Enabling safe aqueous lithium ion open batteries by suppressing oxygen reduction reaction

Due to the non-flammable nature of water-based electrolytes, aqueous lithium-ion batteries are resistant to catching fire. However, they are not immune to the risk of explosion, since the sealing structure adopted by current batteries limits the dissipation of heat and pressure within the cells. Here, we report a safe aqueous lithium-ion battery with an open configuration using water-in-salt electrolytes and aluminum oxide coated anodes. The design can inhibit the self-discharge by substantially suppressing the oxygen reduction reaction on lithiated anodes and enable good cycle performance over 1000 times. Our study may open a pathway towards safer lithium-ion battery designs.


Supplementary Notes Supplementary Note 1. Overall electron transfer number of the ORR and corresponding percentage H 2 O 2 in 1 m SiWE.
The amount of H2O2 was calculated based on the Supplementary Equation 1, where jr and jd are the ring and the disk current, respectively, and N is the collection efficiency (0.37).

Supplementary Note 2. Raman spectrum of discharged carbon cathodes.
The carbon electrode was discharged with a constant current of −0.1 mA for 20 h in the 28 m WiSE. Before the Raman characterization, the carbon cathode was rinsed by DME (Dimethoxyethane) several times. Supplementary Fig. 4 compares the RDE curves of Al2O3, LiTi2(PO4)3/Carbon black (90:5 in weight, the same ratio as in real electrode) and Carbon black with 1 m SiWE. The ORR kinetics at the LiTi2(PO4)3/Carbon black surface are faster than at the Al2O3 surface, but it is slower than at the carbon black surface. The limiting current density at 1600 rpm at the LiTi(PO4)3/Carbon black surface was 0.31 mA cm −2 , which is larger than at the Al2O3 surface (0.107 mA cm −2 ) and lower than at the carbon black surface (0.85 mA cm −2 ). Therefore, the Al2O3 nano layer coating would significantly slow down the ORR kinetics at the LiTi2(PO4)3/Carbon black electrode surface.

Supplementary Note 4. Effects of the ORR on the self-discharge performance of the lithiated Li 3 Ti 2 (PO 4 ) 3 electrode in different atmospheres.
As shown in Supplementary Fig. 5, the open circuit potential of the fully lithiated Li3Ti2(PO4)3 electrode in the 1 m SiWE after exposure to the ambient environment could be sustained at the equilibrium potential only for about 4 h before the potential quickly rose to +0.2V after 10 h, suggesting that the fully lithiated Li3Ti2(PO4)3 electrode was fully self-discharged after 10 h in the ambient environment. When the fully lithiated Li3Ti2(PO4)3 electrode in the 1 m SiWE was introduced with pure O2, it was completely self-discharged after only 0.5 h. However, in the absence of O2 (in a N2 atmosphere), it can maintain its potential for 50 h. These results clearly demonstrate that even in an ambient environment (not pure O2), the O2 in the air can quickly chemically oxidize the lithiated anodes in dilute aqueous electrolytes, resulting in fast self-discharge.

Supplementary Note 5. Effects of the ORR on the self-discharge of the lithiated Al 2 O 3 @LiTi 2 (PO 4 ) 3 electrode in 10 m LiTFSI electrolyte.
The Coulombic efficiency of the ALD-Al2O3 coated LiTi2(PO4)3 in the 10 m LiTFSI electrolyte at a current density of 0.5 A g −1 is 94% ( Supplementary Fig. 8a), which is higher than the 89% Coulombic efficiency in the 1 m SiWE, but lower than the 99.9% Coulombic efficiency in the 28 m WiSE. The capacity retention of the fully lithiated Li3Ti2(PO4)3 electrode after opening it to the air for 10 h is ~52% (Supplementary Fig.  8b). All of these results indicate that the 10 m LiTFSI electrolyte can partially suppress the ORR. The solubility of O2 in the 10 m LiTFSI electrolyte is 1.52 mg L −1 , which is lower than 1.97 mg L −1 in the 1 m SiWE, but it is much higher than in the 28m WiSE (0.95 mg L −1 ). The relatively low O2 solubility reduces the ORR kinetics to a certain degree; however, the 10 m LiTFSI electrolyte cannot form a passivation layer on the anode to further reduce the ORR kinetics. That is another reason for the relatively poor performance of the lithiated Li3Ti2(PO4)3 in the 10 m LiTFSI electrolyte when compared to the performance in the 28 m WiSE.
Supplementary Note 6. Self-discharge performance of the fully charged LiMn 2 O 4 //LiTi 2 (PO 4 ) 3 open-cell with the 1 m SiWE exposed to the ambient environment. We fabricated a LiMn2O4//LiTi2(PO4)3 open full cell with the 1 m SiWE. The full cell was fully charged and then allowed to rest at open-circuit for 10 h under an ambient environment. As demonstrated in Supplementary Fig. 12, almost all of the charge capacity was self-discharged after 10 h of open-circuit resting due to the ORR on the lithiated anode. This is consistent with the results in Supplementary Fig. 5.

Supplementary Note 7. Electrochemical window of the 63 m WiSE.
The electrochemical stability window of the 63 m WiSE was evaluated with LSV on inactive stainless-steel foil electrodes. Supplementary Fig. 17 indicates that the potential for hydrogen and oxygen evolution are 1.76 V and 4.91 V, respectively.
Supplementary Note 9. Picture and electrochemical performance of the 28 m gel WiSE. Supplementary Fig. 19a shows that the 28 m gel WiSE does not flow at room temperature. As shown in Supplementary Fig. 19b and 19c, the LiMn2O4//Al2O3@LiTi2(PO4)3 open-cell with the 28 m gel WiSE delivered a comparable charge/discharge behavior and cycling stability to the LiMn2O4//Al2O3@LiTi2(PO4)3 open-cell with the 28 m liquid WiSE. In fact, the LiTFSI concentration in the PVA-H2O system can reach 35 m, which can further enhance the electrolyte's electrochemical stability window, reduce the oxygen solubility and the ORR rate, and suppress the electrolyte's solvent evaporation.