Fast Track to Cleaner Diesel Exhaust

Company to supply catalytic converters for reducing car pollution. Research collaboration is part of the equation.

air pollution

Many countries are trying to get more electric and hybrid cars on the roads. Nevertheless, combustion engine vehicles will still dominate the traffic for many years to come.

Today, diesel engines are still the cheapest solution for many, especially for heavy vehicles. However, we must reduce the emission of environmentally hazardous and harmful substances from diesel vehicles. This is where catalysis comes in.

A catalyst is a substance that initiates a chemical process without being consumed. There are four main groups of problematic substances in diesel exhaust: soot particles, hydrocarbon (diesel that did not burn up in the engine), nitrogen oxide (NOx), and carbon monoxide (CO).

In treating the exhaust smoke, it is necessary to reduce the emission of all four groups at practically the same within a very short time and with limited space available.

Pär Gabrielsson is the Director of Research & Technology at the catalysis company Umicore, which has taken over Haldor Topsøe’s activities within automotive emission catalysis. Umicore has also take over a five-year collaboration agreement between Haldor Topsøe and DTU Chemical Engineering.

Pär Gabrielsson explains that there is an alternative approach behind the catalysts and the catalytic converters developed by Haldor Topsøe and bought by Umicore:

“For example, the car industry has traditionally based their work on so-called mapping. This means that comprehensive engine tests are carried out, after which the results are converted into tables, where you can look up the state of the emissions under certain conditions. Instead, we have chosen a technical chemistry approach, in which we began by modelling the chemical reactions that take place during the cleaning process.”

Saves months of development time

Kinetics, in particular, has attracted attention. That means being able to calculate reaction time and the amount of catalytic material used.

Based on these models, we are able to predict how a system change considered by the car factory will affect the emission.

It is still necessary to perform practical testing to verify the calculations, but the very extensive measuring campaigns are no longer necessary.

“This means that the car manufacturers—our customers—can get the answers to their questions in a matte of days rather than months. This obviously makes them very happy, and that is a big part of the explanation for how we have managed to establish ourselves within a business area that was completely new to us,” says Pär Gabrielsson.


The collaboration project with DTU Chemical Engineering was named NEXT (Next generation exhaust gas cleaning technologies for diesel vehicles).

In addition to training researchers and graduates, the programme has delivered a number of research results.

“The DTU researchers have modelled the interaction between the different catalytic layers and how porosities in the surfaces of the layers affect the efficiency of the catalyst. My colleague Tonnes Janssens succeeded in incorporating the models into our own software at an early stage,” says Pär Gabrielsson.

Particles can become too small

Another important result relates to the size of the platinum particles catalysing the removal of NOx.

Platinum is an expensive material which must be optimally exploited. As a rule of thumb, it is good to have very small particles, as this will create a greater surface area compared to the amount of catalytic material. However, there is a limit to how small the particles should be.

“When the particle is so small that it only consists of a few atoms, it can have sudden undesirable consequences. First, the geometry changes in a way that reduces the number of active ‘sites’, i.e. there will be fewer platinum atoms available to catalyse the reaction. Secondly, there is a risk that the particle will be oxidized completely. This would block the catalysis,” says Pär Gabrielsson. 
NEXT project testing has revealed that the ideal size of the platinum particles are two to five nanometres.

“It is a remarkable result, which we are now working to put into practice. We have yet to find the exact recipe that can give us the optimal distribution of particle sizes, but now we know the goal that we should aim for,” says Pär Gabrielsson.