Semiconductors took advantage of a single technology (CMOS) to achieve incremental gains, batteries need new chemistries.
There’s been a battery revolution underway, and it’s been fueled by lithium-ion (Li-ion) batteries. Whether it’s your cell phone or laptop, electric or hybrid vehicle, chances are that there’s a lithium-ion battery powering it. Investments in the technology are huge, with market research firm Avicenne Energy estimating the 2016 market size for Li-ion cells at $23B, with a historical growth rate of 17% (Figure 1).
A quick look at the chart above shows the consumer electronic applications relatively flat, with electric vehicles and buses (including hybrid vehicles) driving the growth. The flatness of the consumer applications is likely the result of two forces: larger volumes and lower average selling prices.
With the rising adoption of electric vehicles, including hybrids, the market for Li-ion targeting these applications is expected to continue its ascent, with a CAGR (compounded annual growth rate) of 17% when measured in MWh (megawatt-hours) of capacity, and 12% when measured in revenue (Figure 2).
Will Li-ion batteries meet the key requirements of higher capacity/weight ratios (gravimetric energy density, measured in Wh/kg) and higher capacity/volume ratios (volumetric energy density, measured in Wh/L) demanded by these applications? Many argue that battery technology now has an inherent growth rate in key metrics (between 5 and 8% annual increase in gravimetric energy density), analogous to Moore’s Law for semiconductors. But can we expect this to continue?
Moore’s Law came from a paper published in 1965 by Intel co-founder Gordon E. Moore, that postulated that the number of transistors that can be packed into a given unit of space would double approximately every two years. It’s been remarkably prescient and has served as a forward-looking predictor by much of the high-tech industry. It is, however, not a physical law. As my former undergraduate advisor, Carver Mead, stated in 2005, “Moore’s law is really about economics. [It is] really about people’s belief system, it’s not a law of physics, it’s about human belief, and when people believe in something, they’ll put energy behind it to make it come to pass.”
I’ll note that it also takes a strong economic need and a technology prone to exponential improvements. CMOS had both. It drove the IT and communication revolution, and the figure of merit was the number of transistors, something that lent itself to miniaturization.
Is it the same for Li-ion technology? As stated earlier, the predicted range of gravimetric improvement is between 5 and 8% per year. While not as high as Moore’s Law’s 40% improvement per year, it is nonetheless an exponential increase, but at a lower rate. The issue is whether the technology lends itself to these continuous improvements.
An interesting difference is that improvements in Li-ion technology are largely driven by different chemistries of the cells. In the Li ion Battery presentation by K. Devaki, the following chart was presented. The orange represents different Li-ion technologies. There is a virtual alphabet soup of Li-ion technologies being deployed and being developed (Figure 3).
The issue facing the battery industry, unlike the semiconductor industry, is that different technologies are required to be deployed for comparatively modest gains. Each technology requires significant time until commercialization. To give an example, the LFP battery (lithium iron phosphate) began development in 1996, when the University of Texas discovered that creating a cathode with phosphate enabled a high current rating and long cycle life. Yet, it was only into this decade that they were commercially deployed.
And yet, batteries continue to offer higher gravimetric and volumetric energy density. Some of the improvements are linear improvements within a chemistry, and some are step function improvements through new chemistries. The next breakthroughs in Li-ion technology may be based on lithium-sulfur and lithium-air. Like Moore’s Law, there is a real limit to this trend line, even at 5-8% per year. It’s difficult to get much more energy into a material than 1 eV per atom. A rough calculation leads to approximately 850 Wh/kg. While impressive and useful, it is just a factor of five better than the Li-Cobalt technology shown in Fig. 3.
But up to that point, it’s Moore’s Law in slow motion.
—Larry Desjardin is a regular contributor to EDN’s Test Café. He served in several R&D and executive management positions with Hewlett-Packard and Agilent Technologies. Contact him at firstname.lastname@example.org.