10.967

). The reference PSi cell, without any additional treatments, exhibits an efficiency of 1.511%, serving as the baseline reference. The application of a commercial ZnO layer, composed of rod-like particles, leads to an efficiency of 2.408%, representing an absolute increase of 0.897 percentage points and a relative improvement of approximately 59.4% compared with PSi. Transitioning to Al-doped ZnO (10AlZnO) results in an efficiency of 10.504%, marking an impressive increase of 8.096 percentage points over plain ZnO and an improvement of over 336%. The efficiency obtained with the CSP/ZnO layer is 10.867%, slightly higher than that of 10AlZnO, with a gain of 0.363 percentage points and an increase of 619% compared with PSi. The highest average efficiency is achieved with the CSP/10AlZnO layer, reaching 10.967%, which corresponds to an increase of 0.100 percentage points over CSP/ZnO and 625% over the reference cell. Compared with ZnO, this final layer results in an absolute increase of 8.559 percentage points, yielding an efficiency more than 3.5 times higher. These results highlight the fact that both doping and advanced synthesis via CSP cumulatively contribute to optimizing the conversion of solar energy into electrical energy, and the quantitative differences between the samples support the superiority of nanostructured and doped solutions in solar cell architecture.

3.7. The Effect of Particle Size on the Performance of the Studied Solar Cells

The particle size of the additional layer applied to the solar cells is a critical factor in optimizing photoconversion performance. The comparative analysis between the different samples demonstrates a direct correlation between the reduction in particle size and the increase in both power generation and energy efficiency.

In the case of the ZnO layer, the particles used are 0.115–0.250 µm in size and have a rod-like shape. This configuration offers a modest improvement in performance over the PSi control cell: the average power increases from 12.726 mW to 20.275 mW (+59.3%) and the energy efficiency from 1.511% to 2.408% (+59.4%). However, the increase is relatively limited, suggesting that the large particle size reduces the active surface area available for photon–electron interactions and decreases the efficiency of collecting charge carriers. By switching to smaller nanometer particles, a dramatic improvement in performance is observed. In the case of the 10AlZnO sample, where particles with tens of nanometer sizes are used, the average power reaches 88.421 mW, which means an increase of 68.146 mW (+336%) compared with ZnO. Similarly, energy efficiency rises to 10.504%, with an absolute jump of 8.096 percentage points compared with ZnO (+336%). This change demonstrates that nanoscale particles create a more porous structure with a much larger specific surface area, facilitating the generation and collection of charge carriers, reducing losses through recombination and increasing current density.

Next, in the case of the CSP/ZnO and CSP/10AlZnO samples, where nanometric particles are also used but with an advanced morphology (whiskers), the performances reach the maximum values in the entire experimental set: the average power is 91.484 mW and 92.322 mW, respectively, and the efficiencies reach 10.867% and 10.967%. Compared with micrometric ZnO, we find an increase in power by more than 72 mW and an improvement in efficiency by about 8.56 percentage points, which corresponds to a jump of more than 350%. This confirms that the nano-whiskers of the active layer not only optimize the interface with the silicon substrate but also considerably improve the mobility of the charge carriers within the layer, maximizing the overall yield of the device.

3.8. Effect of Particle Shape on the Performance of the Studied Solar Cells

The shape of the particles used in the additional layers applied to the solar cells significantly influences the efficiency of converting solar radiation into electrical energy by the way it affects both the light absorption and the mobility of the charge carriers within the layer. In this experiment, three morphological types of particles are analyzed—spherical, nanorods, and whiskers—each having a distinct impact on the electrical parameters of the cells.

Rod-like particles are found in the pure ZnO. This shape, although homogeneous and easy to disperse, has an agglomerate and a random organization in the deposited layer, which limits both the efficient absorption of light and the collection of charge carriers (see

Table 4. Sample codes, morphological structures, and associated performance observations.

Sample Code	Form	Observations
ZnO	Nanorod	Modest efficiency
10AlZnO	Spheres	Enhanced efficiency, directed charge transport
CSP/ZnO,	Whiskers	Maximum efficiency collection/favorable interface
CSP/10AlZnO	Whiskers, spheres

). Thus, the cell treated with rod-like ZnO achieves an average power of 20.275 mW and an efficiency of 2.408%, which is only a modest improvement over the control cell (PSi), indicating that the rod-like shape does not provide an efficient network for directed load transport. In contrast, the spherical particles used in the 10AlZnO sample favor the migration of charge carriers along the longitudinal axis and significantly reduce the likelihood of recombination. This morphology generates natural channels for conduction, leading to an average power of 88.421 mW and an efficiency of 10.504%. Compared with spherical ZnO, there is an increase of 68.146 mW in power (+336%) and 8.096 percentage points in efficiency (+336%), clearly attributable to the oriented organization and superior active surface area offered by nanorods. The whisker shape, present in the CSP/ZnO and CSP/10AlZnO samples, introduces an extremely advanced morphology, consisting of thin, elongated, and branched structures, with a very high length/width ratio. These structures create a grid-like spatial architecture, which allows not only an improved absorption of solar radiation due to multiple scattering effects but also an efficient and fast transport of charge carriers to the electrodes. Thus, CSP/ZnO reaches an average power of 91.484 mW and an efficiency of 10.867%, while CSP/10AlZnO rises to 92.322 mW and 10.967%. In relation to spherical ZnO, we find an increase in power of more than 72 mW (+355%) and an efficiency more than 4.5 times higher, clearly demonstrating the superiority of the whisker morphology.

3.9. Technical Economic Analysis for a Residential Photovoltaic System Connected to the Grid

This analysis investigates how the electrical parameters of a photovoltaic panel built using the studied commercial cells are modified, while aligning with the specifications of a commercially available polycrystalline solar panel—the model JA Solar JAP72S10 350/SC [32]. The technical characteristics of this panel are as follows: nominal power 350 Wp; module efficiency: 17.4%; voltage at maximum power V_mpp: 38.43 V; current at maximum power I_mpp: 9.11 A; open voltage V_oc: 46.27 V; short-circuit current I_sc: 9.58 A; dimensions: 2015 mm × 996 mm.

Considering both the dimensional constraints and the electrical parameters of the reference panel and the commercial cells (without additional layers), the following optimal configuration is proposed: 76 cells connected in series; 23 strings connected in parallel; total panel voltage: 38.0 V; total current: 9.2 A; total number of cells used: 1748.

Using the PVSyst 8.0.7 software solution [33], a 5.6 kW residential system is designed (JA_Solar_JAP72_S10_350_SC. PAN (350 Wp)), located in the same geographical location where the measurements on commercial polycrystalline Si solar cells are made. According to the software analysis, it results in a requirement of 16 units—2 strings × 8 in series.

Figure 16 illustrates the energy performance and economic implications of a residen-tial photovoltaic system. Thus, Figure 16a highlights the monthly variation of the energy generated in the selected location, with higher values in the summer months and lower in the winter. Figure 16b compares the total amount of excess electricity produced and the proportion of it injected into the grid over a year, for panels with different types of additional layers, and Figure 16c analyzes the annual energy production of the system, depending on the solar cell technology used, highlighting the impact of the additional layers.

Analysis of the data presented in

Table 5. Influence of additional layers applied on commercial Si polycrystalline solar cells used for the realization of a single solar panel implemented in a residential photovoltaic system.

Solar Panel with Sample	Power [W]		New Efficiency Depending on the Layers [%]	New Power with Increased Efficiency [W]	Power Difference, Obtained from the Layers [W]
	Maximum Values	Average Values
PSi	22.6	22.2	20.24	349.55	0
ZnO	36.2	35.4	21.167	365.56	16.01
10AlZnO	157.2	154.6	29.388	507.54	157.99
CSP/ZnO	161.3	159.9	29.664	512.3	162.75
CSP/10AlZnO	165	161.4	29.918	516.69	167.14

highlights the impact of applying various additional layers on polycrystalline silicon (PSi) solar cells extrapolated to a residential photovoltaic system. The focus is on increasing the power available for daily household consumption and how the synthesis, size, and shape of the particles influence this increase. The performances are evaluated in terms of average power and energy efficiency and also according to the daily energy potentially produced, estimated for an operating regime of 5 h of equivalent solar irradiation per day—a typical scenario for temperate zones.

The type of synthesis used to obtain the active layer plays a decisive role in the operating efficiency of the solar panel. In the basic configuration (PSi), we obtain an estimated daily production of about 1.75 kWh/day. The application of a commercial ZnO layer, obtained without advanced synthesis, brings a moderate increase in power, corresponding to a daily production of 1.83 kWh/day, an additional intake of only 80 Wh/day. In contrast, the use of hydrothermal synthesis for Al-doped ZnO (10AlZnO) allows a daily production of 2.54 kWh/day to be achieved, marking a significant jump of +45% compared with PSi. The synthesis of CSP applied to pure ZnO (CSP/ZnO) brings an increase of up to 2.56 kWh/day, and the one applied to 10AlZnO (CSP/10AlZnO) generates a maximum production of 2.58 kWh/day, i.e., 0.83 kWh/day more than PSi. Annually, these differences translate into increases of up to 300 kWh, representing a significant contribution to residential self-consumption. From another point of view, through the transition to nanometric particles (10AlZnO, CSP/ZnO, CSP/10AlZnO), the power increases by more than 140 W, and the efficiency increases by more than 8.7 percentage points. This evolution translates into real increases in the daily energy produced from ~1.8 kWh/day (ZnO) to ~2.58 kWh/day (CSP/10AlZnO). Thus, the use of nanoparticles is not just a laboratory option but a fundamental condition for increased performance in real applications. The shape of the particles completes the picture of morphological influence. Rod-like particles (ZnO) do not provide a favorable architecture for the directed transport of charges, which is why the additional daily energy obtained remains below the threshold of 100 Wh/day. Spherical particles (10AlZnO) create organized structures, leading to a daily increase of about 0.79 kWh/day compared with PSi, whereas whiskers (CSP/ZnO and CSP/10AlZnO), due to their complex and branched geometry, enhance both light absorption and charge carrier migration, reaching the highest value of daily energy production—2.58 kWh/day, that is, 46% more than the control cell. These differences are directly reflected in the low costs of the energy produced and in the efficiency of the use of the active surface of the panel.

In addition to significant increases in energy performance, the application of advanced coatings (in particular, CSP/10AlZnO) also generates direct economic benefits, quantifiable through additional energy production. Assuming an average irradiation duration of 5 h/day, the production difference between a control cell (PSi) and one treated with CSP/10AlZnO layer is approximately 0.83 kWh/day, equivalent to ~303 kWh/year per module. At an average electricity price for residential consumers in Romania (~EUR 0.22/kWh), this difference translates into an annual saving of ~EUR 66.66/year/module, and, at the EU level (~EUR 0.29/kWh), the annual saving would be ~EUR 87.87/year/module (based on 2024 estimates). In the case of a system with 10 modules, the financial benefit reaches between ~EUR 667 and EUR 879 per year, just by choosing a morphologically and synthetically optimized layer structure. Thus, investing in advanced materials not only increases energy efficiency but also contributes to the accelerated amortization of the photovoltaic system, ensuring both technological and economic sustainability in a European context marked by the energy transition.

4. Conclusions

The results of this study highlight the decisive impact of nanomaterial engineering on the performance of polycrystalline silicon solar cells, demonstrating that photovoltaic conversion efficiency can be significantly improved through the use of additional layers optimized in terms of synthesis method, particle size and shape, and chemical composition. The CSP/10AlZnO layer, obtained via CSP vapor condensation and doped with aluminum, achieves an energy efficiency of 10.967% due to its advanced whisker-type morphology combined with nanometric spherical particles. The spin coating technique enables the uniform and efficient deposition of these layers, supporting their integration into industrial applications. It is confirmed that nanometric particle size and geometry play a critical role in determining the architecture of the active layer, charge transport, and light absorption, with directional morphologies such as nanorods and whiskers leading to superior results. In a residential photovoltaic system, these improvements translate into an increase in daily energy production from 1.75 kWh/day to nearly 2.6 kWh/day per module, generating substantial technical and economic benefits. Future research should focus on integrating synthesis technologies into continuous industrial processes, analyzing behavior under repeated thermal cycles, and testing on large-area surfaces relevant to commercial applications. Additionally, an interdisciplinary approach—combining nanotechnology, optoelectronics, and energy system analysis—will accelerate the widespread adoption of these solutions.

Given the considerable potential of metal oxides in photovoltaic applications, future research must focus on optimizing their properties through advanced nanomaterial synthesis and engineering strategies. The integration of plasmonic nanoparticles, controlled doping, and the use of innovative deposition methods can help increase the efficiency of these materials and expand their use in sustainable technologies.