Reaction of 2-alkyl pyridine N-oxide derivatives with Mosher's acyl chloride: First example of stereoselective Boekelheide rearrangement

Article abstract of DOI:10.1016/j.tetlet.2010.10.020

Treatment of 2-alkyl pyridine N-oxides with acylating reagents represents an established procedure for the introduction of oxygen functionality into alkyl group at the ortho position of N heteroaromatic rings. We have reported the first example of asymmet

Full text of DOI:10.1016/j.tetlet.2010.10.020

Tetrahedron Letters 51 (2010) 6526–6530
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Reaction of 2-alkyl pyridine N-oxide derivatives with Mosher’s acyl
chloride: first example of stereoselective Boekelheide rearrangement
⇑
⇑
Daniele Andreotti , Emanuele Miserazzi, Arnaldo Nalin, Alfonso Pozzan, Roberto Profeta, Simone Spada
Neurosciences CEDD Medicine Research Centre, GSK Spa, Via Fleming 4, 37135 Verona, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 July 2010
Revised 29 September 2010
Accepted 5 October 2010
Available online 27 October 2010
Treatment of 2-alkyl pyridine N-oxides with acylating reagents represents an established procedure for
the introduction of oxygen functionality into alkyl group at the ortho position of N heteroaromatic rings.
We have reported the first example of asymmetric Boekelheide rearrangement applied to a set of 2-alkyl-
pyridine N-oxide derivatives using (R) Mosher’s acyl chloride as activator of the rearrangement to give,
after hydrolysis, enantiomerically enriched 1-(2-pyridinyl)alkyl alcohol. Diastereoselectivity of the pro-
cess was studied at low temperatures in different solvents, and was supported by a preliminary in silico
modeling.
Dedicated to the memory of Alessandro
Fedeli-Bernardini
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Boekelheide rearrangement
Stereoselection
1-(2-Pyridinyl)-alkyl alcohol
Mosher’s acyl chloride
1. Introduction
such as TFAA4 and tosyl chloride5 give access to 1-(2-pyridinyl)-al-
kyl alcohol derivatives 5.
The development of new synthetic methods has reached a level
that allows organic chemists to plan and perform almost any kind
of functional group transformation. Nevertheless, despite this
achievement, advances in the field are still highly desirable partic-
ularly if they can solve problems that allow easy and rapid access
to chiral pool of compounds.
Today this is a well established procedure known as the
Boekelheide rearrangement,6 Scheme 1. Under the classical
Boekelheide’s reaction conditions, the 2-alkyl pyridine N-oxide 3
heated with a large excess of acetic anhydride is converted to 4,
generally in moderate-good yields, and after hydrolysis, the corre-
sponding alcohol 5 is obtained. This can then be resolved using
classical techniques7 to give enantiomers 5-S and 5-R.
Despite much research, the mechanism of the rearrangement
has not been fully understood, and remains controversial.6 The
most common and accepted explanation is an ion-pair mechanism
(see Scheme 2) proposed by Oae8 and Katritzky9c,d. However, we
cannot exclude the possibility that other mechanisms, such as a
hetero-Claisen rearrangement, in relation to the nature of the sub-
strate, could contribute to the Boekelheide rearrangement.
Until now, the Boekelheide rearrangement has found only lim-
ited application, and to the best of our knowledge, there have been
reports describing facial diastereoselective rearrangements medi-
ated by chiral promoters. With the aim of expanding the potential
of the Boekelheide rearrangement, we investigated the possible
development of an asymmetric version of the reaction using a chi-
ral activator.
An interesting and productive way to improve synthetic effi-
ciency, which can give access to a multitude of Csp3 chiral com-
pounds, is to take a new look at poorly exploited and old pioneer
reactions using current knowledge of organic chemistry. In this
manner, we re-investigated the so-called Boekelheide rearrange-
ment1 with the aim of obtaining an asymmetric version of the reac-
tion, and in this way expand the available tools for accessing at
enantiomerically enriched 1-(2-pyridinyl)alkyl alcohol derivatives.
Since its discovery in the late 1940s, the reactivity of pyridine
N-oxide mediated by activators, such as Ac2O has attracted the
interest of several research groups. The conversion of 1 to the cor-
responding 2-pyridone 2 with acetic anhydride performed by Kat-
ada, Scheme 1, was the first example reported in this field.2
Shortly thereafter, it was found that also activation of 2-alkyl
pyridine N-oxides 3 either with Ac2O3 or with other promoters,
The easily available, chemically stable and enantiomerically
pure (R) and (S) Mosher’s acyl chlorides were chosen as activators,
while the 2-alkyl pyridines 8a, 9a, 10a, and 11a derivatives
depicted in Scheme 3 where chosen as substrates. With the
⇑
Corresponding authors. Tel.: +39 0458219953; fax: +39 0458218196 (D.A.).
E-mail addresses: daniele.andreotti@aptuit.com (D. Andreotti), simone.spa-
da@aptuit.com (S. Spada).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.10.020

D. Andreotti et al. / Tetrahedron Letters 51 (2010) 6526–6530
6527
Ac2O
+
N
Reflux
N
H
O
O
-
1
2
Ac2O
+
N
R
R
R
R
R
N
O
N
N
Hydrolysis
N
Reflux
O
OH
O
OH
OH
-
4
5-S
5-R
3
5
Scheme 1. Katada and Boekelheide rearrangements.
+
R
N
Ion pair
O
O
-
7
Ac2O
R
+
N
+
N
R
R
N
O
O
O
O
O
R
-
N
O
3
6
O
4
Anhydrobase
Hetero-Claisen
Scheme 2. Boekelheide rearrangement: ion-pair versus intramolecular pathway.
well below 0 °C. Table 1 summarizes the results obtained.9 The
high reactivity exhibited by the (R)-(À)MTPA-Cl during the
Boekelheide reaction allowed performance of the reaction at tem-
peratures down to À78 °C, entry 5. Interestingly, ester 9c was ob-
tained in 67% molar yield using 2-propanol as the solvent with
an 88:12 diastereomeric ratio with respect to the major isomer.
Surprisingly, 2-propanol was found to be fully compatible with
(R)-(À)MTPA-Cl in the presence of TEA and no formation of the re-
lated i-Pr MTPA ester was detected when the reaction was carried
out at T 6 À30 °C.
N
N
N
N
8a
9a
10a
11a
Scheme 3. 2-Alkyl pyridine substrates.
exception of 10a, all these pyridine derivatives are commercially
available.
When compound 10b was reacted under the same conditions
adopted in Scheme 5 the expected product 10c was not observed
in spite of the complete consumption of the starting material
10b. In that case, the main side products detected were pyridine
10a along with its oxidized form, likely corresponding to the tri-
substituted double bond 14, Scheme 6. We assumed that it could
derive from spontaneous elimination of the ester 10c or proton loss
of intermediate 7 like, Scheme 2.
Since no suitable route has been reported in the literature for
compound 10a, we envisaged the possibility to prepare it starting
from 10d following the chemical route reported in Scheme 4.
Compounds 8a, 9a, 10a, and 11a were converted to the corre-
sponding N-oxides, respectively, 8b, 9b, 10b, and 11b in high yields
according to literature procedures9 by means of standard mCPBA
conditions or by aqueous hydrogen peroxide in the presence of a
catalytic amount of methyl-trioxo-rhenium, Scheme 5.
Saponification with LiOH of 8c, 9c, and 11c led to free 1-(2-
pyridinyl)alkyl alcohol derivatives 8d, 9d, and 11d. Analysis of
their absolute stereochemistry by ORP9a,b in comparison with the
literature data revealed that all the major isomers of 8c, 9c, and
11c have absolute stereochemistry S configuration when (R)-
(À)MTPA-Cl was used as the promoter. To further support this re-
sult, the same reaction was carried out on 9b with (S)-(À)MTPA-Cl
producing, as expected, mainly the R isomer of 9c with the same
diastereoselectivity ratio in entry 3, Table 1.
We found that (R)-(À)-
a-methoxy-a-(trifluoromethyl) phenyl-
acetyl chloride, (R)-(À)MTPA-Cl, is an extremely efficient agent
for the Boekelheide rearrangement. In fact, treatment of 8b, 9b,
and 11b in ethyl acetate with (R)-(À)MTPA-Cl in the presence of
TEA as the base, allowed the fast and complete rearrangement to
the corresponding esters 8c, 9c, and 11c at reaction temperatures
ii)
i)
iii)
Additionally, influence of solvent on the diastereoselection was
investigated using 9b as a model substrate, and results are summa-
rized in Table 2. In this case only a moderate variation in terms of
diastereoselectivity was detected when moving from a polar protic
solvent, such as 2-propanol (D2/D1 75:25) to apolar solvents, such
as toluene (D2/D1 67:33). However, we noticed that the solubility
of the reaction mixture highly affects the kinetics as in the case of
N
N
N
N
OH
O
10d
12
13
10a
Scheme 4. Synthesis of bicyclic pyridine derivative 10a. Reagents and conditions:
(i) Dess–Martin periodinane, DCM, y 85%; (ii) (Ph)3P–CH3Br, tBuOK, THF, y 50%; (iii)
Pd/C, EtOH, H2, y 80%.

6528
D. Andreotti et al. / Tetrahedron Letters 51 (2010) 6526–6530
R3
R3
R1
R3
R1
i)
8c, 9c, 10c, 11c
ii)
R2
R1
+
R2
R2
N
N
N
O
O
-
N
N
N
N
OH
8a, 9a, 10a, 11a
OH
OH
HO
8b, 9b, 10b, 11b
O
CF3
Target
8d
9d
10d
11d
iii)
O
R1
R2
R3
H
H
CH2
Me
CH2
H
CH2
R3
Me
H
CH2
CH2
CH2
R2
R1
N
8d, 9d, 10d, 11d
OH
Scheme 5. Rearrangement of different N-oxide substrates promoted by (R)-Mosher’s acyl chloride and target correlation table. Reagents and conditions: (i) m-CPBA or H2O2/
CH3ReO3 cat.; (ii) solvent, TEA, (R)-Mosher’s acyl chloride, low T; (iii) LiOH, THF/water, 70 °C.
Table 1
Reaction of different N-oxide substrates with (R)-Mosher’s acid chloride via Scheme 59
Entry
Substrate
Target
Solvent
T (°C)
Time (h)
Conversiona
Yieldb%
D2:D1 ratioc
Comments
1
2
3
4
5
6
7
8
8b
8b
9b
9b
9b
11b
11b
10b
8c
8c
9c
9c
9c
11c
11c
—
2-Propanold
Ethyl acetate
Ethyl acetate
Ethyl acetate
2-Propanold
Ethyl acetate
2-Propanold
Ethyl acetate
À30
<1
22
22
8
4
4
Quant.
Quant.
Quant.
80%
Quant.
Quant.
Quant.
0%
69
—
70
—
67
61
—
78:22
71:29
70:30
80:20
88:12
72:28
78:22e
—
i-Propyl ether in traces
À30
À30
À78
À78
i-Propyl ether in traces
i-Propyl ether in traces
À30
À78
2.5
À30/rt
0
a
b
c
Conversion from HPLC a/a TM/SM.
Isolated yield as sum of D1 + D2.
From HPLC D1/D2 a/a.
d
e
No i-propyl Mosher’s acid ester was detected.
19F NMR is consistent with HPLC data.
R3
i)
10b
R2
R1
N
N
10a
N
14
O
8e, 9e, 11e
Scheme 6. Reactivity of 10b. Reagents and conditions: (i) solvent, TEA, (R)-
Mosher’s acyl chloride, À30 °C/rt.
Figure 1. Proposed side products observed using 2-propanol as the solvent.
Table 2
Effect of the solvent on diastereoisomeric ratioa
Entry
Solvent
Temperature (°C)
D1:D2b
N
1
2
3
4
5
6
7
Ethyl acetate
THF
Toluene
DCM
DMF
2-Propanol
MeOH
À30
À30
À30
À30
À30
À30c
À78
70:30
70:30
67:33
63:37
70:30
75:25
17
18
+
+
N
Br
N
O
-
16
15
N
d
—
Scheme 7. Boekelheide rearrangement on neopentyl derivative 16.
a
Screening was conducted on Substrate 9b under the following conditions. TEA
(3 equiv), (R)-Mosher’s acid chloride (2 equiv), C = 0.025 M.
b
From HPLC D1/D2 at 2.5 h reaction time.
i-Pr ether in traces.
No rearrangement reaction.
In fact, conversion time of 9b to 9c was reduced from 22 h at
À30 °C down to 4 h at À78 °C changing solvent from ethyl acetate
to 2-propanol with a moderately positive effect on the diastereose-
lectivity, see entries 3 and 5 Table 2 (D2/D1 70:30; D2/D1 88:12).
Similar behavior was also observed with substrates 8b and 11b,
entry 1 and 7 Table 1. The relatively low influence of solvent polar-
ity on selectivity shown in Table 2 provides little support for a
complete ion-pair pathway (Scheme 2): 2-propanol only afforded
an 8 mol % increase with respect to toluene (75:25 vs 67:33).
It is worth noting that when using 2-propanol as the solvent,
entries 1, 5, and 7 Table 1, traces of a compound incorporating 2-
propanol, which could fit with the corresponding i-propyl ethers
8e, 9e, and 11e were detected by LC–MS (Fig. 1).
c
d
toluene, where the reaction was found to be very slow. Moreover,
as expected MeOH, was not a suitable solvent even at À78 °C, the
reaction with acyl chloride to form the corresponding ester was too
fast with respect to the Boekelheide reaction.
Ethyl acetate and 2-propanol turned out to be the solvents of
choice in terms of conversion and diastereomeric ratio. In particular,
2-propanol accelerated the reaction kinetics with respect to ethyl
acetate affording almost the same range of molar yield, Table 1.

D. Andreotti et al. / Tetrahedron Letters 51 (2010) 6526–6530
6529
Figure 2. Putative transition states for Boekelheide rearrangement of 9b promoted by (R)-(À)MTPA-Cl; on the left, the transition state leading to the (S) isomer, on the right,
the transition state leading to the (R) isomer.
Intrigued by these results we decided to repeat the same exper-
iment carried out by Katritzky9c,d (16 treated with Ac2O) to eluci-
date the mechanism of the Boekelheide rearrangement, Scheme
7. In that case 16, prepared form 15,9c underwent a [1,2] methyl
shift to give alkene 18, corroborating the cationic hypothesis for
the reaction.
In our case, using the procedure described in Scheme 5, only
degradation side products were observed where the main
compound obtained was pyridine 17; however, no trace of 18
was detected.
reaction, through different techniques HPLC, 1H or 19F NMR, pro-
viding then a further reason for using the chiral Mosher’s acyl chlo-
ride as the activator.
We believe that future investigations should be focused on
the design of alternative quaternary chiral acids, possibly driven
by a quantum mechanics analysis, which would be capable of
activating 2-alkyl pyridine N-oxide derivatives with increased
diastereoselectivity.
Acknowledgments
Explanation of the diastereselectivity encountered in the reac-
tion of chiral (R)-MTPA-Cl with 8b, 9b, and 11b would be straight-
forward considering a sigmatropic rearrangement.
However, the facial selectivity could be explained by different
energetic levels associated with the transient diastereomeric inter-
mediates both in the case of hetero-Claisen rearrangement10, or
considering a tight ion-pair pathway. For this reason we deter-
mined the geometry of the two diastereoisomeric transition states
using quantum mechanics calculations.11
We are grateful to GSK Analytical Chemistry Department in
Verona for the experimental collaboration and in particular to Dr.
Ornella Curcuruto for High Resolution Mass generation and to Dr.
Luca Rovatti for the effective LC–MS support.
References and notes
1. (a) Katritzky, A. R.; Lam, J. N. Heterocycles 1992, 33, 1011–1049; b Katritzky, A.
R.; Lagowski, J. M. Chemistry of Heterocyclic N-Oxide; Academic Press: NY, 1971.
2. (a) Katada, M. J. Pharm. Soc. Jpn. 1947, 67, 15–19; (b) McKillop, A.; Bhagrath, M.
K. Heterocycles 1985, 23, 1697–1701; (c) Ochiai, E. J. Org. Chem. 1953, 18, 534–
551.
3. (a) Boekelheide, V.; Linn, W. J. J. Am. Chem. Soc. 1954, 76, 1286–1291; (b)
Boekelheide, V.; Harrington, D. L. Chem. Ind. 1955, 1423–1424; (c) Boekelheide,
V.; Lehn, W. L. J. Org. Chem. 1961, 26, 428–430.
4. Fontenas, C.; Bejan, E.; Haddou, H. A.; Balavoine, G. G. A. Synth. Commun. 1995,
25, 629–633.
5. Sledeski, A. W.; O’Brien, M. K.; Truesdale, L. K. Tetrahedron Lett. 1997, 38, 1129–
1132.
Having well in mind12 that (R)-MTPA-Cl derives from (S)-
(À)Mosher’s acids, it is worth noting that the geometries observed
were apparently characterized by the steric hindrance of the sub-
stituent of the (S)-(À)Mosher’s acids, in the transition states. In
the lowest minimum transition state, the methoxy group of the
acid is always facing the benzylic proton at position 2 of the pyri-
dine ring (R1, Scheme 5). From the enthalpy point of view, we
found that the transition state leading to the (S) isomer is slightly
more stable than the one leading to the (R) isomer (ꢀ1.1 kcal) (see
Fig. 2), which is in good agreement with the experimental findings.
6. Li, J. Name Reactions in Heterocyclic Chemistry; John Wiley & Sons: NJ, 2005.
7. (a) Seemayer, R.; Schneider, M. P. Tetrahedron: Asymmetry 1992, 3, 827–830.
and references cited therein; Nakamura, K.; Fujii, M.; Ida, Y. Tetrahedron:
Asymmetry 2001, 12, 3147–3153.
8. (a) Oae, S.; Kitao, Y. J. Am. Chem. Soc. 1962, 84, 3359–3362; (b) Oae, S.; Kozuka,
S. Tetrahedron 1964, 20, 2671–2676; (c) Tamagaki, S.; Kozuka, S.; Oae, S.
Tetrahedron Lett. 1968, 4765–4768.
9. Typical experimental procedures: Pyridines N-oxides preparation: Compounds 8b,
9b, 10b and 11b were synthesized according to: (a) Kaiser, S.; Smidt, S. P.;
Pfaltz, A. Angew. Chem. Int. Ed. 2005, 45, 5194–5197; (b) Drury, W.;
Zimmermann, N.; Keenan, M.; Hayashi, M.; Kaiser, S.; Goddard, R.; Pfaltz, A.
Angew. Chem., Int. Ed. 2003, 43, 70–74; intermediate 16 was synthesized
according to: (c) Bodalski, R.; Katritzky, A. R. Tetrahedron Lett. 1968, 257–260;
(d) Bodalski, R.; Katritzky, A. R. J. Chem. Soc. (B) 1968, 831–838.
2. Conclusions
We have reported the first example of asymmetric Boekelheide
rearrangement applied to a set of 2-alkyl-pyridine N-oxide deriva-
tives 8a, 9a, and 11a using (R) Mosher’s acyl chloride as activator of
the rearrangement to obtain enantiomerically enriched 1-(2-pyrid-
inyl)alkyl alcohol derivatives 8d, 9d, and 11d. The reaction was ob-
served to work effectively across a range of low temperatures from
À30 to À78 °C, though only a limited effect of temperature was ob-
served on the diastereomeric ratio. Interestingly, the absolute ste-
reochemistry of the major isomers 8d, 9d, and 11d was S in all
cases when using (R)-MTPA-Cl. We noticed that steric hindrance
close to the reaction center had a detrimental effect on the
Boekelheide rearrangement, substrates 10b and 16. Although, at
present it is possible to acquire even large amounts of chiral
Mosher’s acids at an acceptable price, it is possible to recover the
chiral Mosher’s acids after hydrolysis without losing optical purity.
Additionally, it is worth pointing out that the resulting Mosher es-
ter per se can be used to evaluate the diastereomeric ratio of the
Boekelheide rearrangement: preparation of 1-(2-pyridinyl)ethyl (2S)-3,3,3-
trifluoro-2-(methyloxy)-2-phenylpropanoate 8c (entry 1, Table 1). To a solution
of 2-ethylpyridine 1-oxide 8b (100 mg, 0.812 mmol) and TEA (0.340 mL,
2.436 mmol, 3 equiv) in isopropanol (4 mL) under argon at À30 °C, (R)-
Mosher’s acyl chloride (0.304 mL, 1.624 mmol, 2 equiv) dissolved in 0.5 mL of
ethyl acetate was added dropwise. The resulting slurry, become thick with
formation of yellow solid, was left reacting at À30 °C. After 10 min LC–MS
check showed reaction almost complete. After further 40 min, the reaction
mixture was taken up with EtOAc/NaHCO3 saturated solution/water. The
phases were separated and the aqueous one was back extracted with EtOAc.
Combined organics were dried over Na2SO4 and evaporated to dryness to get
crude material (262 mg as yellow foam). LC–MS showed D2/D1 ratio 22:78
(labels 1 and 2 assigned according to elution order). Crude material was
purified by SiO2 flash chromatography targeting mixture of D1 and D2, eluting
with cyclohexane/EtOAc from 95:5 to 60:40 (Rf 0.3 cyclohexane/EtOAc 8:2,

6530
D. Andreotti et al. / Tetrahedron Letters 51 (2010) 6526–6530
isomers D1and D2 did not separate on SiO2 TLC). Evaporation of solvent
afforded title material (190 mg, 0.560 mmol, 69.0% yield) as pale yellow thick
oil. sample of this material was purified by preparative LC–MS for
trifluoro-2-(methyloxy)-2-phenylpropanoate (190 mg, 0.541 mmol, 36.5%
yield) as brownish oil.
Preparation of 5,6,7,8-tetrahydro-8-quinolinyl (2S)-3,3,3-trifluoro-2-(methyloxy)-
A
characterisation purpose to get 2 fractions:
2-phenylpropanoate 11c (entry 6, Table 1). To a solution of 5,6,7,8-
—D1: (1R)-1-(2-Pyridinyl)ethyl (2S)-3,3,3-trifluoro-2-(methyloxy)-2-phenylpro-
tetrahydroquinoline 1-oxide 11b (200 mg, 1.341 mmol) and TEA (0.561 mL,
4.02 mmol) in dry ethyl acetate (8 mL) under argon at À30 °C, (R)-Mosher’s
acyl chloride (0.502 mL, 2.68 mmol) dissolved in 1 mL of ethyl acetate was
added dropwise. The resulting thick slurry was left reacting at À30 °C for 4 h.
LC–MS check showed reaction complete and a D1/D2 ratio 28:72 (labels 1 and
2 assigned according to elution order). Reaction mixture was taken up with
EtOAc/water/NaHCO3 saturated solution. The phases were separated and the
aqueous one was back extracted with EtOAc. Combined organics were dried
over Na2SO4 and evaporated to dryness to get crude material (1.1 g as brown
foam) that was purified by SiO2 flash chromatography eluting with
panoate as colourless thick oil. 1H NMR (400 MHz, CDCl3):
d 1.71 (d,
J = 6.53 Hz, 3H), 3.60 (s, 3H), 6.17 (q, J = 6.69 Hz, 1H), 7.13–7.24 (m, 1H),
7.35–7.44 (m, 4H), 7.50–7.57 (m, 2H), 7.62 (td, J = 7.72, 1.63 Hz, 1H), 8.56 (d,
J = 4.27 Hz, 1H); 13C NMR (100 MHz, CDCl3): d 20.4, 55.5, 75.4, 84.7, 120.0,
120.5, 122.9, 127.4, 128.4, 129.6, 132.2, 136.8, 149.2, 159.2, 165.6; HRMS (ES+)
calcd for C17H16F3NO3 [M+H]+: 340.1161, found: 340.1162 (
D
0.3 ppm).
—D2: (1S)-1-(2-Pyridinyl)ethyl (2S)-3,3,3-trifluoro-2-(methyloxy)-2-phenylpro-
panoate as colourless thick oil. 1H NMR (400 MHz, CDCl3):
1.65 (d,
d
J = 6.53 Hz, 3H), 3.55 (s, 3H), 6.17 (q, J = 6.53 Hz, 1H), 7.24 (m, 1H), 7.33–7.45
(m, 4H), 7.55 (d, J = 6.27 Hz, 2H), 7.66–7.73 (m, 1H), 8.60 (d, J = 4.52 Hz, 1H);
13C NMR (100 MHz, CDCl3): d 20.4, 55.4, 75.5, 85.1, 120.5, 123.0, 124.8, 127.5,
128.4, 129.6, 132.2, 136.9, 149.3, 159.1, 165.8; HRMS (ES+) calcd for
cyclohexane/EtOAc from 95:5 to 1:1. Evaporation of solvent afforded
fractions:
3
—High running spot (TLC Rf 0.55 cyclohexane/EtOAc 7:4, LC–MS D2).
C
17H16F3NO3 [M+H]+: 340.1161, found: 340.1147 (
D
À4.1 ppm).
(8S)-5,6,7,8-Tetrahydro-8-quinolinyl
(2S)-3,3,3-trifluoro-2-(methyloxy)-2-
Preparation of 6,7-dihydro-5H-cyclopenta[b]pyridin-7-yl (2S)-3,3,3-trifluoro-2-
(methyloxy)-2-phenylpropanoate 9c (entry 3, Table 1). To a solution of 6,7-
dihydro-5H-cyclopenta[b]pyridine 1-oxide 9b (200 mg, 1.480 mmol) and TEA
(0.619 mL, 4.44 mmol, 3 equiv) in dry ethyl acetate (6 mL) under argon at
À30 °C, (R)-Mosher’s acyl chloride (748 mg, 2.96 mmol, 2 equiv) dissolved in
1 mL of dry ethyl acetate was added dropwise. The resulting slurry, become
immediately thick and dark brown, was left reacting at À30 °C for 22 h. LC–MS
check at 22 h showed reaction complete and a ratio D1/D2 = 30:70 (labels 1
and 2 assigned according to elution order). Reaction mixture was taken up with
EtOAc/NaHCO3 saturated solution/water. The phases were separated and the
aqueous one was back extracted with EtOAc. Combined organic phases were
dried over Na2SO4 and evaporated to dryness to get crude material (1.06 g as
brownish solid) that was purified by SiO2 flash chromatography eluting with
phenylpropanoate (140 mg, 0.383 mmol, 28.6% yield) as colourless thick oil.
19F NMR (376 MHz, CDCl3): d À71.68; 1H NMR (400 MHz, CDCl3): d 1.79–1.86
(m, 2H), 2.03–2.13 (m, 1H), 2.15–2.23 (m, 1H), 2.69–2.87 (m, 2H), 3.62 (s, 3H),
6.31 (t, J = 4.42 Hz, 1H), 7.18 (dd, J = 7.71, 4.67 Hz, 1H), 7.34–7.41 (m, 3H), 7.44
(d, J = 7.58 Hz, 1H) 7.59–7.67 (m, 2H), 8.51 (d, J = 4.04 Hz, 1H); 13C NMR
(100 MHz, CDCl3): d 18.0, 28.2, 28.8, 55.6, 73.0, 121.9, 123.4, 124.9, 127.6,
128.2, 129.4, 132.7, 133.8, 137.1, 147.7, 152.2, 165.9; HRMS (ES+) calcd for
C
19H18F3NO3 [M+H]+: 366.1317, found: 336.1323 (
D 1.6 ppm).
—Low running spot (TLC Rf 0.5 cyclohexane/EtOAc 7:4, LC–MS D1).
(8R)-5,6,7,8-Tetrahydro-8-quinolinyl (2S)-3,3,3-trifluoro-2-(methyloxy)-2-pheny-
lpropanoate (57 mg, 0.156 mmol, 11.64% yield) as pale yellow oil. 19F NMR
(376 MHz, CDCl3): d À72.15; 1H NMR (400 MHz, CDCl3): d 1.87–2.06 (m, 2H),
2.12–2.22 (m, 1H), 2.25–2.34 (m, 1H), 2.75–2.92 (m, 2H), 3.56 (s, 3H), 6.34 (t,
J = 4.67 Hz, 1H), 7.15 (dd, J = 7.71, 4.67 Hz, 1H), 7.38–7.41 (m, 3H), 7.43 (d,
J = 7.83 Hz, 1H), 7.66 (dd, J = 6.32, 2.78 Hz, 2H), 8.45 (d, J = 4.55 Hz, 1H); 13C
NMR (100 MHz, CDCl3): d 18.8, 28.5, 29.4, 55.7, 73.3, 122.24, 123.6, 125.1,
128.3, 128.4, 129.7, 132.6, 133.7, 137.3, 148.0, 152.5, 166.0; HRMS (ES+) calcd
cyclohexane/EtOAc from 95:5 to 6:4. Evaporation of solvent afforded
fractions:
3
—High running spot (TLC Rf 0.6 cyclohexane/EtOAc 6:4, LC–MS D2): (7S)-6,7-
dihydro-5H-cyclopenta[b]pyridin-7-yl (2S)-3,3,3-trifluoro-2-(methyloxy)-2-phe-
nylpropanoate (130 mg, 0.370 mmol, 25.01% yield) as colourless oil. 1H NMR
(500 MHz, DMSO-d6): d 1.85–2.02 (m, 1H), 2.62 (m, J = 13.70, 6.10 Hz, 1H),
2.84–3.00 (m, 2H), 3.52 (s, 3H), 6.43 (m, J = 7.60, 4.70 Hz, 1H), 7.33 (m, J = 7.60,
4.70 Hz, 1H), 7.35–7.58 (m, 5H), 7.75 (d, J = 8.15 Hz, 1H), 8.49 (d, J = 4.66 Hz,
1H); 13C NMR (100 MHz, DMSO-d6): d 27.2, 29.4, 55.2, 79.0, 83.9, 123.3, 123.9,
127.0, 128.5, 129.8, 131.7, 133.4, 137.5, 148.6, 158.9, 165.3; HRMS (ES+) calcd
for C19H18F3NO3 [M+H]+: 366.1317, found: 336.1314 (
—Mixed fractions.
D
À0.8 ppm).
5,6,7,8-Tetrahydro-8-quinolinyl
(2S)-3,3,3-trifluoro-2-(methyloxy)-2-phenylpr-
opanoate (100 mg, 0.274 mmol, 20.42% yield) as a colourless thick oil.
10. Dalko, P. I.; Langlois, Y. Tetrahedron Lett. 1998, 39, 2107–2110.
11. Initial guess for the transition states were calculated using the transition state
module available within Spartan’02 (www.wavefun.com). STO-3G full
transition state optimization was then carried out until convergence.
Vibration analysis was also carried out to check consistency amongst the
imaginary frequency and the transition state path. The transition state leading
to the (S) isomer had a total energy of À1251.8706985 au, while the one
for C18H16F3NO3 [M+H]+: 352.1161, found: 352.1162 (
D 0.3 ppm).
—Low running spot (TLC Rf 0.53 cyclohexane/EtOAc 6:4, LC–MS D1): (7R)-6,7-
dihydro-5H-cyclopenta[b]pyridin-7-yl (2S)-3,3,3-trifluoro-2-(methyloxy)-2-phen-
ylpropanoate (43 mg, 0.122 mmol, 8.27% yield) as white solid. 1H NMR
(500 MHz, DMSO-d6): d 2.04–2.20 (m, 1H), 2.58–2.73 (m, 1H), 2.83–3.11 (m,
2H), 3.48 (s, 3H), 6.38–6.45 (m, 1H), 7.27–7.35 (m, 1H), 7.45 (m, J = 2.30 Hz,
5H), 7.74 (d, J = 1.00 Hz, 1H), 8.44 (d, J = 1.00 Hz, 1H); 13C NMR (100 MHz,
DMSO-d6): d 27.4, 29.4, 55.1, 79.0, 84.0, 123.7, 124.6, 127.3, 128.4, 129.8, 131.6,
133.5, 137.6, 148.6, 159.1, 165.1; HRMS (ES+) calcd for C18H16F3NO3 [M+H]+:
leading to the (R) isomer had
difference in kcal/mol is 1.11 kcal in favour of the transition state leading to the
(S) isomer.
a
total energy of À1251.8689235 au. The
12. Attention must be paid to the change of the R and S label when the chiral
Mosher’s acid is converted to the corresponding chiral Mosher’s acyl chloride.
352.1161, found: 352.1167 (D 1.7 ppm).
—Mixed fractions: 6,7-dihydro-5H-cyclopenta[b]pyridin-7-yl (2S)-3,3,3-