1. Performance studies of a rapid charging Protocol for Li-ion Cells by Godfrey Sikha, P.Ramadass, Bala S. Haran, Ralph E. White, Branko N. Popov Center for Electrochemical Engineering, Department of Chemical Engineering, University of South Carolina Columbia, SC 29208

2. Objectives  To develop a new protocol for charging Li-ion cells.  To reduce the total charging time.  To reduce capacity fade due to over-charging.  To develop a smart charger after optimizing the cell capacity, charging time and capacity decay.

3. Need for a New Protocol…  The charging time for CC-CV protocol is higher.  Using higher DC currents during CC charge keeps the cell during most part of the time in CV mode, but still there is no considerable decrease in total charging time.  Usage of CV mode for the entire charging time could decrease the total charging time.  But the CV charging needs a very high current during the earlier stages of charging that makes the cost of the charger high.In addition its capacity fade is more  Thus, a new protocol that may be similar to CV charging, but not using very high currents, could be ideal for a smart charger and could reduce the charging time to a great extent.

4. Development of Current Decay Protocol Our Approach.. Development of Current Decay Protocol Our Approach..  Charging the Li-ion cell with linearly descending current with time reduces the charging time as compared to constant current charging *.  Similar mode of charging was tested with commercial Sony 18650 cells.  Based on the preliminary results of LCD charging, the protocol was modified to avoid over charging and to decrease the charging time further.  A new protocol was developed that consists of initial high current DC charging (~3C rate) followed by a current decay with the fixed total charging time.  Cycling studies were carried out with the new protocol and the performance was compared with conventional CC-CV and CV mode of charging. * S.K. Chang, A.A. Andriiko, A.P. Monko and S.H. Lee, Journal of Power Sources 79 (1999) 205-211.

5. Development of New protocol Linear Current Decay Protocol (LCD) Modified Linear Current Decay Protocol (MLCD) The New Protocol High Current DC charging followed by a current decay

6. LCD & MLCD Protocols… LCD-Protocol MLCD-Protocol

7. Variation of Charging Current and Cell Voltage for the New Protocol

8. Experimental Cycling Studies: For the comparison of performance of the New Protocol the following cycling studies were done.  CC-CV Protocol0.9 A charging until 4.2V and a float at 4.2 V until the utilization reached 98% for the first cycle(the corresponding time was kept as a time limit for further cycling)and 1A discharge until the potential drops to 2.5V.  CV ProtocolPotentiostatically controlled at 4.2V until the utilization reached 98% for the first cycle(the corresponding time was kept as a time limit for further cycling) and 1A discharge until the potential drops to 2.5 V  New ProtocolA short 5A pulse until the potential reaches 4.2 V followed by the current decay profile for a total time of 5400 seconds which yielded an utilization ca 98 % and 1A discharge until the potential drops to 2.5 V

9. Post Cycling studies  Rate Capability Studies were done after 150 cycles, where all cells are charged using CC-CV protocol with 1A DC and discharged at different rates namely C/8, C/4, C, 3/2C and 2C.  CV's were obtained at the scan rate of 0.05 mV/s within the voltage range of 2.5-4.2 V at the end of 150cycles.  Impedance measurements were done at fully charged and fully discharged states. (100 SOC & 0 SOC)for fresh and cycled full cells  T-cell studies: The cycled cells were cut open and individual electrodes were cycled against excess lithium as counter and capacity of individual electrodes were measured.  The individual half-cells were subjected to cyclic voltammogram and ac-impedance studies

10. Comparison of Utilization of CV-charging, CC-CV charging and new protocol (cycle 1) 2125 sec 2461 sec 4476 sec

11. Charge Curves of the CC-CV Protocol and CV Protocol CC-CV Protocol CV Protocol

12. Charge Curves of the New Protocol

13. Charge Utilization Curves CC-CVCV New Protocol

14. Discharge Curves CC-CV New protocol CV

15. Variation of Discharge Capacity with Cycling

16. Capacity Fade Comparison after 150 Cycles Mode of Charging % Capacity Fade after… 50 cycles100 cycles150 cycles CC-CV4.065.616.64 CV5.247.5110.42 New Protocol 5.377.549.506

17. Rate Capability after 150 cycles C/2C/8C 3/2C 2C

18. Nyquist Plots for LiCoO 2 half cell(fully Lithiated and delithiated) lithiated delithiated

19. Nyquist Plots for Carbon half cell(fully Lithiated and delithiated) lithiateddelithiated

20.  The overall capacity fade (Q) is the result of two major contributions one from the rate capability (Q1) and other from the secondary material losses (Q2), either carbon on LiCoO 2  A very low rate discharge(C/8) at the end of cycling will give the maximum available capacity on the full cell. The difference between this capacity and the measured capacity will account for the rate capability losses.(Q1)  The T-cells(both LiCoO 2 and Carbon) made from cycled full cells are lithiated against excess Lithium as counter electrode to check the maximum available capacity in the respective secondary material.  The same procedure is done for fresh cell and the degradation of secondary material that is limiting can be found(Carbon showed more degradation as compared to Lithium in all the three protocols)  A capacity balance is performed to find Q3 which is categorized as other losses Quantitative analysis of Capacity fade from half cell measurements

21. Quantitative split of capacity fade Protocol Type Q%Q1% Q2 % ( Carbon) Q3% Constant current constant voltage (CC-CV) 6.642.2793.2131.148 Constant voltage(CV) 10.424.3225.7860.312 New Protocol 9.5064.0145.2590.233

22. Conclusions  New charging protocol was developed for charging commercial 18650 Li- ion cells.  High utilizations can be achieved at short periods of time which is useful for many applications  New protocol shows better performance when compared with CV mode,gains a lot of time despite a poor performance when compared to CC-CV mode of charging.  Effective control of overcharging can be ensured when much better current profiles are chosen.

23. Acknowledgements This work was carried out under a contract with the National Reconnaissance Office for Hybrid Advanced Power Sources # NRO-00-C-1034.