Hydrogen on Silicon

Adsorption Dynamics of Molecular Hydrogen on Silicon

film30.gifThe reaction dynamics of molecular hydrogen with silicon surfaces is strongly influenced by surface structure and lattice distortions. Sufaces diffusion and recombinative desorption experiments showed that the covalent nature of hydrogen bonding on silicon surfaces leads to high diffusion barriers and to desorption kinetics that strongly depend on surface reconstruction. Dissociative adsorption exhibits a pronounced increase in reaction rate with surface temperature and demonstrates the decisive role of the lattice degrees of freedom in the reaction dynamics on semiconductor surfaces.

russ1bThe lattice distortions are the basis of the model of phonon-assisted sticking first proposed by Brenig et al. The essence is shown in the 2-dim. model-potential in the figure to the left. The hydrogen molecule approaching the Si surface experiences a high barrier resulting in a low sticking probability. With increasing surface temperature, more of the vibrational states are occupied and the lattice is more often in the favorable configuration where the hydrogen molecule experiences a lower potential barrier and is consequently leading to a higher sticking coefficient.
In the desorption process the molecules do not gain kinetic energy because most of the energy released is stored in lattice vibration.

Experimental Procedure

hads2bThe high sensitivity of SHG to hydrogen adsorption is exploited to in situ measure the H coverage on silicon surfaces. The red dots in the left hand side figure show the decrease of the SHG signal with the adsorption of molecular hydrogen. Measuring the hydrogen pressure (indicated as a blue line), the flux of molecules on the sample and therefore the sticking coefficient can be calculated.

Second Harmonic Generation (SHG) from Silicon Surfaces

shg_exp4                shg_exp6
shg_exp7

shg_exp2
Second-harmonic generation (SHG) provides an optical probe of surfaces and interfaces. Within the electric dipole approximation SHG only occurs if the medium lacks inversion symmetry. Therefore the nonlinear susceptibility vanishes in the bulk of centrosymmetric media but has a finite value at the surface where the symmetry is broken.
Microscopically SHG can be described by the excitation of an electron from the ground state |g> via an intermediate |n‘> to the excited state |n> by two photons with energy ħω and the subsequent emission of a second-harmonic photon with energy 2ħω. Whenever ħω or 2ħω coincide with real transitions from |g> to |n‘> or |n> the process exhibits resonant enhancement.
In the case of silicon surfaces such levels are provided by surface states in the band gap originating from Si dangling-bonds or reconstruction-induced bonds. The band structure of Si(111)7×7 is shown schematically in the figure to the left with occupied (Si) and unoccupied surface states (Ui).

Photon Energy Dependence

In spectroscopic SH experiments of the clean Si(111)7×7 surface the spectra reveal two resonant structures.
The peak at 3.4 eV is a resonance that almost coincides with the E1 transition between the valence and conduction band of bulk silicon. It arises because the bulk electronic structure is distorted at the surface. This resonance exhibits relatively little sensitivity on adsorbed hydrogen because the 7×7 surface structure remains intact for moderate exposures.
The broad feature emerging for photon energies below 1.3 eV beyond the accessible wavelength range completely disappears if hydrogen is adsorbed. Its origin is the resonant enhancement of SHG by the Si dangling bond states which are quenched upon hydrogen adsorption. Around 1.2 eV hydrogen adsorption reduces the nonlinear susceptibility almost by a factor of 10 which leads to a dramatic decrease of the measured signal by a factor of 100.

Coverage Dependence

si111speThe high sensitivity of the nonlinear susceptibility on hydrogen coverage has been the basis for many applications of SHG. The SH signal is dominated by the contribution of the dangling bonds resulting in a nearly linear dependence of the nonlinear susceptibility on the hydrogen coverage for low coverages. The minimum for higher coverages is a result of a phase shift between the two contributions to the nonlinear susceptibility. Besides the resonant part a weak nonresonant background e.g. from adatom backbonds exists.

Experimental setup for spectroscopy
Experimental setup for adsorption measurements

Further Information

U. Höfer
Nonlinear optical investigations of the dynamics of hydrogen interaction with silicon surfaces

Appl. Phys. A 63, 533-47 (1996)
G. A. Schmitt

Untersuchung der nichtlinearen optischen Eigenschaften von Siliziumoberflächen im nahen Infrarot:
Frequenzabhängigkeit und mikroskopische Mechanismen

(Doctoral thesis, TU München, 1996).

SHG from Semiconductor Surfaces – Worldwide Links

Oleg A. Aktsipetrov Moscow State University, Russia
Mike Downer University of Texas, Austin, USA
Tony F. Heinz Columbia University, New York, USA
Ulrich Höfer Philipps-Universität Marburg, Germany
Dietrich von der Linde Universität Essen, Germany
John F. McGilp Trinity College Dublin, Ireland
Frank Rebentrost MPQ Garching, Germany
Georg Reider TU Wien, Austria
Y. Ron Shen University of California, Berkeley, USA
John E. Sipe University of Toronto, Canada
Takanori Suzuki Riken, Japan
Harry K. Tom University of California, Riverside, USA
Henry M. van Driel University of Toronto, Canada
Arjun G. Yodh University of Pennsylvania, Philadelphia, USA
Helmut Zacharias Universität Münster, Germany

Second-Harmonic Spectroscopy – Experimental Setup

spectopt

For the spectroscopic measurements the Si(111) and Si(100) samples were mounted on a liquid-nitrogen cooled cryostat in a UHV chamber equipped with conventional diagnostics (LEED, Auger, QMS) with a base pressure of p < 5×10-11 mbar.For SHG we used a mode-locked Ti:sapphire laser tunable between 700 and 1080 nm, generating 70-120 fs pulses at 82 MHz with an average power of 50-250 mW incident at the sample. The laser beam was The laser beam was focused on the sample to a 80 µm diameter spot at an angle of incidence of 45°. The fundamental beam and the SH signal were separated by a Pellin-Broca prism and filters. A photomultiplier tube and a gated photon counter detected the SH signal. The SHG spectra were normalized against the SH signal of a plane parallel quartz plate.

Adsorption Experimental Setup

sfbbild3

The adsorption experiments are performed in an ultrahigh vacuum (UHV) chamber operating at a working pressure below 5×10-11 mbar. Surface conditions are checked by low-energy electron diffraction (LEED) and Auger electron spectroscopy (AES). Hydrogen is supplied by a high flux molecular beam apparatus with a nozzle – sample distance of about 30 cm. Sticking coefficients down to 1×10-8 can be measured. For SHG, the pump radiation with a wavelength of 1064 nm and a pulse duration of 10 ns is provided by a Q-switched Nd:YAG laser operating at repetition rates between 10 and 30 Hz. The fluence on the sample varied between 30 and 100 mJ/cm2, depending on sample temperature.

Experimental Results

jcp3cUsing molecular beam techniques and second-harmonic generation as the probing method, we investigated the dependence of the sticking coefficient of molecular hydrogen on kinetic energy and the surface temperature Ts. Results are shown in the figure on the left hand side. The strong dependence on the kinetic energy can be described by s-shaped adsorption functions with a common mean adsorption barrier of about 0.8 eV. With higher surface temperature a wider range of barriers is accessible, resulting in broader adsorption functions and an Arrhenius law for the initial sticking coefficient s0=A exp(-Ea/kTs) at fixed kinetic energy. The resulting activation energy of Ea=0.7 eV for the Si(001) surface gives convincing evidence for the strong influence of the Si lattice on the reaction dynamics.

Perspectives

The current work in our group deals with static distortions of the lattice (steps, pre-adsorbed hydrogen) and their influence on the adsorption dynamics. By this, detailed informations on the adsorption mechanism are obtained.

Literature

P. Bratu, K. L. Kompa, and U. Höfer
Optical second-harmonic investigations of H2 and D2 adsorption on Si(100)2×1:
the surface temperature dependence of the sticking coefficient

Chem. Phys. Lett. 251, 1-7 (1996)

U. Höfer
Nonlinear optical investigations of the dynamics of hydrogen interaction with silicon surfaces
Appl. Phys. A 63, 533-47 (1996)

M. Dürr, M. B. Raschke, and U. Höfer
Effect of beam energy and surface temperature on the dissociative adsorption of H2 on Si(001)
J. Chem. Phys. 111, 10411-4 (1999)

Hydrogen on Silicon – Worldwide Links

John J. Boland Johns Hopkins University, Baltimore, USA
Wilhelm Brenig TU München, Germany
Emily A. Carter University of California, Los Angeles, USA
Douglas Doren University of Delaware, Newark, USA
Steven M. George University of Colorado, Boulder, USA
Philippe Guyot-Sionnest University of Chicago, USA
Eckart Hasselbrink University of Essen, Germany
Tony F. Heinz Columbia University, New York, USA
Ulrich Höfer Philipps-Universität Marburg, Germany
John D. Joannopoulos Massachusetts Institute of Technology, Cambridge, USA
Kenneth D. Jordan University of Pittsburgh, USA
Efthimios Kaxiras Harvard University, Cambridge, USA
Kurt W. Kolasinski University of Birmingham, UK
Alan Cooper Luntz Odense University, Denmark
Jens K. Nørskov Technical University of Denmark, Lyngby, Denmark
Matthias Scheffler Fritz-Haber-Institut, Berlin, Germany
John T. Yates University of Pittsburgh, USA