Water use Efficiency in Immersion Wafer Rinsing
- PDF / 372,830 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 40 Downloads / 180 Views
527 Mat. Res. Soc. Symp. Proc. Vol. 477 ©1997 Materials Research Society
Experimental Procedure The experiments were carried out by adjusting the flow rate of the water going into each rinse bath (the primary bath and the final rinse bath) and adjusting the flow program (percentage and period of time on and off) of the primary rinse bath only. The water was provided by the MOS 12 UPDI plant and its resistivity was measured to be greater than 18 Mohm at the plant. The water in both baths was allowed to flow until it reached the highest resistivity possible, and then the recipe was begun to treat the wafers with the chemical of interest, 30 sec. for our 10:1 HF bath. The wafers were then moved robotically to the primary rinse bath and treated with the rinse process of interest. After the rinse the wafers were moved robotically to the final rinse bath and the resistivity recorded in this second bath to determine the amount of chemical remaining on the wafers after the first rinse. The final rinse bath contained a megasonic transducer to assure efficient rinsing. Analysis of Data The resistivity readings of the water in the final rinse bath were converted to conductivity readings and integrated over the time of the rinse. The water in this rinse bath was always allowed to return to its maximum value during the experiment so we are concerned only with the change in resistivity during the experiment. The integrated conductivity values were converted to the concentration of the ions in solution. To confirm the validity of the experiments performed, in one experiment, known amounts of HF solution were micro-pipetted directly from the process tank into the second rinse tank, in the absence of wafers. The resistivity in this tank was then followed, and the usual calculations for integrated conductivity performed. The mg carryover as calculated from the integrated conductivity was compared to the mg acid added based on the volume pipetted into the tank and the concentration of the chemical solution. The actual amount of HF pipetted into the tank for this experiment was 2.5 mg. The calculated amount based on the integrated conductivity measurement was 1.7 mg. Similar results were obtained for a sulfuric acid experiment. EXPERIMENTAL RESULTS These experiments were done on an HF rinse process which is perhaps the simplest process to describe since it is an overflow process. Due to the hydrophobic nature of the wafers, dump rinses are not done and they were not attempted in these experiments. As mentioned in the experimental section, and unlike previously published results 2 -4 , we used the results from the final rinse to monitor the success of the initial rinse. This was done to determine how the various rinse parameters in the normal rinse would affect the amount of chemical carried out, for example, into the second process chemical in which the wafers would be treated. These results thus allow us to calculate how much HF would be brought into, for example, an SC1 tank when the proposed rinse is used. Effect of Rinse Volume Th
Data Loading...
