Advances in Geosciences: Planetary Science, Vol. 7 9789812709882, 9812709886

Citation preview

A d v a n c e s

i n

Geosciences Volume 7: Planetary Science (PS)

This page intentionally left blank

A d v a n c e s

i n

Geosciences Volume 7: Planetary Science (PS)

Editor-in-Chief

Wing-Huen Ip

National Central University, Taiwan

Volume Editor-in-Chief

Anil Bhardwaj

Vikram Sarabhai Space Centre, India

World Scientific NEW JERSEY

•

LONDON

•

SINGAPORE

•

BEIJING

•

SHANGHAI

•

HONG KONG

•

TA I P E I

•

CHENNAI

Published by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.

ADVANCES IN GEOSCIENCES A 4-Volume Set Volume 7: Planetary Science (PS) Copyright © 2007 by World Scientific Publishing Co. Pte. Ltd. All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.

ISBN-13 ISBN-10 ISBN-13 ISBN-10

978-981-270-781-9 981-270-781-6 978-981-270-986-8 981-270-986-X

(Set) (Set) (Vol. 7) (Vol. 7)

Typeset by Stallion Press Email: [email protected] Printed in Singapore.

EDITORS

Editor-in-Chief:

Wing-Huen Ip

Volume 6: Hydrological Science (HS) Editor-in-Chief: Namsik Park Editors: Chunguang Cui Eiichi Nakakita Simon Toze Chulsang Yoo Volume 7: Planetary Science (PS) Editor-in-Chief: Anil Bhardwaj Editors: C. Y. Robert Wu Francois Leblanc Paul Hartogh Yasumasa Kasaba Volume 8: Solar Terrestrial (ST) Editor-in-Chief: Marc Duldig Editors: P. K. Manoharan Andrew W. Yau Q.-G. Zong Volume 9: Solid Earth (SE), Ocean Science (OS) & Atmospheric Science (AS) Editor-in-Chief: Yun-Tai Chen Editors: Hyo Choi Jianping Gan

v

This page intentionally left blank

REVIEWERS

The Editors and Editor-in-Chief of Volume 7 (Planetary Science) would like to thank the following referees who have helped review the papers published in this volume: Aigen Li Alexander Medvedev Andrew Nagy Anil Bhardwaj Bing-Meng Cheng Cesar Bertucci David J. McComas Georges Durry Gerad Beaudin Guillermo M. Munoz Caro Hiroshi Kimura Jamie Elsila Karri Muinonen Manubo Kato Martin Hilchenbach Masahiko Arakawa

Miriam Rengel Nicolas Fray Norbert Krupp Paul Hartogh Pradip Gangopadhyay Robert Wu Scott Bolton Sonia Fornasier Tai-Sone Yih Takao Nakagawa Takeshi Imamura Tetsuo Yamamoto Tom Charlock Yasushi Yamaguchi Yuriy Shkuratov

vii

This page intentionally left blank

CONTENTS

Editors

v

Reviewers

vii

Some Similarities and Differences Between the Mars and Venus Solar Wind Interactions J.-G. Trotignin Comparison of Microwave Observations of Martian Temperature and Winds with General Circulation Model Simulations T. Kuroda and P. Hartogh Asteroid Compositions: Some Evidence From Polarimetry A. Cellino, M. Di Martino, A.-C. Levasseur-Regourd, I. N. Belskaya, Ph. Bendjoya, R. Gil-Hutton Low Energy Charged Particle Measurement by Japanese Lunar Orbiter Selene Y. Saito, S. Yokota, K. Asamura, T. Tanaka and T. Mukai A Jovian Small Orbiter for Magnetospheric and Auroral Studies with the Solar-Sail Project Y. Kasaba, T. Takashima, H. Misawa and Jovian Small Orbiter Sub-Working Group [with J. Kawaguchi and Solar-Sail Working Group] Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer L. Paganini and P. Hartogh ix

1

13

21

33

45

55

x

Contents

Formation of Alumina Nanoparticles in Plasma M. Kurumada and C. Kaito Infrared Study of UV/EUV Irradiation of Naphthalene in H2 O+NH3 Ice Y.-J. Chen, M. Nuevo, F.-C. Yeh, T.-S. Yih, W.-H. Sun, W.-H. Ip, H.-S. Fung, Y.-Y. Lee and C.-Y. R. Wu New Method of Producing Titanium Carbide, Monoxide and Dioxide Grains in Laboratory A. Kumamoto, M. Kurumada, Y. Kimura and C. Kaito Destruction Yields of NH3 Produced by EUV Photolysis of Various Mixed Cosmic Ice Analogs C. Y. R. Wu, T. Nguyen, D. L. Judge, H.-C. Lu, H.-K. Chen and B.-M. Cheng Formation of CaTiO3 Crystalline Dust in Laboratory K. Yokoyama, Y. Kimura, O. Kido, M. Kurumada, A. Kumamoto and C. Kaito Direct Observation of the Crystallization of Carbon-Coated Amorphous Mg-bearing Silicate Grains C. Kaito, S. Sasaki, Y. Miyazaki, A. Kumamoto, M. Kurumada, K. Yokoyama, M. Saito, Y. Kimura and H. Suzuki Relationship Between Morphology and Spectra Revealed by Difference in Magnesium Content of Spinel Particles M. Saito, M. Kurumada and C. Kaito Ionization of Polycyclic Aromatic Hydrocarbon Molecules around the Herbig Ae/Be Environment I. Sakon, T. Onaka, Y. K. Okamoto, H. Kataza, H. Kaneda and M. Honda Search for Solid O- and N-Rich Organic Matter of Prebiotic Interest in Space G. M. M. Caro and E. Dartois

69

79

93

101

115

125

133

143

155

Contents

Balloon-Borne Telescope System for Optical Remote Sensing of Planetary Atmospheres and Plasmas M. Taguchi, K. Yoshida, H. Nakanishi, Y. Shoji, K. Kawasaki, J. Shimasaki, Y. Takahashi, J. Yoshida, D. Tamura and T. Sakanoi The Strategic Plan for the Integrated Sciences and the Development Status of Japanese Lunar Explorers: SELENE and Lunar-A T. Iwata, S. Tanaka, M. Kato, S. Sasaki, N. Namiki From Nuclear Blasts to Cosmic Bombardment K. O’Brien

xi

169

181

191

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


SOME SIMILARITIES AND DIFFERENCES BETWEEN THE MARS AND VENUS SOLAR WIND INTERACTIONS JEAN-GABRIEL TROTIGNON Laboratoire de Physique et Chimie de l’Environnement, Centre National de la Recherche Scientifique, Orl´ eans University 3A, Avenue de la Recherche Scientifique, F-45071 Orl´ eans Cedex 02, France [email protected]

The plasma environments of Mars and Venus have been explored by spacecraft, such as Mars 2, 3 and 5, Phobos 2, Mars Global Surveyor (MGS), Mars Express for planet Mars and Venera 9 and 10, Pioneer Venus Orbiter, Venus Express for planet Venus. Overall observations of plasma regions and their boundaries, in particular the bow shock, the magnetic pile-up boundary and the magnetic tail, show the solar wind interaction with these two planets to be rather similar. Mars and Venus are both considered as non-magnetic planets, compared with the Earth, in a sense that they do not possess any significant intrinsic magnetic field that could play a significant role in their interactions with the solar wind. At most, the magnetic anomalies discovered at Mars by MGS are thought to slightly influence the lower regions of the Martian ionosphere. Therefore, both Venus and Mars have principally comet-like induced magnetospheres and magnetotails as a result of the atmospheric mass loading and subsequent draping of passing interplanetary flux tubes. Nevertheless, there are many differences between the characteristics and space environment behaviors of the two telluric planets and a lot remains actually to be done, in terms of in situ measurements and modeling efforts, to fully understand how Venus and Mars interact with the interplanetary medium. The objective of the presentation is not to review all the aspects of these interactions but simply to compare the main characteristics of the Mars’ and Venus’ plasma environments and to highlight some similarities and differences between the interactions of these two non-magnetic planets with the solar wind as a function of solar wind dynamic pressure and solar activity.

1. Introduction Despite the great number of missions devoted to the exploration of Earth’s nearest neighbors, Venus and Mars and a prolific scientific output, actually not too much is known about the interaction of these planets with the solar 1

2

J.-G. Trotignon

wind, and this is particularly true for Mars. This is due to limitations of space missions for gathering in situ data especially in situation involving variations in latitude, longitude, altitude, time, and solar activity. This is also because only a very few spacecraft are well equipped to monitor electric and magnetic fields, and the spatial distribution and composition of particles over wide ranges of frequency, mass and energy. This leads one to the conclusion that new missions, in particular low altitude orbiters and entry probes, are required to fully understand the nature of these interactions, compared with the ones of the Earth and mainly comets. They are indeed thought to be much more intricate than those described in most of the current literature. The objective of this paper is actually not to give a comprehensive review of what is known and what is unknown about the plasma and wave environments of Mars and Venus and about their interactions with the interplanetary medium but rather to point out some relevant features that remains to be clarified and/or understood and therefore investigated in future missions. Section 2 recalls the basic facts about the interactions between the solar wind and the two planets. Some major similarities and differences are then presented in Sec. 3, before the conclusion.

2. Basic Facts about Venus and Mars Solar Wind Interactions 2.1. The Venus case Venera 9 and 10 and Pioneer Venus Orbiter (PVO) missions to Venus have considerably enriched our knowledge of the space environment of Venus and its interaction with the solar wind.1,2 Venus turned out to be a non-magnetic planet, with a dense dayside ionosphere. There are extended dayside exospheres of hot hydrogen and oxygen: O dominates over H up to 3000 km where H becomes dominant (the main source of neutral O corona is dissociative recombination of O+ 2 ). Neutral exosphere atoms above the dayside ionopause, ionized by photoionization, charge exchange with solar wind protons, impact ionization, etc., can mass load and slow the solar wind via pickup processes, as for comets. Venus has no intrinsic magnetic field that could stand off the solar wind. The solar wind thus appears to be diverted around the upper boundary of the ionosphere, the ionopause, where the incident solar wind ram pressure is balanced by the ionosphere thermal pressure. At the

Mars and Venus Solar Wind Interactions

3

ionopause, dissipationless currents flow in a thin layer, thus creating a highly conducting spheroid. These currents produce magnetic fields that contribute to stand off the solar wind. As illustrated in Fig. 1,1 the solar wind is shocked and diverted around the ionosphere and continues along its antisolar route. This flow carries the interplanetary magnetic field with it. The interplanetary magnetic field is compressed in front of the Venus’ ionopause, thus creating a magnetic barrier that separates the plasmas of external and internal origin. On the flanks of the planet, the interplanetary magnetic field lines continue to move downward to form a magnetic tail similar to comet tails. This phenomenon is known as the magnetic field draping effect.

2.2. The Mars case Before Mars Global Surveyor (MGS), the only available measurements of the upper environment of Mars come from spacecraft en route to deliver landers and higher-altitude data from orbiters which never encountered the ionosphere.3 This explains why a controversy has long existed regarding the nature of the Martian obstacle to the solar wind flow. The MGS data have confirmed that it is an ionospheric obstacle like that of the unmagnetized

Fig. 1. field.1

Formation of induced Venus’ magnetotail from draped interplanetary magnetic

4

J.-G. Trotignon

planet Venus, while multiple magnetic anomalies of small spatial scale exist in the crust of Mars.4 These magnetic anomalies are at most thought to slightly influence the lower regions of the Martian ionosphere.5 The low Martian gravitational field (compared with those of Earth and Venus) allows the neutral exosphere of Mars to interact significantly with the interplanetary medium. Comet-like features are likely to be more evident at Mars than Venus.6 Neutral particles can escape above the exobase (∼200-km altitude): those whose velocities are larger than 5 km s−1 , the required velocity for escape from Mars, will contribute to the solar wind erosion of the Mars atmosphere; the others make up the gravitationally bound exosphere. Particles can gain energy through photochemical processes or sputtering by energetic particles from above. For example, O+ 2 which are majority ions in the ionosphere of Mars produce energetic oxygen atoms, O∗ , through photodissociative recombination. O∗ may gain sufficient energy to flow sunward and even escape from Mars. At Venus, the above photodissociative recombination cannot supply O∗ whose velocity is larger than the Venus escape velocity (∼10 km s−1 ). By photoionization O+ ions may then be produced and captured by the solar wind and embedded into the interplanetary magnetic field (ion pickup process as for Venus and comets). As ions are continuously produced, the solar wind carries more material along with it and is mass-loaded. Conservation of momentum and energy implies a slowing down of the solar wind flow, thus diminishing the pressure exerted on the planet environment. Extensive reviews of the Martian environment and its interaction with the solar wind may be found in Refs. 3 and 7–9 and references therein.

3. Some Similarities and Differences 3.1. Bow shock and upstream waves Figure 2 shows the Venus’ and Mars’ bow shock models inferred from the data of, respectively, PVO,10 and Phobos 2 and MGS.11 As can be seen in Fig. 2, the Mars’ bow shock has a larger terminator radius than the one of Venus, while the subsolar positions of the two bow shocks are comparable. In addition, the effective obstacle to the solar wind flow is larger at the flanks of Mars. According to Luhmann,12 it could be the result of a larger solar wind ion gyroradius relative to the planet radius for Mars and/or different compositions and scale heights of the two upper atmospheres.

Mars and Venus Solar Wind Interactions

5

Fig. 2. Conic section fits to the positions of the Venus bow shock crossed by PVO (thin solid, dashed and dashed-dotted lines).10 For comparison, the conic section fit to the Mars bow shock positions obtained from the Phobos 2 and MGS observations is shown as a thick solid line.11 For the latter curve, RV must be replaced by RM , the Martian radius.

The electric-field spectra plotted in the right-hand panel of Fig. 3 were recorded in the Martian bow shock ramp by the Plasma Wave System (PWS) onboard Phobos 2. These spectra exhibit two main components: a low-frequency component, below the electron cyclotron frequency, fce , which is attributed to the electric component of the whistler mode noise; and a high-frequency component with a broad peak at the ion plasma

Fig. 3. A comparison between the electric-field spectra measured in the shock ramps of Jupiter, the Earth, Venus and Mars.13–15

6

J.-G. Trotignon

frequency, fpi , and then a cutoff, which is thought to be the signature of Doppler-shifted ion acoustic waves.13 The electric-field spectra measured in the bow shock ramps of Jupiter, the Earth and Venus are also shown, for comparison, in Fig. 3.13–15 It is noteworthy that they all show a remarkably close similarity in shape. The noise amplitudes are nevertheless very different. This is partly due to instrumental effects, the antenna lengths are indeed sometimes lower than the plasma Debye length, so that the effective length of such antennae becomes questionable. Moreover, as shown on the third panel from the left, the PVO plasma wave instrument (orbiter electric field detector, OEFD) suffered from a lack of frequency resolution, only four frequency channels (centered on 30 kHz, 5.4 kHz, 730 Hz and 100 Hz) were indeed available.16 There is therefore a lot to do in this domain, and in particular at Venus. Electron plasma oscillations have, for the first time, been detected at Mars by PWS. These electrostatic waves are generated, at the plasma frequency, in the electron foreshock by suprathermal electrons that are energized and reflected at the shock whenever the interplanetary magnetic field is connected to the shock surface. Further downstream, in the ion foreshock, ion-acoustic and ULF waves are also generated. Such waves were observed in the Venus electron and ion foreshocks, as shown in Fig. 4.16 Note that foreshock waves are different to those generated from exospheric ion pick-up.9 The spatial distribution of electron plasma oscillations observed at Mars by PWS is displayed in Fig. 5.17 As expected, the highest electric-field intensities are seen in the electron foreshock. The observed limited extent of plasma oscillations along the tangent magnetic-field line is thought to be the consequence of the small size of the Martian shock. This is consistent with

Fig. 4. Electron plasma oscillations (left) and ion acoustic waves (right) in, respectively, the electron and ion foreshocks of Venus.16

Mars and Venus Solar Wind Interactions

7

Fig. 5. Electron plasma oscillation intensity in the equatorial B–v plane, which contains the center of Mars and is parallel both to the solar wind velocity and the interplanetary magnetic field (IMF) line that passes through the spacecraft. The white line, curve and circle are, respectively, the typical Parker IMF line tangent to the shock, the average shock surface, bow shock (BS) and planet Mars.17

the argument that the shock curvature controls the electron energization, as is the case for Venus.16

3.2. Ionosphere and ionopause The ionospheric plasma composition at Venus is mainly O+ 2 below 200 km and O+ above. The dayside ionosphere density peak (5–7 × 105 cm−3 ) is located at about 140-km altitude, depending on solar activity. It has only been measured by radio-occultation.18 A precise determination of the ion and electron temperatures remains to be done. The ionosphere upper boundary, the ionopause, is typically located at 300 km in the subsolar direction and at 1000 km close to the terminator, at least, when the solar wind ram pressure does not exceed the ionospheric pressure.19,20

8

J.-G. Trotignon

Information about the ionosphere of Mars actually comes only from in situ measurements made by the two Viking landers (at low latitudes, ∼45◦ SZA, near a minimum of solar activity), and altitude profiles of the electron density obtained from the radio-occultation experiments onboard the Mariner and Viking orbiters.22 A peak density of 105 cm−3 (about one order of magnitude lower compared with Venus) is usually observed at 125-km altitude. The topside termination of the ionosphere (ionopause) has not been firmly observed. This could be attributed to the fact that, at Mars, the ionospheric thermal plasma pressure is most of the time insufficient to balance the solar wind dynamic pressure.3 At Venus, when the solar wind dynamic pressure occasionally exceeds the peak ionospheric plasma pressure, no sharp increase of the ionosphere density is indeed observed at the ionopause.19 The Martian ionosphere is mainly composed of O+ 2 ions, and O , and between mostly created by charge exchanges between CO+ 2 2 23 CO+ 2 and O. Returning now to the fact that, at Mars, the ionospheric thermal plasma pressure is usually lower than the solar wind ram pressure. Luhmann24 claimed that by analogy with Venus, one might expect to find induced magnetic fields on the surface of Mars that could be detected by ground magnetometers. These fields of external origin should thus combine with crustal remanent magnetic fields. At solar maximum, when the solar wind pressure gets very high, large-scale horizontal magnetic fields have been observed in the dayside ionosphere of Venus. They are interplanetary fields incompletely cancelled by the shielding currents in the upper ionosphere. At Mars, these fields could also make their way through the solid mantle of the planet, might be pulled into the wake, and could produce the slingshot magnetic field pattern on the nightside that will accelerate plasma down the tail. It may therefore be the greatest loss source for the Mars atmosphere (C. T. Russell, private communication). 3.3. Plasma clouds Thermal electrons (plasma clouds), possibly scavenged from the top of the ionosphere by Kelvin–Helmholtz instability, were detected in the Venus’ magnetosheath, above the ionopause.25,26 They could be attached streamers analogous to cometary tails and might be the seed population of suprathermal ions (10–90 eV) that are observed in these regions. Cold plasma clouds have been observed in the sunward “planetosphere” of Mars (Fig. 6). Densities as high as 700 cm−3 , and temperatures of the

Mars and Venus Solar Wind Interactions

9

Fig. 6. From top to bottom, spacecraft potential, plasma density, electron gyrofrequency and electric field in three frequency channels measured by the PWS of the Phobos 2 mission to Mars.15 The bow shock (BS) and magnetic pile-up boundary (MPB) are, respectively, crossed at 0535:50 UT and 0548 UT on 8 February 1989. 21 Plasma clouds are seen just after the MPB.

order of 104 K have been reported.15 It is worth noting that the dynamic pressure developed by such clouds is equivalent to that of a 20-nT magnetic field. Plasma clouds could be generated by ionization of plasmaspheric neutrals or could result from ionopause instabilities. They actually look like Venus’ cold plasma clouds. Plasma clouds (>60 cm−3 ; ∼1 eV) were also observed in the night sector of Mars, close to the neutral sheet, in association with a broadband wave activity (from a few Hz to several kHz). The density profiles of these cold plasma clouds displayed fluctuations correlated with those of the magnetic field. They may originate from the dayside ionosphere and be dragged into the night sector by the solar wind flow. Again, they look like the Venus’ plasma clouds and/or tail rays.27 3.4. Magnetic pile-up boundary Attempts to characterize the location and shape of the magnetic pileup boundary (MPB), a boundary in the lower magnetosheath of Mars

10

J.-G. Trotignon

(previously called Protonopause, Planetopause, Ion Composition Boundary, Magnetopause, etc.) were first carried out using Phobos 2 data.21 Then, MGS observations combined with those of Phobos 2 have shown that it is actually a plasma boundary formed by the interaction of the solar wind with the Martian exosphere/ionosphere.8,11,28 At Venus, plasmas of solar and planetary origin appear to be separated from each other by a transition region, called mantle (or magnetic barrier), where plasma clouds were seen.1,26 Recently, Bertucci et al.29 claimed that the upper boundary of this region would be similar to the MPB identified at comets30,31 and planet Mars.21,28 The MPB should therefore be a common plasma boundary in the interaction between the solar wind and non-magnetic objects.

4. Conclusion The solar wind interactions with Venus and Mars appear to be quite similar, but with significant differences.12,19,32 These differences include a greater width of the Martian bow shock associated with a greater width of magnetotail and a much larger O+ ion escape rate. The latter is related to a lower Martian gravitational field compared with those of Venus and the Earth, a larger extent of the exosphere, and efficient energizing processes. A relative insensitivity of the Martian bow shock to the solar cycle compared with the unquestionable Venus’ bow shock variability has also been reported (Refs. 9 and 32, and references therein). Any intrinsic B-fields are too weak at Mars and Venus to stand off the solar wind, therefore the atmosphere (exosphere included) and ionosphere combine to provide an obstacle to the flow. Outermost signatures of Mars’ and Venus’ obstacles are fast magnetosonic shocks and foreshocks. Magnetic pile-up regions of the two planets are dominated by oxygen ions from planetary origin and are bounded outward by the MPB, an internal plasma boundary which could be a common feature of non-magnetized bodies. Topside termination of the ionosphere (ionopause) is not observed at Mars, it could be because the incident solar wind pressure usually exceeds the ionospheric one. Further analyses (in particular, Mars Express and Venus Express data) and, definitely, in situ observations (ESA Cosmic Vision program, other international programs) might yield additional information and most probably contrasts, in particular on the nightside which is, for example, almost unknown at Mars. We indeed do not know if

Mars and Venus Solar Wind Interactions

11

plasma holes similar to the ones observed at Venus33,34 are actually present at Mars.

References 1. C. T. Russell and O. Vaisberg, in Venus, eds. D. M. Hunten, L. Colin, T. M. Donahue and V. I. Moroz (University of Arizona Press, Tucson, AZ, USA, 1983), p. 873. 2. J. G. Luhmann, Space Sci. Rev. 44 (1986) 241. 3. J. G. Luhmann and L. H. Brace, Rev. Geophys. 29 (1991) 121. 4. M. H. Acu˜ na, J. E. P. Connerney, P. Wasilewski, R. P. Lin, K. A. Anderson, C. W. Carlson, J. McFadden, D. W. Curtis, D. Mitchell, H. R`eme, C. Mazelle, J. A. Sauvaud, C. d’Huston, A. Cros, J. L. Medale, S. J. Bauer, P. Cloutier, M. Mayhew, D. Winterhalter and N. F. Ness, Science 279 (1998) 1676. 5. D. H. Crider, Adv. Space Res. 33 (2004) 152. 6. T. K. Breus, S. J. Bauer, A. M. Krymskii and V. Y. Mitniskii, J. Geophys. Res. 94 (1989) 2375. 7. J. G. Trotignon, M. Parrot, J. C. Cerisier, M. Menvielle, W. I. Axford, M. Pa¨etzold, R. Warnant and A. W. Wernik, Planet. Space Sci. 48 (2000) 1181. 8. A. F. Nagy, D. Winterhalter, K. Sauer, T. E. Cravens, S. Brecht, C. Mazelle, D. Crider, E. Kallio, A. Zakharov, E. Dubinin, M. Verigin, G. Kotova, W. I. Axford, C. Bertucci and J. G. Trotignon, Space Sci. Rev. 111 (2004) 33. 9. C. Mazelle, D. Winterhalter, K. Sauer, J. G. Trotignon, M. H. Acu˜ na, K. Baumg¨ artel, C. Bertucci, D. A. Brain, S. H. Brecht, M. Delva, E. Dubinin, M. Øieroset and J. Slavin, Space Sci. Rev. 111 (2004) 115. 10. M. Tatrallyay, C. T. Russell, J. D. Mihalov and A. Barnes, J. Geophys. Res. 88 (1983) 5613. 11. J. G. Trotignon, C. Mazelle, C. Bertucci and M. H. Acu˜ na, Planet. Space Sci. 54 (2006) 357. 12. J. G. Luhmann, Adv. Space Res. 12 (1992) 191. 13. J. G. Trotignon, R. Grard and S. Savin, J. Geophys. Res. 96 (1991) 11,253. 14. F. L. Scarf, D. A. Gurnett and W. S. Kurth, Nature 292 (1981) 747. 15. R. Grard, C. Nairn, A. Pedersen, S. Klimov, S. Savin, A. Skalsky and J. G. Trotignon, Planet. Space Sci. 39 (1991) 89. 16. R. J. Strangeway and G. K. Crawford, Adv. Space Res. 16 (1995) 125. 17. J. G. Trotignon, A. Trotignon, E. Dubinin, A. Skalsky, R. Grard and K. Schwingenschuh, Adv. Space Res. 26 (2000) 1619. 18. A. J. Kliore, R. Woo, J. W. Armstrong and I. R. Patel, Science 203 (1979) 765. 19. J. G. Luhmann, C. T. Russell, F. L. Scarf, L. H. Brace and W. C. Knudsen, J. Geophys. Res. 92 (1987) 8545. 20. J. L. Phillips, J. G. Luhmann and C. T. Russell, J. Geophys. Res. 89 (1984) 10676. 21. J. G. Trotignon, E. Dubinin, R. Grard, S. Barabash and R. Lundin, J. Geophys. Res. 101 (1996) 24,965.

12

J.-G. Trotignon

22. M. H. G. Zhang, J. G. Luhmann, A. J. Kliore and J. Kim, J. Geophys. Res. 95 (1990) 14,829. 23. W. B. Hanson, S. Sanatani and D. R. Zuccaro, J. Geophys. Res. 82 (1977) 4351. 24. J. G. Luhmann, J. Geophys. Res. 96 (1991) 18,831. 25. L. H. Brace, R. F. Theis and W. R. Hoegy, Planet. Space Sci. 30 (1982) 29. 26. L. H. Brace, H. A. Taylor, Jr., T. I. Gombosi, A. J. Kliore, W. C. Knudsen and A. F. Nagy, in Venus, eds. D. M. Hunten, L. Colin, T. M. Donahue and V. I. Moroz (University of Arizona Press, Tucson, AZ, USA, 1983), p. 779. 27. C. M. C. Nairn, R. Grard, A. Skalsky and J. G. Trotignon, J. Geophys. Res. 96 (1991) 11,227. 28. C. Bertucci, C. Mazelle, D. H. Crider, D. Vignes, M. H. Acu˜ na, D. L. Mitchell, R. P. Lin, J. E. P. Connerney, H. R`eme, P. Cloutier, N. F. Ness and D. Winterhalter, Geophys. Res. Lett. 30 (2003) 1099. 29. C. Bertucci, C. Mazelle, J. A. Slavin, C. T. Russell and M. H. Acu˜ na, Geophys. Res. Lett. 30 (2003) 1876. 30. F. M. Neubauer, Astron. Astrophys. 187 (1987) 73. 31. C. Mazelle, H. R`eme, J. A. Sauvaud, C. d’Huston, C. W. Carlson, K. A. Anderson, D. W. Curtis, R. P. Lin, A. Korth, D. A. Mendis, F. M. Neubauer, K. H. Glassmeir and J. Raeder, Geophys. Res. Lett. 16 (1989) 1035. 32. C. T. Russell, M. Ong, J. G. Luhmann, K. Schwingenschuh, W. Riedler and Ye. Yeroshenko, Adv. Space Res. 12 (1992) 163. 33. L. H. Brace, R. F. Theis, H. G. Mayr, S. A. Curtis and J. G. Luhmann, J. Geophys. Res. 87 (1982) 199. 34. H. P´erez-de-Tejada, J. Geophys. Res. 106 (2001) 211.

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


COMPARISON OF MICROWAVE OBSERVATIONS OF MARTIAN TEMPERATURE AND WINDS WITH GENERAL CIRCULATION MODEL SIMULATIONS TAKESHI KURODA∗ and PAUL HARTOGH Max-Planck-Institute for Solar System Research Max-Planck-Str. 2, D-37191 Katlenburg-Lindau, Germany ∗ [email protected]

Microwave observations of the temperature and wind in the middle atmosphere of Mars are compared with the results of simulations with the Martian general circulation model. The simulated global-mean mesospheric temperature during a northern summer solstice is ∼10 K lower than in spring and autumn equinoxes, which is consistent with the James Clerk Maxwell Telescope observation in 1996–1997, although the absolute values are 30–40 K higher than in the observations. The wind velocity in the middle atmosphere in the model is comparable to the observations, except that the easterly wind in the afternoon is ∼100 m s−1 weaker. A brief discussion for the discrepancies between the model and observation is provided.

1. Introduction Microwave observations of the Martian atmosphere from the ground or Earth-orbiter were performed to detect the vertical profiles of the temperature,1–5 wind,5,6 and distributions of molecules3 from the spectral lines of CO, H2 O, and O2 . Microwave observations have three major advantages in comparison with the infrared observations such as the Thermal Emission Spectrometer onboard Mars Global Surveyor (MGS-TES). First, a more accurate retrieval is possible. Radiative transfer calculations have fewer unknowns, because the direct scattering and emission by Martian dust are negligible due to small particle sizes (∼2 µm) relative to the observing wavelength. The change of atmospheric temperature due to the planet-encircling dust storm was observed using the Earth-orbiting Submillimeter Wave Astronomy Satellite (SWAS). The detected change of the global-mean temperature consistent with the MGSTES observations was obtained below ∼ 40 km4 . 13

14

T. Kuroda and P. Hartogh

Second, microwave measurements are possible for higher altitudes. Due to high sensitivity and spectral resolution of the heterodyne technique, the latter can probe temperatures from the surface to above 80 km (compared to ∼40 km for MGS-TES).1,2,4 The SWAS observations detected that the planet-encircling dust storm has little impact on the global temperature structure above ∼60 km, the upper limit of MGS-TES measurements.4 In addition, the James Clerk Maxwell Telescope (JCMT) on Mauna Kea, Hawaii observed the temperature of the Martian mesosphere (height of 50–80 km) in 1996–1997, during the Martian northern spring and summer. It showed that the global-mean dayside temperature becomes 125–140 K for 50–60 km height and drops to ∼120 K at 70–80 km, that is to temperatures at which a local CO2 condensation can occur,2 as observed by the descending Pathfinder spacecraft.7 The observations with the Heinrich Hertz Telescope (HHT, Mount Graham, Arizona) in 1996/1997 also show results overall consistent with JCMT (Hartogh, poster on AGU fall meeting, 1997). Third, the direct observation of wind is possible using microwave instruments. Winds can be retrieved from Doppler shifts detected on the rotational transitions of CO. Observations with JCMT show easterly winds of 120–200 ms−1 at ∼20◦ S and the poleward winds of 30–40 ms−1 in both hemispheres at 35–80 km during Ls = 254◦ (the northern autumn).6 We should emphasize that direct observation of the wind velocity in the Martian atmosphere has never been done with other methods. In this paper, we simulate the Martian atmospheric temperature and wind velocity using a general circulation model (GCM),8 and compare the model results to the results of JCMT microwave observations. The comparison with the SWAS observations before and after the onset of a global dust storm in 2001 was already published.9 Here we show the simulated global-mean temperature of the mesosphere in the northern spring and summer, and wind velocity distributions in the northern autumn. The GCM used in this paper is briefly described in Sec. 2. The results of the simulations are presented in Sec. 3. The discussions and comparison between the model and observations are given in Sec. 4. 2. Model Description A detailed description of the Martian GCM called MAOAM was given in Ref. 8. The radiative CO2 scheme, which accounts for non-LTE (breakdown of the Local Thermodynamic Equilibrium), and the dust radiation scheme

Microwave Observations of Martian Temperature and Winds

15

are the same as described in Ref. 9, except that the dust-mixing ratio is smaller. This is because global dust storms, which can significantly affect the thermal structure of the middle atmosphere, were not observed either in 1996–1997 or in 2003. The “TES2 dust scenario” (described in Ref. 10 in full details) is used for the definition of the dust opacity depending on time and latitude. The global mean visible dust opacity of less than 0.3 is prescribed for the whole year. Effects of dust on the results of simulations presented in this paper are very small when only the atmosphere above ∼50 km is considered.

3. Model Results Figure 1 shows the simulated “disk-averaged” temperature for Ls = 0◦ – 180◦ (from a northern spring equinox to an autumn equinox) at 50–90 km. The local time at disk center varied between 09:00 and 15:00, and the subEarth latitude of the observation also varied between 10◦ N and 27◦ N, during the JCMT observation in 1996–1997.2 They possibly affected the results for the observed temperature; therefore, we considered them for the plots in Fig. 1. To construct “disk-averaged” quantities, we reprocessed the model output in accordance with the appearance of the planet’s disk. The accurate ephemeris was used to obtain the sub-Earth local time and latitude for particular dates, and the weighting function for the “visible” grid points based on the corresponding viewing geometry was utilized, as described in Ref. 9 for Figs. 2 and 3. Figure 2 shows the observational results from JCMT in 1996–1997, Viking and Mars Pathfinder descent measurements

Fig. 1. “Disk-averaged” atmospheric temperature simulated with MAOAM-GCM at Ls = 0◦ –180◦ .

16

T. Kuroda and P. Hartogh

Fig. 2. Temperature profiles in the Martian mesosphere2 obtained with: JCMT submillimeter measurements, Viking and Mars Pathfinder descent measurements, and NRAO/IRAM millimeter observations during the 1994 global dust storm.

and NRAO/IRAM millimeter measurements in 1994 (during a global dust storm).2 The simulated “disk-averaged” temperature (Fig. 1) is 30–40 K warmer than in the JCMT observations, and 10–20 K warmer than in the Viking descend measurements. However, it is consistent with the JCMT observations in that the simulated temperature at 60–70 km during a northern summer solstice is ∼10 K lower than in spring and autumn equinoxes. A characteristic feature of the current MAOAM-GCM is strong winter polar warming that tends to be produced11 due to the non-LTE CO2 radiation scheme12 which underestimates the radiative cooling. If it is replaced by the LTE one13 (e.g., as in Ref. 10), the simulated temperature becomes ∼ 10 K colder. This temperature decrease occurs due to the neglect of non-LTE effects and the associated overestimate of the infrared cooling rates that result in weaker simulated meridional transport. With stronger meridional circulation simulated with the non-LTE scheme, the adiabatic heating in the mesosphere increases, and the diabatic cooling decreases.11 The temperature profile simulated with the LTE CO2 radiation scheme is comparable with the Viking descend temperatures. However, note that Viking did not sample temperatures at exactly same latitudes and seasons compared to the MAOAM-GCM simulations. The microwave observations provide only globally averaged temperatures, and do not have the spatial resolution that would allow separating polar warmings.

Microwave Observations of Martian Temperature and Winds

17

Fig. 3. (a) Zonal and (b) meridional wind velocity at 60 km at Ls = 254◦ simulated with MAOAM-GCM (averaged during half a day, from sub-solar longitude of 45◦ E to 135◦ W). The center of the Martian disk corresponds to the equator during a local noon. The top, bottom, left, and right of the disk correspond to the North Pole, South Pole, morning, and evening limbs, respectively. The shaded circles represent the corresponding telescope beam locations used for obtaining the wind velocity shown in Fig. 4.

Figure 3 shows the simulated zonal and meridional wind velocities at 60 km for Ls = 254◦ . The velocities in this figure are reconstructed by averaging the model output in the “visible” grid points for half a day. The sub-solar longitude varies from 45◦ E to 135◦W during this period; the center of each disk corresponds to a local noon. The top, bottom, left, and right of the disk correspond to the North Pole, South Pole, morning, and evening limbs, respectively. The four shaded circles in this figure represent the corresponding telescope beam locations during the JCMT measurement on September 4, 2003 (Fig. 4). At each sub-solar longitude during this period, the easterly wind velocity in the morning is 40–60 ms−1 larger than in the afternoon, i.e. the local-time dependence of zonal wind velocity is apparently more significant than the longitude dependence. For the meridional wind, the maximum velocity of the northward wind in the northern hemisphere at a sub-solar longitude of 135◦ W is ∼40 ms−1 smaller than of 45◦ E, while that of southward wind in the southern hemisphere is ∼30 ms−1 larger. The meridional wind velocities in both hemispheres and easterly wind in the morning are comparable in the observations and simulations. However, the easterly wind in the afternoon is ∼ 100 ms−1 larger in the observation than in the model.

18

T. Kuroda and P. Hartogh

Fig. 4. Schematic of four antenna beam locations on the Martian disk. Wind velocities and directions at ∼ 60 km are indicated in each location for the September 4, 2003 observation.5

4. Discussions The results of the microwave observations of temperature and wind in the Martian mesosphere are compared with the GCM simulations. The wind velocity was directly measured from Doppler shifts of CO lines. We emphasize that these wind data obtained from the microwave measurements are unique for the Martian atmosphere. The descending landers (Viking and Mars Pathfinder) performed the only other available wind measurements. The simulated temperature in the Martian mesosphere is considerably higher than shown by the JCMT microwave data. However, the model temperature at 60–70 km is ∼10 K lower in summer solstice than in spring and autumn equinoxes, which is consistent with the measurements. One possible reason for this discrepancy is the strong winter polar warming that is simulated in the model.11 Moreover, it is still difficult to determine the characteristics of the mesospheric temperature, because the available observational data are very sparse, and there are differences of 20–30 K between the observational data (Fig. 2). Ref. 2 describes that the higher (compared to other measurements) temperature derived from the Viking descend is due to either the effect of dust raised to higher altitudes, or strong near-infrared heating by CO2 in the northern summer. Because the Viking descend was made during the daytime in summer at the latitude of ∼22◦ N, the observed temperature represents a local snapshot, whereas the JCMT observations cover the global-mean temperature. More observational data are needed for constraining and validating model results.

Microwave Observations of Martian Temperature and Winds

19

As for the wind velocity, the large difference between the measurements and simulations is seen only for the easterly wind in afternoons. Other Martian GCM14 did not reproduce the strong easterly wind observed by JCMT either, as described in Ref. 5. To a large degree, the simulated wind velocity is related to the temperature distribution through the so-called thermal wind relation: 
g ∂T 2u tan φ ∂u =− (1) f+ a ∂z aT ∂φ where f is the Coriolis parameter, u the zonal wind velocity, φ the latitude, a the radius of planet, z the height, g the acceleration of gravity, and T is the temperature. To maintain the easterly wind velocity of ∼180 ms−1 at 60 km from 1, as observed by JCMT, the value of ∂T /∂φ ∼ −16 K is required. For example, the temperature at the latitude of 57◦ S must be ∼16 K higher than that at the equator, from the surface and up to 60 km. In the model, ∂T /∂φ is almost zero from the surface up to ∼40 km at the local times 2–3 pm, which explains a weaker easterly wind. The atmospheric heating in midlatitude and polar region in the southern hemisphere below ∼60 km should be mainly due to the heating by CO2 infrared band or dust. Our GCM uses a simplified parameterization14 for heating effects by CO2 infrared band, therefore the improvement of it or of the dust distribution might contribute to the production of stronger easterlies. In addition, ion drag may affect the wind velocities. Because of the thin air and weaker magnetic field, the effects of ions can influence the middle atmosphere of Mars. Acknowledgments This work was supported by Deutsche Forschungsgemeinschaft (DFG), project HA 3261/1-2. The authors are grateful to Dr. Alexander S. Medvedev and an anonymous reviewer for helpful comments on the original manuscript. References 1. R. T. Clancy, D. O. Muhleman and G. L. Berge, Global changes in the 0–70 km thermal structure of the Martian atmosphere derived from 1975 to 1989 microwave CO spectra, J. Geophys. Res. 95 (1990) 14543. 2. R. T. Clancy and B. J. Sandor, CO2 ice clouds in the upper atmosphere of Mars, Geophys. Res. Lett. 25 (1998) 489.

20

T. Kuroda and P. Hartogh

3. M. A. Gurwell et al., Submillimeter wave astronomy satellite observations of the Martian atmosphere: temperature and vertical distribution of water vapor, Astrophys. J. 539 (2000) L143. 4. M. A. Gurwell, E. A. Bergin, G. J. Melnick and V. Tolls, Mars surface and atmospheric temperature during the 2001 global dust storm, Icarus 175 (2005) 23. 5. R. T. Clancy, B. J. Sandor, G. H. Moriarty-Schieven and M. D. Smith, Mesoscale winds and temperatures from JCMT sub-millimeter CO line observations during the 2003 and 2005 Mars oppositions, Abstract of “Second Workshop on Mars Atmosphere Modelling and Observations”, Granada, Spain (2006), pp. 6. 6. M. Lellouch, J. J. Goldstein, S. W. Bougher, G. Paubert and J. Rosenqvist, First absolute wind measurements in the middle atmosphere of Mars, Astrophys. J. 383 (1991) 401. 7. J. T. Schofield et al., The Mars Pathfinder Atmospheric Structure Investigation/ Meteorology (ASI/MET) experiment, Science 278 (1997) 1752. 8. P. Hartogh, A. S. Medvedev, T. Kuroda, R. Saito, G. Villanueva, A. G. Feofilov, A. A. Kutepov and U. Berger, Description and climatology of a new general circulation model of the Martian atmosphere, J. Geophys. Res. 110 (2005) E11 doi:10.1029/ 2005JE002498. 9. T. Kuroda, A. S. Medvedev and P. Hartogh, Martian atmosphere during the 2001 global dust storm: observations with SWAS and simulations with a general circulation model, in Advances in Geosciences. Vol. 3: Planetary Science (PS) (2006), (World Scientific Publishing, Singapore, 2006), pp. 145– 154 10. T. Kuroda, N. Hashimoto, D. Sakai and M. Takahashi, Simulation of the Martian atmosphere using a CCSR/NIES AGCM, J. Meteorol. Soc. Jpn. 83 (2005) 1. 11. A. S. Medvedev and P. Hartogh, Winter polar warmings and the meridional transport on Mars simulated with a general circulation model, Icarus 186 (2007) 97. 12. O. A. Gusev and A. A. Kutepov, Non-LTE gas in planetary atmospheres, in Stellar Atmosphere Modeling, eds. I. Hubeny, D. Mihalas and K. Werner, ASP Conf. Ser. 288 (2003) 318. 13. T. Nakajima and M. Tanaka, Matrix formulations for the transfer of solar radiation in a plane-parallel scattering atmosphere, J. Quant. Spectrosc. Radiat. Transfer 35 (1986) 13. 14. F. Forget, F. Hourdin, R. Fournier, C. Hourdin, O. Talagrand, M. Collins, S. R. Lewis, P. L. Read and J.-P. Huot, Improved general circulation models of the Martian atmosphere from the surface to above 80 km, J. Geophys. Res. 104 (1999) 24155.

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


ASTEROID COMPOSITIONS: SOME EVIDENCE FROM POLARIMETRY A. CELLINO∗ and M. DI MARTINO† INAF, Osservatorio Astronomico di Torino strada Osservatorio 20, 10025 Pino Torinese, Italy ∗ [email protected] † [email protected] A.-C. LEVASSEUR-REGOURD Univ. P. & M. Curie (Paris VI)/Aeronomie CNRS-IPSL BP 3, 91371 Verrieres, France [email protected] I. N. BELSKAYA Astronomical Institute of Kharkiv National University Sumska str. 35, Kharkiv 61022, Ukraine [email protected] Ph. BENDJOYA LUAN UMR 6525, Universit´ e de Nice, Parc Valrose, 06108 Nice cedex 2, France [email protected] R. GIL-HUTTON Complejo Astronomico El Leoncito (Conicet) and San Juan National University, Av. Espa˜ na 1512 sur, J5402DSP San Juan, Argentina [email protected]

Although it cannot provide direct and unambiguous information on the mineralogical composition of an asteroid surface, polarimetry is a very useful tool to get an improved understanding of parameters which are intimately related to surface composition and regolith structure. In recent times there has been a revival in the field of asteroid polarimetry, on the theoretical side, in relation to experimental simulations, and due to the activity of some teams who are engaged in extensive observational campaigns. Some new discoveries of objects exhibiting unprecedented polarimetric properties have been done. The above subjects are briefly reviewed.

21

22

A. Cellino et al.

1. Introduction The visible light that we receive from the asteroids and other atmosphereless bodies of our solar system is in a state of partial linear polarization, as a consequence of the fact that it consists of solar radiation scattered by the solid surfaces of the objects. The polarization properties of sunlight scattered by atmosphereless solar system bodies have been investigated since a long time, because in principle they can be a source of information about the physical properties of the materials present on the surfaces of these bodies. The first pioneering investigations in this field were carried out by Lyot,1 and were later continued by Dollfus et al. at the Paris-Meudon Observatory, and subsequently by other researchers in different countries. The historical background of asteroid polarimetry was briefly summarized in a classical chapter of the Asteroids II book.2 The observations allow the observers to directly measure the degree of polarization of light coming from an asteroid. The state of polarization of a light beam is described by the Stokes parameters Q and U (giving the degree of linear polarization), V (related to circular polarization), and I (the total intensity of the received light). In asteroid polarimetry, the V parameter is usually negligible, and the light is in a state of partial linear polarization described by the Stokes parameters Q and U . The observations show that the plane of linear polarization is generally either parallel or perpendicular to the scattering plane, which is defined as the plane containing the asteroid, the Sun and the observer at the epoch of observation. This fact is a consequence of the sunlight scattering process across the surface of the body (see also below). The parameter that is usually adopted to describe the polarimetric where P is the degree of linear behavior of asteroids is Pr = P cos(2θ),  polarization, given in module by Q2 + U 2 , and θ is the angle between the measured direction of the plane of partial linear polarization (defined by the observed position angle, given by arctan(U/Q) and the normal to the scattering plane. When the Pr parameter is measured in different conditions of illumination, described by the phase angle (the angle between the directions to the Sun and to the Earth as seen from the asteroid) a well defined relation between Pr and the phase angle is usually found. The typical situation is shown in Fig. 1, for the case of asteroid (1) Ceres. As can be seen, the relation is characterized by the presence of a range of phase angles, between 0 and about 20◦ , in which Pr is negative (the so-called branch of negative polarization). At larger phase angles, Pr

Asteroid Compositions

23

Fig. 1. A typical example of a Pr –phase curve, corresponding to the asteroid (1) Ceres. Data taken from the PDS archive (open circles) and from more recent observations carried out at the CASLEO observatory (filled circles). The phase angle, in degrees, is indicated by the α symbol.

becomes positive. The phase angle at which Pr changes sign is called the inversion angle. Due to obvious geometric constraints, the branch of positive polarization cannot be sampled beyond phase angles of the order of 30–35◦ in the case of main belt asteroids. The polarimetric behavior at much larger phase angles, however, is well documented in the case of some near-Earth asteroids, including (4179) Toutatis and (25143) Itokawa.3,4 Around the inversion angle, the trend of variation of Pr as a function of phase is mostly linear. The slope of this linear trend is usually indicated by the symbol h, and is an important parameter, because an empirical relation is known to exist between h and the geometric albedo pV of the surface (in V light). This relation may be written in the form log pV = C1 log h + C2 , and a similar relation is also found between pV and the absolute maximum of negative polarization, usually called Pmin , which is usually reached at phase angles between 8◦ and 10◦ . The existence of these relations between the albedo and the polarimetric properties of the objects constitutes one of the best available techniques to derive asteroid albedos.2,5,6

24

A. Cellino et al.

2. Theory and Experiments The interpretation of the polarimetric properties exhibited by the asteroids and other atmosphereless solar system bodies is still a challenge, although significant advances in this field have obtained in recent years. At present, there are not exact analytical formulae fully describing the phenomena of light scattering in situations corresponding to the case of sunlight scattered by asteroid surfaces. These situations essentially consist of light scattering by close-packed random media of inhomogeneous particles having sizes larger than the light wavelength.7 Light scattering mechanisms must be responsible not only of the observed polarimetric properties, but also of the photometric properties, including the observed phase–brightness relation. In particular, observations show that there is a mostly linear variation of brightness upon the phase angle (the objects becoming increasingly fainter for increasing phase angle), but at small phase angles a considerable non-linear increase of brightness is observed (the socalled “opposition effect”). In recent times, it has been widely accepted the idea that both the observed polarimetric and photometric properties of the asteroids can be explained in terms of two major mechanisms: coherent backscattering and shadowing. The former mechanism is based on constructive interference of scattered electromagnetic waves in presence of multiple scattering. Interested readers can find a description of this mechanism and appropriate references to previous work in the Muinonen et al. chapter in the Asteroids III book.7 The shadowing mechanism is essentially due to the fact that a photon incident on a particle on the surface can always be scattered back along the same direction of incidence, whereas along other directions it can be blocked by the presence of other surface particles.7 According to current understanding, the coherent backscattering mechanism plays a role in the generation of both the observed brightness and polarization phase relations, while shadowing should contribute mainly in determining the brightness opposition effect. In particular, the latter seems to be nicely explained as a consequence of both a lack of shadowing at zero-phase, as well as by constructive interference of light scattered from the surface.7 Theoretical studies are also complemented by laboratory experiments. In this respect, many authors have produced very useful measurements in the past. Based on laboratory data, it seems that both the depth of the branch of negative polarization and the value of the inversion angle strongly depend on the albedo and microscopic inhomogeneity of the investigated material samples, as well as on their packing density. A problem has been

Asteroid Compositions

25

for a long time the lack of experiments at phase angles smaller than 1◦ , that are very important from the theoretical point of view. The reason is that for this purpose special instruments are needed, with small angle apertures of both the light source and the receiver. Another problem is that laboratory measurements at extremely small phase angles require a very high accuracy, because the polarization degree is typically close to zero at these angles. Some attempts at improving the situation in this respect have been done only in recent years.8 We note that laboratory experiments including polarimetric and albedo measurements at or very close to zero phase angle are very important also from the perspective of complementing available astronomical observations with laboratory data obtained in similar illumination conditions. In particular, the definition of the geometric albedo of the asteroids is based on their reflectivity at zero phase angle, thus any rigorous attempt of deriving in the future a better calibrations of the polarimetric slope–albedo relation (see above) cannot include laboratory experiments if they are not made very close to zero phase angle. The reason is that the measured luminosity at zero phase can be strongly affected by the non-linear brightness opposition surge. To summarize the current situation, it can be said that we have today a better understanding of the most subtle physical effects involved in the generation of the photometric and polarimetric properties of the radiation we receive from the asteroids. In particular, the role of the coherent backscattering mechanism is now fully appreciated. There are still some problems in creating models able to reproduce at the same time and in details both the magnitude–phase relation and the polarimetric properties observed for the asteroids. In particular, the wide extension of the negative polarization branch in a range of phase angles of about 20◦ is still a challenging feature, although wide negative polarization branches are observed in some laboratory experiments, or can be numerically modeled assuming single-particle scattering. On the other hand, it seems likely that further advances in the modeling side are still possible, and the subject is currently being actively investigated by different teams.9

3. The Role of Polarimetry in Asteroid Taxonomy Since many decades it has been realized that asteroids differ in terms of color and albedo. This led to the development of taxonomic classes based on these properties. The general idea, especially at the beginning,

26

A. Cellino et al.

was that the differences in reflectance properties among different objects could be directly related to differences in mineralogic compositions. For this reason, some of the first taxonomic classes identified in the first pioneering works were called using symbols that were strictly related to a mineralogic interpretation (i.e., S for “silicates”, M for “metals”, etc.). Analyses of the first available spectrophotometric and polarimetric data-sets revealed that spectral reflectance properties alone were not sufficient in many cases to derive a full taxonomic classification, since it turned out that different classes existed which exhibited the same spectrophotometric properties in visible wavelengths, but were characterized by huge differences in albedo, based on their polarimetric properties.10 The best example is given by the EM P complex, formed by three separate classes (E, M , and P ), which have fairly identical spectra but largely differ in terms of albedo. In more recent years, some large systematic surveys (SMASS, SMASS211 ) have provided spectral data for big samples of objects, from which a taxonomy has been derived. This classification is no longer based on simple colorindexes using limited numbers of filters, but on full reflectance spectra, covering wavelength ranges approximately between 0.5 and 0.9 µm. For the vast majority of these objects no polarimetric data are available, and the albedo is not known, and attempts have been made to separate different taxonomic classes based on the presence of subtle features of the reflectance spectra.11 In particular, the E, M , and P classes correspond now to different subsets of a bigger complex called X. These subsets are distinguished on the basis of spectroscopic features that have been found to characterize objects belonging to the E, M , and P classes defined in previous, albedo-based classifications. This does not mean, however, that polarimetrically-derived albedos and, more in general, polarimetric properties are no longer useful for taxonomy-related purposes, or for better understanding the properties of asteroid surfaces. In this respect, the albedo per se is a very important parameter, being strictly related to the composition and aging of the surface. It should be stressed that polarimetry must be considered as the best available technique to derive asteroid albedos. The reason is that the existence of a direct relation between the polarimetric slope (and also Pmin ) and the albedo makes it possible to derive the albedo directly from the observed polarimetric parameters, without the need of knowing any other parameter. This is a big advantage with respect to other possible techniques, which cannot measure directly the albedo, but derive it more indirectly, for instance from measurements of the size, and based on nominal values of the absolute magnitude. This is the case, for instance, of thermal radiometry.

Asteroid Compositions

27

Fig. 2. Pmin vs. inversion angle (αinv ) plot for a set of asteroids belonging to different taxonomic classes11,13 (used as symbols in the plot). Data available in the literature. 14–17

Moreover, polarimetry has long been found to be an important tool for taxonomic characterization purposes. As an example, Fig. 2 shows that in a plot of Pmin vs. polarimetric inversion angle, not only there is a clear separation between low (F , B, C, G, D) intermediate (S, M ) and highalbedo (E) objects, but also there is a fairly clear separation even among objects of similar albedo. As a matter of fact, a recent analysis12 has convincingly shown that the availability of a good coverage of the phase– polarization curve for a sample of objects is sufficient to derive a taxonomic classification in very good agreement with that produced by spectroscopic data. In particular, a principal component analysis applied to a set of phase–polarization curves described by a polynomial or trigonometric representation has been found to produce a set of taxonomic classes that very nicely fit the classes obtained by means of purely spectroscopic data. This interesting result suggests that polarimetry and spectroscopy are nicely complementary for taxonomic purposes.

4. Recent Observational Results From the point of view of observations, for a long time and until some years ago, activity in the field of asteroid polarimetry has not been very intense. There are several reasons for that, including (1) a scarce availability of suitable instruments, (2) the fact that asteroid polarimetry is intrinsically

28

A. Cellino et al.

more time-consuming than other observing techniques, (3) the relative rarity of experts in the field, which is also a consequence of the somewhat particular characteristics of polarimetry, being often considered as a fairly obscure technique whose results seem mostly based on empirical and nottoo-well understood laws. The items (1) and (2) in the above list are straightforward: since a polarimetric measurement implies the need of splitting the incident light beam into an ordinary and an extraordinary ray, in order to derive the Stokes parameters, fairly large telescopes are needed for the observations of faint targets. Moreover, asteroid polarimetry is intrinsically time-consuming due to the need of observing each object over fairly long intervals of time, to obtain a sufficient sampling of the phase–polarization curve. In other words, asteroid polarimetry is not for those who want to have one publication per one night of observations. In spite of the above difficulties, in recent years there has been a significant increase of activities in the field. A significant role has been played by the availability of the Torino photopolarimeter, which equips the 2.15 m telescope of the El Leoncito observatory (Argentina). This instrument has produced in recent years a significant amount of data,6,15 nicely complementing the observing activities of other teams, mainly in Ukraine, who have been working in this field since a long time. The most recent observing campaigns carried out by different authors had a variety of purposes. These include a systematic check of the albedo values derived by thermal radiometry observations of small main belt asteroids, an analysis of the polarimetric behavior of different objects observed at very small phase angles, an exploration of the properties of the branch of positive polarization which, in the case of near-Earth objects, is accessible up to very large values of phase angle, and, more pertinent to the subject of the present paper, comparative analyses of the polarimetric properties of objects belonging to different taxonomic classes. In this respect, at least a couple of interesting results have been obtained recently, namely an extensive characterization of the polarimetric properties of F -type asteroids, and the discovery of the unusual polarimetric properties of (234) Barbara, a rare Ld-type object. The above-mentioned examples deal with phase–polarization curves characterized by extreme and opposite properties. In the case of the F -type asteroids, the most striking polarimetric property, as shown also in Fig. 2, is the very low value of the inversion angle.16 The F taxonomic class includes objects characterized by low-albedo, and linear, featureless reflectance spectrum. F -type asteroids are believed to be primitive, representing a

Asteroid Compositions

29

subclass of the big C complex, from which they can be separated based on their spectral behavior at short wavelengths. Since this part of the spectrum was not included in the spectroscopic data used in the most recent SMASS2 taxonomic classification,11 the F class is not included among the classes identified in this survey. On the other hand, polarimetry indicates that the F -type asteroids certainly represent a separate class, exhibiting well defined and unique polarimetric properties. As mentioned above, F -type asteroids have been found to be characterized by unusually small values of the inversion angle. In the vast majority of the cases, for asteroids the Pr parameter changes sign at phase angles around 20◦ . In the case of F -type objects, however, the inversion takes place at much lower phase angles. In particular, 419 Aurelia exhibits an inversion angle at 14◦ , the smallest value ever observed for asteroids. Also the depth of its negative branch seems unusually low, for a low-albedo objects. Small inversion angles, but more usual negative branches, are also exhibited by other objects of the F class.16 The interpretation of these properties is not straightforward, but laboratory experiments and numerical simulations suggest that these might be explained by assuming that the surface regolith consists of particles characterized by very high optical homogeneity down to scales of the order of visible light wavelengths. The case of (234) Barbara is just the opposite. As shown in Fig. 3, in this case we deal with an object exhibiting a very high value of the inversion angle. The first measurements published for this object suggested a possible value of about 30◦ ,18 but more recent, and still unpublished V and R data (shown in Fig. 3) show that Pr seems to tend more rapidly to zero between 24 and 26◦ , and the inversion might take place at a phase angle around 28◦ . This behavior challenges theoretical interpretation. According to current knowledge, a large inversion angle may be expected for a regolith layer consisting of very regularly shaped particles (like spheres or crystals) and/or large optical inhomogeneity. As quoted in Sec. 2, a large value of the inversion angle is just one of the most challenging features to be reproduced by current theoretical models. In this respect, (234) Barbara is very interesting, since it exhibits the largest value of the inversion angle known today for any atmosphereless solar system body, and is possibly the prototype of a previously unknown class of asteroids, from the point of view of polarimetric properties. One of the most interesting facts concerning (234) Barbara is that it has been found to belong to a fairly rare taxonomic class that has been introduced only recently based on spectroscopic data collected by

30

A. Cellino et al.

Fig. 3. The phase–polarization curve of (234) Barbara, from observations in V and R colors by some of the authors of the present article. The data include some unpublished data obtained recently at the El Leoncito observatory. The phase angle, α, is given in degrees.

the SMASS2 survey.11 The objects now classified as Ld are a subclass of a larger class called L. Both L and Ld asteroids were previously classified as S, the dominant taxonomic class in the inner part of the asteroid main belt. Is the unusual taxonomic classification of (234) Barbara directly related to its peculiar polarimetric behavior? We have not yet answered this question, due to a lack of polarimetric data on other Ld-type objects. On the other hand, it is known that a few asteroids belonging to the wider L class for which observations are available exhibit “normal” polarimetric properties, as in the case of (12) Victoria. It is clear that new observations of both L and Ld-type asteroids are needed. What seems also clear at this stage is that polarimetric properties can be a powerful tool for investigating some properties of the surface regolith particles at a microscopic scale, nicely complementing the information that can be obtained by means of other techniques like spectroscopy. Of course, a strong effort on the modeling and theoretical side, as well as in the field of laboratory experiments, is still needed to increase the diagnostic power of polarimetric data. The wealth of new results obtained in recent years seems

Asteroid Compositions

31

to indicate that the field of research in asteroid polarimetry is currently experiencing a new era of rapid development.

5. Future Developments Apart from the perspectives in the field of theory and laboratory experiments, we want to focus here on some perspectives concerning future observational activities. Future observations will be devoted to a continuation of current efforts, but will probably be aimed also at exploring new fields of research, including an extensive analysis of the wavelength dependence of asteroid polarimetric properties. Another field in rapid development seems to be the study of near-Earth objects. These objects are interesting in many respects. First, they allow the observers to study the polarimetric properties at much larger phase angles, where available data are still scarce. Moreover, NEOs are important also from the point of view of the collision hazard. In this respect, a big effort is being produced by many teams to discover the most dangerous objects. A lot of work must be done, however, on the side of the study of the physical properties of these objects. In this respect, polarimetry can play an invaluable role as a powerful tool to derive the surface albedo, and consequently the sizes of the objects. Using instruments like the ESO VLT 8-m telescope, it is possible to efficiently obtain albedos and sizes of dangerous objects, as demonstrated by some recent observations of (99942) Apophis.19 The availability of new polarimeters in the future is needed to ensure a stable development of asteroid polarimetry. Recently, a few new instruments, including a single-color polarimeter using phototube detectors, has entered into operations at the El Leoncito observatory. Compared to the older Torino photopolarimeter, which performs simultaneous measurements in five colors (UBVRI), the new instrument has a more limited spectral capability, but this is compensated by an increased sensitivity in V light, which should allow the observers to observe fainter and darker objects. Another instrument which has been developed recently is the CCD polarimeter built at the Asiago observatory (Italy), which has started recently to produce its first data.17 Coupled with the recent availability of the VLT telescope for a number of asteroid polarimetry campaigns, we hope that the above-mentioned developments can be diagnostic of a real renaissance of asteroid polarimetry. We are convinced that a new burst of activity in this field can produce in

32

A. Cellino et al.

the next years very important advances in several fields of modern asteroid research, including primarily the study of the properties of asteroidal surfaces.

References 1. B. Lyot, Ann. Obs. Paris 8(I) (1929) 1. 2. A. Dollfus, M. Wolff, J. E. Geake, D. F. Lupishko, L. M. Dougherty, in Asteroids II, eds. R. P. Binzel, T. Gehrels and M. S. Matthews (University of Arizona Press, Tucson, 1989), pp. 594–616. 3. A. C. Levasseur-Regourd and E. Hadamcik, J. Quant. Spectrosc. Radiat. Transfer 79–80 (2003) 903. 4. A. Cellino, F. Yoshida, E. Anderlucci, Ph. Bendjoya, M. Di Martino, M. Ishiguro, A. M. Nakamura and J. Saito, Icarus 179 (2005) 297. 5. D. F. Lupishko and R. A. Mohamed, Icarus 119 (1996) 209. 6. A. Cellino, R. Gil Hutton, E. F. Tedesco, M. Di Martino and A. Brunini, Icarus 138 (1999) 129. 7. K. Muinonen, J. Piironen, Yu. Shkuratov, A. Ovcharenko and B. E. Clark, in Asteroids III, eds. W. F. Bottke, A. Cellino, P. Paolicchi and R. P. Binzel (University of Arizona Press, Tucson, 2002), pp. 123–138. 8. Yu. G. Shkuratov, A. Ovcharenko, E. Zubko, D. Stankevich, O. Miloslavskaya, K. Muinonen, J. Piironen, R. Nelson, W. Smythe, V. Rosenbush and P. Helfenstein, Icarus 159 (2002) 396. 9. A. C. Levasseur-Regourd, E. Hadamcik and J. Lasue, Adv. Space Res. 37 (2006) 161. 10. D. J. Tholen and A. Barucci, in Asteroids II, eds. R. P. Binzel, T. Gehrels and M. S. Matthews (University of Arizona Press, Tucson, 1989), pp. 298–315. 11. S. J. Bus and R. P. Binzel, Icarus 158 (2002) 146. 12. A. Penttil¨ a, K. Lumme, E. Hadamcik and A.-C. Levasseur-Regourd, Astron. Astrophys. 432 (2002) 1081. 13. D. J. Tholen, in Asteroids II, eds. R. P. Binzel, T. Gehrels and M. S. Matthews (University of Arizona Press, Tucson, 1989), pp. 1139–1150. 14. B. Zellner and J. Gradie, Astron. J. 81 (1976) 262. 15. A. Cellino, R. Gil Hutton, M. Di Martino, Ph. Bendjoya, I. N. Belskaya and E. F. Tedesco, Icarus 179 (2005) 304. 16. I. N. Belskaya, Yu. G. Shkuratov, Yu. S. Efimov, N. M. Shakhovskoy, R. Gil Hutton, A. Cellino, E. S. Zubko, A. A. Ovcharenko, S. Yu. Bondarenko, V. G. Shevchenko, S. Fornasier and C. Barbieri, Icarus 178 (2005) 213. 17. S. Fornasier, I. N. Belskaya, Yu. G. Shkuratov, C. Pernechele, C. Barbieri, E. Giro and H. Navasardyan, Astron. Astrophys. 455 (2006) 371. 18. A. Cellino, I. N. Belskaya, Ph. Bendjoya, M. Di Martino, R. Gil Hutton, K. Muinonen and E. F. Tedesco, Icarus 180 (2006) 565. 19. M. Delb` o, A. Cellino and E. F. Tedesco, submitted to Icarus (2006), in press.

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


LOW ENERGY CHARGED PARTICLE MEASUREMENT BY JAPANESE LUNAR ORBITER SELENE Y. SAITO∗ , S. YOKOTA, K. ASAMURA, T. TANAKA and T. MUKAI Institute of Space and Astronautical Science Japan Aerospace Exploration Agency 3-1-1 Yoshinodai, Sagamihara Kanagawa 229-8510, Japan ∗ [email protected]

SELenological and ENgineering Explorer (SELENE) is a Japanese lunar orbiter that will be launched in 2007. The main purpose of this satellite is to study the origin and evolution of the Moon by means of global mapping of element abundances, mineralogical composition, and surface geographical mapping from 100 km altitude. Plasma energy Angle and Composition Experiment (PACE) is one of the scientific instruments onboard the SELENE satellite. The scientific objectives of PACE are (1) to measure the ions sputtered from the lunar surface and the lunar atmosphere, (2) to measure the magnetic anomaly on the lunar surface using two electron spectrum analyzers (ESAs) and a magnetometer onboard SELENE simultaneously as an electron reflectometer, (3) to resolve the Moon–solar wind interaction, (4) to resolve the Moon– Earth’s magnetosphere interaction, and (5) to observe the Earth’s magnetotail. PACE consists of four sensors: ESA-S1, ESA-S2, ion mass analyzer (IMA), and ion energy analyzer (IEA). ESA-S1 and S2 measure the three-dimensional distribution function of low energy electrons below 15 keV, while IMA and IEA measure the three-dimensional distribution function of low energy ions below 28 keV/q.

1. Introduction Low energy charged particles around the Moon were vigorously observed by Moon-orbiting satellites and plasma instrumentation placed on the lunar surface in 1960s and 1970s.1–7 Many new discoveries concerning the lunar plasma environment were made during the period. Though there are some satellites that explored the Moon afterwards, most of them were dedicated to the global mapping of the lunar surface.8–10 Except the low energy electron measurement by Lunar Prospector,9 and the lunar wake plasma data obtained by the WIND satellite11 during the Moon fly-by, almost 33

34

Y. Saito et al.

no new information about the low energy charged particles around the Moon was obtained. SELenological and ENgineering Explorer (SELENE) is a Japanese lunar orbiter that will be launched in 2007. SELENE will study the origin and evolution of the Moon by means of global mapping of element abundances, mineralogical composition, and surface geographical mapping from 100 km altitude. One of the scientific instruments, Plasma energy Angle and Composition Experiment (PACE), was developed for making comprehensive three-dimensional plasma measurement around the Moon. After describing the science objectives of PACE, the configuration of the PACE sensors is shown.

2. Science Objectives 2.1. Ions originated from the lunar surface and the lunar atmosphere The research of the lunar atmosphere and lunar surface material is one of the most important aims of PACE on SELENE. Ground-based observations revealed the existence of tenuous alkali-atmosphere around the Moon in the end of 1980s. Potter and Morgan12 discovered the existence of Na and K atmospheres above the sunlit limb of the Moon for the first time. Since then several generation mechanisms are proposed for the rarefied lunar alkali-atmosphere.13 Sputtering caused by the solar wind ions has drawn considerable attention, because it produces secondary particles that reflect the lunar surface composition. On the other hand, Potter et al.14 found that solar photons play a dominant role in desorption of the lunar alkali atmosphere. According to them, sputtering of the lunar surface only contributes to the creation of the lunar Na atmosphere by enhancing diffusion of Na to the lunar surface. Sputtered or desorped particles from the lunar surface are mainly composed of neutrals, which are ionized by solar photons and electrons. Both ionized particles and sputtered/desorped ions are accelerated and transported by the solar wind in cycloidal motion. The ion analyzer on board the AMPTE satellite obtained mass spectra of picked-up ions at several Rm distance behind the Moon.15 The distribution of tenuous lunar atmosphere will be obtained by the in-situ low-energy ion measurement by PACE. PACE will also remotely reveal the global composition of the lunar surface by means of detecting ions sputtered by the solar wind and tracing them back to their sputtering point like to laboratory secondary ion mass spectrometry.16

Low Energy Charged Particle Measurement

35

2.2. Electron reflectometer The solar wind electrons and the magnetotail electrons that reach the Moon will be absorbed if there is no magnetic field on the lunar surface. However, with the existence of the remnant magnetic field on the Moon, the electrons moving with large angle around the ambient magnetic field will be mirror reflected back to SELENE. Measuring the pitch angle distribution of the reflected electrons, the remnant magnetic field on the lunar surface can be deduced.4,17 The previous remnant magnetic field measurements using mirror-reflected electrons were conducted by the Apollo 15,16 sub-satellites whose orbits were limited around the equator region of the Moon.18,19 Lunar Prospector also measured remnant magnetic field on various areas of the Lunar surface using electron reflectometer.20,21 The SELENE PACE-electron spectrum analyzer (ESA) sensors will survey the remnant magnetic field on almost all the lunar surface with higher spatial resolution than previous electron reflectometer measurement.

2.3. Moon–solar wind interaction It has a primary importance to study the structure of the lunar wake, and the behavior of plasma near the limb of the Moon. As widely accepted there is almost no intrinsic magnetic field around the Moon. Thereby, there is no well-defined bow shock as can be found in the terrestrial magnetosphere. Alternatively, the void region of plasma surrounded by the rarefaction region is created just behind the Moon. The lunar wake consists of these void and rarefaction regions. In the rarefaction region, the plasma is in a highly turbulent state, and various kinds of waves are existed and particles are accelerated and heated. We regard the lunar wake as a suitable and ideal region to study the formation of plasma turbulence. It is also known that there is some plasma compression at the limb of the Moon and the degree of the plasma compression has a regional dependence.22 Therefore, it is thought that there is a weak but some magnetic field anomaly on the surface of the Moon, and this deflects the solar wind particles and varies the degree of compression.23 We will be able to know more details of such phenomena and further verify proposed models of the solar wind and Moon interaction by conducting three-dimensional plasma particle measurement, obtaining their velocity momentum, i.e., the number density, the velocity, and the temperature.

36

Y. Saito et al.

2.4. Moon–Earth’s magnetosphere interaction The Moon provides “an eclipse” of the Earth’s magnetotail. Near the Moon in the Earth’s plasma sheet, there often exist earthward flowing plasmas, tailward flowing plasmas, and counter-streaming (earthward and tailward) flows. They are a mixture of solar wind origin and of ionospheric origin. It is very important to know origins and dynamics of these plasmas. With this knowledge, we understand the formation mechanism of the magnetosphere and the structure of the magnetotail. The Moon provides an excellent experimental tool to us. In the first approximation, the Moon is not any obstacle for the magnetic fields. However, the Moon is a strong absorber for plasmas. When the Moon is in the magnetotail, plasmas from the distant tail cannot reach any observer on the earthward side of the Moon, whereas plasmas from the Earth cannot reach any observer on the tailward side of the Moon. Therefore, we will be able to discriminate the tailwardflowing plasmas from the earthward-flowing plasmas. It will be possible to identify correctly the tailward-flowing component and the earthward flowing component in the counter-streaming flows. Furthermore, it will be possible to identify some “hidden” plasmas in the flowing plasmas.

2.5. Plasma measurement of the Earth’s magnetotail The comprehensive plasma measurement near the Moon orbit has been carried out only with the spacecraft Geotail. Geotail provided mainly the survey of plasmas in the Sun–Earth direction, not in the cross section of the magnetotail. The spacecraft around Moon makes a detailed survey of plasmas in the cross section of the magnetotail, so that we would be able to examine various boundaries in the magnetotail (i.e., the distant bow shock, the magnetopause, the magnetopause boundary layer, the low latitude boundary layer, the plasma sheet-tail lobe boundary). In the magnetotail, various plasma populations are controlled by the interplanetary magnetic field (IMF). The loading of ionospheric O+ in the magnetotail depends strongly on the IMF. With measurement of O+ distribution in the tail cross section, we will be able to understand the loading and transport mechanisms of ionospheric ions in the tail lobe/mantle. The draping of the magnetic field lines near the magnetopause and the deformation of the magnetotail will be correctly evaluated, in order to understand changes in the magnetotail structure. Furthermore, it will be possible to find magnetic reconnection process in the tail magnetopause region.

Low Energy Charged Particle Measurement

37

3. Instrument Configuration 3.1. ESA-S1 and ESA-S2 The ESA sensor basically utilizes a method of a top hat electrostatic analyzer with angular scanning deflectors at the entrance and toroidal electrodes inside (see Fig. 1). The field of view (FOV) is electrically scanned between ±45◦ around the center of the FOV that is 45◦ inclined from the axis of symmetry. With two ESA sensors that are installed in the +Z and −Z surface of the spacecraft, the three-dimensional electron distribution function is observed. The upper and lower angular deflectors are supplied with high voltage which are swept between 0 V and +4 kV. The inner toroidal electrode is also supplied with high voltage swept between 0 V and +3 kV simultaneously with the angular scanning deflectors. The electrons coming through the angular scanning deflectors are attracted down toward the inner electrode by the action of the applied potential. Only the electrons with specific energy range can further travel down to the exit of the

Fig. 1.

Cross section of ESA-S1 and ESA-S2.

38

Y. Saito et al.

electrodes. There also placed a slit and a mesh supplied with slightly negative voltage (−2.5 V to −10 V) to reflect the secondary electrons that give spurious counts. The electrons passing through the electrode enter to micro-channel plate (MCP) and are intensified to detectable charge pulses. Finally, the charge pulses are received by one-dimensional circular resistive anode. The positions where the charge pulses are detected correspond to the incident azimuthal directions of the electrons. Tables 1 and 2 summarize the specifications of ESA-S1 and ESA-S2, respectively.

3.2. IMA and IEA The IMA sensor consists of an energy analyzer that has the similar structure as the ESA sensors and a linear electric field (LEF) time of flight (TOF) ion mass analyzer24–26 (see Fig. 2). The IEA sensor consists of only an energy analyzer that is the same as the energy analyzer of IMA (see Fig. 3). The upper and lower angular deflectors of the energy analyzer are supplied with high voltage which are swept between 0 V and +5 kV. The inner toroidal electrode is also supplied with high voltage swept between 0 V and −4 kV simultaneously with the angular scanning deflectors. Between the toroidal electrode and the angular deflectors, there exist a pair of electrodes that serve as sensitivity control electrodes. Since the flux of the solar wind ions Table 1.

Specifications of ESA-S1.

Energy range Energy resolution Energy sweep step Field of view Angular resolution Time resolution FOV sweep range g-factor (5◦ × 22.5◦ )

Table 2. Energy range Energy resolution Energy sweep step Field of view Angular resolution Time resolution FOV sweep range g-factor (5◦ × 22.5◦ )

5 eV–10 keV 15% (FWHM) 32 2π str. 5◦ × 8◦ (FWHM) 1 second 45◦ ± 45◦ 10−3 cm2 str keV/keV

Specifications of ESA-S2. 5 eV–15 keV 10% (FWHM) 32 2π str. 5◦ × 8◦ (FWHM) 1 second 45◦ ± 45◦ 2 × 10−4 cm2 str keV/keV

Low Energy Charged Particle Measurement

Fig. 2.

39

Cross section of IMA.

and the lunar-origin ions differs significantly, the sensitivity of the energy analyzer can be reduced electrically to about 1/100 in case of the solar wind ion observation. The ions transmitted through the energy analyzer of IEA are detected by MCP with one-dimensional circular resistive anode. The ions transmitted through the energy analyzer of IMA are post accelerated and enters into the LEF TOF mass analyzer. Thin carbon foil is placed at the entrance of the LEF TOF mass analyzer, which generates start electrons when ions pass through the carbon foil. The start electrons are accelerated by the electric field inside the mass analyzer and their positions are detected by one-dimensional circular resistive anode that is placed behind the MCP. These start electrons also generate start signals when they pass through a

40

Y. Saito et al.

Fig. 3.

Cross section of IEA.

mesh anode that exist between the position sensitive anode and the MCP. Most of the ions that pass though the carbon foil lose their initial charge state and enter into the mass analyzer as neutral particles. These neutral particles are detected by an anode that is in the center of the position sensitive anode. These signals are used as stop signals. The mass/charge of the incident ions can be calculated from its energy/charge and the TOF. Some of the incident ions enter the analyzer as ions. These ions are reflected by the linear electric field whose intensity is proportional to the distance from the entrance point. The reflected ions generate secondary electrons when they collide with the top part of the mass analyzer. These electrons are accelerated and detected by the center anode, which generate the stop signals. Since the TOF of the reflected ions is proportional to the square root of the mass of the ions, the mass of the incident ions can be determined

Low Energy Charged Particle Measurement Table 3. Energy range Mass range Energy resolution Energy sweep step Mass resolution Field of view Angular resolution Time resolution FOV sweep range g-factor (5◦ × 22.5◦ ) Table 4. Energy range Energy resolution Energy sweep step Field of view Angular resolution Time resolution FOV sweep range g-factor (5◦ × 22.5◦ )

41

Specifications of IMA. 5 eV/q–28 keV/q 1–60 5% (FWHM) 32 m/∆m ∼ 15 2π str. 5◦ × 10◦ (FWHM) 1 second 45◦ ± 45◦ 10−6 –10−4 cm2 str keV/keV (variable) Specifications of IEA. 5 eV/q–28 keV/q 5% (FWHM) 32 2π str. 5◦ × 5◦ (FWHM) 1 second 45◦ ± 45◦ 10−6 –10−4 cm2 str keV/keV (variable)

precisely without being affected by the angular scattering and the energy degradation caused by the ion passage in the carbon foil.24,25

4. Conclusion SELENE will be launched in 2007. One of the scientific instruments, PACE, will measure three-dimensional distribution function of low energy ions and electrons at 100 km altitude around the Moon. With the minimum time resolution of 1 second, high spatial resolution measurement of magnetic anomalies on the lunar surface will be made. Since nobody has ever measured the three-dimensional distribution function of low energy ions at 100 km altitude around the Moon, it is expected that many unresolved problems concerning the lunar plasma environment will be elucidated by the PACE observation in the near future.

Acknowledgments The authors thank all the members of SELENE-MAP-PACE team for their valuable discussion on the specifications of PACE sensors and their

42

Y. Saito et al.

unstinting help in developing PACE sensors. SELENE-PACE sensors were manufactured by Mitaka Kohki Co. Ltd., Meisei Elec. Co., Hamamatsu Photonics K.K., and Kyocera Co.

References 1. E. F. Lyon, H. S. Bridge and J. H. Binsack, J. Geophys. Res. 72 (1967) 6113. 2. D. S. Colburn, R. G. Currie, J. D. Mihalov and C. P. Sonett, Science 158 (1967) 1040. 3. K. A. Anderson, L. M. Chase, R. P. Lin, J. E. McCoy and R. E. McGuire, J. Geophys. Res. 77 (1972) 4611. 4. H. C. Howe, R. P. Lin, R. E. McGuire and K. A. Anderson, Geophys. Res. Lett. 1 (1974) 101. 5. M. Neugebauer, C. W. Snyder, D. R. Clay and B. E. Goldstein, Planet. Space Sci. 20 (1972) 1577. 6. D. R. Clay, B. E. Goldstein, M. Neugebauer and C. W. Snyder, NASA Spec. Publ. 289 (1972) 10. 7. H. K. Hills, J. C. Meister, R. R. Vondrak and J. J. W. Freeman, NASA Spec. Publ. 289 (1972) 12. 8. S. Nozette, P. Rustan, L. P. Pleasance, J. F. Kordas, I. T. Lewis, H. S. Park, R. E. Priest, D. M. Horan, P. Regeon, C. L. Lichtenberg, E. M. Shoemaker, E. M. Eliason, A. S. McEwen, M. S. Robinson, P. D. Spudis, C. H. Acton, B. J. Buratti, T. C. Duxbury, D. N. Baker, B. M. Jakosky, J. E. Blamont, M. P. Corson, J. H. Resnick, C. J. Rollins, M. E. Davies, P. G. Lucey, E. Malaret, M. A. Massie, C. M. Pieters, R. A. Reisse, R. A. Simpson, D. E. Smith, T. C. Sorenson, R. W. V. Breugge and M. T. Zuber, Science 266 (1994) 1835. 9. A. B. Binder, Science 281 (1998) 1475. 10. B. H. Foing, G. Racca, A. Marini, E. Evrard, L. Stagnaro, M. Almeida, D. Koschny, D. Frew, J. Zender, J. Heather, M. Grande, J. Huovelin, H. Keller, A. Nathues, J. Josset, A. Malkki, W. Schmidt, G. Noci, R. Birkl, L. Iess, Z. Sodnik and P. McManamon, Adv. Space Res. 37 (2006) 6. 11. K. W. Ogilvie, J. T. Steinberg, R. J. Fitzenreiter, C. J. Owen, A. J. Lazarus, W. J. Farrell and R. B. Torbert, Geophys. Res. Lett. 23 (1996) 1255. 12. A. E. Potter and T. H. Morgan, Science 241 (1988) 675. 13. S. A. Stern, Rev. Geophys. 37 (1999) 453. 14. A. E. Potter, R. M. Killen and T. H. Morgan, J. Geophys. Res. 105 (2000) 15073. 15. M. Hilchenbach, D. Hovstadt, B. Klecker and E. M¨obius, Adv. Space Res. 13 (1993) 321. 16. R. C. Elphic, H. O. Funsten, B. L. Barraclough, D. J. McComas, M. T. Paffett, D. T. Vaniman and G. Heiken, Geophys. Res. Lett. 11 (1991) 2165. 17. K. A. Anderson, R. P. Lin, R. E. McGuire and J. E. McCoy, Space Sci. Instrum. 1 (1975) 439.

Low Energy Charged Particle Measurement

43

18. K. A. Anderson, R. P. Lin, R. E. McGuire, J. E. McCoy, C. T. Russell and P. J. Coleman, Earth Planet. Sci. Lett. 34 (1977) 141. 19. K. A. Anderson and D. E. Wilhelms, Earth Planet. Sci. Lett. 46 (1979) 107. 20. R. P. Lin, D. L. Mitchell, D. W. Curtis, K. A. Anderson, C. W. Carlson, J. McFadden, M. H. Acuna, L. L. Hood and A. Binder, Science 281 (1998) 1480. 21. J. S. Halekas, D. L. Mitchell, R. P. Lin, S. Frey, L. L. Hood, M. H. Acuna and A. B. Binder, J. Geophys. Res. 106 (2001) 27841. 22. D. R. Criswll, Moon 7 (1973) 202. 23. R. P. Lin, D. L. Mitchell, D. W. Curtis, K. A. Anderson, C. W. Carlson, J. McFadden, M. H. Acuna, L. L. Hood and A. Binder, Science 281 (1998) 1480. 24. D. J. McComas and J. E. Nordholt, Rev. Sci. Instrum. 61 (1990) 3095. 25. S. Yokota, Y. Saito, K. Asamura and T. Mukai, Rev. Sci. Instrum. 76 (2005) 014501. 26. S. Yokota and Y. Saito, Earth Planets Space 57 (2005) 281.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


A JOVIAN SMALL ORBITER FOR MAGNETOSPHERIC AND AURORAL STUDIES WITH THE SOLAR-SAIL PROJECT Y. KASABA∗,†,§ , T. TAKASHIMA† , H. MISAWA‡ and JOVIAN SMALL ORBITER SUB-WORKING GROUP [WITH J. KAWAGUCHI† AND SOLAR-SAIL WORKING GROUP] † Institute of Space and Astronautical Science (ISAS) Japan Aerospace Exploration Agency (JAXA) Sagamihara, Kanagawa 229-8510, Japan ‡ Planetary

Plasma and Atmospheric Research Center Tohoku University, Sendai, Miyagi 980-8578, Japan ∗ [email protected]

The Solar-Sail Project has been investigated by JAXA as an engineering mission with a small orbiter into the Jovian orbit. This paper summarizes the basic design of this project and possible Jovian system studies by this opportunity. The large-scale Jovian mission has been discussed as a long future plan since the 1970s, when the investigation of the future planetary exploration program started in Japan. The largest planet and its complex planetary system would be studied by several main objectives: (1) The structure of a gas planet: the internal and atmospheric structures of a gas planet which could not be a star. (2) The Jovian-type magnetosphere: the structure and processes of the largest and strongest magnetosphere in the solar system. (3) The structure, composition, and evolution of Jupiter and its satellite system. The small Jovian orbiter accompanied with the Solar-Sail Project will try to establish the technical feasibility of such future outer planet missions in Japan. The main objective is the second target, the Jovian magnetospheric and auroral studies with its limited payload resources.

1. Solar-Sail Project 1.1. Objectives The Solar-Sail project (Fig. 1) is another engineering mission, the next plan after the Hayabusa spacecraft to the asteroid Itokawa. Its main objective is to establish the method to explore the outer solar system by the § Corresponding

author. 45

46

Fig. 1. Press).

Y. Kasaba et al.

The Solar-Sail project: an image at the flyby of Jupiter (Courtesy: Newton

developments of (1) the advanced combined propulsion systems by solar-sail and electrical propulsion technologies, (2) the spacecraft bus technologies powered by solar cell in the outer solar system, and (3) the complex system with the mother spacecraft, the daughter spacecraft, and the entry probe (option) which are separated in deep space. This mission will visit main-belt asteroids, Jupiter, and Trojan asteroids. The spacecraft will consist of three modules: the mother spacecraft, the daughter spacecraft, and an entry probe (Fig. 2). The mother spacecraft (300–500 kg) will make the multiple flybys at main-belt asteroids and Jupiter, and finally reach to Trojan asteroids. The daughter spacecraft (∼ 100 kg) will be separated just before the arrival to Jupiter and will be the first Japanese Jovian orbiter. The Jovian entry probe (∼ 30 kg: the

Fig. 2. Three units for the Solar-Sail project: the mother spacecraft, the daughter spacecraft as the small Jovian orbiter, and the Jovian entry probe (option).

Jovian Small Orbiter for Magnetospheric and Auroral Studies

47

option unit at this moment) will be separated from the daughter spacecraft and will enter into the Jovian atmosphere. For those units, the scientists from multiple communities have proposed the observational plans in order to take this important opportunity to travel to the outer solar system and stay there for an extended period in time. The latest Solar-Sail project is, therefore, not only an engineering mission but also the first multi-scientific platform in deep space, which is little similar to the Space Flyer Unit concept, a multi-scientific platform in the low Earth orbit launched in 1995 by the NASDA and the ISAS. The science goals of the mother spacecraft during cruise and the flybys will be (1) the cosmic infrared background observation in low background environment, (2) the gamma-ray burst detection sites located far from the Earth, (3) the dust measurement with a largest detector formed by the sail sheet, and (4) the flyby sciences of main-belt and Trojan asteroids. And, the mother spacecraft, and the daughter spacecraft, and the entry probe (option) will observe the Jovian system by fly-by, in orbit, and the entry to the atmosphere. 1.2. Current development status and future base plan This project started in the late 1990s. Several engineering tests for the establishment of the technical baseline of this mission have succeeded: 2000: The formal base study started. The main objectives were “the observation of the solar pole region associated with the Venus flyby” or “the observation of Trojan group with the Jupiter flyby.”

Fig. 3. First unfurling test of the solar-sail in the world with S-310 sounding rocket in 2005 (http://www.isas.jaxa.jp/e/snews/2004/0809.shtml).

48

Y. Kasaba et al.

Fig. 4.

The original schedule plan for the Solar-Sail project.

2004: The mission was selected as “the high-priority mission.” The sail test units aboard the S-310 sounding rocket successfully unfurl two-type sails (Fig. 3).1 2005: The test with the high latitude balloon. 2006: The test with a Piggy-bag satellite launched by the M-V rocket.

Jovian Small Orbiter for Magnetospheric and Auroral Studies

Fig. 5.

49

The orbits of the Jovian small orbiter after the Jupiter orbit insertion.

The original baseline schedule was shown in Fig. 4. Because of the launch delay, the overall schedule is not fixed yet: 2012: Launch by the M-V rocket: The Earth Swing-by (×2) and the Mainbelt asteroid flyby (×1–2) are planned before the arrival at Jupiter. 2017: Arrival at Jupiter: The mother spacecraft makes swing-by. The daughter spacecraft is separated from the mother spacecraft and enters into the Jovian polar orbit (Fig. 5). The entry probe (option) will be separated from the daughter spacecraft and enters the Jovian atmosphere. 2022: The mother spacecraft makes the flyby of some Trojan asteroids. 2. The Small Jovian Orbiter 2.1. Objectives The daughter spacecraft will be the first Jovian orbiter of Japan.2 The main objective of this orbiter is the technology establishment of future outer planetary orbiters, beyond the heritage of the past planetary projects of ISAS and JAXA, the Nozomi mission to Mars (1998–2003), the Planet-C mission to Venus (Launch: 2010), and the BepiColombo Mercury

50

Y. Kasaba et al.

Magnetospheric Orbiter (Launch: 2013). It will be “the pathfinder” for the future full-scale Jovian mission. The planned small Jovian orbiter will prove the technical feasibility for (1) the power supply with solar cell formed on the sail and the thermal design under weak sunlight condition (∼ 1/30 of the Earth-orbiting spacecraft) and (2) the sustainability under the active radiation environment near Jupiter. The Jovian mission has been discussed as a long future plan since 1970s, when the planetary science studies were started in Japan. The largest planet in the solar system would be solved by several main objectives: 1. The structure of a gas planet: The internal and atmospheric structures of the giant gas planet which could not become a star. It follows the objectives of the Venus and Mercury studies based on the Planet-C and the BepiColombo missions. 2. The Jovian-type magnetosphere: The structure and processes of the largest and strongest magnetosphere in the solar system, driven by its own rotational energy with the strong contribution from its satellites. It follows the terrestrial magnetospheric studies based on the Akebono (1989–), Geotail (1992–), Reimei (2005–), the ERG and SCOPE missions planned as future projects, and the Herman magnetospheric studies by the BepiColombo mission. 3. The structure, composition, and evolution of the satellites: the study of the Jovian system and its evolution. It follows the objectives of the SELENE (Lunar orbiter), Planet-C, and BepiColombo missions. The astrobiology topics might be included as an optional target. The main objective of the small orbiter accompanied with the solarsail project is the second target with its limited payload resources, i.e., the quick-look observations of the Jovian magnetosphere and auroral regions, in order to study the Jovian-type magnetosphere–ionosphere coupling and the planet and satellite coupling processes. Those objectives will be shared with and supported by the collaborative observations with the mother spacecraft, the entry probe, and the Earth-based and Earth-orbiting telescope facilities. The simultaneous observation with the Juno mission planned by United States might be valuable, if it is realized. 2.2. The base plan The small Jovian orbiter will be separated from the mother spacecraft before its flyby at Jupiter, and will enter to the polar orbit around Jupiter

Jovian Small Orbiter for Magnetospheric and Auroral Studies

51

by itself. This orbit is selected for the studies of the polar regions of both hemispheres (Fig. 5), the avoidance of the passage in the Jovian radiation belt, and the targeting of and the telecommunication with the entry probe (option). By its long orbital period (∼ 230 days) and expected nominal lifetime (1 year) which is limited by the fuel resource, we consider that the in-situ observations of the polar region will be possible at least three times. The basic configuration of this orbiter is summarized in Table 1. The scientific payload is relatively limited, as shown in Table 1. In the present assumption the expected allocated resources are less than 4 kg and less than 5 W in average. Under these restrictions, the main scientific objective is confined to the quick-look observation of the most active objects in the Solar system, by the crossing of the Jovian polar regions. It will enable us to study (1) the magnetosphere–ionosphere coupling in the Jovian-type magnetosphere driven with planetary rotation energy, and (2) the interaction between the planet and satellites via plasma process which is also expected between a proto-star and proto-planets (hot Jupiter). Since this orbiter will stay in the magnetosheath and the solar wind for a long time, it will also be expected to act as a solar wind monitor for the cooperated studies with ground-based and Earth-orbiting telescopes for the Jovian auroral and magnetospheric studies. The payload is mainly assumed under the heritage of the instruments developed for the Planet-C mission and the BepiColombo Mercury Magnetospheric Orbiter. The expected model instruments are as follows: Instruments for the Jovian plasmaspheric and auroral studies with following

Table 1.

Basic parameters of the small Jovian orbiter.

General Spacecraft

Power consumption Life Power system

A spin-type spacecraft: diameter: 70 cm, mass: 98 kg 90 W (no communication), 190 W (with communication) > 1 Earth year Solar cell panel formed on the sail film

Orbit

Distance Period

1.2–1.3RJ ∼ 300RJ (< 20RJ : about 2 days) 230 days = 1.5 round/year

Payload

Resource Thermal Telemetry

Mass: < 4 kg, power: < 5 W Nominal: −30◦ C 16 bps/1 kbps (X/Ka, 60 cm Φ, 20 W HGA) Operation: 1 h/day (40 bps/ave = 0.4 MB/day) or 1 h/week (6 bps/ave = 0.06 MB/day)

52

Y. Kasaba et al.

priorities: (a) Remote sensing: Remote sensing capabilities coupled with the mother spacecraft, the entry probe, and other orbiters (ex. Juno), and groundbased and Earth-orbiting telescopes: The priority-1 is the Jovian aurora and lightening observations which will collaborate with in-situ measurements. Priority-2 is the decametric and hectometric radio observations which will collaborate with radio, infrared, and visible observations from the Earth. (b) In-situ observations: The observations in the particle acceleration regions and the monitoring of the magnetospheric and solar wind plasmas. Priority-2 is the magnetospheric and solar wind observations, i.e., activities inside the auroral acceleration region, the large-scale current system, etc., which will collaborate with radio, infrared, and visible observations from the Earth. The current model payload plan is summarized in Table 2: (1) remote sensing by an auroral monitor, a radio monitor, (2) in-situ measurements by a magnetometer, high-energy and low-energy particle detectors, and (3) an integrated control system unified with the spacecraft bus system. For the entry probe (option), which feasibility is not established yet, some payloads with the mass less than 1 kg are considered. At this moment, we consider a magnetometer, a high energy particle monitor, a narrow-band radiometer (lightening detection), and a simple infrared radiometer (H2 O detection), etc. Heavy instruments like a mass spectrometer are hard to be Table 2.

Model payloads: the planned total mass is less than 4 kg.

Payload candidates Remote

In-situ

Common

Line image (Priority-1) Radio (Priority-2) Camera (option) Magnetic field (Priority-2) High-E particles (Priority-2) Low-E ions (Priority-2) Digital/Power

Mass (kg)

Objectives

1.5

Aurora/Lightening

0.5

Jovian radio waves

0.5

Navigation

0.5

B-field and current structure (no boom) Particles in polar and equatorial region Solar wind monitor and outflow ions

0.7 0.7 –

Unified to the system

Jovian Small Orbiter for Magnetospheric and Auroral Studies

53

included. The entry probe will be dropped into the middle to high latitude regions, not covered by the Galileo probe. Some studies and developments for this mission are investigated with the engineering team in order to achieve good scientific results: (1) The design of the sail film: the current canceling for the reduction of magnetic interference and the installation of wire antennas. (2) The development of sensitive and light-weight sensors and optics. (3) The unified data processing and power systems with the bus system. (4) The system studies for the small orbiter itself, including the communication, power, structure and orbit/attitude control systems. The establishment for the collaboration with ground-based and space telescope facilities and US Juno mission will be in near future, after the start of Phase-B. Japanese space science communities will expect the establishment of new-era through this project. Acknowledgments The authors wish to express their sincere gratitude to all members participating and contributing to the studies of the Jovian Small Orbiter Sub-Working Group and the Solar-Sail Working group: K. Kuramoto, K. Sugiyama (Hokkaido Univ.), A. Kumamoto, A. Morioka, S. Okano, K. Sakanoi, Y. Takahashi, F. Tsuchiya, A. Yamazaki (Tohoku Univ.), M. Hoshino, I. Yoshikawa (Univ. Tokyo), A. Nishida (G-Univ. Adv. St.), Y. Miyoshi, T. Ogino, K. Seki (Nagoya Univ.), K. Hashimoto, S. Machida (Kyoto Univ.), T. Murata (Ehime Univ.), M. Yamamoto (Kochi Inst. Tech.), K. Imai (Kochi NCT), T. Hada, K. Nakajima, K. Yumoto (Kyushu Univ.), S. Takeuchi (Fukuoka Univ.), T. Sato (Kumamoto Univ.), H. Nozawa (Kagoshima NCT), M. Taguchi (NIPR), Y. Kasai, T. Kondo, F. Nakagawa (NiCT), J. Ishida (NTSpace), T. Abe, K. Asamura, M. Fujimoto, Y. Futaana, H. Hasegawa, H. Hayakawa, T. Imamura, K. Maezawa, A. Matsuoka, T. Mukai, M. Nakamura, M. N. Nishino, Y. Saito, I. Shinohara, J. Terazono (JAXA). References 1. S. Takeuchi, K. Minesugi, J. Onoda and J. Kawaguchi, Deployment experiment of solar sail using sounding rocket, Proc. 54th IAC/IAF, (2003) pp. 1557–1562. 2. Y. Kasaba, T. Takashima, H. Misawa, Jovian Small Orbiter sub-WG with J. Kawaguchi and Solar-Sail WG, Jovian Small Orbiter for Magnetospheric and Auroral Studies. “Solar Sail Project”, Proc. IAA International Conference on Low-Cost Planetary Missions (ICLCPM), (2005) pp. 08-B1.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


DESCRIPTION OF A NEW 400 MHz BANDWIDTH CHIRP TRANSFORM SPECTROMETER LUCAS PAGANINI∗ and PAUL HARTOGH Max Planck Institute for Solar System Research PO Box 37191, Katlenburg-Lindau, Germany ∗ [email protected]

A new chirp transform spectrometer (CTS) with a bandwidth of 400 MHz and a spectral resolution of 100 kHz has been developed. The CTS is deviced using a digital chirp generator and a preprocessing unit based on a Complementary Metal Oxide Semiconductor (CMOS) and an Application-Specific Integrated Circuit (ASIC). A build in PC 104 computer handles the process control and the external communication via Ethernet and a Transistor-Transistor Logic (TTL) interface. The CTS has been applied to atmospheric science, i.e., a 25-K noise temperature, 22-GHz water vapor, and a 142-GHz ozone system. Astronomical observations have been performed using the Heinrich Hertz submillimeter telescope. In this paper, we describe the function of the CTS and provide information about its functional performance.

1. Introduction Heterodyne spectroscopy is a technique providing practically unlimited spectral resolution. A high frequency signal for instance from the submillimeter range is down-converted by a local oscillator signal to a lower frequency band in which electronic spectrum analysis techniques can be applied. In atmospheric spectroscopy or radioastronomy, the downconverted signals are in general of stochastic nature. As a consequence, the derived power spectra are stochastic as well and require averaging. Thus, a high efficiency or high duty cycle of the spectral analysis method is required. Spectrometers with nearly 100% duty cycle are called realtime spectrometers, since the a priori data rate of the calculated spectrum is the same as the incoming time domain signal. Reduction of the data rate is done by averaging the power spectra. Real-time or nearly realtime spectrometers, such as filterbanks, acousto-optical spectrometers, and autocorrelators, are widely spread especially in radioastronomy. Recently 55

56

L. Paganini and P. Hartogh

high bandwidth Fast Fourier Transform (FFT) spectrometers have been developed. The chirp transform spectrometer (CTS) has a successful history of more than two decades in ground-based, airborne, and space missions. The first millimeter wave heterodyne spectroscopy with CTS has been performed in the middle of the 1980s by detecting the 142 GHz rotational transition of ozone in the Earth middle atmosphere.1–3 Since the early 1990s, continuous ground-based measurements of water vapor (22/183 GHz) and ozone (142 GHz) have been performed. Furthermore, upon the integration of the first CTS spectrometer into the Heinrich Hertz submillimeter telescope (HHSMT) on the Mt. Graham in Arizona, USA, it has been applied to address a wide range of topics of modern astrophysics.4,5 These topics go from questions about comets, planetary atmospheres, and the interstellar medium in the galaxy, to investigations related to the early Universe. Recently, CTS provided a high-quality spectra of comet 2002 T7 (LINEAR) and the Earth from the Microwave Instrument for the Rosetta Orbiter (MIRO), the first deep space mission carrying a submillimeter heterodyne spectrometer.6,7 The CTS has been proven to be a very reliable and accurate spectrometer (see for instance Refs. 8 and 9). The newly developed 400-MHz CTS combines the advantage of broader bandwidth analysis with keeping the characteristics of previous CTSs. The new techniques applied for creating the chirp signal, which also involves similar digital techniques as used in the SOFIA-GREAT (Stratospheric Observatory for Infrared Astronomy — German REceiver for Astronomy at THz frequencies) CTS,10 yield a perfect matching to the dispersive properties of the compressor unit of the CTS and improvement of the signal-to-noise ratio (SNR) up to 60 dB. The CTS is based on the Chirp transform,11,12 an algorithm derived from the Fourier transform and implemented by linear frequency-modulated waveforms and their matched filters. In the CTS, the input signal is first modulated by a chirp. Thereby, for instance a fixed frequency signal becomes linearly modulated. The latter signal is fed into a dispersive filter (called compressor) with a delay time depending on the frequency and equal dispersive characteristics. The filter output for the fixed frequency now looks like a single peak at a specific time giving the spectrum as a function of time. The dispersive elements are (surface acoustic wave) (SAW) filters, which are characterized by the propagation of acoustic energy along the surface of a piezoelectrical crystal base, whose displacement amplitudes undergo exponential decay beneath this surface. The wave pattern of

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer

57

the surface acoustic wave can be modified by microstructures on the substrate surface, translating in amplitude and/or phase changes of the input signal.This publication contains a section with a description of the system design, leading to the improvements achieved in the new CTS. Thereafter, the instrument’s characterization can be found, i.e., analysis of the digitally created chirp signal, spectral resolution, dynamic range, nonlinearity of the frequency scale, and the overall stability. Allan-variance tests were used to retrieve the radiometric performance of the instrument by quantifying the amplitude (or gain) stability, while laboratory sources were used to quantify the instrument’s white noise dynamic range and spectral properties. Furthermore, we present test spectra of the 142-GHz ozone line measured at our institute and of spectral lines of comet 73P/SchwassmannWachmann 3 gained at the HHSMT in May 2006.

2. Development and Design The CTS involves two parts: the analog and the digital. The analog part is integrated by the RF stage, which includes mixers, amplifiers, filters, doublers, splitters, and SAW filters. On the other hand, the digital part involves the chirp generation board, the data processing and synchronization board [Application-specific integrated circuit (ASIC) board], an ISA–ASIC interface and an embedded computer. As a backend system, its duty is to acquire an incoming signal in real time with 100% efficiency and then provide the spectral information supplied by heterodyne frontend systems. The chirp is digitally created using direct digital synthesizers (DDS), which is driven by a 1-GHz clock frequency. The importance of such technique is the possibility to fix every aspect of the chirp signal, in order to achieve the perfect dispersive matching required as above mentioned, providing high dynamic range, since the outcoming chirp waveform has a large SNR. It is important to highlight this because it can be corrected by means of digital properties changes. The created signal in the chirp board has a 400-MHz bandwidth centered at 250 MHz with a dispersion time of 20 µs and then it is frequency up-converted using RF mixers. In the SOFIA CTS, the signal is quadrature-modulated and upconverted before the RF stage where it is frequency-tripled. The new implementation allows a broader bandwidth which only needs to be doubled in the RF stage (Fig. 1), achieving not only a bandwidth up to 800 MHz, but also improving the SNR up to 60 dB.

58

L. Paganini and P. Hartogh

Fig. 1. After the chirp signal is digitally created and up-converted with an RF mixer, it follows a series of filtering, amplifying, and doubling processes to be ready for a modulation with the incoming atmospheric signal and consequently feeding into the compressor (SAW filter). By mixing the signal with a complex source, after the convolution process, it is down-converted and separated in real and imaginary parts. This complex signal is down-converted. Thereafter, it is possible to digitalize it with four pairs of ADCs and continue with the digital squaring, preprocessing, and integration in the ASIC board. At the end, it goes to an embedded computer through an ISA–ASIC interface.

After the doubling process, the mixing stage follows where the incoming atmospheric signal is frequency modulated by the chirp. As a result, a linear changing frequency of 800-MHz bandwidth is created. This signal has a center frequency in the range of 600–1400 MHz, depending on the

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer

59

Fig. 2. Compressed pulse observed at the end of the analog part and before the digital acquisition by an oscilloscope for a 2.1-GHz pulse input.

frequency of the input atmospheric signal, which is in the 1.9–2.3 GHz range. The newly modulated signal is fed into the SAW filters which have a 400-MHz bandwidth (800–1200 MHz range). The output of the filter in the time domain represents the analog spectrum of the input signal, mapped on a 10-µs time interval (Fig. 2). Due to intrinsic properties of the chirp transform, the duty cycle of an expander–compressor scheme is only 50%. This means that the setup takes 20 µs to perform a transform. However, it produces only half the time of useful spectral information. For that purpose, two branches are combined through a commutator with a switching period of 10 µs with the idea of 100% efficiency. Later on, the signal is down-converted to the base-band and the real and imaginary components are obtained by mixing the signal with a complex source. In the ASIC board, these two components are digitally acquired with a set of eight 100 MSPS analog-to-digital converters (ADC). The signal is digitally analyzed by a preprocessor integrated in an ASIC chip with low power consumptions. This preprocessor computes the power from the complex spectrum, where the signal is numerically squared (real and imaginary parts), added, and finally mapped into a 4096-channel memory where it is integrated. Furthermore, it also handles the synchronization signals needed to control the timing of the two branches. This also involves a signal which defines the generation start of the chirp signal at a specific

60

L. Paganini and P. Hartogh

moment. Finally, through an ISA–ASIC interface an embedded computer provides the instrument control and external communication.

3. Test Measurements 3.1. Chirp signal Especially important is the understanding of the behavior of the chirp signal. There are a number of quality requirements for the chirp signal like a small passband ripple, a good SNR, small phase deviation, group delay, and the corresponding chirp rate (µ). This is analyzed in Fig. 3. Most of these methods, based on Fourier transform, analyzed this signal in a stationary regime. This is explained by the factthat the transformtakes +∞ as limit of its integral infinite boundaries F(ω) = −∞ f(t) · e−jωt · dt , i.e., they analyze the spectra without the possibility to recognize which event occurs at a specific time. Because the chirp signal is non-stationary, i.e., change with time, a novel STFT analysis tool was applied. Among others, an STFT is able to calculate the Fourier transform in small time intervals, thus giving the chance to discriminate events vs. defined time slots. This is used with the aim of observing the variation of frequency components with time, e.g., to determine whether harmonics observed in a spectrum analyzer are influencing the 10-µs transformation interval or not. If that is the case, noise is added. This is especially important during the design process of the instrument, when different RF devices are evaluated. Furthermore, it is possible to estimate if the influence of undesired harmonics can be neglected or should be suppressed, e.g., by adding filters. As it can be seen in Fig. 4, it is possible to observe deterioration in the signal after mixing, amplifying, doubling, and filtering processes producing undesirable spurious which affect the overall behavior of the instrument. This was improved by selecting a better RF mixer before the doubling process and the addition of filters.

3.2. Power linearity and dynamic range Three effects constrain the instrument’s dynamic range: the high insertion loss of the SAW devices (more than 40 dB), the noise and interferences introduced during the RF signal processing, and the compression point of the different passive (mixers) and active components. The maximum signal

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer

Fig. 3. Chirp signal study tools by acquiring and digitalizing it in time domain with DSO. Upper left: Calculated magnitude. Upper center and right: Wrapped phase and fitting in the desired frequency range. Down left: phase deviation and root mean square error. Down center: Calculation of chirp rate in MHz/µs. Down right: new analysis approach using a Short Term Fourier Transform (STFT) computation.

61

62

L. Paganini and P. Hartogh

Fig. 4. Left: chirp signal right after the digital generation in the chirp board. Right: signal fed into the SAW filter. It can be seen that after the processes of amplification, doubling, mixing, and filtering spurious frequencies appear producing deterioration in the signal.

in the RF part is the one which drives the main mixer (+10 dBm), while the lowest is the output of the SAW device (−50 dBm). The inclusion of digital techniques into the CTS design and the 60-dB SNR allows a wide range adaptability, which assure a proper dynamic range setup. During this evaluation, a noise source with ∼1 dB flatness and +10 dBm power level was used as input in the frequency range of 1.9–2.3 GHz; together with high accuracy and low repeatability RF step attenuators with ±0.15 dB accuracy. The linear response of the instrument was analyzed in a 60-dB input range by stepping the input power in 1 dB steps. The results in Fig. 5 show a dynamic range with a maximum deviation from linearity of ±1 dB equal to 35 dB and ±0.1 equal to 20 dB; and an optimum input power level of −35 dBm. This analysis allows to predict the overall response of the instrument by calculating the mean of the whole range of channels. But, for a more detailed evaluation of the instrument response, i.e., each single channel, a novel tool plots a 3D graph, which allows to observe the given response of the spectrometer by means of channel number, power input, and output counts/cycle. This is specially important for establishing not only linear dynamic range, but also a wide range of unexpected behaviors not possible to identify in previous analysis, e.g., mismatch of dispersive properties between SAW filters and the fed signal, possible regions which goes into saturation and compression faster than others, etc.

3.3. Stability In order to quantify the stability of microwave heterodyne spectrometers, Allan variance measurements are usually used.13 The radiometer formula

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer

63

Fig. 5. Power linearity study. In the upper part a novel 3D method to observe the output power plotted against input power and number of channel. This method allows to see possible deteriorations in the system. Down left: output power obtained by inserting a variable power noise source at the input. Down right: dynamic range calculation, considering a ±1 and a ±0.1 dB deviation from linearity.

can be applied while the instrument is stable during the observation time between two calibrations. As any additional noise above the radiometric level is unfavorable, one has to find the optimum integration time, where the impact of drift contributions is nearly negligible. In other words, the radiometer equation is valid within the white noise part, i.e., before the Allan variance minimum. This minimum describes the turnover point where the radiometric noise with a slope of −1 in the logarithmic plot becomes dominated by the additional and undesired drift noise. After warming up 43,000 spectra, with 1 s integration time each and an ultra stable noise source input at a constant power level, were acquired. The analysis of the data gives a minimum Allan variance time of 173 s,

64

L. Paganini and P. Hartogh

Fig. 6. Allan variance calculation. Left: output count per cycle for the channel 2045 using a noise source during 43,000 spectra integrated in 1 s each. Right: the Allan variance of a frequency channel versus integration time. The Allan time is defined by the minimum k (173 s).

represented in the plot by the value k = 173 (Fig. 6). The spectroscopic Allan variance, which performs a similar study in two independent channels, showed no independent drift behavior between them. The frequency stability of the spectrometer is linearly related to the temperature stability of the compressor filters. Thus a good thermal stabilization of these devices is essential. The thermal stabilization we used results in a frequency stability of 550 Hz/◦ C.

3.4. Spectral resolution The response of the spectrometer to a sine wave is a sinc2 function with the first zero crossing in the frequency domain at 1/TC , defining the spectral resolution of the spectrometer. The dispersion time of 10 µs gives us a 100kHz spectral resolution. Another approach is to obtain the full width at half maximum (FWHM) of each channel (Fig. 7). This was obtained by calculating the FWHM of the sample curved for each single channel in the whole range, using a step size of a 10th nominal resolution. Moreover, deviations in the spectral scale linearity were further retrieved. This property describes the relationship between the input frequency and the corresponding expected frequency for a specific channel index, obtaining values smaller than 6% of the spectral FWHM for the complete operational bandwidth of the instrument.

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer

65

Fig. 7. Spectral resolution and frequency linearity versus number of channel. Upper: measurement of the spectral resolution (FWHM). Down: deviation of the obtained calculation from the expected nominal value (∼ 0.06 of a channel).

Table 1. Specifications for the 400-MHz bandwidth chirp transform spectrometer. Center frequency Spectral resolution (noise equivalent) 1/TC Spectral resolution (FWHM) Channel spacing Bandwidth (−3 dB) Optimal RF-Power input Noise dynamic range (±1 dB) Noise dynamic range (0.1 dB) Frequency linearity Absolute allan-variance time Channels Power consumption

2.1 GHz 100 kHz 121.2 kHz 97.6 kHz 400 MHz −35 dBm 35 dB 20 dB ±4 kHz 173 s 4096 < 30 W

4. Observations and Results The new 400-MHz bandwidth CTS was tested under real observing conditions in order to assess the expected performance. First light was observed on December 2005 by measuring the O3 line using a 142-GHz ozone system as test facility in Katlenburg-Lindau (Fig. 8). In order to test its behavior by observing astronomical objects, the instrument was installed at the HHSMT on the observing run of the 73P/Schwassmann-Wachmann 3 comet in May 2006, when it had its closest approach. The HHSMT is located at 3200 m altitude on Mt. Graham. It has a 10m-diameter main reflector, the absolute pointing accuracy is about 2′′ , with a tracking accuracy of better that 1′′ .14 The receivers used were the SSB 1.3 mmJT and the MPIfR SIS-345. It is equipped with several backends: a 218-MHz bandwidth CTS, two AOSs, and filterbanks.

66

L. Paganini and P. Hartogh

Fig. 8. First light. Ozone spectra obtained at 142 GHz using the test facilities in Katenburg-Lindau, Germany.

Fig. 9. HCN(3-2) emission line of the 73P/Schwassmann-Wachmann 3 comet during its closest approach on May 17, 2006 at HHSMT on Mt. Graham, Arizona.

The observed components were mainly fragments B and C focusing on the HCN(3-2) and HCN(4-3), H2 CO, CO and CS lines, under good weather conditions (Fig. 9). The scope of this test was not only to show the spectrometer’s high resolution which is significantly important for the study of narrow features (e.g., in cometary emission lines), but also to observe broader lines, and thus demonstrate the different capabilities that the CTS can reach. Some observations were performed in Mars and some

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer

67

Fig. 10. 12 CO(2-1) absorption line of Mars observed at HHSMT during the 73P/S-W 3 observing.

Fig. 11.

Observation of the N7538IRS1 star. CS(5-4) line.

flux standards. Figure 10 shows a sample of the 12 CO(2-1) line in Mars and Fig. 11 the star N7538IRS1, CS(5-4) line.

5. Conclusions and Outlook We have developed a new 400-MHz bandwidth and 100-kHz spectral resolution CTS. The digital techniques used for the chirp signal generation were improved allowing a SNR up to 60 dB. In this publication, a complete characterization of the spectrometer is presented. Furthermore, test results were obtained in atmospheric science and astronomical observation by

68

L. Paganini and P. Hartogh

integration of the spectrometer in a 142-GHz ozone system and by using the HHSMT . Considering the demand for broad bandwidth for various wavelength spectroscopy, future development will concentrate on the following areas: (i) broader bandwidth SAW filters development; (ii) new analysis techniques to fully and completely analyze different and essential stages in the system with orders of accuracy never achieved before, and thus eliminate possible nonlinearities impossible to measure by nowadays methods; (iii) by optimizing power consumption, size and weight, the present sensitivity and resolution of the spectrometer can considerably widen different aspects studied nowadays in science, allowing its inclusion for future space missions. Finally, the goal is to employ careful design choices, test measurements, and advances in technology to ensure that a new CTSs achieve a comparable or, which is always the main aim, better performance of previous CTSs.

Acknowledgments We wish to thank Dr. C. Jarchow for his continuous cooperation and advices and the SMTO staff for their friendly assistance.

References 1. P. Hartogh, Messung der 142 GHz Emissionslinie des atmosph¨ arischen Ozons, PhD thesis (University of G¨ ottingen, 1989). 2. P. Hartogh and G. Hartmann, Meas. Sci. Technol. 1 (1990) 592–595. 3. P. Hartogh and C. Jarchow, Proc. SPIE 2586 (1995) 188–213. 4. G. Villanueva, PhD thesis (Albert-Ludwigs-Universitat zu Freiburg, 2004). 5. M. D. Hosstadter et al., Earth, Moon Planets 78 (1999) 53–61. 6. G. Beaudin et al., Proceedings of the 2nd ESA Workshop on Millimetre Wave Technology and Applications, pp. 43–48. ESA WPP-149 (ESA Publ. Div., ESPOO, Finland, 1998). 7. S. Gulkis et al., Planet. Space Rev., doi: 10.1007/s11214-006-9032-y (2006), available only online, pending paper publication. 8. C. Seele and P. Hartogh, Geophys. Res. Lett. 26(11) (1999) 1517. 9. P. Hartogh, et al., J. Geophys. Res. 109 (2004), D18303, doi: 10.1029/ 2004JD004576. 10. G. Villanueva and P. Hartogh, Exp. Astron. 18 (2004) 77–91. 11. S. Darlington, Bell Syst. Tech. J. 43 (1964) 339. 12. J. R. Klauder et al., Bell Syst. Tech. J. 39 (1960) 745. 13. R. Schieder and C. Kramer, A&A 373 (2001) 746. 14. J. W. M. Baars et al., PASP 111 (1999) 627–646.

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


FORMATION OF ALUMINA NANOPARTICLES IN PLASMA MAMI KURUMADA and CHIHIRO KAITO∗ Department of Physics, Ritsumeikan University, 1-1-1 Nojihigashi Kusatsu-shi, Shiga 525-8577, Japan ∗[email protected]

γ-Alumina nanoparticles with a polyhedral shape based on an octahedral shape were produced with a plasma field of Ar–O2 gas mixture (Ar: 7 Torr, O2 : 3 Torr). The obtained infrared spectrum showed a characteristic strong absorption peak at 7.2 µm, which had never been previously observed in the alumina nanoparticles produced without a plasma field. The plasma field mainly affected the surfaces of the alumina nanoparticles and changed their morphology and gas adsorbability. The 7.2 µm, absorption peak derived from surface hydroxyl was considered to be due to the existence of activated Al in the alumina nanoparticles produced with a plasma field.

1. Introduction Alumina has many metastable polymorphs denoted by ρ-, γ-, δ-, θ-, χ-, o-, κ-, λ- and η-phases in addition to the α-phase, which is the most thermodynamically stable phase well known as corundum. The γ-, δ-, η-, λ- and θ-phases possess structures based on spinel, with the face-centered cubic (fcc) packing of oxygen anions and aluminum cations distributed at their octahedral and tetrahedral sites. γ-Alumina is typically produced from boehmite and is transformed into the δ-, θ- and α-phases by heating, [AlO(OH)] (300–500◦C) → γ (700–800◦C) → δ (900–1000◦C) → θ (1000–1100◦C) → α-alumina.1 These metastable polymorphs with spinelbased structures differ in the distribution pattern of Al cations in an oxygen sublattice. The occupation percentages of the tetrahedral (10%) and octahedral (47%) sites of γ-alumina were estimated by molecular dynamic (MD) simulation.2 The migration of Al cations from tetrahedral to octahedral sites occurs during transformation from γ- to δ-, θ- and finally α-alumina.2 The distribution of Al cations in the oxygen sublattice

69

70

M. Kurumada and C. Kaito

of γ-alumina is random and becomes periodic during transformation. θAlumina has a more ordered distribution than δ- or γ-alumina.1 A γ-alumina grain is a candidate material for 13 µm absorption observed around certain post-AGB stars.3 In our previous study, we produced δ-alumina nanoparticles by evaporating the aluminum in an Ar–O2 gas mixture.4 The aluminum clusters evaporated immediately reacted with oxygen and alumina nanoparticles were formed. We showed that δ-alumina, as well as γ-alumina, has a strong absorption at 13 µm. However, the characteristic infrared absorption peaks of δ-alumina were observed between 10.4 and 19.0 µm. The difference in infrared (IR) spectrum between γ-and δ-alumina is due to the number of Al cations in the tetrahedral sites. On the other hand, we began to realize that a plasma field plays an important role in the formation of grains. A quenched carbonaceous composite, which is considered as a candidate material for 220 nm absorption, was produced with a CH4 plasma.5 A nitride grain, such as Si3 N4 , was also produced with a plasma field.6 The plasma field also deeply affects the formation of crystalline silicate grains. The study based on the ISO spectra of a large number of stars indicates that enstatite (MgSiO3 ) is more abundant than forsterite (Mg2 SiO4 ) by a factor of 3 to 4.7 Although crystalline forsterite grains could be easily produced in our smoke experiments,8,9 the enstatite grains had never been observed. However, we have recently succeeded in forming crystalline enstatite grains with a spherical or needle shape by simultaneously evaporating SiO and Mg vapor in a plasma field.10 We also have observed the formation of fayalite (Fe2 SiO4 ) grains from an amorphous that produced with a plasma field.11 The crystallization temperature of Fe2 SiO4 silicates was lower than those previously reported owing to plasma effects such as the ionization or doping of atoms. The crystallization temperature of iron silicates being higher than that of Mg silicates is considered to be one of the reasons for the lack of crystalline iron silicates in astronomical objects.12 An experiment about the formation of crystalline iron silicates suggested certain upper annealing temperature limits. Thus, the plasma field triggers the formation of crystal grains and it is important to know the effects of the plasma field on grains. In this study, the plasma field was introduced into the formation stage of alumina nanoparticles. In the case of the alumina nanoparticles produced without a plasma, the δ-phase was formed. The stability and morphological

Formation of Alumina Nanoparticles in Plasma

71

change of this metastable phase in a plasma field were considered to be due to the existence of Al radicals.

2. Experimental Procedure A gas mixture consisting of Ar (7 Torr) and O2 (3 Torr) was introduced into the chamber, and a radio frequency (RF) plasma (frequency: 13.56 MHz, output: 300 W) was charged between electrodes. The schematic system used is shown in Fig. 1. Al powder was evaporated from the Ta boat in the plasma field of Ar–O2 gas mixture. The aluminum clusters immediately reacted with oxygen, and then alumina nanoparticles were produced. The observation and spectroscopy of IR spectra were performed for the produced samples. The collected nanoparticles were dispersed in ethyl alcohol and mounted on an amorphous carbon film supported by standard copper electron microscope grids for the analysis of the structures of the nanoparticles using a transmission electron microscope (TEM) operated at 100 kV (Hitachi H-7100R). They were also buried in KBr pellets, and their transmittance was measured with a Fourier transform IR spectrometer

Fig. 1. Schematic image of plasma system. After evacuating to 10−5 Torr, the Ar–O2 gas mixture was introduced into the chamber. Al powder was evaporated from the Ta boat in the RF plasma field of Ar–O2 gas mixture (frequency: 13.56 MHz, output: 300 W).

72

M. Kurumada and C. Kaito

(Horiba Inc., FT210) from 2.5 to 25.0 µm. The wavelength resolution was 2 cm−1 . The beam splitter was a Ge-evaporated KBr substrate, and the detector was deuteriated triglycine sulfate.

3. Results and Discussion 3.1. Phase or morphological change due to plasma Figure 2 shows the TEM micrographs of the alumina nanoparticles produced (a) without and (b) with a plasma field. The phase of the alumina

Fig. 2. TEM micrographs of alumina nanoparticles produced (a) without and (b) with a plasma field. Corresponding ED patterns indicate that γ-alumina is produced with a plasma field, whereas δ-alumina is produced without a plasma field. Black arrows in image (b) show the nanoparticles with a characteristic polyhedral shape.

Formation of Alumina Nanoparticles in Plasma

73

nanoparticles produced with a plasma field is different from that without a plasma field. Corresponding electron diffraction patterns indicate the production of δ-alumina or γ-alumina for the samples produced without or with a plasma field, respectively. The unit cells of tetragonal or orthorhombic δ-alumina have the following relationship with γ-alumina: a = aγ and c = 3aγ (tetragonal) or a = aγ , b = 2aγ and c = 1.5aγ (orthorhombic). Since both δ- and γ-alumina have defective spinel structures, the electron diffraction (ED) pattern of δ-alumina resembles that of γ-alumina. However, δ-alumina has characteristic diffraction rings between (220)γ and (111)γ owing to their large unit cells. Therefore, we conclude that the nanoparticles produced in the plasma field are γ-alumina. This result obtained by TEM observation is consistent with that obtained by IR spectroscopy, as shown in the next chapter. A morphological change was also observed in the nanoparticles produced with a plasma field. In the case of without-plasma-produced alumina nanoparticles, 50-nm-order particles are all spherical (Fig. 2a). However, polyhedral or partially polyhedral nanoparticles were clearly observed in the alumina produced with a plasma field, as indicated by black arrows in Fig. 2b. It is considered that these morphological changes are due to the electric charge disorder on the surface.13 3.2. IR spectral change of alumina nanoparticles produced with a plasma field Figure 3 shows the IR spectra of the (a) γ-alumina, (b) δ-alumina, and (c, d) samples produced in our present experiment. γ-Alumina is a commercial powder of Newmet Koch, and its purity is 99.99%. The spectrum of δ-alumina was obtained from our previous report.4 Spectra (c) and (d) correspond to the nanoparticles produced without and with a plasma field, respectively. Spectrum (c) has many small peaks in the region 10–20 µm and shows almost the same shape as δ-alumina. On the other hand, spectrum (d) corresponds to the spectrum of γ-alumina rather than that of δ-alumina. When the alumina nanoparticles were produced with a plasma field, the γ-phase was formed. The phase of alumina nanoparticles changed owing to the plasma. Both γ- and δ-alumina have been described to have defective spinel structures. The ideal spinel structure AB2 O4 is based on the fcc packing of oxygen anions, with A and B cations occupying the 8a tetrahedral and 16d octahedral Wyckoff positions. The distribution patterns of Al cations

74

M. Kurumada and C. Kaito

Fig. 3. IR spectra of (a) commercial γ-alumina, (b) δ-alumina particles produced in our previous study,4 and (c, d) samples produced in our present experiment correspond to the nanoparticles produced without and with a plasma field, respectively. Arrows P and C mean absorption features by physisorption and chemisorption of H2 O.

are different in both phases. In γ-alumina, Al cations randomly distribute in tetrahedral or octahedral sites. Furthermore, the number of Al cations coordinated in the tetrahedral sites of γ-alumina is larger than that of δ-alumina.2 The migration of Al cations from tetrahedral to octahedral sites occurs during transformation from γ- to δ-, θ- and finally α-alumina. Therefore, the production of γ-alumina indicates that many Al cations are coordinated in the tetrahedral sites of the oxygen sublattice within the plasma field. The ionic radius of Al cations coordinated in the tetrahedral (rt ) or A, respectively. Alvarez et al. calculated octahedral (ro ) sites is 0.39 or 0.54 ˚ the coordination numbers of Al cations as a function of ionic radius (rAl ) by MD simulation.2 Their results indicated that when rAl is less than rt , Al cations tend to occupy tetrahedral sites, whereas when rAl > rt is used for simulation, Al cations tend to occupy octahedral sites. Thus, it is considered that the Al cations in a plasma field have an ionic radius smaller than rt , promoting the coordination of Al cations in the tetrahedral sites. A more marked difference can be observed in spectrum (d). The sharp and strong absorption peak at 7.2 µm, which had never been previously

Formation of Alumina Nanoparticles in Plasma

75

observed in the spectra of any alumina nanoparticles, appeared in the spectrum of the alumina nanoparticles produced with a plasma field. It is considered that this 7.2 µm peak was formed in an O–H bending mode (6.9–8.4 µm) on the surfaces of the alumina nanoparticles from its position. When the nanoparticles were produced with a plasma field exposed to air before measuring the IR spectrum, many H2 O molecules in air adsorbed to their surfaces. There are three reactions involved in water absorption, namely, physisorption, chemisorption, and surface hydroxylation. In the case of physisorption, the IR absorption peaks of H2 O molecules are observed in two regions, namely, 2.5–3.0 and 5.5–7.0 µm, as observed in all the spectra in Fig. 3. As shown in Fig. 3d, we found that the alumina nanoparticles produced with a plasma exhibit a high level of physisorption compared with the other spectra in Fig. 3. The H2 O molecules could be removed from the surfaces of the nanoparticles at 100–200◦C because they were attached to the surfaces by van der Waals’ force. On the other hand, the chemisorption or surface hydroxylation has stronger bonds with the surfaces than physisorption. The H2 O molecules for chemisorption or surface hydroxylation could be removed at about 400◦C or 1000◦C, respectively. Figure 4 shows the spectrum of the alumina nanoparticles after heating at 600◦C (the IR spectrum was measured at room temperature). The 7.2 µm peak disappeared after heating at 600◦ C. That is, the adsorption corresponding to this 7.2 µm peak is an irreversible reaction, whereas physisorption is a reversible reaction corresponding to 2.5–3.0 or 5.5–7.0 µm absorption peaks observed again in the spectrum in Fig. 4. Therefore, it

Fig. 4. Spectrum of the alumina nanoparticles after heating at 600◦ C (IR spectrum was measured at room temperature). The 7.2 µm peak disappeared after heating at 600◦ C. Arrows P mean absorption features by physisorption of H2 O.

76

M. Kurumada and C. Kaito

is considered that the chemisorption of H2 O molecules contributes to the formation of the 7.2 µm peak. Both equilibrium condensation calculation14 and nonequilibrium nucleation accompanying grain growth15 indicate that corundum grains are the first material to condense in the expanding and cooling gas of solar composition in an oxygen-rich atmosphere. As the plasma field exists on our universe, alumina phase with 7.2 µm by chemisorption of H2 O may be observed according to the increase in the resolving power of the instruments accompanying on the development of extra-solar planetary science. The release energy of the chemisorption water by heating at 600◦C may also accelerated the growth into spinel, hibonite, and other compounds.

4. Conclusion Aluminum was evaporated in a plasma field of Ar–O2 gas mixture, and the γ-alumina nanoparticles were formed. The IR spectrum of the alumina nanoparticles produced with a plasma field indicated the production of γ-alumina and a sharp and strong absorption peak at 7.2 µm, which had been never observed in the spectrum of the nanoparticles produced without a plasma. The 7.2 µm peak is considered to be produced in an O–H bending mode from chemisorbed H2 O molecules on the surfaces of the nanoparticles. A morphological change was also observed for the nanoparticles produced with a plasma field. Although the without-plasmaproduced alumina nanoparticles were all spherical, nanoparticles with a polyhedral shape based on an octahedral shape were formed due to the plasma.

Acknowledgment This research was partially funded by Research Fellowships from the Japan Society for the Promotion of Science for Young Scientists.

References 1. 2. 3. 4.

I. Levin and D. Brandon, J. Am. Ceram. Soc. 81 (1998) 1995. L. J. Alvarez, L. E. Le´on and H. Mu˜ noz, Catal. Lett. 26 (1994) 259. T. Onaka, T. de Jong and F. J. Willems, Astron. Astrophys. 218 (1989) 169. M. Kurumada, C. Koike and C. Kaito, Mon. Not. R. Astron. Soc. 359 (2005) 643.

Formation of Alumina Nanoparticles in Plasma

77

5. S. Kimura, C. Kaito and S. Wada, Antarct. Meteorite Res. 13 (2000) 145. 6. T. Sato, C. Kaito and Y. Saito, Surf. Rev. Lett. 10 (2003) 435. 7. K. S. K. Swamy, Dust in the Universe: Similarities and Differences, in World Scientific Series in Astronomy and Astrophysics, Vol. 7, ed. J. V. Narlikar. (World Scientific Publishing: Singapore, 2005), pp. 182–184. 8. C. Kaito, Y. Ojima, K. Kamitsuji, O. Kido, Y. Kimura, H. Suzuki, T. Sato, T. Nakada, Y. Saito and C. Koike, Meteor. Planet. Sci. 38 (2003) 49. 9. K. Kamitsuji, H. Suzuki, Y. Kimura, T. Sato, Y. Saito and C. Kaito, Astron. Astrophys. 429 (2005) 205. 10. T. Sato, K. Kamitsuji, M. Shintaku, Y. Kimura, M. Kurumada, O. Kido, H. Suzuki, Y. Saito and C. Kaito, Planet. Space Sci. 54 (2006) 617. 11. T. Sato, M. Kurumada, K. Kamitsuji, O. Kido, H. Suzuki, M. Shintaku, Y. Kimura, Y. Saito and C. Kaito, Planet. Space Sci. 54 (2006) 612. 12. J. A. Nuth III, F. J. M. Rietmeijer and H. G. M. Hill, Meteor. Planet. Sci. 37 (2002) 1579. 13. I. Sunagawa, Crystals: Growth, Morphology and Perfection (Cambridge University Press, Cambridge, 2000). 14. L. Grossman, Geochim. Cosmochim. Acta. 36 (1972) 597. 15. T. Yamamoto and H. Hasegawa, Prog. Theor. Phys. 58 (1977) 816.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


INFRARED STUDY OF UV/EUV IRRADIATION OF NAPHTHALENE IN H2 O+NH3 ICE Y.-J. CHEN∗,† , M. NUEVO† , F.-C. YEH† , T.-S. YIH† , W.-H. SUN‡ , W.-H. IP‡ , H.-S. FUNG§ , Y.-Y. LEE§ and C.-Y. R. WU¶ †Department of Physics, National Central University Chung-Li 32054, Taiwan, R.O.C. ‡Institute

of Astronomy, National Central University Chung-Li, 32049, Taiwan, R.O.C.

§National

Synchrotron Radiation Research Center Hsinchu 30076, Taiwan, R.O.C.

¶Space Sciences Center and Department of Physics and Astronomy University of Southern California Los Angeles, CA 90089-1341, U.S.A.

We have carried out photon irradiation study of naphthalene (C10 H8 ), the smallest polycyclic aromatic hydrocarbon (PAH) in water and ammonia ice mixtures. Photons provided by a synchrotron radiation light source in two broad-band energy ranges in the ultraviolet/near extreme ultraviolet (4–20 eV) and the extreme ultraviolet (13–45 eV) ranges were used for the irradiation of H2 O+NH3 +C10 H8 = 1:1:1 ice mixtures at 15 K. We could identify several photo-products, namely CH4 , C2 H6 , C3 H8 , CO, CO2 , HNCO, OCN− , and probably quinoline (C9 H7 N) and phenanthridine (C13 H9 N). We found that the light hydrocarbons are preferably produced for the ice mixture subjected to 4–20 eV photons. However, the production yields of CO, CO2 , and OCN− species seem to be higher for the mixture subjected to EUV photons (13–45 eV). Therefore, naphthalene and its photo-products appear to be more efficiently destroyed when high energy photons (E > 20 eV) are used. This has important consequences on the photochemical evolution of PAHs in astrophysical environments.

1. Introduction In the early twentieth century, a series of diffuse interstellar bands (DIBs) were recorded on photographic plates. More than 100 of such bands are observed nowadays in the ultraviolet (UV), visible and near infrared (IR) regions of the electromagnetic spectrum1−4 . The identification of ∗ Corresponding

author. E-mail: [email protected] 79

80

Y.-J. Chen et al.

the carriers responsible for these DIBs has become one of the most active challenges in astrophysical spectroscopy.5,6 Polycyclic aromatic hydrocarbons (PAHs) are now thought to be the best candidates to account for the DIBs in the interstellar medium (ISM), or most likely their cations.7,8 It has been suggested that PAHs could represent ∼17% of the cosmic carbon, and consequently be the most abundant free organic molecules in the Universe [7 and references therein]. PAHs constitute a group of very stable organic molecules made up from carbon and hydrogen only, assembled into aromatic cycles of 6 sp2 -carbon atoms like benzene (C6 H6 ) bound together. Such cycles are flat, so that PAHs are in most cases also flat, like graphite layers. It is believed that these molecules are formed in the outflows of dying carbon-rich stars from which they are ejected into the ISM7 . The presence of PAHs in the ISM has been confirmed by the astronomical observations of their C–H stretching and out-of-plane bending modes, both in emission9−11 and in absorption12−15 in embedded protostars. PAHs may condense onto refractory dust grains with other volatile species (ices), among which H2 O is the most abundant, and are abundant and ubiquitous in many different astrophysical environments.7,16,17 Bernstein et al.18,19 and Sandford et al.20 have been investigating the ultraviolet (UV) processing of PAHs in H2 O ices, emphasizing the possible connections between interstellar and meteoritic PAHs, and have shown that PAHs undergo both oxidation and reduction photo-reactions in ices, resulting in the production of aromatic hydrocarbon species similar to some of those identified in carbonaceous chondrites and interplanetary dust particles (IDPs). In this paper, we report results obtained from an experimental study of photon irradiations of naphthalene (C10 H8 ), the smallest PAH (only two aromatic cycles), in H2 O+NH3 ice mixtures at low temperature. Photons in the 4–20 and 13–45 eV ranges, i.e., from the ultraviolet to the extreme ultraviolet (EUV) ranges, were used to irradiate H2 O+NH3 +C10 H8 = 1:1:1 ice mixtures at 15 K in two separate experiments. EUV photons can excite molecules beyond their ionization continua and produce various neutral and ionic fragments. It has been shown that certain molecular species can be synthesized after irradiation with EUV photons, but not necessarily after vacuum ultraviolet (VUV) irradiation21 and vice versa. Finally, the use of two different energy ranges allows us to study the dependence of the production yields of several photo-products with the photon energy.

UV/EUV Irradiation Study of Naphthalene in H2 O+NH3 Ices

81

2. Experimental Protocol The experimental apparatus and techniques have previously been described elsewhere.21 In the present work, the UV/EUV irradiations of the icy samples containing naphthalene were performed in a stainless steel vacuum chamber (P < 5 × 10−10 torr). The reagents used in this work and their purities are as follows: H2 O (liquid, triply distilled), NH3 (gas, SigmaAldrich, 99.5% purity), and C10 H8 (powder, Sigma-Aldrich, 99% purity). The H2 O+C10 H8 +NH3 samples were vapor-deposited onto a KBr substrate at 15 K, by injecting a gas mixture of H2 O+NH3 and the C10 H8 vapor simultaneously through two separated stainless steel thin tubes (2-mm inner diameter). The relative proportions of the final H2 O+NH3 +C10 H8 = 1:1:1 mixtures were controlled from the partial pressures inside the stainless steel bottles where they were mixed. The typical thickness of the ice films was 1–3 µm, measured by monitoring the variation of interference fringes of a He-Ne laser light reflected by the KBr substrate. A Fourier-transform infrared spectrometer (FTIR) (Perkin-Elmer FTIR-1600) was used to record infrared (IR) spectra between 4000 and 500 cm−1 with a 4 cm−1 resolution. The IR and EUV beams form an angle of 90◦ on the substrate, so that IR spectra could be recorded in situ during the whole experiment, i.e., before, during and after irradiation. Once the gas deposition was completed, the vacuum chamber was left idling for a few hours so that all gases could condense on the KBr substrate. The irradiation experiments were performed after the pressure in the chamber reaches ∼5 × 10−10 torr. The IR spectrum of one of the H2 O+NH3 +C10 H8 =1:1:1 mixtures before irradiation is shown in Fig. 1. The broad-band UV/EUV beams were provided by the high-flux beamline of the National Synchrotron Radiation Research Center (NSRRC) in Hsinchu, Taiwan. The incident photon energy used was the 0th order of the white light in the 4–20 and 13–45 eV ranges using 450 and 1600 lines mm−1 gratings, respectively.22 The incident photon flux was constantly monitored by an in-line gold mesh. The irradiations were performed until a total integrated incident photon dose of about 1.5 × 1020 photons was reached for each irradiation experiment. We could therefore identify the compounds photo- produced after irradiation with 4–20 and 13–45 eV photons, and compare the results for both photon energy ranges.

82

Y.-J. Chen et al. 0.20 H2O+NH3+C10H8 = 1:1:1 at 15 K, before irradiation

Absorbance

0.15

0.10

0.05

0.00 4000

3500

3000

2500

2000

1500

1000

500

Wavenumber (cm-1)

Fig. 1. Infrared absorbance spectrum of one of the H2 O+NH3 +C10 H8 = 1:1:1 ice mixtures at 15 K in the 4000–500 cm−1 range before irradiation.

3. Results and Discussion 3.1. Photo-dissociation of naphthalene Figure 2 shows the IR spectra (difference of absorbances) of the H2 O+NH3 +C10 H8 = 1:1:1 ice mixtures at 15 K irradiated with UV/EUV photons. The measured spectral positions of the absorption features and the identified species are summarized in Table 1. As can be seen from Fig. 2 most of the major features in both spectra look alike although their peak absorbances are different. The signatures of some photo-products however only appear on the spectrum obtained after 4–20 eV irradiation (upper trace), but not on the spectrum obtained after 13–45 eV irradiation (lower trace). These compounds are the aliphatic hydrocarbons ethane (C2 H6 ), proprane (C3 H8 ), and the aromatic benzyl radical (C6 H5 CH2 ), the later being only tentatively identified. It is likely that these compounds are easily destroyed by high energy (E > 20 eV) photons or that they readily react with other chemical species. Two other compounds (CH2 N and C2 O), whose features are marked with an asterisk, were only tentatively identified. In the following, we will discuss the possible photo-induced chemical reaction mechanisms of these compounds, and their link with the irradiation photon energy. The photo-products formed after irradiation with 4–20 eV photons include CH3 OH (1030 cm−1 ), CH4 (1305 cm−1 ), C2 H6 (2873 and 2931 cm−1 ), C3 H8 (2959 cm−1 ), CO (2134 cm−1 ), CO2 (2338 cm−1 ), HNCO −1 ). Three other (2255 cm−1 ), OCN− (2159 cm−1 ) and NH+ 4 (1453 cm

Difference of absorbances: 13-45 eV

CO2

0.06

0.06 C3H8

C2H6

HNCO C2H6

-

OCN

CO

0.04

CH2N?

*

+

NH4

CH4

C2O?

0.02

0.04

CH3OH C6H5CH2?

*

*

0.00

-0.02 4000

0.02

3500

3000

2500

2000

1500

1000

83

Difference ofabsorbances: 4-20 eV

UV/EUV Irradiation Study of Naphthalene in H2 O+NH3 Ices

500

-1

Wavenumber (cm )

Fig. 2. IR spectra (difference of absorbances) of the H2 O+C10 H8 +NH3 = 1:1:1 ice mixtures at 15 K after irradiation with 4–20 eV (upper trace) and 13–45 eV (lower trace) photons. The features marked with* correspond to the tentatively identified compounds listed in Table 1.

Table 1. Photo-products identified in the IR spectra of the H2 O+C10 H8 +NH3 = 1:1:1 ice mixtures at 15 K after UV/EUV irradiation. The species marked with * were tentatively identified. The abbreviation n.d. stands for “not detected”. Peak position (cm−1 )

Species

C3 H 8 C3 H 8 C2 H 6 CH2 N* CO2 HNCO OCN− CO C2 O* NH+ 4 CH4 CH3 OH C6 H5 CH2 *

after 4–20 eV irradiation

after 13–45 eV irradiation

2959 2931 2873 2722 2338 2255 2159 2134 1898 1453 1300 1030 863

n.d. n.d. n.d. n.d. 2338 2253 2160 2135 1902 1450 1299 1030 n.d.

absorption features, marked with an asterisk in Fig. 2, were only tentatively identified. They could be assigned to CH2 N (methylene amidogen) (2722 cm−1 ), C2 O (1898 cm−1 ) and the benzyl radical C6 H5 CH2 (863 cm−1 ).23,24 They are also listed in Table 1.

84

Y.-J. Chen et al.

In the H2 O+NH3 +C10 H8 ice samples, naphthalene is the only molecule containing carbon atoms. Thus, all photo-products containing carbon can only be formed from the UV/EUV photolysis of C10 H8 . This means that the aromatic cycles have to be dissociated into smaller fragments to produce aliphatic hydrocarbon chains (containing only C and H atoms). It is known that breaking aromatic cycles of PAHs is not an efficient process when using photons in the 5–10 eV range. Indeed, using a microwave discharge H2 flow lamp as the light source, Bernstein et al.18 showed that only substitution reactions of H atoms and/or addition of heteroatoms are efficient process in the 5–10 eV energy range. As listed in Table 1, CH4 has been produced after the photon irradiation of the H2 O+NH3 +C10 H8 ice mixtures with both energy ranges. However, C2 H6 and C3 H8 were only identified in the IR spectrum of the mixture subjected to 4–20 eV photons. Separate experiments, where pure C10 H8 and a H2 O+C10 H8 mixture were irradiated with 4–20 eV photons at low temperature, were carried out in our laboratory. In the obtained IR spectra, we identified the features of CH4 , C2 H6 and C3 H8 , indicating that alkanes Cn H2n+2 (n ≥ 1) can be directly produced from the photolysis of naphthalene and/or via secondary chemical reactions involving radicals such as CH3 and C2 H5 . For instance, C2 H6 could be a secondary product of the photo-dissociation of CH4 : CH4 + hν → CH3 • + H• CH3 • + CH3 • → C2 H6 where the species followed by dots are radicals. If the CH3 radical is efficiently produced, then it may also react with the HO radical (produced from the photo-dissociation of H2 O) to form methanol (CH3 OH). Pure methanol ice displays two strong absorption features at 1030 and 1128 cm−1 .25 Unfortunately, in our IR spectra the feature around 1128 cm−1 overlaps with one of the depletion features of naphthalene (see Fig. 2), and thus it can not be observed. However, a discernible feature around 1030 cm−1 can be assigned to methanol, and indirectly support the photochemical pathway leading to the production of CH3 radicals and therefore aliphatic molecules such as C2 H6 and C3 H8 . Furthermore, the feature at 863 cm−1 , probably due to the benzyl radical (C6 H5 CH2 )24 , can clearly be identified in the IR spectrum of the 4–20 eV irradiated sample (upper trace of Fig. 2) but not in the spectrum of the 13–45 eV irradiated sample (lower trace of Fig. 2). This result

UV/EUV Irradiation Study of Naphthalene in H2 O+NH3 Ices

85

indicates that naphthalene is progressively photo-dissociated into fragments becoming smaller and smaller as the photon energy increases from the UV to the EUV ranges. In other words, high energy (E > 20 eV) photons appear to dissociate naphthalene with a higher efficiency than less energetic UV photons. Therefore, we can conclude that although both 4–20 and 13–45 eV photons can dissociate naphthalene, only 13–45 eV photons can totally break the aromatic structure of the naphthalene molecule. In addition to the production of small aliphatic hydrocarbons, we have also identified the IR features of O- and N-containing compounds, namely CO (2134 and 2135 cm−1 in the 4–20 and 13–45 eV experiments, respectively), CO2 (2338 cm−1 ), OCN− (2159 and 2160 cm−1 ) and HNCO (2255 and 2253 cm−1 ). These molecules are probably formed by recombination of small aliphatic compounds (or their radicals) with HO and NH2 , produced from the photo-dissociation of water and ammonia, respectively. The possible mechanisms of formation of such compounds will be discussed in Sec. 3.2. Finally, some weak absorption features in the 1600–1300 and the 820– 720 cm−1 ranges can be assigned to nitrogen-bearing PAHs (see Fig. 3 and Table 2), called polycyclic aromatic nitrogen heterocycles (PANHs).

Relative Difference of Absorbances

0.034

0.034

0.030

0.036 0.032

0.025

0.032 0.034

0.030

0.020 0.030 0.032

0.028

0.015 0.028

0.030 0.026 1590

1560

1530

1500

1470

1440

1380

1350

1320

0.010 780

760

740

-1

Wavenumber ( cm )

Fig. 3. IR spectra (difference of absorbances) of the H2 O+C10 H8 +NH3 = 1:1:1 ice mixtures at 15 K after irradiation with 4–20 (upper trace) and 13–45 eV (lower trace) photons in 4 chosen small wavenumber ranges. The dashed lines correspond to the weak absorption features of quinoline and phenanthridine (see Table 2).

86

Y.-J. Chen et al. Table 2. Weak absorption features have tentatively been assigned to quinoline (C9 H7 N) and phenanthridine (C13 H9 N) in the 1600–1300 and 820–720 cm−1 ranges26 observed for the H2 O+C10 H8 +NH3 = 1:1:1 ice mixtures at 15 K after UV/EUV irradiation (see Fig. 3). Peak position (cm−1 )

Species

after 4–20 eV irradiation

after 13–45 eV irradiation

C9 H 7 N

1439 1375 1322 767

1440 n.d. 1321 767

C13 H9 N

n.d. n.d. 1494 n.d. 1342 756

1584 1529 1493 1462 1341 752

The positions of these features appear to be in good agreement with the absorption features of quinoline (C9 H7 N) and phenanthridine (C13 H9 N) in water ice at 15 K.26 Quinoline is a molecule of naphthalene where one of the carbon atoms has been substituted by a nitrogen atom. We plan to carry out further investigations using a higher resolution FTIR spectrometer (0.1 cm−1 ) and a brighter photon source in order to better resolve the observed features. These important results indicate that naphthalene and possibly other PAHs can be photo-dissociated in cold astrophysical environments by EUV photons and contribute to the reservoir of carbon whose photochemical evolution can lead to the production of complex organic molecules in the ISM.

3.2. Production yields of CO, CO2 and OCN− We have determined the production yields of CO, CO2 and OCN− , photoproduced during the irradiation of the H2 O+NH3 +C10 H8 = 1:1:1 ice mixtures at 15 K with 4–20 and 13–45 eV photons, using data analysis procedures which have been previously described elsewhere.21 The column densities of CO, CO2 and OCN− are plotted as a function of the photon dose in Figs. 4, 5 and 6, respectively. Absorption strengths of A(CO, 2134 cm−1 ) = 1.1 × 10−17 cm molec−1 , A(CO2 , 2340 cm−1 ) =

UV/EUV Irradiation Study of Naphthalene in H2 O+NH3 Ices

87

4 - 20 eV 13 - 45 eV

30

25

15

-2

CO column density (x 10 molec cm )

35

20

15

10

5

0 0

5

10

15

20 19

25

30

-2

Photon dose (x 10 photons cm )

Fig. 4. Plot of the column density of CO as a function of the 4–20 eV (squares) and 13–45 eV (triangles) photon doses.

4 - 20 eV 13 - 45 eV

-2

CO2 column density (x 10 molec cm )

6

5

15

4

3

2

1

0 0

5

10

15

20 19

25

30

-2

Photon dose (x 10 photons cm )

Fig. 5. Plot of the column density of CO2 as a function of the 4–20 eV (squares) and 13–45 eV (triangles) photon doses.

7.6× 10−17 cm molec−1 and A(OCN− , 2160 cm−1 ) = 4× 10−17 cm molec−1 , respectively27,28 , were used for these calculations. Figures 4–6 show that the column densities for CO, CO2 and OCN− , photo-produced during the 13–45 eV experiment (triangles) are respectively

88

Y.-J. Chen et al.

4 - 20 eV 13 - 45 eV 5

4

3

2

-

15

-2

OCN column density (x 10 molec cm )

6

1

0 0

5

10

15

20 19

25

30

-2

Photon dose (x 10 photons cm )

Fig. 6. Plot of the column density of OCN− as a function of the 4–20 eV (squares) and 13–45 eV (triangles) photon doses.

3.6, 2.2 and 1.9 times higher than the corresponding yields for the 4–20 eV experiment (squares) after an integrated photon dose of approximately 1 × 1020 photons cm−2 . This result suggests that small C-bearing molecules are produced from the direct photo-dissociation of naphthalene, which is the only carbon- bearing compound in our starting mixtures, and confirm that the photo- dissociation efficiency for such small molecules increases with the photon energy. Figures 4 and 5 also show that CO is produced with a higher efficiency than CO2 . The column densities for CO are about 3.5 and 5.8 times higher than those for CO2 in the 4–20 and 13–45 eV experiments, respectively. This can be understood because CO2 is most likely a secondary photo-product, formed via the photolysis of CO. A possible mechanism of formation of CO2 from CO would be the photo-excitation of CO and the subsequent reaction with CO29 : CO + hν → CO∗ (A1 Π, a3 Π) CO∗ + CO → CO2 + C•, where A1 Π and a3 Π denote excited states of CO. The carbon atom released in this mechanism does not remain free in the medium, and may react with another CO molecule to produce C2 O. This mechanism is supported by the

UV/EUV Irradiation Study of Naphthalene in H2 O+NH3 Ices

89

presence of two IR features of C2 O at 1898 and 1902 cm−1 in the irradiated samples23 (see Fig. 2 and Table 1). CO2 can also be formed from the reaction between CO and the HO radical, which is produced from photo-dissociation of H2 O30 : H2 O + hν → HO• + H• HO• + CO → •COOH → CO2 + H• . The photolysis of CO to form CO2 and COOH radicals constitute the first steps for the formation of complex organic molecules. In particular, reactions between the parent molecules and their photo-products including radicals such as H, HO, Cn H2n+1 (n ≥ 1) (alkyl radicals), COOH, NH, NH2 at low temperature could lead to the production of molecules as complex as amino acids in such experiments, but also in astrophysical environments.

4. Conclusion We have investigated the effects of the irradiation of H2 O+NH3 +C10 H8 = 1:1:1 ice mixtures with 4–20 eV (UV/near EUV) and 13–45 eV (EUV) photons. In the IR spectra of these samples, we have identified the characteristic features of several photo-products, namely CH4 , C2 H6 , C3 H8 , CO, CO2 , HNCO and OCN− . Methyl amidogen (CH2 N), C2 O and the benzyl radical (C6 H5 CH2 ) have also been tentatively identified. Our work also shows that on the one hand small hydrocarbons such as C2 H6 , C3 H8 and the benzyl radical are significantly produced during the 4–20 eV irradiation experiment, and that on the other hand the production yields of CO, CO2 and OCN− are significantly higher in the 13–45 eV irradiation experiment. Therefore, the photo-products and their production yields strongly depend on the photon energy. The present work shows that EUV photons can efficiently photodissociate naphthalene and probably other PAHs at low temperature. This has important implications on the photochemical evolution of PAHs in astrophysical environments, where the carbon reservoir could contribute significantly to the production of complex organic molecules including amino acids, the building blocks of proteins in all living beings on Earth.

90

Y.-J. Chen et al.

Acknowledgments We are grateful for the support of the staff of the National Synchrotron Radiation Research Center in Hsinchu, Taiwan. This research is supported in part by the Ministry of Education under the Aim for the Top University Program (NCU) and based on the work supported by the NSC grant #NSC-95-2112-M008-028, the NSF Planetary Astronomy Program under Grant AST-0604455 (C.-Y. R. W.), and the NASA Planetary Atmospheres Program under Grant NAG5-11960 (C.-Y. R. W.).

References 1. R. W. Russell, B. T. Soifer and S. P. Willner, Astrophys. J. 217 (1977) L149. 2. K. Sellgren, M. W. Werner and H. L. Dinerstein, Astrophys. J. 271 (1983) L13. 3. J. D. Bregman, H. L. Dinerstein, J. H. Goebel, D. F. Lester, F. C. Witteborn and D. M. Rank, Astrophys. J. 274 (1983) 666. 4. J. P. Simpson, J. D. Bregman, M. Cohen, F. Witteborn and D. H. Wooden, Bull. AAS 16 523 (1984). 5. A. L´eger, L. d’Hendecourt and D. D´efourneau, Astron. Astrophys. 293 (1995) L53. 6. F. M. Johnson, Bulletin of the Am. Astron. Soc. 33 (2000) 716. 7. J. L. Puget and A. L´eger, Annual Review of Astron. Astrophys. 27 (1989) 161. 8. G. Mulas, G. Malloci and P. Benvenuti, Astron. Astrophys. 410 (2003) 639. 9. L. J. Allamandola, D. M. Hudgins and S. A. Sandford, Astrophys. J. 511 (1999) L115. 10. E. Peeters, A. L. Mattioda, D. M. Hudgins and L. J. Allamandola, Astrophys. J. 617 (2004) L65. 11. J. M. Cannon and 24 co-authors, Astrophys. J. 647 (2006) 293. 12. R. G. Smith, K. Sellgren and A. T. Tokunaga, Astrophys. J. 344 (1989) 413. 13. K. Sellgren, T. Y. Brooke, R. G. Smith and T. R. Geballe, Astrophys. J. 449 (1995) L69. 14. J. D. Bregman, Th. L. Hayward and G. C. Sloan, Astrophys. J. 544 (2000) L75. 15. J. D. Bregman and P. Temi, Astrophys. J. 554 (2001) 126. 16. L. J. Allamandola, S. A. Sandford and B. Wopenka, Science 237 (1987) 56. 17. S. J. Clemett, C. R. Maechling, R. N. Zare, P. D. Swan and R. M. Walker, Science 262 (1993) 721. 18. M. P. Bernstein, S. A. Sandford, L. J. Allamandola, J. S. Gillette, S. J. Clemett and R. N. Zare, Science 283 (1999) 1135. 19. M. P. Bernstein, J. P. Dworkin, S. A. Sandford and L. J. Allamandola, Meteoritics & Planetary Science 36 (2001) 351.

UV/EUV Irradiation Study of Naphthalene in H2 O+NH3 Ices

91

20. S. A. Sandford, M. P. Bernstein and L. J. Allamandola, Astrophys. J. 607 (2004) 346. 21. C.-Y. R. Wu, D. L. Judge, B.-M. Cheng, W.-H. Shih, T.-S. Yih and W.-H. Ip, Icarus 156 (2002) 456. 22. T.-F. Hsieh, L.-R. Huang, S.-C. Chung, T.-E. Dann, P.-C. Tseng, C.-T. Chen and K.-L. Tsang, J. Synchrotron Rad. 5 (1998) 562. 23. M. Jacox and D. E. Milligan, J. Chem. Phys. 43 (1965) 3734. 24. E. G. Baskir, A. K. Maltsev, V. A. Koroler, V. N. Khabasheska and O. M. Nefedov, Russ. Chem. Bul. 42 (1993) 1438. 25. D. M. Hudgins, S. A. Sandford, L. J. Allamandola and A. G. G. M. Tielens, Astrophys. J. Suppl. Ser. 86 (1993) 713. 26. M. P. Bernstein, A. L. Mattioda, S. A. Sandford and D. M. Hudgins, Astrophys. J. 626 (2005) 909. 27. P. A. Gerakines, W. A. Schutte, J. M. Greenberg and E. F. van Dishoeck, Astron. Astrophys. 296 (1995) 810. 28. L. B. d’Hendecourt and L. J. Allamandola, Astron. Astrophys Suppl. Ser. 64 (1986) 453. 29. H. Okabe, Photochemistry of small molecules (Wiley, New York, USA, 1978). 30. N. Watanabe and A. Kouchi, Astrophys. J. 571 (2002) L173.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


NEW METHOD OF PRODUCING TITANIUM CARBIDE, MONOXIDE, AND DIOXIDE GRAINS IN LABORATORY AKIHITO KUMAMOTO∗ , MAMI KURUMADA, YUKI KIMURA and CHIHIRO KAITO Department of Physics, Ritsumeikan University, Noji-higashi 1-1-1 Kusatsu, Shiga 525-8577, Japan ∗[email protected]

By making a carbon rod covered with Ti on the surface without exposure to air, TiC grains less than 10 nm in diameter were predominantly produced. The introduction of a small amount of oxygen in Ar gas (partial pressure 1/1000), allowed the continuous formation of TiO2 and TiO–TiC. The infrared spectra of TiO2 , TiO, and TiC were measured. An absorption feature attributed to TiO phase in oxidized TiC grains showed a characteristic peak at 14.7 µm.

1. Introduction Titanium oxides are considered to be the first species to condense in oxygenrich environments.1,2 With respect to dust formation in circumstellar shells, it is important to note that the TiO molecule is rather prominent in the atmosphere of O-rich stars. Solid titanium oxides such as TiO, TiO2 , Ti2 O3 , Ti3 O5 , and Ti4 O7 are considered to play a role in the formation of solid Ti compounds in the gas phase.3 TiO2 exists in three different phases, namely, rutile, anatase, and brookite. The infrared (IR) spectra of these TiO2 phases were amply measured by Posch et al.4 Another phase displayed by Tin O2n−1 compounds is the Magneli phase, which is derived from the rutile structure by introducing oxygen planar defects as elucidated in a previous paper.5 In the previous paper, NaCl-type structure of the ordinary TiO phase did not appear by the reduction of the TiO2 phase.5 TiO of a superstructure, which is an ordered structure observed in order–disorder transition alloys, was found on the sample prepared from mixture of iodide titanium and TiO2 within a tungsten-arc furnace in a gas atmosphere of purified argon.6 93

94

A. Kumamoto et al.

On the other hand, the 20.1 µm feature observed in the spectra of some post-AGB stars was proposed to be due to titanium carbide.7 Subsequent studies showed that the attribution of this feature to titanium carbide is arguable.8–10 A number of TiC grains have been found to be of presolar origin in graphite spherules extracted from the Murchison meteorite. TiC crystals can be formed in AGB star atmospheres only in high-density regions such as the superwindow phase.11 In the laboratory, TiC crystallites (2–3 nm in size) produced by the reaction of Ti metal grains with carbon films,12 and TiC grains (50 nm in size) of cubic shape produced by the gas evaporation method13 showed no peaks at 20.1 µm. Recently, Kimura et al. have produced TiC grains covered by a carbon mantle from CO gas by the Boudouard reaction and have showed the existence of large fullerenes contributing to the 21 µm feature.14 In this paper, a new experimental method of producing TiC, TiO2 , and TiO grains in the laboratory is described. It was found that the TiO phase can be formed with a small amount of oxygen even for TiC grains produced in carbon-rich conditions. It was suggested from theoretical calculation that the condensation temperature of TiC grains is lower than that of titanium oxide.3 Finally, we present the conditions of formation of TiC grains in an oxygen-poor environment, corresponding to an environment where C/O ≈ 1, as for S-stars, in a laboratory experiment.

2. Experimental Methods The evaporation chamber was a glass cylinder, 17 cm in diameter and 30 cm high, covered with a stainless-steel plate on top and connected to a high-vacuum exhaust through a valve at the bottom. A Ti wire (0.25 mm diameter) was placed on the concave position between a pair of carbon rods (0.5 mm diameter) as schematically showed in Fig. 1. After heating the

Fig. 1. Schematic presentation on proposed experimental method. Two carbon rods with molten Ti metal were used as the evaporation source to produce smoke. Smoke grains were produced in Ar gas at 10 kPa or a gas mixture of Ar and O2 at 10 K and 10 Pa, respectively.

New Method of Producing Titanium Oxides

95

system to above 2000 K, the Ti powder or wire melted and wrapped around the carbon rods. Since the temperature distribution of the carbon rods was very uniform, the grain size was controllable. When heating up to 2300 K, TiC grains were produced in the Ar gas at 10 kPa. When oxygen was added with a partial pressure of up to 10 Pa, titanium oxide grains were produced. Then, as oxygen partial pressure decreased, the crystal structure of the formed grains changed from the TiO2 phase to the TiO phase. The samples were observed using a Hitachi H-7100R transmission electron microscope (TEM) equipped with an energy-dispersion X-ray analysis system (Horiba Xerophy), and also using a Hitachi H-9000NAR high-resolution TEM (HRTEM). The transmission IR of the samples embedded in KBr pellets to a concentration of less than 1% in the 5–25 µm range were measured with a Fourier-transform infrared (FTIR) spectrometer (Horiba FT210).

3. Results and Discussion TiC grains covered with a thin carbon layer were produced by heating carbon rods wrapped by the Ti wire at 2300 K in an Ar atmosphere of 10 kPa, as shown in Fig. 2. The produced grains which are seen in Fig. 2a are TiC grains. They have NaCl-type structure, determined from the electron diffraction (ED) pattern (Fig. 2b). Small TiC grains less than 10 nm in diameter were predominantly produced and their external shape is a truncated octahedron as indicated by arrows and a typical HRTEM image shown in the inset of Fig. 2c. In spite of the different shape compared with the cubic form grains observed in a previous paper,13 the same IR absorption peak was seen at 12.5 µm. Therefore, it can be concluded that the differences in size and shape did not influence the spectral feature of TiC in this size range. Since Ti–C phase diagram shows that TiC and C form a eutectic for a relative proportion of C of 30 wt.% (63 at.%), TiC and C evaporate at the same temperature as the eutectic composition. Therefore, carbon atoms in excess are produced and can surround the TiC grain surface. When mixing a small amount of oxygen (10 Pa) with the Ar gas at 10 kPa, TiO2 grains were initially produced. The color of the evaporating grains varied from white to black during the experiment. White is the typical color of TiO2 grains. According to TEM analysis, the black grains were composed of a mixture of TiC and TiO as shown in Fig. 3. Therefore, the structure of the produced grains changed from the TiO2 phase to the

96

A. Kumamoto et al.

(b)

(a)

TiC

(c)

111 200 220 311 222 400 331 420 422

(200)

1) (11

Fig. 2. (a) TEM image and (b) corresponding ED pattern of TiC grains produced in an Ar atmosphere of 10 kPa. (c) the HRTEM image shows that TiC grains are covered with a thin carbon layer.

(a)

(b)

100 nm

Fig. 3. Typical TEM image and corresponding ED pattern of evaporating grains collected at the end of experiment. The color of these grains was black. The formation of TiO could be deduced from the higher-order index of the ED pattern.

New Method of Producing Titanium Oxides

97

TiO phase with decreasing oxygen partial pressure starting from 1/1000 of Ar pressure. Figure 4 also shows an HRTEM image of TiO and mixed TiO/TiC grains. The (111) plane of the upper right TiO grain, with a size of 10 nm, is nearly parallel to the (111) plane of the large central TiC grain. The contrast in the stripes in the TiC (1¯11) lattice image may be due to the existence of a TiO layer. A few layers of TiO are present at the positions indicated by arrows. Since TiC and TiO have the same NaCl-type structure and very close lattice constants, the mismatch between the two phases is small (3.6%). Therefore, these mixed grains may grow well in an oxygenpoor atmosphere. The surfaces of these grains are covered with a thin

(200)TiO 0.21 nm

(110)TiO2- Rutile 0.32 nm

55˚ (111)TiO 0.24 nm (111)TiC 0.25 nm

70.5˚ (111)TiC 0.25 nm

10 nm

(110)TiO2- Rutile 0.32 nm

Fig. 4. HRTEM image of produced grain at oxygen partial pressure of 10 Pa. Grains consisting of a mixture of TiC and TiO phases were produced at the atomic layer level. TiO2 crystallites and amorphous carbon covered the mixture of TiC and TiO grains.

98

A. Kumamoto et al.

carbon layer and contain crystallites with the rutile (TiO2 ) structure. The formation of TiO2 crystallites at/on the surface carbon layer can be deduced from the lattice spacing and the angle of crossed lattice images. When the oxygen relative partial pressure was of the order of 1/1000 at the beginning of the experiment, the TiC–C mixture vapor, which were evaporated with the eutectic reaction, was oxidized and TiO2 grains were produced. Since TiO2 grains consume oxygen gas, the oxygen partial pressure was gradually decreasing during the experiment. Then, TiO and TiC phases were formed. Figure 5 shows the IR spectra of (a) TiO2 and (b) the TiC–TiO mixture. TiO and TiC phases have structures based on corner-sharing octahedrons, and TiO2 also consists of distorted octahedrons having the edge-sharing structure. Two peaks at 14.3 and 16.7 µm appear in Fig. 5a. The feature at 12.5 µm (Fig. 5b) is attributed to TiC.13 The 14.7 µm feature in spectrum (b) may be due to TiO, which have distorted octahedral structures formed by oxidized TiC. The 14.3 and 16.7 µm features (trace (a)) may also be due to the stretching mode of the distorted octahedron.15,16 These IR spectra show that the feature around 14 µm caused by titanium oxide easily appears with the introduction of a small amount of oxygen.

100 90

(a)

80

Transmittance [%]

70 12.5

60 (b)

50

14.7

40 30 16.7 20

14.3

10 0 5

7

9

11

13

15

17

19

21

23

25

Wavelength [µm]

Fig. 5. IR absorption spectra of laboratory grains embedded in KBr pellets in the 5–25 µm range. The spectra correspond to the grains produced during the evolution from (a) white TiO2 grains to (b) a black TiO–TiC grain mixture, following the gradually decreasing partial pressure of oxygen.

New Method of Producing Titanium Oxides

99

4. Summary A new evaporation method for TiC–C mixture compound was achieved for the production of solid TiO on formed TiC dust. TiC grains less than 10 nm in diameter had predominantly truncated octahedral structures covered with a carbon layer. In the case of the formation of TiC grains in Ar atmosphere with a small amount of oxygen, the composition of the produced grains varied from TiO2 to a TiO–TiC mixture as oxygen partial pressure decreased. The characteristic IR features at 12.5 and 14.7 µm assigned to TiC and TiO, respectively, characterized by the change of color of the evaporating grains from white to black during the experiment.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

S. S. Barshay and J. S. Lewis, Ann. Rev. Astron. Astrophysics 14 (1976) 81. E. E. Salpeter, Ann. Rev. Astron. Astrophysics 15 (1977) 267. H. P. Gail and E. Sedlmayr, Faraday Discuss. 109 (1998) 303. Th. Posch, F. Kerschbaum, D. Fabian, H. Mutschke, J. Dorschner, A. Tamani and Th. Henning, Astrophys. J. Suppl. Ser. 149 (2003) 437. C. Kaito, M. Iwanishi, T. Harada, T. Miyano and M. Shiojiri, Trans. Jpn. Inst. Meteoroe. 24 (1983) 450. D. Watanabe, J. R. Castles, A. Jostsons and A. S. Malin, Acta Cryst. 23 (1967) 307. G. von Helden, A. G. G. M. Tielens, D. van Heijnsbergen, M. A. Duncan, S. Hony, L. B. F. M. Waters and G. Meijer, Science 288 (2000) 313. S. Hony, A. G. G. M. Tielens, L. B. F. M. Waters and A. de Koter, Astron. Astrophys. 402 (2003) 211. A. Li, Astrophys. J. 599 (2003) L45. T. Chigai, T. Yamamoto, C. Kaito and Y. Kimura, Astrophys. J. 587 (2003) 771. T. Chigai, T. Yamamoto and T. Kozasa, Astrophys. J. 510 (1999) 999. Y. Kimura, A. Ikegami, M. Kurumada, K. Kamitsuji and C. Kaito, Astrophys. J. Suppl. Ser. 152 (2004) 297. Y. Kimura and C. Kaito, Mon. Not. R. Astron. Soc. 343 (2003) 385. Y. Kimura, J. A. Nuth III and F. T. Ferguson, Astrophys. J. 632 (2005) L159. M. Kurumada, O. Kido, T. Sato, H. Susuki, Y. Kimura, K. Kamitsuji, Y. Saito and C. Kaito, J. Cryst. Growth 275 (2005) e1673. M. Kurumada and C. Kaito, J. Phys. Soc. Jpn. 75 (2006) 074712.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


DESTRUCTION YIELDS OF NH3 PRODUCED BY EUV PHOTOLYSIS OF VARIOUS MIXED COSMIC ICE ANALOGS C. Y. R. WU∗ , T. NGUYEN and D. L. JUDGE Space Sciences Center and Department of Physics and Astronomy University of Southern California, Los Angeles, CA 90089-1341, USA ∗[email protected] H.-C. LU, H.-K. CHEN and B.-M. CHENG National Synchrotron Radiation Research Center 101 Hsin-An Road, Hsinchu Science Park Hsinchu 30076, Taiwan

Experimental measurements of the destruction yields of NH 3 have been carried out by extreme ultraviolet–vacuum ultraviolet (EUV–VUV) photolysis of cosmic ices containing NH3 . The ice systems studied in the present work include pure NH3 ices and icy mixtures of NH3 with CO, H2 O, and CH4 at a temperature of 10 K. A tunable intense synchrotron radiation light source, available at the National Synchrotron Radiation Research Center, Hsinchu, Taiwan, was employed to provide the required EUV–VUV photons. In this study, the photon wavelengths used to irradiate the icy samples were mainly selected to center on the prominent solar lines, namely, the 30.4, 58.4 and 121.6 nm. The photodestruction yields of NH3 in the presently studied ice mixtures are typically higher than 0.5 and can be higher than unity, a very efficient ice photochemical process.

1. Introduction The surfaces of astronomic objects are subjected to continuous irradiation by solar and interstellar photons, stellar winds, and cosmic rays.1 Modification of chemical composition and mass transfer/evolution in the frozen surfaces are thus expected to occur. The study of photolysis effects in ices is important in order to improve our understanding of the surface evolution of the planetary icy satellites and rings, comets, interstellar medium and grains, and dense molecular clouds. Therefore, we need laboratory simulation studies of photolysis/photoprocessing and charged particle impact on realistic cosmic ice analogs. We have recently 101

102

C. Y. R. Wu et al.

implemented a laboratory program to study ice photochemistry and to provide extreme ultraviolet–vacuum ultraviolet (EUV–VUV) photolysis data for interpreting the above-mentioned observations of cosmic objects. In this preliminary work, we report laboratory results on the measurements of destruction yields of NH3 in several ice systems. The NH3 ice has been observed on the surfaces of Uranus’s moon Miranda, Pluto’s moon Charon, a Kuiper Belt object Quaoar, and dense molecular clouds. It is also possible that NH3 exists in the form of ammonia hydrates2 and 3 NH+ 4 in the ice mantles in these objects. An intensive search for evidence of NH3 ices existing under the surface of Saturn’s moon Enceladus is currently being conducted by the Cassini mission. [For more information about the Cassini-Huygens mission the interested reader is recommended to visit http://saturn.jpl.nasa.gov.] It is important to point out that photodestruction cross sections of several pure and binary ice systems have previously been measured4–6 by using undispersed output from a microwave-discharged H2 flow lamp. The spectral outputs of the lamp include the HI 121.6 nm line and the H2 molecular band emissions in the spectral region between 135 and 145 nm, sharp bands in 145–155 nm, and the very broad emission bands in 150–300 nm.7 Its brightness varies according to the operating conditions. At very low H2 flow pressures or 2% H2 in He in the discharge tube the spectral outputs will almost exclusively consist of the atomic HI 121.6 nm emission.8,9 However, at high H2 flow pressures the intensity of the HI 121.6 nm emission decreases significantly and the H2 molecular band emissions become the dominant outputs.4,7,10 The relative output intensities10 of the H2 molecular bands strongly vary with the H2 pressure (i.e., H2 molecular density). The photolysis data generated by using the microwave-discharged H2 flow lamp indeed cover broad bands of the photon wavelength range of interest. Therefore, photolysis study using such a lamp is qualitatively valuable because it can only provide data averaged over a broad spectral region. It may be also important to add that the UV spectrum of the interstellar medium is best simulated by the microwave-discharged H2 flow lamp. The output spectrum of the lamp resembles, but is not identical to, the spectrum of the diffuse interstellar medium.11,12 Furthermore, excitation of H2 in cosmic environments by, e.g., charged particles, photon, cosmic rays, and so on, may produce UV photons similar to that provided by microwave-discharged H2 flow lamp. However, the output spectrum will of course depend on the H2 pressure condition in the given cosmic

Destruction Yields of NH3

103

environment of interest. It is commonly accepted that the diffusive UV background in interstellar environment is longer than 91.2 nm, the HI ionization threshold. The EUV diffusive background emission is known to be weak under such circumstances. However, in a small region around the Sun, hot white dwarfs, subdwarfs, and novae the EUV radiation photons are enough to seriously affect photochemical and photophysical processes on surrounding matters.12 The photolysis data obtained in the EUV region can thus provide crucial parameters for modeling the evolution of prebiotic matters including the possible formation of amino acids in icy grains. For obvious reasons, we really need to use a monochromatic light source in the photolysis study in order to provide meaningful and accurate data for modeling cosmic ice photochemistry throughout the UV and EUV regions. In our work we have employed a synchrotron radiation facility, which provides selectable monochromatic light with a well-defined spectral bandwidth, as previously described.13–15 In the present work, we choose four basic molecules of great importance in the cosmic environments for the make-up of the mixed ice samples. They are CO, H2 O, NH3 , and CH4 . The relative absorbances of solid CO ices in the 106–170 nm spectral region,16 H2 O ices in the 106–180 nm region, NH3 ices in the 106–240 nm region, and CH4 ices in the 106– 160 nm spectral region17 have recently been measured at the National Synchrotron Radiation Research Center (NSRRC). However, there are no such data for the above solid ices in the photon wavelength region shorter than 106 nm, the LiF window cut off. Therefore, our discussion of the EUV photoexcitation processes in ices will mainly rely upon what we know about the molecules of interest in the gas phase. The absorption of 121.6 nm photons by gaseous NH3 will result in dissociation into NH2 + H, NH + H2 , and NH + 2H fragments, and gaseous CH4 yields the CH3 + H and CH2 + H2 fragments.18–21 However, the 121.6 nm photons can only excite CO to its excited valence electronic states.16,22,23 Therefore, the photolyzed products produced through photolysis at different photon wavelengths may be different, with different production and destruction yields. Specifically, we have carried out simulation experiments on EUV– VUV photolysis of pure NH3 ices and mixed ices of CO + NH3 (1:1) and (4:1), NH3 + CH4 (1:1), CO + NH3 + CH4 (1:1:1), H2 O + NH3 + CH4 (1:1:1), and H2 O + CO + NH3 (1:1:1) at a temperature of 10 K. A tunable intense synchrotron radiation light source available at NSRRC, Hsinchu, Taiwan, was employed to provide the required EUV–VUV photons. The

104

C. Y. R. Wu et al.

photon wavelengths used to irradiate the icy samples were selected to center on the prominent solar lines, namely, the HeII 30.4 nm, the HeI 58.4 nm, and the HI 121.6 nm lines. We have previously investigated the identification of the chemical products produced through photolysis of several of the above ice mixtures.24 In this report, we focus on measurements of the destruction of the NH3 molecule in the various ice mixtures. The results obtained from the present study are important to our understanding of photon-induced stability of NH3 -containing ice analogs in cosmic objects. 2. Experimental Arrangements and Apparatus The experimental setup and experimental procedures employed in the present work have previously been described in detail.13,14 Briefly, the experimental setup consists of a synchrotron radiation source and monochromator system, a helium closed-cycle cryostat system (APD HC-DE204S) and a sample handling/preparation system, a Fourier-transform infrared (FTIR) spectrometer (Bomem DA8), and a PC-based data acquisition system. The EUV light source was provided by the High Flux Cylindrical Grating Monochromator (HF-CGM) beamline25 at the NSRRC. The synchrotron radiation was produced by a 1.5 GeV electron storage ring. In the present work the entrance and the exit slit widths were set to give a bandwidth of 0.4 nm at 58.4 nm and at 30.4 nm, and a bandwidth of 1.1 nm at 121.6 nm.10,13–15 The purity of the samples used in the present work is of the highest quality commercially available. Specifically, the ammonia gas was provided by Sigma-Aldrich with a stated purity of 99.5%. All gaseous samples were handled by the standard freeze–pump–thaw–pump–freeze–distill cycles as previously described.13,14,24 The ice samples were deposited onto a KBr substrate mounted on the cold finger, which was maintained at 10 K by the helium closed-cycle cryostat system. The thickness of the ice samples used in the present work is typically from 3 µm to 5 µm. The optical penetration depths14,26 of EUV photons used in the present work vary from 10−3 to 5 × 10−2 µm based on the well known absorption cross sections of relevant molecules in the gas phase.21 Thus, all incident photons are absorbed by the ice sample without impinging on the substrate. The absorption spectra of a given icy sample before and after EUV photon irradiation were obtained by the FTIR spectrometer using a globar source, a CsBr beam splitter, and a HgCdTe detector (cooled to the temperature of liquid nitrogen at 77 K). The spectral ranges employed in

Destruction Yields of NH3

105

the present work cover from 2.5 to 20 mm (i.e., from 4,000 to 500 cm−1 ). A typical resolution of 0.5 cm−1 was used in the present work.

3. Experimental Results and Discussion 3.1. The spectra of the difference of absorbances The IR absorbance spectra of a given icy sample before and after photolysis, a0 and ap , respectively, for a given dosage were obtained by a FTIR spectrometer. To enhance the changes of absorbances of the photon-induced chemical products we chose to present the spectra of the difference of absorbances, (ap − a0 ).13,14 In Fig. 1 we show typical absorbance spectra of pure NH3 ices taken before (a0 , the top panel) and after (ap , the middle panel) the 30.4 nm photolysis in the 5000 to 500 cm−1 region for an irradiation time of 180 min. The spectrum of the difference of absorbances, (ap − a0 ), is plotted in the bottom panel. A peak in the (ap − a0 ) spectrum often reflects the growth of photon-induced chemical products. A dip correlates with depletion of the parent ice molecules, which can be clearly seen by comparing the absorption features of NH3 shown in the upper, middle, and bottom panels. In other words, while new molecular species were formed, the original reactants were depleted due to their conversion to other species. The photolyzed products are indicated in the bottom panel of Fig. 1, showing the NH2 radical feature at 1505 cm−1 and the previously observed, but unidentified feature at 2115 cm−1 .27 We cannot positively observe the 886 cm−1 feature N2 H4 reported by Gerakines et al.27 This is possibly due to the fact that we have used difference light source, a monochromatic synchrotron radiation (a spectral bandwidth less than 1.1 nm13–15 ) vs. an undispersed output from a microwave-discharged H2 flow lamp.27 There are two broad features indicated by “?” mark because they do not correspond to any absorption features of possible products such 28–31 Further study to assign their identities as NH, N2 H4 , (NH3 )2 , or NH+ 4. is required. The absorption feature of the NH3 at 1065 cm−1 27,28,32 appears to be isolated from other features of NH3 , as well as the possible products. Therefore, we chose this feature for our measurements of the destruction yields. In Fig. 2 we show the portion of the spectra of the difference of absorbances between 3000 and 800 cm−1 after 30.4 nm photolysis for 5, 30, 60, 120, and 180 min. The magnitudes of the peaks and dips increase with increasing photon irradiation time, i.e., photon dosage.

106

C. Y. R. Wu et al.

Fig. 1. The top panel: absorbance spectrum of pure NH3 ices before photolysis. The middle panel: absorbance spectrum of pure NH3 ices after photolysis at 30.4 nm for 180 min. The bottom panel: the difference of absorbances of the NH3 ices before and after photolysis at 30.4 nm.

3.2. The destruction yields To quantitatively determine the dependence of the destruction yield as a function of photon dose we first measure the product column density, Np (molecule cm−2 ), which is obtained by dividing the integrated area of the dip of the difference of absorbances by its absorption band strength, A (cm per molecule). The relationship13,26 is given below:  [ (ap − a0 )dν] Np = 2.3 A

Destruction Yields of NH3

107 NH3Fig1b.org

0.02 0.01 0.00

After 30.4 nm Photolysis for 180 min.

-0.01 0.02

Difference of Absorbances

0.01 After 30.4 nm Photolysis for 120 min.

0.00 -0.01 0.02 0.01

After 30.4 nm Photolysis for 60 min.

0.00 -0.01 0.02 0.01 0.00

After 30.4 nm Photolysis for 30 min.

-0.01 0.02 0.01 0.00

Pure NH3 ices at 10 K After 30.4 nm Photol;ysisfor 5 min.

-0.01

3000

2500

2000

1500

1000

Wavenumber (cm-1)

Fig. 2. Spectra of the difference of absorbances of pure NH3 ices at 10 K after 30.4 nm photolysis for 5, 30, 60, 120, and 180 min.

The infrared band strength of NH3 at 1065 cm−1 is A(NH3 , 1065 cm−1 ) = 1.7 × 10−17 cm per molecule.32 The photon dose (photons cm−2 ), ∫ PUV dt, is obtained by dividing the integrated numbers of photons impinging on the ices by the photon beam size, which is 0.145 cm2 , at the ice sample. The destruction yield13,14 can be expressed as Y =

dNp  . d[ PUV dt]

The slope determined from data points that show a linear dependence on the photon dose gives the destruction yield per photon. A typical plot of the

108

C. Y. R. Wu et al.

Fig. 3. Plots of the destruction of the column density of NH3 in the CO + NH3 (4:1) mixed ices at 10 K as a function of photon dose. The results obtained are shown for photolysis at 30.4 nm (top panel), 58.4 nm (middle panel), and 121.6 nm (the bottom panel).

destruction column density (cm−2 ) as a function of photon dose (photons cm−2 ) is displayed in Fig. 3 for photolysis of CO + NH3 (4:1) mixed ices at the three selected photon wavelengths. The photon dose used in the present work has a dynamic range between 10 and 100. As can be seen from Fig. 3 and within the presently studied photon dose range, the product column density of NH3 appears to deviate from a linear decrease at a photon dosage of about 1 × 1017 photons cm−2 in the photolysis at 30.4 and 58.4 nm and at a photon dosage of about 8 × 1015 photons cm−2 in the photolysis at 121.6 nm. We have also carried out similar analyses for the destruction of NH3 in several ice systems and have summarized their results in Table 1. Brief discussions of the results for the ice systems studied are given below. In the case of pure NH3 ices the 1065 cm−1 feature is well separated from other absorption features. Since the destruction of absorbances has

Destruction Yields of NH3 Table 1. at 10 K.

109

Destruction yields of NH3 through photolysis of mixed ice systems

Ice system NH3 CO + NH3 (1:1) CO + NH3 (4:1) CH4 + NH3 (1:1) H2 O + CO + NH3 (1:1:1) CH4 + CO + NH3 (1:1:1) H2 O + CH4 + NH3 (1:1:1)

30.4 nm 1.13 1.06 0.53 0.90 1.29 1.57 0,83

(±0.48) (±0.14) (±0.07) (±0.16) (±0.18) (±0.18) (±0.09)

58.4 nm 0.88 1.56 0.60 0.51 1.17 0.99 0.53

(±0.23) (±0.09) (±0.16) (±0.09) (±0.25) (±0.12) (±0.12)

121.6 nm Not measured 2.76 (±1.38) 2.53 (±1.38) Not measured 0.69 (±0.12) 0.60 (±0.12) Not measured

been accurately measured the data can be fitted to a single exponential function. The half-life for the destruction of NH3 has thus been determined to be 3.9 × 1017 photons cm−2 and 2.6 × 1017 photons cm−2 for the photolysis at 30.4 and 58.4 nm, respectively, under the present experimental conditions. The half-lifetime of pure NH3 ice molecules can be estimated in the cosmic environments if the photon flux at the given photon wavelength is known. The EUV photon fluxes for the prebiotic environs and the current Solar EUV outputs can be very different. As an example, if we use an absolute solar flux of 4 × 1010 photons cm−2 sec−1 for photon wavelength from 5 to 50.4 nm33 of the current days, then the half-lifetimes will be about 107 years. We have examined the CO + NH3 ice system in two different compositions, namely, at a ratio of 1:1 and 4:1. In Fig. 3, we plot the difference of the column densities of the 1065 cm−1 feature produced through the photolysis of the CO+NH3 (4:1) mixed ices as a function of the photon dose. The destruction yields are 0.53, 0.60, and 2.53 for photolysis at 30.4, 58.4, and 121.6 nm, respectively. For the (1:1) ice mixtures the destruction yields are slightly larger than the corresponding values for the (4:1) ice sample, as summarized in Table 1. The error bars in the destruction yields represent the spread of values determined from different experimental runs. It is interesting to note that the destruction yield of NH3 by 121.6 nm photolysis is more than unity. This is possible since reactions of CO with the photolyzed products of NH3 and reactions between excited CO and NH3 molecules can occur in addition to the direct photodestruction of NH3 only. This assertion can be supported by a recent photodestruction study.4 Using a microwave-discharged H2 flow lamp with the spectral output peaking approximately at 162 nm, Cottin et al.4 have found that the destruction for CH4 , CH3 OH, NH3 , and HNCO is faster in a N2 ice matrix than in H2 O

110

C. Y. R. Wu et al.

ices. The photodestruction yields of NH3 in the CH4 + NH3 (1:1) mixed ices measured at 30.4 and 58.4 nm behave similarly as the above-mentioned binary systems. The destruction yields of NH3 in tertiary ice systems, namely, H2 O + CO + NH3 (1:1:1), CH4 + CO + NH3 (1:1:1), and H2 O + CH4 + NH3 (1:1:1) mixed ices have also been measured. In this study, we attempted to learn the ice photochemistry by replacing H2 O with CH4 in the H2 O + CO+ NH3 (1:1:1) ices. As can be seen from Table 1, the destruction yields of NH3 in the ice systems at 30.4 and 58.4 nm are larger than those at 121.6 nm. These results appear to correlate with the fact that the absorption cross section of CO ice at 121.6 nm is relatively small15,16 in comparison with that at 58.4 and 30.4 nm. However, the destruction yields produced through photolysis of the CH4 + CO + NH3 (1:1:1) ices at 58.4 and 121.6 nm are smaller than those in the H2 O + CO + NH3 (1:1:1) ices. This may suggest that H2 O is more reactive than CH4 in the destruction of NH3 in the presently studied mixed ice systems containing CO and NH3 ices. We can compare the results between CH4 +CO+NH3 (1:1:1) and H2 O+ CH4 + NH3 (1:1:1) ices and between H2 O + CO + NH3 (1:1:1) and H2 O + CH4 + NH3 (1:1:1) ices. In the former case H2 O was replaced by CO and in the latter case CH4 was replaced by CO, we find that the destruction yields at 30.4 and 58.4 nm are larger in the mixed ice containing CO suggesting that CO is more reactive in helping destruction of NH3 in these ice systems through photolysis at these two EUV energies. The ice photochemistry is obviously complicated. We plan to carry out further detailed analyses at more photon energies in the near future in order to elucidate and understand the important photolysis and reaction mechanisms.

4. Concluding Remarks The present work shows that the destruction yields of NH3 in several photolyzed mixed ices clearly depend on (1) the ice compositions and (2) the photon excitation energies. We conclude that EUV–VUV photodestructions of NH3 in the presently studied ice mixtures are very efficient photochemical processes with yields higher than 0.5. We have previously identified various photolyzed products in these mixed ice systems, for example, the CN-containing species such as OCN− , HCN, and CH2 N2 , the light hydrocarbons such as the C2 H4 , C2 H6 , and C3 H8 , as well as the HCO, H2 CO, CH3 OH, and HCCO radical products.24 The quantitative analysis of

Destruction Yields of NH3

111

the product yields for the above-mentioned species are currently in progress in our laboratory. We see that in photolysis there are many important processes, such as (1) photodestruction of parent molecules, (2) photo-induced products, (3) subsequent chemical reactions among the photolyzed products and parent molecules, (4) subsequent photolysis of the photolyzed products and reaction products, and (5) depletion and mantle effect of the ice sample. In this preliminary work, we have just begun to explore the physics and chemistry of cosmic ices. Further study to unravel the complex processes is clearly needed. In our previously investigated photo-induced chemical products13–15,23,24 , the EUV photolysis apparently produces more chemical products and with higher production yields than the 121.6 nm counterpart in the molecular ice systems studied. In the present study of destructions of the parent NH3 ice molecule, we find that the VUV photodestruction process can be more efficient than the EUV counterpart. This may be in part due to a large absorbance of NH3 ice in the 121.6 nm region,17 which is expected to be larger than those at 30.4 and 58.4 nm based on absorption cross section data of the gaseous NH3 .21 Our ongoing research effort will further improve our understanding of the correlation of electronic states with the destruction yields of important ice systems, as well as photoninduced chemical reaction products and decay pathways in cosmic ices. The results are important to our understanding of EUV–UV photon-induced ice chemistry in cosmic ice analogs.

Acknowledgments We are grateful for the support of the staff of the Synchrotron Research Radiation Center, Hsinchu, Taiwan. We appreciate valuable suggestions by the referees. This research is based on work supported by the NSF Planetary Atmospheres Program under Grant AST-0604455 (Wu) and the Astrophysics and Astrochemistry program of SRRC (Cheng).

References 1. R. E. Johnson, Energetic Charged-Particle Interactions with Atmospheres and Surfaces (Springer-Verlag, Berlin, 1990). 2. E. Dartois and L. B. d’Hendecourt, Astron. Astrophys. 365 (2001) 144. 3. W. A. Schutte and R. K. Khanna, Astron. Astrophys. 398 (2003) 1049.

112

C. Y. R. Wu et al.

4. H. Cottin, M. H. Moore and Y. Benilan, Astrophys. J. 590 (2003) 874. 5. P. Ehrenfreund, M. Bernstein, S. A. Sandford and L. J. Allamandola, Astrophys. J. 550 (2001) L95. 6. G. A. Baratta, G. Leto and M. E. Palumbo, Astron. Astrophys. 384 (2002) 343. 7. J. A. R. Samson, Techniques of Ultraviolet Spectroscopy (Pied Publications, Lincoln, NE, 1980). 8. D. Davis and W. Braun, Appl. Optics 7 (1968) 2071. 9. M. S. Westley, R. A. Baragiola, R. E. Johnson and G. A. Baratta, Planet. Space Sci. 43 (1995) 1311. 10. C.Y. R. Wu, D. L. Judge, Y.-J. Chen and T.-S. Yih, VUV photolysis of CO ices at 10 K — a detailed study employing different light sources, Presented in the 38th Annual Meeting of the Division for Planetary Sciences of the American Astronomical Society (Pasadena Convention Center, Pasadena, CA, October 8–13, 2006). 11. P. Jenniskens, G. A. Baratta, A. Kouchi, M.-S. deGroot, J. M. Greenberg and G. Strazzulla, Astron. Astrophys. 273(2) (1993) 583. 12. J. B. Holberg, Astrophys. J. 311 (1986) 969. 13. C. Y. R. Wu, D. L. Judge, B.-M. Cheng, C. S. Lee, T. S. Yih and W. H. Ip, J. Geophys. Res. 108(E4) (2003) 5032. 14. C. Y. R. Wu, D. L. Judge, B.-M. Cheng, W.-H. Shih, T.-S. Yih and W. H. Ip, Icarus 156 (2002) 456. 15. C. Y. R. Wu, H.-K. Chen, H.-C. Lu, B.-M. Cheng and D. L. Judge, EUV– VUV Photolysis of Pure CO Molecular Ice Systems at 10 K, submitted to J. Chem. Phys. for publication (2007). 16. H.-C. Lu, H.-K. Chen, B.-M. Cheng, Y.-P. Kuo and J. F. Ogilvie, J. Phys. B 38 (2005) 3693. 17. B.-M. Cheng, unpublished results (2005). 18. C. Romanzin, M.-C. Gazeau, Y. B´enilan, E. H´ebrard, A. Jolly, F. Raulin, S. Boy´e-P´eronne, S. Douin and D. Gauyacq, Adv. Space Res. 36 (2005) 258–267 and references therein. 19. C. Y. R. Wu and D. L. Judge, J. Chem. Phys. 75 (1981) 172. 20. C. Y. R. Wu, J. Chem. Phys. 86 (1987) 5584. 21. J. W. Gallagher, C. E. Brion, J. A. R. Samson and P. W. Langhoff, J. Phys. Chem. Ref. Data 17 (1988) 9. 22. P. A. Gerakines and M. H. Moore, Icarus 154 (2001) 372. 23. C. Y. R. Wu and B.-M. Cheng, NSRRC Activity Report, 2003/2004 (2004) 3–5. 24. C. Y. R. Wu, D. L. Judge and B.-M. Cheng, EUV–VUV photolysis of molecular ice systems of astronomical interest, Proceedings of the NASA Laboratory Astrophysics Workshop (University of Nevada, Las Vegas, NV, 14–16 February, 2006), NASA/CP-2006-214549 p. 284. 25. Y.-F. Song, C.-I. Ma, T.-F. Hsieh, L.-R. Huang, S.-C. Chung, N.-F. Cheng, G.Y. Hsiung, D.-J. Wang, C.T. Chen and K.-L. Tsang, Nucl. Instr. Meth. Phys. Res. A 467–468 (2001) 569. 26. P. A. Gerakines, M. H. Moore and R. L. Hudson, Astron. Astrophys. 357 (2000) 793.

Destruction Yields of NH3

113

27. P. A. Gerakines, W.A. Schutte and P. Ehrenfreund, Astron. Astrophys. 312 (1996) 289. 28. M. E. Jacox, D. E. Millikan, N. G. Moll and W. E. Thompson, J. Chem. Phys. 43 (1965) 3734. 29. S. Suzer and L. Andrews, J. Chem. Phys. 87 (1987) 5131. 30. R. L. Hudson, M. H. Moore and P. A. Gerakines, Astrophys. J. 550 (2001) 1140. 31. K. Rosengren and G. C. Pimentel, J. Chem. Phys. 43 (1965) 507. 32. L. B. d’Hendecourt and L. J. Allamandola, Astron. Astrophys. Suppl. Ser. 64 (1986) 453. 33. H. S. Ogawa, E. Phillips and D. L. Judge, J. Geophys. Res. 102A6 (1997) 11557.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


FORMATION OF CaTiO3 CRYSTALLINE DUST IN LABORATORY KAORI YOKOYAMA∗ , YUKI KIMURA, OSAMU KIDO, MAMI KURUMADA, AKIHITO KUMAMOTO and CHIHIRO KAITO Department of Physics, Ritsumeikan University, 1-1-1 Nojihigashi Kusatsu City, Shiga 525-8577, Japan ∗[email protected]

The crystalline grain formation of perovskite (CaTiO3 ) by the coalescence due to grain–grain collisions between TiO2 and CaO grains in smoke has been demonstrated. Spherical grains with diameters of 100–200 nm were produced. A large quantity of perovskite grains were also produced by the condensation from CaTiO3 vapor evaporated in selecting Ar gas pressure of 10 Torr. These grains contained the WO3 crystal in the center by the peritectic reaction. The perovskite grains showed significant peaks at 14.4 and 21.9 µm in the optical spectra.

1. Introduction The discovery of the existence of the crystalline silicate grains in space by the infrared (IR) space observatory1 has changed views on dust formation. We demonstrated, through laboratory experiments, that crystalline grains can be produced by the coalescence between Mg and SiO smoke grains2 or MgO and SiO2 smoke grains.3 In the present study, it was shown that crystal grains can be produced by the coalescence between CaO and TiO2 grains. Through the present laboratory experiments, it became evident that collisions between different grains can produce the crystalline grains. In the equilibrium condensation theory in our solar system, perovskite (CaTiO3 ) condenses at 1647 K and disappears at 1393 K4 after the condensation of corundum (Al2 O3 ). However, the discovery of perovskite in meteorites5 also suggests that the nonequilibrium condensation occurs as well as corundum. Therefore, this fact greatly interest the formative condition of CaTiO3 dust and its spectra in the laboratory. The thermochemical condition of condensation in late-type M dwarfs is fulfilled due to low temperature and high density.6 The high temperature 115

116

K. Yokoyama et al.

condensates that are the most abundant in the atmospheres of late-type M dwarfs and brown dwarfs with smaller ratio of C/O than unity are ZrO2 for T < 2000 K and Al2 O3 for T < 1800 K. Other stable species to appear at T < 1600 K are Ca2 Al2 SiO7 , Ca2 MgSi2 O7 , and CaMgSi2 O6 , as well as Ti4 O7 and Ti2 O3 . These grains compete with the formation of perovskite and corundum. Ti oxides (particularly CaTiO3 ) are the condensations at high temperature. TiO2 and CaTiO3 are likely to be among the first nuclei to form in brown dwarf atmospheres. The form of Ti oxides depends on the brown dwarf’s effective temperature. For effective temperatures above 2000 K, CaTiO3 is produced, whereas for lower temperatures, Ti3 O5 is produced. Therefore, CaTiO3 is important for the formation of dust around brown dwarfs.7 Therefore, the condensation of perovskite is the principal cause of TiO depletion in the atmosphere of dwarfs later than about M6.8,9 However, CaTiO3 dust is hardly ever produced under laboratory conditions. In the present experiment, CaTiO3 grains were first produced by the same method as that for producing Mg2 SiO4 in which the mixed smoke in the laboratory was due to the demonstration of solid–solid reaction. The massive production of CaTiO3 grains have been succeeded with the order of 50 nm by selecting the gas pressure, and the optical spectra at 14.4 and 21.9 µm can be identified as the perovskite.

2. Experimental The experimental set up for producing TiO2 and CaO grains and for their coalescence was similar to that used in a previous study.3 The evacuation chamber was a glass cylinder 170 mm in diameter and 300 mm in height, covered with a stainless-steel plate at the top and connected to a highvacuum exhaust through a valve at its bottom. After evacuation of the chamber, smoke was produced in a mixture of Ar and O2 gases at 80 Torr (Ar 79 Torr, O2 1 Torr). A V-shaped tantalum boat (50 mm in length, 2 mm in width, and 1 mm in depth) and a conical tantalum basket (10 mm in diameter) were used for the evaporation of Ti and Ca in the mixture gas, respectively, as shown schematically in Fig. 1. The collected samples were observed using Hitachi H-7100R and H-9000NAR transmission electron microscopes (TEMs). The composition of the grains was determined with an energy-dispersive X-ray spectrometer (EDX/Horiba EMAX-5370) attached to the H-7100R TEM. The transmission IR spectra of the collected samples embedded in KBr pellets were measured with a Fourier transform infrared

Formation of CaTiO3 Crystalline Dust in Laboratory

117

Fig. 1. Schematic diagram of apparatuses for producing CaTiO3 by mixing TiO2 and CaO smokes in Ar (79 Torr) and O2 (1 Torr) gases. The boat and the conical basket made of tantalum were used for the evaporation of Ti and Ca in the mixed gas. The grains produced were collected from these regions [(a), (b), and (c)] with the glass plate. The distance between the boat and basket was 5 mm. The two types of smokes were mixed between the heaters.

spectrometer (FTIR/Horiba FT210) in the wavelengths region from 2.5 to 25 µm.

3. Results and Discussion 3.1. Demonstration of CaTiO3 crystalline grain formation by the coalescence of TiO2 and CaO grains TEM images of the grains produced in TiO2 , CaO, and their mixed smoke are shown in Fig. 2. Spherical TiO2 grains with sizes smaller than 100 nm were produced in the TiO2 smoke.10 Electron diffraction patterns show the formation of brookite and anatase. In the CaO smoke, although CaO grains were predominately produced, the ED pattern also indicates the formation of Ca(OH)2 , as shown in Fig. 2c. The CaO grains produced are easily changing to Ca(OH)2 from their surface by exposure to air.11 As a result, diffuse ED pattern of Ca(OH)2 has appeared, as shown in Fig. 2c. CaTiO3 grains were produced in the mixed smoke by the coalescence of TiO2 and CaO grains. TiO2 grains of the rutile phase, which is a high temperature phase of TiO2 were produced, as shown in Fig. 2b. The formation of the high-temperature phase of TiO2 in the mixed smoke was due to higher temperature of the mixed smoke region.12 Figure 3 shows a TEM image of typical grains collected from the mixed smoke region. EDX-based spectroscopy was used to search for typical grains.

118

K. Yokoyama et al.

Fig. 2. TEM images and corresponding ED patterns of typical grains collected at (a), (b), and (c) indicated in Fig. 1.

Formation of CaTiO3 Crystalline Dust in Laboratory

119

Fig. 3. Result of identification of mixed TiO2 , CaO, and CaTiO3 grains. Grains were discriminated by EDX with a focused electron beam. Spherical grains were CaTiO3 with sizes on the order of 100–200 nm. The spherical grains smaller than 100 nm were TiO 2 . The cubic and rod-shaped grains were CaO and Ca(OH) 2 .

It can be concluded that CaTiO3 grains had sizes of 100–200 nm order. The grains with sizes smaller than 100 nm were TiO2 . The coagulated grains smaller than 50 nm with cubic and rod shapes were identified as CaO or Ca(OH)2 . Figure 4 shows the electron diffraction (ED) pattern of the grain indicated by an arrow. The ED pattern can be identified as that of a CaTiO3

Fig. 4. ED pattern of a grain indicated by an arrow. The diffraction spots were attributed to the CaTiO3 single-crystalline grain.

120

K. Yokoyama et al.

single-crystalline grain seen along its [010] zone axis. Therefore, it can be concluded that CaTiO3 (orthorhombic, a0 = 0.544 nm, b0 = 0.764 nm, c0 = 0.538 nm) single-crystalline grains were produced by the coalescence of TiO2 and CaO smoke grains. If a TiO2 grain and a CaO grain with the size of 100 nm were coalesced, the size of a CaTiO3 grain was estimated to be 133.6 nm. Therefore, the large spherical grains were CaTiO3 which is in accordance with the TEM image in Fig. 3. The CaTiO3 crystalline grains can be produced by the coalescence of CaO and TiO2 crystalline grains.

3.2. Production of numerous CaTiO3 crystalline grains and their infrared spectra CaTiO3 grains were produced by the coalescence between TiO2 and CaO grains, as indicated in Sec. 3.1. Because CaTiO3 has a high melting point and is a refractory material, it is generally decomposed by direct evaporation. Coalescence-based growth of grains predominately took place in the smoke, and the results described in Sec. 3.1 suggest the formation of CaTiO3 from solid–solid grain reaction in Ar gas. As is shown by the formation of iron-oxide grains in a similar experiment, the selection of an appropriate gas pressure, which governs the coalescence frequency, can produce the evaporated-material grains in spite of the decomposition phenomenon.13 In the present experiment, the gas pressure of 10 Torr enabled successful the formation of CaTiO3 dust as follows. The evaporation was performed using a V-shaped tungsten boat (50 mm in length, 2 mm in width, and 1 mm in depth) in Ar gas at a pressure of 10 Torr. Figure 5 shows typical smoke samples obtained by the direct evaporation of commercial CaTiO3 powder in Ar gas. Spherical grains with sizes smaller than 50 nm were predominantly produced. The ED pattern clearly indicated the formation of CaTiO3 grains. The ED pattern also indicated the formation of WO3 crystal. The TEM image in Fig. 5 shows the small black dots in CaTiO3 . These dots inside the CaTiO3 grains were WO3 , as indicated in Figs. 6 and 7. Figure 6 shows bright- and dark-field images and corresponding ED patterns of a single grain containing a black dot. The indexed ED pattern indicated both CaTiO3 with a [101] zone axis and WO3 with a [001] zone axis. The two crystals have definite lattice relations with [10¯ 1]CaTiO3 //[001]WO3 and (020)CaTiO3 //(200)WO3 . Therefore, drops of mixed CaTiO3 –WO3 liquid were crystallized from

Formation of CaTiO3 Crystalline Dust in Laboratory

Fig. 5.

121

Grains produced by direct evaporation of commercial CaTiO3 powder in Ar gas.

Fig. 6. (a) Bright-field image, (b) dark-field image, and (c), (d) corresponding diffraction patterns. Strong spots were identified as CaTiO 3 with a [101] zone axis, as shown in (c). Weak spots among CaTiO3 spot were attributed to WO3 , as shown in (d).

the surface accompanying the peritectic reaction. The high-resolution transmission electron microscopy (HRTEM) images shown in Fig. 7 clearly indicate that CaTiO3 with an orthorhombic phase was produced. The WO3 crystallites were derived from the evaporation source. Figure 8 shows IR spectra of grains of CaO, TiO2 , WO3 , and CaTiO3 grains containing WO3 and commercial CaTiO3 powder. The grains of CaO, TiO2 , and WO3 were also produced in our laboratory. Their grain sizes were less than 200 nm and their spectra are clearly different. Our peritectic grains showed a spectrum similar to that of commercial CaTiO3 powder. The 14.1

122

K. Yokoyama et al.

Fig. 7. HRTEM image of a grain. The surrounding region was identified as the CaTiO 3 structure. The magnified lattice image clearly shows the existence of the orthorhombic CaTiO3 structure.

Fig. 8. Mid-IR transmission spectra of grains of TiO2 , CaO, and CaTiO3 containing WO3 and commercial CaTiO3 powder are shown. The features at positions of 14.4 and 21.9 µm are caused by CaTiO3 .

and 21.0 µm features were observed in the work of Posch et al.7 Therefore, the present 14.4 and 21.9 µm features are concluded to be due to CaTiO3 . The full width at half maximums (FWHMs) of 14 and 21 µm features of CaTiO3 in the present specimen in Fig. 8 were compared with Posch

Formation of CaTiO3 Crystalline Dust in Laboratory

123

data.7 The FWHMs of the present peaks at 14 and 21 µm were 4.0 and 1.9 µm, respectively. These values are larger than Posch et al.’s data of 3.2 and 1.8 µm, respectively. The difference of grain sizes of the present specimen and Posch et al.’s sample may account for the differences in the FWHM. The stretched vibration at approximately 14.4 µm becomes broad because the TiO6 octahedron in the CaTiO3 structure is distorted. The broadness due to the distorted octahedron has been demonstrated by the WO3 grains.14 The feature at 21.9 µm corresponds to the TiO6 bending vibration. The spectrum differences between the commercial powder and the present specimen were due to the shape and size effects; commercial CaTiO3 powder grains have irregular shape and a mean size of 10 µm. On the other hand, the present nano grains were uniformly spherical with the mean size of 50 nm. The differences between Rayleigh and Mie scattering may exist between the spectra of CaTiO3 in Fig. 8.

References 1. L. B. F. M. Waters, F. J. Molster and T. deJong, Astron. Astrophys. 315 (1996) L361. 2. C. Kaito, Y. Ojima, K. Kamitsuji, O. Kido, Y. Kimura, H. Suzuki, T. Sato, T. Tanaka, Y. Sito and C. Koike, M & PS 38 (2003) 49. 3. K. Kamitsuji, H. Suzuki, Y. Kimura, T. Sato, Y. Saito and C. Kaito, Meteorit Planet. Sci. 429 (2005) 205. 4. L. Grossman, Geochim. Cosmochim. Acta 36 (1972) 597. 5. A. N. Knot, T. J. Fagan, K. Keil, K. D. McKeegan, S. Sahijpal, I. D. Hutcheon, M. I. Petaev and H. Yurimoto, Geochim. Cosmochim. Acta 68(9) (2004) 2167. 6. T. Tsuji, K. Ohnaka and W. Aoki, Astron. Astrophys. 305 (1996) L1. 7. Th. Posch, F. Kerschbaum, D. Fabian, H. Mutschke, J. Dorschner, A. Tamanai and Th. Henning, Astrophys. J. Suppl. 149 (2003) 437. 8. D. R. Alexander, F. Allard, A. Tamanai and P. H. Hauschildt, Astrophys. Space Sci. 251 (1997) 171. 9. K. Lodders, Astrophys. J. 577 (2002) 974. 10. Y. Atou, H. Suzuki, Y. Kimura, T. Sato, T. Tanigaki, Y. Saito and C. Kaito, Physica E 16 (2003) 179. 11. Y. Kimura and J. A. Nuth, Astrophys. J. 630 (2005) 637. 12. C. Kaito and M. Shiojiri, JJAP 21 (1982) L421. 13. C. Kaito, T. Watanabe, K. Ohtsuka and Y. Saito, Proc. NIPR Symp. Antarct. Meteor. 5 (1992) 310. 14. M. Kurumada, O. Kido, T. Sato, H. Suzuki, Y. Kimura, K. Kamitsuji, Y. Saito and C. Kaito, J. Cryst. Growth 275 (2005) 1673.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


DIRECT OBSERVATION OF THE CRYSTALLIZATION OF CARBON-COATED AMORPHOUS Mg-BEARING SILICATE GRAINS CHIHIRO KAITO, SHINICHI SASAKI, YU MIYAZAKI∗ , AKIHITO KUMAMOTO† , MAMI KURUMADA, KAORI YOKOYAMA, MIDORI SAITO and YUKI KIMURA Department of Physics, Ritsumeikan University, Noji-higashi 1-1-1 Kusatsu, Shiga 525-8577, Japan ∗[email protected] †[email protected] HITOSHI SUZUKI Department of Electric Engineering, Tohoku Gakuin University, Chuoh 1-13-1 Tagajo, Miyagi 985-8537, Japan

The crystallization of amorphous Mg-bearing silicate grains into Mg2 SiO4 crystal covered with a thin carbon layer was directly observed by in-situ transmission electron microscopy. The temperature of crystallization of the sample was observed to be 200◦ C lower than that of the sample without the carbon layer. The graphitization energy of the surface amorphous carbon layer with the mean thickness of 10 nm accelerated the crystallization of the central amorphous Mg-bearing silicate grain of 100 nm order. Sample results for crystallization at room temperature are presented.

1. Introduction In a previous work, we demonstrated that the crystallization of amorphous Mg-bearing silicate grains to Mg2 SiO4 crystals takes place at 800◦C under vacuum.1 The crystallization started from the grain surface. We found that prenucleation occurs in the 650–800◦ C temperature range before crystallization at 800◦ C. The phenomenon in which the prenucleation state corresponded to the stall state was suggested and clarified by Hallenbeck and Nuth using IR spectroscopy.2 Yamamoto and Chigai have proposed a chemical heating model to explain the crystallization mechanism of cometary silicate, and have shown that the chemical heating mechanism leads to the degree of crystallization required to explain the observed strength of the cometary crystalline features.3 125

126

C. Kaito et al.

To elucidate this model using laboratory analogies, amorphous Mgbearing silicate grains crystallizing at 800◦ C were covered with an amorphous carbon layer. Since the amorphous carbon layer on the surface of CdTe crystal nanoparticles crystallized into a graphene layer after being heated to 500◦ C,4 the effect of the graphene structure alteration on the surface of the central Mg-bearing silicate grains has been examined by insitu transmission electron microscopy. On the other hand, the same Mgbearing silicate grains covered with a carbon layer produced in a methane atmosphere clearly showed graphitization at temperatures as low as room temperature. The results for these samples are presented in this paper.

2. Experimental Methods Amorphous Mg-bearing silicate grains used were produced by the coalescence between MgO and SiO2 smoke grains.5 The grains dispersed on glass plates were covered with an amorphous carbon layer by depositing carbon under vacuum using the arc discharge method.6 Electron diffraction (ED) patterns clearly showed that the produced samples were amorphous. The metamorphism of the carbon-coated grains was directly observed using a Hitachi H-9000NAR electron microscope with a special heating holder, which can be heated up to 1500◦C.7

3. Results and Discussion The carbon layer was deposited on the Mg-bearing silicate grains by the arc discharge method in the direction indicated by arrows in Fig. 1. Thus, the thickness of the carbon layer deposited on the grains was not uniform (Fig. 1). As shown in the bottom left inset of Fig. 1, the opposite surface of the grain was also covered with a thin carbon layer. This indicates that carbon atoms diffused efficiently on the silicate grains. The ED pattern of the sample clearly shows amorphous halo rings (up right corner inset of Fig. 1). The temperature of the specimen was increased by 50◦ C steps from room temperature. When the temperature of the sample reached equilibrium, its movement was reduced. Selected temperature for each step was maintained for 15 min. When the phase of the specimen changed, the selected temperature was maintained for more than 15 min. As shown in Fig. 2, the halo diffraction pattern based on the graphitic structure given in Fig. 1 became sharper after heating to 550◦ C. The substrate differences

Crystallization of Amorphous Mg-bearing Silicate Grains

127

Fig. 1. TEM image and ED pattern (upper right corner inset) of Mg-bearing silicate particle covered with carbon layer.

Fig. 2. The sample heated at 550◦ C was hardly altered. The halo ED pattern (upper right corner inset) corresponding to carbon became sharper than the pattern before heating, indicated that graphitization started.

128

C. Kaito et al.

Fig. 3. Crystallization occurred at 600◦ C. The ED pattern can be assigned to the Mg2 SiO4 crystallites. Amorphous grains were crystallized at a lower temperature than the amorphous silicate grains without a carbon layer. The crystallization of Mg-bearing silicate grains without a carbon layer occurred at 800◦ C as elucidated in the previous paper.

of the CdTe crystal and the amorphous silicate may affect the difference of the graphitization temperature. After heating to 600◦ C, the crystallization of the central amorphous Mg-bearing silicate occurred as shown in Fig. 3. The ED pattern can be indexed to correspond to Mg2 SiO4 . In a previous work, the prenucleation state was directly observed at 650◦ C. With the present sample, the prenucleation state could hardly be observed. The crystallization occurred with the graphitization of the amorphous carbon layer at a lower temperature. A typical example of image of a graphitized carbon layer is shown in Fig. 4. A fast Fourier transform (FFT) at one part of the high-resolution transmission electron microscopic (HRTEM) image clearly showed the graphitization of the surface carbon layer as depicted by the bottom left corner inset of Fig. 4. The crystallization heat, which is usually interpreted as occurring at the transformation from the metastable amorphous phase into stable crystalline phase, is generated when the dangling bonds are connected to form stable bonds or are reconnected to form an atomic arrangement as is the crystal. The temperature of the interface is thereby raised and the crystallization may be accelerated. The crystallization heat raises the temperature of the adjacent microcrystallites of graphite structure and effectively promotes the graphitization on the covered layer. In the graphite crystal, the thermal conductivity is 3.1 times

Crystallization of Amorphous Mg-bearing Silicate Grains

129

Fig. 4. HRTEM image of a graphitized carbon layer at 600◦ C. The released graphitization energy excess accelerates the crystallization of Mg-bearing silicate grains. The FFT image of a part of the graphite surface (bottom left corner inset) illustrates the graphitization.

larger in the perpendicular direction of the c-axis than in the direction normal to the c-axis. The carbon layer arranged graphitic layer parallel to the amorphous grain surface as seen in Fig. 4 as indicated by circles. Since the phonon mean free path is proportional to the crystallite size, a part of the energy excess due to the graphitization dispersed to the central amorphous grains. The volume of amorphous carbon layer with mean thickness of 10 nm on the 100 nm amorphous silicate grain was about 0.7 times of the volume of silicate grain. So, the alteration of the amorphous carbon layer accelerated the crystallization on the amorphous grains. Therefore, it is supposed that the energy excess due to graphitization was transferred to the central amorphous Mg-bearing silicate grain. The observed decrement of the crystallization temperature supported the theory proposed by Yamamoto and Chigai on the crystallization of amorphous silicate induced by the chemical reaction energy on the surface layer. To achieve a lower temperature crystallization for amorphous silicate grains, the same Mg-bearing silicate grains were covered with a carbon layer produced by the arc discharge of carbon in a CH4 atmosphere of 10−3 Torr. The carbon layer containing CH4 was graphitized at room temperature

130

C. Kaito et al.

Fig. 5. An amorphous Mg-bearing silicate grain (B) was covered by a carbon layer (A) produced by the arc discharge of carbon in a CH4 atmosphere of 10−3 Torr. Image (a) was seen just 30 min after the grains were produced. Image (b) was observed after 4200 min in air. The white-ring contrast between (A) and (B) is due to the change in density on the deposited carbon layer caused by graphitization. The surface of (B) also changed to ring contrasts at room temperature as indicated by the arrows.

Fig. 6. HRTEM image of part of amorphous silicate surface. Crystal Mg2 SiO4 formation was clearly observed. The graphitization of the carbon layer also occurred in the spots indicated by an arrow.

Crystallization of Amorphous Mg-bearing Silicate Grains

131

by keeping in air, and the partial crystallization of the amorphous Mgbearing silicate occurred at room temperature as shown in Fig. 5. The same positions as previously selected were observed by the special specimen holder. The as-prepared specimen was exposed in air and immediately observed by electron microscope (Fig. 5a). The variation of density due to the crystallization proceeded from the surface to the central region. The circular contrast layer in the carbon layer and the amorphous Mgbearing silicate can be clearly seen. An HRTEM image of the Mg2 SiO4 crystal is shown in Fig. 6. In addition to the graphitization indicated by an arrow, Mg2 SiO4 crystals with a size of 3 nm are seen. The crystallization in Fig. 3 and previous paper took place suddenly from one or some crystallite from the surface of the amorphous silicate grains. In Fig. 5, the crystallization took place concentrically. Detailed experimental results including the spectral alteration of this sample at room temperature will be published elsewhere.

4. Summary Mg-bearing silicate grains crystallized at 600◦C with a carbon layer without a prenucleation stage indicated that the temperature of crystallization decreased due to the energy excess of graphitization. The energy brought to the samples by heating helped significantly the crystallization process to occur. The amorphous Mg-bearing silicate grains covered with the carbon layer containing CH4 crystallized at room temperature in agreement with Yamamoto and Chigai theory.

References 1. K. Kamitsuji, T. Sato, H. Suzuki and C. Kaito, Astron. Astrophys. 436 (2005) 165. 2. S. L. Hallenbeck and J. A. Nuth, Astrophys. Space Sci. 255 (1998) 427. 3. T. Yamamoto and T. Chigai, Highlights Astron. 13 (2005) 522. 4. T. Tanigaki, Y. Kimura, H. Suzuki and C. Kaito, J. Cryst. Growth 260 (2004) 298. 5. K. Kamitsuji, H. Suzuki, Y. Kimura, T. Sato, Y. Saito and C. Kaito, Astron. Astrophys. 429 (2005) 205. 6. Y. Saito, C. Kaito, T. Sakamoto, S. Kimura, Y. Nakayama and C. Koike, Planet. Space Sci. 43 (1995) 1303. 7. S. Kimura, C. Kaito and S. Wada, Antarc. Meteor. Res. 13 (2000) 145.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


RELATIONSHIP BETWEEN MORPHOLOGY AND SPECTRA REVEALED BY DIFFERENCE IN MAGNESIUM CONTENT OF SPINEL PARTICLES MIDORI SAITO∗ , MAMI KURUMADA and CHIHIRO KAITO Department of Physics, Ritsumeikan University, 1-1-1 Nojihigashi Kusatsu-shi, 525-8577 Shiga, Japan ∗[email protected]

Spinel particles containing Mg of different amounts were obtained by flash gas evaporation from different mixture powders of Mg and Al in a mixture gas of argon and oxygen. By decreasing Mg content, the shape of the particles was changed from cubic, octahedral, elliptical to spherical. The characteristic spectra among these produced particles are indicated. The difference between the observed peak positions and the calculated absorption positions was discussed in terms of the effect of the spinel phase on shape.

1. Introduction Spinel is one of the major oxides found in meteorites or chondrites, such as CM2 meteorite Murray or CI chondrite Orgueil.1,2 It condenses at 1240◦C from a cooling gas of cosmic composition at thermodynamic equilibrium and disappears at 1089◦C due to the formation of anorthite (CaAl2 Si2 O8 ).3 However, spinel was not expected to appear in nonequilibrium process theory. Spinel formation due to solid–solid reaction was proposed.4 One of the present authors demonstrated the formation of spinel by MgO solid– Al2 O3 solid reaction using nanosized particles.5 On the other hand, the 12–13 µm emission feature was found in the spectra of oxygen-rich AGB stars by ISO spectra observation. The origin of this feature was identified as γ-alumina particles6 or spinel.7,8 Recently, we have directly produced the δ-alumina grains from the vapor Al phase in a mixture gas of argon and oxygen.9 Although γ-alumina is considered as the carrier of the 13 µm feature, the δ-alumina grains also have a strong absorption at 13 µm. Furthermore, the C/O ratio of AGB stars is closely related to the 13 µm feature; that is, in the case of a C/O ratio close to unity, almost all the oxygen will be consumed by alumina grains.10 Hron et al.11 133

134

M. Saito, M. Kurumada and C. Kaito

also found that the 13 µm feature is observed in stars within a rather narrow range of photosphere and dust shell temperatures. The spinel is also considered as the carrier of the 13, 17 and 32 µm emission features observed in the ISO spectra of oxygen-rich stars.8,12 Fabian et al. calculated the absorption efficiencies of sub-µm-sized spherical particles from the spectra of synthetic spinels, and showed the 13, 17, and 32 µm features.8 In the present experiment, we produced nm-sized spinel particles by flash gas evaporation. Morphological and spectral alterations induced by varying the content of Mg are elucidated. The effect of production condition or size on the spectrum of spinel is also discussed on the basis of the 14 and 18 µm features in their spectra.

2. Experimental Procedure The work chamber used was a glass cylinder of 170 mm diameter and 330 mm height. After evacuating the chamber to approximately 10−5 Torr, argon gas of 45 Torr and oxygen gas of 5 Torr were introduced into the chamber. Flash evaporation in the mixture gas was used as shown in Fig. 1. The mixture powder of Al and Mg was dropped onto the Ta boat which was heated above 1800◦C, and then the produced nanoparticles can be seen in the smoke owing to the scattering of radiation light from the heater. In this smoke, the nanoparticles were grown by coalescence.13 The mixing ratio of

Fig. 1.

Schematic representation of flash gas evaporation method.

Relationship between Morphology and Spectra

135

dropped powder was altered, such as Mg:Al = 1:1, 1:2 or 1:3, with weight ratio. The produced particles were collected on a glass plate, dispersed in ethyl alcohol, and mounted on an amorphous carbon film supported by standard copper electron microscope grids. A transmission electron microscope (TEM) equipped with an energy-dispersion X-ray (EDX) analysis system (Horiba EMAX-5770) was used to observe the morphology and analyze the structure of the particles. High-resolution transmission electron microscopy (HRTEM) was carried out using a Hitachi H-9000NAR electron microscope. The produced particles were also buried in KBr pellets and their transmittances were measured with a Fourier transform infrared (IR) spectrometer (Horiba FT210) from 2.5 to 25.0 µm. The wavelength resolution was 2 cm−1 . The beam splitter was a Ge-evaporated KBr substrate and the detector was deuterium triglycine sulfate.

3. Results and Discussion 3.1. Structure of produced particles The TEM images and corresponding electron diffraction (ED) patterns of typical particles produced by varying mixture ratio are shown in Fig. 2. The morphology of the markedly altered particles depends on the mass ratios between Mg and Al. Because the atomic masses of Mg and Al, which are 24.3 and 27.0, respectively, are close, the mass ratio almost corresponds to the atomic ratio. The mixed metallic powder was flashed in the atmosphere of mixed gases of argon and oxygen; the produced particles were oxides. Therefore, the evaporated Mg and Al vapors were burned in the atmosphere and produced spinel (MgAl2 O4 ), MgO, and δ-Al2 O3 . By incrementing Al content, δ-Al2 O3 particles were produced, which gave the characteristic spectrum in the range of 13–18 µm as elucidated in a previous paper.9 Figure 2a shows the polyhedral shape. The corresponding ED pattern indicates that the produced particles were MgO and MgAl2 O4 . The cubic shape is typical for MgO particles.14 MgO reflection in the ED pattern weakens with decreasing Mg ratio (1:2 or 1:3). In the case of 1:2, the reflection corresponding to spinel becomes stronger than that of the 1:1 sample. That is, the spinel was efficiently produced under the condition of Mg:Al = 1:2, i.e., the stoichiometric composition of MgAl2 O4 . The morphology of the particles changed to spherical with the 1:3 sample.

136

M. Saito, M. Kurumada and C. Kaito

Fig. 2. TEM images and ED patterns of particles produced by flash gas evaporation. (a) Mg:Al = 1:1, (b) Mg:Al = 1:2, and (c) Mg:Al = 1:3.

The morphology of the particles can be classified into four types: (A) cubic, (B) octahedral, (C) elliptical, and (D) spherical shapes, as shown in Fig. 3. The results of EDX analysis for morphological alteration are also shown in Fig. 3. Since an electron beam of 30–50 nm size can be focused on the specimen, nanoprobe analysis was performed on about

Relationship between Morphology and Spectra

137

Fig. 3. Relationship between morphology and composition. The morphology can be classified into four types: (A) cubic, (B) octahedral, (C) elliptical, and (D) spherical.

10 particles for each shape. The atomic ratio for each morphology is shown in Fig. 3. MgO particles were cubic. When the Mg content was 75–100 at.%, the morphology was altered to the (110) truncated form from the cubic shape. With increasing Al content, {111} octahedral or truncated octahedral particles appeared. This morphological change is similar to that of Pb-doped KCl crystal.15 Whereas pure KCl is cubic with the {100} plane, its morphology is changed to octahedral with the {111} plane by impurity doping. In the case of KCl, the impurity concentration is of ppm order. Although the order of impurity concentration differs markedly, these morphological changes seem to be the same phenomenon. The HRTEM images of the cubic-form particles containing Al atoms which were verified by EDX analysis are shown in Fig. 4. The small white dots can be seen in the low-magnification image (A). The HRTEM image (B) clearly shows the (200) lattice of MgO along the cubic edge. Therefore, the cubic shape corresponded to (100) planes of the MgO structure. As can be seen from the enlargement of the lattice image in Fig. 4B, the (200) lattice images had different contrasts in short-range order, which correspond to the small white dots. Since MgO oxide is the characteristic metal excess oxide, the excess Al will be distributed on (100) planes as in the G.P. zone

138

M. Saito, M. Kurumada and C. Kaito

Fig. 4. HRTEM images of an MgO cubic particle containing Al. The excess Al atoms distributed in the (100) planes as in the G.P. zone.

seen in alloys and some ultrafine particles.16,17 If the amount of Al increases, the shape of the produced particles is altered to the octahedral shape. On the other hand, the particles in the C-region (Mg content: 35– 65 at.%) were markedly larger than the particles in the other regions, as shown in Fig. 2c. The fundamental shape in this region was elliptical. When the Mg content was less than the spinel content (about 33 at.%), the produced particles were spherical (D-region). Since the original pure MgO and δ-Al2 O3 particles produced by gas evaporation were cubic and spherical, the particle shapes were altered from MgO to Al2 O3 phases containing different amounts of Mg and Al. Since both δ-Al2 O3 and spinel (MgAl2 O4 ) have the same structure, no considerable morphological change occurred in the D-region. However, in the C-region, Mg content becomes greater than the Mg content in spinel, and excess Mg forms the MgO structure. The two-phase mixture between spinel (MgAl2 O4 ) and MgO phases was observed by HRTEM. Figure 5 shows the octahedral particle containing MgO. Since the lattice images of (400) MgAl2 O4 and (200) MgO were in parallel, the central spinel particles were covered with MgO, in which a definite lattice relation with MgO existed: (100) MgAl2 O4 //(100) MgO. The surface covered with MgO crystals entirely had a definite orientation with respect to the central particles (MgAl2 O4 ). The special moir´e-like contrasts in the central region was due to the parallel relation between MgAl2 O4 and MgO. This parallel growth shows that the MgO and

Relationship between Morphology and Spectra

139

Fig. 5. HRTEM images of the particles with moir´e fringes. This particle is based on the spinel structure and the surface covered with MgO particles.

MgAl2 O4 were obtained through the mixture phase during the growth process. Since the temperature of solidification of MgAl2 O4 from the liquidlike state is lower than that of MgO, MgO solidification took place from the surface as has been recently demonstrated on the crystallization process of amorphous Mg2 SiO4 .18 The MgAl2 O4 grew with a definite lattice relation with the surface of MgO crystals. Figure 6 shows HRTEM images of Al2 O3 -rich spinel particles (Dregion) with a characteristic stripe contrast similar to that of a superlattice. Since EDX analysis indicated that the particle with the stripe contrast contained 4–6 at.% Mg, these stripe contrasts may be due to the coexistence of the spinel and alumina phases. The (040) lattice of alumina was the region with different contrasts. Some of the strong-contrast regions that

Fig. 6.

HRTEM images of spherical particles with stripe contrast.

140

M. Saito, M. Kurumada and C. Kaito

were similar to the G.P. zone appeared partly on the particles. These strongcontrast lattice images correspond to the (400) lattice image of spinel. The (040)Al2O3 // (400)spinel relation existed on these particles. Since the Al2 O3 and MgAl2 O4 have the same spinel structure, they were distributed in the octahedral and tetrahedral sites on the oxygen close-packed structure. The coexistence of the two-phase mixture was easily realized by the diffusion of Mg metal. Therefore, the coexistence of two phases was achieved during the solidification from the mixture state of Mg–Al–O.

3.2. Infrared spectral changes induced by varying Mg–Al ratio Infrared spectral changes depending on the composition ratio of Mg and Al are shown in Fig. 7. The spectra of MgO and δ-Al2 O3 particles produced using a similar method are indicated in the figure. The spectra of Mg– Al–O particles show the differences in the features of MgO and δ-Al2 O3 particles. The spectra, which correspond to the samples shown in Fig. 2, have two characteristic features at 14 and 18 µm. The peaks at 14 and 18 µm correspond to the octahedral and tetrahedral units in the spinel structure, respectively.19 The spinel phase was contained in all the specimens, and the crystallographic shape of produced particles varied from cubic to spherical with increasing Al ratio. Since these peaks are totally different from those of

Fig. 7. IR spectra of MgO, δ-Al2 O3 , and Mg:Al = 1:1, 1:2 and 1:3 specimens. The spectra have two characteristic peaks at 14 and 18 µm.

Relationship between Morphology and Spectra

141

MgO and δ-Al2 O3 phases, the features of the spectrum reflect the alteration of the morphology. Since the peak at Mg:Al = 1:1 was attributable to the content of the MgO phase being markedly higher than those of the other specimens, the overlap of the MgO absorption and spinel absorption peaks gives the broad feature. A predominant growth of MgAl2 O4 was observed at Mg:Al = 1:2 and the feature becomes sharper.

4. Conclusion The morphology of the MgO–Al2 O3 compound markedly changed with composition. The morphologies were cubic for Mg-75–100 at.% particles, octahedral for Mg-70 at.% particles, elliptical with characteristic contrast for Mg-35–65 at.% particles, or spherical for less than Mg-40 at.% particles. HRTEM observation indicated that the structure of octahedral particles was based on spinel, and the structure of spherical particles was based on Al2 O3 . The characteristic moir´e contrast in elliptical particles was due to the coexistence of MgO and Al2 O3 . The IR spectra and ED patterns showed that the spinel phase was predominantly produced at Mg:Al = 1:2. The IR spectrum of the nm-sized spinel showed absorption peaks at 14 and 18 µm, which varied with their morphology.

References 1. G. R. Huss, A. J. Fahey, R. Gallino and G. J. Wasserburg, Astrophys. J. 430 (1994) L81. 2. E. Zinner, L. R. Nittler, P. Hoppe, R. Gallino, O. Straniero and C. M. O’D. Alexander, Geochim. Cosmochim. Acta 69 (2005) 4149. 3. L. Grossman, Geochim. Cosmochim. Acta 36 (1972) 597. 4. T. Yamamoto and H. Hasegawa, Prog. Theor. Phys. 58 (1977) 816. 5. C. Kaito, S. Kimura, K. Kamei, Y. Saito and C. Koike, Geomagn. Geoelectr. 46 (1994) 1043 . 6. T. Onaka, T. de Jong and F. J. Willems, Astron. Astrophys. 218 (1989) 169. 7. T. Posch, F. Kerschbaum, H. Mutschke, D. Fabian, J. Dorschner and J. Hron, Astron. Astrophys. 352 (1999) 609. 8. D. Fabian, T. Posch, H. Mutschke, F. Kerschbaum and J. Dorschner, Astron. Astrophys. 373 (2001) 1125. 9. M. Kurumada, C. Koike and C. Kaito, Mon. Not. R. Astron. Soc. 359 (2005) 643. 10. G. C. Sloan and S. D. Price, Astrophys. J. Suppl. 119 (1998) 141. 11. J. Hron, B. Aringer and F. Kerschbaum, Astron. Astrophys. 322 (1997) 280.

142

M. Saito, M. Kurumada and C. Kaito

12. A. K. Speck and A. M. Hofmeister, Workshop on Cometary Dust in Astrophysics, (2003) 6049. 13. T. Tanigaki, S. Kimura, N. Tamura and C. Kaito, JJAP 41 (2002) 5529. 14. C. Kaito, K. Fujita and H. Shibahara, JJAP 16 (1977) 697. 15. I. Sunagawa, Crystals: Growth, Morphology and Perfection (Cambridge University Press, Cambridge, 2005). 16. V. A. Phillips, Acta Metall. 21 (1973) 219. 17. Y. Kimura, H. Ueno, H. Suzuki, T. Tanigaki, T. Sato, Y. Saito and C. Kaito, Physica E 19 (2003) 298. 18. K. Kamitsuji, T. Sato, H. Suzuki and C. Kaito, Astron. Astrophys. 436 (2005) 165. 19. C. Morterra and G. Magnacca, Catalysis Today 27 (1996) 497.

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


IONIZATION OF POLYCYCLIC AROMATIC HYDROCARBON MOLECULES AROUND THE HERBIG Ae/Be ENVIRONMENT∗ ITSUKI SAKON†,§ and TAKASHI ONAKA Department of Astronomy, Schools of Science, University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan §[email protected] YOSHIKO K. OKAMOTO Institute of Astrophysics and Planetary Sciences, Ibaraki University, Japan 2-1-1 Bunkyo, Mito 310-8512, Ibaraki, Japan HIROKAZU KATAZA, HIDEHIRO KANEDA and MITSUHIKO HONDA‡ Institute of Space and Astronautical Science Japan Aerospace Exploration Agency 3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510, Japan

We present the results of mid-infrared N-band spectroscopy of the Herbig Ae/Be system MWC1080 using the Cooled Mid-Infrared Camera and Spectrometer (COMICS) on board the 8 m Subaru Telescope. The MWC1080 has a geometry such that the diffuse nebulous structures surround the central Herbig B0 type star. We focus on the properties of polycyclic aromatic hydrocarbons (PAHs) and PAH-like species, which are thought to be the carriers of the unidentified infrared (UIR) bands in such environments. A series of UIR bands at 8.6, 11.0, 11.2, and 12.7 µm is detected throughout the system and we find a clear increase in the UIR 11.0 µm/11.2 µm ratio in the vicinity of the central star. Since the UIR 11.0 µm feature is attributed to a solo-CH outof-plane wagging mode of cationic PAHs while the UIR 11.2 µm feature to a solo-CH out-of-plane bending mode of neutral PAHs, the large 11.0 µm/11.2 µm ratio directly indicates a promotion of the ionization of PAHs near the central star.

∗ This work is based on data collected at Subaru Telescope, which is operated by the National Astronomical Observatory of Japan. † I.S. is financially supported by the Japan Society for the Promotion of Science (JSPS). ‡ Current address: Department of Information Science, Kanagawa University, 2946 Tsuchiya, Hiratsuka, Kanagawa, 259-1205, Japan.

143

144

I. Sakon et al.

1. Introduction The unidentified infrared (UIR) bands are a series of emission bands observed at 3.3, 3.4, 6.2, 7.7, 8.6, 11.0, 11.2, 12.0, and 12.7 µm together with some other fainter features. They have been ubiquitously observed in various astrophysical objects, including reflection nebulae, HII regions, planetary nebulae, post-AGB stars,1 diffuse interstellar medium,2,3 external star forming galaxies,4 remote ultraluminous infrared galaxies,5 and submillimeter galaxies even at z = 2.8.6 They are supposed to be carried by small carbonaceous dust including polycyclic aromatic hydrocarbons (PAHs) and/or PAH-like species such as quenched carbonaceous composite (QCCs).7–9 They are stochastically excited by absorbing a single ultraviolet (UV) photon and release the energy with a number of infrared photons in cascades via several lattice vibration modes of aromatic C–C and C–H.10 Note that bulk QCC or amorphous carbon are not likely to be carriers of the UIR bands since the absorbed photon energy will not be confined within the aromatic group within/attached to the bulk QCC or amorphous carbon dust.11,12 The portion of the pumping energy to each vibration mode is supposed to be controlled by the physical and chemical conditions of the carriers such as the charging state, the molecular structure, and the size of the carriers, which follow as a consequence of the physical processing in the incident radiation environment as well as of the molecular evolution in chemically reactive regions. Therefore, understanding the systematic differences in UIR spectra in terms of the variation in the nature of the carriers in various astrophysical environments is, above all, important to use UIR bands as a useful probe of the local physical conditions.13 The ionization state of PAHs is one of the most significant factors to affect the spectral characteristics of the UIR bands. The ionization of PAHs is controlled by U/ne, where U is the strength of the incident radiation field that acts in promoting the photo-ionization of PAHs and ne is the electron density that plays a role in the recombination. A large U/ne ratio favors positively ionized PAHs.14 Past laboratory experiments and theoretical studies have shown that the UIR bands in the 6–9 µm region are much weaker than those in the 11–14 µm region when PAHs are neutral, however, that they become as strong as those in 11–14 µm region when PAHs are ionized.14–15 Actually several studies report that the variations in the ratios of 7.7 µm/11.2 µm and/or 8.6 µm/11.2 µm bands, for example, within a reflection nebula along the distance from the ionizing source,17,18

Ionization of Polycyclic Aromatic Hydrocarbon Molecules

145

among the Herbig Ae/Be stars with different spectral types19 and between the inner and outer Galactic plane20 have been reasonably explained by the changing in ionization status of the carriers of the UIR bands. The variations of 11.0 µm feature in the real astrophysical object have been firstly reported by Sloan et al.21 They observed the reflection nebula NGC1333 SVS 3 using the 5 m Hale telescope at Palomar, and found an excess at 10.8–11.0 µm and a feature around at 10 µm increase relative to the 11.2 µm feature in the close area to the illuminating early B star SVS 3. Recently, Werner et al.22 has reported the increase in 11.2 µm/11.0 µm with increasing distance from the central star based on the observation of reflection nebulae NGC7023 with Infrared Spectrograph on the Spitzer Space Telescope. Bregman and Temi17 have investigated the variation in the band ratio of 11.2 µm/7.7 µm within a reflection nebula along the distance from the central star and have made a quantitative evaluation of the relation between the 11.2 µm/7.7 µm and the ratio of the incident radiation field strength to the electron density. However, the 7.7 µm and 11.2 µm features come from different vibration modes (the former one corresponds to aromatic C–C stretching and the latter one to aromatic C–H out-ofplane bending, see Ref. 10 for details), and their relative strengths are affected by various factors other than ionization, such as, the degree of hydrogenation, the molecular structures, and the molecular sizes.23 On the other hand, the 11.0 and 11.2 µm features come from the same vibration modes but with different ionization status of the carriers and, therefore, the 11.0 µm/11.2 µm band ratio can be used as more direct and quantitative measure for the ionization of PAHs than the 11.2 µm/7.7 µm band ratio. In this work, we aim to quantitatively evaluate the UIR 11.0 µm/ 11.2 µm band ratio in terms of the relation with U/ne by investigating the spectral changes in these features around the Herbig Ae/Be system MWC1080 using the Cooled Mid-Infrared Camera and Spectrometer (COMICS)24 on board the 8 m Subaru Telescope. Quite recently, Habart et al.25 have presented the spatially resolved PAH emission in the inner disks of nearby Herbig Ae/Be stars using adapted optics system on board NACO/VLT. Our attempts would surely be useful to understand the physical conditions of the disk around Herbig Ae/Be stars as well as the evolution of disk materials including carbonaceous dust when the spatially well-resolved mid-infrared spectra of nearby Herbig Ae/Be stars are obtained.

146

I. Sakon et al.

2. Observations and Data Reduction 2.1. Target The mid-infrared spectroscopic data of MWC1080 were taken on the nights of July 16–17, 2005 (UT). MWC1080 is a non-isolated Herbig Ae/Be star located at the distance of 1.0 kpc26 surrounded by a bright reflection nebulae27 and the spectral type of the central star is classified as B0.28 Recently Wang et al.29 report the existence of at least 45 faint young low-mass stars within 0.3 pc radius from the central star based on the observation of the CFHT with the high-resolution adaptive optics. Optical bipolar outflows in the form of Herbig-Haro (HH) objects or HH-like jets with the radial velocity of 400 km−1 s have been discovered predominantly in the east of MWC1080.30 We have obtained the N-band low-resolution spectra with the resolving power of R ∼ 250 using the 0.33′′ slit. We also carried out the imaging observation at 11.7 µm (∆λ = 1.0 µm) to adjust the slit position. In order to cancel the high infrared background radiation, secondary mirror chopping was used at the frequency of 0.45 Hz with 20′′ throw. The spectra were obtained along two slit positions. One is set at a position angle of −88.75◦ so that it went across the nebula (S4), the illuminating star (S1), the companion star (S2), and the nearby source (S3) (hereafter SLIT0716, see Fig. 1a) and the total integration time was 810 sec, which enabled us to obtain the spectra of S1, S2, S3, and S4 with the signal-to-noise ratio larger

Fig. 1. Slit positions of (a) SLIT0716 and (b) SLIT0717 overlaid with the Subaru/COMICS N-band 11.7 µm image of MWC1080.

Ionization of Polycyclic Aromatic Hydrocarbon Molecules

147

than 10. The other one is set at a position angle of −25.00◦ to observe the nebulae and the illuminating star (hereafter SLIT0717, see Fig. 1b) and the total integration time was 1800 sec, which was sufficient enough to observe the variations in the diffuse UIR emission. Non-isolated Herbig Ae/Be stars provide us decisively an ideal environment to investigate the ionization effect of PAHs on their spectra. Herbig Ae/Be stars are a pre-main sequence object of 2–8M⊙ and the ionizing regions of hydrogen gas are restricted only within a few AU from the central star. The low ne environment, where carbon atoms instead of hydrogen dominantly supply electrons, realizes extremely high U/ne in the vicinity of the central star (within several hundred AU) compared to HII regions, taking account of the typical atomic abundance of C/H ∼ 2 × 10−4 for pre-main sequence objects.31

2.2. Data reduction The standard chopping subtraction and flat-fielding by thermal spectra of the telescope cell cover were employed. Then the spectra of MWC1080 along the SLIT0716 and SLIT0717 as well as the standard star HD3712 were obtained. The wavelength calibration was performed using the atmospheric emission lines and the uncertainty was estimated to be less than 0.0025 µm.32 The spectra of MWC1080 along SLIT0716 and SLIT0717 were divided by the standard star HD3712 spectrum for the purpose of correcting the atmospheric absorption, and then we multiplied the resulting spectra by the template spectrum of the standard star provided by Cohen et al.33 Finally, we adjusted the flux measured in the N8.8 and N 12.4 µm imaging bands to correct the slit throughput. The uncertainty in the flux at each wavelength was estimated from the noise in the blank sky. We note that the ozone absorption increases the uncertainty in the 9.3–10.0 µm region.

3. Results 3.1. Obtained spectra along the SLIT0716 Figure 2 shows the obtained spectra along SLIT0716. S1 is the illuminating central B0 star and the obtained spectrum is dominated by the strong emission from the photosphere. A slight dent in 9–10.5 µm seems to be the effect of absorption by amorphous silicate but we note that this wavelength region is suffered by the atmospheric ozone absorption. On the other hand,

148

I. Sakon et al.

Fig. 2. Observed spectra at (a) S1, (b) S2, (c) S3, and (d) S4 along the SLIT0716. A pixel corresponds to an area of 0.165′′ × 0.165′′ .

S2 is the companion located at less than 1 arcsecond west to S1 and the obtained spectrum shows the clear presence of crystalline silicate features. The peaks around at 9.3 and 10.5 µm are supposed to be carried by crystalline enstatite, and those around at 11.3 and 11.9 µm are supposed to be carried by crystalline forsterite. The spectrum at S3 exhibits a feature peaking around 11.3 µm, where the contribution from the UIR 11.2 µm cannot be distinguished from that from the crystalline forsterite. However, we cannot further discuss the dust composition for this spectrum since other features characteristic to PAHs nor to crystalline forsterite are hardly recognized due to the relatively low signal-to-noise ratio. S4 is the diffuse nebula region and the obtained spectrum shows a series of the UIR bands, including those at 8.6 and 11.2 µm.

3.2. Changing in UIR solo-CH bond spectra along the SLIT0717 PAHs in the diffuse nebulae of MWC1080 are supposed to be illuminated by the central B0 star and we assume that the projected distance from the central B0 star corresponds to the actual distance from the heating

Ionization of Polycyclic Aromatic Hydrocarbon Molecules

149

source. The spectrum of the diffuse nebula is actually dominated by the UIR features as observed in S4 on SLIT0716. Along SLIT0717, therefore, we can investigate the spectra of PAHs in various strengths of the radiation field. In order to compare the profiles of these features at different positions, we subtract a local underlying continuum defined by linear interpolation between the average values around 10.9 and 11.65 µm.34 Figure 3 shows the variation in the continuumsubtracted UIR spectra of solo-CH modes normalized by the total intensity of their strengths along SLIT0717. We recognize a small hump located on the blue shoulder of the distinct UIR 11.2 µm feature extending from 10.95 to 11.1 µm (hereafter the UIR 11.0 µm band, see Fig. 3). In this analysis, the strength of the UIR 11.0 µm band is defined by the integration of the continuum-subtracted emission from 10.95 to 11.1 µm, and that of the UIR 11.2 µm band is defined by the integration of the continuum-subtracted emission from 11.1 to 11.6 µm. 1 − σ errors for each band strengths are 2 + σb2 )1/2 , where σm is the measurement error and σb is the defined by (σm uncertainty in the baseline estimation. σm is defined by σm = δsky × ∆λ, where δsky is the standard deviation of flux density of the blank sky

Fig. 3. Variations in the profiles of solo-CH bond features along SLIT0717. Each spectrum is normalized by the total intensity of the solo-CH bond features. The spectrum at the central star position corresponds to zero offset. The spectra are shifted.

150

I. Sakon et al.

spectrum in units of W m−2 µm−2 pixel−1 within the wavelengths used for calculating the strength of each UIR band and ∆λ is the width of each wavelength region. σb is defined by σb = ηbase × ∆λ, where ηbase is the standard deviation of flux density in units of W m−2 µm−2 pixel−1 of the spectrum at each position along SLIT0717 within the wavelength range used for the continuum definition. Figure 4a shows the spatial distribution of the intensities of UIR 11.0 and 11.2 µm features as a function of the offset from the central B0 star along SLIT0717. The UIR 11.0µm feature is significantly detected with

Fig. 4. Spatial variations (a) in the UIR 11.0 and 11.2 µm intensities and (b) in the relative band strengths of UIR 11.0–11.2 µm features as a function of the offset, ∆d (arcsec), from the central B0 star along SLIT0717. A pixel corresponds to an area of 0.165′′ × 0.165′′ . Regions where 11.2 µm feature has local peaks of in its intensity distribution (∆d ∼ −4, −2, 0, and 3) are indicated with shadows.

Ionization of Polycyclic Aromatic Hydrocarbon Molecules

151

generally better signal-to-noise ratios than ∼3 where the 11.2 µm feature has local peaks (i.e., ∆d ∼ −4, −2, 0, and 3; see Fig. 4) in its intensity distribution (Fig. 4a). Such regions are expected to have larger column density of PAHs and/or better supply of UV photons than other regions. Figure 4b shows the spatial variations of the relative band strength of the UIR 11.0–11.2 µm features along SLIT0717. We find the ratio increases up to ∼0.3 in the vicinity of the central star. Among the above four regions around at ∆d = −4, −2, 0, and 3, where the UIR 11.0 µm feature is significantly detected in each spectrum, the nearest position to the central star (∆d ∼ 0) shows the largest ratio while the most distant position from the central star (∆d ∼ −4) shows the smallest ratio. Taking account of the fact that the 11.0 µm feature is assigned to a solo CH out-of-plane wagging mode of cationic PAHs while the 11.2 µm feature to a solo CH out-of-plane bending mode of neutral PAHs,35 the large ratio of the UIR 11.0 µm/11.2 µm in the vicinity of the central star can be interpreted as the promotion of PAHs’ ionization to cationic species. We note that the UIR 11.2 µm feature emitted from large PAHs may somewhat blue shifted in its peak position35,36 and a slight contribution from largest members of neutral PAHs to our calculated intensity of 11.0 µm feature would be possible. However, in the following analysis, we assume that our calculated intensity of 11.0 µm feature is dominated by the emission carried by the CH out-of-plane wagging mode of cationic PAHs. We also note that some types of crystalline silicate (e.g. crystalline forsterite)37 can contribute to the spectra in 11 µm region in the vicinity of the Herbig Ae/Be stars with a broad band feature peaking at 11.3 µm as can be seen in the spectrum of S2 on SLIT0716 (see Fig. 2b), but their band widths are typically broader than those of UIR features. Since our continuum subtracted spectra do not show an increase in the band width of 11.2 µm feature in the vicinity of the central star (Fig. 3), we assume that spectra above the continuum in the 11 µm region along SLIT0717 are dominated by UIR bands in the following analysis.

4. Discussion In this section, we examine a quantitative relation between the band ratio of the UIR 11.0 µm/11.2 µm and the ratio of the interstellar radiation field strength to the electron density U/ne assuming a simple model in which the central B0 star is located inside in the spherically symmetric nebula.

152

I. Sakon et al.

In this model, the interstellar radiation field strength at a distance r from the B0 star U (r) in units of the solar vicinity is given by,  1µm 2 ˚ πBλ (T∗ )(R∗ /r) dλ U (r) = 912A  3 µm ⊙ ˚ 4πJλ dλ 912A

K is the effective temperature, R∗ = 3.2 × 109 m is the where T = 10 effective radius of the central B0 star,38 and Jλ⊙ is the interstellar radiation field of solar vicinity.39 −3 We adopt a constant electron density ne = 100+300 ) from Poetzel −50 (cm 30 et al. in the region in our analysis. We examine the relation between the UIR 11.0 µm/11.2 µm ratio and U/ne each in the northwest part of the slit and in the southeast part of the slit (see Fig. 5). We can clearly see a correlation between the UIR 11.0 µm/11.2 µm ratio and U/ne , suggesting that the 11.0 µm/11.2 µm ratio can be a measure for the interstellar radiation field strength and the electron density. A relatively large scatter at small U/ne is supposed to originate from both the inhomogeneity in the electron density and the underestimation of the radiation field strength at the distant region from the central B0 star, where the contribution from faint low-mass stars29 cannot be neglected. Detailed modeling of the interstellar radiation field strength and the electron density as well as the additional spectroscopic observation of the UIR 11.0 and 11.2 µm around spatially resolved non-isolated Herbig Ae/Be ∗

4.31

Fig. 5. (a) Relative band strengths of 11.0 µm/11.2 µm against U (r)/ne in the northwest part of the slit (−6′′ < d < 0′′ ) and (b) those in the southeast part of the slit (0′′ < d < 6′′ ). The error bars for the horizontal axis are defined by the uncertainties in −3 ) and those for the vertical axis are calculated from electron density ne = 100+300 −50 (cm the uncertainties in the estimation of the UIR band strengths.

Ionization of Polycyclic Aromatic Hydrocarbon Molecules

153

objects are quite important to examine the quantitative and robust relation between the UIR 11.0 µm/11.2 µm ratio and U/ne .

5. Summary We present mid-infrared N-band spectroscopy of the Herbig Ae/Be system MWC1080 using the COMICS on board the 8 m Subaru Telescope. MWC1080 has a geometry such that diffuse nebulous structures extend around the central Herbig B0 type star. We focus on the properties of PAHs and PAH-like species, which are thought to be the carriers of the UIR bands in such environments. A series of the UIR bands at 8.6, 11.0, 11.2, and 12.7 µm are detected throughout the system and we find a clear increase in the UIR 11.0 µm/11.2 µm ratio in the vicinity of the central star. Since the UIR 11.0 µm feature is attributed to a solo-CH out-of-plane wagging mode of cationic PAHs while the UIR 11.2 µm feature to a soloCH out-of-plane bending mode of neutral PAHs, the large 11.0 µm/11.2 µm ratio directly indicates a promotion of the cationic ionization of PAHs. This work suggests an application and robust use of the UIR 11.0 µm/11.2 µm ratio as a valid probe of the local interstellar radiation field strength and the electron density.

6. Acknowledgments The authors are grateful to all the staff members of the Subaru Telescope for the continuous support. I.S. especially thanks Drs. Hideko Nomura, Amit Pathak, and Hiroshi Kimura for useful comments and discussion. This work is supported by a Grant-in-Aid for the Japan Society for the Promotion of Science (JSPS). Y.K.O. is supported by a Grant-in-Aid for young scientists (#17740103) by the Ministry of Education, Culture, Sports, Science and Technology, Japan.

References 1. A. T. Tokunaga, ASP Conf. Ser. 124 (1997) 149. 2. T. Onaka, I. Yamamura, T. Tanabe, T. L. Roellig and L. Yuen, PASJ 48 (1996) L59. 3. K. Mattila et al., Astron. Astrophys. 315 (1996) L5. 4. G. Helou et al., Astrophys. J. 532 (2000) 21. 5. L. Yan et al., Astrophys. J. 628 (2005) 604.

154

I. Sakon et al.

6. D. Lutz et al., Astrophys. J. 625 (2005) L83. 7. A. Leger and J. L. Puget, Astron. Astrophys. 137 (1984) L5. 8. L. J. Allamandola, A. G. G. M. Tielens and J. R. Barker, Astrophys. J. 290 (1985) 25. 9. A. Sakata, S. Wada, T. Tanabe and T. Onaka, Astrophys. J. 287 (1984) 51. 10. L. J. Allamandola, A. G. G. M. Tielens and J. R. Barker, Astrophys. J. Suppl. 71 (1989) 733. 11. A. Li and B. T. Draine, Astrophys. J. 572 (2002) 232. 12. B. T. Draine and A. Li, Astrophys. J. 657 (2007) 810. 13. E. Peeters, H. W. W. Spoon and A. G. G. M. Tielens, Astrophys. J. 613 (2004) 986. 14. E. L. O. Bakes, A. G. G. M. Tielens and C. W. Bauschlicher, Astrophys. J. 556 (2001) 501. 15. D. J. de Frees et al., Astrophys. J. 408 (1993) 530. 16. J. Szczepanski and M. Vala, Astrophys. J. 414 (1993) 646. 17. J. Bregman and P. Temi, 621 (2005) 831. 18. C. Joblin, A. G. G. M. Tielens, T. R. Geballe and D. H. Wooden, Astrophys. J. 460 (1996) L119. 19. G. C. Sloan et al., Astrophys. J. 632 (2005) 956. 20. I. Sakon et al., Astrophys. J. 609 (2004) 203 (erratum 625 (2005) 1062). 21. G. C. Sloan et al., Astrophys. J. 513 (1999) 65. 22. M. W. Werner et al., Astrophys. J. Suppl. 154 (2004) 309. 23. E. Peeters et al., Astron. Astrophys. 390 (2002) 1089. 24. H. Kataza et al., Proc. SPIE Conf. Ser. 4008 (2000) 1144. 25. E. Habart, A. Natta, L. Testi and M. Carbillet, Astron. Astrophys. 449 (2006) 1067. 26. J. A. Eisner, Astrophys. J. 613 (2004) 1049. 27. G. H. Herbig, Astrophys. J. Suppl. 4 (1960) 337. 28. M. Cohen and L. V. Kuhi, Astrophys. J. Suppl. 41 (1979) 743. 29. S. Wang, L. W. Looney, W. Brandner and L. M. Close, IAUS 237 (2006) 238. 30. R. Poetzel et al., Astron. Astrophys. 262 (1992) 229. 31. T. P. Snow and A. N. Witt, Science 270 (1995) 1455. 32. Y. K. Okamoto et al., Proc. SPIE Conf. Ser. 4841 (2003) 169. 33. M. Cohen et al., AJ 117 (1999) 1864. 34. F. C. Witteborn et al., Astrophys. J. 341 (1989) 270. 35. D. M. Hudgins and L. J. Allamandola, Astrophys. J. 516 (1999) L41. 36. A. Pathak and S. Rastogi, Proc. of 36th COSPER Scientific Assembly (2006), p. 432. 37. C. Koike et al., Astron. Astrophys. 399 (2003) 1101. 38. R. Millan-Gabet et al., Astrophys. J. 546 (2001) 358. 39. P. G. Mezger, J. S. Mathis, N. Panagia, Astron. Astrophys. 105 (1982) 372.

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


SEARCH FOR SOLID O- AND N-RICH ORGANIC MATTER OF PREBIOTIC INTEREST IN SPACE ˜ G. M. MUNOZ CARO∗,† and E. DARTOIS‡,§ †Centro de Astrobiolog´ ıa, INTA-CSIC Ctra. de Ajalvir, km 4, Torrej´ on de Ardoz, 28850 Madrid, Spain ‡Institut d’Astrophysique Spatiale, UMR 8617, Bˆ at. 121 Universit´ e Paris XI, 91405 Orsay, France ∗[email protected] §[email protected]

Mainly based on our previous results, this article evaluates the presence of solid organic matter of prebiotic interest in space: from the carbon grains observed toward diffuse interstellar clouds and the organic grain mantles made from ice processing that are likely present in dense interstellar clouds and circumstellar regions, to the carbon component of solar system objects that could have delivered organic species to the early Earth [comets, meteorites, and interplanetary dust particles (IDPs)]. Here the term organic is attributed to hydrocarbon materials rich in O and N that contain prebiotic species or their precursors, and are therefore interesting for astrobiology. Organic residues made in the laboratory from ice ultraviolet-photoprocessing under simulated interstellar conditions are used as representative of organic matter and compared by means of infrared and Raman spectroscopy to carbon-bearing extraterrestrial samples. It is observed that the carbon bulk in grains of the diffuse interstellar medium, carbonaceous chondrites, and the IDPs collected in the stratosphere consists of amorphous carbon, with at most a small percent of organic matter. On the other hand, about 50% of the carbon component in comet Halley is made of organics that formed in the interstellar/circumstellar medium in the absence of liquid water. The characterization of cometary organic matter is therefore vital to constrain the contribution of extraterrestrial matter to the origin of life on Earth.

1. Introduction Carbon is sixth in the scale of cosmic elemental abundances. It is a component of many gas phase molecules in the interstellar medium, from diatomic molecules like CO to large polyaromatic hydrocarbons. Most of the observed solid carbon in the interstellar medium is in the form of amorphous carbon, graphite, diamond, carbides, and ices. Icy grain mantles in dense 155

156

G. M. Mu˜ noz Caro and E. Dartois

interstellar clouds and circumstellar regions are mainly composed of H2 O, carbon-containing molecules (such as CO, CO2 , CH3 OH, OCN− , OCS, 1–3 Ulatraviolet (UV) irradiation H2 CO, HCOOH, CH4 ), and NH3 or NH+ 4. and warm-up of such ice mixtures in the laboratory leads to an organic refractory residue that is rich in prebiotic organic species or their precursors (carboxylic acids and their ammonium salts, alcohols, amides, esters, amino acids, N-heterocycles, etc.).4–10 The solar nebula that preceded our solar system formed by gravitational collapse of a portion of a dense interstellar cloud. Comets and primitive meteorites preserve carbon matter of interstellar or primitive solar nebula origin. As already mentioned, the term organic is attributed to hydrocarbon materials rich in O and N that contain prebiotic species or their precursors, and are therefore interesting for astrobiology. Here, we search for solid organic matter in space that might have contributed to the origin of life on Earth. This is done by comparing organic residues made from ice photoprocessing, used as representative of organic matter that could be present in space, to carbon-bearing matter observed in the diffuse interstellar medium and solar system bodies such as comets, meteorites, and interplanetary dust particles (IDPs). The characterization that enables such comparison is accomplished by infrared spectroscopy and, for laboratory available samples, also Raman spectroscopy. In particular, the profile of the infrared 3.4-µm feature (3000–2800 cm−1 , CH stretching modes in hydrocarbons) and the presence/absence of the bands associated with functional groups serve to characterize the carbon material either as organic in nature or as amorphous carbon. Amorphous carbon consists of an amorphous matrix composed of sp2 , sp3 , and even sp1 , hybridized carbon atoms. a-C refers to amorphous carbon with less than 20% hydrogen. Hydrogenated amorphous carbons (a-C:H) contain 20–60% hydrogen and have as a consequence a low C-C sp3 content. The sp2 /sp3 ratio, the hydrogen content, and the degree of sp2 clustering determine the macroscopic properties of amorphous carbon.11 As it will be shown, the aromatic component of amorphous carbon is well traced by Raman spectroscopy. Two bands dominate the Raman spectrum of amorphous carbon, the D (disorder mode, 1300–1500 cm−1 range) and G (graphite mode, around 1600 cm−1 ) bands. The D band corresponds to a breathing mode in aromatic rings and the G band is due to a C=C stretching mode in olefinic or aromatic carbon. From the intensity ratio of these bands, I(D)/I(G), the value of the aromatic domain size, La , can be obtained.11

Search for Solid O- and N-Rich Organic Matter

157

Section 2 describes the composition of organic refractory residues made from photo- and thermal processing of interstellar/circumstellar ice analogs, as representative of organic matter thought to be present in space. The chemical characterization of carbon grains in the diffuse interstellar medium is summarized (Sec. 3) as well as that of small bodies of the solar system (Sec. 4) and compared to organic refractory residues; the similarities are outlined in Sec. 5. The delivery of organic matter to the primitive Earth is briefly discussed in Sec. 6.

2. Organic Refractory Matter of Prebiotic Interest Made from UV-Irradiation of Interstellar/Circumstellar Ice Analogs Energetic ice processing (UV photons and cosmic rays) is expected to take place in dense interstellar clouds, the sites of star formation, and the outer parts of circumstellar regions. Laboratory experiments simulating the photo- and thermal processing of interstellar ice analogs show the formation of new molecules, radicals, and other fragments.12 Large organic compounds are produced by both ice photoprocessing4–10 and by ion bombardment.13–15 Here we will focus on the organic refractory products of ice UV-photoprocessing. Experiments simulating the UV irradiation of ice mantles in circumstellar/interstellar regions consist in the simultaneous deposition and UVirradiation of a H2 O:CH3 OH:CO:CO2 :NH3 ice layer at T ≈ 12 K and P ≤ 10−7 mbar. The chemical evolution of the ice is monitored in situ by Fourier transform infrared spectroscopy. After warm-up to room temperature a refractory residue is observed. The infrared spectrum of the residue obtained from irradiation of a H2 O:CH3 OH:CO:CO2 :NH3 = 2:1:1:1:1 ice mixture is shown in Fig. 1. Among the strong features in this spectrum are those attributed to the stretching modes of hydroxyl (OH, at 3500–2300 cm−1 ), carbonyl (C=O, at ∼1650–1750 cm−1 ), and carboxylic (COO− , at 1586 cm−1 ) groups.10 The 3.4-µm feature assigned to the aliphatic CH stretching modes, composed of just two shallow bands at ∼2925 and 2875 cm−1 , is quite peculiar. Its profile is strongly affected by the electrophylic oxygen atom incorporated in the aliphatic structures. After integration of the absorbance features it is found that the main component of this residue is hexamethylenetetramine [(CH2 )6 N4 ] as evidenced by its main absorption features at 1236 and 1007 cm−1 , followed by ammonium salts of carboxylic acids [(R-COO− )(NH+ 4 )], amides [H2 NC(=O)-R] and esters

158

G. M. Mu˜ noz Caro and E. Dartois

Fig. 1. Infrared spectrum of the residue obtained after UV-irradiation and warm-up of the H2 O:NH3 :CH3 OH:CO:CO2 = 2:1:1:1:1 ice mixture (adapted from Ref. 10).

[R-C(=O)-O-R].5,10 Amino acids or their precursors and N-heterocyclic molecules are both present with an abundance of the order of ∼1% by number of molecules.8,9 If the concentration of H2 O in the starting ice mixture is low compared to the other ice components, species based on polyoxymethylene [(–CH2 O–)n ] are the most abundant. In general, residues show high O/C ratios; for the one resulting from UV irradiation of H2 O:CH3 OH:CO:CO2 :NH3 = 2:1:1:1:1 ice, O/C ≃ 0.4 was found.10 The visible Raman spectrum of these residues is dominated by a strong photoluminescence, presumably due to the high H/C and O/C ratios, and the first-order D and G bands, characteristic of disordered carbonaceous materials, are not observed. Although these bands could be masked by the photoluminescence continuum, they are likely not present simply because the organic refractory residues have no, or very few, olefinic and aromatic bonds, in agreement with the results obtained from infrared and GC-MS analysis.17

3. Solid Carbon in the Diffuse Interstellar Medium The infrared spectrum due to the absorption of carbon grains toward diffuse interstellar clouds shows a prominent band at 3.4 µm that is

Search for Solid O- and N-Rich Organic Matter

159

very distinct from that of the organic residues described in Sec. 2 (it has three subfeatures at 2923, 2958, and 2865 cm−1 ; respectively, the a-CH3 , a-CH2 , and s-CH3 stretching modes in aliphatics), and bands due to the CH bending modes at 6.85 µm (around 1460 cm−1 ) and 7.25 µm (around 1380 cm−1 ). The profiles and relative intensities of these features are very well matched with the laboratory spectrum of a hydrogenated amorphous carbon polymer, called photoproduced a-C:H, that is made from photoprocessing of simple aliphatic species under interstellar conditions.18,19 This suggests that the composition of this material is similar to that of carbonaceous grains in the diffuse medium. Based on the infrared and Raman analysis of photoproduced a-C:H, carbon grains in the diffuse interstellar medium are expected to be composed of an a-C:H polymer, of low oxygen content, consisting of hydrocarbon chains containing olefinic and aliphatic bonds with CH2 /CH3 ≈ 2 and eventually some embedded small aromatic units (1–2 rings).18,19 An example of a possible substructure unit of photoproduced a-C:H is shown in Fig. 2.

Fig. 2. A typical expected substructure unit for the hydrogen-rich photoproduced a-C:H (adapted from Ref. 18).

160

G. M. Mu˜ noz Caro and E. Dartois

4. The Carbon Fraction of Small Bodies in the Solar System: Comets, Meteorites and Interplanetary Dust Particles Carbonaceous materials are present in various environments of the outer solar system: on moons like Titan, on outer belt asteroids and Kuipert belt objects, and in comets. The delivery of solid organic matter to the early Earth was mainly caused by comets, meteorites, and IDPs; the characterization of the solid carbon fraction of these objects is summarized below.

4.1. Comets Comets are thought to preserve the most pristine material in the solar system as their high abundances of volatiles indicate. These bodies are thought to be formed by agglomeration of dust particles of pre-solar and/or solar nebula origin in the outer parts of the solar nebula.20,21 However, the detection of crystalline silicates in comets, as inferred from infrared observations,22 and preliminary results obtained from the analysis of the dust collected by the Stardust mission during flyby of comet Wild 2, indicate that not all comets are made of pristine material that was kept cold since its formation.23 The presence of high-temperature mineral grains suggests that at least some cometary grains were reprocessed in regions near the early Sun and ejected by radial mixing to the outer regions of the nebula,24,25 where subsequently gas phase species would accrete onto such grains forming ice mantles. Cometary dust is rich in carbonaceous/organic matter, as much as 50% by mass for comet Halley,26 although this proportion varies among comets. A large fraction of the carbon matter in Halley, about 50%, is oxygen-rich (O/C ≥ 0.5); these compounds are consistent with structures of alcohols, aldehydes, ketones, acids and amino acids, and their salts. The exact makeup of these molecules cannot be unambiguously identified.26,27 Several N-heterocyclic compounds are very likely to be present in the dust of comet Halley. These include pyrrole, pyrazole/imidazole, pyridine, pyrimidine, and its derivatives.28 More recently, low O and H and high N contents were reported for comet Wild 2.29 The Stardust samples will allow the first laboratory analysis of carbonaceous/organic cometary matter. The Rosetta mission attempts to go a step further by analyzing in situ the composition of a comet nucleus.

Search for Solid O- and N-Rich Organic Matter

161

4.2. Meteorites The main source of carbon in the meteorites known as carbonaceous chondrites is a kerogen-like material, but they also contain a large variety of organic species, with concentrations of a few parts per million of the carbon abundance. These include carboxylic and dicarboxylic acids, amino acids, hydroxy-acids, amines, amides, nitrogen heterocycles including purines and a pyrimidine, carbonyl compounds and alcohols.30,31 Stable isotope measurements indicate an interstellar origin of these compounds, although the original organics probably underwent reactions in the aqueous phase on the asteroidal parent body, which could have altered their composition.30 Organic species are embedded in the carbonaceous matrix. The right panel of Fig. 3 shows that the Raman D and G bands corresponding to the carbon bulk of the Orgueil meteorite (top spectrum) are similar to those of IDPs (labeled Y, K2, K3, and N) and are characteristic of amorphous carbon. The FWHM of the D and G bands corresponding to Orgueil and Murchison are close to the lower limit for IDPs. From the intensity ratio of these bands, I(D)/I(G), the values of the aromatic domain sizes, La , are 1.3 nm for Orgueil and 1.1 nm for Murchison (respectively, 30 and 20 rings in total, assuming a two-dimensional structure). These values fall within the IDP values. The infrared and Raman spectra of insoluble organic

Fig. 3. Left: Comparison between the infrared spectra of IDP Y (bottom) and a residue (top) made from UV-irradiation of the H2 O:CH3 OH:CO:CO2 :NH3 = 2:1:1:1:1 ice mixture.38 Right: First-order Raman D and G bands of four HF-hydrolized IDPs (abbreviated as Y, K2, K3, and N) and Orgueil acid residue after baseline correction (thick line) and Lorentzian fits (thin line). The intensity of the spectra was shifted for intercomparison.39

162

G. M. Mu˜ noz Caro and E. Dartois

matter in Orgueil and Murchison are consistent with that of a-C:H with a CH2 /CH3 ratio close to the diffuse interstellar value around 2 and much larger aromatic units (20–30 rings in these meteorites compared to 1–2 rings in carbon grains of the diffuse medium). This conclusion is consistent with earlier characterizations. The insoluble organic matter of Murchison was earlier described as a structurally complex, extensively cross-linked, and highly aromatic macromolecule. The O/C and N/C values found for Murchison are 25 and 2.9%.32

4.3. Interplanetary dust particles Interplanetary dust particles are collected by NASA aircrafts from the Earth’s stratosphere. These particles mainly consist of silicate grains coated by carbonaceous material.33 They have an average carbon content of 10– 12%.34 The detection of organic species in IDPs is hindered by their small masses of the order of nanograms. The 3.4-µm feature profile of IDPs is composed of three subfeatures shared by aliphatic species with no electrophylic heteroatoms. Although some IDPs are known to contain carbonyl (C=O) groups35,36 and hydrocarbon chains in some IDPs host N,37 the infrared absorption bands due to functional groups are weak or not present in most IDP spectra compared to infrared spectra of refractory organic residues made from UV irradiation of interstellar/circumstellar ice analogs (see left panel of Fig. 3).38 The Raman spectra of four IDPs are shown on the right panel of Fig. 3. They are all reminiscent of amorphous carbon. The I(D)/I(G) values of IDPs were found to vary between 0.66 and 1.39, i.e. La = 1.1– 1.6 nm (a total number of about 20 to 42 rings, assuming a two-dimensional structure). The infrared spectra of IDPs (left panel of Fig. 3 shows the spectrum of IDP Y) are also well fitted with amorphous carbon, and provide information on the aliphatic component and the organic functional groups. From infrared and Raman spectroscopy of IDPs it is concluded that the bulk of the carbonaceous IDP component consists of a-C or a-C:H for IDPs with an observed 3.4-µm infrared feature. This corresponds to a material with aromatic units of ∼20–40 rings in total, depending on the particle, linked either by aliphatic chains with CH2 /CH3 ratios between 2.8 and 5.5 for IDPs containing a-C:H, or a carbon sp3 -skeleton for IDPs containing a-C. This material presents relatively low O/C and N/C contents.36,38

Search for Solid O- and N-Rich Organic Matter

163

Fig. 4. Left: A typical expected substructure unit for the a-C:H observed in many IDPs showing three aromatic units depicted in two dimensions. Right: The real structure is not planar because it folds gradually as the number of aromatic units increases. The a-C:H polymer in IDPs is build up of numerous aromatic units linked by aliphatic chains, leading to the intricate structure characteristic of a-C:H. Small amounts of functional groups (OH, NH, etc.; not shown in this figure) can be inserted in the aliphatic chains.

An example of the expected substructure unit of the a-C:H present in IDPs is shown in Fig. 4. Thermal annealing of organic refractory residues above ∼300–400 ◦ C, in the vicinity of the corresponding binding energies for C=O and C–H bonds, leads to a material that resembles spectroscopically the a-C or aC:H in IDPs. It is thus possible that the carbon bulk in IDPs resulted from heating of organic grain mantles in the solar nebula or during atmospheric entry heating. That can explain the morphology of cluster IDPs and the high D/H and 15 N/14 N associated with the carbon fraction in IDPs that point to a formation at cryogenic temperatures.38

5. Presence of Organics Made from UV-Photoprocessing of Ice in Small Solar System Bodies Experiments show that photoprocessing of ice mantles which likely took place in the local dense cloud and/or the solar nebula leads to a variety

164

G. M. Mu˜ noz Caro and E. Dartois

of molecules of biochemical relevance. Some of these molecules will be stored in comets and asteroids. There are significant similarities between the organics found in comets and meteorites and the organic refractory residues produced by ice photoprocessing in the laboratory. First, the ubiquitous oxygen-rich complex organic molecules (O/C ≥ 0.5) found in the coma of comet Halley26–28 have an elemental composition very similar to that of carboxylic acids and alcohols which comprise an abundant fraction of the organic residues.4,7 Glycolic acid, the most abundant carboxylic acid produced by UV-photoprocessing and warmup of the H2 O:CO:NH3 = 5:5:1 ice mixture4,7 was found in the Murchison meteorite.40 Second, the Nheterocycles present in organic residues are similar to those inferred from the data of comet Halley; some of them are precursors of biological cofactors.28 Third, amino acids and diamino acids and/or their precursors also result from ice photoprocessing and warm-up.6,9 Most of these amino acids are common to the Murchison meteorite. Based on the detection of diamino acids in the organic residues, the presence of diamino acids in Murchison was first predicted9 and they were later detected.41 Amino acids are the components of proteins and diamino acids are peptide nucleic acid components, a possible precursor of RNA. This suggests that energetic processing of ice in the local dense cloud and the solar nebula cannot be disregarded since the above reported species (carboxylic acids, N-heterocycles, and amino acids) are not produced in the atmospheres of stars or by thermal reactions in the ice without irradiation. Strecker synthesis could explain the formation of species like α-amino acids on meteorite parent bodies where aqueous alteration took place, but does not explain the presence of organics in comets. Furthermore, for the synthesis in space of N-heterocyclic compounds and diamino acids, ice irradiation is to our knowledge the only reported formation mechanism. 6. Delivery of Extraterrestrial Organic Matter to the Early Earth Extraterrestrial delivery of prebiotic molecules could have triggered the appearance of life on Earth.42,43 Shortly after its formation, the Earth suffered numerous impacts by cometary bodies that were ejected from their formation site in the Uranus–Neptune zone by gravitational perturbations of the growing protoplanets,44,45 about 3.9 Gyr ago, known as the era of heavy bombardment. In this way, prebiotic species were delivered to the primitive Earth via comets, and also carbonaceous asteroids and IDPs.46

Search for Solid O- and N-Rich Organic Matter

165

It has been estimated that the mass of asteroids and comets reaching the Earth after core formation was 0.7–2.7×1022 kg, and that comets represented less than 0.1% by mass of the impacting population.47 An important fraction of the incoming extraterrestrial matter is carbonaceous. The presence of extraterrestrial prebiotic molecules in carbonaceous chondrites indicates that a fraction of these species survived the impact with the Earth. The extraterrestrial carbon delivered on the Earth surface can easily account for the surficial biomass of the Earth, i.e. about 1015 kg. As an example, it was calculated that comet Halley organic matter corresponds to about 10% of the current biomass of the Earth.48 It is, however, important to note that organic matter represents a small fraction of the total carbon in meteorites; the carbon bulk consists of amorphous carbon, a priori less attractive for prebiotic chemistry. The same holds for IDPs (Sec. 4). As mentioned in Sec. 4.3, amorphous carbon could result from thermal annealing of pristine organic grain mantles that in turn were made by ice irradiation in space. In this context, cometary matter seems to be the best candidate to trigger chemical evolution, as proposed by Or´ o43 : Although they may not be as pristine as previously thought, comets like Halley are rich in organic matter with high O and N contents, probably dating from interstellar medium or solar nebula grains, which did not undergo aqueous alteration or substantial heating (organic matter is preserved only at temperatures below ∼300 ◦ C before severe annealing occurs leading to graphitization,38 and therefore the organic grains in comet Halley were not exposed to such high temperatures). The same holds for cometary ice. During an impact, a fraction of the very low-density cometary dust could have been ablated off the surface of the comet and land gently on Earth, depositing the cometary organics relatively intact.49 Hopefully, the ongoing analysis of the Stardust samples, and the Rosetta mission, will contribute to determine the composition of organic matter in comets.

Acknowledgments G.M.M.C. was supported by a Marie Curie Individual Fellowship from the European Union and a Ram´ on y Cajal research contract from the MCYT. References 1. E. Dartois and L. d’Hendecourt, Astron. Astrophys. 365 (2001) 144. 2. E. L. Gibb, D. C. B. Whittet and J. E. Chiar, Astrophys. J. 558 (2001) 702.

166

G. M. Mu˜ noz Caro and E. Dartois

3. W. A. Schutte and R. K. Khanna, Astron. Astrophys. 398 (2003) 1049. 4. V. K. Agarwal, W. A. Schutte, J. M. Greenberg et al., Origins Life Evol. Biosphere 16 (1985) 21. 5. M. P. Bernstein, S. A. Sandford, L. J. Allamandola, S. Chang and M. A. Scharberg, Astrophys. J. 454 (1995) 327. 6. M. P. Bernstein, J. P. Dworkin, S. A. Sandford, G. W. Cooper and L. J. Allamandola, Nature 416 (2002) 401. 7. R. Briggs, G. Ertem, J. P. Ferris et al., Origins Life Evol. Biosphere 22 (1992) 287. 8. U. J. Meierhenrich, G. M. Mu˜ noz Caro, W. A. Schutte, W. H.-P. Thiemann, B. Barbier and A. Brack, Chem. Eur. J. 11 (2005) 4895. 9. G. M. Mu˜ noz Caro U. J. Meierhenrich, W. A. Schutte et al., Nature 416 (2002) 403. 10. G. M. Mu˜ noz Caro and W. A. Schutte, Astron. Astrophys. 412 (2003) 121. 11. A. C. Ferrari and J. Robertson, Phys. Rev. B 61(20) (2000) 14095. 12. W. Hagen, L. J. Allamandola and J. M. Greenberg, Astrophys. J. Suppl. Ser. 65 (1979) 215. 13. H. Cottin, C. Szopa and M. H. Moore, Astrophys. J. 561 (2001) L139. 14. R. I. Kaiser and K. Roessler, Astrophys. J. 503 (1998) 959. 15. G. Strazzulla, G. A. Baratta and M. E. Palumbo, Spectrochim. Acta A 57 (2001) 825. 16. G. M. Mu˜ noz Caro and E. Dartois, in prep. 17. G. M. Mu˜ noz Caro, U. Meierhenrich, W. A. Schutte, W. H.-P. Thiemann and J. M. Greenberg, Astron. Astrophys. 413 (2004) 209. 18. E. Dartois, G. M. Mu˜ noz Caro, D. Deboffle, G. Montagnac and L. d’Hendecourt, Astron. Astrophys. 432 (2005) 895. 19. E. Dartois, G. M. Mu˜ noz Caro, D. Deboffle, and L. d’Hendecourt, Astron. Astrophys. 423 (2004) L33. 20. J. M. Greenberg, in Comets, ed. L. L. Wilkening (University of Arizona Press, Tucson, 1982), p. 131. 21. W. M. Irvine, F. P. Schloerb, J. Crovisier, B. Fegley Jr and M. J. Mumma, in Protostars and Planets IV, eds. V. Mannings, A. P. Boss and S. S. Russell (University of Arizona Press, Tucson, 2000), p. 1159. 22. J. Crovisier, K. Leech, D. Bockel´ee-Morvan et al., Science 275 (1997) 1904. 23. D. E. Brownlee and Stardust Team, AAS DPS Meeting no. 38, abs. no. 23.02 (2006). 24. D. Bockel´ee-Morvan, D. Gautier, F. Hersant, J.-M. Hur´e and F. Robert, Astron. Astrophys. 384 (2002) 1107. 25. F. H. Shu, H. Shang, M. Gounelle, A. E. Glassgold and T. Lee, Astrophys. J. 548 (2001) 1029. 26. M. N. Fomenkova, Space Sci. Rev. 90 (1999) 109. 27. M. N. Fomenkova, S. Chang and L. M. Mukhin, Geochim. Cosmochim. Acta 58(20) (1994) 4503. 28. J. Kissel and F. R. Krueger, Nature 326 (1987) 755.

Search for Solid O- and N-Rich Organic Matter

167

29. J. Kissel, F. R. Krueger, J. Sil´en and B. C. Clark, Science 304 (Issue 5678) (2004) 1774. 30. J. R. Cronin and S. Chang, in The Chemistry of Life’s Origins, eds. J. M. Greenberg et al. (Kluwer, Dordrecht, 1993), p. 209. 31. M. A. Sephton, Astron. Geophys. 45(2) (2004) 2.08. 32. G. D. Cody, C. M. O’D. Alexander and F. Tera Geochim. Cosmochim. Acta 66(10) (2002) 1851. 33. D. E. Brownlee, in Cosmic Dust, ed. J. A. M. McDonnell (John Wiley, New York, 1978), p. 295. 34. L. S. Schramm, D. E. Brownlee and M. M. Wheelock, Meteoritics 24 (1989) 99. 35. G. J. Flynn, L. P. Keller, C. Jacobsen and S. Wirick, Adv. Space Res. 33 (2004) 57. 36. G. Matrajt, G. M. Mu˜ noz Caro, E. Dartois et al., Astron. Astrophys. 433 (2005) 979. 37. L. P. Keller, S. Messenger, G. J. Flynn et al., Geochim. Cosmochim. Acta 68(11) (2004) 2577. 38. G. M. Mu˜ noz Caro, G. Matrajt, E. Dartois et al., Astron. Astrophys. 459 (2006) 147. 39. G. M. Mu˜ noz Caro and J. Mart´ınez Fr´ıas, in Dust in Planetary Systems 2005 Proceedings Book, Kauai, Hawaii, September 26–30, 2005 (2006). 40. J. R. Cronin, S. Pizzarello and D. P. Cruikshank, in Meteorites and the Early Solar System, eds. J. F. Kerridge and M. S. Matthews (University of Arizona Press, Tucson, 1988), p. 819. 41. U. J. Meierhenrich, G. M. Mu˜ noz Caro, J. H. Bredeh¨ oft, E. K. Jessberger and W. H.-P. Thiemann, Proc. Natl. Acad. Sci. USA 101(25) (2004) 9182. 42. T. C. Chamberlin and R. T. Chamberlin, Science 28 (1908) 897. 43. J. Or´ o, Nature 190 (1961) 389. 44. K. E. Edgeworth, Mar. Not. R. Astron. Soc. 109 (1949) 600. 45. G. P. Kuiper, in Astrophysics, ed. J. A. Hynek (McGraw-Hill, New York, 1951), p. 357. 46. C. Chyba and C. Sagan, Nature 355 (1992) 125. 47. N. Dauphas and B. Marty, J. Geophys. Res. 107(E12) (2002) 5129. 48. J. M. Greenberg, in The Chemistry of Life’s Origins, eds. J. M. Greenberg and V. Pirronello (Kluwer, Dordrecht, 1993), p. 195. 49. J. M. Greenberg and A. Li, in Chemical Evolution: Physics of the Origin and Evolution of Life, eds. J. Chela-Flores and F. Raulin (Kluwer, Dordrecht, 1966), p. 51.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


BALLOON-BORNE TELESCOPE SYSTEM FOR OPTICAL REMOTE SENSING OF PLANETARY ATMOSPHERES AND PLASMAS MAKOTO TAGUCHI∗ , KAZUYA YOSHIDA† , HIROKI NAKANISHI† , YASUHIRO SHOJI† , KOHEI KAWASAKI† , JUNICHI SHIMASAKI† , YUKIHIRO TAKAHASHI‡ , JUN YOSHIDA‡ , DAISUKE TAMURA‡ and TAKESHI SAKANOI‡ ∗National Institute of Polar Research, 1-9-10, Kaga Itabashi-ku, Tokyo 173-8515, Japan [email protected] †Graduate

School of Engineering, Tohoku University Aramaki, Aoba-ku, Sendai 980-8579, Japan

‡Graduate

School of Science, Tohoku University Aramaki, Aoba-ku, Sendai 980-8578, Japan

This paper reports on the ongoing development of a balloon-borne telescope system for remote sensing of planetary atmospheres and plasmas. In this system, a Schmidt–Cassegrain telescope with a 300-mm clear aperture is mounted on a gondola whose attitude is controlled by control moment gyros, an active decoupling motor, and a Sun sensor. The gondola can float in the stratosphere for periods in excess of 1 week. A pointing stability of 10 arcsec/s will be achieved via the cooperative operation of the following three-stage pointing devices: a gondola-attitude control system, two-axis telescope gimbals for coarse guiding, and a tip/tilt mirror mount for guiding error correction. The first target for the system is Venus. Wind vectors in the Venusian upper atmosphere will be derived from the tracking of cloud patterns observed in the ultraviolet and near-infrared regions. An experiment designed to test the system performance is scheduled to take place in Japan during June 2007, and a long-duration flight in the Arctic is scheduled for 2008.

1. Introduction Ground-based observations of planets have proven to be a fundamental and powerful method in the study of planetary atmospheres and plasmas. However, the time-window available for observing planets using a large ground-based telescope, generally located in the mid to low latitudes, is limited to a period of less than 10 h. Opportunities for observation are also limited by the allocation of telescope-time, which is usually restricted to just several nights per experiment. In contrast, a telescope that floats in 169

170

M. Taguchi et al.

the polar stratosphere is able to continuously monitor planets for periods in excess of 24 h. The thin, clear, and stable air of the stratosphere makes it possible to observe planets under ideal conditions, free from poor weather, with excellent visual conditions and high atmospheric transmittance, especially in the X-ray and infrared regions. In the 1950s and 1960s, large balloon-borne telescope systems (Stratoscope I and II) were launched to undertake astronomical observations from the stratosphere.1 A balloonborne infrared telescope was developed by Shibai et al.2 for far-infrared spectroscopy of galactic nebula and Milky Way. These types of balloonborne telescope systems are also less expensive than large-scale groundbased telescopes or direct planetary probe missions. We are currently developing such a balloon-borne telescope system for the remote sensing of planetary atmospheres and plasmas. Our system is not as large and heavy as Stratoscope II, but it is capable of a long-duration flight. The telescope has a clear aperture of 300 mm, and planetary video images are taken by CCD cameras. The diffractionlimited spatial resolution is 0.34 and 0.75 arcsec at wavelengths of 400 and 900 nm, respectively. Thus, a pointing stability of better than 0.34 arcsec per 33 ms, which is exposure time of a video frame, or 10 arcsec/s is required. The gondola is suspended by a large plastic balloon of 100,000 m3 class at an altitude of 32 km above the ground. The balloonborne telescope will be launched in Kiruna, Sweden, and it will take 5–7 days for a transatlantic flight to North America or 2 weeks for a circumpolar flight back to Scandinavia. For over-horizon telemetry, an Iridium-based low-speed satellite communication link will be used for control and data downlink during the duration of the flight. As the gondola will always be within sunlight, panels of solar cells will be used to ensure the continuous supply of power required to keep the system alive. Observation targets include varied phenomena on almost all of the planets, e.g., the sodium tail of Mercury; lightning, airglow, and aurora in the atmospheres of Venus, Jupiter, and Saturn; escaping atmospheres of the Earth-type planets; and satellite-induced luminous events in Jupiter’s atmosphere. The first target of our project is the global dynamics of the Venusian atmosphere, which is to be investigated by detecting cloud motion in ultraviolet and near-infrared imagery. The first experimental test of the system was planned for June 2006, at the Sanriku Balloon Center (SBC), Japan; however, this experiment has been postponed until 2007 because of delays in system development. This project also represents a step toward

Balloon-Borne Telescope System for Optical Remote Sensing of Planets

171

the development of a planetary satellite telescope; an outline of this second project is also being presented at this meeting.

2. Instrument Description 2.1. Gondola and attitude-control system The design and development of sub-components of the balloon-borne telescope system began in 2003, and a flight model has been under construction since 2005. The important specifications of the system are listed in Table 1, and a block diagram of the system is shown in Fig. 1. The gondola structure consists of aluminum space frames (Fig. 2). A decoupling mechanism and a pair of control moment gyros (CMGs; Fig. 3) are the key mechanisms involved in attitude control [3]; these are mounted on top of the gondola. Heavy components such as the CMGs, pressurized cell, battery, and telescope are aligned along the vertical axis of the gondola to minimize the moment of inertia. The decoupling mechanism isolates the gondola from the balloon, which can generate a large twisting moment, and transfers the excess angular momentum of the CMGs to the balloon. The CMGs function as a torquer in the control of azimuthal attitude. Each CMG has a pair of gyros with a large angular momentum mounted on a gimbal mount; this changes the angular momentum by rotating the axis of the gyro. Compared with a Table 1.

Specifications of the balloon-borne telescope.

Telescope Type Clear Aperture F-number Mount Wavelengths Spatial resolution Detector Attitude control Attitude sensor Size Weight Moment of inertia Power consumption Up-link Down-link

Schmidt–Cassegrain 300 mm 10 Two-axis gimbals 400 and 900 nm 0.34 and 0.75 arcsec CCD video cameras CMGs and active decoupling motor Sun sensor 1 m × 1 m × 2.2 m (H) 290 kg 50 kg m2 170 W (Max. 260 W) Serial command: 1 ch Switch: 2 ch Serial data: 1 ch Analog video: 2 ch

172 M. Taguchi et al. Fig. 1. Block diagram of the proposed balloon-borne telescope system. C1, C2, and C3 are pressurized cells that maintain an internal unit atmospheric pressure to prevent high-voltage discharge and the overheating of electronic components. C4 is a waterproof cell.

Balloon-Borne Telescope System for Optical Remote Sensing of Planets

Decoupling Mechanism

Battery

173

Sun Sensor

CMGs

Solar Cell Panel

Pressurized Cell

Fig. 2.

Telescope

Drawing of the gondola part of the balloon-borne telescope system.

reaction wheel, a CMG can generate large torque over a short response time with a small mass. The azimuthal attitude of the gondola is stabilized at a constant Sun azimuthal angle using a Sun sensor, such that a solar cell panel faces the Sun. As the CMGs comprise a pair of flywheels that tilt in opposite directions, they generate zero torque around the horizontal axes. A Schmidt–Cassegrain telescope is installed at the bottom of the gondola in the shadow of the solar cell panel where it is maintained at near-constant thermal conditions. A pressurized cell (1 atm) contains PCs, a bus bridge, a high voltage power supply for a piezo-electrically controlled mirror, and digital video recorders. Exposure of these components to the thin air of the stratosphere would probably cause electrical discharge or over-heating related to a reduction in cooling efficiency. The pressurized

174

M. Taguchi et al.

Fig. 3. Photograph of the twin control moment gyros. These will be installed within a waterproof box in preparation for an ocean landing.

cells also protect electrical equipment from seawater upon landing in the ocean. The gondola is surrounded by polystyrene foam that acts variously as thermal insulation, a shock-absorber when the gondola falls onto the ocean, and a float. The azimuthal angle is detected by a Sun sensor or a geomagnetic aspectmeter (GA). In the mid-latitudes, a GA sensor is effective in determining the gondola attitude; however, in the polar regions a Sun sensor is more reliable because aurora-induced currents make the GA unreliable. A PC processes sensor output that is used to control DC motors in the decoupling mechanism and CMGs to an accuracy in azimuthal attitude of about 0.1˚. The performance of the attitude control system has been tested in a laboratory configuration, as shown in Fig. 4; test results are shown in Fig. 5. In the test, the target for attitude detection (usually the Sun) was a bright spot on a wall produced by a laser beam. After some adjustment of the feedback circuit, the test confirmed that the system is able to stabilize the gondola attitude to the required accuracy within a minute; however, the test configuration differs from that in actual experiments in terms of external perturbations and the length of the hanging wire. An outdoor experiment with a hanging wire of the same length as that used in flight configuration is planned for the near future to confirm the system performance under more realistic conditions.

Balloon-Borne Telescope System for Optical Remote Sensing of Planets

175

Fig. 4. Photograph of the gondola during an indoor test of the attitude control system. Note that shorter pillars for the gondola structure were adopted to fit the available test space.

2.2. Optical system and pointing The two-axis gimbals mount of the telescope is controlled by the same PC that controls the DC motors, guiding an object within the field-of-view of a guide telescope. The field-of-view of the telescope covers elevation angles from 0◦ to 70◦ without interference of the gondola frame or the balloon. The azimuthal solar angle of the object of interest should be greater than 25◦ to prevent exposure of the telescope to direct solar light. The telescope has a diameter of 300 mm and focal ratio of F/10. A focal extender lens is used to expand the object image by a factor of 5. After passing through the extender lens, the optical path is divided into three paths that reach three focal planes with different colors: the first has wavelengths of less than 450 nm, the second has wavelengths of 550–630 nm, and the third has wavelengths greater than 750 nm. The first and third

176

M. Taguchi et al.

Fig. 5. Results of an indoor test of gondola attitude control by the CMGs and active decoupling mechanism. The plots show temporal variations in gondola azimuthal angle (upper) and CMG gimbal tilt angle (lower) following activation of the attitude control system.

focuses are utilized for imaging ultraviolet and near-infrared, respectively, using bandpass filters and CCD video cameras. The second focus is used for detecting tracking errors. Tracking errors that are beyond the ability of the gondola attitude and gimbal mount control are detected by a positionsensitive photomultiplier tube inserted at the focal plane and corrected by the tip/tilt mirror installed in the optical system. This mirror corrects tracking errors of angle displacement of ±200 arcsec with a resolution of 0.02 arcsec and frequency up to 100 Hz. The focus can be adjusted by a motor-driven mechanism that is not automated but moves back and forth manually as commanded. The tip/tilt mirror mount and photomultiplier tube are kept in pressurized cells to avoid electrical discharge. A second PC controls the

Balloon-Borne Telescope System for Optical Remote Sensing of Planets

177

tip/tilt mirror, high-voltage power supply to the photomultiplier tube, and telescope focus. 2.3. Telemetry and command Video signals from the CCD cameras are transmitted by analog modulation telemetries to the ground for real-time monitoring, as well as being recorded by onboard digital video recorders that are recovered after landing in the ocean. Commands are up-loaded and status and house-keeping data are down-loaded by the first PC via PCM code telemetry at speeds of 2400 bps. Commands and status data for the second PC are transferred by the first PC. Two additional analog switch commands control the power switches of the PCs. The video cameras currently generate analog video signals; however, by the time of the first experiment planned for 2007, the cameras will be replaced by digital video cameras whose output signals are obtained and stored by the PC. 2.4. Weight and power Restrictions in terms of weight and power consumption are not as severe as those for a satellite sub-system; however, the speed and accuracy of attitude control, sizing of the balloon, and float capacity all depend on the gondola weight. The gondola should therefore be as light as possible. Estimates of the current weight of the system are provided in Table 2. The power consumption of the sub-components is listed in Table 3. The solar cell panel and battery sizing have been determined to ensure that they supply sufficient power. The solar cell panel fits into one of the side areas of the gondola and generates 170 W at a solar incident angle of 45◦ . Power is Table 2.

Weight budget (kg).

Gondola structure CMG and decoupling motor Telescope Solar cell panel Battery Pressurized cell Electronics Total

60 35 57 24 17 56 39 288

178

M. Taguchi et al. Table 3.

Power budget (W).

Supply Solar cell panel NiH battery Consumption PC1 PC2 CMGs Decoupling motor F/T sensor Tip/tilt mirror mount HV Preamplifier CCD cameras Telescope Heater Total

240 at the rate of 0◦ 170 at the rate of 45◦ 24 (V) 50 (Ah) Nominal

Maximum

23.5 20.0 34.3 9.3 3.2 20.0 1.5 1.0 4.8 10.0 20.0

47.0 40.0 68.5 18.5 3.2 20.0 1.5 1.0 4.8 10.0 20.0

147.6

234.5

supplied by NiH batteries at times when the solar cell panel generates little power, such as during launching and ascent through the cloudy troposphere. The NiH batteries are designed to provide the minimum required power for a period in excess of several hours.

2.5. Flight plan The gondola is stabilized at a level-flight altitude of 32 km by an auto-ballast system, although for the experiment at SBC the level-flight altitude is subject to change depending on wind speed and direction in the atmospheric layers in which the gondola moves away from and toward the launching site. The gondola can also be floated at a fixed altitude for as long as several weeks when launched in the polar regions during suitable wind condition. The flight duration depends on ballast weight and requirements for accuracy in altitude stabilization. Observations will only be undertaken during daytime. Planetary disks can be observed with high contrast against background sky brightness, as the brightness of the sky due to atmospheric scattering is just 1% of that on the ground. Following completion of the observations, the gondola will land on the ground or the sea surface. In either case, recovery of the gondola is necessary. Most of the sub-components can then be reused in the next experiment.

Balloon-Borne Telescope System for Optical Remote Sensing of Planets

179

3. Schedule of Experiments The system is still under development as of April 2006, and the first experimental flight is scheduled for June 2007 at SBC. After confirming the performance of the system in a test experiment, it will be refurbished and placed into full-scale operation in the polar regions. In terms of location and facilities, Kiruna is the most ideal launching site in the Arctic. A longduration flight in the Arctic is scheduled for 2008. Acknowledgments This research was supported by a Grant-in-Aid for Scientific Research (C: 17540426) from the Japan Society for the Promotion of Science (JSPS). We would like to thank T. Yamagami, Y. Saito, and T. Nakagawa for their helpful discussions. References 1. D. McCarthy, IEEE Trans. Aerospace Electron. Syst. AES-5(2) (1969). 2. H. Shibai, H. Okuda, T. Nakagawa, N. Yajima, T. Maihara, K. Mizutani, H. Matsuhara, Y. Kobayashi, N. Hiromoto and H. Takami, SPIE Proc. 1235 (1990). 3. N. Yajima, S. Kokaji, and S. Hashino, Report of Mechanical Engineering Laboratory, Japan, No. 135 (1986).

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


THE STRATEGIC PLAN FOR THE INTEGRATED SCIENCES AND THE DEVELOPMENT STATUS OF JAPANESE LUNAR EXPLORERS: SELENE AND Lunar-A TAKAHIRO IWATA∗ , SATOSHI TANAKA and MANABU KATO Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency 3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510, Japan ∗[email protected] SHO SASAKI National Astronomical Observatory 2-12 Hoshigaoka, Mizusawa, Ohshu, Iwate 023-0861, Japan NORIYUKI NAMIKI Department of Earth Planet Science, Kyushu University 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan

A new era of lunar explorations is coming by two Japanese missions to the Moon: SELENE and Lunar-A. SELENE will execute the global mappings of the Moon, make technical demonstrations, and acquire the lunar data for future explorations. Fifteen mission instruments on SELENE will observe chemical elements, mineralogical distributions, surface structures, surface environments, gravity fields, and images for outreaches. They will provide wide knowledge of phenomena on the Moon to elucidate its origin and evolution, and also yield information to comprehend the interplanetary space of the solar system. SELENE is at the stage of the satellite integration and environment tests for all the systems in 2006, and is scheduled to be launched in the summer in 2007. Lunar-A is a spacecraft which provides two penetrators into the lunar surface to elucidate structures and compositions of the lunar interior with seismological and heat-flow data. The final confirmation for the penetrator system is ongoing in 2006. We have examined a strategic plan for the integrated sciences of Japanese lunar exploration projects, and designed four stages for the definite accomplishment of the sciences of the Moon by: drawing two-dimensional maps, drawing three-dimensional subsurface structures, joint studies of special topics, and those of advanced topics.

181

182

T. Iwata et al.

1. Introduction The Moon is a celestial body which has an anomalously large mass and size relative to its mother planet, the Earth. Its surface and interior structure cannot, therefore, be explicated without significant effects from the coexistence with the Earth. Hitherto lunar explorations from Apollo to Lunar Prospector have suggested the origin by a giant impact and the evolution with a magma ocean. These hypotheses seem to be, however, still less exhaustive because the diversity of instruments and the coverage of observations on each exploration were insufficient to resolve those issues. Therefore, global mappings with multi-instruments are expected to verify lunar origin and evolution. The Moon–Earth system is, on the other hand, an important probe to understand the physics of the interplanetary space in our solar system. The Earth has so strong magnetosphere that there are interactions with the solar wind. It is helpful to observe ionic and magnetic phenomena at the lunar orbit to comprehend the physics of the ionization activities around the Earth. Moreover, the Moon is recently watched with keen interests as an indispensable base for the manned explorations of our solar system. It is, therefore, expected that data of lunar elements, materials, and environments will produce beneficial knowledge for future lunar utilizations. Under these backgrounds, Japan has started two mission projects for lunar scientific explorations: SELENE and Lunar-A. SELENE is a lunar explorer which will execute the global mappings of the Moon, make technical demonstrations, and acquire the lunar data for future explorations.1 Using 15 mission instruments, we will shed light on to figure out the origin and evolution of the Moon. Lunar-A is a lunar probe which provides two penetrators2 into the lunar surface to elucidate structures and compositions of the lunar interior with seismological and heat-flow data. In this paper, we show the status and the strategy of the integrated sciences of these missions.

2. Mission Outline and Status 2.1. SELENE SELENE is the first Japanese large explorer to the Moon which will be launched by the H-IIA launch vehicle. The total mass of the spacecraft just separated from the rocket measures about 2.9 metric tons. SELENE Main Orbiter will be injected into a lunar orbit after the lunar-transfer phasing orbit with two revolutions. It will separate two small sub-satellites: the

Integrated Sciences and Development Status of Japanese Lunar Explorers

183

Relay Satellite (Rstar) into the elliptical orbit of 100–2400 km in altitudes and the VLBI Radio Satellite (Vstar) into that of 100–800 km. The Main Orbiter will then be maneuvered into the circular orbit of 100 km to make observation for 1 year. Rstar and Vstar are spin-stabilized satellites which have specific purposes for global mappings of the lunar gravity field.3 Figure 1 shows the configurations of these three spacecrafts of SELENE, and Table 1 summarizes characteristics of them. Fifteen mission instruments on SELENE and their main purposes and characteristics are shown in Table 2. Among them, components of the Relay Satellite Transponder (RSAT) are installed on Rstar and Main Orbiter, and those of the Differential VLBI Radio Sources (VRAD) are on Rstar and Vstar. The Radio Science (RS) does not have its own component but utilize signals of VRAD. All other instruments are equipped with Main Orbiter. A cross dipole antenna of the Lunar Radar Sounder (LRS) and a boom of the Lunar Magnetometer (LMAG) are displayed in Fig. 1. SELENE is at the stage of the satellite integration and environment tests for all the systems in 2006. It will be transported to the Tanegashima Space Center in the spring and is scheduled to be launched in the summer in 2007.

2.2. Lunar-A Lunar-A is an explorer to elucidate the lunar interior using penetrators. Figure 1 (lower left) shows the configuration of Lunar-A when it separates LMAG boom

SELENE Rstar/Vstar 50cm

LRS antenna

penetrator

2m

Lunar-A

SELENE Main Orbiter 1m

Fig. 1. Configurations of SELENE Main Orbiter on-orbit (right), SELENE Subsatellite; common to Rstar and Vstar (upper left), and Lunar-A separating a penetrator (lower left).

184

T. Iwata et al. Table 1.

Characteristics of SELENE and Lunar-A.

SELENE Orbits (altitude) Main Orbiter Rstar Vstar

100 ± 30 [km], i = 90◦ 100 to 2400 [km], i = 90◦ ; initial orbit 100 to 800 [km], i = 90◦ ; initial orbit

Mass Main Orbiter Rstar Vstar

2885 [kg]; wet 45 [kg] 45 [kg]

Launch vehicle

H-IIA

Lunar-A Orbit (altitude) Spacecraft Position Penetrators

200 ± 80 [km], i = 20◦ ; orbit for data relay

Mass Space Craft Penetrator Launch vehicle

1) Apollo XII site 2) antipode of (1) 540 [kg]; wet 45 [kg] M-V

a penetrator, and Table 1 presents its basic characteristics. The mass of the spacecraft and the penetrator is 540 and 45 kg, respectively. The spacecraft has been designed to be launched by the M-V launch vehicle, and injected into a lunar orbit after swing by the Moon. Then, penetrators will be separated at the perigee on the lunar elliptical orbit of 45–200 km in altitudes and decelerated to free-fall at 25 km. Each penetrator is controlled to come to a halt at 1–3 m below the lunar surface at the candidate area of Apollo XII site and its antipode. Each of them equips with two seismometers and thermal sensors (Table 2), which will make observations for 1 year. Data will be relayed via the spacecraft on the circular orbit of 200 km. The final confirmations for the penetrator system are ongoing in 2006. By the accomplishment of technologies for penetrators, it is expected to be widely applicable for future missions toward the Moon and planets. 3. Strategy for Sciences 3.1. Scenarios from individual to integrated sciences Table 2 describes a list of mission instruments of SELENE and Lunar-A, and their main purposes and characteristics. Fifteen mission instruments on

Integrated Sciences and Development Status of Japanese Lunar Explorers Table 2.

185

Mission instruments of SELENE and Lunar-A.

Instruments SELENE Chemical elements X-ray Spectrometer (XRS) Gamma-ray Spectrometer (GRS) Mineralogy Multi-band Imager (MI) Spectral Profiler (SP) Surface structure Terrain Camera (TC) Lunar Radar Sounder (LRS) Laser Altimeter (LALT) Gravity field Differential VLBI Radio Source (VRAD) Relay Satellite Transponder (RSAT) Surface environment Lunar Magnetometer (LMAG) Charged Particle Spectrometer (CPS) Plasma Imager (PACE) Radio Science (RS) Plasma Imager (UPI) Imaging High Definition Television Camera (HDTV) Lunar-A Seismometer Thermal sensor

Main purpose or characteristics

Global mapping of Al, Si, Mg, Fe distributions Global mapping of U, Th, K, major elements distributions UV–VIS–NIR CCD and InGaAs imager Continuous spectral profile High-resolution stereo camera Mapping of subsurface structures using active sounding Nd:YAG laser altimeter Differential VLBI observations from ground stations Far-side gravimetry using four-way range rate measurement via Relay Satellite Magnetic field measurements using flux-gate type magnetometer Measurements of high-energy particles Charged particle energy, angle, and composition measurements Detection of the tenuous lunar ionosphere using S- and X-band carriers Observations of terrestrial plasmasphere from lunar orbit Imaging of the Moon and the Earth for outreach Seismometry for lunar interior Heat-flow measurements for lunar interior

SELENE will provide not only wide knowledge of phenomena on the Moon to elucidate its origin and evolution but also information to comprehend the interplanetary space of the solar system. They are classified into six subgroups by their purposes. Global maps of chemical elements are obtained by the X-ray Spectrometer (XRS), and the Gamma-ray Spectrometer (GRS), mineralogical maps by the Multi-band Imager (MI) and the Spectral Profiler (SP), surface structure maps by the Terrain Camera (TC), LRS,

186

T. Iwata et al.

and the Laser Altimeter (LALT), and the gravity field maps by VRAD and RSAT. Surface environments of the Moon are observed by LMAG, the Charged Particle Spectrometer (CPS), the Plasma Analyzer (PACE), RS, and the Plasma Imager (UPI). Images toward the Earth and the Moon are also taken for popularizations by the High Definition Television Camera (HDTV). Instruments in the Lunar-A penetrators are for seismometry and heat-flow experiments, which resolve the physical parameters of the lunar interior. Figure 2 shows the scenarios from the individual scientific theme of each instrument to the integrated sciences, and finally to the scientific goal to elucidate the origin and the evolution of the Moon and to comprehend our solar system. Among them, the scenarios for the integrated sciences of elemental abundances, mineral compositions, geologic structures, gravity anomalies, and magnetic anomalies obtained by the instruments on SELENE are more characteristic and intricate than those of other lunar missions. We, therefore, examined a strategy for them as described in the next section.

SELENE Elemental Abundance (XRS, GRS)

Lunar Chemical Constituents

Mineralogical Composition

(MI, SP)

Lunar Interior Structure

Geological Feature (LRS, TC, LALT)

Dichotomy of Nearand Far-side

Global Gravity Field (RSAT, VRAD)

Differentiation in Magma Ocean

Electromagnetic & Particle Environment (CPS,

PACE, LMAG, RS, UPI)

Lunar-A Interior Survey (Seismometer, Thermal Sensor)

Origin of the Moon

Evolution of the Moon

Origin of Lunar Magnetic Field Lunar Tectonics Lunar &Terrestrial Environment

Science of the Solar System

Fig. 2. A schematic diagram of each scientific theme of SELENE and Lunar-A instruments to the goal for the lunar origin and evolution.

Integrated Sciences and Development Status of Japanese Lunar Explorers

187

3.2. Stages of integrated sciences of the moon by SELENE Figure 3 shows the scheme of the integrated sciences of the Moon by SELENE mission instruments. It consists of four stages as: 1. Drawing two-dimensional maps to integrate various geologic units into a coherent map. 2. Drawing three-dimensional subsurface structures beneath maria and highlands. 3. Joint studies of special topics such as mare tectonics and crustal formations. 4. Joint studies of advanced topics such as dichotomies and bulk compositions. For the first step, we will make two-dimensional maps as results of geological interpretations by integrating various geologic units on the basis of elemental abundances (from XRS and GRS), mineralogy (MI/SP), morphologic and topographic boundaries (LRS, LALT, and TC) at the surface and subsurface into a coherent map. Several lunar geologic maps have been published, however, there are no consensus regarding identifications and classifications of geologic units among these maps. This suggests that surface materials have been scattered by repeated impacts. We will, therefore, produce a coherent map to determine topographic boundaries, such as lava front in mare Imbria, by combining various data.

XRS

GRS

Elemental abundances

MI/SP Mineral composition

LRS

LALT

TC

Topography Geologic structures

RSAT/ VRAD

LMAG

Gravity anomaly

Magnetic anomaly

(1) Drawing two-dimensional maps (2) Drawing three-dimensional subsurface structures (3) Joint study of special topics (4) Joint study of advanced topics Fig. 3.

Four stages in the integrated sciences of the Moon.

188

T. Iwata et al.

For the second step, we will draw three-dimensional maps to establish subsurface structures beneath maria and highlands. The stratigraphic distributions of basaltic lava flows in the lunar maria will be observed by LRS, and examined by MI/SP. We will determine the mineralogical structures of the upper crust, the lower crust, and the upper mantle from the MI/SP spectrum. The thicknesses of layers will be estimated quantitatively by the gravity data of RSAT/VRAD, and the estimated structures will be compared with the distributions of the electric conductivity by LMAG. For the third step, we plan to study the common targets in the cooperation of several instruments on the basis of two-dimensional and three-dimensional maps such as differentiations of mare basalts. LRS, LALT, and TC will provide areas and thicknesses of each lava flow. Mineralogical maps and elemental abundances obtained by MI/SP, GRS, and GRS/CPS are expected to show differentiation of magma reservoirs. Through these processes, we elucidate the formation of the lunar crust which is defined as three major terrains: Procellarum KREEP Terrain (PKT), Feldspathic Highland Terrain (FHT), and South Pole Aitken basin (SPAT).4 Mare Serenitatis and Mare Crisium will be markedly analyzed for the exercises and calibrations of the first geologic mapping, and then for the investigations of mare tectonics. We will also concentrate on examining the North and South Pole regions for the water-ice and terrain surveys, and Orientale Basin as an archetype of multi-ring craters. At the final step, scientific achievements up to the third step will promote further investigations regarding the origin and evolution of the Moon. Bulk compositions, dichotomies, evolution of magma oceans, and the lunar thermal history are targets in this stage.

4. Summary The mission outlines and the status of Japanese lunar explorers: SELENE and Lunar-A, are summarized as follows: SELENE has 15 instruments which observe chemical elements, mineralogical distributions, surface structures, surface environments, gravity fields, and images for outreaches. The satellite integration and environment tests for the systems have been started in 2006. SELENE is scheduled to be launched in the summer in 2007. Lunar-A provides two penetrators into the lunar surface to elucidate the structures and compositions of the lunar interior with seismological and

Integrated Sciences and Development Status of Japanese Lunar Explorers

189

heat-flow data. The final confirmations for the penetrator system have been ongoing. The scenarios for the integrated sciences of the Moon obtained by the instruments on SELENE are more characteristic and intricate than those of other lunar missions. We have, therefore, designed four stages for the definite accomplishment of the integrated sciences of the Moon by: (1) drawing two-dimensional maps, (2) drawing three-dimensional subsurface structures, (3) joint studies of special topics, and (4) those of advanced topics.

Acknowledgments This paper is presented as a result of the Session “Countdown for the Lunar Exploration” in the Japan Geoscience Union Meeting 2006 which was held at the Makuhari Messe International Conference Hall on May 16 and 18, 2006. Authors are thankful to all the scientific principal investigators and participants who gave fruitful proposals and discussions.

References 1. M. Kato, Y. Takizawa, S. Sasaki and SELENE Project Team, Lunar and Planetary Science XXXVII (Lunar and Planetary Institute, Houston, 2006), p. 1233 (CD-ROM). 2. H. Mizutani, A. Fujimura, S. Tanaka, H. Shiraishi and T. Nakajima, Adv. Space Res. 31 (2003) 2315. 3. T. Iwata, T. Sasaki, T. Izumi, Y. Kono, H. Hanada, N. Kawano and F. Kikuchi, in A Window on the Future of Geodesy, Intern. Asocc. Geod. Symp., Vol. 128, ed. F. Sanso (Springer, Berlin, 2005), p. 157. 4. B. L. Jolliff, J. J. Gillis, L. A. Haskin, R. L. Korotev and M. A. Wieczorek, J. Geophys. Res. 105 (2000) 4197.

This page intentionally left blank

Advances in Geosciences Vol. 7: Planetary Science (2006) Eds. Anil Bhardwaj et al. c World Scientific Publishing Company


FROM NUCLEAR BLASTS TO COSMIC BOMBARDMENT KERAN O’BRIEN Department of Physics and Astronomy Northern Arizona University, Flagstaff, Az 86011, USA keran.o’[email protected]

Radiation protection has evolved from pen and pencil studies using tables of cross sections and of mathematical function to large and complex codes written and maintained by highly skilled teams. The author’s pilgrimage through this process; from his pen and pencil days while on Eniwetak Atoll in 1956 to spherical-harmonics transport codes, the use of discrete ordinate and Monte Carlo codes to an analytical transport code for the calculation of cosmicray transport through solar-system atmospheres and finally to a Monte Carlo code to treat cosmic-ray transport through the heliosphere will be described. The application of these calculations include the radiation from radioactive fallout, beta-ray transport, accelerator shielding, hospital physics, cosmic-ray ionization, cosmogenic isotope production, the radiation dose to air crews and space crews, and cosmic-ray fluxes to space craft. Some examples of the results of these calculations will be given.

1. Eniwetak 1.1. Prologue The events which I relate in this little tale of my participation in the Operation Redwing nuclear weapons test series happened 50 years ago, in 1956. Much has faded from my memory, and as happens with memory, some of what I think I remember may be distorted or combined with other events that took place at other times. Friends have urged me to set this story down before the fading and distortion of my memory progresses any further. Fortunately, a study of the effects of Operation Redwing became available in 19971 and I have relied upon it for many details of time and place. At the time this story began, Len Solon, who was my boss at the Health and Safety Laboratory urged me to take notes and to write up this 191

192

K. O’Brien

adventure. I did not follow his advice. It was harder to do then without the resources of computers and word processors, I was much younger and less reflective, and finally the adventure seemed to leave little room in my life to take notes at the time it was taking place. There was no privacy and almost no opportunity to be alone on Eniwetak where I was situated. When the adventure was over, I settled into my routine life without much thought of writing it up. A contributing factor, perhaps, was the presence in the lab of so many who had visited the Pacific Proving Grounds more than once, making my adventure seem rather ordinary and indeed rather less than spectacular by comparison with what so many others of my acquaintance had experienced. Despite all that, the numbers of men (and they were almost all men, perhaps they were all men) who have had this experience were never very large, and the passage of time has severely reduced those numbers. I think my friends were right; it is time and past time to set down what I saw for others to see. 1.2. In the lab Except for a brief lacuna, I had been working for the Health and Safety Laboratory since 1953. The Health and Safety Laboratory was located in New York City, New York in the United States, at the site where Lincoln Center in Manhattan is now. I had spent a summer between my junior and senior year at Fordham University at the lab which at that time was the Health and Safety Division of the New York Operations Office of the U.S. Atomic Energy Commission. I had returned the year before, 1955, from a disastrous year in graduate school at Carnegie Tech where I had once again proven I was uneducable. The laboratory was gearing up for Operation Redwing where the lab’s duty was to monitor the radiation intensities on inhabited Pacific Islands in the path of weapons fallout and to estimate the fraction of the fission product yield that entered the stratosphere as opposed to the faction that remained in the troposphere. Both missions were to be accomplished by measurements of radiation quantities on the ground and in the water by measurement of radiation intensities using aircraft-borne instrumentation. My job would be to relate quantities of radioactivity on the ground and in the water to radiation intensities as a function of altitude in the air above. During a previous test in the Pacific, Operation Castle in 1954, the laboratory had put together Project Dumbo comprising polystyrene rafts

Nuclear Blasts to Cosmic Bombardment

193

with instruments and telemeters which they had dropped into the Pacific prior to test shots. The hope was to fly over them and record the radiation intensity data broadcast by the telemeters. The radiation intensity was to be related to the quantity of radioactivity collected on the known area of the raft. I remember having seen the rafts 2 years before. Project Dumbo was a failure for two reasons: Most of the rafts could not be found again after a weapons test, and the radiation levels from the radioactivity in the surrounding ocean were very much greater than the radiation levels from whatever rafts could be found, making the broadcast data completely useless. During Redwing, the ocean itself would be used as the collector, and the radiation from the sea would be used to determine the quantity of radioactivity it contained, requiring only that the depth of the water occupied by the radioactivity be determined. 1.3. In transit Late in May of 1956, I traveled from New York to San Francisco on a so-called Jet-Stream DC-7. I was carrying orders which I presented to the airline in lieu of a ticket. All my necessities were stowed in a duffel bag — an idea based on the experience of others who had made this trip. I have never carried anything else so clumsy and so hard to manage. I never did it again. I remember an amusing incident on the flight. In the days before jet aircraft, cross-country flights were much longer than they are today, and we flew first class. Meals during these long flights were frequently good, and instead of peanuts, we got macadamia nuts. During the meal on this flight, I bit into a cherry tomato which spattered all over my shirt. A lovely young stewardess came over and helped me mop up. I can remember her saying “I’m enjoying this!” I was too, but shyness paralyzed my tongue. I was 24 years old at the time, but I was a very young 24. The San Francisco Airport terminal was a white building in Greek Revival style above a broad flight of steps falling down from the colonnade. There was a booth on the right-hand side of the steps that sold bus tickets. I bought a ticket for a bus that went to the Greyhound bus station in San Francisco. It was late and I was tired. The Greyhound bus to Travis Air Force Base passed though a number of northern California towns. I can remember the bus driver calling out “San Rafael (San ra fell’),” and “Vallejo (Va lay’ ho),” but I was dozing most of the way.

194

K. O’Brien

At Travis, I produced my orders and they put me in a room in the BOQ, the Bachelors’ Officers’ Quarters. I should observe that since we interacted with the military, we had to be given a fictitious military rank so that they could fit us into their system which, naturally enough, was based on rank. My fictitious rank was lieutenant colonel. I am sure this was far above my real importance. I left Travis the next day on MATS, the Military Air Transport Service. The aircraft was a Boeing Stratoliner. We, the passengers, were seated in what looked like a big brown room on camp chairs, all facing aft. Windows were small and not at all synchronized with the seat rows. I could not look out. I remember without conviction that it was a 12-h flight. We landed at Hickam Air Force Base on Oahu. Hickam shares facilities with the International Airport, and for some purposes, the facilities are indistinguishable. Once again, I was assigned a room in the BOQ. It was several days before a flight was available to take me on to Kwajalein and Eniwetak. Both big B-36s and the giant cargo aircraft, C-120s, I think, flew overhead and landed. The B-36s were a hybrid with both jet and piston engines, four of one and six of the other. They were designed to fly at great heights and carry nuclear weapons. The sound of their engines was unique, an especially deep-throated rumble. All the aircraft, after landing, would follow a jeep with a sign on the back saying “Follow Me.” The giant cargo craft creaked in time with the slow swaying of their wings as they made their way along the concrete runways following the lead jeep. I could have taken a cab to Honolulu, but I didn’t. I was unfamiliar with the city and had no idea what I could do there. I did not know of the existence of the Bishop Museum, for instance. That might have made a difference to me. I was, as I have said, a very young 24, and I am sure that a great deal to do with it. The base had a very fine restaurant, and I remember one very pleasant dinner there with a young officer. It was my first taste of mahi-mahi, and by no means my last. After several days, a flight was made available to me to Eniwetak via Kwajalein. I can remember standing in line with a lot of other young men waiting to board the flight, when Norris Bradbury, the director of Los Alamos, who had been on the aircraft walked by us. He was a strikinglooking man and had iron in his face. I remember he was wearing a business suit which looked very out-of-place in that world of military uniforms and the casual khakis of the civilian scientists and work force.

Nuclear Blasts to Cosmic Bombardment

195

The flight to Kwaj was about as long as the flight to Hawai’i and just as tedious. We sat in the big brown room, facing aft, as before. Eventually we landed at Kwajalein. It was hot and humid. We went into a large mess hall and were fed. I remember Kwaj as rocky, bare and brown. It is an atoll and we landed on one of the islets, probably Meek. Then it was back into the big brown room for the short hop to Eniwetak.

1.4. The fire on the earth We landed on Eniwetak Island late in the day. I was ferried over with some of the others the five miles to Parry Island were the scientific staff were headquartered., in a small boat known, inevitably, as “The African Queen.” The official name of Parry was Elmer. How it got to be called Parry I do not know, but that is how it was known to all of us. Its Polynesian name was Medren. Eniwetak Atoll is circular (cf. Fig. 1, from Noshkin and Robinson1 ) about 25 miles in diameter. Eniwetak Island was at the southern end of the atoll, a bit to the east of the South Channel. The Deep Entrance, just to the north of Parry Island was deep enough so that the Joint Task Force 7 (JTF-7) Fleet could pass through and anchor in the center of the lagoon. Eniwetak was one of the larger islands of the atoll and was the administrative headquarters. It had a good-sized airstrip on which we had landed. All of the islands had both English names as well as the original Polynesian names in case one of the participants in the operation had trouble remembering or saying the local name; the upshot was that we had to remember both names. An atoll is a reef sitting like a crown on a drowned seamount. The small islands were upward extensions of the reef that had weathered and produced soil and hence vegetation: trees, plants, and flowers. Seaward, there was no sloping beach. At the end of the reef, the sea-bottom dropped into the abyss. At the end of the reef, in other words, one stood on the edge of a mountain, atop a cliff which extended down to the very bottom of the ocean, thousands of feet below. Those great sailors, the Polynesians, had colonized this vast ocean, occupying essentially all the habitable islands, living as farmers, fishermen, traders and, occasionally, conquerors. The inhabitants of Eniwetak had been relocated to Meek Island of the Kwajalein Atoll to make way for the weapons tests.

196

K. O’Brien

Fig. 1. Islands of Enewetak Atoll with Marshallese and English code names. Locations of major nuclear craters are indicated.

Parry was on the south–southeast of the atoll. It was about a mile and a half long, in the north–south direction, and about half a mile wide in the east–west direction. The U.S. Army Corp of Engineers had made it habitable for us by bulldozing it flat and knocking down every single tree. We could look to the north to Japtan, a much smaller island that was left intact for “R&R” and see how it must have looked prior to the coming of the Corp of Engineers. Japtan was covered with a bamboo forest. About 10 years ago, when I stood in a stand of Arizona’s native bamboo in the neighborhood of the Kofa mountains, the scene was brought back to me vividly. I was assigned a bunk in a long low aluminum building. The bunks were stacked three high. The building had no glass in the windows, which were just square openings in the walls. There was absolutely no privacy. Even the toilets had no doors. I never got used to that. I was tired and disoriented. I got into a conversation with a roommate about high-fidelity audio units. I had just put together a 20-W Scott audio

Nuclear Blasts to Cosmic Bombardment

197

amplifier with a Klipsch horn and a University multicellular horn tweeter, and I couldn’t remember the brand name of the horn, even though it was new and the pride of my system. It couldn’t have been jet lag. I had made the whole trip in piston aircraft. We were wakened in the early dawn. Still confused I struggled outside with everyone else and discovered that we were being awakened to view a nuclear weapons test. It was known as Zuni and was to be detonated on Bikini, 180 nautical miles to the east. A tower stood behind us, a green light flashing on top. The sea lay before us. As there was no sloping beach, there were no big waves, no “combers” rising up and crashing on the shore. There was only a low sloshing as the ocean’s waves moved in and out of the irregular coral pavement. The green light meant that the test was still on. It if were replaced by a red light, then the test was off. Just before 6 o’clock, a bright light leaped up from the eastern horizon. To our dark-adapted eyes it was as though we were in a brief daylight. I looked at the sharply etched scene, and then, as suddenly, it was dark. Back in my bunk, I calculated that the sound should reach us in about 20 min. So it did, but to my surprise there were several booms, not just one. There must have been some echoing between sea and sky, or perhaps the acoustic shock transmitted to the earth had been re-emitted at other locations, reaching us at other times over different paths. Later that morning several people from the lab came looking for me. I had been assigned to the wrong barracks. I was moved to another building where the bunks were stacked only two high, and there were fewer per room. I shared a room with people from the laboratory. I had a lower bunk. Fred Wilson from the Instrumentation Division had the bunk above me. Fred and I got to be good friends during this adventure. However, when we got back to the lab, our paths almost never crossed. Many years later we became rather serious unfriends. I have no idea how that happened. At the time we had some common tastes and we were thrown together a lot. Another man I made friends with was Gerry Hamada from the lab’s Analytical Branch. Gerry and I both like classical music and were serious about science so we had things to talk about. Gerry left the lab for a commercial firm some years afterward. The sun was quite powerful at that latitude and we were warned about overexposure to the sun. The protocol recommended was to wear long pants and long-sleeved shirts until we became accustomed to the sun. I did just that and suffered no problems, although when I was in Hawai’i, my face got

198

K. O’Brien

badly sun burnt. Eventually I wore no more than bathing trunks and rubber sandals of the Japanese type which are called, I believe, zorris. I got so used to them I wore them a lot at home too. But they afforded no protection to the toes, and I, too often, smacked my toes against the furniture, painfully, and so gave them up. The zorris were comfortable and practical on the island. Wet feet and wet footgear were no problem. They would quickly dry. Later, much later, I was surprised that hats of some sort were not recommended. But we were all young and had plenty of hair on our heads. I suppose, though I do not remember, that the older guys knew enough to wear ball caps. A ball cap might have saved my face from sunburn. A young red-headed lad from the Instrumentation Division, Jerry Coughlin by name, chose to ignore all the cautions and got very badly sun burnt. He had to be sent back to New York. The day-to-day weather was uniform. It rained every afternoon at about 3 o’clock. The weather was warm and the constant sea-breeze made it quite comfortable, especially dressed lightly in bathing shorts and sandals. It meant nothing to get caught out in the rain. The rain was warm and the sun after the rain dried everything quickly. During my stay, I saw many nuclear explosions, most of them in the tens of kiloton range and detonated locally in the atoll. Some of the people on Eniwetak had become so blas´e that they didn’t go out to watch the shots. I was enough impressed with the importance of the events I was witnessing that I went out to watch each one. Of all of them, I remember only two with any distinctness. One of them was Zuni, the shot that was detonated the morning after I arrived on Eniwetak, and the other was Pequod, which, if I remember correctly, was detonated sometime after the midpoint of my stay on Eniwetak. I had spent the previous evening at the Officers’ Club. There were only two forms of entertainment on the island. One was the movies and the other, the Officers’ Club. I had become a bit too fond of cr`eme de menthe and had drunk enough of it the night before so that I had a head in the morning. Pequod was a barge shot, a thermonuclear device to be detonated over the site of Elugelab, a small island at the north of the atoll that had been essentially destroyed by Mike Shot during Operation Ivy in 1952. It was about 20 nautical miles to the northwest of Parry Island. In case the weather people had erred, or the winds suddenly changed, provisions for evacuating the island had been made.

Nuclear Blasts to Cosmic Bombardment

199

We walked down to the lagoon shore a little before 6 o’clock in the morning and sat on the sand. Very dark glasses were handed out so that we could look straight at the fireball. I hazarded a brief glance at the sun through them and it looked like a dull purple disk. I noticed that the water tower was heavily braced and tied down with wire, in case the shock might topple it. The countdown began about 15 min before the shot. “This is Manhunt,” it began. “It is 15 min before zero time.” “Manhunt” warned against looking directly at ground zero unless wearing the dark glasses. Ground zero was “down the beach, toward Yvonne (Yvonne was, of course, the English name; Yvonne was also Runit).” The countdown was in 5-min intervals down to 5 mins, then at 1-min intervals, then at 1 min, “30 s, 15 s, 10 s, 5 s, 4 s, 3 s, 2 s, 1 s, zero time.” At zero time a brilliant white point appeared “down the beach.” I could feel the radiated heat on my face. The ground rocked silently. I wondered, briefly, if perhaps that was the effect of the previous night’s cr`eme de menthe, or the shock of the explosion. The sound wave must have arrived about 2 min after but I don’t remember hearing it. The point of light grew bigger and began to rise. I lifted the bottom of the dark glasses and looked at the sand below me. It was too bright to look at for more than a few seconds, so I left the glasses on and watched the fireball rise. I tried looking at the sand several times. Eventually I could I could look at the white sand without hurting my eyes, so I ventured a glance at the fireball without the dark glasses. It had risen, oh, say, 30◦ above the horizon and had swollen considerably. I was now reddish and streaked. It rose and darkened, leaving a gray column of debris behind it. Eventually, there was a huge mushroom cloud extending over our heads, largely gray in color with some pink highlights. After watching this for awhile, began to think of leaving and going back to work, when suddenly the water at the edge of the lagoon retreated with a hiss, about 3 or 4 ft. Then it surged up the beach a few feet, and retreated again, doing this about seven times. In order to determine the proper time for a shot, the weather people from Los Alamos made measurements of the winds as a function of altitude. The weather in this part of the world at this time of the year was rather steady, and, I suppose, comparatively easy to predict. I do not know what sort of synoptic data were gathered, but locally, the Los Alamos people launched small weather balloons and tracked them as they arose to construct what was called a hodogram which gave a picture of wind

200

K. O’Brien

velocity and direction aloft. They were looking for a pattern of wind shear that would cause the “stem” of the mushroom to fall one way and miss inhabited areas, and the cap to go another to facilitate diagnostics. The cap, or some part of it, would rise into the stratosphere and drift around the world. The radioactive particles that made up the cloud had a half residence time of 10 months, and were slowly falling from the stratosphere though the tropopause down to the surface of the earth. The regions that had the greatest amount of fallout on their surface were those places, as you might expect, where the rainfall was greatest. “Wet fallout” was considerably greater than “dry fallout,” even in the same location. After each shot I would enter the classified compound near the center of the island. We all wore security badges hung about our necks on a chain that indicated whether or not you were cleared to the level necessary to enter the compound. A guard would take the badge from you, hold it in his hand, glance from you to the picture on the badge and back again and then return it to you. I haven’t the faintest idea what information it was that they felt they needed to protect. In those days, the tendency was to overclassify. In the compound I worked with a William Mills. Willy Milly, as we called him, was a tall thin man with closely cropped black hair. He was, at that time, an ensign in the Public Health Service. Our job was to take the radiation levels as measured by aircraft over the islands of the Marshall Island chain and post them on a big wall map, updating them as new data came in. I met Willy Milly again at a Health Physics Society Meeting some years ago. I remembered him and, to my surprise, he remembered me. His hair is no longer black. Mine is no longer brown either — what there is left of it. My basic job, the job I came out to do, was to calculate the radiation intensity over the radioactive ocean and land. Fortunately, all the necessary references were either there, or I had brought them with me. Then I discovered that in a military outfit stationed on Eniwetak for Redwing, a company of infantry, if I remember aright, was a soldier who had a PhD, a “cosmic-ray physicist” who had been drafted. The story was that “W” as I will call him did not get along well with his mates — this hardly surprised me. W, I was told, had been assigned the same task as had I. I went to him and suggested we work together. It would shorten our labor and make error less likely. I was coolly dismissed. “When I have my results,” he said, “I will show them to you.” I can only suppose he was filled

Nuclear Blasts to Cosmic Bombardment

201

with bitterness to overflowing. I was disappointed because it was a big job. I had no one to talk to. I had to work alone and that is always dangerous. For equipment, I had a Marchant electromechanical calculator, a copy of Goldstein and Wilkins calculations of the scatter component of gamma radiation in the format known as “buildup factors,” a table of Gradstein’s gamma-ray cross section, the CRC table of integrals and a copy of Jahnke and Emde. The last three were my own property; I still possess them, battered but usable. I should say, usable, but unused, since in these days, utilities on my computers are the method of choice, rather than these venerable references. Today, one would solve the Boltzmann equation with a code, “off the shelf,” and get the answers much more quickly and more accurately than I got them. I sat in an air-conditioned trailer next to the building we shared with the University of Washington Marine Biology Laboratory, fit build-up factors to polynomials by the method of least squares using Crout’s Method (which I no longer remember) to determine the matrix of coefficients, integrated exponential functions and calculated dose rates for a range of energies and heights over the sea, and for various angles of acceptance, since it was planned to use conical collimators to control the “angle of vision” of the detectors. I then repeated this operation for radioactive material on land surfaces. The results seemed reasonable and I had concocted some “theorems” based on what I called “homogeneous medium theory” to check them at the sea surface. No such results existed for the land surface, but the formalism allowed a transformation between them so that I had a way of checking those results as well. In the end, my results were taken to the New York University Computing Center at the Courant Institute for Mathematical Studies and were cranked out for an immense range of altitudes, photon energies and conical collimator angles. The results appeared to agree well with the measurements. Word got back to me that W was not so fortunate. That information made me feel that I had just escaped from a serious accident “by the skin of my teeth.” It could well have been me. During the first part of my stay, the group from our laboratory was under the direction of Harry LeVine, director of the Instrumentation Division. While Harry and I didn’t get along well back at the lab (neither before nor after Redwing) we seemed to hit if off well in the field. During his stay, he used me as a kind of physics consultant. I would go to meetings

202

K. O’Brien

with him and supply answers. I recall one time when we needed an estimate of the decay of fission products over a period of time. I tried to calculate it using the Way–Wigner relationship at a meeting on my small pocket slide rule and got it all messed up. Somewhat later we were supplied with circular slide rules specialized to solve the Way–Wigner equation. When Harry returned to the lab he left “D” in charge. D was effectively “Second Officer” in the Instrumentation Division. He was quite young, with an authoritarian bent that never left him. Under D we worked long hours and late and quickly grew to resent his management. Perhaps we would have felt differently had he been notably talented. And had he been talented, he might not have had that authoritarian temper. Merril Eisenbud, the Laboratory Director arrived somewhat later and matters immediately improved. Merril simply stated that “If you can’t do it in eight hours, you can’t do it in twelve.” He was aware that morale in this remote area, absent much of the ordinary amenities of life, can quickly go sour, and hurt productivity. After Merril left, D tried to restore the old routine. I remember driving off in the six-by-six truck we had managed to acquire, with the rest of the crew in back, with D stumbling along angrily behind, yelling at us to come back. Merril was the Director of the Health and Safety Laboratory, with an International reputation, and had recently played an important role in the exposure of the fishermen of the Fukuru Maru to fallout, but he was utterly unpretentious. A truly great and fine man, he died a few years ago from leukemia. One Sunday he took me out in a rowboat onto the waters of the lagoon to explain to me his philosophy on the importance of field work to the laboratory, and regretting that my immediate boss, Len Solon, could not be spared to come out to Eniwetak for the experience of field work. I remember looking down through the amazingly clear waters of the lagoon at a coral head that seemed perhaps only a dozen feet below the surface. I dove for it from the rowboat and quickly found that the coral head was far deeper and out of my reach, as the water pressure quickly began to exert too much pressure on my eardrums. The water was absolutely colorless, and were it not for the difference in refractive index between it and air would have been completely invisible. This is not to say that Merril was not above taking advantage of some of the perquisites of rank. He managed to obtain one of those giant clam shells that are supposed to occasionally catch skin divers by the leg and

Nuclear Blasts to Cosmic Bombardment

203

drown them unless they could quickly hack off the caught leg with the knife they always carried with them. The clam shells would be brought up from the bottom of the lagoon and then buried in the sand until the clam inside would have been devoured by a combination of insects and bacteria, leaving it bare of organic matter, after which it could be further cleaned, packed and sent off to “The States.” Some of us tried that with pieces of coral, with varying luck. Our only recreations were swimming in the lagoon, the evening movie, and drinking in the Officers’ Club to which our fictitious rank entitled us. We did a lot of the latter. And then there was eating. The Navy was responsible for the mess. The Navy has a reputation for a good mess. This was certainly the case on Parry Island. The big meal of the day often included steak, and the steak was good, often tenderloin. I gained a good deal of weight while on Parry. There was not, after all, much opportunity for exercise, except swimming. Food was another important aspect of what recreation there was. The beach was of white coral sand. The sand, much lighter than silica, would stick to anything wet, damp or dry. The beach was steep. At one step you were at the edge of the water, at the second step you were thigh deep. At the third step you were chest deep. At the fourth step you were swimming. I enjoyed swimming there, but it was not always possible. A Marine guard always stood watch in case there was any trouble, though what he would do if someone were drowning I have no idea. The helicopter pilots kept watch over the water, as clear as glass was, to guard against the presence of large predators, usually sharks, in the lagoon. One day, I went down to the beach. No one was in sight, not even the Marine guard. I took three steps and was swimming in the beautiful water when the guard hurried out of his little shack and called me back in an agitated voice. Apparently the helicopter pilots had warned of sharks in the lagoon. He should have been on duty on the breakwater to “stage left” of the beach, but he had taken a break just before I came along. Another reason that occasionally closed of swimming was the Escherichia coli count. The JTF-7 fleet was anchored in the center of the lagoon. The water, whatever the E. coli count, was always transparently clear. One day, Admiral R. Ball Hanlon, the admiral in charge of JTF-7, came to the beach to swim. Like any massive primary in this solar system, he came with a large number of satellites. One of the satellites, a Navy captain,

204

K. O’Brien

took a half step into the water. Before he could take another step, a large fish rushed up, jaws extended. The captain was able to recover dry land before his leg was seized. As we shared a building with the University of Washington Marine Biology Lab, we soon heard of the account. I was outside when Admiral Lewis Strauss, then the Chair of the Atomic Energy Commission came up and asked me where the Marine Biology Lab could be found. I heard all about the adventure from others, and watched as Ed Held and Lauren Donaldson pored over reference books trying to find something that looked like the monster that tried to get the commander of JTF-7 (starting with the captain). They found nothing that matched the description. I suspect that the combination of panic and excitement make accurate description difficult. Occasionally we went swimming with the Marine Biology group along the reef. Ed Held taught me a way to equalize the pressure on my eardrums, permitting me to “dive” down to much greater depths than I could previously. Because of the transparency of the water, and the beauty of the corals and brilliantly colored fish, it was great fun to swim and dive off the reef. Not all the fish were harmless. Stone fish and turkey fish had hollow spines that contained a toxin that was exquisitely painful. We were told that if one were to step on one, the pain would be so intense that it would result in collapse and if the water were deep enough, drowning. While I was there, there was a medical emergency. Someone did get impaled by a turkey fish and was rescued. He was hospitalized on one of the ships of the fleet. The coral itself could cause problems. Cuts could lead to a nasty infection. As a precaution against stone fish, turkey fish and coral cuts, we always wore sandals while wading out in water on the reef. How much they would help, I am not sure, but they would afford some protection. Occasionally, I would wander out on the reef alone and look at the fish. I made an unfortunate acquaintance with a surgeon fish. I would occasionally try to catch a fish with my hand. I succeeded with a surgeon fish and found out why they are called “surgeon fish.” The dorsal fin is quite sharp and I got my fingers cut for my pains. The cut was minor, but I never again tried to catch a surgeon fish with my hands — or in any other way. The reef extended a good distance south of Parry, and it was rumored that with care, one could walk all the way to Eniwetak. It was also said to

Nuclear Blasts to Cosmic Bombardment

205

be dangerous if one was caught by the rising tide. I never walked very far in that direction. Sand sharks could often be found wandering in the waters of the reef. I would try to get between them and the sea and chase them into shallow waters. But water was their element, not mine, and they found it easy to elude me. The sand sharks were comparatively small, about 3 or 4 ft long, with black dorsal fins; they were called, colloquially, “black tips.” I was told they were not totally harmless. A marine stationed at Bikini had one tear his leg muscle off. The laborers on Eniwetak, the men who did the manual labor, were Polynesians, recruited from other Pacific islands. They would catch longusta (the so-called “lobster tails”) and roast them. The sand crabs were another delicacy. However, local weapons tests had released a good bit of radioactivity into the water, and both the longusta and the sand crabs were radioactive. It was hard to discourage the laborers from catching and eating their very usual and customary food. I remember standing on the reef with a number of these tall strong young men, a heap of longusta piled up ready to be roasted, and an earnest military radsafe guy urging them to forego their feast. When we left, the hecatomb of longusta was still intact. Whether or not the feast was foregone, I do not know, but I would guess not. There was one break in the routine. I got an opportunity to fly to Wake Island, about 700 miles to the northeast. My being there served no official purpose, it was just a chance to get away from Eniwetak for a few hours. The purpose of the flight was to deliver some item or items, I have no idea what. I had made a few friends with some of the military on site. It was a small island and one would meet the same people often. This military pilot was about to make the trip and wanted to know if I wanted to come along. Of course! I got picked up in an L19, a single-engined high-wing plane and flown over to Eniwetak — or Fred, as it was also named; a short hop of about 5 miles. I was intrigued by the approach to the Eniwetak landing strip. The pilot rose to his cruising altitude and flew almost over the landing strip and then dove steeply down for the landing. That was the standard method of all the military pilots out there. We flew to Wake on a Stratoliner. It was another case of the big brown room, but this time we were free to wander around, to look out the windows and talk to the flight crew. After more than a month’s immersion in the life

206

K. O’Brien

at Parry, Wake was a sudden shock. There were women! I had not realized how I had missed the presence of women. They all looked like angels. I wandered around, visited a few shops, but had no time to really see the island before it was time to return to Eniwetak. Speaking of L-19s, there was a smaller aircraft, an L-18. A story was told that before my arrival, an L-18 tried to land on the strip at Parry, but because the steady sea-breeze provided too much lift, the pilot had some of the ground crew run alongside the aircraft and hold the wings down. There were two other episodes that correspond to “R&R.” One of them was a picnic on Japtan. We sailed on one of the “African Queens” from Parry. As we crossed the Deep Entrance and passed suddenly into deep water, the water turned an intense and brilliant blue, like nothing I had ever seen before. Japtan was covered by a forest of bamboo, and the sight of so much green life, after so long on bulldozed Parry, was refreshing. I hated to leave. The other episode was a flight on a helicopter. We sat in the hold, next to the thick steel rotating shaft that held the blades, the shaft about a foot in diameter, looking out the open door, seat and shoulder harness fastened, of course. We could see deep into the brilliant sea; one could easily see how the helicopter pilots could see sharks in the lagoon. The encircling reef was clearly visible around the whole atoll. It was extremely noisy, and I was alarmed when we landed when I could not hear for about 5 min. 1.5. The return home There was a Post Exchange on Parry which we were, on the basis of our fictitious ranks, permitted to use. I bought a good solid suitcase to replace the duffel bag I had come out with. I kept the duffel bag, stuffing it in the suitcase, although for what I might some day want it, I couldn’t imagine. I never did use it again. I left for home a few days before Test Shot Tewa, for a stay of a little less than 2 months. I told Gerry Hamada, a native of Hawai’i, that I would meet him there and do some sight-seeing with him. We flew directly to Hickam Air Force Base without a stopover at Kwajalein. I traveled with a Public Health Service employee whose name was something like Mahlon. The aircraft’s heater’s thermostat did not work, so we would fly until the aircraft got unbearable hot, then fly until it got unbearably cold.

Nuclear Blasts to Cosmic Bombardment

207

When Mahlon and I got to Hickam we were told that they did not know when the next flight to the mainland would take place, so we took a room at the Royal Hawai’ian on Waikiki Beach. We had left our destination with PAX, the information exchange, so that we could be reached when a flight became available. I had decided not to wait for Gerry. My feet had swollen up with athlete’s foot — a result, no doubt, of wearing shoes for the first time in almost 2 months, plus the heat in the aircraft resulting from the thermostat failure in the aircraft — and I was ready to go home. I left a message with PAX for Gerry. We went swimming and it was quite a disappointment after the clear brilliant water of Eniwetak lagoon. The water was not as clear and I kept bumping my knees on the coral heads. Mahlon and I had dinner together. When we got to our room, we found that the message light on the telephone was lit. They had a flight for us, we were to leave early the next morning. I visited Honolulu much later, in 1988, when my son was married there. The change was appalling. In 1956 there were complaints that Honolulu had become commercialized and crowded. In 1988, there was no space between the hotels. I remember being astonished at the change. Hotels stood in ranks behind one another along the beach. Honolulu, in 1956, was a tranquil village by comparison. The flight back to the mainland was on a commercial jet with a contract to the military. The airline was a small one, and probably exists no longer. I noticed that the single stewardess was older than stewardesses usually were in those days. I remember her face and pleasant smile with pleasure after all these years. It was a relief to sit on a comfortable seat, facing forward, with a real cabin crewmember to provide meat and drink. I bade farewell to Mahlon in San Francisco and flew home on another DC-7.

2. Accelerator Radiation Protection 2.1. Flintlock days Around 1954, the laboratory director received a letter from the Assistant General Counsel of the Atomic Energy Commission. The Navy and the AEC owned almost all of the cyclotrons in the country, and many universities had more than one; a situation quite different from today’s, where fundamental

208

K. O’Brien

particle research is now done on a few very large machines, built at a few sites. The AEC was concerned that there was no “in-house” capability to ascertain the safety of the personnel using these machines, and the researchers appeared to have developed a significant number of cataracts. There was a reasonable fear of litigation on this account and possibly with respect to other health issues. Len Solon bought a radium–beryllium neutron source and a polyethylene-walled, ethylene-filled neutron proportional counter. The neutron counter was calibrated with the radium–beryllium source. The radium–beryllium source also produced an intense gamma-ray field which made its use quite difficult. The first survey measurements were made with this instrument and with a Juno hand-held gamma-ray ionization chamber. The results were then, of course, quite simple to interpret. There were neutron measurement and gamma-ray measurements which could be compared with the permissible levels of National Bureau of Standards Handbook 59. However, the neutron detector’s low-energy cutoff was 0.5 MeV, so we added more instruments. Ultimately, by 1958 we had eight, including several long counters, a zinc-sulfide-impregnated polymethylmethacrylate (Lucite) counter, a BF3 -filled, plastic-walled ionization chamber, a tissue-equivalent ionization chamber, a graphite-walled, CO2 -filled ionization chamber, and, a hand-held, air-filled ionization chamber (cf. Fig. 2). The radium–beryllium calibration source was replaced by a much easier to handle polonium–beryllium source, and it, in turn was replaced by a plutonium–beryllium source. By 1958 we had visited 14 accelerator sites. Two things stand out in my memory. One of them is that we gained enough experience that we could, just by looking at the shielding of a site, tell where the neutron fields would be highest, and where the thermal neutron fluxes would be highest. The other was the number of senior researchers who assured us that there was no neutron problem but who were looking at us through the thick “bottlebottom” glasses that people who had had cataract surgery wore at that time. Radiogenic cataracts were epidemic. The chief contributor to the problem was the use of deuteron beams. Neutron beams were not available at the time, so deuteron beams were used instead, exploiting the Oppenheimer–Phillips reaction to get the neutron into the target nucleus.

Nuclear Blasts to Cosmic Bombardment

Fig. 2.

209

Some of the instruments used at the University of Wisconsin in 1927. 2

2.2. Neutron spectrometry 2.2.1 Nuclear emulsion spectrometry With more instruments, the problem of interpretation grew quite complicated. All the neutron gave different results and had different energy responses, so we blamed it on a “deus ex machine,” the neutron spectrum. We invested considerable effort into developing nuclear-track emulsion spectrometry. We used 400 and 600 micron Ilford plates developed in a darkroom built for the purpose. Recoil proton track lengths were measured with a microscope. The location of each end was marked on a punched card. Pythagoras’ theorem was used to obtain the length, and range-energy tables for emulsion

210

K. O’Brien

were used to obtain the proton energies.3 Proton energies were grouped in bins and then corrected for their escape probabilities — 600 microns corresponds to a proton energy of only 0.6 MeV. In order to determine the neutron spectrum from the protons spectrum, we had to solve the Fredholm equation:

P (Ep ) =

∞

γσn,p (En , Ep )N (En )dEn

(1)

0

where P (Ep ) is the recoil proton spectrum per MeV, γ is the hydrogen ion density per cm3 , σ n,p (En , Ep ) is the n,p scattering cross section and N (En ) is the incident neutron spectrum per cm2 per MeV. We solved this equation by two means, an iterative numerical approach and by expanding the neutron spectrum in Hermite polynomials.4 Figure 3 shows some of the results we got. 2.2.2. Bonner spectrometry Returning from a vacation some time about 1970, I was stunned to find that my then boss, J. E. McLaughlin (Len Solon had left the lab to take the job of director of the New York City Office of Radiation Control) had gutted the darkroom, disposed of the microscopes and painfully machined screws that we used to measure proton-recoil track lengths and had, at a blow, terminated the nuclear emulsion neutron spectrometry program. A man woefully unsure of himself, he often went to the headquarters of the Atomic Energy Commission in Washington seeking guidance in the management of the Radiation Physics Division, of which he was director. He had gotten the notion that Washington wanted us to demonstrate progress by terminating programs, and this was his reaction. Without consulting any of us, he irreparably destroyed what was a successful and interesting program which we were able to use to help unravel the problems posed by the differing responses of our neutron detectors. We were not without resources, however. We had obtained the five polyethylene spheres that comprised the Bonner spectrometer.6 The spheres were used to moderate incoming neutrons so that they could be captured by a thermal detector at the center. With a bare thermal detector we had a six-element spectrometer. The main problem in its use was that it was a mathematically underdetermined system. The equation for the neutron

Nuclear Blasts to Cosmic Bombardment

211

Fig. 3. Measured neutron spectra from a number of accelerator sites.5 AGS, Alternating Gradient Synchrotron; CEA, Cambridge Electron Accelerator; Cos, Cosmotron; Bev, Bevatron, and PPA, Princeton-Penn Accelerator.

spectrum is again a homogenous Fredholm equation of the first kind: Bν =

∞

ϕ(E)Kν (E)dE

(2)

0

where Bν is the counting rate of the νth sphere, φ(E) is the neutron spectrum, and Kν is the energy response of the νth sphere. To calculate a detailed neutron spectrum using just six elements posed an interesting problem. Our first attempt was an iterative method similar to that used for the emulsion spectra.7 The second and most successful approach was the use of a Monte Carlo method.8

212

K. O’Brien

The indeterminateness of the solution to Eq. (2) can be expressed as follows: ∞ Bν = [ϕ(E) + ξ(E)]Kν (E)dE 0

∞ 0

Bν + ε ν =

(3)

ξ(E)Kν (E)dE = 0 ∞

[ϕ(E) + ξ(E)]Kν (E)dE

0

where εv is the counting error in the νth sphere and ξ(E) is an oscillatory term which appears because of the undeterminedness of the system. In the iterative method, we smoothed the solution between each iteration. In the Monte Carlo approach, these terms are randomly generated, so that by taking the mean of several solutions, we could remove them. Figure 4 indicates the power of this method. We took a calculated neutron spectrum, inserted it into Eq. (2) and then unfolded the resulting set of synthesized sphere counts to get the results shown in the figure. The method is quite robust, allowing good results for values of εv as large as 20%. 2.3. Transport theory During the 1960s an immense literature on transport theory appeared in the periodical literature. I had read it avidly, and indeed had successfully applied it to a problem in β-ray transport.9 I ran into trouble with out Instrumentation Division. The Instrumentation Division personnel always felt that they did all the hard work, but the other divisions got all the glory and they liked to make sure you understood how much you depended on them. In fact they could exercise an absolute veto on the work of some of the more junior staff, such as I was at the time. They made it quite plain that they were no longer interested in my accelerator radiation measurements, and so they, perforce, came to an end. But because I had begun to work with transport theory, I realized that I could get the information I needed from basic principles. Radiation transport obeys the equation that Ludwig Boltzmann wrote down in 1872:   t) = ˆi ϕi (r, E, Ω, Qij B j

 · ∇ + σi + di − ˆi = Ω B




∂ ∂Ei



Si

Nuclear Blasts to Cosmic Bombardment

213

Fig. 4. Unfolded Bonner spectrometer neutron flux spectrum (neutrons per cm2 per Me V) by the Monte Carlo method.8

  ∞     ′ → Ω)ϕ  j (r, EB , Ω,  t)  dΩ dEB σij (EB → E, Ω Qij = j

4π

(4)

E

di =



1 − β2 τi cβ

where B is the Boltzmann operator, φi is the angular flux per unit area, energy, unit time and direction, r is the location of the point of interest (such as the outside of an accelerator shield), E is the neutron energy, Ω is the particle direction t is time, σi is the collision cross section of particletype i, di is the decay probability of radioactive particle-type i, S is the

214

K. O’Brien

stopping power if the particle is charged, otherwise S is zero, Q is the “scattering-down” integral, σij is the cross section for the production of a particle of type i at a direction Ω and energy E from a collision with a particle of type j, φj is the angular flux producing secondary particles of type i and energy EB in the scattering-down process, β is the ratio of particle velocity to the velocity of light, τ is the particle mean life, and c is the velocity of light. 2.3.1. Neutron transport through accelerator shields Expanding the fluxes and cross sections of Eq. (4) in spherical harmonics, I was able to write a transport code in the P3 approximation.10 Comparing it with experimental results, indicated quite satisfactory agreement (cf. Fig. 5).11

Fig. 5. Comparison measured and theoretical dose rates in the side shielding of the Alternating Gradient Synchrotron.

Nuclear Blasts to Cosmic Bombardment

215

Using this code I did shielding studies for a number of accelerators, including the Los Alamos Meson Physics Facility, Fermilab, the Stanford Linear Accelerator and the Cambridge Electron Accelerator.10

3. Cosmic-ray Studies At this time, my friend and colleague, R. G. “Tut” Alsmiller of the Oak Ridge National Laboratory was also doing accelerator shielding studies, and because of the business-orientation of ORNL, had to charge for his work. I did not feel it fair to compete with him, because my services were free. Earlier, during the weapons-test days, Commissioner Willard Libby of radiocarbon fame had said, publicly that the fallout from weapons tests carried out on U.S. soil produced no more additional radiation to the U.S. population than living in a brick house would. Then, thinking the better of it, telephoned the director of the lab, and said “Prove it!”

3.1. Cosmic rays in the terrestrial atmosphere My colleagues, Len Solon and Wayne Lowder put together a suite of instruments and began a long career in the study of the natural background radiation in the USA. They quickly found that they needed to separate terrestrial from cosmic sources. Both their terrestrial and cosmic-ray studies became increasingly sophisticated. From large plastic bottles painted on the inside with aquadag they developed argon-filled steel-walled chambers calibrated to take into account the differences between the medium of the chambers and the surrounding atmosphere as it would affect the passage of cosmic rays through the wall and the gas. At this time, I decided to try my hand at cosmic-ray transport. After a couple of false starts, I came across a paper by Tut Alsmiller’s wife, Fran Alsmiller. Franny was a brilliant woman and an excellent mathematician. She suffered from diabetes and during one of her frequent hospital stays, decided it was a wonderful time to write a paper on high-energy radiation transport.12 I made a point of reading everything that either of the Alsmillers wrote and so pored over her paper and came across a simple equation for highenergy radiation transport, originally formulated by C. Passow.13 It proved

216

Fig. 6.

K. O’Brien

The calculated and measured ionization profile at Durham NH in 1965. 16

quite successful in treating cosmic-ray propagation in the atmosphere. It was a “straight-ahead” code, at first, which put too many particles at high altitudes and too few at intermediate depths. The accuracy of the code was greatly improved by extending it to three dimensions,14 using an approach suggested by Elliott15 and Williams16 and by incorporating a routine for transporting cosmic-ray nuclei. Its accuracy is shown in Fig. 6, where the measurements of Lowder et al.17 Solar modulation was treated in the heliocentric approximation. The geomagnetic field was treated by combining St¨ormer’s theorem with the calculated vertical cutoffs.13 The code, called “PLOTINUS” has also been used to calculate the radiation dose to flight crew. The results, show in Fig. 6 show good agreement with the measurements carried out at PTB by Schrewe18 as seen in Fig. 7.

3.2. Solar particles in the terrestrial atmosphere Fast coronal mass ejections from the sun can generate high-energy particles, mostly hydrogen, from the interplanetary medium through the mechanism known as shock acceleration.19 A study of ground-level event, or enhancement number 42 (GLE-42) which began on September 29, 1998, one of the largest in recent years, yielded world-wide dose-rates. These calculations were executed by applying

Nuclear Blasts to Cosmic Bombardment

217

Fig. 7. A comparison of effective dose rate and ambient dose equivalent with measurement on a flight from Frankfurt to Fairbanks.

the high-energy transport code to the shock-accelerated spectra incident on the earth. Local intensities in the atmosphere were obtained by comparing calculated fluxes with the measurements of ground-level neutron monitors distributed over the surface of the earth and then interpolating between them, using their geomagnetic latitude as the interpolation grid, at each altitude of interest.19 GLE-42 was highly anisotropic, as Fig. 8 clearly shows.

Fig. 8. 42.15

Solar particle effective dose rates at flight-level 350, 3 h after the onset of GLE-

218

K. O’Brien

4. Extraterrestrial Cosmic Rays 4.1. Radiation transport through the heliosphere Parker’s equation for the transport of cosmic rays through the heliosphere is20 :
 ∂D 1 ∂ + ∇ · S¯ + (P 2 P˙ D) = Q (5) ∂t P 2 ∂P   ¯ · ∇ D. S¯ = 4πP 2 C V¯ D − K

where D is the phase-space density, S is the differential current density, P is the rigidity, V is the velocity of the solar wind, C is the Compton–Getting coefficient, and K is the diffusion tensor. The source-free, one-dimensional version of Eq. (5) is20 : 


∂D 1 ∂ 1 ∂ ∂D ∂D V + 2 =0 (6) r2 κ − 2 (r2 V )P ∂r r ∂r ∂r 3r ∂r ∂P The flux, φ, is related to the rigidity and the phase-space density by: ϕi =

Ai 2 P D, Zi

(7)

where Ai and Zi are the atomic weights and atomic numbers of cosmic-ray nuclei of type i. The diffusion coefficient κ, in Eq. (6) has collapsed to a scalar, and is given by: κ = κ0 P β

(8)

1Vr 3 U

(9)

and κo is obtained from κ0 =

where r is the distance to the solar-wind termination shock, taken here as 100 AU (one AU is the earth–sun distance, 150 million km), and U is the heliocentric potential. Finally, U is obtained from14 2
P (E) ϕi (E, r = 1) = ϕi (T, r = rb ), T = E + Zi U. (10) P (T ) The flux, φi (E, r = 1) is the cosmic-ray flux of particles of type I, the flux φi (T, r = rb ) is the flux at the solar-wind transition shock.

Nuclear Blasts to Cosmic Bombardment

219

By inserting a range of values of U in Eq. (10) and using the resulting values of the fluxes given by the left-hand side of the equation and using those fluxes to calculate the response of high-latitude neutron monitors, the high-level neutron monitor counting rates can be used to determine U , and hence the scalar diffusion coefficient κ.

4.2. Cosmic rays in extraterrestrial atmospheres These methods have been applied to a number of cases. Ionization profiles have been calculated in the atmospheres of Mars (S. Tripathi, Private Communication, 2006) and Titan,21 and radiation intensity profiles in the atmosphere of Mars.22 Figure 9 shows the ionization profile calculated in the atmosphere of Titan.

4.3. Radiation shielding for a Mars mission Energetic solar-particle events represent a serious danger to the health of astronauts in interplanetary space. As solar-particle events are less frequent at solar minimum than at other phases of the solar cycle, it is anticipated that a Mars mission would be flown at solar minimum. The cosmic-ray gradient between earth and Mars is quite small23 and so the dose rate would

Fig. 9. Ionization components calculated in the atmosphere of Titan at a recent solar minimum.

220

K. O’Brien

Table 1. Radiation doses to astronauts on a two and half-year mission to Mars at a recent solar minimum spending 6 months on the Martian surface. Spacecraft hull

Earth/Mars space (rem/year)

Martian surface (rem/year)

2.5-year mission (rem)

2 g/cm2 Al, 4 g/cm2 CH2

63.75

17.17

136.1

2 g/cm2 Al, 20 g/cm2 CH2

55.02

11.51

115.8

not vary significantly with location (although it would, of course, vary with time). The dose rate through various hull thicknesses of aluminum and polyethylene was calculated at solar minimum and on the surface of Mars, assuming an 18 g/cm2 atmosphere of CO2 .22 The resulting dose to the crew, assuming that a round trip would take 2 years and that they spent 6 months on the surface of Mars is shown in Table 1. The resulting doses would be a little over the mission limit of one Sievert. The thicker shielding combined with a slightly shorter trip could, however, reduce the dose to a level below 1 Sievert.

5. Conclusion I have taken the reader on a voyage from the early pencil and paper days on a very small island in the midst of the Pacific Ocean through computer calculations of radiation transport through accelerator shields, planetary atmospheres, and spacecraft hulls. I have not had time or space to include work in other fields, such as hospital physics, Monte Carlo methods, and cosmogenic isotope production. That would make this a very long paper indeed. It is my hope that the reader finds the variety and evolution of these studies of interest.

References 1. S. Simon and R. J. Vetter, Health Phys. 73 (1997) 3. 2. W. M. Lowder, A. V. Zila and L. J. Goodman, Radiation Survey University of Wisconsin Van de Graaf Generators, (U. S. Atomic Energy Report, HASL-11, April 15–16, 1957). 3. J. C. Allred and A. H. Anderson, Laboratory Handbook of Nuclear Microscopy (Los Alamos Report LA-1510 (revised edition), 1953).

Nuclear Blasts to Cosmic Bombardment

221

4. K. O’Brien, R. Sanna, M. Alberg, J. E. McLaughlin and S. Rothenberg, Nuovo Cimento Suppl. 3 (1965) 409. 5. K. O’Brien, R. Sanna and J. E. McLaughlin, First International Conference on Accelerator Dosimetry and Experience (U. S. Atomic Energy Report CONF-651109, 1965). 6. R. L. Bramblett, R. I. Ewing, and T. W. Bonner, Nucl. Instr. Meth. 9 (1960) 1. 7. R. Sanna and K. O’Brien, Nucl. Instr. Meth. 91 (1971) 573. 8. K. O’Brien and R. Sanna, Nucl. Instr. Meth. 185 (1981) 287. 9. K. O’Brien, S. Samson, R. Sanna and J. E. McLaughlin, Nucl. Sci. Eng. 18 (1964) 90. 10. K. O’Brien and J. E. McLauglin, Nucl. Instr. Meth. 60 (1968) 129. 11. K. O’Brien, Second International Conference on Accelerator Dosimetry and Experience (U. S. Atomic Energy Report CONF-691101, Washington, D.C. 1969). 12. F. S. Alsmiller, A General Category of Soluble Nucleon-Meson Cascade Equations, (Oak Ridge National Laboratory Report ORNL-3746, 1962). 13. C. Passow, Phenomenologische Theorie zur Berechnung einer Kaskade aus schweren Teilchen (Nukleonenkaskade) in der Materie (Deutches Elektronen Synchrotron report, Notiz A 285, 1962). 14. K. O’Brien, The theory of cosmic-ray and high-energy solar-particle transport in the atmosphere, in The Natural Radiation Environment VII, Seventh International Symposium on the Natural Radiation Environment (NRE-VII), eds. J. P. McLaughlin, S. E. Simpopoulos and F. Steinh¨ausler, ISBN: 0-08-044137-8 (Elsevier: Amsterdam, Boston, Heidelberg, London, New York, Paris, San Diego, San Francisco, Singapore, Sydney, Tokyo, 2005), pp. 29–44. 15. J. P. Elliott, Proc. R. Soc. London A 228 (1955) 424. 16. M. M. R. Williams, Nukleonic 9 (1966) 305. 17. W. Lowder, P. Raft and H. Beck, in Proceedings of the National Symposium on Natural and Manmade Radiation in Space, Las Vegas, NV, ed. E. A. Warman (National Aeronautics and Space Administration Report NASA TM X-2440. 1971), pp. 908–913. 18. U. Schrewe, Radiat. Prot. Dosim. 91 (2000) 347. 19. K. O’Brien and H. H. Sauer, Adv. Space Res. 32 (2003) 73. 20. R. A. Caballero and H. Moraal, J. Geophys. Res. 109 (2004) A01101, doi:10.1029/2003JA010098. 21. G. J. Molina-Cuberos, J. J. Lopez-Moreno, R. Rodrigo, L. M. Lara and K. O’Brien, Planet. Space Sci. 47 (1999) 1347. 22. G. J. Molina Cuberos, W. Stumptner, H. Lammer, N. J. K¨omle and K. O’Brien, Icarus 154 (2001) 216. 23. K. O’Brien, Adv. Space Res. 36 (2005) 1731.