Reduction of outgassing rate and photon stimulated desorption yield

Transcription

Reduction of outgassing rate and photon stimulated desorption yield
IUVSTA WS-63 Workshop for “Surface phenomena limiting ultimate pressures inVacuum Systems”,
Residencia Santo Tomás, Ávila, Spain, Sep. 14-19 (2010)
Reduction of outgassing rate and photon stimulated
desorption yield by ozonate water cleaning process for
the large accelerator vacuum system
Gao-Yu Hsiung
Vacuum Group
NSRRC, TAIWAN
E-mail: hsiung@nsrrc.org.tw
1
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Outline
 Why
the ozonate water cleaning
 How to do it
 What we have found
 Conclusions
2
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Why the Ozonate water cleaning
 A Synchrotron Light Source needs a vacuum vessel with
extreme low pressure for accommodating the circulating
electron (positron) beam, without the scattering or
interactions with the residual outgas molecules, that allows
the beam maintains a smallest emittance, highest stable
qualities, and the long life time.
 The problem of the accelerator is limited space for the
pumps to remove the outgas from the beam ducts effectively.
 The only solution for reaching a lowest pressure is to provide
an extreme clean surface that results in an extreme low
outgassing rate for the beam ducts.
3
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Synchrotron Light Sources in Taiwan
Taiwan Light Source (TLS) is the 2nd third
generation light source (1993)
VUV
SX
HX
1E16
U9(1)
Hsinchu Science Park
Flux (Phot/s/0.1%bw)
1E15
(3)
EPU5.6(1)
SP8-IVXU3.2(1)
U5(1)
(3)
(3)
(5)
(5)
(7)
SW6
1E14
1E13
(3)
(5)
(7)
SP8-BM
IASW6
W20
BM
1E12
SWLS
Electron
Storage Ring
(1.5 GeV)
1E11
0.1
LINAC
(50 MeV)
Booster Ring
(1.5 GeV)
1
10
HX
VUV
SX
VUV
SX
 Circumference – 120 m
 Critical Energy – 2.14 keV
 1996 – Ramping to 1.5 GeV
 Natural Emittance – 25 n mrad
 2000 – 1.5 GeV Full Energy Injection (200 mA)
 2004 – Operation with Superconducting RF Cavity  Average Pressure (200 mA) – 0.68 nTorr
 Accumulated Dose > 8000 Ah
 2005 – Top-up Injection at 300 mA
 Life Time – 10 h
 1993 – Operation 1.3 GeV
5
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
100
1000
10000
Photon Energy (eV)
HX
HX
HX
HX
HX
SX
ID in operation
W20 (1995) 4 ~ 15 keV
U10 (1995) 3 ~ 500 eV
U5
(1997) 60 ~ 1500 eV
U9
(1999) 4 ~ 100 eV
EPU5.6 (1999) 60 ~ 1400 eV
SWLS (2002) 4 ~ 30 keV
SW6
(2004) 6.5 ~ 19 keV
IASW6 (2006) 6.5 ~ 19 keV
New ID in planning
2nd IASW6 (2009)
3rd IASW6
EPU4.6
G.Y. Hsiung
The major beam ducts for the electron
storage ring is made of Al-alloys
Undulator
SC Wiggler
SRF
BM
SR
Front End
e-
SGV
Vacuum Chambers
NEG
e6
SIP
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
Electron Storage Ring
G.Y. Hsiung
The pressure increases when installing new insertion
Top-up
devices and needs beam scrabble cleaning
1.3 GeV
200 mA
Averaged Pressure per Beam Current (Pa/mA)_
1.0E-09
Install ID
W20
9.0E-10
1.5 GeV
1.5 GeV
Replace new 300 mA
Kicker Chambers
Install
SRF Cavity
200 mA
Install ID-CH
EPU5.6 + U5
Install ID-CH
U9
8.0E-10
7.0E-10
Install
New FE
6.0E-10
Install ID
IASW6
Replace ID
IASW6
Install ID Install ID
SWLS
SW6
5.0E-10
4.0E-10
3.0E-10
2.0E-10
1.0E-10
07/01/1993
YEAR/MONTH
15.5 years
(iv)
2008/07
2007/07
(iii)
2006/07
2005/07
2004/07
2003/07
2002/07
2001/07
2000/07
(ii)
1999/07
1998/07
1997/07
1996/07
1995/07
(i)
1994/07
1993/07
0.0E+00
12/31/2008
Ref. G.Y. Hsiung, PAC’09
7
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
The beam life time reduces after installation of new
Top-up
vacuum devices (new absorbers)
1.3 GeV
1.5 GeV
1.5 GeV
200 mA
200 mA
300 mA
5000
Beam Current × Life Time (mA h)_
4500
Install ID
W20
4000
Install ID-CH
EPU5.6 + U5
Replace new
Kicker Chambers
Install ID
SWLS
Install
SRF Cavity
Install ID-CH
U9
3500
Install ID
SW6
3000
Install
New FE
2500
Install ID
IASW6
Replace ID
IASW6
2000
1500
1000
07/01/1993
YEAR/MONTH
15.5 years
(iv)
2008/07
2007/07
(iii)
2006/07
2005/07
2004/07
2003/07
2002/07
2001/07
2000/07
(ii)
1999/07
1998/07
1997/07
1996/07
(i)
1995/07
1993/07
0
1994/07
500
12/31/2008
Ref. G.Y. Hsiung, PAC’09
8
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Aluminum Alloys (Al) Vacuum Chambers for the TLS
Storage Ring
Downstream chamber wall of B-chamber, act as the photon absorber, absorbs the non-used
synchrotron light.
B-Chamber
• CNC machining in pure alcohol
• Freon cleaning before welding
• TIG welding in clean room
S-Chamber
• Al Extrusion
• Chemical cleaning by acid
• TIG welding in clean room
ID-Chamber
• Al Extrusion
• Chemical cleaning by acid
9
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Oil-free manufacturing process for Al Bending
chambers provides high reliable quality
3) Surface Cleaning
1) NC Machining
with Ethyl Alcohol
4) DIP Installation
2) Dimension Check
After Machining
5)Welding
in Clean Room
9) Installation
in the Tunnel
8) Pre-assembly
In Lab
10
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
7) Leak Test
6) Deformation Check
AfterWelding
G.Y. Hsiung
Assembly, Welding and Testing (at NSRRC) for the BChamber
Assembly
TIGWelding
Vacuum Test
11
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Need to request the commission period for Beam-cleaning to
reduce the PSD yields (dynamic pressure) (weeks ~ months)
Pressure rise per beam current and
Desorption coefficient in B-ch (ηB)
during beam cleaning.
Desorption coefficient at S-ch (ηS)
and B-ch (ηB)
TMP pumps were useful for removing
the large amount of PSD outgas during
the commissioning.
(a)
(b)
100 Ah
12
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
Ref. G.Y. Hsiung et al, J.Vac. Sci.Technol. A13, 2569 (1995)
G.Y. Hsiung
TLS Operation Experiences
- Pressure Requirement
200 mA × 10 hours
100 Ah
13
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Residual Gases of PSD from TLS
Major PSD Outgas :
H2 (93%) > CO (4.4%) > CO2 (1.2%) >
CH4 (0.5%) ~ H2O (0.5%)
Ref. G.Y. Hsiung et al, J.Vac. Sci.Technol. A12(4), 1639 (1994)
14
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
Ref. G.Y. Hsiung et al, J.Vac. Sci.Technol. A13, 2569 (1995)
G.Y. Hsiung
The 3 GeV TPS Vacuum System
 Vacuum system for TPS (e-beam) storage ring (SR):
 Lattice : 24 DBA, ξ= 2 nmrad
 Straight : 12m (6)+7m (18)
 E = 3 GeV, Imax = 400 mA (top-up)
 Booster (concentric with SR) : 3 GeV full energy injection
 Design safety margin: 28% (heat load from 3.3 GeV, 350 mA)
 Requirements of SR vacuum system:
 Ultra-high vacuum (UHV)
 High reliability
 High stability
15
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Layout of 1/6 Super periods (L = 86.4 m)
12 m
1
7m
7m
2
3
1.
2.
3.
4.
16
One Cell (1/24) Vacuum System
Bending beam ducts
Straight beam ducts
Beam ducts for Insertion Devices (ID)
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
4
7m
G.Y. Hsiung
24 Arc Cells of Vacuum System
SGV
S3
B1
S4
B2
SGV
17
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Vacuum Design of One Cell
 Simple structure of vacuum beam ducts
along the beam channel, no flange, few
absorbers, few bellows, for lowering the
impedance.
 Completely oil-free and precise CNC
machining for the B-chambers and TIG
welding in the clean room to obtain a
clean surface with lowest surface
outgassing rate and the consequent
lowest pressure.
 Strong back supports for the BPM fixed
on the girders for positioning the BPM
precisely and cooling channels drilled
through the B-chambers provides an
uniform temperature control for
assuring the BPM shift < 0.1 micron
against the thermal stress.
18
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
1. Button : φ6.9 mm
2. Gap : 0.3 mm
3. Distance : 17.5 mm
4. Angle : 6.7°
G.Y. Hsiung
Main Chambers for One Cell
S3
BPM
Pumping port
Bellows
B1
S4
B2
19
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
The electron storage ring request an
extreme low outgassing rate
 Vacuum system for 3 GeV TPS electron storage ring requests an average
pressure (P) < 1×10-9 Torr in the chambers, which depends on the rate of
thermal outgas (QTH), the rate of photon stimulated desorption (QPSD), and
the effective pumping speed (S).
P (QTH
Where
QPSD ) / S
K1 q A K2
I
q : surface outgassing rate (Torr·l·s-1·cm-2) of chambers
A : area (cm2) of chamber surface produces outgas
η: yield (molecules/photon) of photon stimulated desorption of photon absorbers
I : electron beam current (mA)
K1, K2 : proportional constants
Typically in case of P < 1 nTorr :
q < 1×10-13 Torr·L/(s·cm2)
η < 1×10-4 Molecules/photon
20
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Cleaning methods for the chambers
 The popular cleaning methods:
 Aluminum chambers
 Chemical cleaning with HNO3/HF acid (CERN*, TLS) for Al extrusion
chambers
 Oil-free CNC machining in pure alcohol and Freon cleaning (TLS) for Al
bending chambers
 Copper chambers
 Chemical cleaning with Citranox (APS)
 Cleaning processes with the solutions of HF, Freon,
etc. will be replaced by the ones without polluting
the environment.
* Ref. A.G. Mathewson et al, J.Vac. Sci.Technol. A7(1), 77 (1989)
21
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Chemical cleaning for Al extrusion
chambers
1.
Immersion in NaOH (45 g/L) at 45 ºC for 1~2 minutes.
2.
Rinsing in demineralized water.
3.
Immersion in an acid bath containing HNO3 (50% by volume) and HF (3%
by volume).
4.
Rinsing in demineralized water.
5.
Ultrasonic bath in demineralized water (R > 10 MΩ) for ~ 20 minutes.
6.
Drying.
* Ref. A.G. Mathewson et al, J.Vac. Sci.Technol. A7(1), 77 (1989)
22
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
AES of Al samples after Chemical
cleaning
3.0E+04
“F” is resided on the surface
after chemical cleaning
2.0E+04
1.0E+04
0.0E+00
0
-1.0E+04
-2.0E+04
C
500
F
1000
O
1500
Al
5.0E+04
O
4.0E+04
-3.0E+04
F
3.0E+04
Al
O
2.0E+04
F
1.0E+04
C
Al
C
0.0E+00
0
23
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
1
2
3
4
G.Y. Hsiung
Motivations of using Ozonate water
 An effective cleaning method with ozone gas for the
reduction of carbon contamination from the surface has
been investigated by T. Momose [1].
 Another method of ozonate water cleaning for the
superconducting rf cavities studied by K. Asano [2] shows
good results of removing the surface contaminations.
 Replace the Freon cleaning process with ozonate water
cleaning for Aluminum B-chambers after ethanol
machining.
 Ozonate water may remove the residual F atom from the
Al extrusion chambers
[1] T. Momose, Y. Maeda, K. Asano and H. Ishimaru, J. Vac. Sci. Technol. A13(3), 515 (1995).
[2] K. Asano, T. Furuya, S. Mitsunobu, T. Tajima and T. Takahashi, “Stable Performance of 508-MHz Superconducting RF
Cavities for KEK B-Factory”, KEK Preprint 95-191 (1996).
24
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
How to do it?
 Key is to produce a very clean surface from the beginning!
 Completed oil-free manufacturing process for Al B-chamber
 Oil-free CNC machine in dust controlled clean room
 Ethanol machining environment with pure alcohol spray
 Dry ambient atmosphere for the machined chambers
 Dry compressed air (H2O < 1 ppm) for spray system
 Dry N2-filled aluminum bag for wrapping the machined chambers
 Ozonate water cleaning for unpacked Al chambers prior to the
TIG welding in the clean room
25
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Machining for the TPS Al B-chambers (~ 4 m)
Oil-free CNC Machining
in Clean Room
CNC Machining Processes (sprayed with pure alcohol)
26
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Ozonate water Cleaning for the B-chamber
Aluminum half plates, for 4 m Bending
chambers, packed after oil-free machining
Ozonated Water Cleaning (> 20 ppm)
4. Transport to Welding room
1. Ozonate water vessel
2. Immerse chamber in
Ozonate water (30 min)
Unpack and Prepare for Cleaning
27
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
3. Drying in Clean booth
G.Y. Hsiung
Welding for the B-chamber in clean room
Upper and Lower Halves
B-Chamber Alignment
Alignment with Pins
28
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Welding for B-chambers
Manual welding for rest parts
Auto welding for straight sides
B1
29
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Summary of welding procedure for Bchambers
1. Manual TIG welding for pumping ports of each halves.
2. Auto TIG welding for both non-parallel straight sides,
with 6 torches ignited simultaneously.
3. Manual TIG welding for rest curved sides.
4. Machining the end ports.
5. Manual TIG welding for end ducts, flanges, and cooling
tubes.
30
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Assembly of 14m vacuum cell, in-situ
welding, and vacuum baking (150°C, 24h)
31
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
What we have found
 Prepare the Aluminum and Copper test samples
 AES - Surface analysis, depth profile
 SIMS - Surface contaminations
 Kevin Probe – Surface Work function
 Thermal outgassing rate
 Photon stimulated desorption yield
32
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Sample preparation for cleaning test
• Ethanol machining : only Ethanol machining
• Chemical cleaning : regular machining and chemical cleaning
• Ozonate water cleaning : Ethanol machining and immerse in
ozonate water for 30 minutes
• Citranox cleaning (Cu) : regular machining and citranox cleaning
Chemical cleaning process:
NaOH (45 g/L) at 45 ℃ for 2 min → bubble bath in DI water for 10 min → (50 % HNO3 + 3 % HF) for 2 min →
bubble bath in DI water for 10 min → ultrasonic rinsing for 2 min → drying with pure nitrogen gas
Citranox cleaning process:
2% Citranox in 60℃ DI water and ultrasonic rinsing for 10 min → DI water rinsing for 10 min → drying with pure
nitrogen gas
33
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
AES surface inspection
Surface analysis of AES for Al samples without photon exposure
Chemical cleaning
6.7ppm ozonated water cleaning
Only ethanol machining

34
The Al samples with ozonate water cleaning provide lowest
carbon resided on the surface in comparison with those samples
cleaned with either chemical cleaning or ethanol machining.
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Al samples - (a) Original (CNC machining) ; (b) O3
water cleaning
(a) Original
O
C
(31%)
Al
Al
O
C
(b) O3 water cleaning
O
C
(7%)
Al
O
Al
C
Low Carbon
35
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Thermal Outgassing Rate (Aluminum)
1.0E-07
(a)
(b)
1.0E-08
(c)
1.0E-09
q(torr L/sec cm2)
♦♦♦ (a) O3 Water Cleaning
臭氧水清洗
---- (b) Ethanol CNC Machining
酒精加工
xxx (c) Chemical Cleaning
化學清洗
1.0E-10
After Baking :
1.0E-11
(c) Chemical cleaning
q72 = 9.5×10-14 Torr·L/s·cm2
1.0E-12
(b) Ethanol CNC Machining
q72 = 1.4×10-14 Torr·L/s·cm2
1.0E-13
1.0E-14
1.0E-15
1.E-04
(a) O3 Water Cleaning
q72 = 4.8×10-15 Torr·L/s·cm2
1.E-03
1.E-02
1.E-01
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
* 1 Torr·L/(s·cm2) ~ 1330 Pa·m/s
TIME (hour)
36
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Photon exposure experiments taken at
19B(PSD) Beam Line of TLS
Sample
Ion gun
Safety
Shutter
Photon
Shutter
X-Y
Slits
SR
QMS
 Yield of photon stimulated desorption
η = 3.5 × 1019.S.ΔP/ΔN
ΔN = 8.05 × 1017.E.Ie.θ/2π
37
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
• Sample has been installed at
the 19B(PSD) beam line for
exposure test.
• The synchrotron radiation is
white light with critical energy
2.14 keV.
• To measure the surface
concentration by in situ SIMS.
S
the effective pumping speed of system
ΔP
pressure increasing when photon exposure
E
electron beam energy (1.5 GeV)
Ie
electron beam current (300 mA)
θ
horiz. photon span on sample (0.0033rad)
G.Y. Hsiung
PSD yield for Aluminum samples
Yield of photon stimulated desorption (η)
for various cleaning processes :
38
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
• η of ozonate water cleaning
is more than one order of
magnitude lower than those
of only ethanol machining
or chemical cleaning at
1mAh.
• η decreases with beam dose
increases.
• Al samples with ozonate
water cleaning maintain the
lowest η at the beam dose
through 1000 mAh.
G.Y. Hsiung
Comparison of PSD partial pressure
- w/wo ozonate water cleaning
Partial pressure rise during photon exposure
• Ethanol machining
• 6.7 ppm ozonate water cleaning
CO
CO2
H2
C
CH4
 At the beginning of photon exposure, all the major outgases of ozonate water
cleaning are lower than those of ethanol machining.
 The major difference comes from carbon related outgases, e.g. CO, CO2, CH4.
39
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Comparison of PSD partial pressure
- different concentration of ozonate water
Partial pressure rise during photon exposure
• 6.7 ppm ozonate water cleaning
CO2
• 30 ppm ozonate water cleaning
H2
CO
C
CH4
 The major PSD outgases are H2 ~ CO ~ CO2 > CH4 ~ C2H6.
 The denser ozonated water effectively reduces all the major PSD outgases.
40
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Can the PSD of Crotch Absorber (OFHC) in Bchamber be reduced with ozonate water cleaning?
For Beam line
< 204 °C (at 3.3 GeV, 350 mA)
BM fan
I.D. fan
< 190 MPa (at 3.3 GeV, 350 mA)
• Crotch absorbers ( > 2.3 m) takes BM fan only.
• 100% I.D. fan will be penetrated through.
Courtesy: Albert Sheng (NSRRC)
41
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
SIMS for ethanol machined Cu cleaned with (a)
Citranox, (b) Ozonate water
Cu-X : Copper cleaned with Citranox
CH
C
C 2H2
C2
CH CC2H2
2
C F
Cl
F
Cu-O3 : Copper cleaned with Ozonate water
NEGATIVE ION MASS (a.m.u.)
Na
K
Ca
Cu
POSITIVE ION MASS (a.m.u.)
42
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
Cl
NEGATIVE ION MASS (a.m.u.)
Na
Al
CxHy- ,Cl- : reduced
Na+, K+ ,Ca+ : reduced
Al
K
Ca
Cu
POSITIVE ION MASS (a.m.u.)
G.Y. Hsiung
PSD for ethanol machined Cu samples
cleaned with (a) citranox, (b) O3-water
Yield of photon stimulated desorption (η) for different cleaning processes :
• η of ozonate water (20 ppm) cleaning is lower than that of
citranox cleaning at the beam dose through 400 mAh.
• η deceases with beam dose increases.
43
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
PSD-RGA for ethanol machined Cu cleaned
with (a) Citranox, (b) Ozonate water
Partial pressure rise during photon exposure
• 20 ppm ozonate water cleaning
• Citranox cleaning
CO2
H2
CO
C
CH4
C2H6
 The major PSD outgases are CO2 > H2 ~ CO > CH4 ~ C2H6.
 The ozonated water cleaning effectively reduces all the major PSD outgases.
44
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
PSD beam cleaning for Cu cleaned with (a)
Citranox, (b) Ozonate-water
Cu-X : Copper cleaned with Citranox
H2
1.0E-08
1.0E-09
1.0E-07
CO2
CO
C
CH3
0.01Ah
0.1Ah
1Ah
1.0E-10
1.0E-11
•
45
PARTIAL PRESSURE (Arbi. Unit)
PARTIAL PRESSURE (Arbi. Unit)
1.0E-07
Cu-O3 : Copper cleaned with Ozonate water
CO2
H2
1.0E-08
1.0E-09
CO
C
0.01Ah
CH3
0.1Ah
1Ah
1.0E-10
1.0E-11
2 12 13 14 15 16 17 18 27 28 29 32 44
2 12 13 14 15 16 17 18 27 28 29 32 44
MASS (a.m.u.)
MASS (a.m.u.)
Pressure rises (dP/I) of H2, CH4, CO, CO2 are lower for Cu (Ozonate water cleaning)
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Yield of PSD with Beam cleaning
Comparison of Al and Cu samples
Aluminum samples
Copper samples
1.0E-02
1.0E-02
Ethanol machining only
Citrinox
Chemical cleaning
20ppmO3-30min
6.7ppm O3 cleaning
YIELD of PSD, η (molecules/photon)…
YIELD of PSD, η (molecules/photon)...
30ppm O3 cleaning
1.0E-03
1.0E-04
1.0E-05
1.0E-03
1.0E-04
1.0E-05
1.0E-06
1.0E-06
0.001
0.001
0.01
0.1
1
10
0.01
0.1
1
BEAM DOSE (Ah)
BEAM DOSE (Ah)
•
•
46
η < 5 × 10-5 molecules/photon at beam dose of 0.1 A·h (for ozonate water cleaned samples)
η of Cu (ozonate water cleaning) < η of Cu (citranox cleaning)
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
10
Summary of cleaning effect for Al and
Cu cleaned with Ozonate water

Comparison of η(PSD) :
 η(Al) & η(Cu) (Ozonate water cleaning) < η(Al) & η(Cu) (Ordinary cleaning methods)

Partial pressure of outgas from PSD for Cu :
 Major outgas of H2 ~ CO ~ CO2 > CH4 : Cu-O (Ozonate water) < Cu-X (Citranox)

Surface analysis for Cu by SIMS
 Negative ions of CxHy, Cl : Cu-O < Cu-X
 Positive ions of Na, K, Ca : Cu-O < Cu-X

The lower carbon resided on surface layers results in the lower yield of PSD.

Ozonate water cleaning method provides lower carbon on surface layer than other cleaning methods.

Recommend :
 OFHC Cu with Ethanol CNC machining can be cleaned with ozonate water.
 OFHC Cu with oil machining can be cleaned with acid (Citranox etc.) prior to ozonate water (rather than DI water)
rinsing.
 Next experiments will analyze the Cu samples with various cleaning methods after vacuum baking.
47
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Try to understand the surface condition
after various cleaning methods
Ethanol
machining
Ozonate water
cleaning
Chemical
cleaning
Work function (mV)
Surface Roughness (µm)
Thickness of Oxide layer (nm)
48
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
Courtesy: Ivan Liu, Master thesis of NTHU,Taiwan, Sep. 12 (2006)
G.Y. Hsiung
A super dry N2 exposure to 1 atm for the baked Al
chamber (Ethanol CNC + 20 ppm O3-water cleaning)
recovers to the UHV in 10 Hours (non-bake)
• q ~ 2 E-13,
• P1 < 2 nTorr
at 10 hours after
pumping (before
vacuum baking)
49
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung
Conclusions
 Ozonate water cleaning process after oil-free ethanol machining
provides the lowest thermal outgassing rate and the least yield of the
photon stimulated desorption for both aluminum and copper samples.
 The surface analysis by SIMS or AES shows the less contaminants of
C, Na, K, etc. resided on the surface after ozonate water cleaning
which results in the lowest PSD yield of carbonaceous outgases such
as CO, CO2 and CH4.
 The work function and surface roughness have been measured on Alsamples for various cleaning methods. The machined surface provides
thinner oxide layer and less roughness which gives lower rate of
surface desorption.
 A further cleaning with ozonate water is helpful for removing the
residual carbon on the surface layer and the reduction of carbonaceous
molecular desorption.
Thanks for your attention.
50
IUVSTA WS-63, Avila, Spain, Sep. 14-19, 2010
G.Y. Hsiung