Tungsten Oxide Coated Liquid Metal Electrodes via Galvanic Replacement as Heavy Metal Ion Sensors

Abstract

Gallium liquid metals (LMs) like Galinstan and eutectic Gallium-Indium (EGaIn) have seen increasing applications in heavy metal ion (HMI) sensing, because of their ability to amalgamate with HMIs like lead, their high hydrogen potential, and their stable electrochemical window. Furthermore, coating LM droplets with nanopowders of tungsten oxide (WO) has shown enhancement in HMI sensing owing to intense electrical fields at the nanopowder-liquid-metal interface. However, most LM HMI sensors are droplet based, which show limitations in scalability and the homogeneity of the surface. A scalable approach that can be extended to LM electrodes is therefore highly desirable. In this work, we present, for the first time, WO-Galinstan HMI sensors fabricated via photolithography of a negative cavity, Galinstan brushing inside the cavity, lift-off, and galvanic replacement (GR) in a tungsten salt solution. Successful GR of Galinstan was verified using optical microscopy, SEM, EDX, XPS, and surface roughness measurements of the Galinstan electrodes. The fabricated WO-Galinstan electrodes demonstrated enhanced sensitivity in comparison with electrodes structured from pure Galinstan and detected lead at concentrations down to 0.1 mmol·L^-1. This work paves the way for a new class of HMI sensors using GR of WO-Galinstan electrodes, with applications in microfluidics and MEMS for a toxic-free environment.

Keywords: galvanic replacement; heavy metal ions; liquid metals; photostructuring; sensor.

Figure A1

Schematic overview of the galvanic replacement of a 10 µL Galinstan droplet immersed in different concentrations (10, 50, 100, 200, and 400 mM) of 1 mL W salt solution at 0, 24, 48, and 72 h with the corresponding optical microscopy images. (ai–aiv) No visible change in droplet shape was observed for the 10 mM W salt droplets over 72 h. (bi–eiv) Definite change from droplet to flattened shape for 50 mM to 400 mM Galinstan droplets indicated successful galvanic replacement. (Scale bar 1 mm).

Figure A2

SEM images of Galinstan droplets after 24 h in different concentrations of W salt solutions with supporting EDX spot scan data and XRF measurements. (a–e) SEM images of Galinstan droplets with the measured spots (1-Galinstan core, 2-W) for EDX and the corresponding elemental concentration in wt. % and error % (Scale bar: 50 µm). (f) Representative EDX spot scan data for spot 1 indicated the expulsion of the Galinstan core due to the vacuum conditions inside the SEM. (g) Plot derived from EDX spot scan data representing the increase in W amount with increasing salt concentration for 24 h. (h) The area density vs. salt concentration plot obtained from XRF measurements for Gallium and Tungsten.

Figure A3

SEM-EDX mapping images for the exfoliated skin of Galvanically replaced Galinstan droplets in 10 mM and 400 mM for 24 h (a,b) and 72 h (c,d), respectively, with the corresponding element weight.%. Tungsten amount for all four exfoliated skins was similar ~70 wt.% with ~20 wt.% Oxygen. (Scale bar: 20 µm).

Figure A4

XRD plots for Galinstan powder and 400 mM 24 h treated W-salt Galinstan powder.

Figure 1

Schematic overview of the photostructuring of Galinstan electrode via lithography followed by galvanic replacement in W salt solution resulting in WO-Galinstan electrode.

Figure 2

SEM images of an untreated bare Galinstan electrode and a galvanically treated WO-Galinstan electrode in different W salt concentrations for 24 h along with the corresponding EDX spot scan data. (A–F) Representative EDX spot scan data and the corresponding elemental composition mean in wt.% for four measured spots on each sample with the standard deviation, indicating a gradual increase in the W amount with the highest W amount for the 400 mM W salt-treated Galinstan electrode (Scale bar: 10 µm).

Figure 3

Characterization of Galinstan and WO-Galinstan electrodes. (A) Bright-field image of a Galinstan electrode with the surface analysis via WLI alongside the measured profile (right), denoting surface roughness of Ra = 42 nm, supported by the SEM image. (B) Bright-field image of a treated WO-Galinstan electrode in 400 mM W salt with the surface analysis via WLI alongside the measured profile (right), denoting surface roughness of Ra = 193 nm, supported by the SEM image clearly indicating a rough surface. (C,D) Surface composition analysis of untreated Galinstan and treated WO-Galinstan electrodes using XPS, confirming the presence of W with a W4f7 peak.

Figure 4

Differential Pulse Stripping Voltammetry (DPSV) experimental setup for Lead (II) detection. (A) DPSV setup schematic with the WO-Galinstan electrode as a working electrode, Ag/AgCl as reference electrode, and platinum wire as a counter electrode, with a positive scan of −800 mV to 0 mV at 3 mv·s⁻¹. (B,C) The resulting DPSV voltammograms with the corresponding output current for Lead (II) concentrations of 0.1, 0.2, and 10 mmol·L⁻¹ and buffer solution for the Galinstan and WO-Galinstan sensors, respectively. (D) Recorded current stability for three consecutive sweeps compared with the fifteenth consecutive sweep for the detection of 0.2 mmol·L⁻¹ Pb²⁺ with the WO-Galinstan electrode. (E,F) SEM images indicated a definite change in the surface morphology of the Galinstan and WO-Galinstan electrodes pre (i) and post HMI (ii) sensing, respectively.