Royal Commission on Environmental Pollution

The views expressed in the paper are those of the authors and do not necessarily represent the thinking of the Royal Commission. Any queries about the paper should be directed to the author indicated * above.

Whilst every reasonable effort has been made to ensure accurate transposition of the written reports onto the website, the Royal Commission cannot be held responsible for any accidental errors which might have been introduced during the transcription.

March 1998

Acknowledgements

The author would specially like to thank Rolf Frischknecht for his help in preparing and checking this report. In addition he is very grateful for the advice received from colleagues at the Centre for Environmental Strategy and from Peter Douben.

Executive Summary

This report assesses the environmental impacts of four different types of energy system delivering: heat (per 1GJ); transportation (per person-km); electricity (per 1GJ); and combined heat and electricity (per 0.64GJ heat and 0.36 GJ electricity). The environmental impacts considered are: resource depletion, greenhouse gas emission, acidification, and eutrophication.

The data is taken from a detailed study on the life cycle inventories of continental European energy systems. Although the UK is not specifically included in this study, most of the energy systems considered are representative. The environmental impacts are quantified using life cycle assessment which includes all stages in the provision of the particular service.

For the delivery of heat to domestic consumers it is found that there is little difference between natural gas and heating oil fuels in terms of greenhouse gas emission and eutrophication. Acidification and resource depletion are higher for the heating oil system. The environmental impacts of the supply of natural gas to the user are small relative those due to its combustion in boilers. For the heating oil this is true only for greenhouse gas emissions; other impacts due to extraction, processing and distribution are significant.

The analysis of the supply to point of use of the transport fuels, diesel and unleaded petrol, shows that unleaded petrol supply causes larger environmental impacts than diesel, although the differences are small. Relative to their use in vehicles, the supply of either transport fuel causes little environmental impact in the categories considered, emphasing that it is in the use phase that most of the environmental impacts arise.

The supply of electricity considers: oil, coal, natural gas, nuclear and solar-derived electricity. Of the fossil fuel electricity production systems, natural gas produces the lowest acidification and eutrophication impact; coal the largest greenhouse gas emissions; and oil the largest impact in resource depletion. Photovoltaic electricity has substantially lower impacts in all categories except acidification where it exceeds that of natural gas electricity. Nuclear power produces the lowest impacts of all electricity production systems in these four environmental impact categories. However the main environmental impact of nuclear power, radioactive emissions, has not been included in the report, as this impact cannot yet be quantified by life cycle analysis.

The supply of heat and electricity to consumers using natural gas as a fuel is found to have the lowest impact if a combined heat and power plant is used. Using the gas to produce electricity which is transformed to heat and electricity at the point of use has the highest impact of the systems considered in all environmental impact categories.

Contents

Executive Summary

1. Introduction

2. Life Cycle Assessment of Energy Systems
  2.1 Heating Fuels
  2.2 Transport fuels
  2.3 Electricity
    2.3.1 Electricity from Fossil Fuels
      2.3.1.1 Electricity from Oil
      2.3.1.2 Electricity from Coal
      2.3.1.3 Electricity from Gas
      2.3.1.4 Comparison of Fossil Fuel-generated Electricity
    2.3.2 Electricity from Nuclear Fuel
    2.3.3 Electricity from Solar Power
    2.3.4 Comparison of all methods of electricity generation
  2.4 Combined Heat and Power Systems

3. Conclusion

4. Appendices
  4.1 Appendix A Life cycle inventory categories
  4.2 Appendix B Life cycle inventories of energy systems
    4.2.1 Life cycle inventory of heating fuels - supply and use for 1GJ heat.
    4.2.2 Life cycle inventory of transport fuels - supply of 1GJ
    4.2.3 Life cycle inventory of transport fuels - supply and use in transport per person-km (20% diesel, 80% unleaded petrol).
    4.2.4 Life cycle inventory of electricity supply - per GJ delivered.
      4.2.4.1 The life cycle of nuclear generated electricity
    4.2.5 Life cycle inventory of CHP provision - supply of 0.64 GJ heat, 0.36 GJ electricity
  4.3 Appendix C. More detailed explanation of the environmental impact categories.
  4.4 Glossary
  4.5 Nomenclature and Acronyms
  4.6 References

Footnotes

1. Introduction

This report summarises the work which has been carried out at ETH¹ in Zurich, detailing the flows of materials and energy in the life cycles of European (excluding the UK) energy systems. Although UK energy systems are not included specifically in the ETH study, many of the systems are similar to those found in the UK. This information is used in this report to examine the environmental impacts of various energy systems using life cycle assessment (LCA).

1.1 Life Cycle Assessment

LCA is a method which quantifies the environmental impacts associated with the delivery of a particular service or product. An accepted definition is:

'LCA is a systematic way of evaluating the environmental impact of products or activities by following a 'cradle to grave' [raw material extraction to waste disposal] approach. This approach implies the identification and quantification of emissions and material consumption which affect the environment at all stages of the entire product life cycle'²

Thus the analysis includes all processes connected to the delivery of this service or product (also termed the functional unit), from the extraction of raw materials to the disposal of wastes. This network of processes forms the life cycle of the product or service. Using this method ensures that all processes which contribute to the environmental impacts of the delivery of a particular service or product are included in the final result. This produces an unambiguous picture of the overall impacts in certain environmental categories of a particular energy system, allowing the comparison of different energy systems to be made on a consistent basis.

The life cycles of energy systems will therefore include processes such as raw material extraction, transportation, intermediate processing and delivery. Figure 1-1 shows those processes included in the life cycle delivering heat from oil to domestic users.

For each process within the life cycle detailed inventories of the material and energy inputs and outputs are produced. This includes all the raw material inputs, waste outputs, energy inputs and useful product outputs of each process. In this way a life cycle inventory (LCI) is produced which accounts for the total inputs and outputs of all energy and material flows attributable to the provision of a particular service or product. The life cycle inventory categories are shown and explained in Appendix A.

These inventories consist of a large number of inflows and outflows which in themselves are difficult to relate to individual environmental problems. To make the results intelligible and relevant these material and energy flows have to be translated into environmental impacts. This is the final assessment stage of the LCA. This relates the overall life cycle inventory to the effect on the environment of providing a certain service or product. The environmental impacts considered can vary according to the purpose of the study. In this report the impact categories are limited to the four categories of: resource depletion, greenhouse gas emission, acidification and eutrophication, shown in Table 1-1. These environmental impacts are of most relevance to energy systems. Other factors which are also often considered in impact assessments are: human toxicity, ozone depletion potential, photochemical oxidant formation and ecological toxicity. However these are not considered in this report as energy systems do not generally contribute significantly to these impacts.

Figure 1-1 Processes in the life cycle of oil-fired domestic heating

The contribution of each input or output flow in the life cycle inventory to each of the environmental impact categories is calculated by multiplying by a relevant factor. Thus an output of CO₂ contributes a certain amount to global warming potential, CFC emissions to both greenhouse gas emissions and ozone depletion potential, and SO₂ and NO_x emissions to acidification potential. In each case the impact described is equal to the maximum impact on the environment which could be caused by the particular emission or consumption of resource. The conversion factors are shown in Table 1-2 for the impact categories considered in this report.

The end result of this procedure is to produce graphs of the environmental impacts in each category caused by the provision of the product or service. This information can be used in two ways. Firstly, when looking only at one life cycle, it can be used to assess which particular processes in the life cycle are responsible for the most significant environmental impacts. This allows improvements to the system to be targeted where they are likely be most effective at reducing each category of environmental impact.

Secondly, the results can be used to compare the environmental impacts of systems with the same functional units. This will provide answers to the questions of which energy system contributes most to each environmental impact category. It is with this aim in mind that LCA is used to describe energy systems in this report.

Thus LCA will give a clear and unambiguous account of the environmental impacts of energy systems and will allow these systems to be compared on a clear and consistent basis. It is apparent though, that there will be occasions when value judgements have to be made regarding which system is 'better' environmentally than another. For instance system A may have very low emission of greenhouse gases, but a higher acidification effect relative to system B. There is no strict procedure in LCA in dealing with such dilemmas and this has to be left as an assessment of priorities dependent on the aims of the interested parties.

Table 1-1 Environmental impact categories.³

Table 1-2 Multiplication factors for the conversion of inventory flows to environmental impacts

In addition, while LCA covers the main areas of environmental impacts, there are some important environmental effects which cannot be captured by LCA. These include site-specific questions which are very dependent on the particular geography or population surrounding a process installation. These precise effects are addressed more effectively by environmental impact assessments (EIA). Examples would include noise pollution, the exact impacts of emissions on local populations, effects of localised resource depletion, e.g. water and soil. The effects of catastrophic accidents, e.g. oil tanker spills and nuclear accidents, are also not included in the analysis. These should be separately addressed using risk assessment techniques.

Also, there is the important environmental impact due to radioactive releases to the environment which is not yet satisfactorily dealt with by LCA. This is because of the difficulties in quantifying the human and ecological-toxicity of the diverse range of radioactive emissions and in satisfactorily including the longevity of some of these effects. Work is continuing in this area⁹, but at present, because it is not adequately quantified, this impact category has not been included in the results of this report except at the inventory stage detailing the radioactivity of emissions. Radioactive emissions are especially relevant to nuclear-derived electricity, and the omission of this impact category should be borne in mind when considering the results based on the four impact categories mentioned above.

1.2 The ETH study and its use in this report

The ETH study has produced detailed life-cycle inventories of energy systems supplying Switzerland and those European countries connected to the European electricity grid (UCPTE)¹⁰. It is intended for use as an information source for the analysis of the environmental impacts of energy systems in these countries. This report summarises some of those energy systems covered in the study which are relevant to the UK. These are:

• The supply and use of domestic heating fuels: heating oil and natural gas;

• The supply and use of transport fuels: unleaded petrol and diesel;

• The supply of electricity to domestic consumers, derived from: coal, gas, oil, nuclear fuels and solar radiation;

• The supply of heat and electricity to domestic consumers using natural gas through three different systems.

These energy systems are grouped according to their functional units. These are: the delivery of 1 GJ of heat, 1km traveled in a car, 1GJ of electricity supplied, and 1 GJ of heat and power supplied to consumers, respectively.

It should be noted however that because the ETH study does not specifically examine UK energy systems, the results will not be precisely representative. As far as possible examples will be used which are similar to the UK situation. Thus energy systems will, where possible, be chosen which derive their oil and gas from the North Sea. Otherwise the European average mix of these energy systems is used and differences highlighted in the relevant section.

One area where there are substantial differences between the UK and continental Europe is in the fuel base for the provision of electricity. Table1-3 shows that the mix of electricity generators supplying to the European grid differs from that of the UK. Particular differences are the absence of combined cycle gas turbine (CCGT) power stations in Europe and the absence of brown coal power stations in the UK. For this reason the ETH study has not covered CCGT stations - a major omission when considering UK electricity provision. The relevance of each electricity energy system to the UK is discussed in later sections.

The life cycle inventory data given in the ETH study is converted into environmental impacts using the factors in Table 1-2. Each of the following sections will compare the life cycle inventories, and also the environmental impacts of the energy systems. Individual tables detailing the inventories and descriptions of each energy system are given in Appendix B.

Table 1-3 Comparison of UCPTE and UK electricity generating mixes.

2. Life Cycle Assessment of Energy Systems

2.1 Heating Fuels

This section compares the life cycles of heating fuels derived from crude oil with those from natural gas. The functional unit used is the delivery of one GJ of heat to domestic consumers. Thus the efficiency of the boiler in using the heat released by combustion is taken into account.

The heating oil is a low sulphur fuel based on the European average extraction and supply system. The oil is supplied to domestic users to fuel a 100kW boiler to provide the 1GJ of heat. The main differences between the average European supply of heating oil and that in the UK will arise in distribution from point of extraction to the refinery, and in distribution from the refinery to the consumer. It is likely that the distances will be less in both cases for the UK. The boilers are be similar to those found in the UK. The processes included in the life cycle are shown in Figure1-1.

The natural gas energy system is based on the Dutch situation. The gas is extracted from the North Sea, similarly to gas supplied to the UK, and is burnt in a 100kW conventional natural gas boiler. The main stages in the provision of domestic heat from gas are shown in Figure 2-1-1.

Figure 2-1-1 Process stages in the provision of domestic heat from natural gas.

The comparison looks at both the supply of these fuels to the point of use (point A in Figure 2-1-1), and at the overall life cycle, i.e. supply and conversion into heat. This will show the contributions to environmental impacts of the supply chain relative to the use stage. Both the life cycle inventories and the environmental impacts are detailed.

First the LCIs for the supply of heating fuels are considered. Figure 2-1-2 shows that in terms of land area use, electricity consumption, material use and distance transported, the natural gas supply uses less. This is mainly due to the extra infrastructural requirements of oil transport relative to North Sea gas. Figure 2-1-3 details the emissions to air from the supply of each fuel. Again, there are greater emissions from the heating oil supply. These arise mainly in the refining of the crude oil and the extraction of the oil. Methane emissions occur principally in the extraction of the crude oil. Perhaps surprisingly, there is relatively little methane release in the supply of natural gas: leakages were estimated to be about 2% of natural gas production, although the accuracy of this figure is admitted to be uncertain. The higher CO₂ emissions from heating oil supply arise principally from energy use in the refinery and also from gas flaring during extraction.

Figure 2-1-4, shows that in all categories except mercury releases, the emissions to water were also higher for heating oil supply, which is mainly due to the extraction and refining stages.

Figure 2-1-2

Figure 2-1-3

Figure 2-1-4

Comparing the LCI for the supply of the gas relative to the LCI for the delivery of 1GJ of heat to consumers - including supply - (Figure 2-1-5, 6 & 7), shows that the emission of CO₂ in the supply of both fuels is very small in comparison to that emitted in use: CO₂ emitted from supply 13.5kg/GJ heat and 0.5kg/GJ heat; CO₂ emitted from use 84kg/GJ heat and 72 kg/GJ heat for heating oil and natural gas respectively. For the heating oil the combustion of the fuel for heat leads to higher NO_x and mercury emissions to air. For the natural gas the land area use, and total materials use are higher. This is because the total life cycle is based on the use of the gas in Switzerland, increasing the land and infrastructural requirements, relative to the supply which was based on supply to the Netherlands. These factors are geography-dependent and so would not necessarily reflect the UK situation. More general factors such as the emission of NO_x and CO₂ will be representative of similar energy systems in the UK.

The combustion of the natural gas in the boiler is the main cause of the CO₂, NO_x and CH₄ emissions to air and in most categories of emissions to water. In total 29kg of crude oil and 30kg of natural gas are required to supply 1GJ of heat.

Figure 2-1-5

Figure 2-1-6

Figure 2-1-7

Figure 2-1-8 shows the environmental impacts of both the delivery of fuels, and the overall impacts including delivery and use in providing 1 GJ heat to domestic consumers. Resource depletion is higher for the fuel oil, which reflects the shorter reserve life for oil relative to natural gas. The greenhouse gas emissions are mainly concentrated in the conversion of the fuel into heat and the impact from both fuels is similar. This suggests that improvements in the efficiency of boilers could reduce greenhouse gas emissions in both cases and that switching fuels would achieve little.

Acidification due to air emissions is higher for the heating oil, with most of the emissions occurring in the supply chain. For natural gas the acidification is caused mainly by combustion and subsequent NO_x emission in the boiler. Eutrophication reflects a similar pattern.

Figure 2-1-8

2.2 Transport fuels

The transport fuels considered are unleaded petrol and diesel. The systems for the supply of these fuels to the consumer, i.e. at the petrol pump, are calculated according to the European average. These will not be dissimilar to the UK equivalent energy systems, although the sources of the crude oil are slightly different with imports from North Africa forming a substantial part of the European source of crude oil. The UK will draw a larger portion from the North Sea, leading to lower transport distances.

The life cycles for supply of unleaded petrol and diesel follow similar stages to that of the heating oil. The main difference between the processes for the two transport fuels is in the refining stage where catalytic cracking is used to make the petrol from heavier fuel fractions in the refinery. This is in addition to distillation of crude oil which is used for both petrol and diesel. Also the processing of lead-substitutes (MBTE and TEL) from naphtha in the refinery is also included in the supply of unleaded petrol. These extra processes combined make the refining stage more energy intensive for the petrol. Apart from refining, the life cycles of the two fuels are almost identical. These LCIs are shown for the supply of 1GJ of fuel (or 23.4kg¹² unleaded petrol and diesel) in Figures 2-2-1 to 2-2-3.

Figure 2-2-1

Figure 2-2-2

Figure 2-2-3

These figures show that in almost all categories the unleaded petrol slightly exceeds the diesel: 28kg of crude oil are required to supply 1GJ of unleaded petrol; for diesel this figure is 26kg. Other large differences arise in the emission of NO_x to air (petrol 80g/GJ, diesel 62g/GJ), due to the extra refining processes for unleaded petrol. Overall though, the supply of the two fuels use and emit the same substances in similar quantities.

Given that these fuels have broadly similar environmental interventions, the comparison of interest is between the supply of these fuels and their use in transport. The ETH study quantifies the main emissions from car use in the provision of one person-km in a typical car. Unfortunately this is not fuel specific, but is based on the European average mix of fuels (20% diesel, 80% petrol; UK average: 20.5% diesel, 79.5% petrol). It does however indicate the relative contributions of supply and use to the life cycle inventory of transport fuels in those categories covered. These are shown in Figure 2-2-4. This shows that 78% of the CO₂, and 88% of the NO_x are emitted in the use of the fuel, while 82% of total methane emissions occur in the supply of the fuels to use. Emissions of non-methane volatile organic compounds (NMVOC) are almost entirely due to the use of the fuels.

This LCI accounts for only a small selection of substances, but does show the relative contributions of supply and use in the emissions of some important substances. Converting into environmental impacts (Figure 2-2-5) re-emphasises the significance of the use phase: 73% of greenhouse emissions, 70% of acidification and 88% of eutrophication are associated with the use of the fuel in cars. This implies that strategies to reduce these environmental impacts in the transport fuel life cycle should targeted at this stage. Any reductions in the environmental impacts achievable by switching fuels in the use phase, i.e. from unleaded petrol to diesel, cannot be currently addressed with the ETH data.

Figure 2-2-4

Figure 2-2-5

2.3 Electricity

This section covers the life cycle assessment of the provision of 1GJ of electricity to consumers. This functional unit is not exactly equivalent between energy systems. For instance, the supply of electricity from fossil-fuel fired and nuclear power stations can be provided on demand. This is not the case for photovoltaic solar-generated electricity (without energy storage) which is reliant on daylight and prevailing weather conditions. There are therefore differences between these means of electricity generation which might have implications for their use in particular applications. These should be considered alongside the results of the analysis.

Five methods of electricity provision are considered fueled by oil, coal, natural gas, nuclear fuel, and solar radiation. Except for photovoltaic electricity, these are based on the performance of average technologies in particular countries. The photovoltaic electricity life cycle is based on a specific power station.

The transmission of electricity in all cases is taken to be distribution from the power station via the high voltage electricity grid to low voltage electricity suitable for domestic use, causing a loss of 13.8% of the electricity produced at the power station. Because this figure is used in all the electricity life cycles it will not affect the relative results but will increase their absolute values.

The provision of electricity from fossil fuels is considered first.

2.3.1 Electricity from Fossil Fuels

2.3.1.1 Electricity from Oil

The analysis of the electricity production from oil-fired power stations is based on the European average supplying the UCPTE. Proportionately Italy is the main supplier of oil generated electricity (72%), followed by Portugal, Germany and Greece (with about 5% each). The type of fuel is assumed to be high sulphur heating oil for all plants (variations due to the use of different fuels e.g. Germany uses 31% diesel and light heating oil as fuels, are assumed to be negligible.)

In the power station stage an oil-air mixture is combusted producing heat which is used to raise steam in a water circuit. This steam is used to drive a turbine which in turn drives an alternator to produce electricity.

The overall energy and material requirements for the oil-fired power stations were calculated by considering the average energy efficiencies of oil-fired generation in each country (Table 2-3-1). Using this efficiency and information on the emissions released from the combustion of heating oil, the overall emissions and resource use per GJ of electricity produced were calculated.

It should be noted that these efficiencies are the measured efficiencies in operation calculated from the amount of electricity produced divided by the amount of fuel supplied. These will be lower than the design, or achievable, efficiencies due to operational practices which keep certain types of power station on-line, without generating electricity in order to prevent spikes in demand overloading the grid. Thus, these efficiency figures and those for other power stations described later do not represent the highest efficiencies possible with those technologies.

Table 2-3-1 Energetic efficiencies of selected European Countries' Oil-fired Power Stations.

2.3.1.2 Electricity from Coal

The inventory for production of electricity from coal is calculated in the ETH report from data on various individual coal-fired plants in each country. The system described here is based on the French hard coal-fired electricity generation system, because of its similarity to the UK system. The UK does not have any brown coal power stations so only hard coal is considered. The main processes considered in the life cycle are shown in Figure 2-3-1.

Figure 2-3-1 Main process steps in the production of electricity from coal.

The power station considered is a conventional thermal power station, operating in a similar way to the oil-fired power stations considered earlier. The French plants considered have 16% (on a basis of MW installed capacity) of flue gas de-sulphurisation units which take out the SO₂ from the exhaust gas. This is similar to the UK percentage of about 10%¹³. There was no de-NO_x equipment fitted to the plants.

2.3.1.3 Electricity from Gas

The gas supply system is identical to that described above for the supply of gas to consumers except that the power stations take their gas direct from the high pressure gas network. Therefore the infrastructural requirements of the local supply network are not included in the life cycle. The gas is used to produce electricity in conventional thermal power stations, similarly to the coal and oil-fired power stations, using the combustion of the gas to raise steam at around 500 ⁰C which drives a steam turbine to produce electricity. The results are based on the UCPTE average for gas generated electricity.

As the UK has a large percentage of the more efficient CCGT stations, the results of this analysis will not be typical for the UK. Therefore the UK environmental impacts for gas-fired electricity generation are likely to be lower than those for the European average due to the higher efficiency in these CCGT plants. In addition in the rest of Europe blast-furnace and coke-oven gases from steel plants are used to generate about 25% of the gas-fired electricity. This too is not representative of the UK where these gases only make up a small fraction (<1%) of the gas-fired electricity supplied nationally.

2.3.1.4 Comparison of Fossil Fuel-generated Electricity

The life cycle inventory comparing oil, coal and gas generated electricity is shown in Figures 2-3-2 to 2-3-4. Overall the delivery of 1GJ of electricity to domestic consumers requires 83kg of oil, 210kg of coal or 67 m³ (56kg) of natural gas. For oil, coal and gas-derived electricity the emissions of CO₂ are 260kg, 315kg and 277kg; for SO₂ 2.7kg, 1.2kg and 0.1kg; and for NO_x 0.57kg, 0.51kg and 0.46kg, respectively.

Comparing the environmental impacts of the three fossil fuel electricity life cycles (Figure 2-3-5) shows that in all factors apart from resource depletion the impact from gas-generated electricity is smallest. Resource depletion is smallest for coal, reflecting the much larger reserves of coal than oil or gas: even though more coal is needed in mass terms per GJ of electricity generated, relative to the size of its resource base it causes less resource depletion. Coal however, generates the most greenhouse gases in the production of electricity and also produces the highest eutrophication effect. Oil produces the heaviest impact in acidification. This is partially due to the fuel refining stage which is not a feature of the other two methods of electricity supply, but is also due to the lack of desulphurisation and de-NO_x equipment on the oil plants considered. If these were fitted to the oil-fired plants one would expect a reduction in acidification.

Figure 2-3-2

Figure 2-3-3

Figure 2-3-4

Figure 2-3-5

2.3.2 Electricity from Nuclear Fuel

The life cycle of electricity from nuclear fuel includes all processes from the extraction of uranium ore to the production of electricity from nuclear fission reactions in the power station and subsequent reprocessing and storage of wastes. Two types of reactor are considered in the ETH study: the boiling water reactor (BWR) and the pressurised water reactor (PWR). Of these only the PWR is used in the UK (Sizewell B). The other types of UK reactor are the advanced gas reactor (AGR) and Magnox power plants. These are peculiar to the UK and so have not been covered in the ETH study. Therefore only electricity generation from the PWR is considered here.

The stages in the life cycle of PWR generated electricity are shown in Figure 2-3-6.

Figure 2-3-6. Stages in the life cycle of the production of nuclear electricity.

The main processes in the life cycle are described in more detail in Appendix B.

2.3.3 Electricity from Solar Power

Photovoltaic cells convert solar radiation directly into electricity using semiconductor technology. Currently these cells operate with efficiencies of around 13-16% and are looked upon as one of the renewable energies with considerable future potential. However although the fuel, solar radiation, is free and inexhaustible there are environmental impacts associated with this means of generating electricity due to the actual construction and installation of the cells.

The ETH study has looked at many different types of individual photovoltaic installations. Here the analysis of the PHALK solar power station situated on the Mont Soleil in Switzerland is selected for direct comparison with the other electricity supply methods described above. The PHALK is smaller scale than previous methods of electricity supply producing a peak output of 500kW using mono-crystalline silicon cells. The UK receives less solar radiation than this location so the environmental impacts are likely to be slightly higher for a similar installation in the UK.

Following the life cycle methodology all the energetic and material inputs and outputs to the manufacture of the solar cells, the construction of the structures and infrastructure, and the maintenance of the power station are included in the analysis.

2.3.4 Comparison of all methods of electricity generation

The inventory results for the nuclear life cycle and for photovoltaic electricity are shown in Figures 2-3-7, 8 & 9, with those for fossil fuel-fired electricity included for comparison.

This shows (Figure 2-3-7) that 9g of uranium is used per GJ supplied by PWR. In addition nuclear-derived electricity causes more disruption to relatively undisturbed land (Land area II-III) than other methods. This also accounts for the fact that land occupied for mining and preparation of radioactive materials cannot be used for other purposes for 80 000 years after decommissioning. Emissions to air (Figure 2-3-8) from the PWR in all categories except radionucleide emission are insignificant relative to those from fossil fuels. This is principally because of the absence of combustion processes in the nuclear cycle. Radionucleide emissions to air are more than 20 times those for photovoltaic electricity - the largest of the other methods in this category. Emissions to water (Figure 2-3-9) follow a similar pattern: generally the PWR electricity is lower in all categories except in the activity of radioactive releases which are between 10 and 100 times greater than for other methods of electricity provision.

The life cycle inventory of photovoltaic electricity shows that although the primary fuel source - solar energy - is renewable, there is still some non-renewable resource use. 4kg of oil, 2 m³ of natural gas, and 10 kg of coal are used in the generation of 1GJ of photovoltaic electricity. This occurs because the manufacture of solar cells uses significant amounts of electricity from the European grid (UCPTE) which is mostly produced from non-renewable fuels. This dependence on non-renewable electricity supply in manufacturing also means there are non-trivial emissions to the environment from electricity generated from photovoltaics.

However it is worth mentioning that if there were a higher proportion of photovoltaics (or other renewable energies sources) used in the production of electricity the overall environmental impacts of this type of electricity generation would fall. Frischknecht et al. estimate in the ETH report that the environmental impacts would fall by around 40% if all the electricity were generated by photovoltaics.

Material use is highest for photovoltaics, reflecting the large infrastructural requirements per GJ for this particular stand-alone photovoltaic installation. This would be much reduced for photovoltaics which are integrated into building facades, which is seen as the most economically effective method of photovoltaic electricity generation¹⁴.

Gaseous emissions from photovoltaic electricity are in all cases, except SO₂ and NMVOCs, much lower than for fossil fuel electricity; yet also considerably higher than PWR electricity, radioactive emissions aside. The NMVOCs are emitted in the manufacture of the solar wafers. Emissions to water follow a similar pattern.

Figure 2-3-7

Figure 2-3-8

Figure 2-3-9

Overall the inventories show that, excepting radioactive emissions, the fossil fuel-derived electricity produces greater interventions in all categories than PWR and photovoltaics electricity. Converting these inventories to environmental impacts reinforces this picture.

Comparing all the life cycles across impact categories (Figure 2-3-10) shows that in these four categories the PWR electricity cycle has the lowest impact. The fossil fuel-derived electricity contributes most to the greenhouse effect, while oil and coal have the largest impact on acidification. Of the fossil fuels, natural gas-fired power stations (not CCGT) have the lowest impacts in all categories except for resource depletion where the larger coal reserves relative to oil and gas reserves mean that coal-fired electricity has the lowest impact in this category. The not-insignificant impact of photovoltaics is also demonstrated, although these result to a large extent from the impacts of current methods of electricity production and (except for acidification) are much lower than fossil fuel-fired electricity generation.

While these results accurately quantify the effects of these energy systems on these four impact categories, there are other environmental impacts which are not shown either because they are difficult to relate directly to the emission of substances in this type of analysis, or because not enough is known about the long term effects of particular emissions. Examples of the former would include impacts such as the risk of catastrophic accidents and the effects of visual intrusion. The effect of long-lived nuclear isotopes on the environment is an example of the latter. This should be borne in mind when examining these results, as should differences between the electricity production systems examined here and those in the UK. If more accuracy is required, further, more detailed examination of the processes relevant to the UK is necessary.

Figure2-3-10

2.4 Combined Heat and Power Systems

This section analyses the supply of thermal energy and electricity to end users. In particular the aim is to analyse the performance of combined heat and power systems in relation to other means of delivering these two forms of energy.

Three systems are analysed all using natural gas as the main energy source and each providing 0.36GJ electricity and 0.64 GJ heat¹⁵ to consumers (Figure 2-4-1):

1. Conversion of natural gas in a conventional power station to electricity which is delivered to consumers. Part of this electricity is then converted to heat through electrical resistance heating in electric radiators.

2. Delivery of natural gas to consumers which is converted to heat in domestic boilers. The electricity supply is provided from natural gas-fired power stations.

3. The supply of electricity and heat from a natural gas-fired combined heat and power (CHP) plant. Both the electricity and the heat are produced together and supplied directly to consumers with no need for conversion at the point of use.

Most of the processes in these energy delivery systems have been analysed in previous sections. The electric heater, the natural gas heater and the CHP plant itself are introduced here.

The electric heater is assumed to have a 100% efficiency of conversion of electricity to heat. The gas heater in the second process fires a condensing boiler with an energy efficiency of 97%. This boiler is more efficient than that described in Section 2-1. The CHP plant is based on a gas-fired Otto cycle motor which drives a generator for electricity production¹⁶. The heat for export to consumers is produced from heat exchange with the motor cooling water and with the motor exhaust. A heat pump driven by the electrical output is also used to provide additional heat. The power rating of this installation is 150kWe and would be used to provide heating for a large block of flats. The heat loss in the distribution system is small and is therefore neglected.

Figure 2-4-1 Three methods of supplying heat and electricity to consumers

The results of this comparison (Figures 2-4-2, 3 &4) reflect the differences in each method of heat and electricity provision. In almost all categories Process 1 uses and emits larger quantities followed by Process 2. Only in SO₂ emissions, some water emissions, and transport t-km does CHP (Process 3) exceed the other processes. The former is because of differences in the gas used in the CHP plant (high sulphur content) relative to that used in the other processes. If the same gas were to be used one would expect the CHP process to produce lower SO₂ emissions. The reasons for the higher water emissions and transport distances for the CHP are not made clear in the ETH report, however these have an insignificant effect on the overall results.

For the provision of 1GJ of heat and electricity the different methods require 50kg, 37kg and 23kg of natural gas for processes (1), (2) and (3) respectively. It is evident that process (1) has the largest environmental impact. This is due to the low energy efficiency of conversion of gas to electricity in the power station (UCPTE average =39% efficiency). The potential of the gas to heat directly at 97% efficiency has been wasted by burning the gas in the power station, then converting to electricity, and then again converting this electricity to heat. Therefore in total, much larger quantities of gas have to be combusted to provide the same amount of electricity and heat, and this leads to higher environmental impacts.

The difference between processes (2) and (3) are smaller but still indicate that the system with the least environmental impact in these categories is the one which generates heat and power together in a CHP plant¹⁷. For reasons to do with the thermodynamic conversion of the chemical energy in the natural gas, it turns out to be more efficient to produce electricity and heat in a single process from a fuel rather than producing them separately. As a result of this the CHP (3) system has clear environmental advantages over the other systems.

Figure 2-4-2

Figure 2-4-3

Figure 2-4-4

Figure 2-4-5

3. Conclusion

This report has used the ETH study to compare energy systems supplying three different functional units to consumers:

1. One gigajoule thermal energy supplied to consumers from heating oil and natural gas.

2. One person-km transported using petrol and diesel fuels.

3. One gigajoule of electricity from: oil, coal, gas, nuclear and photovoltaic power stations

4. One gigajoule total of electricity and heat (36% and 64% respectively) by three different means.

Table 3-1 Summary of results of LCA of energy systems

While the ETH data was found to be detailed and comprehensive it is not based on UK energy systems and if greater accuracy were to be required for particular UK energy systems separate inventories for these would have to be drawn up. Particular omissions from the ETH data concerning the UK are the combined cycle gas turbine, Magnox, and AGR energy systems. These systems apart though, the main conclusions of this report are likely to apply equally to UK energy systems.

Finally, it should be emphasised that LCA is just one tool used for the assessment of particular environmental impacts of an energy system. Many environmental impacts are not yet quantifiable in LCA. For energy systems the main omission is the impact of radioactivity on the environment and therefore LCA does not yet adequately deal with nuclear derived power.

4. Appendices

4.1 Appendix A Life cycle inventory categories

This appendix summarises the main inventory categories used by ETH.

4.2 Appendix B Life cycle inventories of energy systems

4.2.1 Life cycle inventory of heating fuels - supply and use for 1GJ heat.

4.2.2 Life cycle inventory of transport fuels - supply of 1GJ

4.2.3 Life cycle inventory of transport fuels - supply and use in transport per person-km (20% diesel, 80% unleaded petrol).

4.2.4 Life cycle inventory of electricity supply - per GJ delivered.