ENERGY TECHNOLOGY SYSTEMS ANALYSIS PROGRAMME

IRENA
International Renewable Energy Agency

Solar Photovoltaics
Technology Brief

IEA-ETSAP and IRENA© Technology Brief E11 – January 2013

www.etsap.org – www.irena.org
This brief is available for download from the following IEA-ETSAP and IRENA sites
iea-etsap.org/web/Supply.asp
www.irena.org/Publications

Copyright © IEA-ETSAP and IRENA 2013

About IRENA
The International Renewable Energy Agency (IRENA) is an intergovernmental organiza-
tion dedicated to renewable energy. In accordance with its Statute, IRENA’s objective is to
“promote the widespread and increased adoption, and the sustainable use of all forms of
renewable energy”. This concerns all forms of energy produced from renewable sources in
a sustainable manner and includes bioenergy, geothermal energy, hydropower, ocean, solar
and wind energy.

As of December 2012, the membership of IRENA comprises some 160 States and the
European Union (EU), out of which 104 States and the EU have ratified the Statute.

About IEA-ETSAP
The Energy Technology Systems Analysis Programme (ETSAP) is an Implementing Agree-
ment of the International Energy Agency (IEA), first established in 1976. It functions as a
consortium of member country teams and invited teams that actively cooperate to establish,
maintain, and expand a consistent multi-country energy/economy/environment/engineering
(4E) analytical capability.

Its backbone consists of individual national teams in nearly 70 countries, and a common,
comparable and combinable methodology, mainly based on the MARKAL / TIMES family
of models, permitting the compilation of long term energy scenarios and in-depth national,
multi-country, and global energy and environmental analyses.

ETSAP promotes and supports the application of technical economic tools at the global,
regional, national and local levels. It aims at preparing sustainable strategies for economic
development, energy security, climate change mitigation and environment.

ETSAP holds open workshops twice a year, to discuss methodologies, disseminate results,
and provide opportunities for new users to get acquainted with advanced energy-technolo-
gies, systems and modeling developments.

Print compensated Id-No. 1225221

www.bvdm-online.de
Insights for Policy Makers
Solar photovoltaic (PV) cells convert sunlight directly into electricity. Currently,
crystalline silicon (c-Si) and the so-called thin-film (TF) technologies dominate
the global PV market. In a c-Si PV system slices (wafers) of solar-grade (high pu-
rity) silicon are made into cells that are assembled into modules and electrically
connected. TF PV technology consists of thin layers of semiconducting material
deposited onto inexpensive, large-size substrates such as glass, polymer or metal.
Crystalline silicon PV is the oldest and currently dominant PV technology with ap-
proximately 85-90% of the PV market share.

The manufacture of solar PV systems basically comprises of four phases: produc-

tion of the semiconducting material (90% of polysilicon is supplied by a handful
of companies in the United States, Japan, Europe, and China); production of the
PV cells, which often requires sophisticated manufacturing (most solar cells are
produced in China, Germany, the US and Japan); production of PV modules, a
labour-intensive process whereby the cells are encapsulated with protective
materials and frames to increase module strength (around 1,200 companies
worldwide currently produce solar PV cells and modules); and installation of PV
modules, including the inverter to connect the PV system to the grid, the power
control systems, energy storage devices (where appropriate) and the final instal-
lation in residential or commercial buildings or in utility-scale plants. The cost of a
PV module typically ranges between 30-50% of the total cost of the system. The
remaining costs include the balance of system and the installation - which can be
as low as 20% for utility-scale PV plants, 50-60% for residential applications, and
as a high as 70% for off-grid systems, including energy storage (usually batteries)
and back-up power.

PV power has an enormous energy potential and is usually seen as an environ-

mentally benign technology. Over the years a good number of countries have
implemented specific policies and incentives to support PV deployment. This has
led to a rapid increase in the total installed capacity of PV from 1.4 GW in 2000
to around 70 GW at the end of 2011, with about 30 GW of capacity installed in
that year alone. The associated industrial learning and market competition have
resulted in very significant and rapid cost reductions for PV systems. Continued
cost reductions for PV systems are an essential requirement for accelerating the
attainment of grid-parity of electricity generated using on-grid solar PV systems.
In countries with good solar resources and high electricity tariffs, residential solar
PV systems have already reached parity with electricity retail prices, whilst in gen-
eral PV is now fully competitive with power generated from diesel-based on- and
off-grid systems.

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 1
 The choice of solar PV technology for installation is often based on a trade-off
between investment cost, module efficiency and electricity tariffs. Compared with
c-Si-based PV systems, the production of TF PV system is less energy-intensive
and requires significantly less active (semiconducting) material. TF solar PV is
therefore generally cheaper, though significantly less efficient and requires sub-
stantially more surface area for the same power output, than c-Si-based systems.
The module cost of c-Si PV systems have fallen by more than 60% over the last
two years; in September 2012, Chinese-made modules averaged USD 0.75/watt,
while TF PV modules. Consequently, even though TF PV has experienced tremen-
dous growth a few years ago, more recently its market share is decreasing and
the current outlook for further growth in the deployment of this technology is
uncertain and will depend heavily on technology innovation.

Solar PV, as a variable renewable electricity source, can be readily integrated into
existing grids up to a penetration level of about 20% depending on the configu-
ration of the existing electricity generation mix and demand profiles. Increasing
the integration of a high level of variable renewable power from PV systems into
electricity grids requires, in general, re-thinking of grid readiness with regards to
connectivity, demand-side response and/or energy storage solutions. However,
the on-going reduction of financial incentives in many leading markets, together
with the overcapacity of the PV manufacturing industry, suggest that module
prices will continue to decline, leading to parity in off- and on-grid PV. It is note-
worthy that, since 2001, the global PV market has grown faster than even the most
optimistic projections. However, it is not clear whether the deployment of PV will
slow down or continue to grow at the same as in the recent past years.

2 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

Highlights
䡵 Process and Technology Status – Photovoltaic (PV) solar cells directly
convert sunlight into electricity, using the photovoltaic effect. The process
works even on cloudy or rainy days, though with reduced the production and
conversion efficiency. PV cells are assembled into modules to build modular
PV systems that are used to generate electricity in both grid-connected and
off-grid applications, such as residential and commercial buildings, industrial
facilities, remote and rural areas and power plants (i.e. utility PV systems).
Over the past decades PV technology has been constantly improving perfor-
mance and reducing costs. Most recently, rapid cost reductions are enabling
PV plants to become economically competitive not only in niche markets
such as off-grid installations, but also for on-grid applications. As a result, PV
power is expanding rapidly in many countries, even though governmental
support policies and incentives (feed-in tariffs) are being reduced. In coun-
tries with good solar resources and high electricity tariffs, residential solar
PV systems have already reached parity with electricity retail prices, whilst
in general PV is now fully competitive with power generated from diesel-
based on- and off-grid systems.The global cumulative installed PV capacity
grew from 1.4 GW in 2000 to about 70 GW at the end of 2011, with around
30 GW added in 2011 alone. Global annual investment totalled some USD 93
billion in 2011 (NPD Solarbuzz, 2012). Italy, Germany, China, the United States,
France and Japan are the leading countries in terms of cumulative capacity,
annual installed capacity and/or production of PV modules and systems.
European countries accounted for 70% of the newly installed capacity in 2011.
Commercial PV technologies include wafer-based crystalline silicon (c-Si)
(either mono-crystalline or multi-crystalline silicon) and thin-films (TF) using
amorphous Si (a-Si/␮c-Si), cadmium-telluride (CdTe) and copper-indium-
[gallium]-[di]selenide-[di]sulphide (CI[G]S). The c-Si systems accounted for
89% of the market in 2011, the rest being TF.

䡵 Performance and Costs – Crystalline silicon (c-Si) cells have reached a record
efficiency of around 25%. The efficiency of the best current commercial mod-
ules is around 19-20% (with a target of 23% by 2020). The majority of com-
mercial c-Si modules, however, have efficiencies in the range of 13-19% with
more than a 25-year lifetime. Commercial TF modules offer lower efficiency
between 6-12% (with a target of 12-16% by 2020). In addition to the commer-
cial options, a number of new PV technologies is under development (e.g.
concentrating PV, organic PV cells, advanced thin films and novel concepts
and materials) and hold out the promise of high performance and low costs
in the medium-term. Today’s PV systems are fully competitive for off-grid
electricity generation and with diesel-based on-grid systems in countries with

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 3
 good solar resources. In an increasing number of countries with high cost of
electricity and good solar resources, small PV systems are also achieving the
so-called grid-parity between the PV electricity cost and the residential retail
prices for householders. For instance, in 2011, electricity prices for house-
holders in the EU-27 ranged between USD 83-291/MWh, excluding taxes
(Eurostat), while the average cost of PV electricity for large ground-mounted
systems ranged from USD 160- 270/MWh in southern and northern Europe,
respectively. Furthermore, advantages of PV electricity are that it is usually
produced close to the consumption site and can match peak demand profiles.
Owing to their low capacity factors, PV systems are not yet cost competitive
for base-load electric power generation. However, PV capital costs are declin-
ing very rapidly due to technology learning, increasing industrial production
and improved efficiency. With a learning rate between of between 18-22%
for each doubling of installed capacity, PV module prices have dramatically
dropped over the past two decades. A 60% reduction has been achieved over
the last two years and more than a 40% reduction is likely to occur by 2020.
In September 2012, Chinese c-Si module prices had fallen to an average of
USD 0.75/W, although the rate of decline has slowed. In Germany, the costs of
installed rooftop PV system had fallen to USD 2.2/W by mid-2012. PV system
costs have been declining so rapidly that past projections of cost reductions
have become obsolete in a very short time. The latest industry projections
(EPIA, 2012) suggest a slowdown in cost reductions of PV systems, with resi-
dential rooftop systems in the most competitive European markets falling to
between USD 1.8-2.4/W by 2020. The recent cost reductions for c-Si modules
have squeezed TF technologies increasingly into niche markets while novel
PV concepts are expected to reach the market in the medium- to long-term.

䡵 Potential and Barriers – PV power has a virtually unlimited energy poten-

tial with no environmental constraints to market expansion. A main issue is
the limited capacity factors that translate into higher electricity costs than
most base-load electricity generation technologies. To drive cost reductions
through deployment, many governments offer financial incentives (such as
feed-in tariffs and tax incentives). The variable nature of the solar source also
means that appropriate grid management and technology (i.e. smart grids)
and energy storage are required for high PV penetration. Where feed-in tariffs
and incentives are in place, PV is attractive to investors and consumers and
contributes significantly to energy production and the mitigation of green-
house gas emissions. PV technology is perceived as a sustainable business
with jobs creation opportunities, and is usually the least-cost solution when
integrated with storage and/or backup generation to provide electricity to
remote areas. Scenarios analysed by the European PV Industry Association
(EPIA, 2012) project that PV power could reach between 4.9-9.1% of the

4 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

 global electricity generation by 2030 (depending on PV growth and electric-
ity demand) and up to 17-21% of electricity share by 2050. The World Energy
Outlook of the International Energy Agency (IEA, 2012) presents a more
conservative outlook that projects PV power to provide between 2-3.3% of
the global electricity by 2030 assuming continued policy support and cost
reductions. It is noteworthy that since 2001 the global PV market has grown
faster than the most optimistic projections.

Basic Process and Technology Status

Based on the photovoltaic (PV) effect1, PV solar cells can convert solar energy di-
rectly into electricity. PV electricity was discovered in the 19th century, but the first
modern PV cells for electricity generation based on silicon (Si) semi-conductors
were developed only in the 1950s. The large-scale commercialisation of PV devices
started only after 2000, following financial incentives in many countries that are
part of government policies to mitigate CO2 emissions and improve energy secu-
rity. PV electricity is environmentally friendly2 and has virtually unlimited potential.
Currently, PV power provides only a small percentage of global electricity supply,
but the market is expanding rapidly, driven by financial incentives and rapid cost
reductions. Over the past decade, the global cumulative installed PV capacity has
grown from 1.4 GW in 2000 to around 70 GW at the end of 2011 (Figure 1), with
some 30 GW of new capacity installed in 2011 alone and annual revenues that
are estimated at around USD 93 billion3. In terms of newly installed capacity, the
leading countries in 2011 were Italy (9.3 GW), Germany (7.5 GW), China (2.2 GW),
the United States (1.9 GW) and Japan (1.3 GW). In terms of total installed capacity,
Europe has around three-quarters of the global total with the leading countries

1 In the PV effect, two different (or differently doped) semi-conducting materials

(e.g. silicon, germanium), in close contact each other, generate an electrical current
when exposed to sunlight. Sunlight provides electrons with the energy to move
cross the junction between the two materials more easily in one direction than in
the other. This gives one side of the junction a negative charge with respect to the
other side (i.e. p-n junction) and generates a voltage and a direct current (DC). PV
cells work with direct and diffused light and generate electricity even during cloudy
days, though with reduced production and conversion efficiency.
2 No GHG emissions during operations and negligible environmental impact.
3 All investment and cost figures are expressed in 2010 USD or converted from EUR
into USD using the conventional exchange rate of 1E=1.3 USD.

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 5
 being Germany, Italy and Spain. Outside of Europe, China, the United States, Ja-
pan, Australia, Canada and India constitute the largest markets. China, Germany,
the US and Japan are the leading producers of PV components and systems.

PV power can be used for either grid-connected applications (e.g. residential,

commercial, utility systems) or off-grid installations. In fact, more than 90% of the
installed capacity consists of grid-connected systems. Primary applications are
systems for residential and commercial buildings, with unit sizes of up to 10 kWp
and 100 kWp, respectively, followed by utility systems (size greater than 1 MWp),
and off-grid applications (e.g. telecommunication towers, rural supply, consumer
goods). The current share of the residential sector is 60%. There are currently no
material availability or industrial constraints to the growth of the share PV power
in the global energy mix. The PV Industry has quickly increased its production
capacity to meet growing demand. At present, supply exceeds demand by a
large margin. Although there are over 1,000 companies that currently produce PV
products worldwide (e.g. PV cells, modules and systems), 90% of the basic mate-
rial (polysilicon) is produced by only a few companies, mostly in the United States,
Japan, Europe and China.

Figure 1 – Global installed PV capacity (MW) – Source EPIA, 2012

6 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

PV Technologies and Performance
The basic element of a PV system is the PV solar cell that converts solar energy
into direct-current (DC) electricity. PV cells are assembled and electrically inter-
connected to form PV modules. Several PV modules are connected in a series and/
or in parallel to increase voltage and/or current, respectively. An inverter is needed
to convert DC into AC for grid integration and use with most electrical appliances.
Modules and balance of system (i.e. inverter, racking, power control, cabling and
batteries, if any) form a modular PV system with a capacity ranging from a few
kW to virtually hundreds of MW. PV systems can be integrated into building-
structures (i.e. building-adaptive or integrated PV systems, BAPV or BIPV), placed
on roofs or ground-based. A number of PV technologies are either commercially
available or under development. They can be grouped into three categories that
are also referred to as 1st, 2nd and 3rd generation: 1) wafer-based crystalline silicon
(c-Si); 2) thin-films (TF); and 3) emerging and novel PV technologies, including
concentrating PV, organic PV, advanced thin films and other novel concepts. Over
the past two decades, PV technologies have dramatically improved their perfor-
mance (i.e. efficiency4, lifetime, energy pay-back time5) and reduced their costs,
and this trend is expected to continue in the future. Research aims to increase ef-
ficiency and lifetime, as well as reduce the investment costs so as to minimise the
electricity generation cost. Several studies have analysed the development of PV
performance and costs over time. A broad overview of current PV technologies
and their performance is provided in Table 1. Current commercial technologies
include wafer-based crystalline silicon (c-Si) and thin-films (TF). The c-Si technol-
ogy includes three variants: mono crystalline silicon (mono-c-Si), multi-crystalline
silicon (multi-c-Si) and ribbon-sheet grown silicon. The TF technology currently
includes four basic variants: 1) amorphous silicon (a-Si); 2) amorphous and mi-
cromorph silicon multi-junctions (a-Si/␮c-Si); 3) Cadmium-Telluride (CdTe); and
4) copper-indium-[gallium]-[di]selenide-[di]sulphide (CI[G]S). It is worth noting
that module efficiencies are lower than commercial cell efficiencies and commer-
cial cell efficiencies are lower than the best efficiency performance of cells in labo-
ratories. Besides current commercial technologies, Table 1 includes 3rd generation
PV technologies that promise significant advances in terms of performance and
costs. Crystalline silicon (c-Si) accounted for 89% of the global market in 2011. The
TF technology currently accounts for the remaining 11%. Among TF, the market is

4 Efficiency is the ratio of the electrical power to the incident solar power.
5 The time needed for the PV system to produce the energy needed for its manu-
facture, i.e. approximately 1-3 years, depending on location and materials, against
a lifetime of at least 25 years. PV systems are durable as they have no moving or
rotating components.

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 7
 Table 1 – Performance of Commercial PV Technologies (Data from EPIA, 2011)

Cell Module Record Area/kW Life-

effic. effic. commercial and time
(lab) efficiency,
(%) (%) (%) (m2/KW)a) (yr)
c-Si
Mono-c-Si 16 - 22 13 - 19 22 (24.7) 7 25 (30)
Multi-c-Si 14 -18 11 - 15 20.3 8 25 (30)
TF
a-Si 4–8 7.1 (10.4) 15 25
a-Si/␮c-Si 7–9 10 (13.2) 12 25
CdTe 10 - 11 11.2 (16.5) 10 25
CI(G)S 7 - 12 12.1 (20.3) 10 25
Org.Dyes 2–4 4 (6-12) 10 (15) na
CPV na 20 - 25 >40 na na
a) A module efficiency of 10% corresponds to about 100 W/m2

shifting to CdTe and CIGS. Among emerging technologies, concentrating PV and

organic solar cells are just entering the market and are expected to capture some
percentage points by 2020. Table 2 provides the evolution of performance over
time for commercial PV modules. Individual PV technologies are discussed in the
following sections.

䡵 Wafer-based Crystalline Silicon Technology – The manufacturing process

of c-Si modules includes: 1) purification of metallurgical silicon to solar grade
polysilicon; 2) melting of polysilicon to form ingots and slicing these ingots
into wafers6; 3) wafer transformation into cells (typically 15x15 cm, 3-4.5 W)
by creating p-n junctions, metal (silver) contacts and back-coating (metal-
lisation); and 4) cell assembly, connection and encapsulation into modules
with protective materials (e.g. transparent glass, thin polymers) and frames
to increase module strength. Silicon is used in the three forms of single-crystal
(sc-Si), block crystals (multi-crystalline silicon, mc-Si) and ribbon-sheet
grown c-Si. Unlike sc-Si cells offer high efficiency (Tables 2 and 3), mc-Si cells

6 Slicing the wafer by wire saw produces up to 40% silicon wastage. This can be
reduced by using a laser cutter and ribbon/sheet-grown c-Si.

8 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

 Table 2 – Performance and Targets for PV Technology (Data from EU SRA
2008, IEA 2010a, EPIA 2011)

1980 2007 2010 2015-20 2030+

Module effic., %
Mono-c-Si ≤8 13-18 13-19 16-23 25-40
Mult-c-Si 11-15 19 21
TF na 4-11 4-12 8-16 na
c-Si material use, g/Wp 7 3 <3
c-Si wafer thick, mm 180-200 <100 na
Lifetime, yr na 20-25 25-30 30-35 35-40
En. payback, yr >10 3 1-2 1-0.5 0.5

have lower efficiency owing to their random atomic structure which affects
the flow of electrons. However, they are less expensive than sc-Si cells.. The
efficiency of mc-Si cells is lower owing to their random atomic structure which
affects the flow of electrons. However, they are less expensive than sc-Si cells.
A standard c-Si module is typically made up of 60-72 cells, has a nominal
power of 120-300 Wp and a surface of 1.4-1.7 m2 (up to 2.5 m2 maximum).
Factory production capacities of 500-1000 MWp per year are currently com-
mon to achieve economies of scale and reduce manufacturing costs. Special
processes for high-efficiency commercial cells include Buried Contacts (by
laser-cut grooves); Back Contacts (that currently achieve the highest com-
mercial efficiency of 22%); specialised surface texturing to improve sunlight
absorption (Suntech, 19%), and HIT (heterojunction with an intrinsic thin
layer, consisting of a sc-Si wafer between ultra-thin a-Si layers to improve ef-
ficiency - Sanyo Electrics, 19.8%). The record cell efficiency for simple c-Si cells
(24.7%) belongs to SunPower. Higher efficiencies have been achieved using
materials other than silicon and multi-junction cells by Sharp, (35.8%, without
concentration) and Boeing Spectrolab (41.6%, with 364 times concentration).

A standard c-Si module is typically made up of 60-72 cells, has a nominal

power of 120-300 Wp and a surface of 1.4-1.7 m2 (up to 2.5 m2 max). Factory
production capacities of 500-1,000 MWp per year are currently common to
achieve economies of scale and reduce manufacturing costs. Special process-
es for high-efficiency commercial cells include Buried Contacts (by laser-cut
grooves); Back Contacts (that currently achieve the highest commercial ef-
ficiency of 22%); Pluto process to improve sunlight absorption (Suntech, 19%),
and HIT (hetero-junction with an intrinsic thin layer consisting of a sc-Si wafer
between ultra-thin a-Si layers to improve efficiency—Sanyo Electrics, 19.8%).

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 9
 The record cell efficiency for simple c-Si cells (24.7%) belongs to SunPower.
Higher efficiencies have been achieved using materials other than silicon and
multi-junction cells by Sharp, (35.8%, without concentration) and Boeing
Spectrolab (41.6%, with 364X concentration).

The main manufacturing challenge for c-Si cells is to improve efficiency and
reduce costs through learning-by-doing and reducing material use. High sili-
con prices in 2006 spurred a 30% reduction of in the amount of silicon in the
manufacturing process to just 5-10 g/Wp today. This has been made possible
by the use of thinner wafers, process automation and waste recycling. The
target is to reach the level of 3 g/Wp or less between 2030 and 2050. Today’s
wafers have a typical thickness of 180-200 ␮m. Cell interconnection and as-
sembly using glass, polymer and aluminium structures, and techniques (e.g.
metallisation, back-contacts and encapsulation) are continuously improved
to reduce costs and enhance performance. The reduction or substitution of
high-cost materials used in the manufacturing process (e.g. silver, currently
80-90 mg/Wp) also is a key objective. As far as efficiency is concerned, the
maximum theoretical efficiency for c-Si is currently estimated at around 29%.
Record cell efficiencies have been obtained using expensive laboratory pro-
cesses (e.g. clean rooms, vacuum technologies), but only a few commercial
cells have efficiencies above 20%. Current commercial sc-Si modules efficien-
cies, which are lower than for cells, range between 13-19%. They could reach
23% by 2020 and up to 25% in the longer term. The majority of commercial
modules, however, are based on multi-crystalline silicon and low-cost manu-
facturing (screen-printing) and offer efficiencies between 12-15% (17% in the
best cases), with prospects for reaching a 21% target in the long term.

䡵 Thin-film Technologies – The TF technology is based on the deposition of

a thin (␮m) layer of active materials on large-area (m2-sized or long foils)
substrates of materials such as steel, glass or plastic. TF technologies use
small amounts of active materials and can be manufactured at a lower full
cost than c-Si. They have short energy pay-back times (i.e. <1yr in southern
Europe) despite their lower efficiency, good stability and lifetimes compa-
rable to c-Si modules. Plastic TF are usually frameless and flexible and can
easily adapt to different surfaces. Standard TF modules have a typical 60-120
Wp capacity and a size between 0.6-1.0 m2 for CIGS and CdTe, and 1.4-5.7 m2
for silicon-based TF. In comparison with c-Si modules, TF module efficiency
(i.e. 4-12%) is significantly lower. The operational experience is also lower. A
typical TF manufacturing process includes: 1) coating of the substrate with
a transparent conducting layer (TCO); 2) deposition of the active layer by
various techniques (e.g. chemical/physical vapour deposition); 3) back-side
metallisation (contacts) using laser scribing or traditional screen-printing;

10 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

 and 4) encapsulation in a glass-polymer casing. Roll-to-roll (R2R) techniques
are often used with flexible substrates to reduce production time and costs.

Research efforts focus on materials with higher absorption and efficiency, thin
polymer substrates, high-stability TCO, deposition techniques (e.g. plasma-
enhanced chemical vapour deposition, PECVD), hetero-structures, electrical
interconnection, low-cost manufacturing (i.e. R2R coating, sputtering, cheap
and durable packaging), quality control and aging tests. In a few years, the
typical manufacturing plant-scale has increased from less than 50 MW to
hundreds of MW per year. However, the TF manufacturing industry is under-
going significant changes at the moment and the outlook is quite uncertain
because TF’s share in the PV market is being challenged by the current low
costs of c-Si modules. Four types of commercial TF modules are described
below, with typical efficiencies as given in Table 3.

● Amorphous Silicon (a-Si) films consist typically of 1␮m-thick amorphous

silicon (good light absorption, but low electron flow) deposited on very large
substrates (5-6 m2), with low manufacturing costs but also low efficiency

Table 3 – Performance and Targets for TF Technologies

(Data from IEA 2010a, EPIA 2011)

a-Si 2010 2015-2020 2030-

Max. effic., % 9.5-10 15 Na
Commercial effic., % 4-8 10-11 13
a-Si/␮c-Si
Max. effic., % 12-13 15-17 Na
Commercial effic., % 7-11 12-13 15
Cd-Te
Max. effic., % 16.5 na Na
Commercial effic., % 10-11 14 15
CI(G)S
Max. effic., % 20 na Na
Commercial effic., % 7-12 15 18
Key R&D targets: Optimise CVD and plasma deposition process; new roll-to-roll processes; low-cost
packaging; new materials (i.e. ␮c-SiGe, SiC, nano-diamond, cheaper TCO and substrates); replace/re-
cycle scarce materials; better understanding of the physics of advanced concepts (e.g. multi-junctions,
doping, quantum dots, up/down converters, photonic crystals)

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 11
 (4-8%). The best laboratory efficiencies are currently in the range of 9.5-10%.
Among TF technologies, a-Si TF is perhaps the most challenged by the cur-
rent low-cost c-Si. Its future is rather uncertain. Some producers have recently
retired part of manufacturing capacity.

● Multi-junction Silicon (a-Si/␮-Si) films offer higher efficiency than a-Si

films. The basic material is combined with other active layers, e.g. microcrys-
talline silicon (␮c-Si) and silicon-germanium (␮c-SiGe), to form a-Si/␮c-Si
tandem cells, micro-morph and hybrid cells (even triple junction cells) that
absorb light in a wider range of frequency. An a-Si film with an additional 3␮m
layer of ␮c-Si absorbs more light in red and near-infrared spectrum, and may
reach an of efficiency up to 10%. Best laboratory efficiencies are currently in
the range of 12-13% for a-Si/␮c-Si tandem cells and triple junction SiGe cells.
Commercial module efficiencies are between 6.5-9%, but prototype module
efficiencies of up to 11% have been demonstrated for best multi-junctions
(Bailat, 2010). Short-term targets (Table 4) include the demonstration of cell
efficiencies of 15% (17% by 2020) and module efficiency of 12%. Research has
also been exploring further material options such as sc-Si (hetero-junctions,
HIT), SiC, nanocrystalline-diamond, layers with quantum dots, and spectrum
converters, improved TCO and substrates, and alternative low-cost deposition
techniques (e.g. using no plasma)

● Cadmium-telluride (CdTe) films are chemically stable and offer relatively

high module efficiencies (i.e. up to 11%). They are easily manufactured at
low costs via a variety of deposition techniques. The efficiency depends sig-
nificantly on deposition temperatures, growth techniques and the substrate
material. The highest efficiencies (i.e. up to 16.5%) have been obtained from
high temperature (600°C) deposition on alkali-free glass. The theoretical
efficiency limit is around 25%. Approaches to increase the efficiency include
inter-mixing elements, hetero-junctions, activation / annealing treatments
and improved electrical back contacts. In most efficient CdTe films, the sub-
strate faces the sun. In such a configuration, TCO properties are crucial for
module efficiency. Thinner CdTe layers are also key to minimising the use of
tellurium, given that its long-term availability may be a concern.

● Copper-indium-[gallium]-[di]selenide-[di]sulphide film (CI[G]S) has the

highest efficiency among TF technologies (i.e. 20.1% lab efficiency; 13-14% for
prototype modules and 7-12% for commercial modules). However, the manu-
facturing process is more complex and costly than the other TF technologies.
Replacing indium with lower-cost materials or reducing indium use could help
reduce costs (indium is used in liquid crystal displays as well). Cost reduc-
tion and module efficiencies of up to 15% can be achieved using better basic

12 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

 Table 4 – Performance and Targets for Emerging PV Options
(data from IEA 2010a, EPIA 2011)

2010 2015-2020 2030-

CPV
Effic.(lab-effic.),% 20-25 (40) 36 (45) >45
lifetime; optical efficiency (85%), sun-track-
Major R&D areas and targets
ing, high concentration, up-scaling;
Inorganic TF (spheral cells, poly-c Si cells)
Effic.(lab-effic.),% (10.5) 12-14 (15) 16-18
deposition, interconnection, ultra-thin films;
Major R&D areas and targets
up-scaling, light tailoring
Organic cells (OPV, DSSC)
Effic.(lab-effic.),% 4 (6-12) 10 (15) na
Major R&D areas and targets Lifetime (>15 yr), industrial up-scaling
Novel active layers
Effic.(lab-effic.),% Na (>25) 40
Materials, deposition techniques, understand-
Major R&D areas and targets ing quantum effects, up-scaling from lab
production
Up/down converters
Module effic., % +10% over ref. Material
Major R&D areas and targets (nano) materials, physical stability, up-scaling

processes (e.g. interface and grain boundary chemistry, thin-film growth on

substrates), novel materials (e.g. new chalcopirytes, wide band-gap materials
for tandem cells), material band-gap engineering (e.g. spectrum conversion,
quantum effects), non-vacuum deposition techniques, electro-deposition,
nano-particle printing and low-cost substrates and packaging.

䡵 Emerging and Novel PV Technologies – A number of emerging and novel PV

technologies are under investigation with a potential for higher efficiency and
lower cost than c-Si and thin films. They include concentrating PV (CPV), or-
ganic solar cells, advanced inorganic thin-films, thermo-photovoltaics (TPV)
and novel concepts that aim at either tailoring the active layer for better
matching to the solar spectrum or modifying the solar spectrum to improve
the energy capture. Typical efficiencies and R&D targets are provided in Table
4. Some of these technologies are beginning to emerge in the marketplace for

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 13
 niche applications. The feasibility of other options depends on breakthroughs
in material science, nano-technology, plastic electronics and photonics.

● Concentrating Photovoltaics (CPV) is the most mature emerging technol-

ogy. In CPV systems, optical sun-tracking concentrators (i.e. lens) focus the
direct sunlight on highly efficient solar cells. This high efficiency reduces the
need for costly active materials and helps offset to some extent the additional
cost of the concentration system. The CPV technology is currently moving
from pilot and demonstration plants to commercial applications, but further
R&D is needed, particularly to reduce costs. A variety of options for cell ma-
terials and concentrators (with concentration factors from 2-100 and even up
to 1,000 suns) is being tested.

In general, c-Si modules with efficiencies of 20-25% are used with low-me-
dium sunlight concentration while III-V semi-conductors and multi-junction
solar cells (e.g. triple junction GaInP/GaInAs/Ge obtained from metal-
organic CVD) are used for high concentrations (> 250). These high-quality
cells can reach lab efficiencies above 40% (and even higher, when adding
further junctions). CPV research efforts focus on low-cost. multi-junction
cells with efficiency of around 35% and even high-cost, ultra-efficient cells.
Concentration systems include lenses, reflection and refraction systems.
High concentration factors require high accuracy in optical and sun-tracking
systems (0.1 degree) and heat dissipation. Unlike other PV technologies,
CPV uses only the direct sunlight component and will make the most sense
in Sun Belt regions.

● Organic Solar Cells are based on active, organic layers that are also suitable
for liquid processing. This technology is based on the use of very low-cost
materials and manufacturing processes, with low energy input and easy up-
scaling. It might be feasible to achieve costs below USD 0.5/Wp. Major chal-
lenges relate to the low efficiency and stability over time. Organic cells include
hybrid dye-sensitised solar cells (DSSC) - which retain inorganic elements
- and fully-organic cells (OPV). While in 2009, DSSC production amounted
to 30 MW, production in 2012 is estimated on the order of few hundred MW.
Lab efficiency is in the range of 8-12%, while commercial applications still have
efficiency of around 4%. OPV production totalled 5 MW in 2009, with cell ef-
ficiencies of 6% for very small areas and below 4% for larger areas. Both tech-
nologies use R2R techniques and standard printing to reduce manufacturing
costs to USD 0.6-0.7/W in a few years, which means that they still will not
compete with c-Si. The demonstration of lab cell efficiencies of 10% (15% by
2015) and a lifetime of 15 years is needed to confirm feasibility. This involves a
thorough understanding of the basic physics and synergies with organic LED

14 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

 and organic electronics. OPV cells are currently used for niche applications
and their competitiveness has yet to be proven.

● Advanced Inorganic Thin Films include evolutionary TF concepts, such as

the spheral CIS approach (i.e. glass beads covered by a thin multi-ycrystalline
layer with a special interconnection between spheral cells) and the multi-
crystalline silicon thin films obtained from the high-temperature (> 600 °C)
deposition process, which promises lab efficiencies of up to 15% (10.5%
achieved by CSG Solar).

● Other Novel PV Concepts are in a very early stage and their technical feasibil-
ity has yet to be proved. They rely on nanotechnology and quantum effects
to provide high-efficiency solar cells that either match the solar spectrum
using novel and tailored active materials or modify it to increase the energy
absorption of current active materials. In the first approach, quantum effects7
and nano-materials enable a more favourable trade-off between output
current and voltage of the solar cell8. R&D efforts aim to demonstrate cell ef-
ficiency above 25% by 2015 and to characterise nano-materials and cells with
a theoretical efficiency limit as high as 60%. The second approach relies on
up/down-converters9 to tailor the solar radiation and maximise the energy
capture in existing solar cells. Photon absorption and re-emission may shift
the sunlight wave-length and increase the energy capture (i.e. plasmonic
excitation). The target is a 10% increase in the efficiency of existing c-Si cells
and TF. However, a full understanding of these processes will take some years.

䡵 Balance of System (BoS) – The balance of the system (BoS) includes com-
ponents other than the PV modules (e.g. the inverter to convert DC into AC,
the power control systems, cabling, racking and energy storage devices, if
any). The BoS consists of rather mature technologies and components, but
in recent years the BoS cost has declined in line with the price of PV module
in most competitive PV markets. However, it remains to be seen whether this
trend can be maintained in the future. Apart from reducing costs, the main
targets for the inverter are improved lifetime and reliability, and control of

7 Note that quantum wells or quantum dots consist of low-band gap semi-conduc-
tors within a host semi-conductor with a wider band gap to enable increased cur-
rent with high output voltage).
8 In PV cells, current and voltage depend on band gap with opposite trends.
9 Down-converters convert high-energy (e.g. violet) photons into two lower-energy
(near infrared) photons for more efficient sunlight absorption. Up-converters con-
vert two low-energy photons into a higher energy photon.

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 15
 reactive power in grid-connected systems (this may help grid integration).
Inverters are available with capacities from a few kW to as much as 2 MW for
use in large-scale systems. Either single or numerous inverters can be used
for a single PV system, depending on design requirements.

The BoS can also include electricity storage. As far as storage is concerned, in
addition to lead-acid batteries and traditional pumped hydro storage systems
(suitable for large-scale only), a number of new energy storage devices are
being developed. These include new batteries technologies, electric capaci-
tors, compressed air systems, superconducting magnets and flywheels. Apart
from pumped hydro, none of these technologies is currently mature and cost-
effective for large-scale commercialisation. Cost-effective electricity storage
could significantly boost the market penetration of PV power by helping to
manage the variability of the solar energy. For batteries, current R&D efforts
focus primarily on performance, lifetime and cost of electrical batteries (i.e.
Ni-MH and Li-ion batteries), but a number of other options are also under
consideration. In particular, NaS batteries could represent a competitive long-
term, large-scale solution. Off-grid PV systems must also be equipped with
back-up power (e.g. diesel generators, biomass-fired generators, wind power)
to supply energy when sunlight is not available.

Current Costs and Cost Projections

At present, providing up to date PV prices is a challenging task as the market
is evolving on a monthly basis due to increased competition among suppliers,
changing policy incentives in many countries, continuous material and technol-
ogy innovation, growing economies of scale and dramatic cost reductions. PV
power is currently economically competitive for off-grid applications, but recent
PV cost reductions mean that residential and commercial grid-connected systems
have significantly increased their economic attractiveness in the most favourable
geographical locations, even without policy incentives and are likely to become
competitive in the near future. The financial incentives that many governments
- notably those in developed countries – have been offering to promote PV instal-
lations as a part of policies to combat climate change have helped significantly to
spur PV deployment and reduce costs by mass production of components and
systems. The so-called grid parity (i.e. the parity between the PV generation cost

16 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

for residential and commercial systems and the electricity retail price for house-
holders) has been achieved or is nearing that goal in the most favourable loca-
tions. For instance, in 2011, electricity prices for householders in the EU-27 ranged
between USD 83-291/MWh, excluding taxes (Eurostat) while the average cost of
PV electricity for large ground-mounted systems ranged from USD 160-270/MWh
in southern and northern Europe, respectively.

From the utility perspective, in the absence of incentives, the PV generation cost
is not yet competitive with the generation cost of conventional base-load power
technologies (except in some countries with excellent solar resources and high
fossil fuel prices) because of the relatively high investment cost and the limited
capacity factor10 of PV plants. However, this simple comparison does not take into
account the fact that PV systems generally produce during daily peak-demand
hours when the marginal cost of electricity is higher. Following this trend, produc-
ers of PV systems envisage that large-scale utility systems (i.e. the most competi-
tive PV installations in terms of investment and electricity costs) will lead to the
reduction of the levelised cost of electricity (LCOE) from PV systems by between
USD 90-200/MWh by 2020 in southern and northern Europe, and USD 50-70/
MWh by 2030 in Sun Belt countries (Note that, by definition, the LCOE is the cost
per unit of electricity required to cover all investment and operational costs over
the system’s lifetime without profits). These projections account for the annual
solar irradiance variability (e.g. 1,000 kWh/m2 in Scandinavia, 1,900 kWh/m2 in
southern Europe and 2,200 kWh/m2 in the Middle East). Residential PV prices will
also decline sharply but will remain more expensive than large ground-mounted
systems. The PV costs have to be compared with the rising costs of gas- and coal-
fired power, taking into account that in many countries governments still subsidise
conventional power and fossil fuels.

The investment cost of PV systems is rapidly declining. Over the past three dec-
ades, the PV industry has been reducing the price of PV modules by between
18-22% with each doubling of the cumulative installed capacity (EPIA 2011). More
recently, prices have dropped even faster as increased competition and a supply
surplus have pushed down PV module prices. Further reductions of 40-60% by
2020 are feasible. The increased efficiency of PV modules is an important compo-
nent of this cost reduction.

10 The capacity factor of a power plant is defined as the ratio of the actual electricity
produced per year to the electricity that could in theory be produced based on the
nominal peak power and technical availability of the PV system.

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 17
 The current costs of small PV systems in Germany fell to just USD 2,200/kW in the
second quarter of 2012 from an average of USD 3,800/kW in 2010 (IRENA, 2012).
However, not all PV markets are as competitive as Germany’s. In some countries,
small-scale systems (i.e. <10kW) may cost twice as much as in Germany (Seel,
2012). The EPIA forecasts that small-scale rooftop PV system costs in the most
competitive markets could decline to between USD 1,750-2,400/kW by 2020.
Large, utility-scale PV projects could see their average costs decline to between
USD 1,300-1,900/kW by 2020 (EPIA, 2012).

䡵 Cost of PV Modules – Due to significant overcapacity, current prices for

wafer-based c-Si modules fell to around USD 800/kW in September 2012. In
addition to overcapacity, the 60% price decline in just two years was due to
the reduced use of silicon, higher efficiency (i.e. 5-7% cost reduction per 1% ef-
ficiency increase) and industrial learning driven by deployment policies. Thin
film prices are slightly lower than c-Si module prices, but the projected TF
growth in market share, due to lower cost structures than for c-Si, has yet to
materialise in today’s highly competitive market. In the long-term, the differ-
ences between TF technologies are projected to converge, but TF’s future is
uncertain in the current climate and will depend heavily on technology inno-
vation. CdTe modules with efficiency of 11% can compete economically with
the cheapest c-Si modules. They are expected to increase efficiency by up to
15% while cost reductions should keep them competitive with c-Si modules.
Important steps towards this target include a full understanding of CdTe’s
basic properties and the use of lower temperature deposition processes. A
better understanding of the basic physics can also reduce the cost of CI(G)
S modules by introducing novel materials, concepts and manufacturing (e.g.
new chalcopirytes, polymers, metal substrates, quantum effects, spectrum
conversion, electro-deposition and nano-particle printing).

䡵 Cost Breakdown of PV Systems – The typical cost of a c-Si module includes

about 45-50% for silicon, 25-30% for cell manufacturing and 20-25% for cell
assembling into modules. The cost breakdown for a commercial PV system
includes 50-60% for PV modules (TF and c-Si, respectively), 10% for the
inverter, 23-32% for installation of BoS and about 7% for engineering and
procurement (EPIA, 2011). In the past five years, PV modules’ share has de-
clined from about 60-75% to 40-60%, depending on the technology. As a
consequence —and important for PV competitiveness—is the cost reduction
of inverter and BoS prices. Markets like Germany have seen these costs de-
cline in line with module costs, but others have stickier soft-costs, particularly
for residential installations, and BoS costs have not come down as rapidly. In
Europe, BoS prices have fallen to as little as USD 1,300/kWp for residential
roof installations but tend to be lower for ground-based, utility-scale systems.

18 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

Potential and Barriers
PV technology is well-positioned to exploit the huge solar energy potential. Esti-
mates suggest that, in principle, if about 4% of the world’s deserts were covered
by PV modules, this could meet the global primary energy demand, and that
0.34% of the European landmass if covered with solar PV modules would meet
the entire European electricity demand (EPIA). Based on solar irradiance and
electricity consumption, and assuming a 14% PV efficiency, the surface needed to
meet the annual electricity demand of a typical house ranges from 14 m2 in Rome
(i.e. 2,700 kWh/yr/household of electricity) to 33 m2 in Copenhagen (i.e. 4,400
kWh/yr/household) up to 45 m2 in New York (i.e. 11,000 kWh/yr/household). This
supply would, in principle, meet the average annual electricity demand, taking into
account excess production in spring and summer, while additional power would be
needed from the grid in most regions during the winter.

There are no technical constraints (e.g. material availability, energy payback time)
to the full expansion of the PV market. The main issues remain the relatively high
electricity costs (although grid-parity for residential systems will soon be the
norm rather than the exception) and the need for advanced grid management
and energy storage (or back-up power in off-grid installations) to deal with the
intermittent nature of PV power. Recent studies suggest that more than 20% of
intermittent power could be integrated into existing large grids without significant
difficulty while a more sizeable integration would require re-thinking grid man-
agement, larger interconnections, mid-load power plants and energy storage. At
present, many countries are preparing for new grid regulations and technologies11
to accommodate an increasing amount of renewable, non-dispatchable electric-
ity. More challenging is the development of cost-effective energy storage, a key
component when the intermittent electricity reaches a certain share, depending
on grid size and inter-connection.

As a part of the policies to promote local socio-economic development, combat

climate change and increase energy security a good number of countries have
financial incentives in place (e.g. investment subsidies, fiscal allowances, portfolio
standards and feed-in tariffs) aimed at promoting the deployment of PV systems
and providing a level playing field for PV power. Feed-in tariffs (FiT) is the most
common and effective mechanism for PV promotion when accurately set. Under
the FiT mechanism, PV electricity producers are granted the right to feed elec-

11 Smart grids, e.g. with bi-directional flow and metering between grids and users/
producers, and super grids, with international interconnection to complement e.g.
wind power from windy countries with solar power from sunny countries.

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 19
 tricity into the grid and receive a premium tariff per kWh over a fixed period of
time. This compensates for the high investment cost of PV systems and reflects
the social and environmental benefits of solar power. The FiT cost is typically
covered by utilities and recovered from electricity consumers. One advantage of a
FiT mechanism is that it rewards electricity generation rather than capacity addi-
tions. To ensure continued pressure on investment cost reduction, well-designed
FiT should plan for gradual decrease over time, based on the PV cost decline (i.e.
the “corridor” mechanism). In line with this principle, many European countries
(e.g. Germany, Italy, Spain) are significantly reducing their FiT tariffs. In the United
States, FiT, energy credits, loans and other mechanisms are in place in some 19
states while, at the federal level, a 30% tax credit is granted to commercial and
residential PV systems. China is currently the largest PV module producer and until
recently had a small national market. With feed-in tariff incentives introduced in
2011, China is fast becoming the largest PV market. Other countries with incentives
include Israel, Turkey, Thailand, South Africa, India, Indonesia, Uganda, Canada and
Australia.

PV power benefits from a good social acceptance. It is perceived as an environ-

mentally benign technology, a sustainable business and labour opportunity, and
an easy and cost-effective way to provide access to electricity in remote rural
areas. While today’s incentive schemes focus mostly on grid-connected PV sys-
tems, special financing mechanisms are needed for off-grid rural electrification in
developing countries. According to New Energy Finance (2009), private venture
capital investment in solar energy, notably PV, more than doubles each year while
public R&D funds increase at a slower pace. In terms of applications, residential
systems will continue to dominate in the coming decade, but commercial and util-
ity systems will increase their share. Since 2009, a number of large PV plants has
been completed and started operation. As of November 2012, about 40 PV power
plants with capacities between 40-75 MW are in operation or about to open all
over the world (i.e. mostly in Germany, Italy, the USA and Spain, but also in Canada,
France, India and the Ukraine).

In terms of market potential, assuming continued policy support and cost reduc-
tions, the World Energy Outlook published by the International Energy Agency
(IEA, 2012) projects that PV power will provide between 2-3.3% of the global elec-
tricity by 2030, depending on the scenario. The most ambitious projections by the
European PV Industry Association (EPIA) and Greenpeace predict that PV power
could reach between 4.9-5.7% and 7.8-9.1% of the global electricity generation
by 2030 (depending on PV growth and electricity demand) and up to 17-21% by
2050 (EPIA, 2012b). The EPIA’s analysis assumes a global average capacity factor
increasing from today’s 12-17% by 2050 with moderate progress in energy storage.

20 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

It should be noted that, since 2001, the global PV market has been growing faster
than even the most optimistic projection.

PV benefits and impacts have been analysed and monetised by the SET For 2020
study for the European Union. The benefit of reducing GHG emissions in Europe
by producing PV electricity has been estimated at €12/MWh, compared with €23/
MWh at the global scale. This is based on a global average GHG emission factor
from electricity production of 0.6 kgCO2/kWh, which includes 12-25 gCO2/kWh
emitted from the PV lifecycle and assumes a CO2 abatement cost of €20/tCO2.
This estimate is actually a conservative one because the cost of CO2 abatement
in fossil fuel power plants is likely to be well above €20/tCO2 in the long term. All
this means that the real external costs of fossil fuel-based power are not included
in the current electricity prices. Other benefits in terms of external costs include
the reduction of grid losses due to distributed generation (on the order of €5/
MWh); the positive impact on energy security (€15-30/MWh, depending on fossil
fuel prices) and on electricity demand peaks (i.e. peak shaving), thus reducing the
need for additional peak capacity (€10/MWh).

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f 21
 References and Further Information
EPIA, 2012 - European Photovoltaic Industry Association - Global Market Outlook May
2012 www.epia.org.
EPIA, 2011 - European Photovoltaic Industry Association and Greenpeace, Solar Gen-
eration 6, 2011 - www.epia.org.
IEA, 2010 - International Energy Agency, July 2010 - Solar Photovoltaic Energy Technol-
ogy Roadmap - www.iea.org.
IRENA, 2012, Solar Photovoltaics – Renewable Energy Technologies, Cost Analysis
Series, IRENA Working Paper, June 2012, www.irena.org.

Other relevant sources include:

Bailat J et al.2010 - High Efficiency Micromorph Tandem Solar Cells - 5th PV World
Conference, Sept.2010, Valencia, Spain.
CREIA, 2009 - Chinese Renewable Energy Industry Association - www.creia.net.
DOE, 2008 - Solar Energy Technologies Programme – Multi-Year Programme Plan
(2008 – 2012).
European Photovoltaic Technology Platform, 2009, Implementation Plan for a Strategic
Research Agenda.
EU SRA, 2008 - European PV Technology Platform - A Strategic Research Agenda for
PV Solar Energy - www.eupvplatform.org.
IEA, 2012 – International Energy Agency - World Energy Outlook 2012, www.iea.org.
IEA, 2012 - International - Energy Technology Perspectives 2012 - Scenarios and Strate-
gies to 2050 www.iea.org.
IEA, 2008 - International Energy Agency 2008 - Energy Technology Perspectives 2008
Scenarios and Strategies to 2050.
IEA PVPS, 2009 - Trends in Photovoltaic Applications – Survey Report of Selected IEA
countries – www.iea-pvps.org.
Kazmerski L L., 2006 - Solar photovoltaics R&D at the tipping point: A technology
overview, Journal of Electron Spectroscopy.
NEDO, 2009 - New Energy and Industrial Technology Development Organisation, PV
roadmap toward 2030+, 2009.
Neij L., 2007 - Cost Development of Energy Technologies – An Analysis based on
Experience Curves.
New Energy Finance, 2009 - Global Trends in Sustainable Energy Investment 2009.
NPD Solarbuzz, 2012 – 2012 Marketbuzz. http://www.solarbuzz.com
Phoenix Solar, A.T. Kearney, 2010 “The True Value of Photovoltaics for Germany”.
Photon Consulting 2012 - The True Cost of Solar Power 2012 – www.photonconsulting.
com.
World Bank, 2010 - World Development Report, http://siteresources.worldbank.org/
INTWDR2010/Resources/.

22 Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

 Table 8 – Summary Table - Key Data and Figures for PV Technologies

Technical performance Typical current international values and ranges

Energy input/output Sunlight/ Electricity
Current PV technologies Crystalline Si Thin Film CPV
sc-Si mc-Si a-Si/m-Si (m-SiGe) CdTe CI(G)S
Max. (record) cell efficiency, % 22 (24.7) 18 (20.3) 10 (13.2) 11.2 (16.5) 12.1(20.3) (>40)
Max. module efficiency, % 19-20 15-16 9 na na na
Commercial modules effic., % 13-19 11-15 7-9 10-11 7-12 20-25
Land use, m2/kW 6-8 7-9 11-15 9-10 9-15 na
Lifetime, yr 25 (30) 25 na
Energy payback time, yr 1-2 1-1.5 na
Material use, g/W 5-7 na na
Wafer thickness, ␮m <180-200 na na
Market share, % ~85 ~15 na
Typical size (capacity), kW Residential < 10 kWp; Commercial < 100 kWp; Industry 100Kwp -1MWp; Utility > 1MWp
Total cumulative capacity 1.4 GW (2001), 23 GW (2009), 40 GW (2010), 70 GW (2011)
Annual installed capacity 2.8 GW (2007), 5.9 GW (2008), 7.2 GW (2009); 15 GW (2010); 30 GW (2011)
Capacity factor, % From 9-16% (in most favourable locations), based on annual electricity production
CO2 emissions, gCO2eq/kWh Occurring during manufacturing only - between 12-25 gCO2eq/kWh
Avoided CO2 emissions ~ 600 gCO2eq/kWh (based on electricity mix in developed countries); up to 900 gCO2eq/
kWh in countries with coal-based power generation.

Sol a r P h ot ov ol t a ic s | T e c h n o lo g y B r ie f
23
24
Costs Typical current international values and ranges (2012 USD, 1EUR = 1.3 USD)
By technology Crystalline
Thin Film CPV
Si
a-Si/␮-Si;
c-Si CdTe CI(G)S
(␮-SiGe)
Module cost, $/kW (2012)1 770- 3100-
880-1140 650-750 770-1000 (1500)
1000(1500) 4400
BoS cost, $/kW (2012) 820-1660 (best practice to global average)
O&M cost Estimated at 1% of the investment cost per year
Typical cost breakdown PV module 50-60% (TF-c-Si); Inverter 10-11%; BoS & Installation 32-23%; E&P 7%
By applications Residential systems Commercial systems Utility systems
System cost, $/kW 2200 – 4500 1900 - 2500 1700 – 2100
Electricity cost2, $/MWh 190-2003 130-1604 100 -1505
Cost projections 2016 Residential systems Commercial systems Utility systems

Sol ar Pho to vo l tai c s | T e c h n ol og y Br i e f

Module cost, $/kW 920 (c-Si) -950 (TF)
BoS cost, $/kW 1200-1600 (global average)
Electricity cost, $/MWh 150 110-130 80-140
Market share projection 2030 2050
Global electricity share, % 17-21%
2-3.3% (IEA,2012); 4.9-5.7% to 7.8-9.1% (EPIA,2012)
(EPIA,2012)
1 Sources: www. Sologico.com – 2012; Photon Consulting 2012; (overall TF module cost)
2 25-year lifetime, 10% interest rate, 1%/yr O&M cost
3 With reference to the US and Japan markets,
4 With reference to Italian and German market
5 With reference to Chinese and US markets
Disclaimer
The designations employed and the presentation of materials herein do
not imply the expression of any opinion whatsoever on the part of the Sec-
retariat of the International Renewable Energy Agency concerning the le-
gal status of any country, territory, city or area or of its authorities, or con-
cerning the delimitation of its frontiers or boundaries. The term “country”
as used in this material also refers, as appropriate, to territories or areas.
The preparation of the paper was led by
Giorgio Simbolotti (ENEA).
Comments are welcome and should be addressed to
Michael Taylor (MTaylor@irena.org),
Giorgio Simbolotti (giorgio.simbolotti@enea.it)
and Giancarlo Tosato (gct@etsap.org)