COSMIC RADIATION AND RADIOACTIVITY .BY

LOUIS R. MAXWELL, Ph.D .

National Research Fellow.

BARTOL RESEARCH

IN 1898 S . CurrieI suggested that thereFOUNDATION

may exist in all space some unknown highlycommunication No . 37 . penetrating radiation which is capable ofbeing absorbed by the elements of high atomic weight . Thisradiation would manifest itself by its secondary radiations inthe form of Becquerel rays . Later Perrin 2 proposed a theoryof radioactivity in which he assumed that the atoms of theradioactive elements are capable of absorbing ultra pene-trating radiation which passes unabsorbed through non-radioactive elements . Experiments which have afforded atest of this theory are those in which an investigation is madein order to determine if the rate of decay of a radioactivesubstance is altered by placing it underground where it will beshielded from cosmic radiations . Elster and Geitel' havemeasured the activity of joachimisthal (a uranium ore) in amine at a depth of 500 meters and found that its activityremained constant . Piccard and StahelI have found nochange in the activity of UX when left in the Simplon tunnelfor two months at a depth of 2,200 meters . Also A. H .Compton and Hagenow lowered a tube of radium emanationto the bottom of the Calumet Heckla Mine in northernMichigan . After allowing it to remain there for a period oftwo weeks it was compared with a similar tube which had beenleft in the laboratory for the same length of time . Theirresults were not published, but Prof . Compton has informedthe writer that their measurements showed no greater changeof the activity of the two specimens than about two per centin the remaining activity of the emanation . This differencewas of the same order as the experimental error .

In the present work a radioactive substance was takento the bottom of a mine where its activity was measured .These results were then compared with previous measure-

61g

620

Louis R. MAXWELL .

U. F. 1 .

ments which were taken while the substance was at the top ofthe mine . The specimen which was chosen for the investi-gation was polonium because of its relatively short half-valueperiod and its simple transformation . The most accurate ofthe experiments which have been mentioned above are those ofPiccard and Stahel . They prepared four specimens of LUXand for a period of two months they were left in the followingplaces : the first at Zurich (500 m. alt.), the second at Brigue(68o m. alt .), the third at Junfraujoch (3,500 m . alt.) andthe fourth in the Simplon tunnel under 2,200 m . of rock . Themeasurement of the relative intensity of the specimens weremade at Zurich before and after the exposures and were foundto differ by not more than one-tenth of one per cent . If weassume the fundamental postulate of radioactivity the methodused in the present work is equivalent to that of Piccard andStahel . The great difference between the two experiments isthe entirely different measure of transformation which itconcerns . The liX used by Piccard and Stahel emittedp and y rays while in the present work a particles are theprimary agent involved .

The polonium was taken 1,150 feet below the surface tothe bottom of the zinc mine (owned by the New Jersey ZincCompany) at Franklin, New Jersey . The mine contained alarge amount of Willemite (Zn2SiO4) and in order to calculatethe absorption of cosmic radiation for this depth of Willemitein terms of water or lead we can use the following formula ofKlein and Nishmia :

2trNe4 J i + a 12(1 +a) - 1 log(t +

2a Jm2c4 l

a2 L1 + 2a

a

)

+ 2a log ( I + 2a) - ( I +2) 2 ' 'where µ is the absorption coefficient . N the number ofelectrons per cc . which produce scattering, i .e . N = ZLpIAwhere Z is the atomic number, p the density, A the atomicweight and L Loschmidts number (the number of atoms pergram atom) . m is the mass of the electron and e its charge .

a =by

where v is the frequency of the cosmic radiation .

Miac, nn9,l Cosmic RAUr:rrm6 AND RAUIIIACIIVI'IV .

62 1

Thus for a given frequency the absorption depends directlyupon the number of electrons of the atone which produce

Fcc I .

YG

0

V

0

0

0

0

0

M

Experimental Arrangement .

scattering . From this equation we find that the ratio of µfor lead to that for Willemite is 2.8 for any given frequency .Thus 1,150 feet of Willemite should absorb as much of the

622

Louis R . MAXWELL.

[S r.1 .

cosmic radiation as would about 400 feet of lead or 3,200 feetof water . Since Millikan has found that it only requiresapproximately 68 meters of water to absorb practically all ofthe cosmic radiation we are lead to believe that at a depth of1,150 feet in the mine there should he no appreciable cosmicradiation .

The method for measuring the activity of the polonium isillustrated in Fig . i . The polonium plate a was mounted inthe ionization chamber I with a difference of potential of 300volts between a and b supplied by the battery A, while b, the,inner part of the ionization chamber, was connected to thefiber of the single-fiber electrometer . The potential for theplates of the electrometer PP was obtained by using a portionof the potential maintained upon the megohm resistance MM.The ionization chamber was supported by the case of theelectrometer and insulated from it by a hard rubber ring thecross section of which is shown at D . The ionization chamberwas 6.5 cm. high and 5.5 cm . in diameter, and the distancebetween the polonium plate and plate b was greater than therange in air of the alpha particles . The charge accumulatingon b was neutralized by altering the potential of the condenserC by known amounts by means of the battery B and re-sistance R, as shown in the figure . Thus throughout thereadings, the fiber and b were always kept at the samepotential as the case of the electrometer, so that the possibilityof leakage was greatly reduced .

Figure 2 shows a photograph of the portable apparatusused . The box on top of the stand contained the apparatusshown in the preceding figure . It was made nearly airtightand contained drying agents . The readings of the electrom-eter and all of its adjustments could be made withoutopening the box . The longer box in the lower part of thestand contained the batteries and resistances which were alsokept dry . The wires connecting the two boxes were mountedon plugs of sulphur set in the sides of the boxes to insure goodinsulation .

The polonium plate was prepared from old radium emana-tion tubes in which the emanation and products had goneover to RaD . The tubes were broken up and placed indilute HCI . The HC1 was then evaporated on a copper plate .

R,zv l COSMIC RADIATION .~sn RA1nOACTR1TV .

giving an active deposit . By measuring the absorption of theradiation emitted by the plate it was found to consist of nrays, hence the activity must be due to polonium . In fact,over a period of 140 (lays the radiation from pure polonitin:

FIG, 2 .

D"iew o the apparatus ,In u:h was taken into the miuc .

consisting entirely of alpha rays, will diminish to half value .It was found that while the activity of the plate prepared wasdue almost entirely to alpha rays, its strength remainedconstant to two per cent for a period of ninety days. Theinference is, that the amount of Radium E which remained onthe plate after evaporation of the hydrochloric acid was justsufficient to generate polonium at a rate which balanced theloss by decay . While the condition obtained was accidental

VOL. 207, No. 1241-43

624

Louis R. MAXWELL .

[J . F. 1 .

and unexpected it was not without its advantages in thepresent work.

RESULTS AND DISCUSSION.

The activity of the plate was measured at the top of the(nine and then 1,150 feet below the surface . The ionizationcurrent produced in I varied only between the values of9.6 X io_ ,1 amperes and 9.7 X io" amperes during theentire course of the measurements . We thus conclude thatthere is no effect of cosmic radiation on the disintegration ofthe polonium to within one per cent . The natural ionizationof the ionization chamber when at the bottom of the mine was6 X to-" amperes . This current, which must be due to theactivities of the rocks is small compared to the ionizationproduced by the polonium .

Swann 5 has shown that a single cosmic ray, by using all ofits energy in ionization, is capable of producing 5 .4 X los ionsand has pointed out that since Millikan has found thatcosmic rays produce 1 .4 ions per cc . per sec. in air at sea level,we would have 1 .4/5.4 X toy cosmic rays absorbed per cc . persecond in air or 1 .3 X to--' cosmic rays absorbed per secondper gram of absorbing material so that the 1011 grains ofpolonium contained on the plate would absorb 1 .3 X 10- ' 6cosmic rays per second . This would mean that a cosmic raywould be absorbed by one of the polonium atoms once in 1o15seconds or once in about 2 X lo' years, a situation whichentirely precludes any attempt to measure the change inactivity if we assume that the absorption of one cosmic rayis to be associated with the disintegration of one atom .

While the considerations which have already been involvedleave very little hope for a plausible explanation of radio-activity as a result of the action of cosmic rays, it is worth-while to consider the possibilities of a different view point withregard to the action of these rays than the particular mecha-nism which we have already considered . Let us then tracethe results which would follow from the assumption that theray is to be considered as of the nature of a particle, which inits passage by an atom must approach within a certaindistance in order that it should produce the necessary dis-integration . A simple calculation of the number of poloniumatoms on the plate used will show that the total number per

[NLq, [1)211 .1 (ISMiC RA II]Xuov

AND RADIOA,'TOVITV .

62.

sy . cm. is of the order of i X ro 10 , and that therefore thedistance between the atoms is of the order of t X to-' cm,that is about 1,000 atomic diameters . We should naturallyexpect that a cosmic ray operating as a corpuscular entitywould have to approach nearer than this distance in order toproduce a disruption of the atom . The condition is in fact oneof single impact of the cosmic rays in passing through theplate and the maximum effect to be expected is one in whicheach cosmic ray disrupts one atom while penetrating throughthe plate . The problem now remains to calculate thenumber of cosmic rays per second which at sea level will passthrough the polonium plate .

We shall assume that cosmic rays are generated in inter-stellar space and let N(R) be the number of cosmic raysgenerated per unit volume per second between R and R + dR.Then N(R)R2 sin 0 dodOdR (spherical coordinates) will be thenumber of cosmic rays created per second in the polar elementof volume dv in interstellar space . The distance R is measuredfrom an element dS of a surface which incloses a volume V ofair at sea level . The angle B is measured with respect to thevertical . Then the total amount of energy falling on dS persecond due to those rays originating in dv is given by

US cos ik4irR , N(R)C_

e'R sin edod¢dR,

where k is the energy of one cosmic ray and ' is the anglewhich the normal to dS makes with R. H is the height of theearth's atmosphere if it were of uniform density throughout .We have here neglected any absorption which the rays mightexperience before coining into the earth's atmosphere . Theamount of energy dE which will be absorbed in V will be

dE

plkdS cos 1'N(R) sin oe `PH100SrdododR,

47

where the summation is taken with respect to all elements ofsurface dS over one-half of the area of the surface, i .e. overthat part of the surface on which the energy is incident .

6 26 Luis 1%. IAN WELL .

1 is the distance traversed by these rays while going through V.Since the angle B will not change appreciably during thesummation we shall assume that it is constant and hence wehave

dE = µkV N(R) sin Oe `"x1CO8B)dod¢dR .47

If we assume on the average that all of the energy of thecosmic ray goes into energy of ionization, then the numberdQ of ions produced by the absorption of cosmic ray energy toamount dE is given by

dQ = KdE .

where K = r/Ue, Ue being the energy necessary to ionize oneion . Thus we can obtain the following for q the total numberof ions formed per second per cc . in V

µk7q _ 4` .1 N(R) sin Be"-")dodOdR

.,J (t)

The integration will be taken over a portion of the regionoutside of the earth's atmosphere where cosmic rays are beinggenerated, i .e . over the space for which B varies from o to 7r/2,0 from 0 to 2ir, and R between the values R, and ~ . Thefunction N(R) is assumed to become zero at infinity in such a

manner that J N(R)dR will be finite .

The number of cosmic rays do passing through ds, anelement of a horizontal surface, is given by

do = ds f f f N(R) cos 0 sin Oe `""I`080)dodOdR .4

In this equation cos 4' has been replaced by cos 0 since now4' and 0 are equal . Thus we have the following for S thenumber of cosmic rays per second per sq . cm

penetrating

through the horizontal surface .

S= ss = -1~ f f f N(R) cos 0 sin oe(ui10OeO'dod¢dR .

4

jlaY . no-'oI COSMIC R .AulATIoN yyo Rnoto,vc-nvrrv .

From equations i and 2 We now have on integrating

e NH=

dzq

t

-ya _

'S

µkhI

'~CµHr

d.V

where x = sec 0 . kK gives the number of ions a singlecosmic ray is capable of producing and is equal to 5 .4 X 10°for a cosmic ray having an energy corresponding to anabsorption coefficient of approximately i X to -" per cm. ofair . For this value of p we now have

ffCVHx

I w z° dx= .136

and

4H

= . 103,

where H = 10.5 meters for the water equivalent of the earth'satmosphere measured at sea level . q for air has been found tobe 1 .4 . By using these values . S comes out equal to about .2cosmic ray per sq. cm. per sec .

The area of the polonium plate used in this experiment wasto sq . cm . and therefore two cosmic rays would pass throughit per second . There were about 3,000 atoms of poloniumdisintegrating per second and if each cosmic ray striking theplate were capable of causing one atom to disintegrate,irrespective of the fact that it may be necessary for thecosmic ray to approach within a certain distance of the atom,then the total cosmic radiation would contribute to less thanone-tenth of one per cent. of the activity of the polonium .According to these considerations we should not expect tofind an effect of cosmic radiation on the rate of disintegrationfor the present work .

The Writer wishes to thank the New Jersey Zinc Companyfor the permission to carry on the experiment on their mineat Franklin, New Jersey : and desires to express his gratitudeto Dr. F . C . Benson of the Hahnemann Medical College andHospital of Philadelphia for supplying the radium emanation

62

h This expression is of the same for in as that deduced by Swann for the ratioof the influx of atmospheric corpuscles per sq . cm. per sec, into the earth, to thecorpuscular current absorbed per see . per cc . by a mass of ire al .

628

Louis R. MAXWELL .

[J. F. r .

tubes used in the preparation of the polonium, and to Dr . A .Bramley for his assistance in taking the measurements atFranklin, New Jersey . The Writer also wishes gratefully toacknowledge the helpful suggestions of Professor W . F. G .Swann .

REFERENCES .

i . M. Currie, C . R . 126, p . 1101, 1898 .2 . J. Perrin, Ann . de Phys ., t 1, P. 5, 1919.3 . J . Elster and FI . Geitel, Ann . d . Phys ., 66, p. 735, 1898 .4. A. Piecard and E. Stahel, Arch . Se . Phys . and Nat ., 3, P. 542, 1921 .5. W. F. G. Swann, Jour. of the Frank . Inst ., 206, p . 772, 1928 ; Nature, 122, p .

997, 1928 .