Further Investigations of the Penetrating Radiation Citation Cameron, George Harvey (1926) Further Investigations of the Penetrating Radiation. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/6HV7-DP02. https://resolver.caltech.edu/CaltechETD:etd-06022004-142013 Abstract No Abstract. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: Physics Degree Grantor: California Institute of Technology Division: Physics, Mathematics and Astronomy Major Option: Physics Thesis Availability: Public (worldwide access) Research Advisor(s): Unknown, Unknown Thesis Committee: Unknown, Unknown Defense Date: 1 January 1926 Record Number: CaltechETD:etd-06022004-142013 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-06022004-142013 DOI: 10.7907/6HV7-DP02 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 2384 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 02 Jun 2004 Last Modified: 03 Oct 2019 23:17 Thesis Files Preview PDF (Cameron_gh_1926.pdf) - Final Version See Usage Policy. 968kB Repository Staff Only: item control page

Further Investigation of the Penetrating Radiation. Thesis by. G. Harvey Cameron In partiel fulfillment of the requirements of the degree of Doctor of Philosophy. CALIFORNIA INSTITUTE OF TECHNOLOGY Pasadena, Celifornia 1926 bed fH f% the time when the experimerts to be deser were planned, a decided lack of agreement appeared to xist between the results of verious observers. ALL were agreed that the ilonizetion in closed vessels increased with altitude but there was a considerable ariation in the rates of inerease announced. Further- more, screening experiments by Kolhorster in Switzerland led him to support more strongly then ever an extremely penetrating radiaticn of extra-terrestrial origin, while the results obtained by Millikan and Otis were interpreted by them as indicating that all or by far the greater part of the effect was due to radiation cf no extraordinary penetration. The course to be pursued was not hard to see. If doubt were to be cast upon the experiments of Kolhirster it would be on two principal counts in all probabliity. First, the sereening experiments in water were made in shallow pleces at Low altitudes where the whole effect is small in comparison with that due to other causes of ionization. Second, in the experiments in the glacier one could not be entirely sure thet specicus results were not obtained as a result of rocks and waste matter eonteined in the ice. Hence we planned to go to the high- est altitudes available where deep, snow-fed lakes could be found and to use the lake water as screening material. APPARATUS In the werk to be described three electroscopes of a modified Wulf type were used. Details of the const- + Co) 9 k$— /f— Qajustable ir Condenser. ructicn are given in the following teble. Noel No. 2 Now Height £200 CMe J5.0 29.6 Diamster 11.4 20.0 15.0 Thickness of walls - 1,8 mm. & mm. Sma. Volume effective for ionization 1895 ec. 11826 cc. 211 ec. Capacity of fibres 1.41 esu. 1.41 1.10 esu. Reduction constant 1.45 0665 Sensitivity (approx) volts per scale div. 1.2 | Le 1.8 The Lonizgeticon chamber is of brass in all cases except that the top of No. 1 is cast zinc. To determine the cap.citi es of the fibre systems the following procedure was adopted. An apparatus was constructed (see fig.) which consisted of a small plane condenser with the upper plate moveble. The lower plate fastened directly to the charging rod of the instrument by a ball joint. The upper plate was mounted in a rigidly-held, tight-fitting bearing so that it could be raised or lowered at will. A micrometer screw enabled the distance between the plates to be altered by any given amount. The plates were adjusted accurately parallel to one another by lowering the upper one and pressing them together. The ball joint on the lower then permitted it to rotate until they were completely in contact. ve : The capacity of this condenser is given by the following formula, where t is the thickness of the plates, r the radius of the plates, and @ the distence The procedure involved finding first the capacity of the fibres plus the charging rod plus the condenser in the extended position. Call all this C,. C, was a & charged to a potential Vo » read on a voltmeter. Then while C, was isolated from the source of potential the upper plate was lowered to within a@ distance a of the lower. Then the fibres came together somewhat and indicated a potential V, . Hence if G is the capacity of the small condenser when the plates are a distence ad apart, we have boats Vy og Ma . and henee ©, could be determined. In the second part of the problem two distinct methods were used. In the first the fibres alone were charged to a high potential, about 300 volts, by connecting them to the battery through the charging rod and then earthing the rod. This potential was read from the voltmeter. Then the rod was swung beck into contact with the fibres and the final potential detcrmined from the position of the fibres. In the other method the fibres and rod were in- itially uncharged, the rod being just beyond the charging position. Then the rod alone was charged to a given potential read on the voltmeter. Then this charge was spread over the fibres and rod, the final voltage being read from the position of the fibres. Since the combined capacity had been determined in the first part of the determination, the capacity of the fibres alone was readily calculated. The method of measurement was preciselp that used by Otis EXCEPT THAT MUCH LONGER PERIODS OF OSSERVATION WERE GENERALLY USED. At various times the insulation of the fibres was tested by the guerd 4 ring method described by Otis and no leak could ever be detected so that in most cases this preceution was abendoned. AS IN OTIS' YORK, A CALIBRATION CURVE WaS DRAWN FOR EACH READING TO AVOID ERRORS DUE TO CHANGING CHARACTERISTICS OF TH! INSTRUMENT. That saturation was obtained was best indicated by the fact that for long observations where the potential fell to as low as 50 volts, the ionization wes not apprec- iably less than for an observation under the same circumstances lasting only one-fifth of the time. COTTONVOOD LAKES TRIP. In the summer of 1924, the author with the help of ir. C.H.Prescott Jr., and Mir, Alvin Viney, organized and conducted an expedition to the Cottonwood lakes in the Sierra Nevada mountains. The lake chosen for the work was at an elevation of 11200 feet. The avparatus used was electroscope Mo. 2. This instrument consisted of the frame and fibres of No. 1 enclos 3d ina large brass ionization chamber (volume about 11 litres.) The inereased ionization due to the much lerger volume was expected to increase the eccur- acy of the readings especially over short time intervals. In practise, however, the instrument proved undesirable for the following reesons: 1. Temperature changes caused convection currents inside the electroscope which greatly disturbed the measurements. 2. It was difficult to obtain saturation. 3. The electroscope wes heavy and bulky. 4. Due to the large ratio of volume to area of walls and other causes 5 the fonization produced with this instrument per cc, per second was much smaller than with the others used and hence the relative error was large. On account of these difficulties combined with ine clement weather, wholly satisfactory results were not obtained and the experience in organization for high el- titude work gained through this trip was, no doubt, the most theless 7 valuable result of the expedition. Nevere- the observations showed quite clearly — 2 1. A decided increase in ionization with altitude. ee No variation with time of day. 3. A considerable decrease over the #2 DEPTHS PRESENCE in ionization when the electroscope was read water. 4. AN APPRECIASLE DECREASE IN IONIZATION BELOW ONE METRE OF WATER INDICaTI uG THE OF A RADIATION MORE FENETRATING THAN ORDINARY GAMIIA RAYS. A series of 13 one-hour consecutive readings taken with the same instrument on Mt. Wilson{altitude 1760 m,) in the latter part of August 1924, also checked points Nos. 1 and 2 above. LAKaS MUIR AND ARROWHEAD. In the winter of 1924-25 a new electroscope, No. 3, was constructed, tested and calibrated and in August 1925 a second expedition organized consisting of Dr. Millikan, Dr. Oldenberg, Wr. Prescott, Mr. Glenn Millikan and the author. The work this time was cerried on at Muir Lake on Lone Fine Creek near the base of Mt. Whitney at an elevation of 11785 feet. & folding canvas pentcon furnished by the Military 6 Department of the Institute was used for transportation on the water. For the underwater readings the electro- scopes weresuspended from a small raft anchored: in about 40 metres of water. Granite cliffs varying in height from five hundred to two thousand feet at a distances up to three-fourths of a mile surrounded the lake on three sides, The lake itself was about one-fourth of a mile in diameter and roughly circular in shape. On the return from Muir Lake, readings were also taken for a day on a large flat rock in the desert near Lone Pine at an altitude of about 5500 feet. Finally the Muir Lake observations were repeated at Lake Arrowhead in the San Baernardino mountains at an altitude of 5125 feet. The depth of the lake at the point of observation was about 50 metres. RESULTS. The results of the underwater readings are collected in Table 1. For our Burposes it is convenient to regard the atmos- phere as the equivelent of 10.43 metres of water and to consider a given locality, such as Muir Lake, in terms of the water equivalent of the air above this point, The. pressure due to a layer of air of thickness diHis dp«~paii where ¢ is the density. But pak Ply . Then es po T is a function of H but to avoid complication we will assume it to be a constant and substitute a suitable mean value over the region H concerned when we eval- Muir Lake surface .45m 15.3 9.7 13.2 means 15.25 9,7 Arrowhead surface .7m 7.0 5.8 V.2 7.5 / 6.9 / (or ? means 7.0 5.8 Mair Lake surface ,.45m 16.5 13,85 15.8 15.9 means 16.1 13.5 Arrowhead surface .6m 10.5 9.35 means 10.75 9.355 Table No. 1 Electroscope No. 3 im lim dm 5m 5.5 5.15 4.85 4,4 / 4.9 j f f / y Electroscope No. 1 im 2.8m 11.0 9.90 11.0 9.0 Yom Cem tom on > DID WY Lom 3.6 566 GIG GW Onmaoag 7 uate the integral. Then Og fe = EF at AA vi or pe Ae F - Now when - 76%13-6 ana Fe £73 se & om wet Gets 2 : ° ry Therefore pb» Aa ** ur if H is measured in kn. But when “=O . bw /O:53 ” Hence 42/33 and bu 10.53 € where T is the mean temperature over the height H. Now we assume the following temperature distribution. H (km) % (GC) T (abs} 0 22 295 1° 18 291 2 14 287 3 10 283 4 6 279 5 2 275 This distribution, corrected for the average ground temperature, corresponds closely to that observed experimentally in balloons near Paris, according to the Smithsonian tables. The curve connecting altitude with depth of water, shows Muir take at a depth of 6.79 metres of water below the surface of the air, Lone Pine 8.5 m, Arrowhead &.6 m., and Pasadena 9.98 m. Now Muir Lake and Arrowhead differ on this scale by almost exactly 2 metepes. Hence whenever we have a reading in Muir Lake two metres below one at Arrowhead we should expect them to agree. Consequently it is very reassuring to find in the four possible instances where this point could be tested, pointed out by arrows in Table 1, the beautiful confirmation shown there. g The results of the water work are also displayed in the two curves. It will be noticed that at both Muir Lake and Arrowhead, readings taken at the surface come above the curves and this is to be expected because the readings at the surface must be partly due to local rediation of ordinary penetration. Below one metre, however, we should have only the ionization contributed by the instrument itself and the penetrating radiation. This is because the radiation which has the penetration of gamma rays is so reduced as to be practically negligible by one metre of water. The extraordinarily good agreement of the results testifies to the correctness of these views as well as to the validity of the readings themselves. COMPUTATION OF ABSORPTION COEFFICIENT, _ In. computating the absorption coefficisnt from the above data, two alternatives present themselves. In the first place, we may assume that the radiation comes into our atmosphere as a beam of parallel rays or On the other hand, we may assume that the rays are incident to the surface of the atmosphere equally in all directions. In the first case we would use the | | = — fs C8 well known law of absorption Loe £5 @ Tf the radiation did come into our atmosphere as a parallel beam of rays, however, we should expect well-marked variations with the time of day. For, on account of the earth's rotation on its axis, when the rays are coming in at an angle of * degrees say, With the vertical, they would have to go through a layer d/cos#, where @ is the vertical height g of the atmosphere at the point in question. Even if the source of the radiation is not a small region of the heavens but a band such as the Milky Vay, we should expect easily observable variations in the intensity measured at Muir Lake or Ht. Whitney. Now this point has been repeatedly tested not only in the present experiments but also by Otis on Mt. Whitney and by Otis and Millikan on Pike's Peak, and we have never been able to find any indication of a regular variation. Consequently we appear to be justified in using the second alternative given above. Consider the radiation coming from above only, and let its intensity at the surface be Ig. Then the part of I, due to the radiation included in a cone- shaped shell whose angle with the vertical is #@ anda of anglular thickness d@ is, lp H inh tan d «COS & H* cas" 8 The intensity at the depth H in the absorbing medium due to this part of the radiation is f° mt oo a FF GOS aie 257, sind ab e-™ e « =m pH Sea O é am Ff 2 ? r i . 4/7 we elf i fe Sta BE ae i f Now put ec fax Then sec *@ frat Ao = BK de , Se # F Ke 4 3 va a eo ' SO kw a B; Values of the integral Lf a a Aa A are given by Gold - Proc. Roy. Soc. & + 82,62 1909 and these were used in this computation. — 10 The method of determining the absorption coeff- icient was to select the most reliable observation near the top of each curve and to see what value of he best represented the other reedings when compared with it. A little calculation shows that the radiation becomes more penetrating with depth. In other words, the radiation is not homogeneous but consists of a spectrum of wave lengths. For both instrumentsthe upper portion of the curves gave w= 0.30 per metre / of water and the lower ends gave ¢= 0,18. Electroscope ¢t 7 © NO. & W&S more sensitive however, and geve somewhat more reliable results throughout as a superficial examination of the data shows. These coe*ficients, of course, Characterize the radiation only throughout the region studied. Ye might expect still softer compon- ents t at higher altitudes. RESULTS OF LEAD SCREENING In addition to the under water observations, the same lead screen used by Otis, enabled us to obtain additional information. The lead was used with Hleetroscops No.l only end was the equivalent of 55 em. of water. The results of these experiments are collected in the following table. Table No. Cosmic Rays inside 4.8 cm. Fasadena “Lone Pine Muir Lake eltitude altitude 605 m. Tons per cc. ber see. unshielded 13.0 Shielded with 4.8 cm. lead 9.0 External radiations after screening 1.6 Cosmic Rays after sereening (theoretical ) 1.3 Cosmic Rays after sereening (obs, 1.32 2 1676 m. 16.7 10.1 ae” lead. altitude 3590 me 20.0 11.8 444 4,08 Pike's Peak altitude 4298 m. 12.4 eR ad we fw il The first aud second lines contain the readings without, and with the lead shield respectively. From the latter values we subtract 7.4 I which we found due to the walls of the electroscope alone, obtain- ing the values given in the third line, These figures consequently represent the whole effect inside the shield of all external radiations of any kind. Now all the experiments with the lead were made with a view to having the surroundings as nearly the same as possible. In every case the instrument was placed either on granite rocks or on soil which was largely decomposed granite. Consequently we should not expect a large veriction in the etfect due to the surroundings After consulting as many authorities as possibls we took 3 I as a fair mean value and assumed thet thés amount might be attributed to Local causes in ell these places, Now the amount of this rediation which will penetrate the shield is just 0.5 I. Hence we subtract 0.5i from the last figures given and obtain the values given in the fourth line which we calli the "theoretical" values of the cosmical rays after pene- trating the lead. However we have a possibilityof getting these values in another, quite independent way, and these we call "observed! They are obtained by teking the readings on the curve at each location, subtracting 7.4 I and calculating how much of the resulting radiation would penetrate the lead, using the coefficient 12 30 per metre of water and the formula [-Lew* Since now the shape of the shield ap proximates to & sphere and most of the radiation goes through very nearly perpendicularly. This calculation gives surprisingly good agreement with the preceding results as the table shows. Cur one assumption of a constant value for the lécal, softer radiation is not a serious one, Since it could be altered considerably without affecting the agreement very much. SOF? RADIATIONS Millikan and Otis, in commenting on the results of their Pike's Peak work called attention to the large part played by soft radiations in the total effect ob- served. The following table, No. 3, is designed to exhibit as strikingly as possible the properties of these softer radiations and to suggest how they may be accounted for. Some of the Millikan-Otis results are included in the last column for completeness. The first line gives the directly observed results in each location, The second series of values is taken from the curve and includes of course in addition to the cosmic rays, the zero value of the electroscope. The difference of the first two series then must give the ionization due to the softer rays. It is difficult to escape the conclusion that we have here a stimulated radiation of some sort. If we make the same assumption which led to the very satisfactory conclusion in the previous section, namely that 36 I are due to local gauses, we have a residue in every case which increases markedly with altitude. It is interesting to notice Table No. 3 Soft Radiations Mleetroscope To. 1 Faesadena Lone rine “uir Lake Tike'ts Peak Direct Observ- ation 15.0 16.7 20.0 2047 Cosmic Rays (from curve) 8.95 10.0 12.2 15 Soft Rays 4,05 6.7 7.8 10.4 Stimulated Soft Rays Stimulating Cosmic Rays (observed) 1.55 2.6 4.8 5.9 by comparison with the last series of values given, that the increase is roughly proportional to the intensity of the cosmic rays. It would scem probable, therefore in addition to the radiation with the penetration of gamma rays which we attribute to the ra dioactivity of the surroundings, THAY TEERE IS A CONSIDERASLE QUANTITY OF RADIATION OF ABOUT TEE SAME PENETRATIOL , STIMULATED Cua ot sQrooT Rs a ee "Th eytT mere T ata TO Tan pepe R ‘ RS IN SOME “VAY BY {38 PRIMARY, MOR: PLNETRACING RADIATION, SUMMARY 1. The present work seems to us to furnish conclusive evidence to show that a penetrating radiation does exist which a. increases in a perf b. is much more penetr ectly definite way with eltitude. ating than any known radiation. &. Although no very elaborate tests of this point have been made, we have been able to detect no variation of any kind. 6. In addition to the penetrating radiation assumed to be of cosmical origin, thsre is present in relatively large amounts, softer radiation with about the penetration of ordinary gemma rays. It is suggested that this is in part a secondary radiation stimulated by the primary, and some evidence is presented which seems to confirm this snegestion. } 6 fon Ionization in ions THA co per cc. per second. 12 rr 1 10 FE ! 1 | i. Electroscope No. 3 OQ Muir Lake x Arrowhead T H im im SE : “ iH kee rk | Distance below surface of atmosphere ; par aan HPEPEPE Pein Tr EEE | in meters of water, a a | ‘ll i i | EEELEEFE ! TUTTI Trier rrr terrier rrr rr irr rit 6 7 8 9 WO 4 12 138 4 15 16 17 8 19 20 21 22 23 24 25 26 27 { f Jiptip gt {fh pty 16 , ra CeCe tt Tonization in ions TH 15 ana per cc, per secona. Tt) 14 1 +o { im 1S i Electroscope No. 1 12 raphy men © wWuir Lake 11 | nan i oe : x Arrowhead = <U | ! fan me an ry oa 10 = SS pet ~ Ee EES SERS EPEEEEPEEEE EEE EEEEE EEE H Coit 7 i Ss Trt imi QFE oS LH 8 i — = 5 i] = rt 7 Pepi Distance below surface of atmosphere TTT; T cry in meters of water. ; HEE HH TOO CEC 6 7 & 9 10 11 12 «613 14 iS 16 7 18 19 20 21 22 23 24 25 26 27 28