Monitoring trace gases is of great importance in a wide range of applications. Detecting a diverse range of chemical agents requires an adaptable sensor platform capable of identifying threats before they cause harm. Research and development in hazardous-materials detection technology focuses on increasing speed, sensitivity, and selectivity while reducing size and cost. Although the current state-of-the-art vapor detector (Joint Chemical Agent Detector) is lightweight, handheld, and easily attaches to a belt, it still provides added bulk to a soldier on foot. Recently, microcantilever-based technology has emerged as a viable platform due to its many advantages such as small size, high sensitivity, and low cost. However, microcantilevers lack the inherent ability to selectively identify chemicals of interest. The key to overcoming this challenge is to functionalize the top surface of the microcantilever with a sorbent layer (i.e., polymer) that allows for selective binding between the microbeam and analyte(s) of interest.

A simplified schematic of molecularly imprinted polymer fabrication.

Molecular imprinting involves arranging polymerizable functional monomers around a template, followed by polymerization and template removal (see figure). Arrangement is generally achieved by noncovalent or reversible covalent interactions. In both types of molecular imprinting, once the template is removed, three-dimensional cavities are generated within the final materials that are complementary to the template molecule in size, shape, and functionality. Essentially, one creates a molecular “memory” within the imprinted polymer matrix. This allows preparation of polymers that are selective for the adsorption of the target molecule of interest. Other advantages of this technique include robustness and stability under a wide range of chemical and physical conditions, and an ability to easily design recognition sites for a plethora of target chemicals (e.g., pesticides, energetic materials, pharmaceuticals, and proteins).

The polymer materials of interest are sol-gel-derived xerogels, which have been used as a platform for MIP-based sensor development. These materials are attractive because their physicochemical properties can be adjusted by choice of precursor( s) and the processing protocol. Precursors were chosen based on potential interactions with the explosive 2,4,6-trinitrotoluene (TNT). These specific interactions allow for increased target recognition.

A MIP alone does not meet the requirements for a sensor without some form of a transducer to convert the analyte interaction into a measureable signal. There is evidence of a variety of gravimetrical detection techniques applied to convert a MIP into a “sensor.” Low-mass, highfrequency, and low-cost micro/nano sensors utilizing mass loading of microcantilevers have drawn increasing attention in the area of gravimetric sensing, and MIPs have become an attractive thin film coating for many microelectromechanical systems (MEMS)-based sensors. In this work, molecularly imprinted xerogel thin films have demonstrated selectivity and stability in combination with a fixed-fixed beam MEMS cantilever.

The sensors are fixed-fixed beams of varied lengths and widths. The beam thickness was 2 μm. The values were chosen so that the natural frequencies of the beams were less than half the natural frequency of the shear piezo actuator (330 kHz) used to drive the device. The microbeams were fabricated using a standard silicon on insulator (SOI) process. The SOI wafer used was 2 μm silicon (Si) device layer with 1 μm buried oxide and 520 μm Si handle. First, oxide was grown on both sides of the wafer. Silicon nitride (Si3N4) was deposited on the backside on top of silicon oxide; together they serve as masks to protect the backside for potassium hydroxide (KOH) etch in a later step. After front side oxide removal, it was then spun with photoresist and pattern was transferred. Then a deep reactive-ion etching was used to define the device features. Back side mask features were defined using photolithography and inductively coupled plasma. The front side was spun with a ProTEK® coating to protect features during back side release etch. The back side was opened by anisotropic KOH etch and stopped at the buried oxide. The device was then finished with removal of ProTEK and buried oxide layer.

The work reported here validates the MIP-coated microcantilever sensing concept and demonstrates the feasibility of this MEMS sensor for the detection of explosive compounds and CWAs. To date, this is one of the only demonstrations of a MIP-coated microbeam MEMS sensing platform for these targets. Although preliminary, the data suggests that this combination is an effective and robust chemical nanosensing scheme. Further investigations will focus on refinement of the MIP (i.e., xerogel formulation) for improved selectivity. Finally, the MIPcoated microcantilever sensor platform evaluation should be expanded to include other explosives and chemical warfare agents of interest to the Army. A successful MIP-coated microbeam MEMS sensing format could reduce sensor cost and size, while maintaining the high sensitivity, selectivity, and portability needed for military applications.

This work was done by Ellen L. Holthoff of the Army Research Laboratory and Lily Li, Tobias Hiller, and Kimberly L. Turner of the University of California Santa Barbara. ARL-0191

This Brief includes a Technical Support Package (TSP).
A Molecularly Imprinted Polymer (MIP)-Coated Microbeam MEMS Sensor for Chemical Detection

(reference ARL-0191) is currently available for download from the TSP library.

Don't have an account? Sign up here.