Non-thermal desorption is the release in the gas phase of adsorbed species when the system is not at the thermodynamic equilibrium. Dierent types of non-thermal desorption can be considered. In the following we present a summary list.
1. Photodesorption: the absorption of a UV photon by a molecule condensed on a surface can result in its desorption (direct photodesorption) or in the kick-out of a nearby molecule (indirect photodesorption) (i.e. DeSimone et al. 2013; Bertin et al. 2013; Fillion et al. 2014; Yuan&Yates 2013).
2. Sputtering: an atom is ejected from a solid target material (i.e. ice) due to bombardment of the target by energetic particles, i.e. ions or electrons (i.e. Behrisch 1981; Johnson et al. 2013; Cassidy et al. 2013). The condition to observe sputtering is that the kinetic energy of the energetic particles is much higher than conventional thermal energies ( 1 eV). Sputtering can lead to signicant erosion of bombarded materials or ice.
3. Chemical desorption: the energy excess of an exothermic reaction is not dissipated on the surface and provokes the ejection of the newly formed species in the gas phase (Astarita&Savage 1980; Dulieu et al. 2013).
Surface migration: Thermal hopping and tunneling
Another important process for the description of surface physics is the migration of adsorbed atoms on the surface. Atoms and molecules physically adsorbed on a surface nd themself in a potential minimum, and they cannot travel freely along the surface. Nevertheless, depending on the surface temperature, adsorbed species could be mobile in their physisorbed (or even chemisorbed) precursor states. The motion of an adatom on a periodic and regular Surface Physics 9 surface can be thought as a random site-to-site hopping process, namely a random-walk motion. The diusing atom will have a mean square displacement in the time t given by hr2i = an t (1.8).
where is the frequency of hops, a is the jump distance, and n is the number of dimensions. Clearly in the case of surface diusion, n=2. It is possible to dene the diusion coecient D as the ratio of the mean square displacement over time weighted on the number z of neighboring sites where the adatom can hop: D = hr2i z t = an.
In this mechanism, proposed in 1938 by D. D. Eley and E. K. Rideal, only one of the molecules is adsorbed and the other one reacts with it directly coming from the gas phase, without being adsorbed on the surface of the catalyst (see Figure 1.8):
A(gas) + Surface ! A(S) (1.19).
A(S) + B(gas) ! Products (1.20).
ER reactions are somehow similar to gas phase reactions. The impinging particles either collide and react with one adsorbate or have enough energy to hop on the surface before thermalizing and accommodating in an empty adsorption site. In other words, ER mechanism is a nonthermal surface mechanism because it leads to a reaction between a thermally adsorbed surface species and a reactant which has not yet thermally accommodated to the surface. This point is of crucial importance to evaluate the activation barrier of a reaction, as we will discuss in Sec.
Mass spectroscopy: the dierent uses of the QMS
In subSec. 2.1.5, we have briey described the QMS operation. We have claimed that gas species are detected after their ionization. This occurs in the so called ionization zone through electron impacts. Here we pinpoint that electron impacts could produce dissociation of species in addition to ionization. Actually it is necessary to pay attention to the fragment distribution of ionic species which results from dissociation and ionization of multi-atomic molecules of any given species in the ionizer, the so called cracking pattern. The probabilities of ionization and dissociation depend on molecular geometry (and intra-molecular bonds energy), angle of impact between molecules and ionizing electrons of the QMS, and the energy of ionizing electrons. For these reason, signals (and peaks) at dierent masses can derive from the same molecule, reecting its partial fragmentation in the QMS head. Figure 2.8 shows the TPD curves of CH3OH and six major signals are presented. Table 2.1 reports the cracking pattern of some molecules: NO2, O3, H2CO, NH2OH, HCOOH, and CH3OH. The signals are normalized to the highest mass signal. Nitrogen dioxide is the simplest case listed in Table 2.1.
Knowing electronic state of atoms and molecules
The QMS is also used to know the quantum state of the species sent on the substrate. Usually the energy of ionizing electrons is set to 30 eV to have a good ionization cross section and a reduced cracking pattern (which raises in complexity with electrons energy). Nevertheless the electrons energy can be tuned from 0 to 100 eV with a resolution of 0.2 eV . Each molecule has an ionization threshold which depends on the initial quantum state that has to be overcome to detect the species. By tuning the energy of the ionizing electrons of the QMS, we can selectively detect ground state or excited atoms and molecules, as described in Congiu et al. (2009). Here we stress that this method can only distinguish between electronic excited states, due to the 0.2 eV resolution; actually the energy dierence between two vibrational (rotational) states is of 101-102 (103-104) eV . Top and bottom panels of Figure 2.12 show, respectively, signals of O/O2 (Mass16/Mass32) and N/N2 (Mass14/Mass28) as function of electron energy when the discharge is ON (red curves) and OFF (green curves). In the O/O2 case (top panels), we see that the Mass 16 signal drops o at around 13.5 eV (green curve is not present because no O atoms are present in the un-dissociated O2 beam), while Mass 32 signals (red and green curves) fall to zero at 12.1 eV . We can explain these results by looking at the energy of the excited states of O and O2. Atomic orbital theory predicts that O atoms in the ground state (3P) are ionized by 13.6 eV electrons (Moore 1993) while in the rst excited state 1D the minimum energy necessary for ionization is 11.7 eV (-1.9 eV ). Hence, electrons less energetic than 13.6 eV can ionize excited (3P ro-vibrational excited states or 1D electronic excited state) O atoms only. The same argument can be applied to molecularoxygen. O2 molecule has two low-lying excited singlet states, O2(a1g) and O2(b1+g ), while the ground state is the triplet O2(X3g ) state3. The energy dierence between the lowest energy of O2 in the singlet state, and the lowest energy in the triplet state is about Te (a1g – X3g ) = 0.98 eV (Schweitzer&Schmidt 2003). The required energy to ionize an O2 molecule in the ground state is 12.07 eV (Tanaka&Tanaka 1973). This means that electrons less energetic than this value cannot ionize O2 in the ground state, while O2(a1g ) molecules can be ionized by electrons with energy >11.09 eV . Finally electrons with energy below 11.09 eV can ionize O2(b1+g ). By considering these argumentations we can go back to the top panels of Figure 2.12 and claim that the beam did not contain O or O2 in an excited state. The O beam was thus composed of at least 99% ground-state O and O2. In the case of nitrogen, theory predicts that N atoms in the ground state (4S) are ionized by 14.53 eV electrons (Moore 1993) while in the rst excited state 2D the minimum energy necessary for ionization is 12.15 eV (-2.38 eV ). Hence, electrons in the range 14.53-12.15 eV can ionize excited (4S ro-vibrational excited states or 2D electronic excited state) N atoms only. The ground state of N2 molecule is the singlet N2(X1+g ) state, while the rst electronic excited state of N2 molecule is the triplet states, N2(A3+u ) (Lofthus&Krupenie 1977). The energy dierence between the two states is about Te (A3+u – X1+g ) 6 eV.
TPD: Temperature Programmed Desorption
In subSec. 1.1.3 we have described the desorption process. We have claimed that two type of desorption can occur: thermal and non-thermal. In thermal desorption the adsorbed species acquires thermal energy to desorb. One of the most important methods to study thermal desorption is the Temperature Programmed Desorption (TPD). It consists in observing desorbed molecules from a surface when the surface temperature is increased. During the TPD the sample is heated linearly vs time with a rate from its initial value T0 T = T0 + t.
Table of contents :
1 Theoretical background
1.1 Surface Physics
1.1.1 Sub-monolayer and multilayer: surface and bulk
126.96.36.199 Physisorption and chemisorption
188.8.131.52 Thermal desorption
184.108.40.206 Non-thermal desorption
1.1.4 Surface migration: Thermal hopping and tunneling
220.127.116.11 An evaluation of the crossover temperature Ttnth.
1.2 Gas Phase Chemistry
1.2.1 Chemistry in the gas phase
1.2.2 Exothermic and endothermic reactions
1.3 Surface Chemistry
1.3.1 Eley-Rideal mechanism
1.3.3 Hot atom mechanism
2 Experimental apparatus and methods
2.1 Experimental apparatus
2.1.1 The main chamber
2.1.2 The beamlines
2.1.3 The microwave cavities
2.1.4 The sample holder and the sample
2.1.5 The Quadrupole Mass Spectrometer
2.1.6 The Infrared Spectrometer
2.2 Experimental methods
2.2.1 Water ice deposition on the sample holder
2.2.2 Mass spectroscopy: the dierent uses of the QMS
18.104.22.168 Cracking pattern
22.214.171.124 Residual gas analyzer: knowing beam composition
126.96.36.199 Knowing electronic state of atoms and molecules
188.8.131.52 TPD: Temperature Programmed Desorption
184.108.40.206 DED: During-Exposure Desorption
2.2.3 Calibration of the H2 and D2 beams
2.2.4 Calibration of the molecular beams
220.127.116.11 Determination of O3 monolayer and detection eciency .
2.2.5 Can reactions occur on gas phase?
3.1.1 Initial condition
3.1.2 Flux: x
3.1.3 Reaction probability: r
3.1.4 Diusion coecient: kdiff
3.1.5 Desorption coecient: Nxdes
3.2 From model parameters to physical-chemical quantities
3.2.1 Reactivity: reaction probability and activation barrier
18.104.22.168 Teff evaluation
3.2.2 Mobility: kdiff and surface migration
3.2.3 Desorption: desorption probability and binding energy
4 Surface Physics
4.1 Oxygen reactivity and diusion
4.1.2 Oxygen reactivity
4.1.3 Oxygen diusion
22.214.171.124 The role of substrate
126.96.36.199 Armophicity and diusion
4.2 Evaluation of desorption energies
4.2.1 Desorption energy of non-reactive species
4.2.2 Desorption energy of reactive species
4.3 Chemical desorption
4.3.1 The chemical desorption of the O-H system
188.8.131.52 Chemical desorption in O2+H experiments
184.108.40.206 Chemical desorption in O3+H experiments
220.127.116.11 Chemical desorption in O+H experiments
4.3.2 The inuence of the substrate: CD of O2
18.104.22.168 Experimental results
22.214.171.124 Model and discussion
5 Surface Chemistry
5.1 Water formation via O2+H(D)
5.1.2 Results and discussion
5.2 Nitrogen oxides chemistry
126.96.36.199 Result and discussion
188.8.131.52.a Initial conditions and model
184.108.40.206 Results and discussion
220.127.116.11 Results and discussion
18.104.22.168.a Dependence on surface temperature
5.2.4 NO2 reactivity
5.3 O/C/H chemistry
5.3.1 Carbon dioxide formation on cold surfaces
22.214.171.124.b Results and discussion
126.96.36.199.c Model: evaluation of the CO+O barrier
188.8.131.52.b Results and discussion
184.108.40.206.c Model: evaluation of the H2CO+O barrier
5.3.2 The cycle of the CO-H chemistry
220.127.116.11 Chemical desorption pumping process
5.3.3 Preliminary results of CH3OH and HCOOH irradiation with H/O atoms
6.1 Astrophysical conclusion
6.1.1 Astrophysical context
6.1.2 O diusion
6.1.3 Chemical Desorption
6.1.4 H/C/N/O chemical network
18.104.22.168 Interstellar ices
6.2.1 Ice growth: MonteCarlo Model
6.2.2 Further perspectives
Appendix A: atomic and molecular term symbols
Appendix B: list of publications
List of Acronyms