Supporting Information

Transcription

1 Supporting Information Deciphering EC-PC Mystery in Li-ion Battery Lidan Xing * a,ǁ, Xiongwen Zheng a,ǁ, Marshall Schroeder b, Judith Alvarado b, Arthur von Wald Cresce b, Kang Xu *b, Qianshu Li a, Weishan Li *a a. Engineering Research Center of MTEES (Ministry of Education), Research Center of BMET (Guangdong Province), Engineering Lab. of OFMHEB (Guangdong Province), Key Lab. of ETESPG (GHEI), and Innovative Platform for ITBMD (Guangzhou Municipality), School of Chemistry and Environment, South China Normal University, Guangzhou , China b. Electrochemistry Branch, Sensor and Electron Devices Directorate, Power and Energy Division, U.S. Army Research Laboratory, Adelphi, MD 20783, USA. Corresponding Author *xingld@sucn.edu.cn; conrad.k.xu.civ@mail.mil; liwsh@scnu.edu.cn

2 Calculation methods: All calculations were performed using the Gaussian 09 package 1. The geometry of molecules and solvation shells were optimized using the B3LYP level of theory in conjunction with the G (d) basis set 2,3,4,5,6. Frequency analyses were done with the same basis set to confirm the obtained optimized stationary point. The atomic charge distributions were computed form natural population analysis (NPA) by using natural bond orbital (NBO) theory. To investigate the role of the environment, the bulk solvent effect was estimated using the polarized continuum models (PCM) 7. The acetone dielectric constant (20.5) was used to represent the solvent for PCM calculations. Solvation energy (See Figure 1, Figure S6) of the n th solvent is defined as the energy difference between (LiPF 6 : n solvent) and {LiPF 6 : (n-1) solvent} and solvent. Specifically, it is the energy released when {LiPF 6 : (n-1) solvent} interacts with the - new added one solvent. While solvation energy of PF 6 with the (Li + : n solvent) is defined as the energy difference between (LiPF 6 : n solvent) and (Li + : n solvent) and PF - 6, hence it is the energy released when (Li + : n solvent) interacts with PF - 6. Electron affinity energy is the energy difference of the investigated structure after and before reduction. The lower the electron affinity energy it is, the higher reductive activity the structure will have. Experimental methods: Graphite half-cell experiments: The graphite electrodes contained 91.83% Gelon G15 MCMB with 2 wt% C45 conductive additive and 6 wt% Kureha 9300 binder, and were coated on 10 µm thick Cu foil. These electrodes were provided by the CAMP Facility at Argonne National Laboratory. These electrodes were punched to 9/16 diameter (1.6 cm 2 area) and assembled in 2032 coin cells (Hohsen Corp.) with Celgard 3501 separators soaked in various electrolytes (1M, 3.5M LiPF 6 propylene carbonate (PC), 1M LiPF 6 ethylene carbonate (EC), and 1.2M LiPF 6 EC: ethylmethyl carbonate (3:7)) and a 1.5 mm thick,

3 0.5 diameter Li metal counter-electrode (FMC Corp.). Cell assembly was carried out in a dryroom (dew point < -75 C). After resting for 12 hours, the coin cells were cycled in a 30 environmental chamber with a Maccor battery cycler. The galvanostatic cycling protocol included upper and lower voltage limits of 2.0V and 0.05V, respectively, for a formation cycle at C/20 followed by 4.5 cycles at C/5 until the end of the sixth discharge (lithiated graphite). Cells that exhibited exfoliation/decomposition behavior without reaching the lower limit (i.e. 1.0M PF 6 PC) were discharged at C/20 for 80 hours for consistency in the XPS analysis. XPS sample preparation/analysis: Following cycling, all coin cells were disassembled in an argon-filled Vacuum Atmospheres Nexus One glovebox (H 2 O < 1 ppm, O 2 <1ppm). The electrodes were rinsed with anhydrous dimethyl carbonate and dried in vacuum at room temperature to evaporate any residual solvent. The samples were transferred to a PHI Versaprobe III XPS system using a sealed vacuum transfer capsule enabling rigorous air/moisture exclusion, so the electrode surface chemistry is believed to be unaltered. The XPS was operated using Al anode source at 15 kv with a 100 µm x 100 µm spot size and charge compensation was provided by the PHI charge neutralization system to eliminate differential charging. Survey scans were collected with a pass energy of 224 ev and a 1.0 ev step size followed by high-resolution scans with a pass energy of 26 ev and a step size of 0.05 ev. Peak fitting was performed using CasaXPS software (version , Casa Software Ltd.), using 70/30 Gaussian/Lorentzian line shapes on a linear background. Quantification was performed using peak area corrections to account for the photoionization cross section of each element and the instrument geometry. All spectra were shifted relative to the binding energy of the carbon 1s sp 3 (assigned to ev) to compensate for any off-set during the measurement.

4 Results and discussion: Figure S1. Optimized structures of (LiPF 6 :n solvent) (n =1~5) solvation shells.

5 Figure S2. First discharge/charge curve of the investigated electrolytes. The graphite/lithium half cells are cycled at C/20 in the pictured cycle. It is evident that the half cell behavior of the PC-LiPF 6 electrolytes is consistent with that of LiFSI, showing suppression of electrolyte reduction (~0.9V) scaling with concentration. However, highly concentrated FSI - exhibits better capacity than concentrated PF - 6 due to its higher F-donating capability.

6 Figure S3. High resolution F1s XPS spectra of graphite electrodes cycled in different electrolytes following the sixth discharge.

7 Figure S4. NBO atomic charge distribution of the investigated ions and solvents (A), solvation energy of solvent with Li + (B) and with PF - 6 (in kj/mol) (C), respectively. The atomic charge of =O in EC is slightly higher than that of PC, resulting in stronger interaction ability with Li + than that of the later, as shown in Figure S4B. The solvation energy of Li + - with EC (-22.9 kj/mol) is obviously weaker than that of PF 6 (-24.5 kj/mol), explaining the Li + - solvation shell tends to lose EC instead of PF 6 in EC-based electrolyte. The solvation energy difference of Li + with PC and PF - 6 is only 0.8 kj/mol. On the other hand, it should be noted that in the LiPF 6 :4 solvent structure (see Figure S1), the PF - 6 interacts with the alkyl groups of 3 solvents. Hence, a lower charge density of alkyl groups would show weaker interaction ability of solvent with PF - 6, which would be easier to lose PF - 6. The atomic charge of alkyl groups in EC and PC is 0.70 and 0.72, respectively. Considering that the alkyl chain in EC is shorter than that of PC, the charge density of the alkyl group in EC should higher than PC, leading to stronger interaction ability with PF - 6 of the former than the later, as shown in Figure S4C. Therefore, comparing with EC, PC shows stronger interaction ability with Li + and weaker ability with PF - 6, resulting in different desolvation behavior, as the discussion of Figure 3.

8 Figure S5. Optimized structure (left) of Li +, PF 6 -, EC and PC before and after reduction, together with the calculated electron affinity energies (right). Figure S5 presents the optimized structure of Li +, PF 6 -, EC, PC and SL before and after reduction (represent as +e ), together with the calculated electron affinity energies. For EC and PC solvents, the dihedral angle of CO 3 group increases from 0 to 25 after gaining an electron. According to the calculated electron affinity energy, it can be found that the reductive activity of the investigated solvents and ions decreases as EC>PC>PF 6 - >Li +.

9 Figure S6. Optimized structures of (Li + : n solvent) (n=1~4) solvation shell. Figure S7. Optimized structures of (Li + :n solvent) (n =1~4) solvation shells after one electron reduction reaction. The gained electron is mainly distributed on the solvent highlighted with dotted circle.

10 Figure S8. Calculated electron affinity energy of (Li + :n solvent) (n =1~4) solvation shells. Figure S9. Possible decomposition paths of EC-Li + +e and PC-Li + +e. Together with some selected bond length of the optimized structure.

11 Figure S10. Possible termination reactions of the main decomposition products of EC-Li + +e, PC-Li + +e and SL-Li + +e. Decomposition mechanisms of PC-Li + and EC-Li + were studied to illustrate the difference between these solvents. The optimized structure of reactants, transition state and products are presented in Figure S9, together with the corresponding reaction energy profiles. According to the reaction energy profile shown in Figure 9, the most possible initial decomposition products of EC-Li + +e and PC-Li + +e would be EC-M2 and PC-M1, PC-M2, respectively, which may further terminate to generate oligomer LEDC and LPDC, as shown in Figure S10. Figure S10 shows the possible termination reactions of the main decomposition products of EC-Li + +e and PC-Li + +e. It can be noted from Figure S10 that PC-M2 could generate PC-M2-1 via a radical transfer transition state. PC-M2-1 would further terminate with PC-M3 to generate LPDC-L, the structure of which is as flat as LEDC. References: 1. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.;

12 Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A, Gaussian, Inc, Wallingford, CT, Xing, L. D.; Li, W.S.; Wang, C.Y.; Xu, M.Q.; Tan, C.L.; Yi, J. Theoretical Investigations on Oxidative Stability of Solvents and Oxidative Decomposition Mechanism of Ethylene Carbonate for Lithium Ion Battery Use. J. Phys. Chem. B. 2009,113: Xing, L. D.; Wang, C.Y.; Li, W.S.; Xu, M.Q.; Meng, X.L.; Zhao, S.F. Theoretical Insight into Oxidative Decomposition of Propylene Carbonate in the Lithium Ion Battery. J. Phys. Chem. B. 2009,113: Wang, Y. T.; Xing, L. D.; Li, W. S.; Bedrov, D. Why Do Sulfone-Based Electrolytes Show Stability at High Voltages? Insight from Density Functional Theory. J. Phys. Chem. Lett. 2013, 4, Xing, L. D.; Borodin, O. Oxidation Induced Decomposition of Ethylene Carbonate from DFT Calculations-Importance of Explicitly Treating Surrounding Solvent. Phys. Chem. Chem. Phys. 2012, 14, Wang, Y.T.; Xing, L.D.; Borodin, O.; Huang, W.N.; Xu, M.Q.; Li, X.P.; Li, W.S. Quantum Chemistry Study of the Oxidation-Induced Stability and Decomposition of Propylene Carbonate-Containing Complexes. Phys. Chem. Chem. Phys.2014, 16: Xing, L. D.; Borodin, O.; Smith, G. D.; Li, W. S. Density Functional Theory Study of the Role of Anions on the Oxidative Decomposition Reaction of Propylene Carbonate. J. Phys. Chem. A 2011, 115,