Astrochemistry and Astrophysics PDF

Faculté des Sciences
Astrophysics and astrochemistry

Michaël De Becker
– Academic year 2013-2014 –
Masters en Sciences Chimiques

et en Sciences Géologiques
de l’Université de Liège
Contents
Generalities 1
1 Astronomical environments 5
1.1 General concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.1.1 Black body radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.1.2 Distance scales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.1.3 Magnitude and extinction . . . . . . . . . . . . . . . . . . . . . . . . 8
1.2 Stellar objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2.1 Hertzsprung-Russell diagram . . . . . . . . . . . . . . . . . . . . . . 11
1.2.2 Spectral classification . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.2.3 Stellar formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2.4 Stellar nucleosynthesis and evolution . . . . . . . . . . . . . . . . . . 16
1.3 Interstellar medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.4 Solar system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4.1 A few definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4.2 Planets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.4.3 Comets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.4.4 Asteroids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2 Chemical processes 27
2.1 Elementary gas-phase processes . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.1.1 Photodissociation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.1.2 Photoionization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.1.3 Neutral-neutral reactions . . . . . . . . . . . . . . . . . . . . . . . . 31
2.1.4 Ion-molecule reactions . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.1.5 Dissociative electron recombination reactions . . . . . . . . . . . . . 33
2.1.6 Cosmic-ray induced reactions . . . . . . . . . . . . . . . . . . . . . . 34
2.1.7 Charge transfer reactions . . . . . . . . . . . . . . . . . . . . . . . . 36
2.1.8 Radiative association reactions . . . . . . . . . . . . . . . . . . . . . 37
2.1.9 Associative detachment reactions . . . . . . . . . . . . . . . . . . . . 38
2.1.10 Collisional association reactions . . . . . . . . . . . . . . . . . . . . . 39
2.1.11 Collisional dissociation reactions . . . . . . . . . . . . . . . . . . . . 39
2.1.12 Gas-phase chemical networks . . . . . . . . . . . . . . . . . . . . . . 39
2.2 Grain-surface processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.2.1 Schematic view of a grain-surface process . . . . . . . . . . . . . . . 42
2.2.2 The formation of molecular hydrogen . . . . . . . . . . . . . . . . . 46
2.2.3 Grain-surface chemical networks . . . . . . . . . . . . . . . . . . . . 49
i
ii Table of contents
3 Molecules in the Universe 51

3.1 Molecular content of the Galaxy . . . . . . . . . . . . . . . . . . . . . . . . 51
3.1.1 Inorganic compounds . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.1.2 Hydrocarbons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.1.3 Organic compounds with functional groups . . . . . . . . . . . . . . 62
3.2 Important scientific issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.2.1 The quest for amino acids . . . . . . . . . . . . . . . . . . . . . . . . 66
3.2.2 The homochirality problem . . . . . . . . . . . . . . . . . . . . . . . 67
3.2.3 Molecular filiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Bibliography 85
Generalities
What is astrophysics?
Astrophysics is the science devoted to the study astronomical objects that populate the Uni-
verse, including their formation and evolution, through a dedicated application of physical
concepts. Astrophysics started when astronomers began to apply physics to understand
the objects they were studying while contemplating the sky.
To give an idea of the scientific topics which are studied in modern astrophysics, here
is a non-exhaustive list:
- the formation and evolution of stars
- the formation and evolution of galaxies
- the dynamics and evolution of the interstellar medium
- the formation of planets and planetary systems
- the physics of astronomical objects in extreme conditions (black holes, neutron stars,
very hot plasmas...)
- extragalactic objects and large-scale structures
- the evolution of the Universe
- particle physics in astronomical environments
- ...
What is astrochemistry?
Astrochemistry is the science devoted to the study of the chemical processes at work in
astrophysical environments, including the interstellar medium, comets, circumstellar and
circumplanetary regions.
It is important to insist on the difference between two fields that are often confused,
at least in the astrophysicists community, i.e. chemistry and molecular spectroscopy.
Chemistry is a science interested in processes transforming molecular species, and molec-
ular spectroscopy is a technique used to make a qualitative and quantitative census of
molecules in a given environment. In astronomy, the ongoing exploration of the sky leads
to the discovery of many molecules, mainly in the interstellar medium and in circumstellar
environments. These discoveries are made through molecular spectroscopy studies. Beside
that, researchers are trying to understand the processes likely to lead to the formation of
1
2 Generalities
these molecules, as a function of the properties of their environment, along with the poten-
tial filiations which exist between chemical species found at a given place. This is typically
what astrochemists are doing, on the basis of the very valuable information gathered by
molecular spectroscopists.
Astrochemistry constitutes therefore some kind of overlap between chemistry and as-
trophysics, its two parent disciplines. It should also be kept in mind that this inter-relation
is not univoque. On the one hand, astrochemistry extends the domains where chemistry
can be applied. Chemistry is not limited to a laboratory science, and astrochemistry opens
the door of astronomy, in a similar way as biochemistry opens that of living organisms.
On the other hand, astrochemical processes are expected to be important for many issues
of astrophysics in the sense that chemical processes can be used as valuable tracers for
physical properties of astronomical environments to which they are intimately dependent.
Astrochemistry is the natural pluridisciplinary answer that modern science has developed
to address scientific questions which are totally out of the scope of existing individual
scientific disciplines.
The challenges of astrochemistry

In short, the aim of astrochemistry is to understand the chemical processes at work in
space. However, some more specific objectives are worth discussing briefly before going
through the details in the next sections.
One of the tasks of astrochemists is to understand the impact of the astrophysical
environment on chemical processes responsible for the formation of molecules. In this
context, their studies aim at understanding the level of complexity that can be reached by
molecular species in space. This is important in the sense that chemical complexity is a
necessary condition for the development of life in the Universe.
What are the most complex molecules that can be formed in space? How do they
form? Are these molecules likely to resist to the collapse of protostellar clouds leading to
the formation of stellar systems? What is the relation between the molecular populations
found in comets and in the interstellar medium? Is the interstellar medium the reactor
where some molecules important for life are formed? These questions summarize the main
scientific issues that are motivating research in astrochemistry.
The circumstances under which astrochemistry is taking place are somewhat different
from those of laboratory chemistry, and these differences deserve a few words. Astro-
chemistry investigates indeed chemical processes under unusual conditions (low densities,
low temperatures, exposition to strong radiation fields and to high energy particles often
called cosmic rays...). These conditions are difficult to reproduce in laboratory, and this
constitutes a severe problem if one wants to explore the reactivity on the basis of labora-
tory experiments. In addition, the study of chemical reactions in space is based upon an
uncomplete census of potential reaction partners. Molecular spectroscopy studies allow us
to obtain many information on the molecular populations in space, but the task in not
easy. Molecular spectroscopists have to deal with blends of molecules whose spectroscopic
identification is often difficult. As molecular complexity increases, this difficulty increases
rapidly, and we do not have the opportunity to separate compounds in order to facilitate
their identification. The dependence on a lot of physical parameters which are not at all
controlled is an issue as well. The physical parameters can not be tuned at will in order
to study the influence of a parameter or another on reactivity.
Generalities 3
A brief census of molecules in astronomical environments

Molecular spectroscopy studies led to the detection of many molecules in the interstellar
medium. Most of them have been discovered via their rotational signature from radio to
far-infrared frequencies. The first interstellar molecules identified thanks to their sharp
bands in gas phase were CH (Swings & Rosenfeld 1937), CN (McKellar 1940) and CH+
(Douglas & Herzberg 1941), and the first polyatomic molecules (NH3 ) was discovered in
the end of the 60’s (Cheung et al. 1968) thanks to the development of radioastronomy.
Since then, mostly thanks to significant progresses in millimeter astronomy, many other
molecules of increasing complexity have been firmly identified in the ISM. So far, this
amounts up to about 170 molecules. Lists of identified molecules, sorted by increasing
number of constituting atoms, are given in Table 1. It should be noted that, up to now,
most of the molecules that have been detected in the interstellar medium or elsewhere in
space are neutral species. Some cations have also been found, along with only a few anions
whose discovery is quite recent.
The present census1 of interstellar molecules leads automatically to a question which
lies at the core of astrochemistry, i.e. the question of molecular complexity. Many poly-
atomic molecules have been detected and firmly identified, and much more are certainly
waiting to be discovered in the future.
The objectives of this course

The main motivations behind the organization of this course are the following:
- introduce some basic concepts of astrophysics, mainly concerning Galactic objects

and their related environments,
- give a brief overview of the processes and circumstances under which chemical ele-
ments (important for chemical and geological purposes) are synthesized in the Uni-
verse,
- introduce some basic concepts of astrochemistry, i.e. present the main processes
responsible for the formation of molecules in astronomical environments,
- address some highly relevant scientific questions in relation with astrophysics and as-
trochemistry, emphasizing the importance of pluridisciplinary approaches in modern
sciences,
- finally, enhance the scientific culture of students in disciplines which are not directly
related to space sciences.
1
The census of firmly identified interstellar molecules presented here is partly taken from the following
website http://www.astrochymist.org, created and maintained by D.E. Woon.
4 Generalities
Table 1: Temporary census of interstellar molecules.

2 atoms AlCl, AlF, AlO, C2 , CF+ , CH, CH+ , CN, CN− , CO,
CO+ , CP, CS, FeO, H2 , HCl, HF, NH, KCl, N2 ,
NO, NS, NaCl, O2 , OH, OH+ , PN, SH, SH+ , SO, SO+ ,
SiC, SiN, SiO, SiS, P, TiO, ArH+
3 atoms AlNC, AlOH, C3 , C2 H, C2 O, C2 P, C2 S,CO2 , H+ 3 , CH2 ,
H2 Cl+ , H2 O, H2 O+ , H2 S, HCN, HCO, HCO+ , HCS+ , HCP,
HNC, HN+ +
2 , HNO, HOC , KCN, MgCN, NH2 , N2 H ,
+
N2 O, NaCN, OCS, SO2 , c-SiC2 , SiCN, SiNC, FeCN, TiO2

4 atoms l-C3 H, c-C3 H, C3 N, C3 O, C3 S, C3 H+ , H3 O+ ,
C2 H2 , H2 CN, H2 CO, H2 CS, HCCN, HCNH+ ,
HCNO, HOCN, HOCO+ , HNCO, HNCS, HSCN, NH3 , SiC3 ,
PH3 , H2 O2 , HMgNC
5 atoms C5 , CH4 , c-C3 H2 , l-C3 H2 , H2 CCN, H2 C2 O,
H2 CNH, H2 COH+ , C4 H, C4 H− , HC3 N, HCCNC,
HCOOH, NH2 CN, SiC4 , SiH4 , HCOCN, HNCNH
HC3 N− , CH3 O
6 atoms c-H2 C3 O, C2 H4 , CH3 CN, CH3 NC, CH3 OH,
CH3 SH, l-H2 C4 , HC3 NH+ , HCONH2 , C5 H, HC2 CHO,
HC4 N, CH2 CNH, C5 N−
7 atoms c-C2 H4 O, CH3 C2 H, H3 CNH2 , CH2 CHCN, H2 CHCOH,
C6 H, C6 H− , HC4 CN, CH3 CHO, HC5 N−
8 atoms H3 CC2 CN, H2 COHCOH, CH3 OOCH, CH3 COOH,
C6 H2 , CH2 CHCHO, CH2 CCHCN, C7 H,
NH2 CH2 CN
9 atoms CH3 C4 H, CH3 OCH3 , CH3 CH2 CN, CH3 CONH2 ,
CH3 CH2 OH, C8 H, HC6 CN, C8 H− , CH2 CHCH3
> 9 atoms CH3 COCH3 , CH3 CH2 CHO, CH3 C5 N,
HC8 CN, CH3 C6 H, CH3 OC2 H5 , HC10 CN,
C6 H6 , C2 H5 OCHO, C3 H7 CN, CH3 COOCH3 ,
C60 , C70
Deuterated HD, H2 D , HDO, D2 O, DCN, DCO, DNC, N2 D+ ,
+
NHD2 , ND3 , HDCO, D2 CO, CH2 DCCH, CH3 CCD,

D2 CS, NH3 D+
Chapter 1
Astronomical environments
1.1 General concepts

1.1.1 Black body radiation
An important concept in the context of astrophysics is that of black body radiation. A
black body is an ideal body that is able to absorb all the light which hits it, whatever its
wavelength. The behaviour of a black body can be obtained if one considers a mixture of
particles and photons intimately interacting in an enclosure. In such a situation, the system
can reach the thermodynamical equilibrium where each elementary process characterizing
the interactions between particles, and between particles and photons, is compensated by
its reverse process.
In a medium populated by particles and photons, we can consider the following points:
- the energy distribution of electrons is defined by the Maxwell-Boltzmann relation,

characterized by a unique temperature called the electron temperature
- the ionization levels for a given element are distributed according to Saha’s law,
characterized by a unique temperature called ionization temperature
- the level populations of electrons bound to a given atom or ion are distributed
according to the Boltzmann’s law, also characterized by its own temperature called
excitation temperature
- if molecules are present, they are partitioned among their various vibrational levels
or among their numerous rotational levels, and we could assign to these distributions
a vibrational temperature and a rotational temperature, respectively.
Thermodynamical equilibrium is reached when these temperatures (electron, ionization,

excitation, vibrational, rotational...) are equal. The properties of a black body are com-
pletely defined by its temperature. For a given temperature, the black body will produce
a continuum emission spectrum. The distribution of photons emitted by the black body
as a function of wavelength is defined by a given temperature, equal to the temperatures
enumerated above. Such a temperature (T) is the temperature of the black body.
The emission spectrum of a black body is described by the Planck’s law. This law
expresses the spectral energy distribution of the radiation field produced by the black
body. Physically, it is equivalent to an amount of energy emitted per time unit, per
wavelength unit, per surface unit along a direction perpendicular to that surface within
5
6 Astronomical environments
Figure 1.1: Examples of black body spectra at different temperatures.
an unitary solid angle. It can be expressed as follows:
2 h c2 1
Bλ = (1.1)
λ5 ehc/λkT − 1
where λ is the wavelength, and T is the temperature of the black body. The con-
stants are respectively Planck’s constant (h = 6.626 10−26 erg s−1 ), the speed of light (c
= 2.99792 1010 cm s−1 ) and Boltzmann’s constant (k = 1.3806 10−16 erg K−1 ). Planck’s
function can also be expressed as a function of frequency (ν):
2 h ν3 1
Bν = 2 hν/kT
(1.2)
c e −1
In Fig. 1.1, Eq. 1.1 is plotted for several values of the black body temperature. When these
curves are compared, two comments can be formulated:
1. The surface area under the curve, i.e. the energy amount integrated over the elec-
tromagnetic spectrum, increases substantially as the temperature increases. This is
expressed by the relation for the spectral intensity of a black body, also called Stefan’s
law: B = σ T4 /π, where σ (= 5.6696 10−5 erg cm−2 s−1 K−4 ) is Stefan-Boltzmann’s
constant. It is important to note that the relation between B and the flux (F,
expressed in erg cm−2 s−1 ) is very simple: F = πB. The flux is therefore simply ex-
pressed by F = σ T4 . Such a dependence versus temperature explains how hotter
bodies radiate much more radiative energy than cooler bodies, as illustrated by the
curves plotted in Fig. 1.1.
2. The maximum of the curve shifts to shorter wavelengths when higher temperatures
are considered. This is expressed by Wien’s displacement law: λmax = 2.9 107 /T,
with λmax expressed in Å and T expressed in K.
The reason why the concept of black body is so important in astrophysics is the fact
that the continuum spectrum of many objects, including mostly stars, corresponds very
reasonably to that of a black body. Typically, a stellar emission spectrum is that of a
1.1. General concepts 7
black body characterized by a temperature called effective temperature. For instance, the
effective temperature of the Sun is about 5800 K. The corresponding Planck function peaks
at a wavelength close to 5500 Å . On the other hand, the night sky is full of stars displaying
colors suggesting their emission spectrum peaks at different temperatures. This is a result
of the fact that there is a distribution of effective temperature among the stars. Stars
such as Betelgeuse (in Orion), Antares (in Scorpius) or Aldebaran (in Taurus) appear red,
because their atmosphere is cooler than that of the Sun. On the other hand, much hotter
stars, with effective temperatures of several 10000 K exist in the Universe. The continuum
spectrum of such stars can even peak in the ultraviolet domain, in the most extreme cases.
The F quantity (i.e. the flux) defined above can therefore express the amount of
energy emitted by a star, per second, per square centimeter. A very useful quantity for
astrophysics purpose is the total radiative power produced by a star, namely the total
radiative energy per second. Considering that the star emits its spectrum isotropically,
the total power is that emitted by a sphere of radius equal to that of the star, keeping
in mind that each surface element produces the spectrum of a black body at the effective
temperature of the star. The flux given by Stefan’s law has just to be multiplied by
the total surface of the sphere corresponding to the size of the star. We thus obtain the
following relation giving the total luminosity (expressed in erg s−1 ) of a star with radius
R and effective temperature T:
L = 4 π R2 σ T4 (1.3)
The latter relation is very useful for instance to estimate the luminosity of a star as
a function of fundamental parameters (R, T), or to estimate a stellar radius provided
effective temperature and luminosity have been determined.
1.1.2 Distance scales

When astronomical environments are concerned, one has to deal with typical sizes, and
distances that are very unfamiliar. This is why the enonciation of a few definitions and
the discussion of several orders of magnitude are needed.
It is important to have in mind the meaning of at least three distance units often used
in astrophysics:
- The astronomical unit (AU). This distance unit is defined as the mean radius
of the Earth’s orbit. Typically, this is the mean distance between the Earth and
the Sun. It corresponds to 149.6 106 km. In cgs units, it converts into 1.496 1013 cm.
This unit is frequently used to express distances within our Solar system, or even in
circumstellar environments of other stars.
- The light-year (ly). This is the distance crossed by electromagnetic radiation

during one year. In cgs units, its value is 9.463 1017 cm. This unit is typically used
when referring to interstellar distances in the Solar neighbourhood. For instance,
the closest star to the Sun is Proxima Centauri, located at a distance of 4.2 ly.
- The parsec (pc). This is defined as the distance at which an observer would see a
linear distance of 1 AU within an angle of 1 arcsec (i.e. 1/3600 degree). The parsec
(and the kpc) are very frequently used to express distances within our Galaxy. One
parsec is equal to 3.089 1018 cm (∼ 3.26 ly).
1.1.3 Magnitude and extinction

The traditional convention to quantify the brightness of a star relies on a ’naked-eye’
classification. Historically, the stars were ranked according to their brightness into six cat-
egories called magnitudes. The brightest stars were of the first magnitude, and the faintest
seen by the naked eye were of the sixth magnitude. This classification has therefore been
standardised afterwards by astronomers, and it is still used in modern science. This stan-
dardisation comes from the fact that quantitative observations show that first magnitude
stars are about 2.5 times brighter than second magnitude stars to the naked eye, and that
the ratio between first and sixth magnitude stars is nearly 100. The brightness ratio be-
tween two stars separated by one magnitude unit has therefore been conventionally set to
1001/5 = 2.512 in 1850. In addition, some ’standard’ stars have been used to calibrate the
entire magnitude scale. For instance, it has been decided to attibuted the zero magnitude
to the star Vega (α Lyrae, the brightest star of the Lyra constellation).
On the other hand, the natural scientific measure of the relative brightness of stars
comes from their flux. A relation between magnitudes and fluxes (expressed in physical
units) needs therefore to be established:
m1 − m2 = 2.5 log(f2 /f1 ) (1.4)
According to this relation, if star 1 is brighter than star 2, f1 is larger than f2 , and m1
is lower than m2 . The presence of the logarithm comes from the logarithmic nature of
the perception of brightness by the naked eye. The above relation constitutes therefore
the tranformation of a linear scale (fluxes) into a logarithmic scale (brightness perception).
When one has to estimate the brightness of a star, it should be kept in mind that
there is a strong difference between apparent and absolute quantities. A star that appears
to be faint (large magnitude) could be intrinsically very luminous, but located at a large
distance. In order to lift this degeneracy between these distance and luminosity effects,
absolute magnitudes have been introduced by astronomers. The absolute magnitude of a
star is defined to be the magnitude of that star if it was located at a distance of 10 pc. As
all stars are now considered at the same distance, the absolute magnitude scale provides
a valuable basis to compare intrinsic stellar brightnesses. A relation has therefore been
established between apparent and absolute magnitudes, as a function of the distance to
the star.
Starting from Eq. 1.4, let us consider the following quantities:

- m1 = m , the apparent magnitude of a star
- m2 = M , the absolute magnitude of the same star
- f1 = f , the flux received from the star at a distance d expressed in pc
- f2 = F , the flux received from the star at a standard distance D (= 10 pc)
this leads to
m − M = 2.5 log(F/f)
A flux is defined as a power per surface unit (in cgs units, it gives erg s−1 cm−2 ). If
we note the luminous power of the star as L, the fluxes measured at distances d and D,
respectively, are
L
f=
4 π d2
1.1. General concepts 9
and
L
F=
4 π D2
Introducing these quantities in the magnitude relation,
L/4 π D2
m − M = 2.5 log
L/4 π d2
1/D2
= 2.5 log
1/d2
d2
= 2.5 log 2
D
d
= 5 log
D
Considering a standard distance (D) of 10 pc, we finally obtain:
m − M = 5 log d − 5 (1.5)
The quantity (m - M) is often called the distance modulus of the object.
It is important to note that apparent and absolute magnitudes refer generally to quan-
tities in the visual domain (i.e. in the V photometric band). Such magnitudes are therefore
not related to the total energy emitted by the stars. If the total energy emitted by a star
is taken into account, it is expressed in the form of a bolometric magnitude. The relation
between a bolometric and a non-bolometric quantity can be expressed by the following
equation:
BC = mbol − m (1.6)
where BC is called the bolometric correction expressing (in magnitude formalism) the
contribution of the luminous energy produced by the star that is not included in the energy
band used to determine the magnitude m. Generally, the magnitude m is expressed in the
V photometric band, approximately centered in the bandpass of the naked eye: we talk
about visual magnitudes. However, many different photometric bands have been defined
by astronomers. Note that the bolometric correction is expressed by a negative number.
Beside the effect of geometrical dilution of the luminous energy as a function of dis-
tance, there is another phenomenon that affects significantly the flux measured from an
astronomical object: the extinction. The extinction is defined as the combination of ab-
sorption and scattering of stellar photons along their optical path, because of the presence
of intertellar material along the line of sight. This effect causes a dimming that can also
be expressed in magnitude:
m − M = 5 log d − 5 + A (1.7)
where A expresses the extinction, and it is intimately dependent on the waveband consid-
ered. Note that A is preceded by a + sign, as the effect of extinction is a dimming of the
star, i.e. an increase of the apparent magnitude.
Another concept worth to be defined, in relation with magnitudes, is that of color. The
color is simply the difference between two magnitudes measured in two different energy
bands. For instance, the B–V color expresses a ratio between fluxes measured, respectively,
in the B and V bands. The color is an indicator of the slope of the stellar spectrum, which
itself translates into an effective temperature. The bandpasses of different usual standard
photometric bands are illustrated in Fig. 1.2. These bands are frequently used to express
colors of stars.
Figure 1.2: Examples of standard photometric bands frequently used in astronomical

photometry.
Application 1. The Sun has an effective temperature of about 5800 K. What would be
its effective temperature if it was two times brighter than now? We consider here that its
radius would be the same.
Application 2. Betelgeuse and Rigel are the two brightest stars in Orion. Do we expect
their surface colors to be different, knowing that their respective surface temperatures are
about 3600 K and 10000 K?
Application 3. What would be the minimum effective temperature of a star presenting
its emission peak at most in the near ultraviolet domain, typically at a wavelength of
2500 Å ?
Application 4. In the Gemini constellation, the two brightest stars are Castor and Pol-
lux. The apparent magnitude of Castor is 1.96. Knowing that the luminosity of Castor is
47 per cent that of Pollux, determine the apparent magnitude of the latter.
Application 5. Show that the values for the apparent magnitude of the Sun (-26.74) and
for its absolute magnitude (+4.83) are consistent with a distance to the Sun of 1 AU. One
AU is equivalent to about 150 millions of km, and one parsec is equal to 3.086 1018 cm.
Application 6. If Polaris is located at a distance of about 132 pc, determine its absolute
magnitude. The apparent magnitude for this star is 1.97.
Application 7. Sirius A is the brightest star in our night Sky. It is located at a distance
of 2.64 pc and its absolute magnitude is 1.47. Calculate its apparent magnitude, and com-
ment this value regarding the numbers related to Polaris in the previous application.
Question 1. The bolometric correction in the case of the Sun is especially weak. Is it a
coincidence?
1.2. Stellar objects 11
1.2 Stellar objects

Shortly, the Galaxy is populated by stellar objects of different kinds, and between these
objects one finds what is called the interstellar medium (ISM). Even though the ISM is of
crucial importance in the context of astrochemistry, neglecting stellar objects in such an
environment would not make any sense. Stars are indeed the main providers of photons
likely to play a role in interstellar chemistry. On the other hand, stars in formation
and some circumstellar environments constitute priviledged places where many chemical
processes are taking place. In order to clarify our view of the content of the Milky Way,
a brief introduction to stellar astrophysics is needed.
Figure 1.3: Schematic view of the Milky Way.
A schematic view of the Milky Way is presented in Fig. 1.3. Most stars are located in
a plane, with a denser population close to the central part of the Galaxy. In the Galactic
plane, stars are not uniformly distributed, but are concentrated in spiral arms. Most of the
Galactic material is located in this plane, with only a small fraction of the stellar population
evolving in a sphere called the halo. The latter stars are mainly old stars concentrated
in globular clusters that are, as a first approximation, of negligible importance from the
astrochemical point of view. For stellar astrophysics and astrochemical purposes, the most
important regions are indeed concentrated in the Galactic plane.
1.2.1 Hertzsprung-Russell diagram

When astronomers started to study stellar populations, they early noticed that stars don’t
seem to be randomly distributed as a function of their fundamental properties. The two
more obvious characteristics of a star, at least from an observational point of view, are its
luminosity and its color. The first one refers to the amount of energy released by stars in a
given time interval (the power radiated), and the second one is a direct observational fact
(some stars appear bluer and other seem redder). In addition, as emphasized in Sect. 1.1.3,
Figure 1.4: Hertzsprung-Russell diagram of 22000 stars from the Hipparcos Catalogue
together with 1000 low-luminosity stars (red and white dwarfs) from the Gliese Catalogue
of Nearby Stars.
the color of a star is directly related to its effective temperature. Its is also interesting to
note that these two quantities can be discussed in the context of the fundamental concept
of black bodies.
If one plots the luminosity of stars as a function of their color, it is striking to note that
the points are not randomly distributed. It appears first that most stars are concentrated in
a particular region of the diagram: a diagonal going from the upper-left (hot and luminous)
to the lower-right (cooler and less bright). Such a finding was initially made circa 1910 by
Ejnar Hertzsprung and Henry Norris Russell, and they named that particular region where
most stars are located the main sequence. Beside that, many stars are located in other
particular regions, and the advances in stellar astrophysics showed that these regions were
corresponding to different evolutionary stages of the stars. Such a Hertzsprung-Russell
diagram is shown in Fig. 1.4. The abscissa is defined as the B–V color (bottom scale),
translating into an effective temperature scale (top scale). It should be emphasized that
in such a diagram the temperature increases from the right to the left: hotter stars are
on the left. The luminosity of the stars (in solar luminosity units) is increasing from the
bottom to the top of the diagram, and on the right its equivalence expressed in absolute
magnitude is represented.
The stars located on the main-sequence correspond to the evolutionary stage where
the central burning of hydrogen is taking place. Most stars appear in this part of the
diagram because it corresponds to the longer stage of stellar evolution crossed by stars1 .
Other regions such as the giant branch, or the supergiant branch, correspond to evolved
evolutionary stages, which are significantly shorter than the main-sequence, hence the
lower number of stars located there.
1.2.2 Spectral classification

Once astronomers noticed that stars displayed somewhat different properties, they man-
aged to sort them in various categories. Spectral classification criteria have therefore been
established. Beside the continuum emission following quite well Planck’s function, stars
also present spectral features such as absorption and emission lines. These lines are at-
tributed to different chemical elements in various ionized forms. The line strengths depend
intimately on the abundances of chemical elements, and also on the physical properties of
the stellar atmospheres where they are formed (temperature, local gravity...).
Historically, a first stellar classification was established circa 1860 by Angelo Secchi.
The standard spectral classification (the so-called Harvard classification system) was de-
veloped later on, following the work of Antonia Maury and Annie Jump Cannon (among
others). This classification relies on the presence, and on the intensities, of spectral lines
present in the optical spectra of stars. The ionization stage and population levels are
directly related to the effective temperature of stars. This fact allows a direct connection
between spectral classification and effective temperature. The relation between spectral
classification and temperature was discovered and explained by the astronomer Cecilia
Payne in the 1920’s. As a consequence, the investigation of the spectral features in stellar
atmospheres leads to a determination of one of its most fundamental parameters. Histor-
ically, letters have been assigned to spectral types of stars. The standard classification
scheme is summarized in Table 1.1.
Beside the spectral types attributed to stars, it should also be noted that stars of the
same spectral type may differ significantly. An important origin for such differences comes
from the luminosity class of stars. These classes correspond to different evolutionary stages
crossed by the stars during they lifetime (see Sect. 1.2.4). Typically, stars may belong to
the class of main-sequence stars, to the class of giants, or to the class of supergiants, with
increasing luminosity for a given spectral type.
The spectral types given in Table 1.1 are subdivided in sub-categories, called sub-types,
generally ranging from 0 to 9. Stars with the hotter atmospheres are said to be early-type
stars, and star with cooler atmospheres are said to be late-type stars. It is important to
note that the stars producing the stronger UV field likely to play a role in interstellar
photochemistry are early-type stars, belonging the O and B spectral types. The earliest
(hotter, more massive, more luminous...) stars are often put together in the category of
OB stars, including O-type stars and B-type stars not later than B2.
1
This statement assumes that only stars where inner nuclear reactions are occurring are considered.
Stellar remnants such as white dwarfs, neutron stars or black hole are not concerned by this criterion.
Table 1.1: Standard spectral classification of stars.

Spectral Spectral lines Typical effective
type temperature (K)
O He ii and He i lines, H lines, N iii,... > 25000
B He i, Si iii, O ii, increasing H lines 10000-25000
A H lines at their maximum, ionised metallic 7500–10000
lines such as Fe ii and Mg ii
F Decreasing H lines, strong lines from 6000–7500
Fe ii and Mg ii
G Ca ii, strong lines from Fe i and Mg i 5000–6000
K Ca ii, neutral metallic lines, 3500–5000
molecules such as CN and CH
M Strong neutral atom lines, molecules 2500–3500
such as TiO, VO, AlO...
Things are not so simple, as some stars seem to be at odd with the stardard classifica-
tion. Other categories had therefore to be established. One category worth considering in
this context is that of Wolf-Rayet stars: these are evolved OB-type stars with untypical
abundances, they are very bright, and they reject large amounts of stellar material in the
interstellar medium (Crowther 2007). Such stars are generally noted WR stars.
The hottest stars constitute a minority in the galactic stellar population. The Milky
Way should contain about 100 billion stars, among which not more than 100000 stars as
hot as OB and WR stars. Most stars are indeed cooler stars. For instance, the Sun is a
G2 star, still on the main-sequence.
1.2.3 Stellar formation

The densest regions in the interstellar medium are molecular clouds. These clouds consti-
tute the starting material from which stars are formed. Even though typical dynamical
time scales in the interstellar medium can be quite long (with respect to the life-time of
human beings, for instance), molecular clouds are not eternally steady. They can undergo
instabilities, for instance triggerred by supernovae in the vicinity, leading to its collapse.
Once a dense interstellar cloud undergoes a gravitational collapse, it shrinks and becomes
denser, leading to the formation of a new star. Such a process is summarized in Fig. 1.5,
where panels (a) to (f) illustrate various steps, from the initial molecular cloud to the
formation of a stellar system, according to the standard scheme based mainly on infrared
measurements (see Bottinelli 2006):
a. At the beginning of the collapse of the initial molecular cloud, the density increases
by at least two orders of magnitude. The cloud can reach some kind of metastable
state where gravity and thermal pressure equilibrate.
b. When gravity overwhelms the thermal pressure exerted by the gas, the pre-stellar
core goes through a phase of ‘free fall’ collapse, producing a protostar in accretion,
surrounded by a large dust circumstellar envelope more masssive than the protostar
itself (a so-called Class 0 object). Such an object emits in the radio domain (∼ cm),
with a typical black body temperature of 10-30 K.
Figure 1.5: Schematic view of the gravitational collapse of a molecular cloud leading to
the formation of a stellar system. On the left: illustration of the various stages crossed
by the pre-stellar object. On the right: schematic emission spectra corresponding to the
different evolution steps. The figure is taken from Bottinelli (2006).
c. An evolved accreting protostar has been formed, surrounded by a massive disk and
a collapsing residual envelope whose mass is now smaller than that of the protostar
(a so-salled Class I object). This object is hotter than a Class 0, and it emits
predominantly in the infrared, with an additional infrared excess due to the presence
of the disk.
d. In a Class II object, the envelope has (almost) disappeared, but the disk is well
developed (very large infrared excess).
e. The Class III phase is characterized by an infrared excess that is much weaker than in
the previous stage. This results from the ongoing accretion and from the formation
of planetesimals that lead to the dissipation of the disk.
f. The core of the central star has reached a temperature that is high enough to start
the nuclear burning of hydrogen, and the star is now on the main-sequence of the
evolutionary scheme of stars. The disk ends its transformation into a planetary
system.
It should be emphasized that the stellar formation scenario briefly discussed above
concerns mainly low and intermediate mass stars. In the case of the most massive stars,
the formation process is most probably more complicated and is still a matter of debate.
1.2.4 Stellar nucleosynthesis and evolution

Once a star has reached the main-sequence, it enters its evolutionary stage where the
central burning of hydrogen takes place. The result of this process is the synthesis of
helium. This consitutes the energy source of the star, responsible for its emission of
radiation. Such nuclear reaction can only take place provided the central temperature
of the star has reached a given threshold value, typically of the order of 1–2 107 K. Such
temperatures are reached thanks to the gravitational collapse of the material constituting
the star. During the phase of central hydrogen burning, the energy is released leading to an
expansion of the gas. In short, this expansion compensates the gravitational collapse. The
star has reached some kind of equilibrium, at least as far as the central nuclear reactions
are going on.
In its simplest form, the proton-proton chain reaction is the following:
1
1H +11 H → 21 H + e + ν
2
1H +11 H → 32 He + γ
3
2 He +32 He → 42 He + 211 H
The first reaction transforms two protons into a deuterium nucleus, with the release of
a positron (e) and a neutrino (ν). It is followed by the annihilation of a positron produced
by the transformation of a proton into a neutron. The second reaction leads to a 32 He
helium isotope. Finally, the third reaction involves two 32 He helium isotopes and leads to
the synthesis of a 42 He nucleus, and restitutes two protons likely to be involved again in
the first and second reactions.
If chemical elements heavier than hydrogen and helium are available, another process
can be at work in stars: the CNO process. The steps of the CNO process are the following:
12
6 C +11 H → 13
7 N +γ
13 13
7 N → 6 C +e+ν
13
6 C +11 H → 14
7 N +γ
14
7 N +11 H → 15
8 O +γ
15 15
8 O → 7 N +e+ν
15
7 N +11 H → 12
6 C + 42 He
Globally, this process transforms four protons into a helium nucleus. In addition, we can
also consider the following reactions leading to the formation of helium:
15
7 N +11 H → 16
8 O +γ
16
8 O +11 H → 17
9 F +γ
17 17
9 F → 8 O +e+ν
17
8 O +11 H → 14
7 N + 42 He
The nitrogen that is produced in the last step can then participate again in the CNO
process through the capture of a proton, and so on. In such a process, CNO elements
are almost used as ’catalysts’ as they are restituted simultaneously with the synthesis of
helium. At the end of the phase of central burning of hydrogen, the core of the star is
made of helium, surrounded by hydrogen. The cumulated abundances of CNO elements is
not altered by such a process, even though the individual abundances of the three elements
are slightly altered.
Once the star has exhausted the hydrogen located in its central part it leaves the main-
sequence (central p-p and CNO processes stop). The core of the star is mainly made of
helium. As the central nuclear reactions are turned off, the gravitational collapse takes
the advantage. The contraction leads to a rise of the central temperature. Helium burning
can occur in the deep layers of the stellar core once the temperature reaches about 108 K.
The central helium burning step is significantly shorter than the central hydrogen burning
stage. Typically, in the case of solar-type stars, the lifetime on the main-sequence is of the
order of 10 billion years, but the helium burning stage duration is less than 1 billion year.
The triggerring of these helium fusion reactions leads to another expansion of the central
gas, and the collapse stops again. At such high temperatures, the following equilibrium
can be established:
4 4 8
2 He +2 He
4 Be
Even though this equilibrium is significantly shifted to the left (about 1 Be nucleus for
109 He nucleus), this is sufficient to trigger the formation of carbon though the following
reaction:
8 4 12
4 Be +2 He → 6 C + γ
Such a process is called the triple-α process, as is it acting as the addition of three α
particles (helium nuclei). Other reactions are also participating in the burning of helium.
Mainly, we can consider:
12 4 16
6 C +2 He → 8 O + γ
16
8 O +42 He → 20
10 Ne +γ
20
10 Ne +42 He → 24
12 Mg +γ
Nitrogen is also able to capture helium nuclei, and is therefore rapidly destroyed during
the helium burning phase:
14 4 18
7 N +2 He → 8 O + e + ν
18
8 O +42 He → 22
10 Ne +γ
Once the temperature has reached values as high as 5 to 10 108 K, the fusion of carbon
nuclei is likely to start. Here, several channels are possible:
12
6 C +12
6 C →
24
12 Mg + γ
12
6 C +12
6 C →
23
12 Mg + n
12
6 C +12
6 C →
23 1
11 Na +1 H
12
6 C +12
6 C →
20 4
10 Ne +2 He
Among these processes, the last one is the most probable. The burning of carbon yields
therefore essentially neon. Neon nuclei can undergo photodisintegration leading to the
production of oxygen:
20 16 4
10 Ne + γ →8 O +2 He
The consequence of this process is that the burning of carbon is always a source of oxygen
in stars.
The burning of oxygen can start provided the temperature has reached a temperature
of the order of 109 K:
16
8 O +16
8 O →
32
16 S + γ
16
8 O +16
8 O →
31
16 S + n
16
8 O +16
8 O →
31 1
15 P +1 H
16
8 O +16
8 O →
28 4
14 Si +2 He
Once again, the channel leading to the formation of silicon (with the release of an α
particle) is the most probable process.
In the case of silicon, there is a competition between captures of α particles and photo-
disintegration leading to the restitution of α particles. As a result, some kind of equilibrium
slightly shifted to the right establishes:
28
14 Si +42 He
32
16 S + γ
32
16 S +42 He
36
18 Ar + γ
36
18 Ar +42 He
40
20 Ca + γ
40
20 Ca +42 He
44
22 Ti + γ
44
22 Ti +42 He
48
24 Cr + γ
48
24 Cr +42 He
52
26 Fe + γ
52
26 Fe +42 He
56
28 Ni + γ
The situation is even much more complicated, as neutron captures and other disinti-
grations are occurring. All these processes are often referred to as the silicon burning,
and it leads to central layers of the stellar core very rich in iron and nickel. Because
of energetic considerations, nuclear fusion reactions will not lead to heavier nuclei than
iron and nickel. These elements correspond indeed to the most stable nuclei: any fusion
reaction leading to their formation is exothermal, and any fusion reaction involving these
elements is endothermal. On the other hand, any fission reaction starting with these
elements is endothermal, and any fission reaction leading to these elements is exother-
mal. This is illustrated in Fig. 1.6 where the average binding energy per nuclei is plotted
as a function of the mass number of nuclei. The maximum stability is reached close to iron.
Figure 1.6: Plot of the average binding energy per nuclei as a function of the mass number
of nuclei. The maximum corresponds to iron and nickel.
Depending on the mass of the star, the gravitational collapse can lead to various central
temperatures: the highest masses lead to the highest central temperatures. The triggering
of the subsequent nuclear fusion phases is intimately dependent on the central temperature
that can be reached, and therefore on the initial mass of the star. As a result, only the
most massive stars are able to lead to the nucleosynthesis of iron and nickel. The evolution
of stars will be significnatly different depending on their initial mass (see Fig. 1.7). The
less massive stars are indeed unable to reach central temperatures high enough to trigger
the most advanced fusion reactions. Once a star has stopped its central hydrogen burning
stage, it leaves the main-sequence and reaches the red giant stage, where hydrogen fusion
takes place in a thin shell around the core.
In the case of stars with inital mass between 0.6 and about 10 solar mass, after com-
pletion of the central helium burning, the helium combustion will continue in a thin
shell surrounding the carbon-oxygen core. Beside that, the interface between the he-
lium layer and the upper layer rich in hydrogen can reach a temperature high enough to
allow shell hydrogen burning. This particular evolution stage corresponds to a location of
the Hertzsprung-Russell diagram called the asymptotic giant branch (AGB). At this stage,
some elements heavier than carbon can be formed through slow neutron capture reactions
(s-process), providing therefore a source a new chemical elements. Typical neutron fluxes
involved in this process are of the order of 105 to 1011 neutrons cm−2 s−1. In this-process,
a stable isotope captures a neutron, but the radioactive isotope that results decays to its
stable daughter before the next neutron is captured. These decays consist of β-decays:
the conversion of a neutron into a proton, with the ejection of an electron.
For the most massive stars (> 10 solar masses), immediately after the core collapse,
the outer shells of the star that are fastly expanding are submitted to the bombardment of
an intense neutron flux (∼ 1022 neutrons cm−2 s−1) in a high temperature environment. In
such circumstances, the neutron capture is much faster than the β-decay and the process
runs fastly up the scale of neutron number per nucleus. This process (called the r-process)
leads fastly to neutron rich nuclides, that will decay later on to yield stable nuclides.
Both types of neutron capture processes (r- and s-processes), in addition to stellar
fusion reactions, contribute significantly to the nucleosynthesis of many nuclides. In addi-
tion, it should be noted that some elements were not synthesized in stellar environments,
and rather emerged from the primordial nucleosynthesis that took place just a few mo-
ments after the Big Bang. This primordial process led to the formation of deuterium (D),
of the helium isotopes 3 He and 4 He, and the lithium isotopes 6 Li and 7 Li.
Figure 1.7: Summary of the main scenarios followed by stars along their evolution as a
function of their mass.
1.3. Interstellar medium 21
At the evolution stage where the nuclear fuel at the center is exhausted (whatever the
initial mass of the star and whatever the highest combustion level reached in the core),
gravitational forces will prevail and the collapse of the central part of the star will go on.
The next step will once again depend on the stellar mass (see Fig. 1.7):
- The core of the lighter stellar objects (between about 0.6 and 10 solar masses) will
become a white dwarf. For instance, for a star with the mass of the Sun, this stellar
remnant will have a mass of the order of 0.6 solar mass. The stellar material of
the white dwarf is compressed within a volume equivalent to that of the Earth. The
collapse stops when the gravitational forces are compensated by the degeneracy elec-
tron pressure, as a consequence of Pauli exclusion principle. The degeneracy electron
pressure is strong enough to compensate gravitational forces provided the mass of the
stellar core is not larger than about 1.4 solar mass, i.e. the so-called Chandrasekhar
limit. With no fuel left to burn, the white dwarf radiates its remaining heat into
space for billions of years. Typically, the density of a white dwarf material is about
106 g cm−3 . Simultaneously with the collapse of the stellar core, the outer shells of
the star expand and a planetary nebula is formed.
- For massive stars (typically larger than 10 solar masses), even the degeneracy electron
pressure is not enough to compensate the strong gravitational forces, and the collapse
goes on. The increased pressure in the collapsing stellar core forces electron capture,
and protons are converted into neutrons. The new state of the stellar material is
then governed once again by a huge pressure, this time caused by closely confined
neutrons. The typical size of such a degenerate stellar core is 10-20 km, and it is
called a neutron star. The typical density of the material constituting a neutron
star is about 1014 –1015 g cm−3 . For the most massive stars, the collapse can even go
further and the stellar remnant reaches the stage of black hole, provided the mass
of the core is larger than a critical value of the order of 2.5-3.5 solar mass (this
limit is still not accurately constrained). The typical density of a black hole may
reach values of the order of 1025 –1027 g cm−3 When the stellar core shrinks up to
the stage of neutron star or black hole, the outer shells are violently expelled from
the star: this is the supernova phenomenon. Such an explosion sends large amounts
of nucleosynthesized nuclides in the interstellar medium. In addition, the explosion
causes a huge rejection of neutrons that are captured by elements even heavier than
iron, producing unstable heavy elements, that relax through β-decay. As a result,
many elements heavier than iron and nickel are produced, populating the interstellar
medium with chemical elements that will be integrated in the material available for
the next generation of stars, and so on.
To conclude this section devoted to stellar evolution, it should be emphasized once
again that the crucial parameter is the stellar mass. The mass of the star is the parameter
that will decide the fate of stars, and constrain their capabilities to synthesize chemical
elements that will be rejected in the interstellar environment at the end of their evolution.
As a result, stars are responsible for the availability of the starting material that will be
involved in the interstellar chemistry discussed in Chapters 2 and 3.
1.3 Interstellar medium

The volume between stars in a galaxy is not empty. Even though interstellar material is
not very dense, it is populated by a significant amount of particles. Typically, one can
find:
- neutral atoms: mainly H and He, with significant amounts of C, N, O elements. It is

important to note that D abundance is not negligeable with respect to that of main
isotopes of elements such as C, N, O, Si and P, for instance.
- ions: once again, mainly H+ and cations of other abundant elements. Cations are
the dominating ions in interstellar environments.
- electrons: as the result of ionization of many chemical elements in most astrophysical
environments, free electrons are significantly abundant.
- small size molecules: the two main molecules are H2 and CO, but other small size
molecules are present, mainly in molecular clouds.
- larger molecules: mainly, polycyclic aromatic hydrocarbons (PAH, see Chapter 3).
PAH have been found to populate most places investigated so far in galaxies.
- dust particles: small particles with typical sizes of a few tens of Å up to several µm
are also ubiquitous in interstellar environments. They play a crucial role in chemical
processes leading to the formation of many molecules (see Chapter 2).
Depending on the physical conditions of the different regions in the interstellar medium,
different environments can be distinguished:
- H II regions. Because of the presence of bright early-type stars (typically, O and
B-type stars), a very strong UV radiation field is present in many locations of our
Galaxy. This UV radiation field is responsible for the ionization of atomic hydrogen,
therefore present in the form of protons. In usual spectroscopic notations, hydrogen is
noted H I in its neutral form and H II in its ionized form. The regions where hydrogen
is dominantly ionized are called H II regions. Such environments are characterized
by typical temperatures of about 10000 K, and are therefore not priviledged places
to find molecules.
- H I regions. In places where the UV radiation field is not so dominant, hydrogen is
predominantly found in its neutral form. Typically, the temperature of H I regions is
of the order of 100 K. Such regions are mainly detected through the neutral hydrogen
emission line at a wavelength of 21 cm.
- Molecular clouds Some regions are much colder and denser than H I regions, and
are populated by many molecules. In molecular clouds, the typical temperature is of
the order of 10–20 K. Such regions are dominantly populated by molecular hydrogen
(H2 ) and by carbon monoxide (CO), the two most abundant molecules in the Uni-
verse. Such environments constitute privileged regions to search for molecules and to
investigate interstellar chemistry. Many molecules identified so far in astrophysical
environments have been detected in molecular clouds, even though other environ-
ments such as protostellar and circumstellar environments are also very valuable
targets for such investigations.
1.4 Solar system

1.4.1 A few definitions
In order to clarify the content of the Solar ssytem (see Fig. 1.8 for a view of its main com-
ponents), in agreement with the conventions adopted by the International Astronomical
Union, a few definitions may be helpful.
1.4. Solar system 23
- A planetary system consists of the various non-stellar objects orbiting a star such
as planets, dwarf planets, moons, asteroids, meteoroids, comets, and cosmic dust.
- A planet (from Greek meaning ”wandering star”) is a celestial body orbiting a

star or stellar remnant that is massive enough to be rounded by its own gravity, is
not massive enough to cause thermonuclear fusion, and has cleared its neighbouring
region of planetesimals.
- A dwarf planet is a celestial body orbiting the Sun that is massive enough to be
spherical as a result of its own gravity but has not cleared its neighboring region of
planetesimals and is not a satellite.
- A natural satellite or moon is a celestial body that orbits a planet or a dwarf

planet.
- Planetesimals are solid objects thought to exist in protoplanetary disks and in

debris disks.
- A meteoroid is a sand- to boulder-sized particle of debris in the Solar System.
- A comet is a small icy residual of the formation of the Solar system, evolving on
large orbits.
- A meteorite is a natural object originating in outer space that survives impact with
the Earth’s surface.
- Small Solar System body (SSSB) is a term defined in 2006 by the International
Astronomical Union to describe objects in the Solar System that are neither planets
nor dwarf planets, nor satellites of a planet or dwarf planet.
The outer region of the plane where most planetary orbits are located is called the
Kuiper belt: an annulus populated by large amounts of small bodies mostly made of frozen
volatiles. Pluto and other dwarf planets are located there. It extends from the orbit of
neptune (∼ 30 AU) up to about 55 AU. In addition, the Solar system is surrounded by
a huge sphere where countless comets are believed to evolve: Oort’s Cloud. This cloud
extends up to distances of tens of thousands AU from the Sun.
1.4.2 Planets
Our Solar system include 8 planets, that can be subdivided into two categories:
- Terrestrials: this category includes the planets characterized by the higher den-
sities, also called rocky planets. These objects are mainly made of a solid body,
surrounded by a gaseous atmosphere. The members of this class in our Solar system
are: Mercury, Venus, the Earth and Mars. The orbits of these planets are located in
the inner part of the Solar system, i.e. they are closer to the Sun.
- Gas giants: this class includes the biggest planets, made mostly of hydrogen and
helium. As a consequence, their density is quite low with respect to terrestrial
planets. They are also called Jovian planets. Jupiter, Saturn, Uranus and Neptune
belong to this category. Their orbits are larger than those of terrestrial planets.
Figure 1.8: Schematic representation of the main components of our Solar system. The
relative sizes of the Sun and planets are on scale, but their separations are not. The
distinction between the inner part and the outer part of the Solar system is obvious,
with the inner part populated by terrestrial planets. Note the presence of Ceres, a SSSB
located in the asteroid belt whose orbit lies between that of Mars and Jupiter. This belt
is made of planetesimals whose accretion to form a planet has been inhibited because of
the perturbating influence of Jupiter.
The orbits of all planets are approximately confined in the same plane, called the
ecliptic plane. As noted above, there is a noticeable distinction between terrestrials with
inner orbits, and gas giants with outer orbits. This is believed to come from the formation
of the Solar system, when the lighter elements were rejected in the outer regions because
of the action of the strong radiation field from the luminous young Sun.
It is important to note that modern astrophysics is now further investigating planetary
physics, including the physics of the formation of planetary systems, thanks to the discov-
ery of several hundreds of planets orbiting other stars, i.e. exoplanets. Even though almost
all exoplanets discovered so far are gas giants, significant progressses are being made in
order to discover Earth-like planets.
1.4.3 Comets
Comets are small bodies whose nucleus is mostly made of ice, dust, and small rocky par-
ticles. Sometimes, the orbit of a comet brings it closer to the Sun. In such circumstances,
the heat from the Sun causes a fraction of its material to sublimate, producing what is
called the coma of the comet. In addition, the interaction of the solar wind (made of fast
particles) with the comet produces a tail made of ionized gas that is always oriented away
from the Sun, and an antitail made of dust whose orientation is slightly different from that
of the tail.
As residual material formed at the epoch of the formation of the Solar system, comets
are important objects worth being studied in detail. This is especially important in the
sense that prebiotic Earth has most probably known a period of intense cometary bom-
bardment, that may have played a significant role in its early chemical evolution.
1.4. Solar system 25
Figure 1.9: Schematic repre-

sentation of the Oort cloud
known as the main reservoir
of comets in our Solar sys-
tem. The distance scale is ex-
pressed in AU.
1.4.4 Asteroids
Among SSSBs, asteroids deserve a particular attention because they constitute very in-
teresting objects to study mineralogy in interplanetary environments. In addition, such
objects are scrutinized in order prevent potential impacts with our planet likely to cause
important damages. As an example, the impact of a large asteroid close to the present
Yucatan peninsula, about 65 million years ago (end of Cretaceous period), is likely to have
significantly contributed to the mass extinction fatal notably to dinosaurs. The residual of
this event is the Chicxulub crater with a diameter of about 180 km. According to recent
estimates, the size of the body that collided the Earth was 10–20 km, and the energy re-
leased in the impact was of the order of several 109 times the energy of the Hiroshima bomb.
Figure 1.10: Location of the

crater attributed to the im-
pact at the end of the Creta-
ceous period.
Application 8. Considering an asteroid with an incoming velocity of the order 5 km/s,

and a radius of the order of 100 m (assumed to be spherical, for simplicity), what would
be the energy released by the impact of such an object, assuming it is homogeneous with
a density of 5? Express this energy in equivalent of Megaton of TNT (∼ 4.184 1015 J).
Application 9. Considering an asteroid with an incoming velocity of the order 7 km/s

and a density of 4, what would be its diameter to lead to the release of an amout of energy
equal to that of the Hiroshima bomb (∼ 63 1012 J)? The object is assumed to be spherical
and homogeneous, for the sake of simplicity.
Chapter 2
Chemical processes
The interstellar medium is not empty. It is populated by matter in different forms: neutral
atoms, atomic ions, electrons, molecules (neutral and ionic), and dust particles. From the
physical chemistry point of view, the ISM can be seen as a gas phase volume, where gas
phase kinetics can be applied. In such circumstances, it is obvious that the low particle
densities constitute an issue. In addition, interstellar temperatures can be very low, which
is not in favor of a fast chemical kinetics. As a consequence, one can expect reaction
time scales to be very long and it is a priori difficult to envisage the formation of complex
molecules. However, the current census of interstellar molecules provides strong evidence
that efficient chemical processes are at work in space, despite the a priori unfavorable
gas phase kinetics. This apparent contradiction is lifted when the catalytic effect of dust
grains is taken into account in astrochemical models. Dust particles constitute indeed
a solid-state component in the ISM, whose surface can act as catalyst responsible for a
considerable activation of interstellar chemical kinetics. Both approaches – gas phase and
grain surface – are discussed below before being combined in order to provide a more
complete view of the physical chemistry that is governing astrochemical processes.
2.1 Elementary gas-phase processes

During a chemical – or photochemical – process, chemical species are destroyed and other
chemical species are formed. We can consider formation and destruction rates for various
kinds of reactions:
- unimolecular reactions: A → C
d n(A) d n(C)
− = k n(A) =
dt dt
- bimolecular reactions: A + B → C
d n(A) d n(C)
− = k n(A) n(B) =
dt dt
- t(h)ermolecular reactions: A + B + M → C + M
d n(A) d n(C)
− = k n(A) n(B) n(M) =
dt dt
27
28 Chemical processes
In these relations, the temporal derivatives are reaction rates, and the factors (k) pre-
ceding the numerical densities are reaction rate coefficients, also called kinetic constants.
The reaction rates are always expressed in cm−3 s−1 , whilst the units of reaction rate
coefficients depend on the type of reaction. For instance, for bimolecular reactions, k is
expressed in cm3 s−1 , while for unimolecular reactions it is expressed in s−1 .
It should be emphasized that when one is dealing with chemical kinetics, two crucial
factors are essential:
- Density. Typically, densities expressed in particles per cm−3 are quite low, with
respect to terrestrial standards. To give an idea, diffuse clouds and dense molecu-
lar clouds in the interstellar medium are characterized by densities of the order of
102 cm−3 and 104 to 106 cm−3 respectively, and the densest circumstellar clouds have
densities of the order of 1011 -1012 cm−3 ; but at sea level, the atmosphere of the Earth
has a density of the order of 2.5 × 1019 cm−3 . The immediate consequence is a low
probability of interaction between reaction partners, translating into low reaction
rates.
- Temperature. In molecular clouds, i.e. the densest interstellar region where most
interstellar molecules have been identified, temperatures may be as low as a few K,
up to a few K. Other parts of the interstellar medium where elements are ionized may
be characterized by much higher temperatures, but such environments are permeated
by strong radiation fields that compromise at least the existence of small molecules.
In short, the only places in the interstellar medium which are significantly populated
by molecules are very cold media. Gas phase kinetics, in the context of the perfect
gas approximation, teaches us that reaction rate coefficients (k) depend intimately
on the temperature
(T) according to the following proportionality relation: k ∝
1/2 Ea
T exp − kB T . In this relation1 , the first factor is related to the properties of
the gas and to their consequences on the interaction between reaction partners, and
the second factor expresses the need to cross an activation barrier Ea . These two
factors clearly show that the low temperature of molecular clouds leads to low values
for k, which is not in favor of fast chemical kinetics.
This points to an especially strong constraint for astrochemistry, at least in inter-

stellar environments: one is dealing with slow kinetics, and this is a priori a good reason
to consider that the interstellar molecular complexity will suffer from stringent limitations.
In the context of astrochemical studies, it is also not relevant to establish complex

reaction mechanisms or to derive heavy analytical expressions for reaction rate equations.
Even when simple situations are considered, with only a small number of first or second
order processes, analytical relations become rapidly too complex to be manipulated. The
approach adopted consists therefore to derive numerical solutions for systems of equations
involving many elementary processes, i.e. processes for which there is no identified interme-
diate process, occurring consecutively and/or in parallel. The complete problem consisting
in determining abundances of many chemical species as a function of time is therefore split
into a large number of simple problems, for which the solutions can be derived much more
easily. For a given population of chemical species, a census of elementary processes has
1
This kind of proportionality relation is of frequent use in parametric form to express reaction rate
coefficients, as described for most elementary processes below. In practise, some differences are however
expected because of deviations from the perfect gas case.
2.1. Elementary gas-phase processes 29
therefore to be made, with a priori knowledge of the adequate rate coefficients. Several
kinds of elementary processes of common use in astrochemistry are discussed below.
2.1.1 Photodissociation
In the ISM, the dominant destruction agent for small molecules is the far ultraviolet (FUV)
radiation field from early-type stars. These stars are not the most abundant ones, but their
brightness in the ultraviolet and visible domains has a strong impact on their environment
(see De Becker 2014, and references therein). Typical bonding energies are of the order of
5-10 eV, corresponding to wavelengths of about 3000 Å and shorter, in the FUV domain.
A photodissociation reaction is typically a reaction of the type:
AB + hν → A + B
The photodissociation rate in the diffuse ISM can be expressed by
Z νH
kpd = 4 π JIS α(ν) dν)
νd
where JIS is the mean intensity of the interstellar radiation field and α(ν) is the pho-
todissociation cross section at a given frequency ν. The integration is considered from
the dissociation limit (νd ) to the hydrogen photo-ionization limit (νH ). In the interstel-
lar mean radiation field (∼ 108 FUV photon cm−2 s−1 sr−1 ), this translates into a rate of
10−9 -10−10 s−1 , corresponding to a lifetime of 109 -1010 s.
In the presence of dust in the ISM, the FUV radiation field will be attenuated. As
the absorption and scattering by dust is wavelength dependent, the frequency distribution
of the FUV radiation field will vary as a function of the depth into the cloud. In this
context, it is therefore important to distinguish two reaction rates: the unshielded one (in
the absence of dust), and the attenuated one (in the presence of dust). The latter can
be expressed as the former multiplied by an exponential factor depending on the visual
extinction (Av ) due to dust. As we are mostly dealing with photodissociation due to
the UV radiation field, one has to take into account the increased extinction of dust at
ultraviolet wavelengths via the following relation:
kpd = α exp [− γ Av ]
Intensive radiative transfer studies allowed to estimate α and γ for many photodisso-
ciation reactions. A few examples, taken from the UMIST database for astrochemistry
(http://www.udfa.net: Woodall et al. 2007) are provided in Table 2.1.
In the astronomical context, where abundant molecules such as H2 (or even CO) are
considered, we may envisage three situations of growing complexity when one wants to
quantify the photodissociation process:
- the photodissociation occurs in a diffuse cloud in the ISM because of the presence
of a population of massive stars in the vicinity, responsible for the production of a
strong FUV radiation field. Here, the photodissociation rate can be considered as
constant across the cloud, provided that the level population distribution is constant.
- the medium permeated by the FUV photons is not only made of gas, but also contains
dust. The radiation field undergoes therefore a cumulative attenuation as deeper
layers are considered. As a consequence, the photodissociation rate depends on the
dust distribution in the cloud.
Table 2.1: α and γ factors for a few important photodissociation reactions (UMIST
database: Woodall et al. 2007).
Reaction α γ
HCN + hν −→ CN + H 1.3 × 10−9 2.1
HCO + hν −→ CO + H 1.1 × 10−9 0.8
H2 O + hν −→ OH + H 5.9 × 10−10 1.7
CH + hν −→ C + H 8.6 × 10−10 1.2
CH+ + hν −→ C + H+ 2.5 × 10−10 2.5
CH2 + hν −→ CH + H 7.2 × 10−11 1.7
C2 + hν −→ C + C 1.5 × 10−10 2.1
C2 H + hν −→ C2 + H 5.1 × 10−10 1.9
C2 H2 + hν −→ C2 H + H 7.3 × 10−9 1.8
C2 H3 + hν −→ C2 H2 + H 1.0 × 10−9 1.7
CO + hν −→ C + O 2.0 × 10−10 2.5
N2 + hν −→ N + N 2.3 × 10−10 3.8
NH + hν −→ N + H 5.0 × 10−10 2.0
NO + hν −→ N + O 4.3 × 10−10 1.7
O2 + hν −→ O + O 6.9 × 10−10 1.8
OH + hν −→ O + H 3.5 × 10−10 1.7
- the gas is dense, and the radiation field is further attenuated by the molecules them-
selves, in addition to the attenuation due to dust. This is what is called self-shielding.
In this case, the photodissociation rate depends on the abundance of the molecule
and the level population distribution as a function of depth in the cloud.
When dust absorption and self-shielding are considered, in the case of H2 , the pho-
todissociation rate coefficient can be expressed as follows:
kUV (H2 ) = ko (H2 ) βSS exp (−τd )
where ko (H2 ) is the unshielded photodissociation rate in the absence of dust, βSS is the
self-shielding factor, and τd is the optical depth due to dust in the cloud. It should also be
mentioned that ISM clouds are not homogenous. The influence of dust and of self-shielding
will therefore depend intimately on the line of sight.
Carbon monoxide (CO) is a very stable molecule. The bonding energy is 11.09 eV, cor-
responding to 1118 Å . Considering the fact that interstellar photons with energies higher
than 13.6 eV (912 Å ) are almost completely used for the ionization of abundant hydrogen
atoms, the photodissociation of CO can only occur through photons with wavelengths
between 912 and 1118 Å .
2.1.2 Photoionization
The direct effect of photons on atoms or molecules can be a significant source of ionization
in interstellar clouds, provided wealth of photons are available. For instance, in diffuse
clouds (low density clouds which are not opaque to UV radiation), photoionization of C
to yield carbon cations is the triggering step of carbon chemistry. This is a consequence
of the fact that the ionization potential of C is 11.26 eV, i.e. lower than the ionization
potential of atomic hydrogen. As the ionization potentials of O and N are higher than
13.6 eV (respectively, 13.62 and 14.53 eV), these elements can not be photoionized. A few
examples of photoionization reactions are given in Table 2.2, with their α and γ parameters
carrying the same meaning as in the case of photodissociation reactions.
Table 2.2: α and γ factors for a few important photoionization reactions (UMIST database:
Woodall et al. 2007).
Reaction α γ
C + hν −→ C+ + e− 3.0 × 10−10 3.0
C2 + hν −→ C+
2 + e
− 4.1 × 10−10 3.5
CH + hν −→ CH+ + e− 7.6 × 10−10 2.8
NH + hν −→ NH+ + e− 1.0 × 10−11 2.0
OH + hν −→ OH+ + e− 1.6 × 10−12 3.1
2.1.3 Neutral-neutral reactions
This kind of reaction is typically of the type
A+B→C+D
Reaction rate coefficient are considered to be of the form
k = α (T/300)β exp [−γ/T]
with the parameters α, β and γ given in Table 2.3 for a few reactions. It should be noted
that the values attributed to these parameters are valid only for a given temperature in-
terval. In addition, these values will be different according to the approach adopted to
derive them (see the UMIST database for details: Woodall et al. 2007). One should also
note the significant deviation of β with respect to the 0.5 value expected for perfect gases.
The attractive interaction is due to van der Waals forces that is effective only at very
short distances (interaction potential ∝ 1/r6 ). Neutral-neutral processes are generally re-
actions with large activation barriers because of the necessity to break chemical bonds
during the molecular rearrangement. As a consequence, such reactions can be of impor-
tance where the gas is warm enough (i.e. possessing the requested energy): for instance
in stellar ejecta, in hot cores associated to protostars, in dense photodissociation regions
associated with luminous stars, and in clouds crossed by hydrodynamic shocks. In the
colder conditions of dark clouds, only neutral-neutral reactions involving atoms or radicals
can occur significantly because they have only small or no activation barriers. In the latter
case, the reaction rate coefficient takes the form
k = α (T/300)β
Table 2.3: α, β and γ parameters for a few examples of neutral-neutral reactions (UMIST
Reaction α β γ
H + OH −→ O + H2 7.0 × 10−14 2.8 1950
H + CH4 −→ CH3 + H2 5.9 × 10−13 3.0 4045
H + NH3 −→ NH2 + H2 7.8 × 10−13 2.4 4990
H + H2 CO −→ HCO + H2 4.9 × 10−12 1.9 1379
H2 + CN −→ HCN + H 4.0 × 10−13 2.9 820
C + OH −→ CH + O 2.3 × 10−11 0.5 14800
C + NH −→ N + CH 1.7 × 10−11 0.5 4000
CH + O −→ OH + C 2.5 × 10−11 0.0 2381
CH + O2 −→ HCO + O 1.4 × 10−11 0.7 3000
CH + N2 −→ HCN + N 5.6 × 10−13 0.9 10128
CH3 + OH −→ CH4 + O 3.3 × 10−14 2.2 2240
N + HCO −→ CO + NH 5.7 × 10−12 0.5 1000
N + HNO −→ N2 O + H 1.4 × 10−12 0.5 1500
N + O2 −→ NO + O 2.3 × 10−12 0.9 3134
NH + CN −→ HCN + N 2.9 × 10−12 0.5 1000
NH + OH −→ NH2 + O 2.9 × 10−12 0.1 5800
NH + O2 −→ HNO + O 6.9 × 10−14 2.7 3281
O + CN −→ CO + N 4.4 × 10−11 0.5 364
O + HCN −→ CO + NH 7.3 × 10−13 1.1 3742
O + C2 H4 −→ H2 CCO + H2 5.1 × 10−14 1.9 92
OH + OH −→ H2 O + O 1.7 × 10−12 1.1 50
O2 + SO −→ SO2 + O 1.1 × 10−14 1.9 1538
2.1.4 Ion-molecule reactions
In this case, we consider reactions of the type
A+ + B → C + + D
This kind of reaction occurs generally more rapidly than neutral-neutral ones because
of the strong polarization-induced interaction potential (∝ 1/r4 ). Typically, the reaction
rate coefficient is of the order of 2 × 10−9 cm3 s−1 with no temperature dependence. This
is generally at least one or two orders of magnitude faster than neutral-neutral reactions.
For this reason, even a small amount of ions in a given medium can have a strong impact
on interstellar chemistry. In addition, if the reaction involves an ion and a neutral species
with a permanent dipole, the efficiency of the interaction can reach much higher levels,
with improvement of the kinetics of one or two additional orders of magnitude. A few
examples of ion-molecule reactions are given in Table 2.4.
Table 2.4: Reaction rate coefficients for a few examples of ion-molecule reactions (UMIST
Reaction k
H+ +
2 + OH −→ H2 O + H 7.6 × 10−10
H2 + OH+ −→ H2 O+ + H 1.1 × 10−9
H2 O+ + H2 −→ H3 O+ + H 6.1 × 10−10
H2 O+ + CO −→ HCO+ + OH 5.0 × 10−10
H2 O+ + C2 −→ C2 H+ + OH 4.7 × 10−10
C+ + OH −→ CO+ + H 7.7 × 10−10
C+ + CH3 −→ C2 H+2 + H 1.3 × 10−9
C+ + C2 H2 −→ C3 H+ + H 2.2 × 10−9
C+ + C2 H4 −→ C3 H+2 + H2 3.4 × 10−10
+ +
CH + OH −→ CO + H2 7.5 × 10−10
CH+ +
2 + O2 −→ HCO + OH 9.1 × 10−10
CH+ +
3 + OH −→ H2 CO + H2 7.2 × 10−10
+
+
O + CH4 −→ CH3 + OH 1.1 × 10−10
OH+ + CN −→ HCN+ + O 1.0 × 10−9
OH+ + NH3 −→ NH+ 4 + O 1.0 × 10−9
2.1.5 Dissociative electron recombination reactions
In such a process, an electron is captured by an ion to form a neutral species in an excited

electronic state, then follows a dissociation of the neutral compound:
A+ + e− → A∗ → C + D
Rate coefficients are typically of the order of 10−7 cm3 s−1 . Some examples of reaction
rate coefficients, along with their dependence on the temperature (k = α(T/300)β ) are
given in Table 2.5.
This process constitutes an important step for a significant production channel for
many small neutral molecules that could not easily be formed through the single addition
of atomic species in gas phase. Many neutral chemical species are significantly produced
through ion-molecule processes, terminated by a dissociative electron recombination re-
action with the loss a of fragment. For instance, several H-bearing molecules could be
formed via mutliple reactions of cations with H2 molecules, terminated by a dissociative
electronic recombination: e.g. the formation of ammonia initiated by N+ , and the for-
mation of methane initiated by C+ . One may also consider the growth of larger neutral
aliphatic hydrocarbons through the successive addition of C-bearing molecules to a cationic
hydrocarbon, followed by the reaction with an electron. Considering the large value of the
rate coefficients of this process, it constitutes the main terminator of cationic processes.
Table 2.5: α and β parameters for a few examples of dissociative electron recombination
reactions (UMIST database: Woodall et al. 2007) .
Reaction α β
H+ −
2 + e −→ H + H 1.6 × 10−8 -0.43
H2 O+ + e− −→ O + H + H 3.1 × 10−7 -0.5
H2 O+ + e− −→ OH + H 8.6 × 10−8 -0.5
H3 O+ + e− −→ OH + H2 6.0 × 10−8 -0.5
H3 O+ + e− −→ O + H + H2 5.6 × 10−9 -0.5
H+ −
3 + e −→ H2 + H 2.3 × 10−8 -0.52
HCN+ + e− −→ CN + H 2.0 × 10−7 -0.5
HCO+ + e− −→ CO + H 2.4 × 10−7 -0.69
HNO+ + e− −→ NO + H 3.0 × 10−7 -0.69
C+ −
2 + e −→ C + C 3.0 × 10−7 -0.5
CH + e− −→ C + H
+ 1.5 × 10−7 -0.42
CH+ −
2 + e −→ CH + H 1.4 × 10−7 -0.55
+
CH2 + e− −→ C + H + H 4.0 × 10−7 -0.6
CH+ −
3 + e −→ CH + H + H 2.0 × 10−7 -0.4
CH+ −
3 + e −→ CH + H2 2.0 × 10−7 -0.5
CH + e− −→ C + H
+ 1.5 × 10−7 -0.42
CN+ + e− −→ C + N 1.8 × 10−7 -0.5
NH+ − 2
4 + e −→ NH + H2 1.5 × 10−7 -0.47
+ −
NO2 + e −→ NO + O 3.0 × 10−7 -0.5
OH+ + e− −→ O + H 3.8 × 10−8 -0.5
OCS+ + e− −→ CS + O 4.9 × 10−8 -0.62
2.1.6 Cosmic-ray induced reactions

A particularly important example of reaction occurring in the interstellar medium is the
interaction of chemical species in molecular clouds with cosmic-rays. Cosmic-rays are high
energy charged particles accelerated in quite extreme astrophysical environments such as
supernova remnants (Ellison et al. 1997; Prantzos 2012; Drury 2012; Helder et al. 2012),
or other astrophysical environments presenting for instance strong hydrodynamic shocks
likely to accelerate particles (Reimer et al. 2006; De Becker & Raucq 2013). Such processes
are especially efficient in dense molecular clouds, where the probability of the interaction
of the cosmic-ray particle with a partner is quite large.
Cosmic-rays can interact with molecules and lead to their dissociation, as illustrated
by the examples in Table 2.6. Such a process can be an efficient agent of destruction
of molecules in clouds that are too dense to allow significant photodissociation by the
interstellar UV radiation field. The dissociating particles may be the primary cosmic-ray,
or even secondary ions produced by its interaction with the cloud material, such as helium
cations.
Cosmic-rays can also directly ionize neutral species. Such a process can be viewed
as equivalent to a classical photoionization, but it is induced by the interaction with a
cosmic-ray. Examples of cosmic-ray ionization reactions are given in Table 2.7.
Finally, it has been demonstrated that the electrons generated by the interaction of
cosmic-rays with chemical species can collisionally excite hydrogen molecules in dense
clouds. These excited hydrogen molecules relax through the emission of UV photons
Table 2.6: Reaction rate coefficients for a few examples of cosmic-ray induced dissociation
reactions (UMIST database: Woodall et al. 2007).
Reaction k
H2 + CR −→ H+ + H + e− 1.2 × 10−17
He+ + CO −→ C+ + O + He 1.6 × 10−9
He+ + OH −→ O+ + H + He 1.1 × 10−9
He+ + H2 O −→ H+ + OH + He 2.0 × 10−10
Table 2.7: Reaction rate coefficients for a few examples of cosmic-ray ionization reactions
(UMIST database: Woodall et al. 2007).
Reaction k
H + CR −→ H+ + e− 6.0 × 10−18
He + CR −→ He+ + e− 6.5 × 10−18
H2 + CR −→ H+2 + e
− 1.2 × 10−17
C + CR −→ C+ + e− 2.3 × 10−17
CO + CR −→ CO+ + e− 3.9 × 10−17
Cl + CR −→ Cl+ + e− 3.9 × 10−17
N + CR −→ N+ + e− 2.7 × 10−17
O + CR −→ O+ + e− 3.4 × 10−17
(Prasad & Tarafdar 1983). This process constitutes a significant provider of UV photons
in inner parts of clouds which are too dense to be reached by the interstellar UV radiation
field. Several examples of cosmic-ray induced photoreactions (ionization and dissociation)
are given in Table 2.8, for reaction rate coefficients of the form
k = α (T/300)β γ/(1 − ω)
where α is the cosmic-ray photoreaction rate, β is a parameter governing the temperature
dependence of the process (often equal to 0.0), γ is an efficiency factor, and ω is the
average albedo of dust grains that contribute significantly to the decrease of the local UV
radiation field produced by cosmic-rays (typically equal to 0.6 at 150 nm).
The latter process is especially important in the sense that it is capable of recovering
neutral carbon atoms from its main reservoir (i.e. CO), in the absence of significant UV
interstellar radiation field from the neighboring stars. The efficiency of this process is
also expected to be higher than that of the dissociative charge transfer due to He+ (see
Table 2.6) followed by neutralization of C+ through electronic recombination, which is not
very efficient. The combination of these two processes to recover neutral carbon is at
least one order of magnitude less efficient than the cosmic ray induced photodissociation.
Considering this process appears therefore very important in order to understand quite
large values of the n(C)/n(CO) ratio as measured in the case of several dense clouds (e.g.
Phillips & Huggins 1981). Furthermore, this UV production process in the inner part
of dense clouds can play a significant role in the desorption of chemical species adsorbed
on dust grains, in order to release them in the gas phase. Grain surface processes are
discussed in Sect. 2.2.
Beside the fact that cosmic ray induced reactions introduce ions in a medium that
Table 2.8: Reaction rate coefficients for a few examples of cosmic-ray induced photoreac-
tions (UMIST database: Woodall et al. 2007).
Reaction α β γ
H2 O + CR −→ OH + H 1.3 × 10−17 0.0 486
HCO + CR −→ HCO+ + e− 1.3 × 10−17 0.0 585
HCO + CR −→ CO + H 1.3 × 10−17 0.0 211
C + CR −→ C+ + e− 1.3 × 10−17 0.0 155
CH + CR −→ C + H 1.3 × 10−17 0.0 365
C2 + CR −→ C + C 1.3 × 10−17 0.0 120
CO + CR −→ C + O 3.9 × 10−17 1.17 105
NH + CR −→ N + H 1.3 × 10−17 0.0 250
O2 + CR −→ O + O 1.3 × 10−17 0.0 376
is too dense to be ionized through the interstellar radiation field (which is important for
kinetics purpose), these high-energy particles vehiculate large amounts of energy that is
distributed to the products of these cosmic-ray induced reactions (which is important for
thermodynamics purpose). For instance, one may consider the case of a possible way to
produce ammonia in interstellar clouds. This process starts with the ion-molecule reaction
of N+ with molecular hydrogen, in a so-called H-atom abstraction reaction, producing
NH+ . Consecutive similar reactions lead to the NH+ 3 + H2 reaction, with the formation
of the ammonium cation. The latter species is likely to undergo a dissociative electronic
recombination reaction resulting in the formation of neutral ammonia. However, this
chain of processes is initiated by the reaction of N+ with molecular hydrogen, which is an
process that requires the availability of some energy to take place (about ∆E/kB ∼ 85 K).
This process is endothermic for ground state N+ , but is exothermic for excited states
(Ervin & Armentrout 1987). As a result, if N+ cations are at thermal energy in cold
clouds (∼ 10 K), the process described above for the formation of ammonia could not start
(Adams et al. 1984). However, the main source of nitrogen cations is expected to be the
cosmic-ray induced dissociation of N2 , with the resulting nitrogen cation possessing an
excess energy that is sufficient to overcome the energy requirement of the ion-molecule
process. This example emphasizes the important role of kinetically excited ions in the
gas-phase chemistry of cold interstellar clouds.
2.1.7 Charge transfer reactions

Another kind of reaction worth considering is the following:
A+ + B → A + B +
In this reaction, there is no break of chemical bonds. The only transformation is the trans-
fer of an electric charge. The rate coefficient can reach values of the order of 10−9 cm3 s−1
provided there is an energy level in B that is resonant (within ∼ 0.1 eV) with the recom-
bination energy of A+ .
Charge transfer involving atoms are crucial to set the ionization balance in the ISM.
An important example is the ionization balance of trace species in HII regions (parts of the
interstellar medium that are mainly populated by ionized hydrogen), with in particular
the charge exchange reaction between O and H+ . This latter reaction is important in
interstellar chemistry because it transfers ionization to oxygen which can then participate
Table 2.9: Reaction rate coefficients for a few examples of charge transfer reactions
Reaction k
H+ + OH −→ H + OH+ 2.1 × 10−9
N+
2 + H2 O −→ N2 + H2 O
+ 2.3 × 10−9
+
C + CH −→ C + CH + 3.8 × 10−10
C+ + CH2 −→ C + CH+ 2 5.2 × 10−10
C+ + C2 H4 −→ C + C2 H+
4 3.0 × 10−10
N+ + CH2 −→ N + CH+ 2 1.0 × 10−9
O+ + CH −→ O + CH+ 3.5 × 10−10
O+ + H2 O −→ O + H2 O+ 3.2 × 10−9
in the chemistry, and therefore allow the insertion of oxygen in molecules. This reaction
can be characterized a large rate coefficient if the ionization potentials are close. In the case
of ion-molecule charge transfer reactions, the rate coefficient may be rather large because
of the larger number of electronic states available in molecules compared to atoms.
2.1.8 Radiative association reactions

In such a process, the excited reaction product stabilizes through the emission of a photon.
Such a process can be illustrated by the following equation:
A + B → AB∗ → AB + hν
In this context, some typical time-scales are worth discussing. The radiative lifetime
for an allowed transition is τrad ∼ 10−7 s. The collision time-scale (τcol ) is of the order of
10−13 s. This leads to an efficiency for the radiation process of the order of 10−6 , i.e. in a
population of 106 molecules produced by the collision only one will be stabilized through
the emission of radiation. The result of the collision between the two partners can be seen
as an activated complex that can then follow three different reaction pathways:
1. The activated complex relaxes through the emission of radiation to yield a stable
reaction product.
2. The most probable scenario is that the activated complex redissociates.
3. If collisions are significant, other collision partners can take away the excess energy
and stabilize the reaction product. However, in order to become significant, this
pathway requires density conditions that are unlikely met in the interstellar medium.
Only long-lived activated complexes will lead to efficiency enhanced radiative association
reactions.
Considering elementary processes, we can define ka , kd and kr as the rate coefficients

respectively for the association of the reactants, the dissociation of the activated complex
and the radiative stabilization of the activated complex. The resulting radiative association
rate coefficient (kra ) will be expressed as follows:
ka
kra = ( ) kr
kr + kd
If we take into account the fact that the dissociation process is much faster than the
radiative relaxation, we obtain
ka kr
kra =
kd
where the ratio ka /kd can be considered as an equilibrium constant for the first elementary
process. A few examples of radiative association reactions are given in Table 2.10.
Table 2.10: Reaction rate coefficients for a few examples of radiative association reactions
Reaction k
C + H −→ CH + hν 1.0 × 10−17
C + C −→ C2 + hν 4.4 × 10−18
C + N −→ CN + hν 1.4 × 10−18
C + H2 −→ CH2 + hν 1.0 × 10−17
C+ + H −→ CH+ + hν 1.7 × 10−17
C+ + H2 −→ CH+2 + hν 4.0 × 10−16
O + O −→ O2 + hν 4.9 × 10−20
The lifetime of the activated complex is the main factor governing the kinetics of such
processes. The excess energy of the activated complex is temporarily stored in vibrational
form. The density of states where the energy can be stored is an increasing function of the
number of atoms in a chemical species, as this number is directly related to the number of
vibrational modes over which this energy can be distributed. Larger species will therefore
lead to a longer-lived activated complex, and hence to a higher radiative association rate.
For this reason, this process is very important for large molecular species such as polycyclic
aromatic hydrocarbons in the interstellar medium.
2.1.9 Associative detachment reactions

In this process, an anion and an atom collide. The neutral product is then stabilized
thanks to the emission of an electron that takes away the energy excess:
A− + B → AB + e−
The most important example is the following reaction:
H− + H → H2 + e−
whose rate coefficient is 1.3 × 10−9 cm3 s−1 . This process is strongly believed to have been
responsible for the production of molecular hydrogen in the early Universe, before its
enrichment in heavier elements due to stellar activity. Since then, molecular hydrogen is
indeed produced in the presence of dust particles, as discussed in section 2.2.2. Only the
densest astronomical environments are likely to be populated by a significant amount of
H− .
Such a process should not be very relevant for most astrophysical environments. For
instance, the radiative combination of atomic carbon with an electron to produce C− has
a reaction rate constant of 2.3 × 10−15 cm3 s−1 , which is very weak compared to the rate of
photoionization of C− , that is of the order of 10−7 s−1 . As a result, anions should be very
short-lived species, with very low probabilities to combine with neutral partners before
undergoing photoionization. In very dense astronomical environment, the low electronic

association rate could be compensated by large densities of electrons and neutral species,
favoring the existence of anions. However, the larger densities will allow more frequent
collisions likely to remove the additional electrons from the chemical species, leading to its
neutralization. So far, only a few anions have been identified in the interstellar medium
(typically, anionic polyines). Only polyatomic species able to stabilize the negative charge
have indeed a significant chance to exist, and to be detected in interstellar clouds. However,
it is interesting to emphasize the recent detection of the CN− anion in the circumstellar
envelope of IRC +10216 (Agúndez et al. 2010).
2.1.10 Collisional association reactions

One may consider processes where three bodies are interacting according to the following
way:
A + B + M → AB + M
Such processes are generally called termolecular processes because they involve 3 partners.
They are sometimes called thermolecular processes, with the prefix ‘ther’ justified by the
thermal role played by the third partner, as the latter takes away the excess energy in order
to stabilize the reaction product that would otherwise dissociate back to the reactants.
Such a process is unlikely to occur significantly in astrophysical environments for a
very simple reason: the probability of such a three body process is negligible due to the
low densities found in the interstellar medium. However, it may become significant for
the dense gas near stellar photospheres on in circumstellar disks characterized by large
densities (i.e. 1011 cm−3 ). This process is also significantly contributing to the chemistry
of planetary atmospheres.
2.1.11 Collisional dissociation reactions

The breaking of chemical bonds can also be the consequence of the collision of a given
molecule with any other partner. Such a process is especially efficient in high temperature
environments (several kK), where the colliding partners have kinetic energy that is suf-
ficient to promote electrons up to high v levels (v is the vibrationnal quantum number),
and the dissociation continuum of the molecule is easily reached.
An example worth discussing is the comparison of H2 with CO, the two most abundant
molecules in the Universe. In hot environments, both are significantly destroyed by colli-
sions. For instance, main contributors to this collisional dissociation are H, He and H2 . As
molecular hydrogen is homonuclear, it does not possess a permanent dipole. The radiative
relaxation of high v electrons is therefore forbidden by vibrational selection rules, and
there is accumulation of vibrationally excited H2 molecules. Consequently, its dissociation
is fast. On the contrary, CO has a permanent dipole, and the radiative relaxation of these
molecules is permitted. There is no accumulation of vibrationally excited CO molecules
and their dissociation is therefore much slower than that of H2 in the same environment.
2.1.12 Gas-phase chemical networks

Basically, a chemical network is a group of elementary processes governing the transforma-
tions of chemical species and reproducing the potential filiations between molecules in a
given medium. When one has to deal with the chemistry of a given part of the interstellar
medium, a census of molecular species and of their related transformation processes has
to be made.
A chemical network can reach rapidly a rather high level of complexity. Many different
kind of processes as described above are likely to take place, and all species present in the
medium are most probably involved in several processes simultaneously, as reactant or as
reaction product. Quantitatively, chemical networks are modelled using theoretical codes
fed with a lot of data including mainly reaction rate coefficients (determined in laborato-
ries or theoretically) and abundances when made available thanks to spectroscopic studies.
Several general rules based on chemical considerations govern gas-phase chemical net-
works:
- as H2 is the most abundant molecule in the Universe, reactions involving this latter
molecule will dominate the chemistry.
- if H2 is not the main potential partner for ions, their main loss channel will be
electron recombination.
- if electron reactions are also inhibited, reactions with neutral species will take over.
- for neutrals species, reactions with ions will be much more important than reactions
with other neutral species.
- in diffuse clouds (low density), for abundance considerations, the important ions are
C+ , H+ and He+ .
- still because of abundance considerations, the important ions in molecular clouds

(higher density) will be H+ + +
3 , HCO , H3 O and He .
+
- neutral species that do not react with the dominating ions will react with small
radicals (atoms of diatomic radicals).
- generally speaking, involving ions in the reaction network constitutes a crucial leap
to molecular complexity.
- the impact of cosmic rays in interstellar clouds is crucial for astrochemistry as these
high energy charged particles are responsible for the ionization of many chemical
species along their trajectory, and for the production of UV photons leading to the
occurrence of cosmic-ray induced photoreactions.
- the molecular build-up in the interstellar medium starts from atomic gas. Given
the low rates for radiative association reactions, the first step combining two atoms
into a diatomic species forms a bottleneck in the gas phase. The main implication
is that gas phase chemistry is not enough to explain the chemical complexity of the
interstellar medium (see Sect. 2.2).
As an example of chemical network, Fig. 2.1 illustrates oxygen chemistry in diffuse

clouds and photodissociation regions. In diffuse clouds, oxygen chemistry is initiated by
the charge exchange reaction between O and H+ , while in photodissociation regions the
reactions of O with H2 and H+ 2 can be important. Another example is given in Table 2.2
for the carbon chemistry. In the latter case, chemistry is initiated by radiative association
reaction of C+ with H2 .
O2
C,C+
OH
H CH, CH2
O+ O CO
H+ ν O
+
H2 H3
ν H2 ,H2* C Figure 2.1: Schematic chem-
e ical reaction network for O
OH+ OH chemistry in diffuse clouds
e C+
and photodissociation re-
H2 e H2 ,H2* gions.
ν
e
H2O+ H2O
C+
e
H2
H3O+
O H3+
CO HCO+
e
OH O2 H
ν He+ H2
ν
OH
O C C+ CO+
e
H
Figure 2.2: Schematic
e, ν
H,ν H2*
H H2 ,H2* O chemical reaction network
ν +
for C chemistry in diffuse
CH CH
e H2 clouds and photodissocia-
e H H2
tion regions.
O ν
e
CH 2
ν, C+ CH 2
+
e
H2
+
CH 3
2.2 Grain-surface processes

Particle densities in space are much lower than in the Earth’s atmosphere. In such con-
ditions, the probability of an interaction between reaction partners in the gas phase of
the interstellar medium is quite weak. Moreover, we know that the association of some
molecules and/or atoms can lead to activated complexes that have to radiate their excess
energy in order to be stabilized, but the dissociation of the activated complex back to
the reaction partners is often much more likely. In order to be efficient, associations may
require a third partner able to take away the excess energy (see Sect. 2.1.10). However the
probability of such a three-body process is very weak considering usual interstellar particle
densities. In this context, dust grains can play the role of the third body, allowing there-
fore to increase substantially the efficiency of the association between chemical species.
Taking into account grain-surface processes appears thus to be very important in order to
understand the formation of some common molecules – including H2 – along with that of
more complex molecules that have been found in space.
2.2.1 Schematic view of a grain-surface process

Two different surface mechanisms can be considered: the Langmuir-Hinshelwood mech-
anism and the Eley-Rideal mechanism. In short, a Langmuir-Hinshelwood grain-surface
process can involve four major steps:
Accretion. Particles (atoms or molecules) migrate from the gas phase to the surface
of dust grains and are adsorbed onto it.
Migration. Once the particles (adsorbates) are bound to the surface, they can
migrate or diffuse over the surface. This mobility is important in order to allow the
adsorbates to come together and to react.
Reaction. The reaction between adsorbates occurs when two reaction partners
come close enough in order to interact and create new chemical bondings.
Ejection. The reaction products can be ejected from the dust grain and populate
the gas phase.
In the case of the Eley-Rideal mechanism, there is no migration and it is considered that
the accreting species interact directly with previously adsorbed species, with immediate
reaction. In other words, the reaction is considered to occur through collision of gaseous
species with stationary species held on the surface of a solid.
Accretion. Atoms or molecules from the gas phase can adsorb on a grain surface.
The accretion rate coefficient on grains can be expressed as follows:
kac = nd σd v S
where nd is the number density of dust grains, σd is the cross-section of dust grains, v is
the mean speed of particles, and S is the sticking factor translating the probability that a
particle colliding with the dust grain accretes onto it. The sticking factor – that is depen-
dent on the dust and gas temperatures – is generally close to 1. In the particular case of
H, it is evaluated to be equal to 0.8 at 10 K, and to 0.5 at 100 K.
2.2. Grain-surface processes 43
Two adsorption processes must be considered:

- Physisorption: the weak interaction is that of van der Waals forces. These forces are
due to the mutually induced dipole moments in the electron shells of the adsorbate
and atoms on the surface of the dust grain. The typical interaction energy is of
the order of a few 0.1 eV. This interaction energy increases when the fixation site
corresponds to an kink or a corner in the structure of the dust grain.
- Chemisorption: in this case, the interaction can be about one order of magnitude
stronger than in the case of physisorption. Forces responsible for this reflect the
overlap between the respective wave functions of the adsorbate and the surface, i.e.
a chemical bond is created. These forces act on much smaller length scales. The
interaction energy is of the order of a few eV.
The binding energy can also be influenced by the composition of the dust grain, and
by presence of adsorbates on the surface that can strengthen or weaken the surface bond.
Assessment of adsorption energies can therefore become exceedingly complicated.
Surface migration. From the energetic point of view, the dust grain surface can
be considered as an irregular surface made of hills and valleys of different height and
depth, where the adsorbates can deposit. In this analogy, chemisorption sites correspond
to deep holes, and physisorption sites are represented by shallow cavities. On this surface,
the adsorbates may migrate from one site to another following some kind of random walk
between the moment they accrete and the moment they desorb. The mobility of adsorbates
can proceed following two different approaches. First, in the context of our analogy, one
can envisage that adsorbate climb up the hills provided they possess enough energy to do
it before settling in other cavities. In this case, we are dealing with a thermal hopping
migration process. However, even at very low temperature, the mobility of adsorbates is
not completely annealed. In this latter case, the mobility is supported by the quantum
effect called tunneling (see e.g. Cazaux & Tielens 2004).
We may therefore consider three different kinds of motion on the surface: (a) a transi-
tion between a physisorbed site and a chemisorbed site, (b) a transition from a chemisorbed
site to another chemisorbed site, and (c) and a transition from a physisorbed site to an-
other physisorbed site. These three situations are illustrated in Fig. 2.3. Generally speak-
ing, heavier species are less mobile that lighter ones. For this reason, efficient reactions
will mostly involve at least one light (and therefore mobile) species such as H, and also C,
N, and O.
The valleys represented in Fig. 2.3 are separated by saddle points, corresponding to
the top of the barriers that has to be crossed by the atoms in order to migrate to another
nearby adsorption site. Considering case (a), the activation barrier for a transition from
the chemisorbed site to the physisorbed site is the quantity Ec (k) − Es . The reverse sit-
uation will require less energy as the activation barrier is the quantity Ep (k) − Es , with
Ep (k) being much lower than Ec (k).
The migration time-scale dependence on the dust temperature can be expressed as

follows: E
m
τm ∝ exp
k Td
where Em is the energy barrier against migration (typically of the order of one third of
the binding energy), and Td is the dust temperature.
E
(a)
distance to the surface
Echem Es Ephys
Ep(k)
Ec(k)
atom k
Physisorbed site
atom k
Chemisorbed site
E
(b)
distance along the surface
Es
Echem Ec(k)
atom k atom k
Chemisorbed sites
E
(c)
distance along the surface
Es
Ephys Ep(k)
atom k atom k
Physisorbed sites
Figure 2.3: Illustration of the mobility of adsorbed atoms on a grain surface. (a) Barrier
between a physisorbed site and a chemisorbed site for an atom k bound to the surface. The
crossing of the barrier corresponds to a motion perpendicular to the surface. (b) Barrier
between two chemisorbed sites. The transition corresponds to a motion along the surface.
(c) Barrier between two physisorbed sites. The transition corresponds to a motion along
the surface as well. In each case, Es is the energy of the saddle point. Ec and Ep are
respectively the energies of the chemisorbed and physisorbed sites. Ec (k) and Ep (k) are
respectively the binding energy of the atom k in the chemisorbed and physisorbed sites
(Figure adapted from Cazaux & Tielens 2004).
Reaction. During their random walk along the grain surface, adsorbates (potential
reaction partners) may come close enough from each other to interact and create (and/or
break) chemical bonds. Such interactions will allow new chemical species to be formed,
before being rejected in the gas phase.
We may first consider reactions involving radicals – therefore reactions without acti-
vation barrier. Such reactions will involve mostly H, C, N, and O with other potentially
heavier radical partners. Beside these radical reactions, reactions with activation barrier
are also worth considering. As discussed earlier in this chapter, such reactions are inhib-
ited in gas phase. However, the prolongated residence time on grain surfaces allows such
reactions to occur significantly. This is one of the main distinguishing characteristics of
grain surface chemistry with respect to gas phase one.
Let us consider a mobile species adsorbed on a site where it can react with a neighboring
species with a given barrier (Ea ). The probability for a reaction before evaporation (pr )
as given by Tielens (2005) is then
pr = τ θ p
where p is the probability for the penetration of the reaction activation barrier. It has
been assumed that the reactant has much more chance to migrate than to penetrate the
reaction barrier. This relation clearly shows that the probability that the reaction occurs
depends directly on the overall time spent by the atom on the grain surface (τ ), and on
the surface coverage of potential coreactants (θ). The dependence with respect to the
activation barrier of the reaction is included in the factor p. A few examples of reactions
with their activation barrier (expressed in K) are given in Table 2.11.
Table 2.11: Examples of grain surface reactions with activation barriers involving H atoms
(Tielens 2005).
Reactants products Ea (K)

H + CO −→ HCO 1000
H + O2 −→ HO2 1200
H + H2 O2 −→ H2 O + OH 1400
H + O3 −→ O2 + OH 450
H + C2 H2 −→ C 2 H3 1250
H + C2 H4 −→ C 2 H5 1100
H + H2 S −→ SH + H2 860
H + N2 H2 −→ N2 H + H2 650
H + N2 H4 −→ N2 H3 + H 2 650
Evaporation The residence time of an adsorbed chemical species on the surface of a

dust grain can be expressed as follows:
E
b
τev = νo−1 exp
k Td
where Eb is the binding energy of the adsorbed species, and Td is the dust temperature.
νo is the vibrational frequency of the adsorbed species on the grain surface. The residence
time is the typical time-scale for a species to acquire sufficient energy through thermal
fluctuations to evaporate. This time-scale is very sensitive to temperature elevations of the

dust grain. Heavier species will benefit from longer residence times, and this dependence
on the mass of the particle is provided by the factor νo .
As an example, let us consider the case of the adsorption of atomic H on dust grains.
For a typical value of νo of 1012 s−1 , and a bonding energy (Eb /k) of the order of 500 K
for a physisorbed site, the residence time of atomic H is of the order of 5 × 109 s at 10 K,
and 2 × 10−8 s at 50 K. For a chemisorbed site with a typical bonding energy of the order
of 20000 K, the residence time tends to ∞, even at 100 K. The conclusion is that, in the
absence of significantly strong adsorption site, a temperature of a few tens of K is too
large to allow a reaction of adsorbed hydrogen atoms with any other adsorbed partners
during its short residence time. At such temperatures, chemisorbed sites are necessary to
envisage surface reactions of atomic hydrogen.
The desorption of adsorbed species can be substantially influenced by several factors:
- Action of ultra-violet (UV) radiation: typically, photo-desorption is enhanced in

diffuse clouds and photo-dissociation regions (PDRs). Photons provide the energy
to eject species from their adsorption sites.
- Action of X-rays: in the presence of active X-ray sources, notably in the context of
proto-stellar objects whose X-ray activity is significant, X-rays are likely to affect
dust grains at least in some parts of the proto-stellar disk. The action is similar to
that of UV radiation, but with more energetic photons. In addition, the excess en-
ergy is efficient at warming somewhat the dust grains, favoring therefore subsequent
thermal desorption of atoms and molecules.
- Action of cosmic-rays: charged energetic particles can interact with dust grains and
deposit an given amount of energy that is high enough to allow a significant fraction
of adsorbed species to desorb. The efficiency of the cosmic-ray-driven desorption will
depend on its energy and on the size of the dust grain.
- Hydrodynamic shocks: the conversion of kinetic energy into thermal energy is also
very favorable to an efficient desorption. In addition, shocks are also likely to be
responsible for a significant sputtering of dust grains, potentially leading to their
destruction.
2.2.2 The formation of molecular hydrogen

As we are dealing with surface processes, we will not adopt densities expressed in cm−3 ,
but we will talk about monolayer surface coverage translating the fraction of the total dust
grain surface that is covered by a given species. The reaction rates are therefore expressed
in units of monolayer per second. In the discussion below, mostly inspired by the work of
Cazaux & Tielens (2004), only the Langmuir-Hinshelwood mechanism is considered.
Let us first consider the surface reaction process of H atoms leading to the formation
of molecular hydrogen. Here, we consider that H atoms accrete onto the grain surface
with a rate F . The surface coverage of physisorbed and chemisorbed hydrogen is noted
HP and HC respectively, and that of molecular hydrogen is noted H2 . We will also make
use of diffusion coefficients noted αi , with the subscript i depending on the nature of the
starting and arrival sites, i.e. physisorbed (P) or chemisorbed (C). The desorption rate
coefficients will be noted βHP , βHC and βH2 respectively for physisorbed, chemisorbed and
molecular hydrogen. We can write the rate equations of physisorbed and chemisorbed
H atoms, taking into account every gain and loss processes likely to affect their surface
coverage:
ḢP = F (1 − HP − H2 ) − αP C HP (1 − HC ) − αP C HP HC − 2 αP P H2P
+αCP HC (1 − HP − H2 ) − αCP HC HP − βHP HP
ḢC = αP C HP (1 − HC ) − αP C HP HC − αCP HC (1 − HP − H2 )
−αCP HC HP − 2 αCC H2C − βHC HC
1 4 7
P 2 A A2
3 6
C 5 A
Figure 2.4: Elementary processes diagram (EPD) illustrating the census of elementary
processes affecting the surface coverage of physisorbed A species (surrounded by a circle),
for the formation of A2 . The upper and lower parts of the diagram stand respectively for
physisorbed and chemisorbed sites. The boxes in each row illustrate the occupation of
adsorption sites: they can be empty, or occupied by atomic or molecular species.
A convenient way to establish such rate equations is based on elementary processes

diagrams (EDP), such as displayed in Fig. 2.4 for the rate equation of physisorbed A.
Each elementary process is illustrated by a unique arrow, and each arrow translates into
one term in the second hand side of the rate equation. The first term of the first relation is
the rate of incoming atoms that are staying on the surface. It is considered indeed that the
accretion occurs only on physisorbed sites (direct chemisorption is unlikely), taking into
account that a fraction of physisorption sites are already occupied by atoms or molecules
(here, only species made of H are taken into account). The second term expresses the
rate of physisorbed atoms moving to free chemisorption sites. The third term stands for
the recombination of moving physisorbed H atoms with chemisorbed H atoms. The fourth
term stands for the recombination between physisorbed H atoms. The fifth and sixth terms
represent respectively the migration of chemisorbed atoms into free physisorbed sites and
the recombination of chemisorbed atoms with physisorbed ones. The last term expresses
the desorption of physisorbed H atoms.
In the rate equation of chemisorbed hydrogen, the first term expresses the rate of
H atoms coming from physisorbed sites and migrating into free chemisorbed sites. The
second term stands for the recombination between a moving physisorbed atom and a
chemisorbed atom. The third term expresses the migration of a chemisorbed atom into
free physisorbed sites. The fourth term represents recombination of moving chemisorbed
H atoms with physisorbed H atoms. The fifth term expresses the rate of recombination
between two chemisorbed H atoms. The last term stands for the desorption of chemisorbed
H atoms.
The desorption of the hydrogen molecules can proceed following two approaches:
- a first order desorption where the physisorbed molecule leaves the dust grain surface.
- a second order desorption where the molecule evaporates from the surface during the
formation process. Typically, the energy released by the formation of the molecule
is used to leave the energy well of the physisorption site. It is considered that
only a fraction µ of the molecular hydrogen stays in the physisorption site after its
formation. The complementary fraction (1 - µ) desorbs spontaneously during the
formation process.
Taking into account these considerations, we may write the rate equation for molecular
hydrogen:
Ḣ2 = µ (αP P H2P + αCP HC HP + αP C HP HC + αCC H2C ) − βH2 H2

and the total desorption rate is
Ḣ2 = (1 − µ) (αP P H2P + αCP HC HP + αP C HP HC + αCC H2C ) + βH2 H2

where the two terms stand respectively for the second order and first order desorption rates.
The production of H2 in the interstellar medium is of course not so simple, but this
discussion gives a idea of the process that is now believed to be responsible for the formation
of the most abundant molecule in the Universe. I one wants to refine this mechanism,
several aspects of the approach described above should be upgraded:
- Only one isotope of hydrogen has been taken into account. However, the absolute
abundance of the second isotope of hydrogen (i.e. deuterium, D) is not negligible as
compared to that of some major elements that astrophysicists call generally metals,
such as C, N, O, Si or P. Reaction channels involving deuterium should therefore be
taken into account.
- In the mechanism presented above, only reactions of H with other H atoms are
considered. Many other partners, in addition to D, constitute potential reactants for
H atoms on dust grains.
- Only the Langmuir-Hinshelwood mechanism has been considered. If the surface

coverage of reactants is high enough, additional terms in the rate equations should
be included to take into account Eley-Rideal processes.
- If the density of the overlying gas phase is high enough, the accretion rate of chemical
species can increase substantially and the approximation of the monolayer is not valid
anymore.
- Only one typical migration coefficient has been considered for each kind of motion
on the grain surface (C→C, C→P, P→P or P→C). It is obvious that, depending on
the nature of the solid grain and on that of other accreted species, physisorption and
adsorption interaction may vary significantly. This will have a strong impact on the
migration coefficients that should be taken into account in the rate equations.
2.2.3 Grain-surface chemical networks

An example of grain surface chemical network is shown in Fig. 2.5. Atomic carbon can be
hydrogenated on grain surfaces through sequential addition of atomic hydrogen to form
methane. However, in molecular clouds, most of the carbon is in the form of carbon
monoxide (CO). As a consequence, the chemistry of C is dominated by processes starting
from CO. Carbon monoxide may react with O or OH to form carbon dioxide. It is inter-
esting to note that the presence of traces of N, C, and O can lead to increasingly complex
chemical species such as ethanol, acetaldehyde, dimethyl ether, formamide, or formic acid.
These latter species are likely to constitute the starting point of a rather complex organic
chemistry. This very important part of interstellar chemistry will be discussed with some
details in Chapter. 3.
CO
H
C
CCHO HCO
O
H H N
CHCHO H CO HCON HCOO

2
H H H H
CH CHO CH3O HCONH HCOOH

2
H H Carboxylic acid
Aldehydes H
CH CHO CH3OH HCONH 2

3
H C Amide
CH3CH2O
CH3OC
H
H
CH3CH2OH
CH3OCH
Alcohols
H
CH3OCH 2
CH3OCH 3
Ether
Figure 2.5: Illustration of the chemistry of carbon monoxide on grain sufaces.

Chapter 3
Molecules in the Universe
3.1 Molecular content of the Galaxy

3.1.1 Inorganic compounds
In order to illustrate the chemistry that is likely to take place in the ISM, the example of
carbon monoxide is of particular interest. Indeed, CO is the most abundant molecule of
the Universe after molecular hydrogen. Its first interstellar detection comes from Wilson
et al. (1970) thanks to observations in the radio domain. CO is commonly used as a
proxy for molecular hydrogen, as it is generally more easy to detect than H2 and their
abundances are often correlated (see e.g. Dickman 1978). This correlation comes from the
fact that the radio emission of CO (mainly at 2.6 and 1.3 mm) is caused by collisional
excitation with H2 molecules. Through this process, the emission lines of CO are good
tracers of molecular hydrogen whose UV lines are often heavily obscured in dense clouds.
The formation of carbon monoxide is likely to occur rather efficiently in gas phase,
through the radical reactions (neutral-neutral reactions) between C with a small O-bearing
molecule, i.e. between C and OH, between C and O2 , between CH and O, or between CH2
and O. The dissociative electronic recombination reaction involving the ion HCO+ should
also contribute. Some of these reactions are included in the gas-phase chemical network
presented in Fig. 3.1.
CH
O2 e− O
+ H2 +
C CH2 CH3
e−
H2 O
METALS CH2
+ e− He
+
+ O 2,CO2 +
HCO CO C CO H2
+
H3
+ e−
O 2 ,SO 2, CO2 H 2O CH5
H2
Figure 3.1: Simplified chemical network including some of the main reactions involved in
the formation and destruction of CO (adapted from Prasad & Huntress 1980a).
Another important example is that of the cyanide radical (CN). This is one of the first
molecules identified in the interstellar medium. The first interstellar lines originating from
51
52 Molecules in the Universe
CN were reported by McKellar (1940). The chemistry of cyanide is likely related to that
of hydrogen cyanide molecule (HCN) whose first detection in the interstellar medium is
due to Snyder & Buhl (1971). A simplified chemical network describing the chemistry of
CN and of HCN is shown in Fig. 3.2.
LOSS
O, O 2
CH N
LOSS
CN
e− C
+
NH C
e−
+ +
H3 H 2CN HCN
+ +
H 3 ,HCO
+
N CH3
+ N
+ H2 C
HCN
NH2 + NH3 CH2 ,CH3

C
Figure 3.2: Simplified chemical network including the main reactions involved in the for-
mation and destruction of cyanide and hydrogen cyanide (adapted from Prasad & Huntress
1980a).
Several molecules containing a cyano group have been found in the interstellar medium:
NaCN (Turner et al. 1994), MgCN (Ziurys et al. 1995), SiCN (Guélin et al. 2000). In ad-
dition, organic compounds containing CN groups have been detected as well, including
cyanopolyynes. For a discussion of chemical processes related to cyanpolyynes in inter-
stellar clouds, one can refer for instance to Mitchell et al. (1979).
Water is also an important molecule that deserves a few explanation. Water has
been detected for the first time in the interstellar medium in the radio domain (Cheung
et al. 1969). Since then, it has been detected in various environments such as diffuse
clouds, dense star-forming regions in the Milky Way and nearby galaxies, cometary comae,
planetary and exoplanetary atmospheres (see van Dishoeck et al. 2013, and references
therein). One may emphasize for instance the results obtained with the Herschel infrared
space observatory in star forming regions published by van Dishoeck et al. (2011). Water is
the thermodynamically most stable oxygen-containing molecule after CO. In oxygen-rich
environments (where the abundance of oxygen is higher than that of carbon), one may
expect most oxygen atoms in excess of those found in CO molecules to be involved in
water formation.
The formation of water proceeds through both gas-phase and grain-surface processes.
In gas-phase, one may distinguish low-temperature processes dominated by ion-neutral re-
actions (typically in dense and dark clouds where ions are mainly produced by cosmic rays)
and higher temperature environments where neutral-neutral processes become significant
(typically in the densest parts of photo-dissociation regions). Main processes involved in
the production of water in interstellar environments are summarized in Fig. 3.3.
The most straghtforward ionic pathway to produce water is the successive reaction of
an oxygen cation with the most abundant hydrogen carrier, i.e. H2 . Each reaction leads to
the addition of an H atom, with conservation of the cationic nature of the product. When
H3 O+ is formed, the dissociative electronic recombination reaction produces neutral water,
3.1. Molecular content of the Galaxy 53
Figure 3.3: Summary of the main reactions involved in the formation and destruction of
water, considering both gas-phase and grain surface pathways. The thickness of the arrows
illustrates the amplitude of the activation barrier (for neutral-neutral processes). On the
right part, s-X stand for a species X adsorbed onto the grain surface. Taken from van
Dishoeck et al. 2013.
with the ejection of an H atom. It should be mentioned that the dissociative electronic re-
combination reaction will lead to several potential products (parallel reactions): H2 O + H,
OH + H2 , OH + H + H, or O + H2 + H. All these parallel reactions are characterized by
branching ratios whose determination is crucial, and not easy. The most recent measure-
ments suggest that the OH + H + H product dominates (∼ 71%), with only about 17 %
for the formation of water (Buhr et al. 2010). In parallel to the cationic reaction chain,
the equivalent neutral-neutral pathway can take place. The significance of such a reaction
chain is intimately dependent on the availability of energy to cross the activation barrier.
For this reason, this process is expected to be more efficient at higher temperatures. On
dust grains, two successive additions of H to O (and then to OH) leads to the formation of
water. As illustrated in Fig. 3.3, other grain surface pathways can be considered, involving
ozone (O3 ) or hydrogen peroxide (H2 O2 ). The formation of water through the addition of
H atoms to O2 molecules on grain surfaces is expected to be efficient, as demonstrated by
laboratory studies performed by Ioppolo et al. (2008).
3.1.2 Hydrocarbons
Carbon chemistry occupies a privileged position in chemistry, as it constitutes the base-
ment of organic compounds, including the building blocks of life. Historically, the most
simple molecule containing only hydrogen and carbon discovered in the ISM is the CH
radical, and this discovery has been reported at the end of the 30’s by Swings & Rosen-
feld (1937). Since then, many polyatomic hyrocarbons characterized by various degrees of
unsaturation have been identified. In this discussion, we will mostly address the cases of
the two main classes of compounds: aliphatic hydrocarbons and aromatic hydrocarbons. A
few words will also be given on other carbonaceous compounds (likely to be) found in the
interstellar medium.
Aliphatic hydrocarbons.
In order to illustrate the main chemical processes leading to the formation of organic
compounds, we may first consider the chemical reactions involved in the formation of
the simplest saturated hydrocarbon, i.e. methane (CH4 ), that has been detected in the
interstellar medium for the first time by Lacy et al. (1991). The corresponding gas phase
chemical network, as proposed by Prasad & Huntress (1980b), is illustrated in Fig. 3.4.
+
H3 +
C CH
H2 +
CH4 + +
H He
H2
H2 + H2 + H2 + e−
+ CH2 CH3 CH5 CH4
C +
H3
e −,Mg + +
CH3 C
CHn
O O + +
C 2H 5 C 2H 3
e− e− e−
+
CHnO C 2H 4 C 2H 2
Figure 3.4: Simplified chemical network including the main reactions involved in the for-
mation of methane in interstellar clouds (adapted from Prasad & Huntress 1980b).
In this example, we see that the two main pathways leading to low molecular weight
hydrocarbons are:
- the reaction of a neutral carbon atom with cationic H-bearing species. In the chem-
ical network presented here, H+3 , as an abundant cationic molecule containing H, is
considered.
- the reaction of cationic carbon with H2 , which acts as a preferential partner for
obvious abundance reasons.
Of course, other potential partners may also be involved in such a process, but with prob-
ably weaker contributions. It is also important to note that no neutral-neutral reactions
have been considered here because of their significantly lower reaction rates as compared
to ion-molecule processes, as explained in Sect. 2.1.4. It has also been emphasized in
Sect. 2.1.5 that the dissociative electronic recombination is a very important process for
polyatomic cationic species, leading to neutral compounds after the loss of a fragment.
Here, several compounds may undergo such a process, and it is illustrated for instance
in the case of cationic hydrocabons that are precursors of species such as methane, ethy-
lene or acetylene. One should also note the important role of cosmic rays, and secondary
cations such as He+ , in the destruction of methane and other molecules in such molecular
clouds.
Alternatively, the successive addition of H starting from atomic C is a possible for-
mation mechanism on dust grains. Indeed, in agreement with the principles developed in
Sect. 2.2, as mobile species, H atoms constitute valuable reaction partners for less mobile
ones already adsorbed on dust grains, such as C, or hydrocarbons such as CH2 , or other
molecules belonging to the category of hydrocarbons.
Aromatic hydrocarbons.
Investigations of the Universe in the infrared revealed among others interstellar emission
bands at 3.29, 6.2, 8.7 of 11.3 µm forming the core of an important issue in interstellar
astrophysics (see e.g. Gillett et al. 1973). For instance, see the infrared spectrum shown
in Fig. 3.5. Such spectral features have been found to be associated with many astrophys-
ical objects such as planetary and reflection nebulae, H ii regions, or even extragalactic
sources. The ubiquity and the strength of these spectral lines showed that they were due
to widespread and abundant interstellar carriers. It has first been proposed that the mech-
anism responsible for such emission bands was most likely infrared molecular fluorescence
lines pumped by UV radiation (Allamandola & Norman 1978). Although the mechanism
appears to be correct, the idea that large molecules such as aromatic compounds were
the carriers of those lines came a bit later (Duley & Williams 1981; Leger & Puget 1984;
Allamandola et al. 1985). Since then, it is now admitted by the scientific community that
polycyclic aromatic hydrocarbons constitute an abundant component of the interstellar
medium, and that they play a significant role in its energy budget.
Figure 3.5: Infrared spectrum in the direction of the planetary nebula BD +36◦ 3639. Most
of the porminent spectral features are attributed to PAHs ( NASA).
c
The most common aromatic molecule is benzene. This molecule has been discovered
in the interstellar medium through infrared observation by Cernicharo et al. (2001). Poly-
cyclic aromatic hydrocarbons are mainly planar molecules allowing the delocalization of
electrons across more than one aromatic ring. Following this idea, the next aromatic com-
pound starting from benzene contains 4 additional carbons providing each one additional
p electron: i.e. naphtalene. If we go on this way, adding 4 other carbons to increase
linearly the aromatic compound, we obtain anthracene. These molecules start the series
of catacondensed polycyclic aromatic hydrocarbons (see Fig. 3.6). In this series, no carbon
belongs to more than two rings. In catacondensed PAHs, we can envisage two subclasses:
the acenes which are linear, and the phenes which consists in bent rows of rings.
(a) (b) (c)
(d) (e)
(g)
(f)
(h)
(i)
Figure 3.6: Molecular structure of a few catacondensed PAHs: (a) naphtalene, (b) an-
thracene, (c) phenanthrene, (d) tetraphene, (e) naphtacene (or tetracene), (f) chrysene,
(g) pentacene, (h) pentaphene and (i) hexacene. In each case, only one resonance form is
illustrated.
Beside catacondensed PAHs, we can consider also the class of pericondensed PAHs con-
taining carbons that are members of three separate rings. The structure of some members
of this class is shown in Fig. 3.7. Pericondensed PAHs are sometimes called superaromat-
ics. Among pericondensed aromatics, centrally condensed are particularly stable because
their structure generally allows an improved electron delocalization. The general formula
of centrally condensed PAHs is C6r2 H6r , where r is an integer. For r equal to 1, we re-
trieve benzene. For r = 2, we obtain coronene, and so on. For such aromatics, there are
3r2 − 3r + 1 hexagonal cycles arranged in r − 1 rings around the central cycle. Consid-
ering a typical C-C bond length of about 1.4 Å , the surface area of one aromatic ring is
about 6 Å2 . It should also be noted that PAHs containing a few tens of carbons can have
sizes larger than 100 Å . Such sizes are typical of the smallest dust grains in the interstellar
medium. The characteristic size of these molecules is therefore intermediate between small
molecules and dust particles.
(a) (b) (c)
(d) (f)
(e)
(g)
(h)
Figure 3.7: Molecular structure of a few pericondensed PAHs: (a) pyrene, (b) coronene,
(c) antanthrene, (d) ovalene, (e) perylene, (f) hexabenzocoronene, (g) circumcoronene and
(h) pyranthrene. In each case, only one resonance form is illustrated.
The main process likely to lead to the formation of PAHs is the condensation of short
carbon chains. In the case of the formation of aromatic compounds, this process con-
sists in the ion-molecule or the radical-molecule addition of short unsaturated aliphatic
molecules such as alcenes, allenes and alcynes. For instance, we can consider the reac-
tions illustrated in Fig. 3.8 leading to the formation of the phenyl radical (C6 H5 ) through
the successive addition of C2 H2 molecules. From such a phenyl radical, further additions
of acetylene molecules are likely to increase the size of the aromatic molecule through
successive formation of additional rings (see Allamandola et al. 1989).
It should also be noted that we cannot reject a scenario where the addition of small un-
saturated aliphatic chains is catalyzed by metallic ions, via the formation of organometallic
compounds as illustrated in Fig. 3.9. The interaction of metallic cations with π electrons
of alcenes (or alcynes) can lead to the formation of adducts to catalize the approach of
molecules likely to react to form aromatics (see e.g. Bohme 1992).
It has also been proposed that the extension of the size of PAHs may proceed through
the concerted addition of monocyclic aromatics. A concerted addition occurs when bonds
are formed and others are broken simultaneously. Such an approach is therefore more
favorable from the activation point of view. For instance, Mimura (1995) considered the
formation of PAHs by shock waves propagating in the interstellar medium. In this sce-
nario, the energy required for the cycloaddition is taken from the shock. In the context of
concerted cycloaddition, the product of the cycloaddition has to undergo an intramolecu-
lar rearrangement followed by the elimination of H2 in order to yield planar PAHs.
H H
H C 2H2 H C H
H C C H
C .
C
C
.
C
H C
H H
H H H
C C C
H C H C H C
C 2H2 C
C C
C
.
C H .
C
+ H
C
+ H
H H H H
C
H H
H
C
C
C
. H
H
C
C
C
H
C C C C
H C H H C H
H H
C 2H2
H H
C
H C C
C C
C C
+ H ...
H C H
Figure 3.8: Probable formation mechanism of benzene initiated by a radical species, i.e.
hydrogenated acetylene, adapted from Allamandola et al. (1989). The process repeats until
the formation of a phenyl radical. This radical can yield benzene through the addition
of hydrogen, or can undergo additional reactions with C2 H2 molecules leading to more
extended PAHs.
A significantly important aspect of the chemistry of PAHs in the interstellar medium is

photochemistry. As mentioned earlier in this section, the presence of PAHs in the Galaxy
and beyond has been revealed by their infrared emission bands arising from the relaxation
of their vibrational degrees of freedom, after excitation by UV photons. The high capability
of PAHs to relax progressively through cascades in their vibrational states is responsible
for their high stability upon photodissociation. However, some channels in these cascades
may lead to the breaking of chemical bondings with subsequent loss of fragments (see e.g.
Tielens 2005).
The probability of photodissociation can be expressed by the following relation:
k
pd =
k + kIR
where k and kIR are respectively the rate coefficients for the dissociation and for the in-
frared radiative relaxation channels.
H H H H
H H
C C.
C 2H2
C C . C C
H
H Me
+ H Me
+ C
H +
Me
H C C H . C
H
H C 2H2
H C H
H H
H H
C C
C C
−H C C
C C H
H + C H
H C H Me +
. H Me C
H
+
H
C
C
C
H
C
. C
+ Me H C
H H
H
Figure 3.9: Probable formation mechanism of benzene where a metallic cation acts as a
catalyst. The positive charge favors the approach of acetylene molecules, and the presence
of the cation constrains the conformation adopted by the radicalar intermediates. A similar
mechanism as in Fig. 3.8 in the presence of the metallic cation can also lead to the formation
of phenyl radical.
In case of dissociation, two pathways can be envisaged:
- Side σ bond rupture: the absorption of a photon can lead to the breaking of a C-X
bond, where X is hydrogen or a sidegroup such as -OH, -CH3 , -NH2 . Alternatively,
the bond rupture can occur in the sidegroup, with subsequent release of H or for
instance a methyl group from an ethyl sidegroup (Geballe et al. 1989).
- Concerted decondensation: as an alternative reaction channel leading to the breaking

of σ bonds, the possible concerted fragmentation of an aromatic molecule into acety-
lene molecules should also be considered. According to Allamandola et al. (1989),
this process is the only pathway involving the breaking of C-C bonds in aromatics
that may be envisaged, even though it has a higher energy threshold than the rupture
of sidegroup bonds.
Another process that is important for PAHs is an electron attachment process. These
large molecules are indeed likely to give rise to radiative electronic recombinations such as
PAH + e− → (PAH− )∗ → PAH− + hν
The electronic capture (kc ) leads to an activated complex which can itself relax radiatively
(kr ) to yield the negatively charged PAH, or it can undergo an electronic detachment (kd )
leading back to the neutral PAH. Such a two-step process involving an activated complex
is very similar to that of radiative association reactions described in Sect. 2.1.8.
In the case of the electron capture of PAHs, the reaction rate coefficient is generally
expressed the following way:
kc kr
kra = ( ) kr = ( ) kc = Se kc
kr + k d kr + kd
where Se is defined as the sticking coefficient of the electron on the PAH. The same ex-
pression for the recombination of PAH cations holds, but in the latter case the radiative
stabilization is very rapid and the sticking coefficient is essentially unity.
Processes such as charge exchange are also important for PAHs. The abundances of
PAH anions in molecular clouds is regulated by the balance between electron capture and
the recombination with metal or molecular cations.
PAH− + M+ → PAH + M
Such a process is likely the most important one leading to the neutralization of PAH anions
in the ISM.
Beside the charge transfer reaction described above, ion-neutral reactions involving
PAHs can lead to the formation of adducts ions. For instance, the addition of Si+ is
mentioned by Millar (1992):
PAH + Si+ → SiPAH+
Another example of ion-molecule reaction that may be important in interstellar clouds is
the interaction of PAHs with ionized helium (Bohme 1992):
PAH + He+ → PAH+2 + He + e−
This reaction is a significant source of dications in dense clouds where He+ is abundant
because of the efficient ionization of neutral helium by cosmic-rays.
Finally, it should also be noted that PAH-catalysed processes should play a role in
the chemistry of various astrophysical environments. As chemical species are likely to
form adducts with PAHs (see above), it has been proposed that adsorption-like processes
may occur involving PAHs and atomic hydrogen (see e.g. Bohme 1992). As a result, a
process similar to that described in Sect. 2.2.2 for the formation of H2 may take place on
large PAHs. Considering the large abundance of PAHs in some regions of the interstellar
medium, such a process may constitute a significant channel for the formation of molecular
hydrogen. A brief discussion of this scenario is given by Bohme (1992). Following this
idea, other PAH-catalyzed processes are likely to take place in the interstellar medium such
as neutralization reactions, or other chemical reactions unlikely to take place efficiently in
gas phase without the support of a third partner.
Other carbonaceous compounds

Another class of carbonaceous compounds is that of fullerenes: compounds made entirely
of carbon that are similar in structure to graphite, but are not planar and include also
pentagonal (or even heptagonal) rings. The first fullerene (C60 ) was discovered in the
1980’s (Kroto et al. 1985). Their presence in astronomical environments was suspected, but
some of them (C60 and C70 , see Fig. 3.10) were discovered in a planetary nebula (remnant
of an intermediate mass star that ejected large amounts of material) only quite recently
(Cami et al. 2010). Observations of the young planetary nebula Tc1 (∼ 6500 ly) in the
infrared domain performed with the Spitzer satellite1 revealed indeed prominent spectral
features at 7.0, 8.5, 17.4 and 18.9 µm attributable to C60 , along with other features due
to C70 .
1
Spitzer is a NASA infrared space-borne observatory launched in August 2003:
http://www.spitzer.caltech.edu/mission/32-The-Mission
Figure 3.10: Structure of two members of the class of fullerenes recently discovered in an
hydrogen poor planetary nebula: C60 (on the left) and C70 (on the right).
It is interesting to note that the environment where these two fullerenes were firmly
identified is carbon-rich, and hydrogen poor. Indeed, no significant signs of the presence
of aliphatic hydrocarbons or PAHs were found in the circumstellar nebula of Tc1. This
provides a valuable indication on the circumstances likely to lead to the formation of
fullerenes that is inhibitted in the presence of hydrogen. In addition, it seems that the
fullerenes detected in this context are still adsorbed onto carbonaceous dust grains. Finally,
the results of the study of Cami et al. (2010) reveal that C60 and C70 molecules are found
at significantly different distances from the central white dwarf, with the former closer
than the latter, suggesting that C70 might be formed starting from C60 , even though
this assertion still needs confirmation through additional studies. Indeed, as the gas shells
expand and are moving away from the center of the nebula, the radial distance is intimately
related to the time evolution of the gas.
The detection of C60 has also been reported in the NGC 2023 and NGC 7023 reflec-
tion nebulae, respectively located in the Orion and Cepheus constellations (Sellgren et al.
2010), using the Infrared Spectrograph (IRS) onboard the Spitzer space telescope2 . Here,
fullerenes have been detected in an environment where PAHs are also present, in contra-
diction to the case of Tc 1 investigated by Cami et al. (2010). In addition, the study of
Sellgren et al. (2010) provides some evidence for C60 excited by UV radiation in gas-phase,
and not on dust grains. These facts strongly suggest that the physico-chemical conditions
where fullerenes could be formed are wide.
Berné & Tielens (2012) recently proposed that C60 could form through photochemical
processing of large PAH molecules. In this scenario, the strong UV field from massive stars
(or even in inner regions of proto-stellar disks where accretion produces UV radiation)
leads to the dehydrogenation of PAHs (breaking of the weakest chemical bondings) in the
relevant size range (i.e. 60–70 C atoms) and to the formation of pentagonal cycles at the
edge of the molecules. The latter defects in the structure lead to the curvature of the
molecule. Isomerization allows pentagons to migrate and finally to close the molecule in
the stable form of a fullerene.
2
The first spectral features attributable to interstellar fullerenes were in fact reported by Werner et al.
(2004) and Sellgren et al. (2007), but the definitive confirmation was provided by Sellgren et al. (2010).
Figure 3.11: Schematic representation of the interstellar chemistry likely to lead to the
formation of fullerenes. A. Chemical evolution of PAHs in the interstellar medium under
the influence of UV radiation. B. Schematic conversion of graphene into C60 . Figure taken
from Berné & Tielens (2012).
3.1.3 Organic compounds with functional groups

A significant leap forward on the way of chemical complexity is achieved when functional
groups are included in hydrocarbons. Molecules containing various functional groups have
been detected in several astrophysical environments. A few examples are given below:
- Methanol. This is the most simple organic compound containing an hydroxyl group.
It has been detected in the interstellar medium by Ball et al. (1970).
- Ethanol. Ethylic alcohol have been detected in the 70’s (Zuckerman et al. 1975).
- Dimethylether. This is the simplest molecule containing an ether function. Snyder

et al. (1974) reported on its presence in the Orion Nebula. The probable formation
mechanism of this molecule has been discussed by Peeters et al. (2006).
- Metanethiol. This is the simplest mercaptan, i.e. an organic molecule containing

a thiol functional group. The detection of this molecule in the interstellar medium
was first reported by Linke et al. (1979).
- Cyanomethane. This is the smallest organic molecule found in the ISM containing
a nitrile group. Cyanomethane has been detected in molecular clouds by Solomon
et al. (1971).
- Formaldehyde. This is the simplest molecule containing an aldehyde functional

group. It is found in many astrophysical environments, and its first detection was
reported by Snyder et al. (1969).
- Acetaldehyde. This is the equivalent of the formaldehyde, but with a hydrogen

substituted by a methyl group. Several studies led to the detection of this molecule
in the interstellar medium (Gottlieb 1973; Fourikis et al. 1974a; Gilmore et al. 1976).
- Formic acid. This is the simplest carboxylic acid. It has been detected in the inter-
stellar medium by Zuckerman et al. (1971) and Winnewisser & Churchwell (1975).
- Acetic acid. The discovery of acetic acid by Mehringer et al. (1997) added a new
chemical species from the class of carboxylic acids in the census of interstellar
molecules.
- Acetone. This is the simplest organic molecule including a ketone functional group.
This molecule has been reported in the interstellar medium by Combes et al. (1987)
and Snyder et al. (2002).
- Methylformate. This is the simplest organic compound including an ester functional

group. Its presence has been reported in the direction of the Galactic Center by
Brown et al. (1975) and Churchwell & Winnewisser (1975).
- Formamide. This is the simplest organic molecule including an amide functional

group. Its detection dates back in the early 70’s (Rubin et al. 1971).
- Methylamine. This is the simplest primary amine. Its presence has been reported in
various places of the interstellar medium (Kaifu et al. 1974; Fourikis et al. 1974b).
- Ketenimine. This molecule is simultaneously a representant of the classes of imines

and ketenes. Its was discovered in a star forming region by Lovas et al. (2006).
This short (and incomplete) census of interstellar organic molecules shows that, among
the functional groups usually found in organic chemistry, many have already been identified
in space. This is rather promising in the sense that most of the organic representatives
of chemistry are present in the interstellar medium, and one may expect most of their
chemistry to take place in various astrophysical environments. Their detection constitutes
in addition a strong evidence that efficient chemical processes likely to produce complex
molecules are at work in space.
A first illustration of the basic interstellar chemistry likely to lead to such compounds
is shown in Fig. 3.12. This example shows the main processes leading to the formation and
destruction of a rather abundant molecular species that is very important for chemistry:
formaldehyde.
The two main mechanisms for its formation are the following:
- the neutral-neutral reaction between methyl radical and an oxygen atom. As em-
phasized in Sect. 2.1.3, chemical processes between neutral partners become really
significant when radical species are involved, because of the lack of significant activa-
tion barrier likely to affect the kinetics of the reaction. This is typically what happens
here. In addition, the formation of the new chemical bonding with the oxygen occurs
as a H group dissociates, taking away the excess energy likely to destabilize the new
molecule.
CO
OH
ν +
+
H3 e− C
CO 2
ν
S
+ ν +
+ e− C
C + +
HCO METAL S O
+ O
C +
+ C
O He O
CH3 H 2CO HCO CH2
+
H + e−
C
+
+ H3
H 2CO
+
H3
e− e−
+
H 3CO
Figure 3.12: Simplified chemical network including the main reactions involved in the
formation of formaldehyde in interstellar clouds (adapted from Prasad & Huntress 1980b).
- the dissociative electronic recombination of a cationic species. Once again, provided

the cationic precursor is sufficiently abundant, it leads fastly to the formation of
formaldehyde.
On the other hand, ionic processes such as charge transfer and ion-neutral reactions are
efficient processes leading to its destruction. It should also be noted that, provided the
astronomical environment is translucent enough to allow ultraviolet photons to interact
with it, photodissociation is an additional efficient destruction mechanism.
An additional leap forward in the sense of increasing molecular complexity is achieved if

more than one functional group is included in the same molecule. This is typically what is
found in a few molecules that have been identified in space in the last past years. So far, five
interstellar molecules belong to this catogory: glycolaldehyde (Hollis et al. 2000), ethylene
glycol (Hollis et al. 2002), amino acetonitrile (Belloche et al. 2008), cyanoformaldehyde
(Remijan et al. 2008) and cyanomethanimine (Zaleski et al. 2013). These molecules have
been discovered in the Srg B2 molecular cloud complex located close to the center of the
Galaxy. The presence of organic molecules of that complexity is the signature that a
complex chemistry is at work in such molecular clouds.
Among these molecules, it is interesting to find the first molecule of the carbohydrate
series: glycolaldehyde. This is especially important if one keeps in mind the important role
played by sugars in our complex biochemistry. The exact mechanism likely to lead to its
synthesis in the ISM is still not firmly determined. A polimerization-like formation process
involving formaldehyde molecules is a possibility, considering the wealth of formaldehyde
found in interstellar clouds. It has been proposed that a process similar to the formose
reaction3 on grain surfaces in molecular clouds could be efficient (see e.g. Hollis et al.
2000).
3
The formose reaction is a polymerization reaction of formaldehyde catalyzed by water (Larralde et al.
1995).
Cyclic molecules containing heterotatoms, i.e. atoms other than carbon, are also worth
considering. In the context of the study of chemistry in space and of its relation with
Earth’s biochemistry, several heterocyclic molecules such as heterocyclic bases are of crucial
interest. The latter molecules belong to the category of building blocks of the genetic
material of living beings on Earth. To date, only a few heterocyclic molecules have been
found in the interstellar medium: c-C2 H4 O (Dickens et al. 1997), and presumably c-C2 H3 O
and c-C2 H5 N. In addition, the analysis of meteoritic organic matter revealed the presence
of various more complex heterocyclic compounds, containing one or more nitrogen atoms,
including pyrimidines, purines, and derivatives of quinoline and isoquinoline (Stoks &
Schwartz 1981, 1982). In the interstellar medium, recent investigations failed to detect
purines and pyrimidines (see Kuan et al. (2003b) for the case of pyrimidine in interstellar
clouds). However, three points deserve to be mentioned here:
- the large abundance of carbon rich polycyclic molecules (PAHs) provides strong
evidence for the fact that very efficient processes are at work in the interstellar
medium to build cyclic and polycyclic organic compounds.
- the large abundance of compounds such as nitriles or cyanates allows us to expect

some of these compounds to be involved in the processes responsible for the formation
of unsaturated cyclic organic compounds (see e.g. Ricca et al. 2001a,b).
- in the case of comets, for instance, large complexes of organic compounds made
of carbon and nitrogen are likely present. These complexes are expected to con-
tain many cyclic units including simultaneously carbon and nitrogen, in a way that
is reminiscent to the structure of nitrogen-bearing heterocyclic compounds. This
fact constitutes an important hint for the existence of purine- and pyrimidine-like
compounds in space.
Figure 3.13: Influence of the number of N atoms incorporated in monocyclic aromatic

compounds on the stability against photolysis. (taken from Peeters et al. 2005).
However, if a molecule such as adenine (C5 H5 N5 ) is formed in the ISM, it is unlikely

to proceed through the successive neutral-neutral addition of HCN molecules, as demon-
strated by Smith et al. (2001).
It should also be kept in mind that the stability of pyrimidine, and other nitrogenated
heterocycles against UV photolysis is rather limited and far below that of monocyclic
and polycyclic aromatic hydrocarbons (Peeters et al. 2005). In addition, the stability
of N-heterocycles against photolysis decreases significantly as the number of N atoms
incorportated in the ring increases (Fig. 3.13). Any circumstellar N-heterocyclic molecule
formed on dust grains would likely be destroyed by UV photons or cosmic-rays after
rejection in the gas phase of the ISM. Alternatively, such molecules would more probably
be formed on small interplanetary bodies such as those found in the form of meteorites on
planetary surfaces (Peeters et al. 2003).
3.2 Important scientific issues

3.2.1 The quest for amino acids
The discovery of amino acids in space is really a challenge for molecular spectroscopists.
Such a discovery would constitute a significant improvement of our view of the molecular
content of the Universe, as it would contribute to fill the gap between currently confirmed
interstellar molecules and Earth-like biochemistry. A few years ago, the announcement
of the discovery of glycine in the interstellar medium by Kuan et al. (2003a) has been
questioned. Several studies aimed at confirming this putative discovery but confirmation
is still lacking (see e.g. Snyder et al. 2005). To date, the detection of glycine in the
interstellar medium is still not admitted by the astrochemical community.
More recently, glycine has been firmly identified much closer: in the coma of a comet.
This detection is credited to Stardust, a NASA mission designed to collect interplanetary
dust and to launch the sample in the Earth’s atmosphere for detailed analysis (for addi-
tional information, see http://stardustnext.jpl.nasa.gov/). The Stardust spacecraft
passed through gas and dust surrounding the icy nucleus of the comet Wild-2, in 2004. In
addition, the extra-terrestrial origin of the glycine collected by Stardust has been firmly
established thanks to a careful measurement of isotopic ratios (Elsila et al. 2009). The
confirmation of the presence of molecules such as amino acids out of the Earth’s atmo-
sphere lends significant support to the idea that astronomical environments could have
played a pivotal role in prebiotic chemistry.
Scenarios to produce such molecules in astrophysical environments are investigated by

astrochemists. A pathway that might lead to the formation of many complex molecules,
including amino acids, is that of condensed matter chemistry. Such a chemistry is likely
to take place in solid bodies that are found in several astrophysical environments, such as
icy grain mantles (see e.g. Bossa et al. (2009), Woon (2011), and references therein) or
comets. It is important to note that in solid-state matrices, the mobility of molecules is
not completely inhibitted. In addition, the confinement of chemical species in such a small
volume favors interactions, and potentially reactions:
- Among the recent molecular identifications in the interstellar medium, that of amino
acetonitrile deserves a few more comments. The nitrile functional group is indeed
likely to be converted into a carboxylic acid through a hydrolysis reaction, leading
directly to the first natural amino acid: glycine. This is not clear whether astro-
nomical environments are likely to allow for such a reaction (in solution, this is an
acid-catalyzed process), but it is important to note that amino acetonitrile is a direct
precursor of the simplest amino acid. In solid phase, the interaction between organic
compounds containing a cyano group and water molecules is a scenario worth to be
investigated.
3.2. Important scientific issues 67
- The formation of amino acids may proceed through a Strecker-like process, starting
from simple and abundant molecules known to be present in icy grain mantles and
comets, i.e. ammonia, formaldehyde, water, CN-bearing molecules, etc (Rimola et al.
2010). Once again, one should keep in mind that even though the Strecker synthesis
of amino acids occurs in solution, a similar process in solid matrices can not a priori
be completely ruled out. In this context, aldehyde would lead to α-H-amino acids.
In particular, glycine would be synthesized with formaldehyde as a starting point.
- It is strongly believed that comets (and potentially their precursor icy grain man-
tles) contain large amounts of refractory macromolecular compounds, mainly based
on carbon, nitrogen and oxygen. Such a material could contain some kind of HCN
polymers (see e.g. Rettig et al. 1992; Matthews 1995a,b) and similar compounds in-
cluding oxygen (in the form of ether or carbonyl groups). These complex compounds
are likely to undergo for instance hydrolysis, likely to release significant amounts of
smaller organic compounds, including notably amino acids and other relevant prebi-
otic compounds. In addition, photolysis of such compounds will also produce small
radicals likely to react efficiently and to lead to the formation of various organic
compounds.
Many studies are currently in progress to investigate reaction pathways in icy grains,
including laboratory studies performed on ice analogs aiming at simulating interstellar
icy grains (see e.g. Duvernay et al. 2010; Danger et al. 2011). Even though the mecha-
nisms ruling such reactions in icy matrices are not fully understood, laboratory studies on
ice analogs demonstrate that complex and interesting organic compounds can be formed
starting from simple initial conditions: the presence of a few simple reaction partners,
combined with the action of heat or light, or even energetic particles. Such environments
are therefore potentially relevant notably for the formation of amino acids. In this context,
the study of meteorites is especially interesting in the sense that many amino acids have
already been identified. The most famous exemple is that of the Murchison meteorite that
crashed in Australia in 1969, and whose detailed analysis revealed the presence of many
organic compounds, including α-H and α-Me amino acids (Cronin & Pizzarello 1997; Engel
& Macko 1997).
3.2.2 The homochirality problem

Provided chemical compounds have reached a significant level of complexity, new proper-
ties are likely to appear, notably in relation with the stereochemistry of molecules. In the
case of sugars, stereoisomers can be distinguished using conventions leading to the dis-
tinction between D and L sugars. The simplest chiral monosaccharide is glyceraldehyde,
whose natural enantiomer is denoted (+) as it rotates the plane polarized light in a clock-
wise direction. (+)-glyceraldehyde is known to have the R configuration at its stereogenic
center, and its Fischer projection is shown in Fig. 3.14. This enantiomer with the R con-
figuration is conventionnaly called D-glyceraldehyde. For monosaccharides, the generally
adopted conventions consider that the carbonyl is represented on top of the projection.
In the projection of glyceraldehyde, one can see that the hydroxyl group attached to the
chiral center is on the right side. By analogy with the case of glyceraldehyde, all sugars
with the same configuration on the chiral carbon opposite to the carbonyl group will be
referred to as D sugars. A few examples are provided in Fig. 3.14. It is important to note
that D and L notations have no relation to the direction in which a given sugar rotates
plane-polarized light. A D sugar may be dextrorotatory or levorotatory. The notation
only refers to the stereochemistry of the last chiral carbon.
The D and L notations are also used to distinguish enantiomers of amino acids. In the
latter case, the stereochemical similarity of natural serine with L-glyceraldehyde (where
the hydroxyl group is replaced by the amino group) leads to similar conventions: in Fischer
projections, amino acids with the amino group on the left (S configuration) are referred to
as L amino acids. Here again, the optical activity is not only related to the configuration
of the chiral center in the amino acid. Among the 20 natural amino acids, 19 are chiral;
and among these 19 L amino acids, 9 are dextrorotatory!
H O H O
C C
H O
C H OH H OH
H OH H OH HO H
H OH H OH
CH2OH
H OH
CH2OH
CH2OH
D − glyceraldehyde D − ribose D − glucose
H O HO O
C C
HO H H 2N H
CH2OH CH2OH
L − glyceraldehyde L − serine
Figure 3.14: Upper part: Fischer projections of D monosaccharides. The assignement of

the D notation refers to the fact that the configuration of the last chiral carbon is R, i.e.
with the hydroxyl group on the right. Lower part: Fischer projections of the amino acid
serine, whose DL-notation is attributed by analogy with glyceraldehyde.
Homochirality is a term that refers to a group of compounds with the same sense
of chirality. In biology, homochirality is found inside living organisms. Active forms of
amino acids are almost exclusively of the L-form4 and most biologically relevant sugars
are of the D-form. The stereochemistry of organic compounds is of crucial importance
in biochemistry as it is responsible for the three-dimensional configuration of complex
molecules such as proteins and (deoxy)ribonucleic acids. If the stereochemistry of the
sugars in nucleotides and of amino acids in proteins was not strictly controlled, biologically
important molecules would have very different properties. The fact that we find only L-
type amino acids and D-type sugars is therefore a major element in the complex chemistry
of living organisms.
The question here is the following: where did this homochirality come from? This
question is really a challenge for chemistry... and maybe a question worth asking to astro-
4
It should be noted that D amino acids are also used in our biochemistry, in particular in peptidoglycans,
i.e. composite macromolecules made of crosslinked polysaccharides and peptides found in cell membranes
of bacteria.
chemistry.
The problem comes from the fact that without any contraint, the synthesis of chiral
molecules leads to racemic mixtures. This can be illustrated, for instance, in Fig. 3.15
in the case of the nucleophilic addition on a carbonyl group. Two major cases can be
considered:
- Gas phase process. If one considers a plane trigonal carbonyl group in gas phase,
the approach of the nucleophilic partner is likely to occur with equal probability on
any side of the plane. If a stereogenic center is created, this nucleophilic addition
will lead to the two different enantiomers depending on the approaching direction.
- Grain surface process. The carbonyl bearing molecule will first adsorb on the grain
surface. The molecule has a priori equal chance to adsorb on one side or the other.
The approach of the nucleophilic partner will occur on the only side that faces the
gas phase. In this case, the enantiomeric selection occurs when the molecule adsorbs
on the grain surface, and not during the nucleophilic attack.
Gas phase process

Nu
NuH
C
R
OH
R R’
C O 2 enantiomers
R’ R’
OH
R
C
NuH
Nu
Grain surface process
R NuH
C O
Nu
R’
R
C
C O R
OH
R’ R’
1 2 3
2 enantiomers
R’ NuH
C O
Nu
R
R’
C
C O R’
OH
R R
Figure 3.15: Upper part: nucleophilic addition on a carbonyl group in gas phase. Lower
part: nucleophilic addition on a carbonyl group on a grain surface.
In both case, we see that such an addition leads to the two enantiomeric forms of the
reaction product, without any selectivity. Both processes lead to a racemic mixture. Basic
synthesis processes do therefore not suggest that any enantioselectivity can be expected
in the production of chiral molecules.
Symmetry breaking
As basic synthesis processes do not seem to lead to enantiomeric excesses of chiral com-
pounds, two scenarios may be considered to break the symmetry:
- The synthesis of chiral compounds may be constrained by enantioselective interac-

tions in the context of rather complex processes. This is what occurs in living beings
on Earth where for instance enzymes act as enantioselective catalysts. But in inor-
ganic environments, such a selectivity is a priori unlikely. As biological systems make
use of specific enantiomers, where did the first microorganism find the enantiomeric
excess that triggered the homochirality of our biochemistry?
- That is not during the synthesis of chiral molecules that the enantiomeric symmetry
is broken, but during their destruction. This is where the impact of the astrophysical
environment is likely to play a role.
(a)
Aligned dust
particles
Asymmetric photolysis
Stars
Interstellar cloud
Unpolarized UV light Circularly

polarized UV light
(b) Symmetric photolysis
Stars
Interstellar cloud
Figure 3.16: Illustration of the impact of an asymmetry on the astronomical scale on the
asymmetry on the molecular scale, through circular polarization of light due to magnet-
ically aligned dust particles. (a) asymmetric photolysis due to circularly polarized light.
(b) symmetric photolysis in the presence of unpolarized light.
The latter scenario is worth considering in the context of the interstellar medium per-
meated by UV light likely to break chemical bondings, and therefore destroy molecules. As
already discussed, photodissociation of chemical species is a common and efficient process
in the interstellar medium. In addition, it is interesting to note that circularly polar-
ized5 light may constitute the asymmetry seed needed to break the symmetry between
enantiomeric forms of a given compound (i.e. D and L amino acids or sugars). Ultravio-
let circularly polarized light does indeed generate a small enantiomeric excess (ee) in an
initially racemic mixture through asymmetric photochemical destruction of L and D enan-
tiomers at different rates, i.e. asymmetric photolysis. In other words, the kinetic constants
for the photolysis of L and D enantiomers through irradiation by circularly polarized light
are slightly different. Both enantiomers are therefore destroyed, but one of them with a
5
Light from neighboring stars can be circularly polarized after crossing a region of the interstellar
medium where non-spherical dust grains are aligned under the influence of a large scale magnetic field.
rate that is slightly lower than the other. In this model, asymmetry on the molecular
scale results from asymmetry on the astronomical scale. According to such a scenario,
the homochirality should affect the solar neighborhood (and not only the Earth) in the
same sense. However, an opposite homochirality cannot be completely ruled out in other
regions of our Galaxy, or elsewhere in the Universe. The general scenario is summarized
in Fig. 3.16. Useful references for this scenario are Cerf & Jorissen (2000) and Jorissen &
Cerf (2002), along with Nuevo et al. (2006), Takano et al. (2007) and de Marcellus et al.
(2011) for reports on laboratory studies aiming at producing small enantiomeric excesses
through the action of circularly polarized light.
Once the racemic mixture is converted into a scalemic mixture (i.e. with a small enan-
tiomeric excess), we are still very far from homochirality. It has been proposed that the
enantiomeric asymmetry may be amplified by some kind of autocatalytic process, favoring
the synthesis of a particular enantiomer in the presence of a given enantiomeric excess.
Such an asymmetric autocatalysis has been described for instance by Shibata et al. (1998).
It is indeed known – and in several cases understood – that the incorporation of a chiral
agent in a process can lead to an enantioselective synthesis. To some extent, we could
say that asymmetry is contagious. On the other hand, one can also envisage that the
incorporation of small enantiomeric excesses in living organisms may lead to a progressive
selectivity in the incorporation of one particular enantiomer. The latter scenario is also
worth considering in order to investigate the complex question of homochirality, but it is
still a matter of debate.
Racemization processes
Beside the existence of processes potentially responsible for the breaking of the racemic
symmetry, it should be mentioned that processes may also contribute to restitute this sym-
metry: i.e. racemization processes. The importance of racemization processes is discussed
by Cataldo et al. (2005). Typically, racemization denotes any reaction which converts
an enantiomer into its optical counterpart until an equimolar mixture of the two optical
antipodes is achieved. In other words, it is a process that converts a scalemic mixture into
a racemic mixture. In this context, we may consider:
- action of heat, even though heat has only little chance to be high enough to redis-
tribute chemical bondings in most astrophysical environments.
- action of light, i.e. photoracemization, through UV or X-ray photons
- action of high energy particles, i.e. radioracemization. This is typically what is likely
to occur through the interaction of cosmic rays with molecules. In the particular case
of obscured interstellar clouds, this process is more likely than photoracemization.
Schematically, racemization can be illustrated by the following relations
2 R −→ R + S
or
2 S −→ R + S
where R and S represent respectively two enantiomers of a given molecule.
On Earth, in aqueous solution, chiral molecules such as most of the natural amino
acids are likely to undergo a unimolecular transformation called tautomerization. If one
considers the natural chiral amino acids, the chiral carbon is bonded to one hydrogen. The
tautomerization takes place when this latter α-hydrogen is removed before the breaking of
the double bond and the formation of another one between the α and β carbons to yield
an enol (see Fig. 3.17). This enol is not chiral anymore, and can convert to the carboxylic
acid with equal probability to give a R or S configuration at the stereogenic center on the
α position. This tautomerization process has thus the net result to convert a scalemic
mixture into a racemic one. The occurrence of this process should not be ruled out in
astrophysical environments as well, e.g. on dust grains or in icy mantles of dust grain, or
even in comets. However, it should be noted that the carbonyl form is a priori intrinsically
much more stable, and actually in solution the tautomerization requires a basic catalysis.
The enol form may however be more stable if the double C-C bonding participates in a
conjugated system. In such a case, tautomerization is not expected to be a very efficient
racemization process.
S − alanine S − alanine
O O
N C N C
H H
C OH C OH
H H H
O
H C H C
H H H H
N C
H H
H
C OH
H
O O
C
H H
N C N C
H
H H
C OH C OH
H H
Enol
C H C H
H H H H
H H
R − alanine R − alanine
Figure 3.17: Tautomerization of alanine converting to an enol with loss of chirality, followed
by the restoration of the amino acid.
It should be noted that, as far as amino acids are concerned, tautomerization may only
be effective if one is dealing with α-H-amino acids, such as natural amino acids. In the
case of α-methyl amino acids where there is no hydrogen bonded to the chiral carbon,
the tautomerization cannot occur. This is the main reason why α-Me-amino acids are
likely to retain their configuration more efficiently than α-H-amino acids. This is what is
observed in the sample of amino acids that have been found in the Murchison meteorite.
Indeed, the enantiomeric excesses measured for α-Me-amino acids are higher than those
of α-H-amino acids.
The homochirality problem is a complex issue, that is far from being solved by modern
science. However, it seems that some relevant elements may be found in interstellar regions,
and not necessarily only on Earth even if prebiotic ages are considered. This longstanding
issue still needs to be investigated in detail, and most probably following a pluridisciplinary
approach.
3.2.3 Molecular filiation

As emphasized in Chapter 1, planetary systems are made of material initially present in
molecular clouds, themselves corresponding to the densest parts of the interstellar medium.
A natural question arising from this fact is that of the potential filiation of molecular species
across all steps of the transformation of the molecular cloud into a planetary system. A
schematic view of the latter transformation is given in Fig. 3.18.
In order to address the complex issue of the molecular filiation in astrophysical en-
vironments, each step of the evolution described in Fig. 3.18 deserves a discussion, with
emphasis on the physico-chemical conditions governing the chemistry.
(a) The quiet molecular cloud

The part (a) in Fig. 3.18 constitutes the starting point of the interstellar material that is
available to form molecules. As already mentioned earlier, the molecular cloud constitutes
the densest part of the interstellar medium. It is dominantly made of neutral molecules,
with some atoms, ions, electrons and significant amounts of dust grains. The central
parts of a molecular cloud are not permeated by strong radiation fields likely to play a
significant role in any photochemical process. However, the outer part of the cloud might
be interacting with such a radiation field. In addition, UV photons produced in situ
through the interaction of cosmic-rays with the cloud material should also play a role in
its chemical activity.
Typically, the molecular cloud from which our Solar system was made about 4.5 billion
years ago, along with other present molecular clouds in the Galactic plane, are made of
material that has been processed by generations of stars. As a result, the cloud mate-
rial that is considered here is rich in chemical elements with atomic number larger than
2. Many elements such as C, N, O, S, P, Si, or Fe populate such clouds. Most of these
elements were nucleosynthesized in the core of massive stars through nuclear fusion, be-
fore rejection of some of this processed material in the interstellar medium. In addition,
during the supernova explosion, fast neutron captures (followed by beta disintegration)
led to a subsequent processing of nuclides, leading to many other chemical elements in
the interstellar environment. Finally, evolved intermediate mass stars contributed also to
the nucelosynthesis of other nuclides through slow neutron capture (also followed by beta
disintegration), and rejected these elements in the interstellar medium. The role of stellar
evolution on the chemical enrichment of the Galaxy has been briefly described in Chapter 1.
Tens, or even hundreds of stars may have contributed to the enrichment of the pre-Solar
cloud. The chemically enriched interstellar medium is not static, and its dynamic allows
some mixing of its material. When the dynamics of interstellar clouds leads some of its
parts to become denser, the physical conditions become more favorable for the formation
of molecular species: typically, the interstellar cloud is a molecular cloud, including many
chemical elements likely to be involved in the formation of molecular species.
A summary of the chemistry taking place in ’quiet’ molecular clouds (i.e. molecular
clouds that are not collapsing into a proto-stellar object) is given in Fig. 3.19. As explained
in Chapter 2, a molecular cloud gives rise to two classes of processes: (i) gas-phase processes
and (ii) surface processes (the latter are likely to take place on the surface of dust grains,
or even on large molecular species such as extended PAHs). In gas phase, cationic and
neutral reaction chains are likely to be at work, with a significantly enhanced efficiency
for cationic processes (mostly triggered by the interaction of cosmic-rays with molecules
and atoms). Even though cationic processes are a priori more efficient, neutral reaction
chains should not necessarily be neglected for abundance reasons. Indeed, as neutral
(b)
(c)
(a)
(d)
Figure 3.18: Schematic view of the transformation of a molecular cloud into a planetary
system: (a) molecular cloud, (b) hot core made of collapsing gas, (c) proto-stellar system
with its disk, (d) planetary system with the young star at its center. Details on each step
are given in Figures 3.19, 3.20, 3.22 and 3.24.
species are much more abundant than cationic ones, the dependence of kinetics on the
number densities may make neutral reaction chains quantitatively significant. On grain
surface, the efficiency of chemical processes is amplified by the enhanced probability of
interaction between reaction partners adsorbed on the same particle. As a result, even
neutral species can interact significantly and efficiently provided they are significantly
adsorbed on surfaces.
Gas phase processes
Cationic chain
Atoms Neutral chain
(Photochemistry)
Cosmic−ray processes
Electrons
Ions
Molecules Accretion Desorption
Dust grains
Surface processes
Dust grains
PAHs (?)
Figure 3.19: Summary of the processes governing the chemistry in a molecular cloud.
Obviously, the two reaction pathways (gas phase and surface) are not completely unre-
lated. Strong connections exist between these processes. A first connection is the accretion
of chemical species onto surfaces from the gas phase. Such a process is dependent on the
density, and on the abundances of dust (or PAH) particles in the cloud. A second con-
nection is that of desorption of chemically processed or un-processed species, going or
returning to the gas phase. Desorption is dominated by thermal and mass considerations:
high dust temperatures favor desorption, and heavier species desorb more slowly. In a
quiet molecular cloud, we may as a first approximation consider that some kind of sta-
tionary state is achieved, i.e. the size of dust particles is nor increasing through accretion,
neither decreasing through efficient evaporation when density conditions are not changing.
The ’quiet’ phase of a molecular cloud, during which the chemical processes summarized
above should be at work, can be of the order of a few 10 up to a few 100 million years.
Once such a molecular cloud starts to collapse (arrow number 1 in Fig. 3.18), it enters the
star formation phase, and the physical conditions start to change significantly.
(b) The proto-stellar object, in class 0 phase

Once the contraction of the molecular cloud has started, the density of the gas increases.
The cloud collapse can be triggered for instance by the propagation of a shock wave in the
interstellar medium because of a supernova explosion. The first step of the star formation
process is reached, and the collapsing molecular cloud is called a Class 0 proto-stellar ob-
ject (see Fig. 1.5 in Chapter 1). The processes governing the physico-chemistry of a Class
Gas−phase processes
Kinetics favored by
− increased density
− increased temperature
Accretion Desorption
Surface processes
Accretion favored with
respect to desorption
growth of dust grains H2
CHON
H2O H2 NH3
CO
Sublimation
H2
CO H2 H2 NH3
CHON
H2O H2
H2O H2O
H2 H2O H2
CO
CO
NH3 CHON
H2 H2O CO
CHON H2 H2
H2 CHON
NH3 H2
CHON
CO
H2O
H2
Dirty ices CHON

CHON NH3
Sublimation of icy NH3
grain mantles CHON
CHON CHON
rejection of many molecules NH3
in the gas phase, where gas
CHON CHON
phase chemistry goes on
NH3 NH3
significant enrichment of
gas phase with molecules CHON
processed on dust grains CHON
CHON
NH3
NH3
Figure 3.20: Summary of the processes governing the chemistry in a Class 0 proto-stellar
object, at the starting phase of the gravitational collapse of the molecular cloud. Both
gas phase and grain surface processes are at work, with an increased accretion rate lead-
ing to the formation of icy grain mantles containing many molecular species (including
CHONs, i.e. organic molecules made of C, H , O and/or O). In the central part where
the temperature is the highest, sublimation of the icy grain mantles will lead to the rejec-
tion of material in gas phase. A strong dependence of the physico-chemical properties is
established as a function of the radial distance.
0 object are summarized in Fig. 3.20.
The increase of the density will have three main consequences, as far as chemical
processes are concerned:
1. Enhanced kinetics through density factors. As the number densities of chemical

species are increasing, the probability of interaction between reaction partners in
gas phase will increase as well.
2. Enhanced kinetics through rise of the temperature. The temperature dependence

of reaction rate constants of elementaty processes will favor the kinetics of most
processes.
3. Enhanced accretion rate onto dust grains. The increasing density will accelerate the
accretion of chemical species on dust particles. As a result, the surface coverage will
increase, therefore favoring the surface reactions. In addition, these conditions will
favor the occurrence of Eley-Rideal processes, in addition to Langmuir-Hinshelwood
processes governing low surface coverage regimes. Finally, the increased accretion
rate will quickly render the ’monolayer’ approximation irrelevant: the growth of the
dust grain will lead to the formation of icy mantles including many chemical species.
It should be emphasized that a gradient of physical conditions is present: the density

and temperature increase from the outer layers of the collapsing cloud to its center. As a
result, a gradient of chemical behaviour is a priori expected.
Figure 3.21: Temperature and par-

ent molecules abundance profiles
in a hot core. r100K is the ra-
dius where the temperature reaches
about 100 K, i.e. the sublima-
tion temperature of icy mantles
leading to the liberation of parent
molecules that formed initially on
dust grains during the pre-stellar
phase. The figure is taken from Bot-
tinelli (2006).
Hot cores and hot corinos correspond to the central part of Class 0 objects respectively
for massive (> 8 M ) and low mass (a few M ) proto-stars. They are compact (about
0.1 pc for a hot core, and 150 AU for a hot corino). The density is higher than 107 cm−3 ,
up to about 108 – 109 cm−3 in the central part. The temperature of the central part may
be higher than 100 K. The latter temperature is crucial in the sense that is corresponds to
the sublimation temperature of most of the chemical species constituting icy grain mantles.
As a result, many molecular species are restituted to the gas phase, where chemical reac-
tions can lead to the formation of more complex molecules. Typically, some of the most
abundant molecules that sublimate are formaldehyde, methanol, ammonia and water. In
Fig. 3.21, the temperature profile of the hot core and its impact on molecular abundances
are illustrated. The occurrence of sublimation is responsible for the structure of the Class
0 object, with a central hot core and a colder outer envelope. In the latter, the icy grain
mantles are likely to grow further, with a desorption rate that is controlled locally by the
temperature.
As the collapse goes on, the cloud will flatten and it will result in the formation of a
disk orbiting around the protostar. The transition to that proto-stellar evolutionary stage
corresponds to the arrow number 2 in Fig. 3.18.
(c) The proto-stellar object, with accretion disk
Proto-stellar phases coinciding with the existence of a proto-stellar disk correspond to

Class I and Class II objects. In the former case, the accretion of material from the initial
cloud onto the disk and the proto-star is still going on, while in the latter case most of the
cloud material is in the proto-star and in its cricumstellar disk. The situation depicted
in Fig. 3.22 is illustrative of the circumstances governing chemical processes in any disk-
surrounded proto-star.
The ongoing contraction of the cloud material leads to unprecedented densities with
respect to typical molecular cloud values. The central parts of the proto-stellar disk (mid-
plane) can reach values up to several 1013 cm−3 , with a rapid decrease as a function of
the radial distance. At such values, termolecular processes may be significant, and should
not be neglected in any quantitative astrochemical model. Such high values are relevant
to the inner parts of the disk, typically up to several AU from the central proto-star.
These regions are also characterized by temperatures of the order of several 100 K close to
the center, decreasing below the threshold of 100 K at several astronomical units. These
trends computed from a dynamical proto-stellar disk model are illustrated in Fig. 3.23,
taken from the study of Markwick et al. (2002). Such a gradient in physical conditions
will have a strong impact on the chemical processes likely to take place in proto-stellar
disks.
It should also be emphasized that the chemical species present in such an environment
are confronted to peculiar conditions in terms of ionization. On the one hand, the disk
material is likely to be affected by the two ’classical’ ionization processes already taking
place in molecular clouds:
- Photoionization by UV photons from neighboring bright, early-type stars.
- Cosmic-ray induced ionization, through the interaction of high energy particles with
atoms and molecules in the disk.
On the other hand, additional local sources of ionization are also considered by recent
models of proto-stellar disk chemistry:
- Photoionization by UV photons produced by the central proto-star.

Cosmic particle
accelerators
CR
Bright stars
UV
Accretion
f(R, z)
Surface processes
Gas−phase processes Still at work
Enhanced by Icy grain mantles
− increasing temperature and density
− additional ionization sources Sublimation (inner regions)
Condensed matter chemistry (?)
Desorption
f(R, z)
CR UV (IS)
UV
Molecular
species
X−rays
26
Al
Figure 3.22: Summary of the processes governing the chemistry in a proto-stellar object
with accretion disk. Both gas phase and grain surface processes are still at work. The
physico-chemical properties in the disk depend intimately on the radial distance (R) and
on the vertical coordinate measured with respect to the midplane (z). The participation
of additional ionization sources, with respect to the usual interstellar ones, is emphasized.
Note that this schematic view is mostly relevant to Class II objects, even though the
black arrows pointing to the disk and to the central proto-star illustrate an cloud-to-disk
accretion process mostly relevant to the case of Class I objects.
- Photoionization by X-ray photons produced by the central proto-star, through mag-

netic reconnection processes. In addition, it is important to note that photoelectrons
produced by X-rays are likely to collide with chemical species and excite them, before
radiative relaxation generating UV photons likely to photoionize other species.
- Ionization induced by the radioactive decay of unstable isotopes present in significant

amounts in proto-stellar material. In particular, 26 Al decays into 26 Mg that de-
excites radiatively by emitting a photon of energy 1.809 MeV. The decay could also
be accompanied by the rejection of a positron, likely to interact with electrons,
producing additional gamma-ray photons.
It should however be noted that the attenuation length of ionizing photons in disks is
relatively short, and the ionization rate should be of the order of 10−13 –10−15 s−1 in the
range between 1 and 10 AU. This is in the midplane region, i.e. a region that cannot be
reached by photons due to the strong opacity of the medium, that the ionization induced
by radionuclides is the most significant. In the upper layers of the disk, X-ray induced
Figure 3.23: Variation of

the midplane density (up-
per panel) and temperature
(lower panel) in the inner part
of a proto-stellar disk. These
curves have been computed
for typical parameters of a
Sun-like proto-stellar object.
These regions are character-
ized by unprecedented physi-
cal conditions with respect to
’quiet’ molecular clouds. The
figure is taken from Markwick
et al. (2002).
ionization may dominate (see Aikawa & Herbst 1999, for a discussion of ionization by
X-rays in proto-stellar disks). The physical conditions will therefore depend intimately on
the radial distance, and on the height above the midplane. For this reason, astrochemical
models relevant to the case of proto-stellar disks are coupled to dynamic models giving
solutions for their density and temperature structure.
In the parts of the disk where icy mantles do not undergo sublimation or destruction
through any mechanism, the growth of icy grain mantles leads to the formation of icy
particles increasing in size. Such proto-stellar ices play a crucial – even poorly understood
– role in astrochemistry. The fate of a given icy particle will depend strongly on its
location in the disk. Those ices will sublimate in the inner parts of the disk, where a
critical temperature is reached, and even dust grains may be completely destroyed in the
very inner part of the disk and close to the photosphere of the forming star. In addition,
the strong radiation field from the proto-star will also cause some parts of the disk –
including grains and ices themselves – to evaporate.
However, a fraction of the ices present in the disk will not sublimate. These ices
will remain in the disk during the planetary system formation process, and contribute to
planets, asteroids and comets. In the meantime, interesting chemical processes are likely
to take place in these ices where many molecular species are trapped (also called ’dirty
ices’):
- UV and X-ray photons are likely to lead to photodissociations in ices. As a result,

chemical bondings will break, generating free radicals likely to react with any partner
present in the vicinity.
- The heating of the ices (but below the evaporation or sublimation threshold) could
also trigger reactions in the solid icy matrices, allowing complex chemistry to take
place in these ices.
These ices seem therefore a priori to be chemically very active, and are important building
blocks for planets, satellites, and small bodies of the Solar system. The ongoing accretion
of the disk material onto the proto-star, and onto grains leading to the formation of
planetesimals, will subsequently form the final planetary system orbiting the young star.
The latter transition is illustrated by the arrow number 3 in Fig. 3.18.
(d) The planetary system

Once the proto-stellar disk has disappeared and the proto-star has reached the main-
sequence (starting its central hydrogen burning phase), the complete stellar system is
formed: a central star surrounded by a planetary system. Such a system is illustrated in
Fig. 3.24.
Any cell of material in the disk is likely (i) to have accreted onto the proto-star, (ii) to
have been expelled at the outskirt of the proto-stellar cloud because of the strong radiation
field from the proto-star, or (iii) to contribute to the material constituting the planetary
system, including also small bodies such as comets and asteroids.
The main reservoir of comets is the Oort Cloud, consisting in a huge sphere whose
radius is of the order of a few 104 AU. Comets are thus made of accreted material that
did not a priori undergo sublimation in the central parts of the system. Following the
dynamical interactions between small bodies located in the Oort Cloud, some comets may
adopt eccentric orbits leading them to the central part of the system. As a result, these
comets cross regions permeated by the radiation field from the Sun and by the solar wind
made of energetic particles.
Material of comets is a priori complex. It should be made of chemical species originating

from the proto-stellar cloud, along with many other chemical compounds likely synthetized
in the comets. Comets should be good places for many chemical processes:
- Photochemistry (and radiochemistry): photons from the Sun, along with high energy
particles from the Sun (and to some extent cosmic-rays) are likely to break chem-
ical bondings, producing radicals likely to react in solid icy matrices where many
molecules are tight together.
- Action of heat is also likely to activate many chemical processes in comets.

Macromolecular (refractory) compounds may populate cometary nuclei (polymers of
HCN, polymers of formaldehyde...). The photolysis, or thermolysis, or even hydrolysis
(as water is abundant in comets) processes on such compounds should affect significantly
the chemical content of comets, and produce wealth of organic compounds, including for
instance amino acids such a glycine, as already identified by the Stardust mission (Elsila
et al. 2009).
On the other hand, the same kind of chemistry could also operate to some extent on
rocky interplanetary material such as asteroids. Fragments of such small Solar system
bodies are collected onto Earth’s surface in the form of meteorites, displaying in some
82
Oort Cloud
Comet reservoir, to some extent
representative of the pre−stellar
conditions
− almost no radiation field
− cold environment
Dynamical
interactions
Comet
Some comets may adopt
an orbit with a periastron
passage close to the Sun
− stronger radiation field
− higher temperature
Cometary
bombardment
proto-stellar cloud that has been discussed in the previous steps.

Planets and satellites
Circumstellar debris − Rocky and icy material agregate
− Asteroids, ices: reservoir of material to form planetesimals
processed during the previous episodes − Planetesimals add together to form
of planetary system formation planets and satellites
− Planetary composition is a function
− Condensed matter chemistry is likely of the location in the planetary system
to slowly process further the molecular − Typically, the heaviest material is more
content of these residuals abundant in the inner part of the system
Figure 3.24: Schematic view of a planetary system resulting from the evolution of the
Molecules in the Universe
− Lighter elements are more abundant in

outer planets
cases very rich abundances in many organic molecular species (see for instance the case
of the Murchison meteorite). In this context, the issue of the survival of complex organic
compounds during the entrance in the Earth’s atmosphere is still a matter of debate.
Asteroids Comets
UV UV
Atmosphere Gas−phase chemistry

Bimolecular processes
T(h)ermolecular processes
Photochemistry
Lithosphere Hydrosphere
Water−phase chemistry
Solution chemistry
(solvatation, hydrolysis,
acid/base processes...)
Geochemistry
Figure 3.25: Schematic view of a planet including its main contributions: atmosphere,
lithosphere and hydrosphere. The double-headed arrows illustrate the possible connections
between these contributions.
Finally, planets and satellites constitute very valuable places to consider a further evo-
lution of the chemical content of a given stellar system, such as our Solar system. The case
of the Earth in its very young phase is especially important in the sense that it constituted
to craddle of the first living beings, about 3.8 billion years ago. The particular case of
a planet harbouring a rocky crust, liquid water and an atmosphere is illustrated in Fig. 3.25.
Planetary atmospheres are known to be very rich environments where a complex chem-
istry is at work:
- Density is much higher than in molecular clouds and proto-stellar disks.
- Temperature is much higher than in molecular clouds, but not necessarily high
enough to produce thermolysis of chemical compounds.
- Photolysis is also likely to play a role, at least in the upper layers of the atmopshere.
In addition, the hydrosphere such as that found on Earth (and already during its pre-
biotic age) is very important in the sense that it provides a liquid-phase, where solution
chemistry is likely to take place, in intimate interaction with the lithosphere and the
atmosphere. This offers the promising opportunity to allow reaction processes such as
hydrolysis, acid-base reactions, and other reaction pathways well-known in modern chem-
istry, to be efficiently at work.
The intense cometary (and asteroid) bombardment that occurred in the early stage
of the young Earth is believed to have brought significant amount of cometary organic
material onto Earth. The braking of some of the impactors during their penetration in the
atmosphere should have been sufficient to allow a fall with a velocity below the critical value
for complete destruction of incoming organic material (Chyba et al. 1990). As a result,
the prebiotic Earth may have been chemically enriched by cometary material, including
large amounts of water and molecular species of diversified complexity. As a result of this
idea, it seems very reasonable to consider that at least some part of the chemical material
that contributed to the prebiotic Earth was of interstellar origin.
Bibliography
Adams, N. G., Smith, D., & Millar, T. J. 1984, MNRAS, 211, 857
Agúndez, M., Cernicharo, J., Guélin, M., et al. 2010, A&A, 517, L2
Aikawa, Y. & Herbst, E. 1999, A&A, 351, 233
Allamandola, L. J. & Norman, C. A. 1978, A&A, 66, 129
Allamandola, L. J., Tielens, A. G. G. M., & Barker, J. R. 1985, ApJ, 290, L25
Allamandola, L. J., Tielens, G. G. M., & Barker, J. R. 1989, ApJS, 71, 733
Ball, J. A., Gottlieb, C. A., Lilley, A. E., & Radford, H. E. 1970, ApJ, 162, L203+
Belloche, A., Menten, K. M., Comito, C., et al. 2008, A&A, 482, 179
Berné, O. & Tielens, A. G. G. M. 2012, Proceedings of the National Academy of Science,

109, 401
Bohme, D. K. 1992, Chem. Rev., 92, 1487
Bossa, J. B., Theule, P., Duvernay, F., & Chiavassa, T. 2009, ApJ, 707, 1524
Bottinelli, S. 2006, PhD thesis, University of Hawaii at Manoa
Brown, R. D., Crofts, J. G., Godfrey, P. D., et al. 1975, ApJ, 197, L29
Buhr, H., Stützel, J., Mendes, M. B., et al. 2010, Phys. Rev. Lett., 105, 103202
Cami, J., Bernard-Salas, J., Peeters, E., & Malek, S. E. 2010, Science, 329, 1180
Cataldo, F., Brucato, J. R., & Keheyan, Y. 2005, Journal of Physics: Conf. Series, 6, 139
Cazaux, S. & Tielens, A. G. G. M. 2004, ApJ, 604, 222
Cerf, C. & Jorissen, A. 2000, Space Science Reviews, 92, 603
Cernicharo, J., Heras, A. M., Tielens, A. G. G. M., et al. 2001, ApJ, 546, L123
Cheung, A. C., Rank, D. M., Townes, C. H., Thornton, D. D., & Welch, W. J. 1968,
Physical Review Letters, 21, 1701
Cheung, A. C., Rank, D. M., Townes, C. H., Thornton, D. D., & Welch, W. J. 1969,
Nature, 221, 626
Churchwell, E. & Winnewisser, G. 1975, A&A, 45, 229
85
86 Bibliography
Chyba, C. F., Thomas, P. J., Brookshaw, L., & Sagan, C. 1990, Science, 249, 366
Combes, F., Gerin, M., Wootten, A., et al. 1987, A&A, 180, L13
Cronin, J. R. & Pizzarello, S. 1997, Science, 275, 951
Crowther, P. A. 2007, Ann.Rev.Astron.Astroph., 45, 177
Danger, G., Bossa, J., de Marcellus, P., et al. 2011, A&A, 525, A30+
De Becker, M. 2014, Ap&SS, in press
De Becker, M. & Raucq, F. 2013, A&A, 558, A28
de Marcellus, P., Meinert, C., Nuevo, M., et al. 2011, ApJ, 727, L27+
Dickens, J. E., Irvine, W. M., Ohishi, M., et al. 1997, in Bulletin of the American Astro-
nomical Society, Vol. 29, Bulletin of the American Astronomical Society, 1245
Dickman, R. L. 1978, ApJS, 37, 407
Douglas, A. E. & Herzberg, G. 1941, ApJ, 94, 381
Drury, L. O. . 2012, Astroparticle Physics, 39, 52
Duley, W. W. & Williams, D. A. 1981, MNRAS, 196, 269
Duvernay, F., Dufauret, V., Danger, G., et al. 2010, A&A, 523, A79+
Ellison, D. C., Drury, L. O., & Meyer, J.-P. 1997, ApJ, 487, 197
Elsila, J. E., Glavin, D. P., & Dworkin, J. P. 2009, Meteoritics and Planetary Science, 44,
1323
Engel, M. H. & Macko, S. A. 1997, Nature, 389, 265
Ervin, K. M. & Armentrout, P. B. 1987, J. Chem. Phys., 86, 2659
Fourikis, N., Sinclair, M. W., Robinson, B. J., Godfrey, P. D., & Brown, R. D. 1974a,
Australian Journal of Physics, 27, 425
Fourikis, N., Takagi, K., & Morimoto, M. 1974b, ApJ, 191, L139+
Geballe, T. R., Tielens, A. G. G. M., Allamandola, L. J., Moorhouse, A., & Brand,
P. W. J. L. 1989, ApJ, 341, 278
Gillett, F. C., Forrest, W. J., & Merrill, K. M. 1973, ApJ, 183, 87
Gilmore, W., Morris, M., Palmer, P., et al. 1976, ApJ, 204, 43
Gottlieb, C. A. 1973, in Molecules in the Galactic Environment, ed. M. A. Gordon & L. E.

Snyder, 181
Guélin, M., Muller, S., Cernicharo, J., et al. 2000, A&A, 363, L9
Helder, E. A., Vink, J., Bykov, A. M., et al. 2012, Space Sci. Rev., 173, 369
Hollis, J. M., Lovas, F. J., & Jewell, P. R. 2000, ApJ, 540, L107
Bibliography 87
Hollis, J. M., Lovas, F. J., Jewell, P. R., & Coudert, L. H. 2002, ApJ, 571, L59
Ioppolo, S., Cuppen, H. M., Romanzin, C., van Dishoeck, E. F., & Linnartz, H. 2008,
ApJ, 686, 1474
Jorissen, A. & Cerf, C. 2002, Origins of Life and Evolution of the Biosphere, 32, 129
Kaifu, N., Morimoto, M., Nagane, K., et al. 1974, ApJ, 191, L135+
Kroto, H. W., Heath, J. R., Obrien, S. C., Curl, R. F., & Smalley, R. E. 1985, Nature,
318, 162
Kuan, Y.-J., Charnley, S. B., Huang, H.-C., Tseng, W.-L., & Kisiel, Z. 2003a, ApJ, 593,
848
Kuan, Y.-J., Yan, C.-H., Charnley, S. B., et al. 2003b, MNRAS, 345, 650
Lacy, J. H., Carr, J. S., Evans, II, N. J., et al. 1991, ApJ, 376, 556
Larralde, R., Robertson, M. P., & Miller, S. 1995, Proc. Natl. Acad. Sci., 92, 8158
Leger, A. & Puget, J. L. 1984, A&A, 137, L5
Linke, R. A., Frerking, M. A., & Thaddeus, P. 1979, ApJ, 234, L139
Lovas, F. J., Hollis, J. M., Remijan, A. J., & Jewell, P. R. 2006, ApJ, 645, L137
Markwick, A. J., Ilgner, M., Millar, T. J., & Henning, T. 2002, A&A, 385, 632
Matthews, C. 1995a, in Astronomical Society of the Pacific Conference Series, Vol. 74,
Progress in the Search for Extraterrestrial Life., ed. G. S. Shostak, 95–+
Matthews, C. N. 1995b, P&SS, 43, 1365
McKellar, A. 1940, PASP, 52, 187
Mehringer, D. M., Snyder, L. E., Miao, Y., & Lovas, F. J. 1997, ApJ, 480, L71+
Millar, T. J. 1992, MNRAS, 259, 35P
Mimura, K. 1995, Geochim. Cosmochim. Acta, 59, 579
Mitchell, G. F., Huntress, Jr., W. T., & Prasad, S. S. 1979, ApJ, 233, 102
Nuevo, M., Meierhenrich, U. J., Muñoz Caro, G. M., et al. 2006, A&A, 457, 741
Peeters, Z., Botta, O., Charnley, S. B., et al. 2005, A&A, 433, 583
Peeters, Z., Botta, O., Charnley, S. B., Ruiterkamp, R., & Ehrenfreund, P. 2003, ApJ,
593, L129
Peeters, Z., Rodgers, S. D., Charnley, S. B., et al. 2006, A&A, 445, 197
Phillips, T. G. & Huggins, P. J. 1981, ApJ, 251, 533
Prantzos, N. 2012, A&A, 538, A80
Prasad, S. S. & Huntress, Jr., W. T. 1980a, ApJS, 43, 1

88 Bibliography
Prasad, S. S. & Huntress, Jr., W. T. 1980b, ApJ, 239, 151
Prasad, S. S. & Tarafdar, S. P. 1983, ApJ, 267, 603
Pringle, J., Allen, R. J., & Lubow, S. 2001, Monthly Notices of the Royal Astronomical
Society, 327, 663
Reimer, A., Pohl, M., & Reimer, O. 2006, ApJ, 644, 1118
Remijan, A. J., Hollis, J. M., Lovas, F. J., et al. 2008, ApJ, 675, L85
Rettig, T. W., Tegler, S. C., Pasto, D. J., & Mumma, M. J. 1992, ApJ, 398, 293
Ricca, A., Bauschlicher, C. W., & Bakes, E. L. O. 2001a, Icarus, 154, 516
Ricca, A., Bauschlicher, C. W., & Rosi, M. 2001b, Chem. Phys. Let., 347, 473
Rimola, A., Sodupe, M., & Ugliengo, P. 2010, Physical Chemistry Chemical Physics (In-
corporating Faraday Transactions), 12, 5285
Rubin, R. H., Swenson, Jr., G. W., Benson, R. C., Tigelaar, H. L., & Flygare, W. H. 1971,
ApJ, 169, L39+
Sellgren, K., Uchida, K. I., & Werner, M. W. 2007, ApJ, 659, 1338
Sellgren, K., Werner, M. W., Ingalls, J. G., et al. 2010, ApJ, 722, L54
Shibata, T., Yamamoto, J., Matsumoto, N., Yonekubo, S., & Osanai, S. a. K. 1998, J.
Am. Chem. soc., 120, 12157
Smith, I. W. M., Talbi, D., & Herbst, E. 2001, A&A, 369, 611
Snyder, L. E. & Buhl, D. 1971, ApJ, 163, L47+
Snyder, L. E., Buhl, D., Schwartz, P. R., et al. 1974, ApJ, 191, L79+
Snyder, L. E., Buhl, D., Zuckerman, B., & Palmer, P. 1969, Physical Review Letters, 22,
679
Snyder, L. E., Lovas, F. J., Hollis, J. M., et al. 2005, ApJ, 619, 914
Snyder, L. E., Lovas, F. J., Mehringer, D. M., et al. 2002, ApJ, 578, 245
Solomon, P. M., Jefferts, K. B., Penzias, A. A., & Wilson, R. W. 1971, ApJ, 168, L107+
Stoks, P. G. & Schwartz, A. W. 1981, Geochim. Cosmochim. Acta, 45, 563
Stoks, P. G. & Schwartz, A. W. 1982, Geochim. Cosmochim. Acta, 46, 309
Swings, P. & Rosenfeld, L. 1937, ApJ, 86, 483
Takano, Y., Takahashi, J., Kaneko, T., Marumo, K., & Kobayashi, K. 2007, Earth and
Planetary Science Letters, 254, 106
Tielens, A. G. G. M. 2005, The Physics and Chemistry of the Interstellar Medium (ISBN
0521826349: Cambridge University Press, 2005.)
Turner, B. E., Steimle, T. C., & Meerts, L. 1994, ApJ, 426, L97+
Bibliography 89
van Dishoeck, E. F., Herbst, E., & Neufeld, D. A. 2013, Chemical Reviews, 113, 9043
van Dishoeck, E. F., Kristensen, L. E., Benz, A. O., et al. 2011, PASP, 123, 138
Werner, M. W., Uchida, K. I., Sellgren, K., et al. 2004, ApJS, 154, 309
Wilson, R. W., Jefferts, K. B., & Penzias, A. A. 1970, ApJ, 161, L43+
Winnewisser, G. & Churchwell, E. 1975, ApJ, 200, L33
Woodall, J., Agúndez, M., Markwick-Kemper, A. J., & Millar, T. J. 2007, A&A, 466, 1197
Woon, D. E. 2011, ApJ, 728, 44
Zaleski, D. P., Seifert, N. A., Steber, A. L., et al. 2013, ApJ, 765, L10
Ziurys, L. M., Apponi, A. J., Guelin, M., & Cernicharo, J. 1995, ApJ, 445, L47
Zuckerman, B., Ball, J. A., & Gottlieb, C. A. 1971, ApJ, 163, L41+
Zuckerman, B., Turner, B. E., Johnson, D. R., et al. 1975, ApJ, 196, L99

Astrochemistry and Astrophysics PDF

Uploaded by

Copyright:

Available Formats

Astrochemistry and Astrophysics PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Astrochemistry and Astrophysics PDF

Uploaded by

Copyright:

Available Formats

Faculté des Sciences

Astrophysics and astrochemistry

– Academic year 2013-2014 –

Masters en Sciences Chimiques

3 Molecules in the Universe 51

- the formation and evolution of stars

- the formation and evolution of galaxies

- the dynamics and evolution of the interstellar medium

- the formation of planets and planetary systems

- extragalactic objects and large-scale structures

- the evolution of the Universe

- particle physics in astronomical environments

The challenges of astrochemistry

A brief census of molecules in astronomical environments

The objectives of this course

- introduce some basic concepts of astrophysics, mainly concerning Galactic objects

Table 1: Temporary census of interstellar molecules.

N2 O, NaCN, OCS, SO2 , c-SiC2 , SiCN, SiNC, FeCN, TiO2

NHD2 , ND3 , HDCO, D2 CO, CH2 DCCH, CH3 CCD,

1.1 General concepts

- the energy distribution of electrons is defined by the Maxwell-Boltzmann relation,

Thermodynamical equilibrium is reached when these temperatures (electron, ionization,

Figure 1.1: Examples of black body spectra at different temperatures.

an unitary solid angle. It can be expressed as follows:

1.1.2 Distance scales

- The light-year (ly). This is the distance crossed by electromagnetic radiation

1.1.3 Magnitude and extinction

m1 − m2 = 2.5 log(f2 /f1 ) (1.4)

Starting from Eq. 1.4, let us consider the following quantities:

Considering a standard distance (D) of 10 pc, we finally obtain:

The quantity (m - M) is often called the distance modulus of the object.

Figure 1.2: Examples of standard photometric bands frequently used in astronomical

1.2 Stellar objects

Figure 1.3: Schematic view of the Milky Way.

1.2.1 Hertzsprung-Russell diagram

1.2.2 Spectral classification

Table 1.1: Standard spectral classification of stars.

1.2.3 Stellar formation

1.2.4 Stellar nucleosynthesis and evolution

1.3 Interstellar medium

- neutral atoms: mainly H and He, with significant amounts of C, N, O elements. It is

1.4 Solar system

- A planet (from Greek meaning ”wandering star”) is a celestial body orbiting a

- A natural satellite or moon is a celestial body that orbits a planet or a dwarf

- Planetesimals are solid objects thought to exist in protoplanetary disks and in

- A meteoroid is a sand- to boulder-sized particle of debris in the Solar System.

Figure 1.9: Schematic repre-

Figure 1.10: Location of the

Application 8. Considering an asteroid with an incoming velocity of the order 5 km/s,

Application 9. Considering an asteroid with an incoming velocity of the order 7 km/s

2.1 Elementary gas-phase processes

This points to an especially strong constraint for astrochemistry, at least in inter-

In the context of astrochemical studies, it is also not relevant to establish complex

kUV (H2 ) = ko (H2 ) βSS exp (−τd )

2.1.3 Neutral-neutral reactions

This kind of reaction is typically of the type

Reaction rate coefficient are considered to be of the form

k = α (T/300)β exp [−γ/T]

2.1.4 Ion-molecule reactions