A study of plasma etching for use on active metals

Abstract

Active metals can be used as a getter pump, removing impurities in ultra-pure high vacuum environments. To relieve the difficulties involved with the transportation, storage and handling of these metals, a process is being developed to create a protective coating by removing the active metal component from the surface of a mixture of the active metal and a stable element via immersion in a wet chemical bath. It is the purpose of this thesis to investigate the utility of a plasma etching process in the removal of the active metal. A decision tree for the selection of a suitable etchant gas was developed and experiments were conducted to validate the process. Magnesium coated glass slides were etched with chlorine, fluorine, and fluorocarbon gasses alone and in the presence of argon to determine which chemistry would etch the fastest. Magnesium was chosen because it was the easiest to handle among the active metals. It was determined that chlorine gas with argon provided the highest etch rate. Next, a set of factorial experiments were conducted to determine the sensitivity of the etch process to changing radio frequency power input, chamber pressure, and the ratio of chlorine to argon.

(cont.) Factor levels were arranged in an orthogonal array and the main effects of each parameter were estimated using analysis of means. It was found that the etch rate increases strongly with increasing power and decreasing amounts of chlorine. Dependence on pressure was less pronounced. These results indicate that the etching mechanism is dependent on physical sputtering. Etch rates of up to 30 nm/min were achieved at 350 W power, 20 mTorr chamber pressure, and a 10%/90% chlorine to argon ratio. In conclusion, although the etch rate is not currently high enough for practical use, the steps in the decision tree were validated by the first set of experiments while the second set was able to characterize the basic mechanism of the etch process.