Thomas Weisel Partners

1
Thomas Weisel Partners

• Alternative Energy

Sector Weighting Market Weight
Jeff Osborne 212 271-3577 josborne_at_tweisel.com
2
No Silver Bullet Solution to Worlds Energy Crisis
Source Greentech Media
3
Alternative Energy Investability Matrix

• Solar
• Profitability along entire supply chain, though
  rapid commoditization driving margin contraction.
• Global market cap gt 200bn, though several
  emerging privately held players.
• Cumulative installed capacity 9GW as of 2007
• Addressable market for solar PV could be
  100-200bn between 2008-10, based on 5-10/watt
  installed cost while addressable market for
  equipment manufacturers could be 15-20bn based
  on capacity projections.
• Several technologies exist, no silver bullet
  solution. Cost per watt is key. Solar generated
  power currently costs 0.15-0.45/kWh.
• Of late, competition heating up in thin film
  solar, while solar thermal and concentrated PV
  gaining momentum, though few if any public
  participants.
• Traditional PV cells/modules
• -Sharp
• -Q-Cells
• -Suntech
• Traditional PV vertically integrated
• -REC
• -SunPower
• Thin Film PV
• -Energy Conversion Devices
• -First Solar
• Polysilicon

• Wind
• Highly profitable.
• Global market cap gt 500bn, driven by large
  diversified industrials like GE, Siemens, etc.
• Lack of domestically traded pure-plays lowers
  investability.
• Largest pure play producers not traded in US.
• Cumulative installed capacity 90GW as of 2007
• Addressable market for wind could be 175-250bn
  between 2008-2010 based on 2.00-2.50/watt
  installed cost.
• Key variable in cost is raw materials such as
  steel for turbines and towers.
• Wind-generated power costs as little as 0.05/kWh
  but distribution and transmission becoming more
  important.
• Pure play wind energy players
• Gamesa
• Vestas
• Enercon
• Suzlon
• Clipper
• Diversified industrials
• GE Wind
• Mitsubishi Heavy
• Siemens

• Bioproducts
• Biofuels and bio-products linked to volatile
  input and output prices currently with minimal
  IP lowering investability.
• Large potential market for products substitute
  for petrol in many applications such as plastics,
  motor fuel, cleaners, solvents, etc.
• Market cap 100bn domestically and much larger
  internationally extremely diverse encompassing
  technology, agri business, fuels, etc.
• Highly dependent on government subsidies and
  support public support is beginning to wane as
  corn agricultural prices climb.
• Promise of cellulosic ethanol Improvements in
  enzyme tech are required to meet long term
  ethanol production targets. Developers of these
  enzymes should have strong intellectual property
  position and therefore minimal commodity risk but
  are a long way from viable commercial production
  (Most are private companies)
• Ethanol producers
• -Archer Daniels Midland
• -Aventine
• -Pacific Ethanol
• -Panda
• -Verasun/US BioEnergy
• -Cosan
• Other bioproducts
• -Metabolix
• -Nova Biosource
• -Verenium
• -Xethanol

• Smart Grid/ Energy Efficiency
• Attractive opportunity for investors over next
  3-5 years. Grid upgrade process will be
  evolutionary rather than revolutionary, so
  long-term investment horizon is must.
• Market cap is small (20bn domestic) but
  growing still low market cap limits investabilty
• Selling into highly regulated utility industry
  drives lumpy sales
• Demand response aggregation opportunity alone
  could be 8bn annually in 5-10 years. Smart
  meters/AMI could be 8bn over the next few years,
  while hardware for demand response within AMI
  rollouts could be another 800mn. Addressable
  market for efficient HIF lighting retrofits in
  CI is over 9bn.
• Energy Management/ Efficiency
• -Comverge
• -EnerNOC
• -Microfield
• -Orion Energy Systems
• -PowerSecure
• AMI Vendors
• -Badger
• -Cooper
• -Echelon
• -Itron

• Clean Coal/ Environmental Consulting
• An attractive opportunity created by regulations
  regarding CO2, SOx and NOx reductions both in the
  U.S. and in developing countries..
• Domestic market cap lt 20bn which limits
  investabilty
• Addressable market is over 10bn globally for
  NOx/SOx reduction. Utility boiler efficiency
  improvement market through specialty chemicals is
  3-5bn globally.
• Companies in the space are profitable and often,
  products offered pay for themselves without
  government incentives.
• Clean coal players
• -ADA-ES
• -CECO Environmental--Evergreen Energy
• -Foster Wheeler
• -Fuel Tech
• -Headwaters
• -Peerless Mfg
• -Rentech

• Energy Storage
• Battery/storage technology has remained
  relatively stagnant over the last 20 years. The
  lack of a reliable and economical high-energy
  density energy storage device has limited
  somewhat the potential of renewable energies such
  as wind and solar which generate electricity only
  during certain periods of the day, and fully
  electric vehicles.
• Market cap lt 10bn domestically limiting
  investability
• Current development efforts focus on high energy
  density Li Ion batteries that do not catch fire,
  new nanotechnology-based solutions and ultra
  capacitors.
• Energy storage players
• -Advanced Battery Technologies
• -China BAK Battery
• -Exide
• -Lithium Technology
• -Ultralife Batteries

• Fuel Cells
• Still a long way from commercialization and
  profitability. Efficient production of H2 still
  needed.
• Valuation is difficult due to small order
  volumes, negative profits and lack of cash flow,
  small market caps
• Current opportunities include materials handling
  (6bn), portable electronics (4.2bn), backup
  (2bn) and residential cogeneration (1bn).
  Automotive (90bn) still several years away from
  being a reality.
• Fuel Cell Producers
• Ballard Power Systems
• -FuelCell Energy
• -Hydrogenics
• -Plug Power

Investability
4
TWP Alternative Energy Coverage
Small Cap (lt 1 Bil)
Mid Cap (1-10 Bil)
Large Cap (gt10 Bil)
Overweight
Market Weight
Underweight
5
Table of Contents

• Solar
• Demand Response / Energy Efficiency
• Clean Fuels/Products Industry
• Clean Coal
• Fuel Cells

6
Summary of Opinions

• Solar Best way to invest in alternative energy
• Fast growing, profitable but eventually will be a
  very commoditized product
• Focus on low cost producers (Asian or niche
  technology) in a declining price environment
• Tight credit conditions not just in the U.S. but
  globally are pushing out project timelines, a
  majority of which are financed by debt. At the
  same time internal hurdle rates for system owners
  are increasing as appetite for risk decreases.
• FSLR is our top pick in a tough market for solar
  stocks with lowest cost and committed output at
  pre-determined pricing. Long term winner best
  positioned to tap the U.S. utility scale market
  see upside in late 2009-2010
• BTUI is a microcap play on next generation cell
  lines rolling out strong momentum with STP and
  FSLR.
• SOLR is a mid cap play on aggressive polysilicon
  and wafer capacity coming out of China industry
  leading backlog of 1.5bn.
• Energy Efficiency / Demand Response - Attractive
  long-term secular trend
• We see an attractive opportunity over the next
  three to five years to invest in companies that
  are saving energy as opposed to generating it and
  enabling the transformation of the power grid
  from 1960s technology to a modern, smart grid.
• We expect this upgrade process to be evolutionary
  rather then revolutionary. As such, we think
  investors need to take a long term view when
  evaluating the sector
• We believe demand response alone could be as
  large as 8 billion annually in the next 510
  years
• In this industry, we favor companies offering
  fully integrated energy solutions that are able
  to prove their cost effectiveness, while
  improving the product they are replacing.
• Clean Coal
• Increasing regulation forcing coal plant to clean
  up their act
• The US has the largest reserves of coal in the
  world so we find it unlikely the generation
  source will be replaced anytime soon.
• We view solutions that provide extra value in the
  form of greater plant efficiency, safer
  operations, and less downtime as winners.

7
Solar High Risk, High Return Market
Source Matt Wuerker
Source Good Energies
8
Solar Industry
Solar The best way to invest in alternative
energy

• Key Points
• Demand in 2008 will be high, but investors are
  extrapolating near term trends for the long term
  view of the PV sector. My crystal ball is very
  fuzzy for 2009, so why should multiples expand?
• We are concerned about a potential oversupply as
  early as 2009 the top 10 cell producers of 2007
  alone plan for production of greater than our 5GW
  demand estimate.
• Tight credit conditions are pushing out project
  timelines, a majority of which are funded by
  debt. Also as appetite for risk declines,
  internal hurdle rates for system owners are
  increasing.
• Long term polysilicon prices will be higher than
  you think.
• New polysilicon entrants in China are running
  behind, are not funded, and do not know what they
  are doing.
• Upcoming regulatory changes in key markets create
  demand and pricing uncertainty exiting 2008
• UMG is a stop gap solution. The economics dont
  make much sense with sub-70 poly as best we can
  see it.
• Rapid Commoditization will make low cost and high
  quality key integrated business models likely to
  come out on top in short run.
• The long term key investment opportunity lies
  with the solar capital equipment sector those
  that are enabling the commoditization. If PV
  players beat each other up on price, capex
  budgets will go up more than 2x to 3x between now
  and 2010. We would also expect to see an upgrade
  cycle of lower quality early cycle equipment as
  well.

Near Term Market Running on Rails We Are Nervous
on 2009 as Demand Profile Unclear
2008 Firm markets dominated by German demand
pull in (EEG Effect), Spain Bonanza (1 Year
Grandfather Clause for approved projects), and
Southern EU. U.S. market driven by state
subsidies (CA, NJ,) but pricing is tough due to
FX movement and excess competition. Average ASPs
for the industry in the 3.95-4.40 range. 2009
Spain will more than halve in 2009 with lower cap
compounded with FIT drop which will lead to lower
IRRs, therefore we expect deceleration of the
market. SunPower and QCells are highlighting
Italy as the next Spain, but politics and
bureaucracy have plagued PV thus far. If the ITC
passes in the Energy Bill, US demand could pick
up, but ASPs will be challenged. If ITC is
delayed, we could see a very soft start to 2009.
Likely average ASPs in the 3.50-4.00 range.
2010 US likely a leading market, China and
India?, Thin Film more pervasive. Emergence of
the solar service provider (SunEdison, etc.),
oversupply could be more pronounced
9
Our View from the Capital Markets What is on
Investors Minds?

• Main investor concerns
• Commodity Risk Solar is rapidly commoditizing
  with several new entrants offering an
  undifferentiated product. When will the price of
  polysilicon fall and what will the ultimate
  clearing price be?
• Demand Risk The last 2 years have been driven
  by favorable subsidies by Spain and Germany with
  those markets expected to slow who will fuel the
  growth in 2009?
• Supply Risk Excess returns created by favorable
  government programs and tight supply of
  polysilicon have pushed many new players into the
  solar field, especially in China. When these
  players are no longer restricted by the
  polysilicon shortage how much will they produce?
  Will think film startups hit their aggressive
  goals? Where will all the supply go? At what
  price?
• Cloudy profit picture Unstable polysilicon
  prices and uncertain module prices coupled with
  relatively short operating history leave
  investors guessing as to the long run business
  models in the solar space.
• Credit Risk Tight credit markets are pushing
  out project timelines and internal hurdle rates
  for system owners are increasing as appetite for
  risk decreases. LIBOR has increased gt200bps from
  end of August, prompting further fears of module
  price erosion to make up for increased borrowing
  costs.

10
Solar Cheat Sheet
11
Solar Supply Chain Overview
More of a matter of when than if, we believe
when more polysilicon comes into the market the
commoditization of the solar module will
accelerate, compressing margins in the middle of
the supply chain while benefiting the integrators
and installers with scale and brand recognition
and the equipment producers which enable more
efficient production at the cell and module
level.
Source SunPower and Thomas Weisel Partners LLC
Limited Players High Barriers to Entry Capex
Intensive - 500 mil - 1 bn High margins
Numerous Players Low Barriers to Entry lt50
million gets you into business

• Most loss of polysilicon occurs in the sawing
  process to make wafers known as kerf loss.
• Polysilicon, Ingot, and Wafer is becoming
  vertically integrated.

• RD focus surrounding less polysilicon
  consumption via thinner wafers and greater
  efficiencies.
• Multiple Players Suntech and SunPower as well
  as 100 others

• 45 of the COGS of a solar module is in the
  polysilicon.
• High barriers to entry, rising prices and
  margins.
• Limited Players MEMC, Wacker, REC, DC Chemical
  and Hemlock

12
Composition of Total Installed Cost per Watt
U.S. Residential/Small CI
Source Wall Street Journal and Thomas Weisel
Partners LLC estimates
13
Polysilicon continues to be main bottleneck

• We estimate Polysilicon will make up 45 of the
  cost of a traditional solar cell in 2008.
  Investors have been focused on if and when new
  supplies will come online to reduce overall
  module cost and what the implications of that
  will be.
• Spot market prices have been reported as high as
  400/kg recently.
• New polysilicon entrants need to price at an
  average of 60 or more to make up cost of
  capital so silicon prices will remain well above
  long-term equilibrium levels.
• Worries in polysilicon have moved to thin film
  production before full scale production has begun
  as investors question availability of Indium and
  Tellurium.

Source Company reports and Thomas Weisel
Partners Estimates
14
Polysilicon Supply
Source Solarbuzz, Photon International, company
reports, and Thomas Weisel Partners LLC
15
Polysilicon Supply Who Are These New Entrants?
16
Metallurgical Silicon Part of the Mix but Higher
Long-term Cost Structure As Polysilicon Prices
Ease Below 70 per kg

• Cons
• Both CSIQ and QCE are using custom ingot machines
  crystallization technique is not standard given
  high parts per million of boron and phosphorous.
  Expensive custom equipment is not what we need
  for an effective mass produced industry striving
  for low costs.
• Efficiencies by Timminco are reported in the
  11-14 range, QCells is indicating 15.
• Carrier lifetime does anyone know how long
  these cells will really last?
• Pros
• Cheap to add capacity 10 to 20 of the cost of
  polysilicon.
• Production costs are half of polysilicon less
  energy intensive.

17
Major Plans for Poly Production Announced, But
Where Will All The Funding Come From? Do They
Know What They Are Doing?

• The last 12-18 months large numbers of capacity
  announcements, especially from China.
• Chinese government giving free or nearly free
  electricity to many polysilicon producers, saving
  8-12 per kg of cost.
• Key risk is where and when TCS production will
  ramp up. Siberians are keeping more of it for
  themselves Hoku admitted defeat and asked
  Suntech to bring their own.
• There are 60 proposed polysilicon plants, of
  which 25-30 have broken ground.
• A conservative estimate of announced new Chinese
  capacity coming online by 2010 is 100,000 metric
  tons this implies capex of about 8-10bn.
• New entrants are shooting for an avg. price of
  about 60-65 per kg.
• Some choosing flat rate pricing (Hoku).
• Most choosing aggressive slope downward
  (Chinese).

Quote from New Chinese Polysilicon Producer
Polysilicon production is half art, half
science. We have no artists.
18
Polysilicon Production Process

• What Do I Need To Manufacture My Own Polysilicon?
• Investment Capital - 100,000 per MT Typically
  100 million to 600 million per plant.
• Feedstock Gas Extraordinary high purity levels
  demands high level of technology and
  construction experience.
• Reactor Technology Decompose gas to solid,
  while maintaining purity levels needs tight
  operational controls.
• May are trying, the question is how many will
  succeed. There have been no new entrants in the
  market since 1985.
• Fluidized Bed Reactor (FBR) process has 3 major
  advantages
• Low cost of production 25 vs. 30-35 per kg due
  to less energy use. Production process gets up
  to 1,600 C in Siemens and 600 C in FBR.
• Higher purity levels less boron and phosphorous
• Smaller size allows for maximization of crucible

Source GT Solar
MEMC is about 75 FBR and 25 Siemens, REC and
Wacker are trying to enter the FBR market. REC
recently acknowledged that costs for FBR plant
are running over budget.
19
Thomas Weisel PV Supply/Demand Model
Oversupply likely to happen in 2009 but likely
to be priced in much sooner as high growth and
lofty multiples force investors to look further
into the future
Source Solarbuzz, Photon International, company
reports, and Thomas Weisel Partners LLC
20
Upcoming regulatory changes in key markets create
demand and pricing uncertainty exiting 2008

• Germany
• 40 of the market, slow start to 2008 though
  because all modules are flowing into Spain.
• Strong demand expected to year-end as customers
  are rushing to connect PV systems to the grid
  before 2009 reductions.
• Lower house of parliament in June 2008 approved
  8 FIT decline for rooftop installations in 2009
  and 2010. For ground mount, the FIT decline is
  10 in 2009 and 2010. From 2011, degression for
  all systems is 9 /- 1. Existing 0.05/kWh
  bonus for BIPV eliminated. Final vote in upper
  parliament is pending.
• Module pricing would need to come down by 9-10
  to maintain IRRs.
• Spain
• Spain was the story of 2007 and 2008, emerging
  from the woodwork with improved legislation that
  included a very rich FIT of 44 cents per kWh
  enabling IRRs in the region of 15. Over 1500
  MW currently hooked up to the grid today.
• Bonanza of demand ahead of September 29th
  deadline before new FIT 29-31 cents for ground
  mount kicks in. On track to complete over 1200MW
  in 2008.
• Spain is set to more than halve in 2009 with new
  500MW cap.
• Module pricing would need to come in by 45-50 to
  maintain current IRRs of 15. However, we
  believe a 15-20 decline in pricing is more
  reasonable, bringing down systems IRRs to a more
  normalized 10.
• United States
• The U.S. does not follow a performance-based FIT
  system like Europe, rather provides a one-time
  federal tax incentive of 30 of installed cost,
  uncapped for commercial installations and capped
  at 2,000 for residential.
• The federal tax credit is scheduled to expire on
  Dec 31, 2008. An extension of the credit
  financed by the roll back of tax breaks to oil
  companies has already been vetoed by the
  President.
• Commercial installations financed by PPAs have
  been driven primarily by tax credit hungry
  investors with the expiration of the credit fast
  approaching we are seeing a slowdown in activity.
• California (gt70 of domestic market) with its own
  solar subsidy program is also slowing as state
  performance-based incentives decline in step with
  cumulative installations approved.

21
Overview of Solar Subsidy Programs

• Solar demand is artificially created due to
  government subsidies. The most popular
  government inducement is called a feed-in
  tariff, in which investors and consumers are
  paid to sell power generated to the local
  electric utility at predetermined rates per watt.
• We are most excited about Spain, Greece, and
  Italy. We note that Spains annual cap is very
  low and will need to be lifted in the next 12-18
  months for the industry to continue to succeed
  there.

Source Schott Solar
22
Where Will All The Demand Come From?
Source Thomas Weisel Partners Estimates
23
In the U.S. State Renewable Portfolio Standards
(RPS) Help Drive Renewable Energy Adoption In
Addition To Federal Tax Incentives
Source Interstate Renewable Energy Council
24
March to Grid Parity Solar Already At Grid
Parity Depending on Definition of Grid Parity and
Seasonal and Time of Use Rates
Grid Parity On a Levelized Cost of Energy Basis
Source Thomas Weisel Partners LLC estimates
So What is the Cost of Each Type of Generation
For New Generating Assets?
25
East Coast and West Coast Showing Greatest Energy
Cost Increases Future Key Markets for Solar
Evolving
Source Electric Power Monthly, DOE/EIA-0226 and
Electric Power Annual, DOE/EIA-0346
26
Solar is Already Cost Competitive in Some States
Today, And Will Be in More as Energy Prices
Increase Nationwide

• Applying conservative 4.7 inflationary rate
  (same as increase in national retail rates from
  2002-07 per EIA), average US electricity prices
  will reach 16 c/kWh by 2015
• Rates in some states will be higher, with CT
  electricity prices forecast to reach 27 c/kWh in
  2015 by the same logic

c/kWh
Year
Estimates
27
2007 U.S. Residential PV and Electricity Price
Differences (with existing incentives)
Currently PV is financially competitive where
there is some combination of high electricity
prices, excellent sunshine and/or state/local
incentives.
28
Power Generation Continues to Get More Expensive
Recent Rate Increases California 39 since
2000 Colorado 20 in 2007 Connecticut 50
since 2004 Delaware 59 increase in 2006 Hawaii
100 since 2002 Maryland 85 through
2008 Massachusetts 40 since 2004 Nevada
25 since 2003 New Jersey 35 in 2007 New
York 44 since 2001 Texas 75 since
2000 Virginia 30 increase in 2006 Source
SunEdison
Increases in the price and volatility of fossil
fuels continue to force utilities to raise rates
29
PPA driving commercial installations in U.S., but
expiry of federal tax credit a near-term overhang

• Buy solar power, not the system PPAs remove the
  risk of installing a complicated, capital and
  time intensive solar system and the risk to the
  end user of actual power generated being less
  than expected. We expect the majority of
  commercial systems in the US to be installed in
  the future under some type of PPA.
• In a PPA, a third party institutional investor
  group driven by an appetite for tax credits as
  well as IRR, finances the solar installation and
  maintenance and sells the power generated under a
  long-term, fixed price contract (pricing is
  generally competitive with grid power). The
  purchaser of power simply agrees to the use of
  their rooftop or land to host the solar system.
• The PPA (power purchase agreement) financing
  model generally applied to traditional sources of
  energy, presents an alternative to direct
  ownership of a solar installation, and has caught
  on in a big way in the U.S, driven by the federal
  investment tax credit (30 of installed cost for
  CI). Eg SunPower has secured a 190mn credit
  facility from Morgan Stanley to finance future
  installations.
• Equity investors are typically looking for IRRs
  in the 8-12 range while debt investors are
  looking for a solid credit profile and yields in
  the LIBOR2.75-3.75 range.
• Sweet spot of market is big box retailers.
  Publicized announcements include Costco, Kohls,
  Macys, Safeway, Staples, Wal-Mart.
• Expiration of investment tax credit at the end of
  the year poses a significant risk to commercial
  installations in the U.S. driven by PPAs. We are
  already seeing signs of a slowdown.

30
PV Cell Manufacturers Multiple Players,
Utilization Rates Low

• Barriers to entry in cell manufacturing business
  are low, resulting in a more segmented market.
• Number of competing firms has been increasing
  rapidly, as capital requirements are not
  prohibitive and financing for companies leveraged
  to the solar industry is abundant.
• Any time companies that currently make 6-8 gross
  margins (Taiwanese PCB makers) enter your market
  and capital equipment is cheap and easy to add.
• Equipment utilization rates are running at
  alarmingly low levels due to low barriers to
  entry and the silicon shortage.

Source PV News, Solarbuzz, and Thomas Weisel
Partners LLC
Source Solarbuzz and Thomas Weisel Partners LLC
31
Rapid Commoditization Low Cost and High Quality
Key Cell/Module Makers Get Squeezed

• The solar panel is going the way of mass produced
  electronics equipment (DRAM, disk drives)
  distinguishable only on marginally better
  technology and reliability of the brand.
• Low cost producers with integrated business
  models will fend off margin degradation.
• Asian producers with their low cost models will
  ultimately win market share race but they are
  likely to sacrifice margins along the way as they
  have proven they are willing to do in the past
• Cell/Module gross margins will trend toward
  10-12 despite declining silicon costs in 2009.
• Combined, China and Taiwan go from roughly 23
  market share in 2006 to 41 by YE 2007.

?

Source PV News and Thomas Weisel Partners LLC
32
Overview of Major Steps in Making a Solar Cell

• It typically takes between 1.5 hours to 6 hours
  for a wafer to be converted to a finished module
  depending on the level of automation and inline
  metrology. Most players we have spoken to are
  aiming to produce between 2,400-3,000 cells per
  hour with many facilities still producing 1,500
  cells per hour.
• Wafers are the initial product needed in the
  multi-step process of producing solar cells based
  on silicon wafers. Silicon based solar cells
  made up about 93 of the total industry supply
  last year so this method is most prevalent.
  Before a wafer can be used, it must be tested to
  make sure that it can proceed with the solar cell
  production process. Wafers that are cracked,
  chipped, or have too much content of boron are
  typically rejected.
• In the first production step, damage to the wafer
  that is a result of the sawing process is removed
  using a wet chemical etching process. After
  etching, the wafers are then cleaned using
  another wet chemical process and then dried. For
  the most part, the etching and cleaning is
  automated however we have seen some Chinese
  facilities performing the cleaning part by hand,
  which can be quite dangerous to employees.
• Phosphorous diffusion is the step performed next
  once a wafer has been cleaned and dried. In this
  step, the wafer is placed in a furnace and
  exposed to a gas containing phosphorous. The
  temperature inside the furnace gets up to 900C as
  oxygen is added, which causes a phosphorous oxide
  layer to form on the surface of the wafer inside
  the vacuum. Depending on the length of time the
  wafer is inside the diffusion vacuum and the
  temperature, the phosphorous diffuses into the
  silicon wafer at varying depths. An electrical
  field is formed within the wafer at the boundary
  of the areas within the wafer that contain
  phosphorous and the area that is without
  phosphorous atoms. It is this field that
  generates the electrical current within the wafer
  as it is exposed to sunlight.
• The subsequent step is referred to as edge
  isolation, in which the electrical connection
  between the front and backside of the wafer is
  broken. It is at this point that the remaining
  phosphorous that remains on the wafer is removed
  with a wet chemical process. This phosphorous on
  the edge is referred to as phosphorous silicate.
  In addition, companies we spoke with cited the
  need for more inline metrology in this step to
  assure the homogeneity of the coating layer and
  assure there are no finger prints or water stains
  on the wafer.
• After the edge isolation and wet chemical process
  to remove the phosphorous silicate, an anti
  reflective coating (AR Coating) is applied to the
  front of the wafer to reduce reflectivity of the
  solar cell and improve electrical properties on
  the surface of the wafer. This is a key step in
  improving efficiency of the solar cell.
• After the AR Coating is applied to the wafer, the
  screen printing, or metallization process begins,
  by applying thin metal connectors on the front of
  the wafer. These metal connectors, typically
  silver, are referred to as fingers. The
  fingers go across the wafer horizontally and
  larger vertical connectors called bus bands
  connect the fingers together. Typically there
  are two fatter bus bars per wafer however, some
  wafers contain three. Bus bars are also applied
  to the back of the wafer with a paste and then
  another printing step is performed with aluminum
  as well. After these printings, the wafer is
  dried in an oven. Once the wafer is completely
  dried out, a great deal of heat is applied to the
  contacts to sinter them into the wafer to fully
  connect them to the silicon in the wafer. These
  contacts are used to extract the electrical
  current from inside the wafer and connect them to
  the fingers so that it can be connected to other
  cells and laminated to make a complete module.
  Before a module can be made, the cells must be
  tested and sorted to determine their output so
  that the modules have cells with similar levels
  of efficiency.

33
Opportunity Lies in the Companies that are
Enabling Commoditization Equipment

• Equipment typically makes up about 65-70 of the
  capex budget for cell/module producers.
• The race to 1 GW of capacity by 2009 is leading
  to a bonanza for equipment companies.
• Typical lead times remain in the 12-18 month
  range, with most industry participants telling us
  that down payments are 30-40
• Capex per watt for traditional silicon ingot,
  wafer, cells and modules is about 1.00-1.20 per
  watt.
• Currently there is roughly 7 GW of cell
  production supply globally, with estimates of
  2010 demand ranging from 7 GW on the low end to
  23 GW on the high end. We believe demand will be
  in the 10 GW range.
• Systems today are cobbled together with multiple
  suppliers but turnkey lines are becoming more
  pervasive as new players enter the market that
  know nothing about solar (Moser Baer, PCB players
  in Taiwan).
• There are very few publicly traded equities in
  the solar capital equipment segment to date. With
  only a handful already public (Meyer Burger,
  Centrotherm, Roth Rau, Manz Automation, BTU
  International, etc.), we expect a great deal of
  equity issuances in the segment over the next
  12-18 months as many players capitalize on the
  near-term surge in demand. Traditional semi
  players are looking for a solar angle as well.

Source Thomas Weisel Partners
Source Prometheus Institute
34
Solar Equipment Supply Chain (Capex Assumes 300
MW Plant)
Silicon Manufacturing
TCS Production
Capital Expenditure 450 M
Silicon Reactors
Filament Production
STC Converters
Product Handling
Off Gas Recovery

• Degusssa
• LX Engrg
• SCC

• GT Solar
• MSA
• PPP
• Solmic
• Beijing Design

• CDI

• GT Solar
• MSA
• PPP

• GT Solar
• Solmic

• GT Solar

Wafer Manufacturing
Capital Expenditure 270 M
Inspection Sorting
Feed Stock Prep
Wafer Cleaning
Slurry Recovery
Ingot Growth
Wafering
Sectioning

• GT OEM
• ALD
• Crystalox
• PVA Tepla
• JFE
• Daichi
• Chinese Cos.

• HCT
• MeyerBurger

• HCT
• MeyerBurger

• Manz
• Henneke

• GT OEM

• GT OEM
• Rena

• GT Solar
• HCT
• SiC

Cell Manufacturing
Capital Expenditure 180 M
Incoming Wafer Inspection
Nitride Deposition
Oxide Etching
Sintering (Firing)
Edge Isolation
Testing Sorting
Screen Printing
Diffusion
Etching

• Baccini
• Asys
• BTU

• Despatch
• RTC
• Centrotherm
• SierraTherm

• Baccini

• GT Solar
• Manz
• Baccini
• OTB
• Spire

• BTU
• Despatch
• Centrotherm
• SierraTherm
• Amtech

• Rena
• Stangl
• Schmid

• RothRau
• Semco
• Centrotherm
• Applied Films
• Amtech

• Rena
• Stangl
• Schmid

• Manz
• Rena

Module Manufacturing
Capital Expenditure 90 M
Incoming Cell Inspection
Tabbing Stringing
Assembly
Lamination
Prep
Testing

• Meier
• Spire
• NPC
• 3S

• Rena
• Stangl
• Schmid

• Rena
• Stangl
• Schmid

• Misc. Players

• Somot
• T-Technik

35
Energy Efficiency / Demand Response Industry
The power grid is under-invested and likely to
remain so in the foreseeable future with
electricity demand outgrowing investment in
supply. Demand Response and Energy Efficiency
are solutions
A cheaper, quicker and green alternative to
building new capacity is Demand Response i.e. the
curtailment of energy consumption (when requested
by utilities or grid operators) by end-users of
electricity at times of peak demand. Electrical
equipment such as air conditioning, lighting,
motors, etc. can be switched off automatically
through wireless or internet signals, thus
providing virtual capacity in a few minutes. The
demand response aggregator recruits program
participants as an outsourced service provided to
the utility or grid operator, and is paid based
on the capacity committed and/or made available.
The average payout from the utility is 80,000
per megawatt per year. The aggregator shares
50 with the end-user who committed capacity.
Typical contracts with utilities are 3-10 years
with fixed prices and for capacity of 25-150 MWs.
The aggregator recruits participants, installs
the required hardware at the end-user site and is
on standby 24/7 to deploy the committed capacity
when called upon by the utility or grid operator
during a grid event. The aggregator uses a
paging, cellular or radio network to
automatically send signals to shut down
equipment. We view the Commercial and Industrial
(CI) segment of the end-user market as more
lucrative and larger than the Residential segment
it can take 2-3 years to build out capacity in
a residential program, but only 2-3 months for a
CI program since CI end-users can offer much
higher capacity.
Energy Efficiency, creating capacity by never
using it While demand response addresses peak
capacity issues, energy efficiency has the
potential to curb the growing demand for base
load power generation.
On a business as usual case investment in
generation, transmission and distribution assets
to meet required demand is expected to be 10
trillion between 2005 and 2030. Energy
efficiency, especially in buildings, has a chance
to supplant a meaningful proportion of the
projected energy need. Buildings consume 40 of
all energy in the United States and 72 of all
the electricity produced yields 38 of all
carbon dioxide emissions and 36 of all
greenhouse gas emissions and accounts for 80
(or 238 billion) of total U.S. electricity
expenditures. A 5 increase in building
efficiency would translate to roughly 78 billion
kW hours of electricity saved at 2005 levels, or
roughly 5 billion at a national 0.06-0.07 per
kW price. The multidisciplinary nature of the
industry will require products, installers and
services to be effective. In the past, many
consumers have been burned by promises of energy
conservation that resulted in the acceptance of
sub par products such a scenario is unlikely to
play out today. The need to prove energy savings
will also likely lead to the need for whole
building software and metering solutions, a
segment we also feel will grow in the coming
years. So, in this industry, we believe the
companies most likely to succeed are those
offering fully integrated energy solutions that
are able to prove their cost effectiveness
36
Demand Response Snapshot
37
Clean Fuels/Products Industry
-Ethanol Industry-

• Keys to Watch
• Producers have had major liquidity issues, a
  continued tight financial market could put
  ethanol producers in jeopardy.
• Mid-Western flooding had caused concern for the
  corn supply but initial reports from the USDA
  show the situation might not be as dire as
  feared. At this point corn prices should
  continue weaken as we go into the fall.
• Despite the improving corn costs, expectations
  should be tempered as low ethanol prices and
  infrastructure issues are likely to weigh on the
  stocks until investors get more certain clarity
  on industry dynamics exiting 2008. Cash corn
  prices are set to be well in the 5.00-6.00
  range, which threaten to hurt results. We,
  however, are likely to see industry dynamics
  improving with falling corn costs and look for
  sustained ethanol pricing support to become more
  positive.

The bottom line is the 36bn gal Renewable Fuel
Standard, which includes a very important 9bn gal
biofuel mandate for 2008, is a positive, and we
are cautiously optimistic as to the long-run
viability of the biofuels industry.
On Going Issues In Ethanol

• Commodity risk. Industry fortunes are tied to two
  volatile commodities, ethanol and corn. These
  factors limit earnings visibility.
• Minimal IP. Commodity business with minimal
  technology differentiation. We believe
  ultimately results in price pressure and low
  margins.
• The regulatory uncertainty Government support
  remains key and a new RFS could provides support
  but supply could still outstrip regulated demand.
• Minimal Visibility Without the ability to see a
  long term price for ethanol, which a roughly 5
  cent move in the price causes a roughly 5-10 cent
  move in EPS, we have minimal visibility into
  2009.
• The promise of cellulosic ethanol. Improvements
  in enzyme technology are required to meet long
  term ethanol production targets. Developers of
  these enzymes should have strong intellectual
  property position and therefore minimal commodity
  risk. (Most are private companies)

38
Ethanol Basics
Wet-milling Ethanol Production
Dry-Milling Ethanol Production
Source USDA
Source USDA

• Ethanol is a type of alcohol primarily used as a
  blend component in the U.S. gasoline market.
• Currently in the U.S. ethanol is primarily made
  from corn. Although a variety of feedstocks for
  ethanol production are under development, most
  production in the U.S. uses corn-based
  technology. Most of the corn supplied to the
  ethanol industry is grown in Illinois, Minnesota,
  Nebraska and South Dakota.
• Corn is converted to ethanol with either dry or
  wet-milling technology. The primary difference
  between dry milling and wet milling is the
  pretreatment of the corn. In dry milling the
  corn is crushed as opposed to wet milling where
  it is soaked before processing.
• Dry-milling accounts roughly 3/4 of ethanol
  production capacity and is employed in the
  majority of new production facilities due to
  lower capital costs than wet-milling and a higher
  ethanol yield. Dry-mill production yields roughly
  2.7-2.8 gallons of ethanol and 15-17lbs of dry
  distilled grains (DDGS) at about 150-200/ton
  currently.
• Wet-milling accounts for roughly 1/4 of
  production capacity (RFA) and has a higher yield
  of co-products and lower ethanol yield. The wet
  milling process has several outputs a bushel of
  corn put through the wet milling process produces
  2.5 gallons of ethanol, 1.5lbs of crude corn oil,
  12.4lbs of gluten feed (livestock feed) and 3lbs
  of gluten meal (poultry feed). All of these
  products have viable markets.

39
US Ethanol Supply/Demand Analysis

• We expect a more balanced supply/demand situation
  in 2008 as more difficult operating conditions
  (financing and high input prices) will make it
  tough for new entrances to succeed.
• It now takes in the 100,000-300,000 per day in
  working capital to support the corn requirements
  of a typical corn ethanol plant, start-ups and
  poorly funded plants will have trouble
• We believe plant shutdowns and delayed/canceled
  construction will accelerate as we move further
  into 2008

Source Company reports, USDA, Ethanol Producer
Mag, DOE, and Thomas Weisel Partners LLC
40
Theoretical Ethanol Plant
Fully Depreciated and paid for plants with
longer histories of operation will have an
advantage in this low spread environment,
creating barriers to entry in the industry.
Source Company reports and Thomas Weisel
Partners LLC
41
Clean Fuels/Products Industry
-Clean Coal-

• Keys to Watch
• The U.S. air pollution control market is the
  primary driver in Fuel Techs NOx reduction
  technology segment. Domestic policy for coal can
  be traced back to 1970 with the implementation of
  the Clean Air Act. While increased regulation
  can increase the cost and complexity of running a
  coal plant, there are currently minimal viable
  substitutes to switch to that make economic
  sense. We view the current as well as future
  clean coal regulations as major drivers of
  industry growth.

New requirements regarding carbon dioxide,
nitrous oxide and mercury reductions are leading
to an investment in performance improvement,
while being environmentally friendly.
Industry Thesis

• The United States has the largest coal reserves
  in the world, holding roughly a quarter of global
  supply. Scientists estimate that the United
  States has nearly 200 years worth of supply at
  current run rates. As a result, we do not look
  for coal plants to be shut down any time soon. We
  do, however, expect global policy to require
  capital expenditures to force coal facility
  owners to move to production methodologies that
  emit less pollution. The growth in consumption
  of coal over the last several years has largely
  been due to the rapid growth seen in China and
  India, where the bulk of new capacity additions
  have been coal based.
• The idea of Clean Coal has been around for
  decades, but costs and lack of technology
  availability led to stagnant growth in the
  industry. As greater regulations enter the fray
  and greater awareness of green house gas
  emissions and global warming become a part of
  doing business, we expect to see a prolonged (a
  very prolonged in fact) upgrade cycle. Utilities
  do not move on a dime we have seen that with
  Fuel Techs quarterly results as well as numerous
  other players focused on both the coal and
  broader utility market. We expect the upgrade of
  coal plants to last five to 10 years, if not much
  longer.

42
Fuel Cell Industry
The fuel cell industry faces many challenges
before full-scale commercialization of the
technology takes place. In our view, cost,
durability, system size and heat recovery are all
areas that need further development and
improvement to make fuel cells a commercial
reality.

• Keys to Watch
• In response to increasing grid disruptions and
  the disarray that followed Katrina, the FCC on
  October 3, 2007, issued a ruling that within 12
  months local exchange carriers (LEC) and
  commercial mobile radio (CMR) service providers
  file a certified emergency backup power
  compliance plan that describes how the LECs or
  CMRs provider will supply emergency backup power
  to 100 percent of non complaint assets in the
  event of a commercial grid power failure. A
  minimum of 24 hours of backup power is required
  for assets inside central offices and 8 hours for
  other assets including cell sites, remote
  switches and digital loop carrier system remote
  terminals. It also has a provision that requires
  CMRs and LECs to detail their monitoring of
  backup power capabilities to remain compliant. We
  believe this is a positive to fuel cells as its
  solution shows positive economics versus other
  backup solutions at about five hours. We believe
  this and the forklift market will be the largest
  near term market for fuel cells.
• Technology validation in the form of large orders

Ongoing Industry Issues

• Waning enthusiasm Emergence of solar and
  alternative fuel sector has reduced investor
  attention on fuel cells.
• Long road to Commercialization Still waiting for
  commercial products. Will back up power or fork
  lifts be first markets? The fuel cell industry
  faces multiple challenges that must be overcome
  before full scale commercialization of the
  technology takes place.
• Burn Fuel cells companies still burning cash
  and some may run out in 07
• Valuation Very difficult to value.
• Technology not there yet In our view, cost,
  durability, system size, and heat recovery are
  all areas that need further development and
  improvement to make fuel cells a commercial
  reality.
• Efficient Hydrogen Production still needed To
  truly create pollution-free fuel cells, hydrogen
  must be produced by renewable means and will
  likely take many years before reaching cost
  competitiveness. Hydrogen today for fuel cells
  uses more electricity than it saves.

43
Fuel Cell Basics
A fuel cell is an electrochemical device that
converts chemical energy directly into electrical
energy. Electricity is generated through a
reaction in which oxygen and hydrogen combine to
form water, along with heat that can be recovered
and used. A fuel cell is composed of three
primary components a fuel electrode (the anode),
an oxidant electrode (the cathode), and an
electrolyte in between the two electrodes. When
hydrogen reaches the anode, an electrochemical
reaction takes places, splitting the hydrogen
molecules into protons and electrons. The
protons can pass through the electrolyte, while
the electrons are forced through an external
circuit, allowing the energy to be captured as
electricity. The protons and electrons rejoin at
the cathode, where they react with oxygen to form
water.
Individual Fuel Cell
Source htttp//www.wikipedia.com
Source htttp//www.wikipedia.com
44
Fuel Cells Snapshot