Caffeine-derived-iron catalyzed carbonyl-ene and Diels-Alder reactions and development of an NHC-diol ligand family
Thè se
Di Meng
Doctorat en chimie
Philosophiæ doctor (Ph. D.)
Qué bec, Canada © Di Meng, 2018
Caffeine-derived-iron catalyzed carbonyl-ene and Diels-Alder reactions and development of an NHC-diol ligand family
Thè se
Di Meng
Sous la direction de :
Thierry Ollevier, directeur de recherche RÉ SUMÉ
Cette thè se de doctorat met en é vidence l'utilisation de catalyseurs de fer qui pré sentent de nombreux avantages par rapport aux autres mé taux de transition. En effet, le fer est moins coû teux, respectueux de l’environnement et présente des activités catalytiques inté ressantes. Du fait de ces caracté ristiques, la catalyse au fer a connu un ré el essor ces 15 derniè res anné es. Cette thè se pré sente la ré action de type carbonyl-è ne intermolé culaire catalysé e par des sels de fer(II) et de fer(III), utiles pour leur rôle d’acides de Lewis, en employant plusieurs alcè nes avec le 3,3,3- trifluoropyruvate d'é thyle. Les sels de FeII, notamment FeCl2, Fe(OAc)2, Fe(NTf)2, Fe(ClO4)2·6H2O,
Fe(BF4)2·6H2O et Fe(OTf)2, ont é té utilisé s pour catalyser cette transformation. Un systè me efficace utilisant le Fe(BF4)2 anhydre a é té dé veloppé pour catalyser la ré action carbonyl-è ne intermolé culaire de multiples alcè nes avec le 3,3,3-trifluoropyruvate d’éthyle, et aussi la ré action carbonyl-è ne intramolé culaire du (S)-citronellal. Des rendements entre 36-87% en produits-è ne, soit des alcools homoallyliques et de produits de cyclisation du citronellal ont été obtenus par l’utilisation de diffé rents alcè nes disubstitué s. Les carbè nes N-hé té rocycliques (NHC) sont reconnus comme des ligands prometteurs en catalyse avec des mé taux de transition. Trois sels de xanthinium dé rivé s de la café ine ont é té utilisé s comme pré curseurs NHC pour dé velopper des complexes fer-ligand NHC pour les ré actions carbonyl-è ne intra- et intermolé culaires. Les conditions optimales ont é té é tudié es, notamment le choix du sel de fer, du solvant, de la charge catalytique et du contreanion.
Fe(OTf)2 est apparu comme le meilleur catalyseur lorsque complexé au ligand NHC dé rivé du sel de xanthinium café ine mé thylé . Avec [NHC-Fe]2+(SbF6)22− comme catalyseur, des rendements de 22% à 99% en alcools homoallyliques ont é té obtenus pour la ré action carbonyl-è ne en employant divers énophiles et le trifluoropyruvate d’éthyle. De plus, NHC-FeIIICl2[SbF6] s’est avéré être un catalyseur efficace et sélectif pour la transformation du citronellal en produit désiré, l’isopulé gol.
L’aspect recyclable du sel de xanthinium dé rivé de la café ine lié au Fe(OTf)2 a é té é valué dans la ré action de Diels-Alder en employant des solvants verts, comme le dimé thyl carbonate. Le catalyseur a pu ê tre recyclé cinq fois et des rendements identiques ont é té obtenus. Diffé rents substrats ont é té testé s dont des composé s dicarbonylé s bidendates, cé tones, aldé hydes et esters. Les ligands NHC alkoxylé s ont é té dé veloppé s comme famille é mergente de ligands dans les réactions d’addition conjuguées énantiosélectives. Enfin, de nouveaux ligands NHC-diol ont é té synthé tisé s et testé s dans la ré action carbonyl-è ne. Ces derniers sont prometteurs en catalyse asymé trique et notamment en catalyse utilisant des mé taux de transition.
iii
ABSTRACT
Iron has many advantages compared to other transition metals in homogeneous catalysis, such as relatively cheap price, eco-friendly, good catalytic activities. Hence, these features boosted the development of iron catalysis since 15 years ago. In this thesis, various iron salts including FeII and FeIII were examined as Lewis acid catalysts in the intermolecular carbonyl-ene reaction of various alkenes and ethyl 3,3,3-trifluoropyruvate. FeII salts, such as FeCl2, Fe(OAc)2, Fe(NTf)2,
Fe(ClO4)·6H2O, Fe(BF4)2·6H2O, Fe(OTf)2, were found to be effective in catalyzing the reaction. An anhydrous Fe(BF4)2 catalytic system was developed for both of an intermolecular carbonyl-ene reaction of various alkenes and ethyl 3,3,3-trifluoropyruvate and an intramolecular carbonyl-ene reaction of (S)-citronellal. The ene-products, i.e. homoallylic alcohols, were afforded in 36-87% yields giving a scope of various with 1,1-disubstituted alkenes and the cyclization of citronellal. N- heterocyclic carbenes (NHC) are recognized as promising ligands in transition metals catalysis. Three caffeine-derived xanthinium salts were used as NHC precursors to transition metals iron for developing an NHC-iron catalyst in the intermolecular carbonyl-ene reaction and the intramolecular carbonyl-ene reaction of citronellal. Optimized conditions were developed from the screening of iron salts, solvents, catalyst loading and counter anions. Fe(OTf)2 was found to efficiently catalyze the reaction while complexed with NHC ligand derived from methylated caffeine xanthinium salt.
Caffeine-derived-[NHC-Fe]2+(SbF6)22− catalyzed carbonyl-ene reaction of various enophiles with ethyl trifluoropyruvate afforded 22-99% yields in homoallylic alcohols. NHC-FeCl2[SbF6] was efficiently and selectively used as a catalyst to convert citronellal into the desired isopulegol.
Caffeine-derived xanthinium salt was designed with Fe(OTf)2 as a recyclable catalyst for Diels-Alder reaction in dimethyl carbonate used as a green solvent. Several other green solvents were examined to further study the application of green solvents in organic synthesis. The catalyst, derived from a caffeine-derived xanthinium salt and Fe(OTf)2, was recycled up to five times, while maintaining the same level of yields for the Diels-Alder reaction and recyclability. A relative large scope of substrates including bidentate dicarbonyl compounds, ketones, aldehydes, and esters were tested. Alkoxyl- NHC ligands were developed as a rising family of ligands in enantioselective conjugate addition. A series of new NHC-diol ligands were designed and tested in the carbonyl-ene reaction. These newly developed ligands are promising systems in asymmetric catalysis and transition metal catalysis.
iv
Table of Contents
RÉ SUMÉ ...... iii
ABSTRACT ...... iv
Table of Contents ...... v
List of Schemes ...... ix
List of Figures ...... xiv
List of Tables ...... xvi
List of Abbreviations ...... xviii
ACKNOWLEDGEMENTS...... xx
Chapter one Introduction of iron and NHC-iron catalysis & Development of caffeine-derived NHC- Fe catalyst in carbonyl-ene reaction ...... 1
1.1 Iron catalysis and the objectives of the research ...... 1
1.2 General objectives of the thesis ...... 12
1.3 N-Heterocyclic carbene (NHC) chemistry ...... 12
1.4 The development of Iron-NHC chemistry ...... 16
1.4.1 Carbon−halide bond ...... 17
1.4.2 Carbon−carbon double and triple bond ...... 18
1.4.3 Carbon−heteroatom double bond ...... 20
1.5 The development of carbonyl-ene reaction ...... 21
1.5.1 Definition and advantages of carbonyl-ene reaction ...... 21
1.5.2 Development of catalysts ...... 22
1.5.3 Development of catalysts for the ethyl trifluoropyruvate-ene reaction ...... 23
1.6 Synthetic organic chemistry of caffeine ...... 24
1.7 References ...... 27
v
Chapter two Development of iron catalysis in carbonyl-ene reaction ...... 37
2.1 Iron salts as Lewis acid for carbonyl-ene reaction ...... 37
2.2 Development of organo-iron catalyst for carbonyl-ene reactions ...... 44
2.3 Development of caffeine-derived NHC with iron as catalyst for carbonyl-ene reactions ...... 48
2.4 Mechanism of caffeine-derived NHC-iron catalysts for carbonyl-ene reactions and the withdrawing property of caffeine’s pyrimidine-dicarbonyl structure ...... 63
2.5 Quantitative determination of the free carbene from caffeine-derived xanthinium salt─ligand 1 ...... 67
2.6 Synthesis of caffeine-derived NHC-Iron complexes ...... 72
2.7 References ...... 76
Chapter three Development of caffeine-derived imidazolium-Fe(OTf)2 combined catalysts in the Diels-Alder reaction ...... 80
3.1. Introduction and background ...... 80
3.1.1 Catalyzed Diels-Alder reactions ...... 80
3.1.2 Iron-catalyzed Diels-Alder reactions ...... 82
3.2. Development of caffeinium iron catalyzed Diels-Alder reactions ...... 84
3.2.1 Development of the catalyst ...... 84
3.2.2 Optimization of the reaction ...... 85
3.2.3 Optimization of the solvent ...... 87
3.2.4 Recyclability test of catalyst 1 ...... 88
3.2.5 Reaction scope ...... 89
3.2.6 Application of the catalyst in other reactions ...... 92
3.3. References ...... 95
Chapter four Development of NHC-diol ligands for the enantioselective carbonyl-ene reaction ....98
4.1 Introduction and background ...... 98
4.2. Synthesis of carbene-diols ligand family ...... 106
vi
4.2.1 Preparation of (R)-tert-butyloxirane ...... 106
4.2.2 Preparation of ligands 6−11 ...... 107
4.3 Catalytic tests in the enantioselective carbonyl-ene reaction ...... 111
4.3.1 Catalytic test of carbene-diols ligands with iron in carbonyl-ene reaction ...... 111
4.3.2 Catalytic tests of a few C2 symmetric ligands in the iron in carbonyl-ene reaction ...... 113
4.4. References ...... 119
Chapter five Conclusion and Perspectives ...... 121
5.1 Iron-catalyzed carbonyl-ene reaction...... 121
5.2 Caffeine-derived NHC-iron-catalyzed carbonyl-ene reaction ...... 122
5.3 Caffeine-derived imidazolium-Fe(OTf)2 catalyzed Diels-Alder reaction ...... 123
5.4 Carbene-diol family ligand ...... 124
Chapter six Experimental section ...... 126
6.1 General information ...... 126
6.2 Experimental part of the carbonyl-ene reaction ...... 126
6.2.1 General procedure for the carbonyl-ene reaction of alkenes with ethyl 3,3,3- trifluoropyruvate by using unhydrous Fe(BF4)2: ...... 126
6.2.2 General procedure for performing NHC-iron carbonyl-ene reaction of alkenes with ethyl 3,3,3-trifluoropyruvate or cyclization of (S)-(−)-citronellal: ...... 127
6.2.3 Spectral data of ligands (1−5) in carbonyl-ene reaction...... 128
6.3 Experimental part of Diels-Alder reaction ...... 158
6.3.1 General procedure for Diels-Alder reaction ...... 158
6.3.2 General procedure for caffeine-derived imidazolium-Fe(OTf)2 catalyzed acetylation of benzyl alcohol and acetic anhydride...... 159
6.3.3 Spectral data of the products of Diels-Alder reaction ...... 159
6.3.4 Copies of NMR spectra of the products of Diels-Alder reaction ...... 165
6.4 Experimental part of NHC-diol ligands...... 180
vii
6.4.1 General procedure for NHC-diol ligand catalyzed carbonyl-ene reaction ...... 180
5.4.3 Copies of NMR spectra of NHC-diol ligands ...... 190
6.5 References ...... 215
viii
List of Schemes
Scheme 1−1 The discovery of Fenton's reagent ...... 3
Scheme 1−2 The mechanism of Fenton’s reagent ...... 3
Scheme 1−3 The Haber-Bosch process ...... 5
Scheme 1−4 The Fischer-Tropsch process ...... 5
Scheme 1−5 Reppe's alcohol synthetic route ...... 6
Scheme 1−6 FeCl3 catalyzed allylation of aldehydes and allyltrimethyl silane ...... 7
Scheme 1−7 FeCl3·6H2O catalyzed Michael addition ...... 7
Scheme 1−8 Iron catalyzed enantioselective Michael addition by the assistance of a chiral auxiliary ...... 8
Scheme 1−9 Iron catalyzed cycloaddition of diallenes with carbon monoxide...... 8
Scheme 1−10 Iron catalyzed isomerization of allylic alcohols ...... 8
Scheme 1−11 A bis-imino-pyridyl iron complex catalyzed polymerization ...... 9
Scheme 1−12 Trifluoromethylation of terminal olefins by using FeCl2 ...... 9
Scheme 1−13 Aerobic oxidative coupling of naphthols by using a salen-iron complex ...... 10
Scheme 1−14 Asymmetric O−H insertion by using a chiral spiro-bisoxazoline-iron complex ...... 10
Scheme 1−15 Oxidative esterification of aromatic aldehyde by an iron-NHC catalyst ...... 11
Scheme 1−16 The hydrogenation of alkynes to alkenes by using a phosphine-iron complex ...... 11
Scheme 1−17 Second generation of Grubbs’s catalyst with a NHC as a ligand for alkene metathesis reaction ...... 13
Scheme 1−18 A cyclopropanation study of diazoacetate and toluene by Eduard Buchner ...... 14
Scheme 1−19 The first NHC-Iron complex by Ö fele ...... 16
Scheme 1−20 NHC-Fe in Kumada-type cross-coupling reaction ...... 17
Scheme 1−21 Regioselective allylic alkylation by using NHC-Fe ...... 18
Scheme 1−22 Selective NHC-iron catalyzed borylation ...... 18
Scheme 1−23 Hydroboration of alkenes catalyzed by a Py-bisNHC-Fe complex ...... 19
ix
Scheme 1−24 Hydrogenation by Py-bisNHC-Fe complex ...... 19
Scheme 1−25 Atom transfer radical polymerization catalyzed by bisNHC-FeBr2 ...... 19
Scheme 1−26 Oxidation of aromatic hydrocarbons catalyzed by Py-NHC-Fe complex ...... 20
Scheme 1−27 Carbometalation with methyl Grignard reagents catalyzed by NHC-iron ...... 20
Scheme 1−28 Hydrosilylation by NHC-Fe catalysis ...... 21
Scheme 1−29 Bis-oxazoline-Cu catalyzed glyoxylate-ene reaction of an aldehyde and a ketone .22
Scheme 1−30 Carbonyl-ene reaction of ethyl trifluoropyruvate with -methylstyrene ...... 23
Scheme 1−31 Alkylation of caffeine ...... 25
Scheme 1−32 The application of bis-NHC-PdI2 in Suzuki cross-coupling reaction ...... 26
Scheme 2−1 Metal triflate catalyzed carbonyl-ene reaction of ethyl trifluoropyruvate ...... 38
Scheme 2−2 Iron salts catalyzed carbonyl-ene reaction ...... 38
Scheme 2−3 Optimization of solvents ...... 40
Scheme 2−4 Carbonyl-ene reaction of various alkenes and ethyl trifluoromethyl pyruvate ...... 41
Scheme 2−5 Intramolecular carbonyl-ene reaction of (S)-(+)-citronellal...... 43
Scheme 2−6 Caffeine as reactant and a stabilizing ligand ...... 45
Scheme 2−7 Caffeine act as a ligand to iron salts in carbonyl-ene reaction ...... 46
Scheme 2−8 Iron-Bolm’s ligand catalyzed Mukaiyama Aldol reaction ...... 47
Scheme 2−9 Bipyridine as a ligand to ironII triflate in catalyzing carbonyl-ene reaction ...... 47
Scheme 2−10 Formation of caffeinium salts by alkylation of caffeine ...... 49
Scheme 2−11 Ligand 1-derived NHC-Fe(OTf)2 catalyzed carbonyl-ene reaction ...... 50
Scheme 2−12 Examination of different solvents in ligand 1-derived-NHC-iron catalyzed carbonyl- ene reaction...... 52
Scheme 2−13 Ligand 2-derived-NHC with different iron salts in catalyzing carbonyl-ene reaction 53
Scheme 2−14 Study of solvent effect using ligand 2-derived-NHC iron catalyzed carbonyl-ene reaction ...... 54
Scheme 2−15 Ligand 2 derived-NHC with iron salts of non-coordinating anions ...... 55
x
Scheme 2−16 Development of ligand 2-derived-NHC-FeIII in carbonyl-ene reaction ...... 57
Scheme 2−17 Testing different alkenes with the catalyst of ligand 2 derived-NHC-Fe(SbF6)2 in carbonyl-ene reaction of trifluoropyruvate ...... 58
Scheme 2−18 Further optimization of using substrate mehylenecyclohexane by ligand 1 and Fe(OTf)2 ...... 60
Scheme 2−19 Further optimization of using substrate methylenecyclohexane with ligand 2 and Fe(SbF6)2 ...... 61
Scheme 2−20 Development of NHC-iron catalyst in the intramolecular carbonyl-ene reaction of the cyclization of citronellal ...... 62
Scheme 2−21 Comparison of ligand 1 with ligand 4 and ligand 2 with ligand 5 in NHC-iron catalyzed carbonyl-ene reaction ...... 65
Scheme 2−22 Synthesis protocol of a free carbene, 1,3-bis(2,4,6-trimethylphenyl)imidazole-2- ylidene, IMes ...... 67
Scheme 2−23 Structures of caffeine and ligand 1 ...... 68
Scheme 2−24 Determination of caffeine derived NHC by method 1 ...... 70
Scheme 2−25 Determination of caffeine derived NHC using method 2 ...... 70
Scheme 2−26 Determination of caffeine derived NHC using method 3 ...... 71
Scheme 2−27 Synthesis of caffeine-derived NHC-iron complexes ...... 73
Scheme 3−1 An example of asymmetric Diels-Alder reaction ...... 80
Scheme 3−2 Prochiral center as stereochemical control element in N-acyloxazolinone in D-A .....80
Scheme 3−3 FeIII as Lewis acid catalyzed Diels-Alder reaction ...... 83
Scheme 3−4 Preparation of caffeine derived-imidazolium salts and the iron-caffeinium catalysts 1−4 ...... 84
Scheme 3−5 Optimization of Diels-Alder reaction ...... 86
Scheme 3−6 Recyclability test using catalyst C1 ...... 88
Scheme 3−7 Caffenium iron catalyst catalyzed carbonyl-ene reaction ...... 93
Scheme 3−8 FeCl3 promoted acetylation of benzyl alcohol acetic anhydride...... 94
Scheme 3−9 Application of caffeine derived imidazolium-Fe(OTf)2 catalyst in acetylation of benzyl alcohol and acetic anhydride ...... 94
xi
Scheme 4−1 Arnold’s NHC-Cu-alkoxide complex ...... 99
Scheme 4−2 Formation of mono-NHC-Cu by transmetallation of NHC-Li alkoxide with CuCl2 ...... 99
Scheme 4−3 Enantioselective conjugate addition of diethyl zinc to cyclohexanone ...... 99
Scheme 4−4 Synthesis of a bidentate alkoxyl-NHC-Ru complex in Hoveyda’s group ...... 100
Scheme 4−5 Alkoxyl-NHC-Ru complex catalyzed metathesis...... 100
Scheme 4−6 Alkoxyl-Cu-NHC generated through transmetalation with relative alkoxyl-Ag-NHC complex in catalyzing allylic alkylation ...... 101
Scheme 4−7 Alkoxyl-Rh/Ir-NHC complexes in catalyzing hydrosilylation reaction ...... 101
Scheme 4−8 Ionic liquid incorporating hydroxyl group by using chiral L-valine ...... 102
Scheme 4−9 L-Leucinol derived NHC-alkoxyl bidentate ligands with Cu(OTf)2 in enantioselective conjugate addition ...... 102
Scheme 4−10 Synthesis of NHC-diol ligands by Wilhelm ...... 103
Scheme 4−11 Synthesis of NHC-diol ligands by epoxide-opening reaction ...... 104
Scheme 4−12 NHC-diol-copper and -iron catalyzed asymmetric addition of an enantioselective conjugate addition ...... 104
Scheme 4−13 NHC-diol ligand modified Fe3O4/Pd nanoparticle in catalyzing α-arylation of ketones ...... 105
Scheme 4−14 Fe-Bolm’s ligand catalyzed Mukaiyama aldol reaction ...... 105
Scheme 4−15 Pyridine-bisNHCs-iron complexes catalyzed Kumada-type cross-coupling...... 106
Scheme 4−16 Preparation of (R)-tert-butyloxirane by oxidation and resolution...... 107
Scheme 4−17 Synthesis of NHC-diol ligand from (S,S)-1,2-diphenyl-ethylene trans-diamine...... 108
Scheme 4−18 Synthesis of NHC-diol ligand from enantiomeric enriched (R,R)-1,2- diaminocyclohexane ...... 109
Scheme 4−19 Epoxide opening reaction of (R)-tert-butyloxirane by imidazole ...... 109
Scheme 4−20 Synthesis of ligands 4-7 ...... 110
Scheme 4−21 Complexation of ligand 6 with Ag2O and its proposed structure ...... 111
Scheme 4−22 NHC-diol Fe(OTf)2 catalyzed carbonyl-ene reaction ...... 112
xii
Scheme 4−23 Iron-C2 symmetric ligands catalyzed carbonyl-ene reaction ...... 114
Scheme 4−24 Synthesis of “a porphyrin-inspired ligand” and a desired/expected NHC precursor base on the ligand ...... 117
Scheme 4−25 Catalytic tests in carbonyl-ene reaction and epoxide-opening reaction ...... 118
xiii
List of Figures
Figure 1−1 Significant NHC-metal complexes in the development of NHC history ...... 15
Figure 1−2 BisNHC-Fe complexes in cross-coupling reactions ...... 17
Figure 1−3 Other NHC-Fe complexes in the reduction of double bonds ...... 21
Figure 1−4 The development of catalysts for the trifluoropyruvate-ene reaction ...... 23
Figure 1−5 The first caffeine derived-NHC-Hg complex ...... 25
Figure 1−6 Development of various caffeine derived-NHC-metal complexes...... 26
Figure 2−1 Reaction scope of different of alkenes ...... 42
Figure 2−2 Postulated mechanism ...... 43
Figure 2−3 Reaction scope of ligand 2 derived-NHC-Fe(SbF6)2 catalysis in carbonyl-ene reaction ...... 59
Figure 2−4 Postulated mechanism of the NHC-Fe catalyzed carbonyl-ene reaction of ethyl trifluoropyruvate and -methylstyrene ...... 64
Figure 2−5 Electronic property of caffeine derived-NHC-Fe bonding orbital ...... 66
Figure 2−6 1H NMR of caffeine and ligand 1 ...... 69
Figure 2−7 Determination of caffeine derived NHC using method 1 ...... 70
Figure 2−8 Determination of caffeine derived NHC using method 2 ...... 71
Figure 2−9 Determination of caffeine derived NHC by method 3 ...... 72
Figure 2−10 Caffeine derived-NHC generated without Fe(OTf)2 in CH2Cl2 ...... 74
Figure 2−11 Caffeine derived-NHC generated with Fe(OTf)2 in CH2Cl2 ...... 75
Figure 3−1 FeII/FeIII-catalyzed enantioselective Diel-Alder reaction ...... 82
Figure 3−2 Iron complexes catalyzed Diels-Alder reaction ...... 83
Figure 4−1 Modified alkoxyl-NHC by 2,6-di-isopropyl group or adamantyl group ...... 101
Figure 4−2 Synthesis of NHC-diol ligands developed by Wilhelm ...... 103
Figure 4−3 Pyridine bisNHC-diol type of ligands developed in literature ...... 106
Figure 4−4 Application of 1,2-diphenyl-ethylene trans-diamine ...... 108
xiv
Figure 4−5 NHC-diol ligands 6b-11b ...... 111
Figure 4−6 Postulated coordination of Box-FeII and Py-Box-FeII ...... 116
xv
List of Tables
Table 1−1 The development of catalysts for the trifluoropyruvate-ene reaction ...... 24
Table 2−1 Screening of a few metal triflate Lewis acids ...... 38
Table 2−2 Screening of iron salts in carbonyl-ene reaction ...... 39
Table 2−3 Screening of solvents in carbonyl-ene reaction ...... 41
Table 2−4 Caffeine as a ligand in iron-catalyzed carbonyl-ene reaction ...... 46
Table 2−5 Optimization of Bipyridine-Fe(OTf)2 catalyzed carbonyl-ene reaction ...... 48
Table 2−6 Optimization of Ligand 1-derived NHC-Fe(OTf)2 catalyzed carbonyl-ene reaction ...... 51
Table 2−7 Screening of different solvents in ligand 1-derived-NHC-iron catalyzed carbonyl-ene reaction ...... 52
Table 2−8 Screening of different iron salts with Ligand 2-derived-NHC in catalyzing carbonyl-ene reaction of ethyl trifluoromethyl pyruvate ...... 53
Table 2−9 Screening of various solvents using ligand 1-derived-NHC iron catalyzed carbonyl-ene reaction ...... 54
Table 2−10 Development of anhydrous Fe(BF4)2 and Fe(SbF6)2 with ligand 1-3 ...... 56
Table 2−11 Variation of the composition of the desirable catalytic species in the reaction by using FeCl3 and AgSbF6 with ligand 2 ...... 57
Table 2−12 Further optimization for the reaction of mehylenecyclohexane using ligand 1 and Fe(OTf)2 ...... 60
Table 2−13 Further optimization of using substrate mehylenecyclohexane with ligand 2 and Fe(SbF6)2 ...... 61
Table 2−14 Optimization of NHC-iron catalyst in the intramolecular carbonyl-ene reaction of the cyclization of citronellal ...... 63
Table 2−15 Synthesis of caffeine-derived NHC-iron complexes...... 73
Table 3−1 Optimization of the Diels-Alder reaction ...... 86
Table 3−2 Screening of selected “green” solvents ...... 88
Table 3−3 Recyclability in 5 runs and yield of each run ...... 89
Table 3−4 Reaction scope ...... 90
xvi
Table 3−5 Less reactive substrates in the reaction scope ...... 92
Table 3−6 Caffenium iron catalyst catalyzed carbonyl-ene reaction ...... 93
Table 4−1 Application of NHC-diol ligand with iron in testing carbonyl-ene reaction ...... 112
Table 4−2 Iron-C2 symmetric ligands catalyzed carbonyl-ene reaction ...... 115
xvii
List of Abbreviations
Å angströ m
ATRP atom transfer radical polymerization
Bipy 2,2’-bipyridine
Boc tert-butyloxycarbonyl
Box bis-oxazoline
CO carbonyl
COD 1,5-cyclooctadiene
Cp- cyclopentyl
CPME cyclopentyl methyl ether
D-A Diels-Alder
DCE 1,2-dichloroethane
DCM dichloromethane
DIPP 2,6-diisopropylphenyl
DMC dimethyl carbonate
DME dimethyl ether
DMF dimethylformamide
DMSO dimethyl sulfoxide
EtOAc ethyl acetate ee enantiomeric excess
F−T Fischer−Tropsch process
HPLC high-performance liquid chromatography
IL ionic liquid
IMes 1,3-bis(2,4,6-trimethylphenyl)imidazole-2-ylidene
KHMDS potassium bis(trimethylsilyl)amide
LUMO lowest unoccupied molecular orbital m-CPBA meta-Chloroperoxybenzoic acid
xviii
NHC N-heterocyclic carbene
NMP N-methyl-2-pyrrolidone
NMR nuclear magnetic resonance
NP nanoparticle
Ph phenyl
Pybox pyridine-2,6-bis-oxazoline
TBME tert-butyl methyl ether tBu tert-butyl
THF tetrahydrofuran
Tr triphenylmethane
1-PEBr 1-phenyl ethylbromide
xix
ACKNOWLEDGEMENTS
Life is a story for everyone, and one needs to write it with the help of others. I am immensely grateful to all the people who gave me so much help, patience, courage, enlightenment, and love. All the streams flow together, and it becomes a river. All the support together from those who played significant roles during my doctoral studies bolstered my Ph. D. and the completion of this thesis.
At first, I want to give my deep thanks to my supervisor Professor Thierry Ollevier for so much effort he put into each step in the past four years. For starting from learning the basics of organic chemistry to pursuing a challenging project, it was an impossible mission without the guidance of my professor. I am very grateful for helping me face the challenges in my Ph. D. His rich knowledge and experience, optimism, and creative thinking propelled my study and work.
I deeply thank China Scholarship Council for providing me a four year doctoral scholarship.
I wish to acknowledge all the group members for their help, time and enthusiasm. I feel thankful for Mathieu Lafantaisie for his help during my first year in Qué bec. I will always remember that Dr. Martin Pichette Drapeau spent much time in helping me be familiar with the laboratory techniques. A huge thank-you also to my colleagues Dr. Angela Jalba and Dr. Hoda Keipour, who are such diligent people and shared many responsibilities in the laboratory. A warm hug will be given to Dr. Jamil Kraïem for his trust and the short, but humorous, and relaxed time working together. An enormous thank-you to each of the following awesome people: Mao Li, Dazhi Li, Dandan Miao and Wan Xu for the wonderful time together in our lives and so much communication and sharing and help, Samuel Lauzon, Samuel Cashman-Kadri, Virginie Carreras, Nour Tanbouza, Claire Besnard for passing precious time in lab.
I want to give Mr. Pierre Audet a deep thank for many useful discussions and help on NMR and MS. A sincere thank must be given to the departmental staff: Denyse Michaud, Mé lanie Tremblay, Christian Cô té , Magali Goulet, Marie Tremblay and Jean Laferriè re. I would like to thank Professor John Boukouvalas for his courses. Also, I appreciate Ramesh Muddala for his help in laboratory.
Moreover, I would like to thank Xiaomian, Zhijun, Min, Shuang, Sarah, Ghislain, Asende, Diego, Mauricio, Paolo, Cynthia, et al. Last, my family is always the place full of love and courage where I can restart my journey.
xx
Chapter one
Introduction
1.1 Iron catalysis and the objectives of the research
Iron is situated in the middle of the d-block in periodic table. As the second most abundant metal on earth, it is ubiquitous in the crust of our planet. The principal iron ores are hematite (Fe2O3), magnetite
(Fe3O4), and siderite (FeCO3). There are also other ores such as pyrite (Fe2S2), magnetite (FeTiO3), and ferropericlase (Mg, Fe)O. Pure iron that smelted from iron ores has been found as early as 2,000 B.C. Iron chemistry was advanced in a slow pace for thousands of years until the last century. Important applications of iron in heterogeneous catalysis in modern times include its use in the Haber process for ammonia synthesis and the Fishcher−Tropsch process for producing gasoline.1
Since the past 40 years, beyond doubt, iron has emerged as a rising star in emulating other transition metals especially noble metals in homogeneous catalysis. Comprehensive reviews and in-depth analyses of iron catalysis provided insights of its development and future. A significant article entitled as “Iron Catalyzed Reactions in Organic Synthesis” was published in 2004 by Bolm et al.2 After that, tremendous work took place and a vast number of reviews, highlighted articles, and books have been published. Iron was demonstrated to catalyze a large scope of transformations in the reviews “Iron Catalysis in Organic Synthesis”,3 and “Iron Catalysis in Organic Synthesis: A Critical Assessment of What It Takes To Make This Base Metal a Multitasking Champion” provided a profound analysis of the emerging uses of iron.4 Various aspects of iron have been reviewed such as the application of iron(III) chloride,5 iron-catalyzed cycloaddition reactions,6 cross-coupling reactions,7 enantioselective transformations,8 NHC metal catalysis,9 iron phosphine catalysts,10 application of ferrocenes11 etc. The reasons for the development in iron chemistry were addressed in the reviews above.
The abundance of iron makes it much cheaper than other metals applied in catalysis. Due to the increasing price of the transition metals or other rare metals for the past ten years, the demand for cheap alternative was claiming.12 Moreover, many iron salts are readily available as commercial
1
products13 and methods of incorporating different moieties for tuning electronic or steric properties were discovered and readily converted into relevant applications.3
The electronic configuration of iron is 1s22s22p63s23p64s23d6, and the oxidation state of iron varies from −2 to +6. The broad span makes it useful in reduction and oxidation reaction. Another characteristic was the ability to transfer electrons, which facilitates the coupling reaction or polymerization. Lewis acidity varies from moderate to high, which actually is correlated with the oxidation state. Hence, the modification of the electronic property of iron provides a way for new discoveries.3
Iron is categorized as relatively non-toxic and environmental benign. It was disclosed that none of the biocatalyst evolved from nature derived from noble metals, but iron is extensively used by all forms of life.14 Furthermore, iron is considered as causing “minimum safety concern” in drug developing process and is bearing a much higher tolerable limitation than other metals. Research in iron catalysis was described as “the most rewarding”. Considering that iron covers almost the full scope of organic transformations, the large scale industrial application will be the most economically beneficial. There are enormous opportunities for discovering new methodologies of iron catalysis.
Many remarkable applications of iron chemistry were selected from the literature and reviewed herein to highlight to a clear perspective of its development. Fenton oxidation process is deemed as the first application in the history of iron catalysis. In 1876, H. J. H. Fenton described a colored product from the mixture of tartaric acid, hydrogen peroxide, and a low concentration of ferrous salt. In 1894, Fenton reported the phenomenon that adding alkali metal to an aqueous solution of tartaric acid with certain oxidizing agents in the presence of a FeII+ salt makes the solution turn to a violet colour (Scheme 1−1).15 Further experiments by using standard solutions of tartaric acid, ferrous sulphate, and hydrogen peroxide brought more insight into the phenomenon. A small quantity, but essential as a catalyst, of iron salt was enough to promote the oxidation of tartaric acid with hydrogen peroxide. The formula of the product was given in the article and the mechanism was proposed. Tartaric acid was oxidized by hydrogen peroxide to form dihydroxymaleic acid, which formed a violet colored complex with FeIII. Ferrous salt was the catalyst in the process and was regenerated through oxidation-reduction. Oxygen was involved in forming a peroxide radical with tartaric acid carbon- centered radical (Scheme 1−1).16
2
Scheme 1−1 The discovery of Fenton's reagent
Subsequently, the solution of ferrous salt and hydrogen peroxide was named as “Fenton’s reagent”. In the solution, a disproportionation of hydrogen peroxide took place. The generally accepted mechanism of the Fenton’s reagent was demonstrated (Scheme 1−2). Hydroxyl radical was formed in the oxidation process of Fe2+ to Fe3+ by hydrogen peroxide, and Fe2+ was regenerated by the reduction of another molecule of hydrogen peroxide while producing a hydroperoxyl radical. Two free radicals generated in the process promoted secondary reactions. The HO· has high oxidative power and non-selectively oxidizes organic compounds to carbon dioxide and water. (Scheme 1−2)
Scheme 1−2 The mechanism of Fenton’s reagent
The process was largely indluenced by pH, temperature, concentrations of iron and hydrogen peroxide. The hydroxyl radical reacts with most organic molecules and many inorganic substances at a higher rate than other conventional oxidants such as chlorine, O2, O3, or KMnO4. Hence, due to the strong oxidative power, the application of Fenton’s reagent has been extended to industrial wastewaters with a wide range of contaminants. For example, amino acids were oxidized to the corresponding -ketonacid and aldehydes and carboxylic acids containing one less carbon in the deamination-decarboxylation proceses.17 Fenton’s oxidation has been used to process a wide range
3
of industrial wastewaters such as chemical, pharmaceutical, textile, paperpulp, cosmetic, cork processing wastewaters etc. with remarkable progress in decreasing the toxicity to the environment. Fenton’s reaction was promoted to Fenton’s chemistry involving Fenton-like processes generating hydroxyl radicals catalyzed by different Fe-hydroxides or Fe-oxides, which was known as advanced oxidation processes (AOPs). Various ways of AOPs were applicable such as homogeneous Fenton, heterogeneous Fenton, photo-Fenton, photocatalysis, and ozonation etc.18,19
Haber-Bosch process is a milstone in the history of iron catalysis since it was patented in 1910 and started to produce ammonia in 1913.20 Iron was applied as an heterougeneous catalyst in the synthesis of ammonia from elemental nitrogen and hydrogen.7 This artificial nitrogen fixation process has been called the most important invention in the 20th century. The significance of the invention was discussed from the aspects of population growth, environment changes, biodiversity, economy, industrial production etc.21 The reaction itself is intriguing as it demonstrates the role that iron plays.
High pressure and high temperature were applied for the industrial process, and the iron used was generally magnetite (Fe3O4), doped with irreducible oxides K2O on a supportive alumina or silica (Scheme 1−3). Potassium was a promoter as to donate electrons to the neighboring iron. Gaseous nitrogen and hydrogen are adsorbed on iron particles. Kinetic studies show that nitrogen- chemisorption on the surface of iron is the rate-limiting step. As a result, iron catalyst requires high energy input, as well as the high temperature and high pressure, to increase the reaction rate and keep the adsorption-desorption equilibrium to shift toward the product. In the process, ammonia formed from molecular nitrogen and hydrogen on iron surface.22
The mechanism of understanding how nitrogen N2 reacted, whether N2 dissociated or not is a long- lasting question. It was not able to discriminate the real reaction mechanism in the industrial process based on the kinetic study alone. One mechanism is welcomed. According to it, during the process
N2 triple bonds and single bond of H2 were cleaved and elemental N and H form bonds from N-H,
NH2, to NH3.23 However, the nature of reactive species in the mechanisms proposed is still controversial. Studies of new iron catalysts for nitrogen activation are still ongoing.24 Although the development of Ru and Co catalyst for improving the thermodynamic efficiency and transition metals catalysis under homogeneous and ambient conditions, the catalyst used presently in industry is not
4
different from the one developed a century ago. Iron catalysis is still unshakable in the nitrogen- fixation field nowadays.25
Scheme 1−3 The Haber-Bosch process
While Haber-Bosch process propelled the growth of population in providing fertilizers, the Fischer- Tropsch process (F-T) dealt with the energy demand for fuel. It discovers a route of converting natural resources of carbon monoxide and hydrogen into liquid hydrocarbons in a metal catalyzed heterogeneous catalysis process. Only the metals Fe, Ni, Co, and Ru exhibited suitable catalytic efficiency for the F-T process. But Ni is too selective towards CH4 while Ru suffers from extreme high price. Fe and Co are the choices for industrial scale application. Iron pre-catalyst used in the process are currently prepared by precipitation technology from using iron oxides and doped with Cu, K2O and bound with Si. The pre-catalyst needs to be activitated by reducing under H2 or CO atmosphere. Similar as the Haber-Bosch process, the exact structural composition of the active site in iron-based catalyst system is still ambiguous to determine. The mechanism of the production of hydrocarbons on iron surface in the F-T process is described in Scheme 1−4. Iron oxides were activated by hydrogen H2 and carbon monoxide CO, then on the surface the reaction took place in the order of reactant adsorption, chain initiation, chain growth, chain termination. Finally, the alkanes, alkenes, and oxygenates products were generated.
Scheme 1−4 The Fischer-Tropsch process
5
Haber-Bosch and Fischer-Tropsch processes are famous application of iron-based catalysts in heterogeneous chemistry. The Reppe synthesis of iron catalyzed alcohols from CO and water in 1953 is usually viewed as the prelude to the development of homogeneous iron catalysis (Scheme 1−5).27 Significant advancement of transition metal based homogeneous catalysis started in the second half of the twentieth century.
Scheme 1−5 Reppe's alcohol synthetic route
Chronologically, iron was pioneered in the 1970s in the research of cross-coupling reaction. But the real breakthrough in homogeneous iron catalysis was the iron-porphyrin complex for oxidation reactions after the investigation of the complex.14 In the 1970s, cytochromes P-450, as a unique class of hemoproteins, was known for oxygen transfer and was used to catalyze the hydroxylation.28,29 Single oxygen donor, such as hydroperoxides, peroxy acids, and iodosylbenzene, was believed to behave similarly with iron-porphyrin species as hemoprotein enzyme in the biological process.29 In 1979, two years later after the major advance in asymmetric catalytic epoxidation by Sharpless and Yamada,30 Groves reported the hydroxylation and epoxidation of hydrocarbons catalyzed by a porphyrin-iron complex with iodosylbenzene as an oxygen source.28 During this period, oxygen transfer in blood cell was a research interest.31 A few years later, he disclosed an enantioselective version of epoxidation by using a chiral iron porphyrin catalyst.32 Since then, in the 1980s and 1990s, a slow increase of iron-catalyzed reactions in organic synthesis was observed in various branches. Untill the summary of iron was in a general review by Bolm in 2004, the research of iron was still underrepresented compared to other transition metals.2
For 1980 to 2004, major advancements, such as addition reactions, substitution reactions, oxidation reactions, polimerizations etc. have took place. Iron salts performed mainly as catalysts in the non- asymmetric catalysis, while significant breakthrough was achieved in enantioselective transformations such as in Diel-Alder reactions, 1,3-dipolar cycloadditions and sulfoxidations. The compiled results were highly attractive and surprisingly encouraging for the organic chemistry field.2
FeCl3 and its hexahydrated salt FeCl3·6H2O were used extensively in transformation processes.5 For example, a very effective method of allylation of aldehydes with allyltrimethylsilane was
6
developed by using anhydrous FeCl3, and the reaction led to homoallylic alcohols in high yields within short reaction times for a large scope (Scheme 1−6). Only when using electron-rich aromatic aldehyde substrates, such as p-tolualdehyde and p-anisaldehyde, almost no yield was obtained. An improvised way was used by converting the aldehydes to acetals and the reaction promoted smoothly in high yields.33
Scheme 1−6 FeCl3 catalyzed allylation of aldehydes and allyltrimethyl silane
FeCl3·6H2O was considered as the best catalyst for Michael addition. Excellent results had been obtained for the substrates of cyclic and acyclic β-dicarbonyls to various acceptors at low catalyst loadings. But for specific substrates such as 1,2-disubstituted Z-enones as acceptors, there was no efficiency; while for E-substituted substrates, the reactions afforded mixtures of kinetic diastereomers (Scheme 1−7).34
Scheme 1−7 FeCl3·6H2O catalyzed Michael addition
Chiral auxiliaries were used to develop an iron catalyzed enantioselective Michael addition. An - amino acid amide was used with an -ketone ester to form a chiral enamine which was the nucleophile to methylvinyl ketone under the catalyzing of FeCl3·6H2O. In the early stage of asymmetric iron catalysis, an obtained 73% ee was very encouraging (Scheme 1−8).35
7
Scheme 1−8 Iron catalyzed enantioselective Michael addition by the assistance of a chiral auxiliary
Iron salts such as FeBr2, FeCl2, Fe(acac)3, Fe(DBM)2, FeSO4, Fe(ClO4)2, FeCl2[P(OEt)3]3 etc. were developed in various reations.2 Among which the iron carbonyl compoumds were employed in cycloadditions and isomerization. A five-membered carboxylic ring was formed in a [4+1] reaction of carbon monoxide and diallenes catalyzed by Fe(CO)5. The mechanism revealed a “π-facial coordination” between the diallene substrate and iron, and a metallocyclopentene intermediate was formed from the coordinated substrate. The dialkyly denecyclopentenones were produced in a high yield (Scheme 1−9).36
Scheme 1−9 Iron catalyzed cycloaddition of diallenes with carbon monoxide
Iron carbonyl derivatives are effective in the photocatalytic isomerization of alkenes. The initiation of the process requires thermal or photo assistance. Although high yields could be afforded, selectivity of the reaction towards the desired product is difficult to control. But still the isomerization of allylic alcohols was developed (Scheme 1−10).37
Scheme 1−10 Iron catalyzed isomerization of allylic alcohols
The imine group as a functional ligand to iron such as bis-imino-pyridyl iron complexes were developed in polymerization (Scheme 1−11). Polyethylene is one the most produced polymers
8
products in the world. Iron complexes formed with 2,6-bis-iminopyridyl ligands were proved to be very efficient in the polymerization of ethylene to polyethylene. The efficiencies of such iron catalyst were described as remarkably high since the TOFs were achieved more than 107/h and compared with Ziegler-Natta systems.38 The results obtained with iron have been proven to be superior to those with colbalt species.
Scheme 1−11 A bis-imino-pyridyl iron complex catalyzed polymerization
For the past fifteen years, there was a rapid increase in developing new iron catalyst species and breakthrough in homogeneous catalysis.2-11 New transformations of using iron salts have been achieved. For example, the trifluoromethylation of terminal olefins was catalyzed by FeCl2 between 2-arylvinyltrifluoroborates and Togni’s reagent II (Scheme 1−12).39 The yields afforded were from good to high. The configuration of the products was influenced by the original olefins, and high stereoselectivity and mixtures were both afforded.
Scheme 1−12 Trifluoromethylation of terminal olefins by using FeCl2
An efficient approach to enantiometic enriched binaphthalene-diols using atmospheric oxygen as oxidant was developed while benefited from a salan-iron complexes as catalyst (Scheme 1−13). The high yields and high ee rank the aerobic oxidative coupling of naphthols highly valuable to accessing to chiral ligands and organocatalysts. More interestingly, two different substituted 2-naphthols were applicable to the method.40
9
Scheme 1−13 Aerobic oxidative coupling of naphthols by using a salen-iron complex
A limited number of chiral iron catalyst was reported and proven to be more efficient than other transition metals. However, a family of chiral spiro-bisoxazoline-iron complex was designed and verified as higly efficient catalyst for asymmetric O─H bond insertion reactions by Zhou (Scheme
1−14). It is worth to mention that a common iron salt FeCl2·4H2O was chosen. Mild conditions were developed for a wide range of alcohols and even water with diazo ester compounds.41
Scheme 1−14 Asymmetric O−H insertion by using a chiral spiro-bisoxazoline-iron complex
N-heterocyclic carbene was emerging as a ligand to transition metals.9 Esters are important functional groups in natural and synthetic molecules.42 The direct oxidative esterification of aldehydes provides an alternative to conventional routes.42 Upon analysis of the limitation of the existing methods for synthesizing benzoate derivatives such as reaction conditions, scope, and
10
yields in the literature, an aerobic oxidative esterification of aromatic aldehyde with boronic acids was disclosed by using Fe(OTf)2 with NHC as a catalyst. Up to 97% yields with avarious of aromatic aldehydes and boronic acids (Scheme 1-15).43
Scheme 1−15 Oxidative esterification of aromatic aldehyde by an iron-NHC catalyst
Catalytic hydrogenation of olefins in the presence of iron compound as a catalyst dated back to the 1960s.44 (Pyridyl)diimine type of ligand was often used with iron in the reduction transformations.45 The transformation of alkynes to alkene by transition-metal catalysts is employed in many processes.46 Another type was incorporating phosphine as a coordinating ligand to iron in this transformation. An acridine-based PNP iron complex was proven to catalyze a semi-hydrogenation of alkynes to E-alkenes using H2.47
Scheme 1−16 The hydrogenation of alkynes to alkenes by using a phosphine-iron complex
11
1.2 General objectives of the thesis
Transition metal catalysis is one of research interest in our group. Methodologies for asymmetric Mukaiyama reaction, epoxide openning reaction, hydrosilylation, oxidation, Michal addition, etc. were reported using iron bipyridine type catalysts.48 The objectives of this research are to continue to focus on iron catalysis and to discover new iron catalysts on expending the less “iron-involved” transformations. In one of the projects of this research, carbonyl-ene reaction was chosen due to the less developed situation of iron in this type of carbon bond forming process. Another one was Diels- Alder reaction on the purpose of recycling the catalyst and developing alternative solvent to tranditional methods.
1.3 N-Heterocyclic carbene (NHC) chemistry
This part focuses on the development of an NHC-iron catalyst as a Lewis acid for the carbonyl-ene reaction. The opening includes a short summary of the history of caffeine-derived N-heterocyclic carbene chemistry, iron-NHC chemistry and the development of carbonyl-ene reaction.
Carbene is defined as a neutral divalent carbon containing six valence electrons.49 N-heterocyclic carbenes originally derived from imidazolium salts, was later on diversified in their structure and functionalization.50 N-heterocyclic carbenes have been applied as ligands with late transition metals from group 7 to group 11.51 The most famous application of N-heterocyclic carbene is the second generation of Grubbs’s catalysts for alkene metathesis, arisen from the substitution of phosphine ligands by IMes (NHC) (Scheme 1−1).52 The second generation of Grubbs’s catalyst was the main breakthrough for the development of metathesis, which was demonstrated initially by Y. Chauvin, R. H. Grubbs, and R. R. Schrock, respectively, who shared the Nobel Prize in chemistry in 2005.52 Many crucial historical moments became essential for the development of N-heterocyclic carbenes in history.
N-heterocyclic carbene (NHC) has been a hot topic in organic chemistry for the past two decades since the first successful isolation and characterization of a free crystalline carbene by Arduengo in 1991.53 IThis discovery triggered the enthusiasm of researchers for the role that N-heterocyclic carbenes as ligands could play in. Then, for the past 27 years, intensive research activity was invested in this area.54
12
Scheme 1−17 Second generation of Grubbs’s catalyst with a NHC as a ligand for alkene metathesis reaction
This milestone of Arduengo’s carbene was built on the contribution of many pioneers’ work. In the 1800s, the development of coordination chemistry progressively built up since that period.55 Coordination complexes such as copper vitriol, Prussian blue, etc. were known in 1800s, but the structures of these compounds were unclear. The formula of Prussian is Fe7(CN)18 or
Fe4[Fe(CN)6]3·xH2O. Now we know that the Fe(II) centers are surrounded by six carbon ligands of cyanides in an octahedral configuration while the Fe(III) centers are octahedrally surrounded on average by 4.5 nitrogen atoms and 1.5 oxygen atoms.56-57 The copper vitrol (CuSO4·xH2O) usually has a pentahydrate structure while the copper centers are interconnected with a sulfate anion.25 But in those days, because of the absence of X-ray crystallography technology, the coordination of ligands was hardly understood. The breakthrough in coordination chemistry was made by a Swiss chemist, Alfred Werner, in1893. He correctly proposed the structure of CoCl3(NH3)6 and the geometric isomers of the complex [CoCl2(NH3)6]Cl, in which the six NH3 and two Cl¯ occupy the octahedral vertices of Co3+ and one Cl¯ is dissociated. The geometric configuration and the valence theory laid the foundation of modern coordination chemistry.58-60 Because of this significant work, Alfred Werner won the Nobel Prize in chemistry in 1913.
Another event in coordination chemistry was the introduction of the concept of “Ligand”. In the complexes, the coordinated part of transition metals was commonly recognized as a coordinated ion or molecule rather than a ligand. A contemporary German chemistry named Alfred Stock firstly used the concept of “Ligand”, which was derived from a Latin word “ligare” meaning to bind, referring to the counterpart in the complexes in boron and silicon chemistry in 1916.61 But until 1948, H. Irving and R. J. P. Williams adopted the term, which began to raise the attention and to be well accepted
13
in the academic word.62 The development of coordination chemistry including the concept of “Ligand” was crucial for N-heterocyclic carbene chemistry. Contemporarily, the carbene chemistry was advancing slowly since 1903 at the background of the development of coordination chemistry. Approximately 50 years after the two great chemists’ contribution, the field of metal-carbene and NHC started to be explored.
In 1907, German researcher Eduard Buchner won the Nobel Prize for cell-free fermentation. In 1903, Eduard Buchner and Leon Feldmann published an article about a cyclopropanation study of diazoacetate and toluene (Scheme 1−2). In this experiment 400 grams of distilled toluene and 50 grams of diazoacetate were heated together to give 1 gram of cyclopropanation product. Although the yield was low, inspired by the enlightenment and analysis of the result, Buchner firstly postulated the concept of carbene.63
Scheme 1−18 A cyclopropanation study of diazoacetate and toluene by Eduard Buchner
It was worth mentioning that Chugajev’s NHC-Pt complex, which was synthesized in 1915,64 was believed to be the first N-heterocyclic carbene, but the structure was identified several decades later after much progress in carbene chemistry was achieved by Fischer’s pioneered work.65 After several decades, the methodology was proved to be applicable to the synthesis of NHC complexes.66 The term carbene was introduced in organic chemistry in1954 by Doering and Hoffmann67 and into organometallic chemistry in 1964 by Fischer and Maasboel.68 Doering and Hoffmann disclosed the formation of dichlorocarbene [: CCl2] from chloroform in the presence of KOtBu, followed by its addition to olefins.68 Metal-carbene chemistry started from Fischer’s seminal work with tungsten hexacarbonyl (W(CO)6), which represented constituted the first Fischer-type carbene.69 The fischer- type carbene is an electrophilic heteroatom-stabilized carbene coordinated to low-oxidation state metals, such as Cr, Mo, and W.69 After a decade, another type of carbene complex was introduced by Schrock, in which the metals have high oxidation states.70 Schrock-type carbenes are nucleophilic complexes that show Wittig’s ylide-style reactivity and that have been debated as ylides.71 But N- heterocyclic carbenes were firstly studied by Wanzilick in 1962.72 Starting from the statement that a
14
carbene could change from exhibiting an electrophilic to a nucleophilic character by the influence of its neighbouring substituents, Wanzlick discussed the tendency of dissociation of bis-[1,3-diphenyl- 2-imidazolidine] and pointed out that this was a source of nucleophilic carbene as well as thiamine. However, he questioned the existence of such particularly stable carbene since carbene was very reactive and sensitive to chemicals or moisture, and dimerization.72
However, four years later in 1968, Ö fele synthesized an NHC-Cr(CO)5 complex obtained from heating 1,3-dimethyl imidazolium pentacarbonyl hydrido chromate [HCr(CO)5]¯ (Figure 1−1).73 In the same year, Wanzlick synthesized the bis-NHC-Hg complex by using 1,3-diphenyl imidazolium perchlorate salt with mercury acetate Hg(OAc)2 in DMSO in thermal condition (Figure 1).74 The two complexes of Wanzlick and Ö fele were both identified as the first description of organometallic complexes with NHCs as ligands. Although N-heterocyclic carbenes were used as ligands for Hg and Cr respectively in Öfele and Wanzlick’s cases, the significance of these compounds was questioned, and the question of the existence of such carbene potentially as a ligand still lasted till the isolation of Arduengo’s carbene.
Figure 1−1 Significant NHC-metal complexes in the development of NHC history
All these efforts promoted the development of the early stage of carbene chemistry. After that, a realization of NHCs as potential ligands in homogeneous catalysis75 brought a massive input of research notably by Herrmann76, Enders77, Dixneuf, and Ç etinkaya.78 For the past twenty years, a dramatic increase in research and published articles on N-heterocyclic carbene was noticed.79 The properties of NHCs explained the success.79 NHCs are stronger donor ligands than phosphines and the variation of the nitrogen cyclic ring together with the variation of N-substituents could promote a fine-tuning of the electronic properties of the carbene. The N-substituents could provide a chiral induction when used with transition metals. The N-heterocyclic carbenes could even take part in organic transformations as organocatalysts. NHCs were applied in the medicinal and materials fields,
15
as well as in organic chemistry. Numerous N-heterocyclic carbenes were designed, and research related to the synthesis of NHC-metal complexes, electronic properties, efficiency in many different reactions etc. was performed in a flourishing period.80 The primary applications of NHCs were still their use as supporting ligands with late transition metals from d-block in homogeneous catalysis, especially the cases in NHC-Ru catalyzed metathesis and NHC-Pd catalyzed cross-coupling reactions. There was a quite much effort dedicated to other metals, such as Co, Rh, Ir, Ni, Cu, Ag and Au.81 However, the applications of NHCs with Fe was much less studied but still promising due to the work already published.82-84
1.4 The development of Iron-NHC chemistry
Although the first example of an iron-containing NHC compound was published one year after the examples of NHC-metals synthesized by Wanzlick and Ö fele in 1968 respectively, it was rather slow for the development of iron-NHC complex and catalytic applications in organic chemistry. It was worth highlighting the first NHC-Iron complex in history, which was synthesized by Ö fele by the same procedure as the first NHC-transition metal complex in 1969;85 1,3-dimethyl imidazolium tetracarbonylhydrid-iron [HFe(CO)4]¯ in thermal conditions generated the corresponding NHC-
Fe(CO)4 (Scheme 1−3). Later, iron-NHCs were mainly used for the research of hydrogenase modeling studies and organometallic synthesis.86
Scheme 1−19 The first NHC-Iron complex by Ö fele
Iron has been known for centuries and used in coordination and organometallic chemistry for decades.87,88 The trend was evolving from inorganic iron salts to ligands coordinated to iron as molecular complexes. Comprehensive reviews of iron-catalyzed reactions were reported by Bolm in 20042, Liu in 201089 and Gopalaiah in 20138, and the application iron-NHCs by Darcel in 201390 and Kü hn in 201457. Iron, which has many advantages, such as its abundance on earth, low cost, and environmentally benign character, has emerged as a promising substitute for other late transition metals such as Pd, Rh, and Ru. But there are still limited applications of iron-NHCs in a relatively
16
small reaction scope, in which the most relevant developments were C─C bond formation, reduction reactions and carbon-heteroatom bond formation reactions.88-100
After the first iron-NHC complex synthesis, it took three decades for the first application of iron-NHC in homogeneous catalysis to appear. In 2000, Grubbs used an active Bis-NHC-FeCl2/FeBr2 complex for atom transfer radical polymerization (ATRP).91 Since then, more iron-NHCs were developed for catalytic tests. Usually, iron-NHCs catalyzed reactions were described according to reaction names, but those applications could be categorized by the bond nature involved in the transformation.
1.4.1 Carbon−halide bond
In carbon-carbon bond forming reactions, several types of Iron-NHCs were developed, but only in Kumada-type cross-coupling as shown below (Scheme 1−4).92 These complexes take advantage of a coordination of bis-NHC to one equivalent of iron salt (Figure 1−2).93 Both primary and secondary alkyl halides including fluoride to iodide could be used.92 A simplified mechanism was proposed by Bedford involving single-electron transfer to generate an alkyl radical in the process.94 But in combining with other examples, bis-NHC-iron complexes (Figure 1−2) below showed much affinity toward the carbon−halide bond of sp3 carbon or sp2 carbon.95
Scheme 1−20 NHC-Fe in Kumada-type cross-coupling reaction
Figure 1−2 BisNHC-Fe complexes in cross-coupling reactions
17
1.4.2 Carbon−carbon double and triple bond
Allylic alkylation is a versatile tool for C─C bond formation.96 Iron-NHCs showed somewhat efficient catalytic effectiveness in this transformation. Plietker performed a regioselective allylic alkylation between an isobutyl allyl carbonate and various Michael donors. High regioselectivities were observed (Scheme 1−5).97 The author proposed a mechanism involving the coordination of NHC-Fe with the Carbon-carbon double bond to provide steric demand. Different π-allyl-iron complexes were employed to enhance the catalytic activity and to explain the regioselectivity through a π-allyl mechanism. Later, the author succeeded in using thiols and sulfones as sulfur nucleophiles in the reaction with isobutyl carbonates.98
Scheme 1−21 Regioselective allylic alkylation by using NHC-Fe
In 2010, a selective iron-NHC catalyzed borylation of furans and thiophenes was reported. A methylated piano stool complex activated the 2-position of furan and the addition of an intermediary iron hydride to tert-butyl ethylene allowed the regeneration of the activated iron-furan (Scheme 1−6).99 In 2013, an iron bis-imino-pyridine-bis-NHC catalyzed hydroboration of alkenes was reported, which revealed the activation of carbon-carbon double bond by iron-NHC (Scheme 1−7).100
Scheme 1−22 Selective NHC-iron catalyzed borylation
18
Scheme 1−23 Hydroboration of alkenes catalyzed by a Py-bisNHC-Fe complex
The same type of bis-imino-pyridine-bis-NHC-iron compound was used for the hydrogenation of alkenes in 2012 by Chirik (Scheme 1−8).101 The efficiency of the reaction was attributed to the electron-rich property of low-valent Fe(0) which makes it able to activate the alkene. An isotopic labeling experiment using D2 revealed that isomerization was happening during the process.
Scheme 1−24 Hydrogenation by Py-bisNHC-Fe complex
Atom transfer radical polymerization (ATRP) is an important polymerization method.102 It was the first catalytic application of iron-NHC. Grubbs conducted the first catalytic application of iron-NHC in a structural transformation in 2000. The polymerization of styrene and methyl methacrylate were conducted by Grubbs in 2000 (Scheme 1−9).91 The double bonds were activated by bis-NHC-FeBr2 complex. The rate of polymerization detected was in the highest list for ATRP in organic solvents.
Scheme 1−25 Atom transfer radical polymerization catalyzed by bisNHC-FeBr2
In 2017, a molecular iron-NHC complex was used catalytically for the oxidation of aromatic hydrocarbons (Scheme 1−10).103 A bis-(pyridine-NHC)-iron was used to activate the aromatic p-
19
xylene by hydrogen peroxide (H2O2) in acetonitrile. Mechanistic studies revealed a process that includs either an iron-arene activation or an iron-arene oxide path.
Scheme 1−26 Oxidation of aromatic hydrocarbons catalyzed by Py-NHC-Fe complex
The first NHC-iron catalyzed cyclization of triynes was performed by Okamoto in 2005.104 The NHC- iron generated in situ using a catalytic amount of zinc was very efficient in the process. A reduced Fe (Fe3+ to Fe2+) with triynes formed an iron metallacyclopentadiene complex for the further cyclization. Although the mechanism was not proved from direct evidence, the affinity of iron-NHC to carbon-carbon triple bonds was indicated. Hilt reported a ring expansion of epoxides by using a similar procedure.105 In 2016, methyl phenylacetylene was used as a substrate for the carbometalation with methyl Grignard reagent catalyzed by NHC-iron, as shown below (Scheme 1−11).106
Scheme 1−27 Carbometalation with methyl Grignard reagents catalyzed by NHC-iron
1.4.3 Carbon−heteroatom double bond
Compared with other applications of NHC-Fe complexes, much more progress was obtained in the selective reduction of double bonds of aldehydes, ketones, esters, amides, imines and even sulfoxides by hydrosilylation.107 The most frequently used iron-NHCs had same ligands involving cyclopentadienyl, carbonyl (CO), halide and an NHC. A general scheme is given below (Scheme
20
1−12). Other examples of iron-NHCs that were used in double bond reductions are shown in Figure 1−3.107
Scheme 1−28 Hydrosilylation by NHC-Fe catalysis
Figure 1−3 Other NHC-Fe complexes in the reduction of double bonds
Although more and more iron-NHC complexes were designed, the catalytic applications of iron- NHCs were insufficient in homogeneous catalysis within the scope of iron-catalyzed synthetic chemistry. To explore iron-NHC combination, we were very interested in seeking routes in carbonyl- ene reaction due to the excellent reactivity of iron in complexing/coordinating with carbonyl groups such as in the aldol reaction, Michael addition, Nazarov cyclization, and allylation of carbonyl compounds, as highlighted above.
1.5 The development of carbonyl-ene reaction
1.5.1 Definition and advantages of carbonyl-ene reaction
The carbonyl-ene reaction is defined as a pericyclic process between an alkene bearing an allylic hydrogen and a double bond, in which the migration of the double bond and 1,5-hydrogen shift takes place at the same time. The intermolecular ene reaction and the intramolecular ene reaction are both possible. Carbon-carbon bond forming processes with a carbonyl group often involves a metal reagent as a nucleophile. But the carbonyl-ene reaction is potentially 100% atom-efficient and does not generate waste.108
21
1.5.2 Development of catalysts
The carbonyl-ene reaction can proceed without any catalyst or promoter, but to overcome the high activation barriers, it needs moderate to high temperatures.109 However, the development of promoters or Lewis acid catalysts for this reaction has been carried out. Comprehensive reviews on the development of catalytic carbonyl-ene reactions were published by Mikami in 1992109 and 2010,110 and Clarke in 2008,107 respectively. The most commonly used promoters were aluminum salts, such as AlMe2Cl, which promotes the reaction at temperatures as low as −78 °C.111 Lewis acid catalysts, such as SnCl4,112 BF3∙Et2O,113 and others,114 were used for accelerating the reaction without being consumed. The more detailed research was conducted on commercially important citronellal cyclisation and glyoxylate-ene reaction. Moreover, the demand for high enantiomerically pure products boosted the development of chiral catalysts. Classical catalysts such as BINOL-Ti,115 and Bis-oxazoline-Cu116 were developed for the glyoxylate-ene reaction in the 1980s as well as the subsequent Schiff base-Cr117. Other metals were also on the list, such as Co, Ni, Sc, Pd, and Pt.118
Scheme 1−29 Bis-oxazoline-Cu catalyzed glyoxylate-ene reaction of an aldehyde and a ketone
The development of catalysts for the glyoxylate-ene reaction was applied to intermolecular-ene reaction of aldehydes, but the process with ketones is more difficult since ketones are less electrophilic than aldehydes. An example is given in Scheme 1−13, in which the bis-oxazoline catalyzed glyoxylate-ene reaction is performed in a much more effortless way than the ene reaction with a ketone.119 If the ketone is activated, for example with using ethyl trifluoropyruvate as the enophile, the catalyzed reaction may occur at a faster rate than with ethyl glyoxylate.
22
1.5.3 Development of catalysts for the ethyl trifluoropyruvate-ene reaction
Ethyl trifluoropyruvates, used as activated ketones, allow the formation of tetra-substituted stereogenic centers containing a trifluoromethyl group, which is present in numerous building blocks
Scheme 1−30 Carbonyl-ene reaction of ethyl trifluoropyruvate with -methylstyrene
Figure 1−4 The development of catalysts for the trifluoropyruvate-ene reaction for pharmaceuticals and agrochemicals.120 Methyl styrene and ethyl trifluoropyruvate were often used in model reactions for catalytic tests. Since 2004 several catalysts were used for the ethyl trifluoropyruvate-ene reaction including organocatalysts, Brø nsted acids, carbocations, and Lewis acid catalysts derived from transition metals, as shown below (Scheme 1−14, Figure 1−4 and Table
23
1−1). In Table 1−1 we can see that every two years on average a new catalyst was developed. Transition metals, such as palladium ruthenium and indium, belong to the list of chiral catalysts.120- 128 However, to the best of our knowledge, there are only two precedents of stoichiometric use of
FeCl3 in intramolecular carbonyl-ene cyclization reactions.129 Consequently, the development of a new iron-catalyzed ene reaction is in high need due to its high potential in synthetic organic chemistry.
Year Catalyst Yield (%) ee (%)
2004 120 BINAP-Pd 99 93
2007 121 thiourea 97 30 Phosphoric- 2008 122 76 96 amide acid 2010 123 di-phosphine-Pd 64 91
2010 124 Py-box-In 99 95
2012 125 Phosphoric-Ca 68 94
2013 126 Ru2+ 89 93
2015 127 carbocation 61 84
Table 1−1 The development of catalysts for the trifluoropyruvate-ene reaction
1.6 Synthetic organic chemistry of caffeine
Caffeine is a central nervous system stimulant and is a member of the methyl-xanthine class.130 Its unique structure makes it ready for the synthesis of N-heterocyclic carbenes (Scheme 1−15). Caffeine, named as 1,3,7-trimethylxanthine, can be alkylated at the N9 position. One of the easiest ways of the alkylation is to install a methyl group at the N9 position to form 1,3,7,9- tetramethylxanthine. Suitable methods using dimethyl sulfate,131a methyl tosylate,131b and methyl iodide132 as alkylating reagents were reported as early as in 1962, but it took 14 years for caffeine derived-NHC precursors to act as an NHC ligand with transition metal.
24
Scheme 1−31 Alkylation of caffeine NHC derived from caffeine was developed earlier than Arduengo’s carbene. Ru2+ and Ru3+ were complexed with protonated imidazolylidene from xanthine in 1975, but the complexes were not synthesized directly from methylated-caffeine derived-NHC.133 Until 1976, using the same procedure (explanation in Figure 1) as Wanzlick,74 Beck synthesized the first bis-(methylated-caffeine derived- NHC)-Hg complex from heating mercury acetate and two equivalents of 1,3,7,9- tetramethylxanthinium in DMSO at 80 °C (Figure 1−5).134
Figure 1−5 The first caffeine derived-NHC-Hg complex More caffeine-derived NHCs came up in the 2000s, as shown in Figure 1−6. Youngs disclosed the synthesis of caffeine-methylated NHC-Rh that derived from the corresponding silver complex by transmetalation in 2003.135 Hermann reported the synthesis of rhodium and iridium 1,3,7,9- tetramethylxanthine-8-ylidine complexes but without further catalytic applications.136 Further research in the same group led to a 1,3,7,9-tetramethylxanthine-8-ylidine silver complex, which exhibited antibiotic properties in medicinal applications.137 Alkylated caffeine-derived imidazolium salts as NHC precursors were developed with Au,138 Ag,139 Pt,140 and Pd141 by various other groups.138-141 Most research was focused on biological properties or medicinal applications. Complexes with Au, Ag and Pt show either anti-cancer, antibiotic features, and with Pt140a it also exhibits photoluminescent properties.138-140 Landaeta studied the coordination chemistry of bidentate benzylated-caffeine NHC-Pd.141
It is worth emphasizing that Luo reported the only case of using caffeine-derived NHC as a ligand in homogeneous catalysis. A caffeine-derived bis-NHC-PdI2 was very efficient in Heck, Suzuki and Sonogashira coupling reactions in aqueous solution (Scheme 1−16).142
25
Figure 1−6 Development of various caffeine derived-NHC-metal complexes
Scheme 1−32 The application of bis-NHC-PdI2 in Suzuki cross-coupling reaction
Although caffeine-derived NHCs have shown great potential as ligands and were used in various applications, it is much less developed in the organic chemistry field. Moreover, caffeine is eco- friendly, with a massive abundance of natural resources and has accordingly low cost.130 Hence, the combination of iron and caffeine would be very auspicious in the field of homogeneous catalysis.
26
1.7 References
[1] (a) Bolm. C. Nat. Chem. 2009, 1, 420. (b) Kolasinski, Kurt W. (2002). "Where are Heterogenous Reactions Important". Surface science: foundations of catalysis and nanoscience. John Wiley and Sons. Wiltshire. pp. 15–16.
[2] Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Chem. Rev. 2004, 104, 6217.
[3] Ingmar, B.; Knö lker, H-J. Chem. Rev. 2015, 115, 3170.
[4] Fü rstner, A. ACS Cent. Sci. 2016, 2, 778.
[5] (a) Diaz, D. D.; Miranda, P. O.; Padron, J. I.; Martín, V. S. Curr. Org. Chem. 2006, 10, 457. (b) Song, Y.; Tang, X.; Hou, X.; Bai, Y. Chin. J. Org. Chem. 2013, 33, 76.
[6] Wang, C.-X.; Wan, B.-S. Chin Sci Bull 2012, 57, 2338.
[7] Czaplik, W. M.; Mayer, M.; Cvengros, J.; Jacobi, A. ChemSusChem 2009, 2, 396.
[8] (a) Gopalaiah, K. Chem. Rev. 2013, 113, 3248. (b) Ollevier, T.; Keipour, H. "Enantioselective Iron Catalysts", in Iron Catalysis II, Bauer, E. Ed., Topics in Organometallic Chemistry, Springer, 2015. (c) Ollevier, T. Catal. Sci. Technol. 2016, 6, 41.
[9] Riener, K.; Haslinger, S.; Raba, A.; Hö gerl, M. P.; Cokoja, M.; Herrmann, W. A.; Kü hn, F. E. Chem. Rev. 2014, 114, 5215.
[10] Castro, L. C. M.; Li, H.; Sortais, J. -B.; Darcel, C. Green Chem. 2015, 17, 2283.
[11] Peters, R.; Fischer, D. F.; Jautze, S. Top. Organomet. Chem. 2011, 33, 139.
[12] Egorova, K. S.; Ananikov, V. P. Angew. Chem., Int. Ed. 2016, 55, 12150.
[13] Pradyot Patnaik. 2003. Handbook of inorganic chemicals. McGraw-Hill. New York. pp. 410-441.
[14] Bertini, I.; Gray, H. B.; Lippard, S. J.; Valentine, J. S., Eds. Bioinorganic Chemistry; University Science Books: Mill Valley, CA,1994.
[15] H. J. H. Fenton. J. Chem. Soc. Trans. 1894, 65, 899.
[16] Wardman, P.; Candeias, P. L. J. Radiat. Res. 1996, 145, 523.
[17] Stadtman, R. E.; Berlett, S. B. J. Bio. Chem. 1991, 266, 17201.
27
[18] Bautista, P.; Mohedano, F. A.; Casas, A. J.; Zazo, A. J.; Rodriguez, J. J. J. Chem. Technol. Biotechnol. 2008, 83, 1323.
[19] Pereira, C. M.; Olliveira, L. C. A.; Murad, E. Clay Miner. 2012, 47, 285.
[20] (a) V. Smil. Nature 1999, 29, 415. (b) Schlö gl, R. Angew. Chem., Int. Ed. 2003, 42, 2004.
[21] Erisman, W. J.; Sutton, A. M.; Galloway, J.; Klimont, Z.; Winiwarter, W. Nat. Geosci. 2008, 1, 636.
[22] (a) Cherkasov, N.; Ibahadon, A. O.; Fitzpatrick, P. Chem. Eng. Process. 2015, 90, 24. (b) Liu, H.-Z. Chin. J. Catal. 2014, 35, 1619. (c) Baranski, A.; Kotarba, A.; Lagan, J. M.; Pattek-Janczyk, A.; Pyrczak, E.; Reizer, A. Appl. Catal. A 1994, 112, 13.
[23] (a) Boudart, M. Catal. Rev. Sci. Eng. 1981, 23, 37. (b) Tayment, T.; Schlö gl, R.; Thomas, J.M.; Ertl, G. Nature 1985, 315, 311. (c) Altenburg, K.; Bosch. H.; Van Ommen, J.; Gellings, P. J. Catal. 1980, 334, 326. (d) Ertl, G.; Huber, M.; Lee, S. B.; Paá l, Z.; Weiss, M. Appl. Surf. Sci. 1981, 8, 373.
[24] Grubel, Katarzyna, Brennessel, W. W.; Mercado, Q. B.; Holland. L. P. J. Am. Chem. Soc. 2014, 136, 16807. (b) McWilliams, F. S.; Rodgers, R. K.; Lukat-Rodgers, G.; Mercado, Q. B.; Grubel, K.; Holland, L. P. Inorg. Chem. 2016, 55, 2960. (c) Kandemir, T.; Schuster, E. M.; Senyshyn, A.; Behrens, M.; Schlö l, R. Angew. Chem., Int. Ed. 2013, 52, 12723.
[25] (a) Sivasankar, C.; Baskaran, S.; Tamizmani, M.; Ramakrishna, K. J. Organomet. Chem. 2014, 752, 44. (b) Smith, E. B. Science 2012, 297, 1654.
[26] (a) Dry, E. M. Catal. Today 2002, 71, 227. (b) Davis, H. B. Catal. Today 2009, 25, 141. (c) Smit, E.; Weckhuysen, M. B. Chem. Soc. Rev. 2008, 37, 2758.
[27] Reppe, W.; Vetter, H. Liebigs Ann. Chem. 1953, 582. 133.
[28] Groves, T. J.; Nemo, E. T.; Myers, S. R. J. Am. Chem. Soc. 1979, 101, 1032.
[29] Nordblom, G. D.; White, R. E.; Coon, M. J. Arch. Biochem. Biophys. 1976, 175, 524.
[30] (a) Michaelson, R. C.; Palermo, R. E.; Sharpless, K. B. J. Am. Chem. Soc.1977, 99, 1990. (b) Yamada, S.; Mashiko, T.; Terashima, S. J. Am. Chem. Soc. 1977, 99, 1988.
[31] Chang, C. K.; Kuo, M.-J. J. Am. Chem. Soc. 1979, 101, 3413.
28
[32] Groves, T. J.; Myers, S. R. J. Am. Chem. Soc. 1983, 105, 5791.
[33] Wahiki, T.; Oriyama, T. Tetrahedron Lett. 2002, 43, 8959.
[34] Christoffers, J. Chem. Commun. 1997, 943. (b) Christoffers, J. J. Chem. Soc., Perkin Trans. 1997, 1, 3141.
[35] Christeoffers, J.; Mann, A. Angew. Chem., Int. Ed. 2000, 39, 2752.
[36] Eaton, B. E.; Rollman, B. J. Am. Chem. Soc. 1992, 114, 6245. (b) Sigman, M. S.; Eaton, B. E.; Rollman, B. J. Am. Chem. Soc. 1996, 118, 11783.
[37] Cherkaoui, H.; Soufiaoui, M.; Gré e, R. Tetrahedron 2001, 57, 2379.
[38] (a) Small, B. L.; Brookhart, M.; Bennett, A. M.-A. J. Am. Chem. Soc. 1998, 120, 4049. (b) Spaleck, W.; Kuber, F.; Winter, A.; Rohmann, J.; Bachmann, B.; Antberg, M. Organometallics 1994, 13, 954.
[39] Cahiez, G.; Gager, O.; Habiak, V. Synthesis 2008, 2636.
[40] (a) Egami, H.; Katsuki, T. J. Am. Chem. Soc. 2009, 131, 6082. (b) Egami, H.; Matsumoto, K.; Oguma, T.; Kunisu, T.; Katsuki, T. J. Am. Chem. Soc. 2012, 48, 5823.
[41] Zhu, S.-F.; Cai, Y.; Mao, H.-X.; Xie, J.-H.; Zhou, Q.-L. Nat. Chem. 2010, 2, 546.
[42] Kove, K. E.; Wolf, C. Chem. Eur. J. 2008, 14, 6302. (b) Qin, C.; Wu, H.; Chem, J.; Liu, M.; Cheng, J.; Su, W.; Ding, J. Org. Lett. 2008, 10, 1537. (c) Yadav, J. S.; Reddy, B. V. S.; Basak, A. K.; Narsaiah, A. V. Chem. Lett. 2004, 33, 248.
[43] Rosa, J. N.; Reddy, R. S.; Candeias, N. R.; Cal, P. M. S. D.; Gois, P. M. P. Org. Lett. 2010, 12, 2686.
[44] (a) Narayanaperumal, S.; Silva, R. C. D.; Monteiro, J. L.; Corrêa, A. G.; Paixāo, M. W. J. Braz. Chem. Soc. 2012, 23, 1982. (b) Allen, C. L.; Lapkin, A. A.; Wilianms, J. M. J. Tetrahedron Lett. 2009, 50, 4262.
[45] (a) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2004, 126, 13794. (b) Bart, S. C.; Hawrelak, E. J.; Lobkovsky, E.; Chirik, P. J. Organometallics 2005, 24, 5518. (c) Russell, S. K.; Darmon, J. M.; Lobkovsky, E.; Chirik, P. J. Inorg. Chem. 2010, 49, 2782.
29
[46] de Vries, J. G.; Elsevier, C. J. C. J. Handbook for homogeneous hydrogenation, Vol. 1, Wiley- VCH, Weinheim, 2007, p. 375.
[47] Srimani, D.; Diskin-Posner, Y.; Ben-David, Y.; Milstein, D. Angew. Chem., Int. Ed. 2013, 52, 14131.
[48] (a) Ollevier, T.; Plancq, B. Chem. Commun. 2012, 48, 2289. (b) Plancq, B.; Ollevier, T. Chem. Commun. 2012, 48, 3806. (c) Lauzon, S.; Keipour, H.; Gandon, V.; Ollevier, T. Org. Lett. 2017, 19, 6324. (d) Keipour, H.; Ollevier, T. Org. Lett. 2017, 19, 5736. (e) Li, M.; Carreras, V.; Jalba, A.; Ollevier, T. Org. Lett. 2018, 20, 995.
[49] Hopkinson, M. N.; Richter, C.; Schedler, M.; Glorius, F. Nature 2014, 510, 485.
[50] Benhamou, L.; Chardon, E.; Lavigne, G.; Bellemin-Laponnaz, S.; Cé sar, V. Chem. Rev. 2011, 111, 2705.
[51] Díez-Gonzá lez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109, 3612.
[52] (a) Scholl, M.; Trnka, T. M.; Morgan, J. P.; Grubbs, R. H. Tetrahedron Lett. 1999, 40, 2247. (b) Huang, J.; Stevens, E. D.; Nolan, S. P.; Petersen, J. L. J. Am. Chem. Soc. 1999, 121, 2674. (c) Achermann, L.; Fü rstner, A.; Weskamp, T.; Kohl, F. J.; Hermann, W. A. Tetrahedron Lett. 1999, 40, 4787.
[53] Arduengo, A. J.; III, Harlow, R. L.; Kline, M. J. Am. Chem. Soc. 1991, 113, 361.
[54] Cavallo, L.; Cazin, C. S. J. N-Heterocyclic Carbenes in Transition Metal Catalysis and Organocatalysis. In Catalysis by Metal Complexes 32; Springer Science+Bussiness MediaB .V. Berlin, Heidelberg. 2011; pg 1-19.
[55] Kraft, A. Bull. Hist. Chem. 2008, 33, 61.
[56] Buser, H. J.; Schwarzenbach, D.; Petter, W.; Ludi, A. Inorg. Chem. 1977, 16, 2704.
[57] Herren, F.; Fischer, P.; Ludi, A.; Halg, W. Inorg. Chem. 1980, 19, 956.
[58] Ting, V. P.; Henry, P. F.; Schmidtmann, M.; Wilson, C. C.; Weller, M. T. Chem. Comm. 2009, 48, 7527.
[59] Jackson, W. G.; Josephine, A. M.; Cortez, S. Inorg. Chem. 2004, 43, 6249.
[60] James, B, K. Acc. Chem. Res. 2005, 38, 671.
30
[61] William, H. B.; Jensen, K. A.; Jø rgensen, C. K.; Kauffman, G. B. Polyhedron 1983, 2, 1.
[62] Irving, H.; Williams, R. J. P. Nature 1948, 162, 4123.
[63] Buchner, E.; Feldmann, L. Ber. Dtsch. Chem. Ges. 1903, 36, 3509.
[64] (a) Tschugajeff, L.; Skanawy-Grigorjewa, M.; Posnjak, A. Z. Anorg. Chem. 1925, 148, 37. (b) Tschugajeff, L.; Skanawy-Grigorjewa, M. J. Russ. Chem. Ges. 1915, 47, 776.
[65] (a) Rouschias, G.; Shaw, B. L. Chem. Commun. 1970, 183. (b) Burke, A.; Balch, A. L.; Enemark, J. H. J. Am. Chem. Soc. 1970, 92, 2555. (c) Stork, J. R.; Olmstead, M. M.; Balch. A. L. Inorg. Chem. 2004, 42, 7508.
[66] Fehlhammer, W. P.; Fritz, M. Chem. Rev. 1993, 93, 1243.
[67] Doering, W. E.; Hoffmann, A. K. J. Am. Chem. Soc. 1954, 76, 6162.
[68] Fischer, E. O.; Maasboel, A. Angew. Chem. 1964, 76, 645.
[69] (a) Frohlich, N.; Frenking, G. Phys, Organomet, Chem. 1999, 2, 173. (b) Jiang, W.; Fuertes, M. J.; Wulff, W. D. Tetrahedron 2000, 56, 2183. (c) Jacobsen, H.; Ziegler, T. Organometallics, 1995, 14, 224. (d) Ehlers, A. W.; Dapprich, S.; Vyboishchikov, S. F.; Frenking, G. Organometallics 1996, 15, 105.
[70] Schrock, R. R. J. Am. Chem. Soc. 1974, 96, 6796.
[71] (a) Cundari, T. R.; Gordon, M. S. Organometallics 1992, 11, 55. (b) Taylor, T. E.; Hall, M. B. J. Am. Chem. Soc. 1984, 106, 1576. (c) Fox, H. H.; Schofield, M. H.; Schrock, R. R. Organometallics 1994, 13, 2804. (d) Schoeller, W. W.; Rozhenko, A. J. B.; Alijah, A. J. Organomet. Chem. 2001, 617, 435.
[72] Wanzlick, H. W. Angew. Chem., Int. Ed. 1962, 1, 75.
[73] Ö fele, K. J. Organomet. Chem. 1968, 12, 42.
[74] Wanzlick, H. W.; Schö nherr, H. J. Angew. Chem., Int. Ed. 1968, 7, 141.
[75] (a) Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290. (b) Herrmann, W. A.; Kö cher, C. Angew. Chem., Int. Ed. Engl. 1997, 36, 2162.
[76] Herrmann, W. A.; Elison, M.; Fischer, J.; Kö cher, C.; Artus, G. R. J. Angew. Chem., Int. Ed. Engl. 1995, 34, 2371.
31
[77] (a) Enders, D.; Gielen, H.; Raabe, G.; Runsink, J.; Teles, J. H. Chem. Ber. 1996, 129, 1483. (b) Teles, J. H.; Melder, J.-P.; Ebel, K.; Schneider, R.; Gehrer, E.; Harder, W.; Brode, S.; Enders, D.; Breuer, K.; Raabe, G. Helv. Chim. Acta. 1996, 79, 61.
[78] Kü cü kbay, H.; Cetinkaya, B.; Guesmi, S.; Dixneuf, P. H. Organometallics 1996, 15, 2434.
[79] (a) Díez-Gonzá lez, S.; Nolan, S. P. Coord. Chem. Rev. 2007, 251, 874. (b) Cavallo, L.; Correa, A.; Costabile, C.; Jacobsen, H. J. Organomet. Chem. 2005, 690, 5407. (c) Strassner, T. Top. Organomet. Chem. 2004, 13, 1.
[80] (a) Jacobsen, H.; Correa, A.; Poater, A.; Costabile, C.; Cavallo, L. Coord. Chem. Rev. 2009, 253, 687. (b) Radius, U.; Bickelhaupt, F. M. Coord. Chem. Rev. 2009, 253, 678. (c) Frenking, G.; Solà , M.; Vyboishchikov, S. F. J. Organomet. Chem. 2005, 690, 6178.
[81] Nelson, D. J. Nolan, S. P. Chem. Soc. Rev. 2013, 42, 6723.
[82] Hahn, F. E.; Jahnke, M. C. Angew. Chem., Int. Ed. 2008, 47, 3122.
[83] Herrmann, W. A.; Elison, M.; Fischer, J.; Kö cher, C.; Artus, G. R. J. Angew. Chem., Int. Ed. Engl. 1995, 34, 2371.
[84] Díez-Gonzá lez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109, 3612.
[85] Ö fele, K. Angew. Chem., Int. Ed. 1969, 8, 916.
[86] Tard, C. D.; Pickett, C. J. Chem. Rev. 2009, 109, 2245. (b) Lavallo, V.; Grubbs, R. H. Science 2009, 326, 559.
[87] Plietker, B.; Dieskau. A. Eur. J. Org. Chem. 2009, 775.
[88] Bauer, E. B. Curr. Org. Chem. 2008, 12, 1341.
[89] Liu, L. X. Curr. Org. Chem. 2010, 14, 1099.
[90] Bé zier, D.; Sortais, J.-B.; Darcel, C. Adv. Synth. Catal. 2013, 355, 19.
[91] Louie, J.; Grubbs, R. H. Chem. Commun. 2000, 1479.
[92] Bedford, R. B.; Betham, M.; Bruce, D. W.; Danopoulos, A. A.; Frost, R. M.; Hird, M. J. Org. Chem. 2006, 71, 1104.
[93] (a) Mo, A.; Zhang, Q.; Deng, L. Organometallics 2012, 31, 6518. (b) Meyer, S.; Orben, C. M.; Demeshko, S.; Dechert, S.; Meyer, F. Organometallics 2011, 30, 6692. (c) Wu, J.; Dai, W.;
32
Farnaby, J. H.; Hazari, N.; Le Roy, J. J.; Mereacre, V.; Murugesu, M.; Powell, A. K.; Takase, M. K. Dalton Trans. 2013, 42, 7404.
[94] (a) Ghorai, S. K.; Jin, M.; Hatakeyama, T.; Nakamura, M. Org. Lett. 2012, 14, 1066. (b) Silberstein, A. L.; Ramgren, S. D.; Garg, N. K. Org. Lett. 2012, 14, 3796. (c) Agrawal, T.; Cook, S. P. Org. Lett. 2013, 15, 96.
[95] Guisá n-Ceinos, M.; Tato, F.; Bunuel, E.; Calle, P.; Cá rdenas, D. J. Chem. Sci. 2013, 4, 1098.
[96] Trost, B. M.; Crawley, M. L. Chem. Rev. 2003, 103, 2921.
[97] (a) Plietker, B. Angew. Chem., Int. Ed. 2006, 45, 1469. (b) Plietker, B.; Dieskau, A.; Mö ws, K.; Jatsch, A. Angew. Chem., Int. Ed. 2008, 47, 198. (c) Holzwarth, M.; Dieskau, A.; Tabassam, M.; Plietker, B. Angew. Chem., Int. Ed. 2009, 48, 7251. (d) Dieskau, A. P.; Holzwarth, M. S.; Plietker, B. Chem. Eur. J. 2012, 18, 2423.
[98] Jegelka, M.; Plietker, B. ChemCatChem 2012, 4, 329.
[99] Hatanaka, T.; Ohki, Y.; Tatsumi, K. J. Asian Chem. 2010, 5, 1657.
[100] Obligacion, J. V.; Chirik, P. J. Org. Lett. 2013, 15, 2680.
[101] Yu, R. P.; Darmon, J. M.; Hoyt, J. M.; Margulieux, G. W.; Turner, Z. R.; Chirik, P. J. ACS. Catal. 2012, 2, 1760.
[102] (a) Coessens, V.; Pintauer, T.; Matyjaszewski, K. Prog. Polym. Sci. 2001, 26, 337. (b) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921.
[103] Lindhorst, A. C.; Netscher, T.; Bonrath, W.; Kü nh, F. E. Catal. Sci. Technol. 2017, 7, 1902.
[104] Saino, N.; Kogure, D.; Okamoto, S. Org. Lett. 2005, 7, 3065.
[105] Hilt, G.; Bolze, P.; Kieltsch, I. Chem. Commun. 2005, 1996.
[106] Liu, Y-S.; Wang, L-J.; Deng, L. J. Am. Chem. Soc. 2016, 138, 112.
[107] (a) Jiang, F.; Bé zier, D.; Sortais, J. B.; Darcel, C. Adv. Synth. Catal., 2011, 353, 239. (b) Bé zier, D.; Jiang, F.; Roisnel, T.; Sortais, J. B.; Darcel, C. Eur. J. Inorg. Chem. 2012, 1333. (c) Lopes, R.; Cardoso, J. M. S.; Postigo, L. Royo, B. Catal. Lett. 2013, 143, 1061. (d) Warratz, S.; Postigo, L.; Royo, B. Organometallics 2013, 32, 893. (e) Li, H.; Misal Castro, L. C.; Zheng, J.; Roisnel, T.; Dorcet, V.; Sortais, J. B.; Darcel, C. Angew. Chem., Int. Ed. 2013, 52, 8045. (f) Bé zier, D.;
33
Venkanna, G. T.; Sortais, J. B.; Darcel, C. ChemCatChem. 2011, 3, 1747. (g) Misal Castro, L. C.; Sortais, J. B.; Darcel, C. Chem. Commun. 2012, 48, 151. (h) Cardoso, J. M. S.; Royo, B. Chem. Commun. 2012, 48, 4994.
[108] Mikami, K.; Shimizu, M. Chem. Rev. 1992, 92, 1021.
[109] Mikami, K.; Aikawa, K. In Catalytic Asymmetric Synthesis, Ojima. I., Ed.; Thieme: Hoboken, 2010, 683.
[110] Clark, M. L.; France, M. B. Tetrahedron 2008, 64, 9003.
[111] (a) Snider, B. B.; Rodini, D. J. Tetrahedron Lett. 1980, 21, 1815. (b) Snidr, B. B.; Rodini, D. J.; Kirk, T. C.; Cordova, R. J. Am. Chem. Soc. 1982, 104, 555.
[112] Cook, R. C.; Mirza, N. A. Synth. Commun. 1981, 11, 229.
[113] (a) Blomquist, A. T.; Himics, R. J. J. Org. Chem. 1968, 33, 1156. (b) Blomquist, A. T.; Himics, R. J. Tetrahedron Lett. 1967, 3947.
[114] (a) Qian, C.; Huang, T. Tetrahedron Lett. 1997, 38, 6721. (b) Qian, C.; Huang, T.; Tetrahedron: Asymmetry 2000, 11, 2347.
[115] (a) Mikami, K.; Matsukawa, S. J. Am. Chem. Soc. 1993, 115, 7039. (b) Mikami, K.; Matsukawa, S. Tetrahedron Lett. 1994, 35, 3133. (c) Terada, M.; Motoyama, Y.; Mikami, K. Tetrahedron Lett. 1994, 35, 6693. (d) Mikami, K.; Motoyama, Y.; Terada, M. Inorg. Chim. Acta. 1994, 222, 71. (e) Matsukawa, S.; Mikami, K. J. Chem. Soc., Chem. Commun. 1995, 2391.
[116] Evans, D. A.; Burgey, C. S.; Paras, N. A.; Vojkovsky, T.; Tregay, S. W. J. Am. Chem. Soc. 1998, 120, 5821.
[117] (a) Ruck, R. T.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 2882. (b) Ruck, R. T.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2003, 42, 47.
[118] (a) Tudor, M. D.; Becker, J. J.; White, P. S.; Gagné , M. R. Organometallics 2000, 19, 4367. (b) Becker, J. J.; White, P. S.; Gagné , M. R. J. Am. Chem. Soc. 2001, 123, 9478. (c) Evans, D. A.; Wu, J. J. Am. Chem. Soc. 2005, 127, 8006. (d) Kezuka, S.; Kogami, Y.; Ikeno, T.; Yamada, T. Bull. Chem. Soc. Jpn. 2003, 76, 49. (e) Hutson, G. E.; Dave, A. H.; Rawal, V. H. Org. Lett. 2007, 9, 3869.
34
[119] Evans, D. A.; Tregay, S. W.; Burgey, C. S.; Paras, N. A.; Vojkovsky, T. J. Am. Chem. Soc. 2000, 122, 7936.
[120] (a) Mikami, K.; Terada, M. In Comprehensive Asymmetric Catalysis; Jacobsen, E. N.; Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin, 1999; Vol. III, Chapter 32. (b) Snider, B. B. InComprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, U.K., 1991.
[121] Mikami, K; Aikawa, K.; Kainuma, S.; Kawakami, Y.; Saito, T.; Sayo, N.; Kumobayashi, H. Tetrahedron: Asymmetry 2004, 15, 3885.
[122] Clarke, M.; Jones, C. ES.; France, M. B. Beilstein J. Org. Chem. 2007, 3, 24.
[123] Rueping, M.; Theissmann, T.; Kuenkel, A.; Koenigs, R. M. Angew. Chem., Int. Ed. 2008, 47, 6798.
[124] Luo, H.K.; Woo, Y-L,; Schumann, H.; Jacob, C.; Yang, H-Y.; Tan, Y-T. Adv. Synth. Catal. 2010, 352, 1356.
[125] Zhao, J. F.; Tan, B. H.; Zhu, M. K.; Tjan, T. B. W.; Tec Peng, L. Adv. Synth. Catal. 2010, 352, 2085.
[126] Rueping, M.; Bootwicha, T.; Kambutong, S.; Sugiono, E. Chem. Asian. J. 2012, 7, 1195.
[127] Wang, T.; Hao, X-Q.; Huang, J. J.; Niu, J-J.; Gong, J-F.; Song, MP. J. Org. Chem. 2013, 78, 8712.
[128] Lv, J.; Zhang, Q.; Zhong, S.; Luo, S.Z. J. Am. Chem. Soc. 2015, 137, 15576.
[129] (a) Laschat, S.; Grehl, M. Chem. Ber. 1994, 127, 2023. (b) Laschat, S.; Fox, T. Synthesis 1997, 45.
[130] (a) Terada, H.; Sakabe, Y. J. Chromatogr. 1984, 291, 453. (b) Kumazawaa, T.; Senob, H.; Leea, X.P.; Ishiib, A.; Suzukib, K.W.; Satoa, K.; Suzukib, O. Anal. Chim. Acta 1999, 387, 53. (c) Khursheed, T.; Ansari, M. Y. K; D. Shahab. Biol. Med. 2009, 1, 56. (d) Waldvogel, S.R. Angew. Chem., Int. Ed. 2003, 42, 604.
[131] (a) Hori, M.; Kataoka, T.; Shimizu, H.; Imai, E.; Matsumoto, Y. Chem. Pharm. Bull. 1985, 33, 3681. (b) Bredereck, H.; Christmann, O.; Koser, W.; Schellenberg, P.; Nast, R. Chem. Ber. 1962, 95, 1812.
35
[132] Ivanov, E. I.; Kalayanov, G. D.; Yaroshchenko, I. M.; Stepanov, D. E. Khim. Geterotsikl. Soedin. 1989, 11, 1570.
[133] (a) Clarke, M. J.; Taube, H. J. Am. Chem. Soc. 1975, 97, 1397. (b) Ckrentzien, H. J.; Clarke, M. J.; Taube, H. Bioinorg. Chem. 1975, 4, 143.
[134] Beck, W.; Kottmair, N.; Chem. Ber. 1976, 109, 970.
[135] Kascatan-Nebioglu, A.; Panzner, J. M.; Carrison, C. Jered.; Tessier, A. C.; Youngs, J. W.; Organometallics 2004, 23, 1982.
[136] Schü tz, J.; Herrmann, A. W. J. Organomet. Chem. 2004, 689, 2995.
[137] (a) Kascatan-Nebioglu, A.; Melaiye, A.; Hindi, K.; Durmus, S.; Panzner, J. M.; Hogue, A. L.; Mallett, J. R.; Hovis, E. C.; Coughenour, M.; Crosby, D. S.; Milsted, A.; Ely, L. D.; Tessier, A. C.; Cannon, L. C.; Youngs, W. J. J. Med. Chem. 2006, 49, 6811. (b) Kascatan-Nebioglu, A.; Panzner, M. J.; Tessier, C. A.; Cannon, C. L.; Yongs, W. J.; Coord. Chem. Rev. 2007, 251, 884.
[138] Bertrand, B.; Stefan, L.; Pirrotta, M.; Monchaud, D.; Bodio, E.; Richard, P.; Le Gendre, P.; Warmerdam, E.; H. De Jager, M.; Groothuis, G. M. M.; Picquet, M.; Casini, A. Inorg. Chem. 2014, 53, 2296.
[139] Mohanmed, H. A.; Lake, B. R. M.; Laing, T.; Phillips, R. M.; Willans, C. E. Dalton Trans. 2015, 44, 7563.
[140] (a) Hu, J. J.; Bai, S. Q.; Yeh, H. H. Young, D. J.; Chi, Y.; Hor, T. S. Dalton Trans. 2011, 40, 4402. (b) Zhang, J. J.; Che, C-M.; Ott, I. J. Org. Chem. 2015, 782, 37.
[141] Landaeta, V. R.; Rodríguez-Lugo, R. E.; Rodríguez-Lugo, E. N.; Coll-Gó mez, D. S.; Gonzá lez, T. Transition Met. Chem. 2010, 35, 165.
[142] Luo, F.-T.; Lo, H.-K. J. Organomet. Chem. 2011, 696, 1262.
36
Chapter two
Development of iron catalysis in carbonyl-ene reaction
2.1 Iron salts as Lewis acid for carbonyl-ene reaction
For the past decade, research focused on using trifluoropyruvates as activated ketones, allowing the formation of compounds possessing tetra-substituted stereogenic centers containing a trifluoromethyl group, which are important building blocks for pharmaceuticals and agrochemicals. Transition metal Lewis acids, such as Ca Pd, Pt, In, Ru complexes, were developed as efficient catalysts in trifluoropyruvates-ene reactions as shown in Table 1−1. Iron was much less explored compared to other transition metals used in trifluoropyruvates-ene reactions. There is still room for developing catalytic iron-mediated processes for the carbonyl-ene reaction. Consequently, the development of new iron-catalyzed carbonyl-ene reactions is in high need due to its high potential in synthetic organic chemistry.
Metal triflate salts such as iron triflate Fe(OTf)2,1 scandium triflate Sc(OTf)3,2 gallium triflate Ga(OTf)33 and bismuth triflate [Bi(OTf)3·4H2O]4 were developed as Lewis acid in various organic transformations. These metal salts exhibited good catalytic efficiency toward carbonyl groups.
Sc(OTf)3 is resistant to water and hence can work in an aqueous phase. It was initially developed to promote the Mukaiyama-aldol reaction of benzaldehyde with silyl enol ethers. After the reaction is complete, it can be recycled. Methods using gallium salts were developed for reactions involving carbonyl groups such as the direct hydroamination of aldehydes using anilines and the reduction of ketones to methylene using organosilane. Bismuth was categorized as a green catalyst and BiX3 salts were used to various reactions.5a Bi(OTf)3·4H2O was a more efficient catalyst than BiCl3 for the intramolecular carbonyl-ene reaction with citronellal.5b Hence, it will be valuable to test Bi(OTf)3·4H2O in the intermolecular carbonyl-ene reaction of ethyl trifluoropyruvate.
Our studies were initiated on the selected model reaction of -methyl styrene and trifluoropyruvate using a few selected metal triflate salts as Lewis acid (Scheme 2−1, Table 2−1). Fe(OTf)2 led to a low yield (22%), and a higher yield (33%) was achieved using more catalyst loading (Table 2−1,
37
entries 1 and 2). Sc(OTf)3 and Bi(OTf)3·4H2O failed to promote the reaction, and no conversion was observed (Table 2−1, entries 3 and 5). A low yield of 15% was obtained using Ga(OTf)3 (Table 2−1, entry 4). Higher yields were achieved using Fe(OTf)2. Consequently, Fe(OTf)2 was used for further optimization with other iron salts.
Scheme 2−1 Metal triflate catalyzed carbonyl-ene reaction of ethyl trifluoropyruvate
Entry a Metal salts Yield (%)b
1 Fe(OTf)2 22
2 Fe(OTf)2 33 c
3 Sc(OTf)3 −
4 Ga(OTf)3 15
5 Bi(OTf)3·4H2O −
a 1/2 = 1:1, 0.25 mmol methylstyrene and 0.25 mmol ethyl trifluoropyruvate were used in 0.5 mL CH2Cl2; b isolated yields; c 10 mmol% Fe(OTf)2 was used.
Table 2−1 Screening of a few metal triflate Lewis acids
After a comparison of a few selected metal triflate salts, iron triflate (Fe(OTf)2) appeared to be more promising. The studies were continued by testing the model reaction of α-methyl styrene and ethyl trifluoropyruvate with various iron salts (Scheme 2−2, Table 2−2). FeIII catalysts only led to side products of -methyl styrene (Table 2−2, entries 1 and 2). Using FeII salts, the obtained products were a mixture of the ene-product and side products of α-methyl styrene (Table 2−2, entries 3−7).
Scheme 2−2 Iron salts catalyzed carbonyl-ene reaction
38
Entry 1/2 FeXn Yield (%)a
1 2:1 FeCl3 –
2 2:1 Fe(OTf)3 –
3 2:1 FeCl2 58
4 2:1 Fe(OAc)2 44
5 2:1 Fe(NTf)2 63
6 2:1 Fe(ClO4)2·6H2O 53 b
7 2:1 Fe(BF4)2·6H2O 75 b
8 2:1 Fe(OTf)2 62 (66)c
9 4:1 Fe(OTf)2 68
10 1:1.5 Fe(OTf)2 72
11 1:2 Fe(OTf)2 73
12 1:1.2 Fe(OTf)2 54
13 1:1.5 Fe(BF4)2·6H2O 75 b
14 1 : 1.5 FeCl2 + 2AgBF4 82 b
15 1 : 1.5 FeCl2 + 2AgBF4 87 b, d a Isolated yields; b With 4Å molecular sieve (for entries 7 and 13, the yields were the same without 4Å MS); c 10 mol% iron salt, 60 h; d Using filtration on cotton and celite.
Table 2−2 Screening of iron salts in carbonyl-ene reaction
Fe(BF4)2·6H2O afforded homoallylic alcohol 3a in a 75% yield, which encouraged us to optimize the conditions with this specific catalyst (Table 2−2, entry 7). The yield decreased to 62% using Fe(OTf)2 and a more extended reaction time or an increased catalyst loading (10 mol%) had little effect on the reaction efficiency (Table 2−2, entry 8). Optimization of the ratio of -methyl styrene to ethyl trifluoropyruvate was performed in both ways (Table 2−2, entries 9─12). Using an excess of styrene
39
(4 equivalents) did not lead to much increase of yield (Table 2−2, entry 9); but an excess of ethyl trifluoropyruvate afforded a better yield (Table 2−2, entries 10−11). However, using Fe(BF4)2·6H2O, and changing the ratio of 1/2 from 2:1 to 1:1.5 had no impact on the yield (Table 2−2, entries 7 and
13). Fe(BF4)2·6H2O led to 75% which is 3% higher than Fe(OTf)2, but this is not sufficiently enough to state Fe(BF4)2·6H2O is better than Fe(OTf)2 (Table 2−2, entries 10 and 13). Anhydrous Fe(BF4)2, generated from FeCl2 and AgBF4, was also tested and showed better catalytic efficiency than
Fe(BF4)2·6H2O (Table 2−2, entry 14). To suppress any competing catalytic effect resulting from AgCl, a control experiment, via the filtration of AgCl, demonstrated that anhydrous Fe(BF4)2 was a more efficient catalyst, affording 3a 87% yield of 3a (Table 2−2, entry 15). However, Fe(BF4)2·6H2O being a commercial product, it was used for the screening of solvents (Scheme 2−3, Table 2−3).
In MeCN, the carbonyl-ene reaction did not proceed, and the starting materials were recovered
(Table 2−3, entry 1). Et2O and DCE gave moderate yields of 3a (Table 2−3, entries 2 and 3). Toluene led to a higher yield of 3a (Table 2−3, entry 4). Both THF and Me-THF afforded only traces of the product, and ethyl trifluoropyruvate was recovered (Table 2−3, entries 6–7). A few other green solvents were considered (Table 2−3, entries 8–10). Cyclopentyl methyl ether (CPME) gave a promising result (Table 2−3, entry 8). Tert-butyl methyl ether (TBME) and dimethyl carbonate (DMC) led to moderate yields (Table 2−3, entries 9 and 10). Among the list of selected solvents, CH2Cl2 provided the highest yield (Table 2−3, entry 5). Thus, CH2Cl2 was chosen for studying the scope of the reaction using various alkene substrates.
Scheme 2−3 Optimization of solvents
40
Entry Solvent Yield (%)a
1 MeCN –
2 Et2O 52
3 DCE 46
4 Toluene 71
5 CH2Cl2 75
6 THF –
7 Me-THF < 5 b
8 CPME 72
9 TBME 42
10 DMC 59 a Isolated yields; b Estimated by 1H NMR.
Table 2−3 Screening of solvents in carbonyl-ene reaction
To demonstrate the reaction scope, various substituted -methyl styrenes were selected (Scheme
2−4, Figure 2−1), using the best conditions selected in Table 2−2 (5 mol% FeCl2, 10 mol% AgBF4, entry 15).
Scheme 2−4 Carbonyl-ene reaction of various alkenes and ethyl trifluoromethyl pyruvate
41
*isolated yields
Figure 2−1 Reaction scope of different of alkenes
By using substrates possessing an electron-donating group, such as a methyl group in the para position, the yield was good (3b). With a more electron-donating group such as p-MeO, the reaction of side products of alkene (dimerization or polymerization) occurred as a competing reaction and the yield of the carbonyl decreased to 60% (Figure 2−1, 3c). Steric hindrance was another factor influencing the result. Moreover, electron-withdrawing groups (Br, F) on the aryl ring of the styrene derivatives led to moderate yields (3d–e). Methoxy group in ortho position led to further decrease of the yield (3f). More nucleophilic 1,1-disubstituted methylenecyclopentane and methylenecyclohexane were isolated in moderate to good yields (3g–h). By using less nucleophilic 2-isopropenyl naphthalene, the yield dropped to a lower stage (3i). A modest yield was obtained by using 2-(prop-1-en-2-yl)-thiophene as a nucleophile (3j). Only traces of product were obtained by using 2-(1-methylethenyl)-pyridine as a heteroatom aromatic alkene, probably because the pyridine deactivated the catalyst. Less nucleophilic non-cyclic aliphatic alkenes, such as a mono-substituted alkene, i.e., 1-hexene, led only to traces of the expected product. Indeed, 1-hexene has often been used in the literature with a large excess of the enophile.6 Overall, the developed catalytic system was efficient toward 1,1-disubstituted aromatic and aliphatic alkenes. Mono-substituted alkenes such as 1-hexene were not reactive in the reaction conditions.
42
Scheme 2−5 Intramolecular carbonyl-ene reaction of (S)-(+)-citronellal
The Intramolecular carbonyl-ene reaction is a useful method for making C─C bond in organic synthesis.7 The cyclization of (R)-(+)-citronellal affording (−)-isopulegol 5 is a known reaction whichleads to a precursor to the widely employed chemical (−)-menthol.8 A few catalysts have been used for this process, i. e. Sc(OTf)3,9 SnCl4,10 BiCl3,11 and Bi(OTf)3·xH2O.12 FeCl3 (10 mol%) was used in the literature but a low yield (20%) with a 76:24 ratio of isopulegol 5 to other diastereomers was achieved.13 Hence, it was still very valuable to test Fe(BF4)2 in the cyclization of citronellal, since iron salts were less explored compared to other metal salts. In this catalytic system of Fe(BF4)2, by using (S)-(−)-citronellal 4, a total 70% yield was obtained with a 70:30 trans/cis ratio of (+)-isopulegol to (−)-neo-isopulegol in the crude product while the isolated yields were: (+)-isopulegol 5, 45%, (−)- neo-isopulegol 6, 25%, and no (−)-iso-isopulegol or (−)-neo-iso-isopulegol were identified (Scheme 2−5).14
Figure 2−2 Postulated mechanism
Through the evaluation of a variety of iron salts, we have found that the FeII salts were more appropriate to catalyze the studied reaction. Indeed, FeIII salts, such as FeCl3 and Fe(OTf)3, failed to
43
catalyze the carbonyl-ene reaction of -methyl styrene and trifluoropyruvate (Table 2−2, entries 1 and 2). Secondly, it was reported that a marked effect of the counterion affected the conversion of Lewis acid-catalyzed reaction between methylenecyclohexane and ethyl trifluoropyruvate.15
Anhydrous Fe(BF4)2, generated from FeCl2 and AgBF4, led to higher yields than Fe(BF4)2·6H2O. Furthermore, a bidentate coordination can be postulated such that proposed in the CuII catalyzed carbonyl-ene reaction of ethyl glyoxylate disclosed by Evans.16a FeII coordinated with two carbonyl oxygens in one molecule of ethyl trifluoropyruvate was drawn in a bidentate manner in the proposed mechanism (Figure 2−2). The Lewis acid-promoted ene reaction is usually discussed as a continuum from a concerted to a cationic mechanism.16b However, a stepwise radical pathway could also be envisioned.16c, 16d A control experiment using TEMPO as a radical scavenger (methylstyrene 1, 1.5 equiv. trifluoromethyl pyruvate 2, 5 mol% FeCl2, 10 mol% AgBF4, 5 mol% TEMPO) led to a major decrease of the yield of 3a (43% instead of 83%). The observation that the reaction still occurs in the presence of TEMPO does not indicate unambiguously that a radical process is not involved in the process, at least to some extent. To sum up, this was the first iron catalytic system for both inter- and intramolecular carbonyl-ene reactions. Using low catalytic loading of Fe(BF4)2, the reaction proceeds smoothly to afford ene products in low to good yields. The protocol was applied to 1,1- disubstituted alkenes with ethyl trifluoropyruvate and cyclization of citronellal. The carbon-carbon bond forming process was affected by the electronic and steric properties of the substituents on the arene ring in the intermolecular carbonyl-ene reaction. Good yield and selectivity of isopulegol were obtained in the intramolecular carbonyl-ene reaction.17 Further development of iron catalysis in the carbonyl-ene reaction was performed by modifying the Lewis acidity of iron salts with ligands.
2.2 Development of organo-iron catalyst for carbonyl-ene reactions
Caffeine was used as a reactant in a NHC-Cu catalyzed cross-coupling reaction, and the sp2 long pair of nitrogen (N9) indicates that caffeine can act as a base or as a ligand in iron salts.18a Ong reported the C−H activation of caffeine at the C8 position in an NHC-Cu catalyzed coupling reaction (Scheme 2−6).18b The hydrogen was activated by adding two equivalents of alkali t-butoxide as a base (Li, Na, K). Various functionalized aryl iodides were successfully applied in the process. Lamaty reported an efficient catalytic system composed of FeCl2∙4H2O and caffeine as a stabilizing ligand for the transformation of benzamides from alcohols and amine hydrochloride salts in microwave promoted conditions (Scheme 2−6).19 The yield of the reaction was increased from 55% to 71% by
44
lowering the pKa of the base (actually the pKa of the conjugate acid, from pyridine pKa = 5.14 to either
3-chloropyridine pKa = 2.84 or caffeine pKa = 0.81).20, 21 It afforded the same yield (71%) by using 3- chloropyridine or caffeine, but caffeine was preferred due to its lower toxicity and low cost. Using bases of lower pKa was proved to be inefficient in improving the yield. Decreasing the caffeine loading from 1 equivalent to 10 mol% improved the yield from 71% to 81%. The necessity of the complete bicyclic structure of caffeine was confirmed by comparing the result of the structurally simplified analog N-methylimidazole, the latter resulting in a decreased yield of 45%. Herein, caffeine as a weak base was used as a ligand for iron in catalyzing the carbonyl-ene reaction of ethyl trifluoropyruvate (Scheme 2−7 and Table 2−4).
Scheme 2−6 Caffeine as reactant and a stabilizing ligand
Because Fe(OTf)2 exhibited a good reactivity in Table 1−2 and Table 1−3, it was used as a catalyst in this reaction. Without the assistance of caffeine, Fe(OTf)2 alone afforded 33% of the desired product (Table 2−4, entry 2, Table 1−5, entry 1). By adding 10 mol% caffeine (the same ratio of caffeine to iron salt as in Lamaty’s conditions for the formation of benzamides), the same yield was achieved (Table 2−4, entry 2). Changing the ratio of caffeine/iron salt to 4:1 led to a decrease in the yield (Table 2−4, entry 3) but which was still higher than using 5 mol% Fe(OTf)2 alone (Table 2−4, entry 1). KOtBu as a base was used in Ong’s work to activate the C−H bond in the NHC-Cu catalyzed coupling process. Hence, it was used in an attempt to activate caffeine, with the objective of improving the catalytic activity of iron in the carbonyl-ene reaction, through activation of caffeine
(Table 2−4, entry 4). However, the yield was the same as with using 5 mol% Fe(OTf)2 alone (Table
2−4, entry 1). Low yields were obtained with Fe(ClO4)2·6H2O, Fe(OAc)2 and FeCl2 (Table 2−4, entries
45
5−7). A same iron and caffeine loading as in Lamaty’s conditions was used, and the yield was increased to 65% (Table 2−4, entry 8). But a control experiment without using caffeine in CH2Cl2 was performed, and a surprisingly higher yield of 81% was obtained (Table 2−4, entry 9). Conclusively, the expected positive assisting or stabilizing effect as a ligand that caffeine could play was proved to be inefficient or even detrimental to the reaction. Conversely, from the way of considering the pKa of the base, caffeine as a low pKa (pKa = 0.81) base could not help to improve the yield. Then, a base of higher pKa value was expected to be a coordinating ligand in iron-catalyzed carbonyl-ene reaction.
Scheme 2−7 Caffeine act as a ligand to iron salts in carbonyl-ene reaction
FeX AgBF Caffeine KOtBu Entry a n 4 Yield (%)b (mol %) (mol %) (mol %) (mol %)
1 Fe(OTf)2 10% − − − 33
2 Fe(OTf)2 10 % − 10 − 32
3 Fe(OTf)2 5% − 20 − 26
4 Fe(OTf)2 5% − 15 16 21
5 Fe(ClO4)2·6H2O 5% − 15 16 27
6 Fe(OAc)2 5% − 15 16 -
7 FeCl2 5% − 15 16 27
8 c FeCl2 5% 10 10 − 65
9 d FeCl2 5% 10 − − 81 a 1/2 = 1:1, 0.25 mmol methylstyrene and 0.25 mmol ethyl trifluoropyruvate were used in the reaction; b isolated yields; c in CH3CN solution; d in CH2Cl2.
Table 2−4 Caffeine as a ligand in iron-catalyzed carbonyl-ene reaction
46
Pyridine is a base of higher pKa value (pKa = 5.14) compared with caffeine.22 2,2’-bipyridine (Bipy, pKa ≈ 5.1) has the same pKa as pyridine and is a bidentate coordinating ligand that could form a metal complex with many transition metals.23 Fe2+ salts form stable complexes with Bipy24 such as
[Fe(Bipy)2]2+Br2 complex by using ferrous dibromide, which exhibited a low-spin state and was used in a stereoselective photocatalytic reaction in an organocatalyzed alkylation of aldehydes.25 Fe2+ also worked with Bipy-derived ligands such as Bolm’s ligand. Fe(ClO4)2·6H2O was used with Bolm’s ligand to catalyze the Mukaiyama-aldol reaction, for which high yields and enantioselectivities were obtained (Scheme 2−8).26
Scheme 2−8 Iron-Bolm’s ligand catalyzed Mukaiyama Aldol reaction
Scheme 2−9 Bipyridine as a ligand to ironII triflate in catalyzing carbonyl-ene reaction
Based on the pioneered work presented above, the idea was to test Fe(OTf)2 with Bipy as a ligand in the carbonyl-ene reaction of ethyl trifluoropyruvate (Scheme 2−9, Table 2−5). Although Fe2+ forms tri-Bipy coordinating complexes, different ratio of Bipy to Fe(OTf)2 were examined. At a 2:1 Bipy to
Fe2+ ratio, in CH2Cl2 the yield reached 42% (Table 2−5, entry 2) which was higher than Fe(OTf)2 alone (Table 2−5, entry 1). Further increasing the ratio of Bipy to Fe2+ at 4:1, resulted in a decreased yield of 26% (Table 2−5, entry 3). Probably more Bipy ligand deactivated the catalytic iron catalyst. With 2:1 ratio of Bipy to Fe2+ in solvent DCE, the yield was slightly lower at 39% compared to 42%, showing the iron-Bipy complex was less efficient in catalyzing the reaction, even for prolonged reaction time that in DCE (Table 2−5, entry 4).
47
Entry a X (mol%) Y (mol%) Time ( h) Solvent Yield (%)b
1 5 − 24 CH2Cl2 22
2 5 10 24 CH2Cl2 42
3 5 20 24 CH2Cl2 26
4 5 10 48 DCE 39
5 5 6 48 DCE 35
6 5 10 24 Me-THF 39
7 5 10 65 H2O No a 1/2 = 1:1, 0.25 mmol methylstyrene and 0.25 mmol ethyl trifluoropyruvate were used in the reaction; b isolated yields.
Table 2−5 Optimization of Bipyridine-Fe(OTf)2 catalyzed carbonyl-ene reaction
Changing the ratio from 4:1 to 1:1, a lower yield was afforded (Table 2−5, entry 5). Lastly, a 2:1 ratio of Bipy to Fe2+ in Me-THF or H2O gave low yields without further optimization (Table 2−5, entries 6 and 7). With using Bipy ligand the catalytic efficiency of Fe(OTf)2 was not improved much. The work of NHC-Fe catalyst in carbonyl-ene reactionwas started.
2.3 Development of caffeine-derived NHC with iron as catalyst for carbonyl-ene reactions
Caffeine was alkylated with MeI or EtOTf by using literature methods to form the corresponding imidazolium salts, which are NHC precursors (scheme 2−10).27, 28, 29 Three NHC precursors (ligand
1, 2 and 3) from the literature were selected for this research. Through SN2 reaction, caffeine was methylated by methyl iodide, and a caffeinium iodide salt was formed as ligand 1. Under thermal conditions, caffeine reacted with ethyl trifluoromethane sulfonate in nitrobenzene to afford ethylated caffeine as ligand 2. Ligand 3 was prepared by anion exchange between ligand 1 and NH4PF6 in water solution and recrystallization.27 Ligand 1 has a coordinating anion I− while ligand 2 and 3 have non-coordinating anions NTf2− and PF6−. Compared with ligand 1 and 3, ligand 2 has an ethyl group at the N9 position, which indicated that the imidazolium ring of ligand 2 is slightly more electron rich
48
from electron donating group -CH2CH3. Ligand 1-3, were used to fine-tuning the electronic peoperties of NHC-Fe complexes. NHCs of ligand 1-3 were generated-in-situ to coordinate with Fe(OTf)2.
Scheme 2−10 Formation of caffeinium salts by alkylation of caffeine
Methylated caffeine was used most frequently for NHC-metal complesation in literature. Accordingly, ligand 1 was as NHC precursor used with Fe(OTf)2 to catalyze the carbonyl-ene reaction of ethyl trifluoropyruvate (Scheme 2−11, Table 2−6). Excess of -methylstyrene (3 eq.) was used, and the ratio of NHC/Fe was set at 1:1. Commonly used base NaOtBu and KOtBu were compared in the reaction. A 51% yield was achieved with NaOtBu while a higher yield 75% (76%) was obtained using KOtBu (Table 2−6, entries 1 and 2). By adding more -methylstyrene (4:1 and 5:1) to boost the reaction, yields increased slightly to 79% and 82% respectively compared with 75% (Table 2−6, entries 3 and 4). Since with KOtBu, a better yield was obtained than when using NaOtBu, a change of base was performed from strong base NaH or KHMDS to weak base Cs2CO3 or K2CO3, and low to moderate yields were obtained (Table 2−6, entries 5-8). As carbonate salts, using Cs2CO3 the reaction was more efficient than with K2CO3 (Table 2−6, entries 5 and 8). Higher yields were expected when using NaH and KHMDS since the conjugate acids of the two bases have higher pKa than KOtBu and hence could more efficiently deprotonate caffeine-methylated iodide salt to generate NHC. However, these two bases are probably more moisture sensitive to the reaction than KOtBu, and lower yields were obtained. Thus, it needs more strict anhydrous conditions for using these two bases.
Based on the result that much excess of -methylstyrene (5 eq.) could afford an 82% yield (Table 2−6, entry 4), optimization of the reaction from using an excess of ethyl trifluoropyruvate (1.5 eq.) was performed, and an 80% yield was afforded (Table 2−6, entry 9). Further increasing the number of equivalents of an ethyl trifluoropyruvate (2 eq.) had little effect in improving the yield (Table 2−6, entry 10). 2 equivalents of NHC to iron was inefficient on the yield, and it decreased to 56% (Table
49
2−6, entry 11). Interestingly, upon adding twice the amount of -methyl styrene (from 0.25 mmol to
0.5mmol in 0.5 mL CH2Cl2) and ethyl trifluoropyruvate to reach a higher concentration of reactants from 0.5 mol/L to 1 mol/L, hence the catalyst loading was changed from 5 mol% [(0.25 mmol*5 mol% in 0.5 mL)/0.25 mmol] to 2.5 mol% [(0.25 mmol*5 mol% in 0.5 mL)/0.5 mmol], the yield was maintained at the same level (Table 2−6, entry 12). Even further reducing ethyl trifluoropyruvate from 1.5 eq. to 1.2 eq., a similar catalytic efficiency was kept (Table 2−6, entry 13). When using NaOtBu in the same conditions, the yield decreased slightly (Table 2−6, entry 14), but a higher yield of 85% was afforded by using NaH (Table 2−6, entry 15). A reverse condition of using an excess of - methylstyrene led to a lower yield at 65% (Table 2−6, entry 16). The results (Table 2−6) demonstrate that caffeine-derived NHC-Fe catalyst generated-in-situ was superior to iron salts mediated carbonyl- ene reaction of ethyl trifluoropyruvate. At a low catalyst loading and low ethyl trifluoropyruvate (1.2 equivalent) consuming (Table 2−6, entry 15), the reaction proceeded efficiently and achieved 85%. For further optimization of the reaction, we looked at the aspects of solvent, counterions (anions) of NHC ligands and iron salts.
Scheme 2−11 Ligand 1-derived NHC-Fe(OTf)2 catalyzed carbonyl-ene reaction
50
Entry 1 : 2 a X-Y-Z Base Yield (%)b
1 3 : 1 5-6-6 NaOtBu 51
2 3 : 1 5-6-6 KOtBu 75 (76)c
3 4 : 1 5-6-6 KOtBu 79 c
4 5 : 1 5-6-6 KOtBu 82 c
5 3 : 1 5-6-6 Cs2CO3 44
6 3 : 1 5-6-6 KHMDS 58
7 3 : 1 5-6-6 NaH 63
8 3 : 1 5-6-6 K2CO3 5
9 1 : 1.5 5-6-6 KOtBu 80
10 1 : 2 5-6-6 KOtBu 82
11 1 : 1.5 5-10-10 KOtBu 56
12 1 : 1.5 2.5-2.5-4 KOtBu 83
13 1 : 1.2 2.5-2.5-4 KOtBu 81
14 1 : 1.2 2.5-2.5-4 NaOtBu 77 c
15 1 : 1.2 2.5-2.5-4 NaH 85
16 2 : 1 2.5-2.5-4 NaH 65 a The limiting quantity of starting material was 0.25 mmol in 0.5 ml CH2Cl2 in the reaction; b isolated yields; c reaction time: 60 h.
Table 2−6 Optimization of Ligand 1-derived NHC-Fe(OTf)2 catalyzed carbonyl-ene reaction
The effect of solvent according to polarity difference was examined (Scheme 2−12, Table 2−7). From polar solvents such as H2O, DMF, NMP, MeCN, to less polar solvents EtOAc, DME, toluene, dimethyl carbonate and 1,4-dioxane, the yields were lower than using CH2Cl2 (Table 2−7, entry 1). On the
51
one hand, solvents bearing oxygen or nitrogen atoms poisoned the active catalytic iron, hence the yields dropped to a low level; on the other side, due to solubility issue, NHC-Fe was not applicable to toluene.
Scheme 2−12 Examination of different solvents in ligand 1-derived-NHC-iron catalyzed carbonyl-ene reaction
Entry 1 2 3 4 5 6 7 8 9 1,4- Solvents NMP DMF MeCN H O EtOAC DME Toluene DMC 2 dioxane Yield % 0 0 33 0 19 0 10 13 7
Table 2−7 Screening of different solvents in ligand 1-derived-NHC-iron catalyzed carbonyl- ene reaction
Ligand 2, the ethylated caffeine (1,3,7-trimethyl-9-ethyl-xanthine) was used as NHC ligand to iron in the carbonyl-ene reaction (Scheme 2−13, Table 2−8). Various iron salts were tested including different oxidation state Fe0, Fe2+, and Fe3+. Yields (Table 2−8, entries 1-6) obtained in three days were all lower than when using Fe(OTf)2 (Table 2−8, entry 7). A Variation of NHC loadings from 1:1 to 2:1, had a negative impact on the yield which decreased to 42% (Table 2−8, entry 8). A higher NHC loading to 3:1 led to a decrease of the yield dropped to 33% (Table 2−8, entry 9). It indicated that a ratio of NHC/Fe at 1:1 was more efficient than ratios at 2:1 or 3:1 in catalyzing the reaction. More NHC to iron deactivated the Lewis acidity of iron.
52
Scheme 2−13 Ligand 2-derived-NHC with different iron salts in catalyzing carbonyl-ene reaction
Entry a Iron salt X Y Yield (%) b
1 FeCl2 6 7 36
2 FeBr2 6 7 42
3 Fe(OAc)2 6 7 32
4 Fe(acac)3 6 7 44
5 Fe(CO)5 6 7 30
6 Fe(ClO4)2·6H2O 6 7 42
7 Fe(OTf)2 6 7 54
8 Fe(OTf)2 c 10 10 42
9 Fe(OTf)2 15 15 33 a 1/2 = 2:1, 0.5 mmol -methylstyrene and 0.25 mmol ethyl trifluoropyruvate were used; b isolated yields; c 10 mol% of Fe(OTf)2.
Table 2−8 Screening of different iron salts with Ligand 2-derived-NHC in catalyzing carbonyl-ene reaction of ethyl trifluoromethyl pyruvate
Optimization was performed from the aspects of solvents and quantities of catalyst (Scheme 2−14, Table 2−9). Based on the results of Table 2−7 of using ligand 1 in different solvents, a few solvents
53
were chosen. MeCN and toluene led to 30% yields (Table 2−9, entries 1 and 2), and no conversion was found in DME and THF (Table 2−9, entries 3 and 4).
Scheme 2−14 Study of solvent effect using ligand 2-derived-NHC iron catalyzed carbonyl- ene reaction
In a chlorinated solvent, such as dichloroethane and dichloromethane, the reaction proceeded smoothly to give similar yields (Table 2−9, entries 5 and 6). However, using a ratio of 1/2 = 3:1, DCM led to a higher yield (72%) than DCE (61%) (Table 2−9, entries 7 and 8). By further increasing the
Entry a 1/2 Solvent Yield (%) b
1 2 : 1 MeCN 30 2 2 : 1 Toluene 30 3 2 : 1 DME − 4 2 : 1 THF − 5 2 : 1 DCE 55 6 2 : 1 DCM 56 7 3 : 1 DCM 71 8 3 : 1 DCE 61 9 5 : 1 DCM 72 10 1 : 1.5 DCM 61 a The limiting quantity of starting material was 0.25 mmol in 0.5 ml CH2Cl2 in the reaction; b Isolated yields.
Table 2−9 Screening of various solvents using ligand 1-derived-NHC iron catalyzed carbonyl-ene reaction
54
amount of methylstyrene at 5:1 ratio, the yield was not improved (Table 2−9, entry 9). Excess of ethyl trifluoropyruvate (1.5 eq.) led to a decreased yield (Table 2−9, entry 10). From this optimization, the reaction was improved from 54% (Table 2−8, entry 7) to 71% (Table 2−9, entry 7).
A marked effect was reported from the influence of counterion In Lewis acid-(Pd)-catalyzed reaction on conversions and enantioselectivities (SbF6ˉ ≈ BF4ˉ ˃ ClO4ˉ ˃ OTfˉ).29 OTfˉ Anion has also been tested [Fe(OTf)2] (Scheme 2−(1-4)). In order to examine the effect of non-coordinating anions such as SbF6ˉ and BF4ˉ with ligand 1−3, a series of reactions were conducted (Scheme 2−15, Table 2−10).
Anhydrous irons salts Fe(BF4)2 and Fe(SbF6)2 were prepared from FeCl2 and AgBF4 or AgSbF6, respectively. Similar methods for generating metal salts with non-coordinating anions could be found.
In Luo’s work, chiral diphosphines-Pd(SbF6)2 were prepared from diphosphines-PdCl2, and
AgSbF6.30 Silver salts were used to activate the NCN-Ru catalyst in Song’s work.31 In Loh’s method,
InCl3 and AgSbF6 were mixed to prepare pyridine-bisoxazoline-In3+ catalytic active complex with noncoordinating anion SbF6ˉ. In the examples above, the precipitated AgCl by the addition of silver salt on the metal chloride had no influence on the enantioselectivity, which indicated that AgCl was not effective to promote the reaction. In this work, AgCl was filtered off to exclude any catalytic activity.
Anhydrous Fe(BF4)2, prepared from FeCl2 and AgBF4, led to a 82% yield and side products of methyl styrene (Table 2−10, entry 1). Although the yield obtained when using anhydrous Fe(BF4)2 alone was encouraging, decreased yields were obtained when the conditions were applied to other substrates. Fe(BF4)2 with NHC from ligand 1 afforded a reduced yield of 54% (Table 2−10, entry 2). When ligand 2 was used, the yield reached 84%, and an even higher yield was achieved by using ligand 3 (Table 2−10, entries 3 and 4). The yield decreased in the order of PF6ˉ > NTf2ˉ > Iˉ, which were from the ligand when used Fe(BF4)2 (Table 2−10, entries 2-4).
Scheme 2−15 Ligand 2 derived-NHC with iron salts of non-coordinating anions
55
Entry a FeCln Ag salt Ligand Yield (%) b
1 FeCl2 AgBF4 − 82
2 FeCl2 AgBF4 1 54
3 FeCl2 AgBF4 2 84
4 FeCl2 AgBF4 3 88
5 FeCl2 AgSbF6 − 73
6 FeCl2 AgSbF6 2 88
7 FeCl2 AgSbF6 3 84
8 FeCl2 AgSbF6 2 90
9 FeCl2 AgBF4 2 78
10 FeCl2 AgSbF6 2 81 a With 1/2 at 2:1 ratio for entry 1-7; 1/2 = 1:1.5 for entry 8; 1/2 = 1:1.2 for entry 9 and 10; b Isolated yields.
Table 2−10 Development of anhydrous Fe(BF4)2 and Fe(SbF6)2 with ligand 1-3
Anhydrous Fe(SbF6)2, prepared from FeCl2 and Ag(SbF6)2 led to a 73% yield (Table 2−10, entry 5).
An increased yield (88%) was obtained when using ligand 2 and Fe(SbF6)2 compared with Fe(BF4)2 (Table 2−10, entry 6). A slight decrease in yield from 88% to 84% was observed by using ligand 3 (Table 2−10, entry 7). Ligand 2 was derived from ligand 1 by anion exchange; thus ligand 2 was not selected for further reactions. In further optimization by using an excess of ethyl trifluoropyruvate, the yield reached 90% (Table 2−10, entry 8) and was higher than using the conditions of ligand 1 (Table 2−6, entry 15). However, when conditions of concentration and catalyst loading in ‘Table 2−6, entry 15’, in which the yield was 85%, were applied to ligand 2 with Fe(BF4)2 or Fe(SbF6)2, lower yields (78% and 81%) were obtained. Thus, ligand 2 derived-NHC-Fe(SbF6)2 was more efficient to catalyze the reaction.
56
FeIII+ salts were also studied. Fe(SbF6)3, prepared from anhydrous FeCl3 and Ag(SbF6)2, was examined with ligand 2 in the carbonyl-ene reaction (Scheme 2−16, Table 1−11). For testing FeCl3 with ligand 2, factors were considered from the ratio of NHC and AgSbF6 to iron to generate different active species. Firstly, one equivalent of AgSbF6 was used to prepare [FeCl2(SbF6)], and using one equivalent of NHC the yield was 53% (Table 2−11, entry 1). An active catalyst (NHC)2-FeCl2(SbF6) made from two equivalents of NHC improved the yield from 53% to 62% (Table 2−11, entry 2).
FeCl(SbF6)2 was prepared from using two equivalents of AgSbF6, and higher yields (73% and 78%) were obtained from using 1 eq. and 2 eq. of ligand 2 (Table 2−11, entries 3 and 4). Three equivalents of AgSbF6 were used to prepare Fe(SbF6)3, and yields were decreased to 50% and 69%, respectively (Table 2−11, entries 5 and 6).
Scheme 2−16 Development of ligand 2-derived-NHC-FeIII in carbonyl-ene reaction
Designed composition Entry a X Y Z Yield (%) b of active species
1 5 5 6 NHC-FeCl2(SbF6) 53
2 5 5 11 (NHC)2-FeCl2(SbF6) 62
3 5 10 6 NHC-FeCl(SbF6)2 73
4 5 10 11 (NHC)2-FeCl (SbF6)2 78
5 5 15 6 NHC-Fe(SbF6)3 50
6 5 15 11 NHC-Fe(SbF6)3 69 a with 1/2 = 1:1.5 for all entries; b isolated yields.
Table 2−11 Variation of the composition of the desirable catalytic species in the reaction by
using FeCl3 and AgSbF6 with ligand 2
57
The combination of anhydrous Fe(SbF6)2 with ligand 2 (Table 2−11, entry 8) was selected as the best condition and was applied to study various alkenes with ethyl trifluoropyruvate (Scheme 2−17, Figure2−3). Initially, the ene-reaction product of -methylstyrene 3a was obtained with a 90% yield. However, the yield decreased to 79% when the conditions were applied to p-Me-methylstyrene to afford 3b. A more electron-donating group as para methoxy led to 3c in quantitative yield. Due to steric hindrance of an ortho-methoxy group on the phenyl ring, 3f was obtained in a lower yield (75%). Moreover, with halogen substituted aromatic rings: p-Cl/Br/F -methylyrene, the yields of products 3d, 3e and 3f dropped to very low level. The reasons were possibly from two aspects: the electron withdrawing halogens decreased the nucleophilicity of the alkene or there was an affinity between NHC-Fe2+ and halogens or halogen–carbon bonds.32 Furthermore, the yield of 3i decreased to 61%, probably because 2-naphthyl iso-propene was less nucleophilic, or NHC coordinated Fe2+ had an affinity with naphthyl rings.33 Methylenecyclohexane and methylenecyclopentane afforded good to very high yields of 3g and 3k, respectively. High yield was obtained by using thiophene derived alkene (3j) as nucleophile. A moderate yield was achieved when using 2-methoxy-propene (3e). 2-Methoxy-propene was partially possibly decomposed due to its light sensitivity in the reaction process.34 Finally, the catalytic system was the least efficient toward mono-substituted alkyl alkenes such as 1-hexene 3m.
Scheme 2−17 Testing different alkenes with the catalyst of ligand 2 derived-NHC-Fe(SbF6)2 in carbonyl-ene reaction of trifluoropyruvate
When methylenecyclohexane was used as a nucleophile, the yield reached 80% only when 2 equivalents of ligand 2 were used, and the catalyst generated-in-situ was [(NHC)2Fe(SbF6)2]. Thus, a further optimization of the carbonyl-ene reaction of the substrate methylenecyclohexane was conducted by using ligand 1 and 2 respectively (Scheme 2−18, Table 2−12; Scheme 2−19, Table 2−13). The best conditions of involving ligand 1 (Table 2−6, entry 15), in which the concentration of the limiting substrate methylstyrene was 1 mol/L, and the catalyst loading was 2.5 mmol%, were
58
employed for testing the reaction firstly, and a low yield (59%) was obtained (Table 2−12, entry 1). Based on this result, increasing the amount of ethyl trifluoropyruvate from 1:1.2 to 1:1.5 led to a high-
a with 1/2 = 1:1.5 for all entries, isolated yields; for substrate of product 3i, 11 mol% of ligand 2 was used, the generated catalyst was [(NHC)2Fe(SbF6)2].
Figure 2−3 Reaction scope of ligand 2 derived-NHC-Fe(SbF6)2 catalysis in carbonyl-ene reaction
-er yield 71% (Table 2−12, entry 2). The yields were lower than expected; thus the conditions involving ligand 1 (Table2−9, entry 9) which led to a yield of 80% of the product of methylstyrene were tested (Table 2−12, entry 3). But an even lower yield 38% was afforded. After that, a bis-NHC to iron composition of catalyst was used, and the relative catalyst composition was
[(NHC)2Fe(SbF6)2] which derived from ligand 1. The yield was increased from 38% to 57% (Table 2−12, entries 3 and 4). By using of ligand 1 derived NHC, the highest yield reached 71%. Further experiments were performed by using ligand 2.
59
Scheme 2−18 Further optimization of using substrate mehylenecyclohexane by ligand 1 and
Fe(OTf)2
Entry a 1I/2 X-Y-Z Time (h) Yield (%)b
1 1 : 1.2 2.5-2.5-3 60 59
2 1 : 1.5 2.5-2.5-3 48 71
3 1 : 1.5 5-5-6 48 38
4 1 : 1.5 5-5-10 60 57 a for entries 1 and 2, 0.5 mmol methylenecyclohexane and 1.2 eq. of ethyl trifluoroypruvate were used in 0.5 ml CH2Cl2; for entries 3 and 4, 0.25 mmol methylenecyclohexane and 1.5 eq. ethyl trifluoropyruvate were used in 0.5 ml CH2Cl2. b isolated yields.
Table 2−12 Further optimization for the reaction of mehylenecyclohexane using ligand 1 and
Fe(OTf)2
One of the best conditions in Table 2−10 (entry 6) was applied to the reaction of methylenecyclohexane and ethyl trifluoropyruvate. The relative active catalyst was 5 mol% ligand 2 derived-NHC-Fe(SbF6)2 with a 2:1 ratio of 1I/2, a 70% yield was afforded (Table 2−13, entry 1). Increasing one more equivalent of alkene (3 equiv. methylenecyclohexane) loading, the yield was increased to 72% (Table 2−13, entry 2). By just precipitating one chloride anion to generate a NHC-
FeClSbF6 catalyst and loading more excess of ethyl trifluoropyruvate, a lower yield 62% was achieved (Table 2−13, entry 3). 10 mol% of NHC-Fe(SbF6)2 led to 71% yield similar as entries 1 and
2 (Table 2−13, entry 4). Optimization of the factors of starting materials, NHC-FeClSbF6 and NHC-
Fe(SbF6)2 could not lead a yield higher than 72%, hence an (NHC)2Fe(SbF6)2 (with two equivalents
60
of NHC) catalyst composition was explicitly designed for methylenecyclohexane, and a higher yield 80% was obtained (Table 2−13, entry 5).
Scheme 2−19 Further optimization of using substrate methylenecyclohexane with ligand 2
and Fe(SbF6)2
Entry a 1I/2 X-Y-Z Yield (%) b
1 2 : 1 5-10-6 70
2 3 : 1 5-10-5 72
3 1 : 1.5 5-5-6 62
4 1 : 1.5 10-20-10 71
5 1 : 1.5 5-10-12 80 a For entries 1 and 2, 0.5 mmol methylenecyclohexane and 1.2 eq. of ethyl trifluoroypruvate were used in 0.5 ml CH2Cl2; b Isolated yields.
Table 2−13 Further optimization of using substrate mehylenecyclohexane with ligand 2 and
Fe(SbF6)2
According to the results above, we got preliminary information about the reactivity of NHC-Fe(SbF6)2 catalyst in an intermolecular carbonyl-ene reaction. It was also fascinating to test the NHC-Fe chemistry in the intramolecular carbonyl-ene reaction. In the catalytic system involving Fe(BF4)2, by using (S)-(−)-citronellal 4, an overall 70% yield was obtained in a 70:30 trans/cis ratio of (+)- isopulegol to (−)-neo-isopulegol in the crude product, while the isolated yields were: 45% for (+)- isopulegol 5, 25% for (−)-neo-isopulegol 6, and no (−)-iso-isopulegol or (−)-neo-iso-isopulegol was
61
identified (Scheme 21). Hence, NHC-Fe(SbF6)2 catalytic system was applied to the intra-molecular carbonyl-ene reaction of (S)-(−)-citronellal (Scheme 2−20, Table 2−14). FeCl3 (10 mol%) was used as a catalyst in benzene for the cyclization of citronellal in a short reaction time (20 min.), and a low yield (20%) with a 76:24 ratio of isopulegol 5 to other diastereomers was obtained. Firstly, FeCl2 and
FeCl3 as transition metal Lewis acid were mediated in the reaction, and the results were compared to NHC-Fe(SbF6)2 catalyst. Considering the reaction time while compared with the one in the literature, the yields were reasonable (47% and 40% for (+)-isopulegol as major products Table 2−14, entries 1 and 2). But with FeII, neither NHCFe(SbF6)2 nor NHCFeCl(SbF6) led to any conversion (Table 2−14, entries 3 and 4). NHC-FeII was inefficient in intramolecular carbonyl-ene reaction of citronellal. Hence, NHC-iron catalysts based on FeIII were designed and optimized in the carbonyl- ene reaction of citronellal. When three Cl¯ were all precipitated by using three equivalents of silver salts, NHCFe(SbF6)3 did not lead to any conversion of the reaction (Table 2−14, entry 5). But when two Cl− were precipitated from FeCl3, there was 46% conversion, including 32% (+)-isopulegol and 12% (−)-neo-isopulegol (Table 2−14, entry 6). A hypothesis is that Cl¯ could adjust the charge on iron center and modify the Lewis acidity of NHC-FeIII catalyst. In this way, one Cl− was precipitated, and an NHCFeCl2(SbF6) catalyst led to a higher yield of 78%, including 70% (+)-isopulegol and 8% of (−)-neo-isopulegol (Table2−14, entry 7). Conclusively, FeCl2 and FeCl3 were effective in catalyzing the intra-molecular carbonyl-ene reaction of citronellal at moderate yields (58-61) with trans/cis ratio around 70/30. But in the catalytic system of ligand 2-derived NHCFeCl2(SbF6), the reaction proceeded more efficiently and selectively toward (+)-isopulegol.
Scheme 2−20 Development of NHC-iron catalyst in the intramolecular carbonyl-ene reaction of the cyclization of citronellal
62
Yield (%) b Entry a FeCln X Trans/cis 5/6/7/8
1 FeCl2 − 72/28 47/14/−/−
2 FeCl3 − 69/31 40/18/−/−
3 FeCl2 20 − −
4 FeCl2 10 − −
5 FeCl3 30 − −
6 FeCl3 20 76/24 32/14/−/−
7 FeCl3 10 80/20 70/8/−/−
a The reactions were performed at room temperature. In entries 1 and 2, only FeCl2 and FeCl3 salts were used as Lewis acid respectively, there were no use of ligand, silver salt or base; b isolated yields.
Table 2−14 Optimization of NHC-iron catalyst in the intramolecular carbonyl-ene reaction of the cyclization of citronellal
2.4 Mechanism of caffeine-derived NHC-iron catalysts for carbonyl-ene reactions and the withdrawing property of caffeine’s pyrimidine-dicarbonyl structure
The mechanism of anhydrous Fe(BF4)2 catalyzed inter- and intra-molecular carbonyl-ene reaction was postulated (illustrated in Figure 2−3), in which FeII acts as Lewis acid and coordinates with the carbonyl groups of ethyl trifluoropyruvate. Based on this, the mechanism of NHCFeCl2(SbF6) catalyzed carbonyl-ene reaction was postulated as follows (Figure 2−4). Anhydrous Fe(SbF6)2 is prepared from FeCl2 and AgSbF6 through filtration. An NHC-Fe2+ with non-coordinating anion SbF6 is formed through the complexation of Fe(SbF6)2 with caffeine-derived NHC which is generated-in- situ by deprotonation of ligand 2 by KOtBu. NHC-Fe2+ coordinated with trifluoropyruvate with the two carbonyl groups in equilibrium. The carbonyl group of the NHC-Fe2+ coordinated trifluoropyruvate should have a lower LUMO compared with non-coordinated trifluoropyruvate and is then attacked by
63
the nucleophilic alkene of -methylstyrene. Through a pericyclic process, a carbonyl-ene product was formed, and the dissociation between NHC-Fe2+ and the hydroxyl and carbonyl of the product released the product and the catalyst NHCFe(SbF6)2. Hence, a carbonyl-ene reaction proceeded by the catalyst of caffeine derived-NHC-FeII.
Figure 2−4 Postulated mechanism of the NHC-Fe catalyzed carbonyl-ene reaction of ethyl trifluoropyruvate and -methylstyrene
In order to get more insight on the mechanism, it was of interest to get some clue about the function of the ring structure of caffeine, an imidazole ring, and a pyrimidyl carbonyl ring. Ligand 4 (1,3- dimethyl-imidazolium iodide) and ligand 5 (1-ethyl-3-methyl-imidazolium bis-triflimide) were prepared by literature methods62 and tested as NHC ligands in the carbonyl-ene reaction of -methylstyrene and ethyl trifluoropyruvate (Scheme 2−21). The yields 39% and 60% were much lower than the
64
results of using ligand 1 (54%) and ligand 2 (90%) (Scheme 2−21). Although the imidazole ring parts have same structure between ligand 1 and ligand 4, ligand 2 and ligand 5 respectively, the yields of ligand 1 and ligand 2 were much higher than those of ligand 4 and 5. Without caffeine back ring, the yields decreased by 15% from using ligand 1 to 1,3-dimethyl-imidazolium iodide, and by 30% from using ligand 2 to 1-ethyl-3methyl-imidazolium bis-triflimide.
*1/2 = 1:1.5, in 0.5 mL CH2Cl2.
Scheme 2−21 Comparison of ligand 1 with ligand 4 and ligand 2 with ligand 5 in NHC-iron catalyzed carbonyl-ene reaction
The possible reason was explained from the main structural difference among the ligands (Figure 2−5). Firstly, the pyrimidyl carbonyl ring of caffeine-derived NHCs is very electron-withdrawing. An impact of the electron-withdrawing effect on caffeine derived carbene-metal bond would lead to a less electron rich NHC-iron catalyst, and without the pyrimidyl carbonyl ring, NHC-FeIIs derived from ligand 4 and 5 were more electron-rich than caffeine-derived NHC-FeII. Furthermore, there would be a more inductive effect from caffeine’s unique back ring compared with the hydrogens of ligand 4 and 5 derived NHCs, which attracts nitrogen long pairs. Hence, it is reasonable to expect less σ donation from NHC to iron or an increased π-back donation from iron filled-d orbital to the carbene’s empty π orbital or both cases at the same time. As a result, the iron center from caffeine-derived- NHC-FeII complexes was less basic or more Lewis acidic compared with NHC-FeII complexes which
65
were derived the ligand 4 and 5. In literature, it was reported that numerous NHCs were synthesized but few of them have shown increased catalytic activity compared with the initially introduced IMes and SIMes.169 In the case of NHC-iron catalysis, the applications were still limited. To expand this area, much effort in varying the electronic property and Lewis acidity of NHC-iron is necessarily in order to fine-tune the substrates in different organic transformations. Although optimization of the NHC ligands in NHC-iron catalysis was already performed by various means, for example: bidentate and tridentate NHC-iron complexes were used in cross-coupling32 and polymerization35, monodentate NHC-iron with assisting ligands such as Cp-, CO, NO, halogens etc. in reactions such as arylation,36 hydrosilylation37 etc., to improve the situation, simpler and more effective methods are in high preference. However, on the purpose of fine-tuning the electronic property of NHC-iron to modify the Lewis acidity of iron, herein, the caffeine-derived NHC which has an electron withdrawing ring would be a promising solution to the problem.
Figure 2−5 Electronic property of caffeine derived-NHC-Fe bonding orbital
66
2.5 Quantitative determination of the free carbene from caffeine-derived xanthinium salt─ligand 1
Nolan reported the synthesis protocol of one of the most widely used free carbene, 1,3-bis(2,4,6- trimethylphenyl)imidazole-2-ylidene, IMes (Scheme 2−22).38 The isolation was due to the difference of solubility between free carbene and the resulting salts NaBF4, KBF4 or unreacted KOtBu. Free carbene was soluble in THF and hence by filtration unsoluble salts were separated. After evaporation of THF and HOtBu, the free carbene was isolated.
Scheme 2−22 Synthesis protocol of a free carbene, 1,3-bis(2,4,6-trimethylphenyl)imidazole- 2-ylidene, IMes
Nolan’s method was not applicable to the isolation of the free carbene of caffeine-derived NHC precursors since caffeine derived-NHCs have low solubility in THF. The effort was put into the isolation of the free carbene from caffeine-derived imidazolium salt ligand 1, in a qualitative manner.
Caffeine derived-NHC of ligand 1 was prepared using three solvents (CH2Cl2, THF, and a mixture of
DMF and THF respectively). Protons shifts of N1CH3, N3CH3, N7CH3 and C8H in caffeine molecule are at 3.41, 3.59, 3.99 and 7.50 ppm in CDCl3,39 but the 1H NMR of the four methyl groups and C8H of ligand 1 shifted downfield to 3.41, 3.85, 4.25, 4.42 and 10.56 respectively due to the formation of imidazolium salt (Scheme 2−23 and Figure 2−6). The formation of the cation of the imidazolium/xanthinium salt (ligand 1) was electron deficient; hence proton NMR shifts of the adjacent methyl groups were identified (Figure 2−6). 1H NMR of N1CH3 was still at 3.41 ppm, 1H NMR of N3CH3 shifted from 3.59 to 3.85, and 1H NMR of N7CH3 shifted from 3.99 to 4.25. Δδ of the proton shifts of N3CH3 and N7CH3 in caffeine and ligand 1 were around 0.25. A large Δδ of the proton shifts on C8H was about 10.56-7.50 = 3.06 ppm. The phenomenon of the proton shifts of caffeine and
67
ligand 1 is a manner of understanding the electronic property of the relative NHC; hence, the ligand 1 derived-NHC was produced by using three methods, and the spectrums were taken (Scheme 2−(24-26), Figure 2−7).
Scheme 2−23 Structures of caffeine and ligand 1
68
Figure 2−6 1H NMR of caffeine and ligand 1
The formation of caffeine derived-NHC by the deprotonation at C8H gave back the pair electrons of N9 or N7 by resonance (the orbitals of the two nitrogen atoms in the imidazolium ring stabilizing the empty orbital of the carbene carbon C8), and hence a neutral carbene was born and was not electrons-withdrawing from neighboring methyl groups compared with ligand 1. The change was observed by the shifts of the 1H NMR of the four methyl groups. The 1H NMR of the four methyl groups shifted upfield from 3.41, 3.85, 4.25, 4.42 of ligand 1 to 3.38, 3.59, 3.64, 3.73 of the free carbene respectively (Scheme 2−(24-26), Figure 2−(7-9)). In the first method, solvent CH2Cl2 and an excess amount of base NaH were used (Scheme 2−24). Peaks at 0.88 and 1.25 ppm were from hexane in the mineral oil of NaH base (Figure 2−7). In the second method, THF was used as a solvent, and over amount of base NaH and a catalytic amount of KOtBu acted as bases (Scheme
2−25). Peaks at 1.85 and 3.74 ppm were from THF, and surprisingly an H2O peak at 1.56 ppm was identified (Figure 2−8). In the third method, ligand 1 was more soluble in DMF, hence 1.5 eq. base NaH was enough to deprotonate NHC precursor but a longer reaction time was necessary (Scheme 2−26). Peaks of DMF in 1H NMR spectrum were identified at 2.88, 2.96 and 8.02 ppm (Figure 2−9).
69
Scheme 2−24 Determination of caffeine derived NHC by method 1
Figure 2−7 Determination of caffeine derived NHC using method 1
Scheme 2−25 Determination of caffeine derived NHC using method 2
70
Figure 2−8 Determination of caffeine derived NHC using method 2
Scheme 2−26 Determination of caffeine derived NHC using method 3
71
Figure 2−9 Determination of caffeine derived NHC by method 3
The free caffeine derived-NHC was determined qualitatively in three different solvents. MS analysis of the free carbene from the three methods showed m/z = 227.11325 (m/z of NHC = 208, m/z = 227 explained as NHC+H2O+H+), which was quite different from ligand 1, the caffeinium salt, [m/z =
209.10287 exclusively, ESI-MS (m/z) = 209 (C9H13N4O2+)40]. And trace of dimer was identified in the sample of using THF or DMF as solvents with m/z = 475.20174 (dimer = 416 + MeCN, NH4+),
435.2084 (dimer = 416 + H2O + H3O+), 227.11323 (NHC = 208 + H3O+), 199.11810 (NHC – C + 2H + H+ = 208-12+3 = 199). This evidence demonstrated for the first time the determination of the free carbene of caffeine derived-imidazolium salt and the formation of NHC in the caffeine derived-NHC- iron catalysis in carbonyl-ene reaction.
2.6 Synthesis of caffeine-derived NHC-Iron complexes
Since the carbene from caffeine derived-imidazolium salt was clearly identified using these three methods, the relative NHC-iron complex was in high interest if it could be observed. Hence, various conditions had been used for the synthesis of caffeine-derived NHC-iron complexes (Scheme 2−27). But all the synthesized caffeine derived-NHC-Fe complexes were not fully identified due to the
72
solubility issue (Table 2−15). The yields were calculated according to the relative desired products in Scheme 2−27. The synthesis from entries 1−6, 8 and 12 failed, and the results could not be clearly identified. Complexes showed brick color but were insoluble in common solvents such as DMSO,
H2O, THF, CH2Cl2, MeOH, EtOAc etc. (entries 7, and 9−11).
Scheme 2−27 Synthesis of caffeine-derived NHC-iron complexes
Yield Entry a FeX Base Solvent Conditions n (%)
1 1 Fe(OTf)2 KOtBu DCM − −
2 2 Fe(OTf)2 KOtBu DCM − −
3 1 Fe(OAc)2 KOtBu THF Reflux 3 h −
4 2 Fe(OAc)2 KOtBu THF Reflux 3 h −
5 1 FeCl2 KOtBu DCM+THF − −
6 3 FeBr2 KOtBu THF Reflux 3 h −
7 1 FeCl2 NaH DCM − 97%
8 3 FeCl2 NaH 4 eq. DCM+DMF − −
9 1 FeBr2 NaH 4 eq. DCM − 96%
10 1 Fe(OAc)2 NaH 4 eq. DCM − 90%
11 1 Fe(OTf)2 NaH 4 eq. DCM − 99%
12 1 Fe(OTf)2 NaH DMF − −
13 1 − NaH 2 eq. DCM 4Å MS, 6h 100%
14 1 Fe(OTf)2 NaH 2 eq. DCM 4Å MS, 6h 100%
Table 2−15 Synthesis of caffeine-derived NHC-iron complexes
73
However, a modified method from ‘Scheme 2−24’ was utilized to complex with Fe(OTf)2, and a control experiment without using iron salt was performed (Table 2−15, entries 13 and 14). 1H NMR spectrum corresponding to entry 13 showed the formed NHC with unreacted NaH and mineral oil (Figure 2−10), while entry 14 showed four peaks (δ = 3.49, 3.34, 3.20, 3.05) shifted more to upfield than the NHC (δ = 3.74, 3.64, 3.59, 3.38) (Figure 2−11). The complex from entry 14 was insoluble in neither D2O nor DMSO, and not fully characterized. A comparison of the spectra of the four methyl groups of ligand 1 (δ = 4.42, 4.25, 3.85, 3.41), ligand 1-derived-NHC (δ = 3.74, 3.64, 3.59, 3.38) and the relative NHC with addition of Fe(OTf)2 (δ = 3.49, 3.34, 3.20, 3.05) indicated the possible complexation of NHC with the iron salt.
Figure 2−10 Caffeine derived-NHC generated without Fe(OTf)2 in CH2Cl2
74
Figure 2−11 Caffeine derived-NHC generated with Fe(OTf)2 in CH2Cl2
75
2.7 References
[1] Nishikawa, Y. and Yamamoto, H. 2013. Iron(II) Trifluoromethanesulfonate. e-EROS Encyclopedia of Reagents for Organic Synthesis. Fuchs, P. L.; Bode, J. W.; Charette, A. B.; Rovis, T.; Paquette, L. A. Eds.
[2] Kobayashi, S. Eur. J. Org. Chem. 1999, 15.
[3] (a) Gilbert, D. E.; Van Der Marel, G. A.; Van Boom, J. H.; Feigon, J. Proc. Natl. Acad. Sci., 1989, 86, 3006. (b) Prakash, G. K. S.; Do., C.; Mathew, T.; Olah, G. A. Catal. Lett. 2011, 141, 507. (c) Kobayashi, S.; Ogawa, C. Chem. Eur. J. 2006, 12, 5954. (d) Prakash, G. K. S.; Mathew, T.; Olah, G. A. Acc. Chem. Res. 2012, 45, 565.
[4] Hua, R. Curr. Org. Synth. 2008, 5, 1.
[5] (a) Ollevier, T.; Dé syroy, V.; Debailleul, B.; Vaur, S. Eur. J. Org. Chem. 2005, 4971. (b) Anderson, E. D.; Ernat, J. J.; Nguyen, M. P.; Palma, A. C.; Mohan, R. S. Tetrahedron Lett. 2005, 46, 7747.
[6] James, B, K. Acc. Chem. Res. 2005, 38, 671.
[7] (a) Zhao, Y,-J.; Li, B.; Tan, S.; Shen, Z.-L.; Loh, T.-P. J. Am. Chem. Soc. 2010, 132, 10242. (b) Tymann, D.; Klü ppel, A.; Hiller, W.; Hiersemann, M. Org. Lett. 2014, 16, 4062. (c) Okamoto, R.; Takeda, K.; Tokuyama, H.; Ihara, M.; Toyota, M. J. Org. Chem. 2013, 78, 93.
[8] (a) Misono, M.; Nojiri, N. Appl. Catal. 1990, 64, 1. (b) Akutagawa, S. In Chirality in Industry; Collins, A. N.; Sheldrack, G. N.; Crosby, J., Eds.; John Wiley, 1992, 313.
[9] Aggarwal, B. K.; Vennal, G. P.; Davey, P. N.; NewMan, C. Tetrahedron Lett. 1998, 39, 1997.
[10] Nakatani, Y.; Kawashima, K. Synthesis 1978, 147.
[11] Peidro, L.; Le Roux, C.; Laporterie, A.; Dubac, J. J. Organomet. Chem. 1996, 521, 397.
[12] Anderson, E. D.; Ernat, J. J. Nguyen, M. P.; Palma, A. C.; Mohan, R. S. Tetrahedron Lett. 2005, 46, 7747.
[13] Nakatani, Y.; Kawashima, K. Synthesis 1978, 147.
[14] (a) Isolated yields were given for 5 and 6, 1H NMR spectra in agreement with: (b) Kropp. P. J.; Breton, G. W.; Graig. S. L. Crawford, S. D.; Durland, W. F., Jr.; Jones, J. E., III; Raleigh, J. S. J.
76
Org. Chem. 1995, 60, 4146. (c) Imachi, S.; Owada, K.; Onaka, M. J. Mol. Catal. A: Chem. 2007, 272, 174. (d) Jacolb, R. G.; Perin, G.; Loi, L. N.; Pinno, C. S.; Lenardã o, E. J. Tetrahedron Lett. 2003, 44, 3605.
[15] Doherty, S; Knight, Julian G.; Smyth, C. H.; Harrington, R. W.; Clegg, W. J. Org. Chem. 2006, 71, 9751.
[16] (a) Evans, D. A.; Tregay, S. W.; Burgey, C. S.; Paras, N. A.; Vojkovsky, T. J. Am. Chem. Soc. 2000, 122, 7936. (b) Morao, I.; McNamara, J. P.; Hillier, I. H. J. Am. Chem. Soc. 2003, 125, 628. (c) Thaler, W. A.; Franzus, B. J. Org. Chem. 1964, 29, 2226. (d) Huisgen, R.; Pohl, H. Chem. Ber. 1960, 93, 527. (e) Walling, C.; Thaler, W. J. Am. Chem. Soc. 1961, 83, 3877.
[17] Meng, D.; Ollevier, T. Synlett 2018, 29, 640.
[18] (a) Ebrahimi, A.; Habibi-Khorassani, M.; Akher, F. B.; Farrokhzadeh, A.; Karimi, P. J. Mol. Graph. Model. 2013, 42, 81. (b) Huang, H.-J.; Lee, W.-C.; Yap, G. P. A.; Ong, T.-G. J. Organomet. Chem. 2014, 761, 64.
[19] Bantreil, X.; Navals, P.; Martinez, J.; Lamaty, F. Eur. J. Org. Chem. 2015, 417.
[20] E. B. Leffler, H. M. Spencer, A. Burger, J. Am. Chem. Soc. 1951, 73, 2611.
[21] Hazra, D. K.; Lahiri, S. C. Anal. Chim. Acta. 1975, 79, 335.
[22] Scarborough, C. C.; Wieghardt, K. Inorg. Chem. 2011, 50, 9773.
[23] Cozzi, P. G., Gualandi, A., Mengozzi, L. and Manoni, E. 2016. Iron(2+), tris(2,2′-bipyridine-N,N′ )-, Dibromide, (OC-6-11). e-EROS Encyclopedia of Reagents for Organic Synthesis. 1–4. Fuchs, P. L.; Bode, J. W.; Charette, A. B.; Rovis, T.; Paquette, L. A. Eds.
[24] Gualandi, A.; Marchini, M.; Mengozzi, L.; Natali, M.; Lucarini, M., Ceroni, P.; Cozzi, P.G., ACS Catal. 2015, 5, 5927.
[25] Ollevier, T.; Plancq, B. Chem. Commun. 2012, 48, 2289.
[26] (a) Kascatan-Nebioglu, A.; Melaiye, A.; Hindi, K.; Durmus, S.; Panzner, J. M.; Hogue, A. L.; Mallett, J. R.; Hovis, E. C.; Coughenour, M.; Crosby, D. S.; Milsted, A.; Ely, L. D.; Tessier, A. C.; Cannon, L. C.; Youngs, W. J. J. Med. Chem. 2006, 49, 6811. (b) Kascatan-Nebioglu, A.; Panzner, M. J.; Tessier, C. A.; Cannon, C. L.; Yongs, W. J.; Coord. Chem. Rev. 2007, 251, 884.
77
[27] Luo, F.-T.; Lo, H.-K. J. Organomet. Chem. 2011, 696, 1262.
[28] Pinto, R. M. A.; Salvador, J. A. R.; Le Roux, C. Catal. Commun. 2007, 9, 465.
[29] Doherty, S.; Knight, G. J.; Smyth, H. C. Harrington, R. W.; Clegg, W. J. Org. Chem. 2006, 71, 9751.
[30] Rueping, M.; Theissmann, T.; Kuenkel, A.; Koenigs, R. M. Angew. Chem., Int. Ed. 2008, 47, 6798.
[31] Rueping, M.; Bootwicha, T.; Kambutong, S.; Sugiono, E. Chem. Asian. J. 2012, 7, 1195.
[32] (a) Bedford, R. B.; Betham, M.; Bruce, D. W.; Danopoulos, A. A.; Frost, R. M.; Hird, M. J. Org. Chem. 2006, 71, 1104. (b) Mo, A.; Zhang, Q.; Deng, L. Organometallics 2012, 31, 6518. (c) Meyer, S.; Orben, C. M.; Demeshko, S.; Dechert, S.; Meyer, F. Organometallics 2011, 30, 6692. (d) Wu, J.; Dai, W.; Farnaby, J. H.; Hazari, N.; Le Roy, J. J.; Mereacre, V.; Murugesu, M.; Powell, A. K.; Takase, M. K. Dalton Trans. 2013, 42, 7404. (e) Ghorai, S. K.; Jin, M.; Hatakeyama, T.; Nakamura, M. Org. Lett. 2012, 14, 1066. (f) Silberstein, A. L.; Ramgren, S. D.; Garg, N. K. Org. Lett. 2012, 14, 3796. (g) Agrawal, T.; Cook, S. P. Org. Lett. 2013, 15, 96. (h) Guisá n-Ceinos, M.; Tato, F.; Bunuel, E.; Calle, P.; Cá rdenas, D. J. Chem. Sci. 2013, 4, 1098.
[33] Blom, B.; Tan, G.; Enthaler, S.; Inoue, S.; Epping, J. D.; Driess, M. J. Am. Chem. Soc. 2013, 135, 18108.
[34] Whitaker, K. S. and Whitaker, D. T. 2001. 2-Methoxypropene. e-EROS Encyclopedia of Reagents for Organic Synthesis. Fuchs, P. L.; Bode, J. W.; Charette, A. B.; Rovis, T.; Paquette, L. A. Eds.
[35] Louie, J.; Grubbs, R. H. Chem. Commun. 2000, 1479.
[36] Bé zier, D.; Sortais, J.-B.; Darcel, C. Adv. Synth. Catal. 2013, 355, 19.
[37] (a) Jiang, F.; Bé zier, D.; Sortais, J. B.; Darcel, C. Adv. Synth. Catal. 2011, 353, 239. (b) Bé zier, D.; Jiang, F.; Roisnel, T.; Sortais, J. B.; Darcel, C. Eur. J. Inorg. Chem. 2012, 1333. (c) Lopes, R.; Cardoso, J. M. S.; Postigo, L. Royo, B. Catal. Lett. 2013, 143, 1061. (d) Warratz, S.; Postigo, L.; Royo, B. Organometallics 2013, 32, 893.
[38] Bantreil, X.; Nolan, P. S. Nat. Protoc. 2011, 6, 69.
78
[39] Sitkowski, J.; Stufaniak, L.; Nicol, L.; Martin, M. L; Martin, G. J.; Webb, G. A. Spectrochim. Acta. Part A 1995, 51, 839.
[40] Kascatan-Nebioglu, A.; Melaiye, A.; Hindi, K.; Durmus, S.; Youngs, W. J. Med. Chem. 2006, 49, 6811.
79
Chapter three
Development of caffeine-derived imidazolium-Fe(OTf)2 combined catalysts in the Diels-Alder reaction
3.1. Introduction and background
3.1.1 Catalyzed Diels-Alder reactions
The Diels−Alder (D−A) reaction is among the most powerful C─C bond forming transformations in synthetic chemistry (Scheme 3−1).1 ,-Dicarbonyl derivatives have been used as dienophiles in the enantioselective D-A reaction because these substrates could provide a rigid dicoordinating on the Lewis acid site in order to induce a high level of enantioselectivity control.2 This intriguing idea arose from the structural homology between a chiral enolate-based bond construction and N- acyloxazolinones which was potential chiral dienophile with a chiral auxiliary in D-A reaction. The reaction of N-acyloxazolinones with cyclopentadiene was described by Evans in which a stoichiometric amount of Et2AlCl was used as a mediator and excellent stereoselectivities were obtained (Scheme 3−2).3
Scheme 3−1 An example of asymmetric Diels-Alder reaction
Scheme 3−2 Prochiral center as stereochemical control element in N-acyloxazolinone in D-A
80
Later on, the reaction of achiral N-acyloxazolinones was reported by Corey (Bis-oxazoline(BOX)-
FeCl2I complex) and Narasaka (chiral diol-TiCl2 complex), respectively.4 High enantioselectivities had been obtained with these catalytic systems. Consequently, the use of this type of dienophile in the asymmetric D-A reaction and a benchmark D-A reaction between cyclopentadiene and α,β- unsaturated acryloyl-oxazolidin-2-one derivatives were studied extensively with many Lewis acids catalysts, such as Mg,5 Cu,6 Sc,7 Ti,8 Ln,2 Ni,9 Pd,10 Fe,11 Cr12.
Much progress was made in the development of Lewis acid catalyts, the design of superior ligands, improvement of stereoselectivity, effects from substrates, use of additives. Also, recycling of the catalyst has attracted considerable attention due to its high cost and substantial amount use initially for Box ligand-metal complex.13 Methods were designed, such as the immobilization of the catalyst on a polymer or inorganic material, developing heterogeneous catalysis or replacing the reaction solvent by ionic liquids, which were used initially to improve the stereoselectivities.14 As a result, ionic liquids (ILs) have received considerable attention as a powerful reaction media and as an alternative to conventional organic solvents.15a Furthermore, the combination of ionic liquids or ionic solids with transition metals has been reported as a promising area, and their scope of applicability is extending.15b,15c A catalyt-immobilization system of bimim-FeCl4 was developed for aryl Grignard cross-coupling via a liquid-liquid biphasic process, and FeCl3 was trapped in the ionic liquid.15d However, although ILs have many advantages from a green chemistry point of view, they suffer from high prices and other issues like separation of products and sensitivity to water or moisture, which more or less limit their application.15a, 16 Moreover, beside ILs there are still different ways for seeking eco-friendly sources of solvents or chemical processes.17 It appears that the most commonly used solvent in the literature for D-A reaction was the halogenated solvent dichloromethane, in some cases: chloroform, toluene or diethyl ether, and water for a few cases in heterogeneous catalysis. To promote the green aspects of the D-A reaction, we became interested in seeking a green solvent to replace the halogenated solvents.18a Dimethyl carbonate is an alkylating reagent and is classified as a green solvent.18b It is environmentally-benign and does not have irritating or mutagenic effect by neither contacting nor inhalation.18 It would be a highly desirable alternative solvent for D-A reaction.
81
3.1.2 Iron-catalyzed Diels-Alder reactions
3.1.2.1 Iron complexes catalyzed Diels-Alder reactions
Iron has attracted considerable attention for its green features: eco-friendiness, abundance, inexpensive price and its promising applications in organic synthesis.19 Examples are showed below
(Figure 3−1). Corey was the first to use a bisoxazoline (Box) ligand with FeCl2+I2 and obtained high enantioselectivity and endo/exo selectivity (95%, 96/4, 82% ee) in the above-mentioned D-A reaction.3 A few other C2-symmetrical ligands were used in this reaction, such as bis-sulfoxide and bisphosphine oxide and dibenzofurandiyl bisoxazoline.20-22 Khiar reported the good reactivity of bis- sulfoxide-Fe-I2 with these substrates.20 Matsukawa presented the synthesis of P-chirogenic diphosphine oxide ligands and the application of the corresponding FeIII complex in the catalytic asymmetric reactions of N-acrylamide dienophiles.21 However, the results with these two types of chiral iron complexes were not as good as chiral Box-iron complex from Corey’s group.22 The best results (90%, 99/1, 98% ee) were obtained by Kanemasa with the dibenzofurandiyl bisoxazoline ligand. Kanemasa disclosed a tridentate dibenzofuran-Box ligand which was tested with several metal salts and attained the best result with Fe(ClO4)2.23 In these results, the endo adduct was exclusively obtained while achieving 99% enantiomeric excess.
Figure 3−1 FeII/FeIII-catalyzed enantioselective Diel-Alder reaction
Kü ndig first investigated CpFeMe(CO)2 with a trans diol and a pentafluorophenyl derived complex for the dienophiles acrolein, methacrolein, methyl methacrylate and similar structure substrates
(Figure 3−2).24 Interestingly, a Schiff base FeCl2 complex was prepared for the dienophiles of acyl nitroso derivatives, and a Fe3+ porphyrin complex was capable of catalyzing the reactions of using various unactivated aldehydes (Figure 2−2).25-26 Progress was achieved from the study of many other substrates in iron catalysed D-A reaction such as ,-unsaturated acyl CpFe(CO)227, a -4
82
1,3-diene Fe(CO)3 induced enophile28, an o-Me3Si-phenyl triflate derived benzyne29 and a trans- methylene-1,3-dithiolane dioxide as dienophile in homogeneous catalysis30 and ,-unsaturated acetals,31 piperine,32 p-benzoquinones33 and imino D-A reaction in heterogeneous catalysis.34
Figure 3−2 Iron complexes catalyzed Diels-Alder reaction
FeIII as a Lewis acid was tested previously in the Diels-Alder reaction. FeCl3 was used as a catalyst in the study of the dimerization of 1,3-cyclohexadiene by Eberson.35 IronIII 2-ethyl hexanoate had been used as a novel and mild Lewis acid catalyst by Gorman for the stereoselective Diels–Alder reaction of ethyl (E)-4-oxobutenoate with alkyl vinyl ethers (Scheme 3−3).36
Scheme 3−3 FeIII as Lewis acid catalyzed Diels-Alder reaction
2.1.2.2 The use of caffeine as an imidazolium salt
Caffeine is a green natural abundant chemical37 and can be used as a precursor to synthesize imidazolium salts. Virtually, these are similar to 1,3-dialkyl imidazolium ionic liquids. Firstly, in the pioneering work by Le Roux, caffeine-based imidazolium salts combined with bismuth triflate were recyclable catalysts in the Diels-Alder reaction.38 Bi(OTf)3 was trapped as a caffeine-based salt (xanthinium salt), and the combined xanthinium-BiIII mixture was recyclable without losing its efficiency over several runs. Secondly, a combined effect of triflate and ionic liquid such as BMIM salt in promoting Diels-Alder reaction was reported.7 Moreover, ironII triflate (Fe(OTf)2) was chosen with ethylmethylimidazolium bistriflimide for aziridination.39 Thus, it is reasonable to expect that a
83
caffeine-based imidazolium salt, although solid at room temperature, combined with iron triflate
(Fe(OTf)2) could lead to similar enhancement of the D-A reaction properties as ionic liquids.
3.2. Development of caffeinium iron catalyzed Diels-Alder reactions
3.2.1 Development of the catalyst
Alkylated caffeine salts were synthesized by using ethyl triflate or methyl iodide reagents (Scheme
3−4).38, 40 By anion exchange of the alkylated caffeine, imidazolium salts with NTf2ˉ, Iˉ, and PF6ˉ as anions were obtained. In Le Roux's work, Bi(OTf)3·4H2O was used in 1:10 ratio with an imidazolium salt for Diels-Alder reaction.38
Scheme 3−4 Preparation of caffeine derived-imidazolium salts and the iron-caffeinium catalysts 1−4
84
Accordingly, the preparation of caffeine-based imidazolium salt-Fe(OTf)2/Fe(OTf)3 catalysts with different anions was designed. The ratio of Fe(OTf)2 to the imidazolium salt was examined from 2:1 to 10:1. We found that at 8:1 to 10:1 all the iron salt could be solubilized in acetone and there was no loss in the recyclability test of the imidazolium-Fe(OTf)2 complex. Acetone and heptane were indeed used in a recycling process for catalysts C1, C2, C3, and C4. The acetone-heptane solution was layered with a brown liquid, and a light brown solid was obtained after evaporation of the solvent. Catalyst C2 was brown-black, and catalyst C3 exhibited a white color. These iron complexes as well as Fe(OTf)2 alone were tested in the optimization of the Diels-Alder reaction of cyclopentadiene and 3-acryloyl-1,4-oxazolidin-2-one (Scheme 3−5). The four complexes had identical proton NMR spectra with the corresponding imidazolium salts and were characterized by infrared spectroscopy analysis.
3.2.2 Optimization of the reaction
First, Fe(OTf)2 and catalyst 1 were tested in the most commonly used solvent CH2Cl2. The ratio of diene to dienophile was set to be 7:1. The yields and stereoselectivity of Fe(OTf)2 (Table 3−1, entry
1) and catalyst C1 (Table 3−1, entry 2) in CH2Cl2 were similar. Then, higher yield and endo/exo selectivity were obtained by using dimethyl carbonate (DMC) as a green solvent (Table 3−1, entry 3). Less side product from the dimerization of cyclopentadiene was obtained at 2 ℃ vesus room temperature (Table 3−1, entry 4). Then, optimization was performed by lowering the ratio of reactants from 7:1 to 5:1 and down to 3:1 and a short reaction time (3 h). Consequently, quantitative yields with same endo/exo selectivity (91/9) were obtained (Table 3−1, entries 5 and 6). The control experiments were conducted by using Fe(OTf)2 and no catalyst, respectively, but lower yields were afforded (Table 3−1, entries 7 and 8). When further lowering the ratio of reactants to 2:1, the yield was still quantitative but the endo/exo selectivity reached 94/6 (Table 3−1, entry 9). Then, the other three catalysts C2, C3, and C4 were tested in the same conditions. With catalyst C2, endo/exo ratio increased slightly but the yield was reduced (Table 3−1, entry 10). Catalyst C3 could not reach the same efficiency as catalyst C1 (Table 3−1, entry 11). Using FeIII derived catalyst C4 the yield reached 97% but with an endo/exo ratio of 90/10 (Table 3−1, entry 12). Three control experiments were accordingly performed (Table 3−1, entries 13−15).
85
Scheme 3−5 Optimization of Diels-Alder reaction
Temp. endo/ Entry a 1a/2 Catalyst Time (h) Yield (%)b (℃) exo
1 7 : 1 Fe(OTf)2 r.t. 13 87/13 89
2 7 : 1 C1 r.t. 13 88/12 92
3 7 : 1 C1 r.t. 13 91/9 99
4 7 : 1 C1 2 3 91/9 99
5 5 : 1 C1 2 3 91/9 99
6 3 : 1 C1 2 3 91/9 99
7 3 : 1 Fe(OTf)2 2 3 89/11 84
8 3 : 1 − 2 3 93/7 46
9 2 : 1 C1 2 3 94/6 99
10 2 : 1 C2 2 3 94/6 33
11 2 : 1 C3 2 3 90/10 55
12 2 : 1 C4 2 3 90/10 97
13 2 : 1 - 2 3 90/10 34
14 2 : 1 Fe(OTf)2 2 3 90/10 81
15 2 : 1 Fe(NTf2)2 2 3 93/7 100 a Conditions: 1a (1 mmol), 2 (0.5 mmol), DMC (1 mL), isolated yields. b The reactions of entry 1 and 2 were run in CH2Cl2 (1 mL).
Table 3−1 Optimization of the Diels-Alder reaction
86
Without any catalyst the yield dropped to 34%, while using Fe(OTf)2 it was lower than using catalyst
C1. For comparing with catalyst C1, a quantitative yield was obtained when using Fe(NTf2)2 alone
(Table 2−1, entry 15), which indicated that Fe(NTf2)2 was generated-in-situ in the DMC solution when using catalyst C1. Compared with Fe(OTf)2, it demonstrates that catalyst C1 enhanced the yield and stereoselectivity. Although Fe(NTf2)2 alone was as efficient as catalyst C1, it appeared to be non- recyclable in the established conditions.
In the optimization process (Table 3−1), Fe(OTf)2 alone afforded good yields, but still lower than when using catalyst C1. Fe(OTf)2 was lost in the work-up step but catalyst C1 could be recycled by addition of hexane and filtration. Therefore, it was supposed that in the catalytic system there was an anion exchange between triflate and imide, and as a result Fe(NTf2)2 was generated in situ. Coordinating anion I¯ in catalyst C2 could not improve the catalytic effect. Low coordinating ligand
PF6¯ was not expected to afford moderate yield, but as described in Corma's article, the equilibrium
PF6¯ →PF5 + F¯ exists and generates the iron fluoride salt in situ, which may reduce the Lewis acidity of Fe2+. Another explanation would be that Fe2+ has some affinity toward PF6¯ .41 Thus, catalyst C3 did not promote the reaction as efficiently as catalyst C1. Catalyst C4 led to a lower endo/exo diaselectivity and was not as easily recycled as catalyst C1.
3.2.3 Optimization of the solvent
To examine the solvent effect, a few green solvents were selected for the reaction and compared with DCM and THF (Table 3−2). DCM led to the lowest endo/exo selectivity, while THF led to moderate yield (Table 3−2, entries 1 and 2). Me-THF afforded a low yield but the highest endo/exo selectivity (Table 3−2, entry 3). Cyclopentane-methyl ether, ethyl acetate, and methyl tert-butyl ether led to promising results (Table 3−2, entries 4−6), hence they could be suitable green solvents for the substrates of DA reaction. NMP in entry 5 would not be a solvent of choice due to its high boiling point which makes it difficult to seperate from the product (Table 3−2, entry 7). Of all the solvents that were tested, DMC provided high endo/exo ratio combined with an excellent yield and was thus selected for the next part of the study (Table 3−2, entry 8).
87
Entry a Solvent endo/exo Yield (%) b 1 DCM 87/13 92 2 THF 90/10 65 3 Me-THF 95/5 50 4 CPME 88/12 100 5 EtOAc 90/10 98 6 MTBE 89/11 97 7 NMP 94/6 85 8 DMC 94/6 99 a Conditions: 1a (1 mmol), 2 (0.5 mmol), C1 (1 mol %), DMC (1 mL), 2 ° C, 3h. b Isolated yields.
Table 3−2 Screening of selected “green” solvents
3.2.4 Recyclability test of catalyst 1
Scheme 3−6 Recyclability test using catalyst C1
Then, we decided to examine the recyclability of the catalyst in the D-A reaction (Scheme 3−6, Table 3−3). It provides the yields of the reaction and the recycled yields of the catalyst each run in 5 times reaction. 1 mol% catalyst was used for five runs, and after each run, hexane was added to the reaction mixture to precipitate the catalyst. Then, by filtration through a cotton-plugged pipet the catalyst was recollected and washed by acetone and finally recycled by evaporation. Because of the small quantity used, there was a small loss of catalyst after each run resulting in the decrease of both yields of the reaction and recycled yields of the catalyst. But when we increased the loading to 2 mol%, the catalyst was fully recycled after each run up to five times, and the reaction yield was maintained at the same level.
88
Entry/Run a 1 2 3 4 5
Yield (%) 99 98 96 95 95 Catalyst 100 99 97 95 94 recycled (%) Yield (%)a 99 99 99 99 99 Catalyst 100 100 100 100 100 recycled (%)b a Conditions: 1a (1 mmol), 2 (0.5 mmol), C1 (1 mol %), DMC (1 mL); isolated yields. b Reaction performed with 2 mol % of catalyst C1.
Table 3−3 Recyclability in 5 runs and yield of each run
3.2.5 Reaction scope
Finally, given the good reactivity of the above-mentioned caffeine-derived imidazolium-Fe(OTf)2 catalyst in DMC, we applied it to various dienophiles in the D-A reaction (Table 3−4). The substrate 3-crotonoyloxazolidinone required a longer time than acryloyl-oxazolidin-2-one (Table 3−4, entry 1) but afforded a slightly higher stereoselectivity (Table 3−4, entry 2). Instead of a Me- group, a β-CF3 to the carbonyl group led to a higher endo/exo selectivity and high yield (Table 3−4, entry 3). However, when cinnamoyl oxazolidinone was used as a substrate, solubility issues were met in all the solvents from the list, and there was only a trace of conversion (Table 3−4, entry 4). 2-Alkenoyl pyridines were also explored as bidentate dienophiles that can chelate to a metal center by pyridine and carbonyl lone pairs.42 Here, the objective was to test them in the caffeinium assisted-iron catalytic system (Table 3−4, entries 5−8). High yields and high endo/exo ratios were obtained. The stereoselectivity decreased when using substrate cinnamoyl pyridine N-oxide probably due to less coordination of the oxide with iron than that with pyridine (Table 3−4, entry 8). Being similar at the cinnamoyl part, benzylidine acetone and cinnaldehyde were chosen for comparison with the bidentate dienophiles in entry 4 and 5. The ketone (benzylidine acetone) showed good reactivity with higher endo/exo ratio, but the aldehyde hardly reacted even in a longer time (Table 3−4, entries 9 and 10). Using acryloyl chloride, a moderate yield (60%) was obtained (Table 3−4, entry 11). Neat conditions were explored efficiently for methyl acrylate at a very low catalyst loading 0.1 mol% with achieving 90% yield and 77/23 endo/exo selectivity (Table 3−4, entry 12), which was even better than some obtained results using ionic liquids but a low yield was obtained for methyl propiolate.43
89
Entry a Substrates Conditions endo/exo Yield (%)b
1 2 °C, 3 h 94/6 99
2 2 °C, 36 h 89/11 94
3 2 °C, 24 h 94/6 95
2 °C, 36 h,DMC/ 4 c − 5 acetone/NMP/EtOAc 5 r.t., 4 h 92/8 92
6 r.t., 4 h 92/8 99
7 2°C, 20 h 93/7 95
8 r.t., 24 h 85/15 99
9 r.t., 24 h 96//4 88
10 c 2 °C, 48 h - 2
11 2 °C, 15 h 85/15 60
0.1 mol%, 1 °C, 12 77/23 90 24 h, neat
13 2 °C, 24 h 79/21 25
14 2 °C, 24 h 83/17 95
15 2 °C, 48 h 77/23 20
c 16 2 °C, 48 h - 10 a 1a/1b/1c (1 mmol), 2-14 (0.5 mmol), C1 (1 mol %), DMC. b isolated yields. c the yields were calculated by 1H NMR.
Table 3−4 Reaction scope
90
Acryloyl amide only led to a 25% yield which revealed the importance of introducing an oxazolidin- 2-one or a pyridine group to achieve good conversion (Table 3−4, entry 13). On the contrary, keeping the di-carbonyl group acryloyl 2-pyrrolidinone could reach a 95% yield and higher endo/exo selectivity than using acryloyl-oxazolidin-2-one in entry 1 (Table 3−4, entry 14). Finally, we intended to perform a quick screening of the reactivity of other dienes. Cyclohexa-1,3-diene and 2,3-dimethyl 1,3-butadiene were selected to react with 3-acryloyl-1,4-oxazolidin-2-one. But the reactivity was very slow and with decreased stereoselectivity (Table 3−4, entries 15 and 16).
The caffeinium iron catalytic system was efficient in catalyzing the Diels-Alder reaction of the substrates above. However, some substrates were problematic using this catalytic system (Table 3−5). For example, unlike cinnmaldehyde, acrolein and crotonaldehyde led only to moderate to good yields, but the products could not be characterized because of polymerization (Table 3−5, entries 1 and 2). Adduct of cyclopentadiene with methyl propiolate was also polymerized (Table 3−5, entry 3). A methyl group at α position to the carbonyl, 2-oxazolidinone, 3-(2-methyl-1-oxo-2-propen-1-yl)-, was unreactive toward cyclopentadiene and cyclohexadiene in the reaction (Table 3−5, entries 4 and 5). Cinnamoyl oxazolidinone was not only unreactive to cyclopentadiene in solvents: DCM, acetone, NMP, and EtOAc, but also was not appropriate to react with cyclohexadiene and 2,3-dimethyl-1,3- butadiene (Table 3−5, entries 6−8). Although a high yield was obtained for 3-crotonoyloxazolidinone (Table 3−4, entry 2), the substrate was unreactive toward cyclohexadiene and 2,3-dimethyl-1,3- butadiene (Table 3−5, entries 9 and 10).
In conclusion, a green catalytic system using caffeine-derived imidazolium-Fe(OTf)2 complex in DMC for the Diels-Alder reaction was demonstrated to be an efficient Lewis acid catalyst for a large scope of substrates. The caffeine derived xanthinium salt as a solid provides a new way of catalyst- immobilization in comparing to ionic liquid. The xanthinium salt trapped the iron salt Fe(OTf)2 in the solid catalyst and during the work-up process, and in such a way the catalyst was recycled for several runs. Several green solvents were examined There was a limitation on the scope, but promising progress has been obtained. It is the first time that neat conditions have been used with methyl acrylate leading to 90% with 77:23 endo/exo selectivity. Thus, it was highly interesting to test the catalytic system in other reactions, such as carbonyl-ene reaction.
91
Entry Substrates Conditions endo/exo Yield (%)
1 −20 °C, 48 h 75/25 68
2 2 °C, 15 h 80/20 78
0.1 mol%, 2 °C, 3 79/21 26 24 h, neat
4 2 °C, 36 h − trace
5 r.t., 24 h + 40 °C 24 h − −
rt., 48 h, 6 − − DMC/Acetone/NMP/EtOAc
7 2 °C, 48 h − −
8 2 °C, 48 h − −
9 2 °C, 3 h+ rt., 12 h − 5
10 2 °C, 3 h; rt., 12 h − trace
Table 3−5 Less reactive substrates in the reaction scope
3.2.6 Application of the catalyst in other reactions
Due to the simplicity and recyclability of the caffeine derived imidazolium-Fe(OTf)2 catalyst, the catalyst was tested in the carbonyl-ene reaction (Scheme 3−5, Table 3−6). Because caffeine-derived imidazolium bismuth catalyst was very efficient in catalyzing the Diels-Alder reaction in CH2Cl2,38 the iron catalyst and bismuth catalyst were compared for promoting carbonyl-ene reaction (Table 3−6, entries 1 and 2). Good yields were obtained using both catalysts, but with FeII the yield was slightly higher than bismuth in two repetitions. However, it needs more repetitions to confirm the difference.
92
Scheme 3−7 Caffenium iron catalyst catalyzed carbonyl-ene reaction
It was more desirable in applying the newly developed catalytic system of the Diels-Alder reaction of the caffeine derived imidazolium-Fe(OTf)2 catalyst to carbonyl-ene reaction by using DMC as a green solvent. The yield (64%) was lower than using CH2Cl2 in the same conditions (Table 3−6, entry 3) and was decreased to 52% in a short reaction time (48 hours) (Table 3−6, entry 4).
Catalyst Entry 1/2 Solvent Time (h) Yield (%) (mol %)
1 1 : 1.5 1, 1 CH2Cl2 72 80
2 1 : 1.5 Bismuth catalyst, 1 CH2Cl2 72 77
3 1 : 1.5 1, 1 DMC 72 64
4 1 : 1.5 1, 1 DMC 48 52
Table 3−6 Caffenium iron catalyst catalyzed carbonyl-ene reaction
Acetylation of benzyl alcohol under solvent-free conditions was promoted using FeCl3 by Mihara (Scheme 3−8).44 The reaction benefits from low catalyst loading, short reaction time, and high yield, but the iron salt was not recyclable from the workup step. However, when the caffeine derived-
Fe(OTf)2 catalyst was used, better results were achieved with more than 99% yield at 0.1 mol% catalyst loading and 95% by using 0.01 mol% catalyst loading (Scheme 3−9). Although 4 hours reaction time was necessary, the catalyst was recyclable. Hence, it is highly attractive in developing this catalytic system with more substrates in acetylation.
93
Scheme 3−8 FeCl3 promoted acetylation of benzyl alcohol acetic anhydride
Scheme 3−9 Application of caffeine derived imidazolium-Fe(OTf)2 catalyst in acetylation of benzyl alcohol and acetic anhydride
94
3.3. References
[1] Furuta, K.; Shimizu, S.; Miwa, Y.; Yamamoto, H. J. Org. Chem. 1989, 54, 1481.
[2] Desimoni, G.; Faita, G.; Mella, M.; Piccinini, F.; Toscanini, M. Eur. J. Org. Chem. 2007, 1529.
[3] Evans, A. D.; Chapman, K. T.; Bisaha, J. J. Am. Chem. Soc. 1988, 110, 1238.
[4] (a) Narasaka, K.; Iwasawa, N.; Inoue, M.; Yamada, T.; Nakashima, M.; Sugimori, J. J. Am. Chem. Soc. 1989, 111, 5340. (b) Corey, E. J.; Imai, N.; Zhang, H-Y. J. Am. Chem. Soc. 1991, 113, 729.
[5] Ichiyanagi, T.; Shimizu, M.; Fujisawa, T. J. Org. Chem. 1997, 62, 7937.
[6] (a) Barroso, S.; Blay, G.; Pedro, J. R. Org. Lett. 2007, 9, 1983. (b) Rechavi, D.; Lemaire, M. J. Mol. Catal. A: Chem. 2002, 182, 239.
[7] Sarma, D., Appl. Catal. A: 2007, 335, 1.
[8] Bull, S. D.; Davidson, M. G.; Johnson, A. L.; Mahon, M. F.; Robinson, D. E. J. E. Chem. Asian J. 2010, 5, 612.
[9] Suga, H.; Kakehi, A.; Mitsuda, M. Bull. Chem. Soc. Jpn. 2004, 77, 561.
[10] Hiroi, K.; Watanabe, K. Tetrahedron: Asymmetry 2002, 13, 1841.
[11] Sibi, M. P.; Manyem, S.; Palencia, H. J. Am. Chem. Soc. 2006, 128, 13660.
[12] Aikawa, K.; Irie, R.; Katsuki, T. Tetrahedron 2001, 57, 845.
[13] Chollet, G.; Rodriguez, F.; Schulz, E. Org. Lett. 2006, 8, 539.
[14] (a) Doherty, S.; Goodrich, P.; Hardacre, C.; Knight J. G.; Nguyen, M. T,; Pâ rvulescu. V. I,; Paun, C. Adv. Synth. Catal. 2007, 349, 951. (b) Yeom, C-E.; Kim, W. H.; Shin, J. Y.; Kim, M. B. Tetrahedron Lett. 2007, 48, 9053. (c) Meracz, I.; Oh, T. Tetrahedron Lett. 2003, 44, 6465.
[15] (a) Baudequin, C.; Baudoux, J.; Levillain, J.; Cahard, D.; Gaumont, A.-C.; Plaquevent, J,-C. Tetrahedron: Asymmetry 2003, 14, 3081. (b) Chiappe, C.; Malvaldi, M. Chem. Phys., 2010, 12, 11191. (c) Zazybin, A.; Rafikova, K.; Yu, V.; Zolotareva, D.; Dembitsky, M. V.; Sasaki, T. Russ. Chem. Rev. 2017, 86, 1254. (d) Bica, K.; Gaertner, P. Org. Lett. 2006, 8, 733.
95
[16] (a) Stark, A. and Seddon, K. R. 2007. Ionic Liquids. Kirk-Othmer Encyclopedia of Chemical Technology. (b) Meindersma, W. W., Onink, F. A. F. and de Haan, A. B. 2010. Green Separation Processes with Ionic Liquids. Handbook of Green Chemistry. 6:6:137–190.
[17] Capello, C.; Fischer, U.; Hungerbü hler, K. Green Chem. 2007, 9, 927.
[18] (a) Aricò , F.; Tundo, P. Beilstein J. Org. Chem. 2016, 12, 2256. (b) Aricò , F.; Tundo, P. Russian Chem. Rev. 2010, 79, 479.
[19] Bauer, I.; Knö lker, H-J. Chem. Rev. 2015, 115, 3170.
[20] Khiar, N.; Ferná ndez, I.; Alcudia, F. Tetrahedron Lett. 1993, 34, 123.
[21] Matsukawa, S.; Sugama, H.; Imamoto, T. Tetrahedron Lett. 2000, 41, 6461.
[22] Corey, E. J.; Ishihara, K. Tetrahedron Lett. 1992, 33, 6807.
[23] (a) Kanemasa, S.; Oderaotoshi, Y.; Yamamoto, H.; Tanaka, J.; Wada, E. J. Org. Chem. 1997, 62, 6454. (b) Kanemasa, S.; Oderaotoshi, Y.; Sakaguchi, S.; Yamamoto, H.; Tanaka, J.; Wada, E.; Curran, D. P. J. Am. Chem. Soc. 1998, 120, 3074.
[24] (a) Kü ndig, E. P.; Bourdin, B.; Bernardinelli, G. Angew. Chem., Int. Ed. Engl. 1994, 33, 1856. (b) Bruin, M. E.; Kü ndig, E. P. Chem. Commun. 1998, 23, 2635.
[25] Howard, J.; Ilyashenko, G.; Sparkes, H.; Whiting. A. Dalton Trans. 2007, 21, 2108.
[26] Fujiwara, K.; Kurahashi, T.; Matsubara, S. J. Am. Chem. Soc. 2012, 134, 5512.
[27] Herndon, J. W. J. Organomet. Chem. 1986, 51, 2855.
[28] Nakanishi, S.; Kumeta, K.; Sawai, Y.; Takata, T. J. Org. Chem. 1996, 515, 99.
[29] Wang, B.; Mu, B.; Chen, D.; Xu, S.; Z, X. Organometallics 2004, 23, 6225.
[30] Gü ltekin, Z. Clay Miner. 2004, 39, 345.
[31] Chavan, S. P.; Sharma, A. K. Synlett 2001, 667.
[32] Wei, K.; Wang, S.; Liu, Z.; Du. Y.; Shi. X.; Qi, T.; Ji. S. Tetrahedron Lett. 2013, 54, 2264.
[33] Donatoni, M. C.; Ando, R. A.; Brocksom, T. J. Tetrahedron 2014, 70, 3231.
[34] Basavegowda, N.; Mishra, K.; Lee, Y. R.; Job, Y-G. Bull. Korean Chem. Soc. 2016, 37, 142.
[35] Eberson, L.; Olofsson, B.; Svensson, J. O. Acta Chem. Scand. 1992, 46, 1005.
96
[36] Gorman, D. B.; Tomlinson, I. A. Chem. Commun. 1998, 25, 81.
[37] (b) Kumazawaa, T.; Senob, H.; Leea, X.P.; Ishiib, A.; Suzukib, K.W.; Satoa, K.; Suzukib, O. Anal. Chim. Acta. 1999, 387, 53. (c) Khursheed, T.; Ansari, M. Y. K; Shahab, D. Biol. Med. 2009, 1, 56. (d) Waldvogel, S. R. Angew. Chem., Int. Ed. 2003, 42, 604.
[38] Pinto, R. M. A.; Salvador, J. A. R.; Le Roux, C. Catal. Comm. 2007, 9, 465.
[39] Mayer, A. C.; Salit, A.-F.; Bolm, C. Chem. Comm. 2008, 45, 5975.
[40] Luo, F. T.; Lo, H. K. J. Organomet. Chem. 2011, 696, 1262.
[41] Cabrero-Antonino, J. R.; Leyva-Perez, A.; Corma, A. Chem. Eur. J. 2012, 18, 11107.
[42] Barroso, S.; Blay, G.; Pedro, J. R. Org. Lett. 2007, 9, 1983.
[43] Erfurt, K.; Wandzik, I.; Walczak, K.; Matuszek. K.; Chrobok, A. Green Chem. 2014, 16, 3508.
[44] Mihara, M.; Nakai, T.; Iwai, T.; Ito, T.; Ohno, T.; Mizuno, T. Synlett 2010, 253.
97
Chapter four
Development of NHC-diol ligands for the enantioselective carbonyl-ene reaction
4.1 Introduction and background
Based on the fact that functionalization of NHC precursors using different functional groups, such as a phosphine, carbonyl, pyridyl, other NHC etc. is a common way of tuning the electronic property of NHC transition metal complexes, Arnold1 targeted the first NHC-chelating ligand incorporating a hard anionic alkoxide function group.1,2 In the first case, simplicity as the epoxide-opening reaction and
SN2 substitution of Cl were simply applied to the first step, and the zwitterionic di-imidazolium salt was treated with Ag2O to afford the di-NHC-Ag alkoxide complex. Transmetallation from bis-NHC- Ag alkoxide with CuI afforded the bis-NHC-Cu alkoxide complex (Scheme 4−1). In the second example, NHC-Li alkoxide was used for transmetallation with CuCl2 to afford a mono-NHC-Cu alkoxide complex, while introducing a chiral center at the position to N in the NHC-Cu alkoxide complex (Scheme 4−2). Arnold’s two examples demonstrate the feasibility of using an NHC-alcohol as a chelating ligand with a transition metal. Furthermore, the mono-NHC-Cu alkoxide was studied in the conjugate addition of diethyl zinc to cyclohexanone with a very good enantioselectivity for tackling catalytic applications (Scheme 4−3). The possibility that alkoxide acted as a catalyst in the reaction was put forward.1, 2
98
Scheme 4−1 Arnold’s NHC-Cu-alkoxide complex
Scheme 4−2 Formation of mono-NHC-Cu by transmetallation of NHC-Li alkoxide with CuCl2
Scheme 4−3 Enantioselective conjugate addition of diethyl zinc to cyclohexanone
The flexibility of using a hydroxyl group was demonstrated by incorporating it with NHC in an axial chiral binaphthyl group in a ruthenium complex for the catalytic ring-opening metathesis.3 In 2002, based on the hypothesis that a bidentate ligand would induce chirality more efficiently, and encouraged by the activities of a Ru catalyst bearing a phenolic Schiff base, Hoveyda developed a ruthenium catalyst for the metathesis reaction (Scheme 4−4 and Scheme 4−5).4 Although there was some concern about the reactivity, further proved in their experiments, of the substitution of one chloride by a phenoxide, the Ru-NHC complex was air stable and could efficiently promote the
99
metathesis reaction. However, the elevated temperature and the more extended reaction time required in the reaction were explained by the reduced reactivity due to the replacement of a Cl by a less electronegative phenoxide and the bulky binaphthyl axial group. Due to the Ru-O bond from aryloxide, the Ru-NHC complex was tolerant to undistilled solvents and substrates which were easily polymerized with Mo catalysts and could be purified by silica gel chromatography without loss of enantiomeric purity.5
Scheme 4−4 Synthesis of a bidentate alkoxyl-NHC-Ru complex in Hoveyda’s group
Scheme 4−5 Alkoxyl-NHC-Ru complex catalyzed metathesis
Other few ligands were examined with copper salts to promote asymmetric alkylation (Figure 4−1). The results showed that the replacement of the mesityl group by 2,6-di-isopropyl group or an adamantyl group prohibited the reactivity of the addition of Et2Zn to allylic phosphate derivative. To enhance the reactivity, a silver-NHC dimer was obtained and used to synthesize the relative (analogous) Cu-NHC complex through transmetallation. A relative large scope of substrates was disclosed with the catalyst of silver-NHC and copper salt (Scheme 4−6).5
100
Figure 4−1 Modified alkoxyl-NHC by 2,6-di-isopropyl group or adamantyl group
Scheme 4−6 Alkoxyl-Cu-NHC generated through transmetalation with relative alkoxyl-Ag- NHC complex in catalyzing allylic alkylation
Another way of synthesizing the NHC-naphthoxy ligand was performed by Crabtree.6 Newly developed rhodium and iridium NHC complexes were applied to the hydrosilylation reaction while achieving low enantiomeric excess by rhodium and moderate ee using iridium (Scheme 4−7).
Scheme 4−7 Alkoxyl-Rh/Ir-NHC complexes in catalyzing hydrosilylation reaction
In 2004, Mauduit designed and synthesized a new imidazolium salt incorporating a hydroxyl group from L-valine as a starting material (Scheme 4−8). Although the imidazolynium salt was not investigated as a NHC precursor, it highlighted the structure-interaction relationship that the hydroxyl lateral chain and bulky aromatic N-substituent played in obtaining high diasteromeric interaction with the racemic anion. It provided a way of getting evidence of the chiral interaction that a NHC ligand would have with racemic substrates in an organic reaction.7
101
Scheme 4−8 Ionic liquid incorporating hydroxyl group by using chiral L-valine
In Mauduit’s subsequent work, a similar synthetic route was used for developing other bidentate ligands of NHC-hydroxyl imidazolynium salt from natural amino alcohols. The developed salts were evaluated as efficient NHC-alkoxide bidentate ligands with Cu(OTf)2 in an enantioselective conjugate addition. One salt derived from L-Leucinol was found to be the best among the list and selected for optimization. In suspecting that the hydroxide of the NHC ligand could play a role in the enantiocontrol process of the conjugate addition, the hydroxyl functional group was protected by a tert-butyldimethylsilyl group (Scheme 4−9). Consequently, the resulting enantiomeric excess was decreased to 62% compared to NHC with the free hydroxyl group. The requirement of hydroxyl functional group for high enantioselectivity was again proved to be crucial.8
Scheme 4−9 L-Leucinol derived NHC-alkoxyl bidentate ligands with Cu(OTf)2 in enantioselective conjugate addition
Furthermore, Mauduit disclosed the conjugate addition catalyzed by copper with the new NHC family ligand, chiral alkoxy-NHC ligands in a comprehensive way and disclosed various analogs based on different amino alcohols as bidentate ligands.9 The hydroxyl group of chiral alkoxy-NHC ligands was emphasized as a crucial enantioselectivity-control factor of the conjugate addition. The alkoxy-NHC ligands were proposed to serve as LX bidentate ligands.10 Based on the investigation above, new
102
synthetic routes were designed for NHC-di-hydroxyl ligands (NHC-diol ligands) because of the interest in imidazolium salts as ionic liquids in Wilhelm’s group (Figure 4−2, Scheme 4−10).11
Scheme 4−10 Synthesis of NHC-diol ligands by Wilhelm
Figure 4−2 Synthesis of NHC-diol ligands developed by Wilhelm
Wilhelm reported two synthetic routes for new imidazolium carbene ligands incorporating two hydroxyl groups. The first route was designed with two steps including substitution of 1,2- dibromoethane by different amino alcohols and cyclization using triethyl orthoformate in the presence of an ammonium salt (Scheme 4−10). Hence, various imidazolium salts with chiral centers in the α and/or β position to N were prepared in high yields. A second route was started from the epoxide opening reaction of 1,2-(R,R)-diphenyl-1,2-ethylenediamine and chiral or achiral epoxides giving rise to β-amino alcohols, and followed by cyclization to form two imidazolium salts, respectively (Scheme 4−11).11 A bis-amino-alcohol, prepared from (1R,2R)-trans-diaminocyclohexane and cyclohexene oxide by Marson, was used for synthesizing the corresponding imidazolium salt (Figure 4−2). The geometric structure of the N-heterocycle of NHC-diol ligand having a cyclohexyl backbone and two other cyclohexyl rings was tested in the addition reaction of diethyl zinc to benzaldehyde to reach the highest enantiomeric excess of 66%. But when the conditions were applied to the addition of diethyl zinc to 1-naphthaldehyde, the enantiomeric excess decreased.12
103
Scheme 4−11 Synthesis of NHC-diol ligands by epoxide-opening reaction
Catalytic tests were studied with the diethylzinc addition to 1-naphthaldehyde in order to evaluate the efficiency of the NHC-diol ligands. Various metals were selected, and copper was chosen as the best due to the enantioselectivity and yields achieved. Promisingly, two iron salts were tested with the NHC-diol ligand, and moderate enantioselectivities were obtained (Scheme 3−12). These results encouraged the pursuing of the synthesis of the dihydroxy-NHC ligands and the development of iron catalyst with such ligands.13
Scheme 4−12 NHC-diol-copper and -iron catalyzed asymmetric addition of an enantioselective conjugate addition
Glorius reported an heterogeneous catalyst comprising of NHC-diol ligand as chiral modifiers and
Fe3O4/Pd nanoparticle (NP), and its application in the asymmetric -arylation of ketones with aryl halides.14 The process included two steps: the formation of Pd NPs on magnetite surface and the modification of the bimetallic NPs by chiral NHC-diols. Moderate to good enantiomeric excesses were achieved with various aryl halides, and the hydroxyls on the ligand were claimed to be crucial to the enantioenrichment of the α-arylated product. The heterogeneous application was claimed to represent the first efficient use of NHCs as chiral modifiers of NP catalysts (Scheme 4−13).14
104
Scheme 4−13 NHC-diol ligand modified Fe3O4/Pd nanoparticle in catalyzing α-arylation of ketones
Scheme 4−14 Fe-Bolm’s ligand catalyzed Mukaiyama aldol reaction
More importantly, a catalytic system consisting of Fe(ClO4)2∙6H2O with Bolm’s ligand was proved to be very efficient in catalyzing the Mukaiyama aldol reaction with excellent results in our group (Scheme 4−14).15 In the studies of catalyst structure, it appeared that two hydroxyl groups were coordinated to the heptacoordinated chiral FeII center. The chiral centers bearing tert-butyls and hydroxyls were considered to be crucial in the asymmetric induction. Thus, a chiral carbon center bearing tert-butyl and hydroxyl groups was of high interest in designing a chiral NHC-diol ligand.15
Apart from the NHC-diol type of ligands reported in the literature, another type of tridentate NHC ligand incorporating a pyridine and di-imidazolium salts appeared to be highly interesting. Danopoulos disclosed the design and synthesis of the pyridine-bisNHCs and its complexation with iron salts.16 More interestingly, the application of such pyridine-bisNHCs-iron complex was very efficient in catalyzing the Kumada-type cross-coupling reaction (Scheme 4−15).17 The structural and
105
reactivity studies of the pyridine-bisNHCs-iron complexes triggered our interest in developing such type of NHC ligand. Furthermore, relevant research on the complexation of such type of ligand with other metals such as silver18 and palladium19 emphasized the undeveloped and promising potential of pyridine-bisNHCs type ligands (Figure 4−3).
Figure 4−3 Pyridine bisNHC-diol type of ligands developed in literature
Scheme 4−15 Pyridine-bisNHCs-iron complexes catalyzed Kumada-type cross-coupling
4.2. Synthesis of carbene-diols ligand family
4.2.1 Preparation of (R)-tert-butyloxirane
Base on the pioneered work shown above, we first designed synthetic routes for a few alkoxy-NHC ligands utilizing an epoxide opening reaction of enantiomeric enriched (R)-tert-butyloxirane using various amines. It started from the oxidation of 3,3-dimethyl-1-butene by m-CPBA to afford a racemic epoxide, followed by the resolution of using Jacobsen’s method.20 A cobalt-salen compound was used as the catalyst in the resolution where only the (S)-tert-butyloxirane enantiomer was attacked by H2O. As a result, an enantiomerically enriched (R)-tert-butyloxirane was obtained and its ee was determined by chiral HPLC analysis of the 2-benzothiazolesulfide derivative. A high enantiomeric excess (>99%) was obtained, and the oxirane was used for the following steps (Scheme 4−16).
106
Scheme 4−16 Preparation of (R)-tert-butyloxirane by oxidation and resolution
4.2.2 Preparation of ligands 6−11
Three amines were selected for the epoxide opening reaction with the enantiomeric enriched (R)- tert-butyloxirane. Enantiopure 1,2-diphenyl-ethylene trans-diamine was chosen as a ligand or an auxiliary to a ligand in many cases (Scheme 4−17), such as Noyori’s catalyst21, ligands to transition metals22, formation of a Schiff base23, an auxiliary in NHC ligands,24 etc. (Figure 4−4). The performance of its application and its nature including the C2 symmetrical geometry and having a diamine functional group were of high interest in designing a chira l NHC-diol ligand. Thus, (S, S)- 1,2-diphenyl-ethylene trans-diamine was selected as a nucleophile for the epoxide opening reaction to obtain the diamino alcohol in 72% yield. In the next step, a NHC-diol precursor, the imidazolium salt, was prepared by reacting the diamino alcohols with triethyl orthoformate and ammonium
107
tetrafluoroborate in high yield according to a literature method.25 Hence, a new C2 symmetric chiral NHC-diol Ligand (6b) was prepared in high yield and fully characterized.
Figure 4−4 Application of 1,2-diphenyl-ethylene trans-diamine
Scheme 4−17 Synthesis of NHC-diol ligand from (S,S)-1,2-diphenyl-ethylene trans-diamine
Inspired by the synthesis of NHC-diol ligand precursor derived from cyclohexane 1,2-trans-diamine in the group of Wilhelm, a new ligand made from of trans-1,2-diamininocyclohexane and tert- butyloxirane was designed and synthesized. The geometry of trans-1,2-diaminocyclohexane was expected to give high chiral induction from the tridentate chelating NHC-diol ligand. Firstly, the resolution of racemic trans-1,2-diamininocyclohexane was performed by reacting with L-(+)-tartaric acid to form a diastereomer salt in which it contained the enantiomeric enriched (R,R)-1,2- diaminocyclohexane.26 Then, the diamino alcohol was reacted with triethyl orthoformate and ammonium tetrafluoroborate to form the imidazolium salt. In this protocol, ligand 7a and 7b were prepared and characterized (Scheme 4−18).
108
Scheme 4−18 Synthesis of NHC-diol ligand from enantiomeric enriched (R,R)-1,2- diaminocyclohexane
Since imidazole ring has already the basic structure of an NHC carbene, it was profitable to use imidazole as a starting material. The synthetic route started from the epoxide opening reaction of (R)-tert-butyloxirane by imidazole and the desired mono-epoxide opening product, 1-R-tert-butyl-2- imidazoly-ethanol 8a, was achieved in a 72% yield. Interestingly, a by-product, a zwitterion as ligand 8b, arised from a second epoxide opening process in 10-15% yields. This tridentate zwitterion ligand could be further developed for being a major product and acting as a promising NHC-diol ligand due to its C2 symmetrical structure (Scheme 4−19).
Scheme 4−19 Epoxide opening reaction of (R)-tert-butyloxirane by imidazole
The mono epoxide opening product was utilized for the substitution with three different bromo- alkylating reagents (para-tert-butylbenzyl bromide, 1,2-dibromoethane and 2,6-bomomethyl pyridine) in order to obtain a bidentate alkoxy-NHC ligand, a bis-NHC-diol ligand and a pyridine-bis- NHC ligand respectively (Scheme 4−20, Ligand 4-7) in almost all quantitative yields. All salts were fully characterized by 1H NMR, 13C NMR and MS analysis.
109
Scheme 4−20 Synthesis of ligands 4-7
Ligand 11a was selected to generate complexes with metals due to its property to act as a pentadentate ligand. Metal salts Ag2O, Pd(OAc)2, Fe(OAc)2 and Fe(OTf)2 were selected to coordinate with ligand 11a. Ag2O was successfully complexed with different NHCs in many examples in the literature.27 It acted as a base to deprotonate the imidazolium salts and as a transition metal to coordinate with the carbene at the same time. By mixing Ag2O with ligand 11a in THF solution while stirring 1 day at room temperature, a silver-carbene complex was formed in a quantitative yield (Scheme 4−21). The characteristic C−H bond of imidazolium salt disappeared after complexation. But the complexation with palladium acetate or iron salts did not happen under the same conditions. Ligand 6−11 (except Ligand 8b) along with a few selected ligands in literature were tested with iron salts for catalyzing the carbonyl-ene reaction of ethyl 3,3,3-trifluoropyruvate and α-methylstyrene.
110
Scheme 4−21 Complexation of ligand 6 with Ag2O and its proposed structure
4.3 Catalytic tests in the enantioselective carbonyl-ene reaction
4.3.1 Catalytic test of carbene-diols ligands with iron in carbonyl-ene reaction
After the synthesis of NHC-diol ligands (Figure 4−5), it was decided to test their catalytic activities with iron salts in the carbonyl-ene (Scheme 4−22). Due to the performance of Fe(OTf)2 in caffeine- derived NHC-iron catalyzed carbonyl-ene reaction, anhydrous Fe(OTf)2 was used as the iron source in the reaction. Carbene alkoxide iron coordinated complexes were designed as the catalytic active species. Hence KOtBu as a base was used to deprotonate the carbene precursor and the alcohols in each ligand in order to generate carbene-alkoxide chelating ligands toward iron. For ligand 6 and 7, three equivalents of base were used whereas two equivalents base for ligand 9 and four equivalents base for ligand 10−11b.
Figure 4−5 NHC-diol ligands 6b-11b
111
Through a complexation process between iron and alkoxide-NHC, the alkoxide-NHC-FeII was expected to be generated-in-situ. The test results are showed below (Table 4−1).
Scheme 4−22 NHC-diol Fe(OTf)2 catalyzed carbonyl-ene reaction
Entry 1/2 Ligand Base x Time (h) Yield (%) ee (%)
1 2 : 1 6b 18 48 25 1
2 2 : 1 6b 18 72 41 0
3 3 : 1 6b 18 72 65 0
4 2 : 1 7b 18 48 55 0
5 2 : 1 9 12 48 54(65) * 0
6 1 : 1.2 9 12 48 54
7 2 : 1 10 25 48 57 0
8 2 : 1 11a 25 48 55 0
9 2 : 1 11b 25 48 75 0
* 10 mol% FeCl2 + 10 mol% ligand 4 + 25 mol% base were used in the entry.
Table 4−1 Application of NHC-diol ligand with iron in testing carbonyl-ene reaction
Generally, no enantioselectivity was achieved using the various conditions tested (Table 4−1), and low to moderate yields were obtained except in entry 9. For ligand 1 in the condition of entry 1, the yield was as low as 25% after 48 hours reaction time. Prolonging reaction time and using α- methylstyrene in excess were effective in boosting the yield (Table 4−1, entries 2 and 3). Ligand 7 led to a moderate yield which was higher than ligand 6 under the same conditions (Table 4−1, entries 1 and 4). The difference arising from same basic chelating NHC-di-alkoxide structure but giving different yields was possibly explained due to the moisture sensitivity nature of carbene which generated-in-situ, which led to the loss of carbene formation and actual NHC-Fe complex. Thus,
112
further experiments for using ligand 6 in more strict anhydrous conditions was essential to verify the reason for the low yield. Ligand 9 under the same conditions of ligand 7 led to same yield 54% and upon increasing the catalyst loading, the yield was not dramatically enhanced (Table 4−1, entries 5). Changing the ratio of α-methylstyrene to trifluoropyruvate from 2:1 to 1:1.2 had no impact on yield (Table 4−1, entries 5 and 6). Ligand 10 and 11a led to around 55% yields. The same results were obtained with ligand 7 and 9 (Table 4−1, entries 7 and 8). The non-coordinating anion PF6¯ and the pyridine of ligand 11b, when compared with other ligands, were possibly the reason of reaching a higher yield 75% than other entries (Table 4−1, entry 9). Since NHC-alkoxide-iron complexes were not successfully synthesized, it would take more research efforts to investigate the functionalization of alkoxide and the formation of bidentate, tridentate, tetradentate, and pentadentate iron complexes by ligands 6-11b. Furthermore, the ionic alkoxide had an impact on the formation of alkoxide-iron bond, which may cause the failure of chiral induction from the chiral center of carbon bearing hydroxyl and tert-butyl groups. It still needs more investigation for developing the newly designed NHC-diol ligands with different iron salts and other transition metals, and further optimization regarding the aspects of solvent, temperature, and complexation methods, etc. for the carbonyl-ene reaction of ethyl trifluoropyruvate.
4.3.2 Catalytic tests of a few C2 symmetric ligands in the iron in carbonyl-ene reaction
Apart from the new NHC-diol ligands synthesized above, a few other C2-symmetric ligands (imine as a functional group) prebiously disclosed to be effective with transition metals in homogeneous catalysis, were chosen to test the catalytic activity with iron in carbonyl-ene reaction. Both bis- oxazoline (Box) and pyridine-2,6-bis-oxazoline (Py-Box) have been applied as chiral ligands with various metals including iron in organic synthesis.28 The excellent catalytic activity of Box ligand with iron salt (Bis(oxazoline)-FeIII complex) was demonstrated by Corey in the enantioselective Diels- Alder reaction in 1990.29 In the presence of 10 mol% of the catalyst loading, the reaction of 3-acryloyl- 1,3- oxazolidin-2-one and cyclopentadiene afforded an excellent yield and with a high endo selectivity (endo/exo = 96 : 4) and a very good enantioselectivity (ee = 82%) for the endo product. Box-Cu complex was developed as a very efficient catalyst for promoting the carbonyl-ene reaction of glyoxylate ester and a wide range of olefins such as monosubstituted, disubstituted, and trisubstituted alkenes.30 Various iron-Py-Box complexes were developed. The catalytic activity between iron and carbonyl groups was also demonstrated in the application of iron-pybox complexes
113
in various reactions such as asymmetric Mukaiyama aldol, Diels-Alder, hydrosilylation, aziridination, cyclization, etc. in which all the substrates in the reactions involved the participation of carbonyl groups.28b
Scheme 4−23 Iron-C2 symmetric ligands catalyzed carbonyl-ene reaction
Following our studies on iron catalysis such as iron-Bolm’s ligand catalyzed enantioselective Mukaiyama aldol15, meso-epoxide opening31 and iron-Py-Box complexes catalyzed aromatic sulfoxide oxidation reactions32, it was of great interest to test these C2-symmetric ligands with iron in carbonyl-ene reaction of ethyl trifluoropyruvate. Base on the results of iron meditated carbonyl-ene reaction in chapter 1, irons salts, such as iron triflate Fe(OTf)2, Fe(OCl4)2∙6H2O which had been applied with Bolm’s ligand, Fe(SbF6)2 (generated from FeCl2+2AgSbF6) were selected with these ligands (Scheme 4−23).
Generally, FeII worked efficiently with these imino type ligands to promote the reaction in moderate to high yields, but some enantioselectivity was observed only with Box ligand. Good to high yields were achieved by using Fe(OTf)2 and Fe(OCl4)2∙6H2O with pybox ligand bearing a phenyl group (Table 4−2, entries 1 and 2). Using slight excess of ethyl trifluoropyruvate with a prolonged reaction time led a decrease of yields. FeII with non-coordinated anion ClO4¯ led to higher yields (Table 4−2, entry 2). Compared with Py-Box-Ph ligand, pybox ligand bearing an iso-propyl group afforded lower yields from 56 to 79 with four iron salts (Table 4−2, entries 3-6). FeCl2 led to a higher yield than
Fe(OTf)2, Fe(OCl4)2∙6H2O, and Fe(SbF6)2 possibly because the electronegative chloride anion balanced the charge on iron center and changed the Lewis acidity of iron-Py-Box catalyst. However, neither coordinating anion nor coordinating anions led to chiral induction, hence, there was no enantioselective excess by using the four iron salts. With using Fe(SbF6)2 with Bolm’s ligand, two
114
Entry a FeX2 Ligand Yield (%)b ee (%)
1 Fe(OTf)2 Py-Box-Ph 73(65) 0
2 Fe(ClO4)2∙6H2O Py-Box-Ph 89(85) 0
3 Fe(OTf)2 Py-Box-iPr 56 0
4 Fe(ClO4)2∙6H2O Py-Box-iPr 68 0
5 FeCl2 Py-Box-iPr 79 0
FeCl + 10 mol% 6 2 Py-Box-iPr 73 0 AgSbF6 FeCl + 10 mol% 7 2 Bolm’s ligand 73 0 AgSbF6 FeCl + 10 mol% 8 2 Box ligand 98 33 AgSbF6
9 Fe(OTf)2 Box ligand 80 2
FeCl + 10 mol% 10 2 Box ligand 89 c 35 AgSbF6 a 1/2 = 1:1.2 for entries 1 and 2, 1/2 = 2:1; b Isolated yields; c Reaction temperature at -4 ℃.
Table 4−2 Iron-C2 symmetric ligands catalyzed carbonyl-ene reaction pyridines while two hydroxyl groups were expected to coordinate with iron in a tetra-coordinated geometry iron complex with non-coordinating anion SbF6¯ . The yield was the same as Fe(SbF6)2 with Py-Box-iPr ligand (Table 4−2, entries 6 and 7). With using Box ligand with iron salts, both yield and enantioselectivity were improved (Table 4−2, entries 8-9). An excellent yield was achieved using
Fe(SbF6)2 with Box ligand with a 33% enantiomeric excess, but changing to Fe(OTf)2 led to decreased yields and loss of enantioselectivity. Aiming at further improving the enantioselectivity, a lower temperature at −4 °C was used and a slight increase of the enantiomeric excess was obtained with a decreased yield (Table 4−2, entry 10). Since box ligand with iron exhibited an excellent yield and a promising enantioselectivity, further optimization would be designed from the aspects of solvent effect, temperature, other iron salts, catalyst loading, etc. with the use of Box ligand. Although
115
high yield was afforded by Py-Box-Ph and Py-Box-iPr with various iron salts (Table 4−2, entries 2 and 5), when comparing Box ligand and Py-Box ligand, it was found that a relative wide expansion of two coordinating bis-oxazoline ligands resulted from the connecting part of pyridyl group instead of a “Me2C-” group and consequently the chiral centers in pybox ligand became too far from the catalytic active center iron to provide steric hindrance (Figure 4−6). Thus, these two aspects were proposed as the major reason for the loss of enantioselectivity of carbonyl-ene reaction with ethyl trifluoropyruvate.
Figure 4−6 Postulated coordination of Box-FeII and Py-Box-FeII
A new diamino-bis-oxazoline ligand described as “a porphyrin-inspired ligand” was designed with manganese for an asymmetric epoxidation of a wide range of olefins with high yields and up to more than 99% ee by Gao.33 The results demonstrated that the diamino-bis-oxazoline-Mg complex provided an excellent chiral induction. Firstly, since iron-oxazoline catalysis had been applied to a relatively wide range of organic transformations, it was of great interest to test this ligand with an iron salt in the carbonyl-ene reaction. Secondly, secondary diamines were used for preparing NHC precursors as a general method. Hence, a first route for synthesizing an NHC-bis-oxazoline ligand was designed, which included the synthesis of triphenyl diamine-bis-oxazoline ligand in the first step and a cyclization of the diamine to form the imidazolium ring. After the synthesis of the diamino- bisoxazoline ligand by following, triethyl orthoformate in the presence of ammonium tetrafluoroborate was used for the cyclization of the secondary diamine to form an imidazolium salt. However, the cyclization of the synthesis of NHC-diol did not proceed in temperatures between 100-120 °C or milder condition, and the compound decomposed at higher temperatures (Scheme 4−24). The failure of the cyclization possibly resulted from the steric hindrance of the two oxazolines in the ortho position of the phenyl groups. Furthermore, based on this reason, a second route to achieve the preparation of desired product was initiated from the cyclization after the first step of the synthesis (Buchwald-Hartwig reaction).
116
Scheme 4−24 Synthesis of “a porphyrin-inspired ligand” and a desired/expected NHC precursor base on the ligand
However, the successful development of the asymmetric epoxidation of olefins by the di-imino bisoxazoline-manganese complex inspired our interest in testing the ligand in carbonyl-ene and meso-epoxide opening reactions. Only one attempt was performed and a low yield was obtained in carbonyl-ene reaction of ethyl trifluoropyruvate together with no enantiomeric excess. A further experiment could be envisioned from the optimization of the iron salts, the solvents, the coordination of iron-diamino-box or iron-box-diamide type complexes. In the meso-epoxide opening reaction, iron salts Fe(OTf)2 and Fe(OCl4)2∙6H2O were tested. Low to moderate yields and low ee were obtained (Scheme 4−25).
117
Scheme 4−25 Catalytic tests in carbonyl-ene reaction and epoxide-opening reaction
118
4.4. References
[1] Arnold, P. L.; Scarisbrick, A. C.; Blake, A. J.; Wilson, C. Chem. Commun. 2001, 2340.
[2] Arnold, P. L.; Rodden, M.; Davis, K. M.; Scarisbrick, A. C.; Blake, A. J.; Wilson, C. Chem. Commun. 2004, 1612.
[3] Chang, S.; Jones, L.; Wang, C.; Henling, L. M.; Grubbs, R. H. Organometallics 1998, 17, 3460.
[4] VanVeldhuizen, J. J.; Garber, S. B.; Kingsbury, J. S.; Hoveyda, A. H. J. Am. Chem. Soc. 2002, 124, 4954.
[5] Larsen, A. O.; Leu, W.; Oberhuber, C. N.; Campbell, J. E.; Hoveyda. A. H. J. Am. Chem. Soc. 2002, 126, 11130.
[6] Chianese, A. R.; Crabtree, R. H. Organometallics 2005, 24, 4432.
[7] Clavier, H.; Boulanger, L.; Audic, N.; Toupet, L.; Mauduit, M.; Guillemin, J. -C. Chem. Commun. 2004, 1224.
[8] Clavier, H.; Coutable, L.; Guillemin, J. -C.; Mauduit, M. Tetrahedron: Asymmetry 2005, 16, 921.
[9] Clavier, H.; Coutable, L.; Toupet, L.; Guillemin, J. -C.; Mauduit, M. J. Organomet. Chem. 2005, 690, 5237.
[10] Martin, D.; Kehrli, S.; Augustin, M.; Clavier, H.; Mauduit. M.; Alexakis. A. J. Am. Chem. Soc. 2006, 128, 8416.
[11] Jurčík, V.; Gilani, M.; Wilhelm, R. Eur. J. Org. Chem. 2006, 5103.
[12] Cobb, A. J. A.; Marson, C. M. Tetrahedron 2005, 61, 1269.
[13] Gilani, M.; Wilhelm, R. Tetrahedron: Asymmetry 2008, 19, 2346.
[14] Ranganath, K. V. S.; Kloesges, J.; Schä fer, A. H.; Glorius, F. Angew. Chem., Int. Ed. 2010, 49, 7786.
[15] Ollevier, T.; Plancq, B. Chem. Commun. 2012, 48, 2289.
[16] Danopoulos, A. A.; Pugh, D.; Smith, H. Chem. Eur. J. 2009, 15, 5491.
119
[17] (a) Danopoulos, A. A.; Tsoureas, N.; Wright, J. A.; Light, M. E. Organometallics 2004, 23, 166. (b) Bedford, R. B.; Betham, M. D.; Bruce, W.; Danopoulos, A. A.; Frost, R. M.; Hird, M. J. Org. Chem. 2006, 71, 1104.
[18] Melaiye, A.; Simons, R. S.; Milsted, A.; Pingitore, F.; Wesdemiotis, C.; Tessier, C. A.; Youngs, W. J. J. Med. Chem. 2004, 47, 973.
[19] Loch, J. A.; Albercht, M.; Peris, E.; Mata, J.; Faller, J. W.; Crabtree, R. H. Organometallics 2002, 21, 700.
[20] Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.; Hansen, K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 1307-1017.
[21] Ikariya, T.; Hashiguchi, S.; Murata, K.; Noyori, R. Org. Syn. 2005, 82, 10.
[22] Ghosh, S. K.; Lewis, K. G.; Kumar, A.; Gladysz, J. A. Inorg. Chem. 2017, 56, 2304.
[23] Ramin, M.; Jutz, F.; Grunwaldt, J. -D.; Baiker, A. J. Mol. Catal. A., Chem. 2005, 242, 32.
[24] Wan, K. Y.; Lough, A. J.; Morris, R. H. Organometallics 2016, 35, 1604.
[25] Saba, S.; Brescia, A.; Kaloustain, M. K. Tetrahedron Lett. 1991, 32, 5031.
[26] Larrow, J. F.; Jacobsen, E. N. Org. Synth. 1998, 75, 1.
[27] (a) McGuinness, D. S.; Cavell, K. J.; Yates, B. F.; Skelton, B. W.; White, A. H. J. Am. Chem. Soc. 2001, 123, 8317. (b) Kascatan-Nebioglu, A.; Panzner, J. M.; Carrison, C. Jered.; Tessier, A. C.; Youngs, J. W.; Organometallics 2004, 23, 1982. (c) Hohnson, N. A.; Southerland, M. R.; Youngs, W. J. Molecules 2017, 22, 1263.
[28] (a) Sibi, M. P.; Stanley, L. M.; Nie, X.; Venkatraman, L.; Liu. M.; Jasperse, C. P. J. Am. Chem. Soc. 2007, 129, 395. (b) Ollevier, T. Catal. Sci. Technol. 2016, 6, 41.
[29] Corey, E. J.; Imai, N.; Zhang, H-Y. J. Am. Chem. Soc. 1991, 113, 729.
[30] Evans, D. A.; Burgey, C. S.; Paras, N. A.; Vojkovsky, T.; Tregay, S. W. J. Am. Chem. Soc. 1998, 120, 5821.
[31] Plancq, B., Ollevier, T. Chem. Commun. 2012, 48, 3806.
[32] Jalba, A.; Ré gnier, N.; Ollevier, T. Eur. J. Org. Chem. 2017, 1628.
[33] Dai, W.; Li, J.; Li, G.; Yang, H.; Wang, L.; Gao, S. Org. Lett. 2013, 15, 4138.
120
Chapter five
Conclusion and Perspectives
5.1 Iron-catalyzed carbonyl-ene reaction
An iron catalytic system was developed using anhydrous Fe(BF4)2 for an intermolecular carbonyl- ene reaction of various alkenes and ethyl 3,3,3-trifluoropyruvate and an intramolecular carbonyl-ene reaction of (S)-citronellal. The anhydrous Fe(BF4)2 was prepared from FeCl2 and AgBF4. The developed anhydrous Fe(BF4)2 system was proved to be better in catalyzing the reaction than the commercial product Fe(BF4)2·6H2O. A 5 mol% of the anhydrous salt, Fe(BF4)2, was used as a catalyst and the homoallylic alcohols reaction products, were afforded in 36-87% yields with a wide scope of various of 1,1-disubstituted alkenes and the cyclization of citronellal. Only traces of product were obtained using 2-(1-methylethenyl)-pyridine as an heteroatom aromatic alkene, probably because the pyridine deactivated the catalyst. Less nucleophilic non-cyclic aliphatic alkenes, such as mono-substituted alkenes, i.e. 1-hexene, led only to traces of the expected product. The anhydrous Fe(BF4)2 catalytic system was used for the intramolecular carbonyl-ene reaction of (S)- citronellal in achievintg a total 70% yield with 45% (+)-isopulegol and 25% (−)-neo-isopulegol.
Overall, the developed catalytic system was efficient toward 1,1-disubstituted aromatic and aliphatic alkenes. The carbon-carbon bond forming process was affected by the electronic and steric properties of the substituents on the arene ring in the intermolecular carbonyl-ene reaction. Hence, the method is valuble since the catalytic system could be applied to other organic transformations. It is the first method using an iron salt as a Lewis acid for the carbonyl-ene reaction of ethyl trifluoropyruvate as activated ketone. Compared to the methods in the literature, such as the ones using Sc(OTf)3, SnCl4, ZnBr2, BiCl3, and Bi(OTf)3·xH2O, this catalyst was better than bismuth salts and SnCl4, but was not as efficient as ZnBr2 or Sc(OTf)3. There are still aspects that could be improved, such as the effects of temperature and solvents for the cyclization of citronellal. For examples, at a lower temperature of -78 °C, Sc(OTf)3 afforded higher yields and diastereoselectivities than at room temperature; a 20% of total yield of cyclization products was
121
obtained using FeCl3 in benzene as a solvent, while in CH2Cl2 there was 58% yield. Furthermore, from a green chemistry point of view, there are various green solvents that could be potential alternatives to halogenated solvents for this reaction. Since there was only one example of using iron in the cyclization of citronellal, the development of Fe(BF4)2 as catalytic system in the cyclization of citronellal was meaningful in providing an alternative tool in homogeneous catalysis.
5.2 Caffeine-derived NHC-iron-catalyzed carbonyl-ene reaction
N-heterocyclic carbenes (NHC) are recognized as promising ligands in transition metals catalysis. Caffeine-derived NHC-iron catalytic systems were developed for an intermolecular carbonyl-ene reaction of various alkenes with ethyl 3,3,3-trifluoropyruvate and an intramolecular carbonyl-ene reaction of citronellal. Optimized conditions were developed from the screening of iron salts, solvents, catalyst loading and counter anions. Three caffeine-derived imidazolium salts were used as NHC precorsurs and were examined with iron salts. Fe(OTf)2 was found to efficiently catalyze the reaction while complexed with methylated-caffeine-derived NHC ligand. A low catalyst loading of 2.5 mol% methylated-caffeine-derived-NHC-Fe(OTf)2 afforded 85% of the ene product. An anion effect, similar as in the literature was also demonstrated. Three caffeine-derived NHCs were used with anhydrous irons salts Fe(BF4)2 and Fe(SbF6)2, which were prepared from FeCl2 and silver salts
Ag(BF4)2 and Ag(SbF6)2 respectively, and the caffeine-derived-[NHC-Fe]2+(SbF6)22− was found to be efficient in catalyzing carbonyl-ene reaction of various enophiles with ethyl trifluoropyruvate in 22- 99% yields. However, low yields (22-27%) were obtained when using halogenated -methyl styrene as nucleophile with ethyl trifluoropyruvate, which were even lower than using anhydrous Fe(BF4)2 as a catalyst alone. The reasons were attributed to two aspects: the alkenes were less nucleophilic due to the inductive effect of the halogen and possibly due to the affinity between NHC-Fe and halogen atoms or halogen─carbon bonds since the NHC-Fe catalyst was used in the cross-coupling reaction of alkyl halides and aryl Grignards.
FeCl2 and FeCl3 were examined for the cyclization of citronelall, an intramolecular carbonyl-ene rection. An optimized combination of SbF6¯ and Cl¯ with iron, NHC-FeCl2(SbF6), was efficiently and selectively used as a catalyst to convert citronellal into the desired isopulegol. A total 80% yield was achieved with selectively 70% of the desired product isopulegol. The variation of the anions SbF6¯ and Cl¯ demonstrated a way of tuning the electronic property of NHC-iron, or generally NHC-metal,
122
a manner of changing the Lewis acidity of the metal to adapt to different substrates even in other reactions.
Much effort in understanding the mechanism of caffeine-derived-NHC-iron catalyst was put into the functionalization of the caffeine structure. A comparison of the yields between two selected ligands without caffeine pyrimidyl ring and two caffeine-derived NHCs highlighted the importance of caffeine pyrimidyl ring in obtaining better yields. The conjugated effect of the pyrimidyl ring of caffeine reduced the electron-donating of carbene to iron, hence a conjugation effect could be used as a manner of fine-tuning the electronic property of the N-heterocyclic carbene.
For the first time, a caffeine-derived NHC-iron system has been explored as a catalyst in the carbonyl-ene reaction. Moreover, much higher yields and less side products were obtained by using NHC-iron-catalyst system than using irons salts as Lewis acid alone in the carbonyl-ene reaction. Since there was only one application of using caffeine-derived NHC with transition metal in organic synthetic organic chemistry, the development of caffeine-derived NHC with iron was significant in expanding the application of NHC-iron catalysis and in underlining the importance of the unique performance of caffeine-derived NHC in transition metal catalysis. Therefore, caffeine-derived NHCs are promising ligands with other transition metals.
5.3 Caffeine-derived imidazolium-Fe(OTf)2 catalyzed Diels-Alder reaction
A green catalytic system was developed to promote the Diels-Alder (D-A) reaction of cyclopentadiene with various enophiles including bidentate α,β-dicarbonyl compounds, ketones, aldehydes, and esters. The catalyst was derived of a caffeine-derived imidazolium salt and Fe(OTf)2 was shown to be recycled for up to five times while maintaining the same level of yields of the Diels-Alder reaction and recyclability. The caffeine-derived imidazolium salt was not only used as an NHC precursor, but also view as a solid ionic liquid since it has similar structure of an ionic liquid. 16 substrates were tested in the reaction. Dimethyl carbonate as green solvent was used as reaction medium due to its green features. Apart from that, various green solvents were tested with achieving good yields to highlight the importance of green organic chemistry.
The system had attractive features: caffeine, iron, dimethyl carbonate, and recyclability of the catalyst. From a green chemistry perspective, iron and caffeine are naturally abundant, low costly,
123
and environmentally-benign. Consequently, developing new synthetic methods using iron, which is inexpensive, ubiquitous, and relatively less toxic in comparison with other metals, represents a significant advantage. Dimethyl carbonate as a green solvent together with the good recyclability of the catalyst highlighted the green features of the catalytic system. Thus, the green catalytic system is an alternative to traditional methods since almost all the traditional solvents used in the literature were halogenated solvent such as CH2Cl2, DCE, and chloroform, or benzene. However, the reaction was catalyzed in a non-enantioselective way since the catalyst is achiral. Hence, in future work, recyclable chiral catalyst could be developed for the enantioselective Diels-Alder reaction used, and various green solvents could be selected as preferred solvents.
5.4 Carbene-diol family ligand
A list of C2 symmetric NHC precursors 6b, 7b, 10, 11a, 11b, a non-C2 symmetric ligand 9, and a zwitterion 8b were synthesized. Two enantiopure trans diamines were used for the synthesis of 6b and 7b. Ligands 8b, 9, 10, 11a and 11b were derived from the epoxide-opening product of imidazolium and an enantio-enriched oxirane bearing a tert-butyl group. The ligands posses the following features: variations on the structures of the backbone of the imidazolium ring, have a chiral center bearing a tert-butyl group and a hydroxyl group, potentially from bidentate coordination, tridentate, tetradentate to pentadentate coordination. The ligand 11a was selected to complex with metals Ag2O, Pd(OAc)2, Fe(OAc)2 and Fe(OTf)2, but only with silver oxide, a 1H NMR of the complex was obtained and a structure of the complex was proposed. Thus, there is still future work regarding the complexation of these ligands with different transition metals. These ligands were tested with iron in the carbonyl-ene reaction of -methyl styrene and ethyl trifluoropyruvate in an enantioselective way, but no ee was obtained so far while the yields were mainly from moderate to good. In future work, optimization could be envisioned regarding the aspects of complexation with transition metals, such as Ru, Ir, Cu, and Fe, solvent, temperature, and the way NHC generated (in- situ or a carbene compound). Although no such NHC-dioxide-iron complex has been synthesized until now, more strict conditions of anhydrous environment for the synthesis would be benificial. Overall, these newly developed ligands are promising to transition metals in the asymmetric catalysis and transition metal catalysis.
124
A few well-known C2 symmetric ligands, including Box ligand, Bolm’s ligand, and two pybox ligands, were employed with iron salts in the enantioselective carbonyl-ene reaction of ethyl trifluoropyruvate. The obtained yields were from good to excellent with all of these ligands. The enantiomeric excess was 33% with 98% yield by using Box ligand and Fe(SbF6)2. Hence, there could be future work with a further optimization of the use of Fe(SbF6)2 with Box ligand and the parameters of temperature and solvent.
125
Chapter six
Experimental section
6.1 General information
All reactions were performed in flame-dried tubes under an atmosphere of argon. Solvents THF and
CH2Cl2 (DCM) were freshly distilled over CaH2 prior to use. Cotton-celite-plugged pipet and pipet for filtration were pre-dried in an oven overnight. Ethyl 3,3,3-trifluoropyruvate and α-methylstyrene were purchased from VWR. 2-Isopropenylnaphthalene, methylenecyclohexane, methylenecyclopentane, and (S)-(−)-citronellal are commercial products from Sigma-Aldrich. Other substituted α-methyl styrenes were lab-synthesized products by using literature method.1 Thin-layer chromatography (TLC) was carried out on 250 μm commercial silica gel plates and compounds were visualized using
UV absorbance and/or aqueous KMnO4. Flash column chromatography was performed on silica gel
(230−400 mesh). 1H and 13C{H}: NMR spectra were recorded on a 400 MHz spectrometer in CDCl3. For 1H NMR (400 MHz), chemical shifts are reported in ppm (from high ppm to low ppm). IR spectra were recorded on a FT−IR spectrometer with NaCl crystal carrier and are reported in reciprocal centimeter (cm−1).
6.2 Experimental part of the carbonyl-ene reaction
6.2.1 General procedure for the carbonyl-ene reaction of alkenes with ethyl 3,3,3- trifluoropyruvate by using unhydrous Fe(BF4)2:
FeCl2 (1.7 mg, 0.0125 mmol) and AgBF4 (5 mg, 0.025 mmol) were added into a flame-dried test tube. The test tube was put under high vacuum (pumped with argon three times). Afterwards, distilled THF (1 mL) was added and the resulting solution was stirred for 0.3 h. Then, using a pre-dried pipet plugged with cotton and celite to filter the precipitated AgCl under argon atmosphere. To the resulting solution was added 50 mg of 4Å MS and the solution was then evaporated under reduced pressure. The test tube was then put under high-vacuum (1 Torr) for 0.3 h. Under argon atmosphere fresh- distilled CH2Cl2 (0.5 mL) was then added into the tube and stirred the solution for 0.1 h. Then, α-
126
methyl styrene (30 mg, 0.25 mmol) and ethyl 3,3,3-trifluoropyruvate (64 mg, 0.375 mmol) were added via a syringe into the solution. The tube was put under argon protection and the solution was stirred for 48 h. Then, the crude reaction mixture was directly purified by silica flash chromatography (eluent: hexane/ethyl acetate). 63 mg of carbonyl-ene product 3a was obtained (yield = 87%)
6.2.2 General procedure for performing NHC-iron carbonyl-ene reaction of alkenes with ethyl 3,3,3-trifluoropyruvate or cyclization of (S)-(−)-citronellal:
Anhydrous FeCl2 (1.7 mg, 0.0125 mmol) and AgSbF4 (8.6-9mg, 0.025mmol) were added into a small flame-dried vial. The vial was put under high vacuum-argon three times. Then, fresh distilled THF 1mL was injected and the solution was stirred for 20 minutes. Used pre-dried cotton-celite-plugged pipet to filter the white precipitate AgCl under argon atmosphere and 0.5 mL fresh distilled THF was used to wash the cotton-celite-plugged pipet. The resulting solution was added 100 mg 4Å MS and evaporated under reduced pressure of an air-pump. After the evaporation was done, immediately the vial was transferred to be under high-vacuum (1 Torr) for 15-20 minutes. After that, under argon atmosphere it was added imidazolium salt 1,3,7-trimethyl-9-ethylxanthinium bis(trifluoromethanesulfonyl)amide/Ligand 2 (7.5 mg, 0.015 mmol) and injected 0.5 mL fresh-distilled DCM in, the resulting solution was stirred for 20-30 minutes. Then KOtBu-THF (26 μL 1M) solution (commercial product from Sigma-Aldrich) was injected into the solution and the solution was stirred under argon protection for 3 hours for complexation. It was injected the starting materials α-methyl styrene (30 mg, 0.25mmol) and ethyl 3,3,3-trifluoropyruvate (64 mg, 0.375 mmol) directly into the stirring solution. The vial was put under argon atmosphere and stirred for 48 hours. When the reaction was finished, directly loaded the sample onto silica flash chromatography by using eluent of hexane/ethyl acetate 97/3, and the product was dried under high vacuum. 65 mg (Yield = 90%) of ene-product was obtained and characterized by 1H NMR analysis and 13C NMR.
*Specially, for methylene−cyclohexane substrate, FeCl2 (2.6 mg, 0.015 mmol) and AgSbF4 (10.3mg, 0.03mmol) and imidazolium salt 1,3,7-trimethyl-9-ethylxanthinium bis(trifluoromethanesulfon- yl)amide (15 mg, 0.03 mmol) were used, and KOtBu-THF solution (40 μL, 1M) was injected. Other procedures were same as described above.
127
6.2.3 Spectral data of ligands (1−5) in carbonyl-ene reaction
Ligand 1─1,3,7,9-tetramethyl-xanthinium Iodide2
1H NMR (400 MHz, DMSO-d6) δ 9.23 (s, 1H), 4.10 (s, 3H), 4.01 (d, J = 0.7 Hz, 3H), 3.69 (s, 3H),
3.23 (s, 3H). 13C NMR (100 MHz, DMSO-d6) δ 153.72, 150.61, 140.03, 139.68, 108.16, 37.24, 36.05, 31.75, 28.84.
1H NMR (400 MHz, Acetone-d6) δ 9.72 (s, 1H), 4.43 (s, 3H), 4.23 (d, J = 0.7 Hz, 3H), 3.91 (s, 3H),
3.32 (s, 3H). 13C NMR (125 MHz, Acetone-d6) δ 205.26 (C=O), 153.54, 139.53, 109.99, 37.20, 35.55, 31.26, 27.93.
1H NMR (400 MHz, CDCl3) δ 10.56 (s, 1H), 4.43 (s, 3H), 4.26 (d, J = 0.5 Hz, 3H), 3.85 (s, 3H), 3.41
(s, 3H). 13C NMR (125 MHz, CDCl3) δ 139.65, 109.99, 38.32, 36.31, 32.03, 29.04. IR (ZnSe): 3492, 2997, 1712, 1670, 1580, 1541, 1447, 1403, 1340, 1298, 1231, 1181, 1089, 1045, 998, 920, 877, 741 cm−1.
Ligand 2─1,3,7-trimethyl-9-ethylxanthinium bis(trifluoromethanesulfonyl) amide3
1H NMR (400 MHz, Acetone-d6) δ 9.28 (s, 1H), 4.86 (q, J = 7.3 Hz, 2H), 4.25 (d, J = 0.7 Hz, 3H),
3.89 (s, 3H), 3.32 (s, 3H), 1.71 (t, J = 7.3 Hz, 3H). 13C NMR (125 MHz, Acetone-d6) δ 205.26, 138.31, 118.84, 45.77, 35.77, 31.39, 27.94, 15.09. IR (ZnSe): 3164, 3100, 2957, 1722, 1672, 1580, 1544, 1450, 1347, 1334, 1321, 1180, 1047, 854, 773, 761, 747, 740 cm−1.
Ligand 3─1,3,7,9-tetramethyl-xanthinium hexafluoroborate4
1H NMR (500 MHz, Acetone-d6) δ 9.17 (s, 1H), 4.45 (s, 3H), 4.27 (d, J = 0.8 Hz, 3H), 3.94 (s, 3H),
3.36 (s, 3H). 13C NMR (126 MHz, Acetone-d6) δ 205.29, 153.48, 139.37, 109.99, 36.97, 35.61, 31.10, 27.93. IR (ZnSe): 3392, 3181, 3122, 1717, 1661, 1581, 1547, 1461, 1348, 1267, 1182, 1105, 1054, 1031, 812, 743 cm−1.
Ligand 4─1,3-dimethyl-imidazolium iodide5
128
1H NMR (300 MHz, CDCl3) δ 10.16 (s, 1H), 7.31 (s, 2H), 4.09 (s, 6H). 13C NMR (100 MHz, CDCl3) δ 137.6, 123.38, 110.0, 37.10. IR (ZnSe): 3421, 3065, 2941, 2851, 2733, 2453, 1728, 1616, 1573, 1541, 1057, 805, 744, 710 cm−1.
Ligand 5─1-methyl-3-ethyl-imidazolium iodide6
1H NMR (300 MHz, CDCl3) δ 10.49 (s, 1H), 7.46 (s, 2H), 4.40 (q, J = 7.4 Hz, 2H), 1.59 (t, J = 7.4 Hz,
3H). 13C NMR (100 MHz, CDCl3) δ 135.3, 123.7, 122.0, 119.7 (q, J = 320.4 Hz), 45.1, 36.0, 14.8. IR (ZnSe): 3159, 3123, 1573, 1455, 1346, 1167, 1131, 1049, 842, 789, 739 cm−1.
Products of carbonyl-ene reaction
Ethyl 2-hydroxy-4-phenyl-2-(trifluoromethyl)pent-4-enoate (3a).7
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 63 mg, 0.22 mmol, 87%; for NHC-Fe catalyzed reaction: 65 mg, 0.225 mmol, 90%). 1H NMR (400
MHz, CDCl3) δ 7.35–7.26 (m, 5H), 5.39 (d, J = 1.3 Hz, 1H), 5.28 (d, J = 1.3 Hz, 1H), 4.03 (dq, J = 10.6, 7.2 Hz, 1H), 3.76 (d, J = 0.9 Hz, 1H), 3.64 (dq, J = 10.6, 7.2 Hz, 1H), 3.28 (d, J = 14.0 Hz, 1H),
3.04 (d, J = 14.0, 1.0 Hz, 1H), 1.11 (t, J = 7.2 Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 168.9, 141.0, 128.1, 127.9, 126.8, 123.4 (q, J = 286.2 Hz), 119.4, 77.09 (q, J = 28.9 Hz), 63.5, 37.0, 13.5. IR (NaCl): 3491, 2985, 1741, 1629, 1446, 1370, 1312, 1227, 1184, 1136, 1050, 911, 778, 701 cm–1.
Ethyl 2-hydroxy-4-(p-tolyl)-2-(trifluoromethyl)pent-4-enoate (3b).8
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 56 mg, 0.19 mmol, 75%; for NHC-Fe catalyzed reaction: 59.3 mg, 0.198 mmol, 79%). 1H NMR (300
MHz, , CDCl3) δ 7.25-7.21 (m, 2H), 7.16–7.06 (m, 2H), 5.35 (s, 1H), 5.21 (s, 1H), 4.05 (dq, J = 10.5, 7.3 Hz, 1H), 3.73 (s, 1H), 3.68 (dq, J = 10.5, 7.3 Hz, 1H), 3.25 (d, J = 14.4 Hz, 1H), 3.01 (d, J = 14.4
Hz, 1H), 2.32 (s, 3H), 1.13 (t, J = 7.3 Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 169.0, 140.9, 138.0, 137.5, 128.8, 126.6, 123.4 (q, J = 286.1 Hz), 118.5, 77.2 (d, J = 28.8 Hz), 63.5, 37.0, 21.1, 13.6. IR (NaCl): 3487, 2985, 1741, 1514, 1446, 1370, 1312, 1227, 1183, 1134, 1052, 1018, 910, 824 cm–1.
129
Ethyl 2-hydroxy-4-(p-methoxylphenyl)-2-(trifluoromethyl)pent-4-enoate (3c).9
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 36 mg, 0.11 mmol, 60%; for NHC-Fe catalyzed reaction: 78.7 mg, 0.248mmol, 99%). 1H NMR (400
MHz, CDCl3) δ 7.33–7.22 (m, 2H), 6.89–6.78 (m, 2H), 5.32 (d, J = 1.3 Hz, 1H), 5.18 (d, J = 1.3 Hz, 1H), 4.07 (dq, J = 10.7, 7.1 Hz, 1H), 3.80 (s, 3H), 3.75 (s, 1H), 3.73 (dq, J = 10.7, 7.1 Hz, 1H), 3.24
(d, J = 14.0 Hz, 1H), 3.01 (d, J = 14.0 Hz, 1H), 1.15 (t, J = 7.1, 3H). 13C NMR (100 MHz, CDCl3) δ 169.0, 159.3, 140.4, 133.3, 127.9, 123.4 (q, J = 286.2 Hz), 117.8, 113.5, 76.9 (q, J = 28.9 Hz), 63.5, 55.3, 37.2, 13.6. IR (NaCl): 3486, 2986, 2839, 2361, 1742, 1608, 1445, 1370, 1312, 1225, 1181, 1136, 1033, 1017 cm–1.
Ethyl 2-hydroxy-4-(4-bromophenyl)-2-(trifluoromethyl)pent-4-enoate (3d).9
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 42 mg, 0.11 mmol, 46%; for NHC-Fe catalyzed reaction: 22.4 mg, 0.061 mmol, 25%). 1H NMR (400
MHz, CDCl3) δ 7.44 (d, J = 8.7 Hz, 1H), 7.20 (d, J = 8.7 Hz, 1H), 5.38 (s, 1H), 5.28 (s, 1H), 4.12 (dq, J = 10.4, 7.2 Hz, 1H), 3.79 (dq, J = 10.4, 7.2 Hz, 1H), 3.74 (s, 1H), 3.21 (d, J = 14.1 Hz, 1H), 3.03 (d,
J = 14.1 Hz, 1H), 1.17 (t, J = 7.2 Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 168.8, 140.0 (d, J = 19.9 Hz), 131.2, 128.4, 123.3 (q, J = 286.3 Hz), 121.7, 119.8 , 77.2 (q, J = 28.9 Hz), 63.7, 36.8, 13.6. IR (NaCl): 3485, 2986, 1741, 1489, 1312, 1226, 1185, 1135, 1009, 832 cm–1.
Ethyl 2-hydroxy-4-(4-Fluorophenyl)-2-(trifluoromethyl)pent-4-enoate (3e).8
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 36 mg, 0.12 mmol, 47%; for NHC-Fe catalyzed reaction: 21 mg, 0.069 mmol, 27%). 1H NMR (400
MHz, CDCl3) δ 7.33–7.27 (m, 2H), 7.06–6.95 (m, 2H), 5.34 (s, 1H), 5.25 (s, 1H), 4.10 (dq, J = 10.7, 7.2 Hz, 1H), 3.75 (s, 1H), 3.76 (dq, J = 10.7, 7.2 Hz, 1H), 3.23 (d, J = 14.0 Hz, 1H), 3.03 (d, J = 14.0
Hz, 1H), 1.16 (t, J = 7.2 Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 169.0, 162.4 (d, J = 247.2 Hz), 140.1, 137.1, 128.4 (d, J = 8.0 Hz), 123.3 (q, J = 286.2 Hz), 119.3 (d, J = 0.9 Hz), 115.0 (d, J = 21.3 Hz), 77.2 (q, J = 38.8 Hz), 63.6, 37.05, 13.6. IR (NaCl): 3446, 1740, 1604, 1505, 1313, 1223, 1184, 1134, 1051, 840 cm–1.
130
Ethyl 2-hydroxy-4-(o-methoxylphenyl)-2-(trifluoromethyl)pent-4-enoate (3f).10
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 42 mg, 0.11 mmol, 46%; for NHC-Fe catalyzed reaction: 60 mg, 0.189 mmol, 75%). 1H NMR (400
MHz, CDCl3) δ 7.44 (d, J = 8.7 Hz, 1H), 7.20 (d, J = 8.7 Hz, 1H), 5.38 (s, 1H), 5.28 (s, 1H), 4.12 (dq, J = 10.4, 7.2 Hz, 1H), 3.79 (dq, J = 10.4, 7.2 Hz, 1H), 3.74 (s, 1H), 3.21 (d, J = 14.1 Hz, 1H), 3.03 (d,
J = 14.1 Hz, 1H), 1.17 (t, J = 7.2 Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 168.8, 140.1, 140.0, 131.2, 128.4, 123.3 (q, J = 286.3 Hz), 121.7, 119.8 , 77.2 (q, J = 28.9 Hz), 63.7, 36.8, 13.6. IR (NaCl): 3485, 2986, 1741, 1489, 1312, 1226, 1185, 1135, 1009, 832 cm–1.
Ethyl 2-hydroxy-4-(cyclohexyl-1-ene)-2-(trifluoromethyl)but-4-enoate (3g).7
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 40 mg, 0.15 mmol, 60%; for NHC-Fe catalyzed reaction: 53.2 mg, 0.2 mmol, 80%). 1H NMR (400
MHz, CDCl3) δ 5.55 (s, 1H), 4.41–4.26 (m, 2H), 3.77 (s, 1H), 2.66 (d, J = 13.5 Hz, 1H), 2.50 (d, J = 13.5 Hz, 1H), 2.16–1.81 (m, 4H), 1.62–1.50 (m, 4H), 1.34 (t, J = 7.1 Hz, 3H). 13C NMR (100 MHz,
CDCl3) δ 169.7, 130.9, 127.7, 123.4 (q, J = 286.5 Hz), 78.19 (q, J = 28.3 Hz), 63.5, 39.6, 29.8, 25.4, 22.8, 21.9, 14.0. IR (NaCl): 3487, 2933, 1740, 1608, 1437, 1308, 1228, 1177, 1124, 1098, 1014 cm– 1.
Ethyl 2-hydroxy-4-(cyclopentyl-1-ene)-2-(trifluoromethyl)but-4-enoate (3h).12
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 43 mg, 0.17 mmol, 83%; for NHC-Fe catalyzed reaction: 62.4 mg, 0.248 mmol, 99%). 1H NMR (400
MHz, CDCl3) δ 5.53 (s, 1H), 4.40–4.26 (m, 2H), 3.80 (s, 1H), 2.86 (d, J = 14.5 Hz, 1H), 2.67 (d, J = 14.5 Hz, 1H), 2.40–2.14 (m, 4H), 1.86–1.79 (m, 2H), 1.33 (t, J = 7.1 Hz, 3H). 13C NMR (100 MHz,
CDCl3) δ 169.7 , 136.4, 130.1, 123.3 (q, J = 286.6 Hz), 77.71 (q, J = 28.5 Hz), 63.6, 36.0, 33.0 (q, J = 1.6 Hz), 32.4, 23.5, 13.9. IR (NaCl): 3495, 2955, 2852, 1740, 1472, 1311, 1222, 1184, 1120, 1037, 988, 862, 715 cm–1.
Ethyl 2-hydroxy-4-(naphthalen-2-yl)-2-(trifluoromethyl)pent-4-enoate (3i).8
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 30 mg, 0.09 mmol, 36%; for NHC-Fe catalyzed reaction: 50.7 mg, 0.150 mmol, 61%). 1H NMR (400
131
MHz, CDCl3) δ 7.86–7.74 (m, 4H), 7.53–7.41 (m, 3H), 5.53 (d, J = 1.2 Hz, 1H), 5.38 (d, J = 1.2 Hz, 1H), 3.94 (dq, J = 10.7, 7.2 Hz, 1H), 3.79 (d, J = 1.0 Hz, 1H), 3.51 (dq, J = 10.7, 7.2 Hz, 1H), 3.40 (d, J = 14.1 Hz, 1H), 3.15 (dd, J = 14.1, 1.0 Hz, 1H), 1.01 (t, J = 7.2 Hz, 3H). 13C NMR (100 MHz,
CDCl3) δ 169.0, 141.0, 138.2, 133.1, 132.8, 128.1, 127.8, 127.5, 126.3, 126.1, 125.5, 125.0, 123.4 (q, J = 286.5 Hz), 119.9, 77.2 (q, J = 28.9 Hz), 63.5, 37.0, 13.4. IR (NaCl): 3485, 2986, 1740, 1609, 1505, 1445, 1369, 1312, 1236, 1222, 1183, 1138, 1015, 855 cm–1.
Ethyl 2-hydroxy-4-( thiophen-2-yl)-2-(trifluoromethyl)pent-4-enoate (3j).11
The product was obtained as a colorless oil (39 mg, 0.13 mmol, 53%). 1H NMR (400 MHz, CDCl3) δ 7.17 (dd, J = 5.1, 1.2 Hz, 1H), 7.06 (dd, J = 3.7, 1.2 Hz, 1H), 6.96 (dd, J = 5.1, 3.7 Hz, 1H), 5.54 (s, 1H), 5.15 (d, J = 0.9 Hz, 1H), 4.22 (dq, J = 10.7, 7.2 Hz, 1H), 4.00 (dq, J = 10.7, 7.2 Hz, 1H), 3.88 (d, J = 0.9 Hz, 1H), 3.22 (d, J = 14.2 Hz, 1H), 3.04 (dd, J = 14.2, 1.0 Hz, 1H), 1.23 (t, J = 7.2
Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 168.9, 144.6, 134.1, 127.2, 124.9, 124.8, 123.3 (q, J = 286.5 Hz), 116.8, 77.5 (q, J = 28.9 Hz), 36.8, 36.8, 13.7. IR (NaCl): 2595, 2928, 2870, 1750, 1677, 1517, 1464, 1368, 1340, 1165, 1181, 1028, 758, 608 cm–1.
Ethyl 2-hydroxy-4-(4-chlorophenyl)-2-(trifluoromethyl)pent-4-enoate (3k).9
The product was obtained as a colorless oil (for NHC-Fe catalyzed reaction: 18 mg, 0.056 mmol,
22%).1H NMR (400 MHz, CDCl3) δ 7.34–7.21 (m, 4H), 5.38 (s, 1H), 5.28 (s, 1H), 4.10 (dq, 1H), 3.83 (dq, J = 7.2 Hz, 1H), 3.74 (s, 1H), 3.21 (d, J = 14.3 Hz, 1H), 3.03 (d, J = 14.3 Hz, 1H), 1.17 (t, J = 7.2
Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 168.9, 140.1, 139.5, 133.6, 128.3, 128.1, 123.26 (d, J = 286.4Hz), 119.8, 77.0 (d, J = 28.7 Hz), 63.7, 36.8, 13.6. IR (NaCl): 3485, 2986, 1741, 1609, 1493, 1312, 1225, 1184, 1135, 1013, 835 cm–1.
Ethyl 2-hydroxy-4-methoxyl-2-(trifluoromethyl)pent-4-enoate (3l).12
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 36.3 mg, 0.15 mmol, 61%; for NHC-Fe catalyzed reaction: 36.3 mg, 0.15 mmol, 61%). 1H NMR (400
MHz, CDCl3) δ 4.34 (qd, J = 7.2, 1.4 Hz, 2H), 4.10–4.03 (m, 2H), 3.95 (d, J = 0.8 Hz, 1H), 3.48 (s, 3H), 2.93 (d, J = 14.1 Hz, 1H), 2.66 (d, J = 14.1 Hz, 1H), 1.34 (t, J = 7.2 Hz, 3H). 13C NMR (100 MHz,
CDCl3) δ 169.2, 156.4, 123.3 (q, J = 286.0 Hz), 86.0, 76.0 (q, J = 29.1 Hz), 63.5, 55.0, 37.7 (d, J =
132
1.5 Hz), 13.9. IR (NaCl): 3486, 2986, 2840, 1742, 1608, 1514, 1466, 1312, 1225, 1181, 1136, 1098, 1050 cm–1.
(+)-Isopulegol (5).13
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 18 mg, 0.12 mmol, 45%; for NHC-Fe catalyzed reaction: 28 mg, 0.182 mmol, 70%). 1H NMR (400
MHz, CDCl3) δ 4.90 (m, 1H), 4.86 (s, 1H), 3.46 (td, J = 10.3, 4.1 Hz, 1H), 2.05 (dtd, J = 12.5, 4.1, 1.8 Hz, 1H), 1.89 (ddd, J = 13.0, 9.9, 3.3 Hz, 1H), 1.76–1.61 (m, 5H), 1.50 (dddd, J = 18.6, 12.1, 6.6, 3.3 Hz, 1H), 1.38–1.25 (m, 1H), 0.95 (d, J = 6.6 Hz, 3H), 1.02-0.88 (m, 2H). 13C NMR (100 MHz,
CDCl3) δ 146.6, 112.9, 70.3, 54.1, 42.6, 34.3, 31.4, 29.6, 22.2, 19.2. IR (NaCl): 3411, 2951, 2924, 2856, 2360, 1646, 1456, 1376, 1376, 1260, 1052, 1027, 891 cm–1.
(−)-Neo-isopulegol (6).13
The product was obtained as a colorless oil (for anhydrous Fe(BF4)2 catalyzed carbonyl-ene reaction: 10 mg, 0.06 mmol, 25%; for NHC-Fe catalyzed reaction: 4 mg, 0.026 mmol, 10%). 1H NMR (400
MHz, CDCl3) δ 4.95 (s, 1H), 4.79 (s, 1H), 3.99 (m, 1H), 1.98 (dq, J = 13.5, 3.2 Hz, 2H), 1.79 (s, 3H), 1.78–1.65 (m, 3H), 1.45 (dd, J = 13.5, 3.2 Hz, 1H), 1.13 (ddd, J = 14.4, 12.3, 2.5 Hz, 1H), 1.0-0.84
(m, 2H), 0.88 (d, J = 6.5 Hz, 3H). 13C NMR (100 MHz, CDCl3) δ 147.3, 11.3, 66.3, 48.4, 40.9, 34.7, 25.8, 23.9, 22.2. IR (NaCl): 3475, 2922, 2868, 1784, 1714, 1646, 1456, 1376, 1376, 1123, 1026, 1027, 889 cm–1.
133
5.2.4 Copies of NMR spectra
1) Imidazolium salts (Ligand 1−5)
1 13 Ligand 1 HNMR, CNMR and HSQCAD in DMSO d6
134
135
1 13 Ligand 1 HNMR and CNMR in acetone d6
136
1 13 Ligand 1 HNMR and CNMR in CDCl3
137
1 13 Ligand 2 HNMR and CNMR in Acetone d6
138
1 13 Ligand 3 HNMR and CNMR in Acetone d6
139
1 13 Ligand 4 HNMR and CNMR in CDCl3
140
1 13 Ligand 5 HNMR and CNMR in CDCl3
141
2) Carbonyl-ene reaction products
1 13 Ethyl 2-hydroxy-4-phenyl-2-(trifluoromethyl)pent-4-enoate (3a) H NMR, C NMR CDCl3
142
143
Condition: OJ-H, Hex:iPrOH = 99.1:0.9, Flow : 1 mL/min, 220 nm. RT: 12.18, area: 66.4%; 18.50, area: 33.59%. ee = 32.8%.7b
144
1 13 Ethyl 2-hydroxy-4-(p-tolyl)-2-(trifluoromethyl)pent-4-enoate (3b) H NMR, C NMR CDCl3
145
1 13 Ethyl 2-hydroxy-4-(p-methoxylphenyl)-2-((trifluoromethyl)pent-4-enoate (3c) H NMR, C NMR CDCl3
146
1 13 Ethyl 2-hydroxy-4-(4-bromophenyl)-2-(trifluoromethyl)pent-4-enoate (3d) H NMR, C NMR CDCl3
147
1 13 Ethyl 2-hydroxy-4-(4-fluorophenyl)-2-(trifluoromethyl)pent-4-enoate (3e) H NMR, C NMR CDCl3
148
1 13 Ethyl 2-Hydroxy-4-(o-methoxylphenyl)-2-(trifluoromethyl)pent-4-enoate (3f) H NMR, C NMR CDCl3
149
1 13 Ethyl 2-Hydroxy-4-(cyclohexyl-1-ene)-2-(trifluoromethyl)but-4-enoate (3g) H NMR, C NMR CDCl3
150
1 13 Ethyl 2-Hydroxy-4-(cyclopentyl-1-ene)-2-(trifluoromethyl)but-4-enoate (3h) H NMR, C NMR CDCl3
151
1 13 Ethyl 2-Hydroxy-4-(naphthalen-2-yl)-2-(trifluoromethyl)pent-4-enoate (3i) H NMR, C NMR CDCl3
152
Ethyl 2-hydroxy-4-(thiophen-2-yl)-2-(trifluoromethyl)pent-4-enoate (3j) 1H NMR, 13C NMR
153
1 13 Ethyl 2-hydroxy-4-(4-chlorophenyl)-2-(trifluoromethyl)pent-4-enoate (3k) H NMR, C NMR CDCl3
154
1 13 Ethyl 2-hydroxy-4-methoxyl-2-(trifluoromethyl)pent-4-enoate (3l) H NMR, C NMR CDCl3
155
1 13 (+)-Isopulegol (5) H NMR, C NMR CDCl3
156
1 13 (−)-Neo-isopulegol (6) H NMR, C NMR CDCl3
157
6.3 Experimental part of Diels-Alder reaction
6.3.1 General procedure for Diels-Alder reaction
1) Preparation of catalyst 1 2 3 and 4
The relative imidazolium salt 1,3,7-trimethyl-9-ethylxanthinium bis(trifluoromethanesulfonyl) amide
(Caff-Et-NTf2) (2.5 mmol 1.26 g) and anhydrous Fe(OTf)2 (0.25 mmol 89 mg) were added to a 25 mL oven-dried flask. Injected 7 mL acetone to dissolve the salts, then adding 15 mL hexane into the solution. A brown solution formed at the bottom of the hexane-acetone solution in the flask. Then, evaporated the solvent slowly at reduced pressure and put the flask under high vacuum overnight. The catalyst 1 was obtained in quantitative yield. Catalysts 2, 3, and 4 were followed the same procedure.
2) General procedure for cyclopentadiene and bidentate dienophiles in dimethyl carbonate (DMC).
To an oven-dried vial, added catalyst 1 (0.005 mmol, 27.5 mg) and dienophile 3-acryloyl-1,3- oxazoline-2-one (0.5 mmol, 70.5 mg), then injected DMC 1 mL and started stirring for 10 minutes. After that, transferred the vial into icy water bath or ethanol at 2-3 ℃ by cooling equipment and injected distilled cyclopentadiene (1 mmol, 66 mg) into the vial. Kept the reaction stirring for 3 hours. After the reaction was completed, hexane 5 mL was added into the reaction solution to precipitate the catalyst. Through filtration of cotton-plugged pipet, the crude product was transferred to a new vial. The filtrate of crude product was evaporated and the crude product was purified by hexane/EtOAc to obtain quantitative yield (99%, 103 mg). The precipitation of catalyst was washed by acetone and filtered through cotton-plugged pipet to be recycled after evaporation of acetone. All the mass of the catalyst (27.5 mg) was recycled. For the other substrates, 0.25 mmol dienophiles were used with 0.5 mmol diene and the catalyst loading was 0.0025 mmol, 14 mg.
3) General procedure for cyclopentadiene and methyl acrylate/ methyl propiolate in neat condition.
To an oven-dried vial, added catalyst 1 (0.005 mmol, 27.5 mg) and put the vial into icy-water bath or ethanol at 2-3 ℃ by cooling equipment. Injected dienophile methyl acrylate (5 mmol, 430 mg) and
158
distilled cyclopentadiene (1.2 eq., 6 mmol, 396 mg) and started stirring for 1 day. After that, added 5 mL hexane into the vial to precipitate the catalyst and through filtration of cotton-plugged pipet to get the crude product. The filtrate was evaporated and the product was put under high vacuum for 5-10 minutes. Pure product was obtained without further purification at 90% (677 mg) yield. The precipitation of catalyst was washed by acetone and filtered through cotton-plugged pipet to be recycled after evaporation of acetone. All the mass of the catalyst (27.5 mg) was recycled.
6.3.2 General procedure for caffeine-derived imidazolium-Fe(OTf)2 catalyzed acetylation of benzyl alcohol and acetic anhydride.
Catalyst 1 (27 mg, 0.1mol%) was added into a flame-dried testing tube. Benzyl alcohol (5 mmol, 536 mg) and acetic anhydride (6 mmol, 616 mg) were injected via a syringe. Then, the sulotion was stirred for 4 hours. After the reaction was finished, 5 mL hexane was added to precipitate the catalyst, and the product was in the solution. By evaporation of solvent, pure product benzyl methyl ester (748.1 mg, 4.94 mmol) without purification was obtained. The yield was 99.6%. When catalyst loading was 0.01 mmol%, 10mmol benzyl alcohol and 12 mmol acetic anhydride were used. The product obtained was 1.4185 g, 9.45 mmol, and yield was 95%.
6.3.3 Spectral data of the products of Diels-Alder reaction
3-Bicyclo[2.2.1]hept-5-ene-2-carbonyl)oxazolidin-2-one (2a):14
The product was obtained as a colorless semi-solid (103 mg, 0.495 mmol, 99%). 1H NMR (500 MHz,
CDCl3) Endo: δ 6.23 (dd, J = 5.3, 2.9 Hz, 1H), 5.86 (dd, J = 5.4, 2.6 Hz, 1H), 4.43–4.35 (m, 2H), 3.99–3.89 (m, 3H), 3.29 (br, 1H), 2.93 (br s, 1H), 1.94 (m, 1H), 1.49–1.37 (m, 3H). Exo (specific protons): δ 6.16 (s, 2H), 4.05–4.00 (m, 3H), 3.27–3.24 (m, 1H), 3.00 (s, 1H), 1.51 (m, 1H), 1.35 (m,
1H) 13C NMR (126 MHz, CDCl3) Endo: δ 174.7, 153.4, 138.1, 131.6, 62.0, 50.2, 46.3, 43.1, 42.9, 42.9, 29.5. Exo: δ 176.1, 153.3, 138.2, 135.9, 61.9, 46.8, 46.1, 43.1, 43.0, 41.9, 30.4. IR (NaCl): 2975, 1775, 1696, 1386, 1225, 1040, 1005, 761, 705 cm-1.
159
3-Methylbicyclo[2.2.1]hept-5-ene-2-carbonyl)oxazolidin-2-one (3a):14
The product was obtained as a white solid (104.0 mg, 0.470 mmol, 94%). 1H NMR (400 MHz, CDCl3) Endo: δ 6.36 (dd, J = 5.6, 3.1 Hz, 1H), 5.76 (dd, J = 5.7, 2.8 Hz, 1H), 4.39 (td, J = 8.0, 3.1 Hz, 2H), 4.03–3.88 (m, 2H), 3.51 (dd, J = 4.3, 3.5 Hz, 1H), 3.26 (br s, 1H), 2.51 (br s, 1H), 2.12–2.02 (m, 1H), 1.69 (d, J = 8.7 Hz, 1H), 1.44 (dq, J = 8.6, 1.6 Hz, 1H), 1.11 (d, J = 7.1 Hz, 3H). Exo (significant peaks): δ 6.30 (dd, J = 5.6, 3.1 Hz, 1H), 6.14 (dd, J = 5.6, 2.9 Hz, 1H). 13C NMR (126 MHz, CDCl3) Endo: δ 174.4, 153.5, 139.7, 130.9, 61.9, 51.3, 49.5, 47.5, 47.1, 43.0, 36.4, 20.4. Exo: δ 175.54, 153.44, 136.87, 135.53, 61.82, 50.64, 49.52, 47.50, 46.66, 43.07, 37.35, 18.84. IR (NaCl): 2963, 1776, 1696, 1386, 1229, 1100, 770, 735, 705 cm-1.
3-(Trifluoromethyl)bicyclo[2.2.1]hept-5-ene-2-carbonyl)oxazolidin-2-one (4a):
The product was obtained as a white solid (65 mg, 0.236 mmol, 95%). 1H NMR (500 MHz, CDCl3) δ 6.41–6.35 (m, 1H for endo ((6.38, dd, J = 5.0, 3.6 Hz)), 1H for exo), 6.14 (m, 1H), , 4.47–4.39 (m, 2H), 4.14–3.99 (m, 2H), 3.52 (m, 1H), 3.44, d, J = 5.6 Hz, 1H), 3.15 (br, 1H), 3.04 (br, 1H), , 1.58 (d, J = 8.9 Hz, 1H), 1.43 (d, J = 8.9 Hz, 1H). Exo: 5.90 (dd, J = 5.6, 2.8 Hz, 1H) 3.93 (m, 1H), 3.10 (d, J = 1.2 Hz, 1H), 2.76 (qdd, J = 10.3, 5.5, 1.3 Hz, 1H) 1.82 (dd, J = 9.0, 1H), 1.51 (d, J = 9.0 Hz, 1H).
13C NMR (126 MHz, CDCl3) Endo: δ 172.4, 153.3, 137.2, 134.7, 127.1 (q, J = 277.5 Hz), 62.1, 49.3, 46.2 (d, J = 1.0 Hz), 46.00 (d, J = 1.3 Hz), 45.3 (q, J = 26.9 Hz), 43.4 (q, J = 1.7 Hz), 42.9. Exo: δ 171.74, 153.3, 138.7, 133.2, 127.4 (q, J = 278.3 Hz), 62.1, 46.5, 48.3 (d, J = 1.3 Hz), 44.8 (d, J = 1.3 Hz), 45.4 (q, J = 27.0 Hz), 44.0 (d, J = 1.7 Hz), 42.9. IR (NaCl): 2990, 1779, 1696, 1362, 1284, 1125,
676 cm-1. MP: 79−80 °C. HRMS (ESI) [M+Na]+ calcd for C12H12F3NNaO3 requires 298.0662, found 298.0654.
160
3-Phenylbicyclo[2.2.1]hept-5-en-2-yl)(pyridin-2-yl)methanone (6a):15
The product was obtained as a white solid (63 mg, 0.229 mmol, 92%). 1H NMR (400 MHz, CDCl3) Endo: δ 8.69 (dd, J = 4.7, 0.6 Hz, 1H), 8.03 (d, J = 7.8 Hz, 1H), 7.82 (td, J = 7.7, 1.7 Hz, 1H), 7.45 (ddd, J = 7.5, 4.8, 1.1 Hz, 1H), 7.37–7.28 (m, 4H), 7.16 (m, 1H), 6.52 (dd, J = 5.5, 3.2 Hz, 1H), 5.86 (dd, J = 5.6, 2.7 Hz, 1H), 4.57 (dd, J = 5.1, 3.5 Hz, 1H), 3.58 (br s, 1H), 3.49 (d, J = 4.5 Hz, 1H), 3.12 (d, J = 1.2 Hz, 1H), 2.10 (d, J = 8.4 Hz, 1H), 1.64 (dd, J = 8.4, 1.6 Hz, 1H). Exo (specific peaks): δ 8.67 (dd, J = 4.7, 0.7 Hz, 1H), 8.09 (d, J = 7.8 Hz, 1H), 7.27–7.23 (m, 4H), 6.55–6.54 (m, 1H), 6.13 (dd, J = 5.5, 2.8 Hz, 1H), 4.24 (dd, J = 5.3, 0.9 Hz, 1H), 4.06 (dd, J = 5.2, 3.5 Hz, 1H), 3.23 (s, 1H),
3.13 (s, 1H), 1.90 (d, J = 8.5 Hz, 1H), 1.50 (dd, J = 8.5, 1.4 Hz, 1H). 13C NMR (126 MHz, CDCl3) Endo: δ 201.09, 153.57, 148.89, 144.64, 139.44, 136.85, 132.88, 128.40, 127.67, 126.92, 125.85, 122.20, 54.28, 49.37, 48.78, 48.26, 45.59. Exo: δ 202.20, 153.40, 148.97, 143.72, 137.05, 136.76, 136.19, 128.16, 127.91, 126.91, 125.90, 122.40, 52.10, 49.35, 48.99, 47.00, 46.92. IR (NaCl): 2974, 1690, 1600, 1568, 1497, 1332, 1272, 1019, 995, 740 cm-1. MP: 45–46 ° C.
3-(4-Methoxyphenyl)bicyclo[2.2.1]hept-5-en-2-yl)(pyridin-2-yl)methanone (7a):15
The product was obtained as a colorless oil (76 mg, 0.249 mmol, 99%). 1H NMR (400 MHz, CDCl3) Endo: δ 8.70–8.67 (m, 1H), 8.01 (d, J = 7.9 Hz, 1H), 7.81 (td, J = 7.7, 1.7 Hz, 1H), 7.44 (m, 1H), 7.27–7.23 (m, 2H), 6.84 (dd, J = 8.5, 1.6 Hz, 2H), 6.50 (m, 1H), 5.82 (m, 1H), 4.50 (m, 1H), 3.78 (d, J = 1.8 Hz, 3H), 3.54 (br s, 1H), 3.40 (d, J = 5.0 Hz, 1H), 3.04 (br, 1H), 2.07 (d, J = 8.4 Hz, 1H), 1.61 (d, J = 8.4 Hz, 1H). Exo: (specific protons): δ 8.65 (m, 1H), 8.07 (d, J = 7.9 Hz, 1H), 7.14 (d, J = 7.3 Hz, 1H), 6.78 (dd, J = 8.5, 1.6 Hz, 2H), 6.52 (m, 1H), 6.12 (m, 1H), 4.16 (d, J = 5.3 Hz, 1H), 3.96 (m, 1H), 3.76 (d, J = 1.8 Hz, 3H), 3.16 (s, 1H), 3.10 (s, 1H), 1.86 (d, J = 8.5 Hz, 1H), 1.47 (d, J = 8.5 Hz,
161
1H). 13C NMR (126 MHz, CDCl3) Endo: δ 201.2, 157.7, 153.56, 148.9, 139.4, 136.8, 136.6, 132.8, 128.6, 126.90, 122.2, 113.8, 55.3, 54.3, 49.7, 48.7, 48.2, 44.9. Exo: δ 202.3, 157.81, 153.4, 149.0, 137.0, 136.8, 136.2, 135.8, 129.0, 126.9, 122.4, 113.3, 55.2, 52.3, 49.30, 49.1, 47.00, 46.2. IR (NaCl): 2970, 1689, 1610, 1582, 1569, 1512, 1249, 1036, 678, 603 cm-1.
3-(4-Chlorophenyl)bicyclo[2.2.1]hept-5-en-2-yl)(pyridin-2-yl)methanone (8a):15
The product was obtained as a colorless oil (57.9 mg, 0.238 mmol, 96%). 1H NMR (400 MHz, CDCl3) Endo: δ 8.68–8.66 (m, 1H), 8.01 (dt, J = 7.9, 0.9 Hz, 1H), 7.81 (m, 1H), 7.45 (m, 1H), 7.24–7.12 (m, 4H), 6.48 (dd, J = 5.6, 3.2 Hz, 1H), 5.83 (dd, J = 5.6, 2.8 Hz, 1H), 4.47 (dd, J = 5.2, 3.4 Hz, 1H), 3.55 (s, 1H), 3.42 (dd, J = 5.2, 1.7 Hz, 1H), 3.05 (d, J = 1.4 Hz, 1H), 2.01 (d, J = 8.5 Hz, 1H), 1.62 (dd, J = 8.5, 1.7 Hz, 1H). Exo: (specific protons): δ 8.63 (m, 1H), 8.07 (dt, J = 7.8, 0.9 Hz, 1H), 7.42 (m, 1H), 7.29–7.24 (m, 4H), 6.52 (dd, J = 5.6, 3.1 Hz, 1H), 6.06 (dd, J = 5.6, 2.8 Hz, 1H), 4.15 (dd, J = 5.3, 1.2 Hz, 1H), 3.96 (dd, J = 5.2, 3.5 Hz, 1H), 3.16 (s, 1H), 3.11 (d, J = 1.4 Hz, 1H), 1.84 (d, J = 4.1
Hz, 1H), 1.48 (dd, J = 8.6, 1.6 Hz, 1H). 13C NMR (126 MHz, CDCl3) Endo: δ 200.9, 153.4, 148.89, 143.2, 139.2, 136.9, 133.0, 131.5, 129.0, 128.4, 127.0, 122.2, 54.4, 49.2, 48.73, 48.2, 45.0. Exo: δ 202.0, 153.3, 149.0, 142.2, 137.3, 136.8, 135.8, 131.6, 129.5, 127.9, 127.0, 122.4, 52.1, 49.12, 49.0, 47.0, 46.4. IR (NaCl): 2972, 1689, 1582, 1569, 1548, 1491, 1014, 670 cm-1.
3-Phenylbicyclo[2.2.1]hept-5-ene-2-carbonyl)pyridine 1-oxide (9a):16
The product was obtained as a colorless oil (72 mg, 0.247 mmol, 99%). 1H NMR (400 MHz, CDCl3) Endo: δ 8.16 (m, 1H), 7.42 (m, 1H), 7.35–7.27 (m, 6H), 7.16 (m, 1H), 6.46 (dd, J = 5.6, 3.2 Hz, 1H), 5.87 (dd, J = 5.6, 2.7 Hz, 1H), 4.50 (dd, J = 5.1, 3.4 Hz, 1H), 3.38 (s, 1H), 3.35 (d, J = 4.0 Hz, 1H),
162
3.09 (s, 1H), 1.88 (d, J = 8.6 Hz, 1H), 1.56 (ddd, J = 8.6, 3.5, 1.7 Hz, 1H). Exo (specific proton): δ
6.41 (dd, J = 5.6, 3.1 Hz, 1H). 13C NMR (126 MHz, CDCl3) Endo: δ 198.6, 147.4, 143.9, 140.3, 139.9, 133.1, 128.4, 127.6, 127.5, 126.3, 125.7, 125.4, 58.2, 49.1, 47.6, 46.4, 46.4. Exo: δ 199.7, 147.2, 143.0, 140.4, 137.0, 136.2, 128.2, 128.0, 127.9, 126.5, 126.0, 125.6, 56.7, 48.89, 48.4, 47.9, 46.9. IR (NaCl): 2974, 1694, 1600, 1548, 1500, 1427, 1293, 1021, 852, 700, 658 cm-1.
3-Phenylbicyclo[2.2.1]hept-5-en-2-yl]ethanone (10a).17
The product was obtaine as a color less oil (93 mg, 0.438 mmol, 88%). Endo product was nearly the single isomer. 1H NMR (400 MHz, CDCl3) δ 7.33–7.24 (m, 4H), 7.19 (m, 1H), 6.40 (dd, J = 5.7, 3.3 Hz, 1H), 6.03 (dd, J = 5.6, 2.8 Hz, 1H), 3.33 (br, 1H), 3.19 (dd, J = 5.0, 1.6 Hz, 1H), 3.07 (dd, J = 5.0, 3.3 Hz, 1H), 3.02 (d, J = 1.6 Hz, 1H), 2.16 (s, 3H), 1.86 (d, J = 8.6 Hz, 1H), 1.61 (dq, J = 8.6, 1.8 Hz,
1H). 13C NMR (100 MHz, CDCl3) δ 208.20, 144.4, 139.4, 133.1, 128.5, 128.12, 127.5, 126.0, 61.1, 48.5, 47.6, 46.5, 45.3, 29.2.
Bicyclo[2.2.1]hept-5-ene-2-carbonyl chloride (12a).18
The product was obtaine as a color less oil (47 mg, 0.300 mmol, 88%). 1H NMR (400 MHz, Chloroform-d) endo: δ 6.20 (dd, J = 5.6, 3.0 Hz, 1H), 5.99 (dd, J = 5.6, 2.8 Hz, 1H), 3.23 (s, 1H), 2.99 (dt, J = 9.4, 3.8 Hz, 1H), 2.91 (s, 1H), 1.91 (ddd, J = 13.0, 9.4, 3.8 Hz, 1H), 1.42 (m, 3H), 1.28 (d, J = 8.3 Hz, 1H). Exo (specific protons): 6.15 (dd, J = 5.5, 3.0 Hz, 1H), 6.11 (dd, J = 5.5, 2.8 Hz, 1H).
13C NMR (100 MHz, CDCl3) Endo: δ 181.2, 137.9, 132.4, 49.7, 45.7, 43.3, 42.5, 29.1.
5-(Methoxycarbonyl)bicyclo[2.2.1]hept-2-ene (13a).19
163
The product was obtained as a colorless oil (677 mg, 0.891 mmol, 90%). 1H NMR (400 MHz, CDCl3) endo: δ 6.19 (dd, J = 5.6, 3.1 Hz, 1H), 5.93 (dd, J = 5.6, 2.8 Hz, 1H), 3.62 (s, 3H), 3.19 (br, 1H), 2.95 (dt, J = 9.3, 3.9 Hz, 1H), 2.90 (br, 1H), 1.91 (ddd, J = 12.7, 9.4, 3.7 Hz, 1H), 1.46–1.39 (m, 2H), 1.27 (m, 1H). exo (specific protons): δ 6.14 (dd, J = 5.6, 3.0 Hz, 1H), 6.10 (dd, J = 5.6, 3.1 Hz, 1H), 3.69
(s, 3H), 3.04 (s, 1H), 2.90 (s, 1H), 2.23 (dd, J = 9.7, 5.0 Hz, 1H). 13C NMR (126 MHz, CDCl3) endo: δ 175.2, 137.7, 132.3, 51.5, 49.6, 45.6, 43.1, 42.5, 29.2. exo: δ 176.7, 138.0, 13578, 51.7, 46.5, 46.3, 42.9, 41.6, 30.3.
Bicyclo[2.2.1]hept-5-en-2-yl(pyrrolidin-1-yl)methanone (14a):
The product was obtained as a brown oil (11.9 mg, 0.041 mmol, 25%). 1H NMR (400 MHz, CDCl3) Endo: δ 6.16 (dd, J = 5.3, 3.0 Hz, 1H), 6.01 (dd, J = 5.3, 2.8 Hz, 1H), 3.55–3.47 (m, 2H), 3.42–3.33 (m, 2H), 3.11 (br s, 1H), 2.94 (m, 1H), 2.86 (br s, 1H), 1.96–1.89 (m, 3H), 1.84–1.75 (m, 2H), 1.40 (dd, J = 9.3, 4.7 Hz, 2H), 1.25 (d, J = 7.9 Hz, 1H). Exo (specific protons): δ 6.12 (dd, J = 5.4, 2.8 Hz, 1H), 6.09 (dd, J = 5.6, 2.6 Hz, 1H), 2.91 (s, 1H), 2.89 (s, 1H), 1.71 (d, J = 8.3 Hz, 1H), 1.33 (d, J =
8.8 Hz, 2H). 13C NMR (100 MHz, CDCl3) Endo: δ173.1, 136.6, 132.8, 49.6, 46.2, 45.9, 45.1, 43.4, 42.6, 30.4, 26.2, 24.3. Exo: δ 174.4, 138.2, 136.1, 46.5, 46.4, 46.0, 44.9, 43.5, 42.5, 29.7, 26.2, 24.1. IR (NaCl): 2969, 1717, 1633, 1431, 1339, 839, 703 cm-1.
1-Bicyclo[2.2.1]hept-5-ene-2-carbonyl)pyrrolidin-2-one (15a):20
The product was obtained as a colorless oil (50 mg, 0.244 mmol, 95%). 1H NMR (400 MHz, CDCl3) Endo: δ 6.19 (dd, J = 5.5, 3.0 Hz, 1H), 5.81 (dd, J = 5.5, 2.8 Hz, 1H), 3.95 (m, 1H), 3.75–3.63 (m, 2H), 3.22 (br s, 1H), 2.88 (br s, 1H), 2.58 (t, J = 8.1 Hz, 2H, endo), 2.02–1.94 (m, 2H), 1.88 (m, 1H), 1.48–1.36 (m, 3H). Exo (specific protons): δ 6.13 (dd, J = 6.5, 3.2 Hz, 1H), 3.81–3.76 (m, 2H), 2.93
164
(s, 1H), 1.35–1.26 (m, 3H). 13C NMR (100 MHz, CDCl3) Endo: δ 175.5, 175.0, 137.83, 131.8, 50.1, 46.1, 45.9, 44.6, 42.8, 34.0, 29.3, 17.2. Exo: 176.9, 175.0, 138.1, 136.1, 50.1, 46.6, 45.9, 44.4, 41.9, 3.0, 29.7, 17.2. IR (NaCl): 2970, 1723, 1689, 1460, 1387, 1225, 1044, 838, 695 cm-1.
3-Bicyclo[2.2.2]oct-5-ene-2-carbonyl)oxazolidin-2-one (16a):21
The product was obtained as a colorless solid (15.1 mg, 0.068 mmol, 20%). 1H NMR (500 MHz,
CDCl3) Endo: δ 6.34 (m, 1H), 6.16 (t, J = 7.3 Hz, 1H), 4.45–4.30 (m, 2H), 3.98 (t, J = 8.0 Hz, 2H), 3.79-3.71 (m, 1H), 2.84 (m, 1H), 2.62 (m, 1H), 1.84 (m, 1H), 1.71 (m, 1H), 1.64 (m, 1H), 1.53 (m, 1H), 1.30–1.24 (m, 2H). Exo (specific protons): δ 4.05 (m, 1H), 3.58 (m, 1H), 2.77 (m, 1H), 2.02 (m,
1H), 1.34–1.31 (m, 2H). 13C NMR (100 MHz, CDCl3) Endo: δ 175.6, 153.2, 135.0, 131.3, 61.91, 42.9, 42.00, 32.8, 30.2, 29.5, 25.7, 23.9. Exo (specific protons): δ 135.0, 134.0, 61.82, 43.1, 41.8, 32.3, 29.7, 29.4, 27.9, 25.00. IR (NaCl): 2941, 1777, 1699, 1386, 1220, 1041, 761, 695 cm-1. MP: 62–65 ° C.
Phenylmethyl ester:22
1H NMR (400 MHz, CDCl3) δ 7.36 (s, 5H), 5.11 (s, 2H), 2.11 (s, 3H). 13C NMR (100 MHz, CDCl3) δ 170.8, 136.0, 128.6, 128.3, 128.2, 66.3, 20.9. IR (ZnSe): 3065, 3033, 2951, 2892, 1732, 1497, 1455, 1379, 1222, 1023, 963, 919, 835, 734, 695 cm−1.
6.3.4 Copies of NMR spectra of the products of Diels-Alder reaction
1 13 2a+2a’ HNMR, C NMR CDCl3
165
166
1 13 3a+3a’ HNMR, C NMR CDCl3
167
1 13 4a+4a’ HNMR, C NMR CDCl3
168
1 13 6a+6a’ HNMR, C NMR CDCl3
169
1 13 7a+7a’ HNMR, C NMR CDCl3
170
1 13 8a+8a’ HNMR, C NMR CDCl3
171
1 13 9a+9a’ HNMR, C NMR CDCl3
172
1 13 10a+10a’ HNMR, C NMR CDCl3
173
1 13 12a+12a’ HNMR, C NMR CDCl3
174
1 13 13a+13a’ HNMR, C NMR CDCl3
175
1 13 14a+14a’ HNMR, C NMR CDCl3
176
1 13 15a+15a’ HNMR, C NMR CDCl3
177
1 13 16a+16a’ HNMR, C NMR CDCl3
178
1 13 Phenylmethyl ester H NMR C NMR gcosy HSQCAD in CDCl3
179
6.4 Experimental part of NHC-diol ligands
6.4.1 General procedure for NHC-diol ligand catalyzed carbonyl-ene reaction
1) Catalytic test of NHC-diols ligands with iron in carbonyl-ene reaction
In a flame-dried vial, Fe(OTf)2 (4.4 mg, 0.0125 mmol), ligand 6/7/9/10/11a/11b (0.06 mmol) and 4Å
MS (100 mg) was added. Freshly distilled CH2Cl2 was injected and the resulting solution was stirred for 0.5 h. The hydroxyl groups and imidazolium salts were expected to be deprotonated by using a base KOtBu. The amount of base was equivalent to summation of hydroxyls and imidazolium C−H proton(s). KOtBu-THF solution (1M) was injected accordingly into the solution and the solution was stirred for 3 hours. Then, the starting materials α-methyl styrene (30 mg, 0.25mmol) and ethyl 3,3,3- trifluoropyruvate (64 mg, 0.375 mmol) directly into the stirring solution. The vial was put under argon atmosphere and stirred for relative time (h). When the reaction was finished, directly loaded the sample onto silica flash chromatography by using eluent of hexane/ethyl acetate 97/3, and the product was dried under high vacuum. Then carbonyl-ene product was obtained and characterized by 1H NMR analysis and 13C NMR. Anhydrous Fe(SbF6)2 was prepared by following the same procedure of NHC-iron catalyzed carbonyl-ene reaction.
2) Catalytic test of a few C2 symmetric ligands with iron in carbonyl-ene reaction
Iron salt and C2 symmetric lignd were added into a flamed-dried vial, then injected freshly distilled
CH2Cl2. The sulotion was stirred for 2 hours. Then, Then, the starting materials α-methyl styrene (30 mg, 0.25mmol) and ethyl 3,3,3-trifluoropyruvate (64 mg, 0.375 mmol) directly into the stirring solution. The vial was put under argon atmosphere and stirred for relative time (h). When the reaction was finished, directly loaded the sample onto silica flash chromatography by using eluent of hexane/ethyl acetate 97/3, and the product was dried under high vacuum. Then carbonyl-ene products were obtained and characterized by 1H NMR analysis and 13C NMR. Anhydrous Fe(SbF6)2 was prepared by following the same procedure of NHC-iron catalyzed carbonyl-ene reaction.
180
5.4.2 Spectral data of NHC-diol family of ligand
(R)-tert-butyloxirane
1H NMR (400 MHz, CDCl3) δ 2.73 (dd, J = 4.1, 2.8 Hz, 1H), 2.64 (dd, J = 4.8, 4.1 Hz, 1H), 2.59 (dd, J = 4.8, 2.8 Hz, 1H), 0.91 (s, 9H).
(R)-tert-butyloxirane-2-mercaptobenzothiazole derivative
1H NMR (500 MHz, CDCl3) δ 7.86 (dp, J = 8.2, 0.6 Hz, 1H), 7.75 (ddt, J = 8.0, 1.2, 0.6 Hz, 1H), 7.46– 7.39 (m, 1H), 7.34–7.29 (m, 1H), 4.15 (brs, 1H), 3.65 (d, J = 9.6 Hz, 1H), 3.53 (ddd, J = 14.3, 2.0, 0.6 Hz, 1H), 3.33 (ddd, J = 14.3, 9.6, 0.6 Hz, 1H), 1.04 (d, J = 0.7 Hz, 9H). Condition: OD-H, Hex:iPrOH = 95:5, Flow : 1 mL/min, 230 nm. RT: 22.77, area: 0.32%; 28.76, area : 99.68%. ee = 99.3%.
Ligand 6a and 6b
Synthetic route: Added 1mmol 218.6 mg (1S,2S)-(−)-1,2-Diphenylethylenediamine into a flame-dried 10 mL flask and injected 3 mL anhydrous EtOH and 2.2 mmol 220 mg oxyrane, then set the reflux at 85 ℃ under argon for 2.5 days. After that, evaporated the solvent and purified it with
CH2Cl2/MeOH, then the diamine-diol was obtained with 95%. Then, put the diamine-diol into a 10 mL flame-dried flask and added 1.2 eq. triethyl orthoformate and 1.1 eq. ammonium tetrafluoroborate in 4 mL anhydrous EtOH. Reflux the solution at 85 ℃ for 3 days. After the reaction was finished, performed filtration through cotton-pluged pipet and evaporation and purification by CH2Cl2/MeOH. Imidazolium salt was obtained with 92% Yield.
181
6a
1H NMR (400 MHz, CDCl3) δ 7.21–7.14 (m, 6H, CHCHCHCHCHC), 7.08–7.01 (m, 4H,
CHCHCHCHCHC), 3.78 (s, 2H, CHN), 3.24 (d, J = 10.2 Hz, 2H, CH2CHOH), 2.69 (dd, J = 11.2 Hz,
2H, CHHCHOH), 2.33 (dd, J = 11.2 Hz, 10.2 Hz, 2H, CHHCHOH), 0.83 (s, 18H, CC3H9). 13C NMR
(100 MHz, CDCl3) δ 141.03 (NCHCCH), 128.00 (CCHCHCHCHCH), 127.71 (CCHCHCHCHCH),
127.01 (CCHCHCHCHCH), 78.40 (OHCHCH2), 69.98 (NCHCHN), 49.28 (CH2CHOH), 33.87
(CC3H9), 25.89 (CC3H9). ESI-MS, m/z = 412.30963, m/z (Cal.) = 412.30898, Diff.: 2ppm. HRMS: 281.01, 264.97, 211.02, 206.96, 190.93, 166.11, 150.81, 128.92, 115.02, 95.02, 77.03. IR (ZnSe): 3297, 2953, 2870, 1444, 1360, 1289, 1191, 1111, 1075, 1006, 906, 875, 760, 698 cm−1.
6b
1H NMR (500 MHz, CDCl3) δ 8.90 (s, 1H, NCHN), 7.48 (m, 6H, CHCHCHCHCHC), 7.38–7.34 (m,
4H, CHCHCHCHCHC), 5.16 (s, 2H, CHN), 3.51 (d, J = 10.6 Hz, 2H, CH2CHOH), 3.37 (dd, J = 14.2,
2.2 Hz, 2H, CHHCHOH), 3.26 (dd, J = 14.2, 10.6 Hz, 4H, CHHCHOH), 0.79 (s, 18H, CC3H9). 13C
NMR (100 MHz, CDCl3) δ 157.6 (NCHN), 134.7 (NCHCCH), 130.1 (CCHCHCHCHCH), 129.7
(CCHCHCHCHCH), 128.4 (CCHCHCHCHCH), 78.2 (OHCHCH2), 75.6 (NCHCHN), 48.1
(CH2CHOH), 34.3 (CC3H9), 25.3 (CC3H9). ESI-MS, m/z = 423.30083, m/z (Cal.) = 423.30060, Diff.: 1ppm. HRMS: 248.94, 206.96, 121.07, 105.04, 91.07, 77.03. IR (ZnSe): 3526, 2957, 2868, 1637, 1457, 1312, 1203, 1054, 775, 700 cm−1. Mp: 114± 5 ℃.
182
Ligand 7a and 7b
Synthetic route: Added 1mmol 218.6 mg cyclohexane-(1R,2R)-1,2-trans-diamine into a flame-dried 10 mL flask and injected 3 mL anhydrous EtOH and 2.2 mmol 220 mg oxyrane, then set the reflux at 85 ℃ under argon for 2.5 days. After that, evaporated the solvent and purified it with
CH2Cl2/MeOH, then the diamine-diol was obtained with 95%. Then, put the diamine-diol into a 10 mL flame-dried flask and added 1.2 eq. triethyl orthoformate and 1.1 eq. ammonium tetrafluoroborate in 4 mL anhydrous EtOH. Reflux the solution at 85 ℃ for 3 days. After the reaction was finished, performed filtration through cotton-pluged pipet and evaporation and purification by CH2Cl2/MeOH. Imidazolium salt was obtained with 92% Yield.
7a
Mp: 76± 3 ℃. 1H NMR (400 MHz, CDCl3) δ 3.66 (br, 2H, OH), 3.27 (d, J = 13.7 Hz, 2H, CHOH), 2.68 (dd, J = 12.1, 2.6 Hz, 2H, OHCHCHHN), 2.52 (t, J = 11.3 Hz, 2H, OHCHCHHN), 2.23 (br, 2H,
NHCHCH2), 2.09 (d, J = 15.7 Hz, 2H, NHCHCHHCHH), 1.73 (m, 2H, NHCHCHHCHH), 1.23 (m, 2H,
NHCHCHHCHH), 1.04 (m, 2H, NHCHCHHCHH), 0.91 (s, 18H, CC3H9). 13C NMR (100 MHz, CDCl3)
δ 76.3 (OHCHCH2), 59.4 (NHCHCH2), 46.3 (OHCHCH2), 33.9 (CC3H9), 31.2 (NHCHCH2CH2), 25.9
(CC3H9), 25.0 (NHCHCH2CH2). IR (ZnSe): 3297, 2951, 2931, 2859, 1450, 1361, 1244, 1085, 1012, 921, 907, 730 cm−1. HRMS (ESI-MS), m/z = 314.29388, m/z (Cal.) = 314.29333, Diff.: 2 ppm. EI-MS: 281.01, 206.96, 190.93, 164.98, 133.05, 129.06, 103.04, 73.04.
183
7b
Mp: 160± 5 ℃.1H NMR (400 MHz, CDCl3) δ 8.32 (s, 1H, NCHN), 3.54–3.44 (m, 6H,
OHCHCH2NCHCH), 3.34 (s, 2H, CHOH), 3.14 (s, 1H, OH), 2.20–2.10 (m, 2H, NHCHCHHCHH), 1.98 (d, J = 9.5 Hz, 2H, NHCHCHHCHH), 1.61 (s, 2H, NHCHCHHCHH), 1.41 (t, J = 10.1 Hz, 2H,
NHCHCHHCHH), 0.95 (s, 18H, CC3H9). 13C NMR (100 MHz, CDCl3) δ 161.8 (NCHN, 73.1
(OHCHCH2), 65.4 (NCHCH2), 47.9 (OHCHCH2), 34.2 (CC3H9), 26.55 (NHCHCH2CH2), 25.60
(CC3H9), 23.58 (NHCHCH2CH2). F: 150.0. IR (ZnSe): 3443, 3346, 3093, 2856, 2871, 1608, 146, 1365, 1205, 1079, 986, 960, 858, 763 cm−1. HRMS (ESI-MS), m/z = 325.28638, m/z (Cal.) = 325.28495, Diff.: 4 ppm. EI-MS: 280.94, 267.17, 237.16, 207.03, 167.04, 139.10, 112.09, 96.12, 81.09.
Ligand 8a
1(R)-tertButyl-2-imidazolyl-1-ethanol
Synthesis route: Added imidazole (2.2 mmol, 150 mg) into a flame-dried 5 ml flask, injected 1-(R)- tButyl-oxirane (2 mmol, 200 mg) and anhydrous ethanol 2.5 ml. The solution was refluxed at 85 ℃ for 1 day. When the reaction was stopped, cooled down it to room temperature, evaporated the solvent and purified the crude product by chromatography CH2Cl2/MeOH (97/3) and obtained white solid 72% and ligand 8b.
Mp: 85± 3 ℃. 1H NMR (400 MHz, CDCl3) δ 7.25 (s, 1H, NCHN), 6.83 (d, J = 1.3 Hz, 1H,NCHCHN), 6.74 (d, J = 1.3 Hz, 1H, NCHCHN), 5.94 (s, 1H,OH), 3.99 (dd, J = 13.8, 1.8 Hz, 1H,NCHHCH), 3.70
(dd, J = 13.8, 10.0 Hz, 1H, NCHHCH), 3.31 (dd, J = 10.0, 1.8 Hz, 1H, NCH2CHOH), 0.97 (s, 9H,
184
C3H9). 13C NMR (100 MHz, CDCl3) δ 137.4 (NCN), 128.1 (NCHCHNCH2), 119.5 (NCHCHNCH2),
78.3 (CH2CHOHC4H9), 50.0 (NCH2CHOHC4H9), 34.4 (CHCC3H9), 25.9 (CC3H9). IR (ZnSe): 3421, 3140, 3111, 2953, 2867, 2748, 1573, 1517, 1477, 1393, 1364, 1332, 1284, 1234, 1202, 1171, 1084, 1022, 918, 838, 761, 658 cm−1. HRMS (ESI-MS): m/z = 169.13321 (168.12594), m/z (Cal.) = 168.12626, Diff.: 1.9. EI-MS: 168.04, 141.10, 126.06, 81.09, 69.11.
Ligand 8b
Synthesis route: Added imidazole (1 mmol, 68 mg) into a flame-dried 15 ml Schlenk tube, injected 1(R)-tButyl-oxirane (3.2 mmol, 320 mg) and anhydrous ethanol 3 ml. The solution was stirring at 83 ℃ for 2 days. When the reaction was stopped, cooled down it to room temperature, evaporated the solvent and purified the crude product by chromatography CH2Cl2/MeOH (97/3) and obtained white solid 72% 8a and 15% 8b.
Mp: 235± 5 ℃. 1H NMR (400 MHz, DMSO-d6) δ 9.11 (s, 1H, NCHN), 7.69 (d, J = 1.5 Hz, 2H, NCHCHN), 5.16 (d, J = 6.2 Hz, 2H, CCHOHCH2), 4.29 (dd, J = 13.6, 2.2 Hz, 2H, CHCHH), 3.89 (dd,
J = 13.6, 10.5 Hz, 2H, CHCHH), 0.88 (s, 18H, CC3H9). 13C NMR (100 MHz, DMSO-d6) δ 137.4
(NCHN), 123.04 (NCHCHN), 76.71 (CHOHCH2), 51.87 (CHOHCH2), 34.70 (CC3H9), 26.12 (CC3H9). IR (ZnSe): 3342, 3251, 3139, 3083, 2954, 2873, 2722, 1652, 1656, 1479, 1427, 1362, 1327, 1164, 1083, 1021, 920, 849, 766 cm−1. HRMS (ESI-MS), m/z = 268.21553, m/z (Cal.) = 268.21508, Diff.: 2ppm.
Ligand 9
185
Synthesis route: Added 1(R)-tert-Butyl-2-imidazolyl-1-ethanol (1 mmol 168 mg) and 4-tert-Butyl- benzyl bromide (1 mmol 228 mg) in a flame-dried 10 ml Schlenk tube, injected 3 ml dry MeCN. Heat the solution at 83 ℃ under argon for 2 days. After the reaction was finished, cooled it down to room temperature and there was white precipitate. Decanted the solution and washed the rest solution with dry MeCN 6 ml, then using another 5 ml MeCN to transfer the precipitate to a vial. Evaporated the solvent and achieved white powder at 95% yield.
Mp: 242± 5 ℃. 1H NMR (400 MHz, CDCl3) δ 9.79 (s, 1H, NCHN), 7.55 (t, J = 1.8 Hz, 1H,
CHCH2NCHCH NCH2C), 7.36–7.27 (m, 4H, CH2CCHCHCHCHC), 7.24 (d, J = 1.8 Hz, 1H,
CHCH2NCHCHNCH2 C), 5.43–5.28 (m, 2H, CH2(benzyl)), 4.81 (d, J = 7.3 Hz, 1H, OH), 4.32 (dd, J
= 13.7, 2.1 Hz, 1H, OHCHCHHN), 4.13 (m, 1H, OHCHCHHN), 3.57 (t, J = 8.9 Hz, 1H, OHCHCH2),
1.20 (m, 1H, C6H4C4H9), 0.90 (s, 9H, CHC4H9). 13C NMR (100 MHz, CDCl3) δ 152.46
(CH2CCHCHCHCHC), 136.58 (NCN), 129.81 (CCHCHCHCHCC4H9), 128.79
(CCHCHCHCHCC4H9), 126.24 (CCHCHCHCHCC4H9), 123.43 (OHCHCH2NCHCHN), 121.31
(OHCHCH2NCHCHN), 76.79 (OHCHCH2), 53.03 (NCH2C), 52.28 (OHCHCH2N), 34.61 (CCH3,
CCH3), 31.15 (CCC3H9), 26.03 (OHCHCC3H9). IR (ZnSe): 3280, 3058, 2955, 2866, 1554, 1475, 1406, 1329, 1285, 1155, 1080, 1016, 844, 726, 686 cm−1. HRMS (ESI-MS), m/z = 315.24171, m/z (Cal.) = 315.24309, Diff.: 4ppm. EI-MS: 143.03, 129.06, 101.05, 85.08, 77.03.
Ligand 10
Synthesis route: added 1(R)-tertButyl-2-imidazolyl-1-ethanol (2 mmol 336 mg) and 1.2-dibromo- methane (1 mmol 187 mg) in a flame-dried 5 ml flask, injected 3 mL dry MeCN. Refluxed the solution while stirring at 90 ℃ under argon for 1.5 day. After the reaction was finished, there was white precipitate, cooled it down to room temperature. Transferred the precipitate and the solution together to a vial. The solvent was evaporated, and it afforded white powder for 95% yield.
186
Mp: 350± 5 ℃. 1H NMR (400 MHz, DMSO-d6) δ 9.12 (s, 2H, NCHN), 7.78 (s, 2H, NCHCHNCH2CH2),
7.66 (s, 2H, NCHCHNCH2CH2), 5.11 (d, J = 6.4 Hz, 2H, OH), 4.79–4.62 (m, 4H, NCH2CH2N), 4.31 (d, J = 13.5 Hz, 2H, NCHHCHOH), 3.87 (dd, J = 13.5, 10.4 Hz, 2H, NCHHCHOH), 3.33 (d, J = 10.4
Hz, 2H, NCH2CHOH), 0.87 (s, 18H, CC3H9). 13C NMR (100 MHz, DMSO-d6) δ 137.46(NCHN),
123.96 (NCHCHNCH2CH2), 122.48 (NCH CHNCH2CH2), 76.36 (CH2CHOH), 52.09 (NCH2CHOH),
48.79 (NCH2CH2N), 34.68 (CC3H9), 26.14 (CC3H9). HRMS (ESI-MS), m/z: 182.14096, Cal. m/z: 182.14136, Diff.: 2ppm. EI-MS: 143.03, 129.06, 101.05, 75.04. IR (ZnSe): 2399, 2348, 3062, 2953, 2869, 1675, 1616, 1560, 1476, 1446, 1360, 1164, 1082, 1018, 851, 773, 730 cm−1.
Ligand 11a, 11b and 11c
Synthesis route: Added 1(R)-tertButyl-2-imidazolyl-1-ethanol (2 mmol 336 mg) and 2,6- Bis(bromomethyl)-pyridine (1 mmol 265 mg) in a flame-dried 25 ml schlenk tube, injected 5 ml dry MeCN. Heat the solution at 83 ℃ under argon for overnight. After the reaction was finished, there was white precipitate, cooled it down to room temperature. Decanted the solution and washed the rest solution with dry MeCN 6 ml, then using another 5 ml MeCN to transfer the precipitate to another vial. Evaporated the solvent and obtained white powder at 97% yield.
Py-bisNHC-diol-2Br 11a
Mp: 280± 5 ℃. 1H NMR (400 MHz, D2O) δ 8.72 (s, 2H, NCHN), 7.78 (t, J = 7.8 Hz, 1H,
CCHCHCHCN), 7.39 (d, J = 2.0 Hz, 2H, OHCHCH2NCHCHN), 7.31 (d, J = 2.0 Hz, 2H,
OHCHCH2NCHCHN), 7.29 (d, J = 7.8Hz, 2H, CH2CCHCH), 5.37 (s, 4H, NCH2CN), 4.32 (dd, J =
187
14.1, 2.3 Hz, 2H, OHCHCHHN), 3.92 (dd, J = 14.1, 10.8 Hz, 2H, OHCHCHHN), 3.42 (dd, J = 10.8,
2.3 Hz, 2H, OHCHCH2), 0.81 (s, 18H, CC3H9). 13C NMR (100 MHz, D2O) δ 153.1 (CH2CN), 139.8
(p-C of pyridine), 136.3 (m, NCHN, NCDN), 123.0 (CHCH2NCHCHN), 122.8 (CHCH2NCHCHN),
122.7 (CCHCHCHCN), 77.5 (OHCHCH2), 53.3 (NCH2CN), 51.7 (OHCHCH2N), 33.7 (C3H9CCH),
24.8 (CC3H9).
1H NMR (400 MHz, DMSO-d6) δ 9.12 (t, J = 1.6 Hz, 2H, NCHN), 7.93 (t, J = 7.8 Hz, 1H,
CCHCHCHCN), 7.75 (t, J = 1.8 Hz, 2H, OHCHCH2NCHCHN), 7.64 (t, J = 1.8 Hz, 2H,
OHCHCH2NCHCHN), 7.42 (d, J = 7.8 Hz, 2H, CH2CCHCH), 5.56–5.42 (m, 4H, NCH2CN), 5.22 (d, J = 6.0 Hz, 2H, OH), 4.34 (dd, J = 13.5, 2.2 Hz, 2H, OHCHCHHN), 3.92 (dd, J = 13.5, 10.6 Hz, 2H,
OHCHCHHN), 3.33 (ddd, J = 10.6, 6.0, 2.2 Hz, 2H, OHCHCH2), 0.90 (s, 18H, CC3H9). 13C NMR (100
MHz, DMSO-d6) δ 154.10 (CH2CN), 139.36 (p-C of pyridine), 137.75 (NCHN), 123.36
(CHCH2NCHCHN), 123.18 (CHCH2NCHCHN), 122.51 (CCHCHCHCN), 76.60 (OHCHCH2), 53.02
(NCH2CN), 52.00 (OHCHCH2N), 34.72 (C3H9CCH), 26.10 (CC3H9). IR (ZnSe): 3299, 3248, 3141, 3062, 2953, 2869, 1698, 1616, 1560, 1446, 1360, 1164, 1082, 1018, 918, 773, 730 cm−1. HRMS (ESI-MS), m/z = 220.65536, m/z (Cal.) = 220.65464, Diff.: 3ppm. EI-MS: 280.94, 266.90, 252.93, 206.96, 192.92, 164.98, 147.02, 132.98, 91.07, 72.97.
Py-bisNHC-diol-2PF6 11b
Mp: 144± 5 ℃.1H NMR (400 MHz, DMSO-d6) δ 9.12 (s, 2H, NCHN), 7.93 (t, J = 7.8 Hz, 1H,
CCHCHCHCN), 7.75 (s, 2H, OHCHCH2NCHCHN), 7.64 (s, 2H, OHCHCH2NCHCHN), 7.42 (d, J =
7.8 Hz, 2H, CH2CCHCHCHC), 5.57–5.42 (m, 4H, NCH2CN), 5.22 (d, J = 6.0 Hz, 2H, OH), 4.34 (dd, J = 13.4, 2.1 Hz, 2H, OHCHCHHN), 3.92 (dd, J = 13.6, 10.6 Hz, 2H, OHCHCHHN), 3.33 (ddd, J =
10.6, 6.0, 2.2 Hz, 2H, OHCHCH2), 0.90 (s, 18H, CC3H9). 13C NMR (100 MHz, DMSO-d6) δ 154.10
(CH2CN), 139.36 (p-C(pyridine)), 137.75 (NCHN), 123.36 (CHCH2NCHCHN), 123.18
(CHCH2NCHCHN), 122.51 (CCHCHCHCN), 76.60 (OHCHCH2), 53.02 (NCH2CN), 52.00
(OHCHCH2N), 34.72 (C3H9CCH), 26.10 (CC3H9). IR (ZnSe): 3589, 3162, 2964, 2867, 1697, 1563,
188
1446, 1366, 1164, 1087, 998, 828,748 cm−1. HRMS (ESI-MS), m/z = 220.65536, m/z (Cal.) = 220.65464, Diff.: 3ppm. EI-MS: 280.94, 210.95, 206.96, 190.99, 162.85, 147.02, 128.99, 119.01, 89.07, 73.04.
Py-bisNHC-diol-2Br-Ag2O complex 11c
1H NMR (400 MHz, DMSO-d6) δ 7.81 (t, J = 7.7 Hz, 1H), 7.48 (d, J = 1.7 Hz, 2H), 7.39 (d, J = 1.7 Hz, 2H), 7.35 (d, J = 7.7 Hz, 2H), 5.49 (d, J = 14.5 Hz, 2H), 5.32 (d, J = 14.5 Hz, 2H), 4.91 (s, 2H), 3.97 (s, 2H), 3.21 (d, J = 10.1 Hz, 2H), 0.66 (s, 18H).
189
5.4.3 Copies of NMR spectra of NHC-diol ligands
(R)-tert-butyloxirane HNMR CDCl3
190
1 H NMR and HPLC of (R)-tert-butyloxirane 2-mercaptobenzothiazole derivative CDCl3
191
1 13 Ligand 6a H NMR C NMR gcosy HSQCAD in DMSO d6
192
193
1 13 Ligand 6b H NMR C NMR gcosy HSQCAD in CDCl3
194
195
1 13 Ligand 7a H NMR C NMR gcosy HSQCAD in DMSO d6
196
197
1 13 Ligand 7b H NMR C NMR gcosy HSQCAD in CDCl3
198
199
1 13 Ligand 8a H NMR C NMR gcosy HSQCAD CDCl3
200
.
201
1 13 Ligand 8b H NMR C NMR gcosy HSQCAD in DMSO d6
202
203
1 13 Ligand 9 H NMR C NMR gcosy HSQCAD CDCl3
204
205
1 13 Ligand 10 H NMR C NMR gcosy HSQCAD in DMSO d6
206
207
1 13 Ligand 11a H NMR C NMR gcosy HSQCAD in D2O
208
209
1 13 Ligand 11a H NMR C NMR gcosy HSQCAD in DMSO d6
210
211
1 13 Ligand 11b H NMR C NMR gcosy HSQCAD in DMSO d6
212
213
1 11c H NMR of complex of ligand 11a and Ag2O
214
6.5 References
[1] (a) Terent'ev, O. A.; Mulina, M. O.; Pirgach, A. D.; Demchuk, V. D.; Syroeshkin, A. M.; Nikishin. I. G. RSC Adv. 2016, 6, 93476. (b) Ohsugi, S.; Nishide, k.; Node, M. Tetrahedron 2003, 59, 1859.
[2] (a) Luo, F.-T.; Lo, H.-K. J. Organomet. Chem. 2011, 696, 1262. (b) Ivanov, E. I.; Kalayanov, G. D.; Yaroshchenko, I. M.; Stepanov D. E. Khim. Geterotsikl. Soedin. 1989, 11, 1570. (c) Kascatan- Nebioglu, A.; Melaiye, A.; Hindi, K.; Durmus, S.; Youngs, W. J. Med. Chem. 2006, 49, 6811.
[3] (c) Pinto, M. A. R.; Sslvador, A. R. J.; Le Roux, C. Catal. Comm. 2008, 9, 465.
[4] Kascatan-Nebioglu, A.; Panzner, J. M.; Carrison, C. Jered.; Tessier, A. C.; Youngs, J. W. Organometallics 2004, 23, 1982.
[5] Gardner, S.; Kawamoto, T.; Curran, P. D. J. Org. Chem. 2015, 80, 9794.
[6] Finotello, A.; Bara, J. E.; Camper, D.; Noble, D. R. Ind. Eng, Chem. Res. 2008, 47, 3453.
[7] (a) Luo, H.-K.; Woo, Y.-L.; Schumann, H.; Jacob, C.; Meurs, M. V.; Yang, H.-Y.; Tan. Y.-T. Adv. Synth. Catal. 2010, 352, 1356. (b) Rueping, M.; Bootwicha, T.; Kambutong, S.; Sugiono, E. Chem. Asian. J. 2012, 7, 1195.
[8] Rueping, M.; Bootwicha, T.; Kambutong, S.; Sugiono, E. Chem. Asian. J. 2012, 7, 1195.
[9] Rueping, M.; Theissmann, T.; Kuenkel, A.; Koenigs, R. M. Angew. Chem. Int. Ed. 2008, 47, 6798.
[10] Zheng, K.; Lin, L.-L.; Feng, X.-M. Acta Chim. Sinica 2012, 70, 1785.
[11] Zheng, K.; Yang, Y.; Zhao, J.; Yin, C.; Lin, L.; Liu, X.; Feng, X. Chem. Eur. J. 2010, 16, 9969.
[12] Clarke, M. L.; Jones, C. E. S.; France, M. B. Beilstein J. Org. Chem. 2007, 3, 24.
[13] (a) Kropp. P. J.; Breton, G. W.; Graig. S. L. Crawford, S. D.; Durland, W. F., Jr.; Jones, J. E., III; Raleigh, J. S. J. Org. Chem. 1995, 60, 4146. (b) Silva, K. A.; Robles-Deutenhefner, P. A.; Sousa, E. M. B.; Kozhevnikova, E. F.; Kozhevnikov, I. V.; Gusevskaya, E. V. Catal. Commun. 2004, 5, 425.
[14] Kanemasa, S.; Oderaotoshi, Y.; Sakaguchi, S.; Yamamoto, H.; Tanaka, J.; Wada, E.; Curran, D. P. J. Am. Chem. Soc. 1998, 120, 3074.
[15] Li, Y.; Wang, C.; Hao, J.; Cheng, M.; Jia, G.; Li, C. Chem. Comm. 2015, 51, 13174.
215
[16] Barroso, S.; Blay, G.; Pedro, J. R. Org. Lett. 2007, 9, 1983.
[17] Singh, R. S.; Adachi, S.; Tanaka, F.; Yamauchi, T.; Inui, C.; Harada, T. J. Org. Chem. 2008, 73, 212.
[18] (a) Mukhina, O. A.; Kumar, N. N. B.; Arisco, T. M.; Valiulin, R. A.; Metzel, G. A.; Kutateladze, A. G. Angew. Chem. Int. Ed. 2011, 50, 9423. (b) Ripoll, J.-L.; Lasne, M.-C. Tetrahedron Lett. 1978, 19, 5201.
[19] Matuszek, K.; Coffie, S.; Chrobok, A.; Swadźba-Kwaśny, S. Catal. Sci. Technol. 2017, 1045.
[20] Sibi, M. P.; Chen, J.; Stanley, L. Synlett 2007, 0298.
[21] Evans, D. A.; Barnes, D. M.; Johnson, J. S.; Lectka, T.; Matt, P. V.; Miller S. J.; Murry, J. A.; Norcross, R. D.; Shaughnessy, E. A.; K. R. J. Am. Chem. Soc. 1999, 121, 7582.
[22] (a) Janza, B.; Studer, A. J. Org. Chem. 2005, 70, 6991. (b) Glover, S. A.; Mo, G. J. Chem. Soc. Perkin Trans. 2 2002, 10, 1728.
216