Wen-Juan
Shi
a,
Xiao-Lei
Li
a,
Zhao-Wei
Li
a and
Zhang-Jie
Shi
*ab
aBeijing National Laboratory of Molecular Sciences (BNLMS) and Key Laboratory of Bioorganic Chemistry and Molecular Engineering of Ministry of Education, College of Chemistry and Molecular Engineering and Green Chemistry Center, Peking University, Beijing, 100871, China. E-mail: zshi@pku.edu.cn
bState Key Laboratory of Organometallic Chemistry, Chinese Academy of Sciences, Shanghai 200032, China
First published on 14th January 2016
Nickel catalyzed reduction of arenols has been developed with a mutual activation strategy under mild conditions. The reduction featured a broad substrate scope, non-sensitivity to steric hindrance and no over-reduction of the aromatic rings.
Recently, impressive progress in homogeneous catalysis has been made to reduce the C–O bonds of arenol derivatives, including the esters and ethers4 (Scheme 1a). To the best of our knowledge, the in situ generated arenols as byproducts have never been reduced into the desired arenes under mild conditions (under 100 °C). Under high temperature, the Ru/W bifunctional catalyst,5a the stoichiometric LiAlH4/KOtBu combination system,5b and the hydroxycyclopentadienyl iridium complexes5c could promote the reduction of arenols. However, the aromatic rings were hydrogenated to form cyclohexanes, cyclohexanol and others in many cases.
The major challenges for the direct and selective reduction of arenols rely on the stability of the C–OH moiety: (1) the high bond dissociation energy (BDE) of its C–O bond; (2) the poor leaving ability of its hydroxyl group; (3) the deactivation of the catalyst by bonding with a phenolic anion; (4) and the further enhancement of the BDE by the p–π conjugative effect of its anion.3c,12,13 Based on our progress on the mutual activation of arenols,12 we envisioned that the reduction of arenols with suitable reductants assisted by an appropriate Lewis acid should be possible. Herein we describe a novel nickel catalyzed reduction of arenols under mild conditions.
We began our investigations by examining the reactivity of naphthalen-2-ol (1a) with several hydride sources in the combination of Ni(cod)2/PCy3/NaH in toluene (Table 1). The nature of the reductants was critical, and HBPin, BH3·SMe2, HSi(OEt)3, HSiEt3, iPrMgCl, LiAlH4, NaBH4, Zn, and H2, were all ineffective, even when accompanied by a stoichiometric amount of AlMe34g or BEt312a (entry 1). Based on our experience in the Suzuki-type coupling of arenols,12a we considered the important interaction between arenol and boron reagents. After extensive screenings, we found that a cocktail containing Ni(cod)2/PCy3/B2Pin2/K3PO4 promoted the targeted reaction in 91% yield and other diboron reagents did not show better results (entries 2–3). The inclusion of different ligands had a profound influence on the transformation (entry 4). Strikingly, the utilization of a base with regime basicity had deleterious effects on the reactivity (entry 5). The yield slightly decreased in the absence of a base (entry 6). The air and moisture-stable nickel catalyst could also catalyze the reaction (entry 7). Moreover, a difference in the reactivity was found in the diverse solvents (entries 8–9). Amazingly, the reaction occurred at room temperature (entry 10), although the yield was sacrificed (entry 11). The addition of 1.5 equivalents of B2Pin2 delivered the product in excellent yields (entry 12). Notably, the reaction was efficient while the catalyst loading was decreased to 2.5 mol%. Indeed, this is a relatively low catalyst loading in Ni catalyzed C–O bond activations, which was considered as one of the major challenges in this field (entry 13). Importantly, no over-reduction of the aromatic ring was observed.
Entry | Ligand | Reductant | Base | Solvent | Yield (%) |
---|---|---|---|---|---|
a Conditions: 1a (0.2 mmol), reductant (0.4 mmol), Ni catalyst (5 mol%), ligand (20 mol%), base (0.6 mmol), solvent (0.5 mL), 80 °C, 12 h and the yield was determined using GC analysis. b HSiEt3, HSi(OEt)3, BH3·SMe2, HBPin, Zn, iPrMgCl, NaBH4, LiAlH4, H2 were used. c Isolated yield. d Dcype, IPr·HCl, phen were used. e Cs2CO3, K2HPO4, KH2PO4, NaH, DABCO were used. f Ni(acac)2 was used. g The reaction was performed at 20 °C. h The reaction was performed at 60 °C. i B2Pin2 (1.5 equiv.)/K3PO4 (2.5 equiv.) was used. j Ni(cod)2 (2.5 mol%)/PCy3 (10 mol%) was used. k No Ni(cod)2. | |||||
1b | PCy3 | HSiEt3etc. | K3PO4 | Toluene | <5 |
2 | PCy3 | B2(nep)2 | K3PO4 | Toluene | 45 |
3 | PCy3 | B2Pin2 | K3PO4 | Toluene | 91 (80)c |
4d | Dcype etc. | B2Pin2 | K3PO4 | Toluene | <5 |
5e | PCy3 | B2Pin2 | Cs2CO3etc. | Toluene | <5–85 |
6 | PCy3 | B2Pin2 | — | Toluene | 84 |
7f | PCy3 | B2Pin2 | K3PO4 | Toluene | 65 |
8 | PCy3 | B2Pin2 | K3PO4 | THF | 53 |
9 | PCy3 | B2Pin2 | K3PO4 | DMF | <5 |
10g | PCy3 | B2Pin2 | K3PO4 | Toluene | 15 |
11h | PCy3 | B2Pin2 | K3PO4 | Toluene | 65 |
12i | PCy3 | B2Pin2 | K3PO4 | Toluene | 87 |
13j | PCy3 | B2Pin2 | K3PO4 | Toluene | 72 |
14k | PCy3 | B2Pin2 | K3PO4 | Toluene | <5 |
With the optimized conditions in hand, we found that the substrate scope encompassed electron-neutral (2a–4a), electron-poor (18a) and electron-rich naphthols (19a). The chemoselective character of this reduction was nicely illustrated by the good tolerance to diverse substituents in good yields (Table 2). Notably, an excellent yield was obtained with the silyl group (6a),14 which could be easily converted to other functionalities.15–18 Importantly, the boronic ester group survived well (7a).19 Silyloxyl groups in different positions on the aromatic ring showed competing reactivities (8–10a). N,N-Dimethylamino and free amine functionalities were tolerated well (11–12a). Substrates with nitrogen containing heterocycles, such as piperazinyl (13a), morpholinyl (14a), piperidinyl (15a) and tetrahydroquinolinyl (16a) produced the corresponding arenes in excellent yields. A worthy reactivity was found for the carbazole structural unit, a novel building block in pharmaceuticals, natural products and materials science (17a).20 The reduction took place smoothly for the substrate bearing trifluoromethyl group, which is highly valuable for wide application in material and pharmaceutical molecules (18a).21 The survival of the methoxyl and hydroxyl groups on the phenyl ring was achieved by the control of the amount of B2Pin2 (19–20a). The reaction proceeded efficiently with α-naphthol (21a) and fused-aromatic substrates (22a). Notably, the ortho substituted groups did not have an obvious effect on the efficiency, as exemplified by 23–24a giving good yields. Most importantly, the highly sterically hindered 1,3-diethylnaphthalen-2-ol (25a) showed great reactivity. In addition, the dihydroxyl groups gave efficient conversion to the product of the mono reduction (26a). Luckily, the successful preparation of 27–32a illustrated the selectivity profile among the different C–O bonds, showing the practical utility of this method for further transformations in the late-stage modification of biological compounds. Unfortunately, cyano, ester, halide and ether groups on the naphthyl ring could not be retained. Luckily, the benzyl hydroxyl group could also be deoxygenized in a high yield (33a). It is noteworthy that, over-reduction products of the aromatic ring were not observed.
A closer look into the literature indicated that the inclusion of π-extended systems strongly enhanced the reactivity of a C–O bond.22 Encouraged by the results in Table 2, we extended the substrate scope to the simple, yet, challenging phenols without the utilization of ortho-directing groups under the finely modified conditions (Table 3). The substrate bearing an ortho phenyl group generated the product, albeit in a relatively lower yield, compared to its meta and para analogues (34–36a). Both electron-rich (37a) and electron-poor (38a) groups were tolerated. Unfortunately, other π-extended systems, such as (E)-4-styrylphenol were fully recovered. Further efforts to extend this chemistry to general phenols are underway.
a Conditions: 1 (0.2 mmol), B2Pin2 (0.6 mmol), Ni(cod)2 (0.02 mmol), PCy3 (0.08 mmol), K3PO4 (0.6 mmol), benzene (0.5 mL) at 120 °C, 12 h. |
---|
To further understand this reduction, we conducted preliminary mechanistic experiments (Scheme 2). Firstly, when the deuterated naphthalen-2-ol (1aa) was subjected to the standard conditions (eqn (1)), no deuterium incorporation in the product ruled out the direct deoxygenation pathway. The use of benzene-d6 or toluene together with 1 equivalent of D2O only afforded 52% and 68% deuterium incorporation of the products (eqn (2) and (3)). Those results indicated that the solvent was at least one of the hydrogen atom sources in accordance with the literature report.23 An unsymmetrical substrate was subjected to the standard conditions with benzene-d6 as the solvent. And a H/D exchange occurred because the deuterium incorporated was observed at different aryl C–H bonds in the final product (eqn (4)). This result indicated that the C–H bonds in the naphthalene rings could be another source of the hydrogen atoms, since the C–H bonds could be cleaved by low valent metal catalysts with boron reagents,24 which made the labeling experiments complicated. Moreover, according to the result that the reduction was very sensitive to the basicity of the base, we speculated whether the reduction product could originate from the aryl boronic pinacol ester since the arylboronate product in the borylation of C(sp2)–H was reported to be possibly decomposed in the presence of K3PO4.25 The starting material was recovered when we subjected 7b to the standard reaction conditions (eqn (5)), indicating that the borylated product was not the precursor in this transformation. Finally, we found that the catalytic reduction product of 1a was generated in 80% GC yield by the addition of 20 mol% TEMPO (eqn (6)). This observation indicated that a single electron transfer process might not come into play. Finally, the unique effect of B2Pin2 stimulated us to investigate its role in this reduction under such mild conditions. From the 11B NMR spectra in Scheme S1,† no doublet peaks of HBPin were detected,26 when a mixture of 1a, B2Pin2, Ni(cod)2 and PCy3 in toluene was stirred at 80 °C for 30 min, while a new singlet peak at δ = 22.50 ppm27 indicated that a Nap-OBPin intermediate was formed. And in this reaction, a very low conversion to naphthalene was observed. Since Np-OBPin was very prone to being hydrolyzed, we failed to isolate this intermediate, but it was confirmed by GC-MS (Scheme S2†). When B2Pin2 was replaced by 4 equivalents of HBPin under the standard conditions, 1a was converted into trace amounts of 1b and the major product was Nap-OBPin, which was confirmed by GC-MS (Scheme S3†).
Based on the literature report and current mechanism studies, we proposed the catalytic cycle as shown in Scheme 3. The arenol was coordinated with B2Pin2 to form the key intermediate A. With a mutual activation strategy, the phenolic oxygen atom worked as a Lewis base to weaken the B–B bond, thus enhancing the transmetallation of B2Pin2. Meanwhile, the Lewis acidic B atoms decreased the electron density of the C–O bond, reducing the energy barrier of the oxidation addition step. After the oxidative addition of the B–B bond to the low-valent Ni species,28 the intermediate B was transformed in situ to the nickel species D and a relatively active aryl boronic ester C, which was oxidized to generate the nickel species E. The reductive elimination of F generated from the transmetallation between the two nickel species D and E afforded the final product and regenerated the nickel catalyst.29
In summary, the efficient nickel catalyzed reduction of the C–O bonds of arenols under mild conditions with a mutual activation strategy was reported. This new transformation featured a good group tolerance and non-sensitivity to the steric hindrance. Over-reduction of the arene ring was not observed. The vital role of B2Pin2 was the key to this successful transformation of arenols, which could be utilized for the further design of interesting transformations of C–O bonds.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5qo00395d |
This journal is © the Partner Organisations 2016 |