Surface Formation Routes of Interstellar Molecules A Laboratory Study

Surface Formation Routes of Interstellar Molecules A Laboratory Study
Surface Formation Routes of Interstellar Molecules
A Laboratory Study
Surface Formation Routes of Interstellar Molecules - A Laboratory Study
Sergio Ioppolo
Thesis Universiteit Leiden - Illustrated - With summary in Dutch - With references
ISBN/EAN 978-90-9025854-6
Printed by Ipskamp Drukkers
Cover by Petra Vacková - [email protected]
Surface Formation Routes of Interstellar Molecules
A Laboratory Study
PROEFSCHRIFT
ter verkrijging van
de graad van Doctor aan de Universiteit Leiden,
op gezag van de Rector Magnificus prof.mr. P.F. van der Heijden,
volgens besluit van het College voor Promoties
te verdedigen op donderdag 9 december 2010
klokke 15.00 uur
door
Sergio Ioppolo
geboren te Catania, Italië
in 1980
Promotiecommissie
Promotores:
Co-promotor:
Overige Leden:
Prof. dr. H. V. J. Linnartz
Prof. dr. E. F. van Dishoeck
Dr. H. M. Cuppen
Prof. dr. K. Kuijken
Prof. dr. A. G. G. M. Tielens
Prof. dr. E. Herbst
Dr. W. Brown
Dr. M. E. Palumbo
Ohio State University
University College London
Catania University
Contents
1
2
Introduction
1.1 The interstellar medium . . . . . . . . . . . . .
1.2 The cycle of matter . . . . . . . . . . . . . . .
1.3 Interstellar ices . . . . . . . . . . . . . . . . .
1.3.1 Interstellar ice composition . . . . . . .
1.3.2 Interstellar ice chemistry . . . . . . . .
1.4 Laboratory ices . . . . . . . . . . . . . . . . .
1.4.1 RAIR spectroscopy . . . . . . . . . . .
1.4.2 Mass spectrometry . . . . . . . . . . .
1.4.3 Experimental setups used in this thesis .
1.5 This thesis . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
2
6
6
9
12
15
17
18
20
Hydrogenation reactions in interstellar CO ice analogues
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . .
2.2 Experimental procedure . . . . . . . . . . . . . . . . .
2.3 Experimental results . . . . . . . . . . . . . . . . . .
2.3.1 A sample experiment . . . . . . . . . . . . . .
2.3.2 Flux dependence . . . . . . . . . . . . . . . .
2.3.3 Thickness dependence . . . . . . . . . . . . .
2.3.4 Temperature dependence . . . . . . . . . . . .
2.4 Monte Carlo simulations . . . . . . . . . . . . . . . .
2.4.1 The method . . . . . . . . . . . . . . . . . . .
2.4.2 The CO ice layer . . . . . . . . . . . . . . . .
2.4.3 Comparison to the experiment . . . . . . . . .
2.4.4 Effect of diffusion . . . . . . . . . . . . . . .
2.4.5 Effect of H2 molecules on the hydrogenation .
2.5 CO hydrogenation under interstellar conditions . . . .
2.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
23
24
25
27
27
31
32
33
34
34
35
36
37
37
38
42
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
V
Contents
3
4
5
6
VI
Laboratory Evidence for Efficient Water Formation in Interstellar Ices
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4 Determining the reaction rates . . . . . . . . . . . . . . . . . . . . .
3.5 Astrophysical discussion and conclusion . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
49
50
51
52
53
58
Water formation at low temperatures by surface O2 hydrogenation I
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2 Experimental and data analysis . . . . . . . . . . . . . . . . . . .
4.2.1 Experimental . . . . . . . . . . . . . . . . . . . . . . . .
4.2.2 Data analysis . . . . . . . . . . . . . . . . . . . . . . . .
4.2.3 Control experiments . . . . . . . . . . . . . . . . . . . .
4.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . .
4.3.1 Temperature dependence . . . . . . . . . . . . . . . . . .
4.3.2 Structural effect . . . . . . . . . . . . . . . . . . . . . . .
4.3.3 Penetration mechanism . . . . . . . . . . . . . . . . . . .
4.3.4 Thickness dependence . . . . . . . . . . . . . . . . . . .
4.3.5 H2 dependence . . . . . . . . . . . . . . . . . . . . . . .
4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
61
62
64
64
66
70
71
71
74
77
79
82
83
Water formation at low temperatures by surface O2 hydrogenation II
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2 Experimental and data analysis . . . . . . . . . . . . . . . . . . . .
5.2.1 Experimental . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.2 Data analysis . . . . . . . . . . . . . . . . . . . . . . . . .
5.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3.1 H/O2 ratio dependence . . . . . . . . . . . . . . . . . . . .
5.3.2 O3 detection . . . . . . . . . . . . . . . . . . . . . . . . .
5.3.3 Time/fluence dependence . . . . . . . . . . . . . . . . . . .
5.4 Implications for the reaction network . . . . . . . . . . . . . . . . .
5.4.1 Co-deposition experiments . . . . . . . . . . . . . . . . . .
5.4.2 Hydrogenation of H2 O2 . . . . . . . . . . . . . . . . . . .
5.4.3 The role of H2 . . . . . . . . . . . . . . . . . . . . . . . .
5.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
87
88
89
89
90
93
93
94
97
99
100
102
105
106
Water formation by surface O3 hydrogenation
6.1 Introduction . . . . . . . . . . . . . . . . .
6.2 Experimental procedure . . . . . . . . . . .
6.3 Results and discussion . . . . . . . . . . .
6.3.1 Temperature dependence . . . . . .
6.3.2 H/D-atom flux dependence . . . . .
6.3.3 Possible reaction pathways . . . . .
6.4 Conclusion . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
109
110
112
114
114
116
118
122
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Contents
7
8
9
CO + H vs. O2 + H and formation of CO2
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . .
7.2 Experimental details . . . . . . . . . . . . . . . . .
7.3 Data analysis . . . . . . . . . . . . . . . . . . . . .
7.4 Results and discussion . . . . . . . . . . . . . . . .
7.4.1 Hydrogenation of O2 molecules . . . . . . .
7.4.2 Hydrogenation of CO molecules . . . . . . .
7.4.3 Competition between the CO and O2 channel
7.4.4 Formation of solid CO2 . . . . . . . . . . . .
7.5 Astrophysical implications . . . . . . . . . . . . . .
7.6 Conclusions . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
123
124
127
128
129
129
131
132
132
134
135
Surface formation of HCOOH at low temperature
8.1 Introduction . . . . . . . . . . . . . . . . . . .
8.2 Experimental procedure . . . . . . . . . . . . .
8.3 Results and discussion . . . . . . . . . . . . .
8.3.1 Formation of solid HCOOH . . . . . .
8.3.2 Formation temperature . . . . . . . . .
8.3.3 Possible reaction pathways . . . . . . .
8.3.4 Branching ratio of reaction HO-CO + H
8.4 Astrophysical implications . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
137
138
139
140
140
142
144
148
149
Formation of interstellar solid CO2 after energetic processing
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . .
9.2 Experimental procedure . . . . . . . . . . . . . . . . . . .
9.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.3.1 Irradiation of CO bearing mixtures . . . . . . . . .
9.3.2 Irradiation of ice mixtures without CO . . . . . . .
9.3.3 Carbon grains with a water ice cap . . . . . . . . .
9.3.4 Band profiles . . . . . . . . . . . . . . . . . . . .
9.4 Comparison with observations . . . . . . . . . . . . . . .
9.4.1 CO2 towards massive YSOs . . . . . . . . . . . .
9.4.2 The fitting procedure . . . . . . . . . . . . . . . .
9.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . .
9.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
151
152
153
156
156
159
161
163
165
165
166
171
173
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Bibliography
175
Nederlandse Samenvatting
183
Publications
189
Curriculum Vitae
191
Acknowledgements
193
VII
CHAPTER 1
Introduction
1.1 The interstellar medium
Our Galaxy is largely empty. By terrestrial standards the space between the stars can
be considered as a near-perfect vacuum: the average particle density in the solar neighborhood is roughly a factor of 1019 less than in the terrestrial atmosphere at sea level.
However, the highly diluted material present between the stars, the so-called InterStellar
Medium (ISM), plays a central role in the chemical evolution of the Galaxy. The ISM is
the repository of ashes from previous generations of stars and it is itself the birthplace of
new stars and planetary systems. The interstellar matter consists of about 99% gas and
1% (sub)micron size silicate and carbonaceous dust grains by mass. The interstellar gas
is composed of roughly 89% hydrogen, 9% helium and 2% heavier elements. The gas in
the ISM is found in a variety of phases: coronal gas, ionized gas, neutral atomic gas and
molecular gas (Tielens 2005). The physical properties of these phases are summarized
in Table 1.1. Hot ionized gas is observed in X-ray emission and as UV absorption lines
of highly ionized atoms (e.g., C IV, N V, O VI). It is present in the coronal gas of stars
and composes the Hot Ionized Medium (HIM), a hot and tenuous phase of the ISM. The
gas in these regions is heated and ionized through shocks driven by stellar winds from
early type stars and by supernova explosions. Diffuse ionized gas is observed mainly
through emission in the Hα recombination line and resides in both a diffuse component,
the so-called Warm Ionized Medium (WIM), as well as in the classical H II regions surrounding hot O and B stars. Neutral atomic gas is traced though the 21 cm line of atomic
hydrogen, and is found in the Warm Neutral Medium (WNM) and Cold Neutral Medium
(CNM). Molecular gas represents the densest component of the ISM. Although H2 is the
dominant molecular species in space followed by CO (H2 /CO = 104 − 105 ), it cannot be
routinely detected in cold gas by infrared telescopes because of the lack of dipole allowed
transitions. Therefore, molecular gas is commonly traced through the CO rotational transitions, such as J = 1−0 at 2.6 mm. Molecular gas is localized in discrete Giant Molecular
Clouds (GMC), which tend to be irregularly shaped with a density distribution far from
homogeneous. Molecular clouds contain embedded cores in which new stars form.
1
1 Introduction
Table 1.1 Main characteristics of interstellar medium components: coronal gas and
hot ionized medium (HIM), H II regions, warm ionized medium (WIM), warm neutral
medium (WNM), cold neutral medium (CNM), and molecular clouds. The values in this
Table are taken from Tielens (2005) and Visser (2009).
Component
Coronal gas
HIM
WIM
H II regions
WNM
CNM
Molecular clouds
Fractional Volume
(%)
Scale Height
(pc)
Temperature
(K)
Density
(cm−3 )
State of
hydrogen
30−70
20−50
<1
10−20
1−5
<1
1000−3000
1000
70
300−400
100−300
70
106 − 107
8000
8000
6000−10000
50−100
10−50
10−4 − 10−2
0.2−0.5
102 − 104
0.2−0.5
20−50
102 − 106
ionized
ionized
ionized
neutral atomic
neutral atomic
molecular
1.2 The cycle of matter
The evolution of gas and dust in the ISM from stellar birth to death can be described as a
cyclic event as shown in Fig. 1.1. The complete cycle takes some 2×109 years. In the first
step, stellar winds and/or explosions enrich the diffuse interstellar medium in dust and
gas. Carbonaceous dust grains are ejected by carbon-rich stars (C/O>1) at high densities
and temperatures, and silicates originate from oxygen-rich star ejecta. Observations combined with models showed that .10% of the interstellar dust mass has a “stardust” origin
(i.e., formation in the atmospheres of evolved stars), and that interstellar grains are mostly
formed in the ISM by some chemical mechanism which is not yet characterized (Draine
2009, and references therein). The interstellar grains pass through the various aforementioned phases of the interstellar medium several times on a typical scale of ∼3×107 years,
before they take part in the formation of a new star. In the intercloud phase, grains can
coagulate leading to an increase in their size, while strong shocks destroy dust and change
its size distribution. Supernova shocks can also trigger the collapse of diffuse clouds into
dense clouds. In the denser phase, atomic gas is converted into simple molecules, like
CO, through gas-phase reactions, while H-rich species such as H2 O, NH3 , and CH4 are
formed on the grains through surface reactions, resulting in a polar ice mantle. For densities higher than 105 cm−3 most of the gas-phase molecules, mainly CO, freeze-out onto
the dust grains on timescales shorter than the lifetime of a core. The resulting apolar ice,
which is typically a few tens of monolayers (ML) thick, is further involved in chemical
reactions. The composition and evolution of interstellar ices are described in more detail
in § 1.3.1.
Molecular clouds are stable over timescales of ∼3×107 years because of a balance
of turbulent and magnetic pressure and gravity. Physically this is expressed by the virial
theorem, which states that, to maintain equilibrium, the gravitational potential energy
must equal twice the internal thermal energy. The gravitational collapse of the core is
initiated by the loss of turbulent or magnetic support. As it collapses, a molecular cloud
can fragment into smaller pieces. Each of the cold cores collapses in an isothermal manner
2
1.2 The cycle of matter
since the gas (atoms and molecules) releases energy in the form of radiation (Bergin
& Tafalla 2007). While the density increases (105 −107 cm−3 ), the fragments become
optically opaque and are thus less efficient at releasing their energy through radiation.
During the collapse the central cores (i.e., the protostars) gradually warm up and emit
a continuum of infrared radiation. The warm-up of cloud cores (20 K < T < 100 K)
induces the desorption of volatile species, like CO, O2 and N2 , from dust grains and the
segregation of less volatile ones, like CO2 . At such temperatures H atoms are no longer
residing onto the grains, and therefore hydrogenation reactions are no longer a dominant
process. Moreover, the increase in temperature facilitates a higher mobility of the solid
species still frozen on the grains and, therefore, drives a rich grain-surface chemistry.
Figure 1.1 Schematic representation of the lifecycle of cosmic dust. Dust is formed in
stellar ejecta (planetary nebulae, red giants, novae and supernovae) and mainly in the
interstellar medium itself, where it cycles rapidly between the different phases. In the intercloud phase, strong shocks destroy dust and change its size distribution. In the denser
phase, the accretion of gas-phase species onto grains forms an ice mantle, which is exposed to atoms and processed by FUV photons and energetic particles. Coagulation leads
to an increase in the grain size. Ultimately, dust particles are involved in the star and
planet formation process. Figure adapted from Tielens & Allamandola (1987b) and Tielens (2005).
The objects resulting from the cloud collapse are called Young Stellar Objects (YSOs).
At this stage, YSOs are rotating spheres of gas with a central protostar. Stars of different
3
1 Introduction
masses are thought to form by slightly different mechanisms. The process of single lowand intermediate-mass star formation (M? < 8 M ), schematically displayed in Fig. 1.2, is
well understood (e.g., Evans 1999, van Dishoeck 2004, and references therein). Because
of the conservation of angular momentum the collapse of a rotating sphere of gas and dust
(Fig. 1.2b) leads to the formation of an accretion disk through which matter is guided onto
the central protostar (Fig. 1.2c). The temperature in the midplane of the disk drops to 10
K and the density increases to ∼1012 cm−3 . Therefore, gas-phase species will accrete
onto the grains again. Also grains will coagulate, forming most likely larger and larger
boulders, and eventually planets, even though this process is not completely understood
yet. The surface of the disk is affected by a strong UV field from the protostar, the gas
temperature often exceeds 100 K in the outer disk (>1000 K in the inner disk) and the
density is ∼106 cm−3 .
Under such harsh conditions, ices cannot survive and, therefore, only gas-phase species
will participate in the chemistry of the warm inner envelope, known as hot core (Herbst &
van Dishoeck 2009). Hot cores are characterized by a density of 107 −108 cm−3 , a temperature of ∼100 K and size of ∼100 AU. Some excess angular momentum is also dissipated
through large bipolar outflows, launched along the core’s rotation axis near the protostar
(Shu et al. 1991, Bally 2007). The gas present in the circumstellar disk will eventually fall
onto the star, planets and small bodies present in the disk, or will be cleared by irradiation
and winds from the star in a timescale of ∼106 years (Fig. 1.2d-e). For low-mass stars,
the disk will slowly evolve into a planetary system such as our Solar system (Fig. 1.2f).
When the temperature of the protostar is sufficiently high to initiate nuclear burning of
hydrogen into helium, a new star is born.
The formation process of high-mass stars (M? > 8 M ) is not yet fully understood
because of observational problems: they are embedded objects that evolve rapidly to the
main sequence. However, except for the most massive ones, massive stars most likely
form by a mechanism similar to that observed for low-mass stars. It is likely that the
initial phase involves the collapse of infrared dark clouds within giant clouds. High-mass
young stellar objects are associated with ultra- or hypercompact H II regions, masers,
outflows, and/or warm ambient gas at average temperatures of 300 K, known as hot cores
(Herbst & van Dishoeck 2009). The hot cores usually have a size of 104 AU and often are
associated with a rich organic chemistry. It is not yet clear whether disks are formed in
high-mass star regions. If newly formed luminous O and B stars are in the proximity of
cool interstellar matter, their UV radiation heats the neighboring gas and photodissociates
molecules, producing heterogeneous Photon-Dominated Regions (PDRs) (van Dishoeck
2004).
Ongoing and future key new instrumentation, such as the Herschel Space Observatory
and ALMA, will unravel the chemistry and dynamics of star and planet formation with
more detail than possible so far. This will allow us to resolve the physical processes
taking place during the collapse of molecular clouds, to image the structure of protostars
and of protoplanetary disks, and to determine the chemical composition of the material
from which future solar systems are made (van Dishoeck & Jørgensen 2008).
The final stages of the star evolution are well understood. Protostars with masses
less than roughly 0.08 M never reach temperatures high enough for nuclear fusion of
4
1.2 The cycle of matter
hydrogen to begin. These are known as brown dwarfs. For 0.08 < M? < 8 M , the star
will spend a long time on the main-sequence phase (107 −1010 yr) and nuclear fusion will
form elements up to C, O and N. After leaving this phase, stars below 0.23 M become
white dwarfs, while more massive stars will move into the Red Giant and Asymptotic
Giant Branch (RGB and AGB) phases and eventually evolve into a planetary nebula with
a white dwarf core. Red giant winds and planetary nebulae are important sources of
gas and dust enrichment in the ISM. In this way the matter cycle of low-mass stars is
complete.
Figure 1.2 Schematic illustration of the different stages of low-mass star formation. Dense
clouds can form when the diffuse medium is compressed by a shock event such as a
supernova explosion (a). Once the density in the core of dense clouds gets high enough
and support is lost, the core starts to collapse (b). Conservation of angular momentum
results in the formation of a circumstellar disk and bipolar outflows (c). The disk will
eventually evolve into a planetary system (d-f). Figure taken from Visser (2009), based
on Hogerheijde (1998).
High-mass stars spend a shorter time along the evolutionary phases, burning elements
up to Fe. Since no further energetically favorable nuclear reactions can occur, the core
will collapse. Depending on the initial mass, the core can become a neutron star, a pulsar
or a black hole. The outer shells of the star will explode in a violent event, a so-called
supernova explosion, which can trigger the nucleosynthesis of elements heavier than Fe.
These shocks can perturb the surrounding ISM and therefore trigger star formation closing
the cycle of matter.
5
1 Introduction
1.3 Interstellar ices
1.3.1 Interstellar ice composition
Rotational, vibrational, and electronic spectroscopy has established the presence of a large
variety of polyatomic molecules, ions and radicals in space, both in the gas phase and in
the solid state. In fact, over 150 different molecular species (excluding isotopomers) have
been detected in the inter- and circumstellar medium. These species include a variety
of inorganic compounds (e.g., H2 O, CO, CO2 , NH3 , SO and SO2 ), organics (e.g., CH4 ,
H2 CO, CH3 OH, HCOOH, and CH3 CH2 OH), ions (e.g., C6 H− ) and species identified only
in the ice like OCN− and NH+4 ), and unsaturated hydrocarbon chains (e.g., HCN, HC3 N).
Recently, large carbonaceous species like the fullerenes C60 and C70 have been unambiguously detected (Cami et al. 2010, Sellgren et al. 2010). However, aromatic species such as
polycyclic aromatic hydrocarbons (PAHs) are not included in the count, since they are not
uniquely identified. PAHs are detected only as a class through their infrared features. The
spectra of molecular species are probes of the physical conditions and chemical history of
the regions in space where they reside. Hence, high-resolution rotational and vibrational
spectra improve our knowledge about the density and temperature of the gas as well as
collapse of interstellar clouds, whereas vibrational spectra of solid phase molecules give
information on the nature of the ice mantles covering dust grains.
The presence of ice in the interstellar medium was already proposed by Eddington
(1937) before its spectral detection, which came almost four decades later: H2 O ice was
detected at 3 µm by Gillett & Forrest (1973). Most ices can be detected in absorption in
the mid-IR of an embedded object or along the line of sight of a background star. However, ground based observations in the mid-IR spectral window are limited because of
telluric absorption. Airborne and space observatories have therefore been used to identify
solid phase species in space in the last decades. In the 1990’s the launch of the Infrared
Space Observatory (ISO) revolutionized our understanding of interstellar ices. Because
of its limited sensitivity, ISO observed mostly bright sources, such as high-mass YSOs
and quiescent dense clouds toward luminous background stars (Gibb et al. 2000b, 2004).
Observations show that H2 O, CO, CO2 , and, in some cases, CH3 OH represent the bulk of
solid-state species in dense molecular clouds and star-forming regions. Minor ice components, such as CH4 , NH3 , H2 CO, HCOOH, SO2 , OCS, OCN− , NH+4 , and HCOO− ,
are also observed, although their identification is sometimes controversial. Completely
isolated single bands from species such as solid CH4 at 7.67 µm and OCS at 4.92 µm
are confidently assigned (see Table 1.2). However, the identification of other species like
H2 CO, HCOOH and NH3 is more problematic because all their strong vibrational modes
overlap with features of other species (Table 1.2; van Dishoeck 2004, and references
therein).
More recently, the Spitzer Space Telescope characterized the molecular content of icy
grain mantles in the 5−35 µm wavelength range towards more than 40 low-mass protostars within the c2d (cores to disks) program (Boogert et al. 2008, Pontoppidan et al.
2008, Öberg et al. 2008, Bottinelli et al. 2010) and dozens more within other programs
6
1.3 Interstellar ices
Table 1.2 Interstellar ice inventory with respect to H2 O ice towards dark clouds, low- and
high-mass YSOs. Selected solid state infrared transitions are also listed.
Species
H2 O
O-H stretch
H-O-H bend
libration
λ
(µm)
3.05
6.0
12
CO
C=O stretch
4.67
26a
20b
8.1c
CO2
C=O stretch
O=C=O bend
4.27
15.3
24d
21.6e
14.1e
C=O stretch
CH deformation
C-O stretch
5.85
7.25
8.1
≤1.4 f
2.7 f
5.2 f
H2 CO
C-H (asym., sym.)
C=O stretch
CH2 scissor
3.47, 3.54
5.81
6.69
...
6.0 f
3.1c
CH3 OH
O-H stretch
C-H stretch
CH3 deformation
CH3 rock
C-O stretch
3.08
3.53
6.85
8.85
9.75
<2.3 f
6.1g
14.7 f
NH3
N-H stretch
deformation
umbrella
2.96
6.16
9.0
≤8d
6.1g
15c
NH+4
deformation
6.85
5.2 f,i
6.3 f,i
8.1 f,i
CH4
C-H stretch
deformation
3.32
7.67
<3d
5.0h
1.5c
OCN−
C≡N stretch
4.62
<2.3d
≤0.6 j
1.9 j
O=C=S stretch
4.92
<0.27c
<0.04k
0.2c
HCOOH
OCS
Mode
Dark cloud
(Elias 16)
100
Low-mass YSO
(HH 46)
100
High-mass YSO
(W33A)
100
a
Chiar et al. (1995); b Boogert et al. (2004); c Gibb et al. (2004); d Knez et al. (2005);
Pontoppidan et al. (2008); f Boogert et al. (2008); g Bottinelli et al. (2010); h Öberg et al.
(2008); i The entire band is assumed to be due to NH+4 ; j van Broekhuizen et al. (2005);
k
This value is taken from another low-mass YSO, Elias 29 (Gibb et al. 2004).
e
7
1 Introduction
Figure 1.3 Spitzer infrared absorption spectrum combined with L and M band groundbased observations of two low-mass embedded stars: B5 IRS1 (top, multiplied by a factor
of 5 for clarification) and HH46 IRS (bottom). Spectra taken from Boogert et al. (2004).
(e.g., Zasowski et al. 2009). These data have been complemented with spectra at 2−4 µm
obtained with ground-based facilities. These surveys provide the opportunity to unambiguously identify solid H2 O, CO2 , CH4 , CH3 OH, and NH3 . Figure 1.3 shows the Spitzer
infrared absorption spectrum combined with L and M band observations of two low-mass
embedded stars. In addition, Spitzer detected ices towards field stars behind quiescent
dark clouds (e.g., Bergin et al. 2005, Knez et al. 2005). Table 1.2 lists ice abundances
with respect to H2 O ice towards high and low-mass protostars and quiescent dark clouds.
The general formation scenario can be summarized in four steps, as shown in Fig. 1.4.
1. In quiescent dark clouds, H2 O-rich ice is formed via surface reactions with a large
amount of CO2 , and traces of CH4 and NH3 ice. Some CH3 OH is also associated
with the H2 O-rich ice (Fig. 1.4a).
2. In the prestellar core, when densities are as high as 105 cm−3 and temperatures
amount to ∼10 K, CO and probably N2 and O2 molecules freeze-out on top of the
H2 O-rich mantle, forming an apolar (water poor) ice layer. Before the turningon of the protostar, a second CO2 formation phase takes place resulting in a CO
dominated CO:CO2 ice mixture. During this phase, also CH3 OH ice is formed
(Fig. 1.4b). Energetic processing, such as UV photolysis and cosmic rays irradiation, contributes to the production of new molecular species.
8
1.3 Interstellar ices
Figure 1.4 Suggested evolution of ices during star formation. Light-blue indicates a H2 Odominated ice and dark-blue a CO-dominated ice. At each cold stage a small amount of
the ice is released non-thermally. Early during cloud formation (a) a H2 O-rich ice forms.
Once a critical density and temperature is reached CO freezes-out catastrophically (b),
providing reactants for the formation of species like CH3 OH. Energetic processing (UV
photolysis and cosmic rays irradiation) of the CO-rich ice results in the production of
complex species. Closer to the protostar (c), following sublimation of CO, other complex molecules become abundant. Finally, all ice close to the protostar >100 K desorbs
thermally (d). Figure adapted from Öberg et al. (2010a).
3. When the luminosity of the forming star increases, the more volatile species in the
ice desorb, H2 O and CO2 segregate, and more complex solid state chemistry is
driven by the strong UV field coming from the young star (Fig. 1.4c).
4. When the temperature exceeds 90 K, the ice mantles evaporate completely and
complex gas-phase chemistry can proceed in the hot cores (Fig. 1.4d).
The next section discusses the origin of the observed interstellar ices.
1.3.2 Interstellar ice chemistry
In quiescent dark clouds, grains provide a surface on which species can accrete, meet and
react and to which they can donate excess energy. Grain surface chemistry is governed
by the accretion rate of the gas phase species onto the grains, the surface migration rate,
which sets the reaction network, and the desorption rate. The timescale at which gas phase
species deplete-out onto grains is ∼105 years in dense cores. This time is shorter than the
lifetime of dense cores, which is between 105 and 106 years. Hence, in dense regions,
during the first stage of star formation virtually all species (except H2 ) are frozen-out
onto interstellar grains.
9
1 Introduction
desorption
Eley−Rideal
accretion
Langmuir−Hinshelwood
Hot atom
hopping
tunneling
Figure 1.5 Three mechanisms for surface reactions on a regular grain surface: LangmuirHinshelwood (diffusive), Eley-Rideal, and hot atom mechanism. Closed circles are thermalized species and open circles non-thermalized. Reaction can occur when two species
are in the same well. Figure provided by H. M. Cuppen.
As adsorbates accrete onto a cold surface, a complex chemistry occurs, leading to
the production of new molecules in the ices through surface reactions (Herbst & van
Dishoeck 2009, and references therein). There are three major mechanisms for surface
reactions: Langmuir-Hinshelwood (diffusive), Eley-Rideal, and hot atom mechanism (see
Fig. 1.5). In a Langmuir-Hinshelwood (LH) reaction the reactants are initially adsorbed
onto the surface and in thermal equilibrium with the solid. Diffusion of one or more of
the reactants on the surface leads to the formation of new products. In an Eley-Rideal
(ER) reaction a particle coming from the gas phase reacts more or less directly with a
surface-adsorbed species. In the hot atom mechanism, a gas-phase species lands on the
surface and moves considerably before thermalization. In this way it is able to collide and
react with an adsorbate.
In dense clouds, the flux of incoming species on a specific grain is typically very low
(∼1 per day). In contrast, on the grain surface, the abundant accreting atoms, H, C, N,
and O, are relatively mobile and they can move around the grain on timescales less than
the timescale for another gas phase species to accrete on the same grain. Moreover, the
residence time of a species on a grain depends on its binding energy and grain temperature. As a result, for a 10 K grain with a site coverage of ∼106 sites, an accreted H atom
is expected to visit all sites on the surface many times before it desorbs. If co-reactants
are present, H atoms can react before the next radical lands (Tielens & Charnley 1997,
Tielens 2005). In considering possible co-reactants for atomic H, one has to distinguish
between radicals and non-radicals. The former generally react with zero activation barriers; i.e., upon collision on the grain surface. The latter may possess appreciable activation
barriers, but, in each collision, there is a small probability that H will tunnel through the
activation barrier and react. The reaction could also proceed thermally, depending on the
residence time of both species (Herbst & van Dishoeck 2009).
Grain surface chemistry dates back to Allen & Robinson (1977), although the first
realistic gas-grain model was proposed by Tielens & Hagen (1982) in 1982. Their astrochemical model includes a complex grain surface reaction network to explain molecule
formation in quiescent dark clouds. According to this model in a first phase, H2 O ice
(the dominant solid phase species) can be produced by the sequential hydrogenation of
10
1.3 Interstellar ices
O atoms landing on the grain (a process first proposed by van de Hulst; for a review see
van de Hulst 1996). Atomic oxygen can also react with other O atoms to form O2 and
O3 . Reaction of O3 with H reforms O2 and OH. O2 can be hydrogenated to form H2 O,
while the OH radical can react with H or H2 to form again H2 O, or it can form CO2 and
H2 with the CO accreted on the surface. Under these conditions, also other H-rich species
such as CH4 and NH3 can be formed. This first phase of grain-surface chemistry results
in the formation of a polar ice mantle (water-rich) onto the dust grains. During the second
phase, when the density increases in the molecular cloud, CO freezes-out onto the grains
(its accretion rate is higher than that of H) forming an apolar ice layer (water-poor) on
top of the polar one. Under these conditions, the hydrogenation of CO ice can lead to
the formation of H2 CO and CH3 OH (Tielens & Hagen 1982, Charnley et al. 1997). CO2
can be formed through the reaction CO + OH (Goumans et al. 2008). As suggested by
Charnley et al. (2001), the reaction between CO and heavier elements than H may lead to
the formation of more complex molecules such as CH3 COH and C2 H5 OH.
All these astrochemical reaction networks were based on chemical intuition and analogues from gas phase routes. It took several decades before experimental techniques
allowed laboratory astrochemists to put all these reactions to the test. This is now done
by several groups across the world and is the main topic of this thesis (Chapters 2-8). In
this experimental process, several of these reactions were proven to be efficient, whereas
others were not. Also several new reaction routes were revealed. Thus, laboratory work
combined with theoretical models plays an essential role in disentangling the astrochemistry of the ISM.
The mathematical treatment of grain-surface reactions is not straightforward because
of the heterogeneity and small size of the grains (Herbst & van Dishoeck 2009). Rate
equations1 are better used to reproduce the diffusive chemistry on homogeneous larger
grains, on which at least several reactive adsorbate species exist on average (Green et al.
2001, Tielens & Hagen 1982). So-called macroscopic stochastic methods such as the
master equation approach2 and the Monte Carlo method 3 eliminate the problem caused
by the small numbers of adsorbates, which is significant on smaller grains, but do not
solve the problem of inhomogeneity (Barzel & Biham 2007, Charnley 2001, Green et al.
2001, Stantcheva et al. 2002). A special Monte Carlo method known as the “continuoustime, random-walk” approach handles this latter problem (Chang et al. 2007). This Monte
Carlo method follows the species on the surface of the grain; explicitly takes into account
processes, such as accretion, hopping, reaction, and desorption; takes into account the
layering, investigating the penetration of species into the ice; and is a stochastic approach.
Several of these mathematical methods are implemented in Chapters 2 and 3. In Chapter 2, for instance, a Monte Carlo approach is used to simulate CO hydrogenation under
laboratory conditions. Cuppen et al. (2009) extended this model using the continuous1 Rate equations are differential equations which link the reaction rate for a chemical reaction with concentrations of reactants and constant parameters (like rate coefficients and partial reaction orders).
2 The master equation is a set of first-order differential equations describing the time evolution of the probability of a system to occupy each one of a discrete set of states.
3 The Monte Carlo method is based on a class of computational algorithms that rely on repeated random
sampling to compute their results.
11
1 Introduction
time, random-walk Monte Carlo method in order to simulate microscopic grain-surface
chemistry over the long timescales in interstellar space, including the layering of ices
during CO freeze-out. In Chapter 3, reaction rates are obtained by fitting a set of differential equations to the time evolution experimental curves of newly formed H2 O (D2 O) and
H2 O2 (D2 O2 ) through O2 hydrogenation (deuteration).
Surface reactions between thermalized reactants are not the only mechanism leading
to the formation of new molecular species in the solid phase. Interstellar ices undergo
energetic processing due to cosmic ions and UV photons, which may induce non-thermal
desorption of surface molecules as well as the production of complex species. Fast ions
passing through molecular solids release their energy into the target material. As a consequence, many molecular bonds are broken along the ion-track and, in a short time (less
than one picosec), the molecular fragments recombine to produce a rearrangement of the
chemical structure that leads to the formation of new molecular species. In the case of
UV photolysis, the energy is released into the target material by a single photodissociation or photoexcitation event. In this particular case, new molecular species are also
formed. Energetic processing, thus, offers a complementary pathway to atomic addition reactions towards chemical complexity in space (e.g., Hagen et al. 1979, Gerakines
et al. 1995, Hudson & Moore 2000, Palumbo et al. 2008, Chapter 9). This thesis shows
that, for instance, the total observed abundance of molecules like CO2 can be formed in
the solid phase through surface reactions without energetic input (Chapter 7) as well as
through energetic processing of C- and O- bearing molecules that induces surface chemistry (Chapter 9). Recent models supported by laboratory studies show that UV photolysis
and ion irradiation of simple ices, such as CH3 OH, NH3 , and CH4 , trigger a complex surface chemistry. The desorption of these products in the warm inner regions of protostellar
envelopes may explain the observed gas-phase abundances of complex organic molecules
in such environments (e.g., Garrod et al. 2008, Öberg et al. 2009b, Modica & Palumbo
2010).
1.4 Laboratory ices
Laboratory setups that record vibrational spectra of ices typically consist of a radiation
source, a sample chamber, and an instrument for spectra acquisition. These elements are
also present in space: the radiation source is a field star or an embedded source, the sample chamber is the interstellar medium, with dust grains as the surface, and the recording
instrument is the spectrograph mounted onto the telescope. The interpretation of ice data
is achieved by studying interstellar ice analogues in the laboratory spectroscopically and
comparing spectra directly to astronomical observations. Spectra acquired in transmission4 provide information on the ice constituents in space, their corresponding mixing ratios, ice structure and temperature. From such a comparison it was found that dust grains
in cold (<100 K) and dense interstellar environments are covered by 50−100 monolayers
thick icy mantles, which comprise primarily H2 O, but also other species, like CO, CO2 ,
4 The IR light passes through the icy sample, which is deposited onto a cold (10−300 K) infrared transparent
window (CsI, Si or KBr) and then detected by an infrared spectrometer.
12
1.4 Laboratory ices
CH3 OH, HCOOH, NH3 , CH4 , and traces of more complex molecules. These ices are
mainly amorphous and have a layered structure consisting of H2 O- and CO-dominated
ice mixtures, as discussed in §1.3.1. Spectroscopic ice work is still ongoing, even after
more than 30 years of dedicated work. Recent examples comprise work performed in the
Sackler Laboratory for Astrophysics in Leiden on the effect of CO2 and CO on H2 O band
profiles and band strengths by Öberg et al. (2007a) and Bouwman et al. (2007), respectively. Figure 1.6 shows infrared spectra in transmittance of the four H2 O band modes,
ranging between 4000 cm−1 and 500 cm−1 , for six different ice compositions: from pure
H2 O ice (bottom panels) to a H2 O:CO = 1:4 mixture (top panels) (Bouwman et al. 2007).
For different mixture ratios the shape and peak position of the absorption bands vary. This
allows to use the laboratory data to conclude on ice parameters in the ISM.
W avelength /
2.7
0.05
2.8
2.75
3
6
6.5
12
7
x1/2
x1
H O:CO 1:4
m
3.2
14
16
x2
18
x2
2
0.00
H O:CO 1:2
2
Absorbance / arbitrary units
0.05
0.00
H O:CO 1:1
2
0.05
0.00
H O:CO 1:0.5
2
0.05
0.00
H O:CO 1:0.25
2
0.05
0.00
pure water
free OH stretch
libration
bend
bulk stretch
0.05
0.00
3700
3650
3600
3400
3200
1700
W avenumber / cm
1600
1500
900
800
700
600
-1
Figure 1.6 The 15 K infrared absorption spectra in transmittance of the four H2 O band
modes between 4000 cm−1 and 500 cm−1 for six different ice compositions, ranging from
pure water ice (bottom panels) to a H2 O:CO = 1:4 mixture (top panels). For different
mixture ratios the shape and peak position of the absorption bands vary. Note that the
wavelength ranges for separate modes are different. The small structures on the libration
mode are experimental artifacts. Spectra are taken from Bouwman et al. (2007).
The experimental setup used for these measurements consists of a High Vacuum (HV)
setup in which ices are grown onto a cold infrared-transmitting window (10−300 K). The
window is cooled down by a closed-cycle He refrigerator and the sample temperature is
controlled by resistive heating. A Fourier transform infrared (FTIR) spectrometer is used
to record the ice spectra in transmission between 4000−400 cm−1 (2.5−25 µm) with a
typical resolution of 1−2 cm−1 . This allows a straightforward detection in a single pass
13
1 Introduction
experiment. The main weakness of this method is that the base pressure in the main
chamber is 1×10−8 −1×10−7 mbar, which is orders of magnitude higher than the densest
interstellar cloud, and water pollution is not negligible (>100 ML of H2 O deposited per
hour on a 10 K substrate), since the vacuum is mainly composed by background H2 O.
In order to minimize these effects, experiments are performed in a short timescale (∼few
hours) and thick layers of ice are usually deposited on the substrate (>0.1 µm).
The formation of complex species through energetic processing (ions and UV photons) of simple ices have been experimentally investigated for decades using HV setups
(e.g., Hagen et al. 1979, Allamandola et al. 1988, Gerakines et al. 1995, Hudson & Moore
2000, Strazzulla & Palumbo 2001, Mennella et al. 2004, Mennella et al. 2006, Bennett &
Kaiser 2007, Palumbo et al. 2008, Chapter 9). With few exceptions, these experiments
are meant to investigate qualitatively and, more recently, quantitatively the effect of energetic processing on the interstellar chemistry, with the intent to mimic ice composition
in star forming regions. The work presented in Chapter 9 is performed in a HV system
and a quantitative comparison between laboratory and observational infrared spectra is
made. In this chapter the interstellar ice analogs are obtained in situ upon ion irradiation
of selected ice samples.
Another main objective of laboratory work is the characterization of astrophysically
relevant ice processes, such as surface formation, diffusion, segregation, and thermal and
non-thermal desorption of molecules. The investigation of these ice processes started
only roughly a decade ago with the introduction of a new generation of systems derived from standard surface science techniques, called Ultra High Vacuum (UHV) setups
(1×10−10 −1×10−11 mbar). The gas composition in UHV setups reproduces the interstellar
environment in the sense that it is mainly composed of H2 with densities comparable to
the disk midplane. The surface temperatures reached in these chambers are as low as <10
K. The water contamination rate on a 10 K substrate is <1 ML per 3 hours. Two UHV
surface techniques are used as analytical tools: Reflection-Absorption InfraRed Spectroscopy (RAIRS) to investigate species in the solid phase, and Temperature Programmed
Desorption (TPD) using a Quadrupole Mass Spectrometer (QMS) in order to monitor
gas-phase species desorbed from the ice.
Although the importance of surface reactions for interstellar chemistry was already realized in the 1940s, the surface formation of complex molecules in interstellar ice analogs
has only been investigated recently by atomic addition experiments (e.g., Hiraoka et al.
1998, Watanabe et al. 2006a, Matar et al. 2008, Chapters 2-8). Most of the laboratory
work so far focuses on the hydrogenation/deuteration of simple ices such as CO, O2 , and
O3 (e.g., Chapters 2-6 and references therein; see Fig. 1.7). In Chapters 7-8 the hydrogenation of CO:O2 binary mixtures is presented. Chapter 7 focuses on the formation of
solid CO2 through direct dissociation of the HO-CO complex. This chapter investigates
at the same time the competition between the different hydrogenation channels (CO +
H vs. O2 + H), while Chapter 8 focuses on the formation of HCOOH ice through the
hydrogenation of the HO-CO complex. Neither the surface formation of CO2 nor that of
HCOOH have been observed in hydrogenation experiments of pure CO and O2 ices. The
experiments presented in all these chapters are not designed to simulate a realistic interstellar ice, but to test surface reaction pathways. So far few studies have investigated the
14
1.4 Laboratory ices
hydrogenation of more complex species, such as CO2 , HCOOH, and CH3 CHO containing
ices (Bisschop et al. 2007b), or focused on the bombardment of simple ices with heavier
species than hydrogen/deuterium atoms, using a single or a double beam line (e.g., Oba
et al. 2010, Dulieu et al. 2010).
UHV setups have been recently used to investigate UV induced ice chemistry. Also
in this case experimental results are not meant to be directly compared to astronomical
observations. Rather a synergic use of RAIRS and TPD is meant to study the physical
properties of the ice. Examples are the systematic investigation of photodesorption of simple molecules, such as CO, N2 , H2 O and D2 O, as well as the study of photochemistry in
CH3 OH-rich ices and H2 O:CO2 :NH3 :CH4 mixtures (e.g., Öberg et al. 2007b, 2009d,c,b,
2010b).
1.4.1 RAIR spectroscopy
The main advantage of the RAIRS detection technique is that the reactants and products
are monitored in the solid phase at the time and temperature of interest. Figure 1.7 shows
the RAIR spectra from a CO hydrogenation experiment, as presented in Chapter 2. The
top panel of Fig. 1.7 shows the 15 K CO ice spectrum after deposition, while the bottom
panel shows the RAIR difference spectra with respect to the initial deposited ice acquired
during H-atom exposure. Quantifying the formed product with the RAIRS technique is
relatively simple, provided that the RAIRS is calibrated with an independent method, even
though, not all species can be detected in the infrared5 . The column densities of reactants
and products are usually obtained from the integrated intensity of the selected infrared
bands, using a modified Lambert-Beer equation (Bennett et al. 2004):
R
A(ν)dν
(1.1)
NX =
SX
where A(ν) is the integrated absorbance and S X is the corresponding band strength for
species X (see Chapter 2 and 4). Since literature values of transmission band strengths
cannot be used directly in reflectance measurements, an apparent absorption strength of
the various species is obtained from calibration experiments. The determination of this
apparent absorption strength is set-up specific. The calibration method (isothermal desorption of a selected ice) is described in detail in Chapters 2 and 4.
The RAIR technique has both advantages and disadvantages over transmission infrared spectroscopy. Since RAIRS is performed in a grazing incidence configuration with
respect to the substrate, which is often a copper plate covered by a film of polycrystalline
gold6 , the resulting enhancement of the p-polarized electric field at the surface leads to
a sensitivity advantage. Furthermore, in the reflection mode a double-pass geometry is
used. The incident beam must pass once through the surface layer before hitting the reflecting substrate, and a second time on its way to the detector. The adoption of a grazing
5 Diatomic homonuclear molecules like O and N are infrared in-active, except when embedded in an ice
2
2
matrix (Ehrenfreund et al. 1992).
6 The gold substrate is chemically inert, i.e., it does not have a direct effect on the behavior of the ice.
15
1 Introduction
Figure 1.7 The RAIR spectra of a CO hydrogenation experiment taken from Chapter 2.
The top-panel shows the CO ice spectrum at T = 15 K. The bottom panel shows the
RAIR difference spectra with respect to the initial deposited ice acquired during H-atom
exposure.
incidence geometry also leads to a rapid increase in path length, hence increasing the sensitivity for very thin ice layers. Thus, in a surface analysis experiment, RAIRS can probe
ice layers down to the monolayer regime and therefore it can be considered a more sensitive technique than transmission infrared spectroscopy. However, the main disadvantage
is that RAIR spectra cannot be directly compared to astronomical spectra. Therefore, as
previously mentioned, experiments using RAIRS are often not meant to reproduce interstellar ice analogs, but to investigate interstellar relevant solid state processes. Moreover,
the RAIRS detection technique is often complimented by the TPD technique for these
kind of experiments of which Chapters 2 and 8 are two examples. In the experiments
shown in these two chapters, after H-atom addition a TPD experiment is performed and
gas-phase molecules are detected by a QMS. The combination of the two techniques led
to the unambiguous identification of surface CH3 OH and HCOOH formation at low temperatures, respectively.
16
1.4 Laboratory ices
1.4.2 Mass spectrometry
As discussed in the first part of this chapter, in the earliest stages of star formation virtually all species (excluding H2 ) accrete onto grains in dense cold cores. In the later stages,
grains are warmed to temperatures where molecules desorb into the gas phase. Thus, the
less volatile species will still reside on the grain surface and participate in reactions in the
solid state, until the grain temperature will exceed 90 K, at which the entire interstellar ice
mantle is desorbed. This scenario, which takes place at astrophysical timescales of ∼107
years, can be experimentally simulated on laboratory timescales of a few hours through a
TPD experiment. An ice sample is prepared under UHV conditions with monolayer precision and heated linearly with a selected rate (e.g., Fraser et al. 2001, Collings et al. 2003,
2004, Bisschop et al. 2006, Acharyya et al. 2007). The desorbed species are subsequently
recorded as a function of temperature using a QMS, which produces a signal proportional
to the number of incoming molecules as a function of their mass to charge ratio (m/z).
The incoming molecules first enter the ion source of the QMS, where they are ionized
through electron bombardment by electrons released from a hot filament. The resulting
ions are then focussed, selected and directed to the detector. Ions are detected by a Faraday detector, which collects the ions directly, allowing the ion current to be monitored.
Alternatively, for higher sensitivity, a Channel Electron Multiplier (CEM) can be used.
This type of detector is a Secondary Electron Multiplier (SEM) in which a large negative
potential (∼ - 2000 V) is used to attract the ions into the channel entrance. The channel is
coated with a material that readily releases secondary electrons upon ion/electron impact.
This produces a cascade of electrons down to the channel which can be detected, either
as the electron current, or as a series of pulses.
Whilst the TPD experiment itself is relatively straightforward, the interpretation of
the data is often much more challenging. Under conditions where the pumping speed
is sufficiently high, the QMS signal for the selected mass is proportional to the rate of
desorption of that species, rd (molecules cm−2 s−1 ). The rate of desorption is given by the
Polanyi-Wigner equation:
rd = νi NX exp(−Ed /RT)
(1.2)
where νi is the pre-exponential factor for the process leading to the desorption, NX is the
surface concentration of adsorbate X, i is the desorption order, Ed is the activation energy
for desorption per mole, R is the gas constant and T the temperature. Desorption from
multilayers of bulk ice is typically close to zero order (i = 0). For perfect zero order
desorption, the desorption rate does not depend on the surface concentration. In many
cases, desorption of submonolayer coverage results in near first order kinetics (i = 1), i.e.,
indicating that the desorption rate depends linearly on the surface concentration. This two
desorption regimes can be experimentally investigated performing an isothermal desorption experiment, as shown in several chapters of this thesis.
The simplest thermal desorption process is the desorption of a pure ice. Desorption
of binary or more complex mixtures is less understood. Figure 1.8 shows the TPD curves
for CO and CO2 desorption from pure ice (solid lines) and mixed with water (dotted
17
1 Introduction
lines) (Öberg 2009). The desorption energy of the single species depends on the exact ice
composition. In addition, the dominant mantle species may prevent other species from
desorbing. Molecules can, indeed, get trapped in the matrix and then be released at different temperatures than their thermal desorption temperatures. Laboratory experiments
show that ice heating also results in ice segregation of previously mixed ices (e.g., Öberg
et al. 2009a, Chapter 9).
Although, TPD is a more sensitive technique than RAIRS, it has however several disadvantages: the surface reaction products, which remain in the solid phase, cannot be
probed in situ; additional surface reactions during the TPD (during the linear heating of
the ice and before complete desorption of the species) cannot be excluded; quantifying the
desorbing species is not straightforward; and some of the interesting species have equal
(i.e., undistinguishable) masses or are fractionated in the QMS upon electronic bombardment. A TPD experiment finally destroys the ice.
Figure 1.8 TPD curves for CO and CO2 desorption from pure ice (solid lines) and
H2 O:CO = 5:1 and H2 O:CO2 = 4:1 (dotted lines) binary mixtures. The TPD spectra
are normalized with arbitrary factors for visibility and are taken from Öberg (2009).
1.4.3 Experimental setups used in this thesis
SURFRESIDE
Chapters 2-8 are dedicated to the investigation of surface reactions that can lead to the
formation of interstellar relevant molecules. All experiments are performed in the Sackler
laboratory for Astrophysics using an UHV setup SURFace REaction SImulation DEvice
18
1.4 Laboratory ices
(SURFRESIDE), which consists of a stainless steel vacuum main chamber and an atomic
line. A schematic view of the experimental apparatus is shown in Fig. 1.9. The gold
coated copper substrate temperature is controlled between 12 and 300 K. Deposition of
selected gasses proceeds under an angle of 45◦ , with a controllable flow. Gas phase
molecules are monitored during the deposition mass spectrometrically by means of a
QMS, which is placed behind the substrate and opposite to the atomic source. A thermal
cracking source (Tschersich & von Bonin 1998, Tschersich 2000, Tschersich et al. 2008)
is used to hydrogenate the ice sample through heating a capillary pipe, in which H2 flows,
from 300 to 2250 K by a surrounding tungsten filament. A quartz pipe is placed along
the path of the dissociated beam to efficiently thermalize all H atoms to room temperature
through surface collisions before they reach the ice sample. The Hydrogen Atom Beam
Source (HABS) atom fluxes are measured at the substrate position mass spectrometrically,
as described in the appendix of Chapter 4.
Figure 1.9 Schematic top-view of the solid-state experimental UHV set-up (SURFRESIDE): (a) H-atom source; (b) cold finger; (c) IR beam; (d) mass spectrometer; (e) main chamber; ( f ) IR detector; (g) deposition line.
Ices are monitored by means of RAIRS using a FTIR spectrometer, which covers the
range between 4000 and 700 cm−1 (2.5−14 µm). A spectral resolution between 1 and
4 cm−1 is used and several scans are co-added. In Chapters 2-4 and 6-7, the ice is first
deposited and then hydrogenated/deuterated. In this case, RAIR difference spectra with
respect to the initial deposited ice are acquired during H/D exposure. In Chapters 5 and 8,
molecules are co-deposited with H atoms and RAIR difference spectra with respect to the
bare substrate are acquired during co-deposition. In all cases, newly formed solid species
are monitored by RAIRS. Spectra are recorded at different stages during an experiment,
providing time resolved information about the destruction (i.e., use-up) of the precursor
ice (the deposited ice layer) and the formation of new molecules that are identified through
their spectral fingerprints. At the end of the atomic addition a TPD can be performed to
constrain the experimental results. Surface reactions of simple ices, like CO, O2 , O3 , and
19
1 Introduction
CO:O2 mixtures are investigated during a full range of laboratory conditions including
different atomic fluxes, ice temperatures, ice thicknesses, ice structures, and mixture ratios with the intent to unreveal the physics and chemistry of molecule formation more
than simulating chemistry in a realistic interstellar ice. However, in Chapter 3 also the
astronomical implications of H2 O formation are examined in more detail.
HV setup in LASp
The experiments discussed in Chapter 9 are performed in the Laboratory for Experimental Astrophysics (Laboratorio di Astrofisica Sperimentale - LASp) in Catania (Italy). The
experimental setup used to obtain infrared transmission spectra is composed of a stainless
steel HV chamber interfaced with a FTIR spectrometer (4400−400 cm−1 ) comparable to
the setup described in § 1.4. Molecules are injected into the chamber through a needle
valve and subsequently deposited onto a chosen substrate (Si or KBr) placed in thermal
contact with a cold finger (10−300 K). After deposition the samples are bombarded by
200 keV H+ ions. The ions are obtained from an ion implanter interfaced with the vacuum chamber. The used beam produces current densities in the range from 100 nA cm−2
up to a few µA cm−2 in order to avoid macroscopic heating of the target. The ion beam
and the infrared beam are mutually perpendicular, forming an angle of 45◦ with the substrate plane. In this way, spectra can be taken at any time of the experiment without
tilting the sample. The energy released to the sample by impinging ions (dose) is given in
eV/16 u, where u is the unified atomic mass unit defined as 1/12 of the mass of an isolated
atom of carbon-12. The polarization of the infrared radiation is selected by rotating a
polarizer placed in the infrared beam path before the detector. Therefore, for each single
irradiation-step two spectra are acquired in different polarizations: one with the electric
vector parallel (P polarized) and one perpendicular (S polarized) to the plane of incidence.
1.5 This thesis
Chapter 2
Chapter 2 focuses on the formation of formaldehyde (H2 CO) and methanol (CH3 OH) by
hydrogenation of pure CO ice. Reaction rates are determined from RAIR data for different ice temperatures and ice thicknesses, as well as H-atom fluxes (1 × 1012 −2.5×1013 H
atoms cm−2 s−1 ). The formation of new molecules in the ice is confirmed by TPD: H2 CO
and CH3 OH are also found mass spectrometrically. On the basis of these experiments
energy barriers for the H + CO and H + H2 CO reactions are obtained by fitting Monte
Carlo simulation results to the experimental data. Using these barriers, the CH3 OH production can be simulated for interstellar conditions (Cuppen et al. 2009) and from this
work, surface hydrogenation of CO ice can now be safely used to explain the observed
abundance of CH3 OH in the interstellar medium.
20
1.5 This thesis
Chapters 3-6
Water ice formation through surface reactions is extensively discussed here. In 1982 Tielens & Hagen (1982) proposed that interstellar water forms on grain surfaces through
three reaction channels: hydrogenation of atomic oxygen, molecular oxygen and ozone.
In Chapter 3, the molecular oxygen channel is investigated for a large range of temperatures. The main and surprising finding is that the initial formation rate of H2 O2 and H2 O
is much less temperature dependent than the analogous reactions for CO hydrogenation.
Furthermore, O2 hydrogenation results in a much larger yield than the few monolayers
found for CO hydrogenation. This yield is strongly temperature dependent. In Chapter 4, both effects are shown to be a direct consequence of the competition between the
reaction of H atoms with O2 molecules, which is barrierless and therefore temperature
independent, and the H-atom diffusion into the O2 ice, which is temperature dependent.
In Chapter 4, O2 hydrogenation is investigated extensively from a physical approach, i.e.,
studying different ice thicknesses, ice temperatures, ice structures and H2 concentrations
in the atomic beam, whereas Chapter 5 focuses more on the reaction scheme with the
intent to assess reaction routes and branching ratios. The latter Chapter shows that the
initially proposed reaction network of only three channels is too simple and that several
of the channels are actually linked through additional reactions.
The hydrogenation of solid O3 is discussed in Chapter 6. Since this channel is connected to the O2 channel after the first reaction step, special care is taken in Chapter 6 to
deposit a pure O3 ice (free from O2 contamination7 ) by keeping the substrate temperature
between the O2 and O3 desorption temperature during deposition. If such a temperature
is also kept during H-atom addition, the O2 molecules formed upon O3 hydrogenation
will desorb from the surface of the ice. Thus, in this way the reaction of OH to form
water by reaction with H or H2 can be investigated. The hydrogenation of O3 is found
to be more similar to CO hydrogenation in the sense that only the top few monolayers of
O3 are hydrogenated. Moreover, the reaction OH + H2 could be more efficient than the
reaction OH + H: reaction OH + H2 could proceed through tunneling, while reaction OH
+ H needs to dissipate 5.3 eV of excess energy with just one final product, which could
be difficult. The conclusion that the three channels (O/O2 /O3 + H) are strongly linked,
is of importance for astrochemical models focusing on water formation under interstellar
conditions.
Chapter 7
Chapter 7 discusses the hydrogenation of CO:O2 binary mixtures, which results in the
formation of CO2 through the reaction of OH and CO. Surface CO2 formation without
energetic input is found to be an important formation mechanism, which may explain the
formation of CO2 together with H2 O ice during the dense cold core phase prior to star
formation. The competition between CO hydrogenation and O2 hydrogenation reveals
that the penetration depth of H atoms into the ice depends strongly on the ice composition,
7 Some
gas phase O2 can be formed in the deposition line and in the main chamber through O3 dissociation.
21
1 Introduction
and that the CO and O2 channels influence each others final product yields. However,
the formation rate for all the final products is found to be less sensitive on the mixture
composition than the final yield. Therefore, the formation rates found for H2 CO, CH3 OH,
H2 O2 and H2 O in the isolated studies of the CO + H (Chapter 2) and O2 + H (Chapter 3)
channels are valid for use in astrochemical models.
Chapter 8
Chapter 8 deals with the formation of solid formic acid (HCOOH). The aim of Chapter 8
is to give the first experimental evidence for solid HCOOH formation at low temperature
through the hydrogenation of the HO-CO complex, stabilized in the ice by intramolecular energy transfer to the surface, as proposed by Goumans et al. (2008). Formation of
HCOOH is observed in the infrared after co-deposition of CO:O2 mixtures and H atoms,
increasing the ice temperature below the CO and O2 desorption temperature (<30 K) and
therefore increasing the mobility of the ice components only. At these temperatures H
atoms, trapped in the ice matrix or formed through surface reactions, can find and react
with the stabilized HO-CO complex. These experiments demonstrate that the reaction
HCO + OH is inefficient at low temperatures. In the last part of the chapter, experimental results are placed in an astrophysical context. It is shown that the HO-CO complex
channel, which was previously not considered as an important HCOOH formation route,
explains the presence of HCOOH in dense cold clouds, at the beginning of the warm-up
phase of a protostar.
Chapter 9
Chapter 9 discusses the formation of solid CO2 through reactions induced by energetic
processing of C- and O- bearing molecules. Chemical and structural modifications of
the interstellar ice analog samples induced by energetic processing are analyzed using
infrared spectroscopy in transmittance. Therefore, a quantitative study is obtained with
the intent to compare laboratory results to observations. Experiments show that laboratory spectra are a good spectroscopic analogue of the interstellar features. Even if the
comparison between laboratory and observed spectra presented here cannot be considered unique and complete, our results quantitatively show that interstellar solid CO2 can
form after ion irradiation and UV photolysis of icy mantles. The results presented here
complement those shown in Chapter 7, where the CO2 formation without energetic input
is investigated. Both mechanisms can, indeed, contribute to the total CO2 column density
observed in quiescent clouds and star forming regions.
22
CHAPTER 2
Hydrogenation reactions in interstellar CO ice
analogues: a combined experimental/theoretical
approach1
Hydrogenation reactions of CO in inter- and circumstellar ices are regarded as an important starting point in the formation of more complex species. Previous laboratory measurements by two groups of the hydrogenation of CO ices provided controversial results
about the formation rate of methanol. Our aim is to resolve this controversy by an independent investigation of the reaction scheme for a range of H-atom fluxes and different
ice temperatures and thicknesses. To fully understand the laboratory data, the results are
interpreted theoretically by means of continuous-time, random-walk Monte Carlo simulations. Reaction rates are determined by using a state-of-the-art ultra high vacuum
experimental setup to bombard an interstellar CO ice analog with H atoms at room temperature. The reaction of CO + H into H2 CO and subsequently CH3 OH is monitored by a
Fourier transform infrared spectrometer in a reflection absorption mode. In addition, after
each completed measurement, a temperature programmed desorption experiment is performed to identify the produced species according to their mass spectra and to determine
their abundance. Different H-atom fluxes, morphologies, and ice thicknesses are tested.
The experimental results are interpreted using Monte Carlo simulations. This technique
takes into account the layered structure of CO ice. The formation of both formaldehyde
and methanol via CO hydrogenation is confirmed at low temperature (T = 12−20 K).
We confirm that the discrepancy between the two Japanese studies is caused mainly by
a difference in the applied hydrogen atom flux, as proposed by Hidaka and coworkers.
The production rate of formaldehyde is found to decrease and the penetration column to
increase with temperature. Temperature-dependent reaction barriers and diffusion rates
are inferred using a Monte Carlo physical chemical model. The model is extended to
interstellar conditions to compare with observational H2 CO/CH3 OH data.
1 Based on: G. W. Fuchs, H. M. Cuppen, S. Ioppolo, C. Romanzin, S. E. Bisschop, S. Andersson, E. F. van
Dishoeck, and H. Linnartz, 2009, Astronomy & Astrophysics, volume 505, pages 629-639
23
2 Hydrogenation reactions in interstellar CO ice analogues
2.1 Introduction
An increasing number of experimental and theoretical studies have focused on the characterization of solid state astrochemical processes. These studies were triggered by the
recognition that many simple and more complex molecules in the interstellar medium are
most likely to have formed on the surfaces of dust grains. Astronomical observations,
detailed laboratory studies, and progress in UHV surface techniques allow experimental verification of the initial surface reaction schemes, as introduced by Tielens, Hagen,
and Charnley (Tielens & Hagen 1982, Tielens & Charnley 1997). Recently the formation of water was demonstrated in hydrogenation schemes starting from solid molecular
oxygen (Miyauchi et al. 2008, Ioppolo et al. 2008) and the formation of ethanol from acetaldehyde (Bisschop et al. 2007b). The first solid-state astrochemical laboratory studies
focused on the formation of formaldehyde and methanol by H-atom bombardment of CO
ice. Methanol is observed abundantly in interstellar ices and is considered to be a resource
for the formation of more complex molecules through surface reactions and after evaporation in the gas phase (Charnley et al. 1992). The hydrogenation scheme for the solid
state formation of methanol was proposed to be
H
H
H
CO + H → HCO −
→ H2 CO −→ H3 CO −
→ CH3 OH.
(2.1)
Laboratory studies of H-atom bombardment of CO ice were performed independently
by two groups (Hiraoka et al. 2002, Watanabe & Kouchi 2002). Hiraoka et al. (2002)
observed only formaldehyde formation, whereas Watanabe & Kouchi (2002) also found
efficient methanol production. In a series of papers, these conflicting results were discussed (Hiraoka et al. 2002, Watanabe et al. 2003, 2004) and the prevailing discrepancy
between results was proposed to be a consequence of different experimental conditions,
most noticeable the adopted H-atom flux (Hidaka et al. 2004). Understanding the solidstate formation route to methanol became even more important following an experimental
finding that the gas-phase formation route via ion-neutral reactions is less efficient than
previously estimated and cannot explain the observed interstellar abundances (Geppert
et al. 2005, Garrod et al. 2006).
Deuteration experiments were also performed on CO ice, which confirmed the formation of both fully deuterated formaldehyde and methanol, but with substantially lower
reaction rates (Nagaoka et al. 2005, Watanabe et al. 2006a). In the presence of both hydrogen and deuterium it was suggested that first normal methanol forms and is then gradually
converted in the deuterated species by exchange reactions.
The present chapter strongly supports the flux argument given by Hidaka et al. (2004).
Furthermore, in § 2.3 a systematic study of the physical dependencies involved in the
CO-ice hydrogenation is presented to place previous work in a context that allows an extension of solid state astrochemical processes to more complex species. Special emphasis
is placed on the flux and temperature dependence of the formation rate. An analysis of
the spectral changes of CO ice during hydrogenation is included to provide insight into
the structure of the reactive layer. Furthermore, in § 2.4 Monte Carlo simulations are
presented that allow us to interpret the experimental results in greater detail and to vary
24
2.2 Experimental procedure
parameters that are difficult to study independently by experiment. We conclude with a
simulation of H2 CO/CH3 OH formation under interstellar conditions, in particular for low
H-atom fluxes (§ 2.5). The outcome is compared with astronomical observations (§ 2.6).
Nose shaped pipe for
collisional cooling
H−beam shutter
IR Beam
Turnable
cryostat
QMS
Thermal cracking
H−source
Dosing Line
Gold substrate
Main Chamber
Figure 2.1 Schematic representation of the experimental setup. CO ice is deposited
through the dosing line and the products are monitored by means of infrared spectroscopy
and quadrupole mass spectroscopy (QMS).
2.2 Experimental procedure
The experiments are performed under UHV conditions. The room temperature base pressure of the vacuum system is better than 3 × 10−10 mbar. Figure 2.1 shows a schematic
representation of the setup (see Chapter 4 for additional information). Amorphous CO
ices ranging from a few to several monolayers (ML) are grown on a gold coated copper
substrate that is located at the center of the main vacuum chamber and mounted on the
tip of a cold finger of a 10 K He-cryostat. The temperature of the ice can be controlled
between 12 and 300 K with 0.5 K relative precision between experiments. The absolute
accuracy is better than 2 K. During deposition, the layer thickness is monitored by simultaneous recording of reflection absorption infrared (RAIR) spectra. To exclude the effects
of potential pollution, ices are grown using CO, 13 CO, or C18 O isotopologues.
During the experiment the ice layers are exposed to a hydrogen atom beam. The atoms
are produced by a well-characterized commercial thermal-cracking source (Tschersich &
von Bonin 1998, Tschersich 2000) that provides H-atom fluxes on the sample surface of
between 1012 and 1014 atoms cm−2 s−1 . By comparison, the Hiraoka group used fluxes below 1013 atoms cm−2 s−1 and the Watanabe group worked in the 1014 −1015 atoms cm−2 s−1
regime. The hot (∼2000 K) hydrogen atoms are cooled to room temperature by surface
collisions in a nose-like shaped quartz pipe between the atomic source and the ice sample.
In this way, hot hydrogen atoms cannot affect the ice directly. H-atom recombination in
this connecting pipe results in a lower final flux. Details about the flux determination are
given in Appendix A. The absolute fluxes are estimated to be within a factor of two and
the relative fluxes to within 50%.
25
2 Hydrogenation reactions in interstellar CO ice analogues
The relatively high temperature of the incident atoms of 300 K does not affect the process; previous experiments with colder H atoms did not show any substantial temperature
dependence because the atoms are immediately thermalized on the surface (Watanabe
& Kouchi 2002). It is argued that the surface is covered with a thin layer of hydrogen
molecules under these conditions. These molecules are either formed on the surface or
originate from the partially dissociated beam. Since the incoming atoms have to penetrate
this cold H2 layer, they are thermally adjusted to the surface temperature once they come
in contact with the CO molecules.
Information about the reaction products is obtained using two complementary techniques. During the H-atom bombardment, reactants and products are monitored by recording RAIR spectra. The RAIR spectra are recorded using a Fourier transform infrared
spectrometer with 1 and 4 cm−1 resolution and covering the spectral region in which CO
(2143 (s) cm−1 ), formaldehyde (1732 (s), 1479, and 2812 (m), and 1246, 1175, 2991,
2880, and 2812 (mw) cm−1 ) and methanol (1035 (s) and 1125 (w) cm−1 ) exhibit strong
(s), medium (m), or weak (w) absorption bands. The intensity of spectral features is
directly related to the density in the ice. The products are monitored by mass spectrometry using temperature programmed desorption (TPD) once a hydrogenation experiment is
completed.
0.002
∆Abs
0.001
0
-0.001
H2CO
CO
CH3OH
-0.006
2000
1750
1500
1250
1000
-1
wavenumber [cm ]
Figure 2.2 RAIR difference spectrum of a CO ice at 12.0 K exposed to 5.4 × 1017 cm−2
H atoms at a flux of 5 × 1013 cm−2 s−1 . The spectrum after CO deposition is used as the
reference spectrum. Note that the CO peak reaches an absorbance difference of -0.006.
26
2.3 Experimental results
0.2
Abs
submonolayer
regime
CH3OH
0.1
0
CO
10
20
30
40
Time [minutes]
Figure 2.3 The decrease in integrated absorbance of CO and CH3 OH (1035 cm−1 ) following desorption at a constant temperature of 29 and 135 K, respectively. The arrows
indicate the transition points from the multi- to sub-monolayer regime.
2.3 Experimental results
2.3.1 A sample experiment
To illustrate the experimental method, we start by discussing a sample experiment in
which a CO ice of 8 × 1015 molecules cm−2 is bombarded with H atoms with a flux of 5 ×
1013 atoms cm−2 s−1 for three hours at a surface temperature of 12.0 K. This corresponds
to a fluence of 5.4 × 1017 atoms cm−2 . Figure 2.2 shows the RAIR difference spectrum
(∆Abs) after these three hours of exposure (after - before). The CO, the H2 CO and the
CH3 OH spectral signatures are indicated with respect to the spectrum recorded before the
H-atom bombardment started. The CO appears as a negative band, indicating its use-up,
and the other bands are positive, indicating the formation of H2 CO and CH3 OH. Neither
the intermediate species, HCO and H3 CO, nor more complex species are observed.
The column density NX (molecules cm−2 ) of species X in the ice is calculated using
R
A(ν)dν
,
(2.2)
NX =
SX
where A(ν) is the wavelength dependent absorbance. Since literature values of transmission band strengths cannot be used in reflection measurements, an apparent absorption
band strength, S X of species X is calculated from a calibration experiment in which an
27
2 Hydrogenation reactions in interstellar CO ice analogues
ice layer of species X desorbs at constant temperature until the sub-monolayer regime.
This is illustrated in Fig. 2.3, which shows the decrease in integrated absorbance of CO
and CH3 OH during such an experiment. The arrows in the graph indicate the deviation onset from constant desorption, which marks the transition point from the multito sub-monolayer regime. The apparent absorption band strengths of CO and CH3 OH
(1035 cm−1 ) thus obtained are setup specific. The corresponding uncertainty in the band
strengths remains within 50%. The ratio of S CO to S CH3 OH in our reflection experiment
is similar to the transmittance ratio, 0.85. The value for S H2 CO is obtained by assuming
mass balance
R
A(ν)dν
NCO (t) + NCH3 OH (t) = −
(2.3)
S H2 CO
for a set of different experiments. In addition, the results discussed in the present chapter are all in a regime where the proportionality relation (Teolis et al. 2007) still holds
(<3 × 1016 molecules cm−2 ).
0.04
before
0
after--before
CH3OH:CO
Abs
-0.01
after
2150
0.02
0
2145
2140
2140
2135
2130
2130
-1
wavenumber [cm ]
Figure 2.4 Spectral change of the CO 2143 cm−1 RAIR band before and after H-atom
bombardment. The inset shows the corresponding difference spectrum.
The CO band shape can change when molecules other than CO are formed. Figure 2.4
shows the 2143 cm−1 IR peak before and after the H-atom exposure. A clear decrease
in the peak height can be observed caused by the use-up of CO during the experiment,
as expected. However, an additional peak appears at 2135 cm−1 (see inset Fig. 2.5),
which is due to a CH3 OH−CO ice interaction. Transmission IR spectra of a CH3 OH:CO
mixture show a band at 2136 cm−1 (Bisschop 2007, Palumbo & Strazzulla 1993). When
the methanol bands grow, the band at 2135 cm−1 also increases. Figure 2.5 shows how
28
2.3 Experimental results
the peak position of CO shifts with the methanol content in the reflection spectra. The
RAIR spectra on which this graph is based, are taken for ice layers that are formed by codeposition of CO and CH3 OH of known ratio. The CO stretching mode in H2 O:CO and
NH3 :CO mixtures shows a similar behavior (Sandford et al. 1988, Bouwman et al. 2007).
As for both H2 O and NH3 , CH3 OH is able to form hydrogen bonds and these hydrogen
bonds most likely cause the redshift of the CO band. By comparing the position of the
peak in Fig. 2.4 at 2135 cm−1 to the data in Fig. 2.5, we can constrain the methanol fraction
in the top layers, assuming that the formed CH3 OH:CO mixture has the same spectral
behavior as the deposited mixtures. The observed data after three hours correspond to a
CH3 OH:CO mixture of at least 90%. This means that the top layer of the ice is completely
converted into H2 CO and CH3 OH and that no or very little additional mixing with CO
occurs. For the H2 CO and CH3 OH band, no spectral changes are observed during the
experiments.
-1
wavenumber [cm ]
2144
2142
2140
2138
2136
0
0.2
0.4
0.6
0.8
1
CH3OH fraction
Figure 2.5 CO RAIR band position as a function of CH3 OH content in a CO:CH3 OH
mixed ice obtained by co-deposition experiments.
To quantify the use-up of CO and the formation of new products, we have to assume
that the apparent absorption band strength is constant during an experiment, i.e., independent of the ice composition. Bouwman et al. (2007) found that the band strength of the
2143 cm−1 CO feature is indeed unaffected within the experimental error by the water
content in H2 O:CO-ice mixtures up to 4:1. The band strength is expected to behave similarly for a CO:CH3 OH-mixture. Furthermore, if the band strength is strongly affected
by the ice composition, the total ice thickness determined using a constant band strength
should vary in time, whereas the real thickness is constant. Since this does not occur,
we estimate that the change in band strength caused by the change in ice composition is
29
2 Hydrogenation reactions in interstellar CO ice analogues
negligible and well within our margins of error.
Figure 2.6 (a) shows the time evolution of the integrated CO, H2 CO, and CH3 OH
signals in symbols. It shows how the amount of CO decreases as the abundance of H2 CO
increases for four different temperatures. After bombardment with 1×1017 H atoms cm−2 ,
the formation of methanol kicks off at the expense of the growth of the H2 CO abundance.
Similar trends of abundance evolution as a function of fluence are reported by Watanabe
et al. (2006a). This indicates that the fluence is determined with relatively high accuracy
since in both experiments different atomic sources (Tschersich vs. microwave induced
plasma) and different calibration methods are used.
Time [minutes]
0
50
2 Fig. 7
100
150
0
50
100
150
Fig. 2
H2CO
1
15
-2
N(t) [10 molecules cm ]
0
CH3OH
-1
-2
CO
(a)
(b)
2
1
0
-1
-2
(c)
0
(d)
1
2
3
4
5
0
17
1
2
3
4
5
-2
Fluence [10 atoms cm ]
Figure 2.6 Time evolution of the surface abundance (molecules cm−2 ) of CO, H2 CO and
CH3 OH during H-atom bombardment of CO ice with a H-atom flux of 5 × 1013 cm−2 s−1
at surface temperatures of 12.0 K (a), 13.5 K (b), 15.0 K (c), and 16.5 K (d). Experimental
data (symbols) and Monte Carlo simulation results (solid lines) are shown as well.
30
2.3 Experimental results
2.3.2 Flux dependence
As mentioned in § 2.1, the apparent discrepancy between the results by Hiraoka et al.
(2002) and Watanabe & Kouchi (2002) has been attributed to a difference in the H-atom
flux used in the respective experiments. The setup in our laboratory is able to cover
the entire flux range from 1012 to 1014 cm−2 s−1 . For high flux, both formaldehyde and
methanol are formed as can be seen in Figs. 2.2 and 2.6 and in the corresponding work of
Watanabe & Kouchi (2002).
0.002
∆Abs
0.001
0
-0.001
H2CO
CO
CH3OH
-0.002
2000
1750
1500
1250
1000
-1
wavenumber [cm ]
Figure 2.7 RAIR spectrum of a solid CO ice at 12.0 K exposed to a H-atom fluence of
1 × 1016 H atoms cm−2 .
A difference spectrum of a similar experiment but with a lower flux of 1012 cm−2 s−1
is plotted in Fig. 2.7. The exposure time here is four hours to reach a higher total fluence of 1 × 1016 cm−2 , which is still significantly less than the sample experiment shown
in Fig. 2.2. Note that the vertical scales in Figs. 2.2 and 2.7 cover the same range. For
longer exposures, surface contamination would become a problem, but methanol features
would eventually become detectable. As Fig. 2.7 clearly shows, a smaller fraction of CO
is transformed into H2 CO and the sensitivity of the RAIR spectrometer is not sufficiently
high to confirm the formation of CH3 OH in these circumstances. TPD, however, is more
sensitive as a diagnostic tool, although harder to use for a quantitative or time resolved
analysis. Figure 2.8 plots several TPD spectra. It shows a small methanol desorption peak
around 150 K. We checked experimentally that the carrier of this peak is indeed formed
in the ice during the hydrogen exposure and that the observed CH3 OH is not a contaminant in the UHV chamber. This is a strong indication that the formation mechanism of
31
2 Hydrogenation reactions in interstellar CO ice analogues
Partial pressure [mbar]
formaldehyde and methanol does not fundamentally change with varying flux. The H2 O
desorption at 20−30 K originates in frozen background water on the surrounding parts of
the cryohead.
The arrows in Fig. 2.6 (a) indicate the corresponding fluences for the low and high flux
experiments shown in Figs. 2.7 and 2.2, respectively. From this, it is immediately apparent
that only a limited amount of methanol can be formed under low flux circumstances. Note
that Hiraoka et al. (2002) probably used an even lower fluence since their exposure time
was four times shorter than in our experiment. In addition, they used a slightly lower
temperature of 10 K.
H2
-9
1×10
-10
1×10
m/e = 2
m/e = 18
m/e = 28
m/e = 30
m/e = 32
CO
-11
1×10
H 2O
-12
1×10
H2CO
-13
CH3OH
1×10
0
50
100
150
200
Temperature [K]
Figure 2.8 The TPD spectra corresponding to Fig. 2.7.
2.3.3 Thickness dependence
The effect of the initial layer thickness on the formation yield of H2 CO and CH3 OH is
investigated by repeating the sample experiment for different CO layer thicknesses. Figure 2.9 shows the absolute reaction yield after a fluence of 5.4 × 1017 H atoms cm−2
as a function of the layer thickness. In all cases, a steady state value for H2 CO is
reached for this fluence. The figure clearly shows that for CO ice layers thicker than
4 × 1015 molecules cm−2 , the absolute yield is layer thickness independent and the results
are reproducible within the measurement error. The combined H2 CO and CH3 OH yield
of 2 × 1015 molecules cm−2 is lower than the 4 × 1015 molecules cm−2 penetration column.
From these experiments, we conclude that the penetration column of the H atoms into the
CO ice is at most 4 × 1015 molecules cm−2 at 12.0 K. This corresponds to 4 monolayers
of solid (bulk) CO molecules. At least half of the CO molecules in the active layer is
32
2.3 Experimental results
15
-2
Yield [10 molecules cm ]
converted into H2 CO and CH3 OH. The determination of the penetration column by this
experiment is only an upper limit because of the low thickness resolution in Fig. 2.9. It
however agrees with the previous estimate of nearly 100% conversion.
1.4
H2CO
1.2
1
0.8
0.6
CH3OH
0.4
0.2
00
2
4
6
15
8
10
-2
Initial thickness [10 molecules cm ]
Figure 2.9 The absolute reaction yield of H2 CO and CH3 OH after a fluence of 5.4 × 1017
H atoms cm−2 as a function of the layer thickness for experiments at 12.0 K.
2.3.4 Temperature dependence
Several experiments for different surface temperatures have been performed. The initial
layer thickness and flux values are comparable to the values used in the sample experiment. Figures 2.6 (b)-(d) show the results for hydrogenation experiments at 13.5, 15.0,
and 16.5 K, respectively. These clearly indicate the very different evolution of CO, H2 CO,
and CH3 OH abundance with temperature. Table 2.1 gives the initial formation rate of
formaldehyde (slope at t = 0) and the final H2 CO and CH3 OH yields, and also indicates
whether or not a steady state is reached. The table shows that at early times, the formation
rate of H2 CO is much lower for higher temperatures than for 12.0 K. We will return to this
point later. The final yield of CH3 OH is however larger at 13.5 and 15.0 K. For T > 15 K,
the production rate of H2 CO is simply so low that a steady state is not reached. Minimal
amounts of formed methanol were also detected in experiments at 18.0 and 20.0 K, but
since some CO desorption and redeposition occurs at these temperatures, they are not
presented here for a quantitative discussion.
The appearance of the extra CO band at 2135 cm−1 indicates that for temperatures
between 12.0 and 15.0 K a nearly pure methanol layer is formed. We expect a similar
33
2 Hydrogenation reactions in interstellar CO ice analogues
behavior for formaldehyde. This means that the active CO layer involved in the reactions
can be determined directly from the steady state yield of H2 CO and CH3 OH. This active
layer increases with temperature indicating that the penetration column of H atoms into
CO ice increases with temperature as one would expect. The CO molecules in the ice
are more mobile at higher temperatures making it easier for H atoms to penetrate the CO
ice, since the ice becomes less rigid. Note that the absolute temperature calibration in the
set-up of Watanabe and ours appears to differ by 1−2 K (comparing Fig. 3 in Watanabe
et al. (2006a) and Fig. 2.6 here), but the observed trends are identical.
Table 2.1 The production rate and yield of H2 CO and the yield of CH3 OH.
T
(K)
12.0
13.5
15.0
16.5
18.0
20.0
Rate(H2 CO)t=0 a
(10−3 molec./H atom)
9.0
7.3
3.2
1.1
1.0
0.9
Yield (H2 CO)b
(1015 molec. cm−2 )
1.2
1.0
0.9
0.8
0.5
0.4
Yield (CH3 OH)b
(1015 molec. cm−2 )
0.8
1.4
1.6
0.6
0.2
0.1
Steady state
yes
yes
yes
no
no
no
Calc. pen. columnc
(1015 molec. cm−2 )
2.0
2.4
2.5
a Rate
at t = 0 determined from slope.
after three hours of H-atom exposure which corresponds to a fluence of 5.4 × 1017 H atoms cm−2 . The
steady state yield is not reached for all temperatures (fifth column).
c Penetration column obtained from columns 3, 4, and 6.
b Yield
2.4 Monte Carlo simulations
2.4.1 The method
To infer the underlying mechanisms leading to the formation of methanol, a detailed
physical-chemical model is required. The present section discusses an approach based
on the continuous time, random-walk Monte Carlo simulation technique. This method
differs from previous studies based on rate equations and enables the study of surface
processes to be performed in more detail. In addition, it provides a clearer understanding
about what occurs physically on the surface. In contrast to an analysis using rate laws, the
Monte Carlo method determines the H surface abundance by taking into account the layered structure of the ice, the H-atom flux, diffusion, reaction and desorption. This allows
an extension of the results to conditions with much lower fluxes such as in the interstellar
medium (ISM). For a detailed description of the method and program, we refer to Cuppen
& Herbst (2007).
During a simulation, a sequence of processes - hopping, desorption, deposition, and
reaction - is performed, where this sequence is chosen by means of a random number
generator in combination with the rates for the different processes. First, an initial ice
34
2.4 Monte Carlo simulations
layer is created by deposition of CO on a surface. The resulting surface roughness of
this layer depends on temperature and flux. For the experimental conditions simulated
here, the CO ice is compact with a maximum height difference across the surface of only
2−3 monolayers. Hydrogen atoms and hydrogen molecules are subsequently deposited,
following their relative abundance in the H-atom flux, at an angle perpendicular to the
surface to mimic the experimental conditions. They move, react, and desorb according to
rates of a form similar to that used in gas-grain models
E x
,
(2.4)
R x = A exp −
T
where E x is the activation energy for process X, and A is the pre-exponential factor for
which a constant number of ν ∼ kT/h = 2 × 1011 s−1 is used. The activation energies are
not well determined ab initio or by experiment. The desorption energies are determined
from the binding energy, as explained below, and depend on an energy parameter E.
The barriers for reaction are used as a parameter to fit the data. The barrier for hopping
(diffusion) from site i to j is assumed to be
∆Ebind (i, j)
.
(2.5)
2
This expression ensures microscopic reversibility between the different types of sites.
The parameter ξ is another input parameter, which is varied between simulations. Little
quantitative information is available about diffusion rates on these kind of surfaces, which
makes the value of ξ uncertain.
Diffusion into the ice is also considered. Minimum energy path calculations suggest
that CO and H can swap position enabling an H atom to penetrate into the CO ice (see
Appendix B). The barrier for this process strongly depends on the layer in which the H
atom is situated. In the simulations, the barrier for this event is (350 + 2(z1 + z2 )) K
for an H atom to swap between layer z1 and z2 . This compares to a hopping barrier of
H,flat→flat
H,flat
Ehop
= 256 K and a desorption energy of Ebind
= 320 K (see § 2.4.2). Hiraoka et al.
(1998) found that hydrogen atoms can relatively easily diffuse through the CO ice. Moreover, the current experiments show that hydrogen atoms can penetrate into a maximum
of four monolayers for 12.0 K. Hydrogen atoms are also allowed to swap with formaldehyde and methanol, but here the initial barrier is chosen to be higher (450 and 500 K)
since these species are heavier and are more strongly bound within the ice matrix.
Ehop (i, j) = ξE +
2.4.2 The CO ice layer
Although the experimental CO layers are probably amorphous (Kouchi 1990), crystalline
layers are used in the Monte Carlo simulations discussed here. In this way, a lattice-gas
Monte Carlo method can be used, which enables far longer simulation times than in offlattice methods. We expect the crystalline assumption to be reasonable since the local
structure of the CO layers is probably close to crystallinity. The energy released during
deposition may help the molecules to rearrange slightly, leading to micro-crystalline domains. The α-CO structure (Vegard 1930) is used with layers in the (110) orientation. The
35
2 Hydrogenation reactions in interstellar CO ice analogues
dominant faces of a CO crystal will have this crystallographic orientation. The CO surface
consists of alternating carbon and oxygen terminated bi-layers. In the bulk configuration,
each CO molecule has 14 nearest neighbors: five in layers below, five in layers above, and
four in the same layer. The additive energy contribution of these neighbors is 2E for the
layers below and E for the neighbors in the same layer or of lower z, where z is the depth
in layers with respect to the top layer. The different treatment for sites below the particle
is to add a contribution for longer range interactions from the ice layer. E is chosen to
H,flat
be 32 K for atomic hydrogen, and 63 K for CO. This leads to a binding energy of Ebind
H,layer
CO,layer
= 320 K for H on top of a flat CO ice layer, and Ebind = 448 K and Ebind
= 882 K
for H and CO, respectively, embedded in a CO layer. These values agree very well with
binding energies obtained by calculations using accurate H-CO and CO-CO potentials of
320, 440, and 850 K, respectively (see Appendix B).
2.4.3 Comparison to the experiment
The solid lines in Fig. 2.6 represent the results from the Monte Carlo calculations. The
exact mechanisms included in these simulations are discussed in more detail in the following sections. The resulting time evolution series are in very good agreement for 12.0 K.
The agreement for 13.5, 15.0, and 16.5 K is far less good, probably because of missing
mechanisms that promote the penetration into the ice. In the current simulations, only
swapping of species is included. Because of the thermal motion of the CO molecules,
“real” penetration in which the H atoms penetrate in the CO matrix may also be possible.
The shape of the curves is reproduced and only the H2 CO abundance levels off at too low
yields.
The main parameters varied to fit the experimental data are the reaction barriers and
the diffusion rates. The best-fit model barriers are summarized in Table 2.2. Since the
intermediate species HCO and H3 CO are not experimentally detected, the barriers for
hydrogenation of these species are significantly lower than for the other two reactions,
presumably even zero. The HCO and H3 CO abundances stay below detectable levels
in the simulations. The reaction barriers for H + CO and H + H2 CO are temperature
dependent and increase with temperature. Our values are in good absolute agreement with
the barriers found by Awad et al. (2005), who also found a similar temperature behavior.
Their values were obtained using a rate equation analysis for T = 10, 15, and 20 K using
the data from Watanabe et al. (2006a). The temperature dependence suggests that there
is a clear tunnelling component for the reaction at low temperature. The two barriers for
forming H2 CO and CH3 OH show different temperature dependencies. The formation of
methanol becomes relatively more important at higher temperature. Note that the Monte
Carlo method automatically treats a reaction in competition with desorption and hopping.
This is in contrast to gas-grain codes, where it must be included explicitly. To describe
the chemical processes properly, one should introduce this competition into the gas-grain
model.
The errors in the energy barriers reflect the errors due to the uncertainties in the sticking probability, H-atom flux, diffusion, and exact structure of the CO ice.
36
2.4 Monte Carlo simulations
Molecular hydrogen is formed on the surface with efficiencies ranging from 3% at
T = 16.5 K to 70% at T = 12.0 K. However, because of the large excess energy of the
formation reaction, the majority of the formed H2 molecules leaves the surface, and the
H2 surface abundance is predominantly determined by impinging H2 molecules.
Table 2.2 Reaction rates and barriers for CO + H and H2 CO + H for different temperatures.
T
(K)
12.0
13.5
15.0
16.5
CO + H
barrier
rate
(K)
(s−1 )
390 ± 40 2 × 10−3
435 ± 50 2 × 10−3
480 ± 60 3 × 10−3
520 ± 70 4 × 10−3
H2 CO + H
barrier
rate
(K)
(s−1 )
415 ± 40 2 × 10−4
435 ± 50 2 × 10−3
470 ± 60 5 × 10−3
490 ± 70 2 × 10−2
2.4.4 Effect of diffusion
Since the diffusion rates are uncertain, this section discusses the effect of diffusion in
more detail. Minimum energy path calculations of the diffusion of a single hydrogen
atom on a CO (110) surface (see Appendix B) results in energy barriers ranging from 70
to 170 K (ξ = 2−5.3) depending on the direction of diffusion. The Monte Carlo program
considers only one type of diffusion between “flat” sites. This corresponds more closely
to the isotropic nature of an amorphous surface. Amorphous surfaces are usually more
corrugated than crystalline surfaces, increasing the hopping barrier. The second term in
Eq. 2.5 ensures microscopic reversibility. Figure 2.10 shows the influence of the diffusion
parameter ξ on the H2 CO and CH3 OH production. The simulations are carried out in the
presence of H2 for 12.0 K (top) and 15.0 K (bottom). The difference in diffusion appears
to have a larger effect at 15.0 K than at 12.0 K. Faster diffusion (smaller ξ) clearly results
in less CH3 OH and H2 CO production, since the H atoms are more likely to find each
other and to react to form H2 . Slower diffusion provides more time for the H atoms per
CO encounter to cross the reaction barrier and form HCO. In the simulations presented
in Figs. 2.6 and 2.11, we use ξ = 8 to reduce the simulation time. This parameter choice
results in a ratio Ehop (flat, flat)/Ebind (flat) of 0.78, which agrees with the experimentally
found ratio for H atoms on olivine and amorphous carbon (Katz et al. 1999). The amorphocity of the surface may be responsible for such a high ratio.
2.4.5 Effect of H2 molecules on the hydrogenation
All simulations include the deposition of both H atoms and H2 molecules, which results
from the undissociated H2 molecules in the H-beam. If the H2 molecules are excluded
37
2 Hydrogenation reactions in interstellar CO ice analogues
from the simulations, the formation of H2 CO and CH3 OH is affected in only a limited
number of cases of both fast diffusion and high temperature. The presence of H2 appears
to have two principle effects: it limits the penetration into the ice and decelerates the H
atoms, since they move through a “sea” of H2 . The first effect reduces the production rate,
whereas the second depends on the reaction barrier.
The experimental results at temperatures higher than 12.0 K show non-first-order behavior at early times (exponential decay of CO). The H2 CO production rate increases until
30 and 50 minutes of exposure for T = 13.5 and 15.0 K, respectively. After this time, the
H2 CO and CH3 OH follow the expected first order behavior. None of the simulations in
Fig. 2.10 show this trend. The only mechanism capable of describing this phenomenon is
an increasing effective H-atom flux with time. This increase in effective flux can be cased
by an increase in the sticking of atomic hydrogen to the surface. Since the incoming H
atoms are relatively warm, they need to dissipate this extra energy into the surface to stick.
Because CO is relatively heavy compared to the H atoms, this energy dissipation will be
inefficient, and most of the H atoms will scatter back into the gas phase. Once the surface
abundance of the much lighter H2 molecules increases, the sticking of the H atoms to the
surface will increase as well. We assume a 1% sticking for H atoms and H2 molecules
on a bare CO surface and a 65% sticking of H atoms on a surface that is fully covered
with H2 . The sticking probability is further assumed to grow linearly with the H2 coverage. The H2 surface abundance reaches a constant level of 0.39 ML after a few minutes
for T is 12.0 K. This results in a sticking of H atoms to the surface of 26%. For higher
temperatures, it is noticeably longer time before to a steady state is reached, explaining
the non-linear behavior at early times and inferring a lower final sticking probability. The
solid lines in Fig. 2.6 include this mechanism.
As mentioned earlier, Watanabe & Kouchi (2002) concluded that the temperature of
the beam has little effect on the hydrogenation process, which seems to contradict our H2
argument. However, their experiments were carried out at 10 K, where the surfaces are
covered with hydrogen atoms early on in the experiment because of the enhanced sticking at low temperatures. They further reported an unknown flux difference between the
cold and warm beam, which makes quantifying the sticking probability using these experiments not possible. In conclusion, the temperature of the beam can affect the effective
flux of H atoms landing on the surfaces, but it does not introduce additional energetic
effects that influence the crossing of the barrier.
2.5 CO hydrogenation under interstellar conditions
Based on the fitting results in the previous section, the Monte Carlo routine can now be
used to simulate CO hydrogenation reactions under interstellar conditions. An important
ingredient is the H-atom density in the cloud. As in our laboratory beam, the gas in
dense clouds consists of a mix of H and H2 . Under steady-state conditions, the balance of
the rates of H2 formation on grains and H2 destruction by cosmic rays infers an H-atom
density of around 1 cm−3 (Hollenbach & Salpeter 1971). This H-atom number density
is independent of the total density because both the formation and destruction rates scale
38
2.5 CO hydrogenation under interstellar conditions
Time [minutes]
0
2
20
40
60
80 100 120 140 160 180
12.0 K
H2CO
0
CH3OH
15
-2
N(t) [10 molecules cm ]
1
-1
ξ=4
ξ=5
ξ=6
ξ=8
-2
2
CO
15.0 K
CH3OH
H2CO
1
0
-1
-2
CO
0
1
2
3
17
4
5
-2
Fluence [10 atoms cm ]
Figure 2.10 Monte Carlo simulations of the time evolution of the surface abundance of
CO, H2 CO and CH3 OH during H-atom bombardment of CO ice at 12.0 K (top) and 15.0 K
(bottom). Reaction barriers for H + CO and H + H2 CO can be found in Table 2.2. The
diffusion is varied via the parameter ξ (Eq. 5).
with density. Before steady-state is reached, however, the H-atom density may be higher
because the timescale for H to H2 conversion is long (∼107 yr), starting from a purely
atomic low-density cloud (Goldsmith et al. 2007). Our model assumes a constant H-atom
39
2 Hydrogenation reactions in interstellar CO ice analogues
density of 10 cm−3 . Our other model parameters are a gas temperature of 20 K and dust
temperatures of 12.0 and 16.5 K. A CO surface is then simulated for 2 × 105 yr, which
corresponds to a fluence of 10.8 × 1017 atoms cm−2 . Note that half of this fluence was
reached in our experiments. Because the H-atom velocities are low, the sticking of H
atoms to the CO ice is not varied in the simulations, but remains constant at 100%.
The starting configuration for the simulations is a layer of pure CO ice. This is believed to be representative for the top layers of the grain mantles at the center of a highdensity collapsing cloud. Here, the ice layer is observed to consist of predominantly CO
ice as the result of “catastrophic” CO freeze-out (Pontoppidan 2006, Pontoppidan et al.
2008). More heterogeneous ice layers are formed at lower densities where CO and H2 O
are mixed, or towards the center of proto-stellar envelops or proto-planetary disks where
the dust has been heated and CO has desorbed from the top layers.
Figure 2.11 (top) shows the resulting time evolution of CO, H2 CO, and CH3 OH ice
(thick lines) for 12.0 K. The thin lines in Fig. 2.11 represent the direct scaling of the
simulations of the experiment on interstellar timescales. The H2 CO/CH3 OH ratio of the
low flux simulation is very different from the scaled experimental simulation. The reason
for this is that in the laboratory environment twice as many hydrogen atoms react with
each other to form H2 than are involved in the four CO hydrogenation reactions since
the surface density is relatively high. For interstellar conditions, the CO hydrogenation
reactions dominate and only <5% of the reacting H atoms are converted into H2 . A
second effect that changes the time evolution of the ISM is the difference in sticking.
Under laboratory conditions, the sticking probability is much lower since the incoming
H atoms at room temperature cannot release their energy very efficiently into the CO ice.
The presence of H2 on the surface may have a positive effect on the sticking probability.
In the ISM, the incoming atoms are much colder and energy dissipation will not be a
limiting factor for the sticking of H atoms into CO ice. This can be modelled using Monte
Carlo simulations but only after deriving the energy barriers by fitting the laboratory data.
The bottom panel in Fig. 2.11 shows similar trends for 16.5 K. Again the onset of
H2 CO and CH3 OH formation is at much lower fluences than in the experiment. At the end
of the simulation, nearly all H2 CO has been converted into CH3 OH. This is in contrast
to the 12.0 K simulations, where a constant non-zero amount of H2 CO remains after
2 × 105 yr. The crossover point from H2 CO-rich to CH3 OH-rich ice occurs at slightly
later times at 16.5 K compared to 12.0 K. This can clearly be seen in Fig. 2.12, which
plots the H2 CO/CH3 OH ratio for both temperatures. At early times, this ratio is similar
for 12.0 and 16.5 K. At t > 103 yr, the ratio starts to level off at 12.0 K, while it still
decreases rapidly at 16.5 K. The noise in the curve for 16.5 K below t = 5 × 103 yr is due
to the low abundances of H2 CO and CH3 OH.
In space, the H2 CO/CH3 OH ice ratio has been determined directly for only three highmass young stellar objects (YSOs): W 33A, NGC 7538 IRS9, and AFGL 70009S, with
inferred ratios ranging from 0.09 to 0.51 (Keane et al. 2001, Gibb et al. 2004). The laboratory curves for the H2 CO and CH3 OH production show that H2 CO is more or equally
abundant during most of our experiments. Thus, values as low as 0.09−0.51 cannot easily
be reproduced in the experiments. However, the Monte Carlo simulations for interstellar
conditions have a crossover from H2 CO-rich to CH3 OH-rich ice at significantly earlier
40
2.5 CO hydrogenation under interstellar conditions
2
1
12.0 K
CH3OH, ISM
H2CO, ISM
H2CO, Exp.
15
-2
∆N(t) (10 molec. cm )
0
CH3OH, Exp.
CO, Exp.
-1
-2
2
CO, ISM
16.5 K
CH3OH, ISM
1
CH3OH, Exp.
H2CO, ISM
H2CO, Exp.
0
CO, Exp.
-1
CO, ISM
-2
1×10
2
1×10
3
1×10
4
1×10
5
Time (yr)
Figure 2.11 Monte Carlo simulations of CO-ice hydrogenation at 12.0 (top) and 16.5 K
(bottom). A constant atomic hydrogen gas phase density of 10 cm−3 and a gas temperature of 20 K is assumed. Thick lines represent interstellar conditions, thin lines are the
scaled experimental simulations. The results are shown as the change in column density
compared with t = 0 yr.
times than the experimental curves and a H2 CO/CH3 OH ratio of 0.51 is obtained after
5 × 103 yr at T dust = 12 K. Grains at higher temperatures will have this crossover at even
41
2 Hydrogenation reactions in interstellar CO ice analogues
earlier times and for grains with T dust = 16.5 K, a H2 CO/CH3 OH ratio of even 0.09 is
obtained after 2 × 104 yr. Thus, the observed ratios are in agreement with the models discussed above for chemical timescales >2×104 yr, which is consistent with the estimated
ages of these high-mass protostars of a few 104 −105 yr (Hoare et al. 2007).
CH3 OH ice has also been detected toward low-mass YSOs with abundances ranging
from <1% to more than 25% of the H2 O ice abundance (Pontoppidan et al. 2003, Boogert
et al. 2008). An interesting example is the Class 0 protostar Serpens SMM 4, for which
a particularly high CH3 OH abundance of 28% with respect to H2 O ice was deduced for
the outer envelope (Pontoppidan et al. 2004). The upper limit to the H2 CO-ice abundance
implies a H2 CO/CH3 OH ratio <0.18, implying an age >1×104 yr at 16.5 K. This is consistent with the estimated timescale for heavy freeze out in low-mass YSOs of 105±0.5 yr,
including both the pre-stellar and proto-stellar phases (Jørgensen et al. 2005).
Other observational constraints come from sub-millimeter observations of the gas in a
sample of massive hot cores, where a constant ratio of H2 CO/CH3 OH of 0.22 ± 0.05 was
found (Bisschop et al. 2007c). If both the observed H2 CO and CH3 OH have just evaporated freshly off the grains and if they have not been affected by subsequent gas-phase
chemistry, the observed ratio should reflect the ice abundances. This ratio is roughly consistent with the asymptotic value reached by the 12 K model. This remarkably constant
abundance ratio implies that very similar physical conditions (e.g., dust temperatures, Hatom abundances) exist during ice formation.
In contrast, since the CH3 OH ice abundance with respect to that of H2 O is known to
vary by more than an order of magnitude, both local conditions and timescales appear to
play a role. Note, however, that for CH3 OH abundances as large as 25% (columns as large
as 1018 cm−2 ), the CH3 OH layer is approximately 25 ML thick ( 0.25 × n(H2 O)/(ndust ×
<binding sites per grain>) = 0.25×10−4 /(10−12 ×106 ) = 25 ML), much more than can be
produced from just the upper 4 ML of the CO ice. Thus, conversion of CO into CH3 OH
ice must in these cases occur simultaneously with the freeze-out and building up of the
CO layer. Pure CO ice can also easily desorb as soon as the protostar heats up. This
complicates the use of CH3 OH/CO ice as an evolutionary probe. A proper model of
interstellar CH3 OH ice formation should therefore include the effects of the variations in
CO-ice abundances and dust temperatures in the pre- and protostellar phases, and account
for the differences in timescales for CH3 OH-ice formation compared with those of CO
adsorption and desorption. This Chapter provides the necessary molecular data to compile
such a model.
2.6 Conclusion
The present Chapter shows that the formation of methanol by successive hydrogenation of
CO and H2 CO is efficient under various laboratory conditions covering T surf = 12 − 20 K,
ice thicknesses between 1 × 1015 and 1 × 1016 molecules cm−2 equivalent to 1 and 10 ML
bulk CO, and H-atom fluxes between 1 × 1012 and 5 × 1013 cm−2 s−1 . Our results show
that the discrepancy between Hiraoka et al. (2002) and Watanabe & Kouchi (2002) was
indeed caused mainly by the use of different H-atom fluxes and we agree with the latter
42
2.6 Conclusion
100
n(H2CO)/n(CH3OH)
12.0 K
10
16.5 K
1
0.1
0.01 2
1×10
1×10
3
1×10
4
1×10
5
Time (yr)
Figure 2.12 The H2 CO/CH3 OH ratio as a function of time obtained from the Monte Carlo
simulations of CO hydrogenation at 12.0 and 16.5 K under ISM conditions (see Fig. 2.11).
The grey box indicates Spitzer ice observations, the black box gas phase observations.
group of authors that CH3 OH is formed at low temperature. On the basis of this, the
surface hydrogenation of CO can now be safely used to explain the majority of the formed
methanol in the interstellar medium, where it is a key molecule in the synthesis of more
complex molecules.
Energy barriers for the H + CO and H2 CO + H reactions are obtained by fitting Monte
Carlo simulation results to the experimental data. Using these barriers, the methanol production is simulated for interstellar conditions. The obtained H2 CO and CH3 OH abundances do not scale directly with fluence because of the different relative importance of
H2 production and CO hydrogenation in space compared with the laboratory, as can be
clearly seen by comparing the thick and thin lines in Fig. 2.11. However, laboratory
experiments are required to derive the necessary rates that serve as input to the Monte
Carlo program. The obtained H2 CO/CH3 OH ratios for the interstellar simulations are in
closer agreement with observational limits than a direct translation of the experimental
observations.
Monte Carlo simulations of the hydrogenation process show that the presence of H2
has three effects: it promotes the sticking of the warm H atoms, it limits the penetration
into the ice, and it slows down the diffusion of H atoms. The first effect will be negligible
under interstellar conditions since the incoming H atoms will be cold already and the
sticking probability will therefore be high regardless of the substrate. The latter two
effects will be important and are similar to the conditions in the laboratory for also a high
H2 abundance.
43
2 Hydrogenation reactions in interstellar CO ice analogues
The experiments show that the hydrogenation process is thickness independent for
layers thicker than 4 × 1015 cm−2 , and that the active layer, which contains only a limited
amount of CO after a steady state is reached, becomes slightly thicker with temperature.
For temperatures higher than 15.0 K, a clear drop in the production rate of methanol is
observed. This is probably due to two effects: the desorption of H atoms becoming important and the sticking of H atoms being reduced because of the low H2 surface abundance.
Both effects cause the H surface abundance to drop substantially at those temperatures
and therefore reduce the probability of hydrogenation reactions occurring in the laboratory. Simulations of CO hydrogenation in space show a strong temperature dependence
of the H2 CO/CH3 OH ratio over several orders of magnitude. The CH3 OH abundance
changes with time, temperature, and fluence.
Appendix A: Absolute and relative H-atom flux determination
Absolute flux determination
The (accuracy of the) absolute value of the H-atom flux at the ice surface is obtained
by estimating lower and upper limits in two independent ways. We exemplify here the
H-atom flux determination for the case of our standard values with an H2 pressure in the
chamber of pH2 = 1 × 10−5 mbar and a filament temperature of T = 2300 K.
The lower limit to the absolute flux is directly available from the experimental results presented in Chapter 3. That chapter discusses the H2 O2 and H2 O production
from H-atom bombardment of O2 -ice in time using the same setup and settings. During the first hour, H2 O2 and H2 O are produced with an almost constant production rate of
6.0 × 1012 molecules cm−2 s−1 . Since both molecules contain two hydrogen atoms, this
means that the H-atom flux should be at least twice this value. Assuming a conservative
sticking probability of hydrogen atoms at 300 K to O2 -ice between 12 and 28 K of at most
50%, we determine a lower limit to the flux of 2.4 × 1013 cm−2 s−1 .
The determination of the upper limit to the H-atom flux is more elaborate and involves
several steps. Figure 2.1 shows that the hydrogen atoms travel from the source through
the atomic-line chamber to a quartz pipe, where the atoms are collisionally cooled and
then travel through the main chamber onto the substrate. The final H-atom flux is then
determined by
NH,source k1 k2 pr
φH =
,
(2.6)
A
where NH,source is the number of hydrogen atoms leaving the source per second, k1 is
the coupling efficiency between the source and quartz pipe, k2 is the coupling efficiency
between the quartz pipe and the ice surface, p accounts for the pressure drop between the
two chambers, r for the loss in H-atoms because of recombinations in the quartz pipe, and
A is the surface area that is exposed by the H-atom beam.
44
2.6 Conclusion
Our specific hydrogen source, used in the experiments described here, has been tested
prior to delivery at the Forschungszentrum in Jülich where the flux, solid angle, and dissociation rate have been measured for a wide range of H2 pressures and filament temperatures. The set-up used for these calibration experiments is described in Tschersich &
von Bonin (1998). These measurements confirmed that there is little variation between
individual instruments, since nearly identical rates were obtained by Tschersich & von
Bonin (1998) and Tschersich (2000) and later by Tschersich et al. (2008) for different
H-atom sources of the same type. From the flux and dissociation rate measured in Jülich,
NH,source can be obtained as well as k1 using the solid angle information. In our example
case, 4.1 × 1016 H atoms s−1 leave the H-atom source and 44% of these atoms enter the
quartz pipe, which is located at a distance of 1.5 cm.
The pipe is designed so that the atoms cannot reach the substrate directly and that
the number of hydrogen recombinations is kept to a minimum. This is achieved by using
a short pipe with a high diameter/length ratio and choosing quartz, which is known to
have a low recombination efficiency. Following Walraven & Silvera (1982), a theoretical
estimate of the number of recombinations in the pipe can be determined, considering the
specific shape and material. This reduces the H-atom flux by another 27 %. The pipe ends
in close proximity to the cryogenic surface. The use of a pipe instead of a pinhole or a slit
results in a focused H-atom beam for which the flux can be determined to relatively low
uncertainty. From geometric considerations, a minimum solid angle can be estimated.
This will suffice, since our aim is to obtain an upper limit to the flux. The H-atom beam
covers A = 4.9 cm2 of the substrate that is located 3 cm behind the quartz pipe. This spot
falls completely on the surface and k2 can readily be assumed to be unity.
Finally, the pressure drop between the source and the main chamber can be determined
in two ways: by a calculation using the conductance of the pipe and the pumping speed,
and by measuring the pressures in both chambers using undissociated beams. Both results
are in reasonable agreement, leading to p = 3.2 × 10−2 .
Our upper limit to the flux is now
φH =
4.1 × 1016 · 0.44 · 1 · 3.2 × 10−2 · 0.73
= 8.6 × 1013 cm−2 s−1 .
4.9
(2.7)
Deviations from this upper limit are expected to be due to a lower k1 value, because of
misalignments between the source and the entrance of the quartz pipe, an underestimation
of the solid angle of the exiting beam from the quartz pipe (lower k1 and higher A), more
recombinations in the pipe or backscattering of atoms from the quartz pipe, to the chamber
of the H-atom source.
The value of flux adopted in the present chapter is the resultant intermediate H-atom
flux of 5 × 1013 cm−2 s−1 , which is within a factor of 2 of the upper and lower limits. It
should be noted that this is a conservative error, since the true lower and higher flux limits
are likely to be higher and lower, respectively.
45
2 Hydrogenation reactions in interstellar CO ice analogues
Relative flux determination
-2
N(t) [10 molecules cm ]
The accuracy in the relative flux is particularly important to the conclusion presented in
this Chapter, more so than the absolute value. For this we use the CO-hydrogenation data
obtained from the experiments. Figure 2.13 shows the CO, H2 CO and CH3 OH evolution
as a function of fluence for three different fluxes. The fluences are calculated using the
flux determination as described above. The three curves clearly overlap, which means
that the accuracy of the relative fluxes is well within our error bars. We conclude that
the accuracy in the relative flux is substantially higher than the accuracy of the absolute
flux, well below 50%. One of the main conclusions of this chapter, that the discrepancy
between the two Japanese groups is due to a difference in flux, as envisaged by Hidaka
et al. (2004), is therefore robust.
Finally, reproducing the same experiments on different days over the course of several
months showed that reproducibility over periods from day-to-day to months is excellent,
to within a few percent.
1
H2CO
0
2
15
13
-2 -1
12
-2 -1
φH = 7x10 cm s
-1
1
-2 -1
φH = 3x10 cm s
CO
0
13
φH = 5x10 cm s
CH3OH
3
15
Fluence [10 cm
4
-2
5
]
Figure 2.13 Time evolution of the surface abundance (molecules cm−2 ) of CO, H2 CO and
CH3 OH during H-atom bombardment of CO ice at 12.0 K with three different H fluxes of
5 × 1013 , 3 × 1013 , and 7 × 1012 cm−2 s−1 .
Appendix B: Binding energy calculations
To calculate binding energies and barriers to diffusion, recently-developed CO–CO and
H–CO potentials are used. Takahashi and van Hemert (in prep.) have fitted high level
electronic structure (coupled cluster) calculations of the CO–CO dimer to an analytic
46
2.6 Conclusion
potential consisting of partial charges on the atoms and the centers of mass of the CO
molecules, atom-based Lennard-Jones type interactions, and Morse potentials for the intramolecular C–O interaction. In the work by Andersson et al. (in prep.), a potential for
the interaction between a hydrogen atom and CO has been calculated by fitting damped
dispersion and exponential repulsion potentials to coupled cluster calculations.
Using the CO–CO potential, a CO (110) surface has been created consisting of 528
CO molecules in 11 monolayers in a cell with dimensions 33.8 Å × 31.8 Å in the surface
plane. By applying periodic boundary conditions, an infinite surface is created. Binding
energies have been calculated by performing energy minimizations for H atoms at different sites on top of and inside the CO surface and by comparing to the energy when the
hydrogen is far away from the surface. In the same manner, the binding energy for a CO
molecule in the top layer has been calculated. In all instances, the top 3 monolayers of
the ice have been allowed to relax.
To calculate the energy barriers to diffusion both on and into the surface, the nudged
elastic band (NEB) method (Jónsson et al. 1998) is used initially to map out the minimum
energy path (MEP) connecting two potential minima. To fine-tune the barrier height, the
Lanczos method is used to optimize the saddle point of the potential energy (Olsen et al.
2004).
47
CHAPTER 3
Laboratory Evidence for Efficient Water
Formation in Interstellar Ices 1
Even though water is the main constituent in interstellar icy mantles, its chemical origin
is not well understood. Three different formation routes have been proposed following
hydrogenation of O, O2 , or O3 , but experimental evidence is largely lacking. We present
a solid state astrochemical laboratory study in which one of these routes is tested. For
this purpose O2 ice is bombarded by H- or D-atoms under ultra high vacuum conditions
at astronomically relevant temperatures ranging from 12 to 28 K. The use of reflection
absorption infrared spectroscopy (RAIRS) permits derivation of reaction rates and shows
efficient formation of H2 O (D2 O) with a rate that is surprisingly independent of temperature. This formation route converts O2 into H2 O via H2 O2 and is found to be orders of
magnitude more efficient than previously assumed. It should therefore be considered as
an important channel for interstellar water ice formation as illustrated by astrochemical
model calculations.
1 Based on: S. Ioppolo, H. M. Cuppen, C. Romanzin, E. F. van Dishoeck, H. Linnartz, 2008, The Astrophysical Journal, volume 686, pages 1474-1479
49
3 Laboratory Evidence for Efficient Water Formation in Interstellar Ices
3.1 Introduction
Solid water ice has been observed on the surfaces of many different astronomical objects. In the Solar System it is found on planets and minor bodies such as comets, transNeptunian objects and Centaurs. In dense, cold interstellar clouds, infrared observations
show that interstellar dust grains are covered with water-rich ices (e.g., Gillett & Forrest
1973, Gibb et al. 2004, Pontoppidan et al. 2004). The formation of these ice mantles is
especially important in the process of star and planet formation, when a large fraction of
heavy elements can be depleted onto grains. In the dense cloud phase water layers form
on the bare grain surfaces. Then during the gravitational (pre-)collapse, virtually all gas
phase species freeze-out on top of these water layers resulting in a CO dominated layer
that likely also contains traces of O2 but very little water.
The observed H2 O ice abundance cannot be explained by direct accretion from the gas
phase only. The exact mechanism by which water ice is formed is not understood. The
Herschel Space Observatory, to be launched in the near future, will provide important
new information on gaseous water in interstellar space and will measure quantitatively
the water abundance as a function of temperature, UV field and other parameters. Furthermore, the Photodetector Array Camera and Spectrometer (PACS) on Herschel will
cover the 62 µm band of solid H2 O. In this way Herschel will provide a unique opportunity to observe the bulk of the water bands that are unobservable from the ground and
relate them to Spitzer and groundbased mid-IR observations of ices in protostellar envelopes and protoplanetary disks. Understanding the processes by which water forms and
why it is not formed under other circumstances will be essential for the interpretation of
these data.
Tielens & Hagen (1982) proposed a reaction scheme in which water ice is formed
on the surfaces of grains via three different routes: hydrogenation of O, O2 , and O3 .
Models predict that water can indeed be formed through such reactions in dense clouds
(e.g., Tielens & Hagen 1982, d’Hendecourt et al. 1985, Hasegawa & Herbst 1993, Cuppen
& Herbst 2007). Using a Monte Carlo approach Cuppen & Herbst (2007) showed that
the contributions of the different formation channels to water ice formation as well as
its abundance strongly depend on the local environment. However, the initial reaction
scheme with the corresponding rates as proposed by Tielens & Hagen (1982) is based
on old, in some cases outdated, gas phase data of the equivalent reactions. Progress
has been severely hampered by the lack of realistic experimental simulations of these
low-temperature, solid state reactions. Preliminary laboratory studies of water synthesis
testing the first reaction channel have been reported by Hiraoka et al. (1998) and by Dulieu
et al. (2007). Both groups investigated the products of D- and O- reactions on an ice
substrate (N2 O and H2 O, respectively) using temperature programmed desorption (TPD).
In experiments exclusively using this technique, it is hard to rule out any H2 O formation
during warm up. Furthermore, quantitative interpretation can be tricky because unstable
species like H2 O2 are destroyed in the mass spectrometer upon ionization leading to an
artificially enhanced H2 O/H2 O2 ratio. This chapter focuses on the H + O2 channel in
50
3.2 Experiment
which O2 is converted to H2 O via H2 O2 :
H + O2 → HO2
(3.1)
H + HO2 → H2 O2
(3.2)
H + H2 O2 → H2 O + OH
(3.3)
H + OH → H2 O
(3.4)
According to Cuppen & Herbst (2007) this channel is, together with the O3 channel,
responsible for water formation in cold, dense clouds. The exposure of O2 ice to hydrogen and deuterium atoms is investigated by means of reflection absorption infrared spectroscopy (RAIRS) and TPD. These techniques allow one to determine formation yields
and the corresponding reaction rates. The present chapter comprises a study of hydrogenation and deuteration reactions of O2 ice for different temperatures between 12 and
28 K, i.e., up to the desorption temperature of O2 (Acharyya et al. 2007). The formation
of H2 O and H2 O2 is observed at all temperatures. An optimum yield is found at 28 K.
3.2 Experiment
Experiments are performed using an ultra-high vacuum set-up (<5×10−10 mbar) which
comprises a main chamber and an atomic line unit. The set-up is discussed in more
detail in Fuchs et al. (2009). The main chamber contains a gold coated copper substrate
(2.5 × 2.5 cm2 ) that is in thermal contact with the cold finger of a 12 K He cryostat. The
temperature can be varied with 0.5 K precision between 12 and 300 K. A precision leak
valve is used to deposit O2 (99.999% purity, Praxair) on the substrate. Ices are grown at
45◦ with a flow of 1×10−7 mbar s−1 where 1.3×10−6 mbar s−1 corresponds to 1 Langmuir
(L) s−1 . In order to compare results from different experiments, the thickness of the O2
ice is 75 L for all samples studied and the substrate temperature is kept at 15 K during the
deposition. An O2 ice of 75 L consists of roughly 30 monolayers. This thickness is chosen
to exclude substrate induced effects. Because a diatomic homonuclear molecule like O2 is
infrared in-active, gas phase O2 is monitored during the deposition by a quadrupole mass
spectrometer (QMS). After deposition at 15 K the ice is slowly cooled down or heated
(1 K min−1 ) until a selected temperature is reached. Systematic studies are performed for
different temperatures between 12 and 28 K.
H(D)-atoms are produced in a well-characterized thermal-cracking device (Tschersich
& von Bonin 1998, Tschersich 2000). A second precision leak valve is used to admit H2
(D2 ) molecules (99.8% purity, Praxair) into the gas cracking line. In each experiment
the H + H2 (D + D2 ) flow through the capillary in the atomic line is 1 × 10−5 mbar s−1
and the temperature of the heated tungsten filament, which surrounds the gas cracking
pipe, is about 2300 K. The dissociation rate and the atomic flux depend on the pressure
and temperature (Tschersich 2000) and are kept constant during all the experiments. A
nose-shaped quartz pipe is placed along the path of the atomic beam in order to cool down
H(D)-atoms to room temperature before reaching the ice sample by collisions (Walraven
51
3 Laboratory Evidence for Efficient Water Formation in Interstellar Ices
& Silvera 1982). The H(D)-atomic flux nearby the sample is estimated, within 50%, as
5 × 1013 cm−2 s−1 . At temperatures of 12 K and higher, no blocking of surface processes
by the presence of H2 is expected in the ice.
The newly formed species after hydrogenation (deuteration) of O2 ice are monitored
by RAIRS using a Fourier transform infrared spectrometer (FTIR) running at a spectral
resolution of 4 cm−1 in the range between 4000 and 700 cm−1 (2.5−14 µm). Typically the
ice is exposed to the H (or D) beam for 3 (or 2) hours and IR spectra are acquired every
few minutes.
Systematic control experiments have been performed in order (i) to unambiguously
confirm that the products are formed by surface processes and not by gas phase reactions,
(ii) to check that any water present in the system does not affect the final results and
(iii) to verify that water and H2 O2 formation occurs in the solid phase after H(D)-atom
bombardment and not by H2 (D2 )-molecules addition. For (i) co-deposition experiments
are undertaken in which H and O2 are deposited simultaneously. Water is only formed if
the surface temperature is below the desorption temperature of oxygen, confirming that
the presence of the oxygen ice is required for this reaction sequence to occur. Point (ii)
is verified by using inert initial substrates like N2 ice to estimate the background water
contribution, as well as by using different isotopologues (18 O2 , 15 N2 and D). Finally (iii)
is checked by using pure H2 (D2 )-beams, i.e., without any H(D) present.
3.3 Results
The formation of both H2 O2 and H2 O ice is confirmed by the appearance of their infrared
solid state spectral signatures. Figure 3.1 shows typical RAIRS results for hydrogenation
and deuteration of O2 ice at 25 K. From top to bottom a time sequence of four spectra
is plotted. These spectra are difference spectra with respect to the initial oxygen ice.
However, since our initial oxygen ice only consists of 30 ML, no features due to the
intrinsically very weak O2 feature (Ehrenfreund et al. 1992, Bennett & Kaiser 2005) are
observed in the original spectrum. Both the H2 O and H2 O2 clearly grow in time as the Hfluence (H-flux × time) increases. Similar features appear for the deuteration experiment,
although here clearly less D2 O is formed. After fitting the infrared spectra with a straight
baseline the column density (molecules cm−2 ) of the newly formed species in the ice is
calculated from the integrated intensity of the infrared bands using a modified LambertBeers equation (Bennett et al. 2004). In the range of our spectrometer, water ice has two
candidate bands for determining its column density, at 3430 cm−1 and 1650 cm−1 (3 and
6 µm, respectively). Since the strong 3430 cm−1 feature overlaps with the 3250 cm−1
band of H2 O2 , the weak feature at 1650 cm−1 was chosen to quantify the water column
density. Since literature values of transmission band strengths cannot be used in reflection
measurements, an apparent absorption strength is obtained from a calibration experiment
in which a pure water ice layer desorbs at constant temperature until the sub-monolayer
regime (Öberg et al. 2007b). The uncertainty in the band strengths remains within a factor
of two. Quantification of H2 O2 is done using the 1350 cm−1 band. As it is experimentally
very hard to deposit pure H2 O2 ice, the apparent absorption strength has to be obtained
52
3.4 Determining the reaction rates
indirectly by assuming the ratio of the integrated band strengths between the two bands in
transmittance to be the same as in reflectance H2 O/H2 O2 = 0.57 (Gerakines et al. 1995,
Loeffler et al. 2006). The band modes of solid D2 O and D2 O2 ices have systematic peak
position shifts of ∼400 cm−1 with respect to the H2 O and H2 O2 bands. The column density
for deuterated species is obtained in a similar way from a calibration experiment, while
the absorption strength for D2 O2 is estimated assuming that H2 O/H2 O2 = D2 O/D2 O2 .
In Fig. 3.2 the column densities of water and H2 O2 are plotted as a function of the
H(D)-fluence (atoms cm−2 ) for different substrate temperatures. The results for H2 O2 and
D2 O2 are found to be very reproducible (errors within the symbols). Due to their low
column densities the relative errors for H2 O and especially D2 O are larger. The H2 O2 and
D2 O2 results all show the same initial linear increase followed by a very sharp transition
to a steady state column density, with the steady state value increasing with temperature.
The water results show similar behavior although the transition is not as sharp and the
temperature dependence of the steady-state value is not as clear. The observation that the
results for all temperatures show the same initial slope means that the rate of the reaction
to H2 O2 is temperature independent. The final yield is however temperature dependent
and this depends on the penetration depth of the hydrogen atoms into the O2 ices; at higher
temperatures H-atoms can penetrate deeper. This is discussed in more detail later.
During the preparation of this manuscript we received a preprint by Miyauchi et al.
(2008) who performed a similar experiment for one single temperature (10 K). Our results
for 12 K turn out to be close to their results, apart from an absolute scaling due to different
assumptions on the band strengths.
3.4 Determining the reaction rates
Reaction rates are obtained by fitting a set of differential equations to the time evolution
curves of H2 O (D2 O) and H2 O2 (D2 O2 ). Usually a diffusive mechanism is considered to
construct these equations. For H2 O2 that is formed from oxygen and converted into water
as given in Eqs. 3.1-3.4 the rate equation would be
dnH2 O2 (t)
dt
=
O2
H
(khop
+ khop
)k1 nH (t)nO2 (t)
−
H2 O2
H
(khop
+ khop
)k3 nH (t)nH2 O2 (t)
(3.5)
X
with khop
and nX the hopping rate and column density of species X and k1 and k3 the rate
constants of reactions (3.1) and (3.3).
Awad et al. (2005) and Miyauchi et al. (2008) applied similar expressions to determine
the rates of the formation of methanol and water, respectively. They solved these equations under the assumption that the atomic hydrogen abundance on the surface remains
constant. In this way expressions for the surface abundances of the reaction products were
obtained with only the effective reaction rates as fitting parameter
nH2 O2 = nO2 (0)
β1
exp(−β3 t) − exp(−β1 t) ,
β1 − β3
(3.6)
53
3 Laboratory Evidence for Efficient Water Formation in Interstellar Ices
with
O2
H
β1 = (khop
+ khop
)k1 nH .
(3.7)
Fitting this expression to the experimental data points in Fig. 3.2 gives a very poor
agreement because the model has an exponential behavior whereas the experiments for
several temperature values clearly do not. This would moreover result in a different rate
for H + O2 for each temperature whereas the experimental curves show that the rate is
independent of temperature. For this reason we decided to use a different model. We
consider two regimes. In the first regime (t < tt ) the hydrogen atoms get trapped into
the ice with an efficiency that is independent of temperature. Once an atom is trapped it
can diffuse efficiently and find an oxygen molecule to react with. This results in a zeroth
order rate. Once nearly all oxygen molecules within the penetration depth are converted to
H2 O2 (t > tt ), diffusion becomes rate limiting and a diffusive mechanism for the reaction
to H2 O2 and H2 O is applied. Since now the ice is changed from an O2 ice to an H2 O2
ice the penetration depth of the H-atoms into the ice will change as well. This model is
described by the following set of equations
nH2 O2
=
nH2 O
=
pO2 − pH2 O2 β01 pH2 O2
+
1 − exp(−β3 t)
pO2
β3 pO2
!
1 − exp(−β3 t)
0 pH2 O2
t−
β1
pO2
β3
β01 t
(3.8)
(3.9)
for t < tt and
nH2 O2
nH2 O
β01 pH2 O2
exp(−β3 t) exp(β3 tt ) − 1
β3 pO2
nO (tt ) pH2 O2
β1 exp(−β3 (t − tt )) − β3 exp(−β1 (t − tt ))
+ 2
β3 − β1 pO2
−nO2 − pH2 O2 + pO2
β0 pH O
= − 1 2 2 exp(−β3 t) exp(β3 tt ) − 1
β 3 pO 2
nO (tt ) pH2 O2
− 2
β1 exp(−β3 (t − tt )) − β3 exp(−β1 (t − tt ))
β3 − β1 pO2
+pH2 O2
=
(3.10)
(3.11)
for t > tt with pX the penetration depth of H-atoms into ice X in units of column density,
β1 and β01 the effective rates of reaction (3.1) and β3 the rate of reaction (3.3). β1 and β3
represent effective diffusive rates that include both the diffusion rate and the reaction rate,
whereas β01 is mainly determined by the hydrogen flux times the efficiency of H trapping
into the ice.
54
3000
2000
2500
-1
Wavenumber (cm )
1500
(c)
(b)
(a)
7 8 9 10
0.004
0.006
0.008
0.010
0.012
4000
1000
(d)
3000
0.000
6
0.000
3500
5
0.002
3
λ(µm)
4
D2O
D2O2
0.002
0.004
0.006
0.008
0.010
0.012
H2O2
∆Absorbance
2500
4
6
7
2000
1500
-1
Wavenumber (cm )
λ(µm)
5
8
1000
(d)
(c)
(b)
(a)
9 10 11
Figure 3.1 RAIR spectral changes of the O2 ice as a function of H-atom (left) and D-atom (right) bombardment at 25 K. Spectra at
a H(D)-atom fluence of (a) 4 × 1015 , (b) 4 × 1016 , (c) 1 × 1017 , and (d) 2 × 1017 cm−2 are given.
∆Absorbance
H2O
3.4 Determining the reaction rates
55
Figure 3.2 The column densities of H2 O2 /D2 O2 (top) and H2 O/D2 O (bottom) as a function of time and fluence for different surface
temperatures. The symbols indicate experimental data, the solid lines represent the fitted model. The left panel is for hydrogenation,
the right for deuteration.
3 Laboratory Evidence for Efficient Water Formation in Interstellar Ices
56
3.4 Determining the reaction rates
Fitting this model to the data, the three rates are found to be independent of temperature. We therefore apply the same average rate to describe the results at all temperatures
and only the two penetration depths are allowed to vary between the experiments. The
resulting curves are indicated by the solid lines in Fig. 3.2. The obtained rates are given
in Table 3.1. The penetration depths are the steady state values in Fig. 3.2. These clearly
increase with temperature to very high values. This suggests that as the O2 ice reaches its
desorption temperature (∼30 K) the structure becomes more open and the O2 molecules
more mobile, allowing the hydrogen atoms to penetrate deeply into the ice. The H2 O2
structure on the contrary is much more dense and rigid and consequently the H(D)-atoms
cannot penetrate more than a few monolayers even at the highest temperatures. The temperature effect is also much less prominent in cases where the ice is comfortably below
its desorption temperature.
Table 3.1 The reaction rates obtained by fitting the model. For the corresponding uncertainties see the text.
H + O2
D + O2
β01
[molec cm−2 s−1 ]
5.4 × 1012
2.5 × 1012
β1
[s−1 ]
3.3 × 10−3
3.3 × 10−3
β3
[s−1 ]
2.7 × 10−3
2.7 × 10−3
A clear difference in penetration depth between the hydrogenation and deuteration
experiments can be observed. For the oxygen penetration depth a difference of a factor of
two is found whereas for hydrogen and deuterium peroxide the difference can be even as
high as a factor of six.
Like in the models by Awad et al. (2005) and Miyauchi et al. (2008), the rates β1
and β3 are the products between the rates of the reactions and the hydrogen surface abundance. The latter is assumed to be constant during the experiment and is the overall result
of accretion, desorption and reaction with both H and O2 . A comparison of β01 and β1
indicates that the reaction with H only plays a minor role.
The uncertainties in β1 and β3 are mainly determined by the fit and are within 50%.
The error in β01 due to the fit is much less, but here the main uncertainty is determined by
the layer thickness. The error in the calibration of the layer thickness is a factor of two.
Assuming that every hydrogen atom that gets trapped reacts with O2 and considering that
two hydrogen atoms are needed to convert O2 to H2 O2 , the trapping efficiency can be
determined from β01 and the flux (5 × 1013 cm−2 s−1 ). This results in a trapping efficiency
of ∼10% for deuterium and ∼20% for atomic hydrogen. This is very close to the sticking
efficiency of hydrogen atoms to a water ice surface of ∼30% under these circumstances
(Al-Halabi & van Dishoeck 2007). The high efficiency of reaction (3.1) is consistent with
recent theoretical studies of this reaction in the gas phase. Xie et al. (2007) and Xu et al.
(2005) found that this reaction can proceed barrierless for certain incoming angles. The
rates β1 and β3 have very similar values. This suggests that also reaction (3.3) is very
efficient and H2 O formation is only limited by the penetration depth of H2 O2 .
57
3 Laboratory Evidence for Efficient Water Formation in Interstellar Ices
The rate of reaction (3.3) shows no significant isotope effect. This is in contrast with
the results by Miyauchi et al. (2008) who found a significant isotope effect for this reaction
using the diffusive model to fit their 10 K results. If a large barrier is involved in reaction (3.3), barrier crossing would proceed via tunneling and an isotope effect is expected
(Watanabe, private communications). The fact that we do not observe a (large) effect at
12−28 K either suggests that the barrier for this reaction is low or that other mechanisms
for the formation of water should be taken into account in the model to fit the data. We
will address this question in Chapters 4 and 5. For now the conclusion remains that the
formation of water from O2 and H is very efficient.
3.5 Astrophysical discussion and conclusion
The hydrogenation and deuteration experiments presented in this letter show an efficient
mechanism to convert O2 ice to H2 O2 and ultimately H2 O. In the model that describes
these experiments, the rate limiting steps for formation are the trapping of the hydrogen
into the O2 ice and the penetration depth of the hydrogen into the H2 O2 . Astrochemical
models take reaction, diffusion and desorption barriers as input. Our data shows that the
formation of at least H2 O2 proceeds via a reaction with a barrier that is much lower than
the value of 1200 K previously assumed by Tielens & Hagen (1982). It should therefore
be considered as a route for water formation on interstellar grain surfaces.
Care should however be taken when extending these findings from a laboratory environment directly to interstellar ices. The temperature independence of the reaction rate
that is observed, for instance, is directly due to the unique property of the O2 ice that
allows hydrogen to penetrate its structure and thereby preventing desorption back into the
gas phase. In interstellar clouds the mantle surfaces would not consist of pure O2 ice but of
a mixture of different species with water as its main constituent (Whittet et al. 1998). The
structure of these “dirty” ices and the binding energies to it would govern the desorption
and diffusion behavior of adsorbates like O2 and H. Experiments on the binding of atomic
hydrogen on water ice surfaces showed that the surface abundance is generally dependent on temperature (Perets et al. 2005, Hornekær et al. 2003, Dulieu et al. 2005). Also
hydrogenation experiments of CO showed that the decreasing H coverage for increasing
temperature results in lower effective rates (Watanabe et al. 2003, 2006b, Linnartz et al.
2007, Cuppen et al. 2008). Here we discuss the implications of the present work to two
different interstellar environments: hydrogenation of an apolar (water-poor) ice mantle
after freeze-out and cold cloud conditions where ice is formed from direct deposition of
H and O.
Observations show that interstellar ice mantles consist of polar (water-rich) and apolar
(water-poor) layers Tielens et al. (1991). Apolar ices are thought to form during freeze-out
in the densest parts of the cloud. In dense gas, most of the atomic oxygen has converted
to O2 which can become subsequently a constituent of this polar phase. The lack of
observed H2 O2 and H2 O means that most of the O2 in the ice does not react to produce
H2 O2 or H2 O. This laboratory work however shows that hydrogen atoms can penetrate
deeply into O2 ices and then react. In apolar interstellar ices the penetration depth is not
58
3.5 Astrophysical discussion and conclusion
-3
1×10
Old model
O+H channel
-6
This work
Barrier for O + O2
O3+H channel
1×10
n(X)/nH
-9
1×10
n(H2O2)
-12
1×10
-15
1×10
O2+H channel
-18
1×10
1
100
10000 1
100
10000 1
100
10000
Time (year)
Figure 3.3 The contributions of the three different channels to H2 O formation in three
models and the H2 O2 abundance, n(H2 O2 ) from the O2 + H channel. (left) Old network
by Tielens & Hagen (1982), (middle) new network without barriers for reactions (3.1) and
(3.3) and (right) new network with barrier for O + O2 reaction. See text for details.
determined by oxygen ice but by the main constituent of the ice mantle, CO. We therefore
have performed additional laboratory experiments in which a mixture of CO and O2 ice
is exposed to H-atoms. These show indeed that only the top few layers are hydrogenated
and that the main part of the ice stays intact in agreement with the observations. Details
of these experiments are reported in Ioppolo et al. (2010b) (Chapter 7).
In cold and translucent clouds H- and O-atoms deposit onto the grain simultaneously.
The species can then react immediately and O2 is converted all the way to water. Here the
penetration depth observed in the laboratory becomes unimportant. What can we learn
from the present experiments then? The fast reaction kinetics justify treating H + O2 and
H + HO2 in a similar way as H + H are treated in astrochemical models, with the difference that probably only hydrogen is mobile for reaction. The results further suggest
that the continuing reactions leading to water proceed with high efficiency. The original
grain surface network by (Tielens & Hagen 1982) includes two more water formation
routes: via OH and via ozone. Under dense cloud conditions the ozone route was proposed to be the most efficient as is shown in the left panel of Fig. 3.3. This figure plots
the contributions for the three different H2 O formation channels using a reaction network
59
3 Laboratory Evidence for Efficient Water Formation in Interstellar Ices
limited to the three water formation routes and without any dissociation reactions. The
model parameters are taken from model M1 by Ruffle & Herbst (2000) and the initial
conditions from Tielens & Hagen (1982) are used (n(H2 ) = 5 × 104 , n(H) = 1.41 and
n(O) = 2.36 cm−2 ). These calculations include a barrier for reactions (3.1) and (3.3). The
present work shows that these barriers are negligible and the middle panel plots the same
model calculation without these barriers. The figure clearly shows that the O2 channel has
a major contribution to the overall water formation rate. Recent laboratory experiments
by Sivaraman et al. (2007) on the temperature-dependent formation of ozone by irradiation of oxygen ices by high energy electrons suggest that mobile oxygen atoms prefer the
O + O pathway over O + O2 even though the O/O2 ratio in the ice is very small. This implies the presence of a barrier for the formation of ozone. The right panel gives the model
results when a small barrier of 500 K for this reaction is considered. The contribution of
the ozone channel is now further reduced.
Figure 3.3 further shows the H2 O2 abundance, n(H2 O2 ), for all models as produced
through the O2 + H route. Boudin et al. (1998) set an observational constraint of the H2 O2
ice abundance toward NGC 7538 IRS9 of 5.2% with respect to the H2 O ice abundance.
All models are consistent with this number.
60
CHAPTER 4
Water formation at low temperatures by surface O2
hydrogenation I: characterization of ice
penetration1
Water is the main component of interstellar ice mantles, is abundant in the solar system
and is a crucial ingredient for life. The formation of this molecule in the interstellar
medium cannot be explained by gas phase chemistry only and its surface hydrogenation
formation routes at low temperatures (O, O2 , O3 channels) are still unclear and most
likely incomplete. In Chapter 3 we discussed an unexpected zeroth-order H2 O production behavior in O2 ice hydrogenation experiments compared to the first-order H2 CO and
CH3 OH production behavior found in former studies on hydrogenation of CO ice. In this
chapter we experimentally investigate in detail how the structure of O2 ice leads to this
rare behavior in reaction order and production yield. In our experiments H atoms are
added to a thick O2 ice under fully controlled conditions, while the changes are followed
by means of Reflection Absorption InfraRed Spectroscopy (RAIRS). The H-atom penetration mechanism is systematically studied by varying the temperature, thickness and
structure of the O2 ice. We conclude that the competition between reaction and diffusion
of the H atoms into the O2 ice explains the unexpected H2 O and H2 O2 formation behavior.
In addition, we show that the proposed O2 hydrogenation scheme is incomplete, suggesting that additional surface reactions should be considered. Indeed, the detection of newly
formed O3 in the ice upon H-atom exposure proves that the O2 channel is not an isolated
route. Furthermore, the addition of H2 molecules is found not to have a measurable effect
on the O2 reaction channel.
1 Based on: S. Ioppolo, H. M. Cuppen, C. Romanzin, E. F. van Dishoeck, H. Linnartz, 2010, Physical
Chemistry Chemical Physics, volume 12, pages 12065-12076
61
4 Water formation at low temperatures by surface O2 hydrogenation I
4.1 Introduction
The presence of water is a prerequisite for the origin of life as we know it and, even though
the ubiquity and abundance of water ice in space (i.e., dense molecular clouds, protoplanetary disks and solar-like systems) is well established by infrared observations (Gillett &
Forrest 1973, Dartois et al. 1998, Whittet 2003, Gibb et al. 2004, Boogert et al. 2008),
the chemical origin of water is not yet well understood. Among the various molecules
detected in the solid phase in dense molecular clouds, water is the dominant component
of interstellar icy grain mantles. In such environments, the observed abundance of water
ice cannot be explained by direct accretion from the gas phase only. Indeed, theoretical
models predict that grain surface reactions play an essential role in water formation (Tielens & Hagen 1982, d’Hendecourt et al. 1985, Hasegawa et al. 1992, Hasegawa & Herbst
1993, Cuppen & Herbst 2007).
Tielens & Hagen (1982) proposed a reaction scheme in which solid water ice is formed
on grain surfaces by hydrogenation of atomic oxygen, molecular oxygen and ozone. Using a Monte Carlo approach, Cuppen & Herbst (2007) showed that the contribution of
the different formation channels strongly depends on the local environment in interstellar
clouds. They concluded that the atomic oxygen channel is the main route in translucent
and diffuse clouds, while the molecular oxygen channel, together with the ozone route,
is more efficient in dense cold molecular clouds. However, these theoretical results on
surface reactions are largely based on gas phase input data and these data should not be
extrapolated directly to the solid phase. A systematic laboratory investigation of water
ice formation at interstellar relevant temperatures (∼10 K) is therefore highly needed to
confirm these reaction routes.
The suggested water formation channel with the smallest number of steps is the hydrogenation of O atoms:
H + O → OH,
(4.1)
H + OH → H2 O,
(4.2)
H2 + OH → H2 O + H.
(4.3)
and
From gas phase data, it is clear that the radical-radical reactions (4.1) and (4.2) proceed
without activation barriers. Recently, Atkinson et al. (2004) reviewed reaction (4.3), assigning it a gas phase barrier of 2100 K, instead of 2600 K as assumed earlier (Schiff
1973). Preliminary investigations on the H + O (D + O) channel in the solid phase have
been carried out by Hiraoka et al. (1998) and Dulieu et al. (2010). Both groups reported
the formation of D2 O by temperature programmed desorption (TPD) through reactions of
D and O atoms on a N2 O matrix and an annealed amorphous H2 O ice, respectively. Since
the TPD technique does not allow to distinguish between water formed at low temperature
and during heating, a quantitative interpretation of these results is not straight-forward.
The hydrogenation of ozone proceeds through the formation of OH and O2 :
H + O3 → OH + O2 .
62
(4.4)
4.1 Introduction
The hydroxyl radical can be further hydrogenated via reactions (4.2) and (4.3). Mokrane
et al. (2009) reported experimental evidence for water formation via ozone deuteration
at 10 K using again TPD for analysis. Also here, the exclusive use of a quadrupole
mass spectrometer for the analysis of the surface reaction final products led to a qualitative study. Recently, Romanzin et al. (2010) investigated water formation via ozone hydrogenation/deuteration at different temperatures, combining infrared spectroscopy with
mass spectrometry as a probe for the identification of the ongoing surface processes (see
Chapter 6).
The hydrogenation of molecular oxygen results in the formation of water through the
following steps:
H + O2 → HO2
(4.5)
H + HO2 → H2 O2
(4.6)
H + H2 O2 → H2 O + OH.
(4.7)
This O2 channel is the best studied solid-phase route to form water in the literature. Reactions (4.5) and (4.6) have no activation barriers Troe & Ushakov (2008), Keyser (1986)
while reaction (4.7) has a gas phase activation energy of ∼1800 K (Baulch et al. 1992).
Reactions (4.2) and (4.3) are also included in this channel. Several groups have explored
the O2 hydrogenation reaction route independently in the solid phase (Miyauchi et al.
2008, Ioppolo et al. 2008, Matar et al. 2008, Oba et al. 2009) and this channel is also the
topic of the present chapter. The experimental procedure used here is similar to the one
described in (Miyauchi et al. 2008, and Chapter 3).
In Chapter 3 we exposed 15 ML (1 ML ∼1015 molecules cm−2 ) of solid O2 to H/D
atoms, covering a large range of ice temperatures from 12 K to 28 K. A comparison between the experimental results presented in Chapter 3 and obtained here, is reported in
§ 4.3.1 which discusses the temperature dependence. For all investigated temperatures
shown in Chapter 3, we observed an initial linear and temperature independent growth of
the products, H2 O2 /D2 O2 and H2 O/D2 O, which corresponds to a zeroth-order formation
rate. The final yield was found to be temperature dependent. Surface reactions are usually
considered to follow second order dynamics, since they depend on the diffusion of two
reactants. Because of the constant H-atom flux in the surface reaction experiments, which
results in a constant abundance of one of the reactants, namely H, this will effectively
lead to first order behavior. Our observation of zeroth order kinetics therefore came as a
surprise. Hence, we decided to use an effective model to fit our results reflecting a zeroth order behavior. We considered two regimes to describe the initial linear temperature
independent growth and the temperature dependent final yield, respectively. In the first
regime, the H2 O2 /D2 O2 is formed following a zeroth-order rate, while H2 O/D2 O follows a
usual first order rate. In the second regime, both products follow first-order behavior. This
method gives the correct functional form to describe the experimental results, whereas a
first-order model over the entire regime does not. This is especially apparent at high temperatures. In contrast, Miyauchi et al. (2008) used a diffusive, first-order model to fit their
hydrogenation/deuteration results, investigating only experiments at a low temperature of
10 K, where the zeroth-order behavior is less prominent.
63
4 Water formation at low temperatures by surface O2 hydrogenation I
Oba et al. (2009) stated that the physicochemical meaning of this model is unclear.
However, in Chapter 3, we explained our two phase model by the following scenario.
Initially, the incoming H atoms diffuse into the ice with an efficiency that is independent
of temperature. Once an H atom is trapped in the ice, it can diffuse efficiently and find an
oxygen molecule to react with. When nearly all oxygen molecules within the penetration
depth, which is temperature dependent, are converted to H2 O2 /D2 O2 , diffusion becomes
rate limiting. According to our model no large isotope effect was found for reaction (4.7).
Using their diffusive model, Miyauchi et al. (2008) did find a significant isotope effect
for this reaction. Our results, however, suggest that the reaction can proceed with a low
barrier or that additional mechanisms for water formation should be taken into account in
the model to fit the data.
In order to resolve this issue, we provide here the experimental verification of our
scenario for a non-standard behavior, i.e., for an initially linear and temperature independent growth of the reaction products with a final yield that is temperature dependent.
For this purpose, the penetration mechanism of H atoms into the O2 ice is studied experimentally and we discuss in detail the different possible origins for the temperature
dependent penetration depth. Our further objective is to study several surface reaction
mechanisms leading to H2 O formation under laboratory conditions for different temperatures, thicknesses and structures of the O2 ice. In Chapter 5 we focus on unravelling
the reaction scheme with an emphasis on alternative H2 O formation routes to explain the
lacking isotope effect.
In the following sections we present the experimental and data analyzing methods,
the results and discussion, which include temperature dependence, thickness dependence,
structural effect, penetration mechanism, and H2 dependence, and – in the last section –
we summarize the main conclusions of this study.
4.2 Experimental and data analysis
4.2.1 Experimental
All experiments are performed in an ultra high vacuum setup (SURFRESIDE), which
consists of a stainless steel vacuum main chamber and an atomic line. Details are available
from Ioppolo et al. (2008), Fuchs et al. (2009). The room temperature base pressure of
the vacuum system is in the 10−10 mbar regime. A schematic view of the experimental
apparatus is shown in Fig. 4.1. The gold coated copper substrate (2.5 × 2.5 cm2 ), placed
in the center of the main chamber, is in thermal contact with a cold finger of a closecycle He cryostat. The substrate temperature is controlled between 12 and 300 K with
a precision of 0.5 K. The absolute temperature accuracy is better than 2 K. An all-metal
leak valve is used to admit gaseous O2 (99.999% purity, Praxair) into the chamber, where
it condenses onto the substrate for temperatures below ∼30 K (Acharyya et al. 2007).
Deposition proceeds under an angle of 45◦ and with a controllable flow between 10−8
and 10−7 mbar s−1 , where 1.3 × 10−6 mbar s−1 corresponds to 1 Langmuir (L). According
to the measurements presented here 1 ML corresponds to 3 L (see § 4.3.4). Diatomic
64
4.2 Experimental and data analysis
homonuclear molecules like O2 are infrared in-active, except when embedded in an ice
matrix (Ehrenfreund et al. 1992). Hence, gas phase O2 is monitored during the deposition
mass spectrometrically using a quadrupole mass spectrometer (QMS) with a Faraday cup,
which is placed behind the substrate and opposite to the atomic source.
Figure 4.1 Schematic top-view of the solid-state experimental UHV set-up.
A second precision leak valve is used to admit H2 molecules (99.8% purity, Praxair) into the capillary of a well-characterized thermal cracking source (Tschersich & von
Bonin 1998, Tschersich 2000, Tschersich et al. 2008), which is used to hydrogenate our
O2 sample through heating the capillary from 300 to 2250 K by a surrounding tungsten
filament. During the experiments the H + H2 flow through the capillary and the temperature of the tungsten filament are controlled and kept constant. A quartz pipe is placed
along the path of the dissociated beam. The nose-shaped form of the pipe is designed to
efficiently thermalize all H atoms to room temperature through surface collisions before
they reach the ice sample. The geometry is designed in such a way that this is realized
through at least four collisions of the H-atoms with the walls before leaving the pipe. In
this way, hot species (H; H2 ) cannot reach the ice directly. Furthermore, previous experiments with liquid nitrogen cooled atomic beams did not show any H/D-atom temperature
dependence in the O2 hydrogenation reaction process (Miyauchi et al. 2008, Oba et al.
2009). It is important to note that the relatively high temperature of 300 K of the incident
H atoms in our experiments does not affect the experimental results, since H atoms are
thermally adjusted to the surface temperature as has been shown in Chapter 2.
In this work, atomic fluxes are measured at the sample position in the main chamber,
following the procedure described in Hidaka et al. (2007). The H-atom flux used for all
our experiments is 2.5 × 1013 atoms cm−2 s−1 for a filament temperature of 2200 K and an
atomic chamber pressure of 1 × 10−6 mbar and confirms the value used within the errors
in Chapter 2 and 3. The atomic beam is normal to the substrate surface except for the
experiment regarding the angle dependence, which is reported in § 4.3.2. Details about
65
4 Water formation at low temperatures by surface O2 hydrogenation I
O3
HO2
(d)
0.0002
0.014
0.012
∆Absorbance
0.0004
0
H2O
0.01
(c)
0.008
1200
1100
1000
H2O2
0.006
HO2
(b)
0.004
O3
0.002
(a)
0
3600
3200
2800
1600 1400 1200 1000
-1
Wavenumber (cm )
Figure 4.2 RAIR spectral changes of the O2 ice at 25 K as a function of the H-atom
fluence: (a) 4.5 × 1015 , (b) 3.7 × 1016 , (c) 7.5 × 1016 , and (d) 1.5 × 1017 atoms cm−2 .
the H- and D-atom flux determination are given in Appendix A.
Ices are monitored by means of RAIRS using a Fourier Transform InfraRed Spectrometer (FTIR), which covers the range between 4000 and 700 cm−1 (2.5−14 µm). A
spectral resolution of 4 cm−1 is used and 512 scans are co-added. RAIR difference
spectra (∆Absorbance) with respect to the initial O2 ice are acquired every few minutes during H/D exposure. Newly formed solid H2 O/D2 O and H2 O2 /D2 O2 in hydrogenation/deuteration experiments are detected using RAIRS. Table 4.1 shows an extensive list
summarizing different ice thicknesses and temperatures for which the O2 hydrogenation
is investigated. Additional experiments are performed to study the effect of ice structure,
the H-atom penetration mechanism into the ice and the role of H2 in the hydrogenation
reaction scheme. Deuteration experiments are also performed for control purposes.
4.2.2 Data analysis
After fitting the infrared spectra with a straight baseline, the column densities (molecules
cm−2 ) of the newly formed species are determined from the integrated intensity of the
infrared bands using a modified Lambert-Beer equation (Bennett et al. 2004):
R
A(ν)dν
NX =
(4.8)
SX
where A(ν) is the integrated absorbance and S X is the corresponding band strength. This
equation can however only be used for thin layers. Teolis et al. (2007) showed that the
proportionality between the optical depth and the ice abundance breaks down for thick
66
4.2 Experimental and data analysis
Table 4.1 List of experiments. The ice thickness is expressed in monolayers (ML); Rdep is
the O2 deposition rate; T dep is the substrate temperature during O2 deposition; T H/D−add is
the substrate temperature during H/D-atom addition; PAL is the atomic line pressure during the H/D-atom exposure; T AL is the tungsten filament temperature; H/D-atom fluence
is the total fluence at the end of the experiment; t is the time of the H/D-atom addition
which includes, in some cases, the ramping time when changing temperature.
O2 Thickness
Rdep
(ML)
(ML min−1 )
Control experiments
co-deposition(a)
–
25(b)
1.5
(c)
35
0.7
25(d)
1.5
100(e)
1.5
15( f )
0.7
Temperature dependence
35
0.7
35
0.7
35
0.7
35
0.7
35
0.7
35
0.7
35
0.7
Structural effect
25(g)
1.5
25(h)
1.5
(i)
35
0.7
Penetration mechanism
( j)
25
1.5
25(k)
1.5
35(l)
0.7
Thickness dependence
1
0.15
3
0.15
5
0.15
8
0.15
12
0.15
25
0.15
35
0.15
70
1.5
90
1.5
H2 dependence
25
1.5
25
1.5
25
1.5
T dep
(K)
T H/D−add
(K)
PAL
(10−6 mbar)
T AL
(K)
t
(min)
2200
2200
2200
2200
2200
2200
2200
H/D-atom fluence
(1017 atoms cm−2 )
(H/D)
1.3
2.7
2.7
–
4.5
1.5
(H)
2.2
2.2
2.2
2.2
2.2
2.7
3.3
60
15
15
15
15
15
60
15
25
25
28
25
1
1
1
1
1
1
2200
2200
2200
300
2200
2200
15
15
15
15
15
15
15
15
18
20
23
25
26
27
1
1
1
1
1
1
1
15
15
15
15
15
25
1
1
1
2200
2200
2200
1.8
0.22
1.9
125
125
130
15
15
15
15→25
25→15
25→15
1
1
1
2200
2200
2200
1.6
2.1
2.5
130
160
190
15
15
15
15
15
15
15
15
15
25
25
25
25
25
25
25
25
25
1
1
1
1
1
1
1
1
1
2200
2200
2200
2200
2200
2200
2200
2200
2200
2.2
2.2
2.2
1.8
2.2
2.2
2.2
1.9
3.0
150
150
150
125
150
150
150
130
200
15
15
15
25
25
25
10
1
0.5
1850
2200
2250
2.2
2.2
2.2
150
150
150
90
180
180
150
300
100
150
150
150
150
150
180
220
(a)
Simultaneous deposition of O2 and D atoms. (b) An inert layer of 15 N2 ice is used instead of solid O2 . (c) D-atom beam. (d) H2
added to the O2 ice. (e) H-atom beam. ( f ) Thin O2 layer deposited at 15 K on top of 100 ML of compact H2 O ice deposited
at 120 K. (g) The ice is annealed at 25 K before H-atom addition. (h) H-atom addition is stopped upon saturation of the H2 O2 .
(i)
H-atom beam at 45◦ with respect to the substrate. ( j) H-atom addition at 15 K is stopped after 31 minutes; the ice is heated to
25 K with 1 K min−1 rate; H atoms are added for 79 minutes. (k) H-atom addition at 25 K is stopped after 31 minutes; the ice is
cooled to 15 K with 1 K min−1 rate; H atoms are added for 109 minutes. (l) H-atom addition at 25 K is stopped beyond saturation
of the H2 O2 after 100 minutes; the ice is cooled to 15 K with 1 K min−1 rate; H atoms are added for 70 minutes.
67
4 Water formation at low temperatures by surface O2 hydrogenation I
0.25
Multilayer regime
T = 154 K
Int. Absorbance
0.2
0.15
H2O
Deviation onset
0.1
0.05
Submonolayer regime
0
50
60
70
80
90
Time (minutes)
Figure 4.3 Isothermal desorption at 154 K of a thick layer of H2 O ice. The straight line
guides the eye to identify the change in the slope of the curve, which corresponds to the
transition from zeroth-order to first-order desorption.
layers. The integrated band area oscillates as a function of the layer thickness due to
optical interference that is caused by the reflection at both the film-vacuum and filmsubstrate interfaces. The results presented in this chapter are all in the linear regime,
where the modified Lambert-Beer equation still applies. This is verified by depositing
pure H2 O ice and checking for non-linearities in the growth rate.
The RAIR difference spectra acquired during an hydrogenation experiment of 35 ML
of solid O2 at 25 K are shown in Fig. 4.2. Both H2 O and H2 O2 integrated band intensities
clearly grow as the H-fluence (H-flux × time) increases. The inset in Fig. 4.2 also shows
the presence of ozone in our RAIR spectra upon H-atom exposure. The ν3 (O3 ) stretching mode peaks at 1038 cm−1 in our spectra and corresponds to the band position when
solid O3 is formed and mixed in a polar environment like H2 O:O2 ices (Cooper et al.
2008). The presence of solid ozone formed upon hydrogenation of O2 ice provides clear
experimental evidence for the incompleteness of the H2 O formation routes considered for
decades in astrochemical models. In Chapter 5 these models are extended with a more
complete reaction scheme. In this chapter, the focus is on the ice penetration mechanism.
Ozone hydrogenation is an efficient formation channel and can contribute to the total water amount in our experiments (Mokrane et al. 2009, Chapter 6). Figure 4.2 shows trace
amounts of HO2 trapped in the ice. The ν3 (HO2 ) band mode peaks at 1140 cm−1 in our
infrared spectra and corresponds to the band position when HO2 is trapped in H2 O ices
(Cooper et al. 2008). Although the ν3 (HO2 ) mode is weaker than the ν2 (HO2 ) mode, the
latter is not visible in our infrared spectra, since it overlaps with strong H2 O2 bending
modes. We did not detect OH radicals here. This may have two reasons; they have either reacted further to form H2 O or H2 O2 or their infrared features in a H2 O:H2 O2 ice
are too broad to be distinguished from the H2 O and H2 O2 OH-stretching band modes. In
68
4.2 Experimental and data analysis
Chapter 5 we present the spectroscopic detection of OH radicals, which have formed in
the ice through surface reactions at an early stage in the reaction scheme, using a different
experimental approach.
Two vibrational modes of the newly formed H2 O, peaking at 3430 and 1650 cm−1 ,
are above the noise-level in our RAIR spectra (Fig. 4.2). The broad 3430 cm−1 OHstretching modes (ν1 and ν3 ) of H2 O overlap with the 3250 cm−1 stretching modes (ν1
and ν5 ) of H2 O2 . The integrated band area of the OH-stretching modes is at the edge of
the linear regime for the experiment carried out at 28 K, where the optimum in the final
yield is found. Thus, we have chosen the bending modes, peaking at 1650 cm−1 (ν2 ) and
1350 cm−1 (ν2 , ν6 and 2ν4 ), to quantify the column densities of solid H2 O and H2 O2 ,
respectively. The band modes of D2 O and D2 O2 ices exhibit systematic peak position
shifts of ∼400 cm−1 towards lower wavenumbers with respect to the H2 O and H2 O2 band
modes. The column densities of the deuterated species are obtained in a similar way as
described above.
Since literature values of transmission band strengths cannot be used directly in reflectance measurements, an apparent absorption strength of the various species is calculated from calibration experiments. The determination of this apparent absorption strength
is set-up specific. The calibration method is described in Chapter 2. In short, a layer
of the selected ice is deposited at a temperature lower than its desorption temperature.
The sample is then linearly heated, close to its desorption temperature. Infrared spectra
are acquired regularly until the desorption of the ice is complete. The transition from
zeroth-order to first-order desorption is assumed to occur at the onset to the submonolayer regime and appears in the desorption curve as a sudden change in slope. Figure 4.3
shows the isothermal desorption of a H2 O ice layer at 154 K. The arrow in the graph
indicates the onset of the deviation from constant desorption. The apparent absorption
strength in cm−1 ML−1 is then calculated by relating the observed integrated area to 1 ML
in the modified Lambert-Beer equation. The largest uncertainty in the band strengths is
due to the uncertainty in the onset of the first-order desorption, which can be affected by
the non-wetting property of the gold substrate (hydrophobic surface) in the case of H2 O,
for instance. We verified our calibration method by repeating the experiments reported
by Fraser et al. (2001) who performed several TPD experiments of H2 O ice deposited
on a gold substrate for different ice thicknesses, where the deposited amount of H2 O ice
was measured using a quartz crystal microbalance. The desorption order and the position of the desorption peak depend on the deposited amount of H2 O ice. We were able
to reproduce their results quantitatively using the H2 O band strength obtained from the
isothermal desorption experiment. We estimate the uncertainty of the band strength to be
within 50%. In this work, this calibration method is applied to both H2 O and D2 O ice,
giving the same band strength value for both.
Pure hydrogen peroxide and deuterium peroxide are experimentally difficult to deposit, because of their chemical instability. In Chapter 3 the apparent absorption strengths
are obtained indirectly, assuming the ratio of the integrated band strengths in reflectance
between H2 O and H2 O2 (D2 O and D2 O2 ) to be the same as in transmittance H2 O/H2 O2 =
D2 O/D2 O2 = 0.57 (Gerakines et al. 1995, Loeffler et al. 2006). The H2 O bending mode
is very sensitive to the local environment in terms of band shape and width and partially
69
4 Water formation at low temperatures by surface O2 hydrogenation I
overlaps with the H2 O2 bending mode at lower wavenumbers in a mixture. Thus, we obtained the H2 O band strength from calibration experiments, integrating the H2 O bending
mode in the range where it does not overlap with the H2 O2 bending mode, and we determined the band strength ratio in a H2 O2 -rich environment independently. This is done
by assuming mass balance in a H2 O2 hydrogenation experiment between the formed H2 O
and reacted H2 O2 molecules. The H2 O2 layer is formed on top of the gold substrate by
co-depositing H atoms and O2 molecules with a ratio H/O2 = 20 and heating the substrate
to above the O2 desorption temperature. Details of this experiment are reported in Chapter 5. We find the ratio of the integrated band strengths between H2 O and H2 O2 (D2 O and
D2 O2 ) to be 0.31±0.10.
4.2.3 Control experiments
The list of experiments in Table 4.1 contains a number of control experiments. These are
performed
1. to verify that new species are formed in the solid phase by surface reactions and not
in the gas phase;
2. to check whether the deposition of background H2 O on the substrate is indeed negligible;
3. to confirm that final products are due to surface reactions involving H/D atoms and
not H2 /D2 molecules;
4. to verify that the maximum penetration depth of the atomic hydrogen into the oxygen ice is not larger than the actual O2 thickness chosen for each experiment;
5. to verify that the gold substrate does not affect the final results.
Point 1 is investigated by the simultaneous deposition of O2 molecules and D at 60 K.
At this temperature O2 molecules do not stick to the substrate long enough to form an
ice layer, while H2 O and H2 O2 could still freeze-out onto the substrate if they would be
formed via gas phase reactions. After 90 minutes of co-deposition no new solid species
are detected on the substrate. Point 2 is estimated by exposing an inert 15 N2 ice layer
instead of solid O2 to H atoms at 15 K for 180 minutes. In this case no newly formed H2 O
and H2 O2 molecules are detected and the background H2 O contamination does not exceed
0.1 ML after 3 hours of H-addition to 15 N2 ice. Furthermore, we repeated the standard
experiment using D atoms instead of H atoms to confirm that solid H2 O is formed in the
ice and does not originate from the background. In this case, as shown in Miyauchi et al.
(2008), Ioppolo et al. (2008), only solid deuterium peroxide and heavy water are formed.
Point 3 is verified by exposing an O2 layer to H2 molecules. After 2.5 hours of H2 addition
at 25 K, no new species are formed.
Concerning point 4, a thick O2 layer (100 ML) is deposited at 15 K and then exposed
to an H-atom beam for 5 hours at 28 K. Acharyya et al. (2007) showed that under ultra
70
4.3 Results and discussion
high vacuum conditions O2 thermal desorption starts at ∼28 K. Therefore, the O2 ice
thickness can change in time at 28 K upon desorption and experiments at this temperature
are not straight-forward. Nevertheless, as in Chapter 3, we find an optimum temperature
for the formation of H2 O and H2 O2 at 28 K. The control experiment performed in this
work shows that the O2 use-up at 28 K, obtained according to reactions (4.7) and (4.2)
(N[O2 ] = 12 × N[H2 O] + N[H2 O2 ], see § 4.3.4), is ∼35 ML. This has to be considered as
a lower limit for the maximum penetration depth of the atomic hydrogen into the O2 ice
at 28 K, because of the thermal desorption of the oxygen layer. However, experiments at
lower temperatures never reached an O2 use-up higher than 35 ML. For this reason we
decided to use an O2 ice thickness of 35 ML for most of our experiments and to investigate
only temperatures below 28 K, where the O2 layer does not desorb (see Table 4.1). This
thickness is higher than the 15 ML thickness used in Chapter 3 (more details in § 4.3.1).
In this work, special care is taken to use low O2 deposition rates to ensure optimal control
of the ice thickness (see Table 4.1). The number of monolayers is calculated assuming a
constant O2 deposition rate. However, during the first minutes of the O2 deposition the
rate is not in a linear regime and, once the deposition valve is closed again, background O2
molecules still present in the main chamber can still contribute to the total ice thickness.
Thus, longer deposition times result in a smaller error in the O2 ice thickness, because the
relative contribution in time of the non-linear deposition rates becomes negligible.
Point 5 is addressed by depositing a thick layer of compact H2 O ice (100 ML) at 120 K
and on top a thinner layer of O2 (15 ML) at 15 K, which is subsequently hydrogenated
at 25 K. In this case the O2 layer will become completely hydrogenated, because the
maximum penetration depth of H atoms at 25 K is ∼15 ML (see § 4.3.4), but the H
atoms cannot reach the gold substrate because of the thick layer of compact H2 O ice
(Dulieu et al. 2010, Matar et al. 2008). The result of this experiment is the same as for
a similar experiment with the O2 ice directly deposited on the gold substrate. Thus, the
gold substrate is found not to affect the O2 hydrogenation process within the sensitivity
limits of our experiment.
4.3 Results and discussion
In a standard hydrogenation experiment, O2 ice is deposited at 15 K with a thickness of
35 ML. After deposition the ice is slowly heated to the required temperature, with a rate
of 1 K min−1 , and subsequently exposed to an H-atom flux of ∼2.5×1013 atoms cm−2 s−1
for several hours.
4.3.1 Temperature dependence
Experiments are performed for different temperatures in the range between 15 and 27 K.
The aim of this section is to study the effect of the ice temperature on the formation rate of
H2 O and H2 O2 and final yield, investigating in more detail than in Chapter 3 temperatures
above 20 K, where an optimum in the final yield was found. Figure 4.4 plots the column
71
4 Water formation at low temperatures by surface O2 hydrogenation I
Time (minutes)
0
50
100
25
150
200
H2O2
15
-2
Column density (10 molecules cm )
20
15
10
5
0
5
15 K
18 K
20 K
23 K
25 K
26 K
27 K
4
3
H2O
2
1
0
0
1
2
17
3
-2
Fluence (10 atoms cm )
Figure 4.4 Temperature dependence of the H2 O2 (top) and H2 O (bottom) column densities
as a function of the H-atom fluence and time of H-atom exposure.
72
4.3 Results and discussion
densities of H2 O and H2 O2 as a function of the H-fluence for all the considered substrate
temperatures. In each experiment, the final H2 O2 yield is higher than the H2 O yield
resulting in smaller relative errors for the H2 O2 data. For all temperatures, the solid
H2 O2 column density shows the same initial linear increase followed by a relatively sharp
transition to a steady state regime, with the final yield increasing with temperature. The
solid H2 O column densities exhibit a similar behavior, but the transition is not as sharp as
in the H2 O2 case and the temperature dependence of the final yield is not as pronounced.
The experimental results presented in this section reproduce those discussed in Chapter 3 for temperatures below 23 K, once results presented in Chapter 3 are scaled with the
H2 O2 band strength value used in this work. For temperatures above 23 K the H2 O and
H2 O2 final yield is found to be higher in this work than in Chapter 3, while the formation
rate is the same in both studies for all the investigated temperatures. As discussed above,
a different O2 deposition rate at 15 K in the two studies (1.5 ML min−1 Chapter 3 vs.
0.7 ML min−1 this chapter) caused an initial different O2 ice thickness (15 ML in Chapter 3 according to the O2 use-up vs. 35 ML this chapter). The different O2 ice thicknesses
may explain the discrepancy between the two studies in the H2 O and H2 O2 final yield
for temperatures above 23 K, where the hydrogen can penetrate deeper into the ice than
15 ML. Details are reported in § 4.2.3 and § 4.3.4.
The initial constant formation rate and the high final yield of tens of monolayers (see
Fig. 4.4) indicate that the hydrogenation reaction of solid O2 is a bulk process that does not
involve the surface of the ice only. H atoms diffuse into the O2 ice with a penetration depth
which is temperature dependent. This is different from the case of CO hydrogenation, in
which only ∼4 ML are involved in the reactions and minimal amounts of final products
are detected in experiments at temperatures higher than 16 K, due to H-atom desorption at
high temperatures (Watanabe et al. 2004, Fuchs et al. 2009). In the case of O2 , however, H
atoms penetrate deeper into the ice at high temperatures (>20 K). Moreover, the constant
formation rate shows that even at high temperatures the desorption of H atoms from the
ice is a negligible process. Thus, H atoms can penetrate deep and stay into the oxygen ice
long enough to react with O2 .
The temperature dependence of the penetration depth can have several origins. It
is possible that for higher temperatures the O2 ice is less rigid and the mobility of the
O2 atoms facilitates the penetration into the ice by moving to the side. However, more
thermal vibration of O2 molecules leads to larger effective radii, which may also hinder
the H atoms while passing. We would therefore expect the penetration depth to have an
optimum temperature (as in the case of CO ice hydrogenation around 15 K) when there
is high H-atom mobility but not too much thermal vibration of O2 molecules, and this has
not been found. This is therefore not likely the dominant mechanism responsible for the
temperature dependence of the H-atom penetration. A second possibility could be that
at higher temperature the oxygen layer slowly converts from amorphous to crystalline
which is more effective to penetrate. A third scenario is that the hydrogen atoms do
not really penetrate the ice, but that mobile oxygen molecules keep replenishing the top
layers. A final possible scenario involves the ratio between the rate of the barrierless
reaction between hydrogen and solid O2 , which is temperature independent, and the rate
of the thermally activated process of hydrogen diffusion into the ice, which increases with
73
4 Water formation at low temperatures by surface O2 hydrogenation I
temperature. In this scenario, at each O2 layer that the H atom passes, it has a probability
to react with the O2 ice or to penetrate deeper into the ice. At low temperature, reaction
is favored; at high temperature, penetration. Thus, for this explanation the penetration
length of an individual atom is a stochastic process and the average length depends on the
temperature of the ice. In the following sections we will discuss in more detail the latter
three scenarios proposed here: restructuring of the O2 ice, replenishing of the top layers
and competition between reaction and diffusion. We will experimentally show that the
competition between reaction and diffusion is the most likely scenario.
4.3.2 Structural effect
The yield of the final products in the experiments at high temperatures (>20 K) show that
hydrogen atoms can penetrate the ice over several monolayers. One possible explanation
for the efficient diffusion of H atoms into the ice for higher temperatures is, as mentioned
before, a higher crystalline fraction in the structure of the ice at higher temperature. To
verify this hypothesis an O2 layer deposited at 15 K is annealed at 25 K before H-atom addition at 15 K. This experiment is meant to investigate the effects of possible irreversible
structural changes, which could take place at temperatures slightly below the desorption
temperature of O2 (Kreutz et al. 2004). Infrared spectroscopy does not allow us to investigate different phases of solid O2 . Nevertheless, if the change in the ice structure would
play an important role in the O2 hydrogenation experiments, one would expect to obtain a
higher final yield of the products than in a standard O2 hydrogenation experiment at 15 K.
However, the H2 O and H2 O2 formation rates and final yield in the annealing and in the
standard experiment at 15 K are all comparable within the experimental uncertainties (see
Fig. 4.5).
It should be noted that if a structural change from amorphous to crystalline in the ice
is present at high temperatures a different incident angle of the H-atom beam could still
affect the rate reaction and final yield of the end products. We compared results from
a 25 K experiment of H-atom addition with an incident angle of 45◦ to those obtained
from an experiment, where the H-atom beam is normal to the substrate surface. In both
cases the geometry of the infrared optics is kept the same, since the H-atom beam was
stopped and the sample turned back to 90◦ to the the H-atom beam each time an infrared
spectrum was acquired. The infrared beam is in both cases wider than the projection of
the H-atom beam on the substrate. Figure 4.6 shows that formation rate and final yield of
the products are the same in both cases. Thus, the higher penetration depth of the H atoms
into the ice at high temperatures is most likely due to one of the two remaining scenarios:
replenishing or reaction vs. diffusion.
As a further extension, Fig. 4.5 shows also an experiment performed at 15 K in which
H-atom addition is interrupted upon saturation of H2 O2 (triangle). This experiment is
performed to confirm that the decrease in H2 O2 after saturation is due to reaction and not
to desorption. We will come back to this point later.
74
4.3 Results and discussion
Time (minutes)
0
25
50
-2
15
100
125
H2O2
3
Column density (10 molecules cm )
75
2
15 K
15 K (annealed at 25 K)
15 K (H beam stopped upon saturation)
1
0
2
H2O
1.5
1
0.5
0
0
0.5
1
17
1.5
2
-2
Fluence (10 atoms cm )
Figure 4.5 The H2 O2 (top) and H2 O (bottom) column densities as a function of the Hatom fluence and time of H-atom exposure for a standard experiment performed at 15 K
(square) compared to an experiment in which an O2 layer deposited at 15 K is annealed at
25 K before H-atom addition (circle). Also shown is an experiment in which the H-atom
addition is stopped upon H2 O2 column density saturation (triangle).
75
4 Water formation at low temperatures by surface O2 hydrogenation I
Time (minutes)
0
25
50
-2
15
100
125
150
H2O2
15
Column density (10 molecules cm )
75
10
5
0
4
o
45
o
90
H2O
3
2
1
0
0
0.5
1
2
1.5
17
-2
Fluence (10 atoms cm )
Figure 4.6 H2 O2 (top) and H2 O (bottom) column densities as a function of the H-atom
fluence and time of H-atom exposure for an experiment performed at 25 K, in which the
H-atom beam has an incident angle to the O2 ice of 45◦ (circle), compared to a similar
experiment, in which the H-atom beam is normal to the substrate surface (triangle).
76
4.3 Results and discussion
H
H
O2
O2
H
H
O2
H
H
H2O
H2O2
H2O
H2O2
Figure 4.7 Schematic representation of two penetration scenarios: (le f t) competition between reaction and diffusion (mainly bottom-up) and (right) replenishing of the top layers
by O2 mobility (top-down).
4.3.3 Penetration mechanism
In the previous sections we gave experimental evidence for the importance of the hydrogen penetration depth into O2 ice ruling out two possible mechanisms. The purpose of
this section is to investigate whether the high H-atom penetration depth is due to the competition between reaction and diffusion or replenishing of the top layers by mobile O2
molecules. The two scenarios are schematically depicted in Fig. 4.7. In the first case, O2
is hydrogenated mainly bottom to top, in the latter top to bottom.
Using thick O2 ices (see Table 4.1), we performed the experiments shown in Fig. 4.8.
In the first experiment we added H atoms to the O2 ice at 15 K until the column density
of the newly formed species was saturated (O2 use-up = 2.5 ML). After interrupting the
hydrogenation we heated the ice to 25 K and added H atoms until a new saturation level
was found (total O2 use-up = 4 ML). We repeated the same experiment hydrogenating
an identical ice at a temperature of 25 K, until saturation (O2 use-up = 13 ML), then
we cooled it to 15 K and hydrogenated it for 70 minutes (total O2 use-up = 13 ML).
In the third experiment we hydrogenated another O2 ice at 25 K, interrupting the Hatom addition before saturation (O2 use-up = 5.5 ML). Then, we cooled it to 15 K and
hydrogenated it for 109 minutes (total O2 use-up = 6.5 ML).
Figure 4.8 shows that the results from these three experiments differ. The final yield
of the newly formed species and the final O2 use-up are not the same in the first two
experiments. The O2 use-up is ∼4 ML for an O2 ice hydrogenated at 15 K and then at
25 K, while it is ∼13 ML for an O2 ice directly hydrogenated at 25 K. In the competition
scenario, the H atoms have to penetrate 2 ML of O2 molecules already converted into
H2 O and H2 O2 , if we hydrogenate an O2 ice at 25 K, which was previously hydrogenated
at 15 K. In this case the final hydrogen penetration depth into the ice is just 4 ML and not
13 ML, because the penetration depth of the H atoms into solid H2 O and H2 O2 is lower
than into pure O2 ice. This result leads us to hypothesize that the hydrogenation follows
77
4 Water formation at low temperatures by surface O2 hydrogenation I
Time (minutes)
0
25
50
75
100
125
150
175
15
10
15
-2
Column density (10 molecules cm )
H2O2
5
0
1.5
15 K
25 K
15 K
25 K
15 K
25 K
1
H2O
0.5
0
0
0.5
1
1.5
17
2
2.5
-2
Fluence (10 atoms cm )
Figure 4.8 The H2 O2 (top) and H2 O (bottom) column densities as a function of the H-atom
fluence and time of H-atom exposure for an experiment in which H atoms are added to the
O2 ice at 15 K (open symbols) until saturation of the column densities and, then, at 25 K
(closed symbols) until a new saturation level was found (circle); the same experiment,
inverting the order of the temperatures for the H-atom addition (triangle); an experiment
in which the hydrogenation of the O2 ice at 25 K is stopped before saturation and, then,
continued at 15 K (square).
78
4.3 Results and discussion
a distribution that starts mainly from the bottom layers of the ice to the top. If instead the
replenishing of the top layers due to a higher O2 molecules mobility in the ice for higher
temperatures would play an important role in the O2 hydrogenation, we would expect to
get the same final yield for both experiments. This is not the case.
In the third experiment we interrupted the hydrogenation process when 5.5 ML of
the ice were converted into H2 O and H2 O2 at 25 K. If the hydrogenation would start
straight from the bottom layers to the top, the top layers of the ice would have not yet
been hydrogenated and we would expect to hydrogenate up to an extra 2.5 ML of ice at
15 K in the second part of the experiment, whereas we observed the hydrogenation of
only 1 ML of ice. The final O2 use-up is probably not reached because the H atoms are
blocked by some of the H2 O and H2 O2 molecules, which are earlier formed in the upper
layer of the ice according to the competition scenario, which involves mainly the bottom
layers and partially the top layers of the O2 ice in the early stage of the hydrogenation.
Thus, all experiments presented so far are in accordance with the following scenario:
first, the H atoms penetrate into the ice without desorbing. The penetration depth is determined by a temperature dependent competition between reaction with O2 molecules and
further diffusion into the ice. In this way, the ice is hydrogenated following a distribution
that involves mainly the bottom layers and partially the top layers at the beginning. Subsequently, the penetration depth of atoms that enter the ice in a later stage is determined
by the competition mechanism until they reach species other than O2 .
4.3.4 Thickness dependence
Different thicknesses are studied in the range between 1 and 35 ML. The O2 ice is deposited at 15 K with a rate of ∼0.3 ML min−1 to better control the ice thickness during
deposition and then heated to 25 K. The ice layer is subsequently exposed to H atoms, in
most of the cases, for 150 minutes with a total H-atom fluence of 2.2 × 1017 atoms cm−2 .
Figure 4.9 plots the column density of H2 O2 and H2 O versus the fluence of impinging H
atoms on the solid O2 layer for different thicknesses.
Again, the initial formation rate of H2 O and H2 O2 is constant and thickness independent, while the final yield is thickness dependent and is affected by the penetration depth
of the H atoms into the ice. The O2 ice is indeed completely converted into H2 O and
H2 O2 for layers thinner than the maximal penetration depth of the H atoms into the ice at
25 K (low oxygen coverage). For layers thicker than the maximal penetration depth of the
H atoms into the ice, some O2 is not hydrogenated (high oxygen coverage). Figure 4.10
shows the final yield for H2 O and H2 O2 formation as well as the O2 use-up as a function
of the initial O2 thickness. The O2 use-up is estimated by adding half of the newly formed
H2 O column density to the H2 O2 column density (N[O2 ] = 21 × N[H2 O] + N[H2 O2 ]), according to reactions (4.7) and (4.2), where, from each O2 molecule, two H2 O molecules
can be formed. Figure 4.10 shows that the maximal penetration depth at 25 K is reached
in the high oxygen coverage regime (>15 ML) and is ∼16 ML.
79
4 Water formation at low temperatures by surface O2 hydrogenation I
Time (minutes)
0
25
50
-2
15
100
125
150
H2O2
15
Column density (10 molecules cm )
75
10
5
0
4
3
2
1 ML
3 ML
5 ML
8 ML
15 ML
25 ML
35 ML
H2O
0.5
1
1
0
0
2
1.5
17
-2
Fluence (10 atoms cm )
Figure 4.9 Thickness dependence on the H2 O2 (top) and H2 O (bottom) column densities
as a function of the H-atom fluence and time of H-atom exposure at 25 K.
The O2 use-up (N[O2 ]) is fitted with the following function:
− 1
N[O2 ] = (aϑ)−2 + (b)−2 2
(4.9)
where a and b are fitting parameters and ϑ is the initial O2 thickness. Equation 4.9 is a
combination of two linear functions describing the O2 use-up in the two extreme regimes.
For low coverage, the yield is expected to be equal to the coverage N[O2 ] = aϑ, with
a = 1. At high coverage, the yield will be equal to the maximum penetration depth, b.
The fit to the data gives a = 1, as we expected, and b = 16 ML. The transition between the
two linear regimes is not sharp. This can be explained by considering that the penetration
length of the H atoms is not one absolute value, but covers a distribution with an average
value of 16 ML at 25 K. In the case of a 16 ML thick O2 ice, the high end of the distribution
will be missing, resulting in a lower average value. This is again in agreement with the
competition scenario. Results from the control experiment of 15 ML of O2 deposited on
top of a compact H2 O layer, plotted in Fig. 4.10, confirm the scenario described above. In
80
4.3 Results and discussion
20
15
-2
Column density (10 molecules cm )
15
10
H2O2
used O2
5
-2
-2 -1/2
fit curve y=((aθ) +(b) )
0
4
H2O
3
2
1
0
15
30
45
60
75
90
Initial O2 ice thickness (ML)
Figure 4.10 Final yields for H2 O2 (top, triangle) and H2 O (bottom, circle) and O2 use-up
(top, diamond) as a function of the initial O2 thickness at 25 K. The line is the fit curve
to the O2 use-up data. Data from the control experiment of 15 ML of O2 deposited on top
of a compact H2 O ice are also plotted and included in the fit.
this case less than 15 ML of O2 are hydrogenated. Correlating the O2 use-up expressed
in monolayers to our O2 deposition flow in Langmuir, we have an independent estimate
of the O2 ice thickness for all our experiments, which leads to a correspondence of 3 L to
1 ML.
For oxygen films thinner than 5 ML, more H2 O is formed initially and almost all H2 O2
formed at low fluence is later converted into H2 O. This can be explained by the fact that H
atoms are more likely to find H2 O2 to react with in thin O2 layers than in thick O2 layers
for the same absolute amount of H2 O2 . This confirms that H2 O formation can proceed
with H2 O2 as a precursor (reaction (4.7)). Figure 4.5 indeed shows that the decrease in
H2 O2 is due to reaction. The figure clearly shows that reaction (4.7) is stopped and H2 O2
is not converted to form H2 O. In Chapter 5 we will discuss in more detail this reaction
channel.
81
4 Water formation at low temperatures by surface O2 hydrogenation I
Time (minutes)
0
25
50
15
-2
Column density (10 molecules cm )
15
75
100
125
150
H2O2
10
5
0
H2/H=90
H2/H=15
H2/H= 9
3
H2O
2
1
0
0
0.5
1
2
1.5
17
-2
Fluence (10 atoms cm )
Figure 4.11 The H2 O2 (top) and H2 O (bottom) column densities as a function of the Hatom fluence and time of H-atom exposure for different H2 /H ratios at 25 K.
4.3.5 H2 dependence
Reaction (4.3) could play a role in the solid phase, even though it has a barrier in the
gas phase. By controlling the H2 flow into the thermal cracking source and the filament
temperature, different hydrogen fluxes and different H2 /H ratios can be realized. The role
of H2 in surface reactions is investigated comparing the results of three similar experiments at 25 K in which the same H-atom flux, but different H2 /H ratios, are used. The
details about these experiments are reported in Table 4.1. From flux measurements (see
Appendix A) we estimate the H2 /H ratio to vary between 9 and 90. If reaction (4.3) is
important and if a significant amount of OH radicals is present in the ice, the H2 O abundance should be higher in the case of more H2 . Figure 4.11 shows that the formation rate
and the final column densities of both H2 O and H2 O2 are the same in all cases within
the experimental errors. This indicates that H2 molecules, formed through surface reactions in the ice or along the quartz pipe or originating from the not-fully dissociated
82
4.4 Conclusion
H-beam, do not have any measurable effect on the O2 reaction channel and consequently
that reaction (4.3) is not an important route for H2 O formation in our experiments.
4.4 Conclusion
An extensive complementary study of Chapter 3 is presented under ultra high vacuum
conditions at temperatures between 15 and 28 K. RAIR spectra are used to analyze the
results and calibration experiments are performed to measure atomic fluxes and band
strengths of the newly formed species. From this quantitative study we draw the following
conclusions:
1. The initial constant H2 O and H2 O2 formation rate is temperature and thickness
independent. The final yield is temperature and thickness dependent and is probably
governed by the competition between the reaction of atomic hydrogen with the solid
O2 and the hydrogen diffusion into the ice, which is temperature and thickness
dependent.
2. The penetration depth of the H atoms into the ice is affected by the competition
between reaction and diffusion and increases with temperature. H atoms penetrate
into the bulk of the ice with a penetration depth (up to ∼35 ML at 28 K), that is
temperature dependent. Once an H atom is trapped into the ice, it diffuses efficiently
and finds an oxygen molecule to react with.
3. The ice is hydrogenated initially mainly bottom-up and partially from the top layers. The penetration length of the H atoms covers a distribution of values with
an average value which corresponds to the maximum penetration depth measured,
which is temperature dependent. Experimental evidence shows that the model used
in Chapter 3 and described in the introduction of the present chapter to fit the experimental data not only reproduces the laboratory data, but also has physicochemical meaning; it explains the behavior of the H-atom diffusion into the O2 ice for
the conditions investigated here. This also corrects statements made in Oba et al.
(2009).
4. H2 O and H2 O2 formation via the molecular oxygen channel proposed by Tielens &
Hagen (1982) is experimentally proven to be efficient under laboratory conditions.
Nevertheless, other reactions should be considered in the reaction route as well.
The presence of solid ozone formed in the ice, under the experimental conditions
described here, gives a clear experimental evidence for the incompleteness of the
H2 O formation routes considered for decades in astrochemical models (more will
be presented in Chapter 5).
5. H2 molecules do not take a relevant part in the reaction scheme at a detectable level,
for the temperature range investigated.
83
4 Water formation at low temperatures by surface O2 hydrogenation I
Heating Power (W)
-2 -1
D-atom flux (atoms cm s )
96
115
135
156
182
213
PAL (mbar)
14
10
-5
1x10
-6
5x10
-6
1x10
-7
5x10
13
10
-7
1x10
12
10
1800
1900
2000
2100
2200
2300
Temperature (K)
Figure 4.12 The D-atom fluxes at the substrate site plotted versus different parameter
settings: temperature of the tungsten filament, heating power and atomic line pressure.
The lines connecting data points guide the eye and are not best-fits.
Appendix A: Absolute atomic flux determination
Absolute atomic fluxes are measured in the main chamber, following the procedure described in Hidaka et al. (2007). Since the sensitivity of the standard 1−200 amu QMS
does not allow an accurate measurement at mass 1, we decided to measure D-atom fluxes.
For this the substrate was removed and the inlet of the QMS was placed at the center of the
chamber in sight of the atomic line, exactly at the position where the ice is hydrogenated.
Once the source was turned on, the increase in intensity of the D atoms was monitored
with the QMS with a Faraday cup. The QMS measurements do not directly give the Datom flux values, but the increase in intensity of the QMS signal, ∆QD , is proportional to
the increase in pressure in the main chamber, ∆PD :
∆PD = a∆QD
(4.10)
We measured the proportionality factor a, which is independent of the gas species,
introducing D2 molecules instead of D atoms in the main chamber, since the D-atom
beam contains a significant amount of D2 molecules and an exact measurement of ∆PD
is not straight-forward from pressure gauges. A constant factor a was determined for
all main chamber and atomic line pressures that were used during the experiments. The
84
4.4 Conclusion
absolute D-atom flux is then obtained from the following expression:
fD =
cD ∆PD hvi cD a∆QD hvi
=
4kB T
4kB T
(4.11)
where cD is the calibration factor for the pressure gauge for D atoms taken from the manual, hvi is the thermal velocity of the D atoms at 300 K, kB is the Boltzmann constant,
and T is the D-atom temperature. By changing the filament temperature and/or the D2
inlet flow, the D-atom flux can be varied between 1012 −1014 atoms cm−2 s−1 . Figure 4.12
shows the D-atom flux values measured at the substrate site for different parameter settings.
The H-atom flux value used in all the experiments presented here, is determined by
comparing the H2 O2 and D2 O2 formation rate, which is flux dependent, in two identical
25 K experiments with a filament temperature of 2200 K and an atomic chamber pressure
of 1 × 10−6 mbar. The resulting H-atom flux at the surface site is a factor of two higher
than the D-atom flux measured. This can be explained by the difference in mass between
hydrogen and deuterium, which affects the atomic thermal velocity, and the cracking efficiency in the source. The final H-atom flux value of 2.5 × 1013 atoms cm−2 s−1 confirms
the value previously estimated and used in Chapter 2 and 3. The relative error for D-atom
fluxes is within 10%, for H-atom fluxes within 50%. The absolute error is within 50%.
85
CHAPTER 5
Water formation at low temperatures by surface O2
hydrogenation II: the reaction network1
Water is abundantly present in the Universe. It is the main component of interstellar ice
mantles and a key ingredient for life. Water in space is mainly formed through surface
reactions. Three formation routes have been proposed in the past: hydrogenation of surface O, O2 , and O3 . In Chapter 3 we discussed an unexpected non-standard zeroth order
H2 O2 production behavior in O2 hydrogenation experiments, which suggests that the proposed reaction network is not complete, and that the reaction channels are probably more
interconnected than previously thought. In this chapter we aim to derive the full reaction
scheme for O2 surface hydrogenation and to constrain the rates of the individual reactions.
This is achieved through simultaneous H-atom and O2 deposition under ultra-high vacuum conditions for astronomically relevant temperatures. Different H/O2 ratios are used
to trace different stages in the hydrogenation network. The chemical changes in the forming ice are followed by means of Reflection Absorption Infrared Spectroscopy (RAIRS).
New reaction paths are revealed as compared to previous experiments. Several reaction
steps prove to be much more efficient (H + O2 ) or less efficient (H + OH and H2 + OH)
than originally thought. These are the main conclusions of this work and the extended
network concluded here will have profound implications for models that describe the formation of water in space.
1 Based on: H. M. Cuppen, S. Ioppolo, C. Romanzin, H. Linnartz, 2008, 2010, Physical Chemistry Chemical
Physics, volume 12, pages 12077-12088
87
5 Water formation at low temperatures by surface O2 hydrogenation II
5.1 Introduction
Water is the simplest stable compound of the two most common reactive elements, O
and H, and is abundantly present throughout the Universe. It is the main component of
interstellar (Whittet et al. 1988, Boogert et al. 2008) and cometary ices (Bockelée-Morvan
et al. 2000) and both types of ices are believed to play an important role in the delivery
of water to Earth in the early times of our Solar System. Water is considered an essential
ingredient for the formation of life but it is surprising that its own formation mechanism
is not fully understood.
Water in the interstellar medium (ISM) is predominantly formed through surface reactions on interstellar dust particles. Three reaction routes have been proposed: hydrogenation of atomic oxygen, molecular oxygen and ozone (Tielens & Hagen 1982). These
formation routes in the solid phase have been the topic of several laboratory studies in
recent years. The hydrogenation routes through atomic oxygen and ozone have been
studied by Hiraoka et al. (1998) and Dulieu et al. (2010); and Mokrane et al. (2009) and
Romanzin et al. (2010). Here we focus on the hydrogenation of molecular oxygen via the
reaction scheme
H + O2 → HO2 ,
(5.1)
H + HO2 → H2 O2 ,
(5.2a)
H + H2 O2 → H2 O + OH,
(5.3)
H + OH → H2 O,
(5.4)
and
as proposed by Tielens & Hagen (1982). Our and other experimental studies of the
hydrogenation of O2 ice indeed showed the formation of hydrogen peroxide and water
(Miyauchi et al. 2008, Ioppolo et al. 2008, Matar et al. 2008), but the results also raised
several unanswered questions. As discussed in Chapter 4, the formation of H2 O2 shows
zeroth order kinetics, whereas first order kinetics are expected. We hypothesized that
penetration of hydrogen atoms into the oxygen ice causes this effect. Molecular oxygen ice has unique properties, as compared to CO and H2 O ice, and allows hydrogen
atoms to penetrate deep into the ice, depending on the ice temperature. In Chapter 4 this
mechanism was indeed shown to explain the observed zeroth order behavior. Our second
puzzling observation was the fact that we did not observe an isotope effect in reaction
(5.3) whereas this is expected due to its relatively large barrier. This point is made later
by Oba et al. (2009) as well. Reactions with barriers at low temperatures generally proceed via tunnelling and this would here result in a faster hydrogenation than deuteration
rate. This was not observed. In Ioppolo et al. (2008) we promised to address this point in
a later paper and here we suggest that reactions (5.1)-(5.4) may not be the only reactions
involved in the formation of water when hydrogenating O2 ice and that the incomplete
reaction network of the model artificially resulted in an isotope-independent reaction rate.
An additional indication for this is the observation that the water ice formation rate does
not seem to increase with the amount of H2 O2 and this is expected if H2 O2 is its only
precursor.
88
5.2 Experimental and data analysis
The present chapter focuses on the reactions involved in the hydrogenation of pure
oxygen ice. This is done by co-deposition experiments of O2 molecules and H atoms.
This is intrinsically different from the method used in Chapter 4 where the O2 ice was
prepared first and then sequentially exposed to hydrogen atoms. By changing the stoichiometric ratios of O2 and H, different stages of the formation route through reactions
(5.1)-(5.4) become experimentally accessible. This gives us the unique opportunity to
probe also the reactive intermediates. In all previous studies only the stable intermediate H2 O2 - and final H2 O-products were recorded. Oba et al. (2009) performed similar
co-deposition experiments using a very high H/O2 ratio with the aim to study the structure of the obtained water ice. This mainly gave information about the final products but
not about the individual reaction routes. Since the conditions in the interstellar medium
vary and also differ from the laboratory conditions — especially in terms of atom fluxes
— it is very important to obtain detailed information about the surface reaction routes.
In a Monte Carlo study of water ice formation in diffuse, translucent and dark clouds,
Cuppen & Herbst (2007) showed that the dominant water formation route is determined
by the environment (temperature and H/H2 ratio). However, the reaction scheme used in
these simulations was based on gas phase data and not tested for surface reactions. In the
present chapter, a range of different O2 /H ratios are applied to probe different hydrogenation stages. Three different astronomically relevant surface temperatures of 15, 20 and
25 K are used to check for thermally activated processes. The highest temperature, 25 K,
is chosen to be just below the desorption temperature of molecular oxygen (Acharyya
et al. 2007). The overall goal is to derive the full reaction network and to constrain reaction rates for the individual reactions. We will show that indeed a number of extra reaction
paths should be considered to complete the initially proposed reaction network and that
the O2 hydrogenation channel is interconnected with the O and O3 production channels.
5.2 Experimental and data analysis
5.2.1 Experimental
All experiments are performed in an ultra high vacuum setup (SURFRESIDE) with a
room temperature base pressure in the 10−10 mbar regime. A detailed description of the
experiment is given in Chapter 4 and here only a brief explanation is given with the focus
on the difference in methodology with respect to the previous study in Chapter 4. Hydrogen atoms and molecules and oxygen molecules are deposited simultaneously on a gold
coated copper substrate in the center of the main chamber which is temperature controlled
by a close-cycle He cryostat. Temperatures as low as 12 K can be reached with a relative
precision of 0.5 K and an absolute temperature accuracy better than 2 K. An all-metal leak
valve is used to admit O2 gas (99.999% purity, Praxair) into the chamber. Deposition of O2
proceeds under an angle of 45◦ and with a controllable flow between 10−8 and 10−7 mbar.
A pressure of 10−7 mbar corresponds to an O2 flux of 2.5 × 1013 molecules cm−2 s−1 (see
Chapter 4).
89
5 Water formation at low temperatures by surface O2 hydrogenation II
A second precision leak valve is used to admit H2 molecules (99.8% purity, Praxair) into the capillary of a well-characterized thermal cracking source (Tschersich & von
Bonin 1998, Tschersich et al. 2008). For a standard flux, this capillary is heated to 2150 K
by a surrounding tungsten filament. A stable H + H2 flow is obtained in this way. The
beam enters the main chamber through a nose-shape quartz pipe, which is designed to
collisionally cool the H atoms to room temperature while keeping the number of recombinations of H to H2 to a minimum and is guided to the surface under 90◦ . The quartz
pipe is designed such that atoms and molecules have a minimum of four collisions before
impinging on the ice substrate and are therefore fully collisionally cooled. The final Hatom flux at the surface is measured to be 2.5 × 1013 atoms cm−2 s−1 under our standard
conditions within a factor of two. By changing the filament temperature and/or the H2
inlet flow, the H-atom flux can be varied between 1012 −1014 atoms cm−2 s−1 . Absolute
vales are determined as described in Chapter 4. Relative flux accuracies are estimated
to be within a factor of 50%. Between the experiments the H/O2 ratio is varied. This is
achieved by varying the O2 inlet flow and keeping the H-atom flux constant.
Ices are monitored by means of RAIRS using a Fourier Transform infrared spectrometer (FTIR) with a spectral coverage between 4000 and 700 cm−1 . A resolution of 0.5 cm−1
is used and 128 scans are co-added for one spectrum.
5.2.2 Data analysis
Although O2 as a diatomic homonuclear molecule is infrared in-active and only gives a
small contribution in a water-rich environment (Ehrenfreund et al. 1992), deposition of
O2 has an effect on the baseline of the RAIR spectra. This can be seen in Fig. 5.1a which
shows a reference spectrum taken after an O2 and H2 co-deposition experiment. The
spectrum is completely determined by the deposition of O2 and the experiments indicate
that the distortion of the baseline is directly proportional to the amount of O2 present in
the ice.
For the H and O2 co-deposition experiments, we assume that the resulting ice at low
H/O2 ratios mainly consists of O2 and that the baseline distortion is similar to the reference
experiment with H2 . Figure 5.1b shows an example spectrum before baseline subtraction
for H/O2 = 2 at 20 K. To correct for the influence of O2 , the baseline subtraction consists
of two steps for H/O2 = 1 and 2. For H/O2 = 10, we assume that most O2 is converted
to H2 O2 and indeed here the baseline distortion is minimal. First, a reference spectra
(Fig. 5.1a) based on the O2 and H2 co-deposition spectrum after a similar fluence of O2
is subtracted. As a second step, which is applied for all H/O2 ratios, a piecewise straight
baseline is subtracted. A resulting spectrum is shown in Fig. 5.1c for a H/O2 ratio of 2 and
a surface temperature of 20 K. This spectrum (with inset) is also shown in the third panel
from the top in Fig. 5.2 and clearly consists of a forest of different features. The bands
that we have been able to identify are indicated in Fig. 5.2 and summarized in Table 5.1.
All intermediate species from the reaction scheme (5.1)-(5.4) are observed as well as O3
which is not part of this scheme. A small O2 feature becomes visible due to interactions
with water. Two unidentified features appear at 1420 and 1430 cm−1 .
90
5.2 Experimental and data analysis
Absorbance
(a)
0.02
0.01
(b)
(c)
0
4000
3500
3000
2500
2000
1500
-1
Wavenumber (cm )
Figure 5.1 RAIR spectra of co-deposited O2 and H2 without baseline subtraction (a), codeposited O2 and H without baseline subtraction (b), and co-deposited O2 and H with
baseline subtraction (c). H/O2 or H2 /O2 is 2 and the surface temperature is 20 K. The
spectra (b) and (c) are displaced on the vertical axis by 0.01 and 0.02, respectively.
The formation trends are followed by integrating the corresponding band area as a
function time. Because of the overlapping features and because bandstrength information
is not available for unstable species like HO2 and OH, no absolute values are given. The
asterisk in Table 5.1 marks the features that have been used for integration and relative
quantification. In the case of overlapping bands, Gaussian fits are used to separate the
individual contributions.
Table 5.1 Assigned infrared features with their corresponding reference
Position1
(cm−1 )
1037
1100
1272
1282
1296
1370 (∗)
1392 (∗)
1550
1590 (∗)
1600 (∗)
1650 (∗)
2810
3270
3240
3400
3426 (∗)
3463 (∗)
3572 (∗)
3581 (∗)
1
Species
Mode
Reference
O3
HO2
H2 O2
H2 O2
H2 O2
H2 O2 (bulk)
HO2
O2
H2 O
H2 O
H2 O (bulk)
H2 O2 (bulk)
H2 O2 (bulk)
H2 O (bulk)
HO2
OH
OH
H2 O2
H2 O2
ν3
ν3
ν6
ν6
ν6
ν2 , ν6 , 2ν4
ν2
ν1
ν2
ν2
ν2
2ν6
ν1 , ν5
ν1 , ν3
ν1
ν1 (OH-stretch)
ν1 (OH-stretch)
ν5 (OH-stretch)
ν5 (OH-stretch)
Sivaraman et al. (2007)
Bandow & Akimoto (1985)
Catalano & Sanborn (1963), Lannon et al. (1971)
Catalano & Sanborn (1963), Lannon et al. (1971)
Catalano & Sanborn (1963), Lannon et al. (1971)
Giguère & Harvey (1959)
Bandow & Akimoto (1985)
Loeffler et al. (2006)
Bandow & Akimoto (1985)
Bandow & Akimoto (1985)
Gerakines et al. (1995)
Giguère & Harvey (1959)
Giguère & Harvey (1959)
Gerakines et al. (1995)
Bandow & Akimoto (1985)
Acquista et al. (1968)
Acquista et al. (1968)
Catalano & Sanborn (1963), Khriachtchev et al. (2000)
Catalano & Sanborn (1963), Khriachtchev et al. (2000)
Asterisks mark the features used to the determine the integrated absorption.
91
5 Water formation at low temperatures by surface O2 hydrogenation II
3500
0.02
3000
1500
H/O2 = 100
H2O2 + H2O
0.01
H2O2
H2O2
H2O
0
0.02
H/O2 = 10
Absorbance
0.01
0
0.02
0.01
HO2
H/O2 = 2
H2O2
in O2
H2O
H2O2 in O2
0.01
H2O2 bulk
O2
HO2
0.005
0
OH
1500
HO2
1300
1100
H2O2 + H2O
0
0.02
H/O2 = 1
0.01
0
3500
3000
1500
-1
Wavenumber (cm )
Figure 5.2 RAIR spectra of H and O2 co-deposition experiments performed for a surface
temperature of 20 K and different H/O2 ratios of 100, 10, 2, and 1 from top to bottom.
The H-atom fluence is the same for all spectra.
92
5.3 Results
The spectral appearance of both the H2 O and H2 O2 bands strongly depends on the
environment. In an oxygen-rich environment the bands are narrow. In the remainder
of the chapter we will refer to these features as monomer bands, since they are mainly
due to single H2 O or H2 O2 molecules in a hydrophobic environment, in this case the O2
matrix. By increasing the amount of hydrophilic material in the ice, the bands broaden
and the peak positions shift. These we define as bulk bands since they are caused by H2 O
or H2 O2 in an H2 O- or H2 O2 -rich environment. Multiple infrared studies have shown
the presence of both bulk and monomer features (Giguère & Harvey 1959, Catalano &
Sanborn 1963, Lannon et al. 1971, Bandow & Akimoto 1985, Gerakines et al. 1995,
Khriachtchev et al. 2000). For H2 O2 the bulk and monomer contributions are separated
(3572 and 3581 cm−1 vs. 1370 cm−1 , respectively). The OH-stretch monomer features
of H2 O at 3724 and 3732 cm−1 are not observed and we therefore conclude that water
is not abundantly formed in O2 -rich environments. Only the integrated absorption of
the bulk water feature at 1650 cm−1 is given for O2 -poor environments. In general the
water estimation has the largest error, since the 1550−1700 cm−1 range is affected by O2
baseline distortion and the broad bulk H2 O2 feature.
5.3 Results
5.3.1 H/O2 ratio dependence
The ratio between the deposition of H atoms and O2 molecules determines the hydrogenation grade. Four H atoms are required to fully hydrogenate O2 to two H2 O molecules. The
top panel of Fig. 5.2 shows the RAIR spectrum of the highest H/O2 ratio that we can reliably reach, which is H/O2 = 100. This spectrum is clearly dominated by broad H2 O and
H2 O2 bands. In this experiment roughly equal amounts of H2 O and H2 O2 are produced.
Oba et al. (2009) produced even more H2 O dominated ices in this way using a higher
H/O2 ratio of 500.
In the present chapter, we are particularly interested in the oxygen dominated regime,
where full hydrogenation cannot be reached. Because of the constant supply of O2 , intermediate species are locked in the ice mantle. In this way, all intermediate species listed
in Table 5.1 — HO2 , H2 O2 , and OH — can be observed. The first intermediate, HO2 , is
clearly present in a very oxygen-rich environment. This is reflected by monomer bands at
1100, 1392, and 3400 cm−1 in the spectrum of the bottom panel of Fig. 5.2 for an H/O2
ratio of 1. In the same spectrum, also H2 O2 and OH features appear (both monomer features). The presence of the OH features is rather surprising since OH is only formed in
reaction (5.3) of the proposed reaction scheme and this reaction is expected to be reached
at a higher level of hydrogenation. Furthermore, H2 O, which is formed in the same reaction, is not abundantly present in this spectrum. The third panel of Fig. 5.2 shows a
spectrum obtained after exposure of the same H-atom fluence but with an O2 flow that
is reduced by a factor of 2. Here, the HO2 features shrink, whereas the H2 O2 and OH
signals appear to increase slightly. The H2 O features are small and do not grow and are
mostly likely due to background water in the chamber since control experiments of H2
93
5 Water formation at low temperatures by surface O2 hydrogenation II
and O2 co-deposition result in similar amounts of H2 O. If the oxygen flow is further reduced (second panel, Fig. 5.2), broad bulk water bands can be clearly identified, which
are consistent with H2 O formation and in addition the H2 O2 bands broaden and shift. The
spectral features of the intermediates OH and HO2 disappear entirely.
5.3.2 O3 detection
For specific conditions, O3 can be detected as well. Figure 5.3 zooms in on the 1038 cm−1
ozone band for three different temperatures and three different H/O2 ratios. This O3 band
is rather broad and appears to consist of several contributions. As discussed in Sivaraman
et al. (2007) it is very sensitive to the local environment and can shift over more than ten
wavenumbers. Ozone appears to be predominantly present in the low temperature spectra
(15 and 20 K) and for low H/O2 ratios, or at the opposite conditions: high temperature
(25 K) and high H/O2 ratio. Its presence indicates that oxygen atoms are involved at
some stage in the reaction scheme, since ozone is formed from oxygen atoms and oxygen
molecules
O + O2 → O3 .
(5.5)
Two possible O-atom formation routes are through the hydrogenation of HO2
H + HO2 → H2 O + O,
(5.2c)
and the reaction OH with molecular oxygen
OH + O2 → HO2 + O.
(5.6)
Both reactions are discussed in more detail in § 5.4.1.
Sivaraman et al. (2007) also observed a temperature dependence for O3 production
after electron bombardment of an O2 ice. They attributed this to O atoms which are more
likely to react together to form O2 than to form O3 with O2 , even in an O2 dominated
environment. The amount of formed O3 decreases with temperature in oxygen-rich environments for this reason. At higher temperatures, O atoms become mobile and are more
likely to find reactive species like O atoms before reacting with O2 . Similar processes
may be at play here and give a similar temperature dependence at low H/O2 ratios.
For high H/O2 ratios, reaction (5.5) competes with
O + H → OH,
(5.7)
which should proceed without any barrier. We expect that the relative contribution of
reaction (5.5) increases with temperature since the lifetime of H atoms on the surface, responsible for the competing reaction, decreases. A second possible mechanism that could
be responsible for the detection of ozone at high temperatures is the increased penetration
of hydrogen atoms into the O2 ice with temperature as discussed in Chapter 4. Oxygen
atoms are formed through hydrogenation reactions as is addressed in more detail in § 5.4
of the present chapter. At high temperatures H atoms can penetrate deeper into the ice and
94
5.3 Results
therefore oxygen atoms form deeper in the ice, which in turn lead to deeply embedded
ozone molecules. The chance of hydrogenating species that are positioned deep in the ice
is lower than for surface species, even at high temperatures, since newly formed products
at the surface and the constant deposition of O2 can block further penetration. The ozone
molecules therefore remain embedded in the ice, whereas at lower temperatures they can
react further. Ozone can react with hydrogen atoms to form OH and O2
O3 + H → OH + O2
(5.8)
which both can react further to water. The hydrogenation scheme of pure ozone ice is
the topic of a separate paper and confirms water formation upon hydrogenation of a pure
O3 ice (Chapter 6). We expect that the detection of ozone for high H/O2 ratios and high
temperatures is due to a combination of a more effective formation and a less effective
destruction at high temperatures.
Ozone is also detected in Chapter 4 upon hydrogenation of a pre-deposited pure O2
ice and its abundance is observed to increase with ice temperature. The experimental conditions in Chapter 4’s experiments can be best compared to the high H/O2 conditions of
the present chapter. We therefore expect that the effect of penetration at high temperatures
is the dominant mechanism responsible for the ozone formation.
20 K
25 K
H/O2 = 10
15 K
O3
0.001
H2O2
HO2
O3
H/O2 = 2
0.001
0
H/O2 = 1
Absorbance
0
0.001
0
1200
1100
1000
1200
1100
1000
1200
1100
1000
-1
Wavenumber (cm )
Figure 5.3 RAIR spectra of H and O2 co-deposition experiments performed for three
different surface temperature (15, 20, and 25 K) and three different H/O2 ratios (10, 2,
and 1). The spectra are zoomed in on the 1038 cm−1 ozone region.
95
0
50 100 150
50 100 150
0
Time (minutes)
0
50 100 150
HO2 in O2
OH in O2
H2O2 in O2
25 K
(b)
Int. absorbance (cm )
-1
Int. absorbance (cm )
-1
0.0
0.5
1.0
0.0
1.5
0.5
1.0
0.0
1.5
0.5
1.0
1.5
0
50 100 150
15 K
50 100 150
0
Time (minutes)
0
20 K
50 100 150
H2O2 (bulk)
2 x H2O (bulk)
25 K
Figure 5.4 The integrated absorbance for (a) H2 O2 in an O2 -rich environment (circles), OH (squares), and HO2 (triangles) and for
(b) the 1370 cm−1 H2 O2 bulk and the 1650 cm−1 H2 O bulk features as a function of time for three different surface temperatures and
three different H/O2 ratios. The H2 O bulk integrated absorbance (panel (b)) is multiplied by a factor of two.
(a)
0.00
0.05
0.10
0.00
0.05
0.10
0.00
0.05
20 K
H/O2 = 10
H/O2 = 2
H/O2 = 1
H/O2 = 10
H/O2 = 2
96
H/O2 = 1
0.10
15 K
5 Water formation at low temperatures by surface O2 hydrogenation II
5.3 Results
5.3.3 Time/fluence dependence
The production of H2 O2 , OH, HO2 and H2 O is followed by integration of their time resolved infrared features. Figure 5.4a plots the time evolution for the integrated absorbance
of the monomer features of H2 O2 , OH, and HO2 in O2 at different temperatures (15, 20,
and 25 K) and H/O2 ratios (H/O2 = 10, 2, and 1). Figure 5.4b shows the corresponding
evolution of the H2 O2 and H2 O bulk features. Note that the latter signals are stronger. As
mentioned before, the H2 O features in O2 are not observed and the bulk water abundance
is only shown for H/O2 = 10, since for the low ratios, the observed water bending features
are not distinguishable from the background contributions.
The H2 O2 monomer features (black diamonds in Fig. 5.4a) follow the same trends and
curve shapes as the OH abundance (squares in Fig. 5.4a). Also the H2 O and H2 O2 bulk
features (Fig. 5.4b) seem to follow each other, although not as tightly. The HO2 abundance
has its own distinct behavior. The three different trends are discussed separately below,
starting with HO2 . The integrated intensities are plotted as a function of time and not of
fluence, since two different species (H and O2 ) are simultaneously deposited during these
experiments. After 180 minutes an H-atom fluence of 3 × 1017 atoms cm−2 is reached; the
total O2 fluence depends on the H/O2 ratio.
HO2 monomer features.
The HO2 abundance is only detectable for low H/O2 ratios and appears to exhibit only a
small temperature dependence, with 20 K as a rough estimate for the optimum temperature. The total production rate of species in general consists of different components and
depends on the balance between several formation and destruction reactions. The overall
rate of each individual reaction (production rate) is determined by the rate at which the
reactants meet (meeting rate) and by the probability that these species react upon meeting each other (reaction rate). The first depends on the diffusion and desorption rates
of the reactants; the second on the existence of a reaction barrier and the likelihood to
cross this barrier if necessary. The meeting rate first increases with temperature since
the species will become more mobile, but once the desorption temperature of (one of)
the reactants is reached, it decreases again. The reaction rate is probably independent of
temperature when no barrier exists or when the reaction proceeds through tunnelling; in
the case of a thermally activated reaction, the reaction rate will increase with temperature.
In the present chapter, we will try to disentangle both contributions (meeting vs. reaction
rate). For the purpose of astrochemical models, the reaction rates are used as direct input
parameters.
Let us consider to the production rate of HO2 . Since the production of H2 O2 and OH
(monomers) is higher at lower temperatures, the reason for the reduced HO2 abundance
at lower temperature lies probably in the more efficient destruction and not in the reduced
formation of HO2 . At 25 K, the lifetime of H atoms on the surface is significantly shorter
than at 20 K and this is probably the rate limiting factor for HO2 production at higher
temperatures. These arguments suggest that the HO2 formation rate is actually temperature independent, i.e., the observed temperature dependence of the production rates is
97
5 Water formation at low temperatures by surface O2 hydrogenation II
because of a temperature-dependent meeting rate. This is in agreement with gas phase
calculations of reaction (5.1), where for certain incoming angles no barrier was observed
(Sellevag et al. 2008, Troe & Ushakov 2008).
During the co-deposition experiment an ice builds up slowly and surface reactions
will predominantly occur in the top layers, determined by the temperature dependent
penetration depth as discussed in Chapter 4. If the lower layers of the ice are completely
inert, one would expect the absorbance for all species to grow linearly with time. The
HO2 absorbance clearly levels off at later times (Fig. 5.4a), which suggests that some
HO2 is destroyed in the ice. Cooper et al. (2008) suggested a destruction channel via
HO2 + HO2 → H2 O2 + O2
(5.9)
in H2 O+O2 UV irradiated ices. The HO2 radicals, in that study and here, are formed
through reaction (5.1). The hydrogen atoms originate from different sources (H-atom
beam vs. photolysis). Cooper et al. (2008) found reaction (5.9) to be dominant in the case
that HO2 was formed in confined O2 clusters where they were in close vicinity of other
HO2 radicals and the radicals did not have to travel over large distances in order to meet.
In our O2 dominated ices, HO2 radicals are probably formed homogeneously across the
ice and the HO2 will therefore be, for similar densities, at larger average distances from
each other and need to diffuse through the ice before they can react together. We do not
see evidence for an increase in the destruction of HO2 with temperature which would
correspond to a thermally activated process such as diffusion. Furthermore, the products
of reaction (5.9), H2 O2 and O2 , would result in an increase of the H2 O2 monomer features
at the same time that the HO2 disappears. However, these features appear to decrease
instead of increase. We therefore conclude that HO2 most likely falls apart in H atoms
and oxygen molecules.
In Chapter 4, HO2 is observed at the end of hydrogenation experiments at high temperatures (T > 25 K). We expect that HO2 under these circumstances is formed deep in
the ice and that the destruction of HO2 by reaction with H atoms is limited in the same
way as the destruction of ozone, as explained earlier.
H2 O2 and OH monomer features.
The H2 O2 monomer features and the OH abundance follow the same trends and are discussed together. These features are more temperature and H/O2 ratio dependent than the
HO2 features. For H/O2 = 1, they initially increase, then decrease and reach a steady state
for the investigated temperatures, whereas for H/O2 = 2, they only increase, although not
linearly. Since both features follow each other rather tightly, OH and H2 O2 are probably
formed and destroyed by related processes. This implies that OH is formed earlier in the
reaction scheme than through reaction (5.3). We will come back to this later. The decrease
of both the OH and H2 O2 signals (Fig. 5.4a) appears to coincide with the growth of the
H2 O2 bulk and H2 O contributions (Fig. 5.4b). This is a sign typical for segregation and is
caused by diffusion of H2 O2 , by O2 leaving the H2 O2 matrix ,or a combination of both.
Since the interaction between H2 O2 and the O2 matrix is rather weak, H2 O2 may have a
98
5.4 Implications for the reaction network
higher mobility than usually observed in a hydrophilic environment. The mobility of O2 is
probably also rather high, since the temperature is close to the desorption temperature of
O2 . This makes segregation through O2 diffusion the most plausible scenario (Acharyya
et al. 2007). In a similar fashion mobile OH can react with another OH or H2 O2 to form
H2 O2 or H2 O, respectively, through
OH + OH → H2 O2
(5.10)
OH + H2 O2 → H2 O + HO2 .
(5.11)
and
The first reaction is probably rather inefficient judging from the amount of OH that is
present in the co-deposited ices. The hydroxyl radicals are formed in each others vicinity,
since they are formed in pairs in the same reaction (see reaction (5.2b) in the next section) and the reaction of the two OH radicals is therefore not diffusion limited, but limited
by the reaction probability which does not have a 100% efficiency. Another possibility
for mobile OH would be to cluster inside the H2 O2 bulk aggregates. This will probably
lead to a shift and broadening of the OH features causing them to overlap with the broad
3300 cm−1 band. The disappearance of the 3426 and 3463 cm−1 OH features therefore not
necessarily means that OH itself disappears but it may be due to an overlap with the polar
bulk features when OH itself is in a more polar environment. Probably a critical amount
of OH and H2 O2 needs to be present before segregation occurs (Öberg et al. 2009a). This
would explain why the disappearance of the OH and H2 O2 monomer features becomes
more effective at later times. As mentioned in Cooper et al. (2008), Öberg et al. (2009c)
the mobility of OH is thermally activated and only becomes accessible in a water matrix
above 80 K. In an oxygen matrix, which is less rigid, this could proceed at lower temperatures. The present data indicates this to be around 25 K. The strong decrease in OH
and H2 O2 monomers and the increase in H2 O2 bulk and water at 25 K reflects indeed
increased mobility of OH and H2 O2 .
Summarizing § 5.3.3, HO2 forms in a barrierless reaction from H and O2 and it either
reacts further to OH and H2 O2 or it slowly falls apart in H and O2 . Bulk H2 O and H2 O2
are mostly formed for a high H/O2 ratio and they appear to form mostly at later times,
which is consistent with their formation in a late stage of the reaction scheme.
5.4 Implications for the reaction network
In this section a consistent reaction scheme is derived that explains the experimental observations described in the previous sections. This scheme is schematically presented in
Fig. 5.5. This figure indicates the three initially proposed hydrogenation channels: O, O2 ,
and O3 hydrogenation by the black arrows. These three channels run vertically in three
columns and have the last step in common: reaction (5.4) or
H2 + OH → H2 O + H
(5.12)
to form H2 O from OH. In this section we add the reactions indicated by the light gray
and dark gray arrows to this scheme. The arrow type (solid, dashed or dotted) reflects
99
5 Water formation at low temperatures by surface O2 hydrogenation II
O
O
O2
OH
H
H
H
H H
HO 2
O
O3
H
H2
H
H2 H
H2 O 2
H
OH
H
H
H2
H 2O
Figure 5.5 A schematic representation of the reaction network as obtained from the
present study. Four types of reactions are distinguished: efficient, effectively barrierless,
reactions (solid), reactions with a barrier but with detectable efficiency (dashed), reactions of which the efficiency is below the detection limit (dash-dotted), and reactions of
which the efficiency could not be determined in this study (dotted). The light gray arrows
indicate the same entering channel but with different outgoing channels, and the black
arrows the reactions which were in the original reaction scheme.
the efficiency of the reaction. The solid arrows in Fig. 5.5 indicate the reactions that are
effectively barrierless at low temperatures, the dashed lines proceed with a barrier but
have a detectable efficiency, the dash-dotted arrows correspond to reactions that proceed
below the detection limit, and the dotted arrows indicate reactions which were observed
to proceed, but of which the efficiency could not be determined in this study. In the
remainder of this section we will discuss each reaction indicated in Fig. 5.5 separately.
5.4.1 Co-deposition experiments
We first focus on the formation of OH. In the original reaction scheme (black arrows in
Fig. 5.5), OH is only formed in the last reaction step. However, as mentioned before, the
fact that OH is observed for low H/O2 ratios and follows the H2 O2 behavior suggests a
common formation route. Indeed in the gas phase, the reaction of atomic hydrogen with
100
5.4 Implications for the reaction network
HO2 is known not only to lead to H2 O2 through reaction (5.2a)
H + HO2 → H2 O2 ,
(5.2a)
H + HO2 → 2OH,
(5.2b)
H + HO2 → H2 O + O,
(5.2c)
H + HO2 → H2 + O2 .
(5.2d)
but also to result in:
and
In the gas phase branching ratios of 0.90±0.04, 0.08±0.04, and 0.02±0.02 are found for
channels (5.2b)-(5.2d), respectively (Keyser 1986). Channel (5.2a) is very unlikely in the
gas phase without the presence of a third body. This channel is however allowed in the
solid phase. If all four reaction channels would proceed, OH could be formed directly
through channel (5.2b) or indirectly through channel (5.2c) after O has reacted to OH or
to O3 which can further react to OH.
In all experiments with H/O2 ≤ 2, the ratio between the produced OH and H2 O2 abundance is constant. This already suggests that OH is mainly formed directly through channel (5.2b), since OH production through subsequent hydrogenation after channel (5.2c)
would lead to an OH production as function of time differently from the H2 O2 production.
Assuming that all detected OH is indeed formed through channel (5.2b), the branching
ratios between the OH and H2 O2 formation channels in the solid phase can be obtained.
The 2OH channel (5.2b) is found to be 1.6±0.2 times more likely than the H2 O2 channel
(5.2a), provided that the OH-stretch bandstrength per molecule of H2 O2 is twice as large
as that of an OH radical. Another possibility could be that H2 O2 is not formed directly
through reaction (5.2a) but that in (38±5)% of the cases two OH molecules immediately
react and form H2 O2 (reaction (5.10)). Since OH is still abundantly observed and since
most OH is formed through reaction (5.2b) which results in two OH radicals in close
vicinity of each other, this reaction will proceed with some barrier. It is therefore indicated
by a dashed light gray arrow in Fig. 5.5; the double arrow coming from OH reflects the
two OH molecules that are needed in the reaction.
Unfortunately, we cannot quantify channel (5.2d) (H2 +O2 ) since both products are not
infrared detectable and the change in the water-induced O2 feature at 1550 cm−1 caused
by this reaction will be too small to derive a reliable branching ratio.
The branching ratio of the channel leading to H2 O and O (channel (5.2c)) is also hard
to quantify, since O atoms can only be detected indirectly by the production of ozone.
In the low H/O2 regime, the OH-stretch modes which are used to quantify the branching
ratios for the 2OH and H2 O2 channels cannot be used for H2 O, since the OH-stretch
modes for water in O2 are below the detection limit. However, using this detection limit,
the reactive rate for channel (5.2c) can be constrained to an upper limit of 0.2 times the
value of the H2 O2 channel. This upper limit is 0.08 with respect to combined rate of
channels (5.2a) and (5.2b), close to the gas phase branching ratio. The low upper limit
further justifies our assumption that OH is mostly formed through channel (5.2b), since
101
5 Water formation at low temperatures by surface O2 hydrogenation II
only a limited amount of atomic oxygen, needed for the O and O3 routes, is formed
through channel (5.2c).
The light gray arrows in Fig. 5.5 indicate the four different channels for the H +
HO2 reaction. In § 5.3.3 we have argued that reaction (5.1) is barrierless. This reaction
is therefore represented by solid arrows. Since in Chapter 4, HO2 is not observed for
T < 25 K, the reaction of H + HO2 is probably effectively barrierless as well, which
is in agreement with gas phase data where no barrier is observed between 245−300 K
(Keyser 1986). The main channel, (5.2b), is therefore also represented by solid light gray
arrows. Channels (5.2a) and (5.2d) cannot be measured directly as discussed above and
are therefore represented by dotted arrows. For channel (5.2c) only an upper limit is
determined and is therefore represented by a dash-dotted light gray arrow.
Ozone is formed through reaction (5.5) and proceeds with a barrier as discussed earlier. This reaction is therefore indicated by a dashed black arrow in Fig. 5.5. The fact that
O3 is observed, means that O atoms are involved in the reaction network. One O-atom
formation route is through reaction (5.2c). The reaction
OH + O2 → HO2 + O,
(5.6)
which has a gas phase barrier of 220 kJ/mol (Tsang & Hampson 1986), is another likely
candidate, if it could proceed through tunnelling which has little temperature dependence.
As discussed earlier, the OH features are observed to disappear mostly through a thermally
activated diffusion process and reaction (5.6) is therefore thought not to have a large effect
on the total OH abundance. In conclusion, the O atoms are probably formed through two
relatively inefficient reactions: reactions (5.2c) and (5.6). The observed OH is therefore
mainly formed by reaction (5.2b). Since reaction (5.6) is uncertain it is indicated by
dash-dotted dark gray arrows in Fig. 5.5.
5.4.2 Hydrogenation of H2 O2
Water is likely to be formed through a number of different reaction paths in the network:
by the hydrogenation of HO2 , OH or H2 O2 . The first, reaction (5.2c), is relatively inefficient as discussed in § 5.4.1. Leaving the other two as as the dominant routes. In this
subsection we discuss the specific contribution of H2 O2 hydrogenation to the overall H2 O
production. This route proceeds via reactions (5.3) and (5.4). The first has a barrier in
the gas phase of 14.97 kJ/mol (Baulch et al. 1992) and consequently a lower efficiency
is expected for H2 O formation through this reaction. The most straightforward way of
testing this reaction would be to deposit a pure H2 O2 ice and subsequently expose this to
H atoms. However, since the deposition of H2 O2 without simultaneous H2 O deposition is
not experimentally feasible in our set-up, pure H2 O2 ice is produced in a different way. At
the end of a co-deposition experiment with an H/O2 ratio of 10, the ice is dominated by
H2 O2 and O2 (see Fig. 5.2). By heating the ice to 40 K, all the O2 desorbs from the top, reactive layers, and the resulting bulk H2 O2 ice can be used for a hydrogenation experiment
in which the last part of the reaction scheme (reactions (5.3) and (5.4)) can be studied.
In this specific case the ice is exposed to H atoms after it is formed, as in Chapter 4 (in
102
5.4 Implications for the reaction network
Coverage (ML)
6
H2O
H2O2
4
2
0
0
17
1×10
17
2×10
0
17
1×10
17
0
2×10
17
1×10
17
2×10
-2
Fluence (H atoms cm )
Figure 5.6 The H2 O and H2 O2 surface coverage in monolayers at 20 K for a H2 O2 hydrogenation experiment (filled symbols) and O2 hydrogenation experiment (open symbols).
The right panel compares the H2 O2 and the O2 hydrogenation experiment after a fluence
of 7 × 1016 atoms cm−2 (dotted line in middle panel). The solid curve in the middle panel
indicates the calculated contribution of water formation from reaction (5.3) (Eq. 5.13).
contrast to the experiments discussed in the rest of the present chapter where H atoms
and O2 molecules are co-deposited). The left panel of Fig. 5.6 plots the H2 O and H2 O2
surface coverage with respect to the initial H2 O2 ice for a temperature of 20 K. Hydrogen
peroxide is used up whereas H2 O is formed. To obtain the absolute quantities from the
integrated absorbances, the apparent bandstrength for water (0.02 cm−1 ML−1 ) as determined in Chapter 4 is used. The corresponding value for H2 O2 is obtained from this experiment by assuming mass balance. A H2 O(1580−1800 cm−1 )/H2 O2 (1200−1580 cm−1 )
ratio of 0.31 is obtained in accordance with Loeffler et al. (2006).
This experiment can be directly compared to the hydrogenation experiments of solid
O2 as reported in Chapter 4. Both the deposition technique (sequential deposition of
the ice and H atoms instead of simultaneous) and the experimental conditions in terms
of surface temperature and H-atom flux are the same. The middle panel plots the H2 O
and H2 O2 surface coverage as a function of time for an O2 hydrogenation experiment,
again at 20 K. The efficiency of the destruction reaction (5.3) can be determined by
comparison of the combined formation reactions (5.1) and (5.2a). In the middle panel
(5.3±0.7)×10−17 monolayers of H2 O2 are formed per deposited H atom per cm2 (slope
of the first part of the curves with triangles). It takes two H atoms to form one H2 O2
molecule. In the left panel (2.8±0.7)×10−18 monolayers of H2 O2 are destroyed per deposited H atom per cm2 (slope at zero fluence of the curves with triangles). It takes also
two H atoms to destroy one H2 O2 molecule. The H2 O2 destruction reaction (reaction
(5.3)) is the rate limiting step in the formation of water — reaction (5.4) is more efficient.
The rate of reaction (5.3) can therefore be quantified with respect to the rate of reaction
(5.2a), k5.2a , which is (2.8±0.7)×10−18 /(5.3±0.7)×10−17 = (0.05±0.01)k5.2a . This lower
efficiency with respect to k5.2a indicates that there is a barrier for reaction (5.3) and it is
therefore indicated by a dashed black arrow in Fig. 5.5.
The water formed in the O2 hydrogenation reactions (middle panel) can be formed
103
5 Water formation at low temperatures by surface O2 hydrogenation II
through several reaction routes. The most important two are reactions (5.3) and (5.4).
The solid line in the middle panel of Fig. 5.6 shows the contribution of reaction (5.3).
This line is obtained from
Nreaction (5.3) (H2 O) = 2 · 2.8 × 10−18 F
N(H2 O2 )
Nmax (H2 O2 )
(5.13)
with F the hydrogen fluence in atoms cm−2 . The factor of 2 accounts for the stoichiometric ratio in O atoms between H2 O and H2 O2 , the rate of 2.8×10−18 ML cm−2 is taken from
the H2 O2 hydrogenation experiment and the last term in this expression accounts for the
probability of an H atom to meet H2 O2 where the maximum amount of formed H2 O2 corresponds to the starting condition of the H2 O2 hydrogenation experiment. Reaction (5.3)
accounts for (30±5)% of the formed H2 O in the beginning of the O2 hydrogenation experiment by comparing the slope of the solid line and the slope of the experimental water
abundance (open diamonds) at the beginning of the experiment. After 7×1016 atoms cm−2 ,
when the maximum amount of H2 O2 is reached (vertical dotted line), the route accounts
for roughly 70% of the formed water as is shown in both the middle and the right panel
of Fig. 5.6. In the right panel the results of the H2 O2 hydrogenation experiment are overplotted by the O2 hydrogenation results after 7 × 1016 atoms cm−2 . The H2 O2 production
has reached its maximum at that fluence and the resulting ice is probably similar to the
initial condition of the H2 O2 hydrogenation experiment. The hydrogen peroxide use-up in
the O2 hydrogenation experiment is roughly half of the case where H2 O2 is hydrogenated
(comparison open and closed triangles in right panel) while an equal amount of water is
formed (comparison open and closed diamonds in right panel). Part of this is within the
error.
Let us now return to the beginning of the O2 hydrogenation experiment when (30±5)%
of the formed H2 O is formed via reaction (5.3). The remaining (70±5)% is most likely
formed through reaction (5.4) as discussed earlier. From the co-deposition experiments
we know that the reaction of H and HO2 leads to 3.2 times more OH than H2 O2 , however,
only a small amount of water is formed from all these hydroxyl radicals. If OH does
not react to H2 O2 , 3.2 × 5.8 = 19 ML OH should have formed during the first part of
the O2 hydrogenation experiment. Only 0.70 × 1.2 = 0.8 ML of water has been formed
from these OH radicals, which amounts to (4±1)%. Since reaction (5.4) is barrierless,
this low efficiency is rather surprising. A reason for this could be that the majority of the
OH radicals reacts together in the bulk of the ice and would be responsible for part of the
H2 O2 contribution. However, as discussed earlier, we would not expect this to happen
in large quantities based on the co-deposition experiments. Another possibility could be
that H atoms are not able to reach all OH radicals in the ice. An argument against this is
that O2 still reacts, which means that the ice is not impenetrable. However, it could be
that only H-atom approaches under specific incoming angles to OH are reactive, whereas
the number of reactive configurations for H reacting with O2 is much larger. To reflect
the relatively low efficiency of this reaction, this reaction is indicated with a dashed black
arrow in Fig. 5.5.
104
5.4 Implications for the reaction network
H2O2 in O2
0.15
OH in O2
HO2
0.05
0.00
0.15
high H2/H
Absorbance
low H2/H
H2O2 bulk
0.10
0.10
0.05
0.00
0
50
100
150
Time (min.)
Figure 5.7 RAIRS integrated intensities for H/O2 = 1 and at 20 K with two different H/H2
ratios. The standard, low, H2 /H ratio results are plotted in the upper panel; the high H2 /H
ratio results in the lower panel.
5.4.3 The role of H2
All reactions discussed in the previous sections ignore the presence of H2 in the atom
beam. However, H2 will also be present on the surface, mostly from direct deposition of
cold molecules from the atom beam, since formed H2 on the surface is likely to desorb
upon formation. If H2 and O2 are co-deposited, no reactions are observed, only background deposition of H2 O. However, in the presence of H atoms new reactive species are
formed that can react with H2 , in particular OH to form H2 O (reaction (5.12)) or HO2
H2 + HO2 → H2 O2 + H.
(5.14)
By changing the temperature of the filament in the H-atom source and the H2 pressure
in the atomic line, we can keep the H-atom flux constant while increasing the H2 flux.
Figure 5.7 plots the resulting integrated intensities for such an experiment in the lower
part. Here the H2 flux is roughly ten times higher than in the regular experiments (H/H2
= 90 vs. H/H2 = 9 in the standard experiments). The difference in H2 abundance is
therefore definitely due to a change in the cold molecule abundance, since the contribution
of formed H2 molecules remains the same. If reaction (5.12) were efficient, the OH
105
5 Water formation at low temperatures by surface O2 hydrogenation II
radicals that are formed would react further to H2 O in the high H2 /H experiment and
we would be able to observe a significant decrease in the OH surface abundance in the
bottom panel with respect to the top panel. At the same time we would expect to be able to
detect H2 O in the high H2 /H spectra. Figure 5.7 does however not show such a decrease
in OH abundance and also H2 O monomer features were not detected in the IR spectra.
We therefore conclude that reaction (5.12) is not very efficient (dash-dotted black arrow
in Fig. 5.5). Gas phase experiments show a barrier of 12.69 kJ/mol in the low temperature
limit (Orkin et al. 2006).
In the surface abundances of HO2 and H2 O2 on the other hand a change can be observed. The abundance of HO2 decreases in the high H2 /H regime whereas H2 O2 increases. This is in accordance with reaction (5.14) and this reaction is therefore indicated
by a dashed dark gray arrow in Fig. 5.5. The fact that reaction (5.14) proceeds at such
low temperatures, is rather surprising since a high gas phase barrier of 109 kJ/mol was
reported for this reaction (Tsang & Hampson 1986). An explanation for this is not available.
5.5 Conclusions
The present study shows that the water formation reaction network as originally proposed
by Tielens & Hagen (1982) is not complete but that several new reaction paths should be
added. The solid state hydrogenation of O2 exhibits a complex network of reactions as
schematically presented in Fig. 5.5. The original reactions are indicated in black. The dark
gray and light gray reactions are added in the present study. Through this effort we have
shown that the O2 hydrogenation channel is connected to the O and O3 hydrogenation
channels and we have therefore been able to also draw conclusions on some reactions
which are part of the other two hydrogenation channels. We could furthermore quantify
the reaction rates of several reactions.
The solid arrows in Fig. 5.5 indicate the reactions that are effectively barrierless at low
temperatures. These consist of two reactions
H + O2 → HO2 ,
(5.1)
and the reaction of HO2 and H atoms, indicated in dark gray. The latter has probably four
different product channels:


H2 O2
k5.2a ,





(1.6 ± 0.2)k5.2a ,
 2OH
(5.2)
H + HO2 → 


H2 O + O
(< 0.2)k5.2a ,



 H2 + O2
where the last product channel cannot be quantified by the methods used in this experimental study.
The reactions which are indicated by the dashed lines proceed with a barrier but have
a detectable efficiency. These reactions include:
H + H2 O2 → H2 O + OH
106
(0.05 ± 0.01)k5.2a ,
(5.3)
5.5 Conclusions
H + OH → H2 O,
(5.4)
H2 + HO2 → H2 O2 + H,
(5.14)
2OH → H2 O2 .
(5.10)
and
The dash-dotted arrows indicate reactions that have been proposed but which are not
observed to proceed in this study, either because the reaction is too slow or because the
experimental method did not allow us to detect this reaction. The reactions
H2 + OH → H2 O + H,
H + HO2 → H2 O + O
(< 0.2)k5.2a ,
(5.12)
(5.2c)
and
OH + O2 → HO2 + O
(5.6)
were found to proceed with efficiencies below our detection limit. The reaction
H + HO2 → H2 + O2
(5.2d)
cannot be detected by the methods used in this experimental study. The same is true for
the direct channel of
H + HO2 → H2 O2
(5.2a)
which could proceed with 2OH as intermediates.
The dotted arrows indicate reactions that were found to proceed, but of which the
efficiency could not be determined in this study. The reaction
H + O → OH
(5.7)
is most likely barrierless. From the the present study we can conclude that it is more
efficient than the formation of ozone from oxygen atoms. Since the amount of formed
ozone cannot be quantified, the efficiency of the reaction
H + O3 → O2 + OH
(5.8)
could not be determined from this study. Studies of the hydrogenation of ozone indicate
this reaction to be efficient.
This studies shows that in the O2 hydrogenation experiments performed in Chapter 3
and 4, and Miyauchi et al. (2008) water is formed through different reaction paths. Especially in the early stage of experiment, H2 O is not predominantly formed through the
hydrogenation of H2 O2 but through the reaction with OH. By not considering the latter
route in the model to fit to the experimental data, an artificial, isotope-independent reaction rate has been obtained for the H + H2 O2 reaction as explained in the introduction.
This newly determined reaction scheme will have profound implications for models
that model the formation of water under interstellar conditions. Clearly several new reaction paths should be considered through this study. Moreover, several reactions proved to
be much more efficient (H + O2 ) or less efficient (O + OH and H2 + OH) than originally
thought. A dedicated study in which this new scheme will be the input of a new model
needs to be applied to tell us how this will affect the formation of interstellar water under
different interstellar conditions exactly.
107
CHAPTER 6
Water formation by surface O3 hydrogenation1
Three solid state formation routes have been proposed in the past to explain the observed
abundance of water in space: the hydrogenation reaction channels of atomic oxygen (O
+ H), molecular oxygen (O2 + H) and ozone (O3 + H). New data are presented here for
the third scheme with a focus on the reactions O3 + H, OH + H, and OH + H2 , which
were difficult to quantify in previous studies. A comprehensive set of H/D-atom addition
experiments is presented for astronomically relevant temperatures. Starting from the hydrogenation/deuteration of solid O3 ice, we find experimental evidence for H2 O/D2 O (and
H2 O2 /D2 O2 ) ice formation using Reflection Absorption InfraRed Spectroscopy (RAIRS).
The temperature and H/D-atom flux dependence are studied and this provides information on the mobility of ozone within the ice and possible isotope effects in the reaction
scheme. The experiments show that the O3 + H channel takes place through stages that
interact with the O and O2 hydrogenation reaction schemes. It is also found that the reaction OH + H2 (OH + H), as an intermediate step, plays a prominent (less efficient) role.
The main conclusion is that solid O3 hydrogenation offers a potential reaction channel for
the formation of water in space.
1 Based on: C. Romanzin, S. Ioppolo, H. M. Cuppen, E. F. van Dishoeck, H. Linnartz, 2010, submitted to
Journal of Chemical Physics
109
6 Water formation by surface O3 hydrogenation
6.1 Introduction
Water is ubiquitous throughout the Universe and belongs to the more abundant species
in the interstellar medium. Since gas phase formation rates are not efficient at low temperatures, the formation of H2 O ice in cold dense quiescent interstellar clouds (∼10 K) is
expected to take place in the solid state on the surface of dust grains through H-atom addition reactions. Three different hydrogenation channels have been proposed in the past:
O + H, O2 + H and O3 + H (Tielens & Hagen 1982). Several laboratory studies investigated the formation of solid H2 O through the hydrogenation of atomic oxygen (Hiraoka
et al. 1998, Dulieu et al. 2010) and molecular oxygen (Miyauchi et al. 2008, Ioppolo
et al. 2008, Matar et al. 2008, Oba et al. 2009, Ioppolo et al. 2010, Cuppen et al. 2010).
However, only a single study (Mokrane et al. 2009) investigated the third channel so far,
showing that the deuteration of O3 ice on an amorphous H2 O substrate leads to the formation of D2 O by detecting HDO molecules during desorption of the ice using Quadrupole
Mass Spectrometry (QMS). We give here further experimental evidence for H2 O/D2 O ice
formation, presenting for the first time a comprehensive set of H/D-atom addition experiments on solid O3 for astronomically relevant temperatures, using Reflection Absorption
InfraRed Spectroscopy (RAIRS).
Solid O3 can be formed in space through energetic processing (ions, photons, electrons) of O-bearing ices at astronomically relevant temperatures (e.g., Famá et al. 2002,
Loeffler et al. 2006, Cooper et al. 2008, Schriver-Mazzuoli et al. 1995, Gerakines et al.
1996, Lacombe et al. 1997, Bennett & Kaiser 2005, Sivaraman et al. 2007). Tielens & Hagen (1982) proposed the formation of O3 ice through the subsequent oxidation of atomic
oxygen on the surface of the interstellar grains at low temperature and in absence of UV
irradiation. Ozone ice has been observed on the surface of small bodies in the Solar System, like Ganymede, Rhea and Dione (Noll et al. 1996, 1997, Hendrix et al. 1999), but
it has not been observed in the interstellar medium. The non-detection of solid ozone in
dense molecular clouds is consistent with an efficient use-up through hydrogenation, in
the case that O3 + H is an efficient process under interstellar conditions.
Figure 6.1 is taken from Chapter 5 and shows how the three hydrogenation channels
(O/O2 /O3 + H) can interact. Specifically, the hydrogenation of solid O3 comprises the
following solid-state reactions
O3 + H → OH + O2
(6.1)
OH + H → H2 O,
(6.2)
OH + H2 → H2 O + H.
(6.3)
and
Cuppen & Herbst (2007) and Cazaux et al. (2010) showed in their astrochemical models
that the efficiency of this reaction channel strongly depends on the astronomical environment (e.g., diffuse clouds, dense clouds and photon-dominated regions). The experimental
results presented in Chapters 4 and 5 showed that the O + H and the O3 + H channels are
connected via the O2 + H route through common reactive intermediates (see Fig. 6.1).
110
6.1 Introduction
O
O
OH
H H
H
H
O
O2
HO 2
O3
H
H2
H
H2 H
H
H2 O 2
H
OH
H
H
H2
H 2O
Figure 6.1 A schematic representation of the reaction network as obtained from Chapters
4, 5 and 6. The dashed arrows represent the surface reactions investigated here.
The latter channel involves the reactions
HO2 + H → H2 O + O
(6.4)
O2 + OH → HO2 + O,
(6.5)
and
which both lead to the formation of O atoms. These can then react with O2 to form O3
O2 + O → O3 .
(6.6)
Indeed, O3 has been found as a reaction product in hydrogenation experiments of pure O2
ice (Chapters 4 and 5).
In the following sections we investigate the O3 + H scheme under interstellar analog
conditions. We focus in particular on the first reaction step O3 + H as well as the formation
of H2 O from OH through reactions (6.2) and (6.3). For this purpose, most experiments
are carried out at elevated temperatures in order to instantaneously desorb the O2 formed
through reaction (6.1).
111
6 Water formation by surface O3 hydrogenation
6.2 Experimental procedure
Experimental details
The experiments are performed using an ultra high vacuum set-up, which has been described in detail elsewhere (Chapters 2 and 4). It consists of an atomic beam line and
a main chamber (∼10−10 mbar), in which ices are grown on a (12−300 K) cryogenically cooled gold-coated copper substrate by depositing gas under an angle of 45◦ . A
fresh O3 sample is prepared before each experiment in a high-vacuum glass line, following the procedure as described in Berkley et al. (1988). The O3 sample is prepared in
a commercial ozone generator (Fischer-model 502, O2 99.995% of purity, Praxair) and
collected in a liquid nitrogen trap, which is used to purify the sample from O2 pollution.
O2 deposition originating from the dissociation of O3 in the main chamber is kept to a
minimum by maintaining the substrate temperature at 40 K, well above the O2 desorption
temperature (T des (O2 ) ∼30 K, Acharyya et al. 2007). The ice is monitored by means of
RAIRS, using a Fourier Transform InfraRed spectrometer (FTIR). The FTIR covers the
range between 4000 and 700 cm−1 (2.5−14 µm) with a spectral resolution of 1 cm−1 . A
co-addition of 256 scans yields one spectrum. RAIR difference spectra (∆A) with respect
to the deposited O3 ice spectrum are acquired every few minutes during the hydrogenation experiment. According to Sivaraman et al. (2007), shape and position of the ν3 (O3 )
stretching mode is sensitive to the ozone environment. Therefore, the presence of other
molecules should affect this infrared band, but the observed ν3 (O3 ) band in our spectra
after deposition is typical for a rather pure O3 ice (Sivaraman et al. 2007, Chaabouni et al.
2000, Misochko et al. 1999), instead of O3 molecules mixed with O2 (Schriver-Mazzuoli
et al. 1995).
After deposition the ice is subsequently hydrogenated/deuterated at different temperatures (25, 40 and 50 K). H/D atoms are supplied by a well-characterized thermal cracking
source (Tschersich & von Bonin 1998, Tschersich 2000, Tschersich et al. 2008). H2 /D2
molecules are cracked in a capillary pipe surrounded by a tungsten filament, which is
heated to 2200 K. During the H/D-atom exposure, the pressure in the atomic line is kept
constant. Hot H/D atoms are cooled to room temperature via collisions by a nose-shaped
quartz pipe, placed in the H/D-atom beam path towards the substrate. The geometry of the
pipe is designed in such a way that hot species (H/D; H2 /D2 ) cannot reach the ice directly
(more details in Chapters 2 and 4). The H/D-atom fluxes used in our experiments are set
by changing the H2 /D2 pressure in the capillary pipe while the filament temperature is
kept constant. The final flux values (2 × 1013 and 8 × 1013 atoms cm−2 s−1 for H atoms
and 1 × 1013 and 4 × 1013 atoms cm−2 s−1 for D atoms) are measured at the substrate
position in the main chamber using a quadrupole mass spectrometer for the D-atom flux.
The method is described in the Appendix A of Chapter 4. The relative error in the D-atom
flux determination is within 10%, while the relative H-atom flux determination is within
50%. The absolute error for both is estimated to be within 50%.
Several control experiments have been carried out. Deuteration experiments have
been performed to estimate the maximum H2 O contamination, i.e., H2 O contributions
other than those induced in the ice upon H-atom impact. This is essential as H2 O is
112
6.2 Experimental procedure
the prime target of this study. The pollution may originate from H2 O background in
the UHV setup and/or from H2 O in the HV gas line. The contamination is found to
increase with time and to be less ∼1 ML at the end of all experiments. Results presented
in § 6.3 are corrected for this contamination. In the deuteration experiments, naturally,
this contamination does not play a role. Also, a pure O3 ice has been exposed to a D2
beam (at 40 K) to ensure that the D2 molecules do not chemically react with the O3 or
physically change the surface through sputtering. Finally, an unprocessed O3 ice grown
at 40 K and subsequently heated to 50 K with a rate of 1 K min−1 , shows no substantial
O3 loss because of thermal desorption (T des (O3 ) ∼63 K, Famá et al. 2002) over a three
hour period, the length of a typical experiment.
Data analysis
After subtracting the infrared spectra with a piece-wise straight baseline, the column densities (molecules cm−2 ) of the newly
formed species are calculated using the modified
R
Lambert-Beer equation: NX = A(ν)dν/S X , where A(ν) is the wavelength dependent
absorbance. Since literature values of transmission band strengths cannot be used directly in reflection measurements, an apparent absorption band strength, S X of species X,
is determined by individual calibration experiments. This procedure has been described
in detail elsewhere (for the H2 O/D2 O and H2 O2 /D2 O2 band strength determination see
Chapters 4 and 5). Briefly, a layer of the selected ice is deposited at a temperature lower
than its desorption temperature. The sample is then linearly heated, close to its desorption temperature. Infrared spectra are acquired regularly until the desorption of the ice is
complete. Such an isothermal desorption experiment has been performed to determine the
apparent absorption band strength of O3 by recording the transition from zeroth-order to
first-order desorption. This is assumed to occur at the onset of the submonolayer regime
and appears in the desorption curve as a sudden change in slope. The apparent absorption
strength in cm−1 ML−1 is then calculated by relating the observed integrated area to 1 ML
in the modified Lambert-Beer equation. We estimate the uncertainty of the band strength
to be within 50%. The noise in the infrared spectra introduces an extra uncertainty in the
H2 O/D2 O, H2 O2 /D2 O2 and O3 column densities, which is found to be within ±0.5 ML
for all the considered species.
The assignment of the spectral features observed in our experiments is listed in Table 6.1. The band modes peaking at 1650/1210 cm−1 (ν2 ) and 1390/1050 cm−1 (ν2 , 2ν4
and ν6 ) are chosen to quantify the column densities of the newly formed species upon
H/D-atom exposure (solid H2 O/D2 O and H2 O2 /D2 O2 , respectively). The O3 band peaking at 1050 cm−1 (ν3 ) is used to quantify the amount of O3 deposited on the cold substrate,
and, subsequently, the O3 consumed in the surface reactions during H/D-atom addition.
The 1050 cm−1 D2 O2 band overlaps with the ν3 (O3 ) band in our infrared spectra. Thus,
a multi-Gaussian fit is used to separate the contributions and determine the area of the
individual bands.
113
6 Water formation by surface O3 hydrogenation
Table 6.1 Assigned infrared features in the 4000−700 cm−1 region.
Mode
libration
ν3
ν2 , 2ν4 , ν6
ν2
2ν6
ν1 , ν5
ν3
ν3
ν1
ν1 + ν3
Positiona
(cm−1 )
830
888
1390(∗)
1650(∗)
2840
3290
3260
1050
1107
2110
Species
H2 O
H2 O2
H2 O2
H2 O
H2 O2
H2 O2
H2 O
O3
O3
O3
Positiona
(cm−1 )
Species
884
1050(∗)
1210(∗)
2100
2465
2440
D2 O2
D2 O2
D2 O
D2 O2
D2 O2
D2 O
Reference
1, 2
1, 3
1, 3
1, 2
1, 3
1, 3
1, 2
4-6
4-6
4-6
a
Asterisks mark the features used to determine the integrated absorbance.
(1) Giguère & Harvey (1959); (2) Hornig et al. (1958); (3) Lannon et al. (1971); (4) Bennett & Kaiser (2005);
(5) Chaabouni et al. (2000); (6) Brosset et al. (1993).
6.3 Results and discussion
Figure 6.2 shows the RAIR difference spectra acquired during an hydrogenation (left
panel) and a deuteration (right panel) experiment of solid O3 at 25 K. Both H2 O/D2 O
and H2 O2 /D2 O2 integrated band intensities clearly grow as the H/D-fluence (H/D-flux ×
time) increases. Neither species like OH, HO2 , and HO3 , nor the partially deuterated
species, like HDO and HDO2 , are detected in our infrared spectra during H/D-atom addition to the O3 ice. The presence of fully deuterated species gives experimental evidence
for surface formation of water ice in the solid phase with O3 ice as a precursor. The
negative peak shown in Fig. 6.2 indeed reflects the O3 use-up.
6.3.1 Temperature dependence
Figure 6.3 shows the H2 O/D2 O (square) and H2 O2 /D2 O2 (triangle) column densities for
the three investigated temperatures (25, 40, and 50 K) as a function of the H/D-atom fluence. The solid/open symbols correspond to the low/high H/D-atom flux used in our experiments. The amount of O3 use-up (circle) during H/D-atom addition changes with the
substrate temperature from ∼1 ML at 25 K to ∼10 ML at 50 K. This is consistent with the
increase of the H/D-atom penetration depth in the O3 ice at higher temperatures, since the
mobility of O3 molecules in the ice is expected to improve with increasing temperature,
even though the penetration depth of H atoms involves only the surface of the ice and not
the bulk. A similar temperature dependence has been observed for the penetration depth
of H atoms in CO ice (Chapter 2). Another mechanism may also affect the final amount
114
6.3 Results and discussion
H-atom addition
H2O
0.008
D-atom addition
D2O
O3
D2O2
H2O2
∆Absorbance + offset
O3
0.006
H2O
(c)
0.004
(b)
0.002
(a)
0
3500
2800
1800
1200 2800
1800
1200
-1
Wavenumber (cm )
Figure 6.2 Difference infrared spectra of solid O3 ice, with respect to the spectrum before
H/D-atom addition, upon hydrogenation/deuteration at 25 K for three different H/D-atom
exposures: (a) 2.4 × 1016 , (b) 7.2 × 1016 , (c) 2.0 × 1017 H/D atoms cm−2 s−1 (left/right
panel, respectively). Spectra are offset for clarity. The water pollution is visible in the
deuteration experiment (right panel).
of O3 use-up: the erosion of the ice. Each time an H/D atom reacts with an O3 molecule
through reaction (6.1) an O2 molecule is formed. Whether the O2 molecule remains on
the surface of the ice or desorbs, depends on the temperature of the ice. Below 30 K, the
O2 molecules will be further hydrogenated/deuterated according to the scheme shown in
Fig. 6.1 (see also Chapters 4 and 5). At higher temperatures (above 30 K, see Acharyya
et al. 2007) the desorption of the O2 formed through reaction (6.1) will leave the deeper
O3 layers exposed for H/D-atom addition, increasing the final O3 use-up. The H2 O/D2 O
and H2 O2 /D2 O2 column density ratios are also affected by this desorption behavior. Below 30 K, H2 O/D2 O will be formed through both the hydrogenation/deuteration of O2 ice
and reactions (6.2) and (6.3). A significant amount of H2 O2 /D2 O2 will be formed through
the O2 channel as well (Chapters 4 and 5). For increasing temperature, the O2 channel
becomes less important and as a consequence the amount of H2 O2 /D2 O2 decreases, while
H2 O/D2 O formation through reactions (6.2) and (6.3) becomes the dominant process.
As a side-effect of the erosion/restructuring of the ice, the H2 O pollution diluted in
the O3 ice, may rearrange in islands. Consequently, the narrow H2 O bands seen after
deposition of the O3 ice in the region of the H2 O bending mode (1650 cm−1 ) will broaden
115
6 Water formation by surface O3 hydrogenation
Table 6.2 Amounts of O3 use-up, and formed H2 O/D2 O and H2 O2 /D2 O2 in ML after an
exposure of 1.1 × 1017 H/D atoms cm−2 and 4.2 × 1017 H/D atoms cm−2 (low and high
fluxes, respectively) at the three different substrate temperatures investigated. See §6.2 for
the determination of the errors. The Obudget corresponds to the mass-balance of O atoms
in ML: Obudget = -3O3 + H2 O + 2H2 O2 , or the equivalent for deuteration.
H/D-flux
(cm−2 s−1 )
2/1 ×1013
8/4 ×1013
T
(K)
25
40
50
25
40
50
O3
H2 O
0.8
2.7
7.4
0.8
2.9
8.1
1.5
1.8
1.2
3.1
2.9
3.7
H2 O2
(ML)
4.9
2.3
0.7
4.6
2.6
1.8
Obudget
O3
D2 O
8.9
-1.7
-19.6
9.9
-0.6
-17.0
1.5
3.0
8.8
1.5
2.7
8.5
1.0
1.1
1.8
1.6
1.6
2.2
D2 O2
(ML)
2.7
0.6
0.5
2.2
1.4
0.5
Obudget
1.9
-6.7
-23.6
1.5
-3.7
-22.3
upon ice restructuring. This effect increases with time and contributes to the total H2 O
bulk feature peaking at 1650 cm−1 . This effect is shown in the right panel of Fig. 6.2. The
contribution of this effect, which is estimated to be ∼1 ML at the end of all the deuteration
experiments, is taken into account for all the H-atom addition experiments, as mentioned
in §6.2.
6.3.2 H/D-atom flux dependence
Figure 6.3 also indicates the influence of the H/D-atom flux on the amount of reaction
products. The H2 O/D2 O and H2 O2 /D2 O2 column densities follow the same trend for
high and low H/D-atom flux and for all investigated temperatures within the experimental
uncertainties. This observation is in agreement with a scenario in which a reactive system
is limited only by the number of H/D atoms that reaches the ice surface. The O3 column
density follows the same behavior for high and low H/D-atom flux at temperatures below
40 K and at 50 K for a maximum H/D-atom fluence of 1×1016 H/D atoms cm−2 . However,
at higher H/D-atom fluence the O3 column density profile differs for high and low H/Datom flux at 50 K. This is most likely caused by the transition between two different
regimes: in the first regime, reaction (6.1) is limited by the number of H/D atoms, in the
second regime, this reaction is limited by the supply of O3 molecules, since the formed
H2 O/D2 O and H2 O2 /D2 O2 prevent the incoming H/D atoms to reach the O3 molecules in
the lower layers. Further conversion into H2 O/D2 O and H2 O2 /D2 O2 is then only possible
after replenishing of the top layers by fresh O3 . This process is governed by the diffusion
of O3 in the ice, which increases with temperature and is independent of H/D-atom flux.
Thus, this effect is stronger at 50 K than at 40 K and indeed the O3 use-up follows the
same trend for low and high H/D-atom fluxes when plotted as a function of exposure
time instead of fluence for >1×1016 H/D atoms cm−2 . The two regimes are schematically
depicted in Fig. 6.4.
116
6.3 Results and discussion
H-atom addition
D-atom addition
T = 25 K
T = 25 K
O3
H2O
H2O2
5
0
Surface coverage (ML)
-5
5
0
-5
T = 40 K
T = 40 K
5
5
0
0
-5
-5
5
T = 50 K 5
T = 50 K
0
0
-5
-5
-10
-10
0
2
4
6
8
0
17
2
4
6
8
-2
Fluence (10 atoms cm )
Figure 6.3 Column densities for H2 O/D2 O (square), H2 O2 /D2 O2 (triangle) and O3 (circle) for the three temperatures investigated (25, 40, and 50 K) as a function of the H/Datom fluence. The hydrogenated species are plotted in the left panel, and the deuterated species in the right panel. The solid and open symbols correspond to the lower and
higher H/D-atom flux (2 × 1013 and 8 × 1013 atoms cm−2 s−1 for H atoms, 1 × 1013 and
4 × 1013 atoms cm−2 s−1 for D atoms), respectively.
117
6 Water formation by surface O3 hydrogenation
Figure 6.4 Schematic representation of the hydrogenation of O3 ice as a function of the
temperature and H-atom fluence: at temperatures below 40 K (bottom) reaction (6.1) is
limited by the number of H/D atoms, at higher temperatures (top) this reaction is limited
by the supply of O3 molecules. The replenishing of the top layers is induced by diffusion
of O3 in the ice. The erosion of the ice at 50 K is also shown (top).
6.3.3 Possible reaction pathways
The investigation of the mass balance between the formed and consumed species in our
ice after H/D-atom addition allows identifying the most likely reaction channel responsible for the formation of solid H2 O ice. The mass balance for oxygen atoms can be
determined looking at the number of O atoms present in each species (Obudget = -3O3 +
H2 O + 2H2 O2 ). From the comparison of the results listed in Table 6.2 we summarize
three relevant results: (i) the O atoms are found in excess only at 25 K (Obudget = 9.9/8.9
ML for higher/lower H-atom flux and 1.5/1.9 ML for both higher/lower D-atom flux);
(ii) part of the O use-up is not converted into H2 O/D2 O and H2 O2 /D2 O2 at 40 and 50 K
(negative Obudget ); and (iii) there appears to be a strong isotope effect in the formation of
H2 O/D2 O and H2 O2 /D2 O2 (more H2 O and H2 O2 than D2 O and D2 O2 ).
Point (i) can be explained by the presence of an extra O2 poisoning layer deposited on
top of the O3 ice at 25 K. The extra O2 originates from background deposition, while the
substrate was cooled from 40 K to 25 K with a rate of 1 K min−1 . This effect is already
minimized by lowering the surface temperature only after the main chamber pressure has
substantially dropped towards the standard value of 10−10 mbar. However, the deposition
of a maximum of 5 ML of O2 on top of the O3 ice cannot be prevented for the 25 K
experiments (5 ML of O2 correspond to 10 ML of O atoms). The higher value for the
118
6.3 Results and discussion
Obudget in the 25 K hydrogenation experiment with respect to the deuteration experiment
is consistent with a higher penetration depth of H atoms in the O2 ice compared to D
atoms (Chapter 3).
Point (ii) is addressed by the fact that most of the O2 produced through reaction (6.1) is
lost at temperatures higher than the O2 ice desorption temperature. OH/OD and H2 O/D2 O
can desorb upon reaction as well. We will discuss this issue in more detail in the next
paragraph, which deals with point (iii).
Roughly the same amount of O3 is used-up for the hydrogenation and deuteration
experiments. This indicates that the observed isotope effect (point (iii)) is not due to a
different rate for hydrogenation and deuteration of O3 , but that it is probably caused by a
different desorption probability upon reaction. Table 6.2 suggests that D2 O and D2 O2 are
more likely to desorb than H2 O and H2 O2 . Thermal desorption, however, would lead to
the reverse and therefore this effect has to come from the reaction energetics. We will first
consider H2 O/D2 O, which is formed in two steps. In the first step, reaction (6.1), most of
the excess energy will be released in the form of ro-vibrational excitation of OH/OD or
in translational energy. Gas phase calculations show that this translational energy is 5.4%
higher for deuteration than for hydrogenation (Yu & Varandas 1997), which would lead
to a slightly higher desorption probability for D + O3 than for H + O3 and may explain
at least part of the observed effect. If H2 O/D2 O is then mainly formed from OH/OD
through reaction (6.2) (see left side of Fig. 6.5), the large overall difference in desorption
probability still cannot be fully explained. It can however be explained if H2 O/D2 O is
mainly formed through reaction (6.3). In this case, the kinetic energy is distributed over
the products according to the inverse mass which means that D2 O will have nearly twice
the kinetic energy of H2 O after reaction (6.3). Since the total excess energy of ∼1 eV is
close to the desorption energy of H2 O (0.9 eV, Andersson et al. 2006), this difference in
kinetic energy will have a substantial effect on the desorption probability. Therefore, in
this case more D2 O will desorb from the ice.
The observed isotope effect for H2 O/D2 O can thus be explained by reaction (6.3)
instead of reaction (6.2). On first glance one would however expect reaction (6.2) to be
more efficient than reaction (6.3), since the first is barrierless, with an excess of 5.3 eV,
and the second has a barrier of 0.234 eV (Yang et al. 2001), with an excess of 1 eV. The
problem with reaction (6.2) is that one needs to dissipate 5.3 eV of excess energy with just
one product. Part of this could be absorbed by the ice surface, but the weak interactions
between the product and the ice limits the full dissipation. A reaction where only 1 eV
of excess energy is released over two products is therefore more likely, especially since
H2 is abundantly present in our experiment, because the H-atom beam entering the main
chamber contains a large fraction of cold H2 .
Furthermore, gas phase experiments indicate that tunneling becomes important for
OH + H2 below 250 K. The reaction rate at 25−50 K will therefore be substantially
increased through tunneling. This also leads to an extra isotope effect where OH + H2
has probably a higher rate than D2 + OD. In addition, OH/OD is formed “hot” and this
energy can also be used to overcome the reaction barrier. Reaction (6.3) may therefore
be more relevant than reaction (6.2). These two reactions were previously included in
the complete reaction network for O2 surface hydrogenation investigated in Chapter 5,
119
6 Water formation by surface O3 hydrogenation
although no experimental evidence was found for reaction (6.3). However, the method
used there was not very sensitive to the detection of this particular reaction. Therefore,
extra dedicated studies specifically on the OH + H2 reaction are needed to determine
the absolute efficiency of this reaction, especially in light of the present study, which
indicates that this reaction may be crucial as a final step in all three water formation
channels (O/O2 /O3 + H).
Similar arguments can be used for the formation of H2 O2 /D2 O2 from HO2 /DO2 by
HO2 + H → H2 O2
(6.7)
HO2 + H2 → H2 O2 + H,
(6.8)
or
where the latter can again lead to an isotope effect with more D2 O2 than H2 O2 desorption
and a lower rate of reaction for D2 + DO2 through tunneling (see right side of Fig. 6.5).
HO2 + H2 has a high barrier of 1.1 eV and is endothermic by -0.6 eV. However, in the
aforementioned study we have observed this reaction to proceed (Chapter 5). The exothermicity of O2 + H may help to overcome the barrier and the endothermicity, since the total
reaction
O2 + H2 → H2 O2
(6.9)
composed of
O2 + H → HO2 ,
(6.10)
HO2 + H2 → H2 O2 + H,
(6.11)
and
is exothermic by 1.3 eV.
To summarize, the observed isotope effect between H2 O/H2 O2 and D2 O/D2 O2 in the
O3 hydrogenation channel can be explained by a combination of effects. First, OD will
get more translational energy than OH in reaction (6.1). Then, if H2 O2 and H2 O are
formed through reactions with H2 , tunneling leads to a higher rate for hydrogenation than
deuteration and, secondly, the distribution of excess energy can lead to more D2 O/D2 O2
than H2 O/H2 O2 desorption. O3 is destroyed equally for H and D, which indicates that
reaction (6.1) proceeds without substantial tunneling effect.
Finally, all the results discussed here are obtained under laboratory conditions, which
in some cases differ order of magnitudes from interstellar conditions (e.g., time scale to
reach the same H-atom fluence). Therefore, the use of excess energy to allow further reaction steps, which can proceed under laboratory conditions, might not be favorable under
interstellar conditions. For instance, the excess energy of the OH radical formed through
reaction (6.1) can be dissipated in the ice before H2 would reach this OH on interstellar
timescales. However, two reaction steps might still apply to the ISM at low temperature
(10 K), if an H2 layer is available on the surface of the ice for further reactions.
120
Figure 6.5 Proposed reaction mechanism for the formation of H2 O (left side) and H2 O2 (right side) from hydrogenation of O3 ice.
Reactions are shown in brackets.
6.3 Results and discussion
121
6 Water formation by surface O3 hydrogenation
6.4 Conclusion
The present study shows that the water formation through hydrogenation of solid O3 ice
as proposed by Tielens & Hagen (1982) takes place under interstellar ice analog conditions. Hydrogenation of O3 ice exhibits a similar temperature dependency as seen for CO
ice (Chapter 2): the mobility of O3 molecules increases with the temperature, while the
penetration depth of H atoms into the ice involves only the first monolayers. For temperatures above the O2 desorption temperature, hydrogenation of O3 leads to erosion of the
ice, since O2 formed in the reaction O3 + H desorbs. The remaining OH can further react
to H2 O and H2 O2 . The erosion occurs until a layer of H2 O and H2 O2 layer covers the
ice and prevents the incoming H atoms from reaching the underlying O3 ice. It is found
that at high surface temperature (50 K) O3 is mobile enough to slowly diffuse through the
H2 O and H2 O2 layer and to become available for further hydrogenation on the surface of
the ice.
Experimental evidence is found for the connection of the O3 hydrogenation channel to
the O + H and O2 + H channels, as summarized in Fig. 6.1. As a result it has become possible to draw conclusions on several reactions that are part of the other two hydrogenation
channels. The results indicate that the reaction OH + H2 is most likely more efficient than
the reaction OH + H: reaction OH + H2 could proceed through tunneling, while reaction
OH + H needs to dissipate 5.3 eV of excess energy with just one product, which could
be difficult. Our experimental results complete the reaction scheme initially proposed in
Tielens & Hagen (1982) to explain surface water formation in space. The conclusion that
the three channels (O/O2 /O3 + H) are strongly linked, is of importance for astrochemical
models focusing on water formation under interstellar conditions.
122
CHAPTER 7
Competition between CO and O2-ice
hydrogenation channels and surface formation of
CO2 at low temperatures1
In the past decade astrochemical laboratory studies have focused on the investigation of
isolated surface reaction schemes, starting from the hydrogenation of simple and pure
ices, like solid CO or O2 . The reaction products observed for CO hydrogenation are
H2 CO and CH3 OH, while after O2 hydrogenation H2 O2 and H2 O are formed. We present
here the first systematic laboratory study focusing on H-atom addition to a mixed CO:O2
ice. Mixed ices are more relevant from an astrochemical point of view and can elucidate reactions with radicals that are not readily studied otherwise. The aim of this
chapter is to investigate the competition between CO and O2 hydrogenation, and the corresponding surface formation of CO2 for astronomically relevant temperatures. Mixtures
of CO:O2 = 4:1, 1:1 and 1:4 are deposited on a substrate in an ultra high vacuum setup at
low temperatures (15 and 20 K) and subsequently hydrogenated. The ice is monitored by
means of Reflection Absorption InfraRed Spectroscopy (RAIRS). The results show that
the contemporary presence of CO and O2 molecules in the ice influences the final product
yields of the separate CO + H and O2 + H channels, even though the formation rates are
not significantly affected. CO2 is efficiently formed through dissociation of the HO-CO
intermediate complex in all studied CO:O2 mixtures and within the experimental uncertainties no dependency on temperature or ice composition is observed. Moreover, the
CO + O and HCO + O channels are not efficient at low temperature under our experimental conditions. The laboratory results demonstrate that CO2 can be formed in interstellar
ices at low temperatures in the absence of UV radiation and show a correlation between
the formation of CO2 and H2 O, which is consistent with the astronomical observation of
solid CO2 in water-rich environments.
1 Based on: S. Ioppolo, Y. van Boheemen, H. M. Cuppen, E. F. van Dishoeck, and H. Linnartz, 2010,
submitted to Monthly Notices of the Royal Astronomical Society
123
7 CO + H vs. O2 + H and formation of CO2
CO
O
O
O2
OH
H
H
HCO
H
H
H
H
H
H2
H
H2 H
H
OH
H
HO−CO
H
H
CH 3OH
HO 2
O3
H2 O 2
H 2 CO
H 3 CO
H H
O
H
H
H H2
CO 2
H2
H 2O
Figure 7.1 A schematic representation of the reaction network as discussed in the present
study. The CO + H channel (Chapter 2) is shown on the left-side of the figure, while
the O/O2 /O3 + H channels are plotted as presented in Chapter 5 on the right-side. The
possible CO2 formation routes are shown in between the CO + H and O/O2 /O3 + H
channels: the dissociation of the HO-CO intermediate (solid arrow) is one of the topics
of this work; the hydrogenation of the HO-CO complex (solid arrow) is presented in
Chapter 8; the suggested CO + O (dashed arrow) and HCO + O (dotted arrow) routes
are not experimentally confirmed at low temperature (Tielens & Hagen 1982, Ruffle &
Herbst 2001).
7.1 Introduction
Infrared Space Observatory and Spitzer Space Telescope observations have shown that
H2 O, CO, CO2 , and, in some cases, CH3 OH represent the bulk of solid-state species
in dense molecular clouds and star-forming regions. Other ice components, such as CH4 ,
−
NH3 , OCN , H2 CO, HCOOH, SO2 , and OCS have abundances <5% relative to H2 O (e.g.,
Gibb et al. 2004, Boogert et al. 2008, Pontoppidan et al. 2008, Öberg et al. 2008, Zasowski
et al. 2009, Bottinelli et al. 2010). Several of these species are assumed to be formed in
solid state reactions on the surfaces of icy dust grains, first outlined by Tielens & Hagen
(1982). Although these reactions have been postulated nearly 30 years ago, few have been
measured in the laboratory at low temperatures and UHV conditions until recently. Over
the past decade, detailed laboratory studies have started to investigate isolated surface
reaction schemes, starting from the hydrogenation of simple and pure ices, like solid CO
or O2 .
Several groups proved that the hydrogenation of CO ice at low temperatures (12−20 K)
124
7.1 Introduction
leads to the subsequent formation of H2 CO and CH3 OH (e.g., Watanabe et al. 2004,
2006a, Fuchs et al. 2009). The experiments showed that this hydrogenation process involves only the upper monolayers (4 ML, where 1 ML corresponds to 1015 molecules
cm−2 ) of the CO ice and formation rates drop at temperatures higher than 15 K, since the
desorption of H atoms becomes important at these temperatures. The hydrogenation of
CO to CH3 OH proceeds in four steps,
H
H
H
H
CO −
→ HCO −
→ H2 CO −
→ H3 CO −→ CH3 OH,
(7.1)
where the first step from CO to HCO and the third step from H2 CO to H3 CO have a
barrier.
The other surface reaction channel that has been well investigated is the hydrogenation
of O2 ice, which leads to the formation of H2 O2 and H2 O (e.g., Miyauchi et al. 2008,
Ioppolo et al. 2008, 2010, Cuppen et al. 2010). This hydrogenation process
2H
2H
O2 −−→ H2 O2 −−→ 2H2 O
(7.2)
behaves differently compared to the hydrogenation of CO ice (reaction scheme 7.2 shows
the simplified version of this route, as discussed by Tielens & Hagen 1982). In this case,
the penetration depth of H atoms in the O2 ice increases with temperature, even at values
close to the desorption temperature of the O2 layer, involving the bulk of the ice (tens
of monolayers). Thus, H atoms trapped in the ice can diffuse and eventually react up to
much higher temperatures. Moreover, at least the formation of H2 O2 does not exhibit any
noticeable barrier.
The present work is a further step towards a laboratory investigation of surface reactions in a more complex and realistic interstellar ice analogue by studying the competition between the two hydrogenation channels in a binary CO:O2 ice mixture as well as
the inherent formation of solid CO2 . The latter is not found as a reaction product for
the separate channels. CO2 is one of the most common and abundant ices, yet its formation routes are still very uncertain. Figure 7.1 shows a schematic representation of all
the reaction networks investigated in this work (solid arrows) and links the previously
studied CO + H and O2 + H channels through the observed CO2 formation. The dashed
and dotted arrows represent suggested CO2 formation routes in the networks of Tielens &
Hagen (1982) and Ruffle & Herbst (2001), that are not experimentally confirmed at low
temperature in our studies. As discussed in § 7.4.4, CO2 is formed under our experimental conditions through the reaction CO + OH. Here OH radicals are formed through the
hydrogenation of O2 ice, while in space they can also result from the O + H reaction or
from photodissociation of H2 O ice. Our experiments are designed to test the interaction of
the aforementioned individual surface reaction channels rather than simulate a complete
realistic interstellar ice evolution. Hence, our experiments show that OH radicals can get
further hydrogenated, leading to H2 O formation, or, alternatively, can react with the CO
present in the ice, forming solid CO2 . These results give the experimental evidence for the
correlation of H2 O and CO2 formation. Indeed, observations by the Infrared Space Observatory (e.g., Gerakines et al. 1999, Nummelin et al. 2001, Gibb et al. 2004) and the
125
7 CO + H vs. O2 + H and formation of CO2
Spitzer Space Telescope (e.g., Boogert et al. 2004, Bergin et al. 2005, Pontoppidan et al.
2005, 2008, Whittet et al. 2007) show that roughly 23 of the solid CO2 observed in quiescent molecular clouds and star-forming regions is found in water-rich environments,
suggesting that the formation routes of these two molecules are linked. The remaining
CO2 ice is predominantly found in a H2 O-poor, CO-rich environment (Pontoppidan et al.
2008). The origin of this common ice mantle component remains uncertain.
It is widely accepted that CO2 is not formed efficiently in the gas phase, with subsegas
quent accretion onto the grains (CO2 /COice
2 1; van Dishoeck et al. 1996, Boonman
et al. 2003). Therefore, the observed CO2 most likely has to be formed in the solid phase.
Several reaction mechanisms have been proposed with a relevance depending on astronomical environment. Energetic processing, such as UV and ion irradiation of interstellar
ice analogues, has been investigated in various laboratories and proposed as an efficient
CO2 formation mechanism (Chapter 9 and references therein). Furthermore, in the absence of UV irradiation several cold solid-phase reaction channels have been reported in
the past decades as an alternative formation mechanism to explain the CO2 abundance
observed in cold clouds (e.g., Tielens & Hagen 1982, Ruffle & Herbst 2001, Stantcheva
& Herbst 2004, Fraser & van Dishoeck 2004, Goumans et al. 2008, Goumans & Andersson 2010). The most straightforward surface reaction channel is the addition of an O
atom to solid CO ice. The reaction CO + O → CO2 has been experimentally investigated only by temperature programmed desorption experiments using thermal O atoms
below 160 K (Roser et al. 2001) and by energetic O atoms (Madzunkov et al. 2006). This
surface reaction channel has a high reaction barrier, because the CO(1 Σ) + O(3 P) reactants do not correlate directly with the singlet ground state CO2 (1 Σ) (2970 K in the gas
phase; Talbi et al. 2006). Ruffle & Herbst (2001) found in their astrochemical model
that they were only able to reproduce the CO2 abundances observed towards the cold (10
K) cloud Elias 16, if they artificially lowered the barrier to 130 K. Recently, Goumans &
Andersson (2010) used harmonic quantum transition state theory to conclude that whilst
quantum mechanical tunneling through the activation barrier increases the classical reaction rate for reaction CO + O at low temperatures (10−20 K), the onset of tunneling
is at too low temperatures for the reaction to efficiently contribute to CO2 formation in
quiescent clouds.
Solid CO2 is further suggested to be formed through the surface reaction HCO + O,
which presents two exit channels (CO2 + H and CO + OH; Ruffle & Herbst 2001).
Alternatively, solid CO2 can be formed through the surface reaction CO + OH, which
yields a HO-CO intermediate. This complex can directly dissociate, forming solid CO2
and leaving a H atom, or can be stabilized by intramolecular energy transfer to the ice
surface and eventually react with an incoming H atom in a barrierless manner to form
CO2 and H2 (Goumans et al. 2008). Oba et al. (2010) investigated the reaction CO + OH
depositing CO molecules on a cold substrate (10 and 20 K) together with H2 O fragments
(OH, H, O and H2 ) produced by dissociating H2 O molecules in a microwave source. They
confirmed the formation of solid CO2 at low temperature by using CO and OH beams to
initiate surface reactions on a cold substrate. However, their experiments differ from ours.
In the present work we hydrogenate CO:O2 ices and OH radicals are produced in the ice
through the O2 + H channel. Therefore, our experiments give also hints on the interaction
126
7.2 Experimental details
between different surface reaction channels. In chapter 8 we investigated experimentally
the hydrogenation of the HO-CO complex, which presents three exit channels (CO2 + H2 ,
HCOOH, H2 O + CO) with a branching ratio purely statistical as suggested by density
functional theory models and in agreement with our experimental results (Goumans et al.
2008, Ioppolo et al. 2010a). In the present study the HO-CO complex itself is not observed
in the ices under our experimental conditions, since it is efficiently dissociated to form
CO2 . More details are reported in § 7.4. First, in § 7.2 and 7.3 the experimental method
and data analysis are discussed.
7.2 Experimental details
The experiments are performed using an ultra high vacuum setup (SURFRESIDE), which
consists of a main chamber (10−10 mbar) and an atomic beam line. Details are available
in Chapters 2 and 4. The ice is grown on a gold coated copper substrate (12−300 K)
that is mounted on the cold head of a close-cycle He cryostat. Deposition of selected
12 16
C O:16 O2 mixtures (4:1, 1:1 and 1:4) proceeds under an angle of 45◦ and with a rate of
0.7 ML min−1 . The interstellar solid CO:O2 mixing ratio is observationally constrained
to >1:1 (Pontoppidan et al. 2003). After deposition the ice mixture is exposed to a cold
H-atom beam. H2 molecules are dissociated into the capillary of a well-characterized
thermal cracking source (Tschersich & von Bonin 1998, Tschersich 2000, Tschersich
et al. 2008), which is used to hydrogenate the sample. A quartz pipe with a nose-shaped
form is placed along the path of the dissociated beam to efficiently thermalize all H atoms
to room temperature through surface collisions before they reach the ice sample. In this
way, hot species (H; H2 ) cannot reach the ice directly. Furthermore, the relatively high
temperature of 300 K of the incident H atoms in our experiments does not affect the experimental results, since H atoms are thermally adjusted to the surface temperature as has
been shown in Chapter 2. The final H-atom flux (2.5 × 1013 atoms cm−2 s−1 ) is measured
at the substrate position in the main chamber using a quadrupole mass spectrometer, following the procedure as described in the Appendix A of Chapter 4. The absolute error
in the H-atom flux determination is within 50%. Ices are monitored by means of Reflection Absorption InfraRed Spectroscopy (RAIRS) using a Fourier Transform InfraRed
spectrometer (FTIR), which covers the range between 4000 and 700 cm−1 (2.5−14 µm).
A spectral resolution of 1 cm−1 is used and 128 scans are co-added. RAIR difference
spectra (∆A) relative to the initial unprocessed CO:O2 ice are acquired every few minutes
during H-atom exposure.
We performed a control experiment at 15 K in which a CO:O2 ice is exposed to an
H2 molecular beam (i.e., without H atoms) to show that the products detected in the hydrogenation experiments are formed on the surface and do not originate from background
deposition. Only H2 O is detected in this experiment, which gives us an estimate for the
background contamination, that is negligible. None of the other products are detected
in this way. It should be noted that we use in the present experiments the same H-atom
flux as used in Chapters 2 and 4. Therefore, the hydrogenation of our mixtures involves
effectively a lower H-atom flux per (CO and O2 ) channel.
127
7 CO + H vs. O2 + H and formation of CO2
0.003
CO2
CO
CH3OH
∆Absorbance
0.002
H2CO
0.001
0
-0.001
H2O
2250
2000
1750
H2O2
1500
-1
1250
1000
Wavenumber (cm )
Figure 7.2 RAIR difference spectrum of the CO:O2 =1:4 ice, with respect to the spectrum
before H-atom addition, at 15 K after a H-atom fluence of 1.3 × 1017 atoms cm−2 .
7.3 Data analysis
Figure 7.2 shows a RAIR difference spectrum of a CO:O2 = 1:4 mixture acquired after
a H-atom fluence (flux × time) of 1.3 × 1017 atoms cm−2 . The negative peak shown in
Fig. 7.2 is caused by the CO use-up in surface reaction processes. O2 is infrared in-active,
and therefore cannot be observed in the infrared spectrum. All the final reaction products
obtained from the hydrogenation of a pure CO ice (H2 CO and CH3 OH; e.g., Watanabe
et al. 2004, 2006a, Chapter 2) and a pure O2 ice (H2 O and H2 O2 ; e.g., Miyauchi et al.
2008, Chapters 3 and 4) are present. Neither the intermediate species from the separate
CO and O2 channels, like HCO, H3 CO, HO2 , OH, and O3 , nor more complex species,
like the stabilized HO-CO intermediate, HCOOH, and H2 CO3 1 , are observed. However,
a new feature appears at ∼2344 cm−1 , which belongs to the asymmetrical stretching mode
of CO2 ice. This molecule is formed in all our performed experiments at 15 and 20 K,
with different CO:O2 mixing ratios, but was not found in the previous CO + H or O2 + H
experiments.
The infrared spectra are reduced to obtain the column densities of the newly formed
species. As a first step in the infrared data analysis, a straight baseline is subtracted from
all spectra. Some absorption features, like the H2 O bending mode (∼1650 cm−1 ) and the
H2 CO ν(C=O) stretching mode (∼1720 cm−1 ) suffer from spectral overlap. Here a multiGaussian fit is used to determine the area of the selected bands. Since the asymmetric
1440 cm−1 H2 O2 band overlaps with the 1500 cm−1 H2 CO band, a spectrum of pure H2 O2
1 Solid
128
H2 CO3 is observed only in control experiments presented in Chapter 8.
7.4 Results and discussion
ice is fitted in addition to a Gaussian to our infrared spectrum. The spectrum of solid H2 O2
is obtained as discussed in Chapter 5, by co-depositing H atoms and O2 molecules with
a ratio of H/O2 = 20 and subsequently heating the ice to a temperature higher than 30 K,
which is just above the O2 desorption temperature (Acharyya et al. 2007).
−2
The
R column density NX (molecules cm ) of species X in the ice is calculated using:
NX = A(ν)dν/S X , where A(ν) is the wavelength dependent absorbance. Since literature
values of transmission band strengths cannot be used in reflection measurements, an apparent absorption band strength, S X of species X is determined by individual calibration
experiments. These have been described in detail in Chapters 2, 4 and 5. Like for CO,
CH3 OH and H2 O, an isothermal desorption experiment has been performed to determine
the apparent absorption band strength of CO2 by determining the transition from zerothorder to first-order desorption. This is assumed to occur at the onset to the submonolayer
regime and appears in the desorption curve as a sudden change in slope. Since pure H2 CO
and H2 O2 are experimentally difficult to deposit, because of their chemical instability, the
values for S H2 CO and S H2 O2 are obtained by assuming mass balance as reported in Chapters 2 and 5, respectively.
7.4 Results and discussion
7.4.1 Hydrogenation of O2 molecules
Figure 7.3 shows the H2 O2 (top panels) and H2 O (bottom panels) column densities as a
function of the H-atom fluence for the three different mixing ratios (CO:O2 = 4:1 circles,
1:1 squares and 1:4 triangles) and two temperatures investigated (15 K left panels and
20 K right panels). For comparison, results from the hydrogenation of pure O2 ice (Chapter 4) are also plotted (diamonds). Note that the top-right panel has a different scale for
the column density than the other three diagrams. Formation rate and final yield of H2 O2
and H2 O for all the investigated mixtures are lower than those from the pure O2 ice hydrogenation. The differences in the final yield are more evident at higher temperature, where
the yield for the mixed ices only moderately increases, whereas it increases with several
monolayers for the pure O2 experiments. This cannot be explained only by a low effective
H-atom flux for the O2 channel. Hence, the presence of CO in the mixture influences the
final results. In Chapter 2 we showed that H atoms can penetrate only a few layers of CO
ice. Therefore, the presence of CO in the mixture most likely diminishes the penetration
depth of H atoms into the ice compared to O2 + H, and, therefore, desorption of H atoms
from the ice can become important at higher temperatures. This explains the difference in
the H2 O2 and H2 O final yields compared to those from the pure O2 ice, which increases
with temperature.
The formation rates of H2 O and H2 O2 , which are reflected by the initial slopes of the
curves, is also altered by the presence of CO in the ice. In the O2 -rich ice (1:4) the H2 O2
column density shows the same behavior as seen in pure O2 ices: a constant formation rate
is followed by a sharp transition toward saturation (Chapter 4). For high concentration of
CO in the ice, the H2 O2 column density increases with a much lower rate and does not
129
7 CO + H vs. O2 + H and formation of CO2
Time (minutes)
15
-2
H2O2 (10 mol cm )
0
50
100
T = 15 K
3
150
200 0
H2O2
50
100
T = 20 K
150
200
H2O2
6
2
4
1
2
0
T = 15 K
3
H2 O
T = 20 K
H2 O
3
2
2
1
1
15
-2
H2O (10 mol cm )
0
0
0
0
1
2
3
17
0
1
2
3
-2
Fluence (10 atoms cm )
Figure 7.3 H2 O2 (top panels) and H2 O (bottom panels) column densities as a function of
the H-atom fluence and time of H-atom exposure at 15 K (left panels) and 20 K (right panels) for the three mixtures studied: CO:O2 = 4:1 (circle), 1:1 (square), and 1:4 (triangle).
For comparison, results from the hydrogenation of pure O2 ice are plotted (diamond).
Note different vertical scale for upper right panel.
appear to reach a steady state, even at the highest fluence. The H2 O2 final yield increases
with temperature, like for hydrogenation experiments of pure O2 ice. The amount of
H2 O2 formed in the ice is inversely proportional to the amount of CO in the mixture, as
expected.
In the case of CO-rich ice (4:1), a more efficient conversion of H2 O2 ice into H2 O
ice can explain the high H2 O final yield with respect to the O2 -rich ice experiment. In
chapter 4 we showed that H2 O2 is more effectively formed in the bulk of the ice (Chapter
4). However, the presence of CO in the ice limits the hydrogenation reactions to the
surface of the ice. This means that a larger percentage of H2 O2 formed at the surface of
the ice is easily converted into H2 O. This may also explain the lower effective synthesis
of H2 O2 with the increase of the number of CO molecules in the ice. In addition, H2 O can
be formed from OH radicals (see Fig. 7.1 and Chapter 5), which can also react to form
CO2 in our ices. The H2 O column density is constant through almost all our experiments.
This is also the case for CO2 as we will show in § 7.4.4 and indicative for a correlation
between the formation channels of these two species.
130
7.4 Results and discussion
Time (minutes)
15
-2
CH3OH (10 mol cm )
15
-2
H2CO (10 mol cm )
0
2
50
100
T = 15 K
150
200 0
H2CO
50
100
T = 20 K
150
200
H2CO
2
1
1
0
0
2
T = 15 K
CH3OH
T = 20 K
CH3OH
2
1
1
0
0
0
1
2
3
17
0
1
2
3
-2
Fluence (10 atoms cm )
Figure 7.4 H2 CO (top panels) and CH3 OH (bottom panels) column densities as a function of the H-atom fluence and time of H-atom exposure at 15 K (left panels) and 20 K
(right panels) for the three mixtures studied: CO:O2 = 4:1 (circle), 1:1 (square), and 1:4
(triangle). For comparison, results from the hydrogenation of pure CO ice are plotted
(diamond).
7.4.2 Hydrogenation of CO molecules
Figure 7.4 shows the H2 CO (top panels) and CH3 OH (bottom panels) column densities
as a function of the H-atom fluence for the three different mixing ratios (CO:O2 = 4:1
circles, 1:1 squares and 1:4 triangles) and two temperatures investigated (15 K left panels and 20 K right panels). For comparison, results from the hydrogenation of pure CO
ice (Chapter 2) are also plotted (diamonds). For the hydrogenation of pure CO ice, the
H-atom fluence is corrected according to recent H-atom flux measurements (Chapter 4),
which improve the original H-atom flux estimation derived in Chapter 2. The hydrogenation of CO molecules in our mixtures shows the same behavior seen for pure CO ice in
terms of temperature dependence (Chapter 2). An optimum in the final yield for H2 CO
and CH3 OH is found at 15 K, while at 20 K no CH3 OH is formed and H2 CO has a low
formation rate. The H2 CO and CH3 OH formation rates are hardly affected by the presence of O2 in the ice (within the experimental uncertainties). The higher final yield for
H2 CO and CH3 OH in the CO:O2 = 4:1 experiment at 15 K compared to the pure CO ice
131
7 CO + H vs. O2 + H and formation of CO2
cannot be explained by only a difference in the effective H-atom flux. In this case, the
presence of O2 in the ice, as a minor component, increases the penetration depth of H
atoms in the ice compared to pure CO and, therefore, the probability that the H atoms get
trapped and react in the ice. However, if the O2 concentration is increased in the 15 K ice,
the final yield decreases for both H2 CO and CH3 OH molecules. Clearly, the formation of
H2 CO is more sensitive to the O2 concentration in the ice than CH3 OH. As we saw for
the hydrogenation of O2 ice, the intermediate products (H2 O2 and H2 CO respectively) are
more efficiently converted in the final products (H2 O and CH3 OH, respectively), when
the ices are mixed.
7.4.3 Competition between the CO and O2 channel
The results presented in the former sections reflect the competition between the two channels CO vs. O2 , as shown in the left and right part of Fig. 7.1. It is clear from the
experimental results that the presence of one component in the ice influences the reactivity of the other component. The formation rate of the CO hydrogenation reaction products
is less affected by the presence of O2 than the O2 hydrogenation reaction products are affected by the presence of CO. This can be explained by the lower penetration depth of H
atoms in CO ice and by the formation of CO2 as an additional product, since OH radicals,
formed through the O2 channel, are used to yield CO2 instead of H2 O and H2 O2 .
In a CO-rich environment at 15 K, the presence of O2 molecules enhances the production of H2 CO and CH3 OH, since H atoms can penetrate deeper in the ice than in the
pure CO ice experiment. However, the formation rate of the final products seems not to
be affected by the presence of the O2 molecules in the ice. Moreover, in a CO-rich ice
the formation of H2 O2 is limited by the small amount of O2 molecules in the ice, by the
amount of OH radicals used to form CO2 and by the lower penetration depth of H atoms
in the ice, caused by the presence of CO molecules. H2 O2 ice is also more efficiently
converted to H2 O on the surface of the ice. This explains the high final yield for H2 O in a
CO-rich environment at 15 K.
In a O2 -rich environment at 15 K, the formation of H2 CO and CH3 OH is limited by
the small amount of CO molecules, which limits the formation of H2 O2 and H2 O, since
the penetration depth of H-atoms is lower than in the pure O2 ice experiment.
At 20 K, the CO channel is not efficient, although the H atoms penetrate deep in the
ice; at this temperature H atoms prefer to react with O2 molecules. Also in this case the
final yields for H2 O2 and H2 O are lower than those in the pure O2 ice experiment.
7.4.4 Formation of solid CO2
Figure 7.5 shows the CO2 column density as a function of the H-atom fluence, confirming
the CO2 formation for the three different mixing ratios and two temperatures investigated.
Neither the CO2 formation rate nor its final yield depend significantly on either temperature or mixing ratio for the values studied here. Such a behavior is unexpected, since the
132
7.4 Results and discussion
Time (minutes)
-2
0.2
15
CO2 (10 mol cm )
0
0.1
50
100
150
200 0
T = 15 K
50
100
150
200
T = 20 K
0.2
0.1
0
0
0
1
2
3
17
0
1
2
3
-2
Fluence (10 atoms cm )
Figure 7.5 CO2 column density as a function of the H-atom fluence and time of H-atom
exposure at 15 K (left) and 20 K (right) for the three mixtures studied: CO:O2 = 4:1
(circle), 1:1 (square), and 1:4 (triangle).
separate reaction routes CO + H and O2 + H clearly depend on temperature, as shown in
Chapters 2 and 4. The limiting factor for the CO2 synthesis in our experiments is, therefore, the amount of ice that can be penetrated by H atoms, which is only a few monolayers.
This is caused by the presence of CO molecules affecting the penetration depth of the H
atoms in the ice (Chapter 2). Thus, the amount of CO2 formed in all our experiments is
always less than a monolayer. CO2 subsequently does not contribute to further molecular
synthesis in the ice upon ongoing hydrogenation. Bisschop et al. (2007b) showed experimentally that CO2 does not react with H atoms and is a stable molecule under interstellar
ice analogue conditions.
Figure 7.1 summarizes schematically the reaction network which leads to the formation of solid CO2 starting from the combination of the CO + H and O2 + H channels.
Analyzing the species present in our ice after H-atom addition we can identify which
reaction channel is most likely responsible for the formation of solid CO2 ice. The hydrogenation of the HO-CO intermediate (black arrow in the center of Fig. 7.1) should not
occur in our experiments, since HCOOH is not detected in the infrared spectra. Density
functional calculations (Goumans et al. 2008), confirmed by our previous experimental
results (Chapter 8), suggest that the final products from the hydrogenation of the HO-CO
complex have a purely statistical branching ratio. Therefore, HCOOH should be detected
in the ice as well if CO2 would be produced through this route, and this is not the case.
The oxidation of solid CO (dashed arrow) is also not likely to be the main formation
reaction channel, since O3 , which would indirectly prove the presence of abundant O
atoms in the ice, is not observed. O3 ice has been detected in pure O2 hydrogenation
experiments only for temperatures higher than 25 K, when the penetration depth of the H
133
7 CO + H vs. O2 + H and formation of CO2
atoms is higher than a few monolayers and the O2 molecules are most likely more mobile
(Chapter 4). The hydrogenation of O3 ice (Chapter 6) is therefore not considered here.
Furthermore, Roser et al. (2001) tentatively observed CO2 formation through this channel
only during the warm up of the ice and when CO molecules and O atoms were covered
by a thick layer of H2 O ice, which allows the reactants to remain trapped in the ice for
T > 100 K. The CO + O reaction contributes at best at high temperatures. Reaction
HCO + O (dotted arrow) should also be ruled out, since HCO radicals prefer to react in
a barrierless manner with H atoms forming H2 CO rather than with O atoms, as shown in
Chapter 2 and 8. Furthermore, O atoms are not abundant in our ices.
At low temperatures CO2 is therefore formed through the direct dissociation of the
HO-CO complex in the ice (black arrow). The HO-CO complex is efficiently dissociated
and, therefore, is not detected in our infrared spectra as a stable species. In chapter 4 we
observed this complex only in a water-rich environment. H-bonding should, indeed, improve coupling and heat dissipation through the ice, which stabilizes the HO-CO complex
more easily in a polar environment than in an apolar one. Our ice is mainly composed of
CO and O2 , with a polar component on the surface of the ice. The amount of the HO-CO
intermediate stabilized in the polar ice is also under the detection limit. Therefore, in
a water-poor ice the competition between dissociation and further hydrogenation of the
HO-CO complex is in favor of the dissociation. H2 O is also formed through hydrogenation of the OH radicals. Hence, the formation of CO2 is linked to the formation of H2 O
in the ice. This is consistent with the presence of CO2 in polar interstellar ice mantles.
7.5 Astrophysical implications
Recently, results from Spitzer Space Telescope observations (Whittet et al. 2007, Pontoppidan 2006, Pontoppidan et al. 2008) have shown that the formation of CO2 in dark
quiescent clouds occurs in two distinct phases. In the early stages, CO2 forms together
with H2 O on the surface of the interstellar dust grains, creating a polar ice mantle. A
second phase in the CO2 formation occurs during the heavy freeze-out of CO. During this
second phase, a H2 O-poor ice is formed.
Our experimental results indeed make it likely that CO2 and H2 O are formed together
in the early stages of the clouds through surface reactions assuming that both CO and OH
are present in sufficiently high abundances. H2 O ice forms from continued hydrogenation
of OH radicals formed on the surface of the dust grains. Alternatively, OH radicals can
react with nearby CO molecules, which are present in small amounts in the ice before the
strong CO freeze-out phase, forming CO2 ice through the direct dissociation of the HOCO intermediate. CO2 can be also formed at low temperatures through the hydrogenation
of the HO-CO complex, which can lead to the formation of HCOOH and H2 O + CO as
well as CO2 + H2 . The concentration of CO in the ice with respect to H atoms determines
the probability of OH to react with a CO molecule or with another H atom.
In the second stage, during the heavy CO freeze-out, the gas density is >105 cm−3
and the CO accretion rate could be as high as, or even higher than, the H-atom accretion
rate, which makes CO more abundant on the surface than H atoms. OH radicals will
134
7.6 Conclusions
therefore more likely react with a nearby CO molecule than with H atoms. CO2 can thus
be efficiently formed through the dissociation or further hydrogenation of the HO-CO
complex, while just little H2 O ice formed.
Energetic processing (UV irradiation and cosmic ray-induced photons) of polar and
apolar ices is also an efficient mechanism for CO2 formation for specific environments
(Hagen et al. 1979, Mennella et al. 2004, Mennella et al. 2006, Loeffler et al. 2005, Ioppolo et al. 2009). All these channels could contribute to the total CO2 column density
component observed in quiescent clouds.
The experiments presented here are designed to test a possible CO2 formation route
(thermal CO + OH) under interstellar ice analogue conditions rather than simulate a complete realistic interstellar ice evolution. Although our experiments do not exclude other
possible CO2 formation mechanisms, they show that the dissociation of the HO-CO complex is efficient and can contribute to explain the presence of CO2 in polar and apolar
interstellar ices at low temperatures in absence of UV irradiation.
7.6 Conclusions
The present laboratory study shows that the CO and O2 channels influence each others
final product yields, when CO and O2 molecules are mixed and hydrogenated at low
temperature (15 and 20 K). The formation rate for all the final products is found to be
less sensitive on the mixture composition than the final yield. The penetration depth
of the incoming H atoms is the main limiting factor. It depends on the composition of
the ice and decreases when the amount of CO in the ice increases. Our results show
that the formation rates found for H2 CO, CH3 OH, H2 O2 and H2 O are similar within the
experimental uncertainties to those found studying the isolated CO and O2 hydrogenation
channels corrected for the reduced effective H-atom fluxes. Therefore, the formation rates
found in the isolated studies of the CO + H and O2 + H channels are still valid for use in
astrochemical models.
The formation of CO2 from the reaction CO + OH is found here. CO2 is efficiently
formed under our laboratory conditions and no dependence on temperature or ice composition is found. The formation of CO2 is linked to the formation of H2 O and, therefore,
competes with the O2 hydrogenation channel in our experiments. The competition of
these two channels, together with the composition of the ice and the penetration depth of
H atoms into the ice, explains the differences in the H2 O2 and H2 O formation rate between
our results and the hydrogenation of pure O2 ice.
Figure 7.1 shows how the H2 O formation through the O/O2 /O3 + H channels is linked
to the CO2 formation. Here we investigated only the O2 + H channel, even though OH
radicals can be efficiently formed on dust grains through all H2 O formation channels as
well as through the photodissociation of H2 O and CH3 OH ice. Thus, our experimental
result on the efficiency of the CO + OH channel at low temperature has important astrophysical implications on the formation of solid CO2 in cold dense molecular clouds
shielded from strong UV fields and are consistent with the observation of solid CO2 in
H2 O-rich environments.
135
CHAPTER 8
Surface formation of HCOOH at low temperature 1
The production of formic acid (HCOOH) in cold and hot regions of the interstellar medium
is not well understood. Recent gas-phase experiments and gas-grain models hint at a
solid-state production process at low temperatures. Several surface reaction schemes
have been proposed in the past decades, even though experimental evidence for their
efficiency was largely lacking. The aim of this work is to give the first experimental evidence for an efficient solid-state reaction scheme providing a way to form HCOOH under
astronomical conditions. Several surface reaction channels have been tested under fully
controlled experimental conditions by using a state-of-the-art ultra high vacuum setup
through co-deposition of H atoms and CO:O2 mixtures with 4:1, 1:1 and 1:4 ratios. During deposition spectral changes in the ice are monitored by means of a Fourier Transform
InfraRed (FTIR) spectrometer in Reflection Absorption InfraRed (RAIR) mode. After
co-deposition a Temperature Programmed Desorption (TPD) experiment is performed
and gas-phase molecules are detected by a Quadrupole Mass Spectrometer (QMS). Formation of HCOOH is observed at low temperatures mainly through hydrogenation of the
HO-CO complex, while reactions with the HCO radical as intermediate are found to be
inefficient. The HO-CO complex channel, which was previously not considered as an
important HCOOH formation route, can explain the presence of HCOOH in dense cold
clouds, at the beginning of the warm-up phase of a protostar, and, therefore, is likely to
be astrochemically relevant.
1 Based on: S. Ioppolo, H. M. Cuppen, E. F. van Dishoeck, H. Linnartz, 2010, accepted for publication in
Monthly Notices of the Royal Astronomical Society
137
8 Surface formation of HCOOH at low temperature
8.1 Introduction
Formic acid (HCOOH), the smallest organic acid, has been observed in the past decades
toward high and low mass star forming regions and quiescent clouds in the gas phase (e.g.,
Zuckerman et al. 1971, Winnewisser & Churchwell 1975, van Dishoeck et al. 1995, Gibb
et al. 2000a, Ikeda et al. 2001, Liu et al. 2001, 2002, Requena-Torres et al. 2006, Bottinelli
et al. 2007, Bisschop et al. 2007c), and likely also in the solid phase (e.g., Schutte et al.
1998, 1999, Gibb et al. 2000b, 2004, Boogert et al. 2004, 2008, Knez et al. 2005).
Despite the detection of HCOOH in a variety of interstellar environments, its astrochemical origin is still unclear. Leung et al. (1984) discussed in their gas-phase model
that HCOOH can be formed in dense interstellar clouds through the dissociative recombination of protonated formic acid (HCOOH+2 ), which forms by radiative association of
HCO+ and H2 O. Irvine et al. (1990) reported a detection of HCOOH in L134N with a
relative abundance with respect to H2 of 10−10 . They attributed the formation of HCOOH
in this cold dark cloud to the ion-molecule gas-phase reaction CH4 + O+2 followed by a
dissociative recombination of protonated formic acid. According to Vigren et al. (2010),
who combined a laboratory study and a gas-phase model, dissociative recombination of
protonated formic acid is not as efficient as previously thought and the branching ratio of
the channel leading to HCOOH has a maximum of only ∼13%. Therefore, they suggested
that HCOOH is predominantly formed in dense interstellar clouds through surface reactions on grains, even though experimental evidence for the efficiency of surface reactions
is largely lacking.
Several solid-phase reaction channels have been proposed in the past decades. Tielens
& Hagen (1982) included in their astrochemical model the formation of solid HCOOH on
grain surfaces through successive addition of H, O, and H to CO ice:
O
H
CO + H → HCO −
→ HCO2 −
→ HCOOH.
(8.1)
As suggested by Garrod et al. (2006), HCOOH could also be formed through the solidstate reaction:
HCO + OH → HCOOH.
(8.2)
In a recent laboratory study, Öberg et al. (2009b) investigated, among others, reactions
(8.1) and (8.2) experimentally by UV processing of CH3 OH-rich containing ices in which
both HCO and OH are produced. However, only upper limits on HCOOH are found in this
study. The alternative route of direct hydrogenation of solid CO2 was tested by Bisschop
et al. (2007b), who did not detect any HCOOH formation at low temperatures (below
15 K).
In this work an alternative formation route is studied in detail. Goumans et al. (2008)
investigated the formation of CO2 on a carbonaceous surface representing a model grain
with density functional theory. According to their calculations the surface reaction CO
+ OH can yield a HO-CO complex, stabilised by intramolecular energy transfer to the
surface. This intermediate can subsequently react, in a barrierless manner, with an H
atom to form CO2 + H2 , H2 O + CO or HCOOH. Which of these three schemes is followed
appeared to depend only on the orientation of the HO-CO intermediate and the incoming
138
8.2 Experimental procedure
hydrogen atom. Therefore, the authors postulated that statistically about one third of
the events will form HCOOH. We will experimentally focus on solid HCOOH formed
through the latter route:
H
CO + OH → HO-CO −
→ HCOOH.
(8.3)
This chapter is organized according to the following experimental procedure. Starting
from hydrogenation of simple molecules, like CO and O2 , we investigate efficiency and
branching ratios of this surface reaction route. Simultaneous deposition (co-deposition)
of H atoms and CO:O2 mixtures with a selected ratio allows us to control the hydrogenation level and composition of the ice. Therefore, under our experimental conditions,
radicals trapped in a CO:O2 matrix are available for further reactions upon heating of
the ice (§ 8.2). Infrared spectroscopy and mass spectrometry are combined to constrain
the experimental results and to detect the formed and intermediate species (§ 8.3). Our
experiments are not designed to simulate a realistic interstellar ice, but to test the above
reactions. They provide information which can subsequently be included in astrochemical
models of interstellar clouds. Specifically, we discuss the astrophysical importance of this
reaction as an efficient channel for HCOOH formation at low temperatures (10 − 20 K) in
the dense interstellar clouds (§ 8.4).
8.2 Experimental procedure
The experimental set-up (SURFRESIDE) has been described in detail elsewhere (Chapters 2 and 4). Here we give a brief description of the apparatus, focussing more on the
experimental procedure. H atoms are deposited together with CO and O2 molecules (codeposition) on a gold coated copper substrate, placed in the centre of the ultra high vacuum main chamber (10−10 mbar). CO:O2 mixtures with a ratio of 4:1, 1:1 and 1:4 are
prepared in a high vacuum glass line, which comprises a liquid nitrogen trap to prevent
water pollution. During the co-deposition the substrate is kept at a temperature of 15 K by
a close-cycle He cryostat with a relative temperature precision of 0.5 K and an absolute
accuracy better than 2 K. Deposition of the mixtures occurs under an angle of 45◦ with a
flow of 5 × 10−8 mbar, while the H-atom beam is normal to the sample.
H atoms are supplied by a well-characterized thermal cracking source (Tschersich &
von Bonin 1998, Tschersich 2000, Tschersich et al. 2008). H2 molecules are cracked in a
capillary pipe surrounded by a tungsten filament, which is heated to 2200 K. During the
H-atom exposure, the pressure in the atomic line is kept constant at 1 × 10−6 mbar. Hot H
atoms are cooled to room temperature via collisions by a nose-shaped quartz pipe, placed
in the H-atom beam path. The geometry of the pipe is designed in such a way that each
H atom has at least four collisions with the walls before leaving the pipe. In this way, hot
species (H; H2 ) cannot reach the ice directly. Furthermore, previous experiments with liquid nitrogen cooled atomic beams did not show any H/D-atom temperature dependence in
hydrogenation reaction processes (e.g., Watanabe et al. 2006a, Miyauchi et al. 2008, Oba
et al. 2009). The final H-atom flux (2.5×1013 atoms cm−2 s−1 ) is measured at the substrate
139
8 Surface formation of HCOOH at low temperature
position in the main chamber using a quadrupole mass spectrometer, as described in the
Appendix A of Chapter 4. The absolute error in the H-atom flux determination is within
50%.
The ice is monitored by means of reflection absorption infrared spectroscopy, using
a Fourier transform infrared spectrometer. The FTIR covers the range between 4000 and
700 cm−1 (2.5−14 µm) with a spectral resolution of 1 cm−1 and a co-addition of 128
scans. RAIR difference spectra (∆A) with respect to the bare 15 K gold substrate spectrum are acquired every few minutes during the experiment. The QMS that monitors
gas-phase species is placed behind the substrate and in line with the HABS. Following
all our co-deposition experiments a temperature programmed desorption is performed by
heating linearly the co-deposited ice to 200 K with a rate of 0.5 K min−1 . Each separate
experiment is performed twice to link the RAIRS to the QMS data: during the first experiment the sample is kept in the IR optical line also during the TPD phase; in the second
experiment the sample is turned 180◦ to face directly the QMS after co-deposition.
The main goal of this work is to improve our qualitative picture of possible reaction
schemes and to search for an efficient HCOOH formation channel that may be responsible
for the observed abundances of HCOOH in quiescent clouds. To enhance the detection of
intermediates in the reaction schemes, we carry out a so-called co-deposited experiment
in which hydrogenation occurs while the mixture is deposited. Moreover, since a beam of
OH radicals and a H-dominated environment is difficult to produce, we use a mixture of
CO with O2 , because hydrogenation of O2 ice is known to lead to efficient OH production
(Chapter 4 and 5). The ratio between H atoms and CO:O2 mixtures during co-deposition
at 15 K determines the hydrogenation grade in our experiments. We are interested in the
CO and O2 dominated regime, where full hydrogenation is not reached and radicals are
trapped in the ice. Thus, the same H/[CO:O2 ] = 2 ratio is applied to all the H and CO:O2
co-deposition experiments, resulting in a deposited ice mainly consisting of CO and O2
molecules. RAIRS and QMS results from the co-deposition experiments are compared
to those from selected control experiments in order to give experimental evidence for the
unambiguous solid HCOOH formation under our laboratory conditions. For this purpose,
pure HCOOH and H2 CO ices and mixtures of H2 O:HCOOH and H2 O:H2 CO (10:1 and
3:1, respectively) are used as control experiments. The HCOOH containing ices are deposited at 30 K, while the H2 CO containing ices are formed in situ by hydrogenation of
several thin layers (1 ML per step) of pure CO ice or H2 O:CO mixtures at 12 K. The
hydrogenation is stopped once almost all the CO is converted into H2 CO ice: the H2 CO
formation yield has reached its maximum at this point and only traces of CH3 OH ice are
formed, as shown in Chapter 2.
8.3 Results and discussion
8.3.1 Formation of solid HCOOH
Figure 8.1 shows the normalized TPD curves obtained by the QMS with a rate of 0.5 K
min−1 for the 46 amu (HCOOH) and 30 amu (H2 CO) mass signals from the H-atom
140
8.3 Results and discussion
QMS signal (a.u.) + offset
HCOOH
4
H2CO
m/z = 46
m/z = 30
(c)
3
2
(b)
1
(a) x 500
x1
0
90
120
150
180
90
120
150
180
Temperature (K)
Figure 8.1 The 46 amu (HCOOH, le f t column) and 30 amu mass QMS signals
(H2 CO, right column) from the TPD of a H-atom and CO:O2 = 4:1 co-deposition
experiment (a), compared to TPDs of the following experiments: solid H2 CO (b,
solid line), H2 O:H2 CO = 3:1 mixture (b, dashed line), and HCOOH ice (c, solid line),
H2 O:HCOOH = 10:1 mixture (c, dashed line). Spectra are offset for clarity.
and CO:O2 = 4:1 co-deposition experiment compared to the normalized and offset TPD
curves of selected control experiments to prove unambiguously the formation of HCOOH
ice. Mass 46 amu (le f t column) from the co-deposition experiment has a broad desorption
peak at ∼160 K in the TPD curve with a low signal to noise ratio, due to the weakness
of this peak before normalization (Fig. 8.1a). Desorption of pure HCOOH occurs at 142
and 166 K under our experimental conditions (solid line, Fig. 8.1c), while HCOOH in
H2 O desorbs together with H2 O between 150 and 160 K (dashed line, Fig. 8.1c). This is
the desorption temperature range found in the co-deposition experiment for mass 46 amu
(Fig. 8.1a). Similar TPD experiments using pure H2 CO ice (solid line, Fig. 8.1b) and
H2 CO in H2 O ice (dashed line, Fig. 8.1b) reveal no desorption peaks for mass 46 amu.
For this we conclude that the signal from mass 46 amu detected during the TPD of the
co-deposition experiment corresponds to HCOOH desorption from a polar ice and that all
detected HCOOH originates from surface reactions.
The 30 amu mass signal (right column) in the TPD curve from the co-deposition
experiment presents a double peak at 147 and 160 K (Fig. 8.1a). Desorption of pure
H2 CO occurs at 95 K (solid line, Fig. 8.1b), while H2 CO in H2 O desorbs at 97 and 148 K
(dashed line, Fig. 8.1b), which is close to one of the desorption temperatures found in the
141
8 Surface formation of HCOOH at low temperature
co-deposition experiment (Fig. 8.1a). In a TPD experiment of pure HCOOH (solid line,
Fig. 8.1c), the 30 amu mass signal is detected at the same temperature found for mass
46 amu, showing that HCOOH can be fragmented into H2 CO by the electron-emitting
filament in the ionization chamber of the QMS. In the TPD curve of solid HCOOH in
H2 O (dashed line, Fig. 8.1c), the 30 amu mass signal peaks at 148 K, which is the same
desorption temperature of H2 CO in H2 O (dashed line, Fig. 8.1b). Hence, when HCOOH
ice is diluted in H2 O, some of it is already dissociated in the solid phase to form H2 CO at
high temperatures.
In conclusion, we find that HCOOH is formed in the solid phase for temperatures
between the co-deposition of H atoms and CO:O2 mixtures at 15 K and its desorption at
∼160 K. In the following sections we will use infrared data to constrain the temperature
of HCOOH formation and its reaction pathway.
8.3.2 Formation temperature
As a first step in the infrared data analysis, a straight baseline is subtracted from all
spectra. HCOOH ice presents several absorption features in the spectral range from
2000−1000 cm−1 , among which ν(C=O) stretching mode at ∼1700 cm−1 is the strongest
one. Shape and position of the HCOOH infrared spectral features are sensitive to temperature and ice composition as was shown by Bisschop et al. (2007a). In order to further investigate the temperature at which HCOOH ice is formed, we have compared the infrared
spectra acquired during the TPD of the co-deposition experiments to selected laboratory
spectra in the range between 1000 and 2000 cm−1 . We will start close to the HCOOH
desorption temperature since the QMS recorded TPD results gave a positive identification
of HCOOH there and we will then go down in temperature to detect the first occurrence
of HCOOH. The top-left side of Fig. 8.2 shows the RAIR spectrum at 150 K of a H-atom
and CO:O2 = 4:1 co-deposition experiment. The broad 1720 cm−1 HCOOH stretching
mode overlaps with the 1650 cm−1 H2 O bending mode. Solid H2 O2 appears also in the
investigated spectral range as a strong and broad absorption. The top-right side of Fig. 8.2
shows the zoom-in of the RAIR spectrum in the range between 1450 and 1850 cm−1 . In
the same panel a broad component at 1720 cm−1 from the HCOOH:H2 O = 1:10 mixture at
50 K is plotted (dotted line with an offset). This feature not only reproduces the band profile and peak position of the HCOOH stretching mode, but also indicates that HCOOH is
mixed with H2 O in the ice at temperatures slightly below the HCOOH desorption temperature, confirming the mass spectrometry analysis presented in § 8.3.1. A H2 CO:H2 O=1:3
mixture spectrum at 50 K (dashed line) is also plotted to show how shape and position of
H2 CO ν(C=O) stretching mode differ from the 1720 cm−1 spectral feature.
The bottom-left panel of Fig. 8.2 shows the infrared spectrum at 50 K of the same Hatom and CO:O2 = 4:1 co-deposition experiment, while the bottom-right panel of Fig. 8.2
plots the zoom-in of the aforementioned spectrum and the 1740 cm−1 feature from a
pure HCOOH ice at 30 K (dotted line), compared to a spectrum of pure H2 CO at 50 K
(dashed line). The HCOOH feature present in our infrared spectrum at 50 K is narrow and
shifted to higher wavenumbers with respect to the 1650 cm−1 H2 O bending mode. Again,
142
8.3 Results and discussion
0.008
∆Absorbance + offset
0.006
0.006
T = 150 K
0.004
0.004
0.002
0.002
0
0
0.008
0.006
0.006
T = 50 K
0.004
0.004
0.002
0.002
0
0
1800
1600
1400
1200 1800
1700
1600
-1
1500
Wavenumber (cm )
Figure 8.2 Infrared spectra at 150 K (top panels) and 50 K (bottom panels) from the
H-atom and CO:O2 = 4:1 co-deposition experiment (solid lines). The left side of the
figure shows the spectra in the range between 1000 and 2000 cm−1 , while the right
side shows a zoom-in of the 1700 cm−1 region. For comparison two spectra of mixed
H2 O:HCOOH = 10:1 (dotted line) and H2 O:H2 CO = 3:1 (dashed line) are shown in the
top-right panel, and two spectra of pure HCOOH (dotted line) and H2 CO (dashed line)
are plotted in the bottom-right panel. Spectra are offset for clarity.
shape and peak position are better reproduced by the HCOOH than by H2 CO, since the
H2 CO band peaks at 1730 cm−1 and is more narrow than the HCOOH feature. This result
indicates that HCOOH is present in the ice at 50 K and is most likely formed at lower
temperatures. It is also less mixed with H2 O ice than at high temperatures. Hence, at high
temperatures HCOOH molecules can better diffuse in the ice and get mixed with H2 O2
and H2 O ice above 100 K, when H2 O ice starts restructuring.
Figure 8.3a shows the infrared spectrum of the ice after co-deposition of H atoms and
CO:O2 = 4:1 at 15 K, when the ice is hydrogenated but not yet heated. At this stage of
the experiment the HCOOH stretching mode is around our detection limit and a feature
at 1815 cm−1 is detected. The identification of this feature will be discussed in § 8.3.3.
The first clear detection of solid HCOOH in the CO:O2 = 4:1 experiment is shown in
Fig. 8.3b at ∼30 K, when CO and O2 are not completely desorbed yet from the ice (Öberg
et al. 2005, Acharyya et al. 2007). At this temperature the 1815 cm−1 intensity decreases
and the band broadens and shifts to 1820 cm−1 , while the HCOOH stretching mode appears at ∼1750 cm−1 . The HCOOH peak position and band shape are in agreement with
those from a CO:HCOOH = 100:1 mixture shown in Fig. 8.3c. Solid HCOOH is detected at 15 K in the co-deposition experiments of H atoms and CO:O2 = 1:1, and 1:4
143
8 Surface formation of HCOOH at low temperature
(spectra not shown). However, the feature at 1815 cm−1 is detected only in the H-atom
and CO:O2 = 1:1 co-deposition experiment, where an increase of the HCOOH stretching
mode is observed at 30 K, in agreement with the results from the CO:O2 = 4:1 mixture
shown in Fig. 8.3. In the case of the CO:O2 = 1:4 mixture we do not observe the feature
at 1815 cm−1 at 15 K and the increase of the HCOOH stretching mode at 30 K. Thus, the
1815 cm−1 feature requires the presence of CO in the mixture. Fig. 8.3d and e will be
discussed in the next section.
As the previous discussion showed, the HCOOH feature changes strongly in shape
and position as a function of temperature. This is generally true for the entire spectrum.
Figure 8.4 shows this clear temperature dependence in the morphology of the ice in a
CO:O2 = 4:1 co-deposition experiment after an H-atom fluence of 2.7 × 1017 atoms cm−2 .
In a CO- and O2 -rich environment, like the ice at 15 K of our co-deposition experiments, the H2 O2 and H2 O infrared bands are present in both hydrophobic and hydrophilic
contributions (monomer and bulk, respectively; see Fig. 8.4a. For feature assignments
see Chapter 5 and references therein). By increasing the temperature of the sample, the
amount of hydrophobic material decreases and the H2 O2 and H2 O bands broaden and
shift. Thus, the 50 K infrared spectrum presents several broad bulk features originating
from species like H2 O2 , H2 O, and bands due to CO2 , HCOOH and traces of CO, still
trapped in the mixture (Fig. 8.4b). In the 150 K spectrum (Fig. 8.4c) the CO stretching mode has disappeared, traces of solid CO2 are present in the ice, since a very weak
stretching mode band is still visible, and the HCOOH 1740 cm−1 feature has broadened
and shifted towards lower wavenumbers (1700 cm−1 ), overlapping with the 1650 cm−1
H2 O bending mode. The HCOOH feature decreases in the RAIR spectra at the same temperature where the signal of mass 46 amu increases in the mass spectrometer (∼160 K;
see Fig. 8.1a).
In conclusion, the results presented here show that HCOOH ice is formed during codeposition experiments at low temperature (∼15 K) and that its formation increases at
a temperature close to the desorption temperature of volatile species like CO, O2 , when
radicals produced by the hydrogenation of the ice are more mobile (∼30 K). In the next
section we describe the possible reaction mechanism.
8.3.3 Possible reaction pathways
Infrared spectroscopy is used to identify the possible reaction intermediates: HCO according to reactions (8.1) and (8.2) or HO-CO according to reaction (8.3).
As mentioned in § 8.2, the deposited ice mainly consists of CO and O2 molecules.
This is reflected by the saturated CO stretching band mode present in Fig. 8.4a and a very
weak O2 absorption band peaking at 1550 cm−1 (the presence of O2 in the ice causes
also distortions in the infrared spectrum, as discussed in Chapter 5). The infrared spectrum at 15 K presents a forest of bands in absorption due to newly formed species upon
hydrogenation of O2 ice, like H2 O2 , H2 O, HO2 , OH (see Chapter 5). CO ice is not hydrogenated in our experiments, since neither HCO, H2 CO nor CH3 OH features appear in
the RAIR spectra at 15 K (Fig. 8.4a). This is consistent with a previous conclusion that
144
8.3 Results and discussion
0.012
∆Absorbance + offset
(e)
0.008
(d)
(c)
0.004
(b)
(a)
0
1890
1845
1800
1710
1755
-1
Wavenumber (cm )
Figure 8.3 Infrared spectra from a H-atom and CO:O2 = 4:1 co-deposition experiment at
15 K (a), and 30 K (b) compared to a spectrum of CO:HCOOH = 100:1 at 15 K (c), a
spectrum from a CO:O2 = 100:1 co-deposition experiment at 15 K (d), and a spectrum
of CO:HCOOH = 1000:1 at 15 K (e) using higher H-atom flux (2 × 1014 cm−2 s−1 ) and
CO:O2 flow (1 × 1014 cm−2 s−1 ). Spectra are offset for clarity.
CO + H is less efficient than O2 + H (Chapter 2, 4 and 5). Therefore, since HCO and
H2 CO never form in the other co-deposition experiments, reactions (8.1) and (8.2), with
HCO as intermediate, are likely not to take place during our co-deposition experiments.
We performed a control co-deposition experiment, using higher H-atom flux (2 ×
1014 cm−2 s−1 ) and higher CO:O2 = 100:1 deposition rate (∼1×1014 cm−2 s−1 ), with the
intention to give an assignment to the low temperature 1815 cm−1 feature. This experiment allows to observe more reactions products, but with similar matrix properties to
the other co-deposition experiments. During this control experiment several species, like
OH, HO2 , H2 O, H2 O2 , O3 , CO2 , CO3 , HO-CO, HCOOH, HCO, H2 CO are formed in the
ice at 15 K. Table 1 lists the assignments of the features present in the spectral region
between 1700 and 1900 cm−1 , also shown in Fig. 8.3d. The band position and shape of
solid HCOOH formed at low temperature in the control experiment corresponds to the
HCOOH feature in a CO matrix (CO:HCOOH = 1000:1) as shown in Fig. 8.3e. The
cis and trans-HO-CO complexes appear in the control experiment spectrum in both environments: polar (in H2 O matrix at 1775 and 1820 cm−1 , respectively) and apolar ice
(in CO matrix at 1796 and 1833 cm−1 , respectively). The peak at 1815 cm−1 is detected
in our standard co-deposition experiments and can now be assigned by comparison with
the control experiment results to the stabilized trans-HO-CO complex in a polar environment. The cis-HO-CO complex absorption strength, which is normally weaker than that
145
8 Surface formation of HCOOH at low temperature
Table 8.1 Assigned infrared features with their corresponding reference in the range between 1700-1900 cm−1 as found in a control co-deposition experiment.
Species
Position
Reference
Matrix
(cm−1 )
HCO
1860
Milligan & Jacox (1964)
Ar
trans-HO-CO
1833
Milligan & Jacox (1971)
CO
trans-HO-CO
1820
Zheng & Kaiser (2007)
H2 O
cis-HO-CO
1796
Milligan & Jacox (1971)
CO
cis-HO-CO
1775
Zheng & Kaiser (2007)
H2 O
HCOOH
1755
this work
CO
HCOOH
1748
this work
CO
H2 CO
1737
Nelander (1980)
N2
H2 CO
1734
Nelander (1980)
N2
from the trans-HO-CO, is around the detection limit.
Solid HCOOH is therefore most likely formed through reaction (8.3), which can also
lead to the formation of CO2 + H2 and H2 O + CO (Goumans et al. 2008). The presence
of CO2 ice and the stabilized trans-HO-CO complex at 1815 cm−1 (Fig. 8.4), suggests
indeed that CO ice can react with OH-bearing species produced by hydrogenation of O2
ice, since this bond is only present after hydrogenation of CO and O2 containing ices. As
previously discussed, the HO-CO complex is only detected in a water-rich environment.
However, our ice is mainly composed of CO and O2 , with traces of other polar species.
Therefore, we would expect to detect the HO-CO complex also in a water-poor environment. Moreover, the amount of solid HCOOH detected at 15 K is close to the detection
limit for all the mixtures investigated, while CO2 is on average 10 times more abundant
than HCOOH in the ice at low temperature. These experimental results suggest that the
HO-CO complex formed in a CO and O2 matrix is mainly used to form CO2 by direct
dissociation of the complex, and only a small fraction of CO2 comes from reaction (8.3)
at 15 K. Furthermore, the HO-CO complex appears to be stabilized and shielded from
hydrogenation in a water-rich environment at low temperatures. It is indeed likely that
H-bonding improves coupling and heat dissipation through the ice, which would stabilize
the HO-CO complex more easily in a water-rich environment than in a CO and O2 ice (T.
P. M. Goumans, private communication). The HO-CO intermediate formed in a polar ice
is then available again to react at higher temperatures, when it is more mobile and hydrogen can still be present in the ice. In our experiments also solid H2 CO3 can be formed
from the same HO-CO complex through the reaction:
OH
CO + OH → HO-CO −−→ H2 CO3 ,
(8.4)
as shown by Oba et al. Oba et al. (2010). Moreover, reaction (8.4) can also lead to the
formation of CO2 + H2 O and H2 O2 + CO. At temperatures below 100 K the HCOOH
ν(C=O) stretching mode overlaps with the H2 CO3 one, hindering a spectral assignment
146
8.3 Results and discussion
of solid H2 CO3 in the infrared spectra. However, the different desorption temperatures of
these molecules allows an identification, since HCOOH desorbs at ∼160 K and H2 CO3 at
∼250 K.
∆Absorbance + offset
0.08
Thin layer of H2O on gold
(d)
T=190 K
H2O2 + H2O
0.06
HCOOH
H2O2
12
(c)
T=150 K
H2O2 + H2O
0.04
H2O2
H2O
CO2
12
HCOOH
CO2
12
CO
H2O
(b)
0.02
0
T=50 K
12
H2O2
(a)
H2O
HO
12
H2O2 + H2O
CO2
H2O2
CO
H2O2
HO2 H O
2 2
trans-HO-CO
O3
13
H2O O2
CO
HO2
T=15 K
4000
3500
3000
2500
2000
1500
1000
-1
Wavenumber (cm )
Figure 8.4 The 15 K (a), 50 K (b), 150 K (c), and 190 K (d) infrared spectra of the
H-atom and CO:O2 = 4:1 co-deposition experiment after an H-atom fluence of 2.7 ×
1017 atoms cm−2 . Spectra are offset for clarity.
Figure 8.4d shows the 190 K RAIR spectrum from the co-deposition experiment, in
which all the absorption bands have disappeared. Distortions at low wavenumbers in
the difference RAIR baseline spectrum are caused by the high temperature of the gold
substrate. This result excludes the presence of formed solid H2 CO3 in the ice in our codeposition experiments, since H2 CO3 desorbs at temperatures above 220 K (e.g., Zheng
& Kaiser 2007) and should therefore still be present at 190 K.
Oba et al. (2010) studied the formation of solid CO2 through surface reactions between carbon monoxide and non-energetic OH radicals, produced by dissociating H2 O
molecules in microwave-induced plasma. According to their results H2 CO3 was formed
through reaction (8.4), but no HCOOH was detected. The difference in the experimental
results between Oba et al. (2010) and this work can be found in a different flux composition. In their case, H2 O is dissociated and a combination of OH radicals and H atoms
gives the final flux, most likely one to one. In the present work we use an H-atom flux.
The OH radicals present in the ice are formed via surface reactions. Hence, the HO-CO
complex has a higher probability to find and react with a H atom instead of a OH radical,
which is less abundant in the ice.
147
8 Surface formation of HCOOH at low temperature
8.3.4 Branching ratio of reaction HO-CO + H
As shown in the previous sections, solid CO2 and H2 O, which are also observed in our
infrared spectra, can be formed together with HCOOH by reaction HO-CO + H (Goumans
et al. 2008). However, the determination of the branching ratios for this reaction is not
straightforward, since CO2 can also be formed during co-deposition by direct dissociation
of the complex HO-CO (Oba et al. 2010), and H2 O through hydrogenation of molecular
oxygen (Miyauchi et al. 2008, Ioppolo et al. 2008, Oba et al. 2009, Ioppolo et al. 2010,
Cuppen et al. 2010). Here we make an attempt to estimate the branching ratios of reaction
HO-CO + H at higher temperatures, when the other reaction routes to form CO2 and H2 O
are less efficient. At 50 K the three reaction products are all formed but have not yet
desorbed from the ice. In order to calculate the branching ratios based on our experimental
results, we assume that:
1. solid HCOOH is formed only through reaction (8.3);
2. the CO2 formed at 15 K comes mainly through direct dissociation of the complex
HO-CO in a water-poor environment and, therefore, is subtracted to the CO2 column density obtained at 50 K;
3. H2 O can be formed through hydrogenation of O2 and, therefore, the contribution
of this channel is estimated by comparing the results presented here to those from a
similar experiment (H/O2 = 10) shown in Chapter 5 and then subtracted to the H2 O
column density obtained in our 50 K spectrum.
Since literature values of transmission band strengths cannot be used directly in reflectance measurements, an apparent absorption strength of HCOOH, CO2 , and H2 O is
calculated from calibration experiments. The determination of this apparent absorption
strength is set-up specific. The calibration method is described in Chapter 2 and 4.
The values for the branching ratios obtained here after subtracting the contributions
from other possible reaction channels are in agreement with those presented by Goumans
et al. (2008), who suggested that the branching ratio is purely statistical:



HCOOH
k


H 

(8.5)
CO + OH −
→
CO
+
H
(0.7±0.3)k
2
2




H2 O + CO ≤k,
where k is the branching ratio of the channel leading to HCOOH. We conclude from the
50 K infrared spectrum that the amount of solid HCOOH and CO2 formed in our standard
H-atom and CO:O2 = 4:1 co-deposition experiment is ∼1±0.1 ML, while the amount
of H2 O is roughly the same as that produced after the same H-atom fluence in the Hatom and O2 co-deposition experiment (H/O2 = 10) shown in Chapter 5 (∼10±1 ML).
Therefore, H2 O ice is mainly formed through hydrogenation of O2 and the amount of
H2 O formed through reaction HO-CO + H could be equal to that of CO2 and HCOOH
within the experimental uncertainties. Results using the CO:O2 = 1:1 and 1:4 mixtures
confirmed the branching ratio values shown above.
148
8.4 Astrophysical implications
8.4 Astrophysical implications
The origin of the observed HCOOH in the interstellar medium has been unclear. Both
gas-phase reactions and grain-surface processes have been suggested for producing the
HCOOH column densities observed in star forming regions. Results from a two-stage
hot-core gas-grain chemical model (stage 1: collapse; stage 2: warm-up phase) proposed
by Garrod et al. (2006) suggested that HCOOH is mainly formed in the gas phase during
the warm-up phase. According to the authors, at early times of the warm-up, HCOOH
is most likely formed in the gas phase through the dissociative recombination of protonated formic acid, which forms by radiative association of HCO+ and H2 O or by reaction
CH4 + O+2 . However, as mentioned in the introduction, recent gas-phase laboratory experiments showed that the dissociative recombination channel is less efficient than considered before. In the case of L134N for instance, it is found to be 8 times less efficient than
in models using the latest release of the UMIST database for astrochemistry (Woodall
et al. 2007), and a factor of 4 less than the observed value. Hence, the observed HCOOH
abundances in cold regions cannot be explained by gas-phase reactions only, and, therefore, surface reactions should be taken into account for the formation of HCOOH, even
at low temperatures. Especially, since solid HCOOH may be destroyed by energetic processing or in reaction with NH3 to lead to HCOO− and NH+4 (e.g., Hudson & Moore 1999,
Schutte et al. 1999). HCOOH ice was shown in the laboratory to be stable against further
hydrogenation (Bisschop et al. 2007b).
The model used by Garrod et al. (2006) includes only surface reaction (8.2). At low
temperatures, the formation of formaldehyde and methanol through subsequent hydrogenation of solid CO (Chapter 2) is favored over the formation of HCOOH through reaction (8.2). We can roughly estimate the efficiency of reaction (8.2) at low temperatures
considering the abundance of H atoms and OH radicals over CO molecules at the end of
the cloud collapse phase. If we assume that one H2 O molecule is formed on the grain
surface from one OH radical, which gives a ratio one-to-one between H2 O and OH, CO
ice is 4 times less abundant on average in dense regions than OH radicals (e.g., Gibb et al.
2004). However, H atoms are orders of magnitude more abundant than OH radicals in this
environment and H atoms are more mobile on the surface. Therefore, the probability that
an HCO radical finds an OH to react with before it is hydrogenated is almost negligible.
Garrod et al. (2006) found an optimum in HCOOH formation through reaction (8.2) at
∼40 K, when H2 CO resides in the gas phase, but can also accrete again on the grains and
re-evaporate quickly. This allows for reaction with OH radicals in the solid phase to form
HCO available for reaction (8.2). However, grain-surface reaction (8.2) was never considered as a dominant process in their model during the warm-up phase, since HCOOH
can still be formed in the gas phase at these temperatures (>40 K) through the reaction
OH + H2 CO.
The experiments presented here are designed to test the possible reaction channels
and not necessarily to simulate a realistic interstellar ice. In particular, they show a low
efficiency of reaction (8.2) under our laboratory conditions, and even though we do not
exclude that reactions (8.1) and (8.2) can occur under interstellar conditions, our experiments show unambiguously the formation of HCOOH through reaction (8.3) in the solid
149
8 Surface formation of HCOOH at low temperature
phase at low temperatures, giving roughly the same branching ratio for the final reaction
products (HCOOH, CO2 + H2 and H2 O + CO). These results are in agreement with density functional theory models, which suggest that there could be equal branching ratio
between the three possible channels. Under interstellar conditions, the HO-CO complex
is expected to be even more stabilized in the H2 O-rich ices, than in our apolar laboratory
analogues. Furthermore, the HO-CO complex channel should be considered an important
HCOOH formation route, since it could explain the presence of HCOOH in dense molecular clouds and at the beginning of the warm-up phase of the protostar in low and high
mass star forming regions.
150
CHAPTER 9
Formation of interstellar solid CO2 after energetic
processing of icy grain mantles1
Space infrared observations with ISO-SWS and Spitzer telescopes have clearly shown that
solid carbon dioxide (CO2 ) is ubiquitous and abundant along the line of sight to quiescent
clouds and star forming regions. Due to the CO2 low gas-phase abundance, it is suggested that CO2 is synthesized on grains after energetic processing of icy mantles and/or
surface reactions. We study quantitatively the abundance of carbon dioxide synthesized
from ice mixtures of astrophysical relevance induced by ion irradiation at low temperature. We compare the CO2 stretching and bending-mode band profiles observed towards
some young stellar objects (YSOs) for which infrared spectra exist. Using a high vacuum
experimental setup, the effects induced by fast ions (30−200 keV) on several ice mixtures
of astrophysical interest are investigated. Chemical and structural modifications of the
ice samples that form new molecular species are analyzed using infrared spectroscopy.
The formation cross section of solid CO2 is estimated from the increase in column density as a function of the dose fitting of experimental data with an exponential curve. Our
laboratory experiments showed that carbon dioxide is formed after irradiation of ice mixtures containing C- and O- bearing molecules. Furthermore, when the same amount of
energy is released into the icy sample, a larger amount of CO2 is formed in H2 O-rich
mixtures in agreement with previous studies. We also found that the CO2 stretching and
bending mode band profiles depend on the mixture and temperature of the ice sample.
We found that the amount of carbon dioxide formed after ion irradiation can account for
the observed carbon dioxide towards YSOs. Furthermore, we discovered that laboratory
spectra are a good spectroscopic analogue of the interstellar features. Even if the comparison between laboratory and observed spectra presented here cannot be considered unique
and complete, our results quantitatively support the hypothesis that interstellar solid CO2
forms after ion irradiation and UV photolysis of icy mantles.
1 Based on: S. Ioppolo, M. E. Palumbo, G. A. Baratta, V. Mennella, 2009, Astronomy & Astrophysics,
volume 493, pages 1017-1028
151
9 Formation of interstellar solid CO2 after energetic processing
9.1 Introduction
An unanswered question concerning the astrophysics of the interstellar medium is the
origin of solid CO2 observed along the line of sight to embedded young stellar objects
(YSOs) and field stars obscured by dense molecular clouds. Solid CO2 absorption bands
are detected in those regions by the Infrared Space Observatory (ISO; e.g., Gerakines
et al. 1999, Gibb et al. 2004, Nummelin et al. 2001) and by the Spitzer Space Telescope
(e.g., Boogert et al. 2004, Pontoppidan et al. 2005, 2008, Whittet et al. 2007). Since there
is a small abundance of gas phase CO2 in the interstellar medium, about a factor of 100
less than in the solid state (van Dishoeck et al. 1996, Boonman et al. 2003), which is in
agreement with chemical models that predict low CO2 production efficiency by gas phase
reactions (Bergin et al. 1995), it is generally assumed that this molecule is synthesized
onto grains. Solid CO2 could be produced in-situ by means of surface reactions (e.g., Tielens & Hagen 1982, Stantcheva & Herbst 2004, Fraser & van Dishoeck 2004), although
this chemical pathway remains controversial. The reaction CO + O → CO2 was found
to have a prohibitively high barrier in at least one study (Grim & D’Hendecourt 1986),
while, according to Roser et al. (2001), it proceeds with a small barrier or even barrierless.
Extensive laboratory studies of these surface reaction schemes are required. Energetic
processing such as UV and ion irradiation of ices containing C- and O- bearing molecules
is another possible formation mechanism. Laboratory experiments demonstrated that CO2
is formed after energetic processing of pure CO ice and icy mixtures containing CO and
H2 O (e.g., D’Hendecourt & Allamandola 1986, Moore et al. 1991, Gerakines et al. 1996,
Ehrenfreund et al. 1997, Palumbo et al. 1998, Watanabe & Kouchi 2002, Loeffler et al.
2005). Interstellar grains are being continuously exposed to energetic processes such as
cosmic ion irradiation and UV photolysis. Fast ions passing through molecular solids release their energy into the target material. As a consequence, many molecular bonds are
broken along the ion-track and, in a short time (less than one picosec), and the molecular
fragments recombine to produce a rearrangement of the chemical structure that leads to
the formation of new molecular species. In the case of UV photolysis, the energy is released into the target material by a single photodissociation or photoexcitation event. In
this particular case, new molecular species are also formed (Strazzulla & Palumbo 2001,
Palumbo 2005). Mennella et al. (2004), Mennella et al. (2006) suggested that in cold dark
clouds, solid CO and CO2 can be formed by energetic processing of carbon grains with
a water ice cap, and they quantitatively evaluated the amount of CO and CO2 formed.
In this scenario, polar CO2 would be produced mainly on carbon grains when water ice
mantles are formed and nonpolar CO2 would be formed when freeze-out of gas-phase
CO takes place. These results represent a different chemical pathway with respect to the
grain-surface reaction routes proposed by Bergin et al. (2005). Both mechanisms could
contribute to the total solid CO2 column density detected along the line of sight to dense
clouds. Nevertheless, a comprehensive quantitative analysis of the amount of CO2 formed
after irradiation of ice mixtures and a direct comparison with observations is still lacking.
We present results of a quantitative analysis of the CO2 abundance synthesized by ion
irradiation of several ice mixtures at low temperature. In particular, we studied ice mixtures containing carbon monoxide, and ice mixtures that do not contain carbon monoxide
152
9.2 Experimental procedure
but comprise C- and O- bearing molecules. These data and those obtained by Mennella
et al. (2004), Mennella et al. (2006) are used to fit the CO2 stretching and bending-mode
band profiles observed towards some YSOs. The fits presented here support quantitatively
the hypothesis that the formation of interstellar solid CO2 after energetic processing of icy
mantles is an efficient mechanism.
Figure 9.1 Schematic top view of the vacuum chamber.
9.2 Experimental procedure
The experiments are conducted at the Laboratory for Experimental Astrophysics, INAFCatania Astrophysical Observatory (Italy). A schematic view of the experimental apparatus is shown in Fig. 9.1. Experiments are carried out in a stainless steel vacuum chamber
with base pressure of about 10−7 mbar. Inside the vacuum chamber, a crystalline silicon
(or KBr) substrate is placed in thermal contact with a cold finger, whose temperature can
be varied within the 10−300 K range. A needle valve is used to admit pure gases (or
mixtures) into the chamber, where they freeze onto the substrate. Ices are monitored using an FTIR spectrometer (Bruker Equinox 55) at a resolution of 1 cm−1 . A He-Ne laser
is used to monitor the thickness of the icy film during deposition. The vacuum chamber
is connected to an ion implanter (200 kV; Danfysik) from which ions of energy up to
200 keV (400 keV for double ionizations) can be obtained. The ion beam produces a spot
on the target larger than the area probed by the IR beam. Current densities range from
100 nA cm−2 up to a few µA cm−2 . In this set-up, IR spectra can be obtained before and
153
9 Formation of interstellar solid CO2 after energetic processing
after ion irradiation without tilting the sample. The icy samples studied are deposited and
irradiated at 12−16 K. Spectra are acquired before and after irradiation at low temperature
and after warm-up at higher temperatures up to 80−90 K. A different procedure is used
for only one pure CO deposit. In this case, the sample is grown and is irradiated at 16 K.
It is then heated to 70 K and irradiated. It is subsequently cooled to 16 K and irradiated
again. This enabled us to study whether the profile of the CO2 bands changes after further
irradiation at higher temperature and after CO sublimation.
At all the examined irradiation doses and temperatures for each sample two spectra, P
and S, are taken. With a polarizer placed in the path of the IR beam (Fig. 9.1), it is possible
to select and analyze the P and S component of the IR beam separately. The P polarized
component has the electric vector parallel to the plane of incidence (the plane of the
figure), while the S polarized component has the electric vector perpendicular to the plane
of incidence. The corresponding spectrum of the background acquired before deposition
is subtracted from all spectra for a given polarization (Baratta & Palumbo 1998). As
shown by Baratta et al. (2000), spectra taken at oblique incidence in S polarization are
equivalent to spectra taken at normal incidence.
In all examined cases, the penetration depth of impinging ions is larger than the thickness of the sample as verified using the TRIM code (Ziegler 2003). The ions used are
H+ or He+ at energies of 200 keV and 30 keV, respectively. Results do not depend on
the ion used but on the deposited energy, i.e. dose (eV/16 amu), which is calculated from
knowledge of the stopping power (energy loss per unit path length) of the ions and measurement of the number of impinging ions per unit of area, fluence (ions cm−2 ). Some
characteristics of the analyzed samples are listed in Table 9.1. In a few instances, the
energy depositedR is not Runiform throughout
the samples. Then the mean stopping power
R
is calculated as S dx/ dx, where dx is the film depth and S is the stopping power
calculated by TRIM. The doses are given in eV/16 amu to compare the results of experiments performed with different mixture concentrations and/or using species of different
molecular weight.
To estimate the abundance of solid molecules in both laboratory samples and along
the line of sight to molecular clouds, it is essential to convert the transmittance spectra
(I f ) into optical depth units τ(ν) = ln(Io /I f ), where Io is the normalization continuum. We
also needed to know the integrated absorbance (cm molecule−1 ) of their IR bands. For the
integrated absorbance, we assumed 20 × 10−17 cm mol−1 (Allamandola et al. 1988) for
the 3 µm H2 O stretching mode, 7.6 × 10−17 cm mol−1 (Yamada & Person 1964) for the
CO2 asymmetric stretching mode, 1.1 × 10−17 cm mol−1 (Gerakines et al. 1995) for the
CO2 bending mode, 1.1 × 10−17 cm mol−1 (Jiang et al. 1975) for the CO stretching mode,
0.64 × 10−17 cm mol−1 (Schutte et al. 1993, Mulas et al. 1998) for the C-H deformation
mode of CH4 , 1.7 × 10−17 cm mol−1 (Lacy et al. 1998) for the umbrella mode of NH3 ,
and 1.3 × 10−17 cm mol−1 (Palumbo et al. 1999)Rfor the C−O stretching mode of CH3 OH.
After fitting a straight baseline, the band area τ(ν)dν is calculated and divided by the
integrated absorbance to derive the column density for each considered species.
154
(a) Sample irradiated at 16 K; heated to 70 K and irradiated; cooled to 16 K and irradiated. (b) The 13 CO2 band is analyzed.
Table 9.1 Characteristics of the studied samples. All ice mixtures are irradiated at 12−16 K and then heated to 80−90 K.
Ices
Energy Ion Thickness Stopping Power
Dose max
(keV)
(nm)
(eV/Å)
(eV/16 amu)
CO
200
H+
220
6.177
18
CO
200
H+
990
6.440
32
CO(a)
200
H+
1120
6.585
40
CO:N2 = 8:1
200
H+
2000
6.800
23
CO:N2 = 1:1
200
H+
2000
6.900
23
CO:N2 = 1:8
200
H+
2000
7.100
27
H2 O:CO = 1:10
200
H+
750
7.311
28
H2 O:13 CO(b) = 6:1
200
H+
160
9.500
4
H2 O:CO = 8:1
30
He+
90
8.842
57
H2 O:CO:N2 = 1:3:3
30
He+
300
7.381
49
N2 :CH4 :CO = 1:1:1
30
He+
300
8.190
54
CO:NH3 = 2:1
30
He+
300
7.925
19
CH3 OH
30
He+
110
10.477
80
CH3 OH:N2 = 1:1
200
H+
250
9.446
21
H2 O:CH4 = 4:1
30
He+
90
9.902
180
H2 O:CH4 = 1:1
30
He+
300
9.078
60
H2 O:CH4 :N2 = 1:1:1
30
He+
300
8.328
50
9.2 Experimental procedure
155
9 Formation of interstellar solid CO2 after energetic processing
Table 9.2 Best-fit parameters for Eq. (9.1) to the experimental data regarding samples
containing CO ice.
Ices
A(a)
σtot (b)
(16 amu/eV)
CO(c)
0.074 ± 0.002 0.242 ± 0.009
CO:N2 = 8:1
0.092 ± 0.002 0.106 ± 0.005
CO:N2 = 1:1
0.162 ± 0.002 0.090 ± 0.002
CO:N2 = 1:8
0.108 ± 0.004 0.203 ± 0.027
H2 O:CO = 1:10
0.121 ± 0.003 0.125 ± 0.009
H2 O:13 CO = 6:1
0.922 ± 0.128 0.107 ± 0.018
H2 O:CO = 8:1
0.567 ± 0.016 0.103 ± 0.013
H2 O:CO:N2 = 1:3:3 0.246 ± 0.004 0.198 ± 0.009
N2 :CH4 :CO = 1:1:1 0.072 ± 0.001 0.105 ± 0.005
CO:NH3 = 2:1
0.131 ± 0.005 0.094 ± 0.007
(a) The asymptotic value for CO2 column density divided by the initial column density of CO before the irradiation. (b) The total cross section. (c) Sample with a thickness of 990 nm (see Table 9.1).
9.3 Results
9.3.1 Irradiation of CO bearing mixtures
The formation of solid carbon dioxide is investigated in different mixtures that contain
CO ices irradiated at low temperature. All samples considered are listed in Table 9.2.
Among the samples containing CO, pure CO ice is discussed first. The most important
product upon irradiation of pure CO ice is solid carbon dioxide. For all samples listed in
Table 9.2, the column density of solid CO2 (NCO2 ) is estimated as a function of irradiation
dose:
0
NCO2 /NCO
= A[1 − e−(σtot
×D)
],
(9.1)
0
where A = A0 /NCO
is the asymptotic value for CO2 column density (A0 ) divided by the
0
initial column density of CO in the sample before the irradiation (NCO
), σtot is the total
cross section and D is the dose. The best-fit functions fitted to the experimental data are
shown in Figs. 9.2, 9.3, and 9.4, while the values found for A and σtot are listed in Table 9.2. It is interesting to note that the column density of CO2 increases rapidly at low
doses and then reaches a saturation value, indicating that a steady state is reached between the formation and destruction mechanisms as discussed by Mennella et al. (2004),
Loeffler et al. (2005). In the case of a pure CO sample, the percentage of CO2 with respect to the initial CO is in agreement with Loeffler et al. (2005). In the top-left panel of
Fig. 9.2, the results obtained from two different samples of thickness 220 nm and 990 nm
are plotted. To evaluate the amount of CO2 formed by energetic processing, the pure CO
156
9.3 Results
Time (10
0
6
4
years)
Time (10
8
CO + 200 keV H
+
12
16
0
at 16 K
0.10
2
CO:N
2
6
4
years)
6
= 8:1 + 200 keV H
+
8
10
12
at 16 K
0.08
i
/ (CO)
0.06
2
0.04
CO
CO
2
/ (CO)
i
0.08
0.06
0.04
experimental data 1
experimental data
experimental data 2
0.02
0.02
fit curve (data 1)
y=A(1-e
fit curve y=A(1-e
- D
)
- D
)
0.00
0.00
0
5
10
15
20
25
30
35
0
5
Time (10
0
CO:N
0.16
2
2
4
= 1:1 + 200 keV H
6
years)
6
+
10
15
20
25
Dose (eV/16 amu)
Dose (eV/16 amu)
Time (10
8
10
0
at 16 K
0.12
CO:N
2
2
4
= 1:8 + 200 keV H
6
years)
6
+
8
10
12
at 16 K
0.10
/ (CO)
i
0.08
0.06
2
0.08
CO
CO
2
/ (CO)
i
0.12
0.04
experimental data
0.04
fit curve y=A(1-e
experimental data
- D
)
fit curve y=A(1-e
- D
)
0.02
0.00
0.00
0
5
10
15
Dose (eV/16 amu)
20
25
0
5
10
15
20
25
30
Dose (eV/16 amu)
Figure 9.2 Column density of CO2 divided by the initial column density of CO before
irradiation for pure CO ice and CO:N2 mixtures. The column density ratio is plotted as a
function of dose. The dose is transferred to interstellar timescales as indicated on the top
x-axis. Experimental data are fitted to an exponential curve (Eq. (9.1)): y = A[1 − e−(σ D) ],
where A is the asymptotic value for CO2 column density divided by the initial column
density of CO, σ is the total cross section, and D is the dose. Only the points indicated
by data 1 (990 nm thick sample) are used for the fit in the top-left panel. Data 2 (220 nm
thick sample) is plotted for comparison.
sample of thickness 990 nm is fitted with Eq. (9.1). A larger amount of CO2 is formed in
H2 O-rich mixtures, comparing the results reported in Figs. 9.2, 9.3, and 9.4, and considering the same amount of energy released to the icy sample, in agreement with previous
studies (e.g., D’Hendecourt & Allamandola 1986, Moore et al. 1991, Ehrenfreund et al.
1997, Palumbo et al. 1998, Watanabe & Kouchi 2002). Figure 9.3, furthermore, plots the
results obtained from four mixtures containing CO and H2 O in different proportions. CO2
molecules formed after ion irradiation in the H2 O:CO = 1:10 mixture represent ∼12% of
the initial CO and this value is close to the percentage obtained after irradiation of pure
CO, while the column density of CO2 in a H2 O:CO = 8:1 mixture is about 60% of the initial CO. In the case of the H2 O:13 CO = 6:1 mixture, the fit of experimental data provides
157
9 Formation of interstellar solid CO2 after energetic processing
Time (10
0
2
6
4
Time (10
years)
6
8
10
0.0
12
0.2
0.4
6
years)
0.6
0.8
1.0
1.2
1.4
0.4
H O:CO = 1:10 + 200 keV H
+
2
13
H O:
at 16 K
2
CO = 6:1 + 200 keV H
+
at 15 K
0.12
i
/ (CO)
0.08
0.2
13
CO
2
CO
2
/ (CO)
i
0.3
0.04
0.1
experimental data
fit curve y=A(1-e
experimental data
- D
fit curve y=A(1-e
)
0.00
- D
)
0.0
0
5
10
15
20
25
30
0.0
0.5
1.0
Time (10
0
4
8
6
years)
12
1.5
2.0
2.5
3.0
3.5
4.0
Dose (eV/16 amu)
Dose (eV/16 amu)
16
20
Time (10
24
0
4
8
6
years)
12
16
20
24
28
0.30
0.7
H O:CO = 8:1 + 30 keV He
+
2
at 15 K
H O:CO:N
2
0.6
2
= 1:3:3 + 30 keV He
+
at 15 K
0.25
0.5
i
/ (CO)
2
0.15
CO
CO
2
/ (CO)
i
0.20
0.4
0.3
0.10
0.2
experimental data
fit curve y=A(1-e
experimental data
- D
)
fit curve y=A(1-e
0.05
0.1
- D
)
0.00
0.0
0
10
20
30
40
Dose (eV/16 amu)
50
60
0
10
20
30
40
50
60
Dose (eV/16 amu)
Figure 9.3 Column density of CO2 divided by the initial column density of CO before
irradiation for H2 O:CO = 1:10, H2 O:13 CO = 6:1, H2 O:CO = 8:1, and H2 O:CO:N2 = 1:3:3
ice mixtures. The column density ratio is plotted as a function of dose. The dose is
transferred to interstellar timescales as indicated on the top x-axis. Experimental data are
fitted to an exponential curve (Eq. (9.1)): y = A[1 − e−(σ D) ], where A is the asymptotic
value for CO2 column density divided by the initial column density of CO, σ is the total
cross section, and D is the dose.
higher A value (0.92) than in the other considered cases. We note that this is unrealistic.
The experimental data points are far from the saturation value and follow an almost linear
trend, since irradiation is halted after a dose of ∼4 eV/16 amu, which is lower than those
used for the other considered samples. After a dose of ∼30 eV/16 amu, the CO2 column
density in the H2 O:CO:N2 = 1:3:3 mixture decreases because CO2 molecules, like other
species (Baratta et al. 2002), are destroyed after their formation by ion irradiation. We
therefore decided to fit the data up till the dose of ∼20 eV/16 amu, which corresponds to
the maximum CO2 formation level reached in this experiment.
158
9.3 Results
Time (10
0
0.08
5
6
10
15
N :CH :CO = 1:1:1 + 30 keV He
2
4
Time (10
years)
20
+
0
25
at 12 K
2
CO:NH
3
6
years)
4
6
= 2:1 + 30 keV He
+
8
10
12
at 12 K
0.12
/ (CO)
0.09
CO
2
0.04
CO
2
/ (CO)
i
i
0.06
0.06
experimental data
0.02
fit curve y=A(1-e
experimental data
0.03
- D
)
fit curve y=A(1-e
- D
)
0.00
0.00
0
10
20
30
40
50
0
60
5
10
15
20
25
Dose (eV/16 amu)
Dose (eV/16 amu)
Figure 9.4 Column density of CO2 divided by the initial column density of CO before irradiation for N2 O:CH4 :CO = 1:1:1 and CO:NH3 = 2:1 ice mixtures. The column density
ratio is plotted as a function of dose. The dose is transferred to interstellar timescales as indicated on the top x-axis. Experimental data are fitted to an exponential curve (Eq. (9.1)):
y = A[1 − e−(σ D) ], where A is the asymptotic value for CO2 column density divided by
the initial column density of CO, σ is the total cross section, and D is the dose.
9.3.2 Irradiation of ice mixtures without CO
Solid CO and CO2 are the foremost molecules formed at low temperature after processing
of C- and O- bearing ices, which do not contain CO ice initially.
The effects of ion irradiation in a sample of pure solid CH3 OH and in a mixture of
H2 O and CH4 with a ratio 4:1 are quantified. Notice that it is impossible to reproduce
the trend of the CO and CO2 column density with one exponential curve. For both of
the aforementioned mixtures, the experimental column densities of formed CO and CO2
divided by the initial column densities of CH3 OH and CH4 , respectively, are fitted by two
exponential curves with a threshold dose (D0 ) to take into account the change in the CO
and CO2 production rate. In the case of CO2 and for D≤D0 :
1
NCO2 /NX0 = A1 [1 − e−(σtot
× D)
],
(9.2)
while for D>D0 :
1
NCO2 /NX0 = A1 [1 − e−(σtot
× D)
2
] + A2 [1 − e−[σtot
× (D − D0 )]
],
(9.3)
where A1 and A2 are the asymptotic values of CO2 column density divided by the initial
column density of CH3 OH and CH4 (NX0 ), for the deposit of pure solid CH3 OH, and, for
the 4:1 mixture of H2 O and CH4 respectively, σ1tot and σ2tot are the total cross sections, D is
the dose and D0 is the threshold dose. Equations (9.2) and (9.3) are also used to quantify
the CO column density obtained in both of the aforementioned mixtures by energetic
processing at low temperature. The best-fits to experimental data for CO and CO2 column
density divided by NX0 as a function of the dose are shown in Fig. 9.5, while the asymptotic
159
9 Formation of interstellar solid CO2 after energetic processing
values of CO and CO2 column density divided by NX0 and the total cross sections are listed
in Tables 9.3 and 9.4 respectively.
Time (10
0
fit curve y=A (1-e
-
-
D
-
(D-D )
1
y=A (1-e
24
D
)+A (1-e
1
y=A (1-e
Time (10
years)
16
0.4
0.3
6
8
-
0
32
(D-D )
fit curve y=A (1-e
)
0
2
0.3
y=A (1-e
D
-
(D-D )
4
CO / (CH )
3
0.2
)+A (1-e
-
60
(D-D )
0
2
)
)
0
experimental data
i
experimental data
D
)
2
i
2
CO / (CH OH)
-
1
y=A (1-e
)
0
years)
40
-
1
)
6
20
0.2
0.1
0.1
0.0
CH OH + 30 keV He
+
3
0
20
40
60
80
0.0
H O:CH
at 12 K
2
0
100
Time (10
1
y=A (1-e
y=A (1-e
0.06
i
2
Time (10
years)
D
24
)+A (1-e
-
0
32
(D-D )
2
0
D
-
(D-D )
y=A (1-e
-
D
-
(D-D )
1
0
at 12 K
200
y=A (1-e
)
i
)+A (1-e
-
60
(D-D )
2
0
)
0
)
experimental data
0.3
3
4
/ (CH )
D
)
2
experimental data
years)
40
-
1
)
6
20
fit curve y=A (1-e
)
0.4
-
1
/ (CH OH)
-
+
150
Dose (eV/16 amu)
16
fit curve y=A (1-e
0.08
6
8
= 4:1 + 30 keV He
100
Dose (eV/16 amu)
0
4
50
CO
2
CO
2
0.04
0.02
0.2
0.1
0.00
CH OH + 30 keV He
3
0
20
40
60
Dose (eV/16 amu)
80
+
at 12 K
100
0.0
H O:CH
2
0
50
4
= 4:1 + 30 keV He
100
150
+
at 12 K
200
Dose (eV/16 amu)
Figure 9.5 Column density of CO and CO2 divided by the initial column density of
CH3 OH and CH4 for the deposit of pure CH3 OH ice and of H2 O:CH4 = 4:1 mixture, respectively. CO and CO2 are formed after ion irradiation of icy samples containing C- and O- bearing molecules. The column density ratios are plotted as a function of dose. The dose is transferred to interstellar timescales as indicated on the
top x-axis. Experimental data are fitted to two exponential curves with a threshold
dose (Eqs. (9.2) and (9.3)). For D≤D0 : y = A1 [1 − e−(σ1 × D) ], while for D>D0 :
y = A1 [1 − e−(σ1 × D) ] + A2 [1 − e−[σ2 × (D − D0 )] ], where A1 and A2 are the asymptotic
values for CO and CO2 column density divided by the initial column density of CH3 OH
and CH4 , σ1 and σ2 are the total cross sections, D is the dose, and D0 is the threshold
dose.
During irradiation, the column density of methanol, water, and methane decreases,
while the column density of newly formed species, such as CO, CO2 , and H2 CO, increases. We note that at low doses the trend of CO and CO2 production differs from that
at higher doses. Furthermore, at low doses, the main molecules formed are hydrocarbons
(Baratta et al. 2002), whose column density decreases at higher doses as the CO and CO2
160
9.3 Results
column densities increase. First and second generation molecules therefore exist. The
presence of newly formed solid CO in the mixture supports the CO2 synthesis, because
reactions with CO as a precursor, such as CO + OH, must be important. In addition, CO2
could be formed by reactions that do not have CO as a precursor.
9.3.3 Carbon grains with a water ice cap
According to observations, the ratio of solid CO2 to H2 O abundance is almost constant
(varying from 15% to 40%) toward field stars as well as embedded objects, while the
abundance of solid CO with respect to water ice is strongly cloud-density dependent (e.g.,
Gerakines et al. 1999, Whittet et al. 1998, Nummelin et al. 2001, Bergin et al. 2005, Pontoppidan et al. 2008). This suggests that solid water and carbon dioxide may be formed
under similar conditions. During the continuous cycling of dust between diffuse and
dense regions in the interstellar medium, an H2 O-rich ice layer covers hydrogenated carbon grains as they enter in dense clouds. A standard cosmic-ray field and UV field induced
by cosmic-ray fluorescence of molecular hydrogen can process carbon grains covered by
mantles of water ice, leading to the formation of new solid species. Laboratory studies
show that formation of CO and CO2 molecules occurs when hydrogenated carbon grains
with a water ice cap are irradiated at low temperature. The experimental results for ion
irradiation and UV photolysis of H2 O ice covered carbon grains are described by Mennella et al. (2004), Mennella et al. (2006) and represent an alternative efficient formation
route with respect to that proposed by Bergin et al. (2005). Nevertheless, we emphasize
that both energetic processing and surface reactions are not mutually exclusive and could
rather contribute to the total solid CO2 observed along the line of sight to dense clouds.
Mennella et al. (2004), Mennella et al. (2006) derived the formation cross section of
CO and CO2 molecules from the increase in intensity of the CO and CO2 stretching modes
as a function of the ion and UV fluences, and estimated that after t = 3 × 107 yr, column
densities are:
NCO = 7.0 × 1015 AV ,
(9.4)
and
NCO2 = 9.3 × 1015 AV ,
(9.5)
where AV is the visual extinction of the cloud.
These laboratory data will be added to our sample and discussed in the following
sections.
161
162
(a) The threshold dose for the exponential curves. (b) The asymptotic values for CO2 column density divided by the initial column density of CH3 OH and CH4 before
irradiation. (c) The total cross sections.
Table 9.4 Best-fit parameters for Eqs. (9.2) and (9.3) to the experimental CO2 column density regarding samples without CO.
Ices
D0 (CO2 )(a)
A1 (CO2 )(b)
σ1tot (CO2 )(c)
A2 (CO2 )(b)
σ2tot (CO2 )(c)
(eV/16 amu)
(16 amu/eV)
(16 amu/eV)
CH3 OH
7.6 ± 1.1
0.008 ± 0.006 0.126 ± 0.126 0.049 ± 0.006 0.051 ± 0.007
H2 O:CH4 = 4:1
19.2 ± 3.0
0.211 ± 0.033 0.009 ± 0.002 0.327 ± 0.033 0.009 ± 0.001
(a) The threshold dose for the exponential curves. (b) The asymptotic values for CO column density divided by the initial column density of CH3 OH and CH4 before
irradiation. (c) The total cross sections.
Table 9.3 Best-fit parameters for Eqs. (9.2) and (9.3) to the experimental CO column density regarding samples without CO.
Ices
D0 (CO)(a)
A1 (CO)(b)
σ1tot (CO)(c)
A2 (CO)(b)
σ2tot (CO)(c)
(eV/16 amu)
(16 amu/eV)
(16 amu/eV)
CH3 OH
5.5 ± 0.3
0.048 ± 0.004 0.150 ± 0.150 0.174 ± 0.004 0.101 ± 0.004
H2 O:CH4 = 4:1
10.8 ± 2.9
0.021 ± 0.011 0.376 ± 0.376 0.233 ± 0.013 0.017 ± 0.002
9 Formation of interstellar solid CO2 after energetic processing
9.3 Results
CO:N =8:1
Pure CO
CO:N =1:1
2
CO:N =1:8
2
2
3
T=16 K
2
1
0
Optical Depth
0.8
T=40 K
0.4
0.0
0.8
T=60 K
0.4
0.0
0.6
T=80 K
0.3
0.0
2385
2340
2385
2340
2385
2340
2385
2340
-1
Wavenumber (cm )
Figure 9.6 The band profile of the CO2 stretching mode at about 2340 cm−1 formed after
ion irradiation of four different mixtures: (from the left to the right side) pure CO ice,
CO:N2 = 8:1, CO:N2 = 1:1, and CO:N2 = 1:8. The top side of the figure shows the
spectrum taken for each experiment after ion irradiation at 16 K. The following rows show
the spectra taken after irradiation at 16 K and heated at 40, 60, and 80 K, respectively.
9.3.4 Band profiles
In Figs. 9.6 and 9.7, we present the band profile of the CO2 stretching and bending mode
(at about 2340 cm−1 and 660 cm−1 respectively) formed after ion irradiation of four different mixtures. From the left hand side to the right hand side of Figs. 9.6 and 9.7, the case
of pure CO ice, CO:N2 = 8:1, CO:N2 = 1:1, and CO:N2 = 1:8 are shown. From top to
bottom of Figs. 9.6 and 9.7, the spectra taken at 16, 40, 60, and 80 K for each experiment,
after irradiation at 16 K, are shown. Comparing the band profile of the CO2 stretching
mode for the spectra taken at 16 K with those taken at 80 K it becomes apparent that,
after heating the sample, the feature becomes broad and asymmetric. Some differences
in peak positions and band widths can also be discerned by comparing the spectra of dif163
9 Formation of interstellar solid CO2 after energetic processing
CO:N =8:1
Pure CO
CO:N =1:1
2
CO:N =1:8
2
2
0.4
T=16 K
0.2
0.0
Optical Depth
0.2
T=40 K
0.1
0.0
0.2
T=60 K
0.1
0.0
0.2
T=80 K
0.1
0.0
680
660
640
680
660
640
680
660
640
680
660
640
-1
Wavenumber (cm )
Figure 9.7 Similar to Fig. 6 for the CO2 bending mode at about 660 cm−1 .
ferent samples taken at the same temperature. The bending mode appears more sensitive
to temperature variations than the stretching mode. If the sample is heated, the bending
mode feature changes drastically: a double peak appears at about 655 cm−1 and 660 cm−1
and the band becomes broader. The double peak shifts in position and changes in intensity for different N2 abundance in the mixtures, until it effectively disappears in the
spectra for the CO:N2 = 1:8 mixture. At all the investigated temperatures, the spectra of
the CO:N2 = 1:8 sample present a peak due to the KBr substrate at about 670 cm−1 . The
infrared spectra show that, due to the different volatility, after irradiation, CO sublimes
at about 30 K, while CO2 remains in the solid phase up to about 90 K. The band profile
of the CO2 bending mode also depends on the N2 concentration. In general, the results
presented here agree with the study of the CO2 band profile by Ehrenfreund et al. (1997).
Figures 9.8 and 9.9 show the band profile of the CO2 stretching and bending mode
formed after energetic processing at low temperature of other samples. From top to bottom of Figs. 9.8 and 9.9, we present the spectrum of CH3 OH ice irradiated at 16 K, a
mixture of H2 O:CO = 1:10 irradiated at 16 K and heated to 90 K, and irradiated water ice
deposited on hydrogenated amorphous carbon grains called ACARL H (Mennella et al.
164
9.4 Comparison with observations
Stretching mode of CO
2
1.0
CH OH irradiated
3
0.5
at 12 K
0.0
Optical Depth
0.18
H O:CO (1:10) irradiated at 16 K
2
0.09
and warmed up to 90 K
0.00
0.04
H O+ACARL_H irradiated
2
0.02
at 12 K
0.00
0.8
0.4
Pure CO
2
at 15 K
0.0
2370
2360
2350
2340
2330
-1
Wavenumber (cm )
Figure 9.8 The band profile of the CO2 stretching mode at about 2340 cm−1 formed after
energetic processing at low temperature of some samples: (from top to bottom) CH3 OH
ice irradiated at 12 K, H2 O:CO = 1:10 irradiated at 16 K and heated to 90 K, water ice
deposited on hydrogenated amorphous carbon grains ACARL H (Mennella et al. 2006)
and irradiated at 12 K. The band profile of the stretching mode of pure CO2 ice at 15 K is
also shown for comparison.
2006). The stretching and bending mode band profiles of pure CO2 ice are also shown for
comparison. It is interesting to note that, although the peak position changes in each spectrum for both the stretching and bending modes, the bending mode appears more sensitive
to mixture variations.
9.4 Comparison with observations
9.4.1 CO2 towards massive YSOs
Along the line of sight to embedded young stellar objects (YSO), icy mantles could have
been exposed to energetic processing and warming. The observed band profiles due to
solid CO2 indicate the presence of different components along the line of sight. We considered the following YSOs: S140 IRS 1, AFGL 2136, NGC 7538 IRS 9, NGC 7538 IRS 1,
and W33A (Gerakines et al. 1999). A list of the considered young infrared source param165
9 Formation of interstellar solid CO2 after energetic processing
Bending mode of CO
2
0.03
CH OH irradiated
3
at 12 K
Optical Depth
0.00
0.02
H O:CO (1:10) irradiated at 16 K
2
and warmed up to 90 K
0.00
0.003
H O+ACARL_H irradiated
2
at 12 K
0.000
Pure CO
0.1
2
at 15 K
0.0
680
660
640
-1
Wavenumber (cm )
Figure 9.9 Similar to Fig. 8 for the CO2 bending mode at about 660 cm−1 .
eters is reported in Table 9.5.
S140 is an H II region of both high and low mass star formation located 0.9 kpc behind about 23 mag of visual extinction. AFGL 2136 is associated with a bipolar reflection
nebula. Along most of the line of sight to AFGL 2136, the dust and the gas temperature is about 30 K. NGC 7538 IRS 9 is an embedded, luminous, and compact infrared
source. NGC 7538 IRS 1 is a pre-main-sequence object, which is the most luminous in
the NGC 7538 complex. W33A is a massive, luminous source. Along the line of sight to
this source, there is evidence for energetic processing provided by the presence of abundant XCN (Gibb & Whittet 2002). In all the considered sources water, carbon monoxide,
carbon dioxide and methanol are detected in the solid phase (Gibb et al. 2004).
9.4.2 The fitting procedure
Gerakines et al. (1999) considered three categories of laboratory mixtures in attempting
to reproduce qualitatively the CO2 stretching and bending mode of the observed, aforementioned spectral features: polar, nonpolar, and annealed ices. In the cases of polar and
nonpolar ices, they applied particle shape corrections to the laboratory data using different
grain models derived from the real and imaginary parts of the ice’s refractive index (n and
k values).
In this chapter, we have attempted to analyze quantitatively the formation of CO2
166
9.4 Comparison with observations
Table 9.5 Young infrared source parameters: distance (d), luminosity (l), visual extinction
(AV ), and cold component of CO column density detected in gas phase along the line of
sight to some YSOs (Ngas(cold)CO), solid CO2 column density observed along the line of
sight to some YSOs (N solid CO2 ).
Sources
S140 IRS 1
AFGL 2136
NGC 7538 IRS 9
NGC 7538 IRS 1
W33A
d
(kpc)
0.9
2
2.7
2.8
4
l
(L )
2000
7 × 104
4 × 104
1.3 × 105
1.1 × 105
AV
(mag)
23
94
84
119
150
Ngas(cold) CO
(1019 mol cm−2 )
0.45
1.08
1.44
1.30
1.96
N solid CO2
(1017 mol cm−2 )
4.2(0.1)
7.8(0.3)
16.3(1.8)
5.1(0.2)
14.5(1.3)
reference
1, 4
1, 3
1, 2, 3
1, 4
1, 3
(1) Gibb et al. (2004). (2) Chiar et al. (1998). (3) Tielens et al. (1991). (4) Mitchell et al. (1990).
in dense clouds. We have therefore compared the band profiles of CO2 stretching and
bending mode, observed towards embedded YSOs, with laboratory spectra of selected
ion-irradiated samples, to assess the effect of energetic processing on interstellar icy mantles.
According to Baratta et al. (2000, 2002), the P polarized component of the spectrum
can show additional features that are not due to absorption (k) but to the increased reflectivity of the region across the absorption band where n<1. The difference between the
P and S polarized components of the spectra depends on the optical constants of the ice.
Additionally, for a given sample, it depends on the thickness, being higher for thinner
films. Baratta et al. (2000) also demonstrated that if P and S polarized components differ,
then band profiles are unrepresentative of small particle extinction spectra and cannot be
compared with observed interstellar spectra. Figure 9.10 shows the band profile of the
stretching and bending mode for CO2 formed after irradiation of the CO ice, indicating
that the P and S components are similar. We verified that this is true for all band profiles
used in the fits. Given that the signal-to-noise ratio of P spectra is higher than for S spectra, the former are used to fit the observed data. Since P and S spectra are similar, small
particle extinction cross section calculations are in our case unnecessary. For this reason,
we used the laboratory spectra in optical depth units for the fitting procedure.
Looking at the observed interstellar spectra, the 4.27 µm feature due to the CO2
stretching mode is clearly saturated for NGC 7538 IRS 9 and W33A (Fig. 9.11). AFGL
2136 exhibits a shallow shoulder on the long wavelength side of the 4.27 µm feature. It
is suspected that this is the result of an unidentified, broad, underlying component (Gerakines et al. 1999). The shallow absorption shoulders on the short and long wavelength
sides of the stretching mode are due to the unresolved P and R branches of gas-phase CO2 ,
which are consistent with the strong gas-phase CO2 absorptions detected at 14.97 µm toward these sources. Spurious structures also exist in the troughs of the 4.27 µm feature of
NGC 7538 IRS 1, which may be close to saturation. The observed CO2 15.2 µm bending
mode for all the sources can be divided into three components: a feature at 14.97 µm
due to gas-phase CO2 absorption, a pair of sharp peaks at about 15.15 and 15.27 µm that
highlight CO2 segregation induced by thermal processing along the line of sight, and a
shoulder close to 15.4 µm assigned to the acid-base interaction between CO2 and alcohols
167
9 Formation of interstellar solid CO2 after energetic processing
Bending mode of CO
Stretching mode of CO
2
2
Optical Depth (a.u.)
1.0
P
P
S
S
0.5
0.0
2380
2360
2340
2320
680
660
640
-1
Wavenumber (cm )
Figure 9.10 The band profile of the CO2 stretching and bending modes formed after irradiation at 16 K and heating to 80 K of pure CO ice. The comparison between the P
(black line) and S (light gray line) component of the IR beam is plotted, see text in § 9.2.
(Ehrenfreund et al. 1998, Dartois et al. 1999).
Using home-written software, all spectra for samples listed in Table 9.1 are compared
systematically with the observed YSOs data. The software selected spectra of laboratory
mixtures that fitted simultaneously the CO2 bending and stretching mode band profile of
the source considered. The best results are reported in Fig. 9.11.
We note that three or four components are needed to fit simultaneously the bending
and stretching mode band profile of each object. The fit components and percentages are
shown in Table 9.6. The spectra used to fit the observed CO2 band profiles are available
in the Catania database (http://www.oact.inaf.it/weblab/).
The profile of CO2 formed after irradiation of a pure CO or a H2 O:CO = 1:10 mixture
heated to 70 K is required to reproduce the sharp peaks of the bending mode profile at
15.15 and 15.27 µm. The profile of CO2 formed after irradiation of methanol is required
to fit the shoulder at 15.4 µm. This component is due to an acid-base interaction between
the C atom of CO2 and the oxygen atom of a polar molecule. When CO2 molecules are in
the presence of alcohol, they act as a Lewis acid (Ehrenfreund et al. 1998, Dartois et al.
1999). The most abundant alcohol detected in dense clouds is methanol. For this reason,
we considered a sample of methanol deposited and irradiated at low temperature. The
profile of CO2 bending and stretching modes formed after irradiation of H2 O on carbon
grains are broad and smooth (Figs. 9.8 and 9.9), and these are also required to fit the
observed bands.
168
9.4 Comparison with observations
Table 9.6 Contributions in percentage for each listed fit component to the total CO2 column density observed along the line of sight to the considered sources as obtained from
the fitting procedure (Fit columns) and as estimated from Eqs. (9.5), (9.6), (9.7), and (9.8)
which give the maximum possible contribution by each component to the total observed
CO2 (Eq. columns).
Spectra
CO(a)
CO(b)
H2 O:CO(c) = 1:10
CO:NH3 (d) = 2:1
CH3 OH(d)
ACARL H(d)
S140 IRS 1
Fit
Eq.
(%)
(%)
38
71
33
71
28
100
1
83
-
AFGL 2136
Fit
Eq.
(%)
(%)
66
100
19
78
15
100
NGC 7538 IRS 9
Fit
Eq.
(%)
(%)
41
59
44
100
15
43
-
NGC 7538 IRS 1
Fit
Eq.
(%)
(%)
72
100
2
2
20
92
6
100
W33A
Fit
Eq.
(%)
(%)
23
100
44
100
33
96
[Eq.]
[6]
[6]
[6]
[8]
[7]
[5]
(a) Irradiated at 16 K; heated to 70 K and irradiated. (b) Irradiated at 16 K; heated to 70 K and irradiated; cooled
to 16 K and irradiated. (c) Irradiated at 16 K and heated to 90 K. (d) Irradiated at 12 K.
In the case of NGC 7538 IRS 1, the fit improves if we consider a small contribution
(2%) from CO2 formed after irradiation of a CO:NH3 = 2:1 mixture. An upper limit to
the abundance of ammonia (<3.7 × 1017 cm−2 ) is indeed derived along the line of sight
to NGC 7538 IRS 1 (Gibb et al. 2001).
Finally, it is noted that the considered sample of laboratory spectra is by no means
complete and should be regarded only as a starting point for the analysis of the composition of the icy grain mantles. Future studies of the composition of the CO2 ices should
include other laboratory samples of a range of ion and UV irradiation doses, temperatures,
and mixture ratios.
From Fig. 9.11, it is clear that while the bending mode profile is always well reproduced by laboratory spectra, the fit of the stretching mode is not always satisfactory. This
could be due to the fact that the optical depth of the stretching mode band is very large
and our approach could be no longer applicable.
Pontoppidan et al. (2008) performed a fitting analysis of the CO2 ice bending mode
profile for a large sample of low mass embedded young stars. They used five components to reproduce the profile of the considered feature. A CO2 :H2 O ∼ 1:7 mixture and a
CO2 :CO ∼ 1:1 mixture, which are named the red and blue component respectively, generally dominate the bending mode profile and the total CO2 column density observed.
The other three components are needed to reproduce subtle differences due to trace constituents. All components considered by Pontoppidan et al. (2008) correspond to interstellar relevant ice analogues and can be divided into polar and nonpolar ices. Our fitting
results presented here are similar to those of Pontoppidan et al. (2008), although we emphasize that our fitting components are obtained after energetic processing of interstellar
ice analogues and both stretching and bending modes are fitted simultaneously without
any particle-shape correction.
169
9 Formation of interstellar solid CO2 after energetic processing
Wavelength (
4.2
1.2
4.24
4.28
4.32
m)
14.85
15.4
15.95
S140 IRS1
0.2
0.6
0.1
0.0
0.0
AFGL 2136
0.3
2
0.2
1
0.1
0
0.0
Optical Depth
6
0.8
NGC 7538
IRS9
3
0.4
0
2
0.0
NGC 7538
0.2
IRS1
1
0.1
0
0.0
8
0.6
W33A
0.3
4
0.0
0
2380
2360
2340
2320
680
660
640
620
-1
Wavenumber (cm )
Figure 9.11 The CO2 stretching and bending mode band profile of selected YSO sources
are fitted (thick black line) at the same time using different laboratory spectra: CO ice irradiated at 16 K, heated to 70 K and irradiated (light gray line); CO ice irradiated at 16 K,
heated to 70 K and irradiated, cooled to 16 K and irradiated (gray line); H2 O:CO = 1:10
mixture irradiated at 16 K and heated to 90 K (black line); CO:NH3 = 2:1 irradiated at
12 K (dash-dot line); methanol ice irradiated at 12 K (dash line); water ice on hydrogenated carbon grains irradiated at 12 K (dot line). The spectra used to fit the observed
CO2 band profiles are available in the Catania database (http://www.oact.inaf.it/weblab/).
170
9.5 Discussion
9.5 Discussion
Deriving a reliable fit is insufficient to prove the validity of a model; reasonable physical
arguments are also required. We have performed a quantitative study of the CO2 formed
in laboratory C- and O- bearing samples upon energetic processing, and we extended our
results to the interstellar medium by considering the lifetime in solid phase of species
irradiated in dense cores, and the temperature gradient along the line of sight.
In Figs. 9.2, 9.3, 9.4, and 9.5 the top ordinate axes indicate an estimation of the time
(years) required to achieve the same effect on interstellar ices as observed in the laboratory. To estimate this time, we considered the approximation of effective monoenergetic
1 MeV protons assuming that in dense interstellar regions the effective 1 MeV proton flux
is equal to 1 proton cm−2 s−1 (Mennella et al. 2003). The laboratory results are extrapolated to the interstellar medium conditions by deriving the formation cross section due
to 1 MeV protons from that obtained in the irradiation experiments with 30 keV He+ and
200 keV H+ using the ratio of the corresponding stopping power as a scaling factor.
As reported by Greenberg (1982), a dense cloud lifetime ranges between 3 × 107
and 5 × 108 yr. Assuming a density of n0 ∼ 104 cm−3 , the gas takes 109 /n0 ' 105 yr
to condense onto grains (Tielens & Allamandola 1987a). Icy grain mantles therefore
undergo cosmic ion irradiation for about 105 −108 yr. The first estimate refers to the case
of icy mantles that sublime as soon as they form (which could be the case for volatile
species such as CO), the latter estimate refers to the limit case of icy mantles that survive
throughout the cloud lifetime (which could be the case for less volatile species such as
H2 O). The lifetime of molecules in the solid phase is related to the volatility of each
species. The results of the fits discussed in § 9.4 and summarized in Table 9.6, indicate
that a significant percentage of the CO2 band profiles is reproduced by the spectra of
CO2 formed after ion irradiation of pure CO or H2 O:CO = 1:10 ice mixtures. To justify
this result, we should assume that all the CO molecules that we observe today in the gas
phase along the line of sight to YSOs are frozen onto grains during the cloud collapse
phase and have been processed by ion irradiation for about 8 × 106 yr. Then, as indicated
by laboratory experiments, when the temperature of the grains increases, CO molecules
sublimate, while CO2 molecules formed after ion irradiation remain. Thus, the profile of
the bending mode shows the pair of sharp peaks required to fit the observed band profile.
To estimate the amount of solid CO2 that can form after ion irradiation of CO-rich icy
mantles, we therefore use the following equation:
NCO2 = N(CO)Obs,gas × [N(CO2 )/N(CO)]Lab, solid ,
(9.6)
where N(CO)Obs, gas is the CO column density detected in gas phase along the line of
sight to the considered sources, while [N(CO2 )/N(CO)]Lab, solid is the ratio of the CO2 to
CO column density obtained in laboratory spectra.
Solid CO is also observed along the line of sight to YSOs and we then expect that
the spectrum of solid CO2 at low temperatures, formed after ion irradiation of pure CO,
should also be considered in the fit of the CO2 observed band profile. As already discussed
by Loeffler et al. (2005), this component accounts for 1−6% of the observed solid CO2 .
We included this component in the fitting procedure and found that the quality of the fit
171
9 Formation of interstellar solid CO2 after energetic processing
is equivalent to that obtained by considering the mixtures listed in Table 9.6 and shown in
Fig. 9.11. This result agrees with the result obtained by Pontoppidan et al. (2008), who
found that the contribution of the CO2 :CO ∼ 1:25 mixture is almost negligible.
To estimate the amount of solid CO2 formed after ion irradiation of methanol, we used
the following equation:
NCO2 = N(CH3 OH)Obs, solid × [N(CO2 )/N(CH3 OH)]Lab, solid ,
(9.7)
where the time considered is the lifetime of dense clouds (3 × 107 yr), N(CH3 OH)Obs, solid
is the methanol detected in solid phase along the line of sight to the considered sources,
and [N(CO2 )/N(CH3 OH)]Lab, solid is the ratio of the CO2 to CH3 OH column density measured in the laboratory spectrum used for the fit. We point out that this ratio is not the
CO2 /(CH3 OH)i ratio reported in Fig. 9.5.
To estimate the amount of CO2 formed after ion irradiation of CO:NH3 = 2:1 mixture
we used the following equation:
NCO2 = N(NH3 )Obs, solid × [N(CO2 )/N(NH3 )]Lab, solid ,
(9.8)
where N(NH3 )Obs, solid is the column density of NH3 observed along the line of sight
to the sources and [N(CO2 )/N(NH3 )]Lab, solid is the CO2 and NH3 column density ratio
measured in the laboratory spectrum used for the fit. In this calculation, we assumed that
all solid ammonia observed along the line of sight is mixed with CO.
Following Mennella et al. (2006), and considering clouds for which visual extinction
(AV ) is known, we used Eq. (9.5) to evaluated the contribution of CO2 , which is produced
by energetic irradiation of carbon grains with a water ice cap, to the observed column
density.
In Table 9.6, the percentages of the CO2 column density calculated by Eqs. (9.5),
(9.6), (9.7), and (9.8) are compared to the percentages obtained by fits for all young infrared sources considered. These percentages indicated by “Eq” represent the maximum
possible contribution by each component. It is relevant to note that the percentages for
the fit components of each YSOs considered agree with the percentages of the CO2 column density calculated by Eqs. (9.5), (9.6), (9.7), and (9.8). In all cases analyzed, the
percentage values for the fit components are indeed lower than the values derived from
Eqs. (9.5), (9.6), (9.7), and (9.8). Even though the fits presented here are not unique, they
are supported by reasonable astrophysical hypotheses as discussed above.
By assuming that in dense clouds all CO molecules detected in the gas phase (COtot )
are frozen onto grains during the cloud collapse phase, it is possible to evaluate, using
Table 9.7, the contribution, due to irradiation, of pure CO ice to the observed CO2 column
density. For the other ice mixtures listed in Table 9.7, COin is the amount of interstellar
solid CO originally present in icy grain mantles and mixed with other species. For those
lines of sight along which visual extinction are known, it is possible to calculate the
contribution to the solid CO2 due to irradiation of carbon particles covered by amorphous
water ice.
172
9.6 Conclusions
Table 9.7 Contribution of CO2 produced by energetic irradiation of laboratory samples to
the observed column density for those clouds of which the visual extinction and the CO
gas-phase abundance are known.
Icy samples
NCO2 (upper limits)
(mol cm−2 )
CO
0.07 × COtot
CO:N2 = 8:1
0.09 × COin
CO:N2 = 1:1
0.16 × COin
CO:N2 = 1:8
0.11 × COin
H2 O:CO = 1:10
0.12 × COin
H2 O:CO = 8:1
0.57 × COin
H2 O:CO:N2 = 1:3:3 0.25 × COin
N2 :CH4 :CO = 1:1:1 0.07 × COin
CO:NH3 = 2:1
0.13 × COin
9.3 × 1015 AV
ACARL H
9.6 Conclusions
Abundant amount of solid CO2 is detected towards embedded YSOs (both low mass and
high mass protostars) and field stars. Observations towards high mass star-forming regions indicate that some of the observed carbon dioxide is segregated (Ehrenfreund et al.
1998). On the other hand, observations along the line of sight to field stars indicate that
most solid CO2 is mixed with water ice (Bergin et al. 2005). As shown in this article,
CO and CO2 are formed easily after energetic processing of ice mixtures containing Cand O- bearing molecules and carbon grains covered by water ice. Furthermore, given
the same amount of energy released to the icy sample, a larger amount of CO2 is formed
in H2 O-rich mixtures. It has been found that the band profile of the CO2 stretching and
bending modes depends on the mixture and temperature of the ice sample. On the basis of the present laboratory results, it is possible to estimate the contribution of CO2 ,
produced after energetic processing, to the observed carbon dioxide column densities for
several YSOs. Laboratory results presented account not only quantitatively for the column density of observed interstellar CO2 but also provide a good spectroscopic analogue
of the interstellar features supporting the hypothesis that interstellar solid CO2 is formed
after ion irradiation and UV photolysis of icy mantles. This however does not exclude the
possibility that other formation routes, such as grain surface reactions, contribute to the
production of the observed interstellar solid CO2 .
173
Bibliography
Acharyya, K., Fuchs, G. W., Fraser, H. J., van Dishoeck, E. F., & Linnartz, H. 2007,
A&A, 466, 1005
Acquista, N., Schoen, L. J., & Lide, Jr., D. R. 1968, J. Chem. Phys., 48, 1534
Al-Halabi, A. & van Dishoeck, E. F. 2007, MNRAS, 382, 1648
Allamandola, L. J., Sandford, S. A., & Valero, G. J. 1988, Icarus, 76, 225
Allen, M. & Robinson, G. W. 1977, ApJ, 212, 396
Andersson, S., Al-Halabi, A., Kroes, G., & van Dishoeck, E. F. 2006, J. Chem. Phys.,
124, 064715
Atkinson, R., Baulch, D. L., Cox, R. A., et al. 2004, Atmospheric Chemistry & Physics,
4, 1461
Awad, Z., Chigai, T., Kimura, Y., Shalabiea, O. M., & Yamamoto, T. 2005, ApJ, 626, 262
Bally, J. 2007, Ap&SS, 311, 15
Bandow, H. & Akimoto, H. 1985, J. Phys. Chem., 89, 845
Baratta, G. A., Leto, G., & Palumbo, M. E. 2002, A&A, 384, 343
Baratta, G. A. & Palumbo, M. E. 1998, Journal of the Optical Society of America A, 15,
3076
Baratta, G. A., Palumbo, M. E., & Strazzulla, G. 2000, A&A, 357, 1045
Barzel, B. & Biham, O. 2007, ApJ, 658, L37
Baulch, D. L., Cobos, C. J., Cox, R. A., et al. 1992, J. Phys. Chem. Ref. Data, 21, 411
Bennett, C. J., Jamieson, C., Mebel, A. M., & Kaiser, R. I. 2004,
Phys. Chem. Chem. Phys., 6, 735
Bennett, C. J. & Kaiser, R. I. 2005, ApJ, 635, 1362
Bennett, C. J. & Kaiser, R. I. 2007, ApJ, 661, 899
Bergin, E. A., Langer, W. D., & Goldsmith, P. F. 1995, ApJ, 441, 222
Bergin, E. A., Melnick, G. J., Gerakines, P. A., Neufeld, D. A., & Whittet, D. C. B. 2005,
ApJ, 627, L33
Bergin, E. A. & Tafalla, M. 2007, ARA&A, 45, 339
Berkley, D. D., Johnson, B. R., Anand, N., Beauchamp, K. M., & Conroy, L. E. 1988,
Appl. Phys. Lett., 53, 1973
175
Bibliography
Bisschop, S. E. 2007, PhD thesis, Leiden Observatory, Leiden University
Bisschop, S. E., Fraser, H. J., Öberg, K. I., van Dishoeck, E. F., & Schlemmer, S. 2006,
A&A, 449, 1297
Bisschop, S. E., Fuchs, G. W., Boogert, A. C. A., van Dishoeck, E. F., & Linnartz, H.
2007a, A&A, 470, 749
Bisschop, S. E., Fuchs, G. W., van Dishoeck, E. F., & Linnartz, H. 2007b, A&A, 474,
1061
Bisschop, S. E., Jørgensen, J. K., van Dishoeck, E. F., & de Wachter, E. B. M. 2007c,
A&A, 465, 913
Bockelée-Morvan, D., Lis, D. C., Wink, J. E., et al. 2000, A&A, 353, 1101
Boogert, A. C. A., Pontoppidan, K. M., Knez, C., et al. 2008, ApJ, 678, 985
Boogert, A. C. A., Pontoppidan, K. M., Lahuis, F., et al. 2004, ApJS, 154, 359
Boonman, A. M. S., van Dishoeck, E. F., Lahuis, F., & Doty, S. D. 2003, A&A, 399, 1063
Bottinelli, S., Boogert, A. C., Bouwman, J., et al. 2010, ApJ, 718, 1100
Bottinelli, S., Ceccarelli, C., Williams, J. P., & Lefloch, B. 2007, A&A, 463, 601
Boudin, N., Schutte, W. A., & Greenberg, J. M. 1998, A&A, 331, 749
Bouwman, J., Ludwig, W., Awad, Z., et al. 2007, A&A, 476, 995
Brosset, P., Dahoo, R., Gauthierroy, B., Abouafmarguin, L., & Lakhlifi, A. 1993, Chem.
Phys., 172, 315
Cami, J., Bernard-Salas, J., Peeters, E., & Malek, S. E. 2010, Science, 329, 1180
Catalano, E. & Sanborn, R. H. 1963, J. Chem. Phys., 38, 2273
Cazaux, S., Cobut, V., Marseille, M., Spaans, M., & Caselli, P. 2010, A&A in press
Chaabouni, H., Schriver-Mazzuoli, L., & Schriver, A. 2000, Low Temp. Phys., 26, 712
Chang, Q., Cuppen, H. M., & Herbst, E. 2007, A&A, 469, 973
Charnley, S. B. 2001, ApJ, 562, L99
Charnley, S. B., Rodgers, S. D., & Ehrenfreund, P. 2001, A&A, 378, 1024
Charnley, S. B., Tielens, A. G. G. M., & Millar, T. J. 1992, ApJ, 399, L71
Charnley, S. B., Tielens, A. G. G. M., & Rodgers, S. D. 1997, ApJ, 482, L203
Chiar, J. E., Adamson, A. J., Kerr, T. H., & Whittet, D. C. B. 1995, ApJ, 455, 234
Chiar, J. E., Pendleton, Y. J., Geballe, T. R., & Tielens, A. G. G. M. 1998, ApJ, 507, 281
Collings, M. P., Anderson, M. A., Chen, R., et al. 2004, MNRAS, 354, 1133
Collings, M. P., Dever, J. W., Fraser, H. J., & McCoustra, M. R. S. 2003, Ap&SS, 285,
633
Cooper, P. D., Moore, M. H., & Hudson, R. L. 2008, Icarus, 194, 379
Cuppen, H. M., Fuchs, G. W., Ioppolo, S., et al. 2008, in IAU Symposium, Vol. 251, IAU
Symposium, ed. S. Kwok & S. Sandford, 377
Cuppen, H. M. & Herbst, E. 2007, ApJ, 668, 294
Cuppen, H. M., Ioppolo, S., Romanzin, C., & Linnartz, H. 2010, Phys. Chem. Chem.
Phys., 12, 12077
Cuppen, H. M., van Dishoeck, E. F., Herbst, E., & Tielens, A. G. G. M. 2009, A&A, 508,
275
176
Bibliography
Dartois, E., Cox, P., Roelfsema, P. R., et al. 1998, A&A, 338, L21
Dartois, E., Schutte, W., Geballe, T. R., et al. 1999, A&A, 342, L32
D’Hendecourt, L. B. & Allamandola, L. J. 1986, A&AS, 64, 453
d’Hendecourt, L. B., Allamandola, L. J., & Greenberg, J. M. 1985, A&A, 152, 130
Draine, B. T. 2009, in Astronomical Society of the Pacific Conference Series, Vol. 414,
Astronomical Society of the Pacific Conference Series, ed. T. Henning, E. Grün, &
J. Steinacker, 453
Dulieu, F., Amiaud, L., Baouche, S., et al. 2005, Chem. Phys. Lett., 404, 187
Dulieu, F., Amiaud, L., Congiu, E., et al. 2010, A&A, 512, A30
Dulieu, F., Amiaud, L., Fillion, J.-H. ., et al. 2007, in Molecules in Space and Laboratory
Eddington, A. S. 1937, The Observatory, 60, 99
Ehrenfreund, P., Boogert, A. C. A., Gerakines, P. A., Tielens, A. G. G. M., & van
Dishoeck, E. F. 1997, A&A, 328, 649
Ehrenfreund, P., Breukers, R., D’Hendecourt, L., & Greenberg, J. M. 1992, A&A, 260,
431
Ehrenfreund, P., Dartois, E., Demyk, K., & D’Hendecourt, L. 1998, A&A, 339, L17
Evans, II, N. J. 1999, ARA&A, 37, 311
Famá, M., Bahr, D. A., Teolis, B. D., & Baragiola, R. A. 2002, Nucl. Instrum. and Methods in Phys. Res. B, 193, 775
Fraser, H. J., Collings, M. P., McCoustra, M. R. S., & Williams, D. A. 2001, MNRAS,
327, 1165
Fraser, H. J. & van Dishoeck, E. F. 2004, Advances in Space Research, 33, 14
Fuchs, G. W., Cuppen, H. M., Ioppolo, S., et al. 2009, A&A, 505, 629
Garrod, R., Park, I. H., Caselli, P., & Herbst, E. 2006, Faraday Disc., 133, 5
Garrod, R. T., Weaver, S. L. W., & Herbst, E. 2008, ApJ, 682, 283
Geppert, W. D., Thomas, R. D., Ehlerding, A., et al. 2005, J. Phys.: Conf. Ser., 4, 26
Gerakines, P. A., Schutte, W. A., & Ehrenfreund, P. 1996, A&A, 312, 289
Gerakines, P. A., Schutte, W. A., Greenberg, J. M., & van Dishoeck, E. F. 1995, A&A,
296, 810
Gerakines, P. A., Whittet, D. C. B., Ehrenfreund, P., et al. 1999, ApJ, 522, 357
Gibb, E., Nummelin, A., Irvine, W. M., Whittet, D. C. B., & Bergman, P. 2000a, ApJ,
545, 309
Gibb, E. L. & Whittet, D. C. B. 2002, ApJ, 566, L113
Gibb, E. L., Whittet, D. C. B., Boogert, A. C. A., & Tielens, A. G. G. M. 2004, ApJ
Suppl., 151, 35
Gibb, E. L., Whittet, D. C. B., & Chiar, J. E. 2001, ApJ, 558, 702
Gibb, E. L., Whittet, D. C. B., & Schutte et al, W. A. 2000b, ApJ, 536, 347
Giguère, P. A. & Harvey, K. B. 1959, J. Mol. Spec., 3, 36
Gillett, F. C. & Forrest, W. J. 1973, ApJ, 179, 483
Goldsmith, P. F., Li, D., & Krčo, M. 2007, ApJ, 654, 273
Goumans, T. P. M. & Andersson, S. 2010, MNRAS, 406, 2213
177
Bibliography
Goumans, T. P. M., Uppal, M. A., & Brown, W. A. 2008, MNRAS, 384, 1158
Green, N. J. B., Toniazzo, T., Pilling, M. J., et al. 2001, A&A, 375, 1111
Greenberg, J. M. 1982, in IAU Colloq. 61: Comet Discoveries, Statistics, and Observational Selection, ed. L. L. Wilkening, 131
Grim, R. J. A. & D’Hendecourt, L. B. 1986, A&A, 167, 161
Hagen, W., Allamandola, L. J., & Greenberg, J. M. 1979, Ap&SS, 65, 215
Hasegawa, T. I. & Herbst, E. 1993, MNRAS, 263, 589
Hasegawa, T. I., Herbst, E., & Leung, C. M. 1992, ApJ Suppl., 82, 167
Hendrix, A. R., Barth, C. A., & Hord, C. W. 1999, J. Geophys. Res., 104, 14169
Herbst, E. & van Dishoeck, E. F. 2009, ARA&A, 47, 427
Hidaka, H., Kouchi, A., & Watanabe, N. 2007, J. Chem. Phys., 126, 204707
Hidaka, H., Watanabe, N., Shiraki, T., Nagaoka, A., & Kouchi, A. 2004, ApJ, 614, 1124
Hiraoka, K., Miyagoshi, T., Takayama, T., Yamamoto, K., & Kihara, Y. 1998, ApJ, 498,
710
Hiraoka, K., Sato, T., Sato, S., et al. 2002, ApJ, 577, 265
Hoare, M. G., Kurtz, S. E., Lizano, S., Keto, E., & Hofner, P. 2007, in Protostars and
Planets V, ed. B. Reipurth, D. Jewitt, & K. Keil, 181
Hogerheijde, M. 1998, PhD thesis, Dept. of Astronomy, University of California, Campbell Hall, Berkeley, CA 94720, USA
Hollenbach, D. & Salpeter, E. E. 1971, ApJ, 163, 155
Hornekær, L., Baurichter, A., Petrunin, V. V., Field, D., & Luntz, A. C. 2003, Science,
302, 1943
Hornig, D. F., White, H. F., & Reding, F. P. 1958, Spectrochim. Acta, 12, 338
Hudson, R. L. & Moore, M. H. 1999, Icarus, 140, 451
Hudson, R. L. & Moore, M. H. 2000, A&A, 357, 787
Ikeda, M., Ohishi, M., Nummelin, A., et al. 2001, ApJ, 560, 792
Ioppolo, S., Cuppen, H. M., Romanzin, C., van Dishoeck, E. F., & Linnartz, H. 2008,
ApJ, 686, 1474
Ioppolo, S., Cuppen, H. M., Romanzin, C., van Dishoeck, E. F., & Linnartz, H. 2010,
Phys. Chem. Chem. Phys., 12, 12065
Ioppolo, S., Cuppen, H. M., van Dishoeck, E. F., & Linnartz, H. 2010a, accepted for
publication in MNRAS
Ioppolo, S., Palumbo, M. E., Baratta, G. A., & Mennella, V. 2009, A&A, 493, 1017
Ioppolo, S., van Boheemen, Y., Cuppen, H. M., van Dishoeck, E. F., & Linnartz, H.
2010b, submitted to MNRAS
Irvine, W. M., Friberg, P., Kaifu, N., et al. 1990, A&A, 229, L9
Jiang, G. J., Person, W. B., & Brown, K. G. 1975, J. Chem. Phys., 62, 1201
Jónsson, H., Mills, G., & Jacobsen, K. W. 1998, Nudged Elastic Band Method for Finding
Minimum Energy Paths of Transitions, ed. B. J. Berne, G. Ciccotti, & D. F. Coker
(Singapore: World Scientific), 385
Jørgensen, J. K., Schöier, F. L., & van Dishoeck, E. F. 2005, A&A, 435, 177
178
Bibliography
Katz, N., Furman, I., Biham, O., Pirronello, V., & Vidali, G. 1999, ApJ, 522, 305
Keane, J. V., Tielens, A. G. G. M., Boogert, A. C. A., Schutte, W. A., & Whittet, D. C. B.
2001, A&A, 376, 254
Keyser, L. F. 1986, J. Phys. Chem., 90, 2994
Khriachtchev, L., Pettersson, M., Jolkkonen, S., Pehkonen, S., & Räsänen, M. 2000,
J. Chem. Phys., 112, 2187
Knez, C., Boogert, A. C. A., Pontoppidan, K. M., et al. 2005, ApJ, 635, L145
Kouchi, A. 1990, J. Cryst. Growth, 99, 1220
Kreutz, J., Serdyukov, A., & Jodl, H. J. 2004, Journal of Physics Condensed Matter, 16,
6415
Lacombe, S., Cemic, F., Jacobi, K., et al. 1997, Phys. Rev. Lett., 79, 1146
Lacy, J. H., Faraji, H., Sandford, S. A., & Allamandola, L. J. 1998, ApJ, 501, L105
Lannon, J. A., Verderame, F. D., & Anderson, Jr., R. W. 1971, J. Chem. Phys., 54, 2212
Leung, C. M., Herbst, E., & Huebner, W. F. 1984, ApJS, 56, 231
Linnartz, H., Acharyya, K., Awad, Z., et al. 2007, in Molecules in Space and Laboratory
Liu, S., Girart, J. M., Remijan, A., & Snyder, L. E. 2002, ApJ, 576, 255
Liu, S., Mehringer, D. M., & Snyder, L. E. 2001, ApJ, 552, 654
Loeffler, M. J., Baratta, G. A., Palumbo, M. E., Strazzulla, G., & Baragiola, R. A. 2005,
A&A, 435, 587
Loeffler, M. J., Raut, U., Vidal, R. A., Baragiola, R. A., & Carlson, R. W. 2006, Icarus,
180, 265
Madzunkov, S., Shortt, B. J., Macaskill, J. A., Darrach, M. R., & Chutjian, A. 2006,
Phys. Rev. A, 73, 020901
Matar, E., Congiu, E., Dulieu, F., Momeni, A., & Lemaire, J. L. 2008, A&A, 492, L17
Mennella, V., Baratta, G. A., Esposito, A., Ferini, G., & Pendleton, Y. J. 2003, ApJ, 587,
727
Mennella, V., Baratta, G. A., Palumbo, M. E., & Bergin, E. A. 2006, ApJ, 643, 923
Mennella, V., Palumbo, M. E., & Baratta, G. A. 2004, ApJ, 615, 1073
Milligan, D. E. & Jacox, M. E. 1964, J. Chem. Phys., 41, 3032
Milligan, D. E. & Jacox, M. E. 1971, J. Chem. Phys., 54, 927
Misochko, E. Y., Akimov, A. V., & Wight, C. A. 1999, J. Phys. Chem. A, 103, 7972
Mitchell, G. F., Maillard, J., Allen, M., Beer, R., & Belcourt, K. 1990, ApJ, 363, 554
Miyauchi, N., Hidaka, H., Chigai, T., et al. 2008, Chem. Phys. Lett., 456, 27
Modica, P. & Palumbo, M. E. 2010, A&A, 519, A22
Mokrane, H., Chaabouni, H., Accolla, M., et al. 2009, ApJ, 705, L195
Moore, M. H., Khanna, R., & Donn, B. 1991, J. Geophys. Res., 96, 17541
Mulas, G., Baratta, G. A., Palumbo, M. E., & Strazzulla, G. 1998, A&A, 333, 1025
Nagaoka, A., Watanabe, N., & Kouchi, A. 2005, ApJ Suppl., 624, L29
Nelander, B. 1980, J. Chem. Phys., 73, 1034
Noll, K. S., Johnson, R. E., Lane, A. L., Domingue, D. L., & Weaver, H. A. 1996, Science,
273, 341
179
Bibliography
Noll, K. S., Roush, T. L., Cruikshank, D. P., Johnson, R. E., & Pendleton, Y. J. 1997,
Nature, 388, 45
Nummelin, A., Whittet, D. C. B., Gibb, E. L., Gerakines, P. A., & Chiar, J. E. 2001, ApJ,
558, 185
Oba, Y., Miyauchi, N., Hidaka, H., et al. 2009, ApJ, 701, 464
Oba, Y., Watanabe, N., Kouchi, A., Hama, T., & Pirronello, V. 2010, ApJ, 712, L174
Öberg, K. I. 2009, PhD thesis, Leiden Observatory, Leiden University, P.O. Box 9513,
2300 RA Leiden, The Netherlands
Öberg, K. I., Boogert, A. C. A., Pontoppidan, K. M., et al. 2008, ApJ, 678, 1032
Öberg, K. I., Bottinelli, S., Jørgensen, J. K., & van Dishoeck, E. F. 2010a, ApJ, 716, 825
Öberg, K. I., Fayolle, E. C., Cuppen, H. M., van Dishoeck, E. F., & Linnartz, H. 2009a,
A&A, 505, 183
Öberg, K. I., Fraser, H. J., Boogert, A. C. A., et al. 2007a, A&A, 462, 1187
Öberg, K. I., Fuchs, G. W., Awad, Z., et al. 2007b, ApJ Lett, 662, L23
Öberg, K. I., Garrod, R. T., van Dishoeck, E. F., & Linnartz, H. 2009b, A&A, 504, 891
Öberg, K. I., Linnartz, H., Visser, R., & van Dishoeck, E. F. 2009c, ApJ, 693, 1209
Öberg, K. I., van Broekhuizen, F., Fraser, H. J., et al. 2005, ApJ Suppl., 621, L33
Öberg, K. I., van Dishoeck, E. F., & Linnartz, H. 2009d, A&A, 496, 281
Öberg, K. I., van Dishoeck, E. F., Linnartz, H., & Andersson, S. 2010b, ApJ, 718, 832
Olsen, R. A., Kroes, G. J., Henkelman, G., Arnaldsson, A., & Jónsson, H. 2004,
J. Chem. Phys., 121, 9776
Orkin, V. L., Kozlov, S. N., Poskrebyshev, G. A., & Kurylo, M. J. 2006, J. Phys. Chem. A,
110, 6978
Palumbo, M. E. 2005, in IAU Symposium, Vol. 227, Massive Star Birth: A Crossroads of
Astrophysics, ed. R. Cesaroni, M. Felli, E. Churchwell, & M. Walmsley, 37
Palumbo, M. E., Baratta, G. A., Brucato, J. R., et al. 1998, A&A, 334, 247
Palumbo, M. E., Baratta, G. A., Fulvio, D., et al. 2008, Journal of Physics Conference
Series, 101, 012002
Palumbo, M. E., Castorina, A. C., & Strazzulla, G. 1999, A&A, 342, 551
Palumbo, M. E. & Strazzulla, G. 1993, A&A, 269, 568
Perets, H. B., Biham, O., Manicó, G., et al. 2005, ApJ, 627, 850
Pontoppidan, K. M. 2006, A&A, 453, L47
Pontoppidan, K. M., Boogert, A. C. A., Fraser, H. J., et al. 2008, ApJ, 678, 1005
Pontoppidan, K. M., Dartois, E., van Dishoeck, E. F., Thi, W.-F., & d’Hendecourt, L.
2003, A&A, 404, L17
Pontoppidan, K. M., Dullemond, C. P., van Dishoeck, E. F., et al. 2005, ApJ, 622, 463
Pontoppidan, K. M., van Dishoeck, E. F., & Dartois, E. 2004, A&A, 426, 925
Requena-Torres, M. A., Martı́n-Pintado, J., Rodrı́guez-Franco, A., et al. 2006, A&A, 455,
971
Romanzin, C., Ioppolo, S., Cuppen, H. M., van Dishoeck, E. F., & Linnartz, H. 2010,
submitted to J. Chem. Phys.
180
Bibliography
Roser, J. E., Vidali, G., Manrico, G., & Pirronello, V. 2001, ApJ, 555, L61
Ruffle, D. P. & Herbst, E. 2000, MNRAS, 319, 837
Ruffle, D. P. & Herbst, E. 2001, MNRAS, 322, 770
Sandford, S. A., Allamandola, L. J., Tielens, A. G. G. M., & Valero, G. J. 1988, ApJ, 329,
498
Schiff, H. I. 1973, in ASSL Vol. 35: Physics and Chemistry of Upper Atmospheres, ed.
B. M. McCormac (Dordrecht: Reidel), 85
Schriver-Mazzuoli, L., de Saxcé, A., Lugez, C., Camy-Peyret, C., & Schriver, A. 1995,
J. Chem. Phys., 102, 690
Schutte, W. A., Allamandola, L. J., & Sandford, S. A. 1993, Icarus, 104, 118
Schutte, W. A., Boogert, A. C. A., Tielens, A. G. G. M., et al. 1999, A&A, 343, 966
Schutte, W. A., Greenberg, J. M., van Dishoeck, E. F., et al. 1998, Ap&SS, 255, 61
Sellevag, S. R., Georgievskii, Y., & Miller, J. A. 2008, J. Phys. Chem. A, 112, 5085
Sellgren, K., Werner, M. W., Ingalls, J. G., et al. 2010, ApJ, 722, L54
Shu, F. H., Ruden, S. P., Lada, C. J., & Lizano, S. 1991, ApJ, 370, L31
Sivaraman, B., Jamieson, C. S., Mason, N. J., & Kaiser, R. I. 2007, ApJ, 669, 1414
Stantcheva, T. & Herbst, E. 2004, A&A, 423, 241
Stantcheva, T., Shematovich, V. I., & Herbst, E. 2002, A&A, 391, 1069
Strazzulla, G. & Palumbo, M. E. 2001, Advances in Space Research, 27, 237
Talbi, D., Chandler, G. S., & Rohl, A. L. 2006, Chemical Physics, 320, 214
Teolis, B. D., Loeffler, M. J., Raut, U., Famá, M., & Baragiola, R. A. 2007, Icarus, 190,
274
Tielens, A. G. G. 2005, The Physics and Chemistry of the Interstellar Medium (New York:
Cambridge University Press)
Tielens, A. G. G. M. & Allamandola, L. J. 1987a, in ASSL Vol. 134: Interstellar Processes, ed. D. J. Hollenbach & H. A. Thronson, Jr. (Dordrecht: Kluwer), 397
Tielens, A. G. G. M. & Allamandola, L. J. 1987b, in NATO ASIC Proc. 210: Physical
Processes in Interstellar Clouds, ed. G. E. Morfill & M. Scholer, 333
Tielens, A. G. G. M. & Charnley, S. B. 1997, Orig. Life Evol. Biosph., 27, 23
Tielens, A. G. G. M. & Hagen, W. 1982, A&A, 114, 245
Tielens, A. G. G. M., Tokunaga, A. T., Geballe, T. R., & Baas, F. 1991, ApJ, 381, 181
Troe, J. & Ushakov, V. G. 2008, J. Chem. Phys., 128, 204307
Tsang, W. & Hampson, R. F. 1986, J. Phys. Chem. Ref. Data, 15, 1087
Tschersich, K. G. 2000, J. Appl. Phys., 87, 2565
Tschersich, K. G., Fleischhauer, J. P., & Schuler, H. 2008, J. Appl. Phys., 104, 034908
Tschersich, K. G. & von Bonin, V. 1998, J. Appl. Phys., 84, 4065
van Broekhuizen, F. A., Pontoppidan, K. M., Fraser, H. J., & van Dishoeck, E. F. 2005,
A&A, 441, 249
van de Hulst, H. C. 1996, in IAU Symposium, Vol. 178, Molecules in Astrophysics:
Probes & Processes, ed. E. F. van Dishoeck, 13
van Dishoeck, E. F. 2004, ARA&A, 42, 119
181
Bibliography
van Dishoeck, E. F., Blake, G. A., Jansen, D. J., & Groesbeck, T. D. 1995, ApJ, 447, 760
van Dishoeck, E. F., Helmich, F. P., de Graauw, T., et al. 1996, A&A, 315, L349
van Dishoeck, E. F. & Jørgensen, J. K. 2008, Ap&SS, 313, 15
Vegard, I. 1930, Z. Physik, 61, 185
Vigren, E., Hamberg, M., Zhaunerchyk, V., et al. 2010, ApJ, 709, 1429
Visser, R. 2009, PhD thesis, Leiden Observatory, Leiden University, P.O. Box 9513, 2300
RA Leiden, The Netherlands
Walraven, J. T. M. & Silvera, I. F. 1982, Rev. Sci. Instr., 53, 1167
Watanabe, N., Hidaka, H., & Kouchi, A. 2006a, in AIP Conf. Ser., Vol. 855, Astrochemistry - From Laboratory Studies to Astronomical Observations, ed. R. I. Kaiser,
P. Bernath, Y. Osamura, S. Petrie, & A. M. Mebel, 122
Watanabe, N. & Kouchi, A. 2002, ApJ Lett., 571, L173
Watanabe, N., Nagaoka, A., Hidaka, H., et al. 2006b, Planet. Space Sci., 54, 1107
Watanabe, N., Nagaoka, A., Shiraki, T., & Kouchi, A. 2004, ApJ, 616, 638
Watanabe, N., Shiraki, T., & Kouchi, A. 2003, ApJ Lett., 588, L121
Whittet, D. C. B. 2003, Dust in the Galactic Environment (Bristol: Institute of Physics)
Whittet, D. C. B., Bode, M. F., Longmore, A. J., et al. 1988, MNRAS, 233, 321
Whittet, D. C. B., Gerakines, P. A., Tielens, A. G. G. M., et al. 1998, ApJ, 498, L159
Whittet, D. C. B., Shenoy, S. S., Bergin, E. A., et al. 2007, ApJ, 655, 332
Winnewisser, G. & Churchwell, E. 1975, ApJ, 200, L33
Woodall, J., Agúndez, M., Markwick-Kemper, A. J., & Millar, T. J. 2007, A&A, 466,
1197
Xie, D., Xu, C., Ho, T.-S., et al. 2007, J. Chem. Phys., 126, 4315
Xu, C., Xie, D., Zhang, D. H., Lin, S. Y., & Guo, H. 2005, J. Chem. Phys., 122, 244305
Yamada, H. & Person, W. B. 1964, J. Chem. Phys., 41, 2478
Yang, M., Zhang, D. H., Collins, M. A., & Lee, S. 2001, J. Chem. Phys., 115, 174
Yu, H. G. & Varandas, A. J. C. 1997, J. Chem. Soc., Faraday Trans., 93, 2651
Zasowski, G., Kemper, F., Watson, D. M., et al. 2009, ApJ, 694, 459
Zheng, W. & Kaiser, R. I. 2007, Chemical Physics Letters, 450, 55
Ziegler, J. F. 2003, Stopping and Range of Ions in Matter SRIM2003 (available at
www.srim.org)
Zuckerman, B., Ball, J. A., & Gottlieb, C. A. 1971, ApJ, 163, L41
182
Nederlandse Samenvatting
De ruimte tussen de sterren is zeer ijl, ijler dan het beste vacuüm dat we op Aarde kunnen
genereren, maar de ruimte is niet leeg. Waterstof en helium gas en resten van zwaardere
elementen (∼99 %) evenals micrometer grote stofdeeltjes (∼1 %) zijn waargenomen in
de ruimte tussen de sterren. Dit mengsel van gas en stof wordt het interstellaire medium
(ISM) genoemd. Omdat het ISM ook nog eens zeer koud is (enkele tientallen graden
boven het absolute nulpunt) werd lange tijd aangenomen dat in het ISM geen chemische processen plaats vinden die kunnen leiden tot de vorming van moleculen. In de
jaren dertig van de afgelopen eeuw werden echter in donkere interstellaire wolken de
eerste twee-atomige moleculen CN, CH en CH+ waargenomen. Een nieuwe wetenschap
ontstond: de astrochemie. Vandaag de dag, tachtig jaar later, zijn meer dan 150 verschillende moleculen in de ruimte geı̈dentificeerd, zowel in de gasfase als in de vaste stof,
door laboratorium spectra en astronomische waarnemingen te vergelijken. Het betreft
kleine ’alledaagse’ moleculen zoals CO, keukenzout en water, maar ook exotische radicalen, bv. HC11 N en C6 H− , en organische moleculen, zoals ethanol, dimethylether en
ethyleen-glycol, die een mogelijke rol spelen als prebiotische bouwstenen. Een deel van
deze moleculen ontstaat in gasfase reacties, een ander deel in de vaste stof, en ook de wisselwerking tussen de gasfase en de vaste stof speelt een belangrijke rol. De onderliggende
chemische processen zijn sterk afhankelijk van de evolutionaire levensfase waarin een
ster en het haar omringende materiaal zich bevinden.
De nadruk in dit proefschrift ligt op reacties die plaatsvinden in de vaste stof, om precies te zijn in de dunne ijslaagjes op interstellaire stofdeeltjes. Dergelijke ijzige stofdeeltjes ontstaan rond nieuwe sterren wanneer gasfase atomen en moleculen vastvriezen en ze
werken als een soort kosmische katalysator. De dichtheid van deeltjes op het oppervlak is
aanzienlijk hoger dan in de gasfase - het ijs functioneert als een reservoir - en bovendien
biedt het oppervlak een mogelijkheid om overtollige reactie-energie af te voeren. Reacties worden thermisch geı̈nitieerd, door interactie met energierijke straling, bv. hard UV
licht, of deeltjes zoals snelle ionen of waterstofatomen. Om het belang van deze reacties in de ruimte te begrijpen, is het noodzakelijk vergelijkbare reacties op aarde in een
laboratorium en onder gecontroleerde omstandigheden na te bootsen.
Experimentele studies van simulaties van ruimte ijs hebben een lange traditie in Leiden. In de 80- en 90-er jaren van de afgelopen eeuw werd m.b.v. hoog-vacuüm (HV) op183
Nederlandse Samenvatting
Figure 1 SURFRESIDE (zie ook voorkant proefschrift).
stellingen (p > 10−7 mbar, een tienmiljardste van de normale luchtdruk) voor astronomisch
relevante temperaturen onderzocht of in realistische (en complexe) ijsmengsels organische moleculen kunnen ontstaan, bv. door dagenlange UV bestraling van een ijs. Inderdaad
liet een infrarode en massa spectrometrische analyse van het ijs en van het ijs residu een
opmerkelijk scala aan complexe moleculen zien. Het was echter niet mogelijk om individuele reacties te bestuderen en hun afhankelijkheid van relevante parameters zoals bv.
de ijs temperatuur. Dit werd mogelijk rond de millennium wisseling door de introductie van ultra-hoog vacuüm (UHV) experimenten (p > 10−11 mbar), waarin de mogelijke
vervuiling van een ijs door achtergrond gas (voornamelijk H2 ; in HV is dit voornamelijk
H2 O) een aanzienlijk geringere rol speelt. Interstellaire ijs analogen kunnen tegenwoordig
met de precisie van een enkele monolaag worden bereid in het laboratorium en astrofysisch relevante processen worden nagebootst door het ijs te bombarderen met atomen,
ionen en electronen of te bestralen met hard UV-licht. Hiermee kan een aantal processen experimenteel worden gekarakteriseerd, zoals diffusie, segregatie, thermische en
niet-thermische desorptie, en vooral de vorming van nieuwe moleculen in het ijs. Veranderingen in het ijs worden bestudeerd met UHV detectie technieken, zoals RAIRS (Reflection Absorption InfraRed Spectroscopy), een spectroscopische analysemethode die
het mogelijk maakt in-situ, dus in het ijs, signaal sterktes om te rekenen in molecuulabundanties en TPD (Temperature Programmed Desorption), een massaspectrometrische
184
Nederlandse Samenvatting
methode waarmee in de gasfase gedesorbeerde moleculen kunnen worden gedetecteerd.
In dit proefschrift ligt de nadruk op ijs reacties van moleculen met individuele waterstof atomen. Gezien de hoeveelheid waterstof in de ruimte, spelen dergelijke reacties
een belangrijke rol bij het ontstaan van nieuwe moleculen. In de hoofdstukken 2-8 worden vaste-stofreactieschema’s (CO + H, O2 + H en CO:O2 + H) besproken, die resulteren in de vorming van H2 CO, CH3 OH, H2 O, CO2 en HCOOH. Een recentelijk geconstrueerde UHV-opstelling, SURFRESIDE, met daaraan gekoppeld een waterstof atoom
bron (HABS) is in deze experimenten gebruikt (zie hoofdstuk 1). Het merendeel van de
resultaten dat hier wordt beschreven is met deze Leidse opstelling gerealiseerd en RAIRS
en TPD zijn gebruikt om reacties en reactieproducten in het ijs zichtbaar te maken. In
hoofdstuk 9 wordt een hoog-vacuum experiment beschreven waarin wisselwerkingen van
energierijke ionen met een ijs worden gepresenteerd. De betreffende opstelling bevindt
zich in het laboratorium voor experimentele astrofysica in Catania (Italië).
De volgende paragrafen vatten het onderzoek per hoofdstuk kort samen en geven een
overzicht van de belangrijkste conclusies.
De vorming van CH3 OH in de vaste stof (hoofdstuk 2)
Hoofdstuk 2 beschrijft de vorming van formaldehyde (H2 CO) en methanol (CH3 OH)
H
H
H
door opeenvolgende hydrogenatie reacties in puur CO-ijs (CO −→ HCO −
→ H2 CO −
→
H
H3 CO −→ CH3 OH). Methanol ijs is waargenomen in de ruimte en een scenario waarin
CH3 OH vooral in de vaste stof wordt gevormd is recentelijk flink in de belangstelling
gekomen door het inzicht dat gasfasereacties van ionen en neutrale moleculen niet efficiënt genoeg zijn om de waargenomen methanol abundanties in de ruimte te verklaren
(Geppert et al. 2005). Eerdere CO + H experimenten (Hiraoka et al. 2002, Watanabe
& Kouchi 2002) resulteerden echter in conflicterende conclusies. De ene onderzoeksgroep rapporteerde alleen de vorming van H2 CO, terwijl de andere groep ook de vorming
van CH3 OH constateerde. In hoofdstuk 2 wordt bewezen, dat dit een expliciet gevolg
was van de verschillende H-atoom fluxen die door beide groepen werden gebruikt en dat
methanol inderdaad wordt aangemaakt, zelfs bij zeer lage temperaturen. De vorming
van zowel H2 CO als CH3 OH is daarnaast bevestigd m.b.v. massa spectrometrie in een
TPD experiment. Verder zijn in hoofdstuk 2 m.b.v. RAIRS de reactie snelheden bepaald
voor verschillende ijstemperaturen, ijsdiktes en H-atoom fluxen. Monte Carlo simulaties
van de nieuw verkregen experimentele data leveren waarden voor de energiebarrieres van
de reacties CO + H en H2 CO + H, waarmee het mogelijk wordt om de vorming van
methanol te bestuderen onder interstellaire omstandigheden en over astronomische tijdsschalen (Cuppen et al. 2009). De conclusie is dat hydrogenatie reacties van CO-ijs een
effectief mechanisme vormen waarmee de astronomisch waargenomen methanolabundanties kunnen worden verklaard. Dit is een belangrijke uitkomst, omdat recentelijk is
aangetoond (Öberg et al. 2009b) dat de UV-fotolyse van puur CH3 OH-ijs resulteert in de
vorming van een substantieel aantal van de organische moleculen, die reeds in de ruimte
zijn geı̈dentificeerd.
185
Nederlandse Samenvatting
De vorming van H2 O in de vaste stof (hoofdstukken 3-6)
De vorming van water in oppervlakte reacties wordt uitvoerig behandeld in de hoofdstukken 3 t/m 6. In 1982 werd het idee geopperd dat interstellair water ontstaat op stof
deeltjes, door hydrogenatie van atomair zuurstof (O), van moleculair zuurstof (O2 ) en
van ozon (O3 ) (Tielens & Hagen 1982). De astrofysische relevantie van deze individuele reactiekanalen is afhankelijk van de plek in de ruimte waar ze plaatsvinden. Cuppen & Herbst (2007) en Cazaux et al. (2010) concludeerden op basis van Monte-Carlosimulaties dat het eerste kanaal belangrijk is in diffuse wolken, terwijl de beide andere
kanalen vooral in dichte en koude interstellaire wolken een rol spelen. Pas recentelijk
is het mogelijk gebleken de reactie schema’s O + H, O2 + H en O3 + H ook experimenteel te realiseren. Het blijkt dat voor astronomisch lage temperaturen alle drie vaste
stof kanalen inderdaad resulteren in de vorming van water (Dulieu et al. 2010). Vooral
de hydrogenatie van moleculair-zuurstof ijs is intensief bestudeerd in de afgelopen jaren
(Miyauchi et al. 2008, Ioppolo et al. 2008, Matar et al. 2008, Ioppolo et al. 2010, Cup2H
2H
pen et al. 2010). In hoofdstuk 3 wordt het reactie mechanisme O2 −−→ H2 O2 −−→ 2H2 O
beschreven voor een reeks van temperaturen onder de thermische desorptie temperatuur
van O2 . Vergelijking met de hydrogenatie van CO (hoofdstuk 2) leert dat de efficiëntie
waarmee waterstofperoxide (H2 O2 ) en water vormen veel minder temperatuur afhankelijk is dan in het geval van formaldehyde en methanol. Ook blijkt O2 -hydrogenatie te
resulteren in een aanzienlijk hogere conversie dan de paar monolayers in een vergelijkend
CO-hydrogenatie-experiment. De uiteindelijke opbrengst is wel temperatuurafhankelijk.
Deze verschillen zijn een direct gevolg van het feit dat de diffusie van H-atomen in een
O2 -ijs gemakkelijker plaatsvindt dan in een CO-ijs. Dit is het onderwerp van hoofdstuk
4, waar de reactie O2 + H wordt onderzocht voor verschillende ijsdiktes, ijstemperaturen,
ijsstructuren en waterstofconcentraties in de H-atoom bundel. Hoofdstuk 5 gaat vervolgens in op het onderliggende reactieschema, met het doel om de reactiekanalen en hun
relatieve efficiëntie te bepalen. In hoofdstuk 6 wordt tenslotte de hydrogenatie van ozon
ijs besproken. O3 + H is een experimenteel moeilijk te bestuderen reactie, omdat bij de
depositie van ozon ervoor gezorgd moet worden, dat het ijs vrij is van moleculair zuurstof.
Dit wordt gerealiseerd door het ozon te deponeren bij een temperatuur die hoger is dan
de desorptie temperatuur van O2 , maar lager dan die van O3 . Zuurstof moleculen die
tijdens de hydrogenatie van ozon ontstaan, desorberen meteen en nemen niet deel in het
reactieschema. Op deze wijze is het mogelijk gebleken om de vorming van water via de
vervolgreactie van OH met H of H2 te bestuderen. De reactie O3 + H lijkt op de hydrogenatie van CO-ijs; alleen de bovenste lagen nemen deel in het reactieschema. Verder
lijkt de tunnelreactie OH + H2 efficiënter te zijn dan de reactie OH + H en deze reacties verdienen in de toekomst zeker een diepergaande studie. De experimentele resultaten
tonen aan, dat het oorspronkelijk model met separate O + H, O2 + H en O3 + H reactiekanalen te eenvoudig was en dat de drie kanalen in feite sterk onderling gekoppeld zijn
door gemeenschappelijke tussenproducten. Dit is van belang voor astrochemische modellen die de vorming van water verklaren voor interstellaire omstandigheden. De data
die in deze hoofdstukken worden gepresenteerd zijn belangrijk om in de nabije toekomst
186
Nederlandse Samenvatting
astronomische waarnemingen van water (HIFI-Herschel en ALMA) met meer diepgang
te interpreteren.
De vorming van CO2 in de vaste stof: O2 + H vs. CO +
H (hoofdstuk 7)
De resultaten in de hoofdstukken 2-6 zijn gebaseerd op H-atoom interacties met puur
ijs (CO, O2 en O3 ). Astronomische waarnemingen laten zien, dat de interstellaire ijs
samenstelling aanzienlijk complexer is. Dus hoe gedraagt zich een ijs, dat bv. bestaat
uit een mengsel van CO en O2 ? In hoeverre beı̈nvloeden de beide bestanddelen elkaars
chemie - ze zijn nu in competitie om met een H-atoom te kunnen reageren - en ontstaan
daarbij nieuwe moleculen, die niet worden gevormd in een hydrogenatie-experiment van
een enkelvoudig ijs? Deze vragen worden beantwoord in hoofdstuk 7. De gelijktijdige
hydrogenatie van CO en O2 resulteert in de productie van koolstofdioxide (CO2 ) door de
reactie van OH met CO. Hoofdstuk 7 laat zien, dat het CO en O2 ook elkaars reactiviteit
beı̈nvloeden. Dit is te verwachten, gezien H-atoom diffusie en penetratie in CO- en O2 -ijs
zo verschillend zijn. De reactiesnelheid blijkt daarbij minder afhankelijk te zijn van de
mengverhouding als de uiteindelijke productie. De beperkende factor daarbij is vooral
de penetratiediepte van de invallende H-atomen. Deze hangt af van de ijs compositie
en neemt af voor een toenemende hoeveelheid CO in het ijs. De reactie snelheden voor
de vorming van H2 CO, CH3 OH, H2 O2 en H2 O (gecorrigeerd voor de gereduceerde Hatoom flux) blijven vergelijkbaar binnen de experimentele onzekerheid met de waardes
gevonden voor puur CO- en O2 -ijs. De belangrijke conclusie is dat CO2 ijs gevormd kan
worden via een thermische CO + OH reactie.
De vorming van HCOOH in de vaste stof (hoofdstuk 8)
Hoofdstuk 8 behandelt de vorming van mierezuur (HCOOH) in de vaste stof. HCOOH
wordt zowel in koudere als warmere gebieden in het ISM aangetroffen, maar het is niet
duidelijk hoe dit organisch zuur ontstaat. Als alternatief voor gasfase reactie schema’s is
een aantal mogelijke reacties in de vaste stof geponeerd; de opeenvolgende toevoeging
van H-, O- en H-atomen aan CO ijs (Tielens & Hagen 1982), via de reactie HCO + OH
(Garrod et al. 2006) of uitgaande van de hydrogenatie van een HO-CO complex (Goumans
et al. 2008), dat in het ijs door intramoleculaire energieoverdracht stabiliseert. De reactie
HO-CO + H resulteert vervolgens in een van de reactie producten CO2 + H2 , H2 O + CO
of HCOOH. Het doel van hoofdstuk 8 is om te laten zien dat de hydrogenatie van het
HO-CO complex een effectieve methode is om bij lage temperatuur mierezuur te vormen.
Daartoe wordt met RAIRS en TPD gekeken naar ijs dat ontstaat door gelijktijdig waterstof
atomen en een CO:O2 -gasmengsel te deponeren op het cryogeen gekoelde substraat. Na
co-depositie wordt inderdaad HCOOH gemeten, wanneer de temperatuur wordt verhoogd
tot een waarde onder die van de CO- en O2 -desorptietemperatuur (<30 K). Voor dergelijke
temperaturen kunnen H-atomen in de ijsmatrix reageren met het gestabiliseerde HO-CO
187
Nederlandse Samenvatting
complex. Voor lage temperaturen blijkt de reactie HCO + OH langzaam te zijn. De
conclusies van dit onderzoek zijn in dit hoofdstuk in een astronomische context geplaatst.
Hieruit volgt, dat HCOOH-productie uit HO-CO-hydrogenatie een mogelijke verklaring
biedt voor de mierezuur dichtheden, die in dichte interstellaire wolken ten tijde van de
opwarmings fase van de protoster zijn waargenomen.
De vorming van CO2 in de vaste stof door ion beschieting (hoofdstuk 9)
In hoofdstuk 7 werd besproken hoe CO2 ontstaat in een astronomisch relevant temperatuur bereik, door de reactie van CO met OH. In hoofdstuk 9 wordt een alternatieve
route besproken, waarbij chemische reacties worden gestart door beschieting met energierijke (30−200 keV) ionen. Voor verschillende ijs samenstellingen worden veranderingen
zichtbaar gemaakt m.b.v. infrarood transmissie spectroscopie en de resulterende data worden rechtstreeks vergeleken met astronomische waarnemingen. Het laboratorium werk
laat zien dat op deze wijze CO2 wordt gevormd in ijs dat bestaat uit C- en O-houdende
moleculen. De efficiëntie waarmee CO2 vormt, neemt toe in waterrijke mengsels. Het
profiel van de CO2 strek- en buigvibratie blijkt daarbij afhankelijk te zijn van ijsmengsel
en ijstemperatuur (zie ook hoofdstuk 1). De conclusie is, dat ook dit productie proces
resulteert in een efficiënte manier om CO2 onder interstellaire omstandigheden te produceren en dat het reactiemechanisme de hoeveelheid waargenomen CO2 bij protosterren kan
verklaren.
188
Publications
Refereed papers
• Water formation by surface O3 hydrogenation
Romanzin, C., Ioppolo, S., Cuppen, H. M., van Dishoeck, E. F., Linnartz, H. 2010,
submitted to Journal of Chemical Physics (Chapter 6)
• Competition between CO and O2 -ice hydrogenation channels and surface formation of CO2 at low temperatures
Ioppolo, S., van Boheemen, Y., Cuppen, H. M., van Dishoeck, E. F., Linnartz, H.
2010, submitted to Monthly Notices of the Royal Astronomical Society
(Chapter 7)
• The influence of temperature on the synthesis of molecules on icy grain mantles in
dense molecular clouds
Garozzo, M., La Rosa, L., Kanuchova, Z., Ioppolo, S., Baratta, G. A., Palumbo,
M. E., Strazzulla, G. 2010, accepted for publication in Astronomy & Astrophysics
• Surface formation of HCOOH at low temperature
Ioppolo, S., Cuppen, H. M., van Dishoeck, E. F., Linnartz, H. 2010, accepted for
publication in Monthly Notices of the Royal Astronomical Society (Chapter 8)
• Water formation at low temperatures by surface O2 hydrogenation II: The reaction
network
Cuppen, H. M., Ioppolo, S., Romanzin, C., Linnartz, H. 2010, Physical Chemistry
Chemical Physics, 12, 12077-12088 (Chapter 5)
• Water formation at low temperatures by surface O2 hydrogenation I: Characterization of ice penetration
Ioppolo, S., Cuppen, H. M., Romanzin, C., van Dishoeck, E. F., Linnartz, H. 2010,
Physical Chemistry Chemical Physics, 12, 12065-12076 (Chapter 4)
189
Publications
• Hydrogenation reactions in interstellar CO ice analogues. A combined experimental/theoretical approach
Fuchs, G. W., Cuppen, H. M., Ioppolo, S., Romanzin, C., Bisschop, S. E., Andersson, S., van Dishoeck, E. F., Linnartz, H. 2009, Astronomy & Astrophysics, 505,
629-639 (Chapter 2)
• Formation of interstellar solid CO2 after energetic processing of icy grain mantles
Ioppolo, S., Palumbo, M. E., Baratta, G. A., Mennella V. 2009, Astronomy &
Astrophysics, 493, 1017-1028 (Chapter 9)
• Laboratory evidence for efficient water formation in interstellar ices
Ioppolo, S., Cuppen, H. M., Romanzin, C., van Dishoeck, E. F., Linnartz, H. 2008,
Astrophysical Journal, 686, 1474-1479 (Chapter 3)
Conference Proceedings
• Formation of alcohols on ice surfaces
Cuppen, H. M., Fuchs, G. W., Ioppolo, S., Bisschop, S. E., Öberg, K. I., van
Dishoeck, E. F., Linnartz, H. 2008, Organic Matter in Space, International Astronomical Union Symposium, 251, 377-382
• Laboratory study of CO ice hydrogenation
Ioppolo, S., Fuchs, G. W., Bisschop, S. E., van Dishoeck, E. F., Linnartz, H. 2007,
Molecules in Space and Laboratory
• Solid state astrophysics and chemistry: four questions - four answers
Linnartz, H., Acharyya, K., Awad, Z., Bisschop, S. E., Bottinelli, S., Bouwman, J.,
Cuppen, H. M., Fuchs, G. W., Ioppolo, S., Öberg, K. I., van Dishoeck, E. F. 2007,
Molecules in Space and Laboratory
• Ion irradiation of TNO surface analogue ice mixtures: the chemistry
Baratta, G. A., Brunetto, R., Caniglia, G., Fulvio, D., Ioppolo, S., Leto, G., Palumbo,
M. E., Spinella, F., Strazzulla, G. 2007, Memorie della Società Astronomica Italiana Supplementi, 11, 185-189
190
Curriculum Vitae
Originally from Italy, I was born on November 17, 1980 in the city of Catania. In 1994
I began my high school education at the Liceo Scientifico Galileo Galilei, a school for
science. I graduated in five years with the highest possible grades. During these years I
developed passion for physics and mathematics and I decided to continue my education at
the University of Catania at the department of Physics; there I began my specialization in
Astrophysics. My master thesis constituted of experimental investigations of the formation of interstellar solid CO2 after energetic processing of icy grain mantles. The research
was carried out in the LASp (Laboratory for Experimental Astrophyiscs) in Catania, under the supervision of Dr. Maria Elisabetta Palumbo and Prof. Giovanni Strazzulla and
resulted in a publication.
My following doctoral studies began after receiving an invitation from the Raymond
and Beverly Sackler Laboratory for Astrophysics in Leiden, the Netherlands. Supervised
by Prof. Harold Linnartz, Prof. Ewine van Dishoeck, and Dr. Herma Cuppen (daily
supervisor) my PhD project dedicated to the experimental investigation of surface formation routes of interstellar molecules took shape. All experiments were performed using SURFRESIDE which consists of an ultra-high vacuum main chamber and a hydrogen/deuterium atomic line and allows to study H-atom addition reactions in interstellar
ice analogs under astronomically relevant conditions. The doctoral research resulted in
seven publications using SURFRESIDE, which are presented in this thesis. In addition,
the project required the construction and implementation of a second atomic beam line on
SURFRESIDE. I have worked together with Dr. Guido Fuchs and Dr. Claire Romanzin
in the development of the new system. Along with my research, I also supervised the
work of two bachelor students from the Leiden University and one graduate student from
the University of Catania, Italy. Furthermore, I had the opportunity to spend one month
as a visiting researcher at the Laboratory for Experimental Astrophysics (LASp). During
my PhD studies, I took part of several international and national scientific conferences
and summer schools in the Netherlands, the UK, France, Germany, and Iceland where I
presented my work. I will continue my scientific career in the Sackler Laboratory for Astrophysics as a postdoctoral researcher with the intention to test, prove and tune the newly
reconstructed SURFRESIDE setup, and to continue the investigation of more complex
surface formation routes of interstellar molecules by using a double atomic beam line.
191
Acknowledgements
This PhD thesis is the collective result of hard work of many people. That is why I
am sincerely grateful to all persons involved in my research during the past four years.
These following couple of pages will allow me to mention and thank at least some of
them. Among the people that I can name here, I would like to thank the computer department (Aart, David, Erik, and Tycho) as well as the support staff (Anita, Evelijn, Jeanne,
Kirsten, and Liesbeth) for being always kind and helpful. A close collaboration with the
machine workshop FMD/ELD of the Faculty W&N made this work possible. I express
my gratitude to Ewie, Gijsbert and most of all to Martijn who always came to help when
I needed technical support in the laboratory. With Martijn I do not only share the passion for science but also for climbing. Last but not least, my appreciation is extended to
the electronic, glass-blower and cryogenic departments. It was my pleasure to work with
every single one of you. Thank you.
Perhaps the most prominent place within this acknowledgement belongs to the coauthors of this thesis; in particular Claire Romanzin, Guido Fuchs, Herma Cuppen, Stefan
Andersson, Suzanne Bisschop and Yorick van Boheemen. Their dedication and professionalism made our cooperation dynamic and successful. In addition, I am deeply grateful
to the LASp group from Catania (Italy) who introduced me to the astrochemistry several
years ago and with whom I kept collaborating also during my PhD. This inspiring collaboration resulted in a publication included in this thesis. I thank Gianni Strazzulla, Maria
Elisabetta Palumbo, Giuseppe Baratta, Giuseppe Leto and Franco Spinella from Catania
observatory, and Vito Mennella from Naples observatory. Stimulating discussion with
Lou Allamandola, Xander Tielens, Marc van Hemert, Fedor Goumans, Carina Arasa and
Erik Vigren improved the quality of this thesis.
The following paragraph is dedicated to my colleagues and friends from the Sterrewacht. I would like to thank the AstroChem group, but since there are too many of you
I will not attempt to name all of you, since I would surely forget a few. Special gratitude
goes to the Sackler Laboratory for Astrophysics. In the four years of my PhD, I have
been lucky to meet exceptional people and colleagues and to develop close friendships.
Herma, working with you is a pleasure. You taught me a lot, helping me throughout my
PhD. Your friendship is precious to me. Jordy, you are a great friend, without you the
Lab wouldn’t have been the same. Claire, Edith, Guido, Harald, Joseph, Karin, Karoliina,
193
Acknowledgements
Nadine, and Suzanne, I enjoyed the time we spent together in and outside of the Lab.
Michele, Emily, and little Francesca, I feel lucky I met such a great family and friends.
To the new people in this group, Gleb, Jean-Baptiste, and Steven, I wish all the best.
Dominic, Simon, Karoliina, Silvia, Ernst, and Francisco, you are the best office mates.
Outside of work, I would like to mention a few friends that stayed by me throughout the
years. Rafael, we shared so many great and unforgettable moments, you are a real friend.
Kalle, our sailing trip along the West Swedish coast is one of the best adventures of my
life. Nicola, you made our weekly climbing sessions a pleasure. Titti and Carina, you
made me laugh like never before. Franco, Paola, and Daniele, thank you for making my
time abroad less lonely, I consider you to be part of my family.
Last but not least, I would like to dedicate this thesis to my family and friends in Italy.
Without them I wouldn’t be where I am now. I always keep them close to my heart. Petra,
I owe you everything, your boundless support and infinite love pulled me through the
hardest times. Papà, Mamma and Ivana this thesis is for you.
194
Notes:
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement