The food sector uses such electronic noses as a means of quality control, for example in developing and maintaining the right smells for their products.

The technology is also used to pick up the smells produced by bacteria and pathogens as they break down foods, helping processors to remove contaminated products.

However electronic noses can only discern a much smaller range of smells than the natural version. Now a project by researchers at the University of Warwick and Leicester University aims to improve the sensitivity and accuracy of the devices.

The researchers developed an artificial snot to mimic the nasal mucus found in human noses.

The researchers coated the sensors used by odour sensing "electronic noses" with a mix of polymers to mimic human snot. They found their mix improved the performance of the electronic devices allowing them to pick out a more diverse range of smells.

The snot allowed the devices to differentiate between smells such as milk and banana. The two smells had previously been challenging smells for the device, said University of Warwick researcher Julian Gardner.

"Our artificial mucus not only offers improved odour discrimination for electronic noses it also offers much shorter analysis times than conventional techniques," Gardner said.

A natural nose uses over 100 million specialised receptors or sensors which act together in complex ways to identify and tell apart the molecules they encounter.

Electronic noses use the same method but often have less than 50 sensors.

In the natural nose the thin layer of mucus dissolves scents and separates out different odour molecules in a way they arrive at the noses receptors at different speeds and times.

The final device, including the sensors and the artificial mucus, is contained in a relatively thin piece of plastic just a few centimeters square and costing about 7 euros to produce, they said.

The research was published in the current issue of the Proceedings of the Royal Society.