Cyclodextrins containing an acetone bridge. Synthesis and study as epoxidation catalysts

Article abstract of DOI:10.1039/b410098k

Three cyclodextrine derivatives (6^A,6 ^D-di-O-(prop-2-one-1,3-dienyl)-α-cyclodextrin (1), 6-O-(prop-2-one-1-yl)-α-cyclodextrin (2) and 6^A,6 ^D-di-O-(prop-2-one-l, 3-dienyl)-β-cyclodextrin (3)) were synthesised and investigated as epoxidation catalysts. The three compounds were synthesised from the corresponding perbenzylated cyclodextrins which were mono- or didebenzylated in the 6-position using Sinay's method. Reaction with NaH and methallyl chloride in the case of 2, or methallyl dichloride in the case of 1 and 3, followed by dihydroxylation, periodate cleavage and protection group removal gave the target compounds. All three compounds catalysed, in the presence of oxone, the epoxidation of a series of alkenes. Epoxidation was compared to the reaction catalysed by simple ketones and inhibition was studied.

Full text of DOI:10.1039/b410098k

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Cyclodextrins containing an acetone bridge. Synthesis and study as
epoxidation catalysts†
Cyril Rousseau,a Brian Christensen,b Torben Ellebæk Petersenb and Mikael Bols*a
a Department of Chemistry, University of Aarhus, Langelandsgade 140, DK-8000, Aarhus,
Denmark. E-mail: mb@chem.au.dk; Fax: +45 86196199; Tel: +45 89423963
b Protein Chemistry Laboratory, Department of Molecular Biology, University of Aarhus,
DK-8000, Aarhus C, Denmark. E-mail: bc@imsb.au.dk; Tel: +45 89425091
Received 5th July 2004, Accepted 29th September 2004
First published as an Advance Article on the web 1st November 2004
Three cyclodextrine derivatives (6A,6D-di-O-(prop-2-one-1,3-dienyl)-a-cyclodextrin (1), 6-O-(prop-2-one-1-yl)-a-
cyclodextrin (2) and 6A,6D-di-O-(prop-2-one-1,3-dienyl)-b-cyclodextrin (3)) were synthesised and investigated as
epoxidation catalysts. The three compounds were synthesised from the corresponding perbenzylated cyclodextrins
which were mono- or didebenzylated in the 6-position using Sinay¨’s method. Reaction with NaH and methallyl
chloride in the case of 2, or methallyl dichloride in the case of 1 and 3, followed by dihydroxylation, periodate
cleavage and protection group removal gave the target compounds. All three compounds catalysed, in the presence of
oxone, the epoxidation of a series of alkenes. Epoxidation was compared to the reaction catalysed by simple ketones
and inhibition was studied.
Introduction
Mimicking Nature’s complex syntheses has long been an
outstanding challenge for organic chemists. The very specific
transformations during biosynthesis rely on highly developed
molecular recognition performed by enzymes. While recent years
have shown many wonderful applications of enzymes in organic
synthesis,1 their use is nevertheless subject to inherent limitations
in terms of substrate, reaction and solvent. Designed intelligent
catalysts that, like enzymes, can recognise substrates,2 can on
the other hand potentially be engineered to any reaction and
medium. The present work aimed at developing such artificial
enzymes for the epoxidation reaction. Several oxidising artificial
enzymes have been reported2,3 including an elegant epoxidis-
3a
ing metalloenzyme. In the present work we have explored
the combination of dioxirane chemistry with cyclodextrins as
complexing agents.
Fig. 1 Catalysts 1, 2 and 3 and the dioxirane 4 formed when 1 is treated
In the presence of oxone, ketones are known to act as
by oxone.
epoxidation catalysts through the intermediacy of dioxiranes.4
A wide variety of ketones have in recent years been investigated
have previously been investigated. A BINOL derivative was
as epoxidation catalysts particularly with the aim of obtaining
5i,5g
found to be a comparative poor epoxidation catalyst
while
enantioselective epoxidations under a variation of conditions.5
The catalytic efficiency of these ketones vary somewhat un-
predictably, but it has generally been found that electron-
withdrawing groups, such as oxygen and fluorine, in the vicinity
of the ketone is favourable for its catalytic efficiency. Often
stoichiometric amounts of ketone have been necessary, but
epoxidation of trans-stilbene with as little as 1 mol% ketone
1,3-alkoxyacetone derivatives of other diols appeared better
5j
catalysts. A related work on an epoxidising b-cyclodextrin
3e
ketoester has recently been reported. In this work a pyruvic
acid was attached to the primary rim of the cyclodextrin. This
catalyst was used in stoichiometric amounts to epoxidise various
alkenes with moderate enantioselectivity.
5b
has been reported.
Our molecules were designed based on the idea to employ the
cyclodextrin as a bucket-like complexing agent with an epoxidis-
ing entity in the bottom. The epoxidation catalysts made were
the cyclodextrin derivatives 1, 2 and 3 (Fig. 1). Compounds 1 and
3 contain a 1,3-dialkoxyacetone bridge between the 6-oxygen
atoms of the A and D sugar units of an a- or b-cyclodextrin
moiety, respectively. Inspection of computer models suggested
that in 4 (Fig. 1) the dioxirane group would be situated to
interact with a cyclic alkene bound in the cyclodextrin cavity.
Dioxirane epoxidation with other 1,3-alkoxyacetone derivatives
Results and discussion
The synthesis of 1 was carried out as depicted in Scheme 1. From
perbenzylated a-cyclodextrin6 the Pearce–Sinay¨ procedure7 was
used to obtain the diol 5 in high yield. Reaction of 5 with
methalyl dichloride (1.1 equiv.) and NaH (4 equiv.) in DMF
gave a high yield of 6. Through the sequence of dihydroxylation
and treatment with NaIO4–SiO2 in CH2Cl2 the ketone 7 was
obtained, that was deprotected to give 1.
Analogues 2 and 3 were made in a similar manner (Schemes 2
and 3). Compound 2 is a mimic that does not contain the bridge.
It was made from 87 by alkylation with methallylchloride/NaH
in DMF to give the alkene 9 (Scheme 2). Dihydroxylation of
9 with OsO4/NMO to the diol followed by periodate cleavage
with NaIO4 on silica gave the protected ketone 10, that was
8
† Electronic supplementary information (ESI) available: 1H and 13C
nmr spectra for selected compounds. See http://www.rsc.org/suppdata/
ob/b4/b410098k/
3 4 7 6
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 7 6 – 3 4 8 2
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
©

View Article Online
Scheme 1 Synthesis of catalyst 1.
Scheme 2 Synthesis of catalyst 2.
Scheme 3 Synthesis of catalyst 3.
deprotected to 2 by hydrogenolysis with Pd/C in EtOAc–
MeOH–TFA.
cases oxone and NaHCO3 was slowly added over the time period,
but two different solvent systems were used: Either MeCN–
water (3 : 2), which are classical dioxirane epoxidation conditions
In a similar set of reactions the A,D-diol of partially benzy-
lated b-cyclodextrin 117 was converted, by the intemediacy of
alkene 12, to bridged ketone 13 that was hydrogenolysed to 3
(Scheme 3). Noteworthy in this case was that the alkylation of
diol 11 with methylenedichloride appeared less facile than the
reaction of diol 5, but with the addition of NaI as nucleophilic
catalyst the reaction gave a good yield.
Epoxidation experiments were carried out with the ketone 1–3
(30 mol%), oxone (3 equiv.), NaHCO3 (12 equiv.) and different
alkenes (Table 1). Some general comments can be made. In all
5d, 9
(condition A), or pure water (condition B) which was used to
favor host–guest complexation. In water the reaction is faster,
but quite surprisingly epoxidation by oxone itself (background
reaction) occurs quite fast as well and is a greater problem.
Therefore simultaneous control experiments without ketone
were necessary in each case. These controls (not shown) give
negligible formation of epoxide (<10%) when the reaction time
is 1 h. It should be noted that in most cases not only epoxide
but also the corresponding diol was obtained. The efficacy
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 7 6 – 3 4 8 2
3 4 7 7

View Article Online
Table 1 Epoxidation of alkenes at 0 ◦C in the presence of 1 (0.3 equiv.) and oxone (3 equiv.) under condition A or condition B. ACE = acetone,
DCA = 1,3-dichloroacetone.
Entry
Alkene
Catalyst
Conditions
Time
Epoxide
Diol
Alkene
1
2
3
4
14
1
A
A
A
B
6 h
8 h
8 h
1 h
30%
30%
<10%
80%
(1% ee)
35%
10%
60%
—
60%
10%
>90%
—
1
ACE
1
20%
5
6
7
2
3
B
B
B
1 h
1 h
1 h
—
20%
—
65%
30%
14%
50%
86%
DCA
8
9
10
11
12
15
1
A
B
B
B
B
8 h
1 h
1 h
1 h
1 h
30%
30%
25%
40%
40%
—
—
—
20%
—
70%
70%
75%
40%
60%
1
2
3
DCA
21
13
14
16
1
1
A
B
8 h
1 h
30%
50%
50%
20%
—
50%
(12% ee)
18%
50%
15
16
1a
2
B
B
1 h
1 h
28%
—
54%
50%
(0% ee)
13%
19% (0% ee)
5%
12%
18%
17
18
19
20
21
22
23
2a
B
B
B
B
B
B
B
1 h
1 h
20 min
1 h
1 h
1 h
1 h
13%
44%
22%
77%
82%
—
74%
36%
73%
10%
—
30%
17%
3
3
3a
3b
DCA
70%
59%
DCAa
23%
24
25
26
27
28
17
18
19
1
A
B
B
B
B
8 h
1 h
1 h
1 h
1 h
—
—
—
—
21%
—
100%
65%
90%
60%
65%
1
35%
10%
19%
35%
2
3
DCA
29
30
31
32
33
1
A
B
B
B
B
8 h
1 h
1 h
1 h
1 h
—
—
—
—
—
—
—
—
—
—
100%
100%
100%
100%
100%
1
2
3
DCA
34
1
A
8 h
—
—
100%
a 2 equiv. of sodium benzene 2-naphtalenesulfonate was added. b NaHCO3 (20 equiv.) added.
3 4 7 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 7 6 – 3 4 8 2

View Article Online
of the oxidation is therefore based on the formation of both
these oxidation products. For comparison to control ketones,
acetone (ACE) and 1,3-dichloroacetone (DCA), were also tested
as catalysts on the various substrates.
All ketones 1–3 and DCA catalyse oxidation of indene (14),
dihydronaphthalene (15), styrene (16) and a-methylstyrene (17),
while none of them touch either cis or trans stilbene (18). Acetone
was found to be an extremely poor catalyst compared to the
other ketones presumably due to unfavourable electronic effects.
In water the oxidation of indene (14) occurs with an order of
catalyst reactivity that is 1 > DCA > 3 > 2 (entries 4–7). The
lower reactivity of 2 can be explained by electronic effects as
electron deficient ketones are known to be more reactive and
2 only has one electron-withdrawing a-alkoxy group. Also for
electronic reasons DCA is intrinsically the most reactive ketone
because of the larger electron-withdrawing effect of chlorine
(rI = 0.47) compared to alkoxy (rI = 0.25). Therefore the higher
catalyst activity of 1 must be attributed to an inclusion effect
that is expected to be better in 1 than in 3 due to the snugger
fit of 14 into the a-cyclodextrin. A similar pattern is seen for
the oxidation of styrene (16) where the efficiency order was 1
> DCA > 3 > 2 (entries 14, 16, 18 and 22). Again this can be
explained with 16 fitting better into a-cyclodextrin.
Fig. 2 Plot of the change in chemical shift Dd of H-3 in compound 2
versus the ratio of [16] to [2]tot ([G]/[CD.G]).
the presence of oxone and NaHCO3 was necessary to induce
the hydrolysis, a full oxidation experiment, similar to entry 18
(Table 1), was carried out but substituting the alkene 16 with
the product 21. Also here the styrene oxide (21) was recovered
unchanged. Since this suggested, quite surprisingly, that the diol
was formed from the alkene the epoxidation of 16 catalysed by
3 was stopped after 20 min (entry 19), which showed that the
ratio between diol and epoxide was the same as after 1 h, and
thus remained constant over time. The epoxide and diol must
therefore be formed by two parallel reactions.
The catalysts 1–3 reported here display some the basic
characteristics of an enzyme such as substrate recognition and
inhibition. While the binding strength between cyclodextrin
and substrate is within the normal range for natural enzyme–
substrate pairs, at least for 2 and 16, the catalysis found is
very far from showing enzymatic catalysis rate. Therefore the
catalytic step needs to be improved and therefore work is
required to optimise the structure of these catalytic groups either
through variation in its position or modulating its electronic
surroundings. Such work is underway.
With the more bulky alkenes 15 and 17 the catalyst efficiency
in water changed to 3 > DCA > 1 > 2 (entries 9–12 and 25–
28), which is in accordance with these alkenes fitting better into
b-cyclodextrin.
To explore the effect of the binding cavity experiments were
carried out where 2 equiv. of sodium naphthalene-2-sulfonate
(20) was added to the oxidation of 16 (entries 15, 17, 20 and 23).
Compound 20 binds to a-cyclodextrin and b-cyclodextrin with
binding constants of 3.6 × 102 and 2.3 × 105 M,10 respectively
and should therefore be an effective inhibitor. The catalytic
effect of the a-cyclodextrin derivatives 1 and 2 was significantly
inhibited (entries 15 and 17), while no inhibition, rather a slight
rate increase, was observed with 3 and DCA as catalysts (entries
20 and 23). This is in accordance with observations above
that suggested that inclusion increased the rate of styrene (16)
catalysed by 1 because of a better fit of this alkene into the
a-cyclodextrin.
The enantioselectivity of the oxidation of 14 by 1 under
condition B was also determined (HPLC, chiralcel OD column,
hexane–isopropanol 9 : 1), and indene oxide obtained was
found to be essentially racemic (1% ee). This is not entirely
surprising given the highly symmetric nature of the catalyst. For
the oxidation of styrene (16) catalysed by 1 the ee increased to
12% while the similar oxidation catalysed by 2 and and 3 gave
no enantioselectivity (Table 1). Nevertheless it is remarkable
that Chan et al. obtain enantioselectivities up to 40% ee in
epoxidation with a cyclodextrin ketoester being very similar to
Experimental
General
Solvents were distilled under anhydrous conditions. All reagents
were used as purchased without further purification. Evap-
oration was carried out on a rotatory evaporator with the
temperature kept below 40 ◦C. Glassware used for waterfree
reactions was dried for a minimum 2 h at 130 ◦C before use.
Columns were packed with silica gel 60 (230–400 mesh) as the
stationary phase. TLC-plates (Merck, 60, F254) were visualized
by spraying with cerium sulfate (1%) and molybdic acid (1.5%)
3e
2. They use stoichiometric amounts of ketoester, however.
The effectiveness of host–guest binding was determined for
binding of 16 to 2 using NMR as previously described (Fig. 2),
using the change in chemical shift of H-3 protons in 2 to monitor
the binding process. A dissociation constant (Kd) of 3.0 mM
was found for this binding process, which is quite similar to
the binding many enzymes have with their substrates. Similar
measurements for 1 and 3 were not possible because the H-3
NMR signals were not clearly resolved in these.
The diol formation that was observed in almost all the
reactions was puzzling because diol formation has rarely if ever
been reported in dioxirane epoxidations. Initially we attributed
the presence of diol products to the spontaneous hydrolysis
of the epoxides occurring more readily in water. However, the
observation that no diol was formed when DCA (or 2) were used
as catalysts suggested that the formation of diol was related
to the catalyst. Experiments were therefore carried out to see
if the cyclodextrin catalyst also catalysed epoxide hydrolysis.
However, when 3 and styrene oxide (21) was incubated in D2O
and monitored by NMR no hydrolysis was apparent. To see if
1
in 10% H2SO4 and heating until coloured spots appeared. H-
NMR, 13C-NMR and COSY were carried out on a Varian
Mercury 400 instrument. Mass spectra (MALDI MS) were
obtained on a Voyager DE PRO mass spectrometer (Applied
Biosystems) using an a-cyanohydroxycinnamic acid (a-CHCA)
matrix. Spectra were calibrated with angiotensin I m/z 1296.69,
adrenocorticotropic hormone (ACTH) (clip 1–17) m/z 2093.09,
ACTH (clip 18–39) m/z 2465.20, and ACTH (clip 7–38) m/z
3657.93.
6A,6D-Di-O-(prop-2-one-1,3-dienyl)-hexadeca-O-benzyl-a-
cyclodextrin (6)
To a solution of compound 57 (800 mg, 0.33 mmol) in dry DMF
(50 ml) was added NaH (58 mg, 4 equiv.). After stirring for
30 min at room temperature, 3-chloro-2-chloromethyl propene
(42 ll, 1.1 equiv.) was added and the reaction mixture was
stirred over night. The reaction was quenched by addition
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 7 6 – 3 4 8 2
3 4 7 9

View Article Online
of water (30 ml) and AcOEt (50 ml) and then the organic
phase was washed with water (4 × 30 ml), dried over MgSO4,
and concentrated. The remaining oil was purified by column
chromatography on silica gel (eluent: pentane–AcOEt 5 : 1.5),
which resulted in 690 mg (84%) of compound 6 as a white
foam. [a]D +31.4 (c 1.1, CHCl3); MS: calcd for C152H160O30Na
(4C, CH), 71.33 (4C, CH), 71.27 (2C, CH2), 70.55 (2C, CH),
=
62.01 (2C, CH2), 61.06 (2C, CH2). IR (KBr): 1683 (s, C O).
6-O-(Methall-1-yl)-hexadeca-O-benzyl-a-cyclodextrin (9)
To a solution of compound 87 (1.3 g, 0.52 mmol) in dry DMF
(50 ml) was added NaH (50 mg, 4 equiv.). After stirring for
30 min at room temperature, methallyl chloride (256 ll, 5 equiv.)
was added and the reaction mixture was stirred over night.
The reaction was quenched by addition of water (50 ml) and
AcOEt (50 ml), and then the organic phase was washed with
water (4 × 50 ml), dried over MgSO4, and concentrated. The
remaining oil was purified by column chromatography on silica
gel (eluent: pentane–AcOEt 5 : 1), which resulted in 1.06 g (80%)
of compound 9 as a white foam. [a]D +33.3 (c 1.2, CHCl3); MS:
calcd for C159H168O30Na 2580.1518, found 2580.2033; 1H NMR
(400 MHz, CDCl3): d 7.20–7.00 (m, 85H, CHPh), 5.10 (d, 1H, Jgem
11.2 Hz, H-CHPh), 5.09 (d, 1H, Jgem 10.8 Hz, H-CHPh), 5.04
(d, 1H, J 3.2 Hz, H-1), 5.02–4.97 (m, 5H, H-1), 4.82–4.70 (m,
8H), 4.44–4.20 (m, 24H), 4.10–4.01 (m, 6H), 3.99–3.88 (m, 12H),
3.86–3.77 (m, 8H), 3.64 (s, 2H, CH2), 3.52–3.32 (m, 12H), 1.54
(s, 3H, CH3). 13C NMR (100 MHz, CDCl3): d 142.11, 139.55,
139.51, 138.51, 138.33, 138.28 (Cipso), 128.41, 128.27, 128.09,
127.85, 127.72, 127.68, 127.51, 127.41, 127.39, 127.35, 127.03
(CHPh), 112.04 (CH2=), 98.93, 98.82, 98.76, 98.68 (C-1), 81.14,
79.69, 79.46, 79.37, 79.35, 79.16 (CH), 75.69, 75.30, 73.52, 72.86
(CH2), 71.67 (CH), 69.19 (CH2), 19.68 (CH3).
1
2488.0892, found 2488.1388; H NMR (400 MHz, CDCl3): d
7.26–6.88 (m, 80H, CHPh), 5.55–5.45 (m, 4H, CH2, H-1), 5.16
(d, 2H, Jgem 10.8 Hz, H-CHPh), 4.88 (d, 2H, Jgem 10.4 Hz, H-
CHPh), 4.84 (bs, 2H, CH2), 4.78–4.68 (m, 6H), 4.67–4.61 (m,
3H), 4.59 (d, 2H, J 3.2 Hz, H-1), 4.40–4.10 (m, 23H), 4.05–3.95
(m, 8H), 3.89–3.68 (m, 12H), 3.59 (d, 2H, J 9.2 Hz), 3.52 (dd,
2H, J 3.8 Hz, J 9.8 Hz), 3.45 (d, 2H, J 11.2 Hz), 3.40–3.30 (m,
8H), 3.24 (dd, 2H, J 3.2 Hz, J 9.6 Hz). 13C NMR (100 MHz,
=
CDCl3): d 143.09 (C ), 139.77, 139.50, 139.45, 138.88, 138.57,
138.39, 138.25, 137.99 (Cipso), 129.19, 128.41, 128.36, 128.25,
128.22, 128.20, 128.06, 127.99, 127.85, 127.77, 127.73, 127.61,
127.55, 127.18, 127.00, 126.72, 126.07 (CHPh), 114.04 (CH2=),
99.75, 99.46, 98.04 (C-1), 82.11, 81.67, 81.27, 80.76, 80.22, 79.30,
78.97, 77.88 (CH), 76.69, 76.22, 73.62, 73.55, 73.41, 73.27, 72.92,
72.00 (CH2), 72.08, 71.91, 71.85 (CH), 69.16 (CH2).
6A,6D-Di-O-(prop-2-one-1,3-dienyl)-hexadeca-O-benzyl-a-
cyclodextrin (7)
To a solution of compound 6 (1.53 g, 0.62 mmol) in a mixture
of acetone–water 9 : 1 (90 ml) was added N-methyl morpholine
N-oxide NMO, (252 mg, 3 equiv.) and a solution of osmium
tetroxide 2.5% in butanol (743 ll, 10%). The reaction mixture
was stirred overnight and a solution of sodium thiosulfate 10%
in water was added. The mixture was extracted with AcOEt
(3 × 50 ml) and the organic layer was dried over MgSO4.
Evaporation of the solvent gave the corresponding diol as a
white foam. The diol was dissolved in CH2Cl2 (10 ml) and
4.16 g of NaIO4–SiO2 (1.60 g NaIO4/10 g SiO2, 5 equiv.) was
added. After 2 h of stirring, the suspension was filtered through
Celite. Evaporation of the solvent and flash chromatography
(eluent: pentane–AcOEt 5 : 1.5) gave 1.35 g (88%, 2 steps) of
compound 7 as a white foam. [a]D +32.5 (c 1.0, CHCl3); MS:
calcd for C151H158O31Na 2490.0684, found 2490.0571; 1H NMR
(400 MHz, CDCl3): d 7.30−6.90 (m, 80H, CHPh), 5.43 (d, 2H,
6-O-(Prop-2-one-1-yl)-hexadeca-O-benzyl-a-cyclodextrin (10)
To a solution of compound 9 (1.05 g, 0.41 mmol) in a mixture
of acetone–water 9 : 1 (80 ml) was added NMO (166 mg, 3
equiv.) and a solution of osmium tetroxide 2.5% in butanol
(514 ll, 10%). The reaction mixture was stirred over night and
a solution of sodium thiosulfate 10% in water was added. The
mixture was extracted with AcOEt (3 × 50 ml) and the organic
layer was dried over MgSO4. Evaporation of the solvent gave
the corresponding diol as a white foam. The diol was dissolved
in CH2Cl2 (8 ml) and 1.7 g of NaIO4–SiO2 (2.54 g NaIO4/10 g
SiO2, 5 equiv.) was added. After 2 h of stirring, the suspension
was filtered through Celite. Evaporation of the solvent and flash
chromatography (eluent: pentane–AcOEt 5 : 1.5) gave 0.89 g
(84%, 2 steps) of compound 10 as a white foam. [a]D +35.0
(c 1.1, CHCl3); MS: calcd for C158H166O31Na 2582.1310, found
2582.1347; 1H NMR (400 MHz, CDCl3): d 7.20–7.00 (m, 85H,
CHPh), 5.15 (d, 2H, Jgem 10.5 Hz, H-CHPh), 5.12–5.02 (m, 6H),
5.00–4.90 (m, 4H), 4.82–4.74 (m, 6H), 4.46–4.18 (m, 22H), 4.10–
J
gem 10.4 Hz, H-CHPh), 5.30 (d, 2H, J 4.0 Hz, H-1), 5.17 (d, 2H,
Jgem 10.4 Hz, H-CHPh), 4.89 (d, 2H, Jgem 10.4 Hz, H-CHPh),
4.80–4.62 (m, 6H, CH2Ph), 4.63 (d, 2H, J 3.2 Hz, H-1), 4.58 (d,
2H, J 3.2 Hz, H-1), 4.38–4.07 (m, 23H), 4.04–3.85 (m, 12H),
3.81–3.65 (m, 14H), 3.51–3.20 (m, 11H). 13C NMR (100 MHz,
=
CDCl3): d 203.53 (C O), 139.94, 139.65, 139.58, 138.97, 138.66,
138.36, 138.25, 138.10 (Cipso), 128.63, 128.60, 128.55, 128.50,
128.42, 128.40, 128.23, 128.19, 128.16, 128.12, 127.99, 127.95,
127.90, 127.87, 127.84, 127.58, 127.33, 127.19, 127.00, 126.22
(CHPh), 100.64, 100.00, 98.64 (C-1), 82.57, 82.24, 81.14, 80.97,
80.80, 80.15, 79.05, 77.91 (CH), 76.86, 76.35, 74.05, 74.01, 73.62,
73.54, 73.11, 73.02 (CH2), 72.44 (CH), 72.22 (CH, CH2), 69.86,
68.91 (CH2), 68.87 (CH).
3.74 (m, 24H), 3.72 (s, 2H, CH2), 3.50–3.32 (m, 12H), 1.87 (s, 3H,
13
=
CH3). C NMR (100 MHz, CDCl3): d 206.48 (C O), 139.51,
139.48, 138.56, 138.51, 138.48, 138.43, 138.34, 138.29, 138.18
(Cipso), 128.46, 128.42, 128.27, 128.10, 128.00, 127.89, 127.87,
127.83, 127.81, 127.75, 127.71, 127.54, 127.48, 127.43, 127.32,
127.30, 127.05 (CHPh), 99.13, 99.04, 98.83, 98.76, 98.67 (C-1),
81.16, 81.10, 80.97, 80.92, 80.04, 79.88, 79.53, 79.24, 78.98,
78.85, 76.92, 75.79, 75.53, 73.57, 73.52, 73.49, 73.45, 72.96,
72.86, 72.80, 71.77, 71.61, 71.53, 70.55, 69.21 (CH, CH2), 26.46
(CH3).
6A,6D-Di-O-(prop-2-one-1,3-dienyl)-a-cyclodextrin (1)
Compound 7 (3 g, 1.2 mmol) was dissolved in a mixture of
MeOH–AcOEt (1 : 1) (150 ml). Then Pd/C (300 mg) and TFA
(cat.) were added and the mixture was stirred over night under
hydrogen atmosphere. Filtration through silica gel (eluent: n-
BuOH–EtOH–H2O 5 : 4 : 3) and evaporation of the solvent
gave 1.2 g (100%) of compound 1 as a white foam. MS: calcd
for C39H62O31Na 1049.3173, found 1049.3522; [a]D +83.6 (c 1.0,
H2O); 1H NMR (400 MHz, D2O): d 5.00–4.88 (m, 6H, H-1), 4.35
(d, 2H, Jgem 18.8 Hz, H-CHCO), 4.29 (bt, 2H, H-3), 4.16 (d, 2H,
H-CHCO, Jgem 18.8 Hz), 4.00–3.70 (m), 3.62–3.47 (m), 3.34 (m,
2H, H-4), 3.22 (m, 2H, H-4). 13C NMR (100 MHz, D2O): d
6-O-(Prop-2-one-1-yl)-a-cyclodextrin (2)
Compound 2 was obtained from the compound 10 (0.98 g,
0.38 mmol) and Pd/C (100 mg) in a mixture of MeOH–AcOEt
(1 : 1) (70 ml) following the same procedure given above for
the preparation of the compound 1. Gave 390 mg (100%) of
compound 2 as a white foam. [a]D +57.8 (c 1.2, H2O); MS:
1
calcd for C39H64O31Na 1051.3329, found 1051.3799; H NMR
(400 MHz, D2O): d 4.89 (d, 6H, J 2.8 Hz, H-1), 4.26 (d, 1H,
Jgem 18.6 Hz, H-CH), 4.21 (d, 1H, Jgem 18.6 Hz, H-CH), 3.92–
3.80 (m, 8H), 3.78–3.60 (m, 20H), 3.56–3.38 (m, 14H), 2.00 (s,
=
208.84 (C O), 102.02 (2C, C-1), 101.98 (2C, C-1), 98.16 (2C, C-
1), 82.49 (4C, CH), 79.71 (2C, CH), 73.63 (2C, CH), 73.08 (2C,
CH), 72.85 (2C, CH), 72.49 (2C, CH), 72.41 (2C, CH2), 71.50
3H, CH3). 13C NMR (100 MHz, D2O): d 211.05 (C O), 101,80
=
3 4 8 0
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 7 6 – 3 4 8 2

View Article Online
(C-1), 81.13, 76.12, 73.24, 72.16, 71.76, 70.72, 69.45, 60.10 (CH2,
for C45H72O36 1211.3701, found 1211.3876; 1H NMR (400 MHz,
=
CH), 25.68 (CH3). IR (KBr): 1683 (s, C O).
D2O): d 5.11–5.04 (m, 2H, H-1), 5.00 (d, 1H, J 3.6 Hz, H-1),
4.96–4.84 (m, 4H, H-1), 4.28–3.94 (m, 5H), 3.88–3.14 (m, 38H).
13
6A,6D-Di-O-(prop-2-one-1,3-dienyl)-heptadeca-O-benzyl-b-
=
C NMR (100 MHz, CDCl3): d 206.83 (C O), 101.68, 101.19,
101.01, 100.79, 99.99, 98.50 C-1), 94.05, 81.33, 80.70, 80.52,
80.38, 80.22, 80.08, 79.52, 78.82, 74.41, 74.27, 74.13, 73.31,
73.11, 72.85, 72.74, 72.61, 72.42, 72.27, 71.98, 71.93, 71.81,
71.51, 71.18, 70.86, 69.33, 68.75, 68.15, 60.64, 60.58, 60.14 (CH,
cyclodextrin (12)
To a solution of compound 117 (1.7 g, 0.60 mmol) in dry DMF
(60 ml) was added NaH (117 mg, 4 equiv.). After stirring 30
minutes a room temperature, 3-chloro-2-chloromethyl propene
(76 ll, 1.1 equiv.) and NaI (360 mg, 4 equiv.) were added to
the solution and the reaction mixture was stirred over night.
The reaction was quenched by addition of water (50 ml) and
AcOEt (100 ml) and then the organic phase was washed with
water (4 × 50 ml), dried over MgSO4, and concentrated. The
remaining oil was purified by column chromatography on silica
gel (eluent: pentane–AcOEt 5 : 1 to 5 : 1.5), to give 1.38 g (80%)
of compound 12 as a white foam. [a]D +37.4 (c 1.0, CHCl3); MS:
calcd for C179H188O35Na 2920.2828, found 2920.3506; 1H NMR
(400 MHz, CDCl3): d 7.37–6.98 (m, 95H, CHPh), 5.89 (d, 1H,
J 4.4 Hz, H-1), 5.65 (d, 1H, J 4.4 Hz, H-1), 5.47 (d, 1H, Jgem
10.4 Hz, H-CHPh), 5.32–5.25 (m, 3H, CH2Ph), 5.17 (d, 1H, Jgem
10.8 Hz, H-CHPh), 5.00 (bs, 2H, CH2=), 4.92 (d, 2H, J 3.2 Hz,
=
CH2). IR (KBr): 1679 (s, C O).
Epoxidation procedure A
To a solution of alkene (0.5 mmol) and 1 (0.15 mmol, 153 mg)
in MeCN–H2O (3/2, 25 ml) at 0 ◦C were added oxone (115 mg)
and NaHCO3 (63 mg) at the beginning of each hour over a 8 h
period. In total three equivalents of oxone and 12 equivalents of
NaHCO3 were added. Water was added and the aqueous layer
was extracted with CH2Cl2, dried (MgSO4) and evaporated.
Epoxidation procedure B
To a solution of alkene (0.5 mmol) and 1 (0.15 mmol, 153 mg)
in H2O (5 ml) at 0 ◦C were added oxone (153 mg) and NaHCO3
(84 mg) at the beginning of each 10 min over a 1 h period. In total
three equivalents of oxone and 12 equivalents of NaHCO3 were
added. Water was added and the aqueous layer was extracted
with CH2Cl2, dried (MgSO4) and evaporated.
H-1), 4.91 (d, 2H, J 3.6 Hz, H-1), 4.88–4.16 (m, 40H), 4.10–
3.24 (m, 40H). C NMR (100 MHz, CDCl3): d 142.55 (C ),
13
=
139.74, 139.69, 139.61, 139.48, 139.37, 139.30, 139.15, 138.86,
138.76, 138.74, 138.63, 138.42, 138.40, 138.34, 138.30, 138.09,
138.05 (Cipso), 128.41, 128.36, 128.28, 128.23, 128.20, 128.16,
128.10, 128.03, 128.02, 128.00, 127.98, 127.87, 127.82, 127.81,
127.78, 127.69, 127.65, 127.58, 127.21, 127.17, 127.09, 127.04,
126.97, 126.94, 126.51, 126.45 (CHPh), 114.47 (CH2=), 100.47,
99.89, 98.51, 98.27, 98.06, 98.05, 97.37 (C-1), 82.66, 82.17, 81.63,
81.41, 81.35, 81.04, 80.96, 80.84, 80.72, 80.64, 80.46, 80.11,
80.01, 79.17, 79.07, 78.63, 77.85, 77.77, 77.28 (CH), 76.67, 76.58,
76.48, 76.28, 75.40, 74.06, 73.61, 73.46, 73.42, 73.30, 73.21,
73.06, 72.97, 72.63, 72.24, 72.20, 71.98, 71.84, 70.82 (CH, CH2),
69.89 (CH), 69.58, 69.21, 68.93 (CH2), 68.70 (CH).
3e,11
Experimental procedure for NMR titration experiment
To a solution of 2 in D2O (1.5 ml, 0.89 mM) in a NMR tube
at ambient temperature was added styrene in portion (0.2–2.7
equiv.) until the change in chemical shift (DdH-3) of 2 was
negligible. The exact mole ratio of 2 to styrene were determined
1
from the integration ratios of their H NMR signal (CH3 of 2
and olefinic proton of styrene). The chemical shift (CH3) of 2
1
was used as an internal reference. A H NMR titration curve
was obtained by plotting DdH-3 of 2 against the mole ration of
2 and styrene.
6A,6D-Di-O-(prop-2-one-1,3-dienyl)-heptadeca-O-benzyl-b-
cyclodextrin (13)
Calculation of association constant in D2O11,12
Compound 13 was obtained from the compound 12 (1.29 g,
0.44 mmol) and NMO (178 mg, 3 equiv.), OsO4 (560 ll, 10%),
1,4 g of NaIO4–SiO2 (2.54 g NaIO4/10 g SiO2, 5 equiv.) following
the same procedure given above for the preparation of the
compound 7. Flash chromatography (eluent: pentane–AcOEt
5 : 1.5). Gave 1.0 g (77%) of compound 13 as a white foam. [a]D
+37.0 (c 1.8, CHCl3); MS: calcd for C178H186O36Na 2922.2621,
For 1 : 1 complex formation between a cyclodextrin (CD) and
a guest (G) in solution the amount of bound host compared to
the total amound of host (percent complex), P, can be calculated
from the total concentration and the equilibrium constant. The
following equations can be written:
CD + G ꢀ CD · G
[CD · G]
1
found 2922.2444; H NMR (400 MHz, CDCl3): d 7.28–6.80
(m, 95H, CHPh), 5.74 (d, 1H, J 4.0 Hz, H-1), 5.35 (d, 1H, Jgem
10.4 Hz, H-CHPh), 5.28 (d, 1H, J 4.4 Hz, H-1), 5.24 (d, 1H, Jgem
9.6 Hz, H-CHPh), 5.13 (d, 1H, Jgem 10.8 Hz, H-CHPh), 5.12 (d,
1H, Jgem 11.2 Hz, H-CHPh), 5.05 (d, 1H, J 3.2 Hz, H-1), 5.01
(d, 1H, Jgem 10.8 Hz, H-CHPh), 4.82–4.54 (m, 15H), 4.49–4.03
(m, 30H), 4.01–3.42 (m, 31H), 3.39–3.20 (m, 7H). 13C NMR
Ka =
, CCD = [CD · G] + [CD]
[CD · G]
[CD][G]
CG = [CD · G] + [G], P =
[CD · G] + [CD]
, a0 + a1 = 1
CCD
[CD]
[CD · G]
a0 =
, a1 =
=
(100 MHz, CDCl3): d 204.04 (C O), 139.61, 139.35, 139.15,
CCD
138.78, 138.59, 138.52, 138.28, 138.10, 138.03 (Cipso), 128.34,
128.21, 128.15, 128.07, 127.95, 127.82, 127.66, 127.55, 127.34,
127.10, 126.86, 126.66, 126.41 (CHPh), 99.72, 98.84, 98.73, 98.25,
98.10, 97.50 (C-1), 81.88, 81.77, 81.59, 81.33, 81.09, 80.90, 80.63,
79.93, 79.69, 78.99, 78.68, 78.31, 78.21, 77.98, 77.77, 77.40,
76.80, 76.37, 76.14, 75.01, 74.79, 74.36, 73.87, 73.60, 73.43,
73.35, 73.26, 73.14, 72.93, 72.72, 72.56, 72.33, 72.22, 72.03,
71.93, 70.31, 69.78, 69.25, 68.84, 68.60 (CH, CH2).
C
CD and CG being the total concentration of cyclodextrin and
guest, respectively. A combination of the above results give:
ꢀ
ꢁ
ꢂ
2
1
S
1
S
E + 1 +
E + 1 +
− 4E
P =
where E =
2
[G]
and S = [CD]Ka
[CD]
If dCD and dCD·G are the chemical shift of the observed nucleus
in sites CD and CD·G, then, under fast exchange in the NMR
time scale,
6A,6D-Di-O-(prop-2-one-1,3-dienyl)-b-cyclodextrin (3)
Compound 3 was obtained from the compound 13 (1 g,
0.34 mmol) and Pd/C (100 mg) in a mixture of MeOH–AcOEt
(1 : 1) (70 ml) following the same procedure given above for
the preparation of the compound 1. Gave 408 mg (100%) of
compound 3 as a white foam. [a]D +37.2 (c 1.1, H2O); MS: calcd
d = a0dCD + a1dCD · G or Dd = a1Dd0
Dd being the observed chemical shift displacement of the
signalofCDand Dd0 beingthecorrespondingdifferencebetween
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 7 6 – 3 4 8 2
3 4 8 1

View Article Online
pure CD and pure complex CD·G. A combination of the above
10–18; (e) W.- K. Chan, W.-Y. Yu, C.-M. Che and M.-K. Wong,
J. Org. Chem., 2003, 68, 6576–6582.
result in:
4 (a) R. W. Murray, Chem. Rev., 1989, 89, 1187–1201; (b) W. Adam, R.
Curci and J. O. Edwards, Acc. Chem. Res., 1989, 22, 205–211; (c) S. E.
Denmark and Z. Wu, Synlett, 1999, 847–859; (d) M. Frohn and Y.
Shi, Synthesis, 2000, 1979–2000; (e) W. Tu, Z.-X. Wang and Y. Shi,
J. Am. Chem. Soc., 1996, 118, 9806–9807; (f) G. Asensio, R. Mello, C.
Boix-Bernardini, M. E. Gonza´lez-Nu´n˜ez and G. Castellano, J. Org.
Chem., 1995, 60, 3692–3699; (g) D. Yang, X.-C. Wang, M.-K. Wong,
Y.-C. Yip and M.-W. Tang, J. Am. Chem. Soc., 1996, 118, 11311–
11312; (h) Y. Shi, Acc. Chem. Res., 2004, 37, 488–496; (i) D. Yang,
Acc. Chem. Res., 2004, 37, 497–505.
Dd
Dd
[CD · G] =
CCD and [CD] = CCD
−
CCD
Dd0
Dd0
Dd
⇒ P =
Dd0
ꢀ
ꢁ
ꢂ
2
1
S
1
S
E + 1 +
−
E + 1 +
2
− 4E
Dd
⇒
=
Dd0
5 (a) A. Armstrong, W. O. Moss and J. R. Reeves, Tetrahedron:
Asymmetry, 2001, 12, 2779–2781; (b) A. Armstrong and B. R.
Hayter, Chem. Commun., 1998, 621–622; (c) A. Solladie-Cavallo
and L. Bouerat, Org. Lett., 2000, 2, 3531–3534; (d) A. J. Carnell,
R. A. W. Johnstone, C. C. Parsy and W. R. Sanderson, Tetrahedron
Lett., 1999, 40, 8029–8032; (e) A. Armstrong and B. R. Hayter,
Tetrahedron, 1999, 55, 11119–11126; (f) W. Adam, C. R. Saha-
Mo¨ller and C.-G. Zhao, Tetrahedron: Asymmetry, 1999, 10, 2749–
2755; (g) C. E. Song, Y. H. Kim, K. C. Lee, S.-G. Lee and B. W.
Jin, Tetrahedron: Asymmetry, 1997, 8, 2721–2726; (h) K. Matsumoto
and K. Tomioka, Tetrahedron Lett., 2002, 43, 631–633; (i) D. Yang,
M.-K. Wong, Y.-C. Yip, X.-C. Wang, M.-W. Tang, J.-H. Zheng and
K.-K. Cheung, J. Am. Chem. Soc., 1998, 120, 5943–5952; (j) W.
Adam and C.-G. Zhao, Tetrahedron: Asymmetry, 1997, 8, 3995–
3998.
6 T. Sato, H. Nakamra, Y. Ohno and T. Endo, Carbohydr. Res., 1990,
199, 31–35.
7 A. J. Pearce and P. Sinay¨, Angew. Chem., Int. Ed. Engl., 2000, 39,
3610–3612.
8 Y. L. Zhang and T. K. M. Shing, J. Org. Chem., 1997, 62, 2622–2624.
9 Z.-X. Wang, W. Tu, M. Frohn, J. R. Zhang and Y. Shi, J. Am. Chem.
Soc., 1997, 119, 11224–11235.
If Dd0 is unknown, the equation becomes nonlinear and is
solved with the aid of a computer program. The fitting program
EASYPLOT was used.
Acknowledgements
We acknowledge financial support from the Danish National
Science Research Council (SNF) and the Lundbeck foundation
for financial support.
References
1 (a) B. G. Davis and V. Borer, Nat. Prod. Rep., 2001, 18, 618–640;
(b) C. H. Wong and G. M. Whitesides, Enzymes in Synthetic Organic
Chemistry, Pergamon, Oxford, 1994.; (c) K. Faber, Biotransforma-
tions in Organic Chemistry, 3rd ed., Springer, Berlin 1997.
2 (a) R. Breslow and S. D. Dong, Chem. Rev., 1998, 98, 1997–2011;
(b) S. Sasaki and K. Koga, Stud. Org. Chem. (Amsterdam), 1992,
45, 265–310; (c) A. J. Kirby, Angew. Chem., Int. Ed. Engl., 1996, 35,
707–724; J. Rebek, Acc. Chem. Res., 1984, 17, 258–264; (d) V. T. De
Souza and M. L. Bender, Acc. Chem. Res., 1987, 20, 146–152; (e) R.
Breslow, Science, 1982, 218, 532–537.
3 (a) R. Breslow, X. Zhang, R. Xu, M. Maletic and R. Merger,
J. Am. Chem. Soc., 1996, 118, 11678–11697; (b) R. Breslow, X.
Zhang and Y. Huang, J. Am. Chem. Soc., 1997, 119, 4535–4536;
(c) M. E. Deary and D. M. Davies, Carbohydr. Res., 1998, 309, 17–
29; (d) M. E. Deary and D. M. Davies, Carbohydr. Res., 1999, 317,
10 M. V. Rekharsky and Y. Inoue, Chem. Rev., 1998, 98, 1875–1917.
11 A. Botsi, B. Perly and E. Hadjoudis, J. Chem. Soc., Perkin Trans. 2,
1997, 89–94.
12 (a) H. J. Schneider and A. Yatsimirsky, Principle and Methods in
Supramolecular Chemistry, John Wiley & Sons, New York, 2000,
section D; (b) R. Foster and C. A. Fyfe, Prog. Nucl. Magn. Spectrosc.,
1969, 4, 1.
3 4 8 2
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 7 6 – 3 4 8 2