Taking Stock of Phosphorus and Biofuels

Includes: MOS, NTR
by: Eamon Keane

President Franklin D Roosevelt (1938):

I cannot over-emphasize the importance of phosphorus not only to agriculture and soil conservation, but also the physical health and economic security of the people of the nation.

Dana Cordell, Institute for Sustainable Futures (June, 2008):

Phosphorus is as critical for all modern economies as water.

Prof. Isil Aydin, Dicle University, Turkey (August 2008):

High-grade deposits of phosphate rock are being depleted day by day in the world, hence future sources will have to be derived from low-grade rocks.

DOE, National Algal Biofuels Technology Roadmap, (June 2009):

Phosphorous appears to be an especially contentious issue as there have been calculations that the world‘s supply of phosphate is in danger of running out within the century.

Phosphorus is essential to all life forms. It is essential for DNA, RNA, ATP and many other biological functions. There is no substitute - it's present in every cell. Without phosphorus you would quickly die. Fortunately almost every diet contains more than enough of the RDA of 0.8g. Phosphorus makes up 1% of our bodyweight although this is locked up in cells and bones. The phosphorus in food is utilised and then all of it is excreted.

Phosphorus clearly seemed important to Roosevelt, and the reason was because of fertilizer. Phosphorus is an integral part of fertiliser (NPK (nitrogen, phosphorus, potassium). In most cases the reason for reduced yield is insufficient phosphorus. In natural ecosystems, the most common reason they can't expand is because of a lack of phosphorus. One of the reasons early Americans headed west was because eastern fields became phosphorus deficient.

So why is this relevant now? I'll try to explain the situation with several graphs.

Figure 1 shows how farmers have historically used fertilizer (source). It's clear fertilizer is now almost totally dependent on phosphorus derived from phosphorus rock.

Figure 2 shows the breakdown of the phosphorus sector (source).

Figure 3 shows the market share of countries in the phosphorus rock arena (source).

Figure 4 shows the phosphorus reserves (source). The reserve to production ratio is 89.

Figure 5 shows the global resource base (source). The absence of India is notable as is the striking reliance on Morocco.

Figure 6 shows a suspiciously bell shaped decline in phosphate production in the island of Naaru (source).

Figure 7 shows historical production in the US (source).

Figure 8 shows historical global production (source).

Figure 9 (source) shows that demand for food is recession proof. The last thing people cut down on is food. Seventy five million join the global dinner table every year. They require food (although 800 million don't get enough). Diets in India and China are changing. In the last ten years, annual per capita meat consumption has increased 7kg in China. In the same period, per capita Indian consumption of dairy products has increased 6kg (source). Cattle need feed. All this raises pressure, in a world of relatively fixed arable land, to increase yield. That requires fertilizer.

Figure 10 shows projections for the future. The global peak curve is based on modeling done by Dana Cardell for her almost finished PhD. It comes from Dana Cardell's Hubbert Linearization based on "a conservative analysis using industry data" (source). It should be bell shaped. The demand from 2010-2020 is based on November 2009 projections by Mosaic Corp (NYSE:MOS), a large phosphate, potash and fertilizer company. The 2020-2050 projections continue the 2.25% annual (not compound) growth rate to represent an indicative demand scenario. The global population is on the right axis (source).

Figure 11 shows the phosphorus cycle. Again this was loosely adapted from Dana Cardell (source). The parts in read are losses which are opportunities for recycling. The green parts are where recycling takes place.

Figure 12 shows the breakdown of global phosphorus consumption (source).

Figure 12 does not include virtual phosphorus. For example, the US exports a lot of corn and wheat. While I haven't netted the figures for exports minus imports I suspect it is strongly positive. The virtual phosphorus is only about 25% of the applied fertilizer due to losses within the US (crop residues, wastewater, etc.). Figure 13 shows the international trade in corn (source). (As an aside, this also means the export of vast quantities of virtual water.)

The Phosphate Supply Situation

Don Tompkins, Facility Manager at Mosaic's Four Corners Mine in Florida (2005):

We used to go 15 feet down to get 30 feet of matrix. Nowadays, companies have to dig through 30 feet of overburden to access 15 feet of matrix.

Stephen Jasinski, USGS (2009):

Three new phosphate rock mines are planned for development over the next decade in Florida to replace existing mines. The permitting process, however, has been delayed by opposition from local governments concerned about environmental and water use issues. (emphasis added)

Figure 14 shows the phosphate rock price superimposed with the share price of Mosaic Corp (MOS) and Agrium Corp (AGU), two large phosphate rock miners. All three prices are normalised for January 1999. Sources: (Agrium, Mosaic, Phosphate). If ever there was a price signal, this was it - a 900% rise in phosphate rock spot prices. The US switch to ethanol from corn was a contributing factor. In 2007, global grain use for ethanol was 3.3% with US ethanol making up 1.5%. Global use is projected to rise to 5% by 2015 (source).

You might expect this huge spike to entice companies to seriously invest in new capacity. A reading of the November 2009 annual report from Mosaic doesn't give any idication. Despite an excellent cash position and a very bullish projection of phosphate prices going forward, there is no indication they will add extra capacity. Three pages are spent talking about the expansion plans of potash. Mosaic have a 13% share of global phosphate production (down sharply from a couple of years ago due to plant closures). I haven't been following the company but this is at least circumstantial evidence that there is not much easy phosphate left. This lull in capacity is despite the fact (source) that Florida's

phosphate reserves alone contain about 10 billion tons of soluble phosphate rock. Based on the current mining rate in Florida, this would last more than 300 years if economic and technological conditions allow.

The next company I took a look at was Agrium. Agrium has been having difficulty with the decreasing quality of the ore at its Kapuskasing, Ontario phosphate mine. In fact, after drilling extra holes, they were forced to take a $100 million dollar writedown and reduce the economic life of the plant from 2019 to 2013. Figure 15 shows the history of costs to produce Monoammonium Phosphate (MAP), the final phosphorus fertilizer once the rocks are processed with ammonium nitrate and sulfur. The more than doubling of their costs was due to the threefold spike in sulfur prices (an oil production byproduct), a sharp rise in ammonnium nitrate (made with natural gas), and also, significantly, the increased cost to process the decreased ore quality (sources: 1, 2, 3, 4).

There are exploration companies bringing on stream new capacity. Minemakers (MAK) will bring onstream the Australian Wonorrah project in 2010. Although it has large reserves (1.5 Gt), the planned production capacity will be just 3Mt (source). They aim to be "very long term" producers, indicating that rather than have a large capital intensive capacity, they will have a smaller capacity but extract over a longer period. If they believe long term prices are going up, this can make economic sense.

The second place Minemakers are looking is under the sea. 200 metres deep and 60km off the Namibian coast, there are phosphorus nodules spread out over 8,000km^2. These are about 5cm across pebbles on the surface and just below it. These may be accessed by dredging the sea bed and feeding the pebbles to a ship. There may again be 1.5Gt here. This is projected to cost $110/t and again will produce 3Mt/yr.

Another mining company, Widespread Portfolios (WD), is also looking offshore. Off the New Zealand coast, at Chatham Rise, they aim to dredge an area of 227km^2 for $85/t.

I'm not an expert of phosphate by any means, but one possible conclusion is that the cheap, easy phosphate is gone. $85 production cost is a far cry from the $34 1999-2005 selling price. Another possible conclusion is that the best sites on land are largely taken.


Finally, onto the connection with biofuels. As I said at the start, all organisms need phosphorus for their DNA, RNA & for ATP. Algae are no different. They require a constant supply of just the right quantity and ratio of nitrogen, phosphorus and potassium for optimium growth just like any other crop.

I tried to estimate what quantity of phosphorus algae would require to reach certain goals. This brings into question the rather vexed question of yields. Algae proponents, such as the algae companies, claim yields per acre way in excess of any traditional crop. Their figures imply very high photosynthetic efficiencies. There are some who don't buy it.

Take the June 2009 paper by David Alan Walker, a professor of photosynthesis, in the Journal of Phycology (phycology is the study of algae).

He states:

It is frequently claimed that green algae are intrinsically more productive, often by orders of magnitude, than higher plants commonly grown as crops for food. There is no firm evidence for this belief. On the contrary, there is much experience which shows that algae are not more but less productive.

He quotes a contact in the industry

The highest numbers that I have seen based on large scale long production periods for Spirulina are in the range of 4 t per 1,000 m−2. This is an equivalent to a daily productivity of 12–14 g m−2 d−1 and a total of about 300 days of operation. Most of the production facilities are actually doing only 3 t.

He rubbishes claims such as those made by GreenFuels Technologies of

37 times higher than corn and 140 times higher than soybeans.

At any rate, he may be wrong and researchers may yet increase yields. So I took three yield scenarios as laid out in the DOE's 'National Algal Biofuels Technology Roadmap' (although the DOE report recognizes that phosphorus may be THE roadblock, it doesn't attempt to quantify it). The two most commonly cited algae are Spirulina and Chlorella which both have the exact same phosphorus content of 0.895% dry weight.

My calculations are shown in Figure 16, hopefully I haven't made any stupid mistakes.

Figure 17 shows the results graphically.

As is evidenced, algae will require quite a lot of phosphorus. Although I couldn't find any information on it, it's conceivable strains with lower phosphorus intensity could be found. I'm far from a biologist but my inclination would be that any attempt to decrease phosphorus would decrease the yield.

It is thus very clear that if algae are to play a meaningful role at the US and indeed world level, they need to recycle the phosphorus. However even in the run up to build capacity, the levels of phosphorus demand would likely have very detrimental effects on the world markets.

It's entirely possible my analysis is incorrect and that there will actually be lots of phosphorus. Almost universally, those Malthusians claiming shortages in the past have been wrong. On the other hand, I think there are strong grounds that this time could be different. The refrain that demand creates its own supply might not be sufficient.

Cellulosic Biofuels

Although there is no industrial scale commercial technology yet, second generation biofuels are still on the cards. A major portion of their feedstock comes from crop residues. Something not often mentioned is that crop residues are natural fertiliser. The residues decompose and the phosphorus becomes available again for the crops. Wholesale removal of this would require the addition of extra fertiliser to achieve the same yield. In some cases there is too much residue which leads to phosphorus runoff into the local streams. In this case there are grounds for removal. However the optimum removal rate is not known and the DOE is looking into it.

Methods to Improve Phosphorus Availability

Recycle the algae leftover: There are a host of ways you might do this. You could spread it on land (the DOE doesn't know if it's safe to do so yet). You could try to extract the phosphorus and reintroduce it (the DOE doesn't know how you would do that). You could also just dump the leftovers into the pond (the DOE isn't sure what effect this would have). Certainly, though, this should be very high on the priority list. The efficiency of recycling is critical. As I understand it, some of the phosphorus reacts with the dissolved metals in the water, so it isn't all bioavailable. I'm not sure what proportion of introduced phosphorus is bioavailable.

Recycle phosphorus along the food production chain: For instance from Figure 9 it can be seen that 46% of the applied fertiliser is lost from the arable soil through leakage to the water and non arable fields. The Japanese have found a way to recover up to 40% of phosphorus from the sludge at wastewater treatment plants.

Recycle Human Excreta: The Chinese have been doing it for 5,000 years. Stockholm now has toilets which separate the urine and use it to spray on local farmers' fields. Dana Cardell has almost finished a PhD touching on the potential of this to reduce the need for phosphorus rock.

Reduce the applied fertilizer: Precision farming involves taking measurements at one metre intervals on your farm and only applying fertilizer specific to the localised requirements of the soil.

Change to a vegetarian diet: As can be seen from Figure 9, 45% of the phosphorus in crops are fed to domestic animals, with only a small portion of that being eaten by us as meat.

Genetic Modification: I couldn't find much info on the potential to reduce the phosphorus intensity of plants. Some plants use phosphorus more efficiently than others, so there is the possiblity of transplanting genes from the efficient plant into the less efficient. Again, this could potentially have an effect on yield.

Asteroid Mining(!): Some asteroids appear to have 1600ppm of Phosphorus.

Eat Less(!): The average person gets twice the level of the phosphorus RDA.

Procreate Less (!): I estimate the average person requires 50kg of phosphorus from food in their lifetime. Working that back and accounting for inefficiencies, the average person might require 250kg.


It's critical that all new energy sources are examined for bottlenecks and that they are truly sustainable. First generation biofuels have now been shown to be a substantially terrible idea. Critical foresight should be applied to second & third generation biofuels lest they distract us from making a Plan B.

I'm not a mining expert, so the anecdotal evidence presented here may not, in fact, imply that phosphorus production will have difficulty reaching the projected peaks of demand in the future. Additionally, because this issue isn't well publicised, our collective creativity hasn't been focused on this potential crisis. Hopefully that will change and we will find a solution - a technological fix, preferably, but not too energy intensive! Certainly, though, if I wasn't a poor student, I would most likely be investing in this space. The long term trends are very persuasive.

Disclosure: No positions