Co-Electrolysis of Simulated Coke Oven Gas with Carbon Dioxide Using a Solid Oxide Electrolysis Cell

A significant contributor to carbon and air-pollutant emissions is the steel industry, which is responsible for 2.8 Gt of CO 2 annual emissions and accounts for a quarter of industrial CO 2 emissions. A significant and valuable by-product of steelmaking is coke oven gas (COG), which is produced from coke-making via high-temperature dry distillation of coal in the absence of oxygen. COG typically consists of 55-60 vol% hydrogen (H 2 ), 23-27 vol% methane (CH 4 ), 5-8 vol% carbon monoxide (CO) and impurities such as H 2 S and tars [1]. Every ton of coke produced yields 300-360 m 3 of COG and an estimated 650 Mt of COG is produced in the steelmaking industry worldwide each year, with up to 50 % re-utilised within steelmaking [2]. However, the rest is flared, contributing to carbon emissions and wasting valuable and useful gases. In this paper, co-electrolysis of simulated COG (CH 4 /H 2 30/70 vol%) with carbon dioxide using a commercially available solid oxide fuel cell (SOFC) was investigated at 750 °C. The electrochemical performance of an anode supported button cell was characterised using open circuit potential measurements, current-voltage curves and electrochemical impedance spectroscopy. The product gas composition was analysed using quadrupole mass spectrometry (QMS). The QMS measurements show that CH 4 was initially converted at the open circuit potential into synthesis gas (H 2 /CO) via dry reforming of CH 4 and the reverse water gas shift reaction, yielding a gas mixture composed of 39.2 vol% H 2 , 38.3 vol% CO 2 , 22 vol% CO and low levels of unconverted CH 4 (see figure). Increasing the voltage from the OCP to 1.4 V caused the CO 2 to decrease and the CO to increase, yielding a gas mixture consisting of 42.6 vol% H 2 , 26.7 vol% CO 2 and 29.5 vol% COincreased the synthesis gas content (H 2 + CO) of the output gases by 44% relative to the OCP, with the levels of H 2 produced increasing by 9% over this voltage range. In addition, the level of unconverted CH 4 increased slightly, indicating that the CO2 conversion processes were competing with catalytic dry reforming of CH 4 . Possible reaction mechanisms are discussed. The I-V and EIS measurements established that cell electrical performance was increased by addition of CO 2 through dry reforming and also improved diffusion of the fuel gases through the anode. The effects of COG fuel variability have also been investigated and show that increasing the CH 4 /H 2 ratio from 10/90 vol% to 40/60 vol% o