Highly stereoselective chlorination of β-substituted cyclic alcohols using PPh3-NCS: Factors that control the stereoselectivity

Article abstract of DOI:10.1039/b614512d

A variety of trans-β-substituted cyclic alcohols were stereoselectively chlorinated to either the corresponding cis-chloride or trans-chloride (inversion or retention of configuration) with good to excellent yields; the stereochemical outcome is determined by the size of the ring and the nature of the β-substituents, especially the electronegativity of the substituted atom. The Royal Society of Chemistry.

Full text of DOI:10.1039/b614512d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Highly stereoselective chlorination of b-substituted cyclic alcohols using
PPh3–NCS: factors that control the stereoselectivity{
E. A. Jaseer, Ajay B. Naidu, Sreehari S. Kumar, R. Koteshwar Rao, Krishna G. Thakur and G. Sekar*
Received (in Cambridge, UK) 5th October 2006, Accepted 30th October 2006
First published as an Advance Article on the web 28th November 2006
DOI: 10.1039/b614512d
lone pair of electrons) at the b-position of the alcohol (Scheme 1).
We also report that the neighbouring group participation ability of
the b-substituted heteroatom can be controlled using appropriate
electron withdrawing groups or electron releasing groups.
A variety of trans-b-substituted cyclic alcohols were stereo-
selectively chlorinated to either the corresponding cis-chloride
or trans-chloride (inversion or retention of configuration) with
good to excellent yields; the stereochemical outcome is
determined by the size of the ring and the nature of the
b-substituents, especially the electronegativity of the substituted
atom.
First, we have chosen racemic-trans-2-hydroxycyclohexyl benzo-
ate 1, as a model substrate, which was subjected to chlorination
with NCS (1 equiv.) and PPh3 (1 equiv.) in THF at room
temperature. Surprisingly, trans-2-hydroxycyclohexyl benzoate 1
gave cis-2-chlorocyclohexyl benzoate 2 with 23% yield (Table 1;
entry 1). The cis stereochemistry of the chloride was deduced from
the coupling constant of the methine proton (dt, J = 8.8, 2.8, and
The halogenation of alcohols into alkyl halides is a ubiquitous
transformation in organic chemistry and numerous halogenating
agents are available to affect this key step. The conventional
method for the preparation of alkyl halides is by reacting readily
available alcohols with halogenating agents such as hydrohalic
acid,1 SOCl2,2 PX3,3 PX54 and POX3.5 However, these reagents are
corrosive, highly toxic, often result in very poor yields and usually
are intolerable for several acid-sensitive functional groups due to
the very strong acidic conditions. N-Halosuccinimide–PPh3 being a
mild halogenating agent can be used as an alternative (modified
Mitsunobu reaction), which provides inversion of configuration
through a SN2 reaction.6
1
2.8 Hz) at 5.16 ppm (–CH–OCO–Ph) in the H NMR spectrum.
We optimized the reaction conditions using a variety of solvents,
varying the ratio of NCS/PPh3 and the temperature to improve the
yield and the results are summarized in Table 1. Use of
1.5 equivalents of PPh3 with 2 equivalents of NCS in dry THF
at room temperature turned out to be the best conditions that
In recent years, chlorination through modified Mitsunobu
conditions has become very attractive as we discovered the chiral
version of this reaction by using chiral phosphine (BINAP) and
N-chlorosuccinimide (NCS) to produce highly enantiomerically
enriched alkyl chlorides and alcohols through non-enzymatic
kinetic resolution.7 Our continuous interest in this field has led us
to the finding that the stereochemical outcome of this reaction can
be stereoselectively controlled (retention or inversion of config-
uration) by having an appropriate substituent at the b-position of
the alcohol.
Scheme 1 Chlorination of b-substituted cyclohexanol.
Table 1 Optimizing the halogenation of (¡)-trans-2-hydroxycyclo-
hexyl benzoate 1
In general, chlorination of alcohol is a SN1 reaction (mixture of
inversion and retention of configurations) or SN2 reaction
(inversion of configuration) depending on the halogenating agent.
Although neighbouring group participation (NGP) has been
extensively studied in the literature,8 the effect of NGP on
stereoselective halogenation of cyclic alcohols is not well
established.9 For the first time, we report that the stereochemical
outcome of halogenation using PPh3–NCS can be stereoselectively
controlled to give either inversion of configuration (through SN2
reaction) or retention of configuration (through double SN2
reaction) by having an appropriate substituent (heteroatom with
Yielda
(%)
Entry PPh3
NCS
Solvent
THF
Temp. Time
1
2
3
1 equiv.
1 equiv.
1 equiv.
1 equiv.
1.5 equiv. THF
2 equiv.
rt
rt
rt
rt
1 h
23
10 min 39
10 min 47
THF
THF
Diethyl ether rt
4
5
6
7
8
9
10
11
12
13
a
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
1.5 equiv. 2 equiv.
2 h
1 h
24 h
7 h
7 h
86
30
00
82
76
Hexane
Benzene
Toluene
CH2Cl2
CHCl3
CH3CN
EtOAc
Acetone
rt
rt
rt
rt
rt
rt
rt
rt
10 min 68
1 h 59
10 min 55
1 h 15
10 min 51
Department of Chemistry, Indian Institute of Technology Madras,
Chennai, Tamil Nadu, 600 036, India. E-mail: gsekar@iitm.ac.in;
Fax: +91 44 2257 4202; Tel: +91 44 2257 4229
{ Electronic supplementary information (ESI) available: 1H NMR spectra
for cis- and trans-chlorides; X-ray crystal structures and data. See DOI:
10.1039/b614512d
lsolated yield
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 867–869 | 867

View Article Online
provided the highest yield (86%) of the product cis-chloride 2
Table 2 Stereoselective halogenation of b-substituted cyclohexanols
(Table 1; entry 4).{
Under the optimized reaction conditions, we performed
halogenations of several b-substituted cyclohexanols and the
results are summarized in Table 2. The reaction is highly
stereoselective as the reaction provides exclusively either the cis-
chloride or the trans-chloride.10 The inversion of stereochemistry
took place when the b-substituent is aryloxy (–OAr), ester
(–OCOAr), tosyloxy (–OTs), alkyl, aryl, NH–CBZ, and toluene-
sulfonamido (–NH-Ts) groups, whereas retention of configuration
was observed when the b-substituent is hydroxy (–OH), alkyloxy
(–OR), amino (–NH–Ar), amido (–NHCOPh), and phenylthio
groups (–SPh). The ring size of the cyclic alcohols also have a
significant effect in determining the stereochemical outcome of the
product (six-membered vs. eight-membered ring).
Yielda
(%)
Entry Alcohol
Halide
Time
1
2
3
4
5
6
7
Ar = Ph
Ar = Ph
X = Cl
X = Br
X = l
2 h
2 h
20 h
86
60b
08c
Ar = Ph
Ar = –C6H4-p-Cl
X = Cl
Ar = –C6H4-p-OMe X = Cl
10 min 51
1 h 75
10 min 62
10 min 52
Ar = –C6H4-o-Br
Ar = –C6H3-m,p-
(OMe)2
X = Cl
X = Cl
2-Aryloxy cyclohexanols provided cis-chloride through inver-
sion of configuration (entries 9–11), whereas 2-(phenylthio)cyclo-
hexanol (entry 14) provided trans-chloride through a three
membered thiiranium ion intermediate (double SN2 reaction:
retention of configuration). We inferred that this might be because
of the higher electronegativity of oxygen atom which prevents its
lone pair of electrons from neighbouring group participation.11 An
attempt to increase the electron density on the oxygen atom of the
phenoxy group by having electron releasing methoxy groups on
the phenyl ring did not provide retention of configuration for
product chloride (entry 10). Surprisingly, when the aryl group of
the aryloxy was replaced by hydrogen (trans-dihydroxy cyclohex-
ane) or alkyl groups (entries 12 and 13), the b-substituted oxygen
atom participates in NGP and provides the trans-chloride through
a double SN2 reaction.
8
Ar = –C6H3-o-OH, X = Cl
p-OMe
30 min 65
9
10
11
Ar = Ph
Ar = C6H4-p-OMe
Ar = C6H4-p-Ph
10 min 61
1 h 55
10 min 50
12
13
R = H
R = cyclohexyloxy
10 min 40d
10 min 55
14
10 min 55e
As expected, trans-2-phenylaminocyclohexanol produced trans-
2-phenylaminocyclohexyl chloride. This is due to the formation of
the three-membered aziridinium ion intermediate by the lone pair
of electrons on the low electronegative nitrogen atom. We could
not stop the NGP by substituting the highly electron withdrawing
nitro group at the para position of the phenyl group or by
15
16
R = OH
R =
OCO–C6H4-p-NO2
10 min 53
10 min 85
17
18
19
R = Ph
R = C6H4-p-NO2
R = COPh
10 min 30f
10 min 40g
10 min 35h
20
21
R = Ts
R = CBZ
10 min 65
10 min 79
22
10 min 35i
Fig. 1 X-Ray crystal structure of cis-N-(2-chlorocyclohexyl)-4-methyl-
23
24
25
a
R = Ph
R = C6H4–OMe
R = n-hexyl
10 min 75
10 min 60
10 min 37j
benzenesulfonamide (CCDC 621755 ). 30% Probability.
Isolated yields, reasonable quantities of unreacted starting material
was recovered wherever the reaction provided low yield for the
chloride.13 NBS–PPh3 as halogenating reagent. NIS–PPh3 as
halogenating reagent and 70% unreacted alcohol was recovered.
b
c
d
35% unreacted alcohol and 15% cyclohexene oxide were recovered.
Around 50% of the chloride (b-sulfide) was oxidized to the
corresponding b-sulfoxide while doing column chromatography
e
f
g
purification. 32% unreacted alcohol was recovered. An equivalent
amount of unknown compound was also isolated along with the
product. 55% unreacted alcohol was recovered. 25% unreacted
h
i
j
Fig. 2 X-Ray crystal structure of trans-2-hydroxycyclooctyl-4-nitro-
alcohol was recovered. 50% unreacted alcohol was recovered.
benzoate (CCDC 621756 ). 30% Probability.
868 | Chem. Commun., 2007, 867–869
This journal is ß The Royal Society of Chemistry 2007

View Article Online
of the b-substituted heteroatom can be controlled using appro-
priate electron withdrawing groups or electron releasing groups.
We thank DST, CSIR, New Delhi for financial support. We
also thank Dr Babu Varghese and Mr V. Ramkumar for solving
the crystal structures.
Notes and references
{ Typical experimental procedure: To a stirred mixture of trans-2-
hydroxycyclohexyl benzoate 1 (220 mg, 1 mmol), NCS (267 mg, 2 mmol)
and triphenylphosphine (394 mg, 1.5 mmol) was added dry THF (3 mL) at
room temperature. A change from white turbidity to a colorless
homogenous solution showed the completion of the reaction (2 h). The
THF was removed by rotary evaporation and the resulting residue was
purified by silica gel column chromatography (hexane–ethyl acetate) to
provide pure cis-2-chlorocyclohexyl benzoate 2 as a colorless liquid (205 mg,
86% yield). Rf = 0.67 ( 15% ethyl acetate in hexane); IR (neat) 1711 cm21
;
1H NMR (400 MHz, CDCl3) d 1.34–1.49 (m, 2H), 1.60–1.81 (m, 3H), 1.85–
1.94 (m, 1H), 1.95–2.12 (m, 2H), 4.35 (m, 1H, .CH–Cl), 5.16 (dt, J = 8.8,
2.8, 2.8 Hz, 1H, .CH–OCO–Ph), 7.39 (t, J = 7.6 Hz, 2H), 7.50 (t, J =
7.6 Hz, 1H), 8.02 (d, J = 7.2 Hz, 2H); 13C NMR (100 MHz, CDCl3) d 20.4,
21.0, 26.4, 31.2, 59.5, 67.4, 127.2, 128.5, 131.9, 164.6; MS (m/z): 239.1 (M+);
HRMS for calcd mass: 239.0839 (M+); Found: 239.0884.
Scheme 2 Possible mechanism for the formation of cis- and trans-
chloride.
§ Crystallographic data: for C13H18ClNO2S, M = 287.79, triclinic, a =
6.7769(4), b = 9.7641(8), c = 11.4462(8) s, a = 76.916(5), b = 80.416(4), c =
3
replacing the arylamino group by an electron withdrawing amide
group (entries 17–19). Interestingly, we were able to prevent the
NGP ability of the nitrogen atom by replacing the aryl moiety of
the arylamino group by electron withdrawing groups such as CBZ
and tosyl groups (entries 20 and 21). The cis stereochemistry of the
product was confirmed by X-ray crystal structure analysis (Fig. 1).§
When the b-substituent is a phenyl group, it produced cis-
chloride through a SN2 reaction. In this case, the p-electrons of the
phenyl ring did not take part in NGP. Increasing the p-electron
density of the phenyl ring by keeping an electron releasing
methoxy group at the para position did not make the p-electrons
participate in NGP to provide trans-chloride (entries 23 and 24).
The trans-dihydroxycyclohexane yielded trans-chlorohydrine
whereas trans-dihydroxycyclooctane produced cis-chloride (entries
12 vs. 15). This may be because the trans disubstituted cyclohexane
exists in a chair conformation and the dihedral angle between the
two diaxial substituents is nearly 180u.12 trans Disubstituted
cyclooctane, exists in a boat–chair conformation and the dihedral
angle between the two diaxial substituents is not close to 180u. This
phenomena was confirmed from the XRD analysis of trans-1,2-
disubstituted cyclooctane (trans-2-hydroxycyclooctyl-4-nitrobenzo-
ate). The disubstituted cyclooctane has a boat–chair conformation
and the dihedral angle between the two vicinal hydrogens (diaxial)
is 167u (Fig. 2).§ These data clearly show that the b-substituent
should be exactly or nearly 180u (anti) to the hydroxy group (trans;
in the di-axial conformer) to have effective NGP.
¯
74.897(4)u, V = 707.58(9) s , T = 273(2) K, space group P1, Z = 2,
m = 0.411 mm21, Rint = 0.0280 (for 4438 measured reflections), R1 = 0.0626
[for 1996 unique reflections with I . 2s(I)], wR2 = 0.1733 (for all 2382
unique reflections). For C15H19N1O5, M = 293.31, triclinic, a = 7.2564(6),
b = 7.4017(5), c = 15.5988(13) s, a = 88.923(5), b = 85.403(5), c =
3
¯
62.334(4)u, V = 739.47(10) s , T = 293(2) K, space group P1, Z = 2, m =
0.099 mm21, Rint = 0.0237 (for 9003 measured reflections), R1 = 0.0826 [for
2033 unique reflections with I . 2s(I)], wR2 = 0.2637 (for all 2530 unique
reflections), CCDC 621755 and 621756. For crystallographic data in CIF
or other electronic format see DOI: 10.1039/b614512d
1 (a) J. B. Conant and O. R. Quayle, Org. Synth., 1932, Coll. Vol. 1, 292;
(b) J. F. Norris and A. W. Olmsted, Org. Synth., 1941, Coll. Vol. 1, 144.
2 F. F. Caserio, G. E. Dennis, R. H. Dewolfe and W. G. Young, J. Am.
Chem. Soc., 1955, 77, 4182.
3 A. G. Anderson, Jr., N. E. T. Owen, F. J. Freenor and D. Erickson,
Synthesis, 1976, 398.
4 C. W. Shoppee, J. Chem. Soc., 1946, 1138.
5 R. S. Tipson, J. Org. Chem., 1962, 27, 1449.
6 (a) S. Hanessian, M. M. Ponpipom and P. Lavallee, Carbohydr. Res.,
1972, 24, 45; (b) A. K. Bose and B. La1, Tetrahedron Lett., 1973, 3937;
(c) A. K. Bose, B. Lal, W. A. Hoffman, III and M. S. Manhas,
Tetrahedron Lett., 1973, 1619.
7 (a) G. Sekar and H. Nishiyama, J. Am. Chem. Soc., 2001, 123, 3603; (b)
G. Sekar and H. Nishiyama, Chem. Commun., 2001, 1314.
8 J. March, Advanced Organic Chemistry; Reaction, Mechanisms and
Structure, John Wiley & Sons, New York, 4th edn., 1992.
9 For a similar type of selective chlorination of alcohols, see: (a)
R. Boschan and S. Winstein, J. Am. Chem. Soc., 1956, 78, 4921; (b)
C. N. Barry and S. A. Evans, Jr., J. Org. Chem., 1983, 48, 2825.
10 The stereochemistry of the product chlorides was determined by the
coupling pattern of the methine protons in the 1H NMR spectra. All the
cis-chlorides have a doublet of triplets (dt) pattern due to one trans
coupling and two cis couplings. Whereas the trans-chlorides gave a
triplet of doublets (td) coupling pattern due to two trans couplings and
one cis coupling (see ESI{ for 1H NMR spectra of selected compounds).
11 Pauling scale for the electronegativity of O, N, and S: O (3.5) . N (3.0)
. S (2.58). In other words, we can explain this NGP ability by the
nucleophilicity of the b-substituted atoms. The nucleophilicity of O, N,
and S: S . N . O.
We assumed that the stereoselective chlorination takes place as
shown in Scheme 2. However, very detailed mechanistic studies,
complete analysis of the factors that control the stereoselectivity,
kinetic study and the chiral version of this chlorination using chiral
phosphines or chiral halogenating agents are in progress.
In conclusion, the stereochemical outcome of the product
chloride is completely determined by the nature of the b-sub-
stituents and the ring size of the cyclic alcohols. All the
b-substituents with lone pairs of electrons or p-electrons do not
provide retention of configuration. Importantly, the NGP ability
12 The XRD analysis of trans-2-(4-nitrophenylamino)cyclohexanol shows
that the compound exits in a chair conformation and the dihedral angle
between the diaxial hydrogens is 177u. See ESI{ for X-ray structure and
crystal data.
13 Allowing these reactions to continue for longer times or using excess
halogenating agents did not help to improve the yield.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 867–869 | 869