Recycled PETg embedded with graphene, multi-walled carbon nanotubes and carbon black for high-performance conductive additive manufacturing feedstock

Abstract

The first report of conductive recycled polyethylene terephthalate glycol (rPETg) for additive manufacturing and electrochemical applications is reported herein. Graphene nanoplatelets (GNP), multi-walled carbon nanotubes (MWCNT) and carbon black (CB) were embedded within a recycled feedstock to produce a filament with lower resistance than commercially available conductive polylactic acid (PLA). In addition to electrical conductivity, the rPETg was able to hold >10 wt% more conductive filler without the use of a plasticiser, showed enhanced temperature stability, had a higher modulus, improved chemical resistance, lowered levels of solution ingress, and could be sterilised in ethanol. Using a mix of carbon materials CB/MWCNT/GNP (25/2.5/2.5 wt%) the electrochemical performance of the rPETg filament was significantly enhanced, providing a heterogenous electrochemical rate constant, k⁰, equating to 0.88 (±0.01) × 10^-3 cm s^-1 compared to 0.46 (±0.02) × 10^-3 cm s^-1 for commercial conductive PLA. This work presents a paradigm shift within the use of additive manufacturing and electrochemistry, allowing the production of electrodes with enhanced electrical, chemical and mechanical properties, whilst improving the sustainability of the production through the use of recycled feedstock.

Fig. 1. (A) Schematic for the production of conductive recycled PETg additive manufacturing filament. (B) Photographs highlighting the flexibility of the conductive rPETg filament. (C) Thermogravimetric analysis of the conductive rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT). (D) Plot of the modulus obtained from tensile testing, with error calculated from three repeat samples.

Fig. 2. (A) Electrochemical activation profiles for the rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) and commercial conductive PLA. Activated using chronoamperometry in NaOH (0.5 M) at +1.4 V for 200 s and −1.0 V for 200 s. Performed using a nichrome wire counter electrode and Ag|AgCl reference electrode (3 M KCl). (B) The XPS C 1s region for the rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) after electrochemical activation. (C) The XPS O 1s region for the rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) after electrochemical activation. (D) Raman spectra for rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) after electrochemical activation. (E) SEM image of the rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) electrode surface before electrochemical activation. (F) SEM image of the rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) after electrochemical activation.

Fig. 3. (A) Scan rate study (5–150 mV s⁻¹) in [Ru(NH₃)₆]³⁺ (1 mM in 0.1 M KCl) performed with the rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) additive manufactured electrodes as the WE, nichrome coil CE, and Ag|AgCl as RE. (B) Cyclic voltammograms (25 mV s⁻¹) in [Ru(NH₃)₆]³⁺ (1 mM in 0.1 M KCl) performed with the rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT), (25 wt% CB, 5 wt% GNP), and (25 wt% CB, 5 wt% MWCNT) additive manufactured electrodes as the WE, nichrome coil CE, and Ag|AgCl as RE. (C) EIS Nyquist plots comparing rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) with commercial PLA additive manufactured electrodes as the WE. Performed in [Fe(CN)₆]^4−/3− (1 mM in 0.1 M KCl) with a nichrome coil CE, and Ag|AgCl as RE. (D) Cyclic voltammograms (50 mV s⁻¹) comparing rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) with commercial PLA additive manufactured electrodes as the WE. Performed in [Fe(CN)₆]^4−/3− (1 mM in 0.1 M KCl) with a nichrome coil CE, and Ag|AgCl as RE. (E) Cyclic voltammograms (50 mV s⁻¹) comparing rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) with commercial PLA additive manufactured electrodes as the WE after being previously used in [Ru(NH₃)₆]³⁺ (1 mM in 0.1 M KCl) and washed with deionised water. Performed in 0.1 M KCl with a nichrome coil CE, and Ag|AgCl as RE. (F) Plot showing the change in peak current and peak-to-peak separation for additive manufactured electrodes printed from rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) and commercial PLA after being sonicated in ethanol (70%) for 10 min. Calculations based on cyclic voltammograms (50 mV s⁻¹) comparing rPETg (25 wt% CB, 2.5 wt% GNP, 2.5 wt% MWCNT) with commercial PLA additive manufactured electrodes as the WE in [Ru(NH₃)₆]³⁺ (1 mM in 0.1 M KCl) with a nichrome coil CE, and Ag|AgCl as RE.