f National Museum of Natural History, Smithsonian Institution, Washington, DC 20560, USA g reserved. * Corresponding author. Tel.: +1 609 897 8532; fax: +1 609 897 8503. E-mail address: groves@princeton.oil?eld.slb.com (J.L. Groves). Nuclear Instruments and Methods in Physics Research B 241 (2005) 232?237 www.elsevier.com/locate/nimb0168-583X/$ - see front matter  2005 Elsevier B.V. All rightsUniversity of Connecticut, Storrs, CT 06269, USA Available online 15 August 2005 Abstract A pulsed neutron/gamma-ray detection system for use on rovers to survey the elemental concentrations of Martian and Lunar surface and subsurface materials is evaluated. A robotic survey system combining a pulsed neutron gener- ator (PNG) and detectors (gamma ray and neutron) can measure the major constituents to a depth of about 30 cm. Scanning mode measurements can give the major elemental concentrations while the rover is moving; analyzing mode measurements can give a detailed elemental analysis of the adjacent material when the rover is stationary. A detailed map of the subsurface elemental concentrations will provide invaluable information relevant to some of the most fun- damental astrobiological questions including the presence of water, biogenic activity, life habitability and deposition processes.  2005 Elsevier B.V. All rights reserved. PACS: 95.55.Pe; 07.85.Nc Keywords: Neutron generator; Subsurface elemental composition; Planetary surveysPulsed neutron generator system for astrobiological and geochemical exploration of planetary bodies Hatice Akkurt a, Joel L. Groves a,*, Jacob Trombka b, Richard Starr c, Larry Evans d, Samuel Floyd b, Richard Hoover b, Lucy Lim b, Timothy McClanahan b, Ralph James e, Timothy McCoy f, Je?rey Schweitzer g a Schlumberger Princeton Technology Center, 20 Wallace Road, Princeton Junction, NJ 07605, USA b NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA c The Catholic University of America, Washington, DC 20064, USA d Computer Sciences Corporation, 7700 Hubble Drive, Lanham-Seabrook, MD 20706, USA e Brookhaven National Laboratory, Upton, NY 11973, USAdoi:10.1016/j.nimb.2005.07.029 eth. i1. Introduction The analysis of the characteristic neutron and gamma-ray spectra induced by energetic neutrons is one of the most powerful methods for in situ chemical analysis of heterogeneous materials. Orbital and in situ measurements of the gamma- ray and neutron emission spectra induced by cos- mic rays have been used to determine the chemical composition of the near-Earth asteroid 433 Eros by the NEAR probe [1]. Recently, gamma-ray/ neutron spectrometer results from the Mars Odys- sey Orbiter indicate an abundance of ice, likely embedded in or interlayered with the geologic stra- ta [2]. However, cosmic ray induced gamma-ray and neutron spectra require long acquisition times and are not well suited for detailed evaluation of planetary or lunar near surface materials using rovers. Other potential sources of neutrons for geochemical measurements on extraterrestrial bodies include (a,n) reactions (e.g. 241AmBe), spontaneous ?ssion isotopes (e.g. 252Cf), radio- active power generators (RTGs) and pulsed D?T fusion neutron generators (PNGs). The main con- siderations for neutron source selection for plane- tary exploration are launch safety, energy spectra of the emitted neutrons, neutron yield, mass, vol- ume, power requirements and switchability, ability to turn the source on and o?. Furthermore, it is important that the neutron source does not emit gamma rays itself to avoid having interference with the measurements. The ability to switch o? the neutron source is highly desirable both for launch safety reasons as well as to avoid activation of the craft and detection equipment during the mission ?ight time. The 252Cf source and RTGs can not be turned o?; switchable (a,n) reaction based sources are possible but have not been com- mercially developed; whereas, PNGs are easily switchable since they only produce neutrons when the ion source is on and the high voltage is applied. Also, miniature, low power PNGs are commercially available and are routinely used for geochemical logging in the hostile oil?eld environment. PNGs have other signi?cant advantages for H. Akkurt et al. / Nucl. Instr. and Mgeochemical survey applications on extraterrestrialbodies. The concentration measurements of the very important elements, C and O, are most easily accomplished by detecting the gamma rays follow- ing inelastic scattering of high-energy neutrons. The threshold for the C and O inelastic reactions are 4.8 and 6.4 MeV, respectively. All the neutrons generated by the D?T reactions in the PNGs are in the range 13.4?14.7 MeV whereas the other neutron sources only have a small percentage of neutrons above 6 MeV as shown in Fig. 1. A thorough knowledge of the neutron source emission spectrum is very important for extracting the maximum information from measured gamma- ray and neutron spectra. The data inversion pro- cess involves calculating or measuring the system response to the all the elements of interest and then ???tting?? these elemental signatures to the observed data. Much of the system response information is acquired by computational methods for which the source emission characteristics must be known. PNGs are unique compared to other neutron sources since the neutrons are emitted with a well-known distribution in a very narrow energy band. PNGs that operate at constant high voltage with a pulsed ion source also have the very valu- able characteristic that the neutron ?ux can be switched on and o? in less than one microsecond and the neutron output duty cycle can be any value from 1% to 100%. The cycle frequency can be as high as 2 ? 104 Hz. Thus, a neutron on?o? sequence can be generated that allows the maxi- mum elemental information to be extracted from the complex, neutron induced gamma-ray and neutron spectra. For example, gamma rays from inelastic neutron interactions are detected while the PNG is on, gamma rays from neutron capture interactions are detected in the 1?1000 ls time interval just after the neutron pulse is turned o?, and gamma rays from the neutron activation of the surrounding materials are detected in the time interval after the capture gamma rays have gone to zero and before the start of the next neutron pulse. PNGs that are operated with a pulsed high-volt- age supply either have pileup gamma rays that can- not be analyzed by the detector and/or have too long a duty cycle between pulses where neither n Phys. Res. B 241 (2005) 232?237 233inelastic nor capture gamma rays are present. eth. i234 H. Akkurt et al. / Nucl. Instr. and MThus, PNGs that are operated with a constant high voltage and a pulsed ion source are the only practical designs for geochemical logging. In an optimal design, the count rate is set by the pulse rate of the generator. The slowing down time for the 14 MeV neutrons is a strong function of the amount of hydrogen present in the surrounding materials. A Martian or Lunar Rover with a PNG and gamma-ray and neutron detectors will have the capability of sur- veying for the major near surface geochemistry when moving and of analyzing for the elemental fractions and geochemistry of a small area in much more detail when at rest. 2. Description of the instrumentation package The instrumentation package evaluated for planetary rover applications has a PNG with a pulsed ion source and gamma-ray and neutron Fig. 1. The source spectra for AmBe [4], RTG [5] and 252Cf [6] isoto sources can vary signi?cantly in details especially below 2 MeV, depen the thickness of the capsule. The emitted neutron energy from D?T re 14.4 MeV.n Phys. Res. B 241 (2005) 232?237detectors mounted much like that of an oil?eld geochemical logging tool. Such instruments have been routinely used for oil and gas exploration applications for the last few decades. These oil?eld tools are used for subsurface formation elemental measurements, geochemical analysis and behind casing ?uid analysis at depths up to 9000 m, in temperatures ranging from 40 C to beyond +200 C, and on drill collars where shocks exceed 250 g for hundreds of hours. In addition, to the ruggedness of the environ- ment and the di?cult operating conditions, the borehole-logging tool must be as small in diameter as possible in order to enter an oil well through the production tubing (typically less than 5.0 cm) and must be low power since the power must be delivered to the tool by a cable up to 9000 m long. PNGs operating under typical conditions used in the oil?eld require about 15 W for out- puts of 1 ? 108 n/s with life-time on the order of a thousand hours. Less power is required for lower pic neutron sources are shown. Note that the spectra for (a,n) ding on the homogeneity of the target and a emitter mixture and action is almost mono-energetic with average neutron energy of average neutron ?uxes. The total mass of the com- plete instrumentation package including the detec- tors is expected to be about 8.0 kg. The neutron detectors proposed for the system are widely used, rugged 3He detectors. Such detec- tors have been successfully ?own on many previ- ous NASA missions. A 3He detector surrounded with a layer of Cd is used to detect epithermal neu- trons since Cd has large absorption cross-section at thermal energies. An unshielded 3He detector is used for thermal neutron detection. Also, a rover instrumentation system will have a neutron monitor mounted next to PNG to monitor the source neutron output. The optimal detector for gamma-ray spectros- copy measurements has high e?ciency, excellent resolution, a broad temperature operating range and is resistance to radiation damage. Unfortu- nately, there is no gamma-ray detector that simul- taneously satis?es all these criteria. Alternative detectors include scintillators, i.e. NaI, BGO, CsI o?er excellent resolution, which is important for nuclide identi?cation and hence signal processing, but need cooling and su?er from radiation damage when exposed to high-energy neutrons. Of course, MCNP [7] modeling can be carried out for any detector or detector combination; however, the re- sults reported here were calculated using a CsI scintillation gamma-ray detector. 3. Results The instrumentation package used for the MCNP modeling studies reported here consists of a PNG, thermal and epithermal neutron detec- tors (2.5 D ? 10 H cm), and a CsI gamma-ray detector (5.0 D ? 7.5 H cm). The neutron detectors are placed between PNG and the CsI detector. The package is placed on the surface of the soil assum- ing that the instrument will be mounted on one of the arms of the rover. The soil used for the MCNP vary H. Akkurt et al. / Nucl. Instr. and Meth. in Phys. Res. B 241 (2005) 232?237 235and semi-conductors, i.e. HpGe detectors. While scintillators o?er high e?ciency and resistance to temperature, they su?er from poor energy resolu- tion. On the other hand, germanium detectors Fig. 2. Epithermal neutron detector time dependent response for interval is obtained by multiplying by the neutron source output.ing water content in soil [3]. The count rate per acquisition timemodeling has the composition found with the MARS Path?nder APXS instrument [3]. For this con?guration, the epithermal neutron detector response for homogeneous soil with varying water content is shown in Fig. 2. The time dependent response is particularly sensitive to the water content. In addition, the decay curve will provide extra information with regard to proper- Fig. 3. Thermal neutron detector time detector response for varying water content in soil [3]. The count rate per acquisition time interval is obtained by multiplying by the neutron source output. ith th cordin 236 H. Akkurt et al. / Nucl. Instr. and Meth. in Phys. Res. B 241 (2005) 232?237Fig. 4. The CsI gamma-ray detector response for a soil sample w of turning the source on and o? for a given pulse width allows re CsI detector was 8.5%.e APXS instrument composition [3] + 3% water. The capability g inelastic and capture spectra separately. The resolution of the ties of the subsurface since the slope is propor- tional to the slowing down length of the media. The thermal neutron detector response for ing scheme, the burst can be turned on and o? neutrons are deeply penetrating and many of the resulting gamma rays are high energy. Neutron evidence for subsurface ice deposits, Science 2975 (2002) 81. [3] Preliminary Mars Path?nder APXS Results: Elemental H. Akkurt et al. / Nucl. Instr. and Meth. in Phys. Res. B 241 (2005) 232?237 237automatically, which allows recording inelastic and capture spectra separately. This is especially bene?cial when scintillators with poor energy reso- lutions are used, the interpretation and signal pro- cessing becomes much easier since inelastic and capture gammas can be separated. Every element has characteristic inelastic, capture and activation gamma-ray signatures and the resulting gamma- ray spectra are the fractional supposition of all the characteristic elemental spectra. 4. Conclusions and future work The initial modeling results for a PNG and neu- tron/gamma-ray spectrometer system for plane- tary and lunar rover surveys are presented and the modeling results show the considerable analyz- ing power of such instruments. The PNG system is complimentary to the APXS spectrometer used on the Mars Path?nder, which has a very shallow sensitivity depth. The PNG spectrometer instrument easily detects H, C, O and other major constituents over a large vol- ume of the surrounding material since 14.4 MeVvarying water content in soil is shown in Fig. 3. Here, the slope of the decay curve is proportional to the capture cross-section of the media, which is not only proportional to density but also to hydro- gen and absorber content, i.e. chlorine content, of the media. Furthermore, thermal neutron decay time measurements can be performed using PNG as a neutron source and the gamma-ray detector to detect neutron capture. The slope of the decay curve is again proportional to the capture cross- section of the media. The CsI gamma-ray detector response for a soil sample with the APXS instrument composition [3] + 3% water is shown in Fig. 4. Due to the tim-Compositions of Rocks and Soil, 1997. Available from: . [4] G. Burger, R.B. Schwartz, Guidelines on calibration of neutron measuring devices, IAEA Technical Reports Series No. 285 (1988). [5] R.A. Kaminskas, G.W. Ryan, C.A. Smith, RTG/Science instrument radiation interactions for deep space probes, NAS 2-5222, 1969. [6] G.F. Knoll, Radiation Detection and Measurement, third ed., John Wiley, New York, 1999. [7] X-5 Monte Carlo Team, MCNP A General Monte Carlo N-Particle Transport Code, Version 5, LA-UR-03-1987, Los Alamos National Laboratory, April 2003.activated, gamma-ray spectroscopy is the only technique available that can determine the in situ fractional concentrations of H, C and O in the sub- surface region. Further MCNP modeling will be directed at determining the sensitivity of di?erent instrument con?gurations to the critical elements as de?ned by the scienti?c objectives for planetary and lunar missions. Also, in situ measurements with a bore- hole-logging tool mounted near the surface or dragged along the surface on a carrier will be used to benchmark the MCNP modeling results and demonstrate the instrument capabilities. References [1] T.J. McCoy, T.H. Burbine, L.A. McFadden, R.D. Starr, M.J. Ga?ey, L.R. Nittler, L.G. Evans, N. Izenberg, P. Lucey, J.I. Trombka, J.F. Bell III, B.E. Clark, P.E. Clark, S.W. Squyres, C.R. Chapman, W.V. Boynton, J. Veverka, The composition of 433 Eros: a mineralogical?chemical synthesis, Meteor. Planet. Sci. 36 (2001) 1661. [2] W.V. Boynton, W.C. Feldman, S.W. Squyres, T.H. Pretty- man, J. Bruckner, L.G. Evans, R.C. Reedy, R. Starr, J.R. Arnold, D.M. Drake, P.A.J. Englert, A.E. Metzger, I. Mitrofanov, J.I. Trombka, C. dUston, H. Wanke, O. Gasnault, D.K. Hamara, D.M. Jane, R.L. Marcialis, S. Maurice, I. Mikheeva, G.J. Taylor, R. Tokar, C. Shinoha- ral, Distribution of hydrogen in the near surface of mars: