Synthesis, characterization, and applications of hemifluorinated dibranched amphiphiles

Article abstract of DOI:10.1021/jo200835y

Here we describe the synthesis and the physicochemical and preliminary pharmaceutical assessment of a novel class of hemifluorinated dibranched derivatives: M₁diH_xF_y. These compounds have the remarkable ability to completely stop the Ostwald ripening commonly associated with nanoemulsions. The developed synthesis is modular and allows easy incremental structural variations in the fluorophilic (fluorous chains), lipophilic (alkyl spacer head), and hydrophilic (polar head) domains. Furthermore, the synthesis can be easily scaled up and highly pure compounds can be readily obtained through silica gel and fluoro-silica gel column chromatography, without any need to use HPLC or other time-consuming techniques. Surface properties such as micelle formation, critical aggregation concentration (CAC), and emulsion stability studies demonstrated the different behavior of the dibranched hemifluorinated surfactant M₁diH _xF_y with respect to that of single-chain semifluorinated analogues M_zF_y. Remarkably, the new polymer M ₁diH₃F₈ drastically slowed the ripening of nanoemulsions of the commonly used fluorinated anesthetic sevoflurane over a period of more than 1 year. During this time, the nanodroplet size did not increase to more than 400 nm. This result is very promising for inducing and maintaining general anesthesia through intravenous delivery of volatile anesthetics, eliminating the need for the use of large and costly vaporizers in the operating room.

Full text of DOI:10.1021/jo200835y

ARTICLE
pubs.acs.org/joc
Synthesis, Characterization, and Applications of Hemifluorinated
Dibranched Amphiphiles
Maria Cristina Parlato,† Jun-Pil Jee,† Motti Teshite,† and Sandro Mecozzi*,†,‡
†School of Pharmacy and ‡Department of Chemistry, University of Wisconsin—Madison, Madison, Wisconsin 53705, United States
S Supporting Information
b
ABSTRACT: Here we describe the synthesis and the physico-
chemical and preliminary pharmaceutical assessment of a novel
class of hemifluorinated dibranched derivatives: M1diHxFy.
These compounds have the remarkable ability to completely
stop the Ostwald ripening commonly associated with nano-
emulsions. The developed synthesis is modular and allows easy
incremental structural variations in the fluorophilic (fluorous
chains), lipophilic (alkyl spacer head), and hydrophilic (polar
head) domains. Furthermore, the synthesis can be easily scaled up and highly pure compounds can be readily obtained through silica
gel and fluoro-silica gel column chromatography, without any need to use HPLC or other time-consuming techniques. Surface
properties such as micelle formation, critical aggregation concentration (CAC), and emulsion stability studies demonstrated the
different behavior of the dibranched hemifluorinated surfactant M1diHxFy with respect to that of single-chain semifluorinated
analogues MzFy. Remarkably, the new polymer M1diH3F8 drastically slowed the ripening of nanoemulsions of the commonly used
fluorinated anesthetic sevoflurane over a period of more than 1 year. During this time, the nanodroplet size did not increase to more
than 400 nm. This result is very promising for inducing and maintaining general anesthesia through intravenous delivery of volatile
anesthetics, eliminating the need for the use of large and costly vaporizers in the operating room.
’ INTRODUCTION
possibility of using semifluorinated surfactants for the intrave-
nous delivery of the commonly used class of fluorinated volatile
anesthetics.15,16 We found that by using linear diblock copoly-
mers such as the simple M5F13 polymer (Figure 1), we were able
to solubilize up to 25% of sevoflurane in water.
Fluorinated amphiphiles have unique physicochemical prop-
erties, including high surface tension activity, thermal and chemi-
cal stability, and biological inertness.1,2 The fluorous moiety is
characterized by both high hydro- and lipophobicity. Further-
more, linear fluorocarbons are rigid rods due to the size of the
fluorine atom. All these properties allow fluorinated amphiphiles
to form better organized and more stable self-assembling systems
than their hydrocarbon counterparts.1À12 Also, important from a
biological standpoint, the high lipophobicity of fluorous chains
mostly prevents the fusion of these molecules with biological
membranes, therefore strongly reducing their hemolytic activity
and acute toxicity in comparison to their hydrocarbon analogues.
Fluorocarbons are not metabolized in vivo and are eventually
excreted unmodified.13 Due to these unique physicochemical and
biological properties, fluorinated amphiphiles are one of the most
promising and innovative tools for the construction of drug-
delivery devices and other biomedical applications.
Intravenous delivery of fluorinated volatile anesthetics with
little or no aqueous solubility has been a focus area of pharma-
ceutical research for over 40 years, due to the potential for
pharmacodynamics improvements. However, the peculiar prop-
erties of fluorous molecules have prevented a quick solution to
the problem of finding formulations able to stabilize fluorous
anesthetics in water. For instance, the use of lipid-based emul-
sions only allowed formulations containing up to 3.6% of
sevoflurane, a concentration not sufficient for medical applica-
tions as an intravenous anesthetic.14 We originally proposed the
In vivo testing of these fluorous emulsions showed that both
induction and recovery from intravenous emulsified halogenated
anesthetics are more rapid than those with administration as a
vapor.16 Additionally, the intravenous administration of fluorinated
ethers has promising implications regarding end-organ protec-
tion against ischemia and reperfusion injury.17,18 This precondi-
tioning is found to be a separate pharmacologic response from
the drug effects that result in general anesthesia. Therefore, the
potential clinical applications of intravenous emulsified fluori-
nated anesthetics may well extend beyond the operating room.19
Sevoflurane, a moderately water soluble highly fluorinated ether,
used for induction and maintenance of general anesthesia, is the
preferred agent in clinical anesthesia, due to less irritation to
mucous membranes in comparison to that for other fluorous
ethers with anesthetic properties such as isoflurane and
desflurane.20,21 The affinity of fluorinated anesthetics for fluo-
rophilic molecules prompted the investigation of the emulsifica-
tion of sevoflurane using the semifluorinated surfactant M5F13
(Figure 1) with the fluorinated and FDA-approved additive
perfluorooctyl bromide to slow ripening. This new formulation
Received: May 2, 2011
Published: July 07, 2011
r
2011 American Chemical Society
6584
dx.doi.org/10.1021/jo200835y J. Org. Chem. 2011, 76, 6584–6591
|

The Journal of Organic Chemistry
ARTICLE
molecules on the surface of the nanoparticle would then be
dictated by the steric hindrance of the large poly(ethylene glycol)
moieties. In the case of the original polymer M5F13 the large PEG
would prevent tight packing of the surfactant around each
nanoparticle (Figure 2).
Loose packing between the surfactant molecules surrounding
the nanoparticles would then allow anesthetic molecules to
diffuse into the aqueous phase and contribute to the growing
of larger particles by the Ostwald ripening effect. We reasoned
that, by reducing the PEG size and by increasing the volume of
the fluorous moiety, we might surround each droplet with a more
dense, less penetrable polymeric layer and, as a consequence,
reduce the rate of diffusion of the anesthetic out of the nano-
particles (Figure 2). According to this design, we selected a PEG
with an average molecular weight of 1000 as opposed to the PEG
5000 of the original polymer, and we transformed the single
fluorocarbon chain we used in M5F13 into two hemifluorinated
chains connected to the PEG polar chain through a glycerol
moiety. In addition, we introduced a small hydrocarbon segment
between the two fluorous chains and the glycerol moiety to
provide some flexibility to the surfactant fluorous moieties.
We herein describe the synthesis and the preliminary physi-
cochemical and pharmaceutical assessments of this new class of
nonionic dibranched derivatives M1diHxFy (Figure 1). The
number following the letter M indicates the molecular weight
of the PEG. Thus, M5 indicates a methyl-capped poly(ethylene
glycol) of average molecular weight of 5000. The number after
the H indicates the number of carbon atoms composing the
hydrocarbon segment, while the number after the F indicates the
number of carbon atoms in the fluorocarbon chains. All these
new surfactants were able to slow the ripening of sevoflurane
emulsions, in comparison with the original linear diblock copo-
lymers. Remarkably, the new polymer M1diH3F8 drastically
slowed nanoparticle ripening and stabilized sevoflurane emul-
sions over 1 year with a droplet size of under 400 nm.
Synthesis. A series of fluorinated 1,2-di-O-alkylglycerol-PEG
amphiphiles have been synthesized on a gram scale with high
purity. The final surfactants were easily purified by silica gel and
fluoro-silica gel column chromatography without requiring
further purification by HPLC. Their molecular structures follow
a modular design that allows incremental structural variations.
This design involves two small hydrocarbon chains ending with a
fluorocarbon functionality. The two mixed hydrocarbon/fluo-
rocarbon chains are then connected through a glycerol unit to a
PEG (MW 1100) polar head. The various polymers in this series
differ in the size of both the fluorous chain and the hydrocarbon
spacer. We have used two different fluorocarbons, containing
six and eight fluorinated carbons, respectively. The hydrocarbon
spacer between the PEG moiety and the terminal perfluoroalkyl
group contains 3, 5, or 10 methylene groups. These structural
features are expected to influence the fluorophilic/lipophilic/
hydrophilic balance and, consequently, physicochemical and bio-
logical properties such as nanoparticle permeability and nano-
emulsion stability.
Figure 1. Structures of the volatile anesthetic sevoflurane, the additive
perfluorooctyl bromide, the polymer series MzFy, and the new polymer
series M1diHxFy. The number following the letter M indicates the
molecular weight of the PEG. Thus, M5 indicates a methyl-capped
poly(ethylene glycol) of average molecular weight of 5000. The number
after the H indicates the number of carbon atoms composing the
hydrocarbon segment, while the number after the F indicates the
number of carbon atoms in the fluorocarbon chains.
allowed the emulsification of a considerable amount of sevoflur-
ane, up to 25% v/v, by exploiting the greater solubility of the
anesthetic in a fluorous phase. This was a significant improve-
ment (6-fold increase) over the maximum amount of sevoflurane
that can be emulsified in classic lipid emulsion (intralipid), 3.46%
v/v.6 Our formulations have the ability of inducing general
anesthesia by intravenous injection in rats16 and in dogs,22
combined with a rapid recovery profile without causing acute
toxicity. Nevertheless, the sevoflurane/fluoropolymer emulsions
were susceptible to fast particle size growth. For any emulsion
formulation to become clinically useful, the particle droplet size
must be under 500 nm and have a shelf life of 18 months. The
addition of perfluorooctyl bromide as a secondary component
did slow the ripening, but unfortunately the droplet size was over
500 nm after just 30 days. Limiting ripening is fundamental to
provide an effective and convenient means of inducing anesthesia
in human patients. This problem led us to design a new class of
fluorinated surfactants that could provide extended emulsion
stability.
’ RESULTS AND DISCUSSION
Our design started from considering the structure of the
nanoparticles in our emulsions. The original surfactant M5F13
(3b; Figure 1) consisted of a large poly(ethylene glycol) with an
average molecular weight of 5000 attached to a relatively small
rigid perfluorocarbon. Thus, presumably, the surfactant chains
would coat the fluorous nanoparticle composed of the anes-
thetic 1 and the additive 2 by binding the fluorous chain to the
inner fluorous core. The distance between the various surfactant
While the synthesis of dialkyl-semifluorinated amphiphiles is
documented in the literature, the synthesis of PEG-containing
fluorous surfactants is challenging. Our main constraint was
dictated by the need for gram-scale synthesis in order to produce
enough amphiphile as needed for physicochemical characteriza-
tion and for all pharmaceutical and biological studies. The
synthesis and purification of compounds containing fluorous
moieties can be very problematic, due to their unusual reactivities,
6585
dx.doi.org/10.1021/jo200835y |J. Org. Chem. 2011, 76, 6584–6591

The Journal of Organic Chemistry
ARTICLE
Figure 2. Rational design of stable nanoemulsions. (A) The original polymer composed of a large poly(ethylene glycol) and a single small fluorocarbon
chain can self-assemble around the fluorous nanoparticle only in a loose manner, leaving spaces for the diffusion of the anesthetic out of the nanoparticle.
The diffused anesthetic would then contribute to the growing of larger nanoparticles by the Ostwald ripening effect. (B) The newly designed polymers
composed of a smaller PEG and a larger two-chain fluorous moiety is able to better surround the fluorous nanoparticle and strongly reduce the amount of
anesthetic that can diffuse out.
Scheme 1
hydrophobicity, and lipophobicity, which can lead to difficulties in
handling, solubilization, and purification of reagents and inter-
mediates of reaction. All these issues are exacerbated if fluorous
compounds containing reactive functionalities directly attached
to the strongly electron withdrawing fluorous chains are used in
the synthesis. In fact, the electron-withdrawing properties of
fluorocarbons will greatly reduce the reactivity of these molecules.
To improve on the reactivity of fluorous intermediates, we
resorted to the use of short (up to eight carbon atoms) fluorous
chains, containing a small hydrocarbon spacer between the
reactive functionalities and the fluorous tails (Scheme 1). The
next task consisted of identifying the conditions for a quick and
easy way to purify the PEG-containing semifluorinated polymeric
surfactants. Unfortunately, a mixture of different fluorous amphi-
philic PEG derivatives cannot be purified by using common
laboratory techniques such as silica gel column chromatography
or precipitation when the PEG moiety is large compared to the
hydrophobic/fluorophilic moieties. Then, the only possibility of
avoiding time-consuming and expensive purification techniques
such as preparative HPLC resides in the possibility of pushing to
completion all reactions run on PEG-containing molecules and
purifying the final product.
The synthesis of of the di-O-alkylglycerol precursor 7 has been
reported in the literature23 and was performed by alkylation
under phase-transfer-catalysis conditions of benzyl-monopro-
tected rac-1-O-benzylglycerol using perfluoroalkyl mesylates,
followed by hydrogenolysis for removal of the benzyl protecting
group. Purification of compounds 7 would require a tedious
purification by silica gel chromatography because of the close
proximity of the Rf values for the mesyl alcohol 6 and derivative 7.
For this reason, we proceeded to the following hydrogeno-
lysis after just quick filtration on a silica gel pad. The deprotected
6586
dx.doi.org/10.1021/jo200835y |J. Org. Chem. 2011, 76, 6584–6591

The Journal of Organic Chemistry
ARTICLE
Scheme 2
derivatives 8aÀe could then be easily purified from other
impurities by simple silica gel column chromatography.
Several of the partially fluorinated alcohols 11 with the general
structure CmF2m+1(CH2)nOH were not commercially available.
They were synthesized from 9-decen-1-ol or 4-penten-1-ol fol-
lowing a procedure reported in the literature.24,25 A sequence of
redox and radical reactions is used in the synthesis of these
alcohols (Scheme 2).
Table 1
polymer
particle size (nm)a
CAC (M)
M1F13
17.17 ( 1.85
17.23 ( 3.73
36.45 ( 1.64
80.53 ( 7.93
81.48 ( 15.17
83.66 ( 13.58
112.93 ( 1.83
5.50 Â 10À7
7.94 Â 10À7
9.91 Â 10À6
6.11 Â 10À6
5.45 Â 10À6
not available
7.22 Â 10À7
M5F13
M1diH3F6
M1diH3F8
M1diH5F8
M1diH10F6
M1diH10F8
In the last step for the synthesis of the amphiphiles (Scheme 1),
the glycerol derivatives 8 were coupled with polyethylene glycol
monomethyl ether (MW 1100) methanesulfonate ester by fol-
lowing a procedure developed in our laboratory.8 The reactions
a Particle sizes of fluoropolymer-based aggregates. Data are given with the
standard deviation. Each measurement was repeated three times (n = 3).
1
were monitored by H NMR spectroscopy. As shown by FT-
MALDI spectra, a small amount of polyethylene glycol mono-
methyl ether was present in the final compound, possibly deriving
from hydrolysis of the polymeric methanesulfonate ester during
the coupling step. This impurity was easily removed by fluoro
flash silica gel chromatography, while ordinary silica gel chroma-
tography allowed the removal of fluorous alcohols and other side
products. All steps of Schemes 1 and 2 have been performed on a
gram scale with high purity (as shown from both HPLC and mass
spectrometry analysis) and in good yields.
shape formed by self-assembling amphiphiles are dictated by the
ratio of the volumes between the hydrophilic moiety and the
hydrophobic moiety. In the case at hand, a relatively small PEG
(MW 1100, corresponding to 25 ethylene glycol units) is attached
to a wide hydrophobic tail composed of two different chains. This
combination may in principle lead to the formation of various
aggregates, from cylindrical micelles to vesicles and bilayer struc-
tures in general. A transmission electron microscopy study of the
aggregates made by the polymer M1diH10F6 in aqueous solution
showed a variety of particles of different sizes and shapes,
suggestive of the formation of various bilayer structures (see the
Supporting Information). It is important to recognize that the
ability of a surfactant to form either a micelle or other structures
does not have any effect on its ability in stabilizing a nanoemul-
sion. As a matter of fact, the same geometric features that in
polymer M1diH10F6 make impossible the formation of micelles,
namely a relatively small hydrophilic head and a wide hydro-
phobic tail, are also the reasons we expected an improved
stabilization of the corresponding nanoemulsions. Wider hydro-
phobic and fluorophilic tails would pack better around the
fluorophilic nanodroplets, and a smaller PEG (from M5 to M1)
would reduce the spaces between the polymeric chains surround-
ing the nanodroplets, thus reducing the diffusion of the anesthetic
out of the nanodroplet core.
Physicochemical Characterization. All synthesized surfac-
tants were highly soluble in water at room temperature, and the
solutions were clear. Interestingly, M1diH3F8 gave a high-visc-
osity solution that was, however, still suitable for emulsion
preparation. Fluorinated surfactants are characterized by higher
hydrophobicity and start to aggregate at a relatively lower con-
centration than the corresponding hydrogenated compounds.
Furthermore, they form better organized and more stable struc-
tures than their hydrocarbon counterparts. Such interesting prop-
erties mainly result from the stiff structure of the fluorocarbon
chain, which is bulkier and more rigid than a hydrocarbon.1 We
initially expected amphiphiles 4aÀe to form micelles in aque-
ous solution. Dynamic light scattering was applied to measure
the average size of the amphiphilic aggregates in solution. The
particle size data are summarized in Table 1. Interestingly, the size
of the aggregates made by the various polymers radically changes
on the basis of very small changes in the structure of the polymers.
For instance, the average nanoparticle size doubles when M1di-
H3F8 is used in place of M1diH3F6. The aggregate size reaches the
remarkable value of about 113 nm with M1diH10F8. In compar-
ison, the micellar aggregate made by a linear diblock copolymer
such as M1F13 is just 17 nm, as expected. The much larger size of
the aggregates made by the dibranched polymers suggests that
aggregates other than micelles are actually formed. The size and
The critical aggregate concentrations (CAC) of the new
polymers were estimated by measuring the surface tension of
polymer solutions at various concentrations. Increasing the
concentration of the polymer under the CAC leads to a linear
decrease in the surface tension. When the CAC is reached, the
surface tension no longer changes and a plateau value is reached.
The CAC can be quickly estimated by the intersection between
the plateau line and the line formed by the points in which a linear
decrease in surface tension is observed. The CAC data presented
6587
dx.doi.org/10.1021/jo200835y |J. Org. Chem. 2011, 76, 6584–6591

The Journal of Organic Chemistry
ARTICLE
in Figure 3, the new polymer M1diH3F8 very remarkably slowed
the ripening and stabilized sevoflurane emulsions over 1 year.
Furthermore, the droplet size stayed under 400 nm and there was
no visual evidence of either phase separation or sedimentation. In
comparison, the linear diblock copolymers M5F13 and M1F13
showed droplet sedimentation after only 30 and 60 days, respec-
tively. These stability studies were reproduced with three more
identical emulsions containing M1diH3F8, which were followed in
terms of nanoparticle size for 1 year. All emulsions showed
identical stabilities. Thus, the use of dibranched semifluorinated
surfactants allowed us to prepare nanoemulsions that are stable
for extended times. The stability of these emulsions is consistent
with a shelf life acceptable for clinical use.
’ CONCLUSIONS
A novel class of hemifluorinated dibranched PEG-based surfac-
tants was synthesized. The purification of PEG-containing fluo-
rous molecules can be very difficult. The value of the developed
synthesis lies in the possibility of obtaining surfactants on a gram
scale with high purity by using simple purification techniques such
as simple silica gel and fluoro flash silica gel column chromatog-
raphy. The synthesis of these surfactants has been developed with
a modular approach, which easily allows incremental structural
variations in the fluorophilic, hydrophobic, and hydrophilic do-
mains. The relative ratio of the three molecular domains affects the
hydrophobic/hydrophilic balance, leading to the possibility of
tuning physicochemical (stability, fluidity/rigidity, permeability)
and biological properties.
Figure 3. Change in the particle sizes of fluoropolymer-based emul-
sions with time: (9) M5F13-based emulsion; (0) M1F13-based emul-
sion; (b) M1diH3F6-polymer-based emulsion; (O) M1diH3F8-based
emulsion; (2) M1diH5F8-polymer-based emulsion; (4) M1diH10F6-
based emulsion; ([) M1diH10F8 based emulsion. The particle size was
measured over 357 days. Data are given with standard deviation (n = 3).
Data for the last three emulsions are shown in red. These emulsions led
to phase separation after 50 days. Accordingly, particle size measure-
ments were halted after that time. All emulsions contained 20% of
sevoflurane and 10% of perfluoroctyl bromide and were prepared in
saline (0.9% w/w NaCl).
Surface properties such as the critical aggregate concentration
and the nanoemulsion stability were found to be affected by the
size of both the perfluorinated tail and the hydrogenated spacer,
demonstrating the peculiar behavior of dibranched hemifluori-
nated surfactants with respect to single-chain perfluorinated
analogues. All the newly synthesized dibranched surfactants were
able to stabilize nanoemulsions containing the fluorinated anes-
thetic sevoflurane and the additive perfluorooctyl bromide better
than linear diblock copolymers such as M1F13. Remarkably, the
new polymer M1diH3F8 dramatically slowed the ripening and
stabilized sevoflurane emulsions over 1 year with a droplet size of
under 400 nm. The stabilization of nanoemulsions for an extended
period of time is usually very difficult. The fact that the new
polymers were able to achieve exactly this feat is indicative of their
potential for clinical use.
in Table 1 clearly show the influence of small changes in the
polymer structure on the kinds of aggregates that are formed in
aqueous solution. This solution behavior agrees with what is
known about different nonionic surfactants.26,27
Emulsion Stability. In a previous article16 we have reported
on a novel anesthetic intravenous formulation composed of (1)
20% w/v of a semifluorinated surfactant, (2) 10% v/v perfluo-
rooctyl bromide (PFOB), an FDA-approved fluorous additive,
and (3) 20% v/v of a fluorinated anesthetic. This formulation
could be emulsified and the corresponding nanoemulsion could
be used for the intravenous delivery of the volatile anesthetic.
The additive perfluorooctyl bromide increased the stability of the
emulsion due to its fluorophilicity and its reduced water solubi-
lity. In principle, the slow diffusion of this secondary less water
soluble component will lead to a heterogeneous distribution with
smaller droplets enriched in the less soluble component and
larger droplets enriched in the more soluble component. How-
ever, the osmotic pressure will limit composition differences
between droplets and equilibrium will eventually be reached.
This same formulation has been used for studying the stability of
nanoemulsions formed by the new polymers MzdiHxFy. Dy-
namic light scattering was used to measure the change in size of
the droplets with time. The linear diblock copolymer M1F13 was
used as a benchmark.
’ EXPERIMENTAL SECTION
Dry reactions were performed under argon using dry solvents and
reagents. Perfluoroooctyl iodide, 1-iodoperfluorohexane, and perfluoro
alcohols were purchased from SynQuest Laboratories. Methylene
chloride and tetrahydrofuran were dried by flowing through alumina-
containing columns. Acetonitrile, dry diethyl ether, and methanol were
used as provided. Column chromatography: silica gel 60 (Merck, 70À
23 mesh), fluorous flash silica gel (Merck, 70À23 mesh).
As shown in Figure 3, the emulsions formed by the new
dibranched polymers initially show a trend similar to that observed
with M1F13. However, after an initial equilibration time, the
nanoparticle size stabilized and ripening was reduced.
The observed trends show that Ostwald ripening is still causing
an increase in the nanoparticle size, but there is an impressive
decrease in the rate of particle growth. The growing rate was
higher for polymers with larger intermediate alkyl chains, while it
was smaller for polymers with larger fluorinated chains. As shown
HPLC chromatograms for product purity determination were ob-
tained using a Jordi RP-DVB column with a particle size of 5 μm and
pore size of 1000 Å. The solvent gradient started at 10% acetonitrile/
90% water and increased to 100% acetonitrile over 25 min. The flow rate
was 1 mL/min.
General Procedures for Perfluorinated Alcohols 11cÀe.24,25.-
Compound 10e. Perfluorooctyl iodide (3.28 g, 6.00 mmol), NaHCO3
(431 mg, 5.04 mmol), and 85% Na2S2O4 (1.03 g, 5.04 mmol) were added
at 0 °C to a solution of 9-decen-1-ol (787 mg, 5.04 mmol) in CH3CN
6588
dx.doi.org/10.1021/jo200835y |J. Org. Chem. 2011, 76, 6584–6591

The Journal of Organic Chemistry
ARTICLE
(15 mL) and deionized H2O (5 mL). This mixture was stirred for 4 h at
room temperature. The mixture was diluted with deionized H2O and then
extracted with dichloromethane. The organic layers were washed with
saturated aqueous NaCl and then dried over MgSO4. After filtration and
rotary evaporation of the solvent, the residue was used as a crude compound
109.0 (2C), 111.4 (2C), 114.1 (2C), 116.1 (2C), 119.0 (2C), 121.8 (2C),
127.6 (2C), 127.8, 128.4 (2C), 138.5. 19F NMR (CDCl3): δ À82.4 (3F),
À115.5 (2F), À123.0 (2F), À124.0 (2F), À124.4 (2F), À127.4 (2F).
Compound 7b. White solid, 83% yield. Mass (MALDI): calcd for
[C32H24F34O3 + Na]+ 1 125.107 47, found 1 125.103 91. Rf = 0.72
(hexane/ethyl acetate 7/2 v/v). 1H NMR (CDCl3): δ 1.81À1.90
(4H, m), 2.08À2.26 (4H, m), 3.47À4.54 (9H, m), 4.55 (2H, s),
7.24À7.36 (5H, m). 13C NMR (CDCl3): δ 21.2 (d), 28.1 (t), 69.2, 69.3,
70.1, 71.2, 70.2, 71.5, 73.7, 78.5 (2C), 109.0 (2C), 111.4 (2C), 114.1 (2C),
116.1 (2C), 119.0 (2C), 121.8 (2C), 127.9 (2C), 127.8, 128.0, 128.6 (2C),
138.4. 19F NMR (CDCl3): δ À81.4 (3F), À115.5 (2F), À122.4 (2F),
À122.5 (2F), À123.4 (2F), À123.6 (2F), À124.3 (2F), À127.1 (2F).
Compounds 7cÀe. These crude compounds were filtered through a
pad of silica gel and used in the next step.
General Procedure for Hydrogenolysis. To a solution of
BndiHxFy (7; 7.2 mmol) in ethanol (50 mL) and acetic acid (0.7 mL)
was added Pd/C (0.85 g of a 19% mixture), and the reaction mixture was
stirred under a hydrogen atmosphere (1.5 atm) at room temperature for
48 h. The mixture was then filtered, the solvent was rotary evaporated,
and the residue was purified by silica gel chromatography (petroleum
ether/ethyl acetate 6/4).
1
in the next step. H NMR (CDCl3): δ 1.27À1.41 (8H, m), 1.53À1.58
(4H, m), 2.01À2.07 (2H, m), 3.64 (2H, t, J = 6.8 Hz, H1), 4.91À5.02
(2H, m), 5.76À5.86 (1H, m).
Compound 11e. The crude compound (5.04 mmol) was dissolved
in glacial acetic acid (3 mL); then zinc dust was added (988 mg,
15.12 mmol) and the mixture was stirred for 16 h at room temperature.
After filtration and evaporation of the solvent, the residue was purified by
column chromatography (EtOAc 70%/n-hexane 30%) to yield the
desired compound as a white solid (2,36 g, 75%). 1H NMR (CDCl3):
δ 1.29À1.37 (12H, m), 1.53À1.63 (4H, m), 1.98À2.11 (2H, m), 3.64
(2H, J = 6.8 Hz).
1
Compound 10c. H NMR (CDCl3): δ 1.64À1.73 (2H, m, H3),
1.74À1.99 (2H, m, H2), 2.73À3.01 (1H, m, H5), 3.72 (2H, t, J = 6.4 Hz,
H1), 4.35À4.42 (1H, m, H4).
Compound 11c. This compound was prepared with 4-penten-1-ol
and perfluoroooctyl iodide: white solid, 83% yield. 1H NMR (CDCl3): δ
1.44À1.51 (2H, m), 1.59À1.67 (4H, m, H2), 2.01À2.08 (2H, m), 3.69
(2H, t, J = 6.4 Hz).
Compound 8a. Pale oil, 75% yield over two steps. Mass (MALDI)
calculated for [C21H18F26O3 + Na]+: 835.0733, found 835.07218. RF =
Compound 10d. 1H NMR (CDCl3): δ 1.26À1.43 (8H, m), 1.55À
1.58 (4H, m), 1.73À1.89 (2H, m), 3.62À3.67 (2H, m), 4.09À4.13
(2H, m), 4.31À5.37 (1H, m).
1
0.50 (hexane/ethyl acetate 6:4 v/v); HNMR (CDCl3): δ 1.85À1.93
(4H, m), 2.10À2.26 (4H, m), 3.50À3.75 (9H, m); 13CNMR (CDCl3)
δ: 20.9 (d) , 27.8 (t), 61.6, 63.9, 68.6, 70.1, 71.1, 72.2, 80.2, 108.8 (m),
110.8 (m), 113.6 (m), 115.9 (m), 118.7 (m), 121.6 (m); 19F NMR
(CDCl3): δ-82.1 (3F), À115.5 (2F), À123.0 (2F) À124.0 (2F),
À124.2 (2F), À127.2 (2F).
Compound 11d. This compound was prepared with 9-decen-1-ol
1
and 1-iodoperfluorohexane: pale oil, 80% yield. H NMR (CDCl3): δ
1.18À1.23 (12H, m), 1.44À1À56 (4H, m), 1.90À1.97 (2H, m), 3.54
(2H, J = 6.8 Hz).
Compound 8b. White solid, 70% yield over two steps. Mass (MALDI)
calculated for [C25H18F34O3 + Na]+: 1035.06052, found 1035.065695.
RF = 0.53 (hexane/ethyl acetate 6:4 v/v); 1HNMR (CDCl3): δ 1.74À1.83
(4H, m), 2.00À2.13 (4H, m), 3.44À3.62 (9H, m); 13CNMR (CDCl3)
δ: 21.1 (d), 28.0 (t), 62.7, 69.1, 69.1, 70.2, 71.2, 79.4, 108.8 (m), 111.0 (m),
113.7 (m), 116.1 (m), 118.7 (m), 121.6 (m), 128.3; 19F NMR (CDCl3):
δ À81.9 (3F), À115.2 (2F), À122.5 (2F), À122.7 (4F) À123.6 (2F),
À124.2 (2F), À127.1 (2F).
General Procedure for Mesyl Derivatives 6aÀe. Mesyl
chloride (1.3 equiv) was added at 0 °C to a solution of alcohol 11 in
dry DCM and Et3N (2.4 equiv). The reaction mixture was warmed to
room temperature and was then stirred under argon for 16 h. The
mixture was diluted with dichloromethane and washed with water. The
organic layers were dried over MgSO4. After filtration and evaporation of
the solvent, the residue was used in the next step without further
purification. Quantitative yield (from 1H NMR). 6cÀe were obtained as
white solids and 6a,b as pale oils.
Compound 8c. White solid, 50% yield over two steps. Mass (MALDI)
calculated for [C29H26F34O3 + Na]+: 1091.12312, found 1091.12278;
RF = 0.52 (hexane/ethyl acetate 6:4 v/v); 1HNMR (CDCl3): δ 1.37À1.44
(4H, m), 1.52À1.61 (8H, m), 1.92À2.05 (4H, m), 2.62À2.65 (m, 1H)
3.38À3.67 (9H, m). 13C NMR (CDCl3) δ 20.0, 25.7, 25.8, 29.3,
29.7, 30.6, 30.8, 31.1, 62.7, 69.9, 71.0, 71.2, 79.0; 106.3 (m), 108.6 (m),
110.9 (m), 113.8 (m), 115.8 (m), 118.7 (m), 121.3 (m); 19F NMR
(CDCl3) δ: À81.9 (3F), À115.4 (3F), À122.6 (6F), À124.0 (2F),
À124.4 (2F), À127.0 (2F).
1
Compound 6a. H NMR (CDCl3): δ 1.80À2.24 (2H, m), 2.60À
2.28 (2H, m), 2.94 (3H, s), 4.23 (2H, t, J = 6.40 Hz).
1
Compound 6b. H NMR (CDCl3): δ 2.05À2.13 (2H, m), 2.19À
2.32 (2H, m), 3.05 (3H, s), 4.32 (2H, t, J = 5.8 Hz).
Compound 6c. 1H NMR (CDCl3): δ 1.49À1.57 (2H, m, H3),
1.63À1.71 (2H, m, H4), 1.77À1.84 (2H, m, H2), 2.03À2.16 (2H, m,
H5), 3.01 (3H, s, CH3), 4.25 (2H, t, J = 6.4 Hz, H1).
Compound 6d. 1H NMR (CDCl3): δ 1.30À1.40 (12H, m, H3, H4,
H5, H6, H7, H8), 1.56À1.63 (2H, m, H9), 1.72À1.79 (2H, m, H2),
1.98À2.12 (2H, m, H10), 3.03 (3H, s, CH3), 4.23 (2H, t, J = 6.4 Hz, H1).
Compound 6e. 1H NMR (CDCl3): δ 1.29À1.40 (12H, m, H3, H4,
H5, H6, H7, H8), 1.56À1.62 (2H, m, H9), 1.71À1.78 (2H, m, H2),
1.98À2.12 (2H, m, H10), 3.00 (3H, s, CH3), 4.23 (2H, t, J = 6.8 Hz, H1).
Benzylated Glycerol Derivatives Bn-diHxFy. To a solution of
FyHxOM (6; 11 mmol) and 5 (5.5 mmol) in 35 mL of Et2O was added a
solution of 12 N KOH (20 mL) and Bu4N(HSO4) (0.05 equiv). The
mixture was heated for 48 h under reflux conditions. The aqueous phase
was extracted with Et2O, and the combined organic phases were washed,
dried, and rotary evaporated. The residue was purified by silica gel
chromatography (petroleum ether/ethyl acetate 7/3).
Compound 8d. White solid, 30% yield over two steps. Mass (MALDI)
calculated for [C35H46F26O3 + Na]+: 1031.2924, found 1031.28853.
RF = 0.51 (hexane/ethyl acetate 6:4 v/v); 1HNMR (CDCl3) δ 1.17À1.38
(24H, m), 1.48À1.51 (8H,m), 1.86À2.00 (4H, m), 2.8 (1H, broad
peak), 3.30À3.62 (9H, m). 13CNMR (CDCl3): δ 20.8, 25.9, 26.1, 26.2,
29.1, 29.3, 29.4, 29.5, 29.6, 29.7, 30.1, 30.6, 30.8, 31.1 (t), 62.7, 70.4,
70.8, 71.7, 78.8,108.6 (m), 111.2 (m), 113.8 (m), 115.8 (m), 118.8 (m),
121.3 (m); 19F NMR (CDCl3) δ: À82.2 (3F), À115.5 (2F), À123.0 (2F),
À124.1 (2F), À124.8 (2F), À127.4 (2F).
Compound 8e. Pale oil, 33% yield over two steps. Mass (MALDI)
calculated for [C39H46F34O3 + Na]+: 1231.27963, found 1231.284153.
RF = 0.53 (hexane/ethyl acetate 6:4 v/v); 1HNMR (CDCl3) δ 1.17À1.38
(24H, m), 1.47À1.51 (8H,m), 1.86À2.01 (4H, m), 2.8 (1H, broad
peak), 3.33À3.62 (9H, m). 13CNMR (CDCl3): δ 20.2, 26.2, 29.2, 29.3,
29.4, 29.5, 29.6, 29.7, 30.2, 30.7, 31.0, 31.2, 63.1, 70.5, 71.0, 71.9, 78.7,
106.4 (m), 108.5 (m), 110.7 (m), 113.6 (m), 115.8 (m), 118.7 (m),
121.2 (m); 19F NMR (CDCl3) δ: À81.8 (3F), À115.2 (2F), À122.5 (2F),
À122.6 (4F), À123.9 (2F), À124.4 (2F), À127.0 (2F).
Compound 7a. Pale oil, 85% yield. Mass (MALDI): calcd for
[C28H24F26O3 + Na]+ 925.120 25, found 925.124 52. Rf = 0.72 (hexane/
ethyl acetate 7/3 v/v). 1HNMR(CDCl3):δ1.81À1.91(4H, m), 2.08À2.26
(4H, m), 3.47À4.53 (9H, m), 4.54 (2H, s), 7.25À7.36 (5H, m). 13C NMR
(CDCl3) δ: 21.0 (d), 27.9 (t), 69.0, 69.8, 70.1, 71.2, 73.5 (2C), 79.0 (2C),
6589
dx.doi.org/10.1021/jo200835y |J. Org. Chem. 2011, 76, 6584–6591

The Journal of Organic Chemistry
ARTICLE
General Procedure for the Synthesis of Amphiphiles
4aÀe. To a solution of PEG (MW = 1100)-mesylate (0.0032 mmol,
1 equiv) and alcohol (0.0048 mmol, 1.5 equiv) was added NaH
(0.015 mmol, 5 equiv). The reaction mixture was refluxed under argon
andthenmonitoredthroughout thecompletedisappearanceofthemethyl
group in the methanesulfonate functionality by NMR (36 h). The mixture
was cooled and filtered and the solvent evaporated. The residue was
purified by silica gel chromatography (dichloromethane/methanol 9/1)
and then by fluoro flash column chromatography (gradient from 100%
water to 100% acetonitrile). Yield: 50% after purification.
M1F13 (33.8 mg) was dissolved in 25 mL of Millipore water. Nine
solutions were prepared with concentrations ranging from 7.59 Â 10À4
to 7.59 Â 10À11 M.
M1diH3F6 (16.9 mg) was dissolved in 25 mL of Millipore water.
Twelve solutionswere prepared with concentrations ranging from 8.92 Â
10À5 to 4.46 Â 10À11 M.
M1diH3F8 (18.5 mg) was dissolved in 25 mL of Millipore water.
Twelve solutionswere prepared with concentrations ranging from 3.57 Â
10À4 to 1.78 Â 10À9 M .
M1diH5F8 (19.2 mg) was dissolved in 25 mL of Millipore water.
Eleven solutions were prepared with concentrations ranging from 8.92 Â
10À5 to 2.43 Â 10À9 M.
Compound 4a. Mass (MALDI): dispersion around 2 079.965 41. Rf =
0.62 (dichloromethane/methanol 9/1 v/v). 1H NMR (CDCl3): δ
1.65À1.90 (4H, m); 2.10À2.27 (4H, m), 3.38 (s, 3H), 3.38À3.57
(9H, m); 3.58À3.83 (m, PEG). 19F NMR (CDCl3): δ À81.3 (3F),
À114.9 (2F), À122.5 (2F), À123.4 (2F), À124.0 (2F), À126.7 (2F).
Compound 4b. Mass (MALDI): dispersion around 2 105.505 89. Rf =
0.62 (dichloromethane/methanol 9/1 v/v). 1H NMR (CDCl3): δ
1.82À1.90 (4H, m); 2.10À2.26 (4H, m), 3.38 (s, 3), 3.38À3.57 (9H, m);
3.58À3.83 (m, PEG). 19F NMR (CDCl3): δ À81.7 (3F), À114.8 (2F),
À122.2 (2F), À122.4 (4F), À123.6 (2F), À125.1 (2F), À126.5 (2F).
Compound 4c. Mass (MALDI): dispersion around 2 249.696 21. Rf =
0.61 (dichloromethane/methanol 9/1 v/v). 1H NMR (CDCl3): δ
1.43À1.49 (4H, m), 1.57À1.66 (8H, m), 2.03À2.12 (4H, m), 3.82 (s, 3H),
3.38À3.57 (9H, m), 3.58À3.83 (m, PEG). 19F NMR (CDCl3): δ À81.7
(3F), À117.3 (2F), À124.7 (2F), À124.7 (4F), À125.6 (2F), À126.4
(2F), À129.4 (2F).
M1diH10F8 (20.4 mg) was dissolved in 25 mL of Millipore water.
Thirteensolutions were preparedwith concentrations ranging from 3.57Â
10À5 to 1.78 Â 10À9 M.
Surface Tension Measurements. A tensiometer equipped with
a circulator for constant temperature control was used to measure
surface tension. A custom round rod made of platinum with a diameter
of 1.034 mm and wetted length of 3.248 mm was used. First the rod was
submerged into absolute alcohol and flame-dried for 4 s. This same
procedure was repeated after 4 min, after which the rod was hung on
the instrument and cooled for 5 min without touching any surface. The
surface tension of Millipore water was measured to make sure that the
vial and rod were fully cleaned before running any polymer solution
measurements. The measurements were started only after the mea-
sured surface tension value of Millipore water remained within 0.2 μm
of the pure water standard value. The surface tension measurements
were run from the less concentrated solution to the more concentrated
solution. The surface tension at each concentration was measured
four times.
Compound 4d. Mass (MALDI): dispersion around 2 257.89399. Rf =
0.62 (dichloromethane/methanol 9/1 v/v). 1H NMR (CDCl3): δ
1.24À1.43 (28H, m); 1.52À1.63 (4H, m), 1.98À2.18 (4H, m), 3.38
(s, 3H), 3.38À3.57 (9H, m), 3.58À3.83 (m, PEG). 19F NMR (CDCl3):
δ À81.2 (3F), À114.2 (2F), À122.4 (2F), À123.0 (2F), À124.0 (2F),
À126.6 (2F).
Nanoemulsion Preparation. A 500 mg portion of polymer was
solubilized in normal saline solution (11.9 mL). Normal saline solution
was made with 0.9% (w/w) of sodium chloride. Sevoflurane (Abbott
Laboratories, Chicago, IL, 3.4 mL) and perfluorooctyl bromide
(SynQuest Laboratories, Inc., Alachua, FL, 1.7 mL) were added to the
polymer solution. The biphasic final total volume was 17 mL. The
mixture was homogenized with a high-speed homogenizer for 1 min at
21 000 rpm at room temperature. The crude emulsion made with the
homogenizer was then continuously homogenized with a microfluidizer
for 1 min under 5000 psi at 15 °C, achieved with a cooling bath. The
resulting nanoemulsions were stored in 15 mL plastic tubes at 4 °C.
Emulsion Particle Size Analysis. The emulsion particle size was
measured by dynamic light scattering. The emulsions were diluted by a
factor of 300 by adding 20 μL of the emulsion to 3.0 mL of Millipore
water. Each particle size analysis was run for 5 min at room temperature
and repeated three times. The particle diameters were analyzed through
Gaussian analysis and reported as volume-weighted average diameters.
Emulsions were stored at 4 °C. The emulsion particle size was analyzed
at days 0, 2, 7, 14, 21, and 28 and then every 2 weeks up to 357 days if no
phase separation occurred. Day 0 is when the nanoemulsions were
initially prepared.
Compound 4e. Mass (MALDI): dispersion around 2 257.893 99. Rf =
0.62 (dichloromethane/methanol 9/1 v/v). 1H NMR (CDCl3): δ
1.20À1.42 (28H, m); 1.54À1.61 (4H, m), 2.02À2.06 (4H, m), 3.38
(s, 3H), 3.38À3.57 (9H, m), 3.58À3.83 (m, PEG). 19F NMR (CDCl3):
δ À81.2 (3F), À114.8 (2F), À122.1 (2F), À122.3 (4F), À123.1 (2F),
À123.6 (2F), À126.5 (2F).
Preparation of Polymer Micelles. Micelles were prepared using
the solvent evaporation method. the polymer was dissolved in 1 mL of
methanol to achieve a final concentration of 1.0 Â 10À3 M. Methanol
was evaporated at 65 °C under reduced pressure on a rotary evaporator
to produce a thin film of polymer. The thin film was rehydrated with
1 mL of Millipore water heated at 65 °C with gentle agitation, and the
flask was rotated at room temperature for 10 min. The sample was then
filtered through 0.45 μm nylon syringe filters to remove any insoluble
precipitate.
Particle Size Analysis of Polymer Micelles. Particle size
amalysis of polymer micelles was performed by using dynamic light
scattering (Zetasizer Nano ZS, Malvern) on the micellar solutions. Each
particle size analysis was run at room temperature and repeated three
times. Data are reported as Z-average diameters.
Critical Aggregate Concentration. All surface tension vs con-
centration plots are reported in the Supporting Information.
Preparation of M5F13 Stock Solution. A 94.7 mg portion was
dissolved in 25 mL of Millipore water. Ten solutions at concentrations
ranging from 6.77 Â 10À4 to 1.2 Â 10À10 M were prepared by serial
dilution of the stock solution. All solutions were transferred to 20 mL
disposable scintillation vials. After each solution was prepared, it was
heated in a water bath at 40 °C for 2 h. After a further 24 h of equilibration
at room temperature, the solutions were used for surface tension
measurements.
’ ASSOCIATED CONTENT
S
Supporting Information. Figures giving plots for CAC
b
determination by surface tension measurements, electron
microscopy procedures and images, and NMR spectra. This
material is available free of charge via the Internet at http://
pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: smecozzi@wisc.edu.
The following concentrations were used for the other polymer surface
tension measurements.
6590
dx.doi.org/10.1021/jo200835y |J. Org. Chem. 2011, 76, 6584–6591

The Journal of Organic Chemistry
ARTICLE
’ ACKNOWLEDGMENT
The work described in this article was supported by NIG grant
NIGMS 079375 to S.M. J.-P.J. acknowledges support from the
Korean Research Foundation (KRF-2008-357-E00065).
’ REFERENCES
(1) Gladysz, J. A.; Curran, D. P.; Horvath, I., T. Handbook of Fluorous
Chemistry; Wiley-VCH: Weinheim, Germany, 2004.
(2) Kirsch, P. Modern Fluoroorganic Chemistry; Wiley-VCH: Weinheim,
Germany, 2004.
(3) Rosen, B. M.; Wilson, C. J.; Wilson, D. A.; Peterca, M.; Imam,
M. R.; Percec, V. Chem. Rev. 2009, 109, 6275–6540.
(4) Kraft, M., P.; Riess, J. G. Chem. Rev. 2009, 109, 1714–1792.
(5) Riess, J. G. Curr. Opin. Coll. Int. Sci. 2009, 14, 294–304.
(6) Sawada, H. Prog. Polym. Sci. 2007, 32, 509–533.
(7) Krafft, M. P.; Riess, J. G. Act. Chim. 2006, 301, 88–92.
(8) Gladysz, John A. Science 2006, 313, 1249–1250.
(9) Krafft, M. P. J. Polym. Sci., Part A 2006, 44, 4251–4258.
(10) Krafft, M. P.; Muller, P.; Maaloum, M.; Schmutz, M.; Goldmann,
M.; Fontaine, P. Self-Assembly 2003, 367–373.
(11) Tomalia, D. A. Nat. Mater. 2003, 2, 711–712.
(12) Amis, E. J.; Hu, N.; Seery, T. A. P.; Hogen-Esch, T. E.; Yassini,
M.; Hwang, F. Adv. Chem. 1996, 248, 278–302.
(13) Krafft, M. P.; Riess, J. G. J. Polym. Sci. 2007, 45, 1185–1198.
(14) Zhou, J.; Luo, N.; Liang, X.; Liu, J. Anesth. Analg. 2006,
102, 129–34.
(15) Hoang, K. C.; Mecozzi, S. Langmuir 2004, 20, 7347–7350.
(16) Fast, J. P.; Perkins, M. G.; Pearce, R. A.; Mecozzi, S. Anesthe-
siology 2008, 109, 651–656.
(17) Hu, Z.-Y.; Luo, N.-F.; Liu, J. Can. J. Anesth. 2009, 56, 115–125.
(18) Chiari, P. C.; Pagel, P. S.; Tanaka, K.; Krolikowski, J. G.;
Ludwig, L. M.; Trillo, R. A.; Puri, N.; Kersten, J. R.; Warltier, D. C.
Anesthesiology 2004, 101, 1160–1166.
(19) Raphael, J.; Lynch, C. Can. J. Anaesth. 2009, 56, 91–5.
(20) Rudolph, U.; Antkowiak, B. Nat. Rev. Neurosci. 2004, 5, 709–20.
(21) Preckel, B.; Bolten, J. Best Pract. Res. Clin. Anaesth. 2005, 19,
331–348.
(22) Johnson, R. A.; Simmons, K. T.; Fast, J. P.; Schroeder, C. A.;
Pearce, R. A.; Albrecht, R. M.; Mecozzi, S. J. Pharm. Sci. 2011, 100,
2685–2692.
(23) Ravily, V.; Gaentzler, S.; Santaella, C.; Vierling, P. Helv. Chim.
Acta 1996, 79, 405–425.
(24) Takai, K.; Takagi, T.; Baba, T.; Kanamori, T. J. Fluor. Chem.
2004, 125, 1959–1964.
(25) Gopala, P.; Andruzzi, L.; Li, X.; Ober, C. K. Macromol. Chem.
Phys. 2002, 203, 1573–1583.
(26) Matsuoka, K.; Yonekawa, A.; Ishii, M.; Honda, C.; Endo, K.;
Moroi, Y.; Abe, Y.; Tamura, T. Colloid Polym. Sci. 2006, 285, 323–330.
(27) Sadtler, V. M.; Giulieri, F.; Krafft, M. P.; Riess, J. G Chem. Eur.
J. 1998, 4, 1952.
6591
dx.doi.org/10.1021/jo200835y |J. Org. Chem. 2011, 76, 6584–6591