496 
Soluble  Salts  of  Caffeine. 
f  Am.  Jour.  Pharm. 
t       Oct.,  1882. 
nite  and  have  been  described ;  but  they  are  not  stable,  and  water 
decomposes  them  into  free  acid  and  caffeine,  which  precipitates  or  not 
according  to  the  quantity  of  water  employed.  They  are  also  altered 
by  the  air,  and  the  crystals  of  the  hydrochlorate  when  exposed  efflo- 
resce rapidly,  losing  their  acid.  After  a  time,  more  or  less  long, 
according  to  the  temperature,  nothing- remains  but  caffeine.  There- 
fore, from  this  point  of  view,  the  salts  of  caffeine  formed  with  mineral 
acids  do  not  present  any  advantage  over  pure  caffeine. 
In  presence  of  these  negative  results,  I  had  the  idea  of  employing 
the  natural  salt  which  holds  the  caffeine  in  combination  in  coffee,  the 
chlorogenate  of  potash  and  caffeine  of  Payen.  But  that  salt  is  very 
difficult  to  obtain  in  any  considerable  quantity,  and  this  presents  an 
obstacle  to  its  employment  in  the  ordinary  way,  besides  which  when 
in  solution  the  salt  is  very  alterable  in  the  au\  Moreover  it  contains 
only  29  per  cent,  of  caffeine,  and  in  the  cold  requires  to  dissolve  it 
several  times  its  weight  of  water,  which  gives  solutions  still  less 
charged  with  caffeine. 
It  occurred  to  me  then,  in  view  of  the  apparent  similarity  that  exists 
between  chlorogenic  or  caffetannic  acid  and  benzoic,  cinnamic  and  sali- 
cylic acids,  that  double  salts  analogous  to  Payen's  salt  could  be  formed 
with  the  three  last-named  acids,  and  my  conjecture  proved  correct. 
In  the  presence  of  benzoate,  cinnamate  or  salicylate  of  soda  caffeine 
dissolves  in  a  very  small  quantity  of  water  and  forms  thus  very  solu- 
ble double  salts  that  are  very  rich  in  caffeine.  That  there  is  in  this 
case  more  than  a  simple  solution  is  proved  by  the  fact  that  for  a  defi- 
nite weight  of  caffeine  there  is  required  a  definite  weight  of  the  alka- 
line salt;  in  other  words,  combination  takes  place  according  to  the 
equivalents  of  these  various  bodies.  But  this  combination  is  rather 
weak,  since  if  the  salts  be  treated  with  chloroform  they  give  up  to  it 
all  their  caffeine.  In  my  opinion  these  double  salts  may  be  best  com- 
pared with  hydrated  salts,  which  lose  their  water  of  crystallization 
when  treated  with  a  neutral  liquid  in  which  water  is  soluble.  In  this 
way  blue  sulphate  of  copper  loses  its  water  of  crystallization  and 
becomes  white  w^hen  immersed  in  absolute  alcohol. 
Cinnamate  of  soda  dissolves  caffeine  in  w^ater  equivalent  for  equiva- 
lent, 170  parts  of  cinnamate  for  244  of  caffeine.  The  double  salt  thus 
obtained  contains  58*9  per  cent,  of  caffeine. 
Benzoate  of  soda  and  caffeine  contains  for  two  equivalents  of  ben- 
