Gradient flipping algorithm: introducing non-convex constraints in wavefront reconstructions with the transport of intensity equation.

The transport of intensity equation (TIE) is widely applied for recovering wave fronts from an intensity measurement and a measurement of its variation along the direction of propagation. In order to get around the problem of non-uniqueness and ill-conditionedness of the solution of the TIE in the very common case of unspecified boundary conditions or noisy data, additional constraints to the solution are necessary. Although from a numerical optimization point of view, convex constraint as imposed to by total variation minimization is preferable, we will show that in many cases non-convex constraints are necessary to overcome the low-frequency artifacts so typical for convex constraints. We will provide simulated and experimental examples that demonstrate the superiority of solutions to the TIE obtained by our recently introduced gradient flipping algorithm over a total variation constrained solution.

Inline electron holography and VEELS for the measurement of strain in ternary and quaternary (In,Al,Ga)N alloyed thin films and its effect on bandgap energy.

We present the use of (1) dark-field inline electron holography for measuring the structural strain, and indirectly obtaining the composition, in a wurtzite, 4-nm-thick InAlGaN epilayer on a AlN/GaN/AlN/GaN multinano-layer heterosystem, and (2) valence electron energy-loss spectroscopy to study the bandgap value of five different, also hexagonal, 20-50-nm-thick InAlGaN layers. The measured strain values were almost identical to the ones obtained by other techniques for similarly grown materials. We found that the biaxial strain in the III-N alloys lowers the bandgap energy as compared to the value calculated with different known expressions and bowing parameters for unstrained layers. By contrast, calculated and experimental values agreed in the case of lattice-matched (almost unstrained) heterostructures.

FDES, a GPU-based multislice algorithm with increased efficiency of the computation of the projected potential.

While the computational complexity of calculation of the projected potential in the multislice algorithm through reciprocal space scales quadratically with the number of atoms A per slice, a pure real-space calculation scales linearly with A. A hybrid strategy is introduced that has a theoretical complexity of O(AlogA), but that, when measured, outperforms both the reciprocal-space and the real-space approach by approximately an order in A and a large factor, respectively. This strategy is implemented in a new program, dubbed forward dynamical electron scattering (FDES), which simulates high resolution transmission electron microscopy images, diffraction patterns and convergent beam electron diffraction patterns. FDES attains a further increase in speed by running on a graphics processing unit and is made available to the community as open software.

A practical way to resolve ambiguities in wavefront reconstructions by the transport of intensity equation.

The transport of intensity equation (TIE) provides a very straight forward way to computationally reconstruct wavefronts from measurements of the intensity and the derivative of this intensity along the optical axis of the system. However, solving the TIE requires knowledge of boundary conditions which cannot easily be obtained experimentally. The solution one obtains is therefore not guaranteed to be accurate. In addition, noise and systematic measurement errors can very easily lead to low-frequency artefacts. In this paper we solve the TIE by the finite element method (FEM). The flexibility of this approach allows us to define additional boundary conditions (e.g. a flat phase in areas where there is no object) that lead to a correct solution of the TIE, even in the presence of noise.

Hybridization approach to in-line and off-axis (electron) holography for superior resolution and phase sensitivity.

Holography--originally developed for correcting spherical aberration in transmission electron microscopes--is now used in a wide range of disciplines that involve the propagation of waves, including light optics, electron microscopy, acoustics and seismology. In electron microscopy, the two primary modes of holography are Gabor's original in-line setup and an off-axis approach that was developed subsequently. These two techniques are highly complementary, offering superior phase sensitivity at high and low spatial resolution, respectively. All previous investigations have focused on improving each method individually. Here, we show how the two approaches can be combined in a synergetic fashion to provide phase information with excellent sensitivity across all spatial frequencies, low noise and an efficient use of electron dose. The principle is also expected to be widely to applications of holography in light optics, X-ray optics, acoustics, ultra-sound, terahertz imaging, etc.

Toward quantitative core-loss EFTEM tomography.

Core-loss EFTEM tomography provides three-dimensional structural and chemical information. Multiple inelastic scattering occurring in thick specimens as well as orientation-dependent diffraction contrast due to multiple elastic scattering, however, often limit its applications. After demonstrating the capability of core-loss EFTEM tomography to reconstruct just a few monolayers thin carbon layer covering a Fe catalyst particle we discuss its application to thicker samples. We propose an approximate multiple-scattering correction method based on the use of zero-loss images and apply it successfully to copper whiskers, providing a significant improvement of the reconstructed 3D elemental distribution. We conclude this paper by a general discussion on experimental parameters affecting the accuracy of EFTEM 3D elemental mapping.

Direct imaging of surface plasmon resonances on single triangular silver nanoprisms at optical wavelength using low-loss EFTEM imaging.

Using low-loss energy-filtering transmission electron microscopy (EFTEM) imaging, we map surface plasmon resonances (SPRs) at optical wavelengths on single triangular silver nanoprisms. We show that EFTEM imaging combining high spatial sampling and high energy resolution enables the detection and for the first time, to the best of our knowledge, mapping at the nanoscale of an extra multipolar SPR on these nanoparticles. As illustrated on a 276.5 nm long nanoprism, this eigenmode is found to be enhanced on the three edges where it exhibits a two-lobe distribution.

Determining the radial pair-distribution function within intergranular amorphous films by numerical nanodiffraction.

We report on an alternative method to electron nanodiffraction and fluctuation microscopy for the determination of the reduced density function G(r) of amorphous areas with small cross-sections. This method is based on the numerical extraction of diffraction data from the complex-valued exit-face wave function as obtained by HRTEM focal series reconstruction or electron holography. Since it is thus possible to obtain "diffraction data" from rectangular areas of any aspect ratio, this method is particularly suited for intergranular glassy films of only 1-2 nm width, but lengths of several 100 nm. A critical comparison of this method with the already established nanodiffraction and fluctuation microscopy will be made.

Aspects regarding measurement of thickness of intergranular glassy films.

Materials such as Si(3)N(4), SiC and SrTiO(3) can have grain boundaries characterized by the presence of a thin intergranular amorphous film of nearly constant thickness, in some cases (e.g. Si(3)N(4)) almost independent of the orientation of the bounding grains, but dependent on the composition of the ceramic. Microscopy techniques such as high-resolution lattice fringe imaging, Fresnel fringe imaging and diffuse dark field imaging have been applied to the study of intergranular glassy films. The theme of the current investigation is the use of Fresnel fringes and Fourier filtering for the measurement of the thickness of intergranular glassy films. Fresnel fringes hidden in high-resolution micrographs can be used to objectively demarcate the glass-crystal interface and to measure the thickness of intergranular glassy films. Image line profiles obtained from Fourier filtering the high-resolution micrographs can yield better estimates of the thickness. Using image simulation, various kinds of deviation from an ideal square-well potential profile and their effects on the Fresnel image contrast are considered. A method is also put forth to objectively retrieve Fresnel fringe spacing data by Fourier filtering Fresnel contrast images. Difficulties arising from the use of the standard Fresnel fringe extrapolation technique are outlined and an alternative method for the measurement of the thickness of intergranular glassy films, based on zero-defocus (in-focus) Fresnel contrast images is suggested. The experimental work is from two ceramic systems: Lu-Mg-doped Si(3)N(4) and SrTiO(3) (stoichiometric and nonstoichiometric). Further, a comparison is made between the standard high-resolution lattice fringe technique, the standard Fresnel fringe extrapolation technique and the methods of analyses introduced in the current work, to illustrate their utility and merits. Taking experimental difficulties into account, this work is intended to be a practical tool kit for the study of intergranular glassy films.

Iterative phase retrieval without support.

An iterative phase retrieval method for nonperiodic objects has been developed from the charge-flipping algorithm proposed in crystallography. A combination of the hybrid input-output (HIO) algorithm and the flipping algorithm has greatly improved performance. In this combined algorithm the flipping algorithm serves to find the support (object boundary) dynamically, and the HIO part improves convergence and moves the algorithm out of local minima. It starts with a single intensity measurement in the Fourier domain and does not require a priori knowledge of the support in the image domain. This method is suitable for general image recovery from oversampled diffuse elastic x-ray and electron-diffraction intensities. The relationship between this algorithm and the output-output algorithm is elucidated.

Cholescintigraphy with 99mTc-diethyl-IDA for the detection of rejection of auxiliary liver transplants in pigs.

In 15 pigs an auxiliary liver was transplanted and cholescintigraphy with 99mTc-diethyl-IDA was performed to study the function of the livers separately. Serial scintigraphy of the recipient and donor liver was performed on days 3, 8 and 18 after transplantation and time-activity curves over 1 h after injection of the radio-tracer were generated for each liver. From these data the time at which the maximal activity was present in the liver (Tmax) and the time at which during the excretion phase half of this activity was still present in the organ (T1/2) were determined. Liver biopsy material from the recipient liver and the donor liver was obtained at (approximately) the day of one of the scintigraphic examinations and at autopsy. In all cases no histological abnormalities were found in the recipient liver. In 4 pigs there were no histological signs of rejection of the donor liver, in which cases the values for Tmax and T1/2 were normal. In 11 pigs histological signs of rejection were noticed and in all of these cases T1/2 was significantly prolonged, whereas no changes in Tmax were observed. Cholescintigraphy revealed an abnormal value for T1/2 a few days before the biopsy was taken in 10 of these 11 cases. Normal functioning or the prediction of rejection of the auxiliary liver transplant in pigs correlated significantly (P less than 0.01) with normal or prolonged T1/2 values. It is concluded that cholescintigraphy may be helpful in establishing the diagnosis of rejection of liver transplants.

Human skin grafts from mixed lymphocyte culture-positive donors provide help for the rapid rejection of simultaneously transplanted skin grafts from mixed lymphocyte culture-negative donors.

The influence of grafting more than one skin transplant simultaneously on one recipient was investigated. When a mixed lymphocyte culture (MLC)-negative skin was transplanted along with an MLC-positive skin, the MLC-negative skin survived for a significantly shorter time than when transplanted alone. This indicated that the MLC-positive skin provided a stimulus that could provide help to reject the MLC-negative skin. This finding might be important clinically. When an MLC-negative transplant is given to a patient, one should not transfuse this patient with MLC-positive leukocyte-rich blood.

Influence of matching for HLA-DR antigens on skin graft survival.

HLA-DR typing results of 47 skin transplant donor-recipient pairs were analysed. HLA-A, B, and C typing and mixed lymphocyte culture (MLC) testing was also included in this study. Skin transplants exchanged between HLA-A-, B-, and DR-identical, MLC-negative donor-recipient pairs had the longest graft survival (mean survival time, 17 days), whereas skin grafts exchanged between completely nonidentical donor-recipient combinations had the shortest survival (mean survival time, 10 days). Because of the correlation between identity for the DR antigens and the low or nonreactivity in the MLC test, identity for DR will predict a better skin graft survival than nonidentity. It was concluded that the best match between donor and recipient of a graft, using only serological techniques, is one where there is identity for HLA-A, B, and DR.

The immunologic system, immunologic deficiency disorders and lymphoid hyperplasia of the small intestine.

The purpose of this paper is to review our current concept of the systems that subserve immunological functions. It concerns specifically the immunodeficiency diseases probably due to a B cell immunodeficiency. An attempt is made to classify those diseases clinically as well as roentgenologically. At last using the enteroclysis technique some examples of immunodeficiency syndromes are shown with special emphasis on developing a malignant disease.