1. B. Deniz Polat, Levent Eryılmaz*, Özgül Keleş,
Optimization in the Electrochemical Performance of CuSi Thin Film Anode Produced via Magnetron Sputtering
B. Deniz Polat, Levent Eryılmaz*, Özgül Keleş,
Istanbul Technical University
Department of Metallurgical and Materials Engineering
* Argonne National Laboratory, IL, USA
ECS Meeting, 2013
San Francisco, USA 1

2. Acknowledgements Dr. Ali Abouimrane
PhD. Candidate, Billur Deniz Polat,
PhD. Candidate, Nagihan Sezgin
Assoc. Prof. Dr. Kürşat Kazmanlı
Dr. Ali Abouimrane
Dr. Khalil Amine
Tübitak MAG
110M148 2

3. Content Aim of the study Anodes used for Li-Ion Batteries
CuSi Thin Film Anodes
Nanostructured Sn, CuSn, CuSnC Thin Film Anodes
Results and discussion
Conclusions 3

4. Aim of the study
To design new anode materials having architectured structures using thin film technologies to improve the electrochemical performances of the anode materials.
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

5. Why Li-ion Batteries
Higher cell voltage & higher energy/power densities
Ni-MH
Ni-Zn
Ni-Cd
Lead-Acid
Gravimetric Energy Density (Wh/kg)
Volumetric Energy Density (Wh/l)
400
350
300
250
200
150
100 50
Lithium Batteries
(LIB, LPB…)
lighter
smaller

6. Present Lithium Battery Technologies
Capacity 150 mAh/g
V= 3.6 V
Capacity 100 mAh/g
V= 3.8 V
Capacity 150 mAh/g
V= 3.3 V
Capacity 372mAh/g
V= 0.2V

7. Evolution of Anode Materials
1990
2013 I .
Evolution of Anode Materials
1970
1980 I
1990
Mesocarbon Microbeads
SnMx(M: inactive Materials)
SiMx(M: inactive materials)
Lithium Metal
Petroleum Coke
Soft Carbon
Hard Carbon
Sn-SnO
Si based anodes
In 2005 to improve the electrochemical performance of the anode;
It has been proposed using an amorphous electrode material consisting of SnAX
(where A is a transition metal and
X stands for O, F, N, Mg, Ba, Sr, Ca, La, Ce, Si, Ge, C, P, B, Pb, Bi, Sb, Al, Ga,
In, Tl, Zn, Be, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, As, Se, Te, Li, and S)
Then it has been anticipated this idea by using SnMX systems
(where M is one of Ni, Cu, Fe, Co, Mn, Zn, In or Ag, and
X is one of B, C, Al, Si, P, or S.)
High Specific Capacity
SAFETY
High Specific Capacity
SAFETY
SAFETY
Low Specific Capacity
Alloying with inactive metals
Porous morphology
Expansion -Failure
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

8. Anode Materials Commercially available and proposed anode materials C
Materials Properties Li C
Li4Ti5O12 Si Sn Sb Al Mg Bi
Density (gcm-3)
0.53
2.25
3.5
2.33
7.29
6.7
2.7
1.3
9.78
Lithiated Phase
LiC6
Li7Ti5O12
Li4.4Si
Li4.4Sn
Li3Sb
LiAl
Li3Mg
Li3Bi
Theoretical Specific Capacity (mAhg-1)
3862
372
175
4200
994
660
993
3350
385
Theoretical Charge Density (mAhcm-3)
2047
837
613
9786
7246
4422
2681
4355
3765
Volume Change (%)
100 12 1
320
260
200 96
215
Potential vs. Li (V)
0.05
1.6
0.4
0.6
0.9
0.3
0.1
0.8

9. Design Criteria INPUTS OUTPUTS Large reversible capacity
Stable-Good electronic conductivity
Capable of making reversible Li rich intercalation compounds (high Li intercalation density)
Mininum volume changes during intercalation /tolerability volume changes without failure
Compatibility with electrolyte
Large reversible capacity
Small irreversible capacity
Rate capability
Long cycle and calendar life
Safety
PROCESSING CRITERIA
Easy of processing
Low cost

10. Mechanically stable high energy density anode
Design Criteria
Mechanically stable high energy density anode
Performance
THIN FILM
NANO
MATERIALS
Process
Structure
COMPOSITE
MATERIALS
POROUS
Stable-Good electronic conductivity
Reversible Li rich intercalation compound making capability (high Li intercalation density)
Properties
HIGH ELASTIC
STRENGTH
HIGH ELECTRONIC
CONDUCTIVITY
Evaporation Sputtering
Mininum volume changes during intercalation /tolerability volume changes without failure
Compatibility with electrolyte
E-beam Magnetron

11. Design Criteria Glancing Angle Deposition Porosity in the thin film
Target
Glancing Angle
Incident Vapor
Atoms
Source

12. Motivation and Design Anodes Thin Film (Sn or Si based material)
Technology
E-beam
or Magnetron sputtering
Nanoporous
Anodes
inclined nanorods

13. Motivation and Design
Among alternative inactive metals Cu is particularly chosen, because:
It has high electrical conductivity
It has high mechanical tolerance to volumetric changes occurred during cycling
xLi+ + Si(crystal) + xe-LiXSi (amorphous)
LixSi(amorphous) (x-y) Li+ + LixSi(amorphous) +(x-y) e-

14. Experimental Target 2 Target 1 Si Cu Substrate Holder

15. Experimental Basic deposition parameters for CuSi thin films
Cu 30 at%
Cu 20 at%
Cu 10 at%
Power applied to Cu target (W)
285
200
150
Power applied to Si target (W)
1900
2000
3000
Incident flux angle 0o
80o
The deposition time is fixed in order to have the same thickness (3 microns approx.)
for all films

16. Results and Discussions SEM surface analyses of the CuSi thin films
Cu 30 at%-0o
Cu 20 at%-0o
Cu 10 at%-0o
Cu 10 at%-80o
Cu % at.
31.6
23.3
12.9
11.6
Si % at.
68.4
76.7
87.1
88.4

17. Results and Discussions Comparitive XRD analyses of the CuSi thin filmsb)
Cu 10 at%-0o
Cu 20 at%-0o
Cu 30 at%-0o
Cu content in the thin film promotes Cu3Si formation.
The comparison in the characteristic peaks of Cu3Si shows that lower Cu content enhances the nanosized morphology of the thin film

18. Results and Discussions 1st and 2nd cycles CV results of thin films
Cu 30 at%-0o
Cu 20 at%-0o
Cu 10 at%-0o
Cu 10 at%-80o
First cycles are similar
All samples are covered with SEI layer
After the first cycle the shape of CV curvature and the number of peaks change. A possible morphological change in the thin film in turn reveals new active material through SEI layer due to volumetric change.
When the morphology of the thin film becomes more porous, peak intensities gets higher.

19. Results and Discussions Charge/Discharge Capacities of the thin films with different compositions
Cu 10 at%-0o
Cu 20 at%-0o
Cu 30 at%-0o

20. Results and Discussions Charge/Discharge Capacities of the thin films with different morphologies
Cu 10 at%-80o
Cu 10 at%-0o
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

21. Conclusions
Change in the electrochemical performances of the thin films depending on their composition and morphology was seen.
Oblique angle magnetron sputtering technique was used the first time in literature to induce homogenously distributed nanosized porosities in the thin film.
The results show that the nanostructured thin film having 10 %at. Cu and 90 %at. Si performs 1100 mAhg-1for 100 cycles. Considering the advantageous of both magnetron sputtering and GLAD methods, high energy of the sputtered particles forming porous, nanostructured composite thin film having high adherence could explain this performance.

22. B. Deniz Polat, Nagihan Sezgin, Özgül Keleş,
Use of Tin Based Composite Nanorods Anodes for Rechargeable Lithium Applications
B. Deniz Polat, Nagihan Sezgin, Özgül Keleş,
Istanbul Technical University
Department of Metallurgical and Materials Engineering
* Argonne National Laboratory, IL, USA

23. Aim of the study
To design new anode materials having architectured structures using thin film technologies to improve the electrochemical performances of the anode materials.
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

24. Why Li-ion Batteries
Higher cell voltage & higher energy/power densities
Ni-MH
Ni-Zn
Ni-Cd
Lead-Acid
Gravimetric Energy Density (Wh/kg)
Volumetric Energy Density (Wh/l)
400
350
300
250
200
150
100 50
Lithium Batteries
(LIB, LPB…)
lighter
smaller

25. Present Lithium Battery Technologies
Capacity 150 mAh/g
V= 3.6 V
Capacity 100 mAh/g
V= 3.8 V
Capacity 150 mAh/g
V= 3.3 V
Capacity 372mAh/g
V= 0.2V

26. Evolution of Anode Materials
1990
2013 I .
Evolution of Anode Materials
1970
1980 I
1990
Sony
Fuji
Mesocarbon Microbeads
SnMx(M: inactive Materials)
SiMx(M: inactive materials)
Lithium Metal
Petroleum Coke
Soft Carbon
Hard Carbon
Sn-SnO
Si based anodes
In 2005 to improve the electrochemical performance of the anode;
It has been proposed using an amorphous electrode material consisting of SnAX
(where A is a transition metal and
X stands for O, F, N, Mg, Ba, Sr, Ca, La, Ce, Si, Ge, C, P, B, Pb, Bi, Sb, Al, Ga,
In, Tl, Zn, Be, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, As, Se, Te, Li, and S)
Then it has been anticipated this idea by using SnMX systems
(where M is one of Ni, Cu, Fe, Co, Mn, Zn, In or Ag, and
X is one of B, C, Al, Si, P, or S.)
High Specific Capacity
SAFETY
High Specific Capacity
SAFETY
Low Specific Capacity
SAFETY
Alloying with inactive metals
Porous morphology
Expansion -Failure
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

27. Anode Materials Commercially available and proposed anode materials C
Materials Properties Li C
Li4Ti5O12 Si Sn Sb Al Mg Bi
Density (gcm-3)
0.53
2.25
3.5
2.33
7.29
6.7
2.7
1.3
9.78
Lithiated Phase
LiC6
Li7Ti5O12
Li4.4Si
Li4.4Sn
Li3Sb
LiAl
Li3Mg
Li3Bi
Theoretical Specific Capacity (mAhg-1)
3862
372
175
4200
994
660
993
3350
385
Theoretical Charge Density (mAhcm-3)
2047
837
613
9786
7246
4422
2681
4355
3765
Volume Change (%)
100 12 1
320
260
200 96
215
Potential vs. Li (V)
0.05
1.6
0.4
0.6
0.9
0.3
0.1
0.8

28. Design Criteria INPUT OUTPUT Large reversible capacity
Stable-Good electronic conductivity
Reversible Li rich intercalation compound making capability (high Li intercalation density)
Mininum volume changes during intercalation /tolerability volume changes without failure
Compatibility with electrolyte
Large reversible capacity
Small irreversible capacity
Rate capability
Long cycle and calendar life
Safety
PROCESSING CRITERIA
Easy of processing
Low cost

29. Mechanically stable high energy density anode
Design Criteria
Mechanically stable high energy density anode
Performance
THIN FILM
NANO
MATERIALS
Process
Structure
COMPOSITE
MATERIALS
POROUS
Stable-Good electronic conductivity
Reversible Li rich intercalation compound making capability (high Li intercalation density)
Properties
HIGH ELASTIC
STRENGTH
HIGH ELECTRONIC
CONDUCTIVITY
Evaporation Sputtering
Mininum volume changes during intercalation /tolerability volume changes without failure
Compatibility with electrolyte
E-beam Magnetron

30. Design Criteria Glancing Angle Deposition Porosity in the thin film
Target
Glancing Angle
Incident Vapor
Atoms
Source

31. Motivation and Design Sn + Li ⇌ SnxLiy 10Li + 5 Cu6Sn5 ⇌ 5Li2Cu6Sn5
Among alternative inactive metals Cu is particularly chosen, because:
It has high electrical conductivity
It has high mechanical tolerance to volumetric changes occurred during cycling
Sn + Li ⇌ SnxLiy
10Li + 5 Cu6Sn5 ⇌ 5Li2Cu6Sn5
20Li + Li2Cu6Sn5⇌ 5Li4.4Sn + 6Cu

32. (Sn and SnC based material)
Motivation and Design
Intermetallic Anodes
(Sn and SnC based material)
Thin Film
Technology
E-beam
Nanoporous
Anodes
flat
inclined nanorods

33. 15.5 mm diameter, 1.5mm thickness Cu discs
Experimental
Production of Sn based thin film electrodes via glancing angle electron beam deposition method Sn
CuSn
CuSnC
Substrate properties
15.5 mm diameter, 1.5mm thickness Cu discs
The base vacuum in the chamber
᷉10-7mTorr
1st crucible composition
Sn pellets
Mixture of
Cu-Sn pellets
(2/3 weight%)
2nd crucible composition -
Graphite
Incident angle
80 o

34. Results and Discussions
Sn Cu-Sn Cu-Sn-C
Knowing that the source composition was 50%Sn (%at) and 50%Cu (%at), the EDS analyses demonstrate that the Cu-Sn film contains 45%Sn (%at) and 55%Cu (%at ). The difference between the source and the thin film composition might be explained as an outcome of the evaporation point differences of the metals at constant pressure and by the nature of GLAD method. Because in GLAD method, the deposition rate not only has a vertical component (with respect to the substrate surface), but also has a lateral component, which in turn could form thin films with different compositions and thicknesses [16].  Bunlar atomk mi
Cu (wt%)
Sn (wt%)
C (wt%)
Cu-Sn
53.07
46.93
Cu-Sn-C
32.19
59.24
8.57
Figure 1
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

35. Results and Discussions Comparitive thin film XRD of the pristine electrodes
Figure 2 demonstrates the small peak shift on the Cu-Sn film, which could be explained by the internal stress of the thin film due to the excessive amount of Cu atoms present in the Cu6Sn5 crystal. However, the XRD data of Cu-Sn-C containing thin film shows no peak shift. This could be explained by the carbon presence in the thin film, which results in a reduction of the particle size in the thin film. The absence of the characteristic peak at 89.7o for Cu6.26Sn5 also justifies that excessive Cu is not present in the crystal of Cu6Sn5 for Cu-Sn-C thin film. Moreover XRD data of (Cu-Sn-C) does not reveal any peaks related to the carbon presence, which could be related with the amorphous phase of Carbon as revealed by the presence of a bump at low diffraction angles.
Cu-Sn-C containing thin film shows no peak shift
The absence of the characteristic peak at 89.7o for Cu6.26Sn5 also justifies that excessive Cu is not present in the crystal of Cu6Sn5 for Cu-Sn-C thin film.
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

36. Results and Discussions First cycle CV analysis results of the thin films
Cu has no effect on the peak positions, which justifies that it is inactive when reacting with Li; and C enhances the activity of Li+ since the oxidation and reduction peak intensities are much higher than those of the nanostructured Sn and CuSn thin films.
All samples are covered with SEI layer
The reduction peaks around 0.3 and 0.1 V may indicate the formation of Li-Sn (Li-rich) alloys formation. The anodic peaks noted around 0.5 and 0.7V may demonstrate that SEI layer
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

37. Results and Discussions 1st and 3rd cycles EIS analysis of the thin filmsSn
CuSn
CuSnC
a semicircle at high frequency region represents the lithium ion interfacial transfer resistance
a line having 45° angles at low frequency region showing the existence of solid-state diffusion of Li+ into the bulk alloy material.
the nanostrucutred Sn thin film has two semicircles on the EIS data after the first cycle, which proves that the nanostructured film has multiple interfaces with the electrolyte.
C presence in the SEI covering the electrode surface could work as binder in the thin film, which may decrease the contact and internal resistance of the film after.
the nanostrucutred Sn thin film has two semicircles on the EIS data after the first cycle, which proves that the nanstructured film has multiple interfaces with the electrolyte. After the first cycle, the semi-circles gather together to form one circle at the high frequency region and the charge-transfer resistance noted after the third cycle decreases compared to that of the first cycle. This change in impedance could be explained considering the the pulverization of the thin film due to high volume changes, in turn distort the film morphology (the pillar morphology is lost, leading to small soze particles containing thin film) and the interfaces between the thin film and the electrolyte are settled together eventually. This increase in nanostructured Sn thin film’s surface area during cycling affects the detected amount of current in the impedance. This rise in the current recovered from the electrode could explain the decrease in the impedance of the cells after the 3rd cycle. On the other hand,the impedance data of the nanostructured Cu-Sn thin film reveals that an additional semi-circle is formed on the EIS data of the 3rd cycle where the real impedance value between the current collector and the thin film is also decreased. This behavior proves that the morphology of the Cu-Sn nanorods containing thin film is failed partially after three cycles. The partial delamination of the thin film (see Figure 5a) occurred due to the volumetric change, results in the formation of the second circle on its EIS data (Fig 4b). Besides, Figure 5a shows that even the volume of the Cu-Sn thin film is increased during lithiation, the film did not delaminate entirely thanks to the improved adhesion (due to Cu presence) and the homogenously distributed porosities in the composite thin film. When the EIS data (both the as deposited and after the 3rd cycles) of the nanostructured Cu-Sn-C thin film are observed, a decrease in the internal resistance is detected, whereas the real impedance at the interface remains slightly constant. Previous work shows experimentally that C is generally present in the SEI film that formed on the electrode after the first discharge reaction and mostly remains on the surface during the cycling test [18]. Thus, the expected C presence in the SEI covering the electrode surface could work as binder in the thin film, which decreases the contact and internal resistance of the film after the 3rd cycle. The Sem surface view of the Cu-Sn-C thin film after 80cycles justifies the partial delamination in the thin film ans thr pulverization occured on the remained part of the thin film (see Fig 5b).
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

38. Results and Discussions Charge/Discharge Capacities of the thin film electrodes vs Li/Li+Sn
CuSnC
CuSn

39. Results and Discussions SEM analyses of the CuSn and CuSnC thin films after cycling tests

40. Conclusions
In this work, oblique angle deposition process is used to produce nanostructured Sn, Cu-Sn and Cu-Sn-C thin films.
A carbon containing nanostructured Cu-Sn thin film was produced by using oblique angle deposition method.
Introducing another element (Cu and/or C) into the Sn thin film has result in variation in thin film morphology and mechanical properties, which eventually affected the electrochemical performance of the anode materials.
Even the addition of C deteriorated the nanorods formation due to the high mobility of surface adsorbed C atoms, the best performance was achieved when Cu-Sn-C thin film was used as anode: the capacity became fairly constant around 450 mAhg-1upto 80th cycles.
Thin film production process may open a new area of research for the development of metal storage electrodes since there is no risk of hazardous handling of flammable, explosive or cancerogenous metal nanoparticles; nor a need and limitation of (traditional) binder or conductive additives and it is possible toproduce nanosructured films.
pure Sn nanorods have a higher initial anodic capacity around 980 mAhg-1, but the capacity diminishes after 20 cycles due to the morphological changes and the cycling continued upto 50th cycles, then it failed. By introducing approximately 50 at.% Cu into Sn nanorods, an anode material containing Cu-Sn composite nanorods with 100nm in length, 20 nm in diameter was fabricated. This electrode demonstrated approximately 800 mAhg-1 of initial discharge capacity. Its capacity diminished gradually after the first five cycles, then it was stable up to 80th cycles with a discharge capacity around 300 mAhg-1. Then in order to see the effect of C in the thin film nanostructured Cu-Sn-C thin film was also produced.
Go to ”Insert (View) | Header and Footer" to add your organization, sponsor, meeting name here; then, click "Apply to All"

41. Thank you for your attention…