Friday, August 19, 2011

More On the Coming Palladium Paradigm Shift

In early December, 2010, I pointed that there will be a huge paradigm shift in the supply/demand of the global palladium and platinum market. Norilsk Nickel (NILSY.PK), the largest nickel mine in the world who is also responsible for 45% of the world's palladium and 10% of platinum, is abandoning the traditional pyrometallurgy process and adopting the more efficient Activox Process, hydrometallurgy that extracts base metals from mineral ores by chemical leaching. The change is necessary due to severe pollution and deteriorating ore grade.

Knowing the physics in mining, I saw something with hugely significance: While the new technology can extract nickel and copper at near 100% efficiency at low cost, it also means that the chemically stable PGM contents, platinum and palladium, will be extremely difficult to extract. Norilsk will cease to be a major supplier of palladium and platinum. The supply disruption will cause a panic in the global market, sending the price of palladium to the sky!!!

After my article was published, some folks raised some questions. I have done more research on this issue and collected more information. The more I looked at it the more I am convinced of the unescapable conclusion. I will summarize the important points here:

1. Will Norilsk Nickel switch to Activox Process if they could lose PGM revenue?

They have to switch! Maintaining status quo is not sustainable:

a. Due to heavy pollution thanks to the traditional pyrometallurgy, Norilsk City tops the world's most polluted places. 98% of the people are sick. There is not a single live tree within a 30 miles radius. The company is paying 2.6 times higher salary for mining workers, but still can not retain the workforce. If the cost is losing your heath and lose the ability to to raise heathy children, who would want to work in such a filthy place? The government would not turn a blind eye forever.

b. Pyrometallurgy is extremely energy intensive. Even though the technology can extract up to 70% of the base metals and palladium, higher energy cost and deteriorating ore grade means the process could become un-profitable, even if you don't consider the cost to the environment.

c. Activox Process consumes much less energy. The chemicals can be recycled and reused. Not to mention that it extracts nearly 100% of the base metals, versus 70% in pyrometallurgy. The lower cost, more efficient extraction of nickel and copper more than offset loss of PGM revenue.

d. All information indicates that Norilsk is making the switch. They paid US$6.5B cash to acquire LionOre to obtain the Activox Process technology, and rename it to Norilsk Process. They announced aggressive projects to improve mining process. They announced that beginning in 2013, the sulphur emission will be cut from nearly one million tons a year, to only 0.2 million. Such drastic pollution reduction can only happen with the switch to Activox.

Although in the same investor presentation, Norilsk projected continued palladium and platinum production at near flat level. Many readers quoted that report and criticized my conclusion. My response is that Norilsk did not develop the technology, it purchased it. The investor report was prepared by high level management who know very little of the actual physics in mining process. From their point of view, whatever is in the ore, they assume it can be extracted. Whatever technology that do not exist today to extract certain metal content, they assume will become available some years down the road. When making long term projections you are not encumbed by limitation of what is feasible today, but you aim for the long term goal. So I don't blame Norilsk for making the projection that they are going to continue to produce palladium. But the projection is too rosy. It is unrealistic.

But investors need to scrutinize what can or can not be done today or in the near future. We have to get to the intimate details to dig out the facts. With known physics, I do NOT see how they can continue to extract PGM metals once they switch to Activox.

2. Is it really impossible to extract PGM metals once Activox Process is adopted

We have to look at the basic physics of mining process. After mineral ores are collecetd from a mine, the first step, called milling, is to crush the materials into small particles so the rocks and the metal containing grains can then be separated. In traditional mining process, a technology with over a hundred years of history is used, called Froth Flotation. Let me explain it.

Different material stick to oil or other liquid differently. Some will soak in the liquid, some will form oil like droplets as they don't stick. Using this property, we can pour the crushed mineral grains into certain water solutions, and then blow bubbled from the bottom. The metal grains will more likely stick to the bubbles and be bought to the surface, while rock particles remain on the bottom. We can then skim off the top layer which contains metal rich concentrates. This process is called froth flotation.

Mining engineers will tell you that for effective froth flotation, the ores must be grinded to proper particle size. If it is grinded too coarsely, then the metals and rocks are still mixed in the same particles. But if it is over-grinded, it results in particles so tiny that they are almost equally likely to stick to the bubbles, thus it is unable to separate metal and non-metal content. The over-grinding is also called sliming.

How much grinding is over-grinding? This paper and many others suggest that roughly 30 to 80 microns is the ideal particle size for optimum froth flotation recovery. For particles 10 microns or less, the recovery rate quickly falls off.

Chemical leaching, like Activox Process, has a different grinding requirement. For effective leaching, the ores must be ultra fine grinded so as to thoroughly expose the metals to the chemical solution. Literatures on Activox, like this, and this, describe that the ores are grinded to 10 microns or less before feeding into the leach process.

Based on the flow chart, the PGM metals are left in the solid leach residue. The material is then send to a Froth Flotation process to attempt to extract the PGM particles. That was the original design intention. In the actual Tati Nickel demostration plant, the PGM flotation unit was never actually built to test the feasibility of this extra PGM extraction step.

In Norilsk ores, 99.98% of the metal content is nickel and copper. Platinum and palladium is only 0.02%. Starting with 10 microns metal particles, going through chemical leaching which dis-solves 99.98% of the nickel and copper, there is just a tiny bit of the original metal particles remaining. I calculated the resulting PGM particles will be less than 0.6 microns in diameter.

The 0.6 microns number assumes that during the chemical leaching, metal particles do not break down into bits, but remain whole grains. Most likely the particles do break into bits, thus the PGM content probably will become such tiny dusts that they are simply lost within the leach waste. It's just not possibel to recover anything through the conventional froth flotation method, which requires 30 to 120 microns particles. There is no other known technology to pick up such metal dusts efficiently from the waste material.

So the inevitable conclusion, based on physics, is: Norilsk Nickel can NOT produce platinum and palladium any more once they move towards Activox. But they must switch over to the process!

3. How do you leverage the opportunity?

Buy any palladium coins or metal bars you can find; Purchase the PALL and PPLT, which are ETFs backed by physical palladium and platinum metals. But more attractive is to buy shares of the world's only primary palladium producers: Stillwater Mining (SWC) and North American Palladium (PAL).

I encourage metals analysts, and experts from major PGM metal industry users, to really look into this looming palladium supply issue, and consult mining experts to verify my conclusion.

Disclosure: The author is heavily invested in palladium and in stocks of SWC and PAL.

Thursday, August 11, 2011

How Fast Can The US Dollar Collapse?

Recent volatile and relentless surging gold price suggests that we are getting closer to the catastrophic collapse of the US dollar and other likewise fiat currencies of the world.

Remember, no matter how high gold price goes up, it is not the value of gold that's going up, it is the value of the dollar that's going down. Gold is just a commodity that happen to be commonly used as money. Gold does not pay interest or generate income. Gold has no investment value but has storage value. No matter how high gold price goes up, you are not going to be able to buy more stuff with gold, but you merely avoided the loss of value holding ever depreciating dollar.

How fast can the US dollar collapse is the same question as asking how fast can the gold price go up, in dollar terms. Let me try to come up with a model.

The gold price can be written in the following generic formula, which is always correct as long as you choose the proper time dependent function Y(T):

P = P0 * exp(Y(T))

The price of gold should increase or decrease in exponential terms. No matter where the price is, the change is alway in percentages, i.e., the change is reflected in the exponent Y(T). As time elapses, Y(T) increases over time, giving an ever increasing exponent and ever increasing gold price. So let's look at what Y(T) should look like.

In the constant inflation scenary, gold price should increase the same percentage each year, therefore Y(T) would simply be proportional to the time, t, Y(T) = C*T.

But we are not talking about constant inflation. The scenary looks like at first, there is almost no inflation. And then as inflation slowly kicks in, the problem becomes more and more serious, and they print more and more money to spend. Therefore, the higher the inflation pace, the more money they print, and the more money they print, the higher the inflation will go up. It all starts with almost no inflation, and ends with inflation approaching infinity.

In mathematics we can write such correlation simply as that the derivative of Y(T) is directly proportion to Y(T) itself:

dY(T)/dt = C * Y(T)

If you are familiar with math you immediately recognize what it is. It is the exponential function:

Y(T) = EXP(C*(T-T0))

Thus we obtain an elegant math formula of gold price, once you plug in the Y(T):

P = P0 * exp(exp(C*(T-T0)))

Let's pick the two constants. The current gold bull market starts in the beginning of year 2000. Let's use year for the variable T. T0 = 2000. For example today, August 11, 2011 is the 223th day of the year. So for today, T = 2011 + (223/365) = 2011.611

Let's use $200 as starting gold price at the beginning of 2000. That would require that

P = P0 * exp(exp(0)) = $200

Therefore P0 = $200/exp(1) = $73.5

I find that using the constant C = 0.1 gives a perfect gold price match for the past 10 years:

P = $73.50 * exp(exp(0.1*(T-2000)))

For today, T = 2011.611, the formula will give $1791. That's right where we are right now today on the gold price!

Hold your breath. And now, the following chart shows how good this elegant formula matches up with the past eleven years of gold price, and what the future trend will be!

Folks, you still have some time, but clearly time is quickly running out before hyperinflation kicks in in full power. Buy physical precious metals now: Gold, silver, platinum and my favorite, palladium! Buy precious metal and commodity mining stocks: SWC, PAL, PCX, SSRI, CDE, JRCC, PAAS. These are the only things that can protect you in the coming years!