If someone has drunk enough alcohol, this is often recognisable by his/her “alcoholic breath”. Eating significant quantities of garlic or onions is also usually still noticeable by other people the next morning. However, information about what we have consumed is not the only thing given away by breath, because several illnesses are also detectable by specific mouth odours. This is why the analysis of respired air holds attractive prospects for clinical diagnosis. One of its advantages is that it no longer requires the needle stab that is necessary for blood analysis. However, respired air analysis has been very difficult to carry out until now. For example the equipment cost is considerable in most cases. In addition the samples require elaborate preparation before the actual analysis, and up to now only small volatile compounds could be detected reliably. Now ETH researchers have recently developed a mass-spectrometric method that enables a complete fingerprint of breath to be determined quickly and easily. They also succeeded in the quantitative detection of large non-volatile substances. The scientists published their paper in the scientific journal “Angewandte Chemie” (1) .

The method developed by ETH Professor Renato Zenobi and his team is based on what is known as Quadrupole Time-of-Flight Mass Spectrometry (2) , abbreviated QTOF. This involves charging the molecules electrically and separating and identifying them based on their molecular weights. In the QTOF method the molecules are accelerated in an electrical field and are separated in the “Time of Flight” unit according to their mass to charge ratio. The time of flight of the fragments to reach the detector depends on their mass. A preliminary fragmentation of the molecules is carried out in the quadrupole unit. This then yields a spectrum of fragments that is characteristic of the original molecule and allows its identification. Renato Zenobi explains that “It also enables a clear distinction between substances with identical molar masses but different fragmentation patterns.”

However, the decisive new trick in the breath analysis method was the way the sample was input into the mass spectrometer. Usually a sample is extracted first of all and this liquid is then spray-atomised using an electric field. Instead the Zurich researchers now use direct droplet-droplet extraction: the breath sample is passed directly into the electrospray device where it intersects a stream of charged reagent droplets. The purpose of these droplets is to absorb and charge the molecules of interest. As they travel into the mass spectrometer, the droplets lose solvent and are subdivided more and more until finally the charged molecules remain and enter the QTOF mass analyser.

IImmediately testing the new breath analysis on himself: ETH Professor Renato Zenobi.

Renato Zenobi describes the benefits as follows: “This enables the mass analysis to run continuously for a prolonged time. The method is also very sensitive.” The samples no longer need to be pre-treated, which reduces losses. However, the ETH scientist was himself astonished to find that in contrast to conventional methods, this analytical technique also allowed detection of the components of respired air in droplet form, which contain the larger, non-volatile substances. On the question of size. The measurement method enabled the identification of substances with a molecular weight of more than 1000 Daltons ( > 1000 atomic mass units, amu) – an additional, unexpected success.

In their investigations the scientists analysed the breath of test subjects after consuming beer, nicotine or garlic. This enabled them to discover that when for example a European drinks beer this leads to a different breath fingerprint than for an Asian person. It was also possible to determine the urea content in the test persons’ breath. However significant such findings may be, the decisive fact for Zenobi and his team is that the method works. Looking to the future, the ETH Professor sees one possible use for the method in the rapid diagnosis of lung cancer for example, or carrying out quick, simple monitoring in surgical procedures such as lung tissue transplantation.

When asked whether the method will soon be widely usable from the equipment point of view, Zenobi’s reply is: “Yes! The appropriate instruments already exist at many institutes – especially in clinical medicine.” Thanks to ETH research, the doctor’s instruction “take a deep breath” will have much greater meaning in the future.