Chromium nitride coatings were deposited using a Teer Coatings
Ltd. UDP350 closed field unbalanced magnetron sputtering system,
which has been described in detail elsewhere [4]. Control of reactive
gas (nitrogen) flow was achieved by using either a mass flow
controller, or an optical emissions monitoring (OEM) feed back loop,
described elsewhere [9]. The first stage of this work was to investigate
the variation in target voltage with nitrogen flow rate. This was
achieved by initiating a constant 1 kW discharge at the Cr sputter
target (300 mm×100 mm, 99.5% pure) in an argon only atmosphere
at a constant flow rate of 7 sccm, equivalent to a constant partial
pressure of 0.1 Pa. The OEM signal at this point was taken as the 100%
metal signal. The nitrogen flow rate was then increased incrementally
and the target voltage and OEM signal were monitored. Once the
voltage stabilised and no longer increased with increasing flow rate,
the nitrogen flow rate was decreased in the same increments. In order
to identify the best CrN forming region on the curve films were
deposited at 90%, 65% and 40% of the full metal OEM signal, which
matched salient points on the curves. The films were then analysed by
XRD (Seifert&Co.) using Cu Kα radiation in the Bragg–Brentanomode.
For the main experimental array, chromium nitride films were
deposited by DC reactive sputtering at a target power of 1 kW, a
chamber pressure of 0.1–0.2 Pa and a run time of 90 min. Coatings
were deposited onto 50 mm diameter tool steel (Rockwell C hardness
56) and aluminium substrates, which had been polished to a 1.5 mm
finish and then cleaned in propanol using an ultrasound bath. Tool
steel substrates were used for scratch testing and nanoindentation;
aluminium ones for EDX (Edax, Trident) and XRD analysis.
Prior to deposition the substrates were also ion etched at −650 V
substrate bias and varying frequencies for 6 min (details given in
Table 1). During this stage a low power (125 W) discharge was
maintained at the Cr sputter target. Both the target and substratewere
powered by a dual channel Advanced Energy Pinnacle Plus supply. To
improve adhesion, in all cases a 100 nm thick chromium interlayer
was deposited prior to reactive sputtering. Reactive gas control was
again achieved by optical emissions monitoring (OEM). An OEM
turndown setting of 40% of the full metal signal was used, which the
hysteresis studies showed to be well inside the range for stoichio-
metric CrN. The run conditions for the array are listed in Table 1.
The coatings were characterised in terms of their structures and
composition by SEM (Zeiss Supra 40) and XRD. The hardness of the
coatings was measured by nanoindentation (Micro Materials Nanotest
100). In the latter case 50 indentationsweremade at a load of 50mNand
a mean value taken. The adhesion of the coatings was characterised by
single pass scratch testing (Teer Coatings Limited ST3001), failure being
taken as cohesive spallation inside and outside thewear track, or ductile
perforation of the film within the wear track. A small section of each
substrate was shielded during deposition, and film thickness was
measured using a DEKTAK profilometer.