Perovskite-Silicon Solar Cell Hits 30% Efficiency Record

A standard silicon solar cell wastes most of the sunlight that hits it. A-level Physics students learn that a photon must have energy equal to or greater than the semiconductor's band gap to generate an electron-hole pair. For silicon, the band gap is about 1.1 electron volts (eV). Photons with less energy pass straight through. Photons with more energy do generate charge carriers, but the excess energy above 1.1 eV is lost as heat. This thermalisation loss is the single biggest reason why a silicon cell cannot exceed about 29% efficiency, no matter how perfectly it is made. The Shockley-Queisser limit, a calculation from 1961 that A-level students may encounter in extension material, puts the theoretical maximum for a single-junction cell at 33.7%.

A multi-junction cell sidesteps this problem by stacking materials with different band gaps. The top layer has a wide band gap and absorbs high-energy photons efficiently, generating a high voltage per photon. Lower-energy photons pass through to the next layer, which has a narrower band gap suited to absorbing them. In a triple-junction device, three layers divide the spectrum into three portions, each captured near its optimal voltage. The theoretical efficiency limit for a triple-junction cell under normal sunlight is around 49%, far above the single-junction ceiling.

Perovskites are attractive for multi-junction cells because their band gap is tuneable. The general crystal structure is ABX₃, where A is typically a large organic cation (such as methylammonium or formamidinium), B is a metal (usually lead), and X is a halide (iodine, bromine, or chlorine). By varying the ratio of halides, researchers can shift the band gap continuously from about 1.2 eV to over 2.3 eV. This tunability is unusual among semiconductors and makes perovskites ideal partners for silicon in tandem and triple-junction architectures.

The three-step fabrication process for the middle cell is an example of engineering meeting chemistry. Perovskite films are typically deposited by spin-coating a precursor solution onto a substrate and then heating it to crystallise the film. The problem is that a single-step process tends to produce crystals with random orientations and numerous grain boundaries, which act as sites where charge carriers recombine and are lost. By depositing the precursor in three stages, with controlled drying and annealing between each, the team encourages the crystals to grow in a preferred orientation with fewer grain boundaries. The result is a more ordered film that conducts charge more efficiently.

The nanoparticle interlayer between the middle and bottom cells works on a principle familiar from A-level optics: scattering. Photons that pass through the middle perovskite layer without being absorbed hit the nanoparticles and are scattered at wide angles. Many of these scattered photons re-enter the perovskite layer from below, effectively doubling its optical thickness without increasing its physical thickness. This is a form of light trapping, a technique widely used in silicon cell design but here adapted for the internal interface of a multi-junction stack.

For students interested in the maths, the current matching condition is worth understanding. In a series-connected multi-junction cell, the total current is limited by the layer producing the least current. This means the band gaps and thicknesses of all three layers must be carefully balanced so that each generates roughly the same photocurrent. Getting this wrong, even slightly, means one layer becomes a bottleneck. The optimisation is a multi-variable problem that links material composition, film thickness, and optical design, exactly the kind of challenge that makes photovoltaic engineering a rich field for applied physics graduates.