The Heavy Ion Counter on the Galileo spacecraft will monitor energetic heavy nuclei of the elements from C to Ni, with energies from ~6 to ~200 MeV/nucleon. The instrument will provide measurements of trapped heavy ions in the Jovian magnetosphere, including those high-energy heavy ions with the potential for affecting the operation of the spacecraft electronic circuitry. We describe the instrument, which is a modified version of the Voyager CRS instrument.
The Heavy Ion Counter (HIC) is included on the Galileo spacecraft primarily for the purpose of monitoring the fluxes of energetic heavy ions in the inner Jovian magnetosphere and high energy solar particles in the outer magnetosphere in order to characterize the ionizing radiation to which electronic circuitry is most sensitive. The measurements performed will also be of scientific interest, since the instrument's large geometry factors and extended energy range will provide spectral information for ions from C (Z=6) to Ni (Z=28) with energies of ~6 to ~200 MeV/nucleon. In this article we will concentrate on Jovian magnetospheric science. We review here previous scientific results concerning trapped high-energy heavy ions, describe anticipated new findings with Galileo, and provide a brief instrument description.
2. Previous Results
During the Voyager encounters with Jupiter, it was discovered [Krimigis et al., 1979a, b; Vogt et al., 1979a, b] that a major component of the trapped radiation in the inner Jovian magnetosphere is energetic heavy ions. The dominant heavy species inside ~10 Jovian radii (RJ) in the MeV/nuc energy range were found to be oxygen and sulfur, with sodium also present. This is illustrated in Figure 1 which compares the elemental composition between 4.9 and 5.8 RJ with the more normal solar-like composition found in the same energy range in the middle and outer magnetosphere. The abundances in the inner magnetosphere indicate that the source of the ions is the surface or atmosphere of Io.
The phase space density of the energetic oxygen and sulfur ions has a positive radial gradient (i.e., increasing outward) in the inner magnetosphere [Gehrels, Stone, and Trainor, 1981], implying that the diffusive flow at these energies is inward. On the other hand, the density of the oxygen- and sulfur-rich plasma in the inner magnetosphere is highest at the Io plasma torus and decreases outward [Siscoe et al., 1981]. Barbosa et al.  have proposed that charge exchange in the plasma torus produces fast neutrals which escape to the outer magnetosphere where ~0.2% are re-ionized by solar ultraviolet or electron impact and recaptured. Some of these newly created ions are subsequently energized by stochastic acceleration caused by magnetohydrodynamic waves, producing ions with large magnetic moments that adiabatically diffuse inward, some attaining energies in excess of 30 MeV/nuc in the inner magnetosphere.
As the energetic ions flow inward, they are lost from the magnetosphere, presumably by pitch angle scattering of the mirroring particles into the loss cone [see, e.g., Thorne, 1982]. The strength of this loss mechanism determines the rate at which the heavy ions precipitate into the Jovian atmosphere, exciting ultraviolet and x-ray auroral emissions. At the time of the Voyager 1 encounter in March, 1979, it appeared losses due to pitch angle scattering were occurring at nearly the maximum rate, resulting in the precipitation of ~10^^24 ions/s above ~70 MeV/nuc-Gauss, with an extapolation down to 10 MeV/nuc-Gauss suggesting the possibility of a loss rate of ~10^^26 ions/s and an auroral power of ~10^^13 W [Gehrels and Stone, 1983]. The evidence that major losses are occurring in the magnetic moment (energy) range appropriate to HIC is shown in Figure 2, where the total inward flow rate of oxygen ions with magnetic moments greater than several thresholds is given as a function of radial distance (L) from Jupiter. The number of inflowing ions first decreases inside 15 RJ and falls off sharply inside 10 RJ.
The curves in Figure 2 also illustrate that only a small fraction of the trapped particles survive to produce the trapped radiation at 5 RJ. As a result, changes in the rate of pitch angle scattering can result in significant changes in the flux of particles reaching 5 RJ. It is therefore important to better characterize not only the flux at 5 RJ, but also the nature and any long-term variability in the loss mechanism operative outside of 5 RJ.
3. Anticipated New Results
The HIC instrument should provide new information on the spectra of heavy ions at higher energies than previously possible. As illustrated in Figure 3, fluxes at these energies are based on extrapolations of spectra measured at lower energies and as a result are very uncertain.
Two different estimates of the expected fluxes of oxygen ions are shown in panel b for magnetic moments of >=3070 MeV/nuc-G and >=460 MeV/nuc-G, corresponding to particles with E >= 100 MeV/nuc and >=15 MeV/nuc at 5 RJ. The >=15 MeV/nuc flux was directly measured by the Cosmic Ray Science instrument (CRS) on Voyager 1 and is reasonably certain. The >=100 MeV/nuc flux, however, has been determined by extrapolation, and is therefore quite uncertain as indicated by the two different profiles. The dot-dashed line is the result of an assumption that J(>E) E^^4.3 which is based on a least-squares fit of a power law to Voyager 1 data in the energy interval from 7 to 20 MeV/nuc. Because of limited livetime and geometrical factor, Voyager 1 observed no particles with energies >=30 MeV/nuc, so a much softer spectrum is also possible at higher energies as indicated by the dashed line corresponding to J(>E) proportional to E^^-8.5 . These two extrapolations differ by a factor of ~500 in the predicted flux of oxygen with E >= 100 MeV/nuc at 5 RJ.
Measurements with the HIC instrument will significantly reduce the uncertainty in the high energy fluxes. As shown in panel a, the energy of a 3070 MeV/nuc-G oxygen ion is within the HIC energy range into 5 RJ. The geometry factor for E > 20 MeV/nuc oxygen ions (inside 8.5 RJ for 3070 MeV/nuc-G ions) is ~4 cm^^2 sr compared with 0.88 cm^^2 sr for the Voyager CRS instrument. More importantly, the livetime for measuring oxygen and sulfur ions with E >= 50 MeV/nuc will be essentially 100% because a polling priority system (see instrument description section) will discriminate against the large fluxes of protons and electrons that dominate the CRS analysis rate. As a result, a flux of only 10^^-5 ions per cm^^2 s sr should produce 1 analyzed event in 10 hours, the time that Galileo is inside 7 RJ
Figure 3 indicates that the oxygen ions which have >=100 MeV/nuc at 5 RJ have >=10 MeV/nuc at 10 RJ. Thus, as Galileo makes repeated orbital passes in the vicinity of Europa, it will be possible to monitor the fluxes of particles with M >= 3000 MeV/nuc-G throughout the mission. Such long term information is especially important, since there are a number of reasons why the fluxes might vary. For example, the flux of high energy ions could be measurably affected by changes in the density of the Io plasma torus which is the source of the escaping neutral atoms and by changes in the solar ultraviolet which re-ionizes the neutrals in the outer magnetosphere. The HIC instrument will measure any time dependence of the energetic heavy ion fluxes and permit correlative studies of any associated changes in auroral emissions or diffusion processes.
Figure 3 also illustrates that the fluxes at 5 RJ depend strongly on the loss processes occurring inside of ~15 RJ. It is postulated that the losses are due to pitch angle scattering of the mirroring particles into the loss cone, and that the rate of scattering is close to the strong pitch angle diffusion limit in the inner magnetosphere [Thorne, 1982; Gehrels and Stone, 1983]. At this limit there is sufficient pitch angle scattering to refill the loss cone as rapidly as particle precipitation empties it, resulting in a nearly isotropic pitch angle distribution. Since the HIC detectors will view nearly perpendicular to the spacecraft spin axis, essentially complete coverage of the pitch angle distribution will be possible. Investigation of any time dependence of the loss process will also be possible as Galileo repeatedly passes through the radial range between 15 and 10 RJ where the radial profiles in Figure 2 indicate that significant losses occur.
4. Instrument Description
The Galileo Heavy Ion Counter (HIC) consists of two solid-state detector telescopes called Low Energy Telescopes or LETs. Use of these two telescopes over three energy intervals provides the geometry factor and energy range necessary to determine the fluxes of the heavy, penetrating radiation to which solid state memories are most sensitive. Heavy collimation and high discrimination thresholds on all detectors provide the necessary immunity to accidental coincidences from the large proton background. Three-parameter analysis provides additional rejection of background.
Two separate telescopes are included in order to cover a wide energy range while minimizing pulse pileup through the optimum selection of detector and window thicknesses. The LET E is optimized for the detection of nuclei with energies as high as 200 MeV/nuc, requiring thicker detectors. Thick windows protect this system from low energy proton pileup, but also exclude lower energy oxygen and sulfur nuclei. The second telescope, LET B, has a substantially thinner window so that it can detect lower energy nuclei (down to ~6 MeV/nuc), especially in the outer magnetosphere. The properties of the two telescopes are indicated in Table 1 and discussed below.
Detector Radius Thickness Threshold (cm) (µm) (MeV) LB1 0.95 32.1 0.3 LB2 0.95 29.6 0.4 LB3 1.13 421. 3.7 LB4 1.13 440. 2.0 LE1 0.95 30.4 9.3 (1.4)* LE2 0.95 33.4 2.0 (0.3)* LE3 1.13 463. 25. (5.0)* LE4,5 1.66 ~2000 117. (23.)* * High Gain
The LET B telescope is shown in Figure 4. In this telescope ions which penetrate LB1 and LB2 and stop in the thicker LB3 detector are analyzed. Detector LB4 is used in anti-coincidence. (If desired, the command system can be used to allow analysis for ions which penetrate LB1 and stop in LB2, or for ions which stop in LB1. Also detector LB4 can be turned off.) Lighter nuclei (especially hydrogen and helium) are rejected by "slant" discrimination with a weighted sum of the signals from the front three detectors,
The thin window (25 µm Kapton) shown in the figure serves for thermal control and for protection from sunlight. All detectors are surface-barrier type. An important feature is the use of keyhole detectors for LB1 and LB2, which define the event geometry. The active area of these detectors excludes the nonuniform edge of the silicon wafer through the use of keyhole-shaped masks during the deposition of the Au and Al contacts.
This LET B telescope is an improved version of the Voyager CRS LETs [Stone et al., 1977] which have demonstrated charge (Z) resolution of 0.1 charge units at oxygen under solar flare conditions [Cook et al., 1980] and 0.2 charge units at ~5 RJ, Voyager 1's deepest penetration of the Jovian magnetosphere [Gehrels, 1982]. Additional collimation and the thicker window decrease the HIC LET's response to background protons in the Jovian magnetosphere. The flux of protons in detector L1, for example, should be reduced by at least a factor of 10 from that observed on Voyager 1.
The higher energy LET E telescope is illustrated in Figure 5. In order to measure the flux of heavy nuclei at higher energies, LE4 and LE5 are each 2000-µm lithium-drifted detectors. The front detectors, LE1, LE2, and LE3, are identical to their LET B counterparts, providing spectral continuity and overlap. The collimator housing is relatively thick to provide background immunity and has a large opening angle to provide a large geometrical factor. The windows are 76 µm Kapton and 254 µm Al. As in LET B, low charge (low Z) events are recognized and rejected by a slant discriminator,
For intermediate energies, a narrow-angle geometry is defined by LE1 and LE2 for particles stopping in either LE2 or LE3. For the highest energies, where the fluxes are exceedingly small, measurements are made with the wide-angle geometry defined by LE2, LE3, LE4, and the collimator. LE1 is not required but its discriminator is recorded as a tag bit. LE5 distinguishes stopping and penetrating events. The maximum energy observed is determined by the LE2 discriminator threshold. This maximum is ~185 MeV/nucleon for oxygen as noted in Table 2. To allow detection of penetrating galactic cosmic-ray nuclei at higher energies in the outer magnetosphere, the LET E preamplifier gains can be increased by a factor of 5 to7 by command.
With the exception of spacecraft interface circuitry, all of the electronics were originally part of the Proof Test Model of the Voyager Cosmic Ray Science instrument (CRS), and additional detail may be found in Stone et al.  and Stilwell et al. . With the adjustment of amplifier gains and discriminator thresholds and the incorporation of thicker detectors, collimators, and windows, it has been possible to develop an instrument which is optimized for the measurement of high energy heavy ions trapped in the Jovian magnetosphere. Minor modifications to the logic allow the instrument to recognize events of the various types mentioned above, which are summarized in Table 2.
Event data are stored in buffers which are read according to a polling scheme which prevents domination of the telemetry by any one type of event. The buffer polling logic cycles through the five buffers listed in Table 2, reading out one each minor frame (2/3 second) and stepping to the next non-zero buffer on the subsequent minor frame. If a particular type of event occurs less than 0.3 times per second (i.e., is rare) then all of that type will be transmitted regardless of activity in other event types. If a particular type of event occurs more often than 0.3 times per second, it will be readout at least 0.3 times per second and more often if the other event buffers are empty.
Telemetry of counting rates and pulse height analyzed events is rapid compared to the nominal 3 rpm spin rate of the spacecraft; thus pitch angle distributions of the trapped radiation can be measured. Both telescopes have their axes oriented near the spin plane for this purpose. The time resolution of the HIC is in the range from 2/3 second to 2 seconds, implying an angular resolution in the range from 12° to 36°, which is to be compared to the telescope opening angles of 25° in narrow geometry mode and 46° in wide geometry mode.
Many of the functions of the coincidence logic and the buffering/readout scheme can be modified by command to optimize the instrument for changing environments or partial failures. As noted above, commands can also be used to change gain on the LET E preamps.
Name Requirement Geom Z Range Oxygen E Sulfur E Signals Telemetered LETB: LB1.LB2.LB3.!LB4 0.44 C to Ni 6 to 18 9 to 22 LB1, LB2, LB3 DUBL: LE1.LE2.!LE3 0.44 C to Fe 16 to 17 24 to 25 LE1, LE2 TRPL: LE1.LE2.LE3.!LE4 0.44 C to Ni 17 to 27 25 to 38 LE1, LE2, LE3 WDSTP: LE2.LE3.LE4.!LE5 4.0 C to Fe 26 to 46 37 to 70 LE2, LE3, LE4 WDPEN: LE2.LE3.LE4.LE5 4.0 C to Fe 49 to ~185 >=70 LE2, LE3, LE4+LE5 WDPENH: LE2.LE3.LE4.LE5.HG 4.0 Li to O 49 to ~500 - LE2, LE3, LE4+LE5 Geom is the geometrical factor in cm^^2 sr. Oxygen (Sulfur) E is the energy range in Mev/nuc for that element.
Barbosa, D. D., A. Eviatar, and G. L. Siscoe: 1984, J. Geophys. Res. 89, 3789.
Cook, W. R., E. C. Stone, and R. E. Vogt: 1980, Astrophys. J. (Letters) 238, L97.
Gehrels, N.: 1982, Energetic Oxygen and Sulfur Ions in the Jovian Magnetosphere, CIT Ph.D. Thesis.
Gehrels, N., E. C. Stone, and J. H. Trainor: 1981, J. Geophys. Res. 86, 8906.
Gehrels, N. and E. C. Stone: 1983, J. Geophys. Res. 88, 5537.
Krimigis, S. M., T. P. Armstrong, W. I. Axford, C. O. Bostrom, C. Y. Fan, G. Gloeckler, L. J. Lanzerotti, E. P. Keath, R. D. Zwickl, J. F. Carbary, and D. C. Hamilton: 1979a, Science 204, 998.
Krimigis, S. M., T. P. Armstrong, W. I. Axford, C. O. Bostrom, C. Y. Fan, G. Gloeckler, L. J. Lanzerotti, E. P. Keath, R. D. Zwickl, J. F. Carbary, and D. C. Hamilton: 1979b, Science 206, 977.
Siscoe, G. L., A. Eviatar, R. M. Thorne, J. D. Richardson, F. Bagenal, and J. D. Sullivan: 1981, J. Geophys. Res. 86, 8480.
Stilwell, D. E., W. D. Davis, R. M. Joyce, F. B. McDonald, J. H. Trainor, W. E. Althouse, A. C. Cummings, T. L. Garrard, E. C. Stone, and R. E. Vogt: 1979, IEEE Trans. Nuc. Sci. NS-26, 513.
Stone E. C., R. E. Vogt, F. B. McDonald, B. J. Teegarden, J. H. Trainor, J. R. Jokipii, and W. R. Webber: 1977, Sp. Sci. Rev. 21, 355.
Thorne, R. M.: 1982, J. Geophys. Res. 87, 8105.
Vogt, R. E., W. R. Cook, A. C. Cummings, T. L. Garrard, N. Gehrels, E. C. Stone, J. H. Trainor, A. W. Schardt, T. Conlon, N. Lal, and F. B. McDonald: 1979a, Science 204, 1003.
Vogt, R. E., A. C. Cummings, T. L. Garrard, N. Gehrels, E. C. Stone, J. H. Trainor, A. W. Schardt, T. F. Conlon, and F. B. McDonald: 1979b, Science 206, 984.
A special acknowledgment is due A. W. Schardt, whose untimely death prevented his co-authorship of this paper. This project was made possible by the development of the CRS instrument under the leadership of R. E. Vogt. W. E. Althouse (Caltech) and D. E. Stilwell (GSFC) have provided invaluable engineering and programmatic support and advice. We also are pleased to acknowledge the contributions of A. C. Cummings at Caltech; M. Beasley, W.D. Davis, J. H. Trainor, and H. Trexel at GSFC; and D. R. Johnson at JPL. This work was supported by a number of NASA grants and contracts.