Synthesis of dendritic oligo(aryl sulfone)s as supports for synthesis

Article abstract of DOI:10.1016/j.tet.2005.09.103

A short, divergent route to G₁ oligo(aryl sulfone)s and a G ₂ oligo(aryl sulfone) dendrimer using nucleophilic aromatic substitution reactions is described. A range of tetrasubstituted pentasulfones are proposed for applications as homogeneous supports for synthesis. Key to achieving selectivity in the syntheses is the activation of leaving groups by sulfide to sulfone oxidation. Preparation of the G₂ oligo(aryl sulfone) is low-yielding due to competition from SET processes that are interesting from a mechanistic point of view. The utility of the supports is exemplified with a four step synthesis of a dipeptide and by 'react and release' synthesis of amides.

Full text of DOI:10.1016/j.tet.2005.09.103

Tetrahedron 61 (2005) 12314–12322
Synthesis of dendritic oligo(aryl sulfone)s as supports for synthesis
a
b
Paul C. Taylor,a, Michael D. Wall and Peter R. Woodward
*
aDepartment of Chemistry, University of Warwick, Coventry CV4 7AL, UK
bGlaxoSmithKline, Medicines Research Centre, Gunnels Wood Rd, Stevenage SG1 2NY, UK
Received 27 May 2005; revised 25 August 2005; accepted 22 September 2005
Available online 25 October 2005
Abstract—A short, divergent route to G1 oligo(aryl sulfone)s and a G2 oligo(aryl sulfone) dendrimer using nucleophilic aromatic
substitution reactions is described. A range of tetrasubstituted pentasulfones are proposed for applications as homogeneous supports for
synthesis. Key to achieving selectivity in the syntheses is the activation of leaving groups by sulfide to sulfone oxidation. Preparation of the
G2 oligo(aryl sulfone) is low-yielding due to competition from SET processes that are interesting from a mechanistic point of view. The
utility of the supports is exemplified with a four step synthesis of a dipeptide and by ‘react and release’ synthesis of amides.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
Dendrimers are soluble polymers that are highly symmetri-
cal and have a large number of surface functional groups.
These two features make dendrimers highly attractive as
supports for synthesis and for catalysis, since they are
readily characterised, for example by conventional NMR
spectroscopy, all the surface functional groups are identical,
which should lead to high selectivities, and high loadings
can be achieved.7 It should be noted that dendrimers are, in
general, much more expensive than their conventional
polymer analogues.
The use of solid-supports is now widespread in organic
synthesis. In 1963, Merrifield introduced cross-linked
polystyrene for the synthesis of oligopeptides.1 Since then,
this approach has been extended to numerous other classes
of compounds, in particular following the advent of parallel
synthesis in 1985.2 The development of solid supported
analogues of homogeneous catalysts has also been an area
of significant activity for a number of years.3 More recently,
solid supported reagents and scavenger resins have become
popular synthetic tools.4 In all these cases, the salient
advantage of the solid supported technology is that
separation of the supported species from products or by-
products in the solution phase is extremely facile.
There has been a good number of reports of dendrimers as
supports for, in particular, catalysis.8 Separation of the
supported catalysts can be achieved by membrane
filtration.9–11 We felt that there was a need for the de
novo design of dendrimers to be used as supports and that
the new dendrimer supports should meet the following
criteria: be stable to a wide range of reaction conditions;
have very simple NMR spectra to allow integration relative
to the supported species; have good solubility in common
organic solvents; be easy to prepare without needing
protecting group strategies.
In the first two of the areas mentioned above, namely solid-
supported synthesis and catalysis there are significant
disadvantages with the technology.5 In particular, it is
difficult to characterise the supported species using
conventional spectroscopic techniques. The principal draw-
backs of solid supports can, to some extent, be avoided by
using soluble polymer supports, for example poly(ethylene
glycol) (PEG) supports.6 While these supports do not permit
separation from the solution phase by conventional
filtration, separation can be achieved readily by centri-
fugation, precipitation or membrane filtration. Unfortu-
nately, supports such as PEG permit only low loadings.
2. Results and discussion
Our chosen class of dendrimer supports was poly(aryl
sulfone)s, since these compounds appeared to meet all the
criteria listed above. Taking into account the literature
precedent on synthesis of oligo(aryl sulfide) dendritic
species12 and initial experiments in our laboratory, we
decided on a divergent synthesis of our target sulfone 6
(Scheme 1), which has eight functional groups for
Keywords: Dendrimer; Sulfone; Nucleophilic aromatic substitution; Single
electron transfer; Supported synthesis.
Corresponding author. Tel.: C44 24 76524375; fax: C44 24 76524112;
e-mail: p.c.taylor@warwick.ac.uk
*
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.09.103

P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
12315
Scheme 1. (i) 80 8C, 60–80 h; (ii) H2O2/AcOH/H2O.
derivatisation. The key reaction in the series is the repeated
nucleophilic aromatic substitution of chloride by thiolate 1.
At all stages, there are a number of chloride groups that can
be substituted. We reasoned that a good yield of 3 would be
achieved by using an excess of trichlorobenzene 2. In the
other nucleophilic aromatic substitution reaction, the
desired selectivity (i.e., substitution of chloride groups of
the substrate 4 rather than that of the product 5 or of 1) was
to be achieved by prior activation of the sulfide group
through oxidation to the corresponding sulfone, since
electron-demanding groups, even in the meta orientation,
are known to favour nucleophilic aromatic substitution.13
The troublesome substitution reaction was monitored by
HPLC. Authentic samples of di- and tri-substituted products
8 and 9 were obtained by preparative HPLC and
characterised by mass spectrometry. It should be noted
that two isomeric disubstituted products 8 are possible. We
believe from chromatographic evidence that only one of the
two is formed, but insufficient material was isolated to
permit structural characterisation. A simple plot of %
conversion versus time (Fig. 1) suggested that the reaction
could be reaching equilibrium and that only a 10% yield of
the desired product was possible. To prove this hypothesis it
would be necessary to resubmit 6 to the reaction conditions,
but the very low yield precluded this.
Reaction of the thiolate 1, derived from commercially
available 3,5-dichlorobenzenethiol, with excess trichloro-
benzene, using a modification of the conditions developed
by Testaferri et al., led to sulfide 3 in 56% yield.14
Quantitative oxidation to the sulfone 4 was achieved by
modification of a literature procedure.15 Repetition of these
two steps was all that was required to complete the synthesis
of 6, but unfortunately the next substitution step did not
proceed smoothly and led to a mixture of mono- 7, di- 8, tri-
9 and tetra-substituted products 5. Nevertheless, a mixture
of tri-substituted 9 and tetra-substituted 5 products was
isolated and oxidised to the corresponding mixture of
sulfones, from which our target compound 6 was isolated
pure by chromatography. The yield of 6 from the two steps
was a very disappointing 6%.
In an attempt to push the reaction to completion, excess
thiolate 1 was used under numerous different reaction
conditions. In all these cases a very complex mixture of
products resulted, which on one occasion was characterised
by mass spectrometry (Fig. 2). The mass spectrum showed
an apparent series of envelopes of peaks separated on
average by ca. 35 amu, fortuitously leading us to the idea
that carbon–chlorine bonds were being reduced.
The formation of the reduced products is consistent with a
single electron transfer (SET) mechanism competing with,
or supplanting, the nucleophilic aromatic substitution
(SNAr) mechanism we assume to be predominant in the
reactions described above. That aryl halides can react via an
Figure 1. % Di-, tri- and tetra-substituted products in the reaction of sulfone 4 with thiolate 1.

12316
P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
in a similar system.22 Whichever mechanism is operating, it
seems that a substitution reaction requires a sulfone
substituent on the ring being substituted, since higher
oligomers were never observed. The two possible processes
are thus reduction of any carbon chlorine bonds in any of 4,
5, 7, 8 and 9 alongside substitution of any or all of the four
chlorides in 4 to give sulfides 5, 7, 8 or 9. The two processes
correspond to a loss of 34.45 amu or an increase of
142.61 amu, respectively. Therefore, the isotopically
averaged masses for all the possible products are predicted
by MZ356.06C142.61mK34.45n where: mZnumber of
substitutions and 0%m%4; nZnumber of reductions and
0%n%(mC4). The series generated is given below, with
predicted masses that lie in the same region being linked in
parentheses and masses for which there are two isomers in
italics.
Figure 2. Detail from EI mass spectrum of product mixture from reaction of
4 with excess thiolate 1.
SET mechanism is well known.16–18 However, reactivity
decreases through the series ArIOArBrOArClOArF and
indeed there are relatively few examples of SRN1
substitution of chlorobenzenes.16 SRN1 substitutions are
normally ‘stimulated’ photochemically, electrochemically,
by solvated electrons or by transition metal salts, yet the
reactions described here take place on warming with no
external source of electrons other than laboratory lights. A
number of factors may explain the surprising SET reactivity
of these chlorobenzene substrates. In particular, sulfone 4
and substitution products 5, 7, 8 and 9 should be excellent
electron acceptors and the radical that is formed by SET
from thiolate 1 is significantly stabilised.
(926.5), (892.0), (857.6), (823.2), (788.7, 783.9), (754.2,
749.4), (719.8, 715.0), (685.4, 680.5), (650.9, 646.1, 641.3),
(611.6, 606.8), (577.2, 572.4), (542.7, 537.4), (503.5,
498.7), (469.0, 464.2), (434.6, 429.8), (395.3), (360.9,
356.1), (326.4, 321.6), (287.2), (252.7), (218.3).
This series corresponds very closely to the 20 observed
envelopes of peaks in the mass spectrum (Fig. 2), although
we have not analysed the isotope patterns. The only
predicted mass to be totally absent is the highest (926.5).
Thus, we have a remarkable situation in which the mass
spectrum is consistent with the presence of the starting
material and all 40 predicted by-products yet none of the
desired product!
Our suggested mechanism for reaction of 4 with excess 1 is
shown in Scheme 2. The closest precedent for the proposed
intermediates is found in the work of Bunnett and Creary19
and of Amatore et al.20 on the photostimulated substitution
of meta-chloroiodobenzene by phenylthiolate. In contrast to
their work, the substitution processes here appear to be
reversible. This will ensure a high enough concentration of
radical species to permit H-abstraction to become a
significant pathway, thus explaining the preponderance of
reduced products in our case.
In fact, dendrimers as complex as 6 are not required for most
applications. Hence, we proceeded to investigate reactions
of 4 with monosubstituted phenylthiolates, since these were
predicted to be poorer both as leaving groups and as electron
donors, thus avoiding the problems we had encountered
with the dichlorophenylthiolate.
Whether the sulfides observed result from SRN1 or SNAr
processes, or both, is not known. Reaction of thiolates and
aryl radicals has been shown to be slow,21 but there is
evidence of simultaneous operation of the two mechanisms
To provide supports that could be derivatised by reaction
with either nucleophiles or electrophiles, we prepared the
Scheme 2.

P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
12317
Scheme 3.
p-halo and p-hydroxy derivatives 10a–c and 10e and 11a–c
and 11e (Scheme 3). In all these cases the aromatic
substitution reaction proceeded smoothly to yield tetra-
sulfides 10, as did the subsequent oxidation to 11.
Deprotection of the methoxy compounds 10d and 11d was
also high yielding. However, despite repeated purification
attempts, we were unable to remove all traces of complexed
boron species from products 10e and 11e. The range of
compounds 11 prepared will allow derivatisation by
nucleophilic aromatic substitution with both hard (when
XZF) and soft (when XZCl) nucleophiles, by palladium
catalysed coupling reactions (when XZBr) and by simple
reactions with electrophiles (when XZOH). While com-
pounds 10 do not meet all of our design criteria, they are
also potentially useful supports if oxidising agents are not
being used (vide infra).
Firstly, a five step supported synthesis of a dipeptide was
realised. The choice of a peptide synthesis was somewhat
arbitrary and many more classes of synthesis need to be
explored to fully assay the potential of the new supports.
Purities of 90% or greater as judged by NMR spectra were
considered sufficient to proceed to the next step. Boc
protected valine was coupled to the support using an
adaptation of the EDCI–OBt method for peptide coupling.23
After deprotection, Boc protected glycine was coupled
using the same method, again with TFA mediated
deprotection. Finally, sodium hydroxide was used to
hydrolyse the ester, yielding the desired dipeptide 16 pure
in 29% overall yield and recovered support 10e (Scheme 4).
To demonstrate that the new supports have potential in
synthesis, we focussed on the phenolic tetrasulfide support
10e. It should be noted that we did not have facilities for
membrane separation in our laboratories, hence conven-
tional purification methods were used.
Scheme 4.

12318
P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
Scheme 5.
For applications in combinatorial synthesis, it is advan-
tageous for the final cleavage step to permit the further
introduction of diversity, the so-called ‘react and release’
method.24 This was demonstrated by the formation of a
model tetrapropionyl ester of 10e. Reaction of ester 17 with
both benzylamine and cyclohexylamine yielded the desired
amides in good yield, with the support being recovered in
both cases (Scheme 5). Aniline, which is less nucleophilic,
did not function well in the react and release protocol.
flask and heated to reflux. A solution of sulfide 3 (3.0 g,
9.26 mmol), in chloroform (10 ml) was added with care
and the biphasic mixture was heated at reflux for 18 h.
The mixture was cooled and the product extracted with
chloroform (2!25 ml). The combined organic layers
were washed several times with water, NaHCO3 (5%,
2!50 ml) and brine (20 ml). The organic layer was dried
over MgSO4, filtered and the solvent removed under
vacuum. The white crystalline solid was passed through a
silica column to reveal pure sulfone 4 in 98% yield, Rf
(eluent: hexane): 0.63; mp 167–169 8C (Found: C, 40.98;
H, 1.84. C12H6SO2Cl4 requires C, 40.48; H, 1.70); nmax
(cmK1) 1144 (S]O); dH (300 MHz; CDCl3) 7.60 (2H, t,
JZ1.8 Hz, ArH), 7.80 (4H, d, JZ1.8 Hz, ArH); dC
(75.5 MHz; CDCl3) 126.34 (ArH), 134.43 (ArH), 137.08
(quat), 143.45 (quat); m/z (EI) 356, ([M%]C65%), 292
(7), 193 (100), 164 (12); m/z 353.8843 (C12H6SO2Cl4
requires 353.8838).
3. Conclusion
In summary, a short, divergent synthesis of G1 and G2
oligo(aryl sulfones) has been achieved using nucleophilic
aromatic substitution reactions. Activation of the Gn leaving
groups prior to formation of GnC1 by a sulfide to sulfone
oxidation allows good selectivity. However, preparation of
the G2 oligo(aryl sulfone) was largely thwarted by
competition from SET processes, this being unusual for
chlorobenzene substrates. A range of tetrasubstituted
pentasulfones have been prepared for applications as
homogeneous supports for synthesis and the utility of one
of them has been demonstrated through supported pre-
paration of a dipeptide and through ‘react and release’
synthesis of amides.
4.3. Bis-(3,5-bis-(3,5-dichlorophenylsulfanyl)phenyl)
sulfone 5
To a solution of 3,5-dichlorothiophenol (2.1 mmol,
329 mg) in DMF (10 ml), was added potassium carbonate
(1 equiv, 2.1 mmol, 255 mg) and
4
(0.13 equiv,
0.281 mmol, 100 mg). The solution was stirred under a
nitrogen atmosphere at 80 8C for 80 h. The cooled
solution was added to ice and the resulting solid
precipitate was filtered, washed with water, redissolved
in chloroform (20 ml) and washed with water, NaHCO3,
water and brine. The organic fraction was dried over
MgSO4 and concentrated under vacuum to furnish
290 mg of a four-component mixture. The mixture was
subjected to flash chromatography on silica (eluent:
hexanes/ethyl acetate 15:1), to furnish an inseparable
4. Experimental
4.1. Bis-(3,5-dichlorophenyl) sulfide 3
To a round bottomed flask containing 3,5-dichlorothiophe-
nol (5.0 g, 27.9 mmol) was added potassium carbonate
(9.71 g, 69.8 mmol) and 1,3,5-trichlorobenzene (10.6 g,
58.7 mmol). The mixture was heated to 80 8C under a dry
nitrogen atmosphere for 2.5 days. The solution was poured
onto ice and the resulting solid precipitate was collected by
filtration to yield an off-white solid (15.4 g), which was
passed through a silica column using cyclohexane–ethyl
acetate (10/1) yielding 2.72 g of a two component mixture,
containing the desired product and 1,3,5-trichlorobenzene.
This mixture was heated to 75 8C under vacuum to remove
the trichlorobenzene. The liquid product was cooled to
crystallize as a white solid (450 mg, 56%), Rf (eluent:
hexane): 0.75 (Found: C, 44.47; H, 1.86. C12H6SCl4
requires C, 44.36; H, 1.86); nmax (cmK1) 1575; dH
(300 MHz; CDCl3) 7.18 (4H, d, JZ1.8 Hz, ArH), 7.25
(2H, t, JZ1.8 Hz, ArH); dC (75.5 MHz; CDCl3) 128.58
(ArH), 129.55 (ArH), 136.19 (quat), 137.78 (quat); m/z (EI)
324, ([M%]C 50%), 252 (26), 182 (86), 145 (33), 83 (100).
1
mixture of 5 and tri-substituted impurity 9 (ratio by H
NMR, 5:1), Rf (eluent: petrol/ethyl acetate 10:1): 0.31;
nmax (cmK1) 1159, 1128 (S]O); dH (300 MHz; CDCl3)
7.06 (major, 2H, t, JZ1.5 Hz, ArH), 7.20 (major, 8H, d,
JmZ1.9 Hz, ArH), 7.35 (major, 4H, t, JmZ1.9 Hz, ArH),
7.60 (major, 4H, d, JmZ1.5 Hz, ArH), 7.10 (minor, t,
JoZ1.5 Hz, ArH), 7.24 (minor, d, JZ1.9 Hz, ArH), 7.30,
(minor, t, JZ1.7 Hz, ArH), 7.37, (minor, t, JZ1.7 Hz,
ArH), 7.40 (minor, t, JZ1.7 Hz, ArH), 7.66, (minor, t,
JZ1.7 Hz, ArH), 7.70 (minor, t, JZ1.7 Hz, ArH); dC
(75.5 MHz; CDCl3) 126.79 (major, ArH), 129.54 (major,
ArH), 130.84 (major, ArH), 134.45 (major, ArH), 135.72
(major, quat), 136.44 (major, quatH), 140.16 (major,
quat), 143.24 (major, quat), 126.71 (minor, ArH), 127.20
(minor, ArH), 129.74 (minor, ArH), 130.47 (minor, ArH),
130.90 (minor, ArH), 134.37 (minor, ArH), 134.78
(minor, ArH), 140.28 (minor, quat); m/z (EI) 926,
([M%]C100%), 784 ([M%]C, 35%), 748 (14), (CI,
NHC4 ) 944 ([M%]CCNH4C86), 802 ([M%]CCNHC4 ).
4.2. Bis-(3,5-dichlorophenyl) sulfone 4
Hydrogen peroxide (30%, 20 ml) and acetic acid (17 M,
15 ml), in water (10 ml) were added to a round bottomed

P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
12319
4.4. Bis-(3,5-bis-(3,5-dichlorophenylsulfonyl)phenyl)
sulfone 6
(d, JZ22.9 Hz, ArH), 131.21 (ArH), 131.65 (d, JZ
9.9 Hz, ArH), 131.81 (ArH), 135.18 (quat), 143.58 (quat),
146.52 (quat), 164.18 (quat); m/z (EI) 850 ([M%]C45%),
818 (20), 692 (43), 143 (100).
To a solution of crude 5 (720 mg, 0.778 mmol), in
chloroform (20 ml), was added H2O2 (30%, 10 ml) and
glacial acetic acid (9 ml) in water (5 ml). The biphasic
mixture was heated to reflux with vigorous stirring for 4 h.
The mixture was cooled and the organic layer separated,
washed with water, NaOH (10%), water and brine. The
solution was dried over MgSO4 and concentrated under
vacuum to give a crude white solid (710 mg), which was
subjected twice to flash chromatography on silica (eluents:
hexanes/ethyl acetate 15:1; cyclohexane/ethyl acetate 20:1),
to give the target molecule as a white solid (50 mg, 6%) that
was still contaminated by traces of the trisubstituted
compound. Rf (eluent: petrol/ethyl acetate, 10:1): 0.25;
nmax (cmK1) 1150, 1080 (S]O); dH (250 MHz; CDCl3)
7.55 (4H, t, JmZ1.8 Hz, ArH), 7.75 (8H, d, JmZ1.8 Hz,
ArH), 8.55 (6H, s, ArH); dC (75.5 MHz; CDCl3) 126.86
(ArH), 132.18 (ArH), 132.64 (ArH), 135.32 (ArH), 137.60
(quat), 141.99 (quat), 143.69 (quat), 145.54 (quat); m/z (EI)
1054, ([M%]C36%), 880 (39), 844 (14), 670 (20), 496 (53),
333 (42), 193 (100); (CI, NHC4 ) 1072 ([M%]CCNH4C23),
862 (29), 654 (68), 514 (100).
4.7. Bis-(3,5-bis-(4-chlorophenylsulfanyl) sulfone 10b
To a solution of 4-chlorothiophenol (1.95 g, 13.48 mmol) and
potassium-t-butoxide (1.51 g, 13.48 mmol) in DMF (90 ml)
was added sulfone 6. The mixture was heated to 85 8C for
12 days. After cooling the mixture was added to water and
basified with sodium hydroxide (10%). The mixture was
extracted with dichloromethane, washed with water and brine,
dried over MgSO4 and concentrated under vacuum. The
residue was subjected to flash chromatography on silica
(eluent: cyclohexane/dichloromethane 10:1) to yield 10b
(500 mg, 23%) and trisubstituted product (755.0 mg,
1.1 mmol) that was further reacted with 4-chlorothiophenol
(406 mg, 2.83 mmol) and potassium-t-butoxide (315.0 mg,
2.6 mmol) in DMF (50 ml) for 72 h. Usual workup and flash
chromatography on silica afforded a further 420 mg 10b (total
yieldZ43%), Rf (eluent: cyclohexane/dichloromethane 3:1):
0.25; mp 133–136 8C (Found: C, 54.37; H, 2.74.
C36H22Cl4O2S5 requires C, 54.84; H, 2.79); nmax (cmK1
)
1159, 1087 (S]O); dH (300 MHz; CDCl3) 6.82 (2H, t, JmZ
1.6 Hz, ArH), 7.25 (16H, m, ArH) 7.41 (4H, d, JmZ1.6 Hz,
ArH); dC (75.5 MHz; CDCl3) 124.27 (ArH), 130.18 (quat),
130.46 (ArH), 131.21 (ArH), 135.40 (ArH), 136.1 (quat),
141.84(quat), 142.85(quat); m/z (EI) 788, ([M%]C100%), 646
(35), 504 (72); (CI NHC4 ) 805 ([M%]CCNHC4 100).
4.5. Bis-(3,5-bis-(4-fluorophenylsulfanyl) sulfone 10a
To a solution of 4-fluorothiophenol (395.5 mg, 3.1 mmol) in
DMF (25 ml) was added potassium carbonate (390.3 mg,
2.8 mmol). The mixture was stirred at 20 8C for 20 min and
heated to 85 8C. Sulfone 6 (250.0 mg, 0.7 mmol) was added
the mixture and stirred under nitrogen at 85 8C for 7 days.
The mixture was cooled and the solvent removed under
vacuum. The resulting solid was washed several times with
water, dried and subjected to flash chromatography on silica
(eluent: cyclohexane/ethyl acetate 20:1) to reveal pure 11a
in 63% yield, Rf (eluent: hexane): 0.63; mp 154–156 8C
(Found: C, 59.57; H, 3.01. C36H22F4O2S5 requires C, 59.83;
H, 3.05); nmax (cmK1) 1159 (S]O); dH (300 MHz; CDCl3)
6.65 (2H, t, JmZ1.8 Hz, ArH), 7.06 (8H, AA0 of AA0BB0,
C6H4), 7.31 (14H, m, ArH); dC (75.5 MHz; CDCl3) 117.54
4.8. Bis-(3,5-bis-(4-chlorophenylsulfonyl) sulfone 11b
To a solution of tetrachlorosulfide 10b (410 mg, 0.5 mmol) in
dichloromethane (30 ml) was added m-chloroperbenzoic acid
30–50% (2.52 g, 8.32 mmol). The mixture was stirred at room
temperature for 18 h after which calcium hydroxide (1.0 g,
17.5 mmol) was added and themixture stirredfor a further1 h.
The solid precipitate was filtered and the organic layer washed
with sodium hydroxide (10%, 50 ml) water and brine. The
organic layer was dried over MgSO4 and concentrated under
vacuumto give11b asa whitesolid(472 mg, 99%). Rf (eluent:
cyclohexane/ethyl acetate 1:1): 0.85; mpO200 8C (decomp.)
(Found: C, 47.06; H, 2.36. C36H22Cl4O10S5 requires C, 47.18;
H, 2.40); nmax (cmK1) 1149, 1082 (S]O); dH (300 MHz;
CD6SO) 7.65 (8H, AA0 of AA0BB0, JoZ8.7 Hz, C6H4), 8.64
(8H, BB0 of AA0BB0, JoZ8.7 Hz, C6H4), 8.73 (2H, t, JmZ
1.5 Hz, ArH), 9.07 (4H, d, JmZ1.5 Hz, ArH); dC (75.5 MHz;
CDCl3) 130.37 (ArH), 130.75 (ArH), 131.93 (ArH), 132.87
(ArH), 138.17 (quat), 140.26 (quat), 143.35 (quat), 144.50
(quat); m/z (FABC) 917, ([M%]C30%), MALDI (DHB-KC)
955.1 ([M%]CCKC).
2
(d, JCFZ21.7 Hz, ArH), 123.06 (ArH), 126.04 (quat),
129.40 (ArH), 137.04 (d, 3JCFZ15 Hz, ArH), 142.70 (quat),
1
162.14 (quat), 163.64 (d, JCFZ251 Hz, Ar); m/z (EI) 722,
([M%]C100%), 724 (65), 726 (10), 630 (30).
4.6. Bis-(3,5-bis-(4-fluorophenylsulfonyl) sulfone 11a
To a solution of sulfide 10a (84.0 mg, 0.1 mmol) in
chloroform (0.5 ml) was added hydrogen peroxide (30%,
1.2 ml) glacial acetic acid (1 ml, 17 mmol) and water. The
resulting mixture was stirred vigorously under reflux
conditions for 12 h. The mixture was cooled and diluted
with chloroform (20 ml) separated and washed with sodium
hydroxide (10%, 10 ml), water and brine. The organic layer
was dried over MgSO4 and concentrated under vacuum to
give 11a as a white solid (98 mg, 96%). Satisfactory
microanalysis was not obtained, Rf (eluent: cyclohexane/
4.9. Bis-(3,5-bis-(4-bromophenylsulfanyl) sulfone 10c
To a solution of 4-bromothiophenol (581 mg, 3.1 mmol) and
potassium-t-butoxide (378 mg, 3.4 mmol) in DMAC (10 ml)
was added sulfone 6 (250 mg, 0.7 mmol). The reaction was
heated to 80 8C for 72 h. The mixture was cooled and poured
onto ice with stirring then basified with sodium hydroxide
(10%). The resulting off white solid material was collected by
filtration and washed with water (5!50 ml) and sodium
hydroxide (10%, 2!50 ml). The solid was dried to furnish
ethyl acetate 1:1): 0.21; mpO200 8C, (decomp.); nmax (cmK1
)
1152 (S]O); dH (300 MHz; CDCl3) 7.28 (8H, AA0 of
AA0BB0, JoZ8.8 Hz, C6H4), 7.99 (8H, BB0 of AA0BB0,
JoZ8.8 Hz, C6H4) 8.57 (2H, t, JmZ1.5 Hz, ArH), 8.59 (4H,
d, JmZ1.5 Hz, ArH); dC (75.5 MHz; CDCl3) 118.05

12320
P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
tetrabromosulfide 10c (648 mg, 95%), mpO200 8C (Found:
C, 44.62; H, 2.34. C36H22Br4O2S5 requires C, 44.73; H, 2.27);
nmax (cmK1) 1159, 1097 (S]O); dH (300 MHz; CDCl3) 6.88
(2H, t, JZ1.8 Hz, ArH), 7.19 (8H, AA0 of AA0BB0, JZ
10.3 Hz, C6H4)7.42(4H, d,JZ1.8 Hz, ArH),7.48(8H,BB0 of
AA0BB0, JZ10.3 Hz, C6H4); dC (75.5 MHz; CDCl3) 124.1
(quat), 124.6 (ArH), 130.98 (quat), 131.68 (ArH), 133.41
(ArH), 135.42 (ArH), 141.59 (quat), 142.87 (quat); m/z (EI)
966, [M%]C(100%), 888 (80) 814 (98), 640 (96).
then diluted with ethyl acetate/acetone (1:1, 50 ml). The
organic layer was separated, washed with water and brine,
dried over MgSO4 and the solvent removed under vacuum
to furnish, after trituration with cold chloroform, tetraphenol
10e (34 mg, 74%) estimated to be O90% pure by NMR, as
an off-white waxy solid, Rf (eluent: petrol/ethyl acetate 1:1):
0.1; nmax (cmK1) 3320 (OH); dH (300 MHz; CD6CO), 6.78
(2H, t, JmZ1.6 Hz, ArH), 6.95 (8H, AA0 of AA0BB0, JZ
8.8 Hz, C6H4), 7.17 (4H, d, JmZ1.6 Hz, ArH), 7.30 (8H,
BB0 of AA0BB0, JZ8.8 Hz, C6H4), 9.06 (4H, br s, OH); dC
(75.5 MHz; CD6CO), 118.44 (ArH), 119.90 (quat), 121.89
(ArH), 128.04 (ArH), 138.28 (ArH), 143.95 (quat), 145.09
(quat), 160.47 (quat); m/z (FAB) 714, ([M%]C56%), 624
(52), 500 (74), 358 (100).
4.10. Bis-(3,5-bis-(4-bromophenylsulfonyl) sulfone 11c
To a solution of tetrabromosulfide 10c (20 mg, 21 mmol) in
dichloromethane (20 ml) was added m-chloroperbenzoic
acid 35% (163 mg, 0.3 mmol). The resulting solution was
allowed to stir for 16 h at room temperature after which it
was diluted with dichloromethane (30 ml) and washed with
sodium hydroxide 10% (2!50 ml) water (100 ml) and
brine. The organic layer was separated, dried over MgSO4
and concentrated under vacuum to furnish tetrabromo-
sulfone 11c (20 mg, 88%). Satisfactory microanalysis was
not obtained, Rf (eluent: cyclohexane/ethyl acetate1:1):
0.68; mpO200 8C; nmax (cmK1) 1154 (S]O); dH
(300 MHz; CD6SO) 7.83 (8H, AA0 of AA0BB0, JZ8.8 Hz,
C6H4), 8.11 (8H, BB0 of AA0BB0, JZ8.8 Hz, C6H4) 8.69
(2H, t, JZ1.6 Hz, ArH), 9.05 (4H, d, JmZ1.6 Hz, ArH); dC
(75.5 MHz; CDCl3) 128.23 (quat), 129.41 (ArH), 130.7
(ArH), 131.9 (ArH), 133.29 (ArH), 138.54 (quat), 143.35
(quat), 144.54 (quat); m/z (FAB) 1094, [M%]C(61%), 1062
(100), 1032 (43), 690 (100).
4.13. Bis-(3,5-bis-(4-methoxyphenylsulfonyl) sulfone 11d
To a solution of tetramethoxysulfide 10d (600 mg,
0.8 mmol) in dichloromethane (50 ml) was added m-chloro-
perbenzoic acid 57%, (3.6 g, 12.5 mmol). The solution was
stirred at room temperature for 12 h. The solvent was
removed to furnish a white solid which was triturated with
sodium hydroxide 10% (50 ml) for 2 h. The organic solid
was separated by filtration, dissolved in chloroform, dried
over MgSO4 and the solvent removed under vacuum to
furnish tetramethoxy sulfone 11d (689 mg, 98%) as a white
solid, Rf (eluent: petrol/ethyl acetate 10:1): 0.1; mp 150 8C
(decomp.) (Found C, 53.58; H, 3.65. C40H34O6S5 requires
C, 53.45; H, 3.78); nmax (cmK1) 1138, 1102 (S]O); dH
(300 MHz; CDCl3) 3.87 (12H, s, OMe), 7.04 (8H, AA0 of
AA0BB0, JZ9.2 Hz, C6H4), 7.87 (8H, BB0 of AA0BB0, JZ
9.2 Hz, C6H4), 8.50 (4H, d, JmZ1.8 Hz, ArH), 8.54 (2H, t,
JmZ1.8 Hz, ArH); dC (75.5 MHz; CD6SO) 56.20 (OMe),
115.61 (ArH), 130.52 (quat), 130.71 (ArH), 131.03 (ArH),
131.46 (ArH), 143.16 (quat), 145.79 (quat), 164.30 (quat);
m/z (EI) 898, ([M%]C5%), 728 (31), 588 (100), 556 (35), (CI
NHC4 ) 915 ([M%]CCNHC4 10).
4.11. Bis-(3,5-bis-(4-methoxyphenylsulfanyl) sulfone 10d
To a solution of 4-methoxythiophenol (983 mg, 7.0 mmol)
in DMF (25 ml) was added potassium-t-butoxide (803 mg,
7.2 mmol) and sulfone 6 (500 mg, 1.4 mmol). The solution
was heated to 80 8C for 16 h. The crude mixture was cooled
and quenched on ice. The aqueous layer was extracted into
dichloromethane, the solvent dried and removed under
vacuum to reveal an off-white solid. This solid was
triturated in cold acetone to give an off-white solid
(810 mg, 75%) of tetramethoxy 10d. Rf (eluent: petrol/
ethyl acetate 10:1): 0.59; mp. 174–176.5 8C (Found: C,
61.07; H, 4.27. C40H34O6S5 requires C, 62.33; H, 4.41); nmax
(cmK1) 1164, 1025 (S]O); dH (300 MHz; CDCl3), 3.85
(12H, s, OMe), 6.63 (2H, t, JmZ1.8 Hz, ArH), 6.86 (8H,
AA0 of AA0BB0, JZ9.0 Hz, C6H4), 7.21 (4H, d, JmZ
1.8 Hz, ArH), 7.29 (8H, BB0 of AA0BB0, JZ9.0 Hz, C6H4);
dC (75.5 MHz; CDCl3) 55.70 (OMe), 115.75 (ArH), 121.04
(quat), 121.75 (ArH), 127.85 (ArH), 136.99 (ArH), 142.49
(quat), 143.55 (quat), 161.04 (quat); m/z (EI) 770,
([M%]C15%), 632 (63), 492 (100), (CI NHC4 ) 788
([M%]CCNHC4 100).
4.14. Bis-(3,5-bis-(4-hydroxyphenylsulfonyl) sulfone 11e
To a solution of tetramethoxy sulfone 11d (50 mg, 60 mmol)
in dichloromethane (10 ml), which was cooled to K78 8C
under nitrogen, was added boron tribromide 1.0 M in
dichloromethane (0.8 ml, 0.8 mmol) with care. The result-
ing solution was stirred for a further 15 min at K78 8C then
at room temperature for 70 h. Deionised water (5 ml) was
added with care and the mixture stirred for a further 30 min,
then diluted with ethyl acetate/acetone (1:1, 50 ml). The
organic layer was separated, washed with water and brine,
dried over MgSO4 and the solvent removed under vacuum
to furnish tetraphenol 11e (41 mg, 88%) estimated to be
O90% pure by NMR, as an off-white waxy solid, Rf (eluent:
petrol/ethyl acetate 1:1): 0.1; nmax (cmK1) 3310 (OH); dH
(300 MHz; CD6CO), 6.89 (8H, AA0 of AA0BB0, JZ8.8 Hz,
C6H4), 7.78 (8H, BB0 of AA0BB0, JZ8.8 Hz, C6H4), 8.41
(2H, t, JmZ1.5 Hz, ArH), 8.61 (4H, d, JmZ1.5 Hz, ArH),
9.6 (4H, br s, OH); dC (75.5 MHz; CD6CO), 117.76 (2!
ArH), 131.12 (quat), 131.80 (ArH), 131.18 (ArH), 144.55
(quat), 147.80 (quat), 164.1 (quat); m/z (FAB) 843,
([M%]C20%), 667 (94), 638 (100).
4.12. Bis-(3,5-bis-(4-hydroxyphenylsulfanyl) sulfone 10e
To a solution of tetramethoxy sulfide 10d (50 mg, 65 mmol)
in dichloromethane (10 ml), which was cooled to K60 8C
under nitrogen, was added boron tribromide 1.0 M in
dichloromethane (1.3 ml, 1.3 mmol) with care. The result-
ing solution was stirred for a further 15 min at K60 8C then
at room temperature for 18 h. Deionised water (5 ml) was
added with care and the mixture stirred for a further 30 min,

P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
12321
4.15. Supported synthesis of dipeptide 12
tetrapropionate ester 17 (96 mg, 73%) as a crude oil, Rf
(eluent: petrol/ethyl acetate 1:1): 0.77; nmax (cmK1) 1710
(C]O); dH (300 MHz; CD6CO), 1.21 (12H, t, JZ7.5 Hz,
Me), 2.64 (8H, q, JZ7.5 Hz, CH2), 6.94 (2H, t, JmZ1.5 Hz,
ArH), 7.21 (8H, AA0 of AA0BB0, JZ8.7 Hz, C6H4), 7.45–
7.47 (12H, m, ArH); dC (75.5 MHz; CD6CO), 9.41, 28.16,
123.52, 124.20, 135.41, 137.37, 138.70, 142.01, 151.85,
172.85.
To a cooled solution of N-Boc-L-valine (243 mg,
1.12 mmol), and EDCI (1.12 mmol, 215 mg), which had
been stirred in ethyl acetate–THF (1/1), at 0 8C for 1 h was
added HOBT (1.12 mmol, 151 mg). The solution was
allowed to stand for a further 1 h at 0 8C after which
was added 10e (0.14 mmol, 100 mg), and triethylamine
(1.9 equiv, 2.1 mmol, 212.1 mg) in one portion. The
reaction mixture was stirred at room temperature for 48 h,
diluted with ethyl acetate (50 ml) and washed with water
and NaOH (10%). The organic layer was dried over MgSO4
and concentrated to give 12 as a crude oil; dH (300 MHz;
CD3CO) 1.08 (12H, d, JZ6.6 Hz, CHCH3), 1.10 (12H, d,
JZ6.6 Hz, CHCH3), 1.41 (36H, s, C(CH3)3), 2.33–2.41
(4H, m, CH(CH3)2), 4.39 (4H, m, JZ5.8 Hz, CH), 6.50 (4H,
d, JZ8.3 Hz, NH), 6.9–7.6 (22H, m, Ar); dC (75.5 MHz;
CD6CO) 18.95 (Me), 20.02 (Me), 28.97 (t-Bu), 31.67
(CHMe2), 60.88 (CH), 79.94 (quat), 118.56, 124.78, 129.65,
131.75, 136.49, 143.09, 144.20, 153.07, 157.29 (C]O),
171.94 (C]O).
To a solution of tetraester 17 (0.1 mmol) in dichloromethane
(10 ml) was added either cyclohexylamine or benzylamine
(0.5 mmol) and triethylamine (0.1 mmol). The solution was
stirred for 26 h and the solvent was removed in vacuo. The
crude residue was dissolved in ethyl acetate/acetone (1:1,
50 ml) and washed with HCl (1 M) and water. The organic
layer was dried over MgSO4 and concentrated to provide a
mixture of the product amide and the recovered support 10e,
as determined by NMR spectroscopy.
Acknowledgements
To a solution of 12 in dichloromethane (10 ml) was added
TFA (10 ml). The solution was stirred for 2 h after which the
solvent and excess TFA were removed in vacuo below
40 8C to yield 13 as a crude brown oil; dH (300 MHz;
CD3CO) 1.25 (12H, d, JZ6.6 Hz, CHCH3), 1.28 (12H, d,
JZ6.6 Hz, CHCH3), 2.64–2.77 (4H, m, CH(CH3)2), 4.57
(4H, br s, CH), 5.33 (8H, d, JZ6.1 Hz, NH2), 7.1–7.6 (22H,
m, Ar); dC (75.5 MHz; CD6CO) 18.45 (Me), 19.07 (Me),
31.15 (CHMe2), 59.93 (CH), 118.56, 124.45, 131.50,
136.00, 136.14, 151.91, 168.95 (C]O), two Ar signals
not observed.
We thank GlaxoSmithKline and the EPSRC for financial
support, Dr. Alison Rodger and Mr. Muftah Darwish for
assistance with HPLC and Prof. Andrew Livingston for
bibliography on membrane separation.
References and notes
1. Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149–2154.
2. (a) Leznoff, C. C. Chem. Soc. Rev. 1974, 3, 65–85. (b)
Houghton, R. A. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 5131–
5135. (c) Furka, A.; Sebesteyen, F.; Asgedom, M.; Dibo, G.
Abstr. 14th Int. Congr. Biochem. Prague 1988, 5, 47. (d)
Houghton, R. A.; Pinilla, A.; Blondelle, S. E.; Appel, J. R.;
Dooley, C. T.; Cuervo, J. H. Nature 1991, 354, 84–86. (e)
Lam, K. S.; Salmon, S. E.; Hersch, E. M.; Hruby, V. J.;
Kazmierski, W. M.; Knapp, R. J. Nature 1991, 354, 82–84. (f)
Ellman, J.; Stoddard, B.; Wells, J. Proc. Natl. Acad. Sci. U.S.A.
1997, 94, 2779–2782. (g) Terrett, N. K. Combinatorial
Chemistry; Oxford University Press: Oxford, 1998.
3. Dickerson, T. J.; Reed, N. N.; Janda, K. D. Chem. Rev. 2002,
102, 3325–3343.
The two steps above were repeated using crude 13 as
substrate and N-Boc-N-glycine in place of N-Boc-L-valine.
Crude supported dipeptide 15 was then dissolved in ethyl
acetate–acetone (1/1) (50 ml) and sodium hydroxide (10%,
10 ml) was added. The biphasic solution was stirred
vigorously for 30 min. The aqueous layer was neutralised
to pH 7 with HCl (1 M). The water was removed in vacuo
and the solid residue was extracted thoroughly with
methanol. The methanol was removed in vacuo to reveal
the pure target dipeptide in 29% yield for the five steps; dH
(400 MHz; CD3OD) 1.00 (6H, 2!d, J CHCH3), 2.24 (1H,
d, JZ5.8 Hz, CHCH3), 3.80 (2H, s, CH2), 4.41 (1H, d, JZ
5.8 Hz, CH); m/z (EI) 174 (MC100%). The organic layer
was dried over MgSO4 to return the support 10e as a crude
sodium salt; dH (400 MHz; CD3OD) 6.62 (2H, t, JZ
1.5 Hz), 6.92 (8H, AA0 of AA0BB0, JZ8.6 Hz), 7.22–7.25
(12H, m).
4. Ley, S. V.; Baxendale, I. R.; Bream, R. N.; Jackson, P. S.;
Leach, A. G.; Longbottom, D. A.; Nesi, M.; Scott, J. S.; Storer,
R. I.; Taylor, S. J. J. Chem. Soc., Perkin Trans. 1 2000,
3815–4195.
5. Han, H. S.; Janda, K. D. J. Am. Chem. Soc. 1996, 118,
7632–7633.
4.16. React and release synthesis of amides
6. (a) Pillai, V. N. R.; Mutter, M.; Bayer, E. J. Org. Chem. 1980,
45, 5364–5370. (b) Spanka, C.; Wentworth, P.; Janda, K. D.
Comb. Chem. High Throughput Screen. 2002, 5, 233–240.
7. (a) Kim, R. M.; Manna, M.; Hutchins, S. M.; Griffin, P. R.;
Yates, N. A.; Bernick, A. M.; Chapman, K. T. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 10012–10017. (b) Kreiter, R.;
Kleij, A. W.; Klein Gebbink, R. J. M.; van Koten, G. Top.
Curr. Chem. 2001, 217, 163–199.
To a cooled solution of tetrasulfide 10e (100 mg, 140 mmol)
in DMF (10 ml) was added triethylamine (0.71 ml,
0.70 mmol) and propionyl chloride (0.57 ml, 0.63 mmol).
The resulting solution was stirred at room temperature for
18 h. The solution was reduced in volume and diluted with
dichloromethane (30 ml), washed with water (50 ml), HCl
(1 M, 50 ml), water and finally brine. The solution was dried
over MgSO4 and the solvent removed under vacuum to
yield bis-(3,5-bis-(4-hydroxyphenylsulfanyl) sulfone
8. Dijkstra, H. P.; Kruithof, C. A.; Ronde, N.; van de Coevering,
´
R.; Ramon, D. J.; Vogt, D.; van Klink, G. P. M.; van Koten, G.
J. Org. Chem. 2003, 68, 675–685.

12322
P. C. Taylor et al. / Tetrahedron 61 (2005) 12314–12322
9. Dijkstra, H. P.; van Klink, G. P. M.; van Koten, G. Acc. Chem.
Res. 2002, 35, 798–810. Vankelecom, I. F. J. Chem. Rev. 2002,
102, 3779–3810.
17. (a) Bunnett, J.F.;Kearley, F. J. J.Org. Chem. 1971, 36, 184–186.
(b) Montanari, S.; Paradisi, C.; Scorrano, G. J. Org. Chem. 1993,
58, 5628–5631. (c) Baumgarner, C. D.; Malen, A. H.; Pastor,
S. D.; NabiRahni, M. A. Helv. Chim. Acta 1992, 75, 480–486.
´
18. (a) Saveant, J. M. Tetrahedron 1994, 50, 10117–10165. (b)
´ ´
Costentin, C.; Hapiot, P.; Medebielle, M.; Saveant, J. M.
10. White, L. S.; Nitsch, A. R. J. Membr. Sci. 2000, 179, 267–274.
Membrane technology and research (http://mtrinc.nextweb.
net/).
11. (a) Raman, L. P.; Cheryan, M.; Rajagopalan, N. J. Am. Oil
Chem. Soc. 1996, 73, 219–224. (b) Zwijenburg, H. J.; Krosse,
A. M.; Ebert, K.; Peinnemann, K. V.; Cuperus, F. P. J. Am. Oil
Chem. Soc. 1999, 76, 83–87.
J. Am. Chem. Soc. 1999, 121, 4451–4460.
19. Bunnett, J. F.; Creary, X. J. Org. Chem. 1974, 39, 3612.
20. Amatore, C.; Beugelmans, R.; Bois-Choussy, M.; Combellas,
´
C.; Thiebault, A. J. Org. Chem. 1989, 54, 5688–5695.
21. Amatore, C.; Combellas, C.; Pinson, J.; Oturan, M. A.;
12. Van Bierbeek, A.; Gingras, M. Tetrahedron Lett. 1998, 39,
6283–6286.
´ ´
Robveille, S.; Saveant, J. M.; Thiebault, A. J. Am. Chem. Soc.
13. Paradisi, C. In Trost, B., Ed.; Comprehensive Organic
Synthesis; Pergamon: Oxford, 1991; Section 2.1.
14. Testaferri, L.; Tingoli, M.; Tiecco, M. J. Org. Chem. 1980, 45,
4376–4380.
1985, 107, 4846–4853.
22. Grossi, L.; Strazzari, S. J. Chem. Soc., Perkin Trans. 2 1999,
2141–2146.
23. Nozaki, S.; Muramatsu, I. Bull. Chem. Soc. Jpn. 1982, 55,
2165–2168.
15. Sasse, K.; Niedrig, H. Angew. Chem., Int. Ed. Engl. 1981, 20,
780–782.
´
16. Rossi, R. A.; Pierini, A. B.; Pen˜en˜ory, A. B. Chem. Rev. 2003,
103, 71–168.
24. (a) Marhsall, D. L.; Liener, I. E. J. Org. Chem. 1970, 35,
867–868. (b) Breitenbucher, J. G.; Johnson, C. R.; Haight, M.;
Phelan, C. J. Tetrahedron Lett. 1998, 39, 1295–1298.