Perfluorooctanesulfonic Acid Detection Using Molecularly Imprinted Polyaniline on a Paper Substrate

Abstract

Perfluorinated compounds like perfluorooctanesulfonic acid (PFOS) are synthetic water pollutants and have accumulated in environments for decades, causing a serious global health issue. Conventional assays rely on liquid chromatography and mass spectroscopy that are very expensive and complicated and thus limit the large-scale monitoring of PFOS in wastewater. To achieve low-cost and accurate detection of PFOS, we designed a paper-based sensor with molecularly imprinted polyaniline electrodes that have recognition sites specific to PFOS. The calibration curve of resistivity ratios as a function of PFOS concentrations has a linear range from 1 to 100 ppt with a coefficient of determination of 0.995. The estimated limit of detection is 1.02 ppt. We also investigated attenuated total reflectance Fourier-transform infrared spectroscopy (ATR-FTIR) and X-ray photoelectron spectroscopy (XPS) spectra of the surface of the polyaniline (PANI) electrodes to propose the potential recognition sites in polyaniline matrix and the detection mechanism. This electrical paper sensor with low cost and excellent sensitivity and selectivity provides the potential for large-scale monitoring of wastewater.

Keywords: low-cost; molecularly imprinted polymer; paper sensor; perfluorooctanesulfonic acid; polyaniline.

Figure 1

The schematic diagram of the synthesis of PFOS-MIP-PANI and the molecular imprinting process. (a) PFOS, as the template, was first mixed with aniline monomers in HCl aqueous solution, followed by immersing paper strips for the thorough absorption of the monomers on the surface. The polymerization was initiated by adding the oxidant, and the PFOS-imprinted polyaniline (PANI) was synthesized. (b) The PFOS templates were removed by sonication for 4 h in a mixed solution of acetic acid and methanol at a v/v ratio of 1:6, leaving cavities of specific recognition sites. (c) PFOS samples were then dispensed to the surface of the integrated paper sensor, and the PFOS-MIP-PANI was then able to detect PFOS by the molecularly imprinted structures. PFOS targets and PFOS-MIP-PANI performed an equilibrium system of association and dissociation.

Figure 2

ATR-FTIR spectra of the surface of PFOS-MIP-PANI and NIP-PANI. NIP-PANI denotes the polyaniline synthesized in the absence of PFOS templates while the template removal process including sonication was still performed. The y axis based on the transmission mode has been adjusted in order to compare wavelengths of the characteristic peaks.

Figure 3

XPS spectra of the surface of PFOS-MIP-PANI and NIP-PANI. (a) F1s bonding. (b) N1s bonding.

Figure 4

SEM images of the surface morphology of: (a) NIP-PANI before sonication; (b) NIP-PANI after sonication and before the exposure to PFOS; (c) NIP-PANI after the exposure to PFOS; (d) PFOS-MIP-PANI before sonication; (e) PFOS-MIP-PANI after sonication and before the exposure to PFOS; (f) PFOS-MIP-PANI after the exposure to PFOS. The scale bar denotes 1 μm in length.

Figure 5

The calibration curve of PFOS-MIP-PANI exposed to various concentrations of PFOS aqueous solutions. The resistivity ratios were normalized to the resistivity ratio of DI water. The data points denote the averages of repeating measurements of DC resistance of at least three devices, from which the standard deviations are also calculated and marked as error bars.

Figure 6

The selectivity of PFOS-MIP-PANI to PFOS, PFBA, PFHxA, and PFOA. The concentration for each chemical in the aqueous samples was 70 ppt. The data were normalized to water resistivity. Each bar denotes the average of DC resistivity ratios measured from three devices, and the standard deviations are also calculated and marked as the error bars. The asterisks represent significance (p < 0.05) lower than the response of PFOS.

Figure 7

The transformation between the forms of polyaniline emeraldine salt and emeraldine base. A⁻ denotes the counter ion.