Search

We studied the stability of metal/polymer interfaces by measuring the diffusion of calcium into a polymer (OC 1 C 10 PPV) layer during and after deposition of the metal using low energy ion scattering (LEIS) and X-ray photoelectron spectroscopy (XPS). During deposition the calcium diffusion depth in the PPV was found to be comparable for untreated samples and samples prepared in oxygen ambient (10 −7 mbar). In both cases diffusion depths up to ∼7 nm were observed. For PPV layers treated with atomic oxygen, the diffusion depth during deposition was slightly smaller. After deposition, it was observed that calcium diffusion in OC 1 C 10 PPV continues for several hours. When oxygen was present during calcium deposition or during the spin coating of the PPV, the diffusion coefficient for calcium in PPV was decreased considerably. In these cases accumulation of oxygen (adsorbed in the PPV during deposition or spin coating) at the calcium/PPV interface continued for several hours after deposition. Treatment of the PPV with atomic oxygen before calcium deposition resulted in a strong decrease of the calcium diffusion coefficient after deposition. From XPS measurements it was observed that calcium interacts with the chemically bonded oxygen in the PPV and also with the oxygen absorbed in the PPV layer. It can be concluded that oxygen, either chemically bonded to the PPV chain or adsorbed in the film, reduces the calcium diffusion coefficient. The initial performance of PLEDs with atomic oxygen treated PPV layers and PLEDs with calcium deposited in oxygen ambient was worse than the performance of untreated devices, but the stability in the life-test was better.

In this work we address the dynamics and stability of calcium/PPV and barium/PPV interfaces during and after deposition of the metal. Diffusion of calcium and barium into OC/sub 1/C/sub 10/ PPV is studied with low energy ion scattering (LEIS) and x-ray photoelectron spectroscopy (XPS). During metal deposition the diffusivity is found to be orders of magnitude higher than after deposition and the diffusion coefficient was found to be dependent on the metal concentration in the PPV. Furthermore, the amount of metal inside the polymer films was found to depend on the deposition rate. These observations were explained in a two-stage diffusion model. In the first stage atoms land on the surface and diffuse fast into the polymer and in the second stage metal ionizes and is trapped and diffusion is strongly decreased. The diffusion coefficient of barium into PPV at T=298 K is found to be almost an order of magnitude lower than the diffusion coefficient of calcium into PPV [(0.35+or-0.05)*10/sup -23/m/sup 2//s and (2.7+or-0.4)*10/sup -23/ m/sup 2//s, respectively]. Furthermore, the activation energy of the diffusion process of barium into PPV (0.75+or-0.07eV) is significantly higher than the activation energy of the diffusion process of calcium into PPV (0.62+or-0.05 eV). The difference in diffusion coefficient and activation energy between calcium and barium are discussed in terms of an Arrhenius law of diffusion. Finally, polymer LED performance was studied as a function of the amount of metal diffused into the polymer layer. It was observed that the light output and the efficiency decreased as the amount of metal in the PPV increased. This indicates that the metal ions form charge carrier traps and exciton quenching sites in the PPV

In polymer light emitting diodes (PLEDS) with an (ITO/PPV/Ca) structure we obsd. a significant redn. of both the current and the light output at const. voltage after heat treatment for only 30 min at 65 DegC. Electroluminescence spectroscopy expts. showed that the shape as well as the amplitude of the spectra were changed.The redn. of current and light output was investigated by measuring I-V and E-V (current-voltage and brightness-voltage) characteristics of PLEDs, I-V characteristics of single carrier devices, and by performing low energy ion scattering and XPS expts. on the Ca/PPV interface.It was concluded that the current and light output redn. could be ascribed to the degrdn. of the Ca/PPV and the ITO/PPV interfaces. The degrdn. of the ITO/PPV interface resulted in a redn. of the zero field hole mobility and a small increase of the field dependence of the mobility. The degrdn. of the Ca/PPV interface, probably by diffusion of calcium into the PPV, resulted in carrier traps and quenching sites, which influenced the field dependent electron mobility

A combination of low- and high-energy ion beam analysis techniques was used to determine the distribution of indium chloride compounds in model polymeric light-emitting diodes (p-LEDs). Parts of polymeric LEDs (poly-dialkoxyphenylenevinylene (OC/sub 1/C/sub 10/-PPV) on indium-tin-oxide (ITO) substrates) were exposed to a HCl/Ar flow to simulate the processes occurring during conversion of precursor PPVs and acid treatment of polymers. Samples with variable exposure times as well as pristine samples were studied with Rutherford backscattering spectrometry (RBS), low energy ion scattering (LEIS), X-ray photoelectron spectroscopy (XPS) and particle induced X-ray emission (PIXE). The RBS measurements show that after HCl exposure indium is distributed throughout the OC/sub 1/C/sub 10/-PPV layer. LEIS and XPS measurements indicate that the indium and chlorine are present at the outermost surface of the OC/sub 1/C/sub 10/-PPV layer. PIXE measurements in combination with the RBS data demonstrate that the indium chloride in the OC/sub 1/C/sub 10/-PPV layer is distributed in blisters ( phi = 50 mu m) which are orientated perpendicular to the sample surface. To study the OC/sub 1/C/sub 10/-PPV/ITO interface, samples were delaminated by peeling the polymeric layer of the ITO using Scotch tape. LEIS measurements on the fracture planes demonstrate the formation of an interfacial layer between the ITO and the PPV. The implications for p-LEDs are discussed

The effect of controlled exposure to O on the stability of Ca cathode of polymeric light emitting diodes (pLEDs) was studied. The LEDs were fabricated with ITO as anode, poly[2-methoxy-5-(3',7'-dimethyloctyloxy)-1,4-phenylenevinylene] [OC1C10-PPV] as electroluminescent polymer, Ca as cathode and Al as protecting layer. The polymer layers of the LEDs were spin coated under dry N atm. and transported directly into an UHV chamber where the metal electrodes were deposited by evapn. To study the effects of exposure to O of the Ca cathode, deposition of the Ca layer was interrupted in some cases and the samples were exposed to 30-1000 mbar of O. The amt. of O in the layers of the pLEDs was detd. from current-voltage data and elastic recoil detection and was correlated with device performance parameters. Exposing a part of the Ca layer to O at layer thicknesses equal to or less than 10 nm led to a total loss of brightness. The failure of pLEDs is attributed to oxidative degrdn. of OC1C10-PPV in contact with oxidized Ca and O at the interface. The damage and failure can be prevented if fabrication of the devices is carried out in a totally inert atm

The influence of oxygen exposure on three chemically different poly para -phenylene vinylene (PPV) derivatives used in polymeric light emitting devices (PLEDs) has been investigated. During device preparation, PPV layers have been exposed to oxygen either in the dark or during the irradiation with visible light, before the cathode was applied. Device efficiency was studied by electrical and optical characterisation and correlated to oxygen depth profiles measured with Elastic Recoil Detection Analysis (ERDA). Treatment with oxygen during light exposure leads to a decrease in current, light output and efficiency. It was found that two different PPV derivatives show the same current and light output reduction with different oxygen uptake. This behaviour is explained in terms of a different chemical structure and the number of structural defects (tolane-bisbenzyl moieties (TBBs)) incorporated. For two PPV derivatives treated with oxygen in the dark a reduction in current and light output was found, while the efficiency was unchanged. Another PPV derivative, however, shows a shift in the on-set voltage of the light output accompanied by an increased oxygen level at the PPV/Ca interface. It is concluded that after exposure in the dark, oxygen is incorporated in the PPV by Van der Waals interaction. During evaporation of the cathodes, oxygen will diffuse to the cathode and will be gettered by the calcium, which results in the formation of an electron injection barrier.

The energy loss distribution of ions transmitted through a 5.7±0.2 μm thick Si crystal was measured and simulated with the Monte Carlo channeling simulation code FLUX . A general resemblance between the measured and simulated energy loss distributions was obtained after incorporation of an energy dependent energy loss in the simulation program. The energy loss calculations are used to investigate the feasibility to detect the presence of light element dopant clusters in a host crystal from the shape of the energy loss distribution, with transmission ion channeling. A curved crystal structure is used as a model for a region in the host crystal with clusters. The presence of the curvature does have a large influence on the transmitted energy distribution, which offers the possibility to determine the presence of dopant clusters in a host crystal with transmission ion channeling.

Low energy ion implantation at high doses of boron (>1015 cm–2) in Si is necessary for the fabrication of ultrashallow junctions but can result in the undesirable presence of boron clusters. Values for the dimensions of the lattice distortions in the implanted Si are obtained by comparing the enhanced dechanneling and the direct scattering peak in the region with clusters in a channeled Rutherford backscattering spectrometry spectrum to those from Monte Carlo calculations on a curved crystal structure. Values of about 0.17 and 65 nm are found for the maximum deformation and the length of the distortions in the crystal, respectively, which implies that the lattice distortions extend significantly outside the layer in which the B clusters are supposed to be present. ©2001 American Institute of Physics.

We have measured axially channeled Rutherford backscattering spectra of Si1-xGex nanofilms in silicon(001). A step in the yield of the host crystal was found for off-normal axes at the depth of the nanofilm. The step was measured as a function of the angle between the incoming beam and the [011] axis and shows two maxima. It is found that Monte Carlo simulations assuming tetragonal distortion reproduce the experimental results. A universal curve was derived which enables determination of the tetragonal distortion from ion-channeling experiments, for a given film thickness. The results are compared with XRD measurements.

Polymers with an ordered molecular structure can be applied in optical systems for e.g. data transport, data storage and displays. Patterned UV photo-polymerization is used to prepare polymer gratings from a mixture of two acrylate monomers. A 3 MeV proton microprobe is used to study these gratings, prepared from two different monomers, each containing a different easily detectable label element, e.g. Cl, Si or F. During the preparation process, the difference in reactivity and mobility of these two monomers in combination with polymer–monomer interaction results in diffusion of monomers. Since this diffusion process takes place on length scales of micrometers, a scanning ion microprobe is a powerful tool for the quantitative analysis of the polymer films, obtained after complete polymerization. The microprobe is equipped with PIXE, PIGE and RBS, to quantify both the label elements and C and O. This makes it possible to determine the concentration of monomer units as a function of position and thus to study the diffusion process. Two combinations of different monomers are studied. In the case of a 0.5:1 mixture of a monofunctional and a difunctional monomer, both monomers migrate to the illuminated areas and large thickness variations are observed. When a 1:1 mixture of two difunctional monomers is used, opposite migration of the two monomers is observed, while the film shows no variation in thickness.

Raman spectroscopy can be applied to characterise the chemical composition of an atherosclerotic plaque in vivo. In the near future this technique may become available for use in (coronary) arteries of living patients. For this moment, Raman spectroscopy is applied on artery samples in vitro, to study progression and regression of atherosclerotic plaque. Raman spectroscopy provides chemical information on a molecular basis. In this study, micro-particle induced X-ray emission (micro-PIXE) is applied to provide additional information on the elemental composition of the artery. Furthermore, the combined techniques allow for validation of the structures studied with Raman spectroscopy. This study proves that it is possible to combine and compare both techniques using the same region on the same sample if proper sample preparation is applied. Comparison shows that regions appearing in the Raman spectroscopy results can also be distinguished in micro-PIXE and backscattering spectroscopy (BS) distributions and vice versa. Combining both techniques makes it possible to separate phospholipids from triglycerides. Combined Raman spectroscopy and micro-PIXE/BS is recommended for studying progression and regression of atherosclerosis. © 2001 Elsevier Science B.V. All rights reserved.

The influence of gallium ions on the development of the mineralization process in a cell culture model was investigated. To determine the elemental composition the proton microprobe in combination with proton induced X-ray emission (micro-PIXE) and energy-dispersive X-ray fluorescence (XRF) were used. Micro-PIXE allows a precise quantification of elements combined with their accurate localization. In contrast, the XRF results may be used as a rough estimate of the mineralization process development. It was found that gallium influences the mineralization process in a dose-dependent manner. At low doses an increased while at high doses an impaired mineralization was observed. The amount of deposited gallium correlates with the concentration of the element in the cell culture medium. © 2001 Elsevier Science B.V. All rights reserved.

Planar MeV ion-channeling measurements on 2.2 nm thick Si1-xGex nano-films buried in Si are presented. The presence of the nano-film leads to a step in the yield of the host crystal in a {0 1 1} planar channeled RBS spectrum. In previous work we showed that with the help of Monte Carlo (MC) simulations the step height in axial channeling measurements provides useful information on the tetragonal distortion in buried nano-films. Since the step height is larger for planar channeling measurements, we extend this method to planar channeling in the present work. The measurements show that the flux distribution of channeled ions is not in statistical equilibrium, but still oscillates at the depth of the buried films (about 280 nm), This is confirmed by simulations performed with the MC simulation code FLUX and a qualitative understanding is obtained for the observed phenomena. The accuracy of the calculated step height curves appears much more sensitive to the choice of ion-atom potential than those from axial channeling, which can be attributed to the non-equilibrium distribution of channeled ions between the planes. Consequently, from planar channeling measurements, the tetragonal distortion can be determined with about the same accuracy as from axial channeling measurements in spite of the higher steps. (C) 2001 Elsevier Science B.V. All rights reserved.

We recently constructed a high energy ion scattering (HEIS) setup in which samples can be cooled to cryogenic temperatures to suppress ion beam induced damage in polymers. A closed cycle helium gas cooler provides the cooling power of 2 W at 10 K. The setup is equipped with a sample mount that enables sample exchange via a load lock system, while ensuring a good thermal contact with the sample holder by a shrink coupling. We investigated the effects of sample cooling on HEIS measurements on model polymer light emitting diodes and polyacrylate films. It can be concluded that hydrogen loss can be decreased from 50% to 10% for the same irradiation dose, and much more accurate depth profiling of elements is possible.

The chemical composition and crystalline nature of inorganic deposits present in the human aortic wall were investigated. A proton microprobe in combination with proton induced X-ray emission and proton induced gamma-ray emission were used to determine quantitatively the elemental composition of the inorganic deposits while their structure was studied with the use of an X-ray microprobe in combination with X-ray diffraction and Fourier transform infrared microscopy. It was found that deposits in the aortic wall are composed of Ca–P compounds. Fluorine is accumulated in the inorganic phase. However, the mechanism of F deposition awaits further investigations. The only phase detected by the X-ray diffraction was a defective hydroxyapatite. The relative intensities of reflexes indicate that the c axis of the crystallites is parallel to the long axis of the vessel. IR microscopy revealed that the crystals were contaminated with carbonate groups. Moreover, the band characteristic for octacalcium phosphate was detected.

The diffusion of indium into poly-(phenylenevinylene) (PPV) in model polymer light-emitting diodes (p-LEDs) was studied with Rutherford backscattering spectrometry (RBS), X-ray photoelectron spectroscopy (XPS), low energy ion scattering spectroscopy (LEIS) and particle induced X-ray emission (PIXE). The model p-LEDs consisted of a glass substrate, an indium–tin-oxide (ITO) electrode, a PPV layer obtained by thermal conversion of sulfonium precursor PPV, and a patterned aluminium electrode. From RBS measurements it was concluded that about 0.01 at.% indium was present in the PPV, homogeneously distributed in depth. Annealing at 230°C for 19 h caused the amount of In in the PPV layer to increase by roughly an order of magnitude. Under the patterned aluminium electrode, the annealing treatment resulted in accumulation of In at the PPV/Al interface, whereas the depth distribution of In remained homogeneous in the uncovered region of the model LEDs. XPS spectra on annealed model LEDs show that In was present in the near surface region of the PPV films, although LEIS analysis showed that In was not situated in the outermost atomic layer. LEIS measurements on as-prepared model LEDs showed that the patterned Al electrode had caused surface contamination of the uncovered PPV film with Al, which can have impact on the diffusion of In to the outermost surface during annealing treatments.

Strain in and around pyramidal InAs/GaAs quantum dots (QD’s) fabricated by molecular-beam-epitaxy influences the density of states of the confined charge carriers. The presence of strain in QD’s is required to explain their optical properties. In this paper MeV ion-channeling experiments are presented which provide evidence for the presence of strain in and around InAs QD’s in GaAs. The small dimensions of the QD’s (typical height 4 nm) and the presence of a wetting layer complicate the interpretation of channeling measurements, but our experiments show that extended strain fields around the QD’s induce ion steering which accounts for the observed channeling behavior.

We investigated the elemental composition of different layers of polymeric light-emitting diode (pLED) model structures. Of special concern have been poly-(3,4-ethylenedioxythiophene) (PEDOT)/polystyrenesulfonate (PSS) and poly-phenylenevinylene (PPV) films, which are used widely in pLEDs. We found halogen compounds as impurities: bromine in the PEDOT/PSS and chlorine in the PPV layer. It is argued that the presence of halogen containing compounds could be an important reason for the degradation of pLEDs.

Patterned UV photo-polymerization is used to prepare polymer gratings from a mixture of mono- and di-functional acrylate monomers. The difference in reactivity and mobility of these two monomers induces a concentration gradient during the photo-polymerization process. Uniform illumination afterwards fixes the grating. Monomers with different easily detectable elements, e.g., Cl, Br, Si, F, enable the observation of lateral variations of concentration in these gratings with PIXE and PIGE using a scanning ion probe of 3 MeV protons. In addition, backscattering spectrometry is applied to correct for lateral thickness variations.

The proton microprobe in combination with proton induced X-ray emission (micro-PIXE) and with proton induced gamma-ray emission (micro-PIGE) are used to determine quantitatively the modulation of inorganic deposits formation by Sr and F ions in a cell culture model. The results indicate that the process may be investigated by the micro-PIXE determination of the amount of calcium deposited. It was found that F-treatment stimulate bone formation at doses much lower than Sr. At high doses an impaired mineralization is found for both elements. It was found that the mechanisms responsible for of F and Sr incorporations are different. The minimal F and Sr concentrations in the medium at which the incorporation may be investigated by micro-PIGE and micro-PIXE amount to ∼0.02 mg/l and

The aim of the study was to measure with microbeam PIXE elements such as Na, K, Mg, and Ca in cardiac tissue after various treatments in vivo, which affect the cardiomyocyte integrity. It was assumed that local deviations from normal electrolyte levels indicate the degree of cardiac damage. The first step in this feasibility study was comparison of severely damaged cardiac tissues with controls. Severe cardiac damage was introduced by the so-called Ca paradox. Experiments were performed with isolated rat hearts, perfused retrogradely with an oxygenated crystalloid buffer. Results indicated that severe cardiac damage was accompanied with almost complete disappearance of the normal intracellular electrolyte composition as a result of the loss of membrane integrity. Identifications of smaller and more locally present ischemic damages on basis of altered electrolyte levels appeared to be feasible. However, the prerequisite was that the mobility of electrolytes be kept under control during tissue sampling and sample preparation, when physiological mechanisms stop to maintain gradients.

A model for mass transport in biporous packed beds was used to simulate axial concentration profiles of n-hexane measured with positron emission profiling (PEP). Axial dispersion dominated the mass transport behavior in the medium-pore zeolites H-mordenite and H-ZSM-5. Comparison of simulations of experiments using the small-pore zeolites (i.e., H-ZSM-22 and H-ferrierite) shows that micropore diffusion limitation contributes significantly to the peak width of the PEP profiles. As a result, new micropore diffusion constants for H-ZSM-22 and H-ferrierite are reported. The prospects for studying the reaction kinetics of the hydroisomerization reaction of alkanes is discussed.

Positron emission profiling (PEP) has been used to obtain in situ images of the pulses of 11C-labeled alkanes passing through packed beds of acidic zeolites at temperatures typical of the hydroisomerization reaction. Used in this manner, PEP represents an in situ version of tracer pulse chromatography. The two-dimensional activity-position-time profiles measured were fit to an analytical solution. The method of moments was then applied to the subsequent fitted profiles to extract information regarding adsorption and diffusion of the hydrocarbons. Data concerning the mass transfer of alkanes are necessary to perform further modeling of profiles obtained during the hydroisomerization of n-alkanes on Pt/acidic zeolites.

Two alternative detector types have been studied for use in the Eindhoven Scanning Ion Microprobe set-up. First, the applicability of a Charge Coupled Device (CCD) system for X-ray spectroscopy has been explored. Second, some properties of the SiTek type 1L30 Position Sensitive Detector (PSD) for charged-particle spectroscopy have been studied. A literature survey shows that excellent X-ray spectroscopy with a CCD system is feasible, particularly with a deep-depletion backside-illuminated CCD and low speed read-out. If, however, high-speed CCD read-out is required, such as for scanning microprobe experiments, a CCD system cannot be used for spectroscopy due to excess read-out noise. For the PSD, noise theory calculations are presented, which result in a noise shaping time for optimal energy and position resolution. In practice, however, a much longer time is needed to obtain sufficient energy and position linearity. Characterization measurements of the PSD using our 4 MeV He+ microprobe are also described. A position resolution of 0.47 mm and a position linearity of better than 0.15% detector length are found. In addition, an energy linearity better than 0.3% and an energy resolution of 36 keV are measured. The latter will have to be improved, to make the PSD suitable for charged-particle spectroscopy applications.

Ion channeling for materials analysis is usually carried out with He ion beam energies less than 3 MeV. The AVF-cyclotron at Eindhoven University of Technology allows the use of higher beam energies (3–30 MeV) which is of potential interest due to a high angular resolution, a large analysable depth range and the application of nuclear reactions or resonant scattering cross-sections. In this paper the enhanced angular resolution for 12.6 MeV He ions is demonstrated by angular scans from a buried Si 1− x Ge x layer. The enhanced analysable depth range is shown by quantitative defect depth profiling of a 4 μm thick relaxed SiGe multilayer buffer. Finally, a study on vibrational amplitudes in bulk InP was carried out by a combination of channeling and nuclear reaction analysis using the 31 P(α,p) 34 S reaction.

The use of electron irradiation in polymer research for polymer modification is a well known method. Recently the technique of pulsed electron beam polymerization has been introduced as a tool for investigations into the kinetics and mechanisms of free radical polymerization. The technique can be applied to homogeneous and in particular to heterogeneous polymerization systems such as an emulsion polymerization. The major advantages are: (1) no limitation to optical transparency of the system; (2) no additional initiator is required. The 6 MeV linear accelerator (linac) at the Cyclotron Laboratory of the Eindhoven University of Technology is used for the present investigations. In this contribution, the accelerator set-up will be described and pulsed electron beam polymerization experiments will be used to illustrate the new technique. The irradiation set-up operates with pulse repetition rates between 1 and 50 Hz at a monitored dose per pulse adjustable between 0.1 and 3 Gy. The reaction temperature can be varied up to 100°C. Results of the current investigations in the polymerization of styrene in bulk and methyl methacrylate in emulsion will be presented.

In the study of interfacial diffusion processes in polymer light-emitting diodes, the use of high-energy ion-scattering techniques can be of great value due to the possibility of quantitative elemental depth profiling. However, ion irradiation of polymers is known to cause various degradation effects, including the loss of hydrogen. Since the hydrogen loss determines the accuracy of depth profiling, it is an interesting subject for study in order to define experimental conditions hi which the degradation is suppressed. The loss of hydrogen from porphyrins (organic solar cells) has been measured by means of elastic recoil detection analysis with 2, 4, and 7.6 MeV He+ beams. A theoretical model is proposed in which the hydrogen loss is described through the formation and recombination of free hydrogen radicals. A distinct difference is introduced between direct recombination processes and the diffusion of radicals out of the ion track.

Radiation damage effects in a PIPS particle detector have been studied by direct irradiation with a 4 MeV He+ focused ion beam, using the Eindhoven scanning ion microprobe set-up. This set-up enables extreme sensitivity and accuracy in pulse-height measurement. Pulse-height dependence on ion dose and the lateral variations of pulse-height response after damaging were studied, for different bias voltages. Pulse height was found to decrease linearly with ion dose and this could be observed after only a few ions per scan position, corresponding to a dose of 108/cm2. Pulse-height loss was found to be restricted to damaged detector areas. Consequences of this pulse-height dependence on ion dose and scan position for the accuracy of STIM analyses are discussed.

The deconvolution procedure based on the maximum-entropy method is applied to improve both resolution and noise characteristics of micro-PIXE maps. The procedure is simple to execute, fast, easily automated and assumes no a priori information. The best improvement is observed for data with low statistics. Special emphasis is given to the difficulties and limitations of processing micro-PIXE data. Examples of applications for biomedical samples are presented.

In the Eindhoven coincident elastic recoil detection analysis by time-of-flight (CERDA-TOF) spectrometer we have improved sensitivity by focusing the recoils onto the detector by a magnetic quadrupole doublet. In order not to disturb the depth resolution, the focusing was limited to the azimuthal direction, perpendicular to the scattering plane (reaction plane). Sensitivity improves by a factor of 5 to about 1 × 1014 atoms/cm2 for carbon in a carbon foil and by a factor of 4 to 1 × 1014 atoms/cm2 for oxygen in a native oxide layer on a 1 μm amorphous Si-layer.

In elastic recoil detection analysis (ERDA) with a-particles, recoils can be separated from scattered projectiles by using detectors with a thin depletion layer and a pulse shape discrimination (PSD) circuit [S.S. Klein and H.A. Rijken, Nucl. Instr. and Meth. B 66 (1992) 393]. Improved performance is obtained with a multi-parameter data acquisition system; recoil events are more easily and accurately selected and sensitivity is improved by a factor of 5 to 1 X 1014 at/cm2 for oxygen on amorphous silicon. The typical surface depth resolution is 10 nm. Two alternative PSD circuits are proposed neither of them using fast electronics. Examples are given of profiling carbon and oxygen in thin films of Si, C and SiOxNy.

Channeling with high-energy He or H ions offers enhanced angular resolution since the characteristic angle for channeling decreases with increasing energy. In this study channeling experiments with high-energy He ions (4–15 MeV) have been performed with the Eindhoven AVF cyclotron, carefully controlling important parameters such as beam divergence and target alignment. We applied high-energy ion channeling using a three-axis goniometer with high angular accuracy for strain analysis in Si/Si1 − xGex/Si single-layer heterostructures with x = 0.175 and 0.05. The strain is derived from axial angular scans through the 〈110〉 and 〈111〉 directions in the {100} and {110} planes, respectively. The obtained scans are interpreted using the ion-trajectory simulation program LAROSE [J.H. Barrett, Phys. Rev. B 3 (1971) 1527]. The perpendicular lattice constant of the deformed Si1 − xGex layers was determined with an accuracy comparable to the accuracy reached with high resolution X-ray diffraction.

The application of deconvolution procedures (Fourier deconvolution, iterative least-squares, maximum entropy) for the determination of a two-dimensional profile of a proton microprobe is described. It is found that the maximum-entropy method is the best suited deconvolution procedure for the restoration of the microprobe profile. The use of the method is feasible event for data with 40% uncertainty. Differences of the microprobe profiles calculated on the basis of X-ray and secondary electron distributions, are observed. The explanation of the phenomenon is proposed.

As part of a project to study the reaction kinetics in catalysts, a detector system has been designed and built. The detector will measure in one dimension the activity distribution of positron emitters in catalyst reactors under operational conditions as a function of time. The detector consists of two arrays of ten BGO crystals each and has the flexibility to measure with high sensitivity the activity profile in various reactor sizes; the position resolution that can be reached is 3 mm.

In-beam studies of high-spin states in Eu-149 are reported. The level scheme extends up to an excitation energy of 7.1 MeV and a spin of 55/2HBAR. This nucleus is weakly deformed and most of the high-spin structure is interpreted through its multi-particle-hole nature. Octupole-phonon vibrations coupled to multi-particle-hole states are observed and discussed. In contrast with the Z less-than-or-equal-to 62 isotones, there is no conclusive evidence in Eu-149 for octupole deformation induced by rotation.

The construction and experimental results of a novel compact anti-Compton spectrometer system are reported. The device consists of a large BGO scintillation crystal with a cylindrical well. Inside the well, a 56 mm diameter, 57 mm high (n-type) germanium crystal is mounted. The entire assembly is cooled down to 104 K. Scintillation light from the BGO crystal is detected by a 127 mm diameter PM tube. The assembly has the advantage of a compact design and the absence of any material between the Ge and the BGO crystal. A peak-to-total ratio of 0.60 is obtained with a /sup 60/Co source and 0.74 with a /sup 137/Cs source. >

The statistical γ-ray decay of the compound nucleus 154 Dy ∗ formed at an excitation energy of 69 MeV is studied in three angular-momentum windows 〈 J 〉 = 31,42 and 50ℏ. The GDR strength function extracted from the data indicates a constant centroid energy for the resonance E GDR = 15.2 ± 0.5 MeV, a small deformation of the nuclear shape for all spin regions and an increase in the resonance width with increasing angular momentum. The latter might indicate larger shape fluctuations at higher angular momenta. A systematical study of the dependence of the extracted strength function with different methods of analysis is presented, and their discrepancies discussed.

Recoils ejected by α-particles are detected by low-resistivity detectors at low bias. Pulses from scattered α-particles that cross the depletion layer are rejected by pulse shape analysis. Using 500 Ω cm detectors the maximum pulse height recorded for α-particles corresponds to an energy of 4 MeV. Further suppression of pulses caused by α-particles awaits the construction of detectors optimized for operation at low bias as well as the development of faster electronics. Using a 13.4 MeV α-particle beam oxygen has been detected from layers as deep as 250 nm. Sensitivity was improved to 5 × 1014 atoms cm2 and a surface depth resolution of 12 nm was obtained for N in Si2N0.86O3. The depth range and sensitivity may be further improved using higher bombarding energies.

An improved coincidence time-of-flight recoil spectrometer is described. The use of alpha-particles instead of heavy ions to eject the recoils has the advantages of larger depth range and smaller target damage. Multi-element analysis and element-selective depth profiling on thin and moderately thick targets are demonstrated. A surface depth resolution of 3 nm has been obtained for 12C in carbon at a recoil angle φ of 30°. Resolution down to 1 nm is predicted when further compensation of kinematic effects is realized; using very well defined beams it might be possible to obtain atomic layer resolution. Strategies to extend the present sensitivity of about 1014 atoms cm2 by one or two orders of magnitude are discussed.

A method of quantitative analysis of bone samples by simultaneous measurements of proton induced X-ray emission (PIXE) and Rutherford backscattered (RBS) spectra is described. The method is used for the evaluation of the data collected during proton microprobe irradiation. The RBS spectra are used for the determination of the sample thickness. The sample matrix is regarded as a mixture of collagen and hydroxyapatite. The unknown proportion of both compounds at each irradiated point is adjusted until the Ca concentration determined by PIXE is in accordance with the value calculated on the basis of the assumed matrix composition. For quantitative analysis, the calculation of the absolute yield using the thick target PIXE formula was applied. The sources and ranges of systematic errors are discussed.

The new scanning microprobe has been in operation for two years. real time data acquisition takes place on a front end system. The measured data is transported via Ethernet to a workstation for online monitoring, analysis and storage. Hardware for control of the scan magnet and the quadrupoles and a multichannel analyzer have been developed. Data acquisition software for online monitoring and simultaneous display of multiple spectra with user selectable parameters has been written. The user interface is based on the X11-Window system. A relational database gives the user a flexible way to specify all parameters for the experiment. A faster datalink and a faster workstation will be realized this year.

Stopping powers for channeled He ions have been calculated with a modified version of the Monte Carlo code LAROSE [J.H. Barrett, Phys. Rev. B 3 (1971) 1527]. The spatial distribution of the valence electron density in Si was obtained from bandstructure calculations. The stopping power was calculated using Lindhard's free-electron gas approach within the framework of the local density approximation. Spatial variations of the electron density along individual trajectories produce a significant contribution to the energy loss distribution. The average energy loss of 4 MeV He ions channeled in the axial 〈100〉, 〈111〉 and 〈110〉 directions have been calculated and compared with measured values. The contribution of the core electrons to the energy loss is investigated by adding the spherically symmetric Hartree-Fock-Slater electron densities of the closed shells to the valence electron density. Calculations show a high energy loss tail in the spectrum qualitatively in agreement with published experimental results.

Rutherford Backscattering Spectrometry (RBS) is shown to be a powerful tool in the analysis of model catalysts. The surface coverage of various metals on thin SiO2 layers on Si and thin Al2O3 layers on Al can be accurately measured while simultaneously depth profiles of the metals are obtained. The scattering technique is briefly reviewed and several applications concerning the preparation of model catalysts by wet chemical methods are presented.

Elastic recoil detection (ERD) of carbon, nitrogen and oxygen from thin films on thick substrates was demonstrated using 10-15 MeV He2+ ions and thin films detectors. Non-Rutherford scattering was used to alter the relative sensitivities for carbon, nitrogen and oxygen by selecting specific recoil angles and incident energies. A depth resolution of 15 nm near the surface can be easily achieved as well as probing depths up to approximately 500 nm.

Detection of low Z recoils (C, N, O) from scattering of high energy alpha particles (10–15 MeV) is compared with conventional ERDA with heavy ion projectiles. Recoils and scattered a particles are separated using pulse shape discrimination or coincidence methods. Submonolayer sensitivity and depth resolution in the nm range is obtained. Depth/mass ambiguity may be avoided.

In this article the uncertainty in the PIXE yield due to the uncertainty in various parameters such as the ionization cross section, the stopping power, the X-ray attenuation coefficient, the matrix composition and the proton beam energy, is discussed. This is done for both the primary PIXE and the secondary fluorescence yield. A numerical approach to the propagation of an uncertainty in a parameter to the uncertainty in the yield is given. For this a new parameter, the propagation factor, which is the partial relative error in the yield due to the error in a parameter divided by the relative error in the parameter, is introduced. The dependence of the propagation factor on the X-ray and proton energies and or the matrix composition is investigated for the above named parameters. The physical background of these dependencies is explained; this also makes it possible to obtain a better physical insight in the formulas for the primary yield and the secondary fluorescence yield.

In the past particle accelerators were invented and developed to a large extent for scientific purposes. As time proceeded the relatively small machines became dedicated to applications while the fundamntal research required higher and higher energies and improved ion optical beam quality. In many university laboratories machines originally used for fundamental physics are now applied for analyzing techniques and technological developments. Furthermore a very large number of accelerators is installed in industrial sites and hospitals. To a certain extent the fundamental theory of acceleration was developed separately for each type of accelerator. In this paper we will try to give a general theoretical treatment for the smaller accelerators as well as some indroductory historical remarks and a rough extrapolation to the future.

Detector limitations to the energy resolution can be avoided by measuring the time of flight over a sufficient distance. For 6 MeV 12C ions the energy resolution has been improved from 70 to 20 keV on a 3 m flight path. The depth resolution obtained in CERDA (coincident elastic recoil detection analysis) with a 12.6 MeV beam from the Eindhoven AVF cyclotron improves by a smaller factor (from 50 to 30 nm) because the depth resolution is now dominated by kinematic effects of beam emittance and detection solid angle. Without sensitivity loss the depth resolution may be further improved to 15 nm selecting the beam in a more effective way.

The isochronous low-energy cyclotron ILEC is a small AVF-machine which has keen built for acceleration of protons to a fixed energy of 3 MeV. The total weight of the machine is 3 ton and its power consumption is about 20 kW. Apart from providing an experimental facility for machine research, a main objective of this accelerator project is to produce an intense ion beam with low energy spread for element analysis. For this purpose ILEC is equipped with a double 50° dee system for 2nd harmonic excitation at a frequency of 43.5 MHz and a high voltage amplitude of about 30 kV. To attain the requirement of low energy spread also a double 30° system for 6th harmonic flat topping has been constructed and will be installed soon. During the last year the Eindhoven scanning proton muprobe has been updated. The aim of the new design is to obtain a probe diameter of a few μm with a beam current of at least 100 pA. The main application will be the determination of trace elements in biological and medical samples at cellular and subcellular level. In combination with the detection of scattered particles, applications in the field of mu-electronics and materials science are foreseen.

This paper describes the investigation of the possibility to build a compact anti-Compton spectrometer consisting of a Ge detector and a cooled BGO suppressor. Monte Carlo calculations were carried out to obtain information about the performance of two different detector configurations. Light output and timing properties of a BGO-photodiode detector have been measured as a function of temperature and will be compared with computer simulations of light output.