Aerosol Electroanalysis by PILSNER: Particle-into-Liquid Sampling for Nanoliter Electrochemical Reactions

Abstract

Particle-into-liquid sampling (PILS) has enabled robust quantification of analytes of interest in aerosol particles. In PILS, the limit of detection is limited by the factor of particle dilution into the liquid sampling volume. Thus, much lower limits of detection can be achieved by decreasing the sampling volume and increasing the surface area-to-volume ratio of the collection substrate. Unfortunately, few analytical techniques can realize this miniaturization. Here, we use an ultramicroelectrode in a microliter or smaller sampling volume to detect redox active species in aerosols to develop the technique of Particle-into-Liquid Sampling for Nanoliter Electrochemical Reactions (PILSNER). As a proof-of-concept to validate this technique, we demonstrate the detection of K₄Fe(CN)₆ in aerosol particles (diameter ∼0.1-2 μm) and quantify the electrochemical response. To further explore the utility of the method to detect environmentally relevant redox molecules, we show PILSNER can detect 1 ng/m³ airborne Pb in aerosols. We also demonstrate the feasibility of detecting perfluorooctanesulfonate (PFOS), a persistent environmental contaminant, using this technique. PILSNER is shown to represent a significant advancement toward simple and effective detection of a variety of emerging contaminants with an easily miniaturizable and tunable electroanalytical platform.

Figure 1

(A) Illustration of the PILSNER setup during the analysis of Pb aerosol via anodic stripping voltammetry and PFOS aerosol via differential pulse voltammetry. (B) Schematic of the PILSNER system. For PFOS analysis, the pseudoreference/counter electrode was modified with a well, as described below.

Figure 2

PILSNER i–t curves of various sample volumes exposed to 400 mM ferrocyanide aerosol. The sample volume was exposed to aerosol from t = 30 s to t = 60 s. Baseline adjusted.

Figure 3

PILSNER-anodic stripping voltammogram of water and lead(II) nitrate aerosols. Listed concentrations indicate concentrations in the aerosol particles. Baseline adjusted.

Figure 4

(A) DPV response in 20 μL droplets of 1× PBS with 2 mM ferrocene methanol and increasing concentrations of PFOS using a MIP-modified Au ultramicroelectrode, and the same glassy carbon chip quasi-reference electrode and counter used throughout this work. All DPVs were normalized such that the starting current was 0 nA. (B) The resulting calibration curve from the DPVs in panel A, illustrating the dependence of the blank subtracted current response (i – i_o), on the natural log of PFOS concentration, with the x-axis on a logarithmic scale. The equation of best fit was y = 0.06(±0.01) ln(x) + 0.50(±0.04) and the R² value was 0.93 (±0.04). All points were statistically different according to a one-way ANOVA test (n = 3). The LOD of PFOS detection in the collector droplet was determined to be 0.3 pM based on the sensitivity of the calibration curve and an assessment of that slope against 3 times the noise of the blank (0 nM) peak current measurement.

Figure 5

DPV responses in a 20 μL droplet of 1× PBS with 2 mM ferrocene methanol exposed to 1 mM PFOS aerosol for 1, 2, 3, and 5 min compared to a precollection pulse. All DPVs were normalized such that the starting current was 0 nA.