Synthesis of 1-tert-butyl-4-chloropiperidine: Generation of an N-tert-butyl group by the reaction of a dimethyliminium salt with methylmagnesium chloride

Article abstract of DOI:10.1021/jo048047g

(Chemical Equation Presented) Two efficient routes to 1-tert-butyl-4- chloropiperidine are described. In the first route, the key thionyl chloride mediated chlorination reaction features the use of tetrabutyl-ammonium chloride as an additive that effectively suppresses the formation of an elimination-derived side product. In the second route, a novel alternative synthesis of 1-tert-butyl-4-chloropiperidine was developed in which the tertiary butyl group on the nitrogen is efficiently generated through the addition of methylmagnesium chloride to a dimethyliminium salt in 71% overall yield.

Full text of DOI:10.1021/jo048047g

SCHEME 1a
Synthesis of
1-tert-Butyl-4-chloropiperidine: Generation
of an N-tert-Butyl Group by the Reaction of
a Dimethyliminium Salt with
Methylmagnesium Chloride
Joseph S. Amato,† John Y. L. Chung,†
Raymond J. Cvetovich,† Xiaoyi Gong,‡
Mark McLaughlin,*,† and Robert A. Reamer†
Departments of Process Research and Analytical Research,
Merck Research Laboratories, P.O. Box 2000,
Rahway, New Jersey 07065
mark_mclaughlin@merck.com
Received November 3, 2004
a
Reagents and conditions: (a) MeI, acetone, rt; (b) t-BuNH2,
acrylic acid, aq NaOH, 60 °C; (c) Ra-Ni/H2 (40 psi), EtOH; (d)
SOCl2, Bu4NCl, toluene, 85 °C.
rium in the conversion of quaternary salt 3 to 1-tert-
butylpiperidone 5. The details of this particular trans-
formation have been previously documented.2 Despite the
significant improvement made to this transamination
reaction, the overall utility of this piperidone-based route
was diminished by certain problems encountered in the
final chlorination step (6 f 1).
Two efficient routes to 1-tert-butyl-4-chloropiperidine are
described. In the first route, the key thionyl chloride medi-
ated chlorination reaction features the use of tetrabutyl-
ammonium chloride as an additive that effectively sup-
presses the formation of an elimination-derived side product.
In the second route, a novel alternative synthesis of 1-tert-
butyl-4-chloropiperidine was developed in which the tertiary
butyl group on the nitrogen is efficiently generated through
the addition of methylmagnesium chloride to a dimeth-
yliminium salt in 71% overall yield.
As shown in Scheme 1, 1-tert-butylpiperidone 5, ob-
tained from the Michael elimination/readdition sequence,
was reduced almost quantitatively using Raney nickel.1
The conversion of 1-tert-butyl-4-hydroxypiperidine 6 into
4-chloro-piperidine 1 was initially investigated using a
variety of reagents. It became apparent that the thionyl
chloride method1 already described in the literature was
the most applicable to the present case both in terms of
cost and minimizing waste. The major issue associated
with this reaction relates to a fundamental principle of
organic chemistry; attempts to carry out substitution at
a secondary aliphatic center will inevitably result in a
varying amount of elimination side-product.3 Indeed, the
literature conditions for this reaction produced olefin 7
at around 20 mol % of the product mixture. Separation
of olefin 7 from the desired product 1 is not straightfor-
ward and this level of impurity was deemed to be
unacceptably large. Assuming that olefin 7 derives from
an E1 elimination process, it was postulated that an
increase in chloride ion concentration in the reaction
mixture would suppress olefin formation by increasing
the probability of simple recombination of the intermedi-
ate cation with chloride to yield chloropiperidine 1. To
test this hypothesis, we examined the use of various
additives in the chlorination reaction, the results of which
are summarized in Table 1.
Recently, we required an efficient synthesis of 1-tert-
butyl-4-chloropiperidine (1), a useful intermediate from
which the 1-tert-butyl-4-piperidinyl Grignard reagent can
be prepared in high yield. Known in the literature since
1965,1 it transpires that the documented methods for
preparing intermediate 1 are less than satisfactory from
the viewpoint of a multi-kilogram scale synthesis. Con-
sequently, we began research toward the goal of develop-
ing a reliable and productive route to 1-tert-butyl-4-
chloropiperidine (1), the results of which are described
herein.
Initial efforts, directed at modification of existing
methodology to render it amenable to large-scale produc-
tion, led to the eventual development of a four-step (four
isolations) synthesis of 1-tert-butyl-4-chloropiperidine (1)
(Scheme 1) starting from piperidone 2. Of note from this
work is the discovery that acrylic acid can function as
an efficient and selective trap for dimethylamine (vs tert-
butylamine 4), allowing for facile control of the equilib-
The general trends evident in Table 1 are in accord
with those predicted. Increased solvent polarity should
(2) Amato, J. S.; Chung, J. Y. L.; Cvetovich, R. J.; Reamer, R. A.;
Zhao, D.; Zhou, G.; Gong, X. Org. Process Res. Dev. 2004, 8, 939.
(3) For a detailed discussion of the various factors influencing
substitution vs elimination see: Smith, M. B.; March, J. March’s
Advanced Organic Chemistry, 5th ed., Wiley-Interscience: New York,
2001; pages 1319-1322 and references therein.
† Department of Process Research.
‡ Department of Analytical Research.
(1) (a) Robinson, J. B.; Thomas, J. J. Chem. Soc. 1965, 2270. See
also: (b) Fankhauser, R.; Grob, C. A.; Krasnobajew, V. Helv. Chim.
Acta 1966, 49, 690.
10.1021/jo048047g CCC: $30.25 © 2005 American Chemical Society
Published on Web 02/01/2005
1930
J. Org. Chem. 2005, 70, 1930-1933

TABLE 1a
SCHEME 2
entry
solvent
Cl- source
mol % Cl-
% olefin 7
SCHEME 3
1
heptane
cyclohexane
toluene
none
0
0
14
12
18
35
58
17
16
17
5
5
9
13
10
46
35
2
none
3
none
0
4
1,2-DME
toluene
none
0
5
none
0
6
toluene
LiCl
100
100
100
100
50
7
toluene
Me4NCl
Et4NCl
Bu4NCl
Bu4NCl
Bu4NCl
Bu4NCl
Bu4NCl
Bu4NCl
LiCl
8
toluene
9
toluene
10
11
12
13
14
15
toluene
is used only for side-product suppression. An obvious
alternative approach to this problem is to use a com-
mercial starting material that already contains the
4-chloro functionality. Taking this into account, a poten-
tially simplifying retrosynthetic disconnection became
apparent (Scheme 2).
toluene
25
toluene
10
cyclohexane
MeCN
25
100
100
1,2-DME
a Reagents and conditions: (a) SOCl, solvent, 80-85 °C, with
or without chloride additive.
To reach the desired 1-tert-butyl-4-chloropiperidine 1
starting from 8, a method of alkylating the secondary
nitrogen with a tert-butyl group was required. Inspection
of the literature not only revealed sparse attempts at the
direct alkylation of nitrogen using tert-butyl halides, but
also showed these to occur in low yield.4 The prospects
for an efficient direct alkylation appeared poor, so a
stepwise approach was adopted. The condensation of
cyclic secondary amine perchlorates (e.g., pyrrolidine and
morpholine) with acetone is known to yield the corre-
sponding dimethyl iminium salts in good yield.5 Also
documented in the literature is the addition of nucleo-
philic reagents, such as Grignards, to certain iminium
species.6 Hence, it appeared reasonable that the addition
of a reagent such as methylmagnesium chloride to a
preformed dimethyliminium salt 9 of 4-chloropiperidine
(Scheme 3) could provide a means of generating a tert-
butyl group on this secondary nitrogen atom. To our
knowledge, there are no previous literature examples in
which an N-tert-butyl group is accessed in this way.7
4-Chloropiperidine hydrochloride 11 is commercially
available but is not readily obtainable in bulk quantities
and, significantly, is rather expensive. Another com-
mercially available and closely related potential starting
material is 4-chloro-1-methylpiperidine hydrochloride 10.
The N-methyl compound 10 is approximately eight times
less expensive than 4-chloropiperidine hydrochloride 11.
Consequently we elected to pursue a demethylation
strategy (Scheme 4), and indeed, the use of 1-chloroethyl
chloroformate (ACE-Cl)8 led to clean demethylation to
facilitate ionization of the intermediate chlorosulfinyl
ester to the cation, and indeed, increased amounts of
olefin were observed using solvents such as 1,2-DME or
MeCN (entries 4 and 5), compared to the various hydro-
carbon solvents (entries 1-3). In terms of chloride
additives, simple lithium chloride and both tetramethyl-
and tetraethylammonium chloride exhibit insufficient
solubility in hydrocarbons (even at elevated tempera-
tures) and consequently do not increase the concentration
of chloride ion in solution (entries 6-8). Tetrabutyl-
ammonium chloride (Bu4NCl) is soluble in toluene at 85
°C and was found to have a significant effect on the
course of the reaction (entries 9-12). On a small scale, a
50 mol % charge of Bu4NCl was found to have an optimal
effect and led to a 95% isolated yield of the chloride 1,
containing 5% olefin 7. When this optimum procedure
was used on 2.7 kg of 4-hydroxy compound 6, an
improved suppression of the side-product was realized,
with only 2% of olefin 7 being detected and chloropiperi-
dine 1 isolated in 85% yield. This result demonstrates
the practical control over nucleophilic substitution at a
secondary center that is possible through the use of an
appropriate additive that can suppress the formation of
side-products resulting from elimination.
Concurrent with our development of practical experi-
mental protocols for the piperidone-based route to 1-tert-
butyl-4-chloropiperidine 1, an alternative means of access
to this Grignard precursor was investigated. Despite
having made significant improvements with regard to
several aspects of the piperidone route (Scheme 1), the
use of Bu4NCl as an additive in the chlorination reaction
was considered to be a short-term solution for two
reasons. First, the relatively large molecular weight of
Bu4NCl meant that for a 2.7 kg run of this reaction a
charge of 2.5 kg of Bu4NCl was required. As such, the
mass charge efficiency of this reaction is low and this
would almost certainly become an issue on increasing
scale. Also, anhydrous Bu4NCl proves to be relatively
expensive, which is unattractive considering this reagent
(4) For direct tertiary alkylation of oxygen to form tert-butyl ethers,
see: Masada, H.; Gotoh, H.; Ohkubo, M. Chem. Lett. 1991, 10, 1739.
Failed pyridone N-alkylation: Sato, T.; Yoshimatsu, K.; Otera, J.
Synlett 1995, 8, 845. Pyrazole alkylation: Elguero, J.; Jaquier, R.; Tien
Duc, H. C. N. Bull. Soc. Chim. Fr. 1966, 3727.
(5) Leonard, N. J.; Paukstelis, J. V. J. Org. Chem. 1963, 28, 3021.
(6) For examples, see: (a) Courtois, G.; Miginiac, P. Bull. Soc. Chim.
Fr. 1982, 395. (b) Miyano, S.; Yamashita, O.; Sumoto, K.; Shima, K.;
Hayashimatsu, M.; Satoh, F. J. Heterocycl. Chem. 1987, 24, 271.
(7) For a conceptually similar report on the addition of methylmag-
nesium bromide to nitrones as a means of generating tert-butyl groups
on nitrogen, see: Schwartz, M. A.; Hu, X. Tetrahedron Lett., 1992, 33,
1689.
J. Org. Chem, Vol. 70, No. 5, 2005 1931

SCHEME 4a
SCHEME 5a
a Reagents and conditions: (a) MeMgCl, THF, -15 or 25 °C.
1.90 ppm (Ha, d, J ) 0.8 Hz, 3H). Clearly, deprotonation
of one of the methyl groups of the iminium salt was
competing with the desired addition reaction, and indeed,
this served to explain the small amount of effervescence
observed during addition of the Grignard (loss of meth-
ane). With this information in hand, a series of experi-
ments were conducted to uncover the optimal conditions
for this reaction. Sources of methyl anion tested included
methyllithium and methyllithium-lithium bromide com-
plex, methyl Grignards with chloride, bromide and iodide
counterions, and dimethylzinc. Several solvents and
temperatures for the reaction were also investigated since
it is well-known that these parameters can significantly
influence the aggregation state of organometallic species
and therefore the relative nucleophilicity vs basicity of
these reagents.9 From this study it was determined that
the best results toward the preparation of 1 could be
achieved using methylmagnesium chloride in THF at
reduced temperature (-15 °C). Under these conditions,
smooth conversion to the desired 1-tert-butyl-4-chloro-
piperidine 1 is observed with less than 5% of iminium
deprotonation and piperidine 8 can be easily removed
during mildly acidic aqueous workup by taking advan-
tage of its increased basicity relative to the desired
product 1 (unhindered secondary amine vs hindered
tertiary amine). Concentration of the final MTBE solution
affords 95% yield of 1-tert-butyl-4-chloropiperidine 1 (with
99% GC purity) containing none of the olefinic side
product 7 observed during the chlorination step in the
piperidone chemistry.
As noted above, it was necessary to convert the HCl
salt 11 into the corresponding HBr salt 13 for the
subsequent iminium formation. Although fairly simple
on paper, this change of acid salt is less than optimal in
the general sense that it adds an isolation step to the
overall synthesis of 1-tert-butyl-4-chloropiperidine via
this route. Further, and specific to this case, are the
problems associated with the volatility of 4-chloropiperi-
dine (8) if the HBr salt formation requires prior azeotro-
pic drying and/or solvent switching. There are also
operational difficulties linked with the use of a corrosive
gas such as hydrogen bromide. Consequently, an alterna-
tive method was sought. Based on the difference in
solubility between sodium bromide and sodium chloride,
a Reagents and conditions: (a) aq K2CO3/1,2-DCE; (b) ACE-Cl,
0 to 80 °C; (c) MeOH, 65 °C; (d) aq K2CO3/MTBE; (e) HX; (f) 2,2-
dimethoxypropane, MeCN, 70 °C.
afford a 90% yield of 4-chloropiperidine hydrochloride 11
in a one-pot reaction.
The hydrochloride salt 11, obtained directly from the
demethylation reaction for iminium formation, has low
solubility in various solvent mixtures, which precluded
any appreciable rate of reaction. As a result, a variety of
alternative acid salts (12-16) of 4-chloropiperidine were
prepared and then tested in the iminium formation. From
this screen, the triflate and hydrobromide salts (12 and
13) emerged as good candidates for efficient iminium
production and these were then studied in some detail.
The triflic acid salt 12 furnished a homogeneous reaction
mixture using 2,2-dimethoxypropane (DMP) as solvent
and essentially quantitative conversion to the iminium
18 was observed within 3 h by NMR analysis. Under
similar conditions, using MeCN as cosolvent, hydrobro-
mide salt 13 is only partially soluble but nevertheless
steadily converts to the iminium species 19. Significantly,
and in contrast to the triflate case, iminium bromide 19
has low solubility in the DMP/MeCN mixture, causing
19 to precipitate cleanly as a white solid. Importantly,
no Finkelstein-type exchange of the 4-chloro substituent
was observed under the reaction conditions. Optimal
results were achieved at an overall concentration of 0.25
M and using a 1:1 mixture of DMP-MeCN, leading to a
reproducible yield of 87-89% of iminium salt 19, isolated
by simple filtration of the reaction mixture.
The addition of a methyl nucleophile to the iminium
salt 19 was initially attempted using a Grignard reagent
(Scheme 5). A THF slurry of pure iminium salt was
treated with methylmagnesium chloride at 25 °C, which
resulted in a colorless and homogeneous solution upon
warming to room temperature. After aqueous workup,
the desired 1-tert-butyl-4-chloropiperidine 1 was isolated
alongside one other product that was identified by NMR
as 4-chloropiperidine 8. Direct NMR analysis of the
reaction mixture before aqueous quench revealed the
presence of enamine 23 (hydrolyzed upon aqueous work-
up to 8), as indicated by the distinctive signals at 3.86
ppm (Hc, q, J ) 0.8 Hz, 1H), 3.85 ppm (Hb, s, 1H) and
(9) For example: Briggs, T. F.; Winemiller, M. D.; Collum, D. B.;
Parsons, R. L., Jr.; Davulcu, A. H.; Harris, G. D.; Fortunak, J. M.;
Confalone, P. N. J. Am. Chem. Soc. 2004, 126, 5427.
(8) Olofson, R. A.; Martz, J. T.; Senet, J.-P.; Piteau, M.; Malfroot,
T. J. Org. Chem., 1984, 49, 2081.
1932 J. Org. Chem., Vol. 70, No. 5, 2005

SCHEME 6a
iminium formation. As such, the chemistry can be
conducted as a through process beginning with the
N-methyl salt 10 and featuring no isolations until imi-
nium bromide 19. Only 1.2 equiv of NaBr are required
and the resultant mixture of piperidine salt 13, NaBr and
NaCl can simply be solvent switched into the appropriate
quantity of MeCN for the next step. As a result, the
synthesis of 1-tert-butyl-4-chloropiperidine 1 via the
iminium route can be achieved in 71% overall yield
starting from 4-chloro-1-methylpiperidine hydrochloride
10 and involves only two isolations (Scheme 6).
In summary, several important modifications to pub-
lished procedures were developed that allowed for the
rapid synthesis of multi-kilogram quantities of 1-tert-
butyl-4-chloropiperidine (1) starting from 1-methyl-4-
piperidone (2).2 In particular, a degree of control over the
conversion of a secondary alcohol to the corresponding
chloride was achieved through the use of Bu4NCl as an
additive. During this campaign, several long-term draw-
backs with the piperidone route were identified and this
fueled investigations toward an alternative synthesis.
Ultimately, we developed an efficient novel approach,
which furnishes the targeted intermediate in 71% overall
yield and 99% purity and involves only two isolations.
To the best of our knowledge, the iminium salt formation/
methyl Grignard addition sequence has not previously
been documented as a method for the introduction of a
tert-butyl group onto a secondary nitrogen.
a Reagents and conditions: (a) aq K2CO3/1,2-DCE; (b) ACE-Cl,
0 to 80 °C; (c) MeOH, 65 °C; (d) NaBr, MeOH; (e) 2,2-dimethoxy-
propane, MeCN, 70 °C; (f) MeMgCl, THF, -15 °C.
it seemed reasonable that a simple counterion exchange
could be achieved by treating the HCl salt with sodium
bromide in a suitable solvent. Thus, following treatment
of a methanol solution of the HCl salt with 5 equiv of
NaBr at room temperature, a heterogeneous mixture was
obtained. Filtration removed the precipitated NaCl and
concentration of the filtrate left the HBr salt alongside
some excess NaBr that had remained in solution. Capil-
lary electrophoresis on the isolated solid verified that
counterion exchange had occurred. Further, a use-test
of the material in the iminium formation gave essentially
the same result as that obtained using the standard HBr
salt. Based on these findings it was possible to telescope
the anion exchange step back into the previous demeth-
ylation reaction since this step directly affords a MeOH
solution of the HCl salt. Thus, the entire demethylation/
counterion exchange sequence proceeded smoothly and
afforded material that performed satisfactorily in the
iminium formation. Last, in a further refinement to this
process, it was discovered that the NaCl produced during
the counterion exchange need not be filtered off before
Supporting Information Available: Full experimental
procedures and analytical data. This material is available free
of charge via the Internet at http://pubs.acs.org.
JO048047G
J. Org. Chem, Vol. 70, No. 5, 2005 1933