Grecea, Mihail Laurentiu
Citation
Grecea, M. L. (2006, February 23). Light-induced molecular processes on ice. Retrieved from https://hdl.handle.net/1887/4322
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoralthesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/4322
Chapter 2
Experimental setup and
techniques
2.1 Ultra-high vacuum setup
Studying molecular processes on ice requires very well-defined conditions that can be achieved under ultra-high vacuum (UHV). Useful information about these processes can be obtained by means of surface-sensitive techniques specifically running in UHV. The experiments described in this thesis are carried out in the UHV machine POTVIS (“pot voor infrarood spectroscopie”), schematically drawn in Figure 2.1. A brief description is given here, as the setup is described in detail by Jenniskens et al. [1] and Riedmüller et al. [2]. Two quadrupole mass spectrometers (QMS), a Fourier transform infrared spectrometer, a low-energy electron diffraction (LEED) apparatus, and a sputter gun equip the setup.
The main vacuum chamber consists of a stainless steel cylinder with a length of roughly 1 meter and a diameter of 0.25 meter, horizontally mounted on a frame. The base pressure of 2 × 10-11 mbar is achieved by running a turbo-drag pump (450 ls-1 for N
directly attached to the main chamber (Balzers 420, QMS1 in Figure 2.1). Another QMS, a differentially pumped Balzers QS 422 (QMS2 in Figure 2.1), is used for temperature-programmed desorption (TPD) and time-of-flight (TOF) experiments (see below). The quadrupole mass analyzer is situated in a stainless steel housing, which is differentially pumped by a turbo-drag pump (56 ls-1) connected in series with a second pump of the same type which is backed by a diaphragm pump.
The sample is mounted on a manipulator allowing for both translational and rotational sample movement. Translation along the axis of the cylindrical chamber is motorized, while translation in the two perpendicular directions to the axis of the cylinder can be performed manually over a range of 2.5 cm. A rotary feedthrough, pumped with a turbo-drag pump in combination with a membrane
molecular beam sample manipulator QMS 2 QMS 1 LEED sputter gun dosing valve valve infrared cell pressure gauge dosing valve IR 800 nm or 266 nm liquid nitrogen main turbo pump Ti-sublimation pump turbo pump
Figure 2.1. Schematic top view of the UHV setup POTVIS. The dotted circles refer
pump, enables a motorized rotation of 360 degrees. Details about the sample manipulator can be found in ref. [2].
The sample holder consists of a copper block suspended by a stainless steel fork, which is bolted to the rod of the manipulator. Liquid nitrogen can be flowed through the manipulator and the copper block. The single-crystal Pt(533) sample, a cylinder with a diameter of 1 cm and thickness of 3 mm, is fixed to a Ta plate by Ta clamps. The Ta plate and sample holder are electrically isolated by sapphire spacer rings and are screwed together with four Mo screws. Sapphire is known as a thermal diode; it conducts heat at low temperatures very well whereas at high temperatures the thermal conductivity is low. This allows rapid cooling of the sample as well as heating to high temperatures. The sample cooling is assured by heat exchange with the copper block through the sapphire spacers. Heating is provided from the back of the crystal by a filament. The filament is placed in a Ta cup situated in the copper block. Both radiative heating and electron bombardment can be performed. Using radiative heating, temperatures up to ~500 K can be reached whereas higher temperatures (up to 1200 K) are achieved by electron bombardment, which is produced by applying a high voltage (~400 V) to the Ta plate containing the sample. Sample temperatures are measured with a chromel-alumel thermocouple spot-welded to one side of the crystal. Details concerning the sample holder, as well as the cooling and heating of the sample are given in ref. [2].
Valve Buffer QMS Sample manipulator Main vessel Chopper stage Nozzle manipulator Expansionchamber
crystal surface is performed via background dosing or by means of the molecular beam line. The adsorbates exposure is controlled via the dosing time and the (expansion) pressure of the gases.
2.2 Molecular beam line
Figure 2.2 shows a 3D picture of the molecular beam line used to dose water, which is described in detail in ref. [2]. Briefly, it consists of three differentially pumped stages; two stages are implemented into the main vacuum system due to the requirement of a short distance between sample and nozzle.
Figure 2.2. A 3D-picture of the molecular beam line and the differentially pumped
The nozzle (diameter = 200 µm) is situated in the expansion chamber, which is pumped with a 520 ls-1 turbo-drag pump backed by a rotary vane pump. The accurate positioning of the nozzle in front of the skimmer is possible due to high precision XY- and Z-translators. The skimmer (diameter = 0.5 mm) is carried by a XY-stage mounted on a stainless steel plate which separates the expansion and chopper chambers. The chopper chamber is pumped with a 190 ls-1 turbo-drag pump in combination with another rotary vane pump. The beam path is continued, through a plate with an orifice of 3.5 mm, into the buffer chamber, pumped with a 180 ls-1 turbo-drag pump connected to the fore-vacuum pump of the main chamber. A mini-valve and an aperture (diameter = 8 mm) connecting to the main chamber determine the axis of the beam line. A beam shutter that can be actuated from outside the vacuum is located on the UHV side of the aperture. The beam resulting from the mentioned combination of skimmer and apertures has roughly the same size as the crystal sample.
2.3 Laser facilities
2.4 Desorption techniques
Essential information concerning adsorption processes and surface chemical reactions can be obtained from desorption experiments. Learning about the nature of the adsorbate-substrate bonding is possible from the desorption features. These represent authentic “fingerprints” of the adsorption states, as any species adsorbed on a surface is bound to the surface with some specific amount of energy. In certain cases, the bond between adsorbate and substrate could not break and only desorption of particular fragments of the adsorbate occurs (see Chapter 6 and 7). In this thesis, desorption is induced either thermally or by means of laser pulses.
2.4.1 Temperature-programmed desorption
2.4.1.1 Background
state results in a peak in the pressure-temperature curve rather than a plateau. In this thesis TPD experiments are used.
Practically, TPD consists in applying a constant temperature ramp (typically in the range of 0.5-6 Ks-1) to the sample and detecting the desorbing molecules in the gas phase as a function of surface temperature, usually by means of a mass spectrometer. The desorption temperature is related to the bond strength; a higher desorption temperature generally indicates stronger binding of the adsorbate to the substrate. Discrete binding sites with different desorption energies are reflected in the multiple peak structure of a TPD spectrum. As described in Chapter 3, the higher desorption temperature of the species adsorbed on (coordinatively unsaturated) step sites is due to the stronger bond of the adsorbate molecules to the step sites of the surface. In the case of a multilayer system, the bond strength of the molecular layer adjacent to the substrate is generally higher than that experienced in between further layers. For this reason, a multilayer peak usually appears in a TPD spectrum at a distinctly lower temperature than the monolayer peak. In addition to the adsorbants as such, TPD experiments can provide information about intermediate species and reaction products, in connection with a particular surface reactivity [5].
2.4.1.2 Setup
The temperature of the crystal is regulated with a temperature controller (Eurotherm 905S) by adjusting the current of the filament behind the crystal. Linear heating rates between 0.5 Ks-1 and 6 Ks-1 are typically employed. Simultaneous measuring of 16 masses in a mass range up to 511 amu can be performed.
2.4.2 Laser-induced desorption
2.4.2.1 Background
In a laser-induced desorption experiment, laser pulses are used to drive the desorption process. Initially, the laser transfers energy to the electron bath of the substrate. The resulting hot surface electrons relax by heat diffusion into the metal bulk and by coupling to the phonons (lattice vibrations). The time evolution of the electron and phonon temperature is described in detail by Backus et al. [3] based on the two-temperature model [6]. Both hot electrons and hot phonons can provide energy to the adsorbate. If sufficient energy is transferred, a reaction of the adsorbate can be induced: desorption, diffusion or dissociation. The process is schematically shown in Figure 2.3 (left panel).
2000 1500 1000 500 Tsur face (K) 6 5 4 3 2 1 0 Time (ps) Telectron Tphonon metal substrate electrons phonons Tel Tph adsorbate Tads
reaction, e.g. desorption
ηel ηph g laser excitation
A description of the energy flow from the hot electron and/or phonon baths to the adsorbate in the framework of the friction model [7,8] can be found elsewhere [3]. Briefly, in an electron-mediated process, a hot electron is transiently transferred to the adsorbate. This is possible due to a finite number of electrons above the Fermi level that may be resonant with the 2π* orbital of the adsorbate, upon laser irradiation of the sample. During the excited state lifetime, the adsorbate is accelerated on its excited potential energy surface, thus gaining kinetic energy. In a phonon-mediated process, the barrier is overcome by successive transitions occurring between vibrational levels due to the hot phonons, in the electronic ground state.
In a laser-induced desorption experiment, the desorbed species are detected with a mass spectrometer as a function of their flight time. Information about the translational energy of these species is contained in the time-of-flight (TOF) distribution (TOF spectrum). A TOF spectrum as such does not represent the
Figure 2.3. Left panel: Scheme of the energy transfer pathways following the excitation
with a femtosecond laser pulse (Tel: electron temperature, Tph: phonon temperature, g:
electron-phonon coupling constant). The parameters ηel (electron frictional coupling
constant) and ηph (phonon frictional coupling constant) characterize, in the friction
distribution of flight times, because the less likely detection of the faster molecules. Hence, the translational energy distribution of the desorbing species must be extracted from fitting the time-of-flight (TOF) spectrum using the formula:
( )
dt
t
d
kT
m
t
dt
t
n
density⎥
⎥
⎦
⎤
⎢
⎢
⎣
⎡
⎟
⎠
⎞
⎜
⎝
⎛
∝
2 4exp
2
1
(2.1)where d is the distance between the sample and the QMS detector, and T is the translational temperature.
A particular case of laser-driven desorption is represented by the
photodesorption experiments. Laser pulses with sufficient photon energy
(usually 3-10 eV) are used to excite electrons from the adsorbate bonds into antibonding orbitals. This can disrupt intrinsic and/or adsorption bonds of the adsorbate and lead to desorption of fragments and/or parent molecules. The process occurs through direct absorption of laser photons as the substrate electrons are quenched by a purposely deposited spacer layer.
2.4.2.2 Setup
Figure 2.4. Schematic representation of the laser-induced desorption setup. For
photo-desorption experiments the chopper is removed and a mirror at position B switches the beam path to the tripler. The picture is adapted from ref. [3].
Desorbed molecules are detected as a function of their flight time by the differentially pumped mass spectrometer (QMS2 in Figure 2.1), amplified with a fast amplifier and counted with a multichannel scaler after a 70 mm flight path. Time-of-flight (TOF) spectra are recorded by measuring the time difference between the laser trigger on diode 2 (outside the vacuum chamber, see Figure 2.4) and the detection of the molecules in the QMS. The time delay in the QMS is subtracted from the measured flight time [2]. The yield of the desorbed species is obtained from the integrated TOF spectra. A depletion curve is determined as follows: the desorption yield of each pulse is measured on one spot on the sample from 10 sequential laser shots. The first shot yield at the initial coverage is derived by fitting an exponential decay through the yield as a function of the laser shot number, typically averaged four times.
The absorbed laser fluence can be determined from the measured pulse energy and the beam profile recorded with a CCD camera at a position equivalent to that of the sample. The fluence dependence is determined with the yield-weighting procedure, described in detail in ref. [3]. For this measurement, the fluence is varied by placing thin microscope cover glass plates in the beam path, each reflecting 8% of the laser light. A reference window, identical to the viewport of the UHV chamber, is located in front of the camera in order to account for self-focusing effects of the viewport of the vacuum chamber.
Photodesorption experiments are performed in the setup described above. The 266 nm beam is obtained by tripling the 800 nm output of the laser.
2.5 Reflection absorption infrared spectroscopy
2.5.1 Background
strongly with vibrational modes of adsorbed species which have a transition dipole perpendicular to the surface. That means only molecular vibrations resulting in a dynamic dipole moment perpendicular to the surface will yield IR signal, whereas molecules lying down on the surface are not visible with RAIRS. Another way to rationalize this surface selection rule is to consider the response of the valence electrons of the substrate to the molecular vibrations of an adsorbate, as exemplified for CO in Figure 2.5. The enhanced net dipole resulted in upright configuration by summing up the molecular (µM) and image (µI) dipoles gives rise to significant IR absorption. In contrast, for flat lying molecules on the surface, upon vibration, the image and molecular dipoles cancel each other and no IR absorption is observed.
For multilayers adsorbed on a metal substrate, RAIRS probes simultaneously the bulk and the surface vibrational modes of the adsorbate. This can be observed, for example, in the RAIR spectrum of a 40 monolayers (ML) thick
δ
–δ
+μ
Iμ
C
O
δ
+δ
– MC
O
δ
+δ
–μ
Mδ
+δ
–μ
Iwater (H2O) film grown at 100 K on Pt(533) (see Figure 3.9 (ASW curve) in Chapter 3). The broad peak centered at 3401 cm-1 corresponds to the bulk H2O, whereas the free OH species on the surface are characterized by the peak at 3696 cm-1.
2.5.2 Setup
region. A background spectrum is measured, before dosing an adsorbate, from IR reflected from the bare surface. The infrared absorbance A used in Chapters 6 and 7 is defined as A = -log(R/R0), with R and R0 referring to the reflected
Bibliography
[1] H.G. Jenniskens, A. Bot, P.W.F. Dorlandt, W. van Essenberg, E. de Haas, A.W. Kleyn, An ultrahigh vacuum (UHV) apparatus to study the interaction between
adsorbates and photons, Meas. Sci. Technol. 8 (1997) 1313-22.
[2] B. Riedmüller, F. Giskes, D. Glastra van Loon, P. Lassing, A.W. Kleyn, A compact
molecular beam line, Meas. Sci. Technol. 13 (2002) 141-49.
[3] E.H.G. Backus, Driving and probing surfaces with light, Ph.D. Thesis, 2005.
[4] D.P. Woodruff, T.A. Delchar, Modern Techniques of Surface Science - Second
Edition, Cambridge University Press, Cambridge, 1994.
[5] J.W. Niemantsverdriet, Spectroscopy in Catalysis, Wiley, Wienheim, 1993.
[6] S.I. Anisimov, B.L. Kapeliovich, T.L. Perel'man, Electron-emission from surface of
metals induced by ultrashort laser pulses, Sov. Phys. JETP 39 (1974) 375-80.
[7] T.A. Germer, J.C. Stephenson, E.J. Heilweil, R.R. Cavanagh, Picosecond
time-resolved adsorbate response to substrate heating - spectroscopy and dynamics of Co/Cu(100), J. Chem. Phys. 101 (1994) 1704-16.
[8] F. Budde, T.F. Heinz, A. Kalamarides, M.M.T. Loy, J.A. Misewich, Vibrational
