10Aug 2015

Elon Musks’ Gigafactory comes on line in 2017 to build some 35 GWh annually worth of batteries for Tesla. Thirty-five giga….that sounds like a big number. But wait, thirty-five gigabytes is a small capacity for your disk drive. So let’s put things in perspective in this post.

Tesla’s Gigafactory should be capable of producing by 2020 enough battery packs for some 500,000 electric vehicles. That corresponds to a capacity of about 75,000 kWh for each vehicle, with each pack including about 7,000 individual 18650 cells — see this earlier post on the battery pack of a Tesla S. Quick math reveals that the annual output of the Gigafactory equals a little less than 4 billion 18650 cells annually.  The annual worldwide production of lithium-ion batteries stood in 2014 just around 35 GWh, which means that Tesla intends to effectively double the global production — and this is only for making 500,000 cars. So this begs the question, will there be enough lithium to make all, and I mean all, cars on the road electric? Let’s explore.

First, let’s estimate the total usage requirement of lithium. An individual state-of-the-art 3-Ah 18650 cell weighs about 48 g, and contains on average a mere 2 g of lithium. The lithium is in a salt form suspended in a solvent (the electrolyte) or embedded in the cathode material, often a composite metal oxide of lithium (such as lithium cobalt oxide).  Therefore 35 GWh, or equivalently 4 billion 18650 cells, consume 7,900 metric tons of lithium.

But wait, lithium does not exist naturally in a metallic form — remember, lithium metal is reactive. Lithium is mined in the form of lithium carbonate; by weight, it is 5.2x heavier than lithium. In other words, the demand in 2014 corresponds to 41,000 tons LCE (in the lingo of the mining industry, LCE stands for lithium carbonate equivalent to account for other forms of mined lithium salts such as lithium hydroxides). With a selling price of $8,000 per ton (and rising), one can readily see that lithium mining for batteries only is a $320m annual business growing at 7-8% per year. That’s a pretty decent figure but it pales compared to the market size of other metal commodities such as copper. It’s worth noting here that lithium is used for several other applications such as glass manufacturing. In 2014, the total worldwide demand for lithium exceeded 100,000 tons LCE. The vast majority of presently operating lithium mines are in South America, in particular Chile’s Atacama salt flat and Bolivia. Increasingly, deposits in China and Canada, for example the Whabouchi field in northern Quebec, are being explored for further mining as demand and prices for lithium increase.

I still haven’t answered the key question: is there enough of it? Lithium is a very light element and is fairly abundant in the Earth crust — light elements formed early during the evolution of the Sun. The USGS estimates its abundance at 500 atoms per million atoms of Silicon (the most abundant element in the upper Earth crust), or about 0.05%. By comparison, it is as abundant as zinc and copper, and a million times more abundant than gold. We never questioned whether there is enough gold in the world, did we?

There have been no accurate studies to estimate the world reserves of lithium. SQM, the world’s largest lithium mining company, estimates that the world reserves stand near 100 to 300 million tons LCE, or equivalent to 2,400x the demand that the Gigafactory is expected to generate.

So the brief answer is: don’t worry, there is plenty of lithium. If you are a speculative investor, it may be a good time to buy land in Chile or Bolivia, at least what SQM and other mining companies haven’t bought yet. You see, 007 didn’t get it right in Quantum of Solace. The coup d’etat should have been to corner Bolivia’s lithium salt mines, not its water supply.

02Aug 2015

A statement of this devilish nature during the middle ages would have earned its author a burning at the stake. The mere notion of a “battery” would have probably been in the realm of druids and witches, and reviving anything dead would have been…well, enough said, it is the 21st century.

I will digress today a little and talk about your average primary (i.e., non-rechargeable) battery, the type that Energizer, Duracell, and their competitors sell billions of every year, all of which wind up in the trash bin or recycling centers — excepting the batteries that sit on one’s desk for ages as if somehow they will disappear on their own. How do these batteries die, and when they do, are they really dead? Let’s explore.

For the purpose of this discussion, let’s focus on Alkaline batteries. The typical ones, like AA or AAA, are nominally rated at 1.5 V. In other words, when they are fresh and unused, one would measure 1.5 V at the battery terminals. As the battery is inserted into a gadget and gets used, the voltage at the terminals drops, and fast it does. As the voltage drops, it reaches a point where it is no longer sufficient to power the electronics in the gadget. Often, these gadgets, such as toys, employ inexpensive electronics. This means that these electronics do not employ the most modern electronics circuitry. This is parlance for electronics that are not low-voltage and low-power. In other words, these inexpensive electronics draw more current than they need to, and they operate at higher voltages than they should — all in the spirit of saving costs. But these operating requirements place a bigger burden on the battery, the result of which the battery’s voltage drains rapidly and meets an early death.

It should be apparent to the reader that the end of the battery — its “death” —  is now defined as the time at which its terminal voltage is no longer able to power the electronics. This is somewhat subjective because that clearly depends on the quality and sophistication of the electronics in your gadget. Usually, many inexpensive electronics begin to stop operating somewhere between 1.2 V and 1.35 V. Very rarely, one may see electronics get lower in operating voltages but such gadgets would most likely be associated with higher price points, and could very well just use an embedded lithium-ion battery to project an image of a “good” product.

Looking at the Energizer E91 specification sheet, one immediately can observe that this battery has a life of less than 2 hours to hit 1.3 V, and 3 hours to hit 1.2 V (assuming a discharge current of 250 mA). At this point, the electronics begin to stop operating; the cheap display on your child’s toy begins to fade, and voila, you pronounce the battery dead and discard it.

But wait! Is it really true that the battery is dead? Again, it is a matter of definition. For a helpless parent trying to appease a screaming child, the battery is DEAD. But to some engineers and entrepreneurs, they will be quick to observe that this battery continues to hold a lot of charge and energy. Looking at the voltage chart above for the E91, the area under the red curve is the amount of “energy” that the battery holds. So it becomes immediately clear that if the battery is declared dead at 1.2 V, it continues to hold about 75% of its original energy. This is a lot!

So the magic question becomes how to access this extra energy well? and how to do so in a cost-effective and reliable manner? This is where I will put a plug for the company Batteroo Inc.. The team figured out an elegant solution to put a very thin reusable sleeve around the presumed dead battery with low-power electronics that will “boost and regulate” the raw terminal voltage of the battery back up to a higher voltage, say 1.5 V, sufficient now to operate a gadget. This has an effect of reviving this “dead” battery and substantially extending its life. I love clever and simple ideas! Batteroo’s challenge is now to fight off the battery vendors who will not be pleased with selling fewer batteries.

12Jun 2015

Kevin Gibb of TechInsights published recently an article in EE Times that shows a teardown of the lithium-ion battery used inside the iPhone 6 Plus. While the teardown and the article seemed motivated by determining the cost of this battery — somewhere near $4.00 — it contained some very nice cross sectional photographs taken using optical and electron microscopy of the various layers that make the iPhone 6 Plus.  Anything that carries an Apple logo seems to attract a lot of attention, but the battery inside the iPhone 6 Plus is similar in performance and structure to many other Li+ (lithium-ion) polymer batteries used in mobile devices.  For example, the battery capacity of the iPhone 6 Plus is rated at 2.915 Ah, within a rounding error of the capacity of batteries used in the Sony Xperia Z3 and Z3+, the LG G3 and G4. Let’s use this very nice teardown report of the iPhone 6 Plus battery to shed more light onto the inner structure of a lithium-ion battery and its workings especially in view of fast charging.

I described in an earlier post the various shapes of a lithium-ion battery. A 18650 cell is encased in a metallic cylinder, whereas a polymer one is a thin and flat pancake-like without any external metallic protection. Yet, the insides are nearly identical, all consisting of a set of electrodes called anodes opposing another set of electrodes called the cathodes with both sets of electrodes separated by a porous membrane called a — hold your breath — “separator.” The first picture below shows a cross section of the polymer battery inside the iPhone 6 Plus viewed through an optical microscope. For reference purposes, the iPhone 6 Plus battery is approximately 3 mm thick.

In mobile devices, the vast majority of batteries use a metal oxide called lithium-cobalt-oxide (LCO) deposited on an aluminum backplate to act as the cathode (the positive electrode during charging). You can see the bright white aluminum back layer in the photo above, but it is difficult to see the LCO layer at this magnification. The anode is nearly always made of a thin carbon graphite layer deposited on top of a copper backplate. There is a very thin separator layer that sits between each set of anode/cathode layers. During charging, the ions, yes, the lithium ions, travel from the cathode through the porous separator to the anode, and embed themselves inside the graphite. As every skilled engineer should know, charge balance means that there is an opposing current made of electrons that goes through the external circuit between the anode and the cathode. This means that maintaining a low electrical external path resistance is essential to the operation of the battery — one of the reasons why aluminum and copper conductors are used.

The photo above shows a stack of alternating layers of anodes and cathodes. There are 11 anode/cathode layer pairs, which means the pitch is approximately 275 microns. This particular construction is unique to LG Chem with a stack of parallel layers. Other battery manufacturers use what is known as a jelly-roll, with the layers of anodes and cathodes rolled together like a cigar. This mechanical structure, while seemingly immaterial to the novice, plays a big role in the distribution of electrical current inside the battery, and consequently the governing degradation mechanisms. Let’s zoom in a little more.

The second photograph shows a scanning electron micrograph (SEM) of two sets of anode/cathode layers. Now we can see the individual structural materials. The separator is typically near 10 to 20 microns in thickness. The graphite and LCO layers are often around 50 microns but can vary depending on battery capacity and current rating. This SEM now shows that the LCO layer is granular in nature. The graphite layer is granular too.  The grains, varying in size from a few to several microns in diameter, consist of crystalline layers — a lattice-like — where the lithium ions can embed themselves. In charging, they embed themselves in the graphite lattice, and in discharge, in the LCO lattice. The graphite lattice is pictured next using a transmission electron micrograph (TEM). The lattice is made of atomic layers that are a mere 0.34 nm apart — think of it as atomic Swiss cheese.

The LCO and graphite have a limited capacity of how many lithium ions they can “hold” inside their lattice. This determines the amount of LCO and graphite material that is needed for a battery of a given capacity, i.e., of a given mAh. This in turn determines the energy density. Well, sort of, because there is another kink in the design of the battery, and that is the size of the grains (both LCO and graphite) and how tightly packed they are in the electrode layers. If the grains are too tightly packed, then the lithium ions will find it difficult to travel through all the grains; in other words, the maximum current capability of the battery is impaired. So you are hopefully getting a little taste of the various compromises a battery designer needs to go through….and we haven’t even yet gotten to charging.

Now let’s talk about the headaches that come with degradation of this structure especially with fast charging. High capacity and/or faster charging means a lot of ions need to zip in and out of the anode layers — since the anode is primarily responsible for storing the ions during charging. Think of cars on a highway at peak rush hours….it’s not easy; every pothole in the road now contributes to traffic flow. For example, small perturbances in the uniformity of grains means more ions will flow into one grain vs. another, thus creating differences in current density, and excessive stress on some grains (ultimately causing mechanical fracturing of the graphite lattice and loss of capacity). Small disturbances in the voltage distribution across the layers means some portions of the stack may see a potential difference between the anode and cathode that will promote the metallic plating of lithium — a very detrimental failure mode especially present with faster charging. These are only but two examples of the degradation mechanisms. There are several more that are becoming prevalent in modern batteries with high energy density and faster charging. The task is to tame these degradation mechanisms to extract maximum performance, and that is now falling onto the next frontier of clever charging algorithms — and that is what we do at Qnovo.

Fast charging a battery clearly involves a high degree of optimization in order to manage the large flow of ions. Historically, battery vendors did it while sacrificing grain size, or packing density of grain; in other words sacrificing energy density and overall battery capacity. This compromise is no longer acceptable.

13May 2015

It’s all over the media: Moore’s Law just turned 50! What is Moore’s Law? It’s more an observation than a law, but it has stuck around, now 50 years, that we think it is a law, like gravity.

On 19 April 1965, Gordon Moore, at the time the head of R&D at Fairchild Semiconductor, and later the CEO of Intel, made an observation turned prophecy. He predicted at the time* that the number of components and transistors on an integrated circuit (IC or chips) would double every approximately 18 months while holding the cost of the chip constant. In layman’s terms, it means that the industry will be able to double the complexity, and hence the computational power, of these chips every roughly 1.5 years without increasing the cost of the function. And for 50 years, this prediction held remarkably well and has been hailed by the tech and semiconductor industry.

Its implications are just spectacular. Next time you hold a smartphone in your hand, pause for a moment and think about the fact that it is thousands of times more capable than the Apollo Guidance Computer that landed Neil Armstrong and Buzz Aldrin on the moon. Moore’s Law is in some ways the new Bible of the tech industry with an implicit expectation that all new technologies ought to follow this trend. But is that true? and specifically, is it and will it be true of energy storage and battery systems?

Left: Processor power measured in MIPS shown on a logarithmic scale         Right: Battery energy density in Wh/l

The answer in a nutshell is a big fat NO! Whereas history has shown that semiconductors follow Moore’s Law, that same history shows that the trend in batteries is closer to the progress of gastropods, hence, Snail’s Law. The two figures above illustrate the difference. From 1995 to 2015, the computational power of processors made by Intel increased by a factor 300X, effectively doubling every 2 years. In contrast, over that same period of 20 years, the energy density of lithium-ion batteries increased by 4X, or less than 7% annually.  No one disputes this fact because most consumers complain about their batteries, and few, if any, complain about the processor or the electronics in their devices. But why is that? They both involve materials and manufacturing, yet the differences are stark.

It boils down to a balance between the laws of physics and economics. The laws of physics dictate the amount of technical improvement that is possible given a scientific and/or engineering problem. In the case of semiconductors, these were the laws of physics that dictated the shrinking of the dimensions of transistors in a silicon chip. Back in 1965, these transistors did not operate anywhere near the fundamental limits of materials or equipment. So physics were not the limiting factor in this balance, but economics were. In other words, the R&D and manufacturing costs associated with shrinking transistors had to increase at a lesser pace than the technical performance of these transistors. Under such a circumstance, these added costs were amortized over a rapidly increasing technical performance, and hence the benefit of Moore’s Law: more performance for the same cost point. Said differently, shrink the dimensions more, get more benefits, and this equation becomes seemingly a virtuous circle….that is until it starts to hit the limits of physics, at which point the balance tips — something that the industry may be soon facing.

For batteries, that balance between technical limits and economics was really never in place. First, the cost of R&D and manufacturing was not offset by increasing performance, in particular, energy density. In other words, every increase in energy density manifested itself initially as an increase in unit cost. So there really was never an equivalent to Moore’s Law’s cost-constancy. As a matter of fact, as we examine closely the economics of lithium-ion batteries over the past 20 years, we find that the cost of these batteries declines as a function of cumulative production volumes, not annual production volumes. This is a much slower cost curve and is partly responsible for why raising R&D investments for battery research does not make a lot of sense to battery manufacturers. There is more supporting evidence in the fact that battery vendors live on single-digit gross margins, whereas many semiconductor companies have gross margins close to 50% — i.e., much better profitability.

Second, lithium-ion batteries are already hitting some serious material and physical limits. The presently used material systems in lithium-ion batteries seem to saturate right around 650-700 Wh/l. Going above these figures means higher R&D and manufacturing investments for new materials, and these costs are difficult to amortize.

The result is that energy density begins to level off or improve at an increasingly slower rate. Yes, a breakthrough from a university research program or an innovative company may change that, but history has shown that such breakthroughs don’t come from wishful thinking, but rather from years and billions of dollars in research, both of which are becoming scarcities in batteries.

But that may not be such a bad thing. When technologies begin to level off, cost pressures rise immensely as better manufacturing methods are introduced and as more competitors, especially in low-cost geographies, join the fray. So that means costs will drop rapidly — this perhaps may be the implicit corollary and inverse of Moore’s Law. In mathematical form, we can anecdotally write this as Snail’s Law = 1/[Moore’s Law].

In summary, I believe that the battery industry is entering a new era with accelerating cost pressures, accompanied with a shift to improving the performance of the entire battery system that includes the individual cells, the control electronics, algorithms and software. And that will bode well to companies that are skilled in this system integration exercise , regardless of the application and end-market.

*You can read here Dr. Moore’s original article from 1965 reprinted in the Proceedings of the IEEE
11Mar 2015

Did you know that the European semiconductor giant ST Micro makes rechargeable lithium-ion batteries? Probably not. You don’t believe me; go ahead and google “ST Micro thin-film batteries.” It’s really tiny. It  is a square that measures one inch (25.7 mm) on a side, and is a mere 0.2 mm thick (that’s about the thickness of a human hair). But before you jump out of your seat to order one, let me tell you that it has a capacity of 0.7 mAh. That’s right, that’s 0.7 milli Ah. It will take about 2,600 of these little cells to give a capacity equivalent to the iPhone 6 battery.  But they are useful for applications that require very little power and energy, such as RFID tags or smart cards.

The cells from ST Micro and other suppliers such as Cymbet and Front Edge Technology represent a new category of rechargeable lithium-ion batteries that are called solid-state thin-film batteries. The name says it all. They are solid-state, in other words, no gels or liquids inside the structure. They are thin-film, in other words, made of very thin layers (films) of materials. Naturally, this implies that they can be manufactured in similar ways to semiconductor chips. This is a powerful argument for manufacturing with high precision yet delivering extremely low cost. So if that is the case, why don’t we see them more commonly in mobile devices. Before we tackle this question, let’s dive a little into the internal structure of solid-state batteries.

Basic structure of a lithium-ion battery includes two electrodes and an electrolyte in the middle. Courtesy: Wikipedia.

A lithium-ion battery consists fundamentally of two electrodes, an anode and a cathode, sandwiching an electrolyte medium that allows the lithium ions to shuttle back and forth between the electrodes;  in battery parlance, it has to be electrically conductive to the ions. The anode and the cathode are commonly made of carbon and lithium-cobalt-oxide (LCO), both of which are solid materials that can be layered down using semiconductor-like techniques. But the electrolyte is usually a liquid or a gel…hence, it defies our stated objective of thin layer deposition. The hunt has been for decades now to find a electrolyte that is suitable for the transport of ions through it, yet be made of a solid material. Unfortunately, very few candidates present themselves so far as commercially viable — but that has not deterred small and large companies from continuing the search and exploration. Examples include startups such as SEEO in California and Sakti3 in Michigan.

Lithium Phosphorous OxyNitride, or in short LiPON, is a glass initially developed at Oak Ridge National Laboratory in Tennessee and has evolved into the material of choice today for commercially available thin-film solid-state batteries. But it is far from ideal. It exhibits a 1,000X higher resistance to ion flow than do liquid or gel electrolytes.  Not good! That means few ions can shuttle back and forth consequently limiting the capacity of the battery to a very small figure, usually on the order of mAh or less. Other exotic candidates include zirconia-based ceramics but I am not aware of any commercial deployment. The result is that the energy density of these cells is low: for the ST Micro cell, it is a meager 20 Wh/l, or 30X worse than state-of-the-art lithium-ion polymer batteries.

The other challenge is cost, presumably driven by the lack of manufacturing scale, potentially low manufacturing yields, and the high cost of the exotic materials. Presently, a small solid-state cell can retail for $10 – $30 each. That works out to more than $5,000 per Wh vs. $0.20 per Wh for commercial polymer batteries.  But some of the new startups are trying to change this and reduce the cost by several orders of magnitude.

But on the upside, solid-state cells typically exhibit long cycle life and an excellent safety performance — they are not prone to fire the same way liquid or gel electrolytes are.

In summary, solid-state thin-film batteries present a very attractive story but much research and exploration in materials need to be completed first. I will continue to applaud for more breakthroughs in this area but I don’t see one yet on the horizon.