Am 'ji0ne!'i9i5arm' }   Determination  of  Hydrocyanic  Acid.  267 
sometimes  heat  is  applied  after  adding  the  acid.  Others  recom- 
mended that  the  mixture  be  heated,  both  before  and  after  the  addi- 
tion of  acid.  In  testing  by  our  method  heat  is  not  necessary  if 
potassium  fluoride  is  used.  In  using  sodium  chloride  instead  of 
potassium  fluoride,  or  in  the  absence  of  these  salts,  gentle  heat 
hastens  the  formation  of  Prussian  blue. 
Description  of  the  Method. 
As  a  result  of  our  various  experiments,  we  recommend  the 
following  procedure :  Before  concentrating  the  hydrocyanic  acid 
solution,  as  in  the  case  of  a  distillate,  the  portion  to  be  tested  should 
contain  a  slight  excess  of  free  sodium  hydroxide.  We  used  0.02  to 
0.1  g.  This  solution  is  then  concentrated  in  a  round  bottom  flask 
of  200  Cc.  capacity  by  using  a  vacuum  pump  and  condenser.  The 
heat  is  supplied  by  immersing  the  flask  in  a  water-bath  kept  below 
700.  To  avoid  any  loss  by  spattering,  the  flask  is  fitted  to  the  con- 
denser by  means  of  an  adapter  such  as  is  used  in  the  Kjeldahl  method 
for  the  determination  of  nitrogen.  We  concentrate  until  less  than 
1  Cc.  of  liquid  remains  in  the  flask.  Two-tenths  to  one-half  cubic 
centimetre  of  3  per  cent,  freshly-prepared  ferrous  sulphate  solution 
and  about  0.05  g.  of  potassium  fluoride  are  then  added.  The  flask 
is  exhausted  at  once  by  means  of  a  water  vacuum  pump.  The  con- 
tents are  mixed  by  rotating  the  flask.  After  five  to  ten  minutes  the 
flask  is  detached  from  the  pump  and  the  mixture  acidified  with  30  per 
cent,  nitric  acid.  The  blue  color  appears  at  once.  Where  only  traces 
of  hydrocyanic  acid  are  present  it  is  sometimes  necessary  to  warm 
to  about  500  in  a  water-bath  before  the  color  appears.  The  suspen- 
sion is  then  diluted  to  a  volume  that  would  give  a  color  density  con- 
venient to  compare  with  a  suspension  of  Prussian  blue  made  from 
a  known  weight  of  potassium  cyanide.  As  a  standard  we  used  a  sus- 
pension of  Prussian  blue  made  from  1  mg.  of  potassium  cyanide. 
Such  a  suspension  diluted  to  25  Cc.  gave  a  color  of  convenient 
density.    For  comparison  we  used  a  Duboscq  colorimeter. 
If  the  cyanide  solution  to  be  tested  was  sufficiently  concen- 
trated so  that  further  evaporation  was  unnecessary,  the  test  could 
be  made  in  a  test-tube.  We  kept  the  air  out  by  means  of  a  stopper 
and  rotated  the  tube  only  enough  to  mix  the  reagents,  allowing  the 
mixture  to  stand  five  to  ten  minutes  before  acidifying.  Much  shaking 
must  be  avoided  to  prevent  excessive  oxidation  of  the  ferrous 
hydroxide. 
