Design of Zr(IV) and Hf(IV) Co-Ordination Compounds - Precursors for MOCVD Synthesis of Protective Coatings

Design and meation of new materials with unusual, predictable or predeterminated properties become one of the most actual problems of modem applied chemistry. Main demands are discussed, which are made to chemical compounds used in CVD synthesis of bciional materials. A &vision of volatile metal-containing precursors into main classes is gven. On the hasis of volatility and thermal stabiliti criteria elaborated at IGIC-NASU, we have designed, synthesized and characterized a series of new advanced precursors for Zr(lV) and Hf(1V). Their properties are compared to those of know11 species. A possibility of formation of requested properbes and structure of Enal materials is shown. In different branches of modern technique and industry a progress and development depend to a great extent on the possibility of application of new materials with predeterminated properties. Such materials could be synthesized e.g. by one of the most promissing technique Metal-j_)rganic Chemical yapour Deposition. IrrespectiveIy, what is applied either a direct thermal destruction of gaseous substances or an interaction of the latter with a solid substrate, or their destruction in high-frequency plasma, laser beam etc., in any case, the volatile and thermally stable to a certain extent chemical compound plays a dominant role. Thus, one of the main problems of modern chemistq is the development of fundamental research on structure and bonding correlation for chemical compounds, and elaboration of industrial chemical processes of high efficiency on this basis. 2. DISCUSSION It is a well-known fact [I], that chemical substances could be considered as thermalty stable and volatile if they possess a propmty of volatilization and condensation without a composition change at moderate temperatures (below 700-800 K). It is also necessary to note, that the substances in question must possess a whole combination of properties, necessary for industrial realization of CVD processes in its diierent variations : the already mentioned volatility should be sufficiently high at relatively low temperatures (50-250'~): 0 destruction temperatures sould bc sufficienth low (1 50-650 C): Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1995557 JOURNAL DE PHYSIQUE IV * volatile gaseous pyrolysis products. except the final one, should be readily eliminable &om the reaction zone; they should be non-inflammable in air: toxicity of the complexes and their main thermal decomposition productsshould be as low as possible: relatively low costs. Taking the above into a consideration and analyzing the properties requested above a folowing classification of CVD precursors could be given [2]: compounds with monodentate a-donor ligands (halogenides, alcoxidea amides, aorganometallic compound$); hydrides; complexes with n-acceptor type ligands (carbonyls, borohydrides, tri£luorophosphines ); organometallic n-complexes (with cyclopentadienyL arene, ally1 hgands); inaqueous salts of several acids(carbozrilates. nitrates, perchlorates); chelate complexes (p-diketonates and their derivatives, diallcyldithiocarbaminates, tropolonates etc. ); mixed ligands complexes. which usualy contain species from the above list. By elimination of halogenides. hydrides and inaqueous salt of mineral acids the remained part would represent the basis for MOC\Tl. But only a limited number of elements (usually early transition metals) form substances within each class. providing acceptable volatility and thermal stability parameters both for research and technological demands. Probably, for such elements as Si, B, P, As. Sb, etc., their a-organometallic compounds remains up to now the most acceptable precursors despite several properties limiting their technological applications. To our mind, the most universal ones for d-. f-elements Au and platinum group metals are their chelates and mixed ligand complexes. ,4s well as ligands of different chemical nature (dentate number, name of coordinating atom. substitutes nature etc.) could be used, it enables us to vary in a wide range the internal characteristics of the complexes (metal-ligand bond energg, central metal effective charge etc.), hence to control their fragmentation passways. and intermolecular interaction parameters (crystal lattice Q e , dipole moment etc.), and in turn to vary their volatility. We would like to give an example. that illustrates a possibility to design complexes suitable for a purposeful formation of a material with predeterminated properties.On the basis of volatility and thermal stability criteria [3,4] vve have chosen mixed ligand co-ordination compounds of Zr(1V) and Hf(IV). Bearing in mind the electronic structure of these metals and a derivative of it high co-ordination number in condensed phases, revealed in a trend to intermolecular interactions between the central metal atom and donor atoms of hgand's periphery of a neighbour moIecule. a formation of closed structures with the maximum steric shielding of thecentral metal atom are required. Thus. we ftved our attention on metallocene species Cp2HfL2 and Cp2ZrL2 [4] where Lbicyclic alifatic alcoholato-ligands. Such ligands contain only one donor atom oxygen bonded to the metal but shielded from other interactions by voluminous and indifferent bicyclo[2.2.1 .]heptane framework. In order to resolve some theoretical problems of co-ordination chemistry and to realize a purposeful gas-phase synthesis of materiaLs and films, a knowledge of a series of fundamental precursors characteristics such as their volatility. thermal destruction passways and products appear to be very important. A combination of some physical and physico-chemical methods was used for this purpose: sublunation temperature determinations, thenno-gravimetric analysis, mass-spectroscopy (MS) and mass analyzed ion kinetic energy spectroscopy (bmCES). The data obtained are generalized in table 1 in comparison with known species. Table 1. Gm&ed data on volabhty and thermal stabhty of new and traditional Zr and Hf precursors. / bicyclic alcoxides i 200 250 1 600 650 MOC I I -1 i Y 2MS2 1 Type t subl; "C 1 tCVD. "C , Main I 1 I of compound 1 (1 torr) 1 (5-10 ton) I final product 1