Muons reveal the secrets of hydrogen storage in semiconductors
24 Aug 2009



Zinc oxide is a transparent semiconductor material which could play a vital part in a number of new applications such as energy-efficient white lighting.

Arjan Houtepan

​​Arjan Houtepan preparing his II-VI semiconductor nanocrystal sample to search for hydrogen shallow donor states using the EMU muon spectrometer.​


Breakthroughs in this kind of semiconductor technology have occurred over recent years thanks to a better understanding of these materials and their defects at the atomic level. Muons from the ISIS facility have been key in a number of these discoveries.

Muons are elementary particles, commonly found in cosmic rays. At ISIS they are generated alongside neutrons and provide a unique probe in materials science. For Professor Steve Cox and his ISIS colleagues, together with collaborators in the US and Portugal, muons have helped to model how hydrogen behaves inside semiconductor materials. By implanting positively charged muons into semiconductors they can produce hydrogen-like defects. The location and structure of these defects are then studied using a technique known as muon spin resonance (μSR).

An unavoidable impurity

Semiconductors tend to be grown by deposition from hydride gases, so it is very hard to keep hydrogen out.
“Sometimes it reacts chemically with other defects, helping to tie up dangling bonds, that would otherwise trap the conduction electrons”, explains Cox. But too much hydrogen can counteract the desired conductivity, by passivating or compensating the deliberate dopants. Zinc oxide was the first case where hydrogen was predicted theoretically to be a dopant itself – actually acting to increase the conductivity. “This prediction was quickly and easily confirmed by our μSR studies”, says Cox.


Since the zinc oxide discovery, μSR a​t ISIS has been used to survey the effect of hydrogen in many other oxides destined for applications in electronics. Some are candidates for the new generation of thin film insulators – replacing SiO2 as a transistor gate dielectric, for instance – but the ISIS work has shown that by no means all would be suitable.

Cox calls the method muonics. “Thanks to its remarkable sensitivity and selectivity, this is much easier than spectroscopic studies of hydrogen itself, yet it makes hydrogen one of the best characterised and best understood of all relevant defects”, he says.
As other new semiconductors and insulators are proposed or developed, muons offer a convenient and systematic means of screening for hydrogen induced conductivity.

SFJ Cox (ISIS), J Gil and colleagues (Coimbra,Portugal), E Davis (University of Cambridge), JS Lord (ISIS)

Research date: December 2006

Further Information

Experimental confirmation of the predicted shallow donor hydrogen state in zinc oxide, SFJ Cox et al., Phys Rev Lett 86 (2001) 2601

Oxide Muonics: Modelling the electrical activity of hydrogen in wide-gap and high-permittivity dielectrics, SFJ Cox, J Phys Condensed Matter 18 (2006) 1079​