Micron-sized [6,6]-phenyl C61 butyric acid methyl ester crystals grown

by dip coating in solvent vapour atmosphere: interfaces for organic

photovoltaicsw

R. Dabirian,aX. Feng,

bL. Ortolani,

cA. Liscio,

aV. Morandi,

cK. Mullen,*

b

P. Samorı*ad

and V. Palermo*a

Received 10th November 2009, Accepted 1st February 2010

First published as an Advance Article on the web 5th March 2010

DOI: 10.1039/b923496a

We have devised a novel dip coating procedure to form highly crystalline and macroscopic

p-conjugated architectures on solid surfaces. We have employed this approach to a

technologically relevant system, i.e. the electron-acceptor [6,6]-phenyl C61 butyric acid

methyl ester molecule (PCBM), which is the most commonly used electron-acceptor in organic

photovoltaics. Highly ordered, hexagonal shaped crystals of PCBM, ranging between 1 to

80 mm in diameter and from 20 to 500 nm in thickness, have been grown by dip coating the

substrates into a solution containing the fullerene derivative. These crystals have been found to

possess a monocrystalline character, to exhibit a hexagonal symmetry and to display micron sized

molecularly flat terraces. The crystals have been prepared on a wide variety of surfaces such as

SiOx, silanized SiOx, Au, graphite, amorphous carbon–copper grids and ITO. Their multiscale

characterization has been performed by atomic force microscopy (AFM), Kelvin probe force

microscopy (KPFM), X-ray diffraction (XRD), optical microscopy, scanning and transmission

electron microscopy (SEM, TEM).

To test the stability of these electron accepting PCBM crystals, they have been coated with a

complementary, electron donor hexa-peri-hexabenzocoronene (HBC) derivative by solution

processing from acetone and chloroform–methanol blends. The HBC self assembles in a well-defined

network of nanofibers on the PCBM substrate, and the two materials can be clearly resolved by

AFM and KPFM.

Due to its structural precision on the macroscopic scale, the PCBM crystals appear as ideal

interface to perform fundamental photophysical studies in electron–acceptor and –donor blends,

as well as workbench for unravelling the architecture vs. function relationship in organic solar cells

prototypes.

Introduction

[6,6]-Phenyl C61 butyric acid methyl ester molecule PCBM

(Scheme 1a)1 is nowadays the most successful n-type organic

semiconductor for organic-solar cells in terms of device

performance.2 The energy gap between the highest occupied

molecular orbital (HOMO) and the lowest-unoccupied

molecular orbital (LUMO) of PCBM are reduced compared

to C60 allowing for increased photoelectric conversion

efficiencies.3 Another one of its advantages is the high

solubility in various organic solvents. It can be blended with

p-type conjugated polymers to form thin-film organic field

effect transistors (OFETs), featuring ambipolar charge

transport.4 PCBM has also shown promise for use in photo-

detectors.5 In organic photovoltaic blends PCBM is employed

upon mixing it with an electron donor molecular system,

such as polythiophene, poly(phenylene-vinylene) or hexa-peri-

hexabenzocoronene to give phase segregation between the

two electronically dissimilar components on the length scale

typical of the mean exciton diffusion, i.e. about 5–6 nm.6–10

Despite the large number of papers studying the opto-

electronic properties of PCBM in blends and solar cells, the

exact molecular packing of PCBM is still not clear. In blends

with other molecules, PCBM can either disperse on a

molecular scale in the donor matrix, or phase separate forming

irregular crystals or rounded agglomerates, depending on

deposition and post-treatment (e.g. annealing) conditions.11–18

When drop-cast from chlorobenzene (CB), it was found

to assemble into a monoclinic unit cell (a = 13.7, b = 16.6,

a Istituto per la Sintesi Organica e la Fotoreattivita, ConsiglioNazionale delle Ricerche, Via Gobetti 101, 40129 Bologna, Italy.E-mail: palermo@isof.cnr.it

b Istituto per la Microelettronica e Microsistemi, Consiglio Nazionaledelle Ricerche, via Gobetti 101, 40129 Bologna, Italy.

cMax Planck Institute for Polymer Research, Ackermannweg 10,55128 Mainz, Germany. E-mail: muellen@mpip-mainz.mpg.de

d Institut de Science et d’Ingenierie Supramoleculaires (ISIS)—CNRS7006, Universite de Strasbourg 8, allee Gaspard Monge,67000 Strasbourg, France. E-mail: samori@isis-ulp.org

w Electronic supplementary information (ESI) available: TopographicalAFM image of nanoscale amorphous PCBM aggregates on silanizedSiOx formed after dip coating at room temperature (Fig. S1); Opticalmicroscopy images of (a) PCBM crystals on SiOx formed during rapidwithdrawal speed; (b) PCBM crystals on SiOx forming preferentiallyalong the scratched line (Fig. S2); XRD spectrum of PCBM crystals onsilanized SiOx (Fig. S3); AFM images (Fig. S4). See DOI: 10.1039/b923496a

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 | 4473

PAPER www.rsc.org/pccp | Physical Chemistry Chemical Physics

c = 19.0 A, b = 105.21) containing 4 PCBM and 2 CB

molecules. When it was, instead, drop cast from ortho-dichloro-

benzene (ODCB) it yielded a triclinic unit cell (a = 13.8,

b = 15.2, c = 19.2 A, a = 80.2, b = 78.51) containing

4 PCBM and 4 ODCB molecules.19 In a different work,

needle-like structures were obtained by drop casting from

CB and toluene, showing through SAED (selected area

electron diffraction) measurements a whole set of different

d-spacings, ranging from 2.1 to 8.4 A.12 In some works, an fcc

crystal packing was assumed, in analogy with unfunctiona-

lized C60, with a lattice constraint of 14 A.13 Alternatively, an

incomplete unit cell was proposed with a = 15.5, b = 10.1,

g = 901.17 The different results were mainly due to the fact

that PCBM tends to form disordered, amorphous aggregates

composed of nano-crystals with random orientations, whose

structure depends on the solvent used for the deposition, and

often includes solvent molecules in the lattice; it is thus

important to find new ways to form large crystals of PCBM,

having well-defined size and orientation.

The preparation of high-symmetry molecular solids is thus

not only for aesthetic curiosity; the simple physical and

chemical properties of these structures also make it possible

to gain a better understanding of intermolecular interactions

in the low-symmetry analogues.20 Because of this reason the

previous efforts have been addressed towards the fabrication

of macroscopic crystalline architectures based on the insoluble

pure fullerene molecule.20–24

A subtle equilibrium involving substrate, solute and solvent

is known to dictate the self-assembly of organic molecules

when solution is processed on a substrate.25 In particular, the

relative affinity of the substrate for the adsorbed molecule, in

our case PCBM, and the chosen solvent can be expected to

govern the crystalline nature of the films, and eventually

the crystal size and shape. Moreover, the kinetics of the

physisorption process is known to be a crucial parameter

governing the self-assembly. Consequently the solvent

evaporation rate and its affinity for the substrate can be

expected to play relevant roles. It is well known that PCBM

crystallizes when the solvent evaporation is kept slow, leading

to the formation of long needle-like PCBM single crystals.12

Upon spin coating, solvent evaporation and deposition time

are much faster, yielding more uniform but less ordered

material.26,27

In the past, we developed new methods to favour the

controlled self-assembly of different organic semiconductors

using unconventional techniques.28–30 Here, to get large,

ordered self-assembly of PCBM, we describe a new method

based on the dip coating procedure performed in a controlled

environment to achieve higher control over the kinetics of the

physisorption process. The obtained macroscopic PCBM

crystals have been employed as model systems to grow

structurally well-defined electron–acceptor and –donor

interfaces with a strong p-type molecule, i.e. a C3-symmetric

hexa-peri-hexabenzocoronene (HBC).

Experimental

Commercially available PCBM was used without further

purification (Aldrich >99.9%). The C3-symmetric hexa-peri-

hexabenzocoronene was synthesized as previously described.31

A 2.5 mL solution with an initial PCBM concentration of

2.5–3.5 mg mL�1 is used for dip coating. The solvent used

throughout for dip coating was either pure chloroform, or a

chloroform : chlorobenzene 9 : 1 (v/v) mixture. The solution

was poured into a sealed vial, leading to the saturation of the

solvent vapours within the vial and significantly slowing

down solvent evaporation upon the substrate, which in

turn lowered the crystallization speed (Fig. 1). A program-

mable stepper (Lego Mindstorms) was used to remove the

sample out of the solution. This procedure typically led to the

formation of hexagonal crystals averaging 20 mm in diameter.

The system temperature was controlled to a few degrees

precision within a 2–10 1C range, by carrying out the dips in

a refrigerator. Withdrawal speeds in the range of 1–10 cm h�1

were used.

Si/SiOx substrates ([100], p-doped, R = 0.01 O cm) were

cleaned by using a standard RCA procedure.32 Silanization

of the Si/SiOx substrates was performed by exposing the

Scheme 1 Chemical formula of (a) PCBM and (b) HBC derivative with polar side chains.

4474 | Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 This journal is �c the Owner Societies 2010

substrates to the vapour fumes of a heated hexamethyldisila-

zane (Aldrich 99.9%), 3 h at 150 1C inside a closed glass jar, to

make the substrate hydrophobic.33 Graphite substrates were

cleaved with a scotch tape to expose flat surfaces prior to

dipping. Gold substrates were ozonized for ca. 60 min before

dipping. ITO (indium tin oxide) substrates were sonicated in

acetone (10 min) and ethanol (10 min) prior to dipping. All

substrates were ca. 5 � 5 mm2 in area.

The PCBM crystals were coated with the C3-symmetry

hexa-peri-hexabenzocoronene (HBC) by spin-coating. This

molecule was chosen because its amphiphilic type of

substituents 29,34 offers different solubility with respect to

PCBM, and can thus be processed in slightly more polar

solvents. In order to achieve a thin layer of coverage a

0.3 mg mL�1 solution of HBC in either hot acetone or a

chloroform–methanol 2 : 1 (v/v) mixture were used and

samples were spun at 2000 rpm for 60 s. Thick layers (120 nm)

of HBC were prepared by spin coating a 15 mg mL�1 solution

of HBC in chloroform at 1000 rpm for 60 s.

Intermittent contact AFM topographical images were

recorded by using either a Multimode IIIA microscope

(Veeco, S.Barbara, CA, USA) equipped with the Extender

Electronics module or an Autoprobe CP Research

(ThermoMicroscope, Sunnyvale, CA, USA). The measure-

ments were carried out under atmospheric conditions at room

temperature with scan rates of 0.3–2.0 Hz line. Scan sizes

spanning from 95 mm down to 1.0 mm were explored, with a

resolution of 512 � 512 pixels using non-contact Si ultralevers

(RFESPA5 Veeco) with a spring constant k o 4 N m�1.

KPFM measurements were carried out with a Multimode

IIIA microscope (Veeco, S. Barbara, CA, USA) operating in

lift mode: each line was scanned twice, first to measure

the topography in tapping mode and second to measure the

electrostatic potential at a predefined lift height of 50 nm.35

In KPFM technique, the electrostatic interaction between a

scanning tip and the sample was used to get a quantitative map

of the surface potential.36 In order to obtain a sufficiently

large and detectable mechanical deflection, we employed

soft (k o 4 N m�1) highly doped Si cantilevers with

oscillating frequencies in the range of 60 o o o 90 KHz

(SCM, Veeco).

Transmission electron microscopy (TEM) observations was

carried out with a Fei Tecnai F20 TEM equipped with a

Schottky emitter and operating at 200 keV. The scanning

electron microscopy (SEM) images were obtained with a

ZEISS 1530 SEM operating at 10 keV, with an In-Lens

secondary electrons detector.

XRD measurements were carried out on the silicon plates at

room temperature with a Bragg/Brentano diffractometer

(X’pertPro Panalytical) equipped with a fast X’Celerator

detector, using Cu anode as X-ray source (Ka, l =

0.15418 nm).

Results and discussion

Preparation of PCBM interfaces

The substrate was dipped into a vial filled with a concentrated

PCBM solution (Fig. 1a). To process PCBM we used

chloroform, which is an excellent solvent for these kinds of

molecules, and allowed getting high concentrations of PCBM

in solution. The substrate was mounted on a computer-

controlled arm (Fig. 1b), which enabled performing multiple

dipping of the same sample at a highly controlled and very

low speed. The vial was sealed and filled with a saturated

atmosphere of solvent vapours, so that very slow solvent

evaporation was attained, much slower and more controlled

of what is usually obtained in drop casting in a closed

atmosphere.12 Solvent evaporation was thus very slow at the

meniscus formed when the sample was removed from the

solution (Fig. 1c) thus allowing, by means of a high control

over the kinetics of the process, to grow large PCBM crystals

(Fig. 2). Such a procedure required the optimization of various

experimental parameters including the PCBM concentration

in the solution, the temperature, the withdrawal speed and the

final height of the sample in the saturated vapour area inside

Fig. 1 (a) Schematic representation of dip coater used for deposition. (b) Photograph of the home-built dip coater. (c) Depiction of the substrate

and solvent meniscus during dip coating process.

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 | 4475

the tube above the interface, which together dictate the

evaporation rate of the solvent as well as the geometry of

the receding solvent.37

PCBM crystal formation and morphology

We have undertaken a systematic study of the crystal

morphology of the dip coating process by employing various

substrates and different temperature intervals. Fig. 2 displays

the optical microscopy images of the resulting morphologies of

the PCBM crystals supported (a) on silanized SiOx at T =

10 1C, (b) graphite, at T = 10 1C, and (c) Au surface at T

ca. 2 1C. Fig. 2a reveals the presence of hexagonally shaped

crystals, with an average diameter of 20 � 10 mm and thickness

of 300� 100 nm. The crystals were obtained by dipping freshly

prepared silanized SiOx substrates into a 3.5 mg mL�1 PCBM

solution in chloroform at a temperature of 10 1C, with a

withdrawing speed of 1 cm h�1 being applied. The crystals

were found to be either composed of a singular flat hexagonal

layer, or made up of multiple hexagonal layers stacked in a

staggered position on top of each other. Some of the crystals

appeared to have merged with neighboring crystals.

In contrast, Fig. 2b exhibits a sample prepared by dipping a

freshly cleaved graphite substrate into a similar 3.5 mg mL�1

PCBM solution in chloroform at 10 1C, also using a 1 cm h�1

withdrawing speed. Under these conditions, by using freshly

cleaved graphite instead of silicon mostly smaller PCBM

aggregates with a disordered amorphous form were obtained

(Fig. 2b), some having a snowflake shape, with the presence of

only a few sporadic hexagonal crystals with diameters over

15 mm. On the other hand, Fig. 2c depicts a sample prepared

by dipping an Au film into the same 3.5 mg mL�1 PCBM

solution in chloroform by using a 1 cm h�1 withdrawing speed,

but at a temperature of ca. 2 1C. It reveals snowflake and

tripodial shaped polycrystals, whereas no hexagonally shaped

crystals were visualized. In contrast, when dipping an Au

substrate at 10 1C, hexagonal crystals similar to those in

Fig. 2a were obtained.

The hexagonal crystals formed upon silanized SiOx (Fig. 2a)

were found to possess a diameter of up to 80 mm. Their surface

was very flat, as proved by a very low surface root-

mean-square roughness amounting to Rrms = 0.6 � 0.2 nm

on an area of 1 mm2. Such crystals were also obtained on other

substrates such as non-silanized SiOx, Au, ITO and copper

grids and porous carbon supports for TEM by applying the

same dipping conditions. As aforementioned, in order to

obtain these 20 � 10 mm sized hexagonal crystals, the

dipping was performed at 10 1C, using a withdrawing speed

of ca. 1 cm h�1. It was observed that at temperatures exceeding

15 1C, crystals were no longer formed and the substrate (SiOx)

was found to be covered with small amorphous agglomerates

of PCBM having a width of a few hundreds of nm (Fig. S1a in

the ESI).w Moreover, slower withdrawal speeds or multiple

dipping did not seem to increase the crystal size. On the other

hand, by increasing the withdrawal speed to 4 cm h�1 a

noticeable decrease in crystal size was observed, with crystals

over 20 mm becoming rare, while concurrently many amorphous

spheres were monitored (Fig. S2a in the ESI).w Substrate

irregularities were observed to play a role in the distribution

of the crystals: crystal nucleation was in fact found to be

favoured at surface defects such as cracks, scratches, edges and

steps (Fig. S2b in the ESI).wSnowflake shaped crystals are shown in Fig. 2c (Au). Such

crystals were always observed on graphite, Au and SiOx when

dipping at temperatures of below 4 1C. In addition, on

graphite, such snowflake shaped crystals were also detected

at 10 1C (which in the case of Au and SiOx substrates gave

hexagonal crystals).

We focused our attention primarily on crystals formed at

10 1C, as the films formed much above or much below this

temperature did not exhibit wide molecularly flat terraces.

The AFM image in Fig. 3a shows a typically multilayered

hexagonal polycrystal on SiOx, featuring two staggered

hexagonal layers. Other polycrystals were observed being

composed of up to three hexagonal layers. A zoom-in by

AFM on a crystal surface (Fig. 3c) reveals surfaces consisting

of a series of terraces, having a similar step with a height of

ca. 2.5 nm. This is comparable to a reflection peak of an XRD

powder diffraction spectrum recorded on the crystalline films

(see XRD part below and Fig. S3 in the ESI).wIn addition to these monomolecular steps, long channels

were also observed running all along the crystal surface, with a

depth of B1–2 nm and a width of 40 to 60 nm at the crystal

surface (Fig. S4 in the ESI).w They were found to be

exceptionally straight, running over several micrometres, and

parallel to each other, although with variable inter-channel

spacing. Further, they were typically oriented at periodic

601 angles, perfectly reflecting the three-fold symmetry

Fig. 2 Optical microscopy images of PCBM crystal on different surfaces; (a) monocrystalline hexagonal crystals on silanized SiOx, grown at

4–10 1C; (b) amorphous crystals on graphite grown at 4–10 1C, and (c) snowflake-like crystals on Au grown at T ca. 2 1C.

4476 | Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 This journal is �c the Owner Societies 2010

hexagonally shaped crystals. The origin of the channels is

unknown, nevertheless it is envisioned that they might be

cracks that form upon the drying up and evaporation of the

solvent.

To complete the structural characterization, SEM and TEM

observations were performed. The crystals were grown by

dip coating a copper TEM grid exposing a porous amorphous

carbon layer into the PCBM solution. Fig. 4a shows the

optical image of the crystals grown on both the copper

grid and on the porous amorphous carbon film. Fig. 4b

exhibits a SEM secondary electrons image of a PCBM single

crystal of ca. 3 mm size grown on the porous carbon film.

Even in this case, the obtained crystals show hexagonal

symmetry. The vertex of the crystals though, appears

smoothed and more irregular with respect to the SiOx

substrate, likely due to the non-uniform and less polar nature

of the underlying carbon grid. Fig. 5a depicts a TEM low

magnification image of a PCBM crystal superimposed to

the amorphous carbon film. It shows that most of the observed

crystals grown on the carbon were single crystals, some

of them instead being composed of different micron-sized

crystalline domains. In the inset of Fig. 5a is reported the

selected area electron diffraction pattern. It shows a hexagonal

symmetry, with d-spacing of 0.84, 0.49 and 0.42 nm. Fig. 5b

displays a high resolution image where the longitudinal fringes

are clearly visible. The fast Fourier transform in the inset

confirms that the associated distance is ca. 0.85 nm. This

0.85 nm periodicity was observed in XRD, SAED and TEM

measurements, and is in agreement with previous observa-

tions on more amorphous, needle-like structures.12 Unfortu-

nately, up to now it has not been possible to perform XRD

Fig. 3 (a) AFM image of multilayered crystal grown on silanized SiOx grown at 4–10 1C with a corresponding cross section profile, z-range;

290 nm; (b) showing flat surfaces; (c) close-up with six molecular steps are distinguishable, z-range; 32 nm.; (d) cross section indicating step heights

of ca. 2.5 nm.

Fig. 4 (a) PCBM crystals grown on the amorphous carbon film

(dark areas in the image) and on the copper support of a TEM grid;

(b) secondary electrons SEM image of a PCBM single crystal grown

on the carbon film.

Fig. 5 (a) Low magnification TEM image of a PCBM single crystal

on the carbon film of the TEM grid. In the inset is reported the selected

area diffraction pattern; (b) high resolution TEM image of a PCBM

crystal. The fringes are clearly visible in the inset, showing a periodicity

of 0.85 nm as measured by FFT.

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 | 4477

measurements on single PCBM hexagonal crystals, to assess

the unit cell of the crystals.

Electron-acceptor and -donor interfaces: coating of PCBM

crystals with HBC nanofibers

In order to explore the potential use of the obtained PCBM

crystals for the controlled formation of electron–acceptor and

–donor interfaces for organic photovoltaics, we have

coated them with a strong electron-donor molecule, viz.

a hexa-peri-hexabenzocoronene (HBC). Among the various

HBCs we have chosen derivatives exposing in the bay-area

amphiphilic type of substituents determined by the presence of

both polar and apolar side chains, providing good solubility in

more polar solvents than PCBM.29,34 This made it possible to

successfully deposit HBC on PCBM crystals using polar

solvents such as acetone or chloroform–methanol blends,

without re-dissolving the PCBM crystals, also thanks to the

very fast spin coating deposition technique employed and the

low surface–volume ratio of the crystals. The two main

concerns while performing the spin coating were: (a) to

perform it at a sufficient high speed or in a suitable solvent

so as not to dissolve or damage the crystal, (b) to keep the

concentration of HBC low, yielding only a partial coating of

HBC fibers on top of the PCBM crystals, to have the

possibility of comparing the surface of both bare and

HBC-coated areas.

These two conditions were met by using either a hot

0.3 mg mL�1 solution of HBC in acetone or a 0.3 mg mL�1

solution of HBC in chloroform–methanol 2 : 1 (v/v) mixture

and in both cases spinning at 2000 rpm for 60 s. Both these

solutions resulted in the PCBM crystal remaining apparently

intact and being partially covered by HBC fibers, giving rise to

well-defined bare and covered crystalline areas. However, the

use of the two solutions gave markedly different morphologies

of the blended components. Using acetone as solvent led to the

formation of sparse but well-defined distinct fibers spread

evenly on top of the crystals as well as across the substrate

(Fig. 6). The fibers were generally several microns in length,

between 20 and 90 nm in height and from 150 up to 350 nm in

width. In contrast, the solution in chloroform–methanol resulted

in a denser coverage of continuous fiber networks (Fig. 7a).

These latter fibers were generally lower in height (from a few nm

up to 50 nm) and slightly thinner (20 to 100 nm), and they were

interconnected as in a network, resulting in a much higher

degree of PCBM coverage.

In both systems, the HBC fibers could be clearly discrimi-

nated from the PCBM flat surface in the topography images

(Fig. 6a, 7a), and even better in the gradient of the topography

(Fig. 6b, 7b).

To further confirm the composition of the structures

observed, we performed KPFM microscopy36 on the surface

of these composite objects. The surface potential of these maps

obtained in this way shows a sharp contrast between the

nanocrystals and the fibers (Fig. 6c, 7c), with the electron

accepting PCBM being more negatively charged (i.e. darker in

the image) than the electron donating HBC fibers. Some

charge generation and splitting between PCBM and HBC

could be monitored under illumination by KPFM, as observed

previously in many similar acceptor-donor blends.38–40

Unfortunately, the observed variation in surface potential

under illumination was highly dependent on fibers’ density

and size, and comparable to experimental noise, so that it

could not provide any useful information.

This can be due to the reduced interface between the PCBM

and the HBC, as compared to conventional, nanoscale

dispersed bulk heterojunction blends, or to a poor electrical

contact between the nanofibers and the PCBM crystals,

presumably due to the edge-on arrangement of the HBC

molecules into the nanofibers,34 which, in such an ordered

system as the one described here, would put in contact with the

PCBM only the alkylic, insulating side chains of HBC. The

optical properties of these PCBM crystals will be tested with

other electron-donating molecules, different to HBC, to obtain

a more performing acceptor/donor interface.

Conclusions

Micron sized crystals of PCBM have been grown from

solution, by using a newly devised dipping method, on a

variety of substrates. These multilayered hexagonal crystals

were found to exhibit extraordinarily flat terraces extended

over several micrometres. The crystals are hexagonally shaped

and range between 1 to 80 mm in diameter and from 20 to

500 nm in thickness.

The hexagonal PCBM nanocrystals have been coated with

nanofibers of a strong electron-donor, i.e. a functionalized

hexa-peri-hexabenzocoronene (HBC), by spin coating using an

orthogonal solvent such as acetone. By using different solvent

for the HBC deposition, it was possible to obtain fibers

featuring a diverse width and/or thickness. The morphology

Fig. 6 (a) AFM topographic image of a PCBM crystal with HBC deposited from acetone z-range; 155 nm (b) gradient, z-range 1150 nm nm�1,

(c) KPFM image, z-range; 200 mV.

4478 | Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 This journal is �c the Owner Societies 2010

of the PCBM crystal was not affected by the solvent used

for the HBC deposition. Electronic characterization of the

electron–acceptor and –donor blend by KPFM showed a clear

potential difference between the PCBM and the HBC,

although no significant changes in surface potential, indicative

of charge generation at the PCBM :HBC interface, could be

devised up to now.

Due to their structural precision, the macroscopic PCBM

crystals exposing ultra-large and atomically flat terraces

represent model interfaces for both fundamental photovoltaic

studies and for the fabrication of solar cell prototypes featuring

the highest control over the structure of the interface.

Acknowledgements

We thank Dr Massimo Gazzano for performing the XRD

measurements. This work was supported by the ESF-SONS2-

SUPRAMATES, the NanoSci-E+ project SENSORS, the

EC Marie Curie IEF-HESPERUS (PIEF-GA-2008-219770)

and ITN-SUPERIOR (PITN-GA-2009-238177), the EC FP7

ONE-P large-scale project no. 212311, the Regione

Emilia-Romagna PRIITT Prominer Net-Lab, the Inter-

national Center for Frontier Research in Chemistry (FRC)

and the Deutsche Forschungsgemeinsachaft (Special Program

Organic Photovoltaics).

References

1 J. C. Hummelen, B. W. Knight, F. Lepeq, F. Wudl, J. Yao andC. L. Wilkins, J. Org. Chem., 1995, 60, 532–538.

2 J. Y. Kim, K. Lee, N. E. Coates, D. Moses, T. Q. Nguyen,M. Dante and A. J. Heeger, Science, 2007, 317, 222–225.

3 Z. Zhang, P. Han, X. Liu, J. Zhao, H. Jia, F. Zeng and B. Xu,J. Phys. Chem. C, 2008, 112, 19158–19161.

4 E. J. Meijer, D. L. D. M, S. Setayesh, E. Van Veenendaal,B. H. Huisman, P. W. M. Blom, J. C. Hummelen, U. Scherf andT. M. Klapwijk, Nat. Mater., 2003, 2, 678–682.

5 C. J. Brabec, F. Padinger, N. S. Sariciftci and J. C. Hummelen,J. Appl. Phys., 1999, 85, 6866–6872.

6 M. Jorgensen and F. C. Krebs, J. Org. Chem., 2005, 70, 6004–6017.7 A. Pivrikas, N. S. Sariciftci, G. Juska and R. Osterbacka, Progr.Photovolt.: Res. Appl., 2007, 15, 677–696.

8 S. Gunes, H. Neugebauer and N. S. Sariciftci, Chem. Rev., 2007,107, 1324–1338.

9 R. Kroon, M. Lenes, J. C. Hummelen, P. W. M. Blom and B. DeBoer, Polym. Rev., 2008, 48, 531–582.

10 B. R. Saunders andM. L. Turner, Adv. Colloid Interface Sci., 2008,138, 1–23.

11 X. N. Yang, J. K. J. van Duren, R. A. J. Janssen, M. A. J. Michelsand J. Loos, Macromolecules, 2004, 37, 2151–2158.

12 X. N. Yang, J. K. J. van Duren, M. T. Rispens, J. C. Hummelen,R. A. J. Janssen, M. A. J. Michels and J. Loos, Adv. Mater., 2004,16, 802–806.

13 H. Hoppe, T. Glatzel, M. Niggemann, A. Hinsch, M. C.Lux-Steiner and N. S. Sariciftci, Nano. Lett., 2005, 5, 269–274.

14 A. Swinnen, I. Haeldermans, M. vande Ven, J. D’Haen,G. Vanhoyland, S. Aresu, M. D’Olieslaeger and J. Manca, Adv.Funct. Mat., 2006, 16, 760–765.

15 M. Reyes-Reyes, R. Lopez-Sandoval, J. Arenas-Alatorre,R. Garibay-Alonso, D. L. Carroll and A. Lastras-Martinez, ThinSolid Films, 2007, 516, 52–57.

16 M. Campoy-Quiles, T. Ferenczi, T. Agostinelli, P. G. Etchegoin,Y. Kim, T. D. Anthopoulos, P. N. Stavrinou, D. D. C. Bradley andJ. Nelson, Nat. Mater., 2008, 7, 158–164.

17 L. Li, G. Lu, S. Li, H. Tang and X. Yang, J. Phys. Chem. B, 2008,112, 15651–15658.

18 M. Dante, J. Peet and T. Q. Nguyen, J. Phys. Chem. C, 2008, 112,7241–7249.

19 M. T. Rispens, A.Meetsma, R. Rittberger, C. J. Brabec, N. S. Sariciftciand J. C. Hummelen, Chem. Commun., 2003, 2116–2118.

20 S. Pekker, E. Kovats, G. Oszlanyi, G. Benyei, G. Klupp, G. Bortel,I. Jalsovszky, E. Jakab, F. Borondics, K. Kamaras, M. Bokor,G. Kriza, K. Tompa and G. Faigel, Nat. Mater., 2005, 4,764–767.

21 R. Ceolin, J. L. Tamarit, D. O. Lopez, M. Barrio, V. Agafonov,H. Allouchi, F. Moussa and H. Szwarc, Chem. Phys. Lett., 1999,314, 21–26.

22 M. Barrio, D. O. Lopez, J. L. Tamarit, P. Espeau, R. Ceolin andH. Allouchi, Chem. Mater., 2003, 15, 288–291.

23 R. Ceolin, D. O. Lopez, B. Nicolai, P. Espeau, M. Barrio,H. Allouchi and J. L. Tamarit, Chem. Phys., 2007, 342, 78–84.

24 M. Sathish and K. Miyazawa, J. Am. Chem. Soc., 2007, 129,13816.

25 V. Palermo and P. Samorı, Angew. Chem., Int. Ed., 2007, 46,4428–4432.

26 S. E. Shaheen, R. Radspinner, N. Peyghamgarian andG. E. Jabbour, Appl. Phys. Lett., 2001, 79, 2996–2998.

27 S. E. Shaheen, C. J. Brabec, N. S. Sariciftci, F. Padinger,T. Fromherz and J. C. Hummelen, Appl. Phys. Lett., 2001, 78,841–843.

28 G. De Luca, A. Liscio, P. Maccagnani, F. Nolde, V. Palermo,K. Mullen and P. Samorı, Adv. Funct. Mater., 2007, 17, 3791–3798.

29 E. Treossi, A. Liscio, X. L. Feng, V. Palermo, K. Mullen andP. Samorı, Small, 2009, 5, 112–119.

30 E. Treossi, A. Liscio, X. L. Feng, V. Palermo, K. Mullen andP. Samorı, Appl. Phys. A: Mater. Sci. Process., 2009, 95, 15–20.

31 X. Feng, W. Pisula, T. Kudernac, D. Wu, L. Zhi, S. De Feyter andK. Mullen, J. Am. Chem. Soc., 2009, 131, 4439–4448.

32 W. Kern and D. A. Puotinen, RCA Rev., 1970, 31, 187–206.33 G. Scavia, W. Porzio, S. Destri, L. Barba, G. Arrighetti, S. Militia,

L. Fumagalli, D. Natali and M. Sampietro, Surf. Sci., 2008, 602,3106–3115.

34 X. Feng, V. Marcon, W. Pisula, M. R. Hansen, J. Kirkpatrick,F. Grozema, D. Andrienko, K. Kremer and K. Mullen, Nat.Mater., 2009, 8, 421–426.

35 A. Liscio, V. Palermo, D. Gentilini, F. Nolde, K. Mullen andP. Samorı, Adv. Funct. Mater., 2006, 16, 1407–1416.

Fig. 7 PCBM crystal coated with HBC nanofibers, as deposited from chloroform–methanol 2 : 1 (v/v): (a) AFM, z-range; 800 nm. The edge of a

screw dislocation is visible in the lower part of the crystal; (b) gradient, z-range 3860 nm nm�1, (c) KPFM image, z-range; 80 mV.

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 | 4479

36 V. Palermo, M. Palma and P. Samorı, Adv. Mater., 2006, 18,145–164.

37 P. G. de Gennes, Rev. Mod. Phys., 1985, 57, 827–863.38 V. Palermo, G. Ridolfi, A. M. Talarico, L. Favaretto,

G. Barbarella, N. Camaioni and P. Samorı, Adv. Funct. Mater.,2007, 17, 472–478.

39 V. Palermo, M. B. J. Otten, A. Liscio, E. Schwartz, P. A. J. de Witte,M. A. Castriciano, M. M. Wienk, F. Nolde, G. De Luca, J. J. L. M.Cornelissen, R. A. J. Janssen, K. Mullen, A. E. Rowan, R. J. M.Nolte and P. Samorı, J. Am. Chem. Soc., 2008, 130, 14605–14614.

40 A. Liscio, G. De Luca, F. Nolde, V. Palermo, K. Mullen andP. Samorı, J. Am. Chem. Soc., 2008, 130, 780–781.

4480 | Phys. Chem. Chem. Phys., 2010, 12, 4473–4480 This journal is �c the Owner Societies 2010