Biofuels: Thermodynamic Sense - and Nonsense

Jack W Ponton
School of Engineering and Electronics, University of Edinburgh
Edinburgh EH9 3JL, Scotland
j.ponton@ed.ac.uk

Abstract

Much of the current enthusiasm for biofuels amongst politicians (who cannot be expected to know any better) and academics (who should) appears to ignore basic themodynamic and other constraints.

The fundamental problem with growing fuel is that combustible plant matter is almost invariably solid, while the major demand for energy at present is in the form of gas or liquid fuels. All current conversion processes are of low efficiency even for the convertible parts of the plant: for example the energy which could be obtained from burning a kilogram of wheat grain is about twice that available from the ethanol into which it can be converted by fermentation. Furthermore most liquid fuel processes can use only part of the plant.

This paper will seek to identify those biofuel technologies which make sense, such as co-firing straw with coal in power stations, and those which thermodynamic considerations show to be nonsense, such as making ethanol from grain in Europe.

Since arable land is a scarce resource in most of Europe, biofuels are unlikely to become a major replacement for fossil fuels. We will suggest strategies which will help to maximise the contribution which they could make, and promising emerging technologies where it would make sense to concentrate research.

Introduction

Over the last year or so the popular, and popular scientific, press has been full of articles about biofuels: biodiesel and bioethanol. The US has a major, and highly subsidised, corn to bioethanol programme which, enthusiasts claim, will produce fuel at a cost equivalent to 60 cents a US gallon, about 8p/l.

The EU has, without apparently consulting informed scientific sources, imposed an `obligation' for us all to produce 5% of vehicle fuels from biological sources by 2010.

However, more recently a note of scepticism, or realism, seems to have crept in. Some articles have suggested that the net energy yield from maize ethanol, once all energy costs are taken into account, is very low or even negative. Concerns have been raised that the EU's biofuels are not all being grown on CAP `set-aside' land (we note below that this would never have been possible) but are encouraging palm oil growers in Asia to destroy rainforest to grow oilseeds for export. It has also been suggested that using synthetic fertilisers to grow fuel crops in fact releases more damaging greenhouse gases, namely nitrous oxide, than the fuel produced can save in carbon dioxide.

In fact, most of these issues can be understood by elementary thermodynamic analysis, and indeed studies already exist which confirm that many of the concerns and reservations which have been expressed are entirely valid.

Is it Worth Growing Crops for Fuel?

The answer to this question has to be `yes' unless one can devise some other long term source of energy. The alternatives, briefly are as noted below.

Furthermore, nothing other than chemical fuels as yet provide portable energy sources adequate for the kind of transport which societies now expect. Biofuels are chemical fuels and so can, in principle, fill this bill.

Agriculture, properly managed, is environmentally sustainable. It is fundamentally a way of capturing solar energy. Per square metre, it is much cheaper than any manufactured solar collection device. However, overall efficiencies are very low, typically fractions of one percent to usable chemicals.

So biofuels are necessary, and so growing them has to make sense. However, this answer has to be qualified by consideration of what kind of fuel to grow, how to use it and where to grow it.

Initial Considerations

In this discussion we shall consider primarilyy the UK. However, very similar conclusions should result from an analysis of the options for any part of the temperate zone.

First, it is helpful to have an idea of the scale of our energy requirements.

In 2005 the UK [1] used, reducing all energy sources to million tonnes of oil equivalent (Mte), 247 Mte, approximately in the following forms:

In terms of energy content, 1 te oil equivalent is about 42 GJ or 11.7 MWh.

Definitive energy balances for biofuel production are thin on the ground. However a report with estimates for biodiesel and ethanol in the UK was produced by Levington Agriculture [2] for BAFBO, an organisation representing producers of oilseeds and biofuels. A comparison of this and other estimates may be found in [11].

A hectare of arable land used to produce biodiesel would yield about 54 GJ worth of fuel, or about 74 GJ if used for grain based ethanol. So as biodiesel, one would get 1.28 te of oil equivalent. This is a bit less than the 1.5 te of biodiesel which the report predicts.

To quickly put things in context, the total land area of the UK is 241,590 km2 or about 25 Mha. This includes cities and mountaintops, but if all of it could be used to grow biodiesel, it would still produce less than one fifth of the UK fossil fuel consumption.

In fact the total amount of arable land in the UK is about 5.5 Mha. If we put all of this down to biodiesel, we would replace at best 8.25 Mte of oil, less than 10% of our total oil consumption! Ethanol yields are somewhat higher, but it has been estimated that growing grain for ethanol on all this land would provide only the equivalent of about one third of the UK's current gasoline (petrol) requirements. It seems unlikely that more than, say 20% of agricultural land could be divertied to fuel use, so we are looking less than 1% of the national energy requirements.

Furthermore, these are the gross energy figures. The energy required to produce and process the crop is more than 50% of the energy available from biodiesel and nearly 90% for ethanol - so in practice nothing like this proportion of our energy requirements would be replaced.

We can immediately draw some conclusions.

  1. Liquid biofuels grown in the UK can only have a very minor impact.
  2. If biofuels are going to have significant effect on the UK energy economy, they must either:
    • have very much higher per hectare yields than above, or
    • be grown on non-agricultural land, or
    • be grown elsewhere than in the UK.
Others [5] have pointed out the impossibility of producing the EU's biofuel `obligation' on set-aside land, and in general confirm these conclusions on an EU-wide basis.

What Kind of Biofuels?

The next most important thing to appreciate is that chemical transformations of fuel from one form to another normally either involve a reduction of the energy value of the material or the input of energy, or both.

Consider the figures from [2] for ethanol from wheat. For each 1 te of grain we get 0.276 te of ethanol. The calorific value of the grain is about 17 MJ/kg, so if the grain were burned it would produce 17GJ of energy. However, the energy value of the ethanol is only 8.3GJ, a loss of more than 50% without even taking into account the processing costs! In fact the energy value of the wheat straw is 10.9GJ, greater than that of the ethanol. If the entire plant were used as solid fuel it would yield nearly 28GJ of energy instead of 8.3GJ and the production cost would be substantially less, so the net energy yield would be 4 to 5 times greater.

Plant products are normally solids. It is therefore most efficient to use them in this form unless transforming them leads to increased efficiency elsewhere. In this case it is by no means obvious that i.c. engines running on ethanol will be more efficient than, say, power stations fuelled with straw.

The conclusion here is that biofuel crops have to a) allow the whole plant to be used and b) involve minimum additional processing.

A possible exception to this might be gasification. Existing well-established coal gasification technology can be used for biomass on both large and small [20] scale. The gas could be used in a gas turbine which does have a higher thermodynamic efficiency than a steam plant

A Glycerol Lake?

It is worth pointing out that the UK Government's prefered biodiesel strategy, as well as involving unnecessary processing and energy losses, suffers from a further drawback.

Although it is perfectly possible either to use straight vegetable oil [10] (called SVO or pure plant oil, PPO) after only physical processing in slightly modified diesel engines, or to process it along with petroleum oils in existing refineries, a more complex process is being promoted by tax breaks.

The vegetable oil is reacted with methanol to produce methyl ester, see e.g. [18]. This is about one third of the molecular weight of the oil and so can be used in an unmodified engine. In the reaction below R has typically 20 carbons.

C3H5(COOR)3 + 3 CH3OH = 3 CH3COOR + C3H5(OH)3

The byproduct, glycerol, C3H5(OH)3, is a potentially valuable chemical used in the food, cosmetic and pharmaceutical industries. However, it is not a bulk chemical like methanol or diesel fuel. In fact the US market (2000) for glycerol was around 200 kte/yr [8] and was less than installed capacity. Byproduct glycerol from 5% of the UK diesel market alone would provide an additional 150 kte/yr. Excess glycerol from European biodiesel production by 2000 had already dropped US by up to 45% in the late 1990s.

Byproduct Biomass

It is perhaps understandable that the US is obsessed with transportation fuels, but of the 247 Mte oil equivalent used in the UK in 2005, just over one fifth, 59 Mte was used for transportation [1]. This is comparable with 51 Mte used in electricity generation, of which 37 Mte is gas and petroleum, both of which could alternatively be used as transportation fuels.

At present the UK has about 1.9 Mha of wheat and 1 Mha of barley [3]. The straw from this is essentially a waste product. More than 40% [4] is currently ploughed in as it has some minor value as a fertiliser. Taking the the heat value of wheat straw [2] to be applicable also to barley suggests that the UK has at the present the potential to produce (1.9 x 97.5) = 185 million GJ of energy from waste straw. This corresponds to 4.4Mte of oil equivalent to which could be added the straw from 0.58Mha of current rape production, at 60 GJ/ha, a further 0.8Mte, giving a total of 5.2Mte is equivalent to nearly 10% of the energy used in transportation without reducing the land area available for food production.

Of course, there are other issues. Straw is produced over a limited period, although conveniently this is in the autumn, before the winter peak of energy demand. It has to be transported, stored and processed, and the technology of efficient combustors developed. At least one coal burning power station, in Denmark [12], has successfully `co-fired' straw. The last of these problems would appear to already be solved.

In Scotland Cockenzie power station, convenient to the agricultural are of East Lothian, is in fact already co-firing byproduct biomass. However, this presently is olive oil waste from southern Italy. It is hard to see the justfication of transporting a low value fuel more than half the length of Europe.

Biomass Fuel Crops

The relatively high energy content of wheat straw, a plant that has been bred for thousands of years to maximise its grain yield, suggests that there may be other species which could be grown to produce still larger amounts of biomass cellulose. This is indeed the case. Grasses, reeds and some rapidly growing softwoods such as willow and poplar [17], are all crops which could yield biomass in comparable or greater quantities. A particularly promising plant is Miscanthus [14] which offers high yields in most of the UK.

A recent study in the US [21] claims that switchgrass, a fast growing perennial could produce more than five times the energy required to grow it. However, this appears to only marginally better than for oilseed rape for which the factor is about 4.5 [2]. (The yield energy figure in fact applies to a notional and as yet nonexistent process for conversion to ethanol, with uspecified costs and efficiency!)

Effective fuel crops must share the following characteristics:

Provided a strategy for using them directly in power generation can be developed, such plants are possible energy sources to replace fossil fuels. Furthermore, these are plants which have largely evolved naturally. Applying modern plant breeding methods, especially genetic modification, could increase yields still further. One might hope that public hysteria about GM plants will be assuaged if they are to be used for fuel rather than food.

Power generation companies are already carrying out a significant number of tests with biomass fuels, normally co-fired with coal [13]. In the UK, Drax, a large UK coal fired station, has tested co-firing with willow [15].

The above calculation for the total energy available from wheat on a per hectare basis, with a yield of 8.96 te of grain per ha, implies an energy yield of 6.2 te oil equivalent per hectare.

Assume that genetic modification of selected varieties of fuel crop could increase this by about 50% to around 9 te/ha, and that land equivalent to 30% of the current UK arable total were to be used for fuel crops. This could be comprised of 10% set-aside, 10% diversion and 10% of additional marginal land suitable for fuel but not food.

The UK could produce about 0.3 x 5.5 x 9 = 14.8 Mte of fuel crops. Add to this the energy available from straw and other agricultural byproducts and a figure of 20Mte is plausible.

This could either replace about two thirds of the coal used in power generation or two thirds of the oil and gas.

Replacing coal, the worst fuel from a CO2 emission standpoint, would result in a significant reduction in the UK's emissions, providing politicians and concerned members of the public with a warm green glow and meeting everyone's `targets' for greehouse gas reduction. The practical significance of of this, given that China will be continuing to add several times this amount every year is perhaps questionable.

Replacing oil and gas would free these fuels for use in transportation. This would be both ecomomically and environmentally more efficient than making liquid fuels from biomass using any presently available technology.

Prospects for Liquid Fuels

As every chemical engineer knows, fluids, particularly liquids, are much more convenient than solids. It would be satisfactory if we could develop efficient routes from biomass to liquids, and indeed significant effort has gone into the investigation of improved pathways.

Much work has focussed on conversion of cellulose to ethanol. This would enable all of a plant rather than just its sugar or starch content, to be converted. Efficiencies are still too low for this to be a realistic process.

Development of the 1919 Weizmann process for bacterial conversion of carbohydrates to acetone to produce butanol have been investigated since the 1980s. Recent developments, e.g. [6] claim good yields and DuPont [7] claim to have a viable commercial process. Although precise information is hard to come by, it may be that conversion of cellulose using Clostridium bacteria is possible by this route. If so, this is certainly a better propect than ethanol, since butanol may be more readily mixed with gasoline or used as a straight replacement without engine modification. However, it all hinges on the efficiency of conversion of biomass.

Structured plants are very inefficient converters of sunlight to biomass. Much less than 1% of the incident solar energy is released even if the entire plant is used. However, algae, which are much simpler organisms display greater efficiency, and some of them even produce hydrocarbons. Their faster growth rates, and the fact that their production can be essentially continuous rather than seasonal, could make relatively low efficiencies of hydrocarbon production acceptable [16]. Moreover, dried algae are a fine particulate material which could be slurried with liquid fuels and used directly in suitably designed burners or even i.c. engines.

However, present technology does not offer an efficient route to liquid fuels. Much more research and development is needed.

Grow it Elsewhere?

The basic problem with fuel crops in northern Europe is that, since plants are inefficient converters of sunlight, when there isn't much of this, i.e. for about half the year, little growth takes place. In the tropics, on the other hand, it is possible to grow two or three crops a year.

From the standpoint of ecomomics, sugar cane grown for ethanol in e.g. Brazil, appears to make sense. Bioethanol in Brazil is actually sold at a competitive price without any overt subsidies. However, there must be provisos.

If the whole plant is not used then the process is inefficient from either a thermodynamic or environmental viewpoint, or both. Bagasse, the residual plant material from sugar cane, is traditionally burned to provide energy for refining the raw sugar, and it is claimed that it can be used to provide all the energy required in ethanol production.

What are the long term effects of intensive sugar cane cultivation? There is evidence that it can be severely damaging to the environment.

Is cane cultivation really not competing with rainforest? Even if it isn't now, one must fear that if biofuels become a lucrative industry, it will in the future.

Similar arguments apply to e.g. palm oil production in Malasia and Indonesia. Indeed the developed world's enthusiasm for biofuels may seriously threaten such countries [19].

Waste Not?

The UK produces more than 30 Mte/yr of domestic waste. Most of this has gone to landfill, occupying space near cities and towns with which the country is by no means well supplied, and producing, by anaerobic fermentation to methane, a significant contribution to the greenhouse effect.

There are various ways in which this material can be used as an energy source, the simplest being just to burn it. It is fairly simple to remove glass and metal, and most of what remains, about 85%, although not high grade fuel, is combustible, and in 2000 about 6.5Mte was in fact combusted to produce heat and/or power [24].

There are both practical and political difficulties with combusting a larger proportion. Practically, it is a poor fuel, and, with a significant amount of animal material from food waste, may need carefully designed combustors to avoid the production of toxic byproducts. EU directives, for no good scientific or economic reason, also limit the amount of waste which may be combusted to 25% of the national total. The public also dislike the idea of `incinerators'.

However, waste includes 8-10% of plastics, much of which is already separarated by consumers and collected separately. At present this is recycled, in some cases after a complex, expensive and not very efficient further separation, into low value, low grade material with limited uses.

Although it is possible in principle to depolymerise waste plastic to produce a syncrude which can be used as a chemical feedstock, it is questionable as to whether this is worthwhile. Of the 180Mte of oil and gas consumed in the UK in 2005, only about 10Mte was used as a chemical feedstock.

Much more sensible would simply be to use waste plastic as high grade fuel in existing power stations. Since mixed plastic waste is traded at well below its fuel value this makes economic sense. Since it will substitute for new oil or gas it also makes environmental sense. That the most efficient way to recycle plastics is as energy has been known for more than 10 years [25].

There are other proven technologies, such as anaerobic fermentation to methane for fuel which could be more widely used, as well as emerging technologies of algal digestion which can be applied to waste as well as new biomass.

Conclusions

When work on this paper was started (January 2007) the general public and political climate was strongly and uncritically in favour of biofuels. Even from the outset of the work, the author was rather more sceptical [9]. Since that time, a number of better informed sources, e.g. [19, 22] have sounded serious notes of caution. The UK Parliament's Environmental Audit Committee [23] has called for the EU to abandon its renewable transport fuels directive.

In addition to the purely thermodynamic and logistical issues considered here, there are concerns about the net balance of greenhouse gases. Nitrogenous fertilisers can produce nitrous oxide, and ploughed in biomass methane. both of these produce many times the greenhouse effect of carbon dioxide.

Even if biological energy sources are not a panacea, they are potentially of significant value. However it is important to take all factors into account, and choose appropriate rather than fashionable strategies.

We will end with a summary of what we think makes sense, or otherwise, in the development of this type of energy.

Nonsense:

Questionable: Sense:

References

[1] "Energy Consumption in the UK", DTI publication:
www.dti.gov.uk/energy/statistics/publications/ecuk/page17658.html

[2] IR Richards, `Energy balances in the growth of oilseed rape for biodiesel and wheat for ethanol", Report for BAFBO, June 2000.
www.senternovem.nl/mmfiles/27781_tcm24-124189.pdf

[3] UK Agriculture web site www.ukagriculture.com/index.cfm.

[4] www.nnfcc.co.uk/nnfcclibrary/cropreport/download.cfm?id=10.

[5] M Frondel and J Peters, `Biodiesel, a new Oildorado?', Energy Policy 35, 1675-1684, 2007

[6] DE Ramey, US patent 5,753,474, 1998

[7] DuPont website www2.dupont.com/Biofuels/en_US/

[8] www.the-innovation-group.com/ChemProfiles/Glycerine.htm

[9] JW Ponton, `Biofuels, Solution or Snake Oil?' seminar to IES, University of Edinburgh, 2007.

[10] The SVO Users and makers Association http://www.bio-power.co.uk/svoa/index.htm

[11] AP Armstrong et al, `Energy and greehouse gas balance of biofuels for Europe', report 2/02 for CONCAWE, Brussels, 2002 http://www.senternovem.nl/mmfiles/26601_tcm24-124161.pdf

[12] P Overgaard et al, `Two years operational experience and further development of full scale co-firing of straw', 2nd World Conference on Biomass for Energy, Industry and Climate Protection, Rome, 10-14 May 2004

[13] WR Livingston, `Advanced biomass co-firing technologies for coal fired boilers', keynote at International Conference on Coal science and Technology, Nottingham, Sept 2007

[14] MAFF report NF0403 Miscanthus Agronomy: www.defra.gov.uk/farm/crops/industrial/research/reports/rdrep14.pdf

[15] Drax Power Ltd web site, http://www.power-technology.com/projects/drax/

[16]NERC Knowledge Transfer Project web site: http://www.bluemicrobe.com/home.htm

[17] `Yield Models for Energy Coppice of Willow and Poplar', Energy crops and biofuels web site: http://forestry.gov.uk/srcsite/infd-5l8hhr

[18] V Hofman, `Biodiesel Fuel', AE-1240, Feb 2003, http://www.ag.ndsu.edu/pubs/ageng/machine/ae1240w.htm

[19] R Clift and Y Mulugetta, `A plea for common sense (and biomass)', TCE 24-26, October 2007

[20] Ankur Scientific, India, sell small scale biomass gasifiers: http://ankurscientific.com/

[21] M. R. Schmer, K. P. Vogel, R. B. Mitchell, and R. K. Perrin, `Net energy of cellulosic ethanol from switchgrass' Proc Nat Acad Sci, 105, 2, 464-469, 2008 http://www.pnas.org/cgi/content/short/0704767105v1

[22] J. Picket, et al, `Sustainable biofuels: prospects and challenges' The Royal Society, London, 2008 http://royalsociety.org/document.asp?id=7366

[23] `Are biofuels sustainable?', Report by UK Parliamentary Environmental Audit Committee, 2008 http://www.parliament.uk/eacom/

[24] `Incineration of household waste', UK Government Post Note 149, December 2000

[25] W Reid Lea, `Plastic incineration versus recycling: a comparison of energy and landfill cost savings', Jl of Hazardous materials, 47, 295-302, 1996