M. FOUAIDY Thin films applied to superconducting RF cavitiesLegnaro Oct.10, 2006 improved accuracy and sensitivity as compared to the usual RF method RS

M. FOUAIDY Thin films applied to superconducting RF cavities Legnaro Oct.10, 2006

An alternative device, for Nb/Cu samples RF properties characterization purpose was developed. The main feature of this technique, which is based on thermometry, is an improved accuracy and sensitivity as compared to the improved accuracy and sensitivity as compared to the usual RF methodusual RF method. The thermometric method, conjointly with a thermal model, is used for the measurement of the absolute RS distributionRS distribution on superconducting thin film samples. Precise calibration of test-samples RF losses is performed by means of a removable DC heater and temperature sensors pressed on the back of the disk and placed in a vacuum chamber. This new facility allows in-situ determination of all the thermal parameters involved in the model (substrate thermal conductivity and heat transfer coefficient at the solid-Lhe interface). The thermometric technique was first successfully validated and RF properties of several Nb/Cu sample was studiedvalidated and RF properties of several Nb/Cu sample was studied with this new device. Interesting data was obtained and analyzed. In particular, the effect of effect of the copper substrate surfacethe copper substrate surface conditions on the Nb/Cu sample RF propertiesNb/Cu sample RF properties was investigated and the corresponding results discussed.

Determination of Niobium films surface resistance by a calorimetric method

M. Fouaidy, IPN Orsay, France P. Bosland, M. Ribeaudeau, S. Chel, J.P. Charrier, CEA Saclay, France


TOPICS

Motivation for developing such an instrument

Purpose

Main advantages of the calorimetric method

Method principle and thermal modelling

Thermometric system

Measured versus simulated temperature profiles

Sensitivity and accuracy of the calorimetric

method

Validation of the calorimetric method

Test results with sputtered niobium films


Why did we need to develop a new instrument for measuring the

RF surface resistance (Rs) of sputtered superconducting films with

such SRF cavity ?

Improve accuracy and sensitivity of Rs measurement,

Lack of accuracy and sensitivity at 4.2 K for measurements

performed by the end plate replacement method !

Measure exclusively the test-sample RF losses by excluding

any

extra RF losses :

Some of extra RF losses are inherent to the ‘classical’ method

(rest of the cavity, indium gasket, RF coupling loops)

Potentially, anomalous RF losses induced by Field Emitted

electron impacting area other than the sample

Motivations for developing such an instrument


Improve accuracy, reliability and sensitivity of Rs

measurement

Thorough and precise RF characterization of sputtered Nb

and NbTiN films onto Copper substrate

Study the effect of sputtering process parameters and

substrate surface preparation on the films RF properties

Improve SRF performance and master the technology

Investigate Rs(T) in the temperature range: 1.6 K - 4.5 K

Study Rs spatial distribution on the sample .

Progress in the understanding of SRF properties and get

more insight into superconducting film physics and develop new

superconducting material interesting for accelerators

Purpose


Absolute, direct and local method as compared to the usual

RF technique

No reference disk needed

Save time,

No assumption concerning the rest of the Niobium

cavity RF surface

Vacuum insulation and hence a precise temperature

measurement (thermometers in contact with a non-wetted solid

wall)

In-situ measurement of substrate thermal parameters

Main advantages of this method


R=0Rcav=55

Rseal=58

qHF

Nb Cavity RF part

LHe

Thermometric part

qSTAT

Vacuum Heater

Method principle and thermal modelling

Sample


Simulation parameters:-Tbath=1.715 K

-experimental copper thermal conductivity

-kapitza conductance at sample -LHe interface:hK=4000W/m2.K

0

20

40

60

80

100

120

140

0 5 10 15 20 25 30 35 40 45 50 55

r (mm)

T

(mK

)

Pstat

Prf

P R H H S dS

HH

J krJ kr

P H aR bR H cR H

RF s s sS

SSMAX

RF SMAX SMAX SMAX

1

22

11

20 1 2

2

( ) * ( )

( )( )

( )

max

kp

rcav 11

No RF dissipation at the indium seal 2nd order polynomial law parametrization of Rs=f(Hs) Total dissipated RF power PRF depending on the surface magnetic field Hs

p11:J1‘s first zero

For radius r>40mm

PSTAT=PRF TSTAT= T RF

Determination of RS coefficients

(a, b, c) by least square method


Dismountable assembly of 24 Surface thermometers in a vacuum chamber

Four subsets at 90° apart at 6 radial positions from 12.4mm up to 47.4mm (Step r=7mm) Calibration heater (12 OFHC rod) located at the centre of the test-sample Calibrated thermometer (1.5K-60K) placed near the heater/sample boundary control of the heater temperature and determination of the heat leaks Two reference thermometers (Calibrated Germanium and Carbon resistor)

Accurate measurements of Tbath during thermometers

calibration (R vs T curve) and during

measurements sequences (T vs qSTAT and T vs qHF)

Thermometric system (1)


Bulk Niobium cavity : TE011 mode f=4GHz TE012 mode f=5.6GHz Calibration heater 24 thermometersVacuum chamber Heater thermometer (Heat leaks)



Sample

Thermometers

Heater



Measured and simulated temperature profiles

QSTAT

(mW)

Type T12

(mK)

T11

(mK)

T10

(mK)

T 9

(mK)

T 8

(mK)

220 Computed 36.5 29.4 24.4 20.7 17.6

220 Measured 36.8 29.2 25.3 20.9 17.5

1990 Computed 320 260.3 218.7 186.6 160.5

1990 Measured 323 254.6 228.4 187.8 159.2

Hs(A/mm)

Type T12(mK)

T11(mK)

T10(mK)

T 9(mK)

T 8(mK)

14 Computed 25.3 24.8 23.6 21.7 19.2

14 Measured 21.1 20.4 20.1 18.7 18.5

33 Computed 209 205.4 195.9 179.6 159

33 Measured 200.1 194.1 188.3 173.1 162.9


Minimum detectable heating : ~0.1mK at T=1.7K and T=4.2K

Accuracy: calorimetric versus RF method at f=4 GHz

Sensitivity and accuracy of the method

T=1.7K T=4.2K

T=1.7K T=4.2K

The accuracy of calorimetric method is ~5 times better than RF method


Validation of the calorimetric method by comparison with RF measurement (1)

0

50

100

150

200

250

300

350

400

450

0 50 100 150 200Hs(Oe)

RsTh RsHF

5000

6000

7000

8000

9000

10000

11000

12000

13000

0 20 40 60 80 100 120Hs (Oe)

Rs(n

)

RsTh

RsHF

f=5.6 GHz

f=4 GHz

f=4 GHz

f=5.6 GHz

T=1.7KT=4.2K

Tests of a Bulk niobium sample (Solid dots: usual RF method, Solid line : calorimetric method)

Good agreement between the two methodsFor bulk niobium the field is limited by RF heating (Disk cooled by liquid helium at the lateral rim only)

Bs(Oe) Bs(Oe)

RS(n) RS(n)


0

50

100

150

200

0 10 20 30 40Bs (mT)

Rs(

n)

HFHF refroidissement latéralTH refroidissement latéral

5000

6000

7000

8000

9000

0 5 10 15Bs (mT)R

s(n

)

Validation of the calorimetric method by comparison with RF measurement (2)Tests of a Bulk niobium sample (dots: usual RF method, Solid line : calorimetric method)

T=1.7KT=4.2K

f=4 GHzf=4 GHz

Rim cooling

Disk cooling

For bulk niobium, the cooling conditions have a strong effect on the maximum RF field achieved and on surface resistance at high field (Joule heating)


Test of a niobium film sputtered onto a copper substrate at T=1.7K

0200

400600

8001000

12001400

1600

0 10 20 30 40Bs (mT)

Rs(

n)

HFHF refroidissement latéralTh refroidissement latéral

T=1.7K

f=5.6 GHz

Good agreement between the two methods: for six tests performed at 1.7 K the difference is 15%-20%For sputtered niobium films the field is not limited by RF heating Efficient conduction cooling by copper substrate (Disk cooled by liquid helium at the lateral rim only)


Field Limitations For Bs<5mT: thermal instabilities due to switching from natural convection to nucleate boiling (Not hard limit!) For Bs>15mT: power limitation due to dissipations in the cylindrical part of the cavity (bulk Nb)

Test of a niobium film sputtered onto a copper substrate at T=4.2K

RS(n)

Bs(mT)


Effect of substrate roughness on surface resistance at 1.7 K (1)

Copper substrate

Niobium film


<Rs>(B) par méthode thermométriqueMode Te011 Tb=1,7K

0

500

1000

1500

2000

2500

3000

0 5 10 15 20 25 30 35 40B(mT)

Rs(

n

)

d57

d58

d60

d62

Rs vs B measurement of Nb Films on Cu substrate (T=1.7K, f=4GHz)

Nb film residual surface resistance increases with the substrate roughness and defects density Use clean and smooth substrate with intermediate layer for of lattice matching and improve superconducting properties

Bs(mT) Bs(mT)

RS(n)

RS(n)


Rs vs B measurement of Nb Films on Cu substrate (T=1.7K, f=5.6 GHz)

<Rs>(B) par méthode thermométrique et hf Mode Te012 Tb=1,7K

0

1000

2000

3000

4000

5000

6000

0 5 10 15 20 25 30 35 40B(mT)

Rs(n

)

d57

d58

d60

d64

RS(n)

Bs(mT)

Bs(mT)

RS(n)

Nb film residual surface resistance increases with the substrate roughness and defects density Same effect observed at 4 GHz and 5.6 GHz

Documents

M. FOUAIDY Thin films applied to superconducting RF cavitiesLegnaro Oct.10, 2006 improved accuracy and sensitivity as compared to the usual RF method RS