Since mid-July I’ve been an outspoken advocate for advanced lead-acid battery technology and a fairly vocal critic of lithium-ion technology, which I’ve uncharitably compared to airbrushed centerfolds. Understandably, defenders of the true faith have condemned my heresy. Today I’m going to back up a few steps and try to give new readers a better understanding of where the battery industry has been, where it is now and where I believe it is going in the future. I hope this overview of how the industry has developed will make my reasoning more clear and improve everyone else’s understanding of a complex but very exciting investment sector.
A Brief History
To understand the current state of battery technology, one must first consider the historical needs that gave birth to all invention. Around 250 BC, a clever Babylonian found that a magic genie could be released from a clay pot containing the right combination of lead and acid. During the 1800s, people began to find ways to make the genie do useful work beyond electro-plating and parlor tricks. From there technology progressed rapidly to a point where batteries are now a ubiquitous but largely invisible part of our daily lives. We don’t usually think about batteries until they need to be recharged or replaced, but life would be very different without them.
Until the 1960s, there were two primary classes of batteries: rechargeable lead-acid batteries and disposable dry cells. Lead-acid batteries handled the heavy work like starting cars and providing emergency lighting while dry cells were used for flashlights, toys and consumer goods, including the first wave of cheap transistor radios.
In the mid-70s, maintenance free valve regulated lead-acid (VRLA) batteries were introduced and rapidly became the dominant technology. They worked so well in fact that the level of R&D spending on lead-acid technology plummeted. Shortly thereafter, new rechargeable battery chemistries including nickel cadmium (NiCd), nickel metal hydride (NiMH) and lithium ion (Li-ion) emerged on the scene. Since the new chemistries had tremendous potential utility in portable electronics, R&D spending on those chemistries soared in response to intense consumer demand. That trend continued through the early years of the current decade because lead-acid batteries were generally adequate for the work they needed to do while batteries for portable electronics were still frequently inadequate.
Over the last few years, an entirely new market dynamic has emerged as people have been forced to come to grips with the amount of energy they waste. Today we are witnessing a seismic shift in the storage sector because none of the technologies we relied on in the past is durable enough or robust enough to meet the demands of an energy efficient future. In response to this new market dynamic, companies throughout the energy storage sector have:
- Instituted new research programs to improve the performance and durability of lead-acid batteries;
- Refocused existing research to concentrate on making larger NiCd, NiMH and Li-ion batteries;
- Increased research on new and improved flow battery chemistries; and
- Devoted new resources to physical storage systems like pumped hydro, compressed air and flywheels.
The victors’ spoils will be massive new markets that represent an estimated incremental value of up to $70 billion per year – a whopping 233% increase over current global revenues of $30 billion industrywide.
Critical Performance Metrics
Understanding performance claims in the energy storage sector can be difficult because there are several critical performance metrics including “energy,” or the capacity to do work, which is usually measured in watt-hours (Wh); “power,” or the rate at which work can be performed, which is usually measured in watts (W); and “cycle-life,” or the number of times a device can be discharged and recharged before it needs to be replaced. Another key concept is “energy density,” which quantifies the amount of energy a battery pack can deliver per unit of weight measured in kilograms (kg) or volume measured in liters (l).
If you think in terms of the humble electric golf cart, energy limits the distance you can travel on a single charge, power limits your speed of travel, cycle-life limits the number of rounds of golf you can play before replacing the battery and energy density dictates the size of your battery pack. So performance metrics are easy to understand when they are tied to the requirements of a particular application. But if you try to discuss performance metrics in a vacuum without considering how they relate to a particular application, all you get are confusing gee whiz numbers.
I’ve been studying SEC reports from energy storage companies for several years and believe that investors would be well-served if every company presented summary production, revenue and cost data using a uniform watt-hour metric. The disclosures in the prospectus for the proposed A123 Systems IPO come close to my ideal, but are still not quite there. In my opinion, this simple change would make it far easier for investors to make apples to apples comparisons and truly understand the competitive strengths and weaknesses of widely varied storage technologies. But since fair comparability can really take the edge off a story, standardized disclosure may be a long time coming.
Critical Application Requirements
The biggest challenge facing the energy storage industry is an incredible diversity of needs that precludes even the remote possibility of a silver bullet solution. I couldn’t begin to describe or quantify the global scope of the problem, but a couple of concrete examples may be helpful.
In a light HEV where the principal goal is to use energy from recuperative braking to provide extra boost during acceleration, power and cycle-life are the critical metrics. You need a storage solution that can accept a huge charge over a 10 to 15 second braking interval, deliver that charge over a 10 to 15 second acceleration interval and repeat the process many thousands of times over the life of the vehicle. In a PHEV where the principal goal is to run in electric only mode for 40 or 50 miles and then switch over to an internal combustion engine, energy and power are the critical metrics and cycle-life is fairly unimportant because the average user will not recharge his batteries more than 300 to 500 times in any given year.
Similar disparities are common in the utility industry where power and cycle-life are critical metrics for frequency regulation and short-term grid stabilization, but energy and power are the critical metrics for long discharge periods involving rate arbitrage, renewables leveling and diurnal storage.
In the extreme case of an emergency backup or upgrade deferral system that only kicks in if there is a severe grid disruption, energy and power are the only metrics that matter and cycle-life is almost irrelevant.
Size and weight are mission critical constraints in portable electronic device. They are far less important in motive applications and almost irrelevant in stationary applications. Likewise, high cycle-life and power are critical for light HEVs but expensive overkill for an electric runabout that will only be charged a couple thousand times during its useful life. In the final analysis, the fundamental laws of economics will require that every user pick the storage solution that is best suited to his particular needs and budget.
Two Decades of Li-ion Technology
Sony (NYSE:SNE) first introduced commercial Li-ion batteries in 1991 and there have been huge improvements in safety, power and cycle-life over the last two decades. But each major safety improvement has reduced energy density and increased manufacturing costs.
Sony’s original Li-ion batteries had energy densities approaching 200 Wh/kg, were able to deliver their stored energy in an hour and offered between 500 and 1,000 cycles. In comparison, today’s high-end Li-phosphate and Li-titanate batteries offer energy densities of less than 100 Wh/kg; can deliver their stored energy in three to five minutes and offer useful lives of 5,000 to 20,000 cycles. Between these extremes, the variables are almost endless.
While precise cost comparisons are difficult because nobody uses standardized reporting metrics, the bulk of available data indicates that lithium-cobalt batteries based on Sony’s original chemistry cost $0.45 to $0.55 per Wh and high-end Li-phosphate and Li-titanate batteries can cost upwards of $1.50 per Wh. About the only good price news in the group is Li-polymer batteries that cost about $0.35 per Wh to manufacture.
Battery cost per Wh is not a critical issue when a consumer is shopping for a 50 Wh laptop battery. But it will be the primary market driver when that same consumer is shopping for a 2,000 Wh battery for a Toyota Prius, a 16,000 Wh battery for a Chevy Volt or a 26,000 Wh battery for a Th!nk City runabout. After all, the only place a comma and two or three additional zeros don’t matter is Washington DC.
There is no question that today’s Li-ion batteries offer far better power and cycle-life than Sony’s originals. But gains in one performance metric have always reduced energy while increasing manufacturing costs. Over the last two decades, Li-ion technology has seen incremental improvements of 8% to 10% per year, but it's never seen anything even close to the "Moore's Law" type performance gains so many investors have come to rely on.
Since we have not seen disruptive performance improvements over the last two decades when Li-ion technology was rapidly evolving and research chemists had all the R&D funding they could possibly use, I think it is unreasonable to assume that disruptive performance improvements will arise in the future as a mature technology is scaled up to larger sizes. I am also troubled by recurring reports from natural resource analysts who note that Li-ion batteries require raw materials that are not abundant in North America and may not be abundant anywhere else.
I believe Li-ion is a wonderful technology that has a wealth of potential uses. But it is not and never will be a cheap general-purpose solution for all energy storage needs. Julia Child is rumored to have owned a solid gold frying pan that had incredible thermal uniformity but no economic utility in the average kitchen. I remain convinced that many of the highly touted bulk storage applications for Li-ion technology are in a comparable category, technically feasible but impossibly expensive in the real world of paychecks and budgets.
Three Decades of Lead-Acid Technology
After the invention of VRLA batteries in the mid-70s, research on lead-acid technology plummeted and there were no substantive new research and development projects for almost 30 years. VRLA batteries were adequate for the work they needed to do and without the pain of necessity there was no compelling incentive for new invention.
That dynamic began to change a few years ago when it became obvious that new energy storage solutions would be essential to minimize waste. At that point, researchers once again began to look at new ways to improve lead acid battery performance by integrating new materials and technologies that were developed for use in other sectors during the 30-year period when lead-acid research stagnated. Established lead-acid battery producers funded some of the research work, but Firefly Energy, Axion Power International (NASDAQ:AXPW) and Australia’s Commonwealth Scientific and Industrial Research Organisation (CSIRO) initiated the more ambitious projects.
The Firefly project was spun out of Caterpillar (NYSE:CAT) in 2003 and its goal was to use a carbon foam composite to replace lead current collector grids. Firefly’s hope was that its carbon foam technology would reduce the amount of lead used in a battery, minimize lead that was not chemically active and improve energy density. Over the last five years, the Firefly project has grown from a pure R&D initiative to a manufacturing and commercialization partnership between Firefly and C&D Technologies (CHP) that was announced at the end of October. While pricing information hasn’t been released yet, the available performance data indicates that the new Oasis battery will offer a 40% to 50% increase in energy density, higher power and up to 800 cycles at an 80% depth of discharge. My current sense is that the Oasis battery will probably cost $0.20 to $0.30 per Wh, or twice as much as a normal lead-acid battery, but offer four times the performance in suitable applications.
The Axion project was also initiated in 2003 and its goal was to create a true hybrid between a lead-acid battery and a supercapacitor by replacing the lead-based negative electrodes with carbon electrode assemblies. Axion’s hope was that its PbC devices would reduce the amount of lead used in a battery, eliminate sulfation, which is the primary cause of lead-acid battery failure, and bring supercapacitor-like power to the lead-acid world.
Over the last five years, the Axion project has progressed from a pure R&D initiative to a planned commercial rollout that’s expected by mid-2009. While detailed performance and price specifications haven’t been released yet, the available information indicates that Axion’s PbC battery will offer a 400% increase in power and well over 1,200 cycles at a 90% depth of discharge. My sense is that Axion’s PbC batteries will probably cost $0.20 to $0.30 per Wh, or twice as much as a normal lead-acid battery, but offer six to eight times the performance in suitable applications.
The historical details on the CSIRO project are a bit sketchy but the CSIRO ultrabattery appears to have a lot in common with Axion’s PbC battery since both products are a battery-supercapacitor hybrid. While we don’t know much about the design, construction and electrochemistry of the CSIRO ultrabattery, there are some impressive results from a recent 100,000-mile road test in a modified Honda Insight. The bottom line was that the CSIRO device performed flawlessly; got 2.8% less gas mileage because of the added battery weight; but offered a $2,000 cost savings over the factory original NiMH battery.
I am not suggesting that the Firefly, Axion and CSIRO projects embody the pinnacle of lead-acid performance; innovation simply doesn’t work that way. Instead, I believe they’re simply important steps in the ongoing quest for a cheap general-purpose storage solution, But these advances clearly demonstrate that disruptive improvements in lead-acid chemistry are still possible when advanced materials and technologies that were developed in recent years are combined into new products based on inherently cheap lead-acid chemistry. When it comes to cost-effective energy storage, Firefly, Axion and CSIRO have made more progress in five years than the entire Li-ion group has made in two decades. So I think it’s far too early in the game for the press or politicians to be picking a winner.
I’ve spent five years immersed in energy storage because of the work our firm did for Axion. So circumstances and professional standards required that I carefully study the needs of the emerging storage market and the strengths and weaknesses of the leading technologies. The lessons my work taught me beyond any reasonable doubt are:
- Commercial decisions will always be based on detailed studies that carefully weigh the fully loaded cost of storage against the value of the stored energy;
- Consumer decisions will be very sensitive to both front-end costs and back-end energy savings;
- There is no silver bullet solution to the energy storage problem and our future will require the use of several different technologies; and
- The prize will ultimately be shared by dozens of companies instead of being concentrated in one or two.
For the reasons summarized above Li-ion technology has been the headline grabber for the last two decades. During that period the energy requirements of portable electronics have fallen by Moore’s Law multiples and while Li-ion batteries have gotten safer, they’ve also lost energy density and gotten more expensive. For most of the time that Li-ion technology was being actively developed, lead acid technology was the object of benign neglect.
Over the last 5 years, research projects from Firefly, Axion and CSIRO have resulted in disruptive improvements in lead-acid durability and performance. While none of them can claim energy, power and cycle lives that are as good as advanced Li-ion batteries, the size and weight multipliers are now in the 2x to 3x range, rather than the 6x to 8x range that the experts predicted when they first compared advanced Li-ion with conventional lead acid. But what Firefly, Axion and CSIRO lack in performance they more than make up for in price. After all, we Americans have never minded lugging around a few extra pounds if the heavier choice is 40% to 80% cheaper.
In the final analysis I don’t see the future of energy storage as an either-or proposition. I think Li-ion batteries, lead-acid batteries, flow batteries, pumped hydro, compressed air and flywheels will all make important contributions to the energy storage solution. So I believe a balanced portfolio of energy storage stocks is the only sensible approach for investors who don’t have the time or inclination to do their own research. Articles like this one can provide useful ideas, but they should not be relied on as investment advice because every author (including me) has his own agenda, preferences, predilections and prejudices.
As an investor, my goal is to buy low and sell high. Based on five years of work in the sector, I’m convinced that growth in the Li-ion group will be slower than most people expect and growth in the lead-acid group will be faster than most people expect. In the current market, the lead acid group including Exide (XIDE), Enersys (NYSE:ENS), Ultralife (NASDAQ:ULBI), C&D and Axion are trading at far lower valuations than companies in the Li-ion group like Advanced Battery (OTCPK:ABAT), China BAK (NASDAQ:CBAK), Valence (VLNC), Altair (NASDAQ:ALTI) and Ener1 (NASDAQ:HEV). If my basic thesis about differing rates of technological change and sales growth is correct, the companies in the lead-acid group are likely to perform far better over the next few years than the companies in the Li-ion group.
The upcoming IPO from A123 Systems will focus the market’s attention on the storage sector in a whole new way and a rising tide of investor sentiment is certain to lift all of the boats in the marina. Astute investors ought to be doing their boat shopping now.
Disclosure: Author holds a long position in Axion Power International (AXPW) and is a former director of that company.