Synthesis and structure of high-quality films of copper polyphthalocyanine – 2D conductive polymer
Introduction
Two-dimensional electronic conjugation makes metal polyphthalocyanines (PPCs) (Fig. 1) a unique class of elementoorganic semiconductors which are of especial interest for science and technology [1], [2], [3], [4]. In particular, PPCs have been shown to possess a uniquely high dielectric constant [5], [6]. Hyper-branched (which are not truly 2D) PPCs are also investigated as promising materials for solar cells and molecular photovoltaics [7], [8], [9]. Although 2D PPCs are known for more than 50 years [10], [11], their application has been hindered by the lack of processability: they are practically insoluble in all solvents and cannot be melted or evaporated [2]. For these reasons PPCs cannot be treated by the conventional processing methods like spin-coating, sputtering or thermal evaporation. However, in order to access many promising applications of PPCs, it is desirable to obtain them in a form of thin film materials on any arbitrary wafer, in particular on dielectrics.
The reported synthetic routes for PPCs are based on reaction of metals [12] or metal salts [11] with pyromellitic acid tetranitrile (PMTN) (Fig. 2). Note that the wet-chemistry methods, suitable for hyper-branched PPCs [5], [7], [8], [9], are not suitable for the 2D polymer, since they result in a material with low polymerization degree. Hence, the reactions are performed by mixing the powdered reactants and heating under inert atmosphere [11], [12]. Other approaches include: (i) the reaction of pre-sputtered copper (1.5–30 nm layers) with PMTN vapor at T > 350 °C [12] in sealed ampoules; (ii) double-source evaporation of copper and PMTN [13] with subsequent annealing. The significant drawbacks of the first approach are the complexity of experimental set-up and the inevitable oxidation of copper surface when exposed to air even for a short time (which is especially important for the thin films of metal). In a double-source evaporation (iii) the film thickness was not precisely controlled, so the resulting films were quite thick (∼1 μm), which is also undesirable. Based on the reported IR spectra of PPC materials, it can be concluded that PPC materials synthesized by all these methods had significant amounts of low-molecular weight fractions and poor structural uniformity [13], [14]. Recently, the chemical structure of the polymer formed by the reaction (Fig. 2) was confirmed by the scanning tunneling microscopy (STM) for the case of Fe [4]. However, only nano-scale samples were achieved, while the experimental set-up was quite complex (ultra-high-vacuum conditions and atomically clean surfaces).
The fact that for over 50 years the crystal structure of CuPPC, has not been understood, is also a result of the poor processability: it is a very challenging experimental task to obtain high-quality material suitable for XRD measurements. Moreover, once the XRD pattern is obtained, its interpretation is also not straightforward; in the present work it would be impossible to assign the XRD peaks without the help of quantum-chemical studies.
Electronic properties of CuPPC were also a matter of discussions [11], [13], [15], [16]. Until now it was even unclear if CuPPC is semiconductor or semimetal. On one hand, the reliable experimental data could not have been achieved before, since the previously obtained materials always contained high concentration of monomeric and oligomeric units. There are only estimations of charge transport activation energy (0.1–0.2 eV [11], [13]). Also, it has been shown that CuPPC has a high dielectric constant (>10,000), which makes it an attractive material for electro-active composites [6]. On the other hand, to the best of our knowledge, nobody attempted theoretical study of the bulk polymer electronic structure. In fact, such theoretical study would not have any practical value without understanding of the polymer structure, especially the layer stacking mode. The only relevant study by now is the band gap estimation for monolayer made by Zhou [15], suggesting a value of 0.31 eV.
In this work we propose a new experimental approach to the synthesis of CuPPC thin films, namely the reaction of PMTN with copper in a CVD set-up. We prove that in this way the conductive CuPPC films of high uniformity can be obtained. We use infrared spectroscopy, transmission electron microscopy, XRD, UV–visible absorption and sheet resistance measurements to prove the structure and investigate the properties of the films. The quantum-chemical calculations provide support for the interpretation of the X-ray and TEM electron diffraction pattern. In discussion of FTIR spectra and electronic structure, we compare CuPPC with copper phthalocyanine (CuPC), referring to it as a monomer. One should understand however, that this analogy is of limited applicability, because CuPC has 4 benzene rings per molecule, while CuPPC has only 2 per unit cell. But since the symmetry of the former and the latter are the same, this comparison is used hereafter for the reason of convenience.
In this work we use the combination of quantum-chemical calculations and experiment to assign the crystal structure and predict electronic properties of CuPPC. Computational results provide a good support for the interpretation of experimental data, and both approaches give a consistent description of the structure.
Section snippets
Reaction of PMTN with copper and transfer to dielectric wafers
Cu films on dielectric (SiO2/Si, quartz or KBr) were prepared by diode radio-frequency sputtering of Cu targets on a Z-400 installation [17]. Prior to deposition, the working chamber was evacuated to 1×10−4 Pa. The rate of Cu deposition was 0.34 nm/s. For experiments with foil, freshly electropolished Cu foil was used.
Reactions in a CVD set-up were performed in a quartz tube reactor, placed in a two-zone oven. During the reaction, the pressure of 1×103 Pa was maintained. PMTN was evaporated in a
Chemical reaction and structure of the films
After the reaction of copper with PMTN vapors (Fig. 2), notable changes were observed. In most of the experiments with foil, the copper changed the color to purple. By the AFM measurements of the film transferred to SiO2/Si substrate, it was found that the resulting film has a thickness of >100 nm. In the experiments with pre-sputtered metal, color changed to yellow or brown (observed on quartz substrate) for more than 1 h experiments, while in short experiments (0.5 h) it still contained some
Conclusions
In this work we developed an approach to the direct synthesis of thin films of CuPPC. Although the material was obtained by a well-known chemical reaction, in the present work we achieved the polymerization degree and material uniformity much higher than reported before. This conclusion is supported by FTIR, UV–vis spectra and by the complete insolubility of the films in sulfuric acid. TEM, XRD and quantum-chemical studies yield consistent data on the structure of the polymer, including the
Acknowledgments
We acknowledge Dr. D.V. Irzhak (IMT RAS) and Dr. I.S. Bushmarinov (INEOS RAS) for XRD measurements, Dr. Yu.N. Khanin (IMT RAS) for discussions, Dr. N.A. Yarykin (IMT RAS) and Dr. V.M. Senyavin (MSU) for spectral measurements. Yu.V. thanks for support the Alexander von Humboldt Foundation.
References (23)
- et al.
Synth. Met.
(1987) - et al.
Thin Solid Films
(2011) - et al.
Spectrochim. Acta Part A
(2007) - et al.
J. Solid State Chem.
(2004) - et al.
Angew. Chem. Int. Ed.
(2012) Macromol. Rapid Commun.
(2001)- et al.
Angew. Chem. Int. Ed.
(2009) - et al.
J. Am. Chem. Soc.
(2011) - et al.
J. Am. Chem. Soc.
(2006) - et al.
Nature
(2002)