Aston 1922: Difference between revisions

From Mass Spec Terms
← Older editNewer edit →
Line 91: Line 91:


==CHAPTER  IX -  ISOTOPES  AND  ATOMIC  NUMBERS==
==CHAPTER  IX -  ISOTOPES  AND  ATOMIC  NUMBERS==
 
[[Aston 1922/Chapter 9]]
100.  The  relation  between  chemical  atomic  weight
and  atomic  number.  Inasmuch  as  it  is  now  recognised
to  be  in  general  merely  a  statistical  mean  value  the  importance=
 
of  the  chemical  atomic  weight  has  been  greatly  reduced  by
the  discovery  of  isotopes.  Its  position  as  the  natural  numerical=
 
constant  associated  with  an  element  has  been  taken  by  the
atomic  number,  though  from  the  point  of  view  of  chemical
analysis  the  chemical  atomic  weight  is  just  as  important  as
it  ever  was.
 
The  possibility  of  anomalies  in  the  order  of  the  elements
in  the  periodic  table  when  their  chemical  atomic  weights  are
considered,  is  now  obvious  enough.  The  true  weights  of  the
atoms  as  directly  determined,  are  so  intermingled  in  the  order
of  the  natural  numbers  and  the  proportions  present  in  complex
elements  so  varied  that  such  anomalies  are  bound  to  occur,
indeed  it  is  rather  surprising  there  are  not  more.
 
The  following  table  (Fig,  17)  shows  the  masses  of  the  isotopes
of  three  groups  of  elements  now  completely  investigated.  The
approximate  proportions  present  are  indicated  by  the  heights
of  the  columns  ;  plain  for  the  alliali  metals,  black  for  the  in=
ert
gases,  and  hatched  for  the  halogens.  The  anomalous  order
of  argon  and  potassium  is  at  once  seen  to  be  due  to  the  fact=
 
that  whereas  the  heavier  constituent  of  argon  is  present  in
much  the  greater  proportion,  in  potassium  the  reverse  is  the
case.  Had  the  proportions  of  heavier  and  Ughter  isotopes
been  similar  in  each  case  the  atomic  weight  of  potassium  would
have  been  greater  instead  of  less  than  that  of  argon.
 
108
 
 
ISOTOPES  AND  ATOMIC  NUMBERS
 
 
109
 
 
:s
 
 
19
 
 
20
 
 
21
 
 
22        23
 
 
Fluorine  (9)  Neon  (10)
Sodium  (11)
 
 
35        36        37        38        39        40        41
 
(Chlorine  17)  Argon  (18)
Potassium  (19)
 
 
=C2=A5
 
 
1
 
 
I-D
 
 
78
 
 
I
 
127
 
 
79        80        81        82        83        84        85        86  =
      87
Bromine  (35)  Krypton  (36)  Rubidium  (37).
 
 
I    n
 
 
W
 
 
136
 
 
128    129      130    131    132      133      134      135
Iodine  (53)  Xenon  (54)  Caesium  (55)
 
Fig.  17.  Isotopes  of  the  Halogens,  the  inert  gases  and  t=
he  alkali  metals.
 
101.  Statistical  relations  exhibited  by  elements  and
their  isotopes.  -Although  our  knowledge  of  true  atomic
weights  is  far  from  complete,  for  out  of  eighty-seven  existing
elements  only  twenty-seven  have  been  analysed,  of  which
thirteen  are  simple,  interesting  relations  have  already  become
clear  which  are  stated  in  the  form  of  rules  as  follows  : =E2=80=
=94
 
In  the  nucleus  of  an  atom  there  is  never  less  than  one  electron=
  to
every  two  protons.  There  is  no  known  exception  to  this  law.
It  is  the  expression  of  the  fact  that  if  an  element  has  an  ato=
mic
number  N  the  atomic  weight  of  its  lightest  isotope  cannot  be
less  than  2N.  Worded  as  above,  the  exception  in  the  case  of
hydrogen  is  avoided.  True  atomic  weights  corresponding
exactly  to  2N  are  known  in  the  majority  of  the  Ughter  elements=
 
]jp  to  A^^    Among  the  heavier  elements  the  difference  between
 
 
110  ISOTOPES
 
the  weight  of  the  lightest  isotope  and  the  value  2N  tends  to
increase  with  the  atomic  weight ;  in  the  cases  of  mercury  it
amounts  to  37  units.  The  corresponding  divergence  of  the
mean  atomic  weights  from  the  value  2N  has  of  course  been
noticed  from  the  beginning  of  the  idea  of  atomic  number.
 
The  number  of  isotopes  of  an  element  and  their  range  of
atomic  weight  appear  to  have  definite  limits.  Since  the
atomic  number  only  depends  on  the  net  positive  charge  in
the  nucleus  there  is  no  arithmetical  reason  why  an  element
should  not  have  any  number  of  possible  isotopes.  An
examination  of  the  tables  of  results  given  on  p.  89  and
at  the  end  of  the  book  show  that  so  far  the  largest  number
determined  with  certainty  is  6  in  the  case  of  krypton.  It
is  possible  that  xenon  has  even  more,  but  the  majority  of
complex  elements  have  only  two  each.  The  maximum  differ-
ence between  the  lightest  and  heaviest  isotope  of  the  same
element  so  far  determined  is  8  units  in  the  cases  of  krypton
and  xenon.  The  greatest  proportional  difference,  calculated
on  the  lighter  weight,  is  recorded  in  the  case  of  lithium,  where=
 
it  amounts  to  one-sixth.  It  is  about  one-tenth  in  the  case  of
boron,  neon,  argon  and  krypton.
 
The  number  of  electrons  in  the  nucleus  tends  to  be  even.  This
rule  expresses  the  fact  that  in  the  majority  of  cases  even
atomic  number  is  associated  with  even  atomic  weight
and  odd  with  odd.  If  we  consider  the  three  groups  of
elements,  the  halogens,  the  inert  gases  and  the  alkah  metals,
this  tendency  is  very  strongly  marked.  Of  the  halogens  od=
d
atomic  numbers  all  6  (  +  1  ?)  atomic  weights  are  odd.  =
Of
the  inert  gases  even  atomic  numbers  13  (+  2  ?)  =
are  even  and
3  odd.  Of  the  alkali  metals  odd  atomic  numbers  7=
  are
odd  and  1  even.  In  the  few  known  cases  of  elements  of  the
other  groups  the  preponderance,  though  not  so  large,  is  still
very  marked  and  nitrogen  is  the  only  element  yet  discovered
to  consist  entirely  of  atoms  whose  nuclei  contain  an  odd  number=
 
of  electrons.
 
A  further  interesting  result  is  the  absence  of  isobares.  So
far  none  have  been  definitely  identified,  but  it  is  quite  obvious=
 
that  in  the  cases  of  elements  such  as  calcium  and  selenium  they=
 
must  exist,  for  the  supply  of  integers  in  the  region  of  their
 
 
ISOTOPES  AND  ATOMIC  NUMBERS  HI
 
atomic  weights  have  been  exhausted  by  the  needs  of  other
elements.
 
A  table  of  the  first  40  natural  numbers  and  the  true  atomic
weights  corresponding  to  them  is  given  in  Fig.  18.    The  gaps
 
 
H
 
He
 
Li
 
Li
 
Be
 
B
 
B
 
c
 
N
 
O
 
F
 
Ne
 
1
 
2      3      4^      5      6      7      8      9      10    11    =
  12
 
13    14
 
15    16    17
 
18
 
19    20
 
Ne
 
Na
 
Mg
 
m
 
Mg
 
Al?
 
Si
 
Si  Si?
 
P
 
s
 
CI
 
A
 
CI
 
K
 
A
 
21
 
22
 
23
 
24
 
25
 
26
 
27
 
28
 
29
 
30
 
31
 
32
 
33
 
34
 
35
 
36
 
37
 
38
 
39
 
40
 
Fig.  18.  The  first  40  natural  numbers,  showing  those  occ=
upied
by  atomic  weights  of  known  elements.
 
are  particularly  interesting  and  seem  to  show  no  semblance
of  regularity.  It  is  very  clear  that  many  more  experimental
results  will  have  to  be  obtained  before  any  satisfactory  theory
for  the  occurrence  of  these,  or  of  the  other  laws,  is  to  be  fo=
rmu-
lated.
 
102.  The  preponderance  of  elements  of  even  atomic
number.  In  discussing  the  nuclear  structure  of  elements
the  question  of  their  relative  abundance  in  nature  is  one  of
great  interest.  This  may  be  estimated  by  direct  chemical
analysis  of  the  Earth's  crust,  and  such  extra-terrestrial  sources=
 
as  are  available  in  the  form  of  meteorites.  The  spectroscope
will  teU  us  what  elements  are  present  in  the  stars,  but  unfortu-=
 
nately it  does  not  give  much  direct  information  as  to  their
relative  quantities.
 
On  this  question  we  can  classify  to  use  biological  terms=
 
either  by  individuals  or  by  species.  We  may  examine  the=
 
percentage  composition,  which  wiU  give  a  measure  of  the  total
number  of  individual  atoms  of  each  element  present,  or  we
may  inquire  into  the  number  of  different  nuclear  species  which
occur  and  classify  them  without  respect  to  their  individual
abundance.
 
A  very  valuable  discussion  from  the  first  point  of  view  has
been  published  by  Harkins,^  who  considers  the  percentage
composition  of  meteorites  and  of  parts  of  the  Earth's  crust.
He  demonstrates  in  a  most  convincing  manner  that  there  are
 
^Harkins,  Jour.  Amer.  Chem.  Soc,  39,  856,    1917.
 
 
112  ISOTOPES
 
immensely  more  atoms  of  elements  of  even  atomic  number.
This  interesting  preponderance  can,  with  a  reasonable  amount
of  probability,  now  be  extended  to  even  atomic  weight,  by
the  statistics  given  in  the  preceding  paragraphs,  but  it  will
not  be  certain  until  the  constitution  of  certain  abundant
elements  such  as  iron  has  been  actually  determined.
 
The  second  point  of  view  can  be  examined  by  means  of  the
atomic  weights  of  the  radioactive  isotopes  and  also  by  the  true=
 
atomic  weights  given  by  the  mass-spectra.  In  both  cases
nuclear  systems  of  even  atomic  number  are  found  to  predomi-
nate. The  mass-spectra  of  13  elements  of  even,  and  14  ele-
ments of  odd  atomic  number  indicate  32  isotopes  of  even
atomic  number  and  20  of  odd.  The  average  element  of  even
atomic  number  has  therefore  2-5  isotopes  to  1-4  for  each  element=
 
of  odd  atomic  number.
 
The  table  on  p.  15  shows  that  among  the  radioactive  isotopes
the  preponderance  is  greater  32  as  against  10  but=
  it  is
possible  that  the  former  figure  may  include  some  atomic
systems  absolutely  identical  though  of  different  origin.
 
103.  The  constancy  of  chemical  atomic  weights.
 
One  of  the  first  difficulties  in  the  way  of  accepting  the  idea  =
of
the  complex  constitution  of  an  element  such  as  chlorine  was
the  constancy  of  its  atomic  weight.  This  had  been  determined
by  many  different  observers  using  different  methods  and  the
results  were  always  the  same  within  a  very  small  experimental
error.  This  difficulty  may  be  met,  in  the  first  place,  by  noting=
 
that  the  vast  majority,  if  not  all,  of  the  really  accurate  value=
s
were  obtained  from  chlorine  which  must  have  been  originally
derived  from  the  sea.  The  sea  has  been  mixed  for  so  long
that  it  would  be  absurd  to  expect  to  find  chlorines  of  different=
 
chemical  atomic  weights  in  it.  Had  ordinary  galena  been
the  only  source  of  lead  used  in  the  atomic  weight  determina-
tions given  on  page  16  no  difference  would  have  been  found.
It  was  only  by  examining  the  lead  from  extraordinary  radio-
active sources  that  the  results  were  obtained  which  gave  such
definite  and  valuable  support  to  the  theory  of  isotopes.
 
The  atomic  weight  of  chlorine  from  sources  other  than  the
sea  is  now  receiving  the  attention  of  chemists,  though  it  is
 
 
ISOTOPES  AND  ATOMIC  NUMBERS  113
 
naturally  very  difficult  to  be  at  all  sure  that  any  known  source=
 
of  chlorine  is  not  of  marine  origin.  Mile.  Irene  Curie  ^  has
examined  the  atomic  weight  of  chlorine  from  three  minerals
whose  marine  origin  seems  unlikely.  The  values  obtained
from  a  sample  of  sodalite  (sodium  aluminium  chlorosilicate)
from  Canada,  and  from  a  sample  of  calcium  chlorophosphate
from  Norway  agree  with  the  value  for  chlorine  from  sea-water.
The  value  35-60,  for  chlorine  from  a  sample  of  sodium  chloride
from  a  desert  region  in  Central  Africa  was  slightly  high.
 
The  comparison  of  the  atomic  weights  of  terrestrial  and
meteoric  nickel  made  by  Baxter  and  Parsons  ^  is  interesting
in  this  connection.  As  a  mean  of  nine  determinations  with
the  terrestrial  material  the  figure  58-70  was  found,  whilst
three  experiments  with  meteoric  nickel  gave  58-68.  The
standard  value  found  by  Richards  and  Cushman  was  58-68
(Ag  =3D  107-88).  The  difference  found  between  terrestrial  and
meteoric  nickel  is  considered  to  be  within  the  limits  of  experi-=
 
mental error,  but  further  comparisons  are  to  be  made.
 
The  writer  regards  these  negative  results  as  having  a  cause
probably  much  more  fundamental  than  the  mere  mechanical
mixing  of  the  different  constituent  isotopes  during  the  history
of  the  body  containing  them,  namely  a  constancy  of  proportion
during  the  evolution  of  the  elements  themselves.  This  will
be  considered  later.  The  case  of  the  radioactive  leads  is
entirely  exceptional.  These  substances  have  been  produced
continuously  during  the  history  of  the  earth's  crust  and  are
being  so  produced  to-day.  Although  ordinary  lead  may  con-
sist of  isotopes  which  is  practically  certain  and  =
these  isotopes
may  be  identical  in  every  respect  with  those  produced  in  the
last  stage  of  radioactive  disintegration,  yet  there  is  no  reason=
 
whatever  to  assume  that  ordinary  lead  is  itself  the  accumulated
result  of  these  processes.  It  takes  its  place  among  the  other
ordinary  elements  and  would  doubtless  have  done  so  had
thorium  and  uranium  never  existed.
 
104.  The  agreement  between  the  chemical  atomic
weight  and  the  mean  atomic  weight  deduced  from  the
mass  spectrum.  The  mean  atomic  weight  of  the  isotopes
 
II.  Curie,  Compt.  Retid.  172,  1025,    1921.
 
*  Baxter  and  Parsons,  Jour.  Amer.  Chem.  Soc,  43,  607,    1921.
 
I
 
 
114
 
 
ISOTOPES
 
 
of  a  complex  element  can  be  calculated  if  the  relative  intensitie=
s
of  their  lines  in  the  mass-spectrum  is  known.  This  has  been
directly  measured  by  Dempster.  ^  The  charged  particles  of
isotopes  of  the  same  element  are  practically  certain  to  afEect
the  photographic  plate  to  the  same  extent  as  each  other,
hence  we  can  obtain  a  rough  estimate  of  their  relative  pro-
portion by  comparing  the  intensities  of  the  lines.  If  this  is
done  it  is  found  that  the  great  majority  of  the  elements  so
far  tested  give  mean  results  in  good  agreement  with  the
accepted  chemical  values.  The  following  table  gives  the  data
concerning  four  in  which  the  difference  is  noteworthy  : =
 
 
 
Element.
 
Atomic
Weight.
 
Mean  from
Mass-spectrum.
 
Difference.
 
Per  cent.
Difference.
 
Boron  ....
Krypton    .
Xenon
Caesium
 
10-90
82-92
130-2
132-81
 
10-75^0-07
83-5  =C2=B10-3
131-3  =C2=B10-3
133      =C2=B10-3
 
0-15
0-6
1-1
0-2
 
1-37
 
0-72
0-85
0-05
 
The  case  of  boron  is  the  most  difficult  to  account  for.  The
masses  of  its  isotopes  10  and  11  certainly  do  not  differ  from
integers  by  more  than  one  or  two  parts  in  a  thousand.  The
ratio  of  the  intensities  of  their  second  order  lines  5  and  5-5=
 
(and  there  were  no  other  substances  present  which  could
possibly  give  such  lines)  is  equally  certainly  not  as  high  as
9:1.  It  was  for  this  reason  that  a  third  isotope  12  was
suspected,  but  as  no  evidence  of  this  has  been  found  it  seems
most  probable  that  the  chemical  atomic  weight  is  still  slightly
too  high.
 
The  atomic  weights  of  krypton  and  xenon  are  not  of  course
chemical  in  the  ordinary  sense,  as  they  are  deduced  direct
from  density  determinations.  Any  trace  of  the  impurity  most
likely  to  be  present,  argon  in  the  first  case,  krypton  in  the
second,  would  tend  to  make  the  densities  too  low,  and  this
appears  the  most  hkely  explanation.
 
In  the  case  of  caesium  the  chemical  result  may  be  correct,
for  the  probable  error  in  the  determination  of  mass  is  at  least=
 
as  large  as  the  discrepancy.    On  the  other  hand  caesium
 
1  V.  p.  81.
 
 
ISOTOPES  AND  ATOMIC  NUMBERS  115
 
appears  to  be  a  simple  element,  in  which  case  its  chemical
atomic  weight  must  represent  the  true  weight  of  its  atoms.
Any  error  in  this  figure  would  probably  be  of  the  sign  suggested=
,
for  it  is  the  heaviest  member  of  its  chemical  group.  If,  how-
ever, as  is  possible,  the  true  mass  of  its  atom  differs  from  an=
 
integer  by  as  much  as  0-2  it  is  a  fact  of  the  greatest  interes=
t.
 
105.  The    meaning    of    the    word    "  element."  The
 
exact  idea  conveyed  by  the  word  "  element  "  in  chemistry
and  physics  has  given  rise  to  endless  difficulties  in  the  past.=
 
In  this  connection  Crookes  in  1886  sums  up  the  matter  as
follows  :  "  Of  the  attempts  hitherto  made  to  define  or  =
explain
an  element,  none  satisfy  the  demands  of  the  human  intellect.
The  textbooks  tell  us  that  an  element  is  '  a  body  which  has
not  been  decomposed  '  ;  that  it  is  '  a  something  to  which  we=
 
can  add,  but  from  which  we  can  take  away  nothing,'  or  '  a
body  which  increases  in  weight  with  every  chemical  change.'
Such  definitions  are  doubly  unsatisfactory  :  they  are  provi-
sional, and  may  cease  to-morrow  to  be  applicable  to  any
given  case.  They  take  their  stand,  not  on  any  attribute  of
things  to  be  defined,  but  on  the  limitations  of  human  power  ;
they  are  confessions  of  intellectual  impotence."
 
There  was  good  reason  for  this  dissatisfaction.  The  dis-
covery ten  years  later  of  the  electron,  and  the  subsequent
electrical  theory  of  matter  robbed  the  word  of  any  pretence
to  its  original  meaning  ;  for  although  Ramsay  attempted  to
introduce  into  chemistry  electricity  itself  as  an  element,  it
soon  became  obvious  that  this  extension  was  unsuitable.  The
discovery  of  isotopes  brings  us  face  to  face  with  two  possible
alternatives.  The  first  is  to  call  each  isotope,  as  it  is  dis-=
 
covered, a  new  element.  The  second  is  to  fix  the  word  pre-
cisely, now  and  for  the  future,  as  meaning  a  substance  with
definite  chemical  and  spectroscopic  properties  which  may  or
may  not  be  a  mixture  of  isotopes  in  other  words  to  asso=
ciate
it  exclusively  with  the  conception  of  atomic  number.  On
this  view  there  would  be,  corresponding  to  Moseley's  numbers,
92  possible  elements,  of  which  87  are  known.
 
If  we  adopt  the  first  of  these  alternatives  a  new  word  will
be  necessary  to  express  such  substances  as  chlorine  or  mag-
 
 
116  ISOTOPES
 
nesium,  hitherto  called  elements,  and  also  the  word  element
would  mean  something  entirely  different  from  what  it  has
meant  in  all  the  chemical  and  physical  Uterature  of  the  past
century.  It  would  moreover  be  still  subject  to  alterations  in
the  future.
 
In  the  opinion  of  the  writer  the  second  alternative  the=
 
association  of  element  with  atomic  number  is  much  the
more  preferable.  The  difficulties  arising  from  it  are  practi-
cally confined  to  the  radioactive  substances  which  can  differ
from  one  another  even  when  their  atomic  numbers  and  atomic
weights  are  identical.  This  is  not  very  serious,  for  the  radio-=
 
active elements  are  in  a  class  by  themselves  and  the  special
nomenclature  already  applied  to  them  could  be  retained  or  re-
vised as  convenient  without  affecting  that  of  general  chemistry.
 
106.  Disintegration  theory  of  the  evolution  of  the
elements.  A  theory  has  been  put  forward  by  some  writers=
 
that  all  the  elements  occurring  in  nature  are  the  result  of
radioactive  disintegrations  of  the  ordinary  type,  but  continued
far  beyond  the  ordinary  limit  observed  at  present.  For
instance,  if  we  continue  the  a  ray  changes  of  the  thorium  series=
 
far  enough  we  shall  ultimately  reach  helium.  The  emission
of  an  a  particle  is  the  only  change  known  to  occur  which  alters=
 
the  atomic  weight  and  it  always  does  so  by  4  units  at  a  time.=
 
Hence  from  thorium  we  shall  get  a  series  of  elements  or  iso-
topes of  atomic  weights  from  232  to  4  of  the  general  type  4w.
Uranium  in  the  same  way  will  yield  a  similar  series  of  the  type=
 
4:71  +  2.  In  order  to  obtain  isotopes  of  odd  atomic  weight
it  is  necessary  to  postulate  parent  elements  of  the  type  4n  +  1=
 
and  4:%  -f  3.
 
Using  hypotheses  based  on  this  general  idea  Van  den  Broek,  ^
Harkins,^  Kohlweiler,^  Kirchoff  ^  and  others  have  built  up  the
most  elaborate  systems  of  isotopes,
 
1  Van  den  Broek,  Phys.  Zeit.,  17,  260,  579,      1  916  ;    23,  =
164,      1921.
 
"Harkins  and  Wilson,  Jour.  Am.  Chem.  Soc,  37,  1367,  1915  ;  Har-=
 
kins  and  Hall,  ibid.,  38,  169,  1916  ;  Harkins,  Phys.  Rev.,  15,  =
73,
1920;  Nature,  105,  230,  1920;  Jour.  Amer.  Chem.  Soc,  42,  1956,
1920  ;    PhU.  Mag.,  42,  305,    1921.
 
*  Kohlweiler,  Zeit.  fur  physikal.  Chem. ,  94,  51 3,  1 920  ;  Phys.=
  Zeit. ,  21,
311,  543;  22,  243,    1921.  *  Kirchoff,  idid.,  21,  711,    1920.=
 
 
 
ISOTOPES  AND  ATOMIC  NUMBERS  117
 
The  writer  regards  this  view  as  unhkely  and  misleading.
In  the  first  place  it  does  not  appear  to  succeed  in  its  objects=
.
As  an  explanation  of  how  the  elements  may  have  been  evolved
it  starts  with  at  least  fom:  elements  as  complicated  as  any
known  to  exist,  which  does  not  advance  the  inquiry  very
much.  On  the  other  hand  it  may  be  used  to  predict  the
atomic  weights  of  the  isotopes  composing  known  elements,
and  a  great  many  predictions  of  this  kind  have  been  made.
Here,  though  the  measure  of  its  success  has  varied  to  some
extent  with  the  particular  modification  of  the  theory  employed,
it  has  never  been  worthy  of  serious  consideration.  In  cases
where  two  or  three  isotopes  of  a  given  element  were  pre-
dicted they  proved  as  often  wrong  as  right,  and  when  the
number  of  isotopes  of  integral  atomic  weights  was  so  large
that  some  agreements  were  inevitable  the  argument  obviously
loses  all  its  force.
 
Another  objection  is  that  radioactive  transformations  do
not  continue,  as  far  as  we  can  see,  beyond  the  stage  (lead)
indicated  in  the  diagrams  on  p.  15.  The  lighter  elements  are
definitely  not  radioactive.  The  radioactivity  of  potassium
and  rubidium  is  exceedingly  small  and  its  nature  doubtful ;
in  any  case  it  is  best  ascribed  to  minute  vestiges  of  radioactiv=
e
isotopes,  not  to  feeble  radioactivity  of  the  main  constituents.
It  seems  therefore  more  reasonable,  for  the  present,  to  regard
the  property  of  radioactivity  as  absent  entirely  from  the
inactive  elements  than  to  suppose  it  present  but  too  weak  to
be  detected.  It  must  not  be  gathered  from  these  remarks
that  it  is  considered  impossible  to  imagine  physical  conditions
violent  enough  to  disrupt  the  nuclei  of  light  atoms,  but  rather=
 
that  the  mechanism  causing  such  disruption  need  not  be
similar  in  any  way  to  that  causing  normal  radioactivity.
 
107.  Grookes'  theory  of  the  evolution  of  the  elements.
 
  A  more  attractive  theory  than  the  one  given  abo=
ve  is  that
the  complex  atoms  of  matter  have  been  evolved  by  the  aggre-
gation of  simpler  atoms.    This  idea  has  received  a  good  deal  of=
 
attention  in  the  past.    Crookes^  remarks  on  it  as  follows  : =E2=
=80=94
"  Let  us  picture  the  very  beginnings  of  time,  before  geological=
 
 
^  Crookes,  Brit.  Assoc,  address,  1886.
 
 
118  ISOTOPES
 
ases,  before  the  earth  was  thrown  off  from  the  central  nucleus
of  molten  fluid,  before  even  the  sun  himself  had  consolidated
from  the  original  pivtyle.  Let  us  still  imagine  that  at  this
primal  stage  aU  was  in  an  ultra-gaseous  state,  at  a  temperature=
 
inconceivably  hotter  than  anything  now  existing  in  the  visible
universe  ;  so  high  indeed  that  the  chemical  atoms  could  not
yet  have  been  formed,  being  still  far  above  their  dissociation
point.  In  so  far  as  protyle  is  capable  of  radiating  or  reflectmg=
 
light,  this  vast  sea  of  incandescent  mist,  to  an  astronomer  in  a=
 
distant  star,  might  have  appeared  as  a  nebula,  showing  in
the  spectroscope  a  few  isolated  hues,  forecasts  of  hydrogen,
carbon  and  nitrogen  spectra.
 
"  But  in  due  course  of  time  some  process  akin  to  cooling,
probably  internal,  reduces  the  temperature  of  the  cosmic
protyle  to  a  pomt  at  which  the  first  step  in  granulation  takes=
 
place  ;  matter  as  we  know  it  comes  into  existence,  and  atoms
are  formed."
 
This  vivid  picture  may  be  brought  up  to  date  by  the  sub-
stitution of  free  protons  and  electrons  for  the  hypothetical
protyle.  We  can  imagme  regions  containing  matter  where
the  temperature  is  so  high  that  not  only  is  the  dissociation  of=
 
atoms  from  atoms  and  nuclei  from  planetary  electrons  com-
plete but  also  protons  and  electrons  are  in  a  state  of  agitation=
 
so  violent  that  even  the  most  stable  nuclei  cannot  be  formed.
We  should  have  here  matter  of  the  simplest  form  we  can
imagine,  or  rather  of  no  form  at  all,  simply  a  more  or  less
neutral  electric  gas.  Such  a  condition  is  by  no  means  impos-
sible in  our  miiverse  and  may  actually  occur  during  one  of  those=
 
excessively  violent  catastrophes  occurring  in  far  distant  space
and  observed  by  us  as  new  stars.
 
By  some  such  cooling  process  as  that  suggested  by  Crookes
we  easily  imagine  the  free  charges  combining  to  form  the
nuclei  of  elements.  Whether  those  of  heavier  elements  are
formed  direct  by  the  charges  getting  into  particular  geometrical
relations  with  each  other,  or  whether  hehum  nuclei  are  formed
first  and  then  subsequently  coalesce  depends  on  which  theory
of  nuclear  structure  is  adopted.  In  any  case  vast  quantities  of=
 
energy  will  have  to  be  radiated  off  and  this  radiation  may  be
of  such  extremely  high  frequency  that  it  is  capable  of  dis-
 
 
ISOTOPES  AND  ATOMIC  NUMBERS  119
 
rupting  nuclei  themselves,  so  that  there  might  be  at  this  stage=
 
rapid  and  continuous  transformations  from  heavier  to  lighter
nuclei  and  vice  versa.
 
For  the  present  we  are  interested  in  the  number  of  each
type  of  atom  which  survives.  It  is  obvious  that  if  the  con-
ditions of  cooling  are  practically  identical  throughout  the  whole
mass  there  is  no  reason  why  the  composition  of  the  matter
produced  should  vary.  If  3  atoms  of  CP^  are  formed  to  every
1  of  CP^  at  any  one  point  the  same  ratio  must  hold  at  every
point  so  that  a  complex  element  of  constant  atomic  weight
will  be  formed.  But  it  is  much  more  likely  that  different  parts=
 
of  this  primordial  mass  will  undergo  their  transformations
under  different  rates  of  cooling,  etc.,  so  it  is  worth  while
inquiring  if  variation  in  the  mean  atomic  weight  of  a  complex
element  is  to  be  expected.
 
The  quantity  of  one  particular  atomic  nucleus  formed  will
probably  depend  (a)  on  the  probability  of  a  certain  configura-
tion of  charges  happening  as  a  chance  event ;  (b)  the  stability  o=
f
the  particular  nucleus  formed  as  the  result  of  that  event.
Again  to  take  the  case  of  chlorine  each  isotope  may  be  regarded=
 
as  completely  stable  and  the  relative  quantities  formed  will
simply  depend  on  condition  (a).  Now  it  is  not  unreasonable
to  suppose  that  this  is  not  seriously  affected  by  different  rates=
 
of  cooUng,  and  in  this  case  the  isotopes  will  be  evolved  in
constant  proportion.  As  we  know  of  no  natural  process  by
which  the  proportion  of  isotopes  can  be  altered  appreciably
the  complex  elements  will  have  to-day  the  same  chemical
atomic  weight  as  when  they  were  first  formed.
 
The  above  argument  is  of  course  purely  a  speculative  one,
and  the  conclusion  drawn  from  it  would  fall  to  the  ground  at
once  if  noteworthy  differences  of  atomic  weight  in  a  single
complex  element  were  found  supposing  that  element  was  not=
 
the  product  of  a  radioactive  change  at  different  points  o=
n  the
earth's  surface.  It  may  be  worth  noting  that  condition  (a)
suggests  that,  in  general,  the  lighter  atoms  will  outnumber
the  heavier  ones.  In  aU  matter  available  in  nature  this
preponderance  is  actually  enormous.
 
If  the  matter  forming  the  earth  ever  went  through  a  prim-
ordial stage  such  as  that  suggested  above  it  certainly  did  so
 
 
120  ISOTOPES
 
more  than  10^  years  ago.  It  follows  that  of  the  radioactive
elements  then  formed  only  two,  thorium  and  uranium,  wiU
now  be  found  on  the  earth,  for  the  other  radioactive  elements
existing  to-day  are  of  such  short  period  that  they  must  have
been  formed  since.  Hence  we  may  divide  the  original  elements
very  simply  and  definitely  into  two  groups  :  (1)  All  the
inactive  elements,  whose  nuclei  are  sufficiently  simple  to  be
stable ;  (2)  Thorium  and  Uranium,  whose  nuclei  are  so
complex  that  they  are  only  partially  stable.
 
Other  less  stable  elements  vfiay  have  been  formed  then  but
there  can  be  no  proof  of  this  for  they  would,  in  any  case,  hav=
e
disappeared  long  ago,  and  it  is  clear  that  the  other  radioactive=
 
elements  now  found  can  all  be  regarded  as  formed  from  the
two  parent  elements  in  comparatively  recent  times.
 


==CHAPTER  X - THE  SPECTRA  OF  ISOTOPES==
==CHAPTER  X - THE  SPECTRA  OF  ISOTOPES==

Revision as of 18:05, 3 July 2025

ISOTOPES[1]

F. W. ASTON, M.A., D.Sc, A.I.C., F.R.S.

Fellow of Trinity College, Cambridge

LONDON

EDWARD. ARNOLD & CO.

1922

[All rights reserved]

Printed in Great Britain

PREFACE

I have undertaken the preparation of this book on isotopes in response to many requests made to me by teachers of physics and chemistry and others working in these subjects that I should publish the results obtained by means of the mass spectrograph in a form more convenient to the public than that in which they first appeared. This is one of the reasons why the space allotted to the inactive isotopes may appear, in the light of the general title of the book, somewhat disproportion- ately large. Another is that the subject of radioactive isotopes really requires a book to itself, and I am in the hope that the inadequacy of my account may stimulate the production of such a volume by hands more competent than mine to deal with this very special and remarkable field of modern science. The logical order of exposition of a scientific subject is to start with the simple and from that build up the more complex. Unfortunately the sequence of events in experimental research is the exact opposite of this so that a compromise must be effected, unless one is content to sacrifice historical treatment altogether. The latter seems very undesirable in a new subject. I have endeavoured in Chapters I, II and IV, and elsewhere when possible, to adhere strictly to the historical order of events even at the cost of some reiteration.

I wish to take this opportunity of expressing my indebted- ness to Mr. C. G. Darwin for his timely criticism and unfailing assistance throughout the work, and also to Mr. R. H. Fowler for help with the proofs. My thanks are also due to Professor Soddy for his diagram of the radioactive isotopes, to Mr. A. J. Dempster for kindly sending me the illustrations of his work, to the proprietors of the Philosophical Magazine and to the Council of the Chemical Society for permission to use the plates and figures of my original papers, and to Messrs. Macmillan & Co., for the diagram of the radioactive trans- formations.

F. W. Aston
Cambridge,
January, 1922.

CONTENTS

Aston 1922/Contents

CHAPTER I - INTRODUCTION

Aston 1922 Chapter 1

CHAPTER II - THE RADIOACTIVE ISOTOPES

Aston 1922/Chapter 2

CHAPTER III - POSITIVE RAYS

Aston 1922/Chapter 3

CHAPTER IV - NEON

Aston 1922/Chapter 4

CHAPTER V - THE MASS-SPECTROGRAPH

Aston 1922/Chapter 5

CHAPTER VI - ANALYSIS OF THE ELEMENTS

Aston 1922/Chapter 6

CHAPTER VII - ANALYSIS OF THE ELEMENTS (Continued)

Aston 1922/Chapter 7

CHAPTER VIII - THE ELECTRICAL THEORY OF MATTER

Aston 1922/Chapter 8

CHAPTER IX - ISOTOPES AND ATOMIC NUMBERS

Aston 1922/Chapter 9

CHAPTER X - THE SPECTRA OF ISOTOPES

108. The Spectra of isotopes. As has already been stated^ the first experimental work on the spectra of isotopes was that of Russell and Rossi in 1912 who failed to distinguish=

any difference between the spectrum of thorium and that of a mixture of thorium and ionium containing a considerable percentage of the latter. The same negative result was obtained by Exner and Haschek.^ During the fractional diffusion of neon^ no spectroscopic difference was detected between the heaviest and the lightest fraction, though as the separation was small this negative evidence was not very strong. In 1914 Soddy and Hyman showed that the spectrum of lead derived from thorium was identical with that of ordinary=

lead.* Furthermore in the same year the experiments of Richards and Lembert,^ Honigschmidt and HoroAvitz,*^ and Merton proved the same result. Merton concluded from his 1914 experiments that the difference in wave-length for the A 4058 line must be less than 0-003 A. Before going on to consider the more recent results it will be as well to discuss = the magnitude of the difference to be expected from theory.

109. The magnitude of the Gravitational effect. In

the Bohr theory of spectra the planetary electrons of the atom rotate round the central positively charged nucleus in various

1 F. p. 9.

2 Exiier and Haschek, Sitz. Akad. Wiss. Wien, iia, 121, 175, = 1912.

3 V. p. 39.

  • Soddy and Hyman, Jour. Chem. Soc, 105, 1402, 1914.

^ Richards and Lembert, Jour. Amer. Chem. Soc, 36, 1329, 1914.=

^ Honigschmidt and Horowitz, Sitz. Akad. Wiss. Wien, iia, = 123, 1914.

' Merton, Proc. Roy. Soc, 91A, 198, 1914.

121


122 ISOTOPES

stable orbits. The frequencies of the spectral lines emitted by the element are associated in an absolutely definite manner with the rotational frequencies of these orbits which are calculated by what is known as a " quantum " relation. Without going further into the theory it will be seen at once that if we alter the force acting between the central nucleus and its planetary electrons these orbits will change and with them the frequency of the light emitted. It is therefore of interest to examine the magnitude of the change, to be expected=

from this theory, when we alter the mass of the nucleus without=

changing its charge, and so pass from one isotope to another.

The difference in the system which will first occur to one is that although the electrical force remains the same the gravi- tational force must be altered. The order of magnitude of the change expected in the total force will clearly be given by=

considering the ratio between the electrical and gravitational forces acting, to take the simplest case, between the protou and the electron in a neutral hydrogen atom.

Assuming the law of force to be the same in both cases, this ratio is simply e^/GMm  ; where e is the electronic charge 4-77 X 10~i", G the universal gravitational constant 6-6 x 10"^,=

M the mass of the proton 1-66 x lO"^*^ and m the mass of the=

electron 9-0 x 10~ 2^. Putting in these numerical values we obtain the prodigious ratio 2-3 x 10 ^9. In other words the effect of doubling the mass of the nucleus without altering its=

charge would give the same percentage increase in the total pull on the planetary electron, as would be produced in the pull between the earth and the moon by a quantity of meteoric dust weighing less than one million millionth of a gramme falling upon the surface of the former body. The gravitational effect may therefore be dismissed as entirely negligible.

110. Deviation of the Bohr orbits due to change in the position of the centre of gravity of the rotating system. Although we may neglect the gravitational effect there is another, of quite a different order, which arises in th= e following manner. The mass of the electron compared with that of the nucleus is small but not absolutely negligible, hence=

it will not rotate about the nucleus as though that were a


THE SPECTRA OF ISOTOPES 123

fixed point, but both will rotate about their common centre of gravity. The position of this centre of gravity will be shifted by any alteration in the mass of the nucleus. If E, M=

and e, m are the respective charge and mass of the nucleus and=

the rotating electron, the equation of motion is

rM , Ee

M + m r^

where r is the distance between the two charges and w the angular velocity. Bohr ^ introduced this effect of the mass of the nucleus in order to account for the results obtained by Fowler. 2 The Bohr expression for the frequency then becomes

where e, E and m, M are the charges and masses of the electron=

and nucleus respectively. If we suppose that the atomic weight of lead from radium to be one unit less than that of ordinary lead, this theory predicts a difference in wave-length, for the principle line, of 000005 A between the two, a quantity=

beyond the reach of the most delicate methods of spectrum analysis used up to the present.

111. Later experiments of Aronberg and Merton.

In 1917 Aronberg,^ applying the extremely high dispersion derived from the spectrum of the sixth order of a Michelson 10-inch grating to the line A 4058 emitted from a specimen of radio-lead of atomic weight 206-318, observed a difiference of 0-0044 A between this and ordinary lead, of atomic weight 207-20. This remarkable result has been since confirmed by Merton of Oxford* who gives the difference of wave-length between radio-lead from pitchblende and ordinary lead as 0-0050^2 0-0007, Merton made use of a totally different optical system, namely a Fabry and Perot etalon, so that the agreement is very striking.

It is to be noticed that the effect observed was not a mere

1 Bohr, Nature, 92, 231, 1913.

2 Fowler, Nature, 92, 95, 1913.

3 Aronberg, Proc. Nat. Acad. Sci., Z, 710, 1917, and Ast= rophys, Jour., 47, 96, 1918.

4 Merton, Proc. Boy. Soc, 96A, 388, 920.


124


ISOTOPES


broadening of the line but a definite shift, and that, though of the same sign, it is about one hundred times greater than that predicted by the Bohr theory, Merton also found a shift of 0-0022 =C2=B10-0008 A between the wave-length of thorite-lead and ordinary lead, differing in atomic weight by about 0-6. The heavier atom shows the higher frequency in all cases. This remarkable discrepancy between the shift predicted by theory and that actually observed has been discussed by Harkins and Aronberg.^

At a recent discussion on isotopes at the Royal Society ^ Merton commented upon the line 6708 A emitted by the element lithium, which consists of two components 0-151 A apart. If lithium is accepted as a mixture of isotopes 6 and 7,= ^ he calculated that each of these components should be accom- panied by a satellite, some sixteen times as faint, displaced by=

0-087 A. So far he had not been able to observe such satellites= . Previous experiments of Merton and Lindemann* on the expected doubling in the case of neon had given no conclusive results on account of the physical width of the lines. It was hoped that this difficulty could be overcome by the use of liquid hydrogen temperatures.

StiU more recently Merton^ has repeated his experiments on lead, using a very pure sample of uranium lead from Australian Carnotite. His final results are indicated in the following table  :


A

(Carnotite lead)"! . ^(ordinary lead) J

r Wave niimber (ordinary lead) ' . Wave-number (Carnotite lead).

4058 3740 3684 3640 3573

0-011 =C2=B10-0008 0-0074=C2=B10-0011 0-0048=C2=B10-0007 0-0070=C2=B10-0003 0-0048=C2=B10-0005

0-065=C2=B10-005 0-053=C2=B10-008 0-035=C2=B10-005 0-C52=C2=B10-002 0-037=C2=B10-004

1 Harkiiis and Aronberg, Jour. Am. Chem. Soc, 42, 1328, 

Merton,  Proc.  Roy.  Soc.=C2=BB  99A,  87,     1921.

=C2=BB V. p. 86.

  • Lindemann, ibid.
 Merton,  Roy.  Soc.  Proc,  lOOA,  84,     1921.


1920.


THE SPECTRA OF ISOTOPES 125

It will be noticed that the shift for the line A 4058 is rathe= r more than twice that obtained before. Merton suggests that the most probable explanation of this difference is evidently that the Carnotite lead used is a purer sample of uranium lead=

than that obtained from the pitchblende residues. It is also apparent that the differences are not the same for different lines, an interesting and somewhat surprising result.

112. "Isotope" effect on the Infra-red spectrum of molecules. The extreme smaUness of the isotope " shift "=

described above in the case of line spectra emitted by atoms is=

due to the fact that one of the particles concerned in the vibration is the electron itself, whose mass is minute compared with that of the nucleus. Very much larger effects should be expected for any vibration in which two atoms or nuclei are concerned, instead of one atom and an electron. Such a vibration would be in the infra-red region of the spectrum.

This effect was first observed by Imes^ when mapping the fine structure of the infra-red absorption bands of the halogen acids. In the case of the HCl " Harmonic " band at 1-76^, mapped with a 20,000 line grating, the maxima were noticed to be attended by satellites. Imes remarks  : " The apparent tendency of some of the maxima to resolve into doublets in the=

case of the HCl harmonic may be due to errors of observation, but it seems significant that the small secondary maxima are all on the long-wave side of the principal maxima they accom- pany. It is, of course, possible that still higher dispersion applied to the problem may show even the present curves to be composite."

Loomis^ pointed out that these satellites could be attributed to the recently discovered isotopes of chlorine. In a later paper ^ he has shown that, if mi is the mass of the hydrogen nucleus, and ma the mass of the charged halogen atom, the

difference should be expressed by the quanity ^ = ~ the

square root of which occurs in the denominator of the expression=


^ Imes, Astrophysical Journal, 50, 251, 1919.

2 Loomis, Nature, Oct. 7, 179, 1920.

^ Loomis, Astrophysical Journal, 52, 248, 1920.


126 ISOTOPES

for frequency. " Consequently the net difference between the spectra of isotopes will be that the wave-lengths of lines in the spectrum of the heavier isotope will be longer than the=

corresponding lines for the lighter isotope in the ratio 1 + 1/1330  : 1 for chlorine and 1 -f 1/6478  : 1 for bromine.=

Since the average atomic weight of chlorine is 35-46 the amounts=

of CP^ and CP' present in ordinary chlorine must be as 1-54  : 0-46 or as 3-35  : 1 and, if the lines were absolutely = sharp and perfectly resolved, the absorption spectrum of ordinary HCl should consist of pairs of lines separated by 1/1330 of their frequency and the one of shorter wave-length should have about 3-35 the intensity of the other. The average atomic weight of bromine is 79-92, hence the two isotopes are present in nearly equal proportions and the absorption spectrum of HBr should consist of lines of nearly equal intensity separated by 1/6478 of their frequency."

The latter will be too close to be observed with the dispersion=

employed. In the case of the HCl band at IIQ ju the difference=

of wave number on this view should be 4-3. The mean differ- ence of wave number given by Loomis' measurements of 13 lines on Imes' original curves for this band is 4-5 ^ 0-4 corre= - sponding to 14 A in wave-length.

The spectroscopic confirmation of the isotopes of chlorine has also been discussed by Kratzer,! who considers that the oscillation-rotation bands of hydrogen chloride due to Imes^ are in complete accordance with the theory.

1 H. Ivratzer, Zeit. Physik., 3, 60, 1920.

  • Loc. cit.


CHAPTER XI - THE SEPARATION OF ISOTOPES

113. The Separation of Isotopes

The importance, from purely practical and technical points of view, of the theory of isotopes would have been insignificant had its application been confined to the radioactive elements and their products, which are only present in infinitesimal quantities on the Earth. But now that the isotopic nature of many elements in everyday use has been demonstrated, the possi- bility of their separation, to any reasonable extent, raises questions of the most profound importance to applied science. In physics all constants involving, e.g., the density of mercury=

or the atomic weight of silver may have to be redefined, while=

in chemistry the most wholesale reconstruction may be necessary for that part of the science the numerical founda- tions of which have hitherto rested securely upon the constancy of atomic weights.

It is therefore of great interest to consider in turn the various methods of separation proposed and examine how far they have been successful in practice.

114. Separation by Diffusion

he subject of the separation of a mixture of two gases by the method of Atmolysis or has been thoroughly investigated by the late Lord Rayleigh. The diffusion is supposed to take place through porous material. The conditions under which maximum separation is to be obtained are that " mixing " is perfect, so that there can be no accumulation of the less diffusible gas at the surface of the porous material, and that the apertures in the material through which the gases must

iRayleigh, Phil. Mag., 42, 493, 1896. 127


128 ISOTOPES

pass are very small compared with the mean free path of the molecules. If these conditions are satisfied he obtains as an expression for the effect of a single operation  :


X + y _ ^ . _^ Y


r '^


X + Y X + Y "-'^ X + Y "-'-

where (X Y) {x, y) are the initial and final volumes of the gases, /I, V, the velocities of diffusion, and r the enrichment=

of the residue as regards the second constituent.

The velocity of diffusion of a gas is proportional to the square root of the mass of its molecules, so that if a mixture=

of two isotopes is allowed to diffuse a change in composition must be brought about. Now no known isotopes differ from each other much in mass, so the difference between their rates of diffusion will also be small, hence the above equation=

may be written in the approximate form

^- =3D rTc where h =3D ^ a small quantity and,

and, finally, the enrichment by diffusion of the residue as regards the heavier constituent may be expressed with sufficient accm'acy by the expression


mi-m /Initial volume


Final volume


where Wi, mg are the molecular masses of the lighter and heavier isotope respectively. In the most favourable case known at present, that of the isotopes of neon, the number over the root is 21 so that the change in composition obtain- able in a single operation will in practice be very small.

If we take the density of the original mixture as unity, the increase in density of the residual gas to be expected from the=

operation of diffusion will be approximately

(r 1) X ^ X 2 ^


X Wg + Wi

Now neon consists of monatomic molecules differing between each other in mass by 10 per cent, and the heavier is present=

to the extent of 10 per cent. In the diffusion experiments described on p. 39 the effective ratio of the initial volume to=


THE SEPARATION OF ISOTOPES 129

the final volume was estimated as certainly greater than 500 and probably less than 10,000, so that r lies between 1-3 and 1-5. Hence the increase of density of the heavier residue should have been between -003 and -005. It was actually 004.

115. The separation of the isotopes of chlorine by the diffusion of HCl

In the case of other isotopic gaseous mixtures the numerical obstacles in the way of practical separation wiU be correspondingly greater. Thus in the case of HCl the 36th root is involved, and in that of HBr the 80th root. The only way by which measurable increase in density may be hoped for wiU clearly be by increasing the effective ratio of the initial to final volumes to an heroic degree. This can be done by experiments on a huge scale or by a vast number of mechanical repetitions.

Harkins started to attack the HCl problem in 1916 using the first of these two alternatives. In 1920 he mentions a quantity of 19,000 litres of HCl as having been dealt with in these experiments. 2 In the following year^ he published numerical results indicating that a change in atomic weight of 0-055 of a unit had been achieved.

At the recent discussion on isotopes * Sir J. J. Thomson pointed out that a change in the molecular weight of HCl should be caused by allowing a stream of the gas to flow over=

the surface of a material which absorbed it. The higher diffusion coefficient of the lighter isotope would result in it being absorbed more rapidly than the heavier one, so that the residue of unabsorbed gas should give a higher molecular weight. This " free diffusion " without the interposition of porous material has been recently tried in the Cavendish Laboratory by E. B. Ludlam, but no measurable difference has so far been detected.

116. Separation by Thermal Diffusion

It has been

^ Harkins, Jour. Amer. Cheni. Soc, Feb., 1916.

2 Harkins, Science, Mar. 19, 1920 ; Nature, Apl. 22, 1920 ; see=

also Phys. Rev., 15, 74, 1920  ; Science, 51, 289, 1920 ; Jour. = Amer, Chem. Soc, 42, 1328, 1920.

3 Harkins, Science, Oct. 14, 1921  ; Nature, Oct. 3, 1921.=

  • J. J. Thomson, Proc. Roy. Soc, 99A, 98, 1921.

K

shown on theoretical grounds independently by Enskog ^ and Chapman ^ that if a mixture of two gases of different molecular weights is allowed to diffuse freely, in a vessel of which the ends are maintained at two different temperatures T,T', until equilibrium conditions are reached, there will be a slight excess of the heavier gas at the cold end, and of the=

lighter gas at the hot end. The separation attained depends on the law of force between the molecules and is a maximum if they behave as elastic spheres. The effect was experi- mentally verified for a mixture of CO2 and Ha by Chapman and Dootson,^ and recently Ibbs * has demonstrated that the separation can be carried out continuously and that the time for equilibrium to be established is quite short.

Chapman has suggested ^ that thermal diffusion might be used to separate isotopes. He shows that the separating power depends on a constant ^x. And when the difference between the molecular masses mi, ma is smaU the value of this is approximately given by

, _ 17 ma mi AiAj

^^ ~~ 3 ma + mi 9-15 8-25 AiAa where ^1,^2 denote the proportions by volume of each gas in the mixture  ; thus Ai -f Aa =3D=3D1. The actual separation = 

is

given by

Ai A'l =3D (Ai A'a) =3DA;t log T'/T.=


He gives the following numerical example  : " Suppose that it is=

desired to separate a mixture of equal parts of Ne^" and Ne^^,=

then, writing mi =3D 20, ma =3D 22, Ai =3D A3 =3D ^, we find = 

that

Ic,^ =3D 0-0095. Suppose that the mixture is placed in a vessel=

consisting of two bulbs joined by a tube, and one bulb is maintained at 80=C2=B0 absolute by liquid air, while the other is=

heated to 800=C2=B0 absolute (or 527=C2=B0 C). When the steady st= ate has been attained the difference of relative concentration between the two bulbs is given by the equation

1 Enskog, Phys. Zeit., 12, 538, 1911  ; Ann. d. Phys., 38, 75= 0, 1912.

2 Chapman, Phil. Trans., 217A, 115, 1916; Phil. Mag., 34, 146, 1917.

3 Chapman and Dootson, Phil. Mag., 34, 248, 1917.

  • Ibbs, Proc. Boy. Soc, 99A, 385, 1921.

^Chapman, Phil Mag., 38, 182, 1919.


THE SEPARATION OF ISOTOPES 131

Ai A'l =3D (A 2 A' 2) =3D 0-0095 lo= g, 800/80

=3D 0-022

or 2-2 per cent. Thus the cold bulb would contain 48-9 per cent. Ne^" to 51-1 per cent. Ne^^, and vice versa in the hot bulb. By drawing o=C2=A3f the contents of each bulb separately, and by repeating the process with each portion of the gas, the=

difference of relative concentrations can be much increased. But as the proportions of the two gases become more unequal, the separation effected at each operation slowly decreases. For instance, when the proportions are as 3  : 1, the variation=

at each operation falls to 1-8 per cent.  ; while if they are a= s 10  : 1 the value is 1-2 per cent. This assumes that the mole-=

cules behave like elastic spheres  : if they behave like point centres of force varying as the inverse nth. power of the distan= ce, the separation is rather less; e.g., ii n=3D9, it is just over=

half the above quantities."

Chapman points out that for equal values of log p/p and log T/T pressure diffusion (centrifuging) is about three times as powerful as thermal diffusion but suggests that it may be more convenient to maintain large differences of temperature than of pressure.

117. Separation by Gravitation or "Pressure Diffusion"

When a heterogeneous fluid is subjected to a gravitational field its heavier particles tend to concentrate in the direction of the field, and if there is no mixing to co= unter- act this a certain amount of separation must take place. If therefore we have a mixture of isotopes in a gaseous or liquid=

state partial separation should be possible by gravity or centrifuging.

The simplest case to consider is that of the isotopes of neon in the atmosphere and, before the matter had been settled by the mass-spectrograph, analysis of the neon in the air at very great heights was suggested as a possible means of proving its isotopic constitution. 1 The reasoning is as follows: =E2=80= =94

If M be the atomic weight, g the gravitational constant, p the pressure, and p the density, then if no mixing takes place dp =3D gpdh, h being the height. In the isother= mal

1 Lindemann and Aston, Phil. Mag., 37, 530, 1919.


132 ISOTOPES

layer convection is small. If it is small compared with diffusion the gases will separate to a certain extent. Since T is constant

RTp , dp Mp ,,

whence p =3D pffi Rt ,

Po being the density at the height Jiq at 'which mixing by convection ceases, about 10 kilometres, and A^ the height above this level. If two isotopes are present in the ratio 1 to Ko, so that the density of one is po and of the other Kopo=

at height Jiq, then their relative density at height h^ + /SJi = is given by

Putting T =3D 220 as is approximately true in England,

XT

A^ being measured in kilometres. If Mi Ma =3D 2, th= erefore

It might be possible to design a balloon which would rise to 100,000 feet and there fill itself with air. In this case the relative quantity of the heavier constituent would be reduced from 10 per cent, to about 8-15, so that the atomic weight of=

neon from this height should be 20-163 instead of 20-2. If one could get air from 200,000 feet, e.g. by means of a long-=

range gun firing vertically upwards, the atomic weight of the neon should be 20-12.

A more practicable method is to make use of the enormous gravitational fields produced by a high speed centrifuge.

In this case the same equation holds as above except that g varies from the centre to the edge. In a gas therefore <ip__Mv2 dr _ _Mo)'^ ~^ ~ Rf "y ~ RT '

whence p =3D poe 2rt,

Vq being the peripheral velocity. Here again, if Kq is the


THE SEPARATION OF ISOTOPES 133

ratio of the quantities present at the centre, the ratio at the=

edge will be

A peripheral velocity of 10^ cm,/s. or perhaps even 1-3 x 10^ cm./s. might probably be attained in a specially designed

rr

centrifuge, so that:^^ might be made as great as e"=C2=B0'2^^'^'~^=

  • ^ or

even e ~0'^'^^'^>~^2),

If Ml M2 is taken as 2 a single operation would there= fore give fractions with a change of K of 0-65. In the case of neon=

the apparent atomic weight of gas from the edge would be about 0-65 per cent, greater than that of gas from the centre,=

i.e. a separation as great as the best yet achieved in practice=

by any method could be achieved in one operation. By centrifuging several times or by operating at a lower tempera- ture the enrichment might be increased exponentially.

Centrifuging a liquid, e.g. liquid lead, would not appear so favourable, though it is difficult to form an accurate idea of the quantities without a knowledge of the equation of state. If compression is neglected and the one lead treated as a solution in the other, a similar formula to that given above holds. On assumptions similar to these Poole ^ has calculated that a centrifuge working with a peripheral velocity of about 10^ cm. /sec should separate the isotopes of mercury to an extent corresponding to a change of density of 0-000015.

The only experiments on the separation of isotopes by the use of a centrifuge, so far described, are those of Joly and Poole 2 who attempted to separate the hypothetical isotopic constituents of ordinary lead by this means. No positive results were obtained and the check experiments made with definite alloys of lighter metals with lead were by no means encouraging.

118. Separation by Chemical Action or Ordinary Fractional Distillation. The possibility of separating iso-=

topes by means of the difference between their chemical affinities or vapour pressures has been investigated very fully

1 Poole, Phil. Mag., 41, 818, 1921.

2 Joly and Poole, Phil. Mag., 39, 372, 1920.


134 ISOTOPES

from the theoretical standpoint by Lindemann. The thermo- dynamical considerations involved are the same in both cases. The reader is referred to the original papers ^ for the details=

of the reasoning by which the following conclusion is reached  : =

" Isotopes must in principle be separable both by fractiona- tion and by chemical means. The amount of separation to be expected depends upon the way the chemical constant is calculated and upon whether ' NuUpunktsenergie ' is assumed. At temperatures large compared with ^v,^ which are the only practicable temperatures as far as lead is concerned, the difference of the vapour pressure and the constant of the

Bv law of mass action may be expanded in powers of ^. The

Bv most important term of the type log "^ is cancelled by the

chemical constant if this is calculated by what seems the only

Bv reasonable way. The next term in is cancelled by the=


' NuUpunktsenergie ' if this exists. All that remains ar= e

Bv terms containing the higher powers of ^. In practice there- fore fractionation does not appear to hold out prospects of success unless one of the above assumptions is wrong. If the first is wrong a difference of as much as 3 per cent, should occur at 1200 and a difference of electromotive force of one miUivolt might be expected. Negative results would seem to indicate that both assumptions are right."

As regards experimental evidence it has already been pointed out that the most careful chemical analysis, assisted by radio- active methods of extraordinary delicacy, was unable to achieve the shghtest separation of the radioactive isotopes. The laborious efforts to separate the isotopes of neon by a differ-=

ence of vapour pressure over charcoal cooled in hquid air also gave a completely negative result.

119. Separation by evaporation at very low pressure

If a liquid consisting of isotopes of different mass is allowed

1 Lindemann, Phil. Mag., 37, 523, 1919  ; 38, 173, 1919.

  • (iv is the " characteristic " and T the " Absolute " tempera=

ture.


THE SEPARATION OF ISOTOPES 135

to evaporate it can be shown that the number of Hght atoms escaping from the sm'face in a given time will be greater than=

the number of heavier atoms in inverse proportion to the square roots of their weights. If the pressure above the surface is kept so low that none of these atoms return the concentration of the heavier atoms in the residue will steadily increase. This method has been used for the separation of isotopes by Bronsted and Hevesy, who appUed it first to the element mercury.

The mercury was allowed to evaporate at temperatures from 40=C2=B0 to 60=C2=B0 C. in the highest vacuum attainable. The eva= porat- ing and condensing surfaces were only 1 to 2 cms. apart, the latter was cooled in liquid air so that all atoms escaping reached it without coUision and there condensed in the sohd form.

It will be seen that the Uquid surface acts exactly Uke the porous diaphragm in the diffusion of gases. ^ The diffusion rate of mercury can be obtained approximately from the diffusion rate of lead in mercury ^ and is such that the mean=

displacement of the mercury molecule in Uquid mercury is about 5 X 10"^ cm. sec."^. It follows that if not more than 5 X 10"^ c.cm. per cm.^ surface evaporate during one second no disturbing accumulation of the heavier isotope in the surface layer takes place.

The separation was measured by density determination. Mercury is particularly well suited for this and a notable feature of this work was the amazing deUcacy with which it could be performed. With a 5 c.cm. pyknometer an accuracy of one part in two millions is claimed. The first figures pubhshed ^ were  :

Condensed mercury. . . . 0-999981

Residual mercury .... 1-000031

The densities being referred to ordinary mercury as unity.

The later work was on a larger scale.* 2700 c.cm. of mercm-y were employed and fractionated systematically to about

1 V. p. 127.

  • Groh and Hevesy, Ann. der Phys., 63, 92, 1920.

^ Bronsted and Hevesy, Nature, Sept. 30, 1920.

  • Bronsted and Hevesy, Phil. Mag., 43, 31, 1922.


136 ISOTOPES

1/100,000 of its original volume in each direction. The final=

figures were  :

Lightest fraction vol. 0-2 c.c. . . 0-99974

Heaviest fraction vol. 0-3 c.c. . . 1-00023

Mercury behaves as though it was a mixture of equal parts of two isotopes with atomic weights 202-0, 199-2 in equal parts or of isotopes 201-3, 199-8 when the former is four times=

as strong as the latter, and so on.

120. Separation of the isotopes of chlorine by free evaporation

The same two investigators were able to announce the first separation of the isotopes of chlorine ^ by applying the above method to a solution of HCl in water. This was allowed to evaporate at a temperature of 50=C2= =B0 C. and condense on a surface cooled in hquid air. Starting with 1 litre 8-6 mol. solution of HCl 100 c.c. each of the lightest=

and heaviest fraction were obtained.

The degree of separation achieved was tested by two difiEerent methods. In the first the density of a saturated solution of NaCl made from the distillate and the residue respectively was determined with the following results  :

Density (salt from distillate) =3D 1-20222 Density (salt from residue) =3D 1-20235

These figures correspond to a change in atomic weight of 0-024 of a unit.

In the second method exactly equal weights of the isotopic NaCls were taken and each precipitated with accurately the same volume of AgNOg solution, in shght excess. After pre- cipitation and dilution to 2,000 c.c. the approximate concen- tration of the filtrate was determined by titration, also the ratio of Ag concentration of the two solutions was measured in a concentration cell. Calculation showed that the difference in atomic weight of the two samples was 0-021 in good agree- ment with the density result.

121. Separation by Positive Rays

The only method which seems to offer any hope of separating isotopes completely, and so obtaining pure specimens of the constituents of a com-

1 Bronsted and Hevesy, Nature, July 14, 1921.


THE SEPARATION OF ISOTOPES 137

plex element, is by analysing a beam of positive rays and trapping the particles so sorted out in different vessels. It is=

therefore worth while inquiring into the quantities obtainable by this means.

Taking the case of neon and using the parabola method of analysis with long parabolic slits as collecting vessels we find=

that the maximum separation of the parabolas corresponding to masses 20 and 22 (obtained when electric deflexion d is haK the magnetic) is approximately

^ 1 M,-M, _ d_ V2 Ml 28"

Taking a reasonable value of 0 as -3 the maximum angular width of the beam for complete separation =3D 0-01. If the canal-ray tube is made in the form of a slit at 45=C2=B0 to ax= es, i.e. parallel to the curves, the maximum angular length of the beam might be say 5 times as great, which would collect the positive rays contained in a solid angle of -0005 sq. radian= .

The concentration of the discharge at the axis of the positive ray bulb is considerable, and may be roughly estimated to correspond to a uniform distribution of the entire current over a |- sq. radian. One may probably assume that half the current is carried by the positive rays, and that at least half=

the positive rays consist of the gases desired. If neon is analysed by this method therefore the total current carried by the positive rays of mass 20 is

0005 x4:Xixlxi=3D -0005 i.

If i is as large as 5 miUiamperes this =3D 1-5 x 10* E.S.U.=

1-5 X 10*


or


2-7 X 1019 X 4-77 X 10-1"


=3D 1-2 X 10"^ c.c./sec.


i.e. one might obtain about one-tenth of a cubic millimetre of Ne2o and 1/100 cubic miUimetre of Ne^^ per 100 seconds run. It is obvious that even if the difficulties of trapping the rays=

were overcome, the quantities produced, under the most favourable estimates, are hopelessly small.

122. Separation by photochemical methods

A remarkably beautiful method of separating the isotopes of=


138 ISOTOPES

chlorine has been suggested by Merton and Hartley which depends upon the following photochemical considerations. Light falling on a mixture of chlorine and hydrogen causes these gases to combine to form hydrochloric acid. This must be due to the activation of the atoms of hydrogen or those of=

chlorine. Supposing it to be the latter it is conceivable that the radiation frequency necessary to activate the atoms of Cl^^ will not be quite the same as that necessary to activate those of CP'^. CaUing these frequencies 5^35 and V37 respectively=

it would seem possible, by excluding one of these frequencies entirely from the activating beam, to cause only one type of chlorine to combine and so to produce pure HCI^^ or HCI^'. Now ordinary chlorine contains about three times as much CP^ as CP^ and these isotopes must absorb their own activat- ing radiation selectively. In this gas therefore light of frequency V35 will be absorbed much more rapidly than that of frequency V37, so that if we aUow the activating beam to pass through the right amount of chlorine gas V35 might be completely absorbed but sufficient V37 radiation transmitted to cause reaction. On certain theories of photo-chemistry light containing ^37 but no V35 would cause only atoms of CP^ to combine so that a pure preparation of HCP^ would result. Pure CP'^ made from this product could now be used as a filter for the preparation of pure HCP^, and this in its turn would yield pure CP^ which could then be used as a more efficient filter for the formation of more HCP^

Had this very elegant scheme been possible in practice it would have resulted in a separation of a very different order to those previously described and the preparation of un- limited quantities of pure isotopes of at least one complex element. There is however little hope of this, for so far the results of experiments on this method have been entirely negative.

123. Other methods of separation and general conclusions

The following methods have also been suggested. By the electron impact in a discharge tube, in the case of the inert gases, the Ughter atoms being more strongly urged towards


THE SEPARATION OF ISOTOPES 139

the anode  ;^ by the migration velocity of ions in gelatine  ; ^=

by the action of light on metallic chlorides,^

A survey of the separations actually achieved so far shows that from the practical point of view they are very small. In cases where the method can deal with fair quantities of the substance the order of separation is small, while in the case of complete separation (positive rays) the quantities produced are quite insignificant. We can form some idea by considering the quantity

Q =3D (difference in atomic weight achieved) X (average quantity of two fractions produced in grammes). As regards the first of these factors the highest figure so far was 0-13 obtained by the writer in the original diffusion experiments on neon, but as the quantities produced were only a few milli- grams Q is negligibly small. The highest values of Q have been obtained by Bronsted and Hevesy by their evaporation method.* It is 0-5 in the case of Hydrochloric Acid, 0-34 in that of Mercury.

When we consider. the enormous labour and difficulty of obtaining this result it appears that unless new methods are discovered the constants of chemical combination are not likely to be seriously upset for some considerable time to come.=


1 Skaupy, Zeitsch. Phys., 3, 289, 460, 1920.

2 Lindemann, Proc. Roy. Soc, 99A, 104, 1921.

3 Renz, Zeit. Anorg. Chem., 116, 62, 1921.

  • V. p. 134.


APPENDIX I

Table of atomic weights and isotopes of the elements. 

 The  elements  are  given  in  order  of  their  atomic=
 numbers.  The

different periods are indicated by gaps after the inert gases. A curious relation, pointed out by Rydberg, is that the atomic numbers of all the inert gases are given by taking the series 2 (P + 2^ + 22 + 3^ + 3^ + 4^ + = ) and stoppmg the summation at any term. This gives the numbers used by Langmuir (p. 95).

The atomic weights given are the International ones except in the cases marked with an asterisk, where the figures are taken f= rom some of the recent determinations given below.

The isotopes where known are given in order of their atomic masses. The proportion of an isotope in a complex element is indicated by the index letters a, 6, c ... in descending order.=

In the case of isotopes of the radioactive elements 81-92 the ro= man numeral gives the number of them believed to exist. The nomen- clature of some of the rare earths 69-72 is not yet standardised.=

The names here are those used by Moseley. Some of these elements= , though detected by their X-ray spectra, have never been isolated.=

The elements corresponding to atomic numbers 43, 61, 75, 85, 87=

(all odd) have not yet been discovered.

Recent atomic weight determinations. The following is a list of some of the elements whose atomic weights have been re-=

determined quite recently, together with references to the papers in which they were published. Where more than one value is given different methods were used  :

Fluorine 19-001. Moles and Batuecas, Jour. Chim. Phys., 18, 35= 3,

1920. Aluminium 26*963. Richards and Krepelka, Journ. Am. Chem. Soc,=


42, 2221, 1920. Silicon 28-111. Baxter, Weatherelland Holmes, ibid., 42, 1194, = 

1920.

Scandium 45-10. Honigschmid, Zeit. Electrochem., 25, 93, 1919.=

Tin 118-703. Baxter and Starkweather, Journ. Am. Chem. Soc, 42,=


905, 1920.

118-699. Brauner and Krepelka, ibid., 42, 917, 1920.

141


142


APPENDIX I


Tellurium 127-73, 127-79. Bruylants and Michielsen, Bull= . Acad.

Bdg., 119, 1919. Samarium 150 "43. Owens, Balke and Kremers, Journ. Am. Chem= .

Soc, 42, 515, 1920. Thtdium 169-44, 169-66. James and Stewart, ibid., 42, 2022, = 1920. Bismuth 209-02. Honigschmid, Zeit. Electrochem., 26, 403, 1920= .

208-9967. Classen and Wey, Ber., 53, 2267, 1920. Antimony 121-773. Willard and McAlpine, Jouryi. Am. Chem. Soc, = 43,

797, 1921. Lanthanum 138-912. Baxter, Tani and Chapin, Journ. Am. Chem.=


Soc, 43, 1085, 1921. Germanium 72-418. Miller, Journ. Am. Chem. Soc, 43, 1085, 19= 21. Zinc 65-38. Baxter and Hodges, i&id., 43, 1242, 1921. Cadmium 112-411. Baxter and Wilson, ibid., 43, 1230, 1921.


-Q

" m

o^

Element.

2

a

if

Masses of isotopes.

=C2=A3 -2 *^ Hydrogen . .

H

1

1-008

1

1-008

f^^'o Helium . . .

He

2

4-00

1

4

&> 1"

00 Lithivim .

Li

3

6-94

2

-

" Beryllium

Be

4

91

1

9

r^ Boron

B

5

10-9

2

10=C2=BB 11"

3 Carbon .

C

6

12-00

1

12

S Nitrogen .

N

7

14-008

1

14

^ Oxygen . . .

0

8

16-00

1

16

0 Fluorine .

F

9

19-00

1

19

^ Neon ....

Ne

10

20-20

2

20" 22* 23

oQ Sodium .

Na

11

2300

1

^ Magnesium .

Mg

12

24-32*

3

24-=3D 25* 26^

Aluminium .

Al

13

26-96*

_o Silicon

Si

14

28-3

2

28" 29* (30)

3 Phosphorus .

P

15

31-04

1

31

^ Sulphur . . .

s

16

3206

1

32

'S Chlorine . . .

CI

17

35-46

2

35" 37* (39)

^ Argon . . .

A

18

39-9

2

36* 40" 39" 41*

Potassium

K

19

39-10

2

Calcium .

Ca

20

40-07

(2)

40 (44)

Scandium

Sc

21

45-1*

Titanium .

Ti

22

48-1

Vanadium

V

23

510

0

2 Chromium .

Cr

24

52-0

H Manganese .

Mn

25

54-93

' Iron ....

Fe

26

55-84

n

^ Cobalt . . .

Co

27

58-97

J Nickel

Ni

28

58-68

2

58" 60*

P

n Copper .

Cu

29

63-57

J

=3D Zinc ....

Zn

30

65-37

(4)

(64=C2=B0 66* 68 7O<0

  • Galliimi . . .

Ga

31

70-10

Germanivmi .

Ge

32

72-5

Arsenic .

As

33

74-96

1

75

Seleniima .

Se

34

79-2

Bromine .

Br

35

79-92

2

79" 81*

Krypton .

Kr

36

82-92

6

78/ 80 82'^ 83-^ 84=C2=BB

86*

APPENDIX I


143


"S .

^

o *^

O^i

o ^^

Element

o

X!

E >,

00

Masses of Isotopes.

Rubidium

Rb

37

85-45

2

85" 87*

Strontium

Sr

38

87-63

Yttrium .

Y

39

89-33

Zirconium

Zr

40

90-6

Niobium .

Nb

41

93-1

00 Molybdenum

Mo

42

96-0

  • H _ ~


43


'-' Ruthenium .

Ru

44

101-7

'o Rhodium.

Rh

45

102-9

=C2=A7 Palladium

Pd

46

106-7

An Silver ....

Ag

47

107-88

X Cadmium

Cd

48

112-40

"O Indiimi .

In

49

114-8

Tin ... .

Sn

50

118-7

Antimony

Sb

51

120-2

Tellurium

Te

52

127-5

Iodine

I

53

126-92

1

127

L Xenon

X

54

130-2

(7)5

(128) 129" (130) 13P 132=C2=BB 134 136"

Caesium .

Cs

55

132-81

1

133

Barium .

Ba

56

137-37

Lanthanum .

La

57

139-0

Cerium

Ce

58

140-25

Praseodymium .

Pr

59

140-6

Neodymiimi .

Nd

60

144-3



61


Samarium

Sm

62

150-4

Europium

Eu

63

152-0

Gadolinium .

Gd

64

157-3

Terbium .

Tb

65

159-2

Dysprosium .

Ds

66

162-5

c

5 Holmium

Ho

67

163-5

J, Erbium .

Er

68

167-7

=C2=B0 Thulium . . .

Tu

69

168-5

1 Ytterbiiun . .

Yb

70

173-5

'C Lutecuim

Lu

71

175

Pm (Keltium) . .

(Kt)

72

ji Tantalum

Ta

73

181-5

<=C2=BB Tungsten.

W

74

1840



75


Osmium .

Os

76

190-9

Iridium .

Ir

77

193-1

Platinimi .

Pt

78

195-2

1

Gold ....

Au

79

197-2

Mercury .

Hg

80

200-6

(6)

(197-200) 202 204

Thallium . . .

Tl

81

204-0

IV

Lead ....

Pb

82

207-2

XI

Bismuth .

Bi

83

209-0*

V

Poloniuna

Po

84 85

z

VII

L Emanation

Em

86

222-0

III

i

87

.2 Radium . =C2=AE Actinium.

Ra

88

226-0

IV

Ac

89


II

^ Thorium . . .

Th

90

23215

VI

^ Uranium X .

UX

91

II

t_ Uranium

Ur

92

238-2

II

APPENDIX II

The Periodic Table of the Elements. The atomic numbers ar= e given in bold type, the atomic weights in italics and the isotopes, where = known, in ordinary numerals. The roman ntmierals indicate the chemical groups and the most important associated valencies are given below them. Elem= ents are placed to the left or to the right of the columns according= to their chemical properties, those in the same vertical line as each other have s= trong chemical similarities. The Rare Earth group is surrounded by a thick line.= Elements 59-72 have no properties pronounced enough to give them definite = places in the table. The properties of the missing elements can be p= redicted with

PERIODIC TABLE OF


IH

1-008


Valency

0

I

+ 1

II

+ 2

III

+ 3

IV

+ 4

2 He

4-00 4

3 Li

6-94 6, 7

4 Be

9-1

9

5B 10-9 10, 11

60

12-00 12

10 Ne

20-2 20, 22

11 Na

23-00 23

12 Mg

24-32

24, 25, 26

13 AI

26-96

14 Si 28-3 28,29

18 A

39-9 36, 40

19 K

39-1 39, 41

29 Cu

63-57

20 Ca

40-07

30 Zn

65-37

21 Sc 45-1

31 G

70-1

22 Ti 48-1

32 Ge

72-5

36 Kr

82-92

78, 80, 82, 83, 84, 86

37 Rb

85-45

85, 87

47 Ag 107-88

38 Sr

87-83

48 Cd 112-40

39 Y

89-33

49 In

114-8

40 Zr

90-6

50 Sn

118-7

54 Xe

130-2

129, 131, 132, 134, 136

55 Cs

132-81

133

56 Ba

137-37

57 La 58 Ce 139-0 140-25

59 Pr eONd 61 62 Sm 63 Eu = 

    64  Gd           65  Tb

140-6 144-3 150-4 152-0 = 

     157-3           159-2

66 Ds 67 Ho 68 Ev 69 Tu 70 Yb 7= 1 Lu 72 (Kt) 162-5 163-5 1677 168-5 173-5 = 

175

79 Au

197-2

80 Hg

200-6 197-204

81 Tl

204-0

82 Pb

207-2

86 Em

222-0

87-

88 Ra

226-0

89 Ac

90 Th

232-15

144

considerable certainty from the positions of their atomic numbers. From the point of view of the construction of the atom the inert gas= es should mark the end of the periods as they are shown to do ua the hst of = atomic weights in Appendix I, on the other hand it is more usual in chemistry = to start with valency 0. From principles of general convenience of arrangement t= he latter plan is adopted in this table, which is intended to give = the maximum amount of chemical information. Hydrogen, which belongs equally wel= l to group I or group VII, is best omitted from the. table altoget= her.

THE ELEMENTS


V

VI

VII

VIII

3

2

-

-1

7N

80

9F

14-01

16-00

1900

14

16

19

15 P

16 S

17 CI

31-04

32-06

35-46

31

32

35, 37

23 V

24 Cr

25 Mn

26 Fe

27 Co

28 Ni

Sl-O

33 As

74-96 75

52-0

34 Se

79-2

54-93

35 Br

79-92 79, 81

55-85

58-97

58-68 58.60

41 Nb

42 Mo

43

44 Ru

45 Rh

46 Pd

93-5

51 Sb 120-2

96-0

52 Te 127-5

531

126-92 127

101-7

102-9

106-7

73 Ta

74 W

7&-

76 0a

77 Ir

78 Pt

181-5

83 Bi

209-0

184-0

84 Po

85

190-9

1931

195-2

91 UX

ii

92 U

238-2

145

Recent results obtained by Dempster. Thanks to a private=

communication the writer is able to include some further results=

obtained by Dempster and a diagram of his apparatus for obtaining=


Fig. 19. Diagram of Anode in Dempster's latest apparatus.=


positive rays from metals. A full account is to appear in the Physical Review. Fig. 19 shows the new arrangement of vaporising furnace A and ionising filament C. The analysing apparatus has already been described on p, 31 and the results wi= th


.4F


5-9


f

'

1

k

Lithium.

\

1

\

1

\

)

J

[

<=3D/

v..

^^

/

K

9

30


ZO


10


60


6-1


6-9

Atomic Weight.


7-0


7-1


Fig. 20. Curve for Lithium. 146

APPENDIX III

147


magnesium on p. 81. Fig. 20 shows one of the curves obtained with lithium. It will be seen that the relative intensities of t= he isotopes is entirely different from that found by the writer (p. = 

86)

and also disagrees very definitely with the chemical atomic weight= . Dempster describes these relative intensities as varying very considerably. This is a most remarkable phenomenon and further information upon it is very desirable. There seems just a possibi= lity that the 6 line is enhanced by doubly charged carbon but it is = 

not

easy to see where such particles could be produced.

l/oltS 943 928 913-5 899-5 886 873 860 847-5=


J

\

Zinc.

1

t

\

1

\

1

\

f

\

r

\

\

1

1

\

\i

1

1

\

/

\

I

/

1

=C2=AE

l/

\

1

i^

\

^^

62 63 64 65 66 67 Atomic Weight.

Fig. 21. Curve for Zinc.


68 69


70


Fig. 21 gives a remarkable curve obtained from zinc. This indicates three strong isotopes and a faint fourth. The absolute=

scale of atomic weight is not known with certainty, and the valu= es 63, 65, 67, 69 are given by Dempster as those in best agreement=

with the atomic weight 65-37. Considering that the error in th= e


148 APPENDIX III

mean atomic weight of lithium, when calculated on these lines, is about 5 per cent, it would appear possible that these might = be a unit too high or too low. The probability of this is strengthene= d very much by the rule given on p. 110 connecting even atomic number with even atomic weight.

Results with calcium show only one line. This makes it extremely=

probable that this is a simple element of atomic weight 40 and=

therefore an isobare of argon. ^

Note. In a still later communication Dempster states that = 

he

has been successful in using an anode of calcium to which a sma= U quantity of zinc had been added. By this means he is able to compare the masses of the zinc isotopes with the strong calcium=

maximum, assumed as 40. This gives the atomic weights as 64, 66, 68 and 70. The intensities are quite different to those in = the curve given above for zinc. 64 is now the strongest, 66 and 68=

fainter, while 70 is very faint indeed. No explanation is yet advanced for these remarkable irregularities in relative intensity.=


He has also observed a small maximum at 44 invariably accom- panying the strong calcium maximum 40. This he considers to be probably due to an isotope of that element present in smaU quant= ity as suggested by the atomic weight 40 07.

The above values are included provisionally in the tables on pages 89 and 142.

" V. p. 88.


INDEX

Abnormal hydrides, 98

Abundance of the elements, 111

Accuracy of mass-spectrograph, 60

Actinivim chain, 14, 15

Additive law of mass, 99

Alkali metals, mass-spectra of, 83

Alpha ray changes, 13

Analysis of the elements, 63

Andrade and Rutherford, 11

Anode, composite, 80, 86 

      hot,  80,  83,  84

Anticathode, silica, 48

Antimony, 78

Argon, 66

Aronbeeg, 123

,, and Harkins, 124

Atmolysis, separation by, 127

Atomic number, 13, 93 

      theory,  2

,, volume of isotopes, 18 

      weights,  tables  of,  89,  141
      weights  of  radio -elements,  13,

141

Atoms, structure of, 90

Balke, Owens and Kremers, 142 Barkla, 93

Batuecas and Moles, 141 Baxter and Hodges, 142 and Parsons, 113 and Starkweather, 141 and Wilson, 142 Tani and Chapin, 142 Weatherell and Holmes, 73, 142 Beryllium, 88 Beta ray change, 13 Bohr, 94, 95, 121, 122, 123

,, atom, 95 BOLTWOOD, 1, 7 Boron, 72 

     anomalous  atomic  weight  of,

114 

     trifluoride,  73

Bracketing, method of, 59, 69 Brauner and Krepelka, 141 Broek, Van den, 93, 94, 116 Bromine, 76


Bronsted and Hevesy, 135, 136, 139

Brosslera, 102, 104

Bruylants and Michielson, 142

Caesium, 87

,, anomalous atomic weight of, 114 Calcium, 88, 148 Calibration curve, 55 Camera of mass-spectrograph, 51 

      positive  ray,  26

Canalstrahlen, 22 Carbon, 63

Carnotite, lead from, 124 Cathode rays, 22, 24 Chadwick, 94 

 and  Rutherford,  103

Chapin, Baxter and Tani, 142 Chapman, 130 

        and  DooTSON,  130

Chemical action, separation by, 133 

       law  of  radioactive  change,

11 Chlorine, 65, 113 

       separation  of   the  isotopes

of, 136 Classen, 31

and Wey, 142 Claude, 35 Cleveite, lead from, 17 Coincidence, method of, 57 Composite anode, 80, 86 Constancy of chemical atomic weights,

22 Cosmical effect of change of mass, 103 Crookes, 3, 4, 24, 115, 117 ,, dark space, 24, 35 

       theory  of  the  evolution  of

elements, 117 Curie, Mlle. I., 113 

    M.,  18

Dalton's hypothesis, 2 Darwin, 15

Davies and Horton, 68 Deflection of positive rays, 27 Dempster, 31, 80, 81, 86, 114, 146


149


150


INDEX


Dempster's method of analysis, 31,146 Density balance, 35

,, of isotopic leads, 17, 18 Diffusion of neon, 39

separation by, 127 velocity, determination of, 20 Disintegration theory of the evolu- tion of elements, 116 Distillation of neon, 37 Distribution of lines on mass-

spectrum, 64 DooTSON and Chapman, 130 Du Bois magnet, 61

Eddington, 104

Einstein's theory of relativity, 103 Electrical theory of matter, 90 Electric discharge in gases, 23

,, field of mass-spectrograph, 50 Electricity as an element, 115 Electrochemical properties of isotopes,

10 Electron, the, 91

Element, meaning of the word, 115 Enskog, 130 Epstein, 95 ExNER and Haschek, 121

Fa JANS, 11

First order lines, 61

Fleck, 12

Fluorine, 72, 97

Focussing positive rays, 44

FOWLEB, 123 

      and  Aston,  45

Fractional distillation, separation by,

133 Fbanck and Knipping, 68

Gehrcke, 102

,, and Reichenheim, 80, 83, 88 Geigek and Nuttall, 10, 13 Goldstein, 22 Gravitation effect on spectra, 121 

       separation  by,  131

Groh and Hevesy, 20, 135

Hahn, 8 

       and  Meitner,  8

Halation effect, 60 Half-tone plates, 25 Hall and Harkins, 116 Harkins, 102, 111, 116, 129 

        and  Aronberg,  124

        and  Hall,  116

,, and Wilson, 116 Haschek and Exner, 121 Helium, 67, 69, 106


Hevesy, 10, 12, 19 

      and  Bronsted,  136,  136,

139 

      and  Groh,  20,  135
      and  Paneth,  11
      and  Zechmeisteb,  20

Hodges and Baxter, 142 Holmes, Baxteb and Weathebell,

73, 141 Honigschmid, 17, 18, 141, 142 

 and     Horovitz,     18,

121 Horovitz and Honigschmid, 18, 121 HoBTON and Davies, 68 Hot anode, 80, 83, 84 Hydrochloric acid, diffusion of, 129 Hydrogen, 67, 69, 106 Hyman and Soddy, 17, 121

Ibbs, 130

Imes, 125, 126

Indicators, radioactive, 19

Infra-red spectrum of isotopes, 125

Intensity of positive rays, 44

Iodine, 78

Ionic dissociation theory, proof of, 20

lonisation in discharge tube, 24

Ionium, 1, 7, 9, 18

,, atomic weight of, 18 Isobares, 12, 13, 97, 110 Isotopes, definition of, 12

diagrams of, 97

discovery of, 5

melting point of, 18

refractive index of, 18

separation of, 127

solubility of, 18

table of, 89, 141

James and Stewabt, 142 JoLY and Poole, 133

Keetman, 7

Kernel of atom, 98

Kibchoff, 116

Knipping and Franck, 68

kohlweiler, 116

Kratzer, 126

Kremers, Owens and Balke, 142

Krepelka and Bbaun, 141

,, and RiCHABDS, 141

Krypton, 70

,, anomalous atomic weight of, 114

Landaueb and Wendt, 70 Langmuib, 95, 96, 99 Lead, atomic weight of, 16

,, from carnotite, 124

,, from thorite, 17 

     isotopes  of,  14,  15


INDEX


15)


Lembert and Richards, 17, 121 Lewis-Langmuir atom, 95 LmDEMANN, 102, 124, 134, 139

,, and Aston, 131

Lines of first and second order, 61, 76 

     of  reference,  55,  64

Lithium, 86, 97, 146 LooMis, 125, 126

LUDLAM, 129

McAxpiNE and Willard, 142

Magnesimn, 80

Magnetic field of mass-spectrograph,

51 Marckwald, 7, 8 Mass, change of, 100 

     deduced  from  parabolas,  28

    deduced  from  mass -spectrum,

55 Mass-spectrograph, 43 Mass-spectrum, 47, 54 Measurement of lines on mass-

spectrum, 59 Meitner, 21

,, and Hahn, 8 Melting point of isotopes, 18 Mercury, 72, 80 

 parabolas  of,  30

        separation  of  the  isotopes

of, 134 Merton, 121, 123, 124, 125 Mesothorium, 8, 10 Meta-elements, 4

Metallic elements, mass-spectra of, 80 Meteoric nickel, 113 MiCHiELSON and Bruylants, 142 Microbalance for density, 35 MiLLIKAN, 22, 91

Molecular lines of second order, 75 Moles and Batuecas, 141 MOSELEY, 11, 93, 115 Mtjller, 142 Multiply charged rays, 30

Natural numbers and atomic weights,

111 Negatively charged rays, 29, 62 Negative mass-spectra, 62, 66 Neon, 1, 33, 64, 97 Neuberger, 21 Nickel, 79 

     meteoric,  113

Nitrogen, 67, 110 Nomenclature of isotopes, 61 Nucleus atom, 10, 92, 97, 125 

       structure  of,  101

Ntjttall and Geiger, 10, 13

Order, lines of first and second, 61 Owens, Balke and Kremers, 142 Oxygen, 63


Packing effect, 100 Paneth and Hevesy, 11 Parabola method of analysis, 25 Parsons and Baxter, 113 Perforated electrodes, 22, 24 Periodic law, 11, 12, 34 

       table  of  the  elements,  144,

145 Period of radio-elements, 13 Perrin, 104 Phosphonas, 77

Photochemical separation, 137 Photographic plates for positive rays,

25 Planck's quantum, 95 Planetary electrons, 92 Poole, 133 

     and  JoLY,  133

Positive ray paraljolas, 28 

       rays,  22

     separation    by,    136

Potassium, 87 Pressure diffusion, 131 Proton, the, 92 Protyle, 90, 118 Prout's hypothesis, 2, 90, 100


Radioactive isotopes, 7, 14 

       classification  of,

21 

 transformations,  13,  14,

15 Radium B and lead, 11 

       D  and  lead,  11

Ramsay, 115 

        and  Collie,  39
        and  Travers,  33

Ratner, 24 Rayleigh, 127 Reference lines, 55, 64 Refractive index of isotopes, 18 Reichenheim and Gehrcke, 80, 83,

88 Renz, 139

Resolving power of mass-spectro- graph, 60 Richards 17 

        and  Krepelka,  141
        and  Lembert,   17,   121
        and  Wads  WORTH,  17

Richardson, 85 Rossi and Russell, 9, 120 Rubidium, 87 Russell, U 

       and  Rossi,  9,  120

Rutherford, Sir E., 7, 9, 13, 92, 93, 102 

 and  Chadwick,  103

 and  Andrade,  11

Rydberg, 141


162


INDEX


SCHUTZENBERGER, 3

Screens, willemite, 25

Secondary rays, 29

Second order, lines of the, 61

Selenium, 77

Separation of isotopes, 127

Silicon, 72 

      fluoride,  74

Skaupy, 139

Slit system of mass-spectrograph, 49 Smith and Van Haagen, 72 SoDDY, 6, 8, 10, 11, 12, 13, 14, 16, 17, 35 

      and  Hyman,  17,  121

Sodium, 86 Solubility of isotopes, 18

SOMMERFEIiD, 95

Spectra of isotopes, 9, 121,

Spectrum lines, form of, 53

Spencer, 91

Starkweather and Baxter, 141

Stas, 91

Statistical relation of isotopes, 109

Stewart, 11, 12 

        and  James,  142

Sulphur, 76

Tani, Baxter and Chapin, 142 Tellurium, 77 Thermal diffusion, 129 Third order line of argon, 67 

      lines  of,  61

Thomson, G. P., 86, 88

Sir J. J., 1, 22, 29, 33, 62, 70, 72, 75, 84, 91, 129 Thorite, 17, 18 Thorium, 7, 9, 14, 15, 18, 120


Thorium, chain, 17, 18, 116

,, atomic weight of, 18

Tin, 78 Travers, 39 

       and  Ramsay,  33

Triatomic hydrogen, 70

Unitary theory of matter, 90 Uranium, 10, 120 ,, chain, 15

Valency electrons, 98

Van Haagen and Smith, 72

Wadsworth and Richards, 17 Watson, 33 

       and  Aston,  24,  35

Weatherell, Baxter and Holmes,

73, 141 Welsbach, 8

Wendt and Landaueb, 70 Wey and Classen, 142 Whole number rule, 90 WiEN, 22

WiLLARD and McAlpine, 142 Willemite screens, 25 Wilson and Baxter, 142 

       and  Harkins,  116

Xenon, 70

anomalous atomic weight of, 114

X-ray spectra of isotopes, 1 1

Zechmeister and Hevesy, 20 Zinc, 147

Retrieved from "https://msterms.org/wiki/index.php?title=Aston_1922&oldid=21481"