sustainability of large deployment of photovoltaics ......2 source: pv market outlook european...
TRANSCRIPT
1
Sustainability of Large Deployment of Photovoltaics: Environmental & Grid Integration Research
Sustainability of Large Deployment of Photovoltaics: Environmental & Grid Integration Research
www.clca.columbia.eduwww.pv.bnl.gov
Vasilis FthenakisCenter for Life Cycle Analysis
Columbia Universityand
National Photovoltaics Environmental Research CenterBrookhaven National Laboratory
2Source: PV Market Outlook European Photovoltaic Industry Association 2009
Photovoltaic Global Sales and Projections
Doubling of added capacity every 2 years.
MW/yr
3
A Solar Grand PlanA Solar Grand Plan
Energy Policy 37 (2009)
By 2050 renewable energy to supply 69% of electricity, 35% of total energy needs of the U.S.Zweibel, Mason, Fthenakis
.
4
DOE-EERE Solar Vision Study Report is in review, not to be citedDOE-EERE Solar Vision Study Report is in review, not to be cited
PV Capacity Projections: United States 2030PV Capacity Projections: United States 2030
5
PV –Sustainability Criteria PV –Sustainability Criteria
Low Cost
ResourceAvailability
Lowest Environmental Impact
Affordability in a competitive world
Te in CdTe In in CIGS Ge in a-SiGe & III/VAg in c-Si
Lower than alternatives Life Cycle Impacts & Risks
Affordability - Cost ReductionsAffordability - Cost Reductions
Prices and Production Costs of PV Modules
$1
$10
$100
1 10 100 1,000 10,000 100,000 1,000,000
Cumulative Production (MWp)
Mod
ule
Pric
e (2
006$
/Wp)
Historical Prices
1980
2006
2004
a-Si (Unisolar)Thin Films
CdTe (First Solar)
2006
Courtesy R. Margolis, NRELUpdate V. Fthenakis
avg 2010 costs $0.7/W
Crystalline Si
avg 2010 costs $1.5-2 /W2010
7
Affordability: Projected PV Growth and Electricity Price TargetsAffordability: Projected PV Growth and Electricity Price Targets
Source: J. Lushetsky, Solar Technologies Program, US-DOE, 25th EUPV, Valencia, Spain, Sept. 2010
Geographic LocationsPhoenix, AZKansas City, MONew York, NY
Financing ConditionsLow: 8.2% High: 9.9%
8
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Te (M
T/yr
)
Tellurium for PV* from Copper SmeltersTellurium for PV* from Copper SmeltersTellurium Availability for PV* (MT/yr)
•Global Efficiency of Extracting Te from anode slimes increases to 80% by 2030 (low scenario);90% by 2040 (high scenario)
* 322 MT/yr Te demand for other uses has been subtractedAll the growth in Te production is allocated to PV
Low
High
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Te (M
T/yr
)
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
2010
Te (M
T/yr
)
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Te Availability for PV: Primary + RecycledTe Availability for PV: Primary + Recycled
Tellurium Availability for PV (MT/yr)
Low
High
Recycling every 30-yrs 10% loss in collection10% loss in recycling
Fthenakis V., Renewable & Sustainable Energy Reviews 13, 2746, 2009
CdTe PV Production Constraints CdTe PV Production Constraints Annual (GW/yr)
Most likely
Optimistic
Conservative
Most likely
Optimistic
Fthenakis V., Renewable & Sustainable Energy Reviews 13, 2746, 2009
Cumulative (TW)
Optimistic
Most likely
Conservative
Production can increase with direct mining starting at ~2015
11
Life Cycle Environmental ImpactsLife Cycle Environmental Impacts
Raw Material
Acquisition
Material Processing
Manufacture Decommiss-ioning
Treatment Disposal
Installation/Operation
Recycling
M, Q
E
M, Q M, Q M, Q M, Q M, Q
E E E E E
M, Q: material and energy inputs
E: effluents (air, water, solid)
M, Q
E
12
Life Cycle Environmental ImpactsLife Cycle Environmental Impacts
Raw Material
Acquisition
Material Processing
Manufacture Decommiss-ioning
Treatment Disposal
Operation
Recycling
M, Q
E
M, Q M, Q M, Q M, Q M, Q
E E E E E
M, Q: material and energy inputs
E: effluents (air, water, solid)
M, Q
E
Accidental emissions
Separations,Recovery
Experimental Research at BNL
13
Comparative Life-Cycle AnalysisComparative Life-Cycle Analysis
Energy Payback Times (EPBT) Greenhouse Gas EmissionsResource Use (materials, water, land)EH&S Risks
Zero impact technology does not exist èCompare with other energy producing technologies as benchmarks
14
Energy Payback Times (EPBT)Energy Payback Times (EPBT)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Ribbon-Si13.2%
Multi-Si13.2%
Mono-Si14.0%
CdTe 10.9%
EPB
T (Y
ears
)
BOSFrameFramelessModule
1.8 1.9
0.8
1.3
Insolation: 1700 kWh/m2-yr
- Fthenakis et al., EUPV, 2009- deWild 2009, EUPV, 2009- Alsema & de Wild, Material Research Society, Symposium, 895, 73, 2006- deWild & Alsema, Material Research Society, Symposium, 895, 59, 2006- Fthenakis & Kim, Material Research Society, Symposium, 895, 83, 2006- Fthenakis & Alsema, Progress in Photovoltaics, 14, 275, 2006
Based on data from 13 US and European PV manufacturers
15
Greenhouse Gas (GHG) EmissionsGreenhouse Gas (GHG) EmissionsInsolation: 1700 kWh/m2-yr
0
10
20
30
40
50
Ribbon13.2%
Multi-Si13.2%
Mono-Si14%
CdTe 10.9%
CO
2-eq
(g/k
Wh)
BOSFrameFramelessModule
- Fthenakis & Kim, Encyclopedia of Energy, in press- deWild 2009, EUPV, 2009- Fthenakis et al., EUPV, 2009- Fthenakis & Kim, ES&T, 42, 2168, 2008- Alsema & de Wild, Material Research Society, Symposium, 895, 73, 2006- deWild & Alsema, Material Research Society, Symposium, 895, 59, 2006- Fthenakis & Kim, Material Research Society, Symposium, 895, 83, 2006- Fthenakis & Alsema, Progress in Photovoltaics, 14, 275, 2006
16
0
200
400
600
800
1000
1200
1400
Coal (Kim andDale 2005)
Natural Gas(Kim and Dale
2005)
Petroleum(Kim and Dale
2005)
Nuclear(Baseline -
Fthenakis andKim, 2007
PV, CdTe(Fthenakiset al, 2008)
PV, mc-Si,(Fthenakis et al, 2008)
GH
G (g
CO
2-eq
./kW
h)
MaterialsOperationTransportationFuel Production
GHG Emissions from Life Cycle of Electricity Production: ComparisonsGHG Emissions from Life Cycle of Electricity Production: Comparisons
24 18 30
California Energy Commission, Nuclear Issues Workshop, June 2007 Fthenakis & Kim, Life Cycle Emissions…, Energy Policy, 35, 2549, 2007Fthenakis & Kim, ES&T, 42, 2168, 2008
10.9% 13.2%
17
Sinzheim, Germany, with permission from Juwi, 2006
1.4MW
Dual and Ecological-friendly Use of LandDual and Ecological-friendly Use of Land
Does PV use a lot of land ?
18
Land requirement for US surface coal mining: 320 m2 /GWh
More Land is used by the Coal Life Cycle than PVMountain Top Coal MiningRawl, West Virginia
Fthenakis V. and Kim H.C., Sustainable and Renewable Energy Reviews, 2009
Land requirement for PV in the SW:310 m2 /GWh
PV Plant, Tucson Electric Power, Springerville, Arizona
19
CdTe PV Product Life –Accidental ReleasesCdTe PV Product Life –Accidental Releases
Leaching from shuttered modules• 10 mm fragments -Rain-worst-case scenario- “ leached Cd concentration in the
collected water is no higher than the German drinking water concentration.” (Steinberger, Frauhoffer Institute Solid State Technology, Progress in Photovoltaics, 1998)
• < 4 mm fragments “Leached Cd exceeds the limits for disposal in inert landfill but is lower than limits for ordinary landfills”(Okkenhaug, Norgegian Geotechnical Institute, Report, 2010)
• < 2 mm fragments “CdTe PV sample failed California TTLC and STLC tests”
(Sierra Analytical Labs for the “Non-Toxic Solar Alliance”, 2010)
All PV modules would fail the California tests
c-Si for Ag, Pb, and Cu (ribbon), CIGS for Se; a-Si marginally for Ag
Eberspacher & Fthenakis, 26th IEEEPVSC, 1997; Eberspacher 1998
We advocate for all PV modules to be recycled at the end of their life
20
CdTe PV Product Life –Accidental ReleasesCdTe PV Product Life –Accidental Releases
PV Roof-top fires
Negligible emissions during firesFthenakis, Renewable and Sustainable Energy Reviews, 2004,
Fthenakis et al., Progress in Photovoltaics, 2005
Based on standard protocols by the ASTM and ULExpert Peer reviews by:BNL, US-DOE, 2004EC-JRC, 2004German Ministry of the Environment, (BMU), 2005French Ministry of Ecology, Energy, 2009
21
CdTe PV Fire-Simulation Tests: XRF AnalysisCdTe PV Fire-Simulation Tests: XRF Analysis
Heat
-4
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
0 10000 20000 30000
Cd (counts)
posi
tion
(mm
)
-4.5
-4
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
0 10000 20000 30000
Cd (counts)
posi
tion
(mm
)
-8
-7
-6
-5
-4
-3
-2
-1
0
0 5000 10000
Cd (counts)
posi
tion
(mm
)
XRF-micro-probing –Cd Distribution in PV Glass1000 °C, right end of sample
XRF-micro-spectroscopy -Cd Mapping in PV Glass1000 °C, Section taken from middle of sample
Fthenakis, Fuhrman, Heiser, Lanzirotti, Fitts and Wang, Progress in Photovoltaics, 2005
22
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 20000 40000 60000
Cd (counts)
posi
tion
(mm
)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 50 100 150 200
Zr (counts)
posit
ion (m
m)
XRF-micro-probing -Cd & Zr distribution in PV sample Unheated Sample -Vertical Cross Section
23
XRF-micro-probe -Cd distribution in PV sample760 °C, Section taken from middle of sampleXRF-micro-probe -Cd distribution in PV sample760 °C, Section taken from middle of sample
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 20000 40000 60000
Cd (counts)
posi
tion
(mm
)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 20000 40000
Cd (counts)
posi
tion
(mm
)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 20000 40000
Cd (counts)
posi
tion
(mm
)
24
XRF-micro-probe -Cd distribution in PV sample1000 °C, Section taken from middle of sampleXRF-micro-probe -Cd distribution in PV sample1000 °C, Section taken from middle of sample
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
0 10000 20000 30000
Cd (counts)
posi
tion
(mm
)
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
0 10000 20000 30000 40000
Cd (counts)
posi
tion
(mm
)
-3
-2.5
-2
-1.5
-1
-0.5
0
0 2000 4000 6000 8000
Cd (counts)
posi
tion
(mm
)
25
XRF-micro-probing -Cd distribution in PV sample1000 °C, Section taken from right side of sampleXRF-micro-probing -Cd distribution in PV sample1000 °C, Section taken from right side of sample
-4
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
0 10000 20000 30000
Cd (counts)
posi
tion
(mm
)
-4.5
-4
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
0 10000 20000 30000
Cd (counts)
posi
tion
(mm
)
-8
-7
-6
-5
-4
-3
-2
-1
0
0 5000 10000
Cd (counts)
posi
tion
(mm
)
26
XRF-micro-probing -Cd distribution in PV sample1100 °C, Section taken from middle of sampleXRF-micro-probing -Cd distribution in PV sample1100 °C, Section taken from middle of sample
0
1
2
3
4
5
6
7
8
0 1000 2000 3000
Cd (counts)
posi
tion
(mm
)
0
1
2
3
4
5
6
7
8
9
10
0 2000 4000
Cd (counts)
posi
tion
(mm
)
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
0 2000 4000 6000
Cd (counts)
posi
tion
(mm
)
27
Atmospheric Cd Emissions from the Life-Cycle of CdTe PV Modules –Reference CaseAtmospheric Cd Emissions from the Life-
Cycle of CdTe PV Modules –Reference Case
Process (g Cd/ton Cd*) (% ) (mg Cd/GWh)1. Mining of Zn ores 2.7 0.58 0.022. Zn Smelting/Refining 40 0.58 0.303. Cd purification 6 100 7.794. CdTe Production 6 100 7.795. CdTe PV Manufacturing 0.4* 100 0.52*6. CdTe PV Operation 0.05 100 0.067. CdTe PV Recycling 0.1* 100 0.13*TOTAL EMISSIONS 16.55
Fthenakis V. Renewable and Sustainable Energy Reviews, 8, 303-334, 2004
* 2009 updates
28
Total Life-Cycle Cd Atmospheric EmissionsTotal Life-Cycle Cd Atmospheric Emissions
0.02
0.23
0
0.05
0.1
0.15
0.2
0.25
Cd direct emissions Cd indirect emissions
g Cd
/GW
h
Indirect emissions, due to fossil fuels in the electricity mix in the life-cycle of CdTe PV, are more than 10x higher than direct emissions
3.70.3
44.3
0.90
10
20
30
40
50
0.4 0.7 0.6 0.2
g C
d/G
Wh
Fthenakis and Kim, Thin-Solid Films, 515(15), 5961, 2007Fthenakis, Kim & Alsema, Environ. Sci. Technol, 42, 2168, 2008
29
End-of-life Issues of PV modulesEnd-of-life Issues of PV modules Rapid growth of PV market will result in an
eventual waste disposal issue 25+ years after module installation
Potential of environmental impacts from uncontrolled disposal of PV
30
Recycling of Spent ModulesRecycling of Spent Modules
PV recycling will resolve environmental concerns and will create secondary sources of materials that benefit the environment
CdTe PV recycling is technically and economically feasible
31
The Triangle of SuccessThe Triangle of Success
Low Cost
ResourceAvailability Lowest Environmental Impact
Recycling
§ Major PV Sustainability metrics include cost, resource availability, and environmental impacts
§ These three aspects are closely related; recycling spent modules will become increasingly important in resolving cost, resource, and environmental constraints to large scales of sustainable growth
§ Environmental sustainability should be examined in a holistic, life cycle, comparative framework
email: [email protected]
ConclusionConclusion