Sensor Technology - Interviews with Experts

Interview with Udo Weimar

Dr. Udo Weimar (41) is a lecturer and specialist in biological and chemical sensors at the Physical Chemistry department of the University of Tuebingen. He believes the biggest challenge facing detector developers is how to come up with sensor probes capable of detecting odors on a selective basis.

Sensors are currently used for measuring things like acceleration, gas concentration and tire pressure. What will they be able to do in the future?

Weimar: Tomorrow’s sensors will be able to detect chemicals and molecules much more easily and clearly than those of today. This represents a major challenge, since even rough estimates say that about a billion different chemical substances exist. Ultimately, you would have to develop a sensor for each of them.

Do we really need to be able to identify all of these substances?

Weimar: No, of course not. The problem is that we often don’t know what it is we should be looking for. Coffee, for instance, contains around 1,500 substances that influence aroma. But to date, we are only familiar with a portion of those molecules that are decisive for giving it that spicy taste. The real question is: What causes aroma and taste to be perceived by the brain? It’s as if developers of chemical sensors are searching for the right chemicals in a smashed-up pharmacy.

How do you build the appropriate sensors?

Weimar: The important thing is to come up with a sensitive layer that displays a high level of affinity with the substance you’re targeting. This greatly facilitates detection. For example, aromatic compounds or biological molecules can bind on and in open-pored organic surfaces, such as polymers. The result is a change in the weight of the surface layer that can be detected by a unit made of sensitive oscillating quartz. However, the substance you’re trying to detect can’t be allowed to bind too solidly to the sensor, since it has to drop off from the sensor as soon as its concentration in the surrounding environment begins to decrease. As this is not very easy to achieve, it’s often necessary to wash off the biological molecules. Sensors used for measuring physical parameters are significantly less affected by these problems than are biological or chemical sensors.

Does that mean it’s easier to develop physical sensors?

Weimar: Well, the fact is there are only about 100 physical parameters, like acceleration, temperature or pressure. That’s a much more manageable number. The challenge here primarily involves the engineering—in other words, coming up with the ideal design for each application and making sure that it’s both robust and sensitive. In addition, customers are demanding that such sensors become smaller and less expensive. Developers of chemical sensors, on the other hand, are confronted with problems such as cross-sensitivity—in other words, the fact that one sensor can detect more than one substance. For example, natural gas sensors don’t just react to methane; they also often sound an alarm when vapor from a naphtha cleaning agent fills the air. We therefore need to develop sensors that are very selective—in other words, react to only one substance. We can now more or less accomplish this with sensor arrays, whereby sensors of differing affinities and sensitivities are linked together. A mathematical comparison is then used to determine which substance is actually present in the air.

What do you think are the prospects for bio-sensors—can they be used to combat disease, for example?

Weimar: In my opinion, you won’t soon be seeing sensors that travel through the body’s circulation system and radio their diagnosis to the outside. One of the reasons for this is that blood is a rather unfavorable medium, since it can easily clot. At the same time, sensors that measure blood sugar outside the body have been standard for quite some time. Much more exciting and promising, in my opinion, is the possibility of detecting diseases through odor. It is said that the alchemists of old and practitioners of traditional Chinese medicine were able to diagnose illnesses on the basis of body odors. For example, respiratory infections have a typical odor. It’s therefore conceivable that sensors will be used in the future to conduct odor diagnoses. Initially, however, we have to clarify the old question: What exactly causes a typical odor? And which molecules are actually responsible for it?

Interview by Tim Schröder