Ambre Felton 2008 Critique

Critique of Ambre Energy Felton Initial Advice Statement Compared with Alternatives Philip Machanick 5 February 2009

Executive Summary The Felton “clean coal” proposal by Ambre Energy is premised on providing a useful starting point for developing carbon capture and storage technology for coal, with an additional aspect of producing dimethyl ether (DME) as a diesel substitute. The latter aspect of the proposal, while containing claims of improved environmental outcomes relative to conventional diesel, is an add-on to reduce the unattractive economics of carbon capture and storage (CCS) in a coal-fired power plant. The Initial Advice Statement (IAS) fails to present a case for an alternative to carbon emissions, and does not address the combined problems of agriculture facing difficult times as energy becomes expensive, and climate change makes future weather patterns uncertain. The proposal fails to take into account alternatives that would not compromise agricultural output, nor does it contain any proposal to scale up CCS even to the volume of CO2 emitted during the demonstration stage, let alone the final project. Even for the minimal fraction of CO2 for which CCS is described, the annual total of 0.5Mtpy of CO2 emissions in the demonstration stage would amount to 250,000 cubic metres if compressed to a liquid, totaling 0.0125 cubic km by the end of the project, assuming a 50-year lifespan. If scaled up to the full production, a total of 0.2 cubic km of CO2 would have to be sequestered over 50 years. It also fails to account adequately for a range of pollutants, including most non-carbon components of coal (which include toxic metals such as mercury and radioactive elements), and carbon emissions arising from burning DME in diesel engines. There is also no plan to deal with carbon emissions from the power plant, other than an unquantified proposal to plant trees. Sequestering the volume of CO2 produced by this plan would result in planting trees on an impractical scale given that the trees would otherwise be unproductive, amounting to thousands of square kilometres over the lifetime of the project. The CCS proposal is further flawed in that one of the options considered would recycle the CO2 it uses, rather than storing it permanently. The proposal further fails to provide farmers with an answer to escalating energy prices, a significant issue in that food prices are important to everyone, but rather threatens to put a number of farmers out of business. The scale on which CCS is proposed is not only a small fraction of the demonstration stage of the project, but will not demonstrate that CCS works on a larger scale than existing demonstration projects. The CCS aspect of the project therefore does not introduce significant innovation either in advancing understanding of CCS, or in mitigating the pollutants the project will produce. The fact that CCS is not included in discussion of potential environmental effects further suggests that CCS is not a serious aspect of the proposal. There are numerous errors of detail, including variation in the number of days used to annualise a daily value, from 320 to 430, and listing of irrelevant and possibly incorrect data relating to global warming potential of DME. Alternative modes of clean electricity production that have not been considered in the IAS would require less land area, and could be implemented without damage to agricultural output. The number of trees required to offset CO2 generated by the project could be replaced by trees suitable for producing biodiesel from their seeds,

with a similar scale of tree planting required. However, trees used to produce biodiesel would be productive – part of the energy supply chain – not an additional cost, and would become increasingly practical as costs on carbon emissions and the price of oil increased. Some alternatives are presented in this response to illustrate the options; these alternatives are examples drawn from many options that could be considered. Overall the project amounts to a very expensive way of using coal in terms of emissions and probably also in terms of cost per unit of energy (though the latter is not quantified in the IAS). While it is possible that some airborne pollutants will be eliminated, this effect (at least as far as greenhouse gases are concerned) is more than outweighed by the additional inefficiencies relative to burning the coal directly.

ii

Contents 1

Introduction

1

2

Key Features of Ambre Proposal

3

3

Practicalities of DME

6

4

“Clean Coal” Claims

8

5

Food versus Fuel

14

6

Discussion

18

iii

List of Figures 4.1 4.2

IGCC compared with Ambre’s proposal. . . . . . . . . . . . . . . . . Methane versus depth of coal for various types of coal. . . . . . . . .

9 11

5.1 5.2

Wind potential in Queensland . . . . . . . . . . . . . . . . . . . . . . Pongam tree potential biodiesel yields . . . . . . . . . . . . . . . . .

15 16

iv

List of Tables 2.1

2.2

Emissions as reported in the IAS, with some missing numbers derived. If sequestered amounts are subtracted, there would be a substantial reduction in the total especially in stage 4. . . . . . . . . . . . . . . . . Missing emissions. Methane is estimated on the basis of a low limit on how much might be emitted; inadequate numbers are provided in the IAS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

4

4

LIST OF TABLES

LIST OF TABLES

vi

1

Introduction The Felton area was first recognized for its potential to deliver an economic resource by Millmeran coal in 1971 – Felton Clean Coal Project Initial Advice Statement, Ambre Energy, 24 July 2008.

F ELTON AREA may be interested to learn that it takes a coal company to know that their land could be an “economic resource”. The starting point of this response to the Ambre proposal is that farming is a valued activity, and that dealing with the looming energy crisis without addressing the knock-on effects of higher energy costs for farmers will impact society as a whole, especially those most vulnerable, for whom food prices are critical to day to day survival. One way of mitigating the effects of higher energy prices on agriculture is to involve farmers wherever practical in energy production. If they are net energy producers, they are effectively insulated from energy price rises and may even profit from them. At worst, as producers, they will consume some of their energy requirements at producer prices, rather than at retail prices. Further, agriculture is one of the sectors that will be hardest hit by climate change. Taking these points into account, energy production proposals that compromise agricultural output should be a last resort, if they are considered at all. It is particularly obnoxious to propose a development that will not only reduce agricultural output but also contribute to climate change. Ambre energy’s Initial Advice Statement (IAS)1 on their “clean” coal project has a particularly ironic cover picture: a fuel hose with leaves sprouting from the spout. This project aims to replace farm land by coal mines, and to produce fuel of a kind that could be produced as a biofuel instead of from coal. This project appears to be based on the political drive to fund “clean coal” rather than on sound science. While there is a section of the IAS (2.3) headed “Project justification and alternatives”, it is clear that the project is based on the preconception that “clean coal” is a worthy goal, as no alternatives are considered (section heading notwithstanding). Carbon capture and storage, the usual meaning of “clean coal”, is a small part of the project and only applies to CO2 released in the manufacture of fuel

F

ARMERS IN THE

1 Obtainable

in two parts:

www.pdfcoke.com/doc/10841606/Ambre-Energy-Initial-Advice-Statement-2872008-Part-1 www.pdfcoke.com/doc/10841900/Ambre-Energy-Initial-Advice-Statement-2872008-Part-2

1

CHAPTER 1. INTRODUCTION rather than in combustion; this further adds to questions as to what the project is really about, and what it will actually achieve. Nonetheless, given that a fundamental rationale for the project is exploring the “clean coal” concept, it is important to evaluate it on the basis of the extent to which it achieves that goal, i.e., providing an energy source with a significantly lower environmental footprint especially in terms of greenhouse gas emissions. A further consideration is that another alternative not under consideration is continued use of the land in question for farming. Yet another point in the rationale is the development of dimethyl ether (DME: chemical formula CH3 -O-CH3 ) as an alternative fuel, on the basis that DME burns much cleaner than diesel, particularly in terms of reduced particulates and oxides of nitrogen, NOx ) [Advanced Engine Technology 2001]. The DME aspect of the proposal is justified on page 11 of the IAS, where it is indicated that at least 75% of the gasified coal will be converted to DME, and the balance used for “clean coal”. This is a point of critical importance, as this ratio inherently limits the fraction of carbon emissions that can be sequestered at the plant. Any emissions from DME burnt as diesel fuel will be emitted from the tailpipe of many vehicles, and hence not amenable to geosequestration. While the project as described does not allow for sequestration of emissions from the power station either, that could potentially be added later (though one has to wonder why, if it is a practical possibility, it is not part of the initial plan). This response focuses on these principle claims and omissions: that the proposal significantly addresses “clean coal”, that retaining agriculture in the affected area is not an alternative to fuel production and that DME represents a useful alternative to diesel, especially as a coal derivative. Failure to address these issues should be sufficient reason to deny Ambre Energy a license to operate. Prior to entering into a critique of the IAS, this proposal contains a summary of the key features, followed by chapters covering “clean” coal claims, alternatives that do not disrupt agriculture, practicalities of DME as an alternative fuel, and, in conclusion, discussion of the IAS and alternatives.

2

2

Key Features of Ambre Proposal

IAS HAS VARIOUS NUMBERS scattered around, making for difficult reading if you want to quantify the claims and impacts. This chapter therefore contains a summary of the numbers of most significance. Since some aspects of the project are not quantified in the IAS, e.g., the contribution to greenhouse gases from combustion of DME in diesel engines, numbers for these missing quantities are also supplied, to complete the picture. The proposal includes a coal mine, a dimethyl ether (DME) plant, a coal gasification plant to produce an efficient fuel for power generation, a power plant using the fuel produced from the gasified coal using an integrated gasification combined cycle (IGCC) power generator and carbon capture and storage (CCS: by geosequestration). The IAS also mentions fertilizer production as a related technology developed by Ambre (Section 1.2, p 4), but this idea, while briefly mentioned elsewhere, is not further developed in the document. DME is touted as a cleaner alternative to diesel, and is the subject of extensive research including production from biomass [European Commission 2006; BioDME project 2008]. The key reason for the inclusion of DME in the proposal is because “clean coal” on its own (as a means of power generation) is not viable in competition with other alternatives (see p 11 of the IAS). The IAS identifies four stages in total, but stages 2 to 3 are a ramped-up expansion over three years, and breaking the numbers down over those stages is only useful for planning, rather than assessing impacts. In this chapter therefore, we only consider the demonstration stage numbers and the final stage numbers. Table 2.1 contains numbers summarized from subsequent chapters of this document. Some are directly from the IAS; others had to be derived. Other figures that are completely ignored in the IAS are pointed to in Table 2.2. Of these numbers, only an estimate for methane emissions (converted to carbon dioxide-equivalents) is included. It should be the task of the proponent to provide all these numbers, not an outside assessor. Some of these numbers should be reasonably easy to derive (e.g., by taking typical emissions figures for operation of similar mines). Others may be harder, for example, quantifying the effects of land clearing requires an assessment of existing vegetation and soil types, allowing that a large amount of carbon may be sequestered in the soil [Watson et al. 2000].

T

HE

3

CHAPTER 2. KEY FEATURES

stage 1 stage 4

DME production combustion Mtonnes Mtonnes 0.03 0.29 0.46 4

CO2 sequestered Mtonnes 0.55 8

power station combustion Mtonnes 0.5 2

total emissions Mtonnes 1.37 14.46

Table 2.1: Emissions as reported in the IAS, with some missing numbers derived. If sequestered amounts are subtracted, there would be a substantial reduction in the total especially in stage 4. land clearing unknown

mining unknown

factory partial numbers

transport unknown

methane (pa) estimate CO2 -e 1.15Mtpy stage 4

Table 2.2: Missing emissions. Methane is estimated on the basis of a low limit on how much might be emitted; inadequate numbers are provided in the IAS.

Given that carbon capture and storage is touted as a major feature of the project, it is puzzling that this aspect is not mentioned beyond the initial stage and there only as a partial solution. It is therefore hard to conclude that CCS is a serious component of the project. The fact that CCS is not included in potential environmental impacts (p 6) also indicates that this aspect of the project has not been thought through; environmental impacts of CCS are significant especially as regards environmental leakage [Koornneef et al. 2008]. Even given Queensland’s lax environmental impact assessment laws, it is inconceivable that a project for CCS could be seriously contemplated without considering risks; these will affect insurance costs, potential compensation to farmers for loss of production and in an extreme case loss of life. While DME is being widely investigated, it has some downsides. At normal temperatures, it is a gas, requiring that it be stored compressed (with similar general handling to LPG). That is not necessarily a serious problem, as many cars in Australia have been converted to run on LPG. Like any diesel fuel, DME has to have appropriate lubricity (lubrication properties) and a high enough cetane number (representing its ability to combust without a spark) [Radich 2004]. DME does have a relatively high cetane number, but it does not have sufficient lubricity to work on all diesel engine without modifications1 . Also, since it is stored compressed, when pressure is released, it will cool, and some design changes in engines may be needed to avoid this cooling being a problem [Advanced Engine Technology 2001]. In short, although DME has some useful properties in terms of reduced pollutants, it will require changes to engines or possibly may turn out only to be suited to specific engines. Should these problems be overcome, DME has the potential to be a useful biofuel. Making it out of coal significantly reduces its advantages by making it highly carbon-intensive. This summary reveals that the IAS requires detailed scrutiny to determine if the 1 In Australia, many vehicles are modified to run on liquefied petroleum gas, but these are generally sparkignition engines, designed to burn petrol. Running a diesel on pure LPG is even more of a challenge than using DME: http://www.go-lpg.co.uk/diesel.html

4

CHAPTER 2. KEY FEATURES claims stack up. In particular, it is important to understand the implications of claims that carbon emissions can be sequestered since these claims are a key justification for the whole approach.

5

3

Practicalities of DME

DME as a “clean” fuel. In their keenness to demonstrate the “clean” aspect of the project, the authors of the IAS have included some irrelevant information, and have been sloppy on the detail, with some obvious errors. Their claims of lower particulate emissions (IAS p 25) are in line with other reports on DME. However, there are mixed results on carbon monoxide emissions [Arcoumanis et al. 2008]. It is not clear what purpose the data in Table 2.13 (p 22, IAS) is. The global warming potentials listed apply if the gas is in the atmosphere. Their purpose in manufacturing DME is not to vent it into the atmosphere but to burn it as a fuel, which will produce CO2 ; according to their numbers on the same page of the IAS, 1kg of DME produces 1.91kg of CO2 when burnt. This is the figure of most relevance. Their primary source [Good et al. 1998] appears to have neglected to consider the outcome of DME decaying into water (H2 O) and CO2 in the atmosphere as claimed in the IAS. If this was an error in the paper cited and the IAS claim that dimethyl ether decays into water and CO2 , the global warming potential of DME would be almost double that of CO2 . If this is an error in the IAS, it of no significance given that the plan is to burn the DME, not vent it to the atmosphere. Further, Table 2.13 is contradicted by more recent work that shows dimethyl ether as having significantly higher global warming potential (GWP) [Forster et al. 2007; Blowers et al. 2008]. Let us therefore ignore their Table 2.13 and focus on the fact that we must correctly score the global warming potential of the DME produced according to the calculation of 1kg DME = 1.91kg CO2 , arising from combustion of DME, rather than venting it into the atmosphere. Assuming this value, the global warming potential of DME after combustion should be calculated as 1.91, taking CO2 as 1.1 On this basis, the demonstration stage of the project will result in emissions purely from combustion of DME in engines of 286,500 tonnes of CO2 per year (based on annual production of 150,000 tonnes of DME). By stage 4, the plan provides for an increase in DME production to 6,500 tonnes per day, or 2.8Mtpy. Based on this arith-

A

MBRE IS TOUTING

1 Since the IAS authors have made errors elsewhere, let us check the calculation. If we work from atomic masses, and the chemical formula for DME, using standard units (rounded to the nearest whole number, ignoring fractions arising from rare isotopes) of H = 1, C = 12, O = 16 [Mills et al. 1993], DME with the formula CH3 -O-CH3 has an atomic mass of 46. Assuming both carbon atoms convert to CO2 (each atomic mass 44, totaling 88 = 1.91 × 46), 1 molecule of DME should combust to 2 molecules of CO2 at most (this will be an over-estimate given that some carbon monoxide is also produced, though this is a relatively low fraction for DME). On this basis, 1kg of DME should not produce more than 1.91 kg of CO2 by combustion.

6

CHAPTER 3. DME metic, it would appear that Ambre works on a 430-day year; the multiplier for other stages of the project from daily to annual production varies from 320 to 325. Assuming the shortest year, let us be kind to Ambre and correct their number to 6,500 times 320 = 2.08Mtpy (perhaps 2.8 is a typo). Allowing the conversion factor of 1.91, we end up with about 4Mtonnes per year of CO2 emission from DME combustion alone. This CO2 emission may be slightly less than that from conventional diesel, but it has to be compared with other potentially lower emissions alternatives. Further, this number does not take into account emissions from production of the DME, including mining and the chemical plant. Of these numbers, only emissions from production of DME are given in the IAS (p 37) as 220kg/tonne DME for production. It is not clear if this includes mining operations, but seems likely that it does not, since those emissions also apply to fuelling the power station. Some of the emissions from DME production are slated to be used to experiment with carbon capture and storage, but it is important to note that this CO2 does not arise from combustion of fuels, but is a by-product of a fuel conversion process. This point is taken up in chapter 4.

7

4

“Clean Coal” Claims

arising from combustion is being held back from venting to the atmosphere. Numbers have already been derived for DME. For the power plant. the emissions of 400 kg per MWh (IAS p 90) are indeed competitive with best practice for fossil fuels, as would be expected for a modern combined cycle gas power plant. However, these emissions are layered on top of those from DME, the energy requirements of the factory and the pre-combustion CO2 stream. Similar numbers could be achieved with natural gas, without all the collateral environmental damage of a coal mine or the need to bury extra CO2 produced as side-effect. They propose to use an integrated gasification combined cycle (IGCC) power plant. IGCC is a technology that gasifies coal as part of the combustion process,. The proposal in the IAS is to gasify the coal in a separate stage, with some feedback of heat and steam to earlier stages, a variation on standard IGCC designs. An IGCC plant would typically produce 700 kg of CO2 per MWh [Diesendorf 2007]. Apparently if some of the CO2 produced is counted separately the power plant becomes “cleaner”. This can only be described as sleight of hand. Figure 4 illustrates operation of a typical IGCC power plant, showing integration of gasification with power generation1 , by comparison with the Ambre proposal2 , which would integrate the fuel production process they describe into a similar design. The biggest practical differences between the two schemes are that the Ambre design produces DME and (as depicted) does not include a power station, which is off the edge of the diagram. Total emissions (purely from power generation), when the power station is fully operational based on 650MW with a capacity factor of 90% (365 days times 24 hours/day times 90% times 650MW times 400kg/MWh), amount to over 2 Mtonnes of CO2 emissions per year. The only quantification in the IAS of carbon capture and storage is in the Introduction (p 3), where a figure of 1,339 tonnes per day at demonstration stage is mentioned (this figure is confirmed in Figure 4.1(b), which does not however show emissions outside the gasification path). That would be a total of under 0.5Mtpy. Since no indication is given as to the quantity of CO2 generated in the plant, we have to extrapolate from

W

HEN READING THE PROPOSAL , it is important to note that no CO2

1 Source: http://upload.wikimedia.org/wikipedia/commons/9/96/IGCC_ diagram.svg 2 Source: http://www.toowoombachamber.com.au/downloads/AEsurat.pdf

8

CHAPTER 4. “CLEAN COAL”

Gasification - Felton Block Flow Diagram (a) A typical IGCC plant. Steam!for!Process!Use

Coal!Feed

HRSG

tpd 1180 Coal Moisture 140 680 Ash

Oxygen

H2S!!!!!!!!!17.1!tpd Water!!!!!63.0!tpd

720!tpd

H2S!and! water! removal

Fluidised! Bed!Gasifier

tpd H2 77.9 CO 1331.2 CH4 32.4 N2 31.2 CO2 507.4 1980.1

220!tpd Steam

Coal!!!!!!!!!200.3!tpd Ash!!!!!!!!!!680.0!tpd

1339.2!tpd

CO2

Shift!Reactor vol% 38.5% 47.0% 2.0% 1.0% 11.5%

DME Reactor

Tail!Gas!to!Power!Generation

CO2 removal Steam 165.0!tpd

tpd!!!!!!!!!vol% H2 37.8!!!!!60.6% CO!!256.7!!!!!29.4% CH4 32.4!!!!!!6.5% N2 31.2!!!!!!3.6% CO2 0.0!!!!!!!0.0% 358.0 Power!export!!!!!40.9!MW 447.9!tpd

Product!Separation

DME!Product

(b) Ambre’s gassification process.

Figure 4.1: IGCC compared with Ambre’s proposal.

the demonstration stage, assuming that the “precombustion” CO2 stream is a constant fraction of the total. The demonstration stage will mine 800,000 tonnes of coal per year (based on 2000 tonnes per day, evidently using the by-now-familiar Ambre 400 day year), and the planned scale-up is an additional 12 million tonnes per year (3 million extra in stages 2 and 3, and 6 million in stage 4), totalling 12.8 Mtpy. This is a scale-up of 16 over the initial amount. Based on this, the total CO2 available for sequestering would be about 8Mtpy (initial figure times 16). The assumption here is that 9

CHAPTER 4. “CLEAN COAL” the sequestration plan aims to sequester all the CO2 produced from fuel production; the IAS does not make it clear how if at all the initial amount is to be scaled up. In future discussions, the higher number is used for illustration, but the demonstration figure is carried forward as well, since it is not clear what the intent is for carbon capture and storage. Emissions from land clearing are not included nor are emissions from mine machinery; it is not clear if all emissions from the chemical plant are included; a figure is given for DME production. Also, no accounting is made for methane emissions. While methane emissions are dependent on depth and the mining is mostly shallow, for harder grades of coal, methane levels rise very steeply even within the depths contemplated (up to about 85m judging from illustrations in the IAS (Fig. 2.5, p 14). The grade of coal targeted3 would be at the lower end of the curves graphed in Figure 4.2 [Williams and Mitchell 1994]. Nonetheless given the high greenhouse potential of methane, the amount should be quantified even if it is relatively low. Working from Figure 4.2, if the bulk of the coal is relatively shallow, 2cm3 of methane per gramme of coal (which works out as 2m3 per tonne) is a conservative allowance. For 800,000 tonnes of coal (the demonstration stage annual amount), that amounts to 1.6-million m3 of methane per year. Methane has a density4 of 0.67 kg/m3 , so this amounts to over 1,000 tonnes of methane per year. How significant is this output? Methane decays from the atmosphere a little faster than does CO2 so its global warming potential is 72 times that of CO2 , falling to 25 times in 100 years [Forster et al. 2007]. For continuous emissions of methane, the number 72 is a more accurate reflection of the effect. If we look at the longer-term operation of the mine, when coal output will be scaled up by a factor of 16, therefore, the methane output of over 16,000 tonnes per year equates to over 1.15-million tonnes of CO2 (CO2 -equivalent, or CO2 e) per year. Little is said about a variety of toxic chemicals (including mercury, fluorine compounds, arsenic, uranium, among many others) found in coal [CCSD 2008; Riley and Farrell 2002] and what will happen to them. The IAS briefly mentions taking care in burying heavy metals (p 40) but contains no detail of how they will be managed between mining and eventual burying. Putting all the numbers together, the project only contains an aim of sequestering 0.5Mtpy of CO2 emissions in the demonstration stage. No mention is made of the CO2 emissions from operating the mine, building the factory, land clearing, using DME as fuel or running the power station as targets for emissions reduction. These total amounts are in excess of 4Mtpy to which emissions from manufacturing DME (220kg/tonne, a total of 2.08 × 0.220Mtpy, another 0.46Mtpy) plus any other emissions not accounted for. If CO2 sequestration is not scaled up, the extra CO2 emitted before combustion has to be added to the total, making a total of 16.5Mtpy. On the other hand, if CO2 sequestration is scaled up, sequestration of 8Mtpy of CO2 represents a considerable challenge and many risks, none of which are mentioned in the proposal. Let’s consider exactly how big those challenges are. At standard temperature and 3 http://www.engineeringtoolbox.com/classification-coal-d_164.html 4 http://physics.nist.gov/cgi-bin/Star/compos.pl?matno=197

10

CHAPTER 4. “CLEAN COAL”

Figure 4.2: Methane versus depth of coal for various types of coal.

pressure, CO2 has a density5 of 1.98 kg/m3 . Working with uncompressed CO2 in these volumes would require handling over 4-billion cubic metres of material, or over 4 cubic km of gas. Density can be increased to around 1 tonne per cubic metre6 by compressing CO2 to form a liquid7 ; assuming this pressure could be maintained, the volume of CO2 that would have to be stored would be of the order of 0.004 cubic km per year, or about 0.2 cubic km over 50 years. If on the other hand the initial figure is not scaled up, 5 http://www.uigi.com/carbondioxide.html 6 http://encyclopedia.airliquide.com/Encyclopedia.asp?GasID=26 7 http://www.engineeringtoolbox.com/carbon-dioxide-d_1000.html

11

CHAPTER 4. “CLEAN COAL” “only” 0.0125 cubic km (12,500,000 m3 ) of compressed gas would have to be stored over 50 years of operation (250,000m3 per year). Turning to risks, CO2 produced from large-scale combustion does not pose an immediate threat to life provided it is vented into the open air at high temperature, where a combination of convection and turbulence mixes the CO2 into the atmosphere. On the other hand, CO2 vented at low temperature even into an open space can be lethal. There are several documented cases of deaths resulting from emissions from volcanoes; around 15% atmospheric concentration is lethal to humans. This kind of level can arise from emissions into low-lying areas, or near buildings, where the buildings may accumulate the gas. Over 1900 deaths have been reported from volcanic CO2 emissions over the last century [IVHHN 2008]. The proposed approach (p 19) is to compress CO2 then pipeline it. Since it will be piped under pressure it is questionable that an old oil pipeline can be reused. Even with new piping, risks of leakage are considerable as CO2 is denser than air and as noted in above, leakage could result in fatalities. Even if the CO2 is not vented into the atmosphere, if it leaks into aquifers or soil, it can have undesired consequences. CO2 is water-soluble, and forms an acid in solution in water. Another potential unwanted effect is mobilization of heavy metals, resulting in soil and aquifer contamination [Koornneef et al. 2008]. It is not clear how the CO2 flood proposal (p 19) will be beneficial for sequestering CO2 if the CO2 is recycled, as stated in the IAS. Finally, another form of mitigation proposed includes planting trees in various configurations (p 38). Let’s look at the practicality of this. Estimation of carbon sequestration of trees is a complex matter, depending on the type of tree and soil type, among many other factors. Optimum sequestration requires that trees be left to grow for many decades, reducing the options for commercial harvesting. Economics of carbon sequestration in trees must take all these factors, in addition to pricing of carbon emissions, into account. What’s more, to ensure sequestration continues to be effective, constant monitoring is necessary [NSW 2005]. The most generous figure for sequestration that I found in the literature, but which is unlikely to be achieved in practice, is 211tC/ha [Bateman and Lovett 2000], or 774tCO2 /ha. This figure must be related to continuing emissions from the mine, fuel production and fuel use. Let us take this a step at a time. First, the emissions directly attributed to manufacturing DME will be 0.46Mtpy. To offset this amount of carbon, on unrealistically favourable assumptions of how few trees would be required, over 590ha of trees would have to be planted every year to offset this amount of CO2 . Add in the trees to offset burning of DME, and you have another 4Mtpy, requiring over 5,000 ha of trees, again every year. Power generation produces another 2Mtpy, requiring a modest planting of an additional 2500 ha per year. Adding all this together, and you would need to plant over 8,000 ha of trees every year. What’s more, to get anywhere close to the unrealistic assumption of 774tpy per hectare of carbon dioxide sequestration, you would not be able to harvest the trees commercially for many decades. It’s likely that in practical terms, the land needed would be three to four times that calculated here. Since my calculation of methane output is approximate, I will leave that out of the calculation, as well as other unquantified emissions including plant, machinery and land clearing. The duration of the project is not stated, but given 900Mt of coal (p 13) and 12

CHAPTER 4. “CLEAN COAL” 12.8Mtpy (p 5), the project has a maximum life-span of 70 years, depending what fraction of the coal is useful. Let us make an arbitrary assumption of 50 years, which results in at least 400,000ha of trees required to offset the emissions, or 4,000 square kilometres. Since this number is obviously totally impractical (even more so if more realistic assumptions are used, and three to four times the number of trees must be planted), Ambre needs to make it clear exactly what fraction of the CO2 that they are planning on producing that they actually plan to sequester. A few tokenistic plantations will make barely a dent in the emissions produced; quantification is necessary to understand how serious they are. Similarly, the claims that it will be possible to sequester CO2 are not sufficient without quantification of the volumes involved, accurate assessment of where capacity exists to store that volume of CO2 and proper analysis of risks such as leakage from hundreds of kilometres of pipelines.

13

5

Food versus Fuel

frequently raised against biofuels is that they may displace food production. However, replacing farmland by a coal mine does not only have the potential to displace food production, it definitely will. Given the twin imperatives of climate change and peak oil, it is clear that alternative energy sources need to be explored. Given that imperative, exploring a clean break from the past is sensible; trying to do more of the same less efficiently harkens back to the early days of cars, when they were designed as “horseless carriages”, some even to the extent of providing receptacle for the now-unneeded horse whip. When clean energy sources are deployed on a large scale, some will require wide geographic distribution, including solar power and wind power. Including energy generation of these types on farms can be organised to bring income to agriculture, without interfering with food production. Wind power, for example, can generate electricity in a significantly smaller area than an open-cut coal mine feeding a coal power station, and can easily co-exist with agriculture: the negatives are grossly exaggerated especially as compared with fossil fuel alternatives [Diesendorf 2007]. In this case, 650MW of wind power using 1.5MW turbines would require about 66ha1 . It is unlikely that such a wind farm would be sited near Felton, because Queensland’s best wind resources are elsewhere (see Figure 5.1 [Queensland EPA 2002]); nonetheless the lost land is instructive to compare against not only the 500ha of farmland to be destroyed at a time by mining, but the total of 2,800ha eventually to be targeted. If the farm were scaled up to allow for intermittency [Diesendorf 2007], it would need about 100ha: 20% of the area needed by the coal mine. The turbines would be spaced out over a much bigger area to reduce intermittency, but 100ha represents land that would cease to be available for other uses. A point commonly missed in food versus fuel debates is that agriculture is one of the economic sectors hardest hit by increased energy prices. Putting all these points together, it is useful to explore energy options that do not reduce food production, and which cushion farmers from increased energy costs. Examples of these alternatives include energy production from agricultural waste (not a novel concept: sugar farmers, for example, have used bagasse for years as a feedstock for power plants). This approach to electricity generation can be cost effective versus fossil fuels, as well as being a significant employer [van den Broek et al. 2000]. Overall sugar production including using bagasse as a power generation feedstock re-

O

NE OF THE ARGUMENTS

1 http://www.nrel.gov/analysis/power_databook/calc_wind.php

14

CHAPTER 5. FOOD VS. FUEL

Figure 5.1: Wind potential in Queensland (Source: Queensland Environmental Protection Agency).

sults in considerable reduction in carbon emissions, since the burnt bagasse is replaced by the next round of growth [University of Ballarat 2004]. Another alternative is fuel crops that grow on marginal land, such as the pongam tree (Pongamia pinnata: other names include Indian Beech Tree and in Australia, native wisteria), already in use in India to produce biodiesel from its seeds [Karmee and Chadha 2005]. Pongam is being researched at The University of Queensland [Williams and Gresshoff 2006]. This tree 15

CHAPTER 5. FOOD VS. FUEL is a legume, and hence does not need much fertilizer, and is capable of growing in relatively saline soil [Scott et al. 2008]. It can yield up to 5 tonnes of biodiesel per hectare [Biofuels Association of Australia 2008]. Given that pongam has widely differing yields, until researchers have arrived at consistent yields, let us take a conservative estimate of 2.5t/ha of biodiesel. If 8,000ha of trees are planted, matching the annual number needed on a very optimistic calculation of carbon offsets of the Ambre proposal, when those trees reach full production, those trees will produce 20,000tpy of biodiesel, or 0.02Mtpy. It would take 140 such plantings to equal the planned production of 2.8Mtpy of DME. However, the difference is that the trees would have economic value (other than an artificial value accruing from carbon credits or taxes). Once you start planting trees to offset pollution, it becomes a more interesting proposition to replace the source of pollution by trees that do the job directly. The biggest difficulty with establishing pongam as a crop is start-up costs, as a fully-productive plantation is likely to be profitable even at a diesel price of $1 per litre [Odeh et al. 2008]. 1.8

annual biodiesel yield (Mt per year)

1.6

1.4

1.2

5t/ha 2.5t/ha 1

0.8

0.6

0.4

0.2

0

50 years

Figure 5.2: Pongam tree potential on the basis of similar tree plantings to those needed to offset CO2 from the Ambre project. Figure 5.2 illustrates how pongam plantings compare with the peak output of 2.8Mtpy 16

CHAPTER 5. FOOD VS. FUEL of DME. I have only shown output for 10-year old trees, assuming that output as the trees ramp up to full production will pay for initial planting costs. The final position is between 0.8 and 1.6Mtpy of biodiesel, a reasonable fraction of the output of the proposed DME plant, given that it will not destroy agricultural production. Adding in other biofuels options and scaling up tree plantings faster would make this option more competitive with the proposed DME plant, without the downsides of taking farm land out of production and adding to CO2 emissions (biofuels should be carbon neutral with proper mangement – even carbon negative in some circumstances [Tilman et al. 2006]). Note that this comparison is on the basis of highly optimistic assumptions for the number of trees needed to offset the Ambre proposal. If the rate of tree planting is tripled, a more realistic number in terms of offsetting the Ambre proposal, pongam biodiesel output would exceed that of the DME plant somewhere between 27 and 44 years from the earliest plantings (depending on yield). Finally, it should be noted that, once biodiesel becomes a profitable enterprise, farmers will grow it on the scale the market demands whereas tree planting to offset CO2 is an artifact of necessity of reducing pollution, a much harder imperative to drive consistently over a long period. A more recent idea for which there is growing interest is biochar, charcoal produced from organic matter, often agricultural waste. A commonly advocated approach is pyrolysis, chemical decomposition of organic matter in the absence of oxygen. Biochar production can be used to make biofuels, leaving a residue of carbon that can be buried as an enhancer for agricultural land as well as a method of sequestering carbon. Since the sequestered carbon was originally leached from the atmosphere to grow plants, biochar-based processes have been touted as potentially carbon-negative. One estimate puts the break-even point of this process at a CO2 cost of US$37 per tonne [Lehmann 2007]. Agricultural waste is a potential feedstock for this process; I will not however investigate this option in more detail because to do so would require figures on available biomass. There is still significant work to be done on biofuels and biochar, but none of this involves exotic technologies and the probability that at least one will work acceptably is high. Biofuels have an additional advantage in agriculture. By making farmers part of the supply chain, they are able to benefit from being able to access energy lower down the supply chain (i.e., with fewer layers of mark-up), and to add energy into their income stream. Given the inevitable escalation in energy costs, providing these benefits to farmers will have a significant knock-on effect to the rest of the economy, all of which relies on agriculture directly or indirectly (wage settlements are influenced by food prices, for example).

17

6

Discussion

sequestering all of the carbon emissions of the final stage years in trees is not a realistic proposition, even under absurdly favourable assumptions. Ambre needs to clarify just how little of their emissions and other pollutants they really plan to manage. It is further not clear that the plan allows for scaling up CO2 sequestration to a significant degree, especially as it is not clear that the IAS authors have thought through the practicalities of geosequestration, including the fact that the CO2 flood method of enhanced oil recovery will not sequester CO2 if (as described in the IAS, p 19), the CO2 is recycled. If geosequestration does not work on the scale required (amounting to many cubic kilometres of liquified gas), that will be yet more CO2 in the atmosphere for which even more trees will be required. Various details of the IAS show sloppy attention to detail. The number of days used to annualise daily figures varies from 320 to 430. The wrong technology is proposed for the power plant for the proposed fuel, and quantification of emissions omits significant factors, including emissions from combustion of DME as fuel, methane from the mine (even if it is relatively minor), land clearing and operation of the mine machinery. Risks in geosequestration are substantial, and are not mentioned at all in the IAS. Risks relating to disposal of toxic components of coal are dealt with cursorily. Geosequestration is in any case only considered for a by-product CO2 stream not CO2 from producing DME, or from burning fuel. The overall CO2 production therefore must be higher than from a gas-fired power station of similar efficiency to that proposed, but running on natural gas. If research into improving yields of pongam is successful, the equivalent in biodiesel of the 2.8Mtpy of DME could be produced from 560,000ha of trees, a considerable amount compared with the 2,800 hectares the Ambre proposal would cover – but not gigantic compared with at least 400,000ha of trees required to offset the CO2 emissions from the factory, burning DME and the power plant. But these would be productive trees that could be grown on otherwise unproductive land, and they would need no high-risk or expensive carbon sequestration. Furthermore, this tree stock could be built up slowly over time, as the price of oil escalated – and as the price of carbon pollution escalated. To this should be added the option of converting agricultural waste to biofuels; the main cost in this case would be the factory, since the feedstock already exists and by and large has no commercial value.

W

E HAVE ESTABLISHED THAT

18

CHAPTER 6. DISCUSSION Biodiesel puts farmers on the right side of the energy production supply chain. Other options such as wind power may not be suited to particular locations, but including unproductive agricultural land where possible in such schemes aids farmers in coping with higher energy prices by including them on the production side. A wind farm taking up 100ha of unproductive land could effectively replace the power station side of the project, without requiring any risky, highly environmentally damaging technologies. While this wind farm is unlikely to be placed in Felton, it could be part of an overall strategy to place energy production in agricultural areas in non-disruptive ways, giving farmers an income stream to offset rising energy costs. The potential for solar power on farms has not been considered here in detail, but could be an additional factor in providing farmers with an energy-based income stream to offset increased energy costs. As with the other approaches, minimizing impact on agriculture while providing farmers with an additional income stream remain goals. Ambre has managed to put together a proposal with all of the disadvantages of biofuels but very few of the advantages. A large amount of land is involved, productive farmland is to be taken out of production, and costs are likely to be high. On the other hand, the advantages of biofuels of zero net emissions are unlikely to be met, and the Ambre proposal further does nothing to mitigate increased energy costs to farmers. Unless they are serious about a massive tree planting exercise, carbon emissions will be higher than for a regular fossil-fueled power plant, and many farmers will be put out of business. The only real value of the Ambre proposal is that it illustrates how poorly “clean coal” competes with truly clean technologies. Overall the Ambre proposal is not a long-term solution to the declining availability of fossil fuels, and is not a convincing answer to the need to reduce carbon emissions to reduce climate change. It is also a poor answer to the need to protect food supply against the risks caused by climate change, a problem caused largely by coal.

19

References Advanced Engine Technology (2001). A study of dimethyl ether (DME) as an alternative fuel for diesel engine applications. Technical report, CANMET Energy Technology Centre, Natural Resources Canada and Transportation Development Centre, Transport Canada. Arcoumanis, C., Bae, C., Crookes, R., and Kinoshita, E. (2008). The potential of di-methyl ether (DME) as an alternative fuel for compression-ignition engines: A review. Fuel, 87(7):1014–1030. Bateman, I. and Lovett, A. (2000). Estimating and valuing the carbon sequestered in softwood and hardwood trees, timber products and forest soils in Wales. J. Environmental Management, 60:301–323. BioDME project (2008). BioDME web site. http://www.biodme.eu/. Biofuels Association of Australia (2008). Submission by the Biofuels Association of Australia (BAA) to review of Australia’s tax system. http: //www.taxreview.treasury.gov.au/content/submissions/ Biofuels_Association_of_Australia.pdf. Blowers, P., Tetrault, K. F., and Trujillo-Morehead, Y. (2008). Global warming potential predictions for hydrofluoroethers with two carbon atoms. Theoretical Chemistry Accounts: Theory, Computation, and Modeling (Theoretica Chimica Acta), 119(4):369–381. CCSD (2008). CCSD power station emissions handbook. http://www.ccsd. biz/PSE_Handbook/. Diesendorf, M. (2007). Greenhouse Solutions with Sustainable Energy. UNSW Press, Sydney. European Commission (2006). An eu strategy for biofuels. Technical Report SEC(2006) 142, Commission of the European Communities. http://ec. europa.eu/agriculture/biomass/biofuel/com2006_34_en.pdf. Forster, P., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R., Fahey, D., Haywood, J., Lean, J., Lowe, D., Myhre, G., Nganga, J., Prinn, R., Raga, G., Schulz, M., and 20

REFERENCES

REFERENCES

Dorland, R. V. (2007). Chapter 2: Changes in atmospheric constituents and in radiative forcing. In Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K., Tignor, M., and Miller, H., editors, Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, pages 129–234. Cambridge University Press, Cambridge. Good, D., Francisco, J., Jain, A., and Wuebbles, D. (1998). Lifetimes and global warming potentials for dimethyl ether and for fluorinated ethers: CH3 OCF3 (E143a), CHF2 OCHF2 (El34), CHF2 OCF3 (E125). J. Geophysical Research, 103:28181– 28186. IVHHN (2008). Gas and aerosol guidelines carbon dioxide (CO2 ). http://www. dur.ac.uk/claire.horwell/ivhhn/guidelines/gas/co2.html. Karmee, S. K. and Chadha, A. (2005). Preparation of biodiesel from crude oil of pongamia pinnata. Bioresource Technology, 96(13):1425–1429. Koornneef, J., Faaij, A., and Turkenburg, W. (2008). Environmental impact assessment of carbon capture & storage in the Netherlands. In Proc. GHGT8, Trondheim. Lehmann, J. (2007). A handful of carbon. Nature, 447(7141):143–144. Mills, I., Cvitaˇs, T., Homann, K., Kallay, N., and Kuchitsu, K. (1993). Quantities, Units and Symbols in Physical Chemistry. Blackwell Science, 2 edition. http://www. iupac.org/publications/books/gbook/green_book_2ed.pdf. NSW (2005). Greenhouse gas benchmark rule (carbon sequestration) no. 5 of 2003. http://www.greenhousegas.nsw.gov.au/documents/ GtA-CS.pdf. Odeh, I., Tan, D. K., and Ancev, T. (2008). Exploring the suitability and viability of biodiesel crops under climate change in Australia. In Proc. Faculty of Agriculture, Food & Natural Resources Symposium 2008 (FAFNR 08), pages 50–55, University of Sydney. Queensland EPA (2002). Fact sheet wind. http://www.epa.qld.gov.au/ register/p00401aa.pdf. Radich, A. (2004). Biodiesel performance, costs, and use. http://www.eia.doe. gov/oiaf/analysispaper/biodiesel/. Riley, K. and Farrell, O. (2002). Management of trace elements in ash dams. Technical Report 33, QCAT Technology Transfer Centre. Scott, P., Pregelj, L., Chen, N., Hadler, J., Djordjevic, M., and Gresshoff, P. (2008). Pongamia pinnata: an untapped resource for the biofuels industry of the future. BioEnergy Research, 1:1–10. Tilman, D., Hill, J., and Lehman, C. (2006). Carbon-Negative Biofuels from Low-Input High-Diversity Grassland Biomass. Science, 314(5805):1598–1600. 21

REFERENCES

REFERENCES

University of Ballarat (2004). New South Wales sugar industry renewable electricity generation. http://www.ballarat.edu.au/projects/ensus/case_ studies/sugar/index.html. van den Broek, R., van den Burg, T., van Wijk, A., and Turkenburg, W. (2000). Electricity generation from eucalyptus and bagasse by sugar mills in Nicaragua: A comparison with fuel oil electricity generation on the basis of costs, macro-economic impacts and environmental emissions. Biomass and Bioenergy, 19(5):311–335. Watson, R. T., Noble, I. R., Bolin, B., Ravindranath, N., Verardo, D. J., and Dokken, D. J. (2000). IPCC special report on land use, land-use change and forestry. Technical report, IPCC. http://www.ipcc.ch/ipccreports/sres/land_ use/index.htm. Williams, A. and Mitchell, C. (1994). Methane emissions from coal mining. In Hester, R. E. and Harrison, R. M., editors, Mining and Its Environmental Impact. Royal Society of Chemistry. http://www.rsc.org/images/is001006_ tcm18-26096.pdf. Williams, S. and Gresshoff, P. (2006). The good oil. Australasian Science, pages 34–36.

22