memory effect in a lithium-ion battery - nature research › original › nature-assets ›...
TRANSCRIPT
Supplementary Information
Supplementary Information S1: Enhancing the memory effect by repeating the
memory-writing cycle.
Fig. S1. | Charge and discharge curve of LiFePO4 under the memory-effect
conditions with different repetitions of the memory-writing cycle. A sequence of 12
4.5
4.0
3.5
3.0
2.5
2.0
Vol
tage
(V)
100806040200Time (hour)
1 time 3 times 5 times
a
b
3.50
3.48
3.46
3.44
3.42
Vol
tage
(V)
100806040200Time (hour)
Memory effect
Memory effect in a lithium-ion battery
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
NATURE MATERIALS | www.nature.com/naturematerials 1
© 2013 Macmillan Publishers Limited. All rights reserved.
cycles under three memory-effect conditions; a, 1st and 2nd cycle: the memory effect
with one memory-writing cycle; 3rd-6th cycle: the memory effect with three
memory-writing cycles; 7th-12nd cycle: the memory effect with five memory-writing
cycles, between 2.4 and 4.4 V. b, Enlarged view of Fig. S1a between 3.42 and 3.51 V.
Supplementary Information S2: The effect of the current rate in the memory
releasing cycle.
-
Fig. S2. | Charge curves of the memory-releasing cycle of LiFePO4 at different
current rates; a, 1/4 C and 1C rate. b, 1/4, 1C, and 4C rate. The conditions of the
memory-writing cycles were identical. The depth of the memory-writing cycles was
50% and the current rate was 1/4C. The memory writing cycles had 1 hour rest time
3.75
3.70
3.65
3.60
3.55
3.50
3.45
Vol
tage
(V)
100806040200SOC (%)
Memory releasing cycle 4 C 1 C 1/4 C
3.54
3.52
3.50
3.48
3.46
3.44
Vol
tage
(V)
100806040200SOC (%)
Memory releasing cycle 1 C 1/4 C
a b
2 NATURE MATERIALS | www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION DOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
between charge and discharge in the open-circuit state. The rest time between the
memory-writing cycle and the memory-releasing cycle was 10 minutes.
Supplementary Information S3: Demonstration of the memory effect of LiFePO4
in a three-electrode cell.
Fig. S3. | Charge and discharge curve of LiFePO4 under ‘memory-effect
conditions’ at a SOC of 50% in a three-electrode cell with a Li-metal reference
a
b
4.0
3.5
3.0
2.5
Volta
ge (V
) ver
sus
Li/L
i+
302520151050Time (hour)
3.50
3.48
3.46
3.44
3.42
Vol
tage
(V) v
ersu
s Li
/Li+
302520151050Time (hour)
Memory writing cycle
Memory releasing cycle
Memory effect
Normal cycle
NATURE MATERIALS | www.nature.com/naturematerials 3
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
electrode. a, A sequence of three cycles. First cycle (black line): Normal charge and
discharge curve between 2.4 and 4.4 V vs. Li/Li+; second cycle (blue line, memory
writing cycle): charge up to a SOC of 50% and full discharge until 2.4 V; third cycle
(red line, memory releasing cycle): full charge and discharge between 2.4 and 4.4 V. b,
Enlarged view of Fig. S3a between 3.42 and 3.51 V.
Supplementary Information S4: Demonstration of the memory effect of LiFePO4
by using a very thin electrode.
3.50
3.48
3.46
3.44
3.42
Vol
tage
(V) v
ersu
s Li
/Li+
35302520151050Time (hour)
4.0
3.5
3.0
2.5
Vol
tage
(V) v
ersu
s Li
/Li+
35302520151050Time (hour)
Memory writing cycle
Memory releasing cycle
Memory effect
Normal cycle
a
b
4 NATURE MATERIALS | www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION DOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
Fig. S4. | Charge and discharge curve of a very thin electrode under
‘memory-effect conditions’ at a SOC of 50%. The active material loadings of the thin
electrodes were 0.14-0.17 mg/cm2 (standard electrodes in this study: 5 mg/cm2). a, A
sequence of three cycles. First cycle (black line): Normal charge and discharge curve;
second cycle (blue, memory writing cycle): charge up to a SOC of 50% and full
discharge; third cycle (red line, memory releasing cycle): full charge and discharge. b,
Enlarged view of Fig. S4a between 3.42 and 3.51 V.
NATURE MATERIALS | www.nature.com/naturematerials 5
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
Supplementary Information S5: The effect of the current rate in the memory
writing and releasing cycle.
3.48
3.44
3.40
Vol
tage
(V) v
ersu
s Li
/Li+
1614121086420Time (hour)
Memory writing cycle Memory releasing cycle
3.55
3.50
3.45
3.40
3.35
Volta
ge (V
) ver
sus
Li/L
i+
43210Time (hour)
3.6
3.5
3.4
3.3
Vol
tage
(V) v
ersu
s Li
/Li+
0.80.60.40.20.0Time (hour)
Memory effect
Memory effect
Memory effect
0.2 C
0.8 C
3.2 C
4.0
3.5
3.0
2.5
Vol
tage
(V) v
ersu
s Li
/Li+
1614121086420Time (hour)
Memory writing cycle Memory releasing cycle 0.2 C
6 NATURE MATERIALS | www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION DOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
Fig. S5. | Charge and discharge curve of the memory-writing and
memory-releasing cycle of LiFePO4 at different current rates; a, First cycle (blue
line, memory writing cycle): charge up to a SOC of 50% and full discharge until 2.4 V;
second cycle (red line, memory releasing cycle): full charge and discharge between 2.4
and 4.4 V at about 0.2 C rate. b, Enlarged view of Fig. S5a between 3.38 and 3.50 V. c,
Enlarged view between 3.32 and 3.56 V at about 0.8 C rate. d, Enlarged view between
3.2 and 3.7 V at about 3.2 C rate. The current rates of the memory-writing and
memory-releasing cycles were identical. The memory-writing cycle included a rest time
of 1 minute between charge and discharge in the open-circuit state. The rest time
between the memory-writing cycle and the memory-releasing cycle was 1 minute.
NATURE MATERIALS | www.nature.com/naturematerials 7
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
Supplementary Information S6: The effect of rest time between the memory
writing cycle and the memory releasing cycle
Fig. S6. | Charge and discharge curve of LiFePO4 under the memory effect
conditions with different rest time between the memory writing cycle and the
memory releasing cycle; a, 10 minute rest time (standard condition in this study) b, 5
4.5
4.0
3.5
3.0
2.5
2.0
Vol
tage
(V) v
ersu
s Li
/Li+
100Time (hour)
3.50
3.48
3.46
3.44
3.42
Vol
tage
(V) v
ersu
s Li
/Li+
100Time (hour)
6050403020100Time (hour)
6050403020100Time (hour)
5 hourrest
Memory releasing cycle
24 hourrest
Memory writing cycle
10 min.rest
Memory releasing cycle
a b
c d
8 NATURE MATERIALS | www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION DOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
and 24 hour rest times, c, d, Enlarged views of Fig. S6a (c) and S6b (d) between 3.42
and 3.51 V.
NATURE MATERIALS | www.nature.com/naturematerials 9
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
Supplementary Information S7: The effect of the depth of discharge of the
memory-writing cycle
4.5
4.0
3.5
3.0
2.5
2.0
Vol
tage
(V)
50403020100Time (hour)
Partial charge
Partial Discharge
50 %50 %
Memory writing cycle
Memory releasing cycle
Partial charge
Partial Discharge
45 %50 %
Partial charge
Partial Discharge
40 %50 %
3.50
3.48
3.46
3.44
3.42
Vol
tage
(V)
50403020100Time (hour)
Partial Discharge
50 %
Partial Discharge
45 %
Partial Discharge
40 %
a
b
10 NATURE MATERIALS | www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION DOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
Fig. S7. | Charge and discharge curve of LiFePO4 under the memory effect
conditions with different depth of discharge of the memory writing cycle; a, 50%
(standard condition in this study), 45%, and 40%, c, 35% and 45%. b, d, Enlarged views
4.5
4.0
3.5
3.0
2.5
2.0
Vol
tage
(V)
35302520151050Time (hour)
Partial charge
Partial Discharge
35 %50 %
Memory writing cycle
Memory releasing cycle
Partial charge
Partial Discharge
30 %50 %
c
3.50
3.48
3.46
3.44
3.42
Vol
tage
(V)
35302520151050Time (hour)
Partial Discharge
35 %
Partial Discharge
30 %
d
NATURE MATERIALS | www.nature.com/naturematerials 11
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
of Fig. S7a (b) and S7c (d) between 3.42 and 3.51 V. The rest time between the
memory-writing cycle and the memory-releasing cycle was 10 minute in all
measurements.
Figure S7b and S7d show significant difference not only in the memory bumps but also
in the initial overshoots of the charge in the memory-releasing cycles. With decreasing
depth of discharge of the memory writing cycle, the memory bumps were increasing in
size; on the other hand, the initial overshoots of the charge in the memory-releasing
cycles were decreasing. The latter behaviour also supports our proposed mechanism.
According to our mechanism, the initial overshoots come from the first group of Fig. 5f
(upper group) and the memory bumps come from the second group (lower group).
When the depth of discharge of the memory writing cycle is shallow, a part of the first
group still consists of two-phase coexistence particles between Point B and C in Fig. 5.
The triggering condition for the overshooting is “Charging from the condition of Fig.
5d”; in other words, “Charging from the condition without two-phase coexistence phase
particles”. So, the remaining two-phase coexistence particles will weaken the triggering
condition as well as the overshooting. Charging two-phase coexistence particles is
easier than charging particles at point A to point B in Fig. 5. The beginning of charge of
12 NATURE MATERIALS | www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION DOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
the memory-releasing cycle after the shallow-depth discharge is led by the remaining
two-phase coexistence particles, therefore the overshooting doesn’t appear.
NATURE MATERIALS | www.nature.com/naturematerials 13
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
Supplementary Information S8: The effect of long rest time after shallow-depth
discharge on the memory-writing cycle
Fig. S8. | The effect of long rest time after shallow-depth discharge on the
memory-writing cycle. a, Charge and discharge curve of LiFePO4 under the memory
3.50
3.48
3.46
3.44
3.42
Vol
tage
(V)
403020100Time (hour)
Partial Discharge
30 %
Memory releasing cycle
24 hour rest
4.5
4.0
3.5
3.0
2.5
2.0
Vol
tage
(V)
403020100Time (hour)
Partial charge
Partial Discharge
30 %50 %
Memory writing cycle
Memory releasing cycle
24 hour rest
a
b
14 NATURE MATERIALS | www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION DOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.
effect condition with 30% depth of discharge of the memory writing cycle and 24 hour
rest time between the memory-writing cycle and the memory-releasing cycle. b,
Enlarged views of Fig. S8a between 3.42 and 3.51 V.
The overshoot behaviours of the memory-releasing cycles between Fig. S7d and Fig.
S8b are comparable. Even if the overshooting disappeared after 30% discharge in the
memory-writing cycle (see Fig. S7d), the overshooting appeared again after 24 hour rest
time, as shown in Fig. S8b. This can be interpreted as long rest times making the
remaining two-phase coexistence particles disappear. The remaining two-phase
coexistence particles will establish homogeneous Li-poor or Li-rich phases through
interactions with other particles during the long rest time, due to the chemical potential
gaps. The disappearance of the two-phase coexistence particles lets the initial overshoot
of the charge curve appear again.
NATURE MATERIALS | www.nature.com/naturematerials 15
SUPPLEMENTARY INFORMATIONDOI: 10.1038/NMAT3623
© 2013 Macmillan Publishers Limited. All rights reserved.