Measurement of the Positron Annihilation Induced Auger Electron Spectrum from Ag(100)

Research has demonstrated that Positron Annihilation Induced Auger Spectroscopy (PAES) can be used to probe the top-most atomic layer of surfaces and to obtain Auger spectra that are completely free of beam-impact induced secondary background. The high degree of surface selectivity in PAES is a result of the fact that positrons implanted at low energies are trapped with high efficiency at an image-correlation potential well at the surface resulting in almost all of the positrons annihilating with atoms in the top-most layer. Secondary electrons associated with the impact of the incident positrons can be eliminated by a suitable choice of an incident beam energy. In this paper we present the results of measurements of the energy spectrum of electrons emitted as a result of positron annihilation induced Auger electron emission from a clean Ag(100) surface using a series of incident beam energies ranging from 20 eV down to 2 eV. A peak in the spectrum was observed at ~40 eV corresponding to the N2,3VV Auger transition in agreement with previous PAES studies. This peak was accompanied by an even larger low energy tail which persisted even at the lowest beam energies. Our results for Ag(100) are consistent with previous studies of Cu and Au and indicate that a significant fraction of electrons leaving the sample are emitted in the low energy tail and suggest a strong mechanism for energy sharing in the Auger process.


Introduction
In this paper, we present the energy spectrum of Auger electrons emitted from Ag(100) obtained using Positron Annihilation Auger Electron Spectroscopy (PAES). In PAES, the Auger electrons are excited by the creation of core holes through an annihilation process. It has been shown [1] that PAES is significantly more surface selective than Electron stimulated Auger Electron Spectroscopy (EAES) and X-ray induced Auger Electron Spectroscopy (XAES) due to the fact that the positrons are trapped in a surface state localized just above the surface at the time of annihilation. The localization of the positrons in the surface state ensures that almost all of the annihilation induced core hole excitations that result in Auger transitions occur in the topmost atomic layer [2]. More importantly, for the purpose of this paper, the energy of the positron beam used in PAES can be made arbitrarily low making it possible to eliminate most or all of the beam induced secondary electrons leaving the surface [3]. In contrast, the highly penetrating beam used in XAES excites Auger electrons from many inelastic mean free paths below the surface. The beam induced secondary background present in EAES and XAES (which is typically many times larger than the signal) makes the measurement of these spectral features extremely difficult. Here we present the results of measurements of energy spectra of electrons emitted as a result of positron annihilation at the surface of Ag(100). The data indicate that a significant fraction of electrons leaving the sample are emitted in the low energy tail.

Resu
The ener positrons which is containin eV, ~240 (~146 eV the inten and inve ctral weight ttering and i gas covering er transition processes in re electrons. eriment asurements w on described urce modera by 40 Gauss ng (without a ata collection time of fligh tion gamma by a channe lectron occu neous with th timing resolu 0 eV and ~ 16 rmanent mag ds produce a g electrons to lectrons leav e region insi ectrum using lts and Disc rgy spectrum s incident on s due to the ng and order 0 eV, and ~3 V and ~373 nsity of these estigated in fu of the tail is inelastic pro g the atomic n and that a n which the e were perform more fully i ated through s axial magn annealing) an n. ht of the Aug signal from el plate detec urs within ~1 he detection o ution of the s 6.5 eV at 250 med using a in Ref [4]. T a thin (1μm netic field tow nd maintaine ger electrons a BaF 2 -PMT ctor placed af 10 -15 sec of t of the gamm system is ~ 1 0 eV [4]. FPAES  the sample bias. The beam energy used in this experiment was approximately 2 eV. We can clearly see that the secondary electron peak has a broad maximum at the lagging edge which is increasing as the sample bias is increasing.
where is the maximum kinetic energy of the impact is induced secondary electrons, is the kinetic energy of the incident positron, is the positron surface state binding energy, and  is the electron work function. Using eq. 1 with calculated values of E b = 3.12 eV,  4.5 we obtain the upper bounds on the beam induced secondary electron energies of approximately 0 eV, 1 eV, 5 eV and 10 eV, for the incident positron beam energies of 2 eV, 3 eV, 5 eV and 12 eV respectively. It may be seen that the significant low energy tail does not disappear even as the positron beam energy is reduced to the point the incident beam can no longer directly knock out secondary electrons. Mukherjee et al. suggested that the similarly large low energy tail that was observed in Cu and Au was due to multi-electron Auger processes [9]. Another possible explanation is that valence electrons are ejected as a result of the sudden creation of the core hole via the annihilation process. In this case conservation of energy would require that the sum of the energies of the two annihilation gammas would be given by eq. 2: where , are the energies of the two gamma rays, m 0 c 2 is the rest mass energy of the positron and electron, is the total energy of the positron in the surface state, BE core is the binding energy of the core electron, and ∑ is the sum of the binding energy of the low energy valence electrons emitted during the annihilation process.

Summary
In this paper we have presented the results of measurements of the spectrum of electrons emitted as a result of positron annihilation induced Auger transitions on an Ag(100) surface. The measurements were carried out for a series of positron beam energies ranging from 18 eV down to 2 eV. All the spectra show a prominent but broad peak at ~40 eV corresponding to the N 2,3 VV Auger transition in agreement with previous PAES studies. The presence of a large low energy tail (LET) in the spectrum below the Auger peak at the lowest beam energies is consistent with previous experiments with Cu and Au, respectively. Mukherjee et al. present arguments that extrinsic processes such as Auger electrons undergoing inelastic backscattering [9] are not sufficient to account for the majority of the LET observed. They suggest that the majority of the LET is due to intrinsic Auger processes in which the energy associated with the filling of the core hole is shared amongst 2 or more valence electrons. Another possible explanation is that a significant fraction of the tail is due to low energy electrons that are emitted during the core hole creation process. Future experiments are planned that will be aimed at determining the relative importance of these different contributions.

Acknowledgement
This work was supported in part by the National Science Foundation Grant #DMR-0907679.