TIGER Science*

1.0 Introduction

The Trans-Iron Galactic Element Recorder (TIGER) experiment was selected in 1998 as the science payload for flight aboard the first Ultra-Long Duration Balloon (ULDB), a new NASA mission category. In 1997 we flew TIGER on a 23-hour balloon flight in which we demonstrated that TIGER possessed the required charge resolution, and, if flown for ~100 days, had sufficient collecting power to measure the elemental abundances of nuclei with 26 < Z < 40. This led to the selection of TIGER for the first ULDB flight payload. Very recently (3/01) the failure of two ULDB test balloons has resulted in a likely 2 year delay of the ULDB program. To enable us to achieve the science planned for TIGER on a shorter time scale than is likely for the ULDB, we propose a research program based on two LDB flights from the Antarctic. On the first flight (and possibly the second) we are planning for 2 revolutions around the Antarctic. The combination of data from these two flights will enable us to achieve science that is comparable to that we would have obtained from a single ULDB flight.

Previous measurements of UH cosmic rays on HEAO-3 (Binns, et al., 1989) and Ariel-6 (Fowler, et al., 1987) had sufficient charge resolution to discern peaks in the charge spectrum for elements of even-Z (for Z < 60). However they were incapable of measuring elements of odd Z, which are generally significantly less abundant than their even-Z neighbors. TIGER has been demonstrated to possess charge resolution of s < 0.25 charge units (cu). With s < 0.25 cu, distinct charge peaks will be clearly resolvable at every element, including odd-Z elements whose abundance is only one-fifth that of both adjacent even-Z elements. A key measurement is the abundance of 37Rb, which is expected to be at least 0.2 of that of adjacent even-Z elements for any plausible cosmic-ray source propagated to earth. With s < 0.22 cu odd-Z elements would be resolvable with abundances as low as six percent of an adjacent even-Z element, as is to be expected for the most difficult to measure odd-Z UH elements in our range of interest, 31Ga and 33As.

With two LDB flights, one near solar maximum and one during the transition to solar minimum, and flying from the Antarctic instead of New Zealand which was the plan for the ULDB flight, the TIGER instrument will observe large enough numbers of nuclei to make important measurements of both even-Z and odd-Z elements with Z <= 40. While some nuclei will be identified with Z > 40, their numbers will likely be inadequate for definitive conclusions. As described below, significant astrophysical questions will be addressed by our results in the interval 30 <= Z <= 40. TIGER will also measure abundances of elements in the interval 14 < Z < 30. It is valuable to measure these lower-Z elements in the same instrument that is measuring the UH nuclei, so that the UH abundances and energy spectra can be correctly normalized to the well established abundances of these much more common elements.

Additionally, TIGER serves as an engineering model for the ENTICE instrument on the Heavy Nuclei eXplorer (HNX), a mission recently selected by NASA for a SMEX mission concept study in which Washington University is the PI institution and all TIGER investigators are Co-Investigators. In the HNX mission, the ENTICE experiment will measure every UH element, even-Z and odd-Z, up through the heaviest stable element, 83Bi, and the second HNX instrument, ECCO which utilizes passive glass track detectors, will measure the abundances of nuclei with 74 < Z < 110. The ECCO experimenters have demonstrated charge resolution characterized by s = 0.45 cu in the interval 75 < Z < 83 (TREK on MIR, Westphal et al., 1998) and it is expected that ECCO on HNX will have improved charge resolution of < 0.35 cu. Thus we see that the TIGER and TREK experiments are the forerunners of a mission that is capable of definitively completing measurements of the galactic cosmic ray elemental abundances of all elements in the periodic chart.

In addition to measuring the nuclear charges of the incident cosmic rays, the combination of scintillation detectors, plastic Cherenkov detector, and aerogel Cherenkov detector will permit TIGER to measure the energy of the incident nuclei in the energy (E) interval 0.3 to about 10 GeV/nucleon. Since the abundance ratio of secondary elements (produced by interstellar fragmentation of primaries) to primary elements decreases with increasing energy above about 1 GeV/nucleon, we may expect to see some variation of the relative element abundances with energy, although small numbers at the higher energies will limit the precision with which we can measure such variation.

2.0 Scientific Objectives

Among the most fundamental astrophysical problems is understanding the mechanism by which particles are accelerated to the enormous energies observed in the cosmic rays. That problem can be conveniently divided into two questions: (1) What is the source of the energy and the mechanism for converting the energy of that source into the energy of individual cosmic-ray nuclei? (2) What is the source of the material that is accelerated and the mechanism for injecting that material into the cosmic-ray accelerator?

There is a general consensus about the answer to the first of these questions, at least with respect to the bulk of the comic rays, those with energy per nucleus below about 1014 eV. For at least the past forty years the source of the cosmic-ray energy has been understood as supernova explosions (e.g., Ginzburg & Syrovatskii, 1964). For about the past twenty years the acceleration mechanism has been understood as first-order Fermi acceleration at supernova shocks (e.g., Blandford & Ostriker, 1978).

The answer to the second question is still quite uncertain. There are several ways of interpreting available data that lead to quite different models for the source of the material and its injection mechanism. With the TIGER instrument we expect to produce new data that will help to distinguish among these possible models. (The HNX mission is expected to produce further data that would firmly establish one or another of these models.)

2.1 FIP vs. Volatility

A primary clue to the answer of the second question is the comparison between the elemental abundances at the cosmic-ray source (CRS) (i.e., the observed abundances corrected for the effects of nuclear fragmentation during propagation through the interstellar medium) and the general abundances of elements in the solar system (SS). For about twenty years it has been noted (e.g., Cassé & Goret, 1978) that the CRS abundances relative to SS of elements of low first-ionization potential (FIP) are roughly four to eight times those of elements of high FIP. This observation in the galactic cosmic rays leads to a model in which the source material for cosmic rays is the atmosphere of stars, where photospheric temperatures of the order of 104K are found. An alternative model (Epstein, 1980; Cesarsky & Bibring, 1981) noted that most of the elements of low FIP for which CRS abundances had been determined were refractory, while those of high FIP were volatile, suggesting that the material of cosmic rays preferentially originated in interstellar dust. For many years the similarity between galactic CRS abundances and abundances in energetic solar particles, which could not have come from dust, was taken as support for FIP being the governing property, rather than volatility.

Recently, another look at the CRS abundances (Meyer, et al., 1997) combined with a model of shock acceleration in which interstellar grains are accelerated (Ellison, et al., 1997) have given support to the model in which the cosmic-ray fractionation is governed by volatility and the refractory elements are enriched in the cosmic rays because they sputter off of accelerated dust grains and thus are more easily injected and accelerated by supernova shocks.

To distinguish between FIP and volatility as the characteristic that governs CRS fractionation, we need measurements of elements that are present in the CRS in large enough abundances that their source abundances can be inferred from observations near earth and that break the general pattern in which low-FIP elements are refractory and high-FIP elements are volatile. Two of the very few such elements that have not been measured are 37Rb and 55Cs, both of which are in the low-FIP group, but neither of which is refractory. Previous UH cosmic-ray observations either did not have the charge resolution or the collecting power needed to determine the abundances of these odd-Z elements. TIGER will have sufficient collecting power and charge resolution to give a definitive measurement of the 37Rb abundance. A longer duration experiment in space, such as HNX, will similarly determine the abundance of 55Cs. Other elements which also help to break the FIP-volatility degeneracy which have been previously measured (Byrnak et al, 1983, George et al, 2000, Binns et al, 1989) but for which TIGER can be expected to provide measurements of interest are Cu, Zn, Ga, and Ge.

2.2 r-process and s-process

Even within the context of a cosmic-ray source enriched in injection from dust grains, there are still two competing models for the source of the accelerated nuclei. The model of Meyer, et al. (1997) and Ellison, et al. (1997) envisions dust grains accelerated from the ambient interstellar medium, while Lingenfelter, et al. (1998, 2000) envision the refractory elements in the cosmic rays as originating in high-velocity grains that condense out of the supernova ejecta. While supernova ejecta are expected to include products of pre-supernova burning, such as nuclei formed in the slow neutron-capture process (s-process nucleosynthesis), the ejecta are likely to be enriched in products of the rapid neutron-capture process (r-process nucleosynthesis) which occurs in supernova explosions. Lingenfelter, et al. point to the apparent excess of r-process elements around 78Pt (Binns, et al., 1989) as support for their model of cosmic-ray injection from supernova ejecta, while Meyer, et al. explain the same element abundances as arising entirely from volatility-based fractionation of interstellar material.

While previous UH cosmic-ray measurements indicate that for Z<60 the cosmic rays appear to be a mixture of s-process and r-process elements similar to the mixture found in the solar system (Binns, et al., 1989), TIGER will provide a further test of this conclusion by measuring abundances of odd-Z elements that are expected to have a substantial primary component at earth and were previously unresolved -- 37Rb, a primarily r-process element, and 39Y, a primarily s-process element.

2.3 Cosmic-ray propagation in the galaxy

Any inferences about cosmic-ray source abundances that are derived from observations near earth require correction of the observed abundances for the effects of nuclear fragmentation as the cosmic rays propagate through the interstellar medium. The heavier the nuclei the shorter their interaction mean-free-path and thus the more important it is to understand the propagation. To date, inferences about UH cosmic rays have mainly used models that had been derived by comparing abundances of secondaries and primaries among the lighter (Z < 26) elements. (Those models were shown by Binns, et al. (1987) to be generally consistent with the abundances of groups of heavy secondaries, 62 < Z < 69 and 70 < Z < 73, but work with charge groups is less definitive than that with individual elements.)

2.4 Solar flare nuclei in the interval 10 < Z < 30

With the first flight occurring during solar maximum, there is a chance that during the 100-day balloon flight a rare highly energetic solar flare could occur. The atmosphere will shield our instrument from the very abundant flare particles with energy below several hundred MeV/nucleon, but with our large detector area we could measure the fluxes of more energetic heavy nuclei. Because of the rarity of such high-energy flares, the objective of measuring such energetic flare particles is definitely secondary to our UH objectives, but if such a flare should occur during our flight, our instrument would provide important new data.

* Taken from the TIGER proposal