Spatial intensity distributions of vacuum ultraviolet photon beams measured by thermoluminescent image films NOBUYUKI KIKUKAWA'

AND

TSUTOMU MAKITA'

Department of Chemistry, Faculty of Science, Kyoto University, Kyoto, 606 Japan Received January 5, 1988

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

This paper is dedicated to Professor Charles A. McDowell on the occasion of his 70th birthday

NOBUYUKI KIKUKAWA and TSUTOMU MAKITA. Can. J. Chem. 6 6 , 1857 (1988). The thermoluminescent phosphor, calcium sulfate activated by manganese, was found to be a suitable material for imaging films to record the spatial distributionsoand reflections of vacuum ultraviolet (vuv) photon beams used for high-resolution photoelectron spectrometers. He I 584 A photons radiated in a microwave discharge tube were collimated through capillary tubes, and the phosphor films were placed at the position of the photoionization chamber. When the irradiated phosphor films were heated up to 200°C in a darkroom, faint green glow images were observable on the phosphor films and these images were recorded by photographic films. When the films were placed 118 mm from the exit of the collimating capillary tube, the photon beam images were almost circular, about 4.5 mm in diameter. The photon distribution in the collimated beam was interpreted with the aid of computer simulation. The experimental results revealed that the reflections of vuv photons at planes of various kinds were almost specular if the surface was clean and smooth and the glancing angles were small. Nosu~uKrKIKUKAWA et TSUTOMU MAKITA. Can. J. Chem. 66, 1857 (1988). On a trouvk que le sulfate de calcium activk par le mangankse, un produit phosphorescent et thermoluminescent, est un produit utile pour faire des films desquels on peut tirer des images permettant d'enregistrer les distributions spatiales et les rkflexions des faisceaux de photons ultraviolet sous vide (uvv) qui sont utilisks dans les spectromktres photoklectroniques a haute rksolution. On a dirigk les photons du He I 584 A qui proviennent d'une radiation dans un tube a dtcharge a micro-ondes vers des tubes capillaires servant de collimateur et on a place les films phosphorescents ii la place de la chambre a photoionisation. Lorsque les films phosphorescents irradiks ont kt6 chauffks a 200°C, dans un chambre noire, on a pu observer de faibles lueurs d'images vertes que I'on a enregistrk images sur des films photographiques. Lorsqu'on place les films 5 118 mm de la sortie du tube capillaire agissant cornrne collimateur, les images du faisceau de photons sont pratiquement circulaires, avec un diamktre d'environ 4,5 mm. On a interpritk la distribution des photons dans le faisceau qui a pass6 au collimateur en faisant appel a une simulation par ordinateur. Les rksultats exptrimentaux suggkrent que les rkflexions des photons (uvv) sur divers types de plans sont pratiquement sptculaires si la surface est lisse et propre et que les angles de percement sont faibles. [Traduit par la revue]

Introduction Vacuum ultraviolet (vuv) radiation is indispensable for photoionization studies and photoelectron spectroscopy. For instance, high energy photons, He I 584 A, are radiated conveniently by the microwave discharge of low pressure helium gas flowing in a source tube (1). The radiation is collimated through capillary tubes between the discharge tube and the ionization chamber, which contains target molecules. The capillary tubes are necessary to maintain the pressure differential of helium gas between the discharge t!be and the ionization chamber. Furthermore, for the He I 584 A radiation, no material can be used as an efficient window between them. The size and the intensity distribution of collimated vuv photon beams are one of the fundamental design factors for the ionization chamber of high-resolution photoelectron spectrometers (2). Schumann-type plates are not feasible for imaging, because of their sensitivity to visible radiation from the discharge tube. Lyman (3) reported that a thermoluminescent phosphor was sensitive to photon beams of wavelengths from 1300 to 140 A. The phosphor was mainly calcium sulfate containing a few percent of manganese sulfate. Lyman found that atmospheric air was almost transparent to radiation of 1300-1 100 A , but that it absorbed radiation of 2000-1300 A and of wavelengths shorter than 1100 A. Watanabe investigated the properties of 'present address, National Research Institute for Pollution and Resources, Tsukuba, Ibaraki, 305 Japan. 2~resentaddress, Notre Dame Women's College, Chemistry Institute, Kyoto, 606 Japan.

the manganese-activated CaS04 phosphor in detail (4) in connection with its use as a detector of extreme ultraviolet radiation. These thermoluminescent plates were flown in rockets, exposed to the sunlight at an aliitude level between 82 and 128 km, and recovered for dosimetry (5). Image plates made of thermoluminescent phosphor are convenient devices to record the dimensions and intensity distributions of vuv photon beams, for instance, He I 584 A. Because th? plates are not sensitive to light of wavelength longer than 1340 A, they can be handled under room light. The faint green glow images of the thermoluminescent plates have to be detected in the darkroom. This paper reports on the spatial distributions and reflections of He I 584 A hot on beams for a high-resolution photoelectron spectrometer.

Experimental Figure 1 is the schematic diagram of a He I 584 A light source assembly combined with the thermoluminescent image film on copper plate, helium gas flow regulation system, and microwave power supply. Figure 2 is a detailed diagram of the light source to show the quartz discharge tube, capillary tubes I and I1 of Pyrex glass, and the differential pumping. Helium gas was introduced from a cylinder through pressure-reducing valves and a needle valve. The microwave generator was type MR-IS of Ito Electro-Medical and Chemical Instruments Mfg. Co., Tokyo (2450 MHz and 200 watts maximum). The cooking electronic range was, with some modifications, suitable for the microwave generator. The microwave cavity at the discharge tube was a cavity of type 5 as described by Fehsenfeld, Evenson, and Broida (7). The photon beam was mainly He I 584 A, with we+ contaminations such as H I Ly a 1216 A. The intensity of 1216 A photons was found to be reduced remarkably when the helium gas was

CAN. 1. CHEM. VOL. 66. 1988 Needle valve

Rciluci~~g vnlvc

tlcliurn

A i r colllpressor

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

I*---_I I

cavity

power source

I I

-

'l'l~ern~olu~ni~~escc~~t f i l m and i t s s u l ~ p o r t i ~plate ~g

FIG. 1. Schematic diagram of vacuum ultraviolet light source assembly with thennoluminescent imaging film and helium gas flow system.

FIG. 3. Photographic picture of glowing images. (a) Image collimated photon beam falling directly on the phosphoIr film. Images of split photon beam on the identical phosphor film, see Fig

FIG. 4. Picture recorded on photographic film in contact with glowing phosphor film, without photographic lens. To give reference exposures on identical phosphor film, each quarter of the circular film was irradiated with a broad flux of photons for different time intervals (see text). FIG. 2. Detailed diagram of the light source. A; Quartz tube, 8 mm i.d. B; Microwave cavity, Fehsenfeld-Evenson-Broida type (7). C; Capillary I, 1.0 mm i.d., 10 mm length, Pyrex glass. D; Capillary 11, 0.5 mm i.d., 35 mm length. E; Thennoluminescent imaging film on copper plate, 20 X 50 X 1 mm. Distance between E and the end of D is 118 mm. F; From helium cylinder, through pressure reducing valves, a needle valve, and U-shaped copper tube cooled in liquid nitrogen. G; To mechanical pump.

supplied through a U-shaped copper tube, which was cooled in liquid nitrogen. Three different procedures to make the phosphor were tried, as described by the reports of Lyman (3), Watanabe (4), and Novotne et al. (6). These three kinds of the phosphor were exposed to He I 584 photon beams under similar conditions. The intensities of thermoluminescent green glow images were compared visually in a darkroom;

i2

KIKUKAWA AND MAKITA

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

.......................... ............................................. I ::::::::::::;:::::::

FIG. 6. Number of photons arriving at each 0.1 X 0.1 mm meshed area on phosphor film, computer simulation for 500 000 photons. Symbols (right) correspond to the intensities listed. film and exposed for less than 3 s. Figure 4 is an example of Tri-X film images. High temperatures from the tape heater had no effect on the Tri-X film. When the collimated photon beam was directed around the edge of a plane surface, some of the photons can proceed directly to the phosphor film, while the remainder reflect on the surface and may proceed to another position on the same film. Figure 5 is a schematic diagram to show the trajectories of photons to the phosphor film and the positions of the points of arrival at the film for the split photon beams. Figure 3b is an example of the split images on the same phosphor film. The relative intensities of the split beams can be determined by comparing the glowing intensities of the two images. The relative attenuation factors for photon beams reflected on the surface were determined as a function of the incident angles, assuming that the incident angle and reflection angle were equal. FIG. 5. Schematic diagram of the spatial distribution of a collimated photon beam. The horizontal scale is expanded 20-fold compared to the vertical scale. Axes of capillary I and I1 are assumed to be parallel but slightly shifted. The two circles on the imaging film correspond to the limits of the photon trajectories; (a), no reflection and (b) single reflection in capillary 11. the phosphor made following Watanabe (4) was found to be the brightest, and was used throughout with slight modifications as follows. Approximately 8.8 g of CaS04.2H20 was mixed with 0.12 g of MnS04 in 60 mL of concentrated sulfuric acid and a small amount of water. The sluny was left at room temperature for 72 h. Most of the liquid part was then removed by decantation. The excess sulfuric acid was vaporized at 250°C, and the residual mixture was heated to red heat in a porcelain crucible for 90 min. After cooling, the mixture was ground to fine powder, and was suspended in a dilute benzene solution of Dow Coming high vacuum silicone grease. One drop of the suspension was sufficient to make a thin film of 10 mm diameter on a copper plate, 20 x 50 x 1 mm. Such imaging films were strong enough for repetitive exposures to intense vuv photon beams and heatings up to 200°C for thermoluminescence. The imaging film on a copper plate that had been exposed to the photon beam was heated by a tape heater in the darkroom, and the thermoluminescent glowing image was photographed using Polaroid type 107 film in a model 180 camera. Close-up lenses, Cannon 240 and 450, were attached to the camera. Figure 3a is an example of the glow image. It was possible to record the glowing images on a Kodak Tri-X film directly without photographic lens. The Tri-X films were sandwiched between thin glass plates, and were placed on the glowing

Results and discussion The slurry of fine powder of the phosphor was painted on a copper plate to make the thermoluminescent film, 10 mm diameter, about 10 mg/cm2 in thickness. The films were stable in the atmosphere for many months. They were irradiated by He I 584 A photons for up to 160 min. The output power of the microwave generator was usually 50 watts. The copper plate was dismounted from the light source and transferred to a darkroom within 5 min of irradiation to inspect the thermoluminescent image. Before heating the film, it was necessary to wait for 10 min for vision to adjust in the darkroom. The temperature of the film was then increased to 200°C in 15 min. The faint thermoluminescent glow image was observable for the first 5-min period and then faded out. When the irradiated films were exposed under room light for 2 min, there was no appreciable effect on the subsequent intensity of the glow image. When the films were left under room light for 90 h, the glow intensity decreased to one sixth of that of the control films kept in the darkroom. Spatial distribution of the collimated vuv photon beam was determined by measuring the size of the thermoluminescent image of the phosphor film. Figure- 3a is an example of the images. This shows that the He I 584 A photon beams are collimated within a 4.5 mm diameter, 118 mm from the exit of capillary 11. The ionization chambers for photoionization studies were customarily placed ca. 100 mm from

CAN. J. CHEM . VOL. 66. 1988

TABLE1. Replts on the estimated intensity attenuation factors of a He I 584 A photon beam due to specular reflections at various conditions of the reflecting plates Material of the plate

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

Glass Glass Glass Glass LiF crystal LiF crystal LiF crystal LiF crystal FIG.7. Intensity distribution profile of photons along the horizontal central meshes in Fig. 6.

Copper Copper

Surface conditioning

Glancing angle (")

Intensity ratio, reflectedlimpinging

Cleaned, in ethanol Cleaned Cleaned Covered with thin layer of silicone oil Cleaned Cleaned Cleaned Covered with colloidal graphite Sandpapered Amalgamated

*Split beams were detected by two different films

FIG. 8. Schematic geometry for split photon beam images. A, Capillary 11. B, Reflecting plate. C, Phosphor film. D, Image of direct beam. E, Image of reflected beam. the light sources. The divergence of the collimated photon beam was estimated to be 2.2" or 9 x lop4steradians. The resolution of the image film was better than 0.1 mm. Close'examination of the 4.5-mm circular image of Fig. 3 a revealed that the image consisted of two overlapping but not concentric circles, one bright inside and one faint outside. The intensity of the inner bright circle was about three times that of the faint one. Figure 5 is a schematic diagram of the spatial distribution of a collimated photon beam. It is suggested that the two circles, as well as the separation of their centers, resulted from reflections of the photons on the inside surfaces of the

capillaries when the central axes of the capillaries were made parallel but their centres were shifted slightly, indicated as A in Fig. 5. The ratio of the diameters of the two circles in these theoretical considerations is similar to the experimentally determined ratio (Fig. 3a). Reflections of photons at the inside wall of capillary I1 were assumed to be specular. The slight shift of the two centers of the circles could be explained schematically via a corresponding shift of the two capillaries. A preliminary computer simulation study was carried out using a Toshiba ACOS 600 S on the trajectories of the collimating photons through the collimating system. Figures 6 and 7 are examples of the results for 500 000 photons. It was assumed that, ( i ) each photon started at a random point on the circular plane at the entrance to capillary I, (ii) the starting direction of the photon is also random, and (iii) some photons can emerge from the exit circular plane of capillary I1 without reflection, some after a single specular reflection, and some after two specular reflections. The numbers of photons arriving at each 0.1 X 0.1 mm meshed area on the plane of the film are shown as various kinds of dots in Fig. 6. Figure 7 is the intensity distribution profile along the horizontal central line in Fig. 6.' The collimated vuv photon beam source for high-resolution photoelectron spectrometers (2) was designed on the assumption that all reflections of photons at the inside surface of capillary tubes were specular without intensity attenuation when the glancing angles were small. The following considerations refer to this assumption. Figure 3b is an example of split photon beam images, one transmitted directly to a phosphor film, and the other reflected from near the edge of a plate to the'same phosphor film. The geometry used is described in Fig. 8. The relative intensities of the photon beams were estimated by comparing with the glow images of the reference beams, as described below. Figure 4 is an example of the glow image. The phosphor film of Fig. 4 was circular, about 10 mm diameter, and one half part of the circle was covered with a thin plate of lithium fluoride while the film was irradiated by a large-diameter flux of 3~ablesof supplementary material may be purchased from the Depository of Unpublished Data, CISTI, National Research Council of Canada, Ottawa, Ont., Canada KIA 0S2.

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

KlKUKAWA AND MAKITA

He I 584 A photons for 5 min. Then, a different half part of the circle, which was displaced by 90" from the previously irradiated half of the film, was covered with the LiF plate and irradiation continued for 10 min. Each quarter of the circular phosphor film was thus irradiated at different time intervals by this procedure, the relative exposures being 0 , 5 , 10, and 15 min in this case. It was realized that visual observation was more accurate than photographic recording, to estimate the relative size and intensities of the faint glow images of the phosphor film. Therefore, photographic records (Figs. 3a, 36, and 4) are given only for illustration. Table 1 is a summary of the experimental results on the intensity ratio of the specularily reflected photon beam to the impinging photon beam upon various kinds of plate, as investigated via the arrangement of Fig. 8 and the above calibration procedure. It is thereby well established that vuv photon reflections are mainly specular at clean and smooth planes if the glancing angles are less than 10".

Acknowledgements This work was supported in part by a Grant-in-Aid from the

1861

Ministry of Education, Science and Culture, Japan. One of the authors (T.M.) is very grateful to Professor C. A. McDowell and Professor D. C. Frost, University of British Columbia, for their encouragement during this research. The authors are indebted to Dr. N. Itada, Kyoto National Hospital, for helpful discussions. 1. M. I.AL-JOBOURY and D. W. TURNER. J. Chem. Soc. 5141 (1963). D. C . FROST,T. MAKITA, C . A. MCDOWELL, 2. G. R. BRANTON, and I. A. STENHOUS. J. Chem. Phys. 52, 802 (1970). P h y s Rev. 48, 149 (1935). 3. T. LYMAN. 4. K. WATANABE. Phys. Rev. 83, 785 (1951). 5. R. TOUSEY, K. WATANABE, and J. D. PURCELL. Phys. Rev. 83, 792 (1951). 6. J. NOVOTONE?, Z. SPURN?,and M. BINOVA.J. Phys. Chem. Solids, 31, 1412 (1970). 7. F. C. FEHSENFELD, K. M. EVENSON, and H. P. BROIDA. Rev. Sci. Instrum. 36, 294 (1965).