Improvements in the Synthesis Procedure
amounts of K2CO3 and Cu2O used were
decreased significantly, resulting in a more manageable reaction.
reaction temperature window has been accurately defined as between 155* C.
and 165* C; the reaction time is estimated at about 12 hours. Higher
temperatures (170*C) result in much shorter reaction time, cleaner
reaction, but degrade rapidly if carried too long.
from one reaction (containing 2-chloropyridine and toluene) can be
recycled and used in subsequent reaction(s).
single most major improvement made was the use of the HCl salt of the
amine starting material; isolating the freebase is not necessary and this
step can be eliminated. The reaction procedes slowly and smoothly to
completion in 3-4 hrs. at 35-40*C. External
heating is not necessary for large scale runs; the mild exotherm is
sufficient to drive the reaction.
- Use of
less borohydride; only 1.5-1.6 mole equivalent of NaBH(OAc)3
has been found to be required for complete reaction.
method. It was found that a single alkaline quench/workup wash is more
convenient than multiple extractions and acid/base neutralization of the
original method; also, K2CO3 works much better than
NaHCO3 used in the previous method. K2CO3
is calculated based on NaBH(OAc)3; the
final pH of the workup is ~9-9.5 and no product is in the aqueous phase.
Cold trituration in MTBE or acetonitrile affords an easy and convenient
method of purification; MTBE appears to work a little better. Essentially
all the product is recovered after the 3rd crop; wet chemical
manipulation involved removing some lipos with hot heptane and
triturations of the residue in MTBE and acetonitrile can recover most of
should be the limiting reagent; any excess remaining, or it’s reduction product alcohol, make workup and
isolation of final product (recrystallization or trituration) difficult.
(This was the primary difficulty encountered in the 1st
reaction I ran). Excess amine remaining is readily removed by trituration
single best improvement I made for this step was the use of 100%
acetonitrile for recrystallization. The crude product is a clear dark oil; a fine white powder is obtained from
acetonitrile. The technique is a bit tricky; the compound has a tendency
to form a supersaturated oil, especially after warming
to solution. The pure compound is very poorly soluble in cold
acetonitrile. Cold solution / trituration works
best; several crops can be obtained by subsequent evaporations of the
solvent, and essentially all of it can be recovered.
I have been using a slightly different workup method, which may be more efficient and will lessen
the chance of any peroxide in the final product.
For the production run, the reaction solution (mostly
methanol) was quenched with aqueous carbonate, most of the product oiled out. The decant was then stripped of most methanol and extracted
for additional product.
What I did was to strip the methanol 1st (20-30*C
MAX), decant from the oiled out product, then add the carbonate solution to
quench residual peroxide, then extract the decant with
EtOAc to recover all product. No aqueous phases contained any product. The
combined EtOAc solutions then washed twice with brine, and stripped to residue.
All residual peroxide (there really shouldn’t be any at this point, and a
simple check with starch-iodine paper confirms) will remain in the aqueous, not
significant improvement made in this step is the use of 100% acetone for
recrystallization. It works exceptionally well.