Exploring Tandem Ruthenium-Catalyzed Hydrogen Transfer and SNAr Chemistry

A hydrogen-transfer strategy for the catalytic functionalization of benzylic alcohols via electronic arene activation, accessing a diverse range of bespoke diaryl ethers and aryl amines in excellent isolated yields (38 examples, 70% average yield), is reported. Taking advantage of the hydrogen-transfer approach, the oxidation level of the functionalized products can be selected by judicious choice of simple and inexpensive additives. T carbonyl functional group is one of the most prevalent and versatile in chemistry. However, in some cases, carbonyl compounds suffer from poor stability (e.g., air oxidation of aldehydes to carboxylic acids; enolization leading to deleterious side reactions and erosion of enantiopurity) and limited commercial availability. In comparison, alcohols are typically widely available, inexpensive, and relatively inert toward air, moisture, and light, making them highly attractive starting materials for synthesis. Furthermore, the alcohol functional group is ubiquitous in pharmaceuticals, agrochemicals, dyes, fragrances, polymers, functional materials, and catalysts. As such, the development of novel methods that directly functionalize alcohols, diversifying their reactivity profile, is an important pursuit. Hydrogen transfer is a powerful approach that can be employed to access the diverse reactivity of carbonyl compounds from alcohol starting materials. Dehydrogenation of secondary alcohol substrates accesses ketones that can react with both nucleophiles and electrophiles (via enolization) in a variety of important reactions including C−C or C−N bond formation. Dehydrogenation of allylic and benzylic alcohols dramatically alters the properties and reactivity of the olefin and arene, respectively, via electronic activation (Scheme 1, eq 1). In 2013, Williams and co-workers developed a rutheniumcatalyzed transfer hydrogenation/isomerization of aryl allyl alcohols, generating acetophenones that are activated toward nucleophilic aromatic substitution (SNAr) (Scheme 1, eq 2). 6 This redox-neutral approach requires a sacrificial olefin hydrogen acceptor within the substrate, significantly limiting its broader application in organic synthesis. Taking inspiration from these reports, we envisaged developing a more general strategy for catalytic arene functionalization via electronic activation of simple benzylic alcohols through the use of inexpensive additives that serve as oxidants or reductants. This approach would remove the strict requirement for highly Received: November 5, 2017 Published: December 6, 2017 Scheme 1. Concept, Previous Work, And Outline of the Hydrogen Transfer−SNAr Strategy Letter pubs.acs.org/OrgLett Cite This: Org. Lett. 2017, 19, 6716−6719 © 2017 American Chemical Society 6716 DOI: 10.1021/acs.orglett.7b03441 Org. Lett. 2017, 19, 6716−6719 This is an open access article published under a Creative Commons Attribution (CC-BY) License, which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited. D ow nl oa de d vi a 52 .1 1. 21 1. 14 9 on M ar ch 7 , 2 02 0 at 1 9: 24 :0 1 (U T C ). Se e ht tp s: //p ub s. ac s. or g/ sh ar in gg ui de lin es f or o pt io ns o n ho w to le gi tim at el y sh ar e pu bl is he d ar tic le s. specialized aryl allyl alcohol substrates, significantly expanding the potential synthetic applications of the method. Furthermore, taking advantage of the hydrogen transfer approach, it was anticipated that the oxidation level of the functionalized products could be selected as desired by addition of an external oxidant or reductant, generating a diverse array of bespoke ketone and alcohol products, respectively (Scheme 1, eq 3). Herein, we report the successful implementation of this strategy and describe the diverse catalytic functionalization of simple benzylic alcohols via electronic arene activation. To test our hypothesis, we selected 1-(4-fluorophenyl)ethan1-ol 1 as the model substrate and phenol (1.1 equiv) as the nucleophile. Cognizant of the potentially challenging reduction of electron-rich acetophenones via metal-catalyzed hydrogen transfer, the formation of mixtures containing benzylic alcohols 1 and 4 and acetophenones 2 and 3 was anticipated (Table 1). Therefore, the initial target was to achieve full conversion of 1 to a mixture of phenoxy-substituted arenes 3 and 4, with a view to altering the product composition to favor acetophenone 3 or benzylic alcohol 4 via the addition of oxidants or reductants, respectively. After extensive optimization, it was found that Ru(PPh3)3(CO)(H)2 (5 mol %), 1,2bis(diphenylphosphino)ethane (dppe) (5 mol %), and K2CO3 (1.1 equiv) in bench grade DMSO ([1] = 1 M) at 130 °C for 24 h, resulted in full consumption of 1 and formation of 3 and 4 in 62% and 28% NMR yield, respectively (entry 1). The observed net loss of hydrogen is attributed to the challenging hydrogenation of electron-rich acetophenone 3, resulting in H2 expulsion (i.e., acceptorless dehydrogenation) into the headspace of the reaction vessel. No reaction occurs in the absence of ruthenium catalyst (entry 2), with 23% remaining starting material observed in the absence of dppe as ligand (entry 3). Variation of the ligand (entries 4 and 5), base (entries 6 and 7), and solvent (entry 8) all had a deleterious effect on conversion to 3 and 4, as did reducing the concentration (entry 9), reaction temperature (entry 10), and catalyst loading (entry 11). The product distribution could be readily tailored by the addition of simple, inexpensive additives. Acetone (5 equiv) and formic acid (5 equiv), selected due to their low cost and ease of separation, permitted access to acetophenone 3 and benzylic alcohol 4 in 79% (entry 12) and 80% (entry 13) isolated yield, respectively. Using the optimized conditions for the dehydrogenative SNAr process (Table 1, entry 12), various aryl alcohols can be employed as the nucleophile, accessing a range of substituted diaryl ether products in excellent yields (products 3 and 5−12, 70−86% yield) (Scheme 2). Within the aryl alcohol, 4-Me, 3Me, and 2-Me substitution were tolerated in addition to electron-donating substituents (4-OMe). The use of 4chlorophenol and 4-bromophenol as the nucleophile was successful, incorporating an additional functional handle into Table 1. Optimization of Hydrogen Transfer−SNAr Protocol entry variation from “standard” conditions yield (1, 2, 3, 4) (%)

T he carbonyl functional group is one of the most prevalent and versatile in chemistry. 1 However, in some cases, carbonyl compounds suffer from poor stability (e.g., air oxidation of aldehydes to carboxylic acids; enolization leading to deleterious side reactions and erosion of enantiopurity) and limited commercial availability. In comparison, alcohols are typically widely available, inexpensive, and relatively inert toward air, moisture, and light, making them highly attractive starting materials for synthesis. Furthermore, the alcohol functional group is ubiquitous in pharmaceuticals, agrochemicals, dyes, fragrances, polymers, functional materials, and catalysts. As such, the development of novel methods that directly functionalize alcohols, diversifying their reactivity profile, is an important pursuit.
Hydrogen transfer is a powerful approach that can be employed to access the diverse reactivity of carbonyl compounds from alcohol starting materials. 2 Dehydrogenation of secondary alcohol substrates accesses ketones that can react with both nucleophiles and electrophiles (via enolization) in a variety of important reactions including C−C or C−N bond formation. 3,4 Dehydrogenation of allylic and benzylic alcohols dramatically alters the properties and reactivity of the olefin and arene, respectively, via electronic activation (Scheme 1, eq 1). 5 In 2013, Williams and co-workers developed a rutheniumcatalyzed transfer hydrogenation/isomerization of aryl allyl alcohols, generating acetophenones that are activated toward nucleophilic aromatic substitution (S N Ar) (Scheme 1, eq 2). 6 This redox-neutral approach requires a sacrificial olefin hydrogen acceptor within the substrate, significantly limiting its broader application in organic synthesis. Taking inspiration from these reports, we envisaged developing a more general strategy for catalytic arene functionalization via electronic activation of simple benzylic alcohols through the use of inexpensive additives that serve as oxidants or reductants. This approach would remove the strict requirement for highly specialized aryl allyl alcohol substrates, significantly expanding the potential synthetic applications of the method. Furthermore, taking advantage of the hydrogen transfer approach, it was anticipated that the oxidation level of the functionalized products could be selected as desired by addition of an external oxidant or reductant, generating a diverse array of bespoke ketone and alcohol products, respectively (Scheme 1, eq 3). Herein, we report the successful implementation of this strategy and describe the diverse catalytic functionalization of simple benzylic alcohols via electronic arene activation.
To test our hypothesis, we selected 1-(4-fluorophenyl)ethan-1-ol 1 as the model substrate and phenol (1.1 equiv) as the nucleophile. Cognizant of the potentially challenging reduction of electron-rich acetophenones via metal-catalyzed hydrogen transfer, 7 the formation of mixtures containing benzylic alcohols 1 and 4 and acetophenones 2 and 3 was anticipated (Table 1). Therefore, the initial target was to achieve full conversion of 1 to a mixture of phenoxy-substituted arenes 3 and 4, with a view to altering the product composition to favor acetophenone 3 or benzylic alcohol 4 via the addition of oxidants or reductants, respectively. After extensive optimization, 8 it was found that Ru(PPh 3 ) 3 (CO)(H) 2 (5 mol %), 1,2bis(diphenylphosphino)ethane (dppe) (5 mol %), and K 2 CO 3 (1.1 equiv) in bench grade DMSO ([1] = 1 M) at 130°C for 24 h, resulted in full consumption of 1 and formation of 3 and 4 in 62% and 28% NMR yield, respectively (entry 1). The observed net loss of hydrogen is attributed to the challenging hydrogenation of electron-rich acetophenone 3, resulting in H 2 expulsion (i.e., acceptorless dehydrogenation) 4 into the headspace of the reaction vessel. No reaction occurs in the absence of ruthenium catalyst (entry 2), with 23% remaining starting material observed in the absence of dppe as ligand (entry 3). Variation of the ligand (entries 4 and 5), base (entries 6 and 7), and solvent (entry 8) all had a deleterious effect on conversion to 3 and 4, as did reducing the concentration (entry 9), reaction temperature (entry 10), and catalyst loading (entry 11). The product distribution could be readily tailored by the addition of simple, inexpensive additives. Acetone (5 equiv) and formic acid (5 equiv), selected due to their low cost and ease of separation, permitted access to acetophenone 3 and benzylic alcohol 4 in 79% (entry 12) and 80% (entry 13) isolated yield, respectively.
Using the optimized conditions for the dehydrogenative S N Ar process (Table 1, entry 12), various aryl alcohols can be employed as the nucleophile, accessing a range of substituted diaryl ether products in excellent yields (products 3 and 5−12, 70−86% yield) (Scheme 2). 9 Within the aryl alcohol, 4-Me, 3-Me, and 2-Me substitution were tolerated in addition to electron-donating substituents (4-OMe). The use of 4chlorophenol and 4-bromophenol as the nucleophile was successful, incorporating an additional functional handle into the products for subsequent elaboration via cross-coupling methods. 10 2-Naphthol is a competent nucleophile in this process; however, aryl alcohols that are particularly sterically hindered (1-naphthol) or electron-poor (4-nitrophenol) do not readily react, with mostly starting materials returned. 11 A range of 5-, 6-, and 7-membered heterocyclic amines, including pyrrolidine, piperidine, morpholine, and azepane, can be employed as the nucleophile to afford various aryl amine products in high yields (products 15−22, 66−83% yield). 12 Acyclic secondary amines N-methylphenethylamine and diethylamine afforded aryl amines 23 and 24 in 67% and 45% isolated yield, 13 respectively, with less nucleophilic benzylamine generating 25 in only 25% yield. 14 Furthermore, using phenol as nucleophile, a range of fluoroarene substrates undergo functionalization using the optimized reaction conditions, including sterically hindered secondary alcohols (R = i-Pr, Cy) and various trisubstituted arenes (products 26− 34, 52−81% yield). Finally, 1-(2-fluorophenyl)ethan-1-ol was employed in the dehydrogenative S N Ar protocol, accessing diaryl ether derivative 35 in 53% yield. 15 Having successfully demonstrated the dehydrogenative S N Ar process for a variety of nucleophiles and fluoroarenes, we next investigated the scope of the formally redox neutral S N Ar approach, 3 employing formic acid as reducing agent (Scheme 3). Under the optimized reaction conditions (Table 1, entry 13), 1-(4-fluorophenyl)ethan-1-ol 1 couples with a representative selection of substituted aryl alcohol nucleophiles, accessing a range of diaryl ethers in excellent yields (products 4 and 36− 40, 78−89% yield). As observed in the dehydrogenative process, 4-Me, 3-Me, and 2-Me substitution is tolerated within the aryl alcohol in addition to electron-donating substituents (4-OMe). Considering that the reduction of electron-rich amino-substituted acetophenones is challenging via metalcatalyzed transfer hydrogenation processes, 6 it is noteworthy that piperidine can be employed as the nucleophile in this protocol, accessing 41 in 57% yield.
The methods described thus far employ acetone and formic acid as oxidant and reductant, respectively. Despite the low cost of these additives and the ease of product isolation, an alternative approach with increased atom economy was sought. 16 In 2010, Bergman, Ellman, and co-workers described the catalytic C−O bond cleavage of 2-aryloxy-1-arylethanols (Scheme 4, eq 1), 17 which serve as model compounds for lignin depolymerization studies. 18 Inspired by these reports, we envisaged developing a one-pot isomerization of 2-aryloxy-1-arylethanols to diaryl ethers, which proceeds via transfer hydrogenation to generate fluoroarenes that are electronically activated toward a subsequent S N Ar with aryl alcohols (Scheme 4, eq 2). To validate this approach, fluorinated 2-aryloxy-1arylethanol 43 was employed as a model substrate. The reaction conditions developed by Bergman, Ellman, and coworkers were modified to facilitate a subsequent nucleophilic aromatic substitution step by switching to N,N-dimethylacetamide as solvent, adding K 2 CO 3 (1.5 equiv) as base, increasing the reaction time to 24 h, and increasing the catalyst and ligand loading to 2.5 mol %.
Finally, cognizant of the limited commercial availability of 2aryloxy-1-arylethanols, we developed an alternative, telescoped synthesis of diaryl ether 3 that employs commodity epoxide 44 as the starting material. This one-pot transformation proceeds via an initial epoxide ring opening, followed by catalytic alcohol dehydrogenation/C−O bond cleavage and final S N Ar, accessing 3 in 59% isolated yield (Scheme 6).
In conclusion, we have developed a general approach for the catalytic functionalization of benzylic alcohols via electronic arene activation, accessing a diverse range of diaryl ethers and aryl amines in excellent isolated yields. Our method takes advantage of the hydrogen transfer approach to select the Scheme 3. Scope of Redox-Neutral S N Ar Protocol a a Reactions performed using 0.4 mmol of alcohol starting material. All yields are isolated yields after chromatographic purification.

Scheme 4. Previous Work and Outline of the Isomerization Strategy
Scheme 5. Scope of Isomerization Protocol a a Reactions performed using 1.0 mmol of starting material. All yields are isolated yields after chromatographic purification.

Organic Letters
Letter oxidation level of the functionalized products via the addition of simple, inexpensive additives. We have also developed a catalytic isomerization of 2-aryloxy-1-arylethanols and a telescoped synthesis of diaryl ethers directly from commodity epoxide starting materials. Ongoing studies are focused on further applications of hydrogen transfer in catalysis, and these results will be reported in due course.

Notes
The authors declare no competing financial interest. Information about the data that underpins the results presented in this article, including how to access them, can be found in the Cardiff University data catalogue at http://doi.org/10. 17035/d.2017.0044059416.