Cu-Sn Bimetallic Catalyst for Selective Aqueous Electroreduction of ...
Supporting Information
Cu-Sn Bimetallic Catalyst for Selective Aqueous Electroreduction of CO2 to CO Saad Sarfraz, Angel T. Garcia-Esparza, Abdesslem Jedidi, Luigi Cavallo, and Kazuhiro Takanabe*
King Abdullah University of Science and Technology (KAUST), KAUST Catalysis Center (KCC), and Physical Sciences and Engineering Division (PSE), Thuwal, 23955-6900, Saudi Arabia.
[email protected]
S1
Figure S1. CO2 reduction on Sn sheet (0.1 M KHCO3, pH 6.8, saturated CO2). Sn was initially pretreated in 1 M HCl to remove the surface SnOx layer. The circles represent steady-state current densities.
Figure S2. CO2 reduction on Sn deposited on polycrystalline Cu (0.1 M KHCO3, pH 6.8, saturated CO2). Cu was pretreated in nitric acid to remove surface impurities, and a charge of 0.5 C cm−2 was applied to deposit Sn on Cu. The circles represent steady-state current densities.
S2
Figure S3. Effects of Sn electrodeposition charge on CO2 reduction performance after 1 h of CO2 electrolysis at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2). Sn was electrodeposited on OD-Cu without pre-reduction; i.e., Sn electrodeposition occurred concurrent OD-Cu reduction. The circles represent steady-state current densities.
Figure S4. Double layer capacitance of (a) OD-Cu and (b) Cu-Sn (6.3 C cm−2) in 0.1 M KHCO3 solution saturated with CO2 measured in the potential range between +0.36 to +0.46 V vs. RHE.
S3
Figure S5. XRD pattern of electrodeposited compounds after 1 h of CO2 electrolysis at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2). Sn was electrodeposited on OD-Cu without pre-reduction; i.e., Sn electrodeposition occurred concurrent OD-Cu reduction.
Figure S6. XRD pattern of Cu-Sn before and after 14 h of CO2 electrolysis at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2). S4
Figure S7. XPS spectra of (a) Cu 2p (b) O 1s and (c) Sn 4d including valence bands for different samples. Figure S7 shows XPS spectra for Cu 2p, O 1s, and Sn 4d and UV photoelectron spectra (valence electron) and the values are compared in Table S1. From Cu 2p, The reference Cu2O standard was pretreated in acid to remove any possible Cu(II) species on the oxidized surface; the deconvolution of the Cu 2p3/2 resulted in one peak at 933.0 eV and another minor contribution with peak maxima at 935.2 eV. The high-resolution O 1s spectra was resolved, exhibiting a peak at 530.7 eV, which can be assigned to oxygen from Cu2O in reasonable agreement with literature.1 CuO reference sample shows Cu(II) present as CuO or Cu(OH)2 exhibiting the prominent shake-off satellite peaks from 941 to 945 eV from the partially filled Cu 3d9 shell.2,3 The CuO standard Cu 2p3/2 spectra were resolved into two peaks at 933.7 and 935.2 eV, the former peak was assigned to Cu(II). The minor peak observed at 935.2 eV on both the Cu2O and CuO reference samples can be assigned to Cu(OH)2.4 The O 1s spectra for CuO indicated the characteristic oxide peak at 529.9 eV, which has a significant chemical shift compared with Cu2O.1 The peaks in the O 1s region observed at larger binding energies than 532 eV may be attributed to hydroxide oxygen or chemisorbed hydroxyl oxygen.4,5 The metallic Cu sheets were treated with 2 M HNO3, submerged in Milli-Q water, dried under an intense N2 stream and quickly introduced to the vacuum chamber of the XPS apparatus. The acid-treated sheet resulted in one Cu 2p3/2 peak with maxima at 933.1 eV.3 All the OD-Cu and Cu-Sn samples, the O 1s spectra shows the presence of surface oxygen species because of unavoidable exposure to the air in our experiments. Sn 4d spectra also show the mixed valence of the oxidized Sn, consistent with the Sn 3d spectra shown in the main manuscript.
S5
Table S1. XPS peak values for different samples.
eV
Cu 2p
Cu L3M45M45
O 1s
Sn 3d
Sn 4d
Cu
933.1
916.0-916.4
532
-
916.0-916.4
532
-
-
917.6-917.8
529.9
-
-
918.2-918.6 Cu2O
933.0 935.2
CuO
933.7 935.2 940-946
OD-Cu before
OD-Cu after
917.7
-
933.7
529.9
940-946
532.7
933.7
Cu-Sn before 933.1
916.4
530.7
918.5
532.7
916.4 918.4
-
-
-
530.7
484.9
26.6
532.7
486.7 495.3
Cu-Sn after
933.1
916.4
530.7
486.7
918.3
532.7
495.3
S6
26.6
Figure S8. Chronoamperometry profile of Sn electrodeposition on OD-Cu (1.8 cm2) in a twoelectrode cell with Sn sheet as the anode (3.0 cm2) at −0.5 V vs. Sn. The electrolyte consisted of a solution of 0.05 M SnCl2 in 2 M KOH (pH 14.4).
Figure S9. (a) Electroreduction profile of OD-Cu (cathode) in a KOH solution without SnCl2 at 6.3 C cm−2 with Sn plate (anode) at −0.5 V vs. Sn, (b) CO2 reduction on OD-Cu reduced initially in KOH solution at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2).
S7
Figure S10. (a) CO2 reduction of Cu84Sn5 alloy at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2), (b) XRD pattern of alloy electrode. The CuxSn alloy electrode was prepared from commercial bronze powder (Cu84Sn5) by annealing in N2 at 1100 °C for 5 h and cooled to room temperature.
S8
Table S2. Faradaic efficiency (%) and current density comparison of OD-Cu prepared using different methods.6-8 Faradaic efficiency / % Electrolyte
CO
HCOOH
C2H4
C2H6
C2H5OH
Total current density / mA cm-2
Total Faradaic efficiency for CO2 reduction / %
Total Faradaic efficiency for H2 evolution / %
−0.4 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3
38
11
-
-
-
-1
49
51
0.1 M KHCO3
38
11
-
-
-
-0.30
49
51
-
-
-
-
-
-
-
-
-
0.1 M KHCO3
48
3
-
-
-
-0.67
51
49
−0.5 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3
40
33
-
-
-
-2.7
73
27
0.1 M KHCO3
42
12
-
-
-
-0.60
54
44
-
-
-
-
-
-
-
-
-
0.1 M KHCO3
61
13
-
-
-
-1.2
75
26
−0.6 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3 0.1 M KHCO3 0.1 M KHCO3 0.1 M KHCO3
28
33
1
3
-
-3.1
65
31
39
34
-
-
-
-1.0
73
24
34
-
4
-
-
-
38
57
63
25
-
-
-
-2.3
88
11
S9
Faradaic efficiency / % Electrolyte
CO
HCOOH
C2H4
C2H6
C2H5OH
Total current density / mA cm-2
Total Faradaic efficiency for CO2 reduction / %
Total Faradaic efficiency for H2 evolution / %
−0.7 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3 0.1 M KHCO3 0.1 M KHCO3 0.1 M KHCO3
16
32
3
5
-
−6.0
56
40
24
35
-
-
-
−1.6
59
36
7
16
10
-
-
-
33
63
46
41
-
-
-
−3.3
87
11
−0.8 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3
8
23
4
6
-
−10.3
41
54
-
-
-
-
-
-
-
-
-
4
9
19
-
8
-
40
50
32
45
-
-
-
−4.8
77
20
0.1 M KHCO3 0.1 M KHCO3
S10
Electronic structure of Cu-Sn alloys:
Figure S11. The crystal structures of different compositions of the Cu-Sn alloy. From left to right: SQS-Cu24Sn8, Cu3Sn and Cu6Sn5. In this part, we summarize the study of the electronic structures of different compositions of intermetallic compounds (Cu-Sn), as discussed above and shown in Figure S10. The density of states (DOS) of the three examined Cu/Sn alloys is quite similar to that of pure Cu, although the peak heights are different owing to the various compositions (Figure S11). We will focus on the case of Cu6Sn5 as a representative example because there is no change among the three compositions of the nano-alloys. For pure Cu, the DOS (given in states eV−1) at the Fermi level is relatively low but finite in all cases, which indicates a metallic behavior. It increases slightly with Sn content for the Cu–Sn compounds, increasing from 5 states/eV on the pure Cu to approximately 22 states eV−1 for Cu6Sn5 (Figure S12). The most noticeable bonding band is completely occupied and extends from −5.0 to −2.0 eV approximately below the Fermi level. It is mostly defined by the contribution of Cu-d states, with additional minor contributions of Sn s and p states for the Cu–Sn compound. For the PDOS of Sn, the s states are dominant for energies below −7 eV, and the p states are dominant for higher energies up to 4 eV. We also observe that contributions from Sn p states to the PDOS at the Fermi level are more present than those of p and d states of Cu atoms. Figure S12 shows the projection of the d-orbital of Cu on Cu6Sn5 compared to pure Cu. It is clear that in the alloy, the edge of the band corresponding to the Cu-d states is shifted to lower energy, and it is approximately 0.5 eV narrower than that of pure Cu. This result is due to the reduction in the number of Cu-Cu bonds. However, the center of the d-band, which is usually related to catalytic activity, is unaffected because this band in Cu is completely filled and well below the Fermi level.9,10 Figure S12 shows that at low energy (−10 to −6 eV), there is a contribution of Sn-s states, which can be explained by the presence of free electrons. However, Sn-p states increase with increasing energy. This analysis allows us to conclude that the presence of Sn in the compound does not dramatically affect the electronic structure of pure Cu; thus, the different experimental behavior of Cu/Sn alloys from pure Cu is possibly a result of a perturbation of the surface properties, either in terms of the geometry of the active site or in terms of the electronic properties of the active site. S11
Cu6Sn5
Cu3Sn
Cu75Sn25
Figure S12. Total and partial electronic density of states (DOS) for the studied Cu/Sn alloys (Cu6Sn5, Cu3Sn and SQS-Cu75Sn25).
S12
Pure Cu
Cu6Sn5
Figure S13. Comparison of the total and partial DOS between pure Cu and Cu6Sn5 alloy
S13
a ΔE = +0.56 eV
b ΔE = −0.05 eV
c ΔE = −0.06 eV
Figure S14. Optimized geometries and relative energies of H adsorbed on the Cu(111) facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the H atom referenced to 1/2 H2.
a ΔE = +0.59 eV
b ΔE =+0.76 eV
c ΔE = +0.53 eV
Figure S15. Optimized geometry and relative energy of H adsorbed on the Sn modified Cu(111) facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the H atom referenced to 1/2 H2.
a ΔE = +1.24 eV
b ΔE = +1.25 eV
c ΔE = +1.18 eV
Figure S16. Optimized geometries and relative energy of CO adsorbed on the (111) Cu facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the C atom referenced to graphene and the O atom to (H2O-H2).
a ΔE = +1.70 eV
b ΔE = +1.50 eV
c ΔE = +1.30 eV
Figure S17. Optimized geometries and relative energies of H adsorbed on the Sn modified (111) Cu facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the C atom referenced to graphene and the O atom to (H2O-H2).
S14
References 1. McIntyre, N. S.; Sunder, S.; Shoesmith, D. W.; Stanchell, F. W. J. Vac. Sci. Technol. 1981, 18, 714-721. 2.
Larson, P. E. J. Electron. Spectrosc. 1974, 4, 213-218.
3.
Haber, J.; Machej, T.; Ungier, L.; Ziółkowski, J. J. Solid State Chem. 1978, 25, 207-218.
4. 820.
Poulston, S.; Parlett, P. M.; Stone, P.; Bowker, M. Surf. Interface Anal. 1996, 24, 811-
5. Chawla, S. K.; Sankarraman, N.; Payer, J. H. J. Electron Spectrosc. Relat. Phenom. 1992, 61, 1-18. 6.
Li, C. W.; Kanan, M. W. J. Am. Chem. Soc. 2012, 134, 7231-7234.
7. Rasul, S.; Anjum, D. H.; Jedidi, A.; Minenkov, Y.; Cavallo, L.; Takanabe, K. Angew. Chem. Int. Ed. 2015, 54, 2146-2150. 8. Ren, D.; Deng, Y.; Handoko, A. D.; Chen, C. S.; Malkhandi, S.; Yeo, B. S. ACS Catal. 2015, 5, 2814-2821. 9.
Hammer, B.; Nørskov, J. K. Surf. Sci. 1995, 343, 211-220.
10. Hammer, B.; Nørskov, J. K. Theoretical surface science and catalysis—calculations and concepts. In Advances in Catalysis, Bruce C. Gates, H. K., Ed. Academic Press: 2000; Vol. 45, pp 71-129.
S15
Cu-Sn Bimetallic Catalyst for Selective Aqueous Electroreduction of CO2 to CO Saad Sarfraz, Angel T. Garcia-Esparza, Abdesslem Jedidi, Luigi Cavallo, and Kazuhiro Takanabe*
King Abdullah University of Science and Technology (KAUST), KAUST Catalysis Center (KCC), and Physical Sciences and Engineering Division (PSE), Thuwal, 23955-6900, Saudi Arabia.
[email protected]
S1
Figure S1. CO2 reduction on Sn sheet (0.1 M KHCO3, pH 6.8, saturated CO2). Sn was initially pretreated in 1 M HCl to remove the surface SnOx layer. The circles represent steady-state current densities.
Figure S2. CO2 reduction on Sn deposited on polycrystalline Cu (0.1 M KHCO3, pH 6.8, saturated CO2). Cu was pretreated in nitric acid to remove surface impurities, and a charge of 0.5 C cm−2 was applied to deposit Sn on Cu. The circles represent steady-state current densities.
S2
Figure S3. Effects of Sn electrodeposition charge on CO2 reduction performance after 1 h of CO2 electrolysis at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2). Sn was electrodeposited on OD-Cu without pre-reduction; i.e., Sn electrodeposition occurred concurrent OD-Cu reduction. The circles represent steady-state current densities.
Figure S4. Double layer capacitance of (a) OD-Cu and (b) Cu-Sn (6.3 C cm−2) in 0.1 M KHCO3 solution saturated with CO2 measured in the potential range between +0.36 to +0.46 V vs. RHE.
S3
Figure S5. XRD pattern of electrodeposited compounds after 1 h of CO2 electrolysis at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2). Sn was electrodeposited on OD-Cu without pre-reduction; i.e., Sn electrodeposition occurred concurrent OD-Cu reduction.
Figure S6. XRD pattern of Cu-Sn before and after 14 h of CO2 electrolysis at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2). S4
Figure S7. XPS spectra of (a) Cu 2p (b) O 1s and (c) Sn 4d including valence bands for different samples. Figure S7 shows XPS spectra for Cu 2p, O 1s, and Sn 4d and UV photoelectron spectra (valence electron) and the values are compared in Table S1. From Cu 2p, The reference Cu2O standard was pretreated in acid to remove any possible Cu(II) species on the oxidized surface; the deconvolution of the Cu 2p3/2 resulted in one peak at 933.0 eV and another minor contribution with peak maxima at 935.2 eV. The high-resolution O 1s spectra was resolved, exhibiting a peak at 530.7 eV, which can be assigned to oxygen from Cu2O in reasonable agreement with literature.1 CuO reference sample shows Cu(II) present as CuO or Cu(OH)2 exhibiting the prominent shake-off satellite peaks from 941 to 945 eV from the partially filled Cu 3d9 shell.2,3 The CuO standard Cu 2p3/2 spectra were resolved into two peaks at 933.7 and 935.2 eV, the former peak was assigned to Cu(II). The minor peak observed at 935.2 eV on both the Cu2O and CuO reference samples can be assigned to Cu(OH)2.4 The O 1s spectra for CuO indicated the characteristic oxide peak at 529.9 eV, which has a significant chemical shift compared with Cu2O.1 The peaks in the O 1s region observed at larger binding energies than 532 eV may be attributed to hydroxide oxygen or chemisorbed hydroxyl oxygen.4,5 The metallic Cu sheets were treated with 2 M HNO3, submerged in Milli-Q water, dried under an intense N2 stream and quickly introduced to the vacuum chamber of the XPS apparatus. The acid-treated sheet resulted in one Cu 2p3/2 peak with maxima at 933.1 eV.3 All the OD-Cu and Cu-Sn samples, the O 1s spectra shows the presence of surface oxygen species because of unavoidable exposure to the air in our experiments. Sn 4d spectra also show the mixed valence of the oxidized Sn, consistent with the Sn 3d spectra shown in the main manuscript.
S5
Table S1. XPS peak values for different samples.
eV
Cu 2p
Cu L3M45M45
O 1s
Sn 3d
Sn 4d
Cu
933.1
916.0-916.4
532
-
916.0-916.4
532
-
-
917.6-917.8
529.9
-
-
918.2-918.6 Cu2O
933.0 935.2
CuO
933.7 935.2 940-946
OD-Cu before
OD-Cu after
917.7
-
933.7
529.9
940-946
532.7
933.7
Cu-Sn before 933.1
916.4
530.7
918.5
532.7
916.4 918.4
-
-
-
530.7
484.9
26.6
532.7
486.7 495.3
Cu-Sn after
933.1
916.4
530.7
486.7
918.3
532.7
495.3
S6
26.6
Figure S8. Chronoamperometry profile of Sn electrodeposition on OD-Cu (1.8 cm2) in a twoelectrode cell with Sn sheet as the anode (3.0 cm2) at −0.5 V vs. Sn. The electrolyte consisted of a solution of 0.05 M SnCl2 in 2 M KOH (pH 14.4).
Figure S9. (a) Electroreduction profile of OD-Cu (cathode) in a KOH solution without SnCl2 at 6.3 C cm−2 with Sn plate (anode) at −0.5 V vs. Sn, (b) CO2 reduction on OD-Cu reduced initially in KOH solution at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2).
S7
Figure S10. (a) CO2 reduction of Cu84Sn5 alloy at −0.6 V vs. RHE (0.1 M KHCO3, pH 6.8, saturated CO2), (b) XRD pattern of alloy electrode. The CuxSn alloy electrode was prepared from commercial bronze powder (Cu84Sn5) by annealing in N2 at 1100 °C for 5 h and cooled to room temperature.
S8
Table S2. Faradaic efficiency (%) and current density comparison of OD-Cu prepared using different methods.6-8 Faradaic efficiency / % Electrolyte
CO
HCOOH
C2H4
C2H6
C2H5OH
Total current density / mA cm-2
Total Faradaic efficiency for CO2 reduction / %
Total Faradaic efficiency for H2 evolution / %
−0.4 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3
38
11
-
-
-
-1
49
51
0.1 M KHCO3
38
11
-
-
-
-0.30
49
51
-
-
-
-
-
-
-
-
-
0.1 M KHCO3
48
3
-
-
-
-0.67
51
49
−0.5 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3
40
33
-
-
-
-2.7
73
27
0.1 M KHCO3
42
12
-
-
-
-0.60
54
44
-
-
-
-
-
-
-
-
-
0.1 M KHCO3
61
13
-
-
-
-1.2
75
26
−0.6 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3 0.1 M KHCO3 0.1 M KHCO3 0.1 M KHCO3
28
33
1
3
-
-3.1
65
31
39
34
-
-
-
-1.0
73
24
34
-
4
-
-
-
38
57
63
25
-
-
-
-2.3
88
11
S9
Faradaic efficiency / % Electrolyte
CO
HCOOH
C2H4
C2H6
C2H5OH
Total current density / mA cm-2
Total Faradaic efficiency for CO2 reduction / %
Total Faradaic efficiency for H2 evolution / %
−0.7 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3 0.1 M KHCO3 0.1 M KHCO3 0.1 M KHCO3
16
32
3
5
-
−6.0
56
40
24
35
-
-
-
−1.6
59
36
7
16
10
-
-
-
33
63
46
41
-
-
-
−3.3
87
11
−0.8 V vs. RHE OD-Cu (500 °C 12 h, electropolished in 85% phosphoric acid)6 OD-Cu (500 °C, 2 h, cleaned with 1 M HCl)7 8.8 µm Cu2O film deposited on Cu disc8 OD-Cu (this paper)
0.5 M NaHCO3
8
23
4
6
-
−10.3
41
54
-
-
-
-
-
-
-
-
-
4
9
19
-
8
-
40
50
32
45
-
-
-
−4.8
77
20
0.1 M KHCO3 0.1 M KHCO3
S10
Electronic structure of Cu-Sn alloys:
Figure S11. The crystal structures of different compositions of the Cu-Sn alloy. From left to right: SQS-Cu24Sn8, Cu3Sn and Cu6Sn5. In this part, we summarize the study of the electronic structures of different compositions of intermetallic compounds (Cu-Sn), as discussed above and shown in Figure S10. The density of states (DOS) of the three examined Cu/Sn alloys is quite similar to that of pure Cu, although the peak heights are different owing to the various compositions (Figure S11). We will focus on the case of Cu6Sn5 as a representative example because there is no change among the three compositions of the nano-alloys. For pure Cu, the DOS (given in states eV−1) at the Fermi level is relatively low but finite in all cases, which indicates a metallic behavior. It increases slightly with Sn content for the Cu–Sn compounds, increasing from 5 states/eV on the pure Cu to approximately 22 states eV−1 for Cu6Sn5 (Figure S12). The most noticeable bonding band is completely occupied and extends from −5.0 to −2.0 eV approximately below the Fermi level. It is mostly defined by the contribution of Cu-d states, with additional minor contributions of Sn s and p states for the Cu–Sn compound. For the PDOS of Sn, the s states are dominant for energies below −7 eV, and the p states are dominant for higher energies up to 4 eV. We also observe that contributions from Sn p states to the PDOS at the Fermi level are more present than those of p and d states of Cu atoms. Figure S12 shows the projection of the d-orbital of Cu on Cu6Sn5 compared to pure Cu. It is clear that in the alloy, the edge of the band corresponding to the Cu-d states is shifted to lower energy, and it is approximately 0.5 eV narrower than that of pure Cu. This result is due to the reduction in the number of Cu-Cu bonds. However, the center of the d-band, which is usually related to catalytic activity, is unaffected because this band in Cu is completely filled and well below the Fermi level.9,10 Figure S12 shows that at low energy (−10 to −6 eV), there is a contribution of Sn-s states, which can be explained by the presence of free electrons. However, Sn-p states increase with increasing energy. This analysis allows us to conclude that the presence of Sn in the compound does not dramatically affect the electronic structure of pure Cu; thus, the different experimental behavior of Cu/Sn alloys from pure Cu is possibly a result of a perturbation of the surface properties, either in terms of the geometry of the active site or in terms of the electronic properties of the active site. S11
Cu6Sn5
Cu3Sn
Cu75Sn25
Figure S12. Total and partial electronic density of states (DOS) for the studied Cu/Sn alloys (Cu6Sn5, Cu3Sn and SQS-Cu75Sn25).
S12
Pure Cu
Cu6Sn5
Figure S13. Comparison of the total and partial DOS between pure Cu and Cu6Sn5 alloy
S13
a ΔE = +0.56 eV
b ΔE = −0.05 eV
c ΔE = −0.06 eV
Figure S14. Optimized geometries and relative energies of H adsorbed on the Cu(111) facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the H atom referenced to 1/2 H2.
a ΔE = +0.59 eV
b ΔE =+0.76 eV
c ΔE = +0.53 eV
Figure S15. Optimized geometry and relative energy of H adsorbed on the Sn modified Cu(111) facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the H atom referenced to 1/2 H2.
a ΔE = +1.24 eV
b ΔE = +1.25 eV
c ΔE = +1.18 eV
Figure S16. Optimized geometries and relative energy of CO adsorbed on the (111) Cu facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the C atom referenced to graphene and the O atom to (H2O-H2).
a ΔE = +1.70 eV
b ΔE = +1.50 eV
c ΔE = +1.30 eV
Figure S17. Optimized geometries and relative energies of H adsorbed on the Sn modified (111) Cu facet. ΔE is the electronic energy of the state minus the electronic energy of the clean slab associated with that state, with the C atom referenced to graphene and the O atom to (H2O-H2).
S14
References 1. McIntyre, N. S.; Sunder, S.; Shoesmith, D. W.; Stanchell, F. W. J. Vac. Sci. Technol. 1981, 18, 714-721. 2.
Larson, P. E. J. Electron. Spectrosc. 1974, 4, 213-218.
3.
Haber, J.; Machej, T.; Ungier, L.; Ziółkowski, J. J. Solid State Chem. 1978, 25, 207-218.
4. 820.
Poulston, S.; Parlett, P. M.; Stone, P.; Bowker, M. Surf. Interface Anal. 1996, 24, 811-
5. Chawla, S. K.; Sankarraman, N.; Payer, J. H. J. Electron Spectrosc. Relat. Phenom. 1992, 61, 1-18. 6.
Li, C. W.; Kanan, M. W. J. Am. Chem. Soc. 2012, 134, 7231-7234.
7. Rasul, S.; Anjum, D. H.; Jedidi, A.; Minenkov, Y.; Cavallo, L.; Takanabe, K. Angew. Chem. Int. Ed. 2015, 54, 2146-2150. 8. Ren, D.; Deng, Y.; Handoko, A. D.; Chen, C. S.; Malkhandi, S.; Yeo, B. S. ACS Catal. 2015, 5, 2814-2821. 9.
Hammer, B.; Nørskov, J. K. Surf. Sci. 1995, 343, 211-220.
10. Hammer, B.; Nørskov, J. K. Theoretical surface science and catalysis—calculations and concepts. In Advances in Catalysis, Bruce C. Gates, H. K., Ed. Academic Press: 2000; Vol. 45, pp 71-129.
S15
