TY - JOUR
AU - Kübel,, Christian
AB - Abstract Imaging the phase distribution of amorphous or partially crystalline organic materials at the nanoscale and analyzing the local atomic structure of individual phases has been a long-time challenge. We propose a new approach for imaging the phase distribution and for analyzing the local structure of organic materials based on scanning transmission electron diffraction (4D-STEM) pair distribution function analysis (PDF). We show that electron diffraction based PDF analysis can be used to characterize the short- and medium-range order in aperiodically packed organic molecules. Moreover, we show that 4D-STEM-PDF does not only provide local structural information with a resolution of a few nanometers, but can also be used to image the phase distribution of organic composites. The distinct and thickness independent contrast of the phase image is generated by utilizing the structural difference between the different types of molecules and taking advantage of the dose efficiency due to use of the full scattering signal. Therefore, this approach is particularly interesting for imaging unstained organic or polymer composites without distinct valence states for electron energy loss spectroscopy. We explore the possibilities of this new approach using [6,6]-phenyl-C61- butyric acid methyl ester (PC61BM) and poly(3-hexylthiophene-2,5-diyl) (P3HT) as the archetypical and best-investigated semiconductor blend used in organic solar cells, compare our phase distribution with virtual dark-field analysis and validate our approach by electron energy loss spectroscopy. pair/radial distribution function (PDF/RDF), 4D-STEM, polymer blends, phase mapping, local structure analysis Introduction Properties and functionality of organic composite materials are directly related to their molecular geometry, the molecular packing and the morphology of the phase distribution [1–3]. Knowledge of this information is therefore fundamental for polymer processing and device design. This requires characterization techniques with high spatial resolution. (Scanning) transmission electron microscopy ((S)TEM) has been an essential tool, intensively used to image soft materials for decades [4–8]. Most commonly, at the nanoscale, the morphology of organic composites, block copolymers and biological specimens has been studied using selective staining to create mass contrast for the different components [9–13]. However, artifacts can be introduced by staining [14,15], and finding suitable stains can be difficult for some material combinations. High-resolution TEM (HRTEM) has been used for lattice imaging of crystalline and partially crystalline organic materials [5,16–19] and can provide detailed structural information, e.g. on crystal defects [19–21]. The main challenge here is the electron beam sensitivity of the organic materials, where damage due to radiolysis [4,22,23,24] limits the achievable resolution and contrast severely. More recently, electron diffraction techniques have seen a significant development, combining electron beam precession with 3D nanodiffraction to solve the crystal structure of small organic crystals [25]. However, a large number of soft materials are aperiodic or even amorphous, limiting HRTEM and diffraction techniques. Electron energy loss spectroscopy (EELS) techniques, e.g. 4D-STEM-EELS and energy filtered TEM (EFTEM), especially focusing on the low-loss region, are increasingly used to image the phase distribution of soft materials where molecules/polymers possess distinct differences in their valence electron configurations [26–28]. Interpretation of the low-loss signal, relating it to the valence electron state and the plasmon resonance, is usually challenging. Moreover, the image contrast is commonly affected by thickness variations in the specimen [29]. Structural characterization of amorphous organic materials has been carried out by pair distribution function (PDF) analysis obtained from X-ray and neutron diffraction [30,31]. The PDF describes the population of atomic pairs as a function of interatomic distances r and therefore characterizes the atomic configuration of the material. Obtaining PDFs from electron diffraction (ePDF) has been introduced in references [32,33] for characterizing polycrystalline metals and glasses and has been further extended to nanoparticles in catalysis [34] and battery materials [35,36]. Application of ePDF to organic nanocrystals has recently been verified to provide identical information as an X-ray based PDF analysis [37]. The current spatial limit of ePDF is set by conventional diffraction, when using a broad beam with tens or hundreds of nanometer diameter, thus losing local structural information in heterogeneous materials. One solution for probing nano-sized volumes in diffraction is nanobeam electron diffraction implemented in 4D-STEM [38–41]. 4D-STEM records individual diffraction patterns by stepwise scanning a fine probe across a predefined area of interest forming an array of diffraction patterns. This approach has been utilized for structural inhomogeneity measurements of glassy materials [42] and orientation mapping [43,44] of crystalline materials. Recently, with the emerging high-speed cameras and high-efficiency pixelated detectors [45,46], new 4D-STEM developments are rapidly progressing, e.g. for crystal strain mapping [47], differential phase contrast imaging of the mean inner-field [48,49] and low voltage electron ptychography [50,51]. In this paper, we show that ePDF can be used to characterize amorphous organic molecules, polymers and blends thereof. We demonstrate that 4D-STEM-PDF [42], the combination of 4D-STEM and PDF analysis, cannot only be used to perform a local structural analysis of individual phases with nanoscale resolution, but can also be utilized to image the phase distribution without staining and without being significantly affected by thickness variations. 4D-STEM-PDF uses all diffracted electrons, making it sensitive to atomic bonding variation and dose-efficient for structural analysis compared to other techniques. Materials This work uses the fullerene derivative [6,6]-phenyl C61-butyric acid methyl ester (PC61BM) (Fig. 1a, top right) and the polymer poly(3-hexylthiophene-2,5-diyl) (P3HT) (Fig. 1a, top left) as model system to demonstrate 4D-STEM-PDF of organic blends. PC61BM and P3HT form bulk-heterojunctions, which are widely used as light-harvesting semiconductors in organic solar cells [52,53] and, therefore, allow comparison with previous studies in the literature. HRTEM and EFTEM characterizations have been performed previously on this system [28,54]. In our measurements, neat P3HT and PC61BM films were used as reference samples and a blend of both semiconductors was used for the 4D-STEM-PDF analysis. These organic thin films were prepared on glass substrates that were cleaned by ultrasonication in acetone and 2-propanol for 15 min. Afterwards, a sacrificial layer of poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) was prepared by filtering Clevios P VP AI 4083 (PTFE 0.45 μm), diluting it with ethanol 1:3 by volume and spin coating a thin layer at 4000 rpm, which was subsequently annealed at 120°C for 10 min. The neat PC61BM (Solenne, 99%) or P3HT (Rieke Metals, regioregular electronic grade, 4002-E) films were prepared by dissolving 20 g/L in chlorobenzene and spin coating at 1000 rpm on the PEDOT:PSS. For the blend, P3HT and PC61BM were dissolved 1:1 by weight in chlorobenzene at a total concentration of 40 g/L and spin coated at 1500 rpm on the PEDOT:PSS layer. The blend film was annealed afterwards at 140°C for 10 min, which is commonly used to optimize the morphology of the blend for optimum power conversion efficiency of the corresponding solar cells [55]. By dissolving the PEDOT:PSS layer in water, the organic layers of interest were floated off and picked up with bare Cu TEM grids (200 mesh). The blend film on the TEM grids is approximately 150 nm thick as estimated from the EELS zero-loss spectrum. Fig. 1. View largeDownload slide (a) Top: Atomic structure of a single P3HT repeat unit (left) and PC61BM (right). Bottom: SAED patterns of P3HT (left), PC61BM (middle) and their blend (right). (b) Diffraction profiles of P3HT (red line) and PC61BM (green line) by azimuthal integration of the SAED patterns. (c) Structure factors of the blend (top, blue line), P3HT (middle, red line) and PC61BM (bottom, green line) derived from experimentally measured SAED patterns. (d) From top to bottom, the experimental PDFs of the blend (blue solid line), P3HT (red solid line), simulated PDF according to crystalline P3HT model [59] (red dashed line), experimental PDF of PC61BM (green solid line) and simulated PDF according to PC61BM molecular model (green dashed line). For the references to the colors, the reader is referred to the online version of this article. Fig. 1. View largeDownload slide (a) Top: Atomic structure of a single P3HT repeat unit (left) and PC61BM (right). Bottom: SAED patterns of P3HT (left), PC61BM (middle) and their blend (right). (b) Diffraction profiles of P3HT (red line) and PC61BM (green line) by azimuthal integration of the SAED patterns. (c) Structure factors of the blend (top, blue line), P3HT (middle, red line) and PC61BM (bottom, green line) derived from experimentally measured SAED patterns. (d) From top to bottom, the experimental PDFs of the blend (blue solid line), P3HT (red solid line), simulated PDF according to crystalline P3HT model [59] (red dashed line), experimental PDF of PC61BM (green solid line) and simulated PDF according to PC61BM molecular model (green dashed line). For the references to the colors, the reader is referred to the online version of this article. Experimental setup Selected area electron diffraction (SAED) was performed using an aberration (image) corrected FEI Titan 80–300 microscope operated at 300 kV in TEM mode with gun lens 6 and spot size 9. An area of 10 μm diameter was selected by using a 50 μm condenser (C2) aperture, resulting in a beam current density of 5.2 e-nm−2s−1, measured by the calibrated Gatan ultrascan CCD. The recorded maximum scattering angle is 60 mrad, corresponding to 3 Å−1 in reciprocal space. EFTEM maps were acquired with a Gatan Tridium 863 image filter. A 3 eV energy slit was centered on 19 and 28 eV energy loss to image P3HT and PC61BM of the organic blend. 4D-STEM mapping was performed using a Tecnai F20 ST (Philips) operated at 200 kV in microprobe STEM mode, equipped with a NanoMegas Topspin system [43]. For the data acquisition, spot size 8, gun lens 6, extraction voltage of 4.5 kV and a 30 μm C2 aperture were adopted. These settings result in a quasi-parallel beam with a convergence semi-angle of 2.1 mrad, a minimum probe size of approximately 1.0 nm and a beam current of 27 pA measured using the fluorescent screen. The beam scanning was controlled by the NanoMegas Topspin system. Diffraction patterns were recorded by an external Stingray camera with a frame size of 580 × 580 pixels. We set the scanning frequency to 100 fps (0.01 s per pattern) and applied a 3.2 μm defocus resulting in a 5 nm diameter probe. The dose for each diffraction pattern is 8.6 × 104 e-/nm2. A camera length of 100 mm was used, resulting in a maximum recorded angle of 50 mrad (2 Å−1). Virtual annular dark field (ADF) images were created from the 4D-STEM data by choosing a virtual detector defining an area (i.e. collection angle β) for each of the diffraction patterns. The intensity in each pixel in the virtual image is the sum of the signals within the defined area [39]. We adopted β in the range of 28.0−48.5 mrad for generating virtual image with HAADF contrast, and of 1.13−1.14 and 10.9−11.8 mrad for imaging the P3HT and PC61BM phases. ePDF analysis of neat films Distinct features can already be seen in the SAED patterns (Fig 1a, bottom) and the corresponding diffraction profiles (Fig. 1b), even though they are dominantly amorphous haloes, except a relatively sharp ring at 0.26 nm−1 in the diffractions of both neat P3HT and the blend film. It corresponds to a periodic π-π stacking of the thiophene unit [56]. More detailed structural information is obtained from a PDF analysis of the diffraction patterns. As proposed in references [33,57,58], to derive PDF from electron diffraction, first the diffraction profile I(s) has to be normalized by subtracting the atomic scattering factor based on equation (1) φ(s)=I(s)−Nf(s)2Nf(s)2s (1) where s=2sin(θ)λ ⁠, θ is the scattering semi-angle, λ the wavelength of the incident electrons, f(s) the averaged atomic scattering factor of all elements contributing to the diffraction pattern, and N the number of atoms irradiated by the electron beam. In practice, N is determined by fitting the scattering factor (⁠ Nf(s)2 ⁠) to the diffraction profile (I(s)) at large scattering angle s. φ(s) is often referred to as structure factor, describing the structural information in reciprocal space. Figure 1c shows the structure factors of the blend (green), P3HT (blue) and PC61BM (red) derived from experimentally measured SAEDs acquired with 2 s exposure time, resulting in a dose of 10.4 e−/nm2. Most of the peaks of the P3HT structure factor (at s > 0.4 nm−1) are shifted collectively towards lower angles compared to those of PC61BM. This indicates an on average larger bonds length in P3HT compared to PC61BM. The PDF, G(r) ⁠, is obtained by Fourier sine transform of φ(s) ⁠, thus enabling a real space interpretation of the structure information encoded in φ(s) ⁠. It is particularly sensitive to the bond length and coordination, and also carries medium-range information on the intermolecular configuration as demonstrated by T.E. Gorelik et al. [37] Figure 1d shows the PDFs of P3HT, PC61BM and the blend thin film as well as a simulated PDFs based on the structure of crystalline P3HT [59] and a PC61BM molecule [60] (Fig. 1a, top right). The intensity of the PDFs are scaled to the same height of the first peak, as multiple scattering of the electron beam for different film thicknesses influences the absolute height of the PDF. However, the peak position and the relative peak height, which carry most of the information, are robust even for thick specimens [32,42,57]. The excellent agreement between the experimental amorphous and the simulated PDF of PC61BM validates the ePDF approach to characterize amorphous organic molecules. The pronounced first and second peak of the experimental PC61BM PDF corresponds to an average C-C bond length of 1.40 Å and second order C-C correlation of 2.42 Å. This results in an average C-C-C bonding angle of 119.6°. The peak at 2.95 Å corresponds to the distance of opposite carbon atoms in each benzene ring in the fullerene unit. Also a good agreement between the experimental P3HT PDF and a simulated PDF based on the crystalline P3HT model by Kayunkid et al. [59] can be seen. The slight deviations can be attributed to the limited crystallinity of the experimental P3HT film. The PDF of the experimental P3HT exhibits the first two peaks at 1.51 Å and 2.59 Å. They are shifted to significantly larger distances compared to those in PC61BM. The peaks at 3.26 and 3.86 Å highlighted by stars in the P3HT PDF (Fig. 2b, red solid line) are another feature remarkably different to the PC61BM PDF. Based on the crystalline P3HT model, these two peaks can be attributed to the correlation of adjacent thiophene units in the polymer. Fig. 2. View largeDownload slide (a) A typical (HR)TEM bright field image of the blend film. (b) Virtual STEM HAADF image. (c) Procedure for 4D-STEM-PDF analysis. (d) PDF maps of PC61BM and (e) the P3HT phase at the same location as in b. (f) Color mix of d (green) and e (red). (g) A typical EFTEM map of the blend film. (h) PDFs of the P3HT-rich (red solid line) and PC61BM-rich (green solid line) phases of the blend extracted from the 4D-STEM-PDF cube. PDFs of the neat P3HT (red dashed line) and PC61BM (green dashed line) films obtained by using the SAED microscope setup. Black solid line is the PDF of the neat P3HT film obtained using the 4D-STEM-PDF microscope setup. All scale bars are 100 nm. For the references to the colors, the reader is referred to the online version of this article. Fig. 2. View largeDownload slide (a) A typical (HR)TEM bright field image of the blend film. (b) Virtual STEM HAADF image. (c) Procedure for 4D-STEM-PDF analysis. (d) PDF maps of PC61BM and (e) the P3HT phase at the same location as in b. (f) Color mix of d (green) and e (red). (g) A typical EFTEM map of the blend film. (h) PDFs of the P3HT-rich (red solid line) and PC61BM-rich (green solid line) phases of the blend extracted from the 4D-STEM-PDF cube. PDFs of the neat P3HT (red dashed line) and PC61BM (green dashed line) films obtained by using the SAED microscope setup. Black solid line is the PDF of the neat P3HT film obtained using the 4D-STEM-PDF microscope setup. All scale bars are 100 nm. For the references to the colors, the reader is referred to the online version of this article. Overall, PDF enables a concrete interpretation of the diffraction data in terms of interatomic distances and thus delivers reliable information for a structure characterization of organic materials even though their structures are amorphous. The information is particularly useful in association with structural modeling such as molecular dynamic or reversed Monte Carlo simulation for revealing the intermolecular arrangement. 4D-STEM-PDF mapping of the blend film The procedure for 4D-STEM-PDF analysis is illustrated in Fig. 2c [42]. Each diffraction pattern is processed individually to determine a local PDF. The 4D-STEM dataset is then converted to a 3-dimensional (3D) PDF cube, which can be quantitatively analyzed by hyperspectral techniques, such as multiple linear least square (MLLS) fitting to identify and map the phases. By this approach, shown in Fig. 2d and e, the PC61BM and P3HT distribution in the blend film is mapped by the coefficients of MLLS fitting the single-phase PDFs to the PDF cube through the MLLS routine. The phase images are superpositioned as color map in Fig. 2f. The map clearly shows the phase separation and corresponding morphology. These are crucial information for understanding the material’s properties. A typical EFTEM map with a dose in the range of 4 × 104 e−/nm2 taken from the same sample is shown in Fig. 2g for comparison. Its similarity to the 4D-STEM-PDF map indicates that the 4D-STEM-PDF analysis enables meaningful phase mapping of the organic blend. In contrast, it is difficult to distinguish the fullerene derivative and the polymer by using the contrast seen in conventional (HR)TEM imaging, such as the example shown in Fig. 2a. STEM HAADF is only sensitive to the mass thickness (thickness × density) fluctuations in the specimen. No clear features of the phase distribution can be seen in the virtual STEM HAADF image (Fig. 2b) obtained from the same 4D-STEM data. In addition to the phase distribution, 4D-STEM-PDF carries local structural information, which can be used to characterize the structure of individual phases. Figure 2h shows the PDFs of the individual PC61BM-rich and P3HT-rich phases in the blend film. They are obtained by averaging the PDFs of each phase in the PDF cube to reduce noise, followed by normalization of the intensity by the height of the first peak. As shown by the comparison between the PDFs of the neat P3HT film obtained by the 4D-SAED and the 4D-STEM-PDF microscope setup in Fig. 2h, the larger convergence angle of the electron probe in the 4D-STEM-PDF setup dampens the PDF for large correlation distances and, furthermore, the limited recording angle (smax4D-STEM = 2 Å−1, whereas smaxSAED = 3 Å−1) broadens the peaks. However, the key features such as the bond length and higher order correlations (highlighted by stars) are still preserved, which results in significant information to identify the structures of the fullerene derivative and the P3HT polymer. The slight peak shift between the local PDF of the PC61BM rich phase in the blend and the PDF of the neat PC61BM film could result from projection of overlapping domains or local mixing of PC61BM and P3HT in the PC61BM rich phase in the blend. Further analysis pathways As the SAED patterns of the fullerene derivative and the polymer (Fig. 1a and b) are quite distinct, it is possible to create a phase map by choosing suitable virtual ADF settings. However, the large convergence angle used in our current 4D-STEM setup strongly blurs the diffraction patterns, thus weakening the distinction between the two compounds. This can be seen by the 4D-STEM measured structure factors in Fig. 3a. Nevertheless, there are still sufficient differences among them for imaging the corresponding distribution. We positioned the virtual ADF detector in the angular range of 10.9–11.8 mrad (8.7–9.4 nm−1) where the most pronounced peak of the structure factors of PC61BM are. It creates an image where the bright features represent the distribution of PC61BM molecules (Fig. 3b). By positioning the virtual detector in the range of 1.13–1.14 mrad (0.9–0.91 nm−1), another image with reversed contrast is created, where the bright features represent the distribution of P3HT (Fig. 3c). Fig. 3. View largeDownload slide (a) Structure factors of the PCBM-rich and P3HT-rich phase extracted from the 4D STEM data. Virtual ADF maps at the same location of the STEM PDF maps in Fig. 2d, e and f with the detecting angles in the ranges of (b) 10.9–11.8 mrad, and (c) 1.13 to 1.14 mrad. (d) Color mix of b (green) and c (red). (e) Map of the atom number N as a result of STEM PDF process according to equation (1). For the references to the colors, the reader is referred to the online version of this article. Fig. 3. View largeDownload slide (a) Structure factors of the PCBM-rich and P3HT-rich phase extracted from the 4D STEM data. Virtual ADF maps at the same location of the STEM PDF maps in Fig. 2d, e and f with the detecting angles in the ranges of (b) 10.9–11.8 mrad, and (c) 1.13 to 1.14 mrad. (d) Color mix of b (green) and c (red). (e) Map of the atom number N as a result of STEM PDF process according to equation (1). For the references to the colors, the reader is referred to the online version of this article. Although the virtual ADF approach offers an alternative way to process the 4D-STEM data to obtain the phases map, there are noticeable limits. First, mass/thickness variations of the specimen and consequently multiple scattering of the electron beam can strongly influence the contrast in the ADF image. One example can be seen in Fig. 3d showing the PC61BM (green) and P3HT (red) distribution. The strong contrast gradient exhibits the same low frequency features as the STEM HAADF image (Fig. 2b) as a result of a thickness gradient in the sample, maybe also due to some contamination buildup on the sample. Multiple scattering from thicker specimen areas pushes more of the low angle scattered electrons to higher angles [32,57] resulting in a reduced P3HT signal. Furthermore, an amorphous carbon film (and presumably any carbon contamination) has a structure factor with a pronounced peak at 8.90 nm−1 [61] close to that of the PC61BM. They overlay on the ADF angular detection range, seemingly resulting in an enrichment of the PC61BM phase. In contrast, PDF analysis uses the background corrected structure factor, where the fitting parameter N is determined for each diffraction pattern of the 4D-STEM data. Figure 3e shows that the map of N possesses identical features as the HAADF image (Fig. 2b). It indicates that the information of mass fluctuation is extracted from the raw 4D-STEM data by the array of N and thus is eliminated in the final PDF map. This results in the 4D-STEM-PDF maps being (almost) thickness independent, therefore reflecting the true phase distribution without mass-thickness confusion as in STEM ADF images. In addition, the ADF images lack the structural information from the diffraction pattern, so that they do not enable a local (atomic) structure characterization. Electron beam induced damage Dose-dependent measurements were individually performed with parallel illumination using a broad beam SAED setup (Fig. 4a and b) and nanobeam illumination in μP-STEM mode (Fig. 4c). Figure 4a and b show the azimuthally integrated SAED profiles and corresponding PDFs. The diffraction peak at 0.26 nm−1 corresponding to the polymer backbone packing reduces quickly with increasing dose, indicating that the intermolecular stacking is quickly disturbed by the electron beam. This is primarily attributed to damage of the alkyl side chains by the electron beam, resulting in disorder between the polythiophene back bones. However, the characteristic peaks in the PDFs shown in Fig. 1b do not change significantly at a dose of 105 e−/nm−2. This is presumably because of the fairly stable thiophene units dominating the short-range order seen in the PDFs, whereas the more beam sensitive side chains are much more flexible, resulting in a less defined signal in the PDF. At even higher dose, the PDF of P3HT slowly changes, becoming similar to the PDF of PC61BM at a dose of 5 × 107 e−/nm−2 (Fig. 4c, purple). This change can presumably be attributed to the electron beam breaking carbon-hydrogen bonds in the alkyl chains, cross-linking and graphitizing them, thus making the measured atomic distances more similar to that in fullerene. However, the residual difference between PC61BM and partially graphitized P3HT is still larger than the sensitivity of PDF analysis, so that beam damage will not influence the imaged distribution qualitatively, though uncertainty might be caused in evaluating the local phase concentration. However, the actual dose used for the 4D-STEM measurements is below the level incurring detectable beam-induced graphitization, thus enabling a good interpretation. This can also be seen from the features, highlighted by the stars in Fig. 2h, which are preserved in the local PDFs. Fig. 4. View largeDownload slide Test of the beam induced structural changes in the neat P3HT film measured at 300 keV acceleration voltage. (a) Azimuthally integrated diffraction profiles and (b) corresponding PDFs, acquired using a broad beam SAED setup. (c) PDFs of the P3HT film measured with nanobeam illumination with the same beam current but varying probe size and thus dose. The legend notes the probe radius r and dose. The top green and bottom red are PDFs of P3HT and PC61BM acquired with SAED setup and a dose of 10.4 e-/nm−2 (taken from Fig. 1d) for comparison. For the references to the colors, the reader is referred to the online version of this article. Fig. 4. View largeDownload slide Test of the beam induced structural changes in the neat P3HT film measured at 300 keV acceleration voltage. (a) Azimuthally integrated diffraction profiles and (b) corresponding PDFs, acquired using a broad beam SAED setup. (c) PDFs of the P3HT film measured with nanobeam illumination with the same beam current but varying probe size and thus dose. The legend notes the probe radius r and dose. The top green and bottom red are PDFs of P3HT and PC61BM acquired with SAED setup and a dose of 10.4 e-/nm−2 (taken from Fig. 1d) for comparison. For the references to the colors, the reader is referred to the online version of this article. 4D-STEM data with lower dose but good signal to noise ratio would be expected when using high quantum detection efficient (DQE) pixelated detectors. This may make high quality local PDFs available for direct nanovolume analysis, such as to inspect the structural variation across the phase boundary or in different morphological environments. Furthermore, quantifying the local phase concentration with reliable coefficients of the MLLS or principle component analysis will also become feasible. We also expect that cooling the sample to liquid nitrogen temperature would be an alternative to significantly stabilize the organic molecules as has been widely reported [62,63]. Conclusion We illustrated the possibilities 4D-STEM-PDF offers as a new method to map the phase distribution of organic semiconductor blends analyzing elastically scattered electrons. Using PC61BM and P3HT as model system, we showed that ePDF can provide a structural characterization of amorphous organic molecules and polymers. We further demonstrated that 4D-STEM-PDF provides quantitative information on the local structure to image the phase distribution in a blend without applying any staining to the soft materials. As an alternative to 4D-STEM-PDF mapping, virtual dark field imaging was evaluated. The dose used for phase mapping and structural analysis was carefully evaluated and the beam induced radiation damage did not significantly affect the map quality, but limits quantitative evaluation of local concentrations of molecules and a detailed local structural analysis, which might be overcome by using high DQE pixelated detectors and cryo-TEM. In contrast to other (S)TEM approaches, the distinct contrast in the 4D-STEM-PDF images only corresponds to structural information, making it robust in dealing with voids, thickness and density variations, which are common challenges in interpreting data of thin-films of organic blends. The approach could be used to discover the structure of the inter-phase boundary of organic bulk-heterojunction materials and the structural variation of the local environment in soft bulk materials. Acknowledgment X.M. thanks the Deutsche Forschungsgemeinschaft (DFG) for funding of grant MU 4276/1-1 for the support of methodology development of 4D-STEM-PDF. C.S. and A.C. acknowledge funding by the German Federal Ministry of Education and Research (BMBF) under contract 03EK3571 (project TAURUS2). References 1 Richter L J , Delongchamp D M , and Amassian A ( 2017 ) Morphology development in solution-processed functional organic blend films: an in situ viewpoint . Chem. Rev. 117 : 6332 – 6366 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Rivnay J , Mannsfeld S C B , Miller C E , Salleo A , and Toney M F ( 2012 ) Quantitative determination of organic semiconductor microstructure from the molecular to device scale . Chem. Rev. 112 : 5488 – 5519 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Huang Y , Kramer E J , Heeger A J , and Bazan G C ( 2014 ) Bulk heterojunction solar cells: morphology and performance relationships . Chem. Rev. 114 : 7006 – 7043 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Libera M R , and Egerton R F ( 2010 ) Advances in the transmission electron microscopy of polymers . Polym. Rev. 50 : 321 – 339 . Google Scholar Crossref Search ADS WorldCat 5 Martin D C , Chen J , Yang J , Drummy L F , and Kübel C ( 2005 ) High resolution electron microscopy of ordered polymers and organic molecular crystals: Recent developments and future possibilities . J. Polym. Sci. B Polym. Phys 43 : 1749 – 1778 . Google Scholar Crossref Search ADS WorldCat 6 Franken L E , Boekema E J , and Stuart M C A ( 2017 ) Transmission electron microscopy as a tool for the characterization of soft materials: application and interpretation . Adv. Sci. 4 : 1600476 . Google Scholar Crossref Search ADS WorldCat 7 Klein M F G , Pfaff M , Müller E , Czolk J , Reinhard M , Valouch S , Lemmer U , Colsmann A , and Gerthsen D ( 2012 ) Poly(3-hexylselenophene) solar cells: Correlating the optoelectronic device performance and nanomorphology imaged by low-energy scanning transmission electron microscopy . J. Polym. Sci. B Polym. Phys 50 : 198 – 206 . Google Scholar Crossref Search ADS WorldCat 8 Pfaff M , Müller E , Müller P , Gerthsen D , Klein M G , Colsmann A , and Lemmer U ( 2012 ) Nanomorphology of P3HT:PCBM-based organic solar cells analyzed by low-energy scanning transmission electron microscopy . Microsc. Microanal. 18 : 1234 – 1235 . Google Scholar Crossref Search ADS WorldCat 9 Trent J S , Scheinbeim J I , and Couchman P R ( 1983 ) Ruthenium tetraoxide staining of polymers for electron microscopy . Macromolecules 16 : 589 – 598 . Google Scholar Crossref Search ADS WorldCat 10 Haubruge H G , Jonas A M , and Legras R ( 2003 ) Staining of poly(ethylene terephthalate) by ruthenium tetroxide . Polymer (Guildf) 44 : 3229 – 3234 . Google Scholar Crossref Search ADS WorldCat 11 Smith R W , and Bryg V ( 2006 ) Staining Polymers for Microscopical Examination . Rubber Chem. Technol. 79 : 520 – 540 . Google Scholar Crossref Search ADS WorldCat 12 Jinnai H , Spontak R J , and Nishi T ( 2010 ) Transmission electron microtomography and polymer nanostructures . Macromolecules 43 : 1675 – 1688 . Google Scholar Crossref Search ADS WorldCat 13 Breiner U , Krappe U , Thomas E L , and Stadler R ( 1998 ) Structural characterization of the ‘Knitting Pattern’ in methacrylate) triblock copolymers . Macromolecules 9297 : 135 – 141 . Google Scholar Crossref Search ADS WorldCat 14 Chou T M , Prayoonthong P , Aitouchen A , and Libera M ( 2002 ) Nanoscale artifacts in RuO4-stained poly(styrene) . Polymer (Guildf) 43 : 2085 – 2088 . Google Scholar Crossref Search ADS WorldCat 15 Wang Y , Coombs N , Turak A , Lu Z-H , Manners I , and Winnik M A ( 2007 ) Interfacial staining of a phase-separated block copolymer with ruthenium tetroxide . Macromolecules 40 : 1594 – 1597 . Google Scholar Crossref Search ADS WorldCat 16 Martin C , and Thomas L ( 1995 ) Experimental high-resolution microscopy of polymers . Polymer (Guildf) 36 : 1743 – 1759 . Google Scholar Crossref Search ADS WorldCat 17 Kübel C , Eckhardt K , Enkelmann V , Wegner G , and Müllen K ( 2000 ) Synthesis and crystal packing of large polycyclic aromatic hydrocarbons: hexabenzo[bc,ef,hi,kl,no,qr]coronene and dibenzo[fg,ij]phenanthro[9,10,1,2,3-pqrst]pentaphene . J. Mater. Chem. 10 : 879 – 886 . Google Scholar Crossref Search ADS WorldCat 18 Kübel C , Lawrence D P , and Martin D C ( 2001 ) Super-helically twisted strands of poly(m-phenylene isophthalamide) (MPDI) . Macromolecules 34 : 9053 – 9058 . Google Scholar Crossref Search ADS WorldCat 19 Drummy B L F , Kübel C , Lee D , White A , and Martin D C ( 2002 ) Direct imaging of defect structures in pentacene . Adv. Mater. 14 : 54 – 57 . Google Scholar Crossref Search ADS WorldCat 20 Drummy L F , Kübel C , and Martin D C ( 2004 ) Molecular vacancies in herringbone crystals . Philos. Mag. 84 : 1955 – 1968 . Google Scholar Crossref Search ADS WorldCat 21 Kübel C , and Martin D C ( 2001 ) Influence of structural variations on high-resolution electron microscopy images of poly[1,6-di(N -carbazolyl)2,4-hexadiyne] nanocrystals . Philos. Mag. A 81 : 1651 – 1673 . Google Scholar Crossref Search ADS WorldCat 22 Drummy L F , Yang J , and Martin D C ( 2004 ) Low-voltage electron microscopy of polymer and organic molecular thin films . Ultramicroscopy 99 : 247 – 256 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Egerton R F , Lazar S , and Libera M ( 2012 ) Delocalized radiation damage in polymers . Micron 43 : 2 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Egerton R F , Li P , and Malac M ( 2004 ) Radiation damage in the TEM and SEM . Micron 35 : 399 – 409 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Kolb U , Gorelik T E , Mugnaioli E , and Stewart A ( 2010 ) Structural characterization of organics using manual and automated electron diffraction . Polym. Rev. 50 : 385 – 409 . Google Scholar Crossref Search ADS WorldCat 26 Kim G , Sousa A , Meyers D , and Libera M ( 2008 ) Nanoscale composition of biphasic polymer nanocolloids in aqueous suspension . Microsc. Microanal. 14 : 459 – 468 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Du Chesne A ( 1999 ) Energy filtering transmission electron microscopy of polymers - Benefit and limitations of the method . Macromol. Chem. Phys. 200 : 1813 – 1830 . Google Scholar Crossref Search ADS WorldCat 28 Drummy L F , Davis R J , Moore D L , Durstock M , Vaia R A , and Hsu J W P ( 2011 ) Molecular-scale and nanoscale morphology of P3HT:PCBM bulk heterojunctions: Energy-filtered TEM and low-dose HREM . Chem. Mater. 23 : 907 – 912 . Google Scholar Crossref Search ADS WorldCat 29 Pfaff M , Müller E , Klein M F , Colsmann A , Lemmer U , Krzyzanek V , Reichelt R , and Gerthsen D ( 2011 ) Low-energy electron scattering in carbon-based materials analyzed by scanning transmission electron microscopy and its application to sample thickness determination . J. Microsc. 243 : 31 – 39 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Proffen T , and Billinge S ( 2003 ) Structural analysis of complex materials using the atomic pair distribution function—a practical guide . Zeitschrift für Krist 218 : 132 – 143 . WorldCat 31 Takeshi E , and Billinge S J L ( 2012 ) The method of total scattering and atomic pair distribution function analysis. In: Takeshi E , and Billinge S J L (eds) Pergamon Materials Series 16 : pp 55 – 111 (Elsevier, Oxford, UK). Google Preview WorldCat COPAC 32 Anstis G , Liu Z , and Lake M ( 1988 ) Investigation of amorphous materials by electron diffraction—the effects of multiple scattering . Ultramicroscopy 26 : 65 – 69 . Google Scholar Crossref Search ADS WorldCat 33 Cockayne D J H , and McKenzie D R ( 1988 ) Electron diffraction analysis of polycrystalline and amorphous thin films . Acta Crystallogr. A 44 : 870 – 878 . Google Scholar Crossref Search ADS WorldCat 34 Willinger E , Massué C , Schlögl R , and Willinger M-G ( 2017 ) Identifying key structural features of IrOx water splitting catalysts . J. Am. Chem. Soc. . 10.1021/jacs.7b07079 jacs.7b07079. WorldCat 35 Zhu C , Mu X , Popovic J , Weichert K , van Aken P A , Yu Y , and Maier J ( 2014 ) Lithium potential variations for metastable materials: case study of nanocrystalline and amorphous LiFePO4 . Nano Lett. 14 : 5342 – 5349 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Li C , Mu X , Van Aken P A , and Maier J ( 2013 ) A high-capacity cathode for lithium batteries consisting of porous microspheres of highly amorphized iron fluoride densifi ed from its open parent phase . Adv. Energy Mater 3 : 113 – 119 . Google Scholar Crossref Search ADS WorldCat 37 Gorelik T E , Schmidt M U , Kolb U , and Billinge S J L ( 2014 ) Total-scattering pair-distribution function of organic material from powder electron diffraction data . Microsc. Microanal. 21 : 459 – 471 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Alloyeau D , Ricolleau C , Oikawa T , Langlois C , Le Bouar Y , and Loiseau A ( 2008 ) STEM nanodiffraction technique for structural analysis of CoPt nanoparticles . Ultramicroscopy 108 : 656 – 662 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Gammer C , Burak Ozdol V , Liebscher C H , and Minor A M ( 2015 ) Diffraction contrast imaging using virtual apertures . Ultramicroscopy 155 : 1 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Ophus C , Ercius P , Huijben M , and Ciston J ( 2017 ) Non-spectroscopic composition measurements of SrTiO3-La0.7Sr0.3MnO3 multilayers using scanning convergent beam electron diffraction . Appl. Phys. Lett. 110 : 63102 . Google Scholar Crossref Search ADS WorldCat 41 Barnard J S , Johnstone D N , and Midgley P A ( 2017 ) High-resolution scanning precession electron diffraction: alignment and spatial resolution . Ultramicroscopy 174 : 79 – 88 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Mu X , Wang D , Feng T , and Kübel C ( 2016 ) Radial distribution function imaging by STEM diffraction: phase mapping and analysis of heterogeneous nanostructured glasses . Ultramicroscopy 168 : 1 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Rauch E F , Portillo J , Nicolopoulos S , Bultreys D , Rouvimov S , and Moeck P ( 2010 ) Automated nanocrystal orientation and phase mapping in the transmission electron microscope on the basis of precession electron diffraction . Zeitschrift für Krist 225 : 103 – 109 . WorldCat 44 Kobler A , Kashiwar A , Hahn H , and Kübel C ( 2013 ) Combination of in situ straining and ACOM TEM: a novel method for analysis of plastic deformation of nanocrystalline metals . Ultramicroscopy 128 : 68 – 81 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Tate M W , Purohit P , Chamberlain D , Nguyen K X , Hovden R , Chang C S , Deb P , Turgut E , Heron J T , Schlom D G , Ralph D C , Fuchs G D , Shanks K S , Philipp H T , Muller D A , and Gruner S M ( 2016 ) High dynamic range pixel array detector for scanning transmission electron microscopy . Microsc. Microanal. 22 : 237 – 249 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Mir J A , Clough R , MacInnes R , Gough C , Plackett R , Shipsey I , Sawada H , MacLaren I , Ballabriga R , Maneuski D , O'Shea V , McGrouther D , and Kirkland A I ( 2017 ) Characterisation of the Medipix3 detector for 60 and 80 keV electrons . Ultramicroscopy 182 : 44 – 53 . Google Scholar Crossref Search ADS PubMed WorldCat 47 Ozdol V B , Gammer C , Jin X G , Ercius P , Ophus C , Ciston J , and Minor A M ( 2015 ) Strain mapping at nanometer resolution using advanced nano-beam electron diffraction . Appl. Phys. Lett. 106 . WorldCat 48 Krajnak M , McGrouther D , Maneuski D , O’Shea V , and McVitie S ( 2016 ) Pixelated detectors and improved efficiency for magnetic imaging in STEM differential phase contrast . Ultramicroscopy 165 : 42 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Lazić I , Bosch E G T , and Lazar S ( 2016 ) Phase contrast STEM for thin samples: integrated differential phase contrast . Ultramicroscopy 160 : 265 – 280 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Jiang Y , Chen Z , Han Y , Deb P , Gao H , Xie S , Purohit P , Tate M W , Park J , Gruner S M , Elser V , and Muller D A ( 2018 ) Electron ptychography of 2D materials to deep sub-ångström resolution . Nature 559 : 343 – 349 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Wang P , Zhang F , Gao S , Zhang M , and Kirkland A I ( 2017 ) Electron ptychographic diffractive imaging of boron atoms in LaB 6 crystals . Sci. Rep 7 : 1 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Kim K , Liu J , Namboothiry M A G , and Carroll D L ( 2007 ) Roles of donor and acceptor nanodomains in 6% efficient thermally annealed polymer photovoltaics . Appl. Phys. Lett. 90 : 2005 – 2008 . WorldCat 53 Perzon E , Wang X , Zhang F , Mammo W , Delgado J L , de la Cruz P , Inganäs O , Langa F , and Andersson M R ( 2005 ) Design, synthesis and properties of low band gap polyfluorenes for photovoltaic devices . Synth. Met. 154 : 53 – 56 . Google Scholar Crossref Search ADS WorldCat 54 Brinkmann M , and Rannou P ( 2009 ) Molecular weight dependence of chain packing and semicrystalline structure in oriented films of regioregular Poly(3-hexylthiophene) revealed by high-resolution transmission electron microscopy . Macromolecules 42 : 1125 – 1130 . Google Scholar Crossref Search ADS WorldCat 55 Kim Y , Choulis S A , Nelson J , Bradley D D C , Cook S , and Durrant J R ( 2005 ) Device annealing effect in organic solar cells with blends of regioregular poly(3-hexylthiophene) and soluble fullerene . Appl. Phys. Lett. 86 : 1 – 3 . WorldCat 56 Alexiadis O , and Mavrantzas V G ( 2013 ) All-atom molecular dynamics simulation of temperature effects on the structural, thermodynamic, and packing properties of the pure amorphous and pure crystalline phases of regioregular P3HT . Macromolecules 46 : 2450 – 2467 . Google Scholar Crossref Search ADS WorldCat 57 Mu X , Neelamraju S , Sigle W , Koch C T , Totò N , Schön J C , Bach A , Fischer D , Jansen M , and van Aken P A ( 2013 ) Evolution of order in amorphous-to-crystalline phase transformation of MgF2 . J. Appl. Crystallogr. 46 : 1105 – 1116 . Google Scholar Crossref Search ADS WorldCat 58 Cockayne D J H ( 2007 ) The study of nanovolumes of amorphous materials using electron scattering . Annu. Rev. Mater. Res. 37 : 159 – 187 . Google Scholar Crossref Search ADS WorldCat 59 Kayunkid N , Uttiya S , and Brinkmann M ( 2010 ) Structural model of regioregular poly(3-hexylthiophene) obtained by electron diffraction analysis . Macromolecules 43 : 4961 – 4967 . Google Scholar Crossref Search ADS WorldCat 60 Paternò G , Warren A J , Spencer J , Evans G , Sakai V G , Blumberger J , and Cacialli F ( 2013 ) Micro-focused X-ray diffraction characterization of high-quality [6,6]-phenyl-C61-butyric acid methyl ester single crystals without solvent impurities . J. Mater. Chem. C 1 : 5619 – 5623 . Google Scholar Crossref Search ADS WorldCat 61 Mu X TEM study of the structural evolution of ionic solids from amorphous to polycrystalline phases in the case of alkaline earth difluoride systems: Experimental exploration. Dissertation 31–34 ( 2013 ). 62 Knapek E , and Dubochet J ( 1980 ) Beam damage to organic material is considerably reduced in cryo-electron microscopy . J. Mol. Biol. 141 : 147 – 161 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Egerton R F ( 2013 ) Control of radiation damage in the TEM . Ultramicroscopy 127 : 100 – 108 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of The Japanese Society of Microscopy. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Mapping structure and morphology of amorphous organic thin films by 4D-STEM pair distribution function analysis
JF - Microscopy
DO - 10.1093/jmicro/dfz015
DA - 2019-08-06
UR - https://www.deepdyve.com/lp/oxford-university-press/mapping-structure-and-morphology-of-amorphous-organic-thin-films-by-4d-tt8sIlAIxU
SP - 301
VL - 68
IS - 4
DP - DeepDyve
ER -