Recoil Mass Spectrometer (RMS)

Recoil Mass Spectrometer (RMS)
A Recoil Mass Spectrometer Set-up calculator is now available.

The recoil mass spectrometer (RMS) is a zero degree device used to separate masses produced in nuclear reactions. The spectrometer [1-6] is 25 m long and is comprised of a momentum separator for beam rejection and the traditional electric-magnetic-electric dipole mass separator [7-10]. Other recoil separator devices may be found in ref. [11-13].

Fig 1 - A schematic sketch of the RMS. Detectors at the target and focal plane positions may be used to achieve high selectivity of reaction products. Primary beam is usually deposited in the RMS at positions marked with 1 and 3 and in spectrometers without a momentum separator [7-10] at positions 2 and 3. The superior beam rejection capability of the RMS is due to the additional focussing elements between positions 1 and 2. Under special circumstances, the amount of beam reaching position 3 may be reduced through the use of the finger system inside Q3. Photographs of the RMS are available.

The RMS consists of three magnetic dipoles (D), seven quadrupoles (Q), two sextapoles (S), and two electric dipoles (ED). The physical properties of each element may be found in ref. [4]. The first two quadrupoles gather the recoiling nuclei and determine the momentum focus which is located inside Q3. The first magnetic dipole separates by momentum the recoiling nuclei from the beam. The primary beam, originating from a tandem accelerator, has a well-defined momentum. When using very thin targets, this well-defined momentum can remain the beam should be focussed spacially according to its charge-state distribution. The recoils have a large momentum distibution caused by the evaporation process and will thus, fill the available space. Small rods called fingers may be inserted through the split poles of Q3 to intercept the primary beam at its focus and yet, have minimal impact on the overall transport efficiency for recoil products. When used with inverse kinematic reactions where beam particles and recoils have similar rigidities within the acceptances of the RMS, the fingers are designed to provide enhanced primary beam rejection. This feature has never been used because the targets we use are typically 0.3 - 1.0 mg/cm2 and the momentum spread is large.

The second half of the momentum separator, D1-Q4-Q5, provides a second focus which is the object of the mass separator portion of the spectrometer. This position, called the achromat, is energy independent when Q3 is correctly adjusted. The two sextapoles provide second-order corrections in the final focus.

Fig. 2 - A photograph of one of the electric dipole titanium plates. The gap is 10 cm and white cermics support the plates inside the vacuum vessel which is typically maintained at ~3x10-8 Torr. The plates can hold voltages greater than +/- 150 kV although our normal operating voltages are less than 100 kV.

The mass separator section contains a magnetic dipole between two electric dipoles. The electric elements separate the recoils as a function of kinetic energy and charge (E/Q) and the magnetic dipole separates as function of momentum and charge (P/Q). The net result are recoils separated as a function of mass and charge (A/Q). The magnitude of the mass dispersion is determined by the last pair of quadrupoles, Q6 and Q7. Together with Q4 and Q5, the position of the final focal plane is determined.

The performance parameters of the spectrometer, such as the size of the final image and the mass dispersion, often depend only upon specific groups of elements of the RMS. These elements may be adjusted together without affecting the field values of the other elements. Thus, several "knobs" have been incorporated into the control system so that users may adjust these parameters according to the requirements of the experiment.

Various acceptances govern the performance of the RMS and the efficiency to detect recoil products. The RMS has an nonsymmetric, overall solid-angle acceptance. The horizontal acceptance is +/- 30 mrad and the vertical acceptance is +/- 110 mrad and is determined by the positions of Q1 and Q2. The target-to-Q1 distance is approximately 73 cm. However, other aperatures in the system affect the effective overall solid angle which corresponds between 10- and 15 msr. The primary beam spot on target should be a round or vertical line 1 mm in width and 4 mm in length. The energy acceptance is determined, in part, by the length of the plates on the electric dipoles and is measureded to be approximately +/- 10%. The A/Q acceptance, reflecting the various aperatures throughout the flight path is 4.9% in diverging mode and closer to 4.1% in converging mode. The mode is determined by changing the polarity of Q6 and Q7 and reflects whether the separated masses converge or diverge after the focal plane. Converging mode is our standard mode as it has a smaller focal plane (~15 cm instead of 35 cm) and is useful for some decay experiments as two charge states of the same mass can be focussed on to a single large area silicon detector.

Overall mass resolution (M/δM) has been calculated to be 540 although through collimation, it is estimated that this value may approach 1000 (including software corrections). The RMS will typically have mass resolution on the order of 350 and values over 400 have been observed. In our experience, a mass resolution of greater than about 300 is sufficient for most experiments. The resolution usually depends on beam spot size and target thickness.

Fig. 3 - A typical focal plane image of the RMS using the reaction 60Ni + 58Ni at 220 MeV. The masses and resolution are given in the figure. This spectrum is taken with no other requirement than an ion passing through the focal plane detector. The high primary beam suspression of the RMS is evident from the cleanliness of this spectrum although a light spray of beam particles may be observed on the far left of the spectrum. You may access a higher resolution JPEG version of this figure.

The long flight-path of the spectrometer will prohibit detection of nuclei which contain isomeric transitions with halflives ranging from approximately ten nanoseconds to a microsecond. These nuclei decay in flight through internal conversion, causing disruption [14] of the atomic electrons and drastic changes in charge state. In order to reduce this effect, a charge resetting foil is placed some 10 cm away from the target and charge state equilibrium is obtained for all recoils. Due to the gaussian charge state distribution, the maximum total efficiency of the RMS for each mass will be on the order of 45%. This assumes that two charge states are detected at the focal plane. Note that for light ions (A<60) one charge state positioned on the central axis is usually most efficient.

There is a relationship between charge state and energy acceptances of the RMS. All values in this report are quoted for the central ion. Because the RMS is an electromagnetic device it is governed by the magnetic rigidity equation and contrained by the electric dipoles to specific energies. Thus, K = (M/Q)(E/Q) where M is the mass, E is the energy, and Q is the charge state. The RMS acceptance in terms of K is +/-10% for the central ion. In order to detect two charge states of the same mass, (M/Q) has a range of values and hence, (E/Q) must have a range of values. So if (M/Q) is to the left of center (low M/Q), the ions focussed there have higher energy on average than those focussed to the right of center (high M/Q). But since the energy is constrained by the central setting to +/-10%, off axis ions cannot have the full energy acceptance of +/-10%.

When planning for experiments, we are often asked to estimate the overall detection efficiency. This is difficult because it is extremely reaction dependent. It depends on:

  • charge state fraction (number of charge states detected)
  • target thickness (energy spread, multiple scattering)
  • "normal" or inverse kinematics (solid angle acceptance)
  • mass region or type of reaction
Because this device is primarily designed for fusion-evaporation reactions using targets of about 0.5 mg/cm2, a detection efficiency of about 5% for pure nucleon evaporation channels is a good rule of thumb. Alpha channels will be lower although not as bad as comparable devices.

A list of focal plane detectors and other relevent features of the RMS is given below:

RMS hardware and other features
Set-up calculator
target area chambers
charge resetting foil
rotating target (pdf)
RMS target area detectors
CLARION Ge array
HYBALL CsI array
neutron detectors
micro-channel plate detectors (MCP)
RMS focal plane area detectors
moving tape collector (MTC)
position sensitive avalanche counter (PSAC)
micro-channel plate detectors (MCP)
focal plane
ionization chamber
double-sided silicon strip detectors (DSSD)
BESCA - pair spectrometer
3He neutron detectors

Note that for certain RIB experiments such as Coulomb excitation, the RMS is not usually used. However, the 0o exit port of the first dipole can hold a Bragg detector which can be used to sample the beam which enters it. Note that its distance and optional collimation permits only sampling of the beam. However, this is sampling can determine the beam composition throughout the experiment.

This device is funded by the Georgia Institute of Technology, Idaho Nuclear Engineering Laboratory, Louisianna State University, Oak Ridge Associated Universities, Oak Ridge National Laboratory, State of Tennessee, UNISOR, University of Maryland, University of Tennessee, U.S. Department of Energy, and Vanderbilt University.


  1. Performance of the Recoil Mass Spectrometer and its Detector Systems at the Holifield Radioactive Ion Beam Facility, C. J. Gross, et al., Nucl. Instrum. Methods Phys. Res. A 450, 12 (2000). (pdf, postscript)
  2. The New HRIBF Recoil Mass Spectrometer - Performance and First Results, T. N. Ginter, Ph.D. Thesis, Vanderbilt University, unpublished. (pdf version available).
  3. A Recoil Mass Spectrometer for HHIRF Facility, J.D. Cole, T.M. Cormier, and J.H. Hamilton, Exotic Nuclear Spectroscopy, edited by W.C. McHarris, (Plenum Press, New York, 1990), p. 11.
  4. A Large-solid-angle High-resolution Recoil Mass Spectrometer Optimized for Use with GAMMASPHERE, T.M. Cormier, J. Cole, J.H. Hamilton, and A.V. Ramayya, Nucl. Instrum. Methods Phys. Res. A 297, 199 (1990).
  5. A Recoil Mass Spectrometer for HHIRF, J.D. Cole, T.M. Cormier, J.H. Hamilton, and A.V. Ramayya, Nucl. Instrum. Methods Phys. Res. B 70, 343 (1992).
  6. Experimental Apparatus for the ORNL RIB Facility, P.F. Mantica, Nucl. Instrum. Methods Phys. Res. B 99, 338 (1995).
  7. Performance of a Recoil Mass Spectrometer, T.M. Cormier, M.G. Herman, B.S. Lin, and P.M. Stwertka, Nucl. Instrum. Methods, 212, 185 (1983).
  8. S. Spolaore, et al., Nucl. Instrum. Methods A 238, 381 (1988).
  9. C.N. Davids, et al., Nucl. Instrum. Methods Phys. Res. B 70, 358 (1992).
  10. Heavy Ion Reaction Analyzer (HIRA): A Recoil Mass Separator Facility at NSC, A.K. Sinha, N. Madhavan, J.J. Das, P. Sugathan, D.O. Kataria, A.P. Patro, and G.K. Mehta, Nucl. Instrum. Methods Phys. Res. A 339, 543 (1994).
  11. Microsecond Mass Separation of Heavy Compound Nucleus Residues Using the Daresbury Recoil Separator, A.N. James, T.P. Morrison, K.L. Ying, K.A. Connell, H.G. Price, and J. Simpson, Nucl. Instrum. Methods Phys. Res. A 267, 144 (1988).
  12. Construction and Initial Operation of MARS, R. Tribble, J. Bronson, H. Dejbakhsh, C. Gagliardi, S. Hale, W. Lui, D. Semon, H. Xu, S. Yennello, and X. Zhou, Texas A&M Progress Report 1992-1993, p. V-136.
  13. A Recoil Separator for Use in Radioactive Ion Beam Experiments, M.S. Smith, C. Rolfs, and C.A. Barnes, Nucl. Instrum. Methods Phys. Res. A 306, 233 (1991).
  14. Internal Conversion and Evolution of Nuclear Structure at Very High Spin: 154-158Er, T.M. Cormier, P.M. Stwertka, M. Herman, and N.G. Nicolis, Phys. Rev. C 30, 1953 (1984).

For questions about this page please contact the HRIBF User Liaison.

This file last modified Friday August 22, 2008