Contiguous versus segmented hydrophobicity in micellar systems

Article abstract of DOI:10.1021/ja040105s

This paper addresses a question not yet posed systematically in surfactant chemistry: How do the colloidal properties of surfactants respond to insertion of non-hydrocarbon functionalities (i.e., ester groups) within chains that are normally entirely hydrocarbon? In answering this question, two classes of such chain-modified surfactants were discovered. One class forms only small aggregates with noncooperative self-assembly, low foaming, high areas of occupancy at the air/water interface, and weak solid-adsorption and solubilization properties. The other class is much more normal with regard to these properties and, in fact, can even exceed conventional surfactants in mesitylene solubilization. Differences between the two categories of chain-modified surfactants originate from the degree of segmentation of the hydrocarbon and, in particular, upon the location of the longest segment. Segmented hydrophobicity, having in principle a "hydrophobic potential" similar to that of a contiguous hydrophobicity of equal length, can induce aggregation but, concurrently, alters the mode of assembly into films and micelles.

Full text of DOI:10.1021/ja040105s

Published on Web 11/11/2004
Contiguous versus Segmented Hydrophobicity in Micellar
Systems
Fredric M. Menger* and Ashley L. Galloway
Contribution from the Department of Chemistry, Emory UniVersity, Atlanta, Georgia 30322
Received April 12, 2004; E-mail: menger@emory.edu
Abstract: This paper addresses a question not yet posed systematically in surfactant chemistry: How do
the colloidal properties of surfactants respond to insertion of non-hydrocarbon functionalities (i.e., ester
groups) within chains that are normally entirely hydrocarbon? In answering this question, two classes of
such chain-modified surfactants were discovered. One class forms only small aggregates with noncoopera-
tive self-assembly, low foaming, high areas of occupancy at the air/water interface, and weak solid-adsorption
and solubilization properties. The other class is much more normal with regard to these properties and, in
fact, can even exceed conventional surfactants in mesitylene solubilization. Differences between the two
categories of chain-modified surfactants originate from the degree of segmentation of the hydrocarbon
and, in particular, upon the location of the longest segment. Segmented hydrophobicity, having in principle
a “hydrophobic potential” similar to that of a contiguous hydrophobicity of equal length, can induce
aggregation but, concurrently, alters the mode of assembly into films and micelles.
Table 1. Structure of Ester-Modified Surfactants A-F and Two
Conventional Surfactants, DTAB and TTAB
Introduction
Occasionally in the literature one finds examples of people
addressing an important question in colloid chemistry: How
does the presence of non-hydrocarbon functional groups within
a surfactant’s chain affect the self-assembly process and the
properties of the resulting aggregates? One of the first and most
interesting papers confronting this issue was penned by Muller
and Birkhahn in 1967.1 These chemists deduced from solvent-
sensitive 19F NMR signals that the terminal -CF3 group of
micellar CF3(CH2)nCOONa experiences a polarity midway
between water and hydrocarbon. We ourselves, while examining
monomolecular films of 10-hydroxystearic acid, found that the
hydroxyl and polar headgroup maintain contact with an aqueous
subphase while the intervening chains form “loops” above it.2
Despite such work, there has been to our knowledge no
systematic study of what might be termed “interrupted”
hydrophobicity. How, for example, will the relative location of
two semipolar groups within a hydrocarbon chain affect its
propensity to assemble? In this manuscript, we now describe
synthetic pathways to several cationic surfactants bearing ester-
modified chains (Table 1). More pertinently, we discuss the
physical chemistry of fundamentally new surfactant systems that
organic synthesis has now placed at our disposal.3
self-assemble into a micelle that can then solubilize or disperse
insolubles in water. Solubilization/dispersal is by far the most
important property of surfactants;4-6 huge industries rely on the
concept. Now most organic solubilizates reside in the hydro-
phobic portion of the micelles (sometimes near the micelle
surface, sometimes in the interior). Yet a purely hydrocarbon
The decision to incorporate ester groups into the surfactant
chains needs explanation. Conventional surfactants consist of
two sections: a hydrophilic headgroup and a hydrophobic tail.
The headgroup serves one main purpose: to promote solubility
of the hydrophobic tail in water. Once this happens, the tails
(4) Stearns, R. S.; Oppenheimer, H.; Simon, E.; Harkins, W. D. J. Chem. Phys.
(1) Muller, N.; Birkhahn, R. H. J. Phys. Chem. 1967, 71, 957.
(2) Menger, F. M.; Richardson, S. D.; Wood, M. G., Jr.; Sherrod, M. J.
Langmuir 1989, 5, 833.
(3) For an excellent review of new surfactants, consult: Chevalier, Y. Curr.
Opin. Colloid Interface Sci. 2002, 7, 3.
1947, 15, 496.
(5) Uchiyama, H.; Christian, S. D.; Scamehorn, J. F.; Abe, M.; Ogino, K.
Langmuir 1991, 7, 95.
(6) Almgren, M.; Grieser, F.; Thomas, J. K. J. Am. Chem. Soc. 1979, 101,
279.
9
10.1021/ja040105s CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 15883-15889
15883

A R T I C L E S
Menger and Galloway
Scheme 1. Synthetic Route to Surfactants A-D
environment would not be expected to have a particularly great
capacity for dissolving modestly polar organics. Hexane, for
example, is usually a poor solvent for organics with even a single
polar group. If, however, a surfactant chain could be made
somewhat less hydrocarbon-like, but (importantly!) not to such
an extent as to prevent self-assembly, then one might obtain a
much more potent solubilizing entity. In the same way that ethyl
acetate is a better solvent than hexane, the presence of two ester
groups in a surfactant (or 100 ester groups per 50-surfactant
micelle) might enhance the partitioning of organic guests from
the bulk water phase into the micelles. Such was the reasoning
behind the design of the compounds in Table 1.
No more than two ester groups were inserted into the chains
(Table 1) as a conservative compromise between increasing the
micelle’s interior polarity as much as possible while not
preventing the micellization process. It was felt at the outset
that the esters would not prevent micellization if the surfactants
were also provided with sufficient hydrocarbon. The poor
miscibility of ethyl acetate in water supports this expectation,
as does the Handbook of Surfactant Analysis which refers to
esters as lipophilic.7 On a more quantitative footing, the Hansch
π-parameters (based on octanol/water partition coefficients) have
the following values: Et2O ) 7.8; EtOAc ) 9.1; CHCl3 ) 9.5;
and MeOH ) 14.2.8 We concluded from such data that micelles
should be able to tolerate a modest number of ester groups and
that, when they do, the properties of the micelles should likely
be altered.
There was a second and more fundamental reason for ex-
amining ester-modified surfactants. This had to do with the
concept of hydrophobicity, a concept that has been well-
developed over the years owing to the efforts of (among others)
Hildebrand,9 Harkins,10 Tanford,11 Kauzmann,12 Frank,13 Ne´m-
ethy,14 Scheraga,15 Ben-Haim,16 Rekker,17 and Engberts.18 In
brief, contact between two hydrocarbon chains in aqueous
systems releases “structured” water (an entropically favorable
process), and, as a consequence, the hydrocarbons experience
a stronger interaction energy in water than would be expected
solely from van der Waals forces in free space. This is the
hydrophobic effect. This is why surfactants form micelles in
water, and lipids assemble into membranes.
chain sulfates and ammonium salts on the neutral hydrolysis of
1-benzoyl-1,2,4-triazole. The cosolutes bind to the triazole
substrate and inhibit its hydrolysis. The key point is that the
first two or three CH2 groups near the ionic groups of the
cosolutes do not seem to contribute substantially to the cosolute/
substrate binding. It is as if the sulfate and ammonium groups
shield the proximal methylenes and impair their availability for
hydrophobic association. Of course, our ester group is less polar
than the ionic groups, and its shielding would be expected to
be less extensive.
In summary, this work was carried out to alter and, possibly,
improve surfactant properties. (Certainly, the presence of ester
groups would promote surfactant biodegradability!) Also, on a
more basic level, we hoped to learn something about “inter-
rupted” hydrophobic association.
Synthesis
As seen in Table 1, we studied five new diester-loaded
surfactants (A-E) plus a monoester (F) and, for comparison
purposes, two conventional surfactants, DTAB and TTAB. The
synthetic route to A-D is shown in Scheme 1, the route to E
is shown in Scheme 2, and the route to F is shown in Scheme
3. As detailed in the Experimental Section, all new surfactants
were chromatographically pure and characterized by 1H and 13
C
Past work notwithstanding, there exist substantial gaps in our
understanding of hydrophobic association. In particular, it is
not known if hydrophobic association among discrete hydro-
phobic regions is additive. Will, for example, a chain that is 12
methylenes in length provide a greater hydrophobic driving force
than a chain of equal length but separated into three sets of
four methylenes? Asked in another way consistent with this
paper’s title, “How do contiguous and segmented associations
compare?”
NMR, HRMS, and EA. Table 1 also lists the total number of
methylenes in the eight compounds as a qualitative measure of
their “hydrophobic potential”.
Results and Discussion
Our physical-chemical data are largely summarized in Table
2. Let us begin with the critical micelle concentration (cmc)
values listed in column 2 of this table. Cmc’s, the most
representative micelle descriptor, were obtained from the breaks
in the surface tension versus concentration plots, two examples
of which are shown in Figure 1. As a rough rule of thumb, cmc
values decrease by a factor of 4 as the surfactant chain increases
by two carbons.21 In this light, the similarity in cmc values for
F and TTAB (both with 13 methylenes and with cmc’s of 4.6
and 3.3 mM, respectively) would suggest only a slight effect
Engberts and co-workers have carried out kinetic studies
relevant to our concerns.19,20 They studied the effect of short-
(7) Hummel, D. O. Handbook of Surfactant Analysis; Wiley: Chichester, 2000.
(8) Leo, A.; Hansch, C.; Elkins, D. Chem. ReV. 1971, 71, 525.
(9) Hildebrand, J. H. J. Phys. Chem. 1968, 72, 1841.
(10) Harkins, W. D.; Mattoon, R. W.; Corrin, M. L.; Stearns, R. S. J. Chem.
Phys. 1945, 13, 534.
(11) Tanford, C. The Hydrophobic Effect: Formation of Micelles and Biological
Membranes; Wiley: New York, 1980.
(12) Kauzmann, W. AdV. Protein Chem. 1959, 14, 1.
(18) Blokzijl, W.; Engberts, J. B. F. N. Angew. Chem., Int. Ed. Engl. 1993, 32,
1545.
(13) Frank, H. S.; Evans, M. W. J. Chem. Phys. 1945, 13, 507.
(14) Ne´methy, G. Angew. Chem., Int. Ed. Engl. 1967, 6, 195.
(15) Ne´methy, G.; Scheraga, H. A. J. Chem. Phys. 1962, 36, 3382.
(16) Ben-Haim, A. Hydrophobic Interaction; Plenum: New York, 1980.
(17) Rekker, R. F. The Hydrophobic Constant; Elsevier: Amsterdam, 1977.
(19) Noordman, W. H.; Blokzijl, W.; Engberts, J. B. F. N.; Blandamer, M. J. J.
Org. Chem. 1993, 58, 7111.
(20) Hol, P.; Streefland, L.; Blandamer, M. J.; Engberts, J. B. F. N. J. Chem.
Soc., Perkin Trans. 2 1997, 485.
9
15884 J. AM. CHEM. SOC. VOL. 126, NO. 48, 2004

Contiguous versus Segmented Hydrophobicity
A R T I C L E S
Scheme 2. Synthetic Route to Surfactant E
aggregation number (i.e., 5-10) of loosely assembled molecules
(only marginally deserving of the name “micelle”).
The surface tension data such as in Figure 1 show that, in
fact, morphology differences among the surfactants are playing
an important role. Only C, D, and F display surface tension
versus concentration plots that level off sharply into horizontal
lines as happens when there is a cooperative assembly into large
micelles. Surfactants A, B, and E, on the other hand, show a
steady post-cmc decline in surface tension as would be expected
from a less precipitous assembly into small aggregates. Dynamic
light scattering, using a 10 mW laser, confirms the above
conclusions: micelles of C, D, and F in 0.1 M NaCl have 3-4
nm hydrodynamic diameters, whereas A, B, and E aggregates
fall below the 3 nm resolution of the instrument. As a
consequence, the properties of the A, B, E set are not easily
compared to those of the C, D, F set. For example, D (with 18
CH2’s) might have been expected to have a much lower cmc
than A (with only 12 CH2’s), yet the cmc’s are similar.
Additional physical-chemical data, to be revealed presently,
consistently reinforce this same dichotomy among the two sets
of surfactants.
Scheme 3. Synthetic Route to Surfactant F
The presence of two sets of micelle morphologies complicates
answering the interesting question: “Is the ester group hydro-
phobic or hydrophilic?” With surfactant A, B, and E, the ester
groups have a rather minor effect upon the cmc. For example,
E, with a total of 15 CH2’s, has a cmc of 1.0 mM, which is
only slightly larger than the cmc of 0.9 mM for the correspond-
ing surfactant, cetyltrimethylammonium bromide, with 15 CH2’s
and no esters.22 Apparently, when a micelle is small, the ester
linkages play little role in the micellization. Association among
relatively short hydrocarbon segments presumably forms “wet”
aggregates in which ester group hydration, such as it is, is not
impaired. Thus, the esters do not perturb the micelle-forming
equilibrium.
Table 2. Colloidal Properties of Ester-Modified Surfactants and
Two Conventional Surfactants
cmc,a
mM
area,
Å2
foam vol.,
mL
TMS,b
mM
mesitylene,b
mM
surfactant
The situation is different with the larger, more conventional
micelles formed in C, D, and F. Here, the ester groups definitely
impair micellization as reflected by the cmc values. Thus, D
with 18 CH2’s has a cmc that is 14-fold higher than its
hydrocarbon analogue where esters are absent.22 Undoubtedly,
ester polarity and/or packing constraints, imposed in part by
the esters’ s-trans O-CO linkage, contribute to the reduced
tendency for ester-endowed chains to assemble into compact
apolar units. When, however, the three surfactants do reach their
critical concentrations, the resulting micelles are rather normal.
Differences in packing constraints between the two groups
of surfactants, A/B/E and C/D/F, are seen from the area-per-
molecule occupied at the air/water interface (derived in the
standard way from pre-cmc tensiometric data and the Gibbs
adsorption isotherm).23 It is seen from Table 2, column 3, that
A, B, and (especially) E have inordinately large molecular areas.
For example, E has an area of roughly 334 Å2/molecule as
compared to 52 Å2/molecule for TTAB with its all-hydrocarbon
chain. The large areas for A/B/E reveal the reluctance of these
three surfactants to self-assemble into a monomolecular film
just as they are reluctant to self-assemble into large micelles.
It behooves us to explain why A/B/E form small and
presumably loose aggregates, whereas micelles from C/D/F are
A
B
C
D
E
F
2.4
2.5
2.2
1.4
1.0
4.6
13.3
3.3
81
123
75
75
334
54
0.01
0.48
2.5
3.2
0.04
3.5
0
1.7
5.3
38
93
0
27
13
36
0.026
1.2
5.2
0
5.4
0.51
7.9
DTAB
TTAB
64
52
1.9
3.3
a The cmc values for A, B, and E are approximate; see Figure 1A.
b Refers to the amount of TMS or mesitylene solubilized by the surfactant
(see text).
of F’s lone ester group on the tendency to assemble. Yet unfor-
tunately things are not quite so straightforward. Differences in
cmc among A-E are small (covering a range of only 1.0-2.5
mM) despite large differences in total chain-length (12 CH2’s
for A but 18 CH2’s for D). Part of the difficulty in explaining
these data relates to the fact that the cmc is a multi-faceted
thermodynamic parameter reflecting both the solvation proper-
ties of the monomer and the packing constraints within a
spherical micelle. Importantly, the cmc reveals only the
concentration at which a surfactant self-assembles, and not the
size and morphology of the micelles thus formed. As a con-
sequence, two surfactants with an identical cmc might have,
on one hand, a large conventional aggregation number (i.e.,
50-100), while, on the other hand, they might have a small
(22) Haak, J. R.; Van Os, N. M.; Rupert L. A. M. Physicochemical Properties
of Surfactants; Koninklijke/Shell-Laboratorium: Amsterdam, 1989.
(23) Seredyuk, V.; Alami, E.; Nyde´n, M.; Holmberg, K.; Peresypkin, A. V.;
Menger, F. M. Colloids Surf. 2002, 203, 245.
(21) Myers, D. Surfactant Science and Technology; VCH: New York, 1988; p
113.
9
J. AM. CHEM. SOC. VOL. 126, NO. 48, 2004 15885

A R T I C L E S
Menger and Galloway
Figure 1. Surface tensiometry plots of ester-modified surfactants A and F.
more normal in constitution. Inspection of the structures in Table
1 provides a rationale. Thus, the longest contiguous hydrocarbon
segments in A, B, and E are 6, 6, and 8 carbons in length,
respectively. By contrast, the values for C, D, and F are 10,
10, and 11, respectively. Moreover, these latter three segments
are all terminal. It seems critical, therefore, that a long
contiguous stretch of terminal carbons is necessary for formation
of conventional micelles (at least when ester groups constitute
the intervening functionalities). A comparison of C and E drives
this point home. C has a total of 16 CH2’s with a contiguous
terminal segment of 10 carbons. E has a similar total number
of CH2’s (15), but its longest segment is only 8 carbons, and
the segment is not terminal. The main conclusion is, therefore,
that segment length and location play an important role in
micelle formation, not so much in the cmc as in micelle
morphology. Stated in another way, a segmented hydrophobicity,
having in principle a “hydrophobic” potential similar to that of
a contiguous hydrophobicity of equal length, can certainly
induce aggregation. Yet at the same time, packing constraints
imparted by the ester spacers alter the mode of assembly into
films and micelles.
Having two categories of related surfactants on hand led us
to wonder about their differences in various colloidal properties
including foamability. Foamability experiments were carried out
by inverting and uprighting 10 times a 50 mL buret containing
5 mL of surfactant solution (at concentrations 3 times greater
than its cmc), and then measuring the volume of the resulting
foam.24 The data are recorded in Table 2, column 4. As seen,
A, B, and E are capable of sustaining only low volumes of foam
(0.01-0.48 mL) in contrast to C, D, and F and the two controls
(1.9-3.5 mL). Once again, the data support the notion that A,
B, and E, with their short terminal segments, have difficulty
aligning their chains at the air/water interface so as to stabilize
a film. An interesting comparison exists among A, B, and E:
Although all have low foamability, B is by far the best of the
three (0.48 mL as compared to 0.01 and 0.04 mL). An
explanation (tentative as we would need additional compounds
that at the moment have not been synthesized) is based on the
carbon-lengths of the initial and terminal segments of the three
surfactants: A (4, 6), B (6, 6), and E (8, 2). In A, the initial
segment is short, while in E the even more important terminal
segment is only two carbons and contributes little to the
hydrophobic overlap. B is an improvement over A and E in
both respects. If this is correct, then we can predict, for example,
that a (2, 8) would likely emulate B’s foamability.
Figure 2. HPLC traces of ester-modified surfactants A-F as well as
two conventional surfactants, DTAB and TTAB, on a reverse-phase
column.
Samples (1.0 mg/mL) were injected into an Alltech Surfactant/R
column containing a 7 µm polydivinylbenzene-based resin. A
solvent system was comprised of a 60% 10 mM HCl/40%
CH3CN mixture for the first 5 min followed by a 30% 10 mM
HCl/70% CH3CN mixture for the remainder of the experiment.
Nitrogen flow at ambient temperature (1.0 mL/min) was used
for all runs. As seen in Figure 2, surfactants A, B, and E have
shorter retention times on the hydrophobic surface than does
DTAB, whereas surfactants C, D, and F have longer retention
times. These results are consistent with the large adsorption areas
for A, B, and E in Table 2. As compared to surfactants C, D,
and F, surfactants A, B, and E bind rather ineffectively to each
other and to hydrophobic surfaces.
In the introduction of this paper, we mentioned the industrial
importance of solubilization by surfactants. It was, therefore,
natural to investigate the solubilization properties of our new
surfactants. A priori, we could not predict how the ester-
substituted chains would affect solubilization. For one thing, it
was not clear whether solubilization is promoted by solubilizate/
monomer interactions or by solubilizate/micelle adsorption (the
latter being favored at concentrations well above the cmc
values). At least two possibilities exist if micelles dominate the
solubilization mechanism: (a) The esters might increase the
polarity of the micelle interior and thus enhance solubilization
of moderately polar organics (much as ethyl acetate is a better
chromatographic solvent than hexane). (b) The s-transoid esters
might impair chain assembly into a micelle (and, as discussed,
we think this happens with A, B, and E), leading to loose
aggregates that are less effective solubilizers. Because the
solubilization properties could well be solubilizate-dependent,
we examined three of them: tetramethylsilane (TMS, a lipo-
philic compound with no dipole moment), mesitylene (1,3,5-
Reverse-phase HPLC was used to qualitatively measure the
solid-adsorption properties of the eight surfactants in Table 1.
(24) Patist, A.; Axelberd, T.; Shah, D. O. J. Colloid Interface Sci. 1998, 208,
259.
9
15886 J. AM. CHEM. SOC. VOL. 126, NO. 48, 2004

Contiguous versus Segmented Hydrophobicity
A R T I C L E S
plots for A, B, and E are nonlinear (perhaps reflecting aggregate
growth and enhanced solubilization at higher concentrations).
Among the generalizations we can state: (a) Total carbon
content is not always an accurate predictor of solubilization
power (B with 14 CH2’s is better than E with 15 CH2’s; C and
D are similar, although D has 2 more CH2’s). (b) The length of
the terminal chain is critical (C and D with 10 terminal carbons
are the best solubilizers; E, with 2 terminal carbons, is no better
than A, with 6 terminal carbons, although the former has more
total hydrocarbons. (c) All new surfactants are less powerful
solubilizers than TTAB, suggesting that the ester groups’
potential solubilizing ability for polar compounds is overridden
by unfavorable perturbations in the micellar structure.
In summary, we have examined a variety of colloidal prop-
erties of six new surfactants bearing one or more ester groups
within their chains. Systematic studies of chain-modified sur-
factants are rare,25 and even this brief entry into the field has
revealed useful structural relationships. Modification of micelle
chains (and lipid chains in bilayer membranes) clearly merits
further synthetic and physicochemical attention.
Figure 3. Disperse Red 19 solubilization in water as a function of surfactant
concentration. Six data points were used to construct each line, which is
presented as a visual guide. A and E were within experimental error of one
another, so only one line is shown.
trimethylbenzene, an aromatic), and Disperse Red 19 (a polar
dye, drawn below).
Experimental Section
Materials. Solvents used in this synthesis were reagent grade and
dried over 4 Å molecular sieves. Reagents were purchased from Aldrich
or Fluka and used without additional purification.
1
Methods. H NMR and 13C NMR spectra were taken on either a
Varian INOVA 400 mHz (100 mHz for 13C) or a Mercury 300 mHz
(75 mHz for 13C) instrument. Mass spectra experiments were completed
by the Emory University Mass Spectrometry Center, and Atlantic
Microlabs in Norcross, GA performed all elemental analyses. HPLC
experiments were completed on a Shimadzu LC-10AT VP instrument,
and the solubilization of Disperse Red 19 was measured on a Varian
DMS 300 UV-visible spectrophotometer at λmax ) 494 nm.
Excess amounts of TMS or mesitylene (both liquids) were
vortexed with 40 mM surfactant in D2O followed by 1H NMR
analysis of the solubilized material in the D2O after the layers
had separated. The results for TMS and mesitylene are given
in columns 5 and 6 of Table 2, respectively. Note that 40 mM
of surfactant lies appreciably above all of the surfactants’ cmc
values, so that we are dealing here primarily with micellar
behavior.
Syntheses. General Procedure for Compound 1.26,27 5-(Hexyl-
oxy)-5 Oxopentanoic Acid (1, n ) 5). A mixture of glutaric anhydride
(10.73 g, 94.0 mmol, 1 equiv) and hexanol (11.42 g, 111.8 mmol, 1.2
equiv) was heated and stirred in a 90 °C oil bath until the mixture
became homogeneous (1 h). After homogeneity was reached, the
reaction temperature was raised to 105 °C for 30 min. Once the reaction
had cooled, 50 mL of a saturated sodium bicarbonate solution was added
to dissolve the product. The excess hexanol in solution was extracted
with 3 × 50 mL portions of ether. Next, the pH of the aqueous layer
was lowered to 3 with dilute HCl, followed by an extraction of the
desired product with 3 × 50 mL portions of ether. The ether layer was
then dried over MgSO4, and the solvent was evaporated under reduced
pressure to yield a yellow oil. Yield: 19.77 g (97.4%). 1H NMR (300
mHz, CDCl3): δ 4.05 (t, 2H), 2.42 (t, 2H), 2.38 (t, 2H), 1.94 (m, 2H),
1.60 (m, 2H), 1.28 (m, 6H), 0.87 (t, 3H). 13C NMR (75 mHz, CDCl3):
δ 179.5, 173.2, 64.9, 33.4, 33.2, 31.6, 28.7, 25.7, 22.7, 20.0, 14.2. Mass
spectra: M + Li 223.1522 amu, found 223.1512 amu.
Surfactants A, B, and E are seen to be poor solubilizers of
both TMS and mesitylene. For example, A and B solubilize
1.7 and 5.3 mM mesitylene, respectively, as compared to 13
and 36 mM for DTAB and TTAB, respectively. Perhaps not
surprisingly, A and B, with their small, loose, and no doubt
“wet” aggregates, adsorb “little” nonpolar solubilizate. Remark-
ably, surfactant E (despite having a total of 15 CH2’s) failed to
solubilize any TMS or mesitylene at all. Once again, the terminal
tail, only two carbons long in E, reveals its importance.
On the other hand, surfactants C, D, and F are reasonably
good solubilizers. Thus, C and D solubilize 38 and 93 mM
mesitylene, respectively, as compared to only 13 and 36 mM
for DTAB and TTAB, respectively. Surfactant D solubilizes
10 times more TMS than does DTAB. These comparisons are,
of course, not totally fair because C and D possess a greater
number of carbons than do DTAB and TTAB. Probably, the
most straightforward comparison comes from F versus TTAB,
both of which have 13 CH2’s. In this case, F has only a slightly
lower solubilization power for TMS and mesitylene than does
TTAB, indicating that a single ester group neither substantially
assists nor detracts from micellar adsorption.
5-(Decyloxy)-5-oxopentanoic Acid (1, n ) 9). Yield ) 85%, white
1
solid. H NMR (400 mHz, d6-DMSO): δ 4.05 (t, 2H), 2.41 (m, 4H),
1.94 (m, 2H), 1.60 (m, 2H), 1.25 (m, 14H), 0.859 (t, 3H). 13C NMR
(100 mHz, d6-DMSO): δ 179.6, 173.6, 65.26, 33.71, 33.52, 32.38,
30.01 (2 C’s), 29.79, 29.74, 29.08, 26.40, 23.17, 20.33, 14.59.
General Procedure for Compounds 2, 7, 11.26,27 Hexyl 5-Chloro-
5-oxopentanoate (2, n ) 5). Compound 1 (n ) 5) (17.11 g, 79.2 mmol,
1 equiv) and 2 small drops of dry pyridine were stirred in a round-
bottom flask with a condenser and drying tube attached. An excess
(11.99 g, 100.1 mmol, 1.3 equiv) of thionyl chloride was added to the
Figure 3 plots the concentration of solubilized Disperse Red
19 (determined spetrophotometrically) versus surfactant con-
centration for the five diester compounds in Table 2. Only the
(25) Bhattacharya, S.; Dileep, P. V. Bioconjugate Chem. 2004, 15, 508.
(26) Wood, M. G. Synthesis and Study of New Fatty Acids and Phospholipids.
Dissertation, 1988.
(27) Cason, J. Org. Synth. 3, 169.
9
J. AM. CHEM. SOC. VOL. 126, NO. 48, 2004 15887

A R T I C L E S
Menger and Galloway
reaction, and the mixture was heated in an oil bath at 70 °C for 24 h.
Excess thionyl chloride was evaporated under reduced pressure to
produce 13.28 g (71% yield) of a reddish brown liquid. IR (neat, cm-1):
1804, 1733.
(100 mHz, d6-DMSO): δ 172.83, 172.78, 63.61, 60.72, 59.62, 33.40,
32.57, 29.06, 28.88, 28.54, 28.08, 25.51, 25.45, 25.39, 25.32, 24.31,
24.27, 14.09.
Dodecanoic Acid 3-Hydroxypropyl Ester (12). Yield ) 60.3%,
1
off-white solid. H NMR (400 mHz, CDCl3): δ 4.207 (t, 2H), 3.659
Decyl 5-Chloro-5-oxopentanoate (2, n ) 9). Reddish brown liquid,
94.6% yield (immediately carried to next reaction without additional
characterization).
(t, 2H), 2.282 (t, 2H), 1.84, (m, 2H) 1.588 (m, 2H), 1.229 (m, 16H),
0.848 (t, 3H). 13C NMR (100 mHz, CDCl3): δ 174.6, 61.3, 59.3, 34.5,
32.1, 31.9, 29.8 (2C), 29.6, 29.5, 29.4, 29.3, 25.2, 22.9, 14.3.
General Procedure for Compounds 4, 9, 13.30,31 4-Bromobutyl
Hexylpentanedioate (4, n ) 5, m ) 4). A solution of compound 3 (n
) 5, m ) 4) (11.51 g, 39.9 mmol, 1 equiv), pyridine (5.55 g, 70.2
mmol, 1.7 equiv), and 50 mL of acetonitrile was stirred at 0 °C. After
10 min, 21.79 g (51.62 mmol, 1.3 equiv) of solid triphenylphosphine
dibromide was added to the mixture. The reaction was removed from
ice, covered with a drying tube, and stirred at room temperature for 1
h. After 1 h, the solvents were stripped, and the crude product was
triturated in hexanes at reflux for 3 h. Hexanes were collected by
vacuum filtration, and the solvent was evaporated under reduced
pressure. The resulting product was filtered through a 2-in. pad of silica
and rinsed with 200 mL of a 10% ether/pentane solution. After the
solvents were stripped, 9.05 g (64.7%) of yellow oil emerged. 1H NMR
(400 mHz, CDCl3): δ 4.09 (t, 2H), 4.05 (t, 2H), 3.42 (t, 2H), 2.36 (t,
2H), 2.35 (m, 2H), 1.93 (m, 2H), 1.80 (m, 2H), 1.60 (m, 2H), 1.29 (m,
6H), 1.26 (t, 2H), 0.87 (m, 3H). 13C NMR (100 mHz, CDCl3): δ 178.3,
173.2, 64.8, 63.6, 33.5, 33.4, 33.2, 31.6, 29.4, 28.7, 27.4, 25.7, 22.7,
20.3, 14.2. Mass spectra: (M + H)+ (81Br) 353.1150 amu, found
353.1153 amu, (M + H)+ (79Br) 351.1171 amu, found 351.1164 amu.
6-Bromohexyl Hexylpentanedioate (4, n ) 5, m ) 6). Yield )
66.4%, yellow oil. 1H NMR (400 mHz, CDCl3): δ 4.06 (m, 4H), 3.39
(t, 2H), 2.36 (dt, 4H), 1.94 (m, 2H), 1.86 (m, 2H), 1.61 (m, 4H), 1.46
(m, 2H), 1.36-1.28 (m, 8H), 0.878 (m, 3H). 13C NMR (100 mHz,
CDCl3): δ 173.22, 173.21, 64.82, 64.49, 33.88, 33.54, 32.78, 31.61,
28.76, 28.75, 28.63, 27.96, 25.76, 25.34, 22.72, 20.39, 14.19.
6-Bromohexyl Decylpentanedioate (4, n ) 9, m ) 6). Yield )
63.7%, yellow oil. 1H NMR (400 mHz, CDCl3): δ 4.04 (dt, 4H), 3.38
(t, 2H), 2.34 (dt, 4H), 1.92 (m, 2H), 1.59 (m, 6H), 1.23 (m, 18H),
0.848 (m, 3H). Mass spectra: theory (M + H)+ 435.2110 amu, observed
435.2114 amu.
Ethyl 8-Chloro-8-oxooctanoate (7). Yield ) 97.3%, reddish brown
liquid (immediately carried to next reaction without additional char-
acterization).
Dodecanoyl Chloride (11). Yield ) 96%, opaque colorless liquid
(immediately carried to next reaction without additional characteriza-
tion).
General Procedure for Compounds 3, 8, 12.28,29 Hexyl 4-Hy-
droxybutylpentanedioate (3, n ) 5, m ) 4). Compound 2 (n ) 5)
(13.28 g, 56.5 mmol, 1 equiv) was diluted in 50 mL of dry CHCl3 and
placed in a separatory funnel. Butane diol (50.86 g, 564.4 mmol, 10
equiv) and 20 mL of dry pyridine were placed in a round-bottom flask
and put on stir at 0 °C. After the diol mixture was cooled to 0 °C,
compound 2 (n ) 5) in the separatory funnel was slowly dripped into
the round-bottom flask. The reaction was covered with a drying tube
and allowed to sit on ice. After 15 min, the reaction was removed from
ice and stirred at room temperature for 3-4 h. Once the reaction
completed, the mixture was poured into 20 mL of H2O and extracted
with CHCl3 (3 × 50 mL). The chloroform layer was concentrated down
to 75 mL, and then washed with several 50 mL portions of 0.1 N HCl
to remove excess pyridine. After the pyridine was removed, the
chloroform layer was washed with 2 × 50 mL portions of distilled
water and dried over MgSO4. A silica column was then used to separate
the desired alcohol and the disubstituted product using a 1:1 ethyl
acetate/hexane solvent system. Ten milliliter fractions were collected,
and the contents were identified using a Phosphomolybdic acid TLC
stain. Product fractions were collected, and the solvent was then
evaporated under reduced pressure to yield a light brown oil. Yield:
11.73 g (71.7%). 1H NMR (300 mHz, CDCl3): δ 4.06 (t, 2H), 4.01 (t,
2H), 3.61 (t, 2H), 2.32 (t, 2H), 2.31 (t, 2H), 2.1 (br s, 1H) 1.89 (m,
2H), 1.75-1.5 (m, 6H), 1.25 (m, 6H), 0.83 (t, 3H). 13C NMR (75 mHz,
CDCl3): δ 172.8, 172.7, 64.4, 64.0, 62.0, 33.1 (2C), 31.1, 28.8, 28.3,
25.3, 24.8, 22.2, 19.9, 13.7. Mass spectra: M + Li 295.2097 amu,
found 295.2090 amu.
1
8-Bromooctyl Ethylsuberate (9). Yield ) 62.7%, yellow oil. H
NMR (400 mHz, CDCl3): δ 4.09 (t, 2H), 4.03 (t, 2H), 3.38 (t, 2H),
2.26 (dt, 4H), 1.83 (m, 4H), 1.60 (m, 4H), 1.41-1.31 (m, 12H), 1.23
(t, 3H). 13C NMR (100 mHz, CDCl3): δ 173.95, 173.86, 64.47, 60.33,
34.42 (2C’s), 34.08, 32.89, 29.19 (2C’s), 28.91, 28.78, 28.73, 28.2 (2
C’s), 25.97, 24.92, 14.42.
Hexyl 6-Hydroxylhexylpentanedioate (3, n ) 5, m ) 6). Yield )
83.9%, light brown oil. 1H NMR (300 mHz, CDCl3): δ 3.98 (dt, 4H),
3.54 (t, 2H), 2.29 (dt, 4H), 1.86 (m, 2H), 1.53 (m, 6H), 1.30-1.22 (m,
10H), 0.804 (t, 3H). 13C NMR (75 mHz, CDCl3): δ 173.09, 173.08,
64.63, 64.46, 62.53, 33.34, 32.53, 31.42, 29.37, 28.57, 28.49, 25.74,
25.57, 25.42, 22.52, 20.21, 13.99.
Dodecanoic Acid 3-Bromopropyl Ester (13). Yield ) 97%, yellow
oil. 1H NMR (400 mHz, CDCl3): δ 4.188 (t, 2H), 3.444 (t, 2H), 2.285
(t, 2H), 2.157 (m, 2H), 1.597 (m, 2H), 1.265 (m, 16H), 0.859 (t, 3H).
General Procedure for Compounds 5, 10.32 Surfactant A (5, n
) 5, m ) 4). First, 8.8406 g (25.3 mmol, 1 equiv) of compound 4 (n
) 5, m ) 4) and 25 mL of ethanol were placed in a round-bottom
flask and magnetically stirred in a 45 °C oil bath. The solution was
treated with 5.5392 g (30.9 mmol, 1.2 equiv) of 33% trimethylamine
in ethanol. A drying tube covered the flask, and the reaction continued
for 2 days at 45 °C. Upon completion, the reaction was removed from
heat and the solvent was stripped. Ether (50 mL) was added to the
crude product followed by an extraction of the desired product with 3
× 50 mL portions of water. The water was removed by sublimation to
yield 4.54 g (44% yield) of a flakey white powder. 1H NMR (300 mHz,
D2O): δ 4.16 (t, 2H), 4.09 (t, 2H), 3.42 (m, 2H), 3.15 (s, 9H), 2.45
(m, 2H), 2.42 (m, 2H), 1.91-1.63 (m, 8H), 1.31 (m, 6H), 0.88 (t, 3H).
13C NMR (75 mHz, D2O): δ 175.3 (2C’s), 65.3, 64.3, 33.2, 33.1, 31.2,
28.2, 25.3, 25.0, 22.3, 20.0, 19.5, 13.6. Mass spec. theory (M - Br)+
Decyl 4-Hydroxybutylpentanedioate (3, n ) 9, m ) 4). Yield )
1
75.8%, light brown oil. H NMR (400 mHz, CDCl3): δ 4.08 (t, 2H),
4.03, (t, 2H), 3.64 (t, 2H), 2.34 (dt, 4H), 1.92 (m, 2H), 1.68 (m, 2H),
1.58 (m, 2H), 1.23 (m, 16H), 0.85 (t, 3H). Mass spectra: theory (M +
H)+ 345.2641 amu, observed 345.2653 amu.
Decyl 6-Hydroxyhexylpentanedioate (3, n ) 9, m ) 6). Yield )
55.1%, light brown oil. 1H NMR (400 mHz, CDCl3): δ 4.02 (m, 4H),
3.58 (t, 2H), 2.32 (dt, 4H), 1.89 (m, 2H), 1.57 (m, 6H), 1.34-1.21 (m,
18H), 0.829 (t, 3H). 13C NMR (100 mHz, CDCl3): δ 173.21, 173.18,
64.76, 64.55, 62.72, 33.43, 32.66, 31.98, 29.62, 29.40, 29.34 (2C), 28.70
(2C), 26.00, 25.83, 25.69, 25.50, 22.77, 20.29, 14.21. Mass spec. theory
(M + H)+ 373.2954 amu, observed 373.2935 amu.
Ethyl 8-Hydroxyoctylsuberate (8). Yield ) 56.6%, light brown
1
oil. H NMR (400 mHz, d6-DMSO) δ 4.02 (t, 2H), 3.98 (t, 2H), 3.37
(t, 2H), 2.26 (dt, 4H), 1.50-1.252 (m, 20H), 1.16 (t, 3H). 13C NMR
(30) Levy, D.; Stevenson, R. J. Org. Chem. 1965, 30, 3469.
(31) Sandri, J.; Viala, J. Synth. Commun. 1992, 22, 2945.
(32) Gray, A. P.; Schlieper, D. C.; Spinner, E. E.; Cavallito, C. J. J. Am. Chem.
Soc. 1955, 77, 3648.
(28) Traynham, J. G.; Pascual, O. S. J. Org. Chem. 1956, 21, 1362.
(29) Lambert, J. B.; Wang, G. T.; Finzel, R. B.; Teramura, D. H. J. Am. Chem.
Soc. 1987, 109, 7838.
9
15888 J. AM. CHEM. SOC. VOL. 126, NO. 48, 2004

Contiguous versus Segmented Hydrophobicity
A R T I C L E S
330.2644 amu, found 330.2628 amu. Elemental analysis: theory
(surfactant A + 1H2O) 50.56% C, 8.96% H, 3.28% N; found 50.25%
C, 8.74% H, 3.29% N.
28.03, 25.66, 25.26 (2 C’s), 24.28, 24.24, 21.98 (2 C’s), 14.12. Mass
spec. theory (M - Br)+ 372.3114 amu, found 372.3112 amu. Elemental
analysis: theory (2surfactant E + 1H2O) 54.75% C, 9.42% H, 3.04%
N; found 54.92% C, 9.38% H, 3.10% N.
Surfactant B (5, n ) 5, m ) 6). Yield ) 43%, flakey white powder.
1H NMR (400 mHz, d6-DMSO): δ 4.01 (t, 2H), 3.996 (t, 2H), 3.33
(m, 2H), 3.046 (s, 9H), 2.35 (t, 2H), 2.31 (t, 2H), 1.8-1.5 (m, 8H),
1.26 (m, 10H), 0.856 (m, 3H). 13C NMR (100 mHz, d6-DMSO): δ
172.48, 172.46, 65.15, 63.80, 63.64, 52.10, 32.53 (2C), 30.82, 28.05,
27.82, 25.34, 25.00, 24.87, 21.98, 21.92, 19.93, 13.86. Mass spec. theory
(M - Br)+ 358.2957 amu, found 358.2968 amu. Elemental analysis:
theory (surfactant B + 1H2O) 52.72% C, 9.30% H, 3.08% N; found
52.60% C, 9.35% H, 3.23% N.
1
Surfactant F (14), Yield ) 92%, fluffy white powder. H NMR
(400 mHz, d6-DMSO): δ 4.070 (t, 2H), 3.385 (m, 2H), 3.076 (s, 9H),
2.305 (t, 2H), 2.023 (m, 2H), 1.516 (m, 2H), 1.235 (m, 16H), 0.850 (t,
3H). 13C NMR (100 mHz, d6-DMSO): δ 172.8, 60.8, 52.2, 33.4, 31.3,
29.0, 28.9, 28.7, 28.5, 24.3, 22.1, 13.9. Mass spec. theory (M - Br)+
300.2903 amu, found 300.2901 amu. Elemental analysis: theory
(3surfactant F + 1H2O) 56.21% C, 10.29% H, 3.58% N; found 56.41%
C, 10.02% H, 3.36% N.
Surfactant C (5, n ) 9, m ) 4). 1H NMR (300 mHz, D2O): δ 4.13
(t, 2H), 4.04 (t, 2H), 3.36 (m, 2H), 3.13 (s, 9H), 2.43 (t, 4H), 1.87 (m,
4H), 1.84 (m, 2H), 1.72 (m, 2H), 1.26 (m, 14H), 0.85 (m, 3H). 13C
(100 mHz, d6-DMSO): 172.42, 172.39, 64.6, 63.76, 63.03, 52.04, 32.52,
32.48, 31.26, 28.9 (2 C’s), 28.66, 28.62, 28.08, 25.32, 24.99, 22.07,
19.84, 18.92, 13.89. Mass spec. theory (M - Br)+ 386.3270 amu,
observed 386.7836 amu. Elemental analysis: theory (surfactant C +
1H2O) 54.63% C, 9.59% H, 2.90% N; found 54.47% C, 9.56% H,
2.85% N.
8-Ethoxy-8-Oxooctanoic Acid (6).26,33 First, 12.8732 g of suberic
acid was added to a 1 L round-bottomed flask containing 360 mL of
H2O, 300 mL of ethanol, and 3 mL of concentrated H2SO4. The reaction
was continuously extracted with cyclohexane for 2 days using a Kimble
Kontes continuous liquid/liquid, lighter than water extraction apparatus.
After 2 days, the layers were separated and the cyclohexane was allowed
to cool. Upon cooling, the cyclohexane was filtered through fine filter
paper to remove any excess suberic acid that had precipitated out of
solution. After concentrating the cyclohexane layer to 150 mL, 4 ×
200 mL of a saturated NaHCO3 solution was used to extract the mono-
esterified compound from the cyclohexane layer. The water was
collected, and the pH was adjusted to 2 using dilute HCl. Four 100
mL portions of ether were used to extract the product from the aqueous
layer. The ether layer was dried over MgSO4, and the solvent was
stripped to yield 7.0737 g (47.3%) of a yellow oil. 1H NMR (400 mHz,
CDCl3): δ 4.12 (q, 2H), 2.33 (t, 2H), 2.28 (t, 2H), 1.61 (m, 4H), 1.34
(m, 4H), 1.24 (t, 3H). 13C NMR (100 mHz, CDCl3): δ 180.27, 174.03,
60.45, 34.42, 34.15, 28.90, 28.83, 24.90, 24.62, 14.41.
Surfactant D (5, n ) 9, m ) 6). Yield ) 54.7%, fluffy white
1
powder. H NMR (400 mHz, D2O): δ 4.07 (t, 2H), 4.04 (t, 2H), 3.3
(m, 2H), 3.13 (s, 9H), 2.39 (t, 2H), 2.36 (t, 2H), 1.87 (m, 4H), 1.64
(m, 4H), 1.40-1.26 (m, 18H), 0.86 (t, 3H). 13C NMR (100 mHz, d6-
DMSO): δ 172.44, 172.43, 65.06, 63.77, 63.64, 52.05, 32.52 (2 C’s),
31.28, 28.91 (2 C’s), 28.68, 28.62, 28.08, 27.83, 25.34, 24.88, 22.08
(2 C’s), 21.92, 19.91, 13.93. Mass spec. theory (M - Br)+ 414.3583
amu, found 414.3573 amu. Elemental analysis: theory (surfactant D
+ 1H2O) 56.33% C, 9.86% H, 2.74% N; found 56.57% C, 9.74% H,
2.90% N.
Surfactant E (10). Yield ) 78.0%, fluffy white powder. 1H
NMR (400 mHz, d6-DMSO): δ 4.029 (t, 2H), 3.999 (t, 2H), 3.253 (m,
2H), 3.028 (s, 9H), 2.262 (dt, 4H), 1.657 (m, 2H), 1.6-1.45 (m, 6H),
1.35-1.2 (m, 12H), 1.168 (t, 3H). 13C NMR (100 mHz, d6-DMSO): δ
172.90, 172.85, 65.16, 63.61, 59.64, 52.06, 33.40, 28.37 (2 C’s), 28.08,
Acknowledgment. This work was supported by the Petroleum
Research Foundation and the Army Research Office.
JA040105S
(33) Babler, J. H.; Moy, R. K. Synth. Commun. 1979, 9, 669.
9
J. AM. CHEM. SOC. VOL. 126, NO. 48, 2004 15889