Controlling Interface Morphology and Layer Crystallinity in Organic

Subscriber access provided by Macquarie University

Organic Electronic Devices

Controlling Interface Morphology and Layer Crystallinity in Organic Heterostructures: Microscopic View on C60 Island Formation on Pentacene Buffer Layers Andrea Huttner, Tobias Breuer, and Gregor Witte ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b09369 • Publication Date (Web): 28 Aug 2019 Downloaded from pubs.acs.org on August 30, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Controlling Interface Morphology and Layer Crystallinity in Organic Heterostructures: Microscopic View on C60 Island Formation on Pentacene Buffer Layers Andrea Huttner, Tobias Breuer, Gregor Witte* Fachbereich Physik, Philipps-Universität Marburg, 35032 Marburg, Germany ABSTRACT: Controlling the crystallinity of organic thin films is an important aspect in the improvement of organic electronic devices. However, due to high molecular mass, structural anisotropy, and weak intermolecular van der Waals bonding, crystalline ordering is not easily accomplished. While film preparation at elevated substrate temperature often improves the crystalline quality, this approach cannot be applied to temperature-sensitive materials such as plastic foils used as substrates for flexible electronics. Here, we examine in detail a low-temperature approach to improve film crystallinity by using ultra-thin pentacene (PEN) buffer layers that allow crystalline growth of Buckminsterfullerene (C60) thin films while without such buffer layers only amorphous fullerene films are formed upon room temperature deposition on various support substrates. Remarkably, this effect depends critically on the thickness of the PEN buffer and requires a thickness of at least two monolayers to induce crystalline growth whereas a buffer layer consisting of a monolayer of PEN again yields amorphous C60 films. Combining crystallographic investigations by X-ray diffraction and atomic force microscopy measurements we determine distinct nucleation sites on buffer layers of different thickness, which are correlated to the amorphous, respectively crystalline C60 islands. Our microscopic analysis reveals distinct differences for the nucleation and diffusivity of fullerenes on the PEN monolayer and on thicker buffer layers, which are attributed to the molecular arrangement in the PEN monolayer. Finally, we show that the crystalline C60 films are exclusively (111)-oriented and the fullerene islands are even heteroepitaxially aligned on the PEN buffer. Keywords: organic semiconductors; organic heterostructures; pentacene; C60; atomic force microscopy; interfaces; molecular buffer layer

INTRODUCTION Organic electronics has attracted large attention because it offers a perspective for the realization of low-cost, large area and lightweight devices1 and in combination with plastic substrates even enables the fabrication of mechanically flexible devices.2 Small molecule organic semiconductors (OSC) or dyes constitute versatile building blocks that allow controlled growth of pure films using vapor deposition methods, which thus enables deriving a deeper understanding of structure-property relationships in such materials. The realization of organic field effect transistors (OFET), however, also requires a notable degree of crystalline order since grain boundaries and other defects largely reduce an efficient charge carrier transport.3 While the degree of crystallinity and the domain size of OSC films typically increase with processing temperature, the available temperature regime can be limited for plastic substrates. This appears particularly critical for Buckminsterfullerene (C60), which is an n-type OSC with excellent thermal, chemical and (UV-) light stability.4 Although it shows remarkable charge carrier mobility in its crystalline phase, respective crystalline films are only formed upon growth at elevated temperatures.5 An interesting approach to this problem was reported by Itaka et al. who found that precoating the supporting substrate with a (001)-oriented pentacene (PEN) buffer layer as thin as one monolayer largely improves the crystallinity of C60 films on sapphire supports and yields enhanced OFET device characteristics.6 Since similar effects were also observed for such pentacene buffer layer on other inorganic substrates, suggesting that this approach could be transferred also to plastic foil substrates where equivalent molecular orientations of pentacene are found.7-9 Surprisingly, however, rather different values for the necessary PEN buffer layer thicknesses ranging from 2 nm to 20 nm have been reported.10-14 In addition, the micro-structure and growth of C60/PEN heterostructures (and vice versa) is also of great interest because it serves as a model system for molecular acceptor/donor interfaces, as present in organic photovoltaics.15,16 However, previous structural studies are focused either on the very initial stage of hetero-growth (typically first 1-2 monolayers) using scanning

probe microscopy or correlate the crystallinity with resulting device properties so that the physical origin of the improved crystalline growth is still unclear. To gain complementary insights into the structure and development of this interface, the molecular film growth has also been theoretically investigated within the framework of molecular dynamics (MD) simulations. While this allowed to derive valuable details such as diffusion and Ehrlich-Schwöbel barriers for the fullerene17,18 or a coverage dependent molecular reorientation of PEN,19,20 full simulations of the evolution of crystalline hetero-stacks are still very challenging due to the required large number of molecules and the required long simulation time. Therefore, idealized planar bottom layers are often considered, although real systems have defects such as steps that significantly affect the nucleation and growth as demonstrated before for inorganic and organic films on inorganic substrates.21-26 Only in a recent work by Clancy and coworkers also the influence of steps at the underlying PEN layer on the subsequent growth of C60 has been investigated showing a preferential nucleation of C60 at steps27 which is in good agreement with experimental observations.28-30 In addition, the simulations yield a distinctly larger cohesion between fullerenes than the adhesion to PEN. This challenges the explanation of Itaka et al. who attributed the improved crystallinity to the enhanced wetting of PEN by the fullerenes.6 A further complexity has recently been identified for the C60/PEN interface, namely a chemical reaction leading to the formation of a heteromolecular Diels-Alder adduct.31 While temperature dependent studies show that the adduct formation is a thermally activated process which starts at about 330 K,32 the presence of this new species has not yet been taken into account in any theoretical or experimental interface studies. While the previous studies have revealed an improved crystallinity for C60 films grown on top of PEN films, we also want to address heteromolecular epitaxial growth, i.e. a discrete lateral registration between the fullerenes and the underlying PEN lattice. Such an azimuthal arrangement of crystalline C60 films has already been observed for fullerenes on pentacene single crys-

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

tals.33 However, the respective crystals were grown in the crystalline Siegrist phase, which is not adopted in the buffer layers or in thin film devices. Hence, the question arises whether C60 also forms epitaxial layers on PEN films in the relevant polymorph in the buffer layers, which is the so-called thin-film (TF) phase.34 In the present study, we combine atomic force microscopy (AFM) with X-ray diffraction (XRD) to characterize the initial nucleation of C60 on PEN buffer layers and the subsequent evolution of thicker fullerene films. At first, we demonstrate for various substrates (including SiO2, KCl(100) and graphite) that PEN buffer layers enable the growth of crystalline C60 layers at temperatures of about 310 K while no crystalline fullerene films are formed on the bare substrates at that temperature. Our detailed analysis for the case of SiO2 shows further that C60 forms no crystalline films when grown on PEN films of only one monolayer thickness. Instead, crystalline growth occurs for PEN buffer layer thicknesses of two monolayers or above. In addition, we address the influence of the substrate roughness on the fullerene crystallinity by inspecting the hetero-growth also on different supporting substrates. This reveals a subtle interplay of initial nucleation and crystalline island ripening and thus provides valuable insight into the initial stage of hetero-growth.

Page 2 of 9

RESULTS AND DISCUSSION A. Influence of PEN Buffer Layer on C60 Film Growth. Previous works have shown that PEN buffer layers induce the growth of crystalline C60 films on various substrates, even at low temperature6, 10-14. Unfortunately, the microstructure and roughness of these support surfaces have not been systematically analyzed in that works, so that a simple smoothing effect as the origin of the improved crystalline growth can neither be confirmed nor ruled out. Therefore, we first investigate the growth of C60 films on different surfaces and compare the resulting crystallinity for films grown with and without PEN buffer layer.

EXPERIMENTAL SECTION All PEN (Sigma Aldrich, purity ≥ 99.9%) and C60 (Sigma Aldrich, purity ≥ 99.5%) films are grown under high vacuum conditions by means of organic molecular beam deposition (OMBD) from resistively heated Knudsen cells. Typical cell temperatures of 480 K and 650 K are used for PEN and C60, respectively, yielding deposition rates of 0.3 nm/min (PEN) and 0.12 nm/min (C60) that are monitored by quartz crystal microbalance (QCM). All C60 films and PEN buffer layers are deposited at a substrate temperature of 310 K to avoid the formation of Diels-Alder adducts and to suppress thermally induced dewetting of the PEN buffer.35 For most growth experiments natively oxidized Si(100) wafers are used as substrate (referred to as SiO2), which are cleaned by acetone and ethanol. Additional films are also grown on other supporting substrates comprising KCl(100) and graphite. KCl(100) surfaces are prepared by cleaving slices from a single crystal rod (Korth Kristalle) in air. Basal plane surfaces of graphite are prepared by exfoliation of highly oriented pyrolytic graphite (HOPG) substrates (SPI supplies, mosaicity < 0.4°) in air. Before molecular film deposition the substrates are heated at 500 K in vacuum to remove residual water and other contaminations. The film morphology is characterized by atomic force microscopy (AFM, Agilent SPM 5500) operated in tapping mode at ambient conditions and room temperature using cantilevers with a spring constant of 40 N/m (fres = 325 kHz) and a tip radius of 7 nm. All AFM data are processed and visualized using the SPIP 5.1.5 software package (Image Metrology). For better visualization of the fullerene cluster and islands formed on the PEN bottom layer a non-monotonous color coded height scale is used as described in the Supporting Information, Figure S1. The crystalline structure and azimuthal orientation of the molecular films are characterized by means of X-ray diffraction (XRD) using a Bruker D8 Discover diffractometer with a monochromatized Cu Kα radiation (λ= 0.154 nm).

Figure 1. Specular X-ray diffractograms comparing the crystallinity of 10 nm C60 films grown at 310 K onto various substrates without (blue curves) and with 6 nm PEN buffer layer (black curves). A polished Si(100) wafer covered with a native oxide represents an amorphous SiO2 substrate with smooth surface (roughness: rms < 0.3 nm) and no recognizable substrate steps (see Supp. Inf., Figure S2). C60 deposition onto bare SiO2 surfaces at 310 K (and even at elevated temperatures of 460 K) exhibits no detectable crystallinity, i.e. the films are either amorphous or the crystallinity is too low to allow for a detection in our setup. Hence, the film is denoted as quasi-amorphous. By contrast, largely enhanced crystalline order is observed for identical growth conditions on a PEN buffer layer with a thickness of 6 nm. The corresponding X-ray data (cf. Figure 1) reveal that PEN films adopt the thin-film phase,34 while the (111)C60 peak indicates that the fullerenes crystallize in the fcc structure.36 Wide angular range Xray data recorded for a C60 film grown on a thicker buffer layer and their comparison with fullerene powder spectra (see Supp. Inf., Figure S3) reveal no other fullerene related reflections. Consequently, an exclusive (111) texture is observed. Note that the rather large peak width for both, the PEN and C60-related signals are due to the comparably low film thickness, hence limiting the vertical coherent size of the investigated crystallites. Additional

2


Page 3 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


XRD measurements performed on thicker films (cf. Section D and 31) corroborate this assignment. As an example for an atomically smooth and crystalline insulator surface we chose KCl(100) cleavage planes. They exhibit atomically smooth terraces extending over more than 500 nm, separated by surface steps with a height of 0.31 nm and multiple of them (see Supp. Inf., Figure S2). Like in the case of SiO2, the C60 films deposited onto bare alkali halide surfaces at 310K exhibit no detectable crystallinity. Precoating the KCl(100) surface with a PEN buffer layer, however, leads to the formation of crystalline C60 films at the same growth temperature. Remarkably, at significantly larger growth temperatures (above 420 K) isolated C60 crystals are formed on bare KCl(100) surfaces.37 Another, interesting aspect of this substrate is the epitaxial order of PEN adlayers on KCl(100),29,30 since this allows to address the question whether heteroepitaxial fullerene growth also occurs on PEN layers adopting the TF-phase. This will be discussed below in section D. In addition, we also examine graphite as substrate for the C60/PEN hetero-growth. At first glance, this appears to be a surprising substrate choice since crystalline C60 films are formed on the bare graphite substrate.40 Moreover, it was shown previously that PEN forms (022)-oriented films of the Siegrist phase on graphite where molecules exhibit a recumbent orientation.41 While these films are very smooth, no diffraction was observed for subsequently grown fullerene layers indicating quasi-amorphous structure.31 This demonstrates that the molecular orientation in the buffer layer is very crucial for the structure of subsequently grown fullerene films. In order to still allow a comparison with C60 growth on the PEN buffer layers on the before discussed surfaces where molecules are uprightly oriented, the graphite surface is intentionally roughened by brief Ar+ ion sputtering as this destroys the crystalline surface ordering (evidenced by disappearance of the LEED pattern) while the surface roughness is only slightly increased. Subsequent deposition of PEN leads to an upright molecular orientation in the thin-film phase, hence an equal situation as found in the other cases.41, 42 As shown in Figure 1 the growth scenario parallels the before discussed cases: while C60 films without notable crystalline order are formed on the sputtered graphite surface, the use of a PEN buffer layer again induces crystalline order in the fullerene film. B. PEN Buffer Layer Thickness Determines C60 Crystallinity. In the literature, rather different thicknesses of the required PEN buffer layer to achieve crystalline C60 films have been reported. Therefore, this aspect is studied more closely in our work by comparing the grain size and crystallinity of fullerene films as a function of the PEN buffer layer thickness. For that purpose, we use 10 nm thick C60 films that are prepared on top of PEN layers of different thickness grown before on SiO2 substrates. To enable a microscopic comparison of the films and to exclude slight variations of the deposition parameters (e.g. substrate temperature and growth rate), a shadow mask is used to vary the PEN buffer layer thickness at different sample positions while keeping all other parameters identical. The corresponding AFM data (composed of smaller images) is presented in Figure 2. Clearly, various regions (indicated by dashed white lines, marked I-IV) with

Figure 2. Morphology of a 10 nm C60 film grown on a PEN buffer gradient supported by SiO2. (a) Structural sketch of the sample geometry and (b) AFM micrograph showing regions of different PEN bottom layer thickness (0 – 3 ML indicated by white dashed lines) together with magnified AFM images (insets I-IV) and (c) evaluation of C60 island size as function of PEN buffer layer thickness. (d) Specular X-ray diffractograms of individually prepared samples of 10 nm C60 grown SiO2 covered with different PEN buffer layer thickness (0 - 4 ML). different PEN film thickness can be identified as sketched in panel a). While for all buffer layer thicknesses the PEN molecules adopt an upright orientation, the C60 islands atop feature distinctly different sizes on the buffer layers of different thickness as can be recognized in the magnified AFM micrographs presented in the inset of Figure 2b). A statistical analysis of the AFM data shows that the fullerene island diameter steadily increases from about 50 nm (without PEN buffer layer) up to about 350 nm for 3 ML PEN buffer. For even thicker buffer layers this trend saturates as depicted in panel c), due to the PEN island size. While the AFM data clearly indicates an effective ripening process with increasing initial PEN buffer layer thickness, it does not directly allow to distinguish between amorphous and crystalline C60 islands. To derive complementary information on the actual crystallinity of the C60 adlayers formed on top of the different PEN buffer layers, X-ray diffractometry is applied. For these measurements, the films are not prepared with a thickness gradient but instead featured a homogeneous PEN coverage over the entire sample surface as determined by QCM, while in

3



Page 4 of 9

each case a 10 nm C60 film is subsequently grown on top. The corresponding XRD data (cf. Figure 2d) impressively shows that for PEN buffer layer thicknesses of up to 1 ML the C60 films do not grow in crystalline form. Yet, upon further increase of the buffer layer thickness, clear signals revealing a crystalline (111)C60 orientation are observed, without any other reflections of the C60 film (cf. Supp. Inf., Figure S3). We note that supporting AFM investigations of the PEN buffer layers proved the homogeneity of the prepared films and in particular did not show any indications of significant dewetting (cf. Supp. Inf., Figure S4). C. Interface Morphology and C60 Nucleation. To gain further insights into the different C60 film growth on PEN mono- and bilayer buffers, the initial fullerene nucleation is studied in more detail by means of AFM. For this purpose, at first a 1.5 ML (2.3 nm) PEN buffer layer is used, because it simultaneously provides both thicknesses of interest. Figure 3 shows the initial nucleation and evolution of C60 aggregates with increasing fullerene coverage. For better visualization of the morphology a non-monotonously color-coded height scale is used (for details see Supp. Inf., Figure S1), where the PEN mono- and bilayers are depicted in green and yellow, respectively, while the C60 clusters are shown in red as sketched schematically in Figure 3a. The same AFM micrographs are also shown in a monotonous color scale representation in the Supporting Information (cf. Figure S5). Note, that due to the finite resolution of the used AFM only C60 clusters are detected but not individual C60 molecules. After an initial deposition of only 0.02 nm (1/40 ML), isolated C60 clusters are found on the PEN monolayer. For the bilayers, in contrast, they appear exclusively at the rim of the terraces and not in their center. Increasing the fullerene thickness to 0.1 nm (1/8 ML) leads to an increased density of clusters on the monolayer buffer, whereas the center of the bilayer still remains uncovered and C60 clusters are still exclusively located at the rim. Only for an even larger coverage of about 0.2 nm (1/4 ML) when the step edge decoration is well advanced, fullerene aggregates start to form also in the center of the bilayer buffer. Comparing the size of the C60 aggregates on the PEN mono- and bilayer (cf. Figure 3e) moreover indicates a faster ripening of the islands on the bilayer, which suggests an enhanced diffusivity of impinging fullerene molecules. The initial C60 nucleation at PEN steps has been observed also in several earlier studies.28-30 In our previous work we have further demonstrated that the diffusivity of C60 depends sensitively on the substrate temperature which in particular allows to suppress the surface mobility of fullerenes on PEN bottom layers at temperatures below 240 K.29 Interestingly, the C60 clusters are exclusively located on the upper terrace of the steps which indicates the presence of a significant EhrlichSchwöbel barrier. Remarkably, C60 does not agglomerate at steps of the PEN monolayer on the SiO2 substrate, i.e. PEN-substrate steps (cf. blue region in Figure 3d). One possible reason for the different fullerene nucleation on the mono- and bilayer of PEN could be a difference in the film roughness. In this regard, however, it must be noted that the true molecular roughness cannot be determined by AFM due to the finite tip radius. To still investigate the influence of the buffer layer roughness on the C60 nucleation, we have chosen another preparation for the PEN layer yielding a larger mesoscopic roughness. For this purpose, we use a 1.5 ML PEN buffer layer grown onto a

Figure 3. AFM data showing the initial nucleation of C60 on 1.5 ML PEN buffer on a SiO2 substrate. (a) Scheme of the used colorcoded height scale. Micrographs after C60 depositions of (b) 0.02 nm (1/40 ML), (c) 0.1 nm (1/8 ML) and (d) 0.2 nm (1/4 ML), together with e) corresponding line scans across C60 aggregates formed on the PEN mono- and bilayer. sputtered graphite substrate, where molecules adopt an upright orientation and the substrate steps are overgrown by PEN islands thus creating imprinted “micro-steps” of 0.34 nm (corresponding to the graphite interlayer spacing),41,42 while the (001)PEN interlayer spacing amounts to about 1.5 nm. The corresponding AFM micrograph depicted in Figure 4 (a monotonous color scale representation is given in Supp. Inf., Figure S6), again shows equally distributed C60 clusters on the PEN monolayer (green) and an exclusive step decoration on thicker PEN layers (yellow and orange), without nucleation on the terrace center which parallels the situation observed before for SiO2 (cf. Figure 3c). In particular, clearly no preferential fullerene nucleation is found at the micro-steps on the pentacene island thicker than 1ML (yellow regions) as can be clearly seen in the respective micrograph (indicated by dashed lines in Figure 4a).

Figure 4. (a) AFM micrograph showing the interface morphology of 0.1 nm C60 grown on a 1.5 ML PEN buffer layer prepared on an intentionally roughened graphite substrate, together with (b) a scheme of the used color-coded height scale and (c) line scan showing the presence of micro-steps due to graphite surface steps imprinted to the PEN buffer layer (indicated by dashed lines).

4


Page 5 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


To understand the different C60 nucleation on the PEN mono- and bilayer a further phenomenon has to be taken into account. Several studies report that the PEN monolayer adopts a different crystalline phase than the subsequently grown thin films.44-47 The so-called orthorhombic PEN phase is only present in the first PEN layer, in which the molecules are aligned in nearly perfectly upright fashion, while the well-known TF-phase is formed in subsequent layers. This monolayer phase is hardly considered compared to the three dominant PEN polymorphs (TF, Siegrist and Campbell phase)48,49 and its impact on hetero-molecular growth has not been studied so far. Interestingly, Acevedo et al. report on a notable variation of the C60 diffusion on top of PEN substrates of the TF- and Campbell-phase.27 The presently observed differences in the C60 island formation and ripening on the PEN mono- and bilayer suggest significantly altered (surface) properties of the orthorhombic phase, which affects the diffusion as well as the probability of attachment/detachment processes of 2D islands and requires a more precise theoretical analysis to unravel its microscopic origin. So far, we have addressed the initial growth of C60 films on different PEN buffer layers. Now we will discuss the morphology of C60 films of larger thickness to understand the formation of the crystalline regions in these films. For that purpose, C60 films of 1 nm and 10 nm thickness were prepared on top of a PEN buffer layer of 1.5 ML thickness. As shown in Figure 5a,b, initially an efficient step edge decoration takes place, while for larger C60 coverage flattened islands are formed on the PEN bilayer. For the 10 nm C60 film these islands exhibit a distinct hexagonal contour with diameters of more than 250 nm. In addition, high-resolution AFM data reveal individual surface steps on such islands as depicted in Figure 5c. Their height distribution (cf. Figure 5d) reveals characteristic steps of 0.8 nm which agree favorably with the (111) lattice spacing of the face centered cubic C60 crystal structure (0.823 nm),36 which thus proves the crystalline nature of such islands. By contrast, the C60 clusters on the PEN monolayer are more spherical with typical diameters of less than 50 nm without recognizable surface steps as shown by the corresponding line scans in Figure 5e. Such spherical C60 clusters are also found in the AFM micrographs for fullerene films grown on a homogeneous PEN monolayer buffer (see Supp. Inf., Figure S4), for which the corresponding XRD measurements revealed no crystalline texture (cf. Figure 2d), hence indicating that these spherical fullerene clusters are indeed amorphous. Since their height exceeds the fullerene layer thickness, and they only develop during deposition at room temperature or above,29 this indicates that the shape of such nanoaggregates is controlled by the thermodynamic equilibrium. In contrast, the formation of planar, extended islands requires additional stabilization, which may be provided by interface energy due to epitaxial growth. Our investigation shows that the clusters on the PEN monolayers as well as those formed at the step edges apparently lead to amorphous C60 growth. Upon saturation of the step edges, however, also the terrace centers of the PEN bilayer (and multilayers) are covered. These, in contrast, experience distinct ripening with increasing C60 coverage. Their morphological evolution indicates that these are the nuclei for the formation of crystalline C60 islands.

Figure 5. AFM micrographs of (a) 1nm and (b) 10 nm C60 films deposited on top of 1.5 ML pentacene at 310 K, pentacene step edge is highlighted by the dashed white line. (c) Magnified AFM micrograph of a C60 island at the center of the PEN bilayer buffer (indicated in panel (b) showing molecular steps as visible in the height distribution in (d). (e) Line scans along C60 layers on the PEN mono- and bilayer (cf. b) reveal different morphologies. D. Heteroepitaxial C60 Growth on PEN Layers. As detailed in the previous sections, PEN buffer layers enable the preparation of crystalline C60 films at comparably low process temperatures. So far, the provided measurements have focused on the morphological properties as well as the vertical lattice orientation of these films. In addition, we are also interested in the lateral alignment of the fullerene adlayers and the question whether a hetero-epitaxy also exist on the PEN buffer layers, which crystallize in the device-relevant TF-phase. As shown in previous work by Nakayama et al.33 C60 films adopt an epitaxial alignment when grown on top of a PEN single crystal of the Siegrist phase. In that case, the 〈110〉 direction of the fullerene fcc lattice is aligned parallel to the [110] axis of the PEN crystal. To address this issue by means of in-plane X-ray diffraction measurements, one needs PEN buffer layers with distinct azimuthal alignment, for which PEN/SiO2 films are not suitable due to their isotropic in-plane orientation, shown in the Supporting Information, Figure S7. For that purpose, we use KCl(100) as a substrate, as PEN films grow epitaxially in this case.38,39 In order to obtain sufficient diffraction intensity for this analysis we use a thicker heterostructure consisting of a 200 nm C60 adlayer grown on a 150 nm PEN buffer. The corresponding specular X-ray diffractograms shown in Figure 6a reveal sharp diffraction signals reflecting a larger vertical coherence than the thinner films (cf. Figure 1) with a clear separation of the (002)PEN and (111)C60 reflections. The data confirm that the PEN films adopt a (001) orientation of the TF-phase while the fullerenes adopt a (111) orientation as expected. To derive the azimuthal alignment of the crystalline PEN films on KCl(100), in-plane Φ-scans of the (110)PEN reflection are performed. Therefore, the sample is tilted by 82.3° which is the

5



Page 6 of 9

C60 grows epitaxially also on PEN buffers of the TF-phase. Following the grammar of epitaxy,50,51 the epitaxial growth of C60 on PEN crystals of the Siegrist phase was described as line-on-line epitaxy with the [110]C60 direction aligned parallel to the [110]PEN direction.33 Assuming the same epitaxial relation also for PEN layers of the TF-phase, we calculate a lattice mismatch between the [110]PEN and [110]C60 directions of about 4.3%, which is actually smaller than the value of 6% found for heteroepitaxial growth of C60 on PEN crystals of the Siegrist polymorph.33 This suggests that the epitaxial growth of C60 on PEN in the TF-phase might even be more stable than the one on PEN crystals of the Siegrist phase. Finally, we note that also C60 films grown directly onto KCl(100) surfaces at elevated temperatures reveal an epitaxial ordering. In that case, however, the epitaxial alignment is distinctly different,37 so that the observed hetero-epitaxial ordering can be clearly attributed to the PEN buffer and not to fullerene island formation on uncovered KCl regions.

CONCLUSIONS

Figure 6. (a) Specular X-ray diffractograms of 200 nm C60/150 nm pentacene on KCl(100), (b) azimuthal in-plane scans of the (110)PEN reflection (blue) and the 220 C60 reflection (green), together with a visualization of the unit cells of (c) PEN and (d)

C60 showing the lattice planes used for the in-plane XRD measurements.

angle between the chosen (110)PEN lattice plane and the (001)PEN growth direction (cf. Fig. 6c and Supporting Information, Figure S7). As shown in Figure 6b (blue curve), discrete and sharp maxima are observed which prove the epitaxial growth of PEN on KCl(100). This observation as well as the peak positions are perfectly in line with previous results from Kakudate et al. Their quantitative analysis showed that the crystalline PEN islands are aligned with their crystallographic [100], [210] and [210] directions along the 〈010〉KCl azimuth directions.38,39 An equivalent analysis for the C60 film grown on top then allows studying the hetero-epitaxial alignment. A corresponding Φ-scan of the 220 C60 reflection (green curve in Figure 6b) indeed also shows discrete signals, which occur at similar Φ angles as the (110)PEN reflections. Clearly, the observed signals exhibit different broadness, which is attributed to an larger in-plane mosaicity of the C60 layer as compared to the underlying PEN films. Due to the distinctly lower in-plane diffraction intensity compared to outof-plane measurement, larger slits are required to obtain sufficient intensity. Unfortunately, this hampers a precise determination of the registration as well as a quantitative characterization of the mosaicity. Nevertheless, we can safely confirm that

In this study, we elucidate the influence of organic buffer layers on the structural and morphological characteristics of subsequently processed fullerene films. Pre-coating substrates with thin (001)-oriented PEN buffer layers yields crystalline, even epitaxial fullerene films upon deposition at 310 K, while only quasiamorphous C60 films are formed at this growth temperature on the bare substrates. Hence, such buffer layers provide a simple strategy to prepare crystalline C60 films at low temperature. Remarkably, this crystalline ordering is not observed for PEN monolayers as buffer but only occurs for PEN thicknesses larger than 1 ML. Thus, the diffusion of the C60 molecules on top of the first PEN monolayer appears reduced. Excluding a smoothing factor of the PEN layers and an influence of surface roughness by substrate variation, we relate this to the different molecular packing adopted by PEN molecules in the orthorhombic monolayer compared to the TF-phase in thicker layers. This unexpected result certainly requires further theoretical analysis to identify the underlying molecular mechanism. Using amorphous but smooth SiO2 wafer surfaces, sufficiently large and molecularly flat PEN terraces of more than 300 nm diameter are prepared which then allowed us to identify important structural details such as initial C60 nucleation sites and thickness dependent ripening processes, which we correlate with the crystalline respectively amorphous growth of the fullerenes. We derive that the initial decoration of PEN steps causes the formation of amorphous C60 clusters, and only after saturation of these steps crystalline, even epitaxially aligned, fullerene islands are formed at the center of the PEN terraces. Interestingly, steps of lower height such as substrate steps that are overgrown by PEN islands do not seem to affect the diffusion and cause no nucleation. This reveals a minor influence of the PEN surface roughness on the C60 diffusion and nucleation indicating that the fullerene nucleation is distinctly affected by the presence of PEN steps. This can be attributed to an Ehrlich-Schwöbel barrier, while imprinted substrates micro-steps apparently don’t exhibit such barriers. In view of the frequently observed rapid roughening of organic films with increasing thickness,52,53 it seems to be important to use PEN buffers with the lowest possible step density in order to obtain fullerene films with high

6


Page 7 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


crystallinity. Mostly, such idealized Si wafer substrates are not chosen for the fabrication of flexible electronic devices, where typically plastic substrates are used. These, however have such distinctly larger roughness, that a microscopic analysis of nucleation phenomena by means of scanning probe microscopy is not possible. The utilized idealized substrates therefore allow the identification of these effects which then can be transferred also to plastic foil substrates. The present study provides microscopic insights into initial nucleation phenomena upon growth of molecular hetero-layers and especially emphasizes the influence of the buffer layer polymorphs on the templating effect. These results will help optimizing process parameters and deriving strategies for a better understanding of hetero-growth, both, theoretically as well as experimentally.

[3] Karl, N. Charge Carrier Transport in Organic Semiconductors. Synth. Met. 2003, 133-134, 649-657. [4] Frankevich, E.; Maruyama, Y.; Ogata,H. Mobility in Charge Carrieres in Vapor-Phase-Grown C60 Single Crystal. Chem. Phys. Lett. 1993, 214(1), 39-44. [5] Xu, W. Preparing (111)-Oriented C60 Crystalline Films on NaCl Substrate. J. Cryst. Growth 2000, 220, 96-99. [6] Itaka, K.; Yamashiro, M.; Yamaguchi, J.; Haemori, M.; Yaginuma, S.; Matsumoto, Y.; Kondo, M.; Koinuma, H. High-Mobility C60 Field-Effect Transistors Fabricated on Molecular- Wetting Controlled Substrates. Adv. Mater. 2006, 18, 1713-1716. [7] Sekitani, T.; Iba, S.; Kato, Y.; Someya, T. Pentacene Field-Effect Transistors on Plastic Films Operating at High Temperature above 100°C. Appl. Phys. Lett. 2004, 85, 3902.

ASSOCITED CONTENT Supporting Information This material is available free of charge via the Internet at http://pubs.acs.org. Detailed information on the processing of the AFM micrographs, morphological characterization of the various substrate surfaces and PEN buffer layers as well as additional large angular range X-ray diffraction data and comparison of isotropic vs. epitaxial PEN buffer (PDF).

[8] Lim, S. C.; Kim, S. H.; Lee, J. H.; Yu, H.Y.; Park, Y.; Kim D.; Zyung, Z. Organic Thin-Film Transistors on Plastic Substrates. Mater. Sci. Eng., B 2005, 121, 211-215. [9] Brinkmann, M.; Graff, S.; Straupe, C.; Wittmann, J.-C.; Chaumont, C.; Nuesch, F.; Aziz, A.; Schaer, M.; Zuppiroli, L. Orienting Tetracene and Pentacene Thin Films onto Friction-Transferred Poly(tetrafluoroethylene) Substrate. J. Phys. Chem. B 2003, 107, 10531-10539.

AUTOR INFORMATION Corresponding Author *E-mail [email protected]

[10] Salzmann, I.; Duhm, S.; Opitz, R.; Johnson, R. L.; Rabe, J. P.; Koch, N. Structural and Electronic Properties of Pentacene-Fullerene Heterojunctions. J. Appl. Phys. 2008, 104, 114518.

ORCID Gregor Witte: 0000-0003-2237-0953 Tobias Breuer: 0000-0002-9962-9444 Andrea Huttner: 0000-0002-7441-077X

[11] Cosseddu P.; Bonfiglio, A.; Salzmann, I.; Rabe, J.P.; Koch, N. Ambipolar Transport in Transparent and Flexible All-Organic Heterojunction Field Effect Transistors at Ambient Conditions. Org. Electron. 2008, 9(1), 191-197.

Notes The authors declare no competing financial interest.

[12] Xiaoyu, L.; Xiaoman, C.; Boqun, D.; Xiao, B.; Jianfeng, F. Enhanced Performance of C60 N-Type Organic Field-Effect Transistors Using a Pentacene Passivation Layer. J. Semicond. 2013, 34(8), 084002.

ACKNOWLEDEGMENTS We acknowledge support by the Deutsche Forschungsgemeinschaft (Grant SFB 1083, TP A2) and the Friedrich-Ebert-Stiftung (A. H.)

REFERENCES [1] Bredas, J. L.; Marder, S. R. The WSPC Reference on Organic Electronics: Organic Semiconductors, Materials and Energy: Volume 7; World Scientific, 2016. [2] Mzukami, M.; Cho, S.; Watanabe, K.; Abiko, M.; Suzuri, Y.; Tokit, S.; Kido, J. Flexible Organic Light-Emitting Diode Display Driven by Injekt-Printed High-Mobility Organic Thin-Film Transistors. IEEE Electron Device Lett. 2018, 39 (1), 39-42.

[13] Ahn, K.; Kim, J. B.; Park, H.; Kim, H.; Lee, M. H.; Kim, B. J.; Cho, J. H.; Kang, M. S.; Lee, D. R. Enhancing Crystallinity of C60 Layer by Thickness-Control of Underneath Pentacene Layer for High Mobility C60/Pentacene Ambipolar Transistors. Appl. Phys. Lett. 2013, 102, 043306. [14] Tsai, L. S.; Hwang, J. C.; Lee, C. Y.; Lin, Y. T.; Tsai, C. L.; Chang, T.H.; Chueh, Y. L.; Meng, H. F. Solution-Based Silk Fibroin Dielectric in n-Type C60 Organic Field-Effect Transistors: Mobility Enhancement by the Pentacene Interlayer. Appl. Phys. Lett. 2013, 103(23), 233304. [15] Walker, B.; Kim, C.; Nguyen, T. Q. Small Molecule SolutionProcessed Bulk Heterojunction Solar Cells. Chem. Mater. 2011, 23, 470–482.

7



[16] Congreve, D. N.; Lee, J.; Thompson, N. J.; Hontz, E.; Yost, S. R.; Reusswig, P. D.; Bahlke, M. E.; Reineke, S.; Van Voorhis, T.; Baldo, M. A. External Quantum Efficiency Above 100% in a Singlet-Exciton-Fission-Based Organic Photovoltaic Cell. Science, 2013, 340 (6130), 334-337. [17] Cantrell, R.; Clancy, P. A Computational Study of Surface Diffusion of C60 on Pentacene. Surf. Sci. 2008, 602, 3499-3505. [18] Bommel, S.; Kleppmann, N.; Weber, C.; Spranger, H.; Schäfer, P.; Novak, J.; Roth, S. V.; Schreiber, F.; Klapp S. H. L.; and Kowarik, S. Unravelling the Multilayer Growth of the Fullerene C60 in RealTime. Nat. Commun. 2014, 5, 5388. [19] Muccioli, L.; D’Avino, G.; Zannoni, C. Simulation of VaporPhase Deposition and Growth of a Pentacene Thin Film on C60 (001). Adv. Mater 2011, 23 (39), 4532-4539. [20] Fu, Y. T.; Risko, C.; Brédas, J.-L., Intermixing at the Pentacene-Fullerene Bilayer Interface: a Molecular Dynamics Study. Adv. Mater. 2013, 25, 878. [21] Oliver, R. A.; Kappers, M.J.; Humphreys, C.J.; Briggs, G.A.D. Growth Modes in Heteroepitaxy of InGaN on GaN. J. Appl. Phys. 2005, 97, 013707. [22] Evans, J. W.; Thiel, P. A.; Bartelt, M.C. Morphological Evolution During Epitaxial Thin Film Growth: Formation of 2D Islands and 3D Mounds. Surf. Sci. Rep. 2006, 61, 1. [23] Walter, E. C.; Murray, B. J.; Favier, F.; Kaltenpoth, G.; Grunze, M.; Penner, R. M. Noble and Coinage Metal Nanowires by Electrochemical Step Edge Decoration, J. Phys. Chem. B, 2002, 106, 44. [24] Simmondsa, P. J.; and Lee, M. L. Tensile Strained Island Growth at Step-Edges on GaAs(110). Appl. Phys. Lett. 2010, 97, 153101. [25] Wang, W.; Du, C.; Zhong, D.; Hirtz, M.; Wang, Y.; Lu, N.; Wu, L.; Ebeling, D.; Li, L.; Fuchs, H.; Chi, L. Control over Patterning of Organic Semiconductors: Step-Edge-Induced Area-Selective Growth. Adv. Mater. 2009, 21, 4721–4725. [26] Aït-Mansour, K.; Ruffieux, P.; Xiao, W.; Fasel, R.; Gröning, P.; Gröning, O. Nucleation and Growth of C60 Overlayers on the Ag/Pt(111) Dislocation Network Surface. J. Phys. Conf. Ser. 2007, 61, 16. [27] Acevedo, Y.M.; Cantrell, R.A.; Berard, P.G.; Koch, D.L.; Clancy, P. Multiscale Simulation and Modeling of Multilayer Heteroepitactic Growth of C60 on Pentacene. Langmuir 2016, 32 (12), 3045-3056. [28] Conrad, B. R.; Tosado, J.; Dutton, G,; Dougherty, D. B.; Jin, W.; Bonnen, T.; Schuldenfrei, A.; Cullen, W. G.; Williams, E. D.; ReuttRobey, J. E.; Robey, S. W. C60 Cluster Formation at Interfaces with Pentacene Thin-Film Phases. Appl. Phys. Lett. 2009, 95, 213302.

Page 8 of 9

[29] Breuer, T.; Witte, G. Diffusion-Controlled Growth of Molecular Heterostructures: Fabrication of Two-, One-, and Zero-Dimensional C60 Nanostructures on Pentacene Substrates. ACS Appl. Mater. Interfaces 2013, 5, 9740−9745. [30] Kyndiah, A.; Cramer, T.; Albonetti, C.; Liscio, F.; Chiodini, S.; Murgia, M.; Biscarini, F. Charge Transfer and Percolation in C60/Pentacene Field-Effect Transistors. Adv. Electron. Mater. 2015, 1(11), 1400036. [31] Breuer, T.; Karthäuser, A.; Witte, G. Effects of Molecular Orientation in Acceptor-Donor Interfaces between Pentacene and C60 and Diels- Alder Adduct Formation at the Molecular Interface. Adv. Mater. Interfaces 2016, 3(7), 1500452. [32] Breuer, T.; Geiger, T.; Bettinger, H. F.; Witte, G. Diels-Alder Adduct Formation at Solid Interfaces between Fullerenes and Acenes. J. Phys.: Condens. Matter 2019, 31, 034003. [33] Nakayama, Y.; Mizuno, Y.; Hosokai, T.; Koganezawa, T.; Tsuruta, R.; Hinderhofer, A.; Gerlach, A.; Broch, K.; Belova, V.; Frank, H.; Yamamoto, M.; Niederhausen, J.; Glowatzki, H.; Rabe, J.P.; Koch, N.; Ishii, H.; Schreiber, F.; Ueno, N. Epitaxial Growth of an Organic p–n Heterojunction: C60 on Single-Crystal Pentacene. Appl. Mater. Interfaces 2016, 8(21) ,13499-13505. [34] Schiefer, S.; Huth, M.; Dobrinevski, A.; Nickel, B. Determination of the Crystal Structure of Substrate-Induced Pentacene Polymorphs in Fiber Structured Thin Films. J. Am. Chem. Soc. 2007, 129 (34), 10316–10317. [35] Käfer, D.; Wöll, Ch.; Witte, G. Thermally Activated Dewetting of Organic Thin Films: the Case of Pentacene on SiO2 and Gold. Appl. Phys. A: Mater. Sci. Process. 2009, 95, 273. [36] Dorset, D. L.; McCourt, M. P. Disorder and the Molecular Packing of C60 Buckminsterfullerene: A Direct Electron-Crystallographic Analysis. Acta Crystallogr., Sect. A: Found. Crystallogr. 1994, 50, 344−351. [37] Dai, Z.; Naramoto, H.; Narumi, K.; Yamamoto, S.; Miyashita, A. A Three-Step Process for Epitaxial Growth of (111)-Oriented C60 Films on Alkali–Halide Substrates. Thin Solid Films 2000, 360, 28. [38] Kakudate, T.; Yoshimoto, N.; Kawamura, K.; Saito, Y. Observation of Epitaxial Growth of Pentacene Thin Films on KCl Substrate by X-Ray Diffractometry. J. Cryst. Growth 2007, 306, 27. [39] Kiyomura, T.; Nemoto, T.; Yoshida, K.; Minari, T.; Ku-rata, H.; Isoda, S. Epitaxial Growth of Pentacene Thin-Film Phase on Alkali Halides. Thin Solid Films 2006, 515, 810. [40] Seydel, E.; Hoffmann-Vogel, R.; Marz, M. Epitaxial Growth of C60 on Highly Oriented Pyrolytic Graphite Surfaces Studied at Low Temperatures. Nanotechnology 2019, 30, 025703. [41] Götzen, G.; Käfer, D.; Wöll, C.; Witte, G. Growth and Structure of Pentacene Films on Graphite: Weak Adhesion as a Key for Epitaxial Film Growth. Phys. Rev. B 2010, 81, 085440.

8


Page 9 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[42] Breuer, T.; Salzmann, I.; Götzen, J.; Oehzelt, M.; Morherr, A.; Koch, N.; Witte, G. Interrelation between Substrate Roughness and Thin-Film Structure of Functionalized Acenes on Graphite. Cryst. Growth Des. 2011, 11, 4996.

TOC graphic

[43] Trucano, P.; Chen, R. Structure of Graphite by Neutron Diffraction. Nature 1975, 258, 136-137. [44] Mannsfeld, S. C. B.; Virkar, A.; Reese, C.; Toney, M. F.; Bao, Z. Precise Structure of Pentacene Monolayers on Amor-phous Silicon Oxide and Relation to Charge Transport. Adv. Mater. 2009, 21, 2294–2298. [45] Fritz, S. E.; Martin, S. M.; Frisbie, C. D.; Ward, M. D.; Toney, M. F. Structural Characterization of a Pentacene Monolayer on an Amorphous SiO2 Substrate with Grazing Incidence X-Ray Diffraction. J. Am. Chem. Soc. 2004, 126, 4084-4085. [46] Cheng, H. L.; Mai, Y.-S.; Chou, W.-Y.; Chang, L.-R.; Liang, X. W. Thickness-Dependent Structural Evolutions and Growth Models in Relation to Carrier Transport Properties in Polycrytalline Pentacene Thin Films. Adv. Funct. Mater. 2007, 17, 3639–3649. [47] Zhang, L.: Hu, C.; Fu, X.; Jiang, M.; Sui, Q.; Jia, L. Pentacene Crystal Transition During the Growth on SiO2 Studied by in Situ Optical Spectroscopy. Synthetic Metals 2017, 231 65–69. [48] Bouchoms, I.P.M.; Schoonveld, W. A.; Vrijmoeth, J.; Klapwijk, T.M. Morphology Identification of the Thin Film Phases of Vacuum Evaporated Pentacene on SIO Substrates. Synth. Met. 1999, 104, 175. [49] Siegrist, T.; Kloc, C.; Schön, J. H.; Batlogg, B.; Haddon, R. C.; Berg, S.; Thomas, G. A. Enhanced Physical Properties in a Pentacene Polymorph. Angew. Chem., Int. Ed. 2001, 40, 1732−1736. [50] Mannsfeld, S .C .B.; Leo, K.; Fritz, T. Line-on-Line Coincidence: A New Type of Epitaxy Found in Organic-Organic Heterolayers. Phys. Rev. Lett. 2005, 94, 056104.

[51] Hooks, D. E.; Fritz, T.; Ward, M.D. Epitaxy and Molecular Organization on Solid Substrate. Adv. Mater. 2001, 13, 4. [52] Dürr, A.C.; Schreiber, F.; Ritley, K.A.; Kruppa, V.; Krug, J.; Dosch, H.; Struth, B. Rapid Roughening in Thin Film Growth of an Organic Semiconductor (Diindenoperylene), Phys. Rev. Lett. 2003, 90, 016104. [53] Kowarik, S.; Gerlach, A.; Leitenberger, W.; Hu, J.; Witte, G.; Wöll, C.; Pitsch, U.; Schreiber, F. Energy-Dispersive X-Ray Reflectivity and GID for Real-Time Growth Studies of Pentacene Thin Films. Thin Solid Films 2007, 515, 5606-5610.

9


Controlling Interface Morphology and Layer Crystallinity in Organic

Recommend Documents