The environmental scanning electron microscope (ESEM) employs a series of pressure limiting apertures and a differential pumping system to allow for electron imaging at specimen chamber pressures of up to 50 torr. Images rich in secondary electron (SE) contrast can be obtained using the gaseous secondary electron detector (GSED) or ion current (Iion) signals. The GSED and Iion signals are amplified in a gas cascade. SEs emitted from a sample are accelerated through the gas in the specimen chamber by an electric field, EGSED, produced by a positively biased electrode located in the chamber, above the specimen. The accelerated SEs give rise to a cascade ionization process that can amplify the SE signal by up to three orders of magnitude. Electrons produced in the cascade are rapidly swept to the biased electrode and are efficiently removed from the gas. Positive ions produced in the cascade drift away from the electrode with a velocity that is at least three orders of magnitude lower than that of the electrons.

Electron detection in the environmental scanning electron microscope relies upon the presence of a small pressure of a gaseous phase inside the microscope chamber. The presence of this gas controls two important mechanisms by which the microscope functions. The conventional gaseous secondary electron detector (GSED) applies a variable positive bias (0 to +600 V) directly above the specimen. Electrons ejected from the specimen surface due to the incident scanning probe are accelerated by this field towards the positively biased detector. Whilst traversing the gap between specimen and detector the electrons undergo ionizing collisions with gas molecules. These result in an amplification of the incident electron signal through the production of ejected ‘daughter’ electrons and leave behind positively charged ions. After production these ions are repelled by the detector bias and drift back towards the microscope stage where they aid charge neutralization on the surface of the specimen.

The ability of Environmental Scanning Electron Microscopy (ESEM) to provide electron images of volatile and/or insulating samples without need for sample preparation allows imaging of liquid surface at resolutions substantially beyond those obtainable with optical microscopy.

The system chosen for study was a dispersion of di-iso-heptyl-phthalate (DHP) in water, stabilised with a mixed surfactant system comprising equal masses of Tween 20 (Polyoxyethylene sorbitan monolaurate) and Span 20 (Sorbitan monolaurate) at an effective HLB of 12.75. DHP was chosen for its unsaturated electronic structure, which improves secondary electron contrast in the ESEM, and for the similarity of its density (995 kgm-3) to water. Emulsions, containing between 25 and 80 vol% DHP and 0-3 wt% surfactant were produced in a Silverson mixer fitted with an emulsor screen. Samples were placed in a brass holder on a Peltier cooling stage and cooled to 2°C.

The ESEM uses a novel differential pumping system combined with a pressure-limiting aperture to achieve relatively high pressures of up to 10 torr in the specimen chamber. Using this system allows samples to be imaged in the presence of water vapour or other gases. Conveniently the saturated water pressure at 275K lies within this range, allowing the stabilisation of hydrated samples, condensation from water vapour and evaporation to be performed in situ within the microscope chamber.

Wetting of surfaces is of fundamental importance in many different areas of science, from textiles to petroleum engineering. The ESEM now provides an opening for high-resolution studies of these phenomena without the problematic depth-of-field limitations that optical methods present. The contact angle, θ, is defined as the angle between the tangent to the liquid-fluid interface and the tangent to the solid interface. It provides a ready indication to the wetting properties of a system.

The acid etching process is the primary stage for preparation of dental composite restorations. Dentin specimens from extracted teeth are usually prepared by fracturing etched pieces of dentin. Etching creates a demineralized layer on the dentin surface with the thickness of a few microns. The layer consists of wet collagen and dissolved proteins; its drying process is of special interest in dentistry since it affects the bonding of the resin adhesives to dentin. Observations of fractured surfaces can be performed in either a conventional or an environmental SEM.

Three wet dentin samples were observed (XL30 FEG ESEM, FEI/Philips) in wet mode at 5°C Peltier cooling stage temperature and 5.8 MPa water vapor pressure. Samples with excess water on their top surface were put in an ESEM. Thickness measurements of the demineralized zone were begun immediately after the evaporation of this water layer with subsequent measurements taken after 5, 10, 20, 40, 60 and 100 minutes of drying.

It has previously been shown that liquid microstructures can be observed, in their natural state, by use of Environmental SEM (ESEM). In order to take full advantage of the ability to observe the dynamics of complex fluids, efforts are now being directed at developing in situ experimental methodologies. For example, we have recently shown that the coalescence of oil droplets in emulsion systems can be observed in real time. The purpose of the present study is to tackle the challenge of making EM observations of complex fluids under electrical stress (and ultimately electrical & mechanical stress combined) by subjecting particle dispersions to an electric field.

The technique of attracting charged particles to one of two electrodes (electrophoresis) is commonly used as a means of determining the nature and density of charges on such particles. Another technique, using uncharged particle systems, involves applying an electric field across a sample as well as, for example, a shear strain (electro-rheology).

We have investigated the affect of accelerating voltage (E0), pressure (P), GSED bias (VGSED), field line geometry and specimen conductivity on the landing energy of the primary beam in an XL30 ESEM. The landing energy of the primary beam can be measured by using the Duane-Hunt cut-off energy, EDH, in EDS X-ray spectra. We present experimental evidence that the landing energy of the primary beam can be significantly altered when analyzing insulators in an ESEM. Shifts in EDH result from electric fields produced by the presence of positive ions in the ESEM chamber and electrons trapped in the bulk. No shift in EDH was observed with conductive specimens at all E0 and P. Changes in the landing energy of the primary beam can significantly effect the relative intensity of characteristic peaks in EDS spectra measured in the ESEM (Figure 1).

The magnitude and sign of the shift in EDH is determined by E0 and P for a specific insulator specimen and stage geometry.

It is of paramount importance to know the total elastic-scattering cross sections of the gases which are present in the specimen chamber of the variable pressure scanning electron microscope (VP-SEM) or in the environmental SEM (ESEM) since these values are needed to compute the width of the beam broadening skirt by the use of Monte Carlo simulations. However, these values are not well known for even the most common gases because the vibrational modes of the molecules add another contribution to the usual value of the elastic scattering cross-section. To overcome this difficulty a method has been proposed to measure total elastic cross-sections which is based on a measurement on the unscattered beam current as a function of the chamber pressure using a Faraday cup. The total elastic cross-sections are then deduced using the well-known exponential decay law.

Standard Reference Material (SRM) 482 of the National Institute of Standards and Technology is a set of 6 gold/copper wires, ranging in concentration from 0 to 100% Cu in 20% steps, intended for calibration studies of electron beam microanalyzers. This is an appropriate standard to test the accuracy of energy dispersive x-ray spectrometry (EDS) in the Environmental Scanning Electron Microscope (ESEM). While the presence of the gas in the sample chamber gives the ESEM its unique capabilities, it also is the source of complications to x-ray spectrometry. The gas can spread the primary electron beam into a wide skirt of electrons with the consequent production of x-rays many micrometers from the target location of the beam.

The six wires (∼ 500 jam in diameter) were embedded and polished in one epoxy mount. The mount was carbon coated in one set of experiments. The coating was removed and the sample retested.

Recently, a new correction procedure has been proposed in order to perform X-Ray microanalysis in the ESEM or in the VP-SEM1. This new correction procedure is based on this equation:

where I is the measured intensity at a given pressure P, Ip is the intensity that would be generated without any gas in the microscope (the corrected intensity) and Im is the intensity with complete scattering of the electron beam. Im is therefore the contribution of the skirt on I. In equation (1), fp is the fraction of the incident beam, which is not scattered by the gas above the specimen, and it can be obtained from Monte Carlo simulations or from an analytical equation.

In low vacuum scanning electron microscopes, the primary beam is partially scattered by the gas present in the specimen chamber. The development of these microscopes, in particular the so-called ‘Environmental’ SEM, was initiated when it was realized that this scattering does not necessarily compromise the imaging capabilities of the instrument. Indeed, some modern commercial instruments are capable of better than 2 nanometer resolution at gas pressures of several torr. The accepted explanation for this is as follows: The mean-free-path of the high energy primary electrons is several millimeters in one torr of water vapour (for example). Because the actual pathlength of electrons travelling through the gas is only a few millimeters, most of them do not scatter at all. Those that do scatter are supposedly distributed over a relatively large area. Thus, the probe features a high-intensity central region surrounded by a slowly decaying low-intensity skirt. High resolution imaging is possible because the signal-to-(skirt)background ratio is high.

With recent advances in computing power and application specific software it has become practical for smaller microcopy labs to either purchase or build a electron backscatter diffraction (EBSD) systems. This system can be used to probe the crystallography and microstructure of many different materials. One set of such materials is geological samples from deep within the earth's crust. In collaboration with the Geology and Geophysics department an EBSD system has been built to function with the Philips Electroscan E3 Environmental Scanning Electron Microscope (ESEM). Using an existing low light level SIT camera, the system was completed with the addition of a phosphor screen and appropriate software and acquisition cards.

One main advantage of the ESEM for EBSD is that since it is an environmental SEM samples can be imaged without coating. For insulating materials this is means that the sample can be imaged without coating.

In this work, phosphor imaging plate technology is applied to measure electron scattering directly in the environmental scanning electron microscope (ESEM) specimen chamber. The scattering of electrons from the primary electron beam, under relatively high-pressure conditions (266 Pa) in the ESEM sample chamber, degrades the analytical accuracy of elemental analysis. The degree of this degradation is poorly known. To date, attempts to measure experimentally the spatial distribution of the scattered electrons have been limited to observing secondary effects such as the intensity of x-rays produced from copper targets positioned at various distances from the primary electron beam interaction point. A more accurate distribution of the scattered electron intensity can be obtained from a direct measurement of both the scattered and unscattered electrons over a large area with single electron sensitivity. Improvements to the accuracy of Monte Carlo models of the scattering process will be made possible by the direct measurement data.

Multi-photon excitation has been used in microscopy for nearly a decade, providing a number of demonstrated advantages over other methods for fluorescence imaging. Because excitation is achieved using longer, less energetic light, photodamage and photobleaching of the sample are reduced. Furthermore, since excitation occurs only at the focal point, this approach allows the practical collection of three-dimensionally resolved fluorescence images of live cells. However, due to the small two-photon cross-section of most fluorophores, pulsed lasers are required to generate detectable signal levels. This is due to the quadratic dependence of twophoton absorption on the instantaneous power of the laser. Typically, these lasers are pulsed at very high repetition frequencies, on the order of 106 pulses per second with pulse durations of a few hundreds of femtoseconds. For example, a titanium:sapphire (Ti:S) laser mode-locked at 76 Mhz can provide up to 100,000 watts of instantaneous power and is ideal for exciting two-photon events.

The ability to observe the fluorescence arising from single molecules has revolutionized our ability to study the structure and dynamics of materials on a microscopic level and to probe the properties of individual members of a heterogeneous ensemble. A variety of near-field and far-field excitation techniques have been employed to study single molecules. Multiphoton excitation (MPE) techniques have a number of advantages that make them particularly attractive for singlemolecule detection. First, because the excitation and fluorescence wavelengths are significantly different from one another, Rayleigh and Raman scattering can easily be filtered out, leading to a low number of background counts. Second, because the probability for MPE depends on the excitation intensity to the second or higher power, the excitation is localized to the point in space where the excitation beam is most tightly focussed, thus providing three-dimensional resolution.

Two-photon excitation has enabled investigators to image living cells in highly scattering media like the central nervous system (1). We have used a custom-built two-photon microscope to image dendritic spines from living cortical pyramidal neurons. Pyramidal cells form the majority of the neuron in the mammalian cortex and they receive practically all their synaptic contacts through dendritic spines. Dendritic spines are small (<1 μm diameter) appendages that have been practically inaccessible to physiological measurements until the application of two-photon excitation to their study (2). We have concentrated in two questions:

A- Calcium compartmentalization of spines: Mechanisms of calcium decay kinetics.

Dendritic spines can compartmentalize calcium (2). Although the mechanisms of calcium influx into spines have been explored (3), it is unknown what determines the calcium decay kinetics in spines. We investigate calcium dynamics in spines from rat CA1 pyramidal neurons in slices.

Multi-photon fluorescence microscopy has been cited for its advantage in increased depth penetration due to low linear absorption coefficient of biological specimen in the near infrared (NIR) range. Using a pulsed laser, it is possible to efficiently excite two-photon fluorescence with a high peak power while keeping the average power low to avoid thermal and photochemical damages to the specimen. Currently, mode-locked Ti-sapphire and Cr-Forsterite lasers that generate sub-picosecond pulses are used as the light source for multi-photon fluorescence microscopy. Because of the need of high peak power for efficiently exciting two-photon fluorescence, the relationship between cell damage and peak power has become an interesting and much debated topic in the application of multi-photon fluorescence microscopy. It is conceivable that at high illumination intensity, non-linear photochemical processes may have impacts on cell physiology and viability in ways much different from low illumination in the linear domain.

Considering its non-linear nature, two-photon excitation may generate very different spectral response in samples when compared with single photon excitation. It is thus necessary to measure the two-photon spectra of samples, so that the two-photon fluorescence microscopic images can be properly interpreted. Fluorescence spectra obtained from bulk samples may not provide useful information for microscopy. For instance, due to the relatively small contribution to the total fluorescence intensity, a small number of fluorescent particles in a generally fluorescing specimen may escape detection when the spectrum of the specimen as a whole is obtained. Under two-photon excitation, the background noise can be greatly reduced due to the naturally limited excitation volume of focused laser beam. In addition, signals resulted from second harmonic generation (SHG) may be mixed with low level broad-band background autofluorescence which is commonly found in biological specimen. Therefore, measuring fluorescence spectrum from a micro-focused volume is essential for the proper interpretation of multi-photon fluorescence images.

Confocal microscopy is an excellent high resolution method to image fluorescently labeled cells. However, the use of confocal microscopy to monitor physiological events, such as membrane potential changes, in living cells is hampered by photobleaching and phototoxicity. To reduce the collateral damage from excitation of fluorescent probes outside the optical slice, Webb and co-workers introduced the use of two-photon excited (TPE) fluorescence in laser scanning microscopy.1 Two-photon absorption depends on the square of the incident light intensity; this has the effect of confining excitation to the plane of focus where the photon flux density is greatest. The wavelength of the exciting light is in the near infrared facilitating penetration of thick tissues. Due to these significant advantages this methodology is rapidly gaining popularity as a tool for live cell and tissue imaging.

To further exploit non-linear optical processes in laser scanning microscopy, we have developed surface second harmonic generation (SHG) as a powerful new imaging modality.

Tightly focused beams of ultrashort optical pulses such as produced with a high numerical aperture optic as found in a microscope exhibit an interesting property: a strong third harmonic signal is produced when the axial focal symmetry is broken through a change in material properties. This change can be manifested as either a change in refractive index and/or a change in the nonlinear susceptibility of the specimen. Conversely, third harmonic production is inhibited in those cases when the axial focal symmetry is unbroken. This property can be effectively exploited to produce volumetric images of biological and non-biological specimens. The third harmonic is localized to interfaces within the specimen making it possible to produce optical sections in much the same manner as in confocal, and two-photon fluorescence microscopy. Unlike these techniques however, no exogenous fluorophore is necessary for labeling - the signal is produced by endogenous interfaces.