A specially designed Li–H₂O₂ semi-fuel cell: A potential choice for electric vehicle propulsion

Abstract

A specially designed Li–H₂O₂ semi-fuel cell based on hybrid electrolytes is proposed. It is a combination of fuel cell and lithium battery. Five “mechanical charge–discharge” cycles (800 hours in total) of the Li–H₂O₂ semi-fuel cell were conducted. The cell exhibits fast mechanical rechargeability, good stability and high lithium utilization. Output power of the Li–H₂O₂ semi-fuel cell can be flexibly adjusted by changing H₂O₂ concentration.

---

The concern over energy exhaustion and environment pollution has made electric vehicles (EVs) more and more attractive. Driven by electric power transformed from water, wind, nuclear, geothermal and solar energy, EVs can reduce emission of greenhouse gas and reliance on fossil fuels. Better EVs require higher stability, larger capacity, shorter recharge time and lower-cost batteries. Recently, Li–O₂ battery has received considerable attention due to super-high theoretical specific energy, and it is considered as the power battery technology of the next generation.^1–3

However, at the present stage, rechargeable Li–O₂ battery still suffers from several problems:^4–7 (1) relative poor cycle performance. On the one hand, there is a significant capacity fade during charge–discharge cycles caused by electrical passivation at the cathode, growth of lithium dendrite and decomposition of electrolyte.^8,9 On the other hand, the voltage efficiency is relatively low. (2) Limited practical cell capacity. The cell capacity is limited by the capacity of porous air cathode,¹⁰ which is much lower than that of lithium anode (3862 mA h g⁻¹ theoretically) due to clogging. (3) Long recharge time, which is as long as that of traditional batteries. Li–O₂ battery with organic electrolyte exhibits a relatively good charge–discharge cycle performance, but clogging often occurs; using aqueous electrolyte can alleviate the clogging problem, unfortunately the charging behaviour of Li–O₂ battery with aqueous electrolyte is undetermined.¹¹ These problems have plagued efforts to develop Li–O₂ battery technology for practical use.

Since it is difficult to solve the problems, we may avoid them. As charge–discharge cycle performance of Li–O₂ battery is poor, and recharge process takes a long time and causes growth of lithium dendrite and exacerbates decomposition of the electrolyte, it may be better to replace traditional recharge by mechanical recharge. Keeping this in mind, we paid attention to the mechanical rechargeable metal–O₂ and metal–H₂O₂ fuel/semi-fuel cells,^12–15 which use reactive metal (Al, Mg and Zn as usual) as the anode (i.e., fuel) and O₂ or H₂O₂ as the oxidant, and convert stored chemical energy to electrical energy.

Among all the metal anode fuel/semi-fuel cells, Li–H₂O₂ semi-fuel cell may be a good choice for EVs in the future, for lithium is the most negative metal while possessing a super-high capacity. Li–H₂O₂ semi-fuel cell is a combination of lithium battery and fuel cell, enjoying the advantages of high theoretical specific energy, large capacity, and fast mechanical rechargeability. The application of H₂O₂ will not contaminate the environment, furthermore, H₂O₂ is easy to be added into the liquid electrolyte and its concentration can be varied in a wide range as H₂O₂ and water are completely miscible. These properties make it convenient to adjust the cell voltage, i.e., output power, of the Li–H₂O₂ semi-fuel cell by changing H₂O₂ concentration in the catholyte.

Li–H₂O₂ semi-fuel cell had been reported by Gautum Agarwal in 1999.¹⁶ In their cell, a 3% H₂O₂ solution was used as electrolyte and lithium was immersed in the solution. Unfortunately, the cell failed to discharge, and caught fire immediately. To the best of our knowledge, there have been no more reports about Li–H₂O₂ semi-fuel cell to date. To operate a Li–H₂O₂ semi-fuel cell stably and efficiently, the lithium anode must be strictly protected from any contact with H₂O₂, H₂O or air.

In order to ensure good protection of lithium anode in the Li–H₂O₂ semi-fuel cell, hybrid electrolyte type cell^17–25 is preferred, and to enable fast mechanical rechargeability, the cell structure should be specially designed. In our study, a hybrid electrolyte type Li–H₂O₂ semi-fuel cell with novel cell structure was proposed, and the cell performance was investigated.

Fig. 1 shows the cell structure of our Li–H₂O₂ semi-fuel cell. In the assembled cell, the Ni foam was rolled up and placed inside the cathode shell. The current collector was wedged in the anode shell with its hole aligned to the hole in the anode shell so that a lithium inlet was formed. The threads of cathode shell engaged with the mating threads of anode shell were tightened to compress the gaskets and LTAP pellet (purchased from Ohara Inc., Japan) between the two shells. The space inside the tightened shells was divided into two chambers (cathode chamber and anode chamber) by the gaskets, LTAP pellet and a small amount of epoxy glues, and only Li⁺, but no liquid or gas species, could pass through. The assembled semi-fuel cell received fuel, lithium, in an argon protected glove box. Small lithium particles were put into the anode chamber through the lithium inlet and adhered to the inner wall of the current collector; a few drops of organic electrolyte (1 M LiPF₆ in ethylene carbonate and diethyl carbonate (1 : 1 v/v)) were added as well. Then the copper screw was tightened to keep a good electrical contact with the current collector and sealed lithium inside the anode chamber. The fuelled up cell can be removed from the glove box and discharged in ambient air. When the discharge was finished, we moved the cell into the argon protected glove box, twisted the copper screw off, refuelled the cell by adding certain amounts of fresh lithium particles, tightened the copper screw, then mechanical recharge was finished. This entire recharge process took less than 10 minutes.

Fig. 1 Diagram of the opened specially designed Li–H₂O₂ semi-fuel cell.

Fig. 2 shows discharge curves of the tests. During the 800 h tests, the semi-fuel cell was mechanically recharged 5 times. Hence the fast mechanical rechargeability of the designed Li–H₂O₂ semi-fuel cell was confirmed. The average discharge voltage was 2.7 V, which was at the same level as the voltages of most reported Li–O₂ batteries.^26–30 Performance of the Li–H₂O₂ semi-fuel cell could be further improved by employing more advanced positive electrode.^12,31,32 The discharge voltage plateau in each voltage vs. discharge time curves suggested a good stability of our semi-fuel cell.

Fig. 2 Discharge curves of Li–H₂O₂ semi-fuel cell. Current: 0.1 mA; anolyte: 1.0 M LiPF₆ in EC–DEC (1 : 1 by volume); catholyte: 2.0 M LiCl and 0.1 M H₂O₂ solution; flow rate: 10 mL min⁻¹; operation temperature: 25 °C.

The open circuit voltage (OCV) was 3.0 V, which was much lower than the theoretical OCV of 4.83 V given by eqn (1) and (2)

(−): Li → Li⁺ + e⁻ (φ = −3.05 V) (1)

(+): H₂O₂ + 2H⁺+ 2e⁻ → 2H₂O (φ = +1.78 V) (2)

According to previous reports,^33,34 in neutral solution such as the catholyte in our case, reactions like eqn (3), (4) and (5) show may occur at the cathode as well (O₂ was generated from the chemical decomposition of H₂O₂)

H₂O₂ + 2e⁻ → 2OH⁻ (φ = +0.878 V) (3)

O₂ + 2H⁺+ 2e⁻ → H₂O₂ (φ = +0.695 V) (4)

O₂ + 2H₂O + 4e⁻ → 4OH⁻ (φ = +0.401 V) (5)

We investigated the open circuit potential (OCP) of the cathode, the observed OCP was only ∼0.52 V (Fig. S1†), which was much lower than its theoretical value of 1.78 V (see eqn (2)). This result indicates that the reaction couple at the electrode was not solely H₂O₂/H₂O as summarized in eqn (2), it was more likely a mixture of H₂O₂/H₂O, H₂O₂/OH⁻, O₂/H₂O₂ and O₂/OH⁻. These side reactions in cathode resulted in a low OCV. The reaction in cathode is quite complicated, further research is needed to clarify the mechanism.

The gap between OCV and discharge voltage became larger in the third and fourth tests, and discharge voltage slightly decreased with the increase in the discharge depth. This phenomenon could be attributed to the increase in internal resistance of the semi-fuel cell. As LTAP pellet was stable in catholyte, the rise of the internal resistance was probably caused by the degeneration of Ni foam (Fig. S2 and S4†). Actually, after replacing the aged Ni foam with new Ni foam (it is easy to replace through the catholyte outlet without disassembling the cell), the difference between OCV and discharge voltage in the fifth test decreased.

The capacity of rechargeable Li–O₂ battery is fixed and confined by the amount of lithium sealed inside the battery. However, the capacity of the Li–H₂O₂ semi-fuel cell increases with the amount of lithium added (see Table 1). This is as simple as the fact that the distance a car can cover depends on the amount of gasoline in the tank. Actually, mechanical recharge process was almost as fast as the refuel process of a gasoline engine.

Table 1 Cell capacity and Li utilization of Li–H₂O₂ semi-fuel cell

Weight of lithium (mg)	Cell capacity (mA h)	Specific capacity (mA h g^−1 Li)	Lithium utilization (%)
2.9	9.1	3138	81.3
10.4	31.1	2990	77.4
4.2	13.2	3143	81.5
3.6	11.6	3222	83.5
3.1	10.1	3258	84.4

---

Utilization of the fuel, lithium, was calculated by comparing practical specific capacity (based on the mass of lithium) to theoretical one (3862 mA h g⁻¹). Table 1 lists the lithium utilization in Li–H₂O₂ semi-fuel cell. It can be found that high average lithium utilization of 81.6% was gained, and it was almost independent of the mass of lithium added to the cell. More importantly, the specific capacity did not decrease at least during the first five “mechanical charge–discharge” cycles, which lasted for 34 days. High lithium utilization and relative stable specific capacity indicate that lithium in the cell had been well protected, so the corrosion reaction of lithium was very small and capacity fade caused by electrolyte decomposition and lithium dendrite formation was negligible.

H₂O₂ dissolved in LiCl solution acts as an active cathode substance, and its concentration affects the cell performance significantly. Fig. 3 shows the effect the H₂O₂ concentration on OCV of Li–H₂O₂ semi-fuel cell. It can be seen that the OCV of the cell increased with H₂O₂ concentration. The cell could be regarded as a Li–water/dissolved oxygen semi-fuel cell without H₂O₂ in the catholyte, and the OCV was lower than 2.75 V. When adding a small amount of H₂O₂ to the catholyte to get a low H₂O₂ concentration of 0.05 M, an increment of more than 100 mV in OCV was observed. When H₂O₂ concentration increased to 2.0 M, OCV reached 3.24 V, 500 mV higher than the initial one.

Fig. 3 Effects of the H₂O₂ concentration on open circle voltage of Li–H₂O₂ semi-fuel cell. Anolyte: 1.0 M LiPF₆ in EC–DEC (1 : 1 by volume). Catholyte: 2.0 M LiCl + x M H₂O₂ (x = 0, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 1.0 and 2.0). Flow rate: 10 mL min⁻¹. Operation temperature: 25 °C.

According to Nernst equation, it is reasonable to suggest a relationship between OCV and H₂O₂ concentration as:

OCV = A + B ln c(H₂O₂) (6)

where A and B are apparent parameters of the semi-fuel cell system. Using eqn (6), the measured values (except for the value of x = 0) were fitted, and resulted in the red line shown in Fig. 3, which represents an equation of OCV = 3.16 + 0.094 ln c(H₂O₂). As can be seen, the measured values match well with the fitted line. Eqn (6) is only an empirical equation, nevertheless, it is still meaningful. According to eqn (6), OCV of the semi-fuel cell, as well as the cell voltage and output power, could be conveniently adjusted by adjusting H₂O₂ concentration.

To verify this inference, various amounts of 30% H₂O₂ solution was injected directly into the 5 mL cathode chamber while the semi-fuel cell was discharged at a constant current of 0.1 mA. As Fig. 4 shows, when stopping the flow of catholyte and injecting H₂O₂ solution into the cathode chamber, the H₂O₂ concentration inside the chamber increased, and the cell voltage increased consequently. The amount of H₂O₂ solution injected into the chamber was positively related to, but not directly proportional to, the increment of cell voltage, which generally conformed to eqn (6). When starting the flow of catholyte, the increased cell voltage dropped quickly, for the catholyte with high H₂O₂ concentration inside the cathode chamber was diluted. During the experiment, discharge current (I) was kept constant, so output power (P = I × U) of the Li–H₂O₂ semi-fuel cell changed as cell voltage (U) changed.

Fig. 4 Adjusting cell voltage of the Li–H₂O₂ semi-fuel cell. Catholyte: 2.0 M LiCl + 0.1 M H₂O₂. Flow rate: 10 mL min⁻¹. Current: 0.1 mA. Operation temperature: 25 °C. a, c, e, g and h: Stopped the flow of catholyte, and injected 0.1, 0.2, 0.4, 0.6 and 0.6 mL 30% H₂O₂ solution into the cathode chamber (5 mL in capacity), respectively. b, d and f: Started the flow of catholyte.

Catholyte flow rate also affects cell voltage, as shown in Fig. 5. When the catholyte did not flow, cell voltage decreased while discharging, which could be attributed to continuously decrease of H₂O₂ concentration around the cathode. With catholyte flow, there was a supplement of H₂O₂, so that H₂O₂ concentration around the cathode was stable, and discharge voltage was kept relative stable consequently. Furthermore, the higher the flow rate, the higher the discharge voltage, but there was a limit, and too high a flow rate, such as 20 mL min⁻¹ in our experiment, would make the discharge voltage unstable. This results is consistent with ref. 35.

Fig. 5 Effects of the flow rate on the cell voltage. Anolyte: 1.0 M LiPF₆ in EC–DEC (1 : 1 by volume). Catholyte: 0.1 M H₂O₂ + 2 M LiCl. Current: 0.1 mA. Operation temperature: 25 °C.

EV needs different power in different operating conditions. For Li–H₂O₂ semi-fuel cell, output power can be flexibly adjusted to meet EV's different requirement of power by changing H₂O₂ concentration and catholyte flow rate. When low power is needed, operation on the semi-fuel cell is to keep the flow of catholyte, in which H₂O₂ concentration is relatively low; when high power is needed, stop the flow of catholyte, and inject a certain amount of H₂O₂ into the cathode chamber. After the requirement of high power is satisfied, restart the flow of the catholyte to dilute H₂O₂ inside the cathode chamber to keep chemical decomposition of H₂O₂ at a low level, for decomposition rate of H₂O₂ increases with H₂O₂ concentration.³² To confine H₂O₂ concentration in the catholyte, we can increase output power by increasing catholyte flow rate instead of adding H₂O₂.

The results suggested good stability and high lithium utilization of the specially designed Li–H₂O₂ semi-fuel cell. Besides, since Li–H₂O₂ semi-fuel cell is a combination of Li battery and fuel cell, it has advantages of both:

(1) High cell voltage and energy density. The molar mass and electrode potential of H₂O₂ is almost the same as that of O₂. As a result, the cell voltage and energy density (based on the mass of active materials on both electrodes) of Li–H₂O₂ semi-fuel cell is almost the same as that of Li–O₂ battery. Actually, the present semi-fuel cell exhibited a cell performance similar to Li–O₂ battery.

(2) Fast mechanical rechargeability. The special cell structure allowed the semi-fuel cell to get mechanically recharged, which took less than 10 minutes, almost as fast as the refuelling process of a gasoline engine. The mechanical recharge process could be even faster in the future, if conducted in situ by employing an argon protected pump gun to pump lithium particles into anode chamber, without detaching the cell from EV and moving it to an argon protected box.

(3) Large capacity. In the discharge tests, with more lithium stored in the anode chamber, the capacity of the semi-fuel cell became correspondingly larger. It depended on the maximum amount of lithium that can be stored in the anode chamber, or lithium storage tank outside the cell. Theoretically, the energy capacity can be as large as designed, and sustain EV drive for a long distance.

(4) Flexibility in adjusting output power. The cell voltage and output power was very sensitive to H₂O₂ concentration and catholyte flow rate. By injecting a certain amount of H₂O₂ into the cathode chamber and changing flow rate, the output power of the semi-fuel cell can be controlled to meet different power requirements of the EV.

Moreover, combined with a lithium recycle system, the fuel of Li–H₂O₂ semi-fuel cell, lithium, can be cyclically utilized,³⁶ as Fig. 6 shows. The recharge processes in individual batteries, i.e., electrolysis of lithium salts producing lithium metal, are integrated and conducted in electrolysis plants instead, which might be meaningful in energy saving and efficiency enhancement. In the lithium recycle loop, energy (solar, wind, water, tide, nuclear energy and so on) supplied to the electrolysis plants finally drives cars, lithium just acts as an “energy carrier”. H₂O₂ is a common industrial product, and lithium resource is relatively rich on earth, and the method of producing Li from LiOH is well known, Li–H₂O₂ semi-fuel cell with a lithium recycle system may be a promising environmentally friendly power source for EVs in the future. At present, some problems remain. For example, the Li–H₂O₂ semi-fuel cell still suffers a large internal resistance, resulting in a poor high rate discharge performance (see Fig. S5†). Further research is required.

Fig. 6 A concept image for Li–H₂O₂ semi-fuel cell with lithium recycle system for EV propulsion. Cars get refuelled in lithium stations; the only emission is LiOH, which can be used to fabricate metallic Li for fuel in plants.

Conclusions

A hybrid electrolyte based Li–H₂O₂ semi-fuel cell with a specially designed cell structure was proposed. Five “mechanical recharge–discharge” cycles (800 hours in total) were conducted, and the effects of H₂O₂ concentration and catholyte flow rate on the cell performance were studied. The cell exhibited good stability, high lithium utilization, fast mechanical rechargeability and adjustability of capacity, cell voltage and output power. Considering all the features mentioned above, Li–H₂O₂ semi-fuel cell is a promising power battery technology and it provides a new choice for electric vehicle propulsion.

Acknowledgements

The authors would like to thank the National Natural Science Foundation of China for financial support (NSFC-no. 51221291 and 51172119).

Notes and references

1. P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J.-M. Tarascon, Nat. Mater., 2012, 11, 19.
2. M. Park, H. Sun, H. Lee, J. Lee and J. Cho, Adv. Energy Mater., 2012, 2, 180.
3. D. Capsoni, M. Bini, S. Ferrari, E. Quartarone and P. Mustarelli, J. Power Sources, 2012, 220, 253.
4. F. Li, T. Zhang and H. Zhou, Energy Environ. Sci., 2013, 6, 1125.
5. N.-S. Choi, Z. Chen, S. A. Freunberger, X. Ji, Y.-K. Sun, K. Amine, G. Yushin, L. F. Nazar, J. Cho and P. G. Bruce, Angew. Chem., Int. Ed., 2012, 51, 9994.
6. R. Cao, J.-S. Lee, M. Liu and J. Cho, Adv. Energy Mater., 2012, 2, 816.
7. R. Padbury and X. Zhang, J. Power Sources, 2011, 196, 4436.
8. S. A. Freunberger, Y. Chen, Z. Peng, J. M. Griffin, L. J. Hardwick, F. Bardé, P. Novák and P. G. Bruce, J. Am. Chem. Soc., 2011, 133, 8040.
9. Y. Chen, S. A. Freunberger, Z. Peng, F. Bardé and P. G. Bruce, J. Am. Chem. Soc., 2012, 134, 7952.
10. I. C. Jang, Y. Hidaka and T. Ishihara, J. Power Sources, 2013, 244, 606.
11. J. Wang, Y. Li and X. Sun, Nano Energy, 2013, 2, 443.
12. C. J. Patrissi, R. R. Bessette, Y. K. Kim and C. R. Schumacher, J. Electrochem. Soc., 2008, 155, B558.
13. D. J. Brodrecht and J. J. Rusek, Appl. Energy, 2003, 74, 113.
14. S. Zhang, J. Jiang and D. Zhou, Int. J. Energy Res., 2012, 36, 953.
15. P. Sapkota and H. J. Kim, Ind. Eng. Chem. Res., 2009, 15, 445.
16. G. Agarwal, J. Propulsion, 1999, 16, 367.
17. Y. Wang, P. He and H. Zhou, Energy Environ. Sci., 2010, 3, 1515.
18. H. Zhou, Y. Wang, H. Li and P. He, ChemSusChem, 2010, 3, 1009.
19. Y. Wang and H. Zhou, J. Power Sources, 2010, 195, 358.
20. E. Yoo, J. Nakamura and H. Zhou, Energy Environ. Sci., 2012, 5, 6928.
21. T. Zhang, N. Imanishi, Y. Shimonishi, A. Hirano, Y. Takeda, O. Yamamoto and N. Sammes, Chem. Commun., 2010, 46, 1661.
22. L. Li, X. Zhao and A. Manthiram, Electrochem. Commun., 2012, 14, 78.
23. L. Li and A. Manthiram, J. Mater. Chem. A, 2013, 1, 5121.
24. Y. Li, K. Huang and Y. Xing, Electrochim. Acta, 2012, 81, 20.
25. Y. Li, Z. Huang, K. Huang, D. Carnahan and Y. Xing, Energy Environ. Sci., 2013, 6, 3339.
26. D. Xu, Z.-L. Wang, J.-J. Xu, L.-L. Zhang and X.-B. Zhang, Chem. Commun., 2012, 48, 6948.
27. Z. Guo, G. Zhu, Z. Qiu, Y. Wang and Y. Xia, Electrochem. Commun., 2012, 25, 26.
28. Y.-C. Lu, B. M. Gallant, D. G. Kwabi, J. R. Harding, R. R. Mitchell, M. S. Whittingham and S.-H. Yang, Energy Environ. Sci., 2013, 6, 750.
29. W. Zhang, Y. Zeng, C. Xu, H. Tan, W. Liu, J. Zhu, N. Xiao, H. H. Hng, J. Ma, H. E. Hoster, R. Yazami and Q. Yan, RSC Adv., 2012, 2, 8508.
30. H.-G. Jung, J. Hassoun, J.-B. Park, Y.-K. Sun and B. Scrosati, Nat. Chem., 2012, 4, 579.
31. W. Yang, S. Yang, W. Sun, G. Sun and Q. Xin, Electrochim. Acta, 2006, 52, 9.
32. C. Shu, E. Wang, L. Jiang, Q. Tang and G. Sun, J. Power Sources, 2012, 208, 159.
33. X. Jiang, D. X. Cao, Y. Liu, G. L. Wang, J. L. Yin, Q. Wen and Y. Y. Gao, J. Electroanal. Chem., 2011, 658, 46.
34. Y. Wang, Z. Guo and Y. Xia, Adv. Energy Mater., 2013, 3, 713.
35. T. Lei, Y. Tian, G. Wang, J. Yin, Y. Gao, Q. Wen and D. Cao, Fuel Cells, 2011, 11, 431.
36. P. He, Y. Wang and H. Zhou, Electrochem. Commun., 2010, 12, 1686.

---

Footnote

† Electronic supplementary information (ESI) available: Open circuit potential (OCP) curves for cathode (H₂O₂); simple derivation from Nernst equation to eqn (5); picture of the LTAP pellet; SEM images and XRD patterns of the LTAP pellets before and after use; SEM images and EDS results of Ni foam before and after use; detailed information about LTAP pellet and Ni foam; polarization curve of the Li–H₂O₂ semi-fuel cell; picture of the Li–H₂O₂ semi-fuel cell test set-up. See DOI: 10.1039/c3ra47616b

---