Optical fiber probes are routinely used with optical spectrometers to allow measurements to be made on remotely located samples. In most of these systems, however, the optical fibers are used as non-imaging “light pipes” for the transmission of laser light, and luminescence or Raman signals to and from the sample. Thus, while these systems are suitable for remote spectroscopy, they are limited to single-point measurements. In a recent paper, we showed that a small-diameter (i.e., 350 μm) coherent optical fiber bundle can be combined with an AOTF-based imaging spectrometer for fluorescence and Raman spectral micro-imaging with increased flexibility in terms of sample positioning and in-situ capabilities. The previous paper described the operation of the fiber-optic microimaging probe and AOTF imaging system and showed preliminary Raman and fluorescence images for model compounds with 4 μm resolution. We have extended this work to include a discussion of the lateral and vertical spatial resolution of the fiber-optic microprobe in a non-contact proximity-focused configuration.

Two-photon fluorescence microscopy has become an important research tool in both biological and material sciences. The technique uses long wavelength, typically in the near IR, as the excitation light to obtain shorter wavelength fluorescence (e.g. visible light). Because of the low linear absorption coefficient of most biological and polymeric specimens, this technique allows deeper penetration of the excitation beam, achieving optical sectioning to a depth of 250μm or more into the specimen. As a result of the quadratic dependency of the two-photon induced fluorescence to the excitation intensity, the fluorescent emission and photobleaching are limited to the vicinity of focal spot. This capability of addressing a specimen’s 3D space allows exciting possibilities in biological researches, such as 3D photobleaching recovery experiment.

Two-photon confocal fluorescence microscopy is ideal for the study of thick biological and material specimen in 3D. For example, Figure 1 shows a three-dimensional isosurface rendered image of a vascular bundle from a maize stem.

Placing the suspect at the crime scene is an important element in the solution of a crime. A way to do this is to find textile fibers like those from the victim’s environment or the crime scene on the clothing of the suspect. Another way would be to discover fibers like the suspect’s clothing or environment on the victim or at the crime scene. When fibers exhibiting the same microscopic characteristics and optical properties are found in common between a known source and the suspect or victim, an evidentiary value is placed on that association. This value is dependent on many factors that include the type(s) of fibers found (natural or synthetic), the color (s) of the fibers, number of fibers found, the location of fibers on a victim, suspect and/or crime scene and the number of different fibers that “match,” especially if different fibers from the same item of clothing or fibers from more than one item of clothing are found.

Infrared, Raman, and fluorescence/luminescence microspectroscopy/microscopy in many instances seek to provide high sensitivity compositional and functional information that goes beyond mere elemental composition. This goal is shared by NEXAFS microscopy, in which Near Edge X-ray Absorption Fine Structure (NEXAFS) spectroscopy is employed to provide chemical sensitivity and can be relatively easily adopted in a scanning transmission x-ray microscope (STXM). In addition to compositional information, NEXAFS microscopy can exploit the dependence of x-ray absorption resonances on the bond orientation relative to the linearly polarized x rays (linear dichroism microscopy). For compositional analysis, NEXAFS microscopy is analogous to Electron Energy Loss Spectroscopy (EELS) in an electron microscope. However, when utilizing near edge spectral features, NEXAFS microscopy requires a considerable lower dose than EELS microscopy which makes it very suitable to studying radiation sensitive materials such as polymers. NEXAFS has shown to have excellent sensitivity to a wide range of moieties in polymers, including sensitivity to substitution isomerism.

DELTASCAN is a new light microscopy imaging system developed and built in the Department of Physics, Oxford, UK. [1] It is able to separate out the contrast seen in a ‘cross-polars’ image into three components, | sin δ | (a function of the optical retardation), φ (the orientation of a section of the optical indicatrix) and Io (the transmittance). These three variables are plotted as separate coded colour images. With the present computer and apparatus the data is collected, processed and the images simultaneously drawn in approximately 40 seconds.

DELTASCAN has an optical setup [2] based around a polarising microscope (Figure 5). The intensity through this optical setup can be shown to have the form:

I = Io[1 + sin2(ωt−ϕ)sin δ] (1)

where ω = frequency of the analyser, φ = orientation of cross section of the indicatrix and, δ = relative retardation which is related to a samples birefringence by δ = 2πΔnL/λ.

Several spectral bands or regions often are required for contrast generation to utilize the full power Raman imaging for mapping compositional or physical property variations. To achieve this, it is most efficient to use a high throughput imaging spectrograph with a two-dimensional detector operated in pushbroom mode as a line imager. Despite data from several bands, univariate data reduction techniques are inadequate when characterizing complex materials. Multivariate statistical procedures, such as factor analysis, are useful for defining weighted sums of intensities from many Raman shifts (frequencies) and this in turn expresses chemically or structurally useful contrast.

We illustrate this approach with silicate and aluminosilicate glasses. In most cases, no single band can unambiguously identify a particular glass composition. If the glass is heterogeneous, then localized compositional and structural differences may be detected by small changes in several bands, all arising from motions of assorted Si-O moieties.

Vibrational spectroscopy is used in Earth science for both quantitative and qualitative analysis. This report focuses on infrared (IR) spectroscopy, although similar efforts are on-going in Raman spectroscopy.

Qualitative studies utilize the fact that the vibrational spectrum is a characteristic of a material: hence comparison to a set of standards allows for identification of the phase. Most of these types of studies in Earth science involve macrosamples, but measurements of microsamples from meteorites are on interest in order to identify the structure of SiC inclusions and the type of organic compounds in interplanetary dust. As most of these samples are micron sized, which is below the diffraction limit for the mid-IR, the approach has been to compress the sample using a diamond anvil cell (DAC) into a disk of sub-micron thickness, adhere the sample to a KBr plate, and to subsequently remove the disk from the DAC and obtain spectra with the aid of an FTIR microscope.

Raman chemical imaging microscopy is a powerful technique for the characterization of a wide host of materials, including inorganic species. The technique makes use of a liquid crystal tunable filter (LCTF) imaging spectrometer that is integrated within an infinity-corrected optical microscope. The imaging system provides the performance of a dispersive Raman spectrometer at every pixel of the charge-coupled device (CCD) detector used to capture the Raman image.

We are currently applying Raman microscopy to the chemical imaging analysis of mineral composition and phase chemistry found in meteorites and terrestrial minerals. Determination of the chemical composition and structure of mineral components is often useful in developing an understanding of the petrologic process that formed the minerals, including those minerals that have been exposed to water. This is of particular interest in light of recent promising evidence of past life in Martian meteorites.

Zeolites are complex, three-dimensional, hydrated crystalline aluminosilicates that have porous structures with channels or cages of various dimensions. In the simplest way they can be represented by (Na2,K,Ca,Ba) [(Al,Si)O2]nxH2O. They may be either of natural or synthetic origin. As high-surface-area materials, they take on great technological importance, foremost in surface science and catalysis. The structure of the zeolite is critical to its function. Structure information is commonly obtained by x-ray and neutron diffraction, NMR, IR, and Raman spectroscopy techniques. Synthetic zeolites are produced in enormous quantities worldwide and are key to critical technologies. Yet, no real zeolite standards exit that are defined and certified. NIST, with input from industry and academia, has begun a measurement program to certify various physico-chemical properties for a suite of synthetic zeolite powder standard reference materials (SRMs) and research materials (RMs). These proposed "standard" zeolite materials span a range of pore sizes, SiO2/Al2O3 ratios, ring sizes, structural building units, and cage sizes.

A new infrared spectroscopic imaging technique has been described, combining step-scan Fourier transform (FT) Michelson interferometry with indium antimonide (InSb) focal-plane array (FPA) image detection [1-3], for use in the range 3950-1975 cm−1. The coupling of such detector to an interferometer provides an optimized method for infrared spectroscopic imaging by simultaneously realizing both a multiplex and multichannel advantage. Specifically, the multiple detector elements enable spectra at all pixels to be collected simultaneously, while the interferometer allows all the spectral frequencies to be measured concurrently. This technique can rapidly generate very high quality, chemically specific images from a wide variety of samples. Preliminary results from the use of a mid-infrared liquid-helium-cooled arsenic-doped-silicon array have also been reported recently [4]. Focal-plane arrays using mercury-cadmium-telluride (MCT) are now commercially available, giving access to the range 3950-800 cm−1, and greatly broadening the applicability of this technique [5].

FT-IR microscopy does not require exotic sample preparation -- samples can be examined in transmission and reflectance, do not need to be stained or coated, and a vacuum is not required.

Vibrational spectroscopic imaging has already demonstrated enormous potential for studying a variety of chemical and biological systems at both the microscopic and macroscopic level. However, these spectral images are large and complicated, typically consisting of tens of thousands of pixels, each with an associated high-resolution vibrational spectrum, leading to data sizes upwards of 64 megabytes. In order to realize the full potential of these spectral images, we must find ways to query the data so that specific questions can be answered.

We illustrate a multivariate approach to this challenge, where each pixel is considered to be a single point in a multivariate (N-dimensional) space. The variables (coordinates) of the point in N dimensions are simply the intensities of the N-point spectrum associated with the pixel. In this representation, pixels with similar spectra will tend to cluster together in the multivariate space, since they will have similar coordinates.

In today’s laboratory environment, it is important that the performance of analytical instruments be tested and validated. In the past, the testing of infrared micro- spectrometers was based solely upon measuring a signal-to-noise (SNR) ratio of the 100% line. This test is insufficient. It is important to test how well the sample area is defined. In all current commercial infrared spectrometer systems, remote image plane masks are used to define specific areas for infrared spectral analysis. A special test specimen has been developed to evaluate the accuracy of the mask’s ability to define the sample area.

IMS resolution test plates were constructed by photolithographic reproduction of a series geometrically shaped thin polymer films on both infrared transparent windows and gold mirrors. First the surface of the plate is coated with a thin, uniform layer of a photoresist polymer. Then this surface is printed with geometrical patterns such as bands, discs and squares.

Analysis of nanosize particles by x-ray emission spectrometry in the analytical electon microscope (AEM) is limited by the level of electron current that can be placed in a 1-2 nm excitation beam and by the consequent electron beam damage. The current in the beam limits the number of x-ray counts generated in the particle, while the beam damage may cause changes in particle chemistry, structure, orientation, and location. The objective of “ultimate AEM analysis” is to make useful measurements right up to the limit where beam damage prohibits analysis.

Elemental analysis of sub-10nm metal particles supported on a ceramic support requires a field-emission source operating at 100-300 kV to generate a significant number of x-ray counts from the area under the beam in a reasonable time [1]. The high current density in the beam of such an instrument can cause damage to the particle or to the underlying support material.

Perhaps the ultimate aim of analytical electron microscopy in biology is to detect single atoms or ions bound to isolated macromolecular assemblies and small cellular organelles rapidly frozen in their native state. Although the meaningful spatial resolution in such analyses is limited to ∽10 nm or more by radiation damage, the high intrinsic sensitivity of electron energy-loss spectroscopy (EELS) coupled with recent developments seems to make this rather ambitious goal—originally proposed some twenty years ago—within reach. Here we describe examples where EELS has been able to detect surprisingly small numbers of atoms in biological specimens and discuss some fundamental limits that are encountered as well as some possible strategies for circumventing these difficulties.

In our laboratory we have used a scanning transmission electron microscope (STEM) equipped with a field-emission source and parallel-detection EELS because the probe size and collection efficiency of this instrument are optimized to give the lowest detectable number of atoms and lowest detectable atomic fraction.

A probable future development in materials science is the selective manipulation of structure on an atom by atom basis. Along with the ability to manipulate atoms we will need to be able to tell if the correct ordering has been achieved. Thus the ultimate goal of spectroscopic imaging will be to create a species-labeled two or three dimensional map of the atom positions in the sample. In this paper we will show how far we have progressed towards this goal with Auger electron spectroscopy (AES), how much further we have to go and what the prospects are for overcoming the known obstacles.

AES is inherently surface-sensitive because low energy electrons have very small inelastic mean free paths. This scattering mechanism insures that only those electrons created within a few atomic diameters of the sample surface facing the detector will contribute significantly to the measured signal.

The field of semiconductor materials and devices represents a unique challenge to microscopy and microanalysis because the standard of excellence is constantly changing. Driven by the imperative of Moore's law (illustrated in figure 1) - which predicts that the density of devices in a chip doubles every three years - the spatial resolution that must be achieved increases by a factor of two every five years. The problem of meeting this demand is further enhanced by the widespread sentiment that all e-beam tools used on semiconductors should operate at low beam energies (typically below lkeV). Finally, the increase in device packing density and the expected move to 30cm. diameter wafers as substrates for fabrication also means that the microscopist will be faced with the challenge of rapidly and reliably finding a given device or feature among the 108 or so similar features which are present.

The first need is to be able to monitor the basic steps of the fabrication process. Given the assumption that by the year 2001 devices will be built on 0.18μm design rules, it will then be necessary to perform critical dimension metrology with a precision of 2 to 3nm (see figure 2).

The National Technology Roadmap of Semiconductors (NTRS) shows that future semiconductor processing will require the formation of junctions less than 30nm deep with gate lengths approaching, and becoming less than, 0.lum. Techniques must be available to measure these in-depth, as well as spatial, distributions.

Secondary ion mass spectrometry (SIMS) has long been the technique of choice for measuring dopant profiles in semiconductors. The depth of the junctions formed has, up top now, always been sufficiently deep into the semiconductor that the effects of the analysis on the measured profile (i.e. atomic mixing) could be ignored for the most part. In addition, gate lengths have been sufficiently large that diffusion of LDD implants under the gate have not contributed significantly to short-channel effects. However, the relentless decrease in the required junction depths and gate lengths may mean that the situation may change in the near future.

The technique of electron backscatter diffraction (EBSD) in the scanning electron microscope is currently finding a large number of important applications in materials science. The patterns formed through EBSD were first studied over 40 years ago. It has only been in the last 10 years that the technique has really begun to have an impact on the study of materials. The introduction of automatic pattern indexing software has enabled the technique to be used for mapping the orientation of a polycrystalline sample. The more exciting and universally interesting application of the technique has been the identification of micron and sub-micron sized crystalline phases based on their chemistry and crystallography determined by EBSD.

EBSD is obtained by illuminating a highly tilted sample (>45° from horizontal) with a stationary electron beam. Electrons backscattered from the sample may satisfy the condition for channeling and will produce images that contain bands of increased and decreased intensity that are equivalent to electron channeling patterns.

Throughout the history of electron-beam X-ray microanalysis, analysts have made good use of the strong dependence of electron range on incident energy (R ≈ E1,7) to optimize the analytical volume when attacking certain types of problems, such as inclusions in a matrix or layered specimens. The “conventional” energy range for quantitative electron beam X-ray microanalysis can be thought of as beginning at 10 keV and extending to the upper limit of the accelerating potential, typically 30 - 50 keV depending on the instrument. The lower limit of 10 keV is selected because this is the lowest incident beam energy for which there is a satisfactory analytical X-ray peak excited from the K-, L-, or M- shells (in a few cases, two shells are simultaneously excited, e.g., Fe-K and Fe-L) for every element in the Periodic Table that is accessible to X-ray spectrometry, beginning with Be (Ek =116 eV) and extending to the transuranic elements. This criterion is based upon establishing a minimum overvoltage U = E0/Ec > 1.25, which is the practical minimum for useful excitation.

Conventional quantitative X-ray microanalysis in the scanning electron microscope or in the electron microprobe is valid for specimens of bulk homogeneous composition and with flat and polished surfaces. Quantitative methods, using X-ray microanalysis and Monte Carlo simulations of electron trajectories in solids, have been developed for the chemical analysis of spherical inclusions embedded in a matrix and for multilayered specimens. In this paper, the effect of porosity and of the size of the pores are investigated concerning their effect on X-ray emission using Monte Carlo simulation of electron trajectories in solids since porous materials are of great technological importance.

This new Monte Carlo program uses elastic Mott cross-sections to compute electron trajectories and the Joy & Luo modification of the continuous Bethe law of energy loss and the details are given elsewhere. This program assumes that all the pores are spherical and have the same size.