Stereochemical substituent effects: Investigation of the cyano, amide and carboxylate group

Article abstract of DOI:10.1016/j.tet.2004.10.039

Three pairs of diastereomeric piperidines, cis- and trans-2- methylpiperidine-3-carboxylate (6a and 6b), cis- and trans-2-methylpiperidine-3- carboxylamide (9a and 9b) and cis- and trans-2-methyl-3-cyanopiperidine (11a and 11b), were synthesised for the purpose of investigating the effect of the axial versus equatorial carboxylate, carboxamide and cyano group on piperidine base strength. The pK_a values of the six compounds were determined to be 11.0 (6a), 10.4 (6b), 9.5 (9a), 9.3 (9b), 7.8 (11a) and 8.0 (11b). This shows that the strong electron-withdrawing effect of the cyano group and the effect of the amide group are relatively independent of spacial orientation. The carboxylate, on the other hand is considerably less electron-withdrawing when axial. Graphical Abstract.

Full text of DOI:10.1016/j.tet.2004.10.039

Tetrahedron 61 (2005) 115–122
Stereochemical substituent effects: investigation of the cyano,
amide and carboxylate group
*
Christian Marcus Pedersen and Mikael Bols
Department of Chemistry, University of Aarhus, Langelandsgade 140, DK-8000 Aarhus C, Denmark
Received 18 June 2004; revised 20 September 2004; accepted 14 October 2004
Available online 28 October 2004
Abstract—Three pairs of diastereomeric piperidines, cis- and trans-2-methylpiperidine-3-carboxylate (6a and 6b), cis- and trans-2-
methylpiperidine-3-carboxylamide (9a and 9b) and cis- and trans-2-methyl-3-cyanopiperidine (11a and 11b), were synthesised for the
purpose of investigating the effect of the axial versus equatorial carboxylate, carboxamide and cyano group on piperidine base strength. The
pKa values of the six compounds were determined to be 11.0 (6a), 10.4 (6b), 9.5 (9a), 9.3 (9b), 7.8 (11a) and 8.0 (11b). This shows that
the strong electron-withdrawing effect of the cyano group and the effect of the amide group are relatively independent of spacial orientation.
The carboxylate, on the other hand is considerably less electron-withdrawing when axial.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
reversibility, that piperidines with polar substituents may
change conformation on protonation.10 Thus, as shown by
Lankin and Snyder the fluoropiperidinecarboxylic acid 1
changes to the all axial conformation when the amino-group
is protonated (Scheme 1).12
Understanding the relationship between structure and
properties is a fundamental issue in organic chemistry. Of
central importance is the influence, in terms of electronic
effects, of substituents on the chemistry of compounds.
Substituent effects has been quantified by Hammett and
Taft1–3 and subsequently been refined so that very specific
information is known about the electronic influence of
functional groups.4–6 However, relatively little attention has
been given to the study of stereochemistry on substituent
effects. While it was noted early that the base-strength of
amines with polar groups varied widely, it was also
concluded that they appeared to be unpredictable.7,8
However, in recent work it has been observed that the
basicity of piperidines and hexahydropyridazines was
systematically affected by the axial or equatorial positioning
of ring hydroxyl, fluorine or ester-groups.9–11 An equatorial
hydroxyl group was found to be three times more electron
withdrawing than an axial hydroxyl group. This is
exemplified in the difference in base strength of the
epimeric glycosidase inhibitors isofagomine/galacto-isofa-
gomine and 1-deoxynojirimycin/galactostatin (Fig. 1), in
which the axial epimer is the stronger base. Similarly the
piperidines with equatorial ester and fluorine groups were
also found to be less basic than their axial isomers. An
important and fascinating consequence of stereochemical
substituent effects is, due to the principle of microscopic
Scheme 1. Conformational change of cis-3-fluoro-5-piperidinecarboxylic
acid as a result of pH change.
It has been suggested that differences in charge–dipole
interactions between the protonated amine and an axial or
equatorial polar substituent is the prime reason for the
influence of stereochemistry on piperidine base-strength.
The present work is an effort to expand our knowledge about
stereochemical substituent effects especially with the intent
of determining the importance of charge–dipole inter-
actions. For this purpose the nitrile-group was particularly
interesting as it has a strong dipole and is a strongly
electron-withdrawing group. If charge–dipole interactions
are crucial for the base-strength a significant difference
should be observed between piperidine isomers A and B
(Fig. 1). We here report the influence of axial and equatorial
positions on the base-lowering effect of carboxylic acid,
carboxylic amide and nitrile groups. We have synthesised
three pairs of stereoisomeric piperidines having a functional
group in the 3-position and with restricted conformation and
Keywords: Electronic effect; Base strength; Piperidines; Conformation.
* Corresponding author. Tel.: C45 89423963; fax: C45 86196199;
e-mail: mb@chem.au.dk
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.10.039

116
C. M. Pedersen, M. Bols / Tetrahedron 61 (2005) 115–122
Figure 1. Base strength of epimeric glycosidase inhibitors and target molecules 11a and 11b.
recorded their pKa values. We find that stereochemistry and/
or conformation has a minor influence on the base-lowering
effect of the cyano and carboxylic amide groups, while the
base-lowering effect of the carboxylic acid is essentially
eliminated in the axial position.
obtain adduct 3 (Scheme 2).14 Hydrogenation and reductive
amination using Raney Nickel catalyst and 45 atm hydrogen
pressure in EtOH gave the ester 4, as previously reported,14
in good yield. However, we observed that the N-ethyl
derivative was formed as a byproduct in the reaction
lowering yield and purity. This compound is presumably
formed by metal catalysed dehydrogenation from ethanol to
acetaldehyde and subsequent reductive amination of 4. We
therefore changed the solvent to THF which led to a yield of
91%. The reaction gives mainly the cis-isomer with the
content of the trans-isomer not being entirely reproducible
and varying between 0 and 15%. The catalysts Pd/C and Rh/
C did not efficiently reduce the nitrile.
2. Results and discussion
2.1. Synthesis
We decided to synthesise 3-substituted piperidines having a
2-methyl group, which should restrict the conformation of
the molecule by inducing unfavourable 1,3-diaxial inter-
action when axial. These molecules should prefer a chair
conformation with the methyl group equatorial and the 2,3-
cis and trans isomers would therefore have the 3-substituent
equatorial and axial, respectively. Here we rely on the work
of Lapuyade et al. who studied the conformation of various
methyl nipecotic acids and found that the methyl group
preferred to be equatorial.13
By careful chromatography in the polar solvent mixture
EtOAc–EtOH–Et3N 66:33:1 the cis- and trans-isomers of 4,
4a and 4b, could be separated. However, it was typically
more convenient to separate after protection of the amino
group. The piperidine 4 was protected with Boc anhydride/
Na2CO3 in the usual way giving 5 in 91% yield, and was
separated to 5a and 5b by chromatography. To obtain more
5b, the isomerization of 5a with LDA was performed giving
a product with a content of 5b up to 37% (Scheme 2).
The acid, 2-methylnipecotic acid 6, was prepared by
Lapuyade et al.13 as a mixture of cis and trans isomers.
We decided to modify their synthesis to obtain all three sets
of compounds with the anticipation that the stereoisomers
could be separated. Thus the known Michael reaction
between ethyl acetoacetate and acrylonitrile was used to
The cis-2-methylnipecotic acid 6a was made, in 72% yield,
by treatment of 4a with boiling water as described for the
mixture by Lapuyade et al. (Scheme 3).13 The cis isomer 5a,
was converted, in quantitative yield, to the acid 7a by
Scheme 2. Synthesis of esters 5a and 5b. The chiral compounds are racemic.

C. M. Pedersen, M. Bols / Tetrahedron 61 (2005) 115–122
117
Scheme 3. Synthesis of cis isomers 6a, 9a and 11a. All compounds are racemic.
hydrolysis with LiOH. By the use of Boc anhydride/
(NH4)HCO3 on the acid 7a, the primary amide 8a was
2.3. Conformation of the piperidines
15
obtained in 84% yield. Dehydration of 8a with oxalyl
chloride/Et3N16 gave up to 93% yield of the nitrile 10a.
Deprotection of 8a and 10a with HCl gave the amines 9a
and 11a, respectively.
As the conformation can influence the base-strength the
conformational preference of the six piperidines as
hydrochlorides and as free amines is important. If protona-
tion of the piperidine is associated with a conformational
change this may influence the base-strength and complicate
the interpretation. In the following discussion we will, for
simplicity, relate to the enantiomers shown in Schemes 3
and 4 despite the fact that the compounds are racemic. The
arguments are, of course, equally valid for the antipodes. As
mentioned previously, the conformation of 6a and 6b was
determined by Lapuyade et al. to be mainly 4C1 that is have
the methyl group equatorial.13 The trans-amide 9b is clearly
An identical sequence of reactions was carried out on the
trans-isomeric compounds 4b and 5b leading to the
corresponding acid, amide and nitrile 6b, 9b and 11b
(Scheme 4).
2.2. Configuration of the piperidines
1
also in the diequatorial C4 conformation both as hydro-
chloride and as free amine. This is readily seen by the large
The synthesised piperidines are all configurationally related
in two families of cis and trans compounds obtained from
the same two precursors 4a and 4b. Therefore determination
of the configuration of a single member of each family is
sufficient to establish the configuration of all. This
determination is readily done with 6a and 6b. In 6b J23 is
11 Hz showing that it must be diaxial coupling, which is
only consistent with the 2,3-trans configuration of this
molecule. In contrast in 6a J23 is 2.3 Hz which is consistent
with a axial-diequatorial coupling of the cis-configured
molecule.
J
23 proton coupling being 10 Hz in the amine and 10.8 Hz in
the hydrochloride. The cis-amide 9a is also in 4C1
conformation in both protonated and neutral form. This is
particularly evident from the H-3 proton which only has
small couplings and thus must be equatorial. It is also seen
that H-2 and H-3 has an essentially similar chemical shift,
respectively, in both the amino and conjugate acid forms
which confirm that they do not change from equatorial to
axial or vice versa. Finally the 2-methyl group has a 13C
Scheme 4. Synthesis of trans isomers 6b, 9b and 11b. All compounds are racemic.

118
C. M. Pedersen, M. Bols / Tetrahedron 61 (2005) 115–122
Table 1. pKa values of the piperidines
Number
Structure
pKa
Number
Structure
pKa
6a
11.0
9b
9.3
6b
9a
10.4
9.5
11a
11b
7.8
8.0
chemical shift of 16–19 ppm in 9a and 9b both as amines
and conjugated acids, in accordance with it being consist-
ently equatorial. An axial methyl group is 5–7 ppm more
shielded than an equatorial.17 The cis- and trans-nitriles 11a
and 11b were also found to be in the 4C1 conformation. The
13C spectra of the conjugate acids and amines quite clearly
shows the 2-methyl group has essentially the same chemical
shift as in 9a and 9b, and therefore must be predominantly
equatorial. Also the H-2 proton in 11a is found at a similar
chemical shift as was the case in 9a. In 11b J23 is found to be
10.4 Hz in the amine form and 11.2 Hz in the conjugate acid
form clearly consistent with 4C1 conformation for both
forms. So we conclude that the 2-methyl group serves its
purpose as an conformational anchor and that the piper-
idines 6a, 6b, 9a, 9b, 11a and 11b are predominantly in the
conformations as shown in Table 1.
surprisingly the DpKa for the amides 9a/9b is only 0.2,
despite the fact that the very similar ester group differs
largely in axial and equatorial substituent effect. Even more
remarkable is the observation that the very electron-
withdrawing and strongly dipolar nitrile group only has a
pKa difference of K0.2 for 11a/11b, the cis isomer being
slightly more basic.
It is interesting to note that the ss values of the equatorial
isomers are comparable in size or slightly smaller than the sI
values determined by Grob5 on 4-substituted quinuclidines
(Table 2). Evidently an equatorial ss value can be estimated
as 70–100% of a sI value. In contrast, in the axial position,
the substituent effect varies widely. While the CN and
CONH2 groups are similar to sI values for the carboxylate,
ester, hydroxyl and fluorine substituent a partial or complete
elimination of the electron withdrawing effect takes place.
2.4. Determination of base-strength
2.5. PM3 calculations
The pKa values of amino groups of 6a, 6b, 9a, 9b, 11a and
11b were determined by titration and are shown in Table 1.
The pKa values of the carboxylic acid residues of 6a and 6b
were found to 3.3 and 3.2, respectively. The data are the
average of three determinations (errorG0.1). The empirical
formula pKaZ10.7KSss10 was applied to calculate the ss
values for the new substituents, taking into account that the
ss value of an equatorial 2-Me is K0.1. This gave the values
shown in Table 2. The value for an equatorial 3-carboxylate
has previously determined to 0.5 and from that value the pKa
was calculated to 10.3 and thus is underestimated by 0.1.
To get more insight into these differences semi-empirical
calculations (MOPAC, PM3) were made on 3-substituted
piperidine, in ammonium and amine form, with the
substituent placed axial and equatorial. The results are
shown in Table 3. These calculations are valid in the gas
phase and their significance in solution must be interpreted
with caution. Nevertheless some general trends are evident
and consistent with the experimental facts. It is seen that
while the axial and equatorially substituted amines essen-
tially have similar heat of formation, the ammonium ions
generally differ with the equatorial isomer being less stable.
The smallest difference, 6 kJ/mol, is found for the cyano
substituted piperidine, which explains why a large pKa
The results are remarkable in that only the nipecotic acids
6a and 6b show a significant difference in the base-strength
of the amine. The difference (DpKa) between the pKa values
of 6a and 6b is 0.5 (or 0.7 between ss values) which appears
reasonable given the sizeable DpKa found previously for the
ester group in the 3-position of the piperidine.10 However,
Table 3. Heat of formation, in kJ/mol of monosubstituted piperidines
calculated using PM3, MOPAC in Chem3D 6.0 Ultra
Substituent
Amine
80.0
79.6
Ammonium
3-CN (axial)
3-CN (equatorial)
762.8
768.9
Table 2. ss values of substituents in the b or 3-position of piperidines
3-CONH2 (axial)
3-CONH2 (equatorial)
K222.1
K224.2
400.6
437.4
Substituent
ss value (b, equat.)
ss value (b, axial)
sI (Ref. 5)
3-COOK (axial)
K539.7
K538.8
K289.9
K189.0
COOK
CONH2
CN
COOMe
OH
0.5*
1.5
K0.2
1.3
0.58
1.78
3.04
1.7
1.68
2.57
3-COOK (equatorial)
3-COOMe (axial)
3-COOMe (equatorial)
K404.3
K404.3
245.5
273.2
2.8
3.0
1.2*
1.3*
2.3*
0.2*
0.5*
1.5#
3-OH (axial)
3-OH (equatorial)
K251.8
K253.1
399.7
413.6
F
3-F (axial)
3-F (equatorial)
K258.9
K256.4
411.0
423.6
The pKa of the piperidine is calculated using the formula pKaZ10.7KSds,
*value taken from Ref. 10, #value taken from Ref. 11.

C. M. Pedersen, M. Bols / Tetrahedron 61 (2005) 115–122
119
difference not is found between 11a and 11b. Also the
difference observed in the pKa of 6a and 6b is qualitatively
supported by the large difference observed in heat of
formation for the ammonium ions. On the other hand the
size of the pKa difference (0.5–0.7) is relatively small
compared to the huge energy difference calculated. Also the
clear difference in heat of formation between the protonated
amides is not very consistent with the insignificant
difference found in pKa (0.2) between 9a and 9b. This is
contrasted by the ester were a similar energy difference is
calculated between axial and equatorial isomer but a
significant pKa difference (1.0) is observed. It is proposed
that these inconsistencies may be caused by effects of
solvation of the carboxylate and amide groups reducing the
electrostatic interaction with the ammonium ion compared
to the gas phase.
The yield of 4 was 2.14 g (91%) and further purification was
not necessary unless the pure stereoisomers were required.
The cis/trans ratio (4a/4b) was normally 10/1 (as deter-
mined from 1H NMR) but varied between 1:0 and 6:1
depending on the scale. Separation of the 4a and 4b was
carried out by chromatography in EtOAc–EtOH 2:1
containing 1% Et3N added. Separation of 1.0 g of the
mixture gave 634 mg 4a and 74 mg 4b.
1
Compound 4a. Clear oil, H NMR (CDCl3, assigned with
COSY): d 4.1 (q, JZ7.3 Hz, 2H, –OCH2CH3), 3.02 (dt, JZ
13.6, 4.4 Hz, 1H, H6e), 2.9 (dq, JZ6.4, 3.5 Hz, 1H, H2a),
2.62 (dt, JZ10.2, 3.5 Hz, 1H, H6a), 2.48 (q, JZ3.5 Hz, 1H,
H3e), 2.05 (bs, 1H, N–H), 1.95 (dt, JZ5.8, 1.5 Hz 1H, H4e),
1.5–1.8 (m, 2H, H5e, H4a), 1.35 (m, 1H, H5a), 1.22 (t, JZ
7.3 Hz, 3H, –OCH2CH3), 1.08 (d, JZ7 Hz, 3H, –CH3). 13
C
NMR (CDCl3): 4a: dZ174.3 (C]O), 60.1 (–OCH2CH3),
52.5 (C3), 45.4 (C2), 44.3 (C6), 26.5 (C4), 22.9 (–OCH2
CH3), 19.2 (C5), 14.5 (–CH3).
3. Conclusion
1
Compound 4b. Clear oil, H NMR (CDCl3) (assigned with
It was found that the strong electron withdrawing power of
the cyano group is not significantly influenced by its axial or
equatorial orientation. The through space or solvent
component of the electron withdrawing effect must be
minor, and a charge dipole effect cannot be important. The
substituent effect from the amide group was also found
relatively independent on stereochemistry. In contrast the
substituent effect from a carboxylate group was significantly
different in the axial and equatorial positions.
COSY): d 4.1 (q, JZ7.3 Hz, 2H, –OCH2CH3), 3.05 (m, 1H,
H6a), 2.8 (dq, JZ10, 5.8 Hz, 1H, H2a), 2.68 (dt, JZ12,
2.6 Hz, 1H, H6a), 1.9–2.1 (m, 2H, H3a, N–H), 1.4–1.7
(m, 5H, H4e, H5e, H4a, H5a), 1.25 (t, JZ7.7 Hz, 3H,
–OCH2CH3), 1.06 (d, JZ5.8 Hz, 3H, –CH3). MS: Calcd for
C9H17NO2 (MCH)Z172.1338. Found: 172.1328.
4.1.3. cis and trans Ethyl N-tert-butoxycarbonyl-2-
methylnipecotate (5a and 5b). A mixture of 4a and 4b
(2.30 g, 13.4 mmol) was dissolved in 40 mL 50% aq THF.
Na2CO3 (1.46 g, 13.7 mmol) was added under stirring at
0 8C. A mixture of 3.32 g (15.3 mmol) di-tert-butyl
dicarbonate (boc-anhydride) and 1.43 g (16.1 mmol)
Na2CO3 in 30 mL 50% aq THF was added, and the mixture
stirred for 30 min at 0 8C. The reaction mixture was allowed
to reach room temperature and stirred for 2 h and acidified
with 1.0 M HCl. The mixture was extracted with CH2Cl2
and dried over MgSO4. After concentration a white
crystalline product of 5a and 5b was obtained. Yield:
3.32 g (91%, mixture of cis and trans). Compounds 5a and
5b could be separated using chromatography in CH2Cl2/
EtOAc 20:1 (see also below).
4. Experimental
4.1. General
13C-, 1H- and H,H-COSY NMR were recorded on a Varian
Gemini 200 (200 MHz) NMR instrument and when
specifically noted on a Mercury 400 (400 MHz) NMR
instruments. The spectra were referenced to solvent
residues. MS was recorded on a Micromass LC-TOF
instrument. Chromatography was performed in Merck 60
silica. TLC was performed on Merck silica 60 E254 coated
glass plates and developed using either vanillin (3 g in
100 mL EtOH with 1 mL H2SO4 added), potassium
permanganate (aq), Ce-mol (10 g Ce(IV)SO4 and 15 g
(NH4)2MoO4 in 1 L 10% H2SO4) or ninhydrin (2% in
n-BuOH) and subsequent heating.
Compound 5a. White crystals, mpZ72 8C, 1H NMR
(CDCl3): d 4.84* (bs, 0.5H, H2), 4.63* (bs, 0.5H, H2), 4.12
(q, JZ6.6 Hz, 2H, –CH2CH3), 3.99* (d, JZ11.6 Hz, 0.5H,
H6e), 3.86* (d, JZ13.2 Hz, 0.5H, H6e), 2.77 (dt, JZ11.6,
13.2 Hz, 1H, H6a)*, 2.60 (dt, JZ3.6, 12.8 Hz, 1H, H3e),
1.65–1.9 (m, 3H, H4e, H5e, H4a), 1.46 (s, 9H, Boc), 1.40 (m,
1H, H5a), 1.25 (t, JZ6.6 Hz, 3H, –CH2CH3), 1.03 (d, JZ
7.2 Hz, 3H, –CH3) (*Rotamers present).
4.1.1. Ethyl (2-cyanoethyl)-acetoacetate (3).14 This
product, a clear oil, was prepared as described in Ref. 14.
1H NMR (CDCl3): d 4.23 (q, JZ7.8 Hz, 2H, –OCH2CH3),
3.65 (t, JZ6.8 Hz, 1H, H2), 2.45 (t, JZ6.8 Hz, 2H, H6),
2.30 (s, 3H, H4), 2.2 (m, 2H, H5), 1.3 (t, JZ7.8 Hz, 3H,
OCH2CH3).
Compound 5b. White crystals, mpZ74 8C, 1H NMR
(CDCl3): d 4.86 (q, JZ6.4 Hz, 1H, H2e), 4.12 (dq, JZ2.4,
7.4 Hz, 2H, –CH2CH3), 3.92 (bd, JZ13.2 Hz, 1H, H6e),
2.79 (dt, JZ3.2, 13.2 Hz, 1H, H6a), 2.37 (bs, 1H, H3e), 2.02
(dd, JZ3.2, 14 Hz, 1H, H4e), 1.74 (tt, JZ4.0, 13.4 Hz, 1H,
H4a), 1.75 (m, 1H, H5e) 1.62 (tk, JZ4.0, 13.2 Hz, 1H, H5a),
1.43 (s, 9H, Boc), 1.24 (t, JZ6.8 Hz, 3H, –CH2CH3), 1.19
(d, JZ7.2 Hz, 3H, –CH3). MS: Calcd for C14H25NO4 (MC
Na)Z294.1681. Found 294.1680.
4.1.2. cis and trans Ethyl 2-methyl nipecotate (4a and 4b).
Compound 3. (2.52 g, 13.8 mmol) was dissolved in freshly
distilled dry THF, and Raney nickel (0.5 g, prewashed in
THF) was added. The mixture was transferred to an
autoclave with magnet, and the atmosphere replaced by
flushing with a stream of nitrogen. The container was closed
and a pressure of 45 atm. H2 was applied for 15 h at 80 8C.
The mixture was filtered through Celitew and concentrated.

120
C. M. Pedersen, M. Bols / Tetrahedron 61 (2005) 115–122
4.1.4. Epimerisation of 5a. A 50 mL oven dried round-
bottomed flask was purged with nitrogen and closed with
septum. Then 20 mL freshly distilled THF and dry
diisopropylamine (0.50 mL, 3.5 mmol) was injected. The
mixture was cooled, and n-BuLi (2.20 mL, 3.5 mmol, 1.6 M
in hexane) was added. The mixture was stirred on ice for
10 min. The LDA solution was cooled on dry ice/acetone
and 5a (800 mg, 3.0 mmol) in THF was added. The reaction
mixture was stirred for 1 h and quenched with water. The
mixture was acidified with 1.0 M HCl and extracted with
CH2Cl2. The yellowish product was separated by chromato-
graphy in CH2Cl2/EtOAc 20:1 giving 5b (160 mg, 27%) and
5a (428 mg, 47%).
or 7b (0.890 g, 3.27 mmol) was dissolved in 1,4-dioxane
(6 mL) and (0.125 mL, 1.5 mmol) pyridine was added drop-
wise. Boc anhydride (0.75 g, 3.4 mmol) was added and the
reaction was stirred for 15 min. Now ammonium hydro-
gencarbonate (0.27 g) was added and stirring continued for
18 h. EtOAc was added, the mixture separated and
the organic layer washed with 5% H2SO4 and water. The
organic phase was dried with MgSO4 and concentrated. The
products were thick yellowish syrups. After trituration with
ether 8a becomes crystalline and white. Yield: 0.745 g
(84%).
1
Compound 8a. White crystals, mpZ163–165 8C. H NMR
(CDCl3): d 5.65 (bs, 1H, –NH2), 5.5 (bs, 1H, –NH2), 4.64
(dq, JZ4.8, 6.4 Hz, 1H, H2e), 3.92 (bd, JZ13 Hz, 1H, H6e),
2.8 (dt, JZ3.2, 12 Hz, 1H, H6a), 2.5 (dt, JZ4.8, 11.2 Hz,
1H, H3e), 1.8 (dq, JZ13, 4.1 Hz, 1H, H4a), 1.6–1.9 (m, 2H,
H4e, H5e), 1.43 (s, 9H, Boc), 1.2–1.4 (m, 1H, H5a), 1.1 (d,
JZ6.6 Hz, 3H, –CH3). 13C NMR (CDCl3): d 175.7
(CONH2), 154.9 (Boc), 80.7 (–O–C(CH3)3), 48.3 (C2),
46.0 (C6), 38.3 (C3), 28.6 (O–C(CH3)3), 24.8 (C4), 20.6
(C5), 11.9 (–CH3). IR (KBr): 1685 (–COOtBu), 1625
(–CONH2), 2975/2944 (C–H sp3), 3393 (–CONH2). MS:
Calcd for C12H22N2O3 (MCNa)Z265.1528, Found:
265.1516.
4.1.5. cis- and trans-2-Methylnipecotic acid (6a and 6b).
These acids were prepared as described for the mixture in
Ref. 13. Compund 4a (133 mg) or 4b (74 mg) was boiled in
water for 5 h, concentrated and dried. This gave 6a
(115 mg) or 6b (65 mg), respectively. Yield: 95–100%.
1
Compound 6a. White solid, H NMR (D2O): d 3.4 (dt, JZ
1.9, 3.8 Hz, 1H, H6e), 3.3 (dq, JZ2.3, 6.8 Hz, 1H, H2a), 3.0
(m, 1H, H6a), 2.62 (bq, JZ2.3 Hz, 1H, H3e), 1.6–2.1 (m, 4H,
H4a, H4e, H5a, H5e), 1.32 (d, JZ6.8 Hz, 3H, H7). 13C NMR
(D2O): d 180.1 (C]O), 53.1 (C2), 44.0 (C3), 43.5 (C6), 24.7
(C4), 18.3 (C5), 15.6 (C7).
Compound 8b. White solid, mpZ165 8C. 1H NMR
(CDCl3): d 5.41 (bs, 1H, –NH2), 4.71 (q, JZ6.4 Hz, 1H,
H2e), 3.93 (bd, Jz13.5 Hz, 1H, H6e), 2.91 (dt, JZ4.0, 13.2
Hz, 1H, H6a), 2.35 (ddd, JZ2.0 Hz, 1H, H3e), 2.19 (bd,
Jz12.4 Hz, 1H, H4e), 1.85 (tt, JZ4.8, 12.8 Hz, 1H, H4a),
1.46–1.53 (m, 2H, H5e, H5a), 1.46 (s, 9H, Boc), 1.26 (d, JZ
6.8 Hz, 3H, –CH3). 13C NMR (CDCl3): d 177 (amide
C]O), 156 (Boc C]O), 80.5 (–O–C(CH3)3), 46.5 (C2), 45
(C6), 39 (C3), 29.5 (O–C(CH3)3), 22.4 (C4), 22 (C5), 16.5
(–CH3). IR (KBr): 1684 (–COOtBu), 1659 (–CONH2),
2975/2930 (C–H sp3), 3389 (–CONH2). MS: Calcd for
C12H22N2O3 (MCNa)Z265.1528. Found: 265.1526.
1
Compound 6b. White solid, H NMR (D2O): d 3.4 (m, 1H,
H6e), 3.2 (dq, JZ6.5, 11 Hz, 1H, H2a), 2.95 (dt, JZ3.2,
12.5 Hz, 1H, H6a), 2.28 (dt, JZ3.2, 11 Hz, 1H, H3a), 1.5–2.1
(m, 4H, H4a, H4e, H5a, H5e), 1.27 (d, JZ6.5 Hz, 3H, H7). 13
NMR (D2O): d 179.9 (C]O), 53.5 (C2), 50.2 (C6), 43.8
(C3), 26.3 (C4), 21.1 (C5), 16.8 (C7).
C
4.1.6. cis- and trans-N-tert-Butoxycarbonyl-2-methylni-
pecotic acid (7a and 7b). Compound 5a or 5b (1.69 g) was
dissolved in THF and treated with excess 2 M LiOH
solution. After 18 h at room temperature, the solution was
acidified with 1.0 M HCl. The mixture was extracted with
EtOAc, dried over MgSO4 and concentrated giving a white
solid of 7a or 7b. Yield: 1.50 g (100%).
4.1.8. cis- and trans-2-Methylpiperidine-3-carboxylic
amide (9a and 9b). Compound 8a or 8b was treated with
0.2 M HCl in excess and concentrated to dryness to give the
hydrochloride of 9a or 9b. To obtain the amine, the
hydrochloride was dissolved in water and subjected to ion-
exchange chromatography on Amberlite IR-120, HC. The
amine was eluted with 5% aq ammonia.
Compound 7a. White solid, mpZ178 8C. 1H NMR
(CDCl3): d 9.2 (bs, 1H, –COOH), 4.8 (bs, 1H, H2), 3.9
(bs, 1H, H6e), 2.8 (bs, 1H, H3), 2.65 (dt, JZ4.8, 11.9 Hz,
1H, H6a), 1.6–1.9 (m, 3H, H4e, H5e, H4a), 1.46 (s, 9H, Boc),
1.24 (dt, JZ1.6, 7.2 Hz, 1H, H5a), 1.08 (d, JZ6.7 Hz, 1H,
–CH3). IR (KBr): 1660 (–COOtBu), 1734 (–CONH2), 2979
(C–H sp3), 3000–2900 (small bands –COOH). MS: Calcd
for C12H21NO4 (MCNa)Z266.1368. Found: 266.1367.
Compound 9a. Clear syrup, 1H NMR (D2O): d 3.19 (dq, JZ
3.4, 6.6 Hz, 1H, H2a), 2.96 (dt, JZ4.2, 12.6 Hz, 1H, H6e),
2.74 (ddd, JZ3, 9.8, 12.6 Hz, 1H, H6a), 2.62 (q, JZ4.2 Hz,
1H, H3e), 1.85 (m, 1H, H4e), 1.48–1.8 (m, 3H, H4a, H5e,
H5a), 1.14 (d, JZ6.8 Hz, 3H, –CH3). (hydrochloride): dZ
3.42 (dq, JZ6.4, 3.1 Hz, 1H, H2a), 3.36 (m, 1H, H6e), 3.02
(dt, JZ4.6, 12 Hz, 1H, H6a), 2.84 (q, JZ3.1 Hz, 1H, H3e),
1.7–2.05 (m, 4H, H4e, H5e, H4a, H5a,), 1.32 (d, JZ6.4 Hz,
3H, –CH3). 13C NMR (CD3OD): d 179.0 (C]O), 53.7 (C2),
45.1 (C6), 43.5 (C3), 27.0 (C4), 21.1 (C5), 17.4 (–CH3).
(hydrochloride): d 178 (C]O), 54.4 (C2), 45.2 (C6), 42.0
(C3), 27.0 (C4), 19.5 (C5), 16.9 (–CH3). IR (KBr,
hydrochloride): 1608 (–CONH2), 1660 (–C]O), 2753/
2857 (R2NCH2), 2960 (C–H sp3), 3318/3151 (–CONH2).
MS: Calcd for C7H14N2O (MCH)Z143.1184, Found:
143.1080.
Compound 7b. White crystals, mpZ115 8C. 1H NMR
(CDCl3): d 4.82 (q, JZ6.8 Hz, 1H, H2e), 3.90 (d, JZ
12 Hz, 1H, H6e), 2.75 (dt, JZ3.2, 13.2 Hz, 1H, H6a), 2.38
(bs, 1H, H3e), 2.00 (bd, JZ13.2 Hz, 1H, H4e), 1.73 (tt, JZ
4.8, 13.6 Hz, 1H, H4a), 1.58 (tq, JZ4.0, 13.2 Hz, 1H, H5a),
1.45 (bd, JZ13.2 Hz, 1H, H5e), 1.38 (s, 9H, Boc), 1.16 (d,
JZ7.2 Hz, 3H, –CH3). IR (KBr): 1653 (–COOH), 1730
(COOtBu), 3000–2850 (small bands –COOH). MS: Calcd
for C12H21NO4 (MCNa)Z266.1368, Found: 266.1367.
4.1.7. cis- and trans-N-tert-Butoxycarbonyl-2-methylpi-
peridine-3-carboxylic amide (8a and 8b). Compound 7a

C. M. Pedersen, M. Bols / Tetrahedron 61 (2005) 115–122
121
Compound 9b. Clear syrup, 1H NMR (D2O): d 3.0 (bd, JZ
12.8 Hz, 1H, H6e), 2.74 (dq, JZ6.4, 10 Hz, 1H, H2a), 2.6 (dt,
JZ2.8, 12.4 Hz, 1H, H6a), 2.08 (ddd, JZ3.6, 10.0, 10.4 Hz,
1H, H3a), 1.9 (bdd, JZ2.8, 13.6 Hz, 1H, H4e), 1.72 (dt, JZ
2.8, 13.2 Hz, 1H, H5e), 1.55 (dq, JZ2.8, 13 Hz, 1H, H4a),
1.42 (tq, JZ2.8, 13 Hz, 1H, H5a), 1.02 (d, JZ6.4 Hz, 3H,
CH3). (hydrochloride): d 3.06 (bd, JZ12.8 Hz, 1H, H6e),
2.94 (dq, JZ6.4, 10.8 Hz, 1H, H2a), 2.65 (dt, JZ2.5,
12.4 Hz, 1H, H6a), 2.19 (dt, JZ3.6, 10.8 Hz, 1H, H3a), 1.69
(bd, JZ11.6 Hz, 1H, H4e), 1.62 (bd, JZ13.6 Hz, 1H, H5e),
1.39 (ddd, JZ3.2, 13.2 Hz, 1H, H5a), 1.32 (dq, JZ3.2,
11.4 Hz, 1H, H4a), 0.95 (d, JZ6.4 Hz, 3H, –CH3). 13C NMR
(CD3OD): d 180 (C]O), 53 (C2), 50 (C6), 45 (C3), 27 (C4),
23 (C5), 19 (–CH3). (hydrochloride): d 175.5 (C]O), 52.9
(C2), 43.9 (C6), 42.8 (C3), 24.7 (C4), 20.1 (C5), 15.6 (–CH3).
IR (KBr, hydrochloride): 1613 (–CONH2), 1667 (–C]O),
2807 (R2NCH2), 2950 (C–H sp3), 3368/3180 (–CONH2).
MS: Calcd for C7H14N2O (MCH): 143.1184. Found:
143.1086.
Compound 11a. Syrup, 1H NMR (D2O/CD3OD): d 3.2 (bd,
JZ12.2 Hz, 1H, H6e), 2.97 (m, 1H, H3e), 2.85 (dq, JZ6.4,
3 Hz, 1H, H2a), 2.65 (ddd, JZ4, 10.4, 12.2 Hz, 1H, H6a),
2.05 (bd, JZ13 Hz, 1H, H4e), 1.55–1.85 (m, 3H, H4a, H5e,
H5a), 1.24 (d, JZ6.4 Hz, 3H, –CH3). (hydrochloride) d 3.55
(dq, JZ3.7, 6.0 Hz, 1H, H2a), 3.45 (bd, JZ3.7 Hz, 1H, H3e),
3.42 (dd, JZ3, 13 Hz, 1H, H6e), 3.05 (dt, JZ3.6, 12 Hz, 1H,
H6a), 2.18 (m, 1H, H4e), 1.7–2.1 (m, 3H, H4a, H5e, H5a), 1.45
(d, JZ6.4 Hz, 3H, –CH3). 13C NMR (CD3OD): d 121.8
(–C^N), 52.9 (C2), 46.5 (C6), 35.0 (C3), 28.3 (C4), 22.9
(C5), 20.7 (–CH3). (hydrochloride): dZ118.7 (–C^N), 52.5
(C2), 45.1 (C6), 33 (C3), 26.2 (C4), 20.1 (C5), 17.7 (–CH3).
IR (KBr, hydrochloride): 2242 (–C^N), 2700–2800
(R2NCH2), 2940 (C–H sp3). For MS the N-acetate was
prepared by treatment of 8–10 mg sample with 2 mL acetic
anhydride/CH2Cl2 (1:1) for 18 h and evaporating. MS
(N-acetate): Calcd for C9H14NO (MCNa)Z189.1004,
Found: 189.1006.
Compound 11b. Syrup, 1H NMR (D2O): d 3.92 (dt, JZ2.4,
12.8 Hz, 1H, H6e), 2.75 (dq, JZ6.4, 10.4 Hz, 1H, H2a), 2.53
(dt, JZ2.8, 12.4 Hz, 1H, H6a), 2.35 (ddd, JZ3.6, 10.4,
10.4 Hz, 1H, H3a), 2.16 (m, 1H, H5e), 1.58–1.75 (m, 2H,
H4a, H4e), 1.33 (m, 1H, H5a), 1.20 (d, JZ6.4 Hz, 3H, –CH3).
(hydrochloride) d 3.56 (dq, JZ6.56, 11.2 Hz, 1H, H2a), 3.47
(bd, JZ12.9 Hz, 1H, H6e), 3.08 (dd, JZ3.6, 13.2 Hz, 1H,
H6a), 3.04 (dd, JZ3.6, 11.2 Hz, 1H, H3a), 2.34 (ddq, JZ1.6,
3.6, 13.6 Hz, 1H, H4e), 2.05 (dk, JZ3.6, 14.4 Hz, 1H, H5e),
1.95 (ddd, JZ3.6, 12, 12 Hz, 1H, H4a, 1.76 (m, 1H, H5a).
13C NMR (CD3OD): d 121.8 (–C^N), 52.6 (C2), 44.1 (C6),
35.0 (C3), 26.7 (C4), 24.6 (C5), 17.8 (–CH3). (hydrochlor-
ide): dZ118.7 (–C^N), 53.1 (C2), 42.7 (C6), 32.7 (C3),
25.0 (C4), 20.1 (C5), 15.9 (–CH3). IR (KBr): 2254 (–C^N),
2650–2750 (R2NCH2), 2939 (C–H sp3). For MS the
N-acetate was prepared by treatment of 8–10 mg sample
with 2 mL acetic anhydride/CH2Cl2 (1:1) for 18 h
and evaporating. MS (N-acetate): Calcd for C9H14NO
(MCNa)Z189.1004, Found: 189.1005.
4.1.9. cis- and trans-N-tert-Butoxycarbonyl-3-cyano-2-
methylpiperidine (10a and 10b). Compound 8a or 8b
(0.58 g, 2.4 mmol) was dissolved in DCM (7 mL) and
DMSO (0.27 mL, 0.30 g; 3.8 mmol) was added. The
mixture was cooled to K78 8C and oxalylchloride
(0.25 mL, 0.37 g; 2.88 mmol) was added. After 15 min
triethylamine (1.0 mL, 0.73 g; 7.19 mmol) was added
dropwise. After further 15 min of stirring, the reaction
was quenched with 15 mL water and extracted with three
times 15 mL EtOAc. The organic layer was washed with
concd NaCl solution, dried over MgSO4 and concentrated.
Purification by chromatography—eluent DCM/EtOAc
50:1—gave 480 mg 10a or 10b (93%).
Compound 10a. White crystals. mpZ61 8C. 1H NMR
(CDCl3): (assigned using Cosy) d 4.64 (dq, JZ3.5, 6.6,
6.6, 6.6 Hz, 1H, H2e), 3.95 (bd, JZ13.2 Hz, 1H, H6e), 2.80
(dt, JZ3.3, 13.2, 13.2 Hz, 1H, H6a), 2.76 (dt, JZ3.5, 3.5,
11.5 Hz, 1H, H3a), 1.9–2.1 (m, 1H, H4e), 1.90 (ddd, JZ3.5,
13.2, 13.2 Hz, 1H, H4a), 1.70 (ddd, JZ2.0, 3.3, 13.2 Hz, 1H,
H5e), 1.46 (s, 9H, Boc), 1.38 (m, 1H, H5a), 1.31 (d, JZ
6.6 Hz, 3H, –CH3) 13C NMR (CDCl3): d 155 (Boc C]O),
120.5 (–C^N), 80.5 (–O–C(CH3)3), 47 (C6), 37.5 (C2), 31.7
(C3), 28.6 (O–C(CH3)3), 24.5 (C5), 23.0 (C4), 12.4 (–CH3).
IR (KBr): 1690 (C]O), 2240 (–C^N), 2977 (C–H sp2).
MS: Calcd for C12H22N2O3 (MCNa)Z247.1422. Found:
247.1415.
4.1.11. Determination of pKa of piperidine hydrochlor-
ides. The piperidine hydrochloride (30 mg) was dissolved in
15–20 mL distilled water and subjected to titration with
0.1 M NaOH following the pH with a pH electrode. The pKa
was determined from the resulting titration curve, and was
the average of three determinations (errorG0.1).
Compound 10b. White crystals. mpZ69 8C. 1H NMR
(CDCl3): (assigned using Cosy) d 4.63 (q, JZ6.8 Hz, 1H,
H2e),3.99(bd,JZ14 Hz, 1H, H6e),2.75(dt,JZ3.5,13, 13 Hz,
1H, H6a),2.63(bs,1H, H3e),1.81(m, 1H, H4e),1.75(tt,JZ3.6,
13 Hz, 1H, H4a), 1.58 (m, 1H, H5e), 1.41 (s, 9H, Boc), 1.22 (m,
1H, H5a), 1.15(d, JZ6.8 Hz, 3H, –CH3). 13C NMR (CDCl3):d
155 (Boc C]O), 120.5 (–C^N), 80.5 (–O–C(CH3)3), 48
(C6), 38 (C2), 31.6 (C3), 28.5 (O–C(CH3)3), 21.8 (C5), 21.6
(C4), 15.6 (–CH3). IR (KBr): 1693 (C]O), 2239 (–C^N),
2977 (C–H sp3). MS: Calcd for C12H22N2O3 (MCNa)
247.1422, Found: 247.1430.
References and notes
1. Alder, R. W.; Baker, R.; Brown, J. M. Mechanisms in Organic
Chemistry; Wiley: London, 1971.
2. Hammett, L. P. Chem. Rev. 1935, 17, 125–136.
3. Taft, R. W. J. J. Am. Chem. Soc. 1957, 79, 1045–1049.
4. Topsom, R. D. Prog. Phys. Org. Chem. 1976, 12, 1–20.
5. Grob, C. A.; Schlageter, M. G. Helv. Chim. Acta 1976, 59,
264–276.
6. Grob, C. A.; Rich, R. Helv. Chim. Acta 1979, 62, 2793–2801.
7. Clark, J.; Perrin, D. D. Q. Rev. 1964, 18, 295–320.
8. Inouye, S. Chem. Pharm. Bull. 1968, 16, 1134–1137.
9. Jensen, H. H.; Lyngbye, L.; Bols, M. Angew. Chem., Int. Ed.
2001, 40, 3447–3449.
4.1.10. cis- and trans-3-Cyano-2-methylpiperidine (11a
and 11b). Compound 11a or 11b were prepared from 10a or
10b as described for 9 above.

122
C. M. Pedersen, M. Bols / Tetrahedron 61 (2005) 115–122
10. Jensen, H. H.; Lyngbye, L.; Jensen, A.; Bols, M. Chem. Eur.
J. 2002, 1218–1226.
14. Albertson, N. F. J. Am. Chem. Soc. 1950, 72, 2594–2597.
15. Pozdnev, V. F. Tetrahedron Lett. 1995, 36, 7115–7118.
16. Nakajima, N.; Ubukata, M. Tetrahedron Lett. 1997, 38,
2099–2102.
11. Bols, M.; Liang, X.; Jensen, H. H. Angew. Chem., Int. Ed.
2001, 40, 3447–3449.
12. Lankin, D. C.; Chandrakumar, N. S.; Rao, S. N.; Spangler,
D. P.; Snyder, J. P. J. Am. Chem. Soc. 1993, 115, 3356–3357.
13. Laupade, G.; Schlewer, G. S.; Wermuth, C.-G. Bull. Soc.
Chim. 1986, 4, 663–668.
17. For several examples see: Jensen, H. H.; Bols, M. Org. Lett.
2003, 5, 3201–3203. orDong, W.; Jespersen, T. M.;
Skrydstrup, T.; Bols, M.; Sierks, M. R. Biochemistry 1996,
35, 2788–2795.