﻿Determination 
  of 
  Potassium 
  Spectroscopically. 
  

  

  455 
  

  

  Experiment 
  III. 
  

  

  test-solution. 
  

  

  compared 
  with 
  standard. 
  

  

  25 
  cm 
  3 
  

  

  stronger 
  

  

  50 
  " 
  

  

  stronger 
  

  

  80 
  " 
  

  

  stronger 
  

  

  90 
  " 
  

  

  stronger 
  

  

  95 
  " 
  

  

  stronger 
  

  

  105 
  " 
  

  

  stronger 
  

  

  120 
  " 
  

  

  like 
  

  

  (120 
  X 
  0-0001 
  =0 
  0120) 
  

   Potassium 
  found 
  

  

  

  0*0120 
  grm. 
  

  

  " 
  taken 
  

  

  _ 
  0*0100 
  " 
  

  

  Error 
  

  

  .- 
  0-0020 
  " 
  

  

  The 
  result 
  of 
  this 
  experiment 
  was 
  most 
  surprising 
  ; 
  for, 
  in- 
  

   stead 
  of 
  diminishing 
  the 
  delicacy 
  of 
  the 
  test 
  we 
  find 
  that 
  the 
  

   presence 
  of 
  a 
  moderate 
  amount 
  of 
  sodium 
  chloride 
  tends 
  to 
  

   increase 
  the 
  "brilliance 
  of 
  the 
  potassium 
  line. 
  The 
  sodium 
  

   chloride 
  employed 
  was 
  a 
  part 
  of 
  that 
  prepared 
  and 
  tested 
  as 
  

   previously 
  described 
  and 
  used 
  in 
  the 
  experiments 
  upon 
  the 
  

   qualitative 
  determination 
  of 
  potassium. 
  By 
  no 
  possibility 
  

   could 
  the 
  0'1 
  grm. 
  of 
  it 
  taken 
  in 
  the 
  experiment 
  have 
  con- 
  

   tained 
  more 
  than 
  0*000001 
  grm. 
  of 
  potassium. 
  It 
  is 
  evident, 
  

   therefore, 
  that 
  the 
  brilliance 
  of 
  the 
  potassium 
  line 
  gained 
  

   twenty 
  per 
  cent 
  in 
  strength 
  by 
  the 
  influence 
  of 
  sodium 
  

   chloride 
  amounting 
  to 
  ten 
  times 
  the 
  weight 
  of 
  the 
  potassium 
  

   present 
  when 
  the 
  effect 
  due 
  to 
  impurity 
  of 
  the 
  salt 
  could 
  by 
  

   no 
  means 
  exceed 
  a 
  hundredth 
  of 
  one 
  per 
  cent 
  ; 
  that 
  is 
  to 
  say, 
  

   the 
  observed 
  effect 
  is, 
  at 
  the 
  very 
  least, 
  two 
  thousand 
  times 
  

   greater 
  than 
  that 
  which 
  might 
  have 
  been 
  conceivably 
  produced 
  

   by 
  contamination 
  of 
  the 
  sodium 
  salt. 
  

  

  In 
  the 
  following 
  experiment 
  the 
  effect 
  of 
  varying 
  amounts 
  of 
  

   sodium 
  chloride 
  upon 
  the 
  spectrum 
  of 
  the 
  potassium 
  is 
  shown. 
  

   The 
  sodium 
  line 
  was 
  turned 
  out 
  of 
  the 
  field 
  of 
  view 
  to 
  obviate 
  

   the 
  dazzling 
  effect 
  of 
  the 
  sodium 
  light, 
  and 
  a 
  solution 
  of 
  potas- 
  

   sium 
  chloride 
  containing 
  0*01 
  grm. 
  of 
  the 
  element 
  in 
  100 
  cm 
  3 
  , 
  

   was 
  examined 
  spectroscopically 
  after 
  the 
  addition 
  of 
  successive- 
  

   ly 
  increasing 
  amounts 
  of 
  sodium 
  chloride, 
  the 
  strength 
  of 
  the 
  

   line 
  observed 
  being 
  brought 
  into 
  comparison 
  with 
  that 
  produced 
  

   by 
  similar 
  portions 
  of 
  the 
  potassium 
  solution 
  containing 
  no 
  

   sodium. 
  

  

  

  

  

  Characteristic 
  of 
  

  

  

  

  

  line 
  compared 
  

  

  Sodium 
  chloride 
  

  

  Potassium 
  

  

  Ratio 
  of 
  Width 
  

  

  with 
  standard 
  con- 
  

  

  in 
  a 
  coil-full. 
  

  

  in 
  a 
  coil-full. 
  

  

  NaCl 
  : 
  K 
  of 
  slit. 
  

  

  taining- 
  no 
  NaCl. 
  

  

  0-002 
  mg. 
  

  

  0-002 
  mg. 
  

  

  1 
  : 
  1 
  018 
  mm. 
  

  

  like. 
  

  

  o-oio 
  " 
  

  

  0-002 
  " 
  

  

  5:1 
  0-18 
  " 
  

  

  like. 
  

  

  0-020 
  " 
  

  

  0-002 
  " 
  

  

  10:1 
  0-18 
  " 
  

  

  a 
  little 
  stronger. 
  

  

  0-040 
  " 
  

  

  0-002 
  " 
  

  

  20:1 
  0-2*3 
  " 
  

  

  much 
  stronger. 
  

  

  0-200 
  " 
  

  

  0-002 
  " 
  

  

  100:1 
  0-23 
  " 
  

  

  very 
  much 
  stronger. 
  

  

  0-400 
  " 
  

  

  0-002 
  " 
  

  

  200:1 
  0-23 
  " 
  

  

  much 
  stronger. 
  

  

  0-600 
  " 
  

  

  0-002 
  " 
  

  

  300:1 
  023 
  " 
  

  

  much 
  stronger. 
  

  

  Am. 
  Jour. 
  Sol- 
  

   31 
  

  

  —Third 
  Series, 
  Vol. 
  XLII, 
  No. 
  252. 
  

  

  — 
  December, 
  1891. 
  

  

  