Choose timezone
Your profile timezone:
Nb/Cu and Nb/Nb interfacial oxides
→
Europe/Zurich
Participants:
V. Petit, C. Pereirra, S. Leith, M. Himmerlich, G. Rosaz
Guillaume presented the overall objectives of studying the problematic of having interfacial oxides in Nb/Cu and Nb/Nb systems;
Nb/Cu: the presence of interfacial oxide between the Nb layer and Cu susbtrate is of crucial importance to understand:
- the origin of themo-currents building up during cavity cooldown when RF testing could be due to the absence of interfacial oxide with a large contact surface between Nb and Cu that have different Seebeck coefficients.
- The presence or not of interfacial oxide can completely change the growth mode of Nb onto Cu going from a fibrous to an heteroepitaxial mode.
- Heteroepitaxial growth has shown in the past to systematically lead to poorer performance wrt fibrous growth. We suspect that this could be due to defects propagation from the Cu substrate through the Nb layer.
Nb/Nb: this coating has an interest for quantum computing application. FNAL reported the presence of interfacial oxide that could be a problem when performing low temperature annealing as this oxide might spoil the film purity. They asked CERN if we could provide Nb coated cavities with a first step to remove the cavity native oxide.
We first addressed the Nb/Cu topic:
Marcel reported on depth resolved XPS analysis.
Sample : coated at -75V bias using our standard recipe. BO 200C 24h. Thickness ~150nm.
No oxygen is spotted at the Nb/Cu interface. MArcel recalled that the XPS sensitivity is about 0.1 at%.
Valentine reported about the XPS analysis of bare copper (EP 15min, Sulfamic rinse 1min, Ammonium Citrate passivation 5min) samples. She has shown that the surface is already exhibiting a strong Cu Auger peak suggesting that after chemistry the surface already has a strong metallic contribution.
Upon bakeout (200C, 24h, 50C/h), performing analysis at 100C, 150C and 200C she noticed a significant reduction of Cu oxide signal and by the time the 200C setpoint was reached the surface was mostly metallic. Keeping the sample at 200C for 24h did not modify further the Cu surface state.
Valentine noticed some phosphorus signal that she thinks can be attributed to pyro phosphates. Serge confirmed he performed a detergent cleaning after EP to get rid of the EP bath (very viscous) and this could explain the presence of phosphorus.
Guillaume pointed out that the oxide dissolution was previously observed by means of TDS analysis and was traced back to hydrogen driven reduction of the native oxide. He suggested to check if the EP step could provide the necessary amount of hydrogen to dissolve that same oxide.
Valentine will give back to Serge the already characterized samples for a new EP step followed by the same characterization process. The goal is to check if the results are reproducible with a treatment similar to a cavity coated multiple times.
We then discussed the Nb/Nb topic:
Marcel confirmed that the Nb sample coated at -75V exhibits oxygen signal at the layer/substrate interface and thus backed up the observation of FNAL.
He also performed the analysis of sample coated at -400V but he noticed an oxygen signal (from film native oxide) that does not vanish in the bulk of the layer but monotonically decreases. He points out that this is the typical behavior obtained on porous layers.
He suggested providing him a sample coated with a short -400V step followed by a standard -75V coating in order to mitigate the surface roughness problem.
The list of samples to provide/characterize is as following
There are minutes attached to this event.
Show them.
The agenda of this meeting is empty
