Letter pubs.acs.org/NanoLett
Atomically Thin Epitaxial Template for Organic Crystal Growth Using Graphene with Controlled Surface Wettability Nguyen Ngan Nguyen, Sae Byeok Jo, Seong Kyu Lee, Dong Hun Sin, Boseok Kang, Hyun Ho Kim, Hansol Lee, and Kilwon Cho* Department of Chemical Engineering Pohang University of Science and Technology Pohang 790-784 Korea S Supporting Information *
ABSTRACT: A two-dimensional epitaxial growth template for organic semiconductors was developed using a new method for transferring clean graphene sheets onto a substrate with controlled surface wettability. The introduction of a sacrificial graphene layer between a patterned polymeric supporting layer and a monolayer graphene sheet enabled the crack-free and residue-free transfer of free-standing monolayer graphene onto arbitrary substrates. The clean graphene template clearly induced the quasi-epitaxial growth of crystalline organic semiconductors with lying-down molecular orientation while maintaining the “wetting transparency”, which allowed the transmission of the interaction between organic molecules and the underlying substrate. Consequently, the growth mode and corresponding morphology of the organic semiconductors on graphene templates exhibited distinctive dependence on the substrate hydrophobicity with clear transition from lateral to vertical growth mode on hydrophilic substrates, which originated from the high surface energy of the exposed crystallographic planes of the organic semiconductors on graphene. The optical properties of the pentacene layer, especially the diffusion of the exciton, also showed a strong dependency on the corresponding morphological evolution. Furthermore, the effect of pentacene−substrate interaction was systematically investigated by gradually increasing the number of graphene layers. These results suggested that the combination of a clean graphene surface and a suitable underlying substrate could serve as an atomically thin growth template to engineer the interaction between organic molecules and aromatic graphene network, thereby paving the way for effectively and conveniently tuning the semiconductor layer morphologies in devices prepared using graphene. KEYWORDS: Graphene, pentacene, wetting transparency, organic semiconductor, epitaxial growth template
G
been only achieved by chemical vapor deposition (CVD) method on transition metal catalyst.2,3,26−28 However, after the CVD process the synthesized graphene sheet generally needs to be placed onto a target substrate by transfer processes incorporating extra supporting layers directly contacting the graphene surface, which, unfortunately, leaves unavoidable cracks and polymeric contaminants.29−35 Such defects directly interact with graphene to induce accidental doping of graphene and strongly interfere with the interaction between graphene and adsorbed organic molecules, diminishing the template effect of graphene by inducing the growth of randomly or unintentionally oriented organic crystals.29−34 In this sense, it is of high priority that a residue-free and defect-free transfer method of graphene be developed to successfully demonstrate the 2D epitaxial growth template based on graphene sheets. This work addresses the issues on the use of graphene as a controllable epitaxial template for organic crystal growth by
raphene, an atomically thin material, displays fascinating electrical, mechanical, and optical properties.1−12 In particular, the sp2 network of carbon atoms enables graphene to act as a good template for assembling organic and inorganic crystals on its surface.13−19 The extraordinary thickness of monolayer graphene, however, raises issues about the correlations between the properties of graphene surfaces and the underlying substrate.20,21 Recently, the report on the “wetting transparency” of graphene sparked debate about the role of the underlying substrate in the properties of graphene surface.22−25 These arguments clearly state that the hydrophobicity of monolayer graphene could be regulated by the choice of underlying supporting substrate. In this sense, small differences in the wettability of the underlying substrate can introduce significant changes in the crystal growth on the surface, suggesting that it may be possible to tailor the semiconductor layers on a graphene surface without using physical or chemical modifications, which have been known to degrade the electrical properties of graphene. The use of graphene as a 2D growth template requires wellcontrolled surface characteristics on a large scale, which has © 2015 American Chemical Society
Received: December 24, 2014 Revised: March 18, 2015 Published: March 23, 2015 2474
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
Figure 1. Schematic of bilayer transfer method, plane and side views. (a) CVD grown monolayer graphene on Cu after backside-graphene etching by reactive ion etching. (b) PMMA supporting layer coating. (c) Patterning of PMMA-graphene bilayer by reactive ion etching. (d) Cu etching and rinsing in DI water. (e) Transfer of the patterned bilayer onto the 2nd graphene on Cu foil. (f) The 2nd Cu etching and rinsing. (g) Overnight vacuum storage followed by transfer bilayer graphene onto wettability-controlled substrates. (h) Pentacene deposition.
to allow the positioning of graphene sheets onto the desired substrate without any crack formation. Therefore, the second graphene within the pattern of the supporting bilayer could be transferred as a residue-free and crack-free graphene sheet. The size of free-standing graphene sheet was controlled by the pattern size of the supporting bilayer, which was successfully increased up to 1400 μm in diameter. By contrast, transfer process without a sacrificial graphene layer led to failed samples with pattern size larger than 25 μm due to weak interactions between the polymeric supporting layer and facing graphene layer (Supporting Information Figure S2a). The quality of the graphene transferred by the BTM was carefully characterized using optical microscopy (OM), scanning electron microscopy (SEM), Raman spectroscopy, atomic force microscopy (AFM), and X-ray photoelectron spectroscopy (XPS), as displayed in Figure 2. The OM and SEM images revealed that circular graphene areas inside the 250 μm diameter holes remained intact and without noticeable defects. AFM images showed a clean and smooth surface with neither apparent impurities nor fractures, which is in contrast to the graphene surfaces obtained using conventional transfer methods with randomly distributed polymeric residues (Figure 2d and Supporting Information Figure S5). Figure 2c shows the Raman spectrum and its corresponding D peak mapping of the monolayer graphene transferred using the BTM. The G and 2D bands were observed at 1583 and 2640 cm−1. The I2D/IG intensity ratio exceeded 2. The D peak at 1300 cm−1 displayed a negligible intensity. These characteristics correspond to those of the highquality monolayer graphene.30,37 Figure 2e,f show the XPS spectra of transferred graphene sheets on SiO2 substrate. Generally, distinct features of the C−C, C−O, and O−CO groups indicate the contaminated graphene surface, as shown in the graphene transferred by the conventional process (Figure 2f). In contrast, the sharp and narrow C1s core-level peak at
investigating the effects of the underlying substrate surface energy (γ) on epitaxial pentacene films growth on the graphene surface. To this end, a new approach for transferring freestanding graphene sheets onto arbitrary substrates was developed by using a two-step transfer process, the bilayer transfer method (BTM), involving a sacrificial graphene layer. These high quality and residue-free monolayer graphene sheets on wettability-controlled substrates successfully served as a controllable 2D epitaxial template for the organic crystal growth. The crystallographic and microscopic analysis revealed that the morphology and the growth mode of the pentacene film depended strongly on the surface energy of the underlying substrate, exhibiting a clear transition from the vertical to lateral growth mode with enhanced molecular diffusivity on hydrophobic substrate. Moreover, the optical properties of the pentacene layer, especially the diffusion of the exciton, also showed strong dependency on the corresponding morphological evolution. As a result, on our clean and defect-free graphene surface it was clearly shown that pentacene adopted the quasi-epitaxial growth with high crystallinity, and preferentially face-on oriented structures. Figure 1 shows the schematic illustration of the BTM process. A sacrificial graphene layer (1st graphene layer) was first embedded with a polymeric layer, and this bilayer was selectively etched through patterns (Figure 1a−c). The patterned bilayer was then used as the supporting layer for the transfer of a second monolayer graphene, where the first graphene layer functioned as the adhesion layer between polymer and the second graphene layer (Figure 1d−f). The bilayer structure not only reinforced the structural integrity of the pattern of supporting layer but also enhanced the adhesion to the second graphene. The adhesion force was revealed to be strong enough to suspend various sizes of free-standing graphene sheets36 (Supporting Information Figure S2b) and 2475
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
Figure 2. Quality of monolayer graphene in the patterned well from bilayer transfer method. (a) Optical microscope image, (b) SEM image, (c) Raman D peak mapping, (d) AFM image and XPS C1s spectra of graphene transferred by (e) bilayer transfer method and (f) conventional wettransfer method.
284.5 eV, clearly showed that a clean and pristine graphene surface was successfully achieved.30,31,34 To investigate the characteristics of this clean graphene as an epitaxial growth template, the graphene sheets were positioned onto the substrates with various surface energy γ (Figure 1g). The substrates were treated with either UV/ozone treatment or an alkyl self-assembled monolayer (SAM) before carrying out the BTM of graphene sheets, which yielded comparable surface roughness properties, and the surfaces were made hydrophilic or hydrophobic (Table 1 and Supporting Information Table S1). Pentacene was then thermally evaporated onto the graphene-covered surfaces (Figure 1h). The effects of the surface energy of the substrate on the surface characteristics of graphene sheet were demonstrated by characterizing the morphological differences and crystallite structures of the
Table 1. Surface Characteristics of the Background Substrates contact angle (deg)
substrate UV/ozone treated untreated alkyl SAM treated
surface roughness (nm)
water
diiodomethane
surface energy (mJ m−2)
0.38 ± 0.05
7±1
30 ± 1
72.7 ± 0.1
0.45 ± 0.03 0.45 ± 0.05
58 ± 2 100 ± 1
48 ± 4 72 ± 1
45.3 ± 0.5 21.8 ± 0.1
pentacene films grown on the various graphene-covered substrates. First, the nucleation and growth of the pentacene film were examined on each wettability-controlled graphene sheet. The 2476
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
diffusion processes of the adsorbed molecules. During the nucleation stage, it is well established that the nucleus density (N), the diffusivity (D), and the deposition rate (F) of organic molecules follows the following relation
surface topographies were characterized along with the pentacene deposition process as shown in Figure 3. During
F D As the deposition proceeds, the nucleation then competes with the island growth. Because the value of F was always fixed at 0.2 Ås−1, the nucleation density was mainly determined by the D of the pentacene molecules initially adsorbed on the graphene surface. Therefore, the observation of the higher nucleation density formed on hydrophilic substrates than on hydrophobic substrates clearly indicated that D was lower on the higher γ substrate, which could be attributed to the increased interaction between adsorbed pentacene molecules and the substrate. Similarly, the pentacene grown on the substrates without graphene template layer, besides the shape of grains, also showed increased nucleation density on hydrophilic substrates than hydrophobic substrates. (Figure 3j−l) Such findings were not exclusive to any specific substrates, as evidenced by the experimental data obtained using the set of substrates having similar roughness values and the wide range of γ (see the Supporting Information), as shown in Figure 4. However, despite the similar diffusion behavior of N∝
Figure 3. AFM images of pentacene films with various nominal thicknesses on wettability-controlled substrates with (a−i) and without (j−l) graphene template on top. The scale bars are 0.5 μm.
the initial stages of the growth process, the UV/ozone treated hydrophilic substrate induced the formation of a significant number of isolated, thin, long pentacene grains having heights that were many times the nominal thickness of 2 nm. As the underlying substrate γ decreased, the morphology underwent a transformation from scattered islands to conglomerated islands. Notably, the pentacene grains on graphene showed needle-like domains regardless of the substrate γ, which is the characteristic of the preferentially lying-down oriented pentacene crystals. As the nominal thickness increased, the crystal growth expanded along the vertical and lateral directions and the islands coalesced. Uniform plateau-like islands with deep trenches and specific orientations developed all over the surface of the graphene-covered untreated and alkyl SAM-treated hydrophobic substrates, whereas a network of long thin pentacene wires without interconnections formed on the hydrophilic substrate. The differences in the shape, nucleation density, and surface coverage of the pentacene films grown on the variety of graphene templates during the initial stages are related to the
Figure 4. Plots of morphological features of 2 nm thick pentacene films versus surface energy γ of the substrates. (a) The pentacene coverage (θ) on the substrates with and without graphene coating, (b) the average height of pentacene islands (hi) on the graphene-covered substrate. (c) The schematic illustrations of pentacene growth on graphene on various substrates.
adsorbed molecules on substrates with and without graphene, there was an obvious distinction in growth modes of the pentacene on those substrates. Figure 4 shows the relationship of surface coverage (θ) and island’s height of pentacene to the substrate γ at initial growth stage (2 nm nominal thickness). As γ increased, θ showed decrement while the island’s height hi 2477
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
Figure 5. GIXD patterns of pentacene films with various nominal thicknesses on transferred graphene templates on wettability-controlled substrates. The pentacene grown on graphene templates with (a), (a′), (a″) alkyl-SAM treated substrates, (b), (b′), (b″) untreated substrates and (c), (c′), (c″) UV/ozone treated substrates. The thicknesses of the pentacene layer were (a−c) 2 nm, (a′−c′) 10 nm, and (a″−c″) 50 nm.
up to 110 nm (Supporting Information Figure S9), the different growth modes of the pentacene on the substrate with and without graphene resulted in significant deviations in the final film morphologies. As mentioned above, the increase in hydrophilicity resulted in the increased initial nucleation density as in the cases without graphene. On substrates without graphene (Supporting Information Figure S9a), hydrophobicity of the substrate promoted the 3D growth of the initial islands with relatively small nucleation density (Volmer−Weber mode),46 thereby inducing small grains with high interconnectivity. As hydrophilicity increased, initial nucleation density increased and multiple growth of subsequent layers occurred so that the large grains with terrace structure evolved (Stranski− Krastanov mode).46 However, the presence of graphene induced the deviation of the pentacene growth mode from these conventional cases (Supporting Information Figure S9b). Because of the increased grain growth in vertical direction on graphene, the increased nucleation density led to large density of grains with small size. On hydrophobic substrates, a similar trend was observed while coalesce of grains led to larger initial grain accompanied by low nucleation density, which resulted in three-dimensionally grown larger grains. In this sense, on graphene templates the initial nucleation density became the determinant of the final grain size, which resulted in the growth of larger grains with high crystallinity on hydrophobic substrates. These results implied that while the “wetting-transparent” graphene effectively transmitted substrate−molecule interaction to differentiate the nucleation of the organic crystals, the growth mode, and the corresponding morphology were decisively affected by the strong intermolecular interaction due to the high γ of exposed crystal plane of lying-down crystals on graphene. The change of growth mode then resulted in a clear transition from the lateral (2D) to the vertical (3D) growth of pentacene crystals. Our results also agreed well with the previous reports, which found that a lower deposition rate, higher substrate temperature, and appropriate substrate surface could enhance the diffusivity and produce larger grain
increased. Interestingly, the opposite trend held for the substrates without graphene. Considering that the low diffusivity of adsorbed molecules induced higher nucleation density on both graphene covered and noncovered substrates, these results indicated that graphene induced different growth modes of pentacene after the nucleation. To clearly understand these behaviors, thermodynamic parameters such as the balance between the interfacial energy, the surface energy of the pentacene contact plane and of the substrate should be taken into account.38,39 Because the presence of graphene template induced the transition from the standing-up to lying-down orientation of pentacene, the surface energy of pentacene films varied due to the different crystal planes which were in contact with the substrate. On the substrates without graphene, the pentacene crystals adopted the standing-up orientation with the (001) crystal plane normal to the substrates, which bears the lowest γ.40−42 When a molecule lands on the initial island, it preferably adopts the lying-down configuration.43 The admolecules then diffuse to the step-edges of the island due to the low Schwoebel barrier at the island surface with (001) crystal orientation, thereby inducing the lateral growth of the grains.43 In this case, as the surface energy increased, the substrate−admolecule interaction hindered the diffusion of pentacene molecules to produce a large nucleation density with laterally grown grains (Stranski−Krastanov mode), which therefore increased the surface coverage at initial stage. On the contrary, on graphene surface pentacene with lyingdown orientation was formed with the exposed (020) crystal plane, which exhibits the highest γ, regardless of the background substrate.32,42,44,45 In this case, once the pentacene molecules diffused on the initial islands, they preferred to pile on top of those islands, thereby inducing the vertical growth of the grains. Therefore, the high nucleation density of pentacene formed on the graphene surface on the hydrophilic substrate resulted in the reduction of the surface coverage while increasing the islands heights (Figure 4), exhibiting the opposite trend to the case without graphene layers. Furthermore, as the thickness of the pentacene film increased 2478
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
Figure 6. Measured optical properties of the pentacene films on surface-energy controlled templates with and without graphene templates. (a,b) The static photoluminescence spectra with blocking layer (black lines) and with quenching layer (red lines). (a′,b′) The results of the spectrally resolved photoluminescence quenching measurements. The ratio of PL intensities with blocking layer (PLB) and with quenching layer (PLQ) are shown with respect to the absorption coefficient of the pentacene films. Pentacene films (110 nm) were deposited on UV/ozone treated hydrophilic (72.7 mJ/ m2, square), non-treated neutral (45.3 mJ/m2, circle), and alkyl-SAM treated hydrophobic (21.9 mJ/m2, triangle) substrates while 20 nm of BCP and C60 films were used as blocking and quenching layers, respectively. Detailed descriptions of experimental conditions are included in the Supporting Information.
standing-up structure of the thin film phase was detected. Our data agreed with the previous reports that acene compounds mainly form the lying-down configuration on substrate with strong electronic interactions.50−52 Furthermore, the morphological differences between the pentacene films were consistent with the corresponding diffraction patterns. During the early growth stages, the pentacene films grown on the hydrophilic graphene template did not display (003)B patterns (Figure 5c,c′,c″), unlike the others. Moreover, the reflections of the films grown on that template were not as sharp as those of the others and displayed low intensity, probably due to the broadening of crystalline anisotropy and the small crystal sizes. The observed trends in morphology and crystallinity of the pentacene films on wettability-controlled substrates were directly reflected on the optical properties of the pentacene films. Generally, the charge and excited polaron pair transports are facilitated in the direction of π−π stacking. Therefore, the controlled morphology and crystallinity of highly oriented organic semiconductor crystals on graphene would be highly desirable for the vertical diode-type optoelectronic devices, that is, devices that the charge carriers and the excitons necessarily flow along the vertical direction with respect to the substrate, including solar cells and light emitting diodes.53−61 In such devices, the transparency and the conductivity of the graphene enable the graphene itself to serve as both growth template and transparent electrode (or electrode modifier for transparent electrodes such as ITO).61,62 The exciton diffusion length of organic molecule (pentacene), a representative parameter that critically determines the performance of organic optoelectronic devices, was measured
sizes.40,41,47,48 As increasing the deposition rate from 0.2 to 6 Å/s, the tendency of vertical stacking of pentacene was intensified so that the 2D growth mode of pentacene on hydrophobic graphene showed a gradual shift toward 3D mode (Supporting Information Figure S7) and consequently the average lateral grain size of pentacene was significantly diminished. Moreover, increasing the growth temperature from 30 to 70 °C resulted in a slight decrement in the average island height with conglomeration of islands (Supporting Information Figure S8), except for the high temperature (70 °C) which showed prevailed irregular aggregation. More detailed descriptions of the growth behavior at different growth conditions are included in Figure S7 and S8 in the Supporting Information. Notably, although AFM images revealed noticeable differences between the pentacene films formed on the various graphene templates, the grazing incident X-ray diffraction (GIXD) patterns obtained from the pentacene films on graphene substrates were essentially identical. These patterns are illustrated in Figure 5. In all three samples, distinct (00l) planes and (020) plane reflections were observed along angles of 17.5° and 78°, respectively, with respect to the qxy direction. The interplanar spacing of the pentacene crystals based on the (020) reflections was 3.72 Å, which is in good agreement with one-half of the b-spacing value measured in the bulk pentacene.41,49 This result implied that the pentacene molecules adopted the lying-down configuration in which the long axes were nearly parallel to the graphene surface. At a very high thickness, secondary pentacene crystals grew on top of the previous layers. In addition to the recumbent structure, the 2479
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
To this end, SR-PLQ measurements were conducted.64 Using the 1D steady-state exciton diffusion equation, the absorption coefficient (α(λ)) and relative PL intensity of optically thick semiconductor films with blocking (PLB) and quenching (PLQ) layers have the following relationship
by using static photoluminescence quenching (PLQ) and spectrally resolved PLQ (Figure 6).63−65 The experimental details are described in the Figure S10 in the Supporting Information. To estimate the efficiency of exciton diffusion depending on the morphology and orientation of pentacene films, first the quenching of the PL was observed by using the device geometries of substrate/pentacene (110 nm)/bathocuproine (BCP, 20 nm) and substrate/pentacene (110 nm)/ C60 (20 nm). BCP is a well-known and commonly used exciton blocking material while C60 functions as the exciton quenching material for pentacene. Once the exciton reaches the pentacene/C60 interface, the exciton undergoes dissociation so that the PL from the geminate electron−hole pair becomes diminished. Therefore, the degree of the intensity reduction in static PL by replacing BCP with C60, that is, PL quenching, commonly indicates the efficiency of the exciton dissociation as well as the exciton diffusion. As shown in Figure 6a,b, the PL quenching showed two obvious trends; (i) dependence on the size of grain (or crystallinity) and (ii) dependence on the orientation of organic molecules. With the presence of graphene, the pentacene films on hydrophobic substrate exhibited the most efficient quenching of the PL while showing gradual decrement as the hydrophilicity increased, which agreed well with the corresponding decrement of grain sizes and crystallinity of pentacene (Supporting Information Figure S9a). On the contrary, without the presence of graphene the opposite trend was observed, which was also well correlated with the decrement in the grain sizes of pentacene with increasing hydrophobicity. The length of the exciton diffusion mediated by Förster mechanism (LD) can be described by the following equation63 LD =
η=
PL B = α(λ )L D + 1 PLQ
Here, LD corresponds to the slope of η versus α(λ). Because the excitation illumination comes through the blocking or quenching layer side as described in Supporting Information Figure S10, most of the photon absorption happens near quenching (or blocking) interface so that most of excitons reside in close proximitiy to the interfaces. In case that those excitons are in the range of LD, the geminate recombination is suppressed near the quenching interface so that PLQ drops and η becomes a high value. As the absorption coefficient decreases, the average distance (distribution) of the exciton from quenching interface becomes longer and eventually exceeds the LD, leading to the increase in PLQ and decrease in η. Therefore, we can quantitatively measure the exciton diffusion length by measuring η with respect to the absorption coefficient. We used C60 as an exciton quenching layer and BCP as the exciton blocking layer on top of the pentacene films. Figure 6a′, 6b′ show η versus α(λ) for 110 nm pentacene films grown on substrates with different surface energies. The abosprtion coefficients α′(λ) were measured by UV/vis spectroscopy and the coefficients were corrected by the angle of incidence. Consistent with the static PL quenching measurements, the measured LD was increased gradually as the hydrophobicity increased on graphene with the values of 50, 52, and 70 nm. For the cases without graphene, LD showed a gradual decrement as the hydrophobicity increased, giving values of 47, 42, and 37 nm. These results clearly showed that aside from the different energetic states at pentacene/C60 interface depending on the molecular orientation, the lyingdown orientation of pentacene on graphene could promote almost 60% increment in LD regardless of the surface energy of the substrate. Considering that the exciton diffusion length is critically determined by the orientation of dipole moments and the proximity between chromophores, it was obvious that aside from the proximitiy of chromophores in vertical direction, the lying-down oriention of pentacene crystals was also highly beneficial to the diffusion of the excitions. For these reasons, it could be concluded that the use of graphene to promote oriented growth of pentacene and the control of the surface energy of the underneath substrate to optimize the morphology could effectively enhance the optical properties. Another interesting feature in using “wetting transparent” 2D growth template is that the pentacene molecule could undergo the epitaxial growth due to the sp2 carbon-based hexagonal structure of graphene. As shown in topographical and crystallographic analyses, the pentacene molecules adopted the lying-down orientations that indicated direct interactions between the pentacene plane (020) and graphene surface.32,41 We also observed that pentacene grains intersected with their neighbors at the specific angles, suggesting the epitaxial growth of pentacene crystals on the graphene surface.50 In this sense, we chose to use statistical method to observe the epitaxial growth. Although this approach provided indirect observation of physisorbed pentacene layer, there are several theoretical and
κ 2 ΦF σ 8πn 4 a 4
where κ is the dipole orientation factor, n is the refractive index, σ is the Förster overlap integral, ΦF is the fluorescence quantum yield, and a is the average hopping distance of the excitons.64 Generally, the κ is mostly determined by the physical orienation of the molecules and the a is strongly affected by the proximity between the chromophores.56,64,65 In this sense, considering that the intragrain exciton diffusion is much more facilitated than the intergrain diffusion due to the proximity between chromophores, the highly crystalline structure with large coherence is desirable for the optimization of the LD. Therefore, the observed trend in PL quenching efficiency along with the size of the grain (Supporting Information Figure S9) and crystallinity (Figure 5), regardless of the existence of the graphene template, agreed well with the Förster description. Moreover, for all pentacene films on graphene the efficiency of PL quenching was higher than the films on substrates without graphene. Since the exciton transfer between two chromophores depends on the direction of molecular dipole moments κ, the increased PL quenching of pentacene films with lying-down orientation was also consistent with the Förster description. However, aside from the exciton diffusion, it cannot also be ignored that the exciton quenching itself could be different, depending on the molecular orientation of pentacene at pentacene/C60 interfaces.56−58 Therefore, more quantitative measurement of the exciton diffusion length would be necessary to correctly describe the effect of molecular orientation. 2480
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
Figure 7. Sign of pentacene epitaxial growth on graphene. (a) SEM image of 10 nm pentacene on graphene-covered bare substrate, (b) schematic of intersecting point of pentacene islands on graphene; green quadrilateral shows pentacene unit cell. (c−h) Distribution histograms of islands’ intersecting angles of 2 and 50 nm thick pentacene films on transferred graphene on different substrates. (i) Cartoon illustration of serial epitaxial growth of pentacene on graphene.
hydrophilic graphene template yielded both ±60° and ±0° angles. All other angles detected were negligible. Therefore, these findings indicated that the pentacene molecules were preferably underwent epitaxial stacking on the graphene crystals. Moreover, these results also further supported that the hydrophilicity of UV/ozone treated background substrate prevented adsorbents from diffusing easily to form crystals with large grains. As the pentacene layer thickness increased, discrepancies between the φ distributions were less significant (Figure 7f−h). Cartoon images of the pentacene islands’ arrangements on the graphene surface during the growth process are shown in Figure 7i. In this case, the azimuthal orientation of a pentacene grain could extend over different graphene grain boundaries during the lateral growth of the pentacene grains. However, since the initial physisorption (nucleation) of pentacene decides the azimuthal orientation of the pentacene grains, the statistical azimuthal distribution could indicate that the initial growth of pentacene grains exhibited the epitaxial growth behavior. Additional studies on atomic level measurements might be beneficial to the further confirmation of the epitaxial growth of pentacene on graphene surfaces.
experimental observations of epitaxial adsorption of pentacene molecules on graphitic surfaces that show distinct azimuthal distribution yielding 3-fold symmetry of the pentacene monolayer grains.50,66,67 Because the pentacene molecules and graphene crystals were based on a 6-fold symmetric hexagonal unit, the epitaxial stacking of pentacene molecules on a single crystallite graphene would be most favored along three specific directions, thereby intersecting at an angle of 60° for maximizing the π−π interactions.50,66,67 Given that the graphene sheets that were used to compare the different growth behaviors were synthesized in the same batch and the statistics of azimuthal angle distribution were accumulated on multiple (>20) samples, the repeatedly observed distinct trends could be statistically meaningful to support the characteristics of the epitaxial growth. A typical area of intersecting pentacene islands is shown in the SEM image in Figure 7a, and a cartoon illustration of the stacking mode is shown in Figure 7b. Histogram plots of the intersecting angles (φ) are shown in Figure 7c−h. The histograms of the initial layers revealed the dominant intersection angle of ±60° (or its supplementary ±120°), as the pentacene islands crossed one another on the neutral and hydrophobic graphene templates. However, the 2481
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters Finally, to confirm the “wetting-transparency” effect of our graphene sheets, the pentacene island’s height hi and the surface energy γ for the graphene-covered hydrophilic substrate were plotted as a function of the number of present graphene layers (nL) in Figure 8a. As nL increased, the value of hi for the
of the van der Waals cutoff of molecular interactions through graphene on electron-transferring hydrophilic surface, which corresponds to 4−6 layers of graphene.20,22,24,68 These results also suggested that the background substrate dependence of the pentacene growth could be gradually controlled by adjusting the number of graphene layers on hydrophilic surfaces, aside from using wettability-controlled substrates. In summary, the BTM was used to produce residue-free graphene sheets on a desired substrate to demonstrate “wetting-transparent” 2D epitaxial template. We successfully designed graphene surfaces that displayed controlled wettability properties and could be used as growth templates for organic crystals. Graphene sheets with controlled surface properties were achieved by mounting a clean and undoped graphene layer onto arbitrary surface-treated substrates. The growth of pentacene layer on those graphene surfaces firmly depended on the value of γ of the background substrate, indicating that the monolayer graphene was “transparent” to the wetting properties of the underlying substrate surface. As the result of that, the pentacene islands switched from blocklike to single rodlike as γ of the background substrate increased, suggesting that the underlying hydrophobicity firmly altered the pentacene− graphene interactions. The diffusion of the exciton, which is a key parameter in organic optoelectronic devices, also showed strong dependencies on the molecular orientation and the morphological evolution of pentacene films, which resulted in the increment of LD up to 60% for epitaxially oriented pentacene films on the hydrophobic substrates. Furthermore, these effects, or the “transparency”, gradually decreased as the distance between the background substrate and the adsorbed molecules increased. These results confirmed that the monolayer graphene was thin enough to allow the underlying substrate to induce pentacene growth. The profound understanding of graphene’s role in the molecule−substrate interactions, as clarified in this study, represents a step toward applying graphene as the tunable templates and transparent electrodes in flexible electronics.
■
ASSOCIATED CONTENT
S Supporting Information *
Figure 8. Effect of number of graphene layers on the substratepentacene interaction at the initial growth stage of pentacene. (a) Plot of the average island heights of 2 nm thick pentacene films and the surface energy of graphene-covered hydrophilic substrate versus the number of graphene layers. (b−e) AFM images of 2 nm thick pentacene films on different numbers of graphene layers transferred onto the hydrophilic substrate. (b) Single layer, (c) two layers, (d) three layers, and (e) six layers. The scale bars are 0.5 μm.
Experimental and methods, process description of bilayer transfer method, robustness of supporting bilayers in BTM, graphene properties transferred via conventional method and the BTM, characterization of pentacene films on “wettingtransparent” graphene templates, effect of growth conditions on graphene templates, effect of surface-wettability on the final morphology of the pentacene films, measurement of exciton diffusion length in pentacene films, and additional figures, table, and references. This material is available free of charge via the Internet at http://pubs.acs.org.
pentacene grains decreased rapidly until four layers, accompanied by a similar decrement in γ of the graphene template. Further increment in the nL showed a slight decrement of both hi and γ down to the value for the intrinsic hydrophobic graphene surfaces. AFM images of the pentacene films that formed on the 1, 2, 3, and 6 graphene layers on the UV/ozone treated substrates revealed that the nucleation density decreased and the seeds became more uniform at higher values of nL, as shown in Figure 8b−e. Therefore, it was clear that the clean graphene could effectively transmit the interaction between the organic molecule and underlying substrates, which could be regulated by increasing the number of graphene layers until it diminishes completely after four layers of graphene. These results agree well with the previous predictions
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Author Contributions
N.N.N. and S.B.J. contributed equally to this work. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by a grant (Code No.2011-0031628) from the Center for Advanced Soft Electronics under the 2482
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters
(28) Luo, B.; Chen, B.; Meng, L.; Geng, D.; Liu, H.; Xu, J.; Zhang, Z.; Zhang, H.; Peng, L.; He, L.; Hu, W.; Liu, Y.; Yu, G. Adv. Mater. 2014, 26, 3218−3224. (29) Wang, D.-Y.; Huang, I. S.; Ho, P.-H.; Li, S.-S.; Yeh, Y.-C.; Wang, D.-W.; Chen, W.-L.; Lee, Y.-Y.; Chang, Y.-M.; Chen, C.-C.; Liang, C.T.; Chen, C.-W. Adv. Mater. 2013, 25, 4521−4526. (30) Pirkle, A.; Chan, J.; Venugopal, A.; Hinojos, D.; Magnuson, C. W.; McDonnell, S.; Colombo, L.; Vogel, E. M.; Ruoff, R. S.; Wallace, R. M. Appl. Phys. Lett. 2011, 99, 122108. (31) Lin, Y. C.; Lu, C. C.; Yeh, C. H.; Jin, C.; Suenaga, K.; Chiu, P. W. Nano Lett. 2012, 12, 414−419. (32) Lee, W. H.; Park, J.; Sim, S. H.; Lim, S.; Kim, K. S.; Hong, B. H.; Cho, K. J. Am. Chem. Soc. 2011, 133, 4447−4454. (33) Kim, H. H.; Chung, Y.; Lee, E.; Lee, S. K.; Cho, K. Adv. Mater. 2014, 26, 3213−3217. (34) Suk, J. W.; Lee, W. H.; Lee, J.; Chou, H.; Piner, R. D.; Hao, Y.; Akinwande, D.; Ruoff, R. S. Nano Lett. 2013, 13, 1462−1467. (35) Lee, W. H.; Park, J.; Sim, S. H.; Jo, S. B.; Kim, K. S.; Hong, B. H.; Cho, K. Adv. Mater. 2011, 23, 1752−1756. (36) Wang, Y.; Tong, S. W.; Xu, X. F.; Ozyilmaz, B.; Loh, K. P. Adv. Mater. 2011, 23, 1514−1518. (37) Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K. Phys. Rev. Lett. 2006, 97, 187401. (38) Reichelt, K. Vacuum 1988, 38, 1083−1099. (39) Venables, J.; Spiller, G.; Hanbucken, M. Rep. Prog. Phys. 1984, 47, 399. (40) Lee, H. S.; Kim, D. H.; Cho, J. H.; Hwang, M.; Jang, Y.; Cho, K. J. Am. Chem. Soc. 2008, 130, 10556−10564. (41) Ruiz, R.; Choudhary, D.; Nickel, B.; Toccoli, T.; Chang, K.-C.; Mayer, A. C.; Clancy, P.; Blakely, J. M.; Headrick, R. L.; Iannotta, S. Chem. Mater. 2004, 16, 4497−4508. (42) Nabok, D.; Puschnig, P.; Ambrosch-Draxl, C. Phys. Rev. B 2008, 77, 245316. (43) Ruiz, R.; Nickel, B.; Koch, N.; Feldman, L.; Haglund, R.; Kahn, A.; Scoles, G. Phys. Rev. B 2003, 67, 125406. (44) Berke, K.; Tongay, S.; McCarthy, M. A.; Rinzler, A. G.; Appleton, B. R.; Hebard, A. F. J. Phys. Condens. Mater. 2012, 24, 255802. (45) Liu, X.; Grüneis, A.; Haberer, D.; Fedorov, A. V.; Vilkov, O.; Strupinski, W.; Pichler, T. J. Phys. Chem. C 2013, 117, 3969−3975. (46) Yang, S. Y.; Shin, K.; Park, C. E. Adv. Funct. Mater. 2005, 15, 1806−1814. (47) Kim, D. H.; Lee, H. S.; Yang, H.; Yang, L.; Cho, K. Adv. Funct. Mater. 2008, 18, 1363−1370. (48) Kang, B.; Jang, M.; Chung, Y.; Kim, H.; Kwak, S. K.; Oh, J. H.; Cho, K. Nat. Commun. 2014, 5, 4752. (49) Lukas, S.; Sohnchen, S.; Witte, G.; Woll, C. ChemPhysChem 2004, 5, 266−270. (50) Götzen, J.; Käfer, D.; Wöll, C.; Witte, G. Phys. Rev. B 2010, 81, 085440. (51) Käfer, D.; Witte, G. Chem. Phys. Lett. 2007, 442, 376−383. (52) Schroeder, P.; France, C.; Park, J.; Parkinson, B. J. Phys. Chem. B 2003, 107, 2253−2261. (53) Kim, J. S.; Park, Y.; Lee, D. Y.; Lee, J. H.; Park, J. H.; Kim, J. K.; Cho, K. Adv. Funct. Mater. 2010, 20, 540−545. (54) Lu, X.; Hlaing, H.; Nam, C.-Y.; Yager, K. G.; Black, C. T.; Ocko, B. M. Chem. Mater. 2014, 27, 60−66. (55) Kolata, K.; Breuer, T.; Witte, G.; Chatterjee, S. ACS Nano 2014, 8, 7377−7383. (56) Chernikov, A.; Yaffe, O.; Kumar, B.; Zhong, Y.; Nuckolls, C.; Heinz, T. F. J. Phys. Chem. Lett. 2014, 5, 3632−3635. (57) Ayzner, A. L.; Nordlund, D.; Kim, D.-H.; Bao, Z.; Toney, M. F. J. Phys. Chem. Lett. 2014, 6, 6−12. (58) Yi, Y.; Coropceanu, V.; Brédas, J.-L. J. Am. Chem. Soc. 2009, 131, 15777−15783. (59) Tang, F.-C.; Wu, F.-C.; Yen, C.-T.; Chang, J.; Chou, W.-Y.; Gilbert Chang, S.-H.; Cheng, H.-L. Nanoscale 2015, 7, 104−112.
Global Frontier Research Program of the Ministry of Science, ICT, and Future Planning, Korea. The authors thank the Pohang Accelerator Laboratory for providing the synchrotron radiation sources at 4D, 8A2, 3C, and 9A beamlines used in this study.
■
REFERENCES
(1) Geim, A. K. Science 2009, 324, 1530−1534. (2) Rao, C. N.; Sood, A. K.; Subrahmanyam, K. S.; Govindaraj, A. Angew. Chem., Int. Ed. 2009, 48, 7752−7777. (3) Bae, S.; Kim, H.; Lee, Y.; Xu, X.; Park, J. S.; Zheng, Y.; Balakrishnan, J.; Lei, T.; Kim, H. R.; Song, Y. I.; Kim, Y. J.; Kim, K. S.; Ozyilmaz, B.; Ahn, J. H.; Hong, B. H.; Iijima, S. Nat. Nanotechnol. 2010, 5, 574−578. (4) Eda, G.; Fanchini, G.; Chhowalla, M. Nat. Nanotechnol. 2008, 3, 270−274. (5) Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J. H.; Kim, P.; Choi, J. Y.; Hong, B. H. Nature 2009, 457, 706− 710. (6) Lee, C.; Wei, X.; Kysar, J. W.; Hone, J. Science 2008, 321, 385− 388. (7) Jo, S. B.; Park, J.; Lee, W. H.; Cho, K.; Hong, B. H. Solid State Commun. 2012, 152, 1350−1358. (8) Lee, Y.-Y.; Tu, K.-H.; Yu, C.-C.; Li, S.-S.; Hwang, J.-Y.; Lin, C.-C.; Chen, K.-H.; Chen, L.-C.; Chen, H.-L.; Chen, C.-W. ACS Nano 2011, 5, 6564−6570. (9) Ni, G.-X.; Yang, H.-Z.; Ji, W.; Baeck, S.-J.; Toh, C.-T.; Ahn, J.-H.; Pereira, V. M.; Ö zyilmaz, B. Adv. Mater. 2014, 26, 1081−1086. (10) Lee, S.-K.; Jang, H. Y.; Jang, S.; Choi, E.; Hong, B. H.; Lee, J.; Park, S.; Ahn, J.-H. Nano Lett. 2012, 12, 3472−3476. (11) Luo, J.; Kim, J.; Huang, J. Acc. Chem. Res. 2013, 46, 2225−2234. (12) Luo, J.; Jang, H. D.; Huang, J. ACS Nano 2013, 7, 1464−1471. (13) Järvinen, P. i.; Hämäläinen, S. K.; Banerjee, K.; Häkkinen, P.; Ijäs, M.; Harju, A.; Liljeroth, P. Nano Lett. 2013, 13, 3199−3204. (14) Mativetsky, J. M.; Wang, H.; Lee, S. S.; Whittaker-Brooks, L.; Loo, Y.-L. Chem. Commun. 2014, 50, 5319. (15) Pollard, A. J.; Perkins, E. W.; Smith, N. A.; Saywell, A.; Goretzki, G.; Phillips, A. G.; Argent, S. P.; Sachdev, H.; Müller, F.; Hüfner, S. Angew. Chem., Int. Ed. 2010, 49, 1794−1799. (16) Ogawa, Y.; Niu, T.; Wong, S. L.; Tsuji, M.; Wee, A. T. S.; Chen, W.; Ago, H. J. Phys. Chem. C 2013, 117, 21849−21855. (17) Wang, Q. H.; Hersam, M. C. Nat. Chem. 2009, 1, 206−211. (18) Lim, S.; Kang, B.; Kwak, D.; Lee, W. H.; Lim, J. A.; Cho, K. J. Phys. Chem. C 2012, 116, 7520−7525. (19) Shi, Y.; Zhou, W.; Lu, A.-Y.; Fang, W.; Lee, Y.-H.; Hsu, A. L.; Kim, S. M.; Kim, K. K.; Yang, H. Y.; Li, L.-J.; Idrobo, J.-C.; Kong, J. Nano Lett. 2012, 12, 2784−2791. (20) Wang, Q. H.; Jin, Z.; Kim, K. K.; Hilmer, A. J.; Paulus, G. L.; Shih, C. J.; Ham, M. H.; Sanchez-Yamagishi, J. D.; Watanabe, K.; Taniguchi, T.; Kong, J.; Jarillo-Herrero, P.; Strano, M. S. Nat. Chem. 2012, 4, 724−732. (21) Colson, J. W.; Woll, A. R.; Mukherjee, A.; Levendorf, M. P.; Spitler, E. L.; Shields, V. B.; Spencer, M. G.; Park, J.; Dichtel, W. R. Science 2011, 332, 228−231. (22) Rafiee, J.; Mi, X.; Gullapalli, H.; Thomas, A. V.; Yavari, F.; Shi, Y.; Ajayan, P. M.; Koratkar, N. A. Nat. Mater. 2012, 11, 217−222. (23) Raj, R.; Maroo, S. C.; Wang, E. N. Nano Lett. 2013, 13, 1509− 1515. (24) Shih, C.-J.; Wang, Q. H.; Lin, S.; Park, K.-C.; Jin, Z.; Strano, M. S.; Blankschtein, D. Phys. Rev. Lett. 2012, 109, 176101. (25) Li, Z.; Wang, Y.; Kozbial, A.; Shenoy, G.; Zhou, F.; McGinley, R.; Ireland, P.; Morganstein, B.; Kunkel, A.; Surwade, S. P.; Li, L.; Liu, H. Nat. Mater. 2013, 12, 925−931. (26) Su, C.-Y.; Lu, A.-Y.; Wu, C.-Y.; Li, Y.-T.; Liu, K.-K.; Zhang, W.; Lin, S.-Y.; Juang, Z.-Y.; Zhong, Y.-L.; Chen, F.-R.; Li, L.-J. Nano Lett. 2011, 11, 3612−3616. (27) Wei, D.; Liu, Y. Adv. Mater. 2010, 22, 3225−3241. 2483
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484
Letter
Nano Letters (60) Tumbleston, J. R.; Collins, B. A.; Yang, L.; Stuart, A. C.; Gann, E.; Ma, W.; You, W.; Ade, H. Nat. Photonics 2014, 8, 385−391. (61) Zhang, L.; Roy, S. S.; Hamers, R. J.; Arnold, M. S.; Andrew, T. L. J. Phys. Chem. C 2014, 119, 45−54. (62) Singha Roy, S.; Bindl, D. J.; Arnold, M. S. J. Phys. Chem. Lett. 2012, 3, 873−878. (63) Lunt, R. R.; Giebink, N. C.; Belak, A. A.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys. 2009, 105, 053711. (64) Lunt, R. R.; Benziger, J. B.; Forrest, S. R. Adv. Mater. 2010, 22, 1233. (65) Sim, M.; Shin, J.; Shim, C.; Kim, M.; Jo, S. B.; Kim, J.-H.; Cho, K. J. Phys. Chem. C 2013, 118, 760−766. (66) Groszek, A. J. Proc. R. Soc. A 1970, 314, 473−498. (67) Paramonov, P.; Coropceanu, V.; Brédas, J.-L. Phys. Rev. B 2008, 78, 041403. (68) Shih, C. J.; Strano, M. S.; Blankschtein, D. Nat. Mater. 2013, 12, 866−869.
2484
DOI: 10.1021/nl504958e Nano Lett. 2015, 15, 2474−2484