The Reims Journey Towards Discovery and Understanding of Pd-Catalyzed Oxidations The Reims Journey Towards Discovery and Understanding of Pd-Catalyzed Oxidations

: This review recounts the development by the authors of the Pd-catalyzed procedures devoted to various kinds of oxidation. Starting with reactions assisted with UV light, the research has explored reactions under light-free conditions: allylic oxidation, alcohol oxidation, etheriﬁcation, Wacker oxidation and dehydrogenations with, always, accompanying e ﬀ orts towards mechanism determination. Abstract : This review recounts the development by the authors of the Pd-catalyzed procedures devoted to various kinds of oxidation. Starting with reactions assisted with UV light, the research has explored reactions under light-free conditions: allylic oxidation, alcohol oxidation, etherification, Wacker oxidation and dehydrogenations with, always, accompanying efforts towards mechanism determination.


Introduction
Our initial studies on Pd-catalyzed oxidations results from our observation in the early 1980s of the formation of unsaturated carbonyl compounds from irradiation with UV light of bis(µ-chloro)bis(η 3 -allyl)dipalladium complexes in oxygenated acetonitrile (Scheme 1) [1]. That result urged us to look for such reactions under catalytic conditions. This activity progressively led us to study a variety of Pd-catalyzed oxidations under light-free conditions. The aim of the present account is to highlight the main results that we obtained in the area over the years.

Introduction
Our initial studies on Pd-catalyzed oxidations results from our observation in the early 1980s of the formation of unsaturated carbonyl compounds from irradiation with UV light of bis(µchloro)bis(η 3 -allyl)dipalladium complexes in oxygenated acetonitrile (Scheme 1) [1]. That result urged us to look for such reactions under catalytic conditions. This activity progressively led us to study a variety of Pd-catalyzed oxidations under light-free conditions. The aim of the present account is to highlight the main results that we obtained in the area over the years.

Ethylenic Compounds
The easy formation at room temperature of η 3 -allylpalladium complexes from alkenes and Pd(OCOCF3)2 reported by Trost and Metzner [2] urged us to use this Pd salt as a catalyst for the photo-assisted oxidation of alkenes [3,4]. For an example, the reaction of 1-eicosene (1) in MeCN/CH2Cl2 afforded a mixture of saturated and unsaturated ketones 2 to 4 (Equation (1)). Some migration of the double bond of 1 was a competing reaction [5,6]. In contrast to the isomerization, ketones 2 to 4 were not produced in the absence of light. Switching to acetone as the solvent increased the ketone yields. Similar results were obtained using [(η 3 -CH2CHCHC3H7)Pd(OCOCF3)]2 as the catalyst whereas the turnover number was inferior to 1 with Pd(OAc)2 [4]. Under alkene-free conditions, monitoring the irradiation of an acetone solution of Pd(OCOCF3)2 showed the adsorption Scheme 1. Photocatalyzed oxidation of bis(µ-chloro)bis(η 3 -allyl)dipalladium complexes

Ethylenic Compounds
The easy formation at room temperature of η 3 -allylpalladium complexes from alkenes and Pd(OCOCF 3 ) 2 reported by Trost and Metzner [2] urged us to use this Pd salt as a catalyst for the photo-assisted oxidation of alkenes [3,4]. For an example, the reaction of 1-eicosene (1) in MeCN/CH 2 Cl 2 afforded a mixture of saturated and unsaturated ketones 2 to 4 (Equation (1)). Some migration of the double bond of 1 was a competing reaction [5,6]. In contrast to the isomerization, ketones 2 to 4 were not produced in the absence of light. Switching to acetone as the solvent increased the ketone yields. Similar results were obtained using [(η 3 -CH 2 CHCHC 3 H 7 )Pd(OCOCF 3 )] 2 as the catalyst whereas the turnover number was inferior to 1 with Pd(OAc) 2 [4]. Under alkene-free conditions, monitoring the irradiation of an acetone solution of Pd(OCOCF 3 ) 2 showed the adsorption of oxygen [4]. That led us to of oxygen [4]. That led us to suspect the formation of peroxydic species. Nevertheless, ketones 2 to 4 were also produced from the reaction of 1 in the presence of radical and 1 O2 traps [4]. (1) The process was used for the oxidation of allylsulfones [7]. Under the above conditions, 5 led to a mixture of unsaturated aldehyde 6 and alcohol 7 (Equation (2)). The yields were improved with a cocatalyst such as Cu(OCOCF3)2 or Co(OCOCF3)2. (2) Under both Pd(OCOCF3)2 catalysis and UV light, allylsulfones substituted with a trimethylsilyl group underwent cleavage of the allyl-Si bond. Thus, full conversion of 8 occurred in 38 h leading to 6 in 95% yield (Equation (3)) [8]. Lower conversions were mediated by other Pd catalysts. (3) The mechanism of the reaction of allylsilanes was studied using l-phenyl-3-(trimethylsilyl)-lpropene (9) and l-phenyl-1-(trimethylsilyl)-2-propene (10) as substrates, and comparison with the reactivity of corresponding allylpalladium chloride 11 [9]. The latter would be dissymmetric due to the different substitution of the allyl unit extremities [10][11][12]. Cinnamaldehyde (12) was selectively obtained from irradiation of either 9 and 10 in the presence of catalytic Pd(OCOCF3)2, or 11 (Scheme 2). That contrasts from oxidations of the free phenylallyl radical which gave mixtures of oxidation products in 1-and 3-position [9,13,14]. Consequently, the regiospecificity of the reactions of 9, 10 and 11 excluded the formation of the free phenylallyl radical. The formation of η 3 -allylpalladium complexes from allylsilanes and Pd II salts is known [15]. Thus, a common intermediate, which would be a dissymetric η 3 -allylpalladium complex, has been proposed (Scheme 3). Light-mediated cleavage of the longer C-Pd bond would lead to intermediate 3 A [16,17] which reacts with oxygen to afford a peroxopalladium complex. The latter evolves towards aldehyde 12, liberating HOPdX which would be involved in the subsequent catalytic cycle. (1) The process was used for the oxidation of allylsulfones [7]. Under the above conditions, 5 led to a mixture of unsaturated aldehyde 6 and alcohol 7 (Equation (2)). The yields were improved with a cocatalyst such as Cu(OCOCF 3 ) 2 or Co(OCOCF 3 ) 2 .
Catalysts 2020, 10, x FOR PEER REVIEW 2 of 22 of oxygen [4]. That led us to suspect the formation of peroxydic species. Nevertheless, ketones 2 to 4 were also produced from the reaction of 1 in the presence of radical and 1 O2 traps [4]. (1) The process was used for the oxidation of allylsulfones [7]. Under the above conditions, 5 led to a mixture of unsaturated aldehyde 6 and alcohol 7 (Equation (2)). The yields were improved with a cocatalyst such as Cu(OCOCF3)2 or Co(OCOCF3)2. (2) Under both Pd(OCOCF3)2 catalysis and UV light, allylsulfones substituted with a trimethylsilyl group underwent cleavage of the allyl-Si bond. Thus, full conversion of 8 occurred in 38 h leading to 6 in 95% yield (Equation (3)) [8]. Lower conversions were mediated by other Pd catalysts. (3) The mechanism of the reaction of allylsilanes was studied using l-phenyl-3-(trimethylsilyl)-lpropene (9) and l-phenyl-1-(trimethylsilyl)-2-propene (10) as substrates, and comparison with the reactivity of corresponding allylpalladium chloride 11 [9]. The latter would be dissymmetric due to the different substitution of the allyl unit extremities [10][11][12]. Cinnamaldehyde (12) was selectively obtained from irradiation of either 9 and 10 in the presence of catalytic Pd(OCOCF3)2, or 11 (Scheme 2). That contrasts from oxidations of the free phenylallyl radical which gave mixtures of oxidation products in 1-and 3-position [9,13,14]. Consequently, the regiospecificity of the reactions of 9, 10 and 11 excluded the formation of the free phenylallyl radical. The formation of η 3 -allylpalladium complexes from allylsilanes and Pd II salts is known [15]. Thus, a common intermediate, which would be a dissymetric η 3 -allylpalladium complex, has been proposed (Scheme 3). Light-mediated cleavage of the longer C-Pd bond would lead to intermediate 3 A [16,17] which reacts with oxygen to afford a peroxopalladium complex. The latter evolves towards aldehyde 12, liberating HOPdX which would be involved in the subsequent catalytic cycle. (2) Under both Pd(OCOCF 3 ) 2 catalysis and UV light, allylsulfones substituted with a trimethylsilyl group underwent cleavage of the allyl-Si bond. Thus, full conversion of 8 occurred in 38 h leading to 6 in 95% yield (Equation (3)) [8]. Lower conversions were mediated by other Pd catalysts.
Catalysts 2020, 10, x FOR PEER REVIEW 2 of 22 of oxygen [4]. That led us to suspect the formation of peroxydic species. Nevertheless, ketones 2 to 4 were also produced from the reaction of 1 in the presence of radical and 1 O2 traps [4]. (1) The process was used for the oxidation of allylsulfones [7]. Under the above conditions, 5 led to a mixture of unsaturated aldehyde 6 and alcohol 7 (Equation (2)). The yields were improved with a cocatalyst such as Cu(OCOCF3)2 or Co(OCOCF3)2. (2) Under both Pd(OCOCF3)2 catalysis and UV light, allylsulfones substituted with a trimethylsilyl group underwent cleavage of the allyl-Si bond. Thus, full conversion of 8 occurred in 38 h leading to 6 in 95% yield (Equation (3)) [8]. Lower conversions were mediated by other Pd catalysts. (3) The mechanism of the reaction of allylsilanes was studied using l-phenyl-3-(trimethylsilyl)-lpropene (9) and l-phenyl-1-(trimethylsilyl)-2-propene (10) as substrates, and comparison with the reactivity of corresponding allylpalladium chloride 11 [9]. The latter would be dissymmetric due to the different substitution of the allyl unit extremities [10][11][12]. Cinnamaldehyde (12) was selectively obtained from irradiation of either 9 and 10 in the presence of catalytic Pd(OCOCF3)2, or 11 (Scheme 2). That contrasts from oxidations of the free phenylallyl radical which gave mixtures of oxidation products in 1-and 3-position [9,13,14]. Consequently, the regiospecificity of the reactions of 9, 10 and 11 excluded the formation of the free phenylallyl radical. The formation of η 3 -allylpalladium complexes from allylsilanes and Pd II salts is known [15]. Thus, a common intermediate, which would be a dissymetric η 3 -allylpalladium complex, has been proposed (Scheme 3). Light-mediated cleavage of the longer C-Pd bond would lead to intermediate 3 A [16,17] which reacts with oxygen to afford a peroxopalladium complex. The latter evolves towards aldehyde 12, liberating HOPdX which would be involved in the subsequent catalytic cycle.
The mechanism of the reaction of allylsilanes was studied using l-phenyl-3-(trimethylsilyl)-l-propene (9) and l-phenyl-1-(trimethylsilyl)-2-propene (10) as substrates, and comparison with the reactivity of corresponding allylpalladium chloride 11 [9]. The latter would be dissymmetric due to the different substitution of the allyl unit extremities [10][11][12]. Cinnamaldehyde (12) was selectively obtained from irradiation of either 9 and 10 in the presence of catalytic Pd(OCOCF 3 ) 2 , or 11 (Scheme 2). That contrasts from oxidations of the free phenylallyl radical which gave mixtures of oxidation products in 1-and 3-position [9,13,14]. Consequently, the regiospecificity of the reactions of 9, 10 and 11 excluded the formation of the free phenylallyl radical. The formation of η 3 -allylpalladium complexes from allylsilanes and Pd II salts is known [15]. Thus, a common intermediate, which would be a dissymetric η 3 -allylpalladium complex, has been proposed (Scheme 3). Light-mediated cleavage of the longer C-Pd bond would lead to intermediate 3 A [16,17] which reacts with oxygen to afford a peroxopalladium complex. The latter evolves towards aldehyde 12, liberating HOPdX which would be involved in the subsequent catalytic cycle.

Terminal Alkenes
Various conditions have been reported for the synthesis of allylic carboxylates via Pd II -catalyzed allylic oxidation of olefins in carboxylic acids [23,24]. After the observation of the improvement of the Pd(OAc)2-catalyzed allylic acetoxylation of allylbenzene with BQ as the stoichiometric oxidant in the presence of base but with inconsistent yields, we performed the efficient, reproducible and regioselective allylic acyloxylation of terminal alkenes using lithium hydroxide as the additive and propionic acid as the solvent (Equation (8)) [25]. The Pd II -catalyzed oxidation of allylsulfone 5 was carried out with t-BuOOH or oxygen in conjunction with either CuCl or benzoquinone (BQ) (Equations (6) and (7)) [21]. Under the former conditions, the main products were alcohol 7 and peroxide 19, which could be produced from a η 2 -olefin palladium complex formed from 5 and t-BuOOPdOCOCF 3 (Scheme 4) [22].

Terminal Alkenes
Various conditions have been reported for the synthesis of allylic carboxylates via Pd II -catalyzed allylic oxidation of olefins in carboxylic acids [23,24]. After the observation of the improvement of the Pd(OAc)2-catalyzed allylic acetoxylation of allylbenzene with BQ as the stoichiometric oxidant in the presence of base but with inconsistent yields, we performed the efficient, reproducible and regioselective allylic acyloxylation of terminal alkenes using lithium hydroxide as the additive and propionic acid as the solvent (Equation (8)) [25].

Terminal Alkenes
Various conditions have been reported for the synthesis of allylic carboxylates via Pd II -catalyzed allylic oxidation of olefins in carboxylic acids [23,24]. After the observation of the improvement of the Pd(OAc)2-catalyzed allylic acetoxylation of allylbenzene with BQ as the stoichiometric oxidant in the presence of base but with inconsistent yields, we performed the efficient, reproducible and regioselective allylic acyloxylation of terminal alkenes using lithium hydroxide as the additive and propionic acid as the solvent (Equation (8)) [25].

Terminal Alkenes
Various conditions have been reported for the synthesis of allylic carboxylates via Pd II -catalyzed allylic oxidation of olefins in carboxylic acids [23,24]. After the observation of the improvement of the Pd(OAc)2-catalyzed allylic acetoxylation of allylbenzene with BQ as the stoichiometric oxidant in the presence of base but with inconsistent yields, we performed the efficient, reproducible and regioselective allylic acyloxylation of terminal alkenes using lithium hydroxide as the additive and propionic acid as the solvent (Equation (8)) [25]. Submitting allylsilane 8 to the PdCl 2 /CuCl or BQ system under oxygen mainly led to the desilylated compound 5. (Equation (7)) [21]. Aldehyde 6 and alcohol 7 were concomitantly produced.

Terminal Alkenes
Various conditions have been reported for the synthesis of allylic carboxylates via Pd II -catalyzed allylic oxidation of olefins in carboxylic acids [23,24]. After the observation of the improvement of the Pd(OAc) 2 -catalyzed allylic acetoxylation of allylbenzene with BQ as the stoichiometric oxidant in the presence of base but with inconsistent yields, we performed the efficient, reproducible and regioselective allylic acyloxylation of terminal alkenes using lithium hydroxide as the additive and propionic acid as the solvent (Equation (8)) [25].
Catalysts 2020, 10, x FOR PEER REVIEW 5 of 22 (8) The reaction of 1-decene was less selective (Equation (8)). The regioselectivity increased with the more-hindered pivalic acid but to the detriment of the conversion. Improved results were finally obtained using a BQ/MnO2 mixture as the oxidant (Equation (9)) [25]. This modified procedure was well adapted to the oxidation of homoallylic alcohols but led to lower yields with allylarenes (compare Equations (8) and (9)). (9) Careful analysis by Electrospray Ionization Mass Spectrometry (ESI-MS) of the allylic acyloxylation under the above conditions led to identification of different clusters, especially those corresponding to intermediates having the Pd atom (in mauve color) in the proposed catalytic cycle depicted in Scheme 5. Palladium acetate, which is a trimer in the solid state [26], reacts with the insitu formed salt of the carboxylic acid to afford 5 A. Coordination of the substrate to 5 A leads to η 2alkenyl intermediate 5 B, which evolves towards an η 3 -allyl complex 5 C. Subsequent intramolecular acetoxylation delivers the product and Pd 0 . The reoxidation of Pd 0 completes the catalytic cycle.
The reaction of 1-decene was less selective (Equation (8)). The regioselectivity increased with the more-hindered pivalic acid but to the detriment of the conversion. Improved results were finally obtained using a BQ/MnO 2 mixture as the oxidant (Equation (9)) [25]. This modified procedure was well adapted to the oxidation of homoallylic alcohols but led to lower yields with allylarenes (compare Equations (8) and (9)).
Catalysts 2020, 10, x FOR PEER REVIEW 5 of 22 (8) The reaction of 1-decene was less selective (Equation (8)). The regioselectivity increased with the more-hindered pivalic acid but to the detriment of the conversion. Improved results were finally obtained using a BQ/MnO2 mixture as the oxidant (Equation (9)) [25]. This modified procedure was well adapted to the oxidation of homoallylic alcohols but led to lower yields with allylarenes (compare Equations (8) and (9)).

(9)
Careful analysis by Electrospray Ionization Mass Spectrometry (ESI-MS) of the allylic acyloxylation under the above conditions led to identification of different clusters, especially those corresponding to intermediates having the Pd atom (in mauve color) in the proposed catalytic cycle depicted in Scheme 5. Palladium acetate, which is a trimer in the solid state [26], reacts with the insitu formed salt of the carboxylic acid to afford 5 A. Coordination of the substrate to 5 A leads to η 2alkenyl intermediate 5 B, which evolves towards an η 3 -allyl complex 5 C. Subsequent intramolecular acetoxylation delivers the product and Pd 0 . The reoxidation of Pd 0 completes the catalytic cycle. (9) Careful analysis by Electrospray Ionization Mass Spectrometry (ESI-MS) of the allylic acyloxylation under the above conditions led to identification of different clusters, especially those corresponding to intermediates having the Pd atom (in mauve color) in the proposed catalytic cycle depicted in Scheme 5. Palladium acetate, which is a trimer in the solid state [26], reacts with the in-situ formed salt of the carboxylic acid to afford 5 A. Coordination of the substrate to 5 A leads to η 2 -alkenyl intermediate 5 B, which evolves towards an η 3 -allyl complex 5 C. Subsequent intramolecular acetoxylation delivers the product and Pd 0 . The reoxidation of Pd 0 completes the catalytic cycle. acyloxylation under the above conditions led to identification of different clusters, especially those corresponding to intermediates having the Pd atom (in mauve color) in the proposed catalytic cycle depicted in Scheme 5. Palladium acetate, which is a trimer in the solid state [26], reacts with the insitu formed salt of the carboxylic acid to afford 5 A. Coordination of the substrate to 5 A leads to η 2alkenyl intermediate 5 B, which evolves towards an η 3 -allyl complex 5 C. Subsequent intramolecular acetoxylation delivers the product and Pd 0 . The reoxidation of Pd 0 completes the catalytic cycle.

Alcohol Oxidation
The transformation of alcohols into the corresponding carbonyl compounds with metal oxides and metal salts may occur through three pathways (Scheme 6). Instead of the term "oxidation" used for reactions following paths a and b, those arising via path c are often called "dehydrogenation" or "oxidative dehydrogenation". Most Pd-catalyzed oxidation of alcohols occur via paths b and c [27]. As shown below, we have developed procedures for such reactions using various species to regenerate the catalyst.

Alcohol Oxidation
The transformation of alcohols into the corresponding carbonyl compounds with metal oxides and metal salts may occur through three pathways (Scheme 6). Instead of the term "oxidation" used for reactions following paths a and b, those arising via path c are often called "dehydrogenation" or "oxidative dehydrogenation". Most Pd-catalyzed oxidation of alcohols occur via paths b and c [27]. As shown below, we have developed procedures for such reactions using various species to regenerate the catalyst.

With Sodium Percarbonate
Despite its name, sodium percarbonate (SPC) is not a persalt. SPC, which is a versatile oxidizing agent for organic synthesis [28,29], is the association of sodium carbonate with hydrogen peroxide with the formula Na2CO3,1.5 H2O2. In the course of the screening of metal chlorides for the catalytic oxidation of 1-indanol (20) by SPC in 1,2-dichloroethane (DCE) in the presence of Adogen 464 [30], we discovered that the reaction with PdCl2 effectively occurred even in the absence of SPC. That led us to the procedure documented in subsection 4.2. In contrast, SPC is required with solvents such as acetonitrile, hexane and benzene, leading selectively to 1-indanone (21) from 20 (Equation (10)) [31].

With Sodium Percarbonate
Despite its name, sodium percarbonate (SPC) is not a persalt. SPC, which is a versatile oxidizing agent for organic synthesis [28,29], is the association of sodium carbonate with hydrogen peroxide with the formula Na 2 CO 3 ,1.5 H 2 O 2 . In the course of the screening of metal chlorides for the catalytic oxidation of 1-indanol (20) by SPC in 1,2-dichloroethane (DCE) in the presence of Adogen 464 [30], we discovered that the reaction with PdCl 2 effectively occurred even in the absence of SPC. That led Catalysts 2020, 10, 111 7 of 23 us to the procedure documented in Section 4.2. In contrast, SPC is required with solvents such as acetonitrile, hexane and benzene, leading selectively to 1-indanone (21) from 20 (Equation (10)) [31].

With Sodium Percarbonate
Despite its name, sodium percarbonate (SPC) is not a persalt. SPC, which is a versatile oxidizing agent for organic synthesis [28,29], is the association of sodium carbonate with hydrogen peroxide with the formula Na2CO3,1.5 H2O2. In the course of the screening of metal chlorides for the catalytic oxidation of 1-indanol (20) by SPC in 1,2-dichloroethane (DCE) in the presence of Adogen 464 [30], we discovered that the reaction with PdCl2 effectively occurred even in the absence of SPC. That led us to the procedure documented in subsection 4.2. In contrast, SPC is required with solvents such as acetonitrile, hexane and benzene, leading selectively to 1-indanone (21) from 20 (Equation (10)) [31]. (10)

With 1,2-Dichloroethane
The PdCl2-catalyzed oxidation of alcohols in DCE containing sodium carbonate and catalytic amounts of Adogen 464 (Equation (11)) involves the regeneration of active Pd species by the solvent, leading to the formation of ethylene. The formation of the latter has been highlighted by its reaction with iodine, giving 1,2-diiodooethane [32]. The method is efficient for saturated and benzylic secondary alcohols. Some overoxidation of primary alcohols occurred leading to acids which react with the solvent to afford esters. Isomerization of secondary allylic alcohols to saturated ketones may compete with the oxidation (Equation (11)). (10)

With 1,2-Dichloroethane
The PdCl 2 -catalyzed oxidation of alcohols in DCE containing sodium carbonate and catalytic amounts of Adogen 464 (Equation (11)) involves the regeneration of active Pd species by the solvent, leading to the formation of ethylene. The formation of the latter has been highlighted by its reaction with iodine, giving 1,2-diiodooethane [32]. The method is efficient for saturated and benzylic secondary alcohols. Some overoxidation of primary alcohols occurred leading to acids which react with the solvent to afford esters. Isomerization of secondary allylic alcohols to saturated ketones may compete with the oxidation (Equation (11)).
Catalysts 2020, 10, x FOR PEER REVIEW 7 of 22 (11) As the plausible mechanism, we initially suspected the insertion of Pd 0 into a C-Cl bond of DCE to afford ClCH2CH2PdCl which would undergo β-Cl elimination leading to ethylene and PdCl2 [32]. According to a theoretical study, ClCH2CH2PdCl would rather be the active species reacting with the alcohol [33]. The process is, however, carried out in the presence of Adogen 464 which reacts with PdCl2 to afford the soluble palladium salt [PdCl4] 2− [34]. These remarks led us to propose the catalytic cycle depicted in Scheme 7. Hydridopalladium 7 A formed after the first alcohol oxidation leads to Pd 0 species 7 B via elimination of HCl. Insertion of 7 B into DCE affords 7 C. Coordination of the alcohol to the latter gives 7 D. Subsequent elimination of ethylene and HCl leads to an alkoxypalladium intermediate. (11) As the plausible mechanism, we initially suspected the insertion of Pd 0 into a C-Cl bond of DCE to afford ClCH 2 CH 2 PdCl which would undergo β-Cl elimination leading to ethylene and PdCl 2 [32]. According to a theoretical study, ClCH 2 CH 2 PdCl would rather be the active species reacting with the alcohol [33]. The process is, however, carried out in the presence of Adogen 464 which reacts with PdCl 2 to afford the soluble palladium salt [PdCl 4 ] 2− [34]. These remarks led us to propose the catalytic cycle depicted in Scheme 7. Hydridopalladium 7 A formed after the first alcohol oxidation leads to Pd 0 species 7 B via elimination of HCl. Insertion of 7 B into DCE affords 7 C. Coordination of the alcohol to the latter gives 7 D. Subsequent elimination of ethylene and HCl leads to an alkoxypalladium intermediate. alcohol [33]. The process is, however, carried out in the presence of Adogen 464 which reacts with PdCl2 to afford the soluble palladium salt [PdCl4] 2− [34]. These remarks led us to propose the catalytic cycle depicted in Scheme 7. Hydridopalladium 7 A formed after the first alcohol oxidation leads to Pd 0 species 7 B via elimination of HCl. Insertion of 7 B into DCE affords 7 C. Coordination of the alcohol to the latter gives 7 D. Subsequent elimination of ethylene and HCl leads to an alkoxypalladium intermediate.  (22) which led to lactol 23 (96% selectivity) (Equation (12)) although the corresponding saturated diol 25 provided lactone 26 (Equation (13)) [35]. Oxidation of lactol 23 into lactone 24 did not occur under the Pd conditions, but arose using pyridinium dichromate in CH2Cl2 (83% yield [36]) [37] or the Swern oxidation method [38]. The PdCl2/Adogen 464/DCE procedure is however able to oxidize α lactol such as 27 (Equation (14)).  (22) which led to lactol 23 (96% selectivity) (Equation (12)) although the corresponding saturated diol 25 provided lactone 26 (Equation (13)) [35]. Oxidation of lactol 23 into lactone 24 did not occur under the Pd conditions, but arose using pyridinium dichromate in CH 2 Cl 2 (83% yield [36]) [37] or the Swern oxidation method [38]. The PdCl 2 /Adogen 464/DCE procedure is however able to oxidize α lactol such as 27 (Equation (14)).
Catalysts 2020, 10, x FOR PEER REVIEW 8 of 22 The above results clearly demonstrated that the lack of formation of 24 from 22 under Pdconditions was attributable to the C=C bond. Analysis of both the plausible intermediates of the process and literature [39][40][41] [42,43], is prevented in 8 E. That favors alkoxyde exchange with diol 22 to afford 8 F. The latter is in equilibrium with 8 B which evolves towards 23, that is, the more stable isomer [44][45][46]. The apparent absence of 23 evolution under the PdCl2/Adogen 464/DCE conditions would be a "no reaction" reaction [47,48], which involves 8 B, 8 The above results clearly demonstrated that the lack of formation of 24 from 22 under Pdconditions was attributable to the C=C bond. Analysis of both the plausible intermediates of the process and literature [39][40][41] [42,43], is prevented in 8 E. That favors alkoxyde exchange with diol 22 to afford 8 F. The latter is in equilibrium with 8 B which evolves towards 23, that is, the more stable isomer [44][45][46]. The apparent absence of 23 evolution under the PdCl2/Adogen 464/DCE conditions would be a "no reaction" reaction [47,48], which involves 8 B, 8 The above results clearly demonstrated that the lack of formation of 24 from 22 under Pdconditions was attributable to the C=C bond. Analysis of both the plausible intermediates of the process and literature [39][40][41] [42,43], is prevented in 8 E. That favors alkoxyde exchange with diol 22 to afford 8 F. The latter is in equilibrium with 8 B which evolves towards 23, that is, the more stable isomer [44][45][46]. The apparent absence of 23 evolution under the PdCl2/Adogen 464/DCE conditions would be a "no reaction" reaction [47,48], which involves 8 B, 8 The above results clearly demonstrated that the lack of formation of 24 from 22 under Pd-conditions was attributable to the C=C bond. Analysis of both the plausible intermediates of the process and literature [39][40][41]  to the carbonyl unit [42,43], is prevented in 8 E. That favors alkoxyde exchange with diol 22 to afford 8 F. The latter is in equilibrium with 8 B which evolves towards 23, that is, the more stable isomer [44][45][46]. The apparent absence of 23 evolution under the PdCl 2 /Adogen 464/DCE conditions would be a "no Catalysts 2020, 10, 111 9 of 23 reaction" reaction [47,48], which involves 8 B, 8  process and literature [39][40][41] led to the proposal of Scheme 8. The reaction of hydroxyaldehyde 8 B obtained via 8 A leads to alkoxypalladium intermediate 8 C. Subsequent intramolecular reaction occurs through the stereoselective approach of the alkoxypalladium moiety to one face of the aldehyde to afford 8 D. In contrast to the intermediate obtained from 25, 8 D undergoes a ligand exchange leading to palladacycle 8 E. The syn relationship between O-Pd and C-H bonds, which would allow a β-H elimination leading to the carbonyl unit [42,43], is prevented in 8 E. That favors alkoxyde exchange with diol 22 to afford 8 F. The latter is in equilibrium with 8 B which evolves towards 23, that is, the more stable isomer [44][45][46]. The apparent absence of 23 evolution under the PdCl2/Adogen 464/DCE conditions would be a "no reaction" reaction [47,48], which involves 8 B, 8  As depicted in Equation (11), the PdCl2/Adogen 464/DCE procedure efficiently oxidizes 1indanol (20) into 1-indanone (21). The oxidation was also effective using the soluble catalyst (n-Bu4N)2PdCl4 . 0.5 H2O (92% conversion, 90% yield) instead of the PdCl2/Adogen 464 association [31]. Surprisingly, soluble (MeCN)2PdCl2 produced di(1-indanyl) oxide (28) in high yields, even in the absence of the base (Equation (15)) [49]. Scheme 8. The "no reaction" reaction of lactol 23. (11), the PdCl 2 /Adogen 464/DCE procedure efficiently oxidizes 1-indanol (20) into 1-indanone (21). The oxidation was also effective using the soluble catalyst (n-Bu 4 N) 2 PdCl 4 . 0.5 H 2 O (92% conversion, 90% yield) instead of the PdCl 2 /Adogen 464 association [31]. Surprisingly, soluble (MeCN) 2 PdCl 2 produced di(1-indanyl) oxide (28) in high yields, even in the absence of the base (Equation (15)) [49].

As depicted in Equation
Catalysts 2020, 10, x FOR PEER REVIEW 9 of 22 (15) We rationalized the formation of 21 and 28 through two competitive pathways catalyzed with L2PdCl2 (L = Cl − or RCN), that is with [PdCl4] 2− and (RCN)2PdCl2, respectively (Scheme 9) [49]. In contrast to the anionic catalyst (L = Cl -), the neutral catalyst (L = RCN) is electrophilic [50][51][52][53]. Exchange of ligand between L2PdCl2 and 20 affords 9 A. The evolution of 9 A depends on the electrophilicity of the L2PdCl2. Transition metals having Lewis acid properties mediate the formation of ethers from alcohols [31,54]. Consequently, 9 A formed from the anionic catalyst evolves towards 21 via the ketonisation pathway [32,33], while 9 A formed from (RCN)2PdCl2 undergoes heterolytic cleavage of the C-OH bond leading to ionic species 9 B. The latter reacts with 20 to give ether 28, water and the starting catalyst.

With Aryl Bromide
The Yoshida procedure of oxidation of alcohols used a Pd catalyst with an aryl halide as hydrogen acceptor and a base (Scheme 10) [55,56]. We used this procedure for the oxidation with high yields of benzyl-protected sugar hemiacetals into lactones (Equation (16)) [57]. (15) We rationalized the formation of 21 and 28 through two competitive pathways catalyzed with L 2 PdCl 2 (L = Cl − or RCN), that is with [PdCl 4 ] 2− and (RCN) 2 PdCl 2 , respectively (Scheme 9) [49]. In contrast to the anionic catalyst (L = Cl -), the neutral catalyst (L = RCN) is electrophilic [50][51][52][53]. Exchange of ligand between L 2 PdCl 2 and 20 affords 9 A. The evolution of 9 A depends on the electrophilicity of the L 2 PdCl 2 . Transition metals having Lewis acid properties mediate the formation of ethers from alcohols [31,54]. Consequently, 9 A formed from the anionic catalyst evolves towards 21 via the ketonisation pathway [32,33], while 9 A formed from (RCN) 2 PdCl 2 undergoes heterolytic cleavage of the C-OH bond leading to ionic species 9 B. The latter reacts with 20 to give ether 28, water and the starting catalyst.
Catalysts 2020, 10, x FOR PEER REVIEW 9 of 22 (15) We rationalized the formation of 21 and 28 through two competitive pathways catalyzed with L2PdCl2 (L = Cl − or RCN), that is with [PdCl4] 2− and (RCN)2PdCl2, respectively (Scheme 9) [49]. In contrast to the anionic catalyst (L = Cl -), the neutral catalyst (L = RCN) is electrophilic [50][51][52][53]. Exchange of ligand between L2PdCl2 and 20 affords 9 A. The evolution of 9 A depends on the electrophilicity of the L2PdCl2. Transition metals having Lewis acid properties mediate the formation of ethers from alcohols [31,54]. Consequently, 9 A formed from the anionic catalyst evolves towards 21 via the ketonisation pathway [32,33], while 9 A formed from (RCN)2PdCl2 undergoes heterolytic cleavage of the C-OH bond leading to ionic species 9 B. The latter reacts with 20 to give ether 28, water and the starting catalyst.

With Aryl Bromide
The Yoshida procedure of oxidation of alcohols used a Pd catalyst with an aryl halide as hydrogen acceptor and a base (Scheme 10) [55,56]. We used this procedure for the oxidation with high yields of benzyl-protected sugar hemiacetals into lactones (Equation (16)) [57].

Scheme 9.
Dependence of the reaction pathway on the electrophilicity of the catalyst.

With Aryl Bromide
The Yoshida procedure of oxidation of alcohols used a Pd catalyst with an aryl halide as hydrogen acceptor and a base (Scheme 10) [55,56]. We used this procedure for the oxidation with high yields of benzyl-protected sugar hemiacetals into lactones (Equation (16)) [57].

With Aryl Bromide
The Yoshida procedure of oxidation of alcohols used a Pd catalyst with an aryl halide as hydrogen acceptor and a base (Scheme 10) [55,56]. We used this procedure for the oxidation with high yields of benzyl-protected sugar hemiacetals into lactones (Equation (16)) [57].

Dehydrogenation
Over the last thirty years, growing attention has been devoted to the use of ionic liquids and molten salts as solvents for organic synthesis [58][59][60]. The strong interest for catalyzed reactions is due to the immobilization of the catalyst in the ionic liquid or molten salt that would allow the recycling of the tandem catalyst/solvent. These unusual solvents have been used for various catalytic oxidations [61][62][63]. Our above studies with (n-Bu4N)2PdCl4 . 0.5 H2O (see section 4.2) and Heck reaction in molten n-Bu4NBr [64] urged us to carry out Pd-catalyzed oxidations in this medium.
Initial experimentation using 20, catalytic PdCl2 and n-Bu4NBr at 120 °C led to a mixture of 21 and indane. The reductive cleavage of the C-OH bond of 19 indicated in-situ formation of hydrogen [65] and/or [Pd]H2 species [66]. Addition of cyclohexene as a hydrogen acceptor increased the selectivity towards 21. Finally, the best result was obtained under a gentle flow of argon which removes hydrogen gas (Equation (17)) [67]. Under these conditions, the recycling of both catalyst and n-Bu4NBr was relatively efficient. Secondary benzylic alcohols provided the corresponding ketones in good yields. The method is less selective for primary benzylic alcohols and is ineffective from allylic and saturated alcohols. Subsequent experiments showed that recycling was more efficient with Pd(OAc)2 than with PdCl2, the yield obtained from 4th reuse of the catalyst/n-Bu4NBr association being 92% (94% conversion) with the former and 64% (67% conversion) with the latter [68]. (17) (16)

Dehydrogenation
Over the last thirty years, growing attention has been devoted to the use of ionic liquids and molten salts as solvents for organic synthesis [58][59][60]. The strong interest for catalyzed reactions is due to the immobilization of the catalyst in the ionic liquid or molten salt that would allow the recycling of the tandem catalyst/solvent. These unusual solvents have been used for various catalytic oxidations [61][62][63]. Our above studies with (n-Bu 4 N) 2 PdCl 4 . 0.5 H 2 O (see Section 4.2) and Heck reaction in molten n-Bu 4 NBr [64] urged us to carry out Pd-catalyzed oxidations in this medium. Initial experimentation using 20, catalytic PdCl 2 and n-Bu 4 NBr at 120 • C led to a mixture of 21 and indane. The reductive cleavage of the C-OH bond of 19 indicated in-situ formation of hydrogen [65] and/or [Pd]H 2 species [66]. Addition of cyclohexene as a hydrogen acceptor increased the selectivity towards 21. Finally, the best result was obtained under a gentle flow of argon which removes hydrogen gas (Equation (17)) [67]. Under these conditions, the recycling of both catalyst and n-Bu 4 NBr was relatively efficient. Secondary benzylic alcohols provided the corresponding ketones in good yields. The method is less selective for primary benzylic alcohols and is ineffective from allylic and saturated alcohols. Subsequent experiments showed that recycling was more efficient with Pd(OAc) 2 than with PdCl 2 , the yield obtained from 4th reuse of the catalyst/n-Bu 4 NBr association being 92% (94% conversion) with the former and 64% (67% conversion) with the latter [68]. and indane. The reductive cleavage of the C-OH bond of 19 indicated in-situ formation of hydrogen [65] and/or [Pd]H2 species [66]. Addition of cyclohexene as a hydrogen acceptor increased the selectivity towards 21. Finally, the best result was obtained under a gentle flow of argon which removes hydrogen gas (Equation (17)) [67]. Under these conditions, the recycling of both catalyst and n-Bu4NBr was relatively efficient. Secondary benzylic alcohols provided the corresponding ketones in good yields. The method is less selective for primary benzylic alcohols and is ineffective from allylic and saturated alcohols. Subsequent experiments showed that recycling was more efficient with Pd(OAc)2 than with PdCl2, the yield obtained from 4th reuse of the catalyst/n-Bu4NBr association being 92% (94% conversion) with the former and 64% (67% conversion) with the latter [68]. (17)
Catalysts 2020, 10, x FOR PEER REVIEW 11 of 22 (18) The report of Jacobs' team about the phenol-mediated epoxidation of alkenes by H2O2 under metal-free conditions [81] led us to propose in 2005 the mechanism depicted in Scheme 11 [80]. Pdcatalyzed isomerization of 29 affords 32 [82]. Activation by the phenolic OH of the epoxidation of 32 provides 11 A. The high instability of such a compound [83] brings on spontaneous ring opening leading to 30 and 31R. However, the strong acceleration of the palladium-catalyzed reaction of 32 (Equation (19)) indicates some participation of Pd(OCOCF3)2/LH in the process [80]. In fact, epoxides are very sensitive to Pd catalysis [84]. Complementary mechanistic experiments and ESI-MS studies supported the proposed reaction pathway [85]. Scheme 11. Domino reaction of 2-allylphenol. (18) The report of Jacobs' team about the phenol-mediated epoxidation of alkenes by H 2 O 2 under metal-free conditions [81] led us to propose in 2005 the mechanism depicted in Scheme 11 [80]. Pd-catalyzed isomerization of 29 affords 32 [82]. Activation by the phenolic OH of the epoxidation of 32 provides 11 A. The high instability of such a compound [83] brings on spontaneous ring opening leading to 30 and 31 R . However, the strong acceleration of the palladium-catalyzed reaction of 32 (Equation (19)) indicates some participation of Pd(OCOCF 3 ) 2 /L H in the process [80]. In fact, epoxides are very sensitive to Pd catalysis [84]. Complementary mechanistic experiments and ESI-MS studies supported the proposed reaction pathway [85]. catalyzed isomerization of 29 affords 32 [82]. Activation by the phenolic OH of the epoxidation of 32 provides 11 A. The high instability of such a compound [83] brings on spontaneous ring opening leading to 30 and 31R. However, the strong acceleration of the palladium-catalyzed reaction of 32 (Equation (19)) indicates some participation of Pd(OCOCF3)2/LH in the process [80]. In fact, epoxides are very sensitive to Pd catalysis [84]. Complementary mechanistic experiments and ESI-MS studies supported the proposed reaction pathway [85]. Scheme 11. Domino reaction of 2-allylphenol. (19) Recycling of the Pd(OCOCF3)2/LH catalytic system led to gradual loss of activity, the 4th reuse in H2O/MeOH yielding 35% and 24% of 30 and 31Me, respectively, from 29 [85].

Wacker Oxidation
Tsuji's conditions of the Wacker reaction use PdCl2 catalyst in DMF/H2O and oxidants such as copper salts/O2 or benzoquinone [86]. According to investigations through isotope effects, kinetic, stereochemical and theoretical studies [87][88][89][90], the mechanism involves alkene coordination to PdCl2, followed by hydroxypalladation and β-hydride elimination leading to a palladium-enol η 2 -complex. The latter evolves towards the ketone, liberating HCl and Pd 0 . The catalyst is regenerated from Pd 0 through reaction with BQ. Mechanistic details remain however matter of debate [87].
Our interest in the Pd-catalyzed oxidations and in the synthetic properties of DMF [91][92][93] and BQ [94] urged us to investigate the Wacker reaction of terminal alkenes 33 with ESI-MS, using BQ as the terminal oxidant (Equation (20)) [95]. The report of Jacobs' team about the phenol-mediated epoxidation of alkenes by H2O2 under metal-free conditions [81] led us to propose in 2005 the mechanism depicted in Scheme 11 [80]. Pdcatalyzed isomerization of 29 affords 32 [82]. Activation by the phenolic OH of the epoxidation of 32 provides 11 A. The high instability of such a compound [83] brings on spontaneous ring opening leading to 30 and 31R. However, the strong acceleration of the palladium-catalyzed reaction of 32 (Equation (19)) indicates some participation of Pd(OCOCF3)2/LH in the process [80]. In fact, epoxides are very sensitive to Pd catalysis [84]. Complementary mechanistic experiments and ESI-MS studies supported the proposed reaction pathway [85]. Scheme 11. Domino reaction of 2-allylphenol. (19) Recycling of the Pd(OCOCF3)2/LH catalytic system led to gradual loss of activity, the 4th reuse in H2O/MeOH yielding 35% and 24% of 30 and 31Me, respectively, from 29 [85].

Wacker Oxidation
Tsuji's conditions of the Wacker reaction use PdCl2 catalyst in DMF/H2O and oxidants such as copper salts/O2 or benzoquinone [86]. According to investigations through isotope effects, kinetic, stereochemical and theoretical studies [87][88][89][90], the mechanism involves alkene coordination to PdCl2, followed by hydroxypalladation and β-hydride elimination leading to a palladium-enol η 2 -complex. The latter evolves towards the ketone, liberating HCl and Pd 0 . The catalyst is regenerated from Pd 0 through reaction with BQ. Mechanistic details remain however matter of debate [87].

Wacker Oxidation
Tsuji's conditions of the Wacker reaction use PdCl 2 catalyst in DMF/H 2 O and oxidants such as copper salts/O 2 or benzoquinone [86]. According to investigations through isotope effects, kinetic, stereochemical and theoretical studies [87][88][89][90], the mechanism involves alkene coordination to PdCl 2 , followed by hydroxypalladation and β-hydride elimination leading to a palladium-enol η 2 -complex. The latter evolves towards the ketone, liberating HCl and Pd 0 . The catalyst is regenerated from Pd 0 through reaction with BQ. Mechanistic details remain however matter of debate [87].
Our interest in the Pd-catalyzed oxidations and in the synthetic properties of DMF [91][92][93] and BQ [94] urged us to investigate the Wacker reaction of terminal alkenes 33 with ESI-MS, using BQ as the terminal oxidant (Equation (20)) [95].
Catalysts 2020, 10, x FOR PEER REVIEW 12 of 22 (20) ESI-MS monitoring of the reactions showed that dinuclear palladium complexes were more involved as active catalytic intermediates than mononuclear species. Tests of complexation suggested a reoxidation of Pd occurring before the decoordination of the product. These studies associated to kinetic experiments led us to propose the catalytic cycle depicted in Scheme 12, in which intermediates having the Pd atom in mauve color correspond to clusters detected by ESI-MS. (20) ESI-MS monitoring of the reactions showed that dinuclear palladium complexes were more involved as active catalytic intermediates than mononuclear species. Tests of complexation suggested a reoxidation of Pd occurring before the decoordination of the product. These studies associated to kinetic experiments led us to propose the catalytic cycle depicted in Scheme 12, in which intermediates having the Pd atom in mauve color correspond to clusters detected by ESI-MS.
ESI-MS monitoring of the reactions showed that dinuclear palladium complexes were more involved as active catalytic intermediates than mononuclear species. Tests of complexation suggested a reoxidation of Pd occurring before the decoordination of the product. These studies associated to kinetic experiments led us to propose the catalytic cycle depicted in Scheme 12, in which intermediates having the Pd atom in mauve color correspond to clusters detected by ESI-MS. Compared with the usual mechanism, the most important differences are the involvement of mainly dinuclear Pd species and the reoxidation of a Pd-hydride complex before decomplexation of the ketone.

Cyclohexanone Dehydrogenation
In 1982, we disclosed the room temperature Pd(OCOCF3)2-catalyzed dehydrogenation of cyclohexanones under oxygen atmosphere (Equation (21)) [96]. Pd procedures were previously reported but using mainly stoichiometric amounts of Pd II [97]. Compared with the usual mechanism, the most important differences are the involvement of mainly dinuclear Pd species and the reoxidation of a Pd-hydride complex before decomplexation of the ketone.
Catalysts 2020, 10, x FOR PEER REVIEW 13 of 22 (21) Cyclohexenone was selectively produced at low conversion. Increase of the latter led to overoxidation giving phenol. The proposed catalytic cycle (Scheme 13) maintains the formal oxidation state of Pd II throughout the reaction. Coordination of cyclohexanone or its enol form to Pd(OCOCF3)2 provides 13 A, which led to oxo-η 3 -allyl palladium complex 13 B in liberating CF3CO2H. Hydrogen abstraction by palladium provides 13 C which undergoes insertion of oxygen and ligand exchange giving 2-cyclohexenone and hydroperoxy complex 13 D. The latter leads to H2O2 and either 13 B (path a) or 13 A by reacting with CF3CO2H (path b). (21) Cyclohexenone was selectively produced at low conversion. Increase of the latter led to over-oxidation giving phenol. The proposed catalytic cycle (Scheme 13) maintains the formal oxidation state of Pd II throughout the reaction. Coordination of cyclohexanone or its enol form to Pd(OCOCF 3 ) 2 provides 13 A, which led to oxo-η 3 -allyl palladium complex 13 B in liberating CF 3 CO 2 H. Hydrogen abstraction by palladium provides 13 C which undergoes insertion of oxygen and ligand exchange giving 2-cyclohexenone and hydroperoxy complex 13 D. The latter leads to H 2 O 2 and either 13 B (path a) or 13 A by reacting with CF 3 CO 2 H (path b). oxidation giving phenol. The proposed catalytic cycle (Scheme 13) maintains the formal oxidation state of Pd II throughout the reaction. Coordination of cyclohexanone or its enol form to Pd(OCOCF3)2 provides 13 A, which led to oxo-η 3 -allyl palladium complex 13 B in liberating CF3CO2H. Hydrogen abstraction by palladium provides 13 C which undergoes insertion of oxygen and ligand exchange giving 2-cyclohexenone and hydroperoxy complex 13 D. The latter leads to H2O2 and either 13 B (path a) or 13 A by reacting with CF3CO2H (path b). This oxidation process was, in 1982, one of the first reports on the regeneration of active Pd II species using only oxygen [98][99][100]. The pathway leading to PdOOH from PdH and O2, that is 13 D from 13 C remains however a matter of debate [98][99][100][101].
A number of procedures are now available for the Pd-catalyzed dehydrogenation of carbonyl compounds [102]. Moreover, such a dehydrogenation may be a step of a domino reaction involving the Heck reaction, decarboxylative Heck reaction or dehydrogenative Heck reaction [103].

Dehydrogenative Heck Reaction
Disclosed in 1970, the Heck reaction is traditionally the synthesis of an arylalkene from the Pd 0catalyzed cross-coupling of an aryl halide with an alkene [104]. Previously, Fujiwara's team reported the synthesis of stilbene from the reaction of benzene with styrene and PdCl2, leading to two turnovers of palladium [105]. Such a cross-coupling, for which we adopted the name "dehydrogenative Heck reaction" (DHR) [106], may be more respective of the atom economic principle [107], and has been intensively studied over the last twenty years [108].
Our studies focused on the coupling of furan 34 with styrene. Screening various experimental conditions initially led to the best results with catalytic Pd(OAc)2 and the BQ/Cu(OAc)2/O2 oxidizing system (Equation (22)). The method was used for the DHR of various furans and styrenes with high regio-and stereoselectivities (30 examples, 50%-78% yields) [109]. This oxidation process was, in 1982, one of the first reports on the regeneration of active Pd II species using only oxygen [98][99][100]. The pathway leading to PdOOH from PdH and O 2 , that is 13 D from 13 C remains however a matter of debate [98][99][100][101].
A number of procedures are now available for the Pd-catalyzed dehydrogenation of carbonyl compounds [102]. Moreover, such a dehydrogenation may be a step of a domino reaction involving the Heck reaction, decarboxylative Heck reaction or dehydrogenative Heck reaction [103].

Dehydrogenative Heck Reaction
Disclosed in 1970, the Heck reaction is traditionally the synthesis of an arylalkene from the Pd 0 -catalyzed cross-coupling of an aryl halide with an alkene [104]. Previously, Fujiwara's team reported the synthesis of stilbene from the reaction of benzene with styrene and PdCl 2 , leading to two turnovers of palladium [105]. Such a cross-coupling, for which we adopted the name "dehydrogenative Heck reaction" (DHR) [106], may be more respective of the atom economic principle [107], and has been intensively studied over the last twenty years [108].
Our studies focused on the coupling of furan 34 with styrene. Screening various experimental conditions initially led to the best results with catalytic Pd(OAc) 2 and the BQ/Cu(OAc) 2 /O 2 oxidizing system (Equation (22)). The method was used for the DHR of various furans and styrenes with high regio-and stereoselectivities (30 examples, 50%-78% yields) [109].
Kinetic investigations showed an induction period which depends on the nature of the furan, the transformation being faster with electron-rich furans. This led us to investigate the influence of ligands and solvents on the activity of the catalyst. The use of DMSO/AcOH as solvent mixture and BQ as oxidant led to a catalytic system showing no induction period at room temperature, leading to cross-coupling of furans and thiophenes with styrenes, and compatible with halogenated substrates such as 35 and 36 (Equation (23)) [110]. The positive influence of DMSO on the efficiency of Pd(OAc)2catalyzed oxidations is largely documented in the literature [111,112]. Ligation of DMSO to the trimer [Pd(OAc)2]3 affords the dimeric active species Pd(OAc)2(DMSO)2 [113,114], which may react with the arene leading to ArPdOAc(DMSO)2. Subsequent ligand exchange with BQ could give the less electron-rich species ArPdOAc(BQ)(DMSO), which would be susceptible to easily coordinate to electron-rich styrenes [110]. (22) Kinetic investigations showed an induction period which depends on the nature of the furan, the transformation being faster with electron-rich furans. This led us to investigate the influence of ligands and solvents on the activity of the catalyst. The use of DMSO/AcOH as solvent mixture and BQ as oxidant led to a catalytic system showing no induction period at room temperature, leading to cross-coupling of furans and thiophenes with styrenes, and compatible with halogenated substrates such as 35 and 36 (Equation (23)) [110]. The positive influence of DMSO on the efficiency of Pd(OAc) 2 -catalyzed oxidations is largely documented in the literature [111,112]. Ligation of DMSO to the trimer [Pd(OAc) 2 ] 3 affords the dimeric active species Pd(OAc) 2 (DMSO) 2 [113,114], which may react with the arene leading to ArPdOAc(DMSO) 2 . Subsequent ligand exchange with BQ could give the less electron-rich species ArPdOAc(BQ)(DMSO), which would be susceptible to easily coordinate to electron-rich styrenes [110].
ligands and solvents on the activity of the catalyst. The use of DMSO/AcOH as solvent mixture and BQ as oxidant led to a catalytic system showing no induction period at room temperature, leading to cross-coupling of furans and thiophenes with styrenes, and compatible with halogenated substrates such as 35 and 36 (Equation (23)) [110]. The positive influence of DMSO on the efficiency of Pd(OAc)2catalyzed oxidations is largely documented in the literature [111,112]. Ligation of DMSO to the trimer [Pd(OAc)2]3 affords the dimeric active species Pd(OAc)2(DMSO)2 [113,114], which may react with the arene leading to ArPdOAc(DMSO)2. Subsequent ligand exchange with BQ could give the less electron-rich species ArPdOAc(BQ)(DMSO), which would be susceptible to easily coordinate to electron-rich styrenes [110]. (23) The mechanism of the reaction of 34 with tert-butyl acrylate under these conditions (Equation (24)) has been investigated by ESI-MS leading to identification of clusters corresponding especially to Pd II and Pd 0 intermediates 14 A, 14 B and 14 C, leading us to propose the catalytic cycle shown Scheme 14 [115].
The mechanism of the reaction of 34 with tert-butyl acrylate under these conditions (Equation (24)) has been investigated by ESI-MS leading to identification of clusters corresponding especially to Pd II and Pd 0 intermediates 14 A, 14 B and 14 C, leading us to propose the catalytic cycle shown Scheme 14 [115].
ligands and solvents on the activity of the catalyst. The use of DMSO/AcOH as solvent mixture and BQ as oxidant led to a catalytic system showing no induction period at room temperature, leading to cross-coupling of furans and thiophenes with styrenes, and compatible with halogenated substrates such as 35 and 36 (Equation (23)) [110]. The positive influence of DMSO on the efficiency of Pd(OAc)2catalyzed oxidations is largely documented in the literature [111,112]. Ligation of DMSO to the trimer [Pd(OAc)2]3 affords the dimeric active species Pd(OAc)2(DMSO)2 [113,114], which may react with the arene leading to ArPdOAc(DMSO)2. Subsequent ligand exchange with BQ could give the less electron-rich species ArPdOAc(BQ)(DMSO), which would be susceptible to easily coordinate to electron-rich styrenes [110]. (23) The mechanism of the reaction of 34 with tert-butyl acrylate under these conditions (Equation (24)) has been investigated by ESI-MS leading to identification of clusters corresponding especially to Pd II and Pd 0 intermediates 14 A, 14 B and 14 C, leading us to propose the catalytic cycle shown Scheme 14 [115]. Catalytic amounts of metallic co-oxidants are often required for efficient DHRs under oxygen [108]. We observed, however, that such additives are not always beneficial. Indeed, the room temperature Pd(OAc)2-catalyzed reaction of 34 with styrene in oxygenated DMSO/AcOH afforded the cross-coupling product in higher yield in their absence (Equation (25) Catalytic amounts of metallic co-oxidants are often required for efficient DHRs under oxygen [108]. We observed, however, that such additives are not always beneficial. Indeed, the room temperature Pd(OAc) 2 -catalyzed reaction of 34 with styrene in oxygenated DMSO/AcOH afforded the cross-coupling product in higher yield in their absence (Equation (25) The reaction of thiophene 37 with a hindered alkene such as methyl cinnamate under conditions of Equation (24) occurred with 5% conversion leading to traces of DHR product 38 [117]. With AcOH as the solvent instead of the AcOH/DMSO mixture, the conversion increased to 20%. Testing various ligands led to an efficient DHR with 4,5-diazafluorenone (Equation (26)). Moreover, increase of the temperature to 60 °C with O2 instead of BQ led to 38 in 90% isolated yield. Consequently, these conditions have been used for the cross-coupling of furans and thiophenes with various hindered alkenes (19 examples, 51%-96% yield). According to kinetics and competitive experiments as well as (25) The reaction of thiophene 37 with a hindered alkene such as methyl cinnamate under conditions of Equation (24) occurred with 5% conversion leading to traces of DHR product 38 [117]. With AcOH as the solvent instead of the AcOH/DMSO mixture, the conversion increased to 20%. Testing various ligands led to an efficient DHR with 4,5-diazafluorenone (Equation (26)). Moreover, increase of the temperature to 60 • C with O 2 instead of BQ led to 38 in 90% isolated yield. Consequently, these conditions have been used for the cross-coupling of furans and thiophenes with various hindered alkenes (19 examples, 51%-96% yield). According to kinetics and competitive experiments as well as ESI-MS studies, 4,5-diazafluorenone influences the C-H bond activation, the alkene insertion, the stereoselectivity and the regeneration of the catalyst [117].
Under conditions of Equation (24) except the presence of DMSO, a low yield of the DHR product 39 was obtained from the sluggish reaction of 34 with allylbenzene (Equation (27)) [118]. Surprisingly, addition of MeCN as a co-solvent suppressed the formation of 39 and increased the conversion leading to a 1:1 mixture of difurylalkanes 40 and 41. Moreover, switching to Pd(OCOCF3)2 as the catalyst increased the 40 + 41 yield to 94% (Equation (27)). (27) Suspecting the formation of 41 from prop-1-en-1-ylbenzene via isomerization of allylbenzene [82], the reaction was repeated with styrene [118]. Difurylalkanes were also obtained (Equation (28)). Labelling experiments led to assignment of the hydrogen shifts shown in Equation (28) and, (26) Under conditions of Equation (24) except the presence of DMSO, a low yield of the DHR product 39 was obtained from the sluggish reaction of 34 with allylbenzene (Equation (27)) [118]. Surprisingly, addition of MeCN as a co-solvent suppressed the formation of 39 and increased the conversion leading to a 1:1 mixture of difurylalkanes 40 and 41. Moreover, switching to Pd(OCOCF 3 ) 2 as the catalyst increased the 40 + 41 yield to 94% (Equation (27)).
Under conditions of Equation (24) except the presence of DMSO, a low yield of the DHR product 39 was obtained from the sluggish reaction of 34 with allylbenzene (Equation (27)) [118]. Surprisingly, addition of MeCN as a co-solvent suppressed the formation of 39 and increased the conversion leading to a 1:1 mixture of difurylalkanes 40 and 41. Moreover, switching to Pd(OCOCF3)2 as the catalyst increased the 40 + 41 yield to 94% (Equation (27)). (27) Suspecting the formation of 41 from prop-1-en-1-ylbenzene via isomerization of allylbenzene [82], the reaction was repeated with styrene [118]. Difurylalkanes were also obtained (Equation (28)). Labelling experiments led to assignment of the hydrogen shifts shown in Equation (28) Suspecting the formation of 41 from prop-1-en-1-ylbenzene via isomerization of allylbenzene [82], the reaction was repeated with styrene [118]. Difurylalkanes were also obtained (Equation (28)). Labelling experiments led to assignment of the hydrogen shifts shown in Equation (28) and, associated ESI-MS studies, to propose the catalytic cycle depicted in Scheme 15 [118]. (26) Under conditions of Equation (24) except the presence of DMSO, a low yield of the DHR product 39 was obtained from the sluggish reaction of 34 with allylbenzene (Equation (27)) [118]. Surprisingly, addition of MeCN as a co-solvent suppressed the formation of 39 and increased the conversion leading to a 1:1 mixture of difurylalkanes 40 and 41. Moreover, switching to Pd(OCOCF3)2 as the catalyst increased the 40 + 41 yield to 94% (Equation (27)). (27) Suspecting the formation of 41 from prop-1-en-1-ylbenzene via isomerization of allylbenzene [82], the reaction was repeated with styrene [118]. Difurylalkanes were also obtained (Equation (28)). Labelling experiments led to assignment of the hydrogen shifts shown in Equation (28)

Conclusions
Starting at the end of the 1970s with oxidations of CH or CH2 units under UV light, our research evolved towards light-free reactions in various media: organic solvents, water, molten salts, leading to alcohol oxidation, dehydrogenation, etherification or formation of C-C bonds. We have always been strongly focused on the mechanisms; these lead us to various proposals, especially those based on ESI-MS results. Some reactions and mechanisms have been serendipitously discovered [119], but nevertheless, in most cases, they were the fruit of deep investigations and reflection, rather than good fortune.

Conclusions
Starting at the end of the 1970s with oxidations of CH or CH 2 units under UV light, our research evolved towards light-free reactions in various media: organic solvents, water, molten salts, leading to alcohol oxidation, dehydrogenation, etherification or formation of C-C bonds. We have always been strongly focused on the mechanisms; these lead us to various proposals, especially those based on ESI-MS results. Some reactions and mechanisms have been serendipitously discovered [119], but nevertheless, in most cases, they were the fruit of deep investigations and reflection, rather than good fortune.
Funding: This research received no external funding.