Focused ion beam (FIB) microscopes, the use of which is well established in the semiconductor industry, are rapidly gaining attention in the field of materials science, both as a tool for producing site specific, parallel sided TEM specimens and as a stand alone specimen preparation and imaging tool.

Both FIB secondary ion images (FIB SII) and FIB secondary electron images (FIB SEI) contain novel crystallographic and chemical information. The ability to see “orientation contrast” in FIB SEI and to a lesser extent SII is well known for cubic materials and more recently stress-free FIB sectioning combined with FIB imaging have been shown to reveal evidence of plastic deformation in metallic specimens. Particularly in hexagonal metals, FIB orientation contrast is sometimes reduced or eliminated by the FIB sectioning process. We have successfully employed FIB gas assisted etching during FIB sectioning using XeF2 for zirconium alloys and Cl2 for zinc coatings on steels to retain orientation contrast during subsequent imaging.

Gallium focused ion beam (FIB) milling is the method of choice to prepare cross sections of selected features from microelectronic devices for transmission electron microscope (TEM) imaging and analysis. The FIB milling technique is unsurpassed in producing an ultra-thin cross section accurately located through the feature of interest. While much effort has been invested in the development and refinement of Ga+ FIB techniques and instrumentation, there are problems due to the physics of the ion-solid interaction. The problem of surface amorphization limits the quality of the TEM samples and its significance increases as the feature size and the specimen thickness decrease.

In Ga+ FIB milling of silicon, the amorphous damage layer consists of amorphous Si doped with implanted Ga. This damage layer is caused by ions that strike the surface of the silicon and are subsequently scattered laterally as a result of collisions with the Si atoms.

Automated specimen preparation for transmission electron microscopy has the obvious advantage of saving personnel time. While some people may perform labor intensive specimen preparation techniques quickly, automated specimen preparation performed in a timely and reproducible fashion can significantly improve the throughput of specimens in an industrial laboratory. The advent of focused ion beam workstations for the preparation of electron transparent membranes has revolutionized TEM specimen preparation. The FIB lift-out technique is a powerful specimen preparation method. However, there are instances where the “traditional” FIB method of specimen preparation may be more suitable. The traditional FIB method requires that specimens must be prepared so that the area of interest is as thin as possible (preferably less than 50 μm) prior to FIB milling. Automating the initial specimen preparation for brittle materials (e.g., Si wafers) may be performed using the combination of cleaving and sawing techniques as described below.

Focused Ion Beam (FIB) micromachining techniques have gained significant attention over the past few years as a promising method for the preparation of a variety of metallic and nonmetallic materials for subsequent characterization using transmission electron microscopy (TEM) The advantage of the FIB in terms of site specificity and speed for the preparation of uniform electron transparent sections has opened a wide range of potential applications in materials characterization. The ability to image the sample in the FIB can also provide important microstructural data for materials analysis. In this study, both conventionally electropolished and FIB-ed specimens were prepared in order to characterize the microstructure of a commercially-produced tube of Alloy 600 (approximately Ni-15 Cr-10 Fe- 0.05 C). The electropolished samples were prepared using a solution of 20% HClO4 - 80% CH3OH at ∼-40°C. The FIB sections were obtained from a cross-section of the tube that had been mechanically thinned to ∼100 μm. The section was thinned in a Micrion 2500 FIB system with a Ga ion beam at 50 kV accelerating voltage.

The continued decrease in microelectronic feature dimensions has led to a reliance on the focused ion beam (FIB) for site-specific transmission electron microscopy (TEM) specimen preparation. To maximize the capabilities of the FIB, methods must be developed to consistently produce specimens thin enough to generate TEM lattice images. The limiting factor in producing quality TEM specimens by either the traditional or lift-out method is the final thickness of the specimen.

The FIB is used to prepare TEM specimens by removing the bulk material that surrounds a desired feature by sputtering with a focused gallium ion beam. Successively lower beam currents are used to sputter away material until an electron transparent membrane (-0.2 μm) containing the desired feature remains. For a 300 keV TEM, lattice imaging of silicon requires additional membrane thinning to less than 0.05 μm.

The loss of rigidity during the thinning process makes the membrane highly prone to deformation due to residual stresses, linear expansion, and ion beam interaction.

Reduced feature dimensions in microelectronic devices has led to a growing reliance on the focused ion beam (FIB) for scanning electron microscopy (SEM) and transmission electron microscopy (TEM) specimen preparation. The introduction of copper for use in electrical interconnects will increase this reliance. Copper, much more so than aluminum, tends to smear when conventional polishing techniques are employed, rendering mechanical polishing unsuitable for quality specimen preparation.

The sputtering characteristics of copper differ greatly from the aluminum based materials currently in use. Copper exhibits a grain orientation dependent sputtering rate that tends to produce uneven FIB polished surfaces. The sputtering rate for copper is also much higher than the silicon and dielectric materials that must be simultaneously removed. These differences in sputtering rates lead to the formation of curtains and ledges that also affect specimen quality.

Another FIB induced artifact was observed when preparing SEM specimens (Figure 3).

Energy Dispersive Spectrometry (EDS) is generally calibrated for quantification using elemental standards. This can introduce errors when quantifying non-elemental samples and does not provide an accurate detection limit. In addition, variations between analysis tools can lead to values that differ considerably, especially for trace elements. By creating a standard with an exact trace composition, many of the errors inherent in EDS quantification measurements can be eliminated.

The standards are created by high dose ion implantation. For ions implanted into silicon, a dose of 1E16 cm-2 results in a peak concentration of approximately 1E21 cm-3 or 2% atomic. The exact concentration can be determined using other methods, such as Rutherford Backscattering Spectrometry (RBS) or Secondary Ion Mass Spectrometry (SIMS). For this study, SIMS analyses were made using a CAMECA IMS-6f magnetic sector. Measurement protocols were used that were developed for high concentration measurements, such as B and P in borophosphosilicate glass (BPSG).

Micromachined electronic devices present new challenges for established analytical tools in the semiconductor industry. These types of devices have moving components with spacing tolerances in the submicron domain. These spaces can be accumulation sites for mobile particles within the hermetically sealed device. Particles can then cause electrical malfunctions both random - for freely moving particles - and permanent - for entrapped particles. The problems for the failure analyst are: (1) a valid electrical test, (2) confirmation of the failure site without generating particles during the analysis; (3) the ability to locate and analyze known failure sites after the device is opened. A combination of several analytical tools are required for such tasks.

Surface micromachined devices are made by alternating layers of polysilicon and a sacrificial layer, such as oxide. To make a mechanical part from the deposited layer material, an underlying sacrificial layer is dissolved, thus freeing the element except where it is retained by an attachment to the silicon surface.

Microstructural characterization of thin alumina scales formed on oxidized Ni-base alloys using transmission electron microscopy (TEM) has long been a challenge as a result of the many problems encountered during the preparation of thin specimens. Successful and reproducible preparation of uniformly thin, cross-section TEM specimens from these multicomponent “layered” structures (alumina scale on a metallic substrate) is extremely difficult using standard ion beam thinning procedures for several reasons: (1) differential thinning of the various constituents in the system can occur, (2) a weak ceramic-metal interface may lead to separation during the harsh mechanical grinding and thinning procedure, (3) fully intact scales are rarely achieved since native surface structures are usually lost during ion milling, and (4) extensive damage can be created in the scale and alloy during rough grinding and polishing steps as well as during final ion beam thinning.

The cardiac interstitium is a diverse system of extracellular-matrix (ECM) components organized into a stress tolerant three-dimensional network that interconnects the cellular components of the heart. During each contraction of the heart, a complex set of mechanical forces is transmitted through the cardiac interstitium and is applied to the cardiac myocytes and fibroblasts. In comparison to the relaxed heart, each contractile wave results in a rapid reorganization of the extracellular matrix and cell cytoskeleton. A number of studies have shown that by varying the normal pattern of applied mechanical force that occurs during heart contraction, both the phenotype and function of myocytes and fibroblasts in the heart may be altered.

At the molecular level components of the ECM are attached to the cardiac myocytes and fibroblasts via specific transmembrane integrin receptors. Each integrin is comprised of an extracellular domain that attaches to an ECM component, and an intracellular domain that attaches to a cytoplasmic component of the cell.

The wall of saccular brain aneurysms, lesions that develop at the fork regions of human brain arteries, is a layered multidirectional fabric of fibrous collagen. The wall tissue has the mechanical features of high elastic stiffness and a low tensile strength compared to adjacent arteries. Arterial blood pressure within the sac of an aneurysm stresses the wall in all tangential directions; thus an area of mechanical weakness may be characterized by an area lacking fibre strength, or an area with inadequately aligned fibres.

The collagen fibres of the wall are birefringent, with many fibre types, similar to wound healing skin (types I, III, IV, V, VI) The known correlation between fibre birefringence and tissue strength for dermal wound healing3 provided the opportunity in this research to calculate strength maps to compare one lesion from another, and to make regional comparisons around a single lesion.

Cardiovascular disease is the leading cause of morbidity and mortality in diabetic individuals. Diabetes mellitus is associated with a two- to four-fold increase in risk of coronary heart disease. This risk is independent of other classical risk factors including age, smoking, elevated serum cholesterol levels, and hypertension, and the overall risk of cardiovascular disease increases proportionally with the number of risk factors. Many groups are currently investigating the aetiology of diabetic vascular disease, with particular interesting being paid to the processes involved in the formation of highly cross-linked carbohydrate-protein complexes known as melanoids. These complexes appear to be formed by glycation arising from reactions between glucose and the amino groups on proteins, and their formation within the vessel wall resqlts in a decrease in vascular elasticity. This decrease in elasticity has been implicated in the increased incidence of atherosclerosis in diabetic individuals. Previous testing modalities could not test in the radial direction and required assumptions based on incompressibility in order to determine the radial behavior of the artery.

We are developing nondestructive and non-traumatic imaging methods to quantitatively assess structural and mechanical properties of vascular trees and other complex biological systems. We are applying these methods to studies of vascular remodeling in rodent models of disease. Quantitative vascular tree morphometry has, in the past, been carried out using corrosion casting methods which supply valuable information, but are destructive and tend to be tedious, in particular with regard to associating vessel segment measurements accurately with their positions in the intact tree hierarchy. One of our goals is to refine imaging methods capable of quantifying mechanical, as well as anatomical properties, by virtue of their ability to rapidly acquire multiple image volumes of the same structure under various conditions, all in a nearphysiological state.

Using the micro-CT scanner shown in Figure 1, consisting of a microfocal x-ray tube, specimen stage and image intensifier coupled to a 10242, 12-bit CCD camera, we acquire magnified (3-9X), transmitted x-ray projections of contrast enhanced, excised rat lungs or lung lobes.

Fibrin clot formation is necessary for maintaining the integrity of the vasculature via the physiological processes of hemostasis and wound healing and is also involved in pathological processes, such as thrombosis and atherosclerosis. Fibrinolysis, or the dissolution of the clot, is necessary to remove clots so that they do not become obstructive. There must be a dynamic balance between clot formation and fibrinolysis to prevent either bleeding or thrombosis. Although the biochemical process of fibrinolysis is well known, less has been known about the physical process of lysis. In these studies, the surface of digested clots were examined by scanning electron microscopy, the products released were visualized by transmission electron microscopy and the process of lysis was followed in real time by confocal microscopy.

Fibrin is degraded by the fibrinolytic system, in which a plasminogen activator converts plasminogen to plasmin, a serine protease that cleaves specific bonds in fibrin leading to solubilization.

Heart attack and stroke remain a major cause of death and morbidity in the United States. In most cases, these are caused by the formation of an occlusive thrombus overlying an atherosclerotic lesion. Lysis of these occlusive clots by the infusion of lytic agents is now an approved treatment for heart attack patients. However, lytic therapy is not always successful. The reasons for the failures are not clear. Resistance to thrombolysis is multifaceted and involves the key elements of the clot; fibrin, platelets, and endothelial cells. In previous studies we have shown that, in vitro, secretion of plasminogen activator inhibitor-1 (PAI-1) by platelets and endothelial cells is a factor influencing resistance. However, PAI-1 secretion could not account for all of the lysis delay.

For the past 7 years we have been measuring atherosclerosis in murine models of the human disease, constantly honing our methods for quantitating the atherosclerotic response of mice that have been genetically, dietarily or pharmaceutically manipulated to help understand and control the disease process.

With the advent of genetically modified mouse models, investigators now have more robust atherosclerotic responses to quantitate. The surface area of lesions along the entire length of the aorta is a key parameter used to evaluate these responses. We routinely perform quantitative morphometry of intimal atherosclerotic lesions in pinned-out mouse aortas, also called en face lesion analysis.

The number and scale of mouse atherosclerosis studies has increased beyond the capacity of our core laboratory to provide service. While the “wet” processes involved in preparing vascular tissue for atherosclerosis analysis have already been streamlined, we found considerable scope for improving the computer-dependent “back-end” of the analysis.

In the leg venous valves isolate the blood into short columns and thus relieve the vessel wall from the gravitational force of blood columns that would otherwise be several feet in length. Class 6 chronic venous insufficiency is associated with abnormal venous hemodynamics and associated tissue-related abnormalities termed lipodermatosclerosis, accompanied by ulceration. Ingenious surgical techniques have focused on correction of the abnormal hemodynamics by repair, reconstruction, or transplantation of venous valves to the deep venous system of the lower extremity with varying success rates. We have applied free scapular fasciocutaneous flaps to successfully reconstruct recalcitrant grade 6 venous insufficiency ulcers in the lower extremity. We have noticed such ulcers frequently in medial malleolar, lateral malleolar or dorsum pedis but never in the region of the gastrocnemius or peroneus muscles. This fact prompted us to examine the microvenous valvular anatomy in segments of human leg.

Abstract

Hemodynamics and 3D- architecture of the vascular network of the ventral tail fin of tadpoles of the South African Clawed Toad, Xenopus laevis Daudin were studied by intravital video microscopy (IVM) (stage 56) and vascular corrosion casts (VCCs)(stage 58). IVM enabled a detailed study of the circulation in segmental arteries, segmental venules and interposed capillaries. SEM of VCCs and 3-D morphometry allowed exact measurements of vessel diameters.

Anuran tadpoles are good models to study blood vessel growth (=angiogenesis) (1-3) as well as blood vessel regression which occurs physiologically when tadpoles loose larval specific organs such as gills and tail during metamorphosis (4,5). The translucent ventral tail fin enables in-vivo observations of the living circulation by intravital light microscopy (1,2) and allows correlations between vessel fine structure and dynamics of blood flow under various conditions (2).

This study focuses on the 3D-architecture of the blood vascular system in the ventral tail fin of tadpoles of Xenopus laevis Daudin in stage 58, i.e. at the begin of metamorphic climax.

The swordfish ﹛Xiphias gladius), as well as billfishes (Istiophoridae), shows a conspicuous modification of one of its eye muscles: the dorsal rectus muscle. While the other eye muscles show an apparently normal, oculomotoric structure, the musculus rectus dorsalis is modified for heat production. While the remaining oculomotoric portion of this muscle shows normal muscular cells, the bulk of the organ is composed of cells structurally similar to brown adipose tissue (BAT) and blood vessels. The BAT cells are capable of direct (i.e., non-shivering) metabolic heat production. This heat is then transported and disseminated by the blood vessels in this organ.

Microvascular corrosion casting is an excellent technique for studying the morphology of complicate vascular networks. However, studies using this technique on eye muscle vasculature are not frequently found in the literature. Juvenile swordfish, 1.4-1.8kg bodyweight, were collected in the Ionic sea. Immediately after catching, terminal anaesthesia was carried out with tricaine-methanesulfonate.

Vascular corrosion casts provide faithful replicas of the three-dimensional anatomy of blood vessel system of organs and tissues. In addition, corrosion casts can be used to obtain quantitative information regarding vessel and tissue spaces. However, morphometric measurements of SEM images of corrosion casts is difficult due to severe specimen tilt, parallax and the inability to image 3D casts from all angles. We present evidence that confocal microscopy can be used to collect all of the 3D information of a corrosion cast in a z-series of optical slices. Surface and volumetric parameters of casted and non-casted space are readily retrievable from a stack of optical sections through a corrosion cast.

Various tissues and organs were exsanguinated with heparenized saline following cannulization of regional arteries and casts were made by infusing Mercox resin or Mercox diluted with 20% methyl methacrylate monomer. Tissues were macerated with alternating rinses in 5% KOH and hot water.