PDS_VERSION_ID = PDS3 RECORD_TYPE = STREAM OBJECT = INSTRUMENT INSTRUMENT_HOST_ID = "NEAR" INSTRUMENT_ID = "GRS" OBJECT = INSTRUMENT_INFORMATION INSTRUMENT_NAME = "GAMMA RAY SPECTROMETER" INSTRUMENT_TYPE = "SPECTROMETER" INSTRUMENT_DESC = " Instrument Overview =================== The Near Earth Asteroid Rendezvous mission (NEAR) was successfully launched on 17 February 1996. NEAR was the first launch under NASA's Discovery Program, an initiative for small, low-cost planetary missions. As the first spacecraft to orbit an asteroid, the NEAR mission will address fundamental questions about the processes and conditions relevant to planetary formation. The X-ray/Gamma-ray Spectrometer (XGRS) is one of several instruments on-board the NEAR spacecraft that will study Eros during one year of orbital operations beginning in February 2000. The X-ray and gamma-ray spectrometers that comprise the XGRS are independent, but complementary experiments. In this paper we focus on the Gamma Ray Spectrometer, its design, operation, calibration, and procedures for interpretation of the measurements to be made at Eros. History ======= In February 1996, the Near Earth Asteroid Rendezvous (NEAR) mission was launched into space on a planned three-year cruise to the near-Earth asteroid, 433 Eros. The NEAR spacecraft, the first in the Discovery mission program to be launched, was built by the John Hopkins University, Applied Physics Laboratory (APL). The spacecraft will orbit Eros for one year and carries a group of remote sensing instruments including an X-ray and gamma-ray spectrometer system (XGRS). Due to a problem with firing of the main rocket engine in December 1998, the rendezvous was delayed until February 2000. The NEAR spacecraft will be the first spacecraft to orbit a body as small as Eros, which is estimated to be 33 km X 13 km X 13 km. From telescopic, radar, and other observations, it has been inferred that Eros is an S-type asteroid, one of the most common type of near-Earth asteroids. It is not known whether S-type asteroids come from differentiated or undifferentiated parent bodies. One of the prime mission science objectives is to obtain global elemental composition maps. These elemental composition results are needed with sufficient accuracy to enable comparison with major meteorite types. The results would also be used to assess the compositional heterogeneity of the asteroid and help to answer questions about differentiation. The selection of X-ray and gamma-ray spectrometers for this mission was based on their ability to produce global elemental composition maps of the asteroid. Gamma-ray measurements can determine the abundance of elements such as O, Si, Fe, Ti, Mg, K, Th, and U depending on the actual composition. Gamma-Ray Spectrometer Overview =============================== The gamma-ray spectrometer (GRS) on the NEAR mission is a non-dispersive system. For this type of detector, the incident gamma-ray photon is absorbed in the detector material and a signal proportional to the energy absorbed is measured as a voltage from the detector output. An analog-to-digital conversion is performed and the resulting count is binned by 'pulse height' or energy loss in memory. From an analysis of the pulse height spectra, elemental composition can be inferred. The choice among various types of gamma-ray detectors was made early in the planning for NEAR, the first Discovery mission of low-cost, rapid delivery planetary missions. While cooled solid-state detectors dominate laboratory measurements due to their inherent better energy resolution, this was not the case for this space mission. A NEAR GRS design was selected that could meet the required sensitivity in the energy region 0.1-10 MeV, while being consistent with the cost, mass, power, and reliability constraints of the mission. Gamma-ray photons in the energy region 0.1-10 MeV are emitted by excited nuclei and have discrete energies characteristic of that element. The excitation of nuclei can come from the radioactive decay of very-long lived radioisotopes in planetary materials, such as 40K, 238U, and 232Th. For other elements in space, excitation is provided by cosmic-ray bombardment. Except for periods of time during and immediately after major solar flares, this excitation is caused mainly by galactic cosmic rays (GCR) [EVANSETAL1993]. GCR are nuclear particles, mainly protons and helium nuclei (alpha particles), with a broad range in energy, but typically of order a GeV and a nearly isotropic flux of about 2-3 particles/cm2-s. These particles generate nuclear cascades on striking dense matter. The cascade particles most effective in generating gamma rays are neutrons. The high-energy neutrons produced in this manner can undergo further nuclear collisions. They may excite stable nuclei to higher energy levels by inelastic scatter. The resulting de-excitation of the nucleus can result in the emission of a gamma ray and is termed an inelastic scatter reaction; (n,n'gamma). Neutrons may also lose energy by elastic scatter until their energy is comparable to the thermal energy of a nucleus at a given temperature. These thermal neutrons can be captured by a nucleus and the resulting decay to the ground state of this new isotope, can produce a capture gamma ray; (n,gamma). The intensity of gamma rays emitted by a particular element and for a particular process depends on the concentration of that element, the reaction cross-section for the process, and the number of neutrons available with the appropriate energy. The inelastic scatter reactions have a large cross-section for all common nuclei at energies above the reaction threshold energy. Therefore, all the most abundant elements give a useful yield of gamma rays from this process. Thermal neutron cross-sections vary by orders-of-magnitude and yields for neutron capture gamma rays for the most abundant elements also vary widely [EVANSETAL1993]. Material containing large concentrations of elements with large neutron capture cross-sections can alter the thermal flux and lead to a flux depression of neutron capture gamma-rays. The neutron cascade penetrates into a planetary surface to a depth of hundreds of g/cm2, a few meters into the regolith on an object like an asteroid with no atmosphere. Gamma rays are scattered (with loss of characteristic energy) or absorbed on a distance scale of tens of g/cm2. This indicates that only those gamma rays generated in the first tens of g/cm2 can be detected on the surface or from orbit, and the important part of the neutron equilibrium distribution in the planetary body is that near the surface. The moderation and thermalization of the neutrons depend strongly on the composition of the near surface material, particularly on the hydrogen and carbon content, if any. Gamma-ray detectors on the surface or in an orbiting spacecraft can measure the discrete energy gamma rays and determine the elements that emitted these gamma rays [BOYNTONETAL1992], [BOYNTONETAL1993], [TROMBKAETAL1997]. Besides the characteristic gamma rays emerging from the asteroid surface, there will be a number of other sources of gamma rays that will be a background from which the gamma rays of interest will have to be separated. Some of these sources of background produce discrete lines and some appear as part of a continuum. Based on previous spaceflight experience, the major background components measured in orbit are: partial energy deposition in the detector (Compton effect); cosmic-ray activation of the detector and materials surrounding the detector; characteristic gamma rays emitted from the surface material, but scattered before emerging from the asteroid; cosmic-ray activation and natural radioactivity in the spacecraft; and gamma emission from astrophysical sources [BIELEFELDETAL1976]. Detectors --------- Most current non-dispersive gamma-ray detectors utilize either inorganic scintillation material, such as sodium iodide (NaI) or bismuth germinate (BGO), or solid state material, such as high-purity germanium (Ge). The Ge detectors have significantly better energy resolution than scintillation detectors and are generally favored for laboratory measurements. For example, the typical energy resolution (expressed as the full-width at half- maximum, FWHM, of the peak) for a Ge detector is 2 keV measured at 1332 keV from a 60Co calibration source. A similar FWHM for a NaI detector is 80 keV at 1332 keV. In addition, the Ge detector has a much better peak-to-Compton ratio, typically about 50 compared to 2 for a NaI detector [KNOLL1989]. These two factors give a Ge detector a decided advantage in resolving peaks close in energy and detecting peaks in the presence of significant continuum. Figure 1 shows measurement spectra of an extended soil sample irradiated by a neutron generator collected by both Ge and NaI detectors. Identification by element of some of the prominent gamma-ray peaks in the spectrum is indicated. While Ge detectors improve detection capability, they also require operation at cryogenic temperatures (typically <100 K) and suffer serious degradation in performance when exposed to cosmic radiation over long periods of time [BRUCKNERETAL1991]. Typically Ge detectors on planetary missions (such as Mars Observer) are designed with anneal capability to offset the expected resolution degradation during spaceflight. These limitations of a Ge detector along with the added cost and complexity reduced some of the advantages that might be expected for this mission. The question that had to be answered for the NEAR mission was: Could a scintillation detector meet the mission science requirements for elemental sensitivity? Analysis during the preliminary design phase of the NEAR mission indicated that a scintillation detector could meet the science objectives of the mission [EVANSETAL1995]. Experience in both U.S. and Russian planetary missions and in the oil well-logging industry have shown that good results could be obtained using scintillation detectors for such elements as K, Th, O, H, Mg, Si, and Fe. In addition, proposed design changes were expected to greatly improve detector performance over scintillators that have previously flown in space. NaI was chosen as the scintillator for NEAR because it has the best energy resolution of common scintillation materials in combination with a photomultiplier tube (PMT). These systems are rugged and have been used successfully on space flight missions for many years [EVANSETAL1993]. The pulse height spectrum obtained when monoenergetic gamma rays are detected has a shape determined by the gamma-ray energy and the characteristics of the detector. Important factors are: (1) the relative magnitude of the photoelectric, Compton, and pair-production cross-sections as a function of energy, and (2) the statistical fluctuations and losses involved in collecting the signal generated in the detector [KNOLL1989]. A measurement reflects the amount of energy that is lost in the detector and transferred as kinetic energy to electrons. At energies where the photoelectric absorption dominates, the kinetic energy imparted to a secondary electron is equal to the gamma-ray energy minus the electron binding energy. This energy can be reclaimed, in a sense, by the absorption of the X-rays produced by photoelectric absorption. At higher energies, when Compton scattering becomes more important, the gamma ray may lose part of its energy to the detector and escape the crystal or may then be photo absorbed. The gamma ray will lose all or part of its energy in the detector and possible escape with diminished energy. This generates a continuum that adds to the background up to the energy of the initial gamma ray minus the minimum scattered energy. At energies above 1022 keV, electron-positron pair production becomes possible. The electron eventually loses all its kinetic energy in the detector while the positron annihilate with another electron producing two 511 keV photons. The energy of these photons can either be absorbed in the detector or can escape. Therefore, three peaks will be created: (1) pair production with eventual absorption of both 511 keV photons to give a peak at the initial gamma-ray energy; (2) pair production with the absorption of one 511 photon and the escape of the other giving a peak at the initial gamma-ray energy minus 511 keV; and (3) pair production with the escape of both 511 keV photons giving a peak at the initial gamma-ray energy minus 1022 keV. Cosmic-ray interactions in the spacecraft produce gamma-rays characteristic of the spacecraft materials. These gamma rays would constitute an unwanted background signal and could potentially degrade the science return of the mission. Gamma-ray detectors of previous missions have often used a boom to move the gamma-ray detector away from the spacecraft reducing substantially the spacecraft background. On Mars Observer, for example, the gamma-ray detector was mounted on a boom that could be extended 6 meters from the spacecraft. The NEAR detector had to be body mounted on the lower deck along with all the other instruments. This required some other method of reducing the spacecraft background. Plastic scintillators are very effective charged particle shields, but higher density materials are needed to shield gamma rays. Passive shielding is not practical because of the large volume and mass of material needed to absorb the cosmic rays and the secondary radioactive products produced in the shield. An active collimator is used to reduce the charged particle, spacecraft and cosmic gamma-ray background as well as the Compton continuum. Charged particles or photons that interact in the shield produce a corresponding output signal. This signal can be used to trigger an anti-coincidence system to reject any counts in the central detector that are in coincidence with the shield within some time window. Bismuth Germinate (BGO) was chosen for the NEAR shield. It has a density of 7.13 g/cm2 which makes it especially effective for gamma ray interactions. As useful as the NEAR shield would be to reduce the detector background, significant information on the gamma-ray flux from the asteroid could be lost. The first and second escape peaks produced in the central detector will be mostly eliminated in the anti-coincidence spectrum. For a small central detector like the NEAR design, this loss would be unacceptable since the escape peaks would have many more counts than the full-energy peak for many energies of interest. For example, at 6 MeV, calculations indicate that 85% of all the counts would be in the escape peaks and only 15% in the full-energy peak. The NEAR design recovers these peaks with two additional NaI spectra; one in coincidence with 511 keV and one in coincidence with 1022 keV energy deposition in the BGO shield. Calibrations with the NEAR detector indicates the effectiveness of this design. Analysis of the coincidence window spectra during calibrations showed that not only photons generated by pair production were collected in the coincidence spectra. Any photons that Compton scattered in the central detector and then deposited 511 or 1022 keV in the shield were also collected. This was confirmed for energies where no pair production was possible (for incident energies less than 1022 keV), but the first coincidence spectrum showed a peak at the incident energy minus 511 keV. Thus, the number of counts in the peaks for both coincidence spectra were greater than that expected by calculations of efficiency for just pair production [TROMBKAETAL1997]. Subsequent calculations that modeled the coincidence process in the central detector and the shield to understand these measurements showed results similar to the measured efficiencies. This will be discussed further below. A radioactive source for energy calibration was not included in the NEAR design. It was expected that enough discrete lines from background sources would be measured in the spectra that the energy calibration could be monitored. The strongest of these would be the 511 keV line due to electron/positron annihilation in the spacecraft and materials surrounding the detector. Other gamma rays from cosmic ray interactions in the scintillation materials would be expected. Measurements taken during the cruise portion of the mission confirmed these predictions and will be discussed below. A picture of the NEAR GRS detector is shown in Figure 2 and a cross-section diagram in shown in Figure 3. A complete description of the GRS hardware is given in [GOLDSTENETAL1997] and [GOLDSTEN1998]. The central detector is a 2.54 cm X 7.62 cm right circular cylinder of NaI(Tl). A 3 cm diameter metal ceramic PMT is attached to the NaI in the asteroid facing direction. The shield is a BGO cup with outside dimensions of 8.9 cm X 14 cm. A 7.6 cm diameter metal ceramic PMT is attached to the BGO. The measured energy resolution at 662 keV (from 137Cs) was 8.7% for the NaI detector and 14% for the BGO. The energy range of both detectors could be controlled by changing the high-voltage setting for each detector, but was nominally 0.1-10 MeV. A generalized block diagram of the GRS detection scheme is shown in Figure 4 [GOLDSTENETAL1997]. An incoming photon is absorbed by the detector material and produces an output signal proportional to the energy absorbed. The detector signal is amplified, filtered, and its peak value measured using an analog-to-digital converter. A data processor collects the measurements and bins them according to the energy absorbed into a pulse-height spectrum. Five 1024-channel spectra are collected simultaneously. These are: the NaI raw spectrum, with no coincidence or anti-coincidence rejection; the BGO raw spectrum; the NaI spectrum measured in anti-coincidence with the BGO; the NaI spectrum measured in coincidence within a window around 511 keV in the BGO detector; the NaI spectrum measured in coincidence within a window around 1022 keV in the BGO detector. In addition two 21 channel spectra are collected from the BGO detector in coincidence with the NaI and in the windows specified around the 511 keV and 1022 keV energies. Temperature and voltage stability are extremely important to maintain system performance. The light outputs of the NaI and BGO detectors vary significantly with temperature. The detectors are thermally isolated and wrapped with operational heaters to stabilize the temperature to within 0.25 degC. The signal gains of the PMTs are not particularly sensitive to temperature, but are very sensitive to voltage variations. The XGRS uses an external feedback control system to produce ultra-stable high-voltage outputs from the on-board high-voltage power supplies [GOLDSTENETAL1997]. No gain changes due to temperature or high-voltage variation were observed during cruise." END_OBJECT = INSTRUMENT_INFORMATION OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "ADLERETAL1972A" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "ADLERETAL1972B" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "ADLERETAL1972C" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "BIELEFELDETAL1976" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "BOYNTONETAL1992" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "BOYNTONETAL1993" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "BRUCKNERETAL1991" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "CLARKETAL1995" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "EVANSETAL1993" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "EVANSETAL1995" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "GOLDSTEN1998" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "GOLDSTENETAL1997" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "KNOLL1989" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "STARRETAL1999" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "TROMBKAETAL1997" END_OBJECT = INSTRUMENT_REFERENCE_INFO OBJECT = INSTRUMENT_REFERENCE_INFO REFERENCE_KEY_ID = "YINETAL1993" END_OBJECT = INSTRUMENT_REFERENCE_INFO END_OBJECT = INSTRUMENT END