Synthesis of a variety of 2-alkyl-2-azabicyclo[31.1]heptane-1-carbonitriles via a dynamic addition-intramolecular substitution sequence

Article abstract of DOI:10.1055/s-0030-1260811

An improved two-step synthetic approach towards 3-(2-chloroethyl) cyclobutanone is described and used in the synthesis of a class of 2-alkyl-2-azabicyclo[3.1.1]heptane-1-carbonitriles. The key step consists of a reversible addition of hydrogen cyanide ont

Full text of DOI:10.1055/s-0030-1260811

1748
LETTER
Synthesis of a Variety of 2-Alkyl-2-Azabicyclo[3.1.1]heptane-1-carbonitriles
via a Dynamic Addition–Intramolecular Substitution Sequence
S
ynthesis of 2-Alk
n
yl-2-Azabicy
n
c
lo[3. 1.1]heptane-
D
1-carbonitriles e Blieck, Christian V. Stevens*
Department of Sustainable Organic Chemistry and Technology, Faculty of BioScience Engineering, Ghent University,
Coupure Links 653, 9000 Ghent, Belgium
Fax +32(9)2646243; E-mail: chris.stevens@ugent.be
Received 11 February 2011
tanone.5 After extensively optimizing the reaction condi-
Abstract: An improved two-step synthetic approach towards 3-(2-
chloroethyl)cyclobutanone is described and used in the synthesis of
a class of 2-alkyl-2-azabicyclo[3.1.1]heptane-1-carbonitriles. The
key step consists of a reversible addition of hydrogen cyanide onto
the in situ generated imines, followed by an intramolecular nucleo-
philic substitution, thereby leading to the bicyclic skeleton in mod-
erate to good yields (47–92%). These bicyclic compounds are
stable, and the incorporated cyano group can be easily reduced to
the corresponding aminomethyl group in high yields (93–99%),
using lithium aluminum hydride.
tions 2,4-methanoproline (1) could be synthesized on
quite a large scale with an overall yield of 10%.6
Since detailed studies yielding satisfactory results of other
constrained amino acids like 2,4-methanepipecolic acid
(2, Figure 1) are scarce,7 the cyanide-induced dynamic in-
tramolecular cyclization reaction was further developed,
leading to 2-alkyl-2-azabicyclo[3.1.1]heptane-1-carboni-
triles 4 (Scheme 1). These bicyclic compounds possess
the rare 2,4-methanopiperidine as a scaffold. Radchenko
et al.7a published a similar reaction sequence to form 2-
benzyl-2-azabicyclo[3.1.1] heptane-1-carbonitrile (4a) from
3-(2-chloroethyl)cyclobutanone (3), using a threefold ex-
cess amount of acetone cyanohydrin (13, compared to 2
equiv in our protocol) and the established yield of 4a was
much lower (42% compared to our 79%). The present pa-
per highlights the different synthetic strategies towards
these bicyclic systems, starting from the improved synthe-
sis of the key compound 3-(2-chloroethyl)cyclobutanone
(3). Furthermore, our method proved to be applicable on
a wide range of amines, leading to a variety of bicyclic
compounds (Table 1), whereas Radchenko et al. de-
scribed only one example.
Key words: nitriles, heterocycles, bicyclic compounds, ring clo-
sure, nucleophilic addition
Conformationally restricted amino acids are of paramount
value for drug design. In this regard, proline and pipecolic
acid have been excellent building blocks to explore the
synthesis of azabicyclic amino acids. Their incorporation
in targeted peptidomimetics can improve the pharmaco-
kinetic and pharmacodynamic properties of possible drug
candidates.1 An often described example is 2,4-methano-
proline (1), which is a natural nonproteinogenic amino ac-
id, isolated from the seeds of Ateleia Herbert smithii
Pittier.2 The seeds of this legume species are ignored by at
least 100 seed predators. Therefore, it is believed that 2,4-
methanoproline (1), which is present in these seeds, acts
as an antifeedant.3
method A
O
N
R
CN
N
N
COOH
COOH
Cl
3
H
H
4
1
2
method B
Figure 1 2,4-Methanoproline (1) and 2,4-methanepipecolic acid (2)
Scheme 1 Synthesis of the bicyclic compounds 4. Reagents and
conditions: method A: RNH2 (1 equiv), acetone cyanohydrin (2
equiv), Et3N (2 equiv), dry MeOH, temp, 3 d; method B: 14 (1.1
equiv), MeCN, temp, 3 d.
The 2-aza-bicyclo[2.1.1]hexane skeleton has drawn the
attention of several groups, developing different strategies
for the formation of this bicyclic system. Since its isola-
tion in 1980, several synthetic pathways towards 2,4-
methanoproline (1) have been developed.4 Unfortunately,
the reaction conditions and reported yields were unsatis-
factory, and therefore a new route was elaborated in our
research group, starting from 3-(chloromethyl)cyclobu-
A new synthetic protocol for the key compound, 3-(2-
chloroethyl)cyclobutanone (3), was developed for its
preparation on multigram scale (Scheme 2). In the first
step, a [2+2] cycloaddition of in situ generated dichlo-
roketene and homoallyl chloride8 (7) led to the cyclobu-
tanone 9b.9 Many different reaction conditions were
evaluated to improve the yield of this reaction. The best
results were obtained by generating the dichloroketene
from trichloroacetyl chloride and a zinc–copper couple.
SYNLETT 2011, No. 12, pp 1748–1752
x
x
.x
x
.2
0
1
1
Advanced online publication: 29.06.2011
DOI: 10.1055/s-0030-1260811; Art ID: D06311ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
Synthesis of 2-Alkyl-2-Azabicyclo[3.1.1]heptane-1-carbonitriles
1749
ref. xia
95%
OBn
Cl
a
O
O
6
7
Cl
b
OH
a
R
R
ref. viii
80%
9–c
Cl
5
10 R = OBn
3 R = Cl
c
x
a
Br
8
Scheme 2 Synthesis of the key compound 3-(2-chloroethyl)cyclobutanone (3). Reagents and conditions: (a) Zn–Cu couple, 1,2-DME,
Cl3CCOCl, dry Et2O, r.t., overnight; 9a R = OBn, 89%; 9b R = Cl, 78%; 9c R = Br, 70%; (b) Zn, AcOH, temp, overnight; 10 R = OBn, 80%;
3 R = Cl, 86%; (c) 12 N HCl, ZnCl2, temp, 1d.
Formation of the ketene from dichloroacetyl chloride, us- leading to the desired end product. If the reaction is pro-
ing triethylamine as a base, always led to a complex reac- longed to several days, the desired end product 4 is almost
tion mixture, probably as a result of the instability of the completely formed, but some adduct 12 is still present, de-
end product in basic medium. To obtain the 3-(2-chloro- pending on the R group. The nearly constant ratio of the
ethyl)cyclobutanone (3), the cycloaddition product 9b cis/trans adducts (11 and 12) indicates that the isomeriza-
was treated with zinc in acetic acid, reductively removing tion is relatively fast compared to the ring closure. After
both geminal chlorine atoms without the formation of any three days of reflux, the reaction was complete, and the
side product.10
2-alkyl-2-azabicyclo[3.1.1]heptane-1-carbonitrile 4 was
isolated as the sole product, together with a very small
fraction of starting material.12a
Another approach leading to 3-(2-halomethyl) cyclobu-
tanones was evaluated, starting from the known cyclobu-
tanone 10. Slight modification of the literature procedure The obtained bicyclic products 4 can be purified by two
to prepare 10 from 6, using 1,2-dimethoxyethane instead means in order to remove the excess of acetone cyanohy-
of POCl3, gave higher yields than previously reported by drin. The first method consists of a purification by column
other research groups.11 However, when using a concen- chromatography, which has the disadvantage that the sol-
trated hydrochloric acid solution with zinc(II) chloride as id-phase silica has a high affinity for the end products,
a catalyst, cyclobutanone 10 could not be converted into leading to severe losses and consequently lower yields. A
the desired 3-(2-chloroethyl)cyclobutanone (3).
more convenient method to remove the acetone cyanohy-
drin consists of an acid–base extraction. This method is
less time-consuming, and the yields after purification are
significantly higher compared to the first method.
3-(2-Bromoethyl)-2,2-dichlorocyclobutanone (9c) could
also be prepared, starting from the commercially available
homoallyl bromide (8). This precursor would be more in-
teresting than compound 9b, because of the better leaving In a second approach (Scheme 1, method B), based on an
group capacity of the bromine atom, taking into consider- earlier reported method of Komarov and co-workers,13 the
ation the subsequent ring closure to form the bicyclic skel- 2-azabicyclo[3.1.1]heptane skeleton might be formed us-
eton. Unfortunately, in this case the a-dechlorination step ing a ‘tandem Strecker–intramolecular cyclization reac-
failed. Side products were formed during the reaction, and tion’. We investigated the feasibility of applying this
purification of the reaction mixture by distillation led to
R
the formation of dimers.
HN
The best strategy towards the key compound 3 was the use
of a [2+2] cycloaddition with homoallyl chloride (7) in the
presence of a zinc–copper couple to generate dichlo-
roketene from trichloroacetyl chloride, followed by radi-
CN
Cl
11
cal dechlorination of 9b with zinc in acetic acid.
O
N
In an initial experiment to construct the 2-azabicyc-
R
lo[3.1.1]heptane skeleton, cyclobutanone 3 in dry metha-
–CN
RNH2
–CN
Cl
nol was treated with a primary amine, triethylamine, and
acetone cyanohydrin as a hydrogen cyanide source
(Scheme 1, method A). The reaction was performed in a
sealed pressure vessel in refluxing methanol for three
days. If the reaction was stopped after a few hours of re-
flux, the starting material was almost completely convert-
ed into the adduct 12 (Scheme 3) and only traces of the
bicyclic end product 4 could be observed. Especially note-
worthy is that only the cis-isomer 11 allows ring closure,
N
R
CN
Cl
3
4
CN
R
N
H
Cl
12
Scheme 3 Mechanism of ring closure
Synlett 2011, No. 12, 1748–1752 © Thieme Stuttgart · New York

1750
A. De Blieck, C. V. Stevens
LETTER
protocol in our research. Adduct 14 was easily prepared in yield (Scheme 6).17 Applying these protocols, a series of
quantitative yield in a separate reaction, starting from ac- N-alkyl derivatives was synthesized.
etone cyanohydrin 13 (Scheme 4).14 Next, it was added to
a solution of 3-(2-chloroethyl)cyclobutanone (3) in re-
fluxing acetonitrile, again leading to the desired end prod-
ucts. As mentioned before, isolation of the end products
was also achieved using column chromatography or an
acid–base extraction.12b
LiAlH4 (2 equiv)
THF, r.t., overnight, N2
NH2
N
R
N
R
CN
93–99%
15
4
Scheme 5 Reduction of the nitrile functionality (R = Bn; PMB;
i-Pr; n-Pr; i-Bu; n-Bu; 4-MeBn; 2,4-DMB; Et)
CN
RNH2 (1 equiv)
dry MeOH, r.t., 24 h, N2
HO
CN
R
N
H
64–99%
Pd/C
H2 (5 bar)
6 N HCl
13
14
Δ, overnight
MeOH, r.t.
N
R
CN
N
COOH
N
COOH
Scheme 4 Synthesis of adduct 14
HCl⋅ H
HCl⋅ R
4
16a R = Bn (98%)
16b R = PMB (90%)
17 (94%)
However, when comparing the two methodologies, the
previously described method A gave more satisfying re-
sults than this second strategy. An overview of the ob-
tained yields is depicted in Table 1. The significant
discrepancy in yield between entries 4b and 4c can be at-
tributed to difficulties during isolation. Nevertheless, we
were capable of preparing a range of derivatives in mod-
erate to good yields.
Scheme 6 Synthesis of methanepipecolic acid
The results described clearly show the usefulness of 3-(2-
chloroethyl)cyclobutanone 3 as a new building block to
construct the 2-azabicyclo[3.1.1]heptane skeleton. This
key compound is prepared by a two-step methodology, in
a way that allows a multigram synthesis. Due to this short
sequence, the cyclobutanone 3 forms an ideal precursor
for the synthesis of constrained pipecolic acid analogues
in only four steps.
Table 1 Established Yields of Compound 4 via Methods A and B
Compound 4
R
Method A (%) Method B (%)
4a
4b
4c
4d
4e
4f
Bn
79
85
59
73
78
79
70
80
72
92
75
70
62
61
62
47
58
59
60
62
PMB
i-Pr
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synlett.
n-Pr
i-Bu
n-Bu
4-Me-Bn
2,4-DMB
PMP
Et
Acknowledgment
This work was supported by the Ghent University Research Fund.
4g
4h
4i
References and Notes
(1) (a) Patrick, G. L. In An Introduction to Medicinal
Chemistry, 3rd ed.; Oxford University Press: New York,
2005, 210. (b) Hanessian, S.; Auzzas, L. Acc. Chem. Res.
2008, 41, 1241.
(2) Bell, E. A.; Qureshi, M. Y.; Pryce, R. Y.; Janzen, D. H.;
Lemke, P.; Clardy, J. J. Am. Chem. Soc. 1980, 102, 1409.
(3) (a) Montelione, G. T.; Hughes, P.; Clardy, J.; Scheraga,
H. A. J. Am. Chem. Soc. 1986, 108, 6765. (b) Talluri, S.;
Montelione, G. T.; van Duyne, G.; Piela, L.; Clardy, J.;
Scheraga, H. A. J. Am. Chem. Soc. 1987, 109, 4473.
(c) Mapelli, C.; Van Halbeek, H.; Stammer, C. H.
Biopolymers 1990, 29, 407.
(4) General info: (a) Pirrung, M. C. Tetrahedron Lett. 1980, 21,
4577. (b) Hughes, P.; Martin, M.; Clardy, J. Tetrahedron
Lett. 1980, 21, 4579. (c) Hughes, P.; Clardy, J. J. Org.
Chem. 1988, 53, 4793. (d) Gaoni, Y. Org. Prep. Proced. Int.
1995, 27, 185. Via [2+2]cyclisation: (e) Tamura, Y.;
Ishibashi, H.; Hirai, M.; Kita, Y.; Ikeda, M. J. Org. Chem.
1975, 40, 2702. (f) Schell, F. M.; Cook, P. M.; Hawkinson,
S. W.; Cassady, R. E.; Thiessen, W. E. J. Org. Chem. 1979,
44, 1380. (g) Esslinger, C. S.; Koch, H. P.; Kavanaugh, M.
P.; Philips, D. P.; Chamberlin, A. R.; Thompson, C. M.;
a
4j
–
a
4k
Me
–
a Given the volatile properties of EtNH2 and MeNH2 only method A
could be applied using 2 equiv of RNH2.
The bicyclic compounds 4 are stable, and the incorporated
cyano group also offers the possibility to elaborate the
chemistry further. The cyano group can, for example, be
easily reduced to the corresponding aminomethyl group in
excellent yields, using lithium aluminum hydride
(Scheme 5).15 Furthermore, the nitrile function can be
converted into a carboxylic acid moiety in relatively good
yields, using concentrated hydrochloric acid under reflux
conditions.16 After deprotection with Pd/C (10 wt%) 2,4-
methanopipecolic acid (17) was obtained in 92% overall
Synlett 2011, No. 12, 1748–1752 © Thieme Stuttgart · New York

LETTER
Synthesis of 2-Alkyl-2-Azabicyclo[3.1.1]heptane-1-carbonitriles
1751
Bridges, R. J. Bioorg. Med. Chem. Lett. 1998, 8, 3101.
(h) Piotrowsky, D. W. Synlett 1999, 1091. (i) Kwak, Y.-S.;
Winkler, J. D. J. Am. Chem. Soc. 2001, 123, 7429.
(j) Vogler, B.; Bayer, R.; Meller, M.; Kraus, W.; Schell, F.
M. J. Org. Chem. 1989, 54, 4165. (k) Toda, F.; Miyamoto,
H.; Takeda, K.; Matsugawa, R.; Maruyama, N. J. Org.
Chem. 1993, 58, 6208. (l) Tkachenko, A. N.; Radchenko, D.
S.; Mykhailiuk, P. K.; Grygorenko, O. O.; Komarov, I. V.
Org. Lett. 2009, 11, 5674. Via an appropriate
2 × CH2CO), 3.59 (2 H, t, J = 6.6 Hz, CH2Cl). 13C NMR (75
MHz, CDCl3): d = 21.71 (CH), 38.57 (CHCH2), 43.41
(CH2Cl), 52.34 (2 × CH2CO), 207.01 (C=O). ESI-MS:
m/z (%) = 135 (45), 133 (100).
(11) (a) Ripin, D. H. B.; Vetelino, M. Synlett 2003, 2353.
(b) Liotta, D. C.; Mao, S.; Hager, M. WO 2006063281,
Chem. Abstr. 2006, 145, 63117. (c) Kabalka, G. W.; Yao,
M.-L.; Navarane, A. Tetrahedron Lett. 2005, 46, 4915.
(12) Typical Procedure for the Synthesis of 2-Alkyl-2-
Azabicylo[3.1.1]heptane-1-carbonitriles 4
bromohydrine: (m) Krow, G. R.; Lee, Y. B.; Lester, W. S.;
Christian, H.; Shaw, D. A.; Yaun, J. J. Org. Chem. 1998, 63,
8558. (n) Krow, G. R.; Lee, Y. B.; Lester, W. S.; Liu, N.;
Yaun, J.; Duo, J.; Herzon, S. B.; Nguyen, Y.; Zacharias, D.
J. Org. Chem. 2001, 66, 1805. (o) Krow, G. R.; Lester, W.
S.; Liu, N.; Yuan, J.; Hiller, A.; Duo, J.; Herzon, S. B.;
Nguyen, Y.; Cannon, K. J. Org. Chem. 2001, 66, 1811. Via
an intramolecular nucleophilic substitution: (p) Lescop, C.;
Mevellec, L.; Huet, F. J. Org. Chem. 2001, 66, 4187.
(5) Stevens, C. V.; De Kimpe, N. J. Org. Chem. 1996, 61, 2174.
(6) (a) Rammeloo, T.; Stevens, C. V.; De Kimpe, N. J. Org.
Chem. 2002, 67, 6509. (b) Rammeloo, T.; Stevens, C. V.
Chem. Commun. 2002, 250. (c) Stevens, C. V.; Smagghe,
G.; Rammeloo, T.; De Kimpe, N. J. Agric. Food Chem.
2005, 53, 1945.
(a) In a dry, pressure resistant vessel (20 mL volume) 3-(2-
chloroethyl)cyclobutanone (3, 2.00 g, 15 mmol, 1 equiv), a
primary amine (15 mmol, 1 equiv), acetone cyanohydrin
(2.57 g, 30 mmol, 2 equiv), and Et3N (3.05 g, 30 mmol, 2
equiv) were dissolved in dry MeOH (16 mL). The vessel was
closed and heated to 110 °C for 2–3 d. When using ethyl-
(pure) or methylamine (2 M in MeOH), the vessel was
heated for 4 d, using 30 mmol of the volatile amine. Isolation
of the desired end product could be performed by two means.
The first method made use of column chromatography. After
washing of the reaction mixture with a sat. NaHCO3
solution, 3 g of silica were added to the organic phase
(CH2Cl2), followed by removal of the solvent under vacuum.
The end product was then recovered using column chroma-
tography. The second purification strategy was more
convenient and consisted of an acid–base extraction. After
removal of the solvent under reduced pressure, 10 mL of a 2
N HCl solution was added. The solution was extracted with
Et2O (3 × 20 mL) to remove the excess of acetone cyano-
hydrin. A concentrated K2CO3 solution was added to the
H2O layer until basic, followed by an extraction of the H2O
layer with CH2Cl2 (3 × 30 mL). The combined organic
layers (CH2Cl2) were dried with MgSO4. After filtration of
the solids and removal of the volatiles, the pure 2-R0-2-
azabicyclo[3.1.1]heptane-1-carbonitrile (4) was obtained in
moderate to good yields, depending on the R0 group.
(b) In a flame-dried flask of 50 mL, 3-(2-chloroethyl)-
cyclobutanone (3, 1.00 g, 7.5 mmol) was dissolved in
MeCN, together with 14 (8.3 mmol, 1.1 equiv). Hereafter,
the reaction mixture was brought to reflux temperature and
stirred for 3 d. The pure end product was isolated according
to the same procedures as mentioned above in 12a.
2-(4-Methoxybenzyl)-2-azabicyclo[3.1.1]heptane-1-
carbonitrile (4b)
Yellow crystals (3.09 g, 85%). Anal. Calcd (%) for
C15H18N2O: C, 74.4; H, 7.5; N, 11.6. Found: C, 74.3; H, 7.6;
N, 11.3. Rf = 0.35 (PE–EtOAc = 7:3). IR (ATR): 2359 (CN),
1612, 1515, 1495, 1454 (Ar) cm–1. 1H NMR (300 MHz,
CDCl3): d = 1.91 (2 H, td, J = 6.6, 3.3 Hz, CH2), 2.24 (2 H,
dd, J = 7.2, 2.2 Hz, 2 × CqCHaHb), 2.41 (2 H, td, J = 7.2, 2.2
Hz, 2 × CqCHaHb), 2.49 (1 H, ca. sept, J = 3.3 Hz, CH), 2.85
(2 H, t, J = 6.6 Hz, NCH2), 3.79 [5 H, s, NCH2Ar, OCH3
(Ar)], 6.86 [2 H, d, J = 8.8 Hz, 2 × CH (Ar)], 7.29 [2 H, d,
J = 8.8 Hz, 2 × CH (Ar)]. 13C NMR (75 MHz, CDCl3):
d = 27.68 (CH2), 31.25 (CH), 37.21 (2 × CqCHaHb), 42.49
(NCH2), 55.27 [OCH3 (Ar)], 56.60 (NCH2Ar), 58.39 (Cq),
113.72 [2 × CH (Ar)], 120.36 (CN), 130.01 [2 × CH (Ar)],
130.79 [Cq (Ar)], 158.83 [Cq (Ar)]. ESI-MS: m/z (%) = 291
(20), 214 (15), 213 (100) [MH+].
(7) (a) Radchenko, D. S.; Kopylova, N.; Grygorenko, O. O.;
Komarov, I. V. J. Org. Chem. 2009, 74, 5541. (b) Jo, H.;
Fitzgerald, M. E.; Winkler, J. D. Org. Lett. 2009, 11, 1685.
(8) Roberts, J. D.; Mazur, R. H. J. Am. Chem. Soc. 1951, 73,
2509.
(9) 3-(2-Chloroethyl)-2,2-dichlorocyclobutanone (9b)
In an oven-dried two-necked flask of 500 mL, a solution of
homoallyl chloride (15 g, 166 mmol) and a zinc–copper
couple (43.32 g, 663 mmol) in dry Et2O (250 mL) was
cooled to 0 °C under a nitrogen atmosphere. A solution of
trichloroacetyl chloride (60.24 g, 331 mmol) and 1,2-
dimethoxyethane (29.86 g, 331 mmol) in dry Et2O (150 mL)
was added dropwise, after which the reaction mixture was
stirred overnight at r.t. The solution was filtered over Celite®
and washed with Et2O. This filtrate was extracted with H2O
(2 × 100 mL), NaHCO3 (4 × 100 mL), brine (2 × 100 mL).
The organic layer was dried over MgSO4 and the solvent
removed under reduced pressure, leading to the desired 3-(2-
chloroethyl)-2,2-dichlorocyclobutanone (26.24 g, 78%) as a
clear orange oil. IR (NaCl): 1811 (C=O) cm–1. 1H NMR (300
MHz, CDCl3): d = 2.05–2.17 (1 H, m, CHCH2), 2.38–2.50
(1 H, m, CHCH2), 3.10 (1 H, dd, J = 16.2, 9.4 Hz, CH2CO),
3.11–3.24 (1 H, m, CH), 3.44 (1 H, dd, J = 16.2, 9.4 Hz,
CH2CO), 3.66–3.72 (2 H, m, CH2Cl). 13C NMR (75 MHz,
CDCl3): d = 33.94 (CHCH2), 42.12 (CH2Cl), 43.50 (CH),
47.64 (CH2CO), 88.51 (Cq), 191.87 (C=O). ESI-MS: m/z
(%) = 205 (35), 203 (85), 201 (100).
(10) 3-(2-Chloroethyl)cyclobutanone (3)
A solution of 3-(2-chloroethyl)-2,2-dichlorocyclobutanone
(9b, 28.23 g, 140 mmol) in AcOH (100 mL) was vigorously
stirred, while slowly adding 2 equiv of zinc (18.32 g, 280
mmol). Two extra equiv of zinc (18.32 g, 280 mmol) were
added to the reaction mixture, after which it was refluxed
overnight. After cooling, the mixture was filtered over
Celite® and washed with CH2Cl2. The filtrate was neutra-
lized with a sat. NaHCO3 solution. The organic phase was
dried with MgSO4, filtered, and the solvent was removed in
vacuo. 3-(2-Chloroethyl)cyclobutanone (3) was obtained as
a bright yellow oil in 86% yield (15.90 g).
(13) Grygorenko, O. O.; Artamonov, O. S.; Palamarchuk, G. V.;
Zubatyuk, R. I.; Shishkin, O. V.; Komarov, I. V.
Tetrahedron: Asymmetry 2006, 17, 252.
(14) General Procedure for the Synthesis of 2-Methyl-2-
IR (NaCl): 1778 (C=O) cm–1. 1H NMR (300 MHz, CDCl3):
d = 2.08 (2 H, dd, J = 13.8, 6.6 Hz, CHCH2), 2.54–2.69 (1 H,
m, CH), 2.72–2.82 (2 H, m, 2 × CH2CO), 3.15–3.27 (2 H, m,
Alkyl/Aryl Aminopropionitriles (14)
A primary amine (30 mmol) was mixed with acetone
cyanohydrin 13 (2.55 g, 30 mmol) in dry MeOH (25 mL).
Synlett 2011, No. 12, 1748–1752 © Thieme Stuttgart · New York

1752
A. De Blieck, C. V. Stevens
LETTER
The solution was stirred at r.t. for 24 h, while N2 gas was
bubbled through. The solvent was removed in vacuo, and the
propionitriles 14 were obtained in moderate to excellent
yields (64–99%). Analytical samples were obtained after
filtration over silica or recrystallization in MeOH.
2-(4-Methoxybenzylamino)-2-methylpropionitrile (14b)
Yellow crystals (6.07 g, 99%). Anal. Calcd (%) C12H16N2O:
C, 70.6; H, 7.9; N, 13.7. Found: C, 70.3; H, 8.0; N, 13.5. IR
(ATR): 3275 (NH), 2359 (CN), 1611, 1515 (Ar) cm–1. 1H
NMR (300 MHz, CDCl3): d = 1.51 (7 H, s, 2 × CH3, NH),
3.80 [3 H, s, OCH3 (Ar)], 3.83 (2 H, s, NCH2), 6.87 [2 H, d,
J = 8.3 Hz, 2 × CH (Ar)], 7.28 [2 H, d, J = 8.3 Hz, 2 × CH
(Ar)]. 13C NMR (75 MHz, CDCl3): d = 27.44 (2 × CH3),
49.00 (NCH2), 51.68 (Cq), 55.29 [OCH3 (Ar)], 113.98
[2 × CH (Ar)], 122.85 (CN), 129.60 [2 × CH (Ar)], 131.11
[Cq (Ar)], 158.97 [Cq (Ar)]. ESI-MS: m/z (%) = 178 (100)
[M – CN]+).
(100) [MH+]. ESI-HRMS: m/z calcd C15H23ON2: 247.18049
[MH+]; found: 247.17857.
(16) Typical Procedure for the Synthesis of 2-Aryl-2-
azabicyclo[3.1.1]heptane-1-carboxylic Acid
Hydrochlorides 16
An amount of the 2-aryl-2-azabicyclo[3.1.1]heptane-1-
carbonitrile 4 (0.5 mmol) was dissolved in 6 N HCl (4 mL).
The reaction mixture was refluxed overnight. After cooling
to r.t., the mixture was kept in the freezer to crystallize the
acid from the aqueous solution.
2-(4-Methoxybenzyl)-2-azabicyclo[3.1.1]heptane-1-
carboxylic Acid Hydrochloride (16b)
White needles (133.7 mg, 90%). Anal. Calcd (%) for
C15H19NO3: C, 68.9; H, 7.3; N, 5.4. Found: C, 68.9; H, 7.4;
N, 5.35. IR (ATR): 3358 (br, OH), 1739 (CO), 1637, 1548
(Ar) cm–1. 1H NMR (300 MHz, D2O, MeCN): d = 2.20 (2 H,
td, J = 6.6, J = 3.3 Hz, CH2), 2.47 (2 H, dd, J = 9.4, J = 2.8
Hz, 2 × CqCHaHb), 2.75 (1 H, sept, J = 3.3 Hz, CH), 2.94 (2
H, dd, J = 11.8, J = 6.6 Hz, 2 × CqCHaHb), 3.51 (2 H, t,
J = 6.6 Hz, NCH2), 3.86 [3 H, s, OCH3 (Ar)], 4.60 (2 H, s,
NCH2Ar), 7.08 [2 H, d, J = 8.8 Hz, 2 × CH (Ar)], 7.52 [2 H,
d, J = 8.8 Hz, 2 × CH (Ar)]. 13C NMR (75 MHz, D2O,
MeCN): d = 24.72 (CH2), 30.30 (CH), 35.84 (2 × CqCHaHb),
46.50 (NCH2), 50.07 [OCH3 (Ar)], 58.27 (NCH2Ar), 59.63
(Cq), 115.39 [2 × CH (Ar)], 121.02 [Cq (Ar)], 133.43
[2 × CH (Ar)], 161.17 [Cq (Ar), COOH]. ESI-MS:
m/z (%) = 263 (20), 262 (100) [MH+].
(15) General Protocol for the Synthesis of (2-Alkyl-2-
azabicyclo[3.1.1]hept-1-yl)methylamine 15
A solution of LiAlH4 (0.19 g, 5 mmol,2 equiv) in dry THF
(15 mL) was stirred at –78 °C under a nitrogen atmosphere.
A flame-dried syringe was used to slowly add a solution of
2-alkyl-2-azabicyclo[3.1.1]heptane-1-carbonitrile 4 in dry
THF (2.5 mmol). Upon completion of the addition, the
reaction was stirred overnight at r.t., followed by a careful
addition of H2O to neutralize the excess of LiAlH4. The
solution was dried with MgSO4, followed by filtration of the
solids and evaporation of the solvent to give the pure {2-
alkyl-2-azabicyclo[3.1.1]hept-1-yl}methylamine (15, 93–
99%). When necessary, the compounds could be further
purified using filtration over a short silica column (CH2Cl2–
MeOH = 9:1).
(17) Synthesis of 2,4-Methanepipecolic Acid (17)
A suspension of 2-benzyl-2-azabicyclo[3.1.1]heptane-1-
carboxylic acid hydrochloride (16a, 251 mg, 0.94 mmol) and
Pd/C (10 wt%, 125 mg) in dry MeOH (2 mL) was placed in
a Parr apparatus which was degassed and filled with H2. The
mixture was stirred overnight applying a constant pressure
of 5 bar of H2. Filtration of the heterogeneous suspension
over Celite®, followed by evaporation of the solvent in
vacuo delivered the crude 2,4-methanepipecolic acid (17),
which was purified by recrystallization from MeOH;
colorless crystals (124.7 mg, 94%). Anal. Calcd (%) for
C7H11NO2: C, 59.6; H, 7.85; N, 9.9. Found: C, 59.6; H, 7.9;
N, 9.8. IR (ATR): 3358 (br, OH), 3187 (NH), 1742 (CO) cm–1.
1H NMR (300 MHz, D2O, MeCN): d = 1.87 (2 H, dd,
J = 8.3, J = 2.8 Hz, 2 × CqCHaHb), 2.09–2.23 (4 H, m,
2 × CqCHaHb, CH2), 2.60 (1 H, sept, J = 3.3 Hz, CH), 2.81 (2
H, br s, OH, NH), 3.53 (2 H, t, J = 7.2 Hz, NCH2). 13C NMR
(75 MHz, D2O, MeCN): d = 25.75 (CH2), 29.01 (CH), 33.38
(2 × CqCHaHb), 37.66 (NCH2), 65.21 (Cq), 171.65 (COOH).
ESI-MS: m/z (%) = 143 (20), 142 (100) [MH+].
C-{2-(4-Methoxybenzyl)-2-azabicyclo[3.1.1]hept-1-
yl}methylamine (15b)
Bright yellow oil (0.60 g, 97%). IR (ATR): 3365 (NH2),
1684 (Ar) cm–1. 1H NMR (300 MHz, CDCl3): d = 1.37 (2 H,
br s, NH2), 1.65 (2 H, td, J = 6.6, J = 2.8 Hz, 2 × CqCHaHb),
1.90 (2 H, td, J = 6.6, J = 3.3 Hz, CH2), 1.96 (2 H, dd,
J = 6.6, J = 2.8 Hz, 2 × CqCHaHb), 2.43 (1 H, sept, J = 3.3
Hz, CH), 2.66 (2 H, s, CqCH2NH2), 2.92 (2 H, t, J = 6.6 Hz,
NCH2), 3.53 (2 H, s, NCH2Ar), 3.80 [3 H, s, OCH3 (Ar)],
6.86 [2 H, d, J = 8.8 Hz, 2 × CH (Ar)], 7.26 [2 H, d, J = 8.8
Hz, 2 × CH (Ar)]. 13C NMR (75 MHz, CDCl3): d = 28.42
(CH2), 30.13 (CH), 33.84 (2 × CqCHaHb), 43.77 (NCH2),
48.33 (CqCH2NH2), 52.64 (NCH2Ar), 55.42 [OCH3 (Ar)],
68.29 (Cq), 113.82 [2 × CH (Ar)], 129.54 [2 × CH (Ar)],
133.53 [Cq (Ar)], 158.51 [Cq (Ar)]. ESI-MS: m/z (%) = 247
Synlett 2011, No. 12, 1748–1752 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.