Evaluation of Reaction Dynamics of Film Depositions in Plasma CVDs by Using a Remote Plasma CVD System

  • PDF / 328,804 Bytes
  • 6 Pages / 414.72 x 648 pts Page_size
  • 35 Downloads / 167 Views

DOWNLOAD

REPORT


the film deposition region separated with space. The advantages of the use of remote plasma are 1. number of the types of active species with resect to that of direct plasma CVD is substantially less, 2. the source compound can be activated either by excited atoms, radicals or ultraviolet photons, 3. deposition takes place in an electron- and ion-free environment, hence the substrate and the growing film do not get damaged by the bombarding charged particles that often occur in direct plasma CVD, 4. the chemistry of the deposition can be predicted and controlled. As the source materials, three carbosilanes namely, tetramethylsilane (TMS), hexamethyldisilane (HMDS) and tetrakis(trimethylsilyl)silane (TMSS) were selected considering their non-pyrophoric and less-hazardous nature. Use of these source monomers eliminates the necessity of carefully design tubing system with expensive leak detectors which essentially needs when silane is used. The other advantage of these monomers is that the monomer itself is a carbon source as well, therefore, the use of a secondary gas as the carbon source is not required. EXPERIMENTAL Figure 1 shows the experimental set up used for the film deposition. The monomers, TMS ard HMDS are liquid at room temperature, therefore, a bubbler with a 10 seem He flow was used in obtaining the gas-phase of these sources. TMSS is a solid (white powder) at room 145 Mat. Res. Soc. Symp. Proc. Vol. 406 01996 Materials Research Society

temperature. This powder was placed in a glass tube and heated to 70'C. The sublimated TMSS vapor was flushed in to the reaction chamber by a 10 sccm He flow. The monomer + He gas mixture was introduced into the reaction tube through the downstream inlet (see Fig. 1(a)). Hydrogen was used as the upstream gas and plasma was produced by applying 100 W rf power. The pressure inside the flow tube was maintained at 0.20 Torr for all the depositions. c-Si and quartz glass were used as the substrates. Depositions were carried out at different substrate temperatures ranging from room temperature to 400 'C.

Monomer gas

(a)

MHz)

Monomer gas

Heab Thern

(b) gas Vacuum pump Monomer

(c) Downstream gas

Figure 1. A schematic diagram of the CVD reactor with different configurations of the afterglow

tube: (a) straight, (b) with a light trap and (c) with a radical trap.

146

During the deposition of a-SiC:H films using a straight flow tube as shown in Fig.l(a), both H radicals and UV radiation reach the reaction site. Therefore, it is not clear whether UV radiation or H radicals initiate the deposition process. To clarify this point, a flow tube with two bends was used for several depositions (see Fig. 1(b)). With the use of this bent tube, UV radiation could be cut off from the reaction site. Nevertheless, hydrogen radicals generated in the plasma flow to the reaction site. The results of this experiment explains the reactivity between H and the source monomer. In order to determine the reactivity between the source monomer and UV radiation, radicals reaching the reaction site