How a solar cell works: the photovoltaic effect

Customers often ask, "How does a panel make electricity from light?" You should be able to answer in two minutes without jargon, and you should understand it well enough to explain what you see on your multimeter. This lesson covers both.

Silicon, the raw material

Almost every rooftop module in India uses crystalline silicon cells. Silicon is refined from quartz sand, grown into a large crystal (an ingot), and sliced into thin wafers, about 0.13–0.15 mm thick today. Common wafer sizes are 182 mm (called M10) and 210 mm (G12).

Pure silicon is a poor conductor. Its outer electrons are locked in bonds with neighbouring atoms. To make a cell, manufacturers add tiny, controlled amounts of other elements. This is called doping:

• n-type silicon has a little phosphorus added. Phosphorus brings one extra electron per atom, so n-type silicon has spare negative charges.
• p-type silicon has boron (or gallium) added. These atoms have one electron too few, leaving "holes", which behave like positive charges.

A cell is a wafer where one side is n-type and the other is p-type. The boundary between them is the p-n junction, and it is the heart of the cell.

The junction and its built-in field

When n-type and p-type silicon meet, some spare electrons from the n-side drift across and fill holes on the p-side. This leaves a thin zone at the junction with fixed charges: positive on the n-side, negative on the p-side. That zone is the depletion region, and its charges create a permanent electric field inside the cell.

That built-in field is a one-way gate. It pushes electrons towards the n-side and holes towards the p-side, and it never switches off.

The photovoltaic effect, step by step

1. Light arrives. Sunlight is a stream of energy packets called photons.
2. A photon is absorbed. If it carries enough energy (more than silicon's band gap of about 1.12 eV, which covers visible and near-infrared light up to roughly 1,100 nm), it knocks an electron loose from its bond. This creates a free electron and a hole: an electron-hole pair.
3. The junction separates them. The built-in field sweeps the electron to the n-side (front) and the hole to the p-side (rear) before they can recombine.
4. Charge builds up. Electrons pile up at the front fingers and holes at the rear contact. This is a voltage.
5. Current flows. Connect a wire and a load between front and rear, and the electrons flow round the circuit to recombine with holes at the back. That flow is DC current.

No moving parts, no fuel, no noise. As long as photons keep arriving, the cell keeps pumping electrons.

How much voltage and current does one cell give?

Voltage

A modern silicon cell gives about 0.65–0.72 V with no load connected (open-circuit voltage, Voc). That is fixed by the material, not by cell size. To get useful voltages, cells are wired in series.

Check this against a real module. A typical 144 half-cell, 550 Wp module behaves electrically like 72 cells in series. Its datasheet Voc is 49.9 V:

49.9 V ÷ 72 cells ≈ 0.69 V per cell. The physics and the datasheet agree.

Current

Current depends on how much light lands on the cell and on its area. A full 182 mm cell gives around 13–14 A at full sun. A half-cut cell gives half that, and modules connect the halves so the module current still comes out near 13–14 A.

What changes the output

More light: more current

Current rises almost in proportion to irradiance. At 500 W/m² a module gives roughly half its full-sun current. Voltage changes only a little with light; it drops slowly at low light and collapses only near darkness.

More heat: less voltage

Heat is the enemy. Each cell loses about 2 mV per °C of temperature rise. For the 550 Wp module above (Voc temperature coefficient −0.27 %/°C):

• At 25 °C (test condition): Voc = 49.9 V
• At a cell temperature of 65 °C, on a Jaipur roof in May: 49.9 × (1 − 0.0027 × 40) = 49.9 × 0.892 ≈ 44.5 V

That is 5.4 V lost, more than 10%, purely from heat. Power falls with it. This is why yields in hot cities trail their irradiance, and why ventilation under the modules matters.

Cold mornings: higher voltage

The reverse is also true. On a 5 °C winter sunrise in Pune or Nagpur, Voc climbs above the datasheet value. String design (module m06) must allow for this, or the inverter can be damaged.

Why cells are not 100% efficient

A module rated around 21–23% turns that share of the sunlight hitting it into electricity. The rest is lost because:

• photons with too little energy (deep infrared) pass straight through;
• photons with extra energy (blue and ultraviolet) give the surplus away as heat;
• some light reflects off the surface or is blocked by the metal fingers;
• some electrons and holes recombine before they reach the contacts.

A single-junction silicon cell has a theoretical ceiling of about 29%. Cell makers chase the last few percent with better surface passivation, which is what PERC, TOPCon and HJT are about (next lesson).

Linking physics to your multimeter

• Voc reading lower than expected on a sunny morning? Check cell temperature first, then look for a shorted bypass diode (it removes about one-third of the voltage).
• Isc reading low at noon? Check irradiance (thin cloud, haze) and soiling before blaming the module.
• Zero current with normal voltage? The circuit is open: a loose MC4, an open isolator or a blown fuse.

Key points

• A cell is a silicon wafer with a p-n junction whose built-in field separates the electrons and holes that light creates.
• Each cell gives about 0.65–0.72 V open-circuit, whatever its size; series connection builds module voltage.
• Current rises with irradiance and cell area; a full 182 mm cell gives about 13–14 A at full sun.
• Heat lowers voltage by about 0.27–0.35 %/°C; cold mornings raise it, which matters for string design.
• Commercial modules convert about 21–23% of sunlight; physics limits single-junction silicon to about 29%.

Check yourself

1. Which layer of a cell has phosphorus added, and what does it provide?
2. A module has Voc 50 V with 72 cells in series. Roughly what is the voltage per cell?
3. On a hot afternoon, does a module's voltage go up or down? Why does this matter for yield?