Classification of the Modes of Dissociation in Immiscible Cu-Alloy Thin Films
- PDF / 2,004,044 Bytes
- 6 Pages / 417.6 x 639 pts Page_size
- 60 Downloads / 142 Views
1
Mat. Res. Soc. Symp. Proc. Vol. 564 © 1999 Materials Research Society
The electrical resistance of the samples was measured in situ in a purified N 2 atmosphere at a series of heating rates between 0.17 and 6°C/s (10-360°C/min). The maximum temperature of heating was 950'C, except for Cu(Ag) samples where this temperature was limited to 750'C to avoid exceeding the eutectic at 780TC. The room-temperature sheet resistance of the films, before and after these experiments, was measured using a four-point probe. In situ wafer curvature measurements, at a heating rate of 10C/min, were used to monitor the changes in film stress. However, since the initial curvature of the substrate was not measured, only relative stress changes, and not the absolute value of stress, were determined. X-ray diffraction (XRD) experiments were conducted in situ on the NSLS IBMIMIT beamline X20C at Brookhaven National Laboratory. The beamline was equipped with a multilayer monochromator providing energy resolution of 1.5% at 6.9 keV (0.1797 nm) with an average intensity of 3 x 1012 photons/s. For constant-heating-rate experiments, samples were annealed at 3°C/s in purified He to the desired maximum temperature. Isothermal anneals were also carried out for selected samples that were heated to the anneal temperature at 35°C/s. The diffraction information was acquired using a linear position sensitive detector that enabled the monitoring of
an angular window of-10', which was typically chosen to include the Cu( 111) peak at 51'. The temperature was calibrated using a series of eutectic melting points for an accuracy of ±3'C. The samples for transmission electron microscopy (TEM) studies were free-standing films that were heated at 40*C/min to a given maximum temperature using a differential scanning calorimeter furnace. The samples were thinned to electron transparency in a Gatan 691 PIPS and their microstructure was studied using a Philips EM400T operating at 120 kV. RESULTS AND DISCUSSION The variation of resistance with temperature at a heating rate of 0.83°C/s (50*C/min) is shown in Figs. la-c for Cu(Ag)/Ta, Cu(Fe) and Cu(Ta) films, respectively. The Cu(Ag)/Ta film showed a single abrupt decrease in resistance, while the Cu(Fe) film exhibited four resistance drops widely separated in temperature, with the second drop being very small. (At heating rates of 1.5°C/s and higher, the final resistance decrease for the Cu(Fe) film was replaced by a resistance increase.) In contrast to Cu(Fe), Cu(Ta) displayed three distinct resistance drops each immediately following the other. The other films with alloying elements from groups VB and VIB (namely V, Nb, Cr, and Mo), as well as the Cu(Ru) film, showed similar resistance-temperature behavior to that for Cu(Ta). The before(after) room-temperature resistivities of the films depicted in Figs. la-c were respectively 5.2(2.5), 21.8(3.1), and 41.8(4.2) liicm. Thus, annealing clearly resulted in significant reductions in resistivity, but for none of the films studied did the resistivity reduce to that f
Data Loading...
