Note: Descriptions are shown in the official language in which they were submitted.
PROCESS FOR THE PRE'P~RATION OF 2- ~N--(2--
flYDROXYETHYL)-N~LOWER ALKYLAMINOMETHYL]-
BENZHYDROLS
The invention relates to a process for the
production of 2~[N-(2-hydroxyethyl)-N-lower alkyl~ninomethyl]-
benzhydrols (I). The benzhydrol products of this process can
be converted directly to the corresponding phenylbenz(f)-2,5-
oxa~ocines which are valuable physiologically active substances.
Thus see Vnited States Patent 3,830,803, United States Patent
3,978,085, British Patent 1,148,717 and Canadian Patent
863,349.
The direct preparation of compounds of type I
below from compounds of type II has been known heretofore.
This known reaction is carried out by treatment of the start-
ing compound with a metal hydride reducing agent, such as
lithium aluminium hydride, in an inert organic solvent such as
tetrahydrofuran or ether (e.g. see British Patent 1,148,717).
Reduction with lithium aluminium hydride is not J however, a
practical synthetic method for large scale reactions on a com-
mercial basis. The solvents which must be used are fire
hazards, lithium aluminium hydride itself is extremely hazard-
ous, and the costs are prohibitively high.
The present method, on the other hand, can be
commercially feasible and advantageous. It uses less hazard-
ous solvents, can be comparatively less expensive and can pro-
duce high yields of product of good purity.
This invent~on provides a process for the prep-
aration of a 2-[N-(2-hydroxyethyl)-N-lower alkylaminomethyl]-
benzhydrol of the formula:
-- 1 --
~7S5~
~Xn
m ~ ~ I
CH2N (CH2) 20E~
wherein R is methyl or ethyl, X i5 fluorine, chlorine, bromine
or methyl, Y is fluorine, chlorine, methyl or methoxy, and m
and n are independently zero, one or two which comprises re-
ducing a compound of the formula
~Xn
y ~ III
C-N(CH2)2D
0 R
where D is halogen in an inert solvent with sodium borohydride
in the presence of an alkanoic acid. Preferably the compound
of formula III is obtained by treating a compound of the for-
mula
n
II
C-N(CH2)20H
0 R
with a halogenating agent in an inert 501vent.
- The reducti~n of certain amides to amines with
sodium borohydride in the presence of alkanoic acid has been
described in "Sodium Acyloxyborohydride as New Reducing Agents",
Tetrahedron Letters No. 10, pages 763-766, March 1976, and
"Reduction of Amides with Sodium Borohydride", Ventron
Alembic, Issue No. 9,
.. . . . _
pages 6 and 7, but this appears to be di~ferent from our re
duction of -the halogenated compound of formula III. Thus the
prior described reductions recluire a larye excess of sodium
borohydride, with loss of efficiency due to evolution of H2
and probable formation of amine-borane, the reduc~ion allegedly
proceeding by way of the alkyloxyborohydride. The reduction
of the present invention, on the other hand, proceeds readily
in a wide rangè of solvents with a much smaller excess of
sodium borohydride; the presence of the halogen D is essential,
the reaction failing in its absence, and in the preferred pro-
cedure, in which the formula III compound and alkanoic acid
are added gradually to the reaction mixture containing all of
the sodium borohydride, there is no evolution of H2 after the
first small acid addition.
The product I is ordinarily recovered from the
reaction mixture by means of an aqueous workup in solution in
the inert solvent. It can be obtained as a pure substance by
removal of the solvent by conventional methods or the solution
can be used as such, for example in preparing the physiologi-
cally active phenylbenz(f)-~,5~oxazocines.
The starting materials II of the process of the
invention are generally known to the art (see the British and
Canadian patents referred to hereinabove). The first (halo-
genation) step of the process is carried out in an inert sol-
vent, such as a chlorinated hydrocarbon, for example dichloro-
methane or dichloroethane or an aromatic hydrocarbon such as
toluene, benzene or the like. The preferred solvent is di-
chloroethane. Suitable halogenating agents include phosphorus
trichloride, phosphorus tribromide, phosphorus pentachloride
and thionyl chloride. Phosphorus trichloride is presently
preferred due to its relative]y low cost. Generally an equiv-
alent amount or a small excess of up to about 10 percent of
halogenating agent is used.
The temperature required is moderate, e.g. from
20 to 90 C., depending upon the time constraints. At 55 to
80 C., -the reaction is complete in from about one to about
four hours. Excessive reaction temperatures and reaction times
should be avoided to minimiæe the possibility of side reactions.
Completion of the reaction is generally monitored chromato~
graphically. Preferably the reaction mixture is cooled,
neutralized with aqueous base, then separated and dried.
The reduction reaction can be easily and eco-
nomically carried out and produce high yields (e.g. from 80 to
lOQ percent of the theoretical amount). Unlike the prior art
process which uses the powerful reducing agent lithium alu-
minium hydride and difflcult and dangerous reaction conditions,
the reduction step of the present process can be easily con-
trolled even when carried out on a large scale.
The reaction is carried out in the presence of
an alkanoic acid, preferably a lower alkenoic acid containing
not more than four carbon atoms such as acetic acid. When the
acid is omitted, little or none of the desired product is ob-
tained; The amount of alkanoic acid used may for example be
about 0.1 to about 1.0 mole per mole of sodium borohydride.
The reduction is suitably carried out in an inert aliphatic
chlorinated hydrocarbon solvent which has a reflux temperature
in excess of 60 C., preferably dichloroethane. The reaction
temperature is preferably the reflux temperature of the re-
action mixture. It can be dangerous to start the reaction be-
low a temperature of 60 C.
~7~
The reduction ~an be carried ou~ by addingacetic acid to A mixture of sodium borohydride and the pre-
cursor N-(2-haloethyl)-~-lower alkyl-o-henzoylbenzamide (III~
in dichloroethane This reaction is normally quite exotherrnic,
although controllable. Alternatively (and preferably for large
scale reactions) the reaction is hegun on a relatively small
scale with all reaction components bein~ present. This
initial reaction mixture/ which includes ~he entire amount of
sodium borohydride to be used in the lar~r scale reaction, is
maintained at reflux. The balance of the precursor III in di-
chloroethane together with the alkanoic acid is added gradually.
Using this alternate procedure, less alkanoic acid is necessary
(for example less than 0.2 mol~ per mole of sodium boro-
hydride). In addition, less ~dium borohydride may also he
used; for example, it is presently preferred to use about 1.5
moles per mole of III compared to 2.0 moles per mole of I~I
when not using the preferred proc~dure. The reaction can be
carried out under an inert (e.g. nitrogen) atmosphere as a
safety measure and also to minimize the possibility of side
reactions, although this is generally unnecessary.
After the reduction has proceeded to completion
(as conveniently shown by chromatographic analysis), the mix-
ture is added gradually (cautiously at first) to water, then
basified and heated at reflux (to destroy any residual sodium
borohydride and neutralize the alkanoic acid). The product
can then be isolated or alternatively the organic layer con-
taining the product can be separated and used for further re-
action to provide the pharmaceutically active phenylbenz(f)-
2,5-oxazocines.
The following illustrative Examples are pro-
5 "
vided to show the practice of the process of the invent:ion.
Example 2 lllustrates just the preparation of a startiny com-
pound III.
E~ample 1
A complete five-step synthetlc sequence begin-
ning with commercially available materials and ending with a
pharmaceutically active phenylbenz(f)-2,5-oxazocine.
Steps 3 and 4 of this example illustrate a pre-
ferred process of the present invention as part of the sequence.
Step 1 - Acid Chloride Preparation.
To a slurry of ortho-benzoylbenzoic acid (222.2
g., 1.0 mole~ in dichloroethane (230 ml.) was added in one
portion phosphorus trichloride (35.4 ml., 0.46 mole). After
one hour of stirring, the temperature had reached a maximum of
39 C. Stirring at room temperature was continued overnight
after which thin layer chromatographic analysis showed complete
conversion to the acid chloride. The product layer was sepa-
rated by decantation.
Step 2 - Amide Formation.
To a solution of triethylamine (111,39 g., 1.1
mole) and N-methyl ethanolamine (82.62 g.~ 1.1 mole~ in di-
chloroethane (400 ml.) was added dropwise the acid chloride-
dichloroethane solution of step 1 over one hour at 5 to 12 C.
After the addition was complete, the reaction mixture was
stirred for an additional hour, by which time thin layer
chromatographic analysis showed complete conversion to N-(2-
hydroxyethyl)-N-methyl-_-benzoylbenzamide.
Step 3 - Preparation of N-(2-chloroethyl)-N-
methyl-_-benzoylbenæamide.
To a slurry of product obtained in step 2 at
~ 3 5~
room temperature was added phosphorus trichloride (35.4 ml.,
0.406 mole) over five minu-tes. This caused a 20 C. tempera-
ture rise, and -the slurry thinned considerably. After warming
to 55 to 60D C., and maintaininc3 this temperature for one hour,
thin layer chromatographic analysis showed complete conversion
to the desired product. The reaction mixture was cooled to
0 C., and diluted with water (500 ml.) over five minutes
maintaining the temperature below 5 C. The two-phase mixture
was stirred for five minutes and the layers allowed to separate.
The bottom organic layer was washed with an additional 500 ml.
of water and sufficient base (sodium hydroxide~ to raise the
pH to about 7 at below 10 C., and the bottom layer was dried
over anhydrous sodium sulfate (90 g.).
Step 4 - Reduction~
To a slurry of a portion of the solution from
step 3 (108 ml.) and sodium borohydride (28.4 g., 0.75 mole)
was added dropwise acetic acid (3.5 ml.). After rothing and
heat evolution had ceased, the temperature was raised to re-
flux and maintained for 15 minutes. To this refluxing slurry
was added dropwise with stirring over two hours the remainder
of the solution of product from step 3 containing 9 ml. of
acetic acid. The temperature was maintained at reflux by
slight warming of the flask, the addition itself heing fairly
exothermic. After the completion of addition, the reaction
mixture was stirred at reflux for a further hour, at which
time thin layer chromatographic analysis showed complete con-
version to the desired product. The whole of this procedure
was carried out under an atmosphere of nitrogen.
To the above slurry at room temperature was
added dropwise over one hour (very slowly initially) 200 ml.
~ - 7
of water. The addition caused fairly heavy effervescence,
and the temperature rose to 45 C. To the above diluted re-
action mixture was added 40 percent aqueous sodium hydroxide
(75 ml.). The temperature was raised to reflux and maintained
for three quarters o~ an hour. The mixture was then cooled
and the layers separated to provide a solution of 2-[N-(2-
hydroxyethyl)-N-methylaminomethyl]benzhydrol in dichloroethane.
Step 5 - Cyclization.
The benzhydrol from step 4 (still in solution
in dichloroethane) was cyclized with aqueous hydro~romic acid
using the method of U.S. Patent 3~9-78,085. The reaction yielded
131.5 g. of 5-methyl-1-phenyl-1,3,4,6-tetrahydro-5~-benz(f)-
2,5-oxazocine hydrobromide, m.p. 259-261 C. The melting point,
thin layer chromatographic analysis and infrared spectral
analysis all indicate a pure product.
The overall yield of the five steps (beginning
with orthobenzoylbenzoic acid) was 60-65 percent of theoretical.
Example 2
A mixture of N-(2-hydroxyethyl)-N-methyl-_-
20 benzoylbenzamide (56.66 kg., 200 moles), toluene (10 liters)
and phosphorus trichloride (10~07 kg., 10 percent excess~ was
warmed gently until an exothermic reaction ensued with forma
tion of a hot melt at 77 C. After 0.5 hour at 75 to 77 C.,
thin layer chromatographic analysis indicated complete forma~
tion of -the N-(2-chloroethyl)-N-methyl-_-benzoylbenzamide.
Isopropanol (90 liters) was quickly added and the batch rapid-
ly cooled, seeded and chilled to below 0 C. Product was
collected, washed with cold isopropanol (2 x 10 liters) and
dried to give 50.6 kg. (83.3 percent) of product, m.p. 85.5-
30 86.2 C.
s~
_xample 3
A solution of ~44 kg. of N-(2-chloroethyl)-N-
methyl-_-benzoylbenzamide (~81 moles) in 378 kg. of dichloro-
ethane was prepared. A mixture of 50 kg. of this solution,
sodium borohydride (273 kg., 721.5 moles, 1.5 equivalents) and
dichloroethane ~120 litres) at 46 C., was treated with ace~ic
acid (0.5 kg.). The temperature rose steadily to reflux at
86 C. The remaining dichloroethane solution was added slowly,
the exotherm maintaining the mixture at reflux wikhout applied
heat during the addition and for two hours after addition was
complete. Thin layer chromatographic analysis was used to
check for the completeness of reaction (had the reaction been
found to be incomplete, a small furt~er addition of sodium
borohydride could then have been made). The complex was then
decomposed by cautious slow addition of 100 litres of water
at the reflux temperature and then 50 litres of 40% (w/w)
aqueous sodium hydroxide solution to form 2-~N-(2-hydroxy-
ethyl)-N-methylaminomethyl~benæhydrol in dichloroethane. The
benzhydrol in dichloroethane was water washed to pH 8 and
subsequently cyclized to 5-methyl-1-phenyl-1,3,4,6-tetrahydro-
5H-benz(f)-2,5-oxazocine as follows:
The organic phase was separated and cooled to
20 C.; 48-50% (224 kg., 149.2 litres) was added with stirring
over a half hour period, with cooling to keep the temperature
below 50 C. The mixture was gently warmed to reflux, re-
fluxed for 2-1/2 hours, cooled to room temperature and then
chilled to 0 C. The product was collected by centrifuging,
spun dry, washed with a total of 110 litres acetone and spun
dry again. The damp product was returned to a clean vessel
previously charged with toluene (242.8 litres), tap water
(110.4 litres) and sodium hydroxide pearl (20.4 kg.), and the
mixture stirred and heated to 60 C., stirring being continued
for 1/2 hour at 60 C. beEore standing to separate for 1/2
hour. The lower aqueous alkaLine layer was run off and the
toluene layer washed with 2 x 56 litres of tap water at 60 to
70 C. to pH 7. Toluene was gtripped o~f under reduced pres-
sure and the residual oll cooled to room tempera-ture and then
diluted with acetone (220 litres). The solution was clarified
by passage via a cartridge filter to a clean vessel. To the
stirred, water cooled, clear acetone solution was slow]y added
hydrochloric acid (38.02 kg., 31.88 litres) over a 1/2 - 1
hour period (to pH l); the resulting slurry was stirred and
cooled to room temperature, then chilled to 0 C. The product
was collected by centrifuge, washed well with acetone (total
of 110 litres) and sucked dry, and then dried in vacuum oven
overnight at 50-60~ C.
Other substituted 2-[N~(2-hydroxyethyl)-N-lower
alkylaminomethyl]benzhydrols which can be prepared using the
process of the present invention as set forth in the preceding
Examples, are shown in Table I.
TABI,E I
Ex. No. Starting Material Product
- ~ C
4 CH3 ~ C ~ O 3 ~ C~OH
CH3 CH3
0~ \ CH2NcH2cH2oH
1 2C 2 ~I CH3
CH3
-- 10 --
75~L
TABLE I (con.tlnued)
Ex. No. Starting M~terial Product
Cl Cl
~Cl ~Cl
C=O CHOH
[~C-NCHH ~
1~ 1 2C 2~ 2 1 2 20H
O CH3 C~I3
Cl Cl
6 C-O HOH
Il t 2 20H 2 1 2CH20H
O CH3 CH3
~3 ~
7 ~ C=O ~c CHOH
C-NcH2cH20H CH2NCH2CH2H
O CH3 CH3
CH CH
~CH3 ~CH3
8 ~ =O ~CHOH
C-NCH2CH2~)H 2 I CH2CH20H
O CH3 CH3
-- 11 --
S~
~rl~BE~E I (corl t :inued )
Ex~ No. Startin~ M~t~l Proclu~t
~)~
C1130~\ -NCH2cH20H C~130 C~12NCH2CH2H
O CH2CH3 C 2CH3
F C
= ~ F CHOH
\[~`~ C -E~CH2CH20H ~ 21 2 20H
O CH3 CH3
--12--
, .~