The ongoing saga of the Boeing 787 Dreamliner has resulted in a surge of partial or completely misleading stories about modern battery technology. While I’m far from an expert in the field, it’s one I follow closely, and I think I can contribute an “interested outsider” perspective on the state of rechargeable batteries and related technologies, circa 2013.
Let’s start by talking terminology. Lithium-Ion is an umbrella term, which represents a whole family of technologies. Simply knowing that a given application (like an airplane) makes use of “lithium-ion batteries” tells you very little about the performance, safety, and reliability characteristics of those batteries.
Battery technology is a materials-science intensive field, so it should come as no surprise that material choice is the key differentiator between batteries in the lithium-ion family. The three core components of a battery are the cathode and anode, and the electrolyte which separates them.
While there are hundreds of combinations of materials in use, depending on the intended application (and the patent pools of their backers), the most meaningful differentiation to be aware of is the types of positive electrodes (cathodes) in use.
The three primary families of cathode materials, and those worth knowing a little something about, are lithium-cobalt, lithium-iron-phosphate, and lithium-manganese. Each has different pros, cons, and risks.
A further note about terminology here – seeing types of electrodes written in this fashion might cause you to think that other terminology, like lithium-polymer, also refers to electrode choice as well. Unfortunately, it’s just confusing terminology. In fact, lithium-polymer refers to the the electrolyte, and a lithium-polymer battery can use any of the above mentioned electrode materials. Your laptop, for example, almost certainly uses lithium-polymer batteries with lithium-cobalt cathodes.
Now, a battery doesn’t contain pure lithium. That’s why you’re not on fire right now. The lithium is bonded with another material – that’s the cobalt, iron-phosphate, etcetera. These molecules also include oxygen. When exposed to high temperatures, these bonds can break down, resulting in nice, reactive lithium, along with fire’s friend, oxygen. In the case of a battery, a high-temperature situation can result from poor charging circuitry, short circuits, punctures or other external trauma. Since a battery generally consists of many cells, a single failed cell can easily produce enough heat to initiate a chain reaction.
Lithium-cobalt is the most common type of lithium-ion cathode, and delivers high energy density, relatively low cost manufacturing, and decent longevity when managed properly. The primary downside is that the lithium-cobalt bond is relatively weak, meaning these are generally types of lithium batteries which are at fault when you hear about battery fires. See, for example, the 787.
The most common alternative to lithium-cobalt is lithium-iron-phosphate. The A123 Systems batteries I’ve written about in the past are a derivative of this technology. The lithium-iron-phosphate bond is inherently more stable, even when abused or severely heated. The structure of the lithium-iron-phosphate molecule is such that it takes far more energy to free the lithium. Thus, these batteries are ideal for environments in which safety is key – automotive uses for example.
Now, it’s fair to ask why the Boeing 787 doesn’t use this type of battery. I’m obviously not privy to the internal engineering decisions at Boeing, but I can hazard a guess. First off, the battery design for the 787 was locked in 2005 or 2006. Back then, the technology for lithium-iron-phosphate was relatively immature and volume use wasn’t common. Additionally, for a given power output, a lithium-iron-phosphate battery will be larger and heavier than a corresponding lithium-cobalt design – this would have been even more pronounced in 2005.
In addition, the types of situations in which a lithium-iron-phosphate design is “safer” don’t commonly occur on an aircraft. For example, if the relatively small battery of a 787 is engulfed in flames, there are far, far bigger issues to worry about. The risk in an automotive implementation is that a relatively minor accident that damages the battery pack could cause a thermal runaway condition. There don’t tend to be “relatively minor” accidents involving massive jets. The other types of issues that can cause problems with batteries should be able to be mitigated through external controls – smart chargers with fused links in the case of overvoltage, etcetera. When we finally learn (if we learn) what caused the issues on the 787, I would suspect we’ll find that at least part of the cause was poor design or manufacturing issues surrounding these systems, rather than in the battery cells themselves.
There are a variety of other cathode chemistries in various applications. In particular, lithium manganese oxide and related manganese compounds provide better longevity and performance in harsh environments, but don’t yet excel in general purpose situations.
Supercapacitors represent another, related family of energy storage technologies which occasionally spawns a lot of interest, without necessarily a lot of results. Like all capacitors, the supercapacitor (ne ultracapacitor) stores a static charge using a variety of different materials. A supercapacitor can store energy very quickly, for a relatively long time, and survives a far greater number of charge cycles than a chemical battery. Unfortunately, supercapacitors store a relatively small amount of energy and are thus more appropriate to high-output low-duration implementations. Over time, capacity is improving, but the overlap between supercapacitors and traditional batteries is still relatively small – powertools and a few other small gadgets. Cost is still a limiting factor as well.
Longer term, supercapacitors have a lot of potential in energy recovery applications – for example, regenerative braking. But, beware startups promising orders of magnitude advances in supercapacitor technology. There are many out there making such claims, and none have been able to demonstrate solid evidence of their viability.
The reality is that, barring some “out of left field” advance, battery technology looks set to improve in relatively small steps as materials science advances, nanotech manufacturing processes improve, and overall volume drives down costs. An electric car that can charge in seconds and deliver a 500 mile range seems unlikely in the coming decade. But the more relatively-decent electric cars you buy today, the more realistic that future car becomes. I’m sure Tesla, Nissan, and Fisker would appreciate it as well.