Sunday, June 21, 2009

Rare Earths and Neodymium

Rare metals are "the lifeline of industry"
Japanese Rare Metals Task Force

For an excellent primer on rare earth (RE) metals read this article. They are present in small quantities in almost every technological device from TVs to electric windows. However with the world turning to technology to ameliorate our energy crisis, the demand for REs is set to ramp up. Is there enough REs to go around?

The current state of REs may be summarised by a couple of graphs (Source USGS):

Despite having just 30% of RE reserves, China has a virtual monopoly on the production of REs. The reasons for this are that China's RE mines are relativley high grade and low cost, which led to a collapse in production in the US. To analyse the situation I'm going to drill down into the use of Neodymium-Iron-Boron magnets (NdFeB).


"The global market demands for rare earth resources can be satisfied if the demand for the NdFeB industry is satisfied"
Prof. Feng Hong: CEO, China Rare Earth Office

This is true because REs always occur together and thus is Neodymium is going to be extracted, the others will be also. NdFeB magnets are the fastest growing segment of the rare earths.

The US government studied the supply of REs and published a criticality index:

Tonnage of NdFeB magnets is growing at 16% per year:

What this graph shows is that the majority of NdFeB magnets are now made in China (77% based on the above graph) and this share is growing. The three emerging big users of NdFeB magnets are electric bicycles, hybrid cars and wind turbines.

Electric Bicycles

There are 100 million electric bicycles (EBs) on the road in China today; they outnumber cars 4:1. Of the 23 million EBs sold worldwide last year, 21 million were sold in China. The following graph delineates this (Source):

These EBs contain lightweight, compact, NdFeB magnets for their miniature motors. They use approximately 350grams of NdFeB per bicycle. The chemical formula is (Nd-2-Fe-14-B) so this yields 86g Nd/EB. In 2007, EBs accounted for 5800 tons NdFeB or 13% of the worldwide total. I don't have figures for the neodymium produced in 2008 but if it was the same as 2007, the share would have increased to 18%. The average growth rate for the past 8 years was 35%. If this continues then by 2014 Chinese demand would be 100 million/year or 35000 tons NdFeB.

There does not appear to be an alternative to NdFeB in bicycles due to space and weight considerations. The price of NdFeB magnets are about $40/kg so the bicycle contains $14 of magnets and $1.70 of Nd @ current $20/kg.Nd. Its retail price is $290 and neodymium represents 0.6% of that.

Hybrid Cars

Hybrid electric vehicles (HEVs), plug-in hybrids (PHEVs) and pure electric (EVs) all require an electric motor. At present the vast preponderance of HEVs, including the Prius, use a Permanent Magnet Brushless Direct Current (PMDC) motor. These contain NdFeB magnets and there is no alternative (you could argue Sm-11.2%-Co-53.3%-Fe-27.5% (wt%) but the high reliance on cobalt is an Achilles' heel). The best performance one is a sintered magnet of composition Nd-31%-Dy-4.5%-Co-2%-Fe-61.5%-B-1% (wt%). Dysprosium is critical in this application to give resistance to demagnetization at high temperatures as the magnet reaches temeperatures of 160C.

A motor can be up to 100kW although 55kW is a reasonable figure. For a 55kW motor 0.65kg of Nd-Dy-C0-Fe-B is required which gives 200g Nd/Motor (3.6g/kW) and 30g Dy/Motor (0.55g/kW). A 25kW generator is typically required to recoup braking energy so for analysis purposes a hybrid vehicle contains 288g Nd and 44g Dy. At $20/kg a car contains $5.76 worth of Nd and $110/kg Dy a car contains $4.84 worth of Dy. At $10.60 magnets per car and a selling price of, say, $20,000, magnets represent 0.05% of sticker price.

If you accept John Petersen's analysis that binding targets on fuel standards imply an impending widespread adoption of hybrid technology then it is clear use of motors is set to take off. The current use of hybrids is very small (1m Priuses sold to date). If, for example, half of the EU's 15million new cars were hybrids in 2012; 2160 tons Nd (8802 tons NdFeB) and 330 tons Dy (390 tons Dysprosium oxide). Thus 20% extra Neodymium would have to be produced and 25% more Dysprosium (based on 2005 prodution of 1400 tons).

Dysprosium is especially rare and Dysprosium reserves are almost entirely located in China. Japan is painfully aware of this fact and is scouring the globe looking for Dy deposits while also trying to develop magnets without Dy.

This seems like a good point to stop. There is an alternative to PMDC motors - AC induction motors - which I'll discuss in a follow-up if there's any interest. I'll furthermore analyse the use of NdFeB in wind turbines.

RE miners and investors can judge the direction of the industry if they understand the dynamics of NdFeB. My interim observation is that magnets represent a very small proportion of the sticker price of EBs and cars in particular. This would indicate that they are capable of absorbing a higher Neodymium price and manufacturers would be prepared to accept this for diversity and security of supply. This may make production of REs in the US and elsewhere more economic.

USGS: [1],[2]
Jack Lifton: [1], [2], [3], [4], [5],[6],[7],[8],[9],[10],[11],[12]
GWMG: [1], [2]
Hard Assets Investor: [1],[2],[3]
NdFeB: [1]

Friday, February 13, 2009

A brief survey of the Electric Vehicle landscape for beginners

Welcome to this humble blog. The idea is to make it a little bit easier for people trying to get an overview of the direction of the electric vehicle. I recently became very interested in the area, but had to do a lot of googling to get an insight into the issues associated with the expansion of the electric vehicle market. I'll try to distill this down.

We're all searching for the alternative to gasoline and diesel in the context of climate change and the suggested solutions are wide and varied. Here's a sample of the options:

Primary Cells

Mechanically rechargeable; they would be swapped for fresh metal at a refueling station:
Secondary Cells

You plug them in to recharge them or charge them on the go:

Time for a short detour to put the motor industry in context:

Climate Change Context

Here is a chart of anthropogenic emissions:

Another chart showing more detail:

The 16% attributable to road transport is the most intractable slice of the pie. This is because of the millions of individual emitters, with no centralised facility. Electricity generation lends itself to wind, solar thermal, carbon capture etc; manufacturing to more efficient processing and so on. CO2 is an irreconcilable byproduct of gasoline or diesel cars. The cars can be made more efficient, but any gains will be at the margin.

The only way to make a significant reduction is to change to electric cars, drive less, or switch to public transport. The window for action to avert the worst consequences of climate change is, arguably, not that wide.

Which Battery?

Each of them has their own advantages and disadvantages. The main issue I found that is common to most of them is that of scalability. There were some 53m cars produced in 2007 and that will only grow with the development of the Orient. The issue is climate change and to have an impact on CO2 emissions, electric vehicles must make inroads into that 53m figure.

There is much debate over which battery is the best. I believe any discussion should be impassionate and based on clear-headed analysis of the figures. Resource constraints appear to be a key handicap on the widespread adoption of EVs. Suppose Toyota comes out tomorrow with an all electric NiMH that gets 80mpg, is reliable and much cheaper than gasoline vehicles. Everyone would want one, right? So in what timeframe could Toyota deliver?

Let's look at the figures for how much of the building blocks for each battery type were produced last year(click for a clearer picture):

Would Toyota be able to deliver the goods? We know they use a Nickel Metal Hydride(NiMH) battery. The M in NiMH is predominantly Lanthanum, a rare earth metal. The next 2010 Prius will, for example, have 20kg of Lanthanum per vehicle. The motor in the electric vehicle contains a Neodymium-Iron-Boron(NIB) permanent magnet. A typical magnet contains 100 grams of Dysprosium. Both Dysprosium and Neodymium are rare earth elements. 95% of rare earth elements come from China. China is keenly aware of the value of rare earth elements and institutes export quotas and tarriffs. For the last two years they have lowered the amount they allow to be exported.

I don't know how many vehicles per year Toyota would be able to produce but to put it in context Toyota aim for 1 million hybrids in 2011; down from 2-3 million three years ago.

Here is an illustration of the problem along with commentary on the different battery types(click to read):

"Resource-based view of key components in EV supply chain"
(Height of disks vaguely sized to reflect 2008 production of each metal, the main purpose is qualitative, though)

This graph shows the constraints on Toyota were they to be asked to produce immediately. The question then becomes: As demand increases, can't supply of raw materials simply scale to match it?

In the case of rare earth metals the answer is a qualified no. There are small pockets of rare earth materials outside of China but they are firstly more expensive to extract and would at best only supplement not supplant Chinese dominancy. This would seem to place a ceiling on the expansion of NiMH absent any other consideration. I can't put a figure on it, others might, but it would probably be in the low millions.

Regarding lithium, figures are hard to come by although a recent report (NB: Author has a vested interest) shows demand outstripping supply by 2020. The same report puts the total EV market at 3.5m per year (how much is attributable to li-ion is not delineated). This again does not seem to show lithium impinging materially upon the 53m (it will be much greater by 2020) in the medium term.

Then there is the issue with neodymium and dysprosium in the permanent magnets of most EV motors. At 100 grams a pop and with 100 tons produced last year, it would indicate a current ceiling of 1m Priuses. Here are some good articles to get you up to speed on the difference between DC Brushless(Permanent Magnet) & Induction motors. Chorus claim to have a high efficiency low cost induction motor alternative although independent verification appears scant and it is not in any vehicles at present.

My study of the situation leads me to suspect that high volume production of electric vehicles in the short to medium term is unlikely. Stretching to the longer term, the scalability of lithium-ion and NiMH are dubious. A lead-acid battery or rechargeable zinc battery would thus represent the best options - purely from a resource perspective. There is also the zinc-air option which I haven't looked at but am intuitively skeptical of. I think I've said most of what I wanted to get across and hopefully one or two of you find it useful.

My sources are predominantly secondary or tertiary and all based online so a pinch of salt is perhaps required. I would welcome any corrections, clarifications or criticisms of what I've written - I'm still a neophyte in the EV area.