In the rapidly growing field of optical fibres for telecommunication and data transmission, such fibres are often coated for aesthetic, protective and structural reasons, as well as to prevent light intensity being lost to the surroundings. These coatings usually have thicknesses of the order <500 nm, the fibres having a diameter of ~ 200mm.
Mechanical Property Measurements
Measuring the mechanical properties, in particular the adhesion between coating and fibre, is very difficult owing to the curved geometry of the surface and the problems of positioning a pointed diamond tip so that it moves along the longitudinal axis of the fibre during a scratch test.
This problem has been solved by using a specially manufactured diamond knife with a tip angle of 80°, although other angles have also been tested. The knife (see Fig. 1) is made using a high-purity natural diamond with its  orientation along the axis of the blade, ground to the desired geometry and angle, and mounted on a standard stub for fitting to the Micro Scratch Tester (MST) from CSM Instruments. The company producing the knives is Diatome Ltd., Bienne, Switzerland.
A selection of optical fibres having different metallic coatings have been tested with the modified MST. It was found sufficient to hold the samples securely by taping them down to the flat sample table, and leaving a small section exposed for testing. The large blade width (several mm) made positioning very easy without any risk of the blade sliding off the side of the coated fibre. However, it was found important to mount the blade exactly perpendicular to the fibre in order to ensure reproducible results.
A typical load range was from 0 to 100 mN over a length of 5 mm, giving an adequate load ramp and well-defined critical failure points. Scanning Electron Microscopy (SEM) of the knife blade before and after use confirmed that no visible damage had occurred as a result of scratching under load.
Figure 1. Principle of scratch testing on coated optical fibres using a sharp diamond knife of angle 80°.
Figure 2. Results for a 200 nm coating (a) and for a 500 nm coating, (b) and (c). Note the brittle flaking of material for the former, whereas the latter has a distinct point at which the substrate is reached, after which cracking and delamination occur. Scratch direction is from left to right (mag. 500x).
The optical micrographs in Figure 2 show the different types of coating failure encountered using this method. The 200 nm coating had two critical failure points; the first being the onset of chipping, the second being the point at which continuous chipping begins and the substrate is attained. Both modes are shown in Figure 2 (a). For the 500 nm coating, no initial chipping occurred, the only critical failure point being where the substrate was reached. However, it is interesting to note that at loads greater than this, cracking and delamination was evident along the sides of the scratch path, as depicted in Figure 2 (c).
The successful use of a diamond knife have been shown for an industrial application where such a method is the only one capable of adequately controlling the adhesion between such coatings and optical fibres of small diameters. Future work is envisioned with different knife angles to investigate their potential for other sample geometries which cannot be measured by the conventional MST technique.