Life Cycle Assessment Green Computing CS 290 N

Life Cycle Assessment Green Computing – CS 290 N – Winter 2009 – January 7

“Apple's Environmental Technologies Department is an integral part of Apple's product teams, providing input that guides product teams toward more environmentally-friendly product design. The Department performs industry-leading work in • reducing the amount of toxic substances in its products, • increasing the energy efficiency of its products, • and lowering the amount of greenhouse gases emitted by its products. In support of this latter effort, the Environmental Technologies Department seeks an engineer to support its life cycle analysis (LCA) initiative. ” 20 November 2008, Jean L. Lee, Ph. D. , Environmental Technologies Department, Apple Inc. Green Computing – CS 290 N – Winter 2009 – January 7

The BIG picture: Needs & Wants Services Source of: Materials Energy Water Land Products Production Sink for: Wastes & Emissions Anthroposphere Ecosphere Industrial production and consumption systems use the environment as source of resources and sink for wastes and emissions Green Computing – CS 290 N – Winter 2009 – January 7

Note: The following case study is for teaching purposes only Question: Which beverage container has the lowest environmental impact? Green Computing – CS 290 N – Winter 2009 – January 7

Material choice for beverage containers Processes causing environmental impacts: Material production Container manufacturing Use & distribution Recycling or disposal Environmental impact indicator: Primary energy requirements Aluminum PET Glass Primary energy requirements for material production (MJ/kg) 211. 5 82. 7 12. 0 Primary energy requirements for container forming (MJ/kg) 10. 4 15. 5 2. 9 Density(kg/m 3) 2, 700 1, 370 2, 460 Green Computing – CS 290 N – Winter 2009 – January 7

Material choice for beverage containers Materials can not be compared on a mass basis. Definition of Functional Unit: Containing 1 liter of beverage Beverage container Content Mass/content 12 fl. oz. aluminum can 0. 473 liter 19 gram 0. 0402 kg/liter 20 fl. oz. PET bottle 0. 591 liter 26 gram 0. 0440 kg/liter 25. 4 fl. oz. glass bottle 0. 750 liter 325 gram 0. 4333 kg/liter Reference flows: • 40. 2 gram of aluminum cans • 44. 0 gram of PET bottles • 433. 3 gram of glass bottles Green Computing – CS 290 N – Winter 2009 – January 7

Material choice for beverage containers How much energy is required to produce the beverage containers? Beverage container Aluminum Material production (MJ/liter) Container forming (MJ/liter) Total (MJ/liter) 211. 5 * 0. 0402 = 8. 5 10. 4 * 0. 0402 = 0. 4 8. 9 PET 82. 7 * 0. 0440 = 3. 6 15. 5 * 0. 0440 = 0. 7 4. 3 Glass 12. 0 * 0. 4333 = 5. 3 2. 9 * 0. 433 = 1. 3 6. 6 How much energy is required to transport the beverage containers? Beverage container Mass (g/liter) Transportation distance (km) Transportation energy (MJ/tonne-km) (MJ/liter) Aluminum 40. 2 500 2. 5 0. 05 PET 44. 0 500 2. 5 0. 05 Glass 433. 3 500 2. 5 0. 54 Green Computing – CS 290 N – Winter 2009 – January 7

Material choice for beverage containers How much energy is saved through beverage container recycling? Beverage container Collection rate Aluminum 0. 52 Metal yield 0. 95 Material Energy requirements (MJ/kg) recycling Primary Secondary rate production 0. 49 Energy yield recovery rate PET 0. 20 0. 80 Glass yield Glass 0. 23 1. 0 0. 16 211. 5 25. 8 Energy savings (MJ/liter) 3. 6 Feedstock Energy (MJ/kg) 39. 8 0. 3 Material Energy requirements (MJ/kg) recycling Primary Secondary rate production 0. 23 12. 0 7. 2 Green Computing – CS 290 N – Winter 2009 – January 7 0. 5

Material choice for beverage containers Results: Beverage container Material Container Use & Container Total production manufacturing distribution 1) recycling 2) energy Aluminum 8. 5 0. 4 0. 1 -3. 6 5. 4 PET 3. 6 0. 7 0. 1 -0. 3 4. 1 Glass 5. 3 1. 3 0. 5 -0. 5 6. 6 1) Based on 500 km transportation 2) Based on current recycling rates Green Computing – CS 290 N – Winter 2009 – January 7

Material choice for beverage containers Conclusion: Products create environmental impacts at all stages of their life cycles → It is important to consider the entire life cycle of products Green Computing – CS 290 N – Winter 2009 – January 7

Material choice for beverage containers Question: How do we know that primary energy requirements is the right environmental impact indicator? Results from a more comprehensive life cycle assessment: Primary Energy Requirements (MJ NCV) Global Warming Potential (kg CO 2 eq) Terrestrial Eco-toxicity Potential (g DCBeq) Aluminum 4. 66 0. 354 1. 073 PET 3. 94 0. 205 0. 553 Glass 6. 88 0. 426 0. 430 Conclusion: Products create different types of environmental impacts → It is important to consider a wide range of environmental impacts Green Computing – CS 290 N – Winter 2009 – January 7

Life cycle assessment aims at quantifying the environmental impacts across all relevant environmental concerns and all relevant life cycle stages. Environmental impact categories Life cycle stages Climate Eco. Photo-chemical Ozone Change toxicity Smog depletion Production of materials Manufacturing of product Use & Distribution End-of-life management Total life cycle Green Computing – CS 290 N – Winter 2009 – January 7 Etc.

History and definition of Life Cycle Assessment • Late 1960 s, first Resource and Environmental Profile Analyses (REPAs) (e. g. in 1969 Coca Cola funds study on beverage containers) • Early 1970 s, first LCAs (Sundström, 1973, Sweden, Boustead, 1972, UK, Basler & Hofmann, 1974, Switzerland, Hunt et al. , 1974 USA) • 1980 s, numerous studies without common methodology with contradicting results • 1993, SETAC publishes Guidelines for Life-Cycle Assessment: A ‘Code of Practice’, (Consoli et al. ) • 1997 -2000, ISO publishes Standards 14040 -43, defining the different LCA stages • 1998 -2001, ISO publishes Standards and Technical Reports 14047 -49 • 2000, UNEP and SETAC create Life Cycle Initiative • 2006 ISO publishes Standards 14040 & 14044, which update and replace 14040 -43 Definition of LCA according to ISO 14040: LCA is a technique […] compiling an inventory of relevant inputs and outputs of a product system; evaluating the potential environmental impacts associated with those inputs and outputs; and interpreting the results of the inventory and impact phases in relation to the objectives of the study. Green Computing – CS 290 N – Winter 2009 – January 7

Life cycle assessment terminology (ISO 14040: 2006) Elementary flows (e. g. resource extractions) – input flows Functional unit Economy-environment system boundary economic process Intermediate flow Product system Elementary flows (e. g. emissions to air) – output flows Green Computing – CS 290 N – Winter 2009 – January 7 economic process

Life Cycle Assessment Framework Four different phases of LCA are distinguished: Goal and scope definition Inventory analysis Interpretation Direct application: • product development and improvement • Strategic planning • Public policy making • Marketing • Other Impact assessment Source: ISO 14040 Green Computing – CS 290 N – Winter 2009 – January 7

Life Cycle Assessment Goal and scope definition Inventory analysis Interpretation Impact assessment Green Computing – CS 290 N – Winter 2009 – January 7

Goal and Scope Definition Functional unit and reference flows Procedure: 1. Identify the function of the product system studied 3. Specify the function in SI units 4. Determine an appropriate amount of the function 5. Determine and identify the alternative systems studied in terms of reference flows Previous example: Functional Unit: Containing 1 liter of beverage Reference flows: 40. 2 g of alu cans, 44. 0 g of PET bottles, 433. 3 g of glass bottles What are functional units for the comparison of Various paints? 20 m 2 of wall covering with a coloured surface of 98% opacity and a lifetime of 5 years Paper versus plastic bags in supermarkets? Comfortable carrying of X kg and Y m 3 of groceries (what about durability? ) What are the resulting reference flows? Green Computing – CS 290 N – Winter 2009 – January 7

Inventory analysis In the inventory analysis the elementary flows of a product life cycle are quantified. These are all natural resource inputs and waste & emission outputs of all economic processes within the system boundaries. Functional unit (Reference flows) Process flow diagram Unit processes Inventory table for each unit processes Aggregate inventory table for product system Green Computing – CS 290 N – Winter 2009 – January 7

Inventory Analysis Process flow diagram Definition: The process flow diagram is an illustration of all the unit processes to be modeled, including their interrelationships, which are intermediate product flows. trees wood chips logs Wood yard Harvesting paper cup Digester, washing, bleaching Forming Cup use Landfill, recycling adhesive, coating, heat steam, chlorine (? ) oil pulp gas oil, gas Drilling catalyst gas, naphta Refinery catalyst styrene Styrene production PS cup Polymerization, blowing solvent, blowing agent Green Computing – CS 290 N – Winter 2009 – January 7

Information contained in a process inventory INPUTS Intermediate flows Elementary flows Materials Energy Biotic resources Abiotic resources Unit Process OUTPUTS Materials Energy Intermediate flows Emissions to air Emissions to water Emissions to soil Green Computing – CS 290 N – Winter 2009 – January 7 Elementary flows

Main challenges of inventory analysis Even though the methodology of inventory analysis seems relatively straightforward, it is – in fact – complicated by two important issues: • Defining boundaries for the system under analysis: Which processes to include and which to exclude (cut-off problem in LCA) • Allocation of elementary flows if process has more than one economic output: Which output gets which burdens (Allocation problem in LCA) materials energy unit process product A wastes emissions product B Green Computing – CS 290 N – Winter 2009 – January 7

Allocation There are 3 types of processes where allocation is necessary: co-production, waste treatment, recycling and reuse in an open loop. The 3 are treated on the basis of the same allocation rules. materials energy unit process product A wastes emissions product B closed loop open loop Life Cycle A Life Cycle B waste A waste B materials energy unit process wastes emissions A hierarchy of preferred approaches has been defined in ISO 14044, Section 4. 3. 4: 1. Avoiding allocation by dividing the unit process 2. Avoiding allocation by system expansion 3. Allocation on the basis of physical relationship 4. Allocation on the basis of other relationship, i. e. economic value Green Computing – CS 290 N – Winter 2009 – January 7

Mass-based allocation Example: Emissions 1 kg unit process allocated process Emissions 0. 2 kg 20 kg product A 80 kg product B allocated process Emissions 0. 8 kg 80 kg product B On a mass basis, product A is allocated 20% of the emissions. Green Computing – CS 290 N – Winter 2009 – January 7

Economic allocation Example: Emissions 1 kg unit process 20 kg product A $900 80 kg product B $100 allocated process Emissions 0. 9 kg 20 kg product A $900 allocated process Emissions 0. 1 kg 80 kg product B $100 On an economic basis, product A is allocated 90% of the emissions. Green Computing – CS 290 N – Winter 2009 – January 7

Goal and scope definition Inventory analysis Interpretation Impact assessment Green Computing – CS 290 N – Winter 2009 – January 7

Life Cycle Impact Assessment Life Cycle Inventories (LCIs) by themselves do not characterize the environmental performance of a product system. Impact Assessment (IA) aims at connecting the emissions and extractions listed in LCIs on the basis of impact pathways to their potential environmental damages. Life Cycle Inventory results Classification Impact categories Characterization Category indicator results Normalization Environmental profile Valuation One-dimensional environmental score Green Computing – CS 290 N – Winter 2009 – January 7

Elements of LCIA according to ISO 14044 Mandatory elements Selection of impact categories, category indicators and characterization models Classification: Assignment of LCI results to impact categories Characterization: Calculation of category indicator results Category indicator results (LCIA profile) Optional elements: Normalization of category indicator results relative to reference information Grouping Weighting Data quality analysis Green Computing – CS 290 N – Winter 2009 – January 7

Classification LCI 20 kg CO 2 2 kg Methane 5 g CFC-11 2 kg NO 2 1 kg SO 2 Impact Categories Climate change Stratospheric ozone depletion Photochemical oxidant formation Acidification Green Computing – CS 290 N – Winter 2009 – January 7

Classification LCI Characterization Impact Categories 20 kg CO 2 2 kg Methane 5 g CFC-11 2 kg NO 2 1 kg SO 2 Characterization factors Climate change GWP (global warming potential) Stratospheric ozone depletion ODP (ozone depletion potential) POCP (photochemical ozone creation potential) AP (acidification potential) Photochemical oxidant formation Acidification Substance Amount (kg) GWP 100 (kg CO 2 eq/kg) CO 2 20 1 Methane 2 21 CFC-11 0. 005 4000 NO 2 2 SO 2 1 ODP∞ (kg CFC-11 eq/kg) POCP (kg ethylene eq/kg) AP (kg SO 2 eq/kg) 0. 006 1 0. 028 0. 70 1. 00 Green Computing – CS 290 N – Winter 2009 – January 7

Classification LCI Characterization Impact Categories 20 kg CO 2 2 kg Methane 5 g CFC-11 2 kg NO 2 1 kg SO 2 Characterization factors Climate change GWP Stratospheric ozone depletion ODP Photochemical oxidant formation POCP Acidification AP Substance Amount (kg) GWP 100 (kg CO 2 eq/kg) CO 2 20 1 Methane 2 21 CFC-11 0. 005 4000 NO 2 2 SO 2 1 ODP∞ (kg CFC-11 eq/kg) POCP (kg ethylene eq/kg) AP (kg SO 2 eq/kg) 0. 006 1 0. 028 0. 70 1. 00 20· 1 = 20 kg CO 2 eq 2· 21 = 42 kg CO 2 eq 0. 005· 4000 = 20 kg CO 2 eq (20 + 42 + 20) kg CO 2 eq = 82 kg CO 2 eq Indicator Result Green Computing – CS 290 N – Winter 2009 – January 7

Classification LCI Characterization Impact Categories 20 kg CO 2 2 kg Methane 5 g CFC-11 2 kg NO 2 1 kg SO 2 Characterization factors Indicator results Climate change GWP 82 kg CO 2 eq Stratospheric ozone depletion ODP 0. 005 kg CFC-11 eq Photochemical oxidant formation POCP 0. 068 kg ethylene eq Acidification AP Substance Amount (kg) GWP 100 (kg CO 2 eq/kg) CO 2 20 1 Methane 2 21 CFC-11 0. 005 4000 NO 2 2 SO 2 1 ODP∞ (kg CFC-11 eq/kg) POCP (kg ethylene eq/kg) 2. 4 kg SO 2 eq AP (kg SO 2 eq/kg) 0. 006 1 0. 028 0. 70 1. 00 Indicator kg CO 2 eq kg CFC-11 eq kg ethylene eq kg SO 2 eq Results 82 0. 005 0. 068 2. 4 Green Computing – CS 290 N – Winter 2009 – January 7

Impact Assessment The environmental impact pathway Impact pathways consist of linked environmental processes, processes and they express the causal chain of subsequent effects originating from an emission or extraction (environmental intervention). Examples: Increase in effectiveness of communication of results (generally) SO 2 emissions Acid rain Source CFC emissions Acidified lake Dead fish Loss of biodiversity Endpoint Midpoint Tropospheric OD Stratospheric OD UVB exposure Human health Increase in uncertainty for predicting the environmental impact from the initial interventions Green Computing – CS 290 N – Winter 2009 – January 7

Impact Assessment Impact Categories According to ISO 14044, LCI results are first classified into impact categories that are relevant and appropriate for the scope and goal of the LCA study. Example: Carbon dioxide Climate change Methane CFCs Nitrogen oxides Sulphur dioxide Stratospheric ozone depletion Photochemical oxidant formation Acidification A category indicator, representing the amount of impact potential, can be located at any place between the LCI results and the category endpoints. There are currently two main Impact Assessment methods: • Problem oriented IA methods stop quantitative modeling before the end of the impact pathway and link LCI results to so-defined midpoint categories (or environmental problems), like acidification and ozone depletion. • Damage oriented IA methods, which model the cause-effect chain up to the endpoints or environmental damages, link LCI results to endpoint categories. Green Computing – CS 290 N – Winter 2009 – January 7

Impact Assessment Classification and characterization – Example 1 Impact category Climate change LCI results Emissions of greenhouse gases to the air (in kg) Characterization model the model developed by the IPCC defining the global warming potential of different gases Category indicator Infrared radiative forcing (W/m 2) Characterization factor Global warming potential for a 100 -year time horizon (GWP 100) for each GHG emission to the air (in kg CO 2 equivalents/kg emission) Unit of indicator result kg (CO 2 eq) Substance Carbon dioxide Methane CFC-11 CFC-13 HCFC-123 HCFC-142 b Perfluoroethane Perfluoromethane Sulphur hexafluoride GWP 100 (in kg CO 2 equivalents/kg emission) 1 21 4000 11700 93 2000 9200 6500 23900 Source: (Guinée et al. , 2002) Green Computing – CS 290 N – Winter 2009 – January 7

Impact Assessment Classification and characterization – Example 2 Impact category LCI results Characterization model Category indicator Characterization factor Unit of indicator result Substance ammonia hydrogen chloride hydrogen fluoride hydrogen sulfide nitric acid Nitrogen dioxide Nitrogen monoxide Sulfur dioxide Sulphuric acid Acidification Emissions of acidifying substances to the air (in kg) RAINS 10 model, developed by IIASA, describing the fate and deposition of acidifying substances, adapted to LCA Deposition/acidification critical load Acidification potential (AP) for each acidifying emission to the air (in kg SO 2 equivalents/kg emission) kg (SO 2 eq) AP (in kg SO 2 equivalents/kg emission) 1. 88 0. 88 1. 60 1. 88 0. 51 0. 70 1. 07 1. 00 0. 65 Source: (Guinée et al. , 2002) Green Computing – CS 290 N – Winter 2009 – January 7

Outlook and future developments for LCA Issues to be solved: • Money and time required to do LCAs (especially important of SMEs) • Data availability (public databases, e. g. ELCD and U. S. LCI) • Impact assessment methodology not fully mature (especially toxicity indicators) • Multidimensionality (multi criteria decision making) • Relationship with Environmental Management Systems • Product perspective is not whole system perspective (Most important example: economic relationships) Technical developments: • Consequential LCA (to resolve allocation issues) • Hybrid LCA (Process+I/O LCA) (to resolve cut-off issues) • Modeling economic relationships in and between product systems • Modeling non-linear and dynamic relationships in and between product systems • Modeling spatial aspects of LCI and LCIA Green Computing – CS 290 N – Winter 2009 – January 7

Environmental Product Design – Example: Cell Phones Green Computing – CS 290 N – Winter 2009 – January 7

Material Composition of Cell Phones Plastics 40 -50% Glass and Ceramics 15 -20% Ferrous metals ~ 3% Non ferrous metals 22 -37% Other 5 -10% Green Computing – CS 290 N – Winter 2009 – January 7

Cell Phone Evolution Green Computing – CS 290 N – Winter 2009 – January 7

Cell Phone Components • Plastic housing and keypad • Liquid crystal display (LCD) • Printed wiring board (PWB) • Connectors • Active electronic components (e. g. integrated circuits) • Passive electronic components (e. g. capacitors and resistors) • Microphones and speakers Green Computing – CS 290 N – Winter 2009 – January 7

Global Cell Phone Market Green Computing – CS 290 N – Winter 2009 – January 7

Life Cycle of a Cell Phone Integrated Product Policy (IPP) Pilot Project (http: //ec. europa. eu/environment/ipp/mobile. htm) Green Computing – CS 290 N – Winter 2009 – January 7

Environmental Assessments of Cell Phones at Nokia Wright 1999: Life cycle energy analysis • Scope: ‘ 92 -’ 94 (160 gr) and ‘ 95 -’ 96 (130 gr) cell phones, production, use, eol management, exclude battery, charger, network infrastructure • Functional unit: Use of the cell phone for 2. 5 years • Impact categories: Primary energy consumption (PEC) Frey 2002: Environmental footprint analysis • Scope: ‘ 92 -’ 94 (160 gr) and ‘ 95 -’ 96 (130 gr) cell phones, production, use, eol management, exclude battery, charger, network infrastructure • Functional unit: Use of the cell phone for 2. 5 years • Indicator: Total area required to produce required resources and assimilate generated wastes Mc. Laren & Piukkula 2003: Life cycle assessment (using Ga. Bi 3) • Scope: 2000 cell phone (90 gr), production and use, no eol management include battery and charger, exclude network infrastructure • Functional unit: Use of the cell phone for 2 years • Impact categories: Primary energy consumption (PEC), global warming potential (GWP), Ozone depletion potential (ODP), acidification potential (AP), human toxicity potential (HTP), photochemical oxidant creation potential (POCP) Green Computing – CS 290 N – Winter 2009 – January 7

Summary of environmental hotspots of a cell phone • Life cycle stages: Component manufacture, use phase, end of life • Environmental concern: energy consumption, hazardous wastes & emissions • Use phase: Stand-by power consumption of the charger • Component manufacture: Energy consumption of manufacturing processes • Components with highest environmental impacts: PWB, ICs, LCD • Transportation: Airfreight accounts for almost all of environmental impacts • End-of-life: Hazardous substances in products (e. g. Pb, Cr, Ni, Cu, Sb) • Beyond the handset: Energy consumption of radio base station Green Computing – CS 290 N – Winter 2009 – January 7

Cell Phone Life Cycle: Primary Energy Requirements (PER) Life cycle stage PER (in MJ) PER (in % of total) Materials production 1) 25 9 Component manufacture 1) 100 36 Product assembly 1) 25 9 Transportation 25 9 Use 100 36 0 0 275 100 End-of-life disposal 1) Total 1) 2003 Nokia study gives only 150 MJ for product manufacture. Breakdown is from an earlier Nokia study from 1999, as is the end-of-life assessment. Perspective: 275 MJ is the gross calorific value of 7. 9 liters of gasoline, or 52 km in a Lincoln Navigator, or 185 km in a Toyota Prius. Green Computing – CS 290 N – Winter 2009 – January 7

Options for improving life cycle environmental performance of cell phones • Improvement in cell phone design • Optimizing the in-use life-span of cell phone • Less energy and problematic chemicals during component manufacture • Change buying, usage and disposal behavior of consumers • Improve eol management of cell phones • Reduce energy consumption of network infrastructure • Develop environmental assessment methods/tools • Need for policies to support environmental performance improvements Green Computing – CS 290 N – Winter 2009 – January 7

Cell phone end-of-life management options Primary materials production Components manufacture Final phone assembly Phone demand & use End-of-life phone disposal Inspection & sorting End-of-life phone collection Phone refurbishment Component market Metals market Component reuse Phone recycling Green Computing – CS 290 N – Winter 2009 – January 7

Economics of cell phone end-of-life management £ Cost Revenue Recycling Component reuse Refurbishment Green Computing – CS 290 N – Winter 2009 – January 7

Handset mass and gold content have been declining over the past ten years gr % Gold contains: 60% - 80% of the economic value of the materials (depending on the palladium content) 65% - 75% of the energy embodied in the materials Green Computing – CS 290 N – Winter 2009 – January 7

Therefore economic and environmental benefits due to gold recycling has been declining as well MJ £ Green Computing – CS 290 N – Winter 2009 – January 7