The Disintegration of Lithium Citation Delsasso, Lewis Alexander (1937) The Disintegration of Lithium. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/91tm-t458. https://resolver.caltech.edu/CaltechTHESIS:11252024-223905841 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): Lauritsen, Charles Christian Thesis Committee: Unknown, Unknown Defense Date: 19 May 1937 Funders: Funding Agency Grant Number Seeley W. Mudd Fund UNSPECIFIED Record Number: CaltechTHESIS:11252024-223905841 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:11252024-223905841 DOI: 10.7907/91tm-t458 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 16879 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 26 Nov 2024 18:21 Last Modified: 26 Nov 2024 18:34 Thesis Files PDF - Final Version See Usage Policy. 13MB Repository Staff Only: item control page

THE DISINTEGRATION OF LITHIUM Thesis In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. Submitted By LEWIS ALEXANDER DELSASSO CALIFORNIA INSTITUTE OF TECHNOLOGY Pasadena, California May 19, 1937. SUM}!L.l\.RY An a tte mp t i s ma de to summarize the reasonably well esta o lished e xperi n:en tal result s on the disintegration of lithium with particular reference to the work done by the author in thi s f ield. The reactions di s cus se d are classi f ied accordin g to the disi nte gration products. Particular emphasis is p laced on the work on the gamma radi at ion produced f rom lithium under proton bomba rdment , including a detailed descri pt ion of t he experimental set-up and met hods used, anl a di s cussion of the probable orig in of.this gamma r a d iat ion. The subject i s summarized in a table c ompa ring t he e xp er i me ntal values of t be reaction ener g ies a nd t h e corresp on d ing ener g i es o bt ai 11ed fro m tbe mas ses invo lve d . THE DISINTEGRATION OF LITHIUM INTRODUCTION The first successful a ttemp ts at artificial disintegration were those of Rutherford in 1919. He was able to observe the disintegration protons from nitrogen when bombarded with alpha par t icles. Similar experiments on lithium by Rutherford and Chadwick( 33 , 39 , 4 o) and by Kirsh and Petterson( 23 ) proved unsuccessful. Rutherford and Chadwick, using a right angle method of observation that permi tted t hem to observe protons down to 3 cm. range found no effect when lithium was bombarded with alpha par t icles of 7 cm. range, and concluded that disintegrations of this kind did not t ake place. This conclusion has not been con- tradicted up to the present time, although as we s hall later see, disintegrations of lithium with products · other than protons do occur under alpha particle bombardment. The first disin tegration by ar tificially accelerated particles was that by Cockcroft and Walton(B) in 1932. Their disintegration of lithium by protons accelerated up t o about 500 K.V. marked t he beginning of a period of intensive work in nuclear reactions which has continued unabated to the present. In their first experiments they bomba rded a target of lithium placed at 45° to the direction of the beam, and observed the products of disint egration at right angles by means of a scintillat i ng screen and mic roscope. With this arrangement they were -1- able to observe disintegrations down to about 70 K. V. bombarding energy. Later Oliphant and Ruthe rford (33) using a Wynn Willi ams counter detected disintegration down to 30 K. V. and fi nally Traubenberg( 44 ) re ported po sitive obs e rva ti on s at 13 K. V. Cockcroft and Walton supposed t ha t t he scin tillat ions we re due to alpha - particle s. 1 Hl - 2 They suggested t he reaction He4 Q as their origin. Th ey a lso measured t h eir range s and checked t h e simultaneity of the emission of t he t wo par ticles. They t ook cloud chamber pictur e s and used an oscillograph in connection with a Wynn Willi am s chamber to check the nature of t he parti cles. From t hese data and energy considerations t h ey were able to conclude tha t t h eir hyp oth esis for the production of t he particle s was correct. Since Cockcroft's or i ginal report, much exp er i men tal work has been done on t he disintegra t ion of l ith ium -by a l phapar ticles, neutr ons, and by art ifi ci ally accelerated particles. Much of t his work ha s been done in t hi s laborat ory . t he work done el sewhere will b e discussed n ow . - 2- This and a lso DISINTEGRATION ALPHA PARTICLES -Li. 6 3 2) 3Li7 1) t -t 3) 31 l·6 t 4) 3Li7 4 5) 31 l· 6 + 1 Hl Hl - 2 He 4 2He4 4 2 1 H - 2He 1H2 --=-- 2He 4 1 0 nl -- - 2He4 4 2 He3 + Ql -t Q2 + Q3 4 ne4 2n 4 2He 2He4 -+ -1- 1H3 -1- -t 4 0 nl Q4 Q5 REACTION 1) The short range particles fro m this reaction were first reported by Walton & Cockcroft( 47 ). Dee ( l 6 ) showed from cloud chamber ·work that the particles were emitted simultaneously and in opposite directions. Oliphant, Kinsey and Rutherford ( 32 ) " studied this reaction in detail and obtained t h e values of 6 - 8 mm and 11.5 mm for t he ranges of 2 He 4 and 2He 3 respectively. They also found that the efficiency of production of the short range particles is 30 times greater t han t hat for t he long range particles fro m reaction (2) and the variation of number with voltage is the same as t hat for ( 2). Finally, Ol iphant, Shire and Crowther( 34 ) using se parated isotopes were able to show conclusively t hat t he s h ort range particles are produced in reaction (1). REACTION 2) This is t he reaction first studied by Cockcroft and Walton. Their measurement of the range of t h e al pha particles -3 - gave 8.4 cm. Oliphant C32 ) and coll abo rator s showed by mean s of se par a ted isoto pe s t ha t a s previously assumed by Cockcroft the source of t he 8.4 cm. par t icles is reaction ( 2). A precise measurement of the range of these particle s wa s made by Oliphant, Kempton and Rutherford( 3 l). Using air as an absorber and c ompar ing with t he alpha particles from Th c1 (8 .6 cm.), which in turn have been measured very precisely in a magne tic field, t h ey found t h e range to be 8 .29 ± .03 cm. af ter t he correction was made for t he bombarding voltages. REAC TION 3) This reaction was first reported by Lawrence ( 25 ). Due to his high bombarding voltage s he reported 14.0 cm. as t he range. Precise measurements by Oliphant, Kempton and Rut herford ( 3 l) using t he me t h od of comparison already described, ga ve 12.6 ± .05 cm. for t he range of these particles. Finally, Oliphant, Shire & Crowther( 34 ) showed by separated iso topes t he correctness of t he assumption as to t h e source of t hese particles. REAC TION 4) In t his case t here is a continuous distribution for t he alpha par ticles, since t here are three disintegration products. Oli phant, Kinsey and Rutherford ( 32 ) found t he u pper.limit to be 7.3 cm. The neutrons fro m this reaction will be discussed later. -4- REACTION 5) This reaction is reported for slow neutrons by Chadwick & GoldhaberC 6 ). They found the singly charged particles and doubly charged particles to have ranges of 5.5 cm. and less t han 1.5 cm. respectively. TaylorC 43 ) has found tracks of 6.64 ., ¼ .06 cm. equivalent length in pho tographic plates. Acc ording to Chadwi ck and Goldhaber, the cro ss -secti on for t h is reaction is 10- 21 cm: 2 and wa s, therefore, suggested by them as a means of detec ting slow neutrons. DISINTEGRATION NEUTRONS ( 6) 3 Ll.7 ... 1H2~ 2He4 + (7) 3117 + T-2 -- ·4Be 8 1li + 0 ( 8) 3116 + 1 H2 ---- 2He4 + (9) 3117 -t 2He4-.-5Bl0 -t He 4 -1- nl + 2He3 + nl -t 2 0 0 nl Q8 0 nl + Qg REACTI ONS (6) AND (7) The emissi on of · neutr ons from lithium under deuteron bombardment was first r eported by Cr ane, Lauritsen and Solton(l 4 ). They attributed the neutrons t o reaction ( 6). The ener gies of t hese neutrons and thos e of reaction ( 7) have been deter□ ined by Bonner and Brub aker C2 ). For t h~se measur ements t he y empl oyed a high pressur~ cloud chamber whi ch could be op er ated a t pr essure s u p to 15 a t mospheres of methane.~,((1). The results are s hown in figure The upper curve is the dis t ribution of recoil protons, the For t he higher energies a mica absorber had to be pl a ced in the ch amber. -5- dotted curve t h e distributi on of t h e primary neutr on after correcting for t he vari a tion in neutron- proton collision area. We see in this curve a cont i nuous distribution of energy (three body problem) with a most probable ener gy at 2.1 Mev. and a mean energy at about 3.9 Mev. There is a pronounced hump at approximately 13.0 Mev. su per- imposed on t he continuous distribution. This ha s been inter prete d by Bonner and Brubake r a s due to t h e for mation of 4 Be8 as given by . . reaction (7). The area under t his hump is about 5% of t h e total area under t he curve, indicating the react ion (7) occurs about 5% of the time. Calculati ons from these data give t he mass of 4Be 8 as 0.3 ± 0.75 Mev. greater than t hat of two alpha particles. REACTION 8) Although Oliphant, Shire and Crowther C34 ) have reported negative results f or reaction (8), Rumbaugh and Hafstad have shown t hat this reaction does occur. energy of t hese neutrons. No measurements have been mad~ on t h e This reaction is exothermic by only 1.3 Mev. so that we would not expect to find it a ppearing in Bonner & Brubaker' s work on t h e unse parated isotopes. ~:- REACTION 9) The emission of neutr ons under alpha particle bombardment was first observed by Both e and Bec ker( 4 ) • SnetzlBr( 42 ) has s hown t hat the energy of t h e alpha parti cles at which t his reaction begins is ver y close to 4 .7 Mev. This value i s in good agreemen t with t he * The neutrons fr om t his react ion woul d be buri ed under t he max i mum of t h e reaction (6) curve unless reacti on (8 ) occurr ed much more fre quently t h an ( 6). -6- find i ngs of Savel( 4 l)# of approximate l y 5.0 Mev. Meitner(30) has reported additional reactions wit h both 31i6 and 31i7 in which 5B9 and Bel0 are formed, these latter having positron and 4 beta activity respectively. However, careful work by Snetzler has failed to detect such activity, and one must conclude for the . present that Meitner's report is in error. The energies of the neutr ons fro m reaction (9) are not well known, values are given from 0.2 to 0.9 Mev. DISINTEGRATION PROTONS 10) 31i + 1 H -- 3 Li + 1 H + QlO 11) 3Li6 + 1H2-(3Li7)* + 1Hl + Qll The disintegr at i on protons from lithium were fir st reported by Lawrence( 26 ) and later by Cockcroft & WaltonC9) . Reaction (10) was offered for their formation. Cockcroft found, using a mixed beam and ordinary lithium target, only one group at 30.5 ± 1.0 cm. Later Oliphant, Shire & CrowtherC34) proved by the use of separated isotopes that this group came as postulated fro m 31i6. They were unable to find any proton group from Li 7. In an effort to find t he protons emitted in t he presumed reaction producing the radioactive 31i8 (which will be discussed later) it was decided to investigate carefully in this laboratory(lB) the protons from Li under 1 H2 bombardment. These exper iments were done in a cloud chamber in which the pressure could be varied to make the particles stop in the chamber . # The setup was I e ssentially These results will be furt her di s cussed in connect i on with the gamma rays. -7- as s hown in figu re ( 2), t he only dif f erence wa s t he a ddition of baffl e s to confine t he pr otons to with in 5° of t he pl an e of the ch amber. The sto pping powe r of t h e foil, s epar a ting t he t a r ge t from t he cloud c hamb e r, was de t ermine d by t wo me tho d s: by we i ghing , an d by measuring t he r e si du a l r an ge in the cloud chamber of t he a l pha particles f r om 3Li7 bombarded by pr otons. This r ange , a s has been de scribed, ha s b een ca refully determined. The stoppi ng powe r of t he foil (of appr oxima tely one - half mill co pper) was foun d to b e 4.7 cm. The stopping powe r of t he mixtur e of Ethyl alc oho l vap or an d a ir wa s det e r mi ned by t wo methods: (1) by det e r mining t he pr e ssure and temper ature and computing t he sto pp ing power ; an d ( 2) by mea sur ing the range of polonium a l ph a par ti cle s i n t he chambe r, which is well known. The stopp ing power f or al pha parti cle s and pr otons of other .energ i e s wa s comput ed- fr om data given by Mano (29 ). An exampl e of t he photogr aphs obta ine d is shown in f i gur e (3). The d istribution in r ange of the par ticles r e sulting from t h e disint e gr ati on of lithium by 700 K.V. ( peak ) de ut er on s is s hown in figur e (4). The extra pola t ed r a ng e s of 31.7 ± 0 . 5 cm . 13.8 ± 0.7 cm. and 8.9 ± 0 .1 cm. r e s pe ctive ly, fo r t he l onger range pr otons and t wo a l pha particl e gr oups a r e i n good agreeme nt with t he r an ge s whi ch ha ve be en mea sur ed a t Cambridge (7). The gr oup a t 26 f 1 cm. from r eac ti on (11) wa s no t rep or t ed pre vi ously. en ergy of t he pro t ons of this group is 4 .3 ± 0 .1 Me v. This gr oup wa s a tt r ibuted a t t he time to reac ti on (1 2) f ormlng 31 i 8 . ha s since been pr oven not to be c orrec t. The Thi s Eff or ts were made a t the t i me t o obta in separ a t ed isot ope t arge ts to check this conclusi on, but t he s e pr oved unsucces sful . Recently, L. H. -Rumbaugh and L. R. Hafstad(J?) have repeated the measurements of the proton groups, using separated isotopes. They found that both these groups come from the Li 6 reaction . From Li 7 they found no protons of ranges greater tri..an 8 cm.,* which could not be accounted for by contaminations . The possible implications of the two groups of protons from 3Li6 will be discussed later . DISINTEGRATION ELECTRONS (12) 3 1 J..7 .,. 1 H2- 1 31·8 J. .,. 1 H .,. Q12 (12a) 31i 7 + 0 nl _. 31i8 + Q (l 2a) (13) 118 3 4 Be8* -t- e- -t- )) -t- Q13 (14) Be8➔~-2 He4 f 2 He4 + Q 14 4 Artificial radioactivity of lithium under deuteron bombardment was first reported by Crane, Delsasso, Fowler and LauritsenC 11 ). The radioactive element is assumed to be 3118 formed in . . reaction (12) and disintegrating according to ·reaction (13) and finally (14). ( J) here stands for the neutrino). The experimental setup for observing the electrons is shown in fig ure (2). was placed at 45° to the plane of the chamber. The target The electrons entered the chamber through a thin copper foil soldered over a window cut in the target holder. A quartz ring, visible through a glass section in the ion tube, f luoresced under bombardment and made it possible to direct the beam on to the target. Movement of the chamber nee• essary to line up the beam was made possible by a sylphon in the ion tube. The magnetically operated shutter in the ion beam permitted the beam to be cut off at any time desired in relation to the expan- -* It is extremely difficult to measure shorter ranges because of the continuous alpha particles produced. -9- sion of the chamber.* A description of the chamber proper and the coils for producing the magnetic field will be found under the section on gamma rays. Photographs were taken at right angles to the plane of the chamber. The photographs were re-projected through the same opti- cal systems to actual size and the curvature of the tracks measured by comparison with arcs of known radii. graph is shown in figure (5). A typical cloud chamber photo- A careful discussion on the errors of .measurement will be found in the article on "Radioactive Elements of Low Atomic Number, 11 by Fowler, Delsasso and Lauritsen ( 2 0). From the masses we obtain for the sum (Q 12 + Q13 + Q14) 15.6 ~ .2 Mev. Rumbaugh & Hafstad(3?) have shown that the range of the protons in reaction (12) must be less than 8 cm. or Q12 c::::_l.8 Mev. From the experimentally observed end point of the beta spectrum Q13 These data make Q14 approximately is found to approximate 10.5 Mev. 3.3 Mev. From this it was concluded that 31i8 did not go directly to the ground state of 4Be8 in the beta transition, but rather to an excited state of particles. Be8 which then could disintegrate to two alpha 4 On the assumption that the end point of the beta spectrum is a constant, the data were plotted in two different forms, corresponding to two ways of expressing the Konopinski-Ulenbeck modification of the F er mi theory, based on Pauli's neutr1no assumption(20) and are shown in figure (6) and figure (7). The radioactive alpha particles from reaction (14) have recently been reported by Lewis, Burchain and ChangC 27 ) who reported an energy distribution up to approximately 6 Mev., with most of the particles below 2.5 Mev. From the ir limited data they were not able to say whether the particles represented a line structure or a continuous distribution. This distribution has * Essentially the same setup has recently been used to study the alpha particles from reaction (14) . The copper foil in this case is replaced by an aluminum foil of 4 mm. air equivalent. -10- been carefully investigated in this laboratory(50). The distribu- tion in energy has a maximum at 1.3 Mev. and falls sharply to less than half the maximum at 1 . 0 Mev. very close to 0.5 Mev. The width at half maximum is The high energy portion of the distribution I decreases rapidly with increasing energy extending to at least 6.0 Mev. The average energy of the alpha particles is found to be 2.0 Mev. No evidence was found for line structure. With this distri- bution of alpha particles, one then will not have the beta transition with a single maximum energy point, that is Q13 and Q14 constants, but rather the sum of Q and Q only is a constant. 13 14 Under these circumstances, the application of the unmodified theory to the experimental results is not justified, and the only reason one obtains a reasonable agreement is because the distribution of the alpha particles is such that most of the m come within a small range of energy. Knol and VeldkampC 24) have reported t he formation of radioactive 31i 8 from 31i 7 by slow neutrons. repeated this work(5l). They have recently By using a target of LiN0 3 solution which they circulated through a closed system past two thin walled counters spaced along the system they can deter mine the half life of the beta activity, knowing the time for a given part of the solution to pass from one counter to the next. value of 0.8 ! 0.2 sec. In this manner they obt ain a No measurements on the energy of the beta rays are reported. They conclude that according to reaction (12a) 8 31i is formed in an excited state, losing its excess energy as gamma radiation. Such radiation has not been observed so far. -11- DISINTEGRATION GAMMA RAYS - (15) (16) 17 Me v . . _ Q15 14 M~v .. . _ Q16 The gamma rays from lithium under proton bombardme nt have been the source of controversy ever since t heir discovery was report ed . In fact, the controversy ha s ranged from t he complete denial of their existence by Oliphant and We stcott( 35 ) to reports by Traubenberg and collaboratorsC 45 , 46 ) of t he ir existence at 45 to 60 k.v. bombarding voltage; later work explains Oliphan t 1 s failure to find gamma rays at bomba rdlng voltages below 200 k.v. and shows t ha t wha t ever Traubenberg found, it wa s not gamma rays f rom lithium. Crane and LauritsenC 13 ) re ported gamma rays from a LiF tar ge t under 600 k.v. bombardment. Al t hough t he ir existence has no longer been a point of controversy for some time, the nature and origin of these gamma rays have up to t he pre se n t been a matter of concern for the worke rs in this fi eld . Muc h exper i ment a l wo rk has been done in this l a boratory and elsewhere. Th is work will now be discu ssed . The a bsor ption method wa s first used in an effort to determine the energy of· these gamma rays a s reported by Crane, Delsasso, Fowler and Laurits en C1 2 ). The value obta ined by this J me thod ( 6. 3 Mev.) wa s interpreted as t he average va lue of a complex spectrum. Th is conclusion as later experimental work has shown was incorrect. The explanation for the absorption value s obtained will be given l a ter. It was next decided to use the recoil electrons as obs erved in a cloud chamber in a magnet ic field to determine t he energy of this radiation. ~ hi s work wa s reported by Crane, Delsasso, -1 2- Fowler and Laurits en C1 2 ) . used as a scatterer. The glas s wall of the cloud chamber was Single photographs were taken; but it wa s believed that one coul d be reasonably sure of the ori gin of the tr ack s. The energies of 1576 single e lectr ons and 57 pairs were measured. The single electrons, which we re considered a s being mostly recoil electrons from the glass wall were interpreted as a line structure extending from 3 to 17 Mev. The 57 pairs on the other hand were distributed in the energy interval from 10 to 17 Me v., indicating the radiation to be distributed mostly in the · higher energy region. Because of the small number of pairs and because the behavior of r ec oil electrons wa s considered b e tt er known, more e·.!1 phasis wa s pl ace d on the single el ectrons. This choice of quantity over qua lity as we shall see wa s unfortunate ly a poor one. Be cause of t his discrepancy b e t ween t he recoil elec trons and pa irs, it was decided to obtain mor e data on this pr oblem under better experimental conditi ons. EXPERIMEN TAL ARRANGEMEN T The arrangement of apparatus is shown in figure ( 8 ). The cloud chamber us ed is the same as shown in figure (9), wi t h the exception that a 11 knee" has been added between t he piston rod and the operating sylphon to make it pos sibl e to "disconnect" t he chamber from t he operating mechanism, and thus secure a more rapid expansion . The more rap id expansion permi tt ed t he t racks to be photographed c los er to the scatterer, and , t herefore , made t he i r identification easier . The Helmholtz coils are capable of producing a field of about 3000 gauss with out undue heating when operated intermittently, -13- that is, energized about 1 second out of fifteen. Stereoscopic pictur e s were taken by means of a single mirror. When re - proje cted into a model chamber a~d scatterer, t he pictures revealed t he fact t hat mos t of the single ·track s originated outside and in the top and bottom of t he chamber. In order to reduce t he numb er of these unwanted tr acks, the distance between t he chambe r and t arge t wa s increased and a lead colima tor, 18 cm. thick, with an opening just large enough to illuminate the scatterer, wa s interposed. A 1 mm. a luminum window wa s introduced in the wall of the chamber, where the radiation enters, in order to reduce the numb er of electrons fro m the chamber wall. Scatterers of small stopping power we re used to i ncr ea se t h e resoluti on . These changes le ad to a considerable sacrifice in intensity, but t h is sacrifice, as the data obtained has pr oved, wa s we ll worth while. The c a rbon arc used in pre vious work for illumina tion wa s replaced by f our 300 wa tt lamps, t wo on each side. volt lamps were flashed at approximately 190 volts. The 110 This change i mproved the uniformity of t he photography greatly, and also re l eased one D. C. genera tor for other work . The s ynchronized disc carrying t he lead absorber wa s so arranged t hat alternate pictures were taken with t he absorber in the beam. A marker operated from the shaft of t he disc wa s photographed a t each exp osure, thus insuring later certain ty as to whe t he r t he lead wa s in or out for any photograph. The high volt- age equipment for accelerating t he ions for this and al l other exp e riments performed in t h is laboratory and cons id ered here ha s been carefully described .( 15 ) - 14- An automatic contact system provided proper timing for the magnetic field, the filament and anode of the discharge tube, flashing the lamp s, the chamber expan sion and the exposure. Figures (10,11) show examples of photographs obtained wi th t h is arrangement. RESULTS Pairs. Figure (12) shows the distribution in energy of a total of 770 pairs obtained from stereoscopic pictures. 513 of these were obtained with 0.032 cm. lead scatterer and 257 with 0.012 cm. Figure (13) represents the 25 7 pairs plotted separ a t ely . were obtained mo st recently and under the be st conditions·. Thes e It is seen t ha t the t wo curves are in good agreement as far as the distribution in energy is concerned, but t he wid t h a t half maximum is considerably greater in figure (12). This is to be attributed mo stly to greater experimental errors as indicated by the displacemen t of the high energy s ide of the curve. The still greater dis- placement of the low energy side is presumably due to t he somewhat greater energy loss in the thicker scatterer used in mo st of the pictures. These curves indicate clearly a gamma radiation wi th a strong maximum near 17 Mev. To account for the observed distribu- tion of pairs we must consider the following possibilities for the gamma ray spectrum: (1) a continuous spectrum beginning a t about 10 Mev . and ending above 17 Me v. and having a strong maximum at or near the upper limi t; -15- (2) a single line at about 17 Mev. (3) a line at 17 and one or more weaker lines between 10 and 17 Me v. Of these the first seems the least likely for no mechanism is known which might produce such a continuous spectrum and t here i s no evidence for such a spectrum in any nuclear reaction so far observed. Nevertheless, it cannot be ruled out from our data and since the final produce of t he reaction is unknown, there is no direct experimental evidence agains t it. It does not seem possible to account for the observed distribution of pairs as being due to a single line. The observed width of the distribution in energy of t he pairs produced by such a line would be due to the following causes: (1) natural line breadth; ( 2) ionization losses in the scatterer; (3) radiation losses in the scatterer; (4) fluctuations in t he magnetic field; (5) scattering of t he electrons in t he gas; (6) errors in reprojection and measurement of curvature of the t racks. According to Crane, Delsasso, Fowler and Laur its en C12 ) the radiation in question is produced by resonance. The best measurements on the excitation as function of energy are those by Hafstad, Heydenburg and TuveC 22 ) who find strong resonance at 0.440 Me v. with a half width of 0.011 Me v. From this we conclude that t he half width of the gamma ray line is not muc h more than 0.011 Me v. -16- The scatterer used for obtaining most of the da ta in figure (12) was 0.032 cm. of lead and t he ioniz ation losses for 17 Mev . pairs are , therefor e, uniformly distributed between ze ro and 0.800 Me v. The effect of this is to broaden t he line uniformly toward lower energy by this amount. The radiation losses for electrons in this energy range are, according to Bethe and Heitler(l) approxima tely equal to the l os ses by ioniz a tion, but the nurrber of electrons which would suff er a radiative collision in 0.032 cm. of lead comes out to be rather small. Radiation losses will not produce a uniform broadening of t he line but only a tailing off toward lowe r energy. The numb er of pairs contained in t hi s t a il may be calculated from data given by Be t he and Re itl er and is about 10%. Fluctuations in t he magne tic field amount to less t han 2%, causing a symmetrical broadening of not more t han 0 . 340 Mev. The sc attering of these high energy electrons in air at a pressure of one a tmosphere is extremely small and we prefer to include this •· in the error s of measuremen t of curva tur e . The me asurement of curvature is usually reproducible to 1 Mev. for any pair and we consider t he pr obable e rror due to sc a ttering , rep_'ojection and measurement less than t h is amount. The total effect of a ll t he se factors woul d be a nearly s ymme trical broadening of the line but with a shift of t he center of gravity amounting to about half of the ionization loss in t he scatterer. We have indic ated such a symmetrical distribution in figure (12) and it is seen t h at mo st of the pairs obs er ved lie with in t his distribution and may be a tt ribu t ed to a l ine a t 17.1 ~ - 0 .5 -17- Mev., but it is clear that a considerable frac t ion of t ~e pair s of lower energy cannot be attributed to this line directly. Analyzing the da t a in this manner, we obtain the following results: The a verage energy of 580 pairs lying within t h e symmetrical hi gh energy region is 16.7 ± 0 .5 Mev . To this must be ad ded 0 .4 Mev. f or t he mean loss due to ionization in t he scatterer, giving as t he mos t probabl e value of t he energy of t he h i gh energy gamma ray 17.1 ±. O. 5 Me v. The number of pa irs lying belovr t he symme t rica l dis tri bu- t ion in f i gure (13) is 1 90 or about 25% of t h e total . It s eems likely t ha t approxima tely one-fourth of t he s e pairs, tha t is 10% of 580 , can be a ccounted for as being pairs wh i ch have lost from 1 to 10 Me v. i n escaping from t h e scatterer. The remaining pai rs , amounting to some 15 to 20% of t he tot al , can a pparen tly no t be accounted for in this manner and must t hen b e· due to radiation of energy less t han 17 Mev. falling on t he scatterer. This radi a ti on may be due to one or more of the following causes: (1) 11 7 ~ one or more line s or band s of gamma rad i ation from H1 in addition to the 17 Mev. radiation; ( 2) sec ondary radiation pr oduc ed b y t he 1 7 Mev. line in t h e mat e rial surrounding t he cloud chamber and scattered into the ch amber; (3) radi ~tion due to contamina ti on in t he beam or t arge t, or b oth. From the work of Bethe and He itler we can calculate the number of quanta produced by a 17 Me v. quantum and having energies between 10 and 17 Mev . This c ome s out to be less t han 2% even for lead, and can, t he refo re , not account for t he low energy pairs obs erved . The only reac t ion known which mi ght give radiation of s uff icient energy to acco unt i' or t he observed pairs is Ell f - 18- Fl, but it _-i has an excitati on efficiency of t he same order as Li 7 + H1 and hence t he contamination would have to amount to some 20 to 30% , which is obviously out of t h e question. Thus we seem forced to the conclusion that Li 7 + H1 emits some radiation between 10 and 17 Mev. in addition to the radiati on at 17 Mev. It seems highly probable that this radiation consists of a line in the neighborhood of 14 Mev. and t hat . t he intensity amounts to some 20% of t h e total, but it is possible t ha t it is distributed among t wo or more lines between 10 and 17 Mev. From our measure- ments we may further conclude that t he re is no radiation between 2 and 10 Mev. amounting to more than 5% of t h e tota l. Softer r a diation (down to a million volts) was looked fo r, using a weaker magne tic field, but none wa s fo und . RECOIL ELECTRONS It i s much more difficult to obtain relia•ble data on t he rec oil electrons for clearly not all of t h e single trac k s observed belong to this c a tegory. This is particularly true in t he. pictures taken without collima tion. The reduction due to collimation in t h e relative numbers of sing le track s is best seen from Table 1. Table 1 no c ollima t i on collima t ion f Scatterer Pairs Electrons Positrons J?b 513 381 155 Pb Al 257 101 105 12 71 49 Recoil Electr ons 52 93 It seems mo st reasonable to assume t ha t t h e single positrons observed when collima t ion is used a re in reality members of pa irs -19- orig inating in t h e scatterer and t ha t an equa l number of t he single electrons are of the same origin. Presumably t he corres- ponding pair members have esc ape d detection either due to lar ge energy loss and scattering or t o imperfect photography. To obtain t he appr oximate numbe r of recoil electrons we have, t herefore, sub t rac ted t he number of single pos itron s observed from t he number of sing le electr ons . The effect of collimation is also apparent from Table 2 in whi ch we have shown the average energies of t he several groups with and wi t hout collima ti on . The aver ag e energy of r e coil elec- t ·ons obtained with collima t ion and determined as indicated above is 12 .7 ± 0.7 Mev. which is in satisfactory agreement with the value 1 2 .2 Mev. predicted by t h e Klein-Nishina for mula for 17.1 Mev. radiat i on. Table 2 Average energy of pairs Aver a ge energy of electrons Average energy of positrons Average ener gy of recoil electrons With Coll i ma tion With out Collimati on 15.7 ± 0.7 12.2: 0.6 11.1 ± 1.0 1 2 .7 ~ 0.7 15.7 i 0.5 10.7: 0 .3 10.8 ± 0.6 10.7 ~ 0.4 Figure (14) shows t he distribution i n energy of apparent recoil electrons with and without coll i mation . The distribution obtained wi th collima ti on prob ably represents quite accurately t h e true recoil electrons and is in s atisfa ct ory ag reemen t with expe cta ti ons based on the Klein-Nishina t heory and t he radiation indicated by the pairs . From t he data pres ent ed abov e , it seems t ha t no cont radi .ction exis ts between t he results gi ven by t he pa irs and t he recoil electrons, at least at t he s e energie s, and, t herefore, such an ex- 20 - planation a s proposed by Crane(lO) to account for such a discrepancy seems unnecessary. Energy Divi sion Between Pair Membe rs: Bethe and Reitler have calculated t he probability for t he energy division betwe en the two members of pairs of vari ous energies. Th e curve in figure (•15) shovvs this probability for 1 7 Mev. pairs and the points represent t h e number of electrons observed naving a given fraction of the total energy of t h e pairs. Th e deviation a t the low and high end are to be expected due to the great probability that a pair is not measured as such if t h e energy division is very unequal. This systematic error is not included in t he probable errors indicated. The agreement with t h e t h eory is entirely satisfactory. Absorption in 1 cm. of Lead: Up to t he present time the only measu rements of absorption coefficients for radiation in this energy range have been made i n the usual way by means of ionization chambers . Unfortunately, such measurements are not reliable and cannot be taken as valid tests of t he theory developed by Oppenheimer and Plesset ( 35 ) and by Bethe and HeitlerC 1 ). This is evident from an examination of cloud chamber pictures taken under simila.r conditions for the y show tha t mo st of t he ionizat ion is produced by electrons which cannot be attributed to t h e direct be am . With the low int ensity available t he geometrical arrangement is necessar i ly such t ha t stray and scattered radiation c on tributes a l arg e par t of t he ionization and because t h e absorption coefficient for muc h of t his radiation is lower than that for the primary radiation t hi s part becomes relatively greater with inc reas ing absorber thickness. -21- Wi th such an arrangement we should, t he ref ore, expe ct to obtain a value of t he absor p tion coefficient which lies below t he true value and app roache s t he minimum of t h e absor pt ion curve as t he thic kness of absorber is increased. For lead this minimum occurs at about 3 Mev. and t he measured absorp t ion coeffici ent may , t he refo re, corres pond to any value of t he gamma ray ener gy between 3 and 17 Mev., depending on how this measurement is made . Calculations have been made to show thi s(l 9 ). Figure (16) s hoYs the number of primary quan ta and quanta fr om 2 to 6 Mev. a s a func t ion of thickness of t he lead absorber. It is seen t ha t al ready at a depth of one cm. t he number of pene trat ing quan t a (ar ound 3 Mev. ) equals t h e number of primary quanta, and t hey predominate mor e and more wi th t hi ckness of t h e absorber. The calculated value of t h e logarithm of I (th e ionizati on expected in t h e ionization chamber) is plo tted in figure (17), giving an al most s trai ght line wi t h a constant slope corres ponding to ,,...,« ~ 0.50 cm.- 1 , t he average total absorption coeffici ent corresponding t o radiati on in t he interval from 2 to 6 Mev. If calculations were carried out fo r still greater thickness, t he slo pe would gradually .decrease, ulti ma tel y appro a ching the value /4 = 0 . 46 cm.- 1 , which i s t h e minimum abs orpt ion coefficient in lead and c orre s ponds to radiati on of approxima tely 3 Mev . For compar ison , t he t heoretical value of the ab sor pti on c oefficient; A= 0.74 cm. - 1 for 17 Mev. radiation is also shown . It should be noted t hat t he pene trating secondary radia tion builds up so r ap idl y tha t t h e absorption curve as here calculated is straight even for very t hin absorbers, and a t no point is t he abs orption coeffi cien t - 22- of 17 Mev. radiation approached. This shows cle ar ly t he absorpti on method is unsuited for determin g t h e true absorption c oefficient for radi a tion above 3 Mev. To circumvent t his difficulty, we have taken cloud chamber pictures alternately with and without 1 cm. of lead interposed between t h e t a r ge t and the cloud chamber. By comparing the number of pairs obtained wi th lead i n t he beam with t he numbe r ob~ tai ned without lead we have a true measure of t he total attenuation in 1 cm. of lead of the radiation which pro duces t hese pairs. It seems likely t ha t some pairs of low energy a re produced by r adiation which is scattered into the chamber from the lead absorber, hence it would be reasonable to c on sider only pairs having energies near the maximum, say within the s ymmetrical distribution indicated in figure (13). I n Table 3 we have listed the number of pairs observed in t hree ener gy intervals with and without absorber. Table 3 Number of pairs in Remainder s ymmetrical h igh energy above region 10 Mev No absorber Pairs below 10 Mev Total 260 ± 11 66 :f: 6 2 3 28 With 1 cm. lead absorber 135 :1: 8 51 :I: 5 8 194 Th e attenuation in 1 cm. of lead of the radiation producing t he high energy pairs is seen t o be 135 :1: 8 260 ± 11 = 0.52 :1: 0.04 which gives a total absorption coefficient for t his radiation of · • 1 = -log 0.52 = 0.66 ± 0.07 cm -23- The Klein-Nishina formula gives for 17.1 Mev radiation u = 0.09 cm- 1 whi le Bethe and Beitler give for the absorpti on due to pai r formation ,r = 0.64 cm -1 Hence = 0.73 cm- 1 which is in fair agreement with the observed value. Origin of t he Gamma Radiation: That t he gamma radiation here dis cussed is due· to the Li7 isotope is clear from energy consideration and this has recently been verified by Rumbaugh and Hafs tad( 37 ) who, using the separated isotopes of lithium, observed gamma radiation from 4i7 and confirmed t he resonance a t 0.44 Mev. but found no gamma radiation from Li 6 . The energy available may be calculated from the masses and is 7.0182 + 1.0081 - 8.0080 = 0.0183 or 17.0 Mev. To this mus t be added 7/8 of t he kinetic energy of the bombarding proton. Hence t he total energy available is 17.4 Mev. In the ar ticle by Hafs t ad, Heydenberg and TuveC 22 ), Breit gives a discussion of several possible mechanisms to account for t h e radiation under discussion. The one which best accounts for the observa- t ions is based on t he assumption that the proton is captured on a virtual level for mi ng a Be8 nucleus in an excited sta te which is supposed to be odd in order to exclude disintegration into two alpha par ticles. This was first sugg e sted to us by Dr. Elsasser, his assumption being t hat only protons having the correct combination of angular momen tum and spin could be captured on this level. The model of the Be 8 nucleus used by Breit is based on unpublished cal-24- culations by Wigner and Feenberg. The ground state of Be8 is a ls level which is even and there is an even ln level at approximately 3 Mev. The next even level would be a 1 G at about 8 Mev. The virtual level at 17 Mev. is supposed to be an odd P level. The data presented here indicate that the transition between P to G levels occurs rarely if at all. This is to be expected, since this is a strongly forbidden transition. This leaves reactions (15) and (16) as the probable source of the radiation observed, these occurring with a relative probability of 3 to 1. Since no radiation at 3 Mev. was found, it is reasonable to suppose the Be 8 in reaction (15) breaks up into two alpha particles, each having an energy of approximately 1.5 Mev. It must be mentioned that results of experiments recently presented by Gaerttner and Crane( 2l) agree with the results presented above as far as the pair data are concerned, but disagree in results obtained from the single electron data. The fact that in the setup used by Gaerttner and Crane, it is not possible to determine the starting point of the electrons and that they .may also have electrons due to radioactivity induced by the deuteron contamination in the proton beam, may ·e xplain this discrepancy. The above explanation of the origin of gamma rays seems to account for the facts reasonably well. It is, however, based on the assulllption that all the gamma radiation observed is due to the 440 K.V. resonance level. Recent work by Bothe & GentnerC49) shows an addi- tional resonance at approximately 200 K.V. A possible source of this resonance could be Boron contamination, for which they find a resonance at 180 K.V. However, from the relative intensities given by them the contamination would have to be at least 20%, and possibly 50%, -25- which seems too large to be probable . It is impossible to say from the work of Tuve and his collaborators whether this lower resonance does exist or not. Their published data do not extend in any detail to these low voltages. Neither can one draw any conclusions from the · fact that Oliphant was not able to observe gamma radiation in the work reported at the International Conference of Physics in 1934, his maximum voltage being 187 K.V. can not be excluded either. On theoretical grounds such. a resonance This might be the excitation of another member of the triplet state, the separation of appr-oximately 200 K. V. is not unreasonable. Bothe & Gentner also report measurements on the energy of the gamma radiation produced at 300 K.V . and 520 K.V .. these they obtain the same result of 18.0 Mev. For both of However, t heir method of measu~ing would not be able to resolve lines only 200 K.V. apart, so that they do not contradict the preceding discussion. They do show, however, that the 14 Mev. component observed in this laboratory is not produced exclusively by the 200 K.V. resonance. Finally, one must say that before any definite conclusion. about tbis resonance is reac hed, more work is necessary. Gamma radiation from Lithium under alpha part1cle bombardment was first reported by Bothe & BeckerC4). SnetzlerC4 2 ) finds that production of this gamma radiation begins with 2.3 Mev. alpha particles as compared to 4.7 Mev. for the minimum energy for producing neutrons. These results corroborate the results of Save1(4l), who found approxi mately 2.8 Mev. necessary for the production of gaw.ma radiation, compared to 5.0 Mev. fro m the neutron reaction. -26- From these results the conclusion is drawn that this gamma radiation is produced by inelastic impact. Webster's (4 8 ) attempts to show that the energy lost by the alpha particles corresponds to the energy of the gamma radiation are not very conclusive in the case of lithium. Bothe (5) has given the energies of this gamma radiation as .39 Mev. and .59 Mev. From the difference in the proton ranges previously discussed, we might expect a gamma ray of approximately .4 Mev. This is in fairly good agreement with the value 0.39 Mev. obtained by Bothe, although it does not explain the 0.59 Mev. value. -26a- CONCLUSION Table (4) .. . - · -- -.......-- : 1. Li 3 Li 7 + l H1 _ :2. :3. Experimental Reaction ~-....-------,-- 3.8 + 0.3 2 He4 + 2 He4 31i 6 -t l H2 __,_ 2He4 + 2He4 17. 06 ::!: • 04 3 Li7 + H2 ___ He4 + He4 + 0 n 1 1 2 2 3 3Li6 + onl -2He4 + 1H3 + : 7. + 2He4 + onl . 3Li _7 + 1 H2-4Be8 onl 1i6 + H2 _ He4 + He3 : 8. :9. .. + nl 1 2 2 0 3 1i7 + 2He4 ______ Bl0 + 0 nl 5 3 .. . 22.0 14.6 + 0.3 (a) • 14.8 4. 3 + •5 . . Not known .. Not known 4.8 -t- 0.1 .:11. 31i6 1H2 _ 31i7* + 1Hl :12. 3Li7 + 1H2- - 31i8 1Hl 4.4 -t- 0.1 -t . + ; 12a . .3Li 7 + 0 nl __ 1i 8➔~ : 3 -8 8* /3· 3L\; s-13e + e : ,> ; : 14. 4Be -- 2 He 4 + 2 He .:15. 3Li7 1Hl _,_4Be8*'4Be8 .: + 17 Mev.: . .. :16. 1i 7 + 1H~4Be8~4Be8-* 14 Mev.: -t 11 3 L_ 1.8 Not known l 4.7 14.8 .. . 14.8 . . .. 1.5 -2.6 5.0 .: See Note (c) : See note (d) : See note (e) .. . . . See note (f) 17.1 -t .05 . - 17.0 . 14. 3 + . 5 (b) 14.8 + 0.3 · 3.7 22.06 + .04 :10. 31i6 + 1H2~Li7 + 1 Hl . . - - - - - - r - - - - - - - - ~ · - - - - - - -- + 2He3 . :6. 3Li7 t 1H2_ 2He4 . From Ma sses(x):. . . Q's in Mev. 17.4 (g) See note (h) . .. . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.;....•_ _ _ _ _ _ _ _...a___ _ _ _ _ _ _ • Table (4) shows the values of the Q's obtained from the experimental data and those obtained from mass considerations (appropriate corrections for bombarding energies have been included). (x) The masses used here are those given by Bonner & Brubaker(3). They were obtained from mass spectrographic data and from disintegration data, including some of the above reactions. -27- (a) Obtained from maximum range of the alpha particles. (b) Obtained by using mean energies of alpha particles and neutrons . (c) (d) 3Li7 is left in an excited s tate. Mass of 3Li 8 not known . (e) Mass of 3 Li 8 not known; also it is left in excited state. (f) Approximate estimates only can be made at present which do not disagree with the values obtained from the imperfectly known masses . (g) To the mass energy 0.4 Mev . of the bombarding energy must be (h) added . 8 Be left in excited state. 4 From the material presented here and summarized in Table (4) it is seen that so far as energy balances are concerned, the known reactions fr om lithium are in a reasonably satisfact ory state and probably no great amount of new knowledge can be obtained from furt her experimental study of these balances. The work on t he excitation functions and cross-sections for these reactions is, unfortunately, with the exception of isolated ca ses , very poor and has for this reason not been included here. - 28- ACKNOWLEDGMENTS I take this occasion to express my gratitude to Professor Lauritsen for his guidance and constant interest in my . work. I tbank Dr. W. A. Fowler, who with Professor Lauritsen collaborated with me in the work referred to herein under our names. Finally, I wish to acknowledge the finan- cial assistance of the Seeley W. Mudd Fund. 22 0 20 0 18 0 16 0 I •·I I ! 1· z (/) 14 0 I- 12 0 1' I 0 ,· ~ 0 O (l.. 10 0 LL B 0:: CD ~ •0 Z 20 I ! ' i I 1· .,, 0 i ' ll . 0 , 1,1 ....... ,, I i ' I I ..·- . iii!!! . •• ,,.., .. .....■ :_ ill!!II ··=--- ::, 0 0 ll~ 0 · = '~ 0 OF RECOIL rJ Ir.I- · 0 0 IU 10 ENERGY Fig. 1. . i o ·I:. 0:: w I' I! I ■ PROTONS i:i.l 12 IN II 13 !4 I~ M.E.V Energy dis tributi on of recoil pro tons ( upper curve) and primary neutron (dotted curve). Fig. 2. Arrangement of apparatus for de termination of electron distribution in energy. -30- Fig. 3. Fig . 4. Example of proton tracks Distribution in range of alpha parti cles (s hor t range groups ) and proton (long range groups ) -31- Fig. 5. Example of electron t ra cks Fig . 6 . Energy Distribution of electron omit ted fro m Li. bombarded by deuterons . • - 32- Fig . 7. K. V. plots for ~Lis and 0 F 2 0 Fig. 8. Arrangement of appara t us for t he determi nation of gamma ray energies - 33- Fig. 9. Cross - section of cloud chamber. Fig. 10. Examples of pairs fro m lead scatterers. -34- Fig. 11. Example of recoil electron sta rting in t he scatterer and one passing through. Fig. 1 2 . Distribution in energy of a total of 770 pairs . Obtained from a , 0.032 cm. lead scatterer and a .01 2 cm~ lead scatterer. -35- Fig. 13. Distribution in energy of 257 pair s from a 0 . 012 cm. lead scatterer. Fi g. 14 . Distribution of a pparent r e coil elec t ron with and without coll i ma ti on . - 36 - Fi g . 15. Energy rltv~ sio~ between members of 17 Me v. pairs . Fig. 16. Distribution of" 1'1 Mev. primary raaia1, _,_011 and 2 to 6 Mev . radiation as a function of absorber t hi ckness. - 37- Fi g . 17. Log of ioniza ti on as a fun ction ab sorber t hickne ss ·for = 0 . 46 cm. -1 (3 Mev. radiati on) -- 0. 50 cm. - 1 ( 2 -6 Tuiev. r a diati on) -1 (17 Mev . r a diation ) = . o. 74 cm. - 38- BIBLIOGRAPHY (1) Bethe, H.' Heitler w. Proc. Roy. Soc. 146, 83 (1934). (2) Bonner, T. w.' Brubaker, w. M. Phys. Rev. 48, 742, (1935). (3) Bonner, T. w., Brubaker, w. M. Phys. Rev. 50, 308, (1936). (4) Bothe, w. V •, Becker, H. Zeit. f. Phys. 66, 289, (1930). (5) Bothe, w. v. Zeit. f. Phys. 100, 273, (1936). (6) Chadwick, J. Goldhaber, M. Nature, 135, 65, (1935). (7) Cockcroft, J. D. Int. Conf. on Physics, (1934). (8) Cockcroft, J. D. Walton, E. T. s. Proc. R. s. A. 137, 229, (1932). (9) Cockcroft, J. D. Walton, E. T. s. Proc. R. s. A. 14 4, 704, (1934). (10) Crane, H. R. Phys. Rev. 51, 50, (1937). (11) Crane, H. R.' Delsasso,L.A., Fowler, W.A., Lauritsen,C.C.Phys. Rev. 47, 971, (1935). (12) Crane, H. R.,Delsasso, .LJ .A. Fowler, W. W.,Lauritsen, C.C. Phys. Rev. 48, 125, (193 5). (13) Crane, H. R.' Lauritsen, C.C.Phys. Rev. 45, 64, (1934). (14) Crane, H.R., Lauritsen, Solton, A. (15) Crane, H. R. Lauritsen, Solton, A. c.c. Phys. Rev. 44, 692, (1933). c.c. Phys. Rev. 45, 507 (1934). (16) Dee, P. I. Nature, 132, 819, (1934). (17) Dee, P. I ., Gilbert, C. W. Proc. R. S. A. , 154 , 279, . (1936). (18) Delsasso, L.A., Fowler, W.A. Lauritsen, C. C. Phys. Rev. 48 , 84 8 , (1935). (19) Delsasso, L.A., Fowler, W.A. Lauritsen, C.C. in publication. -39- (20) Fowler, W. A., Delsasso, L.A. Lauritsen, C. C. Phys . Rev. 49, 561, (1936). (21) Gaerttner, Crane, H. R. Phys. Rev. 51, 49, (1937). (22) Hafstad, L. R., Heydenburg, N. P., Tuve, M.A. (23) Kirsh, G. (24) Knol, K. Phys. Rev. 50, 504, (1936). Atomzertrllmmerung, Akademishe Verla~s gesellschaft m.b.H. Leipzig, (1926). s., Veldkamp, J. Physica 3:, 145, 1936. (25) Lawrence, E. 0., Livingston, M. s.' Lewis, G. N. Phys . Rev. 44 , 56, (1933). (26) Lewis, G. N.' Livingston, IVi . s .' Lavrrence, E. o. Phys. Rev. 44, 55, (1933). (27) Lewis , W. B ., Burchain, W.E., Chang, W. Y. Nature, 139, 24, (1937). (28) Lyman, E. M. Phys. Rev. 51, 1, (1937). (29) Mano, G. J. de Phys. et Rad . 5, 628, (1934). (30) Mei tner, L. Naturwissensch. 22, 420, (1934). (31) Oliphant, Tu1 . L. E., Kempton, A. R.' Rutherford, E. Proc. R. s. A. 149, 406, (1935). (32) Oliphant, M.L.E., Kinsey, B. B.' Rutherford, E. Proc. R. s. A., 141, 722, (1933). (33) Oliphant, M.L. E ., Rutherford, Proc. R. S . A. 141, 259, (1933). E. (34) Oliphant, M. L. E., Shire, E. S., Crowther, B. M. Proc. R.S.A. 146, 922, (19 34). (35) International Conf.of Physics, (1934). Oliphant, M.L.E., We stcott, (36) Oppenheimer, R. J.' Plesset, Phys. Rev. 44, 53, (1933). (37) Rumbaugh, L. H.' Hafstad, L.R.Phys. Rev. 50, 681, (1936). (38) Rutherford, E., Chadwic~, J. Phil. Mag. 42, 8 09, (1921). (39) Rutherford, E., Chadwick, J. Nature, 113, 457, (1924). -40- (40) Rutherford, E., Chadwick, J. Proc. Physo Soc. Long. 36, 4117, (1924) • (41) Savel, P. Compt. H.endu, 198., 368, (1934). (42) Snetzler, K. v. Zeits f . Priys. 95., 302., (19.35). (43) Taylor, H.J., Dabholkar, V. D. Phys. Soc. Proc. 48,285 , (1936). Traubenberg, H. H. v., Eckardt, A., Gebaur, R. Naturwissensc haften, (45) Traubenberg, R. H. v. Zeit. r. Phys. 80 , 55?, (1933). (Ji,6) Traubenberg, H. H. v. Zeit. f. Phys . (47) Walton, E.T. S., Cockcroft, Jo •D. Natur e, 131, 23 , (1933). (48) Webster, H. c. Proc.R.S.A. 136, 428 , (1932). (49) Bothe, VJ ., Gentner , (50) Lauritsen, (51) Veldkamp, J., Knol, K. S. (44) w. -, l ,:: . ..1.., 26 , (19.33). 89, 582, (1934). Zeit. f. Phys. 104, 685, (193~ c. C. Fowler, Vi .A. In publication. Physica, 4 , 2, i66, (1937). -41-