Life cyclebased air quality modelling for technology assessment

Life cycle-based air quality modelling for technology assessment and policy applications: the concept and technical considerations Weimin Jiang, Steven C. Smyth, Qiangliang Li Oct 16 -18, 2006

Outline • Introduction • Two current approaches in analysing air quality impact of technologies • The concept of life cycle-based air quality modelling (lc. AQM) for technology assessment and policy applications • Technical considerations for conducting lc. AQM • Summary and discussions Oct 16 -18, 2006 CMAS Conference, Chapel Hill 2

Introduction • A crucial and pressing issue facing human civilization: Rapidly expanding human material needs/desire vs. Availability and sustainable use of natural resources • The three pillars of sustainable development: Social, economic, and environmental • Possible civilized solution: New and emerging technologies, e. g. , biofuels • Key question: Which technologies are really “sustainable”? • What can we (air quality modellers) contribute? Understand potential impact of the technologies on air quality • Who need the answers? − Policy community Oct 16 -18, 2006 − Industry − Anyone who breathes CMAS Conference, Chapel Hill 3

Current approach: LCA (1) • LCA = life cycle assessment • Definition by ISO 14040: “the compilation and evaluation of the inputs, outputs, and potential environmental impacts of a product system throughout its life cycle” • Life cycle: from cradle to grave; individual stages or as a whole e. g. from biomass feedstock production to biofuel combustion • ISO 14042: standard on life cycle impact assessment • ISO 14042 Ga. Bi (a LCA software) impact category indicator characterization factor Oct 16 -18, 2006 photo-oxidant formation tropospheric ozone formation photochemical ozone creation potential (POCP) CMAS Conference, Chapel Hill 4

Current approach: LCA (2) • Emissions and impact assessment are based on a functional unit: e. g. , 1 vehicle-km travelled (VKT), 1 liter of fuel, 1 MJ energy, . . . Emissions POCP Tropos. O 3 formation (g/VKT) (g C 2 H 4 -eq/g emis) (g C 2 H 4 -eq/VKT) Species Diesel SD 100 Carbon monoxide 1. 14 E+00 2. 28 E+00 2. 70 E-02 3. 08 E-02 6. 16 E-02 Nitrogen dioxide 5. 31 E-01 4. 01 E-01 2. 80 E-02 1. 49 E-02 1. 12 E-02 Nitrogen oxides 2. 33 E+00 2. 60 E+00 2. 80 E-02 6. 53 E-02 7. 28 E-02 Alkene (unspecified) 2. 49 E-05 8. 90 E-05 1. 00 E+00 2. 49 E-05 8. 90 E-05 Benzene 7. 58 E-05 5. 41 E-05 2. 18 E-01 1. 65 E-05 1. 18 E-05 Butane 7. 29 E-03 7. 95 E-04 3. 52 E-01 2. 56 E-03 2. 80 E-04 • • • • • • Xylene 1. 03 E-04 3. 67 E-04 1. 06 E+00 1. 09 E-04 3. 89 E-04 VOC (unspecified) 2. 03 E-09 1. 04 E-04 1. 13 E-01 2. 30 E-10 1. 18 E-05 TOTAL 11. 516 7. 535 1. 809 0. 535

Current approach: 3 -D AQM • Use a 3 -D air quality model, such as CMAQ, CAMx, CALGRID, . . . • Detailed atmospheric chemical and physical processes • Spatially, temporally, and chemically resolved • Technology scales considered • Impact on atmospheric pollutant concentrations • Most (if not all) focused on certain life cycle stage(s), e. g. , emissions from vehicle engine combustion Oct 16 -18, 2006 CMAS Conference, Chapel Hill 6

The lc. AQM concept • • lc. AQM: life cycle-based Air Quality Modelling – a natural advance of the current AQM practice with life cycle thinking – integrate the LCA framework defined by ISO 14040 with the current AQM approach Example to illustrate the concept and considerations The potential impact of large scale production and application of Sun. Diesel as a transportation fuel on air quality in Canada and the U. S. from a whole life cycle perspective. The results are to be used to support policy decisions regarding biofuel development. Oct 16 -18, 2006 CMAS Conference, Chapel Hill 7

An operational lc. AQM framework Oct 16 -18, 2006 CMAS Conference, Chapel Hill 8

Technical consideration: System boundary definition • Chemical, physical, and engineering processes in different life cycle stages to be included in the analysis • Sun. Diesel: – Introduced by Choren Industries in Germany – Can be used directly to replace petroleum diesel – Made from cellulose, hemicellulose, and lignin, which are major components of a wide variety of biomasses – Produced through two major chemical processes • biomass gasification syngas (CO, H 2, CO 2, etc) • Fisher-Tropsch synthesis: syngas Sun. Diesel Oct 16 -18, 2006 CMAS Conference, Chapel Hill 9

Life cycle of Sun. Diesel as a transportation fuel Oct 16 -18, 2006 CMAS Conference, Chapel Hill 10

The biomass production stage Oct 16 -18, 2006 CMAS Conference, Chapel Hill 11

The Sun. Diesel production stage Oct 16 -18, 2006 CMAS Conference, Chapel Hill 12

Sun. Diesel production: an energy self-sufficient model Oct 16 -18, 2006 CMAS Conference, Chapel Hill 13

Technical consideration: Modelling scenario definition/design (1) • Possible locations and timing of industrial and agricultural operations in different life cycle stages • Technology application scales and penetration levels • Uncertainties in assumptions: sensitivity tests • Sun. Diesel: – Canada and continental US – Full year of 2050: substantial displacement of petroleum oil by bio-fuels (? ) Oct 16 -18, 2006 CMAS Conference, Chapel Hill 14

Technical consideration: Modelling scenario definition/design (2) – Biomass-growing locations: • Land-use coverage in Canada and US • Forest logging site logging wood residues • Agriculture and other suitable land energy crops • Energy crop yields • Conversion efficiencies of biomass Sun. Diesel • Needs of food crops, animal feed, and energy crops – Sun. Diesel production plant locations: • Close to the biomass growth or collection sites • Competing factors of plant sizes and distances from the biomass sites Oct 16 -18, 2006 CMAS Conference, Chapel Hill 15

Technical consideration: Emissions (1) • Life cycle emissions data: scarce, a major barrier, and require significant efforts • Life cycle thinking in E. I. development: – Cross-checking emissions between different life cycle stages – Ensure completeness and self-consistency among life cycle stages • E. I. used in AQM + LCI (life cycle inventory) used in LCA: emissions data in Ga. Bi, Sima. Pro, Eco. Invent, etc. • Spatial surrogate/ratios, temporal factors: based on scenario definition/design assumptions and to be studied through sensitivity tests Oct 16 -18, 2006 CMAS Conference, Chapel Hill 16

Technical consideration: Emissions (2) • Removal of old-technology emissions: analysis of SIC & SCC codes, and emis. source descriptions • Sun. Diesel: – Emissions from some life cycle stages can be assembled or derived: • Ga. Bi: Functional unit-based NOx, VOC, SO 2, PM, and heavy metal emissions for various processes related to different fertilizers, fuels, and power • Analysis of fertilizer and energy needs for biomass growth, and transportation, storage, and dispensing • Speciated VOC emissions VOC speciation profiles for some life cycle stages • Oct 16 -18, 2006 CMAS Conference, Chapel Hill 17

Technical consideration: Emissions (3) – Measured emissions from Sun. Diesel production not publicly available a major challenge for the Sun. Diesel lc. AQM. • Effort in estimating the emissions with great uncertainties • Need emissions data: Choren or other FT processes, flash gas combustion – Spatial surrogate ratios: reflect assumed spatial distributions of agricultural & industrial sources within life cycle stages – Temporal factors: reflect seasonality of agriculture and industrial operational schedules – Emissions from petroleum diesel life cycle: to be partially removed to reflect displacement of the fuel by Sun. Diesel Oct 16 -18, 2006 CMAS Conference, Chapel Hill 18

Technical consideration: Model implementation and result analysis • Emissions grouped by major life cycle stages: e. g. , feedstock generation, fuel production, fuel transportation, storage, and dispensing, and fuel usage for transportation purposes, etc. • Model runs with all the emissions, and sensitivity runs with or without emissions from certain life cycle stages Oct 16 -18, 2006 the air quality impact of individual life cycle stages or whole life cycle CMAS Conference, Chapel Hill 19

Summary and discussions • lc. AQM = 3 -D AQM practice with life cycle thinking for analysing technology impact on air quality. • Special considerations: – System boundary definition: lc. AQM foundation – Modelling scenario design: lc. AQM foundation – Emissions data collection, estimation, and analysis: • • Data availability Life cycle thinking in E. I. development for AQM E. I. + LCI Spatial and temporal information associated with life cycle stages – Model runs and analysis: • Based on the whole life cycle or individual life cycle stages Oct 16 -18, 2006 CMAS Conference, Chapel Hill 20

Acknowledgements • Helmut Roth and Albert Chan, ICPET/NRC: Review and comments • Natural Resources Canada: Funding for Sun. Diesel analysis Oct 16 -18, 2006 CMAS Conference, Chapel Hill 21

Oct 16 -18, 2006 CMAS Conference, Chapel Hill 22