Copper-catalyzed direct C–H fluoroalkenylation of heteroarenes

Abstract

Copper-catalyzed direct C–H fluoroalkenylation of heterocycles using various gem-bromofluoroalkenes as electrophiles is reported. This efficient method offers step-economical, low-cost and stereocontrolled access to relevant heteroarylated monofluoroalkenes. The synthesis of fluorinated analogues of biomolecules and therapeutic agents for the treatment of Duchenne muscular dystrophy as application is reported.

---

Introduction

The importance of fluorinated compounds in agrochemicals,^1,2 pharmaceuticals/medicinals,^1,3 and materials science^1,4 has triggered an explosion of research efforts in developing new and efficient methods to introduce a fluorinated functional group into organic molecules. Of particular relevance is the emergence of fluoroalkenes,⁵ versatile compounds that have found many applications as, for example, peptidomimetics,⁶ drugs⁷ and materials.⁸ In peptide synthesis and fine organic chemistry, fluoroalkenes are widely looked upon as stable isosteric and isoelectronic mimics of the amide bond,⁶ and bioisosteres in the structure/activity relationship studies.

Within the readily available fluorinated building blocks for the construction of fluoroolefins, gem-bromofluoroalkenes are easily accessible⁹ and versatile reagents for the achievement of highly useful cross-coupling reactions.¹⁰ The development of catalytic direct C–H bond functionalization methodologies using transition metals as catalysts has received considerable attention avoiding thus the preparation of organometallic intermediates as coupling partners.¹¹ For a realistic catalyst loading of these precious metals, less expensive transition elements such as copper have received significant attention. Indeed, since the breakthroughs made by Daugulis,¹² Miura¹³ and Piguel,¹⁴ remarkable advances have been made in copper-catalyzed direct arylation,¹⁵ alkynylation¹⁶ and alkenylation^17,18 of azoles from monohalogenoalkenes (Fig. 1, eqn (1) and (2)). However, no example of copper-catalyzed direct C–H halogenoalkenylation so far has been reported from gem-dihalogenoalkenes. Indeed, the latter have been used only for copper-catalyzed C–H alkynylation of heterocycles, the non-coupled second halogen being eliminated during the catalytic process (Fig. 1, eqn (3)).¹⁹ Recently, Cao has described an elegant metal-free base mediated nucleophilic vinylic substitution reaction between tetrasubstituted gem-difluoroalkenes and azoles (Fig. 1, eqn (4)).^20a In this case, only tetrasubstituted gem-difluoroalkenes can be used as substrates, and the trisubstituted ones would directly lead to the dehydrofluorination process. This year, Loh have successfully engaged gem-difluoroalkenes in an innovative Rh(III)-catalyzed ortho-directed C–H activation, migratory insertion into the double bond and a defluorination sequence to produce fluoroalkenylated (hetero)aromatics.^20b Our group has previously reported the first Pd-catalyzed direct C–H halogenoalkenylation of heterocycles demonstrating that gem-bromofluoroalkenes are suitable building blocks for this transformation (Fig. 1, eqn (5)).²¹ During this study, a combination of copper salt with a palladium catalyst led to a cooperative Pd(0)/Cu(I) catalysis which was found highly performant to achieve the direct C–H alkenylation of a broad range of 1,3-diazoles with any gem-bromofluoroalkenes as electrophiles. This efficiency was based upon the catalytic generation of an heteroaryl copper intermediate reacting as a transmetallating agent.²² In our ongoing project devoted to the production of fluorinated biomolecule analogues, we have recently turned our attention to the evaluation of the reactivity of the heteroaryl copper intermediate towards gem-bromofluoroalkenes in view of developing a novel palladium free copper-catalyzed direct C–H fluoroalkenylation of heterocycles offering step-economical, low-cost and stereocontrolled access to heteroarylated monofluoroalkenes (Fig. 1, eqn (6)). Moreover, taking into account that the direct fluoroalkenylation is very scarce in the literature, it could represent an efficient alternative to produce fluorinated fine chemicals. We reported herein this methodology giving access to a wide variety of trisubstituted monofluoroalkene^5,23 derivatives including fluorinated analogues of therapeutic agents.

Fig. 1 Direct functionalization of ubiquitous C–H bonds using halogenoaryls/alkenes and gem-dihalogenoalkenes.

Results and discussion

We initiated our investigation by probing various reaction conditions for the direct C–H fluoroalkenylation of 5-phenyl-oxazole (2a) with easily accessible (E)-gem-bromofluoroalkene⁹1A (Table 1). Indeed, whereas a set of experiments with our model substrate, phenyloxadiazole, under an optimized previously reported procedure was found unsuccessful without the palladium catalyst,²⁴2a proved to be a suitable substrate for direct fluoroalkenylation without a palladium source, providing 3Aa in 63% yield compared to 75% yield under bimetallic Pd/Cu catalysis (entries 1 and 2). Switching the nature of the copper source for CuI led to a slight enhancement of the yield (entry 3). Among the bases, t-BuOLi proved to be the most effective without formation of the alkynylated side-product via a dehydrofluorination process (entries 4 and 5). Subsequently, common ligands of copper have been screened, such as Phen and derivatives, diamine ligands or mono- and bidendate phosphines (entries 4–10), and surprisingly, dppe revealed considerable efficiency affording the desired product in almost quantitative yield (entry 10). Crucially, formation of 3Aa was not observed when copper was omitted from the reaction mixture (entry 11). We then performed the reaction with different copper sources (entries 12–14). The best performance of the reaction was thus attained by using CuI as the catalyst; however, we observed that the monofluoroalkenylation was slightly insensitive to the copper source (Cu(I) or Cu(II)).

Table 1 Optimization of the fluoroalkenylation reaction^a

Entry	[Cu]	Ligand	Base	Yield^b (%)
a All reactions were performed using 1A (1.1 equiv.), 2a (0.2 mmol, 1.0 equiv.), copper source (10 mol%), ligand (20 mol%), and base (3 equiv.) in 1,4-dioxane (0.25 M) at 110 °C. b Yield based on the isolated product after flash chromatography. c In the presence of PdCl2(PPh3)2 (5 mol%). d L1 = 3,4,7,8-(Me)4-1,10-Phen. e L2 = trans-N,N′-dimethylcyclohexane-1,2-diamine.
1^c	CuBr	—	t-BuOLi	75
2	CuBr	—	t-BuOLi	63
3	CuI	—	t-BuOLi	65
4	CuI	Phen	t-BuOLi	51
5	CuI	Phen	K2CO3	—
6	CuI	L1 ^d	t-BuOLi	51
7	CuI	L2 ^e	t-BuOLi	81
8	CuI	PPh3	t-BuOLi	83
9	CuI	PCy3·HBF4	t-BuOLi	51
10	CuI	dppe	t-BuOLi	96
11	—	dppe	t-BuOLi	—
12	CuBr	dppe	t-BuOLi	66
13	CuCl2	dppe	t-BuOLi	73
14	Cu(OTf)2	dppe	t-BuOLi	65

---

Under these optimized reaction conditions, the heteroarylated fluoroalkene 3Aa was produced in 96% isolated yield as a pure (Z)-isomer, demonstrating that the reaction proceeds with complete retention of the stereochemistry.

With the optimized conditions in hand, the establishment of the scope of the direct C–H fluoroalkenylation was undertaken on 5-phenyloxazole (2a) with various readily accessible (E)-gem-bromofluoroalkenes⁹ (Table 2). The latter flanked indifferently with electron-donating or electron-withdrawing groups on the aromatic unit at the ortho, meta and para positions, reacted at the C-2 position of the 5-phenyloxazole producing the desired product in moderate to excellent yields.

Table 2 Copper-catalyzed C–H fluoroalkenylation of 5-phenyloxazole with various gem-bromofluoroalkenes^a

a All reactions were performed using 1 (1.1 equiv.), 2a (0.2 mmol, 1.0 equiv.), CuI (10 mol%), dppe (20 mol%), t-BuOLi (3 equiv.) in 1,4-dioxane (0.25 M) at 110 °C. Yields are based on the isolated product after flash chromatography. b Yield obtained under bimetallic Pd/Cu catalysis.

---

Interestingly, (E)-gem-bromofluoroalkenes bearing the cyano group (1D) or the chlorine atom (1F) on the aromatic ring, both valuable functional groups for further post-functionalizations, displayed good reactivity, even if, in the case of 3Da, the reaction yield was lower than that obtained under Pd/Cu catalysis.²¹

Alkylated or tetrasubstituted gem-bromofluoroalkenes proved to be unreactive under Cu-catalysis whereas the reaction occurred under Pd/Cu catalysis.^21,25 The aromatic ring bearing an electron-withdrawing group at the ortho position proved to be unstable under these reaction conditions.²⁵

We next examined the substrate scope of various aryloxazoles (2b–e) with a panel of gem-bromofluoroalkenes in this transformation (Table 3). The reaction was efficient whatever the electronic nature of the substituents on the benzene ring of the 5-phenyloxazole 2b–d. It is noteworthy that the use of Phen instead of dppe as a ligand with 4-phenyloxazole 2e as the substrate was crucial to obtain trisubstituted Z-fluoroalkenes 3Ae, 3Ce and 3Ee in good yields.²⁵

Table 3 Copper-catalyzed C–H fluoroalkenylation of various aryloxazoles^a

a All reactions were performed using 1 (1.1 equiv.), 2 (0.2 mmol, 1.0 equiv.), CuI (10 mol%), dppe (20 mol%), and t-BuOLi (3 equiv.) in 1,4-dioxane (0.25 M) at 110 °C. Yields are based on the isolated product after flash chromatography. b Phen (20 mol%) has been used as a ligand instead of dppe.

---

We then applied our reaction conditions to various relevant 1,3-diazoles (Table 4). A first set of experiments engaging benzoxazole 4a with the (E)-gem-bromofluoroalkenes 1A and 1F, as coupling partners was addressed. Through the expected monofluoroalkenes, 4Aa was obtained in good (82%) yield whereas the chlorinated benzoxazolylfluoroalkene 4Fa was produced in poor (20%) yield (¹⁹F NMR yield). Nevertheless, when the temperature was increased from 110 to 130 °C, full completion of benzoxazole 4a was attained to provide product 4Fa in 74% yield. The reaction performed with chlorinated benzoxazole 4b at the same 130 °C temperature delivered the expected benzoxazolylfluoroalkene 4Ab in good (70%) yield and importantly, the reaction could be scaled up from 0.2 mmol to 1 mmol without any loss of efficiency.

Table 4 Extension of the reaction to various heterocycles^a

a All reactions were performed using 1 (1.1 equiv.), heterocycle (0.2 mmol, 1.0 equiv.), CuI (10 mol%), ligand (20 mol%), and t-BuOLi (3 equiv.) in 1,4-dioxane (0.25 M) at 110 °C. Yields are based on the isolated product after flash chromatography. b Yield obtained under bimetallic Pd/Cu catalysis. c Reaction performed at 130 °C. d Reaction performed on a 1 mmol scale. e L₂ = trans-N,N′-dimethylcyclohexane-1,2-diamine. f Reaction performed at 90 °C. g CuI (20 mol%) and Phen (40 mol%) were used.

---

We then examined the selective C–H monofluoroalkenylation with N-methyl-benzimidazole 5 as a heterocycle under our optimized conditions and, unfortunately no desired product was obtained. A careful screening of bases, solvents and ligands at different reaction temperatures²⁵ led to select trans-N,N′-dimethylcyclohexane-1,2-diamine as a ligand and a reaction temperature of 130 °C to produce fluoroalkenes 5A and 5H in optimized 54 and 55% yields, respectively.

Finally, the reaction was investigated in a thiazole series. In this case, the phenanthroline ligand was found highly performant to achieve the cross-coupling at 110 °C of para-, ortho- or disubstituted gem-bromofluoroalkenes 1A, 1G and 1H with benzothiazole 6 and 4,5-dimethythiazole 7 giving the fluoroalkenes in fair 54% to excellent 91% yields. However, the reaction performed with the trifluoromethylated (E)-gem-bromofluoroalkenes 1E provided the desired product 6E in poor 20% isolated yield mainly due to the degradation of the coupling partner 1E. Fortunately, the yield was significantly improved to 41% when operating at lower temperature, 90 °C. It has to be noted that the yields obtained for 4Aa and 6A under these experimental conditions were better than the yields obtained under bimetallic catalysis.²¹

Finally, we applied this copper-catalyzed fluoroalkenylation to the synthesis of relevant biomolecules as depicted in Scheme 1. Taking into account the similarities between the fluoroolefin moiety and the amide bond,⁶ we first synthesized a fluorinated analogue 4Ba, in 53% yield, of the antiasthmatic agent 8,²⁶ starting from benzoxazole 4a and gem-bromofluoroalkene 1B. Then, we decided to apply our methodology as an alternative pathway to produce potential active molecules used in the treatment of Duchenne muscular dystrophy exemplified by the molecule 4Bc ²⁷ which was produced in quantitative yield by reacting 5-methoxybenzoxazole 4c with gem-bromofluoroalkene 1B (Scheme 1). Interestingly, a library of compounds may be readily produced for further SAR study introducing various substituents on both aromatic rings bearing heterocycle 4 or gem-bromofluoroalkene partner 1.

Scheme 1 Synthesis of fluorinated therapeutic agents. ^a Yields based on the isolated product after flash chromatography.

Conclusions

In summary, an efficient Cu(I)/t-BuOLi catalyst has been employed for direct C–H fluoroalkenylation of 1,3-diazoles with readily available gem-bromofluoroalkenes as coupling partners. Although phenyloxadiazole remained unreactive as well as the use of alkylated gem-bromofluoroalkenes as electrophiles were unsuitable under these experimental conditions compared to the bimetallic Pd/Cu catalysis, the palladium free copper catalyzed fluoroalkenylation proved to be very efficient with gem-bromofluorostyrenes. Remarkably, a broad scope of 1,3-diazoles was accomplished modulating the nature of the ligand. Notably, the (benzo)oxazole, (benzo)thiazole and benzimidazole series were successfully coupled with various gem-bromofluoroalkenes using diarylphosphine, phenanthroline and diamine ligands. The methodology gave access to innovative and valuable heteroarylated fluoroalkenes 3–7 produced in fair to excellent yields. It was finally applied to the synthesis of valuable benzoxazolylfluoroalkenes 4Ba, a fluorinated analogue of an antiasthmatic agent, and 4Bc which is potentially active in the treatment of Duchenne muscular dystrophy.

Acknowledgements

This work has been partially supported by INSA Rouen, Rouen University, CNRS EFRD, Labex SynOrg (ANR-11-LABX-0029).

Notes and references

1. (a) T. Hiyama in Organofluorine Compounds: Chemistry and Applications, edH. Yamamoto, Springer-Verlag, Berlin, 2000; (b) Approximately 30–40% of agrochemicals and 20–25% of pharmaceuticals on the market are estimated to contain fluorine: A. M. Thayer, Chem. Eng. News, 2006, 84, 15.
2. P. Jiechke, ChemBioChem, 2004, 5, 570.
3. (a) J.-P. Bégué and D. Bonnet-Delphon, J. Fluorine Chem., 2006, 127, 992; (b) K. L. Kirk, J. Fluorine Chem., 2006, 127, 1013; (c) K. Muller, C. Faeh and F. Diederich, Science, 2007, 317, 1881; (d) W. K. Hagmann, J. Med. Chem., 2008, 51, 4359; (e) S. Purser, P. R. Moore, S. Swallow and V. Gouverneur, Chem. Soc. Rev., 2008, 37, 320; (f) J. Wang, M. Sánchez-Roselló, J. L. Aceña, C. del Pozo, A. E. Sorochinsky, S. Fustero, V. A. Soloshonok and H. Liu, Chem. Rev., 2014, 114, 2432.
4. M. Pagliaro and R. Ciriminna, J. Mater. Chem., 2005, 15, 4981.
5. (a) G. Landelle, N. Bergeron, M.-O. Turcotte-Savard and J.-F. Paquin, Chem. Soc. Rev., 2011, 40, 2867; (b) H. Yanai and T. Taguchi, Eur. J. Org. Chem., 2011, 5939; (c) S. Hara, Top. Curr. Chem., 2012, 327, 59.
6. (a) K. Uneyama, Organofluorine Chemistry, Blackwell, Oxford, 2006; (b) S. Couve-Bonnaire, D. Cahard and X. Pannecoucke, Org. Biomol. Chem., 2007, 5, 1151 and references therein; (c) C. Pierry, S. Couve-Bonnaire, L. Guilhaudis, C. Neveu, A. Marotte, B. Lefranc, D. Cahard, I. Segalas-Milazzo, J. Leprince and X. Pannecoucke, ChemBioChem, 2013, 14, 1620.
7. Monofluoroalkenes as (a) Antitumoral: S. Osada, S. Sano, M. Ueyama, Y. Chuman, H. Kodama and K. Sakaguchi, Bioorg. Med. Chem., 2010, 18, 605; (b) Antimicrobial: Y. Asahina, K. Iwase, F. Iinuma, M. Hosaka and T. Ishizaki, J. Med. Chem., 2005, 48, 3194; (c) Anti-HIV: S. Oishi, H. Kamitani, Y. Kodera, K. Watanabe, K. Kobayashi, T. Narumi, K. Tomita, H. Ohno, T. Naito, R. Kodama, M. Matsuoka and N. Fujii, Org. Biomol. Chem., 2009, 7, 2872; (d) Antidiabetic: S. D. Edmondson, L. Wei, J. Xu, J. Shang, S. Xu, J. Pang, A. Chaudhary, D. C. Dean, H. He, B. Leiting, K. A. Lyons, R. A. Patel, S. B. Patel, G. Scapin, J. K. Wu, M. G. Beconi, N. A. Thornberry and A. E. Weber, Bioorg. Med. Chem. Lett., 2008, 18, 2409.
8. (a) F. Babudri, G. M. Farinola, F. Naso and R. Ragni, Chem. Commun., 2007, 1003; (b) F. Babudri, A. Cardone, G. M. Farinola, C. Martinelli, R. Mendichi, F. Naso and M. Striccoli, Eur. J. Org. Chem., 2008, 1977.
9. All gem-bromofluoroalkenes were prepared following our optimized olefination procedure: (a) X. Lei, G. Dutheuil, X. Pannecoucke and J.-C. Quirion, Org. Lett., 2004, 6, 2101; (b) L. Zoute, G. Dutheuil, J.-C. Quirion, P. Jubault and X. Pannecoucke, Synthesis, 2006, 3409; (c) For other preparation see: D. J. Burton, Z.-Y. Yang and W. Qiu, Chem. Rev., 1996, 96, 1641.
10. (a) F. Legrand, K. Jouvin and G. Evano, Isr. J. Chem., 2010, 50, 588; (b) G. Chelucci, Chem. Rev., 2012, 112, 1344 and references therein.
11. For recent reviews see: (a) D. Alberico, M. E. Scott and M. Lautens, Chem. Rev., 2007, 107, 174; (b) B. Seregin and V. Gevorgyan, Chem. Soc. Rev., 2007, 36, 1173; (c) K. Fagnou, Top. Curr. Chem., 2010, 292, 35; (d) I. A. I. Mkhalid, J. H. Barnard, T. B. Marder, J. M. Murphy and J. F. Hartwig, Chem. Rev., 2010, 110, 890; (e) D. A. Colby, R. G. Bergman and J. A. Ellman, Chem. Rev., 2010, 110, 624; (f) C. L. Sun, B.-J. Li and Z.-J. Shi, Chem. Commun., 2010, 46, 677; (g) M. Wasa, K. M. Engle and J.-Q. Yu, Isr. J. Chem., 2010, 50, 605; (h) C. Liu, H. Zhang, W. Sui and A. Lei, Chem. Rev., 2011, 111, 1780; (i) P. Herrmann and T. Bach, Chem. Soc. Rev., 2011, 40, 2022; (j) L. Ackermann, Chem. Rev., 2011, 111, 1315; (k) C. S. Yeung and V. M. Dong, Chem. Rev., 2011, 111, 1215; (l) W. Shi, C. Liu and A. Lei, Chem. Soc. Rev., 2011, 40, 2761; (m) K. M. Engle, T.-S. Mei, M. Wasa and J.-Q. Yu, Acc. Chem. Res., 2012, 45, 788; (n) J. Yamaguchi, A. D. Yamaguchi and K. Itami, Angew. Chem., Int. Ed., 2012, 51, 8960; (o) T. Besset, T. Poisson and X. Pannecoucke, Chem. – Eur. J., 2014, 20, 16830.
12. (a) H.-Q. Do and O. Daugulis, J. Am. Chem. Soc., 2007, 129, 12404; (b) H.-Q. Do, R. K. M. Khan and O. Daugulis, J. Am. Chem. Soc., 2008, 130, 15185.
13. (a) M. Yamashita, K. Hirano, T. Satoh and M. Miura, Org. Lett., 2009, 11, 2337; (b) T. Yoshizumi, H. Tsurugi, T. Satoh and M. Miura, Tetrahedron Lett., 2008, 49, 1598.
14. F. Besselièvre, S. Piguel, F. Mahuteau-Betzer and D. S. Grierson, Org. Lett., 2008, 10, 4029.
15. (a) A. A. Kulkarni and O. Daugulis, Synthesis, 2009, 4087; (b) O. Daugulis, H.-Q. Do and D. Shabashov, Acc. Chem. Res., 2009, 41, 1074; (c) O. Daugulis, Top. Curr. Chem., 2010, 292, 57; (d) K. Hirano and M. Miura, Synlett, 2011, 294; (e) K. Hirano and M. Miura, Chem. Commun., 2012, 48, 10704; (f) K. Hirano and M. Miura, Top. Catal., 2014, 57, 878.
16. (a) F. Besselièvre and S. Piguel, Angew. Chem., Int. Ed., 2009, 48, 9553; (b) N. Matsuyama, M. Kitahara, K. Hirano, T. Satoh and M. Miura, Org. Lett., 2010, 12, 2358; (c) T. Kawano, N. Matsuyama, K. Hirano, T. Satoh and M. Miura, J. Org. Chem., 2010, 75, 1764.
17. For recent reviews of C–H alkenylation see: (a) S. Messaouidi, J. –D. Brion and M. Alami, Eur. J. Org. Chem., 2010, 6495; (b) R. Rossi, F. Bellina and M. Lessi, Synthesis, 2010, 4131.
18. (a) Q. Liao, L. Zhang, S. Li and C. Xi, Org. Lett., 2011, 13, 228; (b) B. Das, G. C. Reddy, P. Balasubramanyam and N. Salvanna, Tetrahedron, 2012, 68, 300; (c) W. Zhang, Y. Tian, N. Zhao, Y. Wang, J. Li and Z. Wang, Tetrahedron, 2014, 70, 6120.
19. (a) B. P. Berciano, S. Lebrequier, F. Besselièvre and S. Piguel, Org. Lett., 2010, 12, 4038; (b) G. C. Reddy, P. Balasubramanyam, N. Salvanna and B. Das, Eur. J. Org. Chem., 2012, 471; (c) K. Jouvin, A. Coste, A. Bayle, F. Legrand, G. Karthikeyan, K. Tadiparthi and G. Evano, Organometallics, 2012, 31, 7933; (d) L. Ackermann, C. Kornhaass and Z. Yingjun, Org. Lett., 2012, 14, 1824.
20. (a) X. Zhang, Y. Lin, J. Zhang and S. Cao, RSC Adv., 2015, 5, 7905; (b) P. Tian, C. Feng and T.-P. Loh, Nat. Commun., 2015, 6, 7472.
21. C. Schneider, D. Masi, S. Couve-Bonnaire, X. Pannecoucke and C. Hoarau, Angew. Chem., Int. Ed., 2013, 52, 3246.
22. F. Bellina, S. Cauteruccio and R. Rossi, Curr. Org. Chem., 2008, 12, 774.
23. K. Rousée, C. Schneider, S. Couve-Bonnaire, X. Pannecoucke, V. Levacher and C. Hoarau, Chem. – Eur. J., 2014, 20, 15000.
24. Reaction of 1A with phenyloxadiazole in 1,4-dioxane with t-BuOLi as a base gave 83% yield under PdCl₂(PPh₃)₂/CuBr catalysis, 0% under CuBr catalysis and 20% under CuI catalysis.
25. See the ESI† for more details.
26. B. Loev, R. E. Brown, F. –C. Huang and H. Jones, Patent, EP0127066 (A2), USV Pharmaceutical Corporation, 1984.
27. (a) S. P. Wren, G. M. Wynne, F. X. Wilson and S. D. Poignant, Patent, WO2010/112093 (A1), BioMarin IGA, Ltd, 2010; (b) S. P. Wren, G. M. Wynne and F. X. Wilson, Patent, WO2010/112092 (A1), BioMarin IGA, Ltd, 2010.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures, optimization tables, characterization data and NMR spectra of new compounds. See DOI: 10.1039/c5ob02213d

---