Surface Morphology and Deposition Rate of Thin Film

The surface morphology of SiO_x films was observed using SEM, and the thickness of the films was measured from the SEM cross section images. Thin films were deposited on the substrate using a moving stage controlled at 10 mm/s per pass and an inner diameter tube of 4 mm. The residence time of one pass is 0.4 s. Figures 3.12 (a) and (b) show SEM cross section images and surface morphology of the SiOx film, respectively. In this case, the total residence

24

time was 80 s for 200 passes and the thickness of the thin film was 120 nm. Thus, the deposition rate was 90 nm/min. The roughness of the thin films measured by AFM and the scan size of the samples was 2x2 μm. In addition, the plasma temperature of 100^oC was measured at 5 mm from the exit of the quartz tube using a K type thermocouple.

3.3.1 Effect of Number of Treatment Passes

The surface morphology of SiO_x thin films observed under various treatment passes are shown in Figure 3.13. The morphology of the surface structure changed from 4 passes to 30 passes. The beginning of thin film grown on the surface can be found via 4 passes and the surface roughness was 2.2 nm (Figure 3.13 (a)). The thickness of the thin films measured about 100, 200, and 350 nm via 10, 20, and 30 treatment passes, respectively. SEM surface images are shown in Figure 3.14. The particle size and roughness of the thin films also increased with the increasing number of treatment passes.

3.3.2 Effect of Input Power

Figure 3.15 shows the effect of the RF power on the thickness of thin films from 35 W to 50 W. If the RF power was less than 35 W, the deposition rate of the thin film was too slow to be measured. The thickness of the thin films exhibits dependence on the RF power, and it

25

corresponds to the increase in the concentration of active species such as excited argon atoms or excited molecules of oxygen in the plasma stream from the discharge region [Alexandrov et al, 2005]. The concentration of argon metastable excited atoms increased with the increase

of RF power and caused partial dissociation and formation of silicon-containing active species which further enhanced the deposition rate of thin films.

3.3.3 Effect of Treatment Distance

Figure 3.16 shows the effect of the thickness of thin films on various treatment distances ranging from 3 mm to 7.5mm (from the end of the quartz tube). The thickness of the thin films deposited exhibits a strong dependence on treatment distance. HMDSO monomer dissociated via plasma formed the silicon-containing active species and further polymerized on the surface. If the substrate was close to the exit of the monomer, the active species increased and the deposition rate was enhanced. Contrarily, the active species were easily quenched in the ambient via far distance treatment. However, the 3 mm treatment distance was not better for higher deposition rate where the thin film exhibited rough surfaces and loose structure of big particles, as shown in Figure 3.16 (a) and Figure 3.17 (a).

26

3.3.4 Effect of Substrate Temperature

The effect of various substrate temperatures heated in the range from room temperature to 300^oC, and thin films deposited when 0.8% oxygen was added to the discharge are shown in Figure 3.18. Not thin film formation, but polymerization of the HMDSO at low substrate temperatures was found, as shown in Figures 3.18 (a) and (b). There was no solid-state structure of thin film formation and a grease surface presented on the substrate. Figure 3.19 (a) shows the same situation when argon is mixed with 4% oxygen and added into the discharge.

In addition, it is interesting to compare the thickness of thin films at the various substrate temperatures.

The thickness of thin films deposited in the same treatment passes showed a strong negative dependence on the substrate temperature and many authors have discussed this situation using atmospheric pressure plasma [Babayan et al., 2001; Huang et al., 2009]. The sticking coefficient of the active molecule species formed from the plasma region decreased as substrate temperature increased. The ability of molecules to absorb the substrate reduced, but thermal diffusion of the molecules was enhanced at high substrate temperatures to diffuse into the suited sites of the thin film and caused dense film formation, the schematic diagram as shown in Figure 3.20.

27

3.3.5 Effect of Oxygen Concentration in the Discharge

The effect of film characteristics on deposition, using argon plasma mixed with various oxygen concentrations, is shown in Figure 3.21. The thickness and surface morphology of thin films deposited in the same treatment passes showed a strong positive dependence on oxygen concentration. The thickness of film deposited using pure argon plasma was hard to measure accurately. Thus the deposition rate was calculated to be 37.5 nm/min via increasing passes to 200. When oxygen added into argon plasma increased to 0.2 and 0.8%, the deposition rate was 200 and 275 nm/min, respectively. Surface roughness of deposited pure argon plasma was 1.38 nm. When oxygen introduced into argon plasma increased to 0.2 and 0.8%, the roughness was 5.26 and 29.8 nm, respectively. High mechanical strength with a smooth and transparent surface of SiO_x thin film can be formed using argon plasma mixed with oxygen below 0.8% at 300^oC substrate temperature. However, oxygen addition was more than 2 %, particles of a hundred nanometer scale were observed and films lost their transparency, powdery structure formed (see Figure 3.21 (d) to (f)). The possible results of reaction between the silicon-containing species and oxygen, and further cause nucleation in gas phase [Kasih et al., 2007].

Figure 3.22 shows the relationship between the deposition rate and oxygen concentration in the discharge at various substrate temperatures. Oxygen plays an important role in the low-pressure PECVD [Theil et al., 1994] and atmospheric-pressure PECVD system [Sawada

28

et al., 1995; Zhu et al., 2005] to SiO_x films. The oxygen radicals generated from argon mixed with oxygen at discharge, and excited oxygen promote decomposition of monomer molecules.

Therefore, the deposition rate increases by adding oxygen.

Further hardness testing briefly is presented in Table 8 under variation conditions. For the 200 ^oC substrate temperature, the lowest deposition rate (90 nm/min) of pure argon plasma has a rating of 5H in thickness of 100 and 300 nm. However, the film exhibited a lower pencil hardness of HB if oxygen added to 0.8%. For the 300 ^oC substrate temperature, the deposition rate was decreased from 1200 nm/min to 275 nm/min and the hardness evidently raised to 4H.