The PV industry relies on multicrystalline and monocrystalline silicon wafers to manufacture solar cells. Together they represent nearly 90% of all wafer substrate material used in the industry. Due to different grain orientations within the same wafer, alkaline etching cannot be used to texture multicrystalline silicon, as this would result in non-uniform texturing on the surface as different grains etch at different rates. Monocrystalline silicon wafers with  orientation are the most common type of monocrystalline wafer in the industry because it can be easily textured using an alkaline etchant, for example KOH. Silicon crystallises in a diamond cubic lattice (two inter-penetrating face-centred cubic lattices) and is depicted in Fig. 1. The blue, green and red lines in Fig. 1 represent the ,  and  planes, respectively.
Figure 1 Representation of the diamond cubic lattice of a silicon crystal and the representation of the different planes as indicated by coloured lines.
Alkaline etchants etch  silicon surfaces much quicker than  silicon surfaces, which is the basis for the anisotropic etching process used to make pyramid texture. The main difference between saw damage etching and texturing is the etch rate. To increase the anisotropy of the process the etch rate needs to be low, i.e. 2 µm/min or lower. To accomplish lower etch rates, either the process temperature can be lowered and/or the etchant concentration decreased. For instance, a typical texturing recipe which uses a KOH concentration of 1-2% (in comparison to a 30-40 % concentration in saw damage removal) at 70-80°C. The result is a surface populated with randomized square-based pyramids where the sides are formed by  planes and the base is the  plane. This is depicted in Fig 2. In reality, the etched pyramids are not perfect square-based tetrahedrons with a base angle, a, of 54.74°. For most industrial texturing processes a is between 49 and 53°. This is because the tip of the pyramid is etched for the longest duration.
The texturing solution also includes isopropanol (or another industrial additive). Isopropanol acts as a surfactant which enhances surface wetting and ensures that H2 gas (released by the etching) does not stick to the surface. If isopropanol is not used, round “hillocks” can form due to H2 bubbles blocking etch rate at the surface. Isopropyl reduces surface tension and allows H2 bubbles to release from the surface more easily.
There are many factors that contribute to the quality of texturing:
- The texturing result depends on the initial surface.
- The process is sensitive to the presence of residual silicates from saw damage etching.
- The balance between pyramid nucleation and pyramid destruction.
- Over etching can lead to the destruction of pyramids.
- The evaporation of Isopropanol occurs after the bath temperature reaches 90 °C.
- Isopropanol has a wetting function – stops H2 bubbles from sticking to the surface.
- Ventilation is important, but can affect isopropanol evaporation rate
- Typical process durations are 15-20 mins, therefore the evaporation rate must be monitored.
- Batch circulation – bubbling with N2 can help keep bath components well mixed.
Correct texturing is important because the surface texture is directly related to the ability of the solar cell to harvest light and to generate current. Texturing the surfaces improves the cell current via three distinct mechanisms.
- Reflection of light rays from one angled surface onto another improves the probability of absorption.
- Photons refracted into the silicon will propagate at an angle, increasing their effective path length within the cell, which in turn increases the chance of electron-hole pair generation.
- Long-wavelength photons reflected from the rear surface encounter an angled silicon surface, improving the chance of being internally reflected (light trapping)
A good texturing should lead to lower reflectivity for the whole visible wavelength range.
 Baker-Finch & MacIntosh (2012), Reflection distributions of textured monocrystalline silicon: implications for silicon solar cells, PIP, DOI: 10.1002/pip.2186. Available: https://onlinelibrary.wiley.com/doi/abs/10.1002/pip.2186