Micro-algae

TJ Lundquist, IC Woertz, NWT Quinn, and JR Benemann, “A Realistic Technology and Engineering Assessment of Algae Biofuel Production”, Lawrence Berkeley National Laboratory, etc. Oct., 2010.
http://www.ascension-publishing.com/BIZ/Algae-EBI.pdf

It is important to keep in mind that most of the above are to some extent written by people who’s livelihoods depend on continued strong support of R&D in photosynthetic algae by governments and investors.

The review article by Schenk et al identifies Seambiotic as being the closest to having commercial algae oil production capability, though they still seem to be years away. The white paper by Bruton et al, available at the Seambiotic website, is remarkable for its thoroughness with respect to current cost and production data. It quotes the production price of one type of widely sold food-supplement algae at $17,000/ton. Some major cost components are as follows: labor, 45%; electricity, 16%; CO2,13%.

Seambiotic also reports that the only current commercially achieved and sustained algae cultivation is in China, and it yields 25 t/ha/yr dry matter, which is only slightly above that often reported for switchgrass. They report a number of other relatively large scale efforts, but none achieves higher annual production. They note that short-term production rates are often confused with long-term production capability. They report seaweed costs range from $200 to $500/ton, but wild seaweed is not cultivated micro-algae.

Race-track algae ponds at Seambiotic’s test facility in Ashkelon.

The clincher: Seambiotic’s 2009 paper reports algae costs are currently over $5000/ton, which is in agreement with a number reported a year ago in the WSJ. The study by Christi indicates that fuel-grade oil at large volume production from algae at $5000/ton with 35% lipid content would cost over $50/gal, though it reports algae costs in 2006 as being $3000/ton.

The outstanding, recent article by Lardon et al shows that the total input energy (not including solar) required by a scaled-up process based on best available technology would be 2.3 times the energy in the biodiesel produced. They believe improvements are possible that could eventually lead to a positive energy balance, but a number of the key requirements have not yet been demonstrated even at the lab scale – even though – well over one billion dollars have been spent toward this end over the past two decades.

All of the recent studies where there is reasonable transparency and evidence of serious cost projections suggest (for reasons we’ll elaborate on below) it is unlikely fuels from photosynthetic algae will be able to be produced for under $50/gal within two decades. In view of this, we find the amount of attention the algae option has received from NETL and the VC community in the past three years perplexing – even disturbing.

Of course, if energy is cheap, a biofuel need not have a positive energy balance in order to be produced for under $40/gal, but it is necessary for any biofuel to have a positive energy balance if it is ever to contribute to our energy needs.

And yes, we are aware that dozens of companies have made claims of breakthroughs over the past three years that will supposedly lead to enormous price reductions. We have looked at most of them and have seen nothing of real significance. Near the bottom of this page, we mention some of the most outlandish claims.

We are aware that Exxon continues to regularly air commercials in which smiling “scientists” say how beautiful their various colors of algae are, but high-level insiders with real perspective will admit off-the-record that their only interest in algae is public relations – and that they do not expect that micro-algae will really be a competitive challenge to oil. Shell has recently given up on algae and exited their venture with Cellana, one of the most mature algae operations in the world.

Carbon Recycling Using Algae.
The belief by many (including NETL) that photosynthetic algae offer a more efficient option for CO₂ capture from exhaust streams (or the atmosphere) than other options is poorly informed. We explain why near the end of this page.

One of the best analysis on the challenges of carbon recycling using photosynthetic algae was by Krassen Dimitrov in 2007. This outstanding scientist (biologist and physicist, with years of high-profile experience in biofuels) shows that photosynthetic algae, even assuming perfect genetic development and ideal physical implementations, can never compete at oil prices under $800/bbl. The implementation he looks at in particular is that by GreenFuel using large photo-bioreactors.

We have checked Dimitrov’s analysis carefully, and find that his science and economics are sound, and his assumptions are generous – always erring in the direction to make the GreenFuel method appear as favorable as possible. The only place Dimitrov gets off base is in his assumptions about the future price of oil, where he follows the work of Farrell and Brandt (which is excellent in most respects except with regard to estimates of costs of shale oil, tar sands, and coal to liquids).

There are better approaches than that pursued by GreenFuel (and Sapphire, Exxon, BP, Aurora...) which could possibly allow photosynthetic algae to compete when oil is above $1200/bbl, but we doubt the average price of oil ever exceeds $600/bbl for longer than a few years.

Not surprisingly, GreenFuel recently went bankrupt. They apparently operated a 100 m² bioreactor for a while. They may have produced a total of about 2 tons of algae at a cost of $70M. That comes to about $200,000/gal.

Photosynthetic Algae Results.
A number of studies going back about 50 years have often concluded that high-oil photosynthetic algae could be cultivated in covered pools in desert regions and produce much more oil per acre than other plants. One of those studies from the mid 1990’s concluded the cost of the algae oil should be competitive in today’s market, but there were numerous serious flaws in that small and poorly executed study. For example, it concluded microalgae could produce over 14,000 gallons of biodiesel per acre. The only large-scale demonstration to date (HRBP) might have achieved ~600 gal/acre/year in its best year – as is shown below. However, closer scrutiny reveals that even this appears optimistic beyond the first year.

Some Algae photobioreactors.

This large-scale demonstration (HRBP, http://www.hrbp.com/ ) reportedly produced 422 GJ/ha/yr of lipids in its best year (2001) using photobioreactors (PBRs, transparent plastic tanks). The particular strain of algae they were cultivating - Haematococcus pluvialis – yielded ~25% of its dry mass as oil with ~70% of the dry mass being protein and carbohydrates. Hence, ~48% of the total bioenergy of the algae they were producing was lipids.

Their best achieved mean production of lipids in race-track ponds was 3.8 g/m²/day, or 14 t/ha/yr. (A hectare is about 2.47 acres.) Using hexane as a solvent, 70% extraction efficiency could be expected, which implies a crude-oil production rate of 10 t/ha/yr, or 1500 gal/acre/yr. However, the only way to sustain the operation was to use PBRs for the initial growth phase and transfer the cultures to the open ponds for final growth. The gross area required by the combined system was not reported, but was probably over three times the pond surface area. (The PBRs must be well spaced to achieve the advertized performance.)

This fuel energy production rate is considerably less than what is projected for switchgrass, and the 2-ha demonstration (in a tropical, humid setting) cost over $20M during the period of 1997-2001. (Switchgrass can produce up to 24 tons/ha/yr of dry biomass, or bioenergy up to 410 GJ/ha/yr by very low-cost agricultural methods.)

The HRBP estimate published in 2006 is that algae-oil would cost $86/bbl, but this estimate appears to be based largely on material, equipment, and energy costs from 1997 to 2001 – when oil and steel cost about one-tenth what they will cost in a few years. It also assumes an oil production rate over 10 times what was actually demonstrated over a two-year period. They estimate initial capital costs in large scale of $300K per (mean) ha and annual operating costs (perhaps from some 2001 data) of about $15K per ha, or about $11/gal. What they actually demonstrated was nothing close to this. Had they succeeded in producing biofuel from the algae they grew, its cost would have been about $40,000/gal.

Most of the more recent algae studies have studiously avoided the task of trying to project what the product costs might be

One recent cost study was that by Paul Frymier et al, Dept of Chemical Engineering, Univ. of Tenn. The median of their cost estimates for hydrogen from algae is about $50/kg. A reasonable extrapolation is that oil from algae might cost 30% less per unit energy, or about $35/gal. This estimate is also very close to the $33/gal estimate by Solix (in the spring of 2009).

The Israeli company, Seambiotic (applauded above for the outstanding study they published in Feb 2009) has reported producing algae in ponds at the rate of 23g/m²/day. If maintained steadily, that would be 80 ton/ha/yr. The algae contains 30% oil and would yield up to 80 gal of crude oil, or 70 gal of fuel, per ton of algae. That would be an impressive 2200 gal/acre/yr.

A Realistic Cost Estimate.
The enormous drop in the cost of desalinated water over the past 15 years means that it is no longer necessary to steer away from open race-track ponds, and even very large PBRs for inoculation are no longer needed. It’s much cheaper to simply clean and sterilize the open ponds every other month – or even every month. This permits a major reduction in capital cost of algae production. Still, just the pond costs (based on analogies to PVC-lined swimming pools) are likely to be ~$400,000/ha at rather large scale (over 100 ha).

The capital cost of the balance of the plant (land, paddle wheels, pumps, systems for carbonation, fertilization, inoculation, flocculation, dewatering, oil extraction, oil cake utilization, pond cleaning, waste-water treatment, residue management, etc.) will be several times greater. Our total estimate of $1.6M/ha is 4 times the earlier estimate by HRBP; and 5 times the recent estimate by Lundquist et al; but it is one seventh of what HRBP actually spent for an incomplete demonstration, and one twentieth of what several others have actually spent for incomplete large demonstrations.

Our estimate may be about half of what Sapphire is projecting for a 450-ha project. It appears they expect it will cost $1B to develop a 1200-acre facility over the next 8 years, though there are many conflicting reports. They expect it to produce 1 Mgal/yr of “green crude”, which probably means about 700,000 gal/yr of jet fuel and gasoline – a reasonable estimate for ~450 ha of open ponds.

The more detailed published studies listed near the beginning of this page (the BC Innovation Council, the review by Williams and Laurens, and the study by Lundquist et al) are excellent in most respects, and the mean of their capital cost estimates, at the scale of 400 ha, is only about $300K/ha.

So again, for emphasis, let’s assume $10M in R&D and operations costs in the HRBP project, and $100M R&D for the Sapphire project. Then, the capital costs at HRBP were about $5M/ha, and the capital costs in the Sapphire project are expected to be about $2M/ha. In view of the closest real-world data we have seen, our estimate of $1.6M/ha for a 100-ha project using open ponds may be optimistic.

Is there finally a real data point on PBR costs? In February 2011, a useful data point available on the cost of PBRs at large scale may have finally been released. As best we can tell, it appears Algenol expects it will cost about $80M to build 3000 PBRs of about 25 m2 each. That’s about $1000/m2, or $10M/ha. Interestingly, that’s exactly at the upper end of what we’ve been estimating for the past three years. (That should speak well for the quality of our analysis.)

Why are most published studies so far off? Partly because they are thinking in terms of what the fifth plant within a localized region might cost. The problem is that the first several demonstrations in any region will be so expensive and perform so poorly from a cost perspective, that algae is highly unlikely to ever progress beyond R&D.

For general perspective:

(1) the largest components of the operating cost (according to all recent studies) will be labor and electricity;
(2) all crop nutrients (especially phosphates) are expected to experience strong inflation over the coming decades;
(3) in addition to the nutrients that must be fed to any crop, the algae must be fed concentrated CO₂, piped in from power plants (over twice as much as for WindFuels);
(4) the oil extraction costs alone from dry algae are greater than the upgrading costs associated with heavy oil;
(5) the water loss from open ponds in arid regions is enormous (about 40 liters water per liter of oil produced – there is a factor of 10 error in the paper by Lardon et al);
(6) the mean efficiency of utilization of solar energy by microalgae is typically about two percent (about one-tenth that of good PV cells).

An open pond Spirulina farm.

A realistic long-term target for sustained, annual production is 28 ton/ha/yr dry algae, or 2000 gal/ha/yr of extracted crude oil (which would go to a biodiesel factory for refining). The cost of just the interest on a large plant, at 5%/yr, would initially be $40 per gallon of crude oil produced.

The operating costs (in likely order: labor, electricity, replacement parts, fertilizers, CO₂, flocculent, water, hexane, sterilizers, etc.) are very high. The BC study estimates operating costs of $28/gal of fuel for the base case in an area where power costs are very low (and they have probably greatly underestimated maintenance costs). They expect the value of the protein and carbohydrate after lipid extraction to be worth the equivalent of about $2.2/gal of fuel (though others see more value in the co-products). In view of the above, it seems unlikely crude algae oil can be produced for under $90/gal from a 100-ha facility.

From an alternative perspective, one could conclude that our above estimate is still extremely optimistic. Fuels from the best previous results that came close to being a real demonstration (HRBP) would have been about 1000 times more expensive than the above estimate. However, a factor of 10 savings in capital cost should be possible from scale-up by a factor of 100, and another factor of 10 improvement could be expected from increasing the operating lifetime from 2 years to 30 years. On this basis, one might expect algae oil from the first $300M plant to cost about $500/gal and be strongly dominated by O&M costs. This estimate is close to the $425/gal Solazyme is selling their biodiesel for, as noted later under Non-photosynthetic algae.

Additional data points are available that support our belief that O&M costs will remain extremely high. Operating costs at small start-ups (such as Aurora) producing under a few gallons per day appear to be about $5000/gal. Current operating costs at a mature plant (NBT, Nature Beta Technologies, Eilat Israel, operating for the past 20 years) producing 70 t/yr dry algae are about $15,000/ton, which suggests the operating cost for producing oil at the scale of 5000 gal/yr in open ponds might eventually drop to $300/gal, assuming about $100/gal for the cost of oil extraction at this scale.

A number of companies (Joule Unlimited, Synthetic Genomics, Algenol, etc.) are expecting to succeed with genetically modified microorganisms that excrete oil or ethanol. This should greatly reduce the dewatering and oil extraction costs, but they are minor components of the above totals. Most importantly, these concepts could not possibly work in open ponds. PBRs are 3 to 6 times more expensive per area, at least over a period of 20 years or more. Earlier, we estimated the capital cost to contribute $40/gal to the cost of the oil from a large plant based on open ponds. .

A few final words on cost reduction with scale up. Algae oil production by all accounts should be a simple operation, where savings upon scale-up are more limited. Producing 70 t/yr dry algae in a well run mature plant currently costs only a little over $1M/yr (the NBT case). With competent and responsible management, it should not be difficult to develop the needed oil extraction process and produce 5000 gal/yr of biodiesel at an operating cost (ignoring the development cost and the cost of capital) of $2.5M/yr, or $500/gal.

Operating costs of complex industrial processes often scale as the 0.6 power of size. If that applied here, the operating cost at the scale of 5 million gal/yr (5 Mgal/yr) would be $31/gal. On top of that is the cost of capital, which we earlier estimated to be $40/gal for open ponds at about this scale (and 3 to 6 times higher for PBRs).

We suspect the scaling law for something as simple as making algae oil will be closer to 0.8, in which case the operating cost at the scale of 5 Mgal/yr would be $125/gal, and it would still be $31/gal at the scale of 5 Bgal/yr.

The study by Williams and Laurens shows that it will probably be impossible to increase productivity from either race-tracks or PBRs by more than about 50% above current state of the art. Yet, they project a median cost of algae fuel at about $9/gal. No mention is made of the scale of the facility that might achieve this. We see no chance of this ever being possible, even at the scale of a $10B facility.

The BC and the Williams studies are much more optimistic than the Lardon study with respect to energy balance, but we point out that one of the best rule-of-thumb indicators on life-cycle energy balance is the selling price of the fuel relative to the price of fossil fuels. It’s hard to imagine how algal oil could have a positive energy balance or have a climate benefit if it’s selling for more than three times the price of fossil fuels. The energy and carbon debts associated with the plant construction cannot be ignored – as they apparently have been in all prior studies.

It will be interesting to see whether photosynthetic algae oil is being produced at $60/gal or $500/gal in 2020. We’re expecting it will be around $200/gal.

Non-photosynthetic Algae.
Solazyme abandoned their efforts in photosynthetic algae a few years ago in favor of non-photosynthetic algae. They identified many strains of algae and other microbes that are very good at converting other forms of biomass (sugar, hay, wood chips, waste streams...) into oils that generally require less processing than most algal oils. Of course, an enormous amount of CO₂ would be released in the process – more so than from the production of cellulosic ethanol (where more is released than in the production of grain alcohol).

Whether or not non-photosynthetic algal oil from hay and wood chips can achieve the efficiency and cost-effectiveness that has been achieved in production of cellulosic ethanol via either of several established processes remains to be seen. If so, or if sugar prices stay low, algae may produce fuels (including some similar to jet fuel) at significant quantities for 4 to 6 years – until the feedstocks become too expensive, as we explain in our discussion of biofuels. However, both the available evidence and simple reasoning suggest that algae will not be the most cost effective way of making fuel from biomass if it must be dried for transport. It is useful to note that Solazyme has raised about $50M to develop the processfor using algae to make fuels from sugar. They were awarded a Navy contract to deliver 20,000 gal of biodiesel for $8.5M, or $425/gal, and the Navy has ordered another 150,000 gallons. Solazyme generated revenues of $38M (and lost $14M) in 2010, but it’s not clear how much biodiesel and other products they delivered to generate that revenue.

The president of Solazyme, Harrison Dillon, has been quoted as saying “It’s around a thousand times cheaper per gallon to make the (algal) oil by feeding it biomass (especially sugar) than by growing it in the sun.” We see hyperbole here. Our estimate (not too different from that in the BC study) is that it’s only 50 times cheaper today, and will be only 20 times cheaper in 2020.

The price of sugar increased from $140/ton in early 2004 to $600/ton in early 2011. Another factor of four increase in its price over the next seven years is not impossible. If that happens, another billion people may be starving.

Some microorganisms (algae, yeast, bacteria) are also capable of non-photosynthetic production of hydrogen from biomass, particularly sugars. In a recently published paper titled “High Hydrogen Yield by Fermentation Using Clostridium Saccharoperbutylacetonicum N1-4”, about 300 mL of H₂ was produced (in several days) from 500 mL of water containing 5 g of sugar. The energy in 300 mL (24 mg) of H₂ is 3.4 kJ, while the energy in 5 g of sugar is 90 kJ, so the efficiency here was less than 4%. Perhaps some microorganisms do better.

On a more positive note, a promising small source of practical bio-fuel is algae feeding off waste streams which must be treated anyway for sanitation reasons before disposal. This makes sense for three compelling reasons: (1) there is very little increase in the cost of the waste treatment from changing the microbes and the sequence of processes so that most of the suspended biomass is converted to oils instead of being released as methane and CO₂; (2) there is no increase in near-term carbon release, as this bio-carbon is destined for the atmosphere anyway within a matter of weeks; (3) oils are much more valuable than methane. Unfortunately, the amount of fuel that can be produced from waste streams currently being processed for sanitation compliance is very limited – perhaps 0.5% of our transportation fuel needs. However, most of this will likely go toward fueling a more competitive process – direct combustion in power plants, as discussed under biofuels.

CO₂ Distribution.
The recent eloquent paper by Kurt House should be a reference point for any discussion on the energy penalties associated with carbon dioxide re-use, either from point sources or from the atmosphere. House shows clearly that it is more efficient (and cost effective) to separate the CO₂ from exhaust streams before compression, and compression is absolutely essential before sending the CO₂ anywhere other than directly out the smoke stack. The thermodynamics is independent of the method used for separation – whether biological or physical or chemical.

OriginOil. http://www.originoil.com/ promulgates one whacky idea after another in an attempt to fool more investors. First they were going to use electromagnetic fields to achieve quantum fracturing of cells. It so happens that only a handful of companies around the world (making MRI radio frequency and gradient coils) know as much or more about electromagnetic fields in tissues than we do at Doty Energy. (See our sister webpage on NMR probes http://www.dotynmr.com/ .) One thing we know (as do all MR scientists and engineers) is that the radio frequency (RF) and static electromagnetic (EM) fields used in MRI are over 100,000 times less intense than would be needed to cause fracturing of cells in a dilute water suspension. OriginOil thinks somehow they can use EM fields 1000 times less intense yet to cause the oil to magically leave the cells and float to the top of a tank. They’re off by at least a factor of 100,000,000 in their physics and cost projections here.

Their latest notion is that separating the algae ponds into multiple layers will allow growth rates 10-20 times faster. Again, the Williams and Laurens study shows conclusively, for fundamental scientific reasons, that current growth systems (after 3 decades of developments) are probably within 30% of practical limits. A multi-layer system of the type pictured on the OriginOil webpage would probably cost 20 times as much as the conventional PBR and produce less.

Confused Science.
AlgaeVS http://www.algaevs.com/products-technology claims to have reduced the cost of dewatering algae from $875/ton to $1.92/ton; and they go through some flawed calculations that indicate they are confused about the difference between energy, its second derivative, and its first derivative (power). Amazingly, investors, General Atomics, and DOE failed to recognize the problems with their analysis.

Unacceptable Algae Idea.
The idea by LiveFuels of feeding algae to fish and extracting the oil from fish to fuel vehicles is the sadest we’ve heard yet from a humanitarian perspective. We regularly ingest fish oil. We pay about $150/gal for some of the cheapest we can find. If LiveFuels can bring the price of fish oil down, maybe more people can have better health, but fish oil will never fuel our vehicles. The world switched from whale oil to petroleum 150 years ago, and that saved the whales. We won’t go back.