Supramolecular catalysis at work: Diastereoselective synthesis of a molecular bowl with dynamic inner space

Article abstract of DOI:10.1021/jo701538g

(Figure Presented) The supramolecular assistance to Pd(0)/Cu(I)-catalyzed cyclotrimerization of stannylated norbornene 7 has been investigated to give molecular bowl 1_syn in a stereoselective fashion. Following a divergent strategy, racemic norbornene 7 was synthesized in satisfactory yield. Self-coupling, promoted by Pd(0)/Cu(I) catalysis acting in synergy with CsF, yielded molecular bowl 1_syn in a moderate 30% yield. The reaction diastereoselectivity is affected by the concentration of Cu(I) and Cs ⁺: increasing quantities of the cations enhanced the synlanti ratio of the isolated cyclotrimer from statistical (1:3) to a more desirable (4.5:1) ratio, in favor of the molecular bowl 1_syn. ¹H NMR spectroscopic studies suggested the coordinating affinity of 1_syn toward transition metals Cu(I), Ag(I), and Au(I), to account for the observed templation effect. In particular, the tridentate 1_syn has been shown to bind to one Ag(I) cation in the assembly process that is driven with enthalpy (ΔH° = -19 ± 2 kcal/mol, ΔS° = -45 eu). The complete coordination was not cooperative, and was hypothesized to be impeded with the adverse entropy. Accordingly, density functional theory (BP86) calculations of 1_syn and its mono-, bis-, and tris-Ag(I) complexes suggested that the coordination of one to three silver cations is highly exothermic. The calculations also revealed that the bowl constriction is necessary for the aromatic arms to become preorganized and bind to a silver cation(s) (ΔE ≈ 8 kcal/mol). Ultimately, Ag(I) has been shown to assist the diastereoselective formation of 1_syn, lending support to the notion of template-directed synthesis.

Full text of DOI:10.1021/jo701538g

VOLUME 73, NUMBER 2
JANUARY 18, 2008
© Copyright 2008 by the American Chemical Society
Supramolecular Catalysis at Work: Diastereoselective Synthesis of a
Molecular Bowl with Dynamic Inner Space
Zhiqing Yan, Troy McCracken, Shijing Xia, Veselin Maslak, Judith Gallucci,
Christopher M. Hadad, and Jovica D. Badjic´*
Department of Chemistry, The Ohio State UniVersity, 100 West 18th AVenue, Columbus, Ohio 43210
badjic@chemistry.ohio-state.edu
ReceiVed July 16, 2007
The supramolecular assistance to Pd(0)/Cu(I)-catalyzed cyclotrimerization of stannylated norbornene 7
has been investigated to give molecular bowl 1syn in a stereoselective fashion. Following a divergent
strategy, racemic norbornene 7 was synthesized in satisfactory yield. Self-coupling, promoted by Pd(0)/
Cu(I) catalysis acting in synergy with CsF, yielded molecular bowl 1syn in a moderate 30% yield. The
reaction diastereoselectivity is affected by the concentration of Cu(I) and Cs+: increasing quantities of
the cations enhanced the syn/anti ratio of the isolated cyclotrimer from statistical (1:3) to a more desirable
(4.5:1) ratio, in favor of the molecular bowl 1syn. 1H NMR spectroscopic studies suggested the coordinating
affinity of 1syn toward transition metals Cu(I), Ag(I), and Au(I), to account for the observed templation
effect. In particular, the tridentate 1syn has been shown to bind to one Ag(I) cation in the assembly process
that is driven with enthalpy (∆H° ) -19 ( 2 kcal/mol, ∆S° ) -45 eu). The complete coordination was
not cooperative, and was hypothesized to be impeded with the adverse entropy. Accordingly, density
functional theory (BP86) calculations of 1syn and its mono-, bis-, and tris-Ag(I) complexes suggested that
the coordination of one to three silver cations is highly exothermic. The calculations also revealed that
the bowl constriction is necessary for the aromatic arms to become preorganized and bind to a silver
cation(s) (∆E ≈ 8 kcal/mol). Ultimately, Ag(I) has been shown to assist the diastereoselective formation
of 1syn, lending support to the notion of template-directed synthesis.
Introduction
and cavitands with controllable conformational dynamics2 and
regulatory operational mechanisms,3 has the prospect of fun-
damentally addressing the relationship between recognition and
reactivity in artificial environments.4 The inspiration comes from
Broadening the scope and deepening the understanding of
molecular recognition and assembly in an unnatural setting
requires a steady development of functional hosts capable of
binding guests in a constrictive manner.1 Our research program,
focused on the design and study of a family of molecular baskets
* Address correspondence to this author. Phone: (614) 247-8342. Fax: (614)
292-1685.
10.1021/jo701538g CCC: $40.75 © 2008 American Chemical Society
Published on Web 10/11/2007
J. Org. Chem. 2008, 73, 355-363
355

Yan et al.
FIGURE 1. (A) The cyclotrimerization of a racemic norbornene, promoted with a transition metal catalyst, typically yields a mixture of syn/anti
diastereomers; the anti compound is statistically favored.7,8 (B) Presumably, a metal cation could assist the annulation to occur in a diastereoselective
fashion.
the natural world where biological molecules regulate the
binding, transition-state stabilization, and trafficking of mol-
ecules via a sequence of controlled and synchronized confor-
mational changes.
compound is typically the principal constituent (Figure 1A).7
This statistical challenge limits access to greater quantities of
molecular baskets and is frustrating for subsequent molecular
recognition studies. A need for a practical diastereoselective
cyclotrimerization exists, and De Lucchi et al. have invested
considerable effort in developing effective synthetic routes and
understanding the annulation mechanisms.7,8
Molecular baskets, of interest to our program, comprise a
C3V symmetrical tris-norbornadiene framework (Figure 1A). The
structural framework provides a concave curvature,5 and is
synthesized via tris-annulation of a functionalized norbornene
with a transition metal catalyst.6 In general, the reaction gives
a syn/anti diastereomeric mixture whereby undesired anti
The study presented here communicates an approach, whereby
the enantiodiscrimination of a racemic norbornene reactant is
assisted with a metal cation as template to direct the formation
of a molecular bowl in a diastereoselective fashion (Figure 1B).
In addition, the Stille-type cyclotrimerization has, for the first
time, been optimized with a palladium(0)/copper(I) catalytic pair
to act in synergy with CsF.9 Molecular bowl 1syn (Figure 2A)
has been designed in a modular way, and incorporates three
identical o-phenylenediamine moieties in the quinoxaline form.
Conceivably, varying the nature of the diamine component will
afford access to a greater number of dynamic receptors and
baskets. More spacious inner space of the synthesized molecular
bowl is capable of reversibly adjusting its shape and size in
response to the presence of silver(I) cations, presenting a stimuli-
responsive and dynamic character of this emerging class of
hosts.
(1) (a) Judice, J. K.; Cram, D. J. J. Am. Chem. Soc. 1991, 113, 2790-
2791. (b) Quan, M. L. C.; Cram, D. J. J. Am. Chem. Soc. 1991, 113, 2754-
2755. (c) Cram, D. J.; Blanda, M. T.; Paek, K.; Knobler, C. B. J. Am. Chem.
Soc. 1992, 114, 7765-7773. (d) Nakamura, K.; Houk, K. N. J. Am. Chem.
Soc. 1995, 117, 1853-1854. (e) Houk, K. N.; Nakamura, K.; Sheu, C.;
Keating, A. E. Science 1996, 273, 627-629. (f) Sheu, C.; Houk, K. N. J.
Am. Chem. Soc. 1996, 118, 8056-8070. (g) Raymo, F. M.; Houk, K. N.;
Stoddart, J. F. J. Am. Chem. Soc. 1998, 120, 9318-9322. (h) Ihm, C.; Kim,
M.; Ihm, H.; Paek, K. J. Chem. Soc., Perkin Trans. 2 1999, 8, 1569-1575.
(i) Chapman, R. G.; Sherman, J. C. J. Am. Chem. Soc. 1999, 121, 1962-
1963. (j) Warmuth, R.; Yoon, J. Acc. Chem. Res. 2001, 34, 95-105. (k)
Yoon, J.; Warmuth, R. Org. Lett. 2007, 9, 2883-2886.
(2) (a) Maslak, V.; Yan, Z.; Xia, S.; Gallucci, J.; Hadad, C. M.; Badjic´,
J. D. J. Am Chem. Soc. 2006, 128, 5887-5894. (b) Yan, Z.; Xia, S.; Gardlik,
M.; Seo, W.; Hadad, C. M.; Badjic´, J. D. Org. Lett. 2007, 9, 2301-2304.
(3) Yan, Z.; Chang, Y.; Mayo, D.; Xia, S.; Maslak, V.; Badjic´, J. D.
Org. Lett. 2006, 8, 1697-3700.
(6) (a) Gassman, P. G.; Gennick, I. J. Am. Chem. Soc. 1980, 102, 6863-
6864. (b) Durr, R.; De Lucchi, O.; Cossu, S.; Lucchini, V. Chem. Commun.
1996, 2, 2447-2448. (c) Durr, R.; Cossu, S.; Lucchini, V.; De Lucchi, O.
Angew. Chem., Int. Ed. Engl. 1997, 36, 2805-2807. (d) Sakurai, H.; Daiko,
T.; Hirao, T. Science 2003, 301, 1878-1882.
(4) (a) Cuevas, F.; Di Stefano, S.; Magrans, J. O.; Prados, P.; Mandolini,
L.; de Mendoza, J. Chem. Eur. J. 2000, 6, 3228-3234. (b) Davis, A. V.;
Yeh, R. M.; Raymond, K. N. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4793-
4796. (c) Zyryanov, G. V.; Rudkevich, D. M. Org. Lett. 2003, 5, 1253-
1256. (d) Gianneschi, N. C.; Bertin, P. A.; Nguyen SonBinh T.; Mirkin, C.
A.; Zakharov, L. N.; Rheingold, A. L. J. Am. Chem. Soc. 2003, 125, 10508-
10509. (e) Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2004, 43, 6748-6751. (f) Yoshizawa, M.; Miyagi, S.; Kawano,
M.; Ishiguro, K.; Fujita, M. J. Am. Chem. Soc. 2004, 126, 9172-9173. (g)
Palmer, L. C.; Rebek, J., Jr. Org. Biomol. Chem. 2004, 2, 3051-3059. (h)
Kaanumalle, L. S.; Gibb, C. L. D.; Gibb Bruce, C.; Ramamurthy, V. J.
Am. Chem. Soc. 2004, 126, 14366-14367. (i) Hooley, R. J.; Rebek, J., Jr.
J. Am. Chem. Soc. 2005, 127, 11904-11905. (j) Luetzen, A. Angew. Chem.,
Int. Ed. 2005, 44, 1000-1002. (k) Lagona, J.; Mukhopadhyay, P.;
Chakrabarti, S.; Isaacs, L. Angew. Chem., Int. Ed. 2005, 44, 4844-4870.
(l) Yoshizawa, M.; Tamura, M.; Fujita, M. Science 2006, 312, 251-254.
(5) (a) Siegel, J. Angew. Chem., Int. Ed. Engl. 1994, 33, 1721-1723.
(b) Frank, N. L.; Baldridge, K. K.; Gantzel, P.; Siegel, J. S. Tetrahedron
Lett. 1995, 36, 4389-4392. (c) Buergi, H-B.; Baldridge, K. K.; Hardcastle,
K.; Frank, N. L.; Gantzel, P.; Siegel, J. S.; Ziller, J. Angew. Chem., Int. Ed.
Engl. 1995, 34, 1454-1456.
(7) (a) Zonta, C.; Cossu, S.; Peluso, P.; De Lucchi, O. Tetrahedron Lett.
1999, 40, 8185-8188. (b) Cossu, S.; De Lucchi, O.; Paulon, A.; Peluso,
P.; Zonta, C. Tetrahedron Lett. 2001, 42, 3515-3518. (c) Peluso, P.; De
Lucchi, O.; Cossu, S. Eur. J. Org. Chem. 2002, 23, 4032-4036.
(8) (a) Fabris, F.; De Martin, A.; De Lucchi, O. Tetrahedron Lett. 1999,
40, 9121-9124. (b) Paulon, An.; Cossu, S.; De Lucchi, O.; Zonta, C. Chem.
Commun. 2000, 19, 1837-1838. (c) Cossu, S.; Cimenti, C.; Peluso, P.;
Paulon, A.; De Lucchi, O. Angew. Chem., Int. Ed. 2001, 40, 4086-4089.
(d) Borsato, G.; De Lucchi, O.; Fabris, F.; Groppo, L.; Lucchini, V.;
Zambon, A. J. Org. Chem. 2002, 67, 7894-7897. (e) Borsato, G.; De
Lucchi, O.; Fabris, F.; Lucchini, V.; Pasqualotti, M.; Zambon, A. Tetra-
hedron Lett. 2002, 44, 561-563. (f) Fabris, F.; Bellotto, L.; De Lucchi, O.
Tetrahedron Lett. 2003, 44, 1211-1213. (g) Zonta, C.; Fabris, F.; De Lucchi,
O. Org. Lett. 2005, 7, 1003-1006.
(9) (a) Liebeskind, L. S.; Fengi, R. W. J. Org. Chem. 1990, 55, 5359-
5364. (b) Han, X.; Stolz, B. M.; Corey, E. J. J. Am. Chem. Soc. 1999, 121,
7600-7605. (c) Casado, A. L.; Espinet, P. Organometallics 2003, 22, 1305-
1309. (d) Mee, S. P. H.; Lee, V.; Baldwin, J. E. Angew. Chem., Int. Ed.
2004, 43, 1132-1136.
356 J. Org. Chem., Vol. 73, No. 2, 2008

DiastereoselectiVe Synthesis of a Molecular Bowl
1
FIGURE 2. (A) Synthesis of compound 1syn and its H NMR spectroscopic assignment. (B) Synthesis of compound 8. (C) Synthesis of com-
pound 10.
Results and Discussion
as a template14 in the formation of the cyclotrimer: the process
of enantiomeric discrimination, in the course of homocoupling
versus heterocoupling of racemic 7, may well be assisted with
the template to favor the syn product.
Synthesis. The synthesis of 1syn is presented in Figure 2A.
Compounds 2 and 3 were obtained independently, and then
subjected to condensation to give 4. The strategy is divergent
to allow for a number of accessible diamines, akin to 1,2-
diaminobenzene 3, to be ultimately used in the synthesis; in
this way, the versatility of the procedure is greatly enhanced.
Furthermore, 5-norbornene-2,3-dione 2 is easily obtained in
gram quantities from a Diels-Alder reaction of vinylene
carbonate10 and cyclopentadiene followed by the hydrolysis and
Swern oxidation of subsequently formed cycloadduct.11 The
bromination of 4 was completed in boiling decalin at 150 °C to
circumvent Wagner-Meerwein skeletal rearrangements (Figure
2A).12 Both cis and trans addition of bromine took place, with
the stereochemistry of the cis addition preferring the exo side
of the bicyclic ring.13 Subsequent elimination of the halogen in
5a/b, assisted with t-BuOK, and stannylation afforded bromo-
(trimethylstannyl) 7. This compound is preset to undergo the
cyclotrimerization, when prompted with a transition metal
catalyst.
With this presumption, we first attempted the cyclotrimer-
ization of 7 using Cu(I) and Cu(II) compounds to catalyze and
simultaneously template the formation of desired 1syn (Table 1,
entries 1-3). Copper-based reagents proved ineffective in
promoting the transformation: a mixture of unidentifiable
oligomers and destannylated products was typically observed.
It is important to note, though, that the cyclotrimerization of a
variety of stannylated norbornenes assisted with copper catalysts
had, as reported earlier,6c,7a,8d been proven useful. The self-
coupling of 7, under the Stille conditions, also failed to give
the desired product.8b It surprisingly, led to a transfer of two
methyl groups to the trigonal carbons in 7 to give 8 in 19%
yield (Figure 1B). The X-ray structural analysis confirmed the
identity of this compound.15 The transfer of an alkyl group to
a trigonal carbon is typically a slow process.16 It is the sole
reason for the widespread use of trialkyltin derivatives in Stille
couplings, with alkyl groups as “nontransferable” ligands. Herein
observed simultaneous formation of two carbon-carbon bonds
Cyclotrimerization Studies. Modularly designed 7 has two
aromatic nitrogens with an intrinsic ability to complex transition
metals, perhaps, in the course of its coupling into 1 (Figure
1B). That is to say, a metal cation can be hypothesized to act
(14) For recent reviews explaining the principles of template-directed
synthesis see: (a) Jager, R.; Vogtle, F. Angew. Chem., Int. Ed. Engl. 1997,
36, 930-944. (b) Hubin, T. J.; Kolchinski, A. G.; Vance, A. L.; Busch, D.
H. AdV. Supramol. Chem. 1999, 5, 237-357. (c) Arico, F.; Badjic´, J. D.;
Flood, A.; Leung, K.; Liu, Y.; Stoddart, J. F. Top. Curr. Chem. 2005, 249,
203-261. (d) Lankshear, M. D.; Beer, P. D. Coord. Chem. ReV. 2006, 250,
3142-3160.
(10) Newman, M. S.; Addor, R. W. J. Am. Chem. Soc. 1955, 77, 3789-
3793.
(11) Kobayashi, T.; Kobayashi, S. Molecules 2000, 5, 1062-1067.
(12) (a) Colter, A.; Friedrich, E. C.; Holness, N. J.; Winstein, S. J. Am.
Chem. Soc. 1965, 87, 378-379. (b) Walling, C. Acc. Chem. Res. 1983, 16,
448-454. (c) Tutar, A.; Taskesenligil, Y.; Cakmak, O.; Abbasoglu, R.;
Balci, M. J. Org. Chem. 1996, 61, 8297-8300.
(15) See the Supporting Information for more details.
(16) (a) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508. (b)
Farina, V.; Krishnamurthy, V.; Scott, W. J. Organic Reactions; Paquette,
L. A., Ed.; John Wiley & Sons, Inc.: New York, 1997; Vol. 50. (c) Espinet,
P.; Echavarren, A. M. Angew. Chem., Int. Ed. 2004, 43, 4704-4734.
(13) Kobayashi, T.; Miki, K. Bull. Chem. Soc. Jpn. 1998, 71, 1443-
1449.
J. Org. Chem, Vol. 73, No. 2, 2008 357

Yan et al.
TABLE 1. The Cyclotrimerization of 7 into 1syn Was Conducted under Modified Stille Conditions and Assisted with Cu(I), Cs+, and Ag(I)
Cations in a Template-Directed Fashion
inorganic component
(molar equiv)a
overall yield
syn/anti
entry
Pd catalyst
Cu salt (%)a
of 1,b %
1
1
2
CuTc
CuI
0
0
LiCl
3
Cu(NO3)2
0
4
5
6
7
8
9
10
11
12
13
Pd(OAc)2
Pd(PPh3)4
PdCl2
LiCl
0
0
CuI (0)
CuI (4)
CsF (2.0)
CsF (2.0)
CsF (2.0)
CsF (2.0)
CsF (2.0)
TBAF (2.0)
NaF (2.0)
CsF (1.0)
CsF (2.0)
CsClO4 (8.0)
CsF (2.0)
AgOTf (2.0)
24
29
35
23
0
0
23
24
1:3
1:1.4
1.3:1
2.3:1
PdCl2
CuI (20)
CuI (20)
CuI (70)
CuI (20)
CuI (20)
CuI (20)
CuI (20)
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
1.5:1
4.5:1
14
Pd(PPh3)4
CuI (20)
20
4.4:1
a With respect to the quantity of 7. b Isolated yield.
(Figure 1B) is unique,15 and likely facilitated by a slow
transmetalation step whereby the transfer of an electron-deficient
vinyl group is retarded in a sterically congested chemical
environment. In a Heck-type promoted self-coupling of 6 to
give 1 (Table 1; entry 4), the starting material was only partially
recovered.7b
Baldwin et al. have recently revealed a synergistic effect of
Cu(I) salts and fluoride anion in the Stille coupling of electron-
deficient stannanes with vinyl/aryl halides and triflates.9c When
the procedure was applied in the cyclotrimerization of 7,
compound 1 was obtained as a mixture of the syn/anti isomers
in a moderate yield (Table 1; entries 6-9). The reaction
diastereoselectivity has, intriguingly, been found to be a function
of the amount of Cu(I): (a) in the absence of copper, there was
no reaction (Table 1; entry 5), (b) low amounts of copper (4%)
led to a 1:3 statistical syn/anti ratio of the diastereomers (Table
1, entry 6), and (c) higher concentrations of Cu(I) (20-70%)
afforded 1syn as the major product (Table 1; entries 7-9). It is
SCHEME 1. A Proposed Mechanism for the Cyclotrimerization of Racemic 7
358 J. Org. Chem., Vol. 73, No. 2, 2008

DiastereoselectiVe Synthesis of a Molecular Bowl
generally believed that Cu(I) accelerates the Stille coupling via
enhancing the transmetallation step whereby a more reactive
organocopper intermediate is formed from the organostannane
(Scheme 1).9a,b The present case, evidently, implicates an
additional Cu(I) role in the catalytic cycle: the cation perhaps
augments the reaction diastereoselectivity via templation (Figure
1B and Scheme 1). Alternatively, F- has been hypothesized to
contribute in removing tin salts from the catalytic equilibrium,
which in turn supports the formation of more reactive organo-
copper intermediate so that the reaction becomes accelerated
(Scheme 1).9c In the trimerization of 7, the nature of the fluoride
counterbalancing cation has been revealed to be critical: cesium,
but not Bu4N+ or Na+, supported the formation of 1 (Table 1,
entries 10-13). Even more, the increased presence of the large
and soft Cs+ cation (from 1.0 to 10.0 molar equiv) enhanced
the reaction diastereoselectivity practically by a factor of 3
(Table 1, entries 12 and 13). Evidently, Cs+ and Cu(I) cations
work in synergy and their mutual presence is necessary for
converting 7 into 1.
1
FIGURE 3. A series of H NMR (500 MHz, 298 K) spectra of the
following: (a) 1syn (9.4 mM, CD3CN/CDCl3, 5:1), (b) 1syn (9.4 mM,
CD3CN/CDCl3, 5:1) and 3.3 mol equiv of CsPF6, (c) 1syn (9.4 mM,
CD3CN/CDCl3, 1:1) and 4.4 mol equiv of [Cu(I)(CH3CN)4]PF6, (d)
1syn (2.3 mM, CD3OD/CDCl3, 1:1) and 4.8 mol equiv of AgOTf, (e)
1syn (5.9 mM, CD3CN) and 6.2 mol equiv of [Au(I)PPh3]Cl, and (f)
1
syn (9.4 mM, CD3CN/CDCl3, 5:1) and 3.3 mol equiv of CsPF6 and 24
mol equiv of [Cu(I)(CH3CN)4]PF6.16 For the H NMR spectroscopic
1
In a control experiment, we conducted the cyclotrimerization
of naphthobenzonorbornadiene derivative 9, lacking the aromatic
nitrogens (Figure 2C). Almost a statistical syn/anti ratio (1:2.6)
of the isolated cyclotrimer 10 was obtained, further supporting
assignment of 1syn, see Figure 2A.
created the greatest perturbations in the NMR signals.16 In a
control experiment,16 AgOTf was added to a solution of
molecular bowl 10syn/anti,
16 lacking the aromatic nitrogens. The
the notion of template-directed formation of 1syn
.
absence of any detectable spectroscopic changes for 10syn is
consistent with the notion that the heterocyclic arms are critical
for coordinating to the transition metal cation.16
Catalytic Mechanism. A tentative mechanism for the chain-
type conversion (Figure 1B) of 7 to 1 is summarized in Scheme
1. Supposedly, Cu(I) and Cs+ cations promote homo- over
heterochiral coupling of racemic 7. That is to say, in case of
homochiral C2 symmetrical intermediates and 1syn, the coordina-
tion of the cations to the aromatic nitrogens can be hypothesized
to assist their formation and the subsequent cross-coupling. The
coordination, in particular, can preorganize the homochiral
intermediate containing three norbornenes to the extent that the
activation energy for the benzene ring annulation is minimized
(Scheme 1). In fact the analogous Cs symmetrical intermediates
(only one shown in Scheme 1) must be twisted upon complexing
the metal cations. The twisted and “out-of-plane” geometry
would disfavor the transmetalation and the overall intramolecular
coupling providing a mechanism for the oligomerization.
In the titration experiment with a standard solution of AgOTf
1
(0.168 M, CD3OD/CDCl3 ) 1:1), the H NMR signals of 1syn
(2.3 mM, CD3OD/CDCl3 ) 1:1) shifted downfield as the
concentration of Ag(I) was gradually increased (Figure 4A). The
break in the titration curve at 1.0 mM concentration of the metal
cation (Figure 4B) points to 1:1 binding stoichiometry;17 indeed,
the method of continuous variation (Job plot) confirmed this
stoichiometric ratio.16,18 In addition, a MALDI-TOF mass
spectrometric measurement of a solution of 1syn and AgOTf also
suggest the existence of the [1syn:Ag]+ complex as the major
component of the mixture.16 The nonlinear curve fitting of the
binding isotherm (Figure 4B) to a 1:1 equilibrium binding model
yielded an apparent association constant of Kavg ) (5.6 ( 1.6)
It is interesting to note that the increasing amounts of the
syn isomer, produced in the catalysis, suppressed the formation
of the anti diastereomer (Table 1; entries 6-9, 12, and 13). In
that respect, the overall reaction yields, for all transformations,
were rather low and comparable (20-35%). One could antici-
pate that the enhanced formation of the syn isomer, promoted
with the “correct” metal cation, would also be accompanied by
the increased overall reaction yield. Apparently, the undesired
“side” reactions (possible oligomerizations and such) are
kinetically favored to a great extent. The presence of the
templating metal cations evidently upsets the kinetics of the
syn/anti formation, yet making the cyclotrimerization still not
competitive to the fast formation of the side product(s).
× 103 M-1 16,17
.
The observed low-field 1H NMR chemical shifts
of the signals in 1syn (Figure 4A), upon addition of Ag(I), are
consistent with the coordination of a positively charged Ag(I)
to two adjacent aromatic heterocyclic rings in 1syn
: the
coordination reorganizes the electron density making the proton
nuclei magnetically deshielded. Interestingly, the signals for the
bridge Ha/Hb protons experienced a considerable change (∆δ
) 0.22), despite their angular position that depresses through-
bond communication with the aromatic appendages.19 To
accommodate a silver cation, the aromatic arms in 1syn have to
fold inside (Figure 6B) causing the bridge carbon and its protons
to move away from the center of the arms. With such
conformational changes, the diamagnetic effect imposed on the
1H NMR Spectroscopic Studies. To examine the complex-
ation affinity of different metals toward molecular bowl 1syn
and to gain more insight into the templation mechanism, we
originally conducted a series of 1H NMR spectroscopic titrations
(Figures 3 and 4). A gradual addition of cesium salt (CsPF6) to
a solution of 1syn (3.5 mM, CD3CN) caused minor changes in
(17) Wilcox, C. Frontiers in Supramolecular Organic Chemistry and
Photochemistry; Schneider, H-J., Durr, H., Eds.; Wiley-VCH: Weinheim,
Germany, 1991; pp 123-143.
(18) (a) Connors, K. A. Binding Constants, The Measurement of
Molecular Complex Stability; Wiley-Interscience: New York, 1987; pp 24-
28. (b) Blanda, M. T.; Horner, J. H.; Newcomb, M. J. Org. Chem. 1989,
54, 4626-4636.
(19) (a) Wilcox, C. F.; Leung, C. J. J. Am. Chem. Soc. 1968, 90, 336-
341. (b) Cole, T. W., Jr.; Mayers, C. J.; Stock, L. M. J. Am. Chem. Soc.
1974, 96, 4555-4557. (c) Grubbs, E. J.; Wang, C.; Deardurff, L. A. J.
Org. Chem. 1984, 49, 4080-4082.
1
its H NMR spectrum (Figure 3b).16 In contrast, when [Cu-
(CH3CN)4]PF6, AgOTf, or [Au(PPh3)]Cl were individually
titrated to a solution of 1syn (Figure 3c-e), the 1H NMR signals
were affected: Cu(I) imposed the least change, while Ag(I)
J. Org. Chem, Vol. 73, No. 2, 2008 359

Yan et al.
1
FIGURE 4. (A) A series of H NMR spectra (500 MHz, 298 K, CD3OD/CDCl3, 1:1) of 1syn (2.3 mM) recorded on gradual addition of 0.168 M
standard solution of AgOTf (CD3OD/CDCl3, 1:1) such that the final mixture comprises (a) 0, (b) 0.24, (c) 0.73, (d) 1.45, (e) 2.91, and (f) 4.84 mol
1
equiv of AgOTf. (B) The nonlinear curve fitting of the H NMR chemical shifts of the Ha resonance in 1syn to a 1:1 equilibrium model.22,23
FIGURE 5. (a) Isothermal titration calorimetry (ITC) data for the titration of a solution of 1syn (0.173 mM, 1.40 mL) with a 1.75 mM standard
solution of AgOTf (CHCl3/CH3OH ) 1:1) at 300 K. (b) Computer simulated curve fitting afforded the thermodynamic parameters for the assembly.
Ha/b protons becomes less pronounced, which, we reason, then
rendered the observed downfield changes. The host-guest
exchange, and the Ag(I) revolving about 1syn, must be rapid on
the NMR time scale, to explain the observed proton signals
corresponding to an averaged C3V symmetrical species (Figure
4A).
host-guest complex (∆H° ) -19 ( 2 kcal/mol), and the
enthalpy/entropy compensation effect.20 The solvent reorganiza-
tion must also contribute, although in a more complicated
fashion.
Theoretical Calculations. Density functional theory (DFT,
BP86 functional) calculations21 of the optimized geometry of
Isothermal Titration Calorimetry (ITC). The thermody-
namic profile for the binding of Ag(I) to 1syn was further
investigated with isothermal titration calorimetry (Figure 5). A
standard solution of AgOTf (1.75 mM, CH3OH/CHCl3, 1:1) was
successively added to a solution of host 1syn (0.173 mM, CH3-
OH/CHCl3, 1:1), contained in the sample cell. The total heat
released in the interaction (∆H°) was measured to give the
binding isotherm (Figure 5A). The experiments showed that the
assembly is exothermic (∆H° ) -19 ( 2 kcal/mol). The fitting
to the one-site model revealed Ka ) (1.4 ( 0.2) × 104 M-1
(∆G° ) -5.7 ( 0.5 kcal/mol, 300.0 K) and N ) 0.5 as
independent fit parameters. The association constant is, appar-
ently, in fair agreement with the one obtained from the 1H NMR
titration (Figure 4B). The observed binding stoichiometry (N)
is noticeably off the scale. Supposedly, the existence of two
other thermodynamic states, in which Ag(I) is coordinated to
the second and third binding site in 1syn, can in part be accounted
for the apparent “deviations” in the theoretical isotherm (Figure
5b) and the off-scale stoichiometry. The self-association of 1syn
in solution can also contribute to the effect (see the paragraph
about the X-ray diffraction studies and Figure 7B). The
formation of single complexed 1syn is to a large extent hampered
by entropy (Figure 5, ∆S° ) -45 eu). This loss of entropy can,
to a first approximation, be ascribed to the rigidity of the formed
1syn and its corresponding complex with a Ag(I) cation [1syn:
Ag]+ suggested a high thermodynamic stability of the complex
(∆E ) -9.0 kcal/mol, Figure 6B). For the coordination to take
place, two aromatic arms in 1syn need to move toward the inner
space to bring the aromatic nitrogens to the proper distance (∆D
) 1.0 Å). The conformational change is accompanied by the
decrease in the dihedral angle (∆R ) 2.0°) and relocation of
the bridge protons somewhat away from the neighboring
quinoxaline aromatic rings (Figure 6B); the calculation is indeed
in agreement with the observed 1H NMR chemical changes for
Ha/b protons (Figure 4). What does preVent three silVer cations
from cooperatiVely coordinating to the nitrogen ligands of the
bowl and thus causing its complete folding?
To address this matter, the synchronized inward/outward
motion of the arms in the bowl itself has been subjected to a
series of DFT calculations whereby each energy minimization
had the distance between the aromatic nitrogens constrained,
with all other parameters being fully optimized (Figure 6A).
The calculated potential energy profile for the bowl “swinging”
turned out to be asymmetric: the bowl constriction about the
(20) Ruloff, R.; Seelbach, U. T.; Merbach, A. E.; Klarner, F.-G. J. Phys.
Org. Chem. 2002, 15, 189-196.
(21) (a) Perdew, J. P. Phys. ReV. B 1986, 33, 8822. (b) Becke, A. D.
Phys. ReV. A 1998, 38, 3098.
360 J. Org. Chem., Vol. 73, No. 2, 2008

DiastereoselectiVe Synthesis of a Molecular Bowl
FIGURE 6. Energy minimized (DFT, BP86) conformation of 1syn, and calculated potential energy diagrams for the synchronized inward/outward
motion of its arms as a function of the diherdal angle R and the distance between the adjacent nitrogens D. (B) Energy minimized structures (BP86)
of mono-, bis-, and tris-coordination complexes of 1syn with Ag(I).
energy well is endothermic, while the energy required to open
is more neutral (Figure 6A). The energy required to constrict
the dihedral angles (∆R ) 3.5°) and at the same time dislocate
the aromatic nitrogens (∆D ) 1.08 Å) in 1syn, to preorganize
the molecule for coordination to three Ag(I) cations is high (∆E
> 8 kcal/mol, Figure 6A). The formation of three Ag-N
coordinative bonds, based on these calculations, should, how-
ever, be sufficiently exothermic to make the overall complex-
ation thermodynamically favorable (at least -20.7 kcal/mol,
Figure 6B). On the other hand, the great entropic cost that
FIGURE 7. (A) Energy minimized (DFT, BP86) conformation of 1syn
accompanies the binding of the first Ag(I) to 1syn (∆S°(expt)
)
coordinated to three Cu(I) cations and containing CsF at its top rim.
(B) Stick representation of the structure of the 1syn dimer in the solid
state (hydrogen atoms are omitted for clarity).
-45 eu, Figure 5B) is likely to propagate to the second and
third coordination, making the overall interaction thermody-
namically unfeasible.
level. The observed synergistic action of two cations, Cu(I) and
Cs+, in the templation of 1syn can thus, to an extent, be
rationalized.
Thermodynamic Considerations and Rational Summation
of the Supramolecular Assistance to Catalysis. The observed
low propensity of Cu(I) and Au(I) to coordinatively bridge the
arms in 1syn (Figure 3) can now be ascribed to the copper size
and the gold complexing character. In particular, the coordina-
tion of small Cu(I) and soft Au(I) cations can be hypothesized
to provide insufficient enthalpic driving force, which is neces-
sary to surmount the energy required for the bowl to acquire
its optimal binding geometry; the entropy effect needs to be
taken into account, as well. In the case of copper, however, its
interaction with 1syn and other intermediates in the course of
the cyclotrimerization (Scheme 1) is strong enough to provide
sufficient transition-state stabilization to boost the reaction
diastereoselectivity.
In consideration of the Ag(I) aptitude toward coordinating
1syn, it is conceivable to presume for this cation to template the
formation of the bowl as well. Indeed, when AgOTf (2.0 mol
equiv) was used in the cyclotrimerization of 7, under the
standard conditions, 1syn was obtained in excess (Table 1, entry
14).22 The reaction diastereoselectivity, in favor of the syn
isomer (4.4:1), lends support to the notion of a template-directed
mechanism (Scheme 1).
X-Ray Diffraction Studies. A single-crystal X-ray study of
1syn revealed a repeating motif comprised of two fully entangled
molecular bowls, bridged with three molecules of water (Figure
7B).23Each bowl places an arm in the cavity of its partner at
the π-π stacking distance (3.5 Å). Notably, the π-deficient
portion of one arm is against the π-rich portion of another. The
directional assembly in the solid state is thus in accord with
the electrostatic Hunter-Sanders model for aromatic interac-
tions.31 Furthermore, the observed ordering allows the bowl to
fill its inner space in the entropically most efficient manner,
The coordinating affinity of Cu(I) toward 1syn, with and
without cesium cation present in solution, is practically indis-
tinguishable (Figure 2c/f).16 Intriguingly, DFT calculations
(BP86) of energy minimized [1syn:Cu(I)3]3+[CsF] suggest the
binding mode in which cesium facilitates the bowl constriction
via its cation-π interaction with the aromatic arms at the top
rim (Figure 7A), thereby shortening the N‚‚‚N distance between
adjacent arms of the bowl.16 The energy minimization of [1syn
:
(22) For an additional role of Ag(I) in improving Stille couplings see:
Gronowitz, S.; Messmer, A.; Timari, G. J. Heterocycl. Chem. 1992, 29,
1049-.
Cu(I)3]Cl3, without cesium present, was proven to be problem-
atic, and did not undergo the expected convergence at the BP86
J. Org. Chem, Vol. 73, No. 2, 2008 361

Yan et al.
i.e., by excluding the solvent molecules as potential external
guests. The calculated bowl interior volume can be approximated
to 205 Å.32 When positioned inside of 1syn, a partially
incorporated quinoxaline arm (108.7 A3) occupies 53% of the
available space, which is interestingly proportionate to packing
coefficients (PC) of liquids.32 This analysis needs to be taken
with some caution, as it is difficult to obtain the accurate inner
volume of 1syn: the molecule encompasses three sizable side
portals and a large opening at the rim. Three water molecules
are organized such that two share both of their hydrogen atoms
with quinoxaline nitrogens of the entangled bowls (Figure 7B).
The remaining one serves as a hydrogen-bonding linker in
between them. On the basis of the experimental N-H‚‚‚N (2.9
Å) and O-H‚‚‚O (2.8 Å) distances, all of the hydrogen bonds
can categorized to be of a moderate strength (2.8-3.2 Å).33 On
the basis of the solid-state data, it can be of interest to examine
the possible dimerization of 1syn in organic or aqueous solvents.
Our preliminary data point to a monomeric form in CDCl3 and
likely dimeric in D2O (10% DCl). The experimentally deter-
mined structure of 1syn, encompassing a dimer, is almost
equivalent to its DFT energy-minimized form (Figure 6). In the
X-ray resolved structure, one of the aromatic arms in 1syn is
moved toward the bowl interior (∆R ) 2.0°), while the other
two are turned more distant (∆R ) -0.27°). The molecule is
thus well preorganized to undergo the dimerization, and without
considerable conformational adjustments fill its inner space in
the solid state.
Conclusions
Transition metal-catalyzed transformations can be a subject
of templation to boost the reaction stereoselectivity and ma-
neuver the desired formation of covalent bonds. Here, we
demonstrated a utility of template-directed synthesis for the
preparation of a new family of bowl-shaped molecules. They
comprise a semirigid frame and adjustable size cavity capable
of coordinating transition metal cations. Studies of encapsulation
and dynamic modulation of chemical reactivity with these
reported hosts are in progress.
Experimental Section
Typical Procedure for the Synthesis of Compound 1. To a
solution of 7 (0.050 g, 0.115 mmol) in anhydrous DMF (0.5 mL),
at room temperature and under an atmosphere of argon, were added
cesium fluoride (0.035 g, 0.230 mmol), copper(I) iodide (0.004 g,
0.023 mmol), and Pd(PPh3)4 (0.013 g, 0.0115 mmol). The reaction
mixture was heated at 45 °C for 17 h. Diethyl ether (20 mL) and
water (10 mL) were added successively, and the resulting solution
was filtered under vacuum. The solid residue was washed with
diethyl ether (50 mL) and ethyl acetate (50 mL), and the combined
organic phase was dried (Na2SO4) and concentrated under reduced
pressure. The resulting solid was purified by column chromatog-
raphy (SiO2, toluene/acetone, 3:1) to yield 1syn as gray (0.0043 g,
20%) and 1anti as orange solids (0.0033 g, 15%).
(23) The data collection crystal was a pale yellow rectangular rod with
a pointed end. Examination of the diffraction pattern on a diffractometer
indicated an orthorhombic crystal system. All work was done at 150 K,
using a cryostream cooler. The data collection strategy was set up to measure
an octant of reciprocal space with a redundancy factor of 3.6, which means
that 90% of the reflections were measured at least 3.6 times. æ and ω scans
with a frame width of 1.0° were used. Data integration, scaling, and merging
were done computationally.24 Merging the data and averaging the symmetry
equivalent reflections resulted in an Rint value of 0.054. The structure was
solved by the direct methods procedure.25 Full-matrix least-squares refine-
ments based on F2 were performed computationally.26, 27 The asymmetric
unit contains two of the bowl-shaped molecules, with each containing a
crystallographic mirror plane, three water molecules, and some regions of
disordered electron density. The solvents used for recrystallization were
ethyl acetate and hexane, but neither of these molecules were recognizable
in the disordered regions. The SQUEEZE procedure28 of PLATON29 was
used to account for this disordered solvent. The total solvent accessible
void is 1549 Å3, and the electron count for this region is 200 electrons/unit
cell. The hydrogen atoms on the bowl-shaped molecules were included in
the model at calculated positions by using a riding model with U(H) )
1.2Ueq(attached atom). The hydrogen atoms of the water molecules were
located on difference electron density maps and refined isotropically. All
of these hydrogen atoms of the three water molecules are involved in an
intermolecular hydrogen bonded network with the bowl-shaped molecules.
The final refinement cycle was based on 8642 intensities and 460 variables
and resulted in agreement factors of R1(F) ) 0.099 and wR2(F2) ) 0.175.
For the subset of data with I > 2σ(I), the R1(F) value is 0.058 for 5143
reflections. The final difference electron density map contains maximum
and minimum peak heights of 0.29 and -0.27 e/Å3. Neutral atom scattering
factors were used and include terms for anomalous dispersion.30
(24) DENZO: Otwinowski, Z.; Minor, W. Methods Enzymology; Vol.
276, Macromolecular Crystallography, Part A; Carter, C. W., Jr., Sweet,
R. M., Eds.; Academic Press: New York, 1997; pp 307-326.
(25) Sheldrick, G. M. SHELXS-97; Universitat Gottingen: Gottingen,
Germany, 1997.
(26) Sheldrick, G. M. SHELXL-97; Universitat Gottingen: Gottingen,
Germany, 1997.
(27) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837-838.
(28) Sluis, P. V. D.; Spek, A. L. Acta Crystallogr. 1990, A46, 194.
(29) Spek, A. L. Acta Crystallogr. 1990, A46, C34.
(30) International Tables for Crystallography; Kluwer Academic Pub-
lishers: Dordrecht, The Netherlands, 1992; Vol. C.
(31) (a) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112,
5525. (b) Cockroft, S. L.; Hunter, C. A.; Lawson, K. R.; Perkins, J.; Urch,
C. J. J. Am. Chem. Soc. 2005, 127, 8594.
(32) The interior volume of 1syn was calculated following Rebek’s
protocol. See: Mecossi, S.; Rebek, J., Jr. Chem. Eur. J. 1998, 37, 3303.
(33) Jeffrey, A. G. An Introduction to Hydrogen Bonding; Oxford
University Press, Inc.: New York, 1997; pp 11 and 12.
Data for Compound 1syn. Mp >430 °C; 1H NMR (CDCl3, 400
MHz) δ 7.78 (m, 6H), 7.42 (m, 6H), 4.86 (s, 6H), 2.96 (d, 3H, J
) 1.5 Hz), and 2.85 (d, 3H, J ) 1.5 Hz); 13C NMR (CDCl3, 100
MHz) δ 163.2, 139.9, 138.8, 128.8, 128.6, 60.8, 49.3; HRMS(ESI)
m/z calcd for C39H24N6H 577.2140 [M + H]+, found 577.2144.
Data for Compound 1anti. Mp >430 °C; 1H NMR (CDCl3, 400
MHz) δ 8.02 (m, 2H), 7.94 (m, 2H), 7.87 (m, 2H), 7.70 (m, 2H),
7.56 (m, 4H), 4.68 (m, 6H), 2.80 (m, 5H), and 2.61 (d, 1H, J )
9.4 Hz); 13C NMR (CDCl3, 100 MHz) δ 164.1, 163.8, 163.4, 139.9,
139.7, 139.3, 139.2, 139.1, 139.0, 129.2, 128.9, 128.9, 128.9, 128.6,
60.2, 60.1, 60.0, 49.3, 49.2; HRMS(ESI) m/z calcd for C39H24N6H
577.2140 [M + H]+, found 577.2143.
Compound 1anti was proven difficult to purify with column
chromatography. Therefore, to verify its quantity produced in a
reaction, we used pentaerythrityl tetrabromide as internal standard
1
to reveal its concentration by H NMR spectroscopy.
Compound 2.11 To a solution of DMSO (4.0 mL) and CH2Cl2
(33.0 mL), at -78 °C and under an atmosphere of argon, was added
trifluoroacetic anhydride (10.73 g, 51.10 mmol) dropwise over a
period of 1 h. Subsequently, bicyclo[2.2.1]hept-5-ene-2,3-diol11
(2.26 g, 17.93 mmol) dissolved in CH2Cl2 (12 mL) was added over
10 min. The reaction mixture was allowed to stir at -78 °C for 2
h, after which it was quenched with triethylamine (9.54 g, 94.31
mmol). The mixture was then left to stir for 3 h at ambient
temperature, diluted with a solution of HCl (20 mL, 2.0 M), and
extracted with CH2Cl2 (4 × 50 mL). The combined organic phase
was dried (Na2SO4) and concentrated under reduced pressure to
yield an orange oil. Pure 2 was obtained in a vacuum distillation
(1.50 g, 60%). Bp 80 °C at 1.0 mm‚Hg (lit.11 bp 145 °C at 10.0
1
mm‚Hg); H NMR (CDCl3, 400 MHz) δ 6.51 (m, 2H), 3.31 (m,
2H), 2.96 (d, 1H, J ) 10.8 Hz), 2.49 (d, 1H, J ) 10.8 Hz); 13C
NMR (CDCl3, 100 MHz) δ 195.6, 137.7, 51.4, 43.6.
Compound 4. To a solution of 2 (0.69 g, 5.65 mmol) in CH2-
Cl2 (25.0 mL) was added 1,2-phenylenediamine 3 (0.61 g, 5.65
mmol) over a period of 30 min. The reaction was allowed to stir
for 1.5 h and then concentrated under reduced pressure. The
resulting solid was purified by column chromatography (SiO2,
362 J. Org. Chem., Vol. 73, No. 2, 2008

DiastereoselectiVe Synthesis of a Molecular Bowl
toluene/acetone, 9:1) to afford 14 as a light-yellow solid (0.726 g,
66%). Mp 129 °C; H NMR (CDCl3, 400 MHz) δ 7.90 (m, 2H),
7.62 (m, 2H), 6.86 (s, 2 H), 3.95 (s, 2H), 2.75 (d, 1H, J ) 8.7 Hz),
2.57 (d, 1H, J ) 8.7 Hz); 13C NMR (CDCl3, 100 MHz) δ 166.4,
142.31, 139.2, 128.6, 128.6, 63.0, 49.6; HRMS(ESI) m/z calcd for
C13H10N2H 195.0922 [M + H]+, found 195.0921.
Compound 8. To a solution of 7 (0.05 g, 0.115 mmol) in dry
DMF (1.0 mL), heated to 70 °C, were added palladium(II) acetate
(0.003 g, 0.011 mmol), triphenylphosphine (0.006 g, 0.023 mmol),
and a trace amount of lithium chloride. The reaction was allowed
to stir for a period of 1 h upon which additional portions (0.003 g,
0.011 mmol) of palladium(II) acetate were added, each at 3, 4, 5,
and 20 h. The solvent was concentrated under reduced pressure
and crude product was purified by column chromatography (SiO2,
toluene/acetone, 40:1) to afford 8 as a light-yellow solid (0.005 g,
19%). 1H NMR (CDCl3, 400 MHz) δ 7.90 (m, 2H), 7.60 (m, 2H),
3.61 (s, 2H), 2.66 (d, 1H, J ) 8.6 Hz), 2.46 (d, 1H, J1 ) 9.4 Hz),
1.78 (s, 6H); 13C NMR (CDCl3, 100 MHz) δ 166.7, 143.4, 139.2,
128.6, 128.3, 58.7, 54.4, 13.6; HRMS(ESI) m/z calcd for C15H14N2H
223.1235 [M + H]+, found 223.1231.
1
Compounds 5a/b. To a cis-decaline (30.0 mL) solution of 4
(0.726 g, 3.74 mmol) at 150 °C was added a solution of bromine
(0.776 g, 4.86 mmol) in cis-decaline slowly over 10 min. The
mixture was stirred at the same temperature for 20 min, upon which
the solvent was removed in vacuo. The residue was washed with
acetone, filtered, and purified by column chromatography (SiO2,
hexanes/CH2Cl2, 1:4) to afford 5a and 5b as tan solids (1.18 g, 90%).
1
Data for Compound 5a. Mp 148 °C; H NMR (CDCl3, 400
MHz) δ 8.03 (m, 2H), 7.74 (m, 2H), 4.39 (s, 2H), 3.91 (s, 2H),
2.97 (d, 1H, J ) 10.8 Hz), 2.39 (d, 1H, J ) 10.8 Hz); 13C NMR
(CDCl3, 100 MHz) δ 159.8, 142.0, 129.3, 55.7, 51.9, 41.9; HRMS-
(ESI) m/z calcd for C13H10Br2N2H 354.9264 [M + H]+, found
354.9263.
Computational Details.34All calculations for energy optimization
of 1syn and its complexes [1syn:Ag], [1syn:Ag2], [1syn:Ag3], and [1syn
:
Cs:Cu3] were completed with the Resolution of Identity Density
Functional Theory (RI-DFT) method and by using formate ions
attached to Ag(I) for charge neutralization.35-36 The Generalized
Gradient Approximation (GGA) BP86 functional21b was adopted.
TheRI-DFTmethodwasusedasimplementedinTURBOMOLE.37-43
The triple-ú plus polarization (TZVP) basis sets were used for the
lighter elements, while for silver and cesium, relativistic effective
core potential (ECP) and def-SV(P) basis sets were applied.41
Compounds 1syn, [1syn:Ag3], and [1syn:Cs:Cu3] were optimized with
assumed C3V symmetry. To compute the potential energy profile
as a function of the N-N distance in 1syn (Figure S30, Supporting
Information), the relaxed potential energy surface was scanned along
the N-N trajectory (Table S12, Supporting Information).
1
Data for Compound 5b. Mp 139 °C; H NMR (CDCl3, 400
MHz) δ 8.11 (m, 1H), 8.04 (m, 1H), 7.74 (m, 2H), 4.84 (s, 1H),
4.12 (s, 1H), 3.84 (br, 1H), 3.81 (m, 1H), 2.76 (d, 1H, 10.9 Hz),
2.45 (d, 1H, J ) 10.6 Hz); 13C NMR (CDCl3, 100 MHz) δ 159.1,
158.9, 142.0, 141.7, 129.7, 129.6, 129.5, 129.2, 54.5, 54.3, 54.1,
52.5, 45.6; HRMS(ESI) m/z calcd for C13H10Br2N2H 354.9264 [M
+ H]+, found 354.9262.
Compound 6. To a solution of 5a/b (1.15 g, 3.26 mmol) in
anhydrous THF (40 mL), cooled to 0 °C, was added potassium
tert-butoxide (2.20 g, 19.5 mmol) portionwise. The reaction was
allowed to stir for 2 h, quenched with water (5 mL), and
subsequently extracted with diethyl ether (4 × 50 mL). The
combined organic layer was dried (MgSO4) and concentrated under
reduced pressure. The resulting crude product was purified by
column chromatography (SiO2, toluene/acetone, 20:1) to afford 6
Acknowledgment. We thank Professor Jon Parquette of the
Ohio State University for useful suggestions. This work was
financially supported with funds obtained from the Ohio State
University, and the National Science Foundation under CHE-
0716355.
1
as a tan solid (0.825 g, 93%). Mp 111 °C; H NMR (CDCl3, 400
MHz) δ 7.94 (m, 1H), 7.92 (m, 1H), 7.65 (m, 2H), 6.87 (d, 1H, J
) 3.3 Hz), 4.02 (m, 1H), 3.94 (m, 1H), 2.99 (m, 1H), 2.62 (m,
1H); 13C NMR (CDCl3, 100 MHz) δ ) 164.4, 164.2, 139.7, 139.3,
139.2, 134.9, 129.1, 129.0, 128.9, 128.7, 62.4, 57.4, 51.0; HRMS-
(ESI) m/z calcd for C13H9BrN2H 273.0022 [M + H]+, found
273.0023.
Supporting Information Available: Additional experimental
and computational results, and spectra for all new compounds. This
materialisavailablefreeofchargeviatheInternetathttp://pubs.acs.org.
Compound 7. To a solution of dry diisopropylamine (1.3 mL,
9.15 mmol) in dry THF (9.0 mL) at 0 °C was added n-butyllithium
(5.7 mL, 1.6 M in hexane) under an atmosphere of argon. The
reaction mixture was cooled to -78 °C, and a solution of 6 (0.500
g, 1.83 mmol) in THF (4.0 mL) was added dropwise over 10 min.
The resulting mixture was stirred for an additional 30 min before
a solution of trimethyltin chloride (0.44 g, 2.2 mmol) in THF (3.0
mL) was added. After 2 h at -78 °C, the reaction mixture was
gradually warmed to room temperature, washed with water (15 mL),
and extracted with diethyl ether (3 × 50 mL). The combined organic
phase was dried (MgSO4) and concentrated in vacuo. The solid
residue was purified by column chromatography (SiO2, toluene/
acetone, 20:1) to afford 7 as an off-white solid (0.591 g, 75%).
JO701538G
(34) For recent studies employing the BP86 functional in investigating
organometallic systems, see: (a) Wang, F.; Li, L. M. J. Comput. Chem.
2004, 25 (5), 669. (b) Silva, V. D.; Dias, R. P.; Rocha, W. R. Chem. Phys.
Lett. 2007, 439, 69. (c) Nakamura, A.; Ohshima, T.; Mashima, K. J.
Organomet. Chem. 2005, 690, 4373. For a recent comparison of DFT
methods for organometallic systems, see: Furche, F.; Perdew, J. P. J. Chem.
Phys. 2006, 124, 044013.
(35) Eichkorn, K.; Treutler, O.; O¨ hm, H.; Ha¨ser, M.; Ahlrichs, R. Chem.
Phys. Lett. 1995, 242, 652.
(36) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor. Chem.
Acc. 1997, 97, 119.
(37) Becke, A. D. Phys. ReV. A 1988, 38, 3098.
1
Mp 104-105 °C; H NMR (CDCl3, 400 MHz) δ 7.97 (m, 1H),
(38) Scha¨fer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
(39) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Kolmel, C. Chem. Phys.
Lett. 1989, 162, 165.
7.93 (m, 1H), 7.65 (m, 2H), 4.07 (dd, 1H, J1 ) 3.4 Hz, J2 ) 1.6
Hz), 3.94 (dd, 1H, J ) 3.2 Hz, J2 ) 1.7 Hz), 2.92 (dt, 1H, J1 ) 8.8
Hz, J2 ) 1.7 Hz), 2.58 (dt, 1H, J1 ) 8.8 Hz, J2 ) 1.6 Hz), 0.27 (t,
9H, J ) 27.6 Hz); 13C NMR (CDCl3, 100 MHz) δ 164.9, 164.6,
153.9, 145.6, 139.6, 139.4, 129.0, 128.9, 128.8, 128.8, 61.9, 59.5,
56.3, -8.8; HRMS(ESI) m/z calcd for C16H17BrN2SnH 436.9662
[M + H]+, found 436.9664.
(40) Ha¨ser, M.; Ahlrichs, R. J. Comput. Chem. 1989, 10, 104.
(41) Von Arnim, M.; Ahlrichs, R. J. Comput. Chem. 1998, 19, 1746.
(42) Deglmann, P.; Furche, F.; Ahlrichs, R. Chem. Phys. Lett. 2002, 362,
511.
(43) The Turbomole basis set library is available via ftp from ftp://
ftp.chemie.uni-karlsruhe.de/pub/basen.
J. Org. Chem, Vol. 73, No. 2, 2008 363