- Email: [email protected]
Effects of CO2 accumulation during cycling of a Li–O2 battery on the transition of discharge product and performance fading
Journal Pre-proof Effects of CO2 accumulation during cycling of a Li–O2 battery on the transition of discharge product and performance fading Li-Bin Chen, Yu-Hao Hong, Liang-Ping Xiao, Jin-Hai You, Wen-Jia Sheng, Ling Huang, Hua Bai, Shi-Gang Sun PII:
S2211-2855(19)30878-X
DOI:
https://doi.org/10.1016/j.nanoen.2019.104171
Reference:
NANOEN 104171
To appear in:
Nano Energy
Received Date: 11 June 2019 Revised Date:
29 September 2019
Accepted Date: 4 October 2019
Please cite this article as: L.-B. Chen, Y.-H. Hong, L.-P. Xiao, J.-H. You, W.-J. Sheng, L. Huang, H. Bai, S.-G. Sun, Effects of CO2 accumulation during cycling of a Li–O2 battery on the transition of discharge product and performance fading, Nano Energy (2019), doi: https://doi.org/10.1016/ j.nanoen.2019.104171. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.
TOC Graphic: A thorough investigation on the effects of CO2 accumulation during cycling of a Li-O2 battery employing RuO2/rGO cathode leads to realize a strategy to preload CO absorbent (CaO) into the battery, which has prolonged significantly the cycle lifespan.
1
Effects of CO2 Accumulation During Cycling of a Li-O2 Battery on the Transition of Discharge Product and Performance Fading
Li-Bin Chen a, Yu-Hao Hong a , Liang-Ping Xiao a, Jin-Hai You c, Wen-Jia Sheng a, Ling Huang a, Hua Bai b, Shi-Gang Sun*a a
State Key Laboratory of Physical Chemistry of Solid Surfaces, Department of Chemistry,
College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, 361005, P. R. China. b
College of Materials, Xiamen University, Xiamen 361005, P. R. China.
c
College of Energy, Xiamen University, Xiamen 361005, P. R. China.
*
Corresponding author, E-mail: [email protected]
Abstract The increase in charge potential during discharge-charge cycling reduces severely the cycle life of a Li-O2 battery, but its origin has not been fully understood yet. The current study focuses on revealing the intrinsic basis behind the increase of charge potential of a Li-O2 battery and developing a strategy to inhibit this phenomenon. Based on results of X-ray diffraction, Fourier transform infrared spectroscopy, scanning electron microscope and online electrochemical mass spectroscopy, we find that the performance fading of a Li-O2 battery evidenced by the increase in charge potential is caused by the discharge product transition from Li2O2 to Li2CO3 due to the accumulation of by-product CO2 in cathode side reaction during cycling. We further demonstrate that the performance of a cycled Li-O2 battery can be completely 1
recovered when the accumulated CO2 is evacuated by a vacuum pumping treatment. A strategy is therefore proposed to suppress the CO2 accumulation by preloading a CO2 absorbent agent (CaO) into Li-O2 battery at its assemblage. As a result, the phenomenon of charge potential increase has been effectively inhibited, which improved significantly the cycleability. In comparison with a Li-O2 battery without preloading CaO, the cycling life of the CaO preloaded Li-O2 battery has been prolonged to 148%. This study reveals an important mechanism of performance fading of Li-O2 battery and develops an efficient approach to increase the cycle life, which has thrown new insight into the design and construction of Li-O2 batteries with long lifespan.
2
1 Introduction With the increasing demands on energy density by moveable electronic devices and electric vehicles, a great number of studies have focused on energy storage systems with higher energy density. Li-O2 battery is considered as the next generation of energy storage device owing to its high theoretical specific energy density.[1, 2] A nonaqueous Li-O2 battery consists of lithium anode, separator, oxygen cathode and nonaqueous electrolyte. In an ideal discharge process, the oxygen cathode undergoes oxygen reduction reaction (ORR), producing Li2O2, which is then oxidized into O2 via oxygen evolution reaction (OER) in the charge process. The discharge product Li2O2, however, is an electrically non-conductive compound and has slow oxidation kinetics. As a consequence, a large overpotential on the cathode is required to charge the Li-O2 battery. High charge potential will instigate side-reactions and accelerate electrolyte decomposition, resulting in poor round-trip efficiency and passivation of the cathode. Therefore, various electrocatalysts, such as manganese dioxide, cobalt oxide, ruthenium dioxide, have been applied to promote the ORR and OER, and they have efficiently decreased the charge overpotential.[3-6] However, even with appropriate electrocatalysts, the Li-O2 batteries usually display poor cycling stability. The charge potential of Li-O2 batteries generally increases as the discharge-charge cycling going on, which represents the performance fading of Li-O2 battery. Up to now, the charge potential increase has not been understood completely, because many factors can impact on the performance of Li-O2 battery.[1, 2] Some studies proposed that the charge potential increase may be caused by the passivation of the catalyst due to its surface covered by undecomposed residues.[7] Other investigations attributed it to side reactions of electrolyte and oxidation of carbon in electrode.[8, 9] Since the fast performance fading is one of the fatal disadvantages of Li-O2 battery for practical application, it is crucial to reveal the mechanism of charge potential increase during discharge-charge cycling and develop strategy to suppress it. 3
In this study, by taking Li-O2 battery with RuO2/rGO (reduced graphene oxide) cathode catalyst and LiCF3SO3-TEGDME (tetraethylene glycol dimethyl ether) electrolyte as a model system, we investigate the mechanism of the charge potential evolution during cycling. RuO2 is the most widely used catalyst for ORR and OER,[3, 10-12] and Li-O2 battery with RuO2/rGO and LiCF3SO3-TEGDME has been reported with good performance. We have revealed that the increase of charge potential is caused by the transition of discharge product form Li2O2 to Li2CO3. The accumulation of CO2 in the electrolyte has been proved as the origin of the transition of discharge product. A strategy is further suggested to inhibit the accumulation of CO2 in the electrolyte, which prevents significantly the charge potential increase and prolongs the reversible cycling life of Li-O2 battery.
2 Experimental Section 2.1 Preparation of RuO2/rGO cathode Graphene oxide (GO) was prepared with a modified Hummers’ method according to the literature.[13] 30 mg RuCl3 was mixed with 30 mL aqueous GO dispersion (0.8 mg mL−1). Then the mixture was incubated for 30 min and reacted hydrothermally at 180
for 12 h.[5] The precipitation was filtered, washed, freeze-dried and vacuum
dried at 80 Electron
. The as-prepared RuO2/rGO was characterized by Transmission
Microscope
(TEM),
Thermogravimetric
Analysis
(TGA),
Raman
spectroscopy and X-ray photoelectron spectroscopy (XPS). As indicated by TGA analysis in Figure 1b, the weight loss before 200 °C is attributed to the removal of absorbed water, and the rapid weight decrease beginning at 200 °C originates from the oxidative decomposition of rGO. From the residual weight at 800 °C, the content of RuO2 in RuO2/rGO is calculated to be 50%. From spectrum of RuO2/rGO collected on a Xplora instrument, the characteristic D band and G band of rGO respectively at 1350 cm-1 and 1601 cm-1 were clearly detected, which evidences the presence of rGO in the synthesized catalyst (Figure 1c). Figure 1d shows the XPS data for Ru 3d peaks, 4
from which RuO2 bonds have detected in the catalyst. The RuO2/rGO powder was dispersed in N-methyl-2-pyrrolidone (NMP) containing polyvinylidene fluoride (PVDF) binder with a weight ratio of 8:2. The obtained slurry was coated on a gas diffusion layer (Bcgdl-1400S-LD, Porous carbon foil) with a loading density of 0.8 ± 0.2 mgRuO2/rGO cm-2.
2.2 Preparation of RuO2/rGO cathode preloaded with CaO CaO was preloaded on a glassfiber separator (Whatman) that tailored into a ring by dropping the CaO slurry in NMP with PVDF as binder on it, followed by vacuum drying at 60 ℃. The inner diameter of the ring was fitted for the RuO2/rGO electrode and the outer diameter of the ring was matched with the cell as illustrated in Figure 6c.
2.3 Vacuum pumping treatment The vacuum pumping treatment (VPT) was conducted with two steps: (1) A chamber containing Li-O2 battery inside was pumped to 6 KPa and kept for 4 hours; (2) Pure O2 was then inflated into the chamber for ensuring the Li-O2 battery under O2 atmosphere for 4 hours.
2.4 Characterizations The cathode materials after discharge-charge cycling to be characterized were disassembled from the discharge-charged Li-O2 battery in glove box. XRD measurements were carried out on Rigaku IV XRD apparatus at a scan rate of 5° min−1. All the cathodes to be subjected to XRD analysis were protected by Kapton film. Scanning electron microscopic (SEM) images were captured on S-4800 electron microscope (Hitachi, Japan), and Fourier transform infrared spectroscopy (FTIR) were collected on NICOLET IN10 apparatus. The cathode materials used in SEM and FTIR characterizations were washed with dimethosyethane and then vacuum dried for 5
10 hours before measurements. Gas generation from Li-O2 battery in charge process was determined through an online electrochemical mass spectrometry (OEMS). The OEMS analysis was conducted under Ar2 flow at a rate of 3 mL min-1. Before each measurement, the Li-O2 battery has been subjected previously to different number of discharge-charge cycling and then discharged to 1000 mAh g-1 at a rate of 200 mA g-1, subsequently the gas evaluation in charge process was monitored by OEMS. 4 conditions were analyzed, i.e. the Li-O2 batteries after respectively 1st, 5th and 20th cycles, and the Li-O2 battery subjected to vacuum pumping treatment at the end of 20th charging.
3 Results and Discussion 3.1 Variation of charge potential in Discharge-charge processes during cycling The evolution of discharge-charge curves of a typical Li-O2 battery with RuO2/rGO catalyst is illustrated in Figure 2a. The charge potential changes
dramatically during cycling. Generally, the charge potential first decreases and then increases along with cycling. Taking the potential in discharge-charge curves at 150 mAh g−1 (V150) as an index, the variation of charge potential can be evaluated quantitatively, and divided into two stages (Figure 2b). In stage 1 (1st ~ 11th cycles), the V150 is gradually decreased from 3.55 V in 1st cycle to 3.36 V in 3rd cycle, and then kept almost stable till 11th cycle. Starting from 12th cycle, i.e. the stage 2, the V150 increased first rapidly till 20th cycle and then gradually to 3.83 V at 47th cycle. As depicted in Figure 2a, at 40th cycle, the charge potential increased sharply to 3.75 V at the capacity of 50 mAh g−1 and kept above 3.75V throughout the rest 950 mAh g−1. Previous reported Li-O2 batteries with MnO2, LaNiO3, RuO2/NiCo2O4, Co3O4 and Ru-FeCoN/rGO as cathode catalysts also showed a similar increasing tendency of charge potential.[3, 4, 6, 14, 15] Thus the charge potential increase is a common issue of Li-O2 batteries with different kinds of catalysts.
6
3.2 Discharge product transition during cycling In order to understand the variation of charge potential during cycling, the discharge products in the 1st, 5th and 20th cycle were analyzed. Figure 3a shows X-ray diffraction (XRD) pattern of cathodes discharged to 1000 mAh g-1 respectively at 1st, 5th and 20th cycle. The characteristic peaks of Li2O2, i.e. two distinct peaks at 32.88° and 34.97° identified as the (100) and (101) peaks of Li2O2 according to the standard XRD card of Li2O2 (PDF#74-01155), could be obviously observed at 1st cycle, while they are almost invisible at 5th and 20th cycles which may be due to the decrease in Li2O2. Meanwhile the characteristic peak of Li2CO3 at 36.96° identified as the (-311) peaks of Li2CO3 raised in 20th cycle. Since the intensity of XRD characteristic peaks can only represent the crystalline degree of Li2O2, Raman spectroscopic study was further conducted. As shown in Figure 3b, the characteristic band of Li2O2 at 806 cm-1 can be detected at all 1st, 5th and 20th cycles, while the signal at the 20th cycle was weakened corresponding to the decrease in Li2O2. By contrast, the characteristic IR bands of Li2CO3 around 1505 cm−1, 1430 cm−1 and 868 cm−1 in the FTIR spectra of the discharged cathode have been strengthened from 1st to 20th cycle (Figure 3c). The results suggest that the discharge product gradually turned from Li2O2 to Li2CO3 during cycling. Moreover, SEM observation, shown in Figure 4a-4c, evidenced that the Li2O2 toroid particles were clearly seen in 1st cycle, [1, 2] while they decreased in 5th cycle, and almost disappeared in 20th cycles. To further investigate the transition of discharge products, the gas released during charge process was examined by online electrochemical mass spectrometry (OEMS). As illustrated in Figure 4d-4f, O2 was always detected during the charge process, and could be attributed to the decomposition of Li2O2. However, CO2 was also detected in the released gas. At the 1st and 5th cycle, O2 was the main gas product, while CO2 only released at the end of charge progress. This result agrees well with SEM observations, and indicates that Li2O2 is the main discharge product in 1st ~ 5th cycles. The small amount of released CO2 at high potential is attributed to the by-product of electrolyte decomposition and the oxidation of carbon in cathode.[8, 9, 16] At 20th cycle, 7
however, the amount of CO2 became the major component in the released gas from the battery. Therefore, at stage 2 the CO2 must not be the by-product, instead, it becomes the main charge product. It has been reported that CO2 is the decomposition product of Li2CO3 and LiRCO3, which are generated by side reactions on cathode.[8, 9] Thus, the large amount of released CO2 suggests that Li2CO3 or LiRCO3 has become a dominated discharge product. All the data above revealed that, although at stage 1 the discharge product was mainly Li2O2, more Li2CO3 was formed and gradually became the principal discharge product along with continuously cycling at stage 2. It is known that the electrochemical oxidation of Li2CO3 (2Li2CO3
2CO2 +
O2 + 4Li+ + 4e-) occurs in above 3.5V in Li-O2 battery. The decomposition of Li2CO3 will start obviously at potentials higher than 3.5V in the charge process. Consequently, the increase of charge potential during cycling is originated from the transition of discharge product from Li2O2 to Li2CO3.
3.3 The origin of discharge product transition during cycling We further reveal the origin of discharge product transition from Li2O2 to Li2CO3 during cycling. Previous reports in literature [6,9] demonstrated that electrolyte decomposition and carbon oxidation in electrode during discharge-charge process could produce Li2CO3 (Scheme 1a). However, it cannot explain why Li2O2 disappears from the discharge product and is replaced by Li2CO3. It is known that, in the presence of both O2 and CO2, oxygen reduction reaction tends to be the first reaction step and a radical nucleophile reagent is generated, which can react vigorously with neutral or cationic species in the battery, such as CO2, Li+, and electrolytes.[8, 17-19] During discharge-charge cycling, CO2 is produced through the decomposition of Li2CO3 derived from side reaction as mentioned above (Scheme 1a). Moreover, the solubility of CO2 is 50 times larger than that of O2.[20] With CO2 accumulation during cycling, the amount of CO2 could become larger than that of O2 in electrolyte. Therefore, the reaction of radical nucleophile reagent with CO2 turn out to be the main reaction during discharge, resulting in Li2CO3 as the major discharge product 8
(Scheme 1b). The effect of CO2 accumulation was further verified by an additional experiment of vacuum pumping treatment (VPT). The accumulated CO2 in electrolyte can be removed by applying a VPT to the battery, i.e. after the 20th charge, the chamber containing the battery inside was pumped to 6 kPa and sealed for 4 h. The low pressure accelerated the escape of CO2 from electrolyte. Subsequently the chamber was swept by O2 flow so that O2 was refilled into the battery, which was then submitted to the 21st discharge-charge cycle. As expected, after the VPT, the charge curve recovered to the original shape similar to that of the 5th cycle, and the V150 decreased from 3.62 V at 20th cycle to 3.26 V at 21st cycle (Figure 5a). Besides, the toroid particles in the SEM image appeared after the VPT, evidencing clearly the regeneration of Li2O2 (Figure 5b). These results also indicate that the RuO2/rGO catalyst did not degrade, further confirming the increase of charge potential is not caused by catalyst fading. The OEMS analysis of the 21st cycle (Figure 5c) demonstrates that, after the VPT, O2 became again the primary gas released at the beginning of the charge process, while CO2 only appeared when charge potential grew higher than 3.75 V. Obviously, when the Li2CO3 was converted into CO2 and then removed, the discharge product is mainly Li2O2 that has a lower decomposition potential than that of the Li2CO3 (Scheme 1c). As a consequence, the charge potential is recovered and decreased to 3.26V. We also studied the CO2 accumulation by directly inflating CO2 into Li-O2 battery. After the 3rd charge process, CO2 was introduced and filled into the battery before the 4th discharge. As shown in Figure 5d, the charge potential in the 4th charge process increased significantly after inflating CO2 into the battery, and the V150 rose from 3.33V to 3.67V. The charge curve also became similar to that of the 40th cycle shown in Figure 2a. Thus, by inflating CO2, we have forced to yield Li2CO3 (2CO2 + O2 + 4Li+ + 4e-
2Li2CO3 ) as the discharge product and raised the charge potential. As
shown in Figure 6a, there were large number of little sheets dispersed on the cathode with some little particle above, instead of the classical disc (Figure 4a). And the signal 9
of Li2O2 at 806 cm-1 was weakened after CO2 inflation (Figure 6b). Meanwhile characteristic peak of Li2CO3 in FTIR spectrum significantly enhanced (Figure 6c). Also does the signal of characteristic peak of Li2CO3 in XRD spectrum, as shown in Figure 6d. Furthermore, by applying VPT to remove immediately CO2 after the 4th cycle, as indicated in Figure 5d, the charge curve was returned closing to the 3rd cycle and the charge potential was decreased to 3.26V. We could therefore conclude that, from the above results, the CO2 plays the key role on the charge potential variation. When the amount of CO2 has surpassed O2 in electrolyte, Li2CO3 becomes the main discharge product, and its high decomposition potential results in rising the charge potential and reducing the round-trip efficiency of Li-O2 battery.
3.4 Strategy to impede performance fading of Li-O2 battery by inhibiting CO2 accumulation Since the accumulation of CO2 is the key step leading to high charge potential of Li-O2 battery, it may expect that the high charge potential can be prevented if the CO2 accumulation could be suppressed. As inspired by results presented in Figure 5a, VPT was applied regularly every 20 cycles of discharge-charge to remove the accumulated CO2. As illustrated in Figure 7a, by applying VPT to Li-O2 battery every 20 cycles of discharge-charge, V150 decreased regularly. Meanwhile, the round efficiency increased, as shown in Figure 7b, i.e. from 72% to 76% after the 1st VPT. And the cycle life was extended from 47 to 65 cycles, i.e. an extension to 138%. Based on the above analysis, we proposed a strategy to suppress the CO2 accumulation during discharge-charge of Li-O2 battery. A CO2 absorbent agent (CaO) is preloaded into Li-O2 battery at its assemblage. CaO was used as commercial CO2 absorbent for the purpose of air purification, because it can react with CO2 to generate CaCO3. The chemical reaction of CO2 with CaO can be written as follows: CaO + CO2 == CaCO3 with ∆G = -130.4 kJ mol-1, which means that the reaction can take place spontaneously. Furthermore, CaO is electrochemical inertness so that it will not affect the electrochemical process of Li-O2 battery. The CaO has also advantages of 10
inexpensive and non-poisonous. CaO was selected therefore as the CO2 absorbent agent in the Li-O2 battery. As illustrated in Figure 7c, CaO was preloaded onto a ring surrounding the cathode catalyst of RuO2/rGO in Li-O2 battery. After preloading CaO, discharge-charge curve of the Li-O2 battery is relatively stable with tiny round efficiency decrease along with cycling till 20th cycles (Figure 7d). As demonstrated by Figure 7e, the stage 1 is extended to 21st cycles with the CaO-preloaded battery. The cycle life of CaO-preloaded battery has been prolonged from 47 cycles of a battery without preloading CaO to 70 cycles, which represents an extension to 148%. The significant improvement is attributed to the effective capture of CO2 by CaO, which increases the proportion of Li2O2 in the discharge product. OEMS results (Figure 7f) demonstrate that at the 20th cycle, the proportion of O2 during charge process in the released gas from the CaO-preloaded battery is 21%, much higher than that of the battery without preloading CaO (16%, Figure 4f). To prove the absorption of CO2 by CaO, XRD characterization of CaO before Li-O2 battery assembling and after 20 discharge-charge cycles was conducted. As shown in Figure 8a, signal peak at 29.47°of CaCO3 appears clearly after discharge-charge cycling. To further prove the absorption of CO2, the loop loaded with CaO after 20 discharge-charge cycles was sealed in a mass spectrometry bottle and 0.08 ml HCl (0.1 mol L-1) was injected into the bottle and then detected by online mass spectrometry (OMS). As shown in Figure 8b, the signal of CO2 after cycling has greatly increased, evidencing obviously the absorption of CO2 during cycling. The above results confirmed that the CaO is effective in capturing CO2 and consequently enhances the cycle stability of Li-O2 battery.
4 Conclusions In this study, we reveal that the increase in charge potential of a Li-O2 battery is due to the transition of the main discharge product from Li2O2 to Li2CO3 that oxidizes at higher potentials. Such a product transition is caused by the accumulation of CO2 in 11
electrolyte during discharge-charge processes. CO2 is initially generated in charge process at high potentials as a by-product of side reaction on the cathode of Li-O2 battery, and it can convert the discharge product Li2O2 into Li2CO3. The formation of Li2CO3 increases the charge potential of the Li-O2 battery, and further promotes the generation of CO2. Such a positive feedback process leads to the accumulation of CO2 in electrolyte, and finally, the main discharge product becomes Li2CO3. Based on a thorough understanding of the mechanism, we proposed a strategy to suppress the charge potential increase, i.e. preloading a CO2 absorbent agent (CaO) into Li-O2 battery at its assemblage. The results confirmed that the approach of preloading a CO2 absorbent agent can effectively inhibit the charge potential increase and prolong significantly the cycle lifespan of the Li-O2 battery. The study shed light on the mechanism of performance fading of a Li-O2 battery, and established a strategy for further design of long-cycle-lifetime Li-O2 batteries.
Acknowledgments This
work
was
supported
by
NSFC
(21621091)
and
the
MOST
(2016YFB0100202)
Conflict of Interest The authors declare no competing interests.
Keywords: Li-O2 Battery; Performance fading; Discharge product transition; Vacuum pumping treatment; CO2 absorbent
12
References: [1] J. Lu, L. Li, J.-B. Park, Y.-K. Sun, F. Wu and K. Amine. Chem. Rev. 114 (2014) 5611-5640. [2] A. C. Luntz and B. D. McCloskey. Chem. Rev. 114 (2014) 11721-11750. [3] Y.-F. Xu, Y. Chen, G.-L. Xu, X.-R. Zhang, Z. Chen, J.-T. Li, L. Huang, K. Amine and S.-G. Sun. Nano Energy 28 (2016) 63-70. [4] R. Black, J.-H. Lee, B. Adams, C. A. Mims and L. F. Nazar. Angew. Chemie 125 ( 2013) 410-414. [5] H.-G. Jung, Y. S. Jeong, J.-B. Park, Y.-K. Sun, B. Scrosati and Y. J. Lee. ACS Nano. 7 (2013) 3532-3539. [6] X. Zeng, C. You, L. Leng, D. Dang, X. Qiao, X. Li, Y. Li, S. Liao and R. R. Adzic. J. Mater. Chem. A 3 (2015) 11224-11231. [7] X. Guo, P. Liu, J. Han, Y. Ito, A. Hirata, T. Fujita and M. Chen. Adv. Mater. 27 (2015) 6137-6143. [8] S. A. Freunberger, C. Yuhui, N. E. Drewett, L. J. Hardwick, B. Fanny and P. G. Bruce Angew. Chem. Int. Ed. 50 (2011) 8609-8613. [9] M. M. Ottakam Thotiyl, S. A. Freunberger, Z. Peng and P. G. Bruce. J. Am. Chem. Soc. 135 (2013) 494-500. [10] J. Xie, X. Yao, I. P. Madden, D.-E. Jiang, L.-Y. Chou, C.-K. Tsung and D. Wang. J. Am. Chem. Soc. 136 (2014) 8903-8906. [11] J.-J. Xu, Z.-W. Chang, Y. Wang, D.-P. Liu, Y. Zhang and X.-B. Zhang. Adv. Mater. 28 (2016) 9620-9628. [12] Y. Chang, S. Dong, Y. Ju, D. Xiao, X. Zhou, L. Zhang, X. Chen, C. Shang, L. Gu, Z. Peng and G. Cui. Adv. Sci. 2 (2015) 1500092. [13] K. Chen, L. Chen, Y. Chen, H. Bai and L. Li. J. Mater. Chem. 22 (2012) 20968-20976. [14] J. Zhang, Y. Zhao, X. Zhao, Z. Liu and W. Chen, Sci. Rep. 4 (2014) 6005. [15] L. Zou, Y. Jiang, J. Cheng, Y. Chen, B. Chi, J. Pu and L. Jian. Electrochim. Acta 262 (2018) 97-106.
13
[16] B. D. McCloskey, A. Speidel, R. Scheffler, D. C. Miller, V. Viswanathan, J. S. Hummelshøj, J. K. Nørskov and A. C. Luntz. J. Phys. Chem. Lett. 3 (2012) 997-1001. [17] H.-K. Lim, H.-D. Lim, K.-Y. Park, D.-H. Seo, H. Gwon, J. Hong, W. A. Goddard, H. Kim and K. Kang. J. Am. Chem. Soc. 135 (2013) 9733-9742. [18] S. A. Freunberger, C. Yuhui, P. Zhangquan, J. M. Griffin, L. J. Hardwick, B. Fanny, N. Petr and P. G. Bruce. J. Am. Chem. Soc. 133 (2011) 8040-8047. [19] S. R. Gowda, A. Brunet, G. M. Wallraff and B. D. McCloskey. J. Phys. Chem. Lett. 4 (2013) 276-279. [20] J. D. Wadhawan, P. J. Welford, E. Maisonhaute, V. Climent, N. S. Lawrence, R. G. Compton, H. B. McPeak and C. E. W. Hahn. J. Phys. Chem. Lett. 105 (2001) 10659-10668.
14
Figures:
Figure 1. (a) TEM image, (b) TGA, (c) Raman spectra and (d) XPS result of RuO2/GO catalyst.
15
Figure 2. (a) Discharge-charge curves of the Li-O2 battery running for 40 cycles to 1000 mAh g−1 at a rate of 200 mA g−1. (b) Plot of V150 value versus cycle number.
16
Figure 3. (a) XRD pattern and (b) Raman spectra and (c) FTIR spectra of 1st, 5th and 20th cycles of the cathodes discharged to 1000 mAh g−1.
17
Figure 4. Change of discharge product on cathode. (a) ~ (c): SEM images of the cathode discharged to 1000 mAh g−1 at 1st (a), 5th (b) and 20th (c) cycle. (d) ~ (f): OEMS analysis of the charge process of Li-O2 battery after 1st (d), 5th (e) and 20th (f) discharge-charge cycle, dash line was the theoretical line for oxygen.
18
Scheme 1. Schematic illustration of reactions on the cathode of Li-O2 battery during discharge-charge processes. (a) ORR and OER in an ideal battery; (b) CO2 accumulation and the consequent reaction change; (c) The discharge-charge process after VPT.
19
Figure 5. Effects of CO2 on charge process. (a) Discharge-charge curves of a Li-O2 battery at a rate of 200 mA g−1 at the 5th, 20th and 21st cycle. The battery was subjected to VPT before the 21st discharge-charge cycle. (b) SEM image of the cathode after the 21st cycle discharge. (c) OEMS analysis of gas released from Li-O2 battery at the 21st cycle charge process. (d) The discharge-charge curves of Li-O2 battery at a rate of 200 mA g−1, subsequent to the 3rd cycle CO2 was inflated into the battery before the 4th discharge (O2 : CO2 ≈ 1:1), and after the 4th cycle the battery was subjected to VPT.
20
Figure 6. (a) SEM image of cathode discharge to 1000 mAh g−1 at 4th cycle (After CO2 inflation). (b) Raman spectra, (c) FTIR spectra and (d) XRD of cathode discharge to 1000 mAh g-1 respectively at 1st and 4th (After CO2 inflation).
21
Figure 7. (a) Plot of V150 value versus cycle number of Li-O2 batteries without VPT and subjected to VPT after 19th, 39th, and 59th cycle, respectively. (b) The discharge-charge curves at the 19th, 20th, 39th, 40th, 59th, 60th, 65th cycle at a rate of 200 mA g−1 of the Li-O2 battery subjected to VPT after 19th, 39th, and 59th cycle, respectively. (c) Schematic presentation of Li-O2 battery preloaded CaO. (d) Discharge-charge curves of 10th, 20th, 40th, 60th, 70th cycle for the Li-O2 battery preloaded CaO at a rate of 200 mA g−1. (e) Plot of V150 versus cycle number of Li-O2 batteries without CaO and preloaded CaO, respectively. (f) OEMS analysis of gas released in 20st charge process of Li-O2 battery preloaded CaO.
22
Figure 8. (a) XRD and (b) CO2 signal detected by OMS of loop loaded CaO before and after cycling.
23
Li-Bin Chen received her M.S. degree from Xiamen University in 2014, and is pursuing his Ph.D. degree in Xiamen University. Her current research interest focuses mainly on lithium oxygen batteries.
Yu-Hao Hong received his B.S. degree from the Department of Chemistry, Xiamen University in 2014, and is pursuing his Ph.D. degree in Xiamen University. His current research interest focuses mainly on online electrochemical mass spectroscopy.
Liang-Ping Xiao received his M.S. degree from Xiamen University in 2017, and is pursuing his Ph.D. degree in Xiamen University. His current research interest focuses mainly on in-situ liquid-cell TEM.
Jin-Hai You received his B.S. degree from Fujian Agriculture and Forestry University
in 2017. He is pursuing his M.S. degree in Xiamen University, and his current research interest is on lithium metal anodes for Li metal batteries.
Wen-Jia Sheng received her B.Eng. degree from Tianjin University in 2016. She is a master degree candidate of Xiamen University and her research interest is on the cathode catalyst of non-aqueous lithium-oxygen batteries.
Hua Bai received his B. S. and Ph. D. degree in Department of Chemistry, Tsinghua University in 2004 and 2009, respectively. From 2009 to 2011, he worked in Department of Chemical Engineering, Tsinghua University, as a postdoctoral researcher. He then join Xiamen University, as an associate professor in College of Materials. His research focuses on energy conversion and storage, including developing new materials for supercapacitors, lithium ion batteries and solar cells.
Ling Huang received his Ph.D. degree from the Department of Chemistry of Xiamen University in 1997. He has worked in Xiamen University from 2000 to date. He is now a professor of chemistry at Xiamen University, and his research interests include synthesis of tin based alloys by electrochemical deposition and high-energy cathode materials of lithium ion batteries.
Shi-Gang Sun is a professor of chemistry at Xiamen University. He received his Bachelor's degree in 1982 from Xiamen University and Doctor at d′Etat in 1986 from Universit´e Pierre et Marie Curie (Paris VI), and carried out one year of postdoctoral research at the Laboratoire d′Electrochimie Interfaciale du CNRS, France. He is a member of Chinese Academy of Sciences, Fellow of the Royal Society of Chemistry, U.K., and Fellow of the International Society of Electrochemistry. His research interests include electrocatalysis, electrochemical surface science, spectroelectrochemistry, and electrochemical energy conversion and storage.
Research highlights： ： •
Effects of CO2 accumulation during cycling of a Li-O2 battery employing RuO2/rGO cathode is studied.
•
The charge potential at capacity of 150 mAh g-1 is taken as an indice to track the health state and evolution of a Li-O2 battery during cycling.
•
The transition of discharge product due to CO2 accumulation during cycling of a Li-O2 battery is revealed as the origin of performance fading.
•
A strategy to preload CO absorbent (CaO) has inhibited effectively CO2 accumulation and prolonged the cycle lifespan of Li-O2 battery.
1
S2211-2855(19)30878-X
DOI:
https://doi.org/10.1016/j.nanoen.2019.104171
Reference:
NANOEN 104171
To appear in:
Nano Energy
Received Date: 11 June 2019 Revised Date:
29 September 2019
Accepted Date: 4 October 2019
Please cite this article as: L.-B. Chen, Y.-H. Hong, L.-P. Xiao, J.-H. You, W.-J. Sheng, L. Huang, H. Bai, S.-G. Sun, Effects of CO2 accumulation during cycling of a Li–O2 battery on the transition of discharge product and performance fading, Nano Energy (2019), doi: https://doi.org/10.1016/ j.nanoen.2019.104171. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.
TOC Graphic: A thorough investigation on the effects of CO2 accumulation during cycling of a Li-O2 battery employing RuO2/rGO cathode leads to realize a strategy to preload CO absorbent (CaO) into the battery, which has prolonged significantly the cycle lifespan.
1
Effects of CO2 Accumulation During Cycling of a Li-O2 Battery on the Transition of Discharge Product and Performance Fading
Li-Bin Chen a, Yu-Hao Hong a , Liang-Ping Xiao a, Jin-Hai You c, Wen-Jia Sheng a, Ling Huang a, Hua Bai b, Shi-Gang Sun*a a
State Key Laboratory of Physical Chemistry of Solid Surfaces, Department of Chemistry,
College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, 361005, P. R. China. b
College of Materials, Xiamen University, Xiamen 361005, P. R. China.
c
College of Energy, Xiamen University, Xiamen 361005, P. R. China.
*
Corresponding author, E-mail: [email protected]
Abstract The increase in charge potential during discharge-charge cycling reduces severely the cycle life of a Li-O2 battery, but its origin has not been fully understood yet. The current study focuses on revealing the intrinsic basis behind the increase of charge potential of a Li-O2 battery and developing a strategy to inhibit this phenomenon. Based on results of X-ray diffraction, Fourier transform infrared spectroscopy, scanning electron microscope and online electrochemical mass spectroscopy, we find that the performance fading of a Li-O2 battery evidenced by the increase in charge potential is caused by the discharge product transition from Li2O2 to Li2CO3 due to the accumulation of by-product CO2 in cathode side reaction during cycling. We further demonstrate that the performance of a cycled Li-O2 battery can be completely 1
recovered when the accumulated CO2 is evacuated by a vacuum pumping treatment. A strategy is therefore proposed to suppress the CO2 accumulation by preloading a CO2 absorbent agent (CaO) into Li-O2 battery at its assemblage. As a result, the phenomenon of charge potential increase has been effectively inhibited, which improved significantly the cycleability. In comparison with a Li-O2 battery without preloading CaO, the cycling life of the CaO preloaded Li-O2 battery has been prolonged to 148%. This study reveals an important mechanism of performance fading of Li-O2 battery and develops an efficient approach to increase the cycle life, which has thrown new insight into the design and construction of Li-O2 batteries with long lifespan.
2
1 Introduction With the increasing demands on energy density by moveable electronic devices and electric vehicles, a great number of studies have focused on energy storage systems with higher energy density. Li-O2 battery is considered as the next generation of energy storage device owing to its high theoretical specific energy density.[1, 2] A nonaqueous Li-O2 battery consists of lithium anode, separator, oxygen cathode and nonaqueous electrolyte. In an ideal discharge process, the oxygen cathode undergoes oxygen reduction reaction (ORR), producing Li2O2, which is then oxidized into O2 via oxygen evolution reaction (OER) in the charge process. The discharge product Li2O2, however, is an electrically non-conductive compound and has slow oxidation kinetics. As a consequence, a large overpotential on the cathode is required to charge the Li-O2 battery. High charge potential will instigate side-reactions and accelerate electrolyte decomposition, resulting in poor round-trip efficiency and passivation of the cathode. Therefore, various electrocatalysts, such as manganese dioxide, cobalt oxide, ruthenium dioxide, have been applied to promote the ORR and OER, and they have efficiently decreased the charge overpotential.[3-6] However, even with appropriate electrocatalysts, the Li-O2 batteries usually display poor cycling stability. The charge potential of Li-O2 batteries generally increases as the discharge-charge cycling going on, which represents the performance fading of Li-O2 battery. Up to now, the charge potential increase has not been understood completely, because many factors can impact on the performance of Li-O2 battery.[1, 2] Some studies proposed that the charge potential increase may be caused by the passivation of the catalyst due to its surface covered by undecomposed residues.[7] Other investigations attributed it to side reactions of electrolyte and oxidation of carbon in electrode.[8, 9] Since the fast performance fading is one of the fatal disadvantages of Li-O2 battery for practical application, it is crucial to reveal the mechanism of charge potential increase during discharge-charge cycling and develop strategy to suppress it. 3
In this study, by taking Li-O2 battery with RuO2/rGO (reduced graphene oxide) cathode catalyst and LiCF3SO3-TEGDME (tetraethylene glycol dimethyl ether) electrolyte as a model system, we investigate the mechanism of the charge potential evolution during cycling. RuO2 is the most widely used catalyst for ORR and OER,[3, 10-12] and Li-O2 battery with RuO2/rGO and LiCF3SO3-TEGDME has been reported with good performance. We have revealed that the increase of charge potential is caused by the transition of discharge product form Li2O2 to Li2CO3. The accumulation of CO2 in the electrolyte has been proved as the origin of the transition of discharge product. A strategy is further suggested to inhibit the accumulation of CO2 in the electrolyte, which prevents significantly the charge potential increase and prolongs the reversible cycling life of Li-O2 battery.
2 Experimental Section 2.1 Preparation of RuO2/rGO cathode Graphene oxide (GO) was prepared with a modified Hummers’ method according to the literature.[13] 30 mg RuCl3 was mixed with 30 mL aqueous GO dispersion (0.8 mg mL−1). Then the mixture was incubated for 30 min and reacted hydrothermally at 180
for 12 h.[5] The precipitation was filtered, washed, freeze-dried and vacuum
dried at 80 Electron
. The as-prepared RuO2/rGO was characterized by Transmission
Microscope
(TEM),
Thermogravimetric
Analysis
(TGA),
Raman
spectroscopy and X-ray photoelectron spectroscopy (XPS). As indicated by TGA analysis in Figure 1b, the weight loss before 200 °C is attributed to the removal of absorbed water, and the rapid weight decrease beginning at 200 °C originates from the oxidative decomposition of rGO. From the residual weight at 800 °C, the content of RuO2 in RuO2/rGO is calculated to be 50%. From spectrum of RuO2/rGO collected on a Xplora instrument, the characteristic D band and G band of rGO respectively at 1350 cm-1 and 1601 cm-1 were clearly detected, which evidences the presence of rGO in the synthesized catalyst (Figure 1c). Figure 1d shows the XPS data for Ru 3d peaks, 4
from which RuO2 bonds have detected in the catalyst. The RuO2/rGO powder was dispersed in N-methyl-2-pyrrolidone (NMP) containing polyvinylidene fluoride (PVDF) binder with a weight ratio of 8:2. The obtained slurry was coated on a gas diffusion layer (Bcgdl-1400S-LD, Porous carbon foil) with a loading density of 0.8 ± 0.2 mgRuO2/rGO cm-2.
2.2 Preparation of RuO2/rGO cathode preloaded with CaO CaO was preloaded on a glassfiber separator (Whatman) that tailored into a ring by dropping the CaO slurry in NMP with PVDF as binder on it, followed by vacuum drying at 60 ℃. The inner diameter of the ring was fitted for the RuO2/rGO electrode and the outer diameter of the ring was matched with the cell as illustrated in Figure 6c.
2.3 Vacuum pumping treatment The vacuum pumping treatment (VPT) was conducted with two steps: (1) A chamber containing Li-O2 battery inside was pumped to 6 KPa and kept for 4 hours; (2) Pure O2 was then inflated into the chamber for ensuring the Li-O2 battery under O2 atmosphere for 4 hours.
2.4 Characterizations The cathode materials after discharge-charge cycling to be characterized were disassembled from the discharge-charged Li-O2 battery in glove box. XRD measurements were carried out on Rigaku IV XRD apparatus at a scan rate of 5° min−1. All the cathodes to be subjected to XRD analysis were protected by Kapton film. Scanning electron microscopic (SEM) images were captured on S-4800 electron microscope (Hitachi, Japan), and Fourier transform infrared spectroscopy (FTIR) were collected on NICOLET IN10 apparatus. The cathode materials used in SEM and FTIR characterizations were washed with dimethosyethane and then vacuum dried for 5
10 hours before measurements. Gas generation from Li-O2 battery in charge process was determined through an online electrochemical mass spectrometry (OEMS). The OEMS analysis was conducted under Ar2 flow at a rate of 3 mL min-1. Before each measurement, the Li-O2 battery has been subjected previously to different number of discharge-charge cycling and then discharged to 1000 mAh g-1 at a rate of 200 mA g-1, subsequently the gas evaluation in charge process was monitored by OEMS. 4 conditions were analyzed, i.e. the Li-O2 batteries after respectively 1st, 5th and 20th cycles, and the Li-O2 battery subjected to vacuum pumping treatment at the end of 20th charging.
3 Results and Discussion 3.1 Variation of charge potential in Discharge-charge processes during cycling The evolution of discharge-charge curves of a typical Li-O2 battery with RuO2/rGO catalyst is illustrated in Figure 2a. The charge potential changes
dramatically during cycling. Generally, the charge potential first decreases and then increases along with cycling. Taking the potential in discharge-charge curves at 150 mAh g−1 (V150) as an index, the variation of charge potential can be evaluated quantitatively, and divided into two stages (Figure 2b). In stage 1 (1st ~ 11th cycles), the V150 is gradually decreased from 3.55 V in 1st cycle to 3.36 V in 3rd cycle, and then kept almost stable till 11th cycle. Starting from 12th cycle, i.e. the stage 2, the V150 increased first rapidly till 20th cycle and then gradually to 3.83 V at 47th cycle. As depicted in Figure 2a, at 40th cycle, the charge potential increased sharply to 3.75 V at the capacity of 50 mAh g−1 and kept above 3.75V throughout the rest 950 mAh g−1. Previous reported Li-O2 batteries with MnO2, LaNiO3, RuO2/NiCo2O4, Co3O4 and Ru-FeCoN/rGO as cathode catalysts also showed a similar increasing tendency of charge potential.[3, 4, 6, 14, 15] Thus the charge potential increase is a common issue of Li-O2 batteries with different kinds of catalysts.
6
3.2 Discharge product transition during cycling In order to understand the variation of charge potential during cycling, the discharge products in the 1st, 5th and 20th cycle were analyzed. Figure 3a shows X-ray diffraction (XRD) pattern of cathodes discharged to 1000 mAh g-1 respectively at 1st, 5th and 20th cycle. The characteristic peaks of Li2O2, i.e. two distinct peaks at 32.88° and 34.97° identified as the (100) and (101) peaks of Li2O2 according to the standard XRD card of Li2O2 (PDF#74-01155), could be obviously observed at 1st cycle, while they are almost invisible at 5th and 20th cycles which may be due to the decrease in Li2O2. Meanwhile the characteristic peak of Li2CO3 at 36.96° identified as the (-311) peaks of Li2CO3 raised in 20th cycle. Since the intensity of XRD characteristic peaks can only represent the crystalline degree of Li2O2, Raman spectroscopic study was further conducted. As shown in Figure 3b, the characteristic band of Li2O2 at 806 cm-1 can be detected at all 1st, 5th and 20th cycles, while the signal at the 20th cycle was weakened corresponding to the decrease in Li2O2. By contrast, the characteristic IR bands of Li2CO3 around 1505 cm−1, 1430 cm−1 and 868 cm−1 in the FTIR spectra of the discharged cathode have been strengthened from 1st to 20th cycle (Figure 3c). The results suggest that the discharge product gradually turned from Li2O2 to Li2CO3 during cycling. Moreover, SEM observation, shown in Figure 4a-4c, evidenced that the Li2O2 toroid particles were clearly seen in 1st cycle, [1, 2] while they decreased in 5th cycle, and almost disappeared in 20th cycles. To further investigate the transition of discharge products, the gas released during charge process was examined by online electrochemical mass spectrometry (OEMS). As illustrated in Figure 4d-4f, O2 was always detected during the charge process, and could be attributed to the decomposition of Li2O2. However, CO2 was also detected in the released gas. At the 1st and 5th cycle, O2 was the main gas product, while CO2 only released at the end of charge progress. This result agrees well with SEM observations, and indicates that Li2O2 is the main discharge product in 1st ~ 5th cycles. The small amount of released CO2 at high potential is attributed to the by-product of electrolyte decomposition and the oxidation of carbon in cathode.[8, 9, 16] At 20th cycle, 7
however, the amount of CO2 became the major component in the released gas from the battery. Therefore, at stage 2 the CO2 must not be the by-product, instead, it becomes the main charge product. It has been reported that CO2 is the decomposition product of Li2CO3 and LiRCO3, which are generated by side reactions on cathode.[8, 9] Thus, the large amount of released CO2 suggests that Li2CO3 or LiRCO3 has become a dominated discharge product. All the data above revealed that, although at stage 1 the discharge product was mainly Li2O2, more Li2CO3 was formed and gradually became the principal discharge product along with continuously cycling at stage 2. It is known that the electrochemical oxidation of Li2CO3 (2Li2CO3
2CO2 +
O2 + 4Li+ + 4e-) occurs in above 3.5V in Li-O2 battery. The decomposition of Li2CO3 will start obviously at potentials higher than 3.5V in the charge process. Consequently, the increase of charge potential during cycling is originated from the transition of discharge product from Li2O2 to Li2CO3.
3.3 The origin of discharge product transition during cycling We further reveal the origin of discharge product transition from Li2O2 to Li2CO3 during cycling. Previous reports in literature [6,9] demonstrated that electrolyte decomposition and carbon oxidation in electrode during discharge-charge process could produce Li2CO3 (Scheme 1a). However, it cannot explain why Li2O2 disappears from the discharge product and is replaced by Li2CO3. It is known that, in the presence of both O2 and CO2, oxygen reduction reaction tends to be the first reaction step and a radical nucleophile reagent is generated, which can react vigorously with neutral or cationic species in the battery, such as CO2, Li+, and electrolytes.[8, 17-19] During discharge-charge cycling, CO2 is produced through the decomposition of Li2CO3 derived from side reaction as mentioned above (Scheme 1a). Moreover, the solubility of CO2 is 50 times larger than that of O2.[20] With CO2 accumulation during cycling, the amount of CO2 could become larger than that of O2 in electrolyte. Therefore, the reaction of radical nucleophile reagent with CO2 turn out to be the main reaction during discharge, resulting in Li2CO3 as the major discharge product 8
(Scheme 1b). The effect of CO2 accumulation was further verified by an additional experiment of vacuum pumping treatment (VPT). The accumulated CO2 in electrolyte can be removed by applying a VPT to the battery, i.e. after the 20th charge, the chamber containing the battery inside was pumped to 6 kPa and sealed for 4 h. The low pressure accelerated the escape of CO2 from electrolyte. Subsequently the chamber was swept by O2 flow so that O2 was refilled into the battery, which was then submitted to the 21st discharge-charge cycle. As expected, after the VPT, the charge curve recovered to the original shape similar to that of the 5th cycle, and the V150 decreased from 3.62 V at 20th cycle to 3.26 V at 21st cycle (Figure 5a). Besides, the toroid particles in the SEM image appeared after the VPT, evidencing clearly the regeneration of Li2O2 (Figure 5b). These results also indicate that the RuO2/rGO catalyst did not degrade, further confirming the increase of charge potential is not caused by catalyst fading. The OEMS analysis of the 21st cycle (Figure 5c) demonstrates that, after the VPT, O2 became again the primary gas released at the beginning of the charge process, while CO2 only appeared when charge potential grew higher than 3.75 V. Obviously, when the Li2CO3 was converted into CO2 and then removed, the discharge product is mainly Li2O2 that has a lower decomposition potential than that of the Li2CO3 (Scheme 1c). As a consequence, the charge potential is recovered and decreased to 3.26V. We also studied the CO2 accumulation by directly inflating CO2 into Li-O2 battery. After the 3rd charge process, CO2 was introduced and filled into the battery before the 4th discharge. As shown in Figure 5d, the charge potential in the 4th charge process increased significantly after inflating CO2 into the battery, and the V150 rose from 3.33V to 3.67V. The charge curve also became similar to that of the 40th cycle shown in Figure 2a. Thus, by inflating CO2, we have forced to yield Li2CO3 (2CO2 + O2 + 4Li+ + 4e-
2Li2CO3 ) as the discharge product and raised the charge potential. As
shown in Figure 6a, there were large number of little sheets dispersed on the cathode with some little particle above, instead of the classical disc (Figure 4a). And the signal 9
of Li2O2 at 806 cm-1 was weakened after CO2 inflation (Figure 6b). Meanwhile characteristic peak of Li2CO3 in FTIR spectrum significantly enhanced (Figure 6c). Also does the signal of characteristic peak of Li2CO3 in XRD spectrum, as shown in Figure 6d. Furthermore, by applying VPT to remove immediately CO2 after the 4th cycle, as indicated in Figure 5d, the charge curve was returned closing to the 3rd cycle and the charge potential was decreased to 3.26V. We could therefore conclude that, from the above results, the CO2 plays the key role on the charge potential variation. When the amount of CO2 has surpassed O2 in electrolyte, Li2CO3 becomes the main discharge product, and its high decomposition potential results in rising the charge potential and reducing the round-trip efficiency of Li-O2 battery.
3.4 Strategy to impede performance fading of Li-O2 battery by inhibiting CO2 accumulation Since the accumulation of CO2 is the key step leading to high charge potential of Li-O2 battery, it may expect that the high charge potential can be prevented if the CO2 accumulation could be suppressed. As inspired by results presented in Figure 5a, VPT was applied regularly every 20 cycles of discharge-charge to remove the accumulated CO2. As illustrated in Figure 7a, by applying VPT to Li-O2 battery every 20 cycles of discharge-charge, V150 decreased regularly. Meanwhile, the round efficiency increased, as shown in Figure 7b, i.e. from 72% to 76% after the 1st VPT. And the cycle life was extended from 47 to 65 cycles, i.e. an extension to 138%. Based on the above analysis, we proposed a strategy to suppress the CO2 accumulation during discharge-charge of Li-O2 battery. A CO2 absorbent agent (CaO) is preloaded into Li-O2 battery at its assemblage. CaO was used as commercial CO2 absorbent for the purpose of air purification, because it can react with CO2 to generate CaCO3. The chemical reaction of CO2 with CaO can be written as follows: CaO + CO2 == CaCO3 with ∆G = -130.4 kJ mol-1, which means that the reaction can take place spontaneously. Furthermore, CaO is electrochemical inertness so that it will not affect the electrochemical process of Li-O2 battery. The CaO has also advantages of 10
inexpensive and non-poisonous. CaO was selected therefore as the CO2 absorbent agent in the Li-O2 battery. As illustrated in Figure 7c, CaO was preloaded onto a ring surrounding the cathode catalyst of RuO2/rGO in Li-O2 battery. After preloading CaO, discharge-charge curve of the Li-O2 battery is relatively stable with tiny round efficiency decrease along with cycling till 20th cycles (Figure 7d). As demonstrated by Figure 7e, the stage 1 is extended to 21st cycles with the CaO-preloaded battery. The cycle life of CaO-preloaded battery has been prolonged from 47 cycles of a battery without preloading CaO to 70 cycles, which represents an extension to 148%. The significant improvement is attributed to the effective capture of CO2 by CaO, which increases the proportion of Li2O2 in the discharge product. OEMS results (Figure 7f) demonstrate that at the 20th cycle, the proportion of O2 during charge process in the released gas from the CaO-preloaded battery is 21%, much higher than that of the battery without preloading CaO (16%, Figure 4f). To prove the absorption of CO2 by CaO, XRD characterization of CaO before Li-O2 battery assembling and after 20 discharge-charge cycles was conducted. As shown in Figure 8a, signal peak at 29.47°of CaCO3 appears clearly after discharge-charge cycling. To further prove the absorption of CO2, the loop loaded with CaO after 20 discharge-charge cycles was sealed in a mass spectrometry bottle and 0.08 ml HCl (0.1 mol L-1) was injected into the bottle and then detected by online mass spectrometry (OMS). As shown in Figure 8b, the signal of CO2 after cycling has greatly increased, evidencing obviously the absorption of CO2 during cycling. The above results confirmed that the CaO is effective in capturing CO2 and consequently enhances the cycle stability of Li-O2 battery.
4 Conclusions In this study, we reveal that the increase in charge potential of a Li-O2 battery is due to the transition of the main discharge product from Li2O2 to Li2CO3 that oxidizes at higher potentials. Such a product transition is caused by the accumulation of CO2 in 11
electrolyte during discharge-charge processes. CO2 is initially generated in charge process at high potentials as a by-product of side reaction on the cathode of Li-O2 battery, and it can convert the discharge product Li2O2 into Li2CO3. The formation of Li2CO3 increases the charge potential of the Li-O2 battery, and further promotes the generation of CO2. Such a positive feedback process leads to the accumulation of CO2 in electrolyte, and finally, the main discharge product becomes Li2CO3. Based on a thorough understanding of the mechanism, we proposed a strategy to suppress the charge potential increase, i.e. preloading a CO2 absorbent agent (CaO) into Li-O2 battery at its assemblage. The results confirmed that the approach of preloading a CO2 absorbent agent can effectively inhibit the charge potential increase and prolong significantly the cycle lifespan of the Li-O2 battery. The study shed light on the mechanism of performance fading of a Li-O2 battery, and established a strategy for further design of long-cycle-lifetime Li-O2 batteries.
Acknowledgments This
work
was
supported
by
NSFC
(21621091)
and
the
MOST
(2016YFB0100202)
Conflict of Interest The authors declare no competing interests.
Keywords: Li-O2 Battery; Performance fading; Discharge product transition; Vacuum pumping treatment; CO2 absorbent
12
References: [1] J. Lu, L. Li, J.-B. Park, Y.-K. Sun, F. Wu and K. Amine. Chem. Rev. 114 (2014) 5611-5640. [2] A. C. Luntz and B. D. McCloskey. Chem. Rev. 114 (2014) 11721-11750. [3] Y.-F. Xu, Y. Chen, G.-L. Xu, X.-R. Zhang, Z. Chen, J.-T. Li, L. Huang, K. Amine and S.-G. Sun. Nano Energy 28 (2016) 63-70. [4] R. Black, J.-H. Lee, B. Adams, C. A. Mims and L. F. Nazar. Angew. Chemie 125 ( 2013) 410-414. [5] H.-G. Jung, Y. S. Jeong, J.-B. Park, Y.-K. Sun, B. Scrosati and Y. J. Lee. ACS Nano. 7 (2013) 3532-3539. [6] X. Zeng, C. You, L. Leng, D. Dang, X. Qiao, X. Li, Y. Li, S. Liao and R. R. Adzic. J. Mater. Chem. A 3 (2015) 11224-11231. [7] X. Guo, P. Liu, J. Han, Y. Ito, A. Hirata, T. Fujita and M. Chen. Adv. Mater. 27 (2015) 6137-6143. [8] S. A. Freunberger, C. Yuhui, N. E. Drewett, L. J. Hardwick, B. Fanny and P. G. Bruce Angew. Chem. Int. Ed. 50 (2011) 8609-8613. [9] M. M. Ottakam Thotiyl, S. A. Freunberger, Z. Peng and P. G. Bruce. J. Am. Chem. Soc. 135 (2013) 494-500. [10] J. Xie, X. Yao, I. P. Madden, D.-E. Jiang, L.-Y. Chou, C.-K. Tsung and D. Wang. J. Am. Chem. Soc. 136 (2014) 8903-8906. [11] J.-J. Xu, Z.-W. Chang, Y. Wang, D.-P. Liu, Y. Zhang and X.-B. Zhang. Adv. Mater. 28 (2016) 9620-9628. [12] Y. Chang, S. Dong, Y. Ju, D. Xiao, X. Zhou, L. Zhang, X. Chen, C. Shang, L. Gu, Z. Peng and G. Cui. Adv. Sci. 2 (2015) 1500092. [13] K. Chen, L. Chen, Y. Chen, H. Bai and L. Li. J. Mater. Chem. 22 (2012) 20968-20976. [14] J. Zhang, Y. Zhao, X. Zhao, Z. Liu and W. Chen, Sci. Rep. 4 (2014) 6005. [15] L. Zou, Y. Jiang, J. Cheng, Y. Chen, B. Chi, J. Pu and L. Jian. Electrochim. Acta 262 (2018) 97-106.
13
[16] B. D. McCloskey, A. Speidel, R. Scheffler, D. C. Miller, V. Viswanathan, J. S. Hummelshøj, J. K. Nørskov and A. C. Luntz. J. Phys. Chem. Lett. 3 (2012) 997-1001. [17] H.-K. Lim, H.-D. Lim, K.-Y. Park, D.-H. Seo, H. Gwon, J. Hong, W. A. Goddard, H. Kim and K. Kang. J. Am. Chem. Soc. 135 (2013) 9733-9742. [18] S. A. Freunberger, C. Yuhui, P. Zhangquan, J. M. Griffin, L. J. Hardwick, B. Fanny, N. Petr and P. G. Bruce. J. Am. Chem. Soc. 133 (2011) 8040-8047. [19] S. R. Gowda, A. Brunet, G. M. Wallraff and B. D. McCloskey. J. Phys. Chem. Lett. 4 (2013) 276-279. [20] J. D. Wadhawan, P. J. Welford, E. Maisonhaute, V. Climent, N. S. Lawrence, R. G. Compton, H. B. McPeak and C. E. W. Hahn. J. Phys. Chem. Lett. 105 (2001) 10659-10668.
14
Figures:
Figure 1. (a) TEM image, (b) TGA, (c) Raman spectra and (d) XPS result of RuO2/GO catalyst.
15
Figure 2. (a) Discharge-charge curves of the Li-O2 battery running for 40 cycles to 1000 mAh g−1 at a rate of 200 mA g−1. (b) Plot of V150 value versus cycle number.
16
Figure 3. (a) XRD pattern and (b) Raman spectra and (c) FTIR spectra of 1st, 5th and 20th cycles of the cathodes discharged to 1000 mAh g−1.
17
Figure 4. Change of discharge product on cathode. (a) ~ (c): SEM images of the cathode discharged to 1000 mAh g−1 at 1st (a), 5th (b) and 20th (c) cycle. (d) ~ (f): OEMS analysis of the charge process of Li-O2 battery after 1st (d), 5th (e) and 20th (f) discharge-charge cycle, dash line was the theoretical line for oxygen.
18
Scheme 1. Schematic illustration of reactions on the cathode of Li-O2 battery during discharge-charge processes. (a) ORR and OER in an ideal battery; (b) CO2 accumulation and the consequent reaction change; (c) The discharge-charge process after VPT.
19
Figure 5. Effects of CO2 on charge process. (a) Discharge-charge curves of a Li-O2 battery at a rate of 200 mA g−1 at the 5th, 20th and 21st cycle. The battery was subjected to VPT before the 21st discharge-charge cycle. (b) SEM image of the cathode after the 21st cycle discharge. (c) OEMS analysis of gas released from Li-O2 battery at the 21st cycle charge process. (d) The discharge-charge curves of Li-O2 battery at a rate of 200 mA g−1, subsequent to the 3rd cycle CO2 was inflated into the battery before the 4th discharge (O2 : CO2 ≈ 1:1), and after the 4th cycle the battery was subjected to VPT.
20
Figure 6. (a) SEM image of cathode discharge to 1000 mAh g−1 at 4th cycle (After CO2 inflation). (b) Raman spectra, (c) FTIR spectra and (d) XRD of cathode discharge to 1000 mAh g-1 respectively at 1st and 4th (After CO2 inflation).
21
Figure 7. (a) Plot of V150 value versus cycle number of Li-O2 batteries without VPT and subjected to VPT after 19th, 39th, and 59th cycle, respectively. (b) The discharge-charge curves at the 19th, 20th, 39th, 40th, 59th, 60th, 65th cycle at a rate of 200 mA g−1 of the Li-O2 battery subjected to VPT after 19th, 39th, and 59th cycle, respectively. (c) Schematic presentation of Li-O2 battery preloaded CaO. (d) Discharge-charge curves of 10th, 20th, 40th, 60th, 70th cycle for the Li-O2 battery preloaded CaO at a rate of 200 mA g−1. (e) Plot of V150 versus cycle number of Li-O2 batteries without CaO and preloaded CaO, respectively. (f) OEMS analysis of gas released in 20st charge process of Li-O2 battery preloaded CaO.
22
Figure 8. (a) XRD and (b) CO2 signal detected by OMS of loop loaded CaO before and after cycling.
23
Li-Bin Chen received her M.S. degree from Xiamen University in 2014, and is pursuing his Ph.D. degree in Xiamen University. Her current research interest focuses mainly on lithium oxygen batteries.
Yu-Hao Hong received his B.S. degree from the Department of Chemistry, Xiamen University in 2014, and is pursuing his Ph.D. degree in Xiamen University. His current research interest focuses mainly on online electrochemical mass spectroscopy.
Liang-Ping Xiao received his M.S. degree from Xiamen University in 2017, and is pursuing his Ph.D. degree in Xiamen University. His current research interest focuses mainly on in-situ liquid-cell TEM.
Jin-Hai You received his B.S. degree from Fujian Agriculture and Forestry University
in 2017. He is pursuing his M.S. degree in Xiamen University, and his current research interest is on lithium metal anodes for Li metal batteries.
Wen-Jia Sheng received her B.Eng. degree from Tianjin University in 2016. She is a master degree candidate of Xiamen University and her research interest is on the cathode catalyst of non-aqueous lithium-oxygen batteries.
Hua Bai received his B. S. and Ph. D. degree in Department of Chemistry, Tsinghua University in 2004 and 2009, respectively. From 2009 to 2011, he worked in Department of Chemical Engineering, Tsinghua University, as a postdoctoral researcher. He then join Xiamen University, as an associate professor in College of Materials. His research focuses on energy conversion and storage, including developing new materials for supercapacitors, lithium ion batteries and solar cells.
Ling Huang received his Ph.D. degree from the Department of Chemistry of Xiamen University in 1997. He has worked in Xiamen University from 2000 to date. He is now a professor of chemistry at Xiamen University, and his research interests include synthesis of tin based alloys by electrochemical deposition and high-energy cathode materials of lithium ion batteries.
Shi-Gang Sun is a professor of chemistry at Xiamen University. He received his Bachelor's degree in 1982 from Xiamen University and Doctor at d′Etat in 1986 from Universit´e Pierre et Marie Curie (Paris VI), and carried out one year of postdoctoral research at the Laboratoire d′Electrochimie Interfaciale du CNRS, France. He is a member of Chinese Academy of Sciences, Fellow of the Royal Society of Chemistry, U.K., and Fellow of the International Society of Electrochemistry. His research interests include electrocatalysis, electrochemical surface science, spectroelectrochemistry, and electrochemical energy conversion and storage.
Research highlights： ： •
Effects of CO2 accumulation during cycling of a Li-O2 battery employing RuO2/rGO cathode is studied.
•
The charge potential at capacity of 150 mAh g-1 is taken as an indice to track the health state and evolution of a Li-O2 battery during cycling.
•
The transition of discharge product due to CO2 accumulation during cycling of a Li-O2 battery is revealed as the origin of performance fading.
•
A strategy to preload CO absorbent (CaO) has inhibited effectively CO2 accumulation and prolonged the cycle lifespan of Li-O2 battery.
1