"The pollution of the air by tiny organic aerosol particles is one of the biggest challenges to our clean-air policy today." This is the information from the Swiss Agency for the Environment, Forests and Landscape (SAEFL) on suspended particles with diameters of less than ten micrometres. On the one hand, this assessment is based on the danger these aerosols pose to health. Owing to their small size they can penetrate very deeply into the lungs, infiltrating the finest branches. On the other hand, it is also known that these particles are involved in climatological processes, such as cloud formation or the radiation balance of our planet.

It is also known that organic material accounts for up to half of the aerosols. These organic substances result from the burning of fossil fuels, as well as from natural sources, like vegetation. Yet the greater part of the aerosols are not directly emitted from chimneys or exhaust pipes but are formed by secondary processes in the atmosphere under the influence of solar radiation. These particles are therefore also called secondary organic aerosols (SOA).

This was how far researchers had come to understanding the particle composition, until now. As far as the chemical composition of SOA was concerned, they were fishing in murky waters. A study, carried out by two collaborating teams of researchers from ETH Zurich (1) and the Paul Scherrer Institute (PSI) (2)now throws light on the matter. In their study, published in the last issue of Science magazine (3), the authors show that polymers are the main constituents of SOA.

Artificial summer days, real sunburn

In order to help them identify the aerosols, researchers at PSI built a smog chamber. In essence this chamber is a Teflon bag with a volume of 27 cubic metres, which stands in an air-conditioned wooden chamber. First, a mixture of the emission gas trimethylbenzene (TMB), nitrous oxides and water vapor was pumped into the sack. The concentration was chosen to correspond to the composition of the air on a summer's day.

Artificial solarium; the smog chamber in the main hall of PSI. (Picture: PSI) large

Now, only the sun is missing. This is simulated using four high-pressure xenon arc lamps, with a total power of 16 kilowatts. Markus Kalberer, first author of the study, explains that these lamps create a light spectrum that corresponds to the sun's own spectrum. The intensity of the lamps is comparable to that of the sun, too. One graduate student learned this at a painful personal cost, having caught sunburn after working for some time in the smog chamber without sun protection.

But how does the TMB react to an artificial sunny day? To find the answer to this question the team first had to collect the aerosols at "different times" of the day. They solved this with an impactor – a device that pumps the air around narrow curves at high velocity. In the curves the aerosols can't keep up with gas flow and can thus be made to adhere to small steel plates. Once the aerosols are captured the team was able to proceed with identification. To do this they used a laser mass spectrometry, which was developed by ETH Professor Renato Zenobi's team.

Amazingly large molecules

And the mass spectra showed something amazing. More than a hundred organic compounds had been produced in the smog chamber. A greater part of the molecules had a molecular mass of more than 400 and up to 1000.

Hunting down the aerosols: researchers in the smog chamber at the Paul Scherrer Institute. (Picture: PSI) large

The team found that by increasing the length of time of radiation the proportion of big molecules also increased. In addition, signal peaks from the mass spectrum occurred in regular patterns. Taken together these results lead to the conclusion that the discovered aerosol consitituants are polymers, some of which are composites of up to nine monomers.

Aerosols not particularly volatile

After identifying the aerosols as polymers, the team wanted to test the volatility of the particles. To do this Urs Baltensperger's team at the Paul Scherrer Institute subjected the aerosols to temperatures of 200 degrees Celsius. And there was another surprise in store for them; the longer the period of time that the aerosols were treated with solar radiation, the more stable they became. This indicated that the particles evaporated less when they were heated. This is how the team deduced that polymers constitute up to 50 per cent – and sometimes more – of aerosol matter. This result naturally has consequences for models of SOA development, which are used in global climate models. Such models can no longer simply assume that there is an equal balance between gas and aerosol phase.

It is clear to Markus Kalberer, Urs Baltensperger and Renato Zenobi that this study is just the beginning. Amongst other work, they now want to find a more exact definition of the chemical nature of polymers. Because, even though polymers can be explained theoretically by a linear combination of the four photooxidation products of TMB (of methylglyoxal, formaldehyde, 3,5-dimethylbenzaldehyde and pyruvic acid), scientists expect to find that further elements are involved in the process.

New molecules will probably be discovered if one looks at compounds other than TMB. Precursor substances that are are expected to be especially important in climatology will be the focus of such investigations. However, Baltensperger's and Kalberer's research goals don't end here. They also plan, working with biologists, to investigate the impact of the aerosol particles identified in the study on lung tissue. Such knowledge would be of great help in assessing the consequences of organic aerosols on health.

The laser mass spectrometer of ETH Professor Renato Zenobi's team, which analyses aerosol samples. large