1 Introduction
Given the limited supply of conventional energy and increasing environmental pressures, many countries around the world have launched a surge in the development and utilization of new energy sources. Among these new energy sources, solar energy—a permanent energy source that continuously replenishes the Earth—is particularly noteworthy. Solar energy is a clean, pollution-free, and inexhaustible natural energy source. Converting solar energy into electricity is a crucial technological foundation for its large-scale utilization, and it is highly valued by countries worldwide. In 1955, Bell Labs in the United States successfully developed the first practical silicon solar cell, which was subsequently used in artificial satellites. my country began research on solar cells in 1958, successfully applying it to its second satellite launch in 1971, and starting ground applications in 1973. In recent years, the photovoltaic market has developed extremely rapidly, with crystalline silicon solar cells becoming the dominant product, accounting for over 80% of the international market share in 1997. However, currently, most of the silicon materials used in solar cells are derived from off-the-shelf semiconductor silicon materials and the leftovers from monocrystalline silicon, which cannot meet the needs of the photovoltaic industry's development. Meanwhile, silicon is a difficult component to reduce in terms of cost for crystalline silicon solar cell modules. Therefore, to meet the needs of high-efficiency, low-cost, and large-scale production of solar cells, the most effective approach is to avoid the traditional process route from silicon raw materials, ingots, and wafers to solar cells, and instead adopt a direct process route from raw materials to solar cells—that is, developing thin-film solar cell technology. Since the 1970s, many new materials for making thin-film solar cells have been developed, such as CuInSe, CdTe films, and organic films; in the past 20 years, numerous researchers have achieved remarkable results in this field. Thin-film solar cells, with their advantages of low cost, high conversion efficiency, and suitability for mass production, have attracted the interest of manufacturers, leading to rapid growth in their production volume. It is precisely to further reduce the cost of crystalline silicon solar cells that photovoltaic researchers worldwide have developed crystalline silicon thin-film batteries in recent years. Polycrystalline silicon thin-film batteries combine the advantages of crystalline silicon cells—high efficiency, stability, non-toxicity, and abundant resources—with the advantages of thin-film batteries—simple processing, material savings, and significantly reduced costs. Therefore, the research and development of polycrystalline silicon thin-film batteries has become a hot topic in recent years.
2. Battery working principle
Polycrystalline silicon thin-film solar cells are produced by growing polycrystalline silicon thin films on low-cost substrate materials. A relatively thin crystalline silicon layer serves as the activation layer, maintaining the high performance and stability of crystalline silicon solar cells while significantly reducing material usage and lowering overall cell costs. The working principle of polycrystalline silicon thin-film solar cells is the same as other solar cells: the photovoltaic effect is formed by the interaction of sunlight and semiconductor materials. The interaction between light and semiconductors generates photogenerated carriers. When the generated electron-hole pairs are separated by the potential barrier formed within the semiconductor to their respective electrodes, an electric potential is generated between the electrodes; this is called the photovoltaic effect.
3. Battery structural features
In semiconductor solar cells, the thickness of the semiconductor film necessary to absorb solar energy can be extremely thin. For silicon, the absorption value is on the order of 10⁴/cm in the region near the peak of the solar spectrum, between 5.0 × 10⁻⁷ m and 6.0 × 10⁻⁷ m. In principle, a thickness of a few μm can absorb most of the energy, but in practice, the thickness of polycrystalline silicon thin films is typically 50 μm. For this reason, thin-film solar cells have been developed. The thinning of solar cells aims to reduce the manufacturing cost of terrestrial solar cells and save on expensive semiconductor cell structural materials. A substrate is needed to mechanically support the active layer of the cell thin film. Naturally, the substrate material should also be inexpensive. Therefore, in most cases, the substrate is not a semiconductor material. The semiconductor thin film formed on the substrate is polycrystalline or amorphous, not necessarily a single crystal. Semiconductor thin films on substrates can be formed through various methods: physical and chemical growth methods, as well as methods such as immersing the substrate in molten semiconductor material. The conversion structure of thin-film cells is the same as that of single-crystal cells, including pn junction, Schottky, MIS, and heterojunction types. The differences lie in the influence of the substrate on the semiconductor thin film formation process, the role of grain boundaries and film thickness, and the unique material and electrical properties of thin films—all of which cannot be ignored. Due to these constraints, the characteristics of silicon thin-film solar cells still lag behind those of monocrystalline silicon solar cells, remaining in the experimental stage and not yet ready for practical application.
4-cell battery configuration
4.1 Battery Structure on Insulating Substrate
This is an n+-p-p+-Al substrate. Because the substrate is an insulator, the electrode on the p+ side needs to be removed. The characteristics of its Si layer are: p+ layer: 20μm~40μm thick, resistivity 10-3 ohm-cm; p layer: 5μm~20μm thick, impurity concentration 10¹⁶/cm³; n layer: 0.14μm~4μm thick, impurity concentration 10¹⁹/cm³. The p layer and n+ layer are grown using SiHCl₃ epitaxial growth, aluminum electrode is vacuum evaporated, and an anti-reflection film is evaporated to form a 4cm~10cm solar cell.
4.2 Battery Structure on Graphite Substrate
The structure of the silicon thin-film solar cell with graphite as the substrate is the same as that in section 1. The silicon layer characteristics are as follows: p+ layer: 10μm～40μm thick, resistivity 2～3×10-8 ohm-cm; p layer: 8μm～10μm, resistivity: 0.2～2 ohm-cm; n+ layer: 0.2μm～0.4μm thick, resistivity 1～2×10-3 ohm-cm; graphite substrate 3cm×3cm.
5 Battery Characteristics
6. Basic Requirements for Thin Films in Polycrystalline Silicon Solar Cells
The basic requirements for fabricating solar cells using polycrystalline silicon thin films are:
(1) The thickness of the polycrystalline silicon thin film is 5μm to 150μm;
(2) Increase photon absorption;
(3) The width of the polycrystalline silicon thin film is at least twice its thickness;
(4) The minority carrier diffusion length is at least twice the thickness;
(5) The substrate must have mechanical support capabilities;
(6) Good back electrode;
(7) Passivate the back surface;
(8) Good grain boundaries.
7. Preparation method
7.1 Semiconductor Liquid Phase Epitaxial Growth Method (LPE Method)
LPE (Lithography-Enhancing) growth technology has been widely used to grow high-quality and compound semiconductor heterostructures, such as GaAs, AlGaAs, Si, Ge, and SiGe. LPE can be grown on both planar and non-planar substrates, yielding materials with perfect structures. In recent years, the use of LPE technology to grow crystalline silicon thin films for the fabrication of high-efficiency thin-film solar cells has attracted widespread interest. LPE growth allows for doping, forming n-type and p-type layers. LPE growth equipment is general-purpose epitaxial growth equipment, with growth temperatures ranging from 300℃ to 900℃, growth rates from 0.2 μm/min to 2 μm/min, and thicknesses from 0.5 μm to 100 μm. The morphology of the epitaxial layer depends on the crystallization conditions, and epitaxial layers with textured surfaces can be directly obtained.
7.2 Zone Melting Recrystallization (ZMR) Method
A SiO2 layer is formed on silicon (or other inexpensive substrate material), and a silicon layer (3μm to 5μm, grain size 0.01μm to 0.1μm) is deposited on it using LPCVD (low-pressure chemical vapor deposition). This layer is then subjected to zone melting recrystallization (ZMR) to form a polycrystalline silicon layer. By controlling the ZMR conditions, the density of etch pits in the recrystallized film can be reduced from 1×107cm-2 to 1.2×106cm-2, while the (100) crystal phase area rapidly increases to over 90%. To meet the requirements of photovoltaic cells for layer thickness, a silicon layer with a thickness of 50μm to 60μm is grown on the ZMR layer using CVD as an activation layer. Scanning heating is used to increase its grain size to several millimeters, thereby forming an insulating silicon (sol) layer. The activation layer is p-type with a resistivity of 1Ω•cm to 2Ω•cm. To obtain a high-quality activation layer, the surface of the ZMR layer is treated with HCl before LPCVD. To prepare polycrystalline silicon thin-film solar cells, an active layer is etched to form a textured structure, and n-type impurities are diffused to form a pn junction. Then, surface passivation and anti-reflection layer are deposited, and electrodes are prepared. The back electrode is fabricated by back etching and hydrogenation, thus producing a polycrystalline silicon thin-film solar cell.
7.3 Plasma Spraying (PSM)
A DC-RF hybrid plasma system was employed, using p-type crystalline silicon with a purity of 99.19999% and a particle size of 50μm–150μm as the raw material. Ar gas was used as the carry gas, and the deposition was performed by DC-RF plasma. The raw material storage box and carry gas pipeline were coated with Si₂C₂N₂O compound to prevent impurity contamination. The silicon powder was heated and melted in the high-temperature plasma, and the molten particles were deposited on the substrate. The substrate was heated by a heater, and its temperature was maintained at 1200℃ using infrared thermocouples before deposition. The deposition chamber was made of stainless steel and evacuated using an oil-free pump to a vacuum level of 1.33 × 10⁻² Pa. The plasma consisted of Ar and a small amount of H, with a deposition pressure of 8 × 10⁻⁸ Pa. The thickness of the deposited polycrystalline silicon film ranged from 200μm to 1000μm. The polycrystalline silicon grain size ranged from 20μm to 50μm, and the deposition rate was greater than 10μm/s. Polycrystalline silicon thin-film solar cells were deposited using plasma spraying, employing a low-temperature plasma CVD process. The deposited polycrystalline silicon layer was etched with an alkaline or acidic solution, and a 200 × 10⁻⁸ cm thick microcrystalline silicon layer was formed on top at 200°C using plasma CVD as the emitter layer. An ITO antireflective layer and silver paste electrodes were then fabricated to complete the solar cell. With an area of 1 cm², under AM1.5 and 100 mW/cm² conditions, the cell conversion efficiency was η = 4.3%.
7.4 Layering Method
Polycrystalline thin films were deposited on pre-fluorinated glass substrates at a relatively low temperature of 300°C using a stacking technique, similar to the A-Si:H thin film method. Large-area polycrystalline silicon thin films were deposited at low temperatures using plasma-enhanced chemical vapor deposition. Typically, p-type doped polycrystalline silicon thin films were deposited using a stacking technique, with thicknesses ranging from 0.28 mm to 5.78 mm. Typical deposition conditions were: SiF4 flow rate 60 sccm, hydrogen flow rate 15 sccm, deposition temperature 300°C, microwave power 200 W, and pressure 53.3 Pa. For p-type doping deposition, 10 ppm PH3 was mixed into the hydrogen gas at a flow rate of 18 sccm. Each deposition and atomic hydrogen treatment time was 10 s. Because the PH3 and original SiF4 used for doping were added to the hydrogen plasma region during deposition, the P to Si ratio in the film could be better controlled. Carrier transport characteristics were determined using Hall effect and conductivity measurements within a temperature range of 100 K to 400 K. Experiments show that the material structure is a function of the film thickness, and the Hall mobility increases with increasing film thickness. The region with the highest mobility in the sample is near the film surface. Carrier conductivity is determined by the grain boundary barrier.
7.5 Chemical Vapor Deposition (CVD)
Silicon films of 3 μm to 5 μm thickness were deposited on an alumina-ceramic substrate using chemical vapor deposition (CVD). To obtain high-quality silicon films, a Si3N4/TiO2 (650 × 10⁻⁸ cm) bilayer antireflection film was pre-deposited on the alumina-ceramic substrate. Boron doping was introduced during silicon film deposition. The deposited silicon film was melted using a CW-Ar laser beam and recrystallized at 400–500 °C in a nitrogen atmosphere. For the fabrication of thin-film solar cells, p-diffusion and ITO film deposition were performed using conventional methods, and crystal defects were passivated using hydrogen plasma. Alternatively, the cell could employ a MgF2 (1.0 × 10⁻⁸ cm)/TiO2 (650 × 10⁻⁸ cm) bilayer antireflection film, with the MgF2 layer deposited by electron beam evaporation and the TiO2 layer deposited by atmospheric pressure CVD. The solar cell prepared by this method has a thickness of 4.2 μm, a short-circuit current of 25.2 mA/cm2, an open-circuit voltage of 0.48 V, an FF of 0.53, and an η of 6.52%.
7.6 Solid-phase crystallization (SPC)
The initial α-Si material is deposited on a planar or textured substrate using SiH or Si-H glow discharge. PH3 is added during deposition to form a p-doped layer, which enhances the crystal nuclei and promotes the formation of larger crystal nuclei. The typical thickness of the p-doped layer is 170 nm, upon which an undoped α-Si layer is deposited. The structure of the undoped α-Si layer is altered by changing deposition conditions, such as pressure and RF power. After deposition, the α-Si layer is annealed at 600°C in a vacuum to allow solid-state crystallization, forming polycrystalline silicon. Raman spectroscopy is used to study the relationship between the undoped α-Si structure and the polycrystalline silicon film. Secco etching exposes the grain boundaries, and scanning electron microscopy is used to measure the grain size and density. The polycrystalline silicon thin-film solar cell fabricated using the SPC method described above has the following structure: a tungsten substrate, an n-type polycrystalline silicon layer with a thickness of ~10 μm after SPC, p-type α-Si and p-type α-Si layers with a thickness of ~10 μm deposited on the n-type polycrystalline silicon, an ITO film of ~70 nm deposited on the p-type α-Si, and a metal electrode deposited. The fabricated polycrystalline silicon solar cell has an area of 1 cm², a conversion efficiency of 6.3%, a collection factor of 51% at a wavelength of 900 nm, a minority carrier diffusion length of 11 μm, and a maximum short-circuit current of 28.4 mA/cm². The p-type doping layer has a p-doping depth greater than 10²⁰ cm⁻³.
7.7 Comparison of growth methods and characteristics
8. Current Status and Trends in the Development of Crystalline Silicon Thin-Film Batteries [4]
In recent years, crystalline silicon thin-film solar cells have developed rapidly abroad. To commercialize crystalline silicon thin-film solar cells and translate laboratory results into market applications, a 100 cm² thin-film solar cell was manufactured in 1998 with a conversion efficiency of 8%. Eighteen months later, its efficiency reached 10.9% for the same area, and three years later, a 12 kW thin-film solar cell system was put on the market. In late 1994, a 17.11 kW silicon thin-film solar cell array system was successfully established in California, USA. The cells in this system were fabricated using a high-temperature thermal decomposition spraying method. An anti-reflective layer was coated on the thin-film cells. The silicon thin film grains are millimeter-scale, possessing macroscopic structure, which reduces the response to blue and far-infrared light. The 26th IEEE PVSC, the 14th European PVSEC, and the World Solar Energy Congress held in 1997 reported on the United Solar System thin-film silicon solar cell with a conversion efficiency of 16.6%. Japan's Hanebo achieved 9.8%, and test results provided by NREL in the United States showed that USSA's Si/SiGe/SiGe thin-film cell, with an area of 903 cm2, had a conversion efficiency of 10.2% and a power of 9.2 W.
Research on crystalline silicon thin-film batteries in my country is still in the laboratory stage. In 1982, Han Guilin et al. from the Changchun Institute of Applied Chemistry prepared crystalline silicon thin-film batteries using CVD and studied the growth law and basic physical properties of polycrystalline silicon thin films. The specific preparation process is as follows: high-frequency heating of graphite is used in the system. After the system is evacuated, neon gas is introduced to remove residual gas. The graphite is heated to the required temperature, and then a mixed gas is introduced. At 1100℃～1250℃, SiCl4 is reduced by H2, causing silicon to be deposited on the substrate. In 1998, Zhao Yuwen et al. from the Beijing Solar Energy Research Institute reported that SiH2Cl2 was used as the raw material gas, and polycrystalline silicon thin films were deposited in a quartz reactor using rapid thermochemical vapor deposition (RTCVD). The growth characteristics and microstructure of the thin film were studied, and a polycrystalline silicon thin-film solar cell was developed. The cell structure is metal grid/p+ polycrystalline film/n polycrystalline silicon film/n+C- silicon/metal contact. Boron diffusion was used to form a p+ layer with a junction depth of approximately 1 μm. The cell area was 1 cm². Under AM115 and 100 mV/cm² conditions, without an antireflective coating, the cell conversion efficiency was 4.54%, Jsc = 14.3 mA/cm², V∞ = 0.460 V, and FF = 0.67. The specific preparation process conditions are as follows: the gas source was a mixture of H₂ and SiH₂Cl₂, and quartz...
The tube contains a graphite sample holder, which is heated to 1200℃ using a programmable light source. The substrate used in the experiment was a heavily phosphorus-doped inactive single-crystal silicon wafer or a non-silicon substrate, with a thin film growth rate of 10 nm/s at 1030℃. my country's research level on crystalline silicon thin-film solar cells lags significantly behind international levels and should be accelerated. The research aims to form high-quality polycrystalline silicon thin films on inexpensive substrates, study the interlayer between the substrate and the silicon film to prevent impurities from diffusing into the silicon film, and develop polycrystalline silicon thin-film solar cells with high conversion efficiency. The goal is to achieve a conversion efficiency of around 10% in the near future, preparing for industrial production and reducing costs to around $1/W.
9. Conclusion
In conclusion, polycrystalline silicon thin-film solar cells hold immense promise for improving solar cell efficiency, saving energy, and significantly reducing costs. However, due to insufficient research on polycrystalline silicon thin-film materials, imperfect film growth technology, and the inherent limitations of polycrystalline thin-film methods, my country's polycrystalline silicon thin-film solar cell technology remains largely in the laboratory stage. Practice has proven that solving scientific challenges requires long-term, persistent effort, not short-sighted actions. As long as we approach the problem from a strategic perspective and persevere, we will surely reach the shore of success.
