Retarding Grid Research Articles

Angle-resolved ups measurements in a modified LEED system

A LEED chamber has been modified to include a differentially pumped discharge lamp (He or Ne) and an additional retarding grid electron energy analyser for UPS. This small analyser is located at right angles to the LEED analyser and does not interfere with normal LEED and Auger operations. The UPS signal is amplified by a channel plate multiplier and accelerated onto a phosphor-coated screen. Directional information is obtained by scanning this screen with a collimated photomultiplier detector. A phase-lock amplifier is used to differentiate the signal from the photomultipler. Alternatively the phosphor screen can be used as a collector to measure a total spectrum. The acceptance angle of the UPS analyser is 90°. In the angular resolving mode it is possible to observe emission from a (100) fcc crystal in the 〈100〉, 〈110〉 and 〈111〉 directions with a fixed incident photon angle in the range 20–40° to the normal. The acceptance angle of the detector was usually ~7° but this can be varied by changing the collimating tube on the photomultipler. The direction dependent features of the d-band spectrum of clean nickel with a (100) surface have been examined. Characteristic features were observed for each of the 〈100〉, 〈111〉 and 〈110〉 directions. These are compared with those reported for crystals with the corresponding surface orientations. The effects resulting from the chemisorption of nitric oxide on this nickel crystal have also been investigated.

Instrumental effects in the electron loss spectra from surfaces due to the use of LEED optics

The use of a conventional retarding grid LEED-Auger apparatus to record low energy loss spectra from single crystal surfaces has been shown to produce spurious structure in the region within approximately 10 eV from the elastic peak. With the aid of miniature electron guns mounted on the crystal manipulator this structure was shown to arise from transmission effects at the retarding grids when a collimated monoenergetic beam of electrons traverses a limited number of retarding grid meshes. This effect was attributed to an uneven distortion of the electron trajectories due to the use of woven mesh in the manufacture of the grids. It was concluded that low energy loss spectra from single crystal surfaces which are free of this kind of instrumental artifact may be ensured with the incident electrons at normal incidence and at an energy which is sufficiently low so that all LEED beams are excluded from the analysis. Under these conditions the loss spectrum from a W(100) surface contained two broad losses at 2.7 and 5.2 eV which were attributed to interband transitions.

熱電離イオンエネルギー分布測定用の逆電位分析器とマスフィルターを組合せた装置について

A specially designed retarding field analyzer has been set in the front of a quadrupole mass filter for studying the energy distribution of thermal ion emitted from a hot metal surface. The distribution curve is obtained by the technique of electrical differentiation of the retarding plot in which a small modulating voltage is applied to the two retarding grids in a method similar to Auger electron spectroscopy with a four-grid low-energy electron diffraction system. The technique has been applied to study the energy distribution curve of K+ thermal ion emitted from a tungsten ribbon. The curve is Maxwellian in full region of retarding field. The value of retarding potential at maximum peak corresponds to the contact potential difference between the ribbon emitter and the retarding grid.

Spurious effects of electron emission from the grids of a retarding field analyser on secondary electron emission measurements. results on a (111) copper single crystal

Spurious effects of a four grid retarding field analyser were studied for low energy secondary electron measurements. Their behaviour was investigated and two peaks in the energy spectrum were interpreted as resulting from tertiary electrons from the grids. It was shown that the true secondary electron peak has to be separated from these spurious peaks. The spectrum and the yields σ and η obtained for a Cu (111) crystal after a surface cleanness control by Auger spectroscopy are given.

Collision processes of drifting O+and N+ions

Rate constants for charge transfer and interchange reactions of O+ and N+ ions with N2, O2, NO and CO have been measured in the injected ion drift tube, using helium as buffer gas. The range of mean impact energies is 0.04-2 eV. Agreement with previously reported data is, on the whole, good. Energy distributions of ions emerging from the drift tube have been measured using planar retarding grids, and the mean energies E are found to be in agreement with the expression E=1/2m+ nu d2+1/2mg nu d2+3/2kT for ion mass m+, buffer gas mass mg, and drift velocity nu d.

Auger electron and characteristic energy loss spectra of plutonium

The Auger electron and characteristic loss spectra of plutonium have been obtained for the first time using a four grid retarding potential analyzer. The surface of this reactive metal was prepared by scribing in situ in good vacuum with a titanium carbide blade. Oxygen and carbon impurities were still present after scribing. The origin of the Auger electron and loss peaks is suggested, and a correlation is made with the peaks observed for uranium dioxide by Ellis and Campbell.

A combined ESCA and Auger spectrometer

A system which accomplishes high−speed, electron−stimulated Auger electron spectroscopy (AES) and x−ray photoelectron spectroscopy with a double−pass cylindrical−mirror analyzer (CMA) is described. The CMA is well suited for AES because of its very high transmission, but has found very limited use in ESCA studies. In the present system the required source definition for use with a large−diameter x−ray beam is achieved with a two−stage CMA. The required energy resolution for ESCA is achieved with spherical retarding grids. A signal level of 16000 cps is attained on the Ag 3−d lines at 1.25−eV FWHM with a 400 W, Mg anode, x−ray source. Adequate luminosity is achieved primarily through high transmission, rather than a large imaged sample area. The small imaged area (less than 2−mm diam) is useful for selected area analysis and simplifies in−depth analysis when combined with inert−gas sputtering.

Electron loss spectroscopy of clean and oxygen covered GaAs(110) surfaces

Electron energy loss spectra of clean and oxygen covered GaAs(110) surfaces have been measured with a four grid retarding field analyser. Loss spectra of clean cleaved p- and n-type surfaces are slightly different and different states of adsorption for the oxygen on the two surfaces are found. The loss peaks which are common in the spectra obtained from clean surfaces of both types of material have been interpreted in terms of bulk and surface excitations. The data associated with the bulk excitations are in good agreement with previous optical and electron transmission data while loss peaks at 11.5 and 18.5 eV are interpreted as the surface plasma loss and a surface state transition respectively. For n-type material extra loss peaks were observed. In the case of oxygen adsorption on these surfaces new loss peaks were found at 13.5, 17.2 and 28.1 eV in both spectra and are assumed to be characteristic of the oxygen. Further, for n-type material an extra peak occurs at 8.2 eV.

Absolute atomic densities determined by auger electron spectroscopy

The secondary electron distributionN(E) obtained with a spherical grid retarding field analyser is stored in a multichannel analyser. The experimental intensities of Auger lines are accurately determined by numerically substracting the background in theN(E) distribution and taking the area under the resulting peaks. Broadening of the lines due to several experimental factors, the multiple structure of the lines and the characteristics energy losses are taken into account. The absolute atomic densities on the surface are deducted from the Auger line intensities by a simple theoretical model. A comparison is made with atomic densities on the surface which are known either from the crystal structure (cleaved muscovite) or from Rutherford backscattering experiments (thin layer of Ag) or simply from the specific weight in the case of bulk materials (C, Cu, Ag). The maximum deviation is smaller than a factor of 2. Generally, the values differ by about 30%, which shows that AES, performed in this way, can give reliable quantitative results for densities ranging from a fraction of a monolayer to the bulk material.

Quantitative Auger Spectroscopy of Physically Adsorbed Xenon on Nickel

The Auger electron spectrum from xenon adsorbed at 77–85 K on evaporated nickel films has been investigated by means of a retarding grid analyser with post-monochromator. The surface concentration of xenon is quantitatively monitored by the 37 eV xenon peak and by the attenuation of the 60 eV nickel peak. The effect of the incident electron beam on the adsorbate has been measured by quantitative comparison with xenon adsorption isotherms determined volumetrically on the same type of surface. The xenon coverage is found to depend on the current density and electron energy of the incident beam. For the conditions typically used for Auger analysis it is shown that at high equilibrium xenon pressures (>10-8 Torr) the temperature rise due to the power input to the surface is the important effect. At lower pressures electron beam desorption is found to be significant. These effects are discussed in connection with the quantitative analysis of physisorbed layers.

Absolute Atomic Densities Determined By Auger Electron Spectroscopy

The secondary electron distribution N(E) obtained with a spherical grid retarding field analyser is digitized and stored in a multichannel analyser. The experimental intensities of Auger lines are accurately determined by numerically substracting the background in the N(E) distribution and taking the area under the resulting peaks. Broadening of the lines due to several experimental factors, the multiple structure of the lines and the characteristic energy losses are taken into account. The absolute atomic densities on the surface are deducted from the Auger line intensities by a simple theoretical model. A comparison is made with atomic densities on the surface which are known either from the crystal structure (cleaved muscovite) or from Rutherford backscattering experiments (thin layers of Cu, Ag, Ti) or simply from the specific weight in the case of bulk materials (C, Cu, Ag, Ti). The maximum deviation is smaller than a factor 2. Generally, the values differ by less than 30%, which shows that AES, performed in this way, can give reliable quantitative results for densities ranging from a fraction of a monolayer to the bulk material.

Characteristic energy loss and Auger electron spectra of GaP (110)

The characteristic energy loss and Auger electron spectra of clean GaP (110) have been measured with a four grid retarding field analyser. A peak in the loss spectrum has been found at 11.2 eV which is probably due to a surface plasma loss. The remaining structure has been assigned to direct interband transitions, to single and double bulk plasma losses and to d-band transitions by analogy with previous optical and electron transmission studies. Suggestions are made as to the origin of the peaks in the Auger spectrum and changes in the spectrum in the presence of oxygen are discussed.

LEED apparatus: design for high-fidelity diffraction patterns in a strong magnetic field

A design is presented for a LEED (low-energy electron diffraction) apparatus in which diffraction patterns are to be displaced by a uniform magnetic field, so that they are not masked partially from view by the electron gun or by the target crystal. Two useful features of the conventional field-free design are retained: diffraction patterns can be displayed (i) with high angular fidelity and (ii) with little background from electrons which have scattered inelastically in the target crystal. Both features derive from the shape of the display screen and of its enclosed system of concentrically nested electrostatic retarding grids. The shape resembles that of an American football and permits all elastically scattered electrons to enter the grid system at the same angle of incidence regardless of their points of entry.

A Nondispersive Electron Energy Analyzer for ESCA

The energy analyzer described is of the spherical mirror-spherical grid retarding potential type. It obtains high resolution through a small energy overlap between low energy pass and high energy pass filter functions. The first filter is a spherical electron mirror operating slightly off-axis in a reflection mode. The second filter is a conventional spherical grid retarding field operating in transmission. Unlike other narrow bandpass monochromators, acceptance angle does not affect resolution and can be made to yield high luminosity. In addition, the resolution varies as the square of the entrance aperture dimension; hence larger entrance aperture areas can be used than for dispersive electron energy analyzers of similar size. This analyzer is examined theoretically, and experimental results as applied to ESCA (electron spectroscopy for chemical analysis) are presented.

Errors in ion and electron temperature measurements due to grid plane potential nonuniformities in retarding potential analyzers

The behavior of a retarding potential analyzer in a laboratory plasma has been studied to investigate the effects of potential nonuniformities in the grid planes. Significant effects are found for the kind of grids used in the Ogo 1, 3, and 4 traps. It is found that in order to explain the laboratory results the retarding grid potential in the usual planar approximation must be replaced by an effective potential, which may be obtained from the laboratory data. Since the potential distribution about a grid of parallel wires is available analytically, this model is used to calculate the ion current through such a retarding grid. When ion trajectories are taken into account, these calculations show that the effective grid retarding potential is very closely related to the saddle point of the potential near the retarding grid. Applying these corrections to the Ogo 4 O+ temperatures results in a downward revision of about 32%, bringing them into much better agreement with radar backscatter results.

Time-dependence of the electron-impact-induced unimolecular decay process C 3H 8+ → C 3H 7+ + H

A time-of-flight mass spectrometer has been modified to study directly the time dependence of metastable decay processes by introducing a variable time delay between the ionization and extraction pulses and by providing a set of retarding plates and grids in the flight tube to select a single parent ion species for study and to resolve the peak into parent ion, daughter ion and daughter neutral components. The experimentally determined time dependence of the unimolecular decay process C 3H 8 + → C 3H 7 + + H is compared with previously published predictions of the quasiequilibrium theory of mass spectra.

The Use of Scanning Auger Microscopy in Molecular Beam Epitaxy of GaAs and GaP

The morphology of epitaxial films of Groups III–V compounds grown by molecular beam epitaxy has been found to be strongly dependent on the presence of surface impurities, notably carbon. In order to obtain a two-dimensional analysis of substrate surface purity, scanning Auger micrographs were obtained using a modified hemispherical retarding grid analyzer. The surface images were obtained rapidly (<1 min/frame) with a sensitivity in some cases of better than 1% monolayer. The resolution was low (0.1 mm); however, surface areas as large as 5×5 mm could be scanned in a single image, a feature which greatly facilitated substrate evaluation. In some cases crystals showed pronounced spatial variations in surface purity. The technique has also been used to measure surface diffusion rates of Te, As, P, and Cs on GaAs with the result that in most cases the diffusion rates were quite low at temperature up to 600 °C.

Auger Electron Spectroscopy Using 2-Grid LEED System and Photoelectron Multiplier

For the purpose of the detection of atomic species in a few atomic layers of the outermost surface of solids, the electron-bombardment excited Auger electron spectroscopy (AES) using 3-grid LEED system as a retarding field electron energy analyzer is now widely used due to the relative simplicity of the apparatus to observe LEED pattern and Auger spectrum for the same surface under “in situ” conditions. In the case of an earlier 2-grid LEED system, however, a capacitance neutralizer should be used to avoid capacitance coupling and to improve resolution and S/N ratio. The another possibility to eliminate the capacitance coupling between a retarding grid and a fluorescent screen as an electron collector, would be the use of photons emitted from the fluorescent screen. In the case of the multi-grid LEED-AES system, however, the use of a photoelectron multiplier as a detector of Auger electron peaks contained in the secondary electrons from the solid surface, has not yet been reported.Only few investigators suggested the use of the photomultiplier to obtain the energy distribution N (E) of secondary electrons from the solid surface by means of the 3-grid LEED system.It is the purpose of this work to show the capability of photoelectron multiplier combined with the 2-grid LEED system as a retarding field analyzer for detecting Auger electrons. A Fe (100) single crystal ribbon was mounted, as a sample, in the diffraction chamber of the 2-cylindrical grid LEED system (Japan Vacuum Eng. Co., EDL-3) and all measurements were done in the UHV region.The sample surface was bombarded by the primary electron beam (600 eV, 0.7 μA) generated from the LEED gun and the incident angle was approximately 70° from the surface normal. The lower electron current was preferable to decrease the perturbation effects against adsorbing or adsorbed molecules. The energy distribution of secondary electrons emitted from the surface was measured by modulating the retarding grid and scanning its dc potential over the required energy range. The modulation frequency was 20 Hz with an amplitude of about 14 Vp-p, and the time contant used was 3 sec with a scan speed of 10 V/min. The emitted light from the central part of the fluorescent screen held at 3 kV, was then detected by the photomultiplier (Hamamatsu TV Co., R-430, 11 stages, gain of 106), and the ac component of the output was amplified and differentiated by the PAR Model-121 lock-in amplifier to obtain the first and second derivative spectra. The Auger spectrum, in the second derivative form, obtained by the present method for the Fe (100) surface having p (1×1) LEED pattern after ion-bombardment-anneal cleaning, showed only Fe Auger peaks and no traces of other impurities have been detected. The c (2×2) surface structure appeared after heating the clean Fe (100) surface was determined to be due to impurity sulphur and due not to carbon. The minimum detectable sensitivity of about 0.5 percent of a monolayer of the stable adatoms, for example sulphur atoms in the Fe (100) c (2×2) -S structure, was obtained by the present method. When the Fe (100) surface was covered by the weakly bound admolecules such as H2S or CO in the disordered state, the reduced sensitivity of about 5 to 10 percent of a monolayer was obtained because of the electron-bombardment induced desorption. Although the resolution was relatively poor due to the field penetration from the high voltage screen to the retarding grid, the very high sensitivity of this method using low primary electron current has been demonstrated.

A measuring method to distinguish Auger from energy-loss peaks of the primary beam in secondary electron analysis

Introduction Auger peaks are measured from the Fermi level of the sample; some of the energy-loss peaks are referred to the energy of the exciting primary beam. There are also energy-loss peaks due to Auger peaks; they cannot be separated from each other. If one modulates both the exciting primary beam and the retarding grid of a spherical analyzer (outer cylinder for the mirror analyzer) one can tune the lock-in amplifier to a frequency and phase angle such that Auger or energy-loss peaks are detected separately.

Transmission Electron Analyzing Microscope (Team) and Scanning Transmission Eiectron Analyzing Microscope (Steam)

After much difficulty in attempting to measure accurately electron intensities in a conventional transmission electron microscope, it was decided that a quantitative electron analyzing capacility should be designed. Such an instrument should be capable of analyzing both the number and energy of electrons at any point in an image or diffraction pattern. Such a system was developed in the following manner: a set of deflection coils was mounted between the objective lens and viewing screen, an interchangeable aperture was placed in the center of a grounded screen, a retarding grid of variable voltage was positioned behind the aperture, and an electron detector was placed behind the grid, as shown in figure la. Many different types of detection devices can be used. A Faraday cup is useful for quantitative measurements and a light pipe-photomultiplier system is useful for high sensitivity and fast response time with good counting accuracy.