In-situ Study of Oxygen Reduction in DMSO Solution
Supplementary Information:
In-situ Study of Oxygen Reduction in DMSO Solution: A Fundamental Study for Development of Lithium-Oxygen Battery Qiao Yu and Shen Ye* Catalysis Research Center, Hokkaido University, Sapporo 001-0021, Japan *[email protected]
Figure S1. AFM images (2µm×2µm) for the gold thin-films on mica substrate surface prepared by (a) sputtering method and (b) vacuum evaporation method, respectively. The sputtered sample surface shows many small gold islands with domain size of approximately 50 nm. In contrast, the vacuum evaporated sample exhibits relatively wide and flat domains on the surface. The difference in the surface morphology is considered as the main reason for totally different Raman sensitives observed on the two substrates. Especially, the sputtered gold surface shows a strong surface-enhanced Raman scattering (SERS) effect. See text for details.
Table S1 Charge (Q, mC/cm2) and amount of the superoxide (µmol/cm2) during potential sweep at different scan rates in O2-saturated 0.1 M TBA-DMSO. As the scan rate is increased, less superoxide is generated with higher overpotential. The higher Coulombic efficiency is also obtained with lower scan rate. 200 mV/s
10 mV/s
2mV/s
ORR
2.49
2.59
2.63
OER
2.90
2.80
2.76
ORR
3.13
15.08
22.14
OER
1.74
6.12
8.02
0.032
0.1565
0.2298
Oxidized superoxide (OER) (µmol /cm )
0.0181
0.0635
0.0833
Remained superoxide after a potential cycle (µmol/cm2)
0.0143
0.093
0.1465
Coulombic efficiency (OER/ORR)
55.68 %
40.58 %
36.26 %
Peak position (V) Accumulated quantity (mC/cm2)
2
Generated superoxide (ORR) (µmol /cm ) 2
Figure S2 in-situ Raman spectra observed on the sputter gold and vacuum evaporate gold electrodes after potential is swept to 2.0V in (a) O2-saturated 0.1 M TBAClO4-DMSO and (b) O2-saturated 0.1 M LiClO4-DMSO. Potential dependence of currents and Raman peaks for (c) adsorbed superoxide in 0.1 M TBAClO4-DMSO and (d) Li2O2 in 0.1 M LiClO4-DMSO observed on two electrodes are also given.
(a)
0.20
Au-O stretch in adsorbed O2
O-O stretch in adsorbed O2
0.1M TBAClO4-DMSO 10mV/s
Raman Intensity of -1 Au-O stretch peak near 490 cm vacuum evaporated sputter gold
0.10 0.05
2
vacuum evaporated sputter gold
(c)
0.15
Current (mA/cm )
Raman Intensity (au.)
0.1M TBA-DMSO charge to 2.0V (10mV/s)
0.00 -0.05 -0.10 -0.15 -0.20 -0.25
400
600
800
2.0
1000
0.1M Li-DMSO charge to 2.0V (10mV/s)
3.5
4.0
0.2
Raman Intensity of (d) O-O stretch in Li O at 788 cm
-1
2
sputter gold vacuum evaporated
2
2
Current (mA/cm )
0.1
Raman Intensity (au.)
vacuum evaporated sputter gold
Raman Intensity (au.)
3.0
Potential (V vs. Li/Li )
Raman Shift (cm )
(b)
2.5
+
-1
O-O stretch in Li2O2
0.0
-0.1
-0.2
0.1M LiClO4-DMSO 10mV/s
-0.3
400
600
800
1000 -1
Raman Shift (cm )
2.0
2.5
3.0
3.5 +
Potential (V vs. Li/Li )
4.0
Figure S3 (a) Raman spectra of potassium superoxide (KO2) powder on a glass substrate (black trace), 5mM KO2-dissloved DMSO on a sputtered Au substrate in the Raman cell (red trace). An in-situ Raman spectrum obtained on a sputtered Au electrode after the potential is swept to 2.0 V in an O2-saturated 0.1M TBA-DMSO (blue trace). (b) UV-vis spectra observed after potential is swept (10mV/s) to 2.0 V in an O2-saturated 0.1M TBAClO4-DMSO (black trace) and LiClO4-DMSO (red trace). The blue trace is a UV-Vis spectrum for KO2 dissolved DMSO. The adsorption state of the superoxide ((adsorbed O2-) on the Au electrode surface is different from that in bulk solution (ion-pair).
(b) KO2 powder on glass substrate(x5 times) 5mM KO2 dissolve into DMSO then inject into Raman cell on sputter gold substrate TBA-DMSO 10mV/s cycle to 2.0V (sputter gold electrode)
0.3
O-O stretch in KO2 (at 1141)
Au-O stretch in
-
O-O stretch in adsorbed O2
-
adsorbed O2
0.4
peak position 251.79 nm
(around 1102 )
Absorbance
Raman Intensity (au.)
(a)
TBA-DMSO 2.0V Li-DMSO 2.0V 20uL KO2 saturated DMSO drop into 2mL DMSO
0.2
0.1
x5 times 400
600
0.0 800
1000 -1
Raman Shift (cm )
1200
250
260
270
280
Wavelength (nm)
290
300
Figure S4 Potential dependence of Raman peaks (1104 cm-1: O-O stretch in adsorbed superoxide and 788 cm-1: O-O stretch in Li2O2) observed on a sputtered Au electrode surface in O2 saturated 0.1M TBAClO4-DMSO containing 2mM Li-ion during potential sweep (10mV/s).
0.2
2mM Li-ion in 100mM TBA-DMSO 10mV/s CV scan
2
0.0 -1
1104 cm (adsorbed O2 )
-0.1
-1
788 cm (Li2O2) -0.2
-0.3 2.0
2.5
3.0
3.5 +
Potential (V vs. Li/Li )
4.0
Raman Intensity (au.)
Current (mA/cm )
0.1
Table S2 Absorbance and amount of superoxide (µmol/cm2) after the potential is swept to 2.0V at scan rates of 10mV/s and 2mV/s in an O2 saturated 0.1 M TBAClO4-DMSO electrolyte containing different amounts of Li-ion (100mM and 2mM). The film thickness of Li2O2 calculated based on the results are given in the bottom column. 100 mM
2 mM
10 mV/s
2 mV/s
10 mV/s
2 mV/s
Absorbance (@252nm)
0.16
0.383
0.196
0.354
Amount of superoxide based on UV-vis (na, µmol/cm2)
0.098
0.235
0.12
0.217
amount of electron in CVs (nb, µmol/cm2)
0.135
0.272
0.141
0.254
0.0184
0.0186
0.0103
0.0185
3.67
3.71
2.06
3.68
2
amount of Li2O2 ((nb-na)/2, µmol/cm ) Li2O2 thickness (nm)
Figure S5 in-situ Raman spectra recorded after discharge to 2.0V in TBA-DMSO solution (black line). Then the electrolyte solution in the cell is totally replaced by fresh Li-DMSO solution at OCP (red line) and hold for 2 min (blue line). The adsorbed superoxide does not change with the increasing Li-ion concentration of LiO2 in solution under the open circuit conditions. This confirms again that no disproportionation reaction occurs.
Raman Intensity (au.)
0.1M TBA-DMSO after discharge to 2.0V change Li-contained DMSO electrolyte Au-O stretch of adsorbed O2
O-O stretch of adsorbed O2 Li2O2 O-O stretch peak position then stay at OCP for 120s change Li-contained electrolyte discharge to 2.0V
400
500
600
700
800
900 -1
Raman Shift (cm )
1000
1100
Figure S6 Deconvoluted current yield for Li2O2 based on the electrochemistry, in-situ UV-vis and SERS measurement during CV sweep: (a) different scan rate in 100mM Li-DMSO solution, (b) different Li-ion concentration in 100mM TBAClO4-DMSO electrolyte.
(a)
(b) 0.10
0.05
(different scan rate)
0.1M TBA-DMSO 10mV/s scan rate current yield for Li2O2 (different Li-ion concentration)
2
Current (mA/cm )
2
Current (mA/cm )
0.05
0.10 0.1M Li-DMSO current yield for Li2O2
0.00
-0.05
10 mV/s 2 mV/s
-0.10
0.00 0 mM 1 mM 2 mM 5 mM 10mM 100mM
-0.05
-0.10
2.0
2.5
3.0
3.5 +
Potential (V vs. Li/Li )
4.0
2.0
2.5
3.0
3.5
4.0 +
Potential (V vs. Li/Li )
Figure S7 (a) UV-vis spectra in KO2 dissolved in DMSO: saturated solution (blue trace) and 5-time diluent solution (red trace). (b) in-situ UV-vis spectra after the potential is swept to 2.0V (10 mV/s) in O2 saturated 0.1M TBAClO4-DMSO and 0.1M LiClO4-DMSO. Temporal dependence for UV-Vis absorbance: (c) at 310 nm (in KO2 saturated DMSO) and 252 nm peak position (5-times diluent KO2-DMSO) after 0.1 M Li-ion is added. (d) 252 nm after the potential is swept to 2.0V (black trace: no Li-ion added into TBA-DMSO, blue trace: 0.1M Li-ion added into TBA-DMSO). Therefore, the possibility for chemical disproportionation reaction of LiO2 ican be almost ignored.
(a)
(c) KO2 saturated DMSO 500uL KO2 saturated DMSO added into 2.5mL DMSO solvent (KO2@DMSO : DMSO=1:5)
7 6
Absorbance
Absorbance
6
8
4
2
after added Li-DMSO to 100mM KO2 saturated DMSO (310 nm) KO2 saturated : DMSO (1:5) (252 nm peak position)
5 4 3 2 1
0
KO2-saturated DMSO (310nm) KO2 saturated DMSO : DMSO (1:5)
0 min 2.3419 5.52
15 min 2.3072 5.258
retention ratio 98.52% 95.25%
0 200
250
300
350
400
450
0
500
3
Wavelength (nm)
(b)
9
12
15
0.25 Absorbacne in DMSO electrolyte after 10 mV/s discharge to 2.0 V 0.20
TBA Li
Absorbance
0.15
Absorbance
(d)
10 mV/s discharge to 2.0 V in DMSO electrolyte
0.20
6
Time (min)
0.10
0.15
TBA Li TBA added Li-ion
0.10
0.05
0.05 0.00
TBA Li TBA added Li-ion
0 min 0.171 0.159 0.171
15 min 0.172 0.1568 0.1675
retention ratio 100.6% 98.37% 97.95%
0.00 250
260
270
280
Wavelength (nm)
290
300
0
3
6
9
12
15
Time (min)
Preparation for KO2 saturated DMSO solution: the excess amount of KO2 powder was added into DMSO solution and then ultrasound for 30min. After stirring for another 2h, the turbid liquid is further centrifuge and the clarify liquid is taken out.
In-situ Study of Oxygen Reduction in DMSO Solution: A Fundamental Study for Development of Lithium-Oxygen Battery Qiao Yu and Shen Ye* Catalysis Research Center, Hokkaido University, Sapporo 001-0021, Japan *[email protected]
Figure S1. AFM images (2µm×2µm) for the gold thin-films on mica substrate surface prepared by (a) sputtering method and (b) vacuum evaporation method, respectively. The sputtered sample surface shows many small gold islands with domain size of approximately 50 nm. In contrast, the vacuum evaporated sample exhibits relatively wide and flat domains on the surface. The difference in the surface morphology is considered as the main reason for totally different Raman sensitives observed on the two substrates. Especially, the sputtered gold surface shows a strong surface-enhanced Raman scattering (SERS) effect. See text for details.
Table S1 Charge (Q, mC/cm2) and amount of the superoxide (µmol/cm2) during potential sweep at different scan rates in O2-saturated 0.1 M TBA-DMSO. As the scan rate is increased, less superoxide is generated with higher overpotential. The higher Coulombic efficiency is also obtained with lower scan rate. 200 mV/s
10 mV/s
2mV/s
ORR
2.49
2.59
2.63
OER
2.90
2.80
2.76
ORR
3.13
15.08
22.14
OER
1.74
6.12
8.02
0.032
0.1565
0.2298
Oxidized superoxide (OER) (µmol /cm )
0.0181
0.0635
0.0833
Remained superoxide after a potential cycle (µmol/cm2)
0.0143
0.093
0.1465
Coulombic efficiency (OER/ORR)
55.68 %
40.58 %
36.26 %
Peak position (V) Accumulated quantity (mC/cm2)
2
Generated superoxide (ORR) (µmol /cm ) 2
Figure S2 in-situ Raman spectra observed on the sputter gold and vacuum evaporate gold electrodes after potential is swept to 2.0V in (a) O2-saturated 0.1 M TBAClO4-DMSO and (b) O2-saturated 0.1 M LiClO4-DMSO. Potential dependence of currents and Raman peaks for (c) adsorbed superoxide in 0.1 M TBAClO4-DMSO and (d) Li2O2 in 0.1 M LiClO4-DMSO observed on two electrodes are also given.
(a)
0.20
Au-O stretch in adsorbed O2
O-O stretch in adsorbed O2
0.1M TBAClO4-DMSO 10mV/s
Raman Intensity of -1 Au-O stretch peak near 490 cm vacuum evaporated sputter gold
0.10 0.05
2
vacuum evaporated sputter gold
(c)
0.15
Current (mA/cm )
Raman Intensity (au.)
0.1M TBA-DMSO charge to 2.0V (10mV/s)
0.00 -0.05 -0.10 -0.15 -0.20 -0.25
400
600
800
2.0
1000
0.1M Li-DMSO charge to 2.0V (10mV/s)
3.5
4.0
0.2
Raman Intensity of (d) O-O stretch in Li O at 788 cm
-1
2
sputter gold vacuum evaporated
2
2
Current (mA/cm )
0.1
Raman Intensity (au.)
vacuum evaporated sputter gold
Raman Intensity (au.)
3.0
Potential (V vs. Li/Li )
Raman Shift (cm )
(b)
2.5
+
-1
O-O stretch in Li2O2
0.0
-0.1
-0.2
0.1M LiClO4-DMSO 10mV/s
-0.3
400
600
800
1000 -1
Raman Shift (cm )
2.0
2.5
3.0
3.5 +
Potential (V vs. Li/Li )
4.0
Figure S3 (a) Raman spectra of potassium superoxide (KO2) powder on a glass substrate (black trace), 5mM KO2-dissloved DMSO on a sputtered Au substrate in the Raman cell (red trace). An in-situ Raman spectrum obtained on a sputtered Au electrode after the potential is swept to 2.0 V in an O2-saturated 0.1M TBA-DMSO (blue trace). (b) UV-vis spectra observed after potential is swept (10mV/s) to 2.0 V in an O2-saturated 0.1M TBAClO4-DMSO (black trace) and LiClO4-DMSO (red trace). The blue trace is a UV-Vis spectrum for KO2 dissolved DMSO. The adsorption state of the superoxide ((adsorbed O2-) on the Au electrode surface is different from that in bulk solution (ion-pair).
(b) KO2 powder on glass substrate(x5 times) 5mM KO2 dissolve into DMSO then inject into Raman cell on sputter gold substrate TBA-DMSO 10mV/s cycle to 2.0V (sputter gold electrode)
0.3
O-O stretch in KO2 (at 1141)
Au-O stretch in
-
O-O stretch in adsorbed O2
-
adsorbed O2
0.4
peak position 251.79 nm
(around 1102 )
Absorbance
Raman Intensity (au.)
(a)
TBA-DMSO 2.0V Li-DMSO 2.0V 20uL KO2 saturated DMSO drop into 2mL DMSO
0.2
0.1
x5 times 400
600
0.0 800
1000 -1
Raman Shift (cm )
1200
250
260
270
280
Wavelength (nm)
290
300
Figure S4 Potential dependence of Raman peaks (1104 cm-1: O-O stretch in adsorbed superoxide and 788 cm-1: O-O stretch in Li2O2) observed on a sputtered Au electrode surface in O2 saturated 0.1M TBAClO4-DMSO containing 2mM Li-ion during potential sweep (10mV/s).
0.2
2mM Li-ion in 100mM TBA-DMSO 10mV/s CV scan
2
0.0 -1
1104 cm (adsorbed O2 )
-0.1
-1
788 cm (Li2O2) -0.2
-0.3 2.0
2.5
3.0
3.5 +
Potential (V vs. Li/Li )
4.0
Raman Intensity (au.)
Current (mA/cm )
0.1
Table S2 Absorbance and amount of superoxide (µmol/cm2) after the potential is swept to 2.0V at scan rates of 10mV/s and 2mV/s in an O2 saturated 0.1 M TBAClO4-DMSO electrolyte containing different amounts of Li-ion (100mM and 2mM). The film thickness of Li2O2 calculated based on the results are given in the bottom column. 100 mM
2 mM
10 mV/s
2 mV/s
10 mV/s
2 mV/s
Absorbance (@252nm)
0.16
0.383
0.196
0.354
Amount of superoxide based on UV-vis (na, µmol/cm2)
0.098
0.235
0.12
0.217
amount of electron in CVs (nb, µmol/cm2)
0.135
0.272
0.141
0.254
0.0184
0.0186
0.0103
0.0185
3.67
3.71
2.06
3.68
2
amount of Li2O2 ((nb-na)/2, µmol/cm ) Li2O2 thickness (nm)
Figure S5 in-situ Raman spectra recorded after discharge to 2.0V in TBA-DMSO solution (black line). Then the electrolyte solution in the cell is totally replaced by fresh Li-DMSO solution at OCP (red line) and hold for 2 min (blue line). The adsorbed superoxide does not change with the increasing Li-ion concentration of LiO2 in solution under the open circuit conditions. This confirms again that no disproportionation reaction occurs.
Raman Intensity (au.)
0.1M TBA-DMSO after discharge to 2.0V change Li-contained DMSO electrolyte Au-O stretch of adsorbed O2
O-O stretch of adsorbed O2 Li2O2 O-O stretch peak position then stay at OCP for 120s change Li-contained electrolyte discharge to 2.0V
400
500
600
700
800
900 -1
Raman Shift (cm )
1000
1100
Figure S6 Deconvoluted current yield for Li2O2 based on the electrochemistry, in-situ UV-vis and SERS measurement during CV sweep: (a) different scan rate in 100mM Li-DMSO solution, (b) different Li-ion concentration in 100mM TBAClO4-DMSO electrolyte.
(a)
(b) 0.10
0.05
(different scan rate)
0.1M TBA-DMSO 10mV/s scan rate current yield for Li2O2 (different Li-ion concentration)
2
Current (mA/cm )
2
Current (mA/cm )
0.05
0.10 0.1M Li-DMSO current yield for Li2O2
0.00
-0.05
10 mV/s 2 mV/s
-0.10
0.00 0 mM 1 mM 2 mM 5 mM 10mM 100mM
-0.05
-0.10
2.0
2.5
3.0
3.5 +
Potential (V vs. Li/Li )
4.0
2.0
2.5
3.0
3.5
4.0 +
Potential (V vs. Li/Li )
Figure S7 (a) UV-vis spectra in KO2 dissolved in DMSO: saturated solution (blue trace) and 5-time diluent solution (red trace). (b) in-situ UV-vis spectra after the potential is swept to 2.0V (10 mV/s) in O2 saturated 0.1M TBAClO4-DMSO and 0.1M LiClO4-DMSO. Temporal dependence for UV-Vis absorbance: (c) at 310 nm (in KO2 saturated DMSO) and 252 nm peak position (5-times diluent KO2-DMSO) after 0.1 M Li-ion is added. (d) 252 nm after the potential is swept to 2.0V (black trace: no Li-ion added into TBA-DMSO, blue trace: 0.1M Li-ion added into TBA-DMSO). Therefore, the possibility for chemical disproportionation reaction of LiO2 ican be almost ignored.
(a)
(c) KO2 saturated DMSO 500uL KO2 saturated DMSO added into 2.5mL DMSO solvent (KO2@DMSO : DMSO=1:5)
7 6
Absorbance
Absorbance
6
8
4
2
after added Li-DMSO to 100mM KO2 saturated DMSO (310 nm) KO2 saturated : DMSO (1:5) (252 nm peak position)
5 4 3 2 1
0
KO2-saturated DMSO (310nm) KO2 saturated DMSO : DMSO (1:5)
0 min 2.3419 5.52
15 min 2.3072 5.258
retention ratio 98.52% 95.25%
0 200
250
300
350
400
450
0
500
3
Wavelength (nm)
(b)
9
12
15
0.25 Absorbacne in DMSO electrolyte after 10 mV/s discharge to 2.0 V 0.20
TBA Li
Absorbance
0.15
Absorbance
(d)
10 mV/s discharge to 2.0 V in DMSO electrolyte
0.20
6
Time (min)
0.10
0.15
TBA Li TBA added Li-ion
0.10
0.05
0.05 0.00
TBA Li TBA added Li-ion
0 min 0.171 0.159 0.171
15 min 0.172 0.1568 0.1675
retention ratio 100.6% 98.37% 97.95%
0.00 250
260
270
280
Wavelength (nm)
290
300
0
3
6
9
12
15
Time (min)
Preparation for KO2 saturated DMSO solution: the excess amount of KO2 powder was added into DMSO solution and then ultrasound for 30min. After stirring for another 2h, the turbid liquid is further centrifuge and the clarify liquid is taken out.
