Carbon-Carbon Bond-Forming Reactions of Carboxylic Acid Derivatives via Decarbonylation and Decarboxylation

Carbon-Carbon Bond-Forming Reactions of Carboxylic Acid Derivatives

via Decarbonylation and Decarboxylation

(

2021.3

Liyan FU

Graduate School of Natural Science and Technology (Doctor’s Course)

OKAYAMA UNIVERSITY

i

Contents

Contents i~ii

Abstract iii~v

Chapter 1 General Introduction

1-1 Introduction 2

1-2 Carbon-Carbon Bond-Forming Reactions of Acyl Halides 3

1-2-1 Acyl Couplings of Acyl Fluorides 3

1-2-2 Decarbonylative Couplings of Acyl Fluorides 9

1-2-3 Decarbonylative Cross-Couplings of Acyl Chlorides 12

1-3 Decarboxylative Cross-Coupling Reactions for Carbon–Carbon Bond Formation 14

1-3-1 General Introduction 14

1-3-2 Decarboxylative Cross-Couplings of Perfluorobenzoates 15

1-4 Summary 19

References 20

Chapter 2 Palladium-Catalyzed Decarbonylative Alkylation of Acyl Fluorides

2-1 Introduction 24

2-2 Results and Discussion 25

2-2-1 Optimization of the Reaction Conditions 25

2-2-2 Scope of Acyl Fluorides 31

2-2-3 Scope of Alkylboranes 34

2-2-4 Applications 35

2-2-5 Mechanistic Studies 36

2-3 Summary 39

2-4 Experimental Section 40

2-4-1 General Instrumentation and Chemicals 40

2-4-2 Experimental Procedures 41

2-4-3 Copies of NMR Charts for Products 57

References 90

Chapter 3 Decarboxylative Cross-Coupling of Acyl Fluorides with Potassium Perfluorobenzoates

3-1 Introduction 94

ii

3-2 Results and Discussion 95

3-2-1 Optimization of the Reaction Conditions 95

3-2-2 Scope of Acyl Fluorides 98

3-2-3 Scope of Perfluorobenzoates 100

3-2-4 Mechanistic Studies 101

3-3 Summary 106

3-4 Experimental Section 107

3-4-1 General Instrumentation and Chemicals 107

3-4-2 Experimental Procedures 108

3-4-3 Copies of NMR Charts for Products 121

References 158

Chapter 4 Palladium-Catalyzed Decarbonylative and Decarboxylative

Cross-Coupling of Acyl Chlorides with Potassium Perfluorobenzoates

4-1 Introduction 162

4-2 Results and Discussion 163

4-2-1 Optimization of the Reaction Conditions 163

4-2-2 Scope of Acyl Chlorides 166

4-2-3 Scope of Perfluorobenzoates 167

4-2-4 Mechanistic Studies 169

4-3 Summary 171

4-4 Experimental Section 172

4-4-1 General Instrumentation and Chemicals 172

4-4-2 Experimental Procedures 173

4-4-3 Copies of NMR Charts for Products 181

References 196

Conclusion and Future Perspective

List of Publications

Presentations

Acknowledgements

iii

Abstract

Carbon-carbon bond-forming reactions are one of the most important parts in the field of chemistry.

However, conventional cross-coupling reactions for this purpose are severely limited to precious transition metal catalysts, toxic organic (pseudo)halides, and expensive organometallic reagents. In this context, carboxylic acids and their derivatives are considered as versatile surrogates for both organometallic reagents and organic (pseudo)halides due to their abundance, ready availability, and environmentally friendliness.

In this PhD Thesis, the Author focuses on carbon-carbon bond formation via decarbonylative and decarboxylative cross-couplings of carboxylic acid derivatives. Firstly, acyl fluorides were utilized as electrophiles to form C(sp²)-C(sp³) bonds with alkylboranes via palladium-catalyzed decarbonylative cross-coupling reaction. Next, a variety of unsymmetrical ketones were synthesized from acyl fluorides and potassium perfluorobenzoates by a catalyst- and additive-free decarboxylative coupling. Finally, perfluorinated biaryls were obtained through a palladium-catalyzed cross- coupling reaction of acyl chlorides with potassium perfluorobenzoates by means of decarbonylation and decarboxylation. This PhD Thesis presents an economical and environmentally friendly approach to construct various carbon-carbon bonds via decarbonylative and decarboxylative cross- couplings, a new protocol for employing carboxylic acid derivatives as surrogates for conventional electrophilic and nucleophilic reagents.

Chapter 2. Palladium-Catalyzed Decarbonylative Alkylation of Acyl Fluorides

The formation of C(sp²)-C(sp³) bond is barely reported because of the unavoidable β-hydride elimination. Therefore, the development of new synthetic methods for C(sp²)-C(sp³) bond is still in high demand. In this Chapter, the Author develops a palladium-catalyzed decarbonylative cross- coupling of acyl fluorides and alkylboranes containing β-hydrogens, which furnishes a highly efficient decarbonylative route for the formation of a variety of C(sp²)-C(sp³) bonds. Although the decarbonylation process requires a high reaction temperature, this strategy enables a diverse of decarbonylation reactions and improves the tolerance of functional groups. The utilization of palladium and bidentate ligand catalytic system avoids the undesired β-hydride elimination.

Chapter 3. Decarboxylative Cross-Coupling of Acyl Fluorides with Potassium

iv

Perfluorobenzoates

Benzoic acids are considered to be ideal aryl nucleophiles via decarboxylation process because of their ready availability and environmental friendliness. However, precious transition metal catalysts, harsh reaction conditions, toxic organic (pseudo)halides, and a limited substrate scope severely limit its application. Hence, decarboxylative cross-coupling reactions of benzoic acids and their derivatives in a transition-metal-free manner are highly desirable. In this Chapter, the Author describes a transition-metal-free formation of unsymmetrical ketones from acyl fluorides with potassium perfluorobenzoates via a decarboxylative cross-coupling reaction. Compared to conventional transition-metal-catalyzed cross-couplings, this reaction presents an extremely environmentally benign pathway to afford perfluorinated biaryl ketones. For the installation of perfluorophenyl groups, this method highlights highly selective, inexpensive, and non-toxic conditions. The reaction system tolerates various functional groups of acyl fluorides. Remarkably, all the starting materials can be prepared from abundant carboxylic acids and the reaction proceeds without catalysts or additives. Mechanistic studies suggest the transformation proceeds via an aromatic potassium intermediate rather than an anhydride intermediate.

Chapter 4. Palladium-Catalyzed Decarbonylative and Decarboxylative Cross- Coupling of Acyl Chlorides with Potassium Benzoates

Benzoic acid derivatives and benzoates have been developed as convenient surrogates to conventional coupling partners through a loss of CO or CO2, known as decarbonylative and decarboxylative couplings. These methods avoid the use of either toxic organic halides or expensive organometallic reagents, but the employment of either is still necessary. In this context, synthetic protocols for unsymmetrical biaryls obtained from both reagents available from benzoic acids have been of great interest. In this Chapter, the Author discloses the palladium-catalyzed decarbonylative and decarboxylative cross-coupling of acyl chlorides with potassium perfluorobenzoates affording unsymmetrical biaryls. This transformation involves a unique reaction in which the decarbonylation and decarboxylation proceeds simultaneously under redox-neutral conditions. In comparison of conventional cross-coupling protocols for the synthesis of unsymmetrical biaryls, both reactants can

v

be readily prepared from abundant and inexpensive aromatic carboxylic acids in this synthetic strategy.

Chapter 1

General Introduction

2

1-1 Introduction

The construction of carbon-carbon bonds is one of the most important reactions in the field of chemistry.¹ In 2010, Nobel Prize in Chemistry was awarded to Akira Suzuki, Ei-ichi Negishi, and Richard F. Heck for their “palladium-catalyzed cross-couplings in organic synthesis,” which are considered to be extremely powerful tools for carbon-carbon bond formation.² However, problems such as the utilization of toxic halides, handling of unstable and expensive organometallic reagents, and generation of halide-containing waste have limited their application.³ In this context, synthetic protocols that are available, sustainable, and environmentally friendly are highly desired. Carboxylic acids in diverse forms, amides, esters, etc. are readily available from both natural and synthetic sources and are therefore inexpensive, stable, and non-toxic. Hence, carboxylic acids and their derivatives are considered as one of the most suitable starting materials for synthesis.⁴ Consequently, benzoic acids and their derivatives have been developed as versatile surrogates for organometallic reagents (nucleophiles) and organic (pseudo)halides (electrophiles) via decarbonylation and decarboxylation, in which carboxy and carbonyl groups become traceless leaving group with high chemo- and regio- selectivity (Figure 1-1).^4-6

Figure 1-1. Comparison of Decarbonylation and Decarboxylation.

Acyl halides are one of the simplest carboxylic acid derivatives and can be readily prepared from related carboxylic acids and other carboxylic acid derivatives.^1,7 Due to the special C(acyl)-X bonds, acyl halides can act as acyl, aryl, and halides sources in a variety of transformations. As common acyl electrophiles, inexpensive and commercially available acyl chlorides are widely used in nucleophilic acyl substitution and acyl coupling reactions.⁸ Recently, acyl chlorides have also been developed as aryl electrophiles in decarbonylative coupling reactions.⁹ On the other hand, acyl fluorides have been drawing extensive attention under late transition metal catalysis due to their unique properties.^9,10 While existing strategies provide an ideal blueprint for the utilization of acyl halides, further development in a sustainable and environmentally friendly manner is desirable.

3

1-2 Carbon-Carbon Bond-Forming Reactions of Acyl Halides

Acyl halides have been employed as versatile types of acyl and aryl building blocks to construct a broad variety of C-C bonds. In this section, the Author introduces (1) acyl couplings of acyl fluorides;

(2) decarbonylative couplings of acyl fluorides; and (3) decarbonylative cross-couplings of acyl chlorides. These contents illustrate the development of C-C bond formation reactions from acyl halides and the application potential of these synthetic protocols. Acyl halides also have powerful capabilities in other areas, such as the construction of C-X bonds, but these topics are beyond the scope of this PhD Thesis.

1-2-1 Acyl Couplings of Acyl Fluorides

1-2-1-1 Cross-Coupling Reactions

Over the past few decades, acyl fluorides have been used as acyl electrophiles in Negishi-type,¹¹ Hiyama-type,¹² Suzuki-type,¹³ and reductive^14,15 cross-coupling reactions to provide a variety of ketones under transition metal catalysis. The mechanisms of these reactions, involving oxidative addition, transmetalation, and reductive elimination, are consistent with those of conventional cross- coupling reactions (Scheme 1-1).¹⁰ In these carbonyl-retentive cross-couplings, acyl fluorides behave similarly to acyl chlorides. The unique properties of acyl fluorides have not been explicitly highlighted.

Scheme 1-1. Proposed Mechanism of Cross-Coupling of Acyl Fluorides.

4

In 2004, Rovis’s group reported a groundbreaking study of the acyl coupling reaction of acyl fluorides (Scheme 1-2).¹¹ They developed a Negishi-type cross-coupling reaction of acyl fluorides with organozinc reagents at 23ºC in the presence of a nickel catalyst. A variety of ketones were synthesized from the corresponding substrates with aryl, alkene, and alkyl groups in good to excellent yield. Other carboxylic acids such as anhydrides, esters, thioesters, acyl nitriles, and acyl chlorides were also well tolerated under the same catalytic system.

Scheme 1-2. Ni-Catalyzed Cross-Coupling of Acyl Fluorides with Organozinc Reagents.

In 2016, Sakai et al. developed a palladium-catalyzed Hiyama-type cross-coupling reaction of acyl fluorides with phenyltrifluorosilane, which is the second report on the acyl coupling protocol of acyl fluorides (Scheme 1-3).¹² Interestingly, the product yields obtained from the acyl fluorides were similar to those of the acyl chlorides under the same conditions, but the yield of the product from acyl bromide was lower.

Scheme 1-3. Pd-Catalyzed Cross-Coupling of Acyl Fluorides with Phenyltrifluorosilane.

Scheme 1-4. Pd-Catalyzed Cross-Coupling of Acyl Fluorides with Organoboron Reagents.

In 2017, Sakai and Ogiwara’s group extended the ketone formation reaction of acyl fluorides from

5

Hiyama coupling to Suzuki-Miyaura coupling and disclosed the palladium-catalyzed reaction of acyl fluorides with organoboron reagents (Scheme 1-4).¹³ This method was found to be more efficient than the protocol of organosilane because of the high product yields, broad substrate scope, and well functional group tolerance.

Acyl fluorides were also used for reductive cross-couplings, avoiding complex pretreatment and cumbersome final treatment with hazardous organometallic reagents. In 2019, Shu and co-workers first developed a nickel-catalyzed reductive coupling between acyl fluorides and vinyl triflates, providing a straightforward route to various enones (Scheme 1-5).¹⁴ This method was compatible with a wide range of cyclic vinyl triflates and acyl fluorides. Both reactants could be derived directly from available carboxylic acids and alcohols.

Scheme 1-5. Ni-Catalyzed Reductive Cross-Coupling of Acyl Fluorides with Vinyl Triflates.

Scheme 1-6. Ni-Catalyzed Reductive Cross-Coupling of Acyl Fluorides with N-Alkyl Pyridinium Salts.

6

Recently, Weix and Watson presented a reductive cross-coupling of in-situ formed or preformed acyl fluorides with N-alkyl pyridinium salts in the presence of nickel catalyst and manganese metal (Scheme 1-6).¹⁵ This synthetic method had a wide range of substrates and provided a wide variety of dialkyl ketones with various functional groups and pharmaceutical intermediates.

1-2-1-2 Nucleophilic Acyl Substitution Reactions

Acyl chlorides are the most widely used electrophilic reagents in nucleophilic acyl substitution reactions.¹ On the other hand, the C(acyl)-F bond of acyl fluorides is more stable than the C(acyl)- Cl bond of acyl chlorides, allowing the substitution process to be carried out with higher selectivity.

Furthermore, the formation of stable by-products such as fluorosilanes and fluoroboranes is a powerful driving force for these transformations.

In 2017, Riant and Gagosz et al. disclosed the copper-catalyzed boroacylation of allenes with acyl fluorides (Scheme 1-7).¹⁶ In this synthetic protocol, acyl fluorides presented the best properties for trapping the intermediately formed allylcopper species than other carboxylic acid derivatives such as acyl chlorides. The mild transformation exhibited good functional group tolerance.

Scheme 1-7. Cu-Catalyzed Boroacylation Reaction of Allenes with Acyl Fluorides.

Scheme 1-8. NHC-Catalyzed β-Lactonization Cyclization Reaction of Acyl Fluorides.

Lupton’s group reported an enantioselective N-heterocyclic carbene (NHC)-catalyzed indenes synthesis reaction in 2017 (Scheme 1-8).¹⁷ α,β-Unsaturated acyl fluorides and trimethylsilyl enol ethers were used as starting materials. The reaction showed broad generality and high

7

enantioselectivity. The key point of this protocol is the formation of acyl azolium intermediates, which are formed in situ from NHC-catalyzed defluorination and desilylation processes.

Scheme 1-9. NHC-Catalyzed Reactions of Acyl Fluorides via Acyl Azolium.

Recently, Lupton’s group has expanded the scope of enantioselective NHC-catalyzed reactions of

8

acyl fluorides via α,β-unsaturated acyl azolium, which could avoid the use of preactivated nucleophiles, additional oxidizing reagents, or bases (Scheme 1-9).¹⁸ The expected products were produced with high enantioselectivity and yields under the similar reaction conditions without additional optimization.

In 2020, Hopkinson et al. disclosed a Diels-Alder reaction of acyl fluorides with trifluoroacetophenones, which led to the synthesis of a variety of isochroman-1-one derivatives via an NHC-catalyzed photoenolizative process (Scheme 1-10).¹⁹ Stoichiometric studies and TD-DFT calculations supported the existence of azolium intermediates, which exhibited high photochemical reactivity under UVA irradiation.

Scheme 1-10. NHC-Catalyzed Photoenolization Diels-Alder Reaction of Acyl Fluorides.

Scheme 1-11. NHC/Photoredox-Catalyzed Addition Reaction of Acyl Fluorides, Alkenes, and CF3SO2Na.

Recently, Studer and co-workers reported a cooperative photoredox and NHC-catalyzed three- component coupling of acyl fluorides, styrenes and CF3SO2Na (Scheme 1-11).²⁰ Due to the mild reaction conditions, various α-substituted and β-trifluoromethyl ketones with broad functional groups could be synthesized with high efficiency.

9

Recently, Tobisu et al. described the first tricyclohexylphosphine-catalyzed acylfluorination reaction (Scheme 1-12).²¹ This method successfully achieved acyl fluorides as fluorinating reagents and enabled transition-metal-free intermolecular addition of acyl groups and fluorine to internal alkynes. In addition, the mild reaction conditions allowed good functional group tolerance.

Experimental and DFT computational studies revealed that the phosphorus intermediate is a key point in the C-F bond formation.

Scheme 1-12. Phosphine-Catalyzed Addition of Acyl Fluorides to Internal Alkynes.

1-2-1-3 C-H Bond Activation Reaction

In 2019, Sakai and Ogiwara’s group reported the palladium/copper-cocatalyzed acylation of acyl fluorides with azoles in air (Scheme 1-13).²² Mechanistic studies suggested that the copper/phosphine catalyst promoted the cleavage of aromatic C-H bonds. Notably, compared with other carboxylic acid derivatives, acyl fluorides were found to be the best balance between stability and reactivity.

Scheme 1-13. Pd/Cu-Catalyzed Coupling of Acyl Fluorides with Azoles.

1-2-2 Decarbonylative Couplings of Acyl Fluorides

In general, acyl fluorides are more stable than anhydrides and other acyl halides in the presence of water and other nucleophiles because of the high energy of the C(acyl)-F bond. On the other hand, acyl fluorides are more active than esters and amides under transition metal catalysis. Moreover, the fluoride anion presents specific reactivity in the transmetalation process, acting as a mild base in certain cases. With a very good balance of these properties, acyl fluorides play a unique role in

10 decarbonylative cross-coupling reactions.

Schoenebeck and co-workers published a pioneering demonstration of palladium-catalyzed decarbonylative trifluoromethylation of acyl fluorides in 2018 (Scheme 1-14).²³ This strategy relied on the intramolecular redistribution of fluoride, which is crucial in the transmetalation process, and avoided exogenous fluorides. Xantphos was successfully applied to catalytic trifluoromethylation for the first time.

Scheme 1-14. Pd-Catalyzed Decarbonylative Trifluoromethylation of Acyl Fluorides.

In 2018, Nishihara et al. disclosed the nickel-catalyzed decarbonylative ethylation/methylation of acyl fluorides using organoboron reagents (Scheme 1-15).²⁴ The formation of the B-F bond provided a powerful driving force and avoided the use of base in the ethylation coupling.

Scheme 1-15. Ni-Catalyzed Decarbonylative Alkylation of Acyl Fluorides.

In 2018, Sanford and co-workers developed a nickel-catalyzed decarbonylative cross-coupling reaction between acyl fluorides formed in situ and boronic acids (Scheme 1-16).²⁵ The synthetic protocol initially achieved a base-free Suzuki–Miyaura-type cross-coupling, which means that base- sensitive boronic acids were well tolerated in this protocol. To demonstrate its application, a wide variety of biologically active carboxylic acids were transformed to the corresponding products in high

11 yields.

Scheme 1-16. Ni-Catalyzed Decarbonylative Coupling of Acyl Fluorides with Boronic Acids.

In 2019, Tobisu et al. reported the decarbonylative arylation of acyl fluorides with (hetero)arenes (Scheme 1-17).²⁶ The acyl fluorides exhibited high reactivity in the iridium-catalyzed C-H bond activation reaction. Other acyl halides were less efficient in this transformation because the iridium fluoride intermediate was responsible for the C-H bond cleavage event.

Scheme 1-17. Ir-Catalyzed Decarbonylative Arylation of Acyl Fluorides and Aromatics.

Recently, Ogiwara and Sakai et al. carried out a palladium-catalyzed cyclization of acyl fluorides with norbornene to provide polycyclic ketones (Scheme 1-18).²⁷ The annulation proceeded via decarbonylation of the acyl fluoride and activation of the ortho C–H bond. Rearrangement and reinsertion of the CO group were essential for this transformation.

Scheme 1-18. Pd-Catalyzed Decarbonylative Cyclization Reaction of Acyl Fluorides.

In 2020, Nishihara et al. reported the bimetallic palladium/copper-catalyzed decarbonylative Sonogashira-Hagihara alkynylation of acyl fluorides with alkynylsilanes (Scheme 1-19).²⁸ This reaction avoided the addition of base, suppressed the formation of undesired homocoupled products, and provided a wide scope of unsymmetrical diarylethynes in moderate to high yields. They also described the first nickel-catalyzed decarbonylative alkynylation of acyl fluorides with terminal

12

silylethynes in the same year (Scheme 1-20). ²⁹ The copper-free reaction conditions allowed a wide range of substrates and afforded a variety of internal silylethynes. In addition, side reactions such as homocoupling and carbonyl-retentive coupling were well suppressed.

Scheme 1-19. Pd/Cu-Catalyzed Decarbonylative Alkynylation of Acyl Fluorides and Alkynylsilanes.

Scheme 1-20. Ni-Catalyzed Decarbonylative Alkynylation of Acyl Fluorides and Terminal Alkynes.

1-2-3 Decarbonylative Cross-Couplings of Acyl Chlorides

Despite the limitation of poor stability, acyl chlorides have become the most widely used carboxylic acid derivatives in academic laboratories and industry due to their low cost, simple structure, and high reactivity. With this background, nucleophilic acyl substitution and acyl coupling reactions of acyl chlorides have already attracted extensive attention. On the contrary, only a few decarbonylative cross-coupling reactions of acyl chlorides have been established so far.

Scheme 1-21. Rh-Catalyzed Decarbonylative Mizoroki-Heck-Type Reaction of Acyl Chlorides.

In 2003, Miura et al. demonstrated a rhodium(I)-catalyzed Mizoroki-Heck-type coupling of acyl chlorides with alkenes, which is the first catalytic strategy of decarbonylative coupling of acyl

13

chlorides (Scheme 1-21).³⁰ Notably, this protocol did not use any phosphine ligands and bases, which rendered the work-up procedure more efficient.

In 2008, Yu et al. reported the rhodium(I)-catalyzed decarbonylative aromatic carbon-hydrogen bond activation of acyl chlorides and N-heteroaromatic substrates (Scheme 1-22).³¹ The results showed that the 2-pyridyl directing group facilitated regioselective functionalization without the phosphine ligand.

Scheme 1-22. Rh-Catalyzed Decarbonylative C-H Bond Activation of Acyl Chlorides.

In 2017, Sanford’s group developed a palladium-catalyzed method for the formation of aryl chlorides from decarbonylative intramolecular coupling of acyl chlorides (Scheme 1-23).³² In turn, the palladium/Brettphos catalytic system allowed the following one-pot cross-couplings to access a variety of C-X and C-C bonds.

Scheme 1-23. Pd-Catalyzed Decarbonylative Cross-Coupling Reactions of Acyl Chlorides.

Scheme 1-24. Pd-Catalyzed Decarbonylative Difluoromethylation of Acyl Chlorides.

In 2018, Ritter and co-workers reported the decarbonylative difluoromethylation of acyl chlorides with organozinc reagents under palladium/Ruphos catalysis (Scheme 1-24).³³ Ruphos played a

14

crucial role in promoting both decarbonylation and reductive elimination at room temperature. This transformation showed a wide range of functional group tolerance and became an efficient route to difluoromethylating arenes.

In 2019, Nishihara and co-workers developed a nickel-catalyzed decarbonylative cyanation of in- situ formed or pre-prepared acyl chlorides using trimethylsilyl cyanide as nucleophilic reagent (Scheme 1-25).³⁴ In this protocol, a variety of acyl chlorides were readily transformed to nitriles under redox-neutral conditions. Meanwhile, the reaction mechanism was investigated step by step.

Scheme 1-25. Ni-Catalyzed Decarbonylative Cyanation of Acyl Chlorides.

Recently, Szostak and Hong et al. developed a palladium-catalyzed decarbonylative Suzuki- Miyaura cross-coupling for a wide range of acyl chlorides and boronic acids (Scheme 1-26).³⁵ This strategy allowed for direct functionalization of natural products and pharmaceuticals containing carboxylic acid groups.

Scheme 1-26. Pd-Catalyzed Decarbonylative Suzuki-Miyaura Coupling of Acyl Chlorides.

1-3 Decarboxylative Cross-Coupling Reactions for Carbon–Carbon Bond Formation

1-3-1 General Introduction

Carboxylic acids are considered to be an ideal class of coupling reagents, highlighted by their commercial availability, renewability, wide structural diversity, low cost, stability, and non-toxicity.⁴ In addition, the special decarboxylative process allows high regio- and chemo-selectivity with a traceless byproduct.⁶ In 1966, Nilsson reported the pioneering work on the decarboxylative cross-

15

coupling of benzoic acids with iodoarenes.³⁶ This stoichiometric copper(I) oxide-catalyzed Ullmann-type coupling proceeded in quinoline at 240 °C. In this case, the harsh reaction conditions and low yields severely limited the application of decarboxylative cross-couplings. By 2006, Gooßen et al. developed an innovative strategy to achieve the catalytic decarboxylative couplings for biaryls using bimetallic palladium and copper catalysts, allowing decarboxylation of benzoic acids at 120 ^oC to 160 ^oC (Scheme 1-27).³⁷ In this context, decarboxylative cross-coupling reactions have been attracted much attention as the powerful surrogate for conventional cross-coupling protocols, and hence have been explosively developed.⁶

Scheme 1-27. Pd/Cu-Catalyzed Decarboxylation Cross-Coupling Reactions.

Previous decarboxylative couplings include, but are not limited to, (1) cross-coupling reactions with electrophiles under redox-neutral conditions, (2) Heck-type vinylation reactions, (3) C-H activation reactions, (4) cross-coupling reactions with organometallic reagents under oxidative conditions, (5) radical and photocatalyzed reactions, and (6) conjugate addition reactions.

Although decarboxylative couplings have unique features and provide a wide variety of products, there are still issues such as high reaction temperature, limited substrate scope, and high catalyst loading. Therefore, advances in new catalytic protocols and the utilization of these strategies in the synthetic industry are particularly attractive.

1-3-2 Decarboxylative Cross-Couplings of Perfluorobenzoates

Synthetic approaches to perfluorinated arenes have attracted much attention in the fields of material³⁸ and pharmaceutical³⁹ sciences due to their appealing properties. As the most atom- economic protocol, the direct installation of perfluorophenyl groups has drawn extensive attention.

In the past few decades, transition-metal-catalyzed cross-coupling reactions of perfluorinated

16

organometallic compounds,⁴⁰ C-H bond activation reactions of perfluorobenzenes,⁴¹ and functionalization reactions of C-F bond in perfluorobenzenes⁴² become the mainstream. In this context, a series of decarboxylative cross-coupling reactions of perfluorobenzoates have been developed due to the considerable advantages of the decarboxylative strategy. In particular, benzoates are considered to be a combination of benzoic acid, base, and water absorbent.

Scheme 1-28. Cu-Catalyzed Decarboxylative Cross-Coupling Reactions of Aryl Iodides and Bromides with Potassium Perfluorobenzoates.

Scheme 1-29. Pd-Catalyzed Decarboxylative Cross-Coupling Reaction of Aryl Bromides, Chlorides, and Triflates with Potassium Perfluorobenzoates.

In 2009, Liu and co-workers firstly discovered the copper-catalyzed decarboxylative cross-coupling reactions of aryl iodides and bromides with potassium perfluorobenzoates (Scheme 1-28).⁴³ This reaction provided a reliable protocol for the synthesis of perfluorobiaryls from readily available and stable perfluorobenzoates. In contrast to previous decarboxylative cross-coupling reactions, the employment of precious palladium catalysts was avoided in this reaction. Furthermore, DFT calculations indicated that oxidative addition is the rate-limiting step. In 2017, Zhao and Cai et al.

designed a MCM-41-Phen-CuI complex-catalyzed decarboxylative cross-coupling reaction.⁴⁴ This

17

heterogeneous catalyst was easily prepared from available copper precursors, inexpensive 1,10- phenanthroline ligand, and MCM-41 substrates, and presented high catalytic activity even after eight filtration cyclic utilizations.

In the following year, Liu and co-workers achieved the first palladium-catalyzed decarboxylative coupling of potassium perfluorobenzoates with aryl bromides, chlorides, and triflates (Scheme 1-29).⁴⁵ Mechanistic studies suggested that decarboxylation may determine the rate of this transformation.

Scheme 1-30. Ni-Catalyzed Decarboxylative Cross-Coupling Reaction of Potassium Perfluorobenzoates with Aryl Halides and Sulfonates.

Scheme 1-31. Ni-Catalyzed Decarboxylative Cross-Coupling Reactions of Phenylmethanol Derivatives with Potassium Perfluorobenzoates.

In 2015, Kalyani’s group developed the first nickel-catalyzed decarboxylative cross-coupling of aryl halides and pseudohalides with potassium perfluorobenzoates in the presence of PtBu3.HBF4,

18

DPPF or CMphos ligands (Scheme 1-30).⁴⁶ Aryl chlorides, bromides, iodides, triflates, and tosylates were well tolerated in this transformation.

In 2018, Le, Yan, and Zhang et al. reported a series of nickel-catalyzed decarboxylative cross- coupling reactions of unactivated phenol and phenylmethanol derivatives with potassium perfluorobenzoates (Scheme 1-31).⁴⁷ This synthetic strategy expanded the range of electrophilic coupling partners in decarboxylation reactions and provided a new pathway to the production of perfluorobiaryls and perfluorinated diarylmethanes. Mechanistic studies revealed that Zn(OAc)2

plays an important role in decarboxylation process.

In 2019, Topczewski et al. developed a palladium-catalyzed cross-coupling of sodium (hetero)aryl carboxylates with (hetero)aryl chlorides via decarboxylation (Scheme 1-32).⁴⁸ This method was noteworthy for its low catalyst loading and the ability to use less reactive sodium carboxylates and aryl chlorides. Not only perfluorobenzoates but also methoxybenzoates and heteroaryl carboxylates were tolerated.

Scheme 1-32. Pd-Catalyzed Decarboxylative Cross-Coupling of Aryl Chlorides with Sodium Benzoates.

19

1-4 Summary

In this Chapter, the Author mainly described the carbon-carbon bond-forming reactions by couplings of acyl halides and decarboxylative cross-couplings of benzoates. The decarbonylative and decarboxylative synthetic protocols have allowed for the wide availability of carboxylic acid derivatives that can act complementarily as substrates with opposite electronic properties under redox- neutral conditions. Encouraged by the unique properties of readily available acyl halides and benzoates, the construction of carbon-carbon bonds via decarbonylation and decarboxylation has been of great interest.

20

References

1. (a) Trost, B. M.; Fleming, I. Comprehensive Organic Synthesis; Pergamon Press, 1991. (b) Smith, M. B.; March, J. Advanced Organic Chemistry; Wiley, 2007. (c) Brown, W. H.; Iverson, B. L.; Anslyn, E. V.; Foote, C. S. Organic Chemistry; Cengage, 2018.

2. (a) Wu, X.-F.; Anbarasan, P.; Neumann, H.; Beller, M. Angew. Chem., Int. Ed. 2010, 49, 9047- 9050. (b) Seechurn, C. C. C. J.; Kitching, M. O.; Colacot, T. J.; Snieckus, V. Angew. Chem., Int.

Ed. 2012, 51, 5062-5085.

3. (a) de Meijere, A.; Diederich, F. Metal-Catalyzed Cross-Coupling Reactions; Wiley-VCH, 2008.

(b) de Meijere, A.; Bräse, S.; Oestreich, M. Metal-Catalyzed Cross-Coupling Reactions and More; Wiley, 2014.

4. (a) Gooßen, L. J.; Rodriguez, N.; Gooßen, K. Angew. Chem., Int. Ed. 2008, 47, 3100-3120. (b) Rodriguez, N.; Gooßen, L. J. Chem. Soc. Rev. 2011, 40, 5030-5048. (c) Dzik, W. I.; Lange, P.

P.; Gooßen, L. J. Chem. Sci. 2012, 3, 2671-2678.

5. For selected reviews of transition-metal-catalyzed decarbonylative transformations of carboxylic acids and their derivatives, see: (a) Guo, L.; Rueping, M. Acc. Chem. Res. 2018, 51, 1185-1195. (b) Guo, L.; Rueping, M. Chem. Eur. J. 2018, 24, 7794-7809. (c) Liu, C.; Szostak, M. Org. Biomol. Chem. 2018, 16, 7998-8010.

6. For selected reviews of transition-metal-catalyzed decarboxylative transformations of carboxylic acids and their derivatives, see: (a) Shang, R.; Liu, L. Sci. China: Chem. 2011, 54, 1670-1687. (b) Cornella, J.; Larrosa, I. Synthesis 2012, 44, 653-676. (c) Patra, T.; Maiti, D.

Chem. Eur. J. 2017, 23, 7382-7401. (d) Wei, Y.; Hu, P.; Zhang, M.; Su, W. Chem. Rev. 2017, 117, 8864-8907. (e) Zhang, T.; Wang, N.-X.; Xing, Y. J. Org. Chem. 2018, 83, 7559-7565. (f) Schwarz, J.; König, B. Green Chem. 2018, 20, 323-361.

7. For selected papers on the synthesis of acyl fluorides from the corresponding carboxylic acids, see: (a) Chen, C.; Chien, C.-T.; Su, C.-H. J. Fluorine Chem. 2002, 115, 75-77. (b) Beaulieu, F.;

Beauregard, L. P.; Courchesne, G.; Couturier, M.; LaFlamme, F.; L’Heureux, A. Org. Lett. 2009, 11, 5050-5053. (c) Scattolin, T.; Deckers, K.; Schoenebeck, F. Org. Lett. 2017, 19, 5740-5743.

(d) Munoz, S. B.; Dang, H.; Ispizua-Rodriguez, X.; Mathew, T.; Prakash, G. K. S. Org. Lett.

2019, 21, 1659-1663.

8. (a) Roughley, S. D.; Jordan, A. M. J. Med. Chem. 2011, 54, 3451-3479. (b) Brown, D. G.;

Boström, J. J. Med. Chem. 2016, 59, 4443-4458. (c) Buchspies, J.; Szostak, M. Catalysts 2019, 9, 53-75.

9. (a) Wang, Z.; Wang, X.; Nishihara, Y. Chem. Asian J. 2020, 15, 1234-1247. (b) Lu, H.; Yu, T.- Y.; Xu, P.-F.; Wei, H. Chem. Rev. 2020, doi:10.1021/acs.chemrev.0c00153.

21

10. (a) Blanchard, N.; Bizet, V. Angew. Chem., Int. Ed. 2019, 58, 6814-6817. (b) Ogiwara, Y.; Sakai, N. Angew. Chem., Int. Ed. 2020, 59, 574-585.

11. Zhang, Y.; Rovis, T. J. Am. Chem. Soc. 2004, 126, 15964-15965.

12. Ogiwara, Y.; Maegawa, Y.; Sakino, D.; Sakai, N. Chem. Lett. 2016, 45, 790-792.

13. Ogiwara, Y.; Sakino, D.; Sakurai, Y.; Sakai, N. Eur. J. Org. Chem. 2017, 4324-4327.

14. Pan, F.-F.; Guo, P.; Li, C.-L.; Su, P.; Shu, X.-Z. Org. Lett. 2019, 21, 3701-3705.

15. Wang, J.; Hoerrner, M. E.; Watson, M. P.; Weix, D. J. Angew. Chem., Int. Ed. 2020, 59, 13484- 13489.

16. Boreux, A.; Indukuri, K.; Gagosz, F.; Riant, O. ACS Catal. 2017, 7, 8200-8204.

17. Zhang, C.; Lupton, D. W. Org. Lett. 2017, 19, 4456-4459.

18. Cao, J.; Gillard, R.; Jahanbakhsh, A.; Breugst, M.; Lupton, D. W. ACS Catal. 2020, 10, 11791- 11796.

19. Mavroskoufis, A.; Rajes, K.; Golz, P.; Agrawal, A.; Ruß, V.; Götze, J. P.; Hopkinson, M. N.

Angew. Chem., Int. Ed. 2020, 59, 3190-3194.

20. Meng, Q.-Y.; Dçben, N.; Studer, A. Angew. Chem., Int. Ed. 2020, 59, 19956-19960.

21. Fujimoto, H.; Kodama, T.; Yamanaka, M.; Tobisu, M. J. Am. Chem. Soc. 2020, 142, 17323- 17328.

22. Ogiwara, Y.; Iino, Y.; Sakai, N. Chem. Eur. J. 2019, 25, 6513-6516.

23. Keaveney, S. T.; Schoenebeck, F. Angew. Chem., Int. Ed. 2018, 57, 4073-4077.

24. Okuda, Y.; Xu, J.; Ishida, T.; Wang, C.-A.; Nishihara, Y. ACS Omega 2018, 3, 13129-13140.

25. Malapit, C. A.; Bour, J. R.; Brigham, C. E.; Sanford, M. S. Nature 2018, 563, 100-104.

26. Sakurai, S.; Yoshida, T.; Tobisu , M. Chem. Lett. 2019, 48, 94-97.

27. Sakurai, Y.; Ogiwara, Y.; Sakai, N. Chem. Eur. J. 2020, 26, 12972-12977.

28. Chen, Q.; Fu, L.; Nishihara, Y. Chem. Commun. 2020, 56, 7977-7980.

29. Chen, Q.; Fu, L.; You, J.; Nishihara, Y. Synlett 2020, 31, DOI: 10.1055/s-0040-1705954.

30. Sugihara, T.; Satoh, T.; Miura, M.; Nomura, M. Angew. Chem., Int. Ed. 2003, 42, 4672-4674.

31. Zhao, X.; Yu, Z. J. Am. Chem. Soc. 2008, 130, 8136-8137.

32. Malapit, C. A.; Ichiishi, N.; Sanford, M. S. Org. Lett. 2017, 19, 4142-4145.

33. Pan, F.; Boursalian, G. B.; Ritter, T. Angew. Chem., Int. Ed. 2018, 57, 16871-16876.

34. Wang, Z.; Wang, X.; Ura, Y.; Nishihara, Y. Org. Lett. 2019, 21, 6779-6784.

35. Zhou, T.; Xie, P.-P.; Ji, C.-L.; Hong, X.; Szostak, M. Org. Lett. 2020, 22, 6434-6440.

36. Nilsson, M. Acta Chem. Scand. 1966, 22, 423-426.

37. Gooßen, L. J.; Deng, G.; Levy, L. M. Science, 2006, 313, 662-664.

38. (a) Mewshaw, R. E.; Edsall Jr, R. J.; Yang, C.; Manas, E. S.; Xu, Z. B.; Henderson, R. A.; Keith Jr, J. C.; Harris, H. A. J. Med. Chem. 2005, 48, 3953-3979. (b) Backes, B. J.; Longenecker, K.;

Hamilton, G. L.; Stewart, K.; Lai, C.; Kopecka, H.; von Geldern, T. W.; Madar, D. J.; Pei, Z.;

22

Lubben, T. H.; Zinker, B. A.; Tian, Z.; Ballaron, S. J.; Stashko, M. A.; Mika, A. K.; Beno, D.

W. A.; Kempf-Grote, A. J.; Black-Schaefer, C.; Sham H. L.; Trevillyan, J. M. Bioorg. Med.

Chem. Lett. 2007, 17, 2005-2012. (c) de Candia, M.; Liantonio, F.; Carotti, A.; De Cristofaro, R.; Altomare, C. J. Med. Chem. 2009, 52, 1018-1028. (d) Gillis, E. P.; Eastman, K. J.; Hill, M.

D.; Donnelly, D. J.; Meanwell, N. A. J. Med. Chem. 2015, 58, 8315-8359. (e) Preshlock, S.;

Tredwell, M.; Gouverneur, V. Chem. Rev. 2016, 116, 719-766.

39. (a) Sakamoto, Y.; Suzuki, T.; Miura, A.; Fujikawa, H.; Tokito, S.; Taga, Y. J. Am. Chem. Soc.

2000, 122, 1832-1833. (b) Nitschke, J. R.; Tilley, T. D. J. Am. Chem. Soc. 2001, 123, 10183- 10190. (c) Tsuzuki, T.; Shirasawa, N.; Suzuki T.; Tokito, S. Adv. Mater. 2003, 15, 1455-1458.

(d) Facchetti, A.; Yoon, M. H.; Stern, C. L.; Katz, H. E.; Marks, T. J. Angew. Chem., Int. Ed.

2003, 42, 3900-3903. (e) Zacharias, P.; Gather, M. C.; Rojahn, M.; Nuyken, O.; Meerholz, K.

Angew. Chem., Int. Ed. 2007, 46, 4388-4392. (f) Babudri, F.; Farinola, G. M.; Naso, F.; Ragni, R. Chem. Commun. 2007, 1003-1022. (g) Tang, M. L.; Reichardt, A. D.; Miyaki, N.; Stoltenberg, R. M.; Bao, Z. J. Am. Chem. Soc. 2008, 130, 6064-6065. (h) Okamoto, T.; Nakahara, K.; Saeki, A.; Seki, S.; Oh, J. H.; Akkerman, H. B.; Bao, Z.; Matsuo, Y. Chem. Mater. 2011, 23, 1646- 1649. (i) Lai, Y.-Y.; Cheng, Y.-J.; Hsu, C.-S. Energy Environ. Sci. 2014, 7, 1866-1883. (j) Shin, H.; Kim, D.; Kim, H. J.; Kim, J.; Char, K.; Yavuz, C. T.; Choi, J. W. Chem. Mater. 2019, 31, 7910-7921.

40. (a) Kremlev, M. M.; Tyrra, W.; Naumann, D.; Yagupolskii, Y. L. J. Fluorine Chem. 2005, 126, 1327-1331. (b) Korenaga, T.; Kosaki, T.; Fukumura, R.; Ema, T.; Sakai, T. Org. Lett. 2005, 7, 4915-4917. (c) Kinzel, T.; Zhang, Y.; Buchwald, S. L. J. Am. Chem. Soc. 2010, 132, 14073- 14075. (d) Sugita, N.; Hayashi, S.; Ishii, S.; Takanami, T. Catalysts 2013, 3, 839-852.

41. (a) Lafrance, M.; Shore, D.; Fagnou, K. Org. Lett. 2006, 8, 5097-5100. (b) Do, H.-Q.; Daugulis, O. J. Am. Chem. Soc. 2008, 130, 1128-1129. (c) Wei, Y.; Su, W. J. Am. Chem. Soc. 2010, 132, 16377-16379. (d) He, M.; Soule, J.-F.; Doucet, H. ChemCatChem 2014, 6, 1824-1859. (e) Lian, Z.; Friis, S. D.; Skrydstrup, T. Chem. Commun. 2015, 51, 1870-1873.

42. Arora, A.; Weaver, J. D. Acc. Chem. Res. 2016, 49, 2273-2283.

43. Shang, R.; Fu, Y.; Wang, Y.; Xu, Q.; Yu, H.-Z.; Liu, L. Angew. Chem., Int. Ed. 2009, 48, 9350- 9354.

44. Lin, Y.; Cai, M.; Fang, Z.; Zhao, H. RSC Adv. 2017, 7, 34722-34729.

45. Shang, R.; Xu, Q.; Jiang, Y.-Y.; Wang, Y.; Liu, L. Org. Lett. 2010, 12, 1000-1003.

46. Sardzinski, L. W.; Wertjes, W. C.; Schnaith, A. M.; Kalyani, D. Org. Lett. 2015, 17, 1256-1259.

47. Chen, Q.; Wu, A.; Qin, S.; Zeng, M.; Le, Z.; Yan, Z.; Zhang, H. Adv. Synth. Catal. 2018, 360, 3239-3244.

48. Daley, R. A.; Liu, E.-C.; Topczewski, J. J. Org. Lett. 2019, 21, 4734-4738.

Chapter 2

Palladium-Catalyzed Decarbonylative Alkylation

of Acyl Fluorides

24

2-1 Introduction

Palladium-catalyzed cross-coupling reactions to form carbon-carbon bonds are an important class in organic chemistry,¹ and the Suzuki-Miyaura cross-coupling reactions, in particular, have been widely developed and utilized to construct a variety of carbon-carbon bonds.² However, when alkylboron reagents are employed as coupling partners, undesired pathways, such as β-hydride elimination, are inevitable.³ As a result, compared to C(sp²)-C(sp²) and C(sp²)-C(sp) bonds, only a few examples of the construction of C(sp²)-C(sp³) bonds have been reported.^4,5 Therefore, there is still a high demand to develop new synthetic approaches to access C(sp²)-C(sp³) bonds.⁶

Scheme 2-1. Decarbonylative Alkylation of Carboxylic Acid Derivatives.

In the last few decades, the transformation of carboxylic acids by carboxylation has attracted much

25

attention in organic synthesis due to their wide abundance, ready availability, and environmentally friendliness.⁷ Consequently, carboxylic acid derivatives such as esters and amides have been developed as alternative coupling partners for halogenated electrophiles through decarbonylation.⁸ In this connection, Rueping,⁹ Newman,¹⁰ and Yamaguchi¹¹ have disclosed a series of decarbonylative alkylation reactions of esters and amides and with alkylboron, -zinc, or -aluminum reagents (Scheme 2-1).

Scheme 2-2. Decarbonylative Ethylation and Methylation of Acyl Fluorides.

In the course of our ongoing research on the transformation of acyl halides, we succeeded in developing nickel-catalyzed decarbonylative ethylation (triethylboron) and methylation (methylboroxine) of acyl fluorides (Scheme 2-2).¹² However, this method could not be applied to other alkylating agents.

Scheme 2-3. Decarbonylative Alkylation of Acyl Fluorides.

In this Chapter, in order to render the previous work more general, the Author developed a Pd- catalyzed decarbonylative alkylation of acyl fluorides with a variety of functionalized alkylboranes (Scheme 2-3).

2-2 Results and Discussion

2-2-1 Optimization of the Reaction Conditions

We first investigated the reaction of benzoyl fluoride (1a) with B-(3-phenylpropyl)-9-borabicyclo [3.3.1]nonane (2a) obtained by in-situ hydroboration of allylbenzene with 9-BBN dimer in mesitylene.

Table 2-1. Screening of Catalysts and Ligands.^a

26

entry cat. ligand yield (%)^b

3aa 4

1 Ni(cod)2 DCYPE 0 1

2 Pd(OAc)2 DCYPE 38 2

3 Pd(dba)2 DCYPE 42 3

4 Pd(acac)2 DCYPE 44 1

5 Pd(tfa)2 DCYPE 39 4

6^c Pd2(dba)3 DCYPE 33 3

7 PdCl2 DCYPE 28 2

8 Pd(acac)2 DPPM 4 14

9 Pd(acac)2 DPPE 44 4

10 Pd(acac)2 DPPP 17 24

11 Pd(acac)2 DPPF 10 12

12 Pd(acac)2 XantPhos 19 0

13 Pd(acac)2 rac-BINAP 20 24

14 Pd(acac)2 PPh3d 10 41

aReactions were carried out with 1a (0.2 mmol, 1.0 equiv), 2a (0.3 mL, 1.0 M in mesitylene, 0.3 mmol, 1.5 equiv) and cat. (0.02 mmol, 10 mol %), [P] (0.04 mmol, 20 mol %), KF (0.3 mmol, 1.5 equiv) in mesitylene (0.7 mL) at 160 °C for 20 h. ^bGC yields, using n-dodecane as the internal standard. ^c5 mol %. ^d40 mol %.

Firstly, we screened transition metal catalysts with DCYPE as the ligand (Table 2-1, entries 1-7).

27

Ni(cod)2, which showed superior catalytic activity for ethylation and methylation in previous studies, was tested, but no product 3aa was detected (entry 1). In contrast, Pd(acac)2 yielded 44% of 3aa and 1% of 4 (entry 4), indicating that the choice of transition metal catalyst was critical for decarbonylative alkylation. In general, β-hydride elimination from alkylnickel complexes is considered more difficult than for alkylpalladium complexes. However, in some cases, nickel complexes are known to cause more β-hydride elimination than palladium complexes.¹³ After further optimization of the ligand, DPPE was also an efficient ligand, affording 3aa in 44% yield (entry 9). DPPE was chosen as a ligand due to its low cost and high stability.

Table 2-2. Screening of Base.^a

entry base yield (%)^b

3aa 4

1 NaF 32 3

2 KF 44 4

3 CsF 29 2

4 Cs2CO3 8 0

5 K2CO3 7 0

6 K3PO4 42 8

7 KOAc 37 12

8 KO^tBu 0 0

aReactions were carried out with 1a (0.2 mmol, 1.0 equiv), 2a (0.3 mL, 1.0 M in mesitylene, 0.3 mmol, 1.5 equiv) and Pd(acac)2 (0.02 mmol, 10 mol %), DPPE (0.04 mmol, 20 mol %), base (0.3 mmol, 1.5 equiv) in mesitylene (0.7 mL) at 160 °C for 20 h. ^bGC yields, using n-dodecane as the internal standard.

Next, we tested various bases using Pd(acac)2/DPPE as the optimal catalysis system. The results showed that the potassium cation presented higher activity than the other cations (Table 2-2). The common weak base, potassium fluoride, was used to obtain the desired product 3aa in high yield (entry 2). Moreover, the reaction was also stopped by the strong base, probably because it reacted directly with benzoyl fluoride (entry 8).

28

Subsequently, the amount of ligand and base were optimized. When 15% DPPE and 3.0 equiv KF were employed, the yield of 3aa increased to 52% (Table 2-3, entry 2). Notably, the increase in ligand could suppress the decarboxylation reaction and promote the retentive reaction (entries 1-5).

Pleasingly, 59% of 3aa was observed when the reaction temperature was 140 ℃ (entry 10).

Table 2-3. Screening of Reaction Conditions.^a

entry x y T (^oC) yield (%)^b

3aa 4

1 10 1.5 160 40 7

2 15 1.5 160 50 8

3 20 1.5 160 44 4

4 30 1.5 160 26 32

5 40 1.5 160 18 36

6 15 1.0 160 48 5

7 15 2.0 160 50 4

8 15 3.0 160 52 5

9 15 3.0 130 47 0

10 15 3.0 140 59 4

11 15 3.0 150 52 3

12 15 3.0 170 49 9

aReactions were carried out with 1a (0.2 mmol, 1.0 equiv), 2a (0.3 mL, 1.0 M in mesitylene, 0.3 mmol, 1.5 equiv) and Pd(acac)2 (0.02 mmol, 10 mol %), DPPE (x mol %), KF (y equiv) in mesitylene (0.7 mL) for 20 h. ^bGC yields, using n-dodecane as the internal standard.

After the completion of the reactions, obvious palladium black was always observed on the inner surface of the Schlenk tube. Thus, we proposed that the deactivation of the palladium catalyst might reduce the yield of the desired product. To avoid catalyst deactivation, we optimized the solvent type and concentration (Table 2-4, entries 1-4). When 0.1 M toluene was employed instead of 0.2 M

29

mesitylene, the yield of 3aa increased to 71% (entry 3). Inspired by the related reports,¹³ we hypothesized that DMSO may act as the ligand to dissolve KF and stabilize the palladium intermediate.

To our delight, the toluene/DMSO co-solvent system was efficient, providing 3aa in 90% yield with only 1.5 equiv of KF, and no ketone 4 was detected (entry 8).

Table 2-4. Screening of Solvent.^a

entry toluene (mL) DMSO yield (%)^b

3aa 4

1 0.7 - 62 2

2^c 0.7 - 59 4

3 1.7 - 71 2

4 2.7 - 66 3

5 1.7 2 equiv 85 1

6 1.6 0.1 mL 87 1

7 1.5 0.2 mL 90 0

8^d 1.5 0.2 mL 90 0

9 1.2 0.5 mL 67 0

aReactions were carried out with 1a (0.2 mmol, 1.0 equiv), 2a (0.3 mL, 1.0 M in toluene, 0.3 mmol, 1.5 equiv) and Pd(acac)2 (0.02 mmol, 10 mol %), DPPE (0.03 mmol, 15 mol %), KF (0.6 mmol, 3.0 equiv) in solvent at 140 °C for 20 h. ^bGC yields, using n-dodecane as the internal standard. ^cMesitylene instead of toluene. ^dKF (0.4 mmol, 1.5 equiv).

Finally, we examined the reaction time. Gratifyingly, the reaction could be completed within 1 h with 90% GC yield and 80% isolated yield, which was highly efficient compared to previous reports (Table 2-5, entry 3).^9-12

In control experiments, the desired product was not obtained in the absence of palladium catalyst or ligand (Table 2-6, entries 2-3). Considering the typical decarbonylative Suzuki-Miyaura coupling and borylation reaction of acyl fluorides in the absence of an external base under nickel catalysis

30

reported by Sanford’s group,¹⁵ 3aa was observed in 38% yield without the addition of an external KF, indicating that the fluorine atom of the acyl fluoride acted as a base in the coupling (entry 4). Most importantly, when benzoyl chloride was employed instead of 1a under the same reaction conditions, the desired product 3aa was not obtained (entry 6), which demonstrated that the acyl fluoride played a vital role in this transformation. Similarly, no desired product was detected when fluorobenzene was used as the reactant (entry 7), which highlighted the specific reactivity of acyl fluorides and their dominance in the decarbonylative reaction. Furthermore, when the reaction was conducted using [(PPh3)2Pd(C6H5)F] complex instead of 1a and catalysis system, 88% of 3aa was obtained at room temperature (entry 8). This result demonstrated that the transmetalation step and the reductive elimination step could not determine the reaction rate. The yield was found to decrease with decreasing alkylborane or catalyst loading (entries 9-10). In addition, other alkylborane compounds, such as alkyl-BF3K and alkyl-Bnep, were inert reactants for this system (entries 11-12).

Table 2-5. Optimization of Reaction Time.^a

entry time yield (%)^b

3aa 4

1 10 min 36 0

2 30 min 78 0

3 1 h 90 (80) 0

4 20 h 90 0

aReactions were carried out with 1a (0.2 mmol, 1.0 equiv), 2a (0.3 mL, 1.0 M in toluene, 0.3 mmol, 1.5 equiv) and Pd(acac)2 (0.02 mmol, 10 mol %), DPPE (0.03 mmol, 15 mol %), KF (0.3 mmol, 1.5 equiv) in toluene (1.5 mL), DMSO (0.2 mL) at 140 °C. ^bGC yields, using n-dodecane as the internal standard. An isolated yield is given in parentheses.

Table 2-6. Control Experiments.^a

31

entry conditions yield (%)^b

3aa 4

1 none 90 0

2 without Pd(acac)2 0 0

3 without DPPE 0 0

4 without KF 38 0

5 room temperature 0 0

6 PhCOCl instead of PhCOF 0 0

7 PhF instead of PhCOF 0 0

8^c [(PPh3)2Pd(C6H5)F] instead of PhCOF, Pd(acac)2

and DPPE at room temperature 88 0

9 1.2 equiv 2a 85 0

10 5 mol % Pd(acac)2 and 7.5 mol % DPPE 81 0

11 instead of 2a 0 0

12 instead of 2a 0 0

aReactions were carried out with 1a (0.2 mmol, 1.0 equiv), 2a (0.3 mL, 1.0 M in toluene, 0.3 mmol, 1.5 equiv) and Pd(acac)2 (0.02 mmol, 10 mol %), DPPE (0.03 mmol, 15 mol %), KF (0.3 mmol, 1.5 equiv) in toluene (1.5 mL), DMSO (0.2 mL) at 140 °C for 20 h. ^bGC yields, using n-dodecane as the internal standard. ^cReactions were carried out with [(PPh3)2Pd(C6H5)F], 2a (0.3 mL, 1.0 M in toluene, 0.3 mmol, 1.5 equiv) and KF (0.3 mmol, 1.5 equiv) in toluene (1.5 mL), DMSO (0.2 mL) at room temperature for 1 h.

2-2-2 Scope of Acyl Fluorides

Based on these results, the range of acyl fluorides 1 with alkylborane 2a as coupling partner was investigated under optimal conditions (Scheme 2-4).

32

Scheme 2-4. Scope of Acyl Fluorides.^a

aReaction conditions: 1 (0.2 mmol), 2a (0.3 mmol), Pd(acac)2 (10 mol %), DPPE (15 mol %), KF (0.3 mmol), in toluene/DMSO (9:1, 0.1 M) at 140 °C for 1 h under Ar. Isolated yields. ^bReaction was performed in 8.0 mmol.

cDCYPE (15 mol %) was used instead of DPPE. ^d20 h.

33

A series of acyl fluorides bearing electron-donating and -withdrawing functional groups could be converted to the corresponding alkylated products 3aa-3na in moderate to high yields. The use of sterically hindered acyl fluoride 1e smoothly provided 3ea in 87% yield. Fluoro (3ja), trifluoromethyl (3ka, 3la), cyano (3ma), and ketone (3na) substituents were well tolerated.

Furthermore, not only naphthyl-derived substrates but also acyl fluorides derived from benzofuran and benzothiophene were compatible, giving 3oa-3ra in 54-81% yields. Remarkably, large-scale synthesis up to 8.0 mmol was also possible with low doses of palladium, providing 0.93 g of 3aa in 59% yield.

This method presented good tolerance to aromatic acyl fluorides, but this transformation could not be applied to cinnamoyl- or 2-(naphthalen-2-yl)acetyl fluorides (Scheme 2-5). Unfortunately, substrates containing nitrogen, chlorine or bromine were also unable to participate in this protocol.

Scheme 2-5. Negative Scope of Acyl Fluorides.^a

aReaction conditions: 1 (0.2 mmol), 2a (0.3 mmol), Pd(acac)2 (10 mol %), DPPE (15 mol %), KF (0.3 mmol), in toluene/DMSO (9:1, 0.1 M) at 140 °C for 1 h under Ar. Isolated yields.

34 2-2-3 Scope of Alkylboranes

Scheme 2-6. Scope of Alkylboranes.^a

aReaction conditions: 1a (0.2 mmol), 2 (0.3 mmol), Pd(acac)2 (10 mol %), DPPE (15 mol %), KF (0.3 mmol), in toluene/DMSO (9:1, 0.1 M) at 140 ºC for 1 h under Ar. Isolated yields.

We then studied a range of B-alkyl-9-BBN reagents 2 using benzoyl fluoride (1a) as an electrophile (Scheme 2-6). The various alkylboranes prepared from the hydroboration of the corresponding alkenes were sufficiently tolerant to afford the decarbonylative products 3ab-3af under optimized conditions. The opposite combination of substrates 1 and 2 also afforded the same compounds 3ha

35

and 3ia in better yields. Relatively, B-alkyl-9-BBN bearing pentafluorophenyl functionality provided the corresponding product 3ag in 72% yield. Furthermore, alkylboranes containing ester and silyloxy groups were also suitable for decarbonylative alkylation, giving the desired products 3aj and 3ak in 76% and 85% yields, respectively. Unfortunately, the present protocol could not be applied to secondary alkylboron reagents such as 2-methyl-3-phenylpropyl and cyclooctyl derivatives, probably due to their steric hindrance effects.

2-2-4 Applications

To demonstrate the usefulness of this protocol, we conducted the reactions of B-alkyl-9-BBN prepared from safrole with acyl fluorides derived from probonecid and febuxostat (Scheme 2-7).

However, only 23% and 37% of the decarbonylative alkylation products were isolated, respectively.

Scheme 2-7. Substrate of Pharmaceutical-Derived Acyl Fluorides.^a

aReaction conditions: 1 (0.2 mmol), 2a (0.3 mmol), Pd(acac)2 (10 mol %), DPPE (15 mol %), KF (0.3 mmol), in toluene/DMSO (9:1, 0.1 M) at 140 °C for 1 h under Ar. Isolated yields.

Finally, the chemoselective reaction of substrates bearing ester and acyl fluoride functional groups were examined (Scheme 2-8). Under our reaction conditions, the acyl fluoride moiety reacted exclusively with 2a to deliver the corresponding alkylated products 3sa and 3ta in 78% and 65% yields, respectively, while the ester moiety remained intact. Furthermore, the product 3ta could be used as a reactant for further alkylation under the previously reported reaction conditions of the ester.¹⁰ The resultant double alkylated product 5 was obtained in 71% yield.

36

Scheme 2-8. Chemoselective and Sequential Alkylation.

aReaction conditions: 1 (0.2 mmol), 2a (0.3 mmol), Pd(acac)2 (10 mol %), DPPE (15 mol %), KF (0.3 mmol), in toluene/DMSO (9:1, 0.1 M) at 140 ºC for 1 h under Ar. Isolated yields. ^b130 ºC.

2-2-5 Mechanistic Studies

2-2-5-1 Proposed Mechanism

According to a related mechanistic study by Schoenebeck¹⁶ and Xie,¹⁷ a reaction mechanism for the present decarbonylative alkylation of acyl fluorides has been proposed (Scheme 2-9). First, oxidative addition of the C(acyl)-F bond to Pd(0) generates the acyl(fluoro)palladium(II) complex A. Then, transmetalation of A with alkylboranes 2 generates the acyl(alkyl)palladium(II) complex B. As the fate of fluorides, B,B-F2-9-BBN is formed,¹⁸ suggesting that the addition of external KF is required for this transformation. In turn, complex B undergoes decarbonylation to give complex C. Most of the reported decarbonylation reactions of carboxylic acid derivatives with alkylmetals are carried out using hemilabile bidentate ligands such as DCYPE.^9-12 Subsequently, CO extrusion from C forms the aryl(alkyl)palladium(II) complex D. Finally, reductive elimination affords the alkylated products 3 and the Pd(0) catalyst is regenerated.

It is worth noting that even at 140 ºC, which is needed for decarbonylation, no β-hydride elimination occurred at all. We carefully monitored the reaction of 1a with 2h by ¹H NMR measurements, and in addition to the desired coupled product, several unidentified byproducts were observed. However, only trace amounts of the corresponding alkenes derived from β-hydride elimination of the alkylboron reagent were observed. It is possible that one of the phosphine atoms of the hemilabile bidentate ligand DPPE opens a coordination site for decarbonylation. Another possible pathway is that the decarbonylation occurs prior to transmetalation from A to the aryl(fluoro)-palladium(II) complex.

Previous density functional theory (DFT) calculations of Pd-catalyzed decarbonylative

37

transformations of acyl fluorides have shown that pre-decarbonylation is kinetically unfavorable due to the high activation free energy barrier, suggesting that transmetalation might occur prior to decarbonylation.^16,17

Scheme 2-9. Proposed Mechanism.

2-2-5-2 Studies of Side Reaction

Exploration of the range of acyl fluorides containing electron-donating groups with DPPE as ligand, such as 1c, 1d, 1e, and 1f, yielded the unexpected byproduct 3aa. In order to investigate the mechanism of the formation of 3aa, control experiments were conducted using 1f (Table 2-7). When DCYPE was employed instead of DPPE under optimal conditions, 3aa was not detected (entry 2), revealing that DPPE was the source of the phenyl group. However, in the absence of acyl fluoride 1h, only trace amounts of 3aa were observed. Most importantly, in the absence of 2a, a considerable amount of benzoyl fluoride 1a was detected by ¹⁹F{¹H} NMR (entry 4).

Based on these results and related reports,¹⁹ we proposed that the formation of 3aa is due to the metathesis between acyl fluoride and DPPE. Under palladium catalysis, the phenyl group in DPPE was exchanged with the aryl group in acyl fluorides to form benzoyl fluoride 1a and the following reaction of 1a with 2a yielded the corresponding product 3aa. From this point of view, this transformation is very attractive for acyl fluoride applications.

38

Table 2-7. Control Experiments of Side reaction.^a

entry conditions Yield of 3aa (%)^b

1 none 15

2 DCYPE instead of DPPE 0

3 without 1f trace

4 without 2a 0

aReactions were carried out with 1f (0.2 mmol, 1.0 equiv), 2a (0.3 mL, 1.0 M in toluene, 0.3 mmol, 1.5 equiv) and Pd(acac)2 (0.02 mmol, 10 mol %), DPPE (0.03 mmol, 15 mol %), KF (0.3 mmol, 1.5 equiv) in toluene (1.5 mL), DMSO (0.2 mL) at 140 °C for 20 h. ^bGC yields, using n-dodecane as the internal standard.

39

2-3 Summary

In this Chapter, the Author has developed an unprecedented technique for the palladium-catalyzed decarbonylative alkylation of acyl fluorides with various alkylboranes. In this work, the Author hypothesized that the palladium catalyst would be far superior to nickel in terms of retarding β-hydride elimination. This method provided a new approach to the formation of diverse C(sp²)-C(sp³) bonds via decarbonylation and emphasized its superior functional group tolerance. The metathesis reaction between acyl fluorides and DPPE was also studied.