A combinatorially-derived structural phase diagram for 42 zwitterionic geminis [20]

Full text of DOI:10.1021/ja003779l

5614
J. Am. Chem. Soc. 2001, 123, 5614-5615
A Combinatorially-Derived Structural Phase
Diagram for 42 Zwitterionic Geminis
Fredric M. Menger* and Andrey V. Peresypkin
Department of Chemistry, Emory UniVersity
1
515 Pierce DriVe, Atlanta, Georgia 30322
ReceiVed October 25, 2000
The phase diagram and the attendant Gibbs phase rule are
components of most introductory physical chemistry courses. Thus
everyone has had at least a passing acquaintance with those
triangular diagrams in which phases are recorded as a function
of concentrations. Temperature and pressure are often included
as additional variables. We have now examined phase properties
for large numbers (dozens!) of related compounds. As recognized
1
by many others previously, only by knowing how phases respond
to structure can one obtain a molecular understanding of colloidal
behavior. Accordingly, we have synthesized, purified, and
examined (by electron and light microscopy, dynamic light
scattering, and calorimetry) a group of 42 zwitterionic gemini
surfactants. The resulting information was then used to construct
a “structural phase diagram”. As will be seen, phase behavior
that might otherwise have escaped attention emerges from this
combinatorial approach to colloid chemistry.2
Figure 1. Structural Phase Diagram of 42 zwitterionic geminis. Numbers
Our work focused on gemini surfactants shown below. Varying
in the right-top corner of the circles represent gel-transition temperatures
of the vesicles (T ,°C).
m
the chain lengths A and B allowed the acquisition of compounds
where A and B are short-short, short-long, long-short, or long-
long, respectively.3 Geminis are attractive candidates for a
structural phase diagram, relative to conventional surfactants,
owing to their structural versatility. 4
Figure 1 provides a structural phase diagram for the 42 geminis
in which four main zones (identified as gels, micelles, coacervates,
and vesicles) are visible. All phases were made by hydrating the
solid gemini (5-50 mg/mL) without sonication for 1 h at 25 °C.
The gels, micelles, and coacervates were found to be stable for
several months. Gels are formed from highly asymmetric geminis
in the upper left and lower right of the diagram (e.g. A22,B6) at
Figure 2. Cryo-HRSEM image of 1mM solution of A14,B8 showing
network of interconnected strands of vesicular particles. Bar)100 nm.
Similar images were obtained for A18,B8.
the two chains. The importance of hydrophobicity is seen from
the following comparisons based on SEM, light microscopy, and
dynamic light scattering (DLS): When both chains are short (e.g.
A8,B8), micelles predominate (DLS diameter <10 nm). When
both chains are sufficiently long (e.g. A12,B16), the geminis
organize into vesicular systems of 30 nm to 100 µm in diameter
(circles in Figure 1). Several gel-transition temperatures are given
in Figure 1. As with phospholipids,1b more symmetrical geminis
(e.g. A14,B16, Tm ) 38 °C) have higher Tm values than
asymmetrical geminis (e.g. A18,B12, Tm ) 21 °C). More
interestingly, when the chains are intermediate in length (e.g.
1-4 wt % concentrations. Cryo-high-resolution scanning electron
microscope images (cryo-HRSEM) as in Figure 2 prove that the
gel consists of a network of interconnected vesicle-sized particles.
Although gels from densely packed vesicles have been previously
5
reported, this appears to be the first instance of water being
rigidified by “strings” of monodisperse vesicle-like aggregates
of 20-30 nm in diameter. Uncovering such a new type of soft
material demonstrates the value of the combinatorial approach.
Inspection of the structural phase diagram shows that morphol-
ogy depends on two parameters: (a) the “hydrophobicity” or total
number of carbons in the two chains and (b) the asymmetry of
6
A8,B10), SEM-visible coacervates are formed (blue crosses).
These are comprised of weak bilayers that tend to interconnect
with one another, resulting in a rather esoteric type of colloid
that has been likened to a “sponge” (Figure 3). Physically, the
coacervates in the region between micelles and vesicles appear
as liquids that are immiscible with water (despite themselves being
82-86 wt % water) owing to the sponge framework that occupies
the entire phase volume.
(
1) (a) Oda, R.; Huc, I.; Candau, S. J. Chem. Commun. 1997, 2105-2106.
(
b) Huang, C. Klin. Wochenschrift, 1990, 68, 149-165 and references therein.
7
See any current masthead page for ordering information and Web access
instructions.
(
2) For an approach to combinatorial catalysis, see: Menger, F. M.; Eliseev,
A. V.; Migulin, V. A. J. Org. Chem. 1995, 60, 6666-6667.
(
3) The 2-steps synthesis was reported. See: Peresypkin, A. V.; Menger,
F. M. Org. Lett. 1999, 1, 1347-1350. Variations were made possible by
reacting the products of alcohols and 2-chloro-1,3,2-dioxaphospholane-2-oxide
with different tertiary amines.
(
4) Menger, F. M.; Keiper, J. S. Angew. Chem., Int. Ed. 2000, 39, 1906-
920 review the subject of gemini surfactants.
5) Gradzielski, M.; Bergmeier, M.; M u¨ ller, M.; Hoffmann, H. J. Phys.
Chem. B 1997, 101, 1719-1722 and references therein.
(6) Menger, F. M.; Peresypkin, A. V.; Caran, K. L.; Apkarian, R. P.
Langmuir 2000, 16, 9113-9116.
(7) Strey, R.; Jahn, W.; Porte, G.; Bassereau, P. Langmuir 1990, 6, 1635-
1639.
1
(
1
0.1021/ja003779l CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/16/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 23, 2001 5615
Figure 3. Sponge morphology of coacervates. (Adapted from ref 7).
Figure 5. 1H NMR spectra of A8,B10 (top) and A10,B8 (bottom) in
D O. (400 MHz).
2
Figure 4. Proposed models of A8,B10 (left) and A10,B8 (right), in which
central (O-CH
2
-CH -N) bond is gauche and the chains are anti.
2
But hydrophobicity cannot be the sole factor controlling phases
because two geminis of identical hydrophobicity, A8,B10 and
A10,B8, for example, form a coacervate and micelles, respec-
tively. This difference, which persists throughout the 0-80 °C
temperature range, is attributable to chain asymmetry. Figure 4
shows space-filling models of the two geminis assuming a gauche
Figure 6. (A) Synergistic effect on coacervate volume upon addition of
A10,B8 to a coacervate from 100 mg of A8,B10. (B) Addition of A8,B10
to itself as a control.
conformation at the central CH
2
2
-CH bond (a conformation that
A plot of coacervate volume upon addition of 0-100 mg of
A10,B8 to 100 mg of A8,B10 in 20 mL of water is given in
Figure 6. One sees that micelle-forming A10,B8 enhances the
coacervate volume from 0.6 mL to 4.5 mL (a remarkable
synergistic effect considering that 200 mg of pure A8,B10
brings the two chains, as well as the oppositely charged ionic
8
groups, into proximity). It is seen that the lengths of the two
chains in A8,B10 are identical, whereas A10,B8 has a phosphorus
chain about 4 carbons longer than its ammonium chain. A10,B8
asymmetry would be expected to destabilize possible bilayer
formation (already weakened from its shorter-than-optimal chain
lengths) and to favor organization into micelles. A8,B10, on the
other hand, can pack smoothly into bilayers.
Hydrophobicity and symmetry effects work in concert with
A11,B8; A11,B9; and A11,B10 to produce a highly structurally
sensitive phase sequence of micelles, coacervates, and vesicles,
respectively.
9
produces a volume of 1.2 mL). Only when the two geminis are
equimolar does the system suddenly revert to micelles (as proven
by DLS). The most reasonable explanation is that A10,B8
incorporates itself into the sponge structure of A8,B10 where it
increases the bilayer curvature and thus the pore size, so that the
coacervate swells. With sufficient A10,B8, the sponge can no
longer sustain the distortions, and the structure breaks down into
high-curvature micelles characteristic of pure A10,B8.
2
NMR spectra of the colloids in D O (Figure 5) are consistent
with the proposed model. Thus, micellar A10,B8, but not the
coacervate A8,B10, has discernible terminal methyl signals
Each point (i.e. each compound) on the structural phase diagram
warrants a detailed physicochemical examination. This is clearly
impractical for a collection of 42 surfactants. The value of a
structural phase diagram, however, comes from its providing a
sweeping view of phase behavior as a function of structure.
Generalizations at the molecular level are thereby possible.
(
arrows). The same is true for micellar A11,B8 and coacervate
A8,B11.
Having discovered, via a combinatorial-style synthesis, that two
closely related geminis, A8,B10 and A10,B8, create totally
different phases, we asked a quite natural question: What happens
when the two geminis are mixed? To address this question, it is
necessary to describe the procedure for preparing the coacer-
vate: A sample of 100 mg of A8,B10 was added to 20 mL of
water. Instantly, coacervate droplets fell out of solution and
coalesced into a bottom layer of 0.6 mL. When 200 mg of gemini
was added to the 20 mL, a coacervate layer of 1.2 mL was formed,
but analyses showed that the two coacervate layers had the same
gemini concentration and >99% of the added gemini. The
coacervate volume, incidentally, is independent of the original
amount of water.6
Acknowledgment. We thank Dr. R. Apkarian and Dr. K. Caran from
the Integrated Microscopy and Microanalytical Facility for assistance with
electron microscopy. This work was supported by the National Science
Foundation.
Supporting Information Available: Experimental details, charac-
terization of compounds, and additional data (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
JA003779L
(8) The gauche conformation is the preferred conformation of the choline
group of natural phospholipids, see: Hauser, H.; Guyer, W.; Spiess, M.;
Pascher, I.; Sundell, S. J. Mol. Biol. 1980, 137, 265-282 and references
therein.
(9) For a recent theoretical analysis of synergistic effects, see: Bergstr o¨ m,
M.; Eriksson, J. C. Langmuir 2000, 16, 7173-7181.