Efficient, mild, parallel and purification-free synthesis of aryl ethers via the Mitsunobu reaction

Article abstract of DOI:10.1016/j.tetlet.2008.04.099

A wide range of commercial diazodicarboxylates and phosphines were screened in an attempt to find purification-free conditions for application in parallel synthesis. The combination of immobilized triphenylphosphine and TMAD proved to be suitable for the synthesis of aryl ethers via the Mitsunobu reaction. Nine ethers were synthesized in good yield and excellent purity, the purification being limited to a filtration step.

Full text of DOI:10.1016/j.tetlet.2008.04.099

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 4182–4185
Efficient, mild, parallel and purification-free synthesis
of aryl ethers via the Mitsunobu reaction
Eric Valeur *, Didier Roche
Merck-Serono, Department of Medicinal Chemistry, 4 Avenue Pre´sident Mitterrand, 91380 Chilly-Mazarin, France
Received 19 March 2008; revised 14 April 2008; accepted 16 April 2008
Available online 20 April 2008
Abstract
A wide range of commercial diazodicarboxylates and phosphines were screened in an attempt to find purification-free conditions for
application in parallel synthesis. The combination of immobilized triphenylphosphine and TMAD proved to be suitable for the synthesis
of aryl ethers via the Mitsunobu reaction. Nine ethers were synthesized in good yield and excellent purity, the purification being limited
to a filtration step.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Mitsunobu; Aryl ethers; Parallel synthesis; Purification-free; Polymer-supported; TMAD
The Mitsunobu reaction provides an efficient synthetic
tool for the synthesis of aryl ethers.1 The reaction has the
advantage to take place under mild conditions, and is as
such compatible with a large array of functional groups.
However, the reagents used in this reaction are often hard
to remove, and the parallel solution-phase approach to
Mitsunobu reactions often remains tedious to carry out,
although some newer methods have been reported.2
DEAD 1 and DIAD 2, which are commonly used in
Mitsunobu reactions, can usually be readily removed from
the product mixture by flash chromatography but this
aspect makes them unattractive for parallel synthesis.
New coupling reagents have been reported such as ADDP3
3 and TMAD4 4 (Fig. 1), and other diazocarbonyl equiva-
lents derived from morpholine and N-methylpiperazine,1
but few are commercially available. Di-tert-butylazodi-
carboxylate 5 (DBAD) has also been used in order to avoid
chromatographic separation.5 After reaction, 4 M HCl in
dioxane is added to the reaction mixture, which results in
the decomposition of the azodicarboxylate products into
volatile compounds and hydrazine. However, this tech-
nique is unsuitable for substrates sensitive to acidic
conditions.
The predominant and persistent hurdle in the Mitsun-
obu reaction has been the tedious purification to remove
after reaction the by-product triphenylphosphine oxide as
well as excess triphenylphosphine 6. Thus, many research
O
O
N
N
R
N
O
N
N
O
N
R
O
O
1
2
DEAD (R = Et)
3
ADDP
DIAD (R =
i
Pr)
Abbreviations: ADDP, 1,10-(azodicarbonyl) dipiperidine; DAP-DP,
O
O
N
(p-dimethylaminophenyl)-diphenylphosphine;
DBAD,
di-tert-butyl
N
N
O
azodicarboxylate; DEAD, diethylazodicarboxylate; DIAD, diisopro-
pylazo-dicarboxylate; PS, polystyrene; TAP, tris(dimethylamino)-phos-
phine; TBS, tert-butyl-dimethyl-silyl; TMAD, N,N,N0,N0-tetramethyl-
azodicarbonamide.
N
O
N
N
O
O
5
DBAD
4
TMAD
Fig. 1. Mitsunobu coupling reagents.
*
Corresponding author. Tel.: +33 169 79 12 00; fax: +44 169 09 92 55.
E-mail address: eric.valeur@merck.fr (E. Valeur).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.04.099

E. Valeur, D. Roche / Tetrahedron Letters 49 (2008) 4182–4185
4183
Table 1
groups have focused on developing modified phosphines to
allow easy removal of the phosphines reagent from the
completed reaction mixture, along with the corresponding
phosphine oxide by-products (Fig. 2). DPPE 10 can replace
advantageously triphenylphosphine in the Mitsunobu reac-
tion,6 as the enhanced polarity of its oxide by-product
makes it insoluble in many organic solvents. Other phos-
phines include DAP-DP 8,7 Py-PPh2 9,5 and TAP.8 Solid-
supported triphenylphosphine 11 has become a standard
procedure for Mitsunobu reactions.9 In particular it has
been used along with DBAD 5 for parallel synthesis of
alkyl aryl ethers.10 However as explained before, DBAD
5 is not suitable for all types of substrates.
As part of a chemical biology project, the synthesis of
aryl ethers synthesized using various alkyl diols was tar-
geted. As over 40 aryl ethers had to be synthesized, the
interest of finding conditions for the Mitsunobu reaction
that would be truly applicable for parallel synthesis and
possibly automatized was obvious. In addition, the pres-
ence of a TBS group in some of the substrates required rel-
atively mild reaction conditions, and reagents like DBAD 5
were not applicable due to the need of strong acidic condi-
tions after reaction.
In order to evaluate several methods, the synthesis of
aryl ether 13 was chosen as a representative reaction for
the targeted library (Scheme 1). Therefore, classical condi-
tions using solid-supported triphenylphosphine (purchased
from Novabiochem) and DIAD 2 were first investigated
(Table 1). Surprisingly, the immobilized phosphine per-
formed very poorly. The experiment was repeated with a
different batch of resin, but it still failed to give good
results. Thus, the use of other immobilized triphenylphos-
phines from different suppliers was investigated (Table 1).
Similar to the Novabiochem PS-PPh3, Fluka’s reagent per-
formed poorly, while Polymer Lab’s reagent performed
extremely well, resulting in total conversion. However,
LCMS analysis showed the presence of DIAD by-products
in the mixture that were not detected by ELSD. The differ-
ences observed between immobilized-triphenylphosphines
Screening of immobilized triphenylphosphine using DIAD as coupling
reagent for the synthesis of ether 13
Entry
Phosphine
Puritya
1
2
3
a
PS-PPh3 (Novabiochem)
PS-PPh3 (Fluka)
PS-PPh3 (Polymer Labs)
11
06
100b
Determined by ELSD.
DIAD by-products were not detected by ELSD but their presence was
b
confirmed by MS.
purchased from different suppliers were striking. One rea-
son could be the likely presence of oxide in the PS-PPh3
from Fluka and Novabiochem. Indeed, Polymer lab’s
reagent is manufactured through copolymerization which
minimizes the extent of immobilized phosphine oxide.
Another explanation could be the loading of the resin, as
Fluka’s 3 mmol/g may result in high steric hindrance and
thus poor accessibility to the reactive sites. Polymer Lab’s
1.5 mmol/g in comparison may represent a good compro-
mise. This explanation is, however, not consistent with
Novabiochem’s 1.3 mmol/g resin.
Due to the cost of PS-PPh3, the possibility of using the
so-called ‘water-extractable’ phosphines was investigated
and replacement reagents for DIAD were screened for
the synthesis of aryl ether 13. The combinations of suitable
and commercially available coupling reagents (DIAD 2,
TMAD 4, and ADDP 3) and phosphines (PPh3 6, PBu3
7, DAP-DP 8, PyPPh2 9, DPPE 10, and PS-PPh311) were
investigated in parallel. The conditions were adapted from
the reported synthesis of alkyl aryl ethers,11 and in partic-
ular alkyl aryl ethers involving diols.12 An excess of diol
(1.5 equiv) to the phenol (1 equiv) was used in order to
Table 2
Screening of phosphines and Mitsunobu coupling reagents for the
synthesis of ether 13
Entry
Coupling reagent
DIAD
Phosphine
Puritya
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
PPh3
PBu3
29
51
40
13
15
100b
2
26
1
0
Py-PPh2
DAP-DP
DPPE
PS-PPh3
PPh3
TMAD
ADDP
PBu3
Py-PPh2
DAP-DP
DPPE
PS-PPh3
PPh3
5
100
51
36
72
21
11
100
PBu3
Py-PPh2
DAP-DP
DPPE
PS-PPh3
Bold values indicate the best 2 results.
a
Determined by ELSD.
b
DIAD by-products were not detected by ELSD but their presence was
confirmed by MS.
Fig. 2. Phosphines for the Mitsunobu reaction.

4184
E. Valeur, D. Roche / Tetrahedron Letters 49 (2008) 4182–4185
prevent a double Mitsunobu reaction, and the phosphine
and the coupling reagent were used accordingly (1.5 equiv).
A mixture of DCM and THF was used to allow swelling of
the resin, and the coupling reagent was added portion-wise
over 2 h. After reaction, the mixtures were filtered through
SPE extractors containing a mixed bed of acidic/basic mac-
roporous ion exchange resin, and were used as a mimetic of
aqueous workup for parallel use. In the case of ADDP 3
and TMAD 4, the hydrazine by-products formed during
the reaction (insoluble in THF) were filtered off at the same
time than the elution through SPE extractors. The filtrates
were analyzed by LCMS and HPLC (ELSD) with attention
focused on the presence of phosphine oxide or coupling
reagent by-products (Table 2, Fig. 3).
OTBS
OTBS
1,3-propanediol
PPh3, DIAD
rt, N2, 16 h
O
NH
O
NH
O
OH
12
13
OH
Scheme 1. synthesis of aryl ethers.
Unfortunately, none of the ‘water-extractable’ phos-
phines proved to be removed efficiently from the reaction
mixtures, contrasting with the supported triphenylphos-
phine which gave excellent results. It was noticeable that
TMAD 4 and ADDP 3 did not leave any trace in the mix-
ture in comparison to DIAD 2. Thus, when the supported
triphenylphosphine was used in combination with TMAD
4 or ADDP 3 (entries 12 and 18), no by-product from
the reagents was present in the mixture at the end of the
reaction. TMAD 4 appeared to be more reactive than
ADDP 3 as the solution color rapidly changed from orange
to colorless. These results highlighted the importance in the
choice of the phosphine and only the immobilized triphen-
ylphosphine proved to promote the Mitsunobu reaction
with high purities. To the best of our knowledge this study
Fig. 3. Comparison of the product purity by HPLC (ELSD) with (a) PS-PPh3/TMAD, (b) DAP-DP/ADDP, and (c) DIAD/PPh3.

E. Valeur, D. Roche / Tetrahedron Letters 49 (2008) 4182–4185
4185
be entirely removed from the reaction simply by an acid/
base workup. The origin of the immobilized phosphine also
seems to be critical, as two out of three proved to give very
poor results. Polymer-supported triphenylphosphine (Poly-
mer Laboratories) in conjunction with TMAD is the
reagent of choice to obtain aryl ethers in a purification-free
manner, making high-throughput synthesis via the Mitsun-
obu reaction finally possible.13
12
O
HO
OH
OTBS
N
H
16
17
HO
OTBS
O
HO
OH
N
14
H
HO
HO
References and notes
O
O
O
HO
OH
1. (a) Varasi, M.; Walker, K. A. M.; Maddox, M. L. J. Org. Chem. 1987,
52, 4235–4238; (b) Crich, D.; Dyker, H.; Harris, R. J. J. Org. Chem.
1989, 54, 257–259; (c) Pautard-Cooper, A.; Evans, S. A. J. Org. Chem.
1989, 54, 2485–2488; (d) Hughes, D. L.; Reamer, R. A.; Bergan, J. J.;
Grabowski, E. J. J. J. Am. Chem. Soc. 1988, 110, 6487–6491; (e)
Mitsunobu, O. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press, 1991; Vol. 6, p 65.
NHAc
15
18
Fig. 4. Substrates for the validation library.
Table 3
Library of aryl ethers
2. Dembinski, R. Eur. J. Org. Chem. 2004, 2763–2772.
3. Camp, D.; Jenkins, I. D. Aust. J. Chem. 1992, 45, 47–55.
4. (a) Tsunoda, T.; Yamamiya, Y.; Kawamura, Y.; Ito, S. Tetrahedron
Lett. 1995, 36, 2529–2530; (b) Tsunoda, T.; Ozaki, F.; Shirakata, N.;
Tamaoka, Y.; Yamamoto, H.; Ito, S. Tetrahedron Lett. 1996, 37,
2463–2466.
5. Kiankarimi, M.; Lowe, R.; McCarthy, J. R.; Whitten, J. P. Tetra-
hedron Lett. 1999, 40, 4497–4500.
6. O’Neil, I. A.; Thompson, S.; Murray, C. L.; Barret Kalindjian, S.
Tetrahedron Lett. 1998, 39, 7787–7790.
7. Camp, D.; Jenkins, I. D. Aust. J. Chem. 1988, 41, 1835–1839.
8. Grynkiewicz, G.; Jurczak, J.; Zamojski, A. Tetrahedron 1975, 31,
1411–1414.
9. (a) Tunoori, A. R.; Dutta, D.; Georg, G. I. Tetrahedron Lett. 1998,
39, 8751–8754; (b) Amos, R. A.; Emblidge, R. W.; Havens, N. J. Org.
Chem. 1983, 48, 3598–3600.
Entry
Phenol
Diol
Yield
Puritya
1
2
3
12
14
15
16
17
18
97
88
94
99
96
100
4
5
6
16
17
18
77
77
95
100
100
98
7
8
9
a
16
17
18
69
82
43
100
96
96
Determined by ELSD.
10. (a) Pelletier, J. C.; Kincaid, S. Tetrahedron Lett. 2000, 41, 797–800; (b)
Gentles, R. G.; Wodka, D.; Park, D. C.; Vasudevan, A. J. Comb.
Chem. 2002, 4, 442–456.
11. Rano, T. A.; Chapman, K. T. Tetrahedron Lett. 1995, 36, 3789–3792.
12. (a) Roussel, P.; Bradley, M. Tetrahedron Lett. 1997, 38, 4861–4864;
(b) Roussel, P.; Bradley, M.; Kane, P.; Bailey, C.; Arnold, R.; Cross,
A. Tetrahedron 1999, 55, 6219–6230.
13. Typical procedure for the synthesis of aryl ethers: Polymer-supported
triphenylphosphine (Polymer Laboratories, 1.50 mmol/g, 3 equiv)
was swollen in THF/DCM (1:1, 1.5 mL/100 mg of resin). Under a
nitrogen atmosphere, phenol (1 equiv) and diol (1.5 equiv) were added
followed by the addition of TMAD (1.5 equiv, in solution in THF/
DCM 1:1, 1 mL) over 2 h. The reaction mixture was shaken at room
temperature for 16 h. The mixture was filtered, and the resin was
washed with three cycles of DCM/MeOH. The filtrates were
combined and concentrated in vacuo. The residue was suspended in
cold THF and filtered, and concentration of the filtrate afforded the
aryl ether without the need for further purification.
constitutes the first practical comparison of the newly
developed Mitsunobu coupling reagents and phosphines.
In order to validate the conditions found, a small library
of three phenols and three diols was synthesized (Fig. 4).
The aryl ethers were synthesized in moderate to excellent
yields and excellent purities (Table 3), thus validating the
conditions developed. No purification was required: after
reaction the resin was washed with DCM and MeOH,
and the filtrates were combined and concentrated in vacuo.
The residues were simply taken up in cold THF and fil-
tered, resulting in the removal of the insoluble TMAD
by-products.
In conclusion, the screening of the so-called water-
extractable phosphines proved that these reagents cannot