Non-Vacuum Processed Polymer Composite Antireflection Coating Films for Silicon Solar Cells

Abstract

A non-vacuum processing method for preparing polymer-based ZrO₂/TiO₂ multilayer structure antireflection coating (ARC) films for crystalline silicon solar cells by spin coating is introduced. Initially, ZrO₂, TiO₂ and surface deactivated-TiO₂ (SD-TiO₂) based films were examined separately and the effect of photocatalytic properties of TiO₂ film on the reflectivity on silicon surface was investigated. Degradation of the reflectance performance with increasing reflectivity of up to 2% in the ultraviolet region was confirmed. No significant change of the reflectance was observed when utilizing SD-TiO₂ and ZrO₂ films. Average reflectance (between 300 nm–1100 nm) of the silicon surface coated with optimized polymer-based ZrO₂ single or ZrO₂/SD-TiO₂ multilayer composite films was decreased down to 6.5% and 5.5%, respectively. Improvement of photocurrent density (Jsc) and conversion efficiency (η) of fabricated silicon solar cells owing to the ZrO₂/SD-TiO₂ multilayer ARC could be confirmed. The photovoltaic properties of Jsc, the open-circuit photo voltage (V_OC), the fill factor (FF), and the η were 31.42 mA cm⁻², 575 mV, 71.5% and 12.91%. Efficiency of the solar cells was improved by the ZrO₂-polymer/SD-TiO₂ polymer ARC composite layer by a factor of 0.8% with an increase of Jsc (2.07 mA cm⁻²) compared to those of fabricated without the ARC.

1. Introduction

Improving the absorption properties and reducing the optical losses in solar cells are crucial to achieve higher efficiencies. In conventional 100–300 μm thick crystalline silicon solar cells micrometer scale alkaline or acidic surface texturing is the primary method for trapping light in the solar cell. Additionally, various materials and thin films have been used for anti-reflection coating (ARC) purposes, including SiO₂, SiN_x, TiO₂, Al₂O₃ thin films, etc., depending to the type of solar cell. On the other hand, several nanoscale light absorption, trapping and antireflection scheme approaches including nanoscale surface texturing and nanoparticle assisted structures, and plasmonic structures has also been investigated due to their good electro-optical properties [1,2,3]. These approaches are based on light scattering and trapping at nanoscale approaches and mainly focus on thin film solar cells where conventional micrometer scale surface texturing is not suitable and on low grade silicon solar cells where the short carrier collection length needs to be considered. Excellent light absorption of silicon nanostructures could be achieved by silicon nanowires [4,5], silicon pillars [6,7], silicon nanoholes [8,9] for microscale thin film solar cell applications.

Today, SiN_x films dominate in the conventional silicon solar cells industry with their effective antireflective behavior and the passivation effect which leads to high bulk lifetimes [10,11]. SiN_x or stack layers (e.g., SiN_x/SiO_x) are deposited by the well-known plasma-enhanced chemical vapor deposition (PECVD) technique [12,13]. However, the need for toxic and hazardous gases such as SiH₄ and NH₃ with vacuum processing for CVD operation and considerably high costs come along as the main drawbacks. In order to meet industrial requirements with simple low-cost technologies with high throughput, cost-effective methods need to be investigated and adapted to the solar cell manufacturing process. TiO₂ has been introduced in earlier decades as an antireflection layer for silicon solar cells [14,15], due to its good optical properties to enhance the light absorption capability because of the high refractive index of silicon solar cells. However, the photocatalytic effect on TiO₂ films whereby it can decompose organic materials under ultraviolet (UV) light [16] deteriorate the photovoltaic properties of solar-cell modules which have been made of organic polymer (ethylene-vinyl acetate: EVA). A considerable amount of literature has been published on sol-gel processes for silicon solar cells, including TiO₂, TiO₂-SiO₂ [14,15] and some Al₂O₃-based Ti doped mixed sol-gel sources have been introduced as well [17,18]. In addition to these metal oxides, ZrO₂ thin films can be another candidate as an antireflection layer with the properties of high refractive index and good thermal stability [19]. Therefore, further investigations are needed to adapt this kind of low-cost materials to the solar cell fabrication processs.

In this work, as a low temperature and non-vacuum method, spin-coating deposition was performed for polymer-metal oxide composite films. TiO₂ is a suitable material for ARC as described above. Due to the TiO₂ photocatalytic effect, however, we have confirmed the photo decomposition of the polymer-TiO₂ composite layer at the first stage, and therefore, we have utilized surface-deactivated TiO₂ nanoparticles (SD-TiO₂), afterwards. Therefore, ZrO₂-polymer (ZrO₂-P) composite films, surface-deactivated TiO₂-polymer (SD-TiO₂-P) composite films, and ZrO₂-polymer composite/surface-deactivated TiO₂-polymer (ZrO₂-P/SD-TiO₂-P) composite films were introduced as alternative anti-reflection coating films for the fabrication of silicon solar cells. The suitable order of the layers according to their refractive indexes is the key point for multilayer structure ARC films [20]. From that point of view, after confirming the structure of TiO₂ and ZrO₂ polymer-based thin films separately, an <air/ZrO₂-polymer/surface-deactivated TiO₂-polymer/Si> multilayer structure was also built considering the gradual increasing order of refractive indexes of approximately 1.5, 2, 4.3 for ZrO₂, TiO₂ and silicon, respectively, where the refractive index of air is equal to 1. Owing to the advantage of this gradual increase of refractive index on the light path, much lower reflectance can be expected [20]. After the experimental evaluations in this work, silicon solar cells were fabricated with and without using ZrO₂-P composite, ZrO₂-P/SD-TiO₂-P multilayer composite films.

2. Experimental

ZrO₂ colloidal nanoparticles (in water, ZR-30AH, Nissan Chemical Industry Co. Ltd., Chiba, Japan), surface-deactivated TiO₂ colloidal nanoparticle (in water, nanoUSE-Ti, Nissan Chemical Industry Co. Ltd.) and TiO₂ nanoparticle (in screen printing paste for dye-sensitized solar cells, PST-18NR, JGC Catalysts and Chemicals Ltd., Kanagawa, Japan) were utilized. The average diameter of particles and the rate of solid components of ZrO₂ nanoparticles, surface-deactivated TiO₂-nanoparticles and TiO₂-nanoparticles are 50 nm (30 wt %), 13 nm (30 wt %) and 18 nm (17 wt %), respectively. Transmission electron microscopy (TEM) images of each nanoparticle are given in Figure 1.

Solutions for composite films were prepared by mixing one of the nanoparticles with ethanol and organic polymer (ethyl cellulose, Figure 2). The ethyl cellulose was a mixture of E0265 (9–11 mPa·s, 5% in toluene + ethanol (80:20) at 25 °C, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) and E0266 (45–55 mPa·s, 5% in toluene + ethanol (80:20) at 25 °C, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan). For the mixing, 50 g of ethyl cellulose (E0265) and 50 g of ethyl cellulose (E0266) were dissolved in 1 L of ethanol, beforehand. The nanoparticle-polymer composite layers were formed on flat or alkaline textured crystalline silicon (c-Si) substrates and primary evaluation of the coating films as an antireflection coating (ARC) film.

After the initial optical analysis of TiO₂ ARC films and the confirmation of the photocatalytic effect of TiO₂, surface-deactivated TiO₂ and polymer based ZrO₂ films were evaluated as ARC. The amount of ethyl cellulose and ethanol were altered for the further analysis. Optimized solutions were named as surface-deactivated TiO₂-polymer (SD-TiO₂-P) composite films and ZrO₂-polymer (ZrO₂-P) composite films. Finally, textured p-type CZ-Si solar cells were fabricated using SD-TiO₂-P composite, ZrO₂-P composite and ZrO₂-P/SD-TiO₂-P composite multilayer structures as an ARC film on the front surface of the wafers and compared to those of solar cells without ARC.

For the optical measurements, square-shaped p-type Si wafers (25 × 25 mm²) were cut out from 6-inch CZ-Si p-type wafers. All wafers were dipped into the 20% diluted HF for 1 min and rinsed in distilled water. UV/O₃ surface treatment was carried out for a complete cleaning process. All thin films in this work were coated by spin-coating method. Spin coating was carried out at 5000 rpm for 25 s (acceleration time is 5 s). All deposited films were annealed at 125 °C for 5 min. Analyses were carried out by scanning electron microscope (SEM) measurements (JSM-6510, JEOL, Tokyo, Japan), reflection analysis (by ultraviolet-visible spectroscopy, Lambda 750 UV/VIS Spectrometer, Perkin Elmer, Waltham, MA, USA) and ellipsometer (UviselErAgms-nds, Horiba Jobin Yvon, Kyoto, Japan) and finally by fabricating c-Si solar cells.

C-Si solar cells were fabricated in four different sets with a variety of surface conditions: without ARC, with SD-TiO₂-P composite, ZrO₂-P composite, or ZrO₂-P/SD-TiO₂-P composite multilayer structured ARC films. For the fabrication of c-Si solar cells, 25 × 25 mm² p-type CZ-Si wafers were used, which were cut out from 6-inch wafers. For surface-textured Si solar cells, alkaline texturing was performed in KOH (5.19 g) solution in H₂O (100 mL) with Alka-Tex (0.28 mL, GP Solar, Konstanz, Germany) at 80 °C (set: 100 °C) for 30 min. The silicon wafers were dipped into the 20% diluted HF for 10 min and rinsed in distilled water. Then, RCA cleaning [21,22] was carried out to remove contaminant particles on the surface of the wafers by using a NH₄OH/H₂O₂/H₂O (1:1:5 in volume) solution, for 10 min at 80 °C. After removal of the natural oxide films by 20%HF, a mixed solution of HCl/H₂O₂/H₂O (1:1:5 by volume) was used to remove metallic contaminants and mobile ions on the surface by dipping the wafers for 10 min at 80 °C. In order to prevent phosphorous diffusion on the back side of the silicon wafer, polysilazane was coated using spin coater by two times at 1500 rpm for 20 s, succeeded by annealing at 600 °C for 60 min in O₂ gas flow. POCl₃ diffusion was performed at 900 °C for 30 min with 0.2 L/min N₂ flow, in order to obtain n+ emitter. Afterwards, wafers were successively rinsed by dipping in 10% HF and pure water. Front and back contacts were formed by screen-printing Ag and Al, respectively. Co-firing was carried out at 780 °C for 1 min. SD-TiO₂-P, ZrO₂-P and ZrO₂-P/SD-TiO₂-P composite multilayer structure were deposited as ARC on the finished cells and compared to those of the cells without ARC. Deposition of ARC films were carried out after the firing step by spin coating process (5000 rpm, 5 s acceleration + 25 s) and subsequent annealing (125 °C, 5 min). For the cells with ZrO₂-P/SD-TiO₂-P composite multilayer structure ARC, first SD-TiO₂-P composite film was deposited and annealed and then ZrO₂-P composite film was deposited and annealed.

3. Results and Discussion

3.1. Evaluation of TiO₂-Polymer, Surface-Deactivated TiO₂-Polymer and ZrO₂-Polymer Composite Films

In order to observe the photocatalytic effect of TiO₂-P film when used for ARC purposes, reflectance analyses were carried out. Special attention was paid to the effect of UV light on short wavelengths and to the long run stability analysis of the TiO₂-P film. Figure 3 shows the reflectance spectra of the silicon wafers coated with TiO₂-P, ZrO₂-P films or SD-TiO₂-P composite films on short wavelengths. Considerable differences were observed on the short wavelength between the reflectance data of each material. Reflectance was measured from 320 nm to 400 nm initially and after 8 h and 102 h from the first measurement. It was observed that the reflectance of the wafers coated with TiO₂-P coated substrate was increased in the short wavelength by the increase of the illuminating time. This increment of reflectance performance by time in shorter wavelengths can be due to the photocatalytic effect of TiO₂. This increment could be eliminated by SD-TiO₂-P in which the surface TiO₂ nanoparticles are deactivated, and then, the decomposition of organic materials under the UV light could be avoided. No significant change of the reflectance was observed on SD-TiO₂-P and ZrO₂-P coated wafers. These results are important to realize the adverse effect of photo catalysis when using TiO₂ films as an ARC. The photo catalysis effect of TiO₂ becomes rather valuable when the use of it considered for low bandgap and multi junction solar cells that the improved absorbance in UV region of the spectrum is important.

On the other hand, the stability of a film under the continuous illumination is needed to be considered when it was aimed to be used for ARC purposes. The time dependence on the average reflectance of the TiO₂-P, ZrO₂-P and SD-TiO₂-P films on silicon substrates were carried out by continually illuminating the films up to 102 h (under constant light illumination, 1 sun, AM 1.5). Dependence of reflectance of each ARC film to the light irradiation time is given in Figure 4 (dots show the measurement values, straight and dotted lines show the fitting for easy understanding). SD-TiO₂-P and ZrO₂-P composite films were relatively stable against light irradiation, with average reflectance of around 7% and 6.5%, respectively. However, the reflectance of TiO₂-P coated wafer increased with the irradiation time and the increase is more significant after 10 h of irradiation.

After the confirmation of the photocatalytic effect of TiO₂-P film, we focused on the SD-TiO₂-P and ZrO₂-P films as ARC layers for silicon solar cells. First, zirconium bare solution was spun on the surface of the textured wafer without any additives with spin speed of 4000 rpm for 20 s (acceleration time is 4 s).

Detrimental cracks were observed between the textured pyramids after drying, as shown in Figure 5a. To avoid these cracks, a mixed solution was prepared with ethanol (1:1 in volume). However, cracks were could not be avoided totally.

In order to achieve a smooth surface without any cracks, ethyl cellulose was added into the bare zirconium solution. Ethyl cellulose is a non-toxic polymer with good mechanical, physical and chemical properties, including high flexibility, high resistance to moisture and excellent elasticity, which makes it a quality film former [23]. Moreover, it dissolves in a wide range of solvents and alcohols, which is important for our solution that contains ethanol [23]. The ratio of ethyl cellulose solution, ethanol and zirconium solution was 1:32:2 (by volume). Spin coating of the latter solution was carried out with similar parameters. Crack-free and smooth surface could be confirmed as shown in Figure 5b (this is the final solution for ZrO₂-P composite film that is used in subsequent experiments). Eliminating the cracks can be attributed to the flexibility and the strength added by ethyl cellulose in the solution. It is also worth mentioning that the good light permeability of the ethyl cellulose makes it an appropriate material to utilize such a solution aimed to be used for ARC material where the transparency is so important. Similarly, surface-deactivated TiO₂ solution was spun on the surface of the textured wafer with a various composition of TiO₂ solution and ethanol and surface condition of the wafers was examined. 1 mL TiO₂ solution mixed with 2, 4, 6 or 8 mL ethanol and spun on the textured silicon wafer with spin speed of 5000 rpm for 25 s (acceleration time is 5 s), as shown in Figure 6a–d. Cracks were observed between the pyramids when the TiO₂ solution was mixed with 2, 4 or 6 mL of ethanol. Surface condition and cracks can be confirmed in the corresponding enlarged images embedded in Figure 6a–c. Fortunately, TiO₂ solution mixed with 8 mL of ethanol combination provided crack-free surface.

For the initial evaluations of the ZrO₂-P films, it was deposited on flat silicon substrate by a spin-on coating method with a spin-speed in a range from 1000 to 4000 rpm for 20 s (acceleration time is 4 s). Figure 7a compares the reflectivity of flat silicon substrates w/ and w/o ZrO₂-P layer coated on the surface. The lowest average reflectance (the reflectance average between the wavelength of 300 nm–1100 nm) of 28% was achieved by ZrO₂ coating with a spin speed of 1000 rpm (Figure 7b), a steady reflectance tendency from wavelength of around 700 to 1000 nm can also be confirmed. Comparing these results it can be confirmed that the average reflectance of a silicon substrate decreased from 40% down to 28% owing to the formed ZrO₂ film on the flat silicon surface.

Figure 8a shows the comparison of the reflectivity of textured silicon wafers covered by ZrO₂-P composite film including different amount of ethanol in the solution. Average reflectance was decreased from 8% down to 6.5%, when 8 mL (91.4%vl) ethanol is used in total zirconium solution (0.5 mL zirconium solution (5.71%vl) + 0.25 mL ethyl cellulose (2.85%vl)). Figure 8b clearly shows that there is a breaking point for the amount of ethanol used in zirconium solution to achieve better reflectance performance which is the main agent to adjust the viscosity.

Reflectance dependence of ZrO₂-P composite film coated alkaline textured silicon substrates on the applied spin speed is given in Figure 9a. Deposition was performed by spin-on coating with a spin-speed in a range of 500–3000 rpm. A gradual decrease of reflectance was observed with increase of spin speed and becomes relatively stable after 1500 rpm. Average reflectance of lower than 7% was confirmed when the spin speeds is 1500 rpm. The dependence of reflectance to the spin speed is given in Figure 9b. These results further support the benefit of ZrO₂-P composite film on decreasing the reflectance when formed on a silicon substrate.

Evaluations of SD-TiO₂ solution were also carried out both on flat or textured silicon substrate. In 1 mL SD-TiO₂ solution, 0.1 mL of ethyl cellulose was added and the amount of ethanol was varied from 4 to 10 mL. Prepared solution was spun on flat silicon surface and annealed on a hot plate at 125 °C for 5 min. Figure 10 shows the reflectivity of flat silicon surface coated with TiO₂ solution with varied amount of ethanol in precursor, with spin speed of 5000 rpm or 8000 rpm. In

Table 1. Average reflectance of flat silicon surface with surface-deactivated TiO₂-polymer ARC coating by different spin-coating speeds. The volume of surface-deactivated TiO₂–sol was 1 mL.

	EtOH (mL)	4	6	8	10
Speeds
5000 rpm		30.00	24.00	19.92	22.66
8000 rpm		25.78	20.83	20.04	24.15

, the average reflectance of flat silicon surface coated with surface deactivated TiO₂-ethyl cellulose film is summarized for each amount of ethanol in the solution.

The lowest average reflectance was observed when the amount of ethanol is 8 mL, as of 19.92% and 20.04% for spin speed of 5000 rpm and 8000 rpm, respectively. Figure 11a presents the reflectivity of textured silicon surface coated with SD-TiO₂-P film. The amount of ethanol was varied similarly from 4 to 10 mL. From the figure it can be confirmed that the coated film with the precursor include 8 mL of ethanol provides the lowest reflectance. The average reflectance tendency depends on the ethanol amount can be seen in Figure 11b where the lowest average reflectance is around 5.6%.

3.2. Simulation and Formation of ZrO₂-Polymer/Surface-Passivated TiO₂-Polymer Composite Multilayer Film

After evaluating the ZrO₂-P and SD-TiO₂-P composite thin films separately, an <air/ZrO₂-polymer/surface-passivated TiO₂-polymer/Si> multilayer structure was also built considering that the reflectance can be reduced significantly with multilayer films [20] where the layers should be deposited with gradual increase order of their refractive indexes, as n_air < n_ZrO2-polymer < n_TiO2-polymer < n_silicon, in this work. The refractive indexes and extinction coefficients of ZrO₂-P and SD-TiO₂-P composite layers achieved by fitting of the ellipsometry measurements are shown in Figure 12a,b, respectively. As such, refractive indexes of air and silicon wafer are 1 and 4.3, respectively.

A simulation study was carried out for the <air/ZrO₂-polymer (ZrO₂-P)/surface-deactivated TiO₂-polymer (SD-TiO₂-P)/Si> structure in order to estimate the short circuit current density (Jsc) density. Fresnel equations were followed by considering the measured refractive indexes of each layer in order to achieve an ideal reflectance of the total structure [24]. Fresnel approach reduces multi number of layer structure to a single imaginary layer by taking account the refractive index and thickness of each layer in the multi structure and is based on the transmittance and the reflectance waves from the corresponding layers.

The thicknesses of the ZrO₂-P and SD-TiO₂-P composite layers were varied from 10 to 100 nm, and “Incident Photon to Current Conversion Efficiency (IPCE)” was assumed 100% for the calculation. Jsc was calculated using the Equation (1) in order to observe the ideal ZrO₂-P and SD-TiO₂-P composite film thicknesses to obtain an efficient ZrO₂-P/SD-TiO₂-P composite multilayer ARC film:

$$J_{sc,max}\le \frac{eS}{hc}{{\displaystyle \int }}_{300}^{1200}{P}_{in}\times Abs\left(\lambda \right) d\lambda$$

(1)

where “Jsc,max” is maximum short circuit current density in mA cm⁻², “e” is stands for charge of an electron (1.602 × 10^–19 C), “h” is Plank’s constant (6.626 × 10^–34 Js), “c” is speed of light (2.998 × 108 m·s⁻¹), “Pin” is the incident power in W/m², and “λ” is the wavelength in nm.

Figure 13 presents the estimated Jsc with ZrO₂-P/SD-TiO₂-P composite ARC film on c-Si solar cells. A summary of the estimated values of Jsc can be seen in

Table 2. Simulated data of photocurrent density with variations of ZrO₂-polymer composite and surface-deactivated TiO₂-polymer composite thickness on flat-surface silicon solar cells.

	TiO2/nm	10	20	30	40	50	60	70	80	90	100
ZrO2/nm
10		29.90	31.07	32.82	34.95	37.02	38.65	39.72	40.27	40.37	40.15
20		30.66	32.12	34.07	36.18	38.03	39.36	40.15	40.46	40.39	40.06
30		31.74	33.42	35.40	37.33	38.88	39.92	40.46	40.58	40.38	39.98
40		33.06	34.81	36.66	38.31	39.55	40.33	40.66	40.65	40.37	39.94
50		34.48	36.13	37.73	39.07	40.03	40.59	40.78	40.68	40.37	39.93
60		35.84	37.27	38.56	39.60	40.33	40.73	40.83	40.69	40.36	39.93
70		37.00	38.13	39.13	39.92	40.47	40.76	40.82	40.66	40.35	39.92
80		37.87	38.72	39.45	40.04	40.47	40.71	40.76	40.60	40.30	39.89
90		38.45	39.04	39.56	40.01	40.37	40.60	40.64	40.50	40.21	39.82
100		38.76	39.13	39.50	39.87	40.20	40.43	40.47	40.35	40.08	39.72

for each thickness combination of the layers.

Maximum J_SC of 40.83 mA cm⁻² was achieved with <t = 60 nm ZrO₂-P/t = 70 nm SD-TiO₂-P> multilayer ARC film. It is notable that increased values of Jsc of ~40 mA cm^-2 become achievable with the thickness of >50 nm of each layer.

To form and characterize such a multilayer film experimentally, first SD-TiO₂-P film was spin coated on the textured silicon surface and annealed at 125 °C for 5 min, then the reflectance of the surface was measured. After the measurement ZrO₂-P film was spin coated on SD-TiO₂-P film and annealed similarly. The comparison of reflectance spectra of textured silicon surface coated with SD-TiO₂-P film or the ZrO₂-P/SD-TiO₂-P composite multilayer film is given in Figure 14. Average reflectance of the silicon surface was improved due to the multilayer ARC where the average reflectance is around 5.5% for multilayer structure (average reflectance without ARC = 16.3%).

Additionally, lifetime analysis was carried out in order to estimate the passivation effect of purposed ARC films. Four sets of samples (three samples per set) were prepared for lifetime studies; TiO₂-P (set 1), ZrO₂-P (set 2), SP-TiO₂-P (set 3), ZrO₂-P/SP-TiO₂-P (set 4). All samples went through an acidic etching of HF:HNO₃ (1:5vl%) for 5 min and then received UV/O₃ treatment for 15 min before the film deposition. ARC films were spin coated on both sides of the wafers at 5000 rpm for 25 s (acceleration time is 5 s) and were annealed at 125 °C for 5 min.

Lifetime measurements were carried out by microwave photoconductive decay (μ-PCD) technique. Bulk lifetime of a reference sample was measured using chemical passivation (ethanol:iodine, 0.025 M) and measured as 527 μs. Samples of each set (neighbor samples cut from the same large area CZ p-type silicon wafer) were measured before (without ARC) and after forming the ARC film. Measured lifetimes are shown in

Table 3. Lifetimes of the wafers measured before and after forming TiO₂-P, ZrO₂-P, SP-TiO₂-P, ZrO₂-P/SP-TiO₂-P ARC films.

	Sample No.	Lifetime/μs (before ARC)	Lifetime/μs (after ARC)	Increment/μs
TiO2-P	1	10.35	10.88	0.53
	2	10.29	10.72	0.43
	3	10.22	10.75	0.53
ZrO2-P	1	10.26	10.02	−0.24
	2	10.1	10.28	0.18
	3	10.22	10.56	0.34
SP-TiO2-P	1	10.22	10.21	−0.01
	2	10.2	10.45	0.25
	3	11.2	11.6	0.4
ZrO2-P/SP-TiO2-P	1	11.06	11.9	0.84
	2	11.22	12.11	0.89
	3	11	11.55	0.55

. Although a slight increase of lifetimes was observed stable for those of the wafers coated with TiO₂-P or ZrO₂-P/SP-TiO₂-P multilayer ARC, however the increase was negligibly small. The maximum lifetime increases of around 0.9 μs were observed on ZrO₂-P/SP-TiO₂-P multilayer ARC coated wafers. These results show that there is no significant passivation effect of our ARC films.

3.3. Evaluation of the Fabricated p-type Crystalline Silicon Solar Cells

After the evaluation of the ARC films, CZ-Si p-type solar cells were fabricated by applying ZrO₂-P, SD-TiO₂-P composite films and ZrO₂-P/SD-TiO₂-P composite multilayer structure films as ARC layer and compared with those of the cells fabricated without any ARC film. The electrical characteristics of the fabricated solar cells with (w/) or without (w/o) ZrO₂-P composite ARC are summarized in

Table 4. Photovoltaic characteristic of fabricated silicon solar cells on flat or textured silicon surface, w/ or w/o ZrO₂-polymer ARC.

ARC	Surface Structure	Jsc/mA cm−2	Voc/V	FF/%	η/%
w/o ARC	flat	24.03	0.527	57.1	7.25
w/ ZrO2-polymer composite	flat	30.22	0.537	60.2	9.78
w/o ARC	textured	28.74	0.511	69.4	10.20
w/ ZrO2-polymer composite	textured	30.62	0.510	69.7	10.89

.

Conversion efficiency of the best cell fabricated on flat silicon substrate w/o ZrO₂-P composite ARC on the surface was 7.25% with Jsc and open circuit voltage (Voc) of 24.03 mA cm⁻², 527 mV, respectively. Conversion efficiency of the cell w/ ZrO₂-P composite ARC, reaches 9.78% with Jsc of 30.22 mA cm⁻², owing to the light trapping by ZrO₂-P composite ARC film. Jsc of the solar cells fabricated on textured silicon substrates increase up to 29.74 mA cm⁻² (w/o ZrO₂-P ARC) and 30.62 mA cm⁻² (w/ ZrO₂-P ARC) due to reduced reflection by textured surface and ZrO₂-P composite ARC film. Conversion efficiencies of the solar cells with textured surface were 10.2% and 10.9 for those cells w/ and w/o ZrO₂-P ARC, respectively. Figure 15 presents the J-V curve comparison of the best cells for each set.

On the other hand, solar cells with SD-TiO₂-P ARC and ZrO₂-P/SD-TiO₂-P composite multilayer ARC were also fabricated. Jsc of 30.93 mA cm⁻² and Voc of 569 mV were achieved when solar cells fabricated w/o any ARC (average of three samples). Average Jsc of the solar cell with SD-TiO₂-P composite film and ZrO₂-P/SD-TiO₂-P composite multilayer film as ARC increased up to 32.44 and 33.00 mA cm⁻², respectively. These increases of Jsc are attributed to the decrease of the reflectance with ARCs which increases the optical performance of the solar cells. Average efficiencies of the three fabricated cells for each set, w/o ARC, w/ SD-TiO₂-P film and w/ ZrO₂-P/SD-TiO₂-P composite multilayer structure film were 11.99%, 12.84%, and 12.85%, respectively.

Table 5. Best and average cell data of photovoltaic characteristics for fabricated silicon solar cells without ARC and with surface-deactivated TiO₂-polymer or ZrO₂-polymer/Surface-deactivated TiO₂ polymer composite ARC films. Average data for each set is for three different samples.

Structure of Solar Cells		Jsc/mA cm−2	Voc/V	FF/%	Rseries/Ω·cm2	Rshunt/Ω·cm2	η/%
Without ARC	Best	29.33	0.570	71.7	0.73	782	11.99
	Average	30.93	0.569	63.1	1.23	406	11.06
Surface-deactivated TiO2 polymer	Best	30.90	0.574	72.3	0.66	548	12.84
	Average	32.44	0.575	63.7	1.18	321	11.84
ZrO2-polymer/Surface-deactivated TiO2 polymer	Best	31.42	0.575	71.5	0.75	504	12.91
	Average	33.00	0.574	62.7	1.32	309	11.85

summarizes the electrical characteristics of the fabricated solar cells. Efficiency of the solar cells improved by a factor of 0.86% with an increase of Jsc (2.07 mA cm⁻²) by the contribution of ZrO₂-P/SD-TiO₂-P ARC composite multilayer compared to those of fabricated without the ARC. The best efficiency of 12.91% for the cells with ZrO₂-P/SD-TiO₂-P composite multilayer structure ARC film was achieved with Jsc of 31.42 mA cm⁻², Voc of 575 mV, and fill factor (FF) of 71.5%. The J-V curve comparison of the best cells for each set is given in Figure 16. The higher average Jsc of the cells than that of the best cells in spite of lower FF may draw the attention which can be attributed to the ARC effect. Because the lower series resistances of the best cells might be due to the lower contact resistance owing to the better contact formation between the metal contacts and silicon during the firing steps which lead higher FF.

The improvement of Jsc by SD-TiO₂-P or ZrO₂-P/SD-TiO₂-P multilayer ARC composite can be clearly seen in Figure 17, where the Jsc increases gradually for all samples by SD-TiO₂-P single layer ARC and ZrO₂-P/SD-TiO₂-P multilayer ARC composite. Jsc increases (average of four samples) from 30.04 to up to 32.07 mA cm⁻². In order to discuss the statistical errors about the resulting silicon solar cells, the number of samples was not enough because just four cells have been fabricated for Figure 17. However, every cell improved the Jsc with sequential addition of the TiO₂-P and ZrO₂-P composite coating on the silicon surface. Hence, it was enough to understand the merit for the polymer composite layers to be the antireflection coatings.

It should be noted that the limited overall cell performance in this work, compared to commercially available, fully state-of-art solar cells can be mainly attributed to the low lifetime of silicon substrates and inadequate bulk passivation which can cause poor Voc and to the low FF s which can be due to the lack of proper edge isolation of the finished cells. These results suggest that ZrO₂-P/SD-TiO₂-P composite multilayer structure ARC film formed by spin coating deposition technique could be beneficial and promising alternative as a low-cost, simple and vacuum-less process for ARC for silicon solar cells. Further investigations are on progress to achieve better performances on ARC utilizing polymer based ZrO₂ composite layers and its multilayer structures.

4. Conclusions

ZrO₂-polymer composite/surface-deactivated TiO₂-polymer composite films were prepared by a low-cost spin coating deposition method as an alternative ARC film for c-Si solar cells. The deteriorative effect of the photocatalytic properties of TiO₂ on reflectance performance of silicon wafers coated with TiO₂-polymer composite film was confirmed. No degradation was observed on ZrO₂-polymer composite film and surface-deactivated TiO₂-polymer composite films coated wafers for the same UV region. Good quality of ZrO₂/TiO₂ based multilayer thin films was observed for films formed at a very low temperature of 125 °C. After the evaluation of thin films, CZ-Si p-type solar cells were fabricated w/ and w/o ARC. The best efficiency of 12.91% for the cells with ZrO₂-polymer composite/surface-deactivated TiO₂ polymer multilayer structure composite ARC film was achieved with Jsc of 31.42 mA cm⁻², Voc of 575 mV, and fill factor (FF) of 71.5%. Efficiency of the solar cells improved by a factor of 0.86% with an increase of Jsc (2.07 mA cm⁻²) owing to the ZrO₂-polymer/surface-deactivated TiO₂ polymer composite ARC layer compared to those of fabricated without the ARC. Considering the non-vacuum process and simplicity, polymer-based ZrO₂/TiO₂ ARC films produced by the spin coating deposition method were found to be very attractive and promising for c-Si solar cells with good optical performance which can be improved by further optimizations and investigations.