Solution-Processed

performance for solution-processed PV,

and the second part discusses production
and application aspects in more detail.

Advances in Device Efficiency

The photoactive (light-absorbing) layer

Organic Solar Cells of solution-processed organic solar cells is

typically based on a bicontinuous blend of
two organic semiconductor materials (see
Figure 1), most often a conjugated polymer
Christoph J. Brabec and James R. Durrant blended with a soluble C60 derivative.
Offsets of the highest occupied molecular
orbital (HOMO) and lowest unoccupied
molecular orbital (LUMO) levels of the two
Abstract
materials result in the formation of a bulk
Organic solar cells, based on polymer/fullerene-blend films, are advancing rapidly
heterojunction with the polymer being the
toward commercial viability. In this article, we review recent progress on two issues critical
electron donor and C60 the electron accep-
for technological applications: device photovoltaic efficiencies and processing
tor. Optical excitation of the blend results in
technologies for high-throughput production. In terms of device efficiencies, we consider
photoinduced charge separation across the
advances in low-bandgap polymers, film morphology, and device structure aimed at heterojunction, and subsequent charge
increasing efficiencies beyond 5%. We then review recent progress in developing high- transport through the two phases allows
throughput, solution-printing-based processes for low-cost device fabrication. current output to an external circuit.
The current workhorse materials emp-
loyed in OSCs are poly(3-hexylthiophene)
(P3HT) and the C60 derivative PCBM
Introduction ([6,6]-phenyl-C61-butyric acid methyl
Today, after 50 years of challenges and light absorbers, including polymer-blend, ester), as shown in Figure 1. Several
excitement, the field of photovoltaics (PV) small-molecule, dye-sensitized, and inor- groups have now reported small-area
is becoming a significant success story, ganic nanoparticulate devices, are all device efficiencies in excess of 4% with
with a market value in 2007 of over $10 bil- receiving extensive attention. P3HT/PCBM blends.3–5 This success can
lion and an annual growth rate of 30–40%. In this article, we focus on solution- be attributed to several different properties
Although PV’s contribution to the global processed organic solar cells (OSCs) based of this materials combination. The P3HT
electricity supply still remains less than on polymer/fullerene-blend films (see optical bandgap (~1.9 eV) and strong
0.1%, the scale of the global solar resource Figure 1). Such devices have great potential absorption coefficient enable efficient
(the solar irradiation incident upon Earth for ultralow costs, not so much because of light absorption up to 650 nm for films
in one hour is equivalent to the world’s the costs of the material classes under con- thicknesses of only ~200 nm. The LUMO
annual energy demand) coupled with the sideration, but more because of the produc- level of PCBM is ~1 eV lower than that
rapid growth of the PV market suggests tivity of the printing and coating of P3HT, providing a sufficient driving
that PV could soon be playing a major role techniques compared to wafer-based pro- force for interfacial charge separation. The
in the global energy supply. duction techniques. The only production free energy of the charge-separated state,
The past 50 years have been dominated technology that can achieve production approximated by the difference in energy
by silicon-based PV. Over the past decade, rates of greater than 10,000 m2/h at pro- between the PCBM LUMO and P3HT
thin-film PV technologies, so called second- duction costs as low as even €(s)/m2 is HOMO levels, is sufficient to generate a
generation technologies such as a-Si (amor- web coating or printing (including vacuum reasonable output voltage of over 600 mV.
phous silicon), CIGS (copper indium web coating). Web coating and printing are The charge mobilities of the holes in P3HT
gallium diselenide), and CdTe (cadmium alternative names for roll-to-roll coat- and electrons in PCBM are on the order of
telluride) have rapidly caught up in terms ing and printing. The potential low costs 10−4 cm2/(V s) or higher, thus avoiding sig-
of performance and production technology for a printed/coated PV technology are nificant resistive losses during charge
and are now undergoing a rapid expansion extremely encouraging and are motivating transport to the device electrodes.
in production. The next few years will the academic and industrial research com- Crucially, P3HT and PCBM are amen-
determine whether this indicates the begin- munities to develop this technology to able to systematic optimization of blend
ning of the end of the dominance of the PV market readiness within this decade. morphology. They are reasonably miscible
market by crystalline Si. The status of organic photovoltaics but show strong tendencies to crystallize
The time frame for a PV technology to go (OPV) was last reviewed in this journal in into separate domains on the nanometer
from laboratory concept to market is January 2005.1,2 In this article, we address length scale3–5 and to form concentration
approximately 20–30 years. With a-Si, some of the subsequent exciting new gradients normal to the device electrodes.6
CIGS, and CdTe now becoming established developments in this field, focusing on A range of different strategies have been
in the market, the question arises as to what two key issues: employed to optimize film morphology
photovoltaic technology platform(s) will 䊏 selection of materials for the photoac- and crystallinity, including control of
follow. Research into such platforms is tive layer to achieve further advances in polymer regioregularity, film deposition
currently proceeding in two directions: device performance and conditions, and postdeposition film
ultrahigh efficiencies and ultralow produc- 䊏 selection of a high-volume, high-speed, processing.3–5 These strategies appear to
tion costs. For lowest cost PV production, and high-yield production process. have been successful in achieving film
solution-processable solar cell technologies This article is divided into two parts. morphologies compatible with efficient
based on both molecular and inorganic The first part describes the state-of-the-art charge separation and collection while

670 MRS BULLETIN • VOLUME 33 • JULY 2008 • [Link]/bulletin

Downloaded from [Link] UNIVASF, on 18 Sep 2019 at 21:53:59, subject to the Cambridge Core terms of use, available at [Link]
[Link]
 Solution-Processed Organic Solar Cells

a LUMO level. Strategies to address this lim-

itation have typically targeted the use of
TCO coated PEDOT Aluminium
PCBM polymers with larger ionization potentials.
substrate :PSS Polymer electrode The alternative strategy is to improve
the photocurrent by enhancing light
absorption by the photoactive layer. This
has typically been addressed by lowering
Charge the polymer’s optical bandgap to enhance
separation the absorption of longer wavelength light.
The past few years have seen an intense
program of polymer design, synthesis, and
Charge
Light + transport
evaluation based on these two strategies
- (see, for example, Reference 8). In practice,
it has proved to be remarkably challenging
to improve device performance beyond
Exciton that of P3HT/PCBM blends. We have, for
generation example, recently suggested that achiev-
and diffusion ing efficient dissociation of photogener-
ated charges might be a more significant
challenge than had previously been appre-
ciated, with dissociation typically requir-
ing a relatively large free energy for charge
separation to give initially generated
External current flow polarons sufficient thermal energy to over-
come their Coulombic attraction.9
Nevertheless, an exciting range of new
b materials have been evaluated for OSC
applications, including not only homopoly-
O mers but copolymers and metallated conju-
O gated polymers. The new materials and
S S their evaluation lead not only to enhanced
device performance but also to improved
S
understanding of the underlying science
relating materials properties to device per-
formance. A particularly notable success has
been the low-bandgap benzothiadiazole-
P3HT PCBM fused thiophene copolymer PCPDTBT
(poly[2,6-(4,4-bis-(2-ethylhexyl)-4H-
cyclopenta[2,1-b;3,4-b’]dithiophene)-alt-4,7-
(2,1,3-benzothiadiazole)).10,11 This class of
materials led to a certified performance of
c over 5% as depicted in Figure 2. When
blended with the C70 analogue of PCBM
and used with a chlorobenzene/octane
dithiol cosolvent system for film deposition,
this polymer yielded efficient photocurrent
generation out to 850 nm and a white-light
power conversion efficiency of up to 5.5%.11
In parallel with these studies aimed at
advancing device efficiencies by tuning
polymer HOMO and LUMO levels, other
Figure 1. (a) Schematic layout of the function of a typical organic solar cell. (b) Chemical
groups have been trying a range of inno-
structures of typical donors and acceptors. (c) Photograph of reel-to-reel-fabricated organic
solar cells. The active layer of the solar cells is a P3HT/PCBM blend. Note: P3HT is poly(3- vative approaches to achieve enhanced
hexylthiophene), PCBM is [6,6]-phenyl-C61-butyric acid methyl ester, PEDOT is poly(3,4- control over the morphology of the pho-
ethylenedioxythiophene), PSS is poly(4-styrenesulfonate), and TCO is transparent toactive layer, including the use of block-
conductive oxide. copolymer and dendritic structures.12
Although such approaches have, to date,
minimizing interfacial charge recombina- P3HT/PCBM.7 Such analyses have identi- led to only modest device efficiencies, the
tion losses, thereby achieving device effi- fied two key strategies to achieve further more systematic control of blend mor-
ciencies in excess of 4%. advances. One strategy aims at improving phology that can be achieved with such
Theoretical analyses of the optimum the voltage output of the solar cell. This structures is very promising.
device performance achievable as a func- voltage is limited by the free energy of the Aside from improvements in the pho-
tion of particular materials properties have photoinduced charge-separated state and, toactive layer, several studies have tar-
guided strategies to make further improve- therefore, by the energy difference between geted advances in device efficiency
ments in device performance beyond the polymer HOMO level and PCBM through overall device design. One

MRS BULLETIN • VOLUME 33 • JULY 2008 • [Link]/bulletin 671

Downloaded from [Link] UNIVASF, on 18 Sep 2019 at 21:53:59, subject to the Cambridge Core terms of use, available at [Link]
[Link]
 Solution-Processed Organic Solar Cells

approach is to use light management

approaches to enhance light absorption.
The top contact of OPV devices is typically
a reflective, metallic electrode, opening the
potential for multiple light pathways
through the active layer, as well as interfer-
ence effects. Examples of light manage-
ment strategies already showing promise
include the use of TiOx optical spacer lay-
ers to optimize the spatial distribution of
light within the device13 and the use of cor-
rugated device structures to ensure multi-
ple light paths through the active layer.14
An alternative approach is to employ tan-
dem device structures. For example, a 6%
device efficiency has recently been
reported for an all-solution-processed tan-
dem polymer cell employing a TiOx inter-
face layer between the two cells.15

Production Technologies
Highest productivity at lowest process
costs is gained from roll-to-roll processing.
Roll-to-roll processing can be divided into
solid-state (transfer), gas-phase (vacuum),
and liquid-state deposition techniques.
Production of photovoltaic modules
requires at least the deposition of stripes
of active layers using a pattern on to a
web. The method electrically separates
the active layers into individual cells,
which later are interconnected in series
(Figure 3). All three classes of deposition
processes can produce either patterned or
unpatterned films. Techniques for pattern-
ing of thin films can be divided into sub- Figure 2. National Renewable Energy Laboratory (NREL) certificate for a solar cell based
tractive or additive methods. In the first on an electron-donating polymer from the class of bridged bithiophene copolymers.29 This
case, an unpatterned film becomes pat- type of polymer already exhibits short-circuit current densities of up to 15 mA/cm2 at peak
terned in a separate process step, by external quantum efficiency (EQE) values of 60% at 800 nm. A fully optimized polymer with
an EQE as high as 80% has the potential to deliver short-circuit currents of up to 18–20
removing film from undesired areas (e.g., mA/cm2 and efficiencies of around 7%.
by lithography, laser patterning, emboss-
ing, or imprinting, to name a few of the
more popular technologies for postpat- 0 1 2 3 4 5
terning). In the case of additive methods, Barrier Substrate Electrode PEDOT Semiconductor Electrode Packaging
film deposition and patterning are done in
the same process step.
Solid-state printing typically starts from
already prepatterned films. Alternatively,
the pattern can be generated during the
transfer process by a technique such as a 3
laser transfer method, in which a laser
Thickness (µm)

beam ablates material from the web. 2

Patterned vacuum deposition is real-
ized by shadow masking or by lift-off 1
processes. Both processes have limitations 0
in terms of resolution and registration.
Shadow-mask processing requires specific −1
precautions to prevent particles and to 0 5 10 15 20 25 30 35 40 45 50
control the distance between the web and Length (mm)
the mask. More sophisticated lift-off
processes are used for the manufacturing Figure 3. Cross section through a four-stripe organic photovoltaic module. Note that the x
of patterned Al films. Here, a fast- and y axes have different scales. Production requires five major steps: (1) deposition of the
evaporating ink (e.g., an oil) is printed electrode, (2) deposition of the electron blocking layer [i.e., poly(3,4-ethylenedioxythiophene)
right before the film is run through the (PEDOT)], (3) deposition of the semiconductor, (4) deposition of the top electrode, and
evaporation unit. During evaporation, the (5) packaging.

672 MRS BULLETIN • VOLUME 33 • JULY 2008 • [Link]/bulletin

Downloaded from [Link] UNIVASF, on 18 Sep 2019 at 21:53:59, subject to the Cambridge Core terms of use, available at [Link]
[Link]
 Solution-Processed Organic Solar Cells

ink is released from the substrate, and the drying kinetics of the wet film dur- The fabrication of modules requires
adhesion of the metal is prevented. The ing blading are quite different from some patterning of the coated layers. At
best resolution and registration are gained those encountered during spin coating least the interconnect region, where adja-
with classical lift-off or photolithography and can lead to different solid-state mor- cent cells are electrically connected to each
methods, both of which require printing phologies.20 However, detailed investiga- other in series, needs to be kept clean from
of a suitable photoresist. tions have shown that blading can result coated material. Most of the coating tech-
Liquid-state printing results directly in in fully optimized solar cells with per- niques do have a limited potential to
patterned films and is typically used formances identical to those of cells pre- deposit patterned layers. The patterning
when high resolution and registration are pared by spin coating.19 of long stripes in the down web direction
required. In contrast, liquid deposition Another coating technology, spray coat- as specified for photovoltaic modules is
methods that result in unpatterned films ing, which is even easier to perform than fortunately compatible with the specifica-
are commonly called “coating” methods. doctor blading, has been suggested as an tions of various coating methods (e.g, by
Coated, unpatterned films can be con- elegant alternative.21 Spray coating trans- using more complex coating heads with
verted into patterned films by subsequent fers small droplets (from microns to tens multiple die channels). Nevertheless, the
combination with (subtractive) patterning of microns) of ink onto a substrate, where resolution and registration of such a
steps (e.g., mechanical scribing or laser they dry upon impact. Ink formulation is process are somewhere in the submillime-
patterning). not an issue, as the method works well for ter range, which is nearly one order of
highly diluted solutions. Spray coating magnitude less than desired. An elegant
Production of Solar Cells by typically results in rough thin-film sur- way to overcome this limitation is to post-
Coating Techniques faces. Because of the transfer of rather pattern coated films. Postpatterning of
Over the years, the processing of bulk large droplets that individually dry on the flood-coated films can be done by either
heterojunction solar cells has become substrate without first forming a closed mechanical or optical methods. Laser pat-
increasingly sophisticated. The first film, the surface roughness is typically in terning of thin films is well-established in
reports on solution-processed devices uti- the tens of nanometers (compared to a few the inorganic thin-film photovoltaic com-
lized spin coating as the manufacturing nanometers for bladed or spin-coated munity and was probed for the produc-
process.16,17 Spin coating is a well- films). Nevertheless, coherent films and tion of organic solar cells in combination
understood process for the deposition of fully functional solar cells and photode- with large-area coating. Initial trials indi-
thin-film coatings with high precision. tectors have been obtained by spray coat- cated the high potential of this method
The method requires little formulation ing the organic semiconductor layer to a with respect to resolution, speed, and
work and is quite forgiving over a consid- dry film thickness of 3–4 times the surface costs.24
erable range of viscosities. Wetting can be roughness.22
guaranteed by surface pretreatment of the All of these pioneering trials in coating Production of Solar Cells by
substrate. The film thickness is controlled organic semiconductor layers and con- Printing Techniques
by a combination of spinning speed, accel- verting them into solar cells have indi- Printing is, of course, the most straight-
eration, solution viscosity, and tempera- cated that reel-to-reel coating of organic forward technology for creating patterned
ture; typical spin-coated films are between solar cells is possible. Indeed, the success- films. However, most printing methods
50 and 5,000 nm. Typical spin-coating ful manufacturing of organic solar cells require relatively high-viscosity inks.
machines handle substrates up to 6–8 in. using reel-to-reel slot-die coating has been Coating is more compatible with low-
in diameter. The disadvantages of spin reported.23 Both of the active layers—the viscosity inks, as it does not generate a
coating are the large material losses electron-blocking layer [PEDOT:PSS, pattern that needs to stay on the substrate
(>90%) and the incompatibility with roll- where PEDOT is poly(3,4-ethylene- while drying. Each of the printing meth-
to-roll processing. Whereas the material dioxythiophene) and PSS is poly(4- ods operates best for specific viscosities,
losses can be minimized by collecting and styrenesulfonate)] and the semiconductor and fully optimized inks can produce the
purifying the residuals from the spin layer (P3HT:PCBM)—were coated at a resolutions and productivities listed in
bowl, the incompatibility with roll-to-roll speed of meter(s) per minute on a 50-mm- Table 1. There are three printing methods
processing remains. wide indium tin oxide-(ITO-)coated poly- that work well with low-viscosity inks:
Shortly after the first presentation of ester film. Small areas of the web-coated inkjet, flexo, and gravure printing. These
spin-coated organic solar cells, doctor film were converted into functional solar methods would be natural choices for the
blading was demonstrated to overcome cells by vacuum deposition of Al as the production of organic solar cells.
many of the disadvantages of spin coat- top electrode (cathode), resulting in device The first report on printed organic solar
ing.18,19 Doctor blading is a coating tech- efficiencies of 1%. cells used a screen printing process to
nology that is comparable to knife coating,
whereby a knife homogenously spreads
ink as the blade is drawn over a substrate. Table I: Characteristic Properties of Printing Methods.
Doctor blading is fully compatible with Method Resolution Layer Thickness Ink Viscosity Maximum
both batch and roll-to-roll processing. Ink
(mm) (mm) (Pa s) Throughput (m2/h)
formulation for doctor blading is quite
comparable to that for spin coating. Four Inkjet >50 0.1–20 0. 001–0.04 >50
parameters determine the film formation: Screen >100 1–15 0. 5–50 >100
the concentration (responsible for the vis-
Flexo >50 0.5–2.5 0. 05–0.5 >50,000
cosity of the ink), the temperature of the
solution, the speed of the blade, and the Gravure >30 0.5–8 0. 05–0.2 >100,000
distance between the blade and the sub-
strate (i.e., the volume of ink being Note: The table summarizes typical performance values for the printing methods. However, deviations can be
deposited). The temperature profile and found in the case of specialized inks or processes.

MRS BULLETIN • VOLUME 33 • JULY 2008 • [Link]/bulletin 673

Downloaded from [Link] UNIVASF, on 18 Sep 2019 at 21:53:59, subject to the Cambridge Core terms of use, available at [Link]
[Link]
 Solution-Processed Organic Solar Cells

deposit the active layer. In screen printing, and control are continuous and drop- ogy formation of the bulk heterojunction
a stencil is patterned by closing selected on-demand (DoD) systems. In the former composite during the drying phase, espe-
areas of a screen. Upon imprint, ink is case, a continuous liquid jet through a cially with respect to vertical phase sepa-
forced through open meshes by means of a nozzle is split into drops, for example, ration of the polymer to the fullerene.
blade. Screen printing typically requires by means of an acoustic wave gener- Only a few research groups worldwide
10–100 times higher viscosity inks than ated by a piezoelectric device. Ejected work on the application of mass printing
those used for spin coating or doctor blad- droplets are then electrically charged and for the fabrication of large-area, flexible
ing. Nevertheless, even the very first deflected according to the desired pattern. electronics. This is probably due to the
report on screen-printed solar cells des- In DoD systems, droplets are discon- extensive know-how and comparatively
cribed excellent performance. Cells in which tinuously created by acoustic (mostly for high effort required for adapting new ink
the active layer was screen printed were non-water-based inks) or heat (mostly systems for mass printing.
directly comparable to the spin-coated ref- for water-based inks) pulses when needed. Two widely discussed mass-printing
erence.25 Screen printing of OPV has been Inkjet systems can reproducibly handle methods for printed electronics are
reported a few more times,26 but the droplets of a few picoliters up to nano- gravure and flexography printing. In flex-
requirement for high viscosities limits the liters, corresponding to pixel sizes of at ography printing, soft and flexible print-
available materials. Only conjugated poly- least a few microns. ing plates are used in combination with
mers with high molecular weights and Inkjet printing of solar cells has differ- low-viscosity inks and low transfer pres-
high solubilities are suitable for high- ent requirements than inkjet printing of sures. The flexibility of the printing plates
quality screen printing. Today’s commer- PLEDs. Photovoltaic applications require contributes to reducing the resolution and
cial screen printing inks are quite complex large and homogenous coated areas. DoD layer quality. No other printing method,
fluid systems, in which various binders, inkjet printers generate single pixels with however, can deposit low-viscosity inks
thickeners, and thixotropy and viscosity diameters according to the droplet size with comparatively high and variable
modifiers are mixed with the pigment. and to the number of droplets fired, thicknesses up to several micrometers on
Thixotropy modifiers decrease the viscos- making the printers unsuitable for PV both flexible and rigid thin films.
ity of the ink upon applied stress. These applications. Successful inkjet printing of In gravure printing, the printing pat-
additives certainly can convert conju- photovoltaic cells requires the develop- tern is engraved into the printing plate
gated-polymer-based fluids into well- ment of inks that dry sufficiently slowly, (or film/cylinder). During inking, the
defined screen printing inks, but so far, no in combination with a printing scheme engraved cells of the plate are filled, and
studies have been conducted to determine that fires droplets sufficiently narrowly during imprinting, the substrate removes
whether these additives have an impact on spaced to make single pixels coalesce over and takes up the ink from the cells.
the photovoltaic performance. a reasonably large area before drying. Gravure printing can give outstandingly
Inkjet printing has evolved as the most Following these design rules, high- high resolution and registration for rela-
popular printing method for the produc- performance P3HT/PCBM solar cells tively low-viscosity inks. These properties,
tion of multicolor polymer light-emitting were inkjet printed from a solvent mix of in combination with a high-throughput
diode (PLED) displays and, by now, is a high-boiling and a low-boiling solvent production capacity, make gravure print-
probably the most widely used printing (Figure 4).27 It was further shown that ing a promising method for the mass pro-
method in organic electronics. The two such solvent combinations can signifi- duction of OPV. The first gravure printing
basic methods for droplet formation cantly influence the solid-state morphol- trials of P3HT/PCBM blends reported
excellent film quality.28 The homogeneity
4 and the surface roughness of gravure-
Inkjet Printed OPV printed P3HT/PCBM films were compara-
2 ble to that of spin-coated layers. The
JSC = 8.4 mA/cm2 gravure-printed solar cells had good diode
VOC = 0.535 V properties and sufficient shunts. However,
Current Density (mA/cm2)

0 the short-circuit current of the gravure-

FF = 0.64
printed solar cells was lower than average
PCE = 2.9% for P3HT/PBCM cells and might reflect the
−2
formation of a less favorable morphology
during the drying process. Nevertheless,
−4
these preliminary studies clearly demon-
strated that organic solar cells can be mass-
−6 produced by printing methods.

−8 Conclusions
Developments in organic photovoltaics
are moving rapidly. The best laboratory
−10 efficiencies are currently in the 5–6% range,
−0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 and the community has developed a clear
Voltage (V) understanding and strategy of how to fur-
ther evolve performance. No fundamental
Figure 4. Current–voltage (J–V) curves for an inkjet-printed organic solar cell under
obstacles have been identified to achieving
illumination. The semiconductor solution, a poly(3-hexylthiophene)/[6,6]-phenyl-C61-butyric efficiencies of 10–15%, and the general
acid methyl ester (P3HT/PCBM) composite, was printed from a mesitylene/ortho- expectation is that the further progress will
dichlorobenzene mixture. Note: Jsc is short-circuit current, Voc is open-circuit voltage, FF is come from both novel optimized materials
fill factor, and PCE is power conversion efficiency. and more efficient device geometries.

674 MRS BULLETIN • VOLUME 33 • JULY 2008 • [Link]/bulletin

Downloaded from [Link] UNIVASF, on 18 Sep 2019 at 21:53:59, subject to the Cambridge Core terms of use, available at [Link]
[Link]
 Solution-Processed Organic Solar Cells

A breakthrough has been achieved in the Giles, I. McCulloch, C.-S. Ha, M. Ree, Nat. 17. J.J.M. Halls, C.A. Walsh, N.C. Greenham,
development of production methods for Mater. 5, 197 (2006). E.A. Marseglia, R.H. Friend, S.C. Moratti, A.B.
organic solar cells during the past 5. J.Y. Kim, S.H. Kim, H.-H. Lee, K. Lee, W. Ma, Holmes, Nature 376, 498 (1995).
few years. Multiple printing and coating X. Gong, A.J. Heeger, Adv. Mater. 18, 572 18. C.J. Brabec, F. Padinger, J.C. Hummelen,
(2006). R.A.J. Janssen, N.S. Sariciftci, Synth. Met. 102,
methods have been demonstrated to 6. M. Campoy-Quiles, T. Ferenczi, T. 861 (1999).
yield the required film specifications for Agostinelli, P.G. Etchegoin, Y. Kim, T.D. 19. F. Padinger, C.J. Brabec, T. Fromherz, J.C.
device operation. Moreover, one company Anthopoulos, P.N. Stravinou, D.D.C Bradley, Hummelen, N.S. Sariciftci, Optoelectron. Rev. 8
recently succeeded in the roll-to-roll pro- J. Nelson, Nat. Mat. 7, 158 (2008). (4), 280 (2000).
duction of photovoltaic modules, where all 7. M.C. Scharber, D. Muehlbacher, M. Koppe, P. 20. P. Schilinsky, C. Waldauf, C.J. Brabec, Adv.
layers were deposited by printing and coat- Denk, C. Waldauf, A.J. Heeger, C.J. Brabec, Adv. Funct. Mater. 16, 1669 (2006).
ing methods. Roll-to-roll manufacturing of Mater. 18, 789 (2006). 21. D. Vak, S. Kim, J. Jo, S. Oh, S. Na, J. Kim,
functional modules was the last missing 8. E. Bundgaard, F.C Krebs, Sol. Energy Mater. D. Kim, Appl. Phys. Lett. 91, 081102 (2007).
piece of technology between development Sol. Cells 91, 954 (2007). 22. O. Hayden, presented at the Materials
9. H. Ohkita, S. Cook, Y. Astuti, W. Duffy, S. Research Society Fall Meeting, Boston, MA,
and commercialization, and the first Tierney, W. Zhang, M. Heeney, I. McCulloch, 2007.
organic solar cell products are expected to J. Nelson, D.D.C. Bradley, J.R. Durrant, J. Am. 23. K. Schultheis, L. Blankenburg, S. Sensfuss,
enter the market within the short term. Chem. Soc. 130 (10), 3030 (2008). M. Schrödner, Proceedings of the Cintelliq con-
10. D. Mühlbacher, M. Scharber, M. Morana, ference, Frankfurt 2007.
Acknowledgments Z. Zhu, [Link], [Link], C. J. Brabec, Adv. 24. G. Dennler, C. Lungenschmied, H. Neuge-
The authors are grateful to their many Mater. 18, 2884 (2006). bauer, N.S. Sariciftci A. Labouret, J. Mater. Res.
colleagues and co-workers for valuable 11. J. Peet, J.Y. Kim, N.E. Coates, W.L. Ma, D. 20 (12), (2005).
discussions and input. JRD acknowledges Moses, A.J. Heeger, G.C. Bazan, Nat. Mater. 6, 25. S.E. Shaheen, R. Radspinner, N. Peygham-
497 (2007). barian, G.E. Jabbour, Appl. Phys. Lett. 79, 2996
financial support from the EPSRC Super- 12. S.M. Lindner, S. Huttner, A. Chiche, M. (2001).
gen program. Thelakkat, G. Krausch, Angew. Chem., Int. Ed. 26. F. Krebs, J. Alstrup, H. Spangaard, K.
45, 3364 (2006). Larsen, E. Kold, Sol. Energy Mater. Sol. Cells 83,
References 13. J.Y. Kim, S.H. Kim, H.H. Lee, K. Lee, W.L. Ma, 293 (2004).
1. R.A.J. Janssen, J.C. Hummelen, N.S. X. Gong, A.J. Heeger, Adv. Mater. 18, 572 (2006). 27. C. Klepek, S.A. Choulis, P. Schilinsky, C.J.
Sariciftci, MRS Bull. 30, 33 (2005). 14. K. Tvingstedt, V. Andersson, F. Zhang, Brabec, Adv. Mater. 19, 3973 (2007).
2. C.J. Brabec, J.A. Hauch, P. Schilinsky, C. O. Inganas, Appl. Phys. Lett. 91, 123514 (2007). 28. M. Tuomikoski, P. Kopola, Technologies for
Waldauf, MRS Bull. 30, 50 (2005). 15. J.Y. Kim, K. Lee, N.E. Coates, D. Moses, T.Q. Polymer Electronics TPE 06, Rudolstadt,
3. G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, Nguyen, M. Dante, A.J. Heeger, Science 317, 222 Germany, 16–18 May 2006.
K. Emery, Y. Yang, Nat. Mater. 4, 864 (2005). (2007). 29. Konarka NREL certificate from July
4. Y. Kim, S. Cook, S.M. Tuladhar, S.A. Choulis, 16. G. Yu, A.J. Heeger, J. Appl. Phys. 78, 4510 2007. ■
J. Nelson, J.R. Durrant, D.D.C. Bradley, M. (1995).

MRS BULLETIN • VOLUME 33 • JULY 2008 • [Link]/bulletin 675

Downloaded from [Link] UNIVASF, on 18 Sep 2019 at 21:53:59, subject to the Cambridge Core terms of use, available at [Link]
[Link]