Investigations of cryopumping for extreme high vacuum …...Cryopump BNNT cryosorber test setup • Removable cryopump motor, displacer: fully bakable • Gas inlet system, orifice

Marcy Stutzman, Philip Adderley,

Veronica Over, Matt Poelker

Thomas Jefferson National Accelerator Facility

Newport News, VA 23601, USA

Investigations of cryopumping for

extreme high vacuum systems

Marcy Stutzman IUVSTA Cryo Workshop 2016



Outline

• Overview of JLab SRF beamline vacuum, operation

• Jefferson Lab polarized electron source

• Cryopumping for XHV

– Conventional charcoal cryopump

– Novel nanomaterial tests

• Summary


JLEIC


US Department of Energy, 12 GeV electron accelerator

Up to 90% polarization from DC photoemission source

Electron currents to 200μA beam (CW) to four experimental halls

Halls A, B, C

Hall DSource

Polarized source


JLab SRF Accelerator

Original CEBAF

20 cryomodules,

4 five cell cavities in each

CEBAF 12 GeV upgrade

Add 5 cryomodules/Linac,

4 seven cell cavities in each

Charlie Reece, Superconductor Science and Technology Special Issue


SRF Cryomodule Performance

A. Freyberger, IPAC2015


SRF system vacuum: operational issues

Primary concerns at CEBAF

• Gradient / Q achieved in each cryomodule

• Particulate transport mitigation

• Particle / Field emitter elimination during processing

Cryomodule Cryomodule

Warm girder

Ion pumps Beamline valves


(out of 1 GeV/pass)


SRF particulate hunting efforts

JLab, Rongli Geng


Beamline pumping

4x Differential pumping (NEG and ion)

beginning and end of each Linac

35 L/s ion pumps

Warm girder between CM

Pump drop for CM vacuum space

Adding NEGs to one Cryomodule, Girder

Goal: mitigate Ti, Ta

particulate from ion

pump, improve H2

pump speed


Outline

• Overview of JLab SRF beamline vacuum, operation

• Jefferson Lab polarized electron source

• Cryopumping for XHV

– Conventional charcoal cryopump

– Novel nanomaterial tests

• Summary

Polarized source


DC Photoemission Source

• Strained superlattice GaAs/GaAsP photocathode

• Residual gasses ionized, limit operational lifetime

• SAES WP NEG modules

• Gamma Vacuum XHV/SEM style ion pump

• Base pressure approaching XHV ≡ P < 1x10-10 Pa

-130kV

light inelectrons out

0.

36

GaAs

GaAsP

GaAs

Strained

Superlattice

Photocathode

14 layers


Existing Electron Source Pumping

• Heat treat to reduce outgassing

– 400°C, 100 hours

– Q=1.33x10-10 Pa·m·s-1

• Installed pumping

– SAES WP1250 modules x 10 (5600 L/s total)

– Gamma SEM/XHV ion pump

(45 L/s)

• Measured pressure ~1.3x10-10 Pa

(extractor gauge, x-ray limit measured and subtracted)

• CEBAF Lifetime: ~100 Coulombs = 10 days at 100 uA

• Extrapolating e-RHIC to 10 mA: 100 Coulomb lifetime < 3 hours


XHV Cryopumping

Prior research toward XHV

cryopumping

– Double shield walls

• Shiokawa 1996

– Reduced cryosorber temperatures

• Iwasa 1996

• Our initial path toward XHV

cryopumping

– Leybold bakable XHV cryopump

• LN2 chill circuit

– 10K cryosorber

– 30K single shield wall• Presented at AVS 55 by Dieter Mueller


Commercially available XHV gauges

extractor

3BG

AxTran

Backgrounds

• x-ray limits (measured, corrected)

• Heating (measured, corrected)

• ESD effect (degas, long stabilization)

Advertised x-ray limit (Pa)

Extractor 1 ×10-10

AxTran 5 ×10-11

3BG1 5 ×10-12

1) F. Watanabe, JVSTA 28, 486 (2010)


Leybold Cryopump test system

• Three XHV Gauges

• Heat treated chamber for lower

outgassing

• NEG / Ion pumps

– Overboard bake ion pump

• 10” all metal gate valve

• Bellows (12” to 10” adapter,

vibration isolation)

• Cryopump

– Overboard bake ion pump

– Compressor

– ColdHead

– LN2 chill line / chill line evacuation

system


• Getters

– 4 x WP950 modules

– 1 GP500 pump

• Ion pump

– 45 l/s SEM style Ti/Tan

• Outgassing rate

– 1.3x10-10 Pa·m·s-1

• Volume: 40 l

• Area: 8000 cm2

Simple calucuations:

expected pressure

5 x 10-11 Pa

Chamber with NEG/Ion pumps

S=2200 l/s60% NEG activation


NEG, Ion Pumps only NEG, Ion and Cryopump

• Getters 2150 l/s

• Ion pump 45 l/s

• Cryopump 800-1300 l/s

(conductance dependent)

• Area: 25,000 cm2

– 8,000 cm2 Chamber

– 3,400 cm2 Bellows

– 13,700 cm2 Valve

• Outgassing rate valve (not treated)

≥ 6x10-9 Pa m s-1

Calculated pressure

~ 5 x 10-11 Pa

• Getters 2150 l/s

• Ion pump 45 l/s

• Area: 8000 cm2

• Outgassing rate (heat treated)

– 1.3x10-10 Pa m s-1

Calculated pressure

5 x 10-11 Pa

Do we gain no benefit

from cryopumping,

or are our calculations

too simple??


Molflow+ simulations

• Molflow+ software: Roberto

Kersevan and Márton Ady

• 3D CAD model cryopump

system, including vibration

isolation bellows and gate

valve

Chamber with

ion and NEG

pumps

Gate valve

representation

Bellows

Cryopump


Test Particle Monte Carlo

• Tracks simulated

particles

– Pump speeds defined by

sticking coefficients

– Outgassing rates as

measured previously

– Compare simulation with

measured results

Ion pump

Cryopump

Valve thickness,

effective

outgassing rate

XHV gauges

NEG

GP500

pump

Bellows

NEG Sorb-AC

WP950 pumps


Molflow+ cryopump simulations

P(simulated)=

2.7x10-12 mbar

P(simulated)=

1.6x10-12 mbar

gauge

position

Valve Closed Valve OpenSimulations show

outgassing from

valve, conductance

severely limit

cryo-pumping on

chamber

Charcoal sticking

coefficient α=0.6

mbar


Molflow+ Cryopump Only

P(simulated)=

7x10-12 mbar

No NEG pumping

• Simulations predict

pressures a factor of 2.5

higher if NEGs removed:

• Cryopumping largely

ineffective in this

geometry

• High conductance

configuration with NEGs

in chamber far superior

for achieving XHV

mbar


Cryopump Testing

• Full system bakeout: Chamber, Cryopump with LN2 chill

• NEGs partially activated during bakeout (60%)

• Gauges on and degassed

• Record pressure vs. temperature until gauges stable >1 month


Chamber pressures

0

1

2

3

4

5

6

25 30 35 40

Pre

ssu

re (

x10

-12 T

orr

≈ 1

0-1

0P

a)

days

Ext (keithley)

Axtran (keithley)

3BG controller

NEG, ion

pumps only

Open Valve

to cryopump

Corrections for gauge heating, x-ray limits

Pressure drop due to

turning off Axtran

filament


Performance of Leybold cryopump

Pressure ~4x10-10 Pa

◦Small discrepancy in

manufacture gauge

calibrations

Simulations either1. Overestimate

Outgassing

2. Underestimate

conductance to

cryopump

3. Sticking coefficients

(from literature)

incorrect for this

system

0

1E-10

2E-10

3E-10

4E-10

5E-10

6E-10

7E-10

Extractor 3BG AxTran

Pre

ssu

re (

Pa)

Equilibrium pressures

NEG, Ion only

NEG, Ion and Cryopumps


• Cryopump during chamber bakeout?

• Cryopump during NEG activation?

– Heat load in addition to gasses

How best to incorporate Cryopump?

• Cryopump during beam operation?

– Vibration issues due to compressor

Cryopump stage 1

and 2 temperatures

Pressure (Torr)

0

10

20

30

40

50

1

10

100

1000

10000

0 1 2 3 4

Tem

per

atu

re (

K)

Pre

ssu

re (

x1

0-1

2 T

orr

)

Time (hours)

NEG activation


Cryopump as process pumping

1E-10

1E-09

1E-08

Initial bake, CP and

chamber

NEG activation into CP chamber bake into CP

Pre

ssu

re (

Pa

)

NEG, ion only

NEG, Ion, Cryo


Leybold Residual Gas Species Analysis: 9K

XHV test chamber with NEG, ion pumps, gauges, VQM

Not fully baked for this gas analysis

(vented, moved, reconfigured, reassembled)

Cryopump cooled down, 4.5 days elapsed time

Pressure: 1x10-9 Torr

VQM spectrum shows partial pressures at 100 AMU

1.E-14

1.E-13

1.E-12

1.E-11

1.E-10

1.E-09

1.E-08

1.E-071 5 9

13

17

21

25

29

33

37

41

45

49

53

57

61

65

69

73

77

81

85

89

93

97

10

1

10

5

10

9Part

ial

pre

ssu

res

(Torr

)

AMU

1e-9


Cryopump with BN Nanomaterial Cryosorber?

• Boron Nitride Nanomaterial (BNNT) as

potential cryosorber

– Freestanding boron nitride nanotubes

• Mechanical attachment possible

– High thermal conductivity

– Chemically inert

– Very porous structure

– Local JLab affiliated company

• Potential benefits over Leybold Cryopump

– Simpler bakes

– No adhesives

– hydrocarbon free

– EUV lithography applications?

1 g BNNT

10 g CNT (similar scale)

US Patent pending


First Try: BNNT installed as cryosorber

• Nanomaterial mechanically

attached (no adhesive)

• Copper mesh for retention

• Glass tube / ceramic beads

insulating diodes

BNNT material:

Stays in place through static electricity


Ion

pump

H2

Stabilion

VQMExtractor

Test

chamber

orifice

motor

Cryopump

BNNT cryosorber test setup

• Removable cryopump

motor, displacer: fully

bakable

• Gas inlet system, orifice

for pump speed and

capacity measurements


BNNT cryopump: first cooldown

• Baked 48 hours ~150°C

• Coldhead temperature: 15.5K (14K before BNNT)

• 39K: pressure drop – Cryopumping

0

50

100

150

200

250

300

350

1.E-11

1.E-10

1.E-09

1.E-08

1.E-07

1.E-06

0 6 12 18 24

Cry

op

um

p T

emp

era

ture

(K

)

Pre

ssu

re (

To

rr)

Hours

P(Extractor)

T1

T2


Hydrogen pump Speed vs. gas dose

• High initial pump

speed

– Modification of

test dome

procedure

– Conductance

limiting orifice

– 1 g BNNT

– 15.5 K base

temperature

• Poor capacity as

expected

0

5000

10000

15000

20000

0.000001 0.0001 0.01 1

Pu

mp

Sp

eed

(L

/s)

Gas dose (TorrL)

Red: First run, varied pressure

Blue: Second run, constant inlet pressure


Mass spectra comparison

1.E-14

1.E-13

1.E-12

1.E-11

1.E-10

1.E-09

1.E-08

1.E-07

1 4 7

10

13

16

19

22

25

28

31

34

37

40

43

46

49

52

55

58

61

64

67

70

73

76

79

82

85

88

91

94

97

10

0

10

3

10

6

10

9Parti

al

pre

ssu

res

(Torr

)

AMU

BNNT warm

BNNT cold

1.E-141.E-131.E-121.E-111.E-101.E-091.E-081.E-07

1 4 7

10

13

16

19

22

25

28

31

34

37

40

43

46

49

52

55

58

61

64

67

70

73

76

79

82

85

88

91

94

97

10

0

10

3

10

6

10

9

Parti

als

(T

orr

)

AMU

Leybold XHV pump

Cold, unbaked

1e-9

GP VQM, total pressure with

Leybold extractor gauge


BNNT limitations

• First BNNT prototype: 1 g of material = 300 m2

• Typical Cryopump: up to 100,000 m2 installed charcoal surface area

• Hydrogen specific cryopumps

– T ≤ 10K

– Charcoal on underside of fins

• BNNT first prototype

– T ~ 15K

– Heat load due to copper mesh absorption

– BNNT applied only to tops of fins

• BNNT second prototype improvements

– Fine wire attachment, no copper mesh

– Total 4.4g BNNT, 1000 m2

– Attachment on both sides of fins


BNNT simulations

Simulated pressure profile: Upgraded BNNT cryopump

Sticking coefficient: top and

bottom of fins

α =0.003 α =0.012

Simulated Gauge Pressure 6.2x10-9 Pa 1.55x10-9 Pa

First BNNT prototype

Sticking coefficient

calculated from

measured pressure

α =0.003

Measured pressure 6.2x10-9 Pa

mbar

as built α = 0.003 α = 0.0122nd

Prototype

model


Summary

• JLab Cryomodule vacuum system

– investigating particulates, addressing hydrogen pump speed

• Evaluation of Leybold XHV cryopump

– Three gauges characterize pressure near XHV

– Experimental results: only small improvement

– Simulations, calculations and measurements find conductance,

valve outgassing limit results

– Mass spectrum: adhesive contamination?

• BNNT cryopump testing

– Potential hydrocarbon-free cryopump

– Relatively low surface area, low effective α, minimal

cryotrapping?

– Next prototype assembled, to be tested shortly


Backup


BNNT properties

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.na

sa.gov/20140004051.pdf

Purity 40 to 50% by mass

Residual hBN flakes and micro-droplets of

elemental boron, by TEM

Walls 1 to 5 walls, mostly 2 or 3 walled tubes

Tube Length up to 200 microns by SEM

Surface Area up to 300 m2/g by BET

Bundles single through bundles of 5 tubes, TEM

Band Gap 5.7 eV (semiconducting) EELS

spectroscopy

Strength in air 800°C (CNT is 400°C)

Thermal

oxidation

resistance

Stable to at least 920˚C in air

Thermal

Conductivity

3000 W/mK (Cu = 400 W/mK,

CNT 60-40,000 W/mK)

TEM

Impurities: B and hBN


BET surface density comparison

Charcoal properties* m2/g Density (g/cm^3)

Roth 1700 0.3

Chemviron GFF30 1600

Degusorb 2300 0.44

Chemviron SCII 1500 0.45

BNNT 121-300 0.001

*Christian Day, “The use of active carbons as cryosorbent”,

Colloids and Surfaces A 187 (2001) 187-206.

Lower surface area, but

Low density = high porosity: effect on capacity, speed?


0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

1.E-10

1.E-09

1.E-08

1.E-07

0.01 0.10 1.00 10.00 100.00

Pu

mp

sp

eed

(L

/s)

Pre

ssu

re (

To

rr)

Time (hours)

Pump Speed and pressure: 1g BNNT, 15.5K

P(Torr) Run 2

Speed (L/s) Run 2

H2 pump speed vs. time, 1g BNNT, 15.5K


0

5000

10000

15000

20000

25000

30000

35000

1.E-10

1.E-09

1.E-08

1.E-07

0.01 0.10 1.00 10.00 100.00

Pu

mp

sp

eed

(L

/s)

Pre

ssu

re (

To

rr)

Time (hours)

Pump Speed and pressure: 1g BNNT, 15.5K

P(Torr) run 1

P(Torr) Run 2

Speed (L/s) Run 1

Speed (L/s) Run 2

H2 pump speed vs. time, 1g BNNT, 15.5K

Documents

Investigations of cryopumping for extreme high vacuum …...Cryopump BNNT cryosorber test setup • Removable cryopump motor, displacer: fully bakable • Gas inlet system, orifice