[Image above] A free-standing 10-µm-thick black silicon wafer (left) and a scanning electron microscope image of its nanotextured surface (right). Credit: Garín et al., Small (CC BY 4.0)
By Laurel Sheppard
The cost of manufacturing solar panels has decreased significantly over the last four decades, which has led to more people being able to afford solar installations. But with most of the easily achieved cost-saving approaches already in effect, identifying further cost reductions is becoming more challenging.
Reducing the thickness of silicon wafers is one area of production that remains a viable pathway to achieving further cost reductions. Slimmer wafers mean more wafers can be produced from a single silicon ingot.
Currently, average wafer thickness is about 155 μm. Researchers have successfully reduced thickness to as low as 40 μm without decreased performance. However, because silicon is a poor absorber of light in the long wavelength region, further reducing the wafer thickness jeopardizes light absorption.
Black silicon offers a way to reduce thickness without jeopardizing performance. Black silicon refers to silicon that has been modified to have a needle-shaped surface structure. This textured surface reduces the inherent high reflectivity of silicon, making the material appear black when viewed. (Hence the name.)
Black silicon was initially discovered as an undesirable side effect of reactive ion etching. But researchers now desire black silicon for its good light-trapping ability, which allows the material to absorb light across a broader range of wavelengths. (A bonus in areas with low-light conditions.)
Several etching methods are used to make black silicon. However, for high-efficiency ultrathin solar cells, deep reactive ion etching (DRIE) at cryogenic temperatures appears to be the most promising method. That is because minimum surface damage occurs during this process, plus only a small amount of silicon is wasted to produce the textured surface.
In a recent open-access paper, researchers in Spain and Finland explored using black silicon to create ultrathin substrates for application in interdigitated back-contact (IBC) solar cells.
The researchers come from Universitat Politècnica de Catalunya in Spain and Aalto University in Finland. They focused on IBC solar cells because this configuration can potentially achieve higher efficiency by moving all or part of the front contact grids to the rear of the device, which reduces shading on the front of the cell.
To create the ultrathin substrates, the researchers first had to address the fact that ultrathin silicon wafers cannot be directly processed by standard equipment; the wafers require temporary mechanical support. So, for all samples processed in this study, the researchers bonded the ultrathin substrates to standard 4-inch polished crystalline silicon wafers during the DRIE process using standard positive photoresist.
“This provided good thermal contact, as required by the cryogenic DRIE process, while allowing us to easily detach the substrates after the processing,” the researchers write.
They successfully produced substrates with three different thicknesses using this method: 40, 20, and 10 µm. However, when processing the 20 and 10 µm substrates, bubbles would sometimes become trapped between the substrate and the support wafer, causing the substrates to break or detach. To avoid this fate, the researchers engraved a square array of shallow grooves on the support wafer, which provided a pathway for trapped gases to escape.
Compared to nontextured silicon substrates, all the black silicon substrates demonstrated improved absorption capabilities up to nearly 100% across the entire ultraviolet–visible spectrum.
Additionally, absorption capability extended toward the infrared spectrum. In the case of nontextured substrates, absorption started to decrease at about 600 nm for the 10-µm thick wafers and about 800 nm for 40 µm-thick wafers. In contrast, for the black silicon substrates, absorption started to decrease at about 800 nm for the 10-µm thick wafer and about 925 nm for the 40-µm thick wafer.
When a back reflector was added to the setup, which bounces the light back into the cell for possible absorption, absorption capability was further extended toward the infrared spectrum in all cases. The researchers write that the absorption values reported with the back reflector “should be indicative of the absorption potential for a finished cell in the best-case scenario.”
Encouragingly, the results that the researchers reported matched the so-called Yablonovitch 4n2 absorption limit.
The 4n2 limit defines the maximum absorption enhancement that can be achieved through light trapping. It states that, with ideal surface texturing and ideal back reflectance, absorption in a low-absorbing material can be enhanced up to a factor of 4n2 with respect to a single-pass absorption, with n being the refractive index of the material.
The researchers note that the experimental reflectance with a back reflector was close to the 4n2 limit even with just the front surface of the black silicon wafer being textured.
Given these promising results, the researchers created a proof-of-concept IBC solar cell using the 40-µm black silicon substrates. The cell included vanadium oxide and laser processed phosphorus-doped silicon carbide stacks as the hole and electron transport layers, respectively. It also featured a surface passivation layer made of aluminum oxide and silicon carbide.
The black silicon-based cell achieved a power conversion efficiency of 16.4%, an open-circuit voltage of 633 mV, a short-circuit current density of 35.4 mA cm−2, and a fill factor of 73.4%. In contrast, a cell without black silicon achieved an efficiency of 11.5%, an open-circuit voltage of 600 mV, a short-circuit current density of 27.1 mA cm−2, and a fill factor of 70.7%.
Based on the optical characterization of the wafer, the cell did not achieve the expected external quantum efficiency, i.e., the conversion efficiency of incident photons to electrical current. The researchers attribute this poorer-than-expected performance to several aspects of the device fabrication process, namely
- Successive cleaning steps caused additional etching of the substrate, changing the desired nanostructure.
- The multiple layers covering the textured substrate reduced the refractive index contrast and impedance mismatch, thus reducing the substrate’s scattering effect.
- Thin film deposition of the layers filled in gaps between the silicon nanopillars, thus smoothing out the textured surface.
The researchers note that these negative effects could be minimized through further process optimization.
The open-access paper, published in Small, is “Black ultra-thin crystalline silicon wafers reach the 4n2 absorption limit–Application to IBC solar cells” (DOI: 10.1002/smll.202302250).