Synthesis of aryl ethers via a sulfonyl transfer reaction

Article abstract of DOI:10.1021/ol301615z

A general synthesis of aryl ethers from primary and secondary alcohols and aryl mesylates is presented. The reaction proceeds via a sulfonyl-transfer mechanism. In this paper, we compare the sulfonyl transfer reaction to Mitsunobu ether formation. The reaction can be employed in a multistep synthesis where the aryl mesylate is used as a phenol protecting group and then as an activating group for ether formation. This protecting/activating group strategy is demonstrated using raloxifene as the target.

Full text of DOI:10.1021/ol301615z

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of Aryl Ethers via a Sulfonyl
Transfer Reaction
ꢀ
Neal W. Sach, Daniel T. Richter, Stephan Cripps, Michelle Tran-Dube, Huichun Zhu,
Buwen Huang, Jean Cui, and Scott C. Sutton*
Pfizer Global Research and Development, 10770 Science Center Drive, San Diego,
California 92121, United States
scott.sutton@pfizer.com
Received June 12, 2012
ABSTRACT
A general synthesis of aryl ethers from primary and secondary alcohols and aryl mesylates is presented. The reaction proceeds via a sulfonyl-
transfer mechanism. In this paper, we compare the sulfonyl transfer reaction to Mitsunobu ether formation. The reaction can be employed in a
multistep synthesis where the aryl mesylate is used as a phenol protecting group and then as an activating group for ether formation.
This protecting/activating group strategy is demonstrated using raloxifene as the target.
The Mitsunobu reaction1 of phenols and aliphatic alco-
hols is considered an essential reaction for aryl ether
formation (Scheme 1). Its popularity arises from the great
abundance of low molecular weight, accessible alcohols
used as synthons of alkyl halides. The utility of the
Mitsunobu reaction has been demonstrated through ether
formation using stable synthons (i.e., amino alcohols)
equivalent to unstable alkyl halides2 and for inversion of
stereochemistry during ether formation. At the same time,
the Mitsunobu reaction suffers from poor atom economy,3
the necessary removal of byproducts, and the potential
explosive properties of azodicarboxylates, such as DEAD,
when heated.4 Because of these drawbacks, alternative
methods to the Mitsunobu reaction have received some
attention.5 In this paper, we describe an alternative that
reacts primary and secondary alcohols with aryl mesylates
under basic conditionstoform aryl ethers. Although not as
efficient as the Williamson ether synthesis,6 the utility of
this transformation fits a recurrent need to use low molec-
ular weight alcohols from commercial and corporate
collections for SAR studies.7 Although ether formation
has been observed as a side product in the deprotection of
aryl mesylates in alcohol solvents,8 the synthetic utility of
this transformation has not been systematically studied.
Scheme 1. Activation of Alcohols for Aryl Ether Synthesis
Aryl mesylates have generally been regarded as pro-
tected phenols, until recently.9 Although the use of the
aryl triflates for cross coupling is ubiquitous in organic
synthesis, aryl mesylates are differentiated because of
their excellent stability to acidic and mildly basic condi-
tions, making them a useful phenol protecting group.10
(1) (a) Mitsunobu, O. Synthesis 1981, 1, 1. (b) Swamy, K. C. K.;
Kumar, N. N. B.; Balaraman, E.; Kumar, K. V. P. P. Chem. Rev. 2009,
109, 2551.
(2) For example, free-base β-chloroamines have a propensity to ring
close spontaneously to the corresponding aziridinium salts. See: Hickmott,
P. W.; Wood, S.; Murray-Rust, P. J. Chem. Soc., Perkin Trans. 1 1985, 2033.
(3) Li, C-J; Trost, B. M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105 (36),
13197.
(4) Berger, A.; Wehrstedt, K. D. J. Loss Prev. Process Ind. 2010, 23,
734.
(5) (a) Barrett, A. G. M.; Braddock, D. C.; James, R. A.; Koike, N.;
Procopiou, P. A. J. Org. Chem. 1998, 63 (18), 6273. (b) Iranpoor, N.;
Firouzabadi, H.; Khalili, D. Org. Biomol. Chem. 2010, 8 (19), 4436.
(c) Hawryluk, N. A.; Snider, B. B. J. Org. Chem. 2000, 65 (24), 8379.
(6) Fuhrmann, E.; Talbiersky, J. Org. Process Res. Dev. 2005, 9 (2), 206.
(7) Zhu, G.-D.; Gandhi, V. B.; Gong, J.; Thomas, S.; Woods, K. W.;
Song, X.; Li, T.; Diebold, R. B.; Luo, Y.; Liu, X. J. Med. Chem. 2007, 50
(13), 2990.
(8) Bensel, N; Pevere, V; Desmurs, J. R.; Wagner, A.; Mioskowski, C.
Tetrahedron Lett. 2002, 43 (23), 4281.
r
10.1021/ol301615z
XXXX American Chemical Society

At the same time, aryl mesylates undergo a synthetically
useful reaction to form aryl ethers.11 Our initial investiga-
tion studied the nature of four different sulfonates. Using
sulfonates derived from 6-hydroxy quinoline (1bÀe)
paired with 3-(pyridin-4-yl)propan-1-ol (2), we found
that methanesulfonate 1c afforded the highest yield
(52%) when using our initial conditions of 1,4-dioxane
as solvent and Cs2CO3 as base at 100 °C (Table 1).
(conditions D). LiO-t-Bu in DMSO led to a 74% yield of the
reaction of 1c with 2 (Figure 1), a significant improvement
over the 52% achieved using Cs2CO3, and 1,4-dioxane at
100 °C. When the A and B conditions were applied to a variety
of aryl mesylates while keeping the alcohol constant, several
interesting features were noted (Table 2). Conditions A using
NaO-t-Bu in acetonitrile at 80 °C was the most general
affording 63À77% isolated yields for five of the six reactions.
The hindered o-methyl substrate (entry 2) furnished a 77%
yield under the NaO-t-Bu in acetonitrile conditions, while the
electron-rich p-OMe mesylate (entry 3) afforded a 77% yield
under the same conditions. The example affording a higher
yield using the weaker base, Cs2CO3, was entry 4 where a CN
group para to the mesylate led to a 32% yield when conditions
A was used and a 61% yield for conditions B.
Table 1. Arylsulfonates Derived from 6-Hydroxyquinoline
Table 2. Mesylate Transfer Using Alcohol 2 and Various ArOMsa
a Isolated, chromatographically purified. b 0.05 equiv of Pd2(dba)3
and 0.1 equiv of BippyPhos added. c Based on LCMS analysis
Based on these preliminary results, we used 1c to further
optimize our reaction conditions using microscale techni-
ques to screen a cross of six solvents and seven bases at
elevated temperatures (80 and 120 °C) in an effort to improve
upon the 52% yield.12 The screening results and subsequent
follow-up work led to identification of four optimal condi-
tions: NaO-t-Bu in acetonitrile, at 80 °C (conditions A),
Cs2CO3 in DMF at 100 °C (conditions B), NaH in DMF
at 70 °C (conditions C), and LiO-t-Bu in DMSO at 150 °C
(9) (a) So, C. M.; Kwong, F. Y. Chem. Soc. Rev. 2011, 40 (10), 4963.
(b) Li, B.-J.; Yu, D.-G.; Sun, C.-L.; Shi, Z.-J. Chem.;Eur. J. 2011, 17
(6), 1728. (c) Rosen, B. M.; Quasdorf, K. W.; Wilson, D. A.; Zhang, N.;
Resmerita, A.-M.; Garg, N. K.; Percec, V. Chem. Rev. 2011, 111 (3),
1346. (d) Molander, G. A.; Shin, I. Org. Lett. 2012, 14 (12), 3138. (e)
Leowanawat, P.; Zhang, N.; Safi, M.; Hoffman, D. J.; Fryberger, M. C.;
George, A.; Percec, V. J. Org. Chem. 2012, 77 (6), 2885. (f) Chang,
J. W. W.; Chia, E. Y.; Chai, C. L. L.; Seayad, J. Org. Biomol. Chem. 2012,
10 (11), 2289. (g) Chung, K. H.; So, C. M.; Wong, S. M.; Luk, C. H.;
Zhou, Z.; Lau, C. P.; Kwong, F. Y. Chem. Commun. (Cambridge, U. K.)
2012, 48 (14), 1967. (h) Leowanawat, P.; Zhang, N.; Percec, V. J. Org.
Chem. 2012, 77 (2), 1018. (i) Muto, K.; Yamaguchi, J.; Itami, K. J. Am.
Chem. Soc. 2012, 134 (1), 169. (j) Leowanawat, P.; Zhang, N.; Resmerita,
A.-M.; Rosen, B. M.; Percec, V. J. Org. Chem. 2011, 76 (24), 9946.
(k) Shen, C.; Yang, G.; Zhang, W. Org. Biomol. Chem. 2012, 10 (17), 3500.
(l) Wu, X.; Fors, B. P.; Buchwald, S. L. Angew. Chem., Int. Ed. 2011, 50
(42), 9943. (m) Gowrisankar, S.; Sergeev, A. G.; Anbarasan, P.; Spannenberg,
A.; Neumann, H.; Beller, M. J. Am. Chem. Soc. 2010, 132 (33), 11592.
(n) Naidu, A. B.; Jaseer, E. A.; Sekar, G. J. Org. Chem. 2009, 74 (10),
3675. (o) Altman, R. A.; Shafir, A.; Choi, A.; Lichtor, P. A.; Buchwald,
S. L. J. Org. Chem. 2008, 73 (1), 284. (p) Shafir, A.; Lichtor, P. A.;
Buchwald, S. L. J. Am. Chem. Soc. 2007,129 (12), 3490. (q) Dooleweerdt, K.;
Fors, B. P.; Buchwald, S. L. Org. Lett. 2010, 12 (10), 2350. (r) Fors, B. P.;
Watson, D. A.; Biscoe, M. R.; Buchwald, S. L. J. Am. Chem. Soc. 2008,
130 (41), 13552.
a Conditions A: NaOt-Bu, CH3CN, 80 °C, 16 h. Conditions B:
Cs2CO3, DMF, 100 °C, 3 h. Yields represent isolated material on
1 mmol scale.
These results suggested that the mesylate-transfer reaction
can tolerate a variety of aryl mesylates including electron-rich,
electron-poor, and ortho-substitution13 when paired with a
primary alcohol. The results also indicated that the nature of
the aryl mesylate substrate can be sensitive to the strength of
the base used. With this information in hand, we turned our
(13) Using 2,6-dimethylphenyl methanesulfonate was attempted but
led to only trace amounts of product.
(14) (a) Hughes, D. L. Org. React. (Hoboken, NJ) 1992, 42. (b) Bisi,
A.; Rampa, A.; Budriesi, R.; Gobbi, S.; Belluti, F.; Ioan, P.; Valoti, E.;
Chiarini, A.; Valenti, P. Bioorg. Med. Chem. 2003, 11 (7), 1353.
(c) Kiesewetter, D. O.; Eckelman, W. C. J. Labelled Compd. Radiopharm.
2004, 47 (13), 953. (d) Miko, T.; Ligneau, X.; Pertz, H. H.; Ganellin,
C. R.; Arrang, J.-M.; Schwartz, J.-C.; Schunack, W.; Stark, H. J. Med.
Chem. 2003, 46 (8), 1523. (e) Inversion and determination of ee were
confirmed using chiral SFC with the appropriate standards. Kaufman,
T. S. Tetrahedron Lett. 1996, 37 (30), 5329. (f) The control Mitsunobu
reaction was ran at Pfizer-LJ.
(10) Looker, J. H.; Thatcher, D. N. J. Org. Chem. 1954, 19, 784.
(11) Sach, N. W.; Sutton, S. C.; Richter, D. T.; Cripps, S.; Zhu, H.;
ꢀ
Tran-Dube, M.; Cui, J. Abstracts of Papers, 239th ACS National Meeting,
San Francisco, CA, United States, Mar 21À25, 2010, ORGN-74.
(12) Results from this microscale screen are presented in the Support-
ing Information.
B
Org. Lett., Vol. XX, No. XX, XXXX

attention to benchmark this transformation alongside the
Mitsunobu reaction in an effort to evaluate the synthetic
utility of this reaction. Comparison of mesylate transfer to the
Mitsunobu reaction is shown in Figure 1. Mitsunobu yields
from the literature14 or from control reactions14f are tabulated
adjacent to yields using mesylate transfer. Although forma-
tion of the aryl mesylate, itself, needs to be considered in
the overall yield of this two-step process, a straightforward
comparison of the yields starting with the ArOMs precursors
is listed in Figure 1 along with the Mitsunobu yields.
form sulfene16 cannot be ruled out.17 In Table 1, we
observed lower yields when the sulfonyl group lacks a
hydrogen on the carbon alpha to the sulfur. These aryl-
sulfonates are unable to go through path b. However, there
are too many other factors involved to infer any mechanistic
details from this observation. In Table 2 and Figure 1, we
observed the weaker base, Cs2CO3, to give improved yields
for the more stabilized 4-cyanophenolate and 4-chlorophe-
nolate leaving groups (entry 4, Table 2 and compounds 12
and 13 of Figure 1). These observations led to a hypothesis
that path b is in play for ArOMs bearing an electron-
withdrawing group para to the sulfonate. It was hypothe-
sized that both paths a and b are possible, depending on the
nature of the aryl mesylate and the base employed.
Scheme 2. Proposed Mechanisms
To study the mechanism further, we studied the inter-
mediacy of the alkyl mesylate which was believed to form
with retention of alcohol stereochemistry during the trans-
fer step. To prove this, we used a stereochemical probe
substrate, namely meso-(2R,4R,6S)-2,6-dimethyltetrahy-
dro-2H-pyran-4-ol having all-cis stereochemistry18 as
the starting alcohol in the synthesis of compound 15
(Figure 1). Interception of the secondary mesylate inter-
mediate with retention of configuration was accomplished
by observing the reaction, in progress, using 1H NMR.19,20
Having established that the mesylate-transfer step goes
with retention of configuration, we then considered the
second step of the mechanism. Since Gordon and co-workers
provided evidence covering a concerted bimolecular dis-
placement at the sulfonyl group (path a),15 we sought a
method to examine path b, especially in the context of the
p-CN substrate (entry 4, Table 2) which was an outlier
giving higher yields under conditions that are unlikely to
form the fully deprotonated alcohol. After conducting a
literature search around the keywords “sulfene trap”,
a paper by Pregel and Buncel21 was found that fully supported
Figure 1. Comparison to Mitsunobu ether formation.
For compounds 3, 11, and 12 equal or improved yields
were obtained using the mesylate transfer method. On the
other hand, the Mitsunobu reaction led to higher yields for
compounds 10, 13, and 15. For compound 14, the use of
NaH in DMF at 70 °C (conditions C) afforded inversion of
the chiral secondary alcohol to the inverted aryl ether in
41% yield and 96% ee compared to the Mitsunobu reaction
leading to a 48% yield and 91% ee. Kaufman demonstrated
that benzylic alcohols with o-methoxy groups are known to
undergo partial racemization during Mitsunobu inversion
thought to arise from a SN1 reaction pathway which could
explain the slight difference in ee results here.14e
Scheme 2 shows two reasonable mechanisms for the
reaction, both going through SN2 inversion as the final step.
For sulfonyl transfer, Gordon and co-workers examined a
variety of potential mechanisms15 and concluded that a
concerted bimolecular displacement at the sulfonyl group
(path a, Scheme 2) is the dominant pathway. On the other
hand, path b involving E2 elimination of the phenolate to
(18) Gelin, S.; Gelin, R.; Henry, R. Bull. Soc. Chim. Fr. 1975, 1À2
(Part 2), 302.
(19) The 1H NMR data for this experiment is shown in the Suporting
Information section.
(15) Gordon, I. M.; Maskill, H.; Ruasse, M.-F. Chem. Soc. Rev 1989,
18, 123.
(20) cis-2,6-Dimethyltetrahydro-2H-pyran-4-yl methanesulfonate:
1H NMR (400 MHz, DMF) δ ppm 4.79À4.92 (m, 1H), 3.49À3.60 (m,
2H), 3.27 (s, 3H), 2.06À2.17 (m, 2H), 1.28 (q, J = 11.37 Hz, 2H),
1.13À1.18 (m, 6H). trans-2,6-Dimethyltetrahydro-2H-pyran-4-yl
methanesulfonate: 1H 1H NMR (400 MHz, DMF) δ ppm 5.13 (t, J =
2.91 Hz, 1H), 3.77 (ddd, J = 1.52, 6.19, 11.49 Hz, 2H), 3.27 (s, 3H), 1.93
(dd, J = 2.78, 14.91 Hz, 2H), 1.49 (ddd, J = 2.65, 11.62, 14.27 Hz, 2H),
1.11 (d, J = 6.32 Hz, 6H).
(16) King, J. F.; Lee, T. S. J. Am. Chem. Soc. 1969, 91 (23), 6524.
(17) For reactions of sulfene with alcohols via 1,2-addition, see: (a)
Skrypnik, Y. G.; Bezrodnyi, V. P.; Panov, V. P.; Baranov, S. N. Katal.
Sint. Org. Soedin. Sery. 1979, 197. (b) Langendries, R.; De Schryver,
F. C.; De Mayo, P.; Marty, R. A.; Schutyser, J. J. Am. Chem. Soc. 1974,
96 (9), 2964. (c) Reich, M. F.; Lee, V. J.; Ashcroft, J.; Morton, G. O.
J. Org. Chem. 1993, 58 (19), 5288. (d) Crossland, R. K.; Servis, K. L.
J. Org. Chem. 1970, 35 (9), 3195.
(21) Pregel, M. J.; Buncel, E. J. Chem. Soc., Perkin Trans. 2 1991, 307.
Org. Lett., Vol. XX, No. XX, XXXX
C

the proposed contingent mechanisms in Scheme 2. Pregel and
Buncel used an enamine as a sulfene trap in their experiments,
and they examined a variety of ArOMs substrates (Ar =
p-NO2, m-NO2, p-CF3) using KOEt as the base. They found
that in the p-NO2 case, 13% of the sulfene adduct, a 4-mem-
bered ring sulfone, was observed. Deuterium exchange studies
further suggested that the elimination pathway goes through
an E1cb-type mechanism. For their p-NO2 case, they con-
cluded it to react predominantly through an E1Cb-type
elimination via a sulfene intermediate with substitution
(path a) as a minor concurrent pathway. Based on similar
electronic properties of the p-NO2 and p-CN substrates, we
believe the same E1Cb-type elimination occurs as the pre-
dominant pathway with concurrent substitution as a minor
pathway in the formation of 7 (Table 2) and possibly 13
(Figure 1). This would help explain mesylate transfer under
weakly basic conditions where a fully deprotonated alkoxide
is not required. Pregel and Buncel found very little sulfene
adduct in the m-NO2, and p-CF3 cases and concluded that
substitution (path a) was the dominant pathway for these less
stabilized phenolate leaving groups.
demonstrated using calculated phenol pKa values to pre-
dict regioselectivity in cases where multiple ArOMs groups
are present in a single intermediate.
Scheme 3. Raloxifene Synthesis
As previously mentioned, the excellent stability of the
aryl mesylate group to acidic and mildly basic conditions
make it a useful phenol protecting group. The current method
is therefore applicable to multistep synthesis by using the
mesylate, first as a protecting group, and then as an activating
group for aryl ether formation. Raloxifene,22 an important
drug used to treat osteoporosis and to decrease the risk of
developing invasive breast cancer in women, was selected as
an suitable target to highlight this strategy. Based on the
calculated phenol pKa values of the tris-phenol (Scheme 3),
we anticipated mesylate transfer on 18 to selectively occur at
the most acidic phenol, para to the ketone (pKa = 7.44).
FriedelÀCrafts acylation of bis-mesylate 1623 with me-
sylate protected 4-hydroxybenzoylchloride (17) provided
18 in 62% yield. The key step involved reaction of 18 with
2-(piperidin-1-yl)ethanol under the NaOt-Bu/acetonitrile
conditions at 80 °C leading to 19 in 22% yield.24 The
identityof19was confirmedafterdeprotection using KOH
in DMSO to afford 20 (67%). Compound 20 was con-
firmed to be Raloxifene by comparison to an authentic
sample. In this sequence, efforts to improve the yield of 19
were not pursued agressively since the primary purpose of
the study was to demonstrate application of the ArOMs
group as both protecting and activating group in the
synthesis of a credible target. Just as important, the study
Insummary, the mesylate transfer reaction represents an
alternative to the Mitsunobu reaction for aryl ether for-
mation. Application of this reaction to the recurrent need
to rapidly examine a diverse pool of readily available
aliphatic alcohols for SAR studies off a phenol template
is where this reaction stands out. This is particularly true in
cases where the Mitsunobu reaction fails to deliver reason-
able yields or purity. The application of microscale screen-
ing led totwo generalreaction conditions: (A) NaO-t-Bu in
CH3CN at 80 °C and (B) Cs2CO3 in DMF at 100 °C. Other
conditions (C) NaH in DMF at 70 °C and (D) LiO-t-Bu in
DMSO at 150 °C were found through more traditional
reaction optimization techniques. Yields comparable to
the Mitsunobu reaction were observed for primary alco-
hols, while secondary alcohols tended to afford lower
yields. Finally, the reaction can be employed in multistep
synthesis where the mesylate is used as a protecting group
and activating group for ether formation. Opportunities
for further improvement include telescoping mesylate
formation and mesylate transfer to a two-step, one-pot
procedure using the same solvent and base for both reac-
tions. If successful, a simplified one-pot procedure could
add to the growing arsenal of green chemistry available to
the synthetic chemist with improved atom economy rela-
tive to the Mitsunobu reaction.25
(22) Pineiro-Nunez, M. Raloxifene, Evista: a Selective Estrogen
Receptor Modulator (SERM) in Modern Drug Synthesis; Li, J. J.,
Johnson, D. S., Eds.; 2010; pp 309À327.
Acknowledgment. This work was supported by senior
management at Pfizer La Jolla. Wegratefully acknowledge
their support to conduct this work which concurrently led
to measurable impact on multiple projects.
(23) Jones, C. D.; Jevnikar, M. G.; Pike, A. J.; Peters, M. K.; Black,
L. J.; Thompson, A. R.; Falcone, J. F.; Clemens, J. A. J. Med. Chem.
1984, 27 (8), 1057.
(24) This is an unoptimized yield with side products identified as
regioisomeric monophenol intermediates/hydrolysis byproduct as well
as unreacted starting material.
(25) A typical Minsunobu reaction produces 464 g of reagent by-
product for each mole of product formed (assuming 100% yield). The
two-step reaction using Et3N as base in the mesylate formation step and
NaO-t-Bu in the MsXfer step produces 329 g of reagent byproduct for
each mole of product (assuming 100% yield). This represents a 29%
improvement in atom economy that can be envisaged by telescoping the
reaction to a one-solvent/one-base two-step, one-pot procedure.
Supporting Information Available. Experimental pro-
cedures and spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX