Bi@Sn Core–Shell Structure with Compressive Strain Boosts the Electroreduction of CO2 into Formic Acid

Abstract As a profitable product from CO2 electroreduction, HCOOH holds economic viability only when the selectivity is higher than 90% with current density (j) over −200.0 mA cm−2. Herein, Bi@Sn core–shell nanoparticles (Bi core and Sn shell, denoted as Bi@Sn NPs) are developed to boost the activity and selectivity of CO2 electroreduction into HCOOH. In an H‐cell system with 0.5 m KHCO3 as electrolyte, Bi@Sn NPs exhibit a Faradaic efficiency for HCOOH (FEHCOOH) of 91% with partial j for HCOOH (j HCOOH) of −31.0 mA cm−2 at −1.1 V versus reversible hydrogen electrode. The potential application of Bi@Sn NPs is testified via chronopotentiometric measurements in the flow‐cell system with 2.0 m KHCO3 electrolyte. Under this circumstance, Bi@Sn NPs achieve an FEHCOOH of 92% with an energy efficiency of 56% at steady‐state j of −250.0 mA cm−2. Theoretical studies indicate that the energy barrier of the potential‐limiting step for the formation of HCOOH is decreased owing to the compressive strain in the Sn shell, resulting in the enhanced catalytic performance.


S4
Ag/AgCl electrodes were used as the counter electrode and reference electrode, respectively.
Bi@Sn NPs were synthesized via electroreduction of Bi 2 Sn 2 O 7 NPs in 0.5 M KHCO 3 at -0.8 V vs RHE for 1 h. The prepared carbon-paper electrode directly used as working electrode without further treatment. For Sn NPs, the working electrode was prepared with the same condition except the process of electroreduction.
Electrochemical measurements. The potentials were controlled via an Autolab potentiostat/galvanostat (CHI660E). All potentials were measured against the Ag/AgCl reference electrode and converted to the RHE reference scale on account of the equation: E (vs RHE) = E (vs Ag/AgCl) + 0.21 V + 0.0591 × pH. The LSV curves were performed in CO 2 -saturated and Ar-saturated 0.5 M KHCO 3 solution with the scan rate of 10 mV s -1 in H-cell system. The cyclic voltammetry (CV) measurement was conducted in CO 2 -saturated 0.5 M KHCO 3 solution using three-electrode cell equipped with a Ag/AgCl reference electrode, a graphite rod counter electrode and a glassy carbon working electrode (5 mm in diameter). C dl was determined by measuring the capacitive current associated with double-layer charging from the scan-rate dependence of CV. The CV ranged from 0.4 V to 0.5 V vs RHE. The C dl was estimated by plotting the Δj (Δj = j a -j c ) at 0.45 V vs RHE against the scan rates. The slope is twice that of the C dl value. Electrochemical impedance spectroscopy of Bi@Sn NPs and Sn NPs were recorded at -0.5 V vs RHE over a frequency range from 1000 kHz to 1 Hz with a sinusoidal voltage amplitude of 5 mV. CO 2 electroreduction was conducted in CO 2 -saturated 0.5 M KHCO 3 solution in the H-cell system under ambient condition at room temperature. After the feeding gas of CO 2 was purged into 0.5 M KHCO 3 solution for at least 30 min to remove residual air in the reservoir, controlled potential electrolysis was performed at each potential for 40 min. The gaseous products of CO 2 electroreduction were monitored via an online gas chromatography (GC) (GC2014, Shimadzu, Japan) equipped with a thermal conductivity detector (TCD) and Molsieve 5A column once every five minutes. The KHCO 3 solution after electrolysis was collected and analyzed on a 400 MHz NMR spectrometer to quantify liquid products. Standard curve was made by using HCOONa•2H 2 O and the internal standard (DSS). 0.5 mL of the KHCO 3 solution after electrolysis was mixed with the addition of 0.1 mL of D 2 O and 0.1 mL of 6 mM DSS solution as an internal standard. The peak-area ratio of the HCOOH peak to the DSS peak was compared to the standard curve to quantify the concentration of HCOOH.

S5
Electrolysis in the flow-cell system. The cathode and anode chambers were each 15 cm 3 in volume. Each chamber had an inlet and an outlet for electrolyte. Ag/AgCl reference electrode was located inside the cathode chamber. The windows for electrolysis were set to 1 cm × 1 cm.
The prepared carbon-paper electrode in H-cell system was directly used as GDE. Silicone gasket with a 1 cm × 1 cm window was placed between GDE and cathode chamber for sealing. CO 2 electroreduction was conducted in 2.0 M KHCO 3 solution in the flow-cell system under ambient condition at room temperature. During the measurements, CO 2 gas was directly fed to the cathode GDE at a rate of 20 sccm. The electrolyte was forced to continuously circulate through the chamber at a rate of 10 sccm.

The calculation method for the Faradaic efficiency.
The Faradaic efficiency for HCOOH production was calculated at a given potential as follows: V, the volume of the electrolyte. N, the number of electron transferred for product formation, which is 2 for HCOOH. F, Faraday constant, 96485 C mol -1 .

Q, quantity of electric charge integrated by i-t curve.
The calculation method for the yield rate of HCOOH.
The yield rate of HCOOH was calculated at a given potential as follow: ν HCOOH , the yield rate of HCOOH.
C HCOOH , the concentration of HCOOH.
V, the volume of the electrolyte. S, the geometric area of the electrode. t, the reaction time.
The calculation method for the energy efficiency for the conversion of CO 2 into HCOOH.
The energy efficiency for the conversion of CO 2 into HCOOH was calculated at a given potential as follow: Φ HCOOH , the energy efficiency for the conversion of CO 2 into HCOOH.
ΔE 0 HCOOH , the difference between the standard half reaction potentials for water oxidation (1.23 V vs RHE) and the reduction of CO 2 into HCOOH (-0.2 V vs RHE). ΔEHCOOH, the difference between the standard water oxidation potential and the working potential at the cathode.
Therefore, the Bi core in Bi@Sn NPs exhibited preferred orientations of (012) facets. As such, we constructed the structure model of Bi/Sn interface based on the Bi (012) (211) were taken into consideration. Figure S18 illustrates the lattice constants of Sn (101), Sn (200), Sn (211), and Bi (012) planes, respectively. Owing to the large difference in lattice constants, the Sn (200) and (211)  DFT calculation was carried out using the Vienna ab initio Simulation Package (VASP). [S1] The projector augmented wave method in conjugation with a generalized gradient approximation of exchange-correlation functional in the Perdew-Burke-Ernzerhof (PBE) form was adopted. [S2] The one-electron wave functions were expanded using a plane-wave basis set with an energy cut off of 400 eV. [S3] Sn (101) slabs with (2 × 2) surface unit cells and 4 atomic layers was performed.
In the geometric optimization, atomic positions were fully relaxed until the associated forces are less than 0.02 eV/Å. 4 × 5 × 1 Monkhorst-Pack sampled k-points were used in all of slabs system.
For projected density of states (PDOS) plotting, a larger set of k-points (8 × 10 × 1) were used. A S7 dipole correction was imposed to eliminate the error induced by the artificial electrostatic interaction between the surface dipole moments of asymmetric repeated slabs. A vacuum region greater than 14 Å was applied to avoid interactions between the neighboring slabs caused by the periodic boundary conditions.
In-situ attenuated total reflection infrared (ATR-IR) spectroscopy was carried out on a Nicolet iS50 with a wavenumber resolution of 8 cm -1 at room temperature.          Figure S19. PDOS of surface atoms on compressive Sn slab and pristine Sn slab.  Figure S20. Photograph of the flow-cell system in this work.