OBSERVATION OF INDIVIDUAL MOLECULES USING AN ELECTRON MICROPROJECTOR

The problem of increasing the resolving power of electron-optical instruments and of the possibility of proceeding to direct observation of molecules and recognition of their shapes has long confronted electron microscopy.

At present, with the resolving power of electron microscopes being several tens of angstroms at best, it is possible to detect individual large organic molecules consisting of thousands of atoms, but not to judge their shape. Individual molecules were first recognized with the aid of the so-called electron projector, an instrument that made it possible, in a number of successful experiments, to obtain, by direct magnification of an object a million or more times, a resolution exceeding the limits technically attainable in the best electron microscopes with magnetic or electrostatic lenses.

Fig. 1. Schematic of an electron microprojector.

The electron projector in its simplest form consists of a sealed glass vessel evacuated to the highest possible vacuum—of the order of \(10^{-8}\) mm of mercury. Inside the vessel there is a metallic (usually tungsten or molybdenum) point with a very small radius of curvature and an auxiliary electrode, often in the form of a ring (Fig. 1). A fluorescent screen is deposited on the walls of the vessel. The point serves as the cathode; a positive potential is applied to the auxiliary electrode. When the gradient of the electric field at the point reaches a value of the order of \(10^7\) V/cm, electrons begin to be emitted from the metal, flying along radii that emerge, practically, from a single point, and bombard the fluorescent screen. If there is an obstacle in the path of the electrons, its shadow appears on the screen.

image. The greatest magnification is achieved if the object lies on the surface of the tip; in the first approximation it is equal to \(\frac{R}{r_0}\), where \(r_0\) is the radius of curvature of the tip, and \(R\) is the distance from the tip to the screen. The radius of curvature of a tungsten tip, produced by electrolytic etching in a weak solution of alkali with alternating current and subsequent ionic bombardment, in some experiments reached 180 and even 110 angstroms. At the same time, in the case of sufficiently pure metal, the end of the tip is a single crystal. The end of such a tip is invisible when observed in a light microscope. The radius of curvature is determined by observation in an electron microscope or from the relation between the cold-emission current and the applied potential difference.

Beginning approximately in 1937, a number of processes were studied with the aid of electron projectors: the regularities of cold electron emission, adsorption phenomena, the migration of atoms over the surface of the cathode, and the deposition of atoms on single crystals during cathode sputtering.

The resolving power of the electron microprojector depends on the accelerating voltage, the radius of curvature of the tip, and the work function of the metal from which the tip is made. The blurring of the image of a point object in the case of a spherical tip is ultimately determined by the tangential components of the velocities of the electrons as they leave the metal into vacuum. Fig. 2 shows the dependence of the diameter of the blur circle corresponding to a point object on the surface of the tip on the accelerating voltage and the work function. It is understood that, when working with the projector, one has to maintain the accelerating voltage such that the emission current produces a sufficiently bright glow of the screen. As is evident from the curves in Fig. 2, in the case of low voltages, i.e., for tips with radii on the order of hundreds of angstroms, the resolving power of the projector exceeds the limit practically attainable for an electron microscope.

Müller^3 succeeded in showing that if a certain number of phthalocyanine molecules are deposited by evaporation on the tungsten tip of a microprojector, these molecules—apparently owing to their semiconducting properties (see, for example, ^4)—produce a local enhancement of electron emission. On the characteristic geometrical figure on the projector screen, corresponding to maxima and minima of emission from different crystallographic directions of tungsten, much smaller bright spots are superimposed. Upon more detailed examination, each of the spots proves to consist of four segments (Fig. 3). Meanwhile, according to chemical and X-ray data, the phthalocyanine molecule is something like a “flower” with four “petals” formed by benzene rings (Fig. 4). The coincidence of the shadow images with the shape of the molecule, whose cross-section reaches approximately 14 angstroms, confirms the fact that in the photograph of Fig. 3 we are dealing with an image of individual molecules.

One cannot, of course, underestimate the difficulties of the method used in comparison with work with an electron microscope. The electron projector cannot be adjusted for changing the object under investigation; because of insufficiently high vacuum, dismountable systems cannot give great resolving power. In addition, the molecules under study must be sufficiently stable, i.e., not decompose at high temperatures.

Naturally, the question arises of the next step—whether, by decreasing the radius of curvature of the projector tip, it is possible to see still finer details of molecular structure. For this it is necessary that

Fig. 2. Dependence of the resolving power of the electron projector on the accelerating voltage and on the work function of the tip metal.

Fig. 3. Copper phthalocyanine molecules. Shadow image on the screen of an electron microprojector.

Fig. 4. Forms of the copper phthalocyanine molecule, according to chemical data. Black circles are carbon atoms, white circles are nitrogen atoms. Hydrogen atoms in the benzene rings are not shown.

the electron-emitting spherical surface had a radius of curvature of the order of 20–30 angstroms. The diameter of the blur circle would then be about 5 angstroms. However, points with \(r_0\) less than 150–200 angstroms usually no longer have a sufficiently regular shape. Sometimes their shape can be corrected by evaporating the point with a tungsten wire heated at very low sputtering intensity. Another, more important obstacle is that the extent of the potential barrier through which electrons are torn out by the field in cold emission is of the same order of magnitude as the details of the structure of the molecule (interatomic distances).

V. S. Vavilov

CITED LITERATURE

1. B. M. Tsarev, “The electron projector as a method of physico-chemical investigations,” UFN 36, 181 (1948).
2. E. Müller, Zeits. f. Physik 120, 277 (1943).
3. E. Müller, Discovery 9, 231 (1950).
4. A. T. Vartanyan, Zhurn. Fiz. Khimii 22, 763 (1948).