RES EARCH

ASTEROIDS of determining its physical and compositional

properties, collecting samples, and return-
Pebbles and sand on asteroid (162173) Ryugu: In situ ing them to Earth. Ryugu is a “spinning
top”–shaped rubble pile, with a mean radius
observation and particles returned to Earth of 448 ± 2 m (4, 5). The surface is ubiqui-
tously dark, with variations in reflectance
S. Tachibana1,2*, H. Sawada2, R. Okazaki3, Y. Takano4, K. Sakamoto1,2, Y. N. Miura5, C. Okamoto6†, spectra that are due to mixing of bluish and
H. Yano2, S. Yamanouchi3, P. Michel7, Y. Zhang7, S. Schwartz8,9, F. Thuillet7‡, H. Yurimoto10, reddish materials (5, 6). Some bright boul-
T. Nakamura11, T. Noguchi3,12, H. Yabuta13, H. Naraoka3, A. Tsuchiyama14,15, N. Imae16, K. Kurosawa17, ders are present, which could be related to
A. M. Nakamura6, K. Ogawa18, S. Sugita1, T. Morota1, R. Honda19, S. Kameda20, E. Tatsumi1,21, Y. Cho1, spectroscopically similar S-type asteroids (7).
K. Yoshioka1, Y. Yokota2, M. Hayakawa2, M. Matsuoka2§, N. Sakatani20, M. Yamada17, T. Kouyama22, The reddish color is thought to have been
H. Suzuki23, C. Honda24, T. Yoshimitsu2, T. Kubota2, H. Demura24, T. Yada2, M. Nishimura2, K. Yogata2, produced by surface alteration and space
A. Nakato2, M. Yoshitake2, A. I. Suzuki25,26, S. Furuya1,2, K. Hatakeda25, A. Miyazaki2, K. Kumagai25, weathering of originally bluish materials
T. Okada2, M. Abe2,27, T. Usui2, T. R. Ireland28, M. Fujimoto2, T. Yamada2, M. Arakawa6, during the past 106 to 10 7 years (6). Hydrous
H. C. Connolly Jr.29,8, A. Fujii2, S. Hasegawa2, N. Hirata24, N. Hirata6, C. Hirose30, S. Hosoda2, silicates are present across the surface (8)
Y. Iijima2†, H. Ikeda2, M. Ishiguro31, Y. Ishihara18, T. Iwata2,27, S. Kikuchi2,17, K. Kitazato24, but are less abundant than in hydrated car-
D. S. Lauretta8, G. Libourel7, B. Marty32, K. Matsumoto33,34, T. Michikami35, Y. Mimasu2, A. Miura2,27, bonaceous chondrites (8) or the B-type (bluish
O. Mori2, K. Nakamura-Messenger36, N. Namiki33,34, A. N. Nguyen36, L. R. Nittler37, H. Noda33,34, and spectroscopically similar to C-type) as-
R. Noguchi2,38, N. Ogawa18, G. Ono30, M. Ozaki2,27, H. Senshu17, T. Shimada18, Y. Shimaki2, K. Shirai2, teroid (101955) Bennu (9, 10). This could be
S. Soldini39, T. Takahashi40, Y. Takei2,30, H. Takeuchi2,27, R. Tsukizaki2, K. Wada17, Y. Yamamoto2,27, due to dehydration of originally hydrous
K. Yoshikawa30, K. Yumoto1, M. E. Zolensky36, S. Nakazawa2, F. Terui2¶, S. Tanaka2,27, T. Saiki2, silicates or minimal aqueous alteration of
M. Yoshikawa2,27, S. Watanabe41, Y. Tsuda2,42 Ryugu’s parent planetesimal (the original
body in the early Solar System from which
The Hayabusa2 spacecraft investigated the C-type (carbonaceous) asteroid (162173) Ryugu. The mission Ryugu formed) (5, 8).
performed two landing operations to collect samples of surface and subsurface material, the latter Hayabusa2 dropped the Mobile Asteroid
exposed by an artificial impact. We present images of the second touchdown site, finding that ejecta Surface Scout (MASCOT) lander onto Ryugu,
from the impact crater was present at the sample location. Surface pebbles at both landing sites show which showed that the surface is not covered
morphological variations ranging from rugged to smooth, similar to Ryugu’s boulders, and shapes with fine regolith (11). A ~3-cm pebble ob-
from quasi-spherical to flattened. The samples were returned to Earth on 6 December 2020. We describe served by using MASCOT had a thermal
the morphology of >5 grams of returned pebbles and sand. Their diverse color, shape, and structure inertia of ~280 J m−2 K−1 s−1/2, which is much
are consistent with the observed materials of Ryugu; we conclude that they are a representative sample lower than the thermal inertia of chondrites
(12). This low thermal inertia indicates that

Downloaded from https://www.science.org on March 07, 2022

of the asteroid.
the pebble had a high porosity, implying a

A
low tensile strength of a few hundred kilo-
steroids are small celestial bodies in the are predicted to contain a record of Solar Sys- pascals (12). Similarly low thermal inertia
Solar System that are left over from the tem evolution (1). Evidence for ongoing activity (~300 J m−2 K−1 s−1/2) was measured for seve-
planet formation process. The C-type (car- on an asteroid surface, including movement ral large (>10 m) boulders and their surround-
bonaceous) group of asteroids appear and ejection of particles, has previously been ings (13). An artificial impact experiment was
to be related to carbonaceous chondrite inferred from analysis of particles returned to performed by using Hayabusa2’s Small Carry-
meteorites, which are known to contain hy- Earth from the S-type (stony) asteroid (25143) on Impactor (SCI), which showed that Ryugu’s
drated silicates and organic matter (1). Such Itokawa (2, 3). surface is composed of a cohesionless material,
hydrated asteroids could have delivered water The Hayabusa2 spacecraft investigated the at least in part (14). Infrared observations of
and organic molecules to Earth during or after C-type near-Earth asteroid (162173) Ryugu from the SCI-made crater have shown that the
its formation. Samples from C-type asteroids June 2018 to November 2019, with the goal subsurface material has spectral properties

1
UTokyo Organization for Planetary and Space Science–Department of Earth and Planetary Science, The University of Tokyo, Tokyo 113-0033, Japan. 2Institute of Space and Astronautical
Science, Japan Aerospace Exploration Agency (JAXA), Sagamihara 252-5210, Japan. 3Department of Earth and Planetary Sciences, Kyushu University, Fukuoka 812-8581, Japan.
4
Biogeochemistry Research Center, Japan Agency for Marine-Earth Science and Technology, Kanagawa 237-0061, Japan. 5Earthquake Research Institute, The University of Tokyo, Tokyo
113-0032, Japan. 6Department of Planetology, Kobe University, Kobe 657-8501, Japan. 7Université Côte d’Azur, Observatoire de la Côte d’Azur, Centre national de la recherche scientifique,
Laboratoire Lagrange, F-06304 Nice CEDEX 4, France. 8Lunar and Planetary Laboratory, University of Arizona, Tucson, AZ 85705, USA. 9Planetary Science Institute, Tucson, AZ 85719, USA.
10
Department of Earth and Planetary Sciences, Hokkaido University, Sapporo 060-0810, Japan. 11Department of Earth Sciences, Tohoku University, Sendai 980-8578, Japan. 12Division of Earth
and Planetary Sciences, Kyoto University, Kyoto, Japan. 13Department of Earth and Planetary Systems Science, Hiroshima University, Higashi-Hiroshima 739-8526, Japan. 14Research
Organization of Science and Technology, Ritsumeikan University, Kusatsu 525-8577, Japan. 15Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China.
16
Polar Science Resources Center, National Institute of Polar Research, Tokyo 190-8518, Japan. 17Planetary Exploration Research Center, Chiba Institute of Technology, Narashino 275-0016,
Japan. 18JAXA Space Exploration Center, JAXA, Sagamihara 252-5210, Japan. 19Department of Information Science, Kochi University, Kochi 780-8520, Japan. 20Department of Physics, Rikkyo
University, Tokyo 171-8501, Japan. 21Instituto de Astrofísica de Canarias, University of La Laguna, E-38205 Tenerife, Spain. 22Information Technology and Human Factors, National Institute of
Advanced Industrial Science and Technology, Tokyo 135-0064, Japan. 23Department of Physics, Meiji University, Kawasaki 214-8571, Japan. 24Aizu Research Center for Space Informatics,
University of Aizu, Aizu-Wakamatsu 965-8580, Japan. 25Marine Works Japan Ltd., Yokosuka 237-0063, Japan. 26Department of Economics, Toyo University, Tokyo 112-8606, Japan. 27Department
of Space and Astronautical Science, The Graduate University for Advanced Studies, SOKENDAI, Hayama 240-0193, Japan. 28School of Earth and Environmental Sciences, The University of
Queensland, St Lucia, Queensland 4072, Australia. 29Department of Geology, Rowan University, Glassboro, NJ 08028, USA. 30Research and Development Directorate, JAXA, Sagamihara
252-5210, Japan. 31Department of Physics and Astronomy, Seoul National University, Seoul 08826, Korea. 32Université de Lorraine, Centre national de la recherche scientifique, Centre de
Recherches Pétrographiques et Géochimiques, F-54000 Nancy, France. 33National Astronomical Observatory of Japan, Mitaka 181-8588, Japan. 34Department of Astronomical Science, The
Graduate University for Advanced Studies, SOKENDAI, Hayama 240-0193, Japan. 35Department of Mechanical Engineering, Kindai University, Higashi-Hiroshima 739-2116, Japan. 36NASA Johnson
Space Center, Houston, TX 77058, USA. 37Carnegie Institution for Science, Washington, DC 20015, USA. 38Department of Science, Niigata University, Niigata 950-2181, Japan. 39Department of
Mechanical, Materials and Aerospace Engineering, University of Liverpool, Liverpool L69 3BX, UK. 40NEC Corporation, Tokyo 183-8501, Japan. 41Department of Earth and Environmental Sciences,
Nagoya University, Nagoya 464-8601, Japan. 42Department of Aeronautics and Astronautics, The University of Tokyo, Tokyo 113-0033, Japan.
*Corresponding author. Email: tachi@eps.s.u-tokyo.ac.jp †Deceased. ‡Present address: CS Group, 31506 Toulouse CEDEX 5, France. §Present address: Laboratoire d’Etudes Spatiales et d’Instrumentation en
Astrophysique, Observatoire de Paris, 92195 Meudon, France. ¶Present address: Department of Mechanical Engineering, Kanagawa Institute of Technology, Kanagawa 243-0292, Japan.

Tachibana et al., Science 375, 1011–1016 (2022) 4 March 2022 1 of 6

RES EARCH | R E S E A R C H A R T I C L E

Fig. 1. Hayabusa2 touchdown

locations and Ryugu surface
properties. (A) Global map of the
spectral slope, which is indicated
by the color bar, superimposed
on a v-band image map. The
spectral slope is measured
between the b-band (0.48 mm)
and the x-band (0.86 mm) (5, 6).
The white arrows indicate the
locations of the first touchdown
(TD1) and the second touchdown
(TD2). (B) Composite map of the
TD2 site, assembled from images
taken by a wide-angle optical
navigation camera (ONC-W1). Also
visible are the locations of a target
marker (arrow labeled TM) that
was used for spacecraft navigation
and the crater produced by the
SCI experiment (dotted arc, diame-
ter of ~18 m) (19). (C and D) Images
of the areas surrounding the TD1
and TD2 sites, respectively, taken
by another wide-angle optical
navigation camera (ONC-W2).
Examples of flattened boulders
and pebbles are indicated with
green arrows. Labels indicate type
1 and type 2 boulders, which have
rugged and smooth surfaces,
respectively (5). The SCI crater

Downloaded from https://www.science.org on March 07, 2022

(14) is visible (dotted ellipse) in
the image of the TD2 site. The
arch-like feature above the surface
in (C) is an artifact.

similar to, but distinct from, those of the sur- The TD2 location hosts fine particle aggre- should access greater depths; nevertheless, we
face (15). The crater and excavated material gates, observed on the surface of a smooth expect the samples collected during TD2 to
have a slightly higher abundance of hydrated type 2 boulder (Fig. 2A), which were not iden- include some SCI ejecta. Spectroscopy of the
silicates, reflecting the extent of aqueous al- tified in proximity images of TD1 or other SCI crater and its surroundings revealed only
teration that occurred on Ryugu’s parent plan- surface locations (6). These particles did not small differences (15), so the identification of
etesimal (15). strongly adhere to the boulder, being blown subsurface materials in the collected sample
off by the subsequent ascent thruster firing requires other analysis methods.
Sample-collection operations (Fig. 2B). We infer that the fine particles are During both landings, the motions of par-
Hayabusa2 made its first landing [designated geologically recent and, most likely, ejecta from ticles kicked up by the sample projectiles and
touchdown one (TD1)] on 2019 February 21, the SCI crater. Images of the impact event show thruster firings were observed with a small
during which it collected samples of the sur- that part of the ejecta curtain fell back on monitor camera head (CAM-H) (Fig. 3 and
face (6). The second landing (TD2) occurred Ryugu’s surface, with simulations predicting movie S1) (20). One second after the projectile
on 2019 July 11, close to the crater made by that the TD2 site was covered with SCI ejecta firing at TD1, ~10 particles were identified
the SCI, to collect impact ejecta, i.e., sub- excavated from a depth of ~1 m below the beneath or nearby the sampler horn (Fig. 3C).
surface samples (Fig. 1). The two landing surface (18, 19). The estimated sizes of ejecta After another second, the number of particles
locations appeared similar in remote imag- particles range from 1 mm to several decimeters increased to ~20 in the next image (Fig. 3D), of
ing, being covered with boulders and pebbles (18). This is consistent with the observed par- which 3 particles were moving toward CAM-H.
(Fig. 1). The two types of boulder commonly ticles on the boulder (Fig. 3). We conclude We estimate the ejection angle and velocity of
observed on Ryugu’s surface [rugged type 1 that the TD2 sample location was covered these particles as ~50° to 60° and 1 to 2 m s−1,
and smooth type 2 (5)] are found at both with ejecta excavated from Ryugu’s subsurface respectively (20). Their ejection angle is within
locations. In both landing operations, a 5-g by the SCI experiment. Sample-collection analog the range measured in projectile experiments
tantalum projectile was fired through the experiments that were performed in Earth grav- in Earth gravity (40° to 60°; fig. S2) (20) and
sampler horn at ~300 ± 30 m s−1 (1, 16, 17) ity indicate that ~50% of the collected particles consistent with the most frequent angle range
when the horn touched the surface, lifting were taken from depths <1.5 mm from the sur- (48° to 54°) in the Hayabusa2 sampling sim-
material into the collector. face (fig. S3) (20). Collection under microgravity ulations (21). We conclude that these particles

Tachibana et al., Science 375, 1011–1016 (2022) 4 March 2022 2 of 6

RES EARCH | R E S E A R C H A R T I C L E

are ejecta produced by the projectile impact. during TD2 was similar to the amount collected pebbles as well. Numerous millimeter-sized
The ejected particles identified near the rocket during TD1. particles were also observed during the TD2
coupling ring (an assembly used during the ascent operation (Fig. 4), which indicates the
spacecraft launch) were centimeter-sized (Fig. 3). Spacecraft ascent presence of more small particles at the TD2
Centimeter-sized pebbles were also found in CAM-H continued to take images of flying location than at the TD1 location, which were
proximity images of the TD1 site (6) and at particles during the ascent after the two land- presumably the SCI ejecta.
different surface locations observed by the ing operations (Fig. 4). Flying particles are The flying pebbles show two morpholog-
lander MASCOT (11) and the MINERVA-II visible as objects moving relative to the sur- ical types: rugged particles and particles
(Micro Nano Experimental Robot Vehicle for face in multiple sequential images (movies S1 with smooth faces (Fig. 4 and fig. S4). These
Asteroid 2) Rover-1A (Fig. 2). The data from and S2). Because no such flying particles were two types are consistent with the morpho-
MINERVA-II Rover-1A also showed centimeter- observed during the spacecraft descent, we logical variations that were observed within
sized particles that were disturbed by its hop interpret these particles as ejecta either due surface boulders observed by the spacecraft
across the surface (Fig. 2D) (20). This suggests to the projectile impact or lifted by the space- (5) and the MASCOT lander (11). We deter-
that centimeter-sized pebbles, which do not craft thruster firing. Images from a wide-angle mined the three-dimensional shapes of fly-
strongly adhere to larger cobbles and boulders, optical navigation camera showed boulders ing pebbles for those visible in multiple
are present over the surface of Ryugu. moving on the surface because of the thruster two-dimensional projections (from different
We estimate the total mass of the three ejecta operation during TD1 (6), so we infer that the angles) in the CAM-H images. We define L
particles observed in Fig. 3 as 0.3 to 3 g, assum- thruster triggered the ejection of most of the and I as the maximum and minimum caliper
ing spherical particles with a diameter of 0.5 to
1 cm and bulk density of ~2 g cm−3. If all the
ejecta particles in Fig. 3 (~20 in number) have
the same ejection velocity as the three par-
ticles observed moving toward CAM-H, the total
amount of ejecta with an ejection velocity of
~1 m s−1 would be ~2 to 20 g. Numerical sim-
ulations of the Hayabusa2 sample process,
which assume a cohesionless granular bed
consisting of grains with an average diameter
of 0.5 cm, predicted that the total ejecta mass
is about an order of magnitude larger than
the mass of ejecta with a velocity of ~1 m s−1
(21). We therefore estimate the total ejecta

Downloaded from https://www.science.org on March 07, 2022

mass as 20 to 200 g. The simulations also
predicted that ~0.5% of the total ejecta par-
ticles are retained in the sample catcher,
which is located above the sampler horn (21).
This leads to an estimated total collected mass
of 0.1 to 1 g for TD1, which meets the 0.1-g
requirement for returned sample analysis
(1, 16). Laboratory experiments conducted
under Earth gravity by using 1-mm glass
spherules with little cohesion force showed
that 150 to 250 mg of samples can be collected
after projectile firing (fig. S1) (20), which was
expected to be increased up to an order of
magnitude under microgravity conditions
(16). Because the surface materials on Ryugu
are not strongly held to the surface by cohesive
forces (14), we regard this experiment as an
appropriate analog of the sample-collection
operation on Ryugu.
In the CAM-H images taken during TD2
(Fig. 3 and movies S2 and S3), dust-like ejecta
appear from beneath the sampler horn (Fig. 3),
followed by the ejection of numerous larger Fig. 2. Pebbles and boulders observed on Ryugu’s surface. (A) ONC-W1 image taken 2 s before TD2.
particles, which is likely due to the projectile Fine particles are visible (within the pink box, which measures ~20 by 20 cm) on the surface of a smooth
firing. We interpret this dust as ejecta with type 2 boulder (5). (B) Same area as shown in (A), taken during the ascent after TD2. The fine particles on
slow velocities or originating from deeper into the smooth type 2 boulder were blown off by the thruster firing. (C) Image of the surface taken by the
the surface that did not enter the sampler MINERVA-II Rover-1A during its hopping operation on 2018 September 28. The shadow of the rover (~7 cm
horn. Three particles visible in Fig. 3 are ~1 to long) is visible in the center of the image. Numerous decimeter- to centimeter-sized pebbles are visible.
2 cm in diameter, which suggests that loosely Boulders with layered structure (pink boxes with layers indicated by dotted lines) are observed, along with a
packed, movable, centimeter-sized pebbles were boulder from which a flattened piece seems to be peeling (arrow in the white box). Both type 1 and
present at the TD2 location. The CAM-H images type 2 boulders are present in this region (labeled). (D) Image taken during a hop of MINERVA-II Rover-1A on
suggest that the amount of material collected 2018 October 16. Ejected centimeter-sized pebbles are visible within the white box.

Tachibana et al., Science 375, 1011–1016 (2022) 4 March 2022 3 of 6

RES EARCH | R E S E A R C H A R T I C L E

distribution centered at (S/L, I/L) of (0.53,

0.69) and (0.35, 0.48) (Fig. 5), which means
that two shape types are present in the pop-
ulation of surface pebbles, which we refer
to as “subequant” and “elongated block” (23).
Pebbles and small boulders with a shape that
is both elongated and flat are found at both
landing locations (Fig. 1). This suggests that
the bimodal distribution of pebble shapes
is indigenous to Ryugu’s surface. However,
because such a bimodal distribution is not
found for boulders larger than 5 m (24), we
conclude that this shape variation results
from boulder fragmentation or foliation (Fig.
2) (5). This elongated and flat morphology
is not typical among clasts (embedded frag-
ments) in carbonaceous chondrite meteor-
ites but is similar to the texture of clasts in
shocked hydrated carbonaceous chondrites
(25–28). Some of the latter show a high den-
sity of parallel fractures that formed because
of sudden volatile loss during the release of
shock pressure (28).
During the TD1 ascent operation, CAM-H
observed a centimeter-sized pebble that passed
between the camera and the rocket coupling
ring (Fig. 4, C and D). The pebble hit the
spacecraft; then 4 s later, a smaller particle
(~4 mm in size) appeared from the spacecraft
side (Fig. 4E). Because no other particles coming
from the spacecraft side were observed during
the TD1 and TD2 operations, the ~4-mm–

Downloaded from https://www.science.org on March 07, 2022

sized particle is likely to be a fragment of this
centimeter-sized pebble that resulted from its
impact with the spacecraft. The CAM-H images
from the ascending spacecraft (~1 m s−1) sug-
gest that the relative velocity of the pebble to
the spacecraft was ~0.1 m s−1. Because frag-
mentation of typical carbonaceous chondrite
material requires an impact velocity of >1 m s−1
(20, 29), this implies that the tensile strength of
the pebble is much lower than that of typical
chondrites (fig. S6) (17, 20). The highly porous
material identified on the surface (30) could be
of similar composition to the fragile pebble.
Alternatively, the pebble might have contained
a crack (or cracks), such as the one that was
observed on a boulder by the MINERVA-II
Rover-1A (Fig. 2).

Fig. 3. CAM-H images of the TD1 and TD2 sample acquisition processes. (A to F) TD1 operation. The Samples returned to Earth
lower left of each panel indicates the time t from projectile firing, ranging from −1 to +4 s. Ejecta particles are Hayabusa2 left Ryugu in November 2019. On
visible after t = 0. A reflection plate for a laser range finder (23 mm by 23 mm) on the sampler horn 6 December 2020, the reentry capsule contain-
and the rim of the rocket coupling ring (4 mm; distance between arrows) are labeled as size references. ing the samples was delivered to Woomera,
The white arrows in (D) and (E) indicate the same group of particles moving toward CAM-H. (G to South Australia. After transfer to a clean room,
L) Equivalent images of TD2. Three particles, indicated with white arrows in (J), are also seen as mirror the sample chambers were opened and found
images reflected on the rocket coupling ring (red arrows). to contain ~5 g of material (31). This is ~50 times
more than the mission minimum requirement
of 0.1 g (1, 16). The samples recovered from
lengths—the distance between two parallel lines to longest axis (S/L) and intermediate axis to chamber A (24 cm3) of the sample catcher,
tangential to the surface—of the maximum-area longest axis (I/L) ratios of the particles (Fig. 4 which was used for the storage of TD1 sam-
projection of each particle and S as the smallest and tables S1 and S2) show that there is a ples, weigh ~3 g. This mass is consistent with
dimension measured in the minimum-area fraction of elongated block-like flat particles the estimate above based on CAM-H images.
projection (fig. S5) (20, 22). The shortest axis on the surface. The 67 particles have a bimodal We therefore conclude that the sample in

Tachibana et al., Science 375, 1011–1016 (2022) 4 March 2022 4 of 6

RES EARCH | R E S E A R C H A R T I C L E

Fig. 5. Shape parameters of flying particles

observed during touchdown operations and par-
ticles returned to Earth. (A) Properties of

Downloaded from https://www.science.org on March 07, 2022

particles observed during TD1 and TD2 operations.
S/L is the ratio of the shortest axis to the longest
axis, and I/L is the ratio of the intermediate axis to
the longest axis (25). Contours indicate the proba-
Fig. 4. CAM-H images of flying particles during the ascent operations. (A) Particle with smooth faces bility distribution function of 67 particles, assuming
indicated with a white arrow. (B) Rugged particle indicated with a white circle. (C and D) Flat particle (white a bimodal distribution (25). (B) Properties of
circle) and particle that appears to have hit the spacecraft (blue arrow). (E) Particle coming from the returned samples, six particles each from chambers
spacecraft side indicated with a blue arrow. Its mirror image is seen on the rocket coupling ring (red arrow). A and C (25), overlain on the same contours from
The particle’s direction of movement is shown with a white dotted arrow. (F) Millimeter-sized particles (A). The distribution is similar. Data for all the
(within the yellow box) observed after TD2. Movies S1 and S2 show the full footage of these operations, in particles in both panels are listed in tables S1 to S3.
which the particle movements are visible. (A) to (E) are from TD1, and (F) is from TD2.

chamber A was lifted by the projectile firing, of the sample container, millimeter-sized sand abusa2 sampler (16, 21). The presence of these
which worked efficiently to collect asteroid and nearly centimeter-sized pebbles were found, large pebbles in chamber C, despite the smaller
surface regolith. This is unlike the particles along with submillimeter-sized fine powder. The total mass, can be explained either by projectile
returned from the asteroid Itokawa by the grain size variation is consistent with expec- destruction of a larger rock or the scoop-up
original Hayabusa spacecraft, as its projec- tations derived from the surface observation by component of the sampler horn, which was
tile failed to fire, which complicated in- the MASCOT lander (11), MINERVA-II rovers, designed to pick up surface pebbles (16).
terpretation of the samples (2). Chamber C Hayabusa2 cameras (Figs. 2 to 4), and polari- Millimeter-sized fine grains and submillimeter-
(12 cm3) of the sample catcher, which was metric observations from Earth (32). All the sized sand particles were also found in cham-
used for samples from TD2, contained ~2 g. particles in the two chambers appeared black ber C, which are likely to include subsurface
This is also consistent with the estimate (Fig. 6), which is consistent with the color and material, as observed on a boulder (Fig. 2A).
above from the CAM-H images, which sug- albedo of Ryugu’s boulders (5, 6). The sizes of Chamber B, which was not used for either
gests that chamber C samples are also the the collected particles are consistent with the landing operation, is located between chambers
ejecta from the projectile firing. Unlike TD1, ejecta observed during each landing operation A and C. A small number of fine particles
the TD2 sample likely includes material from (Fig. 3). The largest grains from chamber A are (smaller than 1 mm) were found in this cham-
the subsurface that was excavated by the SCI ~5 mm in size, whereas there are three pebbles ber. This shows that no extensive mixing of
impact. larger than 5 mm from chamber C (Fig. 6 and particles occurred through the gaps between
We compare the properties of the returned table S3). The longest dimension of the largest the chambers during the return to Earth or
particles to the constraints derived above from pebble in chamber C is 10.3 mm, which is close capsule recovery (16). We are therefore confi-
the sample collection images. In both chambers to the maximum size obtainable by the Hay- dent that the pebbles and sand in chambers A

Tachibana et al., Science 375, 1011–1016 (2022) 4 March 2022 5 of 6

RES EARCH | R E S E A R C H A R T I C L E

sample (Fig. 6) have S/L and I/L ratios (Fig. 5B) 31. T. Yada et al., Nat. Astron. 10.1038/s41550-021-01550-6
that are consistent with those of the flying par- (2021).
32. D. Kuroda et al., Astrophys. J. Lett. 911, L24 (2021).
ticles observed at Ryugu (Fig. 5A). 33. G. Libourel et al., Mon. Not. R. Astron. Soc. 500, 1905–1920
Many returned particles feature curved and (2021).
straight cracks. Pebbles with a smooth surface
AC KNOWLED GME NTS
could be fragments of particles with straight Hayabusa2 was developed and built under the leadership of JAXA,
cracks, possibly formed by shock or thermal with contributions from the German Aerospace Center and the
fatigue (33). The common presence of cracks Centre National d’Études Spatiales (CNES) and in collaboration
with NASA and other universities, institutes, and companies in
in returned pebbles implies that the small
Japan. The sampler system was developed by JAXA, The University
thermal inertia of surface boulders (12, 13) is of Tokyo, Hokkaido University, Kyushu University, Japan Agency
probably due, at least in part, to cracks or for Marine-Earth Science and Technology, and other universities,
fractures in their interior. Microcracks or institutes, and companies in Japan. Funding: S. Tachibana
acknowledges JSPS KAKENHI Grant (JP 20H05846). S.W.
microporosity could also be responsible for acknowledges JSPS KAKENHI Grant (17H06459 and 19H01951).
the low thermal inertia. P.M., B.M., Y.Z., F. Thuillet, and G.L. acknowledge the French space
The color, shape, surface morphology, and agency CNES. P.M. and Y.Z. acknowledge funding from the
European Union’s Horizon 2020 research and innovation program
structure of the returned pebbles and sand under grant agreement no. 870377 (project NEO-MAPP), from
match those of Ryugu’s surface material ob- the Université Côte d’Azur “Individual grants for young researchers”
served from the spacecraft. We therefore con- program and from Academies of Excellence: Complex Systems and
Space, Environment, Risk, and Resilience, part of the IDEX JEDI of
clude that the pebbles and sand inside chambers Université Côte d’Azur. B.M. acknowledges funding from the
A and C are representative samples of Ryugu at European Research Council (grant agreement no. 695618). Author
two surface sites, without substantial alteration contributions: S. Tachibana. coordinated coauthor contributions; led
the sampler development with H. Sawada; performed data analyses,
during the sample collection and return to and interpretations; and wrote the paper, with contributions from
Earth. The variations in physical properties H. Sawada, R.O., Y. Takano, K. Sakamoto, and H. Yano. Sampler
among the pebbles and sand, which were development and operation: H. Sawada, R.O., Y. Takano.,
K. Sakamoto, Y.N.M., C.O., H. Yano, S.Y., T. Noguchi, T. Nakamura,
not expected before spacecraft arrival at the
A.T., N.I., K. Kurosawa, and A.M.N.; CAM-H operation: H. Sawada
asteroid, reflect the geological history of and K.O.; ONC data acquisition and reduction: S. Sugita, R.H.,
Ryugu (1). T. Morota, Y. Iijima, S. Kameda, H. Sawada, E.T., C. Honda, Y. Yokota,
M. Yamada, T. Kouyama, N.S., K.O., H. Suzuki, K. Yoshioka, M.H.,
Y.C., M.I., A. Miura, and M.M.; MINERVA-II rovers operation:
T. Yoshimitsu, T. Kubota, and H.D.; capsule retrieval operation and
RE FERENCES AND NOTES curation: S.N., M.F., T. Yamada, T.R.I., H. Sawada, R.O., K. Sakamoto,
1. S. Tachibana et al., Geochem. J. 48, 571–587 (2014). Y. Takano, Y.N.M., H. Yano, M.N., K. Yogata, A.N., M. Yoshitake, A.I.S.,
2. A. Tsuchiyama et al., Science 333, 1125–1128 (2011). S.F., K.H., A. Miyazaki, K. Kumagai, T.O., M. Abe, H. Yurimoto, T.U.,

Downloaded from https://www.science.org on March 07, 2022

3. K. Nagao et al., Science 333, 1128–1131 (2011). and K.N.-M.; landing site characterization: Y. Tsuda, S.W., T. Saiki,
S. Kikuchi, N.O., Y. Yamamoto, Y.S., K. Shirai, N.H. (Kobe), K.O.,
4. S. Watanabe et al., Science 364, 268–272 (2019).
K. Kitazato, N.H. (Aizu), K.W., H. Yabuta, Y. Ishihara, R.N., T. Morota,
5. S. Sugita et al., Science 364, 252 (2019).
N.S., K.M., H. Senshu, R.H., E.T., Y. Yokota, C. Honda, T. Michikami,
6. T. Morota et al., Science 368, 654–659 (2020). M.M., and A.M.; interpretation and writing contribution: S. Sugita,
7. E. Tatsumi et al., Nat. Astron. 5, 39–45 (2021). Y. Takano, P.M., Y.Z., S. Schwarz, F. Thuillet, H. Yurimoto,
Fig. 6. Microscope images of particles returned 8. K. Kitazato et al., Science 364, 272–275 (2019). T. Nakamura, T. Noguchi, H. Yabuta, H. Naraoka, A.M.N., K. Kitazato,
to Earth inside the sample container. Particles 9. D. S. Lauretta et al., Nature 568, 55–60 (2019). T. Morota, T. Michikami, S. Kameda, E.T., T. Yoshimitsu, T. Yada,
10. V. E. Hamilton et al., Nat. Astron. 3, 332–340 (2019). T.O., T.U., T.R.I., M.F., H.C.C., S. Hasegawa, D.S.L., G.L., B.M., A.N.N.,
cataloged as A0008, A0019, A0046, A0048, A0069,
11. R. Jaumann et al., Science 365, 817–820 (2019). L.R.N., K.W., K. Yumoto, M.E.Z.; spacecraft science operations:
and A0083 are from chamber A of the sample 12. M. Grott et al., Nat. Astron. 3, 971–976 (2019). S. Tanaka, M. Yoshikawa, T.I., Y. Yamamoto, K.M., M. H., T.O., R.N.,
container used for TD1. C0002, C0004, C0013, 13. T. Okada et al., Nature 579, 518–522 (2020). Y.S., N.S., H.N., M.M., H. Yano, R.T., M.O., F. Terui, N.O., H. Sawada,
C0019, C0028, and C0055 are from chamber C of 14. M. Arakawa et al., Science 368, 67–71 (2020). S. Kikuchi, H.T., G.O., Y.M., K. Yoshikawa, T.T., Y. Takei, A.F., C. Hirose,
15. K. Kitazato et al., Nat. Astron. 5, 246–250 (2021). S.N., S. Hosoda, O.M., T. Shimada, S. Soldini, T. Saiki, S.W., and
the sample container used for TD2. The particles Y. Tsuda; project administration: S.W., M. Yoshikawa, S. Tachibana,
16. H. Sawada et al., Space Sci. Rev. 208, 81–106 (2017).
A0008, A0019, A0069, A0083, and C0002 have K. Kitazato, S. Sugita, T.O., N.N., M. Arakawa, M. Abe, H.I.,
17. R. Okazaki et al., Space Sci. Rev. 208, 107–124 (2017).
rugged surfaces, whereas A0046 and C0019 have S. Tanaka, S.N., F.T., T. Saiki, and Y. Tsuda. All authors discussed
18. K. Wada et al., Astron. Astrophys. 647, A43 (2021).
the results and commented on the manuscript. Competing
smooth surfaces. The particles A0048, C0028, and 19. R. Honda et al., Icarus 366, 114530 (2021). interests: The authors declare no competing interests. Data and
C0055 show elongated blocklike morphologies. 20. Materials and methods are available as supplementary materials availability: All images and data used in this study are
materials. available at the JAXA Data Archives and Transmission System
Particles were mounted on individual sample hold-
21. F. Thuillet, P. Michel, S. Tachibana, R.-L. Ballouz, (DARTS) at www.darts.isas.jaxa.jp/pub/hayabusa2/paper/sample/
ers in a nitrogen atmosphere and are lit from the left S. R. Schwartz, Mon. Not. R. Astron. Soc. 491, 153–177 Tachibana_2022/. Other data from the mission are available
and right simultaneously. (2019). at the DARTS archive www.darts.isas.jaxa.jp/planet/project/
22. G. H. Bagheri, C. Bonadonna, I. Manzella, P. Vonlanthen, hayabusa2/ and on the Small Bodies Node of the NASA Planetary
Powder Technol. 270, 141–153 (2015). Data System https://pds-smallbodies.astro.umd.edu/data_sb/
23. S. J. Blott, K. Pye, Sedimentology 55, 31–63 (2008). missions/hayabusa2/. The samples of Ryugu are curated by the
and C are the samples acquired in the TD1 24. T. Michikami et al., Icarus 331, 179–191 (2019). JAXA Astromaterials Science Research Group; distribution for analysis
25. M. E. Zolensky et al., Geochim. Cosmochim. Acta 61, is through an Announcement of Opportunity available at https://jaxa-
and TD2 operations, respectively. ryugu-sample-ao.net. Our particle size and shape measurements are
5099–5115 (1997).
More than 200 pebbles (1 to 10 mm in size) 26. A. E. Rubin, Geochim. Cosmochim. Acta 90, 181–194 listed in tables S1 to S3.
from chambers A and C were individually re- (2012).
SUPPLEMENTARY MATERIALS
moved and observed under an optical micro- 27. R. D. Hanna, R. A. Ketcham, M. E. Zolensky, W. M. Behr,
Geochim. Cosmochim. Acta 171, 256–282 (2015). science.org/doi/10.1126/science.abj8624
scope (Fig. 6). These pebbles show morphological Materials and Methods
28. T. Nakamura et al., Irradiation-energy dependence of the
variations: Grains with rugged surface and with spectral changes of hydrous C-type asteroids based on 4 keV Figs. S1 to S7
smooth surfaces are observed (Fig. 6), which and 20 keV He exposure experiments of Murchison CM Tables S1 to S3
chondrite, 51st Lunar and Planetary Science Conference, 16 to References (34–41)
is consistent with the flying pebbles observed Movies S1 to S4
20 March 2020, The Woodlands, TX, Abstract 1310.
during the TD1 and TD2 operations. Although 29. M. Setoh, A. M. Nakamura, N. Hirata, K. Hiraoka, M. Arakawa, 7 June 2021; accepted 25 January 2022
only a few particles have been measured, the Earth Planets Space 59, 319–324 (2007). Published online 10 February 2022
elongated block-like pebbles in the collected 30. N. Sakatani et al., Nat. Astron. 5, 766–774 (2021). 10.1126/science.abj8624

Tachibana et al., Science 375, 1011–1016 (2022) 4 March 2022 6 of 6

 Pebbles and sand on asteroid (162173) Ryugu: In situ observation and particles
returned to Earth
S. TachibanaH. SawadaR. OkazakiY. TakanoK. SakamotoY. N. MiuraC. OkamotoH. YanoS. YamanouchiP. MichelY.
ZhangS. SchwartzF. ThuilletH. YurimotoT. NakamuraT. NoguchiH. YabutaH. NaraokaA. TsuchiyamaN. ImaeK.
KurosawaA. M. NakamuraK. OgawaS. SugitaT. MorotaR. HondaS. KamedaE. TatsumiY. ChoK. YoshiokaY. YokotaM.
HayakawaM. MatsuokaN. SakataniM. YamadaT. KouyamaH. SuzukiC. HondaT. YoshimitsuT. KubotaH. DemuraT.
YadaM. NishimuraK. YogataA. NakatoM. YoshitakeA. I. SuzukiS. FuruyaK. HatakedaA. MiyazakiK. KumagaiT. OkadaM.
AbeT. UsuiT. R. IrelandM. FujimotoT. YamadaM. ArakawaH. C. Connolly Jr.A. FujiiS. HasegawaN. HirataN. HirataC.
HiroseS. HosodaY. IijimaH. IkedaM. IshiguroY. IshiharaT. IwataS. KikuchiK. KitazatoD. S. LaurettaG. LibourelB. MartyK.
MatsumotoT. MichikamiY. MimasuA. MiuraO. MoriK. Nakamura-MessengerN. NamikiA. N. NguyenL. R. NittlerH. NodaR.
NoguchiN. OgawaG. OnoM. OzakiH. SenshuT. ShimadaY. ShimakiK. ShiraiS. SoldiniT. TakahashiY. TakeiH. TakeuchiR.
TsukizakiK. WadaY. YamamotoK. YoshikawaK. YumotoM. E. ZolenskyS. NakazawaF. TeruiS. TanakaT. SaikiM.
YoshikawaS. WatanabeY. Tsuda

Science, 375 (6584), • DOI: 10.1126/science.abj8624

Obtaining and returning samples of Ryugu

The Hayabusa2 mission investigated the nearby carbonaceous asteroid Ryugu and collected samples for return to
Earth. Tachibana et al. describe Hayabusa2’s second sample collection, which picked up material that was excavated
from Ryugu’s subsurface by an earlier impact experiment. By analyzing footage from both touchdown events, the
authors determined the morphological properties of small pebbles on the asteroid. After the sample return capsule was
opened on Earth, they found the chambers contain pebbles with properties consistent with the touchdown locations,

Downloaded from https://www.science.org on March 07, 2022

indicating that they are representative samples of Ryugu’s surface and subsurface. —KTS

View the article online

https://www.science.org/doi/10.1126/science.abj8624
Permissions
https://www.science.org/help/reprints-and-permissions

Use of think article is subject to the Terms of service

Science (ISSN ) is published by the American Association for the Advancement of Science. 1200 New York Avenue NW, Washington, DC
20005. The title Science is a registered trademark of AAAS.
Copyright © 2022 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim
to original U.S. Government Works