CR visio-conférences

CR visio-conférencesEdit

2012-05-30

INVOLVED
Helene le Sueur  / CSNSM
Stefanos Marnieros / CSNSM
? Louis Dumoulin / CSNSM
Jochem Baselmans / SRON
Pascale Diener / LPS (CSNSM next year)

GOAL
? (minimal goal) get the dark and optical properties for at least one NbSi film

FIRST FILM - STOECHIO AND THICKNESS
Final choice: NbSi 21% rho=570μΩcm Tc=1.49K t=200nm Lk=25.1pH
? More films? Maybe 1 or 2 other films having a different thickness or stoechio? 
Note: need of Lk lower than 20pH for design constraint. 
Possibility at the csnsm to try NbSi with higher Nb content (above 21%), having probably 
a lower Lk (probably not much different resistivity and higher Tc) but needs few months
development. 
Exists one well caracterized csnsm stoechio is NbSi 18%, giving Lk=50pH for t=200nm.

MASK
For wafer 2 inch
Ordered by csnsm
? Designs on it: To be chosen. Proposition from H and P that J choose what to put on the mask

DEPOSITION AND LITHO 
By Helene
(?) Long time knowledge from Louis and Stefanos about NbSi films
Wafer 2 inch Si high resistivity (>1kΩcm) 
No HF substrate cleaning (not possible presently at CSNSM deposition room)
No SiO top layer (risk to introduce more TLS -> x dB noise)

KID MEASUREMENTS
By Jochem
(High) possibility that H and P come from the first test

PLANNING
-Counting 1 week to choose the dc dark opt hybrid designs, 2 weeks to wait for the mask 
when ordered, 2 weeks to make the hybrid film(s), first measurements may be done in mid july
-Helene is (almost) absent between august 2012 and February 2013 
-H&P to J: are you available one week between July 16 and August 3 for the first try?

2012-06-01

INVOLVED: Pascale Diener, Jochem Baselmans, hls

Pascale: Higher Nb content (above 11-12%) you are close to BCS behavior.

precision LN: 
The S-Metal-T is around xNb=11-12% for thick films (>50nm) -> it is not BCS here.
But, at Nb contents around 18%, the gap has been measured with STM spectroscopy, 
and found close to BCS, with as well the delta(Tc) behavior being close to 1.76. 

This is sensible since at 18% we are far from the SIT (SMT), and we anticipate that 
for 21% (the actual target) NbSi will most likely also behave like BCS 
(though not confirmed with STM)

Film parameters 21 % Nb, 89% Si
- 26 pH
- 1.5 K
- rho=570 muOhm cm
- thickness = 200 nm
- sheet impedance 28.5 Ohm
+> same film properties as TiN from JPL wrt KID and radiation absorption.

actions:
mask Jochem B.
- 2inch mask design with 3 inch size for later.
- use same designs as we have for TiN with absorber (target = Lk ~ 25pH, sheet imp ~ 30ohm)
- we use KID duo all 3 inductances (0, 10pH, 100pH for 'dark design' = without absorber)
achat wafers Hls (ask 'university wafers')

- first experiments in the last week july 23-27 together with Helene and Pascale
No SiO coating, no HF, 
substrate: 350 micron Si prefereable, orientation<100>, 10 kOhm cm.   

- Helen wants to measure close to the SIT; More experiments in Utrecht possible 
on this but no appointment made for now.

2013-03-29

IMPLIQUES
Helene le Sueur  / CSNSM
Stefanos Marnieros / CSNSM
Louis Dumoulin / CSNSM
Alessandro Montfardini / NEEL

Design envoyé par mail le 20/03/2013, comporte 9 échantillons sur un wafer 4 pouces. 
Les plus importants sont les 5 de la croix au centre.
Les motifs comportent beaucoup d'inductance et peu de capa. 
L'absorption se fait au niveau de l'inductance (densité de courant non nulle, courant 
constant sur tout le motif - taille petite devant longueur d'onde)
Les motifs d'inductance sont variés, orientés soit dans une direction, pour privilégier une
polarisation, soit dans toutes les directions.
Les résonances sont dessinées pour 200MHz.
La ligne de transmission est en NbSi sur le design, ce qui peut poser pb pour adaptation 
50Ohms -> prévoit de refaire en 2 étapes, 1 masque Alu, 1 masque NbSi.
Design fait par Nello (thésard)
Echantillons futurs à réaliser par CSNSM. 
Mesure à Néel, mais OK pour que l'un de nous vienne mesurer.
Si fonctionne avec NbSi, pour NIKA aura besoin de beaucoup de couches pour mise au point

Mesures déjà effectuées sur NbSi 18%, 50nm. 
Trouvé Tc = 1K, R carré = 160 Ohms -> Lk carré = 220pH (?? D'après Alessandro, 
ce serait ~20pH??)

Intérêt de Néel:
- Ampli paramétrique: trouve intéressant mais n'a pas forcément le manpower. Il y a un ingé
    micro-ondes qui est sur le coup (Christophe Poireau)
- Composants optiques modulables: faire des filtres / polariseurs en NbSi
    (alternances de grilles en NbSi et d'isolants)
- KIDs: peut-on avoir une Tc de ~1K avec une forte résistivité pour avoir une fréquence 
    de coupure de 90GHz? 
    (NB: Al 15nm, gap augmente quand diminue épaisseur. Coupure actuelle @15 nm = 110GHz)

Problème entrevu avec NbSi, idem TiN:
Sur manips effectuées, voit conduction sous le gap, montre un matériau inhomogène
Mesures DC: voit transition raide car dominé par chemin de moindre résistance
Mesures RF: sonde les propriétés globales de l'échantillon -> gêné par inhomogénéités

Mesures prévues:
couches NbSi 18%, 50nm +/-10
wafer Si (100), rho>17.5 - 20kOhms.com, both side polished, 350µm, 3" 
t=270-300µm -> pic d'absorption à 150GHZ

Relation avec SRON:
Montré courbes obtenues à Jochem. Ennuyé avec collaboration car SRON pas clean.
Plutôt compétition, mais ne prend pas ombrage que CSNSM fournisse aux deux labos.

2013-11-15

INVOLVED
Helene le Sueur  / CSNSM
Stefanos Marnieros / CSNSM
Louis Dumoulin / CSNSM
Jochem Baselmans / SRON
Juan Bueno / SRON

Discussion about the data taken in july on the first NbSi samples (200nm, 21%)

1.- Discussion on NbSi dark and optical experiments - PPT: NbSi_data_analysis_v0

- Dark noise levels seems reasonable - scattering due to the different Q’s of the design
- Possible ways of lowering the noise :
	 - HF cleaning of substrate prior to NbSi deposition.
	- Think about RIE process afterwards to remove resist contaminants and oxide layer 
	(rejuvenation process developped by Gregoire Ithier during his PhD thesis on 
	Nb resonators).. 
	Raises a problem of defects caused by RIE, which might in fact increase TLS. 
	Private communication (Louis Dumoulin): ion etching on Ge below 70V does not 
	induce defects. Possibility to improve the RIE recipe ?
- QP lifetime very short - should be higher
- QP lifetime does not seem to follow Kaplan theory
- responsivity increases with optical loading, and anyhow much lower than expected (10%)
 this is observed as well in TiN
- Two different materials, two different deposition techniques 
   ==> same optical loading dependent responsivity
- Not a material property but a more general observation?
- Possibility of doing FTS measurement at Cardiff (does NbSi absorb light?)
- Include Teun Klapwijk and Eduard Driessen in future discussions. 
(may also include Philippe Joyez who works on QPS in NbSi with Helene).

2.- Actions

- Juan: Replot qp lifetime for each resonator (non-Kaplan check)
- Juan: Look at the design and include overview of the Q factors in the presentation
- Juan: Check noise levels with Julie Gold data
- Juan: Look at resonance frequency vs temperature data and fitted with Mattis-Bardeen
- Juan: Check dimensions for possible FTS measurement at Cardiff
- Juan: Email to Teun and Eduard with the updated presentatio
- Jochem : send reference about RIE induced defects and TLS.
- Helene or Louis : check for reference about voltage threshold for the induction of defects.

2013-11

NEEL, first measurements on Nisba v1, 1.8
- resonances around 600MHz, as expected. Line is 50 ohm.
- Total resonance band is too small (only 12MHz for 49 resonators)
Rays are not well separated. But some allow good measurement.
- quality factors are not good between 3000 et 5000.
Maybe this is only parasitic ligth in setup
- le "power handling" is good -> does not reach non linearity too quick 
Can polarize it with power comparable to previous matrices, and increase S/N
Q -> WHAT POWER?
- Do not have excess phase noise like in SRON matrices
At worse, this is factor 2-3 higher than the best matrices.
around 2-3Hz/sqrt(Hz) for frequency noise between 1 and 10Hz (OK for astronomical
 integration times)
- Can see cosmic rays with fast electronics.
Inferior limit to relaxation time: 5-10µs.

Next measurements on Nisba v1, 1.2, in better fridge
- now quality factors are good, around 1e5 !

Veulent une nouvelle couche avec Tc ~ 700mK, et même Lk