Dissipative self-assembly of a molecular gelator by using a chemical fuel

Article abstract of DOI:10.1002/anie.201001511

(Figure Presented) Fueling the future: A fibrillar network (red fibers, see figure) is formed from an activated building block (red), which is obtained from a synthetic gelator (blue) in a dissipative self-assembly process that is fueled by an alkylating agent. When the available energy is depleted, the system reverts to its thermodynamic equilibrium, that is, an isotropic solution.

Full text of DOI:10.1002/anie.201001511

Angewandte
Chemie
DOI: 10.1002/anie.201001511
Self-Organization
Dissipative Self-Assembly of a Molecular Gelator by Using a Chemical
Fuel**
Job Boekhoven, Aurelie M. Brizard, Krishna N. K. Kowlgi, Ger J. M. Koper, Rienk Eelkema,
and Jan H. van Esch*
The construction of energy-dissipating self-assembling sys-
tems,[1] which, like self-assembled structures found in nature
are formed transiently, far from equilibrium, and under the
constant influx of chemical energy, still represents a frontier
in nanoscale assembly.
(DSA).[12] In general, DSA systems consist of non-assembling
entities which, through activation by an energy source,
assemble into ordered structures. Energy dissipation causes
deactivation of the building blocks, hence leading to a
collapse of the formed structures. A typical example is
microtubule assembly that uses guanosine-5’-triphosphate
(GTP) as an energy source, which in turn catalyzes the
hydrolysis of GTP and therefore its own collapse.[13] The
microtubule assembly process is controlled by feedback loops
that lead to self-organization, including oscillatory behavior
and nonlinear responses of microtubule formation, which are
essential for rapid morphogenic alterations, self-healing, and
self-replication.
These fascinating properties of natural DSA systems have
motivated research on their artificial counterparts. Several
artificial DSA systems based on natural building blocks have
been reported.[14] Examples of fully artificial DSA systems are
most commonly found in the top-down engineered meso-
scopic regime[15] with hard inorganic or polymeric objects. The
few examples that concern soft matter are mostly fueled by
light,[16] whereas the dissipation of “chemical fuels” has been
used to drive mechanical motion.[17] It remains a challenge to
develop a DSA system that is chemically fueled.
A first step towards the development of a self-organizing
self-assembly system is the construction of a simple DSA
system without feedback control loops. Such a simple system
typically follows a sequence of processes.[18] Firstly, an energy
source activates the precursor building blocks so that self-
assembly is favored. Upon self-assembly, the activated build-
ing block can dissipate its energy, thus resulting in the
formation of the initial building block and disassembly of the
architecture. A requirement is that the rate of energy
dissipation (Pd) should be lower than the consumption of
fuel (Pc) to allow the formation of self-assembled architec-
tures (Figure 1).
Herein we present a synthetic DSA fibrous network that
uses chemical fuel as an energy source. A gelator precursor is
converted into a gelator by reaction with a chemical fuel, thus
leading to self-assembly. Hydrolysis of the gelator, which is
labile under ambient conditions, leads to energy dissipation
and disassembly of the formed structures.
The self-assembly of small molecules, polymers, proteins,
nanoparticles, colloids, and particles with sizes that approach
the mesoscale under thermodynamic equilibrium conditions
has been a powerful approach for the construction of a variety
of structures of nano- to micrometer dimensions, like vesicles,
capsules, and nanotubules.[2] The reversible nature of self-
assembly processes has been exploited in switchable,[3]
adaptive,[4] and autopoietic[5] self-assembling systems, which
lead to novel responsive materials and artificial systems that
are capable of self-replication and compartmentalization.
Recently, there has also been a strongly growing interest in
self-assembled materials obtained under non-equilibrium
conditions. For instance, the formation of hierarchically
structured membranes in a reaction-diffusion field,[6] and
the orthogonal self-assembly of molecular gels with surfac-
tants,[7] liquid crystals,[8] or other components[9] can be
controlled through the processing conditions, thus leading to
a much richer structural diversity compared to equilibrium-
processed materials. These self-assembled structures offer
new and intriguing opportunities for functional materials and
biomimetic cellular structures. Nevertheless, in all these cases,
the final self-assembling systems reside in a (local) thermo-
dynamic minimum state.
Despite these advances, the permanent nature of these
synthetic self-assembled structures does not compare well to
the complex spatiotemporally confined self-assembly pro-
cesses seen in natural systems, which for instance allow the
dynamic compartmentalization of incompatible processes,
responsiveness, and self-healing. Natural self-assembled
structures such as the cytoskeleton[10] and phospholipid
membranes[11] are formed by dissipative self-assembly
[*] J. Boekhoven, Dr. A. M. Brizard, K. N. K. Kowlgi, Dr. G. J. M. Koper,
Dr. R. Eelkema, Prof. Dr. J. H. van Esch
Department of Chemical Engineering
Delft University of Technology
Reactive gels have been previously reported[19] and the
hydrolysis of ester functions has been exploited to achieve an
enzymatically controlled gel–sol phase transition.[20] The
design of the dissipative self-assembling system presented
here is based on dibenzoyl-(l)-cystine (DBC; Bz = benzoyl), a
well-known pH-responsive hydrogelator .[21] Above their pKa
value (ca. 4.5), intermolecular repulsion occurs between the
anionic carboxylic acid groups of DBC, and therefore DBC
Julianalaan 136, 2628 BL, Delft (The Netherlands)
Fax: (+31)15-278-4289
E-mail: j.h.vanesch@tudelft.nl
Homepage: http://www.dct.tudelft.nl/sas
[**] We thank Marta E. Dobrowolska for obtaining SEM micrographs.
We acknowledge financial support from the Netherlands Organ-
ization for Research (NWO).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201001511.
Angew. Chem. Int. Ed. 2010, 49, 4825 –4828
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4825

Communications
Figure 1. Dissipative self-assembly: a monomeric building block (blue)
Figure 2. a) The net reaction of one DSA cycle. b) Reaction cycle of the
dissipative system. DBC dicarboxylate (top left) can react with MeI
(fuel) to give monoester DBC-OMe. DBC-OMe can dissipate its energy
through hydrolysis to form DBC or react again with the fuel to form
diester DBC-OMe2. The diester can assemble into fibers, which will be
in equilibrium between formation and hydrolysis of the diester until all
the MeI has reacted.
is activated by fuel consumption and is able to assemble (forming red
fibers). In the assembled state it can dissipate its energy and revert to
its monomeric state (blue).Energy is both consumed and dissipated in
one cycle; self-assembly can occur only if sufficient energy is available.
We found that DBC can be converted into its gelating
diester by reaction with MeI (fuel). Hydrolysis of the diester
back to its nongelating carboxylate derivative (DBC) is the
energy-dissipating step. The addition of sufficient fuel to the
system will result in a state in which the diester is continuously
formed and hydrolyzed. When the diester is present in a
sufficiently high concentration, fibrous aggregates or gels can
be formed. However, as soon as the source of energy is
consumed, the system will hydrolyze to the diacid and return
to its thermodynamically stable non-aggregated state (Fig-
ure 2b).
The DSA system was tested by the addition of MeI
(1.0 mmol, resulting in an initial concentration of 100 mm) to
an aqueous solution of DBC (10 mL,100 mm) at pH 7. The
course of the reaction was followed by HPLC–MS, dynamic
light scattering, and the consumption of hydroxide ions.
HPLC–MS showed that the monoester DBC-OMe was at a
concentration of 0.25 mm after 3 hours (Figure 3a). During
the course of the reaction, the concentration of DBC-OMe
increased further until the concentration leveled off around
10 mm after 100 hours, thus suggesting that a steady state
between the rate of formation of the monoester and its
hydrolysis had been reached.
The increased DBC-OMe concentration should result in
an increased rate of formation of the diester DBC-OMe2, and
after 72 hours DBC-OMe2 was indeed detected by HPLC–
MS. Analogously to DBC-OMe, the concentration of DBC-
OMe2 reached a plateau around 1.8 mm after 100 hours of
reaction time. Interestingly, this increase was accompanied by
an increase in turbidity, hence indicating the formation of
large aggregates. Light scattering studies confirmed the
increase in the formation of large structures (Figure 3b) and
bright field microscopy revealed the presence of fibers in the
system (Figure 4a). The fibers were separated from the
solution by two centrifugation and washing steps and HPLC–
MS showed that the fibers consisted of 96% DBC-OMe2 and
remains in the isotropic solution. However, reduction of the
pH value to below the pKa value leads to protonation and
hence neutralization of the carboxylate groups. As a result,
the molecules self-assemble into elongated fibers, which are
stabilized by intermolecular hydrogen bonds and hydro-
phobic interactions. The assembly of DBC diesters, which do
not contain carboxylate groups, is independent of the pH
value.[22] This observation implies that above the pKa value,
esterification of nongelating DBC should lead to the forma-
tion of fibrous aggregates.
Carboxylate groups can be esterified using alkylating
agents.[23] In this particular case, alkylation by methyl iodide
(MeI) in water under ambient conditions was successful,
albeit rather slow. To increase the rate of alkylation, the
reaction was performed at 358C. Under these conditions, the
hydrolysis reaction of MeI was found to be negligible.[24] On
the contrary, the formed ester is prone to hydrolysis, which
reforms the initial nongelator DBC. The rate of hydrolysis can
be controlled by the pH value, which, combined with temper-
ature control of the esterification rate, makes the system
highly tunable. We found that at pH 7 and 358C, the
formation of ester was faster than its hydrolysis, which is a
requirement for a DSA system. In the alkylation of DBC to
form the diester, two molecules of MeI are consumed, and
two iodide ions are released. The subsequent hydrolysis to
DBC consumes two hydroxide ions and produces two
molecules of MeOH (Figure 2a). This net reaction implies a
decrease of the pH value. To counter this decrease, the
reaction was carried out in a pH-stat setup, which provides the
additional advantage that hydroxide ion uptake (and there-
fore the rate of ester hydrolysis) can be monitored (see the
Supporting Information).
4826 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 4825 –4828

Angewandte
Chemie
&
*
Figure 3. a) Concentrations of DBC ( ), DBC-OMe ( ), and DBC-
!
OMe2 ( ) as determined by HPLC–MS over time. After 3 h the
formation of DBC-OMe was observed. After 100 h the maximum
concentrations of DBC-OMe and DBC-OMe2 were reached and the
hydrolysis rate is higher than the methylation rate. b) Quantitative
measurement of aggregation by dynamic light scattering (DLS;
l=633 nm) in counts per second. Aggregation is observed after 70 h.
Figure 4. a) Bright field micrographs of fibers after 100 h. Scale
bars=20 mm. b) SEM images of fine structure of separated fibers after
100 h. Scale bars=0.5 mm. c) Hydroxide ion uptake by the system over
time. The inset shows a magnification of the first 60 h. At t=0,
1 mmol MeI was added and the hydroxide uptake started to grow
exponentially. After 350 h both the esters had been hydrolyzed and the
hydroxide ion uptake leveled off at 1 mmol. After 400 h, a further
1 mmol of MeI was injected and the process was repeated.
4% DBC-OMe (see the Supporting Information). The
centrifuged pellet was also examined by scanning electron
microscopy (SEM), which revealed the fine structure of the
fibers (Figure 4b).
After 120 hours, the concentrations of both esters clearly
started to decrease, which indicated that the rate of hydrolysis
was higher than the rate of ester formation most likely as a
result of the decreasing fuel concentration. After 350 hours,
HPLC–MS showed that both esters had been hydrolyzed to
DBC, and light scattering confirmed that all the fibers had
disassembled.
the formation and hydrolysis of the DBC ester. The same
concentrations of ester were obtained as in the first cycle.
Bright field microscopy revealed the presence of fibers,
however, the scattering intensity did not reach the previous
level, which might indicate that the fuel consumption
products (MeOH and IÀ) disturb fiber formation to some
extent.
The ester hydrolysis was also followed by monitoring
hydroxide ion uptake. By adding fuel to the system, the
hydroxide ion uptake started to grow exponentially, which
confirmed the hydrolysis of the ester (Figure 4c). Moreover,
the exponential growth confirmed that the ester was accu-
mulating. After 350 hours, the uptake leveled off, thus
indicating that hydrolysis had stopped. The total amount of
hydroxide taken up by the system was in excellent agreement
with the initial amount of fuel added.
The concurrent gelator conversion, formation, and dis-
assembly of fibers, and consumption of fuel shows that the
transient self-assembly is driven by energy dissipation. At the
end of the process, when all fuel is consumed, the system
reverts back to its thermodynamic minimum; that is, an
isotropic solution.
We have shown how the dissipative self-assembly of a
synthetic gelator can be brought about by using a chemical
fuel. A nongelating diacid is activated to self-assemble
through a double alkylation by methyl iodide, the chemical
fuel. After a certain period, the formation of the diester is
observed, which leads to the assembly of fibrous aggregates.
These diesters are prone to hydrolysis, thus resulting in energy
dissipation. When all the MeI has reacted, hydrolysis of the
gelator molecules results in the disassembly of the fibrous
aggregates. This system meets all requirements for dissipative
self-assembly; reaction of a building block with a fuel leads to
activation, which favors self-assembly. This activated building
block can be deactivated by energy dissipation. Activation
occurs at higher rates than deactivation, hence leading to a
steady self-assembled state, which is continuously being
formed and degraded as long as energy in the form of
chemical fuel is available. To the best of our knowledge, the
approach described here is the first chemically fueled
synthetic DSA system, which is a new approach towards
Encouraged by the above results, we anticipated that the
addition of a second batch of fuel would again induce self-
assembly. Indeed, after a second addition of MeI (1 mmol) the
formation of large aggregates was observed, accompanied by
Angew. Chem. Int. Ed. 2010, 49, 4825 –4828
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4827

Communications
[10] D. A. Fletcher, R. D. Mullins, Nature 2010, 463, 485.
[11] B. Alberts, D. Bray, J. Lewis, M. Raff, K. Roberts, J. D. Watson,
Molecular Biology of the Cell, 3rd ed., Garland Publishing, New
York, 1994.
[12] a) G. M. Whitesides, R. F. Ismagilov, Science 1999, 284, 89;
b) B. A. Grzybowski, C. E. Wilmer, J. Kim, K. P. Browne,
K. J. M. Bishop, Soft Matter 2009, 5, 1110.
[13] A. Desai, T. J. Mitchison, Annu. Rev. Cell Dev. Biol. 1997, 13, 83.
[14] a) H. Hess, J. Clemmens, C. Brunner, R. Doot, S. Luna, K. H.
Ernst, V. Vogel, Nano Lett. 2005, 5, 633; b) C. Z. Dinu, J. Opitz,
W. Pompe, J. Howard, M. Mertig, S. Diez, Small 2006, 2, 1090;
c) H. Liu, E. D. Spoerke, M. Bachand, S. J. Koch, B. C. Bunker,
G. D. Bachand, Adv. Mater. 2008, 20, 4476.
[15] a) B. A. Grzybowski, H. A. Stone, G. M. Whitesides, Nature
2000, 405, 1033; b) J. Vermant, M. J. Solomon, J. Phys. Condens.
Matter 2005, 17, R187; c) K. V. Tretiakov, K. J. M. Bishop, B. A.
Grzybowski, J. Phys. Chem. B 2009, 113, 7574.
[16] a) T. R. Kelly, H. De Silva, R. A. Silva, Nature 1999, 401, 150;
b) L. N. Lucas, J. H. van Esch, B. L. Feringa, R. M. Kellogg,
Chem. Commun. 2001, 759; c) J. J. D. de Jong, L. N. Lucas, R. M.
Kellogg, J. H. van Esch, B. L. Feringa, Science 2004, 304, 278;
d) S. Yagai, T. Karatsu, A. Kitamura, Chem. Eur. J. 2005, 11,
4054; e) R. Eelkema, M. M. Pollard, J. Vicario, N. Katsonis, B. S.
Ramon, C. W. M. Bastiaansen, D. J. Broer, B. L. Feringa, Nature
2006, 440, 163; f) R. Klajn, P. J. Wesson, K. J. Bishop, B. A.
Grzybowski, Angew. Chem. 2009, 121, 7169; Angew. Chem. Int.
Ed. 2009, 48, 7035; g) T. Soejima, M. A. Morikawa, N. Kimizuka,
Small 2009, 5, 2043.
complex self-assembled structures and smart materials. How-
ever, this system is not comparable with the complexity of
natural DSA systems. The next steps towards this goal would
include improved kinetics (e.g., by the use of enzyme
responsive gelators[4b,c]) to allow faster switching, and the
incorporation of autocatalytic feedback mechanisms that lead
to oscillating behavior through self-organization.[25] Such
systems can evolve into DSA architectures that are intrinsi-
cally capable of self-healing by continuous regeneration.
Received: March 12, 2010
Published online: May 28, 2010
Keywords: aggregation · gels · hydrolysis ·
.
non-equilibrium processes · self-assembly
[1] G. M. Whitesides, B. A. Grzybowski, Science 2002, 295, 2418.
[2] a) G. M. Whitesides, J. P. Mathias, C. T. Seto, Science 1991, 254,
1312; b) J. H. van Esch, B. L. Feringa, Angew. Chem. 2000, 112,
2351; Angew. Chem. Int. Ed. 2000, 39, 2263; c) L. Brunsveld,
B. J. B. Folmer, E. W. Meijer, R. P. Sijbesma, Chem. Rev. 2001,
101, 4071; d) J. M. Lehn, Proc. Natl. Acad. Sci. USA 2002, 99,
4763; e) I. W. Hamley, Angew. Chem. 2003, 115, 1730; Angew.
Chem. Int. Ed. 2003, 42, 1692; f) L. C. Palmer, S. I. Stupp, Acc.
Chem. Res. 2008, 41, 1674.
[17] a) T. J. Huang, B. Brough, C.-M. Ho, Y. Liu, A. H. Flood, P. A.
Bonvallet, H.-R. Tseng, J. F. Stoddart, M. Baller, S. Magonov,
Appl. Phys. Lett. 2004, 85, 5391; b) J. V. Hernꢀndez, E. R. Kay,
D. A. Leigh, Science 2004, 306, 1532; c) S. P. Fletcher, F. Dumur,
M. M. Pollard, B. L. Feringa, Science 2005, 310, 80; d) J. M.
Catchmark, S. Subramanian, A. Sen, Small 2005, 1, 202.
[18] M. Fialkowski, K. J. M. Bishop, R. Klajn, S. K. Smoukov, C. J.
Campbell, B. A. Grzybowski, J. Phys. Chem. B 2006, 110, 2482.
[19] a) J. F. Miravet, B. Escuder, Org. Lett. 2005, 7, 4791; b) J. F.
Miravet, B. Escuder, Tetrahedron 2007, 63, 7321.
[20] a) A. K. Das, R. Collins, R. V. Ulijn, Small 2008, 4, 279; b) R. J.
Williams, R. J. Mart, R. V. Ulijn, Biopolymers 2010, 94, 107.
[21] C. G. L. Wolfe, E. K. Rideal, Biochem. J. 1922, 16, 548.
[22] F. M. Menger, K. L. Caran, J. Am. Chem. Soc. 2000, 122, 11679.
[23] a) J. Koivurinta, A. Kyllonen, L. Leinonen, K. Valaste, J.
Koskikallio, Finn. Chem. Lett. 1974, 1, 239; b) J. G. Avila-
Zꢀrraga, R. Martꢁnez, Synth. Commun. 2001, 31, 2177.
[3] Molecular Gels (Eds.: R. G. Weiss, P. Terech), Springer, Dor-
drecht, 2005; Molecular Switches (Ed.: B. L. Feringa), Wiley-
VCH, Weinheim, 2001.
[4] a) N. Sreenivasachary, J. M. Lehn, Proc. Natl. Acad. Sci. USA
2005, 102, 5938; b) Z. Yang, G. Liang, L. Wang, B. Xu, J. Am.
Chem. Soc. 2006, 128, 3038; c) R. J. Mart, R. D. Osborne, M. M.
Stevens, R. V. Ulijn, Soft Matter 2006, 2, 822; d) P. T. Corbett, J.
Leclaire, L. Vial, K. R. West, J. L. Wietor, J. K. M. Sanders, S.
Otto, Chem. Rev. 2006, 106, 3652; e) R. Nguyen, L. Allouche, E.
Buhler, N. Giuseppone, Angew. Chem. 2009, 121, 1113; Angew.
Chem. Int. Ed. 2009, 48, 1093; f) C. B. Minkenberg, L. Florusse,
R. Eelkema, G. J. M. Koper, J. H. van Esch, J. Am. Chem. Soc.
2009, 131, 11274.
[5] a) P. A. Bachmann, P. L. Luisi, J. Lang, Nature 1992, 357, 57;
b) H. H. Zepik, E. Blochliger, P. L. Luisi, Angew. Chem. 2001,
113, 205; Angew. Chem. Int. Ed. 2001, 40, 199; c) P. L. Luisi,
Naturwissenschaften 2003, 90, 49; d) M. M. Hanczyc, S. M.
Fujikawa, J. W. Szostak, Science 2003, 302, 618.
[24] kOHhydrolysis = 1.6 ꢂ 10À3 (less then 0.1% was hydrolyzed after
10 days under experimental conditions).[23a]
[25] a) A. Pacault, P. Hanuss, P. De Kepper, C. Vial, J. Boissonade,
Acc. Chem. Res. 1976, 9, 438; b) R. Yoshida, T. Takahashi, T.
Yamaguch, H. Ichijo, J. Am. Chem. Soc. 1996, 118, 5134; c) J. R.
Howse, P. Topham, C. J. Crook, A. J. Gleeson, W. Bras, R. A.
Jones, A. J. Ryan, Nano Lett. 2006, 6, 73.
[6] R. M. Capito, H. S. Azevedo, Y. S. Velichko, A. Mata, S. I. Stupp,
Science 2008, 319, 1812.
[7] A. Brizard, M. Stuart, K. van Bommel, A. Friggeri, M. de Jong, J.
van Esch, Angew. Chem. 2008, 120, 2093; Angew. Chem. Int. Ed.
2008, 47, 2063.
[8] T. Kato, Science 2002, 295, 2414.
[9] a) J. R. Moffat, D. K. Smith, Chem. Commun. 2009, 316; b) J.
van Herrikhuyzen, A. Syamakumari, A. P. H. J. Schenning,
E. W. Meijer, J. Am. Chem. Soc. 2004, 126, 10021.
4828 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 4825 –4828