Nonmonotonic assembly of a deep-cavity cavitand

Article abstract of DOI:10.1021/ja200633d

The synthesis and assembly properties of a new water-soluble deep-cavity cavitand are discussed. For a homologous series of alkanes, the host can form a range of approximately isoenergetic 1:1, 2:1, and 2:2 complexes. As a result of this confluence of binding and assembly the host displays an unusual, nonmonotonic, assembly profile. Thus, no or limited assembly is observed with methane through butane, pentane triggers assembly, and hexane through octane again does not promote assembly, whereas nonane and a larger guest again induce assembly. This unusual behavior is discussed in the context of the diversity of nodes of chemical systems (networks).

Full text of DOI:10.1021/ja200633d

COMMUNICATION
pubs.acs.org/JACS
Nonmonotonic Assembly of a Deep-Cavity Cavitand
Haiying Gan, Christopher J. Benjamin, and Bruce C. Gibb*
Department of Chemistry, University of New Orleans, New Orleans, Louisiana 70148, United States
S Supporting Information
b
3,5-dihydroxy-4-methylbenzyl alcohol (R = Me in structure)
via an 8-fold Ullmann ether reaction to yield octol 3. Oxidation
then gave the crude product 1a, which was purified by an
esterificationÀhydrolysis procedure. Purification of octa-ester
4a is the only step in the synthesis requiring chromatography and
ensured the removal of impurities arising from both the weaving
step and the oxidation of octol 3. The cavitand 1a differs from
1b20,21 by possessing four methyl groups that in part project into
the hydrophobic cavity of the host but importantly also project
into the hydrophobic rim of the host that is important in the
predisposition25 of these types of molecules to dimerize and form
supramolecular capsules.18,22À24 As a result of the latter, the
predisposition of host 1a for dimerization is reduced relative to
the octa-acid 1b. This was first apparent in the 1H NMR
spectrum of 1a that showed sharp signals over the concentration
range 1À3 mM (in D2O buffered with 10À30 mM Na2B4O7). In
contrast, 1b showed broad signals at concentrations above 2 mM
(20 mM Na2B4O7) indicative of partial assembly. That host 1a is
monomeric over the 1À3 mM concentration range was con-
firmed by pulse-gradient stimulated spinÀecho (PGSE) NMR
experiments26 which reveal a diffusion rate of D = (1.79À1.90) Â
10À6 cm2 sÀ1, corresponding to a hydrodynamic volume of
between 6.3 and 7.6 nm3.
ABSTRACT: The synthesis and assembly properties of a
new water-soluble deep-cavity cavitand are discussed. For a
homologous series of alkanes, the host can form a range of
approximately isoenergetic 1:1, 2:1, and 2:2 complexes. As a
result of this ‘confluence’ of binding and assembly the host
displays an unusual, nonmonotonic, assembly profile. Thus,
no or limited assembly is observed with methane through
butane, pentane triggers assembly, and hexane through
octane again does not promote assembly, whereas nonane
and a larger guest again induce assembly. This unusual
behavior is discussed in the context of the diversity of nodes
of chemical systems (networks).
omplex systems possessing emergent phenomena are in-
C
herent to the world we live in.1À3 Complexity in chemical
systems arises from intricate networks of chemical entities4À6
linked by both reversible and irreversible covalent bond forma-
tion and noncovalent interactions. Subsuming these networks,
compartmentalization bestows a system with robustness; in
other words it ensures that the system can lie far from
equilibrium.7 Chemistry is beginning to build an understanding
of the different components of chemical systems. For example, it
is adept at synthesizing molecules, the chemical entities that
constitute the individual nodes of such networks and, more
recently, have began to establish a variety of means by which
external stimuli can switch nodes between one state and another
(on and off).8À14 Similarly, chemistry is also adept at controlling
the kinetics and thermodynamics of molecular interactions, the
individual links, or edges of networks.15,16 Be that as it may, our
appreciation of the different kinds of nodes that are components
of complex chemical systems is limited; as is our appreciation of
how different combinations of multiple nodes and edges en-
gender network properties such as autoregulation or feed-
forward loops.4À6 Consequently, there is a need to both improve
our understanding of biological systems4 and devise wholly new
de novo networks. Here we identify a water-soluble deep-cavity
cavitand that functions as an unusual node. An examination of the
binding of a homologous series of n-alkanes (C1ÀC14)17 reveals
three possible supramolecular entities: 1:1, 2:1, and 2:2 hostÀguest
complexes. The small energy differences between many of the
different complexes leads to a ‘confluence’ at the node and an
unexpected, nonmonotonic, assembly profile.
We examined the binding of the homologous series of the n-
alkanes methane through n-tetradecane to cavitand 1a using a
1
1
combination of H and PGSE NMR experiments. H NMR
confirmed guest binding or encapsulation as evidenced by high-
field signals for the bound guest between 0 and À3.5 ppm. These
experiments also confirmed the number and ratio of species
formed by each guest and the ratio of the host and guest in the
different complexes. The PGSE NMR experiments were used to
determine both the extent of assembly and the stoichiometry of
the complexes formed.
The smallest guest examined, methane, was observed to
weakly bind to host 1a. This was apparent from the appearance
of a signal for the bound but fast exchanging guest at À0.02
ppm (Figure 1, cf. 0.2 ppm for free methane). This is in contrast
to the corresponding host 1b which showed no evidence of
methane binding in a study of hydrocarbon gas separation.17b
The binding of methane to host 1a is stronger because the four
methyl groups at the rim reduce the size of the binding pocket
somewhat. In addition, the methyl groups narrow the portal to
the pocket and undoubtedly increase the kinetic stability of the
1:1 complex. In the case of 1a, diffusion NMR experiments
confirmed that the complex possesses the same hydrodynamic
volume (HV) as the free host (6.3 nm3, Figure 2). The signals of
the bound methyl groups of the guests ethane, propane, and
Deep-cavity cavitand 1a (Scheme 1) was formed by an
analogous procedure to that used in the synthesis of a cavitand
1b (octa-acid) that has figured prominently in our research
endeavors.18,19 Briefly, known cavitand 2 was ‘woven’ with
Received: February 1, 2011
Published: March 14, 2011
r
2011 American Chemical Society
4770
dx.doi.org/10.1021/ja200633d J. Am. Chem. Soc. 2011, 133, 4770–4773
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 1. Synthesis of Deep-Cavity Cavitand Host 1a
Figure 2. Graph of the hydrodynamic volume (HV) of the complexes
formed between host 1a/b and alkanes guests, against the number of
carbon atoms in each guest. Data shown in black correspond to host 1a.
Data shown in blue correspond to the previously reported parent
cavitand.17a,b
Figure 1. 1H NMR spectra of the complexes formed between host 1a
and (a) methane; (b) propane; (c) n-pentane; (d) n-hexane; (e) n-
octane; (f) n-nonane; and (g) n-tetradecane. Shown is the guest binding
region (0.50 to À4.00 ppm) and the signal from the H atoms para to the
acetal group in the host (ca. 7.20 ppm). All solutions were 1 mM
complex in D2O, 10 mM Na2B4O7 buffer.
complexes (Figure 1). A lack of baseline resolution in the host
signals and broad signals from the bound guest precluded
measuring the HV values of each species. However, an average
HV value for the two species of 10.1 nm3 (Figure 2) confirmed
that these were the 1:1 and 2:2 host guest complexes. Integration
of the bound host signals confirmed that these two entities exist
in an approximate ratio of 1.8:1.27 This break in the monotonicity
of the extent of assembly continued with guests n-heptane and n-
octane. Both guest signals and many of the host signals were
again broad indicating intermediate exchange rates. Further-
more, the measured HV values for these complexes were
essentially the same as that obtained for ethane (Figure 2). In
other words, n-heptane and n-octane form 1:1 complexes with
host 1a. n-Nonane behaved very differently than n-octane, with
the NMR experiments demonstrating that this guest is an
efficient template for the formation of the corresponding 2:1
hostÀguest complex. More specifically, this second switch in the
behavior of the host was evident from the 1H NMR data
(Figure 1), which showed a kinetically stable 2:1 complex,
and the diffusion experiment that confirmed a capsule approxi-
mately the same size as that produced by pentane (HV =
13.2 nm3, Figure 2). Finally, for n-decane through n-tetradecane
there were monotonic shifts in both the bound guest methyl
signals and the HV values reflecting the induced-fit ‘swelling’ of
the host and concomitant increase in the packing coefficient
(Figures 1, 2 and SI).
n-butane followed the expected monotonic trend appearing
at À0.95, À1.62, and À1.95 ppm respectively (propane shown
in Figure 1). These signals were broader than that of fast binding
methane, an indication that exchange was slower and close to the
(500 MHz) NMR time scale. In line with this notion, a
competition experiment between methane and ethane revealed
a Krel for the latter of 3. Pertinent host signals also showed
evidence of broadening (Supporting Information (SI)). The
guest pentane continued this monotonic trend with a methyl
signal at À2.02 ppm but showed sharper guest signals suggesting
a kinetically more stable complex (Figure 1). Diffusion NMR
studies focusing on both host and guest signals showed
(Figure 2) that while the ethane complex was essentially mon-
omeric (HV = 7.6 nm3), propane and n-butane led to a mixture of
1:1 and 2:2 complexes (HV = 10.3 and 10.6 nm3 respectively),
while n-pentane led to mostly a 2:2 complex (HV = 12.6 nm3).
1
Returning to the H NMR data (Figure 1), we attribute the
broadening of bound guest signals in the case of propane to a
combination of intermediate exchange rates between the free and
the bound state in the 1:1 complex and intermediate exchange
rates between the 1:1 and 2:2 complexes. The sharper signals for
the pentane complex arise because there are only small amounts
of the 1:1 complex, and exchange is slow on the NMR time scale
1
in the case of the 2:2 complex. Unexpectedly, the H NMR
spectrum of the n-hexane indicates two, slow-exchanging
4771
dx.doi.org/10.1021/ja200633d |J. Am. Chem. Soc. 2011, 133, 4770–4773

Journal of the American Chemical Society
COMMUNICATION
With the aid of CPK models we interpret these results as
follows. Guests methane through butane bind to the host but are
small enough such that there is too much void space in the 2:2
complex and the 1:1 complexes are energetically preferred.
These results are in contrast to the cavitand devoid of methyl
groups at its rim (1b), which readily forms thermodynamically
and kinetically stable 2:2 complexes with propane and butane
(butane data shown in Figure 2). Hence, the methyl groups at the
rim of 1a reduce the relative stability of the capsular complex. On
the other hand, the size of n-pentane is such that it leads to little
void space in the 1:1 complex and dimerization to form a 2:2
complex is energetically preferred because this allows complete
dehydration of the hydrophobic surfaces that form the dimeriza-
tion interface. The guests n-hexane through n-octane are how-
ever too large to form stable 2:2 complexes and too small to form
stable 2:1 complexes. As a result it is the 1:1 complex that is
energetically preferred, especially in the case of n-heptane and n-
octane. In contrast however, the still larger guests are of sufficient
size to form a stable 2:1 hostÀguest complex; they do not form
1:1 complexes because a significant portion of the guest would
remain hydrated in free solution.
as an unusual switch because three different types of supramo-
lecular species can be formed, and because many of these species
are approximately isoenergetic. Considering the elaborate
switching properties of many proteins it is likely that nodes
displaying unusual properties are common, and even essential, to
biological networks. With this notion in mind we are continuing
our investigations of 1a to determine the degree of control that is
possible within such switching. We will report on these findings
at a future date.
’ ASSOCIATED CONTENT
Supporting Information. Experimental details, 1H NMR
S
b
and PGSE experiments, and truth table. This material is available
free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
bgibb@uno.edu
Although guests of nine carbons or more are good templates
for the formation of 2:1 hostÀguest capsular complexes, the
limited predisposition of host 1a to form a dimeric capsule means
that many of the different supramolecular species that can form
for the guests ethane through n-octane are approximately iso-
energetic. This ‘confluence’ of supramolecular species is evident
at both a ‘local level’, where more than one species is observed to
exist in solution, and ‘globally’ in the nonmonotonicity of the
diffusion data/assembly state. The confluence is all the more
intriguing when it is recalled that a small energy difference of
0.5 kcal molÀ1 establishes a 70:30 ratio of two equilibrated
species, and that competition experiments involving guest bind-
ing in simple 1:1 hostÀguest systems frequently reveal ΔΔG°
values of 2À3 kcal molÀ1 for guests differing only in a methylene
group.28,29 Hence for the guests ethane through n-octane, many
of the 1:1, 2:1, and 2:2 complexes and assemblies observed here
likely lie within 0À1.0 kcal molÀ1 of each other, a range that is
hard to engineer in a simple hostÀguests system.
’ ACKNOWLEDGMENT
B.C.G. acknowledges the financial support of the NSF (CHE-
0718461). H.G. acknowledges the Post-Katrina Support Fund
Initiative (PKSFI, LEQSF(2007-12)-ENH-PKSFI-PRS-04) for
support.
’ REFERENCES
(1) Bar-Yam, Y. Dynamics of Complex Systems; Westview: Boulder,
1997.
(2) Gibb, B. C. Nature Chem. 2009, 1, 17–18.
(3) Gibb, B. C. Nature Chem. 2009, 1, 252–253.
(4) Alon, U. An Introduction to Systems Biology: Design Principles of
Biological Circuits; Chapman & Hall/CRC: Boca Raton, FL, 2007.
(5) Ludlow, R. F.; Otto, S. Chem. Soc. Rev. 2008, 37, 101–108.
(6) del Amo, V.; Philp, D. Chem.—Eur. J. 2010, 16, 13304–13318.
(7) de Groot, S. R.; Mazur, P. Nonequilibrium Thermodynamics;
Dover: New York, 1984.
The self-assembly of cavitands such as 1a and 1b can be
envisioned in terms of Boolean logic. For example, a pair of
selected guests (A and B) can be considered as inputs, while the
output of the logic gate is either 0 (no assembly) or 1 (assembly
into a 2:1 or 2:2 complex).30 Hence, both 1a and 1b function as
B gates (true whenever B is true) when guests inducing assembly
are defined as B inputs. If an arbitrary threshold of 12 nm3
(Figure 2) is chosen to define responses of 0 and 1, then for host
1a the A inputs are methane to n-butane and n-hexane to n-octane,
whereas the B inputs correspond to n-pentane and n-nonane
through n-tetradecane. For host 1b, an input of A corresponds to
methane and ethane, and an input of B to guests propane and
larger. Defining these systems as two-input, one-output gates does
however fail to illustrate an important difference between hosts 1a
and 1b; that is that when considering adjacent pairs in the
homologous series, 1b is only capable of differentiating between
ethane and propane, whereas 1a can differentiate between
n-butane and n-pentane, n-pentane and n-hexane, and n-octane
and n-nonane. Hence, it is perhaps more appropriate to consider
the nine guests methane through n-nonane as unique inputs and
treat the system as a nine-input, one-output gate (SI).
(8) Molecular Devices and Machines: Concepts and Perspectives for the
Nanoworld; Wiley-VCH: Weinheim, 2008.
(9) Molecular Switches; Wiley-VCH: Weinheim, 2001.
(10) Zhao, Y.-L.; Li, Z.; Kagehie, S.; Botros, Y. Y.; Stoddart, J. F.;
Zink, J. I. J. Am. Chem. Soc. 2010, 132, 13016–13025.
(11) Klajn, R.; Stoddart, J. F.; Grzybowski, B. A. Chem. Soc. Rev.
2010, 39, 2203–2237.
(12) de Silva, A. P.; James, M. R.; Mckinney, B. O. F.; Pears, D. A.
Nat. Mater. 2006, 5, 787–789.
(13) Collier, C. P.; Wong, E. W.; Belohradsky, M.; Raymo, F. M.;
Stoddart, J. F.; Kuekes, P. J.; Williams, R. S.; Heath, J. R. Science 1999,
285, 391–394.
(14) Tian, H.; Wang, Q. C. Chem. Soc. Rev. 2006, 35, 361–374.
(15) Atkins, P.; de Paula, J. Physical Chemistry; W. H. Freeman and
Company: New York, 2006.
(16) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic
Chemistry; University Science Books: Sausalito, CA, 2006.
(17) For previous work detailing the encapsulation of hydrocarbons
in supramolecular nanocapsules, see: (a) Gibb, C. L. D.; Gibb, B. C.
Chem. Commun. 2007, 1635–1637. (b) Gibb, C. L. D.; Gibb, B. C. J. Am.
Chem. Soc. 2006, 128, 16498–16499. (c) Scarso, A.; Trembleau, L.;
Rebek, J., Jr. J. Am. Chem. Soc. 2004, 126, 13512–13518. (d) Ajami, D.;
Rebek, J., Jr. Angew. Chem., Int. Ed. 2008, 47, 6059–6061. (e) Branda, N.;
Wyler, R.; Rebek, J., Jr. Science 1994, 263, 1267–1268. (f) Hooley, R. J.;
Biros, S. M.; Rebek, J., Jr. Chem. Commun. 2006, 509–510.
In conclusion, we have identified a hostÀguest system with an
unusual nonmonotonic assembly profile. The system functions
4772
dx.doi.org/10.1021/ja200633d |J. Am. Chem. Soc. 2011, 133, 4770–4773

Journal of the American Chemical Society
COMMUNICATION
(18) Gibb, C. L. D.; Gibb, B. C. J. Am. Chem. Soc. 2004,
126, 11408–11409.
(19) Liu, S.; Whisenhunt-Ioup, S. E.; Gibb, C. L. D.; Gibb, B. C.
Supramolecular Chemistry 2011, 24, in press.
(20) Liu, S.; Gibb, B. C. Chem. Commun. 2008, 3709–3716.
(21) Laughrey, Z. R.; Gibb, B. C. Chem. Soc. Rev. 2011, 40, 363–386.
(22) Kaanumalle, L. S.; Gibb, C. L. D.; Gibb, B. C.; Ramamurthy, V.
J. Am. Chem. Soc. 2004, 126, 14366–14367.
(23) Natarajan, A.; Kaanumalle, L. S.; Jockusch, S.; Gibb, C. L. D.;
Gibb, B. C.; Turro, N. J.; Ramamurthy, V. J. Am. Chem. Soc. 2007,
129, 4132–4133.
(24) Gibb, C. L. D.; Sundaresan, A. K.; Ramamurthy, V.; Gibb, B. C.
J. Am. Chem. Soc. 2008, 130, 4069–4080.
(25) Rowan, S. J.; Hamilton, D. G.; Brady, P. A.; Sanders, J. K. M.
J. Am. Chem. Soc. 1997, 119, 2578–2579.
(26) Cohen, Y.; Avram, L.; Frish, L. Angew. Chem., Int. Ed. 2005,
44, 520–554.
(27) The ratio of the two complexes was determined by peak
deconvolution of the host signals at ca. 7.2 ppm.
(28) Houk, K. N.; Leach, A. G.; Kim, S. P.; Zhang, X. Angew. Chem.,
Int. Ed. 2003, 42, 4872–4897.
(29) Rekharsky, M. V.; Inoue, Y. Chem. Rev. 1998, 98, 1875–1917.
(30) For an example of how fluorescence resonance energy transfer
(FRET) can be used to differentiate between disassembled (off) and
assembled (on) states, see: Castellano, R. K.; Craig, S. L.; Nuckolls, C.;
Rebek, J., Jr. J. Am. Chem. Soc. 2000, 122, 7876–7882.
4773
dx.doi.org/10.1021/ja200633d |J. Am. Chem. Soc. 2011, 133, 4770–4773