5.1 Silicon Cell Technologies

Quick Answer: Silicon is the semiconductor used in almost every small PV module a 2922 installer will handle. Cells are typically doped with boron to form p-type material or phosphorus to form n-type material, creating the p–n junction that turns light into DC. The City & Guilds 2922 handbook v1.3 (LO3, 3.1.1) groups conversion technologies as mono-crystalline (single crystal, high conversion efficiency), poly-crystalline (multiple smaller crystals, slightly less efficient, more cost effective), and thin-film (semiconductor layers on a substrate, less expensive to produce, generally lower efficiency). On a typical UK roof the usable area is tight, so the higher watts per square metre of mono-crystalline modules usually deliver more annual kilowatt-hours than poly or thin-film on the same slope.

Why silicon, and what doping does

A photovoltaic cell is a large-area semiconductor diode. When a photon with enough energy hits the lattice, it can free an electron–hole pair. The built-in field at the p–n junction separates those charges so they can do work in an external circuit rather than recombining inside the crystal.

Silicon is the material you will see on almost every domestic and small commercial 2922 job. It is abundant, well understood, and the MCS 005 listed-module supply chain is built around it. Compound semiconductors such as gallium arsenide appear in specialist space and concentrating products; they are not the default for a UK house. The handbook doping statement is the one to remember for the knowledge test: silicon is typically doped with boron (p-type) or phosphorus (n-type). Boron has three valence electrons, so substituting it for silicon creates holes (p-type). Phosphorus has five, so it donates electrons (n-type). Modern cells may use more elaborate doping stacks (PERC, TOPCon, heterojunction), but the exam still expects that boron/phosphorus p/n picture as the starting model.

Cells are then series-connected inside a laminate so their voltages add. That is why a single shaded cell can hold back a whole substring — a point section 5.3 returns to when the I–V curve and bypass diodes meet shading.

Mono-crystalline cells

Mono-crystalline (often shortened to mono) wafers are cut from a single, continuous crystal, classically grown by the Czochralski process. Because the lattice is one crystal, there are fewer grain-boundary recombination sites. The handbook therefore classes mono as high conversion efficiency.

On the roof you recognise mono as a uniform dark or black cell field, often with truncated-circle or fully square wafers, behind anti-reflective glass. Higher conversion efficiency means more DC watts from each square metre of module. For a 2922 designer that is not a vanity number: UK dwellings rarely offer a huge, unobstructed south plane. If the customer wants maximum annual energy from a fixed 20 m² of south-facing covering, watts per square metre dominate the conversation.

Typical commercial module efficiencies for current mono rooftop products sit in a band of about 20–23% at Standard Test Conditions (STC: 1000 W/m², 25 °C cell temperature, AM1.5 spectrum). Treat that band as a typical industry range, not a City & Guilds pass mark and not a figure printed as a requirement in handbook v1.3. Premium back-contact or heterojunction modules can sit a little above that band; older or budget mono can sit a little below. Always read the nameplate and datasheet for the product you are installing.

Poly-crystalline cells

Poly-crystalline (multi-crystalline) wafers are solidified as a block of multiple smaller crystals. Grain boundaries increase recombination, so conversion efficiency is slightly less than mono of the same generation. The handbook's second point is the commercial trade-off: poly is more cost effective — historically a lower price per watt.

Appearance is the speckled blue that many people still picture as "a solar panel." In the mid-2010s poly was common on UK roofs. By the mid-2020s it has largely left the new-install mainstream because mono manufacturing scaled and the price gap collapsed, while the efficiency gap did not. You may still inspect or maintain poly arrays, and the knowledge test can still ask you to describe them in handbook language.

Typical commercial poly module efficiencies occupy a typical industry range of about 15–18%. Again, that is industry observation, not a handbook scoring table.

Thin-film modules

Thin-film devices deposit a semiconductor layer (or stack of layers) onto a substrate — glass, metal foil, or a flexible polymer — rather than sawing wafers from a boule. The handbook's points are production cost and performance: thin-film is less expensive to produce and has generally lower efficiency.

Families you will meet in trade literature include amorphous silicon (a-Si), cadmium telluride (CdTe), and copper indium gallium selenide (CIGS). They are not interchangeable. Utility-scale CdTe can approach the lower end of crystalline-silicon module efficiencies; flexible a-Si products used on membranes are usually much lower. For 2922 small-system work, thin-film is the exception (a standing-seam roof, a curved surface, a weight-limited structure), not the default.

Typical industry module-efficiency bands are wide: many a-Si products sit around 6–12%, while some CdTe utility modules sit around 18–19%. Quote those only as typical industry ranges, never as handbook pass figures.

Comparing the three on a UK roof

Temperature and low light — qualitative only

Crystalline silicon loses power as the cells heat. A typical silicon module power temperature coefficient is about −0.3 to −0.4 %/°C relative to the 25 °C STC rating. That is a typical industry figure, not a UK-specific handbook constant and not a pass mark. On a still, sunny July afternoon a dark roof-mounted module can run tens of degrees above 25 °C, so the array delivers fewer watts than its nameplate even at 1000 W/m². Thin-film products often (not always) have a slightly gentler (less negative) power temperature coefficient, which can help in hot, still conditions — useful qualitative knowledge, not a reason to ignore area on a small roof.

Low light and UK weather are a different qualitative story. Much of the UK resource is diffuse rather than a clean 1000 W/m² beam. Thin-film was historically marketed as kinder in poor light, and poly was sometimes said to be less fussy than early mono. Modern mono (PERC, TOPCon and similar) also behaves well under diffuse irradiance. Do not invent a percentage advantage for any technology under UK overcast: the datasheet's low-irradiance curve, if published, beats folklore. What you can say on the exam is that current scales strongly with irradiance (section 5.3) and that a technology with more watts per square metre still tends to harvest more energy from a small roof across a year of mixed weather.

Cool UK winters raise open-circuit voltage — again a 5.3 topic — so mono's higher voltage per module is a string-sizing issue, not just a kWh issue.

Scenario: 20 m² south roof, customer wants maximum kWh

A customer has 20 m² of unshaded south-facing pitched roof and asks for as many kilowatt-hours per year as the roof will physically support, not the cheapest quote.

Annual energy is driven by installed kWp, orientation and tilt (already favourable), shading (assumed clear here), and system losses. For a given area, nameplate kWp is roughly module efficiency × area × 1 kW/m². Using the typical industry bands above — not handbook pass marks — 20 m² of ~22% mono is about 4.4 kWp. The same 20 m² of ~16% poly is about 3.2 kWp. Thin-film at ~10% is about 2.0 kWp. Specific yield (kWh per kWp per year) will be similar if the three arrays share the same tilt, azimuth, and climate; the mono array simply has more kWp in the same rectangle, so it produces more kWh.

Packing, frame borders, fire-break gaps, and in-roof flashing eat area, so real kWp is a little below that envelope. The design choice is still the same: specify high-efficiency mono, pack the plane honestly, and do not "save money" with poly or a-Si if the brief is maximum energy from a limited UK roof. If the brief later becomes lowest capital cost, or a conservation officer demands a tile-like covering, you revisit module type in section 5.2 — that is a different optimisation.