Microfluidic Sensors Based on Molecularly Imprinted Polymers for Theselective Detection of Volatile Organic Compounds

Abstract

Selective and sensitive detection of volatile organic compounds (VOCs) is of great importance in applications involving the monitoring of hazardous chemicals or non-invasive diagnosis. Diffusion-base microfluidic gas sensors are a new generation of gas detectors. These sensors have increased selectivity; this is primarily due to the effect of diffusion and adsorption/desorption (highly influenced by the coating materials and dimensions of the microfluidic channel) of the analyte molecules along the microfluidic channel (1,2). Therefore, the combination of the diffusion and adsorption/desorption rates of different analytes with different concentrations leads to a unique response (i.e., “smell print”) from the embedded sensor (3). In addition to the coating (or the combination of the layers coated) the surface roughness can affect adsorption and desorption phenomena and hence the performance (mainly selectivity) of the sensors. There are several ways to change the coating and roughness of the channel. One interesting method, which changes the surface material of the microchannel and interaction with the analyte, is the integration of molecularly imprinted polymer nanoparticles (MIP NPs) into the channel.MIP is a synthetic polymer possessing selective molecular recognition properties (due to its binding sites) with size, shape and functionalities complementary to the target molecules. MIPs, often called artificial antibodies or receptors, are alternatives to antibodies in a variety of applications. The advantages of MIPs over natural antibodies include easy and low-cost preparation, longer life cycles (up to several years), highly tailorable recognition properties, solvents and thermal stability, robustness and resistance to a wide range of pH, and ease of regeneration and sterilization.For the first time, MIP NPs were synthesized with acetone recognition sites due to its importance as a biomarker for the non-invasive diagnosis of several diseases like diabetes. The synthesized MIP NPs were integrated into a 3D-printed dual-channel platform (one channel without and the other with MIP NPs) to develop a highly selective detection device against various VOCs (alcohols, ketones, nitrile, and aromatic compounds).The MIP NPs with acetone recognition sites were prepared for the first time using the simple precipitation procedure. After polymerization, the solution was centrifuged and the final polymer was ground into a powder. To remove the template, the MIP powder was washed with acetonitrile, followed by heating the powder at 80ºC for 48 h. The morphology and chemical structure of MIP NPs were investigated by field emission scanning electron microscope and Fourier-transform infrared spectroscopy.The microfluidic-based gas platform consists of two microfluidic channels and gas sensors (embedded at the channel terminus). The microchannels were fabricated from VeroClear RGD810 using a 3D-printer . Each microchannel has equal dimensions of 500 µm in height, 3 mm in width, and 3 cm in length. The inner surfaces of the microchannels are coated with Parylene C using a chemical vapor deposition (CVD) coating machine . To investigate the effect of treating the channel with MIP NPs on the selectivity of the sensors, one channel was coated with MIP NPs (referred to as the MIP channel) and another one was left uncoated (referred to as the control channel). For this purpose, a certain amount of synthesized MIP NPs dispersed in acetonitrile was drop-casted on the surface of the MIP channel and dried for a day in a clean air chamber. VOCs flow into the channel inlet, diffuse along the microfluidic channel and reach gas detectors placed at the end of the microchannels when exposed to the device.The synthesized MIP NPs were integrated into a 3D-printed dual-channel platform (one channel without and the other with MIP NPs) to develop a highly selective detection device against various VOCs (alcohols, ketones, nitrile, and aromatic compounds). A simple feature extraction method was used to distinguish signals arising from different VOCs. The differences between the feature extraction of the two channels showed that in general, this platform was specific to acetone, proving the potential application of the MIP NPs in artificial olfaction platforms.Fig. 1 shows the response of the dual-channel detector when exposed to 200, 400, 800, 2000, 4000 ppm concentrations of several VOCs. The responses of both detectors increase with increasing concentrations of the target VOC. Although the response of the two detectors to the same target analyte was significantly different, with the increase in the concentration the magnitude of signals increases; however, the time of the maximum response still remains the same. It was also observed that for the low concentration of acetone, butanone and ethanol, the responses from the nonoMIP-coated detector are not distinguishable. This again confirms the interaction between these analytes and the MIP NPs.