Electronic Supplementary Material (ESI) for Sustainable Energy & Fuels.

This journal is © The Royal Society of Chemistry 2020

Supporting Information

In-situ morphology transformation of bismuth-based catalysts for

effective electroreduction of carbon dioxide

Xiaowei Ana, Shasha Lib, c, Xiaoqiong Haob, Xiao Dud, Tao Yua, Zhongde Wangd, Xiaogang Haod,

Abuliti Abudulaa, and Guoqing Guan*,a, b

aGraduate School of Science and Technology, Hirosaki University, 1-Bunkyocho, Hirosaki 036-

8560, Japan

bEnergy Conversion Engineering Laboratory, Institute of Regional Innovation (IRI), Hirosaki

University, 2-1-3, Matsubara, Aomori 030-0813, Japan.

cCollege of Chemical and Biological Engineering, Taiyuan University of Science and Technology,

Taiyuan 030012, China

dDepartment of Chemical Engineering, Taiyuan University of Technology, Taiyuan 030024,

China.

*Corresponding author

E-mail: [email protected]

1
Physicochemical Characterizations

The morphology and microstructure of the obtained electrocatalysts were observed by a

scanning electron microscope (SEM, Hitachi SU8010) equipped with a Horiba Scientific energy

dispersive spectrometer (EDS) analyzer, and the nanostructure was analyzed with a JEM-2100F

transmission electron microscope (TEM, JEOL, Japan). The crystalline structures of the

electrocatalysts were determined by X-ray diffraction (XRD, RigakuSmartLab X-Ray

Diffractometer) using a Cu-Kα (λ=1.5405 Å) radiation source. The surface compositions and

their valence states on the catalyst were characterized by using a X-ray photoelectron

spectroscopy (XPS) (Thermo ESCALAB 250XI).

Electrochemical Characterizations

Linear sweep voltammetry (LSV) was carried out in a potential range of -0.8 to -2 V at a scan

rate of 10 mV s-1 in either N2 or CO2-saturated 0.1 M KHCO3 solutions. Electric double layer

capacitance was determined by cyclic voltammetry (CV) measurements at various scan rates

(40, 80, 120, 160, 200 mV s-1) in N2-saturated 0.1 M KHCO3 solution. The double layer

capacitance of each electrode was calculated using Equation (1):

J=C×dV/dt (1)

2
where J is the current density, C the capacitance, and dV/dt the scan rate.

Electrochemical impedance spectroscopy (EIS) was performed within a frequency range of 0.01

Hz - 0.1 MHz in a CO2-saturated 0.1 M KHCO3 solution at -1.2 V.

Calculation of the Overpotential

The overpotential in Tafel curve was calculated as the difference between the cathodic

potential and the standard (equilibrium) reduction potential (𝜂 = |𝐸𝐶𝑑 −𝐸𝑒𝑞|). The standard

reduction potential for converting CO2 into HCOOH in aqueous 0.1 M KHCO3 electrolyte

solution is -0.09 V vs. RHE.1 The formula used to convert potentials vs. Ag/AgCl (1M) to vs. RHE

is: E vs. RHE = E (V) vs. Ag/AgCl (saturated KCl) + 0.198 (V) + 0.059 (V) × pH (6.8).

Calculation of the electrochemically active surface area (ECSA)

The ECSA of electrode was calculated from the double layer capacitance (Cdl) according to

equation 2:

ECSA = CdL/α (2)

where α is the specific capacitance of the sample or the capacitance of an atomically smooth

planar surface of the material per unit area under identical electrolyte conditions.

Product Analysis

3
Gas products were analyzed with a gas chromatography (GC, Agilent 7890A GC system, USA), by

which CO and H2 were detected with two thermal conductivity detectors (TCD), respectively.

Liquid products were analyzed using a Shimadzu high performance liquid chromatography

(HPLC) equipped with SPD-20AV and RID-10A detectors. Supelcogel C610-H (30 cm×7.8 mm,

Sigma) and H3PO4 (0.05 %) solution were used as the column and mobile phase, respectively.

The temperature of the oven CTO-20A was set at 40 ºC. The faradaic efficiency of HCOOH, CO

or H2 was calculated as follows:

Faradaic efficiency = 2F×n/Q (3)

where F is the faraday constant, n the total amount of products (in moles), and Q the amount

of total charge (in Coulomb).

DFT Computational Details

Bi and Bi2O2CO3 crystal structures with the ICSD numbers of 053797 and 036245 respectively

were imported. All the models were optimized by using the spin-polarized DFT computations

with an all-electron method by the DMol3 module in Materials Studio software.4-5 The electron

exchange-correlation functions of generalized gradient approximation (GGA) and Perdew-

Burke-Ernzerhof (PBE), and basis set of double numerical plus polarization (DNP) were used for

these calculations.2 Specifically, the van der Waals correction was employed by using DFT-D3 to

accurately describe the long-range electrostatic interactions. The solvent effects were also

4
applied by using a conductor-like screening model (COSMO)4 with the H2O dielectric constant of

78.54.

Other specific parameters selected in this study were consistent with the data reported in

the literature.5, 6 In detail, the convergence criterion was selected as 10-6 a.u. with the

maximum force of 0.002 a.u. Å-1 and the maximum displacement of 0.005 Å. The k points were

set as 5×5×1 by using the Monkhorst-Pack method.7 The real-space global orbital cutoff radius

was chosen as 4.7 Å in all the computations to ensure high quality.

The free energy of electrochemical reduction of CO2 was calculated based on a

computational hydrogen electrode (CHE) model,8 in which the chemical potential of a

proton/electron (H+ + e-) in solution is equal to the half chemical potential of a gas-phase H2.

The adsorption energy of adsorbate was defined as:

Ead = Esurface-adsorbate – Esurface – Eadsorbate (4)

where Esurface-adsorbate, Esurface, and Eadsorbate are the DFT total energy for the adsorbate on the

surface, surface, and adsorbate itself, respectively.

The change in Gibbs free energy (ΔG) was evaluated by the following equation:

ΔG = ΔE + ΔEzpe – TΔS – neU (5)

5
where the ΔE, ΔEzpe and ΔS are the differences of DFT energy, zero point energy and entropy

between products and reactants, respectively. T was set at 298.15 K in this work. Ezpe can be

obtained by the vibrational frequency analysis over all normal modes ν (Ezpe = 1/2Σћν). The

entropies of the free molecules (CO, CO2, HCOOH, H2, and H2O) were obtained from the NIST

database. The configurational entropy in the adsorbed state was neglected by fixing the

surfaces according to the way in most of the literatures.9-11 U is the applied potential vs. normal

hydrogen electrode (NHE), e is the elementary charge transferred and n is the number of

proton-electron pairs transferred.

Investigation of morphology transformation conditions

In order to confirm the formation mechanism of Bi2O2CO3 nanosheets, a series of treatment

processes were performed and the morphology of electrode surface was observed the by SEM.

Because the in-situ treatment process consisted of two steps, the first was electrochemical

treatment step, and the second was the dry step in the air. In order to confirm whether two

steps were necessary, a set of parallel experiments were performed. Four Bi/Cu foam

electrodes were treated with the following four different treatment methods and then the

surface morphologies were observed.

6
The obtained SEM images for the surfaces of four electrodes are shown as follows:

7
Fig. S1 SEM images of surfaces of electrode (a) A, (b) B, (c) C and (d) D.

The freshly prepared Bi/Cu foam electrode was directly dried in the air as the initial electrode

A for comparison. The two same electrodes of A were only soaked and electrochemically

treated in the electrolyte for 2 hours using the above methods, respectively, and then the two

electrodes were taken out and exposed to the air for 12 hours, which are called as electrodes B

and D respectively. As shown in the Fig. S1a, b and d, significant change was observed for the

electrochemically treated electrode D, and the electrodes without electrochemical treatment

had only a slight surface change. Another electrode A was electrochemically treated at first and

8
then placed in a vacuum oven to avoid contact with air to obtain electrode C. As shown by the

SEM image, the electrode surface became rough after the electrochemical treatment.

For the electrode B, since there was no electrochemical treatment process, in this case, as

reported in the literature, Bi2O2CO3 was slowly formed on the electrode by reaction with the

electrolyte (KHCO3) remained on the surface as follows.12

4Bi + 3O2 + 2H2O ⇌ 4BiO+ + 4OH- (6)

2BiO+ + CO32- ⇌ Bi2O2CO3 (7)

As such, even without the electrochemical treatment, the electrode B still underwent a slight

morphological transformation (Fig. S1b).

For the electrode C, as reported in the literature, in the process of electrochemical reduction

of CO2, some reactions like equations of 8-10 may occur on the surface of the electrode at first,

thereby accumulating a large amount of OH- on the surface, resulting in local alkalinity.13

CO2 + H2O + 2e- ⇌ HCOO- + OH- (8)

CO2 + H2O + 2e- ⇌ CO + 2OH- (9)

2H2O+4e- ⇌ H2+2OH- (10)

Bi3++ OH- ⇌ Bi(OH)3 (11)

2Bi(OH)3+CO2 ⇌ Bi2O2CO3+3H2O (12)

9
Then, the OH- reacted with the Bi3+ present on the surface of the electrode to form Bi(OH)3.

Since electrode C was directly dried in a vacuum oven after electrochemical treatment, the

further reaction between Bi(OH)3 formed on the surface and CO2 in the air was avoided. In this

case, since the solubility of CO2 in the aqueous solution is very low, only a small amount of

Bi2O2CO3 could be formed on the electrode surface. As a result, the formation of Bi(OH)3 and a

small amount of Bi2O2CO3 species resulted in a rough electrode surface but without further

morphological transformation (Fig. S1c).

However, for the electrode D, after electrochemical treatment, since the electrode was

directly exposed to air, Bi(OH)3 should react with CO2 in the air to form Bi2O2CO3 as indicated by

the equation 12. Meanwhile, after the electrochemical treatment, the rough electrode surface

could also accelerate the conversion of the metal Bi to Bi2O2CO3 as indicated by the equations

of 6-7. These caused its surface to be transformed to petal-shaped Bi2O2CO3 nanosheets

completely (Fig. S1d). In addition, it should be noted that, in order to make the morphological

transformation more fully, the electrolyte remaining on the electrode surface after

electrochemical treatment should not be washed away.

10
Fig. S2 Calculated XRD patterns of (a) Bi and (b) Bi2O2CO3 crystalline structures based on the

models and compared with the standard PDF cards from ICDD database.

Fig. S3 Optimized structures of different intermediates on the Bi (012) and Bi2O2CO3 (001)

facets.

11
Fig. S4 SEM images of Cu foam with different scales.

Fig. S5 SEM images of Bi/CP electrode (a, b) before and (c, d) after the in-situ treatment.

12
Fig. S6 SEM images of Bi/Cu foam-in situ treated electrode with different magnification scales.

13
Fig. S7 EDS analysis results of (a) Bi/Cu foam and (b) Bi/Cu foam-in-situ treated electrodes.

14
Fig. S8 EDS analysis results of Bi/CP electrode for (a) bulk and (b) nanosheet structures. It can

be seen that the area where the morphological transformation has taken place has a higher

oxygen content.

Fig. S9 (a) XRD patterns and (b) high-resolution Bi 4f XPS spectra of Bi/CP and Bi/CP-in-situ

treated electrodes.

15
Fig. S10 (a, b and c) Determination of electrochemical double layer capacitance and (d)

electrochemical impedance spectroscopy (EIS) tests for Bi/Cu foam and Bi/Cu foam-in-situ

treated electrodes.

16
Fig. S11 Linear sweep voltammetry (LSV) curves of the Cu foam, Bi/Cu foam and Bi/Cu foam-in-

situ treated electrodes in either N2-saturated (dashed) or CO2-saturated (solid) 0.1 M KHCO3

solutions at a scan rate of 10 mV s-1.

Fig. S12 The stability test of the Bi/Cu foam-in-situ treated electrode continuously operated at -

1.6 V in a CO2 saturated 0.1 M KHCO3 solution (without refreshing) for 20 h.

17
Fig. S13 (a) SEM and (b) TEM images of Bi/Cu foam-in-situ treated electrode after 20 h

electrocatalysis.

Fig. S14 (a) XRD patterns and (b) high-resolution Bi 4f XPS spectra of Bi/Cu foam-in-situ treated

electrode before and after 20 h electrocatalysis.

18
Fig. S15 Projected density of states (PDOS) of C of *COOH intermediate and Bi atoms on the (a)

Bi(012) and (b) Bi2O2CO3(001) facets. The Fermi level is denoted with a red dashed line.

Fig. S16 Calculated free energy diagrams for the H2 evolution on Bi(012) and Bi2O2CO3(001)

facets.

19
Fig. S17 Calculated free energy diagrams for the electroreductions of *HCO3 to HCOOH and CO

on (a) Bi(012) and (b) Bi2O2CO3(001)_ Bi2O2 terminal and (c) Bi2O2CO3(001)_CO3 terminal facets.

For Bi2O2CO3, its (001) crystal plane has two terminals, the Bi2O2CO3(001)_CO3 terminal can be

combined with a *H to form *CO3H and then further reduced.

20
Table S1 The quantitative results of XPS spectra for the Bi/Cu foam electrode before and after
in-situ treatment.

At %
Species
Bi/Cu foam Bi/Cu foam-in-situ treated

Bi 43.36 28.41

O 52.91 69.32

Cu 3.73 2.27

Table S2 The lattice spacings of Bi and Bi2O2CO3.

Bi Bi2O2CO3
Facet Lattice spacing d (Å) Facet Lattice spacing d (Å)
003 3.954 002 6.841
101 3.737 011 3.720
012 3.280 004 3.420
104 2.369 013 2.952
110 2.273 110 2.734
015 2.032 020 1.933

Table S3 jHCOOH at -1.2 V (vs. Ag/AgCl, -0.6 V vs. RHE), measured double layer capacitance, and

jHCOOH per ECSA of the Bi/Cu foam and Bi/Cu foam-in-situ treated electrodes.

j
j C HCOOH
HCOOH dl
Sample name -2 -2 per ECSA
(mA cm ) (mF cm ) -2
(mA cm )
Bi/Cu foam 0.08 1.12 0.07*α
Bi/Cu foam-in-situ
0.45 4.02 0.11*α
treated

The α (mF cm-2) represents the specific capacitance of bismuth.

21
Table S4 The calculated binding energies (Eads, eV) of various CO2RR species on Bi(012) and
Bi2O2CO3(001) facets.

Binding energy (eV)

Species
*COOH *OOCH

Bi (012) -1.41 -2.33

Bi2O2CO3 (001) -1.77 -3.41

22
Table S5 Comparison of electrochemical reduction of CO2 performance for Bi-based catalysts.

Applied Faradaic Current

Electrolyt
Electrocatalyst potential effciency density Ref
e
(V) (%) (mA cm-2)
-1.7 V vs.
0.1 M Ag/AgCl
Bi/Cu foam 92.3 10.29 This work
KHCO3 (-1.1 V
vs. RHE)
0.1 M -0.4 V vs. Nano Energy, 2017,
Bi nanoflakes 79.5 -
KHCO3 RHE 39, 44-52.
Electrochemistry
0.5 M -1.50 V 3.7 mA
Bi/BiOCl 92 Communications,
KHCO3 vs. SCE mg-2
2014, 46, 63-66.
Journal of CO₂
Size tunable nano- 0.5 M -1.60 V
98.4 9.7 Utilization, 2017, 20,
Bi KHCO3 vs. SCE
328-335.
Journal of CO₂
Oxide-derived Bi 0.5 M -0.92 V jHCOOH= 8.3
82 Utilization, 2017, 19,
films KHCO3 vs. RHE mA cm-2
276-283.
Ultrathin Bismuth 0.5 M -1.0 V vs. ChemSusChem, 2018,
Over 90% 12.5
nanosheets NaHCO3 RHE 11, 848-853.
-1.70 V Applied Surface
0.1 M
Nano-sized Bi vs. 91.3 6 Science, 2017, 393,
KHCO3
Ag/AgCl 191-196
Applied Catalysis B:
0.5 M -1.45 V
Bi45/GDE 90 4 Environmental, 2017,
KHCO3 vs. SCE
218, 46-50.
Angewandte Chemie
P-orbital
0.5 M -1.16V International Edition,
delocalization 95 55
KHCO3 vs. RHE 2018, 57, 16114-
(POD)-Bi
16119.
0.5 M -0.74 V ACS Catalysis, 2017, 7,
Bi Dendrite 89 2.6
KHCO3 vs. SCE 5071-5077.
Lattice-dislocated Energy &
Bismuth 0.5 M -0.69 V Environmental
95 15
Nanowires/Cu NaHCO3 vs. RHE Science, 2019,12,
foam 1334-1340.
Journal of Materials
0.5 M -0.75 V
Bi2S3-derived Bi 84 5 Chemistry A, 2018,6,
NaHCO3 vs. RHE
4714-4720.

23
References

1. F. Li, L. Chen, M. Xue, T. Williams, Y. Zhang, D. R. MacFarlane and J. Zhang, Nano Energy,
2017, 31, 270-277.
2. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865.
3. J. Harl and G. Kresse, Phys. Rev. Lett., 2009, 103, 056401.
4. A. Klamt and G. Schüürmann, J. Chem. Soc., Perkin Trans. 2, 1993, 5, 799-805.
5. Y. Liu, J. Zhao and Q. Cai, Phys. Chem. Chem. Phys., 2016, 18, 5491-5498.
6. X. An, S. Li, A. Yoshida, Z. Wang, X. Hao, A. Abudula and G. Guan, ACS Sustain. Chem.
Eng., 2019, 7, 9360-9368.
7. H. J. Monkhorst and J. D. Pack, Phys. Rev. B, 1976, 13, 5188-5192.
8. J. K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J. R. Kitchin, T. Bligaard and H.
Jonsson, J. Phys. Chem. B, 2004, 108, 17886-17892.
9. A. A. Peterson, F. Abild-Pedersen, F. Studt, J. Rossmeisl and J. K. Nørskov, Energy
Environ. Sci., 2010, 3, 1311-1315.
10. D.-H. Lim and J. Wilcox, J. Phys. Chem. C, 2012, 116, 3653-3660.
11. S. Kattel, P. Atanassov and B. Kiefer, J. Phys. Chem. C, 2012, 116, 17378-17383.
12. Y. Zhang, X. Zhang, Y. Ling, F. Li, A. M. Bond and J. Zhang, Angew. Chem. Int. Ed., 2018,
57, 13283-13287.
13. F. Zhang and A. C. Co, Angew. Chem. Int. Ed., 2020, 59, 1674-1681.

24