Lawrence Berkeley National Laboratory

Recent Work

Title
THE ADIABATIC MOTION OF CHARGED PARTICLES IN ELECTROMAGNETIC FIELDS

Permalink
https://escholarship.org/uc/item/7x73871x

Author
Northrop, Theodore G.

Publication Date
1961

eScholarship.org Powered by the California Digital Library

University of California
 ,
r I
.....

I
r,
. {
I!

~:)'\;~~~. '\'.
'·_~~ '-3~·.'-~w·· . .UNIVERSITY OF CALIFORNIA
lq : ",

Lawrence ~adiation Laboratory

w . ....

,\,
Berkeley, California

Contract No. W-7405-eng-48

..,_
RECEIVED
LAWRENCE
BERKELEY LABORATORY

AUG 21 1987 .
LIBRARY AND
DOCUMENTS SECTION

:
THF .ADIABATIC MOTION OF CHARGED PARTICLES
~' IN ELECTROMAGNETIC FIELDS
.
:._
~.. . .

Theodore G. Northrop

t.
'·

TWO-WEEK LOAN COPY

This is a Library Circulating Copy
which may be borrowed for two weeks.
 DISCLAIMER

This document was prepared as an account of work sponsored by the United States
Government. While this document is believed to contain correct information, neither the
United States Government nor any agency thereof, nor the Regents of the University of
California, nor any of their employees, makes any warranty, express or implied, or
assumes any legal responsibility for the accuracy, completeness, or usefulness of any
information, apparatus, product, or process disclosed, or represents that its use would not
infringe privately owned rights. Reference herein to any specific commercial product,
process, or service by its trade name, trademark, manufacturer, or otherwise, does not
necessarily constitute or imply its endorsement, recommendation, or favoring by the
United States Government or any agency thereof, or the Regents of the University of
California. The views and opinions of authors expressed herein do not necessarily state or
reflect those of the United States Government or any agency thereof or the Regents of the
University of California.
 -i- UC~L q:) i 7_.=..
·i
TABLE OF CONTENTS
.,
,, .• .

::a? Page
:: .... .... Introduction 1 .,
. ) '
II. The Guiding Center Motion
A. The Equation of ·Motion j,.n Dimensionless Form
·"· ,-
'
B. Derivation of the Guiding Center Equation of '

. ~.

lv1otion 5 · ...
~·
I
.
C. The Drift Velocity and the Longitudinal Equation
of Motion 9 .. ,
~ .·...
''
Do. The Geometric Interpretation of the Drift ,· :· i

·. '·' .. , ...',
1
Equation and the Longitudinal Equation of . ·.. ·. . . i
,.
Motion 20 '·
I.:.·.
'. E~ The Relativistic Case 25 .~ :.· ::. F .
·.
·~

III. A More Formal Derivation of the Non-Relativistic

i
Guiding Center Equation 33
..
. '•. ~

· .. I
'
,· ·<
., :.. ,:· . .·i
The Adiabatic Invariants of the Motion 38
'•. .'l'·j
'.:.' . Ao The Magnetic 1\1oment 4o . . •·. .1.
6 ·,,,

Bo The Longitudinal Adiabatic Invariant" 45

.. ·.: .... ·. ·~ H: f
'·, .. ' ... .·1
~··.' .··.. ·.
·,
·.
Co The Third or Flux Adiabatic Invariant ]! 6o ·. ''. >:' j.
. . \'·;.·. ,· .·' l
''
~;j,.. •' .
-~.

.. '
. .· 'l'
Do Adiabatic InvariantD to Higher Order · 69
. •' . !

.. ·· Eo The Adiabatic Invariants of Singly Periodic·

.:
H
.. ·.. '
·,: Systems
:.···..
.k... v.' Applications of Adiabatic Theory . ~ ·' ~. :
..'··~
r.:. ~ r;., . ·: ,· -.1 .. : ·.;.

1;'. '
/-.1
· A.. · . T'.ne Current Density in a Plasma,

Bo Lose Free Geometries ;· . '

Co ,The Geomagnetic Case ·

.
. ,· .. /~
j
..
·
'
··.· .·
.... :: '~ ·,:
L:
___ I·, .·· ... i
!._,,. . ,·:-· . ·' . ~- :.: ' ..... f.·
·.··,·1 \,
·,.
'·!:. ·': ;. .··. i
. .. ..
... ·"0:·
-

.· '
~

.
;
.
'1

-~' ' ' .

.

··'·
• ' I -~
.\ . ·.:·· ,\' ·: . ,.
,· ' -~

:. '! ~'
f· ·-, •. : . '
~
'\''. . • ·•
. f . . ·~ .. '. . ... ~
I
• f • '· • ; •• ' , ., , : ,, ,' •; 4 . I ' .' •• , • "···. ...
':· ~ ' ;. ' ' .•: ..• ' . i ;· . ·. :
/',:.

.'•
·-ii-
·..
·:
VI. Unresolved Problems 91
,·. ,., ., :,
A. Non ..Adiaba.tic Effects in Mirror Machines-- · ...
"i .:
.. ·
.. /.·
. ·: .. '. '
, ....,.-·;··
91 '.·:·

Numerical
.>·"('< '·. ·... .'1':'
.) I . •' ,{ ~

· · Be Non-Adiabatic Effecta-·Experimental 95 /
/'
.....
·.,

·'
·•,".;

·, '·
.· .
VIIo' 91
. .
...... ·.··
... ·
~: ...
References
·.l'
\ .
.' -;: :• :~ .

/ '.

..... . ·... ··, .. ·

.. ,
·
. ·•. \•

..~
... ' ·- ' . ' .. ,!
' ...
. ,.
,... ·
~
: ...
.'·~··.·; .
'i'. \

•' .,

• •• ·olli·
,
'·
.·•
~. .' ' ' .
.... ·,··,
· ... '

. '{

','t·

>; ,• •.·.
•:··
''
<'!'
).-.

f.i ··:: ··'..

; '~ ;.
·.,
.· .. · ..
...... .' ~ • .I
'·. :.,
•.·:· '·' •.i
. .\:
..··._. ' •
":.;'.i· ,. •'

. ,•
. ·~ .:'•·
.. ·. ·.,.. _ - c

....... ':,1·.··.:
~· ·.; ~ ,· ·,.
~ . . . ·.:.:"

:-,. . '-'.. ~
·.=. ;· ... :. . :, ~
... ··
'":. :.'
, ' ," ~
...
... ·..
 ·:,.-_.
'· : ·.' ·.. ··.-·... ·::·} ·.· ... · -~··:·
'·, ..

')
.
.. :. .. '. ~ .
'
• ·: i ''.
THE ADIABATIC MOTION OF CHARGED PARTICLES . _,.
. ,• ·,· ·. .:·.!. :·:, :·.
>··,' .'·. ·. ;. . ·.· :.:.,
m ELECTROMAGNETIC FIELDS . ··.... t .• , ·;,;
,.. ··.;
·.
'· !,
. • ..
. ,!,
:·,:,.:
. ... ~
' :_ .' . .'.' ... 1
.,._: ·.._. ~ :. ' ,;_'
..
. ··i
~

~ .,: -~-: . ·; . . ., . -~
a·.
' '

Theodore Northrop ·· . · ,' "-• • ~. ' '

" • ·, .";

•
.~...
.,~
::.

•• I • •
'' I

,,:,. 1• • '.'' •. :; .) , ; 'o

' ' I'

' !' '

•'

; :-~
'

-, . .' '•

. .. '.
Lawrence Radiation Laboratory··
· . ::UniV-ersity of California
, ·-'." ...._:'_.;:_·· ..:
i'-,
':··:. :\:···:::{'·:·:_:y·,..;.':.';:··
· ·, '. · . ·: · •· · ::.
;,
. ~~.. '-. ... '.: ..
.. .Berkeley, California ·:··:> ·}f;·:_.;.:·::. .J.
. . ",: (
··~
. :.: >-· ..
. >: ,;,'
1
. •,
1
,:

~
' ..
f • ; ,. • ..

:· '.'i
~~

r:., INTRODUCTION .
.,i

'·-: ··
....

.:: ~ ~ .
· <.:.
. : . ': ~. :
· The trajectory of a charged :particle in an electromagnetic field is in · ' : ·.. _: ::' · <:.·~\
' ' ·,

i> -'·.-..:.:· ,_,:; general. very complicated and must be obtained by a numerical. integration of' .the · · .. ::·j ·
< ' I~. ~·,: •j; • • ;' • }•

~'/ ~ / --) (~
• '

~
•

+. mg-4 (~
r, t ) , ·..·. ..,. . . .. ~ ·. . ·.:. ·. ...._, ~~·
' '• •' ':

eq_uation · o.~.. motion·.

. .., .\.1) dp dt =. ce )
dr dt X B r, t + eJ!i
(-4
r, t
)

, ' . ',' .. :' .. ··:;:

., -4 ·.•. . . . ~ .. . '
where r is the position of the particle, p is the relativistic momentum · ··~' . .:.;:, .··'<)
••• 4 _m0 r ~. ·Jt a:id Jt are J,~e •.-np.agnetie and. electric fields, m . is the
0
:: ~:;-
',·!~J
\'.

,·.·....
;-; ·: ... rest mass,
.-4 :'• I; ·~' ~ •,' '. ·.
'!· is · m/mry and g is the total non-electromagnetic force ~r
.. : ... ,.

·:unit mass. In special cases, analytic. solutions ·can be obtained, such as in · ,: ' .. : :.\.
': ~\'<);:;':,.I I · '
''•

'·· ::.··.':: '{' · ;:_::.:'( ... the trivial. case of the· uniform static magnetic field, rrhere the particle ,· I ~· •

·._:._ . /··/>.: . ·.. ·. gyrat~s hel~,abou~ . B. :possib~e

,. . ~·

.;.:.
in a Usually_ such solutions are only because· ·'.;· ..
f -~ • ; .... ' '

.•/ . ~~-
. ~

_In a general field, a Taylor ex:pa.nsion of r( t) ·. about the initial conditions '
,. . . '
.
would be practical only for short times. Suppose the particle is in a.n · ·..·.. ·..
... ·.,
approximately unif~rm nagnetic ·field (tme which varies. slm11y :l.n space and time ~- .·

;.,. . .. ,.::
slowly compared· to the gyration radius and period); the :particle motion is ."

.. ;:'· ,_r:
~ : ~ · · approximately helical. A Taylor expansion would be practical. only over a ·fraction > ·. ~·: :; ....\
;. ~~-

.. of a gyration period; a different approximation is needed if one wishes to foll<Yvr . ::. · .\,~'ij
J '
""; :' ·-·.
' •• • • -t'

. ~·.,
• .\ • • ; • • '-_. :,

.·
•

. · '. the particle over·; many gyration periodf3.. The liY:ration and motion parallel to · ·. · · -~·:: 1

;~~~:.. \ ·\:··.·~·:·:. the field lin:e should ber introduced explicitly into the expansion and derla.tions •: }•
. i:{~ ·:.f •.\:'
·. . . from. strict· helical motion treated as the perturbation~ . This is· the so called
·.
o lo ''•

.;~·v·. ~:
.
, . rguid.ing Center" Or . ''adiabatiC II apprO~~tiono ; The terminOlOgy tiguiding centeJ:'
' ; _;· ~ 911

. :::.'1 >/·' ; arises because in a slowly varying field the 'pa.:rticle moves a.:pproxi.mately in :a.
'.•

.
. ·... : ,·
.. ·-:
,··;._., .:.
~ I :
' '

•
..
•• •
.. '
~ . . ~: '~ .. •. ··.... ~.. ' ... ~ ":• ~ ~ .' . ·. .··.'
 2

The approximation would be of no use where the ·fields vary at a frequency

comparable rlth th~ gyration frequency_. In spite of the limitation to

:~ .;
. slowly varying fields 7 . there are manY applications. of the guiding·center . •-'"'-: ... '•

•. . . .
. '· ~: .< :: .·
~.

approximation, especial.ly. to plasma.phys'ics and to particle motion in t)le

• . . -:•••.• : t . '
··' .....
.; ; ..
~
'
. ':'.
' ..
terrestrial (Van Allen) radiation belts. 'i.

'.·.· /··
·.-
:;
Not only the equations governing the guiding center motion;
A

·any approXimat~ constants ("adiabatic invariants .. ) of the particle motion. ·.,.· ... ·:.r

will be usef'ul in applications~ :.Tb.e~'guiding center motion and the adiabatic

in~iants will be treated fully in .this·review.
·: ;_ ::; ·. ··..
It should be pointed out that the approximate (actually asymptotic)
(· ... ;.... solution and adiabatic invariants of the charged particle motion in an
electromagnetic field is a special case of a more general theory of'
. ': ;
:.
asymptotic solutions and adiabatic invariants of differential equations of
"
a. ·certain type o · ·The more general theory will be reviewed to. some extent ~
... ,··~··
.. ; . ' " ... ·.. ·\
' . · ·. 'i

. • ..
;.
.~.'
. .';,,'·. ,',·1

" ~.

·;
.·..· .. '. f

... ,,
•'

~ ...
r ·;, ·: ·':. •
. '!

' .....
.. ~.! .

.. .:.
..:._.· ::·.·.
• I, .•: .•
.·.:·.],'

•o
'····
.. '
•·
,· ,·,
( /.: .-·.:.
'.·· . -:.
'•
·, :

··:-·.. '
.· ....
I . '.

.t:'.'
..... • •- .;.1 "· •.·
-

...
.
•.· i,','
~ -~ . ',
. ' -~
..
_,.
.

•
'
-
r'
, ! ..

·.''
·:.

. ··:i:
·-'
.
..
•.
~
·, ,
' .~ .
..
·~ ~

h''
·,
,· ·. \ . '
" ..
. ,., ·-..·,.
·. ·: . >; .··.; ~ >·: ···- -~. -~
.. '~ . :. ;, ;_; ... ·. :;
. :_,-·.
~}
·,
. ~ . ·,.:
· ..• ' .: ... . :. ~ '-.- .
~-;,

k·
,.
. ':> ;·'·::·.·.: .. '' '. · . ·'
.. '. -'·.·\:

.. ·.. ..
. . ,..~
._.·_. ·\:
··.. .·... r : ,, .

.:~ .· : .
', '
'· .· ~· ' I . ,; .
·.'···.

...
.
,....
·'
·, ··,, ., .... •~ ,. ' .-; •• t ..
\'.f.·<_,_· • . . • • .

. ',·

,· i ..
...
·
..:-.
'''I'
.' ... ·
. ;·.·.·._.;:
I : •

'·', .··'
. '. .
·,

~--:
. -~. ·.

·,
. ·,
.. •. :·,: .. ~ '
' ,•, ·. :- .. ~ :.. ·: ;.·;;./) ·
 : ~\
.. J
. ,; <'j
.•
....... ~· '·

',., . .·' .."

.
'
. . ;t

-. ·~--~ .
'l; '"

. •.· !··.·· ...··.'

·:·;
:!; -
II THE. GUIDING CENTER ~10TION
I.

: .~. 'i'' :. ··~

In this section we will first give ari informal derivation of the
• ~I

'.I
,,
guiding center motiono Although it is informalg it makes good physical sense ·. · .;

and of course gives the correct resultso There are, howeyer, points· t.ffiere .·
.,
\,; ... :'!'-·
objections can be raised. These objections cannot be met ~thout the more ·:r.::. ~·

-;.
n
··., formal mathematical proofs, which will be. outlined in Section Ill: The '.>·

I' . ·:_ ... ·•

i •. •',;_'·- author believes that the derivation presented here is a good one for classroom : · : .
1'. '':
. ... ,.... · . ' .. ·· .
.. -.
•' , ..
,.'

•• •. i
usee It may leave the exacting reader wondering if something has been ... ·.
., •...

.. ., overlooked. ·, • .

., The derivation does not proceed by considering separately special .t.,

'•'
~ ' .:.. -. ;
,, .. . :. ~
·:
;

.·:
field.arrangements in which one guiding center drift or another appears .. :
~

alone, and then superposing the various drifts~ This latter method has been

freque~tly used (a,) J (.3) .P but is lengthy. Instead, all the drifts will be . ., · .;
dttlue·hvefjl.. . .. , ... ..
obtained at· once by starting with. a field of genern l geometry and doing a · ·.
A . TM. "'"#ocdl~ Sfwul•" -to l-lb,...,,/lfl..-+ ,~ e=;::.~ f:tln'l..:t 6u1-fwlltr1'11(;"'? J e-_U;'>tr::JJ,/r
small amaunt of vector algebraoA It will be .assumed that the particle has~
equati~n _t;tft:a:t,r,·~ :· : i}
. ~ ~
- ·.•
,-
non-relativistic energy. and the non-relativistic of motion used .·
,.. , .·: . · '· ... \{
as a starting J)oint o After understanding the geometric reason for. each . · .· ·•· · ··.·.· :: ·,
{' ,·
.
.. · ' .
'

.· '.· ·. . .
drift, it is easy to write the modified equations for
. a particle :tdth. . ·,
'·
l
.
·•
.~
·.
··. ,!, •

...
• 1,', ·: ·.; ·;_ ••

·
... relativistic energy, . ·at least for the- case wh~re· E :is sm8.11. ·
: A, . The Equation of Motion in. Dimensionless Form ., ....... 1

. .. ,; (\'_' .

··. As a preliminary; ·it is ~orthwhile I •

writing· in dimeneionlese
,~
form. · · · · ·; ·
• • : . ' . ' ' •' . • "

-·. .... . . .. · .·. ·, ;.t·.-:; ~ '

the non-relativistic equati.on of motion ,.4> ·,
• ~ • • • 'I • .• ' ' ' •
'..: .'.·:-. ., .::··,
•..:' .· ·; ...

[~, ~· g{r~ ·t_il·o ~ .x a(1~. ~) _+ -~(~~. ~). . .''

'.j,
. .. ..,
•. · '
• . . . ' . \,: <_-._:".· -~: ·. ') •
' ,·:·.
' : .· .. :.. ·
' .. ..r:. : ·.. . ' ·•· ... •. ~ ·: ··:···
:
.... .
l•.
. .., '
••·•••· \'

.
' ,
~:

.. ;. ... .,.
' ~
; 'i
.". ... ~ . .
."·. . . ' . '.
·'•..:.: •· ..... ·.· •• ·.• ~ ·:-.•·:. ,' ,· :.·:.:~... ··•• . .. ' . ~
. I
:' r:,.· .
.~
1···. '• .. ·. . :t, .
.

'•'. ' ..

..
·
.' ~ ' ..:, l ·•.
~ \ .....,· • ; • : •. •' .~.' ....... ';.'•'>?
 ·- ',:
... 4 ... ( .· ...
I:
. r·.· . .. . . I ... _,
. ' :'

' . Let
_:·.. : ,:) ..
·~·-·-~ :·
..,., . . L ' ·~ D _1/L,'.. ·. · ....
,•. • : i_ ', ,. • • .. ,·, -~

... .. . . !.:

.·.···;·. \'

...
• l ' ···.;..'

.· ····. ·.l -';

~-; . ..· .. • r I ~

" B~(t) is the magnetic field a.t a. ·. '·'·' .., .

o.'· ia the irlitial velocity, . .
........ :·
·,· '·

: .::·.:··· .."-·::;:~,
'·
•' .
·!.,·
·;,·
t;y·pica.l point at time t, and L is a characteristic dimension or diatanca ··' ,·'

.. ·i :/ .. ; ·.
. .. ·over which the fields ·ch~geo Then the equation of motion becomes

g·
I. 11,
..... •••• l

0\
.. _.; '··, ···.',·:··. ma'~ v 0 [ d 2:;;!1
Cl<.· · . · ,.,...) dCl ";"'' .,f. . . ,_,.,· ...., . .
··.. ·.·
eB L
-
d.T2
....,( ......
--b a.;~
. . . · .
.a. Ci7' . x (fJ. <~,n +e.; <~, n,
0 • . \t •• .
.' r , • ' •

t1 (Y..-c.o..-K . n !& . ·•
.
a- -:;tr/L ·. and 2E
d.u- . . Y
\

· , ' · with. th~ iilitial conditions that at :f."= 0 9 m

A , where
0
1
•• _ •• >'· . \.::·. :,-10 · is the ·initial pc;siti'on of the p~ticle an~ -;0 equals "'ct~o· Equation ~ · -'·~:
., . .. . · ia formally identical with ( g) by the. substitutions . : ,:.i~,;··.t: ....
. ._ ... . r: ··)

-
·,
... 'f.
.. .. . . .... , ~

i B,

i·. >;.' :·: .•.•. . ·:

I .·~-. :.: . . ··: ~ Thus· any r.olution of (a) by these substitutions~~~·

gives a solution of ( I)
l:... < ~ <' '.mc.v ... :
1· .-. ·· ··. ··Now · · Q is .the ratio of the radius of gyration to the characteristi:()~: .
I . . . . ... .: , . , . . . eB0L . ,,
I .:· ··'- <.... ·' ·.distance over wi:)ich fields change and .therefore is the quantity t:l1ich one ··.!•-\: ....

! .·. ., . ,·· .:
1 ·~ .. ·. ·' · :·.:.:... . :, .. expects must be made sooller in order that the adiabatic appro:r.imation :.\' ·,·:··.
j .: ~ ~· ~ .' , • ,

! : :-:·. -_:·:·_;: , · '.:become more

I \... ~

valid$ · But bOOzuae~ ie1,1,1ation (I ) . is formally identical with . :. . (·L.':

I. • ::.. , ( :!.), we can worll with . (I ) and just use the' dimensional quantity m/e u ~ .. :" '.•...· .
0
~ ~{ :{...:. as the smallness parameter. It obviously is ~impossib~~ ·to make the . m/e
~ .· . ' ,. ·. 0 r aay an ele ct:ton. smaller in a series of exp~rimen ts< in order to .make .it~·;:. ' . .:::

:j~.: .') behavior more adiabatic. But it reall;r ia'o~ :~~ . that. m>lst be made,,,:, ·.

:\;;..;;·:. \. smaller, and this ie possible by chang~ ·:'~6~. B0 or . t~· .The advantage
h~ r-·;_·_:;:~.·:.~: :. : _ .. · ·. o.r'ueing m/~. ~stead..,o.~ ·._.J'II~VO ·: is· tha~ one. d~e~ not have to worl< with···
·; .. ·.·. ,·.. ' .. ·,·.,:~··".,~~·; .·,. . . ~- . · :-~eBr)L:,..._' . ·. : .':·:·:_·,·_,·.,·-·· ' : .t: '. ·-.. . ' ·.·':. ', .:.,~. .-::.... ,: '.,
,f ·.··~-~~· ••::.•::... ~:.·
:\i.:.:,·_u,·:.·.• ..· ·, '· '··......:' • .· ~ .~- ,.: . ;~.
·..· .. · .·. ·~· ·... ~··. .. . ..
..
j· :

'- .r ..j_·,~ ··Qo,• I·~""·'·>.~ .·:~: ··,t, .... <.,. ~~ ,, .. , .

I!
! ·.:'...·
·.··. ''·':
1,

- 5'

·., . ........
.· .. ·,·:.l...;._.·>(.
...

dirneneionleas equations, and results~··are obtained dir~cti; in diniensionB.l

. ,'

forme· A ::iOre· complete discussion of the, scaling o.f ili~ ~quation_ ..or'motior{ '~ . -~. '

.• • ' I >' t ',:','' • : · , ; ' · ••

' .. haa peen given .in r.eference (.S) ... • .-.. ,., ·,..• ; '·_·._.1

l;.~

r~·- ·.:

r\~<·: ~ ...
1'-
B.. · Derivation 'or the Guid:ing Center Equation: of
.-
Motion
. '
'

·t·? .:_·
,.: {~0

-r .,... a+~~-.:-:;.·.:.
,,; ~

· 'l'o derive_ the equation of motion of the guiding center, let t''\. . •''
' ''/' ' ' ·: ::'
where is the instantaneous position of the. particle;· is the. position ' ·.'!I, .
;, '
·'

i . -· '
';.··:·.
'
' ' ,;;·
~~ . '
,,
of the guiding center, and e is a. vector from. the gu:iding center tot he
'.·;
. .: . : ~'

!':: '.,:·::/:·~:·~·:·.:·; • i . particle (Fig.. 1). The vector e can be given a p~ecise definition by-
'
. .::··.-.· ...

the equation' e .. :{mc/eB2) Bx (v..; ,M;B2),· where'' E and B are evaluated' ':. .'. . ' '· '• ~
.">·.: ·::
:: .

· This .:OO~ined ·m.th r ... R' + (. gives a precise definition or . .B:.- :<:: ,
.: -'

· ·-; . To lowest order· in e the fields can of course be ·evaluated at eithe~::: -; ·

...,.
'J;
'or
.. •
~J)- ·•the difference
.
being of order . e
2 in the equation· for: e .•.:
' . .' . : . ·:··· .·.· ···. •.

-.::~ :!·.'. .
..... : (:.
. .... . ...,. . ..
'

.. ;.. . ' ... ·~

. '·::·. ,·
'•. ,· -~ ····)

...
·:
·'·
;,··
"
...
.. , ..-'~
/.'
·' ,., -;,·· :~ ' : ..·.
'.'.

',\ ....... ..··

Fig<> l The charged particle·
4 . .
and· its. guiding centero
.
. , ..
; ....,. ~·
··,'
'•\'

'1 R · + f? ',into Eq.· (1!'~ siz\c~i.th~, radius of gyration is: :;:_ :::::
. ·.~ ~i~:
._,,_.
Now substitute l!fl
).
·' '·2·.:. . .::· .. ' ' . ' ·. ··;;,._
proportional. to ·. ~.··I . terma · ·: e '. '...can be neglected i compared t.~ --:"<:. .. '

,:~ .. .
''.··
·: (' •:',.o>'•'
·, .. . '
thoae 'in . e• ,', ''
.,·.:.
'•, .:
·~

.... ' ~ ' ..·

· ..
. I
,, .'
. . t.
•:•

.·.'.·.
'
: f'. \ .
,:·,, ..
· ..
·i,
., ... ;.

;·.· .. ·. /•
:~'•";"·' "/ ·,· ....
... t
 '··

6
.: ..
.... . . . . . . . .

· Tl'le res~lt of substituting· ~ "': ~ + ~ into Eqo ·(A-) and expanding

i
l;', .... ~: '; ..
(1..-.--- .. ··; .:···,:
l~.~; :~:.: , · .;. the .field a. ~.·~.Taylor. series about .
'~ ~ -· .. : . .. . . . ' . rf is ·•,' ·~·
'·-:· ,'• .J
.-.
',

.' -~·

~ + ~ ~ l ~ <~/m{E(il) + e. vECill+Cl/c >(ii+ ~,.f.kR> + e. J

17 B<llil + eC& ~, .. . •. "
.: ~
~) _,-,., .
, ...
l'

I:.1- .. . ..
., ~
.~ ... ".
·--';'

,. ·._-:;::-
•.· ~ ·:. " ... .-'.
·_ must be ~~tained in Eq._ (3); as will become appar~nt shortly, this ter.m
l . '
1.': ,. ;
r.. ... . • • ~,I . ' .' ' •.• '. ::.;

I- ...,..
r -. -... ·:·'
· ... is not ~r order· t: 2 but is of order 6 • Now define three orthogonal ·. ::_::·~>~-- ; >/
i ... '; ' I , '..,(_ ' . . • . , •: .···.'· '·' '., ,-'_-:.

~ ; ... '. ...

'

.. · _' ,:,(: unit vector~:·. · let ~l . equal· B/B, let ~2 be· a. -unit vector direct~ <)/ .: (, ·-{
towards the center of ormhlre of the line of force, and let ~ 'l?e' ' :. <:-<'-:,-.:~>}
._,_ .. A ,: ._. .. - . , . . ·.- ,,_,··-·
:·r::· :~ .e1 ,... .
_1( . a2 , a uni~ vect.or along the binormaL
.
In order to correspond to the · ·
.~ .. ' •:
:, ·, ... :.:
. . . _: ;_ -~ ( .
.. ·>.:: ~- .., '• .~,~
picture of the particle moving about ·a circle of radius e' let '
.. ~
,(

' ; :. '
N. <

./
',

e: e<~2 sin 9 +OJ. cos e), "here e = f~dt, wbe~g the.ll)lro frequen~( : .
eB(R)/mc~·. Then .#"" r.~e (a; cos e - ~-sin e) + (e ~2 )'"~in e + Ce8))" cos a~ .;__;,;. ,·_ '+-;: _:.:;
·The first .term contains. wp and is of zero order ,in e;. since. w-1/6 ·and . :;
• The second. and third terms coniain e or e
.·
! .. ·..-' ·.:· _· ·.
~;·;':_.
.. The reason for retaining ex (f d.? )B. in Eqo
.. ~ :...a.o . .
(3) is now formally apparent, . : .. >:·'::· ·.-:;.
--
: .
-; . '-·....
. ......_.::;.<·~-- .. -~.-·;· . ·:·.

t'
tl':: ._·_·_-_: .. -.
,.. since it is' of order e ' whereas a.' tc-!rm such as ( e ·'>7 )
2 E .in th~_ Taylor·.. ,.:, ( -. ::·.-; '
:- -;. . . -. :2. . '.. . . ' . '' ., . ,.f ,._ ,..

f'- expansion is of order 6 · .• · A second_ differentiation gives .~ · :_;'·; /'\:.::<' ...

i .•.. ·. ~ a- [w2 e (~{•in e + 1i'3 cos efj+.; e~2coS 9- -e;~in G_] < ······ ·.<',, ,:
!:.:_.:.>:_::·_:, .· +. 2 . wge~2)_'" cos ~- (~~)·':sin f)~,+ [~e2)"··. ~in e +· (e~)~· cos_.~ ' .. · ,.•'•. ':, . .' . ;,

r; .· .· :.: . the four terms being of order: '1/"' '· l., .1, and e; respectively. These ' ~ ' <.. . '" . : . ... .

ti /<:--
~-~; ; -_< .
. expr_esaiona for
', . . ' '
·e ' .
e, .·' •and" .~ \'are. now, , ..substi'tuted
• ' <
illto' Eq;. (3) and tha
• ; 7r.
• •• ' .,
-.. ~· ~ . ..
2
[·~ . • ., resulting equation time~avera~ed• QV~ ·," gyration per~od/-~~~ia!dn~ ~O <·,::)d~: ••· '.
I
I . ; , · :. a.nc(considering .coetficients,;:·auch_ as~;- (e~)o 11 to be constants-~ ~-:Then:: ; . _. ~
j· ' ~ .. '., .• ~' ' :. . . ' ' • -: .. '· . ' ' . ·.' .' -. . . . .• ' ' ,. ·. q ', .• •. . . . -.-: .•

\ · ·· .- ·. <.e >~-. = <b l.e>.m _0 ci 1

.where the. angular ·.brackets. den~te ·the aver~ge_e ·:,-The ; ' ... ._· ·.- ..~ -:·, ·

... ·· · ·· · :H; ~-i ;:z·:~· .· ...~ . ,.,i'·.• :.;. rc.i; ~Jtl; ;. J?: ~.·;-~.,·. r:;< · ···· <..;,. ,.;;"P·i:;;~~::;~::,: L) : ·•·...
 ' '.... '•
ll'
.: i

... 7 .... ·.
·/.
. l
~
l
;
·. . .... , :
'!(

:' -~ .
.since '
~' '

. ,~ ,.,.'
... ..
• ••,_
,.j
'"If
.

.~I.· ...
·.: ', ' <o" ' 'I • ' ~ I ''l- ·1
• : . . .-. L~

....
' ... ··. ~ ·· .. ·-~--~_;,.~~ :
.· •....• ·' .•'' =:-..·
...
......
The cosfficien~ . e2 w/2 is an appr_o~ate .invariant of the motion and is
.;

. i.

Mc/e» 'Where M .is the •

well-kn~wn magnetic mommt.
- II , • • o

invariant.
. .. of the particle.
.
motion . has been. demonstrateC;i
. .
in reference ( 6 ) . · ·
'

. ·.'
and in numerous. othe~ places~ The adiabatic invariants wfll be treated ·
in Section Jr{ e The invariance
. of
.
M. .is not used. in deriving
. . the guiding
. ' '

•·.
t'·.
~·. .' .

Now
e

~l: ~..~(~ o'\?)B:~ ;i ·~.(s; ~ t7)(~1B)·~.-(Bf2)(~;/..$.l7)'8

'
....
,. . .'
.
1~: ..+~J o~B a~ .,
o.i7B,

since
.
esi
2.
b
,, ..
1 •. :Therefore F.q •. (b) becomes . . ,·,
.. ''

·..
.,.
·.
·:·. ; !>. _.:' ,•'
..

'< ·.·
... ...
_ '.

·' ......

. ..

' ~3 ;' <~2 .·rf>B ~ _-:,l· ~2,.'· <~2 ·.lnSJ + ~2 .<~ ·17 >s: .' '-': . · · <f> .· .. : ~/ij j

' I ' \ '' '· • . •' l ' ' ' " ' ' • • :• ·•~ : ·.: '~~~ ~ ...

. 'l'h0 .f&et ·tl1.~t ·'. v~ 1 m 0 '.must n~w- b~ ~aede .:>~·e .operato~·v··can~ ba<i.~e..~rea~OO. : . :: >. ·.:::~'
1.: ~· / .. : '.~. '·: .:; ·. ':·::..:.{, ~; ;;;~::,;·- :·.•:.}(~:.~, ':·. ·~· :~;~~-; ::_ ·.:'; ·. :i ));~ . ::~:;. :):~ :: ;•.i;~:.·~ : ,~:; . :~ :: ~. •. (\:·:~.:·;),: }:·::.::.:. ·.~· :· .~. ' .::: ·. . :. ;. :: ·~; : :>.· ':. ~.; :::-:: :~;-, .,.,.,· .. ;,; :..:.:~: :. : ::·· . . '
 8

as

;;,_' .... : .··

i''

I
I!
i . .t

I,· · •. \ '

,·.:····
;:. '~ . ~ ..
r ~-:. ·. ·:
' ,.
! -;~· 't . ' • . {1/)
''· . . ' ·.: ·::)
~ .':~· . -...:~-~·:·.< ·~~·<:- ..
The time average of Eq. (..3) then is
:.··:-::\~.-

:. ·... ~
.
....... :
,• .... •o . • ~ -...~.--~-~:1

g(R) ~ (e/m).[E(il) (~/c). :x; B(R)J: ~ ---

11 .. ::·:.
. .' .. '; . ~ . R :a + (H/m) VB(R) + CY(t:: ), (/a)
'I
,. :. .
: ::;_-~}-
• . ' ,. . . '• ·i

: ·..<'.:· ·... · with an initial. v~looity R(O) equa"i_to.[~e1 ·v. ~

Jt
.... (ci x.B/B 2.)] tmO + e-_( & .>"&··_:·,.. : ·,·:. _.·_;)
I.:.
I
'. . :. . . . ....... "'
t.· ... ,·
.......
Equ~tion (J:t,) is the basic differential equation for the guiding center
I ~ .. ' • • .·_
. "-:..
~ . . ~- . ....-:. ': .
'
. . (tJ) . . . ~ ·. ' . . .. ' .
t" : ~~ ' :•. ·• • .
motion~ . It is the same ·as the equation of motion of a particle in a
• ' 1 • • • t I • ' ~ "' • ' ' . • . •
... ..... ,·· ·, ~

t ,!· .,·.·
: magnet.ic i'iel4. B . and an equivalen~ electric. field 1·-. (M/e)V B, ar.d ' '. . .
:. . ·. ~
·:.
.. ·.· . -~.
~ . . . I
:·~. · ... : . -.. - :·,.· . ~ ' i . -;.
.. , ., . :-.:-r
~. :. therefore. for. numerical integration:.is,mol"e .complica.ted.than was the · ... .. '
. •.
r-. ·,· ,..• ,'·;
::'t·:·:' ..
i .·.• ·, 0 . . ·, J• :. ·.:·.
~- J< ....:. origi.rial equatio,n ·of motion ~ ~ ) " If a.
.·
numerical
. .solution
.
" . of :
(I :J..) were :-_ ~.-:. ·;:_:._ ./
: .. ;:. .·•
:~ ~- ·.. . ' ..
! ~- :. .
t' ~. ~ performed, it would ~e fo~nd that the guiding ce~ter ']. t
c.
rav~ls in roughly:;: .· .. :~".. ·_·, ~; :·.. _.}
I .. 0 0 0 ' 0 OM 0 0 ' •• 0 0 ", '':· 0 ':.· 0 0 '

I • .~ ~ l . (.
<· •.' ¥ • ·,- : •• • : •__

. .. . a helix about the ·field line. just aa the'par'ticla doeso However, it"can i ..
.
:'

~ ' ... . ~ . . ·. . . "' . . ' ~ ,.· ~ : ,: • ;; I ·:: :::,

j.• :·· .. ... ·.. ~

~ ..
be shown that the radius of thi~ helix is one order of e smaller than·the ,·:.
·; :_,> . • .
. ·i .. ,.:_;
~adius of gyi-ation of the particle,· as w~ul.d · be ·expected for the guiding . :" ;- '.,

·:' .·
cent~r. · This small amplitude :oscillation of 'the guiding center' is ..to be
. . .· . . .. .
:. ~ .

igr~red 1 ~inca it
. ~
i~. of_ord~r.
.
~ ~ ' '
and
'
or . .
no significance 1~ a. first~rder.
. . .
' '

•• •
'

1', ••·•
' ·, '~I

. . .

l<)lrthermora in the. p:recedfu.g analysis.· oth~r terms

. ·'.
·, ..
theor;ro _of this order: · .·.
Equati~n. (;~) ~ll n~~;b_s aolv.ed by 1terat1;ont~o ·.:::: . ..... :
. -~ ~ '

: hav® been _negfected •. .. "\'

... · ' l
t • ' -. '·' "' . \~~~.-- ~ . ·' .~ .:.,.
~
. . : . : ·:
'
I

I
·,,. ·' '' l

.
.,·

··~· :y-~.
' ·.' _·-

·.>· :~ .~ .. ; .
1
.

' .. ' ';

,• , ,·,: ··

' - .·, ' , ~ . ,t'

'.... : '.'
i ,j \ • ''

1
. ~ .<· ~ ' ..• ..
.•;.
' . ·.' \ ·-.-· ;" •': • • • • d
··.
·, ·-."~ :·;~-:i~: £.>;·: : ;:'~.; :::::; ~~LLx_::. .~:....:;·;~· -;~: :~ -: ;,{,-~LL -~ -~,i:-; . ,P .~ ~.-~ ': •, . '.
~

' '
'
 . . ·•· ·~

:. :'
- 9 -
..
·._{:

.,
'•
' " • ·.. ·>
..... . 9btain the ~ equations for the guiding center motion parallel and
.:· .• ..
perpen~i~lar to Bo .
~.

' .. " '·:,

' . ~
• ' .' t-' ' ' I . ,· -r .
·... 'r,· r

.·. ~-

··,··:·:.
c. ~...

.
•• 'f ' •

The Drift Velocity and the Longitudinal Equation of Motion

'
. ·.: ,··'
·,
.. ·..';',
~ ' . ' .:
The differential equation for the guidin~ center ·motion can be . : ·.· ~ .

.. . r
• J
' ' . :,. :~.
··.
separ(lted ·into components _pa:z:oallel and. perpendicular. to. B.,; Crossing Eq._ (J;J.J ;. -!'• . ·'.;{;
' . ·'·y

. . •. . . ! . . . . '<,': ~:-•.

.·
I •' ,',

... :.
·on the righi;, }'lith ~l~~ · gives the perpen~icular component of the vector ,'·J.
. ·.
'
! ~-
,, ... :•.
. . . .·
e~a.tion e.a · t,•· , .
··..:··..:·~·
~ •J ..•
··.~
:·.
·"·· ·.. · . , .
.. . ... : . ·.
-;·
'·,_.

.. ·.· .
, ' : ...
:. ;·

.. "~ere 'trJ.; is the eomponent . of it perpendicular t9 ~1 (10. It is called ..' ·. :.... ··:' I

... ..·_'·,
_·, the drift veloci~y." The first term is the usual ,;Ex B" drirt. · The second·· ·::' ·.:·, :: ::;>:'.!;
. . .. ·.· ,-:;

.. · ·te1"1ll _is the ngradient B" drift,. and the third. is the "acceleration driftts·.. ·· . ·.. > :: ... ;_,,·~·<~
:, . ~ ,· ,:.-. ~ .... : ._:. ;')
: :'·.·,
1\ ..... . , , • ~ ~' ,1

By dotting Eq. (I~) . with one ·obtains the scalar ~x:-alle~ component . ,
' • • , I

e1 (R) . '-;-
... ··
:.-,
..
. .. . . . '••
~ . .. .::. ·. ..~ :: ·....... t,
·as
, ..... : ...
··.·.
,,
.. :. . ,' ·, .. ~:· ., ';'.

. ·. ·; •'

~-G
,.... ~-
m '-"i
g . • e""1 + .-E (I¥) : .·.:, :. ;·...~ .·
.1 6 '• ., ,· •.' . .... ;:
· . '1.•·' ., • . . ·.·:·.-.~..:: .• ·. :'· '; .....
... .::I

..
....• · · : ; •.'':· ·.l

In Eqo (/3) the guiding center accelsr{ltion fl is ne~ded to calculate the

·r· . , .•
. ... ··,(
drift. :velocity; bUt becaus~ the term in which it occurs .eJ.ready contains ~
".;; '. : :' -~~-·-.. '•,' ... I

. . ·. ~·.. ~ .:1,
~ zer~
........as coefficient, R is needed only to
'· ·'
order in _6 o
•.
.·. :'• ~

·.,- ..
R; is not .of negative ordal,"; such as . 1/ c .. ·. If i~ were of riegatiye order~ , . ·: :, \ ~;r
.. ·. >~·
\
,. ·. ·. .. .~· ~·~-:

the fields would: change by a large amount in.. a gyration .p9r'iod' ~h~ 6 ia
•••• ' J
.. ·.,.
·..
;·'
.' -; ~ ... ~'
'. ,, :• • I I i

; ~. ~ .:

. srnall 11 and the guiding center picture would not be valid.• · ••·: L . . ., ·. . ':.'d'f:t
:: :·;:' .. '
, The .acceleration. ~~~ ~/dt. = (d/dt}{~.J.. .:t" . e}_·i; ·:o ·~1 ),· a.nd: ~ dj{J. /dt:i.< ··.
i:)
can bs obtained to zero. order .in(: from Eqo (43 ) a~·.,·;~ o/dt co (dfdt)(cEx~~/B)':~ ~~ (S; )o ·/.~
. ·,
.. ·.
~ : .~ .. ,( I '

~- ; Only the fi~at term in the drirt:1~e~t~!~Jed,.aince. th~··iM.rd term ia ~-~ ~d · ~ ·. ,-·· .·.~ . ~ •.. •:j

:_'::.:': :~·,· ~: : : .:': ·'. · . the ·second term contains· lVfJ .·UJ m( e,;, )2/2.eBAI' 6 ~ .· If the perpendicula~ · ·; ·: ...;
' I ~ • > ; • ' ' • •, ' ••t 4 ' • • ; > • • •.. ·, ' ' • •• • I .
•< ·~ 3, ·~ R'·., ''.1"~ r '. '' • ·,,, ••'.' ,·,',•,' :.," ~.;.;,_t·~·· '' '''

. !_ ,!.-~<-:• ; o"":( .. ·; .:· ,..;\ ; ·.. :_.·-.~;;('~.--~-~~~·i.· •-~~.:: . •-•: .. . ·:':.J .. ; ··'

' ;.· '. ,:-.: <~:.:t£~,~,::-:,~;~:;}i;.:}_,;j_:;:.\\. '·.~:l;_·, '.''i': ... ~->· . .,;;:; ,., ; .... ·.· ''. . ·' ;~.,:·!);/' '
~

.
..
 - 10 ·-

.. ','

electric field_ har·pena to be of ord~~ t: instead of zero order, the

reten~i~n o£ cEx~1/B E tlE. would be.~u.nneceasary in the ·ca.l~tion ot

1l"o The acceleration ~ is · · .·. . . . . . , . '>
• 'f.
.• '•,r :

~·' .. dv d~l d~
.R ....._&-( G ) !!:~ __f!_
v, J!;

-
-. ' 1 dt ..... dt . + dt'"
':. (

·~~ + r~
. . ., . '

. + ""'t
u,) - •·r}e ~·"+
t.',
. •,, E · 1 . u t ..

,' ,•

. ·~. ..,.•,';• ..
~ . I

.. . ~ .·. . l
'.~:.··;•::' • -- ' -~· . ·_~J:'';w~ ,'' • _, \~\,

,, ;:~:· · .where vlt·m~ans it" ~1~) •. Other. "parallel". velocities.can be defined, ,. :· '-.~·~/1:(

.
.;~
: ;?; ·..• ,.' ..
, .
::h;:.t~t~:l~)th:rt::n~:):.:::.::::on~~~hea~:: ::an;:· ::tr:p:t:~ (ii) • .•. •• ·.•.•• ,;, :· ;~ ~
'' -~. ~ I

' . . ,:::·::t::n:~::: ::::::.o:::•t::.:::::i:h:i::::·::r:.:: :~.:t:~t:: i~c..:_:\~~

:presence of a zero-order electric field. It should be stated tha~. the . ~' ·· ::.:..~·:·:_: <iJ1
': :- . .·.
preae~ce_of a. ttzero-or~er electric_fieJ_d~' means that in a se:d!3S of·experimenta'.":. ·.·.",-';·~·

>d.th cOnstant . 'i7 , :;:'0 .and

0
·U:'which .m/e .·is successively reduced, the electric .; ' ': :/~l
.. ·.· . ·!. i ·.1 'I;

-·. field is held constant. The electric field is of orde~ ~ .if it is reduced ',; : .·, ·. . . ·..:. .. '·;. : .-. :: .'_,:';~.: ;
.. . ·. ::.: ._:· ·. '

.
:;::o:i::wt:hem~:. :·::::~t::~l::l~:; i::: ::·:.:;·:~: ::~::.. ., \~ ~ ,,~ ·~i!
·:· .. •. ~ ; . ;

of experiments •. ·If the time acalei~.heldconstant, 1;.heJn o/'dt ,·is o.f·zero<~: :;;'~,
.order ~d.,- . ()~/ (} t. contribut·e~ a. f~rat~~r,der. drift.:· . ·I~ ~he time_ acaie /. :_<;. -·~. .. . . :-.-.'·>· ~.
. ' ~ ·. . .. . . . . . . . ' : :: :-~~~:

.,. ia .increased..in proportion to .1/~ .· 1

:then._,.~/~ t : ;s of~ order& ·~d .·· ·\ . , · · ·.· , >~·~

..-·_ ....:. .~~1/,() t ... .ia .not ne~ded~ ( ~.>. _·ot_co~~~~:(.~~~·~ausa.•:·.mi~'; ~a fj.xed·9_·-~~:_:ia __ . , ·.· ."
..•- ••..
lt . • '~ ~ .... ~--
. .
t·< ·.·" ';•
. .. . . . . --·:·:
..
:~ ·.• ·•... ·; ....~_:·. · .• :. ··~ ~-. -~-_-.=·. ·•.· .. ~:. ···:
..
.. . I . -~
·--~.
·::.
.·. <.· .. ;·.·,-.... \ ' ..···. •• ···. -~.::;·~.' . . '.;_:~-~-~ ... -
. -' . ·: .; ..' ·•··. :· ~·-; ' ' '
~. ~ /~ ~
.
1
<• ',' 't ' .. '.'. ··' ••
·~:···· . ~:.¥.·/-·:~ . :·~··<:· ··~··1;.
• • ' • • ' " · . • • :; ':

i"~l, "\ ,._, •·•·.

··. ·.
.. · '· · · · , ·~ '. ··. · •: I ·. • ·. \ ·..., .. · ·· . ~. i ,,. .,. ' •
'f; :
. '.;·:. ,_:;!· .•:.:,\::.):·i·_~:.:~;~~./t~,:;;;:;: . ' .. .~. ;: ..;; .·. _':/;.:;:· _;
 i
l
''
- 11 -
• !•
' ~ .~
:, '.
.·i .
really B0 , v 0 , or L that must be varied· in the series. or. ex~rime:ntss ·. . ) :··~ .-:. · i
.' '
. . . .. . . . ··.:·-· .;.-~_·... ':;: ..~ ·...
Appropriate modifications. of the actual fields. and time sc~le ~st b~ made , ', ··.· 1 .. : : ( · · • ."·:~
.,,

•il
•'

'
.' to keep_ th~ dimensionless. fiel~s. CB , ?.., .a.ndb unchang~ at a given ~ B.l'ld •.. -· .. ·.: ~ '>/;. ;:'·.:i
i ~ . ~
. '
:f.·. Details are given :S,n Ref. ( S'). .i'·:....
~
With expression (lb).for R,Eq.(/3) for the. drift becomes

. "fl.L. = } x{~- .E+

·.

..
~
~
.
.··~· ... ...
:~ \
,'
..1
··: •', '•.·l,'
.
.., ...,,• ,...
,

I .'·· "··.- •• .: .....

·on -~~1
1
':· ,t ·:-.·

vuij + ~-- ~2)

•,r,' .. '\'.1
. .:
~

. . where. ~
..., ...... . ·. ~
I',
~ .'
I',
'. .,'•
• .->\ cExc;_/B •.. :.(.
.. •.. i
. ... _,

.
~~:
, ..
:
. . '~- . -·: ·.:
.,

'; '
..., The longitudinal Eqo (/£/) showa E11 _... E ~l . mu:t o be of. order G ir.·..::~···. :;:· .·);1
··~
"';·:-::·... :·.: .· R.is to be of non-negative order~ Thus. in contrast.· to. E-'- ».which maY' be.· .. ··.· .. ··-( · .·:··-~
,·,::.
..

~/.: ;·.·.·;._, :):

',, ...
.. · If this were riqt s.o, .the parallel> ....
. of zero order, f.
~ ~

must be of ordez: 6 •
11 ' . ' .~·-:· : .. ~ . ·_:·_,-·. .;}'_·~-'· . . . :·::·_~>;
acceleration would be -1/~ ..• -~·--·_i'
. :'
.,
. -. '7'; ..
.. . •
I
;:,l~ t' '· <:·_ . ·.. )
:·· ·. .
..
~

:: ;
•• •• >;,
.: ·,:. ·:· '. :_.~ '/
1: .•• ;·;:_ . ,. Equation (U{) can be put in a form more useful for obtaining an·:·:: ':·:.
. ,· . .";.
''!..'• . ·'·· ~
: '·'··
·-·-- .. '.·:
·-:- .' .. :-,.:'. • • • • ' • • .: 1

'.~ :. . ~--,.;.# _; ___ _ ' ·--~.: ··- .:··.:·

.. ' ' '.::',~
•:.··
'· ..
'.''. :Y . · ; -~.-.:.":
'.''t

.-;·_..-
••· .j'
.-:·. ' ..
;.;.
I ·' .....
~ .... . :·::•
. ...<' lfi;', : . <. ::·. .·.-~.~'.~-~
:.··.....
"'•·,
. .'· ; .. ' . -~. :. . .. ·, ·-;\

·t"' ', '' ~ ''

;_·:·; .
'.\·'
and noting .that. · ..· · · '.
,'.... . .
:· , ... . .. .· .. _;,,-
'•' ''
... '' .'·;·_
·-.·; .' •, .
;·-·-
,t .·'

·. ·.. ~ ..
.... '...
:
. . ·: . ···, ..',. ; ~
. ·.;·····
' : ' .' . . : . . . "' ,.
·.
(::·~. .
,•
.:
(

. ',·• ' . :·. · ..

J
.• -·, •! •.
'

0~~/.)
·,. !''

:..-:. ·:
~' t.); ~~(~vu' ~ ~) ·~val .~<~> .. ' ,;
. }•

! • -: •. ·.: ·.•
·~4 1 .... ,' ~. .... ·':
!
..•
0
.-; +
·.. ~ . . . . . '• . . . ~. :
:
..·."""""~··~ ..
• : ' ." I ~ •

I·
.. : ... } . .
~ ',~
':· .. •'
.' ... '
I' ,. .
'.I. ., •.

.. ;
,• I ..

. ..... ....
I < ,)• • ', .•
·,· .... ,·

. .
'• ..r·' •.) ,·: .

:
r··· .. ··. ,··.·
',.

.
•"1',•
.., .. :, ' . ..
~

'
~

.. ;· .~ .. ..• · ..
··,.
.·.·
:; ', ' .~ ' .. '·
' .. ' = .. ~' 1 :... ..
' ..
.·• : ~ ·... ~ . .:. ( :' :~ .. '
I.
'i- ' ~-
.)

- 12 - . '
~· .
.- ~ '.
,I ; ·, ;.,

In Eq'a. (lS) and (lt:tj we consider only, the col}tribution .or the .zero order, ·.
-~motion to d/dt • . This is all that is requ±red~.sinca _'}r,. ~. haa_~r~~ ~--·-::-'· _x'-,:.·-

coefficient in Eqe .(J!I). The ·longitudinal Eq• ( I'!) then becomes 1

_.
':-· .. ' •. ··:
... ·.. r---~-----:---~--------------___,..~---t~·
'.: ,,

·-- -· ID.
dv11 .
"" ln. g· + E . M· ~ +. !!l -'t ,. d~l. ..... ..o(. --2)
e
.-.lJ,.,

dt e 11 ll - e ua e· ~· dt rr""' I; .f.
...,,
';' ' -:.,
·' •'

. ·J'

• ~ < -·~~

Equations (l'j) and Qd are equivalent to the original: differential Eq. (J;;J,. · . ·- ·-::.;~;;

Let us n_ow.introduca_, a tr.ue. curvilinear coordinate system, ( ot., (J, sf .' :. : .· ..... -~ ~.;
.......
-'·
. ! . • ~ ::~ ~ ·:
. .:.:;
such that t:J..:fr, t) _ and fJ (~ t) are ~wo paramete~~ specifying_ a ~in~ of

.· force and therefore const~t 6n it; s is distance along the line ·aa .
. '.: ;
. . ~. ' '. . .- .

.. ·t:... ·.-- pravibtfsly. For a divergence-free. field such as -B,, ·"'-and . t> can be ,-.
•,;
\'•' I ,\• ;·',···' ·,'

•; -~-' :·.:;· .· · ,_ ' so that the vecto~ potentiB.l 7. is o1. f7 p and To prove that, . _ · · ·.·
I;-~-"; ···_t , ·.;·
~ '\" ' ... ···. · this is PQsaible_,. let ··)·.··,.- .
. '· .

~
are constant on a line of force, but such that A•. is not .· c 'i1 c ..
1 2
The :,' '
>;
,· ..\. ~~
are at. tirnf!l tthe inteJrsectiona. of the two families of surfaces
I • • • . ·• J•O ••••• ,.

!.. :-::: ·.• lines of for.9e.

·-x.
' .
''··-···
.··,,.., Now·
!o • '--~· '• ·, ·:
I ~·'

. ..~· ~ .. ..
· ·: ..

i. :, :-:.:: ..•

.; .
.·.. --;·. . . :·
....
B · is constant .along a.'line -of.for_ce, since
:,

This says that ··::.

.·.·
. ' ~ ·l'iJCl X (7C2' .. ~-' ·;.,

I . ·. .:
,•.
, ....
-ir.

i .··_...·,· ••.
_·;··' .,•
.
Jvc1 · x ·V o2f ~ .·..ami . '1.1,. ~..17.""
. .
.£.. .
_08 •
Th .
us a genera
r c..
.J..' c2 ....
'· ' : ; '
' -~~. • r, B.
..
~- o:~. ~ ~-: ...... coordinate. system has the property that ._· ~ l!7 .cl x 17c / is conatru:t along
·:.,,''. 2
a given .line_, 'b].!t varies from. line· to line.. Starting \dtq a..: .~ .. c 2 !3yat~Sm
' • ..1, <·. . . . .. .. ' .:. ~. •.. • ·; .-~. . ,:. :..... \ . . -~ .
.• . ,. . : . ..r . . I ' • ?, • ' . ,'•, • • I , •, • '_. . • • ' ..
~
···;. = .,_ • •·••• . •• r~. .·. . . . . ··.· .
. .
.• .·'.
·' "· ·., ·. .··. ··: . .• . ·: \ • ,_._. 1'\ "f (. . ·. ~ •. .'. ~-
I!· : ~·:· :,"" -: • • •: . J ' .. ~ t'
•. < • ~ '..... •: •. " ,_' ~. • • . •• ••
' ' ;. • r ' -~ I

·"·
 . ,·.··'
.· :·'
.--·-.-!
... 13 -. .I . >··A
~ .
'';' ·.- ·.··. .' ' ;., ~

'.
~- ; . .' :· . .. . for naming the lines of force, an
.
system can be obtained aa follows: ·_ ·
~ . '
•:·
. ··.
: ... - .. ·' .. ·,.
. · '.-o;_· -' •• ·' :. ~.~

i·-~~-· ,,!.:':,.,_ ·. since B/{ iJc1 x "c2 j is i~dependent of

2
:~.;-
a,, let it equal$; (cl'. c )c. · >:__ ::/:· . :-_,. -~::
...
'!'hen-~ ·.J ~c 1 -x 't)'c2 •. -~ wher~ --.;.; .-·~:- .. :~
~ :
,.. equals Let (3 c2 . and .d- ... oL.(cp.c2 ),

; .~
. the fun,ctional_rorm is to be determincid, Because J1 is to:-·e~al 1;1.,( x ~,1, \:<', , , i~
I ~~ •,' ·, , . . .'

it must be true that . 5 rt~ _.x ·t1 c 2 .""' !7oLx. l7f1 , or ·,that .'.;-~; . /.-_~j
'! •,<
'. '~ .

•• •. ''">
~· :' . ' .
. ' -:--:: :. -~-~~
~- : . ·.. {!' \7 C.. X .'(7 C
:::> .1. . 2
"'" , or ·".- ·',1·.·.
._ ::~- :: .-_: ·· · · ;:.;_
..•. . .. ':•. ·' ' .
~. .' : ·• :
.; .. .
;.
'~-- . -~: . . : ... ·~---~-;.~
c )~c ~. It is therefore possible. to g9 froJJ. · · · · ·
~ . .. ; :. ·, . ·finally that
2 1 cL
.. . .' .
..
' ~ ' ·.

'' . ~~ ~ ~ . . . ·.' .· :: a gen~ral . cl' c · ~ystem to an r)-, p syateno. Because oJ.. is th~ integraJ.
. -.
;

2
. .. ··
! '.' ,.,__
:' .• " \
. of g
with respect .to cl' it is clear that the oL 1 (J sy~tem for a given .
• ~. . ·.
~- '·,.

f~ .. .

,. ;:· ·:·i_<·:·:~·~_:.. .. field .is not Un~que; -~ ~rb_itrary fun.ction of c2 . can b_e added. to of:- (cl" c2)?:·•. /:'·_:_:·::.· ..-::;
·. ,. .. " l !7~ X D'i I . . ... : , .· . -· ,,
·--_ -~-..:·::·/ •.' _ . In an .. rl- 1 (I system B P'~ i~ constant. not .only on a line of for~e . ·. :<- .· :· ".'_ .. ·.
.".:.-~: . \:·-: __ ·but everywhere,. being equal tounity •. This fact_often simplifies :the al~ebra,_ :_•.(,'.' ..':.;-·_.·<;
. ' .. '· . . . "' ·. .: ~ . ' . . . ~- :'/.~: . . : -~ ' ',

-~ .· ·,
~,.· .:
_·. ·- . especially in _connection with the .adiabatic invariants. .
~ •·,: ' • t '• ' -
i'·.:· -.- · · ·' ·
• ~ I : , , f, ·; ,;::·,~·;·. • .
l • • ' ' '). • : ~ •

· ,- · .· .· .. · ., .. '·' , In equation (~o) the parallel electric field is giv~ by

.'·,
:, :·~·-.; .·. :r
~ :. ·. .·.. .·..:
. ·.
,·.i· ;_ ·,......·. ; . ·._.. _, .... ...
·,·'· ·'.
. ., . 1' "Jf: . ;; . .
·,
.··.·.. : \ ·'.
::·:::
E .t:;' ~1- •. ( ..,; - 5"'""t ~
,, . c. v tf ¢) .a .., · ; -
"'8 .
( f +. ¢) p . C~ o· ·:_-.·_·-.·.: .- .... ·_:-; :.!
::'' ..
.' · .. · ...· '· ., ·:·:
-.
.·.' ..•, I
•. '

. '' ~

·...... .
"'
.\. ·, ' .. ·
where ¢(r, t) ~ is the. sqa.lar .pot~ntial !o~, . . E~ 8J1d
-"
Y'. .is QL0 . oPJ!.
()t .•
·.: .' ·.;: ,_:: }!
'. L '· ' '··~ • .:. >~ ";·~·:
.·"--' ·,.
·. ·.
Multiplication,of ,Eq. ~O) _by and aubsti~~ti~n: of:·-'(~'1) for E 11
'•
'.·;;·.
:-. . ; .~ . :.; :' . v .- .

11
I ,~ •

,. ":"
·gives (the i f'ielq·<,.n.n
.. . .
be omitted from. here on,-bu.t:,co~d-·be retained,
. .. . . ' ', '. ''""" .
;.:
. ' .. { . :.~;
. if desired)· ~ ', :
.· ..·.. ~.. ~- ...
······
:": : .. ·
.. -~

:' ··~

...
., '
.. :
..... ~ ... \' .. '
·... -~ \ .· .. o ~
Furth_ennore
'I I •

.:,_ .· ·. \ ..· ·,· .. .. ·. ~

. .. . . ::. ~·:·· .. ,···

· ..
'. ' \•
:

L .( ~) L ( lli'l) +v ;) ( MB)_, ~ ::.· '"( MB.) ( 2) '. -· ·.•

·••.
,., t;
l

......... dt. e
m
ut s . I! "§& ;- ;~-~--~:v ·;;-~· +:&_6 ._ .. : : ..···_· o :' -::I •

• •' • ~ : ' ." 1 '. "; • ·.1 0.' • '' • ' ' '. • • • I .
.· ..··.::
.. ·... ·.· · ..
.
' ,· \
'•.
:. ~ : ', ... .. ., • • • • 1 .' '·1:···.: . ' : -~
. . ' . ~.
. ' ...
Equa.ti'on (aa) can then be :·~:ttte~ a.a (··:'' :- .
.. '.
•.1','.

. :... .·
· . ·~~·.· ._':· . .,.:·· ':(':: ....·:··.· .... ' _.:·.: . . :~·>! -;_·,_;.-;·· ···. "·' ....,_._.:: ...:_•.· '.. • ... . '. j
·,·
' • ;, .~ ••• • ··:·
.
t ' ,··) . .. •
~-
'•
...
I . ·' ·~ ·: ..
 14 -
''. <

v 2 + !1I}. ) ., -v £... ( 1f+ ¢) + £.... ( HB) .+ .. _;;.

. II . e · II os . ut e . • '~ •' I·. ,

· ...·'"_:- ,•
·,.

.. !.1!! ) + +.O"(c ) •
. 2
;:·~
·-;..
\
··:

'· e.

The rate. or c:hange of kinetic energy or the particle, averaged .over ·

·•1,.

··:·
a gyration period; can riowbe calculatedo The average kinetic energy is
'·
~ 2·
•/

. :!
mv II m~.
~ .. +MB + 2 ,.:,since ~ is the energy of rotation about the gu~ding
2
mv :A. · ~ · mu3-·
._·;
center 9 and _ft_ + j',
is the kinetic energy ?f the guiding cents~.· ·--;_
2 2 ·-·-. i
. motion, sinca '· i ~1,1als . ~ 1v 11 +tiE.+ &{ 6) •. Or to gi ye a. more forzn.e.l .. . . .
~ ~ '.
proof,~ start ..wit!t . v "" R + e0

.m
0

1l+ ~"'"'(~2cos wt· - ' .

. · ... -._

' '. . ..· . •·y~o"

l .. . .
. .; . ; ~
. ,.
m···~·2.
.:

-l::l!- ,
~~
- :::· .·'
2
mv ., • me 2·(A,J2 m 2. ' 2) : . :l
.'lben
2e .2s.
+ - -
2e . .
- · .(
.... .213 .u... +-v,
. j!i' el •'.; :~
....·!- ':
.' ' ' ~

.,: :. :~- -~
:_ :~ : .. ' . ,; ·' •. I • ·-.~ ~
. ~·>.
·:·
' . . ,· '. ·.'. r,
. ·.'
;
. '•,
,.;'
I, ~ . , .· •. '. ~

:·.·· ·.... ;:·:.·; .. ·... :· .. l.···~ ( ID._v. 2 .o} r.m + .m2 u... 2 ) 13
- :·:.·~·~:.·;·_·: ...... --·~~-e :.d.t... 2 II :j!i.
..
",
t' -:·::_
.. ·.·.
<!: H · m~ ..d~lc, :·.· .. ;
. ;.·
•·,'
+ ""'>
u.,:..
.
•. -'O'B·+- u.;.. ·Gl .v . ,· ';. ·.,·_-.::'
:.
.l!.i' a· e. .t!.i· 11 dt -~ :.

r••

::
;
....
.•
. '•.
'.

: ..
fL. ', u..., 2 .; &( 6- 2)
.....
dt'' '
~- .. ·....· .· . . . . : •
.. 'I,• ..

. ';: .....')
. : .. . ·,i .- ••
. (, '.
. ·. ':/,! ·'·.. ·;'t
,. I

·.. -.· ., ·.·

.. .. ,
.~ ...
.....
\"., _.'•
~·.
.. .....
'· ... :: ~

'.
'' . .
',l . '• '

·.,!. ._ . -- ,·,',

·· .. ·_... · ··
...:.·-·:·
~ .. ' •
·,:.
_. . . . . • !
·~ ...... .
. . ~ -~- -~
........ :· : .. · •.
•••••• 1_

-··:-· .. . . - .. '' .. .•~

.' \ ··.:. ,! ,.·.

; .··!
.·.. ', '
r·., , ..... I ~ _'
',.
• '

~ ,• '

·.._-, '' .·..

~ ' ; '.'.. . .

;
....·J
,._
''1;-
.+.'

:·.-
' . · ...
.... ·,
,-... · ..
.... .' . : i.
./. .. '~ .. . ·-· ,'- .·.- ··. .·-· :.- '•
. •:·· . ··'
••• II

...·, ·-··.· ::•. .·.,, . ~ ;·

•: i·. ·: •.• 1;,•\.

' . .. :. . ! ;
·:,,··_·.
...
•: ' :
··.
l ~-
-.: _ ,1_· ... ·,_· ••• -··
., ..
'. ',. ,_.··.:~. .,. ' . .. ' I.,· ,1, '·
• ·, ·~.
- .-· • t ' • • ' • '

~ ·1 . ,< ·.~./~: .,:. }\ j~ ;~,~~ ~::. :·.~'.:~(.-~~~~,~·:·~ . :~·.:;~~~f..' ;~. <.:;·~. ;·.~i,\:i.~\:,:i:·:J,::; ~:.;;: ~; :~·~.~ :: :::.'.:;;-;~··'.::. )):,·:::' ~::•.:.:<' '' . ·~ .~J.~: ~-~ ·:~ ';.~-~: ':~ . ~:~.~~·:{~:~~::~~·:;~:~i _;~/ ~ >.~ ~~·~~::·~· :h· -~~ ~,_;. _~. ~ " ..
 - 15 -

"'. ~"' "'

e·"B...+ ;'. vh' dt.
d-e, ~-
""S>
""t }' ...,., cExel M . m _._ m.· du.....
u... e . .J. VB + !!l
· r.. .. e e·
u_
.l:'.i .
o v II .---*- + m_· g_ u.._ 2' ll:ll
dt · 2e : dt . .l\. . B. ·
-
e ·
( ..,.. ...e~ ~
r.. )
, : .

.·. . '
·.'
.. , ' ..
. ..., -a- . d~ . d~ .
I,,. ,t

. E • ..1 x !1.9.va. + !:!ll:.. ·-.:,

. t . ,' ~ ..' . a ·( :.:l. +. me r.. ) . . ..
. ·; ::--·
. ';
.
.. B· e e .. "16 dt ~ · Cit. D. ' ~; ;

.
', ~ .. ·.':_.
··.· ..· :·... ·
.. ,_·

:' . : '.
··:· . "'·
. '; '' -~ . ;_ ., . ·.,.. •. ·J
', ~
't. (."

.. ·. -;
'', ... ..... ' :' i
.,
. ·.6
.. (·'
'~"'
{!;t} ·:.
,'
~- ·~ .
. . ., '

. •, '·

.. (.

.. .
',t.

~
~· .. ~ (~eti~:: energy,) = ... vh fa( V+ ¢.) + ~ o E ~ .~ ~~ +. &-< c-2) ~r· ---·-..:
..

,'·:'
':
~ I' ' 1 '; '

. ·... : ~ .. '~ .. '

.' . ~ : ·..·.. . ':

.. ..
. •,·• ,. '
...
, I;' ;,~· ·",.' ' •
. ·.· ..
. .. ~- ~

: ._'··; ... or
' 0 I ~
... ,·l
·'
'·.
·, .. -·
·:··
.
,·,.

. . ..
·. 1e _ddt (kinetic. energy) ..., i •. E (R, t) + ~, ~
() B(it~ t) · + & ( €- 2)
d.t. _s
.. :~. '· ' .
.·
-~
·,·
'

_;. ····i
•. 'J
,' ..
'• ·.: .::'

This is not a, surprising result and .probably could be "Written dovm directly'.·.::. t ; .; .. • ·-~/·-~

Th~
-b
terpt e.tt o
"'="'
.E . is the rate of. increase of .energy. due ~o work ~oneby tha
./
... '::_, · · '·
::·.·! . ... ~~ 1~- ... .t •
.....
. jB .·
electric_ field ~m the gui_ding cente,r, while H~. is the induction effect

of a time dependent field rmd is due to the curl· of

' ' -·o • ~ ~ ' • '
E o
acting about thG I
. ;

..
..
~

circle· of gyration$ ' . One. might wonder

.
....Why t-he second
.· .
term. doea not have '
·: ·~I ' ' '
:.~~-

dB dB.· . ~B -:· .
• the total time ~erivativEf. dt' ~ ""TI· :t" v 11 d 5 +·~ ~· r; B. The reason ia.t.t11at -~,

·a ~gnetio field gradient (a~

'
represented. by~
- . .
·v,, ~ + UE ,o VB'} .'does not
. .
: ··. ·. · . ... .
.. -.- .
; :... ·.· .. , ....
'_,,,
\ . '
~terch.ange~ energy betNeen.
.',l,
change .the total.· k.inet:j.c ener:gy,, but merely .
. . •••
' ~
t' ·· ... .'

.· . the ~rpendicular and paralleL components Cl

.' : :· >
t
.....
.....
~ < ,' ' . ' ',
:
'
.··.j
~ <
'
o ' • ' • o ' ' • • I 1
1
:
o
.• , •
o .. .
; .
...
.. .· ... .. : . ,.
·.• ·.·:
-.'~·... ~··~-~; . (. •.;' I, ~.. \' ' .. .
:

• ' '' . ' ' ·.~ '•/, .i ': '• . 'I ' •.'' ' ' • I • lj.: ··-: . ·-: '"
. ··..,. : . .
:• ' ·( . • .':':1. '. ,' ·. . ' .. .. ..
··, ·:, ~..-;: . -..:. :.,;ho:~~;.::x; . . · ·','
' .. . ( '.
'

,,
.
~ ·-; )

•'\'
.
:' '; ...' ..-,'
';_•.:.•·_.,·:.-.:.
' .
'• . .} .
. ;.
. •.
'
1.•· ·-.• ..
~
'. 16 ""

6'

.... We now have the three fundamenta.l equations (~/)I ~(j, 'and (;4/J). "·'
•~. . .··
·,. ·. ···.
:·: .. ~·-· ':_'.: ::.
.. ·...
I~ I : ·: ' ' :'
1
'
0
~ • 0 :, ·,f,
1
.~ ;
-· ....· ·,
'
t .• '·:: ... ' . .':
. ,.
.' :
~ '
.~ - ,. (
·. :: . ~- .' -~ . . '·

':-··.
' ::::-- .· '-
'

- .....·. ..
··..._..--.·- ,.-.·
. ' :: ...
and Eq. ~·?') can be writtert as

.'·, -~ ; '

.. i'
. • . • ,I
' .,
•' •,.

' (.!Jf() ' .·' ..

.. ·

+ MB +e.( V" +·¢) is a. constant ·'·:· :'.

.··.....
_

' ~ . ' ' ~-. ''·.

'i of the zero order motion_ along a line qf forcae
.. ·.l· . .--

!- In a _static· field, 1p.,. 0~ .. :·.

! --~ '
:>.~ . ·: _:_._. . ·-:.·:-.
! .... ·,·;.-_,,
f \. ,..:
and is just conserv~tion of energy~
t. ':. •,'
I. .: . . . ,.
.~' .'
I ·:·: : :. -~-- ~-.......
i '·.,
I , ,

-
! ' · doneo Tpis is the case of a st~ti~ magnetic field with rotational symmetry •'} ·.

. j''
(ouch as a mirror machine), c:tnd a static. E!l Where --'
E J.. has no azimuthal.: .
'\' :_:·;.!_ •. ·.

.'
... ·..
;·· ..

. component .and E_
11
,.,. 0 {Fig a,). The 'POtentia~ ¢. is thus a_ constant ,on· ..: ,
'•' j

. '·....
'

~.-.
.
..
.
' .
·. ,,
. -~'
'''

,•
··'
.. .
a line of Jo~ce~ Such a mirror machine has been named Ixion 9 and is ·. . ·.· -., .
. ·,: .
.. dis~ssed by Lon~~, Na.gl~ and.~ba. (~) Th.e zero-order drift ~
!
...
:; . ' ' ..k, . . ~

·......
.

•; ,.
is in the azimuthal the componen~ pa~alle~ to .B .of the dirl!:)ction~.
f • . .: ,, ·,·.
I >., ; ~: .. , ·, ' 2. . .. ·.
. ·'
resulting radial ..centri~al ·force .m~ /r_ . ha.a the desirable property, of .'.:
'· ..•- .. :,
' . '·''.'
. :.
· . !'ll8Jd.ng ·it more dit'f';lcult tor the. particle. t.o escape at the ends;,
. . .. . -~ ! . ! ,. . ·, ·'
:_,:; -·.
·• .. .:
. -_ I. '· ...·. _·:· .. -:·;.,
~- . ...,. ' .
..
. ·.
'·•'. '._·· _: ..
:.·
' . .·. .. ,·:' . .'.··
''
..
• ' • j.
''·
• ,j
·:·: .... ... '. \ '\. ·.,_ i••
' ~
.'··· 'I·'''

: '. . : . ~ '

---.· : '.•. . \

'. '·.
. :
. . '
' !•

I .
. ..:'. :,.. /.:-:·- .·-~ :·' .·'.\• ._.·:··:· -:·
.•
--i-
··:. .; ' . .. ' -~'

''
.F:·~ . ~ ·. ' ..
. ..·, .
. .. '
.·.·. ·: ~/ ~::. <· ·:- ~·
i ' ._,. '· ·',.
. ' , ' ·.
!.
 ,.
... ;

.·. \

I'
•.'
. ,.
. ... ( _; . .,

......
. . :··.
· .. ,.
'·

·• ,' ~ ..·.·· ':•.

':

.,; ... ·. ·' \

. ·,· .
.·.. ··:. ': ·. ;:.. ':·· ·.·.

,: :· ....
,·
... ; ,l. ···.•
..
:,·"
;
'
I ; ~ . '
. . . ., T .

#- • t .'·
., .
·.·· .. ··,·
'.; ....
.' .'•
I ' •
... · .. ·~
,,.
/ ..·._:·~
··:;
Fig.,,;;t, Mirror machine with large electric field perpendicular to 1l. ·i 0.
··.' . ·. ' ~ ... '.•

The effect is just that which would _be observed if a bead were plac~d ·on .....
' ··> ·.'
.:· /' .:.:_·.
·.• ··a smooth wire bent, in the shape of th? line of force,, and the wire then .· .
... .; '.
. ,/

. ...·

....
rotated.about the z a.x:i.so, This analogy will become apparen~ in the .·._,

:· ·, ~ :\ . : ·.'. :·,.

... :

• '
'':..

...
~
/· .....~:-::. .

" ' \ ' · ! , •.

~

;•
~·
·.· ··.·.·
'Jt <
.. r ; , ~

..i ·~.
.·

·: · :::·:··.:· :..... terms on the right side of Eq •. ¢?7) vanish e::r:cept R·"
followine analysis ..

Under the specified restriction on the .E

~
and
-,•
- B .field~, all
. . . ...
.....
-~

.. ,.
..L.
mv11 ...,
to -;--:- ~ • ~ • () ) ~ in this ;3pecial case~ and equal13 1
·:;·
. . ._. ...::! :··:. mv11 .

'·, ..;.
...
~ , ::. · . . e (cE/B)
2
~J :. t~ • r7) ~1 • Because ~3 • e"1 a 0, the factor ,.• .

. ·.·:..~_:\·· ~3 e. (~ o T/ ) ~l equ.ala :.. ~l o. · ( GJ •V ) ~3 • But (~ • i7 ) 'GJ a -~r/re

. .
.where ~r is a'unit vector in the radial direction. Therefore - ~ ~ (~ "Q' )~ : ""· ' · ·
1 3 . :3 : ·,·'
\' ·.
~l o ~;Jr. . ·; :·

over the zero-order,

"-·-·-·

line of force about z), the variation of cE/B and · ~l " 'er/r .with
•,: ·'
. 'l···.'· .···
longitudinal position must be known. ,The following is a proof that
. "' . ·•·' : l·:
.t:•·
':
... t

f•.
··.
··•.
~ ..
.''l

I , • ~ ,' •

,_ ,,.
•wl. . . . : '.•,'

;· ·~· ·.. ,
1 .. ..~.. : . .
• .f .• • ' •' •• :. • . ~' ' •••
.. ·,.
I • •j:•
•. ,1'·
"·,;··
..' . :
••• j

' . •'
:-·.·

'·. '1 ~·-.. ;",

' ' . '' '
' .',: ~. .. .......- .
~
 18. . ·.·.
' . ;-.
'

•• '• ·•. • . ..1>;! . •

:· - .· perpendicular. to ' B, flux surfaces ~re also equipotentials and ~ . is · i ;;> .: · ~· ::;;

.:·.:· !..;., , U.ei-efore a .funC~ion of 'i:. The component." of electric field at"e' :. ~;:_, ' , :,':

: ; ' : Er a - 71¢/ dr .. - (d¢/d ll!) ~'l/Jr 8nd Ez .a -(d¢/d tz) 'a Yl I d z. Thua' f' ' '·:
".· :.~~. ' E_.e ErrY!;/ o r) 2 ~.(~W Jz)2J l/2 0.¢/d~:.~rB(ct¢/d'M?) and cE/rB·~ cd¢/~'!9 ~: ..'·:·.::- •.
. .
.','
. ·:·'.
·.. '
which is constant 9n. a.flux surface. The quantity cE/rB is the.. angular ': 1
.,"''

. ~
. . . . ' ':. :, . ~ ' :~
ve~oc~ty .of the ~ drift about z ~and will be denoted ,by A •. ·.Therefore
mv · . · ·
. II .~ (~ ) ;'\ . m/e ) 2 A A
.''•'
. ·. . : ..
-:-~ '
~-- ·Tl . e 1 . is. ( ':lE .• . ~ .• v li ..a.: :re1 •· er;, .which equals 1 •f •t,,,: ,' .
.. . ''-:
' . . 2 2 '
·: . ··: .. (m/e)(d/dt)(...n.. r /2) ;. because '.,,

....
• ' ; ', I ' ' . I .

':

. 2
•.· 6
a ..0..
.
V·
II
~

' . ·. .. ' :: ·, . ~ ·>:: . ' .

; " ~ .. '
(. .,. . ·. :: ..· :·: .
. · 'rh~ in'l!egral of Eq. (J../) is then
'·.<' ' , •• ,·) '

.: ·~--< . . - ' ..
./ , .

.mv1·1 2/2 + MB ·-:- m..A.~ 2/2. equals a· .constant .of the zero-order motion
on the flux sur face"
...
... ,

If the subscript. c . denotes: quantities..a.t ·.the· median plan.e of

Fig& a. and e at the mirrqr (i.e., ~t the loca.tionoi'fma.ximum magnetic.: .

• ·. field. on the .f~. sur.fa.cs)li Eq. <.3/) becomes.

...
.l
. . . '··
.•,

...
... ,
·.•' v./12a Q% ·,2il c + (2l{Bc/m)(:l .~ Be/Be) -~r2o (1. .,.. r e 2/r~~)e' ::·~· .· :
,,.... .· ·.·;

Therefore
•· .
.~ ~;·f~, .··:.:: ·.
. : . '~ . ~
. ·.
'•
 19
'' '•

,•,.
. . . . . ··.
"'.'.' M a 0 1 the pDoblem is that of the bead slidin~. on the wire described ,.· ·.<
·. : ,· :" t '; : -·~ :
·,, .. ' ..
.. ....
previoUsly.
. ·i !. ..

'.': Terms containing ~ . in the drift Eq. (/7) give a small (order~),;·· ...'·'.:~ '
.. ::. -~ ' . f . . -~ . . ' ,.
'.
·"· . ~otion in or normal to a nux surface,, the zero-order. veloci~~ being Y.'.. ·' .. -· "· ·. ;':,./: · ·:~
.
' •; ,•.
......
+;.in the _surface •. When crossed ~th.
I, " ..

""~1 .vh. "e1 /B, the t~ir~

···.:.1
.. :·
...... R ' I

• . , .Jit.
.··_;.·
_bz:ac_!cet~ azimutha~
' .'
. ~/ -~ '
term in the square is in the direction_. If· E . is .. ·.

·.· .... ! ;·...

•,>,
.· ~
....
,I··. outward as :1,n Fig • .;a. , the fourth term is
!
.
·••.• ·
.. .:
~ '
.
. ...:· .... .•..
. . ··.·
··'•.....
:, '
:
...
' ,
.' where t} is t)1e. azimut.ha:l angle in cylindrical coordinate.s and ~z is a. .': :•' .':'·
·.. ,, : ... ·'·_;

A.
unit vector in the z 'direction. \'Jhen crossed with e , this fourth
1 ···'··.··
.. term gives a drift normal. to the flux surface: The sixth term ir} brac;:ke.ts .~';.. ~::~;'(<· · _;:' .:
.... is v11 _ ( t) U'. E/'.f ~.) -:= - v.
11 (G>/~s)~~3 ) .. -v11..o. ~ ( tJ r/;;a).""
. 3 '
-~v,,..n.
.
~3.(e'"r ~. 1 ),
• .,· :. . :····.<:.<.,,. !
. . •'' ·1,. .... t'':l

. ,\1' .. ...
hence is the s~e in .. this geometry as th:e '·.fourth term. The ~st term in . .-.· }.···<··;·> · . .
. .' .'.,.I ' .
,. - .

., the. a~uare braqk~ta i~. j,_r~ ... V (.n.z:~3 ) = ..o.. r (~3 , ~ 2 2

v:1 )~ =. -.d- r~r· · '>-·::.':>;·_:·:. ,:,

::::::·:da::io~\0 ~:t t::: : 1::·:u:::::: ::::.~

. .
drift 1n tne
I
::.:r,;·. ,:. .
Because of the two order ~
•
drift terms perpendicular to the flux ·
• -1 • • • .. •: ?:. . .,~-. i

·surface, there .
is
in.¢ (and.
ther~Sfore
an ord~r 6 change
. a,f kinetic energy).
. c .' '

: -: ~ : ' ' 1
.· .
surface.,· This change in · ¢ · can be calculated·
' ~ ''
' ... ~ ;·.
as the particle· traverses the ·.... -· ·. '·J _~. ,: ··,:

I
.. . \.
. .: .. -~ .:-~~~-~
I
: '
;..·.,
'

'7t ..:.'
•. · . •'
directly .from: the product or the dri~ · veloc~ty .normal·' ~o the. sur:t'ac® ·and . · ·· · · ~; '... · · :>.,
·., .. • '

. '-":!:
• 0 • • ••

·. '," .
•• ' ••

,;- -·. I•
' . ·... . .. ...... ~
' .,... :. • ' J -~ ; · the. elactric .;·'
. ·.· .. · : ·~ .~: }'>
;':. •'. . '.
·-:; ~
...
.. .
,' ' ' ,l

:·
"""··.
·. ; .. ·. ·,

."
: r ·,•, •

. ' ~.~.~. . .: ;, ,.. :

o';, ', _ _.•.
'• >

:·...
-~- ; .: ..·.. ' ..
.:·.I
. '{
 ! ,

,.
·,-",1
20 ....
.' . ~

•On integrating
. '

. (.3G'J >\· :.,:_:;. ·. ' : \

~
'I'~ 0
. : ,·_
·"·:
.
·:·'
. '• .. ..
' .~ ..
· T'n e change in e¢ caused by the first-order dri_ft off. the surface ~quals ·. ·~· '· . : • · ., . • til , • . , . ' ! l' '

.~ . . . _:
...
,' as the particl~ mov.es .in zero ordez:- on the· · .....
,. ~-

'i·,

. :_· ... ,··, :. ...

surface c. This result can also be obtained,
. . . The
by. energy conserva.tion.
"-: ·.
-. .'
' ..
'
-
'.,{.·:·-
~

.
,. ~- ..
'; · .:' .
totn?-- a~erage ~nergy associated with the perpendicular motio~ is
!
. ...

j· . . -. ~ ' :
.. HB
'21
+ m~ 2 •. Therefo:re
. 2/2). + MB. + (rouE.
, (mv·
. 2/2) rl
+~ 'P , is .a constan~ of
.
. . · ··
.:·-, 11
motion~
_..,. '·'
;"
But from Eq. (31) · (m/2)v 2 ·+ MB -. {m,{.2)u._g2 ·:. :::: ._,
-:·;·._
. ·· .. . the 'zero plus first-order
11

~onstant =.. - .2./l.{~2/2). ,.:·~.-.:· . . ·,

.·,' ::
·• . ·':·.· !-·- is ·a of. the zero-o_rder. motion. ·By subtraction .6 {e¢)
...... ·,··''
This. dr~ft normal 1;.o th~ flux. a~fac~ is not cumulative.~~ sinceJ ..
,, the s~gn o•f. vII . r~verses_ when the pa.rticl,e_ re.flects pear the mirror. . 4:

'· . . . ' . ·'· ..

,')
.•:;,
I
'.
;"._
The Geometric Interpretation~ of the Drift Equation and the Longitudinal·:.>·:·> · ::
·.·.. ·.....
. ·..· .: . : ... . . · ~
Equation of Motion .
'
' jl .., -:~ •• ~ ·,· :;. .
..

::. :::( ...>. ' . , ...... ·.

. ·, •' .. . ' '. .~

.r·~_.· : ._:;· y.~.· ..'r:. ·....

";

: ' ': ' .. /. ..·.:f._·. 1~ , . :-

' .. , }.'i- To get some physical insight _ihto equations. (/f) ·and ~6, it is .·. · /:-:
f ...;: __ .. ·.·,,_:,..,:
,,
I · .-:\ , :;.:.: ·: , .. instructive to l~ok .at the geometric reason for the mcmr:ence of some of.· .. :~ · .< (."'·.: ~
l· . :·; .·; ~~- :· ~- • ~ . . _:,::J,.,~' :-.~ .. '
.. ' .
......
:-._;· > the terms •. All. the drift terms· .in Eq 0 (//) arise because. of a variation· :.i />>··:
•• ;1., , · .
'
:,: ~· :
. ·:' .~ ':, ..
! ~ ·.. . ·. J at the gyration £:r.equency of the curvature of the;.particl~ trajectory., .

..
•
,, ,-1 • •_·
·:. ·: . '~ ' .
·....
This variatio~ results .in a cycloid like. motion at right angles ~o " . .I
·:
. ~ '• . : ·.. '
.. ·.
-~ ..

·.. The r.eason J'or the variation at;. the _gy:rofrequency is different for each

, ~irt. .Th~' f~ar drifts . c~f',c ~1/B~. · :~ .g_x_ ~1 ·~- and -~~ ~1 x~B \"

·..
•'
;.· .

.·have often been.illustra~ed. in·theliteratu're\~) and will net ·be

•.; .. '

·· .. ·
.:
•J • ,· ,· .
-ri. : ~~ ·... 'l

,, <
~ .....-: ..
.... , 4' For example 1 the 'Q' B drift occurs because · B varies
I ,
( .. ' ;
t··,. ..
: '
during a gyioat.ion .. period, a:nd therefore the radius of cu.rvature doe a ...
":'·; ·,.
1 .. ·:
I'
i
·:: ··,
·...
I, :·•.,
·. ~so ... The rem~i~g drift terms. in Eq. (//'). ~me from the . 1£ x ~l ..term.:
_
....·. L: :; _.:·: :_ ..
:··.:
· in Eqo .. ( 13), and are usua.l,ly described as the result of a.' d 1Alembertian
' '~· . I . • • ' I • • •' ·. '
;
•• ~ ;.- l •• ' ' •

,•
 •' . ...,[ .
.~ .
- 2l ....
. ~: . ;
:• '·· ..... ·. • • j; -~
·.,;

does not help one's geometric un~erstanding .. ' . '~ ' : ', ...~ -~
.· '•, ~
'1 .'
·.. j
...•
•·

~-t~b Consider, f"or. example~ a physical situation in which the ..•. :.}." ·:· : :, . : '· ~
.. .
"' d.rif't app~arso Leta magnet with large parallel pole faces as shown·.-:::: ' · .··, ·.:···:.;,
.... ..- \. . . ! • :-~~ ' ': . • • • • •:

in Fig.~ be rotated about. the Z-~s ~to give a ~ ~ 1/ ;> t =..o.},. where. y .· . ,., < . '•
> •
.
• • ~ ,.
.
;

is a unit vectoJ:" along the Y axis and 4 is the ·magnet'~ angular veiocity
. (..a.~'- w)~ .·'Be.cau~e ther~ is ~ : ..JB/ dt there :dll ~· general b~- an. E:. I
.~
.. ,. .~:
'. '
". ~ ~
~

... ' ::·:· . ~ .:. ·.: •. -.J

;
. ·.:·:·.!.:
,·. y ~- ·~· '·._.~·--··)'·:"'~:
·. ··' ··.· .'.'!. · . . ..1 PiS VIA C.V : (/£· . .Il-l ~G •••. :·.•~: •. .
.··,. . .. i 0 Ol 1"40 'J
,I', 0
•
c
...
-:.
.. .
Yl oi:e'. . . ; • .. .. . .
. . ' !' •...

.. ,
''
. ,• · · ··,· · '·Vr:t:-fo,.,..,~ /J.c,.:-,::
I '.J . ·. · . .. I' ;.'>
.. \. ·~ ;· .. ; ; ... 7 ':· dtrc<-T'~ ;:o;
.' .. ' :. .. :. : ·.. .'J.
.. . ~- "; ',
. :'.
..
"·
: ,1· • · ·.. r'>o~ -L~.· hoi;~~tc.i ' ~~
:~ ..~:. <::.·:·~·.. ( ' ~
'
. . ., : liicA-~ . ,· . . . ;
.·,._.
-~·· ; . .. . ··. '.
,. '.
:.-·
·.>.
..
,• :' . ···~·.
... ~- -: ; . ·' .. .;.

;.,. ·
/''
... · ': ... ' : • '': •• f •'~ •
..
·

•: ' · : •• ·.; ~--.... ,; ·: •• ··;~ j. ,.',''·,·:·· ••

··...... -~ .. ••• l . ;
.. •,.:;,•

:·. { .·; . ·'· ' ._.>:•/

.· .. :· .
·:. ··. '
. . '•
,: ~ ' , ·.,
-.. . :
'· • '
'
I '": <

·... :: .:: ', ·,:, ·•:.::r:-:

;_:{:'·-':~::',{:;···.'·.. Fig.3 Rotating magnet .gives a 'J"J1/ Jt drift,. • in the Z direction : ·;·,.,_··; ,:; • .·,.·,

·.. ; <' . .. and therefore there will also Occur ,the drift lrE .~ ~c ;;1 x E/B. -<~e.•lwo' ·F: · . ' }).%
. :·':-.. . ;_.; ·' terms .in. Eq. (J/) t..h,at ~!"~ pr.oportio~al.t?. ~l ~.J'UE/ Jt. and to ·>. ·.·;:~:f>:·· -:.. :- .i-;~
~- ~~:· :?.·
'

.·. ': ·
•. ·.-·

·,. ·
t ,.'

~l x. UE ~ V-; ·are not obviously zer~,, although tl:lis will turn out to . . ' .· .. ! · . ' : •·· ·:(
...

·.··:, be tho case. · From .c V x E' a. - Ji!/di' one finds '.dE/~ x o.n.B/c •at ·,· _<\ >.{ ' ~. :
·..·.. •·.·;; ·::;·:
• · zero time, if we·· :tak·e. Ex. and Ey zero. ··~hen ,. E~ :"' .n.Bx/c + E z (xaO)? · Now ''. .
" __;···.· 1:
", / .
' ;

~-:. ::~<:.:· . :· ...., '.-~ . ~. ~

: Ez(x=O) also ~quala Ez . everywhere for.~ • O. Let us assW?~_.there is': ', : .. : ., >·' .··.:.. '.
•• /'.

,'>
1' ;;

I
':

•
. ,
•. · · .... . ·,.....
,t'i" • • •
no E ·ill th~ absence of rotation,. so .that', ..Ez(x~>. .. is -zero~: ~Th<3: ·~ '
.
' .:
'
'·
.)

.. '·.. .
,'
'
I
.,
"';.. ..~· .

.
I • ' • " • \ ' • ! > '"',

' : ~ ' ·. .
..... .... ::.
::; ·:·
drift· ia. · ' ·· · ·. i~~·~
: ... ·. . '
.... ·
·. :- .:-"..· ... : . ·~ -:' ~
_..
·Us ~
..... "'I
cE_ X e _B ,• x..n...r
-~-.
a.t zero timeo · ·.
. ,.

.. .. _.:.
.... ....
· ' ',·,,".
·· .. · . 1 . ' ·,' . . . '
··;,
~.

•,, .· , ....
'

.. ..
~ I • ~ • •
• , .. 1

-.; .·.

·: :.· ....
 1·.
I, '.
I
. "~

... 22 ' .
•l·'
~ .. .. .
. :~ ' ::.·
•. . . >.
' . .
,· ·t: ;_·:;_ ·.• ~:· ..
leaving , ~
.. I '.

..
···: '

. .
.•
.,
, I
.... ·
'~ .· if.J,.. "' cE
~

~~(~ 1/B) .. + v 11 (mc/eB)

.

~l.x. Q'e1/Jt ""y.n.x + z~~.A/(,t!J

'

o ~) ·>r. --~ .
' ; ' ~ .. •' '' I • "I > ' I_ I' •.·•
'.
I
,i

. ~ '.. ' .. .·," .,

t ~ . ·• ,_.;
0 • ' I I i

"" .· ·~. ·. , ... :·, ' ., ' '

~--~~: _b.~1/at
.:>::-(,·,·_The
.·l
.drift ·is perpendl,cular to .the page ~d of ma.gnitu~e · v A/~o _
11 .. ,

' . . equatio!} of motion ~ {)) becomes

· . ·--· .. The f>8:!&l-lel
.. ~ ~ . .

;. ' .
.. ' ' :~ ,. . ~. .
'··.·· .
··:.• . ·-· . - . :., ...
.
....·.
.-·. '·/·
., .
. ..,
~

. ~·· . .'' ., . This ia just the centri~gal~cceleration a.t a distanc~ x from the axis ::- .
~.
. ~ . ; . ·. ;.. . .. . .. ;

of rotation~
··.: i'· ·.'

The geometric .reason that the curvature .of the trajectory_ varies

~ the presence :of. U~l~ d t , is that the perpendicular velocity J.~ X ~~

·.·:· ,.l.

varies as \ _changes direction, The drift velocity' ~an be derived

., :'.,. . . · (exce~t possibly for a.··.n~erica~ factor). by:,holdi.ng ~i fixed for .half'
\ ;-,: _., . "' I . .
'... ·. >: :.- ..
';•'.:,·.·:::·: _a gyration per.iod,and then changing~its direction !or the next·ha.lf period, __ .:'·
.. ":"· ·: .·,--·;.._:·,·.

.· i
.
.
~: .. '
etco . A view: along· ~he X axis of. Fig. 3 "1-r.ill ~:.appear\:at 1ift:.. s:.t:;:.~.rt.· .~:. tn . ~ '. .
: -~ -~ .f.. . .
' ..-~~- . ~ ,; ....
• ,r ·. ,...:...-··j ·.· ·:. · .. ·...
•, ·'·
Fig. _:'f." ·'.· .• ... '

. ··\

\'
. ;·.·
.... .,_ ....
.. . ·
) ,, : 'I.

·.·
.' . .. ~
'

' .. ~ .. ';

I. -...
f·•. •'/J.

....1,,.
' · I·,· ....

z
'·:
'
' -.:. . ~ . '.•' I

. ., ·.
.... ·~... .- ··
... ·.··-·.
,.·.·
;"
. :· .
·_;, .'.
~ '.

•
,.
. I :... . '
., . . • ''-: ; >
-. " : .,. . ' ~
,-..

' .

i
-:; .·
Fig.¥ Geometric explanation of the .. tfo1/. dt · drift~ ,. · ·
-~~' ·.·
I .. ·.
I

!
'.• •
l@t ~- be. th~. angle bet\\feen. _l . Md. ~ 1 , so that·. · vJ.. is . v ~in£':
_:· .· . .' ?· ' : ·::. and~ .v ~~~is· v cos &·••. ·At the end. of the first_ half periocr (y >O)·:.. let .. · ·:.' .·
! .. ., ' '
~ ·. :_ .~:. .
1
' . ... A ' · A . . \ . ! '

i ... ,:·.·.··~: ·:·-':\. e1 . ch~ge .~Y· .a.ei ~-the_ y· direc'~'~?~.~- . For the· second h&~f' ~rlod .. _ •: .

1. : •• · :=· ·._ _ (y ( ~)_)v~; ~ll b~ ~~n~e.d -~~ :< ~ ~.~.-7;~~::.:~~-a& A! :

0
·,v 18 .A~ .::·:~.:_;·~S.· ~~~ .':::__ ::··__:.
· ~- ·~~~~:~~_:ti. ~\;i:~~;;f.:~:·:~L:_.~_;·;,L -~- ·. · : , \:· .· :~.~:-... -~- :.~ -: ·. ::.-;~,<:~:·< ~<: ~·-";-~ ;;:: , :_.
 23 .·!·...

. ~. '
·.·

. . ... ··· .. ··
velocity equals the . s~Inici,~cles
. the diaineters ..of the ,two,
difference in. . . ~·.
·.;
.. . . . .. '

Since' e equals..• {· : :'.

:~ ... :' ·..~ ~: .
'lhus the . :: ~

.. , ....
dr:i,ft .velocity: equals v 6flw , which in this case. happens to be. correct
.,
:·· • .1, ,., ••••.

, .... · .• ··.··' .. ·even to numerica.l factors e

··' ':..··,.
~ ' ' '
Similar geometric .derivations. can be given ,of the. other:lc::lrifts :· .. ..... :
, :.. ',1.

:i ~. ' " .. ~ .
.
<J{)~
...
'' .

. ~
·~

..,i ..:·.:
.'

~~.' :;.· .•· .' . ' v11 and· \IE. in Eq. : ·{:. ..·,·::
In Eqo (ao) the parallel acceleration .of the guiding center. come~ .'·.·~:~) . .. ·.~:
·:·... '·1 ,• ;. ••

~omponents of. g
' .... ! ·•
....,. from several sources. It is obvious that . the parallel. ··.· :··
.·• ~~ .' . : t .....
. ...
. and E will accelerate the·particle.along the line of force. The third ,. '• •,.·.:,..:_-,·
. ··.... ;..! ,I
,,

... -.:·
H ~B r· · .. -'_:·..!
'; C)s is the well-known mirror effect. It can ba understood by.
·-.;''···
~ ·.. '
('
. .. ~
..
1 {'
' . ·. ~- ·.·.
..·.· . t· ..
field,• (Fig. SJ" The force
·.
on
. .;'-·..,·¥
the 'particle :.

. ·.:·.
,. _l
.•.~ ·. _;;
·:-.•
~ · '""~> ~ e ~' :· · ~
is
e
0 v x ~(r )' .:
"t. _., e "
0
...., ·
v
-:t )
{e1 11 + vJ.) x ~Br. + Bz) "" c v.i.\x: (Br + tsz • The ..
., · ··· ·
•··.'>. ·;· .
~"';A. ·x B~ term is a radial force. and gives the centripetal accelerat~~n. · ·The.·, : ,·;·:'.:· .· .·~: :·
' . .: ·.; : •. ·~~ I.~·,'~; .:._:.
.. ·.-. .• :. . . • _". ~-. !

.: ~- .. \ . ;··:~ ~.:.. ..

. ·:····: .

.··-.. . •.;
~- ; ; ·.•. ; >
·.,. !· . . . .

•.: ..

··.· . , r .....
..
:•''

. ··. .' j

!, :-···
1.
1.
. ·... ·;· • :: 'J. ·• •''.·.

... Themirror effect• . .: ~ .

·.'.,'•
.' '} .. ' ..-~. .. . .... ·.. · f
...
...
.. ,'.···.
.•;
'~ :' .._. .~. . ..... ~.. . 'i : :~ . l .
. ... ·..'
·t. e ~.h· :'i :i rJ.: r:\i.: .i :1J: , :t-:~l~(.\ e ~~~~!.; ,~•i v/: :i. {h r~·· clf\lll. r·.ilJI:t~'.J.. :,; ~~·~;~1/.·l.'/ ~t1·~m .: :··1.i1t~.,
:-.! . '.~ . :" ..... '. . :·.:-: :··~:\': >' ··:_...,.·· ...... ·-~:;~·~ . ;·· ....
'/~~-~;· './':~ . .; : ..
' ~ '·. ;·,
.. (, ··, ... ··. ,. . -~·

:': .. ·, '.•.; · ... :. :.·.

i '· ..; .·.' •" . . ;.·!-
~ ~ J • • _:. • ••
f ·/
~: ~·-·:

··-· .. . '·
. /.'
.....
'-; / _:.. .... ;
~-
·...
·< ;·. ': . -~ ..
:·: :.•· .. :f .;
.. .'...
' ' ,: I_ •' ,.' .:
. ':•".
~ •.·. ., . ··]', .
. -. ,c··
.. . .· . ,;_'
.· •·•.• • • 1.·

r ..
• •• ,! • _._; .••• : . , ·"··
• '~·... • ' • J ••: ::. .. .' :·.:, ,1. . ~
I~
.. •
. r •
: \ .. ·~>
.·,

,·;
.... ,-:--~ ·~ ., -~ . • ··~-·- ·-w...,•··~,-:-.,.-~--~~~
'..,¥ :- ,:.;: <- .•

i· i •
: _ . .J'•
2.4
. I .
· ....... •
-.
•
:
,<,
.·
.··.·

~ -;..&... x Sr term ls dfracte·d to. the· ieft. and eq~als -M ~~~ ·since_;-:·:_';:_:,':_:~:;;_.:
.. . .· ·... . . . ; .·
the. Br at ,a gyratipn radius fr~Hii the z~~is is related to . __,;: . i :·- \_:·
....,.. -.
3B
. '"'TS
by- V• :B: a O. ,.. . -:~-.:·,

The lf:!.st
..
term· in
: . .
Eq. · Q.O)
.
and
. .
·.···. _:.· ',· .··
·: ·" '. arises ·because the :Pl rallel veloci:ty v
11
is ialtered not only.

..
by an acceler~tiOJ?.·. j · c;>f
. .,
the ~id~ng
. .
cent_er 8 • but also· _by a
.. . e
. ..- '.
'
,';'' ..... _
chaz:.ge .in _directlo:q. _of' the. B : fiel<\ withOut. a change in- R., ..
; · ,: ..:.: ··.A specific. examole oi' this t~~'S seen· in the ·illustratio~ c:£ the; :' ;:<··:. . . ,·,:
. . .... ..;. ·- . 'A. ·-:... .
. ···- ·:
'
'.t'.•.' • .
't·' ·.'' 0 t:J . ···. ..,
;-~ ' ___,....:_
drift due to ·___1.
i) t • . ...... ~.-·. ·_
: ·-~
:·._· ..
...... -
'
. ·.··
.. -,,.
. -=· ',·; ~- ..
~ )

' .. . • <'• ·.·

.. ...:
4.....

' . E.- The RelativisticOasa . · ...

:- '. ~ .-.~ ·... ·. ; ..... . . ~-
-~.
I ' . ~ .: ; ·. . ·.· '
<'.,. '. .
·,_··::· ' •• -~he drifts and ;Longitudinal .. equation of motion for a
. .... ·
1

·'"
. : ;·: · . c·._: ~article of rel_a tivistic ~ne:t"gy c~ now be deduced, at least
. ' .. . ·.·· ..

::\for- the ~ase :w~ere .EJ. and. ~ -·~e.' or ~- ( <: )G To- begin with,_ <· ':_ ·:~ >,.;1

. :· let us· consider .the case where, E :~O; the' relati_vistic equation· · ·, •· .;:.... , .·
,'·;
.. '·.
•· .. :· of motion is
i "< <. ,· .•
...
'·. ' >' .; :. •• ~ •

I ,. ·•.. '· . -'~···... ; : ......

.-mo
. ·
(·
I-.·,,·.':. . ,,. d-"
v .;,
i .• .
I·.
.'' ·:~:, ·, ~ ~
.
,. ~
.... '
'···
(37>. :. ·, '_..>': ,:· :-'. :: . _
i .. ' lll-(9 2' at'· ' .. ·.,
..
. . sinc_e e~ is constant ·1n .the absence or i.. Thus the trajactory · , ..... ·
t's corr'ectly. given
-~
in
.
datail. by. the non-relativistic
. . .
equation · ....:·:· ~ ·...
.. ,.
:.·> :::_.··> ......
..
·· ... ·
··.:-. ·,,;
for a pA.rticl~ 9f mass
m-· ·.. : : .·:;.

·. r'·

.... ;

.. ·... ::.....

..
. ·.
. ..
 25 ....

fM::
..

~~ 2
;'\

' .... ~
•
1.L "".
. e
-1
B x
·2
VB+ 1e a:;]' , . W/) . <
·<: :·.·>.;
:

whe:re Mr 1s ~oB . , the l."ele.t1v1st1e magnetic moment~··( /t'P} To .. ;: ·<. =,.· ·.

'.... ,.
understand; physically the relativistic. effectp .consider. ·the.
J.. ~· ,.. ! ...

VB drift term in Eq. (7{0). . It is y times as large as ·the :. ~ < ;

...
·.non-relativistic ·!9xpressi9n .for the .same V;. • Relativistically. ~· ·
the partic:l'e. has · Y times~
:·~-

. .
the. mass,·
. . hen((
.
a Y times the radius .

or gyrat~on, .and ther~fore s~ples. 'I tim~s as much of~ the B

'·· ·- . :

· · field .inhomogeneity as non-~elativistically •. Since the 17B

drift. comes· from. the small difference in. r·adii·. of curvature on .;.. ··.·
'.· .
'_:: .. . ~ -·: . . . . . . . . . l . . : . ,.

! ,.

. , . opposite sides ..or the "ci.r.clfi" of gyration,_ the larger the rad~us, ..: ~ ·,· ·
J' <'. ~~- ,:· ·-"

,·.·.·
.. .·,·......
:.·.·
•'
the Ja. rge~ the difference, and th~refore .
the .larger . the drift . '

. '·:J·
'

·,·-

i. . . ~ ~. ':.<:.··velocity ... The fact. that the period or gyration is inc.rea.sed ... _-_:,
'.: .
>:'<:···by y is just. comp~nsat.ed by _the ra.ct _that _the dif..ferenc~ .lle .·.· •.:" ..
;~

.·, .... ·
·:;: ··
'•
,· : in radius ,of curvature on the t~o, sides __ of the .orbit _for: a .··.·
.....

. ·.·'( ·':.'
. . ~ . ·,. , .. '.,;•·
,,,',I', giyen AB betwsen-the_two siO.ea.is also~. times.larger •. These. :'·": .... .... )'•

.... :. :: ·-~ • ·. .:·. ' .t . .

· · .·.-. :: ·., .· •· statements }'till be cla.rif'ieq by an order ot magnit~de calculation. ;
·' . . .· ~~ . . ...
-:";:·,.:,:::.·.·':·\<··sim.ilA.r .to thq,t.in. Figo (~) u~
',•' ,.;,.

f'or. the drift. The radius . · ... ,. . ; '·"

.

...·,
.r.,'
,· ..
· ot g~ation .is 'II
' .
•'
. ··;·.
' ' ' ;
. .·.. ;•··
.·,,,

....,·;• ..
(~,..'"l>\ . -.. <:>···.
··.:
; : .. . . .
• . ··.· .. ;·. .
~- .
'•' "·'I
.
•

.··'
··''
·.. · . ·.
i '. ··;' :. .·.
I

•" ,,

: -~ ......
. ::~ _: .. : . ·.
'
. '·.'

.. . . .. .~ ' ; ....· •. ·.· ..

, I
····. '

·.. --~ ..... ·..., ·. i:•.

'.. ·'·.,
.' '
,:'.,
.
.' I' 0 '
~ : .· .· ,:-·-.,;, . ',t • ':····?' .•. . '"· ' .·

• ,..i
.·.
·'.'
·-~·
• ,.
.... . .·.. ·,.•'

. . ' ;, ': .': : ·..

. ....
·:-~~-'.._
 .·., .. ·
25 •.•

:··.
.··
. ~. :
...
..
.-.•:. . ...,
; ,

•.;. '.
... .. '
'

.....__, .
... v
--a- fVB
~(; . _( Yff ''>,~··.r';.:·,).
;.'\ . :./. ,·,
,'.i '•
·' .· ... .L..
'·'
·.·.· ..... · '. ··:- '.,":,::

... ~
•·••• 1

··;and· ABo. fVB 1::; Ytimes larger.relativistically than .. . ..'

···:r-t·. ·:·-·
~n~relativis~icallye
.. ,•
···.·.'
. A simil'ar argument. holds. for the term .}n Eq. (~}<ia
.'

aE\1 . . ' 2 0 el , .
containing ') s·. .• Non~rel9.ti vistically,, v 11 vs is vlt <it .s.:.';. (L·
. , .. ·•: :::; . /-:
·and the explanation o:t the dri.ft is .similar. to that gi van ·in · :·: · · · " '' ·
. <secto$l> for the . ~: drift •. Re~at;tvistically,. the increaa·a -. ... .'.:. ··_::···/·_:~··:· .. ,·
1\'o ...-.... ' ·~• ' • ' ~··,J:
:'·'·, ,_:· < ./l,·•

mass. rln1JtJ:P~e-~s. the gyrat.ion period by. t ·· The difference :{;:_'b · :·:.> .. _
-·. ~'. : . ..

. in radii o~ curvature. between one· halt: per1,od. and the

. .
next
. is .. ·......
,·1 ••
,··

,.(from~' ·Eqo·~~) · .· . ·, ··,

.
' ~-' ' ....
·.f··.:,:· .... '
··'·
,.· _..;f.

.:- .. . b. e Q mo t c . ;.•_. ;. ·.

. ·. ·. ··.· -eB A v.J... •

Drift. ~eloci ty due to ~2J.. ts
~s
.,
.,···, ,p }Ae( d' -a ~ b
I e!== A V .,
arlo· ..!... ·.'

• ~... 0

The parallel equation of motion t:?a··.· ..

in the absence.

:... , . · 2B
• ·-:· • < :.~ •• •·.·.~:·: :
.;:_.,. ,", .·
..... ·~
' ~ ... ' ·~· . .. ·:

.. '\'
.:•
·. ~. '
I·
.~ ;
·_.! '.
f ,!:' ''.,."
' ~ ' .

·...

·• :.-.
··.·.:. .. ~. ~-
·,,
; .
'·
 ... ' ·;· .,
·' " o : .. I ~:· :, ' I
.
.' 0 '
· .... -
~. . t'' -~· '. {' .. ' "-'. '
.·.' . ~--
· .....
l , ' : ! ., . ., ~. . ~ 1 ' '

.
.,• • J,;

times lsrge~ tha~ ·.

.·. . . :
,. · .... ;·<<·•:.<From Eq., (47), the parallel. force is y . -
..
~~ ~
.

'. ~
~

{'

~' .. ·.: ~

5
. ~ '

·. /, non-ralati vistically •. This can be understood.from Fig.

·' ·_,·,-.·..,
: ~.:.:::,:.;.·;and the 'non-relativistic explanation. o.f the mirror effect. . . : ., . ·~; · :· ... ·:;
~-,,'."··_-._~···,~ . . .- • . ·.=·.··.:_.····:.r·"·;:···~~
i ~:. /:· ...·, '·._Rel_attvistically,. thi ra¢tius .. of gyration is lB. rge~ by Y, and.. ;.< .. ·,, ·.;
··.· ... :
' •, . - .

. . ·.: · ~:· · therefore B is larger by ~ at the position of the. part~cl's ·. ·,. :-:• · r~ ··,- .. ·. -~
; ... · •· ~·-.". ':"'>-. ·. . . r'... j • • • • • • ••• •· • <::~-- .. ' . ~. :: ~;.·
.• ;r

:·, ·because or the convergence ..or the field lines~. ·.,··<:· _·:~:·~ ;:·.... ::· ·.
• _1.
''· ·xr · 1 is different from zero.and if the fields are . .. ~-'.
; .' ·•. ·.. ·' ·: : :~<
;.<.. .. ·:~ . / ~ .· .
-: -~ \. ';•
·-
'
:~
......: •'
' ..
non~sta~ic~ ,the drift. and parallel equations .cannot l~gical.ly
0
' .::
. ·.:

, .:'.:.·: be obtained from the. non-J;"eli:>tivistic onesg si.nce '( is no. longe~ '', .. ..:' ::. '
' . ~
·:~ .....
:
.r ··~ :.
l . .
·
. :. :. • • ;; . 1' •
......'
,. : ·.. ·· · ' . constant. One might surmise. that the drift equation (.V/) ·would·
:- '·.:... . cExa' . . . : ;: ·' .. _.!
·. ;;

..
'

b~ modified by the addition of: the term l ·, an9:. that _the· ' : ~ • r ' •

..,. ~
'•·-·
~-
;
'.;. : .
,·
B , .. · .·'.
loJ?.gi.tudinal ~quat ion <-¥9) would have ~e term added.· eE ·:" . 7 :- ·._.- ....
.... '
11 __ :·;'.·:.··:··;:·.:. ·.

· · · This surmise. is· correct for the ciase that . E L. is " .( e.)_, and
·... • ::.·
the rela.tivis.tic eq-gations .become. .. ... ' .
~· .. ·. ~

2
.~d·-·,
1•.;

.. ~ . .
-
. p IJ . 1 .
. s
.
·~ ... '
.
.~
\.
:- ·_·

~ .

~~.,----- ......, ' .·..... ·'

.· M
_dpH
dt
= eE - ..!:. as
JB.
· Relativist 4 c&)E
. ;;-:" • 19
)·_,· .. (s~
.11 . ~ ~
II
'
• ·•

p.L 2.- M E.L and rt' _, / .

t::'8· ( !tt_)' ~
' .. 2mB ~ r ""·Constant, whene
.o·
PJ.• m ~-~~/)
., · ·'·· ... ·· ,. . perpendicular relativistic momentu~ · }./'·/' . .
. :..: . . : , ' ·. P;7,f 'to ctfdu
.. ,. . ·. H)J~"' c ~~~~a
.~;·:.:\·;.~·~>These equations were giv~n (without proof} in reference (~o) .. ·. r~J"t·,L})/~·/,j[·
i· ,.;>·',' '. The mora general case where. E.
is <r(l) has been ,~~.i~&!~li~·~
:. : . .· . ( /..I )
,, '
.L .
' (l!J i•: ~-r ~
• J (ft>> (~)
1f
i:.~·:. ::<·::_': . •. ·:. studied by .Vandefvoort. The method. is neo.essarily covariant .. ·:·., -: '. ·
·:. I ,\

. ··-: . .-.' ...:

and· is not restricted. to cases where, E . is
. . u
or '(t. (. G ) ~. The ·

i :'·: c: •. , : \ : : ' analys_is :will not be reviewed in .de.tail. here, but only .the.
:, _.·';\ •.•. ~·~_I .·, ·• .· . ' .. . . ·. . • . ~. . , . • . . • • . . . ·'.:.·
~ d ... . '

·~;-.-.:· ·.·. ·.: · met.nod 1n~1ca'!;ed .and the resulting guiding canter. equations. of · .- , .
1
. .,· .. ··.·........'.:. '· ' ···:· . . . . : . \. . .::~· . ~:>·.:. .:~·· ... ·
... '~ ';·:' ~ ·~· '·.·: ~-' ..
,.··.,,·.··.:· .. ·.motion gfvenJror· the oase .where· . Ell· is <7( If:)),. The. starting·. , ....
. · ~
' '. ·. .·. ·.. . . . . · . . ' ' . . ·. ( ~~ ., . . :' ' . ;

r.·.::: ·~. :-~ .:. :~: ·. :)'~i'~t:. ~e·. ,;.~~e; ~e.la_tiv.~.at~·c· 'fo·~.~di~en~.ional. e~ua~ion ·; of ~:tion .:; .;.: .··:(·::·;:·:.:. ' . .
'(···~··,'. ~~ " ~-· •.'.':'' ...... •·'· .•.. ·.· ~-:.-·';··•.·:·~··· ·,, ' .·.:·.~.·:-'
······.:.,. ,~ ·.:::-::··_:_.:.::.: -~ ,··.· ...~--~··· . . ._.',·'::·.. ·-:'r··.,.·~·~
··~.·-.~~·.\.·~
~.. • ~
'~~,~ :_.~',--~~
... ·.',·.·.·.....:.,·.·.··.·.·,:·..'.:.'.·,'·...·,..,,:,..:_.. ...
• ''
· :~.~ :_·.v~~.: ·,-~·u~, .·~ .· .1:~·•.',:;-.•·....'•...;,•..•••.•., ....~_':
. =.: .. .. ......
'• • ,.\ -~ ~-~ ,
'. .
: •• - . : ; -~ .L~ ~~\·,:~\ .:• ~~
.·, ,
~- - .· ... .' ·._... ·.
'• :~·-'j~,~>.·~·::·~:~~-~~~~·!,:',{:;~·;_.,{:f:~_::::L.~~/3.~;_~·~.-;··.-
. . . • :.: ·: '; .·., -~
•,. ~:-. •
.
0
•.'
' :·
.
 . .. .
. ~ .... . .' .•. ·~·· ....... . -.
\ .!. .

'·_; .. •. . ··: '·

. ,_.

d. .· . •, ..
j·· .·.·: .. :28. ·""
,~,
·· . _;_·
,•, \ ' J
..
: •' :-.•
.,
;,J
. ·.
,'.·
': ; ... ' :~. ':. • ,t

"· . 2 ·. ·.·· ·.•·:·

~ ! .. . :~_:
.

;.-
' i .
·dx _. . ·.·• ~·
1
... . : ::-.
~2 = '1;;,~. d.,... . ··. ~· ... :

-
~ ( '
~. .I '· .
a~ ..~,.A.
••· ' ; .. .'· • • ·.. •• 'I"

,j! - "'
·.·. ;·
c·.··W~r·_·::·-··
_, ·. ·,' ~~-~
' .. : ./ '·· .
_, .,.
. ...·.
·_
·.·· .. .~"·: .
':.~-<· .f
r.~· ~ ..
r-; ·.. ,: .•
t ••
. .. ' . . . ~
i . ~- . ~· ·:.

r.·. ·. ·.. :.:., . .-,-·.~~ : .~ ' ... 0 ·· .. Bz ... '·.•By

...
,.· •iEX
"· ....... .' /.
·' .,1 '(''•
~
.. .. ....... •;.
~... ~ ~
.,;"..- .. \ . ~ ' .
~ :' ~: · . · · -B 0 B -iEy
z X

-B·x. 0. .-iE z,. .., ...

iE 0 .,·.
Y· ·,

. -~

'is the foUr.-vecto~ · (x.!.' Y.:J. z.;,~ ict). ·and 1::" is the proper time ..

1 .•':. ".\.: ~~ '.

··. the· solution .of Eq. · (,).2) exhibits. a gyratio~ in four-dimensional.space at- .: _; .. :'
. ·'·
' •, ~ ..''·
·..... .'

:.. ~. where ~is the angular frequencY,· in radians, per unit proper·. time. The
. ' . d't'
·, .~ ·.-··.
.
. ... . -~ :·.- .· ...

·... actuaJ. gyration :frequency; is w .:=

·: ~- When
eB
- E = o,
= ~
y
·radians. per unit re_al .time~~: ..... ·

.the.. .frequency t.J reduce~? to the· usual relatiVistic. fr_e.quency

w'!;" dt

; .,·.·..
..
. more . ~ ·.·
·"~

The gui<ll.ng center equations.. of motion are the equations of motion-. ; ., > -...
.:. ./'
. with the gyration _at· fr~_quancy, w a~raged. out. __ There are three equations, •.

· cor_responC!-~g to;the f-irst -three components of, Eq. {5;iJ 1 .which give the · ·

actual gUiding center motion. in three-s~ce. And. cort~sponc'ling :to the

_..,..i
I'·

. . . ,1'
.,_ .
::.·_:·::fourth component 9f Eq: (5'ti1) there ··is an equation ..for· the average .(over a . . '";
- :·->,~::·.: ·.. .... ~ _. .
r::r:·
I,;·_·:
'·'·'.
!f.::··::
~· : ~. ~ r•

r:.
I ·,, ·,· ,-
1.'
I

·.. ':'. ...

 .' '· ~ '
.. •.
. ....' -
. ·~ • . . I ... .' '. . .' ~... ·~
.. ·'\.:: . .l . ;. . ~ .•.

. ~

,· ...... · , , r •

·~ :·p :;. ·.; ~ ..<:·' . t1 </ )cE : •.

..1.. ... ._.,·,.... ·: ... ·.·.

' . ' ... ~ - ·.' ' . ··::
........ i
··7 .· ,··: .•:; ... . ·,,
.i \
.. ..:· :·: ·' .·:>.
J • • . •• : .•~ • • '
. . •. :. ;~ .·
~ •. ~-. ' .
.. :· /
~-. ~ • '. - • .... j ::; :;· •: ·:.- ;: 1.···

i ' •• •••
. ·.-... M c E 2
J..
+ -_:;:_
. 1.~-
,. . -.132
: ....
. .....
.. '(.

'• ! f.' :··

'' .. '
. . ..
~-. .

::·.•
' .... ;· ...

.. ,., . ' .. ·-
i .
(55)
. ,·.
,.·
.. .:. . '···

-.:--: ... · :.·.·

..
·.· ·

•• t ~ :..: . , ..
. . ·~· :' ' ,.
' ... ··''
' . ·... ; ·. '..
........ -. •t::
.·: ... •'·.··
.·.
0 ) . ::. ; ' :·.~
d't e;,.. ~l.
I<S , .: . , . ,. ,_· r
,., . .-., ~. ·J
.·.·.... \ .'• I •, . ·~· •

(57) __ ;:f
. ·. . ~ .

'~.. .

...
_..,. .·.
' .. :: __ ~_· . .-:. .--~
···:. :.· .. ·'j

' .....
•. ·~·:
. ·.... ~-

~. ~ '· • •• . . . :, !·..• • ::
' ' .
. . '.
<;· Here *
--:·: ·~~ -~

i..: . .'. ': ·. ::_>: · · '. · p J. ~s the )?er,pendicula.r mainentum the particle has when observed

i ';, ·: ~:. ·:. :_' ., ::','·:·: :~·-· ·,, :f'ram the t'rame r;rt ~eference movirig at ; , and B* is the magnetic fie~ :.~··}'"·.·. :;::-:
::. · ·..:· ..·_: ·. '·,:/i_· observed in that i"rameo It i_s give; by. B* = B(l ""E.f/B2 )1 / 2 + ~(€)o :<>·:·.. .
·,,
i, . . .. . . ··.• •'
1:', .. , ·M is actually proJ?ortional. to tl:;Le :f'l:w::· through the- circle of gyration .. · ... ,
! :·· .... ·.·.,··,,- . : ,._ . . :' :-:··· . .- r .-. ·•.
; •. •' •.• * ·-·. . ~ .

l: ::,.,/.·. _.. _. ·>: .:·

·.·. ' ;i·' . . . ~ .i ., .

as observed: in the frame of ref~renee moving at ~o tfuen is . . ' ' .. .

)/. ·:j,;~;::~:(:>::: .· •· :;',!:h"~- 0' (~), :P ~~ /~QB* _equaJ.s · l? J./~0-a .to :Lowest Qrd.er in. €1 . {

·. -~' .·.
. . _,.,_ .... de:f'ined. J?reviousJ..¥ fen: that c.a.se ..
'··
. ''·
.,·
In Equat~ons (55) "" (58) 1 7 Oscillates at the gyrofrequencye ··,,

this oscUlation aan be a:verag~d aut to give. llla.-vg

. = '7*<
2/~~·)""l/2 :· <.l=E.~.o ~:. _ ."

.:, :<)-:. '. '-: ·. ,:· Equations_ (55) and· (56). ·~a the guiding· canter eqti:at1oo~,- of ·.moti~.· .• . ._ .·
~-! >--~:i',:): ili.tbreHa-spa.ae,· 'Whil<t;(58) .is ·lila· average x-a.te- of exl1!!1rg:f inareaseo · . .: .. ,·. ·. '
; ·~ ·.,. •. . • J· ..
··t'·
. .· .. ·' .. ...· .

.
,_., -:
, ,

· 2!.:;;;:.-;;:{;-~,;;;·,:.~. iL:.'< ·
;·
... .··.. i·,
_. . ·'- ~- .
' '· .• •.'y .••
•" .·. ·
 '~- . ·'~. .' ,',· .
' -·.
.,. 3:{> ..,
.~ '.
..-.~.·
' '
.:
. . "
•. ' .' ~.

-; .''
•.;· '
' ' ..
.' ;. ~...

. , /,'.·:., _ ::_.<.>. ~Because of the denoininator 1- E L 2/B21 it. is apparent that EJ. must
. .··... ·. ·:·<:. :'
/I ,::. ·. .- ·.:
• • •' : : '/ ' . "f ·. : ~' !

:.·:..',I -. !,.1 •• • ''

·~:tf..:~ ~ ~ ..·:·.~-.
'·.·.:
·.. ' · be. less than B for the equations to be valid c. _ _· . _ _ . .:· ·.:;

Equations (.cs'J tJ. (.:){,),'and (S~) are. respec:tively_ the .ge~eraliza.tipna_ .,,.:_·; .. 1
':; ('. ·, ·.' • ,
. '. . •; './ ~.. ' . ' ~- . ' ; ...

, ':\:or the relativistic.equation~ _((/f;, Lr'a, and (.d./) to t~.(!,.·case where .·,·.. ·::o:·, ::';:_:,·~:.<'~;~
. ..... ...., ~-.. ..., ;~~:\.~_-•:

··EJ.. is 8""(1) instead of t!t(G). If E.J. is 6'( ~-- ), ~hen::so.ara·

:/>;-.;_ and E.t./B,. ~d equ~tions. (:S"'..JJ and fS'6) reduce to (~ and· (S~ upon .
. , ,• .

. .· ·,. dropping. terms. of order c 2o • .. ' ;~

•• ,.1
-~ ' : .
'·.- .. '• .: l-: ..

. '·. ·.
Equations (5".5], {Sb), and {.!if/) are respectively. the relativiat.ic;~: ;:,:_.} ..::· · ... ':;'
' I' ~ ' t :• .'

. _:.·;_
fox;ms. of. {f)) 1 ~C), and (~g'). · A comparison of (:!i5J and (l'J ) shows .... , ..
, . . ·._:·\ / .. ~. ! ·-~ ·•• ~. ' •
:. :. . ..
.:~:·.::<. ·:··that
·····
rel~t±.vistic effects_not_only modify the existing terms in {1'7),
: ~ : '.' ·-
;;.',.
·.·'···,- ....
•<
. ·.
. 'y ....
'_·.1,. ,-
~ :.·· but also in~roduc~ two new drift ~erms ~n the direction of
· ..
. ,•.
- .• _' ! r' ~- ·.· . •.
' .~
,. : ~;
·. '· '' .
:. ' :' '•A I>;~ •' ', •

,· .·-:· •·.· .
;• '•. I
·, ':··,.;1 · . .- ·:._·
.and therefore are. indeed purely relativistic effects. To. show 1;-his, .the .
.
•. • • • • • h• • • ' •••

. . ~ ~

. ill:! ·c. ~ must be ~wrj,tt~n

. . ._B. + e~
._,··
. .re1a ti_v~s
. ti c t'1.on .o f. mo.t 1.on dr :X·
. ···- · 0
.,_ '
~qua P.t .., Cit
· .. (, · ,!:· ' in 9.im.ensionless form. The scaling_ is similar to that for the non-. •' .... ··.

·. . . : ··' .· ..
~ ~

· ·; ::;: -:. · ~ ·. · .-re:la:tivistic equation.

·~.. p: t .
Let
· ·
D a{~~

·
t)/B (t),
0
i = · ~o; E(1, t),
110 0 · m ·v-
'l
: j. ,,

i'..". -· -.··
•
t

1',
'j:'

·•'
.
_·, :r r3

'··. 0
~p.'· and.a -'~~jL . . ~
""'r 1 . where; :Po i .
s the in' .i
~t a
1 O__0; ;. _- - - - :::_
moment.um. _...
2 1/2····. ;. .··

..• ~ .- f.,
-~..
.
• \ i,
• .•
;. :· (l:~ ~0 -~ -~· -~.- ~ .. ' :

......·:_.'.".: .':... ·::and

.....
the
. , ... '
other syrp.bola are the same as.non-relp.tivisti.~llyo ' lb 'tei,ns_,or :· .· .
~ensionless
'•
. ;:_:':.• :· ·. :::,?/!.these guantities,, the equat:i_on ·of motion bec_orr.es
',-;

. -~ ..: t",
. .:• . . ~ ~~-~~. '·· ·~ :-·.-

t' [1 - ( ~ . d~).'2l-l/2J
. . "• I

..-- ..
·s....
_
·-· ,··.·.·-··: .:1· .
gi .,.·
J · ·
. ,, I,

'\· -''··' -'' .. -:--· d 3' . d~ . -· m c · d~

.' . . 0 ·. -;,__ . . .
.. ·. . . .·,·.
... ~
- .
. ;j:• . ,: with the initi~ conditi<?fl• that at :r- 'o,_ if Q 1oft,_ ~~ • .Y0 fl. + ( =~· )~ -1/2 ~
· _'.... <:-_ . .': Y0 x ~est energy/initial. :total en:ergy •. The pr.obl~m no~ contain~ the two .· . --

';_:.~~-. :.-;.>; ·. · dimensionless param~t~rs. ·_-p·0 c/eB L _and.

0
p·Jmoc: .whereas :t;henon-r~lativist.io . .-:·:.
. ..- ..... '! ,.:-·~ . . . . .~ • . . ' : •.

·· . ...-.' : . :.>·~·-problem contained just the .first ons.: Th~ net<~_paramet~r. ilr/mr/~ ~s .... ~ .-·. · .··.' -~~
_.·.) ... :. :;;:_._;;:-:·::/· ·_: ... -:. . •.·· ' • .. : . ,· ,; . . . .· . '. . :_:.· . :_ - ... 'i " >. ·. : ,. . . . ' ._·.· .·· '... ..
.,': ·;·: ·:. ,", ·.·· .m'Q 'JI0 ~ofo V"o/c +'r:r( ')To~/c:3),,' An order :oi ~gnituds .compariao."'l. 9f s~: ·. ·:·:-_·· .. :. ' ·. m .

·.... :;:~>;_:~:_;\::·;,;_{:~-.~~~t·?~:,~;-~Ji~~.{:;\~;::.:j.~}·:~~\_:~;i~:~:;·. ,:.:· :. ...' · ·· . . -.. ·-·- . ·... ·. :,:\:.<:·::i~:~L·.::: ,_.--

 . ..
- '.'• ,·

- ... 31"' . I
I .- ·: .. • - .J_
I
'

-' ~{- ., 'l!J ;. .

·: • t,_,
··~
. '/ .
M ~. _M a
...

.. ···~.
:.
·'··
,.•.
',;;

.
..

.
..)'·.-,· t: . .
. •. '
:..: -~. '..!
... -·
-~ ~~.

~\ .. - ;

j.
e a dt
[:
.1. _£ 0 B(l with the term.
e
r

~.
- n
»t:i. .. ' '
'
)

.
,I·'

' ·i
'. ~' ~· .,, .. .,·:
' .· ... ;· ~·~{:·;: <::: . •• '.' ' 1 ,c.

.. '· .. a;ivea·.·.
~;·
'. ,.
~ .. '.
:';'
',• ' ,·_.·. . '. ~:
:~
... :.

. .·E
.,_
cBJI#

2
(PJm
0~ ) (8,~-:. :~
2
_. But· E/cBt equals
_,. . 2 2 .
' ..::·: order _v/c .. _·S1m1)arly {viiE11 /cY;
tb!t ;prece~ing . dri:f't te.rm.s in Eq. (55).
. ~. . ~ ,
"'· . ·. '·. ·~ ...
··.: .. ~·. It may seem strange that drifts in ·the direction of ' ...
,_'',•

.: ~ .: .
R

' ..~ ~
....
.. ,,
,·,.·
.... ' .
·. ', . . . . i•
occur., The origin of the drif't _which is proportional to v E II '-
'··: ., :
11 . ;·' . ;·~ .. ·. ~.:
-. '. ·-...
is easily understood. Because of the para.JJ.el electric fieJ.d, the ' . . ..· '. ·\'·:.·
;:- ..
~
'
~·..
•
. ..
. '.' ; : ->
' ;

: ·.' .. '.: ',.:. :.)

. :· ·: , ·' · :magnetic field is not in the same direction when "Viewed from the ':' !I

:·. ·~.
·::;··.<: ·._·: -~ ., . ', · ·
: .1, :, '.,.:-~·: '/; ~J /:. ...
•
•..
'
' I
.·•
·:~
'l•'.
• ) • ·~ ' I ' '
····,···
•,
!rallle of reference moving at ~ as when "Viewed. f':l:"om ths laboratory __ . . ·.
. -+* :,: ,.-.'' :..:_..·. :· i
,'. : '·', ·.•

.. :f'ralne. Consequently a velocity which is :pa.rallel to· 'B' (asterisk

-
. ,. .i.
. ; . :_·: ...... ~;·' '

.. :refers to U:g :f"rallle) idll have components both parallel and . :perpend.icul.ar
. .
..· · :·-_ . ~

- -+
:· \ .,. .. -: ·:~ '·;'· · to :S when· obse-r'17"ed tram. tha laboratory f:rameo. Su:p:pose that in the . · ·· ·· '.·': ··-.
-;
: '..... '... {· ; ·~ :..'
',.·;·'';I
·, ~ ~ ·~ . . ~
·>~. ·:-: . : ...... :. (.. ~··frame t~ere _is a ~~orm static field B.· and an electric field _, ...

; . '·' '~ :,., .

·' ;, : ·.- Jf :pa.re.llei t.o it, as shown in Fig., 6a.. The guiding center ~locity ., ,_. \' <. .~ ,
.,;
'''
' .: :.:,: :· .' j~ "Will consist of Y.~ only~ - When a J:;orent~ transformation:r~:J~ .. · . . - .':; ~~
,; : :·to the labor~t~ trame1 the .fialds andguid.ing cent~r 'Veloc~ty appea.x- .·; :-.<.-~::>h~~~-_;,
.>::"}:;.,:.-';:.-'as in Fig. 6b. The i and E Vectors lie ~ th~·, .Y.Z: :plane, but· .. ;--.'·":.,··:· ,·:~.<J
' ·::, , .: (: w ii does not :!» al.ong · Z and . E ia ,;ot parallel to ;:, The angle ; / :. .'i
~ ,:: _,.·!·•• ~ ... ,..~·.: .' ; _'.' • ,· • • ~ • I . "·;·: r,. ,._ . : .... • . , .. {

·::.:.J:/::<.'-:; .·:-~ ·:_;:_ ,·:·. be~en :a and tha . z_ a::ds is :;proportional. to E U ~> . As shawn1 . -~ ..__..·,: :.·.···:·::. _.: <·~
·.::;·-'··-\ ::_::.-.' :.;. _ <_:_ >.i 'san be resol-ved .into: two components, one of which is ; _:_and. : . : :;· :'<. · ,- -~-~-
•.. ··:-··,_:·:-. ·. /:~~ .. the other·is th~.drifi,:..(llEUJj)(~~(l~E}/B2). -in·~ d.ireation. of_ .. ··:···;- ..-.:?'
·:.-_,~:·:·:.~\/-~--/.E.~--~·.·:·: _ _ ·_ ··':;:,:·_~·_:: . ~·<-·.
,. :·::· .. ·.· ··.!_-·:····. :'·'". ·_,: :.:·.j','___ :~~::·'·•. ·,•_ \:·.,,
.. . . .. . . . . . - : .· . : -~-~- ~---./
·....~.·- · ht.,·~::;!~/~~·.~·. :~L-:;.:~:/·~~)}, ·:-}:i.:Ji~:.~\:~.; . :·;f\-:;:\::~ .- _:·_ X::<}.: _ · .;: . - · ... . . . .- -_.. · :;· _ ::·<<~:·. ··· ·
 .. :i ._..
~ ••~
32 - ,:·'J
.... f

• .f.
'-'}
:. ·,
. ;·: .... ~: \ .· ·. . ....:._·_,_
··;•:'. y·:. ··.......,
!/ .... :···.
· ~
~ •;• I .'

I· •''\
I
': : . '.
·'.
·....... ~ ~ ~-
,"
The origin of the last drift term in Eqc. (55) 1 which is aJ.so '·,·,'( .

'-+
in the direction·· of -::.EJ,.l is less obvious .. One can start with a. . ~· .

magnetic field having straight lines of force and a gradient ot

-
,. ,... ·. ·.·'. B
:'{, ~~; ~ :',· ~ . "
·' /

at right angles to B.. There is a. drift velocity due to \i'B.. If'

'j :· ~
·,,
.'·* :· ..
one nOW' makes a Lorentz transformation to a frame (denoted. by · . )

-
,'., '
! .~ .. ......
moving at right angles to the drift and to Bl there 'Will appear an ·

and a
em*
--...
.
Cb:-ift velocity1 when transformed to this ·(

ot* ,·''.,

:: ':' ~
new fi'ame, would equal the sum of severaJ. drifts as calculated for . ·
' ·' ~- "
the new :t'.rame f'roni Eq. (55) 1 includi:Dg the second drift term in ··:. ;,. .-~ ':
.,
. ··-::
tha direation of -*
E.J... • Making .the tra.nsforma.tion does not real.ly' · ·.,:~_·,'{:,·:
> .

,,
•:. . ·,'
·::·
' '
) . '. '' ~.. •'
~.

i./'; . _;-::
:..·.

·~ ·.. . : ~ ' ...

. . •
,,. . '~ .. :·
·.. · ... '•:.
. .-;'•.
'
. 2. . ··;· ;';.· ..
: ·:··
' '
':
'
,,. '
' .. ,·::•
,- ,. '·' ·.<.
. . •. : ,' ~· ,,···

. i ·. ' •'
··\

i.

.;.'.·:·:
.......

'1~·. :_. :. ·:
... ..· ...
'

":')

,,~, ~
.....
,8 (1·,~~)
11

., .............. :·.
·:...... ·
·':.
·,
·'
.'.'
.(
. :-' .. •'
.. :.
·.:/,·
';:'.> ·.· .... :.~·'
. ,.:., ;":. :·.:.· .·
' ....
..
.•. ·;·--/
·:·,,: ,_,:• ,: ·>,;.
.·,.'
~ •,. ," :· ' . ·.~ ... '
.~ : . .,.. · .....: . ,:·'
f.:: .. ~ ., . ;.'· I )
. "!' ... •
.. ·· .. ·',
... :.
/, '.'
F:
·-'

. l /, l ..

.. '\"> .. ·. ,· ·. \
. ·.·· ....
.... <, • ~'I ;

•'. I
~- ' ... '
.,., ~-· '

. .' r

';• . '.
...·
:·~ .

.~. ·'·
·. I
<..· . ·. ".
•.:~ !.1 <''•.' I

.-~ ~;·( )~ ::~. ~ . .·.:~.:;;~> ~~:;\,,:: ~;: .. .

.
''

,.·,
.~.I.

.
''

~
~.
/, •

::... ,•. .. ( ·.• .

: ~:.

': . .~ ·. '\: ·~; .·

 -.• ....

-;.,,,
.~.

...
""32a"" ·>·

-_.', ; ~.· .. '

'·''•. . ~.

~~- a: conrpe.rison of Eq. (56) vd.th (20) a.nd Eq.

·..
.... · •,'

'{;/·:~:·With (28) _shows no new terms, only'_ a. modif'iaation .of ax:tsting terms (J
; .
'
",'\

:· ~ ' .
_; ..· ': ,.
·,
i·;·
·... '; ~ ':.'. ... '•.·
',.·, . ~. . .'

.... "• .. :
•· ; ' ., ~ -~. :' .''
·.\,< .
I ,·~.
•·'
,'

r. ·
,/ {: . ~· '.
. \ ' .
. ', ,.~ •· .:.·
... ·
··,·
:·.·, .

.. · .
'•.·
• J ~ • •_,, ••
··.-,:
·,. . .•f..'

. ·' :!,·
.·... ;·.
·:, ·. . .. , ~
·:·.
: ~- 1-. '

.,.
. '
.;
.·. '.·. ,'

· .... ·,,
'.•
.··; · ..
.·_;

'./
...
. ~ ·,
.. •
' ~ ', . 1. .~ ....'.,
·:,· ... :··l'
..:
· .... \ ",,;
. ' ;'·.· ~: ~ ;; .. ' . .,. ',;:
.t.•
t· ..•
·,··

... , ··,:
•
...,·
• j .. ~·.
.. ·.: .... ··.·
J..
.. ,_ ....
I, • / ,:...
·.:
·.. .... '•.•·f
~ .'·. ': ' .
: '\

:, . ~ ' .. ~ .. .;
·· ..··:.
.·: .
~- .
... ; :)~~.. .......,
.·
. _:~ .
.
i ...... ·.·. • \I
(. .. •\'
_., "• :.. ·•
· ... :'/,:
..
:/· ' ~ ..
. .,. ... /
., ·':' :

... · .'1. ,··. . .·

,.. •• r • ~ :,-· ......
.,. ..
.; .
;-~::·... ~' >·. :·'

·· .. ·.
·:.:.. ·
.'. ~.
.. , .... :.. ,.,
.
. , .. '. ·.-~
·> .
·, . . ,.·.
· ...··, ·, .,
. ,.
..•... '~ "', i \. .'·

.·: ,., v ,;· ,;'·

.·.···t.':·'· ... , '. , .. ;.i'.' .···. .. ·:· . ··.·: •, ·..... ::
i.:
.~··.
·.- ·:_;. \•', ,. /,
, :. ' ,. ' ~ .

'~· .
. .... .. ... \,',"
-:· r·.
·:' ...
.... .<.•
'··'
,·;
. ·-
·;. . ." .
;: ! .. ·'. ··;
·
~.

•:
~ :.I ... ·.·
·'\ ~-···~;~~:· .·.,·

.•' ·, .

.·...:···
''l·
..·· .:·'
·.-...·;'.'
.... ·.
·
.. .,i,·
,;

···.·· ... ··,·

··...:. ......
;

' ~
' ~
I .'
.... '··
,...., ,•

.: ···~
·;: .. .:·· •. - '1'. .•
..
.· ·.·
.' ... .. ~

. .,,
. ·:· :.·. . · ·. ;'.',,: ·;:t"· \_:· .:· '•,· ,··: . . • . •, • i.<! '· '··-" .. ,_...:,, __;:J_· \ ' _,:.:":;:~;;LL2<: ~·~: ::,. :'-~·. ·- --'-~·- .-,-,
't,J.,,... -,
~~-
·.·.
_. l ...
,•,t'
•' .,;., ·" .. ·..
.' ' ' -~·

·;,_·
. ,. ,.. ~ .·.
·.,'.
33 .· ... . .., ·· ...
;
I'
,j' .: :-~ ' .': ~·- ··}
III. A HOP.E FORNAL DEHIVATION OF THE NON-RElATIVISTIC.
1'. '/ .
. GUIDING CENTER . EQUATION

The derivation of the. guiding center eq~ations. of motio!l for.

·>_non..:relativistic particles. presented in .Sect.·!£ · requ_:l,r_es rigorous ;

~ . ' .

_:.>ju~t~fi_ca.~ion •. The work of Krl_lskal ( ~- >. and· of B~rkowitz mdt;Gsrdner(/J'~· . ···

:.·_.prorldes
~ 0: . . . •
the. justification
•
of Eq. (-'/ ). Kruskal. ·: der~ve;J equations for .. '
.:. ··... : .. ,
·the.;:1n: apppearing in . a serie.s of the form
-~ .. . ''

~.,. ~-
~ .
6 lnJ Rn.(tMrin
~'Ec
JB(R ) dt)
0. ·
...00. . '
··. ·.··:I
. <<.:t:;
; _~by eq~a~ing~ c~effic_ients .o~ ~qual pow~:r~ ~f-~~c.fo· dt) after_su~stituting I ' ~ . '
'~ • :4
; • ~I.

the series. into _Eq. ( / ). The fields are. expanded in Taylor. aerie~ abou~ . · : .. ;.·

Each a·. is its~lf~

a power series in.e ,·so that . ~(t)
0
(t).:+Ea (t)._.'+.&:·•. /.,
1
=-R
,.·:r~ . ,n
. . 'I~. • . .'. tJ.4t- .
. n . . . . ·_.;·n·.·,_'···-.:;'. '
:~·-.<., '. R must equal the complex conjugate of R 1 :in tzi:de:i7 1'. .be real. . J;t is · · · · ·
··-n· · · · n 1\ · · .

. ~not immediately C?bvious that equating the coe.ffi~ient .of ,each. eine. (~~~~.::_...,
'·<:>:. :;- .-:> 0 ""jw<ff'0 )
·r {:' . ·• ~ , b
dt) to zero .is justified; the. ~ . are functions of
4-.
t~me_, so .. ·.
·:.>:.·_.:_·that the se~ies ,is not simplY, a Fourier series~ HmO\)'ever,~~ Berk.ow:i,.tz and ·
... ";- ;· .
Gardner. prqve .-that. the ~eries. obtaineq by this pr·o.ces.s ia actuallY: ~-c . ::'
;• ... ·. '- ....
.·> -\

•
..: ..
,, ,.

-~ "'. •I . • •
i

", .
asympt~tic
.
expansion of
·· .. r for· small £ ·.,
·'
Their proof is ! . .• '-
ne.cessa~ily.-.

' .. .:.:·,t·.·.·. a formal mathematical. one .and will not be repeated her~$· But it is
.';'

· '•:: · · <'l·rorthwhile to .dis~u~s -wpat .is .usually mean~ by an a~ymP..totic expansion .. ·• .

~ ~
...
• •. . •• ' .. ., ' • •• • 4 • • • ' • • • • • ' :. •

.; '

an asymptotic eJ..1'a.IlSion;·
',• . ~

(';~~:i":,::; and what i'j:,jS ~jthey proved.. The USUal definition of

.
·;;_.··::::~"<:to _b~
·' ..
found in 't\'hi~~aker: and'L-Watson (~ :~or e~ple, _can be stated. as ,. .
. •,
~ ~ _.., .
;. t~: I

. 2 . ·-:.··
:-;.:_.'..';. .follows: given the po'!'ler. serles S(e ) • , A _+·A~ E +,A~~ +. o o • It ·.. .' :.-'··::::.
0 . . :. ¥: .
.- . ~ : ... :
~~

. is called the asymptotic expansion. in. c of ·,;, :furi.ction .f( (; ) i.f . .

r('E=) ... sn(6) I · ·· · ·. . . . . . . ·.. f
. . .!: . .: · ·>·:,.:·.:·:··.-.. J. .•~:·.·· ~,
....... .-•'. ·~"~O . ~ n_ I is .zero,, wher~ ·Sn(_ ~). is the ~um qf~ -~ :'t_l~·.·,:>:.O:.·.~:i.:;··:_._:·_.····::..
c

:·:,-.,' ·:~,;_:_te:rns:(i.,e.,_includin~ -~C.n) of.the ser.iea.• ·.·Or by-the definit~on-,o~_a, ·. · . ··.,:·· .. ·.;

·. ' ; : , ~t, :fo; ·~veey ~er Q. there .is a n]llnb•X: G~(Q) . such. that <> ,; :,}. C; :. ' ·. ,, . .....: /f
'<):..·:··-·_:':·.·_:_.··.-·.-·.-::,:.~- .~-;:_:_:_·. ._··. ·: ,::: . ......
• '"
. r,
. ' ,. < • ~
....~$' .··.-:~_::;·. ;. '''
• '
:·.... ··~·.; ······ '• ., . . ·.. :, . ·:, ··\,.:'. . '.·
o •f • , '· , .~
0

>
.·::::
~' ,'
0, " • . ," • ·:' • O : I '• • 1 ' ;'_. ; .. , ; • :.-. . . . ,
0
0
0
 •','

"'{.
. ..
~ 'J:

': >_.'! :··. . .. ~ (

"·i. .; ·: . '•
. . .;'
. .
. . ·. ·. . .·:~ .··.·-.-~ _; :·.' ~
~

. ...
.
:., .~. _;·_~· ·.... . "' ··. ~.: . -~.' :. .
....
I •.\

'' ,'_ ~:. i; ·. ' I

34
:\.
' ···.,'

·{He take 6 as po~it~~-~ only). .; . '! : :, ... • -~

..
.'

_:;:/) ::..
>. :,. . :~::.·:this is true, one can Dhow: that giv,~n s value of,~ , say E- 1 , a numbe~
.. _;_

... ·~·-·.:_A( l!:l_) can ~e found(/S). such that fr.{€)- Sn~G >/~A(6 ) 6 n+l ·for.
1 .. ... • • l .·,

···.:,.
~ '-. E-1 ., precis~ statemen~. _t-hat· . / f( e.) ... · Sn(~·.. )/.
.
i. 0 '- This is the meaning of the

·· i.s · (} ( t7: n+l) •. N~w ~he series .in Eq. _((p/) is not a simple power aeries .. -···
. . . .
\' •..··. ·.: .. :: •, '. ·.4.~:.-· ·. ·. -~ ~ ·.. ~- .. ;
.. ·· . in G a Each ~erm is _itself an_ infinit~ a·eries in e , .multiplied by- ' . , ··'·· ..

(:' /. ···i>J:~ dt; . - this s~ri:. in/': . i r2 ::: ; . '

expone(ltial ;is not expanAilil.e in a power
: .. Ncvertheless,_Berkowitz and Gardne.r pr?Vf!.' that i f 2:n .is the sum of terms - :·,.·. ·. -~ -

>.~. ><·-,._ -b~trreen and· i~cluding (/h) 1n and U-n 1 then the.re is an. A( 6l) :·_ . :.·, :· . :

,·;·_;.:; .:·. . fo.r which /t_<~ ,t) -~nl .LA( 6

1
) c n-l- if; e-'~lo Th~refor:_e _I~- z;J=" ~ G-n~l)_· ': .:,:':_-·-..-,
,:·. ,·and. in this ser:~~ :~ is the asymptotic_ series for to It is also true ·_. :· :'::.•;.::; _,
1
~~..., 0 ~-n~~ . .'a· o. ~qual follow~ ? :- ~ = 0 {&)G .... ·..-·-~.:
1
_·'.· :: ·:_:::·.. .that \'lith n to 2, .it that_
. , . :_'But ~2 ... Ro:+ e(Rj_ .ei6-· + R_l e-i9)- + ~2 (R2 e2_i9 + -R-2 e-2i9), _so that ·... ,,:_ \: ::?
.•f>~ ... ·. ;. ~2 ~ Ro ... ~(_6 ) •. By. subtractio!}, rr- z;> t·( ~ ~- Ro)-,.. r.-.Ro .... _. :~c~}~.---.(_··:r_::::.::•:'
. \-.:~.-··?--This means that the differenc0 between the actual pa:r.ticl~ position and ;···,·.:-:,: . ,, :~.. :
'~
.
·~·:/~:' ·:'.-tt
. . ..~~:-.· ":·~··"··~-~~' ·.

: ) {, : Ro . is.
. . . '
of first order ·~; the radius 9f gyration, and therefore lfo is ...·· :,rh \(,
...

,; : ·:.·;';::·'·:':··.·_a .suitable definitiol} ( 4 tl)e guiding cent-er _position. '- .-, · ,:''>:~ :·."_·;.-

Foilmdng Krus~al»
. . we will
.
nbw
.
.derive .equations
..
'(some algebraic, ,·.
~--
.
/.-><.·.; .::,:,:·:•.~· s~me ~fferential) f~r the . ~· It will be' found, that the equat~on:' ~or
: . -'!:> .. ' . .
; . ., ... - R0 , the guiding. ~emter p~si tipn.P, ~s ·equation ( 11). Substituting Eq •
.·.. · .. .· • . .. :in9'
, :_: .. _ into Eq .. _( J)
and ~ollecting coefficients or: e ... gives for
! .. ·:'_. .. ·· . '., .... _. '-• ..
\~ ....

....
·'·'. t
' .···
 .·· . •' . ··.:·· .. ,
. ·' ...... ~' . ....'• f ~~ .
·.: .·•, :I
•"··-".
•• '1 ... .: .. ·· . .· ..
·
::'
·.. ;·;·
·~

.··;·... ·. :· .... ·
. ·.· :.. : ;'
,_ ... ·,
i..
. ,1,
. ~. ;. '•';

,·· ....
. :• ... ·, .., ' .
•I··
·,.:. ..
,• r ',,
·'
.' ~. ·~. ,:~~
....
. ......

.'' .. '
. ·:
.. ·•'. ·... ·
i .···.·,
i·". ...
·.··., ... .., ,· ,.: . ,.·.·.
..l.
·... ·'·
.. : ' ··' ~
···:: .·_'';l
...
... :.. ·.
. .,, \ ~. ·:<'1 ..
.. : . :.~ .:··.{

':·. J~
. ':· ... .;-.
. .~· ...~
. ·......
:.. ·.·.· ..·.:. :<·~~~
·. •.. ' ~ . .:: ·,., ~t

a
~­
G ·tl ·.(~_,.·· ... ·;·:··:···;~
:.,.· n ' ·~ ,. I ~ , ', ', ~ '• • , : ' ::~
.• ' . ':' " ' :. •' &
. Y.ihere~ .. · ~ ; ..

..
I '• ' :' '1
. ,' .'· ., •.. ·:: ·.

'.: .
: ,. :.:/ler--f,;a
. '
velocity 9f light. c
~ ;>~·-:·:-.: ::··.~
will be taken .= L. It can be.re-introduced '.· :··
... ~-(
.,;-·.·.~·~·'·'f
.
;.·,,·
··~

· \ _. : :·:.: ·, by dividing
': ~· . _'. I ·,
B. and B .by. c. l In all ·four equations. the fields ~e .·
..
~~ ~:)• ~ .. \

·::_(·<.i<,·. ·evaluated at ·10_· and . to In Eqo (65) the lowest. ·ord~r· \,_denoted· by:
.·,·.·::·· ~ .... . . '

./ ·,:·, .··.
1
. • ~ 0 ,. w:il~ ?-:l,,~eya_ con~ain. only' .R.p.O (1trhere ;1. ~ p~ n. -:-· 1) •. · For .example,,
:'~ · _·' ... ··:~·· .. from Eq. (/;'/), "G20 = ia(R09 t) -IS_~ x. {a _o 17
10
)B_ (R'0 , t). In general
, ...

:. _:·l:: r::· Gnl. 'frill contain' Rn~ I~· is ~hus po.ssi~le to .soh•:e ~gel;l;r:aic~llY fo~·· ,.• ,,·

:-,::. <:\::.: ··
' . •
·~nO in terms 'pf the .~ 0 ~_-·R'20 ,.•••R'~_1 :.d~
. ·. :,.. --1!> .
· To•
·a.o,lv:~,.Eq$ (d.$) for Rn.0 ,' ...
·,' .· . ·, .' . ·'. :. .

. ... . -. .. . take i:ts 53ca.Jar product with' ~ i' B/B · .t~ · get :: · · ; :,- ·. _~ ~-·.
cr ~ . . .... · ....
..... · . ..' ..
. '~ ·. ' ,;·: ..":
,
. .. ~ ' '
. . ., aolt
.J!O ". el ... .·. ... ..
. ; ·.........
·.·. R .. .'-'-·
e1 ·.· .. '•
. ' .. ;
.. ·,
.;. I• ··~

,__ '
··<.nO Cl ·.-·
.·B2 n2 : .. · :· ,'•:
· ... i':. '·.
.. ~ ,·..
' · . . : 'r-~

. ··.. .. .
• ~ !
', . ~ ·,.
.. ' .·.·· .. •'
.. · . !
,, ... •: .·;
·r:. ! '
t(,;•

·: .·:,,-.. _.'"·.'; .. A ... .. .. , . .

. . :.. .· :: and ita· cross P:roduct with e
~ ;

' ..
., . •'
~
"
' ..
•'

1 •'
.:
'··
·' .. ·.... : ·.·-: ....
.
~ •'
~ . . .. .. ·,
'

'·
\..
.. 'l
:.•
'·/ .
' .. ·.
. ~·

~ -'\ J! . ("" _...,R • ~ ~R : .) ·''., i ,·

.. • ' 'J,'~

· 11 ~n0 ~ el. + m! el nO · ·"'l - nO . "" '.·.

~··.- ·~.;: (:
.. / ...
,·, ....
•...
".:-· .
·.. ~
. ·'
'.·.· ~· ~ r , '-:~.".'•. ·• , ., •

... ,
~'
.,
{
:.
,;•, • I ~d .:"R~~.~:~·.~ ~ <n%0 _+_ ~n~- ·>rt:···. -·r>:
.::: . .·
.o ,.,, ... ,
. . , ·.·. ·.•...... :. · .... ·. ·B ..'I .·,, '.'

. : .. ~
·; ..
.... '; ·. ,:
~ 1'
..... .,.· ,• .
.. •·· ...· ::
'•

., '\ ,; ·'
'· • j :- ·..:..: • ' • .~." ', . . .'? ;

?-'~'.' _{>.:~i:~;Li~.~·;~::; ;i LLL<::L3J~_;-~~j;'}:.·:;:. /~, ~:;~~):.:·:;:,:~;~,~f:.;./·. . ·~

 .!· ; : ·.:.

.....
I ·.. ··· :~

- 36, -
•• : •.1· . ::
. /; . ''. ~
·- ..
·:-::··:'::·:~~:·.. :\· from (651 fnto Eq.· "7); the resulting equation can be solved for
.....tp
Rn 0: ·
· .. .:· .,_,
..... '
·=--

I '·
. '·.·

· · : _.",•,_.
•

' ·.
' .
I . ' •. 2 ~· ;o'\ ~ '\, ' ' ~ " . .. . '
~ ~n__·_G.,::;n~O,_+~.(~e-=1~·-G~n~o.>_e...:l::::o....+_·_in_.G..:.n:.:::O:o..·x_.c.. ::::.l
. '; <;
. ..
. nO B2. ·n2 (n2 _ l) r
I
-..
\,.
·- :··,. ... ~ ·-.~~- ':.

.<f> ' where the q~lda and ~l are evalua~ecj. a~ (~, ~) as usual, , ,; /; : 'ill
"'.; I."l order to fi~d ·R~,.· equa.tio~ ~ is writ~en. to next. .order.as ·< :;_· .',:,.· :,_.·.i?.
,,' .
. ·. ~ .··.'

·:....:.n2B ~ - inB.'ir x a = 'd , ·· · "?'J ....., . __.

... · n1 . n1 n1 · . . -~
•, . ....
' ·. -.~
·:~

.... ,.:Y:/:::., where Gnl ~ wil~ contain ~P (p· ~ n-1) ..and Rno• . Thus. there exists a . 11 ·.!~\
. _·;>;·,' .. · recl:U'sion.scl!em~ for· n2 . and higher.., pro~ided ~O. and R;, are known; so ·.. ·. . . · .:~
. . . , - ._.. - . --··--....- - ./1/ot . ;_,
.... ·: ··. " ' .... that the r_ecursion schem~ can ~e i:ffi.tiated •. J~.Jf ~~ations ({,~) and (63) ... · ·.·:·. JJc: .e ·. · :~
,: . determine to·. ,and ·"R:t ~ .The · ~ero order. equ~~~Oll (b.3 is ., . .;: . ~~G:'t""Gph · : \
~.. ,i . • . • '' '.::·_, .;j .·.
' _.,!
. ~ >~
"<
·• ,~: ~
. ;- ~~
·:.' ·~·""_1,
' . -'il ...... ""' ~ .. _··:: _,·_ .._·.;; ··:'
Dotting '4-th ~1 (R0 )_· gives l1.o . e1 = o, so that R10 if! a. v~ctor· ...:: .· · :;;~(
' ~ . •· .· "" ::1
· :·' · · . . perp~J10.:,.cular. t? ..the._magnct~c .fiel~.. Therefore l~t Rio equal (a.+. ib)_~-.~. (c t .id.)e3 ,_:;

:,;: ...: ···::::'.:· whe~~ .'6'2<Ifo> .. an~ ~3 (~)·. _are .P~:pendicular.to .each othe~-~d.to.~1 (R0 ). ·;: .· . :~
;:. ·.' :,_ :~:: .·:
.:... ,.·:·,··';~~ -'~ ·a.~ previously,_ and a and 'b.are real,.. Substitution ~tq }4. l>'~) givee
' ..:·,.... ·:··... :·.·. ·. .. . ' . .•,.

_;:·:.\.';· (';· .· c ....:;;..b · and a ad~ .As a result_ ·.rt;_0 . 1-s of the form .'··
.. "! ·; '. · ~ :. ~- i ' '·: .~ • •: .
' ·:.·:. ' '· --'9 ' ,_. .
·. · : :. : : ' .·. ··. 11.o "'"__.(a. +· .ib)(~ + ·iC))~ ''

~ . ,. '
·:
'. ' ~
. '
'

·.:
.'
... ·
•.'. j.
. . If . Eq. rJJ) is now a~bst:Ltu~eq.
.
into .. the. square ' '
brac;k~ts.in·Eqo
.
{b2, -:~
.. .~.

_·.' -" the result is·· f •• ~ :-

. ; .
.· '
• •• · , o"! • •• .,.

; ·.:
. •.·
.... ·· ·' ......

. ·.\.
 ·,
,. '
·.: ~· ·. ]' .. · ... ·, ,·r,, \
' " ·,·
;,': 37 '!
'
. ·•·
.. -~- ·.

:.::·' · />' ·The coefficient B-;;_0. • "!S_ ~ will n'?w.~e identified .as the ma.gnetic,mom~mt 0
' ' ' *
· :· by, ditf~~entiating . Eq. (6)) with respect to. tirne.. Since ~ .,. R-n. -·.:;
; ;_.-.. \':
', · ·: d!it.terentiation· yi,elds. · ·. ' · ·.·.· ·. ·;: ' ····
•• ~ .: . ;i. ~-."' • .. . , r . .. .
· ...
···_,.

. .·.:,.·;:

- ... ,.
''

:,. ' ..,

*·
' '

'.· ... ~ ""'P iG. ...- -ie . ·. · . .

~

·-.-~.:·· ·..( v . ~ "" iB(~;tO· e . - ~0 ... e. \ ) + 17 (6 .J,

'

n0 J. . '',

!·· .is· perpendicular

. . to B by Eq. <?J) . · Squaring. (llJ)· gives·

-
~6. "'
. ~
rs_0. ""' ·0~ .However,
.
(v-:" RQ) .L. . is .the. per.pendicular or gyration· .. :· · ..
. ~ ' .

'·:.···

:. :ve}-oqit.y _in the !ra,me ,of re.ference ·moving at tho gt1iding center ve:,locity; .·
~ _e 2 f.N2
:"'
;:, . ....., ; . . (02w M. . .~ . . . • .1r·'!.·.

Therefore ~. Bl1_o· o. RlO =-· • «-2B ew . ·

"" ~ = # e '
•, :'.
'·,·
... ~
.
'·.·.:·::'
.•• ~

·:·,-::•.
r. •

qa) beco~es
•
...&>J
.=; .. ~ oJQ
-":(..,. ..)·- .. ~2w [""" ("" . ~·)~B . . .,
R
(s:-2:, . . )-B]
C: R E(R 1 t). +. ....Q. :x B R01 .~ ~-
20 · e 2 _x e3 "'· v_ ..• - e3. ~. . . •.'!7.
ee
0
.••• • ,I
0 . .e.·

I:'
'(}t'J . . ..
~ ' \
~ ·~. :'

· · \~.which is the same as' equation (f') for thus.· jU;s~ified ~he ll,Veragirlg:'(. :. ,· ··::'.:.
. .
' ' . ,:: :' . ··.' '. : .· .··j'(>. ··:' ...•~f.:.
.:··.- :.•.
'
pr~cess u~ed !-o .get ~· (l.f) in Secto ,'Zl: •· 1' '.. • • • ':: '... • • '. ..~
'•)
. i': ' '· ~.•..
··. :. . ,':. ~· . ,',I
In the next Section, equation
• •
(b) ~ill be use~ tq ne.xt. Jligher , ·
··d . • ' ··.
.. 1 ,; :;:· ..
:'• • . ',

.··.order ~o ~h'*{th<,it . ~· .<B~0 .··\0*) =:·-~·-+.~(E_), and.thia1.dll~e a·.·:o~.-·.)':>·~·:><'·

':· proof of the a.di~(b~t.ic.. inv:ariap_ce. of. th~ ~rtl{tgnet:J,.c ·mOJ?~nt ,lin !:-he~ g¢~al:: ·_. ·· ,.-,".'.. , .·

.~.:·: ., <ca~~<~~-~~ -~~.·.an~··.·~

• • ~ ' .• • • • • •• ~ •

. are,.~~ ~~~)e: ·
• • • •

.
( ' •. ' .•

.
• • •.

.. "
~ ••• ~ • . • • ,!, !,' '-:"

····.
•
. ~.. :

> '.',. •i::· :;"· :\; ·:· ;t>··~ :::.>·' .: ':. \/ -: ' ·....., . ''
>"
:~.... .:.· ·, .. '·. :··. . ·''. ( . :- ..
·· :..
-~~-' .·. -:.:--~ ::.· ' .
, : ..·. .
,:. '• :.

·.· . ·..
·. . . . . ~
'
..... ·
··::-·
.... ... '·
-~ . . . ',•

-!
',.'
•. •

' .... , ~ ~ .' .

,., • :1 :: ... :

··:·;. .·,_.~_, ·.'/.1;:_. _:,·:,.·:·:;.-.·. -·• ·.. t -~-~--- .;_. ;

:.· ' ' ... . .. ·-. '~-. •
•r ·
.....' '··.·. :· ..

~,.:;_~-:.:i.~~2A-S.J ~~~~~,; _ :; i~.~,:~j;,..i/L·~r;:.~ -~.~--~~·-,~j:····~ .:~.}:i::·_.;_~:··..·-0~L~d.-·,-i. '-~~.-

 ' ...... ' , ~:
'·, ..,·
·.. .·.'
\ .
.. .
• t
-~ . ' ....
• }
,.
. ' :. ' :·~ . .... ,. . ·.·- ..
:~. :. : ·.. . . . . ·.r ·.i ·
38 ' ;

'· ··. ... ·

•\!·.
•''

• '·i
i'
. ~ ....
. .'"') ' ·*:. ;';
'•' .., .

:'-t>.: .
··,·: .. IV The Adiabatic Invariants of the Motion
• i• :_"_ ....... ' ·.~ : •
. .;, ~ . -~.. ·.

[,' ·:<~::~:_.··.·,:_.,:: :·: .~· :.. SJ.rteo thEI sol~tion o! tho equation o.f rnotion has been obtained
· --.. · · ~ · · ·:. ,. e Tt'- · . .
J,:·:.:.J;. ·i;.:<_·.. :,'as an asymptotic aeries inA the previous section, it is reasonable that .·

:·. ':· ·,· . ' '<··' '·.,·any approJd...ma.te constants (adiabatic invariants) .of the particle motion)~:·.:'.·.
·:·_. ~-., .·

:
"! . ... ~·:·~ ..·-:·:~;_~.· ·~ .' ~ .. '
~·

·, . . : ' · .... should be obtainable as asymptotic series in G · • In analytical dynamics: ;;··

' ' '

·' · . exact constants of the Notion are usually obtained by the canonical
~ ..... i:t,. ~. ) . . .
. . .. . :/ 'formulation: . if the Hamiltonian is independent of a given coor~ate, ·
,.
.. _: :: ~.
~: _: ·.: ~.. ·:·...; ' . ~-:;'-. ::(
. '
:··.; .: .... : .
< • :· ~ ' •, ,; •• •• : •• ' • • • '
::.
:. ...:.<<.: .',.
•,. . .·: · the conjugate momentum is an invariant. By analogy, we expect that an .:·.::·.
.. . •·
~••.• ,:.: ·)· ... ' • ,, .•• t ... ': ··.. -~.
· :.· ·:::_expansion. of the Hamiltonian· in an asymptotic series in
I ~f
e should reveal. : <.:
, I , ~- ' . ,

··.:._:·.the adiabatic .invariants,· provided that at each s~ of the expansion,~ _::.: ·.: .. ·>
.~ ' .. ..
~: ~::- ' . .

.·'. variables can be found which make::~ the Hamiltonian independent of one of,:,'~·. : ..;:· i:): ·. :· ·.
. . ; .
~~i, the coordinates. ·(The last term Hn . in the expansion of H as .... " ,•
,···.

~-~: :· .:· .. ~· ~ :·.

;~i· . : ·/;:: :.:-H0 + ~ H1 . ~ ••• 6~n may contain the coordinate). · The systematic
.' _. .~~ ..
·{~·\ pro~edure for finding the proper variables has been given for the
'·? ·.

· .,···~;~on-~elativistitc case by Ga.rdner.Vfi ,

i'~
·.:.:~;-~J:: · It is ~imilarly true that to find exact invar:ianb.. the Hamiltonian·:.?~:: ..;;.;·:.·
' ; :. ·~; ~ .'. . .. . .
'·. :·;.·:must be expressed :ln:b:u-ms of. the proper variables~· As an exa.m.pla, if ~/::·. ->::'(,:
'·,; " /: ~ .. . ··
··,·,

charg~d particle in a magnetic field having

.• < .
i;· the Hamiltonian for a. .' ,. .. . ~

i ', :<· . . . .:·. . . ,., · .· .· :- ' ·.

.1:···,··.·,.,:·:··.:... ··' ,azimuth~ symrn~try.is written._in r~c~angular coor~in~~.es:.: it is not at·;:·:.·.: . •::'.::·:_..· .....·-'

';., . :. · ,. , _, all evident that' the canonical momenttm,. . P "". mr 2. G+'! r Ae. is an ·; :.~·~,·. · ·.
8
y: :.:'::<.: -. " ,,
:j,'... , ... · ·· , ·)'exact invariant of the motiono It only_ becomes appare.nt when H is · . : ·' '···: ··.: :'·· _r>
...
f. ;... . . . . .)...
.
. ' . ,-
·'." .. :··_,·

':' ... :. ... .· · written in cylindrical coordinates. . .. .. . ·

t' ' ' ~~ ;, • .,; r.': .'. ~ ~; '
J :·,··, :· :_:. ·:; ·: ': : :
l ', : ,

There is an even more general theory of asymptotic aolut ions · ·; · :-:. >. - . i,
1
···::.:_'·_.::~·:·>
f .: :and adia'Qatic.invariants of coupled first ord~r <l:if!erentialequations · ·,:·.::·_):

. .\ ·,.::·:_.:::, :··(~,~- '·. :·of a. certain type. This is dua to Kruskalo (/1}. The eq~ation of motion·".:,.;:.< /;>>.
:::i:·
I i ·, _i: ' '. . ': .t ~ ·:. • • ·' • :· ': . . ' L.

ri . }· . •. -· .:~.W:: ~:::: :art~::ed~a~·~:::~a::. gene:u· ~~ey :~ • . .. ••_.. · ...

:r .·.···.' •,:··:.;;:·>:·.::··,.,. · · .. ,, : .. ·c.·.: .... :... ;'>,;·;.·'; .. : <.: ··; .. :.,:·. •.. :··,,,. ·
<~ , · .' .·, t,"c., ."! ··•. " ) :-. ·:. . ;_._ ....._ ~.: ..,··.·,, .. ~~~: i, 1.·.:.:.: ·.·. · . ;~~::·.~;· ·. . ·) ..

· t···. ' i£&:~L~-~r::~JE; ·1.~~: ,,•· · . •:.;,,2:.~~;~ i;~dlc;;~i::;;~;i~·4;i.i,,;';,i.,~::i.; -~{ ,-,:~:1b,i:; ;} 2•-J,.~:·~;i..~' ~
 ·- .. · -,'} .·• •! t '!_
:··' .·-·· /: .'
. ,·.' . '.' . '.; ..... ~~
.39 .. ,
• t
i
f • ·1
. "· :,

·Although th~ adiabatic inv~iants are asymptotic series

·_; i·:·:,:j
':.·._fo:rm:constant_ ""A +€-Al
0
+ ca;...~ + o••, it is customary to apeako£ ·,,
• ; "' ' ' ' • 1 1' : • f·~ , :.' .

'the. lowest order invariant A as "the" adiabatic irivariant. For 'the': · ·

0
_.,:_;,:.charged particle there arc. as many as three such series 1 one for the
~:.-~:· · ~ r · ·

. ,. ··.':: ,'magnetic moment, one for the longitudinal invariant,· and one for the
·,.-_. ;'_.. ': - . . ·, .

.·.->./"flux~'~ invariant~ 'J:!lese three series ':'ill be designated by M +c'Mo + • ~ ~-~;':_,(:.:·:.: :_~

, .... ·.···
.';of the lowest orders. _M, J, and i will be given below. . ,··

The number of adiabatic invariants is less· than or equal

·. .. ' ..
.number of degrees. of. freedom of the sys~em!' The charged }:9-rti.cle-! .which:_-~;_:;-':,:·:_-:· '• <.\j
~· ~,; ·. ~

.f,.
• ;; • 1 .• • • ~ ' • ... 1 f I
.•..• ·, ·. ·••. '.1:
·.has three degrees _of freedom» may or inay not_ hav~ · M» J ~ and ... · ..
•
···.-. .
;,,··.
• J

·....... c: ··(, -~- ..-~~~~;'

· . depending on th'e fiel_d geometrl• The number of adi.abatic invariants is ·'. -~ ·. .-:. ~ .'>if
.: determined by the number ot periodicities~ · To illua\rate, suppose. that
~ ·. . . ..
;: ..·
. -: ,,
:__B ·
.
is nowhere
. . .
large enough to reflect the particleo The partl. cle motion

·::,,'·:.; ... is nearly periodic because of the gyration about the. line ot for.ce, but
:.' . ", .. there is no semblance of periodic'ity in the motion along :the line of

:·force. There wouid be only ~ne adiabatic invariant series, the one. for

:. ·'the magnetic moment, even th~ugh there are three degrees of freedom.
.. ' ' . . .
now the field is, such that- a particle is ahrays ·trapped and oscillating
. . ' . .
If

..
' · ::.'' · .. · .,. between two mirror a, there will be a· second or longitudinal invariant
·: ..<.'~;~ / r
i:.

J ... ·:._ · . \ · :- .
, .,
>
: ·;, .-,} ~. ~·->.Y~. Finally, i f the drift from .line to l~e as the· particl~ .oscillates between·:·: .:.· ... ·
.· ·. · ::: ·· · mirrors with constant M and J' c_arrfes the particle ·repeatedly around · .

... ··.
<;·_-<-~._-'a
.. '
closed surfa.ce.,· .there is ·a third periodicity associated with the motion·:~. ·
. .

·.
~-
I.·· .... ·

· .....
•... :
.

. . , · · and there 'Will exist aA adiabatic inv~riarit

. • "'t./A{en:4· · ·
t....
The motion of charged

·· particles tmieh compris~ the Vari Allen radiation

. . . possesa·es
. '
all.. .three
.
. .

...: .' :· .. ·. l>·eriOdlin~t.. ; and ~~ariantso {~~: The perio~cit~ee are:r .ih·e.. · .... ·_·, :'':i; _ ,_, :·. ·<:·:d
,.<. · ;' :,.: :_·,. ~a~~on about the geomagneti~ ti~l~ lliie~~ :t~e ·.north..;south· _oscillation~·: ··' .. · . .'·:· _.~::_;<~.
'· ·, :·· . ·: ·.. :. ...........· ,·.; ·;;'. :·;;.<-.,:·· -·>>>.::.\ .·;: .:·,.... ·:·.:.;·:·.;
'·, .. . ·,: :·a:nd· th~ preceasion.: about t~e·ear:tho::._.:i , , ... ·· ..•. '
.:·_;-~:~L"~~~--;~_.:.·,J&tLL.·}~.;.-.:-:·.Lii;:L~.n~.:::.-,·L~:,,:.~_·,~---·).::;ji~L~.:2::2J:~·{'L.·~···. ····-~~- ·...~·. · -·
 ·'
' ·~·;:' ;-;. . . -~ ' . ... '.
' - 40
···,.
'. .·,. li
?~ i, ....... ' .. '
.:.' ...• ·. ' :. ~ .- .;· '
''

.. - ' . .. ... :·· ..> ,; .

···:. ~ ·_.. ·-(:·A.. T'ne Hagnetic Moment '•

~t... . . '. ~ . ~ : .- \ '. :. ;·. • ..

. •. .. '"
,,
.\

*· •• -~ ' '_·,·
. :·-·~ ··: .. ,··'
I~.

The invarianca of the· magnetic moment will now be pr·oven yia.

(t3) and Maxwell's equations. It will be shOWl) tha. t

_,..
Bl1_0 •
-'\>
11_0 . ,
*:
I·~,;
\,'.
which is proportional to the magnetic moment, is iul.ependent of time :i
'··::

·.to lowest order in €. By way of clarification it should be emphasized ·· ···._ · · .

u v ~~~ .
m J.,.(9·--· - .. . ·---~--· SvPcrsr-
. . :_..,·.. .
r.-p-1-
<that the _magnetic moment is not always 2B 1 where v ..L is the ·.-. ~- ; .~ v.La_ ~.- ".
·. ~. -~ .
.. perpendicular velocity in the· laboratory frame of reference.. If there ·'c:_-_:... ::~_:: as._(:'., .
·~ ., . """">
m(v-iro>i ·-·:~:,: :;;:: " .. ·.
··:_,
... · is an electric field EJ.. , then the magnetic moment ia
2B
'. ::,, '·. ·;'-,:~ . where (V' ~ ]0 ~ equals In short, the 11 perpendicular .,
•'.·
....
.
..... . . ·· .
.........
. : .. '
.·
:·~·:·.,::velocity" must be that observed in the ·frame of reference moving at·. ·· .• ;~

i,' .. ' :',,•.'

·, '. '· ... ~-

.,. .
~~
can
be verified from Eq. (75j. -It does not matter whether
This
:. 'I " • ' '

'
'.·.. the component of v
~- perpendicular :to . B(r) . or. to B(RQ) is
I ;.·•
~ .. ~--- '. ,..
.. _:used_, the difference being of c;r( 6) and thE)~efore appearing in the _, :>/ ~-. ~. : ..·r --~-
->:~-:~ '' . . :.- :' ! ·• • .

.· '.: · ilS M1,•:. term· of the' magnetic moment series.

o; ~' ,•

--~- ····.·
' . . ,. ' Equation (t3) is of the form . ' :-_ :: '' :·~ ::; ': :)·.·· ' :.,

. I' '. ~ ·..

+ iR~
' 1 ~ "'1
~ .....
.
c.:l.
. 2 •
•:._,· ..

B
..
·,~
.• •. \: ~- . . ·, ~ . : . :. : , ...'
.,·'
·-·' ~
.
• ' ' I_

~ .

:· . .
.:,··u this .is .fi.Tst dotted and then crossed with '(;'1 ·in an attempt to
.· . . '4,,:

'::·'.
,;-
,··solve for i\ in a. fashion similar to. the solution for, _RnO ..<~~,&)-,
~

' '-
L '.

l
.
::

:'1: :. >·-:·
·-:~-:~-
·-_'.t .'

',':
~~.

...
,.

' the result is (by setting n ... 1 in _Eq._ (6!?.1) .. ··',.•

.. -:- .·.

_:->·
_' ..
' ~

': ·-.-~ .: : ::
:'.; -: .. :.~:.-:;,; -·. . ;·-.:.-
• > ,,,·, I './• •

.. ,·:·
': .. ·'.: i
.··,'
'• ·.·•
,_ .... ·I
\r, , ••

, :· , -v· . <._; ·,. ·;· •.

).
. ·.: ' ..~ . ,·.- ~~; ·-~~; ~- ·, ")

1 • ,·
.
. '>:. ~ ' ': '.· :. ...· .1"7~1
A.
.... · ... ••• . ' l -~-·-. ,.
\f
': ·, ·,:. \· \ > ·. ·. ._:~. :· ..
.,

·_ .'
., • ,I

··.·
,, .. ·· .. ·.·::.·.·
. '·,. .• .. j'·:·
'' .. ·,· ·.··, :.. ·, -;_
.... ''t"
. ~ • .. ·.... : : ..: .:.;· i .,_ ...
J "·1:
: ......... .'.· ,.·
. .... ·' '
~' ..' . ···:
 ~------- ------------:·-------~......~~~,I"
>. • ~· ' : .I •
-
f ... ' ! ,·
. ·" ' ' ~- ., .
~' . ~ 'I ,' ' :-;.: • .·
.. '·
41 -
L. . ..··. ,- ·,.- ·... ·.
l~-----
l
f: ·i
<· ·:~ : ?
:~
1.
.·'
- ~
.. .
•• "~ :"
.
.;
.. '
' . ,·
·,

::. ·:: :·: -· . ~.: This is a. cond.i tion on F'J,. J which to "*(J.) is ·a condition on
\. •< .
:·;
:,where

f" r
10 c ~0 • vE<R;y t )+~0xil(il'O'_t )+Jl0x /5o
!__._-· .;. '·.
L.f·. •.:,_;.
.~
f''. . ·s·-:.:~-~-~ . 1 ••
and _\71) · constitute a differenti~ equat'ion. to~ _ R10·• ::,. ..
,, .. •· ta '
·.
I
·~-. '

: If the terms With .1 ~ 1 are se:parate~. on the. left hand sidei the differential
.

. ·. ---; ,•
. ' ,'
·";.· . . :: ~-

x al = !.B. cr: . . _. iL,. x ~1 >,

J... •
.' ; -~ ' ...

• :1 .•
:.· .·
''··,'

:·; where '"l.~.. = [-~ 0 . ~ '!? E-l~0 x ("R;_ 0 ·•. v )~ .. BR';_0J.L and ~0 is the
· z;ro order.motion~L vll'~l +
~ ..
One .might conclude from Eq., ((tt;) that u; . . ·: ; .. · .·:.

10
(R ~ must equal · BJ. Such is not the case, however;_ . a:ny complex vector
.. ..... .- -

_',-~::: -.quantity of the form ~· can be written as (g + ih)(~2 + i~)): .. :

~ . .
g~2 + h~ ~ and
W= 3
' ..• 1, .::-_.(:

<'·.:·where · g and h._ are complex:- By colle~ting all terms

'on one side of Eq. ((!() ), one finds that

• L.L.
·(R;,_0l- B also is of this
. ·.··.-· .
.:. :·: ;roz'm and is not necessarily zero. • t. • •
Equation (/Jo) must ·be used as it stands,' .

:·:-;,,to prove that '11'lO • -::.'t

d (B ...
cit R10 *> equals zero.
::
.··. _,_..,'
'~: .:-

;., :.'..

d
dt
''.--.···
·, ;,._·

_Eq,uation ~q 0) c~ntains only _ ~ 0 ~; but this is all that ·is required in. ·'.: .· . ,· ..
;, MOO -~ . .
. . . ...., *. -'":1 '~-~ .\.' :.,": ~~-~,- .·:.:·.
Eq. (Jf{) be~ause of the fact ·that ~O . and .R10 ~ve no parallel _ . " <-· ..·;
.. '<7l> 'gives,'-: whencliff~r~ntiat~d. ~t~-/~esl>ect_to<time>· .. :
...
~ -~': ' ~ •• _. ' • • ' f~ ·,

~ : ••: •.• ·.'' 1', ', -~ '

-t',•.
.
~- 1
' •

- r.• ·.·· :- ,: •'

•1·'.. .-: ...': ·, ·.. •

·',·.
;. ~ .
,','

.
... •. ·.
_'r',
~
1

;· .. ··. ..
\

;.·:,. ~ ..... •. ' ·'

'···.
,.·.,
...
.. ····· . '· ·.... ·_,·. ,·. . .. .
. -~ .· ······' . '·' '-~
. ·.
. .
'· .-<,>··:··.·.~--~: :--~ .. '• .... . .
_. ! ... . l
 ·., : .' .. . .·
'. _. ......
. / . ·, .
'

. • • '.j.... . • ·,· ~
'"j ;
·... ....
• • • .. . ~ ~ ;. ·' l
42 - ·'I " ''
. .· ...
'}• ,t ..,. . ~·

·; ~·-
'
\'
. > ,; .·
.. .. .'.
.·. ~· ...
;:
~

,.. ·: ':_\' .:.~. ·:: . . . 'I ' :... i.'

,'',•·

.' .~
·'•'
'·:
\,
'·

~-·
."I'
:,•

'
; ' .I

·..:· . R;.0 • (~ + ib).<~2 · + 1~3 >· + (a + ib)(i + 1i

2 3
r
'. -~: ~ .~.:·' ·. .' .• ; ': . .: ' .... : •. I
._-;··.:::,••.
'',:
~ ',.
iI
:· \'.· , :·,.:·:· Substittition of· thd.s int·o Eq. (Ia) gives (via ..eith~ .th~ ·:._~;-, . ~r':;.·€-;f.··. :-,
":::_:· -., :. ·'component) . - --· · -·. ,
. ~ • ~ •••· : .... : . :.:_ ,· • • ; 1. : •'

':.,·~. '·. . ': ,": : ,_,; ,• I ·:: ' ' ·, ·~ :• t '

\~'. ~' ; .( : ' : ', . . •_' •: • ;~: < I ' ~ ., '

. !·: .<.· •.
···.'.
:... ·.~ : ., : ~ ~ (e'3')'

,.;here ( g + .ih)(~ .+ i~) now stands· tor From

·.
equa.tlions:- ~. ·< -
. ·... •

·_ ~£?2)- and (8~) it_ follows that

·_}··~··

>~>::~ _\
.···\'''!•;,
:.·' .
a._ -i ~ -. A
e -">
FL + "" + i_h
~
· -, 2 -_ 3 --~o 2B(a + ib)
' ..· ,: ~- ·•.
·'· - -'

·. - ... ·......
,-'

. ... ·.:-
··- .'·.·,
:. '.: , ;: ·': ':_de~ined by
'-. ""'LJ,;.- iLJ..
\ ~ ""'
x e 1 ..,
; ·.

:·~_:::;:._.· _ ,_··. ·: :' g + :ill·· by with~· ~2

. -:··
·:· dotting .. : ~

1.:_:·:.:-
~ .' r . -;, •
. . ..,. ,

..·
. ·._.- ·c;'/J . ,.
:I
' .,
. ...
' ~
·- .. I

.-.~ ~e explicit expression for L J.. given immediately following;' Eq·.. :' (9~

, . . , . ',i· mu~t no~,be used; .· tJ.. c()ntain~.~- -~o>whi~hls ~?:ba·:.~eplaced\ii<·~·a :~"-;~_)(~2 .'+ i~!···:J
. :. . -' . ' ~- ·~. ' >.'. ... . '.. ·. ·t. ':'",,), .• -:~··:.:_-:,·:.-r~·. '•!. . .......... [

: ~ ' < :-~.J ];;G~,i;i;C:;~;) ~ ''~~;),,::~ , ..,.- Aci]:;:,dX:. ~-Li:l~~-';; •·- ·'. •. •:
 "'.' •' ~ . , .•
..... · : . ',1 ••
.. :
.....
43
._·, ·.
•.J.:'
I • ·' ~ . ~'

,•,
·,

[~ x <~ ·v% + -e; x <€"2 ·<?rEJ + 1 <t.. >

' .. ....
~~

. ··:·
··'.
···.• .... -1!1 •
. .';
''• I'

' -~. (a,)

·...
t • ~ •

:.· .. ' .· ;'·:·~:

·.The imaginary part does not have to be .evaiuated because only the real '•1.

'· ·: . :~;::~l
•.:J
' part or (og) isi required. . The second term on the right}···side of· Eci, (eo) . .;· ·~·~.:{!i·
.
. :>;!·.·t
: ·::· .• ' ~

~2) [<~ - i~2):~~(B~l)J :~:r~

·'is
\''.< •
•[ (1~3 X

•.B[~J· X (~2·" V) ~1·- 62 X (~ •. V )·~j. .,

:J
:\:C • . ,:· :.·,: :.
·. :··... ·: .. ·,
:;~i1
~····!"
.:.r•; .
.:~:.~. i~~~
·. :.
. :· ., '. '·.: ·, .~:·:.~\~
·. : ,'....:"'<

. ,. :
.~

. ·.:.
. / ·: . '.

" . ··'.
t· • ;

•; .. ·.,. .· ;.•
···: ..........
.; ~.. . '' '' .
'·'
";·. ~
;,
. :. ' . ;·I
.... ~.. '
. . ..... ·.' .~ ·, ....

. .:' -~. :
·:.t··
..· )
.: ....
···.Because . ;.-.·
.. ·.-·.··.·.
·.r ·•... .. -· . .-_, .·
I ' · ·. . . . ;... '.
' ,' .

. _,; . '· ·::·' L Q .:l, .1711 B.,.. {7o (~i B) = ~1 • VB+ B '17•' ~() .•:
·: <

..'
~~''

~
.... ' ' '

' ' ' . .. :,':;

• • ,: •

' ',
'

'•'' .
• ••

. .. ' ' , .:.

1
' ' : :~·· .

<~t fo~that ..~rr V• ~l =~1 ~l VB o .·and .the .real par.t.; of.~e· right~':.;'.:·.:.·-:" , ;:~·~·
.:<·:~,.:;
". ,," . ·'. : : ~ .. ; .· :\. -~s~de·: of Eq •. ('-o)·. ···is..~d-~ply:. ifoo .~ .VBe ,. · ....; . ·.:.~· :.: ·.. · .:.:.J,')

•~; ;•' I,,

''··
,··.. ' :
I '
'
o~i;;:·. ' '. :~: ,, .•. . •. ', .·~ ~<:.> ·'. .: .. '> . . ''<·' ; ' >·; ~r·
 .... ·.
. ..
..
[.·.

!
'• ,., ·,·. .·:· ..•
·· .; :~- :.~
,•)

44 . ~.
-~t . ... ·.. ' ·· .. ,,1. •

·-.::·.·, ·-:. .. ,
;.·' ··... .. ·:· . 1, ~ '
ot (9fl)
_

. ''rho lnmt tann on th0 ri({ht oicte

I·~ . •:. . '.=···-·.
····,,.
I,_'·-,

~ .,r.
~->:
'
....
,-i
..-.
·.> ·~ ,··:·...and
J :. ·.-..·-
~e(
g+:ih 2B" •
a+ ib .. ....
-., •.

.. ': . ~ oB :. .
I;:·
. ' ' ,:
~
- By Maxwell 1 s equation., VxE"'- ~ ·. • t1x E""' ...
1 . ' it' since
~~1
.
't
.,f.
•·- mQ. '•·.· , .r.-··
·,, ~· .
··;,. -·-::-
.. :.
~·: ~-_:. ..
;·.
i ·.,
-: ·.;
·_:-

d..e ( _g+ ih)

. • ·.... ,\_ . ,'
: .•. • VB - 2B ""-BJ ;i
;· ·, ~ a+ ib · ·.;:__ , .,.
·,. ,.. :•
.'f·'
' ' ·,
. ''
-<~ ~\-; . .
+ w .•
-~· .
- ... ·
Thus. it is proven iha~
.00 .. ~-. •' ~- ,• .

.' ..._--~ '

~ <BR;.o • R;.o*> = 0 + <:?' ( G ) . or that equals a.· '

'•
:· .,_. •,"I

constant +. 00.( e) .. ·.·. _.;.· ,.-·

·.-. . ::, ... ' .. ~ ~

' ~ ...· .. ·. ·' .. ·.. '·

·'· : -~-.
.. _~ . '' .. ··. :,
.":'.: .. ·>-··
'• . .. ·-.'

., .,
'· ·.:
.. ·.· .·:
·. ·:· ··:•,:',
.. _: •.,
._

;- .... . '. : ~

.-... ·
-:. :-.-.-·
·. •,-·· __ ;• : ... ~ .. ..
,·

; .

,,
.; :
... ,•
·,, \''

',• /
.I'··',,_../,·,-.
; .. :• '' . . .....
--
:.\.
. -:':i \. ,.,.' ·.\,. '1·_.-
..,.. / ..

. .--
,/
".
....·. ;

'.
. . . ' -·-. , .. __ -_
;.' ~

I' I, .
... :.~ . : '.:
. ',:_. ' _, .'
.,
.·._· ._ '·

......·
•,I .-.•
·····. :.. ·· ... ' .
' .. .•,.
!•
:-,.:-·:
:' .·

..... , •::
~ : ,..; • •• l

.:-:-. .., ·. .. :.~

· .. ·,.
·.'·
·.,;.·~ --~~::..'·_.__,~·-_;-
 ' • <

i· ,:,

•, ··.'
.... 45 r, < <

,.
. ·:· .. <

B~ 'lhe Longitl:ldinal Adiabatic Invariant .

·.,.·

. The next adiabatic invariant to be studied is the longitudinal · · < • < :.

·~ ~. ·.~· ._·. ~ . .:. •.

.- invariant· ....
;·':..

: .. :- ·,;_ ·.
J ... f:P, ds;, . c~~:
]j i11 .... _ '.
· .· .: :< •
,.
I
: '.'
•

::~: , ..
<
i.
i·. ·... ·

' !

is the ·guiding center momentum parallel to the line of force, .- ..

_,,..;.
··-..
~
.
; ·.
~

>-~ ·, and the. integral

~ .· ~-' . .is taken ·over a complete oscillation from one mirror · ··.·.

.... '-:
Point to the other and back againo As stated previously, the longitudinal ...:_ . ;
• ' • ~;:·; ' ~- -~~- i_

. -~ . ·, :: •··.
\ .· ., .; . · · · ;: : motion must be pet-iodic for · J ·to exist o The procedure thus fa.r has
t -' ~ .· .. /. ,;~ .
i' <. ;. _ :··:--·' ; '·. ;' ',' been to start \dth the equation of motion ( l ) of a dBaged particle and
; :. . . :~': .' ':) \7' 1,

\ ' . : '' ,:: . ' '

I ., ·'-' ... ' . , . -:·: .:,-:
to average over the gyrationo The resulting guiding center equations ( /?), :·,
J· ·. . / ....
·-. and (~o) · are new equa~ions of mot'ion. ,... ··· .
The next step is to average over·. ·-·.·-.·.,
', .. - ; ' :. ~ ..
·.,. ·.. - ...
~ ..
·, · .. .·,. . ·; ~:the longitudinal oscillation and obtain a third set of equations of.· · ' . ··-'1-• .
. ,.· ·. :.

i: >-':--'..,',:· . -... ' ", ., , ~~ ;, '.

, ; :; .. motion governing the average drift from line to line, and then to show
,._

';
'··~' ..

i. '·.. ~ ·, ..... ' "

) '·
l·
. . . . '~-.·. ' .·,; ~-! .. . •' <

:f.
i
<:· ·, , : .: ·:;::;{. that this average drift conserves J.
> '
0
I I
0
• •' '' ' ~ '." ', ~ ' '
'I
'• r
. '
\' . ··.
·' -· ·-' .. As the guiding center moves along a line ot force in accord with····
,
. ! . .-...:~ -~ ·.= .~"' '.
.. ·,,•: ..:···..
~ • . -! .
•
1

'i; ('•. {do), it drifts at right angles to the line at a rate given. by,. ·.... ,.
) .-; .- ..: ·- .
. ,•, . :.
·.. , ····<:~ .I
.. . . .: _.: ~' .. ·.
.. ~ ·_ . -··. .~ ,•' : .
. ;
l.
'
'
•
'
:,.;_·

' i
i
,_
·.: ''' :.·~:·:::_~.' '

i_- ••
.· .. ,· ·. . J~_-:~ ':::.-~ "
·- /~ :-
:'.
''
- .. ·.·:
(·, .

. . :::·.. .._\'·:_.-:
_ -
__ .. ; .·.
._ ''

. Gui(j.ing ·:center. oscillates .along. a line .~:f force and drifts. '·
-~ :.
., __
I.
·.-.- ·.·.
_.· slowly.at right. angles. t~~ ito. :··· ·... ,• .· -~ ', ,: . .·.--,_
..
_-:·~- '.: ·: .
...-1,- ·.:: ~--t ~-- .:·.: . ·, . ~- . ; ~ _.., -, I --~ ' . '. '• ,'. ' : '
' ', ~
,,· .: ,•·, j • • •-_ ••
•' ·_!,•,: \•
...._ .. ·:-;
'I •' ·,' •• ·:. : •. • · · . •'• .''

' ~ '; '

• I ·,,,!
.... :. .
.·. ~- :._ .. :-.. ;
;_'·

:.· ·.-j. :':'-._: _(- .. :· . ~. ~- ., .. ;

.:_' ',
. ~ -,. .
.• '·
. . .·) ~ ·-._ ·.·(· . ' .
. .... · .... ' _,.,
,",.: .·· ~·
,I ,·,

.. , :··
:.· ·'
-
", .. ,.·,·'
;:
. . .

. .--.. : .- ' . ·- -:
·'•. . :.. '

.
.. ::· . • I • •
'\_-
... ' ..
I :
. '
.
'· ' ·. .
~ .- -. '. ,.
(.: .. ,· .• : .-.· ' •• : - . ,'j •• '._
.
· :
~- ' '
\,, ..
----'j· ..... "'.
~- ' .• . ' • : '• ' : ~ •t ~ ..
 . . ., :

.
..·:.. ·..
. I
.
' ..
.

·, • I • ' ' ' ! ~ ''o

·, ,.r '.I

._ ... -:.
,,
·. •, t
f) . . .. .
ft. ···.}
and.
..
ar.e of C"(G )), one can calculate. the average .• _:-;
·

' .. :··.{~drift r:ate at right angles· to the line during a longitud~al oscLillation. ·.
.' '•:'~ .)•
•• • • II!

·.;<(.as if the ~iding center did not deviate from the line of force 11 the • !_ :· ;·. : ~-
~ r :· -~
:., '• • ·:. '· r_
• c :• . . . . ••

6-. 2
1
. <. ; : ~. ; .' ... ' .: ":~
·_. ·• · error bein'g of order since the drift rate. R.~. · .is 9 ( f5 ) o . _ 1·_,. • . • , ,; . ,I

·'· '
~J..
• :. l I •

::,is <!'" (l), contains the C? (1) term-~ and the guiding center moves·::'.>· :·~1 - : · ·: ;
. . -~ '~- ~ ' ' .:
.·I-'. ·- ... ·: · i
",'·
.. ,; .a
.
long
.
ways from a given line of force in one ··oscilla. tiono It is then no . ,
·-·: ·r· .
·-.·: ·,.,.
.•
.
I
:
.
-,_ ·.
. ~

.
,/ .
'

,>::,longe~ possible to ·ignore the d~viation from the line· of force; the .·'

~- -.l .
~-:: . . .
..··.·
:. ~ , ::

I'. • ;_.

· .;<guiding center does not remain ·even approximately on the line and it will · :. . . ) .:~
_.. ·:·:-: .. . .· . . · .. · ~ j, c. •. 1 ' , · . • .. · 1 .

., fo~d ~uif~d~ is not conserved •

...
A- proof has been given in reference (:1.0) that the average drift ,, ·. .
~ .: . .. .
r_;.

··:from line to ·l.ine conserves J; the proof is for relativistic particleso · .. .. '

, .. ' ' '

.. ·:,_(Rather than repeating tthat proof in the present··review, a somewhat different·· ..... ·.' ·
.... .'
:- .. ·.·:.
.. . ._. ': ~. ~ ~ .;'.. :
:··one '_'~ill be given \r~hich is more algebraic and less ·geometric. Also it
'

..
·.·F;.'~will be done for non-relativistic energies, but the . modifications of this. ...... . ~- .· ·;::_ t. -~·:
.•• t ,·,_

:proof to relativistic energies are ·slight. since the relativistic expressions ..:. ··.: .· ...
. . t. ~~> ·~·
·..:< ··' ~-·
·.. :. -';
:::(~<\), (s-o) , and (GB) ·are known. ', I, ~ ,., ,:
·.... ~ . ....... ·:·: •. •'
.. ·, ' '

: ·. ;-.'· ~' l·"

?.'.' To forinulate the problem explicit)..y, the o! ,·(I, s · curvilinear ·~ .1' ·.··
(' ..
' .
.,:·~·-.:.~;.
:' ' . :~ coordinate system introduced previously will. be used. Let ': ·.... ...... . .. ;·.:.:. :·
.
.....
· ·::_-.·.·:::_.: . .
2
l- -. +
cs _mv_l...

.'l....,.,1·-:; ··'. .'. .

2
·~, ~.k
+ e(¢ +1'),· where : "P c '~MB ~. it<
'
~
. · (at-)·
,. . ... '. ~ . . K is a constant of. the longitudinal motiOJ?.. Then J · is given .bY .
'·····.
. •.... ·
. .... ·..
,~ ·
..·.·
...
.. ··... ; ·.. <·
' :
···.·. ·..
'· .

·:
 '''
,. '<

I ..
.··:
'. :,: 47 .'
· ..
.. •' -~ i ;

' ,, .· . ,·..
.. .-:.. '
,.

'~r·. m·ft~rK ~ e('f+?)

,• .''

~J~:.d~ (~+?t~);·:
ds

··.
,,,
'· '
· ..
,·...
.... "
M ,· ·.·.·_ ..·.

_.,.
= ~e f ds
v,, oct
?;
<., +~ + -a B_) . , ' , . • .,1 •• '··:···

l. (tt~ .1" fjjl-. . .. '(

} r~ a~ -~! • ·~ ~ l • ··: ,i '_( . . ..."

..
. ~- .'; ·~'
. _: :'. ds o (1P +<P+ MB
-:-- -. -) ·~·· -:<_.(f'J) .;-
,.!···.·.
,.:,-._
.
'li uf3 e ' ..
. -~ '.
. . .. ~ '. .:: ' .• ~-.
'· ·' -.- .
~ ~ f_-fm_U_K.....-..;;;d;....:-(-V-+-ct-)--.~--~-=-1/"1"':. = f ~:
'·,''.

... ~.. > ~: ·~

(IM:!) . .· _:._.:
: = m 2- = T, ···:·_:·_·. • •• ¥ ~-

···' '.'.·
::·.
. . ' .. .'.· ~

i~:··the ._pe:dod of the longitudinal oscillation. ·. :.·.·

oJ
-:::
,., :; :
{)t
.... ·.

·:...~: > ..... · ··.•.

. \.,.
= ·- e ("'f +f+ MB ). (lol) ·
e

·,·
···In equations·\~9')- (/ol), 'It·, 4', -and B are to be considered as
.·... ,.. -:-:

·.· · •• '~; :.;ji

of (a, (3, s,. t) and not of the_ guiding center position _It and of· t~
·...
,· ·t···
;..·.·'
;
.~ ~.
R · is
'I.·
itself a function R(cr, ~~ · s, t). · It must be remembered that . : ~:. : ·'"_,, ~ :'··,,';\ .j
: ~ . ,'

: •. ·~ -~ : . 1 t;;.,~.)_ _ .
... ;· .;;..~_'lf..._(.Jl,;;a;LI;@u·z;.,;;;SJ- = rR(cr, f3, s, t) '.' 0: )'.1
.' :::·· ?J t ~. t ..
'•i
. ~

.. ' ,· .
(1 Oci~J ·~
~"

: . ·...
'·.
·.' t
. "•
.·• .~

J> and $ .. . > ,: ' • > ::·I

·f' ,: . ._... ,' .. ·.. .. ~ ·{
.·,.'··
•.···· ·.. ·.·.' \~ ~! • ~

'H
' .. ) • .... , ·:, _:; )

·s.";_;'-.~·.;.:··J·.~: ~- · · ··r
:: ; ' :: ' ,• Since a(ii, t); and' fl(~ t) e.re constant~ ·.F . : :i.me ?f.·t~rcie, ' •.:.'· ' '
: ,; ,.· ' ~h~~~ time de~i:"'ti~ c~~f"~ ~": and ~ot ~I : , ' '' · • ·.•.• .•..'·' .•.....·· ,, j.
~ .'· .:· :·~;
;.;'~1 , , .
; t
,,..•·:t;····:.
:l. ·: . ·:- .. . . ..
. ,.:.:·: ;:~ .·''...\1 . •.
.:~· ~. .
,···~:•. ~.;
·~
~... ,r .. · ~···· ,••: ~
'' ~: .... •..
· .' c '.'
. . ·. ·~.f.'·.·
~· ~ •;/:> :• ,•' ··; ~· ~ r' '• -~··.·.·. ·,. . '• . •:._·.L
' .. . .:; .·.. .:.> .. ; ' ' ';: J'. :· :- <, : ·. : -: • .; .. . • ;· ' ..: .-:·. : . :
. ~ :. . ' ' . ·1: ::=. .. . .. . ' .. ··. ,",/ .. .' ~ ' ... :··
 ·.··,·
·:i 48 .....
....
-:'.', ·,

da
-·= . ( /&J)
dt ·,·,·
"•,1

~ ~( ..,..2)
.;::
2fl !f?.··.. ""'R• . ( ht§.) ... '.....
; __

dt =· +
. at . J. ·v~+v""
.
'.
,·
. -' ~
. ::. Th® expression _for R.t. (with ) must now be substituted
I I .~ ',,• . : k , ~ :;•: : 1 • ,;: ' . . . . ,~ I '

~ into -equations (I~ and ( /tJlf) •

I.-~~·;

It is the same for (except for a sign.) •

1 ......

. :··
da .· ..·oa . ~1 r· ~ . me
.at - · .._+-
- =. . at . B x -c,t; . . .
+~
e + e ·va + ~< G~)' . '

13, s, t); t] and s = · s['R(a, ~~ '~~· ~~~ t1_. ·. . '·.·. ·,

' ~. ~--'!,•. ·:· .: •. j

:, By implicit. differentiation: of :. ex: ... with respect to .a, 131 s, S.t1d:.t;,. atl.d 'of, ·: _;: · : ,! •

. . ·;·>:/_. •'
With respect to the_ same four variables, 9~e obtains the ·. }· :·
··:_: ·.·:·,

equat~ons: ···..

··":~_;;_:·:1··= •. va(~, t) ·: ~! (a,· ~>a, t)

·:.·\:·.
-: ~.. . . '
·.·/. ::

.. .·.···:.'
.,_. '·' ',• '

·'·

:, .... ·. ·.·· : ..,_._. o = va ·

'-.·'
~· . .:' '
., , ...
·...~~--
-
._,__ ! .:· •·

. •!,
'
''.
"
~ ....
:' . ;., . '0:= ;, ;''

· · lcovt 6 ~ ~ .
. -··.-.... ;
• _$;>(,~ '.·· .
..;_ .-·... ', . ·:·. 0 = :.- :_

.···: 't•;

~ . '' .
... ·
'< ,:_,
! ., ·6 ·,'
.: ~

:- ~- ..
................... ,•
··., j, ' • ·, -~ (. •. : •

f : ..
:-.· ..··:·. _ . ·.·:-.
: ~ ·' . . .
\
~- 6R .... ::
0 = at +Vt3 dt
•,

I;·' ·I': ..,

· ·.• .. I:'C.vt ci .. __,,; . ·- .. ' .
:;,"R ...· : -·_ ...::
• 'i ..• ~ ~ _).... '
... ·.
~:.
·.·. o :=Vs(R~ .t) aa . (a,
,;·

_,;
·'

~ ...

.0' :.:: Vs ·• • u_
·. "R • • -~ l

'· >_:_;':'<:i·<· ·' .

'·':

.· . d 13 '· :-· 'i':

.
1 = .'i! s .: • '· , - '·
, -o~ '>:.:'·.:.:· 0,;>:>· ·>· ·' :-

·.,. ·. ·..•. ~s .·
· ·; ·, C>s. .+.. , · ·.·. < ·d'R . ,.
..... : 9 --~' ~-. •' ' vs .:'\ '3-t ':\
:.. ·' '; -~~ ~'
: .·.
.'
;

. ! .
. ~ : .
. ';,
;_._ .·: .·.
.'
',
··. '-: _, ...... .. ~­

;·:·•
. ~-- :>; ,: :· _: '
'•' -·····
-· .. ::·. ... . (·.

. . ·· ,\ f·

I :,• . :.. , -~ --:~: ' ·.· ',

'.
•'
·• ' ··~ • ~
;I'••
' / ··I
I;':O~.:c··-: ..... - -:-------·--;c··.---- ·,-:----·-··-·,_·: •-·-··• .,. .. ·•'.· ...:. .. ·
! ., • ',• .. · ~ . ' .:: . . ,; .

' .· . '~ .
··.r.
. . ··..:
49 ....
·,'·'
. ·.... ~ . !. '
. ::.;., i ·.;
...,
_,/.··
\ . ,_. ·. <i
·'. :-

...,., .....,.
I; 0 ,'.

.: In Eqo ( los-) v~ can therefore be replaced by d"R X B,

tinea. B(R)::t7oc.x vp and

·or ..
.. lr'.'
.·~· ., .: ~

: - . I· be. ta. ,

.i: '

Also in Eq., ( /o.rl

····.. .J
_: vlith these substitutions Eq. (/or) becomes, after interchanging : ;_1
·.;: ,.
.. -;,
.:-~{
·.the·. dot· and cross 0 and expanding the triple ·vector product.·. -~.~._~. ..
-·:

''·,'• :'· ... ·.

,']·
' ; . p t • ·.:;_;
. ..-•:
·;
....,1.
-~
·.·. .,,'1
.·.:· .. .:-·. .:\~ .
..' ,.
. ·-;--t

'~- ' ! : .,_

.
··.•

,· -· ·,·!.: ..
.. -·:-.
~ ' '.
., i
-:'. -- ...... ~

.. ·.,.;
·-··'

9 equation (/tJ?).

• ·~ l
'' ..
~ -~ ': .... _ .:_
~.. ·_;: .. :..
,. ...

•• J '": ·. . . . . '':,-._
: .. __ ;
.. · ~- ~

. . ·.
... !'\,
'•, r
.··.:: ;·; ~
'. ~ ... ' .
·.\··_, :'· . . ··,;
'._ .
•'
~ ..
·
,·, ~ •J ~. ·, .. '•'. ')'· : ": .
. :. ~:·.··.... :·~·.'>·. -~ ·_·.-
• ' • .. • • .' •: { •. '.· • .; - •
_·, .
., . ~'
.
. ~-.·.·· ..,.
'
'··.
.·,' .·.··, ' . ' '·.·· '1·. ;_· ..
': .;._ ·,··I .. ~.,. ,
,-;,
. . .. ; :.. . ."_:· -~ :·.: ".·.. r·~. ·. '_; ·. : . ·>• ':I) ....... •',••

. . ; .·.
• '-::, •• <- '. . • • •
... ·•·
• • ) -~: .•• -~ <: .·· : .. '. ~ ,,
' ... ·. ' ·. ~ .. ..
;
 . '·
.'~ . ·.·, ' .. ' :\.. ' ..
. ··'··'
.' ' ..
i;.
~ .
. .. . ....... .·· ··,. .._ ..
! . ~' • • • • ' ' • .
f·~·. -~
h .·· '.;'
'.· ···~ . :. ..· '.
·. ' . ,.. ~
.... . .
'f .
.
;;~ lJ ( . 'JK). ~.2,.
r ~··::. :. \: . ~ -r; .:·~.·. ~:s'. =. T.i t. ~ ¥ .·i'
. ~-.'

....
;
'

~-- ~ . ~- ~~- ..~-•. :;.:._. '·,·! ·, • ·•• ·•

. .. (
l'l

. \. ~ ' '
· ... ,:,,;·
•'....' ,'•
= .P.·(~ •. ~i) - ~. /J
·-~-
ai _. ·_.J ·(/1e•o. . :;-) --u·~
. ~ ~,
··'
. ,., .. as . -;ji . ap os- ~s. . ' .. "lj . .. ' . op . ... , .r

·• ..
::_-
:"':

~
' I ' ·,
.• '·.
·· •. ·.
r·. _.. "···· ' ,•.
...
_:
. ' . '·. .

~1 ·= ~R/os •. ~ The last term of Eq_.

. ~ ·..
.... . ~.
···since (110 ) then is ·,;'
...
.. ·'... ~.- . ·'· .. :·: .: .···. i;
,•. '·
. :.'~ ;- ·:;· •·
·: .·.·· .,:
"""".c.v, d (e, .·J~) +-iK~E-a;. ·::,:(_ '·'··;,._).:.
4 dx- . dp .. . . .
.··
. ,.
(110),. combine to
. ':. .· ... : ..
·.' .· .. , .

. ·:-.
. . .·
.'.

' . . .,.. (/13 y.·· ..

···.•
.... '

.. .-·
:·. ,·
An analogous procedure is need~dJor dK{dt.in Eq_. (97). The Clu&'"ltity

• •• • 1_ ••
.·~ .: . ... ·,·.·.·K was defined as the particle kinetic energy plus e(-y + c(). From Eq_. (:2'/ ), . :<.):'·':;' ·. •
···;; ..

.,_·· ;-'\ ;

... ; . :·. i
.;.. ''
 .... -· ... .....,I

' r
·'' • I'
.,)
.,· .
.' ·~ '
.' ..
·.. ··.,·_.-_'
.~ ·. :,: ,>: ~ .: .
51

. ~ '
.
' .
It E.L. and ~t- are C7 ( G ) , both y and pJ are themsel vas ·.· .

·.· ' . B". ( 6 ) and their partial time derivatives. are

. --:

:·.··:

~ .~
...1>p •• ' . '

induction ef!'ect t!~ , plus terms of 0(1::2-) at least when · E ,: '::..-.·. ;·:. . . ·I •:· -~
',, ~! .
' 11.. &1t. . I . ' ,r. .~.,.
,'-,
.... r

;f
'•; .•

and &·· are. .0"': (1). It.is also true that if and are ~
,, ':~
.}
', .;· Ot:_ : .. ~· :: ,: ' .'
.. •. i
·.'

.·.· .... '

. .
:

.·• .~··
... ','
'

~
•• •
·. ; . ..
• 1·• •
. :: ; '
__d(Kine.15±~ enex,gy) - i . .. -'. .\ ( /16) :. ::~..
.. ' ·'
\,:·,. .. ': ' - -~

. edt .·.·.·,·~ .··...... · ·. ' ....

.. --~·.:'
.-'•

·.~ ...... :-:: ~ ... ·. ~- : . ? .:: :: ~: .... :

) \ ' ' ' ..... ~ ... '•'

...... .
~/) ~
.,
'~ ', '

\ . , ··:·· ,.
I A

··.. -~-~ . ~. This can be verified by starting with Eq., ( . for and '· :. ·: .. · · · ·:·:_ :_ \,:;\
):.\:<::_calc-ulating. d(Kinetic energy) ::a e~ ... E(r, t) •. Upon expanding ·_·_· . . _ '.':,.~i_•._·-_._:_;_::··. · · ,.,. ·"
·.. t ' .
;• . . dt . ,- I .

. . · :-._.:_,: it~; ~-) S:bout .E(R09 t) . and tiroa averaging_ over asgy7~tion ·.7.. • ... •...
~;:. ;\:,,.:·,:;·(over. f7 I) that is), equation (II~ ) results,. The Maxwell ... - . ·. -:·:·.:<~
-,\. ·:,:::.·:;_J.,-·-.:·:;equa1a~n ·. V x.E = . . d.B/ ().t ~ust also be used.-
.... .r
;:·._.:,_··, .·.:· .:_·_';.:
: : ·, •• •
:-/i:

;:.
Addition
.
or
'
(1/~) and (//~ L, with E from Eq. (/011); '\ ) • :i ·-dJ
·yields '" ,;

.·.· ... _)/Jj

'
.: '· . .. .
~... .' . ·.
>~~ .·:::.:,. • ),A

::_.:::. :~ . . \' . '. fc . _.'2- (. 'IY+A,+' MJJ) .... - & ~ T 'f -

''I · e •. q"/: . . • . e..
... ..

'
i? •1- ( ~-rlr/. ~J. t7C c-~
'

: 2-c y.,. 4'f !:9) ;-

at. ·. . : 4! · . at
J.. .C.
..;;...
dt:
r?f' 1 t-
j . I ·
.. -:.·.·· .
,(
.. .- .·.'• .
:. t;· ·. where 'If 9 · flJD B ~ r;).., ~ are all,.:,_to be con~idered . .fUl:p ti_ons· > · .'/
.( ·.: ..

:·.. <:': ... of 1L and- t~. The v~ctor. ~Y"i- ~ 1761 · 'app,et;lr~ .frequently ..; .. /~-:;
·. :·and tiill be denoted py_ ·.-; .. The dri;t- ii'.J..:·.:.·:is.now repl~ced :'··.};_:_:_~<;

·. ,:• ··:: .t ~ F~i! ~ i0.~~ ·~~~-?~J . . /• .· •· )'he~e -~ - i- . V$ .• .•.. , .. •• .;', / ;;:!!

'•r
.!:::..
. . .. \:i'-'·.· -~ \1 ( 't/'1-
··.·; ·,·~-.. :t.".'~.' \, .
t/) 1- e .~ ':· . ·~· \ .'. .. .. ·. . .· :. . ~-.
·.-:. ~ ·. -~ ·. ·.· .., ···.··
_;.':.". ·.·.'. .
·,·· ·" ·,,,··~·-.'~· .·.-·., __ .· 1 ....
. _. :·>: ·.::):! ;
~ ~;·~·-. ··~·.::~:~;
• • ' : ' • . ~.1 • • ' . •. • • ;: • :. ' • ':' ;'. ~ :. • • ! .. ' .. >: • .-. • • •• : · ;

.;~~ ~;-.)·,:· ·.. .'.·,J·.' !:'' ,··."·~-.~·_,·.. ;., ·~.. .'.: ., .·• . ;.1 .; ·' ... :.~ir.'. '· '·:·:~;·.·_;·:.. \:~.·:~....... ,,::~
.'• '' l ~;,-:· '~~~ ~~·:_.·.~ ·:~•
.. ,• '·( I~~ • .'•',' ·'•I~ •:•::' \, ·.·~·~'.,r •' ;-' • :·~. ....... :,:,'">;·.:._:· ,;, ,','' ~e •' !.'" '' • ~:.0
• 1 l., ';-:•''~·:; .\:- :·~ '•,:.,' ',' ,•' ' '\·.' ' -.. .,,
.:.~·-~:
·', • , >.

.~ ....
'/ '• ' ,\' ·.'' t ~ ~; .. ~~,
-~··:·.'
. ,.,':' ~~·. ::< • "< ~. ~·:
?·.''
:ll .. • ' , ' .
··~ :.. .....
• • • • ·.·:.. ·.:,( •.
-;:, .. :;
: •. ·,.. . :_:·
. ·_•. ••• ' •·
 # •• ' ''( • ~ •

··,' .t·
.
52 •.-
·-.~

. '·' . :·'
· ... :'•
:-...··. i· :- .•. _:

.i'

' ~ . .
' .~

..; ' : ..tt ·>

., ~I .•. : •:

-_,
.
'K.....--..
. ~ .<:·:
< ;,;·.'
~-
J.. •.PJ.. -
. ~-
....
.. ·
~
-
1:J
~.
-~

; . ~
'·.· ·~ . .
. :·.:. '~ ,' ' ·: .
\· .... ·,
-- I •' .. ~· ' -~
.· ..

.
'
..•
··:.
r'.!' '
·.
'·.

,\•
'.

• .'•
.·;.
-~' 'I '

'- .... ' ' I . ·, ' '.' .' ~ :.. .... -~:
.·· ',..
), ·.4.. t. ~~-
• 4~

::
·-

' .·... .
..·.' ' .. !:· ,: ,. '. ·.•
• ''J' ,'
... .~

. '. ... ··
··'·.

·.• ·,
., '
.. '·
- ',"',• •'
,·'·
····- ... ,: ·.
, ...
.,
... ~
" ··.·-. -_;
.. " .
.., ·.. ··-··.'
~. '
. •, _t '. ! ~ .:.-· . .' ' .

·, . ~- ~: .
'·, ... -.. =:,.
·.'
.. j ...

'i·:,:
:-.:- • .' ··.,. }_ ,. ' • 1 ; • ~~-~

. .. ;_·.;' ... .~ ·'·'. . \: ·~ . .'-{)

'•·' • : j ~ ,, .

~here Vel--has been replaced by ·~'ft(otJ(l' Sg t) X B

~
f'rom Eq •. ( /o?),
-;Ji
.and similarly '.tor .V ~ ., Now :'11. .~:: 'R ( ~ , f3,.s, t)
' ... ::;
. ·. =- 'R [ o~- <r, t >19 . fJ c'-r, t >jj • s <r, t ), t] 0 By :tmpli(:i t. dif~e~enti.atl. on
: '.< · .with ,x•esps ct ·to time J ·:::... ··.,..
·_. .:·.)
,...
i• •. ·r ' '

-1- dX
.. i ,--;;.
· · From (1 11"1 ) and ( /d..O}.,, . and the f'act. that ;;"R..jas ..:: ~ it.· .,
. ····· ?; . . 1.,
.·.. · '·
..·
.~ : ··follows that ::. .~ ~ ~ . .,·
__

'
:· · . ,. '. :.- .·
.•
·.· '· . ''• . ', t.·

.• ll'j ' , : ' ·,· ' .',/. '·,··.' -~ ' ' • ' "', 'I '
,.·::. ~
. ,' .>
' ' .'
~

~,.
• ' .. ~x.-a,
' I -... re

.. ;'···;~-.~·;:,:B ::~:
_::-til!:'

: ..·· ........
..
~ '! ...... _
; ,'

·.·.-.
. . .. ~-}
.,.
. ..... .. .~ ~. . . . . ..
...
-~···, ..
,' ··: : . ' . '• ·,

'·~ . ·.~

'
~ ;·.··
. .·::. . ' . ~ :·..
t. ' ·' ....
~
. ':'
''
.r, • : • ~ •• .. • • ' • t • •
".. ·. . ·. ·,·
. .. •'.,
.·.·1:... '.
... ..... . ., .. ,. :. .~.~ '~--- :' ··.:.: ..
" \ .•
. ': ,;· .,. ..: ;· \.· '' \ ~.
•
. ·:..·' ·,

. : _·,..:,_ ·: ,1,._
~ . . ... ' .. ,. :·· -~ '• ~ -.
'' :
~~~· f ..;.. :, .. •• ' t. '! ~ . :·.: ,: •• ' : ,; .,.: <.· ~- :·, ~ ·' • •' -~; ... • ••• _. ••
·. \ .·.. ··v... .: ··· .
·,. !•. ; ' : ~.-·. •• • • ' ~. • . ·.·.-·
 -~?-::--;,.-r-:::-:--··-·:,::-..;-~~--=-~·--·...._..- "'""'""::-~.;-,.:._,;--~_·'u'-;;...··=.v:""":r,;;,;_'
. . ' - . .....:!_:·;
-.......- .... ... .. :;.:.-;.~:.;;.),
-- ,. •.,;::::...,~~-;:.-;:.;::,·
---.. ~:..:..··-~~-
• " .. ~.
. ~-~....;· ... ;:.~~<:~_';·
. ~-~:.~:.: .:;_~.::;·:'~·:;.:;:~f;-:;~~-= .. -~ ~-----~=~::::::"~·::~~·~.,~~~~~~-.#>~~- ·'
..
..:~ .
. . . '"-" . ,__

.
•. ·'' ,· '·. -? j

·' } ..
\ ·' I';

.... . ·,
:'

·-
f· ' •
... t ...
-~ ·, .
·.· ·"·
;_:•
. ' ·'·, ...
,;' .. •·- . ·' ,··
.;·
'I . . • : : ;,·: .}, ' ':. ' , /
~· .,.... '·•
·•!.
·.. ; ·: ,· and ,. ;· ·.·

:· .;;}?·i;, t~_0 =·L[l{'f+fi- ~~) r ~ 11t #l. c1 -~ ~ ·;tJ ;.. ·

•) ,, ':.
·-:
.·",''
·;::·" ··...
•. . .. '·':
·- ....... ... ...' : ...··
. .·... .
'I·'· ... •,•·.!
. ,_·. ·•
..... ..., .
·.· ' .,~ ·1
.. ~.
.
. ."·- ' . ·-·. ~ :·~

. ·:. ~. .·: _'·r:J·

.' ·'':'.
··;:-
·...· :<: _: ·. ·. ! ,1.:£~; ·. • .·. . . .:
...
··· ·., ·;· ( . ..:;: :· ;I;
. :::-:.-
' •: ·. I ' ~·< '• ~ ...... : • •'.' ' • •;,1.
~- ~
I !, '
...... : .. . ~' .. -~ '.• .
·;··-
. .. ···''
·,.
··:' ....
t' ., ' . ··....
;

. . ·~· ·. ·. ., ._ ....... _.·!...· . •.' .. ':•

. ·... ·.
.'
......
',1
'
.•
. ,;••,, ·'·.'.
,· ·.···.···
.
. \ ; .-~-. ·. .:• ; t'
.'. ~
.. :, .. .,.·
··.··.·.
.', ' ~·,..'I .. • : '~: ·, ' •. =: ,'
. ..
...
I' . ~ << , ._-~:!
'
;'·. '
' •i, •.

': -~~ -~·;.,:

'·,;! .. .. ·.. -:.

.r·:>/
' 1 ·, ~-. ~ •....

~-~. >-r
_, .
.' .·
. ,.
·; 4.; • ~
-:.: · .· ..•. The last term in Eq .. ( ~~~) is
·> ···.:·~·;.:~ .. 'j_ ~it
equals.
. · ::.- 1 · as at.
' .
In. addition

".,
.. ·.

. .::; ..:'~.·:Replacing a
~ ~-<
MB
v+P.d. +- . a:- > by
: :· •, -~- • •: t_ ,, • •
.
.
,, ',
... · .
. :i· ·:~\·:. ..· ffnall.y give~ .•
.·

_·,.·;.:·: ...
,'

.. ,:_.."
.
·7(J.. o_#- .
.. . .·...
:.·
':i..' ;=.· :~ _; J • j •• '·

-· ' .
·'::!.
. ;,..
.\. .
.~

,· '· ·;·.t
.F.

and .:.:}!
' .• '~ t

i ;,_

I.

~-?·;,,._.~·: . As indicated, ,'If, {0, and .B . are .function~of ._ . R :arid ·t, ·while ·
[ · :i R ~ H( o~., ~. a; t )~ .· It now. "1'.• yj;
' and B · are regarde</- as J:unc~ions . · .
: ·>:-,.-'.: .: ·. :.!':.of,::;do ~-;·-~~_.:, __ ~,:.· t ... as in Eq~:-.~< 'l~l )-~ than· by, Fq~·-.. ·( _1o:JJ ·_. :':/•::'\>.-::··· :·_ '_
_. ?. ::~:·:_ : . : :.·:· .
:_ ... "'t ~.::=... ~--~ ::·~· .......·. ';._:,_;, .. j•: --~ .·• ;· •. :·-· • . ;-~. ., i ·.• :·:· . ";·' ·::.1._·,
~-=··... :··,· ~·· . . . ::··.·-~·.:~, ·,,;·:·:.· ··,: ., ·~ ·.·.:··.·:-'.::
• ', I ~ '< : :. :; ·:_. !i . ' I I • ' ' '' ' ·._-;;:
 . ·.·:
I .;
•. • ' · j ' '.', I

;• ~
.
. .:t ' . ','
.~ ''. ·~ ...
~
·. .... . ·. . . . ·' .

L'
.........
54 ..'·'·',
.. ·•
.... . .·.·. .....
.

·.·.• *~ .·-=··1r ' . :, .

'l{l(.i,P, ·~) -~- 1+':.11] "'"~· ::r {"'!, ~ • 'tJ + (J-((i$) .: ·; . <···:tJ~6)
.....
>·

·~

• I
·'·: ..

. . . ·.'
,..

Finally. we are prepared to· evaluate !!...!!" . . . Comp·a.rison

fr
. d;r .
or ~ in Eq., (/1'3 ). with from Eq9. (99) reveals a· ·
··similarity betv-zeell do~- end the integrand; of
. dr

··'.
. .~. ·. . .
'·.
~L .:~~---· . .:.
.... ·.: . ' ~ \- :· ·.. ·. ~-.
:
... _., . ' ,-..
.:: ·. '...... .-::·\-, .: '

' .·'···:;·:.···
1:~ .
. :, ·But ·
~ ~
_ds equals. dtg .the time :for· the gu.iding_center to
. : ;,.:· •' .~. . . . '11,

r···::··:.!'
.:. ·..·.' ... ~.... ·· t;raversa ·. ds, .so ,that the last integral . in.(/~/) is the net

:,i. · ,.: '.·.': · :.:· .. oscillation..

!~
;
'
' .
...... ,\ ~ •
'~·
change in vI/

.
fl'1 • ; : over a per1o!l· of the longitudinal
Thi,s change :+s zero., · ·As a. special ·case if one
. ' '!'
. . ·...·: ~ .;_·--.
·.-.·..

·
',I
... ·,,

' .. , .
I

:
.
.

'.
' ' ...

.... .-,·.

'
'. ·-.~.

':

·::.-
. . :··::,

-:"·
. ;.

•'

. :·..... tak.es. ~he period .fro~ one mirror. reflection to the next ...... ·
'·.··
... , .:. :-. re:flecti.on at the same ~nd of the line or force, _vanishes .... :
.' ~' . :

r::· ),: ,.:;·,/:·:;·:.both time~ .-,and 'therefore. the integral .vanishes.,.· Equa~io.n (/~)'.> . '··
•,' •• I','
, ..
·-!·- ·"r''

:::. -:.:·(_··:_ ... 'can be l-tritten· as

' ' ' ' • • .. .' ·: I ' ~ • ,• , , , '
·.H""· ,'

I·
-~.
.'''
' · , .,
. ·. ' '·, .
·..· ~T( J.7 ~ ~· .lf (rJ ,(1 1 ~-t:)
'; ~ ·, •, . ·~ .. ;.· dt9 .
.
. . ·. .•. ~

~: ....
,.·•..• ·•
~~
'
'.
'.·.
(.
l where (J.) is the average rate of change of· Ql.. over a.·
·:~ ..,

0 .··.1
., -.···1
.,.

,,.
...
............ .
' .
. i·.: '

( I.:J,ll)
'

·.
,· ...
•
-~ ~
'•'
.I

:
;:} ... · '·, ~ : "·

'··
t.,
' ' -~·
' -~·
 .. ' •.' .· ·,·
' . \ "' I
' ,. .
...
·.·-. :'.'

;_..
. .,

;'·,···
·,
i ...
·.:;: .. ·
. . :• .
.•.. ··
:.~ r •
. ':·

. . .."·.·. ...
'.':·· .
"•

.-.'- ...

( /0 0)
. . . ....

',• ..

. · .?,- ~· • : ( 131) : : :tlf.:r

Equations (2~~) - .(J:3J) may be regarded_ as a third set or . . .'.;~.t.: :.t·

·' ··equations . of motionj the)'· govern the average motioii :f'rom one · : . · · .· ·\".:: · ; :A:fl
. line ot force to another:~ : •·'· ·/· .',;(!, \ ,• ,· ·}.•·" r

> · .· · · · ·.· ·~·

'·.'_.1

:.. . dir
. ·~ ·;
Equation ( 'l/ ) for d(: when the guiding centep is at ·. > •. .,. ..

some point . ( f/- 11 iJ s) can, by virtu·a.cof' thes~& equations. ·:-. . :: _.

·" '· . ~ •. ;
. ·, -:.,.
··of. motion,· be wri tteri as / . ..~ ..

+ T [itsJ - <;i->]
.• .

. >..
.. , ...........
- f ds'{.ef·t
.... . --; - ~ s) t1. ·c.s') - o1.. ·Cs)
~~ A- .

•
where is the parall~l·velocity the_guidin_g cen~er has
-::.:.'at sao - There is no raaSOt;l in general that. . ~- should.·
.. ~- ·'-
.,., ;,
k--
-. .
,_,yanisho .·_However, ~

····....
~:.·
'; . ~· .
. ·~
. ~ . '·,··

··;··~J ~ 1£,. [f ~ ~{i'~~~(•'l+~[~isJ .zc$•) -Jc·1~<~·J]J 7~ ·:: · .·.

. t<'•

. ·. ~.
:.~
•• • • ; :.. •

.,. . ~. :' ·. " .'~ .' . . . -:

• ,! '

'·. : .. •. ·~· .·: ·,.:

.J,- ••. ·-· .. . .: :~.' ';
,;.,. '

' :··-·
.... · . '"
~
~ ~
• - • . l
\
• t '·, :': .':':: ~' - -~
~ ·, ' · ......
·.
"

.. '.·::· I ~ ..
;-;
,·,, '·
··. ~'
·····. .. •' .'· .
:. ' .~ .;
:·_.
··.' ·. · ' j -

.. . ....
':
··: . ..
·, ·',.ll't.

. 1\ • ·,.' :'.:
~-

·~· ,·' :::

··.::· ....... ,. ,,.
.,;;.· -~ ·

·~ ,·. ', ·: .... :;·':··· .,·''

: •• v- ....... • '· ...... ",• i ' :'· . ·. ~ ....
 . '
56 ·;
·'

:, :;.· ·.
:
,. .. . -: ~ • •

·. n
• ' ' I

.....,·,., . '
. The double integral vanishes because of the anti;symrnetry. of
. -:l...-:::.
....
' ' ' .
. :
' '\.
'' _:the integrand,in. s and s' • Equation ( 133) means that ..... ··.·'·

a.lthoug~ ~he· inatantaneou~ rate of·chang~ of J is not zero, ..:./:!

' . '. . (~3)
....
,.. the change\, .averaged. over a complete oscillation is zero,. .
'
·')
. '· '•
'
: . ..

. ;, . ::,,
-: .. ·The above proo( of the invariance .of .. J: · has been
. . . ~
. . . . .~.~~~
.<·: carried out. non.. relativistic.ally •. The relativistic.modi~ioations· • ....
. .... . :·:.;: ·.:. (~o) . .:_.;·,·.·:
·'' , ...
..'. .. ' .~. .
: :., . \
>•·.. : . are
_
, .·· .
.

:.:;
·:-:··. I
' ,
:'•
. ·.:
.. :
....· ;'' ·.. ··. :·
., i.

'····
,;
,'
'
··... .. · .
. •'

.... ...
..
.. . c· . •.
'·,,I

• ••• •• : •• ·:. 1 .. '!

··.•. • f' '.
' ', ,I~ o' :, ' ·~; '. '

' ,._l_-, ' ·_~·

'~ >' ~ '

·•'
..:. ·'<.. l 2 &? :: . ~
I'.; • , . .:· •..

'·. ·' '; >:~. :· ... · . .:; .• ·. .

••
.• J
..
• •• •• •
•'<I''
···!
• .J.

. ~ : -:: ; ... . ... .~ ' :··~·:- .

' ', '
.. · ·''·'· 1

.• ~~ ~ ~ : •. )· . . 1, •. ' I' . · .· · • :. ·· i•
~
·:~ <'·.' : :_.. ~ ' ' ':. ' •r ·•. ( ·_. •• -., ·• • ", , ;'

.' ... .{ ~ '.. ..:·· . ~ . ; . ' ; . ' .. : : .. · ...:.. :::J :

,.;· ~. ... ,·, ' ..·' . .
. ~ ".'
' ., ;
. ':
.
,· ,·,~ '.'I' •:

·and the relativistic. guiding center equations (5"/).· ·.:.·

.. ·.' ·....· .. .:.: ...;....>I
With these~ a .Pr<?of. similar to tha:.ipreceding on.e can be carried ., . ·.> ·. ····. .;·1'~·... _;',.
:-. ;. .. , ··'. ~ . I_':>·

· <<:··., . ' . ' •t

.out,~.<:>r ~he proof irr reference (ao) can be used.... In eithe~ · >:>:
• ';,: ••:. -~ :' '
... ,c:1
., ., ~ ::

.''
:· ·····::·,
·.>.;·;:'\ .case: exactly ~he same equations of Il).otion (/'J.U) - ( 131) result e
~· ·:
... ·...
. .. ' . ,•.
.
. '! '·.. ·.
... -~ .·'
·'··'·:_.;
..··,:,
.. ",' A conve.nient spac~ in which to illustrat~ the guiding
{······ .,
.... ·,
... '.,
.. ,-:.-' · · :\'~ '. center mot.ion.:is a. Cartesian. ( cl , (3 , . s)
•'

sp ac~.- as show,n in ;,

. €!$'
.. . :. :~: •.
Fig (8!3. ) o · In.'th.is. space a line. of .forc.e appears~ a. straight .
•••:.· .3
• ~ 'j

:.:':::.-line parallel; to . s" At any 1Z?-stant . of time the guiding '. ~ :\;
•• ;.4 ••
:• ~ ·.: ·:· . :.
·,, ·,

. plane with velocity_components

. . . .
. ·,,

......... ~ and. j · as_.given .in. <ll3) and (//lf) and illus.t.rated.in

.·· ·' _;< ~
·: .:.,··,·. :. .. ,· ' '
. ;_ i' ~

. :~ ..". . "' :
• ' ; · - · ... , I
(, .
. . center mQves along •.·.·• s, ;the· direction
As the. g.u.i.ding, : '
.. ·'.
.
,:'· ~
: (
... ,~ ·. '

. .. . .'· .' . ~. ·'

'
';•,'

.-'.·• :
I .,

.
:.r.
. '\ ::. . . ~ ..... ..·. ··..
.. . . ...
,' _ ' .....
··';·, -:. '•t, .
.·' .

.....· .. .'
:· i < :. • .' '~ • ·.:
,:·
• • '
~ ,. ':.
. ~ . .
. :·· .. , r: ...... · . ......... '
i.

.. ·. :·. ', •, ·' .,

. :·
. ...,···. : :
'•.
\ •',
..
.
.. .· ..
.. .' ' . ·-~~-' ., . ,_ .' ' .

"
.... :-;'.
i::; : • '' I~, '•,. ' ,•.
•,,

'
•

.': . ....... .
. ··, ·.' ·..
,· . --;
,,_;';
~ ' '• ', .: : ' I I

,·· ' ... ...

.~ •'' ·: '_·, .' .. . ' .•.;
:! ~ ~ .~ '
. ·;' . I,'." ,•,, . ~ ... ; ' .'
 \:'
r'· . . .
··. ··.',,

...
.,
' .. 57 ·.,~.·. ··:·~•; I ! '~ ,' ~~-~
·,, '· ..... ,'· ' '· i
~ .. . .. .. ':·· . ., .... ,'

,: . I ':· ~ f,~ ' '~ • ·, ,•. : •

·.
. ;_.... .•. . . : . :~ .,
~ ~
. .~
. ..~
',· . . . . .: ~-;''~'}
>' ' I -. ;
;·.·. •. • . .
~- .,r
~· . . ,,; ,.· ... . •·)

' '

.:.,:
·:· .· ' (
. ',! . . . ' -'

~ ~1 •

• . 1., ,•
,·
',.

: j

:•.·'
'/,·,

,,
.. ·:
·'
· .. • !'•,
' .
·.~

· ...
'

... ·.. e ·..·

.. ·'' .. ·. . ;: .r . .
.
~
., ...
. 4 •

. . :_;·;
;
..
.. .
,·.•
,. .. ..
·
· '.

.
_

'· . I ', ·~ ,·

.'· {'

.•·:·
'
,•·.,
'
......
·:· .,.
·....
' ; •..
:f . A 'line of force 'and t~e ·drift velooity in a Cartesian a, (3, s1 space •.
:· '.

·. :.- .. ;· •: ..•
·. ,:,.
:,·.,·

:.·,.

···, ..
. -... ..
. . ,. ~ . ... .. ·::·f.

\' ,;,·
and magnitude of the drift vector in Fig .. (ip) change, '.··.
'.··..

~
.
since and ·f are functions _of s ~ The guiding. canter.
·,._· ...... . 1:'' • . t.. . .: , : ~ .

,· • 1. . . ."'.

therefore. doea.not ~lways.

drift. towards .the same adjacent .. ·. j: --~ (.: :·>.- :_>
. .·'·' . :· ' .···· .
•
:.: :·. ,'.,· ...
·r· ...
'· line of force ..during .its rapid, motion .along st. In a spacial: ' '
: .

case it could however.always.drif't. towards. the same adjacez:1t

lire o
·,
Considex_- for exam;>le a static f'ielq, so that . K =0.
.
. .:
.
If the·qt;fft ~s towards the same... adjacent line at all s, then
'
. .....•1 .......

: '...' .:.··.-. .'

~·.' .
. J / p . must b.e constant. ~nd the constant ..consequantly is <.J)/(1/ ·
; ,1_.,:

.'. ·...
'•.
·,
Under these circums.tances . d£- is 1nsta.n1faneously zero, by
·'
·...

:· .•... Eq., U3::t.L Such would .. be .the case .in an e.zimutnallY. symnH:>tric ... ,
' . '
. : \ : ~· ::'
. .' ·, -~. J • • •

· .. '·, .
~ ..
I

.
I·,,

..
._;

: ...:
~{' I

~ '~
..... '
.
•• • ••• • •• •• p ' •
:. ,. .,
' ' ' ~. ::·. ~- l_ •
. ' .: ~
....... \.: . .. · . . . .
' ~.
. . ·.,:.,
·,._ ' ,.::.
·· ... .. ·. ... . . ~ :..
·,
·....... .
'
'· ..
,· . ··' :·'.
.
• "t / < I

"I~
·:'. . ~::
.·._, _- .. ·. __ :,· . ~ :. '
' . ,>; ···~. ~- ':
•':
·,·· .. •• "< ••
·; :.;)
,,·.
.
_

•':' ':_:_·; ,•( ·.. ,. '

. ;.•
''
 .. ~ .

•'';
-··
. ..... ·-;
~: \
... :
/' .;
.... ·,
, ..
• '!
·' .. .
•.·. t• • - ~. 4(
t '•,

The equations (J~S) and (/~Cj) for (..t) and<!> .may•awear.·:~.<:.~··._..·

.. ' :---

' ...
to be canonical, with .J as· the Hamiltonian.
··••:.
··...
. ·.::_·not,
.
for the. rea.sol} that T. is also a function of
·-. <; ..._,
~·· ._::_··"" If J= J(~~·fJ, K,t), is-solved as K =K(~,~,J, t),·: . ;·:. ·.o:·.. ·< .. i
·'. :.-,· .. ' : .: ___ ,. ,;
..
: '·-.
\._, ·; -·; .

. . then by implicit differentiation. Jr=--(c>K/~f )/( () K/;; v-. )1 . . ~: ;,, · :·,

. - . . '~ . " - ·. ·..... · ·' ;'
etO e g . and : , . . , ' ' ' ' . · : : .' ·.; .. . : .- .· :1 .

··;'.'.-:,
.~ .
. I_. ~' .. . . .
, ..
·.; .·: ,' 'I o :'.:

~ ' :... ~' ..... ---~ ..''\:!

,. •. )"
_

.
•I .·,

. :·,.- . -·
. I,.
..
,. . '· . --·· ·.·_· ..
!-
' . .;
,,·.. .
.'/
.
: .''! :,. ·.
·.:'
; . .'

.. . . . . _ ... . . :. :. >.:::> ·.·:. ·.:~;;: ·:·: .c l3&) ·:~:~.!·. ' '

-:..:· •
. ·- '·.
I
.

··,
~-\'·:·

'~:.',')..::
.:-:.
..
·.()()
~ ~ . . -- dl(
Tt: .
-. -· . ·-~··
~ .- .. .. .

.. .
• '' :!

. .. ~
. • :::_.-,.

. ·.·· ;~ .-.·
. ·.... ·, .. - ' . ...:"•
'·
•. ·.··
-:;-···.· .. '• '.!,. ,.' .
:~ ~ .; . .-: -: ' ... ·{ ·, : ;_, . :~ ~- . ..... . ' . :~~ ·-

'•-,, ' I ~: '.

·. ,.·'
. ': . . ,..:~. ' ... ~ ... ' .. .-:·. -.- .
. ..
. .· .. '/.'.-.--:-:
'f : ·•• :·.
.•;. _ ... '· .

('<~
~ ·: . ·-.
· :. :.· -
These are canonical_ in form, with the. resuJJ;t that:· a Liouville
. ............ . ·- ..
'• .. ·, .' .-:: . theOrem eXiStS. in (ol- ' .(J 1 rr ) .SpaCe.,, See ·:reference· · (2o) .': :_ : : ·· ~-· ':. ·· : ~~ .'i ~- <,.· .
'•.. . . :. . . ' ~ :
,.-· ....
. ' . ':· ·.!: ;. . -:_ \.~ .'-~. : :. ..; -'
; . '•
.• .
~f ' _:, . ',
,"1
~ .....

_·,·r;· -··.
'', ,!>,· ·,·' ':.
.;.:·.::::>·... ::· . \).
; -; :;. The an:tasJ111Inetry. of the integrand .in Eq. (/33} b:as an ,. ~ '' :' ~ : / .;e '
:' . ;: ·.• •,.
,' •·:·.
• / O ' 0 ~ 0 O

·' ...
-~ .' . . interes_tin& physical meaning. By equation (J3:t.,) the m ntribution ': . ;-:· ·!.-,·.
...,-.-..- ·
..
;
·,..
; .. .;
.

. .' '·.
of ds•- to the rate of change .of J when the guiding center ... .. .
·:'-:--:·
:.

is at- s . cari ba (but does not .have to be, as .will become '
..
, :-.
'
·. ·. ',.,
..·
..·:;... /:

:'I,· ·.•'• :·. • ' '•

· . apparent belo>l) Considered to lie the integrand.· ~:: {• •• • ; s',. ~f.~.: ·< . l ~
i
! 1 ~
.:,:·\ :'. 'i
.
The contr_ibutiqn of . ds • t:o the. change .in·
.
·J . while· _the guiding . · :_·· · :· ·
,. . ··:,~

•.... ' center

. .traverses
. . . ds . .is. dt = ~ · tim~s .this rate 11 .or' · · ~­ ·'• _: .·

I · ·.•
-'

~,·'·· ,...; ~. ~:{ ,· e lSl

.

p 8 I• • ·J· • See ~is.·9·.·

.• vu.··
·.At :·a ·later .time,·: when the
·. /cr <:
...•·' : ~
.. ,. ···II
. vl v,II . ·
~ ·, I •
... .i -~
-_ .·'. ,. ,•
... .' -. . • . : ,t.~
... . f.'\...
··.~ '_j ~
._ ~ . c •• ~ • '
=.· ;\.-:..•
', ..':.· . ~· .;. . . .. ; ..
·;·.
' .
·''
. I
.....
I • ',
"·.'· '· .-i
! -~ •' ,_;.·' :.~ ... . •• - i .. _. . .,' '·'
-:·>_-_. ....._.:·- .. .- ~ .._: .. :: .. . :. . .·, ': . ;;•. ·,.~;.
'... ~ ' ._ ·, . -.
' ·f···;·~:: :_:~ :..:·.;; i;.. ..·, •,:.• . . . :.'i ~ \ . ,,·,.
.' ·· .... ··;'·.

~
.. '.
,, - . ,\
. ; .; .
_·.! :- .-
.
.
''
· .. ;. . • '.
'•"

••.
~ ...
~ ., ' t ;, • . ;, 1
·.~ '_·.-~~ ;:
'

.'.} ,.' ':.'-:~·- ~ . ··' ·:' _'>._ '··· .. . ·.-:'' ; ,,,. '', .. _,,•,
''
.
·:··.··.
.' ·:;, •. ';'- --! • .
• o\' -.' :. . _. .' ~ ,.: ' .-~ . .. : . ;; '
 . ,,... . ·'' '•" . ', 1· ,.. ··-::- .. ,
' . ~ .
.:·, J~
' ... ~ : ' ~
'
... .·~ .~ ( .:. .' ",,« . .'. ~
··: ·'·..
' . .
. ,• : -.i i~ ·, ..
··, .··
. ...;:..:.
.;
: lt :;., ''' .' '
. .. ·. ~

. 1 -:·1 . .:
..
59
·'.
·.
' ., .
' 'I ..

; ~ .. ··..• '.

' .... •.>·

.. : ;· ... .
., :.
. ~
ds'
'~ . . \. '
::- ' ~ ·,; :,_ .' . '· ' ~· . . "·. .,
.·. : : ~:·.~ . . <:.':<·, ;·'
'.... ': ~. ~ • ' r I • '• '
' ••
' . ' ·.~.
J
. ..
··· .. ·; ...
' ; '. '.
• •• • ' ~. t .

·, '.
J. ·.• ·';. '.~ .... ·~
.. . ,·,
'I •: i ',
::· .. . .·· ·. ., ;
.
~ .. ", '•"
I' · .. :'"\ ,.
··. ·.··.'
,, ·:. ·:.
• '· .• 1_ ... }.::

I' 1 .·
·:_. ··.. .... , : ._,.: i . •'

Cancellation of the effects of ds and ds' on J.

~ .. :.· ...
Fig. ·~
.' .... ··-~ ..

guiding center has actually arrived. at dsa,·the contribution ,.•.

'···
.. ~ . ' ~ .

or ds· to the change in J while the guiding center t~averses, . >·. ··.· ... ·, .·
.··. '~ . .: ,_·::::. · ds •
.··: v ,, '~,
is. ds,' ds[ ••• sv,· .s ••• j, which is just the negative or .,.
. . ' ·. '
:·..... ·J'.i
>J. .·.;
. : . ..~~-·· , ...
.'·'' .. the contribution of ds' to the change ,in
'~·:.
J when the guidirg .' :::·'. · · :, '
.·..
. '.·.,
I·'
~ ...
~ .·.· . .' ·- : . . .I (··.
center traverses ds. This.cancellation is an interpretation . .. , ·,. '' ·• ~
... : .
' ... .'~ :' :~
,.'j ,,
.,. ~ '. .. .f· .·: . . ,._·
·,,';,','}.'I,'
.. · ·i <i
··.· · · · ·~ . of the· anti symmetry. 9£ the integrand in Eq. {/3 3) and holds.
..
,· ...
·, ...
:·.~~I
'.'
•'
:.:···
. ·: . ~· i ' '

·..·. ·: .. for all pairs of arc elements . ds . and ds~. Such a cancellation·
c. ' ·. ~;- ', ' I
. ' . \,

... .'.'::·
·
...
.....
:·-,.·. ·is somewhat. remarkable, especially if one recalls1..that the .. . : . . ;
..
,'
', ,.,
. ·.
;
'
., .. ·
~ -~

.. '.:•.
' •• • 'I ,·/

~ ' .: .·'· :" '

guiding center does not even drift towards. the same adjacent .·· · · .' .. :· .
. ·..
•· : '
. '.
·.~.
._.-,_''."\• ...·.·. :.:;-
...
, ...... ! -:
. .' .. ~ .~
; 1 line of ~orca at s and· s'. The .remarkabili ty. dis~pl::e ars . ·.· ·. ·. ' \ ·: .·, ..
:' ' . ;.,: ' .
·'. ·: ·. ·..
... . .... ..
'.
'," ,
,·.·..
... ·..· :::_:, .....
-
;.•,

'
. to s orne extent i:r'·, one realizes ·that Eq.. ( 13 a.) can also·. ' . :: .... -; • ~1
1. .. . . t :-1

~.-\:. ~:
:', . '. ·.. ~
be Wl"itten as (for_., simplicity 1_ at E ~er..1
.. ()t:,
=0) · .. ~. '·
,.<
. ·,, ··:··.:,_·. ·... ···_:'....
: :;,

. ·. . . '
f .' •

~
~
."-~· . ,.. · . <·-~:.'1'.·.·.~:.)~!
:' ': .. ·;. :;_'. ~ •' ': '~ ' .I ' : •~ '

.:cJ31j!. ·. ·'• '

• ' ·, ' : ' I, '

J)~
. '
. ·,·
.•

~~ ..:.J
' . .. 'j 1
.;
•! . .:.;.
. . :! ' ' ' ~.
'·- . ,j.
.~ ~- .·:·~
.·-:'

... :• ·. ::

, I,
. .':.; \•·
 ...
- . .~ ..:.:' .....
.:~
- 60 - .
•, ~
'
;

I ~ ·, >

:· \. ~·!:.:· ·,_ , . when t~e s 1 integration is performed •. In truth, anything can be added to
~ . .'~ : : .
.... · & (a') a.rul J:( s ' ) in (132) which in"t1egrates to zero wer s 1 o
. '
. :,_,.
in. (137) is no longer antisymmetric in
) __

';:·_ ... ... -·· :-

s and s ' 1 and the _nice pbysicru. ·'
'.
'>.-.
'· ,•
. '
'•
inte:r:pretation is no longer presento The interpretation of the e.ntisy.mmetr;y · ·
;<,'

. '.:~ '.
.. .
.• is therefore not unique. It is nevertheless true that
.... ':_ . ~
'. _·1' ·,_·. . ~- .

~. • •• _, f
[ -·~ MB'
( - ) ~(s) ]
o
is zero •.
,-::. ;.·: ~.)::~
,, .•. e
,.. ' ;....

·"
'.·
...
.... -~- /
' ·. :.. ' c.
--_:
;. -~
The Third or Flux Adiabatic Invariant I
. '.:·-
. ··--· ·:!
As ·a. particle oscilla.tes between mirror points it drif'ts. a.qross lines . .. ' ~ '
'-,;

··:::
;,-.;~:..·
:)·· .. :;;;:··,,:·:.:··;:'of ferae on which J is constant. These lines form a. surface in space,. It ~.. ' ·~
• ' ••• / •. • l -. ,'
··-· ·. .'. ,;~
~ -~
'I1JaY" happen_ that these surfaces are closed.
A

Such· is 'the ca.se in la.boratory-_.we .. ' .

'-,''·
.-.·'.- .. ' ~- '

' ;..~ ~

'· , ~or machines and probably ill the geomagnetic field. A "longitudinal invariant·: ,_: : · · :'..:
.... .
;
' ... '· '' ·. ·;· .... ' .· .:
. ·_,::. ·.. _.._;·'. ·.;.-·(;;1
' .' ·. ~:·
surface" _is. ·ilJ.ustra.ted in ~ig. J..Ofor the earth 'a field,; . , .. >:: .. -'::;l :·:.
i·.:·.
1, . ·• . • ;t' ~'1..:
· ... I•,:
' ·' -:.~ ' {. ;.,_:
.·:'
'.. \~·:
; '
.' ... . ···( ~
'
.
•'
.· .,
. ',.' .
'.:~- j ·,. •
~ _-._:· ..
..
..
. ·
"
.. .. ,. ' •• , 1 , -;_ -:' ' . ::· • • :·· .-... · ' ' ••

;
_

·' . . ........ . .. ::·

;.•:
<\• :·-~ ::••• ' : ~-~ • •

!':
... :
... } ,..
... ,.. •. •'
·-,; -··· . -.: '·' ..._... ·~. -
..·.::•.'. .·
:~·
•"',.,
. ·~ :

::.'' I •o: ~j
·_;~:- ·.' - .
. .
~ :-. -~ ;:-, . . :;· . ... .~ .·
L(}n,ji-i:Mc:/t;us{ lnYon~/1-b $G.'"foce : .. ·
:_: • •, ;' '~I ·•' '
~.- .,: ~--

·...... ' .. __ . : ' .... ·-·. '•1

I .... ..
...
l'i:
:· : -~: . '.:·-.:·.> ·'
.:_

·•:,-: ::-_... ....... ·' 1.-.

. ·... ·· { .
._., ··;. .
<. ••. •. ·,., •••• ~ : ..

' .
• I,. ,<' ~ .• .
,·
•'.

., . :
i '• ;;-··.: -·

': ..:·_.

.. -~ .
·...........~... .
..·;. ,J,

·A longitud:tna.l.invaria.nt surface .in the geamagnst1a 'field. .

,.,.
.-._;-
.. ;-

:
... :·.-··
.. . ..
-~ :
- ...,
. '.
I· ·,' ',
·~-
; ... ,._.- ~-
.• ,•; ·''::·.::-.-_ .. · ... •'. •.···

... ::.·: ::-:-;-'•. ..

. ·: '' ':. •-:/::·• • ;- I ..

'.·_.:. .~' ' :~ . \' '', ,, , \II

I •'"

... . ·,- ·.· .;: ._: .'·-. . ·'

. ~· .. '
.
t"\• .,
,( .,
.': .· ..' ... ·.:_~- ,;.
:.·- • ' l. : . . ,, -'~
. '' ;'';
·.t.-:··
; ~· t" '(,
·. ·.··_. ...·
. .•-':,:
.' },','
.... ·. ·'··: ~. ·'. ·. ~ :·, . ·." _::, • ..- c.· L-' :.-:. ·;_.··~~~-- /_;. -. : ·" \. .-.~ 1
·• ;'' :• '
 . . .. ~
-~ :. ,, . : . :. ·,
- 61 - . .!

' 'I In the Cartesian _.!.

a, ~~ .s space of Fig. $a. longitudinal invariB.nt
':·

·>~·< :.: : · surtaces are ,.curved ~ one direction only, the elements of the surfaces being

. ·. :···, ;/· ...straight lines parallel to If the invaria:n.t surfaces are closed1 they s.
~; ·./.. -...: ._; '
,.·' . : ·_ .\·. ·~r. :~]
·:,. :~·.',:.
~ : '
.~are
.! :
·<·
; ·: •
represented by cylinders in a, 13 1 s ·space as shown' iJl Fig. ll a. Tb.e
•
. . ~-··

.... ·· ".·'',\
.l.,._

'<::. >: ;·elements of a cylinder are not all Of equal length because the distance between .· .· · ...... •, .. • '
'
:~~-~--~·.:~·-· . ' '
..... -~

: . •'j
reflection points is not a constant of the guiding center motiono

·:. ·of a cyl:tnd.er With the a, ~ plane appears as a closed curve. in Fig. llb .•. .:.· : ~:·_: .. . . .. ' .)
;,.··..
.
)

.. ' ...
'l'he intersections of the cylind~rs with the a, ·~ plane form a three · ~' · :::/.·. :,-:: ·
·,
:
;_ ..

:·' ,. -~
·.. .::
:· .... .. :
·. parameter famiJ.zy" of curves. The longitudinal invariant J is a function .... · ' -~ .
. . . ... '
: ., . ~ '
,·
,: .. '. -:_-;.._

J(c:t,$ 1 K,M1 t), as can be ·seen from the defining integraJ. in (96). The M
.,,·
', ·'' ' '~. • I

~'s been sui>pressed heretofore, but ex;Plici~~ · .:', >~

i ·.•
'
"· ·:''
'I;

--~
dependence 'Will now be exhibited -'. _:
·,

' . ' '' . : ·.'. :·_!:· <· 'l

T.fle equation J == .r(a,~ 1 K1 M,t) can be'solved as a= o:(~ 1 J',M1 K1 t),.w~~ e.t'·: · ···,i.· : .. <.)
-~ -. .. ~ . . .· . . ... ·: ...)· .·. - ~-_:.. f/ "_ .· .. ;~I .
. ,.
' ......
;
•· e:ny t:illle defines a ·fa.m.ily ·of ~~ with J, M1 and K as the- parameters.
. ::.:
' ;
•-:
< . ..
..; .. ' \. _. :, :~'
r
. . . ' : . i . ~ :, ~. ' •· f;!
·. :. ·>·:· .:·.; ol \.... . .: ··.. ;,:· ·~ ·• ·: ... :. :··. ·..·• .· .
'
..
' Tt, ,·~,e.,./,; nc+ \

. -·.
.-,:
' ~-
" : ,· .. .....-- ?t<-lf't~ll~ I -t:o t"lu
...
:_.,
•.-.
" ',.'

.-,' · ,J.- (1 pta.u

.: . ' . . :: . ' ~ . '. -~ :. ' . . ·.... ·.·.,
···.::.

\ ... '· ·• .. '

. -~-· .:. ' -· ...., . . . ~~
, .. ''. .. _.-·.
:· 'J ··.·, ••
_:
··-·
·n.. ,-s e ,..; tj ',,. t·k ::::_.:·~..·
/ ' rJ., (3 fie.""-< ' '
't .• ').- •

. r·.- .....·. ~· Notl t'"O 4r~i-t$J~(l.. , ''··... '· · · :;\·,:

e . ·. :.· .·:
Y'

~~----,-«--)----.~.--- P .])()fl ~ 'PC.·: cl~-t,;.,,;.j .-~------(}:;---:-->-.---

Oh

A longitud :f nal in-variant surface in a, ~;, s 13pace .: , . ;' . ' '.· : J:(
"1<4 ,..,. l l , . :·.: . . . . . !:A
.... ""'6. ' ' ,· ',.' ~ ·' ' ' .;
., .· .. . .
'
\
,.
·.
,'.'\.

Suppose that at a given instant the time-dependence of the fields were

turned off; K as well as J and M ·. . would be constant and the guiding
. ·. ~

. center would precess about the corresponding (J1 M1 K) surface... The third

:·< :
. adiabatic invariant J is the f'_lux. of .i through this surface.. That if
is a constant is a triv.ta.l statement if' the fields are static, for the .~iding
' .-·~ -· .. •

·~ . ". ·. center repeatedly precesses around the same surface and the .surface does not
· ~·:,·: 1.> .cbanga with time. An anaJ.ogous trivial statement would 'b~·~ha.t the ma~eti<l -· ~

.:· . ·:,)"<: ·:.. ·.:moment ia constat).t. in a uniform static· magneti~ field. I:f1 on tba contrary1 the
. . ' .•
·_·:·~·::·./.: ·>:;:.'fields are· time dependent1 the time dependenc~ being slow compa.red to the
' ~
; ...
. ·~· ~
. . -~~;
· ·. .' :. ·.., preaession time once around th.a surface, ·tha . •. . . · · . . .. . . · .
.. .. .
~· .. .-~
.,
-· . ..... ... . ... ·.·. '· ,. '
.•

: :..
~

:·. . .. .. . . -~-- - ·. ' : '_.: , <".I .~. •' ' : .• ', ·' 'I' -~ ,
,
•. ·.( '
. ~ ..-. : .~·. -:~ . - '"' -.: . . ·..... ·: ·.. ~ ..
' ~.
••• l. . - ~... ~ ·.~
:. •·. '- .: . ~ . .. .. • i ' . :: ··.... ·~ :
 , .,_,

. {.
I
i
'·
..·. ..·, .... -··,
..... ·.

I
i 62
~
•.·' :.
,,-{. , I ,•
... ., ... ~:
~ .: •;' :· . - .. -~ :.. •.
. -::_1..
F·· .&- •. <. • :.'··.
;:., .. -
'.; · " third _invariant applies and yiel-ds non~tri vial inforrnationo · · ,., __

.,_ l,t

' ... Since .K

-~ .:. ':
'

: . ;..
'
"
~·, • •' I ' ' • .. moves- from._one. · J, M, K ~urface to another. as it rapidly_':· .. ·
-·~- a
'"!·'·
: ·. ·'_precesses, and is at all times on surface:.. with the same
. .· ' ' ... . . "
·Thus although one oo nstant of the motion
.,
i,t. . , .. ~~
:1
f
I . .
,.·:-·:~
The proof of the invariance of is fortunately muc_h .. -.,
' . _;,
J.,. ·The flux through -~he inv aris.nt. ·_'. -·_ ·~·
. ::
··· ·· ...- ,,. simpler·. than that. for. M or
-~. . ~ . . ' ·.. •)
:
'_t
,.,
:
"
. .:. -~- . .. .
~ . I
-· . ': 'j
( .·1

·,·.,_~ . ·_.. surface_ is . . . . . .. \

. . ; ~ '. ': ' ' : .i

-.~ . '' '. . . • ' :..·~. :. t

.-~ ' . . ' . '
.··, · - . -r - ..:1
~(Ji14,·K,-r)=·fA.·R ~ fo~-ve·P = pr;J-(P~ 7;1"1,~-t.;et~. .. ~ .
. ....
.' -~ J
t/3~).
: : '

-
• • • '-

~
~ ~. i
.. . ;--~
;i
,. '. . ' f • • • ~
' ....
.. ·, --.·:· .. :.. --.•·:··
'

.
•·• where the con~ou~ is any closed curve lying on -the surface ,•. ·
.

(in real space). and going. qnce around 1 t.

.,

.However f r;~-de
.. ; .·:.:~-

.i . __ :

·,._
. ···. _- ... ,
····•'j
·,.. ·j. -..
· ·., :.~ --~- is the area inside the closed curve in Fig.,/l-b on which .. ' .-

i_• ->~~~-<- ·.··the g:.,iding center is located ..

.!"' • -~· ~ ' ••

.·:.'~ .. ~ ~-:. , ' ....... ·.: .. ·-. ·.· ..

;.·.. ·· ,:.
. ..:· ,· ,.
. ;\ -~ . ''·,
. I "':--

. ..
' ; ," '

_
~ '•,

;_:_.,
'.

_-~·.
•

\
• o

_-.
··cuJtt). -·_.: .·:
••• !·
-·:·-_,.
'· j",.
:··.!
.. -:.·.
,),_ ·,_.
f .will not be. carried eXplioi tlY,':
·,·
.,__ ....
;'' -,:.
The J· and

since they. are constants ... · The average,

M . dependence_· of ·
.
<K), .
has been used .
::1 ·.·:· .
_,_.
:-;

I ·: ... ·
i •o -.
·.·;

''in place. of_ K,~ pecaus~ . . :..

i
t
I ·• •. ••· : '.
,-.
.
_we
. t1re not
.
interested
.
in fluctuations . ':-. :.. • t- :·
I._· -.• -
1 J.
t
.•
.
·of ..
.. ~ .
f over a· l.ongitudin~l ~~c~ll.at,ion~.
. . : .· ; '
No~. • . :·
.• ...
,.. -.
.·_·.,_
. ~··

!
'
. !. ·; : . . . ~ \ ·••. •. 'I : •.

. ',: : . ·....:- . ----~\;·: ~-- ·...· ._.. ·. -~-.-~.:.' >:''·: ..

,,~- . . '•

' ·_ ._-.·_.,_._: ·,- _. _-.-------.~ ~--·.··-··' ...- ---:----} '8;(

() rl- ( {1I' /(,I~) .. ; :;' ' .~- . •, '. : .. ·
j

~ · . ·dP,:·'.:·:~-_,-;: . :_:_ .. ·
' ';'.,
.• ... . • .~,-~-'
.-.
~ ··~ . ·. . '
.~
',

'.' ::,' '. '.:1 } ( .·' : . . .• -- ·-. '"•

J
,I
. ·..... ·~. '\ ;_
fl -. : ,: :~~·.
.~ -~:,.
. . :.• ~:·-: .. \' .· .
..... ; ; :, -· . ;_ i. . '· ' .·
.· .·. ·.
r;: ,_ ...· .. ·...'. \ ~ .·- ..... ~.
,·. :•. . ': •,. . . . : _J: '.-
:~ '

. ~ >, Y,
\ . I '

:,·.:
' . ' ~ :'. '
:: ,;._:.;_·,.._:. 'r ·' .__ '.·.,.:,··
'l :· ,: .•.• : ·~ ·. . ' .' ' -~ ..
·, :_; ''
·- . ; . ~ ~-.' ·. :.·. ·:.... . . \. ' ' .·..... ':·.:.
' . •;

I, "<, •' • ~ f; , , ' '• . ' • •• ·.' .

.,.1'
.'·; -~. .• ~· 1,' ' '
: ! ... ... ~ • •. . ..
 l ·:
.' ... '
~

~ ,. .' .. ..
. ' ' . ' : . . 63 .....

:_, ·:_· .. ·' ; :

.··>·, ·.··
... I ~··:·, ;~ \

...z: ,.,
. By irnplici t differentiation of K:- K ["': ( f , K, t), (1 j tJ ' H'
·'
. ).
...... <,:;
' '
' ' '< ··,
. .. . ~ ' .. ; .;

• .j ~ • we obtain . •. >'

'';:';· :,·'. ·.;: ... \:

·:.l
t):
(.
'" ·.:... ·.-·
!·.

''
.. :·.
..' ··.;·
!. ''' . ..·.·...
.:.·... _.:.·;
d(3 -
';\'

;~ .:--:·
. ; . "··.
.
'·:··.·
•,
~. '. f --~--

:::.· ·.: .· ,: ·. .·. i:he last equality being via Eq. <! 3'-).
1 ...
. •

-~>-·_:·.·.which ·is. the· tiif).e T .. tor the guiding center to ~ ec·ess once <..-:~: ,...... ·: ···::-.
. '<· P· .,.;.: ·. ~- : ·~. : .... {
around the surface!> Therefore l!-::; ~ TP,- and this the analog·· ·.··. ,. :,'I
• .:·*
·,

...
" :,,.)i·;·
..:;
,.

ot Eq., {/'3/ ) ., The Ja st term in ( \3~). is ' '. '·'·) •j

.' '· ,;.·
..:.. ··

· ot: :·.
o! = f~{ r)t: . :
P,l(,-t:> d P =- . ~
..
f 'd/(/Jf::
olflul-
~-" •
. . ·.· ···. •'

-t • •·• ~;.:;.
·--··-·:-·.:
,·'".::'f··
we : :· '
.. :· ... ...... : · :.;;_·::...
j.'

,' .. ·
• ·J . ·.,• \·.' ,.

·.··· .::·1
.•
.. (/4,S'):
. .• ·r. .
:'·/~·
,..,:.·.
' '·, : : ·-~ ' ..
. ·.:)

•.
;''·, :;·_::. .\

:.> . .·•. • .. ···. ;.

~: .
•:
.
•' •• t,
where ((:K'>) is the average of· (K) over... a pra~ess_ion period~ . · .. ·.. · .. ·
. r·,. ...
·Equation (/'IS") . _is the analog ot: Eq .. (/30).,:. The .inst~taneous . . · ·.i·
· ....:J:.''\. ·,•
. -:~. . .•,.;. '·, . { :. . .
1: ,•.

'\ :·.-· rate or change' or ! ', is therefor~ ', .:·.. ,:: ', ,·.-:; '·.;.' ',. ' ·, f. . ',.' ~ . ' ~ . ,.

'.~~. ~. . ·. ;_
• • 'l• ,, ' :··,-,, ·>, ' ! '•: ,.:'. ' : I . ' L ' ' : : ,' :.·· ·.. .. ' ,·.

1~ ~.:[~b -@h?i'
~ ·\
·, .•c'

> ' .. ' > ' .> ' .· .. .-.. ~ ''

. : .: .·

....
.
·'
.. ,'~'.
•'

,:.
.·· .
..,...
: · .... ~~
;:
. :·. ,·
. !,'

·i'
··.
•.. _.,. ' ' .; '' :-1 .. ~-.' : '•
•.'' ·. ·:.
·.' : ~ . .'· . . '' 4 ... · · '

' ' : : ' • .:. .f ' ~) ' : ~ : ' .• ' . • '. '. .l •• . ' . .·: . • ,' . . ' ' . '; ' ' • ( r, ·•· ... '.·,-:.
•.•·
,
J',:

I • ': ''

>. ·.: ·~.·. ~ '/; ·; :

' ' ' ' ' ' ,•, . ?·
. : :.: . . ·.: ·.
'' • ''
.•
.. ;. ' • . •.• •.

' '
~ •.

' .:
l ~ :. ' ~ ·, "'
·.~; :: . ' . '
• '' . • '

•
· .. \.
'
·'
r·
f
... ·.:
•
• ·. I
''
l. •
I) o• • '< , ~ ,; • ' • ' : '· ~. '. ..· ·.•.
,, I \i
'_.;· .·;; .· .· ....:
.. : .... · :_.·.·. .··, .
i·< · · :."_. \ . . : _: . <';~.:·' ...·:.··:··:_:·': ... ··::'
',;. :, .,..-·· .. ·:r: ..;···: ..
,_·
; ,'• ._
··.·
,•

.' . ... . ' . : ~ .~

.~ ' ' ~ ·,· :. , ; •\ ,
. . .;
·~· .

·.' ' ,:··. ·> .:·: -:; .:· · ._:· .· . . ·" ": ·. :.
' 1 , t " :. • ' I'- · ' •.. !' , .•
~ '; . • • •
. ..... : ..-·:
•
·.-L···~·
: .•
:..... ,'
• •' ·,
·~ ,,·, · . ~·· ·.. " .. · ..... : ,,. : ·.: ,·:
• ' ; • • .: J • ". ·, • '
-~
>•
' .~ . ' . . ,{.
,I.;,
... :•,' . ·.f ,'; • ' < ., < ' ~ 0 • '•.t '• ',I \ .~ ~ \ 0 0 0 0 < •u: ' ....
,,
.". '-~ ·.. ~ ....
·.· .. '·•. ·' ' .~

' '' ~ ,.. .. .." . ' . ~ ·~·

·...·, .. :. ·-: .. ;: ·. ..
;
.···ri::··...... :,:'i. • .I
",
 . :·'! ···.·
.. '•'

-
:'•,
64 - ::·:.· ··.··
..... :.·;..:
. ' .. ,
.,._,
· , f..•

. '·
.... ..... ,. •,. J
....

A's with. ::- , we again find that ::- .-:f O, but that .( "-:? >· ·:,_ ..· .· ·, .. . :-'.
·. t_he avera.ge rate .or chanr;e 'over a precession p.eriod~ do.es: ·:: ·· ·
. :. .
·~ ' . ' .. ·.. : ...
~ ' .
~

. ' . <. 'va~ish_,_ '· aga~n by the_ anti.~ym.metry _or the dlntegrand in: '.,I

.....
• .
··.· .··
.: (~),.= ~ f' i!::.' [<;() . -<fi7r.r''] ·::; ; '• '.
; ':. (/If 'I) .. ~..":'
:<.i.

'
,•
: ~ .
... ~:'' ·. . . . '1.1' . q- ,· '.:. _· .i
,

.. ·...:·.. \;·,'. ·.
'..
.. ri-p . . ) ..
.
·. . ' ·. ~-.:.': ,' . ·. , .. ·. ',• •,

. . ·t·'
'
~.'. ' ., ~
~~ . :; '. '. .. ·:,·'
·'·'
...
,. ....,' .
··..
.· ;, . \ .·::·. ·... ~

-~·
1':' ,:·. ·.·.

· ·: ·: ·,.:·; ·. ·
•, ~ 1-, •••••
where d~is ~he element of arc length.:and v /fJ is the · · . ; . ,.
,.

'· .·:.....
· ~·.. ',· .::. _,;elocity at '
·cr.' about the clos~d curve in Fig.#.b· ·~ .
' '·r ~
... ;, "
' ''' .' .' :,
. ',
:.
~ ~
The. new ·and final set of equations of motion .is· .' ~ ~.:
r .-.,::,· ·,,.·.•,

. ,;.:,. ·:. ,···:. ·.

.... ...
.. .
_
r , •
.····:
. ·:·.. :.'. __ '
{!~e). ;··.;:
. I .
.' : . . ....~-. _: .. . :) .' :' . -~ ' . . .. .
._
, .... ·,·:·.. .....
. ~ '
. . . .·', ·'·.;
·>.:.-'1.'_\':•
~ ' .
. ···:· \ .
\· ' i . .
.. f.:
·. ·_,.··.
•·
The '::· . .' · ..

.: .. ·.· ...... :··:{:

... _ . . . 1 .

·.:: ·.'" · · . adiabatic invariants and sets of equations of motion are •. '/ .. . :: '.'·. :. ~- ,\
. ··. .·. ,, .·
·:·,·summarized in·.the following diagram. :

.... .
~- ..... . . ' ·.. ..
.. •.
·. ,. .... \ .
• . .i. - •• _ ,.

... ., . ·, . ··~:· ..
.
. ~ · ~·

... ,··. ;'t·

· .. •. .
.. ~
'' . ''· ... ":'
,·,,. ·.·,_., :_ ,',
,.
'·' t
•'
'.· :: . ~ .:.·;
. ' .·. ;' ,·.
·.•,.. ' : • ·:. :!
·::·.•·· , ~- ... !
...
... . •, .. : .. ·· . ~. ' .. ' .' -.. : <...
. :: · ..,,
-~ .
.'
·:,_,',_I
.·...
.... ~.
. '',',

' . ; ~ .~ . .
•' )
:··~ !'

~
...
....
• .' i •. . ' .
·\ ·_..'
... · :.._··:.:
·.;..
... ·. .
' _·· ',: '
.. ,·
. .. ·.. .·• .. ·...•. · .. ,
',·' .
·

~ .'
/'

. ·.
.. :. ' .....
.... I,_•·•• ·:.:.-..·
·....
·.'.
,.,
' .... · . ... .-··.:: ~-~ '

·:·:_. ...
.....· ~- ... '·:< . '
.• ,•
J . . . .

. :
•

.' :r,.,,.·· . '

) ...l··\ _;\ ··t·

""...
·'' '.·:.·
.• • .'.. t
,.
~
.,::·'
''·.· .. ·
, I •!
.~-
.
:•.· ..;·_ ·..
,· ... : . ·~
••• •• 0
.. · t • ' ,' •' . ~-.. ~-· • • ~

.. . .. ~
......
.'
•

.. .
:. '
. ~·
~-. ·: •.J· . .'· ,. ' :. : .,· .~ ' .. , .. ..·
..
, I ~ :, ; '

,(•: >.
~
... ·. ·.. ·. ·... -.; , ·~.
.
,,"
.. ·
. . -~: ." ··:. ;

. . ,.
. ., . ·..· :- .: '.

.·
~· '
. ·-·-/·····
'
··. ;~· ·'\- . . '! -' • ...

~;.·v·
'
' ...... ,··
. ' • r--. ·•. .• ~
']•
·. :.·
... <: ~- .. ,... ·• ,. ..

.· '·
'.
..·.:.. ·.'
~ '• . . .: ·. . .. .,·.· ..
~
!',-' ·.·:.···.· '.··; • . ..;, · -!
., ·'. ~ ·. ·.·. •. ·~
.... ~ '' . . t,' '•' .. · "
 ,5 . ~ { ~,.._.

'·/ ........
'' •'.

I' •·.
··:., .•
... ...
65 ~- -~-

''
(o
....
I ' ''

~ ; • : #
' .

. JEq~at i~n or mot. ion] _._A_v_e_r_a_g_:;Je_o_v_e_r..,. Guidi~~ center equation.of ·.. ·...
. '

ps.rt~cle
. .
motion (/'7) and (~0),
.

·
•. !";".'
..·' . .
lot. .a '.. .
· (l) ,
. the. gyration and the
: ' f
;

·• , •.• , ·, '., 1i .• J; . • . ~

-~::·~--~ .. ~; -~.
.'

. ~
·,

' ' .
,y • • '·

' : ' ·,.· .: ' .' .. :'· ...

. ,· ••

,· '·,
... adiabatic invaJ;>iant · N
··..- ~ ..
••• ,', ... ·: J
' ;~ .' : . ' ·.
I ' ~ • f > ~ , ~ ! '
·- ~'I ·• •

. ~-;~ : -,·· .. :·. ·. :~-. ....·

• , I' I~ , ,. • '
' (
. Average over .. '
:.:, '\

i.' ! -~ ~·<'. I ...

·.:-::,. the longitudinal ·•

. I
·., .~ '. . .·
' .~ '. ·.
oscillation
.,
• ,t',
' .
;'
Equations· .or motion Average over .the . Average guidirl;g
..
. ·(1'1.~ ) ~d (/ ll1), and
~~
'. '•
l
'·
•..
. .,·. ·,~ .') ... ~ precessional motion .cente~ equations
'r.: ,_.•: ..

'.·
of motion ( /3~),
.
:-:·· the adiabatic.· invarilint ~
·... ,. I' . '
·,-_'r·:. .'·.:-·
...... ·.·.
i;,' '; .. ' '.
~.- _" ~ ~:o· ·.I •, •
.,·
' cnva ~I' fA~ ;r- .· . : .·

. The averaging process used to. obtain the three adiabatic invariants .and · ... · ·. ·. · ·· ·. ·
,, equa t·ions= bf motion. ,,:
I • •
,'I,

'·•
.. . ,....... :/yt ?:-::.:
Fig. )12 ~
....... .
.
, ... ·':·.':. ·. ·. ·......
A few fficts about the family of longitudinal. invarian~. ·.·.r·
··, ·!

, · •• '•,"· •• :.. C I o ''

~ ~.
0

.·:··. : ~urraces shoula be made clear now ... In the.first place, at . ' ·: ' . . ' .
..
.. ::_ -:
, r •.. .'
''' . '~ .
: ' .' .
'

'~ :
any instant " of time .they are not. simply .nested
' . but intersect ~ ' ' ~ • •: <

~ I.· •
.. , in a. very complex fashion;, this is. to be expected. because '. ' ', ',

.,
.... ,·,
.,,
there are .three parameters,
. .
not·. Qne.
. '
For· exarnple, ... two particles
.. \ ~ ~
,•'

·. '.

with the same K which are. oscillatint;. along the same line . :~:;.. ;
~·of force, ·.but which have different ,trlirror points, , hav~ ~i!'~e~<:mt M ).:: .. ·
. ~ ....

. ..... ':· ''

~
· '·
·_·.. ·:
:· :'
", .....···
I,'· .:•.
:-:::. ·.
~-.·~~/:. force a.nywher!3 else·.•. Fig.,J.:} . 111ustrates.how. th~ir.two J· ..
. . ,',.

/<.' .; · -:,: invariant· s.urface~ mgli1;·.t appear 1~: ~he : o! ,·~.-~ .· plane o. · (Of. course
r~·.,<:.:.:.···'.; .·. '. ·:..:. . .... ·· ·. ·:.·.-;.: .... :. . ·: ... '.. .. .· .· :. : ·.. ·. .: ..

r.r:\··,·· ..•::'\ . ·':· ·....} ' / ,: . ..... ··.' . . · . . : •. ..·.. . : < ; ' . . .• ..·. .. • ' . ·< ..•.
l ~ I :· .~. ' I "\,. . :, • \ i . : . : ; : ' .........

.... :_· ·:y:.·.·.;>···..... .·' ,··. i:.' .... , ·.:.~><·:~:··'''

~~< •'- ', • ~, ,• ~~··~. ~·:·~-~~·<.: :>,;'.·.·:.. ,·; j·:: n,, ·, ,·\ ·, ·,·, ·;,·,,·.,, ,'\•_·,:~• ·:,:. ~ •,,'• '•'' ''•:= ... :.·. ::•;.", ... '
... ··
 •:··. ,:. :.·;_.:I.' .
I - • •.
... • ~ ·; • i
,.! •••
.< ~. • I
66 .... ·.. ·. ·: .•. ·.:.:J.::.::
. ··' ., . . •' ' , • ·..'. :I , • . f~ •,
···:·· '< •, ' ., ' '
i '· •• • '.'
~ ' .. · " •;: I

. '
. .' . ~ '
.-·;' .. :• .. ;·

..
',(

.' ': .:
. ,· ··: ·""··"''
~
>
'
. ~: .· -:· . ·.. '. ,·r'_ .
.. ' t •' :. !. ' •...... 5o

.. '
. ~ ... .....
·.· ..
.•' ··..
. ;\· -
. .. : .. ' . ~
.. :i.

..
·.··'
' . ·.:.. ~ . ''.
', .,_ .
' ·,
. ,·
.... ; ',: . · ·. ·. -~..
,\ . ' .. .. . ' . .. .·~ .. . ... ~- .• '
.... • • ' ' • ,• • . . . 1 •. ·;.• ' . . - ..~·

Longitudinal ~variant surfaces are not simply nested and may intersec~~·.... : ··:·.· ·'·.. ,'· .. ,· ·;; :·~
' ' :
: . . ·~ .
.' ,, . -~ ,' . '
. , .. . ..,: . ~;. ... .:, . : .
Fi g. / '="'
.J'·.
.. •,,, .• ; ; V' : • / . ;'/':~;
:.·. ' .. ·' . . : ... ~
r· ;.·. ·.... . . .... ' .:· :'· . . :. -~- ·;;· ' .
.•. 'f
-,,'\_ ) . \ ....· . . ...
., .. ·.,..·. .··.· .. . . .-.::
#,, •

t.·' •;, I''

they !Jlay intersec:t, elsewhere; there is nothing. to prohibit .•. . ,._ .,

.· ... '> • ~(.'

. .-~

•,' ~ : .. ,. . '• •''

..: •'
~ -~·: ·. \ it in general.) An infinite number of surfaces therefore ··. . . . . ~. ";~
' '

....•·. • •• • h

. '- . . .
:'·
·,
- ••

.
r ,•
:·~ '
inters.ect along any line. of force.;. if particles are injected ' :. ·:: : :.: .. ' . ;· :J
.,;,;' • • !,
. .' -~ ' ~
._,·. .. ~ ;:
on one lin~ of. force with a distribution.of mirror ~oints,
':

they sp:read into a layer .of finite :thickness elsewhere.

~

.
f t ' f I ' 'i" •

._· .. Seootidly, in.the . ()1. f plane a surface denoted, by .fiXed

t'·• .
·•' .
1 .

' .'
values of J, M, K will move in time as indicated schematically .. · ··· · ·· ·'·""... · .u . . -~ ~
In . . ... . ...

. . . :·;,. i ;:< ·<:.'. ·~ .

in Fig .. J.!J
. A
, curves .I and II. It is clear ~hat. the surfacE> l ' • . ~'
!.,-.' I .; .. ,·.. . . :_ ~ .. ·~
must move, beqa~se o(. equals o(...( p, ·J, N, K, t )· and therefo~·e .: . :'·:. · ·. ~ :...<·.·~
. ·. . .. -.--:·/
.·:·':: .·:.
the valu~ of o/...· for a given e varies with ·t at fixed J 1 11,·· Ko .. , ·/:~ :· :·~·-.··:· ~
.
' '
..... ~
The fact that a constant . J, M,· K surface moves. with time '. .. . : .. ...' .
....
.: ' ~
. ·.:
. ';'; .:. ··. ~·
' . . :~. . ' .
.. i ..
does not mean that the lines. of force move . in o/., (3 , s space~· . . ' • .' ·. ·, 1.~: ' . . -~,: ·~:
'' ' ·~'
. . .;_:. ,_..
~· .
'' . '·, ' ... 1 • I I ' ·, ~,
In fact, they can be considered fixed, even though they move '':. . :.. ,.··:_j.J
in real. space ... :. The velocity ofA magnetic· line of. force is_
- ' . a. . ·..
. .' .....
'
. . ,: ~:.

not a physically observable quantity, and therefore must be •. : .: :<~·}.:j

· ·. ' ( '1n ) . : . ·...
. . ... ...-:r ... - .''i,• .
. . defined·., ·The usuall:de f'inition. is. the "flux preserving" one.:
1
.... : ... · . >· :::<
• ,.t

.:. let an. a'rbitrary closed loop be drawn in space a~d let the
..-:· ·;··

' :.:. >: ~--. · ,; ' • ' ' • 0 • • •• • ••• • • •

.....
. ··-;
. ' .:.;
. .. • I '•,

~uo.ru:.that~:th~fi ...
. I ...

( '
.
. velocity ";! ·'
of each arc element
..
of the loop l;>e
. '
. ~ ' •.· .•·
• ; .I
• :;._ l''t

. ..,.: . - ....
. 4'
£i1~lo:C....~Be; through the .loop is consta.n~ in timee ~e v:eloci. ty .
' . -~ ~ ;
of the line.' of ·force at each point ~s defined as . -a. It is . ·. · · ......
: '

muatt s~tisfy fE 1- E X B) =·o •. Th~' ..>.·,, ·_::.

. ./

"~ • -...
·Vi ~. ;., • • · · · . : . · •. -:---.. J,t;' .·, •,•
..

•
. ~

•.,:·.
. ! ,.
'·.· . .' .·
·.·1,
.. .
.•
. ' ;:./:· .:.. ·:·. : .- .. : ..•
} ...\-.:~·-... ··_::·:··:.:--·::- .'!··. ---·~:·.·~;> . .·~.:
\

' .. ·. _:\'.: •··· '. ( .. '.··'··· \ •. ~ . ' .••• : j' ~

..... ·.: . '~- .t • . :.~· / . J
 ----~ -~:·-

1
-'+'••"-
•
";'
'• ,I
•. ·:· .....
:-
·.'
' .'
· ......
t" ,.· .; •. ' •

'
'\ . 67 \· .. -~- :)
., :·
... . "~
• J. ••. • ~ #. ;;

i·.' ...,
.,;
velocity is-such a velocity, where : w
B
;..
t •• .
as ..defined previously •..
...
. ·., . .
'·· ... '.
•.'

'· -~ x a
.-
~

''.
'•' E + 1 ,. '

~B
...... "'' :',~
(/~$' ) , . ·E ~quals
·/ . ·, .. ·:
' ,. ..::,::
so ....
.. ' .. ;'·:
.,·. ' .· .~~- :· •
' !
. ., '·
~
·:•:'
;·· ....... · ' .~ ' .
. ·, (

': ;::1
..:,
'····· X B .::
. ,.
J -: • •
·.. -.~ ; •• • ••
. ,., .
: .. ' . ·:· :' ~ :··: : .

. '- '. ,_·. .

. . ~

·.·. '\ ' .:; :~. and indeed the curl vanishes~ It has thus been established
wx-a ...
·that J can be· taken as the velocity of a line of force.'·.. :>:_ :_.' ·
B .
. _., .. It will now be·· prov.en. tha~if an observer moves
''c. I

·~ ~· ·, .
~ .- ~ ;. ' ; . :-: tl~ ;~ t ()
~··' ~

:: ·' •' the total rate of change of ot. (and (3 ) he·

. .
observe.s is zero. ~ ·.. } .' .
I ~ : ·~:
r·
In other words tne· ( oL,f3) .label on a .line of .fore~ is
,, not·• ehang_ed+.l?Y the motion of the line and ,all .li'nes are fixed
in space coi).seq'l,tently •. ·The rate of change 9f g.. under ·'
·-·· . i
.,
(
; "
the time. dependence and the velocity of the line of :force is ,!,

f.'·.·
.•.. ·.
-':
,. , · .

I '
~ ..
.•. ~ ·,-
I'> '

: .. ··
"' (

. ,::·;··
.. '.·
-·....

··:'

., ;: .
'., ;_.,
·.,:·
:~
: ~ .· _,., ..
't:'<. :.·
' ·~"
~ : . .
. ·.. , .·:.. ·, :.. ··.· ,' ,''
. ·.·:_.·
:··.'
. ···.i.. •''
-.·: .. ==o .·,·-· ....
.· ·.

. . .:.
;.
·.·.' ') : •. ·.·
. . .. ..
.'
'
; ~ ~ '. ~
'.
·. ·:.
,·
.;,.. :. :- .•.
•,!.
''· '\• '!"_,'
'·
. ,.
...
·-.:· .. •' '" '~
','
' .... ...
_ •••••
• •
1
j ~
~

::· ' • .' · ' ' .... ·:

!r ... · .. ', •, . ....
·. '•' _:·. ·,. ··'<i-.
..··, ...
·:-· > ,.
. \'•, ,' •
.. <',. '• I • .

'.·.,
.,: . ' ... . .· ,
··.· ,·_
.···::_>!
··._. ',1 . ·. '•

. .' '~. ' ~ ....

•·.· . '.·. ::. .. -.
: . ~
·~ ..· ·.··.·,
I.:.".
. i
. ' J, •
~ . .
' . ·.,'··
. ·'.; .·.; .:·.· .. ·::
. •'. '· " .. ~ ·• ~ . '.··'
.•'·· ...... ,:·· .
·, ;. •.
. . . ~: .' .. . . ..~ .~ .. ~ ' . '
.. '. j·,
·,'"
. .. ~ •"
. ~ . ': ·.. ~-. :
·' .. .._.: ..... '
·:. 1' ':.-: •. '· .; ; · r,' ·,· \ ..· _- ·.
•.,···
. r' .
. .
·.. .. : '.
~- ~
.r. . ·, •• ..:. ~-. : : • \ :"

.F ' ••
-. .i • ~. ; " :
. \":
''
.... \ •. .
; .6l
! ·..• !! ~
L' ." • .... 0 << 0 I

•'· .... ;
J • •. . . : . : .',·,_,
, .. --·r.· _·, ., ., ... ·. .. ; ... _,. .-... - ··. • ~ • • .• ,'I • •
'1-•·"! . .
II ' ' ...
I
I\
. ~

.. ;
'
-
' ,'

. ,, .,~. .: ' f -~-

68 . ' ..: .
\. ·.: ~ ~
: ·:_·.___ .~.~-- .. , .··--...:~;~'1.-
.' '. ·, . . ... .
.. ·..·-· \

-
; ~

...
.. '
' . ~- ~
.,
.·, :.-.:· '·. '•
·....
·.. ·.. _;

···:.
~

because B ---a
- 1
• i -::. "a1 •. ( (Jot.: x Vf3 ) • It is therefore
..
. ' ..
• ·: •
~. ' ~.

,.: ,>• •• ·.~\.:, • ' :. '· ' •

·.:I ..·· ~~~
I •' ~
: ·.: 'j.":

- ., ':
clear. that altnough lines of force do not move in <.~·, 63, s) .. · .: · · \ ,·.
•" .,
.; > ;_ .... -~- ·.:
. ' .. ·.· . space, the locus ' or :lines . which form ~n invariant surface with .' . . ." -. '
'4. - · · .. ', • •• i• ~- ':- . . ~
•' ....... -
' ~ ':.... •• • ~1
•, -:•.'• I

. ' ' ..
< ·.,·. const'ant J, M, K does. In Fig.l~: · curves I and II are
'.-'.1 '·(··
·' :-. ... · . : the proj~ctions on the r1.,p plane of two surfaces having the
. . ·' .:, ''. . ~-
: . :. :.l .. \.;. ·. : ~
... ;~ r :•... · :-.· ·. ·s ~ ..... e
Q...1.U J, M, and ·K at different times t and
0 ';. ·-. ·'.'

A third point ;ts: .. the actual surface ~n ol,, (3 , s .. -·

. ~ ':'
.·: ;. :-~ ·. •.. ' ; ~ 't • : 'o I • ' -: ~'
' ··. . . . . . . ·• '. !'i-
~ . ;- .' ' -: .•'"' . '
' .~ -~. on which·the guiding center.precesses at time is neither ···. •·. ·: 1'.·.;
,: ····· . ·. :· .: :·: ~
•:: ·~_: .·.. '·.- ' ' '

,.
. .
····, curve ·.·I ' ~

. ",' ··.·.
.• .. ' '...
This is necessari~y so because the partiqle~
~

K has '':•
..· :";
;. . .. ~'.l
'.
.. •. .· ·.·.:
·' ~ '
·.·.·~·. :t changed and is.no longer Curve III has the same enclosed· .·,·
.... ·
t '; ·. ·: l
·l·
·.,
>_:-..>·...>: :~ .:_·:·:.area as curve .· I ·.because the flux invariant. i.f is this area •. . .....,
·.)
. ...
• ''•i ...
It.~~· true.that if the.guiding center were on curve,r at time·
.. . . ; ~

' •' .. · .. , .. ;. , .. J
~ •• 'i
:-!
t 1, ~-t wou,l~ have the ·same f. as at t ·- but
0 J '.
its J. would
. <:· J:j·
·.,.
have to be different. · •
·;,
'i
'· .• ; J . ' : ··~
. r· ,,
,; ,• • ' , I
.. v •.•
·i .. ~~ ... ,: .•~ .': '. . .
.: ·.. :··~-.. .\ .' .~ ·.
··.;· .. ,
. . :.. ".: .;' . :. ... .. ,:,.f
. ;.,_.·,·.· t'
. ·~ :~

. ~· . .·.. ·. ....·..:(•;
.... ,
..
w.~ Vo J 14o~ ffc at-·~· ~;.:: ., . :.·: :;·:;
~ '
.: .: ',. ~· .
"•
···.--··. 1·.
.... '1:1
:· .... ·.
. ~··

:.: .. :··.· .·.

:,•',.,
·.~.,·· . ·, ~··

;
..... ' '
.... ' ~ .':
I '.:-•
•.
. ~- ;·
~· 4 ·:
,··' t ......
. '.
•,
····.··. ·· .. : . . '·. ·.·.
. • ·'·
.......
j /

• . .\ . 'A'!~-··. ; ': . .· .. ';,

...
;
. .... ·..~ ..
,~

'>/. •
;,1
· ..t:
·. . ..:.,
.. ..
..•
_:··~~·. '
... ... •.:t.
·

, .' ' ' . I ~ ' •'

:. ..-'."';.. ·:.. · LOngitudinal invariant ·surfaces at different :tiln.es •.. ·. . . .- ·. · .

•, j, ' •• > ,!, '{ ' .: ' . • ., . • :• ' I ,¥' ' > ' • ~ '• ',' < ~ '·. \I ' ,: : ·• ··~;
'·.'.
...... ·..': ., •.·.· ·-~--)·.;Fig.Ilf· .. ·:··:,· ·;·. ·'' . .
.. "'.~
~ '

·'·
·..... ·· •• \t

... . ' ~

..·
\ .: . ..
.·':·

.·.:,, ·'': .,
.•
...
.-~ :

~·.-.: '}
~

.
'••

·::-.
. ''~'..- .: ; :- .....
•'

.":'
.....
.
' -.>::
..
;
.
:, '' ·: ·: .. ~~ .·. .; .
'. ' ' '•
.. ' . ·. -~~

.. .
\
~

,. ' :_.. :: ~ . ~' • \.~ :, .:: l . . . .. · .

. ''
• ,L' ,
:
.
.
·.,
.~ ~ . .. .. ......
: :.ij
.• :·· •'
..
··~

.:.i ...• .. ·,
..·
··.,', . ...'·:1 •... '.:. ~
'
~1'll'"!"'~r.-:l",.-

f~~\*~t;-~c-:-~-~------ ~
, ... ..., .,.,......,.,.•• ., ...-.,~~........,.,.-~c""'~""',..,..,~,_....,., .,.........,,.1_.,,,...,.,.,,7,,.,,.,,....,,,.,,..,.."··-· ..,...,..•.,_ .. - .... ~1""·

--~--·
•-.·• =-_,,.. _ .• ,....,_,r .. ~··•~ '""'"-r-•-

-~~~
,,.,~-...,-v

-• -•
....... -..-~.,... ~.-.. ··"""":'"'..," .,.........., •.-p .. ,,.,..,,,...,,,..~..,-.- •

..
I ..
:~-~' ' .
'· ·.
r :-·
...
D• .Adiabatic Invariants to Higher Order .
.· !)
·' . . . :: . , .. ·)· . " !
It is ·apparent from the foregoing extensive analysis that ~o proceed. to :·y·<:_··:;. ,, ··_·:
.-:: ~)·
: . \ '~ ... ' .. . ',
higher orders iri the three adiabatic invariant series by use of these .. •' ·_ ~ ·: . ·,.
:~f­

.. · ~
· :>.·· : '_:: .: : · · '.: '

direct·methods ~6uld be ~aborious. Indeed, even to gtiess what . €M',

. .. _ ,.:·-· . :·~ . ': :· > ....>.· ··,_:'}.
.·.•-. .:_. €J', and €i' are woula tax one~s imagination. Proo:fu of the' type presented: ··_ '·. , _r-••
- . . , )' ~---· .. -~----~ -· .. -~_ . .
• .· · .':
'-~i:
... ·'···
:C:··: /- for M, J, and
'
f are valuable in producing a physical picture of what is
. . ·-.. '
::

'''!. .. : ..
:-_i-
· · :. , . · . happening, but must be ·abandoned in favor of a systematic, canonical method·. . ' .•. ,i -~ ·,. . . •
•• $~ •
-.-~ ..
.:,•
;.... :·.).~ike that of G~dner(/g) ~skal (If!) ·-, :·· ·.·.
. or to go to the next higher order.
. .. . .'
':'•. -·' .
. \ ..
·;·_
~ .
.-: -·.:-··._ .
•.. · ..-·
.. ' .
·.,
J.
~-

. :· Probab~y nothing is ~est in the way of physical understanding either, since: _. ·.· ... · :"::
-1,.

~~ ~---~ :-- . .:'· • . 'i! .~

·_. effects which are second order in the gyration radius are difficult to
,. '
.•:. -:_ .J
>
-·;.·.
'·:.-: -:J .
. _: visualize anyway. .
: ·j
'. ·. ·:·' ,.
j.
I,. 1 , • ; ~- • •" ~ ;' .~-},,
,, :·. -:-- In Gardner's procedure, to obtain each new term in one of the
. .. . ... ' ~ : -..,, . .·.
·, . .
. .:~; _-.
. ~ j

;· ·' -~-~---~-~~-~-~···.-~
:•,- ,. '
adiabatic inV13.+iant series a canonical transformation must be :made from the _.,..
~.-·:·...
••
.! f

.. . .~
· variables used· in the prev:i,ous order. A prescription is available f'or obtaining
,- ..
.. -~ ..' ·. . ·: : the·generating function of each successive transformation. At any order, al~ ; (_·:
1.", -_.:-
.:.,;:_. ·.·• ' the preceding canonical transfbr,mations must be inverted to express the · .:~ .;
·-.· ·!'
, ·-.'·t
- ~ ~ . , . I
' .\. ·.- - .• adiabatic invariant. series thus far in terms of the original variables . - __ :f
., .,; .;',

.· ·: .·.-: -:,,'. " . (velocity and position of the ~ticle) •. , I.n practi-ae this may be very labor:Lous:, _..:-: : · ·[
.-.·. . . . ,/ ·'; ·i
1'.
·· · ·: · ·. •· :. · ··but at least a· ded~cti ve methOd is availabl~ and no guessing of the gigher . ·.-.' '.!.
. : ·r:,: ... \
.
·-'
~ :;.
~.
;'· ·I
:···... ·. ':·
.
~ ·,. ,·. : ...
.,_ ,'",
·, _
order terms is re~uired. To .the author's\ knowledge, . . EM'
.,
is the only higher . .. .:•-
,--

:-~ j·:
I

-:~ .. - .' :· ~. term in any of the three series that has 'been workeq. out. . It. :i:s ::·il:.n-~efei-e'noe ··. ..-:. .. ' .. !· ,
!:· - 'i

. -~-- ._ .: 1
.·.-.: and:ts in a slightly different (and mo:re usef'uJ., it will turn out) ·f'orm thB.n'
. ··,.

.The series . correct. t~ough · ~ €) ·. ·is given -.as (fOr the

: '
'. ...... one might· expect. .
. . . . ~ . ~ .· .·· . . . ;:. . ·;:. . ·. -.·
,.,;·
.· .· / . . ~ . '. . .. ·• . . '
r·
.
\

',-_ case_:whe;_e-· 1 .=· 0 ) >\ ·' . ~·........ ' . .·.: .::·:,: .. . .-.··.
'. r''
'
,. ·. .
i
~
;
1 •.•
' ,. . ' · : · ..... ,; l.

.
• •• .•

. :-':~ ·:.'.
·., • • . . ·:-.·· ·-- . . . . . . • ;~ • •• ' • 1. • •
.......
·· , 1
., ~ '., " -, .· ·'',-· ·.· ',•, ;.

·....
-· .· ... ' . . ~ ·: '. .. ~ ·:. '·~- ·..
' ' ... ~ . •... ·- t . .:; .
r ,,._.·
,' .,,.., •·:· ·,. \·, .·. · ..., . ·.-. ·. • . ·.. ·i
. _. ..._
•'-
I,.
.. ·. ~- -~.:
. ___ .-.· '.
'f 'I ' -~. . ~
. '
.. ''
..
....
' . .. . . :-. ··•
',;.,
~ ', ~- :- . .. : -. :

,· .. · ..- . -. .. ·.
.
-~; .-
-·.,·_ ·. .· . . ··.
.. . '··-·'.
-~
...
. . ..
'
' ..
"-:' ~ ~·
-·. ........... .
_
. ~.·:·>·.. _,. ·~--···~·;:·· .·. ·-,·:·"~--· ~

' \ -~. ': . : .;

· '. : ;•

,.-.,. : .··-·- .: :'·._:. " . .. ·... ~ '

•• \'•' r
:
.... '.
.: :··.
·. •'
'·
·....
·-· . .. ~
'!
i
;_.
....... ·•·.
,_..· :- ...
.-._·_.· ·.:· -~ . ....
. •''. ; 1·
i.: -... : .· . , '' . .-· . ........ . ·,
.·· .. ·. ·... ·. -·· .. :,- .,
. -·. ·"·
. I•. '•-.
. ,
,,···
.~ '
··· ,.....
 ~.
'·. ~
~. ' '
'.~

: ., . '~ '
f'.
f. . \.'.
'• ··'·
·':· ·.'":···
.
.:\ ..
~

l : . '. : ~ :· • • ••
. i}

.·
.. . I,·','
',':· '.. :· . ~. ;_ .' .. l
..
iti>>~ ~ [ (h
;· : •'

',: I;, < ; ';, : •: I~ • •

.
Oo;'B/a;' 1 + V ;1
~
·1
) ..
' \
r- ~ .

~.. f ;
.. · '. :-·-· · .. · . .. ·.·, .. vss> ·'.:. \.:.' :;· ..:·.; ·
,, .....
••• 1

.· .. ,
,;
. \' :::'i ;:··:·. .·; ,·:.:J
2 . "> .. ', ~; .

... _,
':> .· ;_. .~· / ' .
';. . •'. ~-·
l '

.
'

-. ~... ..
. . • ~ •.
~)('V X B) (v; ~l + 2~.._ ;;-J, • ~)J + ct<2) •

.t .:<
;.-·:

••• . • • . • • • •

.·.-~> :·.-·~---~. ~.
.
. •

.
i
·:
. ~
J . ;"· '.•
·-
• • l;

.·-i_:
,.,
.... . .! ···:·
.... -~:"" .,:
~- ....-,_ ...' \ :· ~
.. ...
~-~·:. •. .,;,· ... / '•, .. .';,.
.~ ;,

,' ;~ , • <- • • ,'! ' • •;', I • ' ' '

-:~:- :>.'··( :-: . . -_. ~~ •, .· ~ ·. ,,,... ,., ...~·.:

' . ~ :. ' ·. ' .·;

... ·····.~:.~;where V: 1,:s the ins~antaneous. particle velocity perpendicular. to ~1 (~) .. ·· .: ·.·;·::'·.·_·:··~ · _·.:·.. ,:.:
,:~ ~./·::·<:-.not ~1 (it0 ) ..: Also note that in the first ter:t11 B is evaluated at "P 8.nd · .· ..... ·· · .:;.·;:;·
'' . '.. ;'·..', .·: : not il'0 • In the. 0"( •) ter,n, it does not matter. whether . the field and ~1 · · ' ,_' ·. ·.: :'> :. :' . T. ,
.'~ ·. ·;~::'.·;·.<.:. are at 1 pr R0 ;,. nbr does_ it matter whether .V'.t_ is perpendicular to , ::. ·.·;::: :~·~. <-.~·~\~:
·;· :::,'i/'~ :: ·. ~1 (~) .. or. el (Ro) I •. since 'the .differenc~ ~s t7( l). The form (/$3) of the . ' . ' .'' :--: . ·::·;··::
·. ;·. '' · ·· ,,. . series is usefu1· for comparison with a numerical integration of the equations c. · ··. ::·;: :;\'.'
· . ~~< ~·.~ .· ':·!,.'.'
.. . .. ·. '
of ·motion because the re.sul t of such a computation wouid most likely be · .the ·
' . ·.•
... ::.·. · ·· : ·. :.:·:r:-
: .......·:; ~ ~~
. ; (:·;··:~:: :. ·::.::·.
:J,.: ~.;.. particle velocity and position as functions of time. The fields at the · ,.
)' . ~::_ ... : ... ·. , . ·:.r;·
::_ .: _. ,';: · · · particle position would therefore already be present ·in the code, whereas ·._:_-,;,:_:_' ·:·. \.f:
:··.'. :; ··:·.:.:.·.·:··.' . the fields at It0 .would require an auxiliary computation.
~ ,:_. '• • ·, I ' ' ' '' ' •

'· .. '·
The first term of Eq~ V&-3) can be con\rerted to velocities
\' ~ •;:-,v~· '/ ·~· .' •' ·, ;

.~ ·.·..: .··. .:.,-..::·to ~l <~o) and to fields at .the guiding ce.nter 10 as.follows:

;,: ' ~.L (i") = i7 - ~1 (1) ~l (~) • 'V = ~ - ~~\:)vii (r)

~
'·. ...
; .~:

t
'

;·, ·.'
.

·.•. t
\
·~:- ·.' . .
= ~ -L~i<Fo):·+.1I. · v ~<Ri)] fel(r\) +~ ~-~l "R;U· ~ ·.·_. } .· ...
= ~1. (~): v.l/ (~o.) tJ •. Wf. -. -e-1. ;7, • ..<P' •' V)ej_.+ Cf'( ~ ).
: '· ij
.• '.• .. ··' .· :!
. ,. ~t~
• :t

:.· ..:: 2
v .1..<1) = v .l. 2 (Ifo) - .:., (""" . -i-7)"' ""' 2)
2v J((Ro) v .•. p :,, ": e1 +u" € (; 5'S) ~x
>· ·' ....
..
:.;•
~: .: '
·'I·
·.··.·
. :.~·

where the ? or R0 following the·J. or II . subscript signifi~s the direction

. -"P
to which the velocity component· is perpendicular. or parallel. The vector.. p
. . . .. ;·. . . . ~- x· ¥.. .. . .·
~· ~' . . . . : -9 i9 . l J. . 2 . . .l
·.: ~-:····:··.;.is e + c • c,) m . :.·:.. ~ (7(€ ) •.. addition B(r) must be transformed: :. ..j,
... . ,FE11.o =·. ::rn ~ \

.. ~-. .. . . I

'.
;' . ·: . ... : . ' • I ~· it
.. ·-(·';'' ';·;
. ;.·,··;:. ·!,. ..
j ••••
~ ·.... : ~·· .
·. •·. . . ll·.
·~ .. . ' •• ·~ / .• : ·.: .:.. .• ~·, '' . : ' I
-.. .·.!·;'' . ::·
! ,•' ... ·,,
·~·
I •
'
,'
!
': ' ; ',• .' ., ·• • . '-!' . . . ':
.: . :. . . ...
....· .. ·· ~1.,.·:·~ ~ ~·.." . ~. .1 ..... 1 .... <._.','.
~- •'
•; ·.I .:. ·. ,;.. r .~-~- · · :, •.. \.·· -·: '. :; • ~.:; · ,l ~< . • l , ,' 'I ....
 ' .i
- .. ·7l ;.; ·:. .. -~ .
·•' 1' .-. ..)

' .. -~ •·. ·, ' ·' . .

·.·.) .·, ···,. ,;·--.~_-\'
;'

' . ·.: .
• • < ~ •• ' : • ' • • ••

. l'. ' ; ~- ·, . ' •.. ' •

1
'·:,.· :·.: B(¥)
. ~)

.
-...
).-.
~·:. . - . ..
';. -'~ .

. ·~ .
.·'> :::;,: _:· .
·, . ~: ··
<~ ~ ::~I
··· · · With the subst:L tutions ( /G'S)
.: ;~ .' • • ' '
and (/S't,) and some vector algebra,. Eq. ·.(/$J')
' '•
·', ...
. ', ' fI •
.
--~· . >.'
.. :.: becomes:
t,,·
--~-
: . . :· ~ _c . , ~~ ~~- . _:·. /
.·.' -·"' -~-- •·, ··.· -; . 2'.
_::;·.... t·:· ., ;; '·- ·y -(~)
~· :: ' .. ·.. : .. ··.L. no 2v;, 2.
.... -:
.· '.·· ~~nstant = ., B(~)
y
. J.. ·. u@l . ·''.
X ¥J.) • \7.~1]+
.
€.
3'
B
·.·_· .. 1:; J.' •
....
-
.'
·
'/ ~
..
'.,. ·..
'.
2E!l
--
~.:.

v
.~ · v_ . (v x !J). ~ ~ 2)
3
. ( . .L )
. + 2v1.. v + ~\..€ •
:B . 1I_ . . 2 11
'~ ... ' .. -. -~ :
•.'
-.•. ,··
:· '·. It may be observed that. if ... vII

· e·";;;;)re.. is constant. to .one higher order in € ••

.
There i~ a. third form. in which the

.of the a$ymptotic peries

-')
;, .:; ., . . .· Eq. (7~). v was obtained correc~ through zero .order in Now, 2 h~::reyer1 . €•
.. ··.:. .
v.J.. (R.t~\:)" .
:. ; .·· :- ..:·.: -::; J. is needed cor:rect .through . 0'( €) 1 because the _first term.
B(~) ..
... ;\

>·:'. will then yield more .- ~ €) terms •. It sho~d :not be overlooke~ in

differentiating the. series. for ~ ·that· : E2~ · 2iS · . ~
.,,'... ·· ..: · :.·...
'.
~· '
,.,
R2 e . .
0
terms :llti
. '1 ""';;;

·:contribute a{€). 'terms. to ~.as a result .of differentiating. the .exponential.

·· ~--·u,·:_.- '.: When Y · is sub!=ltit~ted into VS'/), the.;esult ;ts.'_ (~s.pee-oi1:1~~ :.· i:
.•.·\'

··'
..

constant -
t.' .··
. I :''
(/Sf) ·:.:.
.:i
.. -. '.. -:;; --~ ~ E v 11 Q1·v:J..i') + c:rce) ·' ··.·., ·-·:·;
·.•. :- l

'~ .- ~- ·)
. B~
. :~ ':
·;-·; : ''l'
. .,,
coeffi~ient ,in ·-~ = Rfo +· Efflt + ~ • -~
···_ .. -~·-: i~
·'
.··.where ~ iS the . second. •· This. ifl certainly ; ·
.',' . ' ·.·' •·· :_, ,:.. !
d
'\;
~-,t· ' '
•· -.-~
:~.- .
. :
the least use:f'ul form for com:pa.risol! with n'UII,leric~ ..computation., Because ' ' .•.
._.·.:· ·'
. d
. ·.

th~re. are no
.-;,

:·. ·.· .·" ;. y contains. ein 6 : it me:y· seem surprising: that

''-:' : . . '
exponentials of.
.
, ,. -··.·:.- . ·; thla .type.left~in .:> (/S'j).·:. The·, reason is that the adiabatic invariant series .' ' ~ • ' ~
~ ·.. ~ .
.
-.- ~ i
. ·. ;~
. <...;
: <

.. _,._,_ :_,__ ·' -~·-.· .,.::....,~-~--·:-..>::.·· ~->: '·<---·-· :.. ··.·-~ .. ..,•. .. . .. . . ·(i
·. i~ ::.,·,J~-~iS.:.~5.;::~ r~~>~Z.§fF1±_.,,.;,:~;;t.-.g:/}1i;:::ti:{~:L~x>·i: ,·-·. · ·"'
 .....
,I,

·72 - ~.
'•,•

.: ·: '.
,. ;
; •'

'·· . ,.•'' .'·

. ·'' ' ;. . '·

...... . '·

· ·. · · ,. · is. an integral over e and therefore e cannot ~:P:pear~ This will became
,l, r

•;
.... . , ....·
,. 'l·',
. apparent .in the 'dt."'<l:rs~on 1n .. S¢ion, :E! of adiabatic invariants .of syst~ of
.: ' ..

· , coupled differential. equations •. '2 ~- ::·1... :. ~-,~ ',. · ...' :-~:,

. ''····· .mvJ.
~
The statement is often made that the masnetic moment is
2B :
;(
t ·:.
~ .' .
,< ·•
..:• !' .
: ·.·.···.
· . , •• ' • • • _:

_.-
• •.... ~· 1 •
.. . •' :J: . .·: . .. .::. ~\.
:~;,
),

"constant to- all orders". The meaning ·or this statement is that if • a :particle
. . .. . . '··~<',·.:...
·. ·:.:'·'. . ' ·. ·. \ ;, J'
-.,. ,·
\ . ..· ,.
-+ -+
~ • ··' • •. l
. ,.·,'·f.
f • • .... • • ~
goes from. one regio:c in space and time where E and B are constant to·
.. /.
:;, "
,\•,.

i::. ··;
·::.. another such region via time and space dependent fields, t~t the change
'. ,:_; 2 ., ,., ..... .
·in v J. /B betveen the ·initial and final states vanishes faster than any ..... ;
1'
>·;
::power of €; even though the change at intermediate times may not. · Th:is 'I,;

., . ····.'·
• J

. - .· ·_;
.
. conclusion follows· from the fact that ~(€) and ·higher terms in the ·magnetic' · · . .
:. l
I

. .~ ;
. . . . ,· · .. · }.
·..• . . . .. '.
moment ~eries vanish in Uniform fields. That the tr'( €) . term vanishes when ·,. .. ':I

A
',. _!, :·.

~~:~ . ·. '.. ·.. ·'

B is constant• can b~ ~een in (/S3) or· (JS?).
'

Righer terms always contain

\'
• •
·,
"
·...
I .. ~
'•'
...·.•· t' ;i
. · :, , · ._. , field grad:ients ·an:a vanish in uniform fields. If the magnetic moment series · . I
··,·.. ;• ·!·
. ' ., ·. ~. '. 'were convergent instead of asymptotic, the change in
,. I

v .f./B would be rigaroos3:-y '. :· l

.7. I
.' . . ···i
,,,
'•.·

.:.::.
..
,'1

:.
::~ , ... zero, but because of the asymptotic :property, . all we know is that the change :: ·-:.. .: . . > ., l.
·. {
1

· ....
,; ·I

.. !
........ :: ,, , goes to zero faster than any :power of €. It is frequently suggested that
.... ....·.: .·}. ·. ' ' i
;._·
2 .::. :i
,· the change in v.;_ /B. is :proportional to ··. exp( -constant/ €) 1 ·which does . ', ·.·
... i
.. I
.. ,
.'
.,. indeed vanish faster t~ any :power.of €1 but the change certainly could be ·' ' . :.;J
;. . . .•.' ........ . ' . '~: .
...
. . . ''·: ,,
,·,
. .' ·:·.··.:,: .

. .
same other tuncti9n of € that has no :power series expansion in €•
• ... ·.-
·· .:::'. E •. The Adiabatic Invariants of Singly .P~riodic Systems ·
r,.. , , : ~· ..
A more general· theory of asymptotic solutions .and adiabatic invariai:l.ts 0" .· .. · ·

,. '
:;·1
.
'·1 ··,_:
has b~en .given by ~skal ~'I). for ~OU::pl~· differential equations. of' a certain
. .. ·,
../.: .
'
'.
. ... . ,•.
·'.. ~

· .. .:·.: ........
·.. •·. .. .... .. ' .. .
~

type.· .Let· .. . ·: .. ~~· .. ' .. ". ·.....·:. . ..... ',

~ .· •. i. _)',. ' ., ~:;
~ .' ·,

. . .. . .
. ···'
·: . ... ·.·.
...
•;

·.'
. •.:·
. •·'
.. .-\ ·:. . ' .... ' ., .
:~ :· ::
• ·. •. ·.·'I~- . . . ' . ' .•
,·
,. . . . :.·:r·'~..;:,.-· ;· .,· ' .. ·· ·· ..... ,... .•
.·. ·' . : ·, ~
.: / ... . ··:, .,.·
.·,.
- '· ;~ .~ : 'I.·
. ·-.~-·
,':!:··. .·
.... ·:·· ; ....
'

.. .. ...
::, _j. •, ·,."

-·, ., - ' ·~ ·.' .. . .

~"'·" . . ..... :~ .. ., ' . .. ;
. :"- ...~ ' ....,
·
; : .,

.. ..
,,
•:
. ·:. ·:· .
;.
' .·::.. .
:. .
...•:.... · .. ··.· .. ·.
. '• ~. ~ ·,
. . .··.·... ··' ,•
·. ·:. /, ...
'•'

.. '•
·,». . . •' ·'. .._ '
..
.· ....... . '·' . ·.· . .. :, \. ·
'l'
··.
':"' • '.·
, ~. '
r , .•
... ·. :-'.. . .~ : :_;'

. ~
...
' ~.
.,, ·.· : .. .: ·' ... ,·.
•.
, .•. '~
;:~ • .. : : •• ' ••- • ' •• .:.:· ' ; • ! ' ' .
. .
·' :~) . .. ·, .. .. ·
1:. ...
.· . ', ;· .·' ..; ·.. ·..... •'~ ':.;.·.·'
. ~ I.·. ' .' ,·:~~· -: ,/' ~.,
....
,. ·.:.
. ~-

.. :
; ;
,• . ~
..'.
•'
~· " · t i' , I ':
\ ,r
. I,' ".
•
' ' 1
\ ·... '·'
.. . ~. .

t. .·.
f _ ....

' . .. .'<" ·:.:/; :.> :·... ·: :~;. / .. ... :.··

·' •,;'

·.. ... ·...

•., ... ,· .... ··.:. .·:·- ·,
. ,· .,,
··_' . '.
.. )'.

.-.··
'/

·, ·. . · ·:.!' 1.'\;.··, .
'.
. >. t,
.·._.;.. · .. :.· .. .. •.,. \· ..., ··.· .. :' .
 L 1 .'•
.. ' ' ' '_: .. ~ '~ 73. -
'··.:: :· ..
·,'I . :\.···

: .· -~ ~ .·< ..
;~ .
~
.• 'j

',
'
,.·.
•..
•,;
,.· ds ... ' .
. ,;·... '·' '·
(/. ) .. .. ~
. ;
i ·' .. ·.. '6"91 . · .· . ··' :· ,' . , ._ .;) . .
;

------------------------------
i ' ... .\.:..
;:.;
·.. ·
ds

, .. '•.,;,.
·. , . · '· ;\be a ·~et of coupled first order differential equations in which the independent
. ; ·....
· ..· :· .. Va!riable · s · does not appear e:icplicitJ.Y. on the right harrl.'.Bide; £ . is a I.

L·
;: : . : ·_. .. . .
.· •.
;· ...
) ,:: :, · · . :·parameter. · The set of equations may be1 ..written vectorically. as
.\ •.·
.....
. t '·.· .lr
i .•
~ . :. ·I
, . .·., :.' •.' >·. ! • ·.. \

~ -+
X 1 € ),
ds
... ,. ....
..
,.(· '•
,
: ·. .. :
', . _.,
''r.-. · ·.· .· ·where ·x is the vector Distinguish between the independem:t .
·:; :,: ., . ," .';·i:" .... '
I··
-variable s ....
li
f•'.
used here and <tstance along the line of force; we will use the t;
.·.:

··notation of .reference (/9) . insofar as possible. Let the system (/to) have ·.. ·. :·.:
-+ .·..·.
't· ·· ·' ,·':~::.~::: i. :.;· :'.the property that solutions of
·'. ' . . .
~ = f(i, 0) are simple closed curves in.

.0:~ space, as illustrated·in Fi~.).'? for two dimensions~ .Ali the components
.
•• ,· ••• • . ••••1

. •, .
; . .......
:··.
I . of. x are periodic with, .the same :f\mdamenta.l frequency. It is ·_also assumed· ·
!.
... ·. . . !,

.... ····.. ·~·.

that ·.'1 possesses • · '• ; ' .~

. ..
' • ,.· ... ~ l •
;'
;)' .. •
."·l·. •··••·. •. •· .•.. .. .. \ ; · •:,. . ·• . .
,, '
,_:·:,;; .
. :. :.· ; 'J . ~: •• ..... .. ....
~ ~
,I•.
· ...
....· ·
._ ... ·.:·..
~ :' ' ... ,: ; '•

... , ' ··.\

.1..
' \ . :.·· .. . ... ::,'
• .. •

I,' •.I
• " •. · .:~ •.. .!" .

.. :... • ·::..
•,.
.. .. ~ .·

'.• .. .f.
,1.
··· ...·:.''
''":..:' '. : ....· ·... ~
~: . : p... ·... :~ .... "\ ,'
\.
· ~:•· ·.·The unperturbed .filOlilt:$-:on is . a closed· curve in . ~
,.,.
';·. ··' .. ·'
!.:

.•.. . : . .,-:,
·,
(
•• <• • ...,
. Fig. l5 .. .· ''",'''
. ·''
5:· ...... '

.. ~

'.·_.·,:.~:,::·-::;a
.. . . . .
·power series ·expansion in ·-e. ·Under thea~ conditions there exists a.
•. •.
; ... ·) ·• • • • 'i·· _,. ~ _.. _-..: :. ·~.. • • • •.
-+· ..
~,~ '., ::·::.:::_:._:_,_::::.:·: t~anaf'ormat·1on · x· ·= x. (z, e, . £), where X . i~ periodic in G,' such tha.t. t.he . ~;

: !:.~:Di:~··;;_;_;?·;~~j~,:_ii:. _};;:_:{: ; ,;: : ~::;)/

·, _... <· ···: ..•.. ,.. :· . ' ''
. - . ,..·(.: . . .;.:;·::/(~:<I:.:_~,}·. {\;:~·:.c.·.<::; . ;, :'-.: .:. ,:;:_.
 .:j.
. ,···. ....
•'.

..., .·
• l .·
·... ····
' ·';

..
:<·.· • .. ;· ·. r
' .,
···. ·' . ~:
( ~ -

transf~ed equations · (/6(J ) assume the form.

,.., ,\'
.. ·; '·!t'
..:·
...;!·
,:: ,: ..• ·t'·"7'~'"-
. . ., ·' ' ~·
\• .. j\....-·:.-;..
~ ··.: ~-:· . .. ' .. ·. '\
-+ ·.
- 5 ~ : '-~.
dz
- -+ (...
.= €h z, €)
'•.

... :-, :.:· .:. . ·. ~i .

.. . ds ..
·
~I., :'I~ . .
, . ·~. ~.

·~.: ,•
··.!
d9
: ··.
ds ..; ·.. :r .
. ·. ~·.-.: . '• .1 .' ' •. , .
.. ·. ·' .. ~ .
. ..
...z . /.:.: i;
'•! •.
- ~ . -·· -~. ;
: ...-,:.·:..... .. . :-;'.'' .
-.~

.. · :The vector has only N-1 components .• It is not immediately obvious that •
...
'. ~ ;'.'

~· .. :_· · ...
" 'there even exists' such a transf<;>rrnation., which makes .. h. and . w . in9-ependent_.
.. -.. •... ·. :
·~ . --

I
of e. The first part .of reference (Jq)
. '
is devoted to proving thi~ an~ to
'· .: .
. ·.·
developing the step by steP.· repeysi<;>n metho<I. which gives. ·the .t~ansj;"ormation
.
• : l

;, .... ·.·,.. ~ '

.· ..
., .: ·:~ ·: :· ,~ (i, e, €). and the ·functions. h. and. ID .as series in ·:e.. According_ to. .. . ·....
'·
;.
!·'·

·. /·\: ,:··. ! ·. ::1

...-..
.
... 1: .. . ·'
- ·. ·. Kruskal, it i~ possible .to pro~ that if the .solution_ i'(s,. ~), 9(s~ €) ·Of •'
::~- ·?.: ':,_: . ·;.
. ' . , ... '", (I l6I I) is suostituted into i =i
l'

•..
:·: ..
. ·"·
,1 ·.. ·g, <),_ the result. is an asymptotic (in <) •• · ..
·. ·:.: '
c;;, r-. ;
approxiln.a.tion ~to the exact solution of (J $'~). .,::.·
, ..
•.. ~· : . '
., .,, . ...
. t', •
.- ... ..
~\. ·~;. '::_' ~.... :
'

'· . .. The equation of' motion . of a charged particle may be written . in: ' ~

~ ·:_.~ ~ ·: . .'.·. ·.
-+ -'>' . ·>,1
x
f·, . .
(~ ··': :;<,:·. ;the :form (/Si]) •. . If' were the six-camponen~ vector ( v, r),~ the ~quat.ion ~f . :· . • '..: • -,1

[',. " .',-: ' ' ...... ~.. :..:...

r ..
~;-~·.~·.
motion WOuld not be Of the requireq_ form beC~?-US_e ~he indeiJen9-ent Variab;l.e . t
~-.
·. ~
'.·
.. ·.
f.···;··. ::.· ·would appear. or the right
' · . ·.·'.· • l
hand sides when the fields. are time_ dependent •. But
• • ' • • •
. ',•

f·~. · ·. ~ .. ·:. ·. if
. •· ..'·
time is. treated as a seventh dependent variable via the substitution . ,._ ,

;:.. ~ :~ . . !.:' :; ·.. '! ,. • . . . .. . . • • ~ • . '

_.,
~ = ~e), seven 'equations of. motion. of the rel;l,utred form z:~sul~.
/
·..
;; :. ; ·.: ...... t = €s (where
~. , <·.!. ~-.~.-~. . · _; i . . . ....
~-:- .
f. <'· :.: :,. They
:::- . .• ·.
are ,
. .. '

·-··.
.. ...
···,·- .. ·i'
·'
)" .. . : . ~..,. .;. .'
' .... -+-· . '• .: .. .• . '·'•
·-··
' .
dv
.- E c:r,·. t). ~ ~.
1_.,•.' . f • .
·;
x
ds
. ' ~- :. ' .. . ·.
. . '.
.. .
:. ~ ....
· .....
... .
·
. ·.
." : i
·_.. .
·._.•. · .. ; .
f '•: i .··'
. :;.
.£!_
-.. .... I~ ·_. :

.
. t ' . . ' .• ' . !, I ; ' '
. , .. ~- . ··,' .·
. _,''
~ . ·' ..

ds ..·=
f!_. .\ ;;' €y ~·
' ~ '
... : ..
t ;.· "•·:· ... ·-~ >, I .-•
.
t\l'-"' ~- :' ,;_ ' ' . .-·. ·, '· ~-

. .. ...-. ·. '::,. · .• >.••• • ....:.

dt'
~
;.•:· :· •'
. <--~- - '::!'".€· •.•. \ : : • • •' ,. • <' /'-~ -'.:
· ds.:· .
~
'.·.· ;, ',c,; 0 l_:.• ..... ,.
.

'-! ·'·'·' . ;,
~:",
.": .'·.: I·' '• ,· '
··• .~ ' .. ' . .
, ..
"';.; ·:"' _l• ·:
.··.· .
l····· · . . . .:·~ ,'r.: ~-·
.. . . ' . ..

1-,
:··,. ~. ;
,'A, '. '
'. :': ·.. .
,·;. -· •::
... .. ' .~ ' .'

: ,•
. ·.•
,. ·,·····.-
, .. ~ '.
,.
'
.....
. ' . .'
F' . ·... -
f' .....
.
v
t:·
;·:::: ·•. :
) .: ..
{. . ~-._.
·:
...
. .. .

'• .•.
.
. _.,
. :·
\

-~
~~
~

;_ ::'.:
··~

-~ '
l,:' :. . . ' . ~. : ....
. ·' . ~. .
·:
,: ',
~·

.
:,.
·,·.·
~ r.
...

-?, "-'.:.'

... ,'~ , :, ' ' ' ' ,

..
~-"'
. ~

' ·,
' ·

•I • " '
.. ,.
i'
' .
J
' .

' '
.... ..
, '' • '-. '
·• J·

·'.,

(:'·
I :.
·.'.·.
. ,. • ,
. .... ,.
. '·-~:~~---;~
.
- . ' ..
·,
. . ·. ..
/
.. .· ·.'
.·:.
·: ·. ·.· -~<<:· ··. ., ': ,·
. : -~
..... ·; -~\ :: .. ·7 ,·
.· ; - j : • " '
< . . :: ":·:··
;. •' .
.,. r·_ .,..
.. ·.• ~:
. : . .'...·.' :·
_....'
·.,_,
 ','.:'.. -·.· .
. ·.··
75
'\
,.·, .
.·.. ·. ··~.
•• t

t.

. . .~ .
When , £, ~ .o,·, this' set reduces to ·.: ..

/ ...... •.:.
'~:
·~~.. ' ..
. ;'. -+ .
';'.·.•
dv ~ -v ~·
,.,.
~ .. '
:.·. ',' -<·,· =~ ;~ .1 ...
./"•'
<
·'
- =.t!it
d5 . c: x B
.·/. '' ....· '. ·.:,·' ·:}....
' . ~ ' ..
•.
'{,
-+ . :·; ·• .
.. ' j • .. . . dr. = 0
·. ,.·

. l ~ ·.
ds
.2:1 = 0
~ •·. ~ ...• .... . :
:•
.
ds

.'.·:·:·
dt/d~ /vanish, -; is; con~tant : -;.0' ~
I~ • ~ o >
·:~
'
. :I'>
Because a;/dS.. and and t is
.
•,I .•

t • ! .~ . •·:··· I.:,·
' -+ ~ .. ~ ·. ,· ".·
dv ·v
constant :-,tO' ~o . ds-
·
=.E -+(-+
r 0, t 0 +
)
c x
-+
B. (r , t
-+
0
· )
0 •.. Because .. E and
-+
. ··.. .: .· .....
·.-.· ... ·•.··
~ ·. ......
:~: '
. ·~·: :
,, :.· ..· .::,'
. . ' .
are constantEi when the~r ~~guments are constant, the zero order motion ·. .. :-:. '··:. : .
'
..
)' I'

.......
:;.·:.:<< .·:... : . :.~is that of 9: .E.~ticle. in a USi:form :field~. In -;
,;: .:. .· cE(r0 , t 0 ) x. B_(r0, t 0 ) : .
. . " ..... :..... center a:t. 1 provided is. assumed .of ~€). '

.
,',.
. . ;.
"

..
B2 .
~ ·
·· . ...:.''.·.''·· ·' .• ....
..:..:.; .·.
": ,·. as usual, and therefor~ the motio~,in. the complet~ seven-dimension~ space : ·:' i. ~.
·,,,
..
,fl
.: ' ...
•. r ~ ' .(· ·····' ·.:., ~!
· •.. •'·.'1

is periOdic when € = o. Conseq_uentlY,· the· theory applie:? and :would give. 't:he · .. · 1: · : · !.::
in Eq_. ~I) and its derivative for ;. ·:,~;_·:J:' ,:':.·· /'';:;: I
. \ ' :
•,•,''·,.' I
'I

Thus far the theory has produced: no adiabati~ invariants, only .·. .....
·.:.
··..:· ,·•'

. ~:/ ·, ·:.
·.·.• asymptotic.solutions. It, ·however, the. eq_uations (/$cjJ are of canonical . ·~~>· ·~:

adiabati~
'','.

;: ::.: :form,. 'there exists.' one or more invariants_ in addition to asymptotic ': .::_': ..: -
.\... · ·.· .· . ·< ~: :·,
·. ·,· ~· ·. - ~ .' solutions.~ Let·:; · the vector x. qe • Jp, .·o
q). ·and . s.
'·'/. .t.
eq_uations are '' ,'t. . ·.;:,_).

.:• '.-. . '•. '. :·,~

~. . '! .. '.' :.·.· .fi ' ~~·.

.. ·'. pi:= ;.1

.• ., . ··· ... .;. ~/;:-:
'·
.• ' ···, > :-..

. '•" .·

···;,, . }~,\v·':··,
·....:
......... '·''
.. ·. ;~':
'. i:
i
~
•'l I
•. . . • ' ~ 4 ·t

. . :,:\ !
... ~ . If the Hamiltonian. is time-dependent,, time and energy can be ~: .' ,·.~ I
·::: ...
•'·
~ •. .. '~ I

. .

,; ' ;
use.d as conj~~te variables. and the number. _o:t; degrees of freed~ increased_ .. . ... .:.-.. ~
'.
•.
.. \' . by one. Given the necessary; conditions .on· the_ pe:riodicity of the ~e:ro order ·
'
.. ~:·
.
.'
~

' . . .· . ' ·: . _.:. . ;_ ... : ·. ·.·.. -+ .... ·... . ·.:.: . . ·.... ·- .... .
·.·. ::,.: aplutions1 ·~here exist•·. the_. trans:fr(lma.tio~s; _: ,P .7 p .(~ 1 ~~ • E) . :~d q~~. ~ (z1 9,
I • ',•
·.<,,'

. . .... ,, ,•.,'.'.':' ··.· . . .· .·· I':

::...· ·., ·~.·: •. ·.• •. _:·~-~ ,·::\_.••·.: ·,' . i .. : • .•
....
';' ,..,. ~·:: ·~ ~;:.'',.. ··::·~··.-.~:· ~~,_~:.._:·:1~~!·.··•.~~-:.~.; :.<·:.·}~.~J . ;:< .:_ ..~...~
'>;
•
,:;'·.; :···:,.,>:.: f, .·: •.. •• l··· ':< i ..:,. :.(.
 ....
·.· ... ,-.·.·.
:
'· •.. · :,'·.:t.: ..
·'?. . '· .··, ·.'•
·-· J.
;.•., . : ·'
..·:' :- "• ·:-_.:·
·' ~. ~ ~.. > >· ~ ' .... , ..
l '( • •••• ,r''f· ·- '·
. '," i ' ~:
.... t .• . ·''
I;" :
. I
~ £ .·.··
,·'
I
. . ... •. :· • . :: • · which are periodic in and are ·such that the transformed canonieal : :. . '
1

':..~· ~··· '· ' ·, ,' ·. . . - ... '~ .. '• ' .

.. : .. ·· .. ;·,.. 1'.. ·"

::·: ' · :· :.. ~··equations are:·::.': .... ·· · :·,

'' . ..
.,

·: ~-:; . . ·~
.
' ' • I• .. • • ' ;: . ·,, ~·. •:

.. :.-:··:
·; ~ .. /'','~./-·,;::':; '·· ... , .. · d't
; ~ .. : • ·, • : . ' ' ' • • • ! ••

.'. dt
. ..~ . ' .. , ~ ' . ·: .. .• ~- .. , . ·?: ·.
... ·· . :· ··: ~ . :J
.... · .,_1.
·-:~~- •. ' :..
•• _i ,•
! :• ·,.·: . de
.. dt
'.t

.. ).' ... '' The solutions ; = r (t, €) and e= Q ( t, €) of (/,c-). can be substituted·
. ·.·'
. -~- ··:.J
.:.~;
~: ...,·:·:-·:· (·:·~:to give · p. ;;P·/) (t, €)
'·.;

:' 1 e( t, €) 1 €} as a series in €1 and likew-ise for < .:·:·:· ·.:· ··.:·:.: <(l
r ; ~-·
.. ' "
.'! ~ . ''·. :· ' .•.
,;·/-.. ,: ·.·:· q.... Tb.e.se .are not exac~ solutions of the canonical. equations, only asymptotic ..
. . 1 , · ·•• •
... ·,
~ ·-~·'
'', '.' :. ··.>::. · s·olu't!ipns ... . . .
. :·3 ··-: . •." " ... ,· .··.'·-~\~:' ~ -'·;-:. i
''.; .. ·
~ ;: ,• •
t·-·
'I ' ·...... ' '. <·~: • -. <(

• .··.·.t
• ·_,.,..t

.,,·. •'.:J
:·,I''·,
:~~-. ..: ~; ~ i{
I
t

• ~
I
' •
'
'
~ :''
,.-
' \•,, ,., .;
,... ~ ... . ,·.
, .. ·' .
~
.·::;'!
.., ...
' ...............
~--·. '. . ~-, ...
~
' ;

• . · ./t"".·. . ... \.
(/'') .. .. -.-:·::
i ·:~. ·, ..;.·. :·: <~(:'!'-~:··, ..· .. •,
I
·, ·.. ·..
·,' '
·.!
'I
: '· ... : ~,. . · . : where to. be specific the angle variable.. e is ass'UI!led to have the :period.· . ' ... :..... · .: ··.!
i, >' ·,;' ~ -::.· ·;:;·._ ·.,
.. ': -... ~·
o _'"' ', I' '\.' ' ' ' '.
''\
,

. .;'·'·· ·:, 0 ;.. · "•t,'·'-:· : .t

to 1. ·. · ·The invariant· I.. will in fact be obtained as a series in e:_, · .·!, ~..::. 'r'~_·,· ~ • '_:-.: I '

'·.'· ,'·... ··. ; ' )',·~ •'' :, : ;.· .::. !

.. ....
are themselves series in €. . The :proof of the a.dia~~tic~ .... ·:~. ;:: , <·:./~
. ~

l',_ ...
since. :p. and q :,
' ... :]"')~:,:Y-'':::;.,~<·. inva.l-iance of ·I ·.consists of showing. that ~~ is ·zero, or. that .. ··· ·· . .- ·.;· ·.~.\

·.:<::,.'·~ . %i·<Io + -~.:t ·~.~. €n In)= o + CT(€n+ ·;)

'· • '".:.:
. .<.:. <.<\~'s! ' .. :J ::'l
!, .
.. ' ,._ .....
':
' ~. ,-
... "' ··' •·:
.-. "'
......,, .
dz. ( t, e:)
. .
~: , ... ·· .
. .•.:
' ... : ':;.::.'ii
~ ~
... ., eli( t, e:) =
"' • \'
.
,\'
'
:
.. ·..
·:;.)',} I
dt dt
'·:..
.:., ..
... '. 2
,·
a.z·
.·
·:·>.·:.
'
,, ..•. ~( de ::J. ..' 0
:Pk ~.,;:;o_z_·-- )~
9.k
'· . dt .; . j
,,.

'
...
:-· > • • • '
I,''

•• _ }·
._._;·

•
•

).
:· ··'

::
i- .. ,, ••
, dzj . ···-::
·.where. sums are to be taken over repeated indices.·. The factor · dt is not. : '•, ;·
: /.
' '
• .. '.· . .~ . ,• .·
.
. ··-·.
. \.\ ... :-~- ,'•
a function of. e and therefore .. has been put under· the integral. · .The second
~- ' •. . "~ . '.' r ' • . ' • ·'' •' '',. ,• • . I f '•... ~
.• •• • ' ,, , '. : ' • • T ,, .~ •' \ , ••• "" , , ~.
.. ·i ;_ terin in the integrand may be integrated. by l,>artB to- give ·.. t. ; · ·.··, , .

. ,. ".

·:.~:.:(·':~';:1~.{';>/' ,·::·.,:· :·:·,'':\~~···;~· ?>··· .•. ·:·,···~·::~; ..··. · ~

. ... ··:~:-~-r···_.~;·\···.·. •"·.·· · . ,, ~ ..-.........
: ...... ; .
~.
. •• . •
"t
. ; ~ ' .. ·· • " ·, .... ;.
- ...•,'.·· .., ·7,·. ...t ,_,,·:
·.' ~·" . : : :: .
.·: !·.~··
• . ~-.. ';
 ·:.; ..
. '·..,.. '.

77 ', ,• ., ·~· '

.'· : ' ~ : '•·' ..

....:... ~ ;-
':··.: .'

~-
,,.,, ·di
-, dt,.
'.'. ~

... : ,, ' ~~ .:
•. ' • ' • I r ~
''·· ,•
-+ -+ ··'
vanished because .P and q •'.

integrand in (/(i). is simply '.:· . :

'·" :
..
' ..
' ·:< ..·.
r. .,, .·. .. · . ' ', . ~.; .
~-~·''
!/: ' -+.. . . ' OH SPk
;?·.
r:. · :~) H(p(z; e, · €), q(~, e, €) €)_ = 0 ~ (fe
.:. ,. .. · ·... .. ~. . ..··;:
·· ........ ,·,

·.·.·... ·:
.. ,·'
., . '·. ';,,

r·· . ' ....

I ''··'' , •'. : •

: •
~
~
·,

'
.,

·:·.r,·.
'• ':·

.·.. '·: ... . ' ' ~ ·. . .. '....

dzj
. . ~ ·;··.
' .
..
' .
.· ..
,· ...
•.',
; '
: .,,
. . . ) dt ! ..•. ;
;t· ,·
... ;· ..
·:.'·.
~ ~·. . ··: ,·;

..... •.>:;
'·, ..... Therefore._. .. ~ ' .

·..

r
\•
:. '. '':·'
... .' ' . ~ -~ ':
\ .. · di
'dt = d9
oH
de.= 0; (//o).
. .'~

0
.;

•
-+ -+
the integral· vanish~s·due to the periodicity of p and q in 9. It ·should be

emphasized that the;: adiabatic invariant is the integral of -p -+

• dq around

a closed. curve :on w:hl~:h:.'>t.<is·:·cqnstant··- iri:.:t, -·~-phase space;!!;; and not about
J :~::, .:>· .
•' .. ~ : ....... __ ..........-·~· . -~\ . . . . .··: .: ... ,;·:..
the_ closed curve representing the zero <r.<ler {i.e. .€ = 0) periodic motion~~-
' . . ,·,.: ..\ ~)'. .~ .... :·
The difference .between the two curves is ·shown schematically in Fig. 16' ·for
'

. ,··.'
€ it does turn . ; : • ·_. ;.' .1 ~•

. . curve.
~.·. . ..
' ' . . ; '. . ~

·.: ·.:·
the· difference appears
. • ·· · · · ...
~ . :. . . ·''',
..
,·i ·,' ·'•'/:·: ·.'
'

'.,·
·.. · .. .· .. ·.. (,
'·

t ~·., \. ..... ·;

; .:'·: :·,·:..
''·:··,•,,,
. ..
:. f, ·,; ·,
···.., .
~ ~·

. .i. ~ '. . ':.·: . ~ .. .· ·::' ·.•. ·

 ,·,. '
;·.·.,- ...

; .·

. . <-" ··. :: . ., .. ·
~ . ~~- . "·

~ ~== '-.H-t...r.: & IS p~...·t:. .. ,f~l"' Ji"'';,_; . . ; .

r f>Jl-f,·t?H ,,;., .ct.,"J. c ,.,,c. .;. . - . ·.. ··:·~ ;_:; . .· i' ';;
. : ···<- <:.~
• • t .•' · · · : ,
'.
. •
. ~:
• ..
. . ..;.. . ! :~- j
b-7-- c=e>: $ '" ptA"~" lv~-(} Jostft""' ": ... :·· ·.·.··. .'
Clio -ti..(J tt. r"~ · . ~
..· ' ·.: '·. ~ '
' • .. !-·· .
. ." ,:
·. ,
····, ~

··::, ., ..... ·.·,

~-
'·•' ·:.. .. ·

·.· ....

...
".".:· ',

The unperturbed path in phase. space differs from the constant 1' curve.·
i _: /l, ·, -~-; Fig.'l6,
! /." ·~- ~
f, ··: .. •'' .. ·.·.:.

. .. -( ·.·= ·.. ' ~ . ·::

: .

. The constant . t .curvet are known to be closed due to the periodicity 'of
~ . '.'· .. .
..; ~ . .' . . . ~ = ~- (i, e, €) in e. ..,·: .
.... ~ .. ;

.··,.,
· In the present situation, where the coordinates all have the same .·..·..._
•· .
.·. . ··_;
t~e unperturbed st~te, the adiabatic invariant (Jc,r..) .· .
'' ~
' : . ~
. .fundamental frequency in
.... . . .
', .::· ,. ·.
'· '') ' . has turned out to be the sum. of action variables rather than any one of' them
.... ;."
t• ·' .· :: alone. :. Of .dourse for a system with .one degree of freedom there is. no .
r·
t.
[·
t• .·;.
. ··'• -; . distinction. The: integral for I can be written
'·· ..

~. . ... f
I=
1,

·.·. ·< ·.:

'•.

-+ '·.
where the double integral is over the ~ea of the . z = constant curve, and "'

this is one ~f the· .Poincar{ invariants.. ·It is not surpris~ that I

,.

,.i-' ..· ·; -~. -~>· ·should be a Poincare" inWxiant~ the value. of the adiabat:i,c invariant for. :a
···;.
: .... · .-!. ·. .• ~· ~. ~
.' .~
given_ sys:tem .should :t>e independent of the. canonical variables used in .·...•
.....~
,. tl . ' ·.
. -~ . .~ :~.

. '
.. ·.-''J'

'·.-.1
and indeed. the Poincare integrals .are invariant under .
...
, I',

.,.. ... ·-~

"·.(
'' '
·canonical transformations,
.
-whereas
. . ea~h:.::term of de. P .· 0~ ·is not. '•: ;.'•

:·

'
..........
·, ·,- It is now clear .that .the number ·of d~grees Of freedon1 is not_ n~<?essarily the
,·.'
f·j. .. •,',·,
:· .; .,. ·. ;.. i'·. . . ~ .. . .,
:: · -_._,_-. ::: . ·n;mmer of adiabatic, iri~iants 1• but is .. the numbeT of terms. in tb.e .adiabatic
. l

...: in~j,ant integral.. A system· with_ae.y;

. . two. degrees of freedom might have .
:.·._.-. ·:
. "
·::jus~ 'the ~n~ adiabatic ·invariant·.·: .· ·..:: ·: ·.' '· · ' . .· ~ . : ' • ·"r•' ,:• ...
. :· ··. ·;. ~-:: .: ..•.\ :' :... . .: ; ' . ,:~- . . . ' .. ·· t··· ~ ': '. ·.., .::. ··. ., 4 '.,

. ; :'
.·;· .. :\.;{
: ·• .··:: . ' ; ... ~- ·•.. 7·.'.. . :- ~- . :, ,·J ·, ... •'
::..;··.. ·.::·· ~~-:-~
. ,"·'
~·- :'.
,·.· '•:
. ,.
'.
·--~-:: :·:~ ,•'. ..·
,.. •. .
;; ·.:-_:.:
·. : ' ·; ... · •·.
. '· .
·..•. :. <
:~·-.'
,: .·~...:;.:>-:~·. · ·:
~ ' ,..
.."·--. --
...... .. - -- ---·.

I ,·

- 79 ..
On the other hand, there may be addit;ional adiabatic invariants.. It is

i
;-·. ·. . possible to prove that the Poisson bracket\) G, I] equals
. . . unity. '
The proof is a

little le~thy to be included here, but it is outlined in reference (19) i?

. .

i1· -- sufficient detaU.• .

The fact that (e_..
IJ =1 means that
·,_
e and I can be used as nevT
J .

conjugate variables in a canonical trans·t~rmation of the form (P1 • • ·I\~;q1 • • .qN)~, i

.. ·~- ..l
(P •• iP . I,;Q •.•. .,.Q_• ·. 9).- Let H' (Q,},_I,e) be the transfor-med Hsmiltonian. As
. 1 . . N...l, 1 1\f-1 , . ·
indicated the nevT Hamiltonian is not a function of· Q · because
' f
Thus
. H' - will have one less degree of freedom than · H, ·and I··
and I · is zero~
I
..,... ···will be merely a parameter• !f now the new canonical equations '·'
·'
·' '

(172)

have solutions which are periodic in P, Qsp~ce when e or some other smallness

parameter is zero,· the entire· process from eq_n. (164) on can be repeated and

a second adiabatic invariant ~P . dQ obtained. For the case of-' the clia.rgeCit-
par.ticle .:thisc:·wollld ;~e: the ~longitudinal ~invariant ;·,:'di:P·it us'.· :t:r -it ··s().::rmppenl~ :'::l
that '·the..·ne'\'J; '.carioxr:i::cal;;equatioris-'agai:ri ihave periOdic~: solutions, there, 'tvill be a

third adiabatic invariant, the flux invariant ((> in the case of the charged

particle • Since ·che number of degrees of freedom is reduced by one for each
• •.
adiabatic invariant, it is now clear why for singly pe:dodic systems the

number of adiabatic invariants is at most eq,ual to the number of degrees of ii

il
I I
freedom, and will be less if at any step the nei-1 canonical equations corresponding
T. ·: •

to (172) fail to·have periodic solutions in the unperturbed state. In connection

·..
lvith the charged particle, this woult;i be :found the case if the motion along

the line of force t·rere not periodic •

..

,.
 . ·.•. :..: . ..·
I . .
\·>-
·.,..:·
·:· ,;
: ·.... ; /,' ·. : ~
.·' / .

1.•. ;·.: .i.· I .

I . . . 80 - ;·'·•
.I·,• •. ·• '~
·.'
' '-.~:'~- .·-·:/:
f. ~ .•.' ..
I 'i
I .
. ........ '

·~·:. ·' · By use of th~ 1 fok d~ensional relativistic H.amiltoni~ ~)

r ". •·
•·...·
.•·• •. R( -r) = - ~
2mc
fl,/ 'LK
(p~ ce ~" )(pK
: ec cpK) the method. of 'this section : ... ;· ,' )'-'·
·. ·. .
'·. - ·'··... .. -·-·
:, _,.,
r :-. .,·.·-.
.· ·. · . .' would. ~ish an alternate .(to Ref. II ) way to study the adiabatic motion::-.::t,u:'. ..- ·:··.:.'· . .' ·
· .. ,...

*
.,

·' . _ of telativisti~ p::~.rticles. To the author's knowledge this has not been done.
·. '· ,· .... :,
The .:Preceding proof that is zero would establish I ·as a ~ .' :. ..'., ·:' ~ .
·!·.·.·:.
•.• #.·

-+ ;·

:.: rigorous (not adiabatic) invci.riant of the motion provided the series for P. ·,.·
i ., _.,·:.:.' '· ~
; :-.'-
·_,.-_._. ·:·:
.·. . . , .
.•, .' i ·:>·and ~ were rigorous solutions of the equations of motio~ instead of ,- .. , .. :,
-_.,·.
.
.. . .•:. .' . :.. ~-
' : -~- . . ' •',·. ;i .', I ~; ..• ~ •. . : ',

~ . non-converge~t asymptotic solutions. It is the asymptotic property of the. .. . · . -~·_." .- ..

'·>. ... ...
-+. .... ~ ~·

··series for p and q that makes I ·an adiabatic invariant. · The theorem
.1. ·- .; ..
~ -~
l
t :.·..
. ~
'· '
;'_
:. · establishing a rigorous inVaricmt woU,ld be: · Given (1) two functions · .'·. ·
··. '·
i ,·: ', .. '··

.·. ·. ~-:;: ..... 'p(t, .e)·. and q(t, ·e);· ·which are periodic· ~ (2) that 9 =e (t) ' .' ·~
.··· . ·...· .
...... ·.··. . ··:,.
(t) are such functions of time that p(-; ~ t), 9 ( t)) and '. ..... t.·:
.... . :
''.•

. .',
,· ,.
'
·.
,. ' .. \ .

.. i.hv
~ . ~.
(
( t)) are solutions of eq_uations of motion :, .
i . A ·. 1,:
,.· _ ·' ..
·.. - -+
.·.em(~, q)
._.,
=- . ~
derived from a Hamiltonian. Then

.- .. ~
~·. . .•
'' I : ' ·, ,' ,• '.•· '• '

I
)
•
·.·.•

: .. :: .....
•'
'f
per:1.od.
is inuependent of t. Such
..
"•.'
:·; . ,·."'·-·.
I··.··:;:\·
.,, .... -+
· :,.~ :. a theorem is not of' muchuse unless one knows hovr to find :f'Ul:lctions· · p,' q 1
I ·:·.:,,
..... . . .I ···.. ,·

t. :.~·-
;- ;:,,·,
z, 9 with the.;required
l
properties.
. ~ .
.The•theory discussed in the pres~t :; ., .
1 ,· •I
...• _·_1-'·

-··...
as asymptotic
. .-·,.·
:'.
!-: ·: '·:_,. .,. . ~ .. . . ,· .

v· :··.:.('
·series in .' ._·;,i_
.
. .·.( ',.'.
··.·.. \, ;_ . ·· .... ;_· ... _, .~·.

·:.· . ~ _.,
,• '.
"·'
.,, .
i
I
•,
.<i
-:

'· . ;· ·.:·
L
,· ....
· .. ·, ..
' . ;.·! ·;.
- rI·'· ... .. . '. '
·.. ·.;.I·:.' ...
'
·~:: .... ·-
; ·..:-' ··· .. '' .. ·· :-. \ .
.
.. ··. ~-' :
';·; .· .......
·.\. )

-",·-·
.
. ~' ' ' ·_. - ' .. ~; ;.-: . . -J .' . :. \ .. -.
..
·. ' • ~. ~ ~- 'I :, ' . -',:' .·.
· ... '· ·•.:
..· .. :::
..
' .•, ~ l_: : '
'. I '' • .' ·~ ', ' '• .,_
··.·.
-_..._, : ._·'
. ,•,'.\,
' .. '·.
. ,·. ' .
.. / ·! •.
...,
':. . ·· ..
-·1·· .. · ; .i.··. .. · .....
.·.
~ ...
;. '·. .. :....--~:·.- ·, ' . . : ," ·. . . . ~:.
J _:_l

..
,1 ..
. ~-> :-·: ·'
: l
.··.. ·
,· '· · ...·., _.
. ' ';
·:.
t ., .
/ , ~ . ·.'. ·.. -.. ··... .... · ': ...
·. ·t-. ·:: .·.. _..-. .._.;. .·.-·:: -,·
;: ,:
. ~- ,• ' ._. .
,.'
::'·. ·._ ... ..-.
_
,':
.. : .· : • .''·• r,''
. " ... ':·.
·,,' •'''I ·, ,.,'-
. ·,·::' . ,1·
 .' ~: ... ,
··c-.::,·.
·.: . .') ~ ·.:' ·, \;
. '
..···, •.
'(
·) .
. ·, ..

~-:"·.,
'~
,,
. {·
. .., ..
i: .,: ' .... _':,1
·~4 .•••• , .. ,_
.·,
,·.' .

(:'
. ' Y• APPLICATIONS OF ADIABATIC THEORY

. The Current Density in a Plasma

_·· .. · (2{,)
In ·e}lectricity and magnetism texts it is· usu8.:l.ly proved that the
~,· ' ' ' • I

•·. ~ • ....,, ; •r·_'

equivalent
.
current·
.
density due .to magnetization of' a material is given by· '
..... ·. . . • •" . 'J:. ,·

::-"~> . . ::-'>
...
... ' . ;--

c\1 x ?/i 1 where J?i is the magnetic moment per unit volume. One expects that·

adiabatic conditions the extension to a collisionless plasma would - .·.-···

' .. ~' . : . ·~ .. .
:-·· .
:.· ~to add to . _c\1 x 11l, the current of the guiding center motion. This is

indeed correct, but is not so easy to prove rtgorously for a general -

B
'

.. ;·
.:
.. - :·
~

, field geometry •. The author is aware of no published proof'. and therefore will<:.
,·
·.
outl:ine~Oil~ here, with many of .the details omitted. ·The -complete proof is •>:, :
·'·
1
;.:
...
. ·... · • ..... _)i
··;··

. .:
. ~- ' .. ~- ,;-·. ..··-
. .
.. ~.;

::. _·: · ·... _ We wish._to prove that the current dens;i..ty at . -; ·is
... .' -: -·
',1:: ·..: ·.·: .. ··>~ . . ·:. . ~ '
\.' ·..
·-::
., . : ~' ':' ·'
•'
.·.
'(?3)- ,_.
:·.... """ + c\1 x ~-
-+
J. (r, ''(, I . .,
.. . \ .
.
..... _: ,·

. ~c;..
·, ·.·· .. ·
'·

f.
( ~ ·.
N =
-~
numbe~
·•
of guiding centers per unit volume at - r and t.
-~
-~ . .'
' .. · ; . ·

:; ~ {R ) =.average guidng center' velocitY: of' particles with guiding centers at r.

0
_., . :·'hi· - tot,al magnetic moment per unit JTQlume of ]?articles with guidini!; centers ·,,'
·-;_ .....· ...

-+ . / !'.
at r.

There will be an equation of this type for each charged component, ions and

. electrons for example. ~e~e exist_s a reasonably simple demonstration of

. .. . t . .. . . .

·; the perpendicular,: component, · J.!... ·= N~{: lftcb.. )+ c('V x ·'hi~ .·. The p:r;oof
-· .:·· 's~s with the second moment{.a'l)::· ·of the collisionless Boltz.naim (VJ..a.'sov)
:. (".· .·.,.
. ; ,·;
''·:: .. : · . '
second moment is- · · .. .~ :\ :~ ·. :'•· ..
,.
'
. -·~ -·· .
. -.:: :·
. ,....
. .. : .
·,:: •. i' I :
•'' I'• I
-.- .. l

•..
·.· ,.!
.-,·· ..- :. ·. -
'• ·,I ..
.·~.
'
_; ·....
,·;
...
.. ~. _; ~-~:' <·: •, I, '· '\,
. ;.:_. . .· .>
·-···.:
, .•..
;: .

' ~ I '
. .. . .
.·.·· ·:·. \ ,.
..~ . . .~ ... . . ·:: :~· ·.. ,; .
'-~ ~-
. . ·. . '
. -~. ' . ·-· ~ ... ' '.... ~ ~ .
·- . . -~ . •'-;· ; .•.. ' ·- . . ._--..
-: .... ;.· ,.. ,:·::-, ~ '·
y . ..__ :1.. -:.: '/·
. -·. ~--:
·.· : \~ ~' ·: ·.·. ... "
• f ,:

. ~ ... ----~-·--· :' :·_: : .~·. :.··.:: ~) .·.•,·,~_-: .. :/'·~ .

~
~ .- . ~·-\:::' ,.. · -~ .... ;·. __
 ..•L.; .
••
i.
:•;
;·· <,
:··;··-,;;.:. .,
.,
. ~!; ':.
:·'
- ·~-. .~ . ·. . . . .
,' ..•.• ··~"J:' ::.:

. d(~} (~) ~ .... •·

nm dt = - 'V • 1 ·+ ne
c x B + ne :j!:, ( /';¥) . _: . . :: ;. :'
• >·
...
. \;·.·. •
'·.:...···..'1
.. ·

'• I 1
wher~. { v)is the avera,ae :particle velocity, n is the particle density a.ii.d' .·
: :- ...
. -· :' ~
,
•.
·~·~

p . ,; npl ( (~ - {-; >) (~ - (:;) ) > . The current density perpendicul.ar to B 1 ' ; ,_.:·:-:;_·:'_.,_', •. ,::>
.'. ' :' .-.'· ..·>'. .......:. ;.:'~-~>:..
·. "!"
:.·. · is J.La = ne( v }J. • Conseq_uently can be obtained from (17'1) by ,• '
. ._.(· •
.... ... . ·'

- -+
crossing it with B, ~ut . j LJ cannot. The a.I).alysis has been given in

detail in re~erence '(~).

~· ..

~l .'
In order to derive the parallel component of (173), it is necessay
. ~.- :·.
·. ;·.
·_. to work from the_ asymptotic. series (/,/) ·and the colli'sicinless Bolt.znann ~- :.
.··:·.
...
:: : . ,· eq_uation itself rather than its second I;llGmerit. ·. Eq_uation (J73) complete . ·; .·
:-
. '!
.· ... r ..
· . is obtained· this way and not merely its .parallel component. Let ~ ( r,v,
~~
t
) ··
.: .. i .·..
... ':.
{.
'._:. .

. . ''.-
·.·; ·.:

... be the particle distribution function satisfying the collisionless Boltzmann .. ,.·.. ,
' ~
. . ,_;. ~

: :'_:·.~ :· : ~quati~~ ~ :+ v .v· t· +. : (E + ~ x B) • 'V ._. f

v
= 0. The. current density is ·!< ._;,
'·.···
._·.'.{
'!
.,
_.,
;
·_:·;
.. ., .
•',
,. ~ .

)·._.
;.-.·
-+ ..;:.
j {r,. t) = ne ·{......v) · =. e 1-+ . vf ·(r, v, t) d 3v.
-+ -+

t.' ',·' :..,~- ' .

·.. :·. ·.'·.
;:·
-+
.·. _,,The procedure is first to differentiate the series ('/ ), for r with respect
•.... .. . . -+
r: ·. ::-:,to time and obtain v correct through terms of cr\E)~ Next the "guidi~ ·.
.......

.·:.· center variabl~./J~) R'0 · and V are introduced, where the components of
I, \, ·; .•, ·: /. '.1
-+ . ,.. h ,...,. .
1"-.f • •• ·-

v along the directions: of el., ~2? and e3 are vl = VJI. ' v2 = p cos e,. ·.--:·.:. :.·
i"" .:' . (1)
• ,'I
.... ·.··
'·--~ ~-~ :_:·
:.

and V ~ .- ~ sin e. ;iJ. is the gyration velocity in the ::f'i-am~ of reference

.
•." ,'I

3
·~ .• ., .. '. -+ . -+
~· ' . movin~ at the guidingcente~ _velocity R • The.:;particl.e velocity .v tram the··
0
...
' :-
·,I' asymptotic series can thEm :be :e'x:Pressed in :te~ of 1R and
0
<V: . :.L- ·:· .· .;
'
f.
.. . . _. . :._. ' ,.~ .. ·..:. .' ! ~ ~·: ~ .: .· .. .'.·. .• ' _...... ' ·.' .;'. ·. . ' . '. ' '. . ;_ ·.. -. . ' .. ·, ':.. :. · : .'' . •.
.,.:. ·I!' .'
. . . ···. . . ·. :.. ' .,.•'. ; '
. ' I'•.. ·..
1,·• .:·. :: ..r.:''·' .. ·· - ... ;·:->-.:-- .. :._-.. :- ::;i_.-.!_·'.' ... ,- ·.·:;.' .• ·'' •' ···•· ~·:
.. ;·:::_._.:-... . . ~.-
. . \: ... ·. ·.....
.
'{•, I· _'

: .. ·
0

~.. ~ ~ .·
f" , . , •

. ·. . .' : ·:: : ;.,·,

r I ' • '.
.. ". ~- . .
·'' • · 1
- ;,
'

;
•'.

.
', :-' ' '• ·'
~
) : ;- .. :;
·, ' ·. ;: .. .... : ;. ·. :·' . . . .' ~ ....~: ' . •,' .. .·. ~

. -.; . . :/'.. ·.- ,·· . ·. ~· .·,

. ... :- . •. .; . . ~ ..
' .·:
"/
~
., :·
\ ~ ... . ... -~ '. . .. .. ' ..
:.
-· •.
·:.· .•., ..· ·..... .·. ·..... · •: .. ..
0 Y .... • .. I' .: .. ;, _: ·' '. .·. :· .; ~- ' _; .

:·~ ·~· . . . ; ·:·-· .. ......

,_ ·.·. : ..
.. · ·...... . :,, .
. ~
~.

·. ,·
'
' ,· :-. . '' ~ .··. ·: ...-.·
; ... : ........
..
c .
:. ·''
·. ~ -
• ",J". ·,
.. .. ·. ,.
' ·.: .· ; ~
~ ...
.: '
~ .. . ' ...... ... :,:.-
 .. \.'\·•
•.· '·. ~ . I
I
. 'l, ~ .
•'•
·.;,, :·
;,· ..
. :··-~ .. :.·· ,•
';.·

.' . .i

i'-,'-.·:_::_::.·:·~~\···<:_~}-~-)~o>k·:.+ ~1v1 +.V2~2 + v3~3 + ;;.·~ ~B ~2- ~2).";- : 2 ( ~B ~3·- ~) · . .<..:

. .......·r·..
,,
. ··'; .. ·,· ,. ... ).,
·• '
". ~;
':1 : ' :.' ~. .

·... ·+
.-:_,:.;·>~.- :_ipi .._:(..,.~·-:·. . -~>R *11·. )V + _e:
_ 2 _,·. (-+lll·+ -4~* ) .v + 2ie:2 (-+R20 ~If >.(v 2 _-v.
"\_:. 2 >.::.r
. :.
<.: ' : . 2 p . ' . 3 (l)
• •
.
·.• .•
20 •
2 . 3
'·:
" ':'
i
.' . ·: .... ~.. :' l : •

. ·. u''> >:_·.: - .···: r

. I,
'•1',. •• '
. '
·. + .
' ... . ;
,.
'•.
. .',
'
.J

.. '.
' :~ i .
. ·.. ~ .

-;.:·:.>,where a11 \rectors ana. tields are evaluated at R'0 . It will be noticed that · · '·-
·~
~':
j.
; .f
·~·:· ·i:·.·.·~.~!. ::.;· the ~O term, Which was of'· order €
2
in ;, gives an order € contribution· ...
' ·.
'·
. ·~.~··

:·:·-<.\.
......
>':_-·to ;
. ·\·:.·
due to differentiation of . e 2 ifcriit with resp~ct to time. Terms of' .'· ~·
. ?,'·
.

J~d3v
/ ' •,. ' . ·:.'
· · · s,;.~ T.
· ..i ... .. . are ·needed in .·:; if'. : is to be correct through C"( €) 1 and 'i:.J,.,·~ ·IS hl!t:co:.s.~:r~ .l.
. • .A 'I .
. . -+ . ' · .. · .. ··.:.._._i_,.
cy -~-~: is .of' order . €. . ·.. ',

. The next step is to transform from the particle distribution function

. f
':··.to the guiding center distribution function F defined by:
...
.· (''
· '•
,...
. . ·r . .

(17/')
':'.
i,,

.. ··... ,'

. .
:. ·· .·
. ~, particles with parallel velocity in. dV
1 _· and, gyration velocity in dV av •
2 3
• .' •.. ,: 7~

"< : \ :·.:·'. .'

.. ·:.
Division.. by . d3r giVes
\ : ... ,_.: .
..
· . 3 f
'' '.··.
-.. ). d v, (178)
,.
'.

where# iv. !h~ Jacobian of ~0. with re.spec: to r .'at constant ·. :V.·. _:. ~cause.; ·.·.·
·:·,\.,·
... '
~· .: .

R'0 = -; + . _ ru 1 · +·. c:1" ( 2 ),

€ the· Jacobian is .. · ·; ·. ·. · ••:. ·. . . ..
·..
:r;
.• '· ·,';,
\•<,',

·' . . . .
·::.::...
. ...
,• .. ~.
,; .
.... ·· ..... ··
 t
! .~

l·: .
.,....... ·.- . ,;
84
I,

. ;:_;:_.;· ·,·_' • • I •
...._,
' .. '

~l (R ) v , must be expanded:·.·<: . ·.. -_

, .. -. i ...
Zero order terms 'in the expression for :;;, such as ., ·
.. . ·.-- .. \1 :·. {_
0 2 ~ ~ . • ' .

~ -··. o. ... about ;; because we wish j (;) in· terms of guiding center velocities at - . :. f, ,:,;·,::. ·. · >~
I ~- :, • •', 1 ' ', :;(:

.. . . -+

,; ·: :t an~' :0:~:. F:eth~::i:.:::ntF;;, p~;ic:::t :t:x::sa:::•r•, or 0, .' . .: ; c ;

'.,·~ ... ·.~: ·. 'F(V, ~). F (.v, . ~;) + EF (v, r) + :• ~·.,. ·•... ·· .... :.. ;-
In addition F("v, ~) can be expanc'!-ed as
. ·. . 0 1 . . . -. -. . '• :. :i
:~.: .. ·· ·After these expansions have been made, equations (J'~ ), (//~h and (tJCJ) : , ·.::~ :.·\<
·: · . ·.·· · are substituted ~to J:;; fd3v. Even after terms of c7( €2 ) are dropped, . : :;:, , .
' ·.':.·, :. :•~ '~ . . ' ;_·. . · ·.. ·. :_:· :; .·.: .·. .. . •, ··~i.
. : ::~: .. ·. there still remain many terms in the integrand. A large fraction of the.se · : · ,.. _ . ~/::1
'· .. ·
';.'.:.integrate to zer.o however by virtue of the following facts: (1) F
0
an~ F1 ;.o:· .:0:·--::: ...>'.]
are f:unctions of .v2 and v3 only through the combination v2
2
+ v3 2 ; ~his ··.·<_::./·'<::·f
.
~·:,. ·is not obvious :and must be demonstrated by use of the
.
"Bolt~mann" equation ,':::;.
.. :~,;·,\f '

2. 2 -+ . ::- ' ... . ':J

which F ·satisfies. (2) --~ · is a function of V1 , v2 + v3 , R0 ,t ~ _::: .>. ::::·(
'y',;· ,'· ::· . -+ • 2 2 . '·

.,_-:·:~/·~~ ,. (3): ,R20 is a function of v2 + v3 ' RO' t. The last two statements again ··.
.~~
,. '~:. .
. :~ .
· r, ' .... ·.. are ·not obvious but can be verified by examining the equations ·
~
ab.r,i . ,·
. ··. ·~

·}''• 'l •

..'. it · · satistY. . .· · ·1
.. '20 .. . . .. }
.' ,: --~ ..
,., ;_:,:·.'::·_.,-The expression_for J~a.. v 3 now simplifies to .' i.·.
..,
·'•
'. ' .• 1',
. .. ... !:
·:··.'
.';.
-. -~

F(;, V,· tj}_-r .·,·

•'--
·. r.
·: .·!
.... ' i-

t. __. ',<'. ·····._.·; ....

<; ~:·-:
•' ·~ .. ' ; ' .. ,

•'•\'
•· . .·. ·, '~

,:' '·. .··:!

+ ~~ • :(~
' •'

.... ···1
. . . (/8o)
'\ ''
~-

''
~

I 3 2 3 . ·...:- \'n
·. ..:!
~ .. •. '· : ~- •· :: ;
;

::: .~ ~
,··... -._

:.. · '·
'Where is the·: velocity· of::. guiding centers at The unit.· ... -.,
~: :.t
; . · .. l
. . ~ . '
•:
. . . . ·. ;:
f ..
vectors and co are at r ··too·.··
. .The .long vector expression
.
in the:. last.·. . .. ..~ .. . •.. ' i ~

.. '
.
. . .

'' •
,.
I• •' ·, integral of (/go) turn~· ~t- ~o. b~. : .>:? x ~1 ; . this is by- no means_ obvious. · So,
·~ ••
.- . ~ J
·.·. !-
I}
.. :{ '. . . : '',: .... ··.:' .· • • •• • • 'f • . • ,:

.-··
·•• •• • ' ' • ,._ • l . . .. < f
:. '.' .., .. ~- . ' ·, ' .. : ., .. ,.. -: ...·.-~..: . ···:-:··: ·,· ' . .- .. :,,
.. . .
(,
·, :•.
. • .. . '. , ''· ..
. .·.. ·_·;· ~

:..
.. ;- .. . ... . \

i
'.:
'-' , 1, I

...
. ' _

::·. ... ;•'

•.
~ lo
r
' I~ ·.. . ,;;: ·• . I ~ . • _l . ' . ' '
· ... · ·-:-: ... .
.'
,.. ·, . ~ ~ ~
·~ I '•I;' ;,' :.,'"'' ~

.... ' : .' ··,:'' o 0 '' :· '' '.... 0 0 •' • .' I : .',

' I ,.'.

. ::~ ' .,' . . "'\

.
'' -~ - .;-,.
' ~ . :(
:·, ~ .' . ~- ..... ..: i-:.·. ·.
~ ' ~ J .• ·,·. :.~ ·::~ ,_·=- . • .• :~ .••
': i '
':·.·_.. ·,, . ·. .·.-. ' .~. ~ . ·: ~ ' . _:; .
 .,. '• .. -
•.
~. '' ,...

., ...
• '' • ;r •, ·~,
!. .
r ....., ..•. :
' "I'

. . 85 . •. . ~ ,. .
•,.' '
: .' ·v

.. ...
. ..

:.:.· :>::J ~(!; ~~ d~v ~ Jd3V i£0 F(~,

•
-+.
t) v, t)
" • ' . ' ; • .,_. "' ~. • • ': I .. • • ~ • ' • • '

···;··

..
.':
.. ....... ' ... '·.•. · ...... ···,
; .. ....•.
_· . .. ;·_.
•·
••••· •. _.J; ."···
··-~ ··j~·.
: <The magnetic moment :per unit volume of :particles with guiding centers at '-;:_
1 ".'.· ·:_-· i.s ·. \
I'.·. •.
_(; ·..
'.';' _-. ~ ·.-~·<..-. .... ::;~:
f :' ....
! •,
....
t:': . ··, ..
... '• .. · -: :~ "·
....·. :~_-t-r
r.. i·:r::_:·,·· . ~ . . . ·V 2 +.V 2_ {;,
~ ~ r~l(~·- inc~)
,·I,,

~·.·.\.:: 3 .
.: : = .; . (II~) ::·,(.: -'~ .. ·; :·
.... : . ......
..,· .... ··:
.. ·, ·
·. .-
.·. . :,
'•.;
··.'.: ' · ;..
. ,. and therefore . . . ,•:_,' ,.,
·•
-.. _ .. _..,
_.
.

. ·: ~.
·,•.,··
:.- __ 1.. -
~
:-..··. • f :.

i Jv ~ ~Jc?v ir0 F(r, V, ~v

•... · l·
3
d v t) + x ?;i (I i3) :. ;
. . . . . j···

''· : .'
:. . .e. i.'·:.
,-.;_
... ::.·
•'
'.
;:·· ......
. .i
. . -+ -+ ,: ~ ;
·-·!·:.
.: . .j _{r, t) + c'V x ?::,
..,
.·. ·. ·. : . •·. uar>.:: ..... ·.· . ~h
' .' ~. ·.. ·;

.·· .... ,
}''•
- ..... ;. •'''
.:..-

theorem is proved~
··.. ·

l.

'2h9 fk''d'2Qd i ng -: e~1ui-';e

I ,
e~eki/§'>...kf~~""'-
~il;f.i.;-;J:;,:t~o>;a;~;;;.·~~,..~oo~;;I~' •6•~Ciiii::eru;:;;·ie!-'=bcee
~
51

:.·; .
.... :.·
-'i _.,:
' •, '·• '•{•"

' .,· •' : • I I , • ; •' ~·t.·. •' '.

•·. :1,··.

f ;_'... :.'
B. Loss-Free Ge.ometries .'(.:.
,, :
I'· t ~~
~ ' .
The ordinary 1a9oratory mi~or machine with a magn~tic field like
!-" ·: ·,

~-~
..·_;: . .
,1-'·.
·.that of Fig. (~) has a:"loss cone": If at any point the velocity vector makes
f·.
· :,::.·too small an angle with the field .line, the particle will escape thro~h the ·.
I:~~J, .. . . .. . . . . .. '
!{'.:<:,; .. mirror (by Eq,.~~). In~ the absence 9f dif£\ision.in velocity: spa.c~ ·due to
b" .. ·-···' ' . . . ' .. . ''· .
1:~ ,·". .
;,
• .'·' collisions With neutral or other· charged..:pa.rticlesJ there ·would be no way -,_ .
.. . '
....
' . '. . ~
.. . .... ":
~ adiabati~.uly· for. a part~cle to !!etC thi~
;
, ,.- ; ~ ~

[',. into. loss cone if' it did. not start • ..- .. "!
·.

· k·:.';; :_ :~ · out in it~ ·.But ~ practia~ s~ch collisions. do occur~·. ~e q,uestion · therefore . · ·
~-~ .. ' •. . . ' : • . .. .... - .. .. ·: .. '. . i ' . . . . .· - -_
> I •
··'·
if.; . ·.. arise's: :Are there any static tield.cozifigurations fran whicha parlicle .cannot ' .. . . .... ·::·-
;·
; ~ :

•_·._·_:·.:-<.·: · .. ···::.: -:~-- ...

1:\_·.M:_ . . ,:. ·.·-·.. .·. .·... .· .. .··" .·: . . . , .......:_._:
, •.. ; ·:· ' .: ';· . <,. > f '.: • •.; ,· '-~- •• :· ... ·.~ ,·<; ~: _:.
r.-~ . ~. • ·.. ·~ -~~ .. \. . . ··.-. ·.··. . ·.... ,. - :.~>:·. ·.. ·" . ..· .-..'· -.•~ :·~- __. . _.. _:_:,-~\ : ·.'·.
·,·.·.
. ' ...
r~··:·~ ., . ._. ,. .. -_; <,.";~- .. ~.... ~.. : ·. •. , -....
..
·-.•
--~.

,..~ • . '' • ·: ' : . . .• • -_ ..... - - . • •. • , '! ' • ' ' ~ ..: .. :' 't ••· •••· • . ' ~ • :. ·, .. ·, ~ ........ . . ~}

'J •• . ....
 86
I. ' ..
.... ·.·
'·:
(~'I) ·,
:-,·
. ",;, escape so long as it behaves· adiabatically?.'' (Non-adiabatic behavipr is another · · ·:.':.:

possibility .that must be considered). Although ~he adiabatic _invariants "'do n(?t
··-.:.
. .
. ,•: '.seem to tellba.r to find such a configuratipn, they do provid~ a way to test f1l " .• ~ ~ '
c
l
'.:

....
.. .
prc:pos::il. ore . The criterion is tba t all the longi tu<;linal invariru:t surfaces

passing throug~ an arbitrary point, in the co~taining volume.must at no.other :,.•

.. -._._. ':'·'..

place intercept .a physical obs:tacle,, such as a vacuum chamber. wall or a current·: ': _·: :.. ., ·.

'' · ·. ;~ carrying wire .. Since the s.urfaces··are n.ot simply nesteQ. and differe~t velocity. ~- '

::_,·.. :. ... ·.. · ·._, ... ·,.'

'
. '·'·' . 'vector directions ~t the same point.in space ~orrespond to different surfacesJ
•'
;_·..
. .:.··
·:···
consid~able'effort may be. needed .to calculate where the surfaces lie. One .· .·, ..
. ·. . -~ i'- _:··
,J ••• •

lo~s-free configuratton. is that o~ t:; circular current loop (Fig .17;: ) • For tbf!rti .' .. .~ ;.•::. '-:

.,·
.•.

·•.:
'.
·'.
matter the field of Fig.;L is loss :free :i,f the external returiJ. flux is included . ,' · · '·." '
...... ~. ' .
· '·. · ·· ~ . in the vacuum system.· The guiding cent~r of a partic~e remains. on .a line·. of
.... .. ,
,, : .' .' ~'
force and repeatedly transverses it.·' The problem is to. supply
. .
the ~urrent to .· .. ··
: . . .. . . '
1.:,.
the coil and to support it mechanically withou~ int:r;oducing wires .and other.
[':"'···'·:

'<\
~~\ ·. / .. :_: '
__
·.r:.
,, .
obstacles into the path of the
. .. .· ;..... ·.. :.~·.. ' ' . . .
~ines of force •. · ..
I •, ',•,: ,•;,
·'
.. :· ::· ... . ,
~· ._ .... : ... ~ ...:·~--~~· .·.~;
. . ~ .... ' ..· ·.....·;: . . -,_r . . '
' ·. ·. -~.' '.
. -~ ·... f ..
.·.:"., .
.,.
.·: ..
,·.. ' ;':· . . '.·.

~ .. :
r, • • •.
I . .· .. .

'.' ·.:· . ..·.··

·.. -
·,' . .:.

;·'·
.. ' .. ,·.
·"'
..... ~
':·

..
,_·_
... '
.
·., ::... .... .. ·.·, .
'; ..
.... _.;_- ..
. ·.·. ,•. ,• ·.·.·'
·..
~ ' ·;
''
. : ~-·.. ·.' :,
···r:·
.
'\'.,

·.·, • ,, I 1.'·
.: ;·.

... '·:
....
. . -~--
, ... ....
.-";·.: ..
-: ...' - ·>.·
;·'-,;_:

..·. ~ .... .. :
. . :··
'-.' '····
.· ·
•'

l. '.•
;
.·· ,: ..:
. ....... ;
.•
.. : , ....
·
.·.
,·.i' .

' Field. of a current loop is loss·. cone free·; I •• • I

. •:'
(' . ' .......... .o ' '"I ·~·. '· 'o '
'•,' .. ' '•':
-·U-
. :.._;, . ..
,_<' ••• ·,
:· ... 1 ~: '.
.; . ,._;
I , ~ ' • ' ' ,'

. ·.~ .. .
.. ' ' . ~ :' . •'. ' -.. . ' ''<' ···',.
.k ...:
... .
.. ... -.
·' ..
\ ·:

·.•
! ' •

....
:
,l ., ..

...
·;-

... , ...... .... .·.

.
: ·,

·
. .~: .. ' :: .'
....

..
.. ; .. ......... • ·' 1
•···. ,· . , .·
·:-··
. .... .. .
 . .",!
.•. ·,; '•

'
, I 'r !,;' ',

·.·.. :.: ,.•, .. _•,. ·,

·'
~-. . . ; ~ •. ' ' .
' '· ~ ..
exten~ively
':·,.
' ·:, Another geometry which has been studied (3 D) for loss-c·one: ', J-;
'
.. .
',,_,:
. .. . .' . . --~ : ::~ .
f, ·'
·free :properties is _the: "bumpy. torus". · ~t consists of several. discr~te Gircular
.·
'. /
," ·' ' •: '• I., • ~-~
':' ·.·! .....
' . ~ • i

:'. -,·: windings.arr_anged around .a tor_oidal.vacuum chamber (Fig.J$) much' ~s if several _J."\;(;
.. -~
)

" ·;.
·•. ·, :... . ,·.
mirror machines~ we.re arranged ~nd to end and then ben;t _into a c+rcle. .It does ' t .. •
·:.. · :,_
~.-:.- : t...... . . .. \ I • • • •, i' · ...__ ,.,·- •. -•.. r·. '..... ,_-_. :~
: : · · ·:: .:. indeed have regions .where a ;particle will not be lost· regardless of the direction •. ·.
f·· "' ~.: ' I • ' ' •• /' , ; •.· : ~
'J'!

of its velocity.~ecto~~ and the coils can be supported~ such a .way.t~t.the

'•;"
.·, ' ...··." ;_ •.
•'•
'•.· .... ,,
- ·:,;:
... .:. ·..
I_''
. supports do not interfere, with the containmen~region •.. However, losses.~till ~· ..
',· :--·.
..~·:·.
•r," .· . occur. du~ to a rather no_ve+ type. diffUSion, . which is a consequence of there
.•. . .-· ~

'.
··:· . ·> -..-·: · :· being many _inyariant. surface~ :through any one ;point •. If the velocity yector :·· ..... •;
,•, ">
;.··:· ·.<_
of a ;particle .is change'!- by scattering. at a :point, the ;particle changes. . , ., !'.:
. ·,··
......·-·
·.·invariant surraces.and ~y-change to one that lies nearer:~an obstacle at some .
. :·,.
I'. .,, , • ~

other ;point .in the system. When the.;particle reaches this second po~t, ·it ma~.'_.:··_,··.:-f:<~:--_ \'
again scatte;r onto a third invari~t. surface which lies still ~ther out,. etc.:·;_: :. ·: ..

:r · ;-··._. >.· Thus be.cause of the :presence 9f many invariant surface~. at any one ;poin~,
·.·.· , _. .. . ·'·
/,
. ·.
.
'• :particle rt::B-Y still -w:ork. its. way out of the system, but more slowly. than: by.'>~\_
' . ' ' . . .?D ....t......
· sc·a.tte~ing in, an. ordinary mirror machine._ Lauer, G.·&,,.,~,~~estimate that at
•• ·:;.'. j • ~ ~
\.'
' -~- ' .. . ,:.·
,-! •. ·
. ,. .·
.·.·· ·,· ·;·
~-- ~·
·..,,·
I • • I ...

.';.:
•·. ··.

\' .... .. . ··..

~

..t. ~ . •, . . •
· ... .; :-· 1

.. ; _.,_,
....... ·
. ; •'

..... . ~\ .
.~ . . .
'I' .••
-~ : . ·.
./ ,'
"~ . .. ·,·.
.-., . . L··t· :·;.
. : :- ~ :- .
.... /': .· .. ·.· ·-.:
.., : . :. . _. ::--
:' _\· . ;. .. ~. ~ . : '
.",I

.... .-;,; •'

.. ... :'-.·: ·,.~·· ,' .
. ··:· _. ...•. : r ~-
....... ·.
.. '

.... '
·., • '· ••.. ~ ,' < ~-
.·!' :•;', . .' .
; • ~ ,. ' 4 ,. :~- • ...
! .·
•, ··.: .
.
··. :·'
.... •. .·:· . . ~ .. . '·
. t . ~ .
~I · . •. ·: .> . ·: '......
' ; ',,, :; .

~-. : . ~ .. ,: . : ~-
. • . •'

. ·'."
-~ '·
,,
' '' ~··. . ··:; "·' ...
.~ . '/ ..
' '
.. ... ' ;.·. ;-_-
••. !
.,.· .
'"/-· ·.
· /

.·:
. _\ .. ·
·:-,·' t.' '' ·. /' .. : • ~ I ',~
.·... · Bumpy torus . -: ( ·. ·: : -·:· . '··
·., . .
' ~ .. ., : ·....'; : _:
:· ... . ·.': t'··· .· --; ···--,-
.. :
',• •• ,'..! ~ig • J8 :. :·,' .'. .·; .\ :. : : ·. ''• ~ ··..
·'
,. '
. .. ~. ! . ~ . ~.......... ~ t .'· .'. . • • •• ·: .; . ·.: ~ ; :/ ..
·:
..
~ I .
·' '·' :· ;: ,·...... . •.:.
·.·.. 'i. ~~.. . ·••. ! ... . . .-·:·.-,) ',.:_'·:
,.... ,'··
.· ._( '•.
.".·
'; ~·. ·' ·. ' ; ' ~- .. . / .... • ,·. ;I,:.;·-~
~-:

' ( .. ' ~ . :•
 - 88 . .:·; ·, t ~ ..·J·•. 'r\ ~~ ~
'. ': . '; '. ~
t •• ·.--

. ... .~. . !.'.; '..

least a factor of' 9 increase in containment time aver the ordinary mirror

'.; 'I machine is possible with the bumpy torus.. Of course a loss-tree geometry '
',,
. ~
'l-L.
' ,. :·.
:.f

does.not eliminate losses by ordinary diffUsion across the lines of force •

. . :· . .' '
. . . ·.·.
If' the loss due to the new~ mechanism in the bumpy torus is less than that · ·
,' • r ' ~ .l, '

due to ordinary diffUsion across the l-ines, :J; ~;"tor practical pur;Poses '. { ·:'< ,' · .:,
;\. . ··;f.·::'::··
loss-f"ree. The hydramagnetic stability of };he bumpy torus is another question. : • • ·"·_f
'· .....
····.
Loss-f"ree geometr1,.es are of particular interest in connection· with
.,.·• ..

.. :1.··
. .. ,_·

. :. the chaDge. of e/m inject' ion methods· which have. been p~oposed (3l), (32 ). f~ ~
;,· •• '
• ; ' ~ r'
r, .

..' ~ ~ 'r,,

building up a thermonuclear plasma. In these methods particles such as H+

2
. ,,.·

., . _
or neutral H are injected into a .magnetic field at energies of many ~ovolts .·

or even an !1ev. ·Some of' these particles undergo a change iri e/m by means ..
·-:.:
. ~ . . .
of' collisions with neutral background gas or w:t th previously injected particles, _.,.·
' .··'·...· . .
and thereby become trapped .if' the geometry of' the field. is p:roper. The time :.~· ·. . . · ... .
.~- . . ·.·:
~-- .·. · · · · scale fo:r such systems to build up have been calculated to be as long as ··'

't'·r .. · · severa.l minutes.

,. '
:..~ of; par.thleo al:reaey · tr~:P:P~~
·,'
• ,I :

' . -:'
. One loss
.. mechanism is scattering into a loss .
..· !· ,· :·

·<.·· ; •'region of .velocity s~ce, so that a loss;..f"ree geometrY" wo;u.ld increase the ... ·.. -· .. ··· ... ··
.. · - . •.
•.•
.. •• . • ~-
, • • ·.t
-'<

i•; ., ..•

·....·.·
· :f':tna.l density. ,·,
. '•'
."
..·' Another advantage that a loss-free geometry may possess. is that of . '·\•'
·..
~-~
·.......-,-."\
greater plas.ma stability. Sufficiently anisotropic velocity distributions .
•,",'
-~1

unst~ble( 33) '

. ...
'
• I '\ ~ ,.

,, ·. t ... are knowntheoretica.lly: to be and the instability may pave been ' ; •' . ··'
,. '
f:

• observed experimenta.ll~,.C3 4 ) in an ordinary m.irror machine.. I f a field. had

-,_. I ..: . · I

,;

.. ~ ..
i ;'
' '. '· ..·
,.·
! ,.. .·_.·.·. ,_'. ·,· .. :.
'·. ·- . :::- :..· ·. ~

' . :, :, '. '' • ~ I

'·:
.. ...
,- ,·

. ,: ··' -.. -..._.

.
;' ~
; ' ....
·: .-:: ..~ ·....... ·. :
'· ...... ,l ,'· ••
.'
··'; . '
.)_·· : .. ' . . .. . .... '•
_-. .'

. -.....
. ,1. ·'
·;·., ., •' .··· _ _.. _.,. ···:
i . ' .·· . . .- _. ~ : .. ,·. ;
..· •• -~ J • ••

I· .-.. ; . ·:· . ·. . ..~ -: . . ..· . .·

··',
-· .. :··. ·... ·· ..
... ..·.. .
' ~

I • •.
.. _,...,..
' I ~ •f '• •
. ,· . :'. ·,: _ ._,
. . .. : -~.., ·.
\,

.. ~ .· . .. ·. .-. (;·
(

- .. ·. . . ·r:·.;. :; ·..··:" ... ·:· '.·· .. '., .... ·.

..
.' _· ... '• '
·. ~ ·. ~' '

' ..... - . .- ·. . .. ·.._·;··,...

' ' '
.1_,. I\'
.
~ ' ' ~ . ' ~ .: •, ::. ,'' .. _; . . ·' ..... ··
•,
.. . . .
' . ': . :
~.

~
~ .. ·• :' ·.·. :·:·: __ '_,·. ;. .. .
•: ' .. . -~
\ '
. : •.•. _

,·.
:

•.;
.1·_
 ''
f . • '·•.

,,•, ' '

. .. ~

.'• ·.
I '~·
.i
. ·< ••

c. The Geomagnetic Case

The beh.B;vior tO be expected Of adiabatiC particles in the earth IS '•'

• ~ :. . f':i.eld. -has been studied in some detaiJ.. ~~).. The geomagnetic fiel~ approximates ·.. · ,, ,· ,.··. : ·,
·· · ·' ·~ Car¥~lnp. . · ·': ·." .. ·.·~.·. ; ·:·;
;; ..' ,:··_;_that of a~ ef!. 'iT~ 1 but the system is. not loss-cone :free because of the . ,:<·..:;::-. .._~:::·:~;·f~·:<;j_
presence of the earth. Neither is the field exactly azimuthally symmetric about ;: ,..:, 1; :.. ~/!
1 ;
I ·~J ,·;?. ·-~~~:.>~~:· -~ ·~~- ·,_,_: <····. )~:~ -~
i<· ·· an,y axis. 0

Neverthel~ss 1 : long as the three adiabatic invariants are conserved, · ·,. ·. :' . !i
:.;··:-:.\~-a :trapped charged particle ca.ruiot escape. One interesting theorem in 'connection.'(:·' <.:).:':'·~-~~
. ' • ', •'. • I • ' ' . ~ I ; ·.,: ·.r.:' ~. ''· ·~·~ •: ~~

0:.· with the ~apped (Van Allen) radiation ~~rounding the earth is a res~t of the ·. ·. -~:, · ·; )
;·. ;_:_:·;~. -~

j;>~· '<.". canonical form of equations (13 '-): Contours of constant· B . on an invariant .. :..; ..... ·;: · ·. :'._,-··.i~
,::·•. ·•.. ·: _;;. ·'-· .'. ,. '•1

,. ''' ~:·,;, surface ar.e also contours of constant ;particle density provided there. is a .. / '•> ·.·: ;}
,. ., ' ... :. ·.< ·......
[/,; ~:- _steady state no electriC :rield. and
c~o)

' ,·: '/ ' __, ;,~;

11
:··, .. ·.· · · One as yet lfllSOlved problem in connection with the Van .Allen radiation' : ·,,_ . -> .Li
r>-· . . . ·:.,. .;_ . !. ::_·. ·: .....r.~
: : :<; .is· why the density of the inner or proton radiation belt falls off so rapidly :, _1 .-·-.:· '·: , : : : \;!

i ', ' 'm th radius. .· it observed that the density oi protons is (.36") Vi th energy greater' '' /~ ,(.i~
[~::\:~~:,:':·,·than 75 Mev falls from a ·maximum at about 10, 000 km.. :from the center- of the

~~:.'
~;
>·.·-.·.earth to :practically zero: at 121 000 km. in the equatorial
... .-,'
plan~. If the theory

k :·:··.::is to be tenable that the protons are the product of ~-decay: of cosmi.c ray
··: ... :
!) . ·. -_.peutrons• 1
1.' '• . . :. .
it is necessary to explain the large decrease in density beyond
. . ' '

:.· .·. >:.10, 000 km. The. de~r~ase in 'the source strength is nothing like this rapid. One
.;._ .·: ·" ·~~- ...~. ·, .
. ~-

r,··: . '-· · · :::::::~ :~7~i:::: :~::e:i:::::s:.::::.::~d:~::::a::::tudinal .·-······<····· -.••••(

f·.O: :· oscillations, the particle would eventually be absorbed by the earth 1 s atmosphere. • ·:· ..> : .-~
r:i:; , Such.a :process mi~ht be!described as scattering' intoth~ los.s cone by' the magnetic: <.'··;>·;.·:1·
r :.
I . ' . . .. . .
I .:· . r,_:·)
~- f'ield, rather than by other particles ... Possibly the gyration radius of a proton .._:··:.__ .,: . :;.~

i' . of say 100 Mev energy. is so large at _12,000 kiD. that non-adiabatic effects ar~ ·/ · · · .. • •• .!

I·.~ ..· importan~: ev~n: in a static. dLLpole .field, as suggested' by· ~fuger. C3 ~) . l:t is . ' .' '·' ,:;
t /

L·-~-:·:.. :.: also ;possible that rapid spatial.variatio~-of the field_ due ·to Al.f'veh -wa-ves is·· . ·.. ·
~·-< :.'_ ~es~~~ibie, a~ ,~ested. by we:ch an<l. ~~~r. (a~),' · .· . > _. ,. ·.·. ;-::· '· ;' . . ,.'
I;'· ·;·j::. ··~· ·.·~·· ._ ... · ···.··~ ,. . _;.
·.r~:., ·:~-;~~ .. .. . ... ~-··· ···,.· ·.··.··· ·.. :·;····!.··:· ~--~ .. -·::·-.::_:~:.·:· ···'· .·>·· -~ . .-;: ·f
', t , t ,_ -< <.. .'• ,," I· • 0 ·~ • 'f
 :; ' _,

.. :
-. 90 ·.. • ,.. .;.··

'.~.

Non-adiabatic effects in a static mirror geometry. will ,pe reviewed

::<, }\·' ·.•.
•·:·.
.·'
•' i
,• ,.
in more detail. ·in· the next section. ... .'. ~: .• ..
.·.i".l ''.'
•.'
':, ..
·,. ; ..-··. 1·!,
..,
'.;

... •• .J~.
';.
··• . ..., )
- .: ··<
.... · '.1 ., · ... ~ ·.:'
·.;-·
'.·'· -.. '. ' ~ .
. _ .. :•. ,.'

). :. ~

.'.\., ·. ·..'
,•

. ,.,I ......
..
· .. · ·
,. -~
··':•
'·
..... · .. ~--·:: 1 • •. , ..

.. ·.
' ·. .
~·
~ •,' ' ·,,(·,·.·
··.·.. ' ·.
" '·~
.'•·
·} . ·: ·.. : .. -
',· • ..
.. · ..
·.· .' ~- ;, . ' '
,.
:. ;:·.: ~- ,. ;

·':-,_.
:
' '. '· ~-
. :-
'···. . _.,\ ,· ·,.- ;··, .·.. ~ 11 •

,._.;
./ ··.)··'.
~- !
· ... . .. . --~ ·. ;.·\·

.•·
::..;.:
.:. .. ;· : .·,... ~ : '.
·.. :· • J ;

.t .. ·..
) ~:. · ....
. .. .·.
.... ,!",
_:,·
.._.,...... ·.·
'•·
·~ ' ,-.·.
'.'.'·
.. ... ... : ~-· . .
;) J' ·.·. -;·'r,'.
;•.
·. i
....'/-
-;:
··.'·. ,·
.···
...
. .. : ..
:'••
.., ,.·.
'./-':·:·
.· .

·'· . ,·
. : ; ;i. ....
. . : .
. ~- ' .
'

·'··· .. ·. ; ... ~:.

. '•
·~ ... .•."'
···:·
·.-• ·,-
'.,· ·:,·,
":··
'• '·.:: /. ··. ,·
'•'

·... .- •· '
. . . ,·
· ..
. '
'··

. :..- .·• :: .,· ..

,::- '.

i·;'·
' ' ·~·
;, •, '.' '·~,· ' r .• ,'
i .•

. '.. ' ... :.

; ~
).-
·.;;• ,·, !'
.. ~ ...
. •.·. ' ·
...
.
'·.,
..... ,. ,•,; : ··:
·-·-~ : ·'
.I ,.. ··
' ~· . .. ~· ... '
' ~- ·, · .. .·.· ... -··,;
:."•

.. ··.·'
...· .: ~· : -
_,'
'•. "','\
., I · ... ·
. ; ,· # ~ I . •
., .·
I -~-
1: .. :·.
· ..... :··
.. ~ .
~ ·, .. .. ...
~
'•"':

. ~· .. ·' .•··
~-- "..
··:.:..-··' . : ':'·

:_;.·_;.,. '.. '

'~ f.. o'
..
L ' 0

·,. . ~ ~: r. . .~ ..
~
.. ·.· ·
·.· ..
·· ...... · ::-,". ::; :. ·.····· . .... ·.
 ,,
,_ '·
'·· .
. "(• .. ··: .·- ~-
~ .. . q;:
"~· ;.-
' .
; ·~. '
. , :· ... .. =· ·.·. . '\ ..
. ' t' ..
·,_,'; . ,._.· .·

-.·. 91

vr. tTh"RESOLVED PROBLEMS

Non-adiabatic E.ffects in Mirror Machines-Num~rical ·

r .. : : ,·. A.r ~<~ i

_[;;~, -,·:· ·> ', .' .,

.. ·,
,._ L
: ::,, adiabatic predictio~s. The question is probably not capable ·of a general answer-1 .
. - . . ·.. :''

,
·,-,'
' ; •' ' ~r •
and each situation must be examined .independently. It is possible to follow - - . ~· .. ' .

particle trajectories, by numerical methods and to compare the results with the- •
'-·
·: ' ~.
(a~)
.

~ .. ' . ' . . adiabatic predictions. This has. been dom for the. laboratory type mirror - .-.-..: .. :
!';.
, ... . ·.. ~

. ·: ..'
• 1 •..
_machine, or more precisely for many mirror machines end to end in a straight. :!L~! -· .: '..
~ ' :-,··
(fi~lds taken ·a~ periodic tuncti?ns of
' ~ '

i. ' -
line :z., Fig.l9;).

observation was made that .as the particle oscillated between mirrors the ~etic:
r .• - . . . .
t ..-.; ',

,·~· .. .,
• ~~ •
,'
• •.•• l
, .,
: •

,.. _.. :. :: moment could vary by large amounts _b')it that· the variations were highly self . . .:: . -1":., .....
~
' ··. - ~ .
. . . ·. '.. .:
·cancelling from on~ oscillation to the ne.xt •
·'
T9 be more specific, the pa.rticJ.e · · .:.:
. :;·.-.·· • 4·..: '" .. , .... ,
1
:· :-. ·~:_:_._· ·was starte?- off at the ;plane midway b~tween mirrors as shown iii Fig •/,3. .The·-··.-·-·
. ' '
-- ' -:.· ·.·.:·~~.:.·· -.. ··,- -·
..:..
/ .. ·.
·,_
:..\'
··, ..
;;' ·- ·....
'.
;;
,·

.. ·.. ··-.... .·· -~' '

I. .~·. ' ·.
I',. .
I· .. , __ , ..
... ,.·
..., - '.';

'8 : .. .·
.~ '
j'. . ~ .
,.....
! . .. -·. >
I, •
. ··<.
.}'• :... ·'· ',
·.·.,·.
... ,_ .... .: . ,·

.: ~.
,:.-.· J • ;1.
,_··. ..
.'·
., -:~ . ;,,
,·.;: .... ' • : .: j • ~ • • • " '

· ..- . .... , .;,_· :._ .·· ...... ·' .•:- ·,

i
'. · ...-·
·-,.: ...
' ~
'_.'"'
..., ·-~
. '. )~ ··'.
.;·-·
.. ,' :-'
.·. .. ;: . ·. ''" ~. \ ·. ' ·.' '
' · '< ·.·:·.···
"' -·
,,·,..,' . ,- .. ','
··.·.,

' '•

A periodic mirror geometry for which nunieri.cal calculations have· been:'

compared to. the adiabatic predictions~,:. - ·· ·
,I . •
.. _.: :~
.. .. :' . .
. -~: .··:.. _:::~- F_ig~· >.< .... ' .. '
·:·.>-·.:_ .:1:9.. -'..,:·.... ·. .
~

. ... ~

. -: .. .,_. . :
_.:·~: . ... :.'. . ., .. . . .. . .~·~; .. ': . ·.' '. : . : _·. ',... \ ....
,·
.~·' : . '
.......
< • • ~ ,; • ,.
-~.

. . .:.·:;/.~::~->~- ;:: -~-- ·_., : :..:_·

•:

. : I . :' ·~ • . "
·<. -~~ ·:,:!.-.•.· .. ··..-..-. ·: .....-
'. ~ ~ ' .'~- .....
... -.-.,
, ..... I:--
'·
,
 I·
f.
!'· .. ·

~ (-+r) -+
. angle between -+
v and .o is o, while is the angle between v·
.L.
and.

the :plane of the :page. Each time· the :particle returned to the midplane, new
I '
i

-· 9 'and il. were computed. A cumuJ.a.tive drift o:r o woul.d imply a gradual
.change in the magnetic moment; in an azimuthally symmetric fieldJ conservation
.•. ·w.

of the canonical a.n.gu.;l.ar momentum . ~e :prohibits a radi~ dri:ft, and there:fore · ·

'"·'
i. ~
. ...· ·2 -+
B(R ) •cannot change. Thus a change in v.L /B(R ) must come :from a change
0 0 .. ··.,
. . . ;,·

in v.J..· = v sin o and therefore :from a change in. o ·since v is constant. .::
'fu<t/,·tA~i've/.1
. . . . Phase :plots of the ty:pe shown in Fig. (g.:o) ·.:ere obtained. The numbered :points
" .
~· • •
:'
t

.. '
• •
represent successive traversals of the media...."l :plane,; zero being the initial
t.' •

·' ·.. · ..· :point. There are two types of :particle behavior evidep.t. The first might be

termed stable (curves .A or B) and the second type is unstable as exemplified

. ·'

!,'
· .' · by :points. X.
'
The o of a stable :particle may undergo severe variations, and
~ ': '.

. the l13-rger the gyration radius, the larger the variation. However, the variation . .;. r
', ', ~ I

... ~ . . .. '
I;·,;. ,. is highly self' cancelling; .the :phase :points always appear ~o :>all on a well- ' ..
··.·.·

defined curve. On a large scale :plot the :points appear to fall on a smooth curve. !':·

·>.. ....'to within less '

than 10-3 degrees in o, and that much scatter can be attributed
' : \.... · •. 1 · · '
·. ·~· . ..
..i • to numerical errors. Thus it appears that the stable points a.z:e very stable
:.
indeed as a result of a "memory" from one median· plane traversal to the next.

:.. , By cont~ast the unstable partic1es show no such memory and form no \ ~- . . : ' ·.:· .

• .. ; ·,,
regular pattern. They usually escape within ten or so mirror reflections. The

·.·line of demarcation between the stable and unstable regions also is quite sharp,
I ......:~'· .
,. ': ~· .' .
. .. ·.~. .
: • t .'
alth.o:ugh ju~t how sudden th,e transition is has not been carefully studied.. It·
: .~ . ~ ..
I'·· ;' .....
I . . .· ::
,.. .
,. ' ~
would be interesting to investigate the transition region more minutely. The ·.·'
f.
.•...
•/ '_i .. ' unst8.ble region, which begins at the loss cone and extends u:pward in o
'.
' . to the line o~ demarcation does not a:p~ear to cover the entire range of ~ tram
i ; .·'""
·[·.:_'··...::;.·
0 to Near . ~ .,:: 1r/2 there· are stable ·curves of ty:pe B .even belov the
I . . ..

~~·, >> .: '>line. of demarcation. The fixed points at the center of ty:pe · B ~atte~s . ·
[' '··:. ·:·: : ', are. ~~orously fixed by virtue · of the symtnetry abo':lt the median plane • · . Because
..,. ··· . . . . '

~: . ' .
• •. ,i ~ ' ;. ·. ! ;'.

. ~ . ', l .'•. . • ...

-~ ··,·.~:.: ~. •' ··~·' ,· . ··.·
 .•.·.-·
~.'
.·..
. -~·

·. -~

: .,._-

·'..
,- ~· .' ~ ..

,.
I
._;,
....
! .·

J,_,,
:-··

·.·.. :.... ., ..

·.:: I~. ,

'.
I .. '

3
_ .. -.

: __ ,,,

..... :·:

------..-.......:::1~s-ta.b/.e ~ v~tsrobl­
l•i.& 11-f. ~c.».ra~o.·6wH.

.·: :•:
.'i',
.
,'J .
. !'
~-· :

... .···
'• ;.
. ,',"·

........
"
'> !:'r'-J .· ~..
~ ,·
;'_.__ . '
'I i

._·._ ..
.·
i i

. t:
~

,, -~ .
• • •• ; .. 1_',
...: -~- ·.
·.")
. ·: .•· ~ :..·•
' ~- ..
'; ';_ ~ .....
. ,·
... ·, · ..',
· ...:
r ·~

.. ·. '·:·
... ~-: '
.·., ·.
'\
'·.
:·.
,; .'.:.
..
... _ ·
. ::
:. ; ..· Phase plots for :pa.I.'ticle in the .mirror 'geometr;y' of Fig. J..9. ···
·.-. ·' ' • d ·,1 :·; .. •• ;.
-·:, . .
,.

-..· ~- .... _,··:· .. ,'·•

'·,_:
,. .. ..... ··Fig. r,
..,,_
ao··
'· l.'
·-_:. ...
_;· _'l:
 the unstable particles esca~e in a few reflections, ·the demarcation between

t~e stable and unstable regions is the loss cone for ~ractical ~ur~oses. It

pas been found that the unstable region vanishes when the ratio of gyration
. radius to the distance between mirrors is less than Jome.thing like · 0.03 . ·:··:
. ., .' ' .~;.
. . . '•I,,).
. ·..
,,,·,. I

The behavior of ~ticles within the loss cone as they go from one mirror.
' . . ,'
section to the next in this periodic machine has not been studied. It would be
··.;
worthwhile studying, since these ~articles are analogous to those in the .t

bum~y.torus that.are not tra~ped in one section but go clear around the torus. ...
.,

·.A theory of the behavior observed in Fig. ao,.- has been given by C~ikov. (.:37), '·
·.•i
·.;

.According to the theory ~articles are lost which exhibit a resonance between

the fundamental longitudinal ~scillation frequency (or one of its harmonics)· and -·;•

the gyration frequency. The theory has been given only in the a~~roximation

of straight lines of force. This is equivalent to assum:i..:ag . "Y • BI 0. The

inportance in this theory of ~ine curvature would be worth investigating • It·

....
should be pointed out that the fixed points at the cen·cer of type B ~atterns
...
in Fig. go :r:o ssess this resonance and are ~recisely the ones \,hich cannot be ,.,· ·-~:· .... .:_··, . .
..

lost because of the symmetry. These ·are however very s:pe;cisl cases arid constitute ·
.~~ ..
·:. ;.:
.· · · · a class of measure zero.

'· The type A stable curves are predicted qualitatively by the first
. .,... '
-~

two terms of the magrietic moment series (Eq. 153). From the fact that the·

canonical angular momentum ~e is a constant of the motion; it is ~ossible to ....

write Eq./fil in the form f(o,X.) =·constant".: The field used in the numerical

calculat!.ons .of. Fig. ~0) is derived. (in· cyl'indrical.coordinates) from the ·vector .
.. .
~otential: : .· :~ · ·· ;·, . ·:,'

... ..: :-': ,'' '• •, . ''

. . ···· ·._._:· .
. ·:..
·.
... ·. ·.•
... ,.
.
' ~-

•...
I
•

•. ....
. . -·...
: •
·
!
. .. ·.
··. \" ~ - . ...... ' •' .
~ I '. ' ' ' ,'
. '.,
. .. . '
_.:. ::;

.;
··. \.
,•
.· .. · '
'
. '·
.. .

- ...
 ~'
.
·,: "" • ' .; I.'. •
~- .

'. '·. - 95
. I
'"'•.·
·,.
·· .. ::.-·:.·
i

........
... ··.' '+
1,"·
,· ..
-~'. ·'
.• ... ·. • . ·,.' ~)'· ··:~.··.· •.. ·.·'. . .,:·;:·-~
':_;_.j· j

21rz· · ,. ·~~-,;:;>.;·:if/.
where ·L·. is the distanc~ between mirrors~. p is is L ,a i~ 0.2_:;'. . , .
... .. .~

"'' .·· and B0 :. is :the field .hal:f'way from. the mirro~ to the median :pla.Il~.: . With th•iB . . .. .· ;:;~r
.. .' .: •. . . ·" . \. ' • .. ··~ ·. ;~.!. ~t
':.···,1. •
field ,' .; '
.-~-: .
,<{·I
..
,
·.,
: -~ . '. ... ·
·.'. ·'·./

i.
I.··
..
·
· •. -• . 't
... ·

. ) ... .. f(g,· A)
"
= sin
. 2
5 (sin 5 + a .cos 5.sin A) = constant,
i. '·
.·.-:
.'·
~- :/. ·, ...
I.
I
where ••• J '

I .:.. :- :,·. .
.. ...
~
. '·. ~ ·: _·,:_
i
) ·. . . -... '
·' ;; ......
.v g,)....· .,
1 ':
i' . ."-l '
! :_.;,, '•
}
mcv.
",."" a=
[·. ;.,·.
~-·' :: .· eB t
'
0 '; ; .
·, ..·
. :: ~
.: ·-~ : )
.. '·· ......
·--~ :: ;. '·, ~ ·,.
.. ;
._-
I

: I' -~·," • \ :.: ~.' _.;• .-.-

.:• ·.
. :;-·--.. .
..,.,_..·~::; . :and~.: Po is "the solution of ·. :_: ·~·
. . . ~.. ... ·.. ·... :"
··'
·.·.
. .. '!
''
·: •: :·.. ,·:. . .. ·. ;-:
. : ··.;
....... ~: ..
' ·... (J 2 g) . ··•,.t
.:. . :

\,i
.. ~-- -: ~·
t-·;.
-::..', ::._
Higher terms in the mag~-~tic moment series migh:t give· more de~ails of the stable · ·.
•..

orbits~ -t.ae ~9; H .li! c::~\1'9S fer i~":~~,. But the unstable behavior could not

be predicted f'rom the series.·.

. . .

An investigation of particle orbit~ in. other geametries1 such as the . . ~. ~ .':

.,
' · •. •j

bumpy torus ·or a mirror machine without a median :Plarie qf ·• symme:trY1 or without .
~ . :;.::. ' I •(

az::.:muthal symmetry would be of iriterest·. : · ·~ ··. ~\_./'~<': . A- ....... '.

.. ·.. · .. ' . .' .

'.•
. "

B. Non Adiabatic.. :Effect.s-Exper:imental ·o

\ '.
. ·. . ' . ' ' ... ·.: .
· :. ·. · ·. . .The containment. t:tJ:ne ·of r.elativistic positrons :in a laboratory ·mirror ...
'.' · _:,_·:·
·.·.; ,.,
•
. ... machine has been measured by I.a.uer, Gi~son, and Jordan. (i'o) ]!xponemtial.decay .
. :~ .
' '

-·:·:
. ·'

' of the positron densii?Y is observed~ "With t:tJ:ne constants as l.ong as IIia.Ili .seconds •. :·:·::'·. ·
.. .·· . · .·. ·. .-·
.·. The: obsezyed loss '·rate is quantitat~vely 'attributable to scattering
. ' . . " ' . ~. ' :. . ' . .
rrlml .....the : :,~
. ' ' · •. ,.; . .
.
... ..
•
':.::· . '.". : ........... '
,1,
,.
· .. ·.·
 "
1"
.~

r ..
I
l·
I· background neutral gas, and therefore is not a non-adiabatic effect, even th~h
-~.1 .... !
10 .
. '~ (

a :particle has made of the order of 10 mirror. reflections in the decay_.

time. The accuracy of the self-cancelling effect in the stable orbits A and

B of Fig. ZQ' therefore is quite phenonemaL It would be. impossible to test · · ..

for such long containment by n·U::r@rical methods because of the.-:camputer. time

required and the numerical errors that would accumulate.

'.···
If in the experiment .the gyration r~i~s . is increased so that the ratio
. ::'\. .
·of gyration radius to L is -;> 0. 03. J
evidence .. of .the unstable
.
region and ·.

enhanGed loss eerie appears. ,· ... ·

•'. l
.. !
·.Similar containment experiments with the ·bu,mpy torus or with a field.

resembling. the earth 1 s would be worthwhile •.

. .
" . .)

'·:·. c. Higher O;rC.er Terms. .. ·...

i: ·'
I
I
t. The first order terms .1.
€3. 1
and e:<V: 1 in the adiabatic series for · j
I

and ~ have not been calculate~. The first correction to the longitudinal ·· ...
; . . . . . . '

invariant would be particularly valuable in studying the motion of the . ·..

· · high energy protons in the· earth 1 s field. .. ·

· ..., h

: •. ·.
··,. __

''
·,. . . :, ....
. ....
·.··.'·.
'
. ,_ ...
...

'.
. ~. :

.. -~ : ·· .. ,: .
·..:,1.·:.
:· .....· '
'• . : ''•'
.. . .. ··\~ :
~
' :·: .
. :. ' .. :
•.'
. . .. ' .

.
.•
. ·,{'.
. .
'- .. .. ·...
Oi·
I
I
:'':: ..· ... ' ·.··· .. ··
f.'.· . . ~
~ ." . -. '
.
I

l ·.' ' .. ,; ..
~~ . ' -~ .' . .'.
f.'.::: ' ... ' . '
·····"'.:·'··
.
:•"
...'::,
.~ ' I :
--~-.
.'

·' ..
,' 'I•

. \.,_
1:. ~
·.,• .·. ~- :
..
; 1 • •

·:··: . .·-.: .. ·
 II
t •..
''97'
I
I , .· ..
._..
L
r·: .VII. SUMMARY
.. ,l. ~ .
',. -~
.,
I ·. .
-~.:
.. •. ·.·
I ;
·.. ..
.•
·..... ,_ ..
·~
I . , ·.
!.;-o-- __ __;'------'---.;......,;.- - · - · --'----'-'----- -. ''
,·
. - ----·.··.·.:: -~ >:.
. .... ·;'
......·.
-.~-~-.' ~;:··.;,

If.· 'l'he eq'Uations of motion of a. charged. particle in an electromagnet:t:a '. '·.,

'•.·
• ~l
· ...
' '

:field have been. used to obtain the so-called "guiding center" motion. Ina ; . ,_. _- .
. ' ~ ~ . .
slowly va.rying field the particle motion is approxilna.tely a rapid rotation ~bout :.:· _;·<··, ·
·a circle whose-: center· (the guiding center) hB.s .velocity comp~ts alon~·and at.':''.:.:,(:~·
-· ..

f light angles to the magnetic field. . The eq_uations governing the ~iding center. •>:, J.': . .......
;'

l. • 4 •,: .: : 1 -.:.

r.
;'.·
· motion have been obtained deductively for the general time-dependent field. ··.: ·.· .:,: : .,
' . ' . • • • • ; .---· J
-~:' ..

f 0'
I . ,· ·' , ... ·;·
~- · .. ·.-_· . Several. unfamiliar ·drifts appeared because the time de~ndence and .the ~~-~~-~J · ·· ;· · ··~ ..
l:. '
. .': .·,·... ··.
~-/:: ~ .. perpendicular electric f~eld were assumed 1 for the sake .. of gener~lity,_to.be ·.. :·~.·~>/.'
<·. ··,,(;:, ..
..·.: .· ·:
·. . . '·..

1> .·;_;_: zero order ·in theradius of gyration rather than first order. These· less 'r .. ".
'-·.·.
1; • · · ' ·.·familiar drifts have been :illustrated via prae;-..~.;al examples ·' ·· where .they ·•·
~~.·.: .·. · ·.· .occur. ....
. . ~. :' .

'I •(·, • :\
·"·. . -' ..
: ' ~ ..
>. ·:.:
The modifications o'f the guiding center eq_uations nee;C.. ..":. for particles ... :~: .
-:·
·: '. :. '. ~ ,· :..
~.; ;·,,:
\

:'.,' of relativistic energy have been given. ". ,"·-'. ·.

r . ...
:;. '

F/ :-· '·~. defined and .th~ invariance proven for the

;·.
: The. three adiabatic:: invariants of the :p2..rticle :notion have ',
··- .·
..... ; i
·-.-;·
··,_., ... _ i: . .-:•

r<?
I' ..,
:)f' time deper":::.ent fields. · :>, :. ·:
F:..:_·: .:·.·.Several· applications have beenJ.ma.d.e of the guiding center ~quations-of motion .. ·.·
t··- ' •.. - <·· , ..·_ .
..·.,:... .
(. ~
,..' · ' and the adiabatic invariants • . ''·
',•,'
·•, ,. ··. ,· ::;'.> ''
'~,<~. •, . :· ..
devi;tion·s.: • _: trani· the~. adiabat~c 'predictions 'have'i•been disc~s~~··j:> ·
,_· .. i. . . •. . r . . . ,· ,': ~~:.\.~ -:~_-.. ,. /~>: ;{\:-H' . '.
-~ -~;: t •• :,'• • --~ ••• ···.': '," '. -~:-~5~:j·J·.. ·,. ·· .. __ , . '
~! . .. <
·.... ., ·.
-. -~-. -:- ·.' .. ·. ·; ··: ... · ' .. _:
.,~
..
\_.

,·oo; \~:·.:: • . • . ~

-·.I ..: ., -·-.1:- ·:-".. -

.
4. . . . .
..-:. .
. .- ~ .. -~ :' .
,\ '· (" ...- .. ~ •'
,_,.,- ·::·::_ .·
·,· .. . __

·. ·,_··.. ?:
· .. ·,_,.:.
'•.
•'.
;,.,<\ ;•""0• •'' I .

.. '
.
.. '
. ~ ... .".·,.
' ;
..
'
\ .~ ;. '
;·.·
." ..
~ '
. ~··.·, .
_:;

·::·. !.:·:.
,•"' . ··~ : • • < :

'
'.:·,
,
... .f / .:"".~: '. .,.··
..... :.·.' . ' . ' '.": .
·· ..
. ~ :·
> ••• .'

... ., . .. -~ ' ·,. ' '{, . •. ~·

;.
··:" ; ...- •• "=·'
... ~ .. '
··' \;
·.··.·
. ···. ·- ... '\,_'· ...
, .. ·... : ·; ~ ·. .'•'
'''
·. ,·
 ;' ' ->'8-

REFERENCES AliD FOOTNOTES

.. , ,..
1. The radiation reaction force will be omitted. In m.any practical cases.: '·1 .'·, ·
.· ..·
• its effect' on·the particle motion is neg],.igible. '·:

. ~.·
2 •. H. Al:f.'v~n,· Cosmi~o.l Electrodynamics, (oxior~, Clarendon Pres~,::i95b) •.

:·.:-:: __ ·:-
: " · 3 .... J. Linhart, Plasma Physics, (Interscience, New York, 1960).
..... .=... :-

/:,
' 4. ·' G. Hellwig, Z. Naturforschung lOA, 50~ (1955). : ····:··
. . ·..
~

:,.,'

.... ,. ... 5· T. Northrop,. :"The· Guiding Center Approx::iJna. tion to Charged Particle Motion", : .

Uni.v •. of Ce.+if~;. Ie:w:rence Rad .. Lab.

Report, UCRL-5708-T, March 151 1960 1 ..
,.
{M,eJ /lhn.,./5 -cf Ph-!~tCS },) ')q {! 96 f)
· · . · · :·:: ' :; •"

6• . . M. Kruskal, "The Gyration of a Cha.rged Particle",. Prince~.on pniversity1 ; ):,~ .·/·· ·.: · ...
· ...
. . .. ,.
.· ·.
Project Matterhorn Report PM-S.:.33 (NY0-7903), March 1958. '·.,.'

,, 7• · . ·For a ~iven
piece of_ eXperimental e'luipment it is .never in ·principle · .' "",
r• >·. .• 1. ·,

0 . -t' . .
wrong to :make the more general assumptior. tha.t dt . and '1jj are of
.·- .. -.
' : zero order f;Uld t_o :~;etain all the tei:ms • Some of them may turn/out to .:. ·· '. : :---:·
r:·· . i . . . . . .. ·. , ·• ·: :· ./,,, ,' .~. ,. \

:_ .. .. ·.
~. ~ t ,' '• •• • • ' :·_.:.,
:. ' ··- ..: .·>.. ' . ·.
.·, . {. '•'·
8. · C. Longmire, D. Nagle, F. Ribe,t PhJs. Rev.. U4 1 +H?') (1959) .
.' : . .'

r.· .. 9· . ,.. H. Lamb, Hyd.rodynamics, (Dqver Publications, /Je-w javlf J9LWJ 6th Ed.,

.........
.' · ..
10 •.. It is simple to demonstrate ·~i. ·:. :Mr is the adiabatic inva.:.. :L8.I1~~ 1
for the
\ ·,.
·. r>: ... -;·_. .
; . . special ':!ase. of a\..ut.Liform1 az_... J.thally· s~etric, time-dependen·~ field .·:
·.:_.;. .;~ '>.:> ;,

[.::.
···. 'with the particle gyrating a·o'-..::.t c~e axis It is e.:· '0
. .• . . 2
true (see Ref~ 11:) that · PJ.. /2:m B is the adiabatic invariant ir~ a
r.

0 -;':?··

:-···.··~.:~general geometry '~o long as· E..L is. a(€) 1 :.and is. the flux of B enclosed·:··~ ··~:· ...:' ·
'): .

L:·_.
t: ' •.• ' t ~ ..,
[' by the cL ;le of gyration •. .: .
~ . .
; . . 11·.:_::. ·P. Vander". ::rt, Annais .of Physics· !Q, 4ol (1960).
1. ..·..
···'.f
· 12'.'::... ·p. G. Bergmann~ Introduction td the Theory of Relativity,· (Prentice-Hall,
.: ~· . {

.7 •. .49.·.:- Bergmi:mn us~s a

1
·•. : ! ·.. :: .. : :·.New Yc::k1 :.942 ), equation different. metric· tensor:. .. ··.
. .' '. . ' . . .·.. .
·,

'.
;'.
' .., .. :~
. . • >

··. <... :·~han in tb.is Ia~l. h~~ver ··: •· ~.:'·.

·'
·.
,•
. !. •

~
I ' '
. ;. ' .,~ "

·.·:
'
. • '

'
'.,".·
. :. ..~. ·..
' . r"
..'. .. · .. ·.· ,·'
' .: ........ '. · ' . • ' .' ~ .,
....
. ; ~
. . ~·· . . . .' :;. ,, . . .. .:, . .·.···,
" .~:. ., .
<'·,
. .:
...
~

.. .. · .
 {., • •. ~~ •' '-'"i' ~... ~ r, "'
:: •· :.:-: _.;;· . -~.E: ~
1 •• __ , ; '·,< .:;_· -: ., •.. ,·
-99-

1-
i.
f: 13. J. Berkowitz and C. S. Gardner, "On the Asymptotic Series Expansion of
1': . . .

L_.-~ •.;.· ' .

j-:~
· ·· the Motion of the' Charged Particle in Slovrly Varying Fields", New York

r . ·. University Report NY0-79751 December, 1957. . ...>: ::::.

1·:£-". · 14. :~E. T<"Whittaker and G. N. Wat.son, A Course of Modern.· Analysis_,- (Cambridge1 ·. · .. , >:-- >
:1945)'; 4thEd~ Ch~ VIII. ·:·:·::~· .
1
.i
}:, ·.. ..:·· .··· . ·.- ·
It:: :·: . . ;·. ' ' .. . -~

i.'. . . '15. The converse is also true. If there is an A( E: ) . such that . ,, ·. ... l
!' 1
n+l ;• '',1

€ 'for 0 < e; _< €

1
, then 1
. i
'
.\

. i

. ·:·'lim = o.
: €-()

~ E:(~ (R2 e2 i.~· +

-!.

'16. ·.·:·.·For example · means R0 + e 19 + c. c)) + E:

2 c.· c.)., ..
··' . -~ ~
1::·: . : .· ·.· _17. _R, as defined at the beginning of Section II, and R
0 do not agree to

r•·· . ·>• ..,;_·. .·.. all orders in In ·fact . ~ - R

R -7 """ 2)
= D\ ~ However1 .this is of no
!::: ' •
E: •.
. 0 . ./' ,:
j .... • ••. ••

/.>'. ,': _
_..:. ;;··. . ·:. significance here, since we are considering only effects which are first:.>·: . ·.!:..
.. . ·r \
I·<·':·.:_.:.,. •order in gyzoation radius. ' r.'
Gardner,~ Phys. Rev. 115, 791 (195~1) .
to I '
f-.:_ .} · . . .18 • .. . C•. s. .
.• ' r:
I,

i ·:: ,, ·. ''' _:_X·:

1· . · : :. ·. 19. M. · Kruskal, in The Theory of Neutral Io:·ized ~. (John Wiley, · .: ~ .'
I' . .I; I;
}

1
1,·-~. ~: : ·~ :: .' . New York, 1960) ..
·f.'
• !
·.:·. j
:- ;, :: ... . . ~ . '··...
;

\' ·..• .. .20 • T •. G. Northrop and E. Tellt::::c·, Pl::;;r_,. Rev. 117, 215 (1960~.
•:·.:· j _.- •• -·
1
21. It_ is best to define J by equation (96), rc.ther than by (:;j ;1 . and
,,
I.

!:-: : ' to define K ~s :particle kinetic energy -,- e:.\v + q:>).

/'(.
l·· ..
22. B. B. Kadomtsev, Plasma Physics and the Problem of. Controlled Thermonuclear
t': ...
(·'''
Reactions, Vol: III (Alred. Nauk USSR, 1958), has given . similar
l

l-1<·'
f~·-:·'·. . · ';,'.; .;:oof for 'the case of static· fields:. and E Zero. '.·l:
1:'·:.: ..
! 'v : 23·. --:'If t~i. orders of are followed throUgh,. it is founJ. that, tk-
€ (§;i) =o:: ::_ ·· :··_>)
,1." ••

. .. , 2 1·1 dJ: . . _.. . . ,, • .L

' . ·. ' · : ~€ ) and . ax = 0 ~ crt€)..
m. ·.' / '. :;')'
24. < W. Newc0m.'8~ . ''Motion of Magnetic _Lines of Force "1 PBii.ugeto~: &i Vffl"'£i'H'-Y ·• ..
;: -\ : ' '? . ' : . . ·~ .

~: ·: ~-&l'Y m.selm;;l.ea.l R~port No.• l, .Apj!14J · 1955~ A Nvt.t;/j of· p~;.,,",·o· .1 ,:J'1'> ( hs"J') ·. ·· · ::·
. ,.
: :· ', ., ; ·: '' ', .' ',

,
....) '"·:.~ . .'
, ,
: . ' ·.. .

..• r.
,· '.
,. ·, ,. ...... ,
' • •
. ,., '•
•• '. • • •
"·'t
,. ,; •II,. '.
".~ ·. . . . ' ' r' . ~ ; ·.'.·
:~
'• •"
.,
,'I
,; . : ·~ .. , . ,·:.
. ... ~! ~
'· '.\ ·.·. ··.: . . ·. ;..-· ..·:. . ··.· . ( :·
.. .,.
'.·:.. :~
··,·
·'. ...; ,·,.
'•
• ... ·. ·· ..<
; I I ;, "..
 -100-

25~ Reference 12, e~uation 7.53

26. R. Becker, Theorie der Elektrizit~t, Vol II (Edwards Brothers, Ann Arbor), . ''
' 'I
'
' 1
paragraph 21. : ','1
i·
:

... ,.' '"

f. 27. ~.
' .
~.-a:.. Sp;itzer, Physics of Fully Ionized Gases · (Interscience 1 New York, 1956) ..... 1•
;' ·.· ~..:.
..
...'. P. 94 . I !,!'.

·,
. . \
i .: .· 28. · These variables have been used by Watson · (Pbys. Rev. 102· 12 (1956)) to
.. \ ~
::.:·:. ,.., prove the perpendicular component of e~uation (173) for the special cas~ · · · · ..~
.--:.. '

where the magnetic ;t>:leld is static, torsi'onless, and curl free.

•' .',I

29.' · ' H.·. Margenau and M. Lewis (Rev. Mod. Pbys. 21_, 569 (1959) ):.point out that the.. '/
~-· .' ·..
-~ . '
opposite of ·"ad:iabatic" is "diabatic~' .. However, "non-adiabatic" seems to.·> .···::::
'.

::.:.~- ..:._· '

F I ·J:··.·.·.

be firmly ~mtrEmched in the literature and therefore will be used. here. · ·

i'. -,;_..,· .. · 30. ··.. G. Gibson, W. Jord.8.n, E. J. Lauer, ~hy~. Rev. Letters-~ .217 (1960). · · · . .
;

~,:_. _:_;: . :,·

~J
... See also A. I. Morozov and. L. s. Sc..i.ov·c:y, ZJl, •. Teklm. iz. 2.Q., 26.1 (1960). ; .. ·I

,.,. . 31 G. Gibson, !f. A. S. Le.n;lb) ·E. J. Lauer, ?cys. Rev. 114,. 937 (1959}.' · ·:'··. · ·
I·''
1
;>~: :~ . . :'.. 32 ~ A. Simon, Physics of Fluids .k 495 (l958) ·,.; ..
"'·'· .·· ... . •,.

:<'·, .:._: .:. 33· (alw. Newcomb~ "Stability cfa Plasma in a lv'JB.gnetic Field., " University of ' ··. _ .. ,,·'

·.·/.
ir ·..·.. . .
r~_-.. ~·-·>-
~.\
-~ ~ . . .
California Ls:w~ence Radiation Lab. Repc:...··~ 'C::.~L 6197 ( ::.:.:i ?reparation).
. -.. .::1 -.··.

~··f/;:·. ,··.' _j.::. (b)M. N. Rosenbluth, Proceedings of the Secx~.:.... ·:..inited. ·· ...::;ns International

Conference on the Peaceful Uses of Atomic :~ __ rgy, \ _ - -J

. ;- '

(c)S. Chindrasekhe.~-, A Kaufman, K. Watson, f'roc. Roy. Sec . .A.2l·' ~~35, (1958)
[:~.- .'·' ~>
i"·
i..
l·.'·.
·.'~ ~

. . •.
34. W. A. Perkins ...d. R. F. Post, Bull. Am. Phys. Soc. 2_., 353 (::_ J) - ·..... . .. .. ·.·
' ~- ...
r ':.~. ~ . . ·.
[~ i:- . . 35· C. Y. Fan, P. ~·LJ;,.:.·, J. .f\.· Simpson; "Trapped and gosmic RaC..:....!.·don Measure- '.
'_;,,;

k.:·:.._:.
,.. •·,
· ments from Explorer VI", paper given at the First ~:::1ternational Space Science .
t'< . · ..
~:-i '• :. .
!;.,.t..
'
f.~~ ' ·.Symposium (Nice, France) Univ. of Chicago Enrico :. erm.i Institute for 1

(.'-,:,
Nuclear S~udies, Report EFINS - 59-73 (Dec. 1959)
: ;!,
I

;,: ;:~-~.: .

r-~~t. ' 36~ .. S •. F. Singer, Pbys. Rev. ·Letters :2, ~~ (l959)

·37···. J. A. W~ch1 Jr. andW. A. Whitaker, J~-Geopbys~ Research.§:, 909 (1959)
38.. A. Garren, et al. "Indirtdual. Particle Motion and the Effect of Scattering

in an Axially Symmetric Magnetic Field", Univ. of Calif'. Lawrence Rad.

. . .
Lab; Re:9c:t UCRL 8076 (March 1958),. 'or ~roceedings of the Second United

~ . '· ~

...
... ; ··. . '·
.,

~·-·-·r' ~-~'i ~ ~1 ~: ~~~L~~:::.~~::~i':·_:_~~·::;·~~~ ::_~, ~~-~.:.~.~~----·~---~*·

.·.. ··.· •,
. ·. ·: \ ~ ~ :. .. .... ." ', ··* · -·. ~ ._... ~~-:.:.._:"_.....Ji:;·_:~-~-· ~-~-~-~!. ''";_'-w.,.-7', .. ~-~·~-. -: ••
 ·.'• ..
-J.OJ.-

39· B. V. Chirikov, Atomnaya Errergiya §., 630 (J.959)

1-
, .,, 4o. G. Gibson, W. c. ~orda.n, E. J. Lauer, Phys. Rev. Letters ,2, J.4J. (1~6o)

'

.... . 'i
'j
i

:···' .. --.'
·•J- .. ~ ..
,lj

:. ' • • :, : ~ ,1,
I:.' ~~ '•• j •

. ,·,
;.·_::·

.:·
' ---~
!'· · · .. .1.~.~
!.. },- ..:. •' ,\ ·:

. ~- .
0

·:

: .1
'.·
! ·' ·,. "
I

i<~ ~
.. ,. .- ·.·.,··
I
';_;.:
.>.:
.·'-:'
.. :
.. ',••:

.r~·.:. ·:
--·-· ........ .~ ....

•,
·'. . ·'.•-

\.•
\
'·

).,