THE DESIGN, CONSTRUCTION, AND PERFORMANCE OF A

i1AGNETIC FOCUSING, SEMI-CIRCULAR, LOW ENERGY

BETA-RAY SPECTROMETER

By

Herbert E. Henrikson

Special Technical Report No. 24

Prepared under Contract No. AT(04-3)-63 with the

UNITED ST.h.TES ATOMIC ENERGY COHMISSION

Jesse i<T.M. DuNond, Contract Supervisor

Calif ornia :nstitute of Technology

Pasada~a, California

October - 1956

ACK NOWLEDGM ENTS

The author wishes to acknowledge the important contributions of

Dr. Pierre ?1armier (now at Physikalisches Institut der E. T. H., Zurich,

Switzerland) in the early stages of the design of the spectrometer.

The concept, design, and construction of the field regulator was

by Dr. Joseph }furray (now at the Radiation Laboratory, University of

California, Berkeley, California) .

The experimental work of Dr. Felix Boehm provided the performance

curves reproduced in this report.

The skillful machine work of Alfred Doering is greatly

appreciated.

ii

I.

II.

III.

TABLE OF CONTENTS

Resume • . . . . . . . . . . . . . . . . Detailed description of the semi- circular

spectrometer and i ts construction ••••••

A. Main field coils - construction and specifications . • . . . . . . . •

B. Current re~~ator . • . . . • . . c. Compensating coils for earth ' s

magnetic field • . . . . D. Spectrometer chamber

l. Source •

Page

1

4

4 9

13 15 15

2 . Geiger counter • • • • 18

3. I nstallation of the spectroMeter chamber • • • • • • • • 18

E. Vacuum system and counter filling system • 20

F. Detecting and recording system • • • • • • 23

Performance curves • • • • •

A. :1ediu.11 er.ergy spectrum of an Yb169 source • .

B. Lm-v energy spectrum of Tm169 follovTing

Yb169 decay and Lu175 following ~75 decay • • • • • • •

C. Low energy spectrum of Tm169 following

24 25

26

Er169 decay • • • • • • • • 27

D. L- Auger in Ft and Os spectrum from

I r 192 decay • • • • • • • • • • . . 28

Appendix ... • • • • . . • • . . . • • • • • • . . • • 29

Hagnetic field study • • • • • 29

Circuit diagrams for current regul ator ••• 35, 36

iii

FIGURES

No . Page

1 . Photographic view of complete spectrometer assembly 2

2. Main magnetic field coils 5

3. Cross- section of main field coils 8

4. Current regulator 10

5. Photographic viel·r of open spectrometer char1ber 16

6 . Source holder 17

7. G. H. counter 19

8. Schenatic diagram cf vacuum and counter filling system 21

9· Medium energy spectrum of an Yb169 source 25

10. A. Low energy beta- spectrum of Tm169 following

Yb169 decay 26

B. Lul75 L- Auger spectrum following ~75 decay 26

11. Low energy spectrum of Tm169 f ollowing Er169 decay 27

12 . 1-Auger in Pt ~illd Cs spectrum from Ir192 decay 28

13 . Curves of magnetic fi el d vs. coil spacing 31

14. Curve of coil spacing vs . radius of homogeneous field 32

15. Cross-section of nagnetic field distribution in the region of the electron trajectories 33

16 . Second order focusing 34

17 . Control chassis and interconnect5.ons for current regulator 35

18 . Pm·rer chassis for c•1rrent regulator 36

iv

I Resume

THE DESIGN, CONSTRUCTION, AND PERFORYJ.ANCE OF A

l1AGNETIC FOCUSING, SEMI- CIRCULAR, LOti ENERGY

BETA- RAY SPECTROi1ETER

A semi-circular beta- ray spectrometer (Fig. 1) has been con­

structed for operation between 2 and 120 kev electron energy. This

instrument ,,ras designed to suppl ement the rL'1g- focusing beta-ray spec­

trometer1 whose lower linut is approxinately 25 kev. The 180°

spectrometer has an electron traj ectory radius of 12 em. The field is

produced by t wo iron free coils spaced to give a nearly homogeneous

field (Hith some second order focusing) in the region of the electron

trajectories . The field current is su~plied by a one br motor- generator

set . Stabilization is obtained to 2 parts in 10, 000 with a rotating

electro- mechanical unit and an electronic servo system. Source and

Geiger counter windoH d:'....""l.e~sions for a standard resolution of 0 . 8% are

0 .1 em x 4 em. \hth a 10 ~/cm2 formvar windovT, transmission is assured

dovrn to 2 kev electron energy. A recording potentiometer traces the

counting rate vs . electron energy. Several examples of chart recordings

are reproduced to illustrat e the performance of the semi- circular beta­

ray spectrometer.

The instrument assembly is shmm in Fig. 1 . The spectrometer

proper is shown in the foreground. The framework is welded aluminum 11T"

section, set on four leveling screws . Nonferrous materials have been

used throughout to avoid field distortions . ~he nearest iron is in the

diffusion pump (9) close to the floor . The s~ectrometer chamber, (1) ,

is located synnnetrically between t.r.e tHo main field coils , (2) . Sur­

rounding the assembly are the coils, (3), 1·1hich cancel the earth's

magnetic field. The source holder, (4), and Geiger counter, (5) , are

clamped to the front of the chamber . The counter gas ballast chamber ,

(6), is below the geiger counter and is connected to the filling

1 An Axial-Focusing Magnetic Beta-Ray Spectrometer, by J .H . M.DuHond et al .

-2-

Fig. l. Low energy magnetic beta-ray spectrometer showing: (1) spectrometer chamber, (2) mainfieldcoils, (3) earth 1 sfield compensating coils, (4) source holder, (5) G. M. Counter, (6) G. M. counter filling gas ballast chamber , ( 7) G. M. counter filling system, ( 8) main clapper valve, (9) diffusion pump, ( 10) current regulator, ( 11) generator field power supply, ( 12) control panel (M. G. set, vacuum pumps, etc.), ( 13) current regulator control chassis, ( 14) power supply for compensating coils, ( 15) vacuum gauge, (16) scaler, (17) counting rate meter, (18) cathode follower, ( 19) recording potentiometer.

~-

system,(7). The main clapper valve, (B), is closed to allow the spec­

trometer to be opened to the atmosphere while inserting a source or

replacing the counter >vindow. The diffusion pump, (9), is below the

main valve and is connected by rubber hose and brass tubing to the fore

pump located in a separate room.

The main field is controlled by the current regulator, (10),

seen to the right of the spectrometer. The motor-generator set supplying

current to the main coils is located in a separate room. To the extreme

left of the spectrometer is the rack containing the power supply, (11),

for the field windings of t he de generat or. Above that is the control

panel, (12) , for the vacuum p~~ps and the motor-generator . Next is the

control chassis, (13) , for the current ~eg~lator and the power supply,

(14), for the earth's field compensation coils. The vacuum gauge, (15),

is of the cold cathode ionization type. The decade scaler, (16), and the

counting rate meter , (17), are connected through the cathode follower ,

(18), to the Geiger counter . The counting rate is recorded on the strip

chart recorder, (19) .

The t wo m&L~ magnetic field coils are connected in series

for the lower hal: of the spectrometer range (2 kev to 50 kev) and are

switched to parall el fo~ t~e upper half (50 kev to 120 kev) . Except

for this switching opera~ion t he spectrometer, once started, operates

semi-automat ically, shutting down at either end of the range.

The ultimate calibration of the 180° spectrometer depends

on the energy measurements of the curved-crystal gamma-ray spectrometer.

II Detailed Description of the Spectrometer and its Construction

A. Main Field Coils -- Construction and 3>ecifications

In the prel~tnary design of the 180° spectrometer the field

1-ras to be provided by a pair of Helmholtz coils 68 em in diameter and

spaced 34 em apart . This produces a comparatively homogeneous field in

the central region. At the central electron trajectory, (r = 12 em),

the gradient of the field would be o.S% per em. At the widest portion

of the electron beam (2 em) the variation 1-.rould be 1%. Calculations

indicated that the homogeneity in the region of the trajectories could

be improved by decreasing the distance between the V.-10 coils . It was

found that for a spacing of 29 em ( .8S x me~! coil radius) the variation

in magnetic field, at radius 12 em, was 1 part in 1000. This gradient

is of the proper sign to contribute to second order focusing. A de­

tailed discussion of the field distribution is given in the Appendix.

To achieve the desired field distribution, each turn of wire

in the upper. coil snould have a matching turn in the lower coil of the

same radius, separated o~r . 8S time their radius . To approximate this

condition each coil is ~-round ui th a cross section of .85 inches by

3.86 inches and the l ong dimension lies along t he generators of a cone

as shown in Fig . 2. The specifications for the main field coils are

given in Table I .

The internal surface of each cast brass coil form was

machined to the same dimensions . Each £'om was lined with 20 mil

fish paper. The winding was done on a large turntable with provision

for maintaining a uniform ~·rire tension. Each layer consists of 22 turns.

To maintain uniform winding, strips of lO mil fishpaper were laid in

every fifth layer. The completed Ninding consists of 91 layers .

Experiments with a model indicated that sufficient cooling

could be achieved by filling the coil form with transformer oil and

passing water through one ~urn of S/16 inch diameter copper tubing

soldered to the inside diameter of the coil form.

rc= coil radius (34cm.)

re= electron trajectory z radius ( 12 em.)

L = co i I spacing (2 9 c m) l r c

~-------------------~-~~ ~ ~

cast brass coil forms ~[---___ I /

'lc: •• __ .../-"'_

~

~ ~ 3.865-A I > ~ inches\ <.

\¢? ~ ~5 inch

L

~

Fig. 2. Cross-section of main magnetic field coils. To achieve maximum homogeneity at the radius of the electron trajectory, (r ). the coil spacing, (L). is fixed at 0. 85 times the coil radius, (r ) . e

c

I V\

I

-6-

TABLE I . SPECIFICATIOKS OF THE MAIN FIELD COILS AND PO:JER SUPPLY

Power Supply

Current Re~1lator

Main Field Coils

Hire size

General Electric r·iotor Genera tor Set

Motor 2 HP, 220 volts , AC 3 ¢ induction tne

Geaerator 1 ~1, 250 volts , 4 a~eres full load (shunt field connected to current regulator)

Rotating electrcmechapical servo system

Current staoility - - 2 parts in 10,000

#19 copper wire (formvar coated)

Number of turns (each) 2000

Resistar.ce of each coil

Insulation

Cross section of each coil

Coil form

Cooling

Field strength (pair)

Total oil expansion (each)

Temperature rise (max .)

113 . 4 ohms at 26° C

heavy formvar coating and fishpaper

.855 inch x 3.865 inches

cast brass (iron free)

oil filled, water cooled form

0 to 120 Gauss

-7-

To seal tte outer edge of each form two cast brass rings

Here machined with 110 11 ring grooves (Fig. 3) and clamped in place . A

l/2 inch x l/2 inch groove w~s r.achined in this outer ring to provide the

coil form for the earth's f~eld compensation discussed in Section II, c. A flexible bellows provides for the expansion of the oil, since the sys­

tem must be closed, to prevent moisture from entering the windings . The

two coil leads are brought o~t to terr.Jnal blocks as shown.

The completed coil assembly was evacuated through the pumping

port for a sufficient period to insure the elL~ination of air and mois­

ture. Current was passed th:'ough the vrindings to aid in the out- gassing

process . The transfo~ner oil Has out gassed in a separate vacuum chamber

to elininate air bubbles . The evacuated coil \·Tas then impregnated Hi th

the treated oil through the oil filling hole and the holes were sealed.

The coils iiere mounted on their adjustable supports as shmm

in Fig. l. The loi·Ter coil \vas leveled with a precision level in tHo

directions and centered with respect to a fine plumb line suspended from

the top of the main frame . The upper coil Has spaced by the use of an

indicator mounted on a rod 1-rith one spherical end. The tHo coils were

lined up axially by the use of a plumb line held against the inside

diameter of the upper coil at seve~al places .

wa

ter

be

llow

s sh

ield

\.

/ ..

coo

ling

tu

be

oil

. e

xpa

nsi

on

ch

am

be

r

term

ina

l b

lock

\

oil

fillin

g

hole

\ e

art

hs

fie

ld

com

pe

nsa

ting

co

il --

ma

in

fie

ld

win

din

gs

Fig

. 3

. C

ross-s

ecti

on

of

main

mag

neti

c f

ield

co

il.

I 0

>

I

- 9-

B. Current Regulator

A direct neasurement of the magnetic field (0 to 120 gauss)

is very difficult over the entire r~~ge . The proton resonance method is

practical dovm to approxinately 50 gauss . Below this value the requisite

volume of the proton resonance ~~~le becomes prohibitively large . This

range (0 to 50 gauss) represents, in the present spectrometer, the elec­

tron energy region (0 to 65 kev) of greatest interest in the planned

research program.

An electromechanical rotating unit vias found to be most ade­

quate for the present pur? ose . This rotating unit operates as a current

regQlator rather than a direct fie~d regulator . Since there is a large

overlapping momentum rang<; betl·reen the present instrument and the Ring

Focusing Beta-P~y Spectrometer , and since the latter instrument is pro­

vided uith proton resonance units to measure its magnetic field, the

momentum scale of the present instrument can be readily established in

terms of the current in the field coils . The primary calibration of both

instr~~ents rests on the wavelength measurements made with the curved

crystal ganuna- ray spec·0rometer .

The pr~n~ry pmrer for the 180° spectrometer magnetic fiel d

coils is obtained from a motor- generator set operating from 220 v ac

and proYiding up to 4 amperes at 250 v de. The spectrometer field

current also flows through a small solenoid mounted on the base of the

current stabilizer (see Fig. 4) '1-·Thich provides a magnetic field propor­

tional to the spectrometer field . A ci rcular coil rotating in this field

forms a small ac generator . This is the sensL~g generator . On the same

rotating shaft is another coil operating in a fixed magnetic field pro­

vided b~r an Alnico Jllagnet . This is the reference generator . The output

of the reference generator is divided by a precision potentiometer

mounted on the same rotating shaft . The divided output of the reference

generator and the output of the se~ing generator are connected in series .

The sinusoidal signals from the reference and sensing generators are

adjustable to be exactly 1eo0 out of phase so as to furnish the error

signal vThich is used to stabilize the current produced by the main de

generator . In order to get the error signal out of the rotating system

-10 -

":1-lu'

,------------------ -------- - ----------------------------, I I I I I I I I I I I I ____ _________ __ __ I

CONTROL. OIA.L.

SPEC.TROME1"ER ~ F lELOc:.QILS

I L __

A--tA&u•.IG '-'0\..."TAG.a UEC~AN\CA..L U NIT

r/

i,-1------, I I I I L_ ___ _____ _

lt,HXX:.TtON MOTOR aHP

Fig. 4. Current Regulator showing the mechanical unit and the system block diagram.

-11-

to the external circuits a transformer is used whose primary winding

consists of a circular coil of l-vire coaxial w-ri th and rigidly fastened to

the rotating shaft . The secondary \vinding is also coaxial Hi th the shaft

but is stationary. This trffi1sformer avoids the necessity for slip rings

and brushes •-<hose sliding contacts might introduce i..'1stabili ty in the

error signal .

The ac error signal null indicates equality of the sensing

generator output and the divided output of the reference generator . -The

dividi..'1g potantiometer is a ten- turn Helipot vihich, as before stated,

rotates with the shaft . The setting of this rotating potentiometer is

made through a double differential gear train .

Since the e~or signal is an alternating signal a reversal of

its algebraic sign would be indistinguishable in the external circuits

unless a reference phase signal from the rotating shaft is available .

This phasing voltage is obtained from a vari&ble reluctance generator

operated by the same shaft as the sensing ~•d reference generators . In the external non- rotati ng circuits the ac error signal is amplified and

rectified in a phase s ensitive detector in which the phasing voltage

from the variable reluctance generator serves as a phase reference . This

detector furnishes at its out~ut a correctly polarized de error signal.

Together with a fixed de voltage the de error signal biases control tubes

which provide field current for the primary pm-ver generator .

The system constitutes a servo-mechanism which regulates the

spectrometer field current by tending to maintain equality between the

sensing generator output and the divided outnut of the reference genera­

tor. Control of the stabilized current is achieved by operating the

dividing potentiometer. 7he relative accuracy of current measurement

with this system depends on the constancy of the reference generator out­

put , the sensi tivity of the sensing ger.erator ffid the linearity of the

dividing potentiometer . Table II gi ves the specifications of the regu­

lator components .

Tests have shovm the inherent stability of the system to be

\..Ji thin + 2 parts in 10 ,ooo. The lineari t:r is determined by the precision

potentiometer which is claimed by the manufacturer to be linear within

~ 2.5 parts in 10,000.

-12-

TABLE II SPECIFICATIONS FOR CURRENT REGULATOR COMPONENTS

Earth's Field Compensation C0ils for Sensing Generator

Vertical 1300 turns of #33 formvar wire; 820 ohms; mean s~\are side 8-3/16 inch; .3 gauss

Horizontal 1000 turns of #33 formvar Hire; 463 ohms; mean square side 6 inches; .3 gauss

Reference Generator JL~ature

1400 turns #40 enameled copper; 370 ohms; mea..1 diameter 7/8 inch

Sensing Generator Armature

2800 turns #40 enameled copper; 2300 ohms; mean diameter 2-3/4 inch

Sensing Generator Sole:.:.0id (each of tlvo coils)

216 turns #20 heavy formvar copper; 1.8 ohm; 25 gauss at 2 amps

Rotating Transformer (mild steel core; air gap~ 10% of total reluctance)

Primary

Secondary

11,000 turns; #36 enameled copper; 2200 ohms

16, 500 t Jtl38 1 d 800 hm urns ; , ename e copper; o s

Variable Reluctance Generat or (each of tl-ro cojls connected in series)

11,000 tun1s #38 enaT"leled copper; 3500 ohms

Magnet (shaded) is \·Tater hardened Ketos . By momentarily connecting ungrounded side of phasing connector on control chassis to + 380 V de the Ketos steel is magnetized to proper strength and direction . Renainder o: core is mild steel. Haximum series O'~tput of generator (vrith 0 .5 mfd. across output) is about 8 V ac with minimum clearance bet1·1een wedge armature and core . Normal output set at 3 V ac by increasing clearance.

-13-

C. Compensating Coils for Earth 1 s Magnetic Field

The sensing generator output of the current stabilizer is

affected by the t1·10 components of the earth's magnetic field perpendicular

to the shaft . The component parallel to the shaft has no effect at all .

The horizontal component, perpendicular to the sensing solenoid, causes

a phase shift between the sensing generator and reference generator out­

puts . The vertical component causes a change in calibration of the

regulator . To eliminate the stray field components tuo pairs of Helm­

holtz coils are mounted around the sensing solenoid. Table II includes

the specifications of these compensating coils.

Adjustment of the current in the Helmholtz coils is achieved

by disconnecting the refe~ence coil and running the sensing generator.

The output i s observed on an oscilloscope and a null is obtained when the

stray field is cancelled out.

Since t he main spectrometer field is not measured directly,

it is also necessary to cancel out the vertical (axial) as well as the

horizontal components of the earth's magnetic field over the region of

electron trajectories .

The horizontal component of the earth's field is cancelled

in this case by the two pairs of square Helmholtz coils seen in Fig. 1 .

Each pair consists of t vro scr1are frames made of aluminum channel. The

corners clip together and each coil form is notched so that the four

coils clip to each other and to the main framework of the spectrometer.

The vertical component of the earth's field is cancelled by

a pair of circular coils wound on the sealing rings of the main coils as

shown in Fig. 3.

An air-driven generator is us ed as the detector to determine

the current settings required to eliminate stray magnetic fields in the

vicinity of the electron trajectories .

Table III gives the specifications of the three pairs of

spectrometer compensating coils .

-14-

TABLE III SPECIFICJ..TIONS OF SPECTROHETER CO?-iPENSATING COilS

Each Coil Round Coil (Vertical) Square Coil (Horizontal)

\vire size #33 (formvar coated) #33 (formvar coated)

Number of turns 1000 1000

Resistance 2100 ohms 1600 oh1ns

Insulation Glass tape Glass tape

Cross section 1/2 in. X 1/2 in. 1/2 in. X 1/2 in.

Coil form Cast brass Aluminum channel

Field strength (pair) 0 to 0. 5 gauss 0 to 0.5 gauss

- 1.5-

D. ~ectrometer Chamber

The vacuum chamber is machined from a solid aluminum forging .

The front is provided Hith t~vo round holes, one for the source and the

other for the Geiger counter . The source and the counter are each

mounted on circular aluminum plates which cover these holes and clamp to

the chamber with thumb screws . The mounting plates are located by bronze

dowel pins as shown in Fig • .5a. The solid angle of the electron trajec­

t ory is determined by the \·lidth of the diaphragm shmm in Fig • .5b . To

change the solid angle the diaphragm is slid out and an appropriate one

inserted. A vacuum lock surrounds the counter to protect the thin window

when the chamber is open to the atmosphere. The clapper valves on the

vacuum lock and in the main vacuum line (this valve may be seen in Fig. 1)

are closed during the insertion of the source. A high vacuum port is

provided in the vacuum lock so that a small leak over a long period of

time would not endanger the counter l-Jindow. Fig • .5 also shows the lead

shielding betHeen the source and the counter .

1 . Source

The s ource ~s prepared by first evaporating a layer of the

radioactive material , of the order of atomic thickness, on one side of a

thin mica sheet . A thin layer of aluminum is evaporated over the source

material to remove the positive charge that builds up on the source as

the negative electrons are ejected. I n cases vrhere the absorption by

the aluminum is significant, the process is reversed: the aluminum is

evaporated directly on the mica sheet and then t he radioactive material

i s evaporated on the alUML~~~ surface . For a standard momentum resolu­

tion of 0 . 8% a strip is cut from t he prepared mica sheet 0 .1 em x 4 em.

This strip is cemented to an alumim.un support between two engraved lines .

The aluminum support is located on the source holder with two pins ~~d

held by two scrm-1s as shmm in Fig. 6.

·-16-

Fig. 5. Spectrometer chamber (with cove r removed).

--17-

Fig. 6. Radioactive source and holder.

-18-

2 . Geig~ Counter

The Geiger counter (shown in Fig. ?) is machined froM one

piece of aluminum. The width of the vrindow is varied for different reso­

lutions by replacing the front plate viith one having the desired aperture

width. The window itself mus t be thin enough to allow transmission of

lm1 energy electrons and yet must support a pressure of approximately

10 em of Hg. The vrindow material used is formvar . A solution of 4% fo~var in ethylene dichloride is prepared. A clean glass plate is im­

mersed and witr.draHn and the solution is allowed to run off, leaving a

thin film on the surface of the glass . '\ilhen the formvar film has set it

is peeled up at the edges by the use of a razor blade and floated off in

a beaker of water. The aperture ~late is passed under the floating film,

and the film adheres t o the plate as it is removed from the water .

The 4% formvar solution gives a windmv thickness of the order

of 10 1-Lg/ cm2, Hhich allot-Is trans.11ission down to an electron energy of

2 kev. Usually tlvo layers of formvar film are used to minimize the leakage

of counter gas vJhich may occur 1-1i th s uch a thin film .

The filli~b gas for the coa~ter is a mixture of 2 em Hg of

alcohol vapor and 4 em Hg of argon.

} . Installation of the ~ectrometer Chamber

A flat cover completes t~e ch&llber assembly. All mating sur­

faces are sealed vTi th neoprene 110 11 rings except those in contact with the

alcohol- argon mixture . These are sealed Hith teflon gaskets to p revent

contamination of the colli!ter gas with dissolved hydrocarbons from the

rubber "0" rings .

The chamber asser:bly is mounted on a base plate with bronze

screws and located by t >vo bronze dm.;el pins . The base plate is adjustable

in the vertical di::-ection by the use of three adjusting screws and in the

horizontal direct ion by slots in the base plate and the main frame work.

The base plate is made parallel to the plane of the main field coils and

its vertical position set by the use of an inside rr~crometer . The chamber

is clamped to the base plate and is lined up coaxially Hith the main coils

by use of a plumb line suspended from the main frame .

-19-

Fig. 7. G. M. Counter

-20-

E. Vacuum ~ystem and Geiger Counter Filling ~ystem

The complete vacuum and counter filling system is shown

schematically in Fig. 8. A Welsh 30 liter/min. duoseal mechanical

pump provides the fore vaaL~um for the 60 liter/sec. oil diffusion

pump , 1-1hich achieves a vacuum of the order of 5 x lo-6 mm of Hg.

The cold cathode Phillips type ionization gauge has a range of 0.5 to lo-7 mm of Hg and is unaffected by atmospheric pressure . A set

of baffles protects the Phillips gauge from contamination by the oil

vapors from the diffusion pump. At B is a clapper type . 10\-T impedance

valve which is closed when the source holder is removed, leaving the

spectrometer c hamber open to the atmosphere. During this period the

vacuum lock clapper valve .b. is also closed. Valve C is al-vrays open,

mai ntaining a vacuum around the outside of the counter window except

when the windorr is being replaced. These three valves (A, B, and C)

allm1 the spectrometer to be opened and still maintain a vacuum in the

rest of the system, rett~cing the pumping time required to evacuate the

spectrometer chamber again.

Valves C, D, E, and F are the diaphragm type -v1ith a Kel-F

seal (Hoke Hodel 48 2). V a: ve G is the diaphragm type with a neoprene

seal (Crane J.Vlodel 1600) . All lines are copper or glass tubing. Per­

manent joints are soldered or fuvar- fused to glass . Removable connec­

tions are sealed with neoprene "0 11 rings or Teflon gaskets .

To pump dorm the spectrometer chamber, valve G is closed

and E is opened. This permits the by-passing of the diff1sion pump ,

eliminating the necessity of an additional mechanical pump . After a

feH minutes the by-pass valve, E, is closed, G is opened, and clapper

valves A and B may be opened.

To r emove the source, t i1e cl apper valves A and B are

closed and valves D and Fare openedtobring the spectrometer chamber

to atmospheric pres sure.

To r eplace the Geiger counter '"indm-1, with the vacuum

pumps in operation and valves C, B, and E closed, the following

procedure is follov1ed: Install the ne-w window and clamp the counter

in place . Close valves G and F. Open valves A, D, K, and H. Open

I I I L

G. M. counter

-21-

SPECTROMETER CHAMBER

ballast tank

B

baffle

diffusion pump

G

-, I I I

_J

E

COUNTER FILLING SYSTEM

H

I

mono­alcohol meter

._- _.J

Bourdon gauge

D

fore pump

Fig. 8. Schematic diagram of vacuum and G . M. counter filling system.

F

atmos. or

argon

-22-

valve E slowly to pe~~it the interior of the counter and the spec­

trometer chamber to be pumped dovm at the same rate. The Bourdon

Gauge gives a good indication of the pu."!lpi.."lg rate . \'lhen the system

has been rcughed dmm to 0.5 mm of Hg, close valve E and open G and

B. This allrn-vs the diffusion pump to continue the evacuation more

quickly.

To fill the ccunter, close valve D and K. Open valve I

to the alcohol until the mercury ~anometer reads 2 em. Close I and

open J slowly to admit the Argon until the manometer reads a total of

6 em.

... 2J-

F. Detecting and Recording ~ystem

With the exception of the Geiger counter and the pre­

amplifier, the entire detect ing and recording system is built up of

commercial units . The pmver supplJr of a decade scaler (Atomic Instru­

ment Co., Model No. 1010) provides the high voltage t o the counter .

The counter pulses are a-mplif ied in a prea-nplifier with a cathode

follower stage to shape the pulse for the counting- rate meter (Detecto­

lab, Model .!liT) . The rate mete:· output voltage is fed to a strip chart

recording potentiometer (Heston, 11odel No. 6701) . The chart is driven

by a synchronous motor as is the control knob of the current regulator

for the spectrometer field . Each complete turn of t h e control knob

sends a signal to the recording potentiometer so that an automatic

record is made of beta- ray momentum versus counting rate.

-24-

III PERFOffi.iAUCE CURVES

Z2

-

..J

..J

..J

~

~

z 2

0 co

o

o

0 rt

) N

0

N

N ~

CD

CD

CD

It

) -

-N

.

. .

• .

. .

. •

co

0)

rt)rt)

rt)

0 rt

) rt

) co

0

0)0

)

0)

,., CD

CD

- -

\ I

I

\ I

I

I I I

I I

1 I

- .5.5

.J~

N

N

NC

O

---

~

. .

• .

rt)

rt)

rt)

0)

CD

CD

CD

0 -

\ I

I I

\ I

I

\ I

I \

I I

I II

I

I II

I

'-'-

G)

101

G)

0 ~

~

<(

~

<(

0 ~

CD

-'

. ~

..J

rt)

~

~

0)

I I

I

•• 'r

" ••••••••••••••• '1

'1' •

• 1..

'r"'

...

....

. I

I I.

I I

I I

I. I

I

. ...

....

....

....

....

.. ,. t

...

... , .

....

.. l'

.. ( .

.. j''

.•.

... r

I .1 ••

•• n· ..

. ~ .. , .....

.. ., .... 1' •

'I" •

• •

... .,

. , "

"'T

...,_

• •

• I

• r

I I'

I II

•

. .

' •

. .

• I

I'

Ill

I I

I

I I

I I

I I

A

I I

I •

I "

I I~

I =

..

I I

I' j I

I'

I

~ I

• I

.. "

I ••

I"

I I

• I

• I

• I

. I I .

~···

····

t····

··· ·1

····· ..

~·· ...

. ·i" .

... ·I·

..... ·

t .....

I·~··

. I ... i···

...... i

··· ....

. I .....

.. ·t ....

.... t .

.....

.. , .

......

·t .....

.. ·t ..

..... ·t

........ 1

·· 1

.80

1

.70

1.

60

1.5

0

1.4

0

1.3

0

1.20

1.

10

1.0

0

Fig

. 9

.

CU

RR

EN

T

IN

AM

PE

RE

S

Co

nv

ers

ion

ele

ctr

on

sp

ectr

um

em

itte

d b

y a

Y

b 16

"9 so

urc

e.

Th

e p

eak

s d

ue t

o

gam

ma t

ran

sit

ion

s i

n T

m1

69

are

id

en

tifi

ed

in

th

e f

igu

re.

Th

e m

om

en

tum

re

solu

tio

n i

n t

his

ex

peri

men

t w

as

0.

8o/o.

T

he e

... e

rgy

ran

ge i

s

33

to

83

kev

.

j

1\)

\.11.

I

Fig

. 1

0.

5 w

~4

a::

t93

z 1

--2

z ::>

0 u

2 5

w4

~ 0

::3

t9

z2

1-

­ z :>I

0 u

3

A

3 B

~

(\J

rt)

(()

4 5

.....

~

(\J

~

<X)

6 E

NE

RG

Y

IN

KE

V

4 5

6 7

EN

ER

GY

IN

K

EV

7

... 0

z

- I (\

J ~

~z

<X

)(\J

~

8

8

9

9

Tm

'69

... _

j

1.{)

1:1

1"-

:_j

0L

l)

C\J

I':

0 (\J

a _

j ~

0 (\J

10

II

12

Lu

l75

10

II

12

In F

ig.

10

-A i

s s

ho

wn

th

e T

m 1

69

beta

-ray

sp

ectr

wn

, fr

om

2 t

o 1

2.

5 k

ev

fo

llo

win

g

the d

ecay

fro

m Y

b1

69

, as r

eco

rded

on

th

e c

hart

reco

rder

fro

m t

he o

utp

ut

of

the

sem

icir

cu

lar

sp

ectr

om

ete

r.

Th

e u

nla

bell

ed

peak

s are

pri

ncip

all

y L

-Au

ger

lin

es.

T

o i

den

tify

th

e T

m L

-Au

ger

lin

es

in t

his

sp

ectr

wn

, an

in

dep

en

den

t st

ud

y o

f th

e

Lu

L-A

ug

er

sp

ectr

um

em

itte

d b

y H

f17

5 s

ou

rce w

as

un

dert

ak

en

, si

nce i

n t

he L

u1

75

sp

ectr

um

no

co

nv

ers

ion

ele

ctr

on

s o

f lo

w e

nerg

ies

are

pre

sen

t.

Fig

. 1

0-B

rep

resen

ts a

ty

pic

al

ch

art

reco

rdin

g o

f th

e L

ul7

5 L

-Au

ger

sp

ectr

wn

.

I rv

0'­ i

9 8 7 6 5 C

l) -0 ~ Ol4

c -c :::

J 8 3 2

N1

Nml

NIV

,V 0

l l M

1v,v j

Mm

l Mi'

M1 ~

a~--~~--~--~~--~~~~--~~--~--~~~

.520

.5

00

.4

80

.4

60

.4

40

.4

20

.4

00

F

ield

cu

rren

t in

am

pere

s

Fig

. 1

1.

Co

nv

ers

ion

ele

ctr

on

s o

f 8

. 4

2 k

ev

gam

ma r

ay

in

Tm

16

9 e

xcit

ed

by

Er

16

9 b

eta

d

ecay

. P

ron

ou

nced

Ml,

M

il,

an

d M

ill

peak

s w

ith

6.

1,

6.

33

, an

d 6

. 5

3 k

ev

ele

ctr

on

en

erg

y r

esp

ecti

vely

can

be s

een

, in

ad

dit

ion

to

N a

nd

0

co

nv

ers

ion

p

eak

s.

Th

e s

ou

rce u

sed

in

th

is e

xp

eri

men

t w

as

ob

tain

ed

by

vacu

um

ev

ap

ora

­ti

on

on

an

Al

co

ate

d m

ica s

heet,

as d

escri

bed

in

th

e t

ex

t.

Th

e l

ines

ob

tain

ed

are

pra

cti

call

y f

ree o

f re

tard

ati

on

"ta

ils 11

at

low

en

erg

y s

ide.

Th

e s

ou

rce w

idth

w

as

1 m

m.

Th

e c

ou

nti

ng

rate

on

th

e p

eak

of

the M

I li

ne w

as

ab

ou

t 1

00

0 c

ou

nts

p

er

min

ute

. F

rom

th

e p

resen

t M

su

bsh

ell

co

nv

ers

ion

sp

ectr

um

it

can

be c

on

­clu

ded

th

at

the 8

. 4

2 k

ev

tra

nsit

ion

in

qu

est

ion

has m

ixed

MI

+ E

2 m

ult

ipo

lari

ty.

I 1

\)

-.J I

w .....

5 4

<I

3 a::

(!

) z .....

z :::>

0 2

(.)

Pt

Lm-M

v M

1v,v

~

Pt

Lm-M

v M

m

Os

L111

-Mv

M1v

,v \

Os

Lm-M

v M

m

Pt

Lm-M

v M

u " --

Pt

and

Os

L-A

ug

er

Pt

Lu-M

v M

1v, v

/

O

5 6

7 8

9 10

II

12

13

Fig

. 1

2.

EN

ER

GY

in

ke

v

L-A

ug

er

ele

ctr

on

sp

ectr

um

em

itte

d f

rom

an

Ir

19

2 so

urc

e.

Sin

ce I

r 1

92

decay

s b

y

ele

ctr

on

cap

ture

to

Osl9

2 a

s w

ell

as b

y b

eta

decay

to

Ptl

92

, A

ug

er

ele

ctr

on

s o

rig

inati

ng

in

Os

an

d P

t ato

ms

are

pre

sen

t; t

he P

t A

ug

ers

are

, h

ow

ev

er,

d

om

ina­

tin

g.

Th

e s

ou

rce u

sed

here

was

pre

pare

d b

y v

acu

um

ev

ap

ora

tio

n a

nd

has

lin

ear

dim

en

sio

ns

of

on

ly 0

. 5

mm

tim

es

7 m

m.

Max

imu

m r

eso

luti

on

of

0.

8%

is a

ch

iev

ed

, b

ut

in v

iew

of

the c

om

pli

cate

d s

pectr

um

bett

er

reso

luti

on

wo

uld

be d

esi

rab

le.

Ten

tati

ve i

den

tifi

cati

on

of

the p

eak

s is

giv

en

.

N

())

I

-29-

APPENDIX

Magnetic Field Study

When the main magnetic field coils were assembled on the

spectrometer framework th~ field was mapped using a one inch diameter

flip coil and a ballistic galvanometer . Fig . 13 shows the axial com­

ponent of the field plotted along a line in the Z = 0 plane, passing

through the Z axis of Fig. 2. Eacn curve represents a different coil

spacing (L). Curve (a) shows the field obtained with the coils spaced

according to the Helmholtz condition, L = r , where r is the radius of c c each coil (r = 34 em). In curve (b) the coils are moved closer together

c (L = 32 em) . The maxlimun is no longer at r = O, but is at approximately

r = 5 cmo In cu~ve (c) (L = 29 em) the maximum is at r = 12 em~ This

is the optimQ~ spacing for the given coil diameter (63 em) and an elec­

tron trajectory radius of 12 em. A spacing of less than 29 em moves the

maximum out beyond r = 12 em as sho1-m in (d) .

If points 11?1'1 in Fig. 13 are connected a curve is obtained

which gives the radius of tha ~~ular region of uniform magnetic field

versus coil spc-.cing. Fig. 14 shows this curve with dimensionless co­

ordinates. For a given pair of coils, multiply the coordinates by the

coil radius and read off the proper spacing for any radius electron

trajectory.

This was done for the present case, using the mean coil

radius (34 em) and the mean radius of the electron trajectories (12 em)

which gives a coil spacing of 25.4 em. A plot of the axial component

of the field was made in a plane through the axis of the coils. Fig. 15 shows this as a family of constant flux lines . The rectangle represents

the cross-section of the electron trajecto:i:'i'3s at the Hidest part of the

beam. Point M is the max.imu.'ll shovm in Fig . 13.

The gradient that exists in the region of the electron

trajectories contributes to second order focusing of the electron beam.

If the magnetic field ~ere perfectly homogeneous then, in Fig. 16 the

central trajectory, c, would enter the counter window slightly to the

-30-

Appendix, continued

right of trajectories A and B. With the field distribution as shown

in Fig. 13-c, the field at A and B is less than at C and thus rB and

rA become ~eater than rc' moving the focus of the extreme trajectories

closer to the focus of the central beam c. In the present instrument

the gradient is not sufficient to achieve complete second order focusing.

-31-

M

{a) L= 34 em. r=O

M

{b) L= 32 em.

r = 5 em.

M

{c) L= 29 em.

r = 12 em. M

{d) L= 28 em.

r = 14 em.

Fig. 13. Curves of axial component of magnetic field plotted along a horizontal line, through the axis, midway between the two main coils. Each curve represents a different coil spacing.

-32-

0.6--------------------------------------~

rM~ rc

0.5

0.4

0.3

0.2

0.1

rM= radius of toroidal region of homogeneous field.

'\. rc =radius of field coils.

' '\. L = coil spacing. -'

0 ~------~------~--------------_. ______ __ .75 .80 .85 .90 .95 I. 0

L Ire

Fig. 14. Radius of the toroidal region of maximum magnetic homogeneity for a pair of circular coils versus separation of the two coils. The curve is obtained by passing a line through points marked 11 M 11 in Fig. 13. The dotted portion is not too useful because the height of the uniform region becomes less than the width.

Z a

xis

~1Cl

> ~Cl

>' ~Cl>

t>c ~ro

~

. q

en

N

\0

t"'"

en

0'\

0'\

en

0'\.

0'\.

0 0 o. .....

4t==

= 4

, 1

/ 4

1 l

1 ~

\ \ "'

r-~

+~~~~·

....

.. ~

+2

0 r

I I

I I

I I

I I

I I

' I

• II

I I

I

-2

-4

2 4

6 8

10

12

14

16

em.

radi

us

Fig

. 1

5.

Plo

t o

f th

e a

xia

l co

mp

on

en

t o

f m

ag

neti

c f

ield

in

a v

ert

ical

pla

ne t

hro

ug

h t

he

cen

tral

ax

is.

Th

e d

ista

nce b

etw

een

th

e i

nd

ucti

on

lin

es

rep

resen

ts a

fl

ux

d

iffe

ren

ce o

f l.

5 p

art

s i

n 1

00

0.

Th

e d

ott

ed

recta

ng

le i

nd

icate

s th

e m

ax

imu

m

wid

th o

f th

e e

lectr

on

beam

.

I w

w i

-34-

B

Fig. 16. Second order focusing.

FA,B Fe counter

lj ;,n:

, -~v=;

'' ........

...,.

MN

IP."

T0

""

~a

-~I

.~M-

' -L

_ .... ~ ..

t;C

. ti

R'-

iO ..

.

•tO

!: R

(C.

AMP

~IFI

I!.R

~·~ 2-

""'-

PH

AS

ii.

O~T~CTOB

I() I

<

4.1

"-

1()

..

aiR\•~

~~liCTION

NE

.TW

OR

!:<,

~ ~~

IICJ

usv""

':j c

I I

\ -

~

~----------~----------7 ·•~

oo~v~o~

e~u~N~

R~.~&~.

----~~ ~

5100

o<.

~~~

·or ·T

~ ••ao~K

FR

OM

N

&G

AT

tva

s

•o•

CSE

:Ne.

R.-

..TO

A

OU

TP

U'T

C.O

LL

KT

iiD

O

IAG

AA

M

OI .

.. ,.I

NG

t>

C

OR

A.C

TID

N "-laTy,QR~

6.av

...c

C'I

LA

t•,oU

iNT

'W

.ItM

A

MIC

><

AN

ICA

Lo

li

N IT

J

i..

<o

a«

ou

<A

-n;>

H.

..J I

--------------

1 I

I I

-,

I

I&L

iflll

lliiiT

IUT

E•T

EC

IIIL

IU 1

=:

CONTRO~ CHA~S

I<;!,

tJ I

NT

&I'

ICO

NN

EC

TIQ

NS

":

01

Fig

. 1

7.

Co

ntr

ol

ch

assis

an

d i

nte

rco

nn

ecti

on

s fo

r cu

rren

t re

gu

lato

r.

I \..

>.)

V\ I

"'*U

+4

00

To

Sli

O V

OC

FU

LL

TO

NO

1..

0"0

eo

mp

. ~

~

II I

+2

.UII

I:IC

.

~~~~~4GV

I I

6 I

I ....

. I

Q

·---.-0

+N

O

VO

C.

VN

Ra

&

II liE

~~v~

um

p.

f __ ~

f r

I I

MI;~IT P

..fi

i7S

IU

iO·C

! ·!>

1>0

HiO

mo

..

pM

T <?~

-11--s

lle-. ~

·-~

0~

loow&~

Mk~IT

Ai:

l<l7

1 ~CO ••

VA

C "f

e.-

6J

L4

~

.. II

..A

M&

NT

Fig

. 1

8.

40

-10

5 V

CC

'R&

G

---

=-===

I ::::1

,. -

CILI

FIIII

IIIIT

IJIT

E•TE

CIIIL

III :

::::

.: ..

----

--·-

...._

_..

~---

·'<,7

--.,.

..,._

,. PO\N'E~

CH

AS

SIS

--..

, .. ~-

, ...

.. ~,_ 'r

/ ... -

. ....

v--

~-

~--

I~IG/

--P

ow

er

ch

assis

fo

r cu

rren

t re

gu

lato

r.

I '

\...V

()

'.

I