A hybrid cavitand made by capping permethylated α-cyclodextrin with cyclotriveratrylene

Article abstract of DOI:10.1002/ejoc.201101604

A hybrid C₃-symmetric cavitand 1, in which permethylated α-cyclodextrin (PM α-CDX) is capped with cyclotriveratrylene (CTV), has been prepared in 8 % yield by intramolecular cyclization of a vanillyl alcohol derivative attached to the primary r

Full text of DOI:10.1002/ejoc.201101604

FULL PAPER
DOI: 10.1002/ejoc.201101604
A Hybrid Cavitand Made by Capping Permethylated α-Cyclodextrin with
Cyclotriveratrylene
Frédérique Brégier,[a] Sekar Karuppannan,[a] and Jean-Claude Chambron*[a]
Keywords: Cyclodextrins / Cyclophanes / Cyclotrimerization / Template synthesis / Diastereoselectivity
A hybrid C3-symmetric cavitand 1, in which permethylated
α-cyclodextrin (PM α-CDX) is capped with cyclotrivera-
trylene (CTV), has been prepared in 8% yield by intramolec-
ular cyclization of a vanillyl alcohol derivative attached to
the primary rim of the CDX platform. The reaction proceeds
diastereoselectively (dr ≈ 6:1), the chirality of the α-glucopyr-
anosyl units controlling the chirality of the CTV component.
Interestingly, in polar solvents, 1 shows self-complexation
properties as the primary methoxy groups of the CDX com-
ponent are directed towards the CTV cavity.
to bind substrates with a greater variety of interactions than
separated cryptophanes and cyclodextrins. Such a system is
reminiscent of calix[6]cryptamides in which the small rim
of calix[6]arene is capped with CTV through amide brid-
ges.[10] In addition, CDX-based cavitands that feature or-
ganic[11] or metallo-organic[12] bridges endowed with cata-
lytic properties have been reported. We now disclose our
synthetic efforts to obtain the aesthetically pleasing CDX/
CTV hybrid receptor, which turned out to be accessible in
two steps from known reaction precursors.
Introduction
Cavitands were defined by Cram and co-workers as “syn-
thetic organic compounds that contain enforced cavities
large enough to accommodate simple molecules or ions”.[1]
However, nowadays the name “cavitand” is usually given to
compounds that derive from four-fold symmetric resorcin-
[4]arenes by linking the phenolic oxygen atoms to small car-
bon (e.g., methylenic)[2] or heteroatom (e.g., phosphon-
ito)[3] bridging units. Deep-cavity cavitands result from the
bridging of these resorcinol-derived oxygen atoms with aro-
matic subunits that form vertical “walls”.[4,5] Self-folding
systems represent the most advanced version of these latter
receptors, their C4v cone conformation being stabilized by
a seam of intra-annular hydrogen bonds, as in the case of
resorcin[4]arene itself. Deep-cavity cavitands are unique
molecular containers because, in spite of their open end,
exchange between complexed and free guest species can be
slow on the NMR timescale.[5a] Besides, cyclodextrins
(CDX) have been considered as natural cavitands[1b] and
cryptophanes,[6] made from two cone-shaped cyclotrivera-
trylene (CTV) cyclophanes, have also been included in the
original cavitand class of compounds.[1b,7] Considering the
host–guest properties of cryptophanes[6b,6c] and cyclodex-
trins,[8] the former involving van der Waals interactions, the
latter solvophobic effects, and the symmetry matching be-
tween C3-symmetric CTVs and C6-symmetric α-CDX, we
envisaged that capping[9] the narrow rim of α-CDX with
CTV would produce a deep-cavity C3-symmetric cavitand[5]
having the shape of a shuttlecock and the potential ability
Results and Discussion
Two synthetic strategies can be considered for preparing
a cavitand from CDX and CTV, either coupling of the func-
tionalized separate components (strategy A) or construc-
tion of the CTV-derived subunit by using the CDX plat-
form as a template (strategy B). Both strategies were tested
and only the second produced the target hybrid cavitand.
Note that cryptophanes are also most efficiently synthe-
sized by constructing the second CTV component on a pre-
synthesized CTV platform,[13] the tripod/tripod coupling
strategy affording lower yields.[14] We used permethylated
(PM) α-CDX rather than genuine α-CDX because the for-
mer lends itself easily to selective discrimination of three
out of the six primary alcohol functions by using the trityl
protection technique[15] and also because it shows better
flexibility and water solubility than the latter.[16] The direct
CDX precursor used in strategy A was the known mesylate
derivative 3[17] of tris(hydroxymethyl) PM α-CDX (2; Fig-
ure 1). It was obtained in 94% isolated yield by reaction
of 2 with mesyl chloride in pyridine in the presence of 4-
(dimethylamino)pyridine (DMAP). However, upon reaction
of 3 with cyclotriphenolene (4)[18] in the presence of caes-
ium carbonate in DMF at 70 °C and under high dilution
conditions (2ϫ10–3 m) we were unable to detect any trace
[a] Institut de Chimie Moléculaire de l’Université de Bourgogne
(ICMUB, UMR no 6302 of the CNRS), Université de
Bourgogne,
9, Avenue Alain Savary, BP 47870, 21078 Dijon, France
Fax: +33-3-80396117
E-mail: Jean-Claude.Chambron@u-bourgogne.fr
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201101604.
1920
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1920–1925

Hybrid Cavitand from α-Cyclodextrin with Cyclotriveratrylene
of cavitand. This was quite surprising as we had deliberately
chosen the sterically unhindered cyclotriphenolene (4) as
CTV precursor. However, the tripod/tripod coupling can
also produce misdirected products resulting from unwanted
[2+2] condensation that cannot be corrected to the desired
[3+3] product when the reaction involved is irreversible.[19]
Therefore we directed our efforts towards the synthesis of a
CDX/CTV hybrid cavitand by using strategy B.
Figure 1. Chemical structures of PM α-CDX 3 and CTV 4 precur-
sors used for the tripod/tripod coupling strategy A.
The chemical structures of the two possible dia-
stereomers of target cavitand 1 are shown in Figure 2. In
this molecule three oxygen atoms at the narrow rim of the
CDX platform are connected to three oxygen atoms of the
CTV subunit through ethylene bridges, as in cryptophane
A.[13a] Cavitand 1 was synthesized according to Scheme 1.
The precursor of the CTV cap is the vanillyl alcohol deriva-
tive 5, which was prepared according to literature pro-
cedures.[20] In the subsequent step, triol 2 was treated with
a four-fold excess of sodium hydride in DMF at room tem-
perature followed by the addition of the same amount of
iodo derivative 5. Twice as much reagent was used as mono-
and disubstituted intermediates were still present after 12 h
of reaction. The direct precursor 6 of cavitand 1 was finally
obtained in 49% yield after column chromatography.
Scheme 1. Two-step preparation (strategy B) of cavitand 1 from
PM α-CDX (2) and vanillyl alcohol derivative 5.
ast two diastereomers are formed as the 1H NMR spectrum
shows splitting of the singlet of the 2Ј-OCH3 substituents
at 3.405 and 3.403 ppm, and splitting of several signals in
the 13C{1H} NMR spectrum in [D6]acetone (see Figures S5
and S6 in the Supporting Information). Intramolecular cy-
clization (Scheme 1) was performed following two different
Figure 2. Chemical structures of the two diastereomers of cavitand
1. The atom numbering is shown for the diastereomer M-1. The
double-ended arrows indicate through-space interactions, as shown
1
by 2D ROESY H/1H NMR spectroscopy.
As this compound combines the asymmetric carbon reported procedures, that is, either under classic formic acid
atom of the THP protecting group and the enantiopure conditions[7,21] or by the recently developed Lewis acid ca-
CDX platform, it should exist as four diastereomers. At le- talysis method using Sc(OTf)3.[22] First, precursor
6
Eur. J. Org. Chem. 2012, 1920–1925
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1921

F. Brégier, S. Karuppannan, J.-C. Chambron
FULL PAPER
(2.0ϫ10–3 m) was heated in a 1:1 mixture of HCOOH and the yield of the cyclization reaction,[25] as expected led to
chloroform at 55 °C[23] and the reaction was followed by the breaking up of the CTV backbone (MALDI-TOF MS).
MALDI-TOF mass spectrometry. Examination of the se- In fact, the major product (7), isolated in 39% yield after
quence of spectra (see Figures S27 and S28 in the Support- 3 h (49% yield after 16 h), results from the dimerization of
ing Information) showed that the signal of precursor 6 (as two vanillyl alcohol pendants, the unreacted benzyl alcohol
sodium adduct) disappeared after 8 h giving rise to signals functions being left as formate esters (Figure 3). This is at-
at m/z = 1691.6 and 1783.7, which correspond, respectively, tested by ESI-MS, which shows a signal at m/z = 1783.76
1
to the sodium adducts of cavitand 1 and intermediate 7 (see corresponding to the sodium adduct, and H NMR spec-
Figure 3). The maximum 1:7 ratio was observed after 32 h, troscopy, which shows narrow triplets due to the formate
but the signals of the CDX-containing compounds disap- hydrogen atoms between 8.3 and 8.5 ppm (see Figures S21
peared after 56 h reaction, which points to a complete de- and S25 in the Supporting Information). In fact, there are
gradation of the CDX backbone.[24] However, monitoring four of these signals, which have pairwise relationships, as
1
of the reaction by H NMR spectroscopy led to less opti- expected for the two possible diastereomers 7a and 7b
mistic conclusions as degradation products that were not shown in Figure 3. Integration of these signals gives a ratio
detected by mass spectrometry were clearly apparent in the of 65:35 for the major and minor diastereomers of 7. Such
1H NMR spectrum recorded after 24 h (see Figure S29 in covalent capture of a cyclization intermediate in the tem-
the Supporting Information). In preparative runs, the heat- plate-directed formation of a CTV is unprecedented. To the
ing of 6 (1.1ϫ10–3 m) in a 1:1 mixture of HCOOH and best of our knowledge, no such intermediates have been ob-
chloroform at 55 °C for 3 h allowed us to isolate cavitand 1 served in the course of cryptophane synthesis. Their isola-
in 4% yield after careful column chromatography and the tion in significant yield points to the difficult nature of the
yield of isolated product increased to 8% by extending the cyclization reaction in this case. Employing Sc(OTf)3 in
reaction time to 16 h. A higher temperature and longer re- dichloromethane at reflux for 48 h only slightly improved
action time, which have otherwise been shown to improve the isolated yield of 1 (5%), but changing the solvent to
acetonitrile, again at reflux, produced only traces of the de-
sired product.
Cavitand 1 was characterized by mass spectrometry and
1H and 13C NMR spectroscopy (see the Supporting Infor-
mation). The main peak in the ESI mass spectrum corre-
sponds to the sodium adduct (m/z = 1691.74). The 1H
NMR spectrum in [D6]acetone is reproduced in Figure 4.
The signals were fully assigned by use of 2D 1H NMR tech-
niques (1H–1H COSY and ROESY, 1H–13C HSQC and
HMBC) and it is clearly seen that the CTV and CDX com-
ponents are present in a 1:1 ratio. The signals of the latter
are split, the doublets of 1-H and 1Ј-H excepted, which indi-
cates that the symmetry of the PM α-CDX platform has
been lowered from C6 to C3 upon conjugation with CTV,
as expected. This is confirmed by the 13C NMR spectrum
Figure 3. Chemical structures of diastereomers 7a and 7b.
1
Figure 4. H NMR spectrum ([D6]acetone, 600 MHz) of cavitand 1.
1922
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1920–1925

Hybrid Cavitand from α-Cyclodextrin with Cyclotriveratrylene
in which all the signals of the CDX component are split
(see Figures S13 and S14 in the Supporting Information).
In fact, the most interesting feature of the NMR spectra is
that they correspond to a single species as in theory two
diastereomers are expected to form depending on the direc-
tion of cyclization of the vanillyl alcohol derived fragment
(Figure 2). In fact, the two diastereomers can be distin-
guished in the 1H NMR spectrum of the crude reaction
mixture by the signals of the α-H and αЈ-H protons at δ =
7.15 and 7.11 ppm for the minor diastereomer, and at 7.12
and 7.09 ppm for the major diastereomer, the ratio of the
latter to the former being dr ≈ 6:1. This indicates that the
reaction is diastereoselective to a certain extent, however,
the slightly less polar minor diastereomer could not be iso-
lated in pure form by column chromatography as it is al-
ways accompanied by the major isomer. Note, the synthesis
of cryptophanes on a CTV platform is also diastereoselec-
tive as it produces either D3- or C3h-symmetric stereoiso-
Conclusions
We have shown that PM α-CDX can be used as a plat-
form for the covalently templated cyclotrimerization of a
vanillyl alcohol derivative attached to its primary rim. The
reaction proceeds in low yield but diastereoselectively, the
chirality of the CDX template controlling the direction of
cyclization of the CTV precursor. As the primary 6Ј-OCH3
substituents are probably responsible for the poor cycliza-
tion yields and complexation properties of 1, we are now
concentrating our efforts towards the synthesis of a true α-
CDX/CTV hybrid lacking the methoxy substituents.
Experimental Section
General Methods: Mass spectra were obtained either in MALDI-
TOF reflectron mode by using dithranol (1,8-dihydroxy-9(10H)-an-
thracene) as a matrix or in ESI mode. Column chromatographic
mers depending on the nature of the bridges connecting the separations were carried out by using 0.035–0.070 mm or 0.075–
0.200 mm silica gel 60. Analytical TLC was performed on silica gel
TLC plates with F-254 indicator. Solvents were dried and distilled
prior to use: DMF from CaH2, CH2Cl2 from P2O5, pyridine from
KOH. All other commercially available chemicals were used with-
out further purification. 2-[4-(2-Iodoethoxy)-3-methoxybenzyloxy]-
tetrahydropyran (5) was prepared from vanillyl alcohol.[20]
2I,2II,2III,2IV,2V,2VI,3I,3II,3III,3IV,3V,3VI,6II,6IV,6VI-Pentadeca-O-
methyl-α-cyclodextrin (2) was prepared according to literature pro-
cedures.[15]
CTV platform and the vanillyl alcohol derived end-
groups.[6a] The diastereoselectivity of the cyclization reac-
tion could stem from the fact that the asymmetry of cyclo-
dextrins is amplified by permethylation.[26] Interestingly, the
primary methoxy groups at C-6Ј are shielded by
–0.233 ppm (in [D6]acetone) on switching from triol 2 to
cavitand 1, whereas the secondary methoxy groups (at C-2/
2Ј and C-3/3Ј) show an upfield shift of less than 0.05 ppm.
Moreover, shielding of 6Ј-OCH3 is increased by an ad-
ditional –0.194 ppm in the presence of only around 5%
H2O. These observations suggest that the 6Ј-OCH3 groups
are directed towards the CTV cavity of the receptor rather
than the outside of the PM α-CDX component in these
polar, hydrophilic solvents. Remarkably, shielding as strong
as –1.72 ppm has been noted in the case of a cryptophane
with endo methyl carboxylate groups.[27] Examination of the
2I,2II,2III,2IV,2V,2VI,3I,3II,3III,3IV,3V,3VI,6II,6IV,6VI-Pentadeca-O-
methyl-6I,6III,6V-tri-O-methylsulfonyl-α-cyclodextrin (3): 4-(Dimeth-
ylamino)pyridine (0.301 g, 2.46 mmol) and methanesulfonyl chlor-
ide (0.410 g, 3.58 mmol) were added successively to a solution of 2
(1.185 g, 1.00 mmol) in pyridine (13 mL). The mixture was stirred
for 22 h at room temperature whereupon brine (50 mL) was added.
Then the mixture was extracted with ethyl acetate (4ϫ40 mL). The
combined organic layers were washed twice with 1 m HCl
(2ϫ30 mL), twice with 1 m NaOH (2ϫ30 mL) and dried with
magnesium sulfate. After evaporating the solvent under reduced
pressure the crude product was purified by column chromatog-
raphy on silica gel (MeOH/CH2Cl2, 4:96) to give compound 3 as
a colourless solid (1.329 g, 0.938 mmol) in 94% yield. 1H NMR
(500 MHz, [D6]acetone, 300 K): δ = 5.10 (d, JH,H = 3.5 Hz, 3 H,
1Ј-H), 5.08 (d, JH,H = 3.5 Hz, 3 H, 1-H), 4.61 (dd, JH,H = 11.5,
4.4 Hz, 3 H, 6A-H), 4.54 (dd, JH,H = 11.5, 1.9 Hz, 3 H, 6B-H),
1
2D H–1H ROESY spectrum of cavitand 1 in [D6]acetone
shows remarkable correlations, for example, α-H/OCH3(Ar)
and αЈ-H/8A,8B-H. However, the only intercomponent
through-space correlation is much weaker and involves the
CTV OCH3(Ar) and the PM α-CDX 6Ј-OCH3 groups (see
Figure S20 in the Supporting Information).
Reports on the complexation properties of PM α-CDX
are relatively scarce. These compounds have been shown, in
particular, to host aromatic guests[8] and this property has
been recently used for the generation of [2]rotaxane dimers
by self-association of singly modified species.[28] Besides,
polyrotaxanes made from the threading of PM α-CDX
onto polyTHF have been reported.[29] The unexpected self-
complexation of hybrid compound 1 is reminiscent of the
behaviour of PM α-CDX, which is bowl-shaped in the solid
state, because glucose subunits A and D are strongly tilted
inwards. As a consequence, the corresponding 6Ј-OCH3
moieties are oriented towards the cavity axis and form
van der Waals contacts with each other, closing the narrow
rim of the CDX cavity.[30] This could account for the inabil-
ity of this receptor to form a host–guest complex with, for
example, decanoic acid even in 20% (v/v) water in ace-
tone.[31]
4.01 (ddd, JH,H = 9.6, 4.4, 1.9 Hz, 3 H, 5-H), 3.92 (ddd, JH,H
=
9.7, 5.8, 1.4 Hz, 3 H, 5Ј-H), 3.85 (dd, JH,H = 10.8, 1.4 Hz, 3 H,
6ЈA-H), 3.76 (dd, JH,H = 10.8, 5.8 Hz, 3 H, 6ЈB-H), 3.60 (s, 9 H,
3- or 3Ј-OCH3), 3.58 (s, 9 H, 3- or 3Ј-OCH3), 3.53–3.39 (m, 12 H,
4/4Ј-H, 3/3Ј-H), 3.48 (s, 9 H, 2- or 2Ј-OCH3), 3.47 (s, 9 H, 2- or 2Ј-
OCH3), 3.36 (s, 9 H, 6Ј-OCH3), 3.17 (s, 9 H, SO2CH3), 3.12 (dd,
JH,H = 9.4, 3.5 Hz, 3 H, 2-H), 3.07 (dd, JH,H = 9.9, 3.5 Hz, 3 H,
2Ј-H) ppm. 13C NMR (150 MHz, [D6]acetone, 300 K): δ = 100.72,
99.85 (C-1/1Ј), 83.64, 83.18, 82.95, 82.59, 82.33, 82.05 (C-2/2Ј, C-
3/3Ј, C-4/4Ј), 72.95 (C-6Ј), 72.27 (C-5Ј), 70.86 (C-6), 70.54 (C-5),
61.91, 61.81 (3/3Ј-OCH3), 59.13 (6Ј-OCH3), 58.34, 57.89 (2/2Ј-
OCH3), 37.54 (SO2CH3) ppm. HRMS (ESI): calcd. for
C54H96NaO36S3 [M + Na]+ 1439.47356; found 1439.46916.
Compound 6: Sodium hydride (60% w/w in oil, 0.144 g, 3.60 mmol)
was added to a solution of 2 (0.347 g, 0.293 mmol) and the mixture
was stirred for 1 h. Tetrahydropyran 5 (1.38 g, 3.52 mmol) was then
added and the mixture was stirred overnight at room temperature.
Eur. J. Org. Chem. 2012, 1920–1925
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1923

F. Brégier, S. Karuppannan, J.-C. Chambron
FULL PAPER
As TLC analysis indicated the presence of remaining starting mate-
rial as well as the mono- and disubstituted derivatives, another por-
tion of sodium hydride (60% w/w in oil, 0.143 g, 3.58 mmol) was
added followed by the addition of 5 (1.378 g, 3.51 mmol) 1 h later
and the mixture was stirred for 22 h at room temperature. The same
procedure was repeated once with sodium hydride (60% w/w in oil,
0.161 g, 4.03 mmol) and 5 (0.365 g, 0.931 mmol). The reaction was
stirred for a further 24 h. The solvent was then evaporated under
reduced pressure and the residue was extracted with CH2Cl2. The
organic layer was washed with water, brine and dried with magne-
sium sulfate. After evaporating the solvent under reduced pressure
the crude product was purified by column chromatography on silica
gel (MeOH/CH2Cl2, gradient from 6:94 to 1:9) to afford compound
6 as a colourless solid (0.285 g, 0.144 mmol) in 49% yield. 1H
NMR (500 MHz, CDCl3, 300 K): δ = 6.89 (s, 3 H, βЈ-H), 6.86 (s,
6 H, αЈ-H, α-H), 5.07 (d, JH,H = 3.0 Hz, 3 H, 1- or 1Ј-H), 5.04 (d,
JH,H = 3.3 Hz, 3 H, 1- or 1Ј-H), 4.70 [d, JH,H = 11.7 Hz, 3 H,
traces of formic acid. The solvent was then evaporated under re-
duced pressure and the residue purified twice by column
chromatography on silica gel (MeOH/CH2Cl2, 3:97) to provide first
intermediate 7 as a mixture of diastereomers 7a and 7b (0.169 g,
49%) followed by cavitand 1 as a colourless solid (23.2ϫ10–3 g,
1.39ϫ10–2 mmol) in 8% yield.
Procedure B: A solution of precursor 6 (0.300 g, 0.152 mmol) in
CH2Cl2 (20 mL) was added to a stirred suspension of scandium
triflate (0.076 g, 0.154 mmol) in CH2Cl2 (38 mL) at reflux over a
period of 25–30 h through a syringe pump. After complete ad-
dition, the reaction mixture was stirred at 40 °C for a further 48 h.
The dark solution was then poured into water and extracted three
times with CH2Cl2. The combined organic layers were washed with
brine and dried with magnesium sulfate. After evaporating the sol-
vent under reduced pressure the crude product was purified by col-
umn chromatography on silica gel (MeOH/CH2Cl2, 3:97) to afford
cavitand 1 as a colourless solid (12.7ϫ10–3 g, 7.61ϫ10–3 mmol) in
ArCH2(A)], 4.67 (t, JH,H = 3.6 Hz, 3 H, 9-H), 4.42 [d, JH,H
11.7 Hz, 3 H, ArCH2(B)], 4.16 (m, 6 H, 8A/B-H), 4.10 (dd, JH,H
=
=
1
5% yield. H NMR (600 MHz, [D6]acetone, 300 K): δ = 7.13 (s, 3
H, α-H), 7.09 (s, 3 H, αЈ-H), 4.97 (d, JH,H = 3.3 Hz, 6 H, 1/1Ј-H),
4.78 (d, JH,H = 13.5 Hz, 3 H, Ha), 4.20 (ddd, JH,H = 11.4, 5.0,
4.5 Hz, 3 H, 8A-H), 4.13 (ddd, JH,H = 11.4, 7.2, 4.1 Hz, 3 H, 8B-
H), 3.87 (ddd, JH,H = 11.9, 7.0, 4.0 Hz, 3 H, 7A-H), 3.84 (s, 9 H,
OCH3-Ar), 3.77 (dd, JH,H = 11.7, 4.1 Hz, 3 H, 6ЈA-H), 3.67 (dd,
JH,H = 10.4, 1.4 Hz, 3 H, 6A-H), 3.64–3.56 (m, 12 H, 7B-H, He,
5/5Ј-H), 3.55 (s, 9 H, 3- or 3Ј-OCH3), 3.55 (s, 9 H, 3- or 3Ј-OCH3),
3.50 (dd, JH,H = 10.4, 1.4 Hz, 3 H, 6B-H), 3.48–3.46 (m, 3 H, 4-
or 4Ј-H), 3.45 (s, 9 H, 2- or 2Ј-OCH3), 3.44 (s, 9 H, 2- or 2Ј-OCH3),
3.43–3.40 (m, 6 H, 4- or 4Ј-H, 6ЈB-H), 3.37 (td, JH,H = 4.4, 8.9 Hz,
6 H, 3/3Ј-H), 3.09 (s, 9 H, 6Ј-OCH3), 3.02 (dd, JH,H = 5.3, 3.3 Hz,
3 H, 2- or 2Ј-H), 3.00 (dd, JH,H = 5.2, 3.3 Hz, 3 H, 2- or 2Ј-H) ppm.
13C NMR (150 MHz, [D6]acetone, 300 K): δ = 149.6 (C-γ), 147.7
(C-γЈ), 134.1 (C-β), 133.0 (C-βЈ), 117.6 (C-αЈ), 114.8 (C-α), 100.4,
100.2 (C-1/1Ј), 83.6 (C-4 or -4Ј), 83.38, 83.36 (C-2/2Ј), 82.9 (C-4
or -4Ј), 82.34, 82.26 (C-3/3Ј), 72.4, 72.3 (C-5/5Ј), 71.8 (C-6Ј), 70.8
(C-6), 69.3 (C-8), 69.1 (C-7), 61.79, 61.76 (3/3Ј-OCH3), 58.8 (6Ј-
OCH3), 58.0, 57.9 (2/2Ј-OCH3), 56.5 (OCH3-Ar), 36.4
(ArCH2) ppm. HRMS (ESI): calcd. for C81H120NaO36 [M + Na]+
1691.74515; found 1691.74451.
10.8, 3.1 Hz, 3 H, 6A-H), 3.90 (m, 9 H, 13A-H, 7A/B-H), 3.83 (s,
9 H, OCH3-Ar), 3.82–3.67 (m, 21 H, 6B-H, 6ЈA/B-H, 5/5Ј-H, 4/4Ј-
H), 3.64 (s, 9 H, 3- or 3Ј-OCH3), 3.62 (s, 9 H, 3- or 3Ј-OCH3), 3.53
(m, 9 H, 13B-H, 3/3Ј-H), 3.48 (s, 9 H, 2- or 2Ј-OCH3), 3.43 (s, 9
H, 2- or 2Ј-OCH3), 3.33 (s, 9 H, 6Ј-OCH3), 3.15 (dd, JH,H = 10.0,
3.3 Hz, 3 H, 2- or 2Ј-H), 3.11 (dd, JH,H = 9.2, 3.0 Hz, 3 H, 2- or
2Ј-H), 1.85 (m, 3 H, 10A-H), 1.72 (m, 3 H, 11A-H), 1.58 (m, 12 H,
1
10B-H, 11B-H, 12A/B-H) ppm. H NMR (600 MHz, [D6]acetone,
300 K): δ = 6.97 (d, JH,H = 1.9 Hz, 3 H, α-H), 6.94 (d, JH,H
=
8.2 Hz, 3 H, αЈ-H), 6.87 (m, 3 H, βЈ-H), 5.07 (d, JH,H = 3.2 Hz, 3
H, 1-H), 4.99 (d, JH,H = 3.4 Hz, 3 H, 1Ј-H), 4.66 (t, JH,H = 3.4 Hz,
3 H, 9-H), 4.64 [d, JH,H = 11.7 Hz, 3 H, ArCH2(A)], 4.40 [d, JH,H
= 11.7 Hz, 3 H, ArCH2(B)], 4.20–4.15 (m, 6 H, 8A-H, 7A-H), 4.13–
4.09 (m, 3 H, 8B-H), 3.91–3.79 (m, 15 H, 13A-H, 6A/B-H, 5/5-HЈ),
3.83 (s, 9 H, OCH3-Ar), 3.74–3.67 (m, 12 H, 7B-H, 6ЈA/B-H, 4Ј-
H), 3.58 (s, 9 H, 3- or 3Ј-OCH3), 3.56 (s, 9 H, 3- or 3Ј-OCH3), 3.48
(m, 3 H, 13B-H), 3.46 (s, 9 H, 2- or 2Ј-OCH3), 3.405 and 3.403 (2
s, 9 H, 2- or 2Ј-OCH3), 3.40 (m, 9 H, 3/3Ј-H, 4-H), 3.31 (s, 9 H,
6Ј-OCH3), 3.04 (dd, JH,H = 9.9, 3.4 Hz, 3 H, 2Ј-H), 2.99 (m, 3 H,
2-H), 1.81 (m, 3 H, 11A-H), 1.67 (m, 3 H, 10A-H), 1.52 (m, 12 H,
10B-H, 11B-H, 12A/B-H) ppm. 13C NMR (150 MHz, [D6]acetone,
300 K): δ = 150.61, 150.60 (C-γ), 149.11 (C-γЈ), 132.51, 132.50 (C-
β), 121.13, 121.12 (C-βЈ), 114.56, 114.54 (C-αЈ), 113.31, 113.28 (C-
α), 100.44 (C-1Ј), 99.74, 99.73 (C-1), 98.09, 98.08 (C-9), 83.32 (C-
4), 83.17, 83.12 (C-2/2Ј), 82.56 (C-4Ј), 82.51, 82.48 (C-3/3Ј), 72.76
(C-6Ј), 72.27 (C-5Ј), 71.79 (C-5), 70.52, 70.51 (C-7), 70.41 (C-6),
69.13, 69.10 (C-8), 69.07, 69.06 (ArCH2), 62.29, 62.28 (C-13),
61.82, 61.72 (3/3Ј-OCH3), 58.90 (6Ј-OCH3), 58.20, 57.82 (2Ј-
OCH3), 57.81 (2-OCH3), 56.36 (OCH3-Ar), 31.37 (C-10), 26.31 (C-
12), 20.10 (C-11) ppm. HRMS (ESI): calcd. for C96H150NaO42 [M
+ Na]+ 1997.94939; found 1997.95692.
MALDI-TOF MS Studies: Formic acid (10 mL) was added in one
portion to a solution of precursor 6 (0.042 g, 0.0213 mmol) in chlo-
roform (10 mL) under nitrogen. Aliquots (0.1 mL) were withdrawn
from time to time, neutralized with NaHCO3 and extracted into
dichloromethane (see Figures S26 and S27 in the Supporting Infor-
mation).
Supporting Information (see footnote on the first page of this arti-
cle): 1H and 13C NMR spectra of compounds 1, 3 and 6. 1H NMR
1
spectrum of compound 7. 1H–1H COSY and ROESY, and H–13C
HSQC and HMBC NMR spectra of compounds 1 and 6. ESI-
HRMS spectra of compounds 1, 3, 6 and 7. MALDI-TOF and
1H NMR spectra recorded during the formation of compound 1
(procedure A).
Compound 7: See the preparation of cavitand 1 below. Characteri-
zation: HRMS (ESI): calcd. for C83H124NaO40 [M
+
Na]+
1783.75611; found 1783.76066. 1H NMR (600 MHz, [D6]acetone,
300 K): see the Supporting Information.
Acknowledgments
Synthesis of Cavitand 1
The Conseil Régional de Bourgogne is acknowledged for post-doc-
toral fellowships to F. B. and S. K. (PARI IME SMT 8 program),
and Marie-José Penouilh is thanked for the ESI-HRMS spectra.
Procedure A: Formic acid (92 mL) was added in one portion to a
solution of precursor 6 (0.386 g, 0.195 mmol) in chloroform
(92 mL) under nitrogen. The reaction mixture was gently stirred at
55 °C and monitored by 1H NMR and MALDI-TOF MS. After
16 h, the reaction was quenched with a saturated aqueous solution
of NaHCO3 and extracted into CH2Cl2. The organic phase was
washed with a saturated aqueous solution of NaHCO3 to remove
[1] a) J. R. Moran, S. Karbach, D. J. Cram, J. Am. Chem. Soc.
1982, 104, 5826–5828; b) D. J. Cram, Science 1983, 219, 1177–
1187.
1924
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1920–1925

Hybrid Cavitand from α-Cyclodextrin with Cyclotriveratrylene
[2] D. J. Cram, S. Karbach, H.-E. Kim, C. B. Knobler, E. F. Mav-
erick, J. L. Ericson, R. C. Helgeson, J. Am. Chem. Soc. 1988,
110, 2229–2237.
10333; b) T. Traoré, L. Delacour, S. Garcia-Argote, P. Ber-
thault, J.-C. Cintrat, B. Rousseau, Org. Lett. 2010, 12, 960–962;
c) T. Traoré, L. Delacour, N. Kotera, G. Merer, D.-A. Buisson,
C. Dupont, B. Rousseau, Org. Process Res. Dev. 2011, 15, 435–
437.
[3] R. Pinalli, M. Suman, E. Dalcanale, Eur. J. Org. Chem. 2004,
451–462.
[4] D. J. Cram, H.-J. Choi, J. A. Bryant, C. B. Knobler, J. Am.
Chem. Soc. 1992, 114, 7748–7765.
[5] a) D. M. Rudkevich, J. Rebek Jr, Eur. J. Org. Chem. 1999,
1991–2005; b) S. M. Biros, J. Rebek Jr, Chem. Soc. Rev. 2007,
36, 93–104.
[6] a) A. Collet, Tetrahedron 1987, 43, 5725–5759; b) A. Collet, J.-
P. Dutasta, B. Lozach, Bull. Soc. Chim. Belg. 1990, 99, 617–
633; c) T. Brotin, J.-P. Dutasta, Chem. Rev. 2009, 109, 88–130.
[7] J. Canceill, M. Cesario, A. Collet, J. Guilhem, C. Pascard, J.
Chem. Soc., Chem. Commun. 1985, 361–363.
[8] M. V. Rekharsky, Y. Inoue, Chem. Rev. 1998, 98, 1875–1977.
[9] a) E. Engeldinger, D. Armspach, D. Matt, Chem. Rev. 2003,
103, 4147–4173; b) T. Lecourt, J.-M. Mallet, P. Sinaÿ, C. R.
Chim. 2003, 6, 87–90.
[10] S. Le Gac, I. Jabin, Chem. Eur. J. 2008, 14, 548–557.
[11] L. G. Marinescu, M. Bols, Angew. Chem. 2006, 118, 4706; An-
gew. Chem. Int. Ed. 2006, 45, 4590–4593.
[12] D. Armspach, D. Matt, C. R. Chim. 2011, 14, 135–148.
[13] a) J. Gabard, A. Collet, J. Chem. Soc., Chem. Commun. 1981,
1137–1139; b) J. Canceill, L. Lacombe, A. Collet, C. R. Acad.
Sci., Ser. II 1984, 298, 39–42.
[19]
[20]
L. Krejcˇí, M. Budeˇsˇínský, I. Císarˇova, T. Kraus, Chem. Com-
mun. 2009, 3557–3559.
J. Canceill, L. Lacombe, A. Collet, J. Am. Chem. Soc. 1985,
107, 6993–6996.
A. S. Lindsey, J. Chem. Soc. 1965, 1685–1692.
T. Brotin, V. Roy, J.-P. Dutasta, J. Org. Chem. 2005, 70, 6187–
6195.
[21]
[22]
[23]
[24]
M. Darzac, T. Brotin, L. Rousset-Arzel, D. Bouchu, J.-P. Duta-
sta, New J. Chem. 2004, 28, 502–512.
Heating 2 in CHCl3/HCO2H (50:50) at 55 °C for 6 h did not
lead to the degradation of the CDX backbone, but as expected
resulted in partial esterification of the primary alcohol func-
tions.
P. A. Hill, Q. Wei, T. Troxler, I. J. Dmochowski, J. Am. Chem.
Soc. 2009, 131, 3069–3077.
[25]
[26] K. Kano, K. Yoshiyasu, S. Hashimoto, J. Chem. Soc., Chem.
Commun. 1989, 1278–1279.
[27] C. E. O. Roesky, E. Weber, T. Rambusch, H. Stephan, K. Gloe,
M. Czugler, Chem. Eur. J. 2003, 9, 1104–1112.
[28] T. Fujimoto, A. Nakamura, Y. Inoue, Y. Sakata, T. Kaneda,
Tetrahedron Lett. 2001, 42, 7987–7989.
[14] D. J. Cram, M. E. Tanner, S. J. Keipert, C. B. Knobler, J. Am.
Chem. Soc. 1991, 113, 8909–8916.
[29] N. Kihara, K. Hinoue, T. Takata, Macromolecules 2005, 38,
[15] J. Boger, D. G. Brenner, J. R. Knowles, J. Am. Chem. Soc. 1979,
223–226.
101, 7630–7631.
[30] T. Steiner, W. Saenger, Angew. Chem. 1998, 110, 3628; Angew.
Chem. Int. Ed. 1998, 37, 3404–3407.
[31] H. Sun, C. L. D. Gibb, B. C. Gibb, Supramol. Chem. 2008, 20,
[16] D. Armspach, D. Matt, Inorg. Chem. 2001, 40, 3505–3509, and
references cited therein.
[17] Reported trimesylate 3 has not been characterized in the litera-
ture, see ref.[15] and the Supporting Information.
[18] a) H. A. Fogarty, P. Berthault, T. Brotin, G. Huber, H.
Desvaux, J.-P. Dutasta, J. Am. Chem. Soc. 2007, 129, 10332–
141–147.
Received: November 4, 2011
Published Online: February 21, 2012
Eur. J. Org. Chem. 2012, 1920–1925
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1925