The Recoil Mass Separator Project at Notre Dame

Larry Lammfor the Notre Dame RMS Design Team

SNEAP 2008

Outline

• Motivation – why do we need a recoil mass separator?

• Basic Design• Key Elements and Design Strategy• Status of the project• Expected timeline

This blast wave originated in the Cygnus Loop supernova, which occurred about 15,000 years ago.

“The surface of the Earth is the shore of the cosmic ocean. On this shore, we’ve learned most of what we know. Recently, we’ve waded a little way out, maybe ankle deep, and the water seems inviting. Some part of our being knows this is where we came from. We long to return. And we can, because the Cosmos is also within us. We’re made of star stuff. We are a way for the Cosmos to know itself.”

Carl Sagan, “Cosmos: A Personal Journey”, PBS TV, 1980

Nucleosynthesis

1957 Burbidge, Burbidge, Fowler, Hoyle

From H. Schatz, Nuclear Astrophysics course material, jinaweb.org

Radiative Capture Reactions

• Most stages of stellar evolution involve the radiative capture of protons and α particles, and many of the pertinent (p,γ) and (α,γ) reactions have been extensively studied by γ ray detection. However this method is limited at lower energies by the cosmic and the beam induced background.

Stellar Temperatures and Available Energy

• Solar Temp – surface ~ 6000 K• – core ~ 15 x 106 K (T6 = 15)• Maxwell-Boltzman distribution, with energy

peaked around 3/2 kT, where k= 8.62 x 10-5 ev/K

• Core energy ~ 2 keV• Only p-p chain reactions• VERY LOW energy by laboratory

standards

Stellar Temperatures and Available Energy

• But many other stellar scenarios to consider

• Novae, Supernovae – much hotter• Energies of hundreds of keV to a few MeV• Still low by laboratory standards

S(E) extrapolation for the 6.79 MeV excited state in the 14N(p,γ)15O reaction

S FactorLimited by cosmic and

beam induced

backgrounds

S(E) for the 6.79 MeV excited state in the 14N(p,γ)15O reaction

Reaction rate differs from extrapolation by 2 orders of magnitude at certain energies!!!

Must measure as low as possible!

S Factor

Recoil Mass Separator• One way to try to measure at very low energies

is to reduce the inherent background by detecting the γ rays in coincidence with the reaction products, but this necessitates separating the recoils from the beam using a recoil mass separator (RMS) in inverse kinematics. This is an incredibly difficult task, since for typical reactions, the ratio of beam to reaction products will be on the order of 1017. However, with proper design, an instrument capable of this level of separation can be built.

How much is a part in 1017?

• Lake Michigan – 5 x 1015 liters• 50 ml out of Lake Michigan!

Heavy water in Lake Michigan?

122 x 106 liters !!!

(5 x 106 liters of D217O) !!!

Recoil separators for radiative capture studies

• DRAGON, ERNA:– Two very successful separators designed for specific

benchmark reactions• 15O(α,γ)19Ne for DRAGON• 12C(α,γ)16O for ERNA

• ND design:– Broad range of reactions: A<40– Large acceptance: ∆θ<±40 mrad, ∆E<±7.5%

Notre Dame recoil separator: Design parameters

9 mrad0.97 deg.

1.8 %1.25 MeV12.5 MeV

36Ar(α,γ)40Ca

32 mrad1.8 deg.

6.5 %460 keV3. MeV

22Ne(α,γ)26Mg

40 mrad2.3 deg.

7.4%360 keV2. MeV

18O(α,γ)22Ne

∆θ (mrad)(deg)

∆E/E (%)ECM Ebeam

Reaction

Stable beam from the KN (4MV) Van de Graaff acceleratorBeam intensity up to 100 µA (~1015 pps)

Beam mass < ~40 Acceptance

Sample of the list of reactions

St. George

Exis

iting

bea

m li

ne

Charge Selection

Mass Selection

Detection - ID

Jet gas target

KN accelerator

STrong Gradient Electro-magnetic Online Recoil separator for capture Gamma ray Experiments

Charge selection

• Multiple charge state after gas target– Selection of the most

abundant one (~40%)– RMS total efficiency is

charge state dependant– Clean rejection of the other

beam/recoil charge state• Charge selection can be a

source of background:– ∆Q/Q0 can be large

→ selection in two steps

26°

26°

Recoil and beam profile

24Mg(α,γ)28Si5+ @ 8 MeV

1+2+3+4+5+6+7+8+9+10+11+12+

Mass selection: Wien filter

• Crossed magnetic and electric field

• Particle with velocity = E/Bare not deflected

• Choice for flexibility

• Fields homogeneity andE/B constant are criticalfor mass selection

Detector development

Ion optic example

18O(α,γ)22Ne @ 1.94 MeV

24Mg(α,γ)28Si5+ @ 8 MeV

Magnet status• Several dipole and quadrupole prototypes are built and

mapped, still trying to reach our specifications.

Wien Filter Design• A Wien filter deflects all

particles except those with a velocity of E/B.

• Homogeneous fields are extremely important, and it is critical to maintain the E/B ratio in the fringe field region.

• The field clamp and the special design of the electrodes insure control of the fringe field.

Wien filter fields – transverse

Need good E and B field homogeneity in the center

|∆B/B|<0.2%

Magnetic field in the filter center

6.6 cm

5 cm

Y

X

Z=WFlength/2

100kV

∆E/E|<0.2%

6.6 cm

5 cm

Y

X

Wien filter magnetic fringe field

Without field clamp With field clamp

Z

ZZ

YY

Mobile field clamp tobe fine adjusted using thebeam

Wien filter electrostatic fringe field

Clamped magnetic field

E-field of standard WF

electrodes

E-field of optimized WF

Wien filter fringe fields - longitudinal

-5.0000E-01

0.0000E+00

5.0000E-01

1.0000E+00

1.5000E+00

-600 -500 -400 -300 -200 -100 0 100 200 300 400

Magnetic fieldElectric fieldE/B ratioFigure of merit

Modified Wien filter fringe field

Aberrations correction

Optic calculation up to 4th orderCorrections up to 3rd orderembedded in the magneticdipoles pole faces.

The goal is to minimize spotsize at the mass selectionslits

Horizontal projection

Detection• Now that the charge and mass selection

is done:– Detection … – Yes but: Background from beam scattered and

charge exchanged

Charge selection

Mass selection

Additional cleaning

Detection system

Particle ID to have additional selection

Tof vs Energy (Micro-channel plate + Si)Ionization chamber

Localize eventual background trajectories

• In other recoil separators dedicated to radiative capture, background is ~10-12 relative to initial beam intensity– Difficult to simulate:

• Enough particles• All the effects (scattering and charge exchange) at every

position

• Procedure to identify what part of the beam phasespace reaches the focal plane detector– Backtracking of beam particle filling the face space

Possible background trajectories

18O(α,γ)22Ne4+ @ 2.MeV

18O4+ @ 2.±0.08MeV

Impact on the NSL

• Significant infrastructure improvement– Electric power

• increase available power in the lab– Water cooling

• de-ionized chilled water– Air conditioning

• Temperature stability in the target room– Concrete block moving

Reduced availability of the KN/JN target room

RMS Infrastructure

RMS Infrastructure

RMS Infrastructure

Dry Coolers on Roof

Timeline?

• Who knows?• Bruker is far behind schedule, but we hope

to have magnets by early spring 2009.• Installation and commissioning during the

summer and fall.• New accelerator? New tower for new

accelerator? When? Before RMS is finished?