The Most Important Types of Solar Cells and the Attendant Manufacturing Methods

This section describes some of the most important types of solar cells in use today, as well as the most widely used manufacturing processes for silicon solar cells.

3.5.1 Crystalline Silicon Solar Cells

As the structure and function of crystalline silicon (c-Si) solar cells were discussed in Section 3.1 (Figure 3.10), the present section mainly describes the manufacturing process for such cells.

In order to achieve full absorption of all photons where h ■ n > Eq, the crystalline silicon solar cells must allow for at least a 100 pm path of light through the silicon. For reasons of mechanical stability (a key factor during the manufacturing process in particular), material thicknesses ranging from 150 to 300 pm are often used, so as to ensure that the condition for full radiation absorption is met.

The base material for silicon manufacturing is silicon dioxide (SiO2), which occurs abundantly in nature in the form of quartz sand or as large quartz crystals, and is reduced in electric furnaces using charcoal, resulting in metallurgical silicon with approximately 98% purity. After being ground very finely, this raw silicon is reacted with hydrochloric acid (HCl), resulting in trichlorosilane (SiHCl3), whose boiling point in liquid form is 31 °C and which is ultra-purified via repeated distillation. In a final step,


Figure 3.37 Left: base material for silicon manufacturing, quartz sand, which is abundantly available in many deserts around the world. Right: polycrystalline silicon (Photo: Fabrimex/Arco Solar/Willi Maag)

ultrapure polycrystalline silicon is obtained via gaseous SiHCl3 and hydrogen (H2) using a reactor that is heated electrically to 1000-1200 °C. Figure 3.37 shows the base material (sand) and the polycrystalline silicon that is made from it.

Monocrystalline solar cells are made by melting polycrystalline silicon in a crucible in the presence of inert gas. After a crystal nucleus attached to a tension rod has been immersed in the molten silicon, a monocrystal is drawn out of the fused silicon via continuous rotation of the rod (Czochralski method; see Figure 3.38). The maximum drawing speed is around 30cm/h.


Figure 3.38 Pulling a silicon monocrystal (sc-Si) using the Czochralski (CZ) method for monocrystalline solar cell manufacturing (Photo: Fabrimex/Arco Solar/Willi Maag)


Figure 3.39 Finished round silicon monocrystal (sc-Si) during a quality test (Courtesy of DOE/NREL)

Hence the monocrystal manufacturing process is slow, energy intensive and expensive. To obtain a high packing factor (PF) in today’s solar cell modules, the originally round monocrystal thus produced (see Figure 3.39) is processed in such a way that a virtually square rather than round form is obtained, whereupon the monocrystal is sliced into wafers about 0.15 to 0.3 mm thick using a wire saw (see Figure 3.40). In this process, by virtue of the thickness of the saw wire, a considerable portion of thecost – intensive monocrystal is transformed into sawdust rather than a wafer.

The manufacturing process for polycrystalline or multicrystalline solar cell wafers is considerably simpler and not nearly as energy intensive. The ultrapure polycrystalline silicon is simply cast into square blocks. The polycrystalline silicon rods thus obtained already have the desired square shape and can be sliced into polycrystalline wafers using a wire saw (see Figure 3.41). Because more lattice imperfections occur at crystallite grain boundaries that promote recombination in the charge carriers generated in the solar cells, polycrystalline solar cells are somewhat less efficient.

Using these monocrystalline or polycrystalline wafers, solar cells are then manufactured via numerous processes, an overview of whose number and complexity is shown in Figure 3.42.

Owing to the 100 mm minimum cell thickness that is required for crystalline solar cells and the material loss resulting from sawing the silicon rods, manufacturing these cells is extremely energy intensive.


Figure 3.40 Completed monocrystal (now almost square) after a few wafers have been sliced from it using a wire saw. The almost completely severed wafers are visible on the right edge of the monocrystal (Photo: Fabrimex/Arco Solar/ Willi Maag)


Figure 3.41 Polycrystalline or multicrystalline block (mc-Si) after being sawn into rectangular segments, from which 40 wafers are then sliced as shown in Figure 3.40 (Photo: DOE/NREL)

The polycrystalline solar cell manufacturing process is simpler (no monocrystal drawing is required) and thus less energy intensive. For more on the energy used for solar cell manufacturing, see Chapter 9.

In view of the relatively low voltage of individual solar cells, in order to achieve reasonable voltage values a large number of cells need to be series connected (see Chapter 4) via a process that for crystalline solar cells is relatively labour intensive and has yet to be fully automated.

In the interest of avoiding material loss occasioned by silicon rod sawing, processes have been developed that allow polycrystalline bands to be drawn directly from fused silicon. One such process is known as the EFG process (see Figure 3.43). Figure 3.44 shows an EFG installation at ASE, an American company that makes commercial solar modules using wafers realized via the EFG process. Here, thin cylindrical octagonal tubes are pulled rather than massive monocrystals, whereupon wafers are produced from the eight even lateral surfaces.

Updated: August 4, 2015 — 2:44 am