Novel Strategy for Organic Cocrystals of n-Type and p-Type Organic Semiconductors with Advanced Optoelectronic Properties.

Cocrystallization has been applied widely for material synthesis. Recently cocrystal of organic molecules has been developing rapidly, taking the advantages of the flexibility and self-assembly of organic molecules. Here we report an experimental study of a cocrystal of copper-phthalocyanines and fluorinated ones. We have grown the samples via the vapor-phase deposition of the mixture with different mass ratios from 1:13.5 to 6:1. As suggested by our scanning electron microscopy (SEM), X-ray diffraction (XRD), and Raman spectroscopy, new crystal structures and morphologies through our novel strategy for the cocrystallization of these molecules have been found. Our work will provide a solid foundation to systematically synthesize the cocrystal of phthalocyanine molecules with new crystal structures, thus providing the opportunity to advance material properties.

Introduction

Organic materials have many significant edges compared with inorganic materials, such as low cost, lightweight, ease of self-assembly, and mechanical flexibility.[1−4] Organic molecular donor/acceptor films have already been widely studied to have a deep understanding of the film growth, phase separation mechanism,[5] charge transfer,[6] intermolecular coupling effects,[7] and optical properties,[8,9] which is helpful for developing organic-based device application, such as photovoltaic devices.[10] For example, mixed films of perfluorinated pentacene and diindenoperylene deposited via co-evaporation by organic molecular beam deposition show two structures, one is a standing-up orientation and the other is a lying-down orientation.[11] Recently, molecular cocrystallization, a methodology to design organic materials with advanced functionality, has aroused much attention.[12−14] Taking advantages of organic synthesis and self-assembly, the organic cocrystal engineering can facilitate the design, fabrication, and optimization of new organic materials with novel physical and chemical properties, including improving charge transport,[12] isolating magnetic moments,[13] and enhancing nonlinear optical properties.[14] Using cocrystal engineering, we can have a great opportunity to develop ambipolar charge transport, rather than relying on electron or hole transport alone. In 2004, the combination of bis(ethylenedithio)tetrathiafulvalene (BEDT-TTF) and tetracyanoquinodimethane (TCNQ) was cocrystallized for the first time, which had a low-temperature ambipolar behavior for charge transport.[15] Organic cocrystal engineering has also demonstrated its promising potential for ferroelectronics.[16] The isolation of magnetic moments is very important for protecting magnets from the environment and the interaction with others for single-molecule magnetism and quantum computing, which recently has been realized through a cocrystallization of thin films consisting of metal-free phthalocyanines (H2Pc) and copper phthalocyanine (CuPc).[17] Moreover, the nonlinear optical properties, owing to the charge-transfer state, have also been found in the organic cocrystal materials (styrylpyridine–tetracyanobenzene).[18]

Over the past few decades, considerable efforts have been dedicated to the scientific investigation of phthalocyanine (Pc) molecules because of their outstanding optoelectronic and magnetic properties. Pc molecules can be explored in many research fields, such as organic solar cells,[19] organic field-effect transistors (OFETs),[20] magnetic semiconductors for spintronics, and quantum computing.[17] Pc is a planar π-conjugated organic molecule with a D4 symmetry, whose the center cavity can accommodate almost all of the elements in the Periodic Table, which can allow us to vary the magnetic properties through chemical synthesis.[21−26] For instance, Heutz group has reported that the as-deposited CoPc thin film showed the anti-ferromagnetic magnetic properties that can survive at a temperatures above the boiling point of nitrogen,[21] owing to a large intrachain exchange interaction. Meanwhile, fluorinated metal phthalocyanine, such as F16CuPc, as a typical n-type organic semiconductor, has attracted attention to the application of organic-based electronic devices due to excellent electrical and optical properties as well as its air-stable property.[27−31] In the report of Jiang et al.,[28,29] the change in the electronic properties of organic semiconductors from p- to n-type was observed with different fluoridation substitutions in metal phthalocyanines. The magnetic properties of F16CuPc have also been studied previously, showing that the exchange interaction between neighboring copper ions is weak.[32] Recently, organic cocrystal engineering, or namely organic semiconductor alloying, has been applied in Pcs to further synthesize new organic materials for advanced physicochemical properties.[33] For example, Ballav group has demonstrated a promising route to obtain the room-temperature ferromagnetism in supramolecular aggregates of NiPc–ZnFPc and FePc–ZnFPc systems.[34,35] Leo group has revealed that the band structure of the organic cocrystal can be engineered by alloying Pcs and their halogenated derivatives.[36] The magnetic properties of an alternating CuPc/CoPc chain have been computed from first-principles, in which a ferromagnetic interaction has been found, and the magnetic quantum metamaterials were proposed.[37] The abundance of Pc-based material resources provides us with a good experimental and theoretical platform to cocrystallize different types of Pcs systematically, thus optimizing and designing materials properties. Despite their promise to create excellent optoelectronic and magnetic properties, the studies on Pc cocrystals are still scarce in the literature probably because the synthesis of the cocrystals of different types of Pc molecules is a great challenge that is yet to overcome.

In general, three experimental methods can be used to fabricate organic cocrystals via intermolecular interactions to spontaneously self-assemble, including liquid-phase, vapor-phase, and solid-phase methods.[33] For the synthesization of Pc cocrystals, liquid-phase methods are the most common due to the advantages of simplicity and low cost.[25,26] However, the method provided in the previous work by solution-processed self-assembly method failed to fabricate the CuPc–F16CuPc cocrystal,[38] rather suggesting the infeasibility to synthesize Pc–Pc cocrystals by this method. Although vapor-phase methods have been widely used to fabricate the single crystal of pure phthalocyanines, even p–n-junction single crystalline,[39,40] its application in the synthesis of Pc–Pc cocrystals is yet to be explored. In addition, co-evaporation is another general way to deposit blended molecule thin films.[36,41] In the solid-state phase, the organic cocrystal can be fabricated through nonvalence bonds and π–π intermolecular interactions.[42,43]

Herein, we report a novel strategy for the preparation of the cocrystal of CuPc–F16CuPc using organic vapor-phase deposition (OVPD). This is the first time to cocrystalize CuPc and F16CuPc using this method, which is defined as the zeta-phase CuPc–F16CuPc cocrystal (ζ-F16CuPc–CuPc). Our X-ray diffraction (XRD) results reveal that the crystal structure of the CuPc–F16CuPc cocrystal is different from that of CuPc and F16CuPc individually. We have investigated a range of mixing ratios between CuPc and F16CuPc, suggesting that a range of new organic materials with interesting properties can be obtained through this novel strategy. The following discussion falls into three sections: in Section , we report the results; in Section , we draw some general conclusions; and in Section , we introduce our experimental methods.

Results and Discussion

Figure a shows the schematic illustration of the growth process of the F16CuPc–CuPc cocrystals. Glass substrates were placed at downstream positions at the edge of the single furnace, where a low-temperature area is obtained for the crystallization of CuPc and F16CuPc. The temperature difference is important as crystals are usually crystallized in the low-temperature zone. Meanwhile, the deposition temperature is a critical factor that affects the morphology and crystal structures of the as-prepared crystals as well as the cocrystals,[44,45] as the β-phase CuPc are usually deposited at high temperatures, while the deposition temperature for the α-phase is low. Figure S1 shows the as-prepared samples. The appropriate vaporization temperature for both CuPc and F16CuPc was established to be ∼460 °C through experiments, which was a prerequisite to conducting the fabrication of the cocrystals by a one-step method via OVPD. When deposited with mixed powder, the morphology of the crystals on the glass substrate shows a difference compared with that of pure powders. Large-size crystals can be formed on the top of the substrate with pure F16CuPc powder, while the amount of large-size crystals decreased at the powder mole ratio of 6:1. This may be attributed to the change of seed layer or nuclei aroused by the intermolecular interaction of F16CuPc and CuPc. We infer that the large crystals can be assigned to the CuPc or F16CuPc individual crystal because few large crystals can be found at a mole ratio of 1:1, which is the pure F16CuPc–CuPc cocrystals, as shown in XRD results. Scanning electron microscopy (SEM) images in Figure b reveal the morphology of the F16CuPc–CuPc cocrystals. The morphology of the cocrystal is different from that of the crystal formed by the individual component. The SEM image in Figure b at the mole ratio of 1:1 shows that the F16CuPc–CuPc cocrystal formed a nanorod structure; however, the pure CuPc crystal is a one-dimensional rod structure and pure F16CuPc crystal is a two-dimensional belt structure, as shown in Figure S2. The SEM images for other mole ratios are shown in Figure S3. The energy-dispersive spectrum (EDS) for the F16CuPc–CuPc cocrystal is shown in Figure S4, the estimated element ratio for F and Cu is about 8.1:1, indicating a molecular ratio of 1:1 for CuPc and F16CuPc in the cocrystal. Therefore, we could claim the coexistence of CuPc and F16CuPc molecules in the cocrystal.

Figure 1

(a) Schematic illustration of the setup for crystal growth. (b) SEM image of F16CuPc–CuPc cocrystals deposited on the glass substrate with a mixing ratio of 1:1. Inset is the as-fabricated F16CuPc–CuPc cocrystal materials in a mixing ratio of 1:1.

Figure a,b shows the X-ray diffraction (XRD) pattern of the as-prepared individual CuPc and F16CuPc crystals and F16CuPc–CuPc cocrystals. It can be seen that the XRD pattern of the F16CuPc–CuPc cocrystals is different from those of the individual CuPc and F16CuPc. For pure CuPc crystals, the two peaks at 7.07 and 9.24°, corresponding to (001) and (201̅), respectively, can be attributed to β-CuPc, which is in a good agreement with previous works.[46,47] As for the pure F16CuPc crystal, the characteristic peak at 6.30° is related to the (002) plane, according to that reported by Yoon et al. The feature peaks for the F16CuPc–CuPc cocrystals is located at 6.56 and 6.85° with the interplanar spacing of 13.46 and 12.91 Å, respectively. This suggests that our cocrystals have a new crystal structure compared with pure ingredients. The detailed information, including 2θ, lattice spaces, and relative intensities, for F16CuPc–CuPc cocrystals, β-CuPc, and β-F16CuPc is provided in Table S1.

Figure 2

(a) X-ray diffraction (XRD) pattern with a measured angle from 3 to 35° for the as-fabricated CuPc, F16CuPc, and F16CuPc–CuPc cocrystals with a mixing ratio of 1:1. (b) Zoom-in image of the peak position in the diffraction angle range from 8 to 35°. All of the XRD results were obtained from the crystalline powders.

Similar to the pure components, the cocrystals can also form different crystal structures.[48] However, the F16CuPc–CuPc cocrystals reported in this work prefer to form one particular crystal structure, which appears at the powder mole ratio of 1:1, although different powder mole ratios can be applied in the fabrication of the cocrystals. The possible reason is the deposition temperature, crucially influencing the crystal structure, is the same for each process. Note that when the powder mole ratio was 1:13.5, the XRD patterns shown in Figures and S5 indicate the preferential growth of the β-CuPc crystals along the [001] direction. This may be the consequence of the formation of the F16CuPc–CuPc cocrystals, which affects the nucleation of CuPc, thus leading to the preferential growth of CuPc crystals. Interestingly, we find that even at the low powder mole ratios of 1:13.5 and 6:1, CuPc and F16CuPc molecules can easily form cocrystals, as shown in the fitted XRD curve in Figure and Table S2, perhaps due to the similar molecular structures and similar lattice constants and strong attractive interactions between F16CuPc and CuPc.[39] To obtain pure F16CuPc–CuPc cocrystals, we adjusted the powder mole ratios to 1:1, which, in fact, tune the vaporized molecules concentration of F16CuPc and CuPc. According to the XRD data, three possible phase combinations may be concluded as follows: the β-CuPc mixed with ζ-F16CuPc–CuPc cocrystals, pure ζ-F16CuPc–CuPc cocrystals, and F16CuPc crystal mixed with ζ-F16CuPc–CuPc cocrystals. Notably, the pure ζ-F16CuPc–CuPc cocrystals can be obtained only for a 1:1 powder mole ratio, which is probably due to large interaction energy between CuPc and F16CuPc molecules, which is in agreement with the study by Hinderhofer et al. In their work, pentacene (PEN) and perfluoropentacene (PFP) formed a mixed-crystal structure when the mixing ratio was 1:1, and a phase separation was observed for other mix ratios.[49]

Figure 3

Fitting plot of the XRD peak of CuPc, F16CuPc, and cocrystal deposited on the glass substrate at different powder mole ratios (F16CuPc/CuPc). All of the XRD results were obtained from crystalline powders. Inset is a zoom-in image of the peak position for the mixing ratio 6:1.

Raman spectroscopy has been found to be a very useful technique used in the study of cocrystals to reveal their crystal structure and property relationship. Due to the combination of the donor and acceptor molecules, the electron densities will be redistributed, and the structures of the donor and acceptor molecules will be influenced, generally, leading to a peak shift or appearance of new peaks in the Raman spectra.[33] As shown in our Raman spectra in Figure , although the cocrystal shares a similar feature to the pure ones, a few differences between them can be spotted. Between 680 and 730 cm–1, the two main peaks of Raman spectra for the cocrystal overlap with those of CuPc (∼680 cm–1) and F16CuPc (∼731 cm–1). The peak at 595.0 cm–1 for β-CuPc and at 584.7 cm–1 for β-F16CuPc correspond to the macrocycle ring breathing mode shift to 592.1 cm–1 for the F16CuPc–CuPc cocrystal. However, the peak at 680.3 cm–1 for β-CuPc related to the C–N–C and N–C–C bending modes is divided into two peaks at 675.9 and 686.2 cm–1 for the F16CuPc–CuPc cocrystal. The peak centered at 734.7 cm–1 for β-F16CuPc is mainly because the symmetric stretching of the four C–N–C bridges of vibration changes to 731.8 cm–1 for the F16CuPc–CuPc cocrystal. This type of feature can be also found across the whole range of energies, although there is some change in the peak intensities. Between 1200 and 1600 cm–1, three of them overlap very well. The peaks related to the central in-plane vibration of C–N locate at 1528.7, 1522.8, and 1522.8 cm–1 for the β-CuPc, β-F16CuPc, and F16CuPc–CuPc cocrystal, respectively.[50] Our Raman spectra, therefore, suggest that indeed CuPc and F16CuPc have been cocrystalized, supporting our other measurements for the cocrystal.

Figure 4

Raman spectra of the pure CuPc (red), pure F16CuPc (purple), and the cocrystal (green) with a mole ratio of 1:1. The peak shift and the change of intensities of Raman spectra among pure CuPc, pure F16CuPc, and the cocrystal suggest that the two types of molecules are indeed mixed in the cocrystal.

Conclusions

We have carried out organic cocrystallization by mixing CuPc and F16CuPc in the vapor phase, which was the first time for this methodology. The zeta-phase CuPc–F16CuPc cocrystal with unique crystal structure and properties, ζ-F16CuPc–CuPc, has been successfully fabricated by our novel cocrystal strategy. Systematic investigation on the influence of different mole ratios on the cocrystal structure indicates that the zeta-phase F16CuPc–CuPc cocrystal has a fixed CuPc and F16CuPc mole ratio of 1:1. The unique and different Raman spectra of ζ-F16CuPc–CuPc, compared with those of β-CuPc and β-F16CuPc, agree well with the XRD study and further confirm that a new cocrystal structure of ζ-F16CuPc–CuPc has been produced. This work paves the way for a series of new cocrystals of Pc molecules with new and advanced material properties via vapor-phase cocrystallization. The novel strategy for the rational design of organic cocrystals of n-type and p-type organic semiconductors with advanced optoelectronic properties has significant potential in device applications such as OPV, OFET, organic light-emitting diode (OLED), etc.

Experimental Section

The F16CuPc–CuPc cocrystals were prepared by organic vapor-phase deposition. F16CuPc and CuPc powders were purchased from Tokyo Chemical Industry and used without further purification. A total of 40 mg of F16CuPc–CuPc powders, with a set of mole ratios (F16CuPc/CuPc) from 1:13.5 to 6:1, were loaded in a quartz tube. The quartz tube was then put in a single-zone heating furnace, keeping the powders in the middle of the furnace, and glass substrates were put in the downstream direction. Ambient air was flushed out by N2 gas (99.999%) at a rate of 400 standard cubic centimeters per minute (sccm) for 30 min before heating the mixed F16CuPc/CuPc powders. After that, the mixed powders were heated at 460 °C for 300 min, at a flow rate of 400 sccm. The vaporized F16CuPc and CuPc molecules self-assembled into crystals on the glass substrates, as shown in Figures and S1.

The morphology was observed using a scanning electron microscope (Hitachi 3400N). The crystal structures of the cocrystals mentioned above were measured by means of X-ray diffraction (XRD) using Cu Kα (λ = 0.154056 nm) (Rigaku, TTR III). The Raman spectroscopy was carried out using an inVia Raman Microscope (Renishaw).