ANISOTROPY IN SURFACE DIFFUSION OF POTASSIUM ON THE W(112) REGION

(1)

HAL Id: jpa-00226897

https://hal.archives-ouvertes.fr/jpa-00226897

Submitted on 1 Jan 1987

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.

ANISOTROPY IN SURFACE DIFFUSION OF POTASSIUM ON THE W(112) REGION

J. Båben, Ch. Kleint, R. Måclewski

To cite this version:

J. Båben, Ch. Kleint, R. Måclewski. ANISOTROPY IN SURFACE DIFFUSION OF POTAS- SIUM ON THE W(112) REGION. Journal de Physique Colloques, 1987, 48 (C6), pp.C6-545-C6-550.

�10.1051/jphyscol:1987689�. �jpa-00226897�

(2)

ANISOTROPY IN SURFACE DIFFUSION OF? POTASSIUM ON THE W(112) REGION

J. b b e n , Ch. ~leint*and R. Meclewski

Institute of Experimental Physics, University of Wrockaw, 5 0-205'Wroctaw, Poland

* ~ e p a r t m e n t of Physics, Karl Marx University Leipzig, 7010 Leipzig, G.D.R.

Abstract - Measurements of anisotropic surface diffusion of potasslum submonolayers on the (112) region of tungsten were carried out using a special version of the field emission fluctuation method. A rectangular slit was used as a probe hole which could be rotated in situ. The mean square fluctuations for successive slit positions were measured and the corresponding values of the diffusion coefficient were calculated by using theoretical results obtained recently by Gesley and Swanson. Activation energies were calculated from the temperature dependence of the diffusion coefficient determined in the temperature range between 395 and 490 K. High anisotropy in the surface diffusion coefficient was found at lower temperatures and -also its decrease with increasing temperature. Surface diffusion along the. [llll direction is much faster than along the [I101 direction. This can be explained by the anisotropy of the substrate atomic structure. The ratio A=D[lll]/D[llO] is about 10 at 400 K but only 1.5 at 490 K. The subscripts indicate cristallographic directions on the W(112) p lane.

I

-

INTRODUCTION

Recently a special version of the field emission.(FE) flicker noise method for studies of adsorbate surface diffusion was reported by Gomer and coworkers [I]. This modification consists of using a long,

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1987689

(3)

JOURNAL DE PHYSIQUE

narrow rectangular slit which can be rotated as a probe hole in a FE tube. In this case the correlation function (or the spectral density function) corresponds to the fluctuation of the adsorbate surface density caused by surface diffusion in a direction perpendicular to the longer slit side. The diffusion parameters obtained on the basis of the measured correlation function (or spectral density function) correspond to the mentioned direction. In this way the diffusion parameters can be determined for successive angular positions of the slit, i.e. for the chosen crystallographic directions. In the present work this method was applied for a study of the anisotropy in surface diffusion of potassium on the (112) tungsten region.

I 1

-

EXPERIMENTAL

The measurements of the spectral density functions were made with a sealed-off-, glass FE tube working under UHV .conditions. A metal anode. with a rectangular slit as a probe hole-scould be rotated in situ. The lepght of the probe area was about Id cm, the ratio of the slide sides 1:8. The error in orientation of the slit normal to the (112) row direction is smal ler than 15'. A Faraday collector was located behind the slit. The FE pattern could be observed on the screen as usual and adjusted to the slit position by an external magnetic field. Other details of the FE tube and the technique of the noise measurements were described elsewhereI21.

FREQUENCY

Fig.1. Two chosen spectral density functions for slit- like collector. Crosses are experimental points and solid lines are the one-dimensional approximation according to t31.

(4)

(5)

I 1 1

-

^RESULTS

The spectral density functions were measured for 12 positions of the slit at a constant substrate temperature and constant coverage.

Representative examples of the results are shown in Fis.1. The crosses mark the obtained values and the solid lines correspond to the theoretical spectral density function for one-dimensional diffusion,

s",

derived by Gesley and Swanson (Eq.78 and Fig. 1 of Ref.[31). They considered also the case of the high frequency

limit (Eq.80 of Ref. [3] ) which is applied to the determination of the diffusion coefficient D in the present paper by using a formula:

lZa3 D =

---

2

where

dD(d)

is the spectral density for the one-dimensional diffusion, P the noise power and 1 the length of the slit. The calculations were made for a frequency of f =0/28=12500 Hz.

Fig. 3. Arrhenius plot of the diffusion coefficient D.

(6)

A=D[lll]/D[llO] is about 10 ( subscripts indicate crystallographic directions on the (112) tungsten plane). At higher temperatures the anisotropy decreases and almost disappears at 490 K. From the measured temperature dependence of the diffusion coefficient the activation energy Q was determined. An example is presented in Fig.3. Similar results were obtained for other positions of the slit. The angular dependence of the activation energy Q is presented in Fig.4. The anisotropy is in agreement with results obtained by Bayat and Wassmuth [41, Bgben, Kleint and Meclewski [51, and BZaszczyszyn and Kleint [ 6 ] , where other methods were used. The values of Q obtained in the present paper are smaller than those in Ref. [41 and [51.

Fig,4. The angular dependence of the surface diffusion actlvatlon energy.

(7)

IV

-

SHORT DISCUSSION

Though the slit position was adjusted carefully to the centre of the W(112) plane there might be some influence of the (112) vicinals because of the slit length used. The decreasing anisotropy with temperature is ascribed to an increasing tendency of the adsorbed submonolayer to behave as a two-dimensional gas. The fact that we found 4 maxima in the angular dependence of the diffusion coefficient at low temperature which do not coincide with the row structure might be explained by a soliton mechanism f7.81. It seems plausible that the domain wall movements induced by kink site motion [91 appear for distinct coverages in the maximum D direction observed in the present experiment.

Acknowledgement

Work sponsored by the Polish Ministry of Science and Schools of Academic Rank within the Central Project of Basic Research CPBP 01.08.Al.

REFERENCES

111. D. R. Bowman, R. Gomer, K. Muttalib and M. Tringides, Surf ace Sci. 138 (1984) 581;

M. Tringides and R. Gomer Surface Sci. 155 (1985) 254.

[ 2 1 . J. Bgben. Ch. Kleint and R. Meclewski, 2. Physik /in press/;

Appl. Phys. A40 (1986) 79.

[31. M. A. Gesley and L. W. Swanson, Phys. Rev. B32 (1985) 7703.

[41. B. Bayat, H. W. Wasmuth, Surface Sci. 131 (1983) 1.

[51. J. Bgben. Ch. Kleint and R. Meclewski, J. Physique C9, Suppl. 45 (1984) C9-13.

[61. R. BYaszczyszyn and Ch. Kleint, Surface Sci. 171 (1986) 615.

[71. Ch. Kleint, 8th Inter. Seminar on Surface Physics. Karpacz 1984, Poland. Acta Univ. Wrdtislaviensis 47 (1986) 81.

181. V . L. Pokrovski, _{A .}L. Talapov, Zh. Exp. Teor Fiz. 78 (1980) 269.

[91. J. Beben, Ch. Kleint and A. Pawelek. 11th Inter. Seminar on Surface Physics, Piechowice 1987, Poland.