5.4 nm spatial resolution in biological photoemission electron microscopy.

We report a spatial resolution of 5.4 nm in images of sarcoplasmic reticulum from rabbit muscle. The images were obtained in an aberration-corrected photoemission electron microscope with a hyperbolic mirror as the correcting element for spherical and chromatic aberration. In-situ measurements and numerical simulations confirm the low residual aberration in the instrument and indicate the ultimate resolution in this type of microscopy to be below 2 nm.

Photoelectron imaging of cells: photoconductivity extends the range of applicability.

Photoelectron imaging is a sensitive surface technique in which photons are used to excite electron emission. This novel method has been applied successfully in studies of relatively flat cultured cells, viruses, and protein-DNA complexes. However, rounded-up cell types such as tumor cells frequently are more difficult to image. By comparing photoelectron images of uncoated and metal-coated MCF-7 human breast carcinoma cells, it is shown that the problem is specimen charging rather than a fundamental limitation of the electron imaging process. This is confirmed by emission current measurements on uncoated monolayers of MCF-7 carcinoma cells and flatter, normal Wi-38 fibroblasts. We report here that sample charging in photoelectron microscopy can be eliminated in most specimens by simultaneous use of two light sources--the standard UV excitation source (e.g., 254 nm) and a longer wavelength light source (e.g., 325 nm). The reduction in sample charging results largely from enhanced photoconduction in the bulk sample and greatly extends the range of cells that can be examined by photoelectron imaging. The contributions of photoconductivity, the electric field of the imaging system, and the short escape depths of the photoelectrons combine to make photoelectron imaging a uniquely sensitive technique for the study of biological surfaces.

Low-energy electron microscopy (LEEM) and mirror electron microscopy (MEM) of biological specimens: preliminary results with a novel beam separating system.

Low-energy electron microscopy (LEEM) and mirror electron microscopy (MEM) utilize a parallel beam of slow-moving electrons backscattered from the specimen surface to form an image. If the electrons strike the surface an LEEM image is produced and if they are turned back just before reaching the surface an MEM image results. The applications thus far have been in surface physics. In the present study, applications of LEEM and MEM in the biological sciences are discussed. The preliminary results demonstrate the feasibility of forming images of uncoated cultured cells and cellular components using electrons in the threshold region (i.e. 0-10 V). The results also constitute a successful test of a novel beam-separating system for LEEM and MEM.

Emission microscopy and related techniques: resolution in photoelectron microscopy, low energy electron microscopy and mirror electron microscopy.

A unified treatment of the resolution of three closely related techniques is presented: emission electron microscopy (particularly photoelectron microscopy, PEM), low energy electron microscopy (LEEM), and mirror electron microscopy (MEM). The resolution calculation is based on the intensity distribution in the image plane for an object of finite size rather than for a point source. The calculations take into account the spherical and chromatic aberrations of the accelerating field and of the objective lens. Intensity distributions for a range of energies in the electron beam are obtained by adding the single-energy distributions weighted according to the energy distribution function. The diffraction error is taken into account separately. A working resolution is calculated that includes the practical requirement for a finite exposure time, and hence a finite non-zero current in the image. The expressions for the aberration coefficients are the same in PEM and LEEM. The calculated aberrations in MEM are somewhat smaller than for PEM and LEEM. The resolution of PEM is calculated to be about 50 A, assuming conventional UV excitation sources, which provide current densities at the specimen of 5 x 10(-5) A/cm2 and emission energies ranging up to 0.5 eV. A resolution of about 70 A has been demonstrated experimentally. The emission current density at the specimen is higher in LEEM and MEM because an electron gun is used in place of a UV source. For a current density of 5 x 10(-4) A/cm2 and the same electron optical parameters as for PEM, the resolution is calculated to be 27 A for LEEM and 21 A for MEM.

Design and performance of a high-resolution photoelectron microscope.

The design of a high-resolution photoelectron microscope (photoelectron emission microscope) is described. It is an oil-free ultrahigh-vacuum instrument utilizing electrostatic electron optics. New designs are presented for a specimen translator, cathode stage, aperture stop control, electrostatic hexapole stigmator, beam shutter, and camera system. These components could also be used in a low-energy electron microscope (LEEM). The theoretical resolution of this instrument is 5 nm for UV illumination near the photoemission threshold. The photoelectron microscope is now in operation at the University of Oregon, and it is achieving results within a factor of two of this design limit.

The resolution of photoelectron microscopes with UV, X-ray, and synchrotron excitation sources.

The resolution of emission electron microscopes is calculated by determining the intensity distribution in the image. The object is a small disc of uniform brightness centered on the axis. A finite object, as distinct form a point source, provides a non-zero current in the image without the requirement of infinite object brightness and the consequent infinities in the geometrical intensity distribution. The minimum object size, which in turn affects the resolution of the microscope, depends on the minimum current or contrast required in the image. In photoelectron microscopes with UV illumination just above the threshold for emission the predominant aberrations are the chromatic and spherical aberrations of the accelerating field and the spherical aberration of the objective lens. For higher energies, e.g. in the soft X-ray range, the chromatic aberration of the objective lens must also be taken into account, as the aberration coefficients of the accelerating field are greatly reduced. The intensity distributions in the image are calculated first for single energies. The intensity distribution for a beam with a range of energies is obtained by adding a series of single-energy distribution curves weighted according to the energy distribution function. In the presence of spherical aberration the position of the image formed by the electrons depends on the angle of emission. In image planes between the paraxial and marginal planes the combination of spherical aberration and defocus causes the the image spot to have a retrograde type of behavior as the angle of emission increases. The image spot initially moves away from the axis in the azimuth of emission and then returns to the axis and moves away in the opposite azimuth. As a result the intensity in the central portion of the image plane is enhanced. The single-energy intensity distribution curves calculated as a function of depth in the image reveal the existence of a compact, high-intensity image peak in an image plane located between the paraxial and marginal planes. This peak occurs in the plane in which the image spot has a maximum retrograde displacement equal to its radius. The present analysis shows that the resolution in the high-intensity plane is better than in the plane of least confusion, and the effects of aberrations in these two planes are quite different.(ABSTRACT TRUNCATED AT 400 WORDS)

Photoelectron imaging and photoelectron labeling.

Photoelectron imaging involves the photoejection of low-energy electrons from a specimen surface exposed to UV light. The electrons are then accelerated and focused by an electron-optics system in much the same way fluorescent light is focused in an optical microscope. Thus, photoelectron imaging is the electron-optical analog of fluorescence microscopy. In combination with photoemissive labels it serves to extend the range of studies possible by fluorescence, for example in work on cell surfaces and internal structures of cells that have been exposed by detergent extraction of membranes.

Depth of information in photoelectron microscopy.

The depth of information is defined as the distance below the surface of a specimen from which information is contributed at a specific resolution. A simplified model of photoemission is used to explore the relationship between electron escape depths and depth of information in photoelectron microscopy (PEM or photoemission electron microscopy). The depth of information is equal to the escape depth when the escape depth is small relative to the instrument resolution. When the escape depth is large compared to the instrument resolution or when information is carried for example by reflected light, the image consists of well resolved surface detail at the instrument resolution and dimmer, more diffuse, images of detail below the surface. Thus the same sample can exhibit different depths of information depending on the image details of interest. Other mechanisms of transmitting information to the surface, for example induced topography, are discussed, and experimental examples are given.

Photoelectron microscopy of cell surface topography.

Photoelectron micrographs of gold-palladium coated mouse 3T3 cells and chick embryo fibroblasts are presented. Since the gold-palladium suppresses differences in work function, the cell morphology seen in these micrographs is due to relief contrast. The heights of comparable cells were measured from the parallax present in transmission electron micrograph stereo-pairs of cell surface replicas. The origin of relief contrast and the effect of cell surface relief on the working depth of field in photoelectron images of cells are discussed. The micrographs demonstrate that photoelectron microscopy is very sensitive to fine details of cell surface topography, and that the working depth of field is not a limiting factor in the imaging of well-spread tissue culture cells.

A high vacuum photoelectron microscope for the study of biological specimens.

A photoelectron microscope (photoemission electron microscope) has been designed and built for the study of organic and biological samples. The microscope is an oil-free stainless steel high vacuum instrument pumped by a titanium sublimation pump, an ion pump, and molecular sieve roughing pumps. The electron lenses are of the electrostatic unipotential type. The microscope is equipped with a dewar for sample cooling, an internal cryogenic camera, TV-image intensifier, and vibration isolation support. Applications include studies of biological cell surfaces, photosynthetic membranes and aromatic chemical carcinogens. A representative micrograph of mouse 3T3 cells is included. In some respects, photoelectron micrographs resemble scanning electron micrographs, but the basis for contrast is different in these two techniques.

Photoelectron microscopy: a new approach to mapping organic and biological surfaces.

A general method of imaging organic and biological surfaces based on the photoelectric effect is reported. For the experiments, a photoelectron emission microscope was constructed. It is an ultrahigh vacuum instrument using electrostatic electron lenses, microchannel plate image intensifier, cold stage, hydrogen excitation source, and magnesium fluoride optics. The organic surfaces examined were grid patterns of acridine orange, fluorescein, and benzo(a)pyrene on a Butvar surface. A biological sample, sectioned rat epididymis, was also imaged by the new photoelectron microscope. Good contrast was obtained in these initial low magnification experiments. These data demonstrate the feasibility of mapping biological surfaces according to differences in ionization potentials of exposed molecules. A number of technical difficulties, such as the intensity of the excitation source, must be solved before high resolution experiments are practical. However, it is probable that this approach can be useful, even at low magnifications, in determination of the properties of organic and biological surfaces.