Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2024 Mar 8;9(11):12292–12306. doi: 10.1021/acsomega.4c00468

Palladium Catalysis: Dependence of the Efficiency of C–N Bond Formation on Carboxylate Ligand and Metal Carboxylate or Carboxylic Acid Additive

Jacques Muzart 1,*
PMCID: PMC10955574  PMID: 38524407

Abstract

graphic file with name ao4c00468_0017.jpg

The Pd-catalyzed inter- and intramolecular reactions of nitrogen compounds are often carried out with palladium carboxylates, sometimes in the presence of carboxylic acids or alkali metal carboxylates. This Mini-Review highlights the dependence of the reaction efficiency on the nature of the ligand and the carboxylate additives. The proposed reaction mechanisms are presented with, as far as possible, personal comments.

1. Introduction

The efficiency of the Pd(OCOR)2-catalyzed C–C bond formation may greatly depend on the carboxylate ligand and metal carboxylate or carboxylic acid additive, as highlighted in a number of reports.1 Such dependences of the C–N bond formation of reactions leading to anilines or oxidative amination, hydroamination, transamidation, and annelation2 products are the subject of the present Mini-Review, which is not exhaustive and discards the processes with amino acid additives. We will tentatively clarify the role of the carboxylate unit and the effect of its characteristics on the reaction efficiency.

2. Substitutive Aminations

Kancherla and co-workers3 recently disclosed the efficient aerobic synthesis of aniline from the cross-coupling of phenyl iodide with sodium azide under photochemical conditions and catalysis with the Pd(OAc2)/4,4′-bis(di-t-butyl)-2.2′-bipyridine association in the presence of N,N-diisopropylethylamine in moist DMSO (Scheme 1a). Switching from Pd(OAc)2 to PdCl2, Pd(OCOCF3)2, or Pd(OCOt-Bu)2 was prejudicial to the efficiency. Decreases in the yield also arose in changing N,N-diisopropylethylamine to NEt3, lowering the concentration, using other solvents, and performing the reaction under anhydrous conditions, while a slight increase occurred with bipyridine as the ligand. No reaction was observed in the absence of either the catalyst or the light. The mechanism proposed by the authors was mainly established from density functional theory studies and is summarized in Scheme 1b. Transmetalation of the PdII catalyst with sodium azide provides 1bA. Irradiation at 380 nm of the latter promotes oxidative addition of PhI to provide PdIV intermediate 1bB, which undergoes reductive elimination to give 1bD via transition state 1bC. Photoexcitation of 1bD allows N2 elimination via 1bE to afford 1bF. Subsequent reduction with N,N-diisopropylethylamine produces 1bG, which releases aniline and 1bH. Reaction of the latter with NaN3 starts another catalytic cycle.

Scheme 1. Cross-Coupling Amination via Photoexcitation of Two Distinct PdII Complexes.

Scheme 1

Open in a new tab

3. Oxidative Aminations

Thirty two years after the seminal report of Hegedus and Bozell4 on the Pd-catalyzed oxidative addition of substituted anilines to electron-deficient olefins, different experimental conditions have been independently studied by the teams of Obora5 and Jiang6 for the reactions of o-toluidine with butyl acrylate and aniline with methyl acrylate, respectively. Carried out at 60 °C in N-methylpyrrolidone under an air atmosphere, the former reaction provided better yields under catalysis with Pd(OAc)2 than with PdCl2 or Pd(OCOCF3)2, and a carboxylic acid, preferably t-BuCO2H, as additive (Scheme 2a). The second reaction, carried out at 50 °C in the presence of LiBr and molecular oxygen, showed a similar sensitivity of the yield to the nature of the Pd catalyst and an efficiency requiring a large amount of LiBr (Scheme 2b). According to Obora and co-workers, nucleophilic attack of the amine to the η2-alkenylpalladium complex 2cA provides η1-alkylpalladium complex 2cB, which undergoes β-H elimination to deliver the (Z)-enamine and palladium hydride species 2cC (Scheme 2c). Regeneration of the active catalyst from 2cC is assumed with the oxygen/carboxylic acid association.7 The excellent stereoselectivity was attributed to the intramolecular hydrogen bond of 2cB.5,6,8 The positive role of the carboxylic acid additive in the reaction of Scheme 2a could be due to its participation in the reoxidation step leading to Pd(OCO2R)2 (R = t-Bu or 2,4,6-Me3C6H2), which would be more active than the starting Pd(OAc)2. For the reaction of Scheme 2b, Jiang’s team proposed that “the role of excess bromide anion in the reaction system is to prevent PdII catalyst from deactivation by strong coordination to aromatic primary amines, thus facilitating the catalytic cycle”.6

Scheme 2. Oxidative Amination of Methyl Acrylate or 3-Butenoic Acid.

Scheme 2

Open in a new tab

Recently, Huang, Zhao, and co-workers9 disclosed the synthesis of amino acid 2d1 from the PdII-catalyzed addition of N-methylaniline to 3-butenoic acid (Scheme 2d). Preliminary experiments were performed in toluene with CuII salts and additives, alkali acetates, or carboxylic acids under air. With CuBr2 and AcONa, Pd(OAc)2 was slightly superior to PdCl2 and Pd(OCOCF3)2. With Cu(OAc)2 and alkali acetates, the efficiency depended on the nature of the alkali cation, with AcONa being superior to AcOK and AcOLi. With Cu(OAc)2 and carboxylic acids, the yield was better with t-BuCO2H than with the more acid AcOH.10 Lower yields were obtained under additive-free conditions. Finally, the best result arose with Pd(OAc)2, Cu(OAc)2 and AcONa under N2. The authors assumed that the coordination of 3-butenoic acid generates palladacycle 2eA with the assistance of AcONa (Scheme 2e). Subsequent addition of methylaniline provides palladacycle 2eB, which, according to the authors, is “followed by β-H elimination and protonation to give product” 2d1 via, without supplementary detail, the rearrangement shown in 2eC (path a). Instead of this short explanation, we propose that 2eB evolves via either the usual β-H elimination, which affords hydridopalladium complex 2eD (path b), or participation of AcONa in H abstraction, leading to carboxylate 2eE (path c). Thus, 2d1 would be obtained from either 2eD via reductive elimination or 2eE via acid-mediated hydrolysis. The preferred pathway will depend on the additive: path b uses t-BuCO2H, and path c uses AcONa. Regeneration of the catalyst occurs with Cu(OAc)2.

Ishii’s team11 relayed a very different process: the PdII-catalyzed oxidative allylic amination of 1-decene with diphenylamine (Scheme 3a). In benzotrifluoride with catalytic amounts of (NH4)5H4PMo6V6O40·23H2O (NPMoV) and air pressure, (E)-N-(dec-2-en-1-yl)-N-phenylaniline 3a1 was obtained in 73% yield under Pd(OCOCF3)2 catalysis, while PdCl2 and Pd(OAc)2 afforded no more than 4% yield. Two reaction pathways were proposed by the authors (Scheme 3b) who, however, prioritized path a over path b. Coordination of 1-decene to the catalyst afford η2-alkenylpalladium complex 3bA, which either leads to η3-allylpalladium intermediate 3bB (path a) or undergoes an aza-Wacker process to give alkylpalladium species 3bC (path b). The allylic amination product 3a1 would be obtained from either 3bB via nucleophilic addition of diphenylamine or 3bC via β-H elimination. Both pathways lead to Pd0, which is oxidized with the NPMoV/O2/CF3CO2H association.

Scheme 3. Allylic Oxidative Amination of 1-Decene.

Scheme 3

Open in a new tab

The considerable positive effect of the Pd(OCOCF3)2 catalyst on the reaction efficiency, which could be in part due to the easier formation of η3-allylpalladium complexes from olefins and this Pd salt as demonstrated by Trost and Metzner,12 urges us to also favor path a. Moreover, 3bC (path b) could be sensitive to the acid, which would result in protodepalladation to give N-decyl-N-phenylaniline.13

4. Hydroaminations

Treatment of an equimolecular mixture of N-(quinolin-8-yl)but-3-enamide 4a and succinimide in 0.1 M MeCN at 120 °C with Pd(OAc)2 catalyst provides a low yield of the hydroamination product 4a1 (Scheme 4a).14 Increased efficiency occurred in the presence of a stoichiometric amount of t-BuCO2H or PhCO2H, while the addition of CF3CO2H or HCl led to full consumption of 4a with a 1% or 0% yield 4a1, respectively. Supplementary improvement arose from concentrating the reaction solution and increasing the amount of succinimide, even with a reduced amount of the starting catalyst. Coordination of 4a to 4bA was proposed by Engle and co-workers. Subsequent nucleophilic addition of succinimide provides palladacycle 4bB, which undergoes protodepalladation to form 4a1 with regeneration of the catalyst (Scheme 4b). We suspect that the absence of β-H elimination from 4bB could be due to the large amount of RCO2H, which helps the protodepalladation step and the regeneration of the catalyst. Given the strong improvement in the presence of RCO2H (t-Bu or Ph), we assume that the active catalytic species is the corresponding Pd(OCOR)2 rather than Pd(OAc)2.

Scheme 4. Hydroamination of Unactivated Alkenes.

Scheme 4

Open in a new tab

5. Transamidations

SanMartina’s team15 recently disclosed the PdII-catalyzed transamidation of dimethylacetamide with benzylamine in the presence of carboxylic acids and azole ligands, notably L1 and L2 (Scheme 5a). In toluene with Pd(OAc)2/L1 and 3.6 equiv of AcOH, CF3CO2H, or PhCO2H under oxygen atmosphere, less than 9% of N-benzylacetamide was isolated while 39% yield was obtained with t-BuCO2H. Lowering the amount of t-BuCO2H was detrimental to the efficiency, with only traces of N-benzylacetamide being produced using this procedure under RCO2H-free conditions. Switching to diethyl carbonate as the solvent increased the yield from 39% to 62%. Increased efficiency arose using the Pd(COD)Cl2 catalyst, especially with ligand L2. The low yield obtained under an argon atmosphere led the authors to propose two complementary pathways, namely aerobic and anaerobic. Performing an array of experiments with the Pd(COD)Cl2/L2 system in diethyl carbonate, they notably observed the in situ formation of ethyl benzylcarbamate and its progressive decay.

Scheme 5. PdII-Catalyzed Transamidation.

Scheme 5

Open in a new tab

A detailed mechanism of the aerobic pathway, summarized in Scheme 5b, with Pd(OCOt-Bu)2/L2 and participation of diethyl carbonate was thus assumed by the authors. The production of ethyl benzylcarbamate from the t-BuCO2H-mediated reaction of diethyl carbonate with benzylamine releases EtOH. Pd-catalyzed oxidation of the latter via intermediate 5bA leads to acetaldehyde and hydridopalladium complex 5bB. The latter suffers either reductive elimination giving Pd0 complex 5bC or insertion of oxygen giving palladium hydroperoxide 5bD. Nucleophilic attack of the hydroxyimine tautomer 5bE of dimethylacetamide to 5bD affords 5bF. Next, nucleophilic addition of benzylamine at the electrophilic azomethine carbon leads to 5bG, which suffers prototropy and then release of dimethylamine to provide peroxy complex 5bH. Protonolysis of either the O–Pd or the O–O bond of 5bH produces peroxidic acid 5bI or 5bJ, respectively, the latter rapidly tautomerizing to N-benzylacetamide. The Radziszewski-type reaction16 of 5bI with H2O2, which is produced through the regeneration of the active PdII species from 5bC and O2,7 is another plausible pathway leading to N-benzylacetamide.

Scheme 5c summarizes the catalytic cycle proposed for the aerobic pathway. Exchange of a tert-butyl carboxylate of LPd(OCOt-Bu)2 for the hydroxyiminic tautomer of dimethylacetamide provides 5cA, which undergoes benzylamine addition leading to 5cB. Successive prototropy and release of dimethylamine give 5cC. Then, protodepalladation regenerates the catalyst with the formation of 5cD, which tautomerizes to N-benzylacetamide.

6. Annelations

6.1. N-Heterocycles

Processes related to the present Mini-Review on the synthesis of N-heterocycles via initial formation of the C–C bond have been previously documented.1c

6.1.1. Intramolecular Reactions

Stahl’s team17 disclosed in 2006 the use of previously reported N-heterocyclic carbene-coordinated PdII complexes IMesPd(OCOR1)2OH2 (R1 = Me or CF3)18 to catalyze the annellation of the cis-crotyl tosylanilide 6a under an oxygen atmosphere, leading to a mixture of five- and six-membered N-heterocycles 6a1 and 6a2, with the 5-exo cyclization product being the main product (Scheme 6a). The 6a1/6a2 ratio depended on the ligand, with the more acidic trifluoroacetate leading to the higher selectivity. With this catalyst, the addition of a catalytic amount of AcOH resulted in the selective formation of 6a1 in high yield. PhCO2H, which is somewhat more acidic than AcOH, gave slightly inferior results, while the stronger acid CF3CO2H was strongly detrimental to the process.

Scheme 6. Indole Derivatives from Intramolecular Reactions of Anilines ortho-Substituted with an Alkenyl Tether.

Scheme 6

Open in a new tab

The aerobic (η3-allylPdCl)2/IPr·HCl-catalyzed intramolecular reaction of N-(2-(2-methylallyl)phenyl)isobutyramide 6b, which mainly differs from 6a by the substitution of the alkenyl tether, was subsequently performed by Yang and co-workers19 in the presence of both a base (1.1 equiv) and a carboxylic acid (0.3 equiv) (Scheme 6b). The three- or five-membered-ring fused indoline 6b1 or 6b2 was isolated as the main compound depending on the used base, Cs2CO3 or t-BuOK, respectively. Testing a set of carboxylic acids led to the best results with t-BuCO2H, and lower conversion and yield were observed in its absence. Exchange of (η3-allylPdCl)2 for Pd(OAc)2 or Pd(OCOCF3)2 was detrimental to the process.

A different substitution of the amide unit may increase the selectivity toward the formation of the five-membered ring, as observed with the reaction of N-(2-(2-methylallyl)phenyl)acetamide 6c, which afforded 6c2 in 69–71% yields without production of 6c1 using the (η3-allylPdCl)2/IPr·HCl or Pd(OCOCF3)2/IPr·HCl catalyst with t-BuOK and t-BuCO2H in tert-amyl alcohol (Scheme 6c).

According to Yang’s team, the alkylpalladiumII6dA, formed from aminopalladation of the C=C bond of the anilide, reacts with either pivalate formed from t-BuCO2H and Cs2CO3 (path a) or tert-butoxide (path b) to provide palladium tert-butyl carboxylate 6dB or enolate 6dC, respectively (Scheme 6d). From 6dB, a benzylic C(sp2)–H abstraction, likely via the concerted metalation-deprotonation mechanism (CMD),20 affords 6dD, which releases the three-membered-ring fused indoline via reductive elimination. The formation of the five-membered-ring fused indoline arises from intramolecular addition of enolate amide 6dC leading to six-membered palladacycle 6dE, which suffers from reductive elimination. The preferred reactive pathway, path a versus path b, seems dependent on (i) the relative basicity of t-BuOCOCs (pKa ≈ 10) and t-BuOK (pKa = 19.2)10 and (ii) the substitution of the amide α-C–H, with the primary amide α-C–H of 6c making the formation of enolate 6dC easier and, consequently, favoring path b.

In 2013, Jiao’s team21 synthesized the tricyclic compound 7a1 from the aerobic PdII-catalyzed domino reaction of N-alkynyl aniline 7a in the presence of carboxylic acids (Scheme 7a). PhCO2H gave the best result, especially in the presence of slight amounts of PhCO2Li, which allowed the reduction of both the reaction temperature and the reaction time. Changing the carboxylic acid to a Lewis acid provoked the consumption of 7a without producing 7a1. Coordination of the triple bond of 7a to Pd(OAc)2 leading to 7bA and then to trans-aminopalladation product 7bB and its cis-form 7bD via tautomer 7bC was proposed by the authors (Scheme 7b, R = Me, path a). Subsequent electrophilic aromatic palladation gives the five-membered palladacycle 7bE, which undergoes reductive elimination to produce 7a1 and Pd0; recycling of the catalyst occurred with oxygen and AcOH.7 The formation of 7bE could arise via transition state 7bG, which is a CMD pathway.19

Scheme 7. Indole Derivatives from Intramolecular Reactions of N-Alkynyl Anilines.

Scheme 7

Open in a new tab

Given the presence of a large amount of PhCO2H, we propose that the active catalyst is rather the corresponding palladium carboxylate (Scheme 7b, where R = Ph). The trans-aminopalladation process of 7bA is in agreement with other intramolecular aminopalladation reactions of alkynes, which were, however, carried out under acid-free conditions.22 This reaction could involve the ammonium intermediate 7bF, whose evolution toward 7bB would be favored by the PhCO2Li additive. We considered an alternative pathway (path b) to the one requiring 7bC: coordination of both nitrogen and the triple bond of 7a1 to the catalyst23 could afford 7bH, which would directly provide the required cis-aminopalladation intermediate. The lack of annelation when the ester group of 7a was changed for a phenyl group, however, disfavors path b.

The efficiency of the annelation depends on the structure of the carboxylic acid, but there is no obvious correlation between the efficiency and their acidity or steric hindrance. In contrast, the improvement of the reactivity with PhCO2Li as an additive leads us to suspect the electrophilic aromatic substitution (SEAr) mechanism19 depicted in 7bI for the formation of 7bE.

6.1.2. Intermolecular Reactions

Fused tetracycles 8a1 and 8a2 have been synthesized by the teams of Ke and Jiang24 from the Pd-catalyzed domino reaction of aniline with 3-butenoic acid, which involves the formation of two N–C bonds, two C–C bonds, and one C–O bond (Scheme 8a). Under an air atmosphere at 45 °C in MeCN, catalysis with Pd(OCOCF3)2 provides a low yield, while no reaction occurred under nitrogen. Addition of H2O2 or better t-BuOOH to the aerobic mixture increased the yield to 69%, while the Pd(OAc)2 catalyst was less efficient. Moreover, further improvement with high diastereoselectivity arose in the presence of a bidentate ligand, the neocuporine. Meticulous studies including deuterium labelling experiments and DFT calculations led the authors to propose a highly detailed mechanism, which is summarized in Scheme 8b. Nucleophilic addition of aniline to η2-alkenylpalladium carboxylate 8bA affords η2-alkylpalladium carboxylate 8bB, which undergoes β-H elimination leading to cationic hydridopalladium species 8bC. Isomerization of the latter gives imine complex 8bD, the oxidation of which produces 8bE. Then, activation of an Ar–H bond via probably a CMD process provides 8bF, which undergoes Heck addition to a second molecule of 3-butenoic acid leading to 8bG. Subsequent β-H elimination followed by oxidation gives the ionic intermediate 8bH. Next, two successive annellations produce 8bI. Hydrogen bonding with a third molecule of 3-butenoic acid and subsequent protonolysis of the N–Pd bond with CF3CO2H deliver 8bA and 8bJ. Finally, CF3CO2H-mediated intramolecular amidation provided the fused tetracycles.

Scheme 8. Fused Benzo-aza-oxa-[5-6-5]tetracycles from the Domino Reaction of Aniline with 3-Butenoic Acid.

Scheme 8

Open in a new tab

It seems of interest to point out the decisive role of the monosubstitution of the amine unit in the formation of the tetracycle as exemplified above with N-methylaniline, which, under also PdII-catalyzed conditions, affords 4-(methyl(phenyl)amino)but-3-enoic acid (Scheme 2d).

The synthesis of dimethyl 1H-indole-2,3-dicarboxylate from the reaction of aniline with dimethyl butynedioate in DMA arose in a better yield when Pd(OAc)2 was used as the catalyst instead of Pd(OCOCF3)2 (Scheme 9a).25 Performing the Pd(OAc)2-catalyzed reaction with a small amount of K2CO3 and a large excess of a carboxylic acid may increase the yield. AcOH, t-BuCH2CO2H, and t-BuCO2H improved the results, while Me(CH2)4CO2H and Me(CH2)8CO2H decreased the efficiency. The reason for such differences cannot be based on their pKa or their steric hindrance, as AcOH and t-BuCH2CO2H, which have similar pKa values but different sizes lead to the same yield. Finally, the best conditions were the use of a 4:1 mixture of DMA and t-BuCO2H as the solvent under base-free conditions. Jiao and co-workers assumed that the activation of the alkyne leading to η2-alkynylpalladium complex 9bA precedes the reaction with aniline, which leads to either the hydroamination intermediate 9bB or the aminopalladation complex 9bC (Scheme 9b). Plausible formation of 9bC from 9bB and PdII via “an acid-promoted electrophilic aromatic palladation and subsequent proton abstraction” was proposed but, under such conditions, we are more confident in the activation of an ortho-Ar–H bond resulting in the formation of 9bD. Both 9bC and 9bD may lead to the product.

Scheme 9. Aniline/Alkyne Coupling.

Scheme 9

Open in a new tab

Using palladium, copper, and potassium acetates or pivalates in DMA under oxygen, Liang, Yang, and co-workers26 performed the decarboxylative domino reaction of o-(phenylethynyl)aniline 10a with o-bromobenzoic acid, which afforded dibenzo[a,c]carbazole 10a1 (Scheme 10a). The best yield arose using catalytic amounts of Pd(OAc)2 with overstoichiometric amounts of both Cu(OCOt-Bu)2 and KOCOt-Bu. Changing the O2 atmosphere for air or N2 was detrimental to the yield. The proposed mechanism involves coordination of the triple bond of 10a to palladium, followed by intramolecular amination to afford 10bA (Scheme 10b). Then, formation of five-membered palladacycle 10bB via C(sp2)–H bond activation is followed by tert-butylcarboxylate-mediated dealkylation leading to five-membered palladacycle 10bC. Subsequent addition of o-bromobenzoic acid provides PdIV complex 10bD, which undergoes decarboxylation/reductive elimination to give PdII species 10bE and/or 10bF. Finally, reductive elimination furnishes the carbazole and Pd0, which is oxidized with O2/Cu(OCOt-Bu)2.

Scheme 10. Domino Aminopalladation/C–H Activation/Dealkylation/Decarboxylative Cyclization Reaction.

Scheme 10

Open in a new tab

In the presence of alkali carbonates, the Pd/AgOCOR-catalyzed dimerization of 2,3-diphenyl-2H-azirine 11a selectively provided either pyrrole or pyrimidine derivatives depending on the nature of both the silver carboxylate and the alkali carbonate (Scheme 11a).27 Catalysis with PdCl2(dppf)·CH2Cl2/AgOAc in dioxane with Cs2CO3 led to a mixture of pyrrole 11a1 and pyrimidine 11a2, while 11a1 was exclusively produced when Li2CO3 was used instead of Cs2CO3. Under the latter conditions, switching to the Pd(PPh3)4 catalyst provided a similar result, with the 11a1/11a2 ratio being 33:1. In contrast, 11a2 was the only product identified when using Cs2CO3 and catalysis with Pd(PPh3)4/AgOCOCF3 in DMF.

Scheme 11. Silver Carboxylate-Mediated Chemodivergence.

Scheme 11

Open in a new tab

According to the mechanism proposed by Yao, Miao, and co-workers,27 both methods involve Pd0 catalysis and the chemodivergence results, at least in part, from different reactivities of AgOAc and AgOCOCF3 toward the azirine. The formation of both pyrrole and pyrimidine arises via the four-membered palladacycle 11bA obtained from insertion of Pd0 into the N–C bond of 11a (Scheme 11b). Metathesis of 11bA affords Pd-carbene intermediate 11bB, which attacks a second molecule of 11a to give zwitterionic intermediate 11bC. The latter leads to nitrile ylide 11bD via reaction with a third molecule of 11a that is activated by coordination to AgOAc. Cyclization of 11bD furnishes the highly strained intermediate 11bE, which gives rise to 11bF via the opening of its cyclopropyl ring. Moisture-mediated hydrolysis of the imine unit of 11bF followed by loss of NH3 provides the pyrrole. The other pathway implicates palladacycle 11bG obtained from regioselective insertion of nitrile ylide 11bH formed from AgOCOCF3-assisted ring opening of the azirine into the C–Pd bond of 11bA. Subsequent reductive elimination of Pd0, which affords dihydropyrimidine 11bI, is followed by dehydrogenation under the Pd/Ag conditions to produce the pyrimidine. The intensive studies carried out by the authors reveal that the above proposed explanation of the chemodivergence is oversimplified and could be inadequate. Indeed, the careful screening of bases, solvents, temperatures, and silver and palladium catalysts to optimize the selectivity demonstrated its dependence on various more or less identified parameters.

Recently, Engle’s team28 disclosed that the aerobic domino reaction between ether 12a, which bears the 8-aminoquinoline directing group, and the protected o-iodoaniline 12a′ in the presence of catalytic amounts of Pd(OAc)2 associated with AgOAc and 1-adamantanecarboxylic acid in MeCN produced dihydroindole 12a1 in a high yield (Scheme 12a). The absence of 1-AdCO2H, or especially AgOAc, or the replacement of AgOAc by Cu(OAc)2 or CsOCOt-Bu was detrimental to the process. A plausible mechanism begins with coordination of 12a, leading to 12bA (Scheme 12b). Subsequent C(sp3)–H abstraction provides the alkylpalladacycle 12bB, which undergoes β-heteroatom elimination rather than β-H elimination29 to afford the η2-alkenyl palladium intermediate 12bC. Then, nucleophilic addition of 12a′ produces 12bD, which undergoes intramolecular oxidative addition leading to PdIV complex 12bE. Finally, reductive elimination and ligand exchange releases 12a1 and regenerates the PdII active species.

Scheme 12. Annelation via C(sp3)–H Activation and β-Heteroatom Elimination.

Scheme 12

Open in a new tab

The C(sp3)–H abstraction could arise via either the CMD transition state 12bF or the electrophilic C–H substitution (SEC) mechanism depicted with 12bG. 1-AdCO2H could be involved via an intermediate corresponding to 12bF with 1-AdCO2 instead of AcO as the ligand, which could be more reactive than 12bF for a CMD process; such a role will, however, be minor given the modest yield decrease in the absence of the acid. AgOAc could mediate the SEC, but the inefficiency of the process using the more basic CsOCOt-Bu disfavors this reaction pathway. Consequently, the reaction probably mainly occurs via 12bF. Scavenging the iodide would be the main role of AgOAc that could also stabilize Pd intermediates via the formation of bimetallic species.

6.2. N,O-Heterocycles

Jiang and co-workers30 recently reported the synthesis of 2,3,3a,5-tetrahydro-1H-benzo[d]pyrrolo[2,1-b][1,3]oxazin-1-one from the coupling of 2-aminobenzyl alcohol with 3-butenoic acid in MeCN using t-BuOOH/O2 as the oxidant and Pd(OAc)2 or preferably Pd(OCOCF3)2 as the catalyst (Scheme 13a). The catalytic cycle assumed by the authors begins by the coordination of butenoic acid, leading to 13bA (Scheme 13b). Nucleophilic attack of the amine on the activated C=C bond followed by β-H-elimination provides hydridopalladium intermediate 13bB, which undergoes either oxidation to give 13bC or isomerization followed by oxidation to give 13bD. Subsequent intramolecular Wacker-type reaction of these species affords palladacycles 13bE and 13bF, respectively, both releasing 13bG and the catalyst via protodepalladation. Finally, 13bG undergoes intramolecular amidation to provide the product.

Scheme 13. Domino 1,1-Oxamidation and Amidation.

Scheme 13

Open in a new tab

The effective formation of CF3CO2PdOOt-Bu from Pd(OCOCF3)2 and t-BuOOH demonstrated by Mimoun’s team31 urges us to hypothesize a reaction catalyzed by such species. Besides, we suspect that interaction of the catalyst with 3-butenoic acid will give η2-alkenylpalladium carboxylate complex 13cA (Scheme 13c) rather than 13bA (Scheme 13b). Nucleophilic addition of the aminoalcohol to 13cA would afford five-membered palladacycle 13cB, which undergoes β-H-elimination followed by oxidation to provide 13cC. Then, intramolecular Wacker-type addition leads to 13cD, which evolves as above assumed.

At 120 °C in N-methylacetamide (NMA), the efficiency of the Pd-catalyzed activation of the C(sp2)–H, which mediates intramolecular addition to a nitrile unit and leads to fused polycyclic indoles 14a1 from indole 14a bearing a cyanohydrin tether at the C-3 position, was better with Pd(OAc)2/bpy than with Pd(OCOCF3)2/bpy (Scheme 14a).32 Addition of AcOH as a cosolvent led to a dramatic acceleration of the reaction and a quasi-quantitative yield. Liao and co-workers proposed a reaction beginning with the C-3 palladation of the indole core followed by 1,2-migration of palladium, leading to 14bA (Scheme 14b). Next, elimination of AcOH affords 14bB, which undergoes insertion into the cyano group to give iminopalladium intermediate 14bC. Subsequent intramolecular reaction provided palladium alcoholate 14bD. The authors are very discrete with regard to the transformation of the latter into 14a1. We hypothesize two pathways: β-OEt elimination or protonation of the O–Pd bond followed by hydrolysis of the resulting hemiacetal. The strong improvement of the reaction rate using AcOH as cosolvent leads us to favor the second possibility.

Scheme 14. Pd-Catalyzed Intramolecular C–H Addition to Nitriles.

Scheme 14

Open in a new tab

Two reports from Liao’s team33,34 related the intermolecular C3–H addition of N-methylindole to the nitrile group of cyano(aryl)methyl benzoates using palladium carboxylates associated with 2,2′-bipyridine in N-methylacetamide. With cyano(phenyl)methyl benzoate, the trisubstituted oxazole 15a1 was selectively produced in a better yield with a carboxylic acid additive, CF3CO2H, rather than AcOH and Pd(OCOCF3)2 instead of Pd(OAc)2 (Scheme 15a).33 With an oxidant, especially O2/TEMPO, and cyano(p-tolyl)methyl benzoate under acid additive-free conditions, the selectivity highly depended on the carboxylate (Scheme 15b).34 Pd(OAc)2 afforded α-imino ketone 15b1 as the main compound and low amounts of both oxazole 15b2 and α-diketone 15b3, while only the oxazole was produced with Pd(OCOCF3)2. This dichotomy may be explained by the plausible mechanism in Scheme 15c. Coordination of the cyano group of O-acyl cyanohydrin to C-3 palladated indole 15cA promotes the insertion leading to ketimine palladium complex 15cB, which undergoes intramolecular cyclization to afford 15cC (path a). Then, cleavage of the C–O bond of the heterocycle produces palladium alcoholate 15cD. Protonolysis of the latter with in situ produced RCO2H yields 15cE or 15cF as the intermediate (path a). The authors assumed the subsequent formation of 15b1 via Pd-catalyzed oxidation. Next, an α-diketone is produced from hydrolysis of 15b1. We hypothesize that 15b1 could be directly obtained from 15cD via β-H elimination (path b). In the presence of catalytic amounts of a carboxylic acid stronger than AcOH such as CF3CO2H, 15cC suffers protonolysis leading to active PdII species and 15cG, with the latter affording 15b2 (path c). Under the experimental conditions of Scheme 15a, the larger quantity of CF3CO2H due to its use as an additive may promote the selective formation of 15a1.

Scheme 15. Dependence of the Reaction Pathway on the Nature of the Carboxylate.

Scheme 15

Open in a new tab

7. Conclusion

The present review underlines the plausible influence of the nature of the carboxylate ligand and carboxylate additives on the efficiency of the Pd-catalyzed formation of C–N bonds and, in some cases, the other successive bonds. An optimized carboxylate may confer a high yield to the process. Acidity or/and steric hindrance of the carboxylate could be involved in the efficiency change, but the diversity of the results excludes a general rule. Besides, yield and selectivity may depend on the cation of the carboxylate additive.

As seen in this Mini-Review, significant advancement in C–N bond formation has been achieved in the course of the last years. Future research should prioritize a better understanding of the dependence of the reaction efficiency on the different additives.

Biography

graphic file with name ao4c00468_0001.jpg

Jacques Muzart was born in 1946, in Vienne la Ville, a small village in the Argonne area, 200 km east of Paris. He studied chemistry at l’Université de Reims Champagne-Ardenne and received his degrees (Doctorat de 3ème cycle in 1972, Doctorat d’Etat in 1976) for his work with J.-P. Pète on photochemical rearrangements of α,β-epoxyketones and β-diketones. He spent 15 months as a postdoctoral fellow of National Science Foundation working with Nobel Laureate E. J. Corey at Harvard University. Directeur de Recherche Emérite since 2011, his research interests concentrate on transition metal catalysis.

The author declares no competing financial interest.

References

  1. For reviews, see; a Le Bras J.; Muzart J. Palladium catalysis: dependence of the efficiency of C–C bond formation on carboxylate ligand and alkali metal carboxylate or carboxylic acid additive. Part A: the C(sp2)–C(sp2), C(sp2)–C(sp3) and C(sp3)–C(sp3) bonds. Adv. Synth. Catal. 2023, 365, 3727–3773. 10.1002/adsc.202300903. [DOI] [Google Scholar]; b Muzart J. Palladium catalysis: dependence of the efficiency of C–C bond formation on carboxylate ligand and alkali metal carboxylate or carboxylic acid additive. Part B: the hydro(hetero)arylation, hydroalkenylation and hydroalkylation reactions. Adv. Synth. Catal. 2023, 365, 3774–3783. 10.1002/adsc.202300921. [DOI] [Google Scholar]; c Muzart J. Palladium catalysis: dependence of the efficiency of C–C bond formation on carboxylate ligand and alkali metal carboxylate or carboxylic acid additive. Part C: the domino diarylation and annelation reactions. Adv. Synth. Catal. 2023, 365, 3784–3813. 10.1002/adsc.202301001. [DOI] [Google Scholar]
  2. We use the term “annelation”, which was originally employed by Heck’s team to name a Pd-catalyzed domino reaction leading to naphthalenes:; Wu G.; Rheingold A. L.; Geib S. J.; Heck R. F. Palladium-catalyzed annelation of aryl iodides with diphenylacetylene. Organometallics 1987, 6, 1941–1946. 10.1021/om00152a019. [DOI] [Google Scholar]; Subsequently, the terms “annulation” and “anulation” have also been used in the literature.
  3. Kancherla R.; Muralirajan K.; Dutta S.; Pal K.; Li B.; Maity B.; Cavallo L.; Rueping M. Photoexcitation of distinct divalent palladium complexes in cross-coupling amination under air. Angew. Chem., Int. Ed. 2024, 63, e202314508. 10.1002/anie.202314508. [DOI] [PubMed] [Google Scholar]
  4. Bozell J. J.; Hegedus L. S. Palladium-assisted functionalization of olefins: a new amination of electron-deficient olefins. J. Org. Chem. 1981, 46, 2561–2563. 10.1021/jo00325a024. [DOI] [Google Scholar]
  5. Mizuta Y.; Yasuda K.; Obora Y. Palladium-catalyzed Z-selective oxidative amination of ortho-substituted primary anilines with olefins under an open air atmosphere. J. Org. Chem. 2013, 78, 6332–6337. 10.1021/jo4010734. [DOI] [PubMed] [Google Scholar]
  6. Ji X.; Huang H.; Wu W.; Li X.; Jiang H. Palladium-catalyzed oxidative coupling of aromatic primary amines and alkenes under molecular oxygen: stereoselective assembly of (Z)-enamines. J. Org. Chem. 2013, 78, 11155–11162. 10.1021/jo402117r. [DOI] [PubMed] [Google Scholar]
  7. For the regeneration of active PdII species from HPdX or Pd0 and oxygen, see:; a Gligorich K. M.; Sigman M. S. Mechanistic questions about the reaction of molecular oxygen with palladium in oxidase catalysis. Angew. Chem., Int. Ed. 2006, 45, 6612–6615. 10.1002/anie.200602138. [DOI] [PubMed] [Google Scholar]; b Muzart J. Molecular oxygen to regenerate PdII active species. Chem. Asian J. 2006, 1, 508–515. 10.1002/asia.200600202. [DOI] [PubMed] [Google Scholar]; c Gligorich K. M.; Sigman M. S. Recent advancements and challenges of palladiumII-catalyzed oxidation reactions with molecular oxygen as the sole oxidant. Chem. Commun. 2009, 3854–3867. 10.1039/b902868d. [DOI] [PMC free article] [PubMed] [Google Scholar]; d Scheuermann M. L.; Goldberg K. I. Reactions of Pd and Pt Complexes with molecular oxygen. Chem. Eur. J. 2014, 20, 14556–14568. 10.1002/chem.201402599. [DOI] [PubMed] [Google Scholar]
  8. a Lee J. M.; Ahn D.-S.; Jung D. Y.; Lee J.; Do Y.; Kim S. K.; Chang S. Hydrogen-bond-directed highly stereoselective synthesis of Z-enamides via Pd-catalyzed oxidative amidation of conjugated olefins. J. Am. Chem. Soc. 2006, 128, 12954–12962. 10.1021/ja0639315. [DOI] [PubMed] [Google Scholar]; b Obora Y.; Shimizu Y.; Ishii Y. Intermolecular oxidative amination of olefins with amines catalyzed by the Pd(II)/NPMoV/O2 system. Org. Lett. 2009, 11, 5058–5061. 10.1021/ol902052z. [DOI] [PubMed] [Google Scholar]
  9. Liu Q.; Zhou Z.; Huang Z.; Zhao Y. Palladium-catalyzed E-selective oxidative amination of aromatic amine with 3-butenoic acid. J. Org. Chem. 2023, 88, 15350–15357. 10.1021/acs.joc.3c01843. [DOI] [PubMed] [Google Scholar]
  10. For the pKa values, see:; a Ripin D. H.; Evans D. A.. Evan’s pKa table. University of Pittsburgh. http://ccc.chem.pitt.edu/wipf/MechOMs/evans_pKa_table.pdf.; b Liu S.; Pedersen L. G. Estimation of molecular acidity via electrostatic potential at the nucleus and valence natural atomic orbitals. J. Phys. Chem. A 2009, 113, 3648–3655. 10.1021/jp811250r. [DOI] [PMC free article] [PubMed] [Google Scholar]; The pKa attributed to CF3CO2H in the literature varies from −0.26 to 0.6; the 0.5 value could be retained.; c Namazian M.; Zakery M.; Noorbala M. R.; Coote M. L. Accurate calculation of the pKa of trifluoroacetic acid using high-level ab initio calculations. Chem. Phys. Lett. 2008, 451, 163–168. 10.1016/j.cplett.2007.11.088. [DOI] [Google Scholar]
  11. a Shimizu Y.; Obora Y.; Ishii Y. Intermolecular aerobic oxidative allylic amination of simple alkenes with diarylamines catalyzed by the Pd(OCOCF3)2/NPMoV/O2 system. Org. Lett. 2010, 12, 1372–1374. 10.1021/ol100292g. [DOI] [PubMed] [Google Scholar]; b Obora Y.; Ishii Y. Palladium-catalyzed intermolecular oxidative amination of alkenes with amines, using molecular oxygen as terminal oxidant. Catalysts 2013, 3, 794–810. 10.3390/catal3040794. [DOI] [Google Scholar]
  12. Trost B. M.; Metzner P. J. Reaction of olefins with palladium trifluoroacetate. J. Am. Chem. Soc. 1980, 102, 3572–3577. 10.1021/ja00530a042. [DOI] [Google Scholar]
  13. Hegedus L. S.; Åkermark B.; Zetterberg K.; Olsson L. F. Palladium-assisted amination of olefins. A mechanistic study. J. Am. Chem. Soc. 1984, 106, 7122–7126. 10.1021/ja00335a042. [DOI] [Google Scholar]
  14. Gurak J. A. Jr.; Yang K. S.; Liu Z.; Engle K. M. Directed, regiocontrolled hydroamination of unactivated alkenes via protodepalladation. J. Am. Chem. Soc. 2016, 138, 5805–5808. 10.1021/jacs.6b02718. [DOI] [PubMed] [Google Scholar]
  15. Urgoitia G.; Obieta M.; Herrero M. T.; Lezama L.; SanMartin R. Molecular oxygen-induced transamidation of unactivated amides in diethyl carbonate in the presence of a palladium catalyst. Adv. Synth. Catal. 2023, 365, 4713–4725. 10.1002/adsc.202300984. [DOI] [Google Scholar]
  16. Brauer H.-D.; Eilers B.; Lange A. Formation of singlet molecular oxygen by the Radziszewski reaction between acetonitrile and hydrogen peroxide in the absence and presence of ketones. J. Chem. Soc., Perkin Trans. 2002, 2, 1288–1295. 10.1039/b202620c. [DOI] [Google Scholar]
  17. Rogers M. M.; Wendlandt J. E.; Guzei I. A.; Stahl S. S. Aerobic intramolecular oxidative amination of alkenes catalyzed by NHC-coordinated palladium complexes. Org. Lett. 2006, 8, 2257–2260. 10.1021/ol060327q. [DOI] [PubMed] [Google Scholar]
  18. a Scarborough C. C.; Grady M. J. W.; Guzei I. A.; Gandhi B. A.; Bunel E. E.; Stahl S. S. PdII complexes possessing a seven-membered N-heterocyclic carbene ligand. Angew. Chem., Int. Ed. 2005, 44, 5269–5272. 10.1002/anie.200501522. [DOI] [PubMed] [Google Scholar]; b Scarborough C. C.; Popp B. V.; Guzei I. A.; Stahl S. S. Development of 7-membered N-heterocyclic carbene ligands for transition metals. J. Organomet. Chem. 2005, 690, 6143–6155. 10.1016/j.jorganchem.2005.08.022. [DOI] [Google Scholar]
  19. Du W.; Gu Q.; Li Z.; Yang D. Palladium(II)-catalyzed intramolecular tandem aminoalkylation via divergent C(sp3)–H functionalization. J. Am. Chem. Soc. 2015, 137, 1130–1135. 10.1021/ja5102739. [DOI] [PubMed] [Google Scholar]
  20. Lafrance M.; Rowley C. N.; Woo T. K.; Fagnou K. Catalytic intermolecular direct arylation of perfluorobenzenes. J. Am. Chem. Soc. 2006, 128, 8754–8756. 10.1021/ja062509l. [DOI] [PubMed] [Google Scholar]; b Gorelsky S. I.; Lapointe D.; Fagnou K. Analysis of the concerted metalation-deprotonation mechanism in palladium-catalyzed direct arylation across a broad range of aromatic substrates. J. Am. Chem. Soc. 2008, 130, 10848–10849. 10.1021/ja802533u. [DOI] [PubMed] [Google Scholar]; c Gorelsky S. I.; Lapointe D.; Fagnou K. Analysis of the palladium-catalyzed (aromatic)C–H bond metalation-deprotonation mechanism spanning the entire spectrum of arenes. J. Org. Chem. 2012, 77, 658–668. 10.1021/jo202342q. [DOI] [PubMed] [Google Scholar]
  21. Ren L.; Shi Z.; Jiao N. Pd(II)-catalyzed aerobic oxidative intramolecular hydroamination and C–H functionalization of N-alkynyl anilines for the synthesis of indole derivatives. Tetrahedron 2013, 69, 4408–4414. 10.1016/j.tet.2013.01.031. [DOI] [Google Scholar]
  22. a Lei A.-W.; Lu X.-Y. Palladium(II)-catalyzed tandem intramolecular aminopalladation of alkynes and conjugate addition. Synthesis of oxazolidinones, imidazolidinones, and lactams. Org. Lett. 2000, 2, 2699–2702. 10.1021/ol006266s. [DOI] [PubMed] [Google Scholar]; b Shen Z.-M.; Lu X.-Y. Palladium(II)-catalyzed tandem intramolecular aminopalladation of alkynylanilines and conjugate addition for synthesis of 2,3-disubstituted indole derivatives. Tetrahedron 2006, 62, 10896–10899. 10.1016/j.tet.2006.08.086. [DOI] [Google Scholar]; c Lu X.-Y.; Han X.-L. Cationic Pd(II)-catalyzed tandem reaction of 2-arylethynylanilines and aldehydes: an efficient synthesis of substituted 3-hydroxymethyl indoles. Org. Lett. 2010, 12, 3336–3339. 10.1021/ol1011086. [DOI] [PubMed] [Google Scholar]
  23. Minatti A.; Muñiz K. Intramolecular aminopalladation of alkenes as a key step to pyrrolidines and related heterocycles. Chem. Soc. Rev. 2007, 36, 1142–1152. 10.1039/B607474J. [DOI] [PubMed] [Google Scholar]
  24. Liu C.; Tan X.; Zhan L.; Jing Y.; Wu W.; Ke Z.; Jiang H. Palladium-catalyzed cascade cyclization for the synthesis of fused benzo-aza-oxa-[5–6-5] tetracycles. Angew. Chem., Int. Ed. 2022, 61, e202215020. 10.1002/anie.202215020. [DOI] [PubMed] [Google Scholar]
  25. Shi Z.; Zhang C.; Li S.; Pan D.; Ding S.; Cui Y.; Jiao N. Indoles from simple anilines and alkynes: palladium-catalyzed C–H activation using dioxygen as the oxidant. Angew. Chem., Int. Ed. 2009, 48, 4572–4576. 10.1002/anie.200901484. [DOI] [PubMed] [Google Scholar]
  26. Chen X.; Chen Y.; Xu W.; Deng D.; Liang Y.; Yang Y. Pd(II)-catalyzed synthesis of polycyclic heteroarenes via an aminopalladation/C–H activation/dealkylation/decarboxylative cyclization cascade. Org. Lett. 2022, 24, 7282–7287. 10.1021/acs.orglett.2c02657. [DOI] [PubMed] [Google Scholar]
  27. Zhao Y.; Li R.; Zhao Q.; Yao J.; Miao M. Palladium/silver cocatalyzed divergent dimerization of 2H-azirines for regioselective synthesis of tetrasubstituted pyrroles and pyrimidines. Org. Lett. 2023, 25, 3978–3983. 10.1021/acs.orglett.3c01434. [DOI] [PubMed] [Google Scholar]
  28. Ni H.-Q.; Dai J.-C.; Yang S.; Loach R. P.; Chuba M. D.; McAlpine I. J.; Engle K. M. Catalytic σ-bond annulation with ambiphilic organohalides enabled by β-X elimination. Angew. Chem., Int. Ed. 2023, 62, e202306581. 10.1002/anie.202306581. [DOI] [PubMed] [Google Scholar]
  29. Le Bras J.; Muzart J. β-Elimination competitions leading to C = C bonds from alkylpalladium intermediates. Tetrahedron 2012, 68, 10065–10113. 10.1016/j.tet.2012.09.076. [DOI] [Google Scholar]
  30. Liu C.; Tan X.; Zhang J.; Wu J.; Wu W.; Jiang H. Palladium-catalyzed 1,1-oxamidation and 1,1-diamination of unactivated alkenyl carbonyl compounds. Org. Lett. 2023, 25, 2701–2706. 10.1021/acs.orglett.3c00819. [DOI] [PubMed] [Google Scholar]
  31. Mimoun H.; Charpentier R.; Mitschler A.; Fischer J.; Weiss R. Palladium(II) tert-butyl peroxide carboxylates. New reagents for the selective oxidation of terminal olefins to methyl ketones. The role of peroxymetalation in selective oxidative processes. J. Am. Chem. Soc. 1980, 102, 1047–1054. 10.1021/ja00523a023. [DOI] [Google Scholar]
  32. Wang T.-T.; Zhao L.; Zhang Y.-J.; Liao W.-W. Pd-catalyzed intramolecular cyclization via direct C–H addition to nitriles: skeletal diverse synthesis of fused polycyclic indoles. Org. Lett. 2016, 18, 5002–5005. 10.1021/acs.orglett.6b02460. [DOI] [PubMed] [Google Scholar]
  33. Zhang D.; Song H.; Cheng N.; Liao W.-W. Synthesis of 2,4,5-trisubstituted oxazoles via Pd-catalyzed C–H addition to nitriles/cyclization sequences. Org. Lett. 2019, 21, 2745–2749. 10.1021/acs.orglett.9b00700. [DOI] [PubMed] [Google Scholar]
  34. Cui S.-Q.; Zhang D.-B.; Wei Z.-L.; Liao W.-W. Construction of functionalized α-imino ketones via Pd-catalyzed C–H addition to nitriles/aerobic oxidation sequences. J. Org. Chem. 2023, 88, 16018–16023. 10.1021/acs.joc.3c01934. [DOI] [PubMed] [Google Scholar]

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES