2-Bromocyclohexanone perhydrate - X-ray crystal structure and conformational effects on reactivity in sulfoxidations

Article abstract of DOI:10.1016/S0040-4020(00)00608-6

The remarkably stable crystalline perhydrate 1 derived from a-bromocyclohexanone has been characterised by X-ray crystallography providing the first example of a perhydrate crystal structure. Perhydrate 1 exists in alternative chair conformations in chloroform and tetrahydrofuran with the bromine substituent occupying equatorial and axial positions, respectively. The perhydrate 1 can oxidise sulfides to sulfoxides in good yields in dichloromethane but not in tetrahydrofuran. The requirement for bromine to be equatorial for a stable perhydrate to form is demonstrated using conformationally locked cis and trans 2-bromo-4-tert-butylcyclohexanones in which case only the cis isomer 4a forms a perhydrate. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00608-6

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 6571±6575
2-Bromocyclohexanone PerhydrateÐX-ray Crystal Structure
and Conformational Effects on Reactivity in Sulfoxidations
b
a
a
Andrew J. Carnell,a, William Clegg, Robert A. W. Johnstone, Christophe C. Parsy and
*
William R. Sandersonc
aDepartment of Chemistry, Robert Robinson Laboratories, University of Liverpool, Liverpool L69 7ZD, UK
bDepartment of Chemistry, University of Newcastle, Newcastle upon Tyne NE1 7RU, UK
cConsultant to Solvay, c/o Solvay Interox Ltd, Baronet Works, P.O. Box 7, Warrington WA4 6HB, UK
Received 26 April 2000; revised 8 June 2000; accepted 29 June 2000
AbstractÐThe remarkably stable crystalline perhydrate 1 derived from a-bromocyclohexanone has been characterised by X-ray crystal-
lography providing the ®rst example of a perhydrate crystal structure. Perhydrate 1 exists in alternative chair conformations in chloroform
and tetrahydrofuran with the bromine substituent occupying equatorial and axial positions, respectively. The perhydrate 1 can oxidise
sul®des to sulfoxides in good yields in dichloromethane but not in tetrahydrofuran. The requirement for bromine to be equatorial for a
stable perhydrate to form is demonstrated using conformationally locked cis and trans 2-bromo-4-tert-butylcyclohexanones in which case
only the cis isomer 4a forms a perhydrate. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
recrystallised from hot toluene (Scheme 1). The perhydrate
1 is a white crystalline solid (mp 82±838C) and is isolated as
a single diastereomer in 53% yield. 13C NMR spectroscopy
(CDCl3) shows a characteristic signal at 101.2 ppm for the
perhydrate carbon. 1H NMR spectroscopy (CDCl3) shows a
double doublet for H-2 with J2,3axˆ9.2 Hz and J2,3eqˆ4.4 Hz
indicating the H-2 and H-3ax to be trans diaxial and that the
bromine is equatorial. A characteristic signal for the
peroxide hydrogen was observed as a sharp singlet at
8 ppm (free peroxide dHˆ7.6 ppm).
The use of electron de®cient ketones as precursors to
activated peroxide reagents such as perhydrates1 and
dioxiranes2 has recently attracted considerable interest.
The structural and electronic requirements for activity and
selectivity of the ketone precursors are beginning to be
probed. Hexa¯uoroacetone perhydrate is perhaps the best
known perhydrate but this reagent must be generated at
low temperature (2608C) using 90% hydrogen peroxide
and must be stored at 08C.3 The remarkably stable a-bromo-
cyclohexanone perhydrate 1 was ®rst reported by Kharasch
in 1958 but no details on stereochemistry were given due to
limitations in analytical methods at that time.4 We now
report the full structural characterisation of perhydrate 1,
the ®rst example of a perhydrate characterised by an
X-ray crystal structure and the importance of stereo-
chemical and conformational effects on reactivity for the
oxidation of simple sul®des.
In order to determine the relative stereochemistry of the
bromo and perhydroxy groups we obtained an X-ray crystal
structure, which shows a cis disposition between the two
groups, with the perhydroxy group axial and bromo
equatorial (Fig. 1). The peroxy O±O bond points away
from the bromine atom; a suggestion1 that the structure is
stabilised by intramolecular hydrogen-bonding between
bromine and perhydroxy is not supported by the crystal
structure, although this cannot be ruled out in solution.
Both the perhydroxy and the hydroxy OH groups show
two-fold disorder of orientation in the crystal structure,
and they engage in an extensive network of hydrogen
Results and Discussion
We prepared the perhydrate 1 according to the original
procedure4 that involved admixture of equimolar amounts
of a-bromocyclohexanone and 30% hydrogen peroxide
(Scheme 1). The crystals formed were ®ltered and
Keywords: perhydrate; a-bromocyclohexanone; conformational effects;
sulfoxidation; X-Ray structure.
* Corresponding author. Tel.: 144-151-794-3531; fax: 144-151-794-
3588; e-mail: acarnell@liv.ac.uk
Scheme 1.
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00608-6

6572
A. J. Carnell et al. / Tetrahedron 56 (2000) 6571±6575
oxide; after the reaction the catalyst could be recovered and
reused without noticable decomposition.5 Although we have
not carried out an extensive survey, it appears that the
perhydrate 1 is not a good source of electrophilic oxygen
since epoxidation of substrates such as cyclooctene and
1-phenylcyclohexene was very slow. As a nucleophilic
oxidant, Baeyer Villiger oxidation was only observed for
highly reactive strained ketones such as bicyclo[3.2.0]hep-
tenone (1 equiv. of perhydrate 1 in dichloromethane, 12 h,
92% yield) but not for cyclohexanone or cyclopentanone.
Perhydrate 1 may therefore prove useful as a selective
oxidant in synthesis since it is easily prepared on scale,
although caution should always be exercised when handling
organic peroxides and we have not shock tested this
material.
Importantly we have shown that the perhydrate 1 is the
active oxidant and not hydrogen peroxide, which might be
liberated in situ. Stirring it in dichloromethane for 24 h
showed no decomposition and no free peroxide being
formed. Surprisingly no sulfoxidation was observed when
using tetrahydrofuran as the solvent with any of the
substrates used in Table 1. This lack of reactivity may be
due to stabilisation of the perhydrate by H-bonding inter-
actions with the solvent or might re¯ect the importance of
stereoelectronic effects where the relative position of the
bromine would affect the electrophilicity of the terminal
Figure 1. Molecular structure of a-bromocyclohexanone perhydrate 1 in
the crystalline state.
bonding which links the groups on adjacent molecules;
O´´´O hydrogen-bond distances lie between 2.781 and
Ê
2.812 A, with angles at hydrogen between 152 and 1678.
Ê
The closest Br´´´O intermolecular distance is 3.572 A,
with no hydrogen atom directed even approximately along
this line, so there is no signi®cant involvement of bromine in
the hydrogen bonding scheme.
1
peroxy oxygen. Interestingly in d 8-tetrahydrofuran the H
The perhydrate 1 is able to carry out the sulfoxidation of a
range of sulphides giving sulfoxides in good to excellent
yields (Table 1) with no overoxidation to the sulfones
observed. In comparison with the well known perhydrates
of hexa¯uoroacetone and 1,1,1-tri¯uoroacetone as oxidants
for sulfoxidation, perhydrate 1 possesses intermediate
reactivity. For example, transformation of diphenyl sulphide
to the sulfoxide with hexa¯uoroacetone perhydrate is very
fast (1 min) whereas 1,1,1-tri¯uroacetone perhydrate takes
several days to oxidise a range of related substrates.1
Notably hexa¯uoroacetone perhydrate must be prepared at
2608C using 90% H2O2, making this reagent unsuitable for
use on a large scale. Recently, Sheldon has used per¯uoro-
heptadecan-9-one as a selective and mild catalyst for the
epoxidation of a wide variety of alkenes with hydrogen per-
NMR spectrum of 1 showed a triplet for the H-2 signal with
Jˆ4.1 Hz indicating that the perhydrate adopts the alterna-
tive chair conformation where the bromine is axial (Fig. 2).
In order to test this latter hypothesis we synthesised the
conformationally locked 2-bromo-4-tert-butylcyclohexa-
none epimers 4a and 4b starting from ketone 2 by
bromination of the derived silyl enol ether 36 with N-bromo-
succinimide (Scheme 2). The bromination reaction gave a
3:4 ratio of separable diastereoisomers 4a and 4b in which
bromine was shown to reside in the equatorial and axial
positions respectively, by examination of the coupling
constants for H-2.
Treatment of the isolated equatorial bromoketone 4a in tert-
butyl methyl ether with 30% hydrogen peroxide afforded a
stable perhydrate assumed to have the structure 5, by
analogy to the formation of perhydrate 1 (Scheme 3).
Table 1. All reactions were carried out in dichloromethane (5 cm3), sul®de
(1 equiv.) and perhydrate 1 (1.2 equiv.). The yield was determined by GC
(FFAP)
As for perhydrate 1, 1H NMR spectroscopy (CDCl3) of the
perhydrate 5 shows a double doublet for H-2 with
J2,3axˆ9.2 Hz and J2,3eqˆ4.4 Hz indicating that the bromine
Entry
1
Substrate
Time (h)
4
Sulfoxide Yield%
86
2
3
99
3
4
5
1
80
98
Figure 2.

A. J. Carnell et al. / Tetrahedron 56 (2000) 6571±6575
6573
Scheme 2.
perhydrates derived from the axial a-bromoketones to
experience stabilisation through H-bonding.
Experimental
General
Melting points were determined with a Gallenkamp
apparatus and are uncorrected. Elemental analyses were
performed by the microanalysis service of the University
of Liverpool. Infrared spectra were recorded either on a
Perkin Elmer 883 Infrared spectrophotometer or on a Perkin
Elmer 1720-X FTIR spectrometer. Solid samples were run
as KBr disc and liquid as a thin ®lms. 1H NMR spectra were
recorded either on a Bruker AC 200 (200 MHz), on a Varian
Gemini 2000 (300 MHz) or on a Bruker AMX 400
(400 MHZ) spectrometer for solutions in CDCl3 unless
otherwise stated. 13C NMR spectra were recorded with a
Varian Gemini 2000 (75 MHz) or on a Bruker AMX 400
(100 MHz) spectrometer for solutions in CDCl3 unless
otherwise stated. The chemical shifts are in parts per million
(ppm) using tetramethylsilane as the internal reference.
Coupling constants are in hertz (Hz). Mass spectra were
recorded with a Fisons TRIO 1000. All solvents and
reagents were puri®ed and dried according to standard
procedures. Thin layer chromatography (TLC) was
performed on silica gel 60-F254 (Merck). For column
chromatography, silica gel 60 (Merck, 63±200 mm) was
used without pre-treatment.
Scheme 3.
has remained equatorial. However, in contrast to the per-
hydrate 1, the H NMR in d8-tetrahydrofuran showed no
1
signi®cant change in appearance for the H-2 signal, re¯ect-
ing the conformational rigidity of this compound. Oxidation
of methyl p-tolyl sul®de with the perhydrate 5 in dichloro-
methane proceeded at a slower rate to that observed with
perhydrate 1 (40% conversion in 3 h). The same reaction
carried out in tetrahydrofuran resulted in a much slower
conversion (6.5% conversion in 3 h).
We therefore conclude that the reactivity of these per-
hydrates in dichloromethane arises by intermolecular
H-bonded interactions (OOH to OH or Br) whereas in tetra-
hydrofuran this activation is disrupted. The slower rate of
reaction with conformationally locked perhydrate 5 in
dichloromethane might be explained by a higher energy
barrier for the conversion of 5 to 4a when compared with
the conversion of 1 to 2-bromocyclohexanone. Calculations
using heats of formation give an energy difference of
3.2 kcal/mol for this transformation in favour of the non-
locked system 1.7 This difference presumably represents the
increased energy barrier for re-establishing the sp2 carbonyl
centre in the cyclohexanone ring when its conformational
¯exibility is limited by the 4-tert-butyl substituent. An
attempt to form the perhydrate with an axial bromine by
treatment of ketone 4b with 30% hydrogen peroxide
resulted in recovery of an equimolar mixture of 5 and 4b,
presumably resulting from acid-catalysed equilibration (pH
of H2O2 solutionˆ3.5) followed by perhydrate formation
from ketone 4a. Repeating the same reaction in phosphate
buffer at pH 7 or in dichloromethane with urea-hydrogen
peroxide gave no conversion and only recovered ketone 4b.
These results re¯ect a requirement for an equatorial bromine
in order for a stable perhydrate to form. This might be
explained either by the anomeric8,9 lowering of reactivity
of the carbonyl group by an axial bromine (ef®cient over-
lapping of the sp and p orbitals) or by the inability of
1-Hydroxy-2-bromocyclohexyl hydroperoxide 1. 2-
Bromocyclohexanone (12.93 g, 72.65 mmol) was added to
hydrogen peroxide solution (30%, 9 cm3, 79.38 mmol) and
stirred vigorously over 1 h. The stirring was stopped and the
reaction left overnight. The product crystallised to form a
white solid, which was crushed in toluene/petroleum-ether.
The solid was ®ltered and washed with the organic solvent.
The mother liquor was concentrated under reduced pressure
and more product was recovered and ®ltered. The solids
were combined and recrystallised from hot toluene to give
the pure perhydrate 1 as a white crystaline solid (8.05 g,
52.5%); (Found: C, 34.28; H, 5.30; Br, 37.73. C6H11BrO3
requires C, 34.15; H, 5.25; Br, 37.86%); nmax (KBr)/cm21
:
3260, 3175 (nO±H), 1198 (nC±C±O); dH (400 MHz,
CDCl3): 1.52±1.43 (2H, m), 1.69±1.62 (2H, m), 1.80±
1.74 (1H, m), 2.20±2.05 (2H, m), 2.52±2.38 (1H, m), 3.32
(1H, s, OH), 4.30 (1H, dd, J2,3axˆ9.2 Hz, J2,3eqˆ4.4 Hz,
2-H), 7.98 (1H, s, OOH);); dH (400 MHz, d8-THF): 1.46±

6574
A. J. Carnell et al. / Tetrahedron 56 (2000) 6571±6575
1.39 (2H, m), 1.61±1.54 (2H, m), 2.15±1.89 (4H, m), 4.39
(1H, t, J2,3ˆ4.1 Hz), 5.47 (1H, bs, OH), 10.35 (1H, s, OOH);
dC (100 MHz, CDCl3): 22.00±24.49 (4-C, 5-C), 30.94
(3-C), 33.46 (6-C), 56.56 (2-C), 101.16 (1-C).
3.06 (1H, m, 6-Hax) 4.37 (IH, m, 2-Heq); dC (75 MHz): 27.4
(3£CH3 and 5-C), 31.8 (40-C), 35.6 (6-C), 36.0 (3-C), 40.7
(4-C), 51.5 (2-C), 205.1 (1-C); m/z (EI): 232 and 234 (1.48,
1.21) bromine isotopes [M1H]1, 176 and 178 (3.65%,
3.43%) bromine isotopes, 97 (18.76), 57 (100).
Crystal structure determination for 110
cis-1-Hydroxy-2-bromo-tert-butylcyclohexyl hydroper-
oxide 5. cis-2-Bromo-tert-butylcyclohexanone 4a (2 g,
8.6 mmol) was dissolved in tert-butyl methyl ether
(3 cm3) and stirred vigorously. A 30% hydrogen peroxide
solution was added (1.17 cm3, 10.3 mmol) and stirred for
15 h, the reaction mixture was transferred to a separating
funnel, the organic layer was separated and the aqueous
layer washed with tert-butyl methyl ether (3£1.5 cm3).
The organic solutions were combined and dried over
magnesium sulfate. The organic solvent was evaporated
under reduced pressure to give a white solid. The solid
was recrystallised from pentane at -208C to give white crys-
tals (0.985 g, 43%); mp 57±598C (Found: C, 44.60; H, 7.24;
Br, 29.82, C10H19BrO3 requires C, 44.96; H, 7.17; Br,
29.91%); nmax (KBr)/cm21: 3560, 3211 (nO±H), 1176
(nC±C±O); dH (400 MHz; CDCl3): 0.88 (9H, s, 3£CH3),
1.24±1.16 (1H, m, 4-Hax), 1.38±1.27 (1H, m, 5-Hax), 1.65±
1.57 (1H, m, 6-Hax), 1.90±1.71 (2H, m, 5-Heq and 3-Hax),
2.28±2.22 (1H, m, 3-Heq), 2.64±2.59 (1H, m, 6-Heq), 3.4
(1H, br, ±OH), 4.25 (IH, dd, J2,3axˆ12.5 Hz,
J2,3eqˆ4.8 Hz, 2-Hax), 8.17 (1H, br, ±OOH); dC
(100 MHz): 23.7 (5-C), 28.1 (3£CH3), 28.2 (40-C), 33.1
(6-C), 37.5 (3-C), 49.9 (4-C), 58.2 (2-C), 101.7 (1-C)
Crystal data: C6H11BrO3, Mˆ211.06, triclinic, space group
Ê
Å
P1, aˆ5.5579(6), bˆ6.5299(7), cˆ10.9826(12) A, aˆb
Ê 3
76.363(3), bˆ84.367(3), gˆ85.686(3)8, Uˆ384.92(7) A ,
Zˆ2, Dcˆ1.821 g cm23
,
mˆ5.29 mm21 (MoKa, lˆ
Ê
0.71073 A), Tˆ160 K. Of 3099 re¯ections measured on a
Bruker AXS SMART CCD diffractometer and corrected
semiempirically for absorption, 1732 were unique
(u,28.68, Rintˆ0.0480). The structure was solved by
heavy-atom methods and re®ned on F2 values. Hydrogen
atoms bonded to carbon were constrained with a riding
model. The O±H groups (hydroxy and perhydroxy) were
each found by difference maps to be disordered equally
over two orientations, which were re®ned with geometry
restraints. Rˆ0.0571 (F values, F2.2s), Rwˆ0.1377 (F2
values, all data), goodness-of-®tˆ0.995, ®nal difference
Ê 23
map extremes (close to Br) 11.39 and 21.17e A . Soft-
ware: Bruker smart, saint, and shelxtl.
cis and trans-2-Bromo-tert-butylcyclohexanone (4a and
b). 4-tert-Butyl-1-trimethylsiloxycyclohexene 32 (2.3 g,
10.17 mmol) was dissolved in DMF (10 cm3) and stirred
vigorously. N-Bromosuccinimide (2.53 g, 14.24 mmol)
was dissolved in DMF (10 cm3) and added dropwise to
the reaction mixture. The temperature was maintained at
08C during the addition then slowly raised to room tempera-
ture. After 3 h, the reaction mixture was transferred into a
separating funnel and water (100 cm3) was added. The
organic layer was extracted and the aqueous layer washed
with diethyl ether (4£40 cm3). The organic solutions were
combined, washed with brine (40 cm3) and dried over
magnesium sulfate. The organic solvent was evaporated
under reduced pressure to give a brown oil that was puri®ed
by ¯ash-chromatography on silica gel buffered with tri-
ethylamine (tert-butyl methyl ester - n-hexane, 1:13) to
give the cis and trans isomers 4a and b.
Typical sulfoxidation procedure using 1-hydroxy-2-
bromocyclohexyl hydroperoxide
1-Hydroxy-2-bromocyclohexyl hydroperoxide (0.169 g,
0.80 mmol) was dissolved in DCM (10 cm3) and stirred.
Methylphenyl sulphide (0.100 g, 0.8 mmol) was added to
the reaction mixture. After 4 h the solvent was evaporated
and the crude was puri®ed by ¯ash-chromatography on
silica gel eluting ®rstly with dichloromethane to give the
unreacted sul®de and a-bromocyclohexanone then with
ethyl acetate to give methyl phenyl sulfoxide11 as cloudy
yellow oil, (95 mg, 85%).
cis-2-bromo-tert-butylcyclohexanone 4a as a white solid
(0.905 g, 38%); mp 64±668C; (Found: C, 51.52; H, 7.39,
Br, 34.83, C10H17BrO requires C, 51.52; H, 7.35; Br,
34.27%); nmax (KBr)/cm21: 1726 (nCvO); dH (400 MHz;
CDCl3): 0.93 (9H, s, 3xCH3), 1.57±1.46 (1H, m, 5-Hax),
1.71±1.63 (1H, m, 4-Hax), 1.96±1.86 (1H, m, 3-Hax),
2.18±2.11 (1H, m, 5-Heq), 2.47±2.38 (1H, m, 6-Hax),
2.72±2.64 (2H, m, 3-Heq and 6-Heq), 4.69 (1H, ddd,
J2,3axˆ13.3 Hz, J2,3eqˆ6.1 Hz, Jˆ1.2 Hz, 2-Hax); dC
(100 MHz): 27.7 (3xCH3), 28.0 (5-C), 32.8 (40-C), 40.2
(6-C), 41.2 (3-C), 48.7 (4-C), 56.5 (2-C), 201.8 (1-C); m/z
(EI): 232 and 234 (2.65%, 2.56%) bromine isotopes
[M1H]1, 176 and 178 (6.08, 5.58) bromine isotopes, 97
(25.33), 57 (100).
Acknowledgements
We gratefully acknowledge the Solvay Interox and the
Eschenmoser Trust for the funding of this work (studentship
to C. P.) and the EPSRC for equipment funding.
References
1. Ganeshpue, P. A.; Adam, W. Synthesis 1995, 179.
2. Denmark, S.; Wu, Z. Synlett 1999, 847.
3. Ganem, B.; Heggs, R. P.; Biloski, A. J.; Schwartz, D. R.
Tetrahedron Lett. 1980, 21, 685.
4. Kharasch, M. S.; Sosnovsky, G. J. Org. Chem. 1958, 23, 1322.
5. vanVliet, M. C. A.; Arends, I. W. C. E.; Sheldon, R. A. J.Chem.
Soc. 1999, 263±264.
trans-2-Bromo-tert-butylcyclohexanone 4b as a colourless
oil (0.951 g, 40%);nmax (®lm)/cm21: 1721 (nCvO); dH
(300 MHz; CDCl3) 0.93 (9H, s, 3£CH3), 1.51±1.37 (1H,
m, 5-Hax), 2.01±1.86 (2H, m, 3-Hax and 4-Hax), 2.14±2.04
(1H, m, 5-Heq), 2.36±2.28 (2H, m, 3-Heq and 6-Heq), 3.18±
6. Wang, Z. X.; Tu, Y.; Frohn, M.; Zhang, J. R.; Shi, Y. J. Am.
Chem. Soc. 1997, 119, 11224.
7. Heats of formation in water were calculated by mopac at the

A. J. Carnell et al. / Tetrahedron 56 (2000) 6571±6575
6575
PM3 level (Stewart, J. J. P. mopac 6.00, QCPE No. 455; supplied
by Cache Work Systems v.4.1, Oxford Molecular, Oxford Science
Park, Oxford OX 4 4GA, UK).
9. Corey, E. J.; Burke, H. J. J. Am. Chem. Soc. 1955, 77, 5418±
5420.
È
È
10. G. M. Sheldrick, shelx 97, Universitat Gottingen, 1997.
11. Ochiai, M.; Nakanishi, A.; Ito, T. J. Org. Chem. 1997, 62,
4253.
8. Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic
Compounds; Wiley Interscience: New York, 1994, pp 731±733.