Polytype transformations

Yoder and Eugster (1955) and Velde (1965) have established by hydrothermal experiments that the 2M polytype is the stable form of muscovite and that the lMd polytype is metastable. Velde (1965) found that the 2M polytype is stable at temperatures as low as 125°C and suggested that the sequence of polytype transformation is lMd- TM->-2M with an increase of either time, temperature, or pressure. He concluded ... 

An early study of mica by HRTEM was reported by Buseck and lijima (1974). They clearly observed three dark lines representing a mica layer (the lines correspond to the two tetrahedral sheets and one octahedral sheet) and that cleavage was formed at the interlayer. During a quarter century after this pioneering work, many HRTEM studies for mica and related phyllosilicates have been reported (for instance, see the references in Baronnet 1992). These included many studies of mica, e.g., polytypism, transformations, defects and interface research. In the following section, recent HRTEM and related techniques are briefly reviewed. Next, two topics of HRTEM investigation, polytype and defect analyses are presented based on studies, mainly by the author and his colleagues. 

Zhadanov [7] visualised the SiC stacking sequences by considering the sequence of non-basal tetrahedral planes. The observed zig-zag of layers ([111] direction for cubic, and [0001] direction for hexagonal and rhombohedral) was represented by a notation which denotes the number of consecutive layers without rotation, the total stacking sequence, and a subscript indicating the sequence repetition. A detailed discussion of polytypism, polytypic transformation and notations has been reported by Jepps and Page [8]. 

The transformation of the epitaxial layer polytype to a structure different from 3C, 6H or 15R was first obtained with the use of liquid phase epitaxy by the travelling solvent method [63]. The authors employed scandium-based solutions and they observed the 4H-SiC layer growth on substrates of the 6H polytypes. Later, the detailed studies of Vodakov et al [64] showed that the 6H to 4H polytype transition occurs when the growth is performed onto the carbon face and only if the melt contains excess carbon. No polytype transformation occurred if the melts were free from excess carbon or contained excess silicon. In [63] no carbon was added to the solvent intentionally, so most probably it entered the melt via reaction with the graphite... 

Further studies [65,66] have shown the possibility of growing the 4H polytype without any impurities providing a moderate excess of carbon is provided on the crystal face. Controlled introduction of barium and phosphorus has ensured the transformation of 6H to 15R or to 21R. Development of the polytype transformation methods has permitted the growth of epitaxial layers of 4H-SiC free from foreign polytype inclusions on 6H substrates [65,66]. The growth of 4H ingots on seeds of other polytypes has been reported in [22,25,52]. These authors did not introduce any impurities in the growth system. The probable reason for the polytype transformation is the vapour depletion of silicon. 

Polymorphism/polytypism Transformations/transitions and/or order/disorder phenomena... 

A number of theories have been put forth to explain the mechanism of polytype formation (30—36), such as the generation of steps by screw dislocations on single-crystal surfaces that could account for the large number of polytypes formed (30,35,36). The growth of crystals via the vapor phase is beheved to occur by surface nucleation and ledge movement by face specific reactions (37). The soHd-state transformation from one polytype to another is beheved to occur by a layer-displacement mechanism (38) caused by nucleation and expansion of stacking faults in close-packed double layers of Si and C. 

A progressive etching technique (39,40), combined with x-ray diffraction analysis, revealed the presence of a number of a polytypes within a single crystal of sihcon carbide. Work using lattice imaging techniques via transmission electron microscopy has shown that a-siUcon carbide formed by transformation from the P-phase (cubic) can consist of a number of the a polytypes in a syntactic array (41). 

In nature the Fe-rich illites (glauconite and celadonite) appear to progress from the lMd to the 1M polytype. The Al-rich illites are predominantly the lMd and 2M varieties. If the 1M polytype is an intermediate phase, it is surprising that it is not more abundant in sediments. Recent studies of unmetamorphosed Precambrian sediments (Reynolds,1963 Maxwell and Hower,1967) have shown that the lMd polytype is relatively abundant in ancient sediments, particularly in the extremely fine fraction. The senior author has noted the relative abundance of the lMd polytype in the fine fraction of most Paleozoic rocks but has considered most of it to be mixedlayered illite-montmorillonite rather than illite. Weaver (1963a), Reynolds (1965), and Maxwell and Hower (1967) have shown that during low-grade metamorphism water is squeezed from the expanded layers and the lMd polytype is transformed into the stable 2M polytype. 

Silicon carbide is covalently bonded with a structure similar to that of diamond. There are two basic structures. One is a cubic form, /i-SiC which transforms irreversibly at about 2000 °C to one of a large number of hexagonal polytypes, and the other is a rhombohedral form also with many polytypes. Both the hexagonal and rhombohedral forms are commonly referred to as a-SiC. 

The form that silicon carbide takes depends on many factors including thermal history, impurity type and level, and environment. The p form is generally felt to be the stable phase at low temperatures, whereas the a form is the high-temperature form. There are many exceptions to the rule, as the conversion to a from /3 and the converse have been reported. The stability and transformations of the various polytypes vary among themselves and constitute a subject that is too broad for this effort. The basic a and p descriptors will be used for the remainder of this section. 

Takeuchi Y, Haga N (1971) Structural transformation of trioctahedral sheet sihcates. Shp mechanism of octahedral sheets and polytypic changes of micas. Mineral Soc Japan Spec Paper 1 74-87 (Proc IMA-lAGOD Meetings 70, IMA Vol)... 

---