Size-selective phase-transfer catalysis with interfacially cross-linked reverse micelles

Article abstract of DOI:10.1021/ol203319w

Cross-linking of the reverse micelles (RMs) of a triallylammonium surfactant afforded organic nanoparticles with introverted cationic groups. The cross-linked reverse micelles catalyzed size-selective biphasic reaction between sodium azide and alkyl bromi

Full text of DOI:10.1021/ol203319w

ORGANIC
LETTERS
2012
Vol. 14, No. 3
784–787
Size-Selective Phase-Transfer Catalysis
with Interfacially Cross-Linked Reverse
Micelles
Li-Chen Lee and Yan Zhao*
Department of Chemistry, Iowa State University, Ames, Iowa 50011-3111,
United States
zhaoy@iastate.edu
Received December 12, 2011
ABSTRACT
Cross-linking of the reverse micelles (RMs) of a triallylammonium surfactant afforded organic nanoparticles with introverted cationic groups.
The cross-linked reverse micelles catalyzed size-selective biphasic reaction between sodium azide and alkyl bromides. Size selectivity of up to 9:1
was obtained for alkyl bromides with similar structures. The selectivity was influenced strongly by the size of the water pool and proposed to
happen as a result of the “sieving” effect of the alkyl corona.
Enzymes carry out highly efficient and selective catalysis
in active sites tailored for their specific functions. The
active site not only has different polarity from the bulk
solvent but alsosize, shape, and functional groups essential
to the molecular recognition and catalysis of the enzyme.
Chemists have adopted a similar strategy and devoted
much effort to the development of nanoreactors for
enzyme-like catalysis.1 By encapsulating orthoester in an
anionic water-soluble nanoreactor, Bergman and Raymond
were able to achieve acid catalysis in basic solution.2 Rebek
and co-workers placed a carboxyl group inside a cavitand
for the regioselective ring-opening of epoxides.3 Ramamurthy
and Gibb were able to control the photoreactions of
ketones using water-soluble molecular capsules.4 Fujita
and colleagues demonstrated that, when packed in a metalÀ
organic nanocapsule, unreactive naphthalenes could
undergo regio- and stereoselective DielsÀAlder reactions.5
Badjic and co-workers developed cavitand “baskets” with
conformational gating to control reactivity.6 Warmuth
employed hemicarcerand to modulate the photochemical
and thermal reactions of reactive intermediates such as
nitrene and carbene.7
We recently reported the synthesis of interfacially cross-
linked reverse micelles (ICRMs) from cationic surfactants
1 and 2 (Scheme 1).8 These surfactants form RMs in a
chloroform/heptane mixture in the presence of a small
amount of water. The nanosized water pool in the middle
of the RM concentrates the water-soluble dithiothreitol
(1) (a) Fiedler, D.; Leung, D. H.; Bergman, R. G.; Raymond, K. N.
Acc. Chem. Res. 2004, 38, 349–358. (b) Vriezema, D. M.; Aragones,
M. C.; Elemans, J.; Cornelissen, J.; Rowan, A. E.; Nolte, R. J. M. Chem.
Rev. 2005, 105, 1445–1489. (c) Rebek, J. Angew. Chem., Int. Ed. 2005, 44,
2068–2078. (d) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew.
Chem., Int. Ed. 2009, 48, 3418–3438.
(5) Murase, T.; Horiuchi, S.; Fujita, M. J. Am. Chem. Soc. 2010, 132,
2866–2867.
(6) (a) Wang, B. Y.; Bao, X. G.; Yan, Z. Q.; Maslak, V.; Hadad,
C. M.; Badjic, J. D. J. Am. Chem. Soc. 2008, 130, 15127–15133. (b) Bao,
X. G.; Rieth, S.; Stojanovic, S.; Hadad, C. M.; Badjic, J. D. Angew.
Chem., Int. Ed. 2010, 49, 4816–4819. (c) Rieth, S.; Hermann, K.; Wang,
B. Y.; Badjic, J. D. Chem. Soc. Rev. 2011, 40, 1609–1622.
(7) (a) Warmuth, R.; Makowiec, S. J. Am. Chem. Soc. 2007, 129,
1233–1241. (b) Lu, Z.; Moss, R. A.; Sauers, R. R.; Warmuth, R. Org.
Lett. 2009, 11, 3866–3869.
(2) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Science 2007,
316, 85–88.
ꢀ
(3) Shenoy, S. R.; Pinacho Crisostomo, F. R.; Iwasawa, T.; Rebek, J.
J. Am. Chem. Soc. 2008, 130, 5658–5659.
(4) Gibb, C. L. D.; Sundaresan, A. K.; Ramamurthy, V.; Gibb, B. C.
J. Am. Chem. Soc. 2008, 130, 4069–4080.
(8) Zhang, S.; Zhao, Y. ACS Nano 2011, 5, 2637–2646.
r
10.1021/ol203319w
Published on Web 01/24/2012
2012 American Chemical Society

Table 1. Percent Yield of Benzyl Azide Obtained in the Biphasic
Reaction between NaN3 and Various Benzyl Bromides (RBr)a
Scheme 1. Preparation of the Interfacially Cross-Linked
Reverse Micelle (ICRM) and the Benzyl Bromides Used
in the Study
yield with
yield with
yield with
entry
RBr
ICRM (1) (%)
ICRM (2) (%)
no ICRM (%)
1
2
3
4
5
6
3
4
5
6
7
8
>95
>95
23
>95
>95
38
8
9
0
0
3
2
82
89
74
74
>95
>95
a The reactions were carried out with 0.1 mmol of RBr, 0.3 mmol of
NaN3, and 20 mol % of the cross-linkable surfactant in the ICRMs in a
mixture of water (1 mL) and CDCl3 (1 mL) under vigorous stirring for
24 h. W0 = [H2O]/[surfactant] = 15. The ICRMs were prepared according
to a previously published procedure.8 The reaction yields were determined
by 1H NMR spectroscopy.
similar to 5 electronically but less sterically demanding,
gave 80À90% yield under the same conditions, steric
interactions were mainly responsible for the selectivity.
The ICRMs overall were amazingly “permeable”, as bro-
mide 7 with two tert-butyl groups and a dodecyloxy chain
gave over 70% yield. A single dodecyloxy chain was even
less of a problem;bromide 8 reacted quantitatively in the
biphasic reaction.
(DTT) near the headgroups of the surfactants. The high
local concentrations of alkene and thiol near the waterÀ
surfactant interface facilitate the already efficient thiolÀ
ene radical chain reaction,9 enabling the dynamic self-
assembled RMs to be captured in the original size by
covalent bonds.10
We reasoned that the introverted ammonium groups of
the ICRMs should make them potential phase-transfer
catalysts (PTCs). Unlike conventional PTCs, however, the
ICRMs have the phase-transferred anions located in or
near the nanosized internal cavity. Because both the sur-
face alkyl density and the size of the water pool can be
tuned easily in our synthesis, we hypothesized that only
substrates small enough to access the nucleophilic anions
would be able to react. The size selectivity is akin to the
“reactive sieving” displayed by tRNA synthetase11 and
synthetic foldamers.12
To test the hypothesis, we examined the biphasic reac-
tion between sodium azide and alkyl bromides (3À8) in a
water/chloroform mixture. As shown in Table 1, in the
absence of the ICRMs, most bromides were unreactive
under our experimental conditions. The small bromides
(3 and 4) had somewhat higher background reactivity, prob-
ably because their higher water-solubility allowed them to
enter the azide-containing aqueous phase more easily. In
the presence of both ICRMs, the small azides (3 and 4)
reacted quantitatively. A bulky bromide (5), on the other
hand, was only converted in 23 and 38% yield, respec-
tively, by the two ICRMs. Because compound 6, which is
A surprising result in the phase-transfer catalysis is
the similar activity of ICRM (1) and ICRM (2) for the
majority of the substrates. Although an alkyl bromide may
not have to get into the hydrophilic core of the ICRM to
react with the azide, it has to penetrate the alkyl corona to
a certain degree to access the nucleophiles in or near the
ICRM core. For this reason, one would expect that the
double-tailed surfactant should afford ICRMs with a
stronger “sieving” effect. Nevertheless, the two ICRM(s)
gave essentially indistinguishable results for the majority
of the bromides. For the bulkiest bromide (5), the ICRM
derived from the double-tailed surfactant actually was
more active, giving 1.7 times as much product as that by
the single-tailed one (Table 1, entry 3).
The aboveresult may be explainedby our previous study
of the ICRMs.8 Normally, one would anticipate an alkyl-
covered organic nanoparticle to be fully soluble in non-
polar solvents. The ICRMs prepared from the single-
tailed surfactant, however, have gaps in between the alkyl
chains due to the bulkiness of the headgroup and the
geometry of a spherical particle, i.e., more space at the
periphery than at the center. These features make ICRM
(1) extremely prone to interparticle aggregation even in
nonpolar solvents such as chloroform. As aggregation
occurs, the alkyl chains on the ICRM surface interdigitate,
not only promoting the van der Waals interactions among
the alkyl chains but also expelling solvent molecules
trapped in between the alkyl chains into the bulk;an
entropically favorable process. At the particles get closer,
the (long-range) electrostatic interactions from the charged
micellar cores also become significant. These interactions
are sufficiently strong in ICRM (1) that it is completely
insoluble in highly nonpolar solvents such as hexane.
(9) (a) Hoyle, C. E.; Lee, T. Y.; Roper, T. J. Polym. Sci., Part A:
Polym. Chem. 2004, 42, 5301–5338. (b) Dondoni, A. Angew. Chem., Int.
Ed. 2008, 47, 8995–8997.
(10) Only one other example was reported to capture RMs in the
original size. See: (a) Jung, H. M.; Price, K. E.; McQuade, D. T. J. Am.
Chem. Soc. 2003, 125, 5351–5355. (b) Price, K. E.; McQuade, D. T.
Chem. Commun. 2005, 1714–1716.
(11) Fukunaga, R.; Fukai, S.; Ishitani, R.; Nureki, O.; Yokoyama, S.
J. Biol. Chem. 2004, 279, 8396–8402.
(12) (a) Smaldone, R. A.; Moore, J. S. J. Am. Chem. Soc. 2007, 129,
5444–5450. (b) Smaldone, R. A.; Moore, J. S. Chem.;Eur. J. 2008, 14,
2650–2657.
Org. Lett., Vol. 14, No. 3, 2012
785

Figure 1. Competitive azidation of 5 and 6 at W0 = 5. The 1H
NMR spectra from top to bottom are for 6, 5, and the 1:1:1
mixture of 5/6/NaN3 at room temperature after 24 h.
Figure 2. (a) Competitive azidation of 5 and 6 catalyzed by
ICRM(1) with different W0 = [H2O]/[Surfactant]. The figure is
color-coded according to the molecules, with the black bars
showing the conversion of 5. The numbers over the bars are the
percent yields of the corresponding azides. (b) Comparison of
the substrate-selectivity in the competitive azidation under
different conditions. CTAB = cetyltrimethylammonium bromide.
TBAB = tetrabutylammonium bromide.
In our previous study, ICRM (2) was found to be free of
aggregation because the higher density of the alkyl chains on
the surface makes alkyl-interdigitation sterically impossible.
Quite likely, it was the aggregation of ICRM (1) that
made it much less “permeable” to bromide 5. Although
this PTC has larger gaps in between the surface alkyl
groups, the gaps are closed by the alkyl interdigitation.
Conceivably, the larger the gap, the higher is the driving
force for the interparticle aggregation and the “tighter” the
alkyl shell, making the ICRM of the single-tailed surfac-
tant more discriminating.
The reactions in Table 1 were performed with an excess
of azide. To better characterize the size selectivity, we
carried out competitive azidation with two bromides pre-
sent in the samesolution and only 1 equiv ofazide. Figure 1
shows the 1H NMR spectra of bromides 5, 6, and a 1:1:1
mixture of 5/6/NaN3 after 24 h at room temperature.
According to the integration, while ca. 70% of 6 was
converted to the azide, barely 10% of 5 reacted.
The size of the hydrophilic core can be tuned by the
amount of water used in the RM formulation.13 Figure
2a shows the yields in the competitive azidation of
Figure 3. Competitive azidation of (a) 5 and 7 and (b) 5 and 8
catalyzed by ICRM (1) with different W0. The figure is color-
coded according to the molecules, with the black bars showing
the conversion of 5.
bromides 5 and 6. The ICRMs prepared at W0
=
[H2O]/[surfactant] = 15 was clearly less selective than
those at lower W0. Note that it was the bulkier substrate
(5) that was affected more by the size of the water pool,
not the smaller one. As shown by Figure 2b, the highest
selectivity for 6/5 was ∼7:1, which was significantly
higher than that displayed by conventional phase-trans-
fer catalysts such as CTAB (cetyltrimethylammonium
bromide) or TBAB (tetrabutylammonium bromide).
Similar competitive azidation was performed for 7/5 and
8/5 (Figure 3). When another bulky bromide (7) was
present, 5 became more reactive, affording 14À23% azida-
tion under the same reaction conditions. Meantime, the
total yields decreased, ranging from 51À68% for 7/5 to
74À92% for 6/5 and 79À91% for 8/5. Clearly, bulky
substrates in general have difficulty approaching the
entrapped azide ions.
For all three pairs, the selectivity was the lowest at the
highest W0. In general, the bromides were more reactive at
higher W0 but the effect was more pronounced for the
bulkier bromide (5) than for the less bulky ones (6 and 8).
An increase of W0 from 5 to 15, for example, doubled the
product yield for 5 while 6 and 8 were hardly affected
(Figures 2a and 3b).
The ICRMs could be affected by the water/surfactant
ratio in several ways. As W0 decreases, the water pool
inside the ICRM becomes smaller, the number of the
(cross-linked) surfactants in the ICRM decreases, and
the curvature of the nanoparticle becomes larger. The first
two factors affect the number of azide ions per ICRM
and the last the aggregation of the ICRMs. For an SN2
reaction in a homogeneous solution, the reaction rate
depends on the inherent reactivity of the reactants, their con-
centrations, and temperature. For a microphase-separated
(13) Pileni, M. P. Structure and Reactivity in Reverse Micelles;
Elsevier: Amsterdam, 1989.
786
Org. Lett., Vol. 14, No. 3, 2012

system such as what we have, many factors could affect the
reaction rates including the local concentration ofthe azide
ions and how the bromide approaches the azide.
as a result, making it more difficult for the bulky substrate
to react. The smaller bromides were not affected signifi-
cantly by the W0 possibly because they could penetrate the
alkyl corona fairly easily at all W0.
Although the total concentration of azide ions was the
same under in all the experiments, the local concentration
of azide could be different at different W0. Because all the
bromides in our study became more reactive at higher W0,
it is possible that the local concentration of azide ions
might have increased with the W0. The higher local concen-
tration of azide, however, could not explain why the steri-
cally most demanding bromide (5) always benefited most
from the increase in the water-to-surfactant ratio. Since the
smaller or slimmer bromides should be better able to react
with the entrapped azide (as evident from their higher yields),
they should benefit more from an increase in the effective
concentration of azide ions if no other factors are involved.
Our current postulation is that the alkyl density outside
the ICRM core is the main determining factor for the size
selectivity. For the single-tailed ICRM, as discussed ear-
lier, interparticle aggregation “tightens” the alkyl shell and
increases the selectivity. A lower W0 increases the curva-
ture of the ICRM core, widens the gaps in between the
alkyl chains, and leads to stronger interparticle aggrega-
tion. The alkyl density around micellar core would increase
ICRMs are prepared in a one-step synthesis from the
cross-linkable surfactant. Their facile synthesis represents
a tremendous benefit when they are compared with other
functionalized nanoreactors such as dendrimers. Selectiv-
ity up to 9:1 has been achieved for similar bromides such as
5 and 8. Further derivatization of these organic coreÀshell
nanoparticles should provide additional functions and
convert them into useful biomimetic nanoreactors.
Acknowledgment. We thank the U.S. Department of
Energy Office of Basic Energy Sciences (Grant No.
DE-SC0002142) for supporting the research.
Supporting Information Available. Experimental pro-
cedures, additional 1H NMR spectra for the competitive
azidation, and the NMR data of the newly synthesized
compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 3, 2012
787