174 
PURIFICATION  OF  BROMIDE  OF  POTASSIUM. 
THE  PURIFICATION  OF  BROMIDE  OF  POTASSIUM. 
MM.  Robiere  et  Herbelin  have  been  engaged  in  examining  a 
number  of  samples  of  bromide  of  potassium  for  iodine.  The  test 
thej  employ  is  to  place  several  fragments  of  the  bromide,  moist- 
ened with  water,  upon  a  piece  of  glazed  paper,  and  expose  to  a 
trace  of  bromine  vapor.  If  iodine  be  present,  the  paper  acquires 
a  blue  tint.  The  bromine  vapor  is  poured  from  a  little  flask 
filled  with  asbestos,  wetted  with  bromine  water.  When  the 
quantity  of  iodine  is  great,  the  blue  tint  may  be  partially  or 
completely  masked  by  the  brown  tint  of  free  iodine.  To  avoid 
this,  the  test  may  be  modified  as  follows.  A  crystal  of  the  bro- 
mide is  pulverized  and  put  in  a  watch-glass  standing  on  a  plate. 
A  few  drops  of  bromine  are  poured  on  the  plate  and  the  whole 
covered  with  a  glass.  The  bromide  is  unchanged,  if  it  is  pure  ; 
or  if  it  takes  a  slight  yellow  tint,  it  loses  it  very  rapidly  in  the 
air.  If  it  contains  a  sensible  quantity  of  iodine,  it  becomes  im- 
mediately brown,  the  iodine  being  displaced  by  the  vapor  of 
bromine.  If  it  contains  only  minute  traces  of  iodine  insufficient 
to  give  a  visible  coloration,  the  salt  is  transferred  to  a  corked 
tube  and  agitated  with  benzol.  The  liquid  immediately  assumes 
a  rose  tint  easy  to  recognize. 
To  purify  bromide  of  potassium  completely  from  iodine,  the 
the  authors  recommend  to  dissolve  the  salt  in  a  small  quantity 
of  water,  and  then  add,  little  by  little,  bromine  water  to  the 
solution,  heated  to  boiling,  until  it  is  present  in  excess.  The 
liquid  is  then  boiled  and  evaporated  to  dryness  with  constant 
stirring.  The  bromine  is  thereby  perfectly  deprived  of  iodine. 
It  is  then  crystallized. 
It  has  been  since  pointed  out  that,  in  the  foregoing  tests  for 
detecting  traces  of  iodine,  the  results  may  in  every  case  be 
masked  by  the  presence  of  any  excess  of  free  bromine.  M. 
Duingt  writes  that  if  we  introduce  a  little  chloroform  (or  benzol 
or  sulphide  of  carbon)  into  a  solution  of  bromide  of  potassium 
containing  yo^oo^^  V^^^  iodide,  and  then  add  drop  by  drop, 
shaking  after  each  addition,  some  dilute  bromine  water,  we  see 
the  chloroform  at  first  assume  a  violet  tint,  become  decolorized 
by  the  next  addition,  and  ultimately  take  a  yellow  color  when 
