California’s Energy Future The View to 2050 -...

80
California’s Energy Future - The View to 2050 Moderator: John Weyant, Precourt Energy Efficiency Center Panelists: Jane Long, Lawrence Livermore National Lab Jeff Greenblatt, Lawrence Berkeley National Lab Heather Youngs, UC Berkeley #SVEnergy2011

Transcript of California’s Energy Future The View to 2050 -...

California’s Energy Future -

The View to 2050

Moderator: John Weyant, Precourt Energy Efficiency Center

Panelists: Jane Long, Lawrence Livermore National Lab

Jeff Greenblatt, Lawrence Berkeley National Lab

Heather Youngs, UC Berkeley

#SVEnergy2011

Silicon Valley Energy Summit,

Stanford University24 June 2011

Jane C. S. LongHeather YoungsJeff Greenblatt

Presenters

Jane Long

Introduction, overview and main results

Heather Youngs

Biomass and biofuels

Jeff Greenblatt

Advanced strategies

California Context

• AB 32 Requires reducing GHG emissions to 1990 levels by 2020 - a reduction of about 25 percent,

• Governor’s executive order S-3-05 (2005) requires an 80 percent reduction below 1990 levels by 2050.

• We must go from 480 mmT CO2e today to 80 mmT CO2e in 40 years

• “Existence proof”: Can it be done, and what needs to change to allow us to get there?

• Focus on technology, GHG emissions and other impacts, not economics

Approach

CEF committeeJane Long and Mim John, co-chairs

• Jeff Greenblatt, LBNL (calculations, writing, building efficiency)

• Burt Richter, Stanford (nuclear)• Heather Youngs, UCB (biofuels)• Jim McMahon, LBNL (building &

appliance efficiency)• Max Wei, LBNL (industry efficiency)• Chris Yang, UCD (transportation)• Bryan Hannegan, EPRI (CCS,

Renewables)• Nate Lewis, Caltech (adv. tech.)• Lyn Orr, Stanford (CCS)• Jan Schori, SMUD (nuclear)• Bill McClean, ret, Sandia (CA

innovation)

Bob BudnitzBob EpsteinLinda CohenChris FieldPeter GleickNalu KaahaainaDan KammenJoan OgdenLarry PapayAri Patrinos

Per PetersonMaxine SavitzGeorge ShultzChris SomervilleDan SperlingJim SweeneyMargaret TaylorCarl WeinbergJohn WeyantMason Willrich

The short answer: Yes, we can

• We can achieve 80% cuts in emissions and still meet our energy needs.

• We can get ~60% of the cuts with technology we largely know about.– Technology in use today or in demonstration. – Deployment will depend more on policy.

• Remaining emissions are due to remaining fossil fuel use

• We can get the rest of the cuts to 80% below 1990, but this will require new technology innovation and development.

Logic–> eliminate fossil fuels*1. How much can we control

demand through efficiency measures?

2. How much do we electrify or convert to hydrogen fuel ?

3. How do we de-carbonize enough electricity to meet resulting electricity demand? How do we balance load?

4. How do we de-carbonize enough fuel (hydrocarbons or hydrogen) to meet remaining demand?

Decrease need for electricity and fuel

Increase demand for electricity, decrease demand for fuel

Nuclear, CCS, RenewablesNatural gas, energy storage,

or demand management

Biofuel, fuel from electricity?

*unless emissions are sequestered

2050 BAU Emissions

(830 MtCO2e)

Demand

GH

G

Inte

nsi

ty

Fuels Electricity

Four Actions to Reduce Emissions

GHG Intensity-Demand Diagram

GH

G

Inte

nsi

ty

Demand

Fuels Electricity

1. Efficiency

Demand

Fuels Electricity

2. Electrification

GH

G

Inte

nsi

ty

Demand

Fuels Electricity

3 + 4. “Low-Carb” Fuels + Electricity

“Low-Carb”Fuels + Electricity

GH

G

Inte

nsi

ty

Electrification Efficiency

Demand

Fuels Electricity

2050 Target Emissions

(80 MtCO2e)

Summary

Technology bins

Bin 1: Deployed at scale now

Bin 2: Has been demonstrated,not available at scale

Bin 3: In development

Bin 4: Research concept

Three sectors of efficiencyand electrification

– Buildings

– Industry

– Transportation

Building efficiency

New

Retrofit

New

Retrofit

Cumulative energy decrease

Building stock turnover

Pre-2010New

Retrofit

Efficiency decrease in energy from BAU

Energ

y d

ecre

ase r

ela

tive t

o B

AU

Light-duty vehicle scenario

Projected Energy Demands

Energy Carrier

Units 2005 2050 BAU

2050 E1

Electricity TWh/yr 270 470 510

HCFuels

bgge*/yr 36 68 25

*Billion gallons gasoline equivalent

Double electricity and de-carbonize it

There are 3 Low-Carbon Electricity Options

Nuclear62% nuclear

44GW33% renewables

5% natl gas load balancing

Fossil/CCS62% fossil/CCS

49 GW 33% renewables5% natl gas load

balancing

Renewables90% renewables

(70% intermittent)160 GW

10% natl gasload balancing

(

Natural Gas*

Flexible LoadsEnergy Storage

Increasingemissions

More difficultto implement

Moreexpensive

Load balancing can add emissions:

* May be possible with CCS in future

Zero-Emission Load Balancing (ZELB)

ZELB

• More challenging for the maximum renewables case– GW-days of storage needed

– Smart grid solution is a challenge• Smart meter fiasco

• Completely change business model to demand follows load vs load follows demand

• Need whole different system of system control – but will this ever solve the GW-day problem?

• Would be easier to have significant baseload power– No more hydro likely

– Renew interest in geothermal energy

– Choose nuclear or CCS

Summary: Electricity

Nuclear

Fewest plantsLess load balancing

Existing laws,safety, public

opinion, waste, etc.

Fossil/CCS

Similar to nuclear, CO2 infrastructure

Reserve CCS for decarbonized fuel?Has emissions: 90%

capture

Renewables

3x plants as baseload(160 vs. 50 GW)

More load balancing and commitment to

ZELB

The median electricity case

• About equal parts of nuclear/renewable/CCS

• A robust choice

• Meets current RPS, but doesn’t exceed it

• ZELB “WAG”: Assume

– ½ load balancing is without emissions

– ½ is with natural gas

• Almost all emissions from electricity are from load balancing

Are we going to have base load power?

• If yes, then:– Do we want to deal with the issues of nuclear poweror– Can we do CCS /fossil for baseload?

• If no, then– Accept the emissions associated with load balancing with

natural gas?– Commit to completely restructure the electric utility?)– Will there be a major breakthrough in energy storage

technology to handle GW-days of demand?– Should we decide to give up on electricity reliability?

• Can’t electrify– Heavy duty transport– Airplanes

• Plus fuel for– Load following– High quality heat

• Biofuels are the choice in the pipeline

Still need 27 billion gge/yr fuel that can’t have CCS

California Biomass

41-121mdt/yr

3-10bgge/yr

California Biomass

41-121mdt/yr

5.5bgge/yr

3 GW(2.0 bgge/yr)

+

7.5bgge/yr

+

Imported

Median Case

Biomass GHG Intensity and Supply

Historical and BAU Emissions

Getting to 60%: Any Single Action

Getting to 60%: Any 2 Actions

Getting to 60%: Any 3 Actions

Getting to 60%: All 4 Actions

150MtCO2eper year

Strategy Requirements and build rate

Strategy Assumed plant size

Total plant capacity

needed in 2050

Build rate 2011-2050

(Plants/year)

Nuclear 1.5 GW 44 GW 0.7

Fossil/CCS 1.5 GW 49 GW 0.8

Renewables Mix 160 GW

Wind 500 MW 59 GW 3

Central Solar(CSP and PV)

500 MW 57 GW 3

Distrib’d. Solar PV 5 kW 19 GW 100,000

Biomass/CCS 500 MW 7 GW 0.3

CA Biofuels 50 Mgge/yr 5,500 Mgge/yr 3

Key challenges:

• Need aggressive efficiency and electrification– largely not a technology issue

• To baseload or not to baseload:– Load balancing without emissions needed for predominant

renewables portrait– Nuclear power and Fukushima incident or CCS?

• Biofuels are important, uncertain, and probably limited – more from Heather

• Getting to 80% below 1990 level requires advanced approaches – more from Jeff

Conclusions

• Yes, we can: Achieve 60% below 1990 level with technologies we largely know about

– The magnitude of the changes required and the pace of implementation will not occur without sustained and substantial capital investment and policy interventions.

• And yes we could do the whole 80% with significant innovation…

Median biofuel estimate

• CA has 3-10 bgge/yr from sustainable biomass– About 2 bgge/yr would go to electricity– Estimate about 5.5 bgge/yr would go to gaseous and

liquid fuel

• Estimate domestic plus international imports could be about 7.5 bgge (equal to biomass from CA)

• Estimate total ~13 bgge/yr = ~ ½ that required• Can improve carbon signature, but biomass

supply is likely a limitation in meeting all fuel needs

Buildings efficiency technologyBin

no.Space

conditioning and building envelope

Water heating Appliances Electronics Other

1 High efficiency

furnaces (including

heat pumps), high

efficiency air

conditioning

equipment,

occupancy

sensors, fiberglass

super-insulation, cool roofs

High efficiency

water heaters, on-

demand water heaters

Energy Star

appliances

(~20%), soil

sensing clothes-

and

dishwashers,

horizontal- axis

clothes washers,

high-spin clothes dryers

Automatic sleep

mode, more

efficient transformers

More efficient

motors and

fans, LED

lighting,

magnetic

induction

cooktops

2 Vacuum panel

insulation, whole-

building optimal

energy

management

Heat pump water

heaters, solar hot

water, waste heat

recovery, whole-

system integration

Higher efficiency

appliances (~40-

50%)

Network proxying Organic LED

lighting

3 Non-invasive

insulation retrofits

4 Magnetic

refrigeration

Industry technology maturity –> complex

Bin Technologies

1 Ultra high efficiency furnaces, controls and monitoring systems, waste heat recovery systems

2 Membrane technology for separations, super boilers, advanced/hybrid distillation, solar boiler systems

3 Integrated & predictive operations/sensors, advanced

materials and processing, electrified process heating (e.g. microwave), process intensification

4 New membrane materials, advanced materials/coatings

Technology maturity light duty transportation

Bin Light-Duty Vehicles Charging infrastructure and management

1 Hybrid engines,

lightweight materials,

better aerodynamics, low-resistance tires

Low- and high-voltage

charging hardware, simple

charging (on-demand or timer)

2 Battery- electric and plug-in hybrids

“Smart” charging via signals from utility or control service

3 Advanced batteries Two-way electricity flow (“Vehicle-to-grid”)

4 None

Nuclear and CCS technology binsBin Nuclear

Technology

Coal or Natural Gas CO2 Capture CO2 Storage

1 Generation III+

reactors

High-efficiency coal gasification, high-efficiency

natural gas combined cycle, ultra-supercritical

pulverized coal combustion, solid-oxide fuel cell

(SOFC), solvent separation

Injection into

oil/gas reservoirs

2 Small modular

reactors (LWR)

Post-combustion CO2 capture technologies with 90%

capture efficiency, integrated gasification systems

with CCS, amine solvent separation

Saline aquifer

injection

3 Generation IV

(including small

modular Na-

cooled reactors)

New capture methods with >90% effectiveness,

lower cost CO2 capture technologies of all kinds,

metal-organic framework separations, membrane

separation

Coal bed injection

4 None None Shale injection

Renewable technology binsBin Wind Concentra-

ted Solar

Power

(CSP)

Solar

Photovol-

taic (PV)

Geothermal Hydro

and

Ocean

Biomass

1 Onshore,

shallow

offshore

turbines

Parabolic

trough,

central

receiver

Silicon PV,

Thin-film

PV,

Concen-

trating PV

Conventional

geothermal

Conven-

tional

hydro

Coal/bio-

mass co-

firing,

direct fired

biomass

2 Dish Stirling Biomass

gasification

3 Floating

(deepwater)

offshore

turbines

”Third

generation”

PV

Wave,

tidal and

river

turbines

4 High-

altitude

wind

Enhanced

geothermal

systems

(EGS)

Load balancing technology bins

Bin Natural

Gas

Storage* Demand Side

Management

1 Combustion

turbine

Pumped hydro Commercial-scale

critical peak demand

response

2 “First generation” compressed air energy

storage (CAES), battery technologies (Na/S,

advanced Pb/Acid, Ni/Cd, Li ion as found in

electric vehicles)

Commercial time-of-

use demand-side

management

3 Variable

fossil

generation

with CCS

Battery technologies (some advanced

Pb/Acid, Vanadium redox, Vanadium flow,

Zn/Br redox, Zn/Br flow, Fe/Cr redox, some

Li ion), flywheel, “second generation”

CAES

Residential time-of-

use demand-side

management

Heather Youngs, PhDBioenergy AnalystUniversity of California, Berkeley

Youngs – California's Energy Future 2011

California Fuel Demand: 23 bgge

Youngs – California's Energy Future 2011

2050*

16

bb

ge

38

bb

ge

2006

2030

2050(doubling)

Effi

cien

cy +

El

ectr

ific

atio

n

Liquid Fuel

2005

2050(doubling)

23

.9 b

bge

(28

Bth

erm

s)2050*

7 b

bge

(0.8

Bth

erm

s)

Effi

cien

cy +

El

ectr

ific

atio

n

Gaseous Fuel (an additional 3-6 bgge could be required for electricity load balancing)

Need deep replacement of fuels to meet the GHG Goals

Youngs – California's Energy Future 2011

Assumptions behind decarbonizing fuel

– Decarbonizing fuel could be resource limited• Policy goal is 75% in-state production by 2050

• 60% available biomass residues are used (ag, forest, MSW)

• Limited energy crop production to 50% of abandoned agland and 50% unused timber land

• Imported biofuels are limited to equal the in-state supply

Youngs – California's Energy Future 2011

California Feedstocks: Plants

Agave

Youngs – California's Energy Future 2011

Eucalyptus – salt tolerantAgave - drought tolerant

Woody energy crops(0-20 mtons/yr)*

Woody residues(17-24 mtons/yr)

Herbaceous energy crops(4.5-21 mtons/yr)

Crop residues(4-7 mtons/yr)

MSW(10-40 mtons/yr)

Animal Waste(5.5-9 mtons/yr)

Municipal Wastewater(3 billion gallons per day)**

*technical recoverable yield (50-80% of gross biomass production depending on type)**not currently used for energy production

Biogas(220 Mtherms)

Liquid Biofuel(7.5 bgge)

Electricity(12 GW)

Cogeneration(125 plants)

1

2

Cogeneration(125 plants)

combustion

combustion

4

3

1.3 GW

2.5 bgge

3 bgge

6.6 x1012 btu

10.8 x1012 btu/yr

2 bgge

10 GW

4.5 x1012 btu

0.7 GW

31 mtons/yr

10 mtons/yr

Decisions regarding biomass use

Youngs – California's Energy Future 2011

California Biomass Technical Availability Scenarios (no economics)

Youngs – California's Energy Future 2011

Fuel Yield3-12 billion gallons gasoline equivalent

40-100 mtons = 3.2-8 bgge residues5-40 mtons = 0.5-3.2 bgge energy crops

Scenario Differences- Improved residue recovery (up to 62% from 40%- Increase in MSW production to correlate with population growth- Growth of additional energy crops (woody and herbaceous) on abandoned ag. and non-productive forest lands

How much biofuel can be produced from CA resources?

Youngs – California's Energy Future 2011

• Policy goal is 75% in-state production by 2050 • Our projections indicate only 12-45% of high efficiency liquid fuel demand is possible from in-state resources without substantial effects on agriculture • California will likely have to import biofuels

(the state imported 73% of refined petroleum products in 2007and 96% of ethanol, mostly from the Midwest and Brazil)

• Limitations on biofuel production and import lead to fossil fuel use and increased GHG emissions

CA biomass (high yield)

10 bgge

Imported biofuel 13 bgge

Deficit load balancing 3-

6 bgge

(low yield)3.5 bgge

Imported biofuel

19.5 bgge)

Deficit load balancing 3-

6 bgge

CA biomass median

case7.5 bgge

Imported biofuel

7.5 bgge

Deficit Fossil Use

8 bgge

Deficit load balancing 3-

6 bgge

16

bb

ge2050 high efficiency

target(16 bgge)

23

bb

ge

2050 high eff. target

(7 bgge/0.8 Btherms)

7-1

3 b

bge

(0.8

-1.4

Qu

ads)

Liq

uid

fu

elG

aseo

us

fuel

Load balancing 60% 3 bgge

Load balancing 80% 3 bgge

29

bb

ge

Residues7.5 bgge

Enegycrops20 bgge

low median high max

Current Limitations to Bioenergy in California are Inter-related

• Socio-Economics– Biomass availability, transportation costs

• Corn from the midwest!• Bioelectricity feedstock cost $20-$60/MWh – fixed price contracts at $45-$70/MWh

– Availability of financing – Cost of recovering waste biomass– Cost of establishing new energy crops

• Interactions with current agriculture and forestry industry

– Farmer/Forester adoption of novel energy crops– Biogas and pipeline interconnection standards (developer pays)

• Permitting (even in a generally policy-friendly context!)– Bluefire goes to Mississippi– Fewer than 1% dairy have functioning digestors– Biogas and pipeline interconnection standards

• Policy conflicts– Definitions of “renewable biomass”– Local and state policies for MSW management– Landfill gas flaring v. on-site electricity or pipeline injection

Youngs – California's Energy Future 2011

The state is funding new projects but at same time…California can’t even keep it’s current bioenergy capacity functioning!?!

Youngs – California's Energy Future 2011

Fuels (mgge/yr) 2009 Production Idle Capacity Proposed Projects

Ethanol 21 221 20

Biodiesel 27 78 30

Biomethane <1 8 6

Total 48 307 56

Biomass Electricity (MW) Operating Capacity Idle Capacity Proposed Projects

Solid-fuel biomass 757 139 346

Landfill Gas 422 - 139

Dairy Digestors 3.9 4.6 4.3

Other Digestors 60 - 7.9

Biogas/NG cofiring facilities 210 - 359

Unrecovered MSW 75 - 455

Total 1527 144 1311

CEC Alternative Fuel Program 2011-2012 investment planCEC Bioenergy Action Plan

Youngs – CA biofuels - AAAS 2011

0

1

2

3

4

5

Do

llars

pe

r G

J

Transport Costs

Distance Fixed Cost

Ag Residues

Searcy et al. 2007 James Baker Institute 2010

Conclusions

• Bioenergy is emerging as a critical player in meeting California’s GHG goals– Required to provide baseload power if nuclear and CCS are off

the table (e.g woody biomass to electricity)

– Required as a replacement for natural gas to for industrial use and to provide firming of intermittant renewables such as wind and solar

– Required to decarbonize fuels

• Advanced (drop-in) biofuels from residues or low-input lignocellulosic residues do better than E85 and conventional biodiesel

• California will likely need to import at least half its biofuel(the state imported 73% of refined petroleum products in 2007and 96% of ethanol, mostly from the Midwest and Brazil)

Youngs – California's Energy Future 2011

Residues Alone are Not Enough

• Increase in-state biomass– Policy Need: Supports to adopt efficient, non-food bioenergy

crops on idle lands and encourage residue/waste biomass use– Risk: Inappropriate choices could have impacts on water

resources, soil quality, and ecosystem services

• Rely heavily on imported biofuels – Policy Need: Establish sustainable biomass/biofuel certification

standards (e.g. Council on Sustainable Biomass Production)– Risks: Difficult to enforce compliance and leakage (sources

produce low carbon fuel for CA but increase fossil use locally)

Youngs – California's Energy Future 2011

Two Choices

California’s Energy Future:Getting to 80% GHG Reductions

Jeffery GreenblattLawrence Berkeley National Laboratory

Silicon Valley Energy SummitStanford University

24 June 2011

Demand

80% Reduction Target

(77 MtCO2/yr)

Fossil Fuels

Bio-mass Fuels

60% ReductionPortrait

(150 MtCO2/yr)

Fuel Electricity

030 Ggge/yr 500 TWh/yr

Biomass electricity

NG/CCS

Natural Gas Load Following

Nuclear & Renewables

(net zero)

GH

G In

ten

sity

Rel

ati

ve P

erce

nta

ge

0%

100%

Strategies for Getting to 80%

1. 100% effective CCS2. Eliminate fossil/CCS (use nuclear instead)3. 100% ZELB for load balancing4. Net-zero GHG biomass5. Behavior Change (10% reduction in demand)6. Biomass/CCS (20% of electricity, offsets fuels)7. Hydrogen (30% replacement of HC fuels)8. Double biomass supply9. Biomass/Coal/CCS (make fuels + electricity)10. Fuel from sunlight (need net-zero carbon source)11. Fusion electricity12. Others?

Small

GHG Impact

Moderate

Large

Trans-formative

Getting to 80%: Single Strategies from the median

100% CCS?

Capture technology Main constituents CO2 capture limit

Post-combustion CO2 (dilute), N2, O2, H2O ~90%

IGCC pre-combustion CO2, CO, CH4, H2S ~92%

Oxyfuel combustion CO2, O2, H2O 96-99%

Conclusions:• >90% capture is costly, >95% very costly• 100% capture is unlikely without breakthroughs• Even if technically achievable, >90% capture would not save much CO2 (~6 MtCO2/yr in base case)

Elimination of CCS

• Slightly greater CO2 savings than 100% CCS (due to reduced refining emissions), but:

• CCS is probably needed for more than fossil electricity production, e.g., biomass…

100% Zero-Emission Load Balancing

• Can storage costs beat spinning reserves cost and/or peak generation with natural gas?

• Can storage efficiency be increased?

• How much flexible load capacity is there, and at what cost?

• How do we solve the “GW-day problem”?

Net Zero GHG Biomass

• Can lifecycle costs be reduced to zero?

Behaviors to reduce buildings energy 10%

• Allowing greater extremes in building temperatures, water temperatures, light levels and moisture content of clothing and dishes

• Allowing greater time to provide desired outcomes, e.g. air-drying of clothing

• Smaller appliances such as refrigerators or clothes washers

• Less habitable space (the “small home”)

• Less reliance on mechanical devices in favor of manual effort, e.g., manual egg beaters

• Less reliance on electronic entertainment, e.g., playing the guitar instead of watching TV

• Sacrificing quality, e.g., air-dried clothing is often stiffer than when heat-dried

• Lifestyle decisions regarding single family / less urban vsmulti-family/ more urban and within SF/MF choice, and habitable space (above)

• Interactions with other behaviors such as telecommuting

• Technologies such as room dependent space conditioning and occupancy sensor

Behaviors to reduce buildings energy 10%

Behaviors to reduce industrial energy 10%

– Designs which use less raw materials to produce the same products (dematerialization)

– Designing more integrated products that reduces the total number of products manufactured

– Designing longer-lasting products that need replacement less frequently

– Extending the length of product design cycles, reducing waste in production lines, etc.

EXAMPLE:Electronic Convergence – how many electronic display screens per capita?

– Include life-cycle analysis and impacts in business practices such as supply chain management and product design

– Design for ease of recycling or re-use

– Change to less energy-intensive materials

• e.g. composite replacements for steel

– Minimize packaging

– Change business model from consumer ownership to business rental/service –

• could result in more repair and re-use, better recycling and disposal of products, and better maintained and thus efficient end use products

Behaviors to reduce industrial energy 10%

1.2 MWhbiomass/CCS

electricity

1 dry ton biomass

80 ggebiofuels

-0.7 tCO20.1 tCO2

80 ggefossil fuels

0.6 tCO2

+

Net GHG :-0.1 tCO2

Net GHG:0.1 tCO2

Biofuels Biomass/CCSvs.

Hydrogen case

H2 demand rel. to median case

Buildings 0% Cars 22%

Industry 21% Trucks 9%

Fuel shifts Airplanes 0%

Electricity -49 TWh Bus 100%

Fuels -7.7 bgge Rail 0%

Hydrogen 8.0 bgge Marine 0%

H2

Doubling Biomass Supply

15bgge/yr

total188 mdt/yr

raw biomass

CA Biofuels

Imported Biofuels

Fossil Fuels*

(2 bgge/yrelectricity)

13bgge/yr

fuels

Median case*Not including natural gas for CCS

Doubling Biomass Supply

CA Biofuels

Imported Biofuels

Additional CA

Biofuels

30bgge/yr

total376 mdt/yr

raw biomass (2 bgge/yrelectricity)

CA grown?Requires 12%

of CA land!

Where could this land come from?9 million acres abandoned crop land +

8 million acres unproductive timber land Half this acreage = 12 bgge/yr!

Doubling Biomass Supply

CA Biofuels

Imported Biofuels30

bgge/yrtotal

376 mdt/yrraw biomass (2 bgge/yr

electricity)

Imported?Requires 3x

base imports(22.5 bgge/yr)

Biomass and Coal with CCS

45%Biomass

55%Coal

Low Carbon Fuels

CO2(sequestered)

~60%

Electricity4%

46%

94 mdt/yr biomass 12-15 bgge/yr fuels

Life cycle GHGs:~10% of fossil

fuels

Example plant design

Source: Guangjian Liu, Eric D. Larson, Robert H. Williams, Thomas G. Kreutz and Xiangbo Guo (2010) Making Fischer- Tropsch Fuels and Electricity from Coal and Biomass: Performance and Cost Analysis. Energy Fuels, Article ASAP doi: 10.1021/ef101184e

IGCC/FT/CCSplant

Getting to 80%: Example Combinations from the median

Advanced Technologies

• Fuel from sunlight

• Fusion: Really cheap electricity?

• What else could help?

Conclusions

• 80% solutions are achievable with technical (and for behavior, social) innovation

• Multiple strategies are probably needed• Key uncertainties/challenges:

– Biofuels are uncertain, and greatly expanded supplies would change nature of solution

– CCS is an important enabling technology; will it work at scale?– How should hydrogen best be used?– Load balancing without emissions needed (storage and flexible

loads), particularly for renewables

• Further research needed:– Biomass/CCS for electricity– Biomass/Coal/CCS for fuels– Fuel from sunlight