An acid-catalyzed cyclialkylation that provides rapid access to a twisted molecular basket

Article abstract of DOI:10.1002/chem.201201153

Twist of fate: Expanding the scope of cavity-containing compounds is essential for developing more efficient sensors and catalysts. This report describes an efficient cyclialkylation reaction that provides rapid access to a novel C₃ symmetric a

Full text of DOI:10.1002/chem.201201153

COMMUNICATION
DOI: 10.1002/chem.201201153
An Acid-Catalyzed Cyclialkylation that Provides Rapid Access to a Twisted
Molecular Basket
Keith Hermann, Daniel A. Turner, Christopher M. Hadad, and Jovica D. Badjic´*[a]
Dedicated to Professor Leo Paquette
The inspiration for creating molecules with concave topol-
ogy[1] comes from nature in which chemical processes are
conducted in confined environments. In the early eighties,[2]
D. J. Cram noted a paucity of synthetic compounds possess-
ing a cavity and contemplated about the potential of encir-
cling a guest species with such hosts. In the years that fol-
lowed, numerous studies of various cavitands[3] as well as
self-assembled capsules[4] have shown some benefits of mo-
lecular encapsulation. For example, 1) the modulation of the
outcome of chemical reactions,[5] 2) the enhancement of the
persistence of unstable reactive intermediates,[6] and 3) the
control of the conformational dynamics of molecules.[7] In
spite of much advancement in the area,[8] the utility of
cavity-containing compounds for routine separation,[9] sens-
ing,[10] and catalysis[11] remains limited. We suggest that cavi-
tands with useful topological features and functionalities are
still difficult to come by,[12] which impedes the expansion of
Figure 1. Top: Energy-minimized structure (B3LYP/6-31G*)[18] of basket
the field into more practical areas of science.[13] Although
1syn and its 1H NMR spectrum (400 MHz, CDCl3). Bottom: Energy-mini-
mized structure (B3LYP/6-31G*)[18] of compound 1anti and its 1H NMR
spectrum (400 MHz, CDCl3).
one can use the tools of kinetic and thermodynamic templa-
tion for preparing capsules,[14] there are often other reactions
competing with the desired
macrocyclization. In this vein,
our report delineates an effec-
tive macrocyclization strategy
that can be used for the rapid
preparation of hosts resembling
gated molecular baskets.[15] In
particular, we have used meth-
ods of experimental and com-
putational chemistry for investi-
gating the formation of cup-
shaped 1syn possessing a twisted
framework and chiral inner
space (Figure 1).
Scheme 1. A) In the presence of acids, indene (2, R=H) undergoes a cationic polymerization to give polymer-
ic products. B) In the presence of CH3SO3H, compound 3 gives dibenzobicycloACHTNUGTRNEUNG[3.2.1]octadiene 4 in 10–77%
yield (Table 1).
In the presence of Lewis- or Brønsted acids, indene 2 un-
dergoes cationic polymerization to give polyindenes
(Scheme 1A).[16] The propagation step of the addition in-
cludes the formation of the indanyl cation,[17] which enables
the chain growth. In light of this mechanism, we reasoned
that in the presence of an acid, indene derivative 3 should
undergo a cationic polymerization in competition with
a
´
[a] K. Hermann, D. A. Turner, Prof. C. M. Hadad, Prof. J. D. Badjic
Department of Chemistry and Biochemistry
The Ohio State University
a
AHCTUNGTRENNUNG
Friedel–Crafts annulation to give dibenzobicyclo-
100 West 18thAvenue, Columbus OH 43210 (USA)
Fax : (+) 614-292-1685
[3.2.1]octadiene 4[19] (Scheme 1B); the hypothesis was based
on earlier studies of electrophilic ring closures of aryl-substi-
tuted compounds (cyclialkylations).[20] Interestingly, we
found that 3 (1.0 mm, CH2Cl2), in the presence of methane-
E-mail: badjic@chemistry.ohio-state.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201153.
Chem. Eur. J. 2012, 18, 8301 – 8305
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8301

Table 1. The annulation of compound 3 (1.0 mm) into 4 and of 5a/b into
1syn/anti was promoted with CH3SO3H (50.0 mm), as determined after
HPLC separation.
First, we synthesized 5a/b by completing the coupling of
1,3,5-tris(bromomethyl)benzene with indene (Scheme 2).[18]
Although the separation of diastereomeric 5a and 5b was an
Substrate
Solvent
T
Yield
[%]
1
arduous task, H NMR spectroscopic measurements suggest-
[K]
ed that two species (C3 symmetric 5a and C1 symmetric 5b)
formed in an approximate 1:5 ratio (see the Supporting In-
formation, Figure S6). Indeed, the result of HPLC separa-
tion corroborated this proportion of 5a/5b stereoisomers (see
the Supporting Information, Figure S16).
3
3
3
5a/b
5a/b
5a/b
5a/b
CH2Cl2
298
344
344
298
313
344
344
ꢁ10
62
77
47
57
ClCH2CH2Cl
ClCH2CH2Cl[a]
CH2Cl2
CH2Cl2
ClCH2CH2Cl
75
85
We tested various acids for catalyzing the conversion of
5a/b into 1syn/anti in CH2Cl2 and at 298 K (see the Supporting
Information, Table S1). Markedly, triflic- (CF3SO3H, pKa =
ꢀ14) and methanesulfonic acids (CH3SO3H, pKa =ꢀ2.6)
promoted the reaction giving the desired cyclotrimer as
a mixture of syn/anti diastereomers in a 1:5 ratio and 47%
overall yield (Scheme 2). The cationic polymerization of
5a/b (Scheme 1B) is a bimolecular process with, perhaps,
a negative entropy of activation characterizing the chain ex-
tension. In line with this premise, we increased the reac-
tionꢁs temperature to assist the intramolecular annulation at
the expense of the intermolecular polymerization. Indeed,
the overall yield of 1syn/anti (Table 1) improved in refluxing
CH2Cl2 (57%), and the yield increased to 75% with the
higher boiling ClCH2CH2Cl. In fact, when the cyclialkyla-
tion of 5a/b was completed with a slow addition of the sub-
strate (using a syringe pump), the yield of desired 1syn/anti in-
creased to 85% (Table 1).
ClCH2CH2Cl[a]
[a] The substrate was added by a syringe pump over a period of 2 h.
sulfonic acid (CH3SO3H, 50.0 mm), would give 4 in a rather
low yield (10%, Table 1), with the polymerization reaction
dominating the conversion of starting material, compound 3.
There arose an intriguing question: would trivalent 5a/b give
any of cup-shaped 1syn through a tandem of such cyclialkyla-
tion reactions (Scheme 2)?
Compounds 1syn and 1anti were separated by column chro-
matography (SiO2, hexanes/acetone=10:1). The full assign-
ment of their 1H NMR resonances (Figure 1) was accom-
plished with the assistance of 1H-1H COSY and NOESY cor-
relations (see the Supporting Information, Figure S9–S12).
Importantly, C3 symmetric 1syn comprises three [3.2.1] bicy-
clic rings twisted in the same direction so that the molecule
is helical with either right (P) or left-handed (M) sense of
twist (Figure 2A). Indeed, the HPLC chromatogram of 1syn
Figure 2. A) C3 symmetric 1syn has a screw-shaped structure with either
a right ((R6)-1syn) or left-handed ((S6)-1syn) sense of twist. B) HPLC chro-
matogram of 1syn (hexane/isopropanol=93:7, Chiracel OD-H) shows two
signals corresponding to a racemic mixture of (R6/S6)-1syn baskets.
Scheme 2. The synthesis of 5a/b and an electron-pushing mechanism to de-
scribe the first cyclialkylation with 5a/b transforming into 1syn/anti
.
8302
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8301 – 8305

Rapid Access to a Twisted Molecular Basket
COMMUNICATION
revealed two peaks corresponding to a racemic mixture of
(R6/S6)-1syn basket (Figure 2B).
curring in nonpolar solvents, thereby suggesting the involve-
ment of multiple CH3SO3H molecules in the protonation. In
fact, the rate law pertaining to the cyclization of monomeric
3 (Scheme 1) was also found to be higher order in acid (v=
kobs [3] [CF3SO3H]2, the Supporting Information, Fig-
ure S17). Furthermore, we observed the formation of two
reactive intermediates during the conversion of 5a/b into
1syn/anti (HPLC, the Supporting Information, Figure S18),
which is in line with the sequential nature of the transforma-
tion (Figure 3). Since the experimental rate law (v=kobs
[5a/b] [CF3SO3H]n, n=2 or 3) could be fit to both general
and specific-acid mechanistic scenarios, we examined the ad-
dition of CD3SO3D to 5a/b. In particular, we hypothesized
that a preequilibrium step would cause for deuterium atoms
to be incorporated into the reactant upon the initiation of
the reaction. In line with this premise, we found no deuteri-
um atoms (1H NMR spectra, the Supporting Information,
Figure S15) in 5a/b after ꢁ40% completion of the reaction.
Moreover, we found a primary kinetic isotope effect of
kH/kD =2.2 (the Supporting Information, Figure S19) sug-
gesting that the proton transfer step(s) are rate limiting;[23]
To investigate the mechanism of the macrocyclization of
5a/b, we used methods of both experimental and computa-
tional chemistry (Figures 3 and 4). First, the electron-push-
ing formalism suggests that the formation of the indanyl-
cation intermediate is followed by an intramolecular
Friedel–Crafts reaction to close the seven-membered ring
(Scheme 2); the process is then repeated twice to give
1
syn/anti. If the proton transfer(s) governs the rate law of the
reaction (Figure 3), then the general-acid catalysis (A-SE2
type mechanism) takes place.[21] Alternatively, if the forma-
tion of the s complex is rate-controlling, then the specific-
acid (A1 type, Figure 3) mechanism operates.[21] We moni-
tored the depletion of reactant 5a/b with time to find that the
reaction is first order in this compound (Figure 3). Addition-
ally, we found that the reaction is of a higher order in meth-
anesulfonic acid (Figure 3); note that the excessive amounts
of the acid (10–80 equiv) were used to secure the pseudo-
first order conditions. Importantly, a higher rate order in
acid was previously observed in electrophilic additions[22] oc-
Figure 3. Top: A potential energy diagram for the conversion of 5a/b into 1syn/anti with CH3SO3H (HA). Bottom: The change in the concentration of
5a/b (1.0 mm) with CH3SO3H (10–80 mm) was monitored by HPLC (334 K, ClCH2CH2Cl). The data were analyzed by a nonlinear least-square analysis
(SigmaPlot 12.0) and fitted to a first-order kinetic model to give kobs (R2 > 0.99). When pseudo-first-order coefficients kobs were plotted against the con-
centration of CH3SO3H, a second (R2 > 0.96) or third-order dependence (R2 > 0.97, see the fitted line) on this reactant appeared. The calculated thermo-
dynamic DG values (in kcalmolꢀ1) were computed at the TPSSh/6-311+G**//B3LYP/6-31G* level of theory.
Chem. Eur. J. 2012, 18, 8301 – 8305
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8303

´
J. D. Badjic et al.
note that discerning a complete role of the acid, in promot-
ing the formation of 1syn/anti, would necessitate additional ex-
perimentation.
of tris-indene [D6]5a/b (with the isotopes installed at posi-
tions 1 and 4, Figure 4A) to find product [D6]1syn/anti carrying
deuterium atoms at the original sites (see also the Support-
ing Information, Scheme S3). Evidently, the Wagner–Meer-
wein rearrangements (Figure 4A) were not occurring in the
reaction. This experimental observation is also in agreement
with computational results (TPSSh/6-311+G**//B3LYP/6-
31G*, Figure 4B) showing a considerable activation barrier
of 21.2 kcalmolꢀ1 for the 1,2-hydride shift. The suprafacial
1,4-sigmatropic shift of hydride (Figure 4B) was found to be
even more energy-demanding (DG°298 =53.8 kcalmolꢀ1).[18]
In fact, the conversion of the indanyl cation into the s com-
Along with the kinetic measurements, we utilized a com-
putational approach with density functional theory,[24] and
optimized various geometries of 5a/b, I1–I2 and 1syn/anti at the
B3LYP/6-31G* level of theory and then completed a single-
point energy calculation at the TPSSh/6-311+G** level of
theory (Figure 2).[18] We noted that there is a considerable
difference in thermodynamic stability of these constitutional
isomers. In brief, the conversion 5a/b!I1!I2!1syn/anti follows
a “downhill” trajectory in which the stability of the product
is 17.2 kcalmolꢀ1 (DG298, Figure 3) greater than that of the
plex appears to be the lowest energy pathway (DG°
=
298
reactant. The two diastereomeric products 1syn and 1anti, how-
9.9 kcalmolꢀ1, Figure 4B)[18] available to this high-energy in-
termediate.
syn/anti
ever, have
a
comparable energy content (DG298
=
ꢀ0.8 kcalmolꢀ1).[18]
A rapid access to concave compounds is critical for ob-
taining useful quantities of hosts capable of trapping smaller
molecules, for application in the area of sensing and cataly-
sis.[11,25] Our study describes an effective cyclialkylation reac-
tion that one can use for obtaining cavitands with a chiral
inner space. The stage is now set for exploring the scope
and characteristics of these concave hosts.
Importantly, a mixture of diastereomeric 5a/b (ꢁ1:5) was
found under all reaction conditions give 1syn/anti in the same
ꢁ1:5 ratio (Table 1). In line with the computational results,
we deduce that the reaction is under a kinetic control with
two reaction pathways occurring simultaneously (Scheme 2):
the RRR/SSS stereoisomer (compound 5a) converts into 1syn
basket whereas the other RRS/SSR stereoisomer (compound
5b) turns into 1anti. In line with such a proposition, we treated
1syn/anti with CH3SO3H (50.0 mm) to find no change in the
quantity or the proportion of the two compounds after a pro-
longed period of time (24 h). Furthermore, the intermediate
indanyl cation must be a subject of rapid hydride-shifts (Fig-
ure 4A), which may cause the “crossover” between two par-
allel reaction pathways. First, we completed the annulation
Experimental Section
Preparative procedure for 1syn/anti
: Methanesulfonic acid (6.8 mmol,
442 mL) was added to a flame-dried flask containing anhydrous 1,2-di-
chloroethane (136 mL), and the mixture was brought to reflux. Com-
pound 5a/b (656 mg, 1.42 mmol) was dissolved in anhydrous 1,2-dichloro-
ethane (6 mL) and then added to the reaction mixture with a syringe
pump over 2 h. After complete addi-
tion of the substrate, the reaction was
allowed to reflux for an additional 2 h.
The reaction mixture was then extract-
ed three times with 50 mL of saturated
sodium bicarbonate solution. The or-
ganic layer was dried over sodium sul-
fate and the solvent removed in vacuo.
The crude product was purified by
column chromatography (SiO2, hex-
anes/acetone=10:1) to yield 89.1 mg
of 1syn (14%) and 467.2 mg of 1anti
(71%).
1syn:
1H NMR (400 MHz, CDCl3): d=
7.24 (3H, d), 7.02–6.93 (6H, m), 6.90–
6.84 (3H, m), 4.14 (3H, d, J=4.8 Hz),
3.52 (3H, m), 3.41 (3H, dd, J=
16.8 Hz, 4.8 Hz), 2.81 (3H, d, J=
16.8 Hz), 2.52–2.43 (3H, m), 2.01 ppm
(3H,
d,
J=10.4 Hz);
13C NMR
(100 MHz, CDCl3): d=149.6, 146.65,
139.69, 126.92, 126.54, 122.91, 121.57,
41.00, 40.92, 40.25, 33.58 ppm; HRMS
(ESI): m/z: calculated for C36H30 +Na+
: 485.2240 [M+Na]+; found: 485.2250.
1anti:
1H NMR (400 MHz, CDCl3): d=
7.37–7.20 (3H, m), 7.18–6.94 (9H, m),
4.13 (1H, d, J=4.8 Hz), 4.09 (1H, d,
J=4.8 Hz), 4.06 (1H, d, J=4.8 Hz),
3.57–3.48 (3H, m), 3.40–3.24 (3H, m),
3.00–2.76 (3H, m), 2.49–2.32 (3H, m)
and 1.98–1.86 ppm (3H, m); 13C NMR
Figure 4. A) The occurrence of the Wagner–Meerwein shifts was tested with the [D6]5a/b substrate. B) The mac-
rocyclization of the indanyl cation of 3 was found (TPSSh/6-311+G**//B3LYP/6-31G*) to have a low activa-
tion barrier (DG° =9.9 kcalmolꢀ1) relative to possible hydride migrations.
8304
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8301 – 8305

Rapid Access to a Twisted Molecular Basket
COMMUNICATION
(100 MHz, CDCl3): d=150.46, 150.26, 150.04, 146.80, 146.77, 146.51,
139.78, 139.72, 139.46, 126.73, 126.69, 126.64, 126.61, 126.57 (2C), 126.54,
126.34, 126.34, 123.25, 123.10 (2C), 121.44, 121.28, 121.12, 40.90, 40.81,
40.78, 40.77, 40.46, 40.32 (two signals), 40.28, 34.92, 33.74, 33.46,
33.41 ppm; HRMS (ESI): m/z: calculated for C36H30 +Na+: 485.2240 [M+
Na]+, found: 485.2221.
[8] Molecular Encapsulation: Organic Reactions in Constrained Systems
(Eds.: U. H. Brinker, J.-L. Mieusset), Wiley-VCH, Weinheim, 2010.
[9] C. D. Tran, I. Mejac, J. Rebek, Jr., R. J. Hooley, Anal. Chem. 2009,
81, 1244–1254.
[10] R. Pinalli, M. Suman, E. Dalcanale, Eur. J. Org. Chem. 2004, 451–
462.
[11] J. Meeuwissen, J. N. H. Reek, Nat. Chem. 2010, 2, 615–621.
[12] a) D. Lucas, T. Minami, G. Iannuzzi, L. Cao, J. B. Wittenberg, P. An-
zenbacher, L. Isaacs, J. Am. Chem. Soc. 2011, 133, 17966–17976;
b) K. Suzuki, K. Takao, S. Sato, M. Fujita, J. Am. Chem. Soc. 2010,
132, 2544–2545.
[13] a) E. Arunkumar, C. C. Forbes, B. D. Smith, Eur. J. Org. Chem.
2005, 4051–4059; b) J. Szejtli, Encycl. Nanosci. Nanotechnol. 2004,
2, 283–304.
[14] J. D. Badjic, S. Stojanovic, Y. Ruan, Adv. Phys. Org. Chem. 2011, 45,
1–34.
[15] S. Rieth, K. Hermann, B.-Y. Wang, J. D. Badjic, Chem. Soc. Rev.
2011, 40, 1609–1622.
Acknowledgements
This work was financially supported with funds obtained from the Na-
tional Science Foundation (CHE-1012146) and the Defense Threat Re-
duction Agency (HDTRA1–11–1–0042). The content of the information
does not necessarily reflect the position or the policy of the federal gov-
ernment, and no official endorsement should be inferred. We thank the
Ohio Supercomputer Center for generous computational resources.
[16] L. Thomas, A. Polton, M. Tardi, P. Sigwalt, Macromolecules 1995,
28, 2105–2111.
[17] M. Givehchi, A. Polton, M. Tardi, M. Moreau, P. Sigwalt, J. P.
Vairon, Macromol. Symp. 2000, 157, 77–92.
Keywords: carbenium ions · cavitands · chirality · host–
guest chemistry · kinetics
[18] See the Supporting Information.
[19] S. J. Cristol, J. R. Mohrig, D. E. Plorde, J. Org. Chem. 1965, 30,
1956–1958.
[20] a) Superacid Chemistry, 2nd ed. (Eds.: G. A. Olah, G. K. Surya Pra-
kash, A. Molnar, J. Sommer), Wiley-VCH, Weinheim, 2009; b) Y.
Koltunov Konstantin, G. K. S. Prakash, G. Rasul, A. Olah George, J
Org Chem. 2002, 67, 4330–4336.
[1] a) A. Szumna, Chem. Soc. Rev. 2010, 39, 4274–4285; b) F.-G.
Klaerner, B. Kahlert, Acc. Chem. Res. 2003, 36, 919–932; c) F.
Fabris, C. Zonta, G. Borsato, O. De Lucchi, Acc. Chem. Res. 2011,
44, 416–423.
[2] D. J. Cram, Science, 1983, 219, 1177–1183.
[3] a) Calixarenes: An Introduction, 2nd ed. (Ed.: C. D. Gutsche), RSC,
Cambridge, 2008; b) L. R. MacGillivray, J. L. Atwood, J. Solid State
Chem. 2000, 152, 199–210; c) T. Brotin, J.-P. Dutasta, Chem. Rev.
2009, 109, 88–130; d) Container Molecules and Their Guests (Eds.:
D. J. Cram, J. M. Crams), RSC, Cambridge, 1997; e) J. Lagona, P.
Mukhopadhyay, S. Chakrabarti, L. Isaacs, Angew. Chem. 2005, 117,
4922–4949; Angew. Chem. Int. Ed. 2005, 44, 4844–4870.
[4] C. A. Schalley, J. Rebek, Jr., Stimul. Concepts Chem. (Eds.: J.F.
Stoddart, F. Vçgtle), Wiley-VCH, Weinheim, 2000, 199–210.
[5] a) C. J. Hastings, M. P. Backlund, R. G. Bergman, K. N. Raymond,
Angew. Chem. Int. Ed. 2011, 50, 10570–10573; b) C. L. D. Gibb,
A. K. Sundaresan, V. Ramamurthy, B. C. Gibb, J. Am. Chem. Soc.
2008, 130, 4069–4080.
[21] Comprehensive Chemical Kinetics, Vol. 13: Reactions of Aromatic
Compounds, (Eds.: C. H. Bamford, C. F. H. Tipper), Elsevier, Am-
sterdam, 1972.
[22] a) G. A. Olah, A. Molnar, Hydrocarbon Chemistry, 2nd ed., Wiley,
New York, 2003; b) Y. Pocker, R. F. Buchholz, J. Am. Chem. Soc.
1970, 92, 4033–4038.
[23] Comprehensive Chemical Kinetics, Vol. 8: Proton Transfer, (Eds.:
C. H. Bamford, C. F. H. Tipper), Elsevier, Amsterdam, 1977.
[24] W. Kohn, A. D. Becke, R. G. Parr, J. Phys. Chem. 1996, 100, 12974–
12980.
[25] a) A. E. Hargrove, S. Nieto, T. Zhang, J. L. Sessler, E. V. Anslyn,
Chem. Rev. 2011, 111, 6603–6782; b) P. Ballester, A. Vidal-Ferran,
P. W. N. M. van Leeuwen, Adv. Catal. 2011, 54, 63–126.
[6] R. Warmuth, S. Makowiec, J. Am. Chem. Soc. 2007, 129, 1233–1241.
[7] a) X. Bao, S. Rieth, S. Stojanovic, C. M. Hadad, J. D. Badjic, Angew.
Chem. 2010, 122, 4926–4929; Angew. Chem. Int. Ed. 2010, 49, 4816–
4819; b) A. Jasat, J. C. Sherman, Chem. Rev. 1999, 99, 931–967.
Received: April 4, 2012
Published online: June 4, 2012
Chem. Eur. J. 2012, 18, 8301 – 8305
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8305