Cucurbit[n]uril formation proceeds by step-growth cyclo-oligomerization

Article abstract of DOI:10.1021/ja8013693

In contrast to the high yield formation of cucurbit[n]uril (CB[n]) from a 1:2 ratio of glycoluril to formaldehyde, the condensation of glycoluril with less than 2 equiv of formaldehyde delivers a reaction mixture that contains glycoluril oligomers (2-6) and CB[n] compounds that lack one or more methylene bridges known as nor-seco-cucurbit[n]urils (ns-CB[n]). In this paper we report the chromatographic purification of C-shaped glycoluril oligomers (dimer-hexamer), their characterization in solution, and their X-ray crystal structures. Quite interestingly, despite being acyclic glycoluril pentamer 5 and hexamer 6 retain the ability to bind to guests typical of CB[6] but are also able to expand their cavity to accommodate larger guests like cationic adamantane derivatives. We performed product resubmission experiments with glycoluril oligomers 2-6 and found preferences for the formation of specific ring sizes during CB[n] formation. A comprehensive mechanistic scheme is proposed that accounts for the observed formation of 2-6 and ns-CB[n]. Overall, the experiments establish that a step-growth cyclo-oligomerization process operates during CB[n] formation.

Full text of DOI:10.1021/ja8013693

Published on Web 06/05/2008
Cucurbit[n]uril Formation Proceeds by Step-Growth
Cyclo-oligomerization
Wei-Hao Huang, Peter Y. Zavalij, and Lyle Isaacs*
Department of Chemistry and Biochemistry, UniVersity of Maryland,
College Park, Maryland 20742
Received February 23, 2008; E-mail: lisaacs@umd.edu
Abstract: In contrast to the high yield formation of cucurbit[n]uril (CB[n]) from a 1:2 ratio of glycoluril to
formaldehyde, the condensation of glycoluril with less than 2 equiv of formaldehyde delivers a reaction
mixture that contains glycoluril oligomers (2-6) and CB[n] compounds that lack one or more methylene
bridges known as nor-seco-cucurbit[n]urils (ns-CB[n]). In this paper we report the chromatographic
purification of C-shaped glycoluril oligomers (dimer-hexamer), their characterization in solution, and their
X-ray crystal structures. Quite interestingly, despite being acyclic glycoluril pentamer 5 and hexamer 6
retain the ability to bind to guests typical of CB[6] but are also able to expand their cavity to accommodate
larger guests like cationic adamantane derivatives. We performed product resubmission experiments with
glycoluril oligomers 2-6 and found preferences for the formation of specific ring sizes during CB[n] formation.
A comprehensive mechanistic scheme is proposed that accounts for the observed formation of 2-6 and
ns-CB[n]. Overall, the experiments establish that a step-growth cyclo-oligomerization process operates
during CB[n] formation.
Scheme 1. Synthesis of CB[6]
Introduction
In 1981 Mock reported that the condensation of glycoluril
and formaldehyde delivers the cyclic hexameric macrocycle
cucurbit[6]uril (CB[6]).1 This remarkable reaction (Scheme 1)
results in the formation of 24 new C-N bonds and six eight-
membered rings all with complete control over the relative
orientation of the glycoluril C-H atoms which point out of the
cavity. During the 1980s Mock extensively studied the host-guest
recognition behavior of CB[6] and found that CB[6] exhibits
remarkably tight (Ka up to 108 M-1) and selective binding
toward organic ammonium ions in water.2,3 Subsequent work
by Mock showed that these binding characteristics could be used
to catalyze a dipolar cycloaddition (click-chemistry) and to
create one of the first examples of a molecular shuttle.4,5 As
supramolecular chemists, we were inspired by the outstanding
recognition properties of CB[6] and hypothesized that if we
could tailor the recognition properties of CB[6] by synthetic
chemistry (e.g., the creation of cucurbit[n]uril homologues
(CB[n]),6–8 analogues, derivatives, or congeners) that we might
be in the position to create contemporary molecular devices like
chemical sensors, supramolecular catalysts, ion and molecular
channels, and molecular machines (e.g., molecular shuttles). As
physical organic chemists, we decided to target a thorough
understanding of the mechanism of CB[n] formation with the
expectation that such knowledge would enable the tailor-made
synthesis of CB[n]-type receptors with new geometrical features
and recognition properties.
Since we began our work on the mechanism of CB[n]
formation upon arriving at the University of Maryland in 1998
there have been a number of developments in the CB[n] area
that post facto validate our plan to enhance the range of
applications to which the CB[n] family6,7 can be applied by
the creation of new CB[n]-type receptors. For example, the
isolation of the cucurbit[n]uril homologues (n ) 5, 6, 7, 8) by
the groups of Kim and Day provided a series of macrocycles
(1) Freeman, W. A.; Mock, W. L.; Shih, N.-Y. J. Am. Chem. Soc. 1981,
103, 7367–7368.
(2) (a) Mock, W. L.; Shih, N.-Y. J. Org. Chem. 1986, 51, 4440–4446.
(b) Jeon, W. S.; Moon, K.; Park, S. H.; Chun, H.; Ko, Y. H.; Lee,
J. Y.; Lee, E. S.; Samal, S.; Selvapalam, N.; Rekharsky, M. V.;
Sindelar, V.; Sobransingh, D.; Inoue, Y.; Kaifer, A. E.; Kim, K. J. Am.
Chem. Soc. 2005, 127, 12984–12989. (c) Rekharsky, M. V.; Mori,
T.; Yang, C.; Ko, Y. H.; Selvapalam, N.; Kim, H.; Sobransingh, D.;
Kaifer, A. E.; Liu, S.; Isaacs, L.; Chen, W.; Moghaddam, S.; Gilson,
M. K.; Kim, K.; Inoue, Y. Proc. Natl. Acad. Sci. U.S.A. 2007, 104,
20737–20742.
(3) Liu, S.; Ruspic, C.; Mukhopadhyay, P.; Chakrabarti, S.; Zavalij, P. Y.;
Isaacs, L. J. Am. Chem. Soc. 2005, 127, 15959–15967.
(4) Mock, W. L.; Pierpont, J. J. Chem. Soc., Chem. Commun. 1990, 1509–
1511.
(6) Lagona, J.; Mukhopadhyay, P.; Chakrabarti, S.; Isaacs, L. Angew.
Chem., Int. Ed. 2005, 44, 4844–4870.
(7) Lee, J. W.; Samal, S.; Selvapalam, N.; Kim, H.-J.; Kim, K. Acc. Chem.
Res. 2003, 36, 621–630.
(5) Mock, W. L.; Irra, T. A.; Wepsiec, J. P.; Adhya, M. J. Org. Chem.
1989, 54, 5302–5308.
(8) Kim, K.; Selvapalam, N.; Ko, Y. H.; Park, K. M.; Kim, D.; Kim, J.
Chem. Soc. ReV. 2007, 36, 267–279.
9
8446 J. AM. CHEM. SOC. 2008, 130, 8446–8454
10.1021/ja8013693 CCC: $40.75
2008 American Chemical Society

Mechanism of CB[n] Formation
A R T I C L E S
Chart 1. Chemical Structures of CB[n], iCB[n], bis-ns-CB[10], and
(()-bis-ns-CB[6]
refolding of non-natural oligomers in water and as a host for
metalloporphyrins.19 Most recently, in collaboration with Ki-
moon Kim’s group, we have reported the isolation and recogni-
tion properties of diastereomeric CB[n] known as inverted-CB[n]
(i-CB[n]) in which a single pair of methine C-H groups point
into the central cavity.20,21 The formation of these new CB[n]
under milder kinetically controlled conditions provide an
existence proof for intermediates in the mechanism of CB[n]
formation and have helped guide our mechanistic studies over
the years.20–25
In this paper we explore the consequences of our realization
that the cyclo-oligomerization reaction between glycoluril and
formaldehydestetrafunctional and difunctional monomers,
respectivelysis in many ways related to a classical polymeri-
zation reaction. For example, classical polymerization reactions
between two different multifunctional monomers only proceed
to give high molecular weight material when there is a
stoichiometric balance between the reactive groups of the two
monomers.26 When there is an excess of one of the monomers,
it acts as an end-capping group and shorter oligomers are
obtained. Realizing that cyclic oligomeric CB[n] can be viewed
as infinitely long oligomers raised the question in our mind of
what would happen if we starved the CB[n] forming reaction
of formaldehyde. Would new mechanistic intermediates in the
formation of CB[n] be delivered as kinetically stable isolable
materials? Our previous reports on the isolation, structural
characterization, and recognition properties of bis-nor-seco-
CB[10]27 and (()-bis-nor-seco-CB[6] provide an answer in the
affirmative and we describe our complete mechanistic study in
detail here.
whose sizes (82, 164, 279, 479 Å3) parallel those of R-, ꢀ-, and
γ-cyclodextrins (Chart 1).9,10 These new macrocycles, therefore,
participate in a variety of interesting applications including
supramolecular dye lasers,11 novel drug delivery vehicles,12 as
a mediator of organic reactions,5,13 peptide recognition,14
chemical sensors,15,16 as components of complex self-sorting
systems,3,17 and in the development of molecular machines.18
More recently, our group has reported the isolation of free
CB[10] from its CB[10]·CB[5] complex and the ability of its
870 Å3 cavity to promote folding, forced unfolding, and
Results and Discussion
We begin this results and discussion section with a brief
review of the state-of-the-art concerning the mechanism of
(18) (a) Jeon, W. S.; Ziganshina, A. Y.; Lee, J. W.; Ko, Y. H.; Kang, J.-
K.; Lee, C.; Kim, K. Angew. Chem., Int. Ed. 2003, 42, 4097–4100.
(b) Ko, Y. H.; Kim, K.; Kang, J.-K.; Chun, H.; Lee, J. W.; Sakamoto,
S.; Yamaguchi, K.; Fettinger, J. C.; Kim, K. J. Am. Chem. Soc. 2004,
126, 1932–1933. (c) Jeon, W. S.; Kim, E.; Ko, Y. H.; Hwang, I.; Lee,
J. W.; Kim, S.-Y.; Kim, H.-J.; Kim, K. Angew. Chem., Int. Ed. 2005,
44, 87–91. (d) Ooya, T.; Inoue, D.; Choi, H. S.; Kobayashi, Y.;
Loethen, S.; Thompson, D. H.; Ko, Y. H.; Kim, K.; Yui, N. Org.
Lett. 2006, 8, 3159–3162. (e) Sobransingh, D.; Kaifer, A. E. Org. Lett.
2006, 8, 3247–3250. (f) Sindelar, V.; Silvi, S.; Kaifer, A. E. Chem.
Commun. 2006, 2185–2187. (g) Liu, Y.; Li, X.-Y.; Zhang, H.-Y.; Li,
C.-J.; Ding, F. J. Org. Chem. 2007, 72, 3640–3645. (h) Tuncel, D.;
(9) Kim, J.; Jung, I.-S.; Kim, S.-Y.; Lee, E.; Kang, J.-K.; Sakamoto, S.;
Yamaguchi, K.; Kim, K. J. Am. Chem. Soc. 2000, 122, 540–541.
(10) Day, A.; Arnold, A. P.; Blanch, R. J.; Snushall, B. J. Org. Chem.
2001, 66, 8094–8100.
(11) Mohanty, J.; Pal, H.; Ray, A. K.; Kumar, S.; Nau, W. M. ChemPhys-
Chem 2007, 8, 54–56.
(12) (a) Wheate, N. J.; Day, A. I.; Blanch, R. J.; Arnold, A. P.; Cullinane,
C.; Collins, J. G. Chem. Commun. 2004, 1424–1425. (b) Bali, M. S.;
Buck, D. P.; Coe, A. J.; Day, A. I.; Collins, J. G. Dalton Trans. 2006,
5337–5344. (c) Wheate; N. J.; Buck, D. P.; Day, A. I.; Collins, J. G.
Dalton Trans. 2006, 451–458. (d) Jeon, Y. J.; Kim, S.-Y.; Ko, Y. H.;
Sakamoto, S.; Yamaguchi, K.; Kim, K. Org. Biomol. Chem. 2005, 3,
2122–2125.
¨
¨
Ozsar, O.; Tiftik, H. B.; Salih, B. Chem. Commun. 2007, 1369–1371.
(i) Ko, Y. H.; Kim, E.; Hwang, I.; Kim, K. Chem. Commun. 2007,
1305–1315.
(19) (a) Day, A. I.; Blanch, R. J.; Arnold, A. P.; Lorenzo, S.; Lewis, G. R.;
Dance, I. Angew Chem., Int. Ed. 2002, 41, 275–277. (b) Liu, S.;
Zavalij, P. Y.; Lam, Y.-F.; Isaacs, L. J. Am. Chem. Soc. 2007, 129,
11232–11241. (c) Liu, S.; Shukla, A. D.; Gadde, S.; Wagner, B. D.;
Kaifer, A. E.; Isaacs, L. Angew. Chem., Int. Ed. 2008, 47, 2657.
(20) Isaacs, L.; Park, S.-K.; Liu, S.; Ko, Y. H.; Selvapalam, N.; Kim, Y.;
Kim, H.; Zavalij, P. Y.; Kim, G.-H.; Lee, H.-S.; Kim, K. J. Am. Chem.
Soc. 2005, 127, 18000–18001.
(13) (a) Jon, S. Y.; Ko, Y. H.; Park, S. H.; Kim, H.-J.; Kim, K. Chem.
Commun. 2001, 1938–1939. (b) Choi, S.; Park, S. H.; Ziganshina,
A. Y.; Ko, Y. H.; Lee, J. W.; Kim, K. Chem. Commun. 2003, 2176–
2177. (c) Pattabiraman, M.; Natarajan, A.; Kaliappan, R.; Mague, J. T.;
Ramamurthy, V. Chem. Commun. 2005, 4542–4544. (d) Wang, R.;
Yuan, L.; Macartney, D. H. J. Org. Chem. 2006, 71, 1237–1239.
(14) (a) Bush, M. E.; Bouley, N. D.; Urbach, A. R. J. Am. Chem. Soc.
2005, 127, 14511–14517. (b) Heitmann, L. M.; Taylor, A. B.; Hart,
P. J.; Urbach, A. R. J. Am. Chem. Soc. 2006, 128, 12574–12581. (c)
Rekharsky, M. V.; Yamamura, H.; Ko, Y. H.; Kim, K.; Inoue, Y.
Peptide Sci. 2006, 43, 393–394.
(21) Liu, S.; Kim, K.; Isaacs, L. J. Org. Chem. 2007, 72, 6840–6847.
(22) (a) Wu, A.; Chakraborty, A.; Witt, D.; Lagona, J.; Damkaci, F.; Ofori,
M. A.; Chiles, J. K.; Fettinger, J. C.; Isaacs, L. J. Org. Chem. 2002,
67, 5817–5830. (b) Witt, D.; Lagona, J.; Damkaci, F.; Fettinger, J. C.;
Isaacs, L. Org. Lett. 2000, 2, 755–758.
(23) Chakraborty, A.; Wu, A.; Witt, D.; Lagona, J.; Fettinger, J. C.; Isaacs,
L. J. Am. Chem. Soc. 2002, 124, 8297–8306.
(15) (a) Sindelar, V.; Cejas, M. A.; Raymo, F. M.; Chen, W.; Parker, S. E.;
Kaifer, A. E. Chem.sEur. J. 2005, 11, 7054–7059. (b) Ling, Y.; Wang,
W.; Kaifer, A. E. Chem. Commun. 2007, 610–612.
(24) Lagona, J.; Fettinger, J. C.; Isaacs, L. Org. Lett. 2003, 5, 3745–3747.
(25) Lagona, J.; Fettinger, J. C.; Isaacs, L. J. Org. Chem. 2005, 70, 10381–
10392.
(16) Lagona, J.; Wagner, B. D.; Isaacs, L. J. Org. Chem. 2006, 71, 1181–
1190.
(26) (a) So, Y.-H. Acc. Chem. Res. 2001, 34, 753–758. (b) Flory, P. J.
Chem. ReV. 1946, 39, 137–197.
(17) (a) Mukhopadhyay, P.; Wu, A.; Isaacs, L. J. Org. Chem. 2004, 69,
6157–6164. (b) Mukhopadhyay, P.; Zavalij, P. Y.; Isaacs, L. J. Am.
Chem. Soc. 2006, 128, 14093–14102.
(27) Huang, W.-H.; Liu, S.; Zavalij, P. Y.; Isaacs, L. J. Am. Chem. Soc.
2006, 128, 14744–14745.
9
J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008 8447

A R T I C L E S
Huang et al.
Scheme 2. Proposed Mechanism of CB[n] Formation
CB[n] formation that sets the stage for a discussion of the results
reported in this paper.
building block of the CB[n] family of macrocycles we initially
studied derivatives of 2 and 2S that contained solubilizing CO2Et
groups on their convex face and capping o-xylylene groups.22,23
These model studies of the earliest steps of the mechanism of
CB[n] formation (e.g., dimer formation and interconversion)
allowed us to establish the high thermodynamic preference for
the C-shaped forms (∆∆G > 2 kcal mol-1), elucidate the
intramolecular nature of the S-shaped to C-shaped interconver-
sion, and to propose and validate the great potential of building
block strategies in the formation of CB[n]-type receptors.22–25,31
In combination, the results described above shed significant light
on the earliest and latest steps of the mechanism of CB[n]
formation. What was lacking was information about what goes
on in the middle.
Previous Mechanistic Studies. Scheme 2 shows the funda-
mental steps of the mechanism of CB[n] formation that were
previously presented by the groups of Isaacs and Day.10,23 First,
glycoluril 1 reacts with 2 equiv of formaldehyde to potentially
deliver a mixture of diastereomers (2 and 2S) that differ in the
relative orientation of the pairs of methine C-H groups. It was
further hypothesized that these dimers undergo further oligo-
merization via 3-8 to ultimately yield a substance known as
Behrend’s polymer.28 The length of the polymer chain and
diastereomeric orientation of methine C-H groups in Behrend’s
polymer was ill-defined. At this stage, Behrends polymer must
undergo an isomerization reaction that converts its S-shaped to
C-shaped subunitsspotentially aided by templating groups29sto
create an oligomer (e.g., 9 or 10) that is poised to undergo
macrocyclization to enter the CB[n] family manifold by either
end-to-end cyclization or back-biting mechanisms.
Our group and the group of Anthony Day have been heavily
involved in elucidating the mechanistic details of the CB[n]
forming reaction.10,23,25,29 Day’s group has focused on the final
steps of the mechanism of CB[n] formation and have examined
the influence of numerous potential templating agents on the
distribution of CB[n] obtained (e.g., ammonium ions and alkali
metal ions).30 Further transformations can occur inside the
manifold of CB[n]-type receptors. For example, Day showed
that heating CB[8] under the reaction conditions, where CB[5],
CB[6], and CB[7] are stable, results in a partial ring contraction
to deliver the smaller CB[n] homologues.10 Similarly, we
recently reported product resubmission experiments that show
that i-CB[6] and i-CB[7] are converted into CB[n] which
establishes that these diastereomeric CB[n] are kinetically
controlled intermediates in the mechanism of CB[n] formation.21
Based on our realization that the methylene-bridged glycoluril
dimer substructure (bold in Chart 1) constituted the fundamental
Reaction Mixtures Deficient in Formaldehyde Deliver
Glycoluril Oligomers 2-6 as Isolable Species. On the basis of
the connections described above between a classical copoly-
merization between multifunctional monomers and CB[n]
formation we decided to conduct the condensation of glycoluril
(1) with less than 2 equiv of formaldehyde under aqueous acidic
conditions (Scheme 3). From these reaction mixtures, we were
able to isolate the series of glycoluril oligomers 2-6 by a
combination of Dowex ion-exchange chromatography and
recrystallization.32 We were able to establish the constitution
of glycoluril oligomers 2-6 by mass spectrometry. Given the
large number of potential diastereomers of these oligomers 2
1
(2), 3 (3), 4 (6), 5 (10), 6 (21), structural elucidation by H
NMR spectroscopy alone is challenging (Figure 1). Compounds
2-6 with all methine C-H groups on a single side of the
molecule are C-shaped and C2V-symmetric. Symmetry consid-
erations dictate that dimer 2 show two doublets for the methine
C-H groups (Ha and Hb) and two doublets for the diastereotopic
CH2-groups (Hc and Hd) as is observed experimentally. The 1H
NMR spectrum for trimer 3 consists of two doublets and one
singlet for the glycoluril methine C-H groups (Ha, Hb, Hc) in
(28) Previous reports have referred to the product of the initial stage of
Behrend’s synthesis of CB[n] (aq. HCl, 100 °C, 3 equiv of formal-
dehyde) as “Behrend’s polymer”. The results reported in this paper
suggest that this substance may, in fact, consist of a mixture of
glycoluril oligomers, nor-seco-CB[n], and other kinetically controlled
intermediates in the mechanism of CB[n] formation.
(31) (a) Day, A. I.; Arnold, A. P.; Blanch, R. J. Molecules. 2003, 8, 74–
84. (b) Zhao, Y.; Xue, S.; Zhu, Q.; Tao, Z.; Zhang, J.; Wei, Z.; L., L.;
Hu, M.; Xiao, H.; Day, A. I. Chin. Sci. Bull. 2004, 49, 1111–1116.
(32) In addition to oligomers 2-6, bis-ns-CB[10], (()-bis-ns-CB[6], and
ns-CB[6] the crude reaction mixtures typically contain some CB[6]
and CB[7]. We have been unsuccessful in developing an HPLC method
to more precisely quantify the components of the reaction mixture
for several reasons including (1) incompatibility of silica gel with the
aq acid required for solubility, (2) irreversible adsorption during Dowex
ion-exchange chromatography, and (3) lack of a convenient chro-
mophore for detection.
(29) Day, A. I.; Blanch, R. J.; Coe, A.; Arnold, A. P. J. Inclusion Phenom.
Macrocyclic Chem. 2002, 43, 247–250.
(30) The influence of glycoluril concentration, acid identity and concentra-
tion and temperature have also been studied by Day and co-workers.
The point in the mechanistic pathway where these variable exert their
influence remains unclear.
9
8448 J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008

Mechanism of CB[n] Formation
A R T I C L E S
Scheme 3. Synthesis of C-Shaped 2-6
addition to two doublets for the diastereotopic CH2-groups. The
1H NMR spectra for 4-6 are complicated by spectral overlap
in the methine CH region, although the number and relative
intensity of the resonances for the diastereotopic CH2-groups
are in accord with symmetry considerations, and complete
assignment was not possible on the basis of mass spectrometry
of 2-6 relative to CB[5]-CB[8] we measured the angles
between the adjacent glycoluril rings (θ, Figure 2) through their
1
and H NMR spectroscopy alone.
X-ray Crystal Structures of 2-6. Although we were quite
confident in our assignment of the structures of the oligomers
2-6 based on spectroscopic techniques and previous model
studies which indicated a significant thermodynamic preference
for the C-shaped diastereomers we wanted to obtain final
structural proof in the form of X-ray crystal structures. With
some effort we were able to obtain single crystals of 2, 3, 4, 5,
and 6 (Figure 2). The structures of 2-6 are striking for several
reasons: (1) the anticipated C-shaped geometries are observed
in all cases, (2) the curvature of 2-5 is mainly restricted to a
single plane (e.g., out of plane twisting is not significant), and
(3) the degree of curvature maps well onto that of CB[6] which
is in accord with theoretical calculations which show that CB[6]
is the least strained and thermodynamically most stable CB[n]
on a per glycoluril basis.7 To quantify the degree of curvature
Figure 2. Cross-eyed stereoviews of the crystal structures of (a) 2, (b) 3,
(c) 4, (d) 5, and (e) 6. Solvating CF3CO2H and H2O molecules have been
removed for clarity. Color code: C, gray; H, white; N, blue; O, red; H-bonds,
red-yellow striped.
Figure 1. 1H NMR spectra (400 MHz, 35% DCl, room temp) for (a) 2,
(b) 3, (c) 4, (d) 5, and (e) 6.
9
J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008 8449

A R T I C L E S
Huang et al.
Table 1. Values of θ Measured from the X-ray Structures of 2-6
and CB[5]-CB[8]1,9
Table 2. The Molar Ratio of Glycoluril and Formaldehyde Units
Needed to Construct CB[n], i-CB[n], ns-CB[n], and Oligomers 2-6
compound
θ
average θ
compound
glycoluril
formaldehyde
2
3
4
5
6
123.5
123.6, 123.6
124.2, 120.0, 124.2
124.8, 123.6, 123.6, 124.8
132.9, 143.6, 130.5, 116.3, 117.9
123.5
123.6
122.8
124.2
128.2
108.0
119.9
128.5
CB[n]
i-CB[n]
ns-CB[6]
bis-ns-CB[10]
1
1
1
1
1
1
1
1
1
1
2
2
1.84
1.80
1.67
1.67
1.60
1.50
1.33
1
(()-bis-ns-CB[6]
CB[5] 103.4, 109.7, 109.7, 103.3, 113.8
CB[6] 118.7, 116.6, 124.5, 118.7, 116.7, 124.4
CB[7] 124.4, 130.4, 129.5, 129.6, 123.3, 133.8, 128.4
6
5
4
3
2
CB[8] 135.4, 135.9, 139.5, 129.2, 135.4, 135.8, 139.5, 129.2 135.0
equatorial methine C-atoms (Table 1). An examination of Table
1 shows that oligomers 2-5 have an average curvature (θ) that
is intermediate between that of CB[6] and CB[7]. Hexamer 6,
in contrast, has a wider spread of values of θ and an average θ
that is close to that of CB[7] which indicates the inherent
flexibility of longer glycoluril oligomers.
Reaction Mixtures Deficient in Formaldehyde Also Deliver
nor-seco-CB[n] as Isolable Species. From similar reaction
mixtures we have previously isolated bis-ns-CB[10], (()-bis-
ns-CB[6], and ns-CB[6].27,35,36 Although the quantitation of the
amounts of each oligomer and nor-seco-CB[n] in a given
reaction mixture has proved challenging because of extensive
Beyond the connectivity of the molecular structures of 2-6
and their curvature there are intriguing aspects of the solvation
and packing of 2-6 in the crystal (Supporting Information).
For example, compound 2 crystallizes (Figure 2a) as the
CF3CO2H (TFA) solvate which segregates the molecules of 2
from one another by forming H-bonds to the ureidyl groups of
2. Trimer 3 also crystallizes (Figure 2b) as the TFA solvate by
H-bonding interactions with the ureidyl groups of 3. In this case,
however, multiple molecules of 3 segregate themselves into slabs
in the ab-plane that are separated by solvating slabs of TFA
molecules along the c-axis. The packing of tetramer 4 (Figure
2c) in the crystal is facilitated by the interaction of the center
two ureidyl CdO groups of 2 equiv of 4 forming a head-to-tail
dimeric entity by ion-dipole interactions with Na+. Interest-
ingly, these dimers of 4 are further organized by NH · · ·OdC
H-bonds between the tips of 4 and the CdO rim of another
molecule of 4 to form a complex network motif (Supporting
Information). The packing of pentamer 5 which begins to look
like CB[5] is even more interesting. Each molecule of pentamer
5 contains one solvating molecule of TFA within its caVity.
Furthermore, these molecules of 5 organize themselves into
infinite 1-dimensional tapes along the a-axis by four H-bonds
between the N-H tips of 5 with the ureidyl CdO of the adjacent
equivalents of 5 (Supporting Information). Most interesting and
relevant toward the mechanism of CB[n] formation is the
packing observed in the crystal structure of 6. Individual
molecules of 6 are distorted from a symmetric CB[6]-like shape
and exhibit an out-of-plane twist toward the tips of the molecule.
Both features speak to the inherent flexibility of the growing
glycoluril oligomer chains which is less apparent by examination
of the macrocyclic CB[n]. Furthermore, two molecules of 6
dimerize in the crystal driven by NH · · ·O H-bonds. Although
glycoluril derived supramolecular structures33 are well-known
to undergo self-association in organic and aqueous solution by
H-bonds or π-π interactions this is the first example that we
are aware of that implicates self-association of water soluble
CB[n] fragments.34 The implication is that (self)-association of
glycoluril oligomers may be considered as a relevant side
pathway during the CB[n] forming reaction.
1
peak overlap in the H NMR spectrum and losses that occur
during Dowex ion exchange we have developed a guiding
principle to maximize the yield of a given compound from
condensations of glycoluril and formaldehyde that contain a
deficiency of formaldehyde.32This guideline is based on the
theoretical ratio of glycoluril to formaldehyde in a given
oligomer or ns-CB[n] (Table 2) and can be considered a
consequence of Le Chatelier’s principle. For example, dimer 2
is composed of 2 equiv of glycoluril and 2 equiv of formalde-
hyde and is best targeted with reaction mixtures containing a
1:1 ratio of the two starting materials. Similarly, the recently
described ns-CB[n] (ns-CB[6], bis-ns-CB[10], and (()-bis-ns-
CB[6]) can be efficiently isolated from reaction mixtures
comprising a 1:1.5-1:1.67 molar ratio of glycoluril/formalde-
hyde.37 This ratio is slightly lower than the calculated ratio since
entry into the CB[n] manifold is an irreversible process that
reduces overall yield and complicates product isolation. Lastly,
the synthesis of CB[n] is known to be most efficient at a
stoichiometric ratio of 1:2.10
Implications of the Isolation of 2-6, Bis-ns-CB[10], (()-Bis-
ns-CB[6], and ns-CB[6] Toward the Mechanism of CB[n]
Formation. The isolation of glycoluril oligomers 2-6 and nor-
seco-CB[n] allows us to add levels of detail to the mechanism
of CB[n] formation that was not previously possible. Scheme
4 illustrates in detail our current level of understanding of the
CB[n] formation process. The isolation of oligomers 2-6 from
reaction mixtures comprising glycoluril and a deficiency of
formaldehyde provides an existence proof for these structures
as kinetically controlled intermediates in the formation of CB[n].
Accordingly, Scheme 4 shows the stepwise interconversion of
glycoluril 1 into hexamer 6 by stepwise addition of monomer
(34) The formation of CB[10] ·CB[5] and the gelation of purified CB[7]
provide evidence for the (self)-association of CB[n] in aqueous
solution. Although the presented x-ray structure is a solid state structure
we find it highly likely that glycoluril oligomers will also undergo
(self)-association in aqueous solution.
(35) Huang, W.-H.; Zavalij, P. Y.; Isaacs, L. Angew. Chem., Int. Ed. 2007,
46, 7425–7427.
(36) Huang, W.-H.; Zavalij, P. Y.; Isaacs, L. Org. Lett. 2008, 10, 2577–
2580.
(37) The isolation of 2-6, bis-ns-CB[10], (()-bis-ns-CB[6], and ns-CB[6]
from reaction mixtures containing a deficiency of formaldehyde was
possible because this stoichiometric imbalance slows down their
transformation to the thermodynamically more stable CB[n]. In
particular, the precipitation of certain compounds (e.g. bis-ns-CB[10])
from the reaction mixture can result in significant kinetic stability.
(33) (a) Rebek, J. J. Acc. Chem. Res. 1999, 32, 278–286. (b) Rowan, A. E.;
Elemans, J. A. A. W.; Nolte, R. J. M. Acc. Chem. Res. 1999, 32,
995–1006. (c) Wu, A.; Chakraborty, A.; Fettinger, J. C.; Flowers,
R. A., II.; Isaacs, L. Angew. Chem., Int. Ed. 2002, 41, 4028–4031. (d)
Isaacs, L.; Witt, D. Angew. Chem., Int. Ed. 2002, 41, 1905–1907.
9
8450 J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008

Mechanism of CB[n] Formation
A R T I C L E S
Scheme 4. Comprehensive Mechanistic Scheme for the Formation of CB[n]a
a Color coding: chain growth, aqua arrows; step growth; red arrows (addition of 2), blue arrows (addition of 3); green arrows (addition of 4).
1 (aqua arrows). Such a process where monomer is added
stepwise is known in polymer chemistry as chain growth
polymerization.26 Although we have not isolated glycoluril
heptamer or octamer (7 and 8) we depict them as highly likely
mechanistic intermediates. As the oligomer lengthens from 1
to 8 the formation of undesired S-shaped diastereomers is likely.
As depicted for 2S and 2sand studied in detail in model systems
by us22,23san equilibration occurs that dramatically favors the
C-shaped forms 3-8. Once these oligomer chains have grown
long enough (5-8) they may undergo direct intramolecular end-
to-end cyclization by way of chiral hydroxymethylated inter-
mediate (()-11(n) then ns-CB[n], and finally enter into the
CB[n] manifold by addition of the final CH2-group. This portion
of Scheme 4 is in essence an enhanced version of Scheme 2
which accounts in detail for the presence of oligomers 3-6 and
also takes into account the intermediacy of ns-CB[6].
oligomer chains is also possible. Consider, for example,
intermediate (()-11(n) (n e 4) which is too short to undergo
direct intramolecular end-to-end cyclization. Intermediate (()-
11(n) is, therefore, forced to undergo intermolecular reaction
with an oligomer of identical or different length. The sets of
colored equilibrium arrows in Scheme 4 indicate the step-growth
occurring by the addition of dimer 2 (blue arrows), trimer 3
(red arrows), and tetramer 4 (green arrows) and 2 equiv of
formaldehyde to give longer C-shaped oligomers. Analysis of
the intermediate stage of the addition of 1 equiv of formaldehyde
is instructive. Despite the fact that 1-8 are achiral, their ureidyl
NH groups are prochiral and pairs of NH groups are either
homotopic (pairs of red or pairs of green NH groups in structure
of 2) or enantiotopic (pairs of red and green NH groups in
structure of 2). Connection of a pair of homotopic groups
between two oligomers results in intermediate (()-12(m,n)
whereas connection of a pair of enantiotopic NH groups by
means of a CH2-bridge results in the formation of 13(m,n).38
Intermediates 13(m,n) and (()-12(m,n) may add a second CH2-
The isolation of bis-ns-CB[10] and (()-bis-ns-CB[6], com-
prising two oligomer chains (e.g., 5 and 3, respectively) linked
by two CH2-groups, alerted us to the potential operation of a
second type of mechanistic pathway known as a step-growth
process. In a step-growth process, the reaction between two
(38) Compounds 13(m,n) and 17(m,n) are chiral when m * n.
9
J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008 8451

A R T I C L E S
Huang et al.
Table 3. The Distribution of CB[n] Obtained from Reaction of 1-6
group in the middle of the molecule to form longer C-shaped
oligomers 4-8 or diasteromeric oligomers with one S-shaped
subunit, respectively. Alternatively, if intermediates (()-12(m,n)
and 13(m,n) are long enough (m + n g 5) they may make a
new N-CH2-N bridge between the ends of the oligomer chain
to deliver macrocyclic compounds. In this manner, (()-12(m,n)
leads to (()-14(m,n) and 15(m,n) by connection of homotopic
NH groups. The previously isolated (()-bis-ns-CB[6] and bis-
ns-CB[10] serve as examples of these types of macrocycles and
provide strong evidence for the depicted mechanistic pathway.
To provide even stronger evidence for this mechanistic scheme
we performed product resubmission experiments with trimer 3
and pentamer 5. When trimer 3 was treated with 1 equiv of
formaldehyde in HCl we observed the formation of (()-bis-
with Formaldehyde (2 equiv) Alone and in Combination
entry
reactant(s)
CB[5]
CB[6]
CB[7]
CB[8]
1
2
3
4
5
6
1
2
3
4
5
6
15%
12%
8%
5%
84%
0%
14%
20%
35%
27%
3%
26%
10%
41%
11%
48%
57%
53%
68%
75%
22%
10%
100%
52%
43%
44%
67%
90%
41%
53%
24%
42%
36%
21%
30%
17%
17%
33%
6%
2%
3%
0%
40%
0%
0%
3%
4%
3%
0%
0%
3%
7%
0%
5%
3%
7%
0%
7
1 + 2
1 + 3
1 + 4
1 + 5
1 + 6
2 + 3
2 + 4
2 + 5
3 + 4
3 + 5
4 + 5
31%
33%
18%
6%
8a
9
10
11
12
13
14
15
16
17
7%
30%
17%
35%
42%
13%
15%
1
ns-CB[6] by H NMR at partial conversion.35,39 When 5 was
treated similarly, we observed and quantified the formation of
1
CB[6] and bis-ns-CB[10] (0.39:1 ratio) by H NMR spectros-
copy of the crude reaction mixture. In combination, these results
provide strong evidence for the operation of a step-growth cyclo-
oligomerization reaction in the mechanism of CB[n] formation.
This mechanistic analysis that can be used to rationalize the
formation of 2- 8, ns-CB[6], (()-bis-ns-CB[6], and bis-ns-
CB[10] can also be used to predict the structures of ns-CB[n]
that have not yet been isolated.40 For example, although (()-
11(n) may cyclize to yield ns-CB[n] by connection between
enantiotopic NH groups it may also react to form (()-ns-CB[n]
by connection between homotopic NH groups. Similarly, either
13(m,n) or (()-12(m,n) may be transformed into (()-16(m,n)
by the addition of an appropriate bridging group. Two additional
bis-ns-CB[n] can be conceived by addition of a CH2-bridge
between enantiotopic NH groups of 13(m,n) which delivers
17(m,n) and 18(m,n).38 We are particularly intrigued by the
geometrical features of 18(m,n) which is reminiscent of the
calixarenes with bridging at the lower rim and flexibility at
the upper rim. We predict that 18(m,n) will display quite
interesting hybrid recognition and dynamic behavior.
Oligomer Resubmission Experiments. Although we viewed
the evidence for the mechanistic scheme presented above as
strong given the isolation of 2-6, bis-ns-CB[10], (()-bis-ns-
CB[6], and ns-CB[6] and previous mechanistic studies by
us21–25,27,35 and others10,29,41 we wanted to obtain further
evidence that applies directly to CB[n] formation rather than
relying on evidence from ns-CB[n] formation. For this purpose
we decided to perform product resubmission experiments
between formaldehyde (2 equiv) and oligomers 2-6 alone and
in binary mixtures. We assess the outcome of these reactions
by 1H NMR spectroscopy of the crude reaction mixtures in 20%
DCl in which CB[5], CB[6], CB[7], and CB[8] display separate
easily integrated resonances (Table 3).42
a A small amount (4%) of i-CB[6] was also detected in this reaction
mixture.
for the results described below, delivers a reaction mixture that
contains mainly CB[5] (15%), CB[6] (53%), and CB[7] (30%);
CB[8] content is low (2%). In contrast, use of dimer 2 and trimer
3 lead to reaction mixtures that contain higher percentages of
CB[6] and diminished amounts of CB[5] and CB[7] (Table 3,
entries 2 and 3). This result is in accord with the portion of our
mechanistic scheme which is based on step-growth polymeri-
zation in that 2 would be expected to give enhanced yields of
CB[n] where n is an even number (e.g., a multiple of 2) and 3
would be expected to give enhanced yields of CB[n] where n
is a multiple of 3. As anticipated based on the step-growth cyclo-
oligomerization model for CB[n] formation, the reaction of
tetramer 4 was found to give a high yield of CB[8] (40%) and
lesser amounts of CB[6] and CB[7] (Table 3, entry 4). The
reactions of pentamer 5 and hexamer 6sthe first oligomers
capable of direct unimolecular macrocyclizationswere particu-
larly interesting (Table 3, entries 5 and 6). Pentamer 5 delivers
mainly CB[5] (84%) and 6 delivers only CB[6] (100%) which
indicates that these oligomers are preorganized to undergo
macrocyclization in preference to further oligomerization. In
these reactions conducted under aqueous acidic conditions, with
the use of purified oligomeric building blocks 2-6, the operation
of step-growth processes is not exclusiVe as can be seen for
example in the reaction of 3 (Table 3, entry 3) which gives
CB[5] and CB[7] as side products. This indicates that the under
the reaction conditions a given oligomer can undergo fragmen-
tation processes (e.g., tetramer fragments to two dimers or to a
trimer and monomer) to yield shorter oligomers which can then
recombine to also give longer oligomers.10 We have indicated
this reversibility explicitly in Scheme 4.
Reactions Conducted Between Formaldehyde and 1-6.
Initially, we conducted the separate reaction between formal-
dehyde (2 equiv) and 1-6 (Table 3). The reaction of monomer
alone (Table 3, entry 1), which serves as our point of reference
Reactions Conducted Between Formaldehyde and Binary
Combinations of Building Blocks 1-6. Although a complete
analysis of the reactions conducted between formaldehyde (2
equiv with respect to total building block concentration) and
binary mixtures of building blocks 1-6 is not possible given
the mixtures that are generally obtained from CB[n] forming
reactions it is possible to tease out some information that we
believe is significant (Table 3, entries 7-17). For example,
combinations of oligomers 2-4 with monomer 1 increases the
proportion of the first oligomer capable of macrocyclization,
CB[5], which is consistent with the increased importance of a
chain-growth process when the amount of monomer is
(39) If the reaction is run to completion we observe a more complex mixture
whose analysis by 1H NMR spectroscopy is challenging.
(40) This analysis is theoretically based and ignores the potential influence
of strain and intramolecular NH · · ·O H-bonds that are known to impact
the stability and geometry of CB[n] and (()-bis-ns-CB[6].
(41) Blanch, R. J.; Sleeman, A. J.; White, T. J.; Arnold, A. P.; Day, A. I.
Nano Lett. 2002, 2, 147–149.
(42) In none of these reactions did we observe the formation of CB[10]·CB[5]
which suggests that the formation of this aggregate is probably more
complex than the reaction of two molecules of 5 to give CB[10]
followed by stabilization with CB[5].
9
8452 J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008

Mechanism of CB[n] Formation
A R T I C L E S
significant.10,43 The results obtained with combinations of
building blocks intended to lead to CB[7] (1 + 6, 2 + 5, and
3 + 4; Table 3, entries 11, 14, 15) are interesting. Of these
combinations, 1 + 6 delivers mainly CB[6], 2 + 5 delivers a
significant portion of CB[5] due to competing cyclization of 5
along with CB[7] (35%), and 3 + 4 delivers a substantial amount
of CB[6] presumably due to competing formation of CB[6]
formed from two molecules of 3 along with CB[7] (42%). In
combination, these reactions shed light on one of the reasons
why CB[6] is the dominant CB[n] formed under a variety of
conditions. Although there are a variety of pathways that lead
to CB[6] in high relative yield (e.g., 6, 1 + 5, 2 + 4, 3 + 3)
there are fewer pathways that lead to CB[7] (e.g., 2 + 5 and 3
+ 4) or CB[8] (only 4 + 4) and those do so only in modest
yield due to competing dimerization, cyclization, or fragmenta-
tion (e.g., tetramer to trimer and monomer) pathways. In this
regard, the quantitative cyclization of 6 suggests that this
cyclization (Table 3, compare entries 5 and 6) may be
particularly favorable from a kinetic standpoint.
Is CB[n] Formation Under Kinetic or Thermodynamic Con-
trol? Previous experimental work from Day’s group10 has clearly
demonstrated by product resubmission experiments that CB[8]
undergoes ring contraction to yield a mixture of CB[5], CB[6],
and CB[7]. This result strongly suggests that CB[5], CB[6], and
CB[7] are thermodynamically more stable than CB[8]. This
conclusion is backed up by the theoretical calculations at various
levels of theory.7,44 Day also reported that similar product
resubmission experiments with purified individual samples of
CB[5], CB[6], and CB[7] do not deliver mixtures of CB[n]
homologues. This result suggests that CB[5], CB[6], and CB[7]
are thermodynamically sufficiently stable such that their forma-
tion is irreVersible under the reaction conditions (conc. HCl,
100 °C). Accordingly we have indicated the entry into the CB[n]
(n ) 5, 6, 7) manifold (Scheme 4) as an irreversible step.45
Conversely, the isolation of 2-6, i-CB[n], and various ns-CB[n]
demonstrates that kinetic products do accumulate during CB[n]
synthesis under milder conditions. Perhaps most informative are
the product resubmission experiments performed with 2-6 (Vide
supra) that demonstrate that both kinetic and thermodynamic
considerations are relevant under the standard reaction condi-
tions. Although CB[5], CB[6], and CB[7] may be the ultimate
thermodynamically controlled products of CB[n] forming reac-
tions, the operation of a step-growth cyclo-oligomerization
process during the reaction suggests routes to the tailor-made
synthesis of new CB[n]-type receptors (e.g., i-CB[n] and nor-
seco-CB[n]) that are kinetically stable and display exciting
recognition properties (e.g., chiral recognition and homotropic
allostery).20,21,27,35,36
Figure 3. 1H NMR spectra (400 MHz, D2O, room temp) for (a) 23, (b)
6·23, (c) a mixture of 6·23 and excess 23, (d) 24, and (e) 6·24.
by forming oligomer·guest complexes that either promote or
disfavor certain cyclization reactions although the precise kinetic
or thermodynamic factors influencing such templated reactions
remain unclear. Given access to oligomers 2-6 we decided to
test their ability to form complexes with representative guests.
We did not observe any evidence of binding between 2-4 and
19 or 20 in water. This observation is not particularly surprising
given that ion-dipole interactions between guest and fully
formed ureidyl CdO portals of CB[n] provide substantial driving
force in the formation of CB[n]·guest complexes.7,16,35,46 In
contrast, we find that pentamer 5 and hexamer 6 begin to exhibit
recognition properties characteristic of CB[n]. For example, 5
forms a complex with 19 that displays slow exchange on the
chemical shift time scale. Slow exchange kinetics on the
chemical shift time scale are commonly observed for CB[n]
host-guest complexes but are less common for other molecular
containers like cyclodextrins and calixarenes. Evidence of
binding between 5 and other guests (21-24) is apparent based
1
on induced upfield shifts in the H NMR spectrum although
1
exchange is faster than the H NMR chemical shift time scale
(Supporting Information). The recognition behavior of 6 is
equally interesting (Figure 3). For example, hexamer 6 retains
the ability to bind with common guests for CB[6] (e.g., 19-21
and 25) and does so with slow exchange on the chemical shift
time scale which is noteworthy given the acyclic nature of 6.
Other guests that are too large for CB[6] (e.g., 23, 24, and 26)
are readily complexed by 6. The ability to bind these guests,
which are good guests for CB[7] (23 and 26) or even CB[8]
(24), indicates that hexamer 6 is quite flexible in solution and
is able to expand its cavity and wrap itself around larger guests
(e.g., 24) to form complexes (Figure 4). One particularly
interesting aspect shown in Figure 4 is that the NH tips of 6
are held apart from one another in a way that would disfavor
direct cyclization to CB[6]. Overall, these results show that
glycoluril oligomers 5 and 6 and by extension 7 and 8 are
capable of binding guests typical of the CB[n] family. This
suggests that the use of suitable templating groups that either
Glycoluril Oligomers 5 and 6 Retain the Ability to Bind
Ammonium Ions. Previously, Day’s group has extensively studied
the ability of organic and alkali metal cations to act as templates
for a specific CB[n] during the CB[n] forming reaction from 1
and formaldehyde (2 equiv).29 The influence of such guests as
templates on the ratio of CB[n] formed is generally modest.
Day hypothesizes that cationic templates exert their influence
(43) Day has previously shown that CB[n] forming reactions when
conducted at high dilution lead mainly to CB[5]. At high dilution
intermolecular reactions are slowed and chain-growth rather than step-
growth processes tend to dominate.
(44) (a) Oh, K. S.; Yoon, J.; Kim, K. S. J. Phys. Chem. B 2001, 105, 9726–
9731. (b) Maslii, A. N.; Grishaeva, T. N.; Kuznetsov, A. M.; Bakovets,
V. V. J. Struct. Chem. 2007, 48, 552–557.
(45) The reaction of ns-CB[8] to CB[8] is reversible as established by Day’s
product resubmission experiments.
(46) Ong, W.; Kaifer, A. E. Organometallics 2003, 22, 4181–4183.
9
J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008 8453

A R T I C L E S
Huang et al.
Figure 4. Cross-eyed stereoview of the MMFF minimized geometry of the 6·24 complex. Color code: C, gray; H, white; N, blue; O, red; H-bonds,
red-yellow striped.
Chart 2. Guests for 5 and 6
CB[7] or CB[8] and the pronounced kinetic tendency of 6 to
undergo cyclization to CB[6] rather than oligomerization or
fragmentation. The formulation of the mechanism of CB[n]
formation as a step growth cyclo-oligomerization pathway also
allows us to predict the structures of different-sized relatives
of the previously isolated bis-ns-CB[10], (()-bis-ns-CB[6], and
ns-CB[6] and formulate the structures of classes of nor-seco-
CB[n] that have not yet been isolated including (()-ns-CB[6]
and 17(m,n) and 18(m,n). We predict that 18(m,n) will show
particularly interesting recognition properties that blend the
advantages of the CB[n] family of macrocycles with those of
the calixarenes.
promote or disfavor cyclization of a particular oligomer may
be useful in directing the CB[n] forming reaction toward an
enhanced yield of a single CB[n] compound.
Of particular interest is the observation that 5 and 6 are
capable of binding CB[6] sized guests (e.g., 19 and 20) and
can even expand their cavities to associate with larger guests
(e.g., 23 and 24) typically bound within CB[7] or CB[8]. This
result establishes a high level of flexibility of the growing
methylene-bridged glycoluril oligomer chain that suggests that
the presence of suitable templating guests in the CB[n] forming
reaction may be able to direct the reaction toward enhanced
yield of a single sized CB[n]. This result also raises the prospect
of using suitable guests to template the formation of larger ns-
CB[n] (n g 10), ns-CB[n] containing multiple cavities and
lacking more than two CH2-groups,27 and even the enantiose-
lective synthesis of chiral ns-CB[n] (e.g., (()-14(m,n), bis-ns-
CB[n + m]). Given the wide range of applications to which
CB[n] (e.g., sensing, molecular machines, drug delivery)6 and
its derivatives can be applied (e.g., polymer nanocapsules,
affinity chromatography, and artificial ion-channels)7,8 we expect
that the ability to target tailor-made CB[n], CB[n] derivatives,
and ns-CB[n] which is enabled by the mechanistic insights
described herein will lead to new application areas that currently
benefit from calixarene or cyclodextrin molecular containers.
Conclusions
In summary, we have described the synthesis, purification,
and solution and solid state characterization of methylene-
bridged glycoluril oligomers 2-6. As expected based on
previous model studies, glycoluril oligomers 2-6 possess the
energetically favored all C-shaped geometry. The curvature of
2-5 maps well onto that exhibited by CB[6] which is calculated
to be the least strained member of the CB[n] family. The packing
ofoligomers4-6inthecrystalisdominatedbyCdO· · ·Na· · ·OdC
and NH · · ·OdC interactions. Most interesting is the dimeric
geometry of 6 in the crystal which demonstrates the inherent
flexibility of the methylene-bridged glycoluril oligomer chain
and suggests that self-association should be considered as a side
pathway in the mechanism of CB[n] formation. On the basis of
the isolation of 2-6 and the previously isolated bis-ns-CB[10],
(()-bis-ns-CB[6], and ns-CB[6], we were able to formulate a
detailed picture of the mechanism of CB[n] formation based
on a step-growth cyclo-oligomerization process. To further
support the step-growth mechanism we performed product
resubmission experiments between formaldehyde (2 equiv) and
oligomers 1-6 and binary combinations thereof and observed
the effect on the ratio of CB[5]-CB[8] formed. Finally, we
find that pentamer 5 and hexamer 6, but not 2-4, retain the
ability to bind organic ammonium ions (e.g., 19-26) in water.
In conclusion, this study has painted a much more detailed
picture of the mechanism of CB[n] formation, based on step-
growth cyclo-oligomerization, than was previously possible. For
example, this study allows us to rationalize the preferred
formation of CB[6] in CB[n] forming reactions as a consequence
of the multiplicity of pathways that leads to CB[6] relative to
Acknowledgment. We thank the National Science Foundation
(Grant CHE-0615049) and the University of Maryland (Ann Wylie
Dissertation Fellowship to W.-H. H.) for financial support of this
work. We thank the reviewers for their thoughtful comments that
helped improve the manuscript.
Supporting Information Available: Experimental procedures
and characterization data, crystallographic information files for
2, 4, 5, and 6 (cif), and selected NMR spectra for host-guest
complexes. This material is available free of charge via the
Internet at http://pubs.acs.org.
JA8013693
9
8454 J. AM. CHEM. SOC. VOL. 130, NO. 26, 2008