The design and synthesis of highly branched and spherically symmetric fluorinated oils and amphiles

Article abstract of DOI:10.1016/j.tet.2007.03.004

A new surfactant design principle, based on concepts borrowed from protein science, is proposed. Using this principle, a class of highly branched and spherically symmetric fluorinated oils and amphiles has been designed and synthesized, for potential appl

Full text of DOI:10.1016/j.tet.2007.03.004

Tetrahedron 63 (2007) 3982–3988
The design and synthesis of highly branched and spherically
symmetric fluorinated oils and amphiles
*
Zhong-Xing Jiang and Y. Bruce Yu
Department of Pharmaceutics and Pharmaceutical Chemistry, College of Pharmacy, University of Utah,
Salt Lake City, Utah, UT 84112, USA
Received 5 January 2007; revised 15 February 2007; accepted 1 March 2007
Available online 6 March 2007
Abstract—A new surfactant design principle, based on concepts borrowed from protein science, is proposed. Using this principle, a class of
highly branched and spherically symmetric fluorinated oils and amphiles has been designed and synthesized, for potential applications in the
construction of fluorocarbon nanoparticles. The Mitsunobu reaction was employed as the key step for introducing three perfluoro-tert-butoxyl
groups into pentaerythritol derivatives with excellent yields and extremely simple isolation procedures. Due to the symmetric arrangement of
the fluorine atoms, each fluorinated oil or amphile molecule gives one sharp singlet 19F NMR signal.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
core and the outer shell of the nanoparticles, respectively,
the fluorocarbon moiety of F-surfactants will nonetheless
be in direct contact with F-oils at the interface of the inner
core and the outer shell). Size and shape matching is
achieved by appending a hydrophilic segment to the fluoro-
carbon moiety of an F-oil, thereby turning it into an F-sur-
factant. This requirement rules out macrocyclic type of
molecules such as perfluoro-15-crown-5.1c This is because
appending a hydrophilic segment to a macrocyclic F-oil
will destroy its cyclic symmetry, thereby rendering the
fluorine atoms in the resulting F-surfactant non-equivalent,
leading to violation of the first requirement.
We are interested in using fluorocarbon liquid nanoparticles,
formulated as microemulsions, as multifunctional drug
delivery vehicles for 19F magnetic resonance image (MRI)
guided targeted drug therapy.1 To this end, we embarked
on an effort to synthesize fluorinated oils (F-oils) and surfac-
tants (F-surfactants) as modules for such nanoparticles. We
choose to synthesize F-oils as well as F-surfactants because
fluorocarbon microemulsions with F-surfactants as emulsi-
fiers are much more stable than those with phospholipids
as emulsifiers.2 We choose to design and synthesize new
F-oils and F-surfactants instead of using existing ones for
two reasons. First, all fluorine atoms in an F-oil or an F-
surfactant molecule should have identical chemical environ-
ment with no adjacent hydrogen atoms. This ensures that all
fluorine atoms together will give one singlet 19F signal,
thereby maximizing signal intensity of 19F MRI. This re-
quirement rules out fluorocarbon chain molecules, such as
perfluorooctyl bromide. This requirement also rules out
dendrimer-type molecules that contain fluorocarbon chain
segments.3 Second, the fluorocarbon moieties of F-oils and
F-surfactants should be identical (size and shape matching).
The rationale, based on the molecular theory of the liquid
state,4 is that the ‘cavity’ vacated by an F-oil molecule can
be filled in snuggly by an F-surfactant molecule and vice
versa as a result of size and shape matching (note that
even though F-oils and F-surfactants constitute the inner
Here, we wish to point out that, to this date, surfactants used
in pharmaceutical emulsion formulations have one common
feature: their hydrophobic moieties are chain-like. The ratio-
nale is that chain-like molecules can entanglewith each other
(like arms), leading to stable emulsions. The paradigm of us-
ing chain-like surfactants has been practiced for decades with
no rigorous proof that, as emulsifiers, chain-like surfactants
are indeed superior to highly branched ones. The outcome
of such a practice is dismal: currently, there is only one com-
mercially available injectableemulsion in the US market (Di-
privanÒ).5 It is time to explore new paradigms for surfactant
design. In our approach, the fluorocarbon moieties of F-oils
and F-surfactants are highly branched. The rationale is that
these branches will interdigitate with each other tightly
(like fingers), leading to stable emulsions. This is based on
the observation that branched side chains are often better
than straight side chains at stabilizing protein molecules.
The most convincing case comes from coiled-coils where ex-
haustive comparison of 20 natural amino acids concluded
that leucine and isoleucine, two branched amino acids, are
Keywords: Fluorinated oils; Fluorinated amphiles; Perfluoro-tert-butyl;
Mitsunobu reaction; 19F NMR.
* Corresponding author. Tel.: +1 801 581 5133; fax: +1 801 585 3614;
e-mail: bruce.yu@utah.edu
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.03.004

Z.-X. Jiang, Y. B. Yu / Tetrahedron 63 (2007) 3982–3988
3983
(F3C)3CO
OC(CF3)3
(F3C)3CO
(F3C)3CO
OC(CF3)3
(CF3)3COH, DEAD, PH3P
4A MS, THF, 45 °C
78%
(F3C)3CO
(F3C)3CO
OC(CF3)3
HO
HO
OH
(OCH2CH2)3OCH2Rs
(F3C)3CO
Ro = Me,
Ro
1a
3
1a
1b
1c
1d
Rs = CH2OH,
CO2H,
2a
2b
2c
2d
CH2OBn,
Scheme 1. Synthesis of F-oil 1a.
CH2OH
CH2NH2,
CH2SH
CH2OMe
to form perfluoro-tert-butyl ethers.9 Initially, treating com-
pound 3 with DEAD and Ph3P at 0 ꢀC for 30 min and then
with perfluoro-tert-butanol for 40 h led to no desired product.
Elevating reaction temperature from 0 ꢀC to room tempera-
ture or refluxing after the addition of perfluoro-tert-butanol
also led to no desired product. As perfluoro-tert-butanol is
a volatile substance with a low boiling point (45 ꢀC), the
reaction was then carried out in a sealed vessel at 45 ꢀC for
30 h. To our delight, the reaction proceeded smoothly and
Figure 1. Structures of F-oils (1a–1d) and F-amphiles (2a–2d).
more effective at stabilizing coiled-coils than any other
amino acid.6 Tight interdigitation also occurs in fluorinated
coiled-coils between branches made of the –CF3 group, its
bulkiness notwithstanding.7 The importance of size and
shape matching in optimizing side chain interdigitation was
revealed in a study on protein interior packing.8
Based on these considerations, a class of highly branched F-
oils (1a–1d) and F-amphiles (2a–2d) was designed (Fig. 1).
As the first step toward the engineering of fluorocarbon nano-
particles, the focus of this paper is to propose a new surfactant
design principle based on concepts borrowed from protein
science and to explore the synthesis of F-oils and F-surfac-
tants. Since no surface activity is measured in this work,
we decided to call 2a–2d F-amphiles rather than F-surfac-
tants. These F-amphiles are prototypes of F-surfactants
that will be eventually used in fluorocarbon nanoparticle for-
mulation. They serve the purpose of demonstrating the
synthesis feasibility of amphilic molecules with a highly
branched fluorocarbon moiety. Surface activities and the ef-
fectiveness of using highly branched F-oils and F-surfactants
in microemulsion formulation will be evaluated in future
studies.
˚
ether 1a was isolated with a 67% yield. When 4 A molecular
sieves were added to the reaction, the yield can be improved
to 78%. Interestingly, the resulting F-oil 1a can bevery easily
isolated from the reaction mixture: after adding water to
quench the reaction at room temperature, the fluorinated
ether 1a automatically settled to the bottom of the reaction
vesseldue to separation of fluorous compounds from aqueous
and hydrocarbon phases.10 Thus, the product 1a was col-
lected through simple phase separation. Due to no or low
fluorine content, all the reagents and side products remained
in the reaction mixture.
Our synthesis of F-oil 1b started with the modification of
commercially available pentaerythritol 4 (Scheme 2).11 In
order to selectively protect one of its four hydroxyl groups,
pentaerythritol 4 was first converted to the hydroxylmethyl
orthoester 5. After protecting the remaining hydroxyl group
with the benzyl group, the orthoester intermediate was then
hydrolyzed to give the corresponding pentaerythritol mono-
benzyl ether 6 with good yield on a 50 g scale. F-oil 1b was
then synthesized from 6 with an 85% yield using the same
reaction condition and the same isolation method as that
used for 1a.
In each F-oil or F-amphile molecule, fluorine was intro-
duced by three symmetrically positioned perfluoro-tert-
butoxyl groups with all 27 fluorine atoms residing on the
surface of a sphere to ensure identical chemical environment
for all fluorine atoms and to avoid 19F–19F or 19F–1H cou-
pling. The stem of the hydrophilic moiety of F-amphiles
(2a–2d) is an oligo-ethylene glycol chain. Ethylene glycol
enhances aqueous solubility and is biocompatible. The oligo-
ethylene glycol chain is terminated with various functional
groups (–OH, –COOH, –NH2, and –SH) to demonstrate
synthesis feasibility and to facilitate future derivatization of
these F-amphiles. The fluorocarbon moiety and the hydro-
carbon moiety are connected through a pentaerythritol core
(in the case of the F-oil 1a, a derivative of pentaerythritol
was used). In this work, the hydrophilic moiety of the F-
amphiles is not branched. However, if needed, branching
can certainly be introduced into the hydrophilic moiety
without compromising the spherical symmetry of the fluoro-
carbon moiety.
F-oils 1c and 1d were synthesized sequentially from F-oil 1b
(Scheme 3). Initial attempts to remove the benzyl group in
1b through hydrogenolysis with either palladium on carbon
or palladium hydroxide on carbon as the catalyst in methanol
under a hydrogen atmosphere of 30 bars were unsuccessful.
Two factors might have hampered the hydrogenolysis reac-
tion. Firstly, the benzyl ether is surrounded with three very
bulky perfluoro-tert-butoxyl ethers, posing significant steric
barriers to the catalyst. Secondly, the highly fluorinated 1b
has limited solubility in methanol, the reaction solvent. As
benzyl ether is susceptible to Lewis acid-catalyzed acidol-
ysis, aluminum chloride was then employed to remove the
benzyl group. Treatment of benzyl ether 1b with aluminum
chloride and anisol in dichloromethane gave the anticipated
product 1c with quantitative yield. Compound 1c was also
isolated from the reaction mixture through phase separation.
To our surprise, no side reactions, such as cleavage of the
perfluoro-tert-butyl ether or rearrangement of the pentaery-
thritol derivatives, took place to any appreciable extent dur-
ing the reaction based on the quantitative yield and the 19F
NMR spectrum of isolated 1c (Supplementary data). This
indicates that the perfluoro-tert-butyl ether bond is stable
under acidic conditions. This property will facilitate future
2. Results and discussion
From commercially available 1,1,1-tris(hydroxymethyl)-
ethane 3 and perfluoro-tert-butanol, F-oil 1a was synthesized
(Scheme 1). To introduce three perfluoro-tert-butoxyl groups
into compound 3 in just one step, the Mitsunobu reaction
was employed. This is because the three electron-withdraw-
ing –CF3 groups in perfluoro-tert-butanol can significantly
enhance the acidity of the hydroxyl group. Thus perfluoro-
tert-butanol is a good substrate for the Mitsunobu reaction

3984
Z.-X. Jiang, Y. B. Yu / Tetrahedron 63 (2007) 3982–3988
CH3C(OEt)3,TosOH
Toluene, 80 °C
1. KOH, DMSO, BnBr, rt.
2. HCl(0.01N), MeOH, rt.
57%
O
HO
HO
OH
OH
OH
O
92%
O
4
5
(CF3)3COH, DEAD, PH3P
4A MS, THF, 45 °C
85%
HO
HO
OH
(F3C)3CO
(F3C)3CO
OC(CF3)3
OBn
OBn
6
1b
Scheme 2. Synthesis of F-oil 1b.
AlCl3, Anisol
CH2Cl2, 0 °C
99%
MeI, KH
THF, rt.
93%
(F3C)3CO
(F3C)3CO
OH
OC(CF3)3
(F3C)3CO
(F3C)3CO
OMe
(F3C)3CO
(F3C)3CO
OBn
OC(CF3)3
OC(CF3)3
1b
1c
1d
Scheme 3. Synthesis of F-oils 1c and 1d.
TBDMSCl, imidazol
DMAP, CH2Cl2, 0 °C
MsCl, Et3N
DCM, rt.
99%
HO(CH2CH2O)4H
HO(CH2CH2O)4TBDMS
7
MsO(CH2CH2O)4TBDMS
8
87%
Scheme 4. Modification of tetraethylene glycol.
derivatization and formulation of this class of compounds.
Methylation of alcohol 1c was carried out with iodomethane
and potassium hydride. After quenching the reaction with
water, 1d was isolated by simple phase separation with
a 93% yield. In summary, all F-oils were isolated from reac-
tion mixtures using simple phase separation (For purity, see
NMR spectra of these compounds in Supplementary data).
isolated by column purification with only a 40% yield.
Then a much stronger base, potassium hydride, was em-
ployed as the base for this reaction. Fortunately, ether 9
was isolated with an 82% yield. Removal of the silyl protec-
tive group in compound 9 gave F-amphile 2a as a viscous
liquid with an 87% yield. It is noteworthy that, in contrast
to F-oils 1a–1d, compounds 9 and 2a can neither be phase
separated from the quenched reaction mixture nor be ex-
tracted from the reaction mixture (diluted by 95/5 (v/v) ace-
tonitrile/water after concentrating under vacuum) with FC72
(perfluorohexanes). Instead, compounds 9 and 2a were puri-
fied by flash column chromatography on silica gel.
F-oil 1c serves as the starting material for the synthesis of F-
amphiles 2a–2d. The synthesis started with the modification
of tetraethylene glycol (Scheme 4). Selective protection of
one of the two hydroxyl groups in tetraethylene glycol
with TBDMS gave the alcohol 7 with good yield (87%).
Treatment of the alcohol 7 with methanesulfonyl chloride
and triethylamine afforded the methanesulfonate 8 with
a 99% yield.
F-amphile 2a serves as the starting material for the synthesis
of other F-amphiles (2b–2d). Compound 2b was synthe-
sized as a viscous wax with an 84% yield by Jones oxidation
of the primary alcohol in compound 2a to carboxylic acid
(Scheme 6).
Compound 9 was then synthesized from methanesulfonate 8
and the alcohol 1c (Scheme 5). In order to form the ether
bond in compound 9, the alcohol 1c was treated with sodium
hydride followed by methanesulfonate 8. Product 9 was
From 2a, F-amphile 2c was also synthesized (Scheme 7).
After transforming the hydroxyl group in 2a into the
MsCl, Et3N
KH, THF, rt.
(F3C)3CO
(F3C)3CO
(F3C)3CO
(F3C)3CO
O(CH2CH2O)4Ms
OC(CF3)3
O(CH2CH2O)4H
OC(CF3)3
(F3C)3CO
(F3C)3CO
CH2Cl2, rt.
OH
(F3C)3CO
(F3C)3CO
O(CH2CH2O)4TBDMS
OC(CF3)3
then 8
97%
82%
OC(CF3)3
10
2a
1c
9
88%
TBAF
THF, rt.
NaN3, DMF, 90 °C
87%
Pd/C, H2
(F3C)3CO
(F3C)3CO
(F3C)3CO
(F3C)3CO
(OCH2CH2)4N3
OC(CF3)3
(F3C)3CO
(F3C)3CO
O(CH2CH2O)4H
OC(CF3)3
(OCH2CH2)4NH2
OC(CF3)3
MeOH, rt.
85%
11
2c
2a
Scheme 7. Synthesis of F-amphile 2c.
Scheme 5. Synthesis of F-amphile 2a.
Jones Reagent
Acetone, 0 °C
O(CH2CH2O)4H
OC(CF3)3
(F3C)3CO
(F3C)3CO
O(CH2CH2O)3CH2CO2H
OC(CF3)3
(F3C)3CO
(F3C)3CO
84%
2a
2b
Scheme 6. Synthesis of F-amphile 2b.

Z.-X. Jiang, Y. B. Yu / Tetrahedron 63 (2007) 3982–3988
3985
CH3COSK
DMF, 90 °C
O(CH2CH2O)4Ms
OC(CF3)3
(OCH2CH2)4SCOCH3
OC(CF3)3
(F3C)3CO
(F3C)3CO
(F3C)3CO
(F3C)3CO
83%
12
10
93%
NaOH(aq), MeOH, rt.
(OCH2CH2)4SH
OC(CF3)3
(F3C)3CO
(F3C)3CO
2d
Scheme 8. Synthesis of F-amphile 2d.
methanesulfonate 10 with a 97% yield, 10 was treated with
sodium azide to give the azide 11 with an 88% yield. Hydro-
genolysis of the azido group in compound 11 with palladium
on carbon afforded F-amphile 2c as a viscous liquid with an
85% yield.
3. Conclusion
A class of highly branched and spherically symmetric fluo-
rinated oils and amphiles has been efficiently synthesized.
Three perfluoro-tert-butoxyl groups were simultaneously
introduced into the substrates by the Mitsunobu reaction
(Schemes 1 and 2) with high yields and straightforward iso-
lation procedures (phase separation). As a result of spherical
symmetry, all 27 fluorine atoms in each F-oil or F-amphile
give one sharp singlet 19F NMR signal.
Finally, F-amphile 2d was synthesized from 10 (Scheme 8).
Treatment of the methanesulfonate 10 with potassium thiol-
acetate afforded the thiolester 12 with an 83% yield. After
hydrolysis of the thiolester 12 with sodium hydroxide, F-
amphile 2d was obtained as a viscous liquid with a 93%
yield.
4. Experimental
19F NMR spectra of F-oils and F-amphiles were acquired.
As designed, each of the F-oils and F-amphiles gave only
one sharp singlet 19F NMR signal with a peak width around
0.03 ppm (Fig. 2).12 Hence, from the 19F signal intensity
standpoint, these F-oils and F-amphiles are ideal for 19F
magnetic resonance imaging applications.
4.1. F-oil 1a
To a stirred suspension of 1,1,1-tris(hydroxymethyl)-
ethane 3 (6.0 g, 50.0 mmol), triphenylphosphine (59.0 g,
˚
225.1 mmol) and 4 A molecular sieves (6 g) in tetrahydro-
furan (300 mL) at 0 ꢀC was added dropwise diethylazodi-
carboxylate (39.2 g, 225.1 mmol). Afterward, the reaction
mixture was allowed to warm to room temperature and was
stirred for an additional 20 min. Then perfluoro-tert-butanol
(53.2 g, 225.1 mmol) was added in one portion and the result-
ing mixture was stirred for 30 h at 45 ꢀC in a sealed vessel.
Water (30 mL) was added to the reaction mixture and stirred
for an additional 10 min. Then the mixture was transferred
to a separatory funnel and the lower phase was collected.
Removal of the perfluoro-tert-butanol under vacuum gave
the product 1a as a clear oil (30.3 g, 78%). 1H NMR
(400 MHz, neat) d 3.95 (s, 6H), 1.05 (s, 3H); 19F NMR
(376 MHz, neat) d ꢁ71.25 (s); 13C NMR (100.7 MHz,
neat) d 120.2 (q, J¼291.8 Hz), 79.7 (m), 69.0, 41.3, 14.3;
MS (CI) m/z 759 (M+–Me, 24); HRMS (MALDI-TOF) calcd
for C17H10F27O3 775.0199, found 775.0193.
4.2. Alcohol 5
To a suspension of pentaerythritol 4 (68.0 g, 0.5 mol) in
toluene (50 mL) were added triethyl orthoacetate (81.0 g,
92 mL, 0.5 mol) and p-toluenesulfonic acid monohydrate
(300 mg). Ethanol was distilled from the mixture at 80 ꢀC
overnight. After all the ethanol had been distilled away,
the bath temperature was raised to 125 ꢀC and toluene was
distilled off until the solution was homogeneous. The residue
was purified by column chromatography on neutral alumi-
num oxide (CH2Cl2/MeOH¼10/1) to give alcohol 5 as
1
Figure 2. 19F NMR (376 MHz) spectra of F-oils (1a–1d) and F-amphiles
(2a–2d).
a white solid (73.6 g, 92%). H NMR (400 MHz, CDCl3)
d 4.00 (s, 6H), 3.44 (s, 2H), 1.44 (s, 3H).

3986
Z.-X. Jiang, Y. B. Yu / Tetrahedron 63 (2007) 3982–3988
4.3. Triol 6
4.7. Alcohol 7
Powdered KOH (123.2 g, 2.2 mol) was added to stirred
dimethyl sulfoxide (750 mL) and the resulting mixture
was stirred at room temperature for 10 min. Alcohol 5
(73.6 g, 460.0 mol) was added, followed quickly by benzyl
bromide (94.7 g, 65.9 mL, 554.0 mmol). The reaction mix-
ture was stirred for 2 h and then diluted with water
(3000 mL) and extracted with diethyl ether. The combined
extract was washed with brine and water, dried with
magnesium sulfate, and concentrated to afford the 4-benzyl-
oxymethyl-1-methyl-2,6,7-trioxa-bicyclo[2.2.2]octane inter-
mediate as a white solid. The intermediate was dissolved in
methanol (300 mL) and treated with 0.1 N HCl (600 mL).
The resulting mixture was stirred at room temperature for
4 h, treated with sodium bicarbonate (42.5 g, 506.0 mmol),
stirred for an additional 1 h, and concentrated under vac-
uum. The residue was purified by flash chromatography
on silica gel (CH2Cl2/methanol¼10/1) to give the triol 6
as a viscous oil (59.3 g, 57%). 1H NMR (400 MHz,
CDCl3) d 7.26–7.33 (m, 5H), 4.46 (s, 2H), 3.64 (s, 6H),
3.39 (s, 2H).
To a stirred solution of tetraethylene glycol (9.7 g,
50.0 mmol), imidazol (10.2 g, 150 mmol), and 4-dimethyl-
aminopyridine (3.0 g) in CH2Cl2/dimethylformamide
(450 mL/50 mL) at 0 ꢀC was added dropwise tert-butyl-
chlorodimethylsilane (9.6 g, 50 mmol) over 3 h. The result-
ing mixture was stirred at room temperature overnight and
quenched with water (300 mL). The organic phase was
collected and the aqueous phasewas extracted with ethyl ace-
tate. The combined organic phase was dried over anhydrous
magnesium sulfate. After concentrating under vacuum, the
residue was purified by flash chromatography on silica gel
(CH2Cl2/methanol¼10/1) to give alcohol 7 as a clear oil
1
(13.4 g, 87%). H NMR (400 MHz, CDCl3) d 3.50–3.70
(m, 16H), 0.83 (s, 9H), 0.01 (s, 6H).
4.8. Methanesulfonate 8
To a stirred solution of alcohol 7 (6.2 g, 20 mmol) and pyr-
idine (4.8 g, 60 mmol) in CH2Cl2 (200 mL) at 0 ꢀC was
added methanesulfonyl chloride (5.5 g, 24 mmol). The
resulting mixture was stirred at room temperature for 2 h
and quenched with water (100 mL). The organic phase
was collected and the aqueous phase was extracted with
ethyl acetate. The combined organic phase was dried over
anhydrous magnesium sulfate. After concentrating under
vacuum, the residue was purified by flash chromatography
on silica gel (n-hexane/ethyl acetate¼1/1) to give methane-
sulfonate 8 as a clear oil (7.7 g, 99%). 1H NMR (400 MHz,
CDCl3) d 4.29–4.32 (m, 2H), 3.68–3.71 (m, 4H), 3.57–3.60
(m, 8H), 3.47–3.49 (m, 2H), 3.01 (s, 3H), 0.83 (s, 9H), 0.00
(s, 6H).
4.4. F-oil 1b
Same procedure as described for the synthesis of 1a, clear
1
oil, yield 85%. H NMR (400 MHz, CDCl3) d 7.25–7.35
(m, 5H), 4.47 (s, 2H), 4.08 (s, 6H), 3.45 (s, 2H); 19F NMR
(376 MHz, CDCl3) d ꢁ73.47 (s); 13C NMR (100.7 MHz,
CDCl3) d 137.3, 128.6, 128.1, 128.0, 120.3 (q, J¼
292.5 Hz), 79.7 (m), 73.9, 65.6, 65.5, 46.4; MS (CI) m/z
881 (M++1, 100), 880 (M+); HRMS (CI) calcd for
C24H15F27O4 880.0539, found 880.0515.
4.5. F-oil 1c
4.9. Compound 9
To a stirred solution of 1b (29.3 g, 33.2 mmol) and anisol
(14.4 g, 132.9 mmol) in CH2Cl2 (500 mL) at 0 ꢀC was added
aluminum chloride powder (13.3 g, 99.7 mmol) slowly. The
resulting mixture was stirred at 0 ꢀC for 1 h and then water
(100 mL) was added slowly. The lower layer was collected
as a clear oil of 1c (25.9 g, 99%). 1H NMR (400 MHz, ace-
tone-d6) d 4.27 (s, 6H), 3.74 (s, 2H); 19F NMR (376 MHz,
acetone-d6) d ꢁ71.20 (s); 13C NMR (100.7 MHz, acetone-
d6) d 121.5 (q, J¼291.8 Hz), 80.9 (m), 67.5, 58.7, 47.9;
MS (CI) m/z 791 (M++1, 100); HRMS (CI) calcd for
C17H10F27O4 791.0150, found 791.0131.
To a suspension of alcohol 1c (15.8 g, 20.0 mmol) in tetra-
hydrofuran (100 mL) at 0 ꢀC was added KH (4.8 g, 25% in
mineral oil, 30 mmol) and the resulting mixture was stirred
at 0 ꢀC for an additional 10 min. Methanesulfonate 8
(11.6 g, 30 mmol) was added at 0 ꢀC and the resulting mix-
ture was stirred overnight at room temperature. After adding
slowly 100 mL of 2 N HCl aqueous solution, the mixture
was extracted with ethyl acetate and the combined organic
phase was dried over anhydrous magnesium sulfate. After
solvent removal under vacuum, the residue was purified
by flash chromatography on silica gel (n-hexane/ethyl
acetate¼1/1) to give compound 9 as a clear oil (17.7 g,
1
4.6. F-oil 1d
82%). H NMR (400 MHz, CDCl3) d 4.01 (s, 6H), 3.71 (t,
J¼5.6 Hz, 2H), 3.49–3.60 (m, 14H), 3.41 (s, 2H), 0.83 (s,
To a stirred solution of alcohol 1c (7.9 g, 10.0 mmol) in tet-
rahydrofuran (50 mL) at 0 ꢀC was added KH (2.4 g, 25% in
mineral oil, 15 mmol). The resulting mixture was stirred at
0 ꢀC for 10 min and then iodomethane (4.26 g, 30 mmol)
was added. After stirring overnight at room temperature,
water (100 mL) was added slowly to the mixture. The lower
layer was collected as a clear oil of 1d (7.5 g, 93%). 1H NMR
(400 MHz, acetone-d6) d 4.19 (s, 6H), 3.42 (s, 2H), 3.34
(s, 3H); 19F NMR (376 MHz, acetone-d6) d ꢁ71.30 (s);
13C NMR (100.7 MHz, acetone-d6) d 121.3 (q, J¼
292.6 Hz), 76.9 (m), 68.1, 66.6, 59.2, 47.4; HRMS
(MALDI-TOF) calcd for C18H11F27NaO4 827.0124, found
827.0118.
9H), 0.00 (s, 6H); 19F NMR (376 MHz, CDCl3) d ꢁ73.19
(s); 13C NMR (100.7 MHz, CDCl3)
d 120.2 (q,
J¼292.5 Hz), 79.7 (m), 72.7, 70.8, 70.75, 70.7, 70.6, 70.3,
66.4, 65.5, 62.7, 46.2, 25.7, 18.2, ꢁ0.4, ꢁ5.6; HRMS
(MALDI-TOF) calcd for C31H40F27O8Si 1081.2063, found
1081.2058.
4.10. F-amphile 2a
Tetrabutylammonium fluoride (12 mL, 1 N solution in THF)
was added dropwise to a stirred solution of compound 9
(10.8 g, 10.0 mmol) in tetrahydrofuran (100 mL) at 0 ꢀC.
The reaction mixture was stirred at room temperature for

Z.-X. Jiang, Y. B. Yu / Tetrahedron 63 (2007) 3982–3988
3987
4 h and quenched with water (200 mL). After extracting the
reaction mixture with ethyl acetate, the combined organic
phase was dried over anhydrous magnesium sulfate. After
concentrating under vacuum, the residue was purified by
flash chromatography on silica gel (n-hexane/ethyl
acetate¼1/1) to give alcohol 2a as a clear viscous liquid
(11.1 g, 87%). 1H NMR (400 MHz, CDCl3) d 3.99 (s, 6H),
3.53–3.60 (m, 16H), 3.39 (s, 2H); 19F NMR (376 MHz,
CDCl3) d ꢁ73.47 (s); 13C NMR (100.7 MHz, CDCl3)
d 120.2 (q, J¼292.6 Hz), 79.7 (m), 72.5, 70.7, 70.6, 70.5,
70.32, 70.3, 66.4, 65.9, 65.5, 61.7, 46.2; MS (CI) m/z 967
(M++1, 100); HRMS (CI) calcd for C25H26F27O8
967.1198, found 967.1174.
the residue was purified by flash chromatography on silica
gel (n-hexane/ethyl acetate¼1/1) to give azide 11 as a clear
oil (9.9 g, 88%). 1H NMR (400 MHz, CDCl3) d 4.05 (s, 6H),
3.56–3.68 (m, 14H), 3.44 (s, 2H), 3.38 (t, J¼5.2 Hz, 2H);
19F NMR (376 MHz, CDCl3) d ꢁ73.57 (s); 13C NMR
(100.7 MHz, CDCl3) d 120.3 (q, J¼292.5 Hz), 79.8 (m),
71.0, 70.9, 70.87, 70.85, 70.8, 70.5, 70.3, 66.6, 65.7, 50.9,
46.2; MS (CI) m/z 992 (M++1, 100); HRMS (CI) calcd for
C25H25F27N3O7 992.1261, found 992.1263.
4.14. F-amphile 2c
A suspension of palladium on carbon (1.0 g, 10% Pd) in
methanol (100 mL) was stirred under vacuum for 5 min
and then stirred under an atmosphere of hydrogen for
10 min. A solution of azide 11 (9.9 g, 9.9 mmol) was added
to the suspension and the resulting mixture was stirred under
an atmosphere of hydrogen overnight. After filtration, the
solvent was removed under vacuum and the residue was pu-
rified by flash chromatography on silica gel (CH2Cl2/meth-
anol¼1/10) to give amine 2c as a clear viscous liquid
4.11. F-amphile 2b
To a stirred solution of compound 2a (3.87 g, 4.0 mmol) in
acetone (40 mL) at 0 ꢀC was added dropwise a solution of
Jones reagent (8 mL). After the addition, the resulting mix-
ture was stirred at room temperature for 1 h and then iso-
propanol (10 mL) was added slowly to the reaction mixture
at 0 ꢀC. After solvent removal under vacuum, the residue
was dissolved in water and extracted with ether. The com-
bined organic phase was dried over anhydrous magnesium
sulfate and concentrated under vacuum. The residue was pu-
rified by flash chromatography on silica gel (n-hexane/ethyl
acetate¼1/3) to give the acid 2b as a viscous wax (3.3 g,
1
(8.1 g, 85%). H NMR (400 MHz, CDCl3) d 3.98 (s, 6H),
3.37–3.56 (m, 16H), 2.79 (m, 2H); 19F NMR (376 MHz,
CDCl3) d ꢁ73.35 (s); 13C NMR (100.7 MHz, CDCl3)
d 119.7 (q, J¼293.3 Hz), 78.9 (m), 72.8, 70.2, 70.0, 69.7,
69.6, 65.8, 65.0, 45.7, 41.1; MS (CI) m/z 966 (M++1, 100);
HRMS (CI) calcd for C25H27F27NO7 966.1356, found
966.1353.
1
84%). H NMR (400 MHz, CD3OD) d 4.14 (s, 6H), 4.08
(s, 2H), 3.57–3.69 (m, 12H), 3.50 (s, 2H); 19F NMR
(376 MHz, CD3OD) d ꢁ71.23 (s); 13C NMR (100.7 MHz,
CD3OD) d 174.9, 121.6 (q, J¼292.5 Hz), 80.8 (m), 72.0,
71.6, 71.57, 71.55, 71.5, 71.4, 69.7, 67.6, 67.2, 47.5; MS
(CI) m/z 981 (M++1, 100); HRMS (CI) calcd for
C25H24F27O9 981.0991, found 981.0989.
4.15. Thiolester 12
To a stirred solution of methanesulfonate 10 (7.5 g,
7.0 mmol) in dimethylformamide (50 mL) was added potas-
sium thiolacetate (1.7 g, 15.0 mmol) and the resulting mix-
ture was stirred at 90 ꢀC overnight. After solvent removal,
the residue was purified by flash chromatography on silica
gel (n-hexane/ethyl acetate¼1/1) to give thiolester 12 as
a clear oil (5.9 g, 83%). 1H NMR (400 MHz, CDCl3)
d 4.03 (s, 6H), 3.54–3.60 (m, 14H), 3.42 (s, 2H), 3.06 (t,
J¼6.8 Hz, 2H), 2.29 (s, 3H); 19F NMR (376 MHz, CDCl3)
d ꢁ73.33 (s); 13C NMR (100.7 MHz, CDCl3) d 195.4,
120.1 (q, J¼292.6 Hz), 79.6 (m), 70.7, 70.6, 70.54, 70.5,
70.3, 69.8, 66.3, 65.5, 46.2, 30.3, 28.7; MS (MALDI-TOF)
m/z 1047 (M++Na, 100); HRMS (MALDI-TOF) calcd for
C27H27F27NaO8S 1047.0893, found 1047.0920.
4.12. Methanesulfonate 10
To a stirred solution of compound 2a (11.6 g, 12 mmol) and
triethylamine (7.3 g, 72 mmol) in CH2Cl2 (200 mL) at 0 ꢀC
was added methanesulfonyl chloride (4.2 g, 36 mmol). The
resulting mixture was stirred at room temperature overnight
and quenched with water (100 mL). The organic phase was
collected and the aqueous phase was extracted with ethyl
acetate. The combined organic phase was dried over anhy-
drous magnesium sulfate. After concentrating under vac-
uum, the residue was purified by flash chromatography on
silica gel (n-hexane/ethyl acetate¼1/1) to give methanesul-
fonate 10 as a clear oil (12.0 g, 97%). 1H NMR (400 MHz,
CDCl3) d 4.35 (m, 2H), 4.04 (s, 6H), 3.73–3.76 (m, 2H),
3.53–3.67 (m, 12H), 3.43 (s, 2H), 3.04 (s, 3H); 19F NMR
(376 MHz, CDCl3) d ꢁ73.37 (s); 13C NMR (100.7 MHz,
CDCl3) d 120.1 (q, J¼292.6 Hz), 79.6 (m), 70.7, 70.6,
70.5, 70.2, 69.2, 69.1, 69.0, 66.3, 65.4, 63.5, 46.2, 37.5;
MS (MALDI-TOF) m/z 1067 (M++Na, 100); HRMS
(MALDI-TOF) calcd for C26H27F27O10SNa 1067.0791,
found 1067.0786.
4.16. F-amphile 2d
To a solution of thiolester 12 (4.7 g, 4.6 mmol) in methanol
(30 mL) was added 1 N sodium hydroxide solution (15 mL)
and the resulting mixture was stirred at room temperature for
4 h. The reaction mixture was acidified with 2 N HCl aque-
ous solution (8 mL) and the solvent was then removed under
vacuum. The residue was purified by flash chromatography
on silica gel (n-hexane/ethyl acetate¼1/1) to give thiol
2d as a clear viscous liquid (4.2 g, 93%). 1H NMR
(400 MHz, CDCl3) d 4.02 (s, 6H), 3.51–3.59 (m, 14H),
3.41 (s, 2H), 3.65 (m, 2H); 19F NMR (376 MHz, CDCl3)
d ꢁ73.36 (s); 13C NMR (100.7 MHz, CDCl3) d 120.1 (q,
J¼293.3 Hz), 79.6 (m), 72.9, 70.8, 70.6, 70.5, 70.3, 70.3,
70.2, 66.3, 65.5, 46.2, 24.2; MS (CI) m/z 983 (M++1, 100);
HRMS (CI) calcd for C25H26F27O7S 983.0968, found
983.0971.
4.13. Azide 11
To a stirred solution of methanesulfonate 10 (11.8 g,
11.3 mmol) in dimethylformamide (60 mL) was added
sodium azide (1.6 g, 25.0 mmol) and the resulting mixture
was stirred at 90 ꢀC overnight. After removal of solvent,

3988
Z.-X. Jiang, Y. B. Yu / Tetrahedron 63 (2007) 3982–3988
3. (a) Caminade, A.-M.; Turrin, C.-O.; Sutra, P.; Majoral, J.-P.
Acknowledgements
Curr. Opin. Coll. Interface Sci. 2003, 8, 282–295; (b) Percec,
V.; Glodde, M.; Johansson, G.; Balagurusamy, V. S. K.;
Heiney, P. A. Angew. Chem., Int. Ed. 2003, 42, 4338–
4342.
This research was supported by the National Institutes of
Health under grants EB002880 and EB004416.
4. Chandler, D. Annu. Rev. Phys. Chem. 1978, 29, 441–
471.
Supplementary data
5. Strickley, R. G. Pharm. Res. 2004, 21, 201–230.
6. Tripet, B.; Wagschal, K.; lavigne, P.; Mant, C. T.; Hodges, R. S.
J. Mol. Biol. 2000, 300, 377–402.
Supplementary data associated with this article can be found
in the online version, at doi:10.1016/j.tet.2007.03.004.
7. Tang, Y.; Ghirlanda, G.; Vaidehi, N.; Kua, J.; Mainz, D. T.;
Goddard, W. A.; DeGrado, W. F.; Tirrell, D. A. Biochemistry
2001, 40, 2790–2796.
References and notes
8. Yu, Y. B.; Lavigne, P.; Privalov, P. L.; Hodges, R. S. J. Am.
Chem. Soc. 1999, 121, 8443–8449.
9. (a) Sebesta, D. P.; O’Rourke, S. S.; Pieken, W. A. J. Org. Chem.
1. (a) Yu, Y. B. J. Drug Target. 2006, 14, 663–669; (b) Lanza,
G. M.; Wickline, S. A. Prog. Cardiovasc. Dis. 2001, 44, 13–
31; (c) Morawski, A. M.; Winter, P. M.; Yu, X.; Fuhrhop,
R. W.; Scott, M. J.; Hockett, F.; Robertson, J. D.; Gaffney,
P. J.; Lanza, G. M.; Wickline, S. A. Magn. Reson. Med. 2004,
52, 1255–1262; Other excellent reviews on the biomedical ap-
plications of fluorocarbon microemulsions include: (d) Krafft,
M. P. Adv. Drug Del. Rev. 2001, 47, 209–228; (e) Riess, J. G.
Tetrahedron 2002, 58, 4113–4131; (f) Kuznetsova, I. N.
Pharm. Chem. J. 2003, 37, 415–420.
ꢀ
€
1996, 61, 361–362; (b) Szabo, D.; Bonto, A.-M.; Kovesdi, I.;
ꢀ
€ €
ꢀ
Gomory, A.; Rabai, J. J. Fluorine Chem. 2005, 126, 641–
652; (c) Jiang, Z.-X.; Yu, Y. B. J. Org. Chem. 2007, 72,
ꢀ
1464–1467; (d) Rabai, J.; Balint, A.-M.; Szıjjarto, C.; Szabo,
D. QSAR Comb. Sci. 2006, 25, 761–765.
ꢀ
ꢀ ꢀ ꢀ
ꢀ
10. Handbook of Fluorous Chemistry; Gladysz, J. A., Curran, D. P.,
ꢀ
Horvath, I. T., Eds.; Wiley-VCH: Weinheim, 2003.
11. Dunn, T. J.; Neumann, W. L.; Rogic, M. M.; Woulfe, S. R.
J. Org. Chem. 1990, 55, 6368–6373.
2. (a) Postel, M.; Riess, J. G.; Weers, J. G. Art. Cells, Blood Subs.
Immob. Biotech. 1994, 22, 991–1005; (b) Riess, J. G.; Krafft,
M. P. Chem. Phys. Lipids 1995, 75, 1–14; (c) Riess, J. G.
Chem. Rev. 2001, 101, 2797–2919.
12. Solvents for 19F NMR: 1a (neat), 1b (CDCl3), 1c (acetone-d6),
1d (acetone-d6) 2a (CDCl3), 2b (CD3OD), 2c (CDCl3), 2d
(CDCl3). Chemical shifts are in parts per million (ppm).