234

Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

Embed Size (px)

Citation preview

Page 1: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)
Page 2: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

A d v a n c e s i n

GeosciencesVo l u m e 7 : P l a n e t a r y S c i e n c e ( P S )

Page 3: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

This page intentionally left blankThis page intentionally left blank

Page 4: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

N E W J E R S E Y • L O N D O N • S I N G A P O R E • BE IJ ING • S H A N G H A I • H O N G K O N G • TA I P E I • C H E N N A I

World Scientific

A d v a n c e s i n

GeosciencesVo l u m e 7 : P l a n e t a r y S c i e n c e ( P S )

Editor-in-Chief

Wing-Huen IpNational Central University, Taiwan

Volume Editor-in-Chief

Anil Bhardwaj Vikram Sarabhai Space Centre, India

Page 5: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

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

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

ISBN-13 978-981-270-781-9(Set)ISBN-10 981-270-781-6 (Set)ISBN-13 978-981-270-986-8(Vol. 7)ISBN-10 981-270-986-X (Vol. 7)

Typeset by Stallion PressEmail: [email protected]

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 retrievalsystem now known or to be invented, without written permission from the Publisher.

Copyright © 2007 by World Scientific Publishing Co. Pte. Ltd.

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

Printed in Singapore.

ADVANCES IN GEOSCIENCESA 4-Volume SetVolume 7: Planetary Science (PS)

Page 6: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in fm 1

EDITORS

Editor-in-Chief: Wing-Huen Ip

Volume 6: Hydrological Science (HS)Editor-in-Chief: Namsik ParkEditors: Chunguang Cui

Eiichi NakakitaSimon TozeChulsang Yoo

Volume 7: Planetary Science (PS)Editor-in-Chief: Anil BhardwajEditors: C. Y. Robert Wu

Francois LeblancPaul HartoghYasumasa Kasaba

Volume 8: Solar Terrestrial (ST)Editor-in-Chief: Marc DuldigEditors: P. K. Manoharan

Andrew W. YauQ.-G. Zong

Volume 9: Solid Earth (SE), Ocean Science (OS) &Atmospheric Science (AS)

Editor-in-Chief: Yun-Tai ChenEditors: Hyo Choi

Jianping Gan

v

Page 7: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in fm 1

This page intentionally left blankThis page intentionally left blank

Page 8: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in fm 1

REVIEWERS

The Editors and Editor-in-Chief of Volume 7 (Planetary Science) would liketo thank the following referees who have helped review the papers publishedin this volume:

Aigen LiAlexander MedvedevAndrew NagyAnil BhardwajBing-Meng ChengCesar BertucciDavid J. McComasGeorges DurryGerad BeaudinGuillermo M. Munoz CaroHiroshi KimuraJamie ElsilaKarri MuinonenManubo KatoMartin HilchenbachMasahiko Arakawa

Miriam RengelNicolas FrayNorbert KruppPaul HartoghPradip GangopadhyayRobert WuScott BoltonSonia FornasierTai-Sone YihTakao NakagawaTakeshi ImamuraTetsuo YamamotoTom CharlockYasushi YamaguchiYuriy Shkuratov

vii

Page 9: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in fm 1

This page intentionally left blankThis page intentionally left blank

Page 10: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in fm 1

CONTENTS

Editors v

Reviewers vii

Some Similarities and Differences Between the Marsand Venus Solar Wind Interactions 1J.-G. Trotignin

Comparison of Microwave Observations of MartianTemperature and Winds with General CirculationModel Simulations 13T. Kuroda and P. Hartogh

Asteroid Compositions: Some EvidenceFrom Polarimetry 21A. Cellino, M. Di Martino, A.-C. Levasseur-Regourd,I. N. Belskaya, Ph. Bendjoya, R. Gil-Hutton

Low Energy Charged Particle Measurementby Japanese Lunar Orbiter Selene 33Y. Saito, S. Yokota, K. Asamura, T. Tanaka and T. Mukai

A Jovian Small Orbiter for Magnetospheric andAuroral Studies with the Solar-Sail Project 45Y. Kasaba, T. Takashima, H. Misawa and Jovian SmallOrbiter Sub-Working Group [with J. Kawaguchi and Solar-SailWorking Group]

Description of a New 400 MHz Bandwidth ChirpTransform Spectrometer 55L. Paganini and P. Hartogh

ix

Page 11: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in fm 1

x Contents

Formation of Alumina Nanoparticles in Plasma 69M. Kurumada and C. Kaito

Infrared Study of UV/EUV Irradiation of Naphthalenein H2O+NH3 Ice 79Y.-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, Monoxideand Dioxide Grains in Laboratory 93A. Kumamoto, M. Kurumada, Y. Kimura and C. Kaito

Destruction Yields of NH3 Produced by EUVPhotolysis of Various Mixed Cosmic Ice Analogs 101C. Y. R. Wu, T. Nguyen, D. L. Judge, H.-C. Lu, H.-K. Chenand B.-M. Cheng

Formation of CaTiO3 Crystalline Dust in Laboratory 115K. Yokoyama, Y. Kimura, O. Kido, M. Kurumada,A. Kumamoto and C. Kaito

Direct Observation of the Crystallization ofCarbon-Coated Amorphous Mg-bearing Silicate Grains 125C. Kaito, S. Sasaki, Y. Miyazaki, A. Kumamoto,M. Kurumada, K. Yokoyama, M. Saito, Y. Kimura and H. Suzuki

Relationship Between Morphology and SpectraRevealed by Difference in Magnesium Content ofSpinel Particles 133M. Saito, M. Kurumada and C. Kaito

Ionization of Polycyclic Aromatic Hydrocarbon Moleculesaround the Herbig Ae/Be Environment 143I. Sakon, T. Onaka, Y. K. Okamoto, H. Kataza,H. Kaneda and M. Honda

Search for Solid O- and N-Rich Organic Matterof Prebiotic Interest in Space 155G. M. M. Caro and E. Dartois

Page 12: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in fm 1

Contents xi

Balloon-Borne Telescope System for Optical RemoteSensing of Planetary Atmospheres and Plasmas 169M. 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 Sciencesand the Development Status of Japanese LunarExplorers: SELENE and Lunar-A 181T. Iwata, S. Tanaka, M. Kato, S. Sasaki, N. Namiki

From Nuclear Blasts to Cosmic Bombardment 191K. O’Brien

Page 13: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

This page intentionally left blankThis page intentionally left blank

Page 14: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

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

SOME SIMILARITIES AND DIFFERENCES BETWEENTHE MARS AND VENUS SOLAR

WIND INTERACTIONS

JEAN-GABRIEL TROTIGNON

Laboratoire de Physique et Chimie de l’Environnement, Centre National de laRecherche Scientifique, Orleans University

3A, Avenue de la Recherche Scientifique, F-45071 Orleans Cedex 02, [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 Expressfor planet Mars and Venera 9 and 10, Pioneer Venus Orbiter, Venus Expressfor planet Venus. Overall observations of plasma regions and their boundaries,in particular the bow shock, the magnetic pile-up boundary and the magnetictail, show the solar wind interaction with these two planets to be rather similar.Mars and Venus are both considered as non-magnetic planets, compared withthe Earth, in a sense that they do not possess any significant intrinsic magneticfield that could play a significant role in their interactions with the solar wind.At most, the magnetic anomalies discovered at Mars by MGS are thoughtto slightly influence the lower regions of the Martian ionosphere. Therefore,both Venus and Mars have principally comet-like induced magnetospheresand magnetotails as a result of the atmospheric mass loading and subsequent

draping of passing interplanetary flux tubes. Nevertheless, there are manydifferences between the characteristics and space environment behaviors of thetwo telluric planets and a lot remains actually to be done, in terms of in situmeasurements and modeling efforts, to fully understand how Venus and Marsinteract with the interplanetary medium. The objective of the presentationis not to review all the aspects of these interactions but simply to comparethe main characteristics of the Mars’ and Venus’ plasma environments and tohighlight some similarities and differences between the interactions of these twonon-magnetic planets with the solar wind as a function of solar wind dynamicpressure and solar activity.

1. Introduction

Despite the great number of missions devoted to the exploration of Earth’snearest neighbors, Venus and Mars and a prolific scientific output, actuallynot too much is known about the interaction of these planets with the solar

1

Page 15: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

2 J.-G. Trotignon

wind, and this is particularly true for Mars. This is due to limitations ofspace missions for gathering in situ data especially in situation involvingvariations in latitude, longitude, altitude, time, and solar activity. Thisis also because only a very few spacecraft are well equipped to monitorelectric and magnetic fields, and the spatial distribution and compositionof particles over wide ranges of frequency, mass and energy. This leads oneto the conclusion that new missions, in particular low altitude orbitersand entry probes, are required to fully understand the nature of theseinteractions, compared with the ones of the Earth and mainly comets. Theyare indeed thought to be much more intricate than those described in mostof the current literature.

The objective of this paper is actually not to give a comprehensivereview of what is known and what is unknown about the plasma and waveenvironments of Mars and Venus and about their interactions with theinterplanetary medium but rather to point out some relevant features thatremains to be clarified and/or understood and therefore investigated infuture missions.

Section 2 recalls the basic facts about the interactions between the solarwind and the two planets. Some major similarities and differences are thenpresented 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 haveconsiderably enriched our knowledge of the space environment of Venus andits interaction with the solar wind.1,2

Venus turned out to be a non-magnetic planet, with a dense daysideionosphere. There are extended dayside exospheres of hot hydrogen andoxygen: O dominates over H up to 3000km where H becomes dominant(the main source of neutral O corona is dissociative recombination ofO+

2 ). Neutral exosphere atoms above the dayside ionopause, ionizedby photoionization, charge exchange with solar wind protons, impactionization, 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 solarwind. The solar wind thus appears to be diverted around the upperboundary of the ionosphere, the ionopause, where the incident solar windram pressure is balanced by the ionosphere thermal pressure. At the

Page 16: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

Mars and Venus Solar Wind Interactions 3

ionopause, dissipationless currents flow in a thin layer, thus creating ahighly conducting spheroid. These currents produce magnetic fields thatcontribute to stand off the solar wind.

As illustrated in Fig. 1,1 the solar wind is shocked and diverted aroundthe ionosphere and continues along its antisolar route. This flow carriesthe interplanetary magnetic field with it. The interplanetary magnetic fieldis compressed in front of the Venus’ ionopause, thus creating a magneticbarrier that separates the plasmas of external and internal origin. On theflanks of the planet, the interplanetary magnetic field lines continue to movedownward to form a magnetic tail similar to comet tails. This phenomenonis known as the magnetic field draping effect.

2.2. The Mars case

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

Fig. 1. Formation of induced Venus’ magnetotail from draped interplanetary magneticfield.1

Page 17: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

4 J.-G. Trotignon

planet Venus, while multiple magnetic anomalies of small spatial scale existin the crust of Mars.4 These magnetic anomalies are at most thought toslightly influence the lower regions of the Martian ionosphere.5 The lowMartian gravitational field (compared with those of Earth and Venus)allows the neutral exosphere of Mars to interact significantly with theinterplanetary medium. Comet-like features are likely to be more evidentat 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 forescape from Mars, will contribute to the solar wind erosion of the Marsatmosphere; the others make up the gravitationally bound exosphere.Particles can gain energy through photochemical processes or sputtering byenergetic particles from above. For example, O+

2 which are majority ionsin the ionosphere of Mars produce energetic oxygen atoms, O∗, throughphotodissociative recombination. O∗ may gain sufficient energy to flowsunward and even escape from Mars. At Venus, the above photodissociativerecombination cannot supply O∗ whose velocity is larger than the Venusescape velocity (∼10 km s−1). By photoionization O+ ions may thenbe produced and captured by the solar wind and embedded into theinterplanetary magnetic field (ion pickup process as for Venus and comets).As ions are continuously produced, the solar wind carries more materialalong with it and is mass-loaded. Conservation of momentum and energyimplies a slowing down of the solar wind flow, thus diminishing the pressureexerted on the planet environment. Extensive reviews of the Martianenvironment and its interaction with the solar wind may be found in Refs. 3and 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 thedata of, respectively, PVO,10 and Phobos 2 and MGS.11

As can be seen in Fig. 2, the Mars’ bow shock has a larger terminatorradius than the one of Venus, while the subsolar positions of the two bowshocks are comparable. In addition, the effective obstacle to the solar windflow is larger at the flanks of Mars. According to Luhmann,12 it could bethe result of a larger solar wind ion gyroradius relative to the planet radiusfor Mars and/or different compositions and scale heights of the two upperatmospheres.

Page 18: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

Mars and Venus Solar Wind Interactions 5

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

The electric-field spectra plotted in the right-hand panel of Fig. 3 wererecorded 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 rampsof Jupiter, the Earth, Venus and Mars.13–15

Page 19: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

6 J.-G. Trotignon

frequency, fpi, and then a cutoff, which is thought to be the signatureof 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 Itis noteworthy that they all show a remarkably close similarity in shape.The noise amplitudes are nevertheless very different. This is partly dueto instrumental effects, the antenna lengths are indeed sometimes lowerthan the plasma Debye length, so that the effective length of such antennaebecomes 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, 730Hz and 100Hz) 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 plasmafrequency, in the electron foreshock by suprathermal electrons that areenergized and reflected at the shock whenever the interplanetary magneticfield is connected to the shock surface. Further downstream, in the ionforeshock, ion-acoustic and ULF waves are also generated. Such waves wereobserved in the Venus electron and ion foreshocks, as shown in Fig. 4.16

Note that foreshock waves are different to those generated from exosphericion pick-up.9

The spatial distribution of electron plasma oscillations observed atMars by PWS is displayed in Fig. 5.17 As expected, the highest electric-fieldintensities are seen in the electron foreshock. The observed limited extent ofplasma oscillations along the tangent magnetic-field line is thought to be theconsequence 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

Page 20: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

Mars and Venus Solar Wind Interactions 7

Fig. 5. Electron plasma oscillation intensity in the equatorial B–v plane, which containsthe center of Mars and is parallel both to the solar wind velocity and the interplanetarymagnetic 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 averageshock 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 hasonly been measured by radio-occultation.18 A precise determination of theion and electron temperatures remains to be done. The ionosphere upperboundary, the ionopause, is typically located at 300km in the subsolardirection and at 1000km close to the terminator, at least, when the solarwind ram pressure does not exceed the ionospheric pressure.19,20

Page 21: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

8 J.-G. Trotignon

Information about the ionosphere of Mars actually comes only fromin situ measurements made by the two Viking landers (at low latitudes,∼45 SZA, near a minimum of solar activity), and altitude profiles of theelectron density obtained from the radio-occultation experiments onboardthe Mariner and Viking orbiters.22 A peak density of 105 cm−3 (about oneorder of magnitude lower compared with Venus) is usually observed at125-km altitude. The topside termination of the ionosphere (ionopause) hasnot been firmly observed. This could be attributed to the fact that, at Mars,the ionospheric thermal plasma pressure is most of the time insufficientto balance the solar wind dynamic pressure.3 At Venus, when the solarwind dynamic pressure occasionally exceeds the peak ionospheric plasmapressure, no sharp increase of the ionosphere density is indeed observed atthe ionopause.19 The Martian ionosphere is mainly composed of O+

2 ions,mostly created by charge exchanges between CO+

2 and O2, and betweenCO+

2 and O.23

Returning now to the fact that, at Mars, the ionospheric thermalplasma pressure is usually lower than the solar wind ram pressure.Luhmann24 claimed that by analogy with Venus, one might expect to findinduced magnetic fields on the surface of Mars that could be detectedby ground magnetometers. These fields of external origin should thuscombine with crustal remanent magnetic fields. At solar maximum, whenthe solar wind pressure gets very high, large-scale horizontal magneticfields have been observed in the dayside ionosphere of Venus. They areinterplanetary fields incompletely cancelled by the shielding currents in theupper ionosphere. At Mars, these fields could also make their way throughthe solid mantle of the planet, might be pulled into the wake, and couldproduce the slingshot magnetic field pattern on the nightside that willaccelerate plasma down the tail. It may therefore be the greatest loss sourcefor the Mars atmosphere (C. T. Russell, private communication).

3.3. Plasma clouds

Thermal electrons (plasma clouds), possibly scavenged from the top ofthe ionosphere by Kelvin–Helmholtz instability, were detected in theVenus’ magnetosheath, above the ionopause.25,26 They could be attachedstreamers analogous to cometary tails and might be the seed population ofsuprathermal ions (10–90eV) 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

Page 22: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

Mars and Venus Solar Wind Interactions 9

Fig. 6. From top to bottom, spacecraft potential, plasma density, electron gyro-frequency and electric field in three frequency channels measured by the PWS of thePhobos 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 Plasmaclouds are seen just after the MPB.

order of 104 K have been reported.15 It is worth noting that the dynamicpressure developed by such clouds is equivalent to that of a 20-nT magneticfield. Plasma clouds could be generated by ionization of plasmasphericneutrals or could result from ionopause instabilities. They actually looklike Venus’ cold plasma clouds.

Plasma clouds (>60 cm−3; ∼1 eV) were also observed in the night sectorof Mars, close to the neutral sheet, in association with a broadband waveactivity (from a few Hz to several kHz). The density profiles of these coldplasma clouds displayed fluctuations correlated with those of the magneticfield. They may originate from the dayside ionosphere and be dragged intothe 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 pile-up boundary (MPB), a boundary in the lower magnetosheath of Mars

Page 23: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

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 isactually a plasma boundary formed by the interaction of the solar windwith the Martian exosphere/ionosphere.8,11,28

At Venus, plasmas of solar and planetary origin appear to be separatedfrom each other by a transition region, called mantle (or magnetic barrier),where plasma clouds were seen.1,26 Recently, Bertucci et al.29 claimedthat the upper boundary of this region would be similar to the MPBidentified at comets30,31 and planet Mars.21,28 The MPB should thereforebe a common plasma boundary in the interaction between the solar windand non-magnetic objects.

4. Conclusion

The solar wind interactions with Venus and Mars appear to be quitesimilar, but with significant differences.12,19,32 These differences include agreater width of the Martian bow shock associated with a greater width ofmagnetotail and a much larger O+ ion escape rate. The latter is related toa lower Martian gravitational field compared with those of Venus and theEarth, a larger extent of the exosphere, and efficient energizing processes. Arelative insensitivity of the Martian bow shock to the solar cycle comparedwith 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 thesolar wind, therefore the atmosphere (exosphere included) and ionospherecombine to provide an obstacle to the flow. Outermost signatures of Mars’and Venus’ obstacles are fast magnetosonic shocks and foreshocks. Magneticpile-up regions of the two planets are dominated by oxygen ions fromplanetary origin and are bounded outward by the MPB, an internal plasmaboundary which could be a common feature of non-magnetized bodies.

Topside termination of the ionosphere (ionopause) is not observed atMars, it could be because the incident solar wind pressure usually exceedsthe ionospheric one. Further analyses (in particular, Mars Express andVenus Express data) and, definitely, in situ observations (ESA CosmicVision program, other international programs) might yield additionalinformation and most probably contrasts, in particular on the nightsidewhich is, for example, almost unknown at Mars. We indeed do not know if

Page 24: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

Mars and Venus Solar Wind Interactions 11

plasma holes similar to the ones observed at Venus33,34 are actually presentat 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. Acuna, J. E. P. Connerney, P. Wasilewski, R. P. Lin, K. A. Anderson,

C. W. Carlson, J. McFadden, D. W. Curtis, D. Mitchell, H. Reme, 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.

Paetzold, 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. Acuna, K.Baumgartel, 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. Acuna, 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.

Page 25: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:54 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch01

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 andV. 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. Acuna, D. L.Mitchell, R. P. Lin, J. E. P. Connerney, H. Reme, P. Cloutier, N. F. Ness andD. Winterhalter, Geophys. Res. Lett. 30 (2003) 1099.

29. C. Bertucci, C. Mazelle, J. A. Slavin, C. T. Russell and M. H. Acuna, Geophys.Res. Lett. 30 (2003) 1876.

30. F. M. Neubauer, Astron. Astrophys. 187 (1987) 73.31. C. Mazelle, H. Reme, 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 andYe. 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. Perez-de-Tejada, J. Geophys. Res. 106 (2001) 211.

Page 26: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

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

COMPARISON OF MICROWAVE OBSERVATIONSOF MARTIAN TEMPERATURE AND WINDS WITHGENERAL CIRCULATION MODEL SIMULATIONS

TAKESHI KURODA∗ and PAUL HARTOGH

Max-Planck-Institute for Solar System ResearchMax-Planck-Str. 2, D-37191 Katlenburg-Lindau, Germany

[email protected]

Microwave observations of the temperature and wind in the middle atmosphereof Mars are compared with the results of simulations with the Martiangeneral circulation model. The simulated global-mean mesospheric temperatureduring a northern summer solstice is ∼10 K lower than in spring and autumnequinoxes, which is consistent with the James Clerk Maxwell Telescopeobservation in 1996–1997, although the absolute values are 30–40 K higherthan in the observations. The wind velocity in the middle atmosphere in themodel is comparable to the observations, except that the easterly wind inthe afternoon is ∼100m s−1 weaker. A brief discussion for the discrepanciesbetween the model and observation is provided.

1. Introduction

Microwave observations of the Martian atmosphere from the groundor Earth-orbiter were performed to detect the vertical profiles of thetemperature,1–5 wind,5,6 and distributions of molecules3 from the spectrallines of CO, H2O, and O2. Microwave observations have three majoradvantages in comparisonwith the infrared observations such as the ThermalEmission Spectrometer onboard Mars Global Surveyor (MGS-TES).

First, a more accurate retrieval is possible. Radiative transfercalculations have fewer unknowns, because the direct scattering andemission by Martian dust are negligible due to small particle sizes(∼2 µm) relative to the observing wavelength. The change of atmospherictemperature due to the planet-encircling dust storm was observed usingthe Earth-orbiting Submillimeter Wave Astronomy Satellite (SWAS). Thedetected change of the global-mean temperature consistent with the MGS-TES observations was obtained below ∼ 40 km4.

13

Page 27: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

14 T. Kuroda and P. Hartogh

Second, microwave measurements are possible for higher altitudes. Dueto high sensitivity and spectral resolution of the heterodyne technique, thelatter can probe temperatures from the surface to above 80 km (comparedto ∼40km for MGS-TES).1,2,4 The SWAS observations detected that theplanet-encircling dust storm has little impact on the global temperaturestructure above ∼60 km, the upper limit of MGS-TES measurements.4 Inaddition, the James Clerk Maxwell Telescope (JCMT) on Mauna Kea,Hawaii observed the temperature of the Martian mesosphere (height of50–80km) in 1996–1997, during the Martian northern spring and summer.It showed that the global-mean dayside temperature becomes 125–140K for50–60km height and drops to ∼120 K at 70–80km, that is to temperaturesat which a local CO2 condensation can occur,2 as observed by the descend-ing Pathfinder spacecraft.7 The observations with the Heinrich HertzTelescope (HHT, Mount Graham, Arizona) in 1996/1997 also show resultsoverall consistent with JCMT (Hartogh, poster on AGU fall meeting, 1997).

Third, the direct observation of wind is possible using microwaveinstruments. Winds can be retrieved from Doppler shifts detected on therotational transitions of CO. Observations with JCMT show easterly windsof 120–200 ms−1 at ∼20S and the poleward winds of 30–40 ms−1 in bothhemispheres at 35–80km during Ls = 254 (the northern autumn).6 Weshould emphasize that direct observation of the wind velocity in the Martianatmosphere has never been done with other methods.

In this paper, we simulate the Martian atmospheric temperature andwind velocity using a general circulation model (GCM),8 and comparethe model results to the results of JCMT microwave observations. Thecomparison with the SWAS observations before and after the onset ofa global dust storm in 2001 was already published.9 Here we show thesimulated global-mean temperature of the mesosphere in the northernspring and summer, and wind velocity distributions in the northern autumn.The GCM used in this paper is briefly described in Sec. 2. The results ofthe simulations are presented in Sec. 3. The discussions and comparisonbetween the model and observations are given in Sec. 4.

2. Model Description

A detailed description of the Martian GCM called MAOAM was given inRef. 8. The radiative CO2 scheme, which accounts for non-LTE (breakdownof the Local Thermodynamic Equilibrium), and the dust radiation scheme

Page 28: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

Microwave Observations of Martian Temperature and Winds 15

are the same as described in Ref. 9, except that the dust-mixing ratio issmaller. This is because global dust storms, which can significantly affectthe thermal structure of the middle atmosphere, were not observed eitherin 1996–1997 or in 2003. The “TES2 dust scenario” (described in Ref. 10in full details) is used for the definition of the dust opacity depending ontime and latitude. The global mean visible dust opacity of less than 0.3 isprescribed for the whole year. Effects of dust on the results of simulationspresented 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 50–90km.The local time at disk center varied between 09:00 and 15:00, and the sub-Earth latitude of the observation also varied between 10N and 27N, duringthe JCMT observation in 1996–1997.2 They possibly affected the results forthe observed temperature; therefore, we considered them for the plots inFig. 1. To construct “disk-averaged” quantities, we reprocessed the modeloutput in accordance with the appearance of the planet’s disk. The accurateephemeris was used to obtain the sub-Earth local time and latitude forparticular dates, and the weighting function for the “visible” grid pointsbased on the corresponding viewing geometry was utilized, as describedin Ref. 9 for Figs. 2 and 3. Figure 2 shows the observational results fromJCMT in 1996–1997, Viking and Mars Pathfinder descent measurements

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

Page 29: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

16 T. Kuroda and P. Hartogh

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

and NRAO/IRAM millimeter measurements in 1994 (during a global duststorm).2

The simulated “disk-averaged” temperature (Fig. 1) is 30–40K warmerthan in the JCMT observations, and 10–20K warmer than in the Vikingdescend measurements. However, it is consistent with the JCMT obser-vations in that the simulated temperature at 60–70km during a northernsummer solstice is ∼10 K lower than in spring and autumn equinoxes. Acharacteristic feature of the current MAOAM-GCM is strong winter polarwarming that tends to be produced11 due to the non-LTE CO2 radiationscheme12 which underestimates the radiative cooling. If it is replaced bythe LTE one13 (e.g., as in Ref. 10), the simulated temperature becomes∼ 10 K colder. This temperature decrease occurs due to the neglect ofnon-LTE effects and the associated overestimate of the infrared coolingrates that result in weaker simulated meridional transport. With strongermeridional circulation simulated with the non-LTE scheme, the adiabaticheating in the mesosphere increases, and the diabatic cooling decreases.11

The temperature profile simulated with the LTE CO2 radiation schemeis comparable with the Viking descend temperatures. However, note thatViking did not sample temperatures at exactly same latitudes and seasonscompared to the MAOAM-GCM simulations. The microwave observationsprovide only globally averaged temperatures, and do not have the spatialresolution that would allow separating polar warmings.

Page 30: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

Microwave Observations of Martian Temperature and Winds 17

Fig. 3. (a) Zonal and (b) meridional wind velocity at 60 km at Ls = 254 simulatedwith MAOAM-GCM (averaged during half a day, from sub-solar longitude of 45E to135W). 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 correspondingtelescope beam locations used for obtaining the wind velocity shown in Fig. 4.

Figure 3 shows the simulated zonal and meridional wind velocities at60 km for Ls = 254. The velocities in this figure are reconstructed byaveraging the model output in the “visible” grid points for half a day.The sub-solar longitude varies from 45E to 135W 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 figurerepresent the corresponding telescope beam locations during the JCMTmeasurement on September 4, 2003 (Fig. 4). At each sub-solar longitudeduring 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 windvelocity is apparently more significant than the longitude dependence. Forthe meridional wind, the maximum velocity of the northward wind in thenorthern hemisphere at a sub-solar longitude of 135W is ∼40 ms−1 smallerthan of 45E, while that of southward wind in the southern hemisphere is∼30 ms−1 larger.

The meridional wind velocities in both hemispheres and easterlywind in the morning are comparable in the observations and simulations.However, the easterly wind in the afternoon is ∼100 ms−1 larger in theobservation than in the model.

Page 31: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

18 T. Kuroda and P. Hartogh

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

4. Discussions

The results of the microwave observations of temperature and wind inthe Martian mesosphere are compared with the GCM simulations. Thewind velocity was directly measured from Doppler shifts of CO lines. Weemphasize that these wind data obtained from the microwave measurementsare unique for the Martian atmosphere. The descending landers (Viking andMars Pathfinder) performed the only other available wind measurements.

The simulated temperature in the Martian mesosphere is considerablyhigher than shown by the JCMT microwave data. However, the modeltemperature at 60–70km is ∼10 K lower in summer solstice than in springand autumn equinoxes, which is consistent with the measurements. Onepossible reason for this discrepancy is the strong winter polar warmingthat is simulated in the model.11 Moreover, it is still difficult to determinethe characteristics of the mesospheric temperature, because the availableobservational data are very sparse, and there are differences of 20–30Kbetween the observational data (Fig. 2). Ref. 2 describes that the higher(compared to other measurements) temperature derived from the Vikingdescend is due to either the effect of dust raised to higher altitudes, orstrong near-infrared heating by CO2 in the northern summer. Because theViking descend was made during the daytime in summer at the latitude of∼22N, the observed temperature represents a local snapshot, whereas theJCMT observations cover the global-mean temperature. More observationaldata are needed for constraining and validating model results.

Page 32: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

Microwave Observations of Martian Temperature and Winds 19

As for the wind velocity, the large difference between the measurementsand simulations is seen only for the easterly wind in afternoons. OtherMartian GCM14 did not reproduce the strong easterly wind observed byJCMT either, as described in Ref. 5. To a large degree, the simulated windvelocity is related to the temperature distribution through the so-calledthermal wind relation:(

f +2u tanφ

a

)∂u

∂z= − g

aT

∂T

∂φ(1)

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 isrequired. For example, the temperature at the latitude of 57S 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 thelocal times 2–3pm, which explains a weaker easterly wind. The atmosphericheating 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 distributionmight contribute to the production of stronger easterlies. In addition, iondrag may affect the wind velocities. Because of the thin air and weakermagnetic field, the effects of ions can influence the middle atmosphere ofMars.

Acknowledgments

This work was supported by Deutsche Forschungsgemeinschaft (DFG),project HA 3261/1-2. The authors are grateful to Dr. Alexander S. Medve-dev and an anonymous reviewer for helpful comments on the originalmanuscript.

References

1. R. T. Clancy, D. O. Muhleman and G. L. Berge, Global changes in the0–70 km thermal structure of the Martian atmosphere derived from 1975 to1989 microwave CO spectra, J. Geophys. Res. 95 (1990) 14543.

2. R. T. Clancy and B. J. Sandor, CO2 ice clouds in the upper atmosphere ofMars, Geophys. Res. Lett. 25 (1998) 489.

Page 33: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch02

20 T. Kuroda and P. Hartogh

3. M. A. Gurwell et al., Submillimeter wave astronomy satellite observationsof the Martian atmosphere: temperature and vertical distribution of watervapor, Astrophys. J. 539 (2000) L143.

4. M. A. Gurwell, E. A. Bergin, G. J. Melnick and V. Tolls, Mars surfaceand 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 lineobservations during the 2003 and 2005 Mars oppositions, Abstract of “SecondWorkshop 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 StructureInvestigation/ 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 anew 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 duringthe 2001 global dust storm: observations with SWAS and simulations witha general circulation model, in Advances in Geosciences. Vol. 3: PlanetaryScience (PS) (2006), (World Scientific Publishing, Singapore, 2006), pp. 145–154

10. T. Kuroda, N. Hashimoto, D. Sakai and M. Takahashi, Simulation of theMartian 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 meridionaltransport 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 solarradiation 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 modelsof the Martian atmosphere from the surface to above 80 km, J. Geophys.Res. 104 (1999) 24155.

Page 34: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

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

ASTEROID COMPOSITIONS: SOME EVIDENCEFROM POLARIMETRY

A. CELLINO∗ and M. DI MARTINO†

INAF, Osservatorio Astronomico di Torinostrada Osservatorio 20, 10025 Pino Torinese, Italy

[email protected][email protected]

A.-C. LEVASSEUR-REGOURD

Univ. P. & M. Curie (Paris VI)/Aeronomie CNRS-IPSLBP 3, 91371 Verrieres, France

[email protected]

I. N. BELSKAYA

Astronomical Institute of Kharkiv National UniversitySumska str. 35, Kharkiv 61022, Ukraine

[email protected]

Ph. BENDJOYA

LUAN UMR 6525, Universite de Nice,Parc Valrose, 06108 Nice cedex 2, France

[email protected]

R. GIL-HUTTON

Complejo Astronomico El Leoncito (Conicet) andSan Juan National University,

Av. Espana 1512 sur, J5402DSP San Juan, [email protected]

Although it cannot provide direct and unambiguous information on themineralogical composition of an asteroid surface, polarimetry is a very usefultool to get an improved understanding of parameters which are intimatelyrelated to surface composition and regolith structure. In recent times there hasbeen a revival in the field of asteroid polarimetry, on the theoretical side, inrelation to experimental simulations, and due to the activity of some teamswho are engaged in extensive observational campaigns. Some new discoveriesof objects exhibiting unprecedented polarimetric properties have been done.The above subjects are briefly reviewed.

21

Page 35: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

22 A. Cellino et al.

1. Introduction

The visible light that we receive from the asteroids and other atmospherelessbodies of our solar system is in a state of partial linear polarization, asa consequence of the fact that it consists of solar radiation scattered bythe solid surfaces of the objects. The polarization properties of sunlightscattered by atmosphereless solar system bodies have been investigatedsince a long time, because in principle they can be a source of informationabout the physical properties of the materials present on the surfaces ofthese bodies. The first pioneering investigations in this field were carriedout by Lyot,1 and were later continued by Dollfus et al. at the Paris-MeudonObservatory, and subsequently by other researchers in different countries.The historical background of asteroid polarimetry was briefly summarizedin a classical chapter of the Asteroids II book.2

The observations allow the observers to directly measure the degreeof polarization of light coming from an asteroid. The state of polarizationof a light beam is described by the Stokes parameters Q and U (givingthe degree of linear polarization), V (related to circular polarization), andI (the total intensity of the received light). In asteroid polarimetry, theV parameter is usually negligible, and the light is in a state of partiallinear polarization described by the Stokes parameters Q and U . Theobservations show that the plane of linear polarization is generally eitherparallel or perpendicular to the scattering plane, which is defined as theplane containing the asteroid, the Sun and the observer at the epoch ofobservation. This fact is a consequence of the sunlight scattering processacross the surface of the body (see also below).

The parameter that is usually adopted to describe the polarimetricbehavior of asteroids is Pr = P cos(2θ), where P is the degree of linearpolarization, given in module by

√Q2 + U2, and θ is the angle between

the measured direction of the plane of partial linear polarization (definedby the observed position angle, given by arctan(U/Q) and the normalto the scattering plane. When the Pr parameter is measured in differentconditions of illumination, described by the phase angle (the angle betweenthe directions to the Sun and to the Earth as seen from the asteroid) awell defined relation between Pr and the phase angle is usually found. Thetypical 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 arange 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

Page 36: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

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 observationscarried 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 theinversion angle. Due to obvious geometric constraints, the branch of positivepolarization cannot be sampled beyond phase angles of the order of 30–35

in the case of main belt asteroids. The polarimetric behavior at much largerphase angles, however, is well documented in the case of some near-Earthasteroids, including (4179) Toutatis and (25143) Itokawa.3,4 Around theinversion angle, the trend of variation of Pr as a function of phase is mostlylinear. The slope of this linear trend is usually indicated by the symbol h,and is an important parameter, because an empirical relation is known toexist between h and the geometric albedo pV of the surface (in V light). Thisrelation may be written in the form log pV = C1 log h + C2, and a similarrelation is also found between pV and the absolute maximum of negativepolarization, usually called Pmin, which is usually reached at phase anglesbetween 8 and 10. The existence of these relations between the albedoand the polarimetric properties of the objects constitutes one of the bestavailable techniques to derive asteroid albedos.2,5,6

Page 37: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

24 A. Cellino et al.

2. Theory and Experiments

The interpretation of the polarimetric properties exhibited by the asteroidsand other atmosphereless solar system bodies is still a challenge, althoughsignificant advances in this field have obtained in recent years.

At present, there are not exact analytical formulae fully describingthe phenomena of light scattering in situations corresponding to the caseof sunlight scattered by asteroid surfaces. These situations essentiallyconsist of light scattering by close-packed random media of inhomogeneousparticles having sizes larger than the light wavelength.7 Light scatteringmechanisms must be responsible not only of the observed polarimetricproperties, but also of the photometric properties, including the observedphase–brightness relation. In particular, observations show that there isa mostly linear variation of brightness upon the phase angle (the objectsbecoming increasingly fainter for increasing phase angle), but at small phaseangles a considerable non-linear increase of brightness is observed (the so-called “opposition effect”).

In recent times, it has been widely accepted the idea that both theobserved polarimetric and photometric properties of the asteroids can beexplained in terms of two major mechanisms: coherent backscattering andshadowing. The former mechanism is based on constructive interferenceof scattered electromagnetic waves in presence of multiple scattering.Interested readers can find a description of this mechanism and appropriatereferences to previous work in the Muinonen et al. chapter in the AsteroidsIII book.7 The shadowing mechanism is essentially due to the fact that aphoton incident on a particle on the surface can always be scattered backalong the same direction of incidence, whereas along other directions it canbe blocked by the presence of other surface particles.7 According to currentunderstanding, the coherent backscattering mechanism plays a role in thegeneration of both the observed brightness and polarization phase relations,while shadowing should contribute mainly in determining the brightnessopposition effect. In particular, the latter seems to be nicely explained asa consequence of both a lack of shadowing at zero-phase, as well as byconstructive 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 measurementsin the past. Based on laboratory data, it seems that both the depth of thebranch of negative polarization and the value of the inversion angle stronglydepend on the albedo and microscopic inhomogeneity of the investigatedmaterial samples, as well as on their packing density. A problem has been

Page 38: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

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 reasonis that for this purpose special instruments are needed, with small angleapertures of both the light source and the receiver. Another problem isthat laboratory measurements at extremely small phase angles require avery high accuracy, because the polarization degree is typically close tozero at these angles. Some attempts at improving the situation in thisrespect have been done only in recent years.8 We note that laboratoryexperiments including polarimetric and albedo measurements at or veryclose to zero phase angle are very important also from the perspective ofcomplementing available astronomical observations with laboratory dataobtained in similar illumination conditions. In particular, the definition ofthe geometric albedo of the asteroids is based on their reflectivity at zerophase angle, thus any rigorous attempt of deriving in the future a bettercalibrations of the polarimetric slope–albedo relation (see above) cannotinclude laboratory experiments if they are not made very close to zerophase angle. The reason is that the measured luminosity at zero phase canbe strongly affected by the non-linear brightness opposition surge.

To summarize the current situation, it can be said that we havetoday a better understanding of the most subtle physical effects involvedin the generation of the photometric and polarimetric properties of theradiation we receive from the asteroids. In particular, the role of thecoherent backscattering mechanism is now fully appreciated. There are stillsome problems in creating models able to reproduce at the same timeand in details both the magnitude–phase relation and the polarimetricproperties observed for the asteroids. In particular, the wide extension ofthe negative polarization branch in a range of phase angles of about 20 isstill a challenging feature, although wide negative polarization branches areobserved in some laboratory experiments, or can be numerically modeledassuming single-particle scattering. On the other hand, it seems likely thatfurther advances in the modeling side are still possible, and the subject iscurrently 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 termsof color and albedo. This led to the development of taxonomic classesbased on these properties. The general idea, especially at the beginning,

Page 39: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

26 A. Cellino et al.

was that the differences in reflectance properties among different objectscould be directly related to differences in mineralogic compositions. For thisreason, some of the first taxonomic classes identified in the first pioneeringworks were called using symbols that were strictly related to a mineralogicinterpretation (i.e., S for “silicates”, M for “metals”, etc.).

Analyses of the first available spectrophotometric and polarimetricdata-sets revealed that spectral reflectance properties alone were not suffi-cient in many cases to derive a full taxonomic classification, since it turned outthat different classes existed which exhibited the same spectrophotometricproperties in visible wavelengths, but were characterized by huge differencesin albedo, based on their polarimetric properties.10 The best example is givenby the EMP complex, formed by three separate classes (E, M , and P ),which have fairly identical spectra but largely differ in terms of albedo. Inmore recent years, some large systematic surveys (SMASS, SMASS211) haveprovided spectral data for big samples of objects, from which a taxonomyhas been derived. This classification is no longer based on simple color-indexes using limited numbers of filters, but on full reflectance spectra,covering wavelength ranges approximately between 0.5 and 0.9 µm. For thevast majority of these objects no polarimetric data are available, and thealbedo is not known, and attempts have been made to separate differenttaxonomic classes based on the presence of subtle features of the reflectancespectra.11 In particular, the E, M , and P classes correspond now to differentsubsets of a bigger complex called X . These subsets are distinguished on thebasis of spectroscopic features that have been found to characterize objectsbelonging to the E, M , and P classes defined in previous, albedo-basedclassifications.

This does not mean, however, that polarimetrically-derived albedosand, more in general, polarimetric properties are no longer useful fortaxonomy-related purposes, or for better understanding the propertiesof asteroid surfaces. In this respect, the albedo per se is a veryimportant parameter, being strictly related to the composition and agingof the surface. It should be stressed that polarimetry must be considered asthe best available technique to derive asteroid albedos. The reason is thatthe existence of a direct relation between the polarimetric slope (and alsoPmin) and the albedo makes it possible to derive the albedo directly from theobserved polarimetric parameters, without the need of knowing any otherparameter. This is a big advantage with respect to other possible techniques,which cannot measure directly the albedo, but derive it more indirectly, forinstance from measurements of the size, and based on nominal values of theabsolute magnitude. This is the case, for instance, of thermal radiometry.

Page 40: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

Asteroid Compositions 27

Fig. 2. Pmin vs. inversion angle (αinv) plot for a set of asteroids belonging to differenttaxonomic 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 toolfor taxonomic characterization purposes. As an example, Fig. 2 shows thatin a plot of Pmin vs. polarimetric inversion angle, not only there is a clearseparation between low (F , B, C, G, D) intermediate (S, M) and high-albedo (E) objects, but also there is a fairly clear separation even amongobjects of similar albedo. As a matter of fact, a recent analysis12 hasconvincingly shown that the availability of a good coverage of the phase–polarization curve for a sample of objects is sufficient to derive a taxonomicclassification in very good agreement with that produced by spectroscopicdata. In particular, a principal component analysis applied to a set ofphase–polarization curves described by a polynomial or trigonometricrepresentation has been found to produce a set of taxonomic classes thatvery nicely fit the classes obtained by means of purely spectroscopic data.This interesting result suggests that polarimetry and spectroscopy arenicely complementary for taxonomic purposes.

4. Recent Observational Results

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

Page 41: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

28 A. Cellino et al.

more time-consuming than other observing techniques, (3) the relativerarity of experts in the field, which is also a consequence of the somewhatparticular characteristics of polarimetry, being often considered as a fairlyobscure technique whose results seem mostly based on empirical and not-too-well understood laws. The items (1) and (2) in the above list arestraightforward: since a polarimetric measurement implies the need ofsplitting the incident light beam into an ordinary and an extraordinaryray, in order to derive the Stokes parameters, fairly large telescopes areneeded for the observations of faint targets. Moreover, asteroid polarimetryis intrinsically time-consuming due to the need of observing each objectover fairly long intervals of time, to obtain a sufficient sampling of thephase–polarization curve. In other words, asteroid polarimetry is not forthose who want to have one publication per one night of observations.

In spite of the above difficulties, in recent years there has been asignificant increase of activities in the field. A significant role has beenplayed by the availability of the Torino photopolarimeter, which equipsthe 2.15m telescope of the El Leoncito observatory (Argentina). Thisinstrument has produced in recent years a significant amount of data,6,15

nicely complementing the observing activities of other teams, mainly inUkraine, who have been working in this field since a long time.

The most recent observing campaigns carried out by different authorshad a variety of purposes. These include a systematic check of the albedovalues derived by thermal radiometry observations of small main beltasteroids, an analysis of the polarimetric behavior of different objectsobserved at very small phase angles, an exploration of the properties ofthe 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 tothe subject of the present paper, comparative analyses of the polarimetricproperties of objects belonging to different taxonomic classes.

In this respect, at least a couple of interesting results have beenobtained recently, namely an extensive characterization of the polarimetricproperties of F -type asteroids, and the discovery of the unusual polarimetricproperties of (234) Barbara, a rare Ld-type object.

The above-mentioned examples deal with phase–polarization curvescharacterized by extreme and opposite properties. In the case of the F -typeasteroids, the most striking polarimetric property, as shown also in Fig. 2, isthe very low value of the inversion angle.16 The F taxonomic class includesobjects characterized by low-albedo, and linear, featureless reflectancespectrum. F -type asteroids are believed to be primitive, representing a

Page 42: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

Asteroid Compositions 29

subclass of the big C complex, from which they can be separated based ontheir spectral behavior at short wavelengths. Since this part of the spectrumwas not included in the spectroscopic data used in the most recent SMASS2taxonomic classification,11 the F class is not included among the classesidentified in this survey. On the other hand, polarimetry indicates thatthe F -type asteroids certainly represent a separate class, exhibiting welldefined and unique polarimetric properties. As mentioned above, F -typeasteroids have been found to be characterized by unusually small values ofthe 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 -typeobjects, however, the inversion takes place at much lower phase angles.In particular, 419 Aurelia exhibits an inversion angle at 14, the smallestvalue ever observed for asteroids. Also the depth of its negative branchseems unusually low, for a low-albedo objects. Small inversion angles, butmore usual negative branches, are also exhibited by other objects of the F

class.16 The interpretation of these properties is not straightforward, butlaboratory experiments and numerical simulations suggest that these mightbe explained by assuming that the surface regolith consists of particlescharacterized by very high optical homogeneity down to scales of the orderof 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 theinversion angle. The first measurements published for this object suggesteda 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 tozero between 24 and 26, and the inversion might take place at a phase anglearound 28. This behavior challenges theoretical interpretation. Accordingto current knowledge, a large inversion angle may be expected for a regolithlayer consisting of very regularly shaped particles (like spheres or crystals)and/or large optical inhomogeneity. As quoted in Sec. 2, a large valueof the inversion angle is just one of the most challenging features to bereproduced by current theoretical models. In this respect, (234) Barbarais very interesting, since it exhibits the largest value of the inversion angleknown today for any atmosphereless solar system body, and is possibly theprototype of a previously unknown class of asteroids, from the point of viewof polarimetric properties.

One of the most interesting facts concerning (234) Barbara is thatit has been found to belong to a fairly rare taxonomic class that hasbeen introduced only recently based on spectroscopic data collected by

Page 43: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

30 A. Cellino et al.

Fig. 3. The phase–polarization curve of (234) Barbara, from observations in V and Rcolors by some of the authors of the present article. The data include some unpublisheddata 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 alarger class called L. Both L and Ld asteroids were previously classified asS, the dominant taxonomic class in the inner part of the asteroid main belt.Is the unusual taxonomic classification of (234) Barbara directly related toits peculiar polarimetric behavior? We have not yet answered this question,due to a lack of polarimetric data on other Ld-type objects. On the otherhand, it is known that a few asteroids belonging to the wider L class forwhich 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 regolithparticles at a microscopic scale, nicely complementing the information thatcan 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 oflaboratory experiments, is still needed to increase the diagnostic power ofpolarimetric data. The wealth of new results obtained in recent years seems

Page 44: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

Asteroid Compositions 31

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

5. Future Developments

Apart from the perspectives in the field of theory and laboratoryexperiments, we want to focus here on some perspectives concerningfuture observational activities. Future observations will be devoted to acontinuation of current efforts, but will probably be aimed also at exploringnew fields of research, including an extensive analysis of the wavelengthdependence of asteroid polarimetric properties. Another field in rapiddevelopment seems to be the study of near-Earth objects. These objectsare interesting in many respects. First, they allow the observers to studythe polarimetric properties at much larger phase angles, where availabledata are still scarce. Moreover, NEOs are important also from the point ofview of the collision hazard. In this respect, a big effort is being producedby many teams to discover the most dangerous objects. A lot of work mustbe done, however, on the side of the study of the physical properties ofthese objects. In this respect, polarimetry can play an invaluable role as apowerful tool to derive the surface albedo, and consequently the sizes of theobjects. Using instruments like the ESO VLT 8-m telescope, it is possible toefficiently obtain albedos and sizes of dangerous objects, as demonstratedby some recent observations of (99942) Apophis.19

The availability of new polarimeters in the future is needed toensure a stable development of asteroid polarimetry. Recently, a few newinstruments, including a single-color polarimeter using phototube detectors,has entered into operations at the El Leoncito observatory. Compared to theolder Torino photopolarimeter, which performs simultaneous measurementsin five colors (UBVRI), the new instrument has a more limited spectralcapability, but this is compensated by an increased sensitivity in V

light, which should allow the observers to observe fainter and darkerobjects. Another instrument which has been developed recently is theCCD polarimeter built at the Asiago observatory (Italy), which has startedrecently to produce its first data.17

Coupled with the recent availability of the VLT telescope for a numberof asteroid polarimetry campaigns, we hope that the above-mentioneddevelopments 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

Page 45: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch03

32 A. Cellino et al.

the next years very important advances in several fields of modern asteroidresearch, including primarily the study of the properties of asteroidalsurfaces.

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 (Universityof 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 andM. 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. Penttila, 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. Delbo, A. Cellino and E. F. Tedesco, submitted to Icarus (2006), in press.

Page 46: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

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

LOW ENERGY CHARGED PARTICLE MEASUREMENTBY JAPANESE LUNAR ORBITER SELENE

Y. SAITO∗, S. YOKOTA, K. ASAMURA, T. TANAKA and T. MUKAI

Institute of Space and Astronautical ScienceJapan Aerospace Exploration Agency

3-1-1 Yoshinodai, SagamiharaKanagawa 229-8510, Japan

[email protected]

SELenological and ENgineering Explorer (SELENE) is a Japanese lunar orbiterthat will be launched in 2007. The main purpose of this satellite is to studythe origin and evolution of the Moon by means of global mapping of elementabundances, mineralogical composition, and surface geographical mappingfrom 100 km altitude. Plasma energy Angle and Composition Experiment(PACE) is one of the scientific instruments onboard the SELENE satellite. Thescientific objectives of PACE are (1) to measure the ions sputtered from thelunar surface and the lunar atmosphere, (2) to measure the magnetic anomaly

on the lunar surface using two electron spectrum analyzers (ESAs) and amagnetometer 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-dimensionaldistribution function of low energy electrons below 15 keV, while IMA and IEAmeasure the three-dimensional distribution function of low energy ions below28 keV/q.

1. Introduction

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

33

Page 47: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

34 Y. Saito et al.

no new information about the low energy charged particles around theMoon was obtained. SELenological and ENgineering Explorer (SELENE)is a Japanese lunar orbiter that will be launched in 2007. SELENE willstudy the origin and evolution of the Moon by means of global mapping ofelement abundances, mineralogical composition, and surface geographicalmapping from 100 km altitude. One of the scientific instruments, Plasmaenergy Angle and Composition Experiment (PACE), was developed formaking comprehensive three-dimensional plasma measurement around theMoon. After describing the science objectives of PACE, the configurationof 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 ofthe most important aims of PACE on SELENE. Ground-based observationsrevealed the existence of tenuous alkali-atmosphere around the Moon inthe end of 1980s. Potter and Morgan12 discovered the existence of Naand K atmospheres above the sunlit limb of the Moon for the first time.Since then several generation mechanisms are proposed for the rarefiedlunar alkali-atmosphere.13 Sputtering caused by the solar wind ions hasdrawn considerable attention, because it produces secondary particles thatreflect the lunar surface composition. On the other hand, Potter et al.14

found that solar photons play a dominant role in desorption of the lunaralkali atmosphere. According to them, sputtering of the lunar surface onlycontributes to the creation of the lunar Na atmosphere by enhancingdiffusion of Na to the lunar surface. Sputtered or desorped particles fromthe lunar surface are mainly composed of neutrals, which are ionized bysolar photons and electrons. Both ionized particles and sputtered/desorpedions are accelerated and transported by the solar wind in cycloidal motion.The ion analyzer on board the AMPTE satellite obtained mass spectra ofpicked-up ions at several Rm distance behind the Moon.15 The distributionof tenuous lunar atmosphere will be obtained by the in-situ low-energyion measurement by PACE. PACE will also remotely reveal the globalcomposition of the lunar surface by means of detecting ions sputtered by thesolar wind and tracing them back to their sputtering point like to laboratorysecondary ion mass spectrometry.16

Page 48: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

Low Energy Charged Particle Measurement 35

2.2. Electron reflectometer

The solar wind electrons and the magnetotail electrons that reach theMoon 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 fieldwill be mirror reflected back to SELENE. Measuring the pitch angledistribution of the reflected electrons, the remnant magnetic field onthe lunar surface can be deduced.4,17 The previous remnant magneticfield measurements using mirror-reflected electrons were conducted bythe Apollo 15,16 sub-satellites whose orbits were limited around theequator region of the Moon.18,19 Lunar Prospector also measured remnantmagnetic field on various areas of the Lunar surface using electronreflectometer.20,21 The SELENE PACE-electron spectrum analyzer (ESA)sensors will survey the remnant magnetic field on almost all the lunarsurface with higher spatial resolution than previous electron reflectometermeasurement.

2.3. Moon–solar wind interaction

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

Page 49: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

36 Y. Saito et al.

2.4. Moon–Earth’s magnetosphere interaction

The Moon provides “an eclipse” of the Earth’s magnetotail. Near the Moonin 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. Itis very important to know origins and dynamics of these plasmas. With thisknowledge, we understand the formation mechanism of the magnetosphereand the structure of the magnetotail. The Moon provides an excellentexperimental tool to us. In the first approximation, the Moon is not anyobstacle for the magnetic fields. However, the Moon is a strong absorber forplasmas. When the Moon is in the magnetotail, plasmas from the distanttail cannot reach any observer on the earthward side of the Moon, whereasplasmas from the Earth cannot reach any observer on the tailward sideof the Moon. Therefore, we will be able to discriminate the tailward-flowing plasmas from the earthward-flowing plasmas. It will be possibleto identify correctly the tailward-flowing component and the earthwardflowing component in the counter-streaming flows. Furthermore, it will bepossible 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 beencarried out only with the spacecraft Geotail. Geotail provided mainly thesurvey of plasmas in the Sun–Earth direction, not in the cross section ofthe magnetotail. The spacecraft around Moon makes a detailed survey ofplasmas in the cross section of the magnetotail, so that we would be able toexamine various boundaries in the magnetotail (i.e., the distant bow shock,the magnetopause, the magnetopause boundary layer, the low latitudeboundary layer, the plasma sheet-tail lobe boundary). In the magnetotail,various plasma populations are controlled by the interplanetary magneticfield (IMF). The loading of ionospheric O+ in the magnetotail dependsstrongly on the IMF. With measurement of O+ distribution in the tailcross section, we will be able to understand the loading and transportmechanisms of ionospheric ions in the tail lobe/mantle. The draping ofthe magnetic field lines near the magnetopause and the deformation of themagnetotail will be correctly evaluated, in order to understand changes inthe magnetotail structure. Furthermore, it will be possible to find magneticreconnection process in the tail magnetopause region.

Page 50: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

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 electrostaticanalyzer with angular scanning deflectors at the entrance and toroidalelectrodes inside (see Fig. 1). The field of view (FOV) is electrically scannedbetween ±45 around the center of the FOV that is 45 inclined from theaxis of symmetry. With two ESA sensors that are installed in the +Z and−Z surface of the spacecraft, the three-dimensional electron distributionfunction is observed. The upper and lower angular deflectors are suppliedwith high voltage which are swept between 0V and +4kV. The innertoroidal electrode is also supplied with high voltage swept between 0V and+3kV simultaneously with the angular scanning deflectors. The electronscoming through the angular scanning deflectors are attracted down towardthe inner electrode by the action of the applied potential. Only the electronswith specific energy range can further travel down to the exit of the

Fig. 1. Cross section of ESA-S1 and ESA-S2.

Page 51: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

38 Y. Saito et al.

electrodes. There also placed a slit and a mesh supplied with slightlynegative voltage (−2.5V to −10V) to reflect the secondary electrons thatgive spurious counts. The electrons passing through the electrode enter tomicro-channel plate (MCP) and are intensified to detectable charge pulses.Finally, the charge pulses are received by one-dimensional circular resistiveanode. The positions where the charge pulses are detected correspond tothe incident azimuthal directions of the electrons. Tables 1 and 2 summarizethe 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 structureas the ESA sensors and a linear electric field (LEF) time of flight (TOF) ionmass analyzer24–26 (see Fig. 2). The IEA sensor consists of only an energyanalyzer that is the same as the energy analyzer of IMA (see Fig. 3). Theupper and lower angular deflectors of the energy analyzer are supplied withhigh voltage which are swept between 0V and +5kV. The inner toroidalelectrode is also supplied with high voltage swept between 0 V and −4 kVsimultaneously with the angular scanning deflectors. Between the toroidalelectrode and the angular deflectors, there exist a pair of electrodes thatserve as sensitivity control electrodes. Since the flux of the solar wind ions

Table 1. Specifications of ESA-S1.

Energy range 5 eV–10 keVEnergy resolution 15% (FWHM)Energy sweep step 32

Field of view 2π str.Angular resolution 5 × 8 (FWHM)Time resolution 1 secondFOV sweep range 45 ± 45g-factor (5 × 22.5) 10−3 cm2 str keV/keV

Table 2. Specifications of ESA-S2.

Energy range 5 eV–15 keVEnergy resolution 10% (FWHM)Energy sweep step 32Field of view 2π str.Angular resolution 5 × 8 (FWHM)Time resolution 1 secondFOV sweep range 45 ± 45g-factor (5 × 22.5) 2 × 10−4 cm2 str keV/keV

Page 52: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

Low Energy Charged Particle Measurement 39

Fig. 2. Cross section of IMA.

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

Page 53: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

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 chargestate and enter into the mass analyzer as neutral particles. These neutralparticles are detected by an anode that is in the center of the positionsensitive anode. These signals are used as stop signals. The mass/charge ofthe 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 reflectedby the linear electric field whose intensity is proportional to the distancefrom the entrance point. The reflected ions generate secondary electronswhen they collide with the top part of the mass analyzer. These electronsare accelerated and detected by the center anode, which generate the stopsignals. Since the TOF of the reflected ions is proportional to the squareroot of the mass of the ions, the mass of the incident ions can be determined

Page 54: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

Low Energy Charged Particle Measurement 41

Table 3. Specifications of IMA.

Energy range 5 eV/q–28 keV/qMass range 1–60Energy resolution 5% (FWHM)Energy sweep step 32Mass resolution m/∆m∼15Field of view 2π str.Angular resolution 5 × 10 (FWHM)Time resolution 1 secondFOV sweep range 45 ± 45g-factor (5 × 22.5) 10−6–10−4 cm2 str keV/keV (variable)

Table 4. Specifications of IEA.

Energy range 5 eV/q–28 keV/qEnergy resolution 5% (FWHM)Energy sweep step 32Field of view 2π str.Angular resolution 5 × 5 (FWHM)Time resolution 1 secondFOV sweep range 45 ± 45g-factor (5 × 22.5) 10−6–10−4 cm2 str keV/keV (variable)

precisely without being affected by the angular scattering and the energydegradation 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 andelectrons at 100km altitude around the Moon. With the minimum timeresolution of 1 second, high spatial resolution measurement of magneticanomalies on the lunar surface will be made. Since nobody has evermeasured the three-dimensional distribution function of low energy ionsat 100km altitude around the Moon, it is expected that many unresolvedproblems concerning the lunar plasma environment will be elucidated bythe PACE observation in the near future.

Acknowledgments

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

Page 55: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

42 Y. Saito et al.

unstinting help in developing PACE sensors. SELENE-PACE sensors weremanufactured by Mitaka Kohki Co. Ltd., Meisei Elec. Co., HamamatsuPhotonics 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, Science266 (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. Mobius, 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.

Page 56: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

Low Energy Charged Particle Measurement 43

18. K. A. Anderson, R. P. Lin, R. E. McGuire, J. E. McCoy, C. T. Russell andP. 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. Acunaand 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.

Page 57: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:55 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch04

This page intentionally left blankThis page intentionally left blank

Page 58: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

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

A JOVIAN SMALL ORBITER FOR MAGNETOSPHERICAND AURORAL STUDIES WITH THE

SOLAR-SAIL PROJECT

Y. KASABA∗,†,§, T. TAKASHIMA†, H. MISAWA‡ and JOVIAN SMALL ORBITERSUB-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 CenterTohoku University, Sendai, Miyagi 980-8578, Japan

[email protected]

The Solar-Sail Project has been investigated by JAXA as an engineering missionwith a small orbiter into the Jovian orbit. This paper summarizes the basicdesign of this project and possible Jovian system studies by this opportunity.

The large-scale Jovian mission has been discussed as a long future plansince the 1970s, when the investigation of the future planetary explorationprogram started in Japan. The largest planet and its complex planetary systemwould 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 astar. (2) The Jovian-type magnetosphere: the structure and processes of thelargest and strongest magnetosphere in the solar system. (3) The structure,composition, and evolution of Jupiter and its satellite system. The smallJovian orbiter accompanied with the Solar-Sail Project will try to establishthe technical feasibility of such future outer planet missions in Japan. The

main objective is the second target, the Jovian magnetospheric and auroralstudies with its limited payload resources.

1. Solar-Sail Project

1.1. Objectives

The Solar-Sail project (Fig. 1) is another engineering mission, the nextplan after the Hayabusa spacecraft to the asteroid Itokawa. Its mainobjective is to establish the method to explore the outer solar system by the

§Corresponding author.

45

Page 59: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

46 Y. Kasaba et al.

Fig. 1. The Solar-Sail project: an image at the flyby of Jupiter (Courtesy: NewtonPress).

developments of (1) the advanced combined propulsion systems by solar-sailand electrical propulsion technologies, (2) the spacecraft bus technologiespowered by solar cell in the outer solar system, and (3) the complex systemwith 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 Trojanasteroids. The spacecraft will consist of three modules: the motherspacecraft, the daughter spacecraft, and an entry probe (Fig. 2). The motherspacecraft (300–500kg) will make the multiple flybys at main-belt asteroidsand Jupiter, and finally reach to Trojan asteroids. The daughter spacecraft(∼ 100 kg) will be separated just before the arrival to Jupiter and will bethe first Japanese Jovian orbiter. The Jovian entry probe (∼ 30 kg: the

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

Page 60: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

Jovian Small Orbiter for Magnetospheric and Auroral Studies 47

option unit at this moment) will be separated from the daughter spacecraftand will enter into the Jovian atmosphere.

For those units, the scientists from multiple communities have proposedthe observational plans in order to take this important opportunity to travelto the outer solar system and stay there for an extended period in time. Thelatest Solar-Sail project is, therefore, not only an engineering mission butalso the first multi-scientific platform in deep space, which is little similarto the Space Flyer Unit concept, a multi-scientific platform in the low Earthorbit launched in 1995 by the NASDA and the ISAS.

The science goals of the mother spacecraft during cruise and the flybyswill be (1) the cosmic infrared background observation in low backgroundenvironment, (2) the gamma-ray burst detection sites located far from theEarth, (3) the dust measurement with a largest detector formed by the sailsheet, 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 tothe atmosphere.

1.2. Current development status and future base plan

This project started in the late 1990s. Several engineering tests for theestablishment of the technical baseline of this mission have succeeded:

2000: The formal base study started. The main objectives were “theobservation 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 in2005 (http://www.isas.jaxa.jp/e/snews/2004/0809.shtml).

Page 61: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

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 testunits aboard the S-310 sounding rocket successfully unfurl two-typesails (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.

Page 62: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

Jovian Small Orbiter for Magnetospheric and Auroral Studies 49

Fig. 5. The orbits of the Jovian small orbiter after the Jupiter orbit insertion.

The original baseline schedule was shown in Fig. 4. Because of thelaunch delay, the overall schedule is not fixed yet:

2012: Launch by the M-V rocket: The Earth Swing-by (×2) and the Main-belt asteroid flyby (×1–2) are planned before the arrival at Jupiter.

2017: Arrival at Jupiter: The mother spacecraft makes swing-by. The daugh-ter spacecraft is separated from the mother spacecraft and entersinto the Jovian polar orbit (Fig. 5). The entry probe (option) willbe separated from the daughter spacecraft and enters the Jovianatmosphere.

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 Themain objective of this orbiter is the technology establishment of futureouter planetary orbiters, beyond the heritage of the past planetary projectsof ISAS and JAXA, the Nozomi mission to Mars (1998–2003), thePlanet-C mission to Venus (Launch: 2010), and the BepiColombo Mercury

Page 63: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

50 Y. Kasaba et al.

Magnetospheric Orbiter (Launch: 2013). It will be “the pathfinder” for thefuture full-scale Jovian mission. The planned small Jovian orbiter will provethe technical feasibility for (1) the power supply with solar cell formed onthe sail and the thermal design under weak sunlight condition (∼ 1/30 ofthe Earth-orbiting spacecraft) and (2) the sustainability under the activeradiation 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 planetin the solar system would be solved by several main objectives:

1. The structure of a gas planet: The internal and atmospheric structuresof the giant gas planet which could not become a star. It follows theobjectives of the Venus and Mercury studies based on the Planet-C andthe BepiColombo missions.

2. The Jovian-type magnetosphere: The structure and processes of thelargest and strongest magnetosphere in the solar system, driven by itsown 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 SCOPEmissions planned as future projects, and the Herman magnetosphericstudies by the BepiColombo mission.

3. The structure, composition, and evolution of the satellites: the studyof the Jovian system and its evolution. It follows the objectives of theSELENE (Lunar orbiter), Planet-C, and BepiColombo missions. Theastrobiology topics might be included as an optional target.

The main objective of the small orbiter accompanied with the solar-sail project is the second target with its limited payload resources, i.e., thequick-look observations of the Jovian magnetosphere and auroral regions, inorder to study the Jovian-type magnetosphere–ionosphere coupling and theplanet and satellite coupling processes. Those objectives will be shared withand 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 UnitedStates might be valuable, if it is realized.

2.2. The base plan

The small Jovian orbiter will be separated from the mother spacecraftbefore its flyby at Jupiter, and will enter to the polar orbit around Jupiter

Page 64: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

Jovian Small Orbiter for Magnetospheric and Auroral Studies 51

by itself. This orbit is selected for the studies of the polar regions of bothhemispheres (Fig. 5), the avoidance of the passage in the Jovian radiationbelt, and the targeting of and the telecommunication with the entry probe(option). By its long orbital period (∼ 230 days) and expected nominallifetime (1 year) which is limited by the fuel resource, we consider that thein-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. Inthe present assumption the expected allocated resources are less than 4 kgand less than 5W in average. Under these restrictions, the main scientificobjective is confined to the quick-look observation of the most activeobjects in the Solar system, by the crossing of the Jovian polar regions.It will enable us to study (1) the magnetosphere–ionosphere coupling inthe Jovian-type magnetosphere driven with planetary rotation energy, and(2) the interaction between the planet and satellites via plasma processwhich 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 along time, it will also be expected to act as a solar wind monitor for thecooperated studies with ground-based and Earth-orbiting telescopes for theJovian auroral and magnetospheric studies.

The payload is mainly assumed under the heritage of the instrumentsdeveloped for the Planet-C mission and the BepiColombo MercuryMagnetospheric 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 A spin-type spacecraft:diameter: 70 cm, mass: 98 kg

Spacecraft Power 90W (no communication),consumption 190 W (with communication)Life > 1 Earth yearPower system Solar cell panel formed on the sail film

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

Payload Resource Mass: < 4 kg, power: < 5WThermal Nominal: −30CTelemetry 16 bps/1 kbps (X/Ka, 60 cm Φ, 20W HGA)

Operation: 1 h/day (40 bps/ave = 0.4MB/day) or1 h/week (6 bps/ave = 0.06MB/day)

Page 65: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

52 Y. Kasaba et al.

priorities:

(a) Remote sensing: Remote sensing capabilities coupled with the motherspacecraft, the entry probe, and other orbiters (ex. Juno), and ground-based and Earth-orbiting telescopes: The priority-1 is the Jovianaurora and lightening observations which will collaborate with in-situmeasurements. Priority-2 is the decametric and hectometric radioobservations which will collaborate with radio, infrared, and visibleobservations from the Earth.

(b) In-situ observations: The observations in the particle accelerationregions and the monitoring of the magnetospheric and solar windplasmas. Priority-2 is the magnetospheric and solar wind observations,i.e., activities inside the auroral acceleration region, the large-scalecurrent system, etc., which will collaborate with radio, infrared, andvisible observations from the Earth.

The current model payload plan is summarized in Table 2: (1) remotesensing by an auroral monitor, a radio monitor, (2) in-situ measurements bya magnetometer, high-energy and low-energy particle detectors, and (3) anintegrated 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-bandradiometer (lightening detection), and a simple infrared radiometer (H2Odetection), 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 Mass (kg) Objectives

Remote Line image 1.5 Aurora/Lightening(Priority-1)Radio 0.5 Jovian radio waves(Priority-2)Camera 0.5 Navigation(option)

In-situ Magnetic field 0.5 B-field and current structure(Priority-2) (no boom)High-E particles 0.7 Particles in polar and equatorial(Priority-2) regionLow-E ions 0.7 Solar wind monitor(Priority-2) and outflow ions

Common Digital/Power – Unified to the system

Page 66: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

Jovian Small Orbiter for Magnetospheric and Auroral Studies 53

included. The entry probe will be dropped into the middle to high latituderegions, not covered by the Galileo probe.

Some studies and developments for this mission are investigated withthe engineering team in order to achieve good scientific results: (1) Thedesign of the sail film: the current canceling for the reduction of magneticinterference and the installation of wire antennas. (2) The developmentof sensitive and light-weight sensors and optics. (3) The unified dataprocessing and power systems with the bus system. (4) The system studiesfor the small orbiter itself, including the communication, power, structureand orbit/attitude control systems.

The establishment for the collaboration with ground-based and spacetelescope facilities and US Juno mission will be in near future, after thestart of Phase-B. Japanese space science communities will expect theestablishment of new-era through this project.

Acknowledgments

The authors wish to express their sincere gratitude to all membersparticipating and contributing to the studies of the Jovian Small OrbiterSub-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 (KochiInst. Tech.), K. Imai (Kochi NCT), T. Hada, K. Nakajima, K. Yumoto(Kyushu Univ.), S. Takeuchi (Fukuoka Univ.), T. Sato (KumamotoUniv.), 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 experimentof 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 withJ. Kawaguchi and Solar-Sail WG, Jovian Small Orbiter for Magnetosphericand Auroral Studies. “Solar Sail Project”, Proc. IAA International Conferenceon Low-Cost Planetary Missions (ICLCPM), (2005) pp. 08-B1.

Page 67: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch05

This page intentionally left blankThis page intentionally left blank

Page 68: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

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

DESCRIPTION OF A NEW 400MHz BANDWIDTH CHIRPTRANSFORM SPECTROMETER

LUCAS PAGANINI∗ and PAUL HARTOGH

Max Planck Institute for Solar System ResearchPO Box 37191, Katlenburg-Lindau, Germany

[email protected]

A new chirp transform spectrometer (CTS) with a bandwidth of 400MHz anda spectral resolution of 100 kHz has been developed. The CTS is deviced usinga digital chirp generator and a preprocessing unit based on a ComplementaryMetal Oxide Semiconductor (CMOS) and an Application-Specific IntegratedCircuit (ASIC). A build in PC 104 computer handles the process controland the external communication via Ethernet and a Transistor-TransistorLogic (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 ozonesystem. Astronomical observations have been performed using the HeinrichHertz 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 unlimitedspectral resolution. A high frequency signal for instance from the sub-millimeter range is down-converted by a local oscillator signal to a lowerfrequency band in which electronic spectrum analysis techniques canbe applied. In atmospheric spectroscopy or radioastronomy, the down-converted 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 methodis required. Spectrometers with nearly 100% duty cycle are called real-time spectrometers, since the a priori data rate of the calculated spectrumis the same as the incoming time domain signal. Reduction of the datarate is done by averaging the power spectra. Real-time or nearly real-time spectrometers, such as filterbanks, acousto-optical spectrometers, andautocorrelators, are widely spread especially in radioastronomy. Recently

55

Page 69: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

56 L. Paganini and P. Hartogh

high bandwidth Fast Fourier Transform (FFT) spectrometers have beendeveloped.

The chirp transform spectrometer (CTS) has a successful history ofmore than two decades in ground-based, airborne, and space missions.The first millimeter wave heterodyne spectroscopy with CTS has beenperformed in the middle of the 1980s by detecting the 142GHz rotationaltransition of ozone in the Earth middle atmosphere.1–3 Since the early1990s, continuous ground-based measurements of water vapor (22/183GHz)and ozone (142GHz) have been performed. Furthermore, upon the integra-tion of the first CTS spectrometer into the Heinrich Hertz submillimetertelescope (HHSMT) on the Mt. Graham in Arizona, USA, it has beenapplied to address a wide range of topics of modern astrophysics.4,5 Thesetopics go from questions about comets, planetary atmospheres, and theinterstellar medium in the galaxy, to investigations related to the earlyUniverse. Recently, CTS provided a high-quality spectra of comet 2002 T7(LINEAR) and the Earth from the Microwave Instrument for the RosettaOrbiter (MIRO), the first deep space mission carrying a submillimeterheterodyne spectrometer.6,7

The CTS has been proven to be a very reliable and accurate spectro-meter (see for instance Refs. 8 and 9). The newly developed 400-MHz CTScombines the advantage of broader bandwidth analysis with keeping thecharacteristics of previous CTSs. The new techniques applied for creatingthe chirp signal, which also involves similar digital techniques as used inthe SOFIA-GREAT (Stratospheric Observatory for Infrared Astronomy —German REceiver for Astronomy at THz frequencies) CTS,10 yield a perfectmatching to the dispersive properties of the compressor unit of the CTSand improvement of the signal-to-noise ratio (SNR) up to 60 dB.

The CTS is based on the Chirp transform,11,12 an algorithm derivedfrom the Fourier transform and implemented by linear frequency-modulatedwaveforms and their matched filters.

In the CTS, the input signal is first modulated by a chirp. Thereby, forinstance a fixed frequency signal becomes linearly modulated. The lattersignal is fed into a dispersive filter (called compressor) with a delay timedepending on the frequency and equal dispersive characteristics. The filteroutput for the fixed frequency now looks like a single peak at a specific timegiving 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 thesurface of a piezoelectrical crystal base, whose displacement amplitudesundergo exponential decay beneath this surface. The wave pattern of

Page 70: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer 57

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

2. Development and Design

The CTS involves two parts: the analog and the digital. The analogpart is integrated by the RF stage, which includes mixers, amplifiers,filters, doublers, splitters, and SAW filters. On the other hand, thedigital part involves the chirp generation board, the data processingand 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 inreal time with 100% efficiency and then provide the spectral informationsupplied 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 suchtechnique is the possibility to fix every aspect of the chirp signal, in orderto achieve the perfect dispersive matching required as above mentioned,providing high dynamic range, since the outcoming chirp waveform has alarge SNR. It is important to highlight this because it can be corrected bymeans of digital properties changes. The created signal in the chirp boardhas a 400-MHz bandwidth centered at 250MHz with a dispersion time of20 µs and then it is frequency up-converted using RF mixers.

In the SOFIA CTS, the signal is quadrature-modulated and up-converted before the RF stage where it is frequency-tripled. The newimplementation allows a broader bandwidth which only needs to be doubledin the RF stage (Fig. 1), achieving not only a bandwidth up to 800MHz,but also improving the SNR up to 60 dB.

Page 71: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

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 intothe compressor (SAW filter). By mixing the signal with a complex source, after theconvolution 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 fourpairs of ADCs and continue with the digital squaring, preprocessing, and integration inthe ASIC board. At the end, it goes to an embedded computer through an ISA–ASICinterface.

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

Page 72: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

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 digitalacquisition by an oscilloscope for a 2.1-GHz pulse input.

frequency of the input atmospheric signal, which is in the 1.9–2.3GHzrange. The newly modulated signal is fed into the SAW filters which have a400-MHz bandwidth (800–1200MHz range). The output of the filter in thetime domain represents the analog spectrum of the input signal, mappedon a 10-µs time interval (Fig. 2).

Due to intrinsic properties of the chirp transform, the duty cycle of anexpander–compressor scheme is only 50%. This means that the setup takes20 µs to perform a transform. However, it produces only half the time ofuseful spectral information. For that purpose, two branches are combinedthrough a commutator with a switching period of 10 µs with the idea of100% efficiency.

Later on, the signal is down-converted to the base-band and the real andimaginary components are obtained by mixing the signal with a complexsource. In the ASIC board, these two components are digitally acquiredwith a set of eight 100MSPS analog-to-digital converters (ADC). Thesignal is digitally analyzed by a preprocessor integrated in an ASIC chipwith low power consumptions. This preprocessor computes the power fromthe complex spectrum, where the signal is numerically squared (real andimaginary parts), added, and finally mapped into a 4096-channel memorywhere it is integrated. Furthermore, it also handles the synchronizationsignals needed to control the timing of the two branches. This also involvesa signal which defines the generation start of the chirp signal at a specific

Page 73: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

60 L. Paganini and P. Hartogh

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

3. Test Measurements

3.1. Chirp signal

Especially important is the understanding of the behavior of the chirpsignal. There are a number of quality requirements for the chirp signallike a small passband ripple, a good SNR, small phase deviation, groupdelay, and the corresponding chirp rate (µ). This is analyzed in Fig. 3.

Most of these methods, based on Fourier transform, analyzed this signalin a stationary regime. This is explained by the fact that the transform takesas 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 eventoccurs 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 isused with the aim of observing the variation of frequency components withtime, e.g., to determine whether harmonics observed in a spectrum analyzerare influencing the 10-µs transformation interval or not. If that is the case,noise is added. This is especially important during the design process ofthe instrument, when different RF devices are evaluated. Furthermore,it is possible to estimate if the influence of undesired harmonics can beneglected or should be suppressed, e.g., by adding filters. As it can be seenin Fig. 4, it is possible to observe deterioration in the signal after mixing,amplifying, doubling, and filtering processes producing undesirable spuriouswhich affect the overall behavior of the instrument. This was improved byselecting a better RF mixer before the doubling process and the additionof filters.

3.2. Power linearity and dynamic range

Three effects constrain the instrument’s dynamic range: the high insertionloss of the SAW devices (more than 40 dB), the noise and interferencesintroduced during the RF signal processing, and the compression point ofthe different passive (mixers) and active components. The maximum signal

Page 74: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May

29,2007

6:5

6sp

i-b486

AO

GS

2006

-P

SVolu

me

9in

x6in

ch06

Descrip

tion

ofa

New

400M

Hz

Bandwid

thChirp

Tra

nsfo

rmSpectro

meter

61

Fig. 3. Chirp signal study tools by acquiring and digitalizing it in time domain with DSO. Upper left: Calculated magnitude. Uppercenter 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.

Page 75: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

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), whilethe lowest is the output of the SAW device (−50 dBm). The inclusion ofdigital techniques into the CTS design and the 60-dB SNR allows a widerange adaptability, which assure a proper dynamic range setup.

During this evaluation, a noise source with ∼1 dB flatness and +10 dBmpower level was used as input in the frequency range of 1.9–2.3GHz;together with high accuracy and low repeatability RF step attenuators with±0.15 dB accuracy. The linear response of the instrument was analyzed ina 60-dB input range by stepping the input power in 1 dB steps. The resultsin Fig. 5 show a dynamic range with a maximum deviation from linearityof ±1 dB equal to 35 dB and ±0.1 equal to 20 dB; and an optimum inputpower level of −35 dBm.

This analysis allows to predict the overall response of the instrumentby calculating the mean of the whole range of channels. But, for a moredetailed evaluation of the instrument response, i.e., each single channel, anovel tool plots a 3D graph, which allows to observe the given response ofthe spectrometer by means of channel number, power input, and outputcounts/cycle. This is specially important for establishing not only lineardynamic range, but also a wide range of unexpected behaviors not possibleto identify in previous analysis, e.g., mismatch of dispersive propertiesbetween SAW filters and the fed signal, possible regions which goes intosaturation 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

Page 76: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

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 outputpower plotted against input power and number of channel. This method allows to seepossible deteriorations in the system. Down left: output power obtained by insertinga 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 timebetween two calibrations. As any additional noise above the radiometriclevel is unfavorable, one has to find the optimum integration time, wherethe impact of drift contributions is nearly negligible. In other words, theradiometer equation is valid within the white noise part, i.e., before theAllan variance minimum. This minimum describes the turnover point wherethe radiometric noise with a slope of −1 in the logarithmic plot becomesdominated by the additional and undesired drift noise.

After warming up 43,000 spectra, with 1 s integration time each andan 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,

Page 77: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

64 L. Paganini and P. Hartogh

Fig. 6. Allan variance calculation. Left: output count per cycle for the channel 2045using a noise source during 43,000 spectra integrated in 1 s each. Right: the Allan varianceof 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 spectroscopicAllan 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 tothe temperature stability of the compressor filters. Thus a good thermalstabilization of these devices is essential. The thermal stabilization we usedresults 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 thefirst zero crossing in the frequency domain at 1/TC, defining the spectralresolution of the spectrometer. The dispersion time of 10 µs gives us a 100-kHz 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 ofthe sample curved for each single channel in the whole range, using a stepsize of a 10th nominal resolution. Moreover, deviations in the spectral scalelinearity were further retrieved. This property describes the relationshipbetween the input frequency and the corresponding expected frequency fora specific channel index, obtaining values smaller than 6% of the spectralFWHM for the complete operational bandwidth of the instrument.

Page 78: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

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 obtainedcalculation from the expected nominal value (∼ 0.06 of a channel).

Table 1. Specifications for the 400-MHz bandwidth chirptransform spectrometer.

Center frequency 2.1GHzSpectral resolution (noise equivalent) 1/TC 100 kHzSpectral resolution (FWHM) 121.2 kHzChannel spacing 97.6 kHzBandwidth (−3 dB) 400 MHzOptimal RF-Power input −35 dBmNoise dynamic range (±1 dB) 35 dBNoise dynamic range (0.1 dB) 20 dBFrequency linearity ±4 kHzAbsolute allan-variance time 173 sChannels 4096Power consumption < 30W

4. Observations and Results

The new 400-MHz bandwidth CTS was tested under real observingconditions in order to assess the expected performance. First light wasobserved on December 2005 by measuring the O3 line using a 142-GHzozone system as test facility in Katlenburg-Lindau (Fig. 8).

In order to test its behavior by observing astronomical objects, theinstrument was installed at the HHSMT on the observing run of the73P/Schwassmann-Wachmann 3 comet in May 2006, when it had its closestapproach.

The HHSMT is located at 3200m altitude on Mt. Graham. It hasa 10m-diameter main reflector, the absolute pointing accuracy is about2′′, with a tracking accuracy of better that 1′′.14 The receivers used werethe SSB 1.3mmJT and the MPIfR SIS-345. It is equipped with severalbackends: a 218-MHz bandwidth CTS, two AOSs, and filterbanks.

Page 79: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

66 L. Paganini and P. Hartogh

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

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

The observed components were mainly fragments B and C focusingon the HCN(3-2) and HCN(4-3), H2CO, CO and CS lines, under goodweather conditions (Fig. 9). The scope of this test was not only to showthe spectrometer’s high resolution which is significantly important for thestudy of narrow features (e.g., in cometary emission lines), but also toobserve broader lines, and thus demonstrate the different capabilities thatthe CTS can reach. Some observations were performed in Mars and some

Page 80: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

Description of a New 400 MHz Bandwidth Chirp Transform Spectrometer 67

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

Fig. 11. Observation of the N7538IRS1 star. CS(5-4) line.

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

5. Conclusions and Outlook

We have developed a new 400-MHz bandwidth and 100-kHz spectralresolution CTS. The digital techniques used for the chirp signal generationwere improved allowing a SNR up to 60 dB. In this publication, a completecharacterization of the spectrometer is presented. Furthermore, test resultswere obtained in atmospheric science and astronomical observation by

Page 81: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch06

68 L. Paganini and P. Hartogh

integration of the spectrometer in a 142-GHz ozone system and by usingthe HHSMT .

Considering the demand for broad bandwidth for various wavelengthspectroscopy, future development will concentrate on the following areas:(i) broader bandwidth SAW filters development; (ii) new analysistechniques to fully and completely analyze different and essential stages inthe system with orders of accuracy never achieved before, and thus eliminatepossible nonlinearities impossible to measure by nowadays methods; (iii) byoptimizing power consumption, size and weight, the present sensitivity andresolution of the spectrometer can considerably widen different aspectsstudied 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 comparableor, which is always the main aim, better performance of previous CTSs.

Acknowledgments

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

References

1. P. Hartogh, Messung der 142 GHz Emissionslinie des atmospharischen Ozons,PhD thesis (University of Gottingen, 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.

Page 82: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

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

FORMATION OF ALUMINA NANOPARTICLESIN PLASMA

MAMI KURUMADA and CHIHIRO KAITO∗

Department of Physics, Ritsumeikan University, 1-1-1 NojihigashiKusatsu-shi, Shiga 525-8577, Japan

[email protected]

γ-Alumina nanoparticles with a polyhedral shape based on an octahedralshape were produced with a plasma field of Ar–O2 gas mixture (Ar: 7Torr,O2: 3Torr). The obtained infrared spectrum showed a characteristic strongabsorption peak at 7.2 µm, which had never been previously observed in thealumina nanoparticles produced without a plasma field. The plasma fieldmainly affected the surfaces of the alumina nanoparticles and changed theirmorphology and gas adsorbability. The 7.2 µm, absorption peak derived fromsurface hydroxyl was considered to be due to the existence of activated Al inthe 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 mostthermodynamically stable phase well known as corundum. The γ-, δ-, η-,λ- and θ-phases possess structures based on spinel, with the face-centeredcubic (fcc) packing of oxygen anions and aluminum cations distributedat their octahedral and tetrahedral sites. γ-Alumina is typically producedfrom boehmite and is transformed into the δ-, θ- and α-phases by heating,[AlO(OH)] (300–500C) → γ (700–800C) → δ (900–1000C) → θ

(1000–1100C) → α-alumina.1 These metastable polymorphs with spinel-based structures differ in the distribution pattern of Al cations in anoxygen sublattice. The occupation percentages of the tetrahedral (10%)and octahedral (47%) sites of γ-alumina were estimated by moleculardynamic (MD) simulation.2 The migration of Al cations from tetrahedralto octahedral sites occurs during transformation from γ- to δ-, θ- andfinally α-alumina.2 The distribution of Al cations in the oxygen sublattice

69

Page 83: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

70 M. Kurumada and C. Kaito

of γ-alumina is random and becomes periodic during transformation. θ-Alumina has a more ordered distribution than δ- or γ-alumina.1

A γ-alumina grain is a candidate material for 13 µm absorptionobserved around certain post-AGB stars.3 In our previous study, weproduced δ-alumina nanoparticles by evaporating the aluminum in anAr–O2 gas mixture.4 The aluminum clusters evaporated immediatelyreacted with oxygen and alumina nanoparticles were formed. We showedthat δ-alumina, as well as γ-alumina, has a strong absorption at 13 µm.However, the characteristic infrared absorption peaks of δ-alumina wereobserved 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 cationsin the tetrahedral sites.

On the other hand, we began to realize that a plasma field playsan important role in the formation of grains. A quenched carbonaceouscomposite, which is considered as a candidate material for 220nmabsorption, was produced with a CH4 plasma.5 A nitride grain, such asSi3N4, was also produced with a plasma field.6 The plasma field alsodeeply affects the formation of crystalline silicate grains. The study basedon the ISO spectra of a large number of stars indicates that enstatite(MgSiO3) is more abundant than forsterite (Mg2SiO4) by a factor of 3to 4.7 Although crystalline forsterite grains could be easily produced inour smoke experiments,8,9 the enstatite grains had never been observed.However, we have recently succeeded in forming crystalline enstatite grainswith a spherical or needle shape by simultaneously evaporating SiO and Mgvapor in a plasma field.10 We also have observed the formation of fayalite(Fe2SiO4) grains from an amorphous that produced with a plasma field.11

The crystallization temperature of Fe2SiO4 silicates was lower than thosepreviously reported owing to plasma effects such as the ionization or dopingof atoms. The crystallization temperature of iron silicates being higher thanthat of Mg silicates is considered to be one of the reasons for the lack ofcrystalline iron silicates in astronomical objects.12 An experiment aboutthe formation of crystalline iron silicates suggested certain upper annealingtemperature limits. Thus, the plasma field triggers the formation of crystalgrains 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 stageof alumina nanoparticles. In the case of the alumina nanoparticles producedwithout a plasma, the δ-phase was formed. The stability and morphological

Page 84: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

Formation of Alumina Nanoparticles in Plasma 71

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

2. Experimental Procedure

A gas mixture consisting of Ar (7 Torr) and O2 (3Torr) was introduced intothe chamber, and a radio frequency (RF) plasma (frequency: 13.56MHz,output: 300W) was charged between electrodes. The schematic system usedis shown in Fig. 1. Al powder was evaporated from the Ta boat in theplasma field of Ar–O2 gas mixture. The aluminum clusters immediatelyreacted with oxygen, and then alumina nanoparticles were produced.The observation and spectroscopy of IR spectra were performed for theproduced samples. The collected nanoparticles were dispersed in ethylalcohol and mounted on an amorphous carbon film supported by standardcopper electron microscope grids for the analysis of the structures of thenanoparticles using a transmission electron microscope (TEM) operatedat 100kV (Hitachi H-7100R). They were also buried in KBr pellets, andtheir 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 Taboat in the RF plasma field of Ar–O2 gas mixture (frequency: 13.56MHz, output: 300W).

Page 85: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

72 M. Kurumada and C. Kaito

(Horiba Inc., FT210) from 2.5 to 25.0 µm. The wavelength resolution was2 cm−1. The beam splitter was a Ge-evaporated KBr substrate, and thedetector 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 pro-duced (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) witha plasma field. Corresponding ED patterns indicate that γ-alumina is produced witha plasma field, whereas δ-alumina is produced without a plasma field. Black arrows inimage (b) show the nanoparticles with a characteristic polyhedral shape.

Page 86: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

Formation of Alumina Nanoparticles in Plasma 73

nanoparticles produced with a plasma field is different from that withouta plasma field. Corresponding electron diffraction patterns indicate theproduction of δ-alumina or γ-alumina for the samples produced withoutor with a plasma field, respectively. The unit cells of tetragonal ororthorhombic 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 spinelstructures, the electron diffraction (ED) pattern of δ-alumina resemblesthat of γ-alumina. However, δ-alumina has characteristic diffraction ringsbetween (220)γ and (111)γ owing to their large unit cells. Therefore, weconclude that the nanoparticles produced in the plasma field are γ-alumina.This result obtained by TEM observation is consistent with that obtainedby IR spectroscopy, as shown in the next chapter.

A morphological change was also observed in the nanoparticlesproduced with a plasma field. In the case of without-plasma-producedalumina nanoparticles, 50-nm-order particles are all spherical (Fig. 2a).However, polyhedral or partially polyhedral nanoparticles were clearlyobserved in the alumina produced with a plasma field, as indicated byblack arrows in Fig. 2b. It is considered that these morphological changesare 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 commercialpowder of Newmet Koch, and its purity is 99.99%. The spectrum ofδ-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 andshows 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, theγ-phase was formed. The phase of alumina nanoparticles changed owing tothe plasma.

Both γ- and δ-alumina have been described to have defective spinelstructures. The ideal spinel structure AB2O4 is based on the fcc packingof oxygen anions, with A and B cations occupying the 8a tetrahedral and16d octahedral Wyckoff positions. The distribution patterns of Al cations

Page 87: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

74 M. Kurumada and C. Kaito

Fig. 3. IR spectra of (a) commercial γ-alumina, (b) δ-alumina particles produced inour previous study,4 and (c, d) samples produced in our present experiment correspondto the nanoparticles produced without and with a plasma field, respectively. Arrows Pand C mean absorption features by physisorption and chemisorption of H2O.

are different in both phases. In γ-alumina, Al cations randomly distributein tetrahedral or octahedral sites. Furthermore, the number of Al cationscoordinated in the tetrahedral sites of γ-alumina is larger than that ofδ-alumina.2 The migration of Al cations from tetrahedral to octahedralsites occurs during transformation from γ- to δ-, θ- and finally α-alumina.Therefore, the production of γ-alumina indicates that many Al cations arecoordinated in the tetrahedral sites of the oxygen sublattice within theplasma field.

The ionic radius of Al cations coordinated in the tetrahedral (rt) oroctahedral (ro) sites is 0.39 or 0.54 A, respectively. Alvarez et al. calculatedthe 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, Alcations tend to occupy tetrahedral sites, whereas when rAl > rt is used forsimulation, Al cations tend to occupy octahedral sites. Thus, it is consideredthat 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 sharpand strong absorption peak at 7.2 µm, which had never been previously

Page 88: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

Formation of Alumina Nanoparticles in Plasma 75

observed in the spectra of any alumina nanoparticles, appeared in thespectrum of the alumina nanoparticles produced with a plasma field. Itis 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 airbefore measuring the IR spectrum, many H2O molecules in air adsorbed totheir surfaces.

There are three reactions involved in water absorption, namely,physisorption, chemisorption, and surface hydroxylation. In the case ofphysisorption, the IR absorption peaks of H2O molecules are observed intwo regions, namely, 2.5–3.0 and 5.5–7.0 µm, as observed in all the spectrain Fig. 3. As shown in Fig. 3d, we found that the alumina nanoparticlesproduced with a plasma exhibit a high level of physisorption comparedwith the other spectra in Fig. 3. The H2O molecules could be removedfrom the surfaces of the nanoparticles at 100–200C because they wereattached to the surfaces by van der Waals’ force. On the other hand, thechemisorption or surface hydroxylation has stronger bonds with the surfacesthan physisorption. The H2O molecules for chemisorption or surfacehydroxylation could be removed at about 400C or 1000C, respectively.Figure 4 shows the spectrum of the alumina nanoparticles after heatingat 600C (the IR spectrum was measured at room temperature). The7.2 µm peak disappeared after heating at 600C. That is, the adsorptioncorresponding to this 7.2 µm peak is an irreversible reaction, whereasphysisorption is a reversible reaction corresponding to 2.5–3.0 or 5.5–7.0 µmabsorption peaks observed again in the spectrum in Fig. 4. Therefore, it

Fig. 4. Spectrum of the alumina nanoparticles after heating at 600C (IR spectrum wasmeasured at room temperature). The 7.2 µm peak disappeared after heating at 600C.Arrows P mean absorption features by physisorption of H2O.

Page 89: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

76 M. Kurumada and C. Kaito

is considered that the chemisorption of H2O molecules contributes to theformation of the 7.2 µm peak.

Both equilibrium condensation calculation14 and nonequilibrium nu-cleation accompanying grain growth15 indicate that corundum grains arethe first material to condense in the expanding and cooling gas of solarcomposition in an oxygen-rich atmosphere. As the plasma field exists onour universe, alumina phase with 7.2 µm by chemisorption of H2O may beobserved according to the increase in the resolving power of the instrumentsaccompanying on the development of extra-solar planetary science. Therelease energy of the chemisorption water by heating at 600C may alsoaccelerated the growth into spinel, hibonite, and other compounds.

4. Conclusion

Aluminum was evaporated in a plasma field of Ar–O2 gas mixture, andthe γ-alumina nanoparticles were formed. The IR spectrum of the aluminananoparticles produced with a plasma field indicated the production ofγ-alumina and a sharp and strong absorption peak at 7.2 µm, whichhad been never observed in the spectrum of the nanoparticles producedwithout a plasma. The 7.2 µm peak is considered to be produced in anO–H bending mode from chemisorbed H2O molecules on the surfacesof the nanoparticles. A morphological change was also observed for thenanoparticles produced with a plasma field. Although the without-plasma-produced alumina nanoparticles were all spherical, nanoparticles with apolyhedral shape based on an octahedral shape were formed due to theplasma.

Acknowledgment

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

References

1. I. Levin and D. Brandon, J. Am. Ceram. Soc. 81 (1998) 1995.2. L. J. Alvarez, L. E. Leon and H. Munoz, Catal. Lett. 26 (1994) 259.3. T. Onaka, T. de Jong and F. J. Willems, Astron. Astrophys. 218 (1989) 169.4. M. Kurumada, C. Koike and C. Kaito, Mon. Not. R. Astron. Soc. 359 (2005)

643.

Page 90: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

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 (CambridgeUniversity Press, Cambridge, 2000).

14. L. Grossman, Geochim. Cosmochim. Acta. 36 (1972) 597.15. T. Yamamoto and H. Hasegawa, Prog. Theor. Phys. 58 (1977) 816.

Page 91: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch07

This page intentionally left blankThis page intentionally left blank

Page 92: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

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

INFRARED STUDY OF UV/EUV IRRADIATIONOF NAPHTHALENE IN H2O+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 UniversityChung-Li 32054, Taiwan, R.O.C.

‡Institute of Astronomy, National Central UniversityChung-Li, 32049, Taiwan, R.O.C.

§National Synchrotron Radiation Research CenterHsinchu 30076, Taiwan, R.O.C.

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

We have carried out photon irradiation study of naphthalene (C10H8), thesmallest polycyclic aromatic hydrocarbon (PAH) in water and ammoniaice mixtures. Photons provided by a synchrotron radiation light source intwo broad-band energy ranges in the ultraviolet/near extreme ultraviolet(4–20 eV) and the extreme ultraviolet (13–45 eV) ranges were used for theirradiation of H2O+NH3+C10H8 = 1:1:1 ice mixtures at 15K. We couldidentify several photo-products, namely CH4, C2H6, C3H8, CO, CO2, HNCO,OCN−, and probably quinoline (C9H7N) and phenanthridine (C13H9N). Wefound that the light hydrocarbons are preferably produced for the ice mixturesubjected to 4–20 eV photons. However, the production yields of CO, CO2,and OCN− species seem to be higher for the mixture subjected to EUVphotons (13–45 eV). Therefore, naphthalene and its photo-products appear tobe more efficiently destroyed when high energy photons (E > 20 eV) are used.This has important consequences on the photochemical evolution of PAHs inastrophysical 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 bandsare 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

Page 93: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

80 Y.-J. Chen et al.

the carriers responsible for these DIBs has become one of the mostactive challenges in astrophysical spectroscopy.5,6 Polycyclic aromatichydrocarbons (PAHs) are now thought to be the best candidates toaccount for the DIBs in the interstellar medium (ISM), or most likelytheir cations.7,8 It has been suggested that PAHs could represent ∼17%of the cosmic carbon, and consequently be the most abundant free organicmolecules in the Universe [7 and references therein].

PAHs constitute a group of very stable organic molecules made up fromcarbon and hydrogen only, assembled into aromatic cycles of 6 sp2-carbonatoms like benzene (C6H6) bound together. Such cycles are flat, so thatPAHs are in most cases also flat, like graphite layers. It is believed thatthese molecules are formed in the outflows of dying carbon-rich stars fromwhich they are ejected into the ISM7.

The presence of PAHs in the ISM has been confirmed by theastronomical observations of their C–H stretching and out-of-planebending modes, both in emission9−11 and in absorption12−15 in embeddedprotostars. PAHs may condense onto refractory dust grains with othervolatile species (ices), among which H2O is the most abundant,and are abundant and ubiquitous in many different astrophysicalenvironments.7,16,17 Bernstein et al.18,19 and Sandford et al.20 have beeninvestigating the ultraviolet (UV) processing of PAHs in H2O ices,emphasizing the possible connections between interstellar and meteoriticPAHs, and have shown that PAHs undergo both oxidation and reductionphoto-reactions in ices, resulting in the production of aromatic hydrocarbonspecies similar to some of those identified in carbonaceous chondrites andinterplanetary dust particles (IDPs).

In this paper, we report results obtained from an experimental studyof photon irradiations of naphthalene (C10H8), the smallest PAH (only twoaromatic cycles), in H2O+NH3 ice mixtures at low temperature. Photonsin the 4–20 and 13–45 eV ranges, i.e., from the ultraviolet to the extremeultraviolet (EUV) ranges, were used to irradiate H2O+NH3+C10H8 = 1:1:1ice mixtures at 15 K in two separate experiments. EUV photons can excitemolecules beyond their ionization continua and produce various neutraland ionic fragments. It has been shown that certain molecular species canbe synthesized after irradiation with EUV photons, but not necessarilyafter vacuum ultraviolet (VUV) irradiation21 and vice versa. Finally, theuse of two different energy ranges allows us to study the dependenceof the production yields of several photo-products with the photonenergy.

Page 94: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

UV/EUV Irradiation Study of Naphthalene in H2O+NH3 Ices 81

2. Experimental Protocol

The experimental apparatus and techniques have previously been describedelsewhere.21 In the present work, the UV/EUV irradiations of the icysamples containing naphthalene were performed in a stainless steel vacuumchamber (P < 5× 10−10 torr). The reagents used in this work and theirpurities are as follows: H2O (liquid, triply distilled), NH3 (gas, Sigma-Aldrich, 99.5% purity), and C10H8 (powder, Sigma-Aldrich, 99% purity).The H2O+C10H8+NH3 samples were vapor-deposited onto a KBr substrateat 15K, by injecting a gas mixture of H2O+NH3 and the C10H8vaporsimultaneously through two separated stainless steel thin tubes (2-mminner diameter). The relative proportions of the final H2O+NH3+C10H8 =1:1:1 mixtures were controlled from the partial pressures inside thestainless steel bottles where they were mixed. The typical thicknessof the ice films was 1–3 µm, measured by monitoring the variationof interference fringes of a He-Ne laser light reflected by the KBrsubstrate.

A Fourier-transform infrared spectrometer (FTIR) (Perkin-ElmerFTIR-1600) was used to record infrared (IR) spectra between 4000 and500 cm−1 with a 4 cm−1 resolution. The IR and EUV beams form anangle of 90 on the substrate, so that IR spectra could be recordedin situ during the whole experiment, i.e., before, during and afterirradiation. Once the gas deposition was completed, the vacuum chamberwas left idling for a few hours so that all gases could condense on theKBr substrate. The irradiation experiments were performed after thepressure in the chamber reaches ∼5× 10−10 torr. The IR spectrum ofone of the H2O+NH3+C10H8 =1:1:1 mixtures before irradiation is shownin Fig. 1.

The broad-band UV/EUV beams were provided by the high-fluxbeamline of the National Synchrotron Radiation Research Center (NSRRC)in Hsinchu, Taiwan. The incident photon energy used was the 0th orderof the white light in the 4–20 and 13–45 eV ranges using 450 and1600 lines mm−1 gratings, respectively.22 The incident photon flux wasconstantly monitored by an in-line gold mesh. The irradiations wereperformed until a total integrated incident photon dose of about 1.5× 1020

photons was reached for each irradiation experiment. We could thereforeidentify the compounds photo- produced after irradiation with 4–20and 13–45 eV photons, and compare the results for both photon energyranges.

Page 95: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

82 Y.-J. Chen et al.

4000 3500 3000 2500 2000 1500 1000 500

0.00

0.05

0.10

0.15

0.20

Wavenumber (cm-1)

H2O+NH3+C10H8 = 1:1:1 at 15 K, before irradiationA

bso

rban

ce

Fig. 1. Infrared absorbance spectrum of one of the H2O+NH3+C10H8 = 1:1:1 icemixtures at 15K 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 theH2O+NH3+C10H8 = 1:1:1 ice mixtures at 15 K irradiated with UV/EUVphotons. The measured spectral positions of the absorption features and theidentified species are summarized in Table 1. As can be seen from Fig. 2most of the major features in both spectra look alike although their peakabsorbances are different. The signatures of some photo-products howeveronly appear on the spectrum obtained after 4–20 eV irradiation (uppertrace), but not on the spectrum obtained after 13–45eV irradiation (lowertrace). These compounds are the aliphatic hydrocarbons ethane (C2H6),proprane (C3H8), and the aromatic benzyl radical (C6H5CH2), the laterbeing only tentatively identified. It is likely that these compounds are easilydestroyed by high energy (E > 20 eV) photons or that they readily reactwith other chemical species. Two other compounds (CH2N and C2O), whosefeatures are marked with an asterisk, were only tentatively identified. Inthe following, we will discuss the possible photo-induced chemical reactionmechanisms of these compounds, and their link with the irradiation photonenergy.

The photo-products formed after irradiation with 4–20 eV photonsinclude CH3OH (1030 cm−1), CH4 (1305 cm−1), C2H6 (2873 and 2931cm−1), C3H8 (2959 cm−1), CO (2134 cm−1), CO2 (2338 cm−1), HNCO(2255 cm−1), OCN− (2159 cm−1) and NH+

4 (1453 cm−1). Three other

Page 96: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

UV/EUV Irradiation Study of Naphthalene in H2O+NH3 Ices 83

4000 3500 3000 2500 2000 1500 1000 500-0.02

0.00

0.02

0.04

0.06

0.02

0.04

0.06

Dif

fere

nce

ofa

bso

rban

ces:

4-2

0 eV

CH3OH

C6H

5CH

2?C

2O?

CH2N?

*

* *

C2H

6

C2H

6C

3H

8

CH4

NH4

+

COOCN-

HNCO

CO2

Wavenumber (cm-1)

Dif

fere

nce

of

abso

rban

ces:

13-

45 e

V

Fig. 2. IR spectra (difference of absorbances) of the H2O+C10H8+NH3 = 1:1:1 icemixtures 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 compoundslisted in Table 1.

Table 1. Photo-products identified in the IR spectra of theH2O+C10H8+NH3 = 1:1:1 ice mixtures at 15 K after UV/EUVirradiation. The species marked with * were tentatively identified.The abbreviation n.d. stands for “not detected”.

Species Peak position (cm−1)

after 4–20 eV irradiation after 13–45 eV irradiation

C3H8 2959 n.d.C3H8 2931 n.d.C2H6 2873 n.d.CH2N* 2722 n.d.CO2 2338 2338HNCO 2255 2253OCN− 2159 2160CO 2134 2135C2O* 1898 1902

NH+4 1453 1450

CH4 1300 1299

CH3OH 1030 1030C6H5CH2* 863 n.d.

absorption features, marked with an asterisk in Fig. 2, were only tentativelyidentified. They could be assigned to CH2N (methylene amidogen)(2722 cm−1), C2O (1898 cm−1) and the benzyl radical C6H5CH2 (863cm−1).23,24 They are also listed in Table 1.

Page 97: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

84 Y.-J. Chen et al.

In the H2O+NH3+C10H8 ice samples, naphthalene is the only moleculecontaining carbon atoms. Thus, all photo-products containing carbon canonly be formed from the UV/EUV photolysis of C10H8. This means thatthe aromatic cycles have to be dissociated into smaller fragments to producealiphatic hydrocarbon chains (containing only C and H atoms). It is knownthat breaking aromatic cycles of PAHs is not an efficient process when usingphotons in the 5–10 eV range. Indeed, using a microwave discharge H2 flowlamp as the light source, Bernstein et al.18 showed that only substitutionreactions of H atoms and/or addition of heteroatoms are efficient processin the 5–10 eV energy range.

As listed in Table 1, CH4 has been produced after the photon irradiationof the H2O+NH3+C10H8 ice mixtures with both energy ranges. However,C2H6 and C3H8 were only identified in the IR spectrum of the mixturesubjected to 4–20 eV photons. Separate experiments, where pure C10H8

and a H2O+C10H8 mixture were irradiated with 4–20 eV photons atlow temperature, were carried out in our laboratory. In the obtained IRspectra, we identified the features of CH4, C2H6 and C3H8, indicating thatalkanes CnH2n+2 (n ≥ 1) can be directly produced from the photolysisof naphthalene and/or via secondary chemical reactions involving radicalssuch as CH3 and C2H5. For instance, C2H6 could be a secondary productof the photo-dissociation of CH4:

CH4 + hν → CH3• + H•CH3• + CH3• → C2H6

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 H2O) to formmethanol (CH3OH). Pure methanol ice displays two strong absorptionfeatures at 1030 and 1128 cm−1.25 Unfortunately, in our IR spectra thefeature around 1128 cm−1 overlaps with one of the depletion features ofnaphthalene (see Fig. 2), and thus it can not be observed. However, adiscernible feature around 1030 cm−1 can be assigned to methanol, andindirectly support the photochemical pathway leading to the production ofCH3 radicals and therefore aliphatic molecules such as C2H6 and C3H8.

Furthermore, the feature at 863 cm−1, probably due to the benzylradical (C6H5CH2)24, can clearly be identified in the IR spectrum of the4–20 eV irradiated sample (upper trace of Fig. 2) but not in the spectrumof the 13–45 eV irradiated sample (lower trace of Fig. 2). This result

Page 98: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

UV/EUV Irradiation Study of Naphthalene in H2O+NH3 Ices 85

indicates that naphthalene is progressively photo-dissociated into fragmentsbecoming smaller and smaller as the photon energy increases from the UVto the EUV ranges. In other words, high energy (E > 20 eV) photonsappear to dissociate naphthalene with a higher efficiency than less energeticUV photons. Therefore, we can conclude that although both 4–20 and13–45eV photons can dissociate naphthalene, only 13–45 eV photons cantotally break the aromatic structure of the naphthalene molecule.

In addition to the production of small aliphatic hydrocarbons, wehave 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) andHNCO (2255 and 2253 cm−1). These molecules are probably formed byrecombination of small aliphatic compounds (or their radicals) with HOand NH2, produced from the photo-dissociation of water and ammonia,respectively. The possible mechanisms of formation of such compounds willbe 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. 3and Table 2), called polycyclic aromatic nitrogen heterocycles (PANHs).

1590 1560 15300.026

0.028

0.030

0.032

0.034

1500 1470 1440

0.030

0.032

0.034

0.036

1380 1350 1320

0.028

0.030

0.032

0.034

780 760 7400.010

0.015

0.020

0.025

0.030

Wavenumber ( cm-1 )

Rel

ativ

e D

iffe

ren

ce o

f A

bso

rban

ces

Fig. 3. IR spectra (difference of absorbances) of the H2O+C10H8+NH3 = 1:1:1 icemixtures at 15K 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 weakabsorption features of quinoline and phenanthridine (see Table 2).

Page 99: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

86 Y.-J. Chen et al.

Table 2. Weak absorption features have tentatively been assigned toquinoline (C9H7N) and phenanthridine (C13H9N) in the 1600–1300 and820–720 cm−1 ranges26 observed for the H2O+C10H8+NH3 = 1:1:1 icemixtures at 15 K after UV/EUV irradiation (see Fig. 3).

Species Peak position (cm−1)

after 4–20 eV irradiation after 13–45 eV irradiation

C9H7N 1439 14401375 n.d.1322 1321767 767

C13H9N n.d. 1584n.d. 15291494 1493n.d. 14621342 1341756 752

The positions of these features appear to be in good agreement with theabsorption features of quinoline (C9H7N) and phenanthridine (C13H9N) inwater ice at 15K.26 Quinoline is a molecule of naphthalene where one of thecarbon atoms has been substituted by a nitrogen atom. We plan to carryout further investigations using a higher resolution FTIR spectrometer(0.1 cm−1) and a brighter photon source in order to better resolve theobserved features.

These important results indicate that naphthalene and possibly otherPAHs can be photo-dissociated in cold astrophysical environments by EUVphotons and contribute to the reservoir of carbon whose photochemicalevolution can lead to the production of complex organic molecules inthe ISM.

3.2. Production yields of CO, CO2 and OCN−

We have determined the production yields of CO, CO2 and OCN−, photo-produced during the irradiation of the H2O+NH3+C10H8 = 1:1:1 icemixtures at 15K with 4–20 and 13–45eV photons, using data analysisprocedures which have been previously described elsewhere.21 The columndensities of CO, CO2 and OCN− are plotted as a function of thephoton dose in Figs. 4, 5 and 6, respectively. Absorption strengths ofA(CO, 2134cm−1) = 1.1 × 10−17 cm molec−1, A(CO2, 2340cm−1) =

Page 100: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

UV/EUV Irradiation Study of Naphthalene in H2O+NH3 Ices 87

0 5 10 15 20 25 300

5

10

15

20

25

30

35

4 - 20 eV 13 - 45 eV

CO

co

lum

n d

ensi

ty (

x 10

15 m

ole

c cm

-2)

Photon dose (x 1019 photons cm-2)

Fig. 4. Plot of the column density of CO as a function of the 4–20 eV (squares) and13–45 eV (triangles) photon doses.

0 5 10 15 20 25 300

1

2

3

4

5

6

CO

2 co

lum

n d

ensi

ty(x

1015

mo

lec

cm-2)

4 - 20 eV 13 - 45 eV

Photon dose (x 1019 photons cm-2)

Fig. 5. Plot of the column density of CO2 as a function of the 4–20 eV (squares) and13–45 eV (triangles) photon doses.

7.6× 10−17 cm molec−1 and A(OCN−, 2160cm−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

Page 101: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

88 Y.-J. Chen et al.

0 5 10 15 20 25 300

1

2

3

4

5

6

4 - 20 eV 13 - 45 eV

OC

N- c

olu

mn

den

sity

(x

1015

mo

lec

cm-2)

Photon dose (x 1019 photons cm-2)

Fig. 6. Plot of the column density of OCN− as a function of the 4–20 eV (squares) and13–45 eV (triangles) photon doses.

3.6, 2.2 and 1.9 times higher than the corresponding yields for the 4–20eV experiment (squares) after an integrated photon dose of approximately1×1020 photons cm−2. This result suggests that small C-bearing moleculesare produced from the direct photo-dissociation of naphthalene, which isthe only carbon- bearing compound in our starting mixtures, and confirmthat the photo- dissociation efficiency for such small molecules increaseswith the photon energy.

Figures 4 and 5 also show that CO is produced with a higher efficiencythan CO2. The column densities for CO are about 3.5 and 5.8 times higherthan those for CO2 in the 4–20 and 13–45 eV experiments, respectively. Thiscan 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 reactionwith CO29:

CO + hν → CO∗(A1Π, a3Π)

CO∗ + CO → CO2 + C•,where A1Π and a3Π denote excited states of CO. The carbon atom releasedin this mechanism does not remain free in the medium, and may react withanother CO molecule to produce C2O. This mechanism is supported by the

Page 102: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

UV/EUV Irradiation Study of Naphthalene in H2O+NH3 Ices 89

presence of two IR features of C2O at 1898 and 1902 cm−1 in the irradiatedsamples23 (see Fig. 2 and Table 1).

CO2 can also be formed from the reaction between CO and the HOradical, which is produced from photo-dissociation of H2O30:

H2O + hν → HO• + H•HO• + CO → •COOH → CO2 + H• .

The photolysis of CO to form CO2 and COOH radicals constitute thefirst steps for the formation of complex organic molecules. In particular,reactions between the parent molecules and their photo-products includingradicals such as H, HO, CnH2n+1 (n ≥ 1) (alkyl radicals), COOH, NH,NH2 at low temperature could lead to the production of molecules ascomplex as amino acids in such experiments, but also in astrophysicalenvironments.

4. Conclusion

We have investigated the effects of the irradiation of H2O+NH3+C10H8 =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 thecharacteristic features of several photo-products, namely CH4, C2H6, C3H8,CO, CO2, HNCO and OCN−. Methyl amidogen (CH2N), C2O and thebenzyl radical (C6H5CH2) have also been tentatively identified. Our workalso shows that on the one hand small hydrocarbons such as C2H6, C3H8

and the benzyl radical are significantly produced during the 4–20 eVirradiation experiment, and that on the other hand the production yieldsof CO, CO2 and OCN− are significantly higher in the 13–45 eV irradiationexperiment. Therefore, the photo-products and their production yieldsstrongly depend on the photon energy.

The present work shows that EUV photons can efficiently photo-dissociate naphthalene and probably other PAHs at low temperature. Thishas important implications on the photochemical evolution of PAHs inastrophysical environments, where the carbon reservoir could contributesignificantly to the production of complex organic molecules includingamino acids, the building blocks of proteins in all living beings onEarth.

Page 103: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

90 Y.-J. Chen et al.

Acknowledgments

We are grateful for the support of the staff of the National SynchrotronRadiation Research Center in Hsinchu, Taiwan. This research issupported in part by the Ministry of Education under the Aimfor the Top University Program (NCU) and based on the worksupported by the NSC grant #NSC-95-2112-M008-028, the NSF PlanetaryAstronomy Program under Grant AST-0604455 (C.-Y. R. W.), andthe 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. Leger, L. d’Hendecourt and D. Defourneau, Astron. Astrophys. 293 (1995)

L53.6. F. M. Johnson, Bulletin of the Am. Astron. Soc. 33 (2000) 716.7. J. L. Puget and A. Leger, 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.

Page 104: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

UV/EUV Irradiation Study of Naphthalene in H2O+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. Chenand 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.

Page 105: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch08

This page intentionally left blankThis page intentionally left blank

Page 106: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

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

NEW METHOD OF PRODUCING TITANIUMCARBIDE, MONOXIDE, AND DIOXIDE GRAINS

IN LABORATORY

AKIHITO KUMAMOTO∗, MAMI KURUMADA, YUKI KIMURAand CHIHIRO KAITO

Department of Physics, Ritsumeikan University, Noji-higashi 1-1-1Kusatsu, Shiga 525-8577, Japan

[email protected]

By making a carbon rod covered with Ti on the surface without exposure toair, TiC grains less than 10 nm in diameter were predominantly produced. Theintroduction of a small amount of oxygen in Ar gas (partial pressure 1/1000),allowed the continuous formation of TiO2 and TiO–TiC. The infrared spectraof TiO2, TiO, and TiC were measured. An absorption feature attributed toTiO 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 oxygen-rich environments.1,2 With respect to dust formation in circumstellar shells,it is important to note that the TiO molecule is rather prominent in theatmosphere of O-rich stars. Solid titanium oxides such as TiO, TiO2, Ti2O3,Ti3O5, and Ti4O7 are considered to play a role in the formation of solidTi compounds in the gas phase.3 TiO2 exists in three different phases,namely, rutile, anatase, and brookite. The infrared (IR) spectra of theseTiO2 phases were amply measured by Posch et al.4 Another phase displayedby TinO2n−1 compounds is the Magneli phase, which is derived from therutile structure by introducing oxygen planar defects as elucidated in aprevious paper.5 In the previous paper, NaCl-type structure of the ordinaryTiO phase did not appear by the reduction of the TiO2 phase.5 TiO of asuperstructure, which is an ordered structure observed in order–disordertransition alloys, was found on the sample prepared from mixture of iodidetitanium and TiO2 within a tungsten-arc furnace in a gas atmosphere ofpurified argon.6

93

Page 107: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

94 A. Kumamoto et al.

On the other hand, the 20.1 µm feature observed in the spectra of somepost-AGB stars was proposed to be due to titanium carbide.7 Subsequentstudies showed that the attribution of this feature to titanium carbide isarguable.8–10 A number of TiC grains have been found to be of presolarorigin in graphite spherules extracted from the Murchison meteorite. TiCcrystals can be formed in AGB star atmospheres only in high-densityregions 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 carbonfilms,12 and TiC grains (50 nm in size) of cubic shape produced by the gasevaporation method13 showed no peaks at 20.1 µm. Recently, Kimura et al.have produced TiC grains covered by a carbon mantle from CO gas bythe Boudouard reaction and have showed the existence of large fullerenescontributing to the 21 µm feature.14

In this paper, a new experimental method of producing TiC, TiO2, andTiO grains in the laboratory is described. It was found that the TiO phasecan be formed with a small amount of oxygen even for TiC grains producedin carbon-rich conditions. It was suggested from theoretical calculation thatthe condensation temperature of TiC grains is lower than that of titaniumoxide.3 Finally, we present the conditions of formation of TiC grainsin an oxygen-poor environment, corresponding to an environment whereC/O ≈ 1, as for S-stars, in a laboratory experiment.

2. Experimental Methods

The evaporation chamber was a glass cylinder, 17 cm in diameter and30 cm high, covered with a stainless-steel plate on top and connected toa high-vacuum exhaust through a valve at the bottom. A Ti wire (0.25mmdiameter) was placed on the concave position between a pair of carbonrods (0.5mm diameter) as schematically showed in Fig. 1. After heating the

Fig. 1. Schematic presentation on proposed experimental method. Two carbon rodswith molten Ti metal were used as the evaporation source to produce smoke. Smokegrains were produced in Ar gas at 10 kPa or a gas mixture of Ar and O2 at 10K and10 Pa, respectively.

Page 108: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

New Method of Producing Titanium Oxides 95

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

3. Results and Discussion

TiC grains covered with a thin carbon layer were produced by heatingcarbon rods wrapped by the Ti wire at 2300K in an Ar atmosphereof 10 kPa, as shown in Fig. 2. The produced grains which are seen inFig. 2a are TiC grains. They have NaCl-type structure, determined fromthe electron diffraction (ED) pattern (Fig. 2b). Small TiC grains less than10 nm in diameter were predominantly produced and their external shape isa truncated octahedron as indicated by arrows and a typical HRTEM imageshown in the inset of Fig. 2c. In spite of the different shape comparedwith the cubic form grains observed in a previous paper,13 the same IRabsorption peak was seen at 12.5 µm. Therefore, it can be concluded thatthe differences in size and shape did not influence the spectral feature ofTiC in this size range.

Since Ti–C phase diagram shows that TiC and C form a eutectic for arelative proportion of C of 30wt.% (63 at.%), TiC and C evaporate at thesame temperature as the eutectic composition. Therefore, carbon atoms inexcess are produced and can surround the TiC grain surface.

When mixing a small amount of oxygen (10 Pa) with the Ar gas at10 kPa, TiO2 grains were initially produced. The color of the evaporatinggrains varied from white to black during the experiment. White is thetypical color of TiO2 grains. According to TEM analysis, the black grainswere 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

Page 109: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

96 A. Kumamoto et al.

TiC

220311222400331420422

200111

(a)

(c)

(111)

(200

)

(b)

TiC

220311222400331420422

200111

220311222400331420422

200111

(a)

(c)

(111)

(200

)

(b)

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

with a thin carbon layer.

100 nm

(a) (b)

100 nm

(a) (b)

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

Page 110: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

New Method of Producing Titanium Oxides 97

TiO phase with decreasing oxygen partial pressure starting from 1/1000 ofAr pressure.

Figure 4 also shows an HRTEM image of TiO and mixed TiO/TiCgrains. The (111) plane of the upper right TiO grain, with a size of 10nm,is nearly parallel to the (111) plane of the large central TiC grain. Thecontrast in the stripes in the TiC (111) lattice image may be due to theexistence of a TiO layer. A few layers of TiO are present at the positionsindicated by arrows. Since TiC and TiO have the same NaCl-type structureand very close lattice constants, the mismatch between the two phases issmall (3.6%). Therefore, these mixed grains may grow well in an oxygen-poor atmosphere. The surfaces of these grains are covered with a thin

70.5˚

10 nm

(111)TiC

0.25 nm

(111)TiC

0.25 nm

55˚

(200)TiO

0.21 nm

(111)TiO

0.24 nm

(110)TiO2- Rutile 0.32 nm

(110)TiO2- Rutile 0.32 nm

Fig. 4. HRTEM image of produced grain at oxygen partial pressure of 10Pa. Grainsconsisting 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.

Page 111: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

98 A. Kumamoto et al.

carbon layer and contain crystallites with the rutile (TiO2) structure. Theformation of TiO2 crystallites at/on the surface carbon layer can be deducedfrom the lattice spacing and the angle of crossed lattice images. When theoxygen relative partial pressure was of the order of 1/1000 at the beginningof the experiment, the TiC–C mixture vapor, which were evaporated withthe eutectic reaction, was oxidized and TiO2 grains were produced. SinceTiO2 grains consume oxygen gas, the oxygen partial pressure was graduallydecreasing during the experiment. Then, TiO and TiC phases were formed.

Figure 5 shows the IR spectra of (a) TiO2 and (b) the TiC–TiOmixture. TiO and TiC phases have structures based on corner-sharingoctahedrons, and TiO2 also consists of distorted octahedrons having theedge-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 featurein spectrum (b) may be due to TiO, which have distorted octahedralstructures 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 bytitanium oxide easily appears with the introduction of a small amount ofoxygen.

0

10

20

30

40

50

60

70

80

90

100

5 7 9 11 13 15 17 19 21 23 25

Wavelength [µm]

Tra

nsm

itta

nce

[%]

(a)

(b)

12.5

14.7

14.316.7

Fig. 5. IR absorption spectra of laboratory grains embedded in KBr pellets in the5–25 µm range. The spectra correspond to the grains produced during the evolutionfrom (a) white TiO2 grains to (b) a black TiO–TiC grain mixture, following the graduallydecreasing partial pressure of oxygen.

Page 112: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

New Method of Producing Titanium Oxides 99

4. Summary

A new evaporation method for TiC–C mixture compound was achieved forthe production of solid TiO on formed TiC dust. TiC grains less than 10 nmin diameter had predominantly truncated octahedral structures coveredwith a carbon layer. In the case of the formation of TiC grains in Aratmosphere with a small amount of oxygen, the composition of the producedgrains varied from TiO2 to a TiO–TiC mixture as oxygen partial pressuredecreased. The characteristic IR features at 12.5 and 14.7 µm assigned toTiC and TiO, respectively, characterized by the change of color of theevaporating grains from white to black during the experiment.

References

1. S. S. Barshay and J. S. Lewis, Ann. Rev. Astron. Astrophysics 14 (1976) 81.2. E. E. Salpeter, Ann. Rev. Astron. Astrophysics 15 (1977) 267.3. H. P. Gail and E. Sedlmayr, Faraday Discuss. 109 (1998) 303.4. Th. Posch, F. Kerschbaum, D. Fabian, H. Mutschke, J. Dorschner, A. Tamani

and Th. Henning, Astrophys. J. Suppl. Ser. 149 (2003) 437.5. C. Kaito, M. Iwanishi, T. Harada, T. Miyano and M. Shiojiri, Trans. Jpn.

Inst. Meteoroe. 24 (1983) 450.6. D. Watanabe, J. R. Castles, A. Jostsons and A. S. Malin, Acta Cryst. 23

(1967) 307.7. 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) 313.8. S. Hony, A. G. G. M. Tielens, L. B. F. M. Waters and A. de Koter, Astron.

Astrophys. 402 (2003) 211.9. A. Li, Astrophys. J. 599 (2003) L45.

10. T. Chigai, T. Yamamoto, C. Kaito and Y. Kimura, Astrophys. J. 587 (2003)771.

11. T. Chigai, T. Yamamoto and T. Kozasa, Astrophys. J. 510 (1999) 999.12. Y. Kimura, A. Ikegami, M. Kurumada, K. Kamitsuji and C. Kaito, Astrophys.

J. Suppl. Ser. 152 (2004) 297.13. Y. Kimura and C. Kaito, Mon. Not. R. Astron. Soc. 343 (2003) 385.14. Y. Kimura, J. A. Nuth III and F. T. Ferguson, Astrophys. J. 632 (2005)

L159.15. M. Kurumada, O. Kido, T. Sato, H. Susuki, Y. Kimura, K. Kamitsuji, Y. Saito

and C. Kaito, J. Cryst. Growth 275 (2005) e1673.16. M. Kurumada and C. Kaito, J. Phys. Soc. Jpn. 75 (2006) 074712.

Page 113: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch09

This page intentionally left blankThis page intentionally left blank

Page 114: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

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

DESTRUCTION YIELDS OF NH3 PRODUCED BY EUVPHOTOLYSIS OF VARIOUS MIXED COSMIC

ICE ANALOGS

C. Y. R. WU∗, T. NGUYEN and D. L. JUDGE

Space Sciences Center and Department of Physics and AstronomyUniversity 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 Center101 Hsin-An Road, Hsinchu Science Park

Hsinchu 30076, Taiwan

Experimental measurements of the destruction yields of NH3 have been carriedout by extreme ultraviolet–vacuum ultraviolet (EUV–VUV) photolysis ofcosmic ices containing NH3. The ice systems studied in the present workinclude pure NH3 ices and icy mixtures of NH3 with CO, H2O, and CH4

at a temperature of 10K. 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 thisstudy, the photon wavelengths used to irradiate the icy samples were mainlyselected to center on the prominent solar lines, namely, the 30.4, 58.4 and121.6 nm. The photodestruction yields of NH3 in the presently studied icemixtures are typically higher than 0.5 and can be higher than unity, a veryefficient ice photochemical process.

1. Introduction

The surfaces of astronomic objects are subjected to continuous irradiationby solar and interstellar photons, stellar winds, and cosmic rays.1

Modification of chemical composition and mass transfer/evolution in thefrozen surfaces are thus expected to occur. The study of photolysiseffects in ices is important in order to improve our understanding ofthe 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 andcharged particle impact on realistic cosmic ice analogs. We have recently

101

Page 115: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

102 C. Y. R. Wu et al.

implemented a laboratory program to study ice photochemistry and toprovide extreme ultraviolet–vacuum ultraviolet (EUV–VUV) photolysisdata for interpreting the above-mentioned observations of cosmic objects.

In this preliminary work, we report laboratory results on the mea-surements of destruction yields of NH3 in several ice systems. The NH3

ice has been observed on the surfaces of Uranus’s moon Miranda, Pluto’smoon Charon, a Kuiper Belt object Quaoar, and dense molecular clouds.It is also possible that NH3 exists in the form of ammonia hydrates2 andNH+

4 in the ice mantles3 in these objects. An intensive search for evidence ofNH3 ices existing under the surface of Saturn’s moon Enceladus is currentlybeing conducted by the Cassini mission. [For more information about theCassini-Huygens mission the interested reader is recommended to visithttp://saturn.jpl.nasa.gov.]

It is important to point out that photodestruction cross sections ofseveral 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 121.6nm line and theH2 molecular band emissions in the spectral region between 135 and145 nm, sharp bands in 145–155nm, and the very broad emission bands in150–300nm.7 Its brightness varies according to the operating conditions.At very low H2 flow pressures or 2% H2 in He in the discharge tubethe spectral outputs will almost exclusively consist of the atomic HI121.6 nm emission.8,9 However, at high H2 flow pressures the intensityof the HI 121.6nm emission decreases significantly and the H2 molecularband emissions become the dominant outputs.4,7,10 The relative outputintensities10 of the H2 molecular bands strongly vary with the H2 pressure(i.e., H2 molecular density). The photolysis data generated by using themicrowave-discharged H2 flow lamp indeed cover broad bands of the photonwavelength range of interest. Therefore, photolysis study using such a lampis qualitatively valuable because it can only provide data averaged over abroad spectral region.

It may be also important to add that the UV spectrum of theinterstellar medium is best simulated by the microwave-discharged H2 flowlamp. The output spectrum of the lamp resembles, but is not identicalto, 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 providedby microwave-discharged H2 flow lamp. However, the output spectrumwill of course depend on the H2 pressure condition in the given cosmic

Page 116: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

Destruction Yields of NH3 103

environment of interest. It is commonly accepted that the diffusive UVbackground in interstellar environment is longer than 91.2 nm, the HIionization threshold. The EUV diffusive background emission is known tobe weak under such circumstances. However, in a small region around theSun, hot white dwarfs, subdwarfs, and novae the EUV radiation photonsare enough to seriously affect photochemical and photophysical processeson surrounding matters.12 The photolysis data obtained in the EUV regioncan thus provide crucial parameters for modeling the evolution of prebioticmatters including the possible formation of amino acids in icy grains.

For obvious reasons, we really need to use a monochromatic light sourcein the photolysis study in order to provide meaningful and accurate datafor modeling cosmic ice photochemistry throughout the UV and EUVregions. In our work we have employed a synchrotron radiation facility,which provides selectable monochromatic light with a well-defined spectralbandwidth, as previously described.13–15

In the present work, we choose four basic molecules of great importancein the cosmic environments for the make-up of the mixed ice samples.They are CO, H2O, NH3, and CH4. The relative absorbances of solidCO ices in the 106–170nm spectral region,16 H2O ices in the 106–180nmregion, NH3 ices in the 106–240nm region, and CH4 ices in the 106–160nm spectral region17 have recently been measured at the NationalSynchrotron Radiation Research Center (NSRRC). However, there are nosuch data for the above solid ices in the photon wavelength region shorterthan 106nm, the LiF window cut off. Therefore, our discussion of theEUV photoexcitation processes in ices will mainly rely upon what weknow about the molecules of interest in the gas phase. The absorptionof 121.6 nm photons by gaseous NH3 will result in dissociation intoNH2 + H, NH + H2, and NH + 2H fragments, and gaseous CH4 yields theCH3 +H and CH2 +H2 fragments.18–21 However, the 121.6 nm photons canonly excite CO to its excited valence electronic states.16,22,23 Therefore,the photolyzed products produced through photolysis at different photonwavelengths may be different, with different production and destructionyields.

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), H2O + NH3 + CH4

(1:1:1), and H2O + CO + NH3 (1:1:1) at a temperature of 10K. A tunableintense synchrotron radiation light source available at NSRRC, Hsinchu,Taiwan, was employed to provide the required EUV–VUV photons. The

Page 117: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

104 C. Y. R. Wu et al.

photon wavelengths used to irradiate the icy samples were selected tocenter on the prominent solar lines, namely, the HeII 30.4nm, the HeI58.4 nm, and the HI 121.6 nm lines. We have previously investigated theidentification of the chemical products produced through photolysis ofseveral of the above ice mixtures.24 In this report, we focus on measurementsof the destruction of the NH3 molecule in the various ice mixtures. Theresults obtained from the present study are important to our understandingof 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 pre-sent work have previously been described in detail.13,14 Briefly, the experi-mental setup consists of a synchrotron radiation source and monochromatorsystem, a helium closed-cycle cryostat system (APD HC-DE204S) and asample 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 GratingMonochromator (HF-CGM) beamline25 at the NSRRC. The synchrotronradiation was produced by a 1.5GeV electron storage ring. In the presentwork the entrance and the exit slit widths were set to give a bandwidthof 0.4 nm at 58.4 nm and at 30.4 nm, and a bandwidth of 1.1 nm at121.6nm.10,13–15

The purity of the samples used in the present work is of the highestquality commercially available. Specifically, the ammonia gas was providedby Sigma-Aldrich with a stated purity of 99.5%. All gaseous samples werehandled by the standard freeze–pump–thaw–pump–freeze–distill cycles aspreviously described.13,14,24 The ice samples were deposited onto a KBrsubstrate mounted on the cold finger, which was maintained at 10K by thehelium closed-cycle cryostat system. The thickness of the ice samples usedin the present work is typically from 3 µm to 5 µm. The optical penetrationdepths14,26 of EUV photons used in the present work vary from 10−3 to5 × 10−2µm based on the well known absorption cross sections of relevantmolecules in the gas phase.21 Thus, all incident photons are absorbed bythe ice sample without impinging on the substrate.

The absorption spectra of a given icy sample before and after EUVphoton irradiation were obtained by the FTIR spectrometer using a globarsource, a CsBr beam splitter, and a HgCdTe detector (cooled to thetemperature of liquid nitrogen at 77K). The spectral ranges employed in

Page 118: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

Destruction Yields of NH3 105

the present work cover from 2.5 to 20mm (i.e., from 4,000 to 500 cm−1). Atypical 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 FTIRspectrometer. To enhance the changes of absorbances of the photon-inducedchemical products we chose to present the spectra of the difference ofabsorbances, (ap − a0).13,14 In Fig. 1 we show typical absorbance spectraof pure NH3 ices taken before (a0, the top panel) and after (ap, themiddle panel) the 30.4 nm photolysis in the 5000 to 500 cm−1 region for anirradiation time of 180min. The spectrum of the difference of absorbances,(ap − a0), is plotted in the bottom panel. A peak in the (ap − a0) spectrumoften reflects the growth of photon-induced chemical products. A dipcorrelates with depletion of the parent ice molecules, which can be clearlyseen by comparing the absorption features of NH3 shown in the upper,middle, and bottom panels. In other words, while new molecular specieswere formed, the original reactants were depleted due to their conversionto other species. The photolyzed products are indicated in the bottompanel of Fig. 1, showing the NH2 radical feature at 1505 cm−1 and thepreviously observed, but unidentified feature at 2115 cm−1.27 We cannotpositively observe the 886 cm−1 feature N2H4 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 than1.1 nm13–15) vs. an undispersed output from a microwave-discharged H2

flow lamp.27 There are two broad features indicated by “?” mark becausethey do not correspond to any absorption features of possible products suchas NH, N2H4, (NH3)2, or NH+

4 .28–31 Further study to assign their identitiesis required.

The absorption feature of the NH3 at 1065 cm−1 27,28,32 appears tobe isolated from other features of NH3, as well as the possible products.Therefore, we chose this feature for our measurements of the destructionyields. In Fig. 2 we show the portion of the spectra of the difference ofabsorbances between 3000 and 800 cm−1 after 30.4 nm photolysis for 5, 30,60, 120, and 180min. The magnitudes of the peaks and dips increase withincreasing photon irradiation time, i.e., photon dosage.

Page 119: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

106 C. Y. R. Wu et al.

Fig. 1. The top panel: absorbance spectrum of pure NH3 ices before photolysis. Themiddle panel: absorbance spectrum of pure NH3 ices after photolysis at 30.4 nm for180 min. The bottom panel: the difference of absorbances of the NH3 ices before andafter photolysis at 30.4 nm.

3.2. The destruction yields

To quantitatively determine the dependence of the destruction yield as afunction of photon dose we first measure the product column density, Np

(molecule cm−2), which is obtained by dividing the integrated area of thedip of the difference of absorbances by its absorption band strength, A (cmper molecule). The relationship13,26 is given below:

Np = 2.3[∫(ap − a0)dν]

A

Page 120: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

Destruction Yields of NH3 107

-0.01

0.00

0.01

0.02

NH3Fig1b.org

After 30.4 nm Photolysisfor 180 min.

-0.01

0.00

0.01

0.02

Diff

ere

nce

of

Ab

sorb

an

ces

After 30.4 nm Photolysisfor 120 min.

-0.01

0.00

0.01

0.02

After 30.4 nm Photolysisfor 60 min.

-0.01

0.00

0.01

0.02

After 30.4 nm Photolysisfor 30 min.

3000 2500 2000 1500 1000

-0.01

0.00

0.01

0.02

Pure NH3 ices at 10 K

After 30.4 nm Photol;ysisfor 5 min.

Wavenumber (cm-1)

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

The infrared band strength ofNH3 at 1065 cm−1 is A(NH3, 1065 cm−1)=1.7 × 10−17 cm per molecule.32

The photon dose (photons cm−2), ∫ PUVdt, is obtained by dividing theintegrated numbers of photons impinging on the ices by the photon beamsize, which is 0.145 cm2, at the ice sample. The destruction yield13,14 canbe expressed as

Y =dNp

d[∫

PUVdt].

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

Page 121: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

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 10K as a function of photon dose. The results obtained are shown forphotolysis 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 (photonscm−2) is displayed in Fig. 3 for photolysis of CO+ NH3 (4:1) mixed ices atthe three selected photon wavelengths. The photon dose used in the presentwork has a dynamic range between 10 and 100. As can be seen from Fig. 3and within the presently studied photon dose range, the product columndensity of NH3 appears to deviate from a linear decrease at a photon dosageof about 1 × 1017 photons cm−2 in the photolysis at 30.4 and 58.4nm andat a photon dosage of about 8 × 1015 photons cm−2 in the photolysis at121.6 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. Briefdiscussions 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 separatedfrom other absorption features. Since the destruction of absorbances has

Page 122: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

Destruction Yields of NH3 109

Table 1. Destruction yields of NH3 through photolysis of mixed ice systemsat 10K.

Ice system 30.4 nm 58.4 nm 121.6 nm

NH3 1.13 (±0.48) 0.88 (±0.23) Not measuredCO + NH3 (1:1) 1.06 (±0.14) 1.56 (±0.09) 2.76 (±1.38)CO + NH3 (4:1) 0.53 (±0.07) 0.60 (±0.16) 2.53 (±1.38)CH4 + NH3 (1:1) 0.90 (±0.16) 0.51 (±0.09) Not measuredH2O + CO + NH3 (1:1:1) 1.29 (±0.18) 1.17 (±0.25) 0.69 (±0.12)CH4 + CO + NH3 (1:1:1) 1.57 (±0.18) 0.99 (±0.12) 0.60 (±0.12)H2O + CH4 + NH3 (1:1:1) 0,83 (±0.09) 0.53 (±0.12) Not measured

been accurately measured the data can be fitted to a single exponentialfunction. The half-life for the destruction of NH3 has thus been determinedto be 3.9 × 1017 photons cm−2 and 2.6 × 1017 photons cm−2 for thephotolysis at 30.4 and 58.4 nm, respectively, under the present experimentalconditions. The half-lifetime of pure NH3 ice molecules can be estimated inthe cosmic environments if the photon flux at the given photon wavelengthis known. The EUV photon fluxes for the prebiotic environs and the currentSolar EUV outputs can be very different. As an example, if we use anabsolute solar flux of 4 × 1010 photons cm−2 sec−1 for photon wavelengthfrom 5 to 50.4nm33 of the current days, then the half-lifetimes will be about107 years.

We have examined the CO + NH3 ice system in two different compo-sitions, namely, at a ratio of 1:1 and 4:1. In Fig. 3, we plot the differenceof the column densities of the 1065 cm−1 feature produced through thephotolysis 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 yieldsare slightly larger than the corresponding values for the (4:1) ice sample, assummarized in Table 1. The error bars in the destruction yields representthe spread of values determined from different experimental runs.

It is interesting to note that the destruction yield of NH3 by 121.6nmphotolysis is more than unity. This is possible since reactions of CO with thephotolyzed 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 Usinga microwave-discharged H2 flow lamp with the spectral output peakingapproximately at 162nm, Cottin et al.4 have found that the destruction forCH4, CH3OH, NH3, and HNCO is faster in a N2 ice matrix than in H2O

Page 123: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

110 C. Y. R. Wu et al.

ices. The photodestruction yields of NH3 in the CH4 + NH3 (1:1) mixedices measured at 30.4 and 58.4 nm behave similarly as the above-mentionedbinary systems.

The destruction yields of NH3 in tertiary ice systems, namely, H2O +CO+NH3 (1:1:1), CH4 +CO+NH3 (1:1:1), and H2O+CH4 +NH3 (1:1:1)mixed ices have also been measured. In this study, we attempted to learnthe ice photochemistry by replacing H2O with CH4 in the H2O+CO+NH3

(1:1:1) ices. As can be seen from Table 1, the destruction yields of NH3 inthe ice systems at 30.4 and 58.4nm are larger than those at 121.6 nm. Theseresults appear to correlate with the fact that the absorption cross section ofCO ice at 121.6 nm is relatively small15,16 in comparison with that at 58.4and 30.4 nm. However, the destruction yields produced through photolysisof the CH4 + CO + NH3 (1:1:1) ices at 58.4 and 121.6 nm are smaller thanthose in the H2O + CO + NH3 (1:1:1) ices. This may suggest that H2O ismore reactive than CH4 in the destruction of NH3 in the presently studiedmixed ice systems containing CO and NH3 ices.

We can compare the results between CH4+CO+NH3 (1:1:1) and H2O+CH4 + NH3 (1:1:1) ices and between H2O + CO + NH3 (1:1:1) and H2O +CH4 +NH3 (1:1:1) ices. In the former case H2O was replaced by CO and inthe latter case CH4 was replaced by CO, we find that the destruction yieldsat 30.4 and 58.4 nm are larger in the mixed ice containing CO suggestingthat CO is more reactive in helping destruction of NH3 in these ice systemsthrough photolysis at these two EUV energies. The ice photochemistry isobviously complicated. We plan to carry out further detailed analyses atmore photon energies in the near future in order to elucidate and understandthe important photolysis and reaction mechanisms.

4. Concluding Remarks

The present work shows that the destruction yields of NH3 in severalphotolyzed mixed ices clearly depend on (1) the ice compositions and (2) thephoton excitation energies. We conclude that EUV–VUV photodestructionsof NH3 in the presently studied ice mixtures are very efficient photochemicalprocesses with yields higher than 0.5. We have previously identifiedvarious photolyzed products in these mixed ice systems, for example,the CN-containing species such as OCN−, HCN, and CH2N2, the lighthydrocarbons such as the C2H4, C2H6, and C3H8, as well as the HCO,H2CO, CH3OH, and HCCO radical products.24 The quantitative analysis of

Page 124: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

Destruction Yields of NH3 111

the product yields for the above-mentioned species are currently in progressin our laboratory.

We see that in photolysis there are many important processes, suchas (1) photodestruction of parent molecules, (2) photo-induced products,(3) subsequent chemical reactions among the photolyzed products andparent molecules, (4) subsequent photolysis of the photolyzed products andreaction products, and (5) depletion and mantle effect of the ice sample.In this preliminary work, we have just begun to explore the physics andchemistry of cosmic ices. Further study to unravel the complex processes isclearly needed.

In our previously investigated photo-induced chemical pro-ducts13–15,23,24, the EUV photolysis apparently produces more chemicalproducts and with higher production yields than the 121.6nm counterpartin the molecular ice systems studied. In the present study of destructionsof the parent NH3 ice molecule, we find that the VUV photodestructionprocess can be more efficient than the EUV counterpart. This may be inpart due to a large absorbance of NH3 ice in the 121.6 nm region,17 whichis expected to be larger than those at 30.4 and 58.4 nm based on absorptioncross section data of the gaseous NH3.21 Our ongoing research effort willfurther improve our understanding of the correlation of electronic stateswith the destruction yields of important ice systems, as well as photon-induced chemical reaction products and decay pathways in cosmic ices. Theresults are important to our understanding of EUV–UV photon-induced icechemistry in cosmic ice analogs.

Acknowledgments

We are grateful for the support of the staff of the Synchrotron ResearchRadiation Center, Hsinchu, Taiwan. We appreciate valuable suggestionsby the referees. This research is based on work supported by the NSFPlanetary Atmospheres Program under Grant AST-0604455 (Wu) and theAstrophysics and Astrochemistry program of SRRC (Cheng).

References

1. R. E. Johnson, Energetic Charged-Particle Interactions with Atmospheres andSurfaces (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.

Page 125: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

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, Presentedin the 38th Annual Meeting of the Division for Planetary Sciences of theAmerican Astronomical Society (Pasadena Convention Center, Pasadena,CA, October 8–13, 2006).

11. P. Jenniskens, G. A. Baratta, A. Kouchi, M.-S. deGroot, J. M. Greenbergand 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 10K, 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. B38 (2005) 3693.

17. B.-M. Cheng, unpublished results (2005).18. C. Romanzin, M.-C. Gazeau, Y. Benilan, E. Hebrard, A. Jolly, F. Raulin, S.

Boye-Peronne, S. Douin and D. Gauyacq, Adv. Space Res. 36 (2005) 258–267and 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 NASALaboratory 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.

Page 126: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

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.

Page 127: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch10

This page intentionally left blankThis page intentionally left blank

Page 128: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

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

FORMATION OF CaTiO3 CRYSTALLINE DUSTIN LABORATORY

KAORI YOKOYAMA∗, YUKI KIMURA, OSAMU KIDO, MAMI KURUMADA,AKIHITO KUMAMOTO and CHIHIRO KAITO

Department of Physics, Ritsumeikan University, 1-1-1 NojihigashiKusatsu City, Shiga 525-8577, Japan

[email protected]

The crystalline grain formation of perovskite (CaTiO3) by the coalescence dueto grain–grain collisions between TiO2 and CaO grains in smoke has beendemonstrated. Spherical grains with diameters of 100–200 nm were produced.A large quantity of perovskite grains were also produced by the condensationfrom CaTiO3 vapor evaporated in selecting Ar gas pressure of 10 Torr. Thesegrains contained the WO3 crystal in the center by the peritectic reaction. Theperovskite grains showed significant peaks at 14.4 and 21.9 µm in the opticalspectra.

1. Introduction

The discovery of the existence of the crystalline silicate grains in space bythe infrared (IR) space observatory1 has changed views on dust formation.We demonstrated, through laboratory experiments, that crystalline grainscan 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 thatcrystal grains can be produced by the coalescence between CaO and TiO2

grains. Through the present laboratory experiments, it became evident thatcollisions between different grains can produce the crystalline grains.

In the equilibrium condensation theory in our solar system, perovskite(CaTiO3) condenses at 1647K and disappears at 1393K4 after thecondensation of corundum (Al2O3). However, the discovery of perovskitein meteorites5 also suggests that the nonequilibrium condensation occursas well as corundum. Therefore, this fact greatly interest the formativecondition of CaTiO3 dust and its spectra in the laboratory.

The thermochemical condition of condensation in late-type M dwarfsis fulfilled due to low temperature and high density.6 The high temperature

115

Page 129: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

116 K. Yokoyama et al.

condensates that are the most abundant in the atmospheres of late-type Mdwarfs and brown dwarfs with smaller ratio of C/O than unity are ZrO2

for T < 2000 K and Al2O3 for T < 1800 K. Other stable species to appearat T < 1600 K are Ca2Al2SiO7, Ca2MgSi2O7, and CaMgSi2O6, as well asTi4O7 and Ti2O3. These grains compete with the formation of perovskiteand corundum. Ti oxides (particularly CaTiO3) are the condensations athigh temperature. TiO2 and CaTiO3 are likely to be among the first nucleito form in brown dwarf atmospheres. The form of Ti oxides depends onthe brown dwarf’s effective temperature. For effective temperatures above2000K, CaTiO3 is produced, whereas for lower temperatures, Ti3O5 isproduced. Therefore, CaTiO3 is important for the formation of dust aroundbrown dwarfs.7 Therefore, the condensation of perovskite is the principalcause 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 Mg2SiO4 in which the mixed smoke inthe laboratory was due to the demonstration of solid–solid reaction. Themassive production of CaTiO3 grains have been succeeded with the orderof 50 nm by selecting the gas pressure, and the optical spectra at 14.4 and21.9 µm can be identified as the perovskite.

2. Experimental

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

Page 130: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

Formation of CaTiO3 Crystalline Dust in Laboratory 117

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

spectrometer (FTIR/Horiba FT210) in the wavelengths region from 2.5 to25 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 smokeare shown in Fig. 2. Spherical TiO2 grains with sizes smaller than 100 nmwere produced in the TiO2 smoke.10 Electron diffraction patterns show theformation of brookite and anatase. In the CaO smoke, although CaO grainswere predominately produced, the ED pattern also indicates the formationof Ca(OH)2, as shown in Fig. 2c. The CaO grains produced are easilychanging 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 andCaO grains. TiO2 grains of the rutile phase, which is a high temperaturephase of TiO2 were produced, as shown in Fig. 2b. The formation of thehigh-temperature phase of TiO2 in the mixed smoke was due to highertemperature of the mixed smoke region.12

Figure 3 shows a TEM image of typical grains collected from the mixedsmoke region. EDX-based spectroscopy was used to search for typical grains.

Page 131: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

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.

Page 132: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

Formation of CaTiO3 Crystalline Dust in Laboratory 119

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

It can be concluded that CaTiO3 grains had sizes of 100–200nm order. Thegrains with sizes smaller than 100 nm were TiO2. The coagulated grainssmaller than 50 nm with cubic and rod shapes were identified as CaO orCa(OH)2.

Figure 4 shows the electron diffraction (ED) pattern of the grainindicated 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 wereattributed to the CaTiO3 single-crystalline grain.

Page 133: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

120 K. Yokoyama et al.

single-crystalline grain seen along its [010] zone axis. Therefore, it can beconcluded 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 of100 nm were coalesced, the size of a CaTiO3 grain was estimated to be133.6 nm. Therefore, the large spherical grains were CaTiO3 which is inaccordance with the TEM image in Fig. 3. The CaTiO3 crystalline grainscan 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 CaOgrains, as indicated in Sec. 3.1. Because CaTiO3 has a high meltingpoint and is a refractory material, it is generally decomposed by directevaporation. Coalescence-based growth of grains predominately took placein the smoke, and the results described in Sec. 3.1 suggest the formationof CaTiO3 from solid–solid grain reaction in Ar gas. As is shown bythe formation of iron-oxide grains in a similar experiment, the selectionof an appropriate gas pressure, which governs the coalescence frequency,can produce the evaporated-material grains in spite of the decompositionphenomenon.13 In the present experiment, the gas pressure of 10Torrenabled successful the formation of CaTiO3 dust as follows.

The evaporation was performed using a V-shaped tungsten boat(50mm in length, 2 mm in width, and 1mm in depth) in Ar gas at a pressureof 10Torr. Figure 5 shows typical smoke samples obtained by the directevaporation of commercial CaTiO3 powder in Ar gas. Spherical grains withsizes smaller than 50 nm were predominantly produced. The ED patternclearly indicated the formation of CaTiO3 grains. The ED pattern alsoindicated the formation of WO3 crystal. The TEM image in Fig. 5 showsthe small black dots in CaTiO3.

These dots inside the CaTiO3 grains were WO3, as indicated inFigs. 6 and 7. Figure 6 shows bright- and dark-field images andcorresponding ED patterns of a single grain containing a black dot.The indexed ED pattern indicated both CaTiO3 with a [101] zoneaxis and WO3 with a [001] zone axis. The two crystals have definitelattice relations with [101]CaTiO3//[001]WO3 and (020)CaTiO3//(200)WO3.Therefore, drops of mixed CaTiO3–WO3 liquid were crystallized from

Page 134: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

Formation of CaTiO3 Crystalline Dust in Laboratory 121

Fig. 5. 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) correspondingdiffraction patterns. Strong spots were identified as CaTiO3 with a [101] zone axis, asshown in (c). Weak spots among CaTiO3 spot were attributed to WO3, as shown in (d).

the surface accompanying the peritectic reaction. The high-resolutiontransmission electron microscopy (HRTEM) images shown in Fig. 7 clearlyindicate 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 wereless than 200nm and their spectra are clearly different. Our peritectic grainsshowed a spectrum similar to that of commercial CaTiO3 powder. The 14.1

Page 135: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

122 K. Yokoyama et al.

Fig. 7. HRTEM image of a grain. The surrounding region was identified as the CaTiO3

structure. The magnified lattice image clearly shows the existence of the orthorhombicCaTiO3 structure.

Fig. 8. Mid-IR transmission spectra of grains of TiO2, CaO, and CaTiO3 containingWO3 and commercial CaTiO3 powder are shown. The features at positions of 14.4 and21.9 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 featuresof CaTiO3 in the present specimen in Fig. 8 were compared with Posch

Page 136: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

Formation of CaTiO3 Crystalline Dust in Laboratory 123

data.7 The FWHMs of the present peaks at 14 and 21 µm were 4.0 and1.9 µm, respectively. These values are larger than Posch et al.’s data of3.2 and 1.8 µm, respectively. The difference of grain sizes of the presentspecimen and Posch et al.’s sample may account for the differences in theFWHM. The stretched vibration at approximately 14.4 µm becomes broadbecause the TiO6 octahedron in the CaTiO3 structure is distorted. Thebroadness due to the distorted octahedron has been demonstrated by theWO3 grains.14 The feature at 21.9 µm corresponds to the TiO6 bendingvibration. The spectrum differences between the commercial powder andthe present specimen were due to the shape and size effects; commercialCaTiO3 powder grains have irregular shape and a mean size of 10 µm. Onthe other hand, the present nano grains were uniformly spherical with themean size of 50nm. The differences between Rayleigh and Mie scatteringmay 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, MeteoritPlanet. 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.

Page 137: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch11

This page intentionally left blankThis page intentionally left blank

Page 138: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

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

DIRECT OBSERVATION OF THE CRYSTALLIZATIONOF 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 1-1-1Kusatsu, Shiga 525-8577, Japan

[email protected][email protected]

HITOSHI SUZUKI

Department of Electric Engineering, Tohoku Gakuin University, Chuoh 1-13-1Tagajo, Miyagi 985-8537, Japan

The crystallization of amorphous Mg-bearing silicate grains into Mg2SiO4

crystal covered with a thin carbon layer was directly observed by in-situtransmission electron microscopy. The temperature of crystallization of thesample was observed to be 200C lower than that of the sample withoutthe carbon layer. The graphitization energy of the surface amorphous carbonlayer with the mean thickness of 10 nm accelerated the crystallization of thecentral amorphous Mg-bearing silicate grain of 100 nm order. Sample resultsfor crystallization at room temperature are presented.

1. Introduction

In a previous work, we demonstrated that the crystallization of amorphousMg-bearing silicate grains to Mg2SiO4 crystals takes place at 800C undervacuum.1 The crystallization started from the grain surface. We foundthat prenucleation occurs in the 650–800C temperature range beforecrystallization at 800C. The phenomenon in which the prenucleation statecorresponded to the stall state was suggested and clarified by Hallenbeckand Nuth using IR spectroscopy.2 Yamamoto and Chigai have proposeda chemical heating model to explain the crystallization mechanism ofcometary silicate, and have shown that the chemical heating mechanismleads to the degree of crystallization required to explain the observedstrength of the cometary crystalline features.3

125

Page 139: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

126 C. Kaito et al.

To elucidate this model using laboratory analogies, amorphous Mg-bearing silicate grains crystallizing at 800C were covered with anamorphous carbon layer. Since the amorphous carbon layer on the surfaceof CdTe crystal nanoparticles crystallized into a graphene layer after beingheated to 500C,4 the effect of the graphene structure alteration on thesurface of the central Mg-bearing silicate grains has been examined by in-situ transmission electron microscopy. On the other hand, the same Mg-bearing silicate grains covered with a carbon layer produced in a methaneatmosphere clearly showed graphitization at temperatures as low as roomtemperature. The results for these samples are presented in this paper.

2. Experimental Methods

Amorphous Mg-bearing silicate grains used were produced by the coale-scence between MgO and SiO2 smoke grains.5 The grains dispersed onglass plates were covered with an amorphous carbon layer by depositingcarbon 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 usinga Hitachi H-9000NAR electron microscope with a special heating holder,which can be heated up to 1500C.7

3. Results and Discussion

The carbon layer was deposited on the Mg-bearing silicate grains by thearc 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 surfaceof the grain was also covered with a thin carbon layer. This indicates thatcarbon atoms diffused efficiently on the silicate grains. The ED pattern ofthe sample clearly shows amorphous halo rings (up right corner inset ofFig. 1). The temperature of the specimen was increased by 50C stepsfrom room temperature. When the temperature of the sample reachedequilibrium, its movement was reduced. Selected temperature for each stepwas maintained for 15min. When the phase of the specimen changed, theselected temperature was maintained for more than 15min. As shown inFig. 2, the halo diffraction pattern based on the graphitic structure givenin Fig. 1 became sharper after heating to 550C. The substrate differences

Page 140: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

Crystallization of Amorphous Mg-bearing Silicate Grains 127

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

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

Page 141: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

128 C. Kaito et al.

Fig. 3. Crystallization occurred at 600C. The ED pattern can be assigned to theMg2SiO4 crystallites. Amorphous grains were crystallized at a lower temperature thanthe amorphous silicate grains without a carbon layer. The crystallization of Mg-bearingsilicate grains without a carbon layer occurred at 800C as elucidated in the previouspaper.

of the CdTe crystal and the amorphous silicate may affect the difference ofthe graphitization temperature.

After heating to 600C, the crystallization of the central amorphousMg-bearing silicate occurred as shown in Fig. 3. The ED pattern can beindexed to correspond to Mg2SiO4. In a previous work, the prenucleationstate was directly observed at 650C. With the present sample, theprenucleation state could hardly be observed. The crystallization occurredwith the graphitization of the amorphous carbon layer at a lowertemperature. A typical example of image of a graphitized carbon layeris shown in Fig. 4. A fast Fourier transform (FFT) at one part of thehigh-resolution transmission electron microscopic (HRTEM) image clearlyshowed the graphitization of the surface carbon layer as depicted by thebottom left corner inset of Fig. 4. The crystallization heat, which isusually interpreted as occurring at the transformation from the metastableamorphous phase into stable crystalline phase, is generated when thedangling bonds are connected to form stable bonds or are reconnectedto form an atomic arrangement as is the crystal. The temperature of theinterface is thereby raised and the crystallization may be accelerated. Thecrystallization heat raises the temperature of the adjacent microcrystallitesof graphite structure and effectively promotes the graphitization on thecovered layer. In the graphite crystal, the thermal conductivity is 3.1 times

Page 142: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

Crystallization of Amorphous Mg-bearing Silicate Grains 129

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

larger in the perpendicular direction of the c-axis than in the directionnormal to the c-axis. The carbon layer arranged graphitic layer parallelto 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 thecentral amorphous grains. The volume of amorphous carbon layer withmean thickness of 10 nm on the 100 nm amorphous silicate grain wasabout 0.7 times of the volume of silicate grain. So, the alteration of theamorphous carbon layer accelerated the crystallization on the amorphousgrains. Therefore, it is supposed that the energy excess due to graphitizationwas transferred to the central amorphous Mg-bearing silicate grain. Theobserved decrement of the crystallization temperature supported the theoryproposed by Yamamoto and Chigai on the crystallization of amorphoussilicate induced by the chemical reaction energy on the surface layer.

To achieve a lower temperature crystallization for amorphous silicategrains, the same Mg-bearing silicate grains were covered with a carbon layerproduced by the arc discharge of carbon in a CH4 atmosphere of 10−3 Torr.The carbon layer containing CH4 was graphitized at room temperature

Page 143: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

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 30min after the grains were produced. Image (b) was observed after4200 min in air. The white-ring contrast between (A) and (B) is due to the change indensity on the deposited carbon layer caused by graphitization. The surface of (B) alsochanged to ring contrasts at room temperature as indicated by the arrows.

Fig. 6. HRTEM image of part of amorphous silicate surface. Crystal Mg2SiO4 formationwas clearly observed. The graphitization of the carbon layer also occurred in the spotsindicated by an arrow.

Page 144: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

Crystallization of Amorphous Mg-bearing Silicate Grains 131

by keeping in air, and the partial crystallization of the amorphous Mg-bearing silicate occurred at room temperature as shown in Fig. 5. Thesame positions as previously selected were observed by the special specimenholder. The as-prepared specimen was exposed in air and immediatelyobserved by electron microscope (Fig. 5a). The variation of density dueto the crystallization proceeded from the surface to the central region.The circular contrast layer in the carbon layer and the amorphous Mg-bearing silicate can be clearly seen. An HRTEM image of the Mg2SiO4

crystal is shown in Fig. 6. In addition to the graphitization indicated by anarrow, Mg2SiO4 crystals with a size of 3 nm are seen. The crystallizationin Fig. 3 and previous paper took place suddenly from one or somecrystallite from the surface of the amorphous silicate grains. In Fig. 5,the crystallization took place concentrically. Detailed experimental resultsincluding the spectral alteration of this sample at room temperature willbe published elsewhere.

4. Summary

Mg-bearing silicate grains crystallized at 600C with a carbon layer withouta prenucleation stage indicated that the temperature of crystallizationdecreased due to the energy excess of graphitization. The energy broughtto the samples by heating helped significantly the crystallization process tooccur. The amorphous Mg-bearing silicate grains covered with the carbonlayer containing CH4 crystallized at room temperature in agreement withYamamoto 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.

Page 145: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch12

This page intentionally left blankThis page intentionally left blank

Page 146: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

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

RELATIONSHIP BETWEEN MORPHOLOGY ANDSPECTRA REVEALED BY DIFFERENCE IN MAGNESIUM

CONTENT OF SPINEL PARTICLES

MIDORI SAITO∗, MAMI KURUMADA and CHIHIRO KAITO

Department of Physics, Ritsumeikan University, 1-1-1 NojihigashiKusatsu-shi, 525-8577 Shiga, Japan

[email protected]

Spinel particles containing Mg of different amounts were obtained by flash gasevaporation from different mixture powders of Mg and Al in a mixture gasof argon and oxygen. By decreasing Mg content, the shape of the particleswas changed from cubic, octahedral, elliptical to spherical. The characteristicspectra among these produced particles are indicated. The difference betweenthe observed peak positions and the calculated absorption positions wasdiscussed 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 asCM2 meteorite Murray or CI chondrite Orgueil.1,2 It condenses at 1240Cfrom a cooling gas of cosmic composition at thermodynamic equilibriumand disappears at 1089C due to the formation of anorthite (CaAl2Si2O8).3

However, spinel was not expected to appear in nonequilibrium processtheory. Spinel formation due to solid–solid reaction was proposed.4 Oneof the present authors demonstrated the formation of spinel by MgO solid–Al2O3 solid reaction using nanosized particles.5

On the other hand, the 12–13 µm emission feature was found in thespectra of oxygen-rich AGB stars by ISO spectra observation. The originof 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 phasein a mixture gas of argon and oxygen.9 Although γ-alumina is consideredas the carrier of the 13 µm feature, the δ-alumina grains also have a strongabsorption at 13 µm. Furthermore, the C/O ratio of AGB stars is closelyrelated 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

Page 147: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

134 M. Saito, M. Kurumada and C. Kaito

also found that the 13 µm feature is observed in stars within a rather narrowrange of photosphere and dust shell temperatures.

The spinel is also considered as the carrier of the 13, 17 and 32 µmemission features observed in the ISO spectra of oxygen-rich stars.8,12

Fabian et al. calculated the absorption efficiencies of sub-µm-sized sphericalparticles from the spectra of synthetic spinels, and showed the 13, 17, and32 µm features.8

In the present experiment, we produced nm-sized spinel particles byflash gas evaporation. Morphological and spectral alterations induced byvarying the content of Mg are elucidated. The effect of production conditionor size on the spectrum of spinel is also discussed on the basis of the 14 and18 µm features in their spectra.

2. Experimental Procedure

The work chamber used was a glass cylinder of 170mm diameter and330mm height. After evacuating the chamber to approximately 10−5 Torr,argon gas of 45Torr and oxygen gas of 5Torr were introduced into thechamber. 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 washeated above 1800C, and then the produced nanoparticles can be seen inthe smoke owing to the scattering of radiation light from the heater. In thissmoke, the nanoparticles were grown by coalescence.13 The mixing ratio of

Fig. 1. Schematic representation of flash gas evaporation method.

Page 148: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

Relationship between Morphology and Spectra 135

dropped powder was altered, such as Mg:Al = 1:1, 1:2 or 1:3, with weightratio.

The produced particles were collected on a glass plate, dispersedin ethyl alcohol, and mounted on an amorphous carbon film supportedby standard copper electron microscope grids. A transmission electronmicroscope (TEM) equipped with an energy-dispersion X-ray (EDX)analysis system (Horiba EMAX-5770) was used to observe the morphologyand analyze the structure of the particles. High-resolution transmissionelectron microscopy (HRTEM) was carried out using a Hitachi H-9000NARelectron microscope.

The produced particles were also buried in KBr pellets and theirtransmittances were measured with a Fourier transform infrared (IR)spectrometer (Horiba FT210) from 2.5 to 25.0 µm. The wavelength resolu-tion was 2 cm−1. The beam splitter was a Ge-evaporated KBr substrateand 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 oftypical particles produced by varying mixture ratio are shown in Fig. 2.The morphology of the markedly altered particles depends on the massratios 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 almostcorresponds to the atomic ratio. The mixed metallic powder was flashed inthe atmosphere of mixed gases of argon and oxygen; the produced particleswere oxides. Therefore, the evaporated Mg and Al vapors were burnedin the atmosphere and produced spinel (MgAl2O4), MgO, and δ-Al2O3.By incrementing Al content, δ-Al2O3 particles were produced, which gavethe characteristic spectrum in the range of 13–18 µm as elucidated in aprevious paper.9 Figure 2a shows the polyhedral shape. The correspondingED pattern indicates that the produced particles were MgO and MgAl2O4.The cubic shape is typical for MgO particles.14 MgO reflection in the EDpattern 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 1:1sample. That is, the spinel was efficiently produced under the conditionof Mg:Al = 1:2, i.e., the stoichiometric composition of MgAl2O4. Themorphology of the particles changed to spherical with the 1:3 sample.

Page 149: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

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 shownin Fig. 3. The results of EDX analysis for morphological alteration arealso shown in Fig. 3. Since an electron beam of 30–50nm size can befocused on the specimen, nanoprobe analysis was performed on about

Page 150: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

Relationship between Morphology and Spectra 137

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

10 particles for each shape. The atomic ratio for each morphology isshown in Fig. 3. MgO particles were cubic. When the Mg content was75–100at.%, the morphology was altered to the (110) truncated form fromthe cubic shape. With increasing Al content, 111 octahedral or truncatedoctahedral particles appeared. This morphological change is similar to thatof Pb-doped KCl crystal.15 Whereas pure KCl is cubic with the 100plane, its morphology is changed to octahedral with the 111 plane byimpurity doping. In the case of KCl, the impurity concentration is of ppmorder. Although the order of impurity concentration differs markedly, thesemorphological changes seem to be the same phenomenon.

The HRTEM images of the cubic-form particles containing Al atomswhich were verified by EDX analysis are shown in Fig. 4. The small whitedots 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. Ascan be seen from the enlargement of the lattice image in Fig. 4B, the (200)lattice images had different contrasts in short-range order, which correspondto the small white dots. Since MgO oxide is the characteristic metal excessoxide, the excess Al will be distributed on (100) planes as in the G.P. zone

Page 151: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

138 M. Saito, M. Kurumada and C. Kaito

Fig. 4. HRTEM images of an MgO cubic particle containing Al. The excess Al atomsdistributed 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, asshown 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-Al2O3 particles produced by gasevaporation were cubic and spherical, the particle shapes were alteredfrom MgO to Al2O3 phases containing different amounts of Mg and Al.Since both δ-Al2O3 and spinel (MgAl2O4) have the same structure, noconsiderable morphological change occurred in the D-region. However, inthe 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 (MgAl2O4) and MgO phaseswas observed by HRTEM. Figure 5 shows the octahedral particle containingMgO. Since the lattice images of (400) MgAl2O4 and (200) MgO were inparallel, the central spinel particles were covered with MgO, in which adefinite lattice relation with MgO existed: (100) MgAl2O4//(100) MgO.The surface covered with MgO crystals entirely had a definite orientationwith respect to the central particles (MgAl2O4). The special moire-likecontrasts in the central region was due to the parallel relation betweenMgAl2O4 and MgO. This parallel growth shows that the MgO and

Page 152: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

Relationship between Morphology and Spectra 139

Fig. 5. HRTEM images of the particles with moire fringes. This particle is based onthe spinel structure and the surface covered with MgO particles.

MgAl2O4 were obtained through the mixture phase during the growthprocess. Since the temperature of solidification of MgAl2O4 from the liquid-like state is lower than that of MgO, MgO solidification took place from thesurface as has been recently demonstrated on the crystallization process ofamorphous Mg2SiO4.18 The MgAl2O4 grew with a definite lattice relationwith the surface of MgO crystals.

Figure 6 shows HRTEM images of Al2O3-rich spinel particles (D-region) with a characteristic stripe contrast similar to that of a superlattice.Since EDX analysis indicated that the particle with the stripe contrastcontained 4–6 at.% Mg, these stripe contrasts may be due to the coexistenceof the spinel and alumina phases. The (040) lattice of alumina was theregion with different contrasts. Some of the strong-contrast regions that

Fig. 6. HRTEM images of spherical particles with stripe contrast.

Page 153: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

140 M. Saito, M. Kurumada and C. Kaito

were similar to the G.P. zone appeared partly on the particles. These strong-contrast lattice images correspond to the (400) lattice image of spinel. The(040)Al2O3 // (400)spinel relation existed on these particles. Since the Al2O3

and MgAl2O4 have the same spinel structure, they were distributed in theoctahedral and tetrahedral sites on the oxygen close-packed structure. Thecoexistence of the two-phase mixture was easily realized by the diffusionof Mg metal. Therefore, the coexistence of two phases was achieved duringthe 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 Alare shown in Fig. 7. The spectra of MgO and δ-Al2O3 particles producedusing 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-Al2O3

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 µmcorrespond to the octahedral and tetrahedral units in the spinel structure,respectively.19 The spinel phase was contained in all the specimens, and thecrystallographic shape of produced particles varied from cubic to sphericalwith increasing Al ratio. Since these peaks are totally different from those of

Fig. 7. IR spectra of MgO, δ-Al2O3, and Mg:Al = 1:1, 1:2 and 1:3 specimens. Thespectra have two characteristic peaks at 14 and 18 µm.

Page 154: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

Relationship between Morphology and Spectra 141

MgO and δ-Al2O3 phases, the features of the spectrum reflect the alterationof the morphology. Since the peak at Mg:Al = 1:1 was attributable to thecontent of the MgO phase being markedly higher than those of the otherspecimens, the overlap of the MgO absorption and spinel absorption peaksgives the broad feature. A predominant growth of MgAl2O4 was observedat Mg:Al = 1:2 and the feature becomes sharper.

4. Conclusion

The morphology of the MgO–Al2O3 compound markedly changed withcomposition. The morphologies were cubic for Mg-75–100at.% particles,octahedral for Mg-70 at.% particles, elliptical with characteristic contrastfor Mg-35–65at.% particles, or spherical for less than Mg-40 at.% particles.HRTEM observation indicated that the structure of octahedral particleswas based on spinel, and the structure of spherical particles was based onAl2O3. The characteristic moire contrast in elliptical particles was due tothe coexistence of MgO and Al2O3.

The IR spectra and ED patterns showed that the spinel phase waspredominantly produced at Mg:Al = 1:2. The IR spectrum of the nm-sizedspinel showed absorption peaks at 14 and 18 µm, which varied with theirmorphology.

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.

Page 155: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch13

142 M. Saito, M. Kurumada and C. Kaito

12. A. K. Speck and A. M. Hofmeister, Workshop on Cometary Dust inAstrophysics, (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.

Page 156: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

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

IONIZATION OF POLYCYCLIC AROMATICHYDROCARBON MOLECULES AROUND THE HERBIG

Ae/Be ENVIRONMENT∗

ITSUKI SAKON†,§ and TAKASHI ONAKA

Department of Astronomy, Schools of Science, University of Tokyo7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

§[email protected]

YOSHIKO K. OKAMOTO

Institute of Astrophysics and Planetary Sciences, Ibaraki University, Japan2-1-1 Bunkyo, Mito 310-8512, Ibaraki, Japan

HIROKAZU KATAZA, HIDEHIRO KANEDA and MITSUHIKO HONDA‡

Institute of Space and Astronautical ScienceJapan Aerospace Exploration Agency

3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510, Japan

We present the results of mid-infrared N-band spectroscopy of the HerbigAe/Be system MWC1080 using the Cooled Mid-Infrared Camera andSpectrometer (COMICS) on board the 8m Subaru Telescope. The MWC1080has a geometry such that the diffuse nebulous structures surround the centralHerbig B0 type star. We focus on the properties of polycyclic aromatichydrocarbons (PAHs) and PAH-like species, which are thought to be thecarriers of the unidentified infrared (UIR) bands in such environments. A seriesof UIR bands at 8.6, 11.0, 11.2, and 12.7 µm is detected throughout the systemand we find a clear increase in the UIR 11.0 µm/11.2 µm ratio in the vicinity ofthe 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 asolo-CH out-of-plane bending mode of neutral PAHs, the large 11.0 µm/11.2 µmratio directly indicates a promotion of the ionization of PAHs near the centralstar.

∗This work is based on data collected at Subaru Telescope, which is operated by theNational 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, 2946Tsuchiya, Hiratsuka, Kanagawa, 259-1205, Japan.

143

Page 157: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

144 I. Sakon et al.

1. Introduction

The unidentified infrared (UIR) bands are a series of emission bandsobserved at 3.3, 3.4, 6.2, 7.7, 8.6, 11.0, 11.2, 12.0, and 12.7 µm togetherwith some other fainter features. They have been ubiquitously observedin various astrophysical objects, including reflection nebulae, HII regions,planetary nebulae, post-AGB stars,1 diffuse interstellar medium,2,3 externalstar forming galaxies,4 remote ultraluminous infrared galaxies,5 andsubmillimeter galaxies even at z = 2.8.6 They are supposed to be carriedby 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 incascades 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 carriersof the UIR bands since the absorbed photon energy will not be confinedwithin the aromatic group within/attached to the bulk QCC or amorphouscarbon dust.11,12 The portion of the pumping energy to each vibrationmode is supposed to be controlled by the physical and chemical conditionsof the carriers such as the charging state, the molecular structure, andthe size of the carriers, which follow as a consequence of the physicalprocessing in the incident radiation environment as well as of the molecularevolution in chemically reactive regions. Therefore, understanding thesystematic differences in UIR spectra in terms of the variation in thenature of the carriers in various astrophysical environments is, aboveall, important to use UIR bands as a useful probe of the local physicalconditions.13

The ionization state of PAHs is one of the most significant factorsto affect the spectral characteristics of the UIR bands. The ionizationof PAHs is controlled by U/ne, where U is the strength of the incidentradiation 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 andtheoretical studies have shown that the UIR bands in the 6–9 µm region aremuch 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 whenPAHs are ionized.14–15 Actually several studies report that the variationsin 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

Page 158: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

Ionization of Polycyclic Aromatic Hydrocarbon Molecules 145

among the Herbig Ae/Be stars with different spectral types19 and betweenthe inner and outer Galactic plane20 have been reasonably explained by thechanging in ionization status of the carriers of the UIR bands.

The variations of 11.0 µm feature in the real astrophysical object havebeen firstly reported by Sloan et al.21 They observed the reflection nebulaNGC1333 SVS 3 using the 5m Hale telescope at Palomar, and found anexcess at 10.8–11.0 µm and a feature around at 10 µm increase relative tothe 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 µmwith increasing distance from the central star based on the observationof reflection nebulae NGC7023 with Infrared Spectrograph on the SpitzerSpace Telescope. Bregman and Temi17 have investigated the variation in theband ratio of 11.2 µm/7.7 µm within a reflection nebula along the distancefrom the central star and have made a quantitative evaluation of therelation between the 11.2 µm/7.7 µm and the ratio of the incident radiationfield strength to the electron density. However, the 7.7 µm and 11.2 µmfeatures come from different vibration modes (the former one correspondsto aromatic C–C stretching and the latter one to aromatic C–H out-of-plane bending, see Ref. 10 for details), and their relative strengths areaffected by various factors other than ionization, such as, the degree ofhydrogenation, the molecular structures, and the molecular sizes.23 On theother hand, the 11.0 and 11.2 µm features come from the same vibrationmodes but with different ionization status of the carriers and, therefore, the11.0 µm/11.2 µm band ratio can be used as more direct and quantitativemeasure 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 investigatingthe spectral changes in these features around the Herbig Ae/Be systemMWC1080 using the Cooled Mid-Infrared Camera and Spectrometer(COMICS)24 on board the 8 m Subaru Telescope. Quite recently, Habartet al.25 have presented the spatially resolved PAH emission in the innerdisks of nearby Herbig Ae/Be stars using adapted optics system on boardNACO/VLT. Our attempts would surely be useful to understand thephysical conditions of the disk around Herbig Ae/Be stars as well asthe evolution of disk materials including carbonaceous dust when thespatially well-resolved mid-infrared spectra of nearby Herbig Ae/Be starsare obtained.

Page 159: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

146 I. Sakon et al.

2. Observations and Data Reduction

2.1. Target

The mid-infrared spectroscopic data of MWC1080 were taken on the nightsof July 16–17, 2005 (UT). MWC1080 is a non-isolated Herbig Ae/Bestar located at the distance of 1.0 kpc26 surrounded by a bright reflectionnebulae27 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 younglow-mass stars within 0.3 pc radius from the central star based on theobservation of the CFHT with the high-resolution adaptive optics. Opticalbipolar outflows in the form of Herbig-Haro (HH) objects or HH-like jetswith the radial velocity of 400km−1s have been discovered predominantlyin the east of MWC1080.30

We have obtained the N-band low-resolution spectra with the resolvingpower of R ∼ 250 using the 0.33′′ slit. We also carried out the imagingobservation at 11.7 µm (∆λ = 1.0 µm) to adjust the slit position. In order tocancel the high infrared background radiation, secondary mirror choppingwas used at the frequency of 0.45Hz with 20′′ throw. The spectra wereobtained 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), thecompanion star (S2), and the nearby source (S3) (hereafter SLIT0716, seeFig. 1a) and the total integration time was 810 sec, which enabled us toobtain 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 theSubaru/COMICS N-band 11.7 µm image of MWC1080.

Page 160: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

Ionization of Polycyclic Aromatic Hydrocarbon Molecules 147

than 10. The other one is set at a position angle of −25.00 to observe thenebulae and the illuminating star (hereafter SLIT0717, see Fig. 1b) and thetotal integration time was 1800 sec, which was sufficient enough to observethe variations in the diffuse UIR emission.

Non-isolated Herbig Ae/Be stars provide us decisively an ideal envi-ronment to investigate the ionization effect of PAHs on their spectra. HerbigAe/Be stars are a pre-main sequence object of 2–8M and the ionizingregions of hydrogen gas are restricted only within a few AU from the centralstar. The low ne environment, where carbon atoms instead of hydrogendominantly supply electrons, realizes extremely high U/ne in the vicinityof the central star (within several hundred AU) compared to HII regions,taking account of the typical atomic abundance of C/H ∼ 2 × 10−4 forpre-main sequence objects.31

2.2. Data reduction

The standard chopping subtraction and flat-fielding by thermal spectra ofthe telescope cell cover were employed. Then the spectra of MWC1080along the SLIT0716 and SLIT0717 as well as the standard star HD3712were obtained. The wavelength calibration was performed using theatmospheric emission lines and the uncertainty was estimated to be lessthan 0.0025 µm.32 The spectra of MWC1080 along SLIT0716 and SLIT0717were divided by the standard star HD3712 spectrum for the purpose ofcorrecting the atmospheric absorption, and then we multiplied the resultingspectra by the template spectrum of the standard star provided by Cohenet al.33 Finally, we adjusted the flux measured in the N8.8 and N12.4 µmimaging bands to correct the slit throughput. The uncertainty in the flux ateach wavelength was estimated from the noise in the blank sky. We note thatthe 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 illuminatingcentral B0 star and the obtained spectrum is dominated by the strongemission from the photosphere. A slight dent in 9–10.5 µm seems to be theeffect of absorption by amorphous silicate but we note that this wavelengthregion is suffered by the atmospheric ozone absorption. On the other hand,

Page 161: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

148 I. Sakon et al.

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

S2 is the companion located at less than 1 arcsecond west to S1 and theobtained spectrum shows the clear presence of crystalline silicate features.The peaks around at 9.3 and 10.5 µm are supposed to be carried bycrystalline enstatite, and those around at 11.3 and 11.9 µm are supposed tobe carried by crystalline forsterite. The spectrum at S3 exhibits a featurepeaking around 11.3 µm, where the contribution from the UIR 11.2 µmcannot be distinguished from that from the crystalline forsterite. However,we cannot further discuss the dust composition for this spectrum sinceother features characteristic to PAHs nor to crystalline forsterite are hardlyrecognized due to the relatively low signal-to-noise ratio. S4 is the diffusenebula 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 illuminatedby the central B0 star and we assume that the projected distance fromthe central B0 star corresponds to the actual distance from the heating

Page 162: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

Ionization of Polycyclic Aromatic Hydrocarbon Molecules 149

source. The spectrum of the diffuse nebula is actually dominated by theUIR features as observed in S4 on SLIT0716.

Along SLIT0717, therefore, we can investigate the spectra of PAHs invarious strengths of the radiation field. In order to compare the profilesof these features at different positions, we subtract a local underlyingcontinuum defined by linear interpolation between the average valuesaround 10.9 and 11.65 µm.34 Figure 3 shows the variation in the continuum-subtracted UIR spectra of solo-CH modes normalized by the total intensityof their strengths along SLIT0717. We recognize a small hump located onthe blue shoulder of the distinct UIR 11.2 µm feature extending from 10.95to 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 thecontinuum-subtracted emission from 10.95 to 11.1 µm, and that of the UIR11.2 µm band is defined by the integration of the continuum-subtractedemission from 11.1 to 11.6 µm. 1 − σ errors for each band strengths aredefined by (σ2

m + σ2b)1/2, where σm is the measurement error and σb is the

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. Eachspectrum is normalized by the total intensity of the solo-CH bond features. The spectrumat the central star position corresponds to zero offset. The spectra are shifted.

Page 163: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

150 I. Sakon et al.

spectrum in units of Wm−2 µm−2 pixel−1 within the wavelengths usedfor calculating the strength of each UIR band and ∆λ is the width of eachwavelength region. σb is defined by σb = ηbase × ∆λ, where ηbase is thestandard deviation of flux density in units of Wm−2µm−2 pixel−1 of thespectrum at each position along SLIT0717 within the wavelength rangeused for the continuum definition.

Figure 4a shows the spatial distribution of the intensities of UIR 11.0and 11.2 µm features as a function of the offset from the central B0 staralong 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 therelative 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 areaof 0.165′′ × 0.165′′. Regions where 11.2 µm feature has local peaks of in its intensitydistribution (∆d ∼ −4,−2, 0, and 3) are indicated with shadows.

Page 164: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

Ionization of Polycyclic Aromatic Hydrocarbon Molecules 151

generally better signal-to-noise ratios than ∼3 where the 11.2 µm featurehas local peaks (i.e., ∆d ∼ −4,−2, 0, and 3; see Fig. 4) in its intensitydistribution (Fig. 4a). Such regions are expected to have larger columndensity of PAHs and/or better supply of UV photons than other regions.Figure 4b shows the spatial variations of the relative band strength of theUIR 11.0–11.2 µm features along SLIT0717. We find the ratio increasesup to ∼0.3 in the vicinity of the central star. Among the above fourregions around at ∆d = −4,−2, 0, and 3, where the UIR 11.0 µm feature issignificantly detected in each spectrum, the nearest position to the centralstar (∆d ∼ 0) shows the largest ratio while the most distant position fromthe central star (∆d ∼ −4) shows the smallest ratio.

Taking account of the fact that the 11.0 µm feature is assigned to asolo CH out-of-plane wagging mode of cationic PAHs while the 11.2 µmfeature to a solo CH out-of-plane bending mode of neutral PAHs,35 thelarge ratio of the UIR 11.0 µm/11.2 µm in the vicinity of the central star canbe interpreted as the promotion of PAHs’ ionization to cationic species. Wenote that the UIR 11.2 µm feature emitted from large PAHs may somewhatblue shifted in its peak position35,36 and a slight contribution from largestmembers of neutral PAHs to our calculated intensity of 11.0 µm featurewould be possible. However, in the following analysis, we assume that ourcalculated intensity of 11.0 µm feature is dominated by the emission carriedby the CH out-of-plane wagging mode of cationic PAHs. We also note thatsome types of crystalline silicate (e.g. crystalline forsterite)37 can contributeto the spectra in 11 µm region in the vicinity of the Herbig Ae/Be stars witha broad band feature peaking at 11.3 µm as can be seen in the spectrum ofS2 on SLIT0716 (see Fig. 2b), but their band widths are typically broaderthan those of UIR features. Since our continuum subtracted spectra do notshow an increase in the band width of 11.2 µm feature in the vicinity of thecentral star (Fig. 3), we assume that spectra above the continuum in the11 µm region along SLIT0717 are dominated by UIR bands in the followinganalysis.

4. Discussion

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

Page 165: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

152 I. Sakon et al.

In this model, the interstellar radiation field strength at a distance r fromthe B0 star U(r) in units of the solar vicinity is given by,

U(r) =

∫ 1µm

912AπBλ(T∗)(R∗/r)2dλ∫ 3 µm

912A4πJ

λ dλ

where T ∗ = 104.31 K is the effective temperature, R∗ = 3.2 × 109 m is theeffective radius of the central B0 star,38 and J

λ is the interstellar radiationfield of solar vicinity.39

We adopt a constant electron density ne = 100+300−50 (cm−3) from Poetzel

et al.30 in the region in our analysis. We examine the relation between theUIR 11.0 µm/11.2 µm ratio and U/ne each in the northwest part of the slitand in the southeast part of the slit (see Fig. 5). We can clearly see acorrelation between the UIR 11.0 µm/11.2 µm ratio and U/ne, suggestingthat the 11.0 µm/11.2 µm ratio can be a measure for the interstellarradiation field strength and the electron density. A relatively large scatterat small U/ne is supposed to originate from both the inhomogeneity in theelectron density and the underestimation of the radiation field strength atthe distant region from the central B0 star, where the contribution fromfaint low-mass stars29 cannot be neglected.

Detailed modeling of the interstellar radiation field strength and theelectron density as well as the additional spectroscopic observation of theUIR 11.0 and 11.2 µm around spatially resolved non-isolated Herbig Ae/Be

Fig. 5. (a) Relative band strengths of 11.0 µm/11.2 µm against U(r)/ne in the north-west 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 inelectron density ne = 100+300

−50 (cm−3) and those for the vertical axis are calculated fromthe uncertainties in the estimation of the UIR band strengths.

Page 166: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

Ionization of Polycyclic Aromatic Hydrocarbon Molecules 153

objects are quite important to examine the quantitative and robust relationbetween 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 systemMWC1080 using the COMICS on board the 8 m Subaru Telescope.MWC1080 has a geometry such that diffuse nebulous structures extendaround the central Herbig B0 type star. We focus on the properties ofPAHs and PAH-like species, which are thought to be the carriers of theUIR 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 clearincrease in the UIR 11.0 µm/11.2 µm ratio in the vicinity of the centralstar. Since the UIR 11.0 µm feature is attributed to a solo-CH out-of-planewagging 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 µmratio directly indicates a promotion of the cationic ionization of PAHs. Thiswork suggests an application and robust use of the UIR 11.0 µm/11.2 µmratio as a valid probe of the local interstellar radiation field strength andthe electron density.

6. Acknowledgments

The authors are grateful to all the staff members of the Subaru Telescopefor 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 thePromotion of Science (JSPS). Y.K.O. is supported by a Grant-in-Aidfor 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.

Page 167: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:56 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch14

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.

Page 168: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

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

SEARCH FOR SOLID O- AND N-RICH ORGANIC MATTEROF PREBIOTIC INTEREST IN SPACE

G. M. MUNOZ CARO∗,† and E. DARTOIS‡,§†Centro de Astrobiologıa, INTA-CSIC

Ctra. de Ajalvir, km 4, Torrejon de Ardoz, 28850 Madrid, Spain‡Institut d’Astrophysique Spatiale, UMR 8617, Bat. 121

Universite Paris XI, 91405 Orsay, France∗[email protected]

§[email protected]

Mainly based on our previous results, this article evaluates the presence ofsolid organic matter of prebiotic interest in space: from the carbon grainsobserved toward diffuse interstellar clouds and the organic grain mantles madefrom ice processing that are likely present in dense interstellar clouds andcircumstellar regions, to the carbon component of solar system objects thatcould 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 speciesor their precursors, and are therefore interesting for astrobiology. Organicresidues made in the laboratory from ice ultraviolet-photoprocessing undersimulated interstellar conditions are used as representative of organic matterand compared by means of infrared and Raman spectroscopy to carbon-bearingextraterrestrial samples. It is observed that the carbon bulk in grains of thediffuse interstellar medium, carbonaceous chondrites, and the IDPs collectedin the stratosphere consists of amorphous carbon, with at most a small percentof organic matter. On the other hand, about 50% of the carbon component incomet Halley is made of organics that formed in the interstellar/circumstellarmedium in the absence of liquid water. The characterization of cometaryorganic matter is therefore vital to constrain the contribution of extraterrestrialmatter to the origin of life on Earth.

1. Introduction

Carbon is sixth in the scale of cosmic elemental abundances. It is acomponent of many gas phase molecules in the interstellar medium, fromdiatomic molecules like CO to large polyaromatic hydrocarbons. Most of theobserved solid carbon in the interstellar medium is in the form of amorphouscarbon, graphite, diamond, carbides, and ices. Icy grain mantles in dense

155

Page 169: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

156 G. M. Munoz Caro and E. Dartois

interstellar clouds and circumstellar regions are mainly composed of H2O,carbon-containing molecules (such as CO, CO2, CH3OH, OCN−, OCS,H2CO, HCOOH, CH4), and NH3 or NH+

4 .1–3 Ulatraviolet (UV) irradiationand warm-up of such ice mixtures in the laboratory leads to an organicrefractory residue that is rich in prebiotic organic species or their precursors(carboxylic acids and their ammonium salts, alcohols, amides, esters, aminoacids, N-heterocycles, etc.).4–10

The solar nebula that preceded our solar system formed by gravitationalcollapse of a portion of a dense interstellar cloud. Comets and primitivemeteorites preserve carbon matter of interstellar or primitive solar nebulaorigin.

As already mentioned, the term organic is attributed to hydrocarbonmaterials rich in O and N that contain prebiotic species or their precursors,and are therefore interesting for astrobiology. Here, we search for solidorganic matter in space that might have contributed to the origin oflife on Earth. This is done by comparing organic residues made fromice photoprocessing, used as representative of organic matter that couldbe present in space, to carbon-bearing matter observed in the diffuseinterstellar medium and solar system bodies such as comets, meteorites,and interplanetary dust particles (IDPs).

The characterization that enables such comparison is accomplishedby infrared spectroscopy and, for laboratory available samples, alsoRaman spectroscopy. In particular, the profile of the infrared 3.4-µmfeature (3000–2800cm−1, CH stretching modes in hydrocarbons) and thepresence/absence of the bands associated with functional groups serve tocharacterize the carbon material either as organic in nature or as amorphouscarbon. Amorphous carbon consists of an amorphous matrix composed ofsp2, sp3, and even sp1, hybridized carbon atoms. a-C refers to amorphouscarbon with less than 20% hydrogen. Hydrogenated amorphous carbons(a-C:H) contain 20–60% hydrogen and have as a consequence a low C-Csp3 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 welltraced by Raman spectroscopy. Two bands dominate the Raman spectrumof amorphous carbon, the D (disorder mode, 1300–1500cm−1 range) andG (graphite mode, around 1600 cm−1) bands. The D band corresponds toa breathing mode in aromatic rings and the G band is due to a C=Cstretching mode in olefinic or aromatic carbon. From the intensity ratio ofthese bands, I(D)/I(G), the value of the aromatic domain size, La, can beobtained.11

Page 170: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

Search for Solid O- and N-Rich Organic Matter 157

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

2. Organic Refractory Matter of Prebiotic Interest Made fromUV-Irradiation of Interstellar/Circumstellar Ice Analogs

Energetic ice processing (UV photons and cosmic rays) is expected to takeplace in dense interstellar clouds, the sites of star formation, and the outerparts of circumstellar regions.

Laboratory experiments simulating the photo- and thermal processingof interstellar ice analogs show the formation of new molecules, radicals,and other fragments.12 Large organic compounds are produced by both icephotoprocessing4–10 and by ion bombardment.13–15 Here we will focus onthe organic refractory products of ice UV-photoprocessing.

Experiments simulating the UV irradiation of ice mantles in circum-stellar/interstellar regions consist in the simultaneous deposition and UV-irradiation of a H2O:CH3OH:CO:CO2:NH3 ice layer at T ≈ 12K and P ≤10−7 mbar. The chemical evolution of the ice is monitored in situ by Fouriertransform infrared spectroscopy. After warm-up to room temperature arefractory residue is observed.

The infrared spectrum of the residue obtained from irradiation ofa H2O:CH3OH: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 tothe stretching modes of hydroxyl (OH, at 3500–2300cm−1), carbonyl(C=O, at ∼1650–1750cm−1), and carboxylic (COO−, at 1586 cm−1)groups.10 The 3.4-µm feature assigned to the aliphatic CH stretchingmodes, composed of just two shallow bands at ∼2925 and 2875 cm−1,is quite peculiar. Its profile is strongly affected by the electrophylicoxygen atom incorporated in the aliphatic structures. After integrationof the absorbance features it is found that the main component of thisresidue is hexamethylenetetramine [(CH2)6N4] as evidenced by its mainabsorption features at 1236 and 1007 cm−1, followed by ammonium saltsof carboxylic acids [(R-COO−)(NH+

4 )], amides [H2NC(=O)-R] and esters

Page 171: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

158 G. M. Munoz Caro and E. Dartois

Fig. 1. Infrared spectrum of the residue obtained after UV-irradiation and warm-up ofthe H2O:NH3:CH3OH: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-heterocyclicmolecules are both present with an abundance of the order of ∼1% bynumber of molecules.8,9

If the concentration of H2O in the starting ice mixture is lowcompared to the other ice components, species based on polyoxymethylene[(–CH2O–)n] are the most abundant. In general, residues showhigh O/C ratios; for the one resulting from UV irradiation ofH2O:CH3OH: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 strongphotoluminescence, presumably due to the high H/C and O/C ratios, andthe first-order D and G bands, characteristic of disordered carbonaceousmaterials, are not observed. Although these bands could be masked by thephotoluminescence continuum, they are likely not present simply becausethe organic refractory residues have no, or very few, olefinic and aromaticbonds, in agreement with the results obtained from infrared and GC-MSanalysis.17

3. Solid Carbon in the Diffuse Interstellar Medium

The infrared spectrum due to the absorption of carbon grains towarddiffuse interstellar clouds shows a prominent band at 3.4µm that is

Page 172: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

Search for Solid O- and N-Rich Organic Matter 159

very distinct from that of the organic residues described in Sec. 2 (ithas three subfeatures at 2923, 2958, and 2865 cm−1; respectively, thea-CH3, a-CH2, and s-CH3 stretching modes in aliphatics), and bandsdue to the CH bending modes at 6.85µm (around 1460cm−1) and7.25µm (around 1380 cm−1). The profiles and relative intensities ofthese features are very well matched with the laboratory spectrum of ahydrogenated amorphous carbon polymer, called photoproduced a-C:H,that is made from photoprocessing of simple aliphatic species underinterstellar conditions.18,19 This suggests that the composition of thismaterial is similar to that of carbonaceous grains in the diffuse medium.Based on the infrared and Raman analysis of photoproduced a-C:H, carbongrains in the diffuse interstellar medium are expected to be composedof an a-C:H polymer, of low oxygen content, consisting of hydrocarbonchains containing olefinic and aliphatic bonds with CH2/CH3 ≈ 2 andeventually some embedded small aromatic units (1–2 rings).18,19 Anexample of a possible substructure unit of photoproduced a-C:H is shownin Fig. 2.

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

Page 173: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

160 G. M. Munoz 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 outersolar system: on moons like Titan, on outer belt asteroids and Kuipertbelt objects, and in comets. The delivery of solid organic matter to theearly Earth was mainly caused by comets, meteorites, and IDPs; thecharacterization of the solid carbon fraction of these objects is summarizedbelow.

4.1. Comets

Comets are thought to preserve the most pristine material in the solarsystem as their high abundances of volatiles indicate. These bodies arethought to be formed by agglomeration of dust particles of pre-solar and/orsolar nebula origin in the outer parts of the solar nebula.20,21 However,the detection of crystalline silicates in comets, as inferred from infraredobservations,22 and preliminary results obtained from the analysis of thedust collected by the Stardust mission during flyby of comet Wild 2, indicatethat not all comets are made of pristine material that was kept cold since itsformation.23 The presence of high-temperature mineral grains suggests thatat least some cometary grains were reprocessed in regions near the early Sunand ejected by radial mixing to the outer regions of the nebula,24,25 wheresubsequently gas phase species would accrete onto such grains forming icemantles.

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 make-up of these molecules cannot be unambiguously identified.26,27 SeveralN-heterocyclic compounds are very likely to be present in the dust of cometHalley. These include pyrrole, pyrazole/imidazole, pyridine, pyrimidine, andits derivatives.28 More recently, low O and H and high N contents werereported for comet Wild 2.29 The Stardust samples will allow the firstlaboratory analysis of carbonaceous/organic cometary matter. The Rosettamission attempts to go a step further by analyzing in situ the compositionof a comet nucleus.

Page 174: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

Search for Solid O- and N-Rich Organic Matter 161

4.2. Meteorites

The main source of carbon in the meteorites known as carbonaceouschondrites is a kerogen-like material, but they also contain a large variety oforganic species, with concentrations of a few parts per million of the carbonabundance. These include carboxylic and dicarboxylic acids, amino acids,hydroxy-acids, amines, amides, nitrogen heterocycles including purinesand a pyrimidine, carbonyl compounds and alcohols.30,31 Stable isotopemeasurements indicate an interstellar origin of these compounds, althoughthe original organics probably underwent reactions in the aqueous phase onthe asteroidal parent body, which could have altered their composition.30

Organic species are embedded in the carbonaceous matrix. The rightpanel of Fig. 3 shows that the Raman D and G bands corresponding to thecarbon bulk of the Orgueil meteorite (top spectrum) are similar to thoseof IDPs (labeled Y, K2, K3, and N) and are characteristic of amorphouscarbon. The FWHM of the D and G bands corresponding to Orgueil andMurchison are close to the lower limit for IDPs. From the intensity ratioof 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 20rings in total, assuming a two-dimensional structure). These values fallwithin the IDP values. The infrared and Raman spectra of insoluble organic

Fig. 3. Left: Comparison between the infrared spectra of IDP Y (bottom) and aresidue (top) made from UV-irradiation of the H2O:CH3OH:CO:CO2:NH3 = 2:1:1:1:1ice 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 forintercomparison.39

Page 175: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

162 G. M. Munoz Caro and E. Dartois

matter in Orgueil and Murchison are consistent with that of a-C:H with aCH2/CH3 ratio close to the diffuse interstellar value around 2 and muchlarger aromatic units (20–30 rings in these meteorites compared to 1–2rings in carbon grains of the diffuse medium). This conclusion is consistentwith earlier characterizations. The insoluble organic matter of Murchisonwas earlier described as a structurally complex, extensively cross-linked,and highly aromatic macromolecule. The O/C and N/C values found forMurchison are 25 and 2.9%.32

4.3. Interplanetary dust particles

Interplanetary dust particles are collected by NASA aircrafts from theEarth’s stratosphere. These particles mainly consist of silicate grains coatedby carbonaceous material.33 They have an average carbon content of 10–12%.34 The detection of organic species in IDPs is hindered by theirsmall masses of the order of nanograms. The 3.4-µm feature profile ofIDPs is composed of three subfeatures shared by aliphatic species withno electrophylic heteroatoms. Although some IDPs are known to containcarbonyl (C=O) groups35,36 and hydrocarbon chains in some IDPs hostN,37 the infrared absorption bands due to functional groups are weak ornot present in most IDP spectra compared to infrared spectra of refractoryorganic residues made from UV irradiation of interstellar/circumstellar iceanalogs (see left panel of Fig. 3).38

The Raman spectra of four IDPs are shown on the right panel ofFig. 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-dimensionalstructure). The infrared spectra of IDPs (left panel of Fig. 3 shows thespectrum of IDP Y) are also well fitted with amorphous carbon, and provideinformation on the aliphatic component and the organic functional groups.From infrared and Raman spectroscopy of IDPs it is concluded that thebulk of the carbonaceous IDP component consists of a-C or a-C:H for IDPswith an observed 3.4-µm infrared feature. This corresponds to a materialwith 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 5.5for IDPs containing a-C:H, or a carbon sp3-skeleton for IDPs containinga-C. This material presents relatively low O/C and N/C contents.36,38

Page 176: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

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 manyIDPs showing three aromatic units depicted in two dimensions. Right: The real structureis not planar because it folds gradually as the number of aromatic units increases. Thea-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 functionalgroups (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 IDPsis 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–Hbonds, leads to a material that resembles spectroscopically the a-C or a-C:H in IDPs. It is thus possible that the carbon bulk in IDPs resulted fromheating of organic grain mantles in the solar nebula or during atmosphericentry heating. That can explain the morphology of cluster IDPs and thehigh D/H and 15N/14N associated with the carbon fraction in IDPs thatpoint to a formation at cryogenic temperatures.38

5. Presence of Organics Made from UV-Photoprocessing of Icein Small Solar System Bodies

Experiments show that photoprocessing of ice mantles which likely tookplace in the local dense cloud and/or the solar nebula leads to a variety

Page 177: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

164 G. M. Munoz Caro and E. Dartois

of molecules of biochemical relevance. Some of these molecules will bestored in comets and asteroids. There are significant similarities between theorganics found in comets and meteorites and the organic refractory residuesproduced by ice photoprocessing in the laboratory. First, the ubiquitousoxygen-rich complex organic molecules (O/C ≥ 0.5) found in the comaof comet Halley26–28 have an elemental composition very similar to thatof carboxylic acids and alcohols which comprise an abundant fraction ofthe organic residues.4,7 Glycolic acid, the most abundant carboxylic acidproduced by UV-photoprocessing and warmup of the H2O:CO:NH3 = 5:5:1ice mixture4,7 was found in the Murchison meteorite.40 Second, the N-heterocycles present in organic residues are similar to those inferred fromthe data of comet Halley; some of them are precursors of biologicalcofactors.28 Third, amino acids and diamino acids and/or their precursorsalso result from ice photoprocessing and warm-up.6,9 Most of these aminoacids are common to the Murchison meteorite. Based on the detectionof diamino acids in the organic residues, the presence of diamino acidsin Murchison was first predicted9 and they were later detected.41 Aminoacids are the components of proteins and diamino acids are peptidenucleic acid components, a possible precursor of RNA. This suggests thatenergetic processing of ice in the local dense cloud and the solar nebulacannot be disregarded since the above reported species (carboxylic acids,N-heterocycles, and amino acids) are not produced in the atmospheresof stars or by thermal reactions in the ice without irradiation. Streckersynthesis could explain the formation of species like α-amino acids onmeteorite parent bodies where aqueous alteration took place, but does notexplain the presence of organics in comets. Furthermore, for the synthesisin space of N-heterocyclic compounds and diamino acids, ice irradiation isto our knowledge the only reported formation mechanism.

6. Delivery of Extraterrestrial Organic Matterto the Early Earth

Extraterrestrial delivery of prebiotic molecules could have triggered theappearance of life on Earth.42,43 Shortly after its formation, the Earthsuffered numerous impacts by cometary bodies that were ejected from theirformation site in the Uranus–Neptune zone by gravitational perturbationsof the growing protoplanets,44,45 about 3.9Gyr ago, known as the era ofheavy bombardment. In this way, prebiotic species were delivered to theprimitive Earth via comets, and also carbonaceous asteroids and IDPs.46

Page 178: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

Search for Solid O- and N-Rich Organic Matter 165

It has been estimated that the mass of asteroids and comets reachingthe Earth after core formation was 0.7–2.7×1022 kg, and that cometsrepresented less than 0.1% by mass of the impacting population.47 Animportant fraction of the incoming extraterrestrial matter is carbonaceous.The presence of extraterrestrial prebiotic molecules in carbonaceouschondrites indicates that a fraction of these species survived the impact withthe Earth. The extraterrestrial carbon delivered on the Earth surface caneasily account for the surficial biomass of the Earth, i.e. about 1015 kg. Asan example, it was calculated that comet Halley organic matter correspondsto about 10% of the current biomass of the Earth.48

It is, however, important to note that organic matter represents asmall fraction of the total carbon in meteorites; the carbon bulk consistsof amorphous carbon, a priori less attractive for prebiotic chemistry. Thesame holds for IDPs (Sec. 4). As mentioned in Sec. 4.3, amorphous carboncould result from thermal annealing of pristine organic grain mantles that inturn were made by ice irradiation in space. In this context, cometary matterseems to be the best candidate to trigger chemical evolution, as proposedby Oro43: 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, whichdid not undergo aqueous alteration or substantial heating (organic matteris preserved only at temperatures below ∼300 C before severe annealingoccurs leading to graphitization,38 and therefore the organic grains in cometHalley were not exposed to such high temperatures). The same holds forcometary ice. During an impact, a fraction of the very low-density cometarydust could have been ablated off the surface of the comet and land gentlyon Earth, depositing the cometary organics relatively intact.49 Hopefully,the ongoing analysis of the Stardust samples, and the Rosetta mission, willcontribute to determine the composition of organic matter in comets.

Acknowledgments

G.M.M.C. was supported by a Marie Curie Individual Fellowship from theEuropean Union and a Ramon 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.

Page 179: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

166 G. M. Munoz 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. Munoz Caro, W. A. Schutte, W. H.-P. Thiemann,

B. Barbier and A. Brack, Chem. Eur. J. 11 (2005) 4895.9. G. M. Munoz Caro U. J. Meierhenrich, W. A. Schutte et al., Nature 416

(2002) 403.10. G. M. Munoz 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. Munoz Caro and E. Dartois, in prep.17. G. M. Munoz Caro, U. Meierhenrich, W. A. Schutte, W. H.-P. Thiemann and

J. M. Greenberg, Astron. Astrophys. 413 (2004) 209.18. E. Dartois, G. M. Munoz Caro, D. Deboffle, G. Montagnac and L. d’Hende-

court, Astron. Astrophys. 432 (2005) 895.19. E. Dartois, G. M. Munoz 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. Bockelee-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. Bockelee-Morvan, D. Gautier, F. Hersant, J.-M. Hure 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.

Page 180: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

Search for Solid O- and N-Rich Organic Matter 167

29. J. Kissel, F. R. Krueger, J. Silen 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. Munoz 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. Munoz Caro, G. Matrajt, E. Dartois et al., Astron. Astrophys. 459

(2006) 147.39. G. M. Munoz Caro and J. Martınez Frıas, 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 ArizonaPress, Tucson, 1988), p. 819.

41. U. J. Meierhenrich, G. M. Munoz Caro, J. H. Bredehoft, E. K. Jessbergerand 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. Oro, 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.

Page 181: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch15

This page intentionally left blankThis page intentionally left blank

Page 182: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

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

BALLOON-BORNE TELESCOPE SYSTEM FOR OPTICALREMOTE SENSING OF PLANETARY ATMOSPHERES

AND PLASMAS

MAKOTO TAGUCHI∗, KAZUYA YOSHIDA†, HIROKI NAKANISHI†, YASUHIROSHOJI†, 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, [email protected]

†Graduate School of Engineering, Tohoku UniversityAramaki, Aoba-ku, Sendai 980-8579, Japan

‡Graduate School of Science, Tohoku UniversityAramaki, Aoba-ku, Sendai 980-8578, Japan

This paper reports on the ongoing development of a balloon-borne telescopesystem for remote sensing of planetary atmospheres and plasmas. In thissystem, a Schmidt–Cassegrain telescope with a 300-mm clear aperture ismounted on a gondola whose attitude is controlled by control moment gyros,an active decoupling motor, and a Sun sensor. The gondola can float in thestratosphere for periods in excess of 1week. A pointing stability of 10 arcsec/swill be achieved via the cooperative operation of the following three-stagepointing devices: a gondola-attitude control system, two-axis telescope gimbalsfor 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 upperatmosphere will be derived from the tracking of cloud patterns observed inthe ultraviolet and near-infrared regions. An experiment designed to test thesystem performance is scheduled to take place in Japan during June 2007, anda long-duration flight in the Arctic is scheduled for 2008.

1. Introduction

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

169

Page 183: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

170 M. Taguchi et al.

the polar stratosphere is able to continuously monitor planets for periodsin excess of 24h. The thin, clear, and stable air of the stratosphere makes itpossible 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, largeballoon-borne telescope systems (Stratoscope I and II) were launched toundertake astronomical observations from the stratosphere.1 A balloon-borne infrared telescope was developed by Shibai et al.2 for far-infraredspectroscopy of galactic nebula and Milky Way. These types of balloon-borne telescope systems are also less expensive than large-scale ground-based telescopes or direct planetary probe missions.

We are currently developing such a balloon-borne telescope system forthe remote sensing of planetary atmospheres and plasmas. Our systemis not as large and heavy as Stratoscope II, but it is capable of along-duration flight. The telescope has a clear aperture of 300mm, andplanetary video images are taken by CCD cameras. The diffraction-limited spatial resolution is 0.34 and 0.75 arcsec at wavelengths of 400 and900 nm, respectively. Thus, a pointing stability of better than 0.34 arcsecper 33ms, which is exposure time of a video frame, or 10 arcsec/sis required. The gondola is suspended by a large plastic balloon of100,000m3 class at an altitude of 32 km above the ground. The balloon-borne telescope will be launched in Kiruna, Sweden, and it will take5–7 days for a transatlantic flight to North America or 2 weeks for acircumpolar flight back to Scandinavia. For over-horizon telemetry, anIridium-based low-speed satellite communication link will be used forcontrol and data downlink during the duration of the flight. As thegondola will always be within sunlight, panels of solar cells will be usedto ensure the continuous supply of power required to keep the systemalive.

Observation targets include varied phenomena on almost all of theplanets, e.g., the sodium tail of Mercury; lightning, airglow, and aurorain the atmospheres of Venus, Jupiter, and Saturn; escaping atmospheres ofthe Earth-type planets; and satellite-induced luminous events in Jupiter’satmosphere. The first target of our project is the global dynamics of theVenusian atmosphere, which is to be investigated by detecting cloud motionin ultraviolet and near-infrared imagery. The first experimental test of thesystem was planned for June 2006, at the Sanriku Balloon Center (SBC),Japan; however, this experiment has been postponed until 2007 because ofdelays in system development. This project also represents a step toward

Page 184: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

Balloon-Borne Telescope System for Optical Remote Sensing of Planets 171

the development of a planetary satellite telescope; an outline of this secondproject 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-bornetelescope system began in 2003, and a flight model has been underconstruction since 2005. The important specifications of the system arelisted in Table 1, and a block diagram of the system is shown in Fig. 1. Thegondola structure consists of aluminum space frames (Fig. 2). A decouplingmechanism and a pair of control moment gyros (CMGs; Fig. 3) are the keymechanisms involved in attitude control [3]; these are mounted on top of thegondola. Heavy components such as the CMGs, pressurized cell, battery,and telescope are aligned along the vertical axis of the gondola to minimizethe moment of inertia.

The decoupling mechanism isolates the gondola from the balloon, whichcan generate a large twisting moment, and transfers the excess angularmomentum of the CMGs to the balloon. The CMGs function as a torquerin the control of azimuthal attitude. Each CMG has a pair of gyros witha large angular momentum mounted on a gimbal mount; this changes theangular momentum by rotating the axis of the gyro. Compared with a

Table 1. Specifications of the balloon-borne telescope.

Telescope Type Schmidt–CassegrainClear Aperture 300 mmF-number 10

Mount Two-axis gimbalsWavelengths 400 and 900 nmSpatial resolution 0.34 and 0.75 arcsecDetector CCD video camerasAttitude control CMGs and active decoupling motorAttitude sensor Sun sensorSize 1m × 1m × 2.2m (H)Weight 290 kgMoment of inertia 50 kg m2

Power consumption 170 W (Max. 260 W)Up-link Serial command: 1 ch

Switch: 2 chDown-link Serial data: 1 ch

Analog video: 2 ch

Page 185: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May

29,2007

6:5

7sp

i-b486

AO

GS

2006

-P

SVolu

me

9in

x6in

ch16

172

M.Tagu

chiet

al.

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

Page 186: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

Balloon-Borne Telescope System for Optical Remote Sensing of Planets 173

Telescope

Battery

PressurizedCell

CMGs

Decoupling Mechanism

SolarCell

Panel

Sun Sensor

Fig. 2. Drawing of the gondola part of the balloon-borne telescope system.

reaction wheel, a CMG can generate large torque over a short responsetime with a small mass.

The azimuthal attitude of the gondola is stabilized at a constant Sunazimuthal angle using a Sun sensor, such that a solar cell panel facesthe Sun. As the CMGs comprise a pair of flywheels that tilt in oppositedirections, they generate zero torque around the horizontal axes.

A Schmidt–Cassegrain telescope is installed at the bottom of thegondola in the shadow of the solar cell panel where it is maintained atnear-constant thermal conditions. A pressurized cell (1 atm) contains PCs,a bus bridge, a high voltage power supply for a piezo-electrically controlledmirror, and digital video recorders. Exposure of these components to thethin air of the stratosphere would probably cause electrical discharge orover-heating related to a reduction in cooling efficiency. The pressurized

Page 187: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

174 M. Taguchi et al.

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

cells also protect electrical equipment from seawater upon landing in theocean. The gondola is surrounded by polystyrene foam that acts variouslyas thermal insulation, a shock-absorber when the gondola falls onto theocean, and a float.

The azimuthal angle is detected by a Sun sensor or a geomagneticaspectmeter (GA). In the mid-latitudes, a GA sensor is effective indetermining the gondola attitude; however, in the polar regions a Sun sensoris more reliable because aurora-induced currents make the GA unreliable.A PC processes sensor output that is used to control DC motors in thedecoupling mechanism and CMGs to an accuracy in azimuthal attitude ofabout 0.1˚.

The performance of the attitude control system has been tested in alaboratory 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 brightspot on a wall produced by a laser beam. After some adjustment of thefeedback circuit, the test confirmed that the system is able to stabilize thegondola attitude to the required accuracy within a minute; however, the testconfiguration differs from that in actual experiments in terms of externalperturbations and the length of the hanging wire. An outdoor experimentwith a hanging wire of the same length as that used in flight configurationis planned for the near future to confirm the system performance undermore realistic conditions.

Page 188: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

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 testspace.

2.2. Optical system and pointing

The two-axis gimbals mount of the telescope is controlled by the same PCthat controls the DC motors, guiding an object within the field-of-view ofa guide telescope. The field-of-view of the telescope covers elevation anglesfrom 0 to 70 without interference of the gondola frame or the balloon.The azimuthal solar angle of the object of interest should be greater than25 to prevent exposure of the telescope to direct solar light.

The telescope has a diameter of 300mm and focal ratio of F/10. A focalextender lens is used to expand the object image by a factor of 5. Afterpassing through the extender lens, the optical path is divided into threepaths that reach three focal planes with different colors: the first haswavelengths of less than 450nm, the second has wavelengths of 550–630nm,and the third has wavelengths greater than 750 nm. The first and third

Page 189: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

176 M. Taguchi et al.

Fig. 5. Results of an indoor test of gondola attitude control by the CMGs and activedecoupling 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 usedfor detecting tracking errors. Tracking errors that are beyond the ability ofthe gondola attitude and gimbal mount control are detected by a position-sensitive photomultiplier tube inserted at the focal plane and correctedby the tip/tilt mirror installed in the optical system. This mirror correctstracking errors of angle displacement of ±200 arcsec with a resolution of0.02 arcsec and frequency up to 100Hz. The focus can be adjusted by amotor-driven mechanism that is not automated but moves back and forthmanually as commanded.

The tip/tilt mirror mount and photomultiplier tube are kept inpressurized cells to avoid electrical discharge. A second PC controls the

Page 190: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

Balloon-Borne Telescope System for Optical Remote Sensing of Planets 177

tip/tilt mirror, high-voltage power supply to the photomultiplier tube, andtelescope focus.

2.3. Telemetry and command

Video signals from the CCD cameras are transmitted by analog modulationtelemetries to the ground for real-time monitoring, as well as being recordedby onboard digital video recorders that are recovered after landing in theocean. Commands are up-loaded and status and house-keeping data aredown-loaded by the first PC via PCM code telemetry at speeds of 2400bps.Commands and status data for the second PC are transferred by the firstPC. Two additional analog switch commands control the power switches ofthe PCs.

The video cameras currently generate analog video signals; however,by the time of the first experiment planned for 2007, the cameras will bereplaced by digital video cameras whose output signals are obtained andstored by the PC.

2.4. Weight and power

Restrictions in terms of weight and power consumption are not as severe asthose for a satellite sub-system; however, the speed and accuracy of attitudecontrol, sizing of the balloon, and float capacity all depend on the gondolaweight. The gondola should therefore be as light as possible. Estimates ofthe current weight of the system are provided in Table 2.

The power consumption of the sub-components is listed in Table 3. Thesolar cell panel and battery sizing have been determined to ensure that theysupply sufficient power. The solar cell panel fits into one of the side areas ofthe gondola and generates 170W at a solar incident angle of 45. Power is

Table 2. Weight budget (kg).

Gondola structure 60CMG and decoupling motor 35Telescope 57Solar cell panel 24Battery 17Pressurized cell 56Electronics 39

Total 288

Page 191: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

178 M. Taguchi et al.

Table 3. Power budget (W).

SupplySolar cell panel 240 at the rate of 0NiH battery 170 at the rate of 45

24 (V) 50 (Ah)

Consumption Nominal Maximum

PC1 23.5 47.0PC2 20.0 40.0CMGs 34.3 68.5Decoupling motor 9.3 18.5F/T sensor 3.2 3.2Tip/tilt mirror mount 20.0 20.0HV 1.5 1.5Preamplifier 1.0 1.0CCD cameras 4.8 4.8Telescope 10.0 10.0Heater 20.0 20.0

Total 147.6 234.5

supplied by NiH batteries at times when the solar cell panel generates littlepower, such as during launching and ascent through the cloudy troposphere.The NiH batteries are designed to provide the minimum required power fora 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-ballastsystem, although for the experiment at SBC the level-flight altitude is subjectto change depending on wind speed and direction in the atmospheric layersin which the gondola moves away from and toward the launching site. Thegondola can also be floated at a fixed altitude for as long as several weekswhen launched in the polar regions during suitable wind condition. Theflight duration depends on ballast weight and requirements for accuracy inaltitude stabilization. Observations will only be undertaken during daytime.Planetary disks can be observed with high contrast against background skybrightness, as the brightness of the sky due to atmospheric scattering is just1% of that on the ground. Following completion of the observations, thegondola will land on the ground or the sea surface. In either case, recoveryof the gondola is necessary. Most of the sub-components can then be reusedin the next experiment.

Page 192: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

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 firstexperimental flight is scheduled for June 2007 at SBC. After confirming theperformance of the system in a test experiment, it will be refurbished andplaced into full-scale operation in the polar regions. In terms of locationand facilities, Kiruna is the most ideal launching site in the Arctic. A long-duration 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 theirhelpful 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 EngineeringLaboratory, Japan, No. 135 (1986).

Page 193: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch16

This page intentionally left blankThis page intentionally left blank

Page 194: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

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

THE STRATEGIC PLAN FOR THE INTEGRATEDSCIENCES AND THE DEVELOPMENT STATUS OF

JAPANESE LUNAR EXPLORERS: SELENEAND Lunar-A

TAKAHIRO IWATA∗, SATOSHI TANAKA and MANABU KATO

Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510, Japan

[email protected]

SHO SASAKI

National Astronomical Observatory2-12 Hoshigaoka, Mizusawa, Ohshu, Iwate 023-0861, Japan

NORIYUKI NAMIKI

Department of Earth Planet Science, Kyushu University6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan

A new era of lunar explorations is coming by two Japanese missions to theMoon: SELENE and Lunar-A. SELENE will execute the global mappings ofthe Moon, make technical demonstrations, and acquire the lunar data for futureexplorations. Fifteen mission instruments on SELENE will observe chemicalelements, mineralogical distributions, surface structures, surface environments,gravity fields, and images for outreaches. They will provide wide knowledgeof phenomena on the Moon to elucidate its origin and evolution, and alsoyield information to comprehend the interplanetary space of the solar system.SELENE is at the stage of the satellite integration and environment tests forall 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 surfaceto elucidate structures and compositions of the lunar interior with seismologicaland heat-flow data. The final confirmation for the penetrator system is ongoingin 2006.

We have examined a strategic plan for the integrated sciences ofJapanese lunar exploration projects, and designed four stages for the definiteaccomplishment 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

Page 195: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

182 T. Iwata et al.

1. Introduction

The Moon is a celestial body which has an anomalously large mass andsize relative to its mother planet, the Earth. Its surface and interiorstructure cannot, therefore, be explicated without significant effects fromthe coexistence with the Earth. Hitherto lunar explorations from Apolloto Lunar Prospector have suggested the origin by a giant impact and theevolution with a magma ocean. These hypotheses seem to be, however, stillless exhaustive because the diversity of instruments and the coverage ofobservations on each exploration were insufficient to resolve those issues.Therefore, global mappings with multi-instruments are expected to verifylunar origin and evolution. The Moon–Earth system is, on the other hand,an important probe to understand the physics of the interplanetary spacein our solar system. The Earth has so strong magnetosphere that there areinteractions with the solar wind. It is helpful to observe ionic and magneticphenomena at the lunar orbit to comprehend the physics of the ionizationactivities around the Earth. Moreover, the Moon is recently watched withkeen interests as an indispensable base for the manned explorations ofour solar system. It is, therefore, expected that data of lunar elements,materials, and environments will produce beneficial knowledge for futurelunar utilizations.

Under these backgrounds, Japan has started two mission projectsfor lunar scientific explorations: SELENE and Lunar-A. SELENE is alunar explorer which will execute the global mappings of the Moon,make technical demonstrations, and acquire the lunar data for futureexplorations.1 Using 15 mission instruments, we will shed light on to figureout the origin and evolution of the Moon. Lunar-A is a lunar probe whichprovides two penetrators2 into the lunar surface to elucidate structures andcompositions of the lunar interior with seismological and heat-flow data. Inthis paper, we show the status and the strategy of the integrated sciencesof these missions.

2. Mission Outline and Status

2.1. SELENE

SELENE is the first Japanese large explorer to the Moon which will belaunched by the H-IIA launch vehicle. The total mass of the spacecraft justseparated from the rocket measures about 2.9 metric tons. SELENE MainOrbiter will be injected into a lunar orbit after the lunar-transfer phasingorbit with two revolutions. It will separate two small sub-satellites: the

Page 196: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

Integrated Sciences and Development Status of Japanese Lunar Explorers 183

Relay Satellite (Rstar) into the elliptical orbit of 100–2400km in altitudesand the VLBI Radio Satellite (Vstar) into that of 100–800km. The MainOrbiter will then be maneuvered into the circular orbit of 100 km to makeobservation for 1 year. Rstar and Vstar are spin-stabilized satellites whichhave 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 andcharacteristics are shown in Table 2. Among them, components of the RelaySatellite Transponder (RSAT) are installed on Rstar and Main Orbiter,and those of the Differential VLBI Radio Sources (VRAD) are on Rstarand Vstar. The Radio Science (RS) does not have its own component bututilize signals of VRAD. All other instruments are equipped with MainOrbiter. A cross dipole antenna of the Lunar Radar Sounder (LRS) and aboom of the Lunar Magnetometer (LMAG) are displayed in Fig. 1.

SELENE is at the stage of the satellite integration and environmenttests for all the systems in 2006. It will be transported to the TanegashimaSpace Center in the spring and is scheduled to be launched in the summerin 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

Lunar-A

SELENE Rstar/Vstar

SELENE Main Orbiter

penetrator

LMAG boom

LRS antenna

1m

50cm

2m

Lunar-A

SELENE Rstar/Vstar

SELENE Main Orbiter

penetrator

LMAG boom

LRS antenna

1m

50cm

2m

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

Page 197: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

184 T. Iwata et al.

Table 1. Characteristics of SELENE and Lunar-A.

SELENEOrbits (altitude)

Main Orbiter 100 ± 30 [km], i = 90Rstar 100 to 2400 [km], i = 90; initial orbitVstar 100 to 800 [km], i = 90; initial orbit

MassMain Orbiter 2885 [kg]; wetRstar 45 [kg]Vstar 45 [kg]

Launch vehicle H-IIA

Lunar-AOrbit (altitude)

Spacecraft 200 ± 80 [km], i = 20; orbit for data relayPositionPenetrators 1) Apollo XII site

2) antipode of (1)

MassSpace Craft 540 [kg]; wetPenetrator 45 [kg]

Launch vehicle M-V

a penetrator, and Table 1 presents its basic characteristics. The massof the spacecraft and the penetrator is 540 and 45 kg, respectively. Thespacecraft has been designed to be launched by the M-V launch vehicle,and injected into a lunar orbit after swing by the Moon. Then, penetratorswill be separated at the perigee on the lunar elliptical orbit of 45–200km inaltitudes and decelerated to free-fall at 25 km. Each penetrator is controlledto come to a halt at 1–3m below the lunar surface at the candidate area ofApollo XII site and its antipode. Each of them equips with two seismometersand thermal sensors (Table 2), which will make observations for 1 year. Datawill be relayed via the spacecraft on the circular orbit of 200km.

The final confirmations for the penetrator system are ongoing in 2006.By the accomplishment of technologies for penetrators, it is expected to bewidely 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

Page 198: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

Integrated Sciences and Development Status of Japanese Lunar Explorers 185

Table 2. Mission instruments of SELENE and Lunar-A.

Instruments Main purpose or characteristics

SELENE

Chemical elementsX-ray Spectrometer (XRS) Global mapping of Al, Si, Mg, Fe distributionsGamma-ray Spectrometer (GRS) Global mapping of U, Th, K, major elements

distributionsMineralogy

Multi-band Imager (MI) UV–VIS–NIR CCD and InGaAs imagerSpectral Profiler (SP) Continuous spectral profile

Surface structureTerrain Camera (TC) High-resolution stereo cameraLunar Radar Sounder (LRS) Mapping of subsurface structures using active

soundingLaser Altimeter (LALT) Nd:YAG laser altimeter

Gravity fieldDifferential VLBI RadioSource (VRAD)

Differential VLBI observations from groundstations

Relay Satellite Transponder (RSAT) Far-side gravimetry using four-way range ratemeasurement via Relay Satellite

Surface environmentLunar Magnetometer (LMAG) Magnetic field measurements using flux-gate

type magnetometerCharged Particle Spectrometer(CPS)

Measurements of high-energy particles

Plasma Imager (PACE) Charged particle energy, angle, andcomposition measurements

Radio Science (RS) Detection of the tenuous lunar ionosphereusing S- and X-band carriers

Plasma Imager (UPI) Observations of terrestrial plasmasphere fromlunar orbit

ImagingHigh Definition Television Camera(HDTV)

Imaging of the Moon and the Earth foroutreach

Lunar-ASeismometer Seismometry for lunar interiorThermal sensor Heat-flow measurements for lunar interior

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

Page 199: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

186 T. Iwata et al.

and the Laser Altimeter (LALT), and the gravity field maps by VRADand RSAT. Surface environments of the Moon are observed by LMAG, theCharged Particle Spectrometer (CPS), the Plasma Analyzer (PACE), RS,and the Plasma Imager (UPI). Images toward the Earth and the Moon arealso taken for popularizations by the High Definition Television Camera(HDTV). Instruments in the Lunar-A penetrators are for seismometry andheat-flow experiments, which resolve the physical parameters of the lunarinterior.

Figure 2 shows the scenarios from the individual scientific theme ofeach instrument to the integrated sciences, and finally to the scientificgoal to elucidate the origin and the evolution of the Moon and tocomprehend our solar system. Among them, the scenarios for the integratedsciences of elemental abundances, mineral compositions, geologic structures,gravity anomalies, and magnetic anomalies obtained by the instruments onSELENE are more characteristic and intricate than those of other lunarmissions. We, therefore, examined a strategy for them as described in thenext section.

Elemental Abundance(XRS, GRS)

Mineralogical Composition(MI, SP)

Geological Feature(LRS, TC, LALT)

Global Gravity Field(RSAT, VRAD)

Electromagnetic & ParticleEnvironment (CPS,

PACE, LMAG, RS, UPI)

Differentiation inMagma Ocean

Originof the Moon

Evolution of the Moon

Origin of Lunar Magnetic Field

Lunar ChemicalConstituents

Lunar InteriorStructure

Dichotomy of Near-and Far-side

Lunar Tectonics

Interior Survey(Seismometer,

Thermal Sensor)

Lunar &TerrestrialEnvironment

SELENE

Lunar-AScience of

the Solar System

Elemental Abundance(XRS, GRS)

Mineralogical Composition(MI, SP)

Geological Feature(LRS, TC, LALT)

Global Gravity Field(RSAT, VRAD)

Electromagnetic & ParticleEnvironment (CPS,

PACE, LMAG, RS, UPI)

Differentiation inMagma Ocean

Originof the Moon

Evolution of the Moon

Origin of Lunar Magnetic Field

Lunar ChemicalConstituents

Lunar InteriorStructure

Dichotomy of Near-and Far-side

Lunar Tectonics

Interior Survey(Seismometer,

Thermal Sensor)

Lunar &TerrestrialEnvironment

SELENE

Lunar-AScience of

the Solar System

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

Page 200: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

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 bySELENE mission instruments. It consists of four stages as:

1. Drawing two-dimensional maps to integrate various geologic units intoa coherent map.

2. Drawing three-dimensional subsurface structures beneath maria andhighlands.

3. Joint studies of special topics such as mare tectonics and crustal for-mations.

4. Joint studies of advanced topics such as dichotomies and bulk com-positions.

For the first step, we will make two-dimensional maps as results ofgeological interpretations by integrating various geologic units on the basisof elemental abundances (from XRS and GRS), mineralogy (MI/SP),morphologic and topographic boundaries (LRS, LALT, and TC) at thesurface and subsurface into a coherent map. Several lunar geologicmaps have been published, however, there are no consensus regardingidentifications and classifications of geologic units among these maps. Thissuggests that surface materials have been scattered by repeated impacts.We will, therefore, produce a coherent map to determine topographicboundaries, such as lava front in mare Imbria, by combining various data.

XRS GRS MI/SP LRS LALTRSAT/VRAD

LMAGTC

Elementalabundances

Mineralcomposition

TopographyGeologic structures

Gravityanomaly

Magneticanomaly

(2) Drawing three-dimensional subsurface structures

(3) Joint study of special topics

(4) Joint study of advanced topics

(1) Drawing two-dimensional maps

XRS GRS MI/SP LRS LALTRSAT/VRAD

LMAGTCXRS GRS MI/SP LRS LALTRSAT/VRAD

LMAGTC

Elementalabundances

Mineralcomposition

TopographyGeologic structures

Gravityanomaly

Magneticanomaly

Elementalabundances

Mineralcomposition

TopographyGeologic structures

Gravityanomaly

Magneticanomaly

(2) Drawing three-dimensional subsurface structures(2) Drawing three-dimensional subsurface structures

(3) Joint study of special topics(3) Joint study of special topics

(4) Joint study of advanced topics(4) Joint study of advanced topics

(1) Drawing two-dimensional maps(1) Drawing two-dimensional maps

Fig. 3. Four stages in the integrated sciences of the Moon.

Page 201: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

188 T. Iwata et al.

For the second step, we will draw three-dimensional maps to establishsubsurface structures beneath maria and highlands. The stratigraphicdistributions of basaltic lava flows in the lunar maria will be observedby LRS, and examined by MI/SP. We will determine the mineralogicalstructures of the upper crust, the lower crust, and the upper mantle from theMI/SP spectrum. The thicknesses of layers will be estimated quantitativelyby the gravity data of RSAT/VRAD, and the estimated structures will becompared with the distributions of the electric conductivity by LMAG.

For the third step, we plan to study the common targets in thecooperation of several instruments on the basis of two-dimensional andthree-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 crustwhich 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 forthe exercises and calibrations of the first geologic mapping, and then forthe investigations of mare tectonics. We will also concentrate on examiningthe North and South Pole regions for the water-ice and terrain surveys, andOrientale Basin as an archetype of multi-ring craters.

At the final step, scientific achievements up to the third step willpromote further investigations regarding the origin and evolution of theMoon. Bulk compositions, dichotomies, evolution of magma oceans, andthe lunar thermal history are targets in this stage.

4. Summary

The mission outlines and the status of Japanese lunar explorers: SELENEand Lunar-A, are summarized as follows:

SELENE has 15 instruments which observe chemical elements, mine-ralogical distributions, surface structures, surface environments, gravityfields, and images for outreaches. The satellite integration and environmenttests for the systems have been started in 2006. SELENE is scheduled tobe launched in the summer in 2007.

Lunar-A provides two penetrators into the lunar surface to elucidatethe structures and compositions of the lunar interior with seismological and

Page 202: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

Integrated Sciences and Development Status of Japanese Lunar Explorers 189

heat-flow data. The final confirmations for the penetrator system have beenongoing.

The scenarios for the integrated sciences of the Moon obtained by theinstruments on SELENE are more characteristic and intricate than thoseof other lunar missions. We have, therefore, designed four stages for thedefinite accomplishment of the integrated sciences of the Moon by: (1)drawing two-dimensional maps, (2) drawing three-dimensional subsurfacestructures, (3) joint studies of special topics, and (4) those of advancedtopics.

Acknowledgments

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

References

1. M. Kato, Y. Takizawa, S. Sasaki and SELENE Project Team, Lunar andPlanetary 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.

Page 203: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch17

This page intentionally left blankThis page intentionally left blank

Page 204: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Advances in GeosciencesVol. 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 AstronomyNorthern Arizona University, Flagstaff, Az 86011, USA

keran.o’[email protected]

Radiation protection has evolved from pen and pencil studies using tables ofcross sections and of mathematical function to large and complex codes writtenand maintained by highly skilled teams. The author’s pilgrimage through thisprocess; from his pen and pencil days while on Eniwetak Atoll in 1956 tospherical-harmonics transport codes, the use of discrete ordinate and MonteCarlo codes to an analytical transport code for the calculation of cosmic-ray 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 radioactivefallout, beta-ray transport, accelerator shielding, hospital physics, cosmic-rayionization, cosmogenic isotope production, the radiation dose to air crews andspace crews, and cosmic-ray fluxes to space craft. Some examples of the resultsof these calculations will be given.

1. Eniwetak

1.1. Prologue

The events which I relate in this little tale of my participation in theOperation Redwing nuclear weapons test series happened 50years ago, in1956. Much has faded from my memory, and as happens with memory,some of what I think I remember may be distorted or combined with otherevents that took place at other times. Friends have urged me to set thisstory down before the fading and distortion of my memory progresses anyfurther.

Fortunately, a study of the effects of Operation Redwing becameavailable in 19971 and I have relied upon it for many details of time andplace.

At the time this story began, Len Solon, who was my boss at theHealth and Safety Laboratory urged me to take notes and to write up this

191

Page 205: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

192 K. O’Brien

adventure. I did not follow his advice. It was harder to do then without theresources of computers and word processors, I was much younger and lessreflective, and finally the adventure seemed to leave little room in my life totake notes at the time it was taking place. There was no privacy and almostno opportunity to be alone on Eniwetak where I was situated. When theadventure was over, I settled into my routine life without much thought ofwriting it up. A contributing factor, perhaps, was the presence in the labof so many who had visited the Pacific Proving Grounds more than once,making my adventure seem rather ordinary and indeed rather less thanspectacular by comparison with what so many others of my acquaintancehad 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 verylarge, and the passage of time has severely reduced those numbers. I thinkmy friends were right; it is time and past time to set down what I saw forothers to see.

1.2. In the lab

Except for a brief lacuna, I had been working for the Health and SafetyLaboratory since 1953. The Health and Safety Laboratory was located inNew York City, New York in the United States, at the site where LincolnCenter in Manhattan is now. I had spent a summer between my junior andsenior year at Fordham University at the lab which at that time was theHealth and Safety Division of the New York Operations Office of the U.S.Atomic Energy Commission. I had returned the year before, 1955, from adisastrous year in graduate school at Carnegie Tech where I had once againproven I was uneducable. The laboratory was gearing up for OperationRedwing where the lab’s duty was to monitor the radiation intensities oninhabited Pacific Islands in the path of weapons fallout and to estimate thefraction of the fission product yield that entered the stratosphere as opposedto the faction that remained in the troposphere. Both missions were to beaccomplished by measurements of radiation quantities on the ground andin the water by measurement of radiation intensities using aircraft-borneinstrumentation. My job would be to relate quantities of radioactivity onthe ground and in the water to radiation intensities as a function of altitudein the air above.

During a previous test in the Pacific, Operation Castle in 1954, thelaboratory had put together Project Dumbo comprising polystyrene rafts

Page 206: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 193

with instruments and telemeters which they had dropped into the Pacificprior to test shots. The hope was to fly over them and record the radiationintensity data broadcast by the telemeters. The radiation intensity was tobe related to the quantity of radioactivity collected on the known area ofthe raft. I remember having seen the rafts 2 years before.

Project Dumbo was a failure for two reasons: Most of the rafts couldnot be found again after a weapons test, and the radiation levels from theradioactivity in the surrounding ocean were very much greater than theradiation levels from whatever rafts could be found, making the broadcastdata completely useless. During Redwing, the ocean itself would be usedas the collector, and the radiation from the sea would be used to determinethe quantity of radioactivity it contained, requiring only that the depth ofthe 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 aso-called Jet-Stream DC-7. I was carrying orders which I presented to theairline 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 havenever carried anything else so clumsy and so hard to manage. I never didit again.

I remember an amusing incident on the flight. In the days before jetaircraft, cross-country flights were much longer than they are today, andwe flew first class. Meals during these long flights were frequently good,and instead of peanuts, we got macadamia nuts. During the meal on thisflight, I bit into a cherry tomato which spattered all over my shirt. A lovelyyoung stewardess came over and helped me mop up. I can remember hersaying “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 GreekRevival 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 inSan Francisco.

It was late and I was tired. The Greyhound bus to Travis Air ForceBase passed though a number of northern California towns. I can rememberthe bus driver calling out “San Rafael (San ra fell’),” and “Vallejo (Va lay’ho),” but I was dozing most of the way.

Page 207: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

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 interactedwith the military, we had to be given a fictitious military rank so that theycould 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 myreal 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 inwhat looked like a big brown room on camp chairs, all facing aft. Windowswere small and not at all synchronized with the seat rows. I could not lookout. I remember without conviction that it was a 12-h flight.

We landed at Hickam Air Force Base on Oahu. Hickam shares facilitieswith the International Airport, and for some purposes, the facilities areindistinguishable. Once again, I was assigned a room in the BOQ.

It was several days before a flight was available to take me on toKwajalein 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 withboth jet and piston engines, four of one and six of the other. They weredesigned to fly at great heights and carry nuclear weapons. The sound oftheir engines was unique, an especially deep-throated rumble.

All the aircraft, after landing, would follow a jeep with a sign on theback saying “Follow Me.” The giant cargo craft creaked in time with theslow swaying of their wings as they made their way along the concreterunways following the lead jeep.

I could have taken a cab to Honolulu, but I didn’t. I was unfamiliarwith the city and had no idea what I could do there. I did not know ofthe existence of the Bishop Museum, for instance. That might have madea difference to me. I was, as I have said, a very young 24, and I am surethat a great deal to do with it.

The base had a very fine restaurant, and I remember one very pleasantdinner there with a young officer. It was my first taste of mahi-mahi, andby no means my last.

After several days, a flight was made available to me to Eniwetak viaKwajalein. I can remember standing in line with a lot of other young menwaiting to board the flight, when Norris Bradbury, the director of LosAlamos, who had been on the aircraft walked by us. He was a striking-looking man and had iron in his face. I remember he was wearing a businesssuit which looked very out-of-place in that world of military uniforms andthe casual khakis of the civilian scientists and work force.

Page 208: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 195

The flight to Kwaj was about as long as the flight to Hawai’i and justas tedious. We sat in the big brown room, facing aft, as before. Eventuallywe landed at Kwajalein. It was hot and humid. We went into a large messhall and were fed. I remember Kwaj as rocky, bare and brown. It is an atolland we landed on one of the islets, probably Meek. Then it was back intothe 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 someof the others the five miles to Parry Island were the scientific staff wereheadquartered., 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 donot know, but that is how it was known to all of us. Its Polynesian namewas Medren.

Eniwetak Atoll is circular (cf. Fig. 1, from Noshkin and Robinson1)about 25 miles in diameter. Eniwetak Island was at the southern end ofthe 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 TaskForce 7 (JTF-7) Fleet could pass through and anchor in the center of thelagoon. Eniwetak was one of the larger islands of the atoll and was theadministrative headquarters. It had a good-sized airstrip on which we hadlanded.

All of the islands had both English names as well as the originalPolynesian names in case one of the participants in the operation hadtrouble remembering or saying the local name; the upshot was that wehad to remember both names. An atoll is a reef sitting like a crown on adrowned seamount. The small islands were upward extensions of the reefthat 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 otherwords, one stood on the edge of a mountain, atop a cliff which extendeddown 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 hadbeen relocated to Meek Island of the Kwajalein Atoll to make way forthe weapons tests.

Page 209: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

196 K. O’Brien

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

Parry was on the south–southeast of the atoll. It was about a mile anda half long, in the north–south direction, and about half a mile wide inthe east–west direction. The U.S. Army Corp of Engineers had made ithabitable 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 leftintact for “R&R” and see how it must have looked prior to the coming ofthe Corp of Engineers. Japtan was covered with a bamboo forest. About10 years ago, when I stood in a stand of Arizona’s native bamboo in theneighborhood of the Kofa mountains, the scene was brought back to mevividly.

I was assigned a bunk in a long low aluminum building. The bunkswere stacked three high. The building had no glass in the windows, whichwere 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 roommateabout high-fidelity audio units. I had just put together a 20-W Scott audio

Page 210: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

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 wasnew and the pride of my system. It couldn’t have been jet lag. I had madethe whole trip in piston aircraft.

We were wakened in the early dawn. Still confused I struggled outsidewith everyone else and discovered that we were being awakened to view anuclear weapons test. It was known as Zuni and was to be detonated onBikini, 180 nautical miles to the east. A tower stood behind us, a greenlight 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 ofthe irregular coral pavement.

The green light meant that the test was still on. It if were replaced bya 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 about20min. So it did, but to my surprise there were several booms, not justone. There must have been some echoing between sea and sky, or perhapsthe acoustic shock transmitted to the earth had been re-emitted at otherlocations, 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 buildingwhere 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. FredWilson from the Instrumentation Division had the bunk above me. Fred andI got to be good friends during this adventure. However, when we got backto the lab, our paths almost never crossed. Many years later we becamerather serious unfriends. I have no idea how that happened. At the time wehad some common tastes and we were thrown together a lot.

Another man I made friends with was Gerry Hamada from the lab’sAnalytical Branch. Gerry and I both like classical music and were seriousabout science so we had things to talk about. Gerry left the lab for acommercial firm some years afterward.

The sun was quite powerful at that latitude and we were warned aboutoverexposure to the sun. The protocol recommended was to wear long pantsand long-sleeved shirts until we became accustomed to the sun. I did justthat and suffered no problems, although when I was in Hawai’i, my face got

Page 211: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

198 K. O’Brien

badly sun burnt. Eventually I wore no more than bathing trunks and rubbersandals of the Japanese type which are called, I believe, zorris. I got so usedto them I wore them a lot at home too. But they afforded no protection tothe 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 andwet footgear were no problem. They would quickly dry.

Later, much later, I was surprised that hats of some sort were notrecommended. 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 towear ball caps. A ball cap might have saved my face from sunburn.

A young red-headed lad from the Instrumentation Division, JerryCoughlin by name, chose to ignore all the cautions and got very badlysun burnt. He had to be sent back to New York.

The day-to-day weather was uniform. It rained every afternoon at about3 o’clock. The weather was warm and the constant sea-breeze made it quitecomfortable, especially dressed lightly in bathing shorts and sandals. Itmeant nothing to get caught out in the rain. The rain was warm and thesun after the rain dried everything quickly.

During my stay, I saw many nuclear explosions, most of them in thetens of kiloton range and detonated locally in the atoll. Some of the peopleon Eniwetak had become so blase that they didn’t go out to watch the shots.I was enough impressed with the importance of the events I was witnessingthat I went out to watch each one. Of all of them, I remember only twowith any distinctness. One of them was Zuni, the shot that was detonatedthe morning after I arrived on Eniwetak, and the other was Pequod, which,if I remember correctly, was detonated sometime after the midpoint of mystay on Eniwetak.

I had spent the previous evening at the Officers’ Club. There were onlytwo forms of entertainment on the island. One was the movies and theother, the Officers’ Club. I had become a bit too fond of creme de mentheand had drunk enough of it the night before so that I had a head in themorning.

Pequod was a barge shot, a thermonuclear device to be detonated overthe site of Elugelab, a small island at the north of the atoll that had beenessentially destroyed by Mike Shot during Operation Ivy in 1952. It wasabout 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.

Page 212: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 199

We walked down to the lagoon shore a little before 6 o’clock in themorning and sat on the sand. Very dark glasses were handed out so thatwe could look straight at the fireball. I hazarded a brief glance at the sunthrough them and it looked like a dull purple disk.

I noticed that the water tower was heavily braced and tied down withwire, in case the shock might topple it.

The countdown began about 15min before the shot. “This is Manhunt,”it began. “It is 15min before zero time.” “Manhunt” warned against lookingdirectly 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 5mins, then at 1-minintervals, 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 couldfeel the radiated heat on my face. The ground rocked silently. I wondered,briefly, if perhaps that was the effect of the previous night’s creme dementhe, or the shock of the explosion.

The sound wave must have arrived about 2 min after but I don’tremember 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. Itwas too bright to look at for more than a few seconds, so I left the glasseson 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 nowreddish and streaked. It rose and darkened, leaving a gray column of debrisbehind it. Eventually, there was a huge mushroom cloud extending over ourheads, largely gray in color with some pink highlights.

After watching this for awhile, began to think of leaving and going backto work, when suddenly the water at the edge of the lagoon retreated witha hiss, about 3 or 4 ft. Then it surged up the beach a few feet, and retreatedagain, doing this about seven times.

In order to determine the proper time for a shot, the weather peoplefrom 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 rathersteady, and, I suppose, comparatively easy to predict. I do not knowwhat sort of synoptic data were gathered, but locally, the Los Alamospeople launched small weather balloons and tracked them as they aroseto construct what was called a hodogram which gave a picture of wind

Page 213: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

200 K. O’Brien

velocity and direction aloft. They were looking for a pattern of wind shearthat would cause the “stem” of the mushroom to fall one way and missinhabited areas, and the cap to go another to facilitate diagnostics.

The cap, or some part of it, would rise into the stratosphere anddrift around the world. The radioactive particles that made up the cloudhad a half residence time of 10 months, and were slowly falling from thestratosphere though the tropopause down to the surface of the earth. Theregions that had the greatest amount of fallout on their surface were thoseplaces, 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 centerof the island. We all wore security badges hung about our necks on a chainthat indicated whether or not you were cleared to the level necessary toenter the compound. A guard would take the badge from you, hold it inhis hand, glance from you to the picture on the badge and back again andthen return it to you.

I haven’t the faintest idea what information it was that they felt theyneeded to protect. In those days, the tendency was to overclassify.

In the compound I worked with a William Mills. Willy Milly, as wecalled 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 takethe radiation levels as measured by aircraft over the islands of the MarshallIsland chain and post them on a big wall map, updating them as new datacame in.

I met Willy Milly again at a Health Physics Society Meeting some yearsago. I remembered him and, to my surprise, he remembered me. His hair isno 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 radiationintensity over the radioactive ocean and land. Fortunately, all the necessaryreferences were either there, or I had brought them with me.

Then I discovered that in a military outfit stationed on Eniwetak forRedwing, a company of infantry, if I remember aright, was a soldier whohad a PhD, a “cosmic-ray physicist” who had been drafted. The story wasthat “W” as I will call him did not get along well with his mates — thishardly surprised me. W, I was told, had been assigned the same task ashad I.

I went to him and suggested we work together. It would shorten ourlabor and make error less likely. I was coolly dismissed. “When I have myresults,” he said, “I will show them to you.” I can only suppose he was filled

Page 214: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

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 copyof Goldstein and Wilkins calculations of the scatter component of gammaradiation in the format known as “buildup factors,” a table of Gradstein’sgamma-ray cross section, the CRC table of integrals and a copy of Jahnkeand 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 thesevenerable references.

Today, one would solve the Boltzmann equation with a code, “off theshelf,” and get the answers much more quickly and more accurately than Igot them.

I sat in an air-conditioned trailer next to the building we sharedwith the University of Washington Marine Biology Laboratory, fit build-upfactors 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 ofenergies and heights over the sea, and for various angles of acceptance, sinceit was planned to use conical collimators to control the “angle of vision”of the detectors. I then repeated this operation for radioactive materialon land surfaces. The results seemed reasonable and I had concocted some“theorems” based on what I called “homogeneous medium theory” to checkthem at the sea surface. No such results existed for the land surface, butthe formalism allowed a transformation between them so that I had a wayof checking those results as well.

In the end, my results were taken to the New York UniversityComputing Center at the Courant Institute for Mathematical Studies andwere cranked out for an immense range of altitudes, photon energies andconical collimator angles. The results appeared to agree well with themeasurements.

Word got back to me that W was not so fortunate. That informationmade me feel that I had just escaped from a serious accident “by the skinof my teeth.” It could well have been me.

During the first part of my stay, the group from our laboratory wasunder the direction of Harry LeVine, director of the InstrumentationDivision. While Harry and I didn’t get along well back at the lab (neitherbefore nor after Redwing) we seemed to hit if off well in the field. Duringhis stay, he used me as a kind of physics consultant. I would go to meetings

Page 215: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

202 K. O’Brien

with him and supply answers. I recall one time when we needed an estimateof the decay of fission products over a period of time. I tried to calculateit using the Way–Wigner relationship at a meeting on my small pocketslide rule and got it all messed up. Somewhat later we were supplied withcircular 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, withan authoritarian bent that never left him.

Under D we worked long hours and late and quickly grew to resenthis management. Perhaps we would have felt differently had he beennotably talented. And had he been talented, he might not have had thatauthoritarian temper.

Merril Eisenbud, the Laboratory Director arrived somewhat later andmatters immediately improved. Merril simply stated that “If you can’t doit in eight hours, you can’t do it in twelve.” He was aware that morale inthis remote area, absent much of the ordinary amenities of life, can quicklygo sour, and hurt productivity.

After Merril left, D tried to restore the old routine. I remember drivingoff in the six-by-six truck we had managed to acquire, with the rest of thecrew in back, with D stumbling along angrily behind, yelling at us to comeback.

Merril was the Director of the Health and Safety Laboratory, with anInternational reputation, and had recently played an important role in theexposure of the fishermen of the Fukuru Maru to fallout, but he was utterlyunpretentious. A truly great and fine man, he died a few years ago fromleukemia. One Sunday he took me out in a rowboat onto the waters of thelagoon to explain to me his philosophy on the importance of field work tothe laboratory, and regretting that my immediate boss, Len Solon, couldnot be spared to come out to Eniwetak for the experience of field work.

I remember looking down through the amazingly clear waters of thelagoon at a coral head that seemed perhaps only a dozen feet below thesurface. I dove for it from the rowboat and quickly found that the coral headwas far deeper and out of my reach, as the water pressure quickly beganto exert too much pressure on my eardrums. The water was absolutelycolorless, and were it not for the difference in refractive index between itand air would have been completely invisible.

This is not to say that Merril was not above taking advantage of someof the perquisites of rank. He managed to obtain one of those giant clamshells that are supposed to occasionally catch skin divers by the leg and

Page 216: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 203

drown them unless they could quickly hack off the caught leg with theknife they always carried with them.

The clam shells would be brought up from the bottom of the lagoonand then buried in the sand until the clam inside would have been devouredby 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 fora good mess. This was certainly the case on Parry Island. The big meal ofthe day often included steak, and the steak was good, often tenderloin. Igained a good deal of weight while on Parry. There was not, after all, muchopportunity for exercise, except swimming. Food was another importantaspect 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 onestep you were at the edge of the water, at the second step you were thighdeep. At the third step you were chest deep. At the fourth step you wereswimming.

I enjoyed swimming there, but it was not always possible. A Marineguard always stood watch in case there was any trouble, though what hewould 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 thelagoon. One day, I went down to the beach. No one was in sight, not eventhe Marine guard. I took three steps and was swimming in the beautifulwater when the guard hurried out of his little shack and called me back inan agitated voice. Apparently the helicopter pilots had warned of sharks inthe 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 theEscherichia coli count. The JTF-7 fleet was anchored in the center of thelagoon. The water, whatever the E. coli count, was always transparentlyclear.

One day, Admiral R. Ball Hanlon, the admiral in charge of JTF-7, cameto the beach to swim. Like any massive primary in this solar system, hecame with a large number of satellites. One of the satellites, a Navy captain,

Page 217: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

204 K. O’Brien

took a half step into the water. Before he could take another step, a largefish rushed up, jaws extended. The captain was able to recover dry landbefore his leg was seized.

As we shared a building with the University of Washington MarineBiology Lab, we soon heard of the account. I was outside when AdmiralLewis Strauss, then the Chair of the Atomic Energy Commission came upand asked me where the Marine Biology Lab could be found. I heard allabout the adventure from others, and watched as Ed Held and LaurenDonaldson pored over reference books trying to find something that lookedlike the monster that tried to get the commander of JTF-7 (starting withthe captain). They found nothing that matched the description. I suspectthat the combination of panic and excitement make accurate descriptiondifficult.

Occasionally we went swimming with the Marine Biology group alongthe reef. Ed Held taught me a way to equalize the pressure on myeardrums, permitting me to “dive” down to much greater depths than Icould previously. Because of the transparency of the water, and the beautyof the corals and brilliantly colored fish, it was great fun to swim and diveoff the reef.

Not all the fish were harmless. Stone fish and turkey fish had hollowspines that contained a toxin that was exquisitely painful. We were toldthat if one were to step on one, the pain would be so intense that it wouldresult in collapse and if the water were deep enough, drowning. While I wasthere, there was a medical emergency. Someone did get impaled by a turkeyfish 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 nastyinfection.

As a precaution against stone fish, turkey fish and coral cuts, we alwayswore sandals while wading out in water on the reef. How much they wouldhelp, I am not sure, but they would afford some protection.

Occasionally, I would wander out on the reef alone and look at thefish. I made an unfortunate acquaintance with a surgeon fish. I wouldoccasionally try to catch a fish with my hand. I succeeded with a surgeonfish and found out why they are called “surgeon fish.” The dorsal fin isquite sharp and I got my fingers cut for my pains. The cut was minor, but Inever again tried to catch a surgeon fish with my hands — or in any otherway.

The reef extended a good distance south of Parry, and it was rumoredthat with care, one could walk all the way to Eniwetak. It was also said to

Page 218: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 205

be dangerous if one was caught by the rising tide. I never walked very farin 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 shallowwaters. But water was their element, not mine, and they found it easy toelude me.

The sand sharks were comparatively small, about 3 or 4 ft long, withblack dorsal fins; they were called, colloquially, “black tips.” I was told theywere not totally harmless. A marine stationed at Bikini had one tear hisleg muscle off.

The laborers on Eniwetak, the men who did the manual labor, werePolynesians, recruited from other Pacific islands. They would catch longusta(the so-called “lobster tails”) and roast them. The sand crabs wereanother delicacy. However, local weapons tests had released a good bitof radioactivity into the water, and both the longusta and the sand crabswere radioactive. It was hard to discourage the laborers from catching andeating their very usual and customary food.

I remember standing on the reef with a number of these tall strongyoung men, a heap of longusta piled up ready to be roasted, and an earnestmilitary radsafe guy urging them to forego their feast. When we left, thehecatomb 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 WakeIsland, about 700 miles to the northeast. My being there served no officialpurpose, 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 ideawhat. I had made a few friends with some of the military on site. It was asmall island and one would meet the same people often. This military pilotwas 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 flownover to Eniwetak — or Fred, as it was also named; a short hop of about5miles. I was intrigued by the approach to the Eniwetak landing strip. Thepilot rose to his cruising altitude and flew almost over the landing strip andthen dove steeply down for the landing. That was the standard method ofall the military pilots out there.

We flew to Wake on a Stratoliner. It was another case of the big brownroom, but this time we were free to wander around, to look out the windowsand talk to the flight crew. After more than a month’s immersion in the life

Page 219: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

206 K. O’Brien

at Parry, Wake was a sudden shock. There were women! I had not realizedhow 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 seethe island before it was time to return to Eniwetak.

Speaking of L-19s, there was a smaller aircraft, an L-18. A story wastold that before my arrival, an L-18 tried to land on the strip at Parry, butbecause the steady sea-breeze provided too much lift, the pilot had some ofthe ground crew run alongside the aircraft and hold the wings down.

There were two other episodes that correspond to “R&R.” One of themwas a picnic on Japtan. We sailed on one of the “African Queens” fromParry. As we crossed the Deep Entrance and passed suddenly into deepwater, the water turned an intense and brilliant blue, like nothing I hadever seen before.

Japtan was covered by a forest of bamboo, and the sight of so muchgreen 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, nextto the thick steel rotating shaft that held the blades, the shaft about a footin 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 howthe helicopter pilots could see sharks in the lagoon. The encircling reef wasclearly visible around the whole atoll.

It was extremely noisy, and I was alarmed when we landed when I couldnot hear for about 5min.

1.5. The return home

There was a Post Exchange on Parry which we were, on the basis of ourfictitious ranks, permitted to use. I bought a good solid suitcase to replacethe duffel bag I had come out with. I kept the duffel bag, stuffing it in thesuitcase, 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 littleless than 2 months. I told Gerry Hamada, a native of Hawai’i, that I wouldmeet him there and do some sight-seeing with him.

We flew directly to Hickam Air Force Base without a stopover atKwajalein. I traveled with a Public Health Service employee whose namewas 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 gotunbearably cold.

Page 220: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 207

When Mahlon and I got to Hickam we were told that they did not knowwhen the next flight to the mainland would take place, so we took a roomat the Royal Hawai’ian on Waikiki Beach. We had left our destination withPAX, the information exchange, so that we could be reached when a flightbecame available. I had decided not to wait for Gerry. My feet had swollenup with athlete’s foot — a result, no doubt, of wearing shoes for the firsttime in almost 2 months, plus the heat in the aircraft resulting from thethermostat failure in the aircraft — and I was ready to go home. I left amessage with PAX for Gerry.

We went swimming and it was quite a disappointment after the clearbrilliant water of Eniwetak lagoon. The water was not as clear and Ikept bumping my knees on the coral heads. Mahlon and I had dinnertogether. When we got to our room, we found that the message light on thetelephone was lit. They had a flight for us, we were to leave early the nextmorning.

I visited Honolulu much later, in 1988, when my son was married there.The change was appalling. In 1956 there were complaints that Honolulu hadbecome commercialized and crowded. In 1988, there was no space betweenthe hotels. I remember being astonished at the change. Hotels stood inranks behind one another along the beach. Honolulu, in 1956, was a tranquilvillage by comparison.

The flight back to the mainland was on a commercial jet with a contractto the military. The airline was a small one, and probably exists no longer.I noticed that the single stewardess was older than stewardesses usuallywere in those days. I remember her face and pleasant smile with pleasureafter all these years.

It was a relief to sit on a comfortable seat, facing forward, with a realcabin crewmember to provide meat and drink.

I bade farewell to Mahlon in San Francisco and flew home on anotherDC-7.

2. Accelerator Radiation Protection

2.1. Flintlock days

Around 1954, the laboratory director received a letter from the AssistantGeneral Counsel of the Atomic Energy Commission. The Navy and the AECowned almost all of the cyclotrons in the country, and many universities hadmore than one; a situation quite different from today’s, where fundamental

Page 221: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

208 K. O’Brien

particle research is now done on a few very large machines, built at a fewsites. The AEC was concerned that there was no “in-house” capabilityto ascertain the safety of the personnel using these machines, and theresearchers appeared to have developed a significant number of cataracts.There was a reasonable fear of litigation on this account and possibly withrespect to other health issues.

Len Solon bought a radium–beryllium neutron source and apolyethylene-walled, ethylene-filled neutron proportional counter. Theneutron counter was calibrated with the radium–beryllium source. Theradium–beryllium source also produced an intense gamma-ray field whichmade its use quite difficult.

The first survey measurements were made with this instrumentand with a Juno hand-held gamma-ray ionization chamber. The resultswere then, of course, quite simple to interpret. There were neutronmeasurement and gamma-ray measurements which could be compared withthe permissible levels of National Bureau of Standards Handbook 59.

However, the neutron detector’s low-energy cutoff was 0.5MeV, so weadded more instruments. Ultimately, by 1958 we had eight, including severallong counters, a zinc-sulfide-impregnated polymethylmethacrylate (Lucite)counter, a BF3-filled, plastic-walled ionization chamber, a tissue-equivalentionization 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 easierto handle polonium–beryllium source, and it, in turn was replaced by aplutonium–beryllium source.

By 1958 we had visited 14 accelerator sites. Two things stand out in mymemory. 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 wouldbe highest, and where the thermal neutron fluxes would be highest. Theother was the number of senior researchers who assured us that there wasno neutron problem but who were looking at us through the thick “bottle-bottom” glasses that people who had had cataract surgery wore at thattime. 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 usedinstead, exploiting the Oppenheimer–Phillips reaction to get the neutroninto the target nucleus.

Page 222: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 209

Fig. 2. 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 quitecomplicated. All the neutron gave different results and had different energyresponses, so we blamed it on a “deus ex machine,” the neutron spectrum.

We invested considerable effort into developing nuclear-track emulsionspectrometry. We used 400 and 600 micron Ilford plates developed in adarkroom built for the purpose.

Recoil proton track lengths were measured with a microscope. Thelocation of each end was marked on a punched card. Pythagoras’ theoremwas used to obtain the length, and range-energy tables for emulsion

Page 223: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

210 K. O’Brien

were used to obtain the proton energies.3 Proton energies were groupedin bins and then corrected for their escape probabilities — 600micronscorresponds to a proton energy of only 0.6MeV. In order to determine theneutron spectrum from the protons spectrum, we had to solve the Fredholmequation:

P (Ep) =

∞∫0

γσn,p(En, Ep)N(En)dEn (1)

where P (Ep) is the recoil proton spectrum per MeV, γ is the hydrogen iondensity per cm3, σn,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 approachand by expanding the neutron spectrum in Hermite polynomials.4 Figure 3shows some of the results we got.

2.2.2. Bonner spectrometry

Returning from a vacation some time about 1970, I was stunned to findthat my then boss, J. E. McLaughlin (Len Solon had left the lab to takethe job of director of the New York City Office of Radiation Control) hadgutted the darkroom, disposed of the microscopes and painfully machinedscrews that we used to measure proton-recoil track lengths and had, at ablow, terminated the nuclear emulsion neutron spectrometry program.

A man woefully unsure of himself, he often went to the headquartersof the Atomic Energy Commission in Washington seeking guidance in themanagement of the Radiation Physics Division, of which he was director. Hehad gotten the notion that Washington wanted us to demonstrate progressby terminating programs, and this was his reaction. Without consultingany of us, he irreparably destroyed what was a successful and interestingprogram which we were able to use to help unravel the problems posed bythe differing responses of our neutron detectors.

We were not without resources, however. We had obtained the fivepolyethylene spheres that comprised the Bonner spectrometer.6 The sphereswere used to moderate incoming neutrons so that they could be capturedby a thermal detector at the center. With a bare thermal detector we hada six-element spectrometer. The main problem in its use was that it wasa mathematically underdetermined system. The equation for the neutron

Page 224: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 211

Fig. 3. Measured neutron spectra from a number of accelerator sites.5 AGS, AlternatingGradient 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 =

∞∫0

ϕ(E)Kν(E)dE (2)

where Bν is the counting rate of the νth sphere, φ(E) is the neutronspectrum, and Kν is the energy response of the νth sphere.

To calculate a detailed neutron spectrum using just six elements posedan interesting problem. Our first attempt was an iterative method similarto that used for the emulsion spectra.7 The second and most successfulapproach was the use of a Monte Carlo method.8

Page 225: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

212 K. O’Brien

The indeterminateness of the solution to Eq. (2) can be expressed asfollows:

Bν =

∞∫0

[ϕ(E) + ξ(E)]Kν(E)dE

∞∫0

ξ(E)Kν(E)dE = 0

Bν + εν =

∞∫0

[ϕ(E) + ξ(E)]Kν(E)dE

(3)

where εv is the counting error in the νth sphere and ξ(E) is an oscillatoryterm which appears because of the undeterminedness of the system. In theiterative method, we smoothed the solution between each iteration. In theMonte Carlo approach, these terms are randomly generated, so that bytaking the mean of several solutions, we could remove them.

Figure 4 indicates the power of this method. We took a calculatedneutron spectrum, inserted it into Eq. (2) and then unfolded the resultingset of synthesized sphere counts to get the results shown in the figure. Themethod 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 inthe periodical literature. I had read it avidly, and indeed had successfullyapplied it to a problem in β-ray transport.9 I ran into trouble with outInstrumentation Division. The Instrumentation Division personnel alwaysfelt that they did all the hard work, but the other divisions got all the gloryand they liked to make sure you understood how much you depended onthem.

In fact they could exercise an absolute veto on the work of some of themore junior staff, such as I was at the time. They made it quite plain thatthey were no longer interested in my accelerator radiation measurements,and so they, perforce, came to an end. But because I had begun to workwith transport theory, I realized that I could get the information I neededfrom basic principles. Radiation transport obeys the equation that LudwigBoltzmann wrote down in 1872:

Biϕi(r, E, Ω, t) =∑

j

Qij

Bi = Ω · ∇ + σi + di −(

∂Ei

)Si

Page 226: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 213

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

Qij =∑

j

∫4π

∞∫E

dEBσij(EB → E, Ω′ → Ω)ϕj(r, EB, Ω, t)

(4)

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 isthe particle direction t is time, σi is the collision cross section of particle-type i, di is the decay probability of radioactive particle-type i, S is the

Page 227: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

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 ofa particle of type i at a direction Ω and energy E from a collision witha particle of type j, φj is the angular flux producing secondary particlesof type i and energy EB in the scattering-down process, β is the ratio ofparticle velocity to the velocity of light, τ is the particle mean life, and c isthe velocity of light.

2.3.1. Neutron transport through accelerator shields

Expanding the fluxes and cross sections of Eq. (4) in spherical harmonics, Iwas able to write a transport code in the P3 approximation.10 Comparingit with experimental results, indicated quite satisfactory agreement(cf. Fig. 5).11

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

Page 228: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

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 StanfordLinear Accelerator and the Cambridge Electron Accelerator.10

3. Cosmic-ray Studies

At this time, my friend and colleague, R. G. “Tut” Alsmiller of the OakRidge National Laboratory was also doing accelerator shielding studies,and because of the business-orientation of ORNL, had to charge for hiswork. I did not feel it fair to compete with him, because my serviceswere free.

Earlier, during the weapons-test days, Commissioner Willard Libby ofradiocarbon fame had said, publicly that the fallout from weapons testscarried 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 ofit, 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 ofinstruments and began a long career in the study of the natural backgroundradiation in the USA. They quickly found that they needed to separateterrestrial from cosmic sources. Both their terrestrial and cosmic-ray studiesbecame increasingly sophisticated.

From large plastic bottles painted on the inside with aquadag theydeveloped argon-filled steel-walled chambers calibrated to take into accountthe differences between the medium of the chambers and the surroundingatmosphere as it would affect the passage of cosmic rays through the walland the gas.

At this time, I decided to try my hand at cosmic-ray transport. Aftera couple of false starts, I came across a paper by Tut Alsmiller’s wife, FranAlsmiller. 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 radiationtransport.12

I made a point of reading everything that either of the Alsmillers wroteand so pored over her paper and came across a simple equation for high-energy radiation transport, originally formulated by C. Passow.13 It proved

Page 229: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

216 K. O’Brien

Fig. 6. The calculated and measured ionization profile at Durham NH in 1965.16

quite successful in treating cosmic-ray propagation in the atmosphere. Itwas a “straight-ahead” code, at first, which put too many particles at highaltitudes and too few at intermediate depths. The accuracy of the code wasgreatly improved by extending it to three dimensions,14 using an approachsuggested by Elliott15 and Williams16 and by incorporating a routine fortransporting cosmic-ray nuclei.

Its accuracy is shown in Fig. 6, where the measurements of Lowderet al.17 Solar modulation was treated in the heliocentric approximation.The geomagnetic field was treated by combining Stormer’s theorem withthe calculated vertical cutoffs.13 The code, called “PLOTINUS” has alsobeen used to calculate the radiation dose to flight crew. The results, showin Fig. 6 show good agreement with the measurements carried out at PTBby 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 mechanismknown 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

Page 230: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 217

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

the high-energy transport code to the shock-accelerated spectra incident onthe earth. Local intensities in the atmosphere were obtained by comparingcalculated fluxes with the measurements of ground-level neutron monitorsdistributed over the surface of the earth and then interpolating betweenthem, using their geomagnetic latitude as the interpolation grid, at eachaltitude of interest.19 GLE-42 was highly anisotropic, as Fig. 8 clearlyshows.

Fig. 8. Solar particle effective dose rates at flight-level 350, 3 h after the onset of GLE-42.15

Page 231: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

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 heliosphereis20:

∂D

∂t+ ∇ · S +

(1

P 2

)∂

∂P(P 2〈P 〉D) = Q (5)

S = 4πP 2(CV D − K · ∇)

D.

where D is the phase-space density, S is the differential current density, P isthe rigidity, V is the velocity of the solar wind, C is the Compton–Gettingcoefficient, and K is the diffusion tensor.

The source-free, one-dimensional version of Eq. (5) is20:

V∂D

∂r+

1r2

[∂

∂r

(r2κ

∂D

∂r

)]− 1

3r2

∂r(r2V )P

∂D

∂P= 0 (6)

The flux, φ, is related to the rigidity and the phase-space density by:

ϕi =Ai

ZiP 2D, (7)

where Ai and Zi are the atomic weights and atomic numbers of cosmic-raynuclei of type i.

The diffusion coefficient κ, in Eq. (6) has collapsed to a scalar, and isgiven by:

κ = κ0Pβ (8)

and κo is obtained from

κ0 =13

V r

U(9)

where r is the distance to the solar-wind termination shock, taken here as100AU (one AU is the earth–sun distance, 150million km), and U is theheliocentric potential.

Finally, U is obtained from14

ϕi(E, r = 1) =(

P (E)P (T )

)2

ϕi(T, r = rb), T = E + ZiU. (10)

The flux, φi(E, r = 1) is the cosmic-ray flux of particles of type I, the fluxφi(T, r = rb) is the flux at the solar-wind transition shock.

Page 232: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 219

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

4.2. Cosmic rays in extraterrestrial atmospheres

These methods have been applied to a number of cases. Ionization profileshave been calculated in the atmospheres of Mars (S. Tripathi, PrivateCommunication, 2006) and Titan,21 and radiation intensity profiles in theatmosphere of Mars.22 Figure 9 shows the ionization profile calculated inthe atmosphere of Titan.

4.3. Radiation shielding for a Mars mission

Energetic solar-particle events represent a serious danger to the health ofastronauts in interplanetary space. As solar-particle events are less frequentat solar minimum than at other phases of the solar cycle, it is anticipatedthat a Mars mission would be flown at solar minimum. The cosmic-raygradient 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 solarminimum.

Page 233: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

220 K. O’Brien

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

Spacecraft hull Earth/Mars space Martian surface 2.5-year mission(rem/year) (rem/year) (rem)

2 g/cm2 Al, 63.75 17.17 136.14 g/cm2 CH2

2 g/cm2 Al, 55.02 11.51 115.820 g/cm2 CH2

not vary significantly with location (although it would, of course, vary withtime). The dose rate through various hull thicknesses of aluminum andpolyethylene 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 monthson the surface of Mars is shown in Table 1.

The resulting doses would be a little over the mission limit of oneSievert. 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 dayson a very small island in the midst of the Pacific Ocean through computercalculations of radiation transport through accelerator shields, planetaryatmospheres, and spacecraft hulls. I have not had time or space to includework in other fields, such as hospital physics, Monte Carlo methods, andcosmogenic isotope production. That would make this a very long paperindeed. It is my hope that the reader finds the variety and evolution ofthese 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).

Page 234: Advances in Geosciences: Planetary Science (PS) (Advances in Geosciences)

May 29, 2007 6:57 spi-b486 AOGS 2006 - PS Volume 9in x 6in ch18

Nuclear Blasts to Cosmic Bombardment 221

4. K. O’Brien, R. Sanna, M. Alberg, J. E. McLaughlin and S. Rothenberg, NuovoCimento Suppl. 3 (1965) 409.

5. K. O’Brien, R. Sanna and J. E. McLaughlin, First International Conferenceon Accelerator Dosimetry and Experience (U. S. Atomic Energy ReportCONF-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 CascadeEquations, (Oak Ridge National Laboratory Report ORNL-3746, 1962).

13. C. Passow, Phenomenologische Theorie zur Berechnung einer Kaskade ausschweren Teilchen (Nukleonenkaskade) in der Materie (Deutches ElektronenSynchrotron report, Notiz A 285, 1962).

14. K. O’Brien, The theory of cosmic-ray and high-energy solar-particle transportin the atmosphere, in The Natural Radiation Environment VII, SeventhInternational Symposium on the Natural Radiation Environment (NRE-VII),eds. J. P. McLaughlin, S. E. Simpopoulos and F. Steinhausler, 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. Komle and

K. O’Brien, Icarus 154 (2001) 216.23. K. O’Brien, Adv. Space Res. 36 (2005) 1731.