Nelson Voldeng Monmouth College Project Title: Sputtering of Alloy Thin Films Advisor: Prof. King

Introduction

Thin films have had many applications in technology namely electronics in which thin films are especially useful as resistors and semiconductors. Our research was aimed at sputtering a thin film alloy and studying its properties (such as composition, thermal properties etc…) to find out what type of applications it could be use in. Another goal we had in mind was for the alloy to have columnar single crystals from top to bottom on our substrate. A possible application in thin film alloys that have columnar single crystal could be for semiconductors. Possibly an alloy thin film semiconductor could have different areas on its surface that conduct electricity at different rates. Unfortunately unlike lattices and superlattices, alloy thin films are relatively new and do not have much prior information and research behind it. Hopefully the research we do will help the field develop into useful future research.

Objectives

To see if thin film alloys have any useful properties to electronics, insulators, etc.

Approach

• Our first goal was to sputter single element thin films. We sputtered Cu, Ag, Ti, Ni, Mo, and Fe. For each element we sputtered 6 films on glass substrates, each at different distances away from our targets and different times.
• After sputtering the films we had to measure the thickness as a function of substrate position.
• Using the measurements we can construct a sputter rate for each element. We would use this data in order to determine which elements would be good to alloy together i.e. if we wanted elements to have same concentration we could pick two elements with similar sputter rates.
• The next step would be to sputter an alloy film and measure its composition.
• From this point we want to test other properties of our thin film alloys such as electrical properties.

Collecting Data

Measuring our films proved to be the hardest and most time consuming area of our research. We tried using a profilometer which failed because it wasn’t precise enough and the measurements had far too much noise to be considered useful. We tried using the DIC (differential interference contrast) method using the ZEISS microscopes and again we could see a difference in height from the different colors, yet this method also was not nearly precise enough. DIC failed because the wave length of light was larger than our film thickness in some areas of our films. Our research came to a slow steady crawl once we started measuring our films on the AFM. We had to section all of our films into 16 pieces with a diamond scribe first. Then measure each piece individually. At first we couldn’t get any sufficient data because our films had a mask running down the middle and trying to measure from glass substrate to film gave bad results. The mask had deflected atoms from the film near the mask, so the film near the mask rose like a hill until it reached the actual height. The bad results were due to the scanning area of the AFM not being large enough to get over the mask to the actual high area of our film, which appeared to be in the middle of our film. Even when we maximized the scanning area, the scan would take 2 to 3 hours to get a photo, and it was bad data. To remedy this problem we developed the method of putting a scratch on the film in the middle and measuring from the scratch (which went down to the glass) up to the highest area of our film. After this step measuring the films was still not a fast process, but we started seeing good results.

Results

Unfortunately, the only good data we collected was from the Ag films because they don’t have strong adhesion to the substrate. When we attempted to scratch films such as Cu and Mo (which have strong adhesion to the substrate), the films would often elevate and shatter on the surface of our substrate. When measurements were taken for these films the data was very inaccurate.

Fortunately, the composition measurements taken on the SEM gave good data for what we expected to see with our Alloy film of AgCu. From our SEM measurements, Ag appears to sputter at a quicker rate than Cu.

Conclusion

Films with low surface tension like Ag are easy to scratch and collect feasible data for. However, in order to measure these other films such as Cu and Mo a new scratching technique or different means of a measuring device need to be applied. The composition measurements however are good and give accurate data for further research.

Figure 1: AFM Film Measurement

Figure 2:Composition Measurement of AgCu Thin Film