Introduction

Mikhail Ya. Schelev, Mikhail A. Monastyrskiy, Nikolai S. Vorobiev, Sergei V. Garnov and Dmitriy E. Greenfield

According to the brilliant historic review by Lincoln L. Endelman, an outstanding member of high-speed photographic society (Endelman, 1988), the very beginning of high-speed photography may be associated with an experiment by William Henry Fox Talbot in 1851, who took a wet-plate photograph of a few square inches of a page from the London Times newspaper attached to a rotating wheel and illuminated by a spark from Leyden jars.

In the late 1860s, after the technique of Talbot’s experiment had been in many ways perfected and directed toward different applications, including military ones, Alfred A. Pollock and Sir John F. W. Herschel suggested that it would be possible to take a series of instantaneous pictures of different moving objects (men, animals, etc.) using a camera with exposure times as short as 1/10 of a second. In 1872, the outstanding professional photographer Eadweard James Muybridge (Figure 1(a)) was invited by California governor Leland Stanford to resolve the popularly debated question of the day—whether all four feet of a horse were simultaneously off the ground when the horse was trotting and galloping. By using a number of large glass-plate cameras positioned in a line along the track, each camera’s shutter triggering by a thread when a horse was passing, Muybridge obtained a series of consequent images displaying the dynamics of the horse’s motion (Figure 1(b); Haas, 1976; http://en.wikipedia.org/wiki/Eadweard_Muybridge).

Needless to say, the importance of this remarkable result goes far beyond merely setting a dispute, or indicating to painters and illustrators their mistake in imaging the galloping horse with its legs extended to the front and back. (Muybridge’s photographs showed that the horse’s legs, while off the ground, are collected beneath its body as it switches from “pulling” with the front legs to “pushing” with the back legs (Leslie, 2001)). Indeed, considering the most up-to-date knowledge, we should always remember that Muybridge's “moving pictures” and the Zoopraxiscope—the device invented later by Muybridge to demonstrate his “movie” on a screen (Muybridge, 1887)—were the first bricks laid into the foundation of high-speed photography and its applications.

An improved version of Muybridge’s camera was used in 1882 by Dr. Étienne-Jules Marey (Figure 2(a)), a French scientist, physiologist, and chronophotographer, to study animal locomotion (Figure 2(b); Braun, 1994; http://en.wikipedia.org/wiki/Etienne-Jules_Marey).

Marey invented his own electric photographic gun that was capable of producing 12 consecutive frames per second with an exposure time of 1/720 s, and carried out a number of first-class research projects associated with the dynamics of moving objects and substances (Marey, 1878). This is the reason why Marey, like Muybridge, is widely recognized as a pioneer of high-speed photography.

Nowadays, the time resolution of opto-mechanical high-speed photography has practically reached its physical limit: ∼10−9 s, determined by the mechanical strength of the rotating parts in those devices (Shnirman, 1959; Reichenbach, 1993; Schelev, 2003). In most cases, the opto-mechanical cameras are now replaced by high-speed videography based on the use of computer-controlled charge-coupled devices (CCDs) or complementary metal–oxide–semiconductor (CMOS) active pixel sensors (see the review by Balch, 1999).

The period between the end of the 1940s and the beginning of the 1950s was marked by the emergence of a new and promising trend in technical physics—high-speed electron-optical photography. A reliable basis for this trend had been formed by great achievements in theoretical electron optics during the 1930s and 1940s (Brüche and Scherzer, 1934; Brüche and Recknagel, 1941; Artsimovich, 1944; Glaser, 1952) and exceedingly successful experimental developments in the field of electron-optical image intensifiers and their application to night vision. After the pioneering works by G. Holst, J. H. de Boer, and P. T. Farnsworth with coworkers, carried out by the end of the 1930s (Holst et al., 1934; Farnsworth, 1930), a wide variety of single-stage and multistage electron-optical image intensifiers operating in infrared and X-ray spectral ranges was designed in Germany, England, France, the United States, and Russia.

After World War II, during which image intensifiers had mainly served as night-vision devices for the military, they began to be used in high-speed photography as ultrafast electronic shutters. In 1949, using a pulsing power source, M. P. Vanyukov (Figure 3(a)) at Leningrad State Optical Institute designed a millisecond framing camera (Vanyukov and Nilov, 1954), while J. S. Courtney-Pratt (Figure 3(b)), in his experiments in streaking the photoelectron slit images by means of a rapidly changing magnetic field, was able to record the photoelectron images of individual phases of an exploding substance with nanosecond time resolution (Courtney-Pratt, 1949, 1952, 1953; 1973).

A real breakthrough in high-speed imaging, which happened a few years later, was the introduction of oscilloscope-type deflection plates inside the streak tube volume. Indeed, Courtney-Pratt’s magnetic streak tube was limited in speed considerably by the value of magnetic coil inductance. Nevertheless, nobody before had attempted to use an electric deflecting field for streak purposes—skeptics even argued that any time-dependent deflecting electric field would inevitably destroy the streaked electron image. M. Butslov (Figure 3(c)), thanks to the encouragement and practical help of Academician E. Zavoisky (Figure 4(c)) and his talented pupil and colleague S. Fanchenko, was able to design his famous UMI-95 streak camera, in which electric deflection plates were used to sweep the electron image across the screen for the first time (Butslov 1958, 1959).

The UMI-95 image sweep technique was very simple: a synchroscan-type, elliptically shaped sweep system was comprised of two mutually perpendicular pairs of deflection plates fed by the sinusoidal electric fields with a phase shift. A 300-MHz, 100-W oscillator providing maximum elliptical deflection speed up to 2 × 109 cm/s was employed. Owing to the multistage amplification of brightness (105–106), every single photoelectron leaving the input photocathode could be recorded.

The exciting experiment undertaken by E. Zavoisky and his coworkers was the evaluation of UMI-95 ultimate time resolution. It was decided to study the minimal-size electric sparks in high-pressure nitrogen (Zavoisky and Fanchenko, 1955). The discharged circuit-ringing frequency was about 1010–1011 Hz. The experimental scheme was as follows: an optical lens projects the spark image onto the UMI-95 input photocathode. When the oscillator is off, the electron image at the output phosphor screen of the streak tube represents a small bright spot. When the oscillator is on, the output image is travelling over an ellipse. The total time of the spark light emission was shown to be about 200–400 ps, with the duration of separate light pulses being less than 10 ps. In 1961, the same authors (see details in Fanchenko, 1961) created a streak image tube of the UMI-95V...