Optical Emission Spectroscopy as a tool for online monitoring of Diamond Like Coating deposition in a plasma enhanced chemical vapor deposition process

Collectively, tetrahedral amorphous carbon and their hydrogenated variants, are termed as Diamond Like Carbon (DLC). The ratio of the two variants of the hydrogenated carbon and the hydrogen content present are the controlling factors that determine the structure and properties of the diamond like carbon. Unfortunately, the sp3 species has overtime been seen to provoke high internal compressive stresses within a coating which restricts the achievable film thickness.

As a fix, annealing with structurally graded DLC has been applied successfully and has even been seen to enable tailoring of properties of the coating as desired. Designing of such graded coatings requires one to know the correlation between the structure of nanocoating and the plasma parameters, such as plasma power and pressure. These parameters vary and so does the coating’s structure. Consequently, reproducing identical coatings becomes quite tedious as it involves trial and error procedure. Several plasma diagnostic techniques have therefore been developed but each has its pros and cons. Thus, it is imperative that Optical Emission Spectroscopy be thoroughly investigated, as it is the most promising amongst all the techniques.

To this note, Laval University researchers led by Professor Gaétan Laroche established a predictive model of DLC properties using a multivariate analysis method. In addition, they aimed at carrying out a comparison of the efficiency of both the deposition process parameters and Optical Emission Spectroscopy data as predictive tools to monitor DLC structure and its mechanical properties. Their work is currently published in the research journal, Thin Solid Films.

The research method employed commenced with the pre-deposition process. Next, DLC deposition was undertaken after which deposition parameters were determined. They then designed the experiments and evaluated the predictive potential of Optical Emission Spectroscopy using partial least square regression (PLSR) analysis.

In their first model, the authors got an excellent prediction of the coating properties while using the plasma process experimental parameters such as bias voltage, plasma duty cycle, plasma pressure, and power used to generate the discharge. However, they also succeeded to get an equivalent prediction while using parameters derived from optical emission spectroscopy.

In summary, Professor Gaétan Laroche and his research team successfully presented the application of partial least square regression analysis to predict plasma enhanced chemical vapor deposited DLC coating properties using Optical Emission Spectroscopy data recorded from the plasma. In general, the results obtained showed that the Optical Emission Spectroscopy derived data was capable of replacing some process parameters to predict the DLC properties. Altogether, their paves the way for the possible utilization of Optical Emission Spectroscopy as a monitoring tool for DLC deposition process, especially in multilayer or gradient DLC film deposition, where in situ film properties have to be modified during the deposition process.