Synthesis and properties of fluorous arenes and triaryl phosphorus compounds with branched fluoroalkyl moieties ("split pony tails")

Article abstract of DOI:10.1016/S0022-1139(03)00171-4

Most compounds designed for immobilization in fluorous media feature linear pony tails of the formula (CH₂) _m(CF₂)_n-1CF₃ [(CH₂)_mR_fn]. This paper presents a first-gener

Full text of DOI:10.1016/S0022-1139(03)00171-4

Journal of Fluorine Chemistry 124 (2003) 45–54
Synthesis and properties of fluorous arenes and triaryl phosphorus
compounds with branched fluoroalkyl moieties (‘‘split pony tails’’)
*
Marc Wende, Florian Seidel, J.A. Gladysz
Institut f u¨ r Organische Chemie, Friedrich-Alexander-Universit a¨ t Erlangen-N u¨ rnberg, Henkestraße 42, Erlangen 91054, Germany
Received 28 April 2003; received in revised form 19 June 2003; accepted 23 June 2003
Abstract
Most compounds designed for immobilization in fluorous media feature linear pony tails of the formula (CH ) (CF ) CF [(CH ) R ].
2
m
2 nꢀ1
3
2 m fn
This paper presents a first-generation approach to compounds with branched or ‘‘split’’ pony tails of the formula (CH
tri(n-butyl)tin is reacted twice with perfluorooctyl iodide (Rf8I; first, photochemical, 78–81%; second, thermal with radical initiator, 71%; 13–
2 l 2 m fn 2
) CH[(CH ) R ] . Allyl
1
yields the branched alkene H
8 g scales) to give the secondary alkyl iodide ICH(CH R ) (3). A subsequent Ni(Cl) (PPh ) -catalyzed reaction with allyl tri(n-butyl)tin
2 f8 2 2 3 2
2
C¼CHCH
2
CH(CH
2
R
f8
)
CH(CH
2
(74%). A palladium-catalyzed Heck coupling with O¼P(p-C
6
H
4
Br)
3
gives the
2 3
(CH ) -
fluorous phosphine oxide O¼P(p-C
6
H
4
CH¼CHCH
2
2
R
f8
)
2
)
CH(CH R ) ) (>99%). Reduction with SiHCl gives P(p-C H (CH ) CH(CH R ) ) , which is protected as the air-stable borane adduct
3
(84%), and Pd/C-catalyzed hydrogenation affords O¼P(p-C
6
H
4
2
f8 2 3
3
6
4
2 3
2 f8 2 3
H
3
BÁP(p-C
CH
with C
2003 Elsevier B.V. All rights reserved.
6
H
4
(CH
2
)
3
CH(CH
2
R
f8
)
2
)
3
(9, 64%). The CF
3
C
6
F
11/toluene partition coefficient of 9 is much higher than that of the analog with p-
at 175–250 8C. However, a CuBr-catalyzed reaction
2 f8 2
R ) , which also exhibits a high partition coefficient (97.9:2.1).
(
2
)
3
R
f8 groups (96.6:3.4 versus 37.3:62.7). The iodide 3 is unreactive towards PAr
MgBr gives C CH(CH
3
6
H
5
6
H
5
#
Keywords: Fluorous; Tin; Radical substitution; Radical addition; Heck reaction; Phosphine-boranes
1
. Introduction
‘‘tuning element’’, the length of which modulates the elec-
tron-withdrawing effect of the perfluoroalkyl groups on the
reaction center [6]. Not surprisingly, a variety of transition-
metal-containing fluorous catalysts has been prepared,
[4] and many of these have been based upon fluorous
phosphines. There are also many reactions catalyzed by
phosphines alone [7], and the applicability of fluorous
phosphines to such processes has been demonstrated [8].
Fluorous phosphines have further been used to prepare
Wittig reagents that give easily recycled phosphine oxides
[9].
Over the last decade, many new catalysts with high
affinities for fluorous solvents have been synthesized [1–4].
This has been prompted by the rapid growth, following the
initial report of Horv a´ th and R a´ bai [1] of ‘‘fluorous biphase
catalysis’’. As most often practiced, this protocol exploits
the markedly temperature-dependent miscibilities of organic
and fluorous solvents [5]. Most combinations give two
phases at room temperature, as illustrated in Scheme 1.
However, with moderate heating, one phase is obtained.
Reactions can be catalyzed under monophasic conditions at
the high temperature limit, and the products and catalyst
are separated under biphasic conditions at the low tempera-
ture limit. Most fluorous solvents in current use are saturated
perfluorocarbons.
Triaryl phosphines offer unique properties, and several
fluorous versions have been developed. The first examples
1
involved simple para-R substituents [10–12]. These phos-
f6
phines were much less basic than non-fluorous analogs due
to the lack of (CH ) segments, and displayed only modest
2
m
High catalyst affinities for fluorous phases can be achieved
byattaching‘‘ponytails’’oftheformula(CH ) (CF )
fluorous phase affinities. Aryl-containing compounds are
difficult to render highly fluorophilic, and studies have
shown that at least two R pony tails per ring are necessary
CF3,
2
m
2 n ꢀ 1
abbreviated (CH ) R so that n represents the number of
2
m
fn
f8
fluorinated carbons. The (CH ) segment can be viewed as a
2
[13]. Both issues were nicely addressed by van Koten,
Deelman, and co-workers [14,15], who used para-silicon
substituents as anchors for one, two, or three pony tails of
m
*
Corresponding author. Tel.: þ49-9131-852-2540;
fax: þ49-9131-852-6865.
1
E-mail address: john.gladysz@organik.uni-erlangen.de (J.A. Gladysz).
For Rf8 analogs, see [10] and [11].
0022-1139/$ – see front matter # 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0022-1139(03)00171-4

46
M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
Scheme 1. A common protocol for fluorous biphase catalysis and catalyst recycling.
the formula (CH ) R (n ¼ 6, 8, 10) [16]. Others have des-
A terminal alkene with a substituent of the formula
(CH ) CH[(CH ) R ] was sought. We established earlier
that the photolysis of commercially available perfluorooctyl
2
2 fn
cribed systems with alkoxide aryl substituents O(CH ) R
2
m
fn
2 l
2 m fn 2
(m=n ¼ 1=7, 3/8) [17].
Although the silicon-substituted fluorous triaryl phosphi-
iodide (R I) and allyl tri(n-butyl)tin gives the alkene
f8
nes have proved to havewide utility in catalysis [15], there has
remained a demand for well ‘‘insulated’’, highly fluorophilic,
non-heteroatom-based systems. Several groups have synthe-
sized phosphines with one(CH ) R or (CH ) R substituent
per aryl ring [12,18–22]. However, in every case analogous
routes to more fluorophilic systems with two such substitu-
tents have been problematic. This includes our own efforts
R CH CH¼CH (2), as shown in Scheme 2 [27]. This free
f8
2
2
radical chain substitution can be conducted on 13–15 g scales
and reliably gives 78–81% yields [27,28]. A subsequent
thermal reaction of 2 and additional R I was conducted in
2
2
fn
2 3 fn
f8
0
the presence of the free radical initiator VAZO (1,1 -azo-
bis(cyclohexanecarbonitrile)). The addition product, second-
ary iodide ICH(CH R ) (3), was obtained in 71% yield after
f8 2
2
3
with reactions of PCl and magnesium or lithium derivatives
3
recrystallization (Scheme 2). Compound 3 could easily be
prepared on 13–18 g scales, and like many alkyl iodides,
solutions were moderately light sensitive.
of the fluorous aryl iodides IC H ((CH ) R ) [13c,23].
6
3
2 3 f8 2
Therefore, we sought to develop routes to arenes and triaryl
phosphines with branched pony tails of the general formula
As illustrated in Scheme 2, 3 was then refluxed with allyl
tri(n-butyl)tin in the presence of a Ni(II) catalyst. This Stille-
like but possibly radical cross-coupling [30] gave the
target alkene H C¼CHCH CH(CH R ) (4) in 74% yield
(
CH ) CH[(CH ) R ] . Such substituents offer ‘‘two pony
2 l 2 m fn 2
tails for the price of one’’, and we refer to them as ‘‘split pony
tails’’. In this paper, we describe our successful first-genera-
tion route to such compounds. There are scattered reports of
other types of fluorous molecules with branched pony tails
2
2
2 f8 2
on multigram scales. Compounds 3 and 4 were characterized
by NMR ( H, 13C), mass spectrometry, and microanalysis,
as summarized in Section 4. All features were routine, and
fully supported the assigned structures. Compound 2 exhib-
ited good solubility in both fluorous and non-fluorous sol-
vents. Compounds 3 and 4 showed lower solubilities in non-
1
[
24], and these are summarized in Section 3.
2
. Results
fluorous solvents (e.g. moderate in CDCl ), but dissolved to
3
2
.1. Syntheses of triaryl phosphorus compounds
appreciable extents in acetone, ether, and THF. When hot
hexane solutions of 3 were cooled, gels were reproducibly
obtained. The flasks could be inverted without loss of
material. More polar solvents also gave gels, although
seemingly with less entrained solvent.
The synthesis of fluorous triaryl phosphines can be
approached by phosphorus–carbon bond disconnection, or
carbon–carbon bond disconnection. Most of the above
examples have been prepared by reactions of PCl with aryl
3
As shown in Scheme 3, the phosphine oxide 1 and alkene
4 (3.3 equivalents) were reacted under Heck conditions
similar to those reported by Xiao and co-workers [20].
The Herrmann–Beller dimeric palladacycle catalyst 5
(Scheme 4) was employed (1.8 mol% or 3.6:100 Pd/1). A
mixed DMF–CF C H solvent system was used to ensure
magnesium or lithium compounds. However, in an impor-
tant development, Xiao and co-workers established the
viability of three-fold Heck reactions of the phosphine oxide
O¼P(p-C H Br) (1) and a variety of terminal alkenes,
6
4
3
including cases with perfluoroalkyl substituents [20]. In
3
6 5
view of our mixed experience with the PCl strategy, we
3
that all reactants were at least partially dissolved. Workup
gave the unsaturated fluorous phosphine oxide O¼P(p-
C H CH¼CHCH CH(CH R ) ) (6) in 84% yield. In
set out to test the efficacy of the Xiao protocol for the
introduction of split pony tails. Accordingly, 1 was prepared
from p-dibromobenzene in a one-flask lithiation–PCl addi-
6
4
2
2 f8 2 3
3
accord with Xiao’s precedent, only the all-trans C¼C isomer
was detected. A subsequent reaction with hydrogen (1 atm,
3 d) in the presence of 10% Pd/C afforded the saturated
phosphine oxide O¼P(p-C H (CH ) CH(CH R ) ) (7) in
tion–oxidation sequence in 38% yield following previously
2
established methods [25,26].
6
4
2 3
2 f8 2 3
2
Since full details have not been reported for these steps, we give our
3
procedure in Section 4. No effort was made to optimize the yield of 1.
We thank L.J. Alvey for a preliminary experiment.

M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
47
hν
R I
+
f8
SnBu3
Rf8
2
(78-81%)
Rf8 = (CF ) CF
R I,
2 7
3
f8
VAZO
110 ˚C
SnBu3
0 mol% Ni(Cl) (PPh )
Rf8
Rf8
Rf8
Rf8
1
2
3
toluene/CF C H
3
6
5
I
4
reflux
3
(74%)
(71%)
Scheme 2. Syntheses of building blocks with split pony tails.
>99% yield. Wilkinson’s catalyst effected only partial con-
version under comparable conditions.
nucleophilic phosphine P(p-tol) (72 h, 250 8C) as well as
3
P(OEt) (22–60 h, 160 8C) gave similar results, and further
3
The phosphine oxide 7 was reduced using an excess of a
standard reagent, HSiCl . Best results were obtained with a
experiments and observations are detailed elsewhere [28].
Additional carbon–carbon bond forming reactions of 3
were investigated. Our attention was drawn to copper-cat-
alyzed cross-couplings of Grignard reagents that had been
applied to fluorous secondary alkyl iodides [32]. As shown
in Scheme 5, 3 and C H MgBr reacted in the presence of
3
two-stage addition (room temperature and 110 8C).
Although the resulting fluorous triaryl phosphine P(p-
C H (CH ) CH(CH R ) ) (8) should not, by analogy to
6
4
2 3
2 f8 2 3
4
related systems [22], be highly air sensitive, it was none-
theless protected in situ as the borane adduct H BÁP(p-
6
5
CuBr (6 mol%) in THF to give the fluorous arene
C H CH(CH R ) (10) in high yield. However, purification
proved challenging. In our hands, the removal of residual 3
3
C H (CH ) CH(CH R ) ) (9). Such borane derivatives
f8 2 3
6
4
2 3
2
6
5
2 f8 2
are usually more convenient to handle, and can often be
converted to metal complexes without prior deprotection
and the byproduct R (CH ) R could not be achieved by
f8
2 3 f8
[
31]. A chromatographic workup gave 9 as a yellow wax in
4% yield. All new phosphorus-containing compounds were
recrystallization or conventional column chromatography.
However, chromatography on fluorous reverse phase silica
gel [33] provided pure 10 in 63% yield. The solubility
properties of 10 were similar to those of 4.
6
characterized by H, 13C, and 31P NMR. Although 9 gave an
acceptable microanalysis, 6 and 7 did not. Compound 9 was
very slightly soluble in hexane. Both 9 and 7 were essentially
insoluble in other organic solvents, and only moderately
soluble in fluorous solvents.
1
An analogous sequence was attempted with the Grignard
reagent p-BrC H MgBr, which was generated from p-
BrC H Br according to literature procedures [34]. Although
6
4
6
4
the corresponding product p-BrC H CH(CH R ) (11)
2 f8 2
6
4
2
.2. Other experiments
formed, conversions were lower than with 10 (ca. 80% by
1H NMR). Upon scale-up, conversions decreased. Despite
intensive efforts, analytically pure 11 could not be isolated.
5
In the course of the preceding investigations, other reac-
tions that might provide convenient routes to compounds
with split pony tails were examined. Since the secondary
iodide 3 is easily prepared in quantity, various functionali-
Attempts to effect bromine–lithium or bromine–magnesium
exchange with crude 11, and subsequent reactions with PCl3
are described elsewhere [28]. A variety of other carbon–
carbon bond-forming cross-coupling reactions were also
investigated, many of which are similarly described else-
where [28]. Of these, a Heck reaction of phenyl iodide and
the alkene 4 was the most promising.
zations were attempted. For example, PPh and the fluorous
3
primary alkyl iodides R (CH ) I react to give phosphonium
2 2
fn
salts that can be applied in Wittig reactions [13a]. However,
these displacements require more forcing conditions (DMF,
1
05–110 8C) than with non-fluorous iodides. Interestingly,
The CF C F /toluene partition coefficients of the bor-
3
6 11
PPh and 3 were recovered unchanged after 50 h at 175 8C
without solvent in a sealed tube (Scheme 5). The more
ane–phosphine adduct 9 and arene 10 were measured by 19
F
NMR and GC as described in Section 4. The results from
two independent runs (9, 96.4:3.6 and 96.9:3.1; 10, 98.0:2.0
3
4
Any air or other sensitivity of fluorous phosphines should be viewed in
the context of their molecular weights. That of 8 (3021.7) is more than 10
5
The maximum purity after fluorous reverse phase chromatography was
times that of PPh (262.3). Therefore, per unit mass of dissolved triaryl
phosphine, a much higher fraction of the former can be decomposed
by a given concentration of a reactive impurity in the solvent.
3
95%, as assayed by 1H NMR (d, CF
JHH ¼ 8:3 Hz, 2H of C
(quint, JHH ¼ 6:8 Hz, C
3
C
6
F
5
/CDCl
), 7.17 (d, JHH ¼ 8:3 Hz, 2H of C
CH), 2.52 (m, 2CH ).
3
, 1:4 (v/v)): 7.46 (d,
6
H
4
6 4
H
), 3.60
6
H
4
2

48
M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
Rf8
(12) was treated with H B under similar conditions to give
3
the adduct H BÁP(p-C H (CH ) R ) (13). Fluorous arenes
3
6
4
2 3 f8 3
O P
Br
+
3
of the formula C H ((CH ) R ) (14) provide comparisons
6
4
2 3 f8 2
3
for 10. The trends are analyzed in Section 3.
1
4
Rf8
KOAc, palladacycle 5 (cat.)
DMF / CF C H , 125 ˚C
3. Discussion
3
6
5
3.1. Merits of syntheses
O P
6
(84%)
As shown in Schemes 2 and 3, the synthesis of the
protected fluorous triaryl phosphine 9 entails seven steps
Rf8
Rf8
Rf8
from commercially available materials. The overall yield
from R I (Scheme 2) is 22%. Apart from the free phosphine
Rf8
3
f8
8
, all compounds are stable in air for extended periods. Only
H , Pd/C,
2
EtOH / CF C H
the reduction of the phosphine oxide 7 has not yet been
tested on a multigram scale, and there would seem to be little
chance for complications. There is a conspicuous reliance on
radical-based carbon–carbon bond forming reactions in
Scheme 2. The Heck reaction in Scheme 3, which completes
the carbon skeleton, nicely illustrates the versatility of the
methodology developed by Xiao and co-workers [20].
To our knowledge, 8 represents the only fluorous triaryl
phosphine with more than one Rfn segment of appreciable
length per ring, outside of the silicon-containing systems of
van Koten and co-workers [14,15]. Three of their com-
pounds that are closely related to 8 are depicted in Fig. 2
3
6 5
O P
7
(
>99%)
Rf8
3
SiHCl , Et N,
3
3
CF C H , reflux
3
6 5
P
8
(15a–c). Different synthetic approaches to 15a–c were also
investigated [14a]. One involving a three-fold lithiation of
(
not isolated)
the phosphine P(p-C H Br) , followed by addition of the
6
4
3
appropriate silyl halide, was distinctly superior to routes
based upon PCl and aryl magnesium or lithium compounds.
Rf8
3
3
Thus, there appears to be a general problem associated with
either the generation or reactivity of high-fluorous-content
aryl magnesium and lithium species, the exact nature of
which can at present only be speculated about [28].
For this reason, we chose not to investigate the halogena-
tion of the fluorous arene 10 (Scheme 5). Although this
otherwise attractive building block might have provided pure
H B·THF
3
H B
P
9
3
(
64%)
Rf8
11 or the corresponding iodide, these are not in our view
Rf8
3
promising precursors to triaryl phosphorus compounds. We
note in passing that several fluorous triaryl phosphites with
more than one pony tail per ring have been reported, but some
of these syntheses are not very efficient [35,36].
Scheme 3. Syntheses of triaryl phosphorus compounds with split pony tails.
and 97.7:2.3) were averaged. Thesevalues are listed in Fig. 1,
together with those for selected reference molecules. In order
to provide a good comparison for 9, the previously char-
acterized fluorous triaryl phosphine P(p-C H (CH ) R )
3.2. Other fluorous compounds with branched pony tails
6
4
2 3 f8 3
The triaryl phosphines 15b and c can be viewed as having
doubly- and triply-branched pony tails, respectively. These
involve a silicon ‘‘stem’’ or branch point. Similar dipho-
sphinesorfluorousdppe(Ar PCH CH PAr ) derivativeshave
2
2
2
2
been synthesized [14b]. Curran and co-workers has reported
the fluorous triaryl phosphine 16 (Fig. 2), which features
a triple branch point within the six-carbon perfluoroalkyl
segment [19]. Such moieties are conveniently constructed
via the addition of CsF to the alkene (CF ) C¼CFCF CF .
Scheme 4. Structure of palladacycle 5.
3
2
2
3

M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
49
Rf8
Rf8
R
MgBr
I
R
6
mol% CuBr, THF
Rf8
Rf8
3
10, R = H (63%)
1, R = Br
crude material,
up to 95% purity)
1
(
Ar P,
3
175-200 ˚C
Rf8
Ar P+
X
3
-
I
Rf8
Scheme 5. Other reactions of 3.
Thisgeneratesa tertiaryperfluoroalkylcesiumcompoundthat
reacts with benzyl halides.
acetate [24d]. These have been used as protecting groups
for carboxylic acids, and many partition coefficients of the
resulting esters have been measured. The tertiary alcohol
21 has been prepared in 40% yield by the addition of
R (CH ) MgI to dimethyl carbonate, and elaborated into
the triply-branched benzylic bromide 22 (Fig. 2) [24c]. This,
as well as the silicon analog 17, have been used to prepare
chiral fluorous b-amino alcohols, one of which exhibited a
high partition coefficient (97:3, FC-72/toluene).
Curran et al. has also described a convenient synthesis of
the fluorous benzylic bromide 17 (Fig. 2), and applied it as a
protecting group for carbohydrates [16]. Suitable disacchar-
ides derivatives with three such moieties showed no tendency
to partition into organic phases. Vincent has synthesized the
fluorous carboxylic acid 18 via a malonate ester synthesis,
and the corresponding dicopper tetracarboxylate 19 [24a,b].
The former can be used to extract transition metal ions into
fluorous solvents, and the latter pyridines.
f6
2 2
3.3. Physical properties
The tertiary alcohols 20a and b, which feature carbon
branch points, have been prepared in 70–72% yields by
additions of the Grignard reagents R (CH ) MgI to ethyl
One obvious question is the degree to which branching
within pony tails affects solubilities. In some cases, molecules
fn
2 2
3 6
Fig. 1. Summary of partition coefficients (CF C F11/toluene, 25 8C).

50
M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
CF3
CH C-CF CF CF
P
SiMe3-n(CH CH R )
P
2
2
f8 n
2
2
2
3
CF3
3
3
1
5 n = a, 1 (68.8 : 31.2)
b, 2 (88.6 : 11.4)
16 (43:57, FC-72/toluene)
c, 3 (67.7 : 32.2)
R
(
CF C F11/toluene)
3
6
Rf6
O
O
R
O
Rf8
O
Cu
O
Cu
O
COOH
Si
O
R
O
Br
Rf6
Rf8
1
7
18 (97.1 : 2.9,
CF C F11/toluene)
Rf6
3
6
R
1
9, RCO - = 18-
2
(
>99 : <1, several
solvent systems)
Rf6
Rfn
Rf6
OH
Rf6
OH
Rf6
Rfn
Br
Rf6
CH3
2
0
21
22
Rf6
n = a, 6
b, 8
(and meta
isomer)
Fig. 2. Some other fluorous compounds with branched pony tails, with partition coefficients were available.
with a high fluorous content are poorly soluble in nearly
all solvents at room temperature [8]. For some applications,
this is advantageous. It would not be surprising if branching
were to enhance solubilities. However, among the species
given above, there are not yet enough data or suitable model
compounds for comparisons.
the aryl groups in the borane adducts 9 and 13 (Fig. 1)
feature pony tails of the formulae (C H )(R ) and
6
11
f8 2
(C H )R , respectively. Not surprisingly, the former exhi-
f8
3
6
bits a much higher partition coefficient (96.6:3.4 versus
37.6:62.7). Since 13 has a much lower partition coefficient
than the corresponding phosphine 12 (66.6:33.4), the parti-
tion coefficient of the unprotected phosphine 8 can be
estimated as ꢁ99:ꢂ1. Thus, 8 constitutes the first highly
fluorophilic triaryl phosphine that lacks a skeletal heteroa-
tom.
With regard to fluorophilicities, the most obvious question
is whether a doubly branched pony tail has the same affect as
two linear pony tails of the same aggregate formula. As
shown in Fig. 1, the arene 10 exhibits a somewhat higher
CF C F /toluene partition coefficient than the two-pony-
When available, partition coefficients for the compounds
with branched pony tails in Fig. 2 are given. The phosphines
15 exhibit an anomalous trend [14a]. The partition coeffi-
cient for 15c, which has three pony tails per aryl ring, is
lower than that of 15b, which has only two. This represents,
to our knowledge, the only case where an increase in the
number of pony tails results in a decrease in fluorophilicity.
The corresponding fluorous dppe (Ar PCH CH PAr ) sys-
3
6 11
tailed arenes 14 (97.9:2.1 versus ca. 91:9). However, the
latter contain three extra methylene groups. When the usual
affect of such a perturbation is considered—for example,
with fluorous pyridines [13b]—the values become approxi-
mately equal. Unfortunately, no constitutionally isomeric
compounds are yet available for comparison.
Another question is how the fluorophilicity of a molecule
with a branched pony tail compares to one with one linear
pony tail that has half the number of atoms. For example,
2
2
2
2
tems show the expected monotonic increase (28.6:71.4,
92.3:7.7, >98:<2) [14b].

M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
51
3
.4. Conclusion
by rotary evaporation and oil pump vacuum to yield a white
solid (7.830 g). The crude product was dissolved in a
mixture of acetone (41 ml), chloroform (17 ml), and ethyl
ether (35 ml). Then hexane (40 ml) was added, and the
sample was stored in the refrigerator. After 14 h, the white
precipitate was collected by filtration. The filtrate was
concentrated to ca. 100 ml, and a second crop was similarly
precipitated. The combined crops were dried under oil pump
vacuum to give 1 as a white powder (3.231 g, 6.274 mmol,
38%), mp 178.3 8C (DSC; lit 179–180 8C) [25].
The highly fluorophilic triaryl phosphine 8, which fea-
tures a branched or ‘‘split’’ pony tail on each ring, has been
synthesized and fully characterized as its protected borane
adduct 9. The convenient seven-step procedure features air
stable intermediates that can be prepared on large scales.
Several of these intermediates hold promise as building
blocks for additional types of compounds with branched
pony tails. Other fluorous compounds with branched pony
tails are becoming available (Fig. 2), and this architectural
motif is certain to see increasing use in fluorous chemistry.
1
NMR (d, CDCl ): H 7.65 (m, 6H of 3 C H ), 7.52 (m, 6H
3
1
6 4
1
3
6
of 3 C H ); C{ H} 133.3 (d, J ¼ 10:6 Hz), 132.0 (d,
CP
6
4
JCP ¼ 12:8 Hz), 130.6 (d, JCP ¼ 106:0 Hz, i-C H ), 127.7
6
4
3
1
1
(
d, JCP ¼ 3:2 Hz, p-C H ); P{ H} 27.8 (s). MS (positive
6
4
þ þ
4
. Experimental
FAB, 3-NBA, m/z): 1030 ([2M] , 15%), 515 (M , 100%),
359 ([M–C H Br] , 18%). Calculated for C H Br PO: C,
41.98; H, 2.35. Found: C, 42.05; H, 2.47.
þ
6
4
18 12
3
4
.1. General
Reactions were conducted under inert atmospheres in
4.3. ICH(CH R ) (3)
2 f8 2
solvents that were distilled as follows: toluene, ethyl ether,
THF, and Et N, from Na and benzophenone; CH Cl and
DMF, from CaH ; CF C H and CF C F , from P O .
A Fisher–Porter bottle was charged with R CH CH¼
3
2
2
f8
2
2
3
6
5
3
6
11
2
5
CH (2; [27] 12.000 g, 26.077 mmol), R I (14.237 g,
2 f8
Workups were conducted under aerobic conditions with
distilled reagent grade solvents. Perfluorooctyl iodide
26.077 mmol), and VAZO (0.510 g, 2.09 mmol), briefly
evacuated and refilled with nitrogen (3ꢃ), and pressurized
with nitrogen (3.8 bar). The mixture was stirred at 110 8C
for 4 h and allowed to cool. The bottle was vented and the
off-white solid dissolved in refluxing hexane (ca. 75 ml).
The solution was cooled to 0 8C and gelled. The gel was
suction filtered and dried by oil pump vacuum to give 3
as soft, white flakes (18.604 g, 18.491 mmol, 71%), mp
55.7 8C (DSC) [29]. In some cases, purification required
a second gel precipitation from acetone. Although solid 3
is light-stable, solutions show decomposition after several
hours.
(
9
R I; Lancaster, 97%), allyl tri(n-butyl)tin (Lancaster,
f8
0
7%), 1,1 -azobis(cyclohexanecarbonitrile) (VAZO; Fluka,
97%), palladacycle 5 (Strem, ꢁ98%), p-BrC H Br (Lan-
2 3 2
ꢁ
6
4
caster, 97%), Ni(Cl) (PPh ) (ABCR, 99%), KF (Acros,
9%), MgSO4 (Riedel-de Ha e¨ n, extra pure), CH CO K
3 2
9
(
in H O, reagent grade), 10% Pd/C (Acros), SiHCl (Aldrich,
reagent grade), Mg turnings (reagent grade), H O (30%
2 2
2
3
ꢁ
99%), CuBr (Aldrich, 98%), H BÁTHF (Acros, 1.0 M in
3
THF), and H BÁSMe (Acros, 2.0 M in THF) were used as
3 2
received. Solutions of n-BuLi (Acros, 1.6 M in hexane) [37]
and C H MgBr (Fluka, 0.8 M in THF) were standardized.
1
NMR (d, CF C F /CDCl , 1:4 (v/v)): H 4.63 (quint,
6
5
3
6
5
3
1
3
1
Most instrumental procedures were described earlier [8b].
DSC measurements were recorded using a Mettler-Toledo
DSC-821 instrument [38].
JHH ¼ 6:6 Hz, CHI), 3.01 (m, 2 CH ); C{ H} (partial)
2
36.6 (t, J ¼ 23:0 Hz, CH ), ꢀ1.2 (s, ICH). MS (positive
CF
2
þ
þ
FAB, 3-NBA, m/z): 1005 (M , 20%), 987 ([M–F] , 18%),
þ þ
79 ([M–I] , 25%), 573 ([C F CH CHI] , 10%). Calcu-
8 17 2
8
4
.2. O¼P(p-C H Br) (1)
lated for C H F I: C, 22.68; H, 0.50. Found: C, 22.99; H,
19 5 34
6
4
3
0.37.
ASchlenkflask,equippedwithathermometerandaseptum,
was charged with p-BrC H Br (11.796 g, 50.000 mmol) and
6
4
4.4. H C¼CHCH CH(CH R ) (4)
2
2
2 f8 2
THF (50 ml) [25]. The mixture was cooled to ꢀ80 8C (ace-
tone and CO ). A solution of n-BuLi (33.1 mL, 1.51 M in
2
A Schlenk flask, equipped with a septum and a condenser,
was charged with 3 (5.025 g, 5.000 mmol), allyl tri(n-butyl)-
tin (4.65 ml, 4.96 g, 15.0 mmol), Ni(Cl) (PPh ) (0.327 g,
hexane, 50.0 mmol) was slowly added, keeping the inner
temperature below ꢀ75 8C. A white suspension formed. A
2
3 2
solution of PCl (1.44 ml, 2.27 g, 16.5 mmol) in THF
3
0.500 mmol), toluene (15 ml), and CF C H (5 ml). The
3 6 5
(
25 ml) was added dropwise over 2 h. The suspension turned
mixture was kept at 120 8C for 24 h. The initially deep red
solution decolorized and a gray solid precipitated. Then KF
(2.905 g, 50.00 mmol) was added. The mixture was stirred
at room temperature for 24 h. The dark gray solid was
removed by filtration and washed with CF C H (50 ml).
to a grape color and was allowed to warm to room tem-
perature overnight. Then, H O (6% in H O, 50 ml; prepared
from a 30% solution) was added. The mixture was stirred
for 30 min, and the phases were allowed to separate. The
aqueous phase was extracted with ethyl ether (3 ꢃ 50 ml).
The organic phases were combined, washed with water
2
2
2
3
6 5
6
The ipso, ortho, meta, and para carbons are designated with reference
to the phosphorus atom.
(
3 ꢃ 30 ml) and dried (MgSO ). The volatiles were removed
4

52
M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
The combined organic phases were washed with water
(
by rotary evaporation and oil pump vacuum. The oil or greasy
crude product was washed (with shaking) with acetonitrile
and ethanol (10 ml), purged with hydrogen, and fitted with
a balloon filled with hydrogen. The suspension was vigor-
ously stirred for 3 days. The solid was allowed to settle, and
the supernatant removed using a filter syringe. The residue
was analogously extracted with CF C H (2 ꢃ 20 ml). The
2 ꢃ 15 ml) and dried (MgSO ). The volatiles were removed
4
(
2 ꢃ 20 ml) and methanol (2 ꢃ 20 ml). If needed, the mix-
3
6 5
tures can be cooled to freeze the lower fluorous product and
facilitate solvent decantation. Drying by oil pump vacuum
gave 4 as a white grease (3.395 g, 3.690 mmol, 74%), which
was pure enough for further use. Kugelrohr distillation
organic phases were combined and the volatiles removed
by rotary evaporation and oil pump vacuum to give 7 as a
yellow gum (0.100 g, 0.0330 mmol, >99%).
1
NMR (d, CF C F /CDCl , 4:1 (v/v)): H 7.59 (m, 6H
3
6
5
3
(
wax.
110 8C, 7 ꢃ 10ꢀ6 bar) gave 4 as an analytically pure white
of 3C H ), 7.28 (m, 6H of 3C H ), 2.82 (t, J ¼ 7:0 Hz,
6
4
6
4
HH
3C H CH ), 2.56 (m, 3CH), 2.29 (m, 6CH R ), 1.81 (m,
6 4 2 2 f8
1
13
1
NMR (d, CF C F /CDCl , 1:4 (v/v)): H 5.74 (m, ¼CH),
3CH CH CH ), 1.77 (m, 3CH CH CH ); C{ H} (partial)
2 2 2 2 2 2
3
6
5
3
5
2
1
2
.17–5.12 (m, H C¼), 2.53 (m, CH), 2.34 (m, ¼CHCH ),
(see footnotes 6–8) 132.6 (d, JCP ¼ 10:0 Hz, o-C H ),
6 4
2
2
1
3
1
7
.22 (m, 2CH R ); C{ H} (partial) 133.8 (s, ¼CH),
128.7 (d, JCP ¼ 12:7 Hz, m-C H ), 35.9 (s, C H CH ),
2 f8 2
2
f8
6 4 6 4 2
19.2 (s, H C¼), 39.1 (s, ¼CHCH ), 33.9 (t, JCF ¼
34.6 (overlapping m and s, CH R , CH CH), 27.8
(s, CH CH CH ), 25.9 (s, CHCH R ); P{ H} 30.0 (s).
2 2 2 2 f8
2
2
3
1
1
1:1 Hz, CH R ), 25.3 (s, CH). MS (positive FAB,
2
f8
þ
þ
3
-NBA, m/z): 919 (M , 65%), 487 ([M–CH C F ] ,
17
Calculated for C H F102PO: C, 33.20; H, 1.49. Found: C,
84 45
34.12; H, 1.76.
2
8
1
C, 28.73; H, 1.18.
00%). Calculated for C H F : C, 28.71; H, 1.10. Found:
2
2 10 34
4
.7. H BÁP(p-C H (CH ) CH(CH R ) ) (9)
3
6
4
2 3
2 f8 2 3
4
.5. O¼P(p-C H CH¼CHCH CH(CH R ) ) (6)
6
4
2
2 f8 2 3
A Schlenk flask, equipped with a septum and a condenser,
was charged with 7 (0.149 g, 0.0491 mmol), Et N (0.070 ml,
A Schlenk flask, equipped with a condenser, was charged
with 4 (1.104 g, 1.200 mmol), 1 (0.170 g, 0.330 mmol),
palladacycle (0.0056 g, 0.0060 mmol), CH CO K
3
0.50 mmol), and CF C H (5 ml). Then SiHCl (0.0510 ml,
3
6
5
3
5
0.0677 g, 0.500 mmol) was added via syringe over 20 min
with stirring. After an additional hour, the mixture was kept
3
2
(
(
0.1081 g, 1.320 mmol), CF C H (3 ml), and DMF
3 6 5
6 ml). The mixture was kept at 125 8C for 24 h. It turned
at 110 8C for 4 h. A second portion of SiHCl (0.0510 ml,
3
dark brown, and was allowed to cool. The volatiles were
removed by oil pump vacuum. The solid was extracted with
CF C H (50 ml). The yellow solution was washed with
0.0677 g, 0.500 mmol) was added. After an additional 15 h
at 110 8C, the mixture was cooled to ambient temperature,
and the excess SiHCl was carefully removed (partial oil
3
3
6
5
water (2 ꢃ 10 ml) and dried (MgSO ). The solvent was
pump vacuum, leaving solvent). Then H BÁTHF (0.100 ml,
4
3
removed by rotary evaporation. The dark brown gum or
oil (1.009 g) was dissolved in a minimum of CF C H .
Hexane was added, and 6 precipitated as a solid or oil,
1.0 M in THF, 0.100 mmol) was slowly added by syringe to
the crude phosphine 8. The mixture was stirred for 18 h, and
water (10 ml) and CF C H (50 ml) were added. The aqu-
3
6 5
3
6 5
which was isolated by decantation and washed with hexane
(0.839 g, 0.277 mmol, 84%). A second crop of 6 could often
be obtained by storing the filtrate in the refrigerator.
eous phase was extracted with CF C H (2 ꢃ 20 ml). The
3
6 5
combined organic phases were dried (MgSO ). The cloudy
4
filtrate was concentrated to ca. 10 ml and centrifuged at
4000 rpm for 10 min. The clear supernatant was decanted
from the beige residue, which was discarded. Silica gel (2 g)
was added to the supernatant. The mixture was dried in
vacuo and loaded on top of a silica gel column (10 cm).
The column was extracted with ethyl acetate/hexane (1:10
(v/v)). The solvent was removed from the extract by rotary
evaporation. The residue was dissolved in CF C F (10 ml).
1
NMR (d, CF C F /CDCl , 4:1 (v/v)): H 7.59 (m, 6H of
3
6
5
3
3
3
C H ), 7.43 (m, 6H of 3C H ), 6.51 (d, JHH ¼ 15:9 Hz,
6
4
6 4
C H CH), 6.23 (m, 3¼CHCH ), 2.63 (m, 3CH), 2.52
6 4 2
13
1
(
(
(
m, 3¼CHCH ), 2.25 (m, 6CH R ); C{ H} (partial)
2
2 f8
8
see footnotes 6 and 7) 133.9 (s, C H CH¼), 132.8
6 4
d, JCP ¼ 10:6 Hz, o-C H ), 128.2 (s, ¼CHCH ), 126.4 (d,
6
4
2
JCP ¼ 12:2 Hz, m-C H ), 38.4 (s, ¼CHCH ), 34.3 (t, J
¼
6
4
2
CF
3 6 11
3
1
1
2
0:6 Hz, CH R ), 26.1 (s, CHCH R ); P{ H} 28.5 (s).
f8
The solution was washed with toluene (2 ꢃ 5 ml), and the
solvent was removed by oil pump vacuum to give 9 as a
yellow wax (0.0958 g, 0.0316 mmol, 64%).
2
2 f8
4
.6. O¼P(p-C H (CH ) CH(CH R ) ) (7)
6
4
2 3
2 f8 2 3
1
NMR (d, CF C F /CDCl , 4:1 (v/v)): H 7.54 (m, 6H of
3
6
5
3
A
.0330 mmol), Pd/C (10%, 0.013 g), CF C H (12 ml),
Schlenk flask was charged with
6
(0.100 g,
3C H ), 7.37 (m, 6H of 3C H ), 2.81 (t, J ¼ 7:0 Hz,
6
4
6
4
HH
0
3C H CH ), 2.60 (m, 3CH), 2.33 (m, 6CH R ), 1.82 (2 m,
6 4 2 2 f8
3
6
5
1
3
1
3
1
CH CH CH ); C{ H} (partial) (see footnotes 6–8)
2 2 2
7
In order to reduce the number of interfering peaks, the CF
3 6 5
C F was
33.7 (d, JCP ¼ 10:3 Hz, o-C H ), 129.0 (d, J ¼ 14:8 Hz,
6
4
CP
replaced by C
6 6
F
for 13C NMR spectra. The non-aromatic carbon signals
m-C H ), 35.9 (s, C H CH ), 34.6 (overlapping m and s,
2
6
4
6
4
were assigned by 1H–13C correlation (HETCOR) experiments. The
aromatic carbon signals of 6, 7, and 9 were assigned by analogy to
those of related fluorous phosphine oxides [22].
CH R ,CH CH),28.0(s,CH CH CH ),26.0(s,CHCH R );
2
f8
2
2
2
2
2 f8
3
1
1
P{ H}20.8(brs).CalculatedforC H BF102P:C,33.22;H,
84 48
8
The signals for the ipso and para carbon atoms were not detected.
1.59. Found: C, 33.71; H, 1.52.

M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
53
4
.8. H BÁP(p-C H (CH ) R ) (13)
4.10. Partition coefficients
3
6
4
2 3 f8 3
A solution of P(p-C H (CH ) R ) (12; 0.0420 g,
6
(A) A 5 ml flask was charged with 9 (0.0368 g,
0.0121 mmol) and CF C F (2.00 ml). After complete
4
2 3 f8 3
0
.0256 mmol) [22] in CF C H (1.5 ml) was treated with
5
3
6
3 6 11
H BÁSMe (2.0 M in THF; 0.0200 ml, 0.0400 mmol). After
dissolution of 9, toluene (2.00 ml) was added and the
mixture was vigorously shaken (20 min). The flask was
kept at 25 8C for 48 h. Then aliquots (each 0.500 ml) were
taken from both phases. The CF C F aliquot was evapo-
3
2
4
8 h, the volatiles were removed by oil pump vacuum. The
white powder was taken up in a mixture of toluene and
CF C F (16 ml, 50:50 (v/v)). The solvent was removed
3
6
11
3 6 11
from the fluorous phase by rotary evaporation and oil pump
vacuum to give 13 as a white solid (0.0141 g, 0.00851 mmol,
rated to dryness. A solution of the internal standard C6F6
(0.0738 g, 0.397 mmol) in CF C H (12.7016 g) was
3
6 5
3
3%), mp 137.0 (DSC).
1
prepared. Portions of this standard solution were added
gravimetrically to the above aliquots (CF C F : 0.6413 g
NMR (d, CDCl ): H 7.49 (m, 6H of 3C H ), 7.24 (m, 6H
3
6
4
3 6 11
of 3C H ), 2.73 (t, JHH ¼ 7:7 Hz, 3C H CH ), 2.07 (m,
solution, 0.0199 mmol C F ; toluene: 0.0596 g solution,
6 6
6
4
6
4
2
13
1
CH R ), 1.94 (m, CH CH CH ); C{ H} (partial) (see
2
0.00185 mmol C F ). Then C D was added (0.05 ml each)
6 6 6 6
f8
2
2
2
footnote 6) 144.2 (s, p-C H ), 133.4 (d, JCP ¼ 9:3 Hz,
and the samples were analyzed by 19F NMR (integration of
CF signal against C F ). The procedure was repeated, giving
6
4
o-C H ), 128.8 (d, JCP ¼ 10:4 Hz, m-C H ), 127.1 (d,
6
4
6
4
3
6 6
J
¼ 58:9 Hz, i-C H ), 34.9 (s, C H CH ), 30.3 (t,
an average partition coefficient of 96.6:3.4 (0.00840 g of 9
in 0.500 ml of CF C F ; 0.000292 g of 9 in 0.500 ml of
toluene). A 2.00/0.500 scale factor gives a total mass recov-
CP
6
4
6
4
2
31
1
JCP ¼ 22:3 Hz, CH R ), 21.6 (s, CH CH CH ); P{ H}
2
f8
2
2
2
3 6 11
1
Found: C, 36.06, H, 1.95.
9.5 (br s). Calculated for C H BF P: C, 36.98; H, 2.01.
5
1
33
51
ery of 0.0348 g (95%). (B) A 5 ml flask was charged with
10 (0.0307 g, 0.0321 mmol) and CF C F (2.00 ml). After
3 6 11
4
.9. C H CH(CH R ) (10)
6
complete dissolution of 10, toluene (2.00 ml) was added and
the mixture was vigorously shaken (20 min). The flask was
kept at 25 8C for 48 h. Then aliquots (each 0.500 ml) were
taken from both phases. A solution of the internal standard
n-undecane (0.0508 g, 0.325 mmol) in CF C H (24.155 g)
5
2 f8 2
A Schlenk flask, equipped with a septum and a condenser,
was charged with 3 (0.503 g, 0.500 mmol), CuBr (0.0040 g,
.028 mmol), and THF (6 ml). The suspension was briefly
refluxed, and a portion of a C H MgBr solution (ca. 0.1 ml
0
3
6 5
was prepared. Portions of this standard solution were
added gravimetrically to the above aliquots (CF C F :
6
5
of 0.625 ml, 0.80 M in THF, 0.500 mmol total) was added
via syringe with vigorous stirring. The resulting yellow color
discharged after a few seconds. The remaining C H MgBr
was added within 5–10 min, and the sample was briefly
refluxed. The orange suspension was stirred at ambient
temperature for 20 h, quenched with saturated aqueous
3
6 11
0.5906 g solution, 0.00798 mmol n-undecane; toluene:
0.5944 g solution, 0.00793 mmol n-undecane). The samples
were analyzed by GC (two injections each). The procedure
was repeated, giving an average partition coefficient of
97.9:2.1.
6
5
NH Cl, and extracted with ethyl ether (3 ꢃ 50 ml). The
4
combined organic phases were dried (MgSO ). The solvent
4
was removed by rotary evaporation. The yellowish-white
solid was dissolved in a minimum of ethyl ether and loaded
on a fluorous reverse phase silica gel column (1:5 cm ꢃ
Acknowledgements
We thank the Deutsche Forschungsgemeinschaft (DFG;
GL 300-3/1) for support.
2
5 cm) [33]. The column was successively eluted with
1
50 ml portions of acetonitrile and THF–acetonitrile mix-
tures (1:1.6, 1:1.5, 1:1.4, and 1:1.3 (v/v)), and finally hexane
250–300 ml). The solvent was removed from the second
(
References
and third fractions by rotary evaporation and oil pump
vacuum to give 10 as a clear oil (0.251 g). The fourth
fraction yielded impure 10 (0.074 g), which could be pur-
ified by a second analogous column (total yield: 0.302 g,
[
1] I.T. Horv a´ th, J. R a´ bai, Science 266 (1994) 72.
2] J.A. Gladysz, D.P. Curran, Tetrahedron 58 (2002) 3823 (and
following articles in this special issue entitled ‘‘Fluorous Chem-
istry’’).
[
0
.316 mmol, 63%). A portion was dissolved in ethyl ether
5 ml). The solvent was allowed to slowly evaporate, and
[
3] (a) I.T. Horv a´ th, Acc. Chem. Res. 31 (1998) 641;
(
white needles of 10 formed. However, the single crystals
(
b) E. de Wolf, G. van Koten, B.-J. Deelman, Chem. Soc. Rev. 28
1999) 37;
c) M. Cavazzini, F. Montanari, G. Pozzi, S. Quici, J. Fluorine
Chem. 94 (1999) 183;
d) E.G. Hope, A.M. Stuart, J. Fluorine Chem. 100 (1999) 75;
(e) J. Yoshida, K. Itami, Chem. Rev. 102 (2002) 3693;
f) A.P. Dobbs, M.R. Kimberly, J. Fluorine Chem. 118 (2002) 3.
(
diffracted poorly.
NMR (d, CF C F /CDCl , 1:4 (v/v)): H ¼ 7.21ꢀ7.36
(
1
3
6
5
3
(
(
CH2); C{ H} (partial) 142.0, 129.1, 127.6, 126.9 (4 s,
m, C H ), 3.60 (quint, J ¼ 6:8 Hz, C H CH), 2.52 (m, 2
6
5
13 1
HH
6 5
(
C H ), 37.0 (t, JCF ¼ 20:8 Hz, CH ), 31.8 (s, C H CH). GC:
6
5
2
6 5
[
4] R.C. da Costa, J.A. Gladysz, in: M. Beller, C. Bolm, (Eds.),
Transition Metals for Organic Synthesis, Wiley–VCH, Weinheim,
2004, in press.
9
1
9.3 area% purity. Calculated for C H F : C, 31.40; H,
25 10 34
.05. Found: C, 31.26; H, 1.52.

54
M. Wende et al. / Journal of Fluorine Chemistry 124 (2003) 45–54
[
[
[
5] L.P. Barthel-Rosa, J.A. Gladysz, Coord. Chem. Rev. 190–192 (1999)
[19] Q. Zhang, Z. Luo, D.P. Curran, J. Org. Chem. 65 (2000) 8866.
587 (survey of practical considerations and underlying physical
[20] (a) W. Chen, L. Xu, J. Xiao, Org. Lett. 2 (2000) 2675;
(b) W. Chen, L. Xu, Y. Hu, A.M. Banet Osuna, J. Xiao, Tetrahedron
58 (2002) 3889.
principles).
6] H. Jiao, S. Le Stang, T. So o´ s, R. Meier, P. Rademacher, K. Kowski,
L. Jafarpour, J.-B. Hamard, S.P. Nolan, J.A. Gladysz, J. Am. Chem.
Soc. 124 (2002) 1516.
[21] S. Darses, M. Pucheault, J.-P. Genet, Eur. J. Org. Chem. 1121 (2001).
[22] T. So o´ s, B.L. Bennett, D. Rutherford, L.P. Barthel-Rosa, J.A.
Gladysz, Organometallics 20 (2001) 3079.
7] (a) G. Buono, O. Chiodi, M. Wills, Syn. Lett. 377 (1999);
(
(
b) X. Lu, C. Zhang, Z. Xu, Acc. Chem. Res. 34 (2001) 535;
c) E. Vedejs, O. Daugulis, J.A. MacKay, E. Rozners, Syn. Lett.
[23] D. Rutherford, B.L. Bennett, L.P. Barthel-Rosa, T. So o´ s, unpublished
results.
1
499 (2001);
d) D. Basavaiah, Chem. Rev. 103 (2003) 811.
8] (a) M. Wende, R. Meier, J.A. Gladysz, J. Am. Chem. Soc. 123
2001) 11490;
b) M. Wende, J.A. Gladysz, J. Am. Chem. Soc. 125 (2003) 5861.
9] A. Galante, P. Lhoste, D. Sinou, Tetrahedron Lett. 42 (2001) 5425.
[24] (a) J. Loiseau, E. Fouquet, R.H. Fish, J.-M. Vincent, J.-B. Verlhac, J.
Fluorine Chem. 108 (2001) 195;
(
[
[
(b) M. El Bakkari, N. McClenaghan, J.-M. Vincent, J. Am. Chem.
Soc. 124 (2002) 12942;
(
(
(c) Y. Nakamura, S. Takeuchi, K. Okumura, Y. Ohgo, Tetrahedron
57 (2001) 5565;
[
10] (a) B. Betzemeier, P. Knochel, Angew. Chem. 109 (1997) 2736;
(d) J. Pardo, A. Cobas, E. Guiti a´ n, L. Castedo, Org. Lett. 3 (2001)
3711.
B. Betzemeier, P. Knochel, Angew. Chem., Int. Ed. Engl. 36 (1997)
2
623;
b) P. Bhattacharyya, D. Gudmunsen, E.G. Hope, R.D.W. Kemmitt,
D.R. Paige, A.M. Stuart, J. Chem. Soc., Perkin Trans. 1 24 (1997)
609.
[25] R. Benassi, M.L. Schenetti, F. Taddei, P. Vivareli, P.J. Dembech, J.
Chem. Soc., Perkin Trans. 2 (1974) 1338.
(
[26] V. Ravindar, H. Hemling, H. Schumann, J. Blum, Synth. Commun.
22 (1992) 841.
3
[
[
11] W. Chen, J. Xiao, Tetrahedron Lett. 41 (2000) 3697.
12] S. Schneider, W. Bannwarth, Angew. Chem., Int. Ed. 39 (2000) 4142;
S. Schneider, W. Bannwarth, Angew. Chem. 112 (2000) 4293.
13] (a) C. Rocaboy, D. Rutherford, B.L. Bennett, J.A. Gladysz, J. Phys.
Org. Chem. 13 (2000) 596;
[27] L.J. Alvey, D. Rutherford, J.J.J. Juliette, J.A. Gladysz, J. Org. Chem.
63 (1998) 6302.
[28] M. Wende, Doctoral Dissertation, Friedrich-Alexander-Universit a¨ t
Erlangen-N u¨ rnberg, 2003.
[
[29] G. Gambaretto, L. Conte, G. Fornasieri, C. Zarantonello, D. Tonei,
A. Sassi, R. Bertani, J. Fluorine Chem. 121 (2003) 57 (this paper
describes an independent, slightly longer synthesis of 3 that appeared
when this manuscript was being reviewed).
(
b) C. Rocaboy, F. Hampel, J.A. Gladysz, J. Org. Chem. 67 (2002)
863;
c) C. Rocaboy, J.A. Gladysz, Chem. Eur. J. 9 (2003) 88.
6
(
[
[
14] (a) B. Richter, E. de Wolf, G. van Koten, B.-J. Deelman, J. Org.
Chem. 65 (2000) 3885;
[30] R. Shimizu, T. Fuchikami, Tetrahedron Lett. 37 (1996) 8405.
[31] (a) N. Brodie, S. Jug e´ , Inorg. Chem. 37 (1998) 2438;
(b) C. Darcel, E.B. Kaloun, R. Merd e` s, D. Moulin, N. Riegel, S.
Thorimbert, J.P. Gen eˆ t, S. Jug e´ , J. Organomet. Chem. 624 (2001)
333;
(b) E. de Wolf, B. Richter, B.-J. Deelman, G. van Koten, J. Org.
Chem. 65 (2000) 5424.
15] (a) B. Richter, A.L. Spek, G. van Koten, B.-J. Deelman, J. Am.
Chem. Soc. 122 (2000) 3945–3951;
(c) M. Ingleson, N.J. Patmore, G.D. Ruggiero, C.G. Frost, M.F.
Mahon, M.C. Willis, A.S. Weller, Organometallics 20 (2001) 4434.
[32] R. Shimizu, E. Yoneda, T. Fuchikami, Tetrahedron Lett. 37 (1996)
5557.
(b) E. de Wolf, E.A. Speets, B.-J. Deelman, G. van Koten,
Organometallics 20 (2001) 3686.
16] D.P. Curran, R. Ferritto, Y. Hua, Tetrahedron Lett. 39 (1998) 4937
[
[
(
for the first report of a fluorous arene with a Si((CH
17] (a) D. Sinou, G. Pozzi, E.G. Hope, A.M. Stuart, Tetrahedron Lett. 40
1999) 849;
b) D. Sinou, D. Maillard, G. Pozzi, Eur. J. Org. Chem. (2002) 269.
18] (a) S. Kainz, D. Koch, W. Baumann, W. Leitner, Angew. Chem. 109
1997) 1699;
2
)
2
R
fn
)
3
moiety).
[33] (a) D.P. Curran, Z. Luo, J. Am. Chem. Soc. 121 (1999) 9069;
(b) D.P. Curran, Syn. Lett. 9 (2001) 1488.
(
[34] S. Kainz, Z. Luo, D.P. Curran, W. Leitner, Synthesis 1425 (1998).
[35] (a) T. Mathivet, E. Monflier, Y. Castanet, A. Mortreux, J.-L.
Couturier, Tetrahedron Lett. 40 (1999) 3885;
(
[
(
(b) T. Mathivet, E. Monflier, Y. Castanet, A. Mortreux, J.-L.
Couturier, Tetrahedron 58 (2002) 3877.
S. Kainz, D. Koch, W. Baumann, W. Leitner, Angew. Chem., Int. Ed.
Engl. 36 (1997) 1628;
[36] S. Le Stang, R. Meier, C. Rocaboy, J.A. Gladysz, J. Fluorine Chem.
119 (2003) 141.
(b) S. Kainz, D. Koch, W. Baumann, W. Leitner, Angew. Chem., Int.
Ed. 36 (1998) 1628;
S. Kainz, D. Koch, W. Baumann, W. Leitner, Angew. Chem. 109
[37] A.F. Burchat, J.M. Chong, N. Nielsen, J. Organomet. Chem. 542
(1997) 281.
(
1997) 1699;
c) S. Kainz, A. Brinkmann, W. Leitner, A. Pfaltz, J. Am. Chem.
Soc. 121 (1999) 6421.
[38] H.K. Cammenga, M. Epple, Angew. Chem. 107 (1995) 1284;
H.K. Cammenga, M. Epple, Angew. Chem., Int. Ed. Engl. 34 (1995)
1171.
(