Article pubs.acs.org/JPCC
Thin Film Growth of Pentacene on Polymeric Dielectrics: Unexpected Changes in the Evolution of Surface Morphology with Substrate T. V. Desai,† A. R. Woll,‡ and J. R. Engstrom*,† †
School of Chemical and Biomolecular Engineering, and ‡Cornell High Energy Synchrotron Source, Cornell University, Ithaca, New York 14853, United States S Supporting Information *
ABSTRACT: We have examined the thin film growth of pentacene on SiO2 and on three different polymeric dielectrics using in situ synchrotron X-ray scattering and ex situ atomic force microscopy (AFM). The polymeric dielectrics investigated spanned the range from a low surface energy hydrophobic surface (polystyrene, PS) to a high surface energy hydrophilic surface [poly(ethylene imine), PEI]. On all surfaces, pentacene forms a polycrystalline thin film, whose structure is that of the previously identified “thin film” phase. From in situ real-time X-ray scattering, we find that pentacene exhibits layer-by-layer (LbL) growth on all surfaces investigated, but the extent of LbL growth is a strong function of the underlying substrate. In particular, LbL growth is significantly more prolonged on PEI and least extended on PS. The roughness and the in-plane feature sizes of thick, ∼10 monolayer, pentacene thin films also vary with the surface energy of the substrategrowth on the high surface energy polymer thin film, PEI, is the smoothest and is characterized by the largest features. It appears that interlayer transport is influenced by the underlying substrate, even for layers that are not in direct contact with the polymer dielectric.
I. INTRODUCTION The study of complex conjugated molecules, e.g., pentacene, for applications in organic thin film devices for electronics and photonics has received much attention owing to their ability to form crystalline thin films with excellent electrical properties.1−4 Previous work has shown that the interface between the organic semiconducting layer and the dielectric is critical to charge transport and that the majority of charge carriers is generated in the first few monolayers (MLs) of the organic layer.5−8 Thus, there is considerable motivation to investigate in detail the first few monolayers of organic thin film growth. Despite the importance of the structure of this interface, there is still a significant lack of understanding of many features of organic crystal growth, especially concerning the potential role played by the substrate and its effect on small molecule organic thin film growth. Due to their ease in processing, flexibility, and tunable surface properties, polymeric dielectrics are attractive for use in devices.9 Indeed, the great promise of organic electronics is based on the idea of using flexible polymeric substrates. Previous studies have shown that the nature of the polymeric dielectric can strongly affect both the morphology and the electrical properties of the deposited pentacene thin films, although the exact nature of these effects is still a matter of debate. One early study showed that the electrical mobility of pentacene thin films deposited on a polymeric substrate could be higher than a comparable film deposited on clean silicon dioxide (SiO2),10 an observation which spurred subsequent investigations into polymeric dielectrics. In one study a systematic variation of the chemical structure of styrenic polymers did not reveal any clear correlations between thin film © 2012 American Chemical Society
structure and electrical properties, even for cases where the electrical mobility varied by nearly an order of magnitude.11 In other work,12,13 however, investigating a similar set of styrenic polymers, but also poly(methylmethacrylate) (PMMA)12 and polyvinyl alcohol (PVA),13 a connection between thin film morphology (“grain size”) and electrical mobility was reported, where there apparently exists a minimum grain size to achieve superior electronic properties. In this work, perhaps the most compelling identification was the importance of mobility of the polymeric chains at the dielectric surface, as gauged by the glass transition temperature. In more recent work, these same investigators have found that the degree of cross-linking in the polymeric dielectric,14 which also reduces the mobility of the polymeric chains, can have similar effects on the electrical properties of the pentacene thin films. Other work has investigated the possible effects of the surface energy of the polymeric dielectric on the electrical properties of the pentacene thin films.15−17 In one of these studies,15 superior properties were reported on a polymeric dielectric that produced pentacene thin films characterized by smaller, not larger, grain sizes. Despite these somewhat contradictory observations concerning the effects of thin film structure on the electrical properties, it should be fairly obvious that microstructure is an important factor in determining these properties. How one characterizes the thin film microstructure is an equally important issue. In most studies to date of organic small molecule thin film growth Received: January 18, 2012 Revised: May 10, 2012 Published: June 1, 2012 12541
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
Article
chambers: a main scattering chamber, a source and antechamber, which act to produce the supersonic beam, and a fast entry load-lock. All chambers are pumped by highthroughput turbomolecular pumps. The base pressure of the chamber was typically ∼4 × 10−9 Torr, and samples were loaded via the load-lock chamber, which was evacuated to ∼10−7 Torr prior to sample transfer into the main chamber. Xray reflectivity (XRR) and grazing incidence diffraction (GID) experiments, conducted ex situ, were carried out in the G2 station at CHESS. Substrates were Si(100) wafers (Wacker-Siltronic, p-type, 100 mm dia., 500−550 μm thick, 38−63 Ω-cm) subject to a SC-1 clean, 15 s HF dip, and a SC-2 clean followed by growth of ∼300 nm thick SiO2 films by wet thermal oxidation at 1100 °C. Next, these wafers were cleaned and degreased by sonication for 15 min in anhydrous CHCl3 solution (99%+), sonicated in deionized (DI) water for 15 min, washed with DI water, dried with N2, and exposed to UV-ozone for 15 min. These processes provided a clean and reproducible hydrophilic surface. The chemical structures of the polymers investigated here are shown in Figure 1. PS (Mn = 200 K g-mol−1, 1 wt % in
on polymeric dielectrics, two techniques are most commonly employed to quantify the thin film morphology and structure: ex situ atomic force microscopy (AFM) and ex situ X-ray diffraction (XRD). AFM can be used to examine submonolayer thin films, as well as the morphological evolution of thicker thin films. In one study, the submonolayer nucleation of pentacene was examined on SiO2 and a handful of polymeric dielectrics. The most significant result of this study was that the critical nucleus, i*, for the growth of pentacene on these surfaces was constant, i* ∼ 3−4 molecules, independent of the polymeric dielectric,18 in good agreement with other work conducted on SiO2.19−22 Also notable in this work was the lack of observation of multilayer high islands for submonolayer coverages reported elsewhere.15,16 In subsequent work on submonolayer nucleation on polymeric dielectrics,23 it was confirmed that the critical nucleus of pentacene was on the order of i* = 3−6, while the density of islands formed at a constant growth rate varied by as much as a factor of 5 on the different dielectrics. Such a variation was associated with a change in the surface diffusivity of pentacene on the different dielectric surfaces, a result that has also been observed concerning the submonolayer nucleation of perfluoropentacene on surfaces terminated by different self-assembled monolayers.24 Here, we investigate the growth of pentacene thin films on three polymeric dielectric thin films and on clean unmodified SiO2, concentrating on the evolution of the thin film morphology using a combination of both in situ and ex situ surface sensitive techniques. Concerning the polymers, we consider three thin film materials that differ in their chemical nature: polystyrene (PS), poly(methylmethacrylate) (PMMA), and poly(ethylene imine) (PEI), where one of these (PEI) also differs from the other two in terms of its thickness. For comparison, we also consider unmodified SiO2. We shall see that, in terms of surface energies, the surfaces follow the order (high-to-low): PEI, unmodified SiO2, PMMA, and PS. We deposit thin films of pentacene in ultrahigh vacuum (UHV) using a collimated supersonic molecular beam, which affords precise control of the kinetic energy of the incidence molecules.22,25,26 We make use of ex situ AFM to probe the thin film morphology and employ in situ real-time synchrotron X-ray scattering measurements at the “anti-Bragg” configuration26−28 to directly probe the filling of each successive molecular layer of pentacene during thin film growth. A great advantage of using in situ real-time X-ray scattering techniques is that it essentially eliminates any artifacts associated with postgrowth thin film reorganization events, such as “dewetting”. Sole reliance on ex situ techniques, such as AFM, can be problematic concerning the growth of small molecule organic crystalline thin films on the relatively low energy surfaces associated with self-assembled monolayers29 and polymers. Using a combination of in situ real-time X-ray scattering and ex situ AFM, we will find that the nature of the polymer affects significantly both the growth mode and the thin film roughness of the pentacene thin films, even for thicknesses where the underlying surface is completely covered by the deposited thin film.
Figure 1. Molecular structure of the polymers examined in this study: polystyrene (PS); poly(methylmethacrylate) (PMMA); and branched poly(ethylene imine) (PEI).
toluene) and PMMA (Mn = 495 K g-mol−1, 2 wt % in anisole) thin films were deposited on the cleaned SiO2 substrates by spin coating. The spin-coated PMMA films were annealed for 15 min at 170 °C on a hot plate. Branched PEI (Mw = 750 K gmol−1, Mn = 60K g-mol−1, 0.1 wt % in DI water) films were deposited by dipping cleaned SiO2 substrates in solution for 15 min followed by drying with N2. The SiO2 substrates modified with PS, PMMA, and PEI were characterized by contact angle, XRR, and AFM. Contact angles were measured in two solvents (water and formamide), and using the Young−Dupre equation30 we calculated the energy of the surfaces modified with thin films of PS, PMMA, and PEI and found values of 25.0, 35.6, and ≥73.4 mJ m−2, respectively. The inequality for the value for PEI is due to the fact that both water and formamide completely wet the surface. In comparison, the surface energy of clean, unmodified SiO2 has been reported to lie between 50 and 60 mJ m−2,31,32 and the surface energy of the (001) crystal plane of pentacene is
II. EXPERIMENTAL PROCEDURES The experiments that were conducted in situ and in real time were carried out in the G3 station of the Cornell High Energy Synchrotron Source (CHESS) in a custom-designed UHV system fitted with Be windows that is described elsewhere.25 Briefly, the system consists of four separately pumped 12542
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
Article
Table 1. Properties of the Polymer Thin Films polymer
contact angle (H2O/formamide)
surface energy (mJ m−2)
thickness, XRR (nm)
roughness, AFM (nm)
electron density (Å−3)
PS PMMA PEI
92°/74° 70°/56° ∼0°/0°
25.0 35.6 ≥73.4
23.9 50.2 0.86 ± 0.03
0.29 0.36 0.34
0.44 0.49 0.35 ± 0.04
reported to be 50−82 mJ m−2.33−35 Fits to the XRR data were performed, as described in detail elsewhere,28 with the Parratt32 software package36 (based on the Parratt formalism37), from which we obtain the thickness of the organic layers and the mean electron density. Details concerning the properties of the SAMs are given in Table 1. Supersonic molecular beams of pentacene were generated by using He as a carrier gas as described in detail elsewhere.27 By varying the He flow rate, the mean kinetic energy of the pentacene molecules in the beam, Ei, could be varied from 2.5 to 7.0 eV as determined from time-of-flight measurements.26 Multiple experiments could be carried out on the same substrate, by translating the substrate perpendicular to the supersonic molecular beam and due to the high beam-tobackground flux ratio. During deposition the substrate temperature was kept at Ts = 40 °C, and in all cases the beam was incident along the surface normal. Time-resolved and in situ measurements of the scattered Xray synchrotron intensity were made using a silicon avalanche photodiode detector (APD, Oxford Danfysik, Oxford, UK). During pentacene thin film growth, the intensity was monitored at the anti-Bragg position (001/2; qz = qBragg/2 = 0.41/2 Å−1), which is an effective monitor of the nature of growth, i.e., layerby-layer (LbL) vs 3D islanded growth.38 Following deposition and X-ray analysis, the samples were removed for ex situ analysis using AFM, conducted in tapping mode using a DI 3100 Dimension microscope. The X-ray data at the anti-Bragg position was fitted using a modified version27,39 of the mean-field rate equation model of growth first proposed by Cohen and co-workers.40 Briefly, the equations for the coverage of individual layers (θn) are given by
∞
I(t ) = |rsubse−iϕ + rfilm ∑ θn(t )e−iqzdn|2 n
(2)
where rsubs and rfilm are the scattering amplitudes of the substrate and the film; ϕ is the phase change upon reflection; qz is the out-of-plane scattering vector; and d is the out-of-plane interplanar spacing. At the anti-Bragg position, qzd = π, which results in a change in the sign of the thin film terms in the summation. If each layer fills sequentially, such as in perfect LbL growth, an oscillation in the intensity results.
III. RESULTS AND DISCUSSION A. Characterization of the Polymer Thin Films. The polymer thin films were characterized using synchrotron X-ray reflectivity measurements, contact angle measurements, and AFM, as described above in Section II. First, we consider the results from XRR. In Figure 2, we plot the reflected intensity as a function of the out-of-plane scattering vector, qz, for clean
dθn = Sn − 1F[(θn − 1 − θn) − αn − 1(θn − 1 − θn)] dt + SnFαn(θn − θn + 1)
(1)
where n = 0 represents the substrate, n = 1 the first molecular layer, etc. Sn is the probability of adsorption for molecules incident on the nth layer; F is the incident molecular flux (ML s−1); and αn is the fraction of molecules that initially impact and land on top of the nth layer but rather than staying on the top of that layer drop down and become part of that layer via some mechanism. In this model, we also assume that there are two values for the probability of adsorption: one for adsorption on the substrate (S0) and one for that on previously existing molecular layers, independent of their thickness (S1 = S2 = S3 ...). Concerning interlayer transport, we will assume that three values are possible (note, as the substrate cannot be penetrated, α0 = 0), namely, α1, α2, and αn≥3. “Upward” interlayer transport (movement from the n to the n + 1 layer) is not included in the model. Once layer coverages have been calculated by integrating eq 1, these can then be used to calculate the scattered X-ray intensity as a function of time.27,38−40 The intensity of the scattered beam (I) depends upon the layer population, θn(t), according to the following relationship
Figure 2. Scattered X-ray intensity as a function of the out-of-plane scattering vector, qz, for thin films of (a) PEI and (b) PS. The solid curves represent fits of the data to a model based on the Parratt formalism (ref 37). 12543
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
Article
SiO2 modified with (2a) PEI and (2b) PS. In these figures, for PEI|SiO2 (PS|SiO2) we plot only 10% (50%) of the data [1 of every 10 (2) points] to show the quality of the fits, which are shown by the smooth curves. These two polymer thin films exhibit dramatically different X-ray reflectivity curves, which is a direct consequence of their thickness. Many more thickness fringes are observed for PS|SiO2, and the spacing between these (Δqz) is significantly smaller that the single feature observed for PEI|SiO2. From a fit to the data, we extract thicknesses of dPEI = 0.86 nm (8.58 Å) and dPS = 23.9 nm, and these values and other properties for the polymer thin films are given in Table 1. For the other polymer thin film we consider here, PMMA|SiO2, the XRR data are similar to that shown for PS|SiO2, and we find a thickness of dPMMA = 50.2 nm. To evaluate the roughness and spatial uniformity of the polymer thin films, we have employed AFM. We find that the as-deposited polymer thin films in all three cases are uniform and smooth, with the rms roughness laying in the range 0.29−0.36 nm (cf. Table 1). The static contact angles (water) and surface energies for PS and PMMA agree well with values reported in the literature,9,13,17,18 and these are also given in Table 1. B. Thin Film Growth of Pentacene. To characterize the growth of pentacene thin films on SiO2 and on the polymer thin films, we have conducted the following experiments. First, we deposit pentacene thin films under UHV conditions, monitoring in situ and in real time the scattered X-ray intensity at the “anti-Bragg” condition. Thin films of a nominal thickness of ∼8−10 MLs are grown in all cases. Following deposition, these thin films are then removed from the UHV system, and ex situ AFM is employed to evaluate the morphology of these multilayer thin films. We have not attempted the growth of subor near-monolayer thin films, coupled with ex situ AFM, due to concerns over postdeposition reorganization and “de-wetting”, particularly on the low-energy surfaces of PS|SiO2 and PMMA| SiO2. In previous work, we have observed significant postdeposition reorganization for the growth of sub- and near-monolayer pentacene thin films on self-assembled monolayers, which also represented low-energy surfaces.29 In this same work, however, postdeposition reorganization was not observed (or occurred at a rate not observable in the time scale of the experiments) concerning thicker films, such as we consider here. Thus, here we will rely upon in situ real-time Xray scattering to characterize the early stages of growth, and we use AFM only to characterize the thicker multilayer thin films. 1. Growth on SiO2. We begin with the results for growth of pentacene on SiO2, which we have reported on previously26 but we reproduce here in a separate set of experiments conducted in concert with the study of growth on the polymer thin films. As noted in the Introduction, a number of investigators have pointed out reasons as to why growth on polymers may be different from that on polymer dielectrics. Thus, SiO2 will provide an important point of comparison. In Figure 3, we present results from both in situ X-ray scattering and ex situ AFM for growth of pentacene on SiO2. In Figure 3a, we plot the scattered intensity measured in situ and in real time for the growth of pentacene on SiO2 at Ei = 2.5 eV and a growth rate, GR = 0.0069 ML s−1. As may be seen, for growth on this surface we observe a small peak, corresponding to the approximate completion of the first ML, followed by a much more intense cusp-like peak, which coincides with completion of the second ML. The expected small maximum corresponding to completion of the third ML is largely obscured, while the larger maximum corresponding to
Figure 3. (a) X-ray intensity at the anti-Bragg condition as a function of exposure to the molecular beam for a thin film of pentacene deposited on SiO2 at Ei = 2.5 and Ts = 40 °C. Thick solid (blue) line (right ordinate) indicates a fit of the data to a model, and thin solid curves (left ordinate) represent predicted coverage of individual layers. (b) AFM image, 5 × 5 μm2, of a pentacene thin film deposited on SiO2 at Ei = 2.5 eV. The thickness of this film is ∼9 MLs.
completion of the fourth ML is clearly observed but is clearly damped. This result is largely the same as we reported previously,26 except that we did not resolve the first small maximum in our earlier work. Beyond the deposition of ∼4 MLs, the intensity remains constant. This suggests that growth becomes 3D after completion of the first ∼4 MLs. The intensity oscillation has been fit using a modified version39 of the mean-field, rate equation model of growth first proposed by Cohen and co-workers.40 The fit to the intensity is indicated by the solid blue line, and we see that the fit to the experimental data is excellent. In Figure 3a we also show the coverage (occupancy) of each layer with solid black lines that are predicted by the fit to the intensity oscillations. After a total growth of 2 MLs, the second layer is ∼85% full, whereas after 4 MLs the fourth layer is ∼69% full. These results indicate that pentacene grows in a layer-by-layer (LbL) mode for approximately 4 MLs before significant roughening begins to occur. In Figure 3b, we display the AF micrograph (5 × 5 μm2) for pentacene deposited on clean SiO2 at Ei = 2.5 eV and a nominal thin film thickness of 8.8 ML. As may be seen, the AF micrograph reveals the formation of tall, pyramid-shaped islands with dendritic features, which have been observed in a number of studies, including our previous work.26 The spacing between the highest parts of the features is ∼1−2 μm in all cases. We have also examined the growth of pentacene on SiO2 at Ei = 4.7 and 7.0 eV, where for these two cases GR = 0.0088 and 0.0052 ML s−1, respectively. Results similar to those shown in 12544
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
Article
Figure 3 are presented in the Supporting Information. Summarizing, we observe only slight changes in the intensity oscillations, indicating that the incident kinetic energy has minimal impact on the morphological evolution of pentacene thin films grown on clean SiO2, consistent with our previous work.26 In addition, for all three values of Ei examined, the AF micrographs of ∼9 ML thick pentacene thin films on SiO2 are very similar (cf. Figure 3b and Supporting Information), and no obvious changes in morphology are observed with the changes in Ei. In Figure 4, we present the rms roughness predicted by our fits to the anti-Bragg oscillations as a function of the thin film
Figure 4. Thin film roughness as a function of pentacene thickness deposited on SiO2 at Ei = 2.5, 4.7, and 7.0 eV. Ts The points shown as symbols represent roughness values directly from 5 × 5 μm2 AFM images, such as that given in
Figure 5. (a) Scattered X-ray intensity as a function of the out-of-plane scattering vector, qz, for a thin film of pentacene (∼9 MLs) deposited on PS at Ei = 2.5 eV. Ts = 40 °C. (b) Scattered X-ray intensity as a function of the in-plane scattering vector, qxy, for this same pentacene deposited on PS.
thin film = 40 °C. obtained Figure 3.
similar to previous reports.26,41 In Figure 5b, we display GID data for this same pentacene thin film deposited on a thin film of PS. As may be seen, we observe several diffraction peaks, which have been indexed based upon previous reports,26,41 and these are indicative of the random distribution of the crystalline domains in the plane of the surface. These same peaks are observed on all surfaces investigated here (i.e., also on PMMA and PEI) and are in agreement with the previously reported “thin film” phase for pentacene.36 A lower limit for the coherent in-plane crystallite size, D∥, can be determined from the Scherrer formula, using the full width half-maximum of the GID peaks. Using the (110) peak, we find D∥ = 47 ± 1 nma value that is significantly smaller than the feature sizes observed for pentacene thin films on all surfaces investigated here (vide infra Figure 3 and Supporting Information). Thus, if this lower limit represents the actual grain size, then the terraces approaching widths of ∼1 μm observed in AFM [cf. Figure 3] must possess defects undetectable in the topographic image. We next consider the dynamics of growth of pentacene deposited on a thin film of PS. In Figure 6a we plot the scattered intensity measured in situ and in real time for the growth of pentacene on PS at Ei = 2.5 eV and a growth rate, GR = 0.011 ML s−1. As may be seen, for growth on this surface we observe a single, cusp-like peak, which coincides with completion of the first monolayer, but the anticipated second (small, similar to the zero-coverage intensity) and third (large, similar to the first peak) are more damped. These oscillations obviously differ in character from those observed on SiO2, where the largest peak corresponded to completion of the second ML. We have observed these effects in earlier work,27,28
thickness for the growth of pentacene on clean SiO2 for all Ei investigated. Here, to convert to a physical length scale we assume the thickness of a monolayer is 1.55 nm (vide infra). Also shown as individual points are the values for the rms surface roughness found from AFM (cf. Figure 3b and Supporting Information). We see from Figure 4 that there are minimal differences in the evolution of surface roughness with Ei. In addition, the difference between the final rms roughness values predicted by XRR and those measured by AFM are less than 10%, which adds to our confidence in the modeling of the intensity oscillations. Thus, the results from both in situ X-ray scattering and ex situ AFM indicate that incident kinetic energy does not produce significant changes in the evolution of surface morphology for both the range of Ei and the range of coverages considered here. 2. Growth on Polystyrene (PS). We now consider growth of pentacene on a thin film of PS. We first consider characterization of a pentacene thin film grown on PS at a single fixed value of Ei, where we will confirm that a crystalline thin film is formed on top of the amorphous polymer thin film. In Figure 5a, we display specular XRR of a pentacene thin film that was deposited on PS at Ei = 2.5 eV. The scattered intensity exhibits Bragg reflections (up to the second order) from the pentacene crystalline thin film and well-defined thickness fringes arising from the combination of the smooth underlying PS thin film and the pentacene thin film. From the (00l) Bragg peaks, the average d001 spacing for the pentacene thin films on all surfaces investigated was determined to be 15.54 ± 0.02 Å, a value 12545
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
Article
thin films on PS are very similar (cf. Figure 6b and Supporting Information), and no obvious changes in morphology are observed with the changes in Ei. From the naked eye, somewhat smaller features are formed on the thin film of PS vs SiO2. In Figure 7, we present the rms roughness predicted by our fits to the anti-Bragg oscillations as a function of the thin film
Figure 7. Thin film roughness as a function of pentacene thickness deposited on SiO2 at Ei = 2.5, 4.7, and 7.0 eV. Ts The points shown as symbols represent roughness values directly from 5 × 5 μm2 AFM images, such as that given in
Figure 6. (a) X-ray intensity at the anti-Bragg condition as a function of exposure to the molecular beam for a thin film of pentacene deposited on PS at Ei = 2.5 eV. Ts = 40 °C. Thick solid (blue) line (right ordinate) indicates a fit of the data to a model, and thin solid curves (left ordinate) represent predicted coverage of individual layers. (b) AFM image, 5 × 5 μm2, of a pentacene thin film deposited on PS at Ei = 2.5 eV. The thickness of this film is ∼9 MLs.
thin film = 40 °C. obtained Figure 6.
thickness for growth of pentacene on a thin film of PS for all Ei investigated. Also shown as individual points are the values for the rms surface roughness found from AFM, where the difference between these values and the final rms roughness values predicted by XRR is less than 9%. We see from Figure 7 that there are minimal differences in the evolution of surface roughness with Ei, suggesting that incident kinetic energy does not produce significant changes in the evolution of surface morphology for pentacene deposited on a thin film of PS. 3. Growth on Poly(methylmethacrylate) (PMMA). We next consider growth of pentacene on a thin film of PMMA. In Figure 8, we present results similar to those shown above for pentacene growth on SiO2 (Figure 3) and a thin film of PS (Figure 6). Here Ei = 2.5 eV, and the growth rate GR = 0.0087 ML s−1. As may be seen, these results on PMMA are comparable to those observed on SiO2 and PS. The intensity oscillation at the anti-Bragg condition for all values of Ei exhibits a small cusp-like maximum at completion of the first ML, followed by a larger maximum as the second ML completes. This behavior is similar to that observed on SiO2 (cf. Figure 3), and we see that the thin film grows in a LbL mode for ∼4 MLs before significant roughening commences. Results obtained at the higher incident kinetic energies (Ei = 4.7 and 7.0 eV), given in the Supporting Information, are similar to this result at Ei = 2.5 eV. We take note of the fact, however, that the fit to the data at Ei = 7.0 eV (see Supporting Information) is the poorest of the three sets of conditions and four surfaces (including PEI, vide infra) we consider here, which lends some uncertainty to the predictions of the model for this particular condition and surface. In Figure 8b, we display the AF micrograph (5 × 5 μm2) for pentacene deposited on a thin film of PMMA at Ei = 2.5 eV and a nominal thin film thickness of 9.0 ML. The AF micrograph reveals the formation of tall, pyramid-shaped islands with
and they are associated with changes in the reflection amplitude(s) and the phase change produced by the underlying substrate. Returning to the data presented in Figure 6a, we see that beyond the deposition of ∼3 MLs the intensity remains constant. This suggests that growth becomes 3D quickly after completion of the first ∼3 MLs. Again, we have fit these data to the model given by eqs 1 and 2, and the fit of the intensity is shown by the solid blue line. In Figure 6a, we also show the coverage (occupancy) of each layer with solid black lines that are predicted by the fit to the intensity oscillations. After a total growth of 2 MLs, the second layer is ∼76% full, whereas after 4 MLs the fourth layer is only ∼63% full. These results indicate that pentacene grows in a layer-by-layer (LbL) mode on a thin film of PS for approximately 3 MLs before significant roughening begins to occur. Indeed, we observe nearly identical growth behavior concerning the intensity oscillations at the anti-Bragg condition for growth at higher incident kinetic energies (Ei = 4.7 and 7.0 eV), results which are shown in the Supporting Information. As for growth on SiO2, the incident kinetic energy has minimal impact on the morphological evolution of pentacene thin films grown on a thin film of PS. In Figure 6b, we display the AF micrograph (5 × 5 μm2) for pentacene deposited on a thin film of PS at Ei = 2.5 eV and a nominal film thickness of 9.1 ML. As for growth on SiO2, the AF micrographs reveal the formation of tall, pyramid-shaped islands with dendritic features. In addition, for all three values of Ei examined, the AF micrographs of ∼9 ML thick pentacene 12546
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
Article
Figure 9. Thin film roughness as a function of pentacene thin film thickness deposited on PMMA at Ei = 2.5, 4.7, and 7.0 eV. Ts = 40 °C. The points shown as symbols represent roughness values obtained directly from 5 × 5 μm2 AFM images, such as that given in Figure 8.
Figure 8. (a) X-ray intensity at the anti-Bragg condition as a function of exposure to the molecular beam for a thin film of pentacene deposited on PMMA at Ei = 2.5 eV. Ts = 40 °C. Thick solid (blue) line (right ordinate) indicates a fit of the data to a model, and thin solid curves (left ordinate) represent predicted coverage of individual layers. (b) AFM image, 5 × 5 μm2, of a pentacene thin film deposited on PMMA at Ei = 2.5 eV. The thickness of this film is ∼9 MLs.
dendritic features, similar to SiO2 and PS, although the similarities are stronger in the case of SiO2. In previous work, growth on PMMA resulted in feature sizes larger than on PS.12 Finally, as may be seen from AF micrographs shown in the Supporting Information (for Ei = 4.7 and 7.0 eV), no obvious changes in morphology for ∼9 ML thick pentacene thin films on PMMA are observed with Ei. In Figure 9, we present the rms roughness predicted by our fits to the anti-Bragg oscillations as a function of the thin film thickness for growth of pentacene on a thin film of PMMA for all Ei investigated. Also shown as individual points are the values for the rms surface roughness found from AFM, where the difference between these values and the final rms roughness values predicted by XRR is less than 7%. We see from Figure 9 that there are essentially no differences in the evolution of surface roughness at Ei = 2.5 and 4.7 eV, but smoother growth at low and moderate thin film thicknesses at Ei = 7.0 eV is indicated. Due to the poor fit to the data concerning the latter condition (see Supporting Information), we do not place significant weight on this observation. Indeed, the similarity of the results from AFM (rms roughness and nature of features) argues against an effect of Ei on the growth of pentacene on this surface. 4. Growth on Polyethyleneimine (PEI). We next consider growth of pentacene on a thin film of PEI, and we begin first with the results from X-ray scattering. In Figure 10a, we plot the scattered intensity measured in situ and in real time for the growth of pentacene on a thin film of PEI at Ei = 2.5 eV, where the thin film growth rate was 0.0088 ML s−1. As may be seen, here the intensity oscillation is strongest for completion of the
Figure 10. (a) X-ray intensity at the anti-Bragg condition as a function of exposure to the molecular beam for a thin film of pentacene deposited on PEI at Ei = 2.5 eV. Ts = 40 °C. Thick solid (blue) line (right ordinate) indicates a fit of the data to a model, and thin solid curves (left ordinate) represent predicted coverage of individual layers. (b) AFM image, 5 × 5 μm2, of a pentacene thin film deposited on PEI at Ei = 2.5 eV. The thickness of this film is ∼9 MLs.
first, third, and fifth MLs, and it is more sustained compared to growth on SiO2, PS, and PMMA. Here, on PEI, LbL growth is extended up to ∼6 MLs. As for growth on the other surfaces, the incident kinetic energy has minimal impact on the morphological evolution of pentacene thin films grown on a thin film of PEI. These results for Ei = 4.7 and 7.0 eV are presented in the Supporting Information. 12547
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
Article
In Figure 10b, we display the AF micrograph (5 × 5 μm2) for pentacene deposited on a thin film of PEI at Ei = 2.5 eV and a nominal thin film thickness of 9.3 ML. As may be seen, the AF micrograph reveals the formation of tall, pyramid-shaped islands with dendritic features. In the Supporting Information, we find no obvious changes in morphology for Ei = 4.7 and 7.0 eV, although somewhat larger terraces are observed at Ei = 7.0 eV. Indeed, most interesting is the presence of these larger terraces on PEI (for all values of Ei), some approaching 2 μm in length (cf. Figure 10b and Supporting Information). On the basis of the AFM images, compared to the growth of pentacene on SiO2, and thin films of PS and PMMA, thin film growth is smoother on a thin film of PEI, corroborating the observation of more extended LbL growth on this surface indicated by the X-ray scattering data. In Figure 11, we present the rms roughness predicted by our fits to the anti-Bragg oscillations as a function of the thin film
Figure 12. Thin film roughness as a function of pentacene thin film thickness deposited on PS, PMMA, SiO2, and PEI at Ei = 2.5 eV. Ts = 40 °C. The points shown as symbols represent roughness values obtained directly from 5 × 5 μm2 AFM images as shown in Figures 3, 6, 8, and 10.
thickness for growth of pentacene on all surfaces investigated at Ei = 2.5 eV. We also show as individual points the values for the rms surface roughness found from AFM. As may be seen, there is good agreement between the final roughness predicted by the model and that measured directly by AFM (in all cases the deviation is 0.02), but only the first layer has a coverage >0.95. In contrast, on PEI at this coverage, the first two layers have coverages >0.95, and only four layers, including the first two, are represented by coverages >0.005. Obviously the interlayer transport between the second, third, fourth, etc. is being affected by the underlying substrate. In other work, a change in the interlayer spacing with thin film thickness, and hence the molecular tilt, was cited as the reason for a thickness-dependent ES barrier. In one study, the layer spacing was estimated from AFM and involved the growth 12550
dx.doi.org/10.1021/jp300635u | J. Phys. Chem. C 2012, 116, 12541−12552
The Journal of Physical Chemistry C
■
of para-sexiphenyl (6P).46 In the other, X-ray scattering was employed, and the growth of diindenoperylene (DIP) was examined.47 In one case, a change in the molecular tilt of 6P to more upright was associated with an increase in the ES barrier, while a decrease in the tilt of DIP was associated with an increase in the ES barrier. If these explanations were to apply here we should have observed a change in the layer spacing of the pentacene thin films with the identity of the underlying substrate. For the thin films directly examined here using X-ray reflectivity, cf. Figure 5a, and GID, cf. Figure 5b, we did not observe changes in the crystal structure of the pentacene thin films with the underlying substrate. We note that these thin films were ∼9 MLs in thickness, and thus we cannot exclude changes in the layer spacing, possibly transient, for thin films of ∼1−4 MLs in thickness. However, our model assumes no further changes in interlayer transport rates for the third and higher layers. The excellent agreement between the roughness predicted by the model for the ∼9 ML thick pentacene thin films and that measured by AFM would argue for the relevance of the structure (layer spacing) of these thick films to those formed at thinner
