Society of Automotive Engineers (SAE)
2012 Powertrain Electric Motors Symposium
Good morning. My name is Larry Thomas from Primet Precision Materials.
It’s a real pleasure to be here with you today to talk about rare Earth’s and roast beef and to give you my perspectives on why process technology matters.
First let me introduce you to my firm. Primet is a small company located in Ithaca New York. We’re focused on developing process technology making highly engineered particles. For the last three years, we’ve been focused exclusively on electrode materials for lithium-ion batteries.
I want to say upfront, I’m not an expert in rare earths, and most of what I know about them, I learned in a 30 minute conversation with Jack. But I’m going to try to tie together the perspectives from the other talks this morning, and talk about a common problem that underlies all of the issues that were discussed for rare earths and battery materials, and it’s not something that usually gets discussed at sessions like this.
Let me start by showing you pictures of two cars. On the left in this slide, you see a Cadillac Escalade with the platinum trim level. And on the right is an entry-level subcompact from the opposite end of the General Motors line, a Chevy Aveo.
Now let me ask you, when you look at these cars, what numbers would you use to describe them? Rare Earths and Roast Beef: Why process technology matters SAE Symposium April 2012 Larry Thomas President & CEO
If you’re looking at this as a consumer, you’re likely to use numbers like MSRP or miles per gallon.
If you’re looking at them as an automotive engineer, you might use numbers like horsepower or curb weight.
Me, I’ve spent my whole career in the chemical industry. And so I describe these vehicles using the same numbers that people from the chemical industry use to describe every transaction between us and our customers. We do business in dollars per pound. And when you look at these cars that way, you see that the Cadillac Escalade costs $15 a pound, while the Chevy costs about five.
This doesn’t make sense to me. Both cars are made from the same raw materials — steel, aluminum, plastics, glass – in about the same proportion. So why should one costs as much as filet mignon while the other one costs about the same as lean ground beef?
Come to think of it, why is there such a big difference between the price of filet mignon and ground beef? After all, they both come from the same cow.
Well, let’s take a look at the cow. We see that the tenderloin is a relatively small part of the cow. There’s a lot more of it that’s chuck then there is that’s tenderloin.
So, the engineer in me looks at the weight percent of the cow that’s chuck and compares it to the weight percent of the cow that’s tenderloin. And with that I can draw a correlation between the price of chuck and the price that I’d expect to pay for tenderloin. And so for the purposes of this talk, I’m going to do that calculation and call this the “scarcity price” of tenderloin as compared to chuck.
So you’re probably wondering how that relates to mining rare earths. To explain that, let me do a thought experiment.
Let’s look at the simplest mining operation there is. I can go down to a building supply yard and buy sand for seven dollars a ton. Let’s assume that nobody in the sand industry ever makes any money, and so the cost of shoveling sand into a truck is seven dollars per ton. That’s the simplest and cheapest mining operation on the planet.
Rare earths are a bit more complex that that. But let’s keep it simple.
Jack tells me that a typical deposit of rare earths has about two weight percent rare earths. For the purposes of this thought experiment I’m going to assume that the rest of this is sand. And once again I’m going to assume that all I have to do is stick a shovel in the ground, pour the contents of that shovel into a truck, and the constituents of that truck will magically separate themselves into a big pile of sand and a bunch of little piles of the valuable minerals.
If the cost of putting that shovel in the ground and pouring its contents into the truck is seven dollars a ton, then I should be able to calculate a scarcity price for each of the rare Earth constituents, just like I did for the constituent parts of the cow.
And so applying my formula, I see that there’s about 100x as much sand as there is cerium, so I can calculate a scarcity price for cerium that’s about 100x the price of the sand. Then I continue on down for each constituent based on its percentage of the sand, and so on down to the smallest ones in the mixture.
When I do this, we see a series of scarcity prices that follows a log curve. That shouldn’t surprise anyone. The materials with the lowest concentrations in the ore body have the highest scarcity price.
But when I compare these scarcity prices to the market price for these minerals, I find that the market price doesn’t track with the scarcity price. Yes, they both follow logarithmic curves, but you can see the magnitude of the numbers of the market price is significantly higher than what we would predict by looking at the scarcity price.
The difference between the two is what I’m going to call a processing premium. And this is clearly an oversimplification but it’s a thought experiment so I hope you’ll allow me that. This processing premium, I’m going to throw in all of the costs beyond the cost of sticking a shovel in the ground. So that’s going to cover the capital and operating costs of my separation operation as well as the costs of preparing, permitting, and remediating the site after the mining takes place.
So let’s take a look at these numbers, and ask ourselves if it makes sense.
As I said, I’m not an expert in this field, but we know that the industry uses an iterative process to isolate and purify the various rare earths in the mix. James Gadolin used fractional crystallization when he first explored the Lanthanide series in the 18th Century. About 100 years ago the industry moved on to solvent exchange and extraction. And that’s basically the same process that’s still used today.
It takes about 30 solvent extractions to isolate the lighter rare earths, and another 50 extractions to get to the heaviest elements. And so, as a chemical engineer, I’m not surprised to see this kind of a log curve in the price premium as we go through progressively finer separations of progressively lower and lower concentration species.
I also know this isn’t the kind of plant I’d want to own or operate, because it looks like it’s going to be a very tedious operation that takes a lot of capital, and consumes a lot of energy.
But let me ask you: We’ve heard a lot of talk about the cost and availability of rare earths. As a room full of engineers, if you were trying to solve the problem of the cost and availability of rare Earth’s where would you focus your efforts to have the greatest leverage on reducing the price of this very rare roast beef?
Would you focus your attention on the cost of the ore itself? Clearly there is a multiplier effect associated with that, but it’s still a relatively small contribution to the overall cost. Focusing on this is the equivalent on trying to reduce the price of tenderloin by trying to find cheaper cows.
Or, you could focus your attention on trying to find ore bodies that have a higher percentage of the desirable minerals. That’s the equivalent of trying to find cows with bigger tenderloins.
I think any engineer looking at this chart would do their sensitivity analysis and say that the greatest leverage would come from focusing their attention on the processing cost to extract the valuable minerals from the ore body.
But if you listen to the politicians, and the media, and even the people in the industry itself, you don’t hear that conversation. Instead, you hear fear and panic about Chinese control of the reserves and the production of rare earths.
Does that make any sense? Not if I look at this data from the USGS. China has a lot of the proven reserves in the world, but not all of them. If control of reserves was the most critical factor, then these two pie charts wouldn’t be so radically different.
So why the disconnect?
I’d like to try to answer that question for you by taking a turn through the other set of metals we’ve talked about today, the electrode materials for lithium-ion batteries.
So here’s the battery pack from a Chevy volt. It’s a 465-pound system that GM buys for about $700 per kilowatt hour. I do my chemical industry math, and I figure out that this pack costs $25 per pound. We’re far beyond the Cadillac Escalade or the filet mignon here. Now, we’re into the territory of Maserati’s and hand-rubbed Kobe beef.
And so like good engineers when we’re tasked with reducing the cost of the system, we do our Pareto analysis and find that the largest line item in the bill of materials for the battery pack is the cathode powder – between 15 and 20% of the cost of the fully assembled system.
When we look at the cost of any chemical product, we know that the largest part of its cost is the cost of the raw materials that go into it. Typically in chemicals, your raw materials are about 50% of the cost of the finished product.
So the lithium-ion battery industry has been on a 20-year journey to replace the expensive and volatile cobalt in its cathodes with lower-cost ores, things like nickel, aluminum, manganese and iron. What could be cheaper than iron?
And we’ve been successful. Just ten years ago, almost every lithium ion battery in the world had a cobalt-based cathode. Today, lithium batteries use a variety of cathode materials – a veritable alphabet soup of transition metals in various combinations.
But a strange thing has happened. Even though we’ve been successful at replacing expensive cobalt with less expensive ores, the price of cathode materials hasn’t come down. In fact, in many cases, the price of the next generation cathode materials based on these cheaper ores is actually higher than the “expensive” cobalt that they’ve replaced.
Why is that? The base ores are cheaper, that’s for sure. But in order to make these cheap ores into good battery materials, the electrochemists have had to design more and more complex crystals and structures to get those materials to perform.
And that means that those cheap ores have to go through a whole series of transformations that are progressively more exacting to make these very demanding crystal structures.
And so for every decrease in raw material cost, we’ve seen an offsetting increase in processing cost.
How did that happen? From my perspective, it’s because the lithium battery industry is dominated by chemists. And when you ask a chemist to solve a problem of cost or performance, they’re going to give you a new molecule. It’s what they know how to do.
It’s just like when you ask the beef industry to solve the problem of the cost of tenderloin, they’re going to solve it by doing what they know how to do: finding cheaper cows with bigger tenderloins.
And in the case of rare earths, you’re depending on mining industry to be your supplier of specialty chemicals. And when you go to the mining industry with a cost problem, they’re going to solve it by focusing on what they know how to do: find cheaper mining sites, and find ore bodies with better assay results. Cheaper cows, with bigger tenderloins.
And what’s missing in all of this, is a focus on the process technology to turn the basic materials into value added products that the industry needs.
This slide, apparently, hasn’t won me any friends at the Department of Energy. But when I put it together, I didn’t mean it as a criticism. I meant as a commentary on an inherent bias that permeates the battery industry.
We are an industry controlled by chemists. So we try to solve our problems with new molecules.
And that bias carries through, so when the Department of Energy decides where to allocate its research funding, you can see it’s heavily geared towards new molecules and almost completely ignores process technology.
That’s not a fault of the DOE. If you go to any battery conference you’ll hear endless conversations about molecules and electrochemical performance, and you’ll almost never hear a conversation about process.
And I think the same bias holds true for rare earths.
And so when we look at value-added materials, I hope you can see that it has very little to do with the source of the basic raw materials, and it has everything to do with the process technology needed to turn those basic materials into the value added products that our industry needs.
If control of reserves determined control of the value added products, then the countries with the largest deposits of lithium, cobalt, and manganese would also be the biggest manufacturers of value added materials for the lithium battery industry.
They’re not. These countries among them don’t make a single pound of electrode materials, and these countries among them don’t make a single commercial lithium ion battery. It’s the same as in rare earths, where the reserves are abundant around the world but the process technology to separate them from those reserves is resident and practiced only in China.
So my message to you is simple: Process technology matters, and it matters now.
If you leave the industries that support you to their own devices, they’ll focus on the issues they know how to solve, and in many cases, they’re going to make your problems worse.
The molecules you need are worthless until you can make them reproducibly in large volume at low cost. That means process technology, not the control of the basic ores, sets the price and the availability of the value-added material.
So it’s up to you. You need to demand from your suppliers, from your industry, from the government, and from the universities that support you with research, to focus their resources on innovations in process technology.
Because that is what is going to drive your long-term costs and availability, and ultimately, your success. And that’s what Primet Precision Materials is trying to address.