HAL Id: jpa-00223206
https://hal.archives-ouvertes.fr/jpa-00223206
Submitted on 1 Jan 1983
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
APPLICATION OF INFRARED LASER
PHOTOACOUSTIC SPECTROSCOPY TO SURFACE STUDIES
H. Coufal, T. Chuang, F. Träger
To cite this version:
H. Coufal, T. Chuang, F. Träger. APPLICATION OF INFRARED LASER PHOTOACOUSTIC
SPECTROSCOPY TO SURFACE STUDIES. Journal de Physique Colloques, 1983, 44 (C6), pp.C6-
297-C6-300. �10.1051/jphyscol:1983647�. �jpa-00223206�
page C6-
A P P L I C A T I O N OF INFRARED LASER PHOTOACOUSTIC SPECTROSCOPY TO SURFACE STUD I ES
H. Coufal, T.J. Chuang and F. ~ r z ~ e r *
ISM Research Laboralory, Sun Jose, CuZifornia 9 5 1 9 3 , U.S.A.
Rksumk
-
Sous des conditions UHV des me sures ont Btabli la sensibilitk de detection photoacoustique d'une submonocouche.Abstract
-
Adsorption studies have been performed under UHV conditions establishing a submonolayer sensitivity of photoacoustic detection.INTRODUCTION
We report on experiments establishing photoacoustic spectroscopy as a technique for surface studies.
This detection method in combination with laser excitation offers a unique combination of advantages such as high sensitivity, high spectral resolution and instrumental simplicity. Taking advantage of real time compensation techniques the substrate signal can be eliminated making this technique applicable to all combinations of adsorbates and substrates in various environments. This technique permits recording of vibrational spectra at coverages of a fraction of a monolayer. Thus, the detailed adsorbate-substrate as well as adsorbate-adsorbate interactions can be investigated.
Furthermore, it is anticipated that this method is of value for analytical studies.
NONABSORBING SUBSTRATES
In our experiments a cw C02-laser linetunable in the infrared spectral region between 9 and 11 pm with an output power in the order of 1W is used to excite vibrational transitions of adsorbed molecules. The unfocused beam was incident at 7S0'from the surface normal and covered the entire sample area about 7 mm in diameter. The light beam is intensity modulated at frequencies of approximately 10 Hz. Due to radiationless decay this results in a periodic release of heat in the sample which causes thermal expansion and induces acoustic waves. These can be detected with a pyro-resp. piezoelectric transducer which is directly attached to the substrate and designed to meet stringent UHV conditions. A piezoceramic disc 10 mm diameter and 1 mm thickness, metallized on both sides and sealed around its perimeter with a glass film to prevent outgassing, was used as a detector. T o achieve maximum sensitivity the laser intensity incident on the sample had to be stabilized with an electronic feedback system.
The sample under study is mounted in a n UHV system (10-lo Torr) equipped with a n ESCA-Auger spectrometer, a sputter ion gun and a mass spectrometer. X-ray photoemission measurements are used to analyze the surface and to determine the amount of surface coverage. Simultaneous photoacoustic and X-ray photoemission measurements allow after a suitable calibration /1/ to evaluate the photoacoustic signal as a function of coverage.
So far, SF6, NH3 and pyridine adsorbed o n silicon and on silver films at liquid nitrogen temperature have been studied. The results show that the method has, indeed, submonolayer sensitivity.
Without employing signal averaging techniques a sensitivity corresponding to a surface coverage of 0.002 of a monolayer of SF6 on silver could be achieved. That high a sensitivity can be readily
*Permanent address : Physikalisches Institut der Universitat Heidelberg, Philosophenweg 12, D-6900 Heidelberg, Federal Kepublic of Germany
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1983647
C6-298 JOURNAL DE PHYSIQUE
Ti
me (seconds)Fig. 1 - Time dependence of PA and XPS-Ag(3d(3/2)) signals recorded during the adsorption of NH3 on a Ag surface a t 90K for C 0 2 laser at v=1031.5 cm" and v=1063.7 cm-l. The final NH3 coverages are 8=4.2 and 8=2.8, respectively.
achieved for non absorbing, i.e., highly reflecting or transparent substrates using a stabilized laser of sufficient output power. Time dependent phenomena like the adsorption/desorption cycle of SF6 o n silver /2/ or the adsorption of NH3 can be readily observed (Fig. 1). In addition, vibrational spectra from submonolayer to multilayer coverages have been obtained; they indicate distinctive adsorbate-substrate interactions and a nonlinear relation between PA-signal and coverage (Fig. 2).
ABSORBING SUBSTRATES
Photoacoustic spectroscopy applied to adsorbed species or thin films on a substrate suffers from a serious problem: the signal originating from the material o n the surface is superimposed on a signal due to the substrate which can be several orders of magnitude larger and gives rise to strong noise fluctuations. This makes the detection of adsorbed species or thin films o n the surface of a n absorbing substrate or the accurate determination of their thickness difficult. This background problem cannot be overcome by electronic techniques like zero suppression o r out of phase detection.
If one, however, could suppress this background signal the detection sensitivity can b e largely enhanced o r a unstabilized lightsource used. In addition, the technique can be much more convenient to use for analytical purposes, i.e., for the detection of a certain type of thin film or adsorbate on the surface.
We therefore explored a new scheme of general applicability which permits complete background suppression for a large variety of experimental purposes. It has been shown to permit ultrahigh sensitivity photoacoustic probing (0.01 of a monolayer) of thin film and adsorbate materials on absorbing substrates.
Fig. 2 - Photoacoustic spectra of the v2 transition of NH3 adsorbed on a Ag surface at 90K for coverages of 0=0.5 and 0= 1.7.
Substrate and thin film, respectively adsorbate are excited by a suitable pulsed or modulated lightsource thus generating a thermal wave; a second source with an appropriate amplitude and phase is used to generate a thermal wave with an amplitude identical to that one due to the substrate but with a 180 degree phase shift. This second thermal wave adds to the first thermal wave.
Contributions due to the substrate therefore result in a DC heating of the sample and cannot contribute t o the AC PA-signal. The thermal wave due to the thin film respectively absorbate, however, still causes an A C signal which is no longer buried by any background. The amplitude and phase adjustment of the compensation source can be readily achieved by zeroing the signal of the uncovered substrate.
Several versions of this new technique have been tested succesfully which illustrate that the scheme can be realized in completely different ways to achieve the same goal e.g., by applying a single light beam or in a two beam arrangement. In addition, different physical principles can be used.
In the first one a dielectric coating on a conducting substrate was studied. A particular surface property was used here: the difference in absorption between s- and p-polarized light is different for the substrate and the adsorbate. Therefore, by illuminating the substrate with alternating polarizations (i.e., s- and p-polarized light waves phase shifted by 180 degrees) the relative amplitude of the electric field vector for the two polarizations was adjusted in such a way to obtain a zero substrate signal. The resulting intensity modulation causes a photoacoustic signal originating only from the dielectric or adsorbate layer.
C6-300 JOURNAL DE PHYSIQUE
In another experiment the thermal wave due to the substrate was cancelled by a second thermal wave generated at the backside of the sample with a suitable amplitude and phase shift to achieve zero substrate signal. The thermal wave from the thin film deposited on the front side, however, is not cancelled and causes a photoacoustic signal. Using the same laser to excite both thermal waves eliminates long term drifts efficiently and therefore lowers the requirements for laser stabilization.
Using two lightsources with different wavelenghts /3/ does not have this advantage but instead allows to take advantage of differences in the absorption of substrate and adsorbate and to measure deposition rates differentially.
Preliminary experiments o n conducting substrates showed that instead of undergoing the formidable task of stabilizing a laser essentially the same sensitivity can be achieved by resistance heating of the substrate. The phase of the current and the gain of the feed back loop is adjusted prior to the adsorption; during the exposure the fluctuations of the laser are compensated efficiently by the feed back loop.
Experiments using an oscillating beam spectrometer /4/ to compare two samples for differential probing / 5 / of their photoacoustic signal are under way.
CONCLUSION
In conclusion, the feasibility of infrared laser photoacoustic spectroscopy at submonolayer coverages and under ultrahigh vacuum conditions has been demonstrated. This new surface analytical technique, alone o r in combination with the compensation schemes that are under development, is not only able to provide important information at high sensitivity and spectral resolution (0.1 cm-') but should also have a high potential for applications to chemical systems in various environments, whether in vacuum or not.
REFERENCES
1. CHUANG, T. J., J. Appl. Phys.
51
(1980) 2614 and references therein.2. TRAGER, F., COUFAL, H., and CHUANG, T. J., Phys. Rev. Lett.
2
(1982) 1720.3. COUFAL, H. and PACANSKY, J., IBM Tech. Disclosure Bu11.g (1980) 4681.
4. SEKI, H. and ITOH, U., Rev. Sci. Instrum. (1980) 22.
5 . COUFAL, H. and PACANSKY, J., IBM Tech. Disclosure Bull.
