In-Situ X-Ray Reflectivity Study on Growth Dynamics of Sputter Deposited Gold on Silicon

  • PDF / 242,124 Bytes
  • 5 Pages / 420.48 x 639 pts Page_size
  • 99 Downloads / 158 Views

DOWNLOAD

REPORT


IN-SITU X-RAY REFLECTIVITY STUDY ON GROWTH DYNAMICS OF SPUTTER DEPOSITED GOLD ON SILICON R.P.CHIARELLO, H.K.KIM, T.ROBERTS, D.J.MILLER, R.T.KAMPWIRTH, K.E.GRAY, AND H.YOU Argonne National Laboratory, MSD, Argonne, Il 60439 ABSTRACT We performed in-situ x-ray reflectivity measurements to study the growth dynamics of gold sputter deposited onto silicon using an x-ray scattering chamber equipped with a faced magnetron source where the substrate is held at a right angle relative to the sputtering guns. By operating the guns at low power (I watt) and under 20 mTorr Ar, we could control the gold deposition rate to less than IA/sec. The observed x-ray reflectivity for gold deposited onto a silicon substrate at 300 K and 400 K is consistent with nucleated island growth for average gold particle sizes less than 50 A. Above 50 A, the reflectivity data indicates that the gold film uniformly covers the silicon surface, and that as the film thickness is increased the gold-vacuum interface gets progressively rougher. Detailed analysis of room temperature data is in progress, as is the temperature dependence on the roughness of the gold vacuum interface. INTRODUCTION The growth dynamics of thin solid films is sensitively dependent upon the deposition method and conditions. Thin solid film formation by particle deposition under nonequilibrium conditions has been an active area of theoretical and computer simulation research for over ten years[l], but only a limited number of experimental work has been done. Here we report preliminary results of the dependence of growth conditions on sputter deposited gold films on silicon studied by in-situ synchrotron x-ray reflectivity, a technique which is a sensitive structural probe on microscopic length scales. EXPERIMENTAL Synchrotron x-ray reflectivity measurements were performed on beamline X6B at the National Synchrotron Light Source, Brookhaven National Laboratory. The wavelength of the incident x-rays was 1.55 A. Two pairs of tantalum slits were used to define the unfocused incoming beam to 2mm by 2mm. An x-ray compatible sputtering chamber, whose details are presented elsewhere[ 2), was used with a faced magnetron sputtering system. The chamber was mounted on a Huber four circle diffractometer and had a base pressure of 5x10-5 torr. Research grade argon gas was used for sputtering, and an Ortec high voltage power supply, tunable from 0-300 volts at 0-20

Mat. Res. Soc. Symp. Proc. Vol. 237. 01992 Materials Research Society

412

mA, powered the sputtering guns. Silicon samples were ultrasonically cleaned for five minutes each in acetone, methyl alcohol, and deionized water. They were then baked in vacuum to 475 K in order to remove surface contaminants such as hydrocarbons and water. The deposition rate was 0.2-1 A/sec for data presented here as determined from the reflectivity data. The reflectivity scans were recorded after sputtering for a fixed time. RESULTS The x-ray reflectivity geometry is shown in figure 1 for xrays incident upon a film having index of refraction nj which was