2.1
Goal Definition
The ‘Definition of goal and scope’ must be present in any standard life cycle assessment (LCA) study as the first component.1)Here, the fundamental concepts of the study are specified within the framework of the standard. While an iterative approach within the standard is explicitly intended (see double arrows in Figure 1.4), any change of goal and scope must be documented during the conduct of an LCA.
The International Standard 140442)reads:
The goal and scope of an LCA shall be clearly defined and shall be consistent with the intended application. Due to the iterative nature of LCA, the scope may have to be refined during the study.
The goal definition is a declaration made by the organisation (such as companies, industry or trade associations, environmental offices, NGOs, etc.) commissioning an LCA, by providing an explanation to the following3):
• Range of application:What is the objective of the study?
• Interest of realisation:Why is an LCA study conducted?
• Target group(s):For whom will an LCA study be conducted?
• Publication or other accessibility for the public:Are comparative assertions intended in the study?4)
The depth and accuracy of the study have to be considered during the goal definition.
The fundamental standard ISO 14040 explicitly points out that the goal definition and therefore also the application of an LCA represent the commissioner’s free will decision and as such shall not be challenged by the critical review (see Section 2.2.7.3 and Chapter 5).5)Thus, a multiplicity of possible applications (for 1) ISO (2006a).
2) ISO (2006b, Section 4.2.1).
3) SETAC (1993), DIN NAGUS (1994), Neitzel (1996) and ISO (2000b,c).
4) Comparative assertions in the sense of ISO standards mean that product A under environmental aspects is alike or better than product B; products in the sense of LCA standards are any goods andservices.
5) This of course will not apply for ethically non-acceptable goals!
Life Cycle Assessment (LCA): A Guide to Best Practice,First Edition.
Walter Kl¨opffer and Birgit Grahl.
c 2014 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2014 by Wiley-VCH Verlag GmbH & Co. KGaA.
examples, see Section 1.3.2, Table 1.1) are feasible – among others, those preparing environmental policy measures. Since the international standards are quite flexible with regard to the details of the conduct of an LCA (this is valid in particular for the phase of life cycle impact assessment, LCIA; see Chapter 4), first an adaptation of the general methodology to the problem in question must be specified. This is achieved by defining the scope of the study.
2.2 Scope 2.2.1
Product System
First, the examined product system or, in case of comparative LCAs product systems must be clearly described. This includes, above all, the functions of the systems as basis for the definition of the functional unit (fU) (see Section 2.2.5).
The description should be brief, but as precise as possible in this early phase.
A product system is best described in a system flow chart. Figure 2.1 shows a simplified system flow chart of a poly(vinyl chloride) (PVC) window.
In a system flow chart, unit processes and their interrelations are usually represented by boxes. The entire, often very complex, pattern reminds of a tree and is therefore often calledproduct tree. Since an essentially linear system definition is aimed at, branches occur only at the boxes (by several inputs with pre-chains
Oil production and transport
Production PVC
Iron ore mining and transport
Coal mining and transport
Extrusion
Cutting to length Cutting to length Punching, deep-drawing
Production sheet steel
Setting in and screwing steel profile Welding PVC profile
Assembly window Installation
use Demounting
Landfill Material recycling Thermal recovery
Screws Teflon foil Fittings PVC-lip seal glass Fittings seals
System boundary
Figure 2.1 Simplified flow chart of the product system PVC window.
or by several outputs in waste treatment), but no network. An exception is the treatment of recycling, which is discussed in Section 3.3. Within a complete LCA, the presentation ends at the disposal or at a point where co-products, by-products or waste for reutilisation exceed the system boundary (thus leaving the product system).
A special problem arises during the omission of parts of the life cycle. This can, by all means, be justified, if, for example, a provisional estimation showed that the overall system contribution is only very small (criteria: mass, energy, environmental relevance) (see Section 2.2.2.1). However, it must always be examined whether, within comparative studies, and thus in the majority of cases, no asymmetry of systems results from omission. Here, particular attention should be paid to the LCIA, because, compared to mass, very small emissions can nevertheless show large effects. Within comparative LCAs, large parts of the life cycle may be omitted in principle if they match accurately in all systems compared (black box method).
In Figure 2.1, for example, the construction elements on the right of the system boundary (screws, Teflon foil, fittings, etc.) are not considered. This is adequate, if, for instance, different windows (PVC, wood or aluminium windows) are to be compared with each other and if these construction elements are used similarly in all variants regarded. An estimation of relevance of the omitted sections should nevertheless be made so that comparison of systems is not based on completely insignificant differences. If, for example, two systems only differ in waste treatment (End-of-LifeStage) and if these can be neglected in both, the ‘black box’6)approach is inadmissible: both systems are – within an error limit – identical in their environmental behaviour analysed in the specific LCA.
A precise description and quantification of material and energy flow is conducted in the stage ‘life cycle inventory analysis’ (LCI) (see Chapter 3). Should details in the context of LCI analysis indicate an inadequate description of the product system, the description of the scope must be iteratively modified.
2.2.2
Technical System Boundary 2.2.2.1 Cut-Off Criteria
The specification of system boundaries is one of the most important steps in an LCA. When two studies on a similar topic (e.g. single-use vs re-usable packaging) contradict themselves, which may incidentally be the case, usually one or several of the following reasons are responsible:
1. different methodology, 2. different data quality, 3. different system boundaries.
6) ‘Black box’ means a life cycle stage or unit process that may be omitted within comparative LCAs because it is identical within all life cycles to be compared. It should nevertheless be applied sparingly because some most important environmental aspects may be blinded out.
To the first point, great progress has already been made by the International Standardisation (see Section 1.4). To the second, a uniform data format for data bases and data communication has been initiated.7) To the third criterion, no general presetting can be provided because system boundaries depend on the specific problem in question. If, for example, a product is manufactured only in Italy using native raw materials and pre-products and distributed solely in Italy, the European Union as geographical system boundary makes little sense.
Nevertheless, in the component LCIA, transnational emissions and their respective potential impacts have to be considered (see Chapter 4). In this context, as in LCA everywhere,transparencyis very important (see Section 5.4).
The necessity forcut-off criteria, regulating the exclusion of insignificant inputs into the product system, results from the following consideration:
Product systems are embedded into the large systems ‘technosphere’ and
‘environment’.8) It is a fundamental realisation of system analysis that all subsystems are linked, even though more or less intensely. To be able to study a subsystem for itself, numerous less important links must be broken.
For this, rules are necessary. An important rule states that, for example, the infrastructure (roads, rails, etc.) is usually neglected (there are important exceptions,9)however). Something similar is true for capital goods (e.g. the production of machines to manufacture the products), provided these are not the ones to be compared in a study.
ISO 1404410)states three cut-off criteria applied for the entire product system as well as for individual unit processes:
1. mass 2. energy
3. environmental relevance.
Often, a proportion of 1% (mass, energy, etc.) of the overall system is chosen as the cut-off criterion. If a first analysis has, for example, shown that for the manufacture of a product 12 different materials are needed, their percentage ratio is determined in a first step. In the fictitious example of Figure 2.2, component ratios of 5, 6, 9 and 12 are below 1%. The cut-off criterion ‘mass<1%’ alone entails that these components are not balanced over their entire life cycle. However, a first estimation of the energy consumption shows that component 9 has a mass ratio of only 0.2%, although for its production, 2.7% of the total energy is needed.
Therefore, component 9 would be examined through its entire life cycle.
In addition, the rule is often applied that the portion to be cut off shall not exceed 5% per unit process (one box in the product tree). In Figure 2.3, a unit process with 7) ISO (2002) and EC (2010).
8) Both together result in the world in which we live; the technosphere, according to this functional definition, is ‘everything under human control’, and the environment is ‘all that is not technosphere’. Frischeet al. (1982) and Kl¨opffer (1989, 2001).
9) Frischknechtet al. (2004, 2005).
10) ISO (2006b, Section 4.2.3.3.3).
Cut-off rules prevent arbitrariness in the choice of system boundaries Example : Analysing material input
Mass fraction (%) Raw material
Pre-product Ancillary material
Sum
1 73.8 12.0
54.7 23.3 0.9 0.1 < 0.1
< 0.1 0.6 0.7 2.7 4.5 0.4
99.9 1.2
0.1 0.1 1.7 1.4 0.2 19.8 1.7
100.0 < 0.1 3
4 5 6 7 8 9 10 11 13 12 2
Energy (%)
Figure 2.2 Application of the cut-off criteria ‘mass’ and ‘energy’.
Mass percentage input 152.2%
23.7%
9.5%
7.4%
0.9%
0.9%
0.9%
0.9%
0.9%
0.9%
0.9%
0.9%
2 3 4 5 7 8 9 10 11 12 13
Production of product X
Downstream unit processes
Figure 2.3 Application of cut-off criteria: the 5% rule.
13 inputs is presented. The first analysis shows that the mass ratios of the inputs 5–13 are below 1% each. However, the cumulative mass ratio adds up to 7.2%, which would not be traced back to the raw materials in case cut-off criterion of 1% is applied. Therefore, the sole application of the 1% rule would result in large asymmetries when in a second variant, for example, just 1.5% would be the overall cut-off result.
In systems with high energy or mass throughput and simultaneous long life time of the product, the cut-off of less important branches of the product tree,
Energy supply
Raw material extraction
Production, processing, formulation
Distribution/transport
Use, re-use, maintenance
Recycling
Waste management
Inputs Outputs
Energy
Raw material
Solid waste
Products Sewage water
Exhaust gas
Other emissions System boundary
Other
Figure 2.4 System boundary of the inventory modified according to Society of Environmen- tal Toxicology and Chemistry (SETAC) (1991).
infrastructure and so on, contributing less than 1% related to the entire life cycle is usually without problems.11)In any case, an error estimation is required.
The cut-off criterion ‘environmental relevance’ is to prevent, for example, the omission of highly toxic emissions (say polychlorinated dibenzodioxins) in the investigated product systems due to too small masses.
The primarily highly interlaced systems of the complete system analysis become one-dimensional approximations by cutting off links. Interlaced subsystems (loops) are either calculated iteratively or by other suitable mathematical tools.12)Branches without feedback represent no deviation of the linear sequence; they may, however, represent allocation problems (see Section 3.3).
2.2.2.2 Demarcation towards System Surrounding
The system surrounding13) is composed by the ecosphere (‘environment’; see Section 2.2.2.1, ‘all that is not technosphere’) plus the large remainder of the technosphere not included in the analysis. In Figure 2.4, this boundary is called14) system boundary. The system under examination receives input from the system surrounding and delivers output to it.
11) Hunt, Sellers and Franklin (1992).
12) Heijungs and Frischknecht (1998) and Heijungs and Suh (2002).
13) The system surrounding is often called ‘system environment’, which can be misleading.
14) Society of Environmental Toxicology and Chemistry (SETAC) (1991).
Inputs specified in Figure 2.4 originate ‘from the earth’ (+air+water) or are directly or indirectly made available by solar power. The following inputs from the environment are to be considered:
• All processes necessary for the extraction of raw materials (mining industry, oil production, forestry, etc.) belong to the investigated system (‘exploitation of raw materials’).
• In addition, inputs which, due to the cut-off criteria, are not traced over their entire life cycle must be considered (‘miscellaneous’). These inputs may be pre-products, ancillary material or lubricant produced in the remainder of technosphere not included into the system under study.
• The entry ‘energy’ on the input side should actually be named ‘energy raw materials’, because the energy is produced from fossil, nuclear and regenerating raw materials. Exceptions are solar energy, potential energy of water (hydro power) and kinetic energy of wind (wind power). The energy supply, for example, in power stations, is within the system boundary.
Outputs, such as usable products and releases into the environment, are delivered to the system’s surrounding. The investigated product in the centre of the system remains within the system boundary.
• ‘Usable products’ are the product under study, co-products and secondary raw materials(see below), which remain in the technosphere.
• Material emissions are delivered into the ecosphere by waste water and exhaust air. The plants for waste water treatment and exhaust air purification are within the system boundary.
• The allocation of solid wastes(landfill) has in former times occasionally been rated as ‘releases into soil’, which means they would leave the system. Today, controlled landfills are regarded as part of the technosphere and thus lie within the system boundaries. Only degassing and contamination of the groundwater due to leaky landfills are regarded as outputs into the environment. For waste incineration, analogous considerations apply. In the early days of LCA (‘proto- LCAs’),15)the sum of solid wastes has been an important aggregated parameter of the inventory.16)
• ‘Other emissions’ can be radiation, biological releases, noise and similar non- chemical emissions.
The handling of co-products and secondary raw materials requires special attention during the definition of the system boundary.
2.2.2.2.1 Co-products During a chemical synthesis (or any other production process), besides the desired output within the examined product system, further useful products, materials or substances may be generated and covered by the generic termco-products.17)In particular, co-products are frequently formed in the 15) Kl¨opffer (2006).
16) BUWAL (1991).
17) Riebel (1955).
chemical industry, but agriculture and its downstream industries are known for their co-product problem as well. Thus, for example, with the production of grain, straw is produced as a co-product that is transferred as ‘usable product’ to the system surrounding. In this case, environmental loads of the processes must be allocated, according to defined rules, both to the examined product and the co-product (see also Section 3.3, ‘Allocation’, and particularly Section 3.3.2.5, ‘System Expansion’).
Co-products can play a role in different unit processes of a product tree.
2.2.2.2.2 Secondary Raw Material Non-directly usable by-products are usually called residual material. Depending upon the recycling potential, distinction is drawn between ‘secondary raw materials’ (after cleaning or other processing) and
‘wastes’. The ‘Closed Substance Cycle and Waste Management Act’18)in Germany has resulted in different designations for the same issue: wastes for reutilisation and wastes for disposal. Secondary raw materials that are gained from waste for reutilisation leave the system and are used as input in other product systems.
Recycling of materials, which lead to new products, where the materials thus become parts of other systems, is called open loop recycling.19) The respective secondary raw materials (e.g. scrap, waste paper, waste glass, plastic wastes, etc.) leave the product system under study, of which where they are the residual material, as respective wastes for reutilisation.
Within the system boundaries remain materials in those recycling processes that lead back to the same product (the one under investigation), that is, it remains in the product system (closed loop recycling).20) Moreover, in the case of product re-use, these materials remain in the investigated system (usually after cleaning).
Examples of closed loop recycling are the re-feed of plastic shreds, punching, cutting-off, and so on, into the extruder. A good example ofre-useis the refilling of returnable bottles.
Rules to be applied for allocation (Allocation, see Section 3.3.4) shall already be specified within the goal and scope definition; if not avoidance of allocation, for example, by system expansion, becomes compulsory in a specific case21)(see Section 3.3.2.5).
The system boundary requires further explanations, the most important concern being the geographical and temporal system boundary.
2.2.3
Geographical System Boundary
The geographical system boundary results from the economic context and from the product definition:
• Is the concerned special product manufactured in factory A, at site B, and so on, or a group of very similar products manufactured in multiple factories all over 18) German: ‘Kreislaufwirtschaftsgesetz’.
19) Kl¨opffer (1996), Hunt, Sellers and Franklin (1992) and Boustead (1992).
20) SETAC Europe (1992), Curran (1996), Kl¨opffer (1996) and Hunt, Sellers and Franklin (1992).
21) ISO (1998, 2006b).
Europe (North America, Japan, etc. world-wide)? Similar considerations are valid for agricultural products, services, and so on.
• Even if a relatively close framework is selected, for example, production and sales in only one country, the geographical system boundary always has extensions beyond the selected range, because certain raw materials may be missing in the concerned country and thus have to be imported. Therefore, pollution of the environment also occurs in the countries of origin and in transportation from these. For export products, it must be noted that transportation, use and disposal predominantly take place in other countries. The international distribution of tasks in the context of progressive globalisation of the world economy (supplier) must also be considered within the geographical system boundary.
• In LCIA (see Chapter 4), global effects are considered for some impact categories (e.g. climate change/greenhouse effect, stratospheric ozone depletion), while for others regional or local effects (e.g. eutrophication potential) are considered. Local boundaries can, however, be clearly assigned only in rare cases, for instance, if a special product is manufactured in one factory only. In this case, at least one point in the life cycle can unambiguously be assigned geographically. Something similar is valid in agriculture if the farming region can be determined.
Altogether, the definition of the geographical system boundary is straightforward;
it is a question of data availability. Commodities (e.g. metals, mass plastics, chemicals of very large production volume) often do not reveal their origin; in these cases, a regional allocation of impacts is difficult, if not impossible (see Chapter 4).
2.2.4
Temporal System Boundary/Time Horizon
The temporal system boundary is more difficult to define than the geographical boundary.
The minimum specification to the system boundary ‘time’ is a year of reference or another time period for data acquisition. For long-lived products, a determined or estimated lifetime or time of use provides a boundary shifted into the future of the inventory: disposal or re-use will only occur in the future. Accordingly, the modelling of these life cycle phases is difficult and uncertain.
For long, those problems with time have not been sufficiently considered in LCA research.22)This did not play a role as long as predominantly short-lived products were examined such as packaging. Problems related to time became evident when LCAs of building materials, buildings and other long-lived products were carried out:
• How may LCA experts know which (perhaps not even yet invented) methods of waste disposal will predominate in 50 a, how recycling will be organised, and so on?
22) Hofstetter (1996) and Held and Kl¨opffer (2000).