A miniaturized multisensor platform for explosives detection

Land mines pose a constant threat to the military personnel and civilian population living and operating in current and former war zones. A couple of incidents with land mines, or even the suspicion of the presence of these devices, are enough to cause an area to be avoided. Therefore, land mines left over from an armed conflict hinder the use of farmland or roads in a mine-infested area many years after combat has ended. Today, mines are primarily detected with a combination of metal detectors and mechanical probing. This is extremely labor intensive and slow. Trained dogs are much more effective. However, they have the disadvantage of being expensive to train and maintain, and they require a highly skilled handler. Thus, there is a demand for very sensitive chemical detectors that can be produced at low cost and operated by minimally skilled personnel.

Realizing that no sensing principle is perfect, we set out to combine four fundamentally different sensing principles into one device.¹ The reasoning is that each sensor will complement the others and provide redundancy under various environmental conditions. As each sensor can be created using micro-fabrication, the inherent advantages associated with microelectromechanical-systems technologies, such as low-fabrication costs and small-device size, allow us to integrate the four sensors into one portable device at low cost (see Figure 1).

Figure 1. Micro devices used in the Xsense project for explosives sensing. SERS: surface enhanced Raman scattering.

Calorimetric sensing is based on measuring the temperature change associated with chemical reactions or physical changes, such as melting or evaporation. We have successfully fabricated a 400μm×100μm×400nm suspended-silicon membrane with integrated heating elements and a temperature measuring resistor.² This silicon ‘bridge’ has very little thermal mass and can thus be heated to 500°C within 50ms to cause first melting and then deflagration of any energetic material on the bridge. Using electronics developed in collaboration with our industrial partners in Unisensor A/S, we heat two bridges in parallel where only one is exposed to the gas to enable a differential measurement. As each explosive has a unique exothermic temperature profile when deflagrating, it is possible to distinguish between different explosives. A particular advantage of the calorimetric sensor is that it is self-cleaning, in the sense that all material is burnt off the bridge after each heating cycle.

The colorimetric sensor is based on the detection of color change of specific chemo-selective compounds (dyes) as these react to environmental conditions such as the presence of explosives' molecules. Each dye reacts to particular airborne substances enabling simultaneous screening of multiple materials of interest. The selected dyes are highly sensitive, fast in response and low in cost. We have synthesized a new class of chemical compounds that changes color when exposed to trinitrotoluene, most commonly know as TNT.³ Also included in the array are dyes that can distinguish between similar volatile organic compounds such as methanol, ethanol, and propanol.⁴ The change of color is based on non-covalent host-guest interaction and redox interaction between the dye (host) and the substance (guest). By first splitting the colors into their red, green, and blue components, a difference map of the substance prior to and after exposure can be created. This is used to create a unique fingerprint library for each compound of interest.

The third detection principle is surface enhanced Raman scattering (SERS). With SERS it is possible to identify a molecule absorbed onto a gold or silver nanostructure by mapping its vibrational states as the molecule is excited by a laser source. SERS has gained popularity as a chemical-sensing platform since it allows for trace levels of detection in addition to spectrographic information. We have developed a process which creates large areas of highly Raman enhancing surfaces using only two fabrication steps.⁵ Since only two steps are needed, the process is very cost effective. These metal-coated freestanding nanopillar substrates have demonstrated a chemical-detection sensitivity more than two orders of magnitude better than current commercial substrates. They are currently being integrated with chip-based micro-spectrometers provided by our partners at Serstech AB.

Finally, micro-cantilevers are capable of detecting minute changes in surface stress and, as such, have been used for various sensing applications in the last decade. By modifying one surface of the micro-cantilever with a selective layer, we can detect various explosives. We have developed a method of reading out multiple cantilevers on the same chip using a standard commercial DVD pickup head and its built-in transverse motor.⁶ The cantilevers are read out optically, and the optics in the DVD pickup head allow for fast auto focusing. Cantilever bending is determined based on an astigmatism of the focused laser beam, and it will ultimately allow for rapid alignment of a multitude of cantilevers. In addition to the DVD pickup head being relatively compact, using commercially available optics drastically reduces the price of the device.

Sample collection and pre-concentration of collected air samples are not within the scope of the Xsense project. However, we are currently pursuing the strategy of combining all four sensors onto a common platform which will allow fast analyte transport from one sensor to the next. As the end-user is only interested in a ‘yes’ or ‘no’ signal, we have an entire work package dedicated to data analysis.⁷ The challenge is to extract the significant signal features from each of the sensors while taking varying environmental conditions into consideration. To this end, machine learning will be employed to minimize the number of false positives and negatives.

The Xsense project has shown proof-of-concept for each sensor technology to detect trace amounts of explosives in gases. We believe that combining these sensing principles in a single device will significantly improve the ability of minimally trained personnel to detect land mines and other explosives with much greater efficiency than is possible today.