Ruthenium-catalyzed room-temperature coupling of α-keto sulfoxonium ylides and cyclopropanols for δ-diketone synthesis

Article abstract of DOI:10.1039/d1cc02576g

Previous transition metal-catalyzed synthesis processes of δ-diketones are plagued by the high cost of the rhodium catalyst and harsh reaction conditions. Herein a low-cost, room temperature ruthenium catalytic method is developed based on the coupling of α-keto sulfoxonium ylides with cyclopropanols. The mild protocol features a broad substrate scope (47 examples) and a high product yield (up to 99%). Mechanistic studies argue against a radical pathway and support a cyclopropanol ring opening, sulfoxonium ylide-derived carbenoid formation, migratory insertion C-C bond formation pathway.

Full text of DOI:10.1039/d1cc02576g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Ruthenium-catalyzed room-temperature coupling
of a-keto sulfoxonium ylides and cyclopropanols
for d-diketone synthesis†
Cite this: Chem. Commun., 2021,
57, 7386
Received 17th May 2021,
Accepted 28th June 2021
Lili Fang, Shuaixin Fan, Weiping Wu, Tielei Li and Jin Zhu
*
DOI: 10.1039/d1cc02576g
rsc.li/chemcomm
Previous transition metal-catalyzed synthesis processes of d-diketones b-keto xanthates to vinyl carbinols, and multicomponent radi-
are plagued by the high cost of the rhodium catalyst and harsh reaction cal carbonylation showcase the diverse reaction manifolds that
conditions. Herein a low-cost, room temperature ruthenium catalytic can be achieved. However, the involvement of either highly
method is developed based on the coupling of a-keto sulfoxonium oxidative or high pressure conditions inevitably was sought as
ylides with cyclopropanols. The mild protocol features a broad sub- an alternative reaction modality.
strate scope (47 examples) and a high product yield (up to 99%).
Transition metal catalysis, known for its capability of mediating
Mechanistic studies argue against a radical pathway and support a a rich set of transformations, has recently made inroads into this
cyclopropanol ring opening, sulfoxonium ylide-derived carbenoid for- important field. Rhodium catalysis has been successfully exploited
mation, migratory insertion C–C bond formation pathway.
to promote two mechanistically distinct cascade processes: ring
opening of vinyl cyclobutanol/transfer hydrogenation/hydroacyla-
Long-chain, acyclic d-diketones are valuable building blocks tion and hydroacylation of homopropargyl alcohol/deconjugative
enabling the synthesis of divergent classes of natural products, isomerization (Scheme 1a). Albeit effective, the high cost of the
pharmaceutical compounds, and functional materials.1–3 catalyst, harsh reaction conditions (140 1C and 110 1C), and
Therefore, tremendous efforts have been dedicated to the potential competition of the decarbonylative pathway are the major
synthesis of these important structural motifs.4–11 The most hurdles that need to be addressed.4,8 To this end, herein, we wish
adopted conventional methods in this area involve the Michael to disclose a ruthenium-catalyzed room-temperature (RT) coupling
addition of metal enolates to a,b-unsaturated ketones.9–11 of a-keto sulfoxonium ylides and cyclopropanols for the synthesis
Despite the advances in stannyl-, silyl-, and titanium-based of d-diketones (Scheme 1b). During the preparation of the manu-
systems, the instability of metal enolates, narrow substrate script, we noticed a patent describing a ruthenium catalytic proto-
scope, and toxicity of metals manifest significant drawbacks col working at 80 1C.13
associated with these approaches. Indeed, large excesses of
metal enolates were typically engaged to drive the reaction
to completion, and post-treatment hydrolysis could produce
substantial amounts of metal wastes; the reactivity of metal
enolates can cause functional group compatibility issues, and
organotin reagents can cause severe damage to human organs
such as the liver and kidneys. Although enol acetates can be
employed as metal-free surrogates, an extra organotin-catalyzed
transesterification step is required for access to the authentic
d-diketone target. In view of these limitations, radical coupling
has been sought as an alternative reaction modality.12 Radical
ring-opening acylation of cyclobutanols, radical addition of
Department of Polymer Science and Engineering, School of Chemistry and Chemical
Engineering, State Key Laboratory of Coordination Chemistry, Nanjing University,
Nanjing 210023, China. E-mail: jinz@nju.edu.cn
† Electronic supplementary information (ESI) available. CCDC 2087645. For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/
Scheme 1 Synthetic strategy for the construction of d-diketones.
d1cc02576g
7386
|
Chem. Commun., 2021, 57, 7386–7389
This journal is © The Royal Society of Chemistry 2021

View Article Online
Communication
ChemComm
Table 1 Substrate scope of sulfoxonium ylidesab
a-Keto sulfoxonium ylides are easy-to-synthesize, bench-stable
reagents that have demonstrated their versatile utility, especially as
precursors to carbenoids, in transition metal catalysis.14–17 Thus
far, the majority of catalytic reactions is based on rhodium and
iridium systems. Cyclopropanols are also readily available com-
pounds that can undergo facile strain-release b-carbon elimination/
ring opening as a means for further elaboration.18 The induction of
this reactivity mode can be effected by a broad class of reagents,
including transition metals, such as palladium, copper, rhodium,
and cobalt.19–23 We have a long-standing interest in transition
metal catalysis for novel C–H functionalization and have therefore
attempted to use the directing groups developed by our group for
C–H coupling with either a-keto sulfoxonium ylides or cyclopropa-
nols (see ESI† for details). For a-keto sulfoxonium ylide 1a, no
reactions were observed for coupling with 1,2,4-oxadiazol-5(4H)-
one, N-chloroamide, and enaminone.24–26 For cyclopropanol
2a,27,28 no coupling proceeded under enaminone; ring opening
occurred for 2a under N-nitroso, with the formation of a ketone
product; and direct coupling of the two rings was observed, without
the participation of the C–H bond, under 1,2,4-oxadiazol-5(4H)-one.
Inspired by the emerging use of a-keto sulfoxonium ylide as
the directing group,17,29–32 we have shifted our focus on the
possibility of C–H coupling with cyclopropanol. Under all
experimental conditions tested herein, no C–H coupling
occurred; instead, d-diketone was the sole product uncovered.
Thus, for the reaction between 1a (0.2 mmol, 1 equiv.) and 2a
(1.25 equiv.), under rhodium catalysis, 3aa was afforded, albeit
with the maximum attainable yield of merely 20% at 80 1C. In
an attempt to seek a low-cost catalytic system and simulta-
a
All reactions were performed on 0.2 mmol scale (sulfoxonium ylide),
1.25 equiv. of the cyclopropanol with [Ru(p-cym)Cl2]2 (3 mmol%) in the
b
HFIP(2 mL) under nitrogen. Isolated yield.
With the optimized conditions established, the substrate
neously improve the yield, we switched the metal catalyst from scope of a-keto sulfoxonium ylides was evaluated (Table 1).
rhodium to palladium, copper, and ruthenium. For either a-Benzoyl sulfoxonium ylides bearing both electron-donating
palladium or copper, no 3aa could be obtained. To our delight, and electron-withdrawing groups on the phenyl ring can
extensive exploration of ruthenium complexes provided a high- successfully participate in the reactions with 2a. For para-
yield catalytic system. Under [Ru(Z6-p-cym)Cl2]2 (abbreviated as substituted sulfoxonium ylides, the Me variant (1b) gives the
[Ru(p-cym)Cl2]2 hereafter; 3 mol%), although no conversion highest yield (94%); the OMe (1c) substitution lowers the yield
could be discerned after 16 h rt reaction in THF, 1,4-dioxane, to 75%; the F (1d) and Cl (1e) substitutions offer similar yields
CH3CN, and DMSO, trace amount of 3aa was formed in (85%, 87%); and the yield is decreased to 60% and 29%,
CH3OH. Alcohols have been often used as solvents for ring- respectively, for CF3 (1f) and NO2 (1g) variants. For ortho-
opening of cyclopropanols.23,33 In particular, hexafluoroisopro- substituted sulfoxonium ylides, sterics dictate the product
panol (HFIP) has been extensively used in organic synthesis yield, with the yield of Me variant (1h, 65%) lower than that
because of its appealing properties (e.g., strong hydrogen of the F variant (1i, 84%). The electronic and steric effects are
bond-donating capability, low nucleophilic reactivity, and high negligible in meta-substituted sulfoxonium ylides (Me, 1j, 89%;
ionizing capacity).34 For example, the carbon–carbon bond- and Cl, 1k, 83%). The replacement of the phenyl ring with other
forging reactions have proven to proceed well in HFIP. Satis- aromatic rings provides equally viable substrates. The 2-furyl
factorily, the use of HFIP enabled the dramatic improvement in (1l) and 2-thienyl (1m) variants furnish the respective products
yield. 3aa could be obtained in 75% yield under open-air in excellent yields (89% and 96%). By contrast, the yields for the
condition and in 95% yield under nitrogen protection (referred 2-benzothienyl (1n), 1,10-biphenylyl-4-yl (1o), and (E)-2-phenylethenyl
to as the optimized conditions hereafter). Elevation of the (1p) variants are slightly lower (61%, 43%, and 67%). Significantly,
temperature is detrimental to the reaction, with the yield being a-alkylketo sulfoxonium ylides are also compatible coupling part-
reduced to 85%, 71%, and 70% respectively at 40 1C, 60 1C, ners. The linear n-pentyl variant (1q) delivers a higher yield (86%)
and 80 1C. Replacement of [Ru(p-cym)Cl2]2 with either RuCl3, than the branched isoamyl variant (1r, 48%). The cyclopropyl (1s,
Ru(acac)3, or Ru3(CO)12 led to no product formation. The 87%), cyclohexyl (1t, 90%), and benzyl (1u, 78%) variants, despite
effectiveness of [Ru(p-cym)Cl2]2 as a catalyst is speculated their relatively unfavored sterics, can react effectively. The bulkier
to be caused by the lability of the p-cym ligand, which 1-phenylpropyl variant (1v) gives a diminished yield (67%).
enables the facile generation of vacant coordination sites
for catalysis.
With the reactivity of a-keto sulfoxonium ylides thoroughly
examined, the substrate scope of cyclopropanols was next
This journal is © The Royal Society of Chemistry 2021
Chem. Commun., 2021, 57, 7386–7389 | 7387

View Article Online
ChemComm
Communication
assessed (Table 2). First, the substituent on the phenyl ring of effect is apparently reversed in the para substitution case (OMe,
1-benzylcyclopropanol is systematically varied to probe its 2u, 72%; and Cl, 2v, 98%). With a bulkier 1-naphthyl group
impact on the reaction with 1a. For para substitution, the (2w), a lower yield (81%) is acquired. The yield returns to 95%
electron-donating group exerts a positive effect on the product with a 1,10-biphenylyl-4-yl group (2s). The transformation is
yield, essentially enabling a quantitative conversion (Me, 2b, surprisingly highly efficient for sterically congested, even spiro-
98%; tert-butyl, 2c, 98%; and OMe, 2d, 99%). The effect of type 1,2-disubstituted cyclopropanols (1-phenyl, 2-Me, 2y, 89%;
electron-withdrawing group is negative, but still good yield can and 1-phenyl, 2-spiro[5], 2z, 98%). With the substrate scope
be obtained (F, 2e, 73%; Cl, 2f, 75%; and Br, 2g, 88%). The ortho completely surveyed, the utility of the protocol developed
substitution displays a similar reactivity pattern, with the herein is demonstrated with the scale-up of the reaction to
exception of the Me variant (2h), which affords a slightly lower 1 mmol level (84% yield).
yield (91%); the performance of OMe (2i, 99%), F (2j, 75%), Cl
The mild conditions and broad substrate scope of the
(2k, 76%), and Br (2l, 82%) variants is virtually identical to that reaction prompted us to investigate its mechanistic basis. The
of their corresponding para analogs. This suggests that the ring stoichiometric reaction between [Ru(p-cym)Cl2]2 and 1a did not
opening and coupling processes are not affected by the sterics. generate any new species, likely suggesting the absence of
In comparison, the meta substitution provides a less reactive direct inner-sphere coordination (Scheme 2, eqn (1)). By
substrate with the OMe group (2m, 89%), but an equally contrast, a ring-opened ketone product was acquired in high
competent substrate with the F (2n, 78%) and Br (2o, 83%) yield (72%) in a reaction between [Ru(p-cym)Cl2]2 and 2a,
groups. The reaction also proceeds in good yield with the supporting the initial occurrence of cyclopropanol coordination
1
replacement of the benzyl group by an alkyl group (n-pentyl, and b-carbon elimination (Scheme 2, eqn (2)). H NMR mon-
2p, 71%; n-hexyl, 2q, 82%; and n-octyl, 2r, 84%). The alteration itoring of the reaction mixture between 1a and 2a under
to a diphenylmethyl group (2x) reduces the yield to 65%. With a ruthenium catalysis revealed the formation of DMSO, a side
phenyl group (2t), the yield reaches 90%, and the electronic product that is consistent with the ruthenium carbenoid path-
way. Ruthenium has been previously shown to be capable of
undergoing redox pathway to promote radical reactions.35,36 To
Table 2 Substrate scope of cyclopropanolsab
explore such a possibility, two radical-related reactions have
been performed. A reaction carried out between 1a and 2a in
the presence of a radical scavenger, butylated hydroxytoluene
(BHT), gives an essentially identical yield (94%, versus 95% in
the absence of BHT) (Scheme 2, eqn (3)). A radical clock
experiment of 1a and 2aa witnesses no ring-opening of the
2-cyclopropyl substituent, and the expected product 3aaa is
furnished in 83% yield (Scheme 2, eqn (4)). These mechanistic
studies argue against a radical reaction course. Based on these
lines of experimental evidence and previous demonstration of
ruthenium(II) complexes as effective catalysts,37,38 the following
mechanistic pathway is proposed (Scheme 3): cyclopropanol
a
All reactions were performed on 0.2 mmol scale (sulfoxonium ylide),
1.25 equiv. of the cyclopropanol with [Ru(p-cym)Cl2]2 (3 mmol%) in the
HFIP(2 mL) under nitrogen. Isolated yield.
b
Scheme 2 Mechanistic studies.
7388 | Chem. Commun., 2021, 57, 7386–7389
This journal is © The Royal Society of Chemistry 2021

View Article Online
Communication
ChemComm
5 X. Zhang, B. Tutkowski, A. Oliver, P. Helquist and O. Wiest,
ACS Catal., 2018, 8, 1740–1747.
6 M. Grigalunas, T. Ankner, P. O. Norrby, O. Wiest and P. Helquist,
J. Am. Chem. Soc., 2015, 137, 7019–7022.
7 C. Che, Z. Qian, M. Wu, Y. Zhao and G. Zhu, J. Org. Chem., 2018, 83,
5665–5673.
8 R. Guo and G.-Z. Zhang, J. Am. Chem. Soc., 2017, 139, 12891–12894.
9 M. Fabregas, A. Gomez-Palomino, M. Pellicena, D. F. Reina,
P. Romea, F. Urpi and M. Font-Bardia, Org. Lett., 2014, 16,
6220–6223.
10 Y. Onishi, Y. Yoneda, Y. Nishimoto, M. Yasuda and A. Baba,
Org. Lett., 2012, 14, 5788–5791.
11 M. Yasuda, K. Chiba, N. Ohigashi, Y. Katoh and A. Baba, J. Am.
Chem. Soc., 2003, 125, 7291–7300.
12 Z. P. Yin, T. Meyer and X.-F. Wu, Tetrahedron Lett., 2019, 60,
864–867.
13 China Pat., CN 112028762 A, 2020. We have independently initiated
the studies and for the detailed timeline, please see ESI.
14 J. D. Neuhaus, A. Bauer, A. Pinto and N. Maulide, Angew. Chem., Int.
Ed., 2018, 57, 16215–16218.
15 A. Bayer and J. Vaitla, Synthesis, 2018, 612–628.
16 K.-C. Yan, M.-Y. He, J.-L. Li, H. He, R.-Z. Lai, Y. Luo, L. Guo and
Y. Wu, Chem. Commun., 2020, 56, 14287–14290.
17 S. Zhu, K. Shi, H. Zhu, Z.-K. Jia, X.-F. Xia, D. Wang and L.-H. Zou,
Org. Lett., 2020, 22, 1504–1509.
18 T. R. McDonald, L. R. Mills, M. S. West and S. A. L. Rousseaux,
Chem. Rev., 2021, 121, 3–79.
Scheme 3 Proposed reaction mechanism.
19 H. Zhang, G. Wu, H. Yi, T. Sun, B. Wang, Y. Zhang, G. Dong and
J. Wang, Angew. Chem., Int. Ed., 2017, 56, 3945–3950.
20 B. Q. Cheng, S. X. Zhang, Y. Y. Cui, X. Q. Chu, W. Rao, H. Xu,
G. Z. Han and Z. L. Shen, Org. Lett., 2020, 22, 5456–5461.
21 J.-C. Li, Y. Zheng, M.-X. Huang and W.-F. Li, Org. Lett., 2020, 22,
5020–5024.
22 D. Rosa and A. Orellana, Chem. Commun., 2013, 49, 5420–5422.
23 X.-K. Zhou, S.-J. Yu, L.-H. Kong and X.-W. Li, ACS Catal., 2016, 6,
647–651.
coordination to ruthenium(II) with the simultaneous release of
proton, b-carbon elimination generating a b-keto alkyl-metal inter-
mediate, formation of a-keto carbenoid species, 1,1-migratory
insertion/C–C bond formation, and proto-demetalation release of
product and regeneration of ruthenium catalyst.
24 X.-L. Yu, K.-H. Chen, F. Yang, S.-K. Zha and J. Zhu, Org. Lett., 2016,
18, 5412–5415.
25 X.-L. Yu, K.-H. Chen, Q. Wang, W.-J. Zhang and J. Zhu, Org. Chem.
Front., 2018, 5, 994–997.
26 S.-G. Zhou, J.-H. Wang, L.-L. Wang, C. Song, K.-H. Chen and J. Zhu,
Angew. Chem., Int. Ed., 2016, 55, 9384–9388.
27 B.-Q. Liu, C. Song, C. Sun, S.-G. Zhou and J. Zhu, J. Am. Chem. Soc.,
2013, 135, 16625–16631.
28 X.-L. Yu, K.-H. Chen, Q. Wang, S. Guo, S.-K. Zha and J. Zhu,
Angew. Chem., Int. Ed., 2017, 56, 5222–5226.
29 P. Zhou, W.-T. Yang, A. U. Rahman, G.-G. Li and B. Jiang, J. Org.
Chem., 2020, 85, 360–366.
In summary, a ruthenium catalytic method has been developed
for the construction of d-diketones via the coupling of readily
available a-keto sulfoxonium ylides and cyclopropanols. The pro-
tocol features mild reaction conditions, a broad substrate scope,
and a high product yield. The ruthenium system reported herein
expands the repertoire of sulfoxonium ylide-derived carbenoid
chemistry and suggests the promising prospect of this under-
explored, low-cost metal as a major player in this synthetically
important field.
30 B. Shu, X.-T. Wang, Z.-X. Shen, T. Che, M. Zhong, J.-L. Song,
H.-J. Kang, H. Xie, L. Zhang and S.-S. Zhang, Org. Chem. Front.,
2020, 7, 1802–1808.
J. Z. gratefully acknowledges the support from the National
Natural Science Foundation of China (21774056, 52073141).
31 X.-F. Cui and G.-S. Huang, Org. Biomol. Chem., 2020, 18, 4014–4018.
32 G.-F. Zheng, M.-M. Tian, Y.-W. Xu, X.-H. Chen and X.-W. Li,
Org. Chem. Front., 2018, 5, 998–1002.
33 B. V. Pati, A. Ghosh and P. C. Ravikumar, Org. Lett., 2020, 22,
2854–2860.
Conflicts of interest
There are no conflicts to declare.
¨
34 I. A. Shuklov, N. V. Dubrovina and A. Borner, Synthesis, 2007,
2925–2943.
Notes and references
35 P. Gandeepan, J. Koeller, K. Korvorapun, J. Mohr and L. Ackermann,
Angew. Chem., Int. Ed., 2019, 58, 9820–9825.
36 Y.-H. Cheng, Y.-H. He, J. Zheng, H. Yang, J. Liu, G.-H. An and
G.-M. Li, Chin. Chem. Lett., 2021, 32, 1437–1441.
37 B. M. Trost, J. Xie and N. Maulide, J. Am. Chem. Soc., 2008, 130,
17258–17259.
1 N. L. Wendler, R. P. Graber, C. S. Snoddy, Jr. and F. W. Bollinger,
J. Am. Chem. Soc., 1957, 79, 4476–4487.
2 T. M. Penning, J. Pharm. Sci., 1985, 74, 651–654.
3 C. R. Rausch, E. J. Jabbour, H. M. Kantarjian and T. M. Kadia,
Cancer, 2020, 126, 1152–1160.
38 E. Gyanchander, S. Ydhyam, N. Tumma, K. Belmore and J. K. Cha,
Org. Lett., 2016, 18, 6098–6101.
4 G. You, Z.-X. Chang, J. Yan, C. Xia, F.-R. Li and H.-S. Li, Org. Chem.
Front., 2021, 8, 39–45.
This journal is © The Royal Society of Chemistry 2021
Chem. Commun., 2021, 57, 7386–7389 | 7389