Using a semiempirical rule based on a range of existing hard materials, Liu
and Cohen predicted that some of the crystalline forms of C3N4 might have
values of bulk modulus close to, or even greater than, that of diamond. Sub
sequent theoretical work has supported this result, although modifying whic
h of the crystal structures had the most extreme properties. Based on the a
ssumption that hardness is primarily determined by the bulk modulus, a larg
e number of groups initiated research to try to make this 'harder-than-diam
ond' material. Although today there are doubts as to the validity of this a
ssumption, the ability to synthesise materials with such a large bulk modul
us is of great interest because such substances can be expected to have ext
reme properties, such as thermal conductivity, compression strength, acoust
ic velocity, etc. A review is presented of the methods used, and the result
s obtained, by a variety of groups in their attempts to prepare carbon nitr
ide films. The preparation techniques have been somewhat arbitrarily divide
d into the following sections: (1) atmospheric-pressure chemical processes,
including pyrolysis and explosive shock; (2) ion-beam deposition, includin
g cathode are; (3) laser techniques; (4) chemical vapour deposition process
es, including direct-current (DC) are, hot-filament chemical vapour deposit
ion (HFCVD) and plasma-enhanced chemical vapour deposition (PECVD); and (5)
reactive sputtering, with and without deliberate substrate biasing. We hav
e also included a section describing some of the problems associated with t
he analysis of carbon nitride films. A brief description of the techniques
is provided and we discuss some of the problems associated with the charact
erisation methods that have been used in the analysis of the deposited mate
rial. A somewhat speculative set of conclusions is presented which we hope
will be of assistance to colleagues working in this field. (C) 1999 Elsevie
r Science S.A. All rights reserved.