Morphology control of nanofibril donor-acceptor heterojunction to achieve high photoconductivity: Exploration of new molecular design rule

Article abstract of DOI:10.1021/ja407024u

Donor-acceptor nanofibril composites have been fabricated, and the dependence of their photocurrent response on the structure and morphology of the donor part has been systematically investigated. The nanofibril composites were composed of template nanofibers, assembled from an electron acceptor molecule, perylene tetracarboxylic diimide (PTCDI), onto which (through drop-casting) various electron donor molecules (D1-D4) were coated. The donor molecules have the same π-conjugated core, but different side groups. Due to the different side groups, the four donor molecules showed distinctly different propensity for intermolecular aggregation, with D1-D3 forming segregated phases, while D4 prefers homogeneous molecular distribution within the film. It was found that the nanofibril composites with D4 exhibit the highest photocurrent, whereas those with aggregation-prone D1-D3 exhibited much lower photocurrent under the same illumination condition. Solvent annealing is found to further enhance the aggregation of D1-D3 but facilitate more uniform molecular distribution of D4 molecules. As a result, the photocurrent response of PTCDI fibers coated with D1-D3 decreased after vapor annealing, whereas those coated with D4 further increased. The detrimental effect of the aggregation of donor molecules on the PTCDI fiber is likely due to the enhanced local electrical field built up by the high charge density around the aggregate-nanofiber interface, which hinders the charge separation of the photogenerated electron-hole pair. The results reported in this study give further insight into the molecular structural effect on photoconductivity of hybrid materials, particularly those based on donor-acceptor composites or interfaces, and provide new molecular design rules and material processing guidelines to achieve high photoconductivity.

Full text of DOI:10.1021/ja407024u

Article
pubs.acs.org/JACS
Morphology Control of Nanofibril Donor−Acceptor Heterojunction
To Achieve High Photoconductivity: Exploration of New Molecular
Design Rule
Helin Huang,†,§ Ching-En Chou,‡,§ Yanke Che,† Ligui Li,† Chen Wang,† Xiaomei Yang,†
Zhonghua Peng,*,‡ and Ling Zang*,†
†Department of Materials Science and Engineering, University of Utah, 36 S. Wasatch Dr., Salt Lake City, Utah 84112, United States
‡Department of Chemistry, University of Missouri-Kansas City, Kansas City, Missouri 64110, United States
S
* Supporting Information
ABSTRACT: Donor−acceptor nanofibril composites have
been fabricated, and the dependence of their photocurrent
response on the structure and morphology of the donor part
has been systematically investigated. The nanofibril composites
were composed of template nanofibers, assembled from an
electron acceptor molecule, perylene tetracarboxylic diimide
(PTCDI), onto which (through drop-casting) various electron
donor molecules (D1−D4) were coated. The donor molecules
have the same π-conjugated core, but different side groups.
Due to the different side groups, the four donor molecules
showed distinctly different propensity for intermolecular aggregation, with D1−D3 forming segregated phases, while D4 prefers
homogeneous molecular distribution within the film. It was found that the nanofibril composites with D4 exhibit the highest
photocurrent, whereas those with aggregation-prone D1−D3 exhibited much lower photocurrent under the same illumination
condition. Solvent annealing is found to further enhance the aggregation of D1−D3 but facilitate more uniform molecular
distribution of D4 molecules. As a result, the photocurrent response of PTCDI fibers coated with D1−D3 decreased after vapor
annealing, whereas those coated with D4 further increased. The detrimental effect of the aggregation of donor molecules on the
PTCDI fiber is likely due to the enhanced local electrical field built up by the high charge density around the aggregate−
nanofiber interface, which hinders the charge separation of the photogenerated electron−hole pair. The results reported in this
study give further insight into the molecular structural effect on photoconductivity of hybrid materials, particularly those based on
donor−acceptor composites or interfaces, and provide new molecular design rules and material processing guidelines to achieve
high photoconductivity.
Recently, we have developed a novel approach based on
nanofibril heterojunction to achieve high photoconductivity for
organic materials.3 The approach relies on interfacial structural
modification of nanofibers fabricated from electron acceptor
(A) molecules, onto which a layer of electron donor (D)
molecules is coated. The A molecules are based on derivatives
of perylene tetracarboxylic diimide (PTCDI), which represent a
robust class of semiconductor materials with high thermal and
photo stability.25 The strong hydrophobic interdigitation
between the long alkyl side chains of D and A molecules
enables efficient charge transfer between the two under
photoexcitation. Such nanofibril heterojunctions possess two
prominent features that are critical for efficient photocurrent
generation: one is that the nanofibers both create a wide D/A
interface for increased charge separation and act as long-range
transport pathway for photogenerated charge carriers toward
the electrodes; the other is that the alkyl side chains employed
INTRODUCTION
■
Fabrication of effective donor−acceptor heterojunction struc-
tures remains critical for developing high-performance organic
photovoltaics and the relevant photoelectric switches and
sensors.1−5 However, most of the research efforts to date have
been focused on bulk heterojunctions, which are formed
through the not-easily controlled phase separation process.6−12
More ordered heterojunction structures, particularly those
controllable at the nanometer scale through molecular design
and engineering, are desirable in order to achieve high
photoconversion efficiency.13 Self-assembly has proven to be
an efficient bottom-up method to construct well-defined
nanostructures that may afford high photoconductivity.14−16
A few examples have been reported on the design and
fabrication of nanostructured heterojunction systems through
molecular self-assembly.3,17−24 Most of these systems, however,
either form well-ordered structure only on a small area or
require complicated molecular synthesis and thus are not suited
for practical applications.
Received: July 16, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja407024u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
not only enable effective surface adsorption of D molecules on
the nanofibers for effective electron-transfer communication
but also facilitate spatial separation the photogenerated charge
carriers to prevent their recombination. The surface coating
method thus developed represents a simple, adaptable method
that will allow for further improvement of organic photo-
conductivity through molecular design and supramolecular
engineering. In the present work, we aim to explore the effects
of aggregation of D molecules on the photoconductivity of
nanofibril heterojunctions. A series of D molecules with varying
side-chain modifications were synthesized26 and investigated for
the different intermolecular arrangements caused by π−π
stacking in balance with steric hindrance of side chains.
Interestingly, it was observed that the different molecular
assemblies of D resulted in distinctive phase segregation
between D and A (the nanofiber), which significantly affects the
interfacial charge transfer and separation as indicated from the
measurements of fluorescence quenching and photocurrent
generation.
Side group modification has proven to be an effective way for
tuning the intermolecular interactions, leading to different
aggregated morphologies and electronic properties.28 Such
molecular structural effect can be more clearly manifested
under solvent vapor annealing, which facilitates the self-
assembly of molecules to reach the thermodynamically stable
(energy minimal) state by removing the grain boundaries
formed during the fast solvent vaporization process. Indeed,
solvent vapor annealing has been extensively studied for organic
semiconductor materials and devices in order to improve the
crystallinity and charge carrier mobility.29−32 In this study, we
adopted this annealing process to facilitate the assembly of D
molecules so that the effect of peripheral side group
modification can be revealed more illustriously. Among the
four D molecules, D4 is expected to have the most severe steric
hindrance for π−π stacking due to its bulkier side groups. The
other three molecules on the other hand possess relatively small
peripheral groups and are thus prone to π−π staking
aggregation. As shown below, the different extent of
intermolecular aggregation in the four D molecules has a
dramatic effect on the photocurrent response of the core
PTCDI nanofibers. With the coating of D4, which shows no
π−π stacking even after solvent annealing, the PTCDI
nanofibers exhibit the highest photocurrent. When coated
with D1, D2, and D3 molecules, which are prone to
aggregation, the photocurrent response is significantly lower.
The difference is more dramatic for the D−A nanofibril
heterojunctions after solvent vapor annealing.
RESULTS AND DISCUSSION
Nanofibril Heterojunction Fabrication. Scheme 1 shows
the structures of the four D molecules employed in this study,
■
Scheme 1. Schematic Illustration of Core-Shell Structured
Nanofibril Heterojunction Composed of D1−D4 Coating
and PTCDI Nanofiber
The nanofiber of A was fabricated from a PTCDI molecule
(Scheme 1) following the previously published self-assembly
protocol.27 The PTCDI molecule has a dodecyl chain at both
imide positions to ensure strong hydrophobic alkyl chain
interdigitation with the D molecules. The nanofibers were
subsequently surface-coated with D molecules using the drop
casting method previously employed for creating similar
nanofibril heterojunctions.3 The as-prepared nanofibers possess
well-defined morphologies and large aspect ratios, providing a
large interface for surface adsorption of D molecules. Figure 1
which are based on the same tribenzopentaphene (TBP) core.
The rigid, planar π-conjugated TBP core is conducive for
cofacial π−π stacking that often leads to the formation of a one-
dimensional molecular assembly. The highest occupied
molecular orbital (HOMO) of TBP molecules is significantly
higher than that of PTCDI, providing sufficient driving force
for the photoinduced TBP to PTCDI electron transfer. All four
TBP molecules (D1−D4) have two dodecyl chains attached at
the same positions. These linear alkyl chains help anchoring the
D molecules onto the nanofiber of A through hydrophobic alkyl
chain interdigitation as observed before.3,25,27 At the periphery
of the TBP core, different side chains were attached for D1−D4
so that the intermolecular arrangement between TBP cores can
be modulated, whereas the interfacial interaction between
PTCDI and the four D molecules is expected to be comparable
due to their identical dodecyl anchoring chains. Systematic
investigation on how the intermolecular aggregation of the
coated D molecules affect the photocurrent response of the
PTCDI nanofibers would provide deeper understanding of the
photoinduced charge separation process at the nanofibril
heterojunction and open more options to further increase the
photocurrent through structural optimization of the D
component.
Figure 1. AFM images of PTCDI nanofibers (molecular amount of 7
nmol) deposited on a 5 mm × 5 mm silicon wafer covered with 300-
nm thick SiO2 before (a) and after (b) surface coating of D4 (4 nmol).
shows the in situ AFM images of PTCDI nanofibers before and
after surface coating of D4 molecules. Almost all of the D
molecules are coated onto the surface of nanofibers after drop
casting, leaving little residue on the silicon oxide substrate.
Similar phenomena were observed for other three D molecules,
D1−D3 (Supplementary Figure S1). The clean and selective
deposit of D molecules onto the PTCDI fibers is attributed to
the strong hydrophobic interdigitation between the dodecyl
chains of D and that of PTCDI molecules on the nanofiber
surface. Meanwhile, the morphology of individual nanofibers
B
dx.doi.org/10.1021/ja407024u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
and the intertwining configuration remain unchanged after
drop casting, indicating the robustness of nanofibril structures
on the substrate. Such a feature enables comparative studies of
PTCDI nanofibers modified with different D molecules,
allowing the pinpointing of different photocurrent responses
on the surface-coated D molecules.
Effect of Intermolecular Aggregation Photoconduc-
tivity. Figure 2a shows the fluorescence spectra of the PTCDI
Figure 2. (a) Fluorescence spectra of PTCDI nanofibers shown in
Figure 1 before (black) and after (red) drop-casting of D4. (b) I−V
curves measured over the D4-coated PTCDI nanofibers in the dark
(black) and under white light irradiation of 0.17 mW/mm2 (red).
Efficient fluorescence quenching of PTCDI nanofibers by coating of
D4 was also confirmed with fluorescence microscopy imaging
(Supplementary Figure S12). Similar fluorescence quenching was
also observed for the coating of D1−D3 molecules (Supplementary
Figures S5 and S13).
Figure 3. (Top) Comparison of fluorescence quenching (red) and
photocurrent generation (blue) between the PTCDI nanofibers coated
with the four donor molecules, D1−D4. Fluorescence quenching and
photocurrent measurements were conducted under the same
conditions as employed in Figure 2. Photocurrent values used in
this plot were obtained at a bias voltage of 10 V. Example fluorescence
quenching and I−V curves measured over D1−D4 are shown in
Figure 2 and Supplementary Figures S5 and S6. (Bottom) Electronic
energy levels of PTCDI and D1−D4. The HOMO levels of D1−D4
were determined by cyclic voltammetry using Fc/Fc+ as an internal
standard.26 The energy levels of PTCDI (A) and D1−D4 (in
parentheses) were calculated with density-functional theory (B3LYP/
6-311g**//B3LYP/6-31g*) using the Gaussian 09 package.
nanofibers before and after the coating of D4 molecules, which
indicates that 80% of the PTCDI fluorescence was quenched
after D4 coating. The significant fluorescence quenching is due
to the forward electron transfer from D4 molecules to the
photoexcited PTCDIs. This photoinduced electron transfer is
thermodynamically favored with a driving force of 1.13 eV,
calculated between the highest occupied molecular orbital
(HOMO) levels of D4 and PTCDI molecules (Figure 3).
Consistent with the efficient fluorescence quenching, high
photoconductivity was observed for the D4/PTCDI nanofibril
heterojunction (Figure 2b), for which an on/off ratio of ca. 104
was obtained under a 10 V bias. The high photoconductivity
was due to an optimal balance between the forward and back
electron transfer between D4 coating and the PTCDI
nanofibers as discussed below. It is possible that the increased
photocurrent may be due to the conductivity of the D4 coating
itself. However, this possibility was excluded since the dark and
photo current of D4 film under the same light illumination
(Supplementary Figure S7) was only around 1 pA under 10 V
bias, which is 3 orders of magnitude lower than that of D4/
PTCDI nanofibers (Figure 2b). Moreover, the D4 film does
not demonstrate photoconductivity response, i.e., the I−V
curves obtained under dark and light illumination remained
almost the same (Supplementary Figure S7).
The same experiments shown in Figure 2 were also
performed on the PTCDI nanofibers coated with other three
D molecules, D1−D3, in order to investigate the effect of
molecular structure of D on the photoconductivity of D/A
nanofibril heterojunctions. Figure 3 shows the results of
fluorescence quenching and photocurrent measurements
performed on the four different nanofibril heterojunctions.
While all four nanofibers showed comparable fluorescence
quenching after D-coating, the photocurrents generated were
dramatically different, with the D4/PTCDI nanofiber giving the
highest photocurrent. The efficient fluorescence quenching is
consistent with the energetically favorable forward electron
transfer from the D molecule to the photoexcited A, for which
the driving force was calculated to be in the range of 0.7−1.1
eV (Figure 3). However, efficient forward electron transfer does
not necessarily produce high electrical current, which also
depends on other factors such as the subsequent charge
separation of the photogenerated D+−A− pair to free charge
carriers and the charge carrier mobility. Charge recombination
(or back electron transfer) within D+−A− pair is often one of
the major reasons for low photocurrent.
It is particularly intriguing to compare D3 and D4. Both have
alkoxy side groups and very close HOMO levels. When coated
on PTCDI fibers, they exhibit an identical extent of
fluorescence quenching. The photocurrent response of the
two heterojunctions, however, differs by almost five times. The
difference in photocurrent was even more dramatic for the
nanofibers coated with D1 and D2, which showed photo-
current values more than 10 and 50 times, respectively lower
than that of D4-coated fibers (Figure 3). Considering the
similar fluorescence quenching efficiency among the four
nanofibers, we suspected that the lower photocurrent observed
with the nanofibers coated with D1−D3 (compared to that of
D4) was primarily due to the less efficient charge separation (or
faster charge recombination) of the photogenerated D+−A−
pair, which has something to do with the phase segregation of
surface-coated D molecules. The bulky side group of D4
prevents intermolecular π−π stacking, resulting in uniform
molecular distribution of D4 onto the PTCDI nanofibers. In
C
dx.doi.org/10.1021/ja407024u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. KPFM measurement of D3/PTCDI nanofibril heterojunction. (a) Topograhy image of the nanofibril heterojunction (Z height range 400
nm). PTCDI nanofiber, D3 coating, and isolated D3 aggregate are marked as domains I, II, and III, respectively. (b, c) Line scan surface potential
profiles obtained by tracing the corresponding lines marked (red, blue) in the dark (black plot) and under white light illumination of 10 mW/cm2
(red plot).
contrast, D1−D3, with relatively small side groups, are prone to
π−π stacking, which leads to the formation of segregated
phases of aggregated D molecules. AFM images of the drop-
cast films of D1−D3 confirmed their self-assembly. Particularly
for D1 and D3, the preferred columnar stacking enabled
formation of nanofibril structures (Supplementary Figures S2
and S10).
Results in Figure 3 suggest that aggregates of D1−D3 can
quench the fluorescence of PTCDI nanofiber in similar (or
comparable) efficiency as the homogeneous coating of D4. This
can be explained by the long distance exciton migration within
the single crystalline nanofiber of PTCDI, where the exciton
migration distance was in the range of hundreds of nanome-
ters.25,33 Once the quenchers (molecules or aggregates) are
distributed on the surface with sufficient density (i.e., spatially
separated within the distance of exciton migration), the
fluorescence of nanofibers can be effectively quenched, as an
exciton can always encounter a quencher when migrating along
the nanofiber within its lifetime.34 Considering the high molar
ratio (4/7) of D/A used in this study, it is reasonable to assume
that the aggregates of D1−D3 can cover the PTCDI nanofiber
with high enough density and the separation between the
aggregates is significantly smaller than the exciton migration
distance of PTCDI.
delocalization along the nanofiber, leading to the generation of
high photocurrent as shown in Figure 3.
It is interesting to note that among the three D molecules
(D1−D3) that produced relatively low photocurrent (Figure
3), D2 demonstrated the lowest current value, significantly
lower than the other two, although the fluorescence quenching
efficiency of D2 is even the highest. This additional lowering of
photoconductivity is likely due to the hydroxyl side groups of
D2 (−OH), which affords redox interaction with electrons,
functioning as a charge carrier trap as previously observed in n-
type organic field effect transistors.38−40
Surface Potential of Nanofibril Heterojunction under
Illumination. To gain further insight into the photoinduced
charge separation process within the nanofibril heterojunction
of PTCDI, Kelvin probe force microscopy (KPFM) was used to
image the surface potential change upon illumination in a
similar way as previously performed on other organic
heterojunctions.41,42 D3 was chosen as a representative of the
three donors (D1−D3) to form D/A heterojunction with
PTCDI; these three donors prefer to form aggregates, which
are easy to be imaged by AFM and distinguished from PTCDI
fiber. The D3/PTCDI nanofibers were deposited onto an
indium-doped tin oxide (ITO) coated glass slide, and measured
simultaneously for topography and surface potential images
both in the dark and under white light illumination.
Although the aggregation of D1−D3 does not affect
fluorescence quenching efficiency, it caused significant decrease
in photocurrent compared to the case of D4. This is likely due
to the locally enhanced recombination (rather than separation)
of the D+−A− charge pair photogenerated around the aggregate
of D. The interfacial charge separation of D+−A− at the PTCDI
nanofiber is primarily caused by the intermolecular electron
delocalization among the π−π stacked PTCDI molecules,3,5,35
and this charge delocalization can be further extended under
electrical bias applied to the nanofiber. At the fibril section with
an aggregated D domain, there may be multiple pairs of D+−A−
photogenerated by quenching multiple excitons (as discussed
above), making the electron delocalization along the PTCDI
stacks more difficult due to the increased charge density.36,37
Moreover, multiple charge pairs form a strong local electrical
field, which can prevent the charge separation. These two
effects result in enhanced charge recombination of D+−A−. In
contrast, the homogeneous coating of D4 produces more
uniform distribution of D+−A− pairs on the nanofiber, which
can be effectively separated through the intermolecular electron
As shown in Figure 4a, well-defined morphology was
observed for the PTCDI nanofiber with surface coating of
D3. Figure 4b shows the line scan profile of surface potential
corresponding to the red line marked in Figure 4a. In the dark,
the surface potential PTCDI fiber (domain I as marked)
appears higher than that of D3 aggregate (domain II),
consistent with their HOMO level difference as shown in
Figure 3. Upon white light illumination, PTCDI gets excited,
initiating electron transfer from D3 to the fiber. This charge
separation was imaged as an increase in surface potential of the
D3 phase, but a decrease for the PTCDI fiber (red plot in
Figure 4b). For comparison, the same KPFM measurements
were also performed on a bare D3 aggregate (domain III), for
which no surface potential change was observed before and
after white illumination (Figure 4c). This is simply because that
D3 can barely be excited by visible light (Figure S3) and thus
no charge separation can be initiated.
Solvent-Annealing Influence on Photoconductivity. It
is well-known that fast evaporation of solvents during the drop-
D
dx.doi.org/10.1021/ja407024u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
casting process rarely produces the thermodynamically stable
(energy minimized) state with regard to the intermolecular
arrangement and phase growth within organic materials. Large
number of defects and grain boundaries can be formed during
the rapid assembly of molecules, and the growth of small nuclei
may be quenched when the solvent is dried. Postassembly
treatment, like solvent vapor annealing (i.e., aging of organic
materials in saturated solvent vapor), has proven to be an
effective method to reorganize and optimize the intermolecular
arrangement, and facilitate the phase growth to reach the
thermodynamic stable state.29,43−46 Indeed, solvent vapor
annealing has been commonly used in solar cell materials
processing to improve the crystalline organization of D and A
phases, with the aim to facilitate the charge transporta-
tion.9,47−53 We expected that the above-mentioned difference
of photocurrent between the nanofibers coated with D4 and
D1−D3 would be more profound if the nanofibers were subject
to solvent vapor annealing, considering that the difference of
phase morphology and its dependence on the molecular
structure between D4 and D1−D3 can be maximized at the
thermodynamic stable state.
D3 was chosen as a representative of D1−D3 to study the
effects of solvent vapor annealing on the surface aggregation
and the subsequent photoconductivity; the results were
compared to those observed with D4. The annealing was
performed in ethanol vapor, as ethanol is a good solvent for
D1−D4, but a poor one for the PTCDI. During the annealing
process the crystalline structure of PTCDI nanofibers was
expected to remain intact. Under the same annealing
conditions, D3 and D4 coatings underwent dramatically
different transition of the aggregation state. For D3, optical
microscopy imaging showed the growth of small aggregates
into large needle like structures (Figure S10). AFM imaging
revealed that the small aggregates of D3 are actually composed
of fine nanofibril structures, which were formed during the
drop-cast process. Formation of nanofibril structure is
indicative of the favorable columnar π−π stacking of D3,
which is consistent with planar geometry of TBP π-conjugation.
Upon solvent vapor annealing these small nanofibers grew into
needle like chunks, though still remaining in an elongated shape
(Figure S10). This observation indicates that D3 is prone to
crystallize into large elongated structures driven by the strong
π−π stacking interaction. The D4 coating on the other hand
showed no obvious aggregation even after solvent vapor
annealing (Figure S11).
The different behavior of D3 and D4 under solvent vapor
annealing led to distinct effect on the photocurrent response as
measured over the PTCDI nanofibers coated with these two
molecules (Figure 5a). For D3, the photocurrent decreased to
14% of its preannealing value, whereas for D4, the photocurrent
increased by 60%. The significant decrease in photoconductiv-
ity of D3/PTCDI nanofiber is likely due to the increased
aggregation of D3 induced by the solvent vapor annealing. As
discussed above, the enlarged aggregated domains of D
molecules are detrimental to the charge separation of D+−A−
pairs due to the enhanced local electrical field. In contrast, for
the D4/PTCDI nanofiber the bulky isopropoxyl substitution at
D4 prevents molecular aggregation and instead favors the
homogeneous distribution of molecules on the surface. Such
molecular distribution can be further facilitated under solvent
vapor annealing, producing a coating layer with minimized
aggregation, as indicated by the 60% increase in photocurrent
shown in Figure 5. This annealing-enhanced molecular
Figure 5. (a) I−V curves measured over D4-coated PTCDI nanofibers
under the same light irradiation as employed in Figure 2 before (black)
and after (red) solvent vapor annealing. Inset: relative photocurrent
change (in percentage) for D3- and D4-coated nanofibers after solvent
vapor annealing. Photocurrent values used in this plot were obtained at
a bias voltage of 10 V. Example I−V curves for D3- and D4-coated
nanofibers are shown in Supplementary Figure S8. (b) Fluorescence
spectra of D4 drop-cast on a glass surface modified with trichloro-
(octadecyl)silane (OTS) before (black) and after (red) solvent vapor
annealing. The two spectra are normalized at 465 nm.
distribution of D4 was supported by the solid-state fluorescence
spectral measurement on a thin D4 film drop-cast on a glass
slide. The glass slide was pretreated with trichloro(octadecyl)-
silane (OTS) to generate a hydrophobic surface similar to that
of the PTCDI nanofiber. Before solvent vapor annealing, the
fluorescence spectrum of D4 film possesses a major peak at 465
nm, along with a shoulder peak at a shorter wavelength, 440
nm. Compared to the fluorescence spectrum of D4 molecularly
dispersed in ethanol (Supplementary Figure S3), the
fluorescence emission in the solid state becomes more
dominant at longer wavelength (Figure 5b), consistent with
the enhanced intermolecular interaction as commonly observed
for molecular assemblies. The 440 nm emission (characteristic
of the fluorescence of individual molecules) remains in the solid
film of D4, indicating the relatively weak intermolecular π−π
interaction in comparison with the film of D3, where the strong
π−π stacking results in almost no emission observed for the
individual molecules (Supplementary Figures S3 and S4). After
solvent vapor annealing, the 440 nm peak of D4 film was
significantly enhanced (Figure 5b), implying more molecules
originally “frozen” as aggregates during the fast evaporation of
drop-casting now transformed into homogeneous molecular
distribution. In contrast, for the D3 film there was no obvious
change observed in the fluorescence spectrum upon solvent
vapor annealing under the same condition (Supplementary
Figure S9). This is consistent with the strong intermolecular
stacking, which leads to the formation of stable aggregates in
the specific nanofibril morphology (Supplementary Figure
S10). Although solvent vapor annealing facilitates the growth of
the nanofibers of D3 into larger elongated crystals, the
electronic property of the solid phase still remains dominant
with the π−π stacking as indicated by the unchanged
fluorescence spectra.
CONCLUSIONS
■
In summary, we have investigated a series of electron donor
molecules (D1−D4) that share the same π-conjugation core,
but which are modified with different side groups. PTCDI
fibers coated with such donor molecules showed dramatically
different photocurrent response. It was found that the
nanofibers coated with homogeneously and molecularly
distributed donor molecules (such as D4) exhibit the highest
photocurrent, whereas those coated with segregated donor
E
dx.doi.org/10.1021/ja407024u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(12) He, Z.; Zhong, C.; Huang, X.; Wong, W.-Y.; Wu, H.; Chen, L.;
Su, S.; Cao, Y. Adv. Mater. 2011, 23, 4636−4643.
(13) Chang, C.-Y.; Wu, C.-E.; Chen, S.-Y.; Cui, C.; Cheng, Y.-J.; Hsu,
C.-S.; Wang, Y.-L.; Li, Y. Angew. Chem., Int. Ed. 2011, 50, 9386−9390.
(14) Hoeben, F. J. M.; Jonkheijm, P.; Meijer, E. W.; Schenning, A. P.
H. J. Chem. Rev. 2005, 105, 1491−1546.
(15) De Greef, T. F. A.; Smulders, M. M. J.; Wolffs, M.; Schenning,
A. P. H. J.; Sijbesma, R. P.; Meijer, E. W. Chem. Rev. 2009, 109, 5687−
5754.
(16) Aida, T.; Meijer, E. W.; Stupp, S. I. Science 2012, 335, 813−817.
(17) Percec, V.; Glodde, M.; Bera, T. K.; Miura, Y.; Shiyanovskaya, I.;
Singer, K. D.; Balagurusamy, V. S. K.; Heiney, P. A.; Schnell, I.; Rapp,
A.; Spiess, H. W.; Hudson, S. D.; Duan, H. Nature 2002, 417, 384−
387.
aggregates (such as D1−D3) show much lower photocurrent
under the same illumination conditions. The aggregation of
donor molecules on the surface of the PTCDI fibers may lead
to the buildup of a local electrical field that hinders the charge
separation of the photogenerated electron−hole pairs. The
different morphologies of molecular aggregates were mostly the
result of side group modification of the donor molecules. Such
structural effect was more clearly manifested by investigating
the structure and morphology change of the drop-cast films
upon solvent vapor annealing. The findings presented provide
new insight into the molecular structural effect on photo-
conductivity of organic semiconductor materials, particularly
those based on donor−acceptor composites or interfaces, and
open alternative ways to improve the photoconductivity by
structural design and modification.
(18) Wurthner, F.; Chen, Z.; Hoeben, F. J. M.; Osswald, P.; You, C.-
̈
C.; Jonkheijm, P.; Herrikhuyzen, J. v.; Schenning, A. P. H. J.; van der
Schoot, P. P. A. M.; Meijer, E. W.; Beckers, E. H. A.; Meskers, S. C. J.;
Janssen, R. A. J. J. Am. Chem. Soc. 2004, 126, 10611−10618.
(19) Yamamoto, Y.; Fukushima, T.; Suna, Y.; Ishii, N.; Saeki, A.; Seki,
S.; Tagawa, S.; Taniguchi, M.; Kawai, T.; Aida, T. Science 2006, 314,
1761−1764.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental methods and materials, additional spectral
measurements, optical microscopy and AFM images. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(20) Sisson, A. L.; Sakai, N.; Banerji, N.; Furstenberg, A.; Vauthey, E.;
̈
Matile, S. Angew. Chem., Int. Ed. 2008, 47, 3727−3729.
(21) Yamamoto, Y.; Zhang, G.; Jin, W.; Fukushima, T.; Ishii, N.;
Saeki, A.; Seki, S.; Tagawa, S.; Minari, T.; Tsukagoshi, K.; Aida, T.
Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 21051−21056.
(22) Sakai, N.; Bhosale, R.; Emery, D.; Mareda, J.; Matile, S. J. Am.
Chem. Soc. 2010, 132, 6923−6925.
AUTHOR INFORMATION
■
Corresponding Authors
pengz@umkc.edu
(23) Zhang, W.; Jin, W.; Fukushima, T.; Saeki, A.; Seki, S.; Aida, T.
Science 2011, 334, 340−343.
(24) Sakai, N.; Matile, S. J. Am. Chem. Soc. 2011, 133, 18542−18545.
(25) Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596−
1608.
lzang@eng.utah.edu
Author Contributions
§These authors contributed equally to this work.
(26) Chou, C.-E.; Li, Y.; Che, Y.; Zang, L.; Peng, Z. RSC Adv. 2013,
Notes
3, 20666−20672.
The authors declare no competing financial interest.
(27) Balakrishnan, K.; Datar, A.; Naddo, T.; Huang, J.; Oitker, R.;
Yen, M.; Zhao, J.; Zang, L. J. Am. Chem. Soc. 2006, 128, 7390−7398.
ACKNOWLEDGMENTS
(28) Nguyen, L. H.; Hoppe, H.; Erb, T.; Gunes, S.; Gobsch, G.;
̈
■
Sariciftci, N. S. Adv. Funct. Mater. 2007, 17, 1071−1078.
(29) Datar, A.; Oitker, R.; Zang, L. Chem. Commun. 2006, 1649−
1651.
This work was supported by DHS (2009-ST-108-LR0005) and
NSF (CHE 0931466). Z.P. acknowledges the support of NSF
(DMR1308577) and the Army Research Office (W911NF-10-
1-0476).
(30) Dickey, K. C.; Anthony, J. E.; Loo, Y. L. Adv. Mater. 2006, 18,
1721−1726.
(31) Lu, G.; Li, L.; Yang, X. Small 2008, 4, 601−606.
(32) Chen, H.; Hu, S.; Zang, H.; Hu, B.; Dadmun, M. Adv. Funct.
Mater. 2013, 23, 1701−1710.
REFERENCES
■
(1) Gunes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007, 107,
̈
(33) Chaudhuri, D.; Li, D.; Che, Y.; Shafran, E.; Gerton, J. M.; Zang,
L.; Lupton, J. M. Nano Lett. 2011, 11, 488−492.
(34) Naddo, T.; Che, Y.; Zhang, W.; Balakrishnan, K.; Yang, X.; Yen,
M.; Zhao, J.; Moore, J. S.; Zang, L. J. Am. Chem. Soc. 2007, 129, 6978−
6979.
1324−1338.
́
(2) Thompson, B. C.; Frechet, J. M. J. Angew. Chem., Int. Ed. 2008,
47, 58−77.
(3) Che, Y.; Huang, H.; Xu, M.; Zhang, C.; Bunes, B. R.; Yang, X.;
Zang, L. J. Am. Chem. Soc. 2011, 133, 1087−1091.
(4) Konstantatos, G.; Sargent, E. H. Nat. Nanotechnol. 2010, 5, 391−
400.
(5) Che, Y.; Yang, X.; Liu, G.; Yu, C.; Ji, H.; Zuo, J.; Zhao, J.; Zang, L.
J. Am. Chem. Soc. 2010, 132, 5743−5750.
(6) Halls, J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.;
Friend, R. H.; Moratti, S. C.; Holmes, A. B. Nature 1995, 376, 498−
500.
(35) Che, Y.; Datar, A.; Yang, X.; Naddo, T.; Zhao, J.; Zang, L. J. Am.
Chem. Soc. 2007, 129, 6354−6355.
(36) Delgado, M. C. R.; Kim, E.-G.; Filho, D. t. A. d. S.; Bredas, J.-L.
J. Am. Chem. Soc. 2010, 132, 3375−3387.
(37) Coropceanu, V.; Cornil, J.; da Silva Filho, D. A.; Olivier, Y.;
́
Silbey, R.; Bredas, J.-L. Chem. Rev. 2007, 107, 926−952.
(38) Chua, L.-L.; Zaumseil, J.; Chang, J.-F.; Ou, E. C. W.; Ho, P. K.
H.; Sirringhaus, H.; Friend, R. H. Nature 2005, 434, 194−199.
(39) Yoon, M.-H.; Kim, C.; Facchetti, A.; Marks, T. J. J. Am. Chem.
Soc. 2006, 128, 12851−12869.
(7) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science
1995, 270, 1789−1791.
(8) Peumans, P.; Uchida, S.; Forrest, S. R. Nature 2003, 425, 158−
(40) Baeg, K.-J.; Facchetti, A.; Noh, Y.-Y. J. Mater. Chem. 2012, 22,
21138−21143.
162.
(9) Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery, K.;
Yang, Y. Nat. Mater. 2005, 4, 864−868.
(10) Kim, J. Y.; Lee, K.; Coates, N. E.; Moses, D.; Nguyen, T.-Q.;
Dante, M.; Heeger, A. J. Science 2007, 317, 222−225.
(11) Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J.
S.; Moses, D.; Leclerc, M.; Lee, K.; Heeger, A. J. Nat. Photonics 2009,
3, 297−302.
(41) Liscio, A.; De Luca, G.; Nolde, F.; Palermo, V.; Mullen, K.;
̈
Samorì, P. J. Am. Chem. Soc. 2008, 130, 780−781.
(42) Palermo, V.; Otten, M. B. J.; Liscio, A.; Schwartz, E.; de Witte,
P. A. J.; Castriciano, M. A.; Wienk, M. M.; Nolde, F.; De Luca, G.;
Cornelissen, J. J. L. M.; Janssen, R. A. J.; Mullen, K.; Rowan, A. E.;
̈
Nolte, R. J. M.; Samorì, P. J. Am. Chem. Soc. 2008, 130, 14605−14614.
F
dx.doi.org/10.1021/ja407024u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(43) Campoy-Quiles, M.; Ferenczi, T.; Agostinelli, T.; Etchegoin, P.
G.; Kim, Y.; Anthopoulos, T. D.; Stavrinou, P. N.; Bradley, D. D. C.;
Nelson, J. Nat. Mater. 2008, 7, 158−164.
(44) De Luca, G.; Liscio, A.; Battagliarin, G.; Chen, L.; Scolaro, L.
M.; Mullen, K.; Samorì, P.; Palermo, V. Chem. Commun. 2013, 49,
̈
4322−4324.
(45) De Luca, G.; Liscio, A.; Maccagnani, P.; Nolde, F.; Palermo, V.;
Mullen, K.; Samorì, P. Adv. Funct. Mater. 2007, 17, 3791−3798.
̈
(46) De Luca, G.; Liscio, A.; Nolde, F.; Scolaro, L. M.; Palermo, V.;
Mullen, K.; Samorì, P. Soft Matter 2008, 4, 2064−2070.
̈
(47) Hoppe, H.; Sariciftci, N. S. J. Mater. Chem. 2006, 16, 45−61.
(48) Yang, X.; Loos, J.; Veenstra, S. C.; Verhees, W. J. H.; Wienk, M.
M.; Kroon, J. M.; Michels, M. A. J.; Janssen, R. A. J. Nano Lett. 2005, 5,
579−583.
(49) Kim, H. J.; Lee, H. H.; Kim, J.-J. Macromol. Rapid Commun.
2009, 30, 1269−1273.
(50) Zhao, Y.; Xie, Z.; Qu, Y.; Geng, Y.; Wang, L. Appl. Phys. Lett.
2007, 90, 043504.
(51) Jo, J.; Na, S.-I.; Kim, S.-S.; Lee, T.-W.; Chung, Y.; Kang, S.-J.;
Vak, D.; Kim, D.-Y. Adv. Funct. Mater. 2009, 19, 2398−2406.
(52) Bavel, S. S. v.; Sourty, E.; With, G. d.; Loos, J. Nano Lett. 2008,
9, 507−513.
(53) Li, G.; Yao, Y.; Yang, H.; Shrotriya, V.; Yang, G.; Yang, Y. Adv.
Funct. Mater. 2007, 17, 1636−1644.
G
dx.doi.org/10.1021/ja407024u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX