Vacuum System Considerations for Ion Source and RFQ of the Proton Driver

1
Vacuum System Considerations for Ion Source and
RFQ of the Proton Driver

• Outline
• 1. Analysis of vacuum requirements
• for the Ion Source
• 2. Preliminary look at expectations
• for the RFQ vacuum
• 3. Pumping stations assignment and
• location (a sketch)

2
Vacuum Requirements for Ion Source of the Proton
Driver

• Ion Source vacuum system counters two sources of
  gas
• - Non-ionized H2 gas feeding the
  Ion Source
• - Out-gassing from the surfaces of
  the Ion Source
• system (possible contamination
  of ion beam)
• Consequently, there are 2 vacuum states of
  considerably
• different strength (by order of magnitudes) to be
  considered
• in evaluating the required pumping speed for the
  Ion Source

3
Ion and Current Production from H2 Gas

• - Number of H2 molecules per 1 cm3 _at_ 760
  Torr
• 1 mol (H2) 2 g
• H2 density 9 E-5 g/cm3
• Number of molecules/cm3 ½ x 9 E-5
  x 6.022 E23
• 
  2.7 E19
• - Theoretical charge/current produced
• If all H2 ? 2 H (or 2 H-),
• Maximum charge gt 2.7 E19 x 1.6E-19 C
  4.3 C/cm3
• - For 1 cm3/sec gas flow, maximum current gt
  4.3 A
• - Duty factor ion production inefficiency gt
  lt 0.001
• H/H- currents gt few mA

4
Ion Source Vacuum Requirements

• To minimize the scattering of produced ions the
  vacuum in the ion production mode is typically
  kept at
• lt 10 mTorr
• A flow of 10 cm3/sec _at_ 760 Torr is equivalent of
  the
• out-gassing rate of 7.6 Torr_L/s,
• The pumping speed of 1000 L/s will keep vacuum
• level at 7.6 mTorr
• Most Ion Sources use 3,000 L/s of total pumping
• speed which probably gives an effective speed
  at the source in the range 1000 L/s

5
Ion Source Vacuum Requirements

• Sketch of the Duoplasmatron Source Assembly
  of the Accelerating Column
• 
  of the Duoplasmatron Source at MS6

6
Ion Source Vacuum Requirements

• Typical Ion Source has inner surface area
  10,000 cm2
• There are multiple large O-rings
• System is un-bake-able, so H2O is a dominant
  source of out-gassing with a rate of 3 x
  10-10 Torr_L/s_cm2
• giving a total H2O out-gassing of 3 x 10-6
  Torr_L/s
• With an effective pumping speed of 1000 L/s, the
• projected vacuum level with no H2 gas feed
  is
• 3 x 10-9 Torr, or
• 1 ppm of the fed H2 gas
• As the H2 gas purity is also 1 ppm, we have a
  match!

7
RFQ Vacuum Considerations

• The RFQ is a massive, large object primarily made
  out of copper, with multiple brazed joints and
  multiple function ports, and very few (2 ?)
  vacuum ports
• There are two options considered for the RFQ
  vacuum system
• - RFQ is embedded inside a large vacuum
  chamber
• - The pump-out ports are directly on the
  RFQ body
• The required vacuum level is 10-8 Torr, or
  better

8
RFQ Vacuum Considerations
RFQ Inside Look
RFQ Outside Look
9
RFQ Vacuum Considerations

• RFQ inner surface area (Cu) 40,000
  cm2
• RFQ outer surface area (Cu), est. 35,000
  cm2 Vacuum chamber area (SS), est. 40,000
  cm2
• For an un-baked system the out-gassing is
  dominated by H2O, at a rate of 3 x 10-10
  Torr_L/s_cm2.
• Therefore, the estimated out-gassing rate is
• (a) RFQ inside the vacuum vessel
• 115,000 cm2 x 3 x 10-10
  Torr_L/s_cm2 gt
• 3.5 10-5 Torr_L/s
• (b) Pump-out port(s) directly on RFQ
• 1.2 10-5 Torr_L/s
• To obtain a vacuum level of 10-8 Torr,
  the effective pumping speed must be
• (a) (3.5 10-5 Torr_L/s)/10-8 Torr gt
  3500 L/s for,
• (b) (1.2 10-5 Torr_L/s/10-8 Torr gt
  1200 L/s for,
• (that is when ignoring the pump -gt RFQ
  conductance)

10
RFQ Vacuum Considerations

• For a well-baked system
• (150 deg. for Cu, 300 deg. for SS _at_ 50h)
• the out-gassing is dominated by hydrogen at
  rate of
• 10-12 Torr_L/s_cm2 for
  Cu, and
• 5 x 10-13 Torr_L/s_cm2 for SS
• leading to an estimated total out-gassing
  rate of
• (Cu) 75,000 cm2 x 10-12 Torr_L/s_cm2
  
• (SS) 40,000 cm2 x 5 x 10-13
  Torr_L/s_cm2
• 10-6 Torr_L/s
• So, for the RFQ mounted inside the vacuum
  chamber, the required pumping speed would be only
  100 L/s.
• This is unlikely, in practice it is not
  possible to reduce that much the out-gassing of
  H2O, but it indicates that with a moderate baking
  (part of brazing process?), and a few 1000 L/s
  pumps, a 10-8 Torr vacuum is feasible

11
RFQ Vacuum Considerations

• In summary, we conclude that with the appropriate
  heat
• treatment of RFQ assembly to lower the
  out-gassing rate of H2O, and with the uhv-type
  cleaning of the RFQ vacuum associated surfaces,
  the vacuum level of lt 10-8 Torr is probably
  feasible for both systems (with or without vacuum
  vessel)
• The spill of the residual H2 gas from the Ion
  Source to the RFQ probably scales down with the
  Ion Source to RFQ beam pipe conductance, which we
  estimate as (0.1-1) .
• So the spill is (0.1-1) x 3 x 10-6
  Torr_L/s, or
• lt 3 x
  10-8 Torr_L/s ,
• and so it can be intercepted at the Ion
  Source RFQ beam pipe using a modest (Turbo or
  Ion) pumping station

12
Ion Source and RFQ Vacuum System