A metal-free electrocatalyst for carbon dioxide reduction to multi-carbon hydrocarbons and oxygenates

Electroreduction of carbon dioxide into higher-energy liquid fuels and chemicals is a promising but challenging renewable energy conversion technology. Among the electrocatalysts screened so far for carbon dioxide reduction, which includes metals, alloys, organometallics, layered materials and carbon nanostructures, only copper exhibits selectivity towards formation of hydrocarbons and multi-carbon oxygenates at fairly high efficiencies, whereas most others favour production of carbon monoxide or formate. Here we report that nanometre-size N-doped graphene quantum dots (NGQDs) catalyse the electrochemical reduction of carbon dioxide into multi-carbon hydrocarbons and oxygenates at high Faradaic efficiencies, high current densities and low overpotentials. The NGQDs show a high total Faradaic efficiency of carbon dioxide reduction of up to 90%, with selectivity for ethylene and ethanol conversions reaching 45%. The C2 and C3 product distribution and production rate for NGQD-catalysed carbon dioxide reduction is comparable to those obtained with copper nanoparticle-based electrocatalysts.

Multi-carbon oxygenates of ethanol (C2H5OH), acetate (AcO -), and n-isopropanol (n-PrOH). The Cu nanoparticles data were adopted from reference (5).  . A small percentages of 12 C contained products were also obtained as evidenced by a tiny 1 H NMR peak in between the two coupling 1 H NMR peaks in the products of HCOOand CH3COObecause some 12 CO2 involves in the 13 CO2 gas.

Graphene quantum dots synthesis
Improved Hummer's method was used to synthesize graphene oxide (GO) from SP1 Graphite powder. The procedure involves mixing 3 g of graphite powder with 18 g of KMnO4 and 3 g NaNO3 followed by slow addition of 360 mL H2SO4. It was then kept under stirring for 12 hours. After that the solution was poured onto ice (made from 500 mL DI water) and 14 mL H2O2 was added carefully. After stirring for an hour, the solution was allowed to stand for a day. The yellowish brown slurry that settled down was collected and the procedure for GO synthesis was again repeated on the collected product for further increasing the oxidation degree of GO, which is very important for controlling the final size of graphene quantum dots (GQDs). The collected slurry after the second oxidation step having a volume of ~300 mL, was then washed using 30% HCl, ethanol and DI water in sequence to remove any impurities. 500 mL 30% HCl was added into the 300 mL slurry, and stirred magnetically for 5 min under 1000 rpm. The subsidence was collected after centrifuging. The washing with HCl was repeated three times. After HCl washing, the subsidence was added with 500 mL ethanol, stirred for 5 min under 1000 rpm, and collected after centrifuging. The washing with ethanol was repeated three times. At last, the sample was washed with 500 mL DI water for three times. After first round DI water washing, the conductivity of 0.2 mg/mL GO solution ranged 50-100 µs/cm. Repeat the DI water washing process until the conductivity decreases down to 1.600 µs/cm, which is very close to the 1.014 µs/cm of DI water.
Moreover, after washing, the trace metals concentration is ultra-low, like Na (0.06 at%) and Mn (0.05 at%) while no K is detected as shown in the XPS analysis.
The resulted GO was used as the precursor for hydrothermal alike synthesis of Ndoped and pristine graphene quantum dots (GQDs). In the case of N-doped GQDs, typically 300 mg GO was dispersed in 30 ml dimethylformamide (DMF) and then sonicated in bath ultrasonicator for 30 min. 1 Afterwards, the GO suspension was transferred to a 50 ml PTFE liner. The NGQDs was formed in a hydrothermally analogous process at 200 °C for 10 h during which GO was exfoliated and cut at the weak sites with oxygen containing groups, and simultaneously doped by N into the carbon lattice with N source from DMF and its derived produce of dimethylamine, methylamine and ammonia. The pristine GQDs were synthesized using the same GO precursors and process except replacing DMF by a mixture of IPA and H2O (1:1 by volume). The ratio is optimized to match the surface energy component of IPA/H2O cosolvent to that of GO, so that to maximize the exfoliation and cutting efficiency. 2,3 The N-doped reduced graphene oxide was prepared in a tube furnace at 800 °C while flowing ammonia for 1 h.

Gas diffusion electrode preparation
The cathodes were prepared using an air-brush method as previously reported. 4 Cathode catalyst inks for QDs were prepared by firstly mixing QDs solution (10 ml) and Nafion® solution (26 μL, 5 wt%, Fuel Cell Earth), and then being sonicated for 5 min.
The cathode ink for NRGOs were prepared in the same manner except using NRGOs The flow rate was set at 0.5 mL min -1 when applying cell potentials more negative than -2 V, otherwise using a slower flow rate of 0.1 mL min -1 to increase the concentration of the liquid products at a relative lower current density.
For each applied voltage, after the cell reached steady state, 1 mL of the effluent gas stream was periodically sampled and diverted into a gas chromatograph (Thermo Finnegan Trace GC) equipped with both the thermal conductivity detection (TCD) and flame ionization detector (FID), and a Carboxen 1000 column (Supelco). Three successive injections of effluent gas stream with 1.6 mins interval between each injection were directed into the GC wherein the second and third injection occurred before the first was allowed to elute to save the overall running time. The gas peaks for respective injection are separated as shown in Figure S5A. Helium as the carrier gas flows at a rate of 20 SCCM. Meanwhile, the exit catholyte was collected at each applied voltage followed by identifying and quantifying using 1 H NMR (nuclear magnetic resonance The onset potential is defined as the lowest cathode potential at which product was detected from either GC or NMR. The gas products from 13 CO2 were identified by VG 70S double-focusing magnetic sector mass spectrometer. The Faradaic efficiency (FE) for a specific product is calculated using the following