Specific Process Knowledge/Thin film deposition/Deposition of Gold/Adhesion layers: Difference between revisions
| Line 141: | Line 141: | ||
== Adhesion layer effect on bilayer thin-film electrical resistivity == | == Adhesion layer effect on bilayer thin-film electrical resistivity == | ||
The change in nanostructure of the Au thin �lm due to the presence of the | |||
adhesion layers observed above, could have an important impact on the �lm | |||
macroscopic properties such as electrical resistivity. Electrical resistivity | |||
in polycrystalline �lms is dependent on electron scattering at surfaces and | |||
grain boundaries, and it was expected that the grain size change measured | |||
with TKD to be re | |||
ected in the electrical properties. Furthermore, for the | |||
case of Cr, a Cr-Au alloy was formed, which also was expected to have an | |||
impact on the thin-�lm electrical resistivity. | |||
To verify these hypotheses, the sheet resistance (R) of the three samples | |||
(20-Au, 2-Ti/20-Au, and 2-Cr/20-Au) investigated with TKD above | |||
was measured using micro 4-point probe (�4PP). Fig. 5.14a shows that | |||
the Ti/Au sample had a lower sheet resistance than pure Au, which can | |||
be attributed to the two layers acting as parallel resistors [124]. Data normalization | |||
was performed with respect to the average sheet resistance (R) | |||
measured at the 20-Au sample. To exclude tip wear e�ects of the �4PP, the | |||
measurements were performed measuring with the same probe alternatively | |||
on the 20-Au and 2-Ti/20-Au sample, respectively. | |||
Moreover, to rule out thin-�lm thickness variation e�ects, Au TEM crosssection | |||
thickness measurements were performed on 36 points along the 20- | |||
Au and the 2-Ti/20-Au samples. They revealed a slightly thicker Au �lm | |||
thickness in the Ti/Au sample compared to the pure Au �lm (23.6 ±0.5 | |||
nm vs 21.9 ±0.5 nm, respectively). Since electrical resistivity is inversely | |||
proportional to �lm thickness, the thicker Au �lm contributes to decrease | |||
the sheet resistance in the Ti/Au sample with respect to pure Au. | |||
Fig. 5.14b shows that the Cr/Au system had a higher sheet resistance | |||
than pure Au. In this case, the formation of a single layer due to Cr-Au | |||
inter-di�usion compromised the multilayer assumption. The sheet resistance | |||
increase is in line with the general resistivity increase of Cr-Au alloys, which | |||
increases linearly with the Cr concentration [125]. Data acquisition and normalization | |||
were done as in the Ti/Au case above, and TEM cross-section | |||
thickness measurements of the Au layer were done on 29 points of the Cr/Au | |||
sample, giving a mean thickness of 22.7 ±0.4 nm. Compared with the thickness | |||
of pure Au, this value is slightly higher. However, the sheet resistance | |||
decrease due to this thickness di�erence was not enough to compensate the | |||
increase due to the Cr-Au alloy formation. | |||
The Ti/Au parallel behavior and Cr/Au inter-di�usion seem to have a | |||
larger impact on the electrical properties of the multilayer systems than the | |||
nanostructure change observed by TKD. For both samples, the increase of | |||
grain boundary scattering due to the higher density of grain boundaries, | |||
compared to pure Au, could not be measured with setup used, but cannot | |||
be excluded a priori. | |||
= Adhesion layer model = | = Adhesion layer model = | ||