oleg malishev development of thin films for superconducting rf cavities in as te-c

Development of thin films for

superconducting RF cavities in ASTeC

O.B. Malyshev and R. Valizadeh

on behalf of collaboration team

ASTeC Vacuum Science Group,

STFC Daresbury Laboratory, UK

O.B. MalyshevThin films and new ideas for SRF,

6-8 October 2014, Legnaro, Italy 2

• Motivation

• PVD deposition

• CVD/ALD deposition

• Surface analysis

• Superconductivity evaluation

• RF sample test

• Future plans

• Conclusions

Outlook



Motivation

• The aim is to develop the PVD and CVD coating

technologies of superconducting materials (including high

temperature ones) for RF cavities and apply it on the RF

cavities

• Objectives:

• a systematic study correlation between

• Deposition condition

• Film morphology, structure, chemistry

• AC and DC superconductivity characteristics such as Tc, Bc,

RRR, etc.

• RF evaluation of samples

• Cavity deposition and test



UHV PVD facility• Bakeable

• Load-lock chamber

• 100 mm diam.

• Three planar

concentric targets

with the variable

distance to the

substrate: 10-15 cm

• Substrate rotation

• Ion beam assist

• 20 Ts 950 ºC

• Differential RGA

pumping to analyse

the sputter gas.



• Base pressure = 10-9 mbar

• Kr pressure during sputtering

• P = 10-3 mbar

• Power at magnetron:100-600 W

• DC sputtering

• Pulsed DC sputtering 100-350 kHz

with an off time of 1.1 µs.

• HiPIMS

PVD deposition

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0

50

100

150

200

250

300

350

400

450

500

100 200 300 400 500 600

Cu

rren

t (A

)

Vo

ltag

e (V

)

Power (W)

Voltage

Current

• On a substrate:

• DC and RF Bias on a substrate:

• A bias voltage to the substrate was varied 0 - 150 V.

• Ion beam assisted



Sample insertion system

- MgO, Si and Cu substrates deposited together.

- Different substrates can be compared when

deposited with same plasma in same conditions.

Up to 6 samples load



• Base pressure:

• 1.5 x 10-5 mbar at 120 ºC

• Gas flows:

• Argon, Max 5 l/min, 200

sccm MFC

• Hydrogen, Max 1 l/min, 100

sccm MFC

• Heater tested up to 500 ºC

(could go 950 ºC)

PECVD/ALD deposition

• First Nb film using NbCl5 precursor was deposited

on last Friday



ALD rig

ALD valves:

tested with

Arduino control,

switching <1 ms

Plasma

waveguide

Reactor

chamber

Sample holder

SAES gas purifiers



• XRD analysis

• average grain size and lattice orientations within the film.

• XPS analysis

• film composition and impurity

• EBSD analysis

• an accurate value for the grain size at the surface of the

film.

• SEM analysis

• to determine the film thickness and growth rates.

• to give an indication of the type of film that has been

deposited i.e., columnar with voids or densely packed

grains.

Film Morphology



Multi-Probe UHV XPS, AES, AFM, STM, LEED and ISS



• Nb grain sizes within our films:

• 9±2 nm for P < 400 W

• 18±3 nm for 400 W < P < 600 W.

• This is similar in size to the

grains produced in other studies.

• A peak 2θ intensity:

• Most films at ~40⁰ which corresponds to the (220) lattice orientation.

• Only 3 samples: at 71⁰ (321).

• RRR values show no preferred orientation between the two values

of 2θ as the highest RRR values were recorded for both.

Film Morphology: XRD



XPS analysis

C 1s

O 1s

Nb 3d5/2

Nb 3d3/2

Nb 3p3/2

Nb 3p1/2

Nb 3s

As received

Ar+ bombarded

x 102

10

20

30

40

50

60

70

80C

PS

600 500 400 300 200 100 0

Binding Energy (eV)



Nb: 3d3/2 (205) and 3d5/2 (202)

As receivedAr+ bombarded

Nb

3d

Nb

3d

x 102

0

10

20

30

40

50

60

70

CP

S

214 212 210 208 206 204 202 200 198 196

Binding Energy (eV)



O 1s (531)

O 1

sO

1s

x 102

0

5

10

15

20

25

30

35

40

CP

S

540 538 536 534 532 530 528 526

Binding Energy (eV)




C 1s

C 1

s

C 1

s

x 102

0

2

4

6

8

10

12

14

CP

S

294 292 290 288 286 284 282 280 278

Binding Energy (eV)




samples deposited

with a 50-V bias

Film Morphology: SEMsamples deposited without a bias

02468

101214161820

100 300 500

Gro

wth

Rat

e (

nm

/min

)

Power (W)

0 V

50 V



• Only one sample has been analysed using EBSD

• EBSD data shows grains larger than 18 nm present in the

sample, the largest of which are of the order 250 nm across

their longest axis. The larger grains are surrounded by a

matrix of smaller grains of a similar size to those described by

XRD.

Film Morphology: EBSD



• RRR measurements

• have been performed using a purpose built cryostat

housing a four point probe.

• DC SQUID measurements

• were performed using a Quantum Design PPMS. The

measurement gives values for both the first and second

critical fields, Hc1 and Hc2.

Superconductivity evaluation



• In all cases

• RRR is higher for a

biased substrate than

with unbiased

• However, increasing

the bias further does

not always result with

increasing RRR

Superconductivity evaluation: RRR

• 2 RRR 22 for 70 samples studied.

• Samples with RRR ≥ 10 were deposited with

300 ≤ P ≤ 600 W.



Hc1

Hc1

Hc1

Hc2

Hc1

Hc1

Hc2

Hc2

Hc2

Hc2

Hsmp Hsmp

Hsmp Hsmp

Hsmp

Superconductivity evaluation: DC SQUID



• 5 samples studied for B perpendicular to the sample surface

• 2 samples studied for B parallel to the sample surface

Superconductivity evaluation: DC SQUID

A typical DC magnetic susceptibility

measurement with the sample parallel to the

magnetic field.

Some results are quite

unexpectable



HC2 increases with RRR at T = 6 K

RRR vs Hc2



What we are looking for?

We have to find a criteria for

a simple evaluation of

deposited film to predict its

behaviour in RF field

• To stay in Meissner state would

be ideal (Q0 conditions)

• T<T1

• Should the criteria for a good

film be Hsmp?

• Is it valid for RF?



We would like to make a

simple RF evaluation of

deposited film

• Tangential magnetic field

• Surface resistance

(power loss)

measurements at each

part

• Nb coated sample

comparison to bulk Nb

and other samples

RF sample test: idea

Nb plate

Nbcylinder

Nb coated Cu plate



RF modelling for 3.9 GHz

H-field

E-field• The idea was found to be

working

• There is an RF leakage

in the gaps

• Optimisation work

on bandwidth and

behaviour at very

high conductivity

with Microwave

Studio



• Mock up 3.9 GHz pill

box cavity has been

fabricated to validate

the simulation results

obtained with

Microwave Studio.

• An order for a niobium

7.8 GHz cavity is in

place.

• Samples:

• a 100-mm diam. Nb disk

• a 100-mm diam. copper

disk with thin films of Nb

deposited on the

surface.

Aluminium mock up RF cavity test



• HiPIMS sample deposition

• Nb, NbN, NbCN, Nb3Ge MgB2

• Plasma ALD deposition

• Nb, NbN, NbCN, MgB2

• RRR measurements

• AC and DC magnetisation measurements

• RF pill-box test facility

• Building up a test facility (Dec 2014)

• Testing with a bulk Nb disk (Jan 2015)

• Sample measurements (beginning 2015)

• 3D coating

Future plans



New surface analysis rigs: SIMS



New surface analysis rigs: Auger microscope



• Lakeshore 7000

AC Susceptometer

• Complex susceptibility

’ and ”

• Differential

susceptibility

• Frequency effect

• AC amplitude effect

• Temperature effect

• Sensitivity 210-7 Oe

Superconductivity evaluation:

New AC susceptibility



• Quality of the film (morphology, RRR, Hc2)

depends on

• deposition parameters such as

• Substrate temperature,

• Ion/atom arrival ratio,

• Substrate bias,

• Plasma generation at the target

• pulsed or not

• HiPIMS

• Substrate crystallography

Conclusions



• Sample evaluation

• RF pill-box cavity is the best

• Cost of manufacturing and cost of LHe,

• Time consuming

• AC and DC susceptibility:

• Not direct for RF (Hc1 and Hc2) and a cost of LHe

• Quicker

• TC and RRR is an initial of evaluation

• RRR > 10-15, but quite indirect

• Quick, cheap

• Surface analysis (to explain what’s gone wrong):

• Dense or columnar

• Grain size

• Composition and impurity

• Defect’s density

Conclusions (2)



Dr. R. Valizadeh, Dr. O.B. Malyshev,A. Hannah, D.O. Malyshev, S. Pattalwar, J. HerbertA. Wheelhouse, P. McIntosh

Dr. G. Stenning

The UK’s SRF collaboration team

S. Wilde, Dr. B. Chesca

P. Pizzol, Prof. P. ChalkerF. Lockwood-Estrin, Prof. J. Bradley



ESCAlab-II: monochromated XPS,

mapping Auger, UPS



• If the desired material is not a binary compound but a single element one, such

as metals or semiconductors, it is very difficult to obtain them using the standard

ALD process. This is one of the reasons why Plasma enhanced ALD was

invented (PEALD).

• Through the use of a plasma source it is possible to generate hydrogen radicals

necessary to reduce the metal or semiconductor precursor. There are different

types of PEALD according to where the plasma source is positioned: if the

radicals are generated away from the reaction chamber and injected like any

other precursor the technique is called Radical enhanced ALD (Figure (a)).

Direct plasma ALD (Figure (b)) is when the plasma develops very close to the

substrate, while Remote plasma ALD (Figure (c)) is when the plasma is tens of

centimeters away from the substrate.

Plasma enhanced atomic layer deposition (PEALD)

oleg malishev development of thin films for superconducting rf cavities in as te-c

Science

malyshevthin films

new ideas

rf bias

superconducting rf cavities

rf cavitiesobjectives

mbar power

single crystal substrate

cpecvdald deposition