Energy-dispersive X-ray fluorescence (EDXRF) spectrometry is a well-established technique with high sensitivity for a wide range of elements. Recently, portable or handy EDXRF spectrometers have been developed for art, 1-4 archaeological, 5-7 environmental, [8][9][10] and forensic analyses. Radioisotopes, small air-cooled X-ray tubes, or pyroelectric X-ray emitters are used. [11][12][13][14][15][16][17][18][19] Terasawa found that the charge-up of an insulator produced X-rays. 20 He invented an X-ray device, as shown in Fig. 1, 21 where applying a high voltage on an insulator produced strong X-rays. Then, Kawai et al. 22 found that a vacuum of between 0.03 and 0.04 Torr was the best for X-ray emission. Brownridge 23 earlier proposed a pyroelectric insulator to produce X-rays, and many interesting phenomena were found for pyroelectric crystals. 24 In the present work, the X-ray generator proposed by Terasawa 20 was made and an independent verification of Terasawa is conducted. We measured X-ray spectra from the X-ray emitter and observed visible-light emission.Its correlation with the X-ray intensity was considered.
ExperimentalA compact cylinder-type vacuum tube, a schematic illustration of which is shown in Fig. 2, with 20 mm radius and 50 mm height was made. A photo is shown in Fig. 3, left. The cylinder body was polypropylene, made of a medical disposal syringe. The top and bottom were made of brass. Teflon as an insulator was fixed between the needle-shaped copper electrode and a plate electrode with a spring. The cylinder was roughly evacuated (about 6 Pa = 0.04 Torr) by a rotary pump and a high voltage was applied from 1500 to 2900 V by a DC power supply. The bottom brass was grounded, and to a feedthrough electrode was applied a minus high voltage. In the present paper we assign a high voltage, such as 1500 V, but this means that the feedthrough was applied -1500 V, and the bottom brass was grounded. We also tried that a high voltage was applied between the top brass and the feedthrough; this way of producing an electric field also produced high intensity X-rays. The vacuum of the cylinder was adjusted by a valve between the rotary pump and the X-ray cylinder, to be 6 Pa, which was suitable for the strongest X-ray emission, according to Ref. 22. The X-rays were emitted through a 150 µm thick beryllium window. The profiles of the measured X-ray spectra with the applied voltages of 1800 V, 2000 V and 2400 V were as follows: (1) The peak positions were 1.71, 1.79, and 1.86 keV, respectively. (2) The higher energy side edges of the spectra were 1.97, 2.14, and 2.54 keV, respectively. Above these energies, the spectra became zero counts. The spectra were weak, but similar to those measured by a glass cylinder X-ray tube described below.The X-ray spectra shifted to higher energy as the applied voltage became higher. During X-ray emission, a moving visible light emitting phenomenon was observed inside the cylinder, as shown in Fig. 4, where three representative patterns of light emission are pictured. The light ...