At first glance, an array of nanowires makes for a lousy solar cell. Their surface-to-volume ratio is huge. So when an absorbed photon produces an electron–hole pair, the charges may separate only briefly (typically driven by a p–n junction fabricated within the wire) before becoming trapped at one of the numerous defects that often reside at the surface. There, the charge carriers are more likely to recombine and thus give up their electrical energy before it can be harnassed in a circuit.

Still, a nanowire architecture has natural appeal: Scarce but highly absorptive semiconductors can be configured into photovoltaics that use far less material than those made from conventional thin films. Some III–V semiconductors, for example, have absorption coefficients a thousand times greater than silicon. But none of the earlier nanowire devices based on them have surpassed power-conversion efficiencies of 3–5%. That’s well below the roughly 20% that can be...

You do not currently have access to this content.