Life Cycle Assessment and Its Application To Process Selection, Design and Optimisation

Review article
Life cycle assessment and its application to

process selection, design and optimisation
Adisa Azapagic
*
Department of Chemical and Process Engineering, University of Surrey, Guildford, Surrey GU2 5XH, UK
Accepted 1 March 1999
Abstract
As the pressures on the chemical and process industries to improve their environmental performance are increasing, the need to move
away from narrow system denitions and concepts in environmental system management is becoming more apparent. Life Cycle
Assessment (LCA) is gaining wider acceptance as a method that enables quantication of environmental interventions and evaluation of the
improvement options throughout the life cycle of a process, product or activity. Historically, LCA has mainly been applied to products;
however, recent literature suggests that it can assist in identifying more sustainable options in process selection, design and optimisation.
This paper reviews some of these newly emerging applications of LCA. A number of case studies indicate that process selection must be
based on considerations of the environment as a whole, including indirect releases, consumption of raw materials and waste disposal. This
approach goes beyond the present practice of choosing Best Practicable Environmental Option (BPEO), by which it is possible to reduce the
environmental impacts directly from the plant, but to increase them elsewhere in the life cycle. These issues are discussed and demonstrated
by the examples of end-of-pipe abatement techniques for SO
2
, NO
x
and VOCs and processes for the production of liquid CO
2
and O
2
. The
integration of LCA into the early stages of process design and optimisation is also reviewed and discussed. The approach is outlined and
illustrated with real case studies related to the mineral and chemical industries. It is shown that a newly emerging Life Cycle Process Design
(LCPD) tool offers a potential for technological innovation in process concept and structure through the selection of best material and
process alternatives over the whole life cycle. The literature also suggests that LCA coupled with multi-objective optimisation (MO)
provides a robust framework for process design by simultaneously optimising on environmental, technical, economic and other criteria.
Pareto-optimum solutions obtained in MO provide a number of options for improved design and operation throughout the whole life cycle.
This approach therefore provides a potentially powerful decision making tool which may help to identify more sustainable solutions in the
process industries. # 1999 Elsevier Science S.A. All rights reserved.
Keywords: Life cycle assessment; BPEO; BATNEEC; Waste minimisation; Multi-objective optimisation; Process design; System analysis
1. Introduction
The development of industrial technology has enabled the
transformation of the environment in different ways, chan-
ging the nature and extent of the environmental impacts of
industrial activities. Resource depletion, air, water and land
pollution, are examples of the environmental problems
which have emerged as a result of intensied interventions
into the environment. One of the main problems associated
with these activities is that they may not have an immediate
effect and some may have a more global impact on the
environment. This is becoming apparent with the increasing
scientic awareness of the cumulative and synergistic
effects of some of the environmental impacts over space
and time. For instance, emissions of greenhouse gases can
occur locally, but the resulting greenhouse effect will have a
global character.
It is therefore not surprising that pressures on those
responsible for the environmental interventions to improve
their performance are rising. Among these, the chemical and
process industries nd themselves constantly under the
scrutiny of various pressure groups demanding more envir-
onmentally acceptable processes, products and practices
through the ideas of `waste minimisation', `zero emission',
`producer responsibility', etc. One of the potential dangers
of this is that the companies exposed to environmental
pressures may simply respond to satisfy a particular group.
However, this short-term approach may lead to costly long-
term mistakes with little environmental improvement and no
Chemical Engineering Journal 73 (1999) 121
*Tel.: +44-1483-259-170; fax: +44-1483-259-510; e-mail:
[email protected]
1385-8947/99/$ see front matter # 1999 Elsevier Science S.A. All rights reserved.
PII: S1 3 8 5 - 8 9 4 7 ( 9 9 ) 0 0 0 4 2 - X
net business benet. To avoid this, environmental issues
must be assessed in a holistic way, alongside nancial,
technical and other criteria.
Life Cycle Assessment (LCA) is an environmental man-
agement tool that enables quantication of environmental
burdens and their potential impacts over the whole life cycle
of a product, process or activity. Although it has been used in
some industrial sectors for about 20 years, LCAhas received
wider attention and methodological development only since
the beginning of the 1990s when its relevance as an envir-
onmental management aid in both corporate and public
decision making became more evident. Examples of this
include incorporation of LCA within the ISO 14000 Envir-
onmental Management Systems (EMS) [1], EU Eco-Man-
agement and Audit Schemes (EMAS) [2]
1
, and EC
Directive on Integrated Pollution Prevention and Control
(IPPC) [3,4] which require companies to have a full knowl-
edge of the environmental consequences of their actions,
both on- and off-site.
Integration of life cycle thinking into environmental
system management started to change the way environmen-
tal problems were seen and tackled. It pointed out that, if
sustainable solutions to environmental problems are to be
found, then they must be sought on a more global level.
Today, LCA is being used widely as a decision making tool;
however, the methodology is still developing and a number
of issues remain to be resolved. This paper reviews the state-
of-the-art of methodological development and uses of LCA.
In particular, it focuses on the application of LCA in process
selection, design and optimisation as a tool for identifying
clean technologies [58]. The procedures for incorporating
into the system optimisation framework the environmental
criteria alongside the economic and technical criteria are
reviewed and discussed. It is shown that this approach can
provide a potentially powerful decision making tool for
managers, process engineers and designers.
2. Life cycle assessment
Life Cycle Assessment is a technique for assessing the
environmental performance of a product, process or activity
from `cradle to grave', i.e. from extraction of raw materials
to nal disposal. Today's LCA originates from `net energy
analysis' studies, which were rst published in the 1970s
[911] and considered only energy consumption over a life
cycle of a product or a process. Some later studies included
wastes and emissions [1216], but none of them went
further than just quantifying materials and energy use. At
this point it was clear that a more sophisticated approach to
complex environmental issues was needed.
As a result, in 1990, the Society for Environmental
Toxicology and Chemistry (SETAC) initiated activities to
dene LCA and develop a general methodology for con-
ducting the LCA studies. Soon afterwards, the International
Organisation for Standardisation (ISO) started similar work
on developing principles and guidelines on the LCA meth-
odology [17]. Although SETAC and ISO worked indepen-
dently of each other, a general consensus on the
methodological framework between the two bodies has
started to emerge, with the difference being in the matter
of detail only. While the ISO methodology is still being
shaped, the methodology developed by SETAC remains
widely accepted among LCA practitioners. The latter is
briey described in the following section with reference to
the ISO methodology where appropriate.
2.1. Methodological framework for Life Cycle Assessment
Life Cycle Assessment, as dened by SETAC, is `à
process to evaluate the environmental burdens associated
with a product, process, or activity by identifying and
quantifying energy and materials used and wastes released
to the environment; to assess the impact of those energy and
material uses and releases to the environment; and to
identify and evaluate opportunities to effect environmental
improvements'' [18,19]. It follows the life cycle of a
product, process or activity from extraction of raw materials
to nal disposal, including manufacturing, transport, use,
re-use, maintenance and recycling (Fig. 1). Its main advan-
tage over other, site-specic, methods for environmental
1
Although LCA is not mentioned explicitly in EMAS, its use is implied
in Annex 1, section C (see [2]). Fig. 1. Stages in the life cycle of a product (from Azapagic [20]).
2 A. Azapagic / Chemical Engineering Journal 73 (1999) 121
analysis, such as Environmental Impact Assessment (EIA)
or Environmental Audit (EA), lies in broadening the system
boundaries to include all burdens and impacts in the life
cycle of a product or a process, and not focusing on the
emissions and wastes generated by the plant or manufactur-
ing site only.
The methodological framework for conducting LCA, as
dened by both SETAC [18] and ISO [17], comprises four
main phases. The two approaches are compared below:
SETAC [18] ISO-14040 [17]
1. Goal Definition and Scoping Goal and Scope Definition
(ISO14041) [21]
2. Inventory Analysis Inventory Analysis
(ISO14041) [21]
3. Impact Assessment Impact Assessment
(ISO14042) [22]
4. Improvement Assessment Interpretation
(ISO 14043) [23]
As indicated, the methodological framework proposed by
ISO is similar to that defined by SETAC; the only sub-
stantial difference is noted in the final phase, as discussed
below. The interactions among the LCAphases are shown in
Fig. 2.
LCA is based on the kind of thermodynamic and system
analyses which are central to process engineering [24].
Therefore, the rst step in any analysis must be denition
of the system under study. In LCA, this is done in the Goal
Denition and Scoping phase [18]. The environment is then
interpreted in the thermodynamic sense as `that which
surrounds the system', i.e. the whole universe except the
system under study. Thus for these purposes, `the environ-
ment' is dened along with the system, by exclusion. On this
basis, Fig. 3 shows schematically the general problem of
environmental system analysis. The system of interest exists
because it produces goods and services, which are treated
together as outputs. To generate these outputs, inputs of
energy and materials are required. In a site-specic envir-
onmental analysis, such as EIA or EA, the system is the
plant or manufacturing site and the inputs are related to the
inputs of material and energy to that plant. In the LCA
context, system boundaries are drawn from `cradle to grave'
to include all burdens and impacts in the life cycle of a
product or a process, so that the inputs into the system
become primary resources.
The system function is also specied within Goal Deni-
tion and Scoping and it is expressed in terms of the func-
tional unit(s) as a measure of the function(s) that the system
delivers. For instance, the function of packaging is to store a
certain amount of liquid. If different packaging is to be
compared, then the comparison should be based on an
equivalent function. Therefore, the functional unit in this
case can be dened as the amount of packaging needed to
contain a certain amount of liquid under specied condi-
tions and for a specied period of time.
In setting the system boundaries, it is useful to distinguish
between `foreground' and `background' systems (or,
strictly, sub-systems). The foreground system is dened
as the set of processes directly affected by the study
delivering a functional unit specied in Goal and Scope
Denition [25]. The background system is that which
supplies energy and materials to the foreground system,
usually via a homogeneous market so that individual plants
and operations cannot be identied. Differentiation between
foreground and background systems is also important for
deciding on the type of data to be used. The foreground
system should be described by specic process data, while
the background is normally represented by data for a mix or
a set of mixes of different technologies or processes [25,26].
In the second, Inventory Analysis phase, material and
energy balances are performed and the environmental bur-
dens are quantied. The burdens are dened by resource
consumption and emissions to air, water and solid waste.
Aggregation of the burdens into a smaller number of impact
categories (Classication) and evaluation of their potential
impacts (Characterisation) is part of the third, Impact
Assessment, phase (see Fig. 2). A number of methods have
been suggested for the identication and quantication of
environmental impacts [2735]; however, the problem- Fig. 2. Interactions between LCA stages (from Fava et al. [18]).
Fig. 3. Environmental system analysis (adapted from Azapagic and Clift
[24]).
A. Azapagic / Chemical Engineering Journal 73 (1999) 121 3
oriented method, developed by Heijungs et al. [36], is the
most widely used. In this approach, the burdens are aggre-
gated according to the relative contributions to specic
potential environmental effects, such as global warming
potential, acidication, ozone depletion etc. For instance,
CO
2
is a reference gas for determining the global warming
potential of other related gases, such as CH
4
and other
VOCs.
Within the Impact Assessment phase, the impacts can be
aggregated further into a single environmental impact func-
tion by attaching weights to the impacts to indicate their
relative importance. This step, known as Valuation [18] or
Value-choices [22], has proven to be the most controversial
part of LCA because it implies subjective value judgments
in deciding on the importance of different impacts. Valua-
tion is typically not based on natural science [22] but on
expressing preferences either by decision makers, èxperts'
or by the public [3742]. At present, there is no consensus
on how to aggregate the environmental impacts into a single
environmental impact function [22,37] nor even on whether
such aggregation is conceptually and philosophically valid.
The nal phase in the SETAC methodology is Improve-
ment Assessment and is aimed at identifying the possibi-
lities for improving the performance of the system. In the
ISO methodology, this phase is known as Interpretation and,
in addition to improvements and innovations, it covers
identication of major stages in the life cycle contributing
to the impacts, sensitivity analysis and nal recommenda-
tions [43]. In the SETAC methodology, these additional
steps are included within Goal Denition and Scoping and
Inventory Analysis.
While the methodology is being developed, the use of
LCA continues to increase. Some examples of LCA studies
include assessment of the environmental impacts of con-
sumer products; others are aimed at improvements of
environmental performance or development of a new pro-
duct or a process. Prior to focusing on the application of
LCA for process selection and design in Section 4, different
uses of LCA are reviewed and discussed next.
3. Applications of LCA
LCA is generally accepted as an application of system
analysis whose prime objective is to provide a picture of the
interactions of an activity with the environment, thus ser-
ving as a tool for environmental management. As such, LCA
has two main objectives. The rst is to quantify and evaluate
the environmental performance of a product or a process and
so help decision makers choose among alternatives. Another
objective of LCA is to provide a basis for assessing potential
improvements in the environmental performance of the
system. The latter can be of particular importance to engi-
neers and environmental managers, because it can suggest
ways to modify or design a system in order to decrease its
overall environmental impacts.
LCAhas been used for both corporate and public decision
making. Some of the more recent examples of LCA appli-
cations in corporate decision making include energy [44
48] and transport [49,50] sectors, chemical [5155], nuclear
[56,57], metal [5861], polymer [62], paper and forest
[63,64], textile and leather [65,66], water [67,68], electronic
[69,70] and other industries. These applications have mainly
included the following uses, but are not limited to:
v strategic planning or environmental strategy develop-
ment,
v product and process optimisation, design, and innova-
tion,
v identification of environmental improvements opportu-
nities,
v environmental reporting and marketing,
v creating a framework for environmental audits.
Some of the uses of LCA, as identied by product
manufacturers, are shown in Fig. 4. These results are based
on an international survey of a number of organisations
actively involved in LCA [71]. The data indicate that the
most common reasons for performing an LCA are to
improve environmental performance through analysis of
products and to inform short and long-term policy decisions
through system optimisation and design. The use of LCA in
marketing is also frequent. A similar survey [72,73] has
identied that the main barriers to wider use of LCA in
industry are the relatively high costs and the time necessary
for carrying out an LCA coupled with uncertainty about the
potential commercial benets.
LCA has also served as a tool for environmental system
management and environmental reporting. For instance, in
the UK alone there are already over 200 companies which
hold the ISO 14001 certicate [74], which means that LCA
has been or will be used within the EMS to indicate and
track their environmental performance. Unilever, for exam-
ple, has developed an environmental reporting approach,
termed Overall Business Impact Assessment (OBIA) which
estimates the emissions associated with the life cycle of a
Fig. 4. The uses of LCA by industry (from Baumann [71]).
group of products and business areas [75]. ICI has devel-
oped a similar, Environmental Burden approach, using the
impact categories in LCA to assess the effects of its activ-
ities on the environment [76,77].
End-of-life waste management regulations and cross-
sectorial market competition have been the main drivers
for LCA activity in European rms [72]. One such example
of end-of-life management is packaging legislation [78,79]
which requires that certain amount of the packaging must be
recycled. Manufacturers of products that may be affected by
end-of-life management regulations have tried to inuence
the regulatory process and the impacts of these regulations
in the market by using LCA to support their claims. LCA-
based claims have been used in cross-sectorial competition,
both by the commodity and nal product producers. Exam-
ples of these include plastic versus paper packaging [16,80]
and phosphates versus perborates in the detergent industry
[34].
This trend of increased LCA activity in industry has not
be followed by governments, who have been much slower in
adopting LCA as a tool in policy making. Although there
have been a number of attempts to incorporate LCA in
public decision making world-wide, LCA has still not
become an integral part of this process. Many governments
concentrate on reducing pollution to one medium or from a
single life cycle stage. Attempts to integrate life cycle
thinking into policy making include EU eco-labelling
schemes [8183], EC directive on Packaging and Packaging
Waste [78,79], and the IPPC Directive [3]. Other policies
that are starting to use the LCA approach are related to
taxation on pollution. One such example is provided by
France which introduced a tax on CO
2
emissions based on
the results of an LCA study [84]. The Belgian and Norwe-
gian governments are also considering programmes to
introduce taxation on packaging for which information
has been obtained through life cycle studies. In the USA,
the use of LCA in policy making is being encouraged by the
EPA through various projects, such as the establishment of
subsidies or tax credits for alternative fuels and using a life
cycle approach to develop a maximum achievable control
technology (MACT) standard under the 1990 Clean Air Act
[85].
However, although these schemes promote the use of life
cycle approaches in decision making, most of them con-
centrate only on a limited number of burdens and usually in
one or two stages in a life cycle. Nevertheless, this approach
is still preferred to single issue considerations, related
normally to the recyclability of the product or its biode-
gradability. Although these examples represent only the
beginning of application of LCA in policy making, they
are important developments because they demonstrate that
governments are starting to consider broader life cycle
thinking and are prepared to integrate it into the decision
making process.
While a wider acceptance of LCAin policy making is still
to come, there are indications that the use of LCA in
industry is constantly increasing [71,74,76,77,85,86]. The
rest of this paper reviews and discusses some of these
applications, in particular the use of LCA for process
selection, design and optimisation.
4. LCA for process selection, design and optimisation
Historically, most of the LCA literature has been product-
focused [e.g. [18,36,37,42,80,8795]]. However, more
recently several authors have also demonstrated the so far
unexplored potential of LCA as a tool for process selection
[9699], design [100104] and optimisation [20,24,105
111]. These approaches are discussed in the following
section.
4.1. LCA for process selection
In 1997 the U.K. Environment Agency published its
guidelines for assessing the `Best Practicable Environmen-
tal Option' (BPEO) for processes regulated under Integrated
Pollution Control (IPC) [112]. The guidance note sets out a
procedure for assessing environmental harm and comparing
options for specic industrial processes to determine BPEO.
The application of the guidance should enable the environ-
mental consequences (level of harm) of releases to be
properly assessed and ensure that the process or abatement
option chosen represents the `Best Available Technique Not
Entailing Excessive Cost' (BATNEEC) and the site specic
BPEO.
However, the approach proposed by the Agency has
already attracted criticism, mainly because only site-speci-
c considerations are included in choosing the BPEO. As
pointed out by the House of Commons Select Committee on
the Environment in connection with regulation of the
cement industry [113], this narrow approach impeded the
Agency from considering the overall environmental effects
as required under the 1995 Environment Act. This criticism
emphasized once again the fundamental aw in the BPEO
concept, which only considers emissions from the plant
itself, with no account of emissions arising from other
sources in the life cycle, such as production of raw materi-
als, transport and waste disposal. Thus it is possible for a
BPEO technology to reduce a particular pollutant from a
plant, but to increase the emissions of this or other pollutants
elsewhere in the life cycle. This has been demonstrated, for
instance, by the examples of the end-of pipe abatement
techniques for SO
2
[96,98], NO
x
[99], and VOCs [99,114].
The ndings of these studies conrm that if a real BPEO
is to be chosen, it has to be assessed in the LCA context.
The importance of LCA for process selection has also
been recognised by the EC Directive on IPPC [3], due
to be implemented in October 1999 in the member states,
which requires that the Best Available Technique (BAT)
must be chosen by considering the environment as a
whole, including indirect releases, consumption of raw
materials and waste disposal. The use of LCA as a tool for
identifying BPEO and BAT is now illustrated on several
case studies.
4.1.1. BPEO for SO
2
abatement
Vyzi and Azapagic [98] demonstrate, for instance, how
LCA can assist in process selection in the context of BPEO
by comparing the three most common ue gas desulphur-
isation (FGD) processes: wet limestone/gypsum, double
alkali and dry sodium carbonate processes. In the wet
limestone/gypsum process, the sulphur oxides in the ue
gas are absorbed in calcium hydroxide to yield calcium
sulphate which after dewatering can be used in the con-
struction industry as gypsum (see Fig. 5). The double alkali
is similar to the gypsum process, but in addition to calcium
carbonate it also uses sodium carbonate. In the dry system,
the sodium carbonate particles are injected into the ue gas
to react with SO
2
and are then removed, usually in an
electrostatic precipitator. Since the purpose of the study
is to compare different SO
2
abatement techniques, the
functional unit has been dened as `treatment of one tonne
of SO
2
in the ue gas'.
The case study considered in this research is based on the
example of Drax, a U.K. coal-red power station with an
output of 4000 MW. The power station burns approximately
430 kg of coal per second with an average content of 2% of
sulphur which generates 17 kg/s of SO
2
. The wet limestone/
gypsum process, which is fully operational at Drax, is
designed to remove 90% of the SO
2
emissions from the
plant, for which it uses approximately 32.5 kg/s CaCO
3
and
produces 48 kg/s gypsum cake. In this way, 482,400 t/year
of SO
2
are not emitted into the atmosphere, thus avoiding
the associated acidication problem.
However, if this abatement technique is assessed on the
basis of LCA, including the environmental burdens of
extraction of raw materials, production and transport of
CaCO
3
and energy consumption in the process (see Fig. 5),
it turns out that additional 900 t/year of SO
2
are emitted
elsewhere in the life cycle. Furthermore, a number of other
emissions are also generated, of which the most signicant
is 82,000 t/year CO
2
, mainly associated with the life cycle
of CaCO
3
. Thus, in an attempt to reduce the emissions of
SO
2
, the acidication problem has been exchanged for
global warming. This comes as a direct consequence of a
narrow system boundary denition and it clearly demon-
strates the drawbacks of not considering the whole life cycle
of a process in choosing the BPEO.
However, overall, the wet/gypsum process is still pre-
ferred over the other two processes, as it generates the least
environmental impacts (Fig. 6). In addition, it produces
gypsum as a by-product thus avoiding disposal of solid
waste, a problem which remains with the other two pro-
cesses. The dry system is the poorest option in environ-
mental terms since it would generate 2 million tonnes per
year of CO
2
and thus contribute to global warming.
Therefore, it appears that in this case the site-specic and
`cradle to grave' considerations result in the same BPEO,
although no LCAwas attempted at the time the decision on
the best FGD process was being made. One of the motiva-
tions for choosing this process as the BPEO then was the
assumption that the by-product could be used in the con-
struction industry. This idea is in agreement with the con-
cept of ìndustrial ecology', whereby the waste from one
system or life cycle becomes an input material into another
[116]. However, if all coal-red stations in the UK followed
the same assumption and had limestone FGD, then the
Fig. 5. Life cycle of a wet limestone/gypsum
*
process (adapted from de Nevers [115]) (
*
In the double alkali process, sodium carbonate is also added into the
thickener overflow tank).
demand for gypsum would be exceeded and a large amount
of solid waste would have to be disposed off instead.
Furthermore, a particular difculty with the Drax gypsum
is that it contains a relatively large amount of chloride due to
the high chlorine content in the coal, which makes it almost
unusable. Under these circumstances, the choice of the
gypsum process as the BPEO becomes questionable.
This example therefore shows that the choice of BPEO
largely depends on the background economic system within
which the process operates. A further illustration of this is
provided by Golonka and Brennan [96]. They have shown
how the choice of BPEO changes with the way the system
boundary is dened in relation to the background system by
comparing four options for the treatment of SO
2
from
metallurgical smelters. The processes assessed have
included the production of sulphuric acid, elemental sul-
phur, saleable gypsum and disposable calcium sulphite
sludge. Following the LCA philosophy, the authors have
also considered the environmental impacts of the down-
streamprocessing of the by-products of the desulphurisation
processes, i.e. sulphuric acid and elemental sulphur, when
used to manufacture either single superphosphate (SSP) or
diammonium phosphate (DAP).
2
In the analysis without the
downstream processing, the H
2
SO
4
process was found to be
the BPEO in terms of resource depletion and acidication
(Fig. 7). In terms of global warming, the elemental sulphur
process was preferred as it generated 140 kg CO
2 eqv
/t SO
2
(the negative value is due to the heat recovery from the
process). The gypsum and disposable sludge processes were
the worst options as they contributed to global warming in
the order of 1 tonne of CO
2 eqv
per tonne of SO
2
removed.
However, if the background system related to down-
stream processing of elemental S and H
2
SO
4
is included
within the system boundaries, i.e. in effect brought into the
foreground, then H
2
SO
4
SSP and H
2
SO
4
DAP become
worse options for resource depletion, having three orders of
magnitude higher impact than, for instance limestone scrub-
bing. In terms of global warming, H
2
SO
4
DAP is worse
than gypsum and sludge, while the elemental S SSP
process now represents the best environmental option
(see Fig. 7). Similar to the ndings of the FGD case study,
this comparison also demonstrates that the system boundary
is of key importance in choosing BPEO. Furthermore, it also
shows how LCA ensures the link to industrial ecology by
considering entire energy and material supply chain along
the life cycle of a product or process.
4.1.2. BPEO for NO
x
abatement
Other authors have also demonstrated the value of LCAas
a tool for choosing BPEOfor clean-up processes. Yates [99],
for instance, has compared the end-of-pipe technologies for
abatement of NO
x
from metal dissolution processes in the
electronics and nuclear industries. Four technologies, sui-
table for treating mixtures of 5005000 mg/m
3
of NO
2
and
NO in the inlet gas down to 300 mg/m
3
, were neutralisation
with NaOH, extended absorption in water, selective cata-
lytic reduction, and adsorption onto zeolite. Neutralisation
of NO
x
in NaOH in a packed column, in which the NO
x
gas
reacts with NaOH yielding an aqueous sodium nitrite
nitrate efuent, is at present the most widely used technique
for NO
x
abatement. However, although very efcient in
removing NO
x
, this method generates liquid waste and
therefore transfers the burdens from one medium to another.
In the extended absorption process, the efciency of a single
water absorption plant is improved by the addition of a
second tower in a series with the existing column. To
minimise the size of the additional absorption tower, the
inlet gas is pressurised and absorbed in chilled water, to
Fig. 6. Comparison of environmental impacts of the FGD systems (from Vyzi and Azapagic [98]).
2
Both SSP and DAP processes use sulphuric acid as a raw material,
hence elemental sulphur must first be converted to sulphuric acid. DAP
also requires the production of phosphoric acid as an intermediate. These
two processes have been chosen as they represent the most common uses
of H
2
SO
4
worldwide [96].
yield nitric acid which can then be used in the dissolution
process. Selective Catalytic Reduction (SCR) reduces NO
x
gases to nitrogen and water by reaction with ammonia over a
catalyst at a temperature of 2508C. A considerable amount
of heat can be recovered to be reused in the process.
Adsorption onto zeolite is carried out by continuous adsorp-
tion/desorption of the NO
x
on to a packed bed of zeolite.
Regeneration of the zeolite is achieved by heating the bed
with steam to remove rich NO
x
fumes which can be
absorbed in water to yield nitric acid. Although very pro-
mising, this technique is still not available commercially.
The life cycles of these techniques, encompassing all activ-
ities from extraction and manufacture of raw materials to
generation and use of energy, are shown in Fig. 8 [99].
Fig. 7. LCA comparison of options for the treatment of SO
2
from metallurgical smelters (from Golonka and Brennan [96]).
Fig. 8. LCA flow diagrams of the NO
x
abatement techniques (adapted from Yates [99]).
The results of LCA suggest that over the concentration
range of 5005000 mg/m
3
the most favourable NO
x
abate-
ment technology for all burdens is adsorption onto zeolite.
As shown in Fig. 9, this technique has negative emissions
(only CO
2
has been shown since the trends are the same for
other emissions) and wastes due to the recovery of nitric
acid from the desorbed NO
x
, which avoids the need to
manufacture it elsewhere. Only at NO
x
inlet concentrations
below 1000 mg/m
3
do the burdens become positive. How-
ever, this technology is not at present available commer-
cially, so the next best available option is caustic scrubbing.
This technique is, however, let down by the high quantities
of waste sodium nitrate/nitrite generated as a by-product:
1.5 kg/kg NO
x
removed, compared to an average of 0.04 kg/
kg NO
x
for catalytic reduction. This then leaves SCR and
extended absorption which exhibit similar values for the
burdens over the life cycle for NO
x
concentrations above
2000 mg/m
3
. Below that, their environmental impacts
increase signicantly (see Fig. 9). For instance, a treatment
of 500 mg/m
3
NO
x
in the SCR and the extended absorption
process generates 25 and 40 kg of CO
2
per kg of NO
x
removed over the whole life cycle. Thus, the best practic-
able environmental option overall appears to be either SCR
or extended absorption for concentrations above 2000 mg/
m
3
, whilst zeolite adsorption remains the best option over
the whole range of concentrations examined. These ndings
therefore conrm a generally accepted belief that recovery
and reuse of pollutants are more environmentally benecial
than the non-recovery options. However, the choice of the
best technique will depend on the specic characteristics of
the process such as gas owrate and the concentration of the
pollutant in the waste stream.
4.1.3. BPEO for VOC abatement
Yates reached similar conclusions in an LCA study of the
end-of-pipe technologies for VOC removal from a dyestuffs
manufacturing plant [99]. The case study investigated the
effect of ow rate (100020,000 m
3
/h) and concentrations
of mainly xylene (2001200 mg/m
3
) in the waste stream on
the choice of BPEO. Four techniques were examined:
activated carbon adsorption with steam regeneration
(ACA-SR), catalytic oxidation (CO), cryogenic recovery
(CR) and biological oxidation (BO). Since the ACA-SG had
previously been identied as the BPEO without using LCA
principles, the study re-examined this choice and compared
it with the other three techniques.
In the ACA-SR process, steam is used to regenerate the
carbon bed and recover xylene in situ. The desorbed VOC
off-gas stream is condensed to yield a water/organic mixture
which is then phase-split. The catalytic oxidation process
destroys gaseous VOCs through oxidation over a catalyst at
temperatures of 2503508C. Efcient heat recovery from
the oxidation of VOC usually satises the heat requirements
of the system and avoids the need for additional heat supply.
Cryogenic recovery, on the other hand, operates by lowering
the temperature of the gas stream below the VOCs' dew
point. Temperatures aslowas1508Ccanbereachedbyusing
liquid nitrogen. If the VOC condensate is of a sufcient
purity, i.e. contains mainly one component, it may be re-
covered directly for reuse. Finally, the biological treatment
utilises bacteriatooxidise organicmaterial intheliquidphase.
The life cycles of these processes are shown in Fig. 10.
It was found that in terms of CO
2
, NO
x
, and VOC
emissions, ACA-SR was the preferable option over all ow
rates and concentrations investigated; furthermore, these
emissions were negative due to the recovery of xylene
and the associated avoided burdens, which would otherwise
arise from the energy-intensive primary production of this
solvent (Fig. 11). The only burden for which this process
was not the BPEO was SO
2
; however, the difference
between this and the other processes was small and could
be neglected. This again points to the conclusion that
Fig. 9. Life cycle CO
2
emissions and waste for NO
x
abatement technologies (from Yates [99]) (Flow rate: 10000 m
3
/h, outlet NO
x
conc. 300 mg/m
3
;
Adsorption onto zeolite: zeolite loading = 0.14 g/g; Extended absorption: system pressure = 100,000 N/m
2
).
recovery of a pollutant is an environmentally better option
than destroying it.
A counter-example is provided by the cryogenic recovery
process, which was found to be the worst option for most
of the environmental burdens. The main reason for this
is a high energy requirement for the production of liquid
nitrogen, which exceeds the benets of VOC recovery.
However, as the amount of VOC recovered increases, the
Fig. 10. Life cycles of VOC abatement techniques (adapted from Yates [99] and Meier [114]).
Fig. 11. Life cycle CO
2
and VOC emissions for VOC abatement technologies (from Yates [99]) (Flow rate: 5000 m
3
/h, Outlet VOC conc. 50 mg/m
3
;
Catalytic oxidation: T
oxidation
= 3508C; Cryogenic recovery: T
cooling
= 508C).
cryogenic process becomes more competitive and, for inlet
VOC concentrations above 600 mg/m
3
, it becomes the
second most favourable option to ACA-SR in terms of
CO
2
emissions (see Fig. 11). Furthermore, in cases where
the existing nitrogen usage elsewhere on site is sufcient
to meet the VOC recovery demand, the environmental
burdens of CR approximate those of ACA-SR [99]. Thus
depending on the operating conditions of the foreground
and the activities in the background systems, the CR process
ranges from being the worst option to representing the
BPEO.
In a similar LCA study of VOC abatement techniques,
which considered activated carbon adsorption, catalytic
oxidation and biological ltration, Meier [114] shows
how the choice of BPEO changes if the waste gas instead
of one compound contains a mixture of 20 different VOCs.
Since the VOCs are not recovered but incinerated in the
activated carbon process considered in this study, there are
no avoided burdens for the primary manufacture of the
VOCs and additional emissions are generated by the incin-
eration. For instance, the total CO
2
emissions are in this case
equal to 23.9 kg/kg VOCs removed, in comparison to
0.5 kg/kg in the ACA-SR process for the same VOC con-
centration in the ue gas of 200 mg/m
3
. Asimilar increase is
observed for the other burdens so that the activated carbon
process no longer represents the BPEO and the biological
lter becomes the best option. Thus, this is another instance
where industrial ecology, i.e. the possibility of re-using
waste materials, inuences the choice of BPEO. This study
also included economic evaluation of the three options
to identify BATNEEC; biological oxidation was found
to be the best option overall. Since there was little dif-
ference between the activated carbon and catalytic
oxidation processes in terms of environmental performance
and catalytic oxidation had lower economic costs, the
latter was the second preferred option overall after the
biological lter.
4.1.4. BPEO for liquid CO
2
and O
2
production
The application of LCA for process selection is not
limited to end-of-pipe techniques only, as demonstrated
by a number of LCA studies of various industrial processes.
For instance, Rice [97] shows the usefulness of LCA in
process selection on an example of the production of liquid
carbon dioxide. Three sources of CO
2
were compared:
chemical waste gas, a natural deposit and fossil fuel com-
bustion. The waste CO
2
in this example arises from the ICI
ammonia plant and if it was not utilized as a by-product it
would normally be released into the atmosphere. The
second source of CO
2
considered in this work is from a
natural deposit, while the third source is via combustion of
fossil fuel (kerosene). The CO
2
from all the three sources is
cleaned, compressed and liqueed before its downstream
use which, in Rice's case, included aeration of soft drinks.
The results of the analysis showed that in terms of global
warming, the best environmental option is to source raw
CO
2
from chemical waste gas. This reduces the ultimate
volume of CO
2
that would otherwise be released into the
atmosphere and ensures that the utilization efciency asso-
ciated with the system is high. However, as shown in
Fig. 12, in terms of all other impacts, CO
2
from natural
gas is environmentally the best option. Sourcing the liquid
CO
2
from fossil fuel is the worst option overall as it has the
highest environmental impacts.
Rice took this analysis further to show the importance of
the upstream CO
2
process selection for the total environ-
mental impacts when the system boundary is extended to
include the use of CO
2
[97]. If CO
2
is used as an input
material into a system of relatively low energy intensity, the
selection of the process for its generation can play an
important role. For instance, when used for a water soft-
ening treatment, the source of the CO
2
becomes an impor-
tant factor in the overall impact of the treatment system. If,
on the other hand, the liquid CO
2
is an input for a process of
relatively high energy intensity, the source of the CO
2
Fig. 12. Environmental impacts of the processes for generation of liquid CO
2
(from Rice [97]).
becomes much less important. One such example is the use
of CO
2
in the soft drinks industry, where the impacts of CO
2
generation become negligible in comparison to the process
of manufacturing soft drinks. These two examples are
particularly interesting as they demonstrate that if a BPEO
that benets the environment as a whole is to be identied, it
has to be done on a case-specic basis. In policy terms, this
suggests that in choosing BPEO it would seem appropriate
to move away from prescriptive legislation with set limits
and hierarchies to a more exible approach which allows
assessment of different options on a case-by-case basis. This
view has already been adopted by the IPPC Directive, which
recommends that the `èmission limit values,. . . or equiva-
lent technical measures should be based on the best avail-
able technologies [BATs], without prescribing the use of one
specic technique or technology and taking into considera-
tion the technical characteristics of the installation con-
cerned, its geographical location. . .and the environment as a
whole'' [3].
Rice [97] also showed how the choice of BPEO can
dependonsomeof thefactorsidentiedbytheIPPC, including
technical characteristics, geographical location, the scale of
operation etc. He compared processes for production of
oxygen used for aeration of activated sludge in waste water
treatment. The alternative technologies considered were
liquid O
2
from small (140 t/day) and large scale (600 t/day)
cryogenic air separation units (ASU), gaseous O
2
from pres-
sure swing adsorption (PSA), and gaseous O
2
froma vacuum
swing adsorption (VSA) unit. The results showed that per kg
of O
2
at the point of use, the best environmental alternative
is to generate O
2
in a cryogenic unit and deliver it to the point
of use. As economies of scale play a signicant role, it was
found best to source O
2
in a large-scale ASU. The choice of
BPEO is also affected by O
2
demand. If the oxygen require-
ment is between20and70 t/day, ASUremains the best option
for delivery distances smaller than 1000 miles. Above that,
an on-site VSA becomes the best environmental option.
PSA units are only attractive if the demand for O
2
is less
than 20 t/day (i.e. below the minimum VSA production
level) and the delivery distance is greater than 1500 miles.
Furthermore, Rice [97] shows how process efciency can
inuence the choice of BPEO by comparing the life cycles
of air-based mechanical aeration techniques (ne and coarse
bubble-diffused air aeration, jet aeration and surface paddle
aeration) with an oxygen-based technology (Vitox
TM
) used
in the activated sludge process. If the Vitox
TM
technology,
with oxygen sourced from the large-scale ASU, is operated
at the design capacity of 2.5 kg O
2
/kWh of energy needed to
supply O
2
, it is clearly the best environmental option ahead
of the ne bubble, jet aeration, coarse bubble and paddle
aeration options, respectively. However, if the operation of
Vitox
TM
moves away from the design capacity, the asso-
ciated life cycle impacts increase so that at the energy
efciency of 0.9 kg O
2
/kWh, its overall impact becomes
second only to paddle aeration and ne bubble and jet
aeration systems now represent the BPEO.
The above examples demonstrate the usefulness of LCA
for process selection. If applied correctly, this approach can
ensure that the best environmental option is identied
throughout the life cycle. However, in many cases, possi-
bilities for further improving and optimising the perfor-
mance of the selected process will exist. Life cycle thinking
can also be applied to identify optimum options for process
improvements. The application of LCA to system optimisa-
tion is the subject of the next section.
4.2. Process optimisation
Traditionally, system optimisation in chemical and pro-
cess engineering applications has focused on maximising
the economic objectives. Over the past 10 years, considera-
tions for improving the environmental performance have
started to be integrated into system optimisation alongside
economic criteria. These included various waste minimisa-
tion approaches from the concept of mass pinch as a tool to
derive cost-optimal Mass Exchange Networks with mini-
mum emissions waste [117], through minimum waste water
generation in process plants (e.g. [118]) and waste treat-
ments costs [119,120], to the concept of Zero Avoidable
Pollution [121,122]. Although these approaches may have
both environmental and economical benets through
reduced wastes and costs of treatment [123], their disad-
vantage is that they concentrate on emissions from the plant
only, without considering other stages in the life cycle. More
recently, life cycle thinking has started to be incorporated
into the process design and optimisation procedures
[20,24,101111], thus establishing a link between the envir-
onmental impacts, operation and economics of the process
[100]. These developments are still underway and the
published literature on this subject is quite limited.
In general, the approach for incorporating LCA into
system optimisation comprises three main steps
[20,24,106111]:
(i) Carrying out a Life Cycle Assessment study;
(ii) Formulation of the multi-objective optimisation
problem in the LCA context;
(iii) Multi-objective optimisation and choice of the best
compromise solution.
(i) The methodology for the rst step of the procedure has
been briey explained in Section 2; a more detailed account
can be found in Consoli [19], Fava [18] and the ISO 14040
series [17].
(ii) The optimisation problem in the context of LCA is
equivalent to a conventional optimisation model except that
in addition to an economic function it also involves envir-
onmental objectives, represented by the burdens or impacts.
Thus a single objective optimisation problem is transformed
into a multi-objective one. The system is optimised simul-
taneously on both economic and environmental perfor-
mance, subject to certain constraints encompassing all
activities from cradle to grave. This results in an n-dimen-
sional non-inferior or Pareto surface with a number of
optimum solutions for system improvements. By denition,
none of the objective functions on the Pareto surface can be
improved without worsening the value of any other objec-
tive function. Therefore, some trade-offs between objective
functions are necessary in order to reach the preferred
optimum solution in a given situation. Evaluation of
trade-offs and elicitation of preferences to identify the best
compromise solution is part of step (iii), as discussed later in
the paper.
In general, a multi-objective optimisation (MO) problem
of a system formulated in the LCA context can take the
following form:
Minimise f (x; y) = [f
1
f
2
. . . f
p
[
s:t: h(x; y) = 0
g(x; y) _ 0
x X _ R
n
y Y _ Z
q
(1)
where f is a vector of economic and environmental objective
functions; h(x,y) = 0 and g(x,y)_0 are equality and inequal-
ity constraints, and x and y are the vectors of continuous and
integer variables, respectively. Equality constraints may be
dened by energy and material balances; the inequality
constraints may describe material availabilities, heat
requirements, capacities etc. A vector of n continuous
variables may include material and energy ows, pressures,
compositions, sizes of units etc., while a vector of q integer
variables may be represented by alternative materials or
technologies in the system or a number of trucks for
transport of raw materials. If the integer set Z is empty
and the constraints and objective functions are linear, then
(1) becomes a Linear Programming (LP) problem; if the set
of integer variables is nonempty and nonlinear terms exist in
the objective functions and constraints, formulation (1) is a
mixed-integer nonlinear programming (MINLP) problem.
Mixed integer linear programming (MILP) problems incor-
porate integer and linear variables only.
An economic objective typically involves a cost or prot
function as dened by:
Minimise F = c
T
y f (x) (2)
where c is a vector of cost or prot coefcients for integer
variables and f(x) is a linear or non-linear function related to
continuous variables. The environmental objectives in this
context represent the burdens B
j
or impacts E
k
:
Minimise B
j
=
X
N
n=1
b
j;n
x
n
(3)
Minimise E
k
=
X
J
j=1
ec
k;j
B
j
(4)
where bc
j,n
represents emission coefcients associated with
continuous variables x
n
. In Eq. (4), ec
k,j
represents the
relative contribution of burden B
j
to impact E
k
, as dened
by the `problem-oriented' approach to Impact Assessment
[36]. In this approach, for example, Global Warming Poten-
tial (GWP) factors, ec
k,j
, for different greenhouse gases are
expressed relative to the GWP of CO
2
, which is therefore
dened to be unity. If a different Characterisation approach
is used, then Eq. (4) may be redened accordingly. Note that
at present the LCA approach assumes that environmental
burdens and impacts functions are linear, i.e. they are
directly proportional to the output of functional unit(s)
and there are no synergistic or antagonistic effects.
Depending on the characteristics of the system, the
problem (1) can be formulated as (mixed integer) linear
or nonlinear. The theory for solving such problems is well
established (e.g. [124,125]) and a number of commercial
software packages are available to deal with large-scale
problems, of which GAMS [126] is probably the most
widely used in process and chemical engineering applica-
tions. Literature on techniques for solving general single
objective optimisation problems is plentiful (e.g.
[124,125,127]); multi-objective optimisation problems have
also been extensively reviewed (e.g. [128,129]).
(iii) The system is then optimised on a number of
objective functions and optimum solutions are found on
the multi-dimensional non-inferior or Pareto surface. Which
environmental objectives will be chosen for optimisation
depends on the Goal and Scope of the study. Thus, opti-
misation can be performed either at the Inventory or Impact
Assessment levels, in which case the environmental objec-
tives are dened as either burdens or impacts, respectively.
Hence, local and global system improvements are found by
rst moving the system to conditions on the Pareto surface,
and then moving on it. As already mentioned, all objectives
on the surface are optimal in the Pareto sense and some
trade-offs between the objectives are necessary to identify
the best compromise solution. For example, if the system is
optimised simultaneously on two objectives one eco-
nomic and one environmental the resulting Pareto opti-
mum does not necessarily mean that these functions are at
their respective optima achieved when the system is opti-
mised on each of them separately (see Fig. 13). The Pareto
Fig. 13. Pareto curve obtained in multi-objective optimisation (from
Azapagic [20]).
optimum, however, does mean that the set of best possible
options has been identied for a system in which both
objectives should be improved.
Ultimately, environmental and economic objectives
could be aggregated into a single function by attaching
weights to indicate their importance, so that the problem
reduces to a single objective one. However, the main
advantage of MO is that it does not require a priori articu-
lation of preferences, so that the whole non-inferior set of
solutions can be explored. The emphasis then is on the
range of choices from the set of non-inferior solutions,
rather than explicit denition of preferences before analys-
ing all the trade-offs among objectives. The trade-offs
show explicitly what can be gained and what lost by
choosing each alternative. Where there are multiple deci-
sion makers with conicting interests, this technique can
help to resolve disputes by generating different alternative
solutions. Decision makers who understand the trade-offs
and the alternatives are more likely to understand the
interests of other parties and, therefore, to compromise.
Although the evaluation of trade-offs among the objectives
to choose the best compromise solution will still imply
certain preferences and value judgments, at least the choice
will be made from all possible non-inferior solutions, unlike
other methods where the bulk of non-inferior solutions may
be ignored. This is particularly relevant in the LCA context,
because it enables avoiding the controversial and debatable
concept of aggregation of environmental impacts into a
single environmental impact function in the Valuation
stage. Furthermore, by being able to trade-off incommen-
surable objectives, e.g. environmental impacts and eco-
nomic requirements, this approach avoids the well-known
problems encountered, for instance, in costbenet analysis
[130133], i.e. reducing individual preferences to a market
value or trying to express the quality of the environment
in nancial terms.
One of the possible ways to choose the `best' solution
is to consider a graphical representation of the non-inferior
set and then choose the best compromise solution on the
basis of the trade-offs. However, this approach is limited to
two or three objective functions at most, because graphical
representation becomes less than helpful with more than
three objectives. Another way to look at it is to express the
values of objectives at non-inferior solutions in terms of the
percentage distance from their individual optima. If all
objectives are considered to be of the same importance,
then the best compromise solution could be that which
equalizes the percentage by which all objectives differ from
their optimumvalues. However, should any of the objectives
be considered more important than the others, then other
techniques that allowordering of preferences, such as Multi-
attribute Utility Theory [134] or Analytic Hierarchy Process
[135], could be used to identify the best compromise solu-
tion. Although this implies eliciting preferences for the
objectives, these preferences are at least articulated in the
post-optimal analysis of all non-inferior solutions and their
trade-offs, as distinct from expressing preferences and
aggregating the objectives prior to identifying all non-
inferior solutions.
An approach similar to this was used by Kniel et al. [100]
for the optimisation of a nitric acid plant. The plant produces
roughly 30 t of 56% nitric acid per hour and generates
107 kg/h nitrogen oxides in the waste gas. To improve
the environmental performance of the plant, two process
alternatives to the basic design were considered: addition of
a SCRunit and increasing the operating pressure in the nitric
acid absorber. LCA identied the latter alternative as envir-
onmentally superior to both the existing operations and the
installation of a SCR unit. Economic analysis revealed that
the rate of return was comparable to that of the existing
operations and about 28% higher than that of the SCR. To
illustrate the approach, the system was optimised on two
objectives: economic returns and environmental index func-
tion. The latter represented a linear combination of impacts,
aggregated by using the marginal values which relate
changes in the environmental impacts to process variables
and are obtained at the solution of the optimisation problem.
As discussed above, one of the drawbacks of the a priori
aggregation of objectives is that a number of non-inferior
solutions could be lost before considering the trade-offs.
Furthermore, as acknowledged by the authors, the aggrega-
tion method used in this particular case is based on the way
the system is operated and fails to include the sociological
aspects into valuation.
As illustrated in Fig. 14, the Pareto curve shows the trade-
offs between the environmental index and the economic
returns, depending on pressure in the system. The environ-
mental index reaches a minimum when pressure is max-
imum because recovery of NO
x
is maximised. The
maximum economic return occurs at an intermediate pres-
sure, where recovery of NO
x
is relatively high, but com-
pression costs are relatively low. It is also interesting to note
that in this case the curve attens in its upper regions,
suggesting that a slight increase in operating pressure would
cause little economic sacrice but substantial environmental
benet. As discussed above, this quantication of losses and
gains provides a useful tool to the decision makers in
choosing the best compromise solution.
Fig. 14. Non-inferior curve for optimisation of a nitric plant (from Kniel
et al. [100]).
Azapagic [20] and Azapagic and Clift [136,137] have
taken this approach further by optimising on a number of
environmental and economic objectives, thus avoiding
Valuation. This is illustrated on a real mining and processing
example of the system producing ve boron products:
10 mol borate (Na
2
B
4
O
7
10H
2
O), 5 mol borate (Na
2
B
4
O
7
4.67H
2
O), boric acid (H
3
BO
3
), anhydrous borax (Na
2
B
4
O
7
),
and anhydrous boric acid (B
2
O
3
). A simplied LCA ow
diagram of the system, showing the distinction between the
foreground and background subsystems, is shown in
Fig. 15. After extraction and crushing, the boron minerals,
borax and kernite, are transported to the plant and dissolved
in water to produce 5 and 10 mol borates. Boric acid (BA) is
produced in a separate plant, by reacting kernite ore with
sulphuric acid. Anhydrous borax (AB) and anhydrous boric
acid (ABA) are made in high-temperature furnaces from
5 mol borate and BA, respectively. All products are then
either packed or shipped in bulk. Electric energy and steam
for the system are provided by the on-site natural gas
cogeneration facility with additional steam generated in
the steam plant. All activities from the extraction of raw
materials to the production of the boron products and
materials used are included in the system; however, the
use and disposal phases of the products are not considered in
this study (`cradle-to-gate' approach). The functional unit
was dened as total yearly production of the boron products
of 1,062,000 t/year.
The objective of this work was to optimise the system on
environmental and economic performance to identify a
range of possibilities for minimising total environmental
impacts from the system, while maximising production
subject to total product demand and keeping the production
costs to their minimum. The optimisation model also
included several process alternatives to identify the BPEO.
These included different product dryers (uid, tray and
rotary), conveyors instead of trucks in the mine, and the
generation of additional steam in the Cogeneration instead
of in the Steam plant. To demonstrate the approach, the
system model dened by Eqs. (1)(4), was optimised on
three objective functions: global warming potential (GWP),
total production (P) and production costs (C). The three-
dimensional non-inferior surface ABCD, generated in a
series of multi-objective optimisations, is shown in
Fig. 16. Depending on the position on the non-inferior
surface, the optimum solutions offer different options for
improvements and BPEOs. For instance, Point A represents
the minimum of the cost-objective function; however, the
Fig. 15. LCA flow diagram of the boron system (from Azapagic and Clift
[110]).
Fig. 16. Non-inferior surface for the boron system (from Azapagic [20]) (C
*
, GWP
*
and P
*
optimum values obtained in single-objective optimisation; C,
GWP, P optimum values obtained in multi-objective optimisation).
production is at the minimum and GWP is 31% above its
optimumvalue. The BPEOat this point includes transport in
the mine by trucks and steam production in the Steam plant.
The KuhnTucker multipliers
3
at this solution suggest that if
a further decrease in the GWP objective by one tonne is
required, a cost increase of 95 would be incurred for
constant product output; similarly, if the production were
to increase by one tonne, the resulting increase in the costs
would be equal to 34 for constant GWP. The KuhnTucker
multipliers therefore determine the marginal costs of redu-
cing GWP and increasing production, respectively.
By moving from Point A along the non-inferior curve for
constant GWP, both costs and production increase, to reach
their maximum feasible values at point B. The Cost function
here is 4% above its optimum value. If Production is
increased by one tonne, 300 of the Costs objective have
to be given up. Similarly, one tonne change in GWP is
associated with a cost change of 100. At this solution, the
BPEO is dened by steam generation in the Steam plant and
the preferred transportation means in the mine is by con-
veyors.
If, however, the system were to be operated at Point C,
GWP would be 3.3% above its optimum value obtained in
single objective optimisation. Production would be at the
minimum, and the costs would increase by 14%. An
improvement in GWP of one tonne would worsen the values
of the Costs objective by 36, while a tonne increase in P
would result in 100 increase in the costs. At this point, 93%
of the steam is generated by the Cogeneration plant and the
rest is produced in the Steam plant. The conveyors still
remain the best transport option in the mine.
Furthermore, if, for example, Point D were to be chosen
as the best compromise solution, then for the same value of
GWP as at Point C, production would reach the maximum;
however, costs would have to increase by 17%. It can be
noted here that both GWP and Production exhibit similar
effect on Costs: a decrease in GWP by one tonne increases
Costs by 3,600, while increasing production by one tonne,
increases the costs by 3,500. At this solution, the best
practicable environmental option is dened by truck trans-
port in the mine and steam production in the Cogeneration
plant.
These results demonstrate how optimum solutions and
therefore BPEO change with the operating state of the
system. The same analysis can be done for other points
on the non-inferior surface which are all optimal in the
Pareto sense. By trading-off the values of different objec-
tives at these points, decision makers can select any solution
on the surface, depending on how much of one objective
they are prepared to give up in order to gain in another. The
value of multi-objective optimisation in the context of LCA,
therefore, lies in offering a range of choices for environ-
mental and economic improvements of the system and so
enabling preferences to be identied after analysing all the
trade-offs among objectives.
Although the discussion in this section has mainly
focused on optimisation of the existing processes, a similar
approach can be applied for the design of new processes.
The use of LCA and multi-objective optimisation in the
process design is reviewed in the next section.
4.3. LCA for process design
One of the newly emerging applications of LCA is in
product and process design. This has resulted in the devel-
opment of a new LCA-related tool Life Cycle Product/
Process Design (LCPD). The LCPD methodology is still in
its infancy and published literature is scant (e.g.
[92,94,138]).
Although the methodologies for life cycle product and
process design are similar, the following discussion will
focus on Life Cycle Process Design. As outlined in Fig. 17,
environmental considerations are incorporated at an early
stage of the design, alongside the more traditional technical
and economic criteria [139]. LCA is used throughout the
design process, initially on a reference process. The con-
ventional system boundary is extended to include the life
cycles of different technologies and raw materials, all the
way from extraction of primary resources through to pro-
duction. This enables a quantitative comparison of different
technological routes for production of the same set of raw
materials as well as an assessment of different raw materi-
als. It may be noted that the same rigorous owsheeting
procedures used in conventional design can directly be
linked with the environmental analysis described here.
Furthermore, as Product Stewardship initiatives are seeking
to build alliances between manufacturers and suppliers
3
Kuhn-Tucker multipliers indicate how much of one objective function
has to be given up in order to improve the value of the other; for more
detail see e.g. [128].
Fig. 17. General methodological framework for Life Cycle Process
Design (from Azapagic [139]).
[140], it is also desirable to include the supply chain within
the design framework. This enables identication of best
suppliers, in terms of their product quality, environmental,
health, safety and other performance criteria. For instance,
one of the guiding principles in choosing the supplier could
be their compliance with ISO 9000 or 14000 or perhaps an
international environmental label of their product. More-
over, the design process must include compliance with
relevant legislation, such as health and safety regulations
and environmental emissions limits. Finally, customer and
consumer requirements in terms of the specications and
performance of the product can also be addressed during
process design. Once all of the criteria of interest have been
identied, a multi-objective optimisation model of the
general form Eqs. (1)(4) can be dened. The system is
then optimised on a number of objectives, dened by
environmental burdens or impacts and socio-economic
functions, subject to the constraints on material and energy
balances, productive capacities, technical, legislative and
other requirements. As a result, a plethora of non-inferior
solutions is obtained, enabling a quantitative evaluation of
options for environmental, technical, economic and other
improvements of the process system. This whole procedure
is dynamic with a continuous exchange of information
within and outside the design team to explore systematically
the possibilities for improvements.
Therefore, LCPD offers a potential for technological
innovation in the process concept and structure through
the selection of the best technologies and raw materials over
the whole cycle. As already discussed, this can be of
particular importance if placed within the context of EMAS
and ISO 14000 EMS, as well as the IPPC Directive, which
require companies to have a full knowledge of the environ-
mental consequences of their actions, both on- and off-site.
Furthermore, as the `polluter pays' and `producer respon-
sibility' initiatives are starting to force manufacturers to
reduce waste at source and manage the waste associated
with their processes, LCPD can provide a powerful frame-
work for the design of cleaner processes which are envir-
onmentally benign and economically protable.
Similar approaches have been proposed by Kniel et al.
[100], Pistikopoulos et al. [102], Stewart and Petrie [104],
and others [103,105]. Pistikopoulos et al. [102,103,105], for
instance, use a so-called Minimum Environmental Impact
(MEI) methodology which embeds LCA principles within a
formal process optimisation framework. This approach
extends the existing waste minimisation design techniques
by providing a more complete description of the environ-
mental impacts of the process. The authors [102,103,105]
use an example of the design of a vinyl chloride monomer
(VCM) plant to show that the `zero emission' target may not
always be the best environmental policy and that an optimal
degree of abatement may be preferable instead. This is
illustrated in Fig. 18 which reveals that there is a minimum
mass of dichloroethane (DCE) discharged beyond which the
global environmental impact increases due to the trade-offs
in waste generation over the whole life cycle between
system inputs and outputs [102]. This example indicates
that from an environmental point of view, minimisation of
the output emissions only, as normally carried out in con-
ventional system optimisation, can in fact lead to sub-
optimal solutions. Furthermore, it also emphasizes the
importance of multi-objective optimisation in which all
functions have to be considered simultaneously if global
environmental improvements are to be achieved. Targeting
for minimum `global' impacts may also result in less
expensive plant operation, as was the case in this study.
5. Conclusions
There is a growing need to move away from narrow
denitions and concepts in environmental system manage-
ment. Life Cycle Assessment offers a systematic way to
incorporate the entire material and energy supply chain into
strategic planning and policy development. This is demon-
strated by an increasing number of applications of LCA,
both by industry and governments. Some of the newly
emerging applications of LCA reviewed in this paper are
in process selection, design and optimisation.
The application of LCA to process selection shows that
the Best Practicable Option (BPEO) that benets the envir-
onment as a whole must be identied and chosen in the LCA
context. The case studies conrm that the choice of BPEO is
dependent on the boundaries, operating state of the system
and on the background economic system in which it oper-
ates. Some of the examples indicate that recycling and
recovery of pollutants may not always be the best option
environmentally and that the BPEO has to be determined on
a case-by-case basis.
LCA can also be coupled with multi-objective optimisa-
tion techniques to provide a powerful tool for process design
and optimisation. A newly emerging Life Cycle Process
Design (LCPD) tool offers a potential for technological
Fig. 18. Effect of the degree of abatement on water discharges from the
VCM plant (from Pistikopoulos et al. [102]).
innovation in process concept and structure through selec-
tion of best material and process alternatives over the
whole life cycle. This approach provides a robust frame-
work for process design by simultaneously optimising on
environmental, technical, economic and other criteria.
Multi-objective optimisation in this context serves to
identify a number of Pareto-optimum options for improved
design and operation throughout the whole life cycle. This
approach provides a potentially powerful decision making
tool which may help process industries identify sustainable
options for the future.
Acknowledgements
I wish to express my thanks to Prof. Roland Clift for
introducing me to LCA some years ago and for his invalu-
able contributions to some of the works quoted here. I am
also grateful for his and Prof. John Smith's suggestions for
improvements to this paper.
References
[1] ISO 14001, Environmental Management Systems Specification
with Guidance for Use, HMSO, London, 1996.
[2] Council Regulation (EEC) No. 1836/93 of 29 June 1993 Allowing
Voluntary Participation by companies in the Industrial Sector in a
Community Eco-Management and Audit Scheme 1993. Official
Journal of the European Communities, No. L 168, 10 July, HMSO,
1993, London, pp. 118.
[3] Council Directive 91/61/EC 1996. Concerning Integrated Pollution
Prevention and Control, Official Journal of the European Commu-
nities, No. L 257, 10 October, HMSO, London, 1996.
[4] Department of the Environment, Transport and the Regions, U.K.
Implementation of EC Directive 96/61 on Integrated Pollution
Prevention and Control, Second Consultation Paper, DETR,
London, 1998.
[5] R. Clift, A.J. Longley, Introduction to clean technology, in: R.C.
Kirkwood, A.J. Longley (Eds.), Clean Technology and the
Environment, Chap. 6, Blackie Academic and Professional,
Glasgow, 1994.
[6] R. Clift, Clean technology The idea and practice, J. Chem. Tech.
Biotech. 68 (1997) 347350.
[7] G. Allen, R. Clift, J.F. Davidson (Eds.), Clean Technology:
The Idea and the Practice. Philosophical Transactions: Mathema-
tical, Physical and Engineering, 355, Royal Society, 1997,
p. 1728.
[8] R. Clift, Engineering for the environment: The new model engineer
and her role, Process Safety and Environ. Protection 76(B2) (1998)
151160.
[9] I. Boustead, The Milk Bottle, Open University Press, Milton
Keynes, 1972.
[10] B. Hannon, System Energy and Recycling: A Study of the
Beverage Industry, Center for Advanced Computation, University
of Illinois, Urbana, IL, 1972.
[11] G. Sundstrom, Investigation of the energy requirements from raw
materials to garbage treatment for 4 Swedish Beer Packaging
Alternatives, Report for Rigello Park AB, Sweden, 1973.
[12] R.G. Hunt, W.E. Franklin, Resources and Environmental Profile
Analysis of 9 Beverage container Alternatives. Contract (68-01-
1848), U.S. EPA, Washington, DC, 1974.
[13] E. Barber, K. Fanganas, F. Iannazzi, R. Shamel, K. Halloack, H.
Frost, Selected Characteristics of Disposable and Reusable
Napkins, (PB-275 222/8), Arthur D. Little, Cambridge MA, 1977.
[14] R.U. Ayres, Process classification for the industrial material sector,
Technical Report, United Nations Statistical Office, New York,
1978.
[15] M.P. Lundolm, G. Sundstrom, Resource and Environmental Impact
of Tetra Brik Carton and Refillable and Non-Refillabel Glass
Bottles, AB Tetra Pak, Malmo, 1985.
[16] I. Boustead, Environmental Impact of the Major Beverage
Packaging Systems U.K. Data 1986 in Response to the EEC
Directive 85/339, INCPEN, London, 1989, pp. 14.
[17] ISO/DIS 14040, Environmental Management Life Cycle Assess-
ment Part 1: Principles and Framework, 1997.
[18] J. Fava, R. Dennison, B. Jones, M.A. Curran, B. Vigon, S. Selke, J.
Barnum (Eds.), A Technical Framework for Life-Cycle Assess-
ment, SETAC and SETAC Foundation for Environmental Educa-
tion, Washington, DC, 1991.
[19] F. Consoli, D. Allen, I. Boustead, J. Fava, W. Franklin, A.A. Jensen,
N. de Oude, R. Parrish, R. Perriman, D. Postlethwaite, B. Quay, J.
Seguin, B. Vigon (eds.), Guidelines for Life-Cycle Assessment: A
`Code of Practice', SETAC, Brussels, 1993.
[20] A. Azapagic, Environmental system analysis: The application of
linear programming to Life Cycle Assessment, Ph.D dissertation,
University of Surrey, 1996.
[21] ISO/TC 207/SC 5/WG 2 N 112, ISO/DIS 14041: Environmental
Management Life Cycle Assessment Part 2: Goal and Scope
Definition and Life Cycle Inventory Analysis, Voting draft, 1997.
[22] ISO/TC 207/SC 5/ N 112, ISO/DIS 14042.3: Environmental
Management Life Cycle Assessment Part 3: Life Cycle Impact
Assessment, Committee draft, 1998.
[23] ISO/TC 207/SC 5/ N 112, ISO/DIS 14043: Environmental
Management Life Cycle Assessment Part 4: Life Cycle
Interpretation, Voting draft, 1998.
[24] A. Azapagic, R. Clift, Allocation of environmental burdens by
whole-system modelling The use of linear programming. In: G.
Huppes, F. Schneider, (Eds.), Allocation in LCA, SETAC, Brussels,
1994, pp. 5460.
[25] R. Clift, R. Frischknecht, G. Huppes, A.-M. Tillman, B. Weidema,
(Eds.), Toward a coherent approach to Life Cycle Inventory
Analysis, Report of the Working Group on Inventory Enhancement,
SETAC-Europe, Brussels, 1998.
[26] H.A. Udo de Haes, J.-F. Bensahel, R. Clift, C.R. Fussler, R.
Griesshammer, A.A. Jensen, Guidelines for the application of Life
Cycle Assessment in the EU ecolabelling programme, European
Commission, DG XI-A-2, Brussels, 1994.
[27] H. Hallay, R. Pfriem, O
ko-Controlling Umweltschutz in mittel-

standischen Unternehmen, Campus, Frankfurt, 1992.
[28] S.H. Kalisvaart, J.A.M. Remmerswaal, The Met-points Method: A
new single figure environmental performance indicator, in: H.A.,
Udo de Haes, A.A. Jensen, W. Klopffer, L.-G. Lindfords, (Eds.),
Integrating Impact Assessment into LCA, SETAC-Europe, Brus-
sels, 1994.
[29] O. Jolliet, Impact Assessment of Human and Eco-toxicity in Life
Cycle Assessment. In: H.A. Udo de Haes, (Ed.), Towards a
Methodology for Life Cycle Impact Assessment, SETAC-Europe,
Brussels, 1996.
[30] K. Habersatter, O
ekobilanz von Packstoffen Stand 1990; Schrif-

tenreiche Umwelt 132, Bundesamt fur Umwelt, Wald und Land-
schaft, Bern, 1991.
[31] M. Goedkoop, The Eco-indicator 95; vol. 9524 (Manual for
Designers) and 9523 (Final report), Netherlands Agency of Energy
and Environment (NoVEM), Amersfoort, 1995.
[32] S. Ahbe, A. Braunschweig, R. Muller-Wenk, Methodik fur
O
kobilanzen auf der Basis okologischer Optimierung. Schriften-

reihe Umwelt Nr. 133, Bundesamt fur Umwelt, Wald und
Landschaft (BUWAL), Oktober, Bern, 1990.
[33] B. Steen, EPS-Default Valuation of Environmental Impacts from
Emission and Use of Resources Version 1996. Annual Report,
AFR Report 111, Swedish Environmental Research Institute IVL,
Stockholm, April, 1996.
[34] B. Wilson, B. Jones, The Phosphate Report: A Life Cycle study to
evaluate the environmental impact of phosphate and zeolite A-PCA
as alternative builders in U.K. laundry detergent formulations,
Landbank Environmental Research and Consulting, January,
London, 1994.
[35] F. Schmidt-Bleek, Wieviel Umwelt braucht der Mensch? MIPS-
Das Mass fur okologisches Wirtschaften. Birkhauser Publikationen,
Berlin, 1993.
[36] R. Heijungs, et al. (Eds.), Environmental Life Cycle Assessment of
Products: Background and Guide, MultiCopy, Leiden, 1992.
[37] J. Fava, F. Consoli, R. Dennison, K. Dickson, T. Mohin, B. Vigon,
(Eds.), A Conceptual Framework for Life-Cycle Impact Assess-
ment. SETAC and SETAC Foundation for Environmental Educa-
tion, Pensacola, 1993.
[38] K.P. Yoon, L. Ching, Multiple Attribute Decision Making
Introduction, Sage, London, 1995.
[39] R.L. Keeney, H. Raiffa, Decisions with Multiple Objectives:
Preferences and Value Trade-offs, Wiley, New York, 1976, pp. 569.
[40] D.W. Pearce, K. Turner, Benefits Estimates and Environmental
Decision-Making, OECD, Paris, 1992.
[41] P. Miettinen, R.P. Hamalainen, How to benefit from decision
analysis in environmental life cycle assessment (LCA), Eur. J. Op.
Res. 102(2) (1997) 279294.
[42] J.B. Guinee, R. Heijungs, H.A. Udo de Haes, G. Huppes,
Quantitative life cycle assessment of products. 2: Classification,
valuation and improvement analysis, J. Cleaner Production 1(2)
(1993) 8191.
[43] K. Saur, Life cycle interpretation A brand new perspective? Int. J.
LCA 2(1), (1997), 810 .
[44] K. Tahara, T. Kojima, A. Inaba, Evaluation of CO
2
payback time of
power plants by LCA, Energy Conversion and Management,
38(SS), (1997), S615S620.
[45] S. Kato, N. Nomura, Hydrogen gas-turbine characteristics and
hydrogen energy system schemes, Ener. Conv. Manage. 38(1013),
(1997), 13191326 .
[46] R. Matsuhashi, K. Hikita, H. Ishitani, Model analyses for
sustainable energy supply taking resource and environmental
constraints into consideration energy conversion and management
37(68), (1996), 12531258.
[47] H. Audus, IEA greenhouse gas R&D programme: Full fuel cycle
studies, Ener. Conv. Manage. 37(68), (1996), 837842, .
[48] R. Dones, R. Frischknecht, Life-cycle assessment of photovoltaic
systems: Results of swiss studies on energy chains, Progr. in
Photovoltaics 6(2) (1998) 117125.
[49] E. Eriksson, M. Blinge, G. Lovgren, Life Cycle Assessment of the
road transport sector, Sci. Total Environ. 190 (1996) 6976.
[50] N.L.C. Steele, D.T. Allen, Life-cycle assessment An abridged
life-cycle assessment of electricvehicle batteries, Environ. Sci.
Technol. 32(1) (1998) A40A46.
[51] R. Bretz, P. Fankhauser, Life-cycle assessment of chemical
production processes: A tool for ecological optimization, CHIMIA
51(5) (1997) 213217.
[52] M. Aresta, I. Tommasi, Carbon dioxide utilisation in the chemical
industry, Ener. Conv. Manage. 38(SS) (1997) S373S378.
[53] E. Ophus, V. Digernes, Life-Cycle Assessment of an alkyd
emulsion: Improvement in environmental performance, Jocca
Surface Coatings International 79(4) (1996) 156.
[54] I.D. Dobson, Life Cycle Assessment for painting processes: Putting
the VOC issue in perspective, Progress in Organic Coatings, 27(1
4) 5558 (1996) .
[55] M. Franke, H. Kluppel, K. Kirchert, P. Olschewski, Life-cycle
assessment Life-cycle inventory for detergent manufacturing,
Tenside Surfactants Detergents 32(6) (1995) 508514.
[56] B. Solberg-Johansen, Environmental LCA of the nuclear fuel cycle,
Ph.D dissertation, vols. 1 and 2, University of Surrey, 1998.
[57] N.L. Griffin, A clean technology approach for nuclear fuel
reprocessing, J. Chem. Technol. Biotechnol. 68(4) (1997) 361366.
[58] J.G.S. Robertson, J.R. Wood, B. Ralph, R. Fenn, Analysis of lead/
acid battery life cycle factors: Their impact on society and the lead
industry, J. Power Sour. 67(12), 225236 (1997).
[59] K. Shibata, Y. Waseda, Newmodel for assessment of metal production
and recycling systems, J. Jpn. Inst. Metals 61(6) (1997) 494501.
[60] M. Finkbeiner, E. Hoffmann, G. Kreisel, The functional unit in the
life cycle inventory analysis of degreasing processes in the metal-
processing industry, Environ. Manage. 21(4) (1997) 635642.
[61] S.T. Chubbs, B.A. Steiner, Life cycle assessment in the steel
industry, Environ. Progr. 17(2) (1998) 922995.
[62] N. Yoda, Life Cycle Assessment in polymer industry toward the
21st century, J. Macromol. Sci. Pure and Appl. Chem. A33(12)
(1996) 18071824.
[63] J. Seppala, M. Melanen, T. Jouttijarvi, L. Kauppi, N. Leikola,
Forest industry and the environment: A life cycle assessment study
from finland, Resour. Conser. Recycling, 23(12), 87105, (1998).
[64] B. Backlund, Scandinavian collaboration develops Life Cycle
Assessment (LCA) as a tool for the forest industry, (in Swedish),
Svensk PapperstidningNordisk Cellulosa 101(3) (1998) 4950.
[65] T.L. Kuusinen, R.H. Barker, D.A. Alexander, Life cycle assessment
in woven textiles, Tappi J. 81(3) (1998) 179182.
[66] A.G. Puntener, Risk assessment of leather dyestuffs, J. Soc. Leather
Technol. Chem. 82(1) (1998) 14.
[67] P.J. Roeleveld, A. Klapwijk, P.G. Eggels, W.H. Rulkens, W. van
Starkenburg, Sustainability of municipal wastewater treatment,
Wat. Sci. Technol. 35(10) (1997) 221228.
[68] F.J. Dennison, A. Azapagic, R. Clift, J.S. Colbourne, Assessing
management options for wastewater treatment works in the context
of Life Cycle Assessment, Wat. Sci. Technol., (1998), in press.
[69] S. Miyamoto, M. Tekawa, Development of life cycle assessment
software and application to personal computer assessment, Nec
Res. Dev. 39(2) (1998) 7781.
[70] P. de Langhe, S. Criel, D. Ceuterick, Green design of telecom
products: The ADSL high speed modem as a case study. IEEE
Transactions on Components Packaging and Manufacturing
Technology Part A. 21(1), (1998), 154167.
[71] H. Baumann, LCA use in swedish industry, Int. J. LCA. 1(3) (1996)
122126.
[72] F. Berkhout, Life Cycle Assessment and industrial innovation.
Global Environmental Change Programme briefings, No. 14, June,
EPSRC, Programme Office, 1997.
[73] F. Berkhout, R. Howes, The adoption of life-cycle approaches by
industry: Patterns and impacts, Resour. Conser, Recycling 20(2)
(1997) 7194.
[74] Environmental Data Services Ltd., ENDS Report 267, April, 1997.
[75] A.P. Taylor, D. Postlethwaite, Overall Business Impact Assessment
(OBIA), 4th LCA Case Studies Symposium, SETAC-Europe,
Brussels, 1996, pp. 181187.
[76] M.D. Wright, D. Allen, R. Clift, H. Sas, Measuring corporate
environmental performance: The ICI environmental burden system,
J. Industrial Ecol. 1(4) (1997) 117127.
[77] R. Clift, Relationship between environmental impacts and added
value along the supply chain, 2nd Int. Conf. Technology Policy and
Innovation, 35 August 1998, Lisbon.
[78] Official Journal of the European Communities No. L 365/10, 1994.
[79] HMSO, Producer Responsibility Obligations (packaging waste)
Regulations, HMSO, London, 1997.
[80] I. Boustead, Eco-balance Methodology for Commodity Thermo-
plastics, PWMI, Brussels, 1992.
[81] Official Journal of the European Communities No. L 99/1, 1992.
[82] H.A. Udo de Haes, R. Clift, L. Griesshammer, A.A. Jensen,
Practical Guidelines for life Cycle Assessment for the EU
Ecolabelling Programme, Final report of third phase, 1996..
[83] R. Clift, Life Cycle Assessment and Ecolabelling, J. Cleaner Prod.
1(34), 155-159, 1993.
[84] OECD, Life Cycle Summaries of OECD Countries, January, 1995.
[85] M.A. Curran, life-cycle based government policies, Int. J. LCA
2(1) (1997) 3943.
[86] J.J. Lee, P. O'Callaghan, D. Allen, Critical review of life cycle
analysis and assessment techniques and their application to
commercial activities, Resour., Conser. Recycling 13 (1995) 3756.
[87] B.P. Weidema, I. Kruger (Eds.), Environmental Assessment of
Products: ATextbook on Life Cycle Assessment, 2nd edn., UETP-
EEE, Finnish Association of Graduate Engineers, Helsinki, 1993.
[88] B. Pedersen, K. Christiansen, A meta-review on product Life Cycle
Assessment, in: Product Life Cycle Assessment Principles and
Methodology. Series: Nordic Council of Ministers, Copenhagen,
Nord, p. 9 , 1992.
[89] B. Pedersen (Ed.), Environmental Assessment of Products. ACourse
on Product Life Cycle Assessement, UETP-EEE, Helsinki, 1993.
[90] A-M. Tillman, H. Baumann, E. Eriksson, T. Rydberg, . Life-Cycle
Analysis of Packaging Materials. Calculation of Environmental
Load, Chalmers Industri Teknik, Gothemburg, 1991.
[91] B.W. Vigon, D.A. Tolle, B.W. Cornaby, H.C. Latham, C.L.
Harrison, T.L. Boguski, R.G. Hunt, J.D. Sellers (Eds.), Life Cycle
Assessment: Inventory Guidelines and Principles, U.S. EPA,
Washington DC, 1993.
[92] A. Azapagic, Life Cycle Assessment: A tool for innovation and
improved environmental performance, 1st Int. Conf. Technology
Policy and Innovation, 24 July, Macau, 1997.
[93] G. Fleischer, W.-P. Schmidt, Iterative screening LCA in an
ecodesign tool, Int. J. LCA 2(1) (1997) 2024.
[94] RMIT A Guide to EcoReDesign, Royal Melbourne Institute of
Technology, Melbourne, Australia, 1996.
[95] G.A. Kekoleian, The application of Life Cycle Assessment to
design, Cleaner Prod. 1(34), (1993), 143-150,.
[96] K.A. Golonka, D.J. Brennan, Application of Life Cycle Assessment
to process selection for pollutant treatment: A case study of sulphur
dioxide emissions from Australian metallurgical smelters, Trans.
IChemE, vol. 74, Part B, May, (1996), pp. 105119.
[97] G. Rice, The Application of Life CycleAssessment to industrial
process selection, EngD Portfolio, University of Surrey, 1997.
[98] E. Vyzi, A. Azapagic, Life Cycle Assessment as a tool for
identifying the Best Practicable Environmental Option (BPEO),
Research report, University of Surrey, 1998.
[99] A. Yates, LCA: Clean-up technologies and abatement of gaseous
pollutant emissions from chemical processing plant, EngD
Portfolio, University of Surrey, 1998.
[100] G.E. Kniel, K. Delmarco, J.G. Petrie, Life Cycle Assessment
applied to process design: Environmental and economic analysis
and optimisation of a nitric acid plant, Environ. Progr. 15(4) (1996)
221228.
[101] C. Pesso, Life cycle methods and applications: Issues and
perspectives, J. Cleaner Prod. 1, 34, (1993), 139142.
[102] E.N. Pistikopoulos, S.K. Stefanis, A.G. Livingston, A methodology
for minimum environmental impact analysis, in: M.M. El-Halwagi,
(Ed.), Pollution Prevention via Process and Product Modification,
AIChE Symposium Series, 90(303), AIChE, New York, 1996.
[103] S.K. Stefanis, A.G. Livingston, E.N. Pistikopoulos, A framework
for minimizing environmental impact of industrial processes. Proc.
The 1995 IChemE Research Event, vol. 1, IChemE, Rugby,
pp. 164166, 1995.
[104] M. Stewart, J.G. Petrie, Life Cycle Assessment for process design
The case of minerals processing, in: M.A. Sanchez, F. Vergara, S.H.
Castro (Eds.), Clean Technology for the Mining Industry,
University of Concepcion, Chile, pp. 6781.
[105] S.K. Stefanis, A.G. Livingston, E.N. Pistikopoulos, . A framework
for minimizing environmental impact of industrial processes. Proc.
The 1995 IChemE Research Event, vol. 1, IChemE, Rugby,
pp. 164166, 1995.
[106] A. Azapagic, R. Clift, 1995a. Whole system modelling and Life
Cycle Assessment, Proc. The 1995 IChemE Research Event, vol. 1,
ICheme, Rugby, pp. 429431.
[107] A. Azapagic, R. Clift, Life Cycle Assessment and linear
programming Environmental optimisation of product system,
Comp. & Chem. Eng. 19 (suppl), (1995b), 229234.
[108] A. Azapagic, R. Clift, S.C. Cowell, J. Lamb, Environmental
management of product system Application of Multi- objective
linear programming to Life Cycle Assessment, Proc. The 1996
IChemE Research Event, vol. 2, ICheme, Rugby, 1996.
[109] A. Azapagic, R. Clift, J. Lamb, Application of Multi-objective
linear programming to environmental process optimisation, AIChE
1996 Spring National Meeting, 2529 February, New Orleans,
1996, LA.
[110] A. Azapagic, R. Clift, Linear programming as a tool in Life Cycle
Assessment, Int. J. LCA 3(6) (1998) 305316.
[111] A. Azapagic, R. Clift, Applying Life Cycle thinking to process
design and operation for the environment, 8th National Meeting of
SAIChE, 1618 April, Cape Town, South Africa, 1997.
[112] The Environment Agency, Best Practicable Environmental Option
Assessment for Integrated Pollution Control, vol. I: Principles and
Methodology, HMSO, London, 1997.
[113] Environmental Data Services, ENDS, Report 269, June, 1997, pp.
3637.
[114] M.A. Meier, Eco-efficiency evaluation of waste gas purification
systems in the chemical industry, in: W. Klopffer, (Ed.), LCA
Documents, vol. 2, Ecomed Publishers, Germany, 1997.
[115] N. de Nevers, Air Pollution Control Engineering, McGraw-Hill,
London, 1995.
[116] T.E. Graedel, B.R. Allenby, Industrial Ecology, Prentice Hall,
Englewood Cliffs, NJ, 1995.
[117] M.M. El-Halwagi, V. Manousiouthakis, Automatic synthesis of
mass-exchange networks with single-component targets, Chem.
Eng. Sci. 45(9) (1990) 2813.
[118] Y.P. Wang, R. Smith, Waste water minimisation, Chem. Eng. Sci.
49(7) (1994) 98.
[119] A.R. Ciric, S.G. Huchette, Economic sensitivity analysis of waste
treatment costs in source reduction projects: Discrete optimisation
problems, Ind. Eng. Chem. Res. 32 (1993) 2636.
[120] A.R. Ciric, T. Jia, Economic sensitivity analysis of waste treatment
costs in source reduction projects: Continuous optimisation
problems, Comput. Chem. Eng. 18(6) (1994) 481495.
[121] A.A. Linninger, C. Han, S.A. Ali, E. Stephanopoulos, G.
Stephanopoulos, Concept of ZAP (zero avoidable pollution) in
the synthesis and evaluation of batch pharmaceutical processes.
Paper presented at the 1994 Annual AIChE Meeting, San
Francisco, CA, 1994.
[122] A.A. Linninger, E. Stephanopoulos, S.A. Ali, C. Han, G.
Stephanopoulos, Generation and assessment of batch processes
with ecological considerations, ESCAPE-5 Conf., 1114 June,
Bled, Slovenia, 1995.
[123] C.A. Wentz, Waste minimisation and resource recovery. In:
Hazardous Waste Management, McGraw-Hill, New York, 1989.
[124] G.B. Dantzig, Linear Programming and Extensions, Princeton
University Press, Princeton, NJ, 1963.
[125] C.A. Floudas, Nonlinear and Mixed-integer Optimization: Funda-
mentals and Applications, Oxford University Press, Oxford, 1995.
[126] GAMS Software GmbH, Giessen, Germany, 1998.
[127] T.F. Edgar, D.M. Himmelblau, Optimization of Chemical Pro-
cesses, McGraw-Hill, New York, 1988.
[128] J.L. Cohon, Multiobjective programming and planning. In:
Mathematics in Science and Engineering, Academic Press, New
York, 1978.
[129] C.L. Hwang, S.R. Paidy, K. Yoon, Mathematical programming with
multiple objectives: A tutorial, Comput. Op. Res. 7 (1980) 531.
[130] D.W. Pearce, Cost-benefit Analysis, 2nd edn., Macmillan, London,
1983.
[131] D.W. Pearce, A. Markandya, E. Barbier, Blueprint for a Green
Economy, Earthscan, London, 1989.
[132] D.W. Pearce, K. Turner, Benefits Estimates and Environmental
Decision-making, OECD, Paris, 1992.
[133] R.K. Turner, D.W. Pearce, I. Bateman, Environmental Economics
An Elementary introduction, Harvester Wheatsheaf, New York,
1994, 328 pp.
[134] R.L. Keeney, H. Raiffa, Decisions with Multiple Objectives:
Preferences and Value Trade-offs, Wiley, New York, 1976.
[135] K.P. Yoon, L. Ching, Multiple Attribute Decision Making:
Introduction, Sage, London, 1995.
[136] A. Azapagic, R. Clift, Life Cycle Assessment and Multi-objective
optimisation, J. Cleaner Prod. 7(2) (1998) 135143.
[137] A. Azapagic, R. Clift, The application of Life Cycle Assessment to
process optimisation. Comp. and Chem. Eng., (1998), in press.
[138] L. Alting, M. Hauschild, H. Wenzel, Environmental assessment in
product development, in: Clean Technology: The Idea and the
Practice. Philosophical Transactions: Mathematical, Physical and
Engineering, 355(1728), Royal Society, London, 1997, pp. 1373
1388.
[139] A. Azapagic, Design for optimum use of resources - cascaded use
of materials. Proc. 2nd Int. Conf. Technology Policy and
Innovation, 35 August, Lisbon, 1998.
[140] Chemical Industries Associations, The UK Indicators of Perfor-
mance 199097, CIA, London, 1998.

Life Cycle Assessment and Its Application To Process Selection, Design and Optimisation

Uploaded by

Copyright:

Available Formats

Life Cycle Assessment and Its Application To Process Selection, Design and Optimisation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Life Cycle Assessment and Its Application To Process Selection, Design and Optimisation

Uploaded by

Copyright:

Available Formats

Review article

Life cycle assessment and its application to

ko-Controlling Umweltschutz in mittel-

ekobilanz von Packstoffen Stand 1990; Schrif-

kobilanzen auf der Basis okologischer Optimierung. Schriften-

You might also like