Synlett 2021; 32(15): 1555-1559
DOI: 10.1055/s-0040-1705943
cluster
Modern Nickel-Catalyzed Reactions

Decarbonylative Synthesis of Aryl Nitriles from Aromatic Esters and Organocyanides by a Nickel Catalyst

Keiichiro Iizumi
a   Department of Applied Chemistry, Waseda University, 513 Wasedatsurumakicho, Shinjuku, Tokyo, 162-0041, Japan   Email: junyamaguchi@waseda.jp
,
Miki B. Kurosawa
a   Department of Applied Chemistry, Waseda University, 513 Wasedatsurumakicho, Shinjuku, Tokyo, 162-0041, Japan   Email: junyamaguchi@waseda.jp
,
Ryota Isshiki
a   Department of Applied Chemistry, Waseda University, 513 Wasedatsurumakicho, Shinjuku, Tokyo, 162-0041, Japan   Email: junyamaguchi@waseda.jp
,
Kei Muto
b   Institute for Advanced Study, Waseda University, 513 Wasedatsurumakicho, Shinjuku, Tokyo, 162-0041, Japan
,
a   Department of Applied Chemistry, Waseda University, 513 Wasedatsurumakicho, Shinjuku, Tokyo, 162-0041, Japan   Email: junyamaguchi@waseda.jp
› Author Affiliations

This work was supported by JSPS KAKENHI Grant Number JP19H02726 (to J.Y.), JP20H04829 (hybrid catalysis), and JP19K15573 (to K.M.).
 


Abstract

A decarbonylative cyanation of aromatic esters with aminoacetonitriles in the presence of a nickel catalyst was developed. The key to this reaction was the use of a thiophene-based diphosphine ligand, dcypt, permitting the synthesis of aryl nitrile without the generation of stoichiometric metal- or halogen-containing chemical wastes. A wide range of aromatic esters, including hetarenes and pharmaceutical molecules, can be converted into aryl nitriles.


#

Aryl nitriles are important structures in pharmaceuticals and organic electronic materials.[1] A method widely used for synthesizing these molecules is the metal-catalyzed cyanation of haloarenes with cyanating reagents (Figure [1a]).[2] Recently, this area of chemical methodology has evolved to make the process of synthesizing aryl nitriles cleaner and safer. Conventionally, metal cyanides such as CuCN,[3] Zn(CN)2,[4] KCN,[5] TMSCN,[6] or others[7] are often used. Although these reagents are reliable, the generation of hazardous HCN and stoichiometric metal waste is often problematic. Replacing metal cyanides with organic cyano compounds such as cyanohydrins or acetonitrile has permitted a safe and practical synthesis of aryl nitriles.[8] Another consideration in this area is the use of aryl electrophiles other than haloarenes to avoid the production of corrosive halogen-containing wastes during the reactions. To this end, alternative catalytic methods using arenol- or aniline-based electrophiles have emerged.[9] Despite this progress toward streamlined cyanation protocols, methods that achieve the replacement of both coupling partners with organic cyanides and non-halogen-based aryl electrophiles remain rare; our previous Ni-catalyzed cyanation of arenols with aminoacetonitriles[10] is one such example (Figure [1b]).[11] [12]

Zoom Image
Figure 1 Metal-catalyzed cyanations

Meanwhile, our group and others have extensively studied the development of metal-catalyzed decarbonylative cross-coupling of aromatic esters, because these compounds are abundant and easily prepared.[13] [14] [15] Our group discovered that the use of the thiophene-based diphosphine 3,4-thiene-2,3-diylbis(dicyclohexylphosphine) (dcypt)[16] as an ancillary ligand for nickel and palladium catalysts showed a good performance in various decarbonylative coupling reactions of aromatic esters with nucleo­philes. We then sought to extend the scope of these decarbonylative reactions to cyanation. Although it is known that aromatic esters can be transformed into aryl nitriles through a stepwise procedure involving hydrolysis, primary amide formation, and dehydration, a decarbonylative coupling would make the synthesis of aryl nitriles shorter and more straightforward. Before this study, the Rueping group reported a decarbonylative cyanation of aromatic esters by using nickel catalysis, albeit with Zn(CN)2 as a cyano source.[17] Here, we report the development of a decarbonylative cyanation of aromatic esters with aminoacetonitriles as nonmetal cyanating reagents (Figure [1c]).

Table 1 Condition Optimizationa

Entry

Ligand

Base

Solvent

Yieldb (%)

 1

dcypt

Na3PO4

1,4-dioxane

23

 2

dcype

Na3PO4

1,4-dioxane

 7

 3

dppe

Na3PO4

1,4-dioxane

 0

 4

PCy3·HBF4

Na3PO4

1,4-dioxane

 0

 5

PBu3

Na3PO4

1,4-dioxane

 0

 6

ICy·HBF4

Na3PO4

1,4-dioxane

 0

 7

dcypt

K3PO4

1,4-dioxane

15

 8

dcypt

Li3PO4

1,4-dioxane

 0

 9

dcypt

K2CO3

1,4-dioxane

17

10

dcypt

Na2CO3

1,4-dioxane

44

11

dcypt

none

1,4-dioxane

21

12c,d

dcypt

Na2CO3

1,4-dioxane

62 (57)e

13c

dcypt

Na2CO3

toluene

63

14c

dcypt

Na2CO3

t-AmylOH

47

a Reaction conditions: 1A (0.20 mmol), 2a (0.40 mmol), Ni(OAc)2 (5.0 mol%), ligand (bidentate: 10 mol%; monodentate: 20 mol%), base (1.5 equiv), solvent (0.80 mL), 150 °C, 12 h.

b GC yield.

c Ni(OAc)2 (10 mol%), dcypt (20 mol%).

d 160 °C.

e Isolated yield.

f Reaction conditions: 1A (0.20 mmol), 2ag (0.40 mmol), Ni(OAc)2 (10 mol%), dcypt (20 mol%), Na2CO3 (1.5 equiv), 1,4-dioxane (0.80 mL), 150 °C, 12 h.

Zoom Image
Scheme 1 Substrate scope. Reagents and conditions: 1 (0.40 mmol), 2a (0.80 mmol), Ni(OAc)2 (10 mol%), dcypt (20 mol%), Na2CO3 (1.5 equiv), 1,4-dioxane (1.6 mL), 160 °C, 12 h. a 24 h. b Toluene solvent instead of 1,4-dioxane. c Na3PO4 instead of Na2CO3. d 150 °C. e 1H NMR yield.

At the outset of this study, we explored the reaction conditions by using phenyl 2-naphthoate (1A) and morpholin-4-ylacetonitrile (2a) as model reactants (Table [1]). Pleasingly, our initial attempt using Ni(OAc)2 (5 mol%), dcypt, and Na3PO4 in 1,4-dioxane at 150 °C gave 2-naphthonitrile (3A) in 23% yield (Table [1], entry 1). Encouraged by this result, we first surveyed the effect of various ligands. Despite being a structurally similar ligand, ethane-1,2-diylbis(dicyclohexylphosphine) (dcype) decreased the yield of 3A to 7% (entry 2). 1,2-Bis(diphenylphosphino)ethane (dppe), a less electron-donating ligand, did not give 3A at all (entry 3). Monodentate phosphines and N-heterocyclic carbene ligands were totally ineffective, resulting in no reaction (entries 4–6). With dcypt, the effect of base was then investigated. Changing from Na3PO4 to K3PO4 or Li3PO4 did not improve the yield (entries 7 and 8). We found that carbonate bases were superior to phosphates; in particular, the use of Na2CO3 gave the best result (entries 9 and 10). A reaction without a base also furnished the product, albeit in only 21% yield (entry 11). At this time, the precise role of the base is unclear. Increasing the catalyst loading to 10 mol% and the temperature to 160 °C gave a better yield of 3A (62%; entry 12). We found that toluene was also an effective solvent, giving 3A in almost comparable yield (entry 13). tert-Amyl alcohol (t-AmylOH) also delivered 3A in 47% yield, but this was accompanied by significant decomposition of 1A (entry 14). We also screened various organic cyanides and we found that morpholin-4-ylacetonitrile (2a) was the best option. Piperidin-1-yl-, pyrrolidin-1-yl, 4-methylpiperazin-1-yl, and (dimethylamino)acetonitriles also gave 3A, albeit in lower yields. The use of aminonitriles was important, because neither phenylacetonitrile nor (phenylsulfanyl)acetonitrile gave any 3A. Finally, we identified the optimal conditions as involving the use of Ni(OAc)2/dcypt catalyst and Na2CO3 in 1,4-dioxane or toluene at 160 °C.

With the optimized conditions in hand, we examined the scope of this decarbonylative cyanation (Scheme [1]). 1-Naphthonitrile (3B) was generated in a lower yield (47%) than 2-naphthonitrile (3A), even at higher temperatures; this decrease was probably due to steric hindrance. Naphthalenes bearing electron-donating substituents were converted into the corresponding aryl nitriles 3C and 3D in moderate yields. Anthracene and pyrene derivatives also showed good reactivities in this decarbonylative cyanation, giving 3E and 3F, respectively. Various benzoic acid derivatives were also readily cyanated: 4- and 3-methylbenzonitrile (3G and 3H, respectively) were obtained from the corresponding aromatic esters. Although the yield of 2-methylbenzonitrile was less than 5% (see Supporting Information), phenyl 9H-fluorene-1-carboxylate (1I) gave nitrile 3I in a moderate yield (51%). 2-, 3-, and 4-Methoxybenzonitriles (3J, 3K, and 3L, respectively) were all generated in acceptable yields. Nitriles containing other electron-donating substituents such as dimethylamino (3M, 3N) or phenoxy (3O) groups were also readily obtained. Moreover, the present protocol was compatible with various electron-deficient aromatic esters, providing the corresponding aryl nitriles bearing cyano (3P), methoxycarbonyl (3Q), and or sulfone groups (3R) in moderate to good yields. A pharmaceutically important aromatic ester derived from probenecid, containing a sulfonamide group, readily reacted under the conditions to give nitrile 3S in 55% yield. Furthermore, the present conditions could be applied to heteroaromatic esters, and various quinolinyl and pyridyl nitriles 3TX were obtained in moderate yields. Finally, we succeeded in synthesizing cinnamonitrile (3Y), albeit in only 33% yield.[18]

Zoom Image
Scheme 2 (A) Reactions using other aryl sources. (B) A reaction with NaCN. (C) A reaction of acyl nitrile 9 (D) (Aminoalkyl)phenol 10 as a byproduct with 3A. (E) Postulated mechanism for the release of cyanide from 2a.

To gain insights into the mechanism of this reaction, we conducted several control experiments (Scheme [2]). First, we subjected various arylcarboxylic acid derivatives such as the acid chloride 4 [17b] acid anhydrides 5 and 6, aldehyde 7, and methyl ester 8 to the reaction conditions (Scheme [2]A). However, none of these gave the desired aryl nitrile 3A. NaCN, a typical cyanating agent, was then used instead of 2a, giving 3A in 26% yield (Scheme [2]B). Moreover, the use of acyl nitrile 9 under the standard conditions gave a good yield of 1-naphthonitrile (3B; Scheme [2]C).[17a] These results support the involvement of a cyanide (CN) species under the present catalytic conditions, generating an aryl–Ni–CN species as a catalytic intermediate. During our investigations, we found that the reaction of nitrile 2a gave the (aminoalkyl)phenol 10 as a byproduct in 41% yield (Scheme [2]D). This result indicated that the aminoacetonitrile 2a releases cyanide along with the generation of an iminium species. To shed light on the mechanism of this cyanide-releasing step, we conducted the reaction between 2a and sodium phenoxide (11) in the presence of Na2CO3 (Scheme [2]E). This reaction did not proceed at all; however, the same reaction with a nickel catalyst afforded 10 in 40% yield. Although the detailed role of nickel is unclear at this stage, this result shows that the nickel catalyst is involved in the cyanide-releasing process.

Taking these control experiments into consideration, we postulated the following Ni(0)/Ni(II) catalytic cycle (Scheme [3]). The reaction might be initiated by oxidative addition of 1 to a Ni(0) species, followed by decarbonylation to give an aryl–Ni(II)–OPh species. Cyanide transfer from the aminoacetonitrile to the Ni(II) species could occur to give an aryl–Ni(II)–CN species that might finally release product 3 through reductive elimination.

Zoom Image
Scheme 3 A plausible catalytic cycle

In summary, we have developed a decarbonylative cyanation of aromatic esters with aminoacetonitriles in the presence of a Ni/dcypt catalyst.[19] The method not only provided a direct synthesis of aryl nitriles from esters, but also permits halogen- and metal-waste-free catalytic cyanation. Further studies to shed light on the detailed mechanism of this reaction and to develop a method with a broader scope are underway in our laboratory.


#

Acknowledgment

The Materials Characterization Central Laboratory in Waseda University is acknowledged for their support of the HRMS measurements.

Supporting Information


Corresponding Author

Junichiro Yamaguchi
Department of Applied Chemistry, Waseda University
513 Wasedatsurumakicho, Shinjuku, Tokyo, 162-0041
Japan   

Publication History

Received: 24 August 2020

Accepted after revision: 17 September 2020

Article published online:
16 October 2020

© 2020. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


Zoom Image
Figure 1 Metal-catalyzed cyanations
Zoom Image
Scheme 1 Substrate scope. Reagents and conditions: 1 (0.40 mmol), 2a (0.80 mmol), Ni(OAc)2 (10 mol%), dcypt (20 mol%), Na2CO3 (1.5 equiv), 1,4-dioxane (1.6 mL), 160 °C, 12 h. a 24 h. b Toluene solvent instead of 1,4-dioxane. c Na3PO4 instead of Na2CO3. d 150 °C. e 1H NMR yield.
Zoom Image
Scheme 2 (A) Reactions using other aryl sources. (B) A reaction with NaCN. (C) A reaction of acyl nitrile 9 (D) (Aminoalkyl)phenol 10 as a byproduct with 3A. (E) Postulated mechanism for the release of cyanide from 2a.
Zoom Image
Scheme 3 A plausible catalytic cycle