Developing a prototype sensor for selectively sensing chemical exposure to monocyclic aromatics

Outline of reasons and objectives
Volatile Organic Compounds (VOCs) are among the most common pollutants. Industrial production and use of motor cars has led to an increased supply of these organic chemical substances in the atmosphere. Benzene derivatives (i.e. benzene, toluene, p-, m-, and o-xylene, and BTX, i.e. mixtures of benzene, toluene, and xylenes) are a very specific category of VOCs because of their particular hazards for the environment and for human health (CMR).
The main objective of this study was to design a transportable real-time analyser prototype for analysing BTX that was usable for taking measurements in the workplace atmosphere. The general principle was to develop a sensitive material, or sensor, for sensing the concentration of BTX and for taking a UV spectroscopy measurement.

Approach
Initiated in 2009 with the Laboratoire Francis Perrin (LFP) of the CEA (France’s Alternative Energies and Atomic Energy Commission), the project did not produce the expected results, mainly due to difficulties with the reproducibility of the dimensions of the sensors. Reoriented in 2012, the study continued with a new partner, the Laboratoire de Chimie Physique et Microbiologie pour l’Environnement (LCPME), a research unit run jointly by CNRS and Universite de Lorraine. A PhD thesis was attached to this work whose core was to develop a material having porosity and functionality that were controlled for BTX concentration.
Numerous tests were performed and they led to silicated sensors being produced that were of geometrical configurations of the monolithic type (thickness of about one millimetre; sol-gel technique) and of the thin film type (a few microns thick; technique of deposition by dip coating with a solution of silica nanoparticles).
Functionalisation tests were conducted to improve the hydrophobicity of the material and to increase the BTX detection sensitivity. In parallel, a testing facility was designed, validated, and used for characterising the various sensors that were manufactured.

Main results
The performance levels achieved with the thin films made it possible to validate the feasibility of the principle of the analyser, in particular for detecting toluene and xylenes over a range from 1 ppmv to 100 ppmv. But the effect of the humidity of the air on adsorption of BTX was significant and the presence of other solvent vapours, such as ketone vapours, prevented quantification of BTX. The benzene detection threshold is not yet satisfactory (it was approximately a few tens of ppmv), compared with the limit value of 1 ppmv for benzene. Initial tests showed drift of about 40% for the same sensor during one hour of testing, and a reproducibility of about 40% over about forty
sensors produced. Conversely, the experience acquired during this work made it possible, on the basis of a different film structure, to envisage making an improvement in reproducibility (about 10%).

Discussion
While the work was taking place, it was observed that the design of the sensor (type of the material and dimensions) and its functionalisation raised technical and scientific problems, which had repercussions on keeping to schedule.
That difficulty prevented the metrological performance of the sensors from being evaluated and therefore prevented a prototype from being finalised. The avenues for improving the new film should be confirmed before considering resuming testing on a generation setup. The experience acquired during the work also made it possible to acquire new skills (UV detection, Nafion tubing, thin-film microcell, etc.) that could be developed in future studies.