Palladium-Catalyzed Synthesis of Bis-Substituted Sulfoxonium Ylides

Article abstract of DOI:10.1021/acs.orglett.8b03744

The lack of general access to bis-substituted sulfoxonium ylides is addressed by developing a palladium-catalyzed C-H cross-coupling of α-ester sulfoxonium ylides with (hetero)aryl iodides, bromides, and triflates. Three different catalysts have been eval

Full text of DOI:10.1021/acs.orglett.8b03744

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Palladium-Catalyzed Synthesis of Bis-Substituted Sulfoxonium
Ylides
Christopher Janot,† Pierre Palamini,† Benjamin C. Dobson,‡ James Muir,‡ and Christophe Aïssa*,†
†Department of Chemistry, University of Liverpool, Crown Street, Liverpool L69 7ZD, United Kingdom
‡Pharmaceutical Technology and Development, AstraZeneca R&D, Silk Road Business Park, Charter Way, Macclesfield, Cheshire
SK10 2NA, United Kingdom
S
* Supporting Information
ABSTRACT: The lack of general access to bis-substituted sulfoxo-
nium ylides is addressed by developing a palladium-catalyzed C−H
cross-coupling of α-ester sulfoxonium ylides with (hetero)aryl iodides,
bromides, and triflates. Three different catalysts have been evaluated.
This method is amenable to the late-stage functionalization of active
pharmaceutical ingredients.
Scheme 1. General Access to (Hetero)aryl-Substituted
Sulfoxonium Ylides by Palladium-Catalyzed C−H Cross-
Coupling
ulfoxonium ylides have recently gained prominence as
Spotentially safer surrogates of diazo compounds in metal-
catalyzed reactions. Thus, the iridium-catalyzed formation of
carbon−heteroatom bonds from monosubstituted α-carbonyl
sulfoxonium ylides has been optimized in industry for the
multi-kilogram synthesis of drug candidates,1,2 and similar
transformations can be promoted by other metal catalysts on
smaller scale.3−7 In addition, α-ketone sulfoxonium ylides can
be used in rhodium-catalyzed functionalization reactions of
carbon−hydrogen bonds of non-nucleophilic substrates,8−21
and similar cross-couplings have been developed with
ruthenium22,23 and cobalt complexes,24 whereas the C−H
functionalization of nucleophilic α,β-unsaturated β-amino-
esters has been described with an iridium catalyst.25 Finally,
the cross-dimerization of sulfoxonium ylides and diazo acetate
derivatives has been reported recently.26
reaction of α-carbonyl sulfoxonium ylides with arynes;30
however, the structural variation of the precursors of these
intermediates will be limited by the required multistep
synthesis, and the regioselectivity will remain in many cases
an issue that is inherent to aryne chemistry.31 Inspired by the
palladium-catalyzed cross-coupling of aryl iodides with ethyl
diazoacetate,32 we assumed that a similar C−H functionaliza-
tion of simple and stable α-ester sulfoxoniums could proceed
efficiently. Whereas the conditions described previously for the
synthesis of donor−acceptor diazo compounds by C−H
functionalization32−36 failed to deliver the expected sulfoxo-
nium ylides in respectable amounts during our preliminary
investigations, we report herein the successful deployment of
this strategy after identification of novel conditions that are
applicable to a wide set of substrates. To the best of our
The superior thermal stability of sulfoxonium ylides over the
parent diazo compounds is easily illustrated with compounds 1
and 2 (Scheme 1). Thus, the decomposition of diazo
compound 1 occurs at 115 °C and liberates 706 J/g27 while
sulfoxonium ylide 2 undergoes decomposition at 172 °C,
which liberates only 435 J/g. Accordingly, it would be desirable
to develop a general access to (hetero)aryl-substituted
sulfoxonium ylides like 2 that do not rely on the synthesis
and use of donor−acceptor diazo compounds like 1, in
contrast to the current state-of-the-art.4 Although iodoniums
have also been reported as precursors of sulfoxonium ylides,28
the optimized conditions are not suitable for the synthesis of
ylides that are substituted by both an electron-acceptor and an
electron-donor like in 2. Moreover, it is possible to conduct the
heteroarylation of dimethylsulfoxonium methylide followed by
an acylation to obtain bis-substituted sulfoxonium ylides, but
the direct nucleophilic aromatic substitution is limited to
reactive heterocycles.29 An alternative approach relies on the
Received: November 22, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03744
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
knowledge, the reaction described in this Letter is the first
example of a palladium-catalyzed transformation of sulfoxo-
nium ylides.37
We began our study with model substrate 3 and
iodobenzene (Table 1). A rapid survey of bases (entries 1−
the scope of this reaction. We thus examined the combination
of Pd2(dba)3 (5 mol %) either with a reduced amount of PPh3
(entry 17), or with other ligands (entries 18−21).
The choice of these alternative ligands was based on the
assumption that the reductive elimination from III might be
the rate-limiting step of the catalytic cycle (Scheme 2). While
a
Table 1. Optimization of the Reaction Conditions
Scheme 2. Plausible Mechanism of the Cross-Coupling
(Ligands Are Omitted for Clarity)
b
entry
X
catalyst
base
solvent
yield
c
1
2
3
4
5
6
7
8
I
I
I
I
I
I
I
I
I
I
I
I
I
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
DBU
Et3N
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
toluene
dioxane
1,2-DCE
DMF
2%
2%
2%
45%
5%
2%
29%
25%
18%
22%
14%
24%
63%
80%
87%
c
c
c
c
c
c
c
c
c
c
c
t-BuOK
Cs2CO3
K2CO3
Li2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
further studies of the mechanism would be required to
demonstrate our hypothesis, two observations support it. First,
the better results obtained with bromobenzene as compared to
iodobenzene (Table 1, entries 13 vs 14) indeed suggest that
the oxidative addition of the catalyst into the carbon−halogen
bond to give I is not rate-limiting. Second, using Ag2CO3 as
base instead of Cs2CO3 did not promote the formation of 4,
which suggests that attempting to accelerate the trans-
metalation of I to II by forming a cationic palladium
intermediate is not a favorable strategy. To the extent that
the deprotonation of II into III is facile, the reductive
elimination from III would therefore be the rate-limiting step,
and either sterically demanding or electron-poor ligands should
accelerate that step.
9
10
11
12
13
14
15
16
17
18
19
20
EtOH
THF
d
d
d
e
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
Br
OTf
Br
Br
Br
Br
Br
Br
73%
64%
0%
0%
80%
86%
f
d
Pd2(dba)3 + 4L1
Pd2(dba)3 + 4L2
Pd2(dba)3 + 4L3
Pd2(dba)3 + 4L4
Pd2(dba)3 + 4L5
f
c
c
e
e
We found that reducing the amount of PPh3 did not prevent
the formation of its oxide as side product (entry 17). Using
electron-poor phosphites L2 and L3 did not give any product
(entries 18 and 19). Gratifyingly, bulky phosphines L4 and L5
gave 4 in high yield after purification by flash chromatography
without need for recrystallization (entries 20 and 21). Overall,
we have thus identified three suitable catalysts: Pd(PPh3)4
(method A), Pd2(dba)3 + 4L4 (method B), and Pd2(dba)3 +
4L5 (method C). Although method A required further
recrystallization of compound 4, we envisioned that less
polar products might be more easily separated from
triphenylphosphine oxide, and we decided to evaluate the
three methods on a set of commercially available (hetero)aryl
bromides.
As envisioned after the optimization study, methods B and C
performed generally better than method A (Figure 1). Indeed,
compounds 5, 6, 8−10, 13, 16, and 18−22 were obtained in
superior yields when using these two methods. However,
method A is not without merits, and compounds 7, 12, 15, and
17 could be obtained in high yields and in pure form by this
method after flash chromatography without recourse to further
purification by recrystallization. Moreover, method A gave only
monofunctionalized compound 14 from 1,3-dibromobenzene,
even in the presence of an excess of reagent 3, whereas
methods B and C led to an intractable mixture of mono- and
bis-functionalized products. In general, electron-poor or
electron-neutral aryl bromides led to excellent conversion
(5−10, 12−14), whereas para-methoxybenzene bromide failed
to deliver 11, which is in agreement with our hypothesis that
f
f
f
21
a
The PhX/3 ratio is 1.15:1 in entries 1−12 and 1:2.5 in entries 13−
b
c
21. 2 equiv (entries 1−15) or 1.1 equiv (entries 16−21). Yield
determined by HPLC after 4 h. Yield of isolated product after 16 h;
the product contains 5−10 mol % of triphenylphosphine oxide. Yield
of pure isolated product after 16 h. 5 mol % of Pd2(dba)3 + 20 mol %
of L1−L5. 1,2-DCE, 1,2-dichloroethane; DMF, dimethylformamide;
THF, tetrahydrofuran.
d
e
f
6) showed the most promising results when Cs2CO3 was used.
Among the solvents examined (entries 4, 7−12), acetonitrile
appeared optimal. Increasing the ratio of 3 to iodobenzene had
a positive effect on the yield of product 4 (entry 13).
Remarkably, in contrast to the reactivity described for ethyl
diazoacetate,2 the reaction of 3 with bromobenzene gave a
better result than that obtained with iodobenzene (entry 14),
whereas benzene triflate also gave an excellent yield (entry 15).
We could decrease the amount of base to 1.1 equiv without
negative effects (entry 16). These results were therefore
encouraging. However, polar product 4 was always obtained
with small amounts of triphenylphosphine oxide after flash
chromatography. Presumably, this side product is formed by
the oxidation of free triphenylphosphine by dimethyl
sulfoxide.38 Although compound 4 could be obtained in pure
form after recrystallization in 73% yield, we decided to identify
alternative conditions before embarking on the exploration of
B
DOI: 10.1021/acs.orglett.8b03744
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
method C, illustrating the good functional group tolerance of
this reaction and its synthetic utility (Scheme 3). We used the
API either directly, as in the case of nicergoline, or its triflate,
as in the case of estrone, and compounds 25 and 26 were
obtained in very good yields.
Scheme 3. Late-Stage Modification of API
Sulfoxonium ylides are amenable to a wide array of
transformations.37,39 For example, 4 and para-hydroxythiophe-
nol reacted smoothly to give 27 (Scheme 4),6 a precursor of a
Scheme 4. Utility of α-Ester Bis-Substituted Sulfoxonium
Ylides for the Synthesis of Drug Candidates and for Further
Functionalization
Figure 1. Scope of cross-coupling products obtained from (hetero)-
aryl bromides and unsubstituted α-ester sulfoxonium ylides. Yields of
pure isolated products from the reactions of (hetero)aryl bromides
(0.20 mmol, 1 equiv), unsubstituted α-ester sulfoxonium ylides (0.50
mmol, 2.5 equiv), and Cs2CO3 (0.22 mmol, 1.1 equiv) in MeCN (2
mL) at 80 °C for 14−16 h in the presence of either Pd(PPh3)4 (0.02
mmol, 0.1 equiv) (method A), Pd2(dba)3 (0.01 mmol, 0.05 equiv),
and L4 (0.04 mmol, 0.2 equiv) (method B) or Pd2(dba)3 (0.01 mmol,
0.05 equiv) and L5 (0.04 mmol, 0.2 equiv) (method C). (a) Phenyl
bromide (7 mmol, 1 equiv), Pd2(dba)3 (0.18 mmol, 0.025 equiv), and
L5 (0.7 mmol, 0.1 equiv) were used. (b) Pd(Pt-Bu3)2 (0.02 mmol, 0.1
equiv) was used.
the reductive elimination might be the rate-limiting step in the
catalytic cycle. Ortho-substitution in the aryl bromide was
tolerated in the case of cyano (15) and nitro (16) groups, but
ortho-bromotoluene failed to deliver the expected product in
either of the methods. The reactions of heteroaryl bromides
are exemplified with a 3-pyridyl (17), a 2-quinolyl (18), and a
3-thiophenyl (19) group. Besides the reaction of ethyl ester 3,
other sulfoxonium ylides bearing an alkyl or aryl group also
underwent the cross-coupling to give the expected products (2,
20−22) in excellent yields. The synthesis of 4 could be
performed on the gram scale in 83% yield according to method
C, and decreasing the amount of Pd2(dba)3 to 2.5 mol % on
that scale still afforded 4 in 69% yield.
fatty acid synthase inhibitor.40 Therefore, the shelf-stable white
solid 4 that is now conveniently available by our method also
offers a valuable alternative feedstock to ethyl α-bromopheny-
lacetate, a lacrymator that was used previously for the synthesis
of this bioactive compound.40 Finally, the ease of access to α-
ester bis-substituted sulfoxonium ylides, as enabled by the
cross-coupling described herein, could allow a full exploration
of the scope of these ylides in the iridium-catalyzed N−H
insertion reaction pioneered by Merck,4a as illustrated with the
reaction of 10 leading to 28.
In conclusion, we have described the first palladium-
catalyzed cross-coupling of sulfoxonium ylides for the rapid
synthesis of bis-substituted α-ester-α-(hetero)aryl sulfoxonium
ylides from easily available (hetero)aryl bromides and triflates.
This method provides a valuable alternative to the metal-
catalyzed decomposition of diazo precursors that have a high
thermal potential.4 Moreover, this work expands the scope of
metal-catalyzed cross-coupling of carbene precursors.41 Further
studies to generalize this reaction to other coupling partners
are ongoing in our laboratories.42
As observed in our optimization study, aryl triflates are also
convenient substrates of the cross-coupling with sulfoxonium
ylides. Indeed, we could couple chromene derivative 23 and
ylide 3 according to method B and obtain compound 24 in
77% yield (eq 1).
Moreover, we could also couple structurally complex active
pharmaceutical ingredients (API) in high yields by following
C
DOI: 10.1021/acs.orglett.8b03744
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(19) You, C.; Pi, C.; Wu, Y.; Cui, X. Adv. Synth. Catal. 2018, 360,
4068−4072.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b03744.
■
S
(20) Chen, G.; Zhang, X.; Jia, R.; Li, B.; Fan, X. Adv. Synth. Catal.
2018, 360, 3781−3787.
(21) Wu, X.; Xiong, H.; Sun, S.; Cheng, J. Org. Lett. 2018, 20, 1396−
1399.
(22) Shi, X.; Wang, R.; Zeng, X.; Zhang, Y.; Hu, H.; Xie, C.; Wang,
M. Adv. Synth. Catal. 2018, 360, 4049−4053.
Detailed experimental procedures and characterization
of all new compounds including DSC trace and NMR
spectra (PDF)
(23) Liang, Y. F.; Yang, L.; Rogge, T.; Ackermann, L. Chem. - Eur. J.
2018, 24, 16548−16552.
(24) Ji, S.; Yan, K.; Li, B.; Wang, B. Org. Lett. 2018, 20, 5981−5984.
(25) Vaitla, J.; Hopmann, K. H.; Bayer, A. Angew. Chem., Int. Ed.
2017, 56, 4277−4281.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: aissa@liverpool.ac.uk.
ORCID
(26) Neuhaus, J. D.; Bauer, A.; Pinto, A.; Maulide, N. Angew. Chem.,
Int. Ed. 2018, 57, 16215−16218.
(27) Bartrum, H. E.; Blakemore, D. C.; Moody, C. J.; Hayes, C. J.
Chem. - Eur. J. 2011, 17, 9586−9589.
Christophe Aïssa: 0000-0003-0750-9435
Notes
(28) Vaitla, J.; Hopmann, K. H.; Bayer, A. Org. Lett. 2017, 19,
6688−6691.
(29) (a) Yamanaka, H.; Konno, S.; Sakamoto, T.; Niitsuma, S.; Noji,
S. Chem. Pharm. Bull. 1981, 29, 2837−2843. (b) Faragher, R.;
Gilchrist, T. L. J. Chem. Soc., Perkin Trans. 1 1977, 1, 1196−1200.
(30) Talero, A. G.; Martins, B. S.; Burtoloso, A. C. B. Org. Lett.
2018, 20, 7206−7209.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful for the financial support from the University of
Liverpool for a studentship to C.J. and AstraZeneca for a
CASE award.
(31) Selected reviews: (a) Pellissier, H.; Santelli, M. The use of
arynes in organic synthesis. Tetrahedron 2003, 59, 701−730.
(b) Bhunia, A.; Yetra, S. R.; Biju, A. T. Chem. Soc. Rev. 2012, 41,
3140. (c) Goetz, A. E.; Garg, N. K. J. Org. Chem. 2014, 79, 846−851.
(32) Peng, C.; Cheng, J.; Wang, J. J. J. Am. Chem. Soc. 2007, 129,
8708−8709.
REFERENCES
■
(1) Mangion, I. K.; Ruck, R. T.; Rivera, N.; Huffman, M. A.; Shevlin,
M. Org. Lett. 2011, 13, 5480−5483.
(2) Molinaro, C.; Bulger, P. G.; Lee, E. E.; Kosjek, B.; Lau, S.;
Gauvreau, D.; Howard, M. E.; Wallace, D. J.; O’Shea, P. D. J. Org.
Chem. 2012, 77, 2299−2309.
(33) Babinski, D. J.; Aguilar, H. R.; Still, R.; Frantz, D. E. J. Org.
Chem. 2011, 76, 5915−5923.
(34) Eidamshaus, C.; Hommes, P.; Reissig, H. U. Synlett 2012, 23,
1670−1674.
(3) Baldwin, J. E.; Adlington, R. M.; Godfrey, C. R.; Gollins, D. W.;
Vaughan, J. G. J. Chem. Soc., Chem. Commun. 1993, 1434−1435.
(4) (a) Mangion, I. K.; Nwamba, I. K.; Shevlin, M.; Huffman, M. A.
Org. Lett. 2009, 11, 3566−3569. For earlier methods using silver or
copper catalysts, see: (b) Dost, F.; Gosselck, J. Tetrahedron Lett. 1970,
11, 5091−5093.
(35) Ye, F.; Qu, S.; Zhou, L.; Peng, C.; Wang, C.; Cheng, J.;
Hossain, M. L.; Liu, Y.; Zhang, Y.; Wang, Z.-X.; Wang, J. J. Am. Chem.
Soc. 2015, 137, 4435−4444.
(36) Fu, L.; Mighion, J. D.; Voight, E. A.; Davies, H. M. Chem. - Eur.
J. 2017, 23, 3272−3275.
(37) For reviews, see: (a) Burtoloso, A. C. B.; Dias, R. M. P.;
Leonarczyk, I. A. Eur. J. Org. Chem. 2013, 2013, 5005−5016.
(b) Neuhaus, J. D.; Oost, R.; Merad, J.; Maulide, N. Top. Curr. Chem.
2018, 376, 1−47. (c) Wu, X.; Sun, S.; Yu, J.-T.; Cheng, J. Synlett
2019, 30, 21. (e) Oost, R.; Neuhaus, J. D.; Merad, J.; Maulide, N.
Sulfur Ylides in Organic Synthesis and Transition Metal Catalysis. In
Structure and BondingModern Ylide Chemistry; Gessner, V. H., Ed.;
Springer International Publishing AG: Cham, 2018; Vol. 177, pp 73−
115.
(5) Mangion, I. K.; Weisel, M. Tetrahedron Lett. 2010, 51, 5490−
5492.
(6) The formation of C−S bond from sulfoxonium ylides is also
̈
possible in the absence of metal catalysts: (a) Konig, H.; Metzger, H.;
Seelert, K. Chem. Ber. 1965, 98, 3712−3723. (b) Dias, R. M.;
Burtoloso, A. C. B. Org. Lett. 2016, 18, 3034−3037.
(7) Vaitla, J.; Bayer, A.; Hopmann, K. H. Angew. Chem., Int. Ed.
2018, 57, 16180−16184.
(8) Barday, M.; Janot, C.; Halcovitch, N. R.; Muir, J.; Aïssa, C.
Angew. Chem., Int. Ed. 2017, 56, 13117−13121.
(9) Xu, Y.; Zhou, X.; Zheng, G.; Li, X. Org. Lett. 2017, 19, 5256−
5259.
(38) Ray, S. K.; Shaw, R. A.; Smith, B. C. Nature 1962, 196, 372.
(39) Gololobov, Y. G.; Nesmeyanov, A. N.; Lysenko, V. P.;
Boldeskul, I. E. Tetrahedron 1987, 43, 2609−2651.
(40) Christensen, S. B. IV.; Daines, R. A.; Lee, J.; Xiang, J.-N. Fatty
acid synthase inhibitors. Patent WO 2002009561 A2.
(41) For recent reviews, see: (a) Xia, Y.; Qiu, D.; Wang, J. Chem.
Rev. 2017, 117, 13810−13889. (b) Wang, K.; Wang, J. Synlett 2018.
DOI: 10.1055/s-0037-1611020
(42) Attempts of cross-coupling of α-ketone or α-amide sulfoxonium
ylides have so far failed under the condition described herein.
Similarly, the attempted cross-coupling of n-BuBr with 3 failed to
deliver any cross-coupled product.
(10) Xu, Y.; Zheng, G.; Yang, X.; Li, X. Chem. Commun. 2018, 54,
670−673.
(11) Zheng, G.; Tian, M.; Xu, Y.; Chen, X.; Li, X. Org. Chem. Front.
2018, 5, 998−1002.
(12) Oh, H.; Han, S.; Pandey, A. K.; Han, S. H.; Mishra, N. K.; Kim,
S.; Chun, R.; Kim, H. S.; Park, J.; Kim, I. S. J. Org. Chem. 2018, 83,
4070−4077.
(13) Zhou, C.; Fang, F.; Cheng, Y.; Li, Y.; Liu, H.; Zhou, Y. Adv.
Synth. Catal. 2018, 360, 2546−2551.
(14) Hu, P.; Zhang, Y.; Xu, Y.; Yang, S.; Liu, B.; Li, X. Org. Lett.
2018, 20, 2160−2163.
(15) Halskov, K. S.; Witten, M. R.; Hoang, G. L.; Mercado, B. Q.;
Ellman, J. A. Org. Lett. 2018, 20, 2464−2467.
(16) Hoang, G. L.; Ellman, J. A. Tetrahedron 2018, 74, 3318−3324.
(17) Zhu, J.; Sun, S.; Cheng, J. Tetrahedron Lett. 2018, 59, 2284−
2287.
(18) Yang, R.; Wu, X.; Sun, S.; Yu, J. T.; Cheng, J. Synthesis 2018,
50, 3487−3492.
D
DOI: 10.1021/acs.orglett.8b03744
Org. Lett. XXXX, XXX, XXX−XXX