Lithium-Air Batteries Are Great, But So Are Their Problems

November 2014
A lithium-air cell. Courtesy of the Nazar Group, University of Waterloo.

Lithium-air batteries, also known as Li-O2, promise to transform energy use this century as lightweight methods of storing electricity. They have great potential, providing up to five times more energy than the lithium-ion batteries which currently power our cell phones and tablets. It may even be possible to have a rechargeable battery of up to 1,000 watt-hours per kilogram, and all it will need is oxygen. Such a battery could be used to fuel electric automobiles and store the electricity generated by solar panels and wind turbines.

Li-O2 batteries consist of a lithium metal anode whose atoms supply the electrons for the electric circuit when it is being used. The residual Li+ ions then migrate across an electrolyte to the cathode where the incoming electrons from the circuit attach to oxygens (O2) from the atmosphere, forming peroxide ions, O22─. The overall chemical process is: 2Li + O2 → Li2O2.

When the battery is recharged, the reverse reaction occurs. Lithium metal atoms and O2 are regenerated, with the oxygen being retained in a closed system or supplied anew in an open system. Although the chemistry seems simple, there are hurdles to be overcome before such batteries reach the marketplace.

One chemist who is at the forefront of Li-O2 research is Linda Nazar of the University of Waterloo at Ontario, Canada. She was named a Thomson Reuters Highly Cited Researcher in June this year, and is well aware of the problems associated with lithium-air batteries.

As she tells ScienceWatch, “We are studying all aspect of lithium-air chemistry, and that includes the development of better and more stable electrolytes. This is coupled with new active-host materials for oxygen reduction and which are resistant to deleterious side reactions that affect performance. Our aim is to understand of the mechanisms that control oxygen reduction in order to determine what cell chemistries are primarily responsible for side-reactions during discharge and charge, and how to overcome them.”

One obvious problem with lithium-air batteries is lithium itself. This is a reactive metal and yet it needs to be in contact with an electrolyte with which it must not react. Nor must its ions react with the peroxide ions that are formed.

The cathode, too, has its problems. There has to be easy diffusion of oxygen from the air into this so it has to be porous and yet not allow H2O and CO2 to gain access as these will react to form lithium hydroxide (LiOH) and lithium carbonate (Li2CO3), both of which will not regenerate the lithium when the battery is recharged. An oxygen-differentiating polymer membrane is therefore needed to prevent these atmospheric gases gaining access, and such membranes are available.

Carbon is seen as a suitable cathode and has been used in most batteries. It is now recognized that carbon reacts with lithium peroxide and that this hinders recharging. Cathode materials that form a passive layer in contact with the peroxide are necessary.  

Despite these hurdles, there is a lot of research going into Li-O2 batteries. The prize of creating the first such battery, and which must go through many discharge/re-charge cycles, will not only be a grateful world, but possibly great financial reward. Not surprisingly, there has been an upsurge lately in publications in this field, as captured by the Web of Science and documented in the accompanying table.  This lists those papers of 2012 to 2014 that have collected most citations, and all have more than 70.

Highly Cited Li-O2 Papers, 2012 to 2014

# Paper Citations
1 P.G. Bruce, et al., “Li-O2 and Li-S batteries with high energy storage,” Nature Materials., 11 (1):19-29, 2012. [U. St. Andrews, Scotland] 717
2 Z. Peng, et al., “A reversible and higher-rate Li-O2 battery,” Science, 337: 563-6, 2012. [U. St. Andrews, UK] 256
3 N-S. Choi, et al., “Challenge facing lithium batteries and electrical double-layer capacitors,” Angew. Chemie Int. Ed.., 51 (40), 9994-1--24, 2012. [Ulsan National Institute of Science & Technology, Ulsan, Korea] 243
4 H.-G. Jung, et al., “An improved high-performance lithium-air battery,” Nature Chemistry, 4 (7), 579-85, 2012. [Hanyang U., Seoul, Korea] 203
5 B.D. McCloskey, et al., “Twin problems of interfacial carbonate formation in nonaqeous Li-O2 batteries,” J. Phys. Chem. Lett., 3 (8), 997-1001, 2012. [SLAC National Accelerator Lab., Menlo Park, CA] 166
6 R. Black, et al., “Screening for superoxide reactivity in Li-O2 batteries; effect on Li2O2/LiOH crystallization”, J. Amer. Chem. Soc., 134 (6), 2902-5, 2012. [U. Waterloo, Ontario, Canada] 152
7 Y. Shao, et al., “Electrocatalysts for non-aqueous lithium-air batteries: Status, challenges, and perspective,” ACS Catalysis, 2 (5), 844-57, 2012. [Pacific Northwest National Lab.,  Richland, WA] 98
8 H. Wang, et al., “Rechargeable Li-O2 batteries with a covalently coupled MnCo2O4-graphene hybrid as an oxygen cathode catalyst,” Energy & Environ. Sci., 5 (7), 7931-5, 2012. [Stanford U., CA] 89
9 M.M.C Thotiyl, et al., “The carbon electrode in nonaqueous Li-O2 cells,” J. Amer. Chem. Soc., 135 (1), 494-500, 2013. [U. St. Andrews, UK] 89
10 S.H. Oh, et al., “Synthesis of a metallic mesoporous pyrochlore as a catalyst for lithium-O2 batteries,” Nature Chem., 4 (12), 1004-10, 2012. [U. Waterloo, Ontario, Canada] 87
11 Y. Chen, et al., “Li-O2 battery with a dimethylformamide electrolyte,” J. Amer. Chem. Soc., 134 (18), 7952-7, 2012. [U. St. Andrews, UK] 84
12 B.M. Gallant, et al., “Chemical and morphological changes of Li-O2 battery electrodes upon cycling,” J. Phys. Chem., 116 (39), 20800-5, 2012. [MIT, Cambridge, MA] 82
Source: Thomson Reuters Web of Science

Paper #1 is a noteworthy and timely review by Peter Bruce, a leading figure in this area. He is now based at Oxford University. Paper #3 is also a review of which he was a co-author along with others based in Korea, Atlanta and Canada. This focuses on the issues that have to be tackled if Li-O2 batteries are ever to be used commercially, in particular the electrolyte and the cathode.  

THE ELECTROLYTE

Several kinds of electrolyte have been used in lithium-air batteries, including aqueous ones which clearly need to incorporate an impervious polymer barrier to protect the lithium metal. Non-aqueous ones do not require this protection provided they are inert toward lithium, and organo-carbonates and ethers are suitable ones.

Those that showed the most promise are dimethoxyethane, CH3OCH2CH2OCH3, and diethyl carbonate, (CH3CH2O)2CO, but they are not without problems, such as the formation of a monolayer of lithium carbonate on the cathode; this interferes with the recharging of such a cell. Paper #5 from a group headed by Alan Luntz at the SLAC National Accelerator Laboratory at Menlo Park, California, used X-ray photoelectron spectroscopy and isotope labelling to show that the carbonate originated from the carbon electrode itself. However, Nazar has found that replacing the CH2 hydrogens of dimethoxyethane with methyl (CH3) groups to form 2,3-dimethyl-2,3-dimethoxybutane gives a much better electrolyte which forms and ionic liquid with lithium and this greatly enhances battery performance (Adv. Energy Mater., 4: 1-11,2014).

Paper #2 offers the best answer yet to the electrolyte issue and reports a battery that can undergo 100 cycles and retain 95% of its capacity. This battery has nanoporous gold as the cathode and dimethyl sulfoxide (CH3)2SO as the electrolyte, and the formation of Li2O2 is virtually 100%. However, nanogold is not without its drawbacks as Bruce has pointed out and he suggests titanium carbide is an even better cathode (Nature Materials, 12 (11), 1049-1055, 2013), while researchers at MIT have also questioned the use of dimethyl sulfoxide because of its reactivity with Li2O2 and the formation of LiOH (J. Chem. Phys. Letters, 29 July 2014).

The electrolyte is the main feature of paper #4 from Hanyan University of Seoul, Korea. Yang-Koo Sun and colleagues used tetra(ethylene) glycol dimethyl / ether-lithium triflate, CH3(OCH2CH2)4OCH3 / CF3SO3Li, as the electrolyte and produced a battery that was able to operate successfully over 50 charge/discharge cycles and delivered a higher energy density than other lithium-air batteries so far reported. Nor did their carbon electrode require a catalyst.

In paper #11, Bruce reports the investigation of N,N-dimethylformamide, (CH3)2NC(O)H, as a possible solution but concludes that any advantages it offers are not sufficient for it to be used. His group have also looked into the carbon electrodes used in such cells, and in paper #9 he discusses their limitations especially with respect of the formation of lithium carbonate when charging the batteries above 3.5 volts.

CATHODE AND CATALYST

It has yet to be decided which kind of cathode works best. Carbon is the most popular, although Paper #2, mentioned above, uses nanogold. Paper #7 by Jun Lui of the Pacific Northwest National Laboratory, Richland, Washington , is a review of the carbon cathode catalyst problem and the challenges which need to be overcome.

Another Nazar paper is #10, which focuses on the catalyst. It is this which enables oxygen to be evolved during recharging, and while the carbons of the electrode itself will catalyze the formation of peroxide, they are not effective for the reverse process. Now Nazar reports that lead ruthenium oxide pyrochlore, Pb2Ru2O7-8  is a good catalyst. The paper also gives the synthesis of a new version of this compound and shows that it forms a high-performance cathode with reversible capacities that show a significant improvement.

Hongiie Dai and colleagues at Stanford University have devised a Li-O2 battery with a manganese cobalt oxide / graphene hybrid cathode catalyst –see #8. The hybrid was produced by direct nucleation and growth of MnCo2O4 nanoparticles on reduced graphene oxide. The result was an impressive performance and longer cycle life of the battery, and better than catalysts like platinum.

Nazar and her team have been looking into the role of superoxide (O2─) which is thought to be an intermediate in the formation of peroxide (O22─).  Her paper, #6, reveals that superoxide ion can react with polyvinylidene fluoride which is also used in these cells and which was assumed to be inert. It results in the formation of unwanted lithium hydroxide, LiOH.

Paper #12 comes from a group at MIT headed by Yang Shao-Horn. This too addresses the problem of chemical and structural changes in carbon nanotube electrodes in a battery that had dimethoxyethane as the electrolyte. By means of X-ray analysis they showed how the formation of lithium carbonate built up over several discharges of the battery, much to its detriment.

The most recent work on Li-O2 electrodes comes from collaboration between groups based at Seoul National University and the University of Texas at Dallas which describes a woven porous carbon nanotube electrode in combination with lithium iodide as a catalyst (see H.-D. Lim, et al., Angew. Chem. Int. Ed., 53 (15): 3926-31, 2014), and this was put through 900 charge/re-charge cycles and continued to perform well.

So when might we expect to see lithium-air batteries in general use?

Nazar: “It is too soon to make a prediction because major breakthroughs can change the timing very significantly, and we shouldn’t expect all the bugs to be ironed out for up to 10 years. Maybe the battery technology will never work, or perhaps practical cells could be available earlier. The reason for the uncertainty is that the chemistry is quite complex, and the individual components of the cell—the positive and negative electrodes, the electrolyte, and the interfaces between them—are highly interdependent. Optimization of each component requires a deep understanding of the chemistry.”

Quite rightly, Nazar strikes a cautionary note. In the past two years, however, things have been moving rapidly, so it’s possible to end this article on a positive note: It’s light, it’s long-lasting, it’s looking good – it’s lithium. Metal number 3 is on its way!


Dr. John Emsley is based at the Department of Chemistry, Cambridge University, UK.

The data and citation records included in this report are from Thomson Reuters Web of ScienceTM. Web of ScienceTM is a registered trademark of Thomson Reuters. All rights reserved.