M. Drechsler’s research while affiliated with French National Centre for Scientific Research and other places

It is well known that a metal tip blunts if it is heated in vacuum. Sharpening may occur in the case of an evaporation of the tip material. To test this, molybdenum tips were heated in vacuum. A typical result: at 2400K, the tip radius decreases from 8 mu m to 0.3 mu m thus confirming the hypothesis. Measured and calculated final radii agree roughly. The pure material evaporation can be replaced by a surface reaction followed by the evaporation of the reaction products. This is shown on tungsten tips heated (1750K) in the presence of oxygen (0.5 mTorr). The evaporation of tungsten oxides results in a sharpening to a radius to 0.05 mu m. Under special conditions, in connection with the formation of solid drops, radii down to 0.01 mu m are obtained.

For controlled heating of metal microcrystals up to the melting point, the end of the field emitter tips are bombarded by an electron beam (8 keV, 400 mu A, 200 mu m beam diameter). To measure local temperatures on small areas down to 2 mu m diameter a special one-wavelength micropyrometer was built. Using an optical microscope, an image of the glowing tip was projected on to a screen outside the vacuum chamber. The screen has a movable hole behind which a one-wavelength filter and a photomultiplier are placed to measure local radiant intensities. The relative temperatures are obtained from Wien's law, and are absolutely calibrated by a thermocouple spot-welded on the emitter shank. Local relative temperatures and temperature gradients for a tungsten emitter are measured in the range between 2800K and 3650K, with an error and a stability better than 10K. For absolute temperatures the error is 50K when no evaporation occurs.

In the course of in-situ transmission electron microscope studies, we found and analyzed a phenomenon of self-formation of small evacuated chambers. The chamber wall is partly metal and partly graphite film. The chamber formation is explained as an interface diffusion transport induced by capillarity. The chamber is resistant against atmospheric pressure. Such chambers can be used to study kinetic phenomena by in-situ electron microscopy.

Une étude des cous stables des pointes en microscopie électronique transmission montre l'existence d'une oscillation des pointes avec des amplitudes de ~ 200 Å et des fréquences de ~ 0.3 Hertz (Cu, Au, 900°C). L'oscillation est expliquée par une interaction de deux forces : (1) une force électro-statique causée par l'impact des électrons et annihilée par une décharge d'émission de champ et (2) une force élastique d'une couche graphitique à la surface de la pointe. A detailed study of stable tip necks by in situ transmission electron microscopy shows an existence of tip oscillations with amplitudes in the order of 200 Å and frequencies in the order of 0.3 Hertz (Cu, Au at ~ 900°C). The oscillation is explained by an interaction of two forces : (1) an electrostatic force caused by electron impact and annihilated by a field emission discharge and (2) an elastic force of a graphitized layer on the tip surface.

On a metal tip (W, Ni, Au, Cu) heated in an electron microscope (SEM) a formation of stable necks is observed. As this does not agree with the theory, experiments are made for clarification. In situ electron microscopy (TEM, STEM, EELS) shows the existence of graphitized surface layers, which surprisingly remain stable up to near the melting point. These layers hinder a complete separation of a solid metal drop from the tip end. On the basis of this result spectacular tip and neck shape changes are explainable as an Ostwald ripening.

A technique is described to study in situ the matter transport and the morphological evolution on crystals. Used for it is a transmission electron microscope, specimens in form of tips, a heating chamber and video registration. The test system studied is Cu-C. Results found are: (1) techniques to measure in situ the thickness of a layer on a metal surface; (2) an indication that a layer formation begins on surface regions of maximum curvature; and (3) the finding of an oscillating matter transport on tip necks.

Former studies of grain boundaries in metal tips concern static crystallographic problems. In this study we describe temperature increase induced evolutions of grain boundaries in very clean tips (Ni, W, Ta). Results of scanning electron microscope studies are: (1) The initial boundaries migrate to form a system of parallel boundaries. (2) A bamboo-structure boundary migrates towards the tip end but stops at the tip end neck. (3) Measurable boundary grooves are formed. Results of field electron microscope studies are: (1) A migration of the intersection line of the boundary and the surface is directly visible on the screen. (2) Twin boundaries are visible and analysable. (3) The tip end crystal can rotate at the grain boundary until the boundary disappears. It is discussed why such type of experiments are promising for future studies of fundamental recrystallization problems.