Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A new method for obtaining long optical paths and its use in spectroscopy Marshall, James Kelso 1941

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1941_A8 M3 N2.pdf [ 10.61MB ]
Metadata
JSON: 831-1.0105207.json
JSON-LD: 831-1.0105207-ld.json
RDF/XML (Pretty): 831-1.0105207-rdf.xml
RDF/JSON: 831-1.0105207-rdf.json
Turtle: 831-1.0105207-turtle.txt
N-Triples: 831-1.0105207-rdf-ntriples.txt
Original Record: 831-1.0105207-source.json
Full Text
831-1.0105207-fulltext.txt
Citation
831-1.0105207.ris

Full Text

•A. HEW METHOD FOR OBTAIMIMGi- LOUS OPTICAL PATHS AND ITS USE IN SPECTROSCOPY by-James Kelso Marshall A Thesis submitted in Partial Fulfilment of The Requirements for the Degree of MASTER OF ARTS in the Department TABLE OF COmBJTS Introduction Description of Apparatus (a) New device for obtaining long opt i c a l paths (b) Apparatus employed for gases at various pressures (c) Apparatus employed for lique-Experiments and Results Bibliography Appendix Reprint of paper published on part of this work. Acknowledgements A NEW METHOD FOR OBTAINING- LONG OPTICAL PATHS AND ITS USE IN SPECTROSCOPY 1. INTRODUCTION In order to obtain absorption spectra for certain substances, i t i s necessary to employ very long path, lengths of the material being inves-tigated. For example, in the case of hydrogen gas, i t can be calculated that a path length of 1000 kilometers at atmospheric pressures would be required to enable predicted bands in the near infra-red region due to quadrupole radiation of the hydrogen molecule to be observed, and, for the absorption, even in liquid hydrogen, to be detected, i t would be necessary to have nearly 13 meters of path. The problem of radiation makes the use of a straight tube of such a length impracticable. It can be seen that, i f an apparatus were designed to pass a beam of light many times through a short column of the liquid, the problem of radiation would be greatly minimized. However, i f gases are being studied, i t i s possible in certain cases to increase the working pressure, making possible a corresponding decrease in path length. There are, however, two reasons why i t i s prefer-able to use long paths at relatively low pressures rather than short paths at higher pressures. In the f i r s t place, at excessively high pressures, there may be a change in position, relative intensity and width of the absorption bands. For example, i t was thought that work carried out using a 6-ft. path and a pressure of 1700 l b . per sq.in. and vork using 34 f t . path length and 45 lb. per sq.in. might yield different absorption spectra. In the second place, i f only a small quantity of the gas were available, i t would be imperative to use the same column of gas repeatedly to get sufficient path length. This long path may be used to advantage also in studying the Fara-day effect, which is the rotation of the plane of polarization of light by a substance placed in a magnetic f i e l d . Here, a traversal of the tube by the beam of light in the direction opposite to that of the i n i t i a l beam w i l l give double the rotation, sinee the reflected beam i s now traveling in the opposite direction in the magnetic f i e l d , and n traversals of the tube produce, n times the angle of rotation of the plane of polarization of the light. This, however, i s not the case in natural rotation by sugar solutions and quartz, where a reflection of a beam of light back on i t s e l f cancels the rotation. 2. DESCRIPTION OF APPARATUS (a) New Device for Obtaining Long Optical Paths Attempts were made, therefore, to devise an apparatus that could be used in the-study of long path lengths of liquefied gases and gases at low and high pressures. The f i r s t t r i a l was made using a separate mirror for each reflection. The i n i t i a l beam from the source, on striking the spherical mirror M as shown in Fig. I was reflected to a small ^  in. square of aluminized mirror M T , fastened to a universal swivel b a l l and socket joint, placed at approximately 45° to the beam. From this, i t went to a similar small mirror Mg, then down to the spherical mirror M again. This process was repeated for another set of small mirrors Mg and MAt and thence to a third set M5 and Mg. The beam could f i n a l l y be reflected out to the spectrograph. This apparatus was too clumsy for any compact work, since the space available i s governed by the inside dimensions of the tube or vacuum flask containing the substance under investigation. Fig. I The,second method used was suggested when, during the work, two sets of reflections were obtained unexpectedly on the set of mirrors Mi and Mg. This new system consisted of two pairs of 4 in. by 1 in. alumi-nized mirrors with reflections occurring as in Fig. 2, where the central rays of the incident and reflected beams are shown i n the vicinity of the upper plane mirrors, the arrows indicating the directions of the rays. The lower spherical mirror M i s not exhibited in this or succeeding optical diagrams. Fig. 2. The-holder consisted of a' rectangular plate of brass PQRS with rectangular hole cut symmetrically out of the center as shown in Fig. 3& The construction of one i s shown in Figs. 3(b) and 3(c). Pig. 3(b) Fig. 3(c) A strip of brass J in. wide and 1/8 in. thick was bent into the shape GHJKLM. Two screws at B and B* and single screws at A, 0 and D held the strip of brass in place by pressing against the large rectangle. At the mid-point of section JK of each strip a hole was bored and a collar E attached, through which passed an axle FN, to provide a surface to which the mirror could be glued. By means of this axle and the screws, three translational and three rotational degrees of freedom for a mirror were obtainable. However, this apparatus was large and cumbersome, and i t was found almost impossible to get a regular series of reflections at the mirrors. During this experiment, as a possible alternative, two 90° totally reflecting prisms were tried instead of the two pairs of mirrors, but were found unsatisfactory for the following reasons: I. When a prism i s used, not more than one beam traversing the tube w i l l enter the long face of the prism at normal incidence. A ray striking this surface obliquely from without w i l l give rise to a reflected ray at the glass surface which may travel back and forth between the spheri-cal mirror M and the prism. Also, because of multiple reflections within the prism, other extraneous beams w i l l originate there which cannot, with-out d i f f i c u l t y , be separated or distinguished from the desired beam. II. For best results, the angle of the prism, or the equivalent angle between the mirrors, i s dependent on the radius of curvature of the spherical mirror. "When mirrors are employed, the required angle for a given spherical mirror may be obtained readily. Concerning the extent of variation of this angle, i t was found that for a particular experimental arrangement where a series of mirrors of radii of curvature ranging from 8 in. to 6 f t . were used, the angle between the mirrors varied from 87°45' to 89°41». III. Under certain conditions better results may be obtained by reducing the length of one of the plane reflecting surfaces. When employ-ing mirrors, this may be done by overlapping one mirror with the other. TV. For a given spherical mirror, one may obtain quickly a wide range 'of path lengths, without moving spectrograph, tube, or source, by a rotation of the plane mirrors about their axes, accompanied by an adjust-ment of the spherical mirror, which is not particularly c r i t i c a l . (b) Apparatus Employed for Gases at Various Pressures The f i n a l optical arrangement used proved highly successful. It is described f u l l y in a paper entitled "Method for Obtaining Long Optical Paths" published in the Journal of the Optical Society. A reprint of this paper i s attached to the present report. To get the greatest amount of light, the image of the hot crater of the horizontal positive carbon in the carbon arc source must be focussed on the diaphragm D attached to the back of the mirror (Fig. 4), with the distance between the condensing lens and the diaphragm the same as that between the diaphragm and the spherical mirror, i f the same diameter beam at this mirror as at the lens is desirable. This holds only i f the aper-ture through which the light enters the absorption tube does not restrict the size of the beam. To get a maximum path length, the second downwardly directed beam must be as close as possible to the original beam, since then a l l beams are closer together and therefore more traversals of the tube may be obtained. The aperture H (Fig. 4) in diaphragm D is placed almost vertically above the bevelled edge AB of the mirror M^ , and the f i r s t hori-zontal beam from Mg strikes as close as possible to A. The f i n a l beam, # Smith and Marshall; J.O.S.A., 30, 338, 1940. Fig. 4 reflected to the spectrograph by the small mirror M , would be cut off slightly by the edge of the mirror Mg and consequently have a cross-section that was not completely circular. In the apparatus used for gases at various pressures, the absorp-tion tube consisted of a 4 f t . length of extra heavy steel pipe, 5 in. inside diameter with -J in. walls, and 10 in . flanges welded on each end. Steel end plates were bolted to the flanges. The top plate, as shown in Fig. 5, had three holes A, B and G drilled along a diameter. The light enters and leaves the system through A and C respectively, while B serves as an observation port during the lining-up of the various mirrors. The f i r s t aperture A was 3/4 in, in diameter, while the other two were -g- in. in diameter.. A l l were recessed on the under side and circular glass win-dows -J in. i n thickness were waxed i n place. The circular windows in A and C were placed with axes passing through the spherical mirror M, so that the light would enter and leave perpendicular to the glass windows to cut down reflection losses. This was done by boring the recessions at the required angle. Attached to the under side of this plate was the suspension for - 9 -- l o -n g . 6(a) holding the plane mirrors ffi , M„ and 1L as shown in Fig. 6(a). Since I S 3 these mirrors had to be rotated about a horizontal axis parallel to the line of centers of the three windows, the actual block PQ holding the mirrors was joined to the suspension by a strip of spring brass at ES with two screws TU to get the required adjustment. In order that these screws could be adjusted from outside the apparatus, two holes were drilled in the upper plate at points half way between A and B, and B and C, and 1§ inches away from the line of centers of the windows. - 12 -When pressure i s applied to the tube, these holes are closed by 5/8" threaded machine plugs. Fig. 6(b) is a photograph of this plate and sus-pension. The lower plate supports a 4 inch stand A on which the spherical mirror rested. In order to move the spherical mirror H, a strip of brass B was bent i n the form of a ci r c l e around i t . A nut D was soldered on one side, and opposite this, a hole was bored in the pipe G in a line with the line of centers of the three windows in the top plate. Through this hole, a threaded rod E was passed, screwed into the nut D. This hole i s closed with a threaded plug when gases under pressure are being studied. Fig. 7 - 13 -Since the mirror H had to be t i l t e d slightly while in the pipe, three lugs C were fastened to the collar on which the mirror rested. The collar also had three nuts F soldered onto i t in a horizontal plane through which pointed screws pressed on the bottom plate A. By turning these screws, the mirror could be t i l t e d . In the wall of the pipe at the level of M-j_ and Mg,(Fig. 6(a) , and directly below the center window, a 3/8 inch hole was bored through which Mj, Mg and Mg could be moved, and the gas introduced. The mirrors were fastened to a pair of f l a t brass plates A which were screwed to the axles B (Fig. 8(a)). The front washer C of each pair had a threaded collar D projecting back. The rear washer E slipped over this and a nut F clamped i t tight on the block G supporting the mirrors. G consists of a -5 inch brass plate with a rectangular hole cut symmetrically i n i t . Fig. 8(b) is a photograph of this without the plate G. A ri £ 8 Fig. 8(a) Fig. 8(b) - 14 -(c) Apparatus Employed for Liquefied Gases In the ease of liquefied gases contained in glass vacuum flasks, the apparatus used resembled that employed in the study of gases at various pressures in many respects, but a l l adjustments must be made from above. The top consisted of a disc of balsa wood, completely enclosed in thin sheet steel, with circular apertures provided for the entrance and exit of the light beams. The apparatus for holding the plane mirrors was similar to that used for the pressure apparatus. The spherical mirror was sup-ported from the top by three adjustable rods, and was moved in the line of centers of the holes by the device shown in Fig. 9(a). This consists Fig. 9(a) Fig. 9(b) - 15 -of the rod A: stipported from above with a thick washer clamped at B. The rod A is free to move in the sleeve C which i s held a fixed distance above plate H by means of rods D. The lower end of A i s threaded onto a lever arm E. This i s pivoted at F to a fixed clamp K. Rigidly fastened to E i s a rod having a peg G which extends through a slot in the plate H. G projects into a hole in the attachment at the rear of the mirror, and so, when A i s turned, the mirror moves across the plate H. Fig. 9(b) is a photograph of the device. In order to cut out reflections at the surface of the liquefied gas, the plane mirrors M 1 and Mg must be completely immersed, and so the supports for these mirrors, as well as for the spherical mirror, must be made of some material of low conductivity, such as stainless steel or German silver. 3. EXPERIMENTS AND RESULTS For this work on pressure oxygen, a set-up was obtained using twelve passages of light between the upper mirrors and Mg and the lower mirror M, the equivalent of a 34 f t . path length. The light from the sys-tem was reflected from Mg to the s l i t of a Hilger constant deviation spectrograph. For the region between 7000 A 0 and 8600 A°, the most satisfac-tory plate was the Eastman I - R spectrographic plate, sensitized with a solution of three parts 28% ammonia, two parts of Ethyl alcohol, and seven parts of water. The plate was placed in this for one minute and then dried with a fan. - 16 -Using such a plate, the absorption spectrum of oxygen under a pressure of 45 l b . per sq.in. was investigated. A print of the spectro-gram obtained i s shown in Fig. 10. It w i l l be noted that the band at 7660 A i s very narrow and actually appears double on the plate. J i g . 10 For comparison with this plate, another plate of the same sensi-t i v i t y was taken using a 6 f t . path length with a pressure of 1700 lb. per sq.in. A reproduction of this plate i s shown in Fig. 11. It w i l l be noted that, although the exposure is longer, shown by a more complete spectrum, which would tend to make the lines narrower, the band at 7660 A° i s actually broader, and only with considerable dif f i c u l t y can i t be seen to be a double band. o Fig. 11 In order to get a complete, general picture of the absorption bands of oxygen, a spectrogram was taken using the 6 f t . tube and 1500 lbs. per sq.in. pressure. An Eastman I - L spectrographic plate, sensitive up to 9000 A°, was used. Well defined bands are shown at 7660 A 0, 6920 A°, 6360 A 0, 5830 A 0, 5360 A 0, 4980 A° and 4800 A°. A print of this plate i s - 17 -shown as Fig. IE. Fig. 12 As a f i n a l exposure, an Eastman I - Z spectrographic plate, sen-sitive in the extreme infra-red up to 12,000 A° was used, and the bands at o ° 9100 A and 10,400 A f i r s t observed, using absorption of light by liquid oxygen, by McLennan, Smith and Wilhelm,^ and, as far as i s known, never before observed with gaseous oxygen, were clearly visible. A print of thii i s included as Fig. 13. 5. BIBLIOGRAPHY McLellan, Smith and Wilhelm; Trans. Roy. Soc. Can. I l l , 24, 1, 1930. Smith and Marshall; J.0.S.A,, 30, 338, 1940. J Fig. 13 # McLennan, Smith and Wilhelm; Trans. Roy. Soc. Can. I l l , 24, 1, 1930. Reprinted from JOURNAL OF T H E OPTICAL SOCIETY OF A M E R I C A , Vol. 30, No. 8, 338-342, August, 1940 Printed in U. S. A. Method for Obtaining Long Optical Paths H . D . , SMITH AND J. K . M A R S H A L L University of British Columbia, Vancouver, British Columbia (Received June 2, 1940) WH I L E studying the absorption Spectra of certain liquids, and gases at high pres-sures, it became necessary to go to very thick layers of the substances under investigation. Many of the methods previously described for obtaining long optical paths could not be em-ployed due to the fact that the substances being studied were available only in columns of re-stricted length. A simple apparatus was devised finally by means of which an absorption spec-' trum, for a path length of several hundred feet, of a substance confined in a limited space, could be obtained with a relatively short exposure time. A brief description of the optical system evolved is given here, accompanied by a series of photo-graphs of actual light paths illustrating several variations of the system which were used under special experimental conditions. The general experimental arrangement is indi-cated in Fig. 1, which is a schematic diagram of one modification of the optical system. The posi-tions of the various optical parts were deter-mined roughly in the following manner. The spherical mirror M% was placed at one end of the tube containing the absorbing medium in such a way that two plane mirrors Mi and Mi, situated at the other end of the tube, were at the center of curvature of Ms. A horizontal beam of light from the source S was condensed by the lens L and reflected by a small plane mirror M in a direction designated by beam A in Fig. 1. Suc-cessive reflections of the beam at Ms, Mi, Mi and again at Ms, gave rise to the beams B, C,D as shown in the diagram. The light emerged in a horizontal direction after reflection at the upper part of Mi and was focused on the slit of a spectrograph. All mirrors used were front-surfaced with evaporated films of aluminum or silver and were completely immersed in the ab-sorbing medium. Regarding the finer adjustment of the apparatus, it was found that when the longest possible path lengths were desired, best results could be obtained by rotating either Mi or Mi about a horizontal axis until the portions of the beams between the two mirrors Mi and M% were parallel to one another as shown in Fig- 7a. Under these conditions it is obvious that the angle between Mi and Mi is slightly less than a right angle. An abstract of some recent work of Kratz and Mack1 gives a method for obtaining a long optical path in a limited space by the use of a truncated totally reflecting prism and a spherical mirror. A somewhat similar arrangement, in-volving the use of a 90° totally reflecting prism, was tried out during the present work. However, it was considered advantageous to replace the prism by the mirrors Mi and Mi, for the following reasons— I. When a prism is used, not more than one beam of the group traversing the tube will enter the long face of the prism at normal incidence. A ray striking this surface obliquely from without will give rise to a reflected ray at the glass surface which may travel back and forth between Ms arid the prism. Also, because of multiple reflections within the prism, other extraneous beams will originate there which cannot, without difficulty, be separated or distinguished from the desired beam. F I G . 1. Schematic diagram of one modification of the optical system. 1 Kratz and Mack, Phys. Rev. 57, 1059A (1940). 338 M E T H O D F O R O B T A I N I N G L O N G O P T I C A L P A T H S a F I G . 3. b H . D . S M I T H A N D J . K . M A R S H A L L 340 a F I G . 7. b H . D . S M I T H A N D J . K . M A R S H A L L 342 II. For best results, the angle of the prism, or the equivalent angle between the mirrors M\ and M%, is dependent on the radius of curvature of the spherical mirror. When mirrors are employed the required angle for a given spherical mirror may be obtained readily. Concerning the extent of the variation of this angle, it was found that for a particular experimental arrangement where a series of mirrors of radii of curvature ranging from 8 in. to 6 ft. were used, the angle between Mi and ikf2 varied from 87° 45' to 89° 41'. III. Under certain conditions better results may be obtained by reducing the length of one of the plane reflecting surfaces. When employing mirrors, this may be done as illustrated in Fig. 5a by overlapping one mirror with the other. IV. For a given spherical mirror one may ob-tain quickly, a wide range of path lengths, with-out moving spectrograph, tube or source, by the rotation of either Mi or Mi about its axis, ac-companied by an adjustment of M% which is not particularly critical. V. Some of the modifications of the' optical system exhibited in the photographs could not be obtained with a prisrrl without considerable difficulty. Figures 2a-7a show the actual paths of rays in the vicinity of the plane mirrors for a number of variations of the optical system. These plates were obtained by photographing the light scattered by smoke particles introduced in the paths of the beams. The central rays of the beams are shown in the diagrams accompanying the plates, the arrows indicating the directions of the rays. If we suppose "that the oblique rays con-verge at a point 0 on the spherical mirror, then the path in each case may be written in terms of the letters in the diagrams. Thus, in Fig.-4b, the light follows the path ABOCDOEFOGHOJKLM. For the photographs reproduced here, the spher-ical mirror had a radius of curvature of 5 ft., and the plane mirrors, Mi and Mi were 2 in. long and 1 in. wide. A carbon arc, drawing 12 amp. in a 110-volt, d.c. circuit, served as a source of light. Figure 2a shows a system, in which there are 10 traversals of the tube and where the absorp-tion path is approximately 50 ft. in length. Figure 3a gives a modification of this method in which the small mirror M is eliminated, the initial beam striking Mi. It may be noted that, for the first two methods, the reflections at the plane mirrors M\ and Mi after successive tra-versals of the tube form a series approaching the junction of Mi and Mi and emerging there. Figures 4a and 5a show variations of a system in which the series of reflections at the plane mirrors approaches the junction, crosses over and then recedes from the junction. In Fig. 4a a small mirror at L is used to deflect the final beam com-ing from the outer edge of Mi, while in Fig. 5a this mirror is eliminated by moving Mi in its plane so that it overlaps Mi at the junction, and the emergent ray is reflected from Mi. In Fig. 6a the initial beam enters between the two plane mirrors and after a series of reflections which approach the edges, the emergent beam is re-flected from the spherical mirror to the spectro-graph. Fig. 7a shows the pattern obtained when the light traversed the tube 15 times. The final beam could be reflected out from the center of the group by means of a small mirror. As may be expected, it is difficult to obtain a photograph of a larger number of traversals due to the fact that the intensity of the light along the path di-minishes rapidly as it is scattered by the smoke particles. In general; one sees that for every two tra-versals of the tube there are three reflections at the mirrored surfaces, one being at almost normal incidence, and two, at approximately 45° inci-dence. In one case with a mirror of 85 cm radius of curvature 37 traversals were obtained without undue difficulty. With a mirror of longer radius of curvature an even greater number of traversals could be obtained. It was found that the above methods could be employed equally well with the absorption tube in either a horizontal or vertical position. The system may be used to advantage also in light scattering experiments and in the determination of the Verdet constants of gases and liquids. The latter part of this work was carried out at the University of California and the authors are indebted to the Physics Department of that institution for the use of its facilities. Thanks are due particularly to Professors Birge, Brode, Jenkins and White for their kindness in providing special equipment. AOEITOILBDQ-EMENTS Thanks are due to Dr. H. D. Smith, who suggested this problem and assisted in the work. Part of this work was carried out at the University of California, and the writer i s indebted to the Physics Department of that i n s t i -tution for the use of i t s f a c i l i t i e s . Thanks are due particularly to Professors Birge, Brode, Jen-kins and White for their kindness in providing special equipment. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0105207/manifest

Comment

Related Items