Role of Crystallinity In Roughness of Ru/c Multilayers -An Amorphization Study
- PDF / 1,534,162 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 51 Downloads / 211 Views
Mat. Res. Soc. Symp. Proc. Vol. 382 ©1995 Materials Research Society
homogeneous mixture, and that a small percentage of oxygen in the film would suffice to intrude on the close packing of the ruthenium atoms and leave an amorphous or highly-disordered structure. PROCEDURE Ruthenium single layers and Ru/C multilayers were deposited by dc (for Ru) and rf (for C) magnetron sputtering on Si (100) and (111) substrates. No attempt was made to remove the native oxide from the substrates before deposition. Deposition was carried out in Ar/0 2 atmospheres at 2.5mtorr total pressure, with gas composition monitored by an MKS quadrupole residual-gas analyzer. The sputtering gas was premixed separately, and its composition monitored for 10 minutes as the magnetron system stabilized before depositing material on the substrates. Film thickness was controlled by prior rate tests at various oxygen partial pressures. Uniformity of the d-spacing was assured by some spot checks of the rate before and after a long run. After each sputtering run with oxygen the guns were run for 10 minutes in pure argon while monitoring the chamber gas composition, in order to be sure there was no lasting oxidation of the targets. After sputtering the wafers were removed and stored in air. Ruthenium single layers were about 200A thick, and the Ru/C multilayers had d-spacing 52A and 50 layer-pairs. Before sputtering the samples for measurement, we performed a series of rate tests of single Ru and C layers under various Ar/0 2 atmospheres. Many samples were made at low oxygen content and their peak reflectivities measured by hard x-ray diffraction. There was no effect on the hard x-ray reflectivity for 0% to I% oxygen in the sputtering gas, so progressively more oxygen was added until the reflectivity began to drop. Hard x-ray studies were done on a Rigaku rotating-anode diffractometer, operating at 20kV and 200mA. The beam was formed by a graphite monochromator and had a width of 3mm at the sample. The diffracted signal was measured unattenuated by a Bicron P-12 scintillation counter. 8 Soft x-ray studies were done on a laser-plasma reflectometer at the Center for X-Ray Optics. This device uses a Continuum 661-10 Nd-YAG laser at 532nm to excite a target rod of copper on aluminum to produce a plasma emitting in the soft x-ray. The analytical beam was selected by a 50 line/mm monochromator and passed through a B filter before striking the sample. The reflected intensity was measured by an avalanche photodiode operating at 30V bias. TEM plan-view specimens were also 52A d-spacing but 10 layer pairs, and were prepared by sputtering on a 400-mesh copper grid bonded to a glass slide by Crystalbond hard wax. After sputtering the wax was dissolved with acetone and the grids and thin films collected. The films were examined without further treatment in a Topcon 002B microscope operating at 200kV. RESULTS Soft x-ray reflectivities of single Ru films, for incident angle 0' < 0 < 30', are shown in Fig. 1. In the accompanying table are the oxygen contents of the s
Data Loading...
