We have developed a system (1) for automatically acquiring large numbers of high quality transmission electron micrographs under low dose conditions. This system has been implemented on a Philips CM200 Transmission Electron Microscope equipped with a Gatan MSC CCD camera. Implementing the automated data acquisition system requires that target locations be identified in a low magnification image and then accurately located at the center of the viewing area. The magnification is subsequently increased, the image is focused on an area adjacent to the target area and the final image is acquired. Centering an identified target location for subsequent high magnification imaging typically requires moving the specimen by many thousands of nm and accurately locating the target to within a few hundred nm. This movement is too large to be achieved using the image shift coils, which would be very accurate, and instead must be achieved using the goniometer.

We have measured the accuracy of the goniometer on the Philips CM200 and the results are shown in fig 1. Data were obtained by selecting a target area from a low magnification image [660x], moving to this target and then measuring the accuracy of the requested movement by cross correlation.

Since it's introduction into TEM in 1986 [1] SSCs have become an effective and easy-to-use solution for acquiring high quality digital images electronically. Their main advantages are excellent linearity, very high dynamic, high sensitivity, and low noise. These advantages have made SSCs a nondispensable tool for quantitative image analysis. Moreover, since SSCs make high quality TEM images available in a computer within fraction of seconds, on-line image processing and software driven automated TEM tuning has become possible [2]. Many of the experiments which are performed nowadays in the area of high resolution, low dose, holographic reconstruction, and EFTEM [3] would not be possible without the digital input coming from SSCs.

Like recording images with a film sheet camera, digital recording of images with SSC requires control of the TEM beam blanker (shutter) by the SSC synchronised with the image acquisition process.

This blanker (shutter) control is a critical link in the chain to the digital image. Many older TEMs do not have direct access to beam blanking coils, or contrary to modern TEMs their coils are not designed to support hysteresis-free fast beam blanking, what is essential for acquisition of high quality digital images by a SSC.

Recent advances with the cone-beam approach to tomography at the Advanced Microscopy and Imaging Laboratory (AMIL) have opened the possibility of employing tomography with the Shadow Projection X-Ray Microscope routinely when the specimen properties require it. The AMIL approach has been tested on small objects, without magnification, using conventional dental x-ray sources and a modestly diverging beam angle. The shadow projection Microscope, by contrast has a tenfold greater divergence and magnifies the specimen from unity to three or more orders of magnitude in resolution beyond the previous tests. The reader is reminded of the simplicity of the xray projection method and of the clarity of its images in fig. 1 and 2. The microscope can be operated in any laboratory and readily transported to alternate locations. Now that the AMIL data base and algorithms for tomography can be handled by a Pentium II processor the data gathering and reconstructions can also be managed at any laboratory installation.

Micro X-ray fluorescence (micro-XRF) is a useful nondestructive technique for multielement analysis of samples from a few hundred to a few micrometers in size. Micro-XRF is usually performed at synchrotron sources where element sensitivities are not compromised by micrometer spatial collimation of high flux beams. Ideally, micro-XRF would not have the burden of transporting samples to a synchrotron facility, but could be performed on a laboratory or portable scale. At the National Institute of Standards and Technology we are developing a micro-XRF system that combines a polycapillary optic with a standalone X-ray tube. This optic not only transmits the X-ray beam over 80 mm, but also focuses the beam to an area smaller than the unfocused beam size. The fine resolution combined with a high intensity produces a tool for laboratory scale micro-XRF analyses.

The X-ray focusing optic is a tapered, glass polycapillary optic. This optic, contained in an approximately 5 × 70 mm steel cylinder, transmits and focuses incoming X-rays to a spot 9.5 mm from the optic exit.

In photoelectron microscopy (PEM; also known as PEEM) ultraviolet light is used to liberate electrons from the surface of a specimen through the photoelectric effect. The emitted electrons (photoelectrons) have low kinetic energies (∼1 eV) and are accelerated by an electric field between the specimen and an anode, before being imaged by a lens system. The lens system forms the image of the specimen in parallel mode, as in TEM, rather than in scanning mode, as in SEM. Image contrast arises from differences in work function and from electric or magnetic fields over the specimen surface. Micro-fields associated with surface topography are also a source of contrast. Because of the low emission energies of the electrons PEM is a highly surface-specific technique and also is relatively non-destructive.

Several versions of the photoelectron microscope have been designed and constructed jointly by the University of Oregon (U0) and Portland State University (PSU), and the instrument has evolved to an advanced state of development.

The optimal method for evaluating fluorescence microscope has not been proposed. In this study we developed two quantitative methods using biological and non-biological materials as standard specimens. Based on these methods, a new water immersion lens was tested and its high sensitivity to fluorescence was demonstrated.

First, particles of a fluorescent glass (Luminass-1; Sumita Optical Glass) were used for the standard specimen. Particles with a few micrometer diameter were embedded in silicon oil on a quartz slide glass covered with a quartz glass slip and then observed. Under the fluorescence microscope, the spectra of emission light from the excited area with or without the particle were measured by a monochromator and a photomultiplier tube (PMT). The signal-to-background ratio (S/B) was calculated from these spectra (Fig. 1, Table 1). This specimen showed a significant advantage due to less photobleaching that permits a stable and long observation time.

The analysis of Auger and other excited electrons in the energy range from about 50eV to 2500eV provides a nearly non-destructive means to determine the surface composition of materials. It is both highly surface sensitive (a few atomic monolayers) and capable of high spatial resolution. This makes it an attractive alternative to energy dispersive x-ray analysis for the examination of ever smaller structures such as in semiconductor devices. For many applications however one important drawback has been the relatively long time required to make useful measurements - usually many seconds or minutes. Spectrum-imaging, in which an entire spectrum is acquired for each pixel in a digital scanned image, is one example for which Auger, spectroscopy becomes a prohibitively lengthy process. Similarly, the study of electron beam sensitive materials is difficult because the measurement time is often longer than the damage time. One method of reducing the measurement time is by acquiring the whole energy spectrum of interest in parallel. This is achieved with the Hyperbolic Field Analyser. (In electron energy loss spectroscopy this approach has lead to the highly successful Gatan energy filter.)

We applied phase-shifting digital holography to microscopy in this paper. At first lensless microscopy is proposed, in which no optical adjustment is necessary. Then, the method is applied to relax the limitation of focal depth in traditional optical microscopy. A theory for image formation and experimental verification using a few specimens are described.

keywords: microscopy, digital holography, phase shifting

Due to the finite focal depth of an imaging lens, a limitation to normal optical microscopy-is that, only the 2-dimensional (2-D) information of an object can be obtained at one time. Besides, it is not convenient for quantitative analysis the observed image. Optical sectioning microscopy (OSM) and scanning confocal microscopy (SCM) which use opto-electronic detection have been proposed for quantitative analysis of a 3-D object. However, the former requires critical mechanical adjustment, while the latter uses timeconsuming mechanical 3-D scanning. Holographic microscopy can solve these problems because it can record 3-D information at one time. But, the chemical processing of holograms and the mechanical focusing at the reconstructed images cause more or less trouble. A 3-D imaging technique without use of photographic recording called optical scanning holography has recently been reported. However, there are also some trouble owing to the twin-image noise.

A multiport-readout, frame-transfer charge coupled device (CCD) digital imaging system has been successfully constructed and tested for intermediate-high voltage electron microscopy (IVEM) applications up to 400 keV. The system employs a back-thinned CCD, made by MIT Lincoln laboratories, that comprises 2560 × 1960 pixels and a pixel size of 24 μm × 24 μm. With a frame transfer design, the imager fills nearly the entire usable area of a 100 mm-diameter silicon wafer (FIG. 1). In the current implementation, four of the eight on-chip readout ports are utilized in parallel each operating at a pixel rate of 1 or 2 MHz so that the entire CCD array can be read out in as short as ∼0.6 seconds. The frame-transfer readout functions as an electronic shutter which permits the rapid transfer of charges in the active pixels to four light-shielded buffers (FIG. 1) where the charges are readout and digitized while the active area of the CCD is integrating the next frame. At 2 MHz, charge transfer of 980 CCD rows will be completed in under 0.5 ms, which is much shorter than a typical exposure time of a few seconds. The camera head (FIG.2) and control electronics for CCD sensor were packaged by Photometries (Tuscon, Arizona). Two MaxVideo 200 image processing boards from Datacube (Peabody, MA) are used for high speed online imaging processing. The CCD sensor is electronically cooled to ∼ −40°C during operation.

For many years, we have been viewing stereo images created from a wide variety of microscopic sources. These images need to be easily viewed, shared and presented for stereo imagery to become more mainstream. We have been working for several years now to create affordable high resolution stereo display systems using the modern digital computers’ display capabilities. In the last couple of years, two methods have come into widespread use in our laboratory. One using shuttered goggles and the other using an active polarizing screen in front of the monitor. Both are synchronized with the display at refresh rates that allow for flicker free viewing. We use standard video display cards and shuttered glasses that cost under $300 (complete).

Unfortunately, the software and video cards turn over at such a fast rate that we have found our hardware “orphaned” several times.

One of the problems with remote imaging which purports to represent the real world (Tele-presence) is that the images are two-dimensional and the “real” world is three dimensional. When viewing many objects, the lack of depth perceptiorican be a serious deficiency. For example, one would want a micro-surgeon to have a good idea how deep to cut into tissue during a tele-presence operation. There is a wealth of three dimensional information in light, SEM, AFM, confocal, and computer reconstructed or simulated objects. To transfer that information faithfully over the networks using broadcast quality high definition video transmission technologies such as MPEG2 may require as much as 20 Mbps per channel. Furthermore, stereo imaging is particularly sensitive to any phase delays and jitter between the two eye channels. Control of these parameters requires advance quality of service features that current internet does not provide, but that may be available on high-performance intranets. All this may present a problem for routine use of the stereo facility .

The next generation of Internet, however, will be able to routinely provision much higher bandwidth and other quality of service, in general through protocols that will allow for not only full screen full resolution television but also the use of two channels (left and right) that can stay in fairly close synchrony.

A stacked CMOS active pixel image-sensor (APS) has been developed for detecting various kinds of charged particles and its noise performance has been measured and analyzed. The sensitivity for ions and electrons with keV energy level utilizes for ion microscopy such-as SIMS and electron microscopy, respectively.

Charge particles such as ions and electrons with kinetic energy of keV order are useful probes for surface analysis of material. A measurement system which yields two-dimensional image of charge particles is highly demanded. The conventional two-dimensional detection system is composed of a micro channel plate, a florescent plate which receives multiplied secondary electrons and generates a visible image, and a visible image sensor. However, its limited dynamic range and non-linearity in the ion-electron-to-photon conversion process make a quantitative measurement difficult. The proposed system using a stacked CMOS APS has several advantages over the conventional system such as high spatial resolution, no insensitive time, high S/N, wide dynamic range, nondestructive readout capability, high robustness, and low power consumption.

Many of the repetitive measurements required in the analysis of images in the semiconductor field can be replaced by partially or fully automated measurements. Automated measurements provide a number of benefits, including speed, repeatability, and removal of operator bias. The question of accuracy and precision in metrology applications is important to understand.

Recently we implemented a tool for automated analysis of resist bar images that was designed for use in acceptance testing-of lithographic stepper machines. The steppers are required to meet a defined set of criteria over the entire area of the wafer and through the focus range of the lens. Prior to the development of this package, the measurements were all done with a ruler and protractor on film.

The software identifies the boundary lines of the resist bar by locating a series of points along each edge and then fitting a line to those points. Edge identification is complicated by coating artifacts and by the variety in the images being analyzed. After identifying edge points along the base, side, and top of the resist bar, the software automatically generates information about the width, height, side wall angles, and various other measurements of the resist bar and stores the data in spread sheet format.

The development of high and intermediate voltage electron microscopes provides the ability to investigate the ultrastructure of thick biological samples. This produces a unique view of the 3-D interrelationships of cells and organelles. However, the analysis and display of this 3-D information has presented challenges, particularly to quantitative biologists. Major advances in adapting computed tomography (CT) to electron microscopy have provided the major means of 3-D analysis of cellular ultrastructure (1,2). In particular, tremendous gains in determining the structure of macromolecules have occurred with advances in electron microscopy CT (3). However, CT remains an arduous undertaking and one not well suited to analysis of the structural alterations within large populations of cells requiring many reconstructions. We have adapted an alternative approach, Tuned Aperture Computer Tomography (TACT®), for use with high and intermediate voltage electron microscopes. Although with EM specimens TACT® does not usually produce as high Z axis resolution as CT, the simplicity of the TACT® algorithms lend themselves to the rapid analysis of multiple samples. TACT® is already gaining acceptance in radiology (4, 5). Like CT, TACT® uses a series of images of the sample taken from different angular perspectives. However, unlike CT, for TACT® only a few projections are required and thus good reconstructions of beam sensitive material are possible. The quality of the reconstruction is determined by the quality of the input images, the total angular spread, and the individual tilt increments between images. Better images, a greater angular spread, and smaller angles between images will produce more accurate results.

Molecular microscopy is one of the most important structural approaches in cell biological investigations and can provide insight into complex biological questions that no other technique can provide. Currently, the technique typically requires the acquisition of very large numbers of transmission electron micrographs from frozen hydrated specimens using low dose techniques. The field is constrained by manual data acquisition methods that are slow, labor-intensive and result in a very low percentage of suitable images. We have developed a system, called Leginon, for automatically acquiring images from a transmission electron microscope. Our first prototype of this system demonstrated that we could acquire 1000 high magnification images per day from negatively stained catalase crystals. We have now extended this system to acquire low dose images of specimens embedded in vitreous ice.

Methods: Specimens were prepared on Quantifoil grids using techniques which have been described previously. The Leginon system uses a Philips CM200 TEM and a Gatan MSC CCD camera and is controlled by the emScope software library. The overall acquisition protocol requires (i) obtaining a low magnification image [660x] of a grid square from a Quantifoil grid (fig. 1a); (ii) automatically identifying holes containing ice of suitable thickness; (iii) acquiring an intermediate magnification image [6600x] of the identified hole (fig. 1b); (iv) identifying features of interest within the hole; (v) focusing at high magnification [38,000x] and finally (vi) acquiring a high magnification image (fig 1c,d).

Treatment of human congenital malformations and for loss of tissue or organ function is a serious clinical concern, frequently supported by harvesting autogenous tissue for reconstructive surgery. This paper summarizes an alternate approach through novel tissue engineering methods and describes cell-polymer composites that could be fundamental in developing advances for digit and joint replacement. The work utilizes cells from different sources transplanted on biodegradable polymer matrices to form structures with defined shape, morphological features and composition resembling those of human phalanges and intact joints.

Polyglycolic acids (PGA) fibers (Albany International, Mansfield, MA) 15 μm in diameter were used as a mesh support for cell attachment and matrix formation. For cartilage and tendon cells, PGA alone served as a flat surface without specific shape. For bone cells, PGA was copolymerized with poly-L-lactic acid (PLLA) (Polysciences, Warrington, PA) and shaped after a human phalanx molded from cadaveric distal and middle phalangeal bones. Polymers were sterilized with ethylene oxide before use.

Study of the microscopic structure of bone dates back to the earliest use of the optical microscope. Understanding the complexities of the structure of this material has roughly paralleled advances in microscopic technique. Relatively recently a new powerful microscopic technique, atomic force microscopy (AFM), has been invented. So it should not be surprising to expect a new step in our understanding of bone. There have been several recent publications applying AFM to bone and its constituents. None of these studies have examined the shape of naturally occurring mineralites in bone. The issue we addsgss in this work is the elucidation of the three dimensional structure of the apatite mineralites in bone. Chiefly, two methods have been used previously to ascertain this structure: transmission electron microscopy (TEM) and X-ray diffraction (XRD,). The major difficulty with TEM is that it is a projection technique. Each image only yields information about geometry in a single plane. Landis et al. showed how to circumvent this problem using tomographic TEM. AFM provides a corroborative and perhaps more gentle method of directly measuring the structure of very small bone mineralites. It is possible that the interaction energy of the electron beam with very small mineralites is sufficient to vaporize them. The problem with XRD is that the line broadening used can have contributions from sources not directly related to mineralite size. In particular, if a single mineralite contains more than one crystalline domain, then the line broadening yields a measure of the domain size, not the overall mineralite size.

In order to understand the systematic fabrication of hierarchical structures found in biominerals, the process nature uses in the biosynthesis of systems with complex morphology and precise internal molecular organization must be mimicked. The multi-component organic supramolecular system reported here is a kinetically controlled model that demonstrates the three stages of biomineralization: supramolecular pre-organization, interfacial molecular recognition, and cellular processing.

Three molecular components are used in this system: acetic acid (HOAc), benzene, and Kemp's Triacid (l,3,5-Trimethylcyclohexane-l,3,5-tricarboxylic acid or KTA). KTA and HOAc cocrystallize as hydrogen bonded aggregates of KTA dimers capped by four HOAc molecules. KTA and benzene co-crystallize with benzene acting as spacers between one dimensional hydrogen bonded rod-like structures of KTA. Growth of microcrystalline KTA/benzene was initiated by immersing single crystals of KTA/HOAc in 1 mL of benzene for 10-120 sec. The product was isolated on filter paper and mounted with carbon tape on an aluminum stub. The samples were coated with ∼10 nm of 60/40 Au/Pd prior to imaging at 15 kV in a DS-130/LaB6 SEM or at 5kV in a DS-130FE SEM.

The need to optimize adhesion between the femoral stem and the surrounding tissue has led to the development of plasma sprayed HA-coated implants as attachment veichles for hip prostheses. The strength of attachment of bone to the HA surface cannot be fully explained by mechanical bonding at the macroscopic or microscopic level; thus, molecular interactions must contribute significantly to bone bonding, facilitating adhesion at the nanometer level. Observations of apatite crystallites and cell adhesion proteins in the direct vicinity of coated implants indicates that these play a role in modifying the surface for osteoblast attachment and subsequent collagen deposition.

An understanding of the mechanism of “bone bonding” could lead to improvements in the design of HA coatings, leading to enhanced performance of orthopedic implants. Transmission electron microscopy (TEM) has been used to elucidate a sequence of early bone bonding events at the interface of PSHA-coated titanium implants.

The deposition of biological apatite and subsequent formation of a bone/HA bond seem to depend on the dissolution of the coating and the reprecipitation of carbonated apatite. Previous investigations in vitro have shown that the degree of dissolution and reprecipitation decreases as the coating crystallinity increases. Thus, coatings which had undergone heat treatment were used to look at the effect of solubility of the coating on mechanism of bone bonding.

The enamel of human teeth is composed of parallel rods (bundles) of long, closely-packed, ribbonlike hydroxyapatite (HAP) Ca10(PO4)6(OH)2 crystallites oriented normal to the tooth surface. Both the unique structure of the crystallites and the higher order arrangement of the crystallites are believed to be a consequence of interactions between proteins secreted by the enamel forming tissues and the growing crystallites. A better understanding of the precise molecular mechanism that controls enamel crystallite formation could potentially lead to both novel enamel repair strategies and crystal growth control systems. It has been previously reported that the unidirectional flow of calcium ions across a membrane can be used to form ribbon-like structures of a related form of calcium phosphate. We have created a similar model system which utilizes the unidirectional flow of calcium ions to generate ribbon-like HAP crystallites. Ultimately, this system will be modified so that enamel proteins can be introduced into the crystallization milieu to better characterize their role in enamel formation.

SEM and TEM samples were prepared using the experimental setup shown in Fig. 1. Several tubes containing a 50mM HAP solution dissolved in 0.1M HC1 were placed upside down into a 50mM MOPS pH buffering solution at pH 7.0.