Table of Contents


The positron, which is the antiparticle of the electron, has the same mass and the same spin, but the opposite charge and magnetic moment. The positron is a stable particle in vacuum, although in metals it rapidly theramlizes, and annihilates with an electron predominently via two gamma-ray decay (511 keV) with a mean life time that is typically only a few hundred pico-seconds. Positrons are emitted by artificially produced radioactive iostopes with proton rich nuclei such as 22Na. When a proton within a nucleus decays into a neutron (n) through beta decay, a positron and a neutrino are produced.



The PAES mechanism can be outlined as follows: 1. Positrons implanted at low energy diffuse to and become trapped at the surface. 2. A few percent of the trapped positrons annihilate with core electrons leaving some of the surface atoms in an excited state. 3. The atoms relax via emission of an Auger electron. The PAES mechanism is contrasted with that of electron induced Auger Electron Spectroscopy (PAES) in Fig. 1



Serious limitations are imposed on conventional electron excited Auger electron Spectroscopy (EAES) by the fact that an electron beam of sufficient energy to excite a core hole creates a background of backscattered and secondary electrons which is typically many times larger than the Auger signal. In PAES the secondary background under the Auger peak can be eliminated by using a positron beam less than that of the Auger electrons (see Figs 2.a and 2.b).
An important result of the PAES measurements has been an indication that published Auger line shapes are in error due to problems with background substraction. (See Figs. 3a and 3b).



The large improvement in signal to background permits PAES measurements to be made with energy doses about 5 orders of magnitude lower than doses required for EAES. Weiss et al. estimate that PAES measurements can be made with charge doses ~1000 times lower than EAES measurements for a given value of the signal to noise ratio. In addition the beam energy used in PAES (~25eV) is about 0.01 times the energy of the electron beam used in EAES. Thus PAES can reduce the energy dose required for the measurements by ~5 order of magnitude as compared to EAES. The large reduction in charge may make PAES useful for the study of fragile systems, such as weakly physisorbed adsorbate systems. In EAES, the incident beam (few keV) can cause damage, and a charging problem in insulators, and study of adsorbed layers, et cetera.



The positron diffuses back to the surface and becomes trapped in an image-correlation potential well before it annihilates. Because the positron wave function is localized within a few angstroms of the surface, annihilation takes place preferentially with atoms in the topmost atomic layer. This gives PAES a very high degree of selectivity for the elemental composition of the top atomic layer. In contrast, the higher energy beam used in EAES excites atoms deep below the surface restricting the surface selectivity of EAES to the 4 - 10 A escape depth of Auger electons. Thus the EAES signal represents an average over several atomic layers.


Calculated atomic potential (top) and ground state wave function (bottom) for a positron trapped in a surface state at clean Cu(100) surface. Note the localization of the positron at the surface which gives PAES its high degree of surface specificity.

(middle) PAES study of intermixing of Pd deposited on Cu(100)

PAES spectra taken at three different temperatures for ~0.69 monolayers Pd deposited on Cu(100) at 173K are shown in Fig. The PAES spectrum for the as deposited surface at 173K (top) shows predominantly the Pd N23N45N45 (40 eV) Auger peak. The middle figure shows the PAES spectrum after the sample was heated to 303K without further Pd deposition. Here the appearance of the Cu M23M45M45 (60 eV) Auger peak is clearly observed as well as the decrease in the Pd Auger peak. As shown in bottom figure, heating the sample to 423K causes the Pd peak to almost disappear as the Cu peak grows, indicating that the Pd is moving below the top surface layer. The corresponding EAES spectra display only a small decrease in the Pd peak as the sample is annealed with temperature. A quantitative analysis of the PAES and EAES data taken together indicates that the Pd is only going one layer below the Cu surface.

(right) Comparison of Specificity, PAES vs.Surface EAES

Plots of "surface concentration versus the amount of Pd deposited on Cu(100) measured using PAES (top) and by EAES (bottom) demonstrate the higher surface selectivity that is possible with PAES. Note that in the PAES data the measured concentration of Pd saturates at 100% and the Cu intensity drops to zero by the time one monolayer of Pd has been deposited at 173K. In contrast, in the EAES data, the measured "surface" concentration of Pd is only ~50% after one monolayer deposition due to the fact that EAES averages over several layers. Saturation of the Pd concentration as measured by EAES does not occur until many layers of Pd are deposited.



[Main] [Introduction] [Current Research] [Publications] [Personnel]