On the origin of molecular oxygen
in comets
Vianney Taquet, Catherine Walsh, Kenji Furuya, Ewine van Dishoeck
Leiden Observatory, Leiden University
Water in the Universe, 13/04/2016, ESTEC
1
Importance of Molecular oxygen
Importance of Molecular Oxygen:
- Dominant component of Earth’s atmosphere (21 % by volume)
- Byproduct of photo-synthesis → Potential marker for biological activity
- O is the third most abundant element → O2 potentially abundant in the ISM
- O2 is a key molecule for the water chemical network
2
O2: an elusive interstellar molecule
Zero electric dipole moment → difficult to detect O2 in the cold ISM
Only two detections in dark clouds:
Orion and ρ Oph A
etection of O toward ρ Ophiuchi A
cof
s).
5.
′′
;
nd
2
(Goldsmith et al. 2011; Liseau et al. 2012)
ρ Oph A
Fig. 4. The WBS spectra toward the positions O1 to O4 are shown
for both the 774 GHz (upper) and 487 GHz (lower) lines in each
frame. The average 774
to the 487 GHz
2 GHz
obs data are convolved
2 mod
′′
beam of 44 . Offsets relative to the origin O4 are shown along
the upper and right-hand scales.
→ X(O )
<< X(O )
line is due to the C17 O (7−6) transition at 786.3 GHz (Larsson
No detection toward protostars:
Fig. 2. Full
spectrum
taken with
HIFI, with
the H- and V-polarization spectr
strong
upper
limit
in NGC1333-IRAS4A
left to right. The entire bandwidth is-95.35 GHz. The O2 line is centred near V
X(O2) < 6 x 10 (Yildiz et al. 2013)
Fig. 3. Spectrum of Fig. 2 magnified around the O2 33 –12 line. The blue
dashed line indicates the LSR velocity of the IRAS 4A envelope at
7.0 km s−1 and the red dashed line shows the velocity at 8.0 km s−1 .
+ No clear detection in comets either
the “Water remote
in Star-forming
regions with
Herschel” (WISH)
through
or in-situ
observations
guaranteed-time key program (van Dishoeck et al. 2011) in a
beam size of 40′′ and reported in Yıldız et al. (2012). Beam efficiencies are 0.77, 0.63, and 0.76 for the 1–0, 3–2, and 5–4 lines,
respectively. The calibration uncertainty for HIFI band 1a is
15%, whereas it is 20% for the IRAM 30 m and JCMT lines.
3
The HIFI beam size at 487 GHz of ∼44′′ corresponds to a
F
ca
li
(H
3
A
O
so
fr
ound to contain a rich array of other molecules, including at 10 km from the nucleus centre. Black symbols are data from 1
c compounds and complex hydrocarbons. Molecular oxygen September 2014 to 31 March 2015. For both periods, O2 clearly shows
however, despite its detection on other icy2bodies such as the the strongest correlation with H2O. While CO shows a high correlation
of Jupiter and Saturn2,3, has remained undetected in comet- with H2O from 17 to 23 October 2014, the correlation for the whole
mas. Here we report in situ measurement of O2 in the coma data set is fairly low. N2 shows the weakest correlation with H2O of all
met 67P/Churyumov–Gerasimenko,
with local abundances
The strong
between H2O and O2, with a
First detection of molecular
oxygenthree
in species.
a comet
by correlation
Rosetta:
g from one per cent to ten per cent relative to H2O and with Pearson correlation coefficient of 0.88 (and even 0.97 for the
n value of 3.80 6 0.85 per cent. Our observations indicate October data), indicates that they are of similar origin in the nucleus
e O2-/HHigh
is isotropic in theof
coma
and does
not change anddistinction
that their release of
mechanisms
are linked,
in contrast
to COof
and
the
ROSINA-DFMS
O2 from
other
species
2O ratio resolution
atically with heliocentric distance. This suggests that primsame mass 32
O2 was incorporated into the nucleus during the comet’s
1 August 2014 (pre-encounter)
tion, which is unexpected given the low upper limits from
O2
18 June (after thruster firing)
4
e sensing
observations
.
Current
Solar
System
formation
- X(O2) / X(H2O) = 3.8 ± 0.9 %
30 km orbit, 11 September
100
s do not predict conditions that would allow this to occur.
20 km orbit, 1 October
→
forth
most
abundant
molecule
10 km orbit, 22 October
surements of the coma of 67P/Churyumov–Gerasimenko
ter 67P) were made between September 2014 and March
N18O
CH3OH
5
with the ROSINA-DFMS mass spectrometer on board the
10–1
S
+ Re-analysis
Giotto
a spacecraft.
For the present of
study,the
we analysed
3,193 data
mass
H15NO
taken
in
this
time
period.
Because
of
the
high
mass
resolving
toward 1P/Halley results in a similar
and sensitivity of ROSINA-DFMS, it was possible to differentiabundance
/ Hspecies
2O ! present in the
H2NO
ambiguously
between of
the 3.7
three %
main
10–2
(Rubin
et al. at2015)
w mass
range centred
mass-to-charge (m/z) ratio 32 Da/e,
N2H4
y, molecular oxygen (O2), atomic sulfur (S) and methanol
H); such differentiation has not been achieved by previous in
remote sensing measurements at comets. Figure 1 shows several
10–3
measurements centred at the O2 peak. The green and orange
how data taken before the close encounter with 67P. Only minor
ures from the tenuous neutral gas atmosphere of the Rosetta
raft can be identified, and even after long thruster firing man10–4
s, which use N2O4 as an oxidizer, the contamination of the O2
emains small (see the green line in Fig. 1). Measurements taken
rbiting 67P, shown as the light blue, dark blue and purple lines
31.94
31.96
31.98
32.00
32.02
32.04
32.06
1, show a clear increase of the O2 signal, indicative of cometary
m/z (Da/e)
Bieler
et
al.
(2015)
ese three measurements were taken at decreasing distances (r)
Figure 1 | DFMS mass spectra around 32 Da/e normalized to the spectrum
he comet nucleus, and follow the predicted 1/r2 signal depend- with the largest signal. The black labels indicate the three major species 4
Intensity (arbitrary units)
Detection of O in the comet 67P
b, H2O and CO; c, H2O and N2. All three panels share a common y axis.
Numbers on x and y axes are proportional to number density but in
arbitrary units. Red crosses mark a subset of data for which N2 data are also
available. Panel a shows the strong correlation between H2O and O2, which is
over time, which leads to a low overall correlation between those two species.
N2 has the lowest correlation with H2O of the compared species for the
October data (c).
Abundant O2 trapped into water ice matrix
N2 which have a similar volatility but do not show a strong correlation surface. Taking account of 67P’s continuous mass loss through outwith H2O (see Fig. 2 for correlation coefficient values). The O2/H2O gassing, we estimate the actively outgassing surface areas to be lost to a
ratio decreases for high H2O abundances, which might be caused by depth of several centimetres over the time from August 2014 to March
surface water ice produced by a cyclic sublimation–condensation pro- 2015. If recent production by radiolysis or photolysis (only affecting
LETTER
data
cess8, although the total amount of surface ice is limited9.
the top few micrometres) were the source of the measured
O2, ourRESEARCH
A plausible mechanism for the strong O2/H2O correlation would be would show a continuous decrease of the O2/H2O ratio over the examthe production of O2 by radiolysis or2photolysis of water
2 ice. Here we ined time period as the active surface continues to be shed over that
related
follow the convention
that
refers to ultraviolet
photons
Full photolysis
data set
Full data settime. Apart from the
17–23 Oct.
2014to H2O abundance, Fig. 3
a
b
c variations
17–23
Oct.
2014
17–23
Oct.
2014
breaking bonds,
radiolysis refers to more energetic photons shows that we do not observe a systematic change in the O2/H2O ratio
27
10whereas
or fast electrons and ions depositing energy into the ice and ionizing over several months. Instantaneous creation of the measured O2 by
molecules. Creation of sputtered O2 by radiolysis has been demon- radiolysis or photolysis seems, overall, unlikely, and would lead to
strated in laboratory
experiments10 and is observed for the icy moons variable O2 ratios due to different illumination conditions. Given that
26
10
of Jupiter—Europa, Ganymede and Callisto11–13—as well as for the radiolysis and photolysis, on any of the discussed timescales, do not
rings of Saturn3. Comets are subject to radiolysis over various time- seem to be plausible production mechanisms, the preferred explanascales: (a) over billions of years, while they reside in the Kuiper belt; (b) tion of our observations is the incorporation of primordial O2 into the
25
over the period10of a few
System; cometary nucleus.
R =years
0.88 once they enter the inner Solar
R = 0.60
Despite great efforts by
sensing campaigns, information on
and (c) on very shortRtimescales,
as for the present radiolysis.
In the
R =remote
0.71
= 0.97
R = 0.90
Kuiper belt, the skin depth for producing O2 is in the range of metres, primordial O2 is still limited. Solid O2 has not yet been detected in
1024
although the produced
O2 may diffuse deeper into the porous nucleus. interstellar ices, and upper limits for the O2/H2O ice ratios of ,0.5
Once a comet begins its residence in the inner Solar System, it loses its and for O2/CO ratios of ,1 are in agreement with our findings, but such
surface material to a depth of several metres during each orbit around high upper limits do not provide useful constraints14,15. Gaseous O2 has
the Sun, therefore
1023we can safely assume that no O2 from radiolysis in only been detected in two interstellar clouds so far4,16,17, and is generally
23
24
23
41022
1021remains
1022in 67P
10at
1024
1025
1022Radiolysis
1023 known
10to
1025 1020 low abundances
1021
. Reports of10very low upper
the Kuiper belt phase
the percentage
level.
have surprisingly
CO for O in a protostellar envelope suggest
N2
2
the material infalling to the
and photolysis by solar wind and Oultraviolet
radiation in the inner limits
2
Solar System only affect the top few micrometres of the cometary accretion disk is very poor in molecular oxygen18. This has been ascribed
Figure 2 | Correlation between H2O and O2, CO and N2. a, H2O and O2;
observed for all data. In contrast, the correlation of CO with H2O (b) varies
2
2
b, H2O and CO; c, H2O and N2. All three panels share a common y axis.
over time, which leads to a low overall correlation between those two species.2
0.15
N2 has the lowest correlation with H2O of the compared species for the
Numbers on x and y axes are proportional to number density but in
October data (c).
arbitrary units. Red crosses mark a subset of data for which N2 data are also
available. Panel a shows the strong correlation between H2O and O2, which is
O2 is trapped into a likely pristine water ice matrix
H2O
Strong correlation of O and H O suggests similar spatial origin and mechanisms
The O /H O abundance ratio remains roughly constant over time (3.8 ± 0.9 % / H O)
0.10
O2/H2O
N2 which have a similar volatility but do not show a strong correlation surface. Taking account of 67P’s continuous mass loss through out0.05
coefficient values). The O2/H2O gassing, we estimate the actively outgassing surface areas to be lost to a
with H2O (see Fig. 2 for correlation
ratio decreases for high H2O abundances, which might be caused by depth of several centimetres over the time from August 2014 to March
surface water ice produced by a cyclic sublimation–condensation pro- 2015. If recent production by radiolysis or photolysis (only affecting
0.00 of surface ice is limited9.
cess8, although the total amount
the top few micrometres) were the source of the measured O2, our data
A plausible mechanism for the strong O2/H2O correlation would be would show a continuous decrease of the O2/H2O ratio over the examthe production of O2 by radiolysis or photolysis of water ice. Here we ined time period as the active surface continues to be shed over that
–0.05
Apart
the2015
variations
related to H2O abundance, Fig. 3
the(2015)
convention that photolysis Oct.
refers
2014to ultraviolet
Nov. 2014 photons
Dec. 2014time. Jan.
2015 from Feb.
Mar. 2015
Bieler follow
et al.
shows
that we do not observe a systematic change in the O2/H2O ratio
breaking
bonds,
radiolysis refers to more energetic photons of
the O2 ratio from October to the end of December 2014 can be attributed
Figure
3 | Owhereas
2/H2O ratio over several months. There seems to be no
over
several
months.
Instantaneous
creation
thecometary
measured O2 by
or fast electrons
and ions
depositing
into The
the variances
ice and happen
ionizing
systematic increase
or decrease
of the energy
O2/H2O ratio.
orbital
changes
of the spacecraft
or to physical
changesof
of the
on to
2 molecules.
nucleus.
very short
timescales
and can be explained
by the decrease
the Odemonradiolysis or photolysis seems, overall, unlikely, and would lead to
hasofbeen
Creation
of sputtered
O2 by radiolysis
2 ratio
→ O was likely already present in the ice mantle prior to comet formation5
10 fully understood if the higher variability
for high H O abundances. It is not
Interstellar chemistry of O2
O2 is involved in the chemical network forming interstellar (icy) water
O2 formation and survival in interstellar ices depend on:
1) Gas phase abundance of H and O atoms
T
O2
H
OH
Ion-Molecule
H+
→ O2 production should be
accompanied by O3, HO2 and
H2O2 but ROSINA measured low
abundances in 67P:
- X(HO2) ~ X(H2O2) = 6 10-4 / O2
- X(O3) < 3 10-5 / O2
~10 000
~3 000
~2 000
0
CO
H+3
HCO+
O OH
T
+
EA [K]
s-O2
hv
2) Mobility of O atoms
3) Activation barriers of key reactions
Surface
Neutral Neutral
O
O
H
H
s-O3
s-O
hv
H+3
O O
H2
H
hv
hv H
H2
H
s-HO2
H, H2
OH+
H2
e-
H2O+
e-
T
OH
H
hv
s-OH
hv
H
s-H2O2
H
H2
hv
H H2
H2
H2O
HCO+
H3O+
-
H2O
T
hv
s-H2O
e
Gas Phase
Grain Surface
Credit: M. Persson
6
Objectives of this work
Give an explanation of the observations of 67P/C-G by Rosetta/ROSINA:
1) Primordial formation of O2 prior to comet formation
2) High abundance of 3.8 % of O2 relative to water but low abundance of the
chemically related species HO2, H2O2, and O3 (lower than 6 x 10-4 / O2)
3) Strong correlation between O2 and H2O signals but weak correlation
between N2, CO and H2O signals
Exploration of three different scenarios:
1) O2 formation and survival in molecular clouds
2) O2 formation and survival during protoplanetary disk formation
3) O2 formation and survival within protoplanetary disks
→ Two multi-phase (bulk, surface, gas) astrochemical models are used to
study the cold and warm gas-grain chemistries
(Taquet et al. 2014, Furuya et al. 2015)
7
O2 formation in dark clouds ?
A&A proofs: manuscript no. O2_aa_v9
Table 2. List of model grid input parameters.
Multi-phase gas-grain model applied
to a parameter approach via a grid
of models:
Parameter
Total density (cm−3 )
Temperature (K)
ζ (s−1 )
AV (mag)
Ed /Eb
Eb (O) (K)
Ea (O+O2 ) (K)
Ea (H+O2 ) (K)
Ea (H+H2 O2 ) (K)
Range of explored values
Physical conditions explored in the model grids
103 − 104 − 105 − 106
10 - 15 - 20 -25 - 30
10−18 - 3 × 10−18 - 10−17 - 3 × 10−17 - 10−16
2 -4 - 6 - 8 - 10
Chemical parameters explored for the ρ Oph A case
0.3 - 0.8
800 - 1700
0 - 300
0 - 1200
0 - 2500
An intermediate temperature of 20 K is also favoured because it enhances the mobility of oxygen atoms on the grain
surfaces whilst at the same time allowing the efficient evaporation of atomic H. This additionally enhances the rate of the
oxygen recombination reaction forming O2 , with respect to the
competing hydrogenation reactions described above. In addition,
because the density of gas-phase H atoms increases linearly with
the cosmic-ray ionisation rate, ζ, a low value of ζ also tends to
favour the survival of O2 ice (see the next Section). On the other
hand, the visual extinction does not seem to have a strong impact
on the abundance of solid O2 as the distributions of abundances
obtained for the five visual extinction values are very similar,
showing that the final abundances are more strongly dependent
upon the assumed gas density, temperature, and cosmic-ray ionisation rate.
3.2. The ρ Oph A case
([H]/[O]ini " 3 × 10−2 ) in
rates (" 10−16 s1 ). as it incr
to H2 O ice. For ζ lower tha
in the gas phase is dominate
ing O atoms. The gas phase
the abundance of O2 and ot
longer influenced by ζ. The
abundance of O2 , at a level
not only requires a high de
higher temperature (T ∼
found in dark clouds, but al
atomic form (X(O) > 10−4
tion rate (ζ < 10−16 s−1 ), w
abundance of the chemicall
chemical desorption upon s
Figure 5 shows the chem
for the model using the ph
(nH ∼ 106 cm−3 and T =
tion rate and the surface a
actions that best reproduc
8
standard surface parameter
ρ Oph highly
A core, located
at a distance of
120 physical
pc, constitutes theconditions
→ Abundances in interstellarThe
ices
depend
on
best test case for the water surface network and the production
of O2 in dark clouds. This is because, to date, it is the only inter-
O2 formation in dark clouds ?
A&A proofs: manuscript no. O2_aa_v9
Table 2. List of model grid input parameters.
Multi-phase gas-grain model applied
to a parameter approach via a grid
of models:
Parameter
Total density (cm−3 )
Temperature (K)
ζ (s−1 )
AV (mag)
Ed /Eb
Eb (O) (K)
Ea (O+O2 ) (K)
Ea (H+O2 ) (K)
Ea (H+H2 O2 ) (K)
Range of explored values
Physical conditions explored in the model grids
103 − 104 − 105 − 106
10 - 15 - 20 -25 - 30
10−18 - 3 × 10−18 - 10−17 - 3 × 10−17 - 10−16
2 -4 - 6 - 8 - 10
Chemical parameters explored for the ρ Oph A case
0.3 - 0.8
800 - 1700
0 - 300
0 - 1200
0 - 2500
An intermediate temperature of 20 K is also favoured because it enhances the mobility of oxygen atoms on the grain
surfaces whilst at the same time allowing the efficient evaporation of atomic H. This additionally enhances the rate of the
oxygen recombination reaction forming O2 , with respect to the
competing hydrogenation reactions described above. In addition,
because the density of gas-phase H atoms increases linearly with
the cosmic-ray ionisation rate, ζ, a low value of ζ also tends to
favour the survival of O2 ice (see the next Section). On the other
hand, the visual extinction does not seem to have a strong impact
on the abundance of solid O2 as the distributions of abundances
obtained for the five visual extinction values are very similar,
showing that the final abundances are more strongly dependent
upon the assumed gas density,2temperature, and cosmic-ray ionisation rate.
([H]/[O]ini " 3 × 10−2 ) in
rates (" 10−16 s1 ). as it incr
to H2 O ice. For ζ lower tha
in the gas phase is dominate
ing O atoms. The gas phase
the abundance of O2 and ot
longer influenced by ζ. The
abundance of O2 , at a level
not only requires a high de
higher temperature (T ∼
found in dark clouds, but al
atomic form (X(O) > 10−4
tion rate (ζ < 10−16 s−1 ), w
abundance of the chemicall
chemical desorption upon
6 s
Figure 5Hshows the chem
for the model using the ph
(nH ∼ 106 cm−3 and T =
tion rate and the surface a
actions that best reproduc
9
standard surface parameter
→ O abundance derived in 67P can
be reproduced with dense (n = 10
3.2. The ρ Oph A case
cm-3) and lukewarm (T = 20 K)
The ρ Oph A core, located
at a distance of 120 pc, constitutes the
conditions
best test case for the water surface network and the production
of O2 in dark clouds. This is because, to date, it is the only inter-
Chemical properties in ices
Physical conditions needed to reproduce the O2 abundance seen in 67P
are consistent with those of ρ Oph A (nH = 106 cm-3; Tdust = 21 K):
- Low X(O3) and X(HO2) in 67P reproduced with Ea(O2+O) ~ Ea(O2+H) ~ 300 K
→ consistent with Monte-Carlo models by Lamberts et al. (2013)
(but H2O2 still overproduced by a factor of 10)
- O2 is trapped into the inner part of
the ices, unlike CO (and N2) which are
mostly present at the surface
→ Explanation of the correlation
between O2 and H2O signals seen in 67P,
and anticorrelation for CO and N2
CO2
H 2O
O2
CO
H2O2
HO2
10
O2 formation during the disk formation ?
Multi-phase astrochemical model applied to a 2D semi-analytical model
of core contraction (Visser et al. 2009)
X(O2)ini = 5 %
X(O2)ini = 0 %
-4
10
8
-6
6
-8
4
-10
2
-1
6
-2
4
-3
2
z [AU]
80
50
(d) log(O2 gas [nH])
Ω=
10-13
s-1
10
-4
100
60
-8
40
-10
20
20 30 40
R [AU]
(e) log(ice O2/H2O)
40
-3
20
-4
-5
-12
50 100 150 200 250 300
R [AU]
6
-2.0
4
50 100 150 200 250 300
R [AU]
-2.5
2
-3.0
1
0
-2
-1.0
-1.5
10
-1
60
(c) log(ice O2/H2O)
8
50
80
-6
z [AU]
100
20 30 40
R [AU]
10
-4
-5
-12
10
1
0
z [AU]
z [AU]
s-1
(b) log(ice O2/H2O)
100
20 30 40
R [AU]
50
(f) log(ice O2/H2O)
80
z [AU]
Ω=
8
10-14
z [AU]
10
(a) log(O2 gas [nH])
-1.0
-1.5
60
-2.0
40
-2.5
20
-3.0
50 100 150 200 250 300
R [AU]
- Efficient formation of O2 vapour, but no production of O2 ice into water matrix
- O2 ice formed in ISM can survive during its journey to the disk
11
O2 formation in protoplanetary disks ?
Can O2 be formed and trapped into water ice in the solar nebula ?
→ Gas phase formation of O2 is only efficient in the upper layers of the disk
20
20
-4
-4
(b) log([H2O ice])
15
-5
10
-6
-7
5
-7
-8
0
-8
15
-5
10
-6
5
0
0
10
20
30
R [AU]
40
50
Z [AU]
Z [AU]
(a) log([O2 gas])
0
10
20
30
40
50
R [AU]
Can O2 be formed during luminosity outbursts and trapped with water
A&A proofs: manuscript no. O2_aa_v9
during the cooling
?
- Luminosity outbursts are too short
to produce O2 in quantities observed in
comets
- CO and O2 are trapped together
during the cooling
→ cannot explain the correlation of O2
and non-correlation of CO/N2 with
water
mposition
of each monolayer
10 luminosity
outbursts
for the six models considered here. The standard
ng
the 10 luminosity
outburstswithin
for theices
six during
modelsthe
considered
here. The
standard
12
Conclusions
High abundance of O2 trapped into water ice observed in 67P by
Rosetta/ROSINA can be explained by:
- an efficient formation in dense and lukewarm molecular clouds
- a survival of the O2-H2O ice mixture in the solar nebula
→ consistent with some properties of our Solar System, suggesting that
it was born in a dense cluster of stars (see Adams 2010)
13
Conclusions
High abundance of O2 trapped into water ice observed in 67P by
Rosetta/ROSINA can be explained by:
- an efficient formation in dense and lukewarm molecular clouds
- a survival of the O2-H2O ice mixture in the solar nebula
→ consistent with some properties of our Solar System, suggesting that
it was born in a dense cluster of stars (see Adams 2010)
Thank you !
14