Towards Cheap, Miniaturized, Robust Sensors For Explosives 

The compound 2,4,6-Trinitrotoluene (TNT) is an explosive material widely used for military, industrial, and mining applications. Its reduction products are known to be toxic and carcinogenic to humans and may contaminate and accumulate in soils and drinking water.

Furthermore, being one of the most commonly-used industrial explosives, the procurement of TNT by terrorists to build improvised explosive devices (IEDs) poses a real and ongoing threat and presents a significant safety concern in peoples’ daily lives. Due to these environmental and security needs, the development of sensors for explosives such as TNT is of huge interest.

Electrochemistry is emerging as a viable technique for explosives detection “in the field” due to its many advantages, including low-cost instrumentation, portability, durability, sensitivity, selectivity, and fast response times. One common electrochemical technique employed in chemical sensors is amperometry. Amperometry is based upon applying a voltage on the sensor electrode and measuring the current generated – this current is directly related to the concentration of the target analyte. Common methods to measure the concentrations of TNT in water samples typically involve extraction and pre-concentration steps, before transportation to the lab for analysis. Cheap, portable (and potentially disposable) sensors that can be used to detect and quantify TNT on-site (e.g. directly in soil or river water) are therefore highly desirable.

Amperometric sensors are comprised of three electrodes (working, counter, and reference) connected by an electrolyte containing the analyte of interest. In recent years, there has been a trend towards miniaturized lab-on-a-chip sensors, such as the one presented in Figure 1. In this device, all three electrodes are contained within a very small circular area of just 2 mm in diameter. An electrolyte volume as small as 1 microliter can be used, and room temperature ionic liquids (RTILs) are ideal solvents for this purpose due to their extremely low volatility (i.e. they do not readily evaporate). Their other properties – intrinsic conductivity, wide electrochemical windows, high chemical and thermal stability, high polarity, and the ability to dissolve a wide range of compounds – also make RTILs favorable electrolyte materials.

Figure 1. Photo of a commercially available thin-film electrode device from MicruX (Oviedo, Spain) that can be used with a microliter droplet of ionic liquid for electrochemical sensing. The finger is used to show the very small dimensions of the low-cost, planar electrode device. Reprinted with permission from Trends Anal. Chem. 97 374–384. Copyright 2017 Elsevier.

A research team led by Dr. Debbie Silvester from Curtin University in Perth, Western Australia, have devised a new electrochemical technique to detect and quantify trace amounts of the harmful TNT contaminant in water samples. They have mixed RTILs with common and commercially available methacrylate polymers (a type of polymer found in contact lenses, bathroom fixtures, signboards, bone cements, dental restorations, etc.) to produce highly viscous “gel polymer electrolytes” (GPEs) which do not readily flow. The GPE can be easily casted as a film on top of the miniaturized planar electrode device.

The choice of ionic liquid and polymer in the mixture was found to be extremely important – only the combination of the very hydrophobic RTIL trihexyltetradecylphosphoniumbis(trifluoromethylsulfonyl)imide[P14,6,6,6][NTf2] and the polymer poly(hexyl methacrylate) was found to be sufficient for direct immersion into water samples for short time periods (see Figure 2). This RTIL/polymer combination gave a sensor current response towards TNT that was much less affected by moisture compared to the neat RTIL. Additionally, the amperometric sensing of TNT in this particular GPE was also found to be unaffected by the presence of oxygen in the air. These factors are crucial requirements to enable these sensors to be utilized in real environments.

Figure 2. Photographs of three different GPEs on a glass slide after increasing immersion times in water, followed by removal and drying in air at room temperature for 20 hours and 2 weeks. Reprinted with permission from Anal. Chem. 89, 4729−4736. Copyright 2017 American Chemical Society.

It was previously uncovered that RTILs possess a unique ability to preconcentrate explosives such as TNT. Hence, by simply dipping the GPE into the analyte sample, followed by 15 minutes of stirring, TNT selectively becomes accumulated into the GPE, thus eliminating the need to perform as a pre-analysis or extraction step (see Figure 3).

The sensor device was able to quickly and easily quantify TNT concentrations at typical groundwater contamination levels (e.g. sub-micrograms per milliliter concentrations) of TNT, with a low limit of detection of 0.37 µg/mL. The low-cost and portability of the sensor device, along with the minimal amounts of GPE materials required, make this a very promising technology for the onsite monitoring of explosives. Furthermore, this hydrophobic polymer/RTIL based sensors can be potentially be extended to enable the detection and sensing of other analyte species including gases and other explosive compounds.

Figure 3. Illustration of the sensing mechanism for TNT using the new electrochemical method. TNT is initially present in an aqueous phase, and this sample solution is placed into contact with the miniaturized electrode device covered with gel-polymer electrolyte. A cumulative method was employed, and the peak current (first reduction peak) from the square-wave voltammetry was recorded at different cumulative concentrations of TNT. Reprinted with permission from Anal. Chem. 89, 4729−4736. Copyright 2017 American Chemical Society.

These findings are described in the article entitled Detection of 2,4,6-Trinitrotoluene Using a Miniaturized, Disposable Electrochemical Sensor with an Ionic Liquid Gel-Polymer Electrolyte Film, recently published in the journal Analytical Chemistry. This work was led by Debbie Silvester from Curtin University.