Valorisation of e-waste by means of direct liquefaction

Abstract

Global E-waste generation has increased rapidly in recent decades owing to a decrease in the lifespan of electronic equipment and an exponential increase in supply and demand. Due to inadequate treatment and disposal practices, the majority of e-waste is landfilled, polluting the air, soil, and water. E-waste can be a valuable source of secondary materials, with most end processing methods aimed at recovering the valuable metals. However, many of these procedures are not feasible without government support and cause secondary pollution in the form of pollutants released by emissions and effluents. Furthermore, the presence of organic components and low metal concentrations in end-products impedes economic viability. As a result, there is a need to investigate novel end-processing routes that can overcome the drawbacks of present technologies and lead to a more environmentally friendly and economically viable treatment method for e-waste. One such novel route is the thermochemical depolymerisation process known as hydrothermal liquefaction or direct liquefaction. The potential depolymerisation and conversion of organic fractions into other valuable organics, as well as the potential immobilisation of metals through co-processing with lignocellulosic biomass, could result in liquefaction becoming a promising option for e-waste valorisation and end-stage treatment. This study aimed to determine the suitability of direct liquefaction as a treatment method for locally obtained printed circuit board (PCB) e-waste co-processed with lignosulphonate biomass. The aim was achieved by studying the effect of hydrothermal liquefaction process conditions while co-processing PCB e-waste with lignosulphonate biomass in an SS-316 batch reactor. The effect of liquefaction temperature (250-330°C), residence time (20-80 min), and waste loading (0-100 wt.% of feed) was studied by analysing the recovery of metals and formation of organic compounds in the residues, and aqueous phases obtained from processing. The intertwining effect of residence time and temperature on the product yields was determined through calculated severity factors. The formation of organics was established by elemental analysis, proximate analysis, total phenolic analysis, Fourier-transform Infrared Spectroscopy (FTIR), Gel Permeation Chromatography (GPC), and Gas Chromatography-Mass Spectrometry (GC-MS). Metal recovery was determined through the digestion of reactor products, followed by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) to determine the concentration of metals and mineral elements. This study found low yields of bio-oils at all process conditions, however, the oils mostly consisted of phenolic compounds. Residue products consisted of inorganics, such as ceramics and metals, as well as unreacted epoxy resin and heavy carbon molecules. Temperature influenced the yields and composition of products more than residence time. Generally, longer residence times mostly affected the products by allowing HTL reactions to progress, whereas temperature heavily influenced the rate of these reactions. Both the bio-oil yield and the phenolic content increase with increasing temperature and reached a turning point as the rate of cracking reactions overtook the recombination. At 250°C, longer residence times increased the bio-oil and residue yields, however at 330°C a dynamic equilibrium between decomposition and repolymerisation reactions was reached. Furthermore, at lower temperatures, longer residence times increased the phenolic content of bio-oils. The optimum conditions for bio-oil production were determined as 300°C for 80 min. Nearly all of the metals remained in a stable state within the residues, with only small concentrations detected in the bio-oils and aqueous phases after HTL. It was concluded that the migration from the residues was in relation to the initial concentration of each metal in the feedstock. Both temperature and residence time produced mostly inconsistent trends. At both short and long residence times, several metals (Au, Cu, Fe, Ni, Sn and Zn) showed high concentrations in the residues at 250°C, as a result of the slow decomposition rate of the waste PCBs. However, as the temperature increased, an initial decrease in concentration was noted as more metals were released, after which the concentrations slightly increased, possibly due to the adsorption from the biochar structures. Longer residence times at 250°C decreased most of the metal concentrations, however, no conclusive trends were seen at other temperatures. It was observed that increasing the waste loading increased both the bio-oil and residue yields. The carbon content of the residues decreased from 49 wt.% at 0 wt.% PCB loading to 5 wt.% at a 100 wt.% PCB loading, confirming that almost all of the organic epoxy resin was depolymerised by HTL. Furthermore, the phenolic content of the bio-oils increased from 6 to 93 area%, confirming that the epoxy resin was primarily depolymerised and recombined into useful phenolic compounds. Increasing the waste loadings resulted in a non-linear increase for most metal concentrations, with the metal concentrations in the residues at 90 and 100 wt.% waste loadings being significantly higher than the original feedstock concentrations, with less than 6 wt.% carbon. This study revealed that hydrothermal liquefaction is an effective treatment approach for waste PCBs by confirming the organics in waste PCBs transform to valuable phenolic compounds while the majority of metals remain stable during processing, resulting in the production of metal-rich residues.