Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Electron attachment and negative ion formation in oxygen Christy, Robert Frederick 1937

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

Item Metadata


831-UBC_1937_A8_C3_E4.pdf [ 3.17MB ]
JSON: 831-1.0085310.json
JSON-LD: 831-1.0085310-ld.json
RDF/XML (Pretty): 831-1.0085310-rdf.xml
RDF/JSON: 831-1.0085310-rdf.json
Turtle: 831-1.0085310-turtle.txt
N-Triples: 831-1.0085310-rdf-ntriples.txt
Original Record: 831-1.0085310-source.json
Full Text

Full Text

L G 3 •& 1 (1 I ^ li ELECTRON ATTACHMENT AND NEGATIVE ION FORMATION IN OXYGEN Robert Frederick Christy A Thesis submitted for the Degree of Master of Arts i n the Department of Physics THE UNIVERSITY OF BRITISH COLUMBIA August, 1937. TABLE OF CONTENTS I. INTRODUCTION. I I . THEORY, I I I . APPARATUS. (a) Tube (i) Construction (11}.Operation ( i i i ) Measurement of Mobility (b) Oscillator (c) Quadrant Electrometer (d) The Vacuum System IV. CONCLUSION. BIBLIOGRAPHY. I L L U S T R A T I O N S Figure 1 Facing page. 2. Figure 2 * " 3. Figure 3 " " 3. Figure 4 " " 4. Figure 5 " rt 6„ Figure 6 " " 6. Figure 7 " " 9• Figure 8 " " 9. ELECTRON ATTACHMENT AND NEGATIVE ION FORMATION IN OXYGEN • . CHAPTER I INTRODUCTION^ It was observed "by many workers ^ )^ that the mobil-i t i e s of photoelectrically generated ions, measured soon after l i b e r a t i o n from a metal plate i n the measuringfield, began to become abnormally great at 100 mm. pressure i n a i r . This was ascribed by numerous observers to a breaking up of the negative ion cluster by impacts with gas molecules. To Wellisch(3,4,5) ;LS due the creditfor having investigated, by means of a modification^) of the Rutherford^ 7^A.C. method, the mobilities of negative ions when formed behind a gauze, at low air pressures., after they had come through the meshes of the gauze. He found two classes of c a r r i e r s , one of which he showed were normal ions of a mobility constant close to the value accepted for normal negative ions, the other of which he asserted consisted of free electrons. The effect of the gauze and the weak auxilia r y f i e l d back of i t was to delay the photoelectrically liberated electrons, or electrons which were freed by radium radiations, u n t i l they attached to neutral molecules to form ions. This Wellisch did.not - know. In fact he did not compare his results with the results of other workers who had used no gauze. He assumed that an electron, when liberated, must have a certain minimum energy for attachment and that I f i t has not this energy i t w i l l never attach and w i l l remain permanently free. In 1920, L o e b ^ undertook the problem and repeated Wellisch'a experiments. He completely corroborated the observations of Wellisch. But he further observed that the re l a t i v e number of ions and free electrons depended on the pressure and aux i l i a r y f i e l d strength which allowed of only one interpretation. I t showed that the free electrons were not permanently free. In fact, i t showed that ion formation was contingent on the condition that the electrons spend a s u f f i c i e n t time i n the au x i l i a r y f i e l d before being studied. I f the pressure and f i e l d strength were such that the time was short, only electrons were obtained; i f the time was long, only ions were obtained. By increasing the a u x i l i a r y f i e l d strength so that the electrons had a very high energy, no increase' i n the number of ions formed was observed by Wahlin^), Thus the Wellisch theory was found to be wrong. The experimental work and viewpoint on this problem Oo) were very much aided by a theory of J. J. Thomson. He att-empted to explain the abnormal increase of mobility of photoelectrically liberated ions at low pressures as follows: The u l t r a - v i o l e t l i g h t used liberates electrons. These do not attach to molecules to form negative ions at their f i r s t impact. I f i t be assumed that the electron requires, on the average, n impacts before i t can attach to form an ion, where n may be a large number, then i t i s possible to explain the phenomenon. This amounts to assuming that the attachment of an elecrtron i s a chance phenomenon, depending on where i t H i . strikes the molecule or under what energy conditions the impact takes place. For each chemically different gas this would be different, depending on i t s chemical nature. For sim p l i c i t y , Thomson assumed n to be a constant, independent of yelocity, characteristic of each gas. That t h i s i s not ent i r e l y correct w i l l be seen l a t e r . As a f i r s t approximation, i t i s , however, s u f f i c i e n t . That electrons do make negative ions i n some gases and not i n others i s an observed fac t . The term electron a f f i n i t y may then be applied to describe the behaviour of a molecule towards the electron. It i s then necessary to find more than a qualitative measure of this property. Two methods are available. The one would be to find out what the ionizing potential of the extra electron of the negative ion i s . This, of course, could be determined by the frequency of the short-est wavelength of the l i g h t emitted when an electron i s attach-ed to a molecule to form an ion. This multiplied by the Planck constant would give the energy necessary to remove the electron. Up to the present, investigations of this nature have yielded no result and one cannot measure electron a f f i n i t y i n this manner. The second procedure would be to get a quantit-ative measurement of the average number of the impacts required by an electron with a given type of molecule before i t could attach. This quantity i s precisely the quantity defined by J. J . Thomson as n, his constant of attachment. .' The f i r s t attempts made to. measure n were due to LoeV°''and W a h l i n w ' . The effect of the attachment, and i t s i v . consequent change of mobility of the car r i e r , on the shape of the current-voltage curves obtained with a square- wave-form alternating current has been rigorously worked out by Mooney and loeb, I t i s given, together with the attempt at experimental v e r i f i c a t i o n , i n an a r t i c l e by loeb ) .The paper doesn't give accurate quantitative agreement because of the variation i n the attachment constant with electron velocity. B a i l e y ( 1 4 ) , i n Townsend's laboratory, measured the attachment constant n, or, better, i t s reciprocal, the probability of attachment h, by the analysis of the rate of loss of electrons and ions by l a t e r a l d i f f u s i o n i n a beam of what were i n i t i a l l y electrons moving under a f i e l d i n a i r at low pressures. He observed the values of h or n for a i r and found that these depended on the velocity, or, better, the energy of the electrons, h decreasing and n increasing as the value of the energy increased. Hence n or h i s a function of X/p (X i s f i e l d strength i n volts/cm. and p i s pressure In mm.) which must be included i n the equations of Loeb i n order to evaluate h. The complexity of the equations and the d i f f i c u l t y i n obtaining accurate values.of electron mobilities at higher pressures made an attempt to deduce the variation of h as a f(X/p) from these data impossible. B a i l e y ^ 1 4 ) , using the method of measuring the l a t e r a l d i ffusion of a current beam going through holes i n a series of plates, found that n increases as the velocity of thermal agitation increases. This means that n i s not a constant of the chemical nature of the gas alone but depends on the average energy of agitation of the electron. This would be expected as V . i t seems that as the electron travels faster a capture would be more d i f f i c u l t to effect. This agrees with Wahlin's results, who found that as the f i e l d increased the number of free electrons i s not decreased as Wellisch 5s theory would make i t . Cravath t i n loeb's laboratory, applied a new method devised by loeb to the measurement of n or h. In this a stream of electrons from a hot filament or other source i s driven through a known distance x of a i r or gas at low press-ure by an e l e c t r i c a l f i e l d . At x cm. from the source there i s an equipotential surface having a grid of fine p a r a l l e l wires separated from each other by 1 or 2 mm. and insulated. Alter-nate wires are connected to the terminals.of a high-frequency alternating supply of potential of the order of 10 6 cycles per. second. This high-frequency f i e l d served to remove any of the very mobile electrons l e f t i n the beam after the distance x had been traversed, without removing the slow negative ions. The l a t t e r reached the other electrode creating the f i e l d , and the r a t i o of the current to the grid wires to the t o t a l curr-ent gave the frac t i o n of the electrons which had not attached i n the distance x. Using the data of Townsend and his pupils on electron mobilities i n ai r at low pressures, Cravath was able to obtain h or n as a function of the terminal velocity or energy of the electron. His results lay close to those of Bailey i n a i r , but i n Qg they extended to far higher values of X/p than did the previous measurements i n a i r . Cravath found that h f e l l to a minimum at about 1 or 2 electron volts energy and started to r i s e again. He found that h appeared to v i . vary with pressure when larger ranges of pressure were used. He also observed what seemed to be a detachment of electrons from negative ions i n the intense f i e l d s about the grid wires when a high alternating potential was used. The energy of detachment was about 0.9 volt as estimated from the f i e l d s . The great defect i n Gravath's method lay i n having to use electron-mobility data fron measurements not made on the same gases with which he was working. I t also suffered from d i f f i c u l t i e s involved .in the use of the absolute values of the currents, to the grid and plate which required correction for loss of positive ions to the grid wires, the f a i l u r e to cap-ture a l l the electrons, etc. Bradburyt 1 7) i n 1933 perfected an elaborate and beautiful method. He used two grids at distances x 1 and xg from the source. He used an outgassed a l l glass system (giving greater purity than previous workers had achieved) and photo-electrons instead of the contaminating hot filament. F i n a l l y he used a method of determining electron mobilities which gave results to far higher values of X/p than before and which were taken i n the same chamber on the same gas and nearly concurr-ently with the attachment data. The method Involved an equation given by J . J. Thomson(18) rel a t i n g the photoelectric current between p a r a l l e l plate electrodes to the mobility of the elec-trons and their i n i t i a l energy (which was also determined i n the same tube). The use of the two grids eliminated the errors due to f a i l u r e to capture electrons, arid due to the capture of ions, by giving similar values due to these losses so that the v i i . r e l a t i v e current values for x^ and x 2 gave the r e a l r a t i o of ions and electrons. The results showed no pressure variation i n Og "but showed that the curve for h as a function of ener-gy f e l l rapidly to a minimum at about 1 volt electron energy , and rose again at about 1.62 v o l t s , reaching a maximum and then f a l l i n g again. The r i s e at 1.62 volts was due to the fact that at 1.62 volts i n Og the electrons lose energy i n i n e l a s t i c impacts and thus suddenly have an increase i n h as the energy i s decreased by i n e l a s t i c impacts. F i n a l l y the average energy of. the electrons increases again as they pick up energy i n the f i e l d after the loss of the 1.62 volts and the curve ultimately f a l l s . Bradbury's values agree with the p a r t i a l results of Bailey(3-4) and of Cravath (-^ ) where they overlap. Bradbury also observed the detachment of electrons at high grid-wire f i e l d s observed by Cravath and at about the same energy. It i s possible that this may lead to a method of obtaining the electron a f f i n i t i e s of neutral molecules on a . true energy basis. Recently Bloch and Bradbury( 1 5) explained the form-ationof negative ions by electron capture i n gases, i n which a dissociation process does not occur, by a unimoleeular pro-cess involving the excitation of molecular vibration levels and the subsequent loss of energy by c o l l i s i o n or resonance. In order to obtain a proper order of magnitude to agree with experimental observations they assumed a change of only one vibrational quantum number. This sets an upper l i m i t on the electron a f f i n i t y . For the case of Og this l i m i t i s 0.17 v i i i . v o l t s , consistent with other observations. The theory also yields a dependence of the attachment probability h on the average electronic energy. This dependence i s i n agreement with "Bradbury's observations i n the energy range .2 to 1.2 volts ( i . e . u n t i l the electrons start making i n e l a s t i c impacts). I t , however, shows a maximum for h below the range of Bradbury's observations at .11 v o l t . I t i s the purpose of this work to look for this (17) maximum i n h, using, essentially, Bradbury's two grid method and a recent method of h i s , to be described l a t e r , for meas-uring the mobility. 1. CHAPTER II THEORY l e t c = random velocity of electrons, W= d r i f t velocity of electrons, 1= number of electrons' per c c , A= mean free path, of electrons, Immobility of electrons i n cm. /sec .Jvolt./em., 1= free electronic current passing through the gas at some point x, X= f i e l d strength i n volts/cm. Then the number of c o l l i s i o n s of an electron per sec. C- _c_ i s \ and the number of c o l l i s i o n s i n a cm. of d r i f t i s W A » Thus the chance of an electron attaching to a molecule i n h c j r d r i f t i n g a distance dx i s \J\ then &I = - I -^fxcLx but by d e f i n i t i o n W - K X Integrating between two points (the two grids) x^ and xg, a distance x apart, gives / Iu hex . - - (i) where I^ and Ig are the free electronic currents at x-j_ and Xg respectively. It has been customary to employ the mobility equation to evaluate c and A for equation (l) . r\ ~ 3(r0 ^yTc as k i s in the units cm./see .^volt/cm. and not erne/see .^st. v l t / Then substituting i n (l) we get , _ 3 av /m k^X , I, ^ 2. CHAPTER I I I APPARATUS (a) Tube. (i) Construction. It was f i r s t intended that the tube should be as nearly as possible the same as that which Bradbury^ 7) used. However it, was thought that the revolving gri d and guard ring combinations would, make the enclosing bulb too large for the available glass blowing f a c i l i t i e s so I t was decided to constr-uct the apparatus with two fixed grids (see f i g . 1). An anal-ysis of the operation of the tube showed that this construction would give, satisfactory operation. To further simplify the construction i t was decided to put the guard rings P, G-, H, I, J, ( f i g . 1) on the outside of the tube containing the rest of the apparatus. Also, as some d i f f i c u l t y was experienced i n making a T tube to admit u l t r a v i o l e t l i g h t , the zinc plate as a source of photoelectrons was replaced by a filament B. Even with the hot filament, how-ever , the system could s t i l l be outgassed quite s a t i s f a c t o r i l y . I t was found that sealing the connecting wires through the side of the pyrex tubing strained the glass too much, giving i t a tendency to crack. To avoid this the tube was de-signed to have seven connecting wires leading through the base and one (to the quadrant electrometer) through the top. For reasons to be mentioned la t e r i n the discussion of the o s c i l l a t o r the grids should have a small capacity. Also F i g . 3. for s i m p l i c i t y the number of lead In wires through the base should be kept as small as possible. For these reasons the metal supports for the grid wires were not connected, l i k e a guard r i n g , to the potential divider but were connected to the o s c i l l a t o r just as the wires themselves. The grid design appears1 i n f i g . 3. The grids and other internal parts are made to f i t i n a piece of 2-inch pyrex tubing. The grids are made a f a i r l y close f i t to ensure correct placing when the tube i s finished. One set of wires A i s mounted on a piece of thin sheet brass B, the other set C i s mounted on another piece of brass D, B and 3) being fastened together to a piece of sheet mica E . In order that both grids s h a l l be most e f f i c i e n t under the same conditions and for other reasons to be given la t e r they are made i d e n t i c a l . Each grid A, B, i s supported by i t s two lead i n wires , ( f i g . . 2;)tc-The. filament C i s an oxide coated filament taken from a radio tube and Is supported by i t s two connecting wires. The plate D i s used as one end of the uniform f i e l d as, without i t , the filament would probably dist o r t the f i e l d . The connecting and supporting wires can be spot welded or brazed to the tungsten sealing wire: i n the f i r s t case they should be made of iron as copper doesn't weld to tungsten. Fi g . 2 i s made as a uni t , wires being welded or ' brazed to the outside ends of the tungsten before sealing i n glass. The f i r s t grid B should be near enough the filament C so that not too many electrons are attached before reaching i t as i t i s desirable to have most of the attachment between the grids, where the currents are measured. Also B should be far enough away from the filament to ensure that the electrons * make enough c o l l i s i o n s to acquire their constant average energy before passing B. For this reason CB was chosen about 1 cm. and BA about 2 cm.; the other distances are not c r i t i c a l at a l l and were chosen to be convenient. A l l the tungsten wires are then sealed through the end of a glass tube about an inch i n diameter. This tube i s then sealed to the drawn down end of the 2-inch pyrex tubing. The collecting plate i s made to f i t closely inside the tube, attached by a connecting wire to a tungsten seal, and is sealed to the drawn down end of another piece of the 2-inch tubing. The open ends of the tubing are sealed together so that the coll e c t i n g plate i s about 1 cm. from the grid D ( f i g . 1). This l a s t seal i s much simpler i f the two surfaces f i t perfectly, having been formed o r i g i n a l l y by cracking the tubing with a hot wire. A tube i s provided as i n f i g . 1 to connect to the evacuating system, ( i i ) Operation. Referring to f i g . 4, a filament B serves as a source of electrons. The guard rings F, G, H, I, J, K, together with the plates A and E, maintain a uniform variable f i e l d i n which the electrons from B travel i n their passage through the gas to the collecting electrode 'E. Some of the electrons are captured by neutral molecules en route from B to E. C and D are grids such as that f i r s t used by Cravath; ^' Between alternate wires of each of these grids a high frequency alternating f i e l d may be produced by means of the o s c i l l a t i n g c i r c u i t L (see l a t e r ) . Such a grid serves to separate electrons from a mixed current of negative ions and electrons. Electrons "reaching the grid as they move through the space between A and E are swept out to the grid wires during one-half cycle of the alternating f i e l d , but negative ions, whose mobility may be 1000 times less, do not have time to reach a grid wire i n a single half cycle, and the continual reversal of the grid f i e l d permits the driving f i e l d to carry them beyond the influence of the gri d . The filament and the grids are maintained at a mean potential equal to that of the guard ring i n which they are placed and consequently do not disto r t the main f i e l d far on either side. In actual practice conditions are not as ideal as des-cribed above since some electrons may escape and some Ions may be caught by the grids. I t Is to avoid this d i f f i c u l t y that two grids C and D are employed, which enable the electronic comp-osi t i o n of the current stream to be studied at C and D. The actual measurements are carried out by measuring the current I Q of ions and electrons when the high frequency potential i s applied at neither g r i d . The high frequency potential i s then applied at C and the current i - ^ of ions at C i s measured as the remaining current. The high frequency i s next cut off at C and applied at D, and the remaining current i g of ions at D i s measured. A careful symbolic analysis showed that.using two grids thus eliminated a l l the serious errors of the single grid and to potentiAI t i g . 5. The Oscillator + /2 0 -F i g . 6. 6. that the electronic current r a t i o i n the theory on page / X S M * * m—T To determine h i t remains to measure, the electron mobility. ( i i i ) Measurement of Mobility. The method Bradbury(16) describes i s available, but a better one and more convenient i s that given by Bradbury and Nielsen(20) and extended(21) to oxygen (see below). The principle i s the application of the grid electron ion f i l t e r s to the e l e c t r i c a l shutter method of measuring ion mobilities f i r s t used by Van de G r a a f f ^ 2 2 ) . Both grids are connected to the o s c i l l a t i n g c i r c u i t (q.v.) at the same time as i n f i g . 5. The voltage on the grids i s less than that necessary to sweep the electrons ent i r e l y out of the f i e l d to the grid wires and instead allows them to pass when the al t e r -nating grid voltage Is low or zero and cuts them off when i t i s high. Thus a series of pulses of electrons pass through C to D. Now D has an alternating f i e l d i n phase with that of C and acts i n the same way. Thus the average current passing 33 w i l l depend on the phase of the h. f. voltage on D when the pulse of electrons reaches i t . I f the o s c i l l a t o r frequency be now adjusted to make the collected current a maximum, then the maximumof the electron pulse must reach 33 when the o s c i l l a t o r y voltage i s zero, and w i l l be a maximum i f the maximum of the electron pulse reaches D when the o s c i l l a t i n g voltage i s zero. Now as this maximum of the pulse l e f t C when the o s c i l l a t i n g voltage was zero, the electrons must have passed from C to D i n one half cycle or some multiple thereof. Using the lowest frequency at which i t i s possible to get a maximum i n the collector current ensures that the time occupied between C and D i s one half a cycle. As the distance between the grids and the o s c i l l a t o r frequency aie known, the d r i f t velocity and mobility of the electrons can be immediately calculated. This method i s not quite as satisfactory i n gases which attach electrons (such as oxygen) as i n gases which do :,i not attach, but by working at somewhat reduced pressures oxy-gen, etc., can also be investigated. Under these circumstances there appears a negative ion background current upon which i s superimposed the electron current to be measured. When the -4 -6 probability of attachment i s of the order of 10 to 10 , the background current can be reduced s u f f i c i e n t l y to permit accurate measurement of the electron current maxima by restr-i c t i n g the pressure to values between 2 and 10 mm. When the measuring tube was constructed, i t was i n -tended to either use this method for measuring mobilities or to use the data of Brose^ 2 2)and of Nielsen and Bradbury^ 2 1). The o s c i l l a t o r has not yet been adapted to these measurements, (b) Osc i l l a t o r. The o s c i l l a t o r i s a tuned plate-tuned grid type of conventional design. By changing the tank c o i l s i t osc i l l a t e s at 10 to 100 metres. A wavemeter was made and i t and the os c i l l a t o r calibrated against a standard wavemeter. As the os c i l l a t o r was of low power i t was desirable to have the largest possible voltage induced i n the o s c i l l a t -8. ing c i r c u i t I so that a large variation i n voltage would be available for operation of the grids. To achieve this the grid capacity was used wi th the pick up c o i l as a resonating c i r c u i t , ' the number of turns of the c o i l was arranged to make the c i r -cuit o s c i l l a t e at nearly the desired frequency and then the o s c i l l a t o r frequency was adjusted for resonance which could be detected by the meter i n the plate c i r c u i t of the o s c i l l a t o r or by a neon lamp near the grid leads. A more convenient method of detecting when the o s c i l l a t o r was tuned at the correct frequency would be through the use of a high frequency milliammeter i n the o s c i l l a t i n g grid c i r c u i t . To get maximum voltage i n the grid c i r c u i t the coupling c o i l E should have as many turns as possible for resonance, thus the grid should have no unnecessary capacity. In order that both grids s h a l l operate most ef f e c t i v e l y with the same coupling c o i l and at the same frequency they thus should be i d e n t i c a l . I t was found that the tuned plate-tuned grid o s c i l l -ator was not as convenient as one with a single frequency con-t r o l as the Hartley. The tuning of the o s c i l l a t o r to the coupled c i r c u i t L necessitated simultaneous adjusting of grid and plate capacities to vary the frequency and to keep the os c i l l a t o r i t s e l f tuned, and made the resonance point in the coupled c i r c u i t d i f f i c u l t to detect as the o s c i l l a t o r frequency could not thus be quickly swept up and down. For measuring the mobility the required o s c i l l a t o r frequency i s much less: from 800 to 4000 meters wavelength. For this purpose It was intended to use a more powerful Hartley To face page 9. to e Itctt-ometef I hi via tot-F i g . 7. o s c i l l a t o r which was available. It would make frequency variat-ion easier and the grid potentials wouldn't be so c r i t i c a l to control. (c) Quadrant Electrometer. The electrometer used i s of the Dolezalek type with a s e n s i t i v i t y of 2000 mm. per v o l t . I f necessary this could be increased by the use of a finer suspension. I t was set, togeth-er with an earthing key ( f i g . 7) with mercury contacts, i n a metal shielding box as described by Makower and Geiger (24) t T h e wire leading from the collecting plate in the tube was also shielded. To adjust the electrometer, the quadrants, the needle and the ease are earthed. The needle i s then adjusted to take a symmetrical position with respect to the quadrants. The needle i s then connected to a source of potential (100 or 200 volts) above earth through a large (about a megohm) resistance for protection i n case of a short c i r c u i t . Charging usually dis-places the needle from i t s zero position to which i t i s returned by adjusting the leveling screws and then discharged. This, i s repeated several times u n t i l there i s no appreciable change i n the position of the needle on charging. As the s e n s i t i v i t y of the instrument increases with needle potential to a certain point and then decreases with further increase, the needle was given the potential of maximum se n s i t i v i t y at about 150 volts and the electrometer was ca l i b -rated for deflection against quadrant voltage. It was thought that the current might be measured by 10. a direct or a n u l l method using some electronic device (some of the high amplification tubes now on the market) rather than, by the time rate method of the quadrant electrometer. However the high frequency potential in the grids would be picked up easily on the collector plate and i t would be quite a problem to eliminate i t i n the amplifying system, (d) The vacuum system. The vacuum system consisted of two parts (in function): the system for evacuating the tube and the system for purifying and admitting pure Oxygen to the tube. See f i g . 8. The vacuum pumps were a double stage mercury vapour diffusion pump backed by a Genco hyvac rotary o i l pump. A Mcleod guage measured pressures from 20 mm. to 10~^ mm. and gave an estimate at 10"^ mm. A mercury manometer measured pressures from 10 mm. to atmospheric pressure. A U tube trap which could, be Immersed i n l i q u i d a i r was provided to prevent mercury vapour from the pump and guages from diffusing to the tube and contaminating the gas. An e l e c t r i c heating oven for the tube which could be moved In and out of place, was wound on asbestos over a thin metal tube and well insulated from the outside with asbestos. A second c o i l was provided inside the asbestos covering to act as a resistance thermometer. The tube was baked out and outgassed at about 125° G. as the solder softens at higher temperatures. The filament was run well above operating temperature during the outgassing procedure to avoid contamination la t e r from i t . The oxitgassing 11. was,carried on to 10~ s mm. The Oxygen pu r i f i c a t i o n system consisted of an i n l e t to take Oxygen from a commercial tank and an'escape valve E "(an 80 cm. tube dipping into a dish of mercury) which could be cut off i f pressures above atmospheric were required. The Oxy-gen then passed over PgOg i n a drying tube F and could be con-densed into a bulb by a c o i l G- immersed In l i q u i d a i r . If nec-essary for condensation the pressure could be raised somewhat above atmospheric. When the bulb was f u l l of l i q u i d Oxygen (or as much as was required) i t was cut off from the tank and was ,: fr a c t i o n a l l y d i s t i l l e d . The f i r s t t h i r d was pumped off v i a I, then the middle t h i r d was stored under about two atmospheres pressure, measured by a mercury pressure guage L (with a closed end and a i r inside), i n J, the l a s t t h i r d was again pumped off. H was another escape valve which could be cut i n i n case of trouble. The l i t e r bulb J also contained some PgOg to ensure dryness. When the tube was to be f i l l e d with the pure Oxygen the l i q u i d a i r was put around the c o i l s It and M and the gas was slowly admitted from J. Passing through the l i q u i d a i r traps K and M" completed the purifying process. JL2 © CHAPTER IY CONCLUSION Owing to considerable technical d i f f a c u l t y i n con-struction of the tube which is the important part of the apparatus, the work was not completed. Apart from a few minor details the apparatus has a l l been constructed, but no operating runs hare been made, I would l i k e to thank Dr. G. M. Shrum of the Physics department for suggesting the problem and for suggestions i n the course of the work, Also I would l i k e to acknowledge the assistance of Mr. ¥. R. Eraser, the mechanic in the Physics department, i n construction of the grids. BIBLIOGRAPHY (1) L.B. Loeb, "Kinetic Theory of Gases", McGraw-Hill,613 (1935) (2) A.F. Kovarik, Phys. Rev., 30, 415 (1910) (3) E.M. Wellisch, Am. Jour. Sc., 39, 583 (1915) (4) E.M. Wellisch, Am. Jour, Sc., 44, 1 (1917) (5) E.M. Wellisch, Proc. Roy. S o c , A134, 427 (1931) (6) J. Franc k and Pohl, Verh. der Deuts. Phys. Ges.,IX, 69,(1907) (7) Rutherford, Proc. Camb. P h i l . Soc, IX, 401 (1898) (8) L.B. Loeb, Phys. Rev., 17, 89 (1921) (9) H.B. Wahlin, Phys... Rev., 19, 173 (1922) (10) J.J. Thomson, P h i l . Mag., 30, 321 (1915) (11) 0. Oldenberg, Phys. Rev., 43, 534 (1933) (12) L.B. Loeb, P h i l . Mag., 43, 230 (1922) (13) L.B. Loeb, Jour. Frank. Inst., 197, 45 (1924) (14) V.A. Bailey, P h i l . Mag., 50, 825 (1925) (15) A.M. Cravath, Phys. Rev., 33, 605 (1929) (16) J.S. Towns end and 'i'izzard, P r o c Roy, Soc, A88, 336 (1913) J.S. Townsend, Jour. Frank. Inst., 200, 563 (1925) (17) N.E. Bradbury, Phys. Rev., 44, 883 (1933) (18) J.J. Thomson, "Conduction of E l e c t r i c i t y through Gases", University Press, Cambridge, 466 (1928). (19) F. Bloch and N.E. Bradbury, Phys. Rev., 48, 689 (1935) (20) I.E. Bradbury and R.A. Nielsen, Phys, Rev., 49, 388 (1936) (21) R.A, Nielsen and N.E. Bradbury, Phys. Rev., 51, 69 (1937) (22) Tan de Graaf, P h i l . Mag., 6, 210 (1929) (23) H.L. Brose, P h i l . Mag., 50, 536 (1925) (24) Makower and Geiger, "P r a c t i c a l Measurements i n Radio-a c t i v i t y " , Longmans, Green and Co., London, 9 (1912) 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items