Improved Methodology for the Preparation of Chiral

(Important Chiral Building Blocks in Pharmaceutical Drugs and Natural Products Synthesis)

Mohamed Mahmoud El-Shazly

A thesis submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy in Organic Synthetic Chemistry

Approved Thesis Committee Prof. Dr. Thomas Nugent Professor of Organic Chemistry Jacobs University Bremen Prof. Dr. Nikolai Kuhnert Professor of Organic Chemistry Jacobs University Bremen Dr. Pralhad Ganeshpure Indian Petrochemicals Corporation Limited, India Date of Defense: August 03, 2009

School of Engineering and Science, Jacobs University, Bremen, Germany.

Declaration

I herewith declare that this thesis is my own work and that I have used only the sources listed. No part of this thesis has been accepted or is currently being submitted for the conferral of any degree at this university or elsewhere.

Mohamed El-Shazly Bremen

This dissertation is dedicated to all those people who have always given me the love, trust, and support to come to this stage of my life

-To My Family-

Abstract

The importance of α-chiral amines as building blocks in pharmaceutical drugs, natural products, fine chemicals and agrochemicals have encouraged scientists to develop different methodologies for their preparation. Their main goal was to develop a step wise efficient and low waste production methodology which utilizes inexpensive starting material for the synthesis of α-chiral amines in high yields and enantioselectivity. Different methodologies have been developed aiming to meet these criteria. These strategies are discussed and their importance and limitations are critically analyzed.

Reductive amination is a powerful methodology for the synthesis of chiral amines in high yields and enantioselectivity. It is a two step strategy beginning from the prochiral carbonyl compound to the primary chiral . The historical development and the latest milestones in this field are discussed in chapter three. Different drugs and natural products which are prepared utilizing reductive amination as a key step in their synthesis are summarized in chapter four.

Reductive amination utilizing chiral auxiliary/Lewis acid/ heterogeneous catalyst/ molecular hydrogen has been investigated in our group over the last five years. This combination allowed the preparation of alkyl-alkyl’ α-chiral amines in mediocre to good yields and enantioselectivities. This group of amines is known historically to be difficult synthetic task.

We developed a new asymmetric reductive amination procedure using Yb(OAc)3 (50-110 mol %) that allows increased diastereoselectivity (6-15% units) for alkyl-alkyl’ α-chiral amines that previously only provided mediocre to good diastereoselectivity. Different Lewis acids were tested under different reaction conditions of temperature, pressure and solvents and the results of these experiments are discussed in chapter five.

O (S,S)-2d HN Ph NH 1d Yb(OAc)3,MeOH-THF Pd-C (S)-3d 2 + H2N Ph Raney-Ni, H2 (120 psi) H2 (60 psi) (S)-α-MBA 86% de 85% ee The use of catalytic Lewis acids in reductive amination has never been reported in literatures.

We demonstrated the beneficial use of 10-15 mol % of Yb(OAc)3 or Ce(OAc)3 or Y(OAc)3 in

i

suppressing formation and promoting reductive amination in good yield but without enhanced . Despite the fact that the use of Brønsted acids in reductive amination is well established no literature reports are available. We have performed and extensive study on the use of commercially available Brønsted and mineral acids in reductive amination. The scope of the reaction and the substrate categories are summarized in chapter six.

A mechanism for the reaction has been proposed and the basic mechanistic experiments have been performed. An in situ cis- to trans-ketimine isomerization mechanism, promoted by

Yb(OAc)3, has been proposed to account for the observed increase in diastereoselectivity. The experiments and the proposed mechanism are summarized in chapter seven

ii

Acknowledgement

All the work reported in this thesis have been carried out at the Department of Chemistry, School of Engineering and Science, Jacobs University, Bremen, Germany since joining here on August 2006 till August 2009. I would like to thank Jacobs University for the financial support and all the laboratory facilities during my stay here. In this regard I would like to thank Prof. Dr. h. c. Bernhard Kramer for approving my PhD scholarship.

I would like to convey my kind regards to my supervisor Prof. Thomas C. Nugent and thank him for all his kind suggestions and deeply appreciate his skillful guidance throughout my research. It was due to his relentless efforts that I could master the various techniques and learn to solve the different scientific challenges that came by my way. Lastly, I would also acknowledge his patience and kind understanding.

I would thank Prof. Nikolai Kuhnert for his kind consent to become the internal examiner of this thesis.

I would also thank Dr. Pralhad Ganeshpure, Research Centre, Indian Petrochemicals Corporation Limited, 391 346 Vadodara, India (B-21, Kinnari Duplex Ellora Park, Vadodara, Gujarat 390023, India) for his kind consent to become the external examiner of this thesis.

My sincere appreciation goes to all my lab mates, Dr. Rashmi R. Mohanty, Dr. Vijay N. Wakchaure, Dr. Abhijit Ghosh, Ahson J. Shaikh, Mohammad Naveed Umar, Mohammad Shoaib, A. Alvaradomendez, Abdul Sadiq, Dan Hu, Ahtaram Bibi, Satish Wakchaure, Andrei Dragan, Andrei Iosub and Daniela Negru for their constant help and encouragement in all respect. I would also thank Mrs. Müller for her continuous help.

All my deepest veneration goes to my parents for everything that they have given to me. I would convey my regards to my sister and all my uncles and aunts for their constant support. I would also thank all professors and colleagues in Egypt. Especially I would like to thank iii

Prof. Mohamed El-Azizi, Prof. Abdel-Nasser Singab and Prof. Nahla Ayoub for their support and help over the past years.

I would like to thank all friends at Jacobs University, Iyad Tumar, Khaled Hassan, Dr. Raed Mesleh, Mohamed Noor, Hamdy El-Sheshtawy, Salahaldin Juba, Ahmed Moussa, Ahmed El- Moasry, Hany Elgala and all other friends in Germany and Egypt for their continuous support

,

,

iv

Abbreviations

Ac Acetyl AcOH Acetic acid aq. Aqueous Ar Aryl bs Broad singlet (1H-NMR) BINOL 1,1'-Bi-2-naphthol BINAP 2,2'-Bis(diphenylphosphino)-1,1'-binaphthyl. BOC tert-Butyl carbamates iBu iso-Butyl nBu n-Butyl conv. Conversion cat. Catalyst

CDCl3 Deuterated chloroform COD Cycloctadiene d Doublet (1H-NMR) dd Doublet of doublet (1H-NMR) DCM Dichloromethane de Diastereomeric excess DIBAL-H Diisobutyl aluminium hydride DME 1,2-Dimethoxyethane DMF N,N’-Dimethylfomamide DMSO Dimethylsulfoxide δ Chemical shift (1H-NMR) ee equiv. Equivalent ESI Electron spray ionization (Mass spectroscopy)

v

Et Ethyl EtOH Ethanol EtOAc Ethylacetate GC Gas chromatography h Hours HPLC High performance liquid chromatography HRMS High resolution mass spectrometry Hz Hertz J Coupling constant (1H-NMR) KHMDS Potassium hexamethyldisilazide LDA m Multiplate (1H-NMR) M Molar MBA Methyl Benzyl Amine Me Methyl min. Minutes MS Molecular sieves MS Mass spectroscopy MTBE Methyl-tert-butyl ether MW Molecular weight m/z Mass/charge m Meta NaOtBu Sodium tert-butoxide NBD N-Bornadiene NMR Nuclear Magentic Resonance o Ortho p Para Pd-C Palladium on carbon Ph Phenyl iPr iso-Propyl nPr n-Propyl Pt-C Platinum on carbon pyr Pyridine vi

q Quartet (1H-NMR) Raney-Ni Raney-Nickel Ref. Reference Rh-C Rhodium on carbon s Singlet (1H-NMR) t Triplet (1H-NMR) t-Bu tert-Butyl tert Tertiary temp Temperature TFA Trifluoroacetic acid THF Tetrahydrofuran TLC Thin layer chromatography TMS Trimethylsilane Ts Tosyl TsOH p-Toluenesulfonic acid tBuLi tert-Butyllithium i Ti(O Pr)4 Titanium(IV) isopropoxide

vii

Table of Contents

Abstract. i Acknowledgment. ii List of Abbreviations. v

1. Introduction to 1.1. Chiral Drugs 1 1.2. Isomers and Isomerism 2 1.3. Nature is Chiral 3 1.4. Chirality and Drug-Receptor Interaction 8 1.5. Sources of Enantiopure Substances 8 1.5.1. Synthesis of Enantiomerically Pure Compounds 9 1.5.2. Resolution 9 1.5.2.1. Preferential 9 1.5.2.1. Crystallization 10 1.5.2.2. 11 1.5.3. Approach 12 1.5.4. Stereoselective Conversion of Prochiral Substrates to Enantiopure Compounds (Asymmetric Synthesis) 15 1.5.5. Asymmetric Synthesis vs Kinetic Resolution vs Chiral Pool 18 1.6. α-Chiral Amines Defining Terms 19 1.7. α-Chiral Amines Importance 20 1.8. α-Chiral Amine Synthesis Different Methodologies 22 1.8.1. Imine and Enamide Synthesis 23 1.8.2. Enantioselective Reduction of Enamides 23 1.9. Conclusion 28 1.10. References 28

2. Imine Reduction 2.1. Historical View 34 2.2. Asymmetric Reduction of N-Phosphinoyl Imines 35 2.2.1. Synthesis of N-Phosphinoyl Imines 35

viii

2.2.2. Different Substrates Categories 36 2.2.3. Nguyen Special Substrates 39 2.3. Asymmetric Reduction of N-aryl imines 40 2.3.1. Synthesis of N-Aryl Imines 40 2.3.2. Different Substrates Categories 43 2.4. Reduction of Miscellaneous Imines 49 2.5 Conclusion 49 2.6. References 50

3. Reductive Amination 3.1. Historical View 53 3.1.1. Reductive Amination Utilizing Heterogeneous Catalyst 53 3.1.2. Reductive Amination Utilizing Homogenous Catalysis 55 3.2. Reductive Amination the Current State of Art 57 3.3. Asymmetric Reductive Amination 60 3.3.1. Asymmetric Reductive Amination Utilizing Chiral Catalysts 60 3.3.2. Reductive Amination Utilizing Chiral Auxiliary 64 3.3.3. Reductive Amination Utilizing Molecular Hydrogen 65 3.3.4. Asymmetric Reductive Amination Utilizing Transfer Hydrogenation Conditions 66 3.4.5. Organocatalytic Asymmetric Reductive Amination 67 3.4. Green Chemistry and Reductive Amination 73 3.4.1. Green Chemistry Basic principle 73 3.4.2. Hydrogenation and Green Chemistry 75 3.5. Conclusion 76 3.6. References 76

4. Drugs and Reductive Amination 4.1 Reductive Amination in the Synthesis of Drugs and Natural Products 81 4.1.1. Synthesis of Delavirdine 81 4.1.2. Synthesis of Muraglitazar 82 4.1.3. Synthesis of 83 4.1.4. Synthesis of Sertraline 84 ix

4.1.5. Synthesis of Emitine 85 4.1.6. Synthesis of Taltobulin 86 4.1.7. Synthesis of Perzinfote 87 4.1.8. Synthesis of Namindinil 88 4.1.9. Synthesis of Ezlopipant 89 4.1.10.Synthesis of Monomorine 90 4.1.11. Synthesis of Ontazolast 91 4.1.12. Synthesis of Pamaquine 92 4.1.13. Synthesis of Torcetrapib 92 4.1.14. Synthesis of Polyaminocholestanol Derivatives 93 4.1.15. Synthesis of piperazinylpropylisoxazoline Analogues 94 4.1.16. Synthesis of Ritonavir and Lopinavir 95 4.1.17. Synthesis of Tetrahydrocarbazoles 95 4.2. Conclusion 97 4.3. References 97

5. Stoichiometric Use of Ytterbium Acetate in Reductive Amination. 5.1. Introduction 98 5.1.1. Ytterbium 100 5.1.1.1. Electronic Overview 100 5.1.1.2. Ytterbium Discovery 101 5.1.1.3. Ytterbium Reactions 101 5.1.1.4. Ytterbium Acetate 102

5.1.2. Commercial Yb(OAc)3 vs Dried Yb(OAc)3 110 5.1.3. Optimized Conditions and Useful Substrate Range 112 5.2. Conclusion 115 5.3. References 115

6. Catalytic Lewis Acids in Reductive Amination 6.1. Introduction 117 6.2. Brønsted Acid Promoted Reductive Amination 122 6.3. Conclusion 125 6.4. References 125 x

7. Stereochemical Considerations of Proposed Mechanistic Models 7.1. Introduction 127

7.1.1. Mechanism Behind Enhanced Stereoselectivity with Yb(OAc)3 128 7.1.2. Reasons behind Enhanced Diastereoselectivity for Different 133 Substrate Categories 7.1.3. Key Findings for Reductive Amination with α-MBA 135 7.2. Conclusion 137 7.3. References 138

8. Appendix Experimental Section 140 Curriculum Vitae 152

xi

Chapter 1 Introduction

1.1. Chiral Drugs

Chiral molecules form a large proportion of therapeutic agents. Drug chirality is considered a major theme in the design, discovery, development, launching and marketing of new drugs. Awareness of the importance of chirality comes from the fact that stereoselectivity is an essential dimension in pharmacology. The recent development of bioanalytical tools led to a better understanding of the importance of stereoselective phramacodyanmics and pharmacokinetics of chiral drugs. In 1984 it was estimated that the total proportion of drugs having chiral centre in the European market (Swedish survey) was 53%.[1] The percentage increased up to 57% within less than three years.[2]

It was also estimated that 55% of the chiral drugs are used as a and the rest is marketed as a single . By the end of the last century, the market for chiral drugs established major place in the overall global drug market. The situation totally changed in this century. Pharmaceutical companies stopped developing racemic drugs; they only focus on the synthesis of single enantiomeric drug entities.

First we have to clarify the concept of chirality. Chirality or handedness comes originally from the Greek word cheir which means hand. One of the simplest definition of the word chiral is given by Mislow: An object is chiral if and only if it is not superposable on its mirror image; otherwise it is chiral.[3] From the definition it is clear that the term chiral refers to the spatial property of the objects including molecules. It defines that the molecule is non- superposable on its mirror image and does not refer to the stereochemical composition of the bulk material.[4]

1

It should be clear that the term chiral drug does not indicate that the drug is marketed as a single isomer it may be a racemic or unequal mixture of isomers. Through investigating the origin of chirality it was revealed that the concept was introduced long ago. Archimedes designed Archimedean water screw and studied its chiral structure. Dominique Arge (1811) discovered the rotation of plan polarized light in quartz crystal. Later the French chemist Jean Baptiste Biot was the first to introduce the modern concept of chirality when he discovered rotation of light in a sugar solution.[5]

The major breakthrough in understanding the concept of chirality and its significance in chemistry was achieved by Louis Pasteur through recrystallization of sodium ammonium tartrate (optically inactive). He noticed that the crystals were of two types which he physically separated. The two types of crystals were optically active, but rotated the plane of polarized light in the opposite directions. He proposed that the molecules came in two forms, “left handed” and “right handed”. Together, the mixture of the two forms is optically inactive. This finding prompted his famous statement that the universe is chiral (l’univers est dissymme´trique).[6]

Later Van’t Hoff, a Dutch young scientist proposed that the carbon atom is attached to four different substituents in space having a tetrahedral arrangement. This proposition was faced by strong opposition from scientists all over the world. Later they discovered that his proposed shape of the molecule was absolutely right and he was awarded the first noble prize in chemistry for his work.[7] Chirality is manifested by centre of dissymmetry, but it can also be represented in axes or planes of dissymmetry.[8]

1.2. Isomers and Isomerism

Isomerism is the phenomenon of two or more compounds having the same number and kind of atoms.[9] Isomers can be subdivided into structural isomers, the difference between isomers is due to a different structural arrangements of the atoms that form molecules, e.g. butane and isobutene. The other division is stereoisomers, the isomers have the same structural formula, but differ in the spatial arrangement of atoms.[9]

There are two types of stereoisomers: 2

1. Cis-trans or geometric isomers. 2. Optical isomers

Optical isomers have the ability to rotate plane-polarized light.[8] are part of the optical isomers, together with . Enantiomers are mirror image optical isomers having only one chiral centre. Enantiomers posses the same physical properties but they differ in their biochemical properties. They behave differently only in a chiral medium, such as when exposed to a polarized light or when participating in a chemical reaction catalyzed by a chiral catalyst, particularly an enzyme in the body. (+)-Glucose (“blood sugar”) is used for metabolic energy whereas (-)-glucose is not. (+)-Lactic acid is produced by reactions occurring in muscle tissue, and (-)-lactic acid is produced by the lactic acid bacteria in the souring of milk. Diastereomers are non mirror image optical isomers having more than chiral centre. Diastereomers have different physical properties allowing their separation. Enantioselectivity and diastereoselectivity are terms used to express the preferential formation of one enantiomer or diastereomer over the other and it is normally expressed as an enantiomeric excess (ee) or diastereomeric excess (de).

R(%) - S(%) D1(%) - D2(%) ee (%) = + 100 de (%) = + 100 R(%) + S(%) D1(%) + D2(%)

1.3. Nature is Chiral

Many naturally occurring substances possess chirality, which is the property that a substance and its mirror image are not superimposable.[10] In every-day life, many examples can be found as well. Human hands are perhaps the most universally-recognized example of chirality. The left hand is a non-superimposable mirror image of the right hand; no matter how the two hands are oriented, it is impossible for all the major features of both hands to coincide.[11]

In particular life depends on molecular chirality, with many biological functions/processes inherently based on the interaction of dissymmetric molecules. Many physiological phenomena arise from highly preferential molecular interactions in which a chiral host

3

molecule recognizes one of two enantiomeric guest molecules. There are numerous examples of enantiomeric effects which are frequently dramatic. Thus, the enantiomers of , both are found in nature, smell differently, because our nasal receptors are made of chiral molecules that interact with these enantiomers differently. Similarly one enantiomer of the asparagine tastes sweet while the other tastes bitter. Clearly living systems are very sensitive to chirality and many pharmaceutical drugs consist of chiral moieties. Chiral drugs are a subgroup of drug substances that contain one or more chiral centres. It is well established that the opposite enantiomer of a chiral drug often differs significantly in its pharmacological,[12] toxicological,[13] pharmacodynamic and pharmacokinetic properties.[14]

A renowned example of how chirality affects the pharmacological action of the drugs, a chiral drug is thalidomide (Thalidomid, Contergan) which was prescribed to pregnant women in the 1960s to alleviate morning sickness. One of the enantiomeric forms of thalidomide does indeed have sedative and antinausea effects, but the other enantiomer is a potent teratogen. The racemic drug was approved in Europe for the treatment of pregnant women suffering from nausea and its use caused severe birth defects. Even formulation of the pure nontoxic (R)-enantiomer of thalidomide would have been unsafe because racemization takes place in vivo and the teratogenic (S)-enantiomer is rapidly generated in the human body. Since the thalidomide tragedy, the significance of the stereochemical integrity of biologically active compounds has received increasing attention and the investigation of the stereodynamic properties of chiral molecules has become an integral part of modern drug development.[15]

4

O O H NH * N * O OH 2 O Thalidomide Ethambutol (R)-active agent (R,R)-blinding agent Limonene Limonen (R)-organge odor (S)-teratogenic (S,S)-tuberculostatic (S)-lemon odor

O O O O

H2N NH2 OH HO OH2N H H NH2O Asparagine Asparagine Carvone Carvone (S)-bitter (R)-sweet (S)-caraway (R)-spearmint

HO O O H O OH H O O N O H N H NH2 O H O H2N H Aspartame Aspartame (S,S)-sweet (R,R)-bitter

Figure 1.1 Absolute Configuration vs Biological Activity.

Another example showing the importance of distinguishing the two enantiomers is the distinguished effect of different isomeric forms of the nonsteroidal anti inflammatory drugs (NSAIDs). They include ibuprofen (Advil), naproxen (Aleve), ketoprofen (Oruvail), and flurbiprofen (Ansaid), which have found widespread use as pain relievers. The anti- inflammatory activity of these profens resides primarily with the (S)-enantiomer. The enantiomers of flurbiprofen possess different pharmacokinetic properties and show substantial racemization under physiological conditions. Although (S)-naproxen is the only profen that was originally marketed in enantiopure form, in vivo interconversion of the enantiomers of NSAIDs is an important issue in preclinical pharmacological and toxicological studies.[16]

Also beta-blokers, the most widely used pharmaceutical agents for angina, hypertension, and arrhythmias. It is known that for most beta-blockers the (S)-enantiomer is the most active enantiomer. The S-enantiomer has the same three dimensional structure as the adrenergic

5

hormone noradrenaline. The (R)-enantiomer of the betablocker does not give serious side- effects, but it does not add to the pharmacological effect either, so it can be considered as ‘isomeric ballast’. The most sold beta-blockers (propranolol, atenolol, metoprolol) were developed in the 1970s and are still marketed as racemate. If these substances would have been developed today, it can be expected that they would have been introduced as a single enantiomer.[17]

Therefore from the points of view of safety and efficacy, the pure enantiomer is preferred over the racemate in many marketed dosage forms. In past decades the pharmacopoeia was dominated by racemates, but since the emergence of new technologies in the 1980s that allowed the preparation of pure enantiomers in significant quantities, the awareness and interest in the of drug action has increased. Although some ‘‘blockbuster’’ drugs, such as fluoxetine hydrochloride (Prozac) is still marketed as racemates. However, the recent trend is toward marketing a single-enantiomer drugs.[18]

Previously the chiral drug is often synthesized in the racemic form, and it is frequently costly to resolve the racemic mixture into the pure enantiomers. Another approach by pharmaceutical companies is what is called racemic switch. This fashionable approach involves the development of a pure enantiomer of the drug that is already marketed as a racemate. This means if a patent on a drug that is marketed as a racemic mixture is expiring; it is sometimes possible to obtain a new patent for the active enantiomer. In this way, the pharmaceutical company retains the exclusive rights on the substance for another period, but they will have to change their manufacturing method as well.[19]

Although only a minority of all racemic drugs has proved to be suitable for a racemic switch, this development has boosted the development of new manufacturing and separation methods. An example of a successful racemic switch is the local anaesthetic bupivacaine (AstraZeneca's Marcain). The (S)-isomer is now marketed under the trade name Chirocaine. This isomer was found to be substantially less cardiotoxic than the (R)-isomer, and therefore a new patent was granted. Furthermore, the (S)-isomer of omeprazole (a proton pump inhibitor by AstraZeneca, known as Losec/Prilosec) is now marketed as a single enantiomer under the trade name Nexium.[20]

6

One enantiomer may be responsible for the activity; its paired enantiomer could be inactive, possess some activity of interest, be an antagonist of the active enantiomer or have a separate activity that could be desirable or undesirable. To market drug as racemate or as the enantiomeric pure form is mainly based on pharmacology, toxicology and economics. From a pharmaceutical perspective, the physical properties of both the racemate and the enantiomer should be characterized in detail in order to develop a safe, efficacious, and reliable formulation, no matter whether the racemate or the enantiomer pure form is chosen as the marketed form. Chirality of a drug can also influence the efficiency of delivery, which has not been well investigated in the pharmaceutical field.[21]

Density, solubility, dissolution behaviour, stability, and mechanical properties which are the many physical properties of a crystalline solid, are governed by the crystal structure.[22] Understanding the relationship between the crystal structure and the physical properties, and their influence on drug release, may therefore provide a clear picture of chirality–delivery relationship.

This discussion supports the fact that using single enantiomeric pure form of a drug has major advantages as reducing the overall administered dose, improving drug therapeutic window, reducing any intersubject variability and finally estimating the dose response relationship accurately.[23]

All previously reported reasons have led to an increasing preference for production of the single enantiomers in both industry and regulatory authorities. Regulation regarding control of chiral drugs began in the US with a publication in 1992 about the formal guidelines on the development of chiral drugs in a document entitled Policy Statement for the Development of New Stereoisomeric Drugs by FDA and European Union. The major outlines for the guidelines state that the drug applicants must recognize the occurrence of chirality in the new drugs, attempt to separate the stereoisomers, assess the contribution of the various stereoisomers to the activity of interest and make a rational selection of the stereoisomeric form that is proposed for marketing.

Global sales of chiral drugs in single-enantiomeric form continue to grow. The annual sales of chiral drugs as a single enantiomeric form increased dramatically, from 27% (US $74.4 7

billion) in 1996, 29% (1997), 30% (1998), 32% (1999), 34% (2000), 38% (2001) to an estimate of 39% (US $151.9 billion) in 2002.[24]

1.4. Chirality and Drug-Receptor Interaction

As mentioned before biological systems are based on chirality. For example, enzymes are considered as chiral biological polymers consisting of solely L-amino acids. They are highly structured compounds: their secondary and tertiary structure is determined by the amino acid constituents. Enzymes function as molecular receptors by binding selectively to specific molecules. Due to their chirality, they commonly interact much stronger with one enantiomer of the ‘target’ molecule; or what is known as chiral recognition. ‘lock-and-key’ concept was introduced in 1894 by Fischer to explain enzyme selectivity. This simple concept sates that one enantiomer ‘fits’ in the enzyme cavity, the other enantiomer does not. This concept was reformulated later to more complicated model (three point model).[25]

To get a high degree of enantioselection, a substrate must be held firmly in three dimensional space. There must be at least three different points of attachment of the substrate onto the active site. Variations and refinements to this rule have been reported. The most important one is that the interactions may be attractive or repulsive. Steric hindrance often plays an important role in chiral recognition.Chirality has also an important role in the field of fine chemical industry. Large applications are found in agrochemicals, food and fragrance industry. There are handful examples of enantiomers showing different fragrances because one is (R) and one is (S).

For a chiral herbicide from the class of α-aryloxypropionic acids, the activity is present almost solely in the (R) enantiomer. Although currently most herbicides are applied as racemates, attempts towards the development of convenient large scale methodologies for the synthesis of herbicides were awarded by development of metalochlor. This trend will result in 50% reduction of dosage, which means 50% less environmental pollution.

1.5. Sources of Enantiopure Substances

8

1.5.1. Synthesis of Enantiomerically Pure Compounds

The importance of chiral compounds and the strong need for enantiomerically pure substances has led to develop versatile methodologies to meet this objective. There are three main approaches for the preparation of chiral compounds as shown in figure 1.2:

1.Resolution of racemates; 2.Chiral pool approach; 3. Stereoselective conversion of prochiral substrates to enantiopure compounds (asymmetric synthesis via catalytic or stoichiometric process).

Racemates Chiral Pool Prochiral Substrates

Synthesis Asymmetric Synthesis Preferential Kinetic Diastereomer Resolution Crystallization Crystalization

Chemical Enzymatic

Figure 1.2. Sources of Enantiopure Substances.

1.5.2. Resolution

Resolution technique is the most classical route to enantiopurity. Although it has many drawbacks and recently it has been overtaken by asymmetric synthesis, this method is still persisted with numerous examples on the industrial scale till present days. Resolution can be subdivided into three main techniques.

1.5.2.1. Preferential Crystallization

Preferential crystallization is possible for racemates which form conglomerates. The conglomerates are mechanical mixtures of enantiomerically pure crystals of one enantiomer 9

and its opposite enantiomer. Molecules in the crystal structure have a greater affinity for the same enantiomer than for the opposite enantiomer. The melting point of the racemic conglomerate is always lower than that of the pure enantiomer. Addition of a small amount of one enantiomer to the conglomerate increases the melting point. Success in this method depends on the fact that for a conglomerate the racemic mixture is more soluble than either of the enantiomers. Generally only 5-10% of racemates form conglomerates.[26]

1.5.2.1. Diastereomer Crystallization

L. Pasteur was the one who first to fully develop this methodology in 1854.[26] In this approach, a racemate interacts with an enantiopure compound to form diastereomeric salt which can then be separated by crystallization due to unequal solubility in a given solvent. These enantiopure compounds are called resolving agents and are obtained from the chiral pool, e.g. L-tartaric acid, D-camphor sulfonic acid or some alkaloid bases. In general this approach is extremely limited to few examples. One example of such process is the crystallization of the salt of one enantiomer of 1,2 diamino cyclohexane obtained from the interaction of the racemic mixture with enantiopure tartaric acid.[27]

O SO3H O HO3S

H

OH X

N Quinine (X=OMe) Chinchonidine (X=H)

H2N HO OH H2O/AcOH H3N K2CO3 NH 90oC-5oC COO 3 H2O/EtOH H2N HOOC COOH H2N NH2 HO COO >98% ee HO

Scheme 1.1. Preparation of 1,2 Diamino Cyclohexane

10

1.5.2.2. Kinetic Resolution

Kinetic resolution is based on the difference in reactivity rate of the two enantiomers with a chiral entity which is used in catalytic amount.[28] The chiral entity can be either a biological catalyst (e.g. enzyme) or a chemical catalyst (e.g. chiral metal complex or organocatalyst). Rule of thumb for kinetic resolution to be successful is that one enantiomer must react faster than the other. In such situation theoretically, 50% of the product from one enantiomer and 50% of the unreacted enantiomer should be obtained. Scheme 1.2 showing one example describing this condition in which racemic β-aryl-β-hydroxy with different substitution patterns on the aryl moiety provides preferably the (R)-enantiomer with 93-98% ee and 32- 41% isolated yield.[29]

Ph Ph N Ph N H HO

BrZnCH2CO2tBu(8equiv.) CO2tBu ligand (5.0 mol%) CO 2tBu CO2tBu Ar OH THF, reflux Ar OH Ar

Scheme 1.2. Kinetic Resolution in Asymmetric Synthesis. .

If the unwanted enantiomer is racemized in situ during resolution, a 100% theoretical yield of the enantiopure product can be theoretically reached, this is known as a kinetic dynamic resolution. This approach was successfully applied utilizing enzymes as resolving agents. [30]

11

CaLB, EtOAc Toluene N ~48h N N NH NH NHAc 2 2 >60% yield

N N N O HN

racemization Scheme 1.3. Dynamic Kinetic Resolution in Asymmetric Synthesis.

For chemical synthesis, one of the earliest demonstrations of this method is an adaptation of the Noyori .[31]

O O OH O H2 R1 OR3 R1 OR3 (R)-BINAP-Ru R2 R2

OH O O O H2 R1 OR3 (R)-BINAP-Ru R1 OR3 R2 R2

a: R1=R2=CH3;R3=C2H5 b: R1=R3=CH3; R2=NHCOCH3 c: R1=3,4-methylenedioxyphenyl; R2=NHCOCH3;R3=CH3 d: R1=3,4-methylenedioxyphenyl; R2=NHCOCH2C6H5;R3=CH3 e: R1=R3=CH3;R2=CH2NHCOC6H5 Scheme 1.4. Chemoselective Dynamic Kinetic Resolution

1.5.3. Chiral Pool Approach

Natural sources are often referred as the ‘chiral pool’. The most important classes of chiral pool substances are amino acids, carbohydrates, hydroxy acids, and alkaloids.[32] These substances are incorporated into products by chemical processes which involve retention of configuration, inversion or chirality transfer. The chiral starting material is called chiral synthon which introduces chirality in the final compound. This strategy is unlike chiral auxiliary approach (will be discussed later) in which the chirality is installed into the achiral

12

compound by the auxiliary. The auxiliary is later deattached from the final product. Despite the breadth of functionality available from nature, limited examples are available in optically pure form on a large scale. This means that incorporation of a “chiral pool” material into a synthesis can result in a multistep sequence. However, with the recent advances in synthetic methods which added new compounds to the chiral pool they are still limited.

Typically chiral pool material should be available on large scale in a reasonable price. One example is L-aspartic acid, where the chiral material can be cheaper than the racemate. An example of the application of chiral pool for synthesis of pharmaceutical drugs is the synthesis of (S)-Vigabatrin, a potent GABA-T inhibitor from (R)-methionine by Knaus and Wei in 96% yield and >98% ee as shown in scheme 1.5.[33]

S COO 1 S COOMe 2 S COOR2

NH3 NHCOOR1 NHCOOR1

a) R1=PhCH2 b) R1=Me a) R1=PhCH2,R2=Et b) R1=PhCH2,R2=Me 3 c) R1=Me, R2=Et d) R1=Me, R2=Me

COO 5 4 O SMe N O N NH3 H H

1. i)MeOH/SOCl2. ii) NaHCO3/ClCO2R1. yield 82%-86%; 2. (R2O)P(O)CH2CO2R2/tBuLi/DiBAL-H,62-78% yield; 3. Mg/MeOH. 92-95% yield; 4. i) NaIO4. ii)190 °C. 56%yield; 5) KOH/iPrOH/H2O. 96% yield. 98% ee.

Scheme 1.5. Synthesis of (S)-vigabatrin from (R)-methionine, chiral pool approach.

13

Chiral Pool Compounds

Amino acids Hydroxy acids

NH2 OH OH OH NH2 COOH COOH COOH COOH Ph COOH HOOC OH lactic acid mandelic acid Valine phenylalanine tartaric acid

Sugars Terpenes OH OH OH OH O O O HO HO OH OH HO OH HO camphor glucose mannose

Figure 1.3. Examples of Chiral Pool Compounds.

Another example utilizing this approach is the synthesis of herbicide (R)-flamprop-isopropyl starting from L-lactic acid.[34]

OH OSO Me MeSO Cl 2 i 2 COO Pr i base COO Pr (S)-lactic acid

Cl

inversion F NH2

Cl O Cl i F N PhCOCl COO Pr F N iPrOOC H

(R)-(-)-flamprop-isopropyl

Scheme 1.5.Synthesis of Enantiopure Herbicide from L-lactic acid.

14

1.5.4. Stereoselective Conversion of Prochiral Substrates to Enantiopure Compounds (Asymmetric Synthesis)

In asymmetric synthesis a stereogenic centre is created under the influence of some external or internal chiral inducing agents. This strategy can be subdivided into three approaches: substrate-controlled approach; chiral auxiliary approach; and catalyst controlled approach. In substrate controlled approach, chirality is present internally within the molecule directing remaining groups or faces in stereoselective manner. Limitations of this approach come from the fact that enantiopure starting materials are not easily available and the reacting sites should be within close proximity to the chiral centre.

Regarding the other two approaches, achiral molecule is converted into chiral entity utilizing either a stoichiometric quantity of the chiral auxiliary or a catalytic quantity of chiral catalysts. In the chiral auxiliary approach, chirality is induced in achiral molecule utilizing external chiral entity through forming covalent bond with the achiral starting material. This auxiliary is then cleaved from the final product in an additional step. Special precautions should be taken to avoid any racemisation of the final product in the deportation step. One example of this auxiliary approach is shown in scheme 1.6 in which (1S,2S)-(+)- is used as the chiral auxiliary to produce diastereomeric alkylated pseudoephedrine which can form enantioenriched carboxylic acids(by hydrolysis), and (by reduction).[35]

15

O 1. 2LDA, LiCl O R 2. R1X N R N OH THF OH R1 80-99% yield 94-99% de

O H2SO4 dioxane 87-97% yield R 95-97% ee O HO R1 R N BH3 Li N R 80-88% yield OH R HO 88-99% ee 1 THF R1

O R 75-92% yield H 90-98% ee R 1 Scheme 1.6. Chiral Auxiliary Approach in Asymmetric Synthesis

The third approach which is the catalytic asymmetric transformation, is promoted by a chiral entity which is generally used in a catalytic amount enhancing the economic value of the process. The chiral entity can be chiral catalysts (e.g. chiral Lewis acid or base, chiral organocatalysts, chiral organometallic complexes) or even bio catalysts. One of the most fascinated examples was the synthesis of L-DOPA developed by Knowles.[36]

H COOH H COOH COOH + H [Rh(DiPAMP)] H3O H NHAc H NH2 NHAc H2 AcO AcO AcO OMe OMe OMe L-DOPA 97.5% ee

P P CH O 3 OCH3

Scheme 1.7. Synthesis of L-DOPA

Another example showing the importance of this approach, was developed by Royoji Noyori,[37] In 1980 he developed different derivatives of chiral BINAP ligands which were widely used as chiral ligands for Ru and Rh hydrogenation reactions. He was successful in applying his catalytic system on industrial scale for the (-)- synthesis from myrcene.

16

It is estimated that 3000 tonnes (after new expansion) of menthol are produced (in 94% ee) by Takasago International Co., using Noyori's method every year. The key step was the asymmetric isomerization of geranyldiethylamine, promoted by an (S)-BINAP-Rh complex in THF and forming (R)-citronellal enamine, which upon hydrolysis gives (R)-citronellal in 96-99% ee. This enantiopurity is higher than naturally available product (ee 80%) obtained from rose oil (scheme 1.8).

H Li, (C2H5)2NH R Hs N(C2H5)2 myrcene diethylgeranylamine

[Rh(S)-BINAP]+

HR CHO + Hs H3O (R)-citronellal N(C2H5)2 (R)-citronellal enamine 96-99% ee ZnBr2

OH OH

H2,Nicat

isopulegol (-)-menthol

Scheme 1.8 Rhodium-BINAP in Synthesis of Menthol.

Noyori BINAP system was applied successfully in the synthesis of many important pharmaceutical drugs as the anti-inflammatory drug, naproxen, in 97% ee from α-aryl-acrylic acid.[38] and the antibacterial levofloxacin obtained from hydroxyacetone through asymmetric hydrogenation of (R)-1,2-Propanediol.[39,40]

Third major breakthrough in the field of asymmetric synthesis was introduced by Barry Sharpless.[41] He developed a highly enantioselective epoxidation of allylic alcohols. His successful result was obtained by the use of a titanium-tartrate complex as the catalyst and water in a ratio of 1:2:1. For the process to be catalytically useful only a slight modification was required. Before catalyst formation 4 Å molecular sieves had to be added. The molecular sieves act as a moisture scavenger and, therefore, control the amount of water present in the reaction mixture. In addition, the formation of other, undesired titanium species which lead to

17

non-enantioselective pathways is diminished (scheme 1.9). Recently, a further decrease of catalyst loading to 10 mol % has been achieved by replacing water with isopropanol.

R R R1 R2 i 1 2 tBuOOH, Ti(O Pr)4 O o OH OH L-(+)-DET, CH2Cl2,-20 C R R3 3 70-90% yield 90-98% ee

HO COOEt L-(+)-DET = HO COOEt

Scheme 1.9. Enantioselective Epoxidation of Allylic Alcohols.

The work of those great minds was rewarded with a Noble prize in 2001 by the Royal Swedish academy of sciences.

1.5.5. Asymmetric Synthesis vs Kinetic Resolution vs Chiral Pool:

From the above discussions it can be concluded that each approach of the three major approaches has advantages and disadvantages. Resolution suffers from a major drawback which is the low yield of the desired product; the maximum obtainable yield is only 50% from the racemates. In case of Kinetic dynamic resolution utilizing enzymes or chiral catalysts the yield can be improved. The yield in the kinetic resolutions can be improved by fast conversion of the (S)-enantiomer into the racemic mixture and the (R)-enantiomer reacts preferably to form the desired product in high yield and ee. The ideal dynamic kinetic resolution reaction which approaches 100% conversion of 100% enantiomerically enriched product is the one in which krac>>kR>>kS (figure 1.4). If krac was in fact closer to or even slower than kR, the ee of the product would be lowered because the amount of (R) in solution would not be produced fast enough to make kS negligible.

18

kR R P1

krac

S P2 kS

Figure 1.4. Reaction Constants for Dynamic Kinetic Resolution.

1.6. α-Chiral Amines Defining Terms:

Amino compounds with a stereogenic centre at the position α-to the amino group are known as α-chiral amines.

Ph Ph Et tBu COOtBu NH2 NH2 NH2 NH2

NH2 NH2 NH2

Figure 1.5 Examples of α-Chiral Amines.

They can be addressed as chiral amine for simplicity and we will try to stick to this nomenclature throughout the whole thesis. Chiral amines are useful intermediates for alkaloid natural product synthesis, eg: morphine, codeine and tropane alkaloids. They are also incorporated in different block buster drugs as the billion dollar drugs, e.g. several ACE inhibitors and Flomax.[42]

To understand the importance of this moiety in the asymmetric synthesis it is estimated that at least 40% of all optically active pharmaceutical drugs contain this moiety. Unfortunately,

19

80% of the synthetic methods still rely on the classical resolution methods.[43] Searching literature in the last 30 years revealed that there is a great lack in efficient methodologies for synthesis of chiral amines.[44] Different synthetic strategies have been developed for the synthesis of chiral amines but as a general conclusion most of these strategies suffer from low yield or stereoselectivity. One of the main challenges in the synthesis of chiral amines comes from the lack of efficient methodology for the synthesis of alkyl-alkyl amines.[45]

This class of chiral amines are accessible in high yield and enantioselectivity through long tedious procedures. Also starting materials are often expensive and requires the use of stoichiometric quantities of chirality inducing agents. The overall process is not atom economical and the waste production is high. As a conclusion the available processes for chiral amine synthesis suffer from many disadvantages resulting in an extreme difficulty for their synthesis on industrial scale utilizing the available methodologies. In this study we will try to identify the existing problem, demonstrate all possible available solutions and show our developed approach to solve this problem.

1.7. α-Chiral Amines Importance:

Enantiomerically pure amines with an α- plays an important role in organic synthesis. Their applications are innumerable: as chiral resolving agents,[46] chiral auxiliaries,[47] ligands in various asymmetric transformations[48] and as advanced building blocks in pharmaceutical and agrochemical industries.[49] They are also fruitful as chiral ligands in metal-complex catalysis.[50]

Ph NH NH2 2 OH H2N NH (R)-or (S)-α-methylbenzylaine (R)-or (S)-phenylglycinol 2 (1S,2S)-cyclohexane 1,2- diamine

CO2H NH2 N N H (S)-3-aminoquinuclidine L-proline

Figure 1.6. α-chiral amines Available Commercially. 20

(S)-(α)-Methylbenzylamine and its enantiomer (R), appear to be ideal compounds as chiral auxiliaries or chiral building blocks for pharmaceutical and chemical industry. (S) and (R) enantiomers are inexpensively available in very high enantiomeric purity, which makes them attractive as stereodifferentiating agents even for industrial scale operations.[51]

It has been used in the synthesis of biologically active molecules such as Labetalol (β- blocker) and Tamsulosin.[52] Another example is α-amino acids, eg; proline. Proline has a unique value in asymmetric processes; it is used as a ligand in transition metal-catalysis.[53] Recently proline and its derivatives were applied as highly efficient organocatalysts in different organic transformation as asymmetric Aldol,[54] Mannich[55] and Michael reactions.[56]

Another important class of chiral α-chiral amines which are used in synthesis of pharmaceutical building blocks is the quinuclidine family. An example of this class includes enantiopure 3-aminoquinuclidine, an important intermediate in the synthesis of 5-HT3 serotonin ligands,[57] such as zacopride. Also diamines as (1S,2S)-Cyclohexane-1,2-diamine is used as chemotherapeutic agents,[58] chiral auxiliary, transition metal-catalysis and in .[59]Of course there are more examples of the available chiral amines which are used in pharmaceutical and agrochemical industry with great success for synthesis of natural products and drugs.[60] The following figure shows some examples of drugs having chiral amines(figure 1.7).

O NH2 NMe2 OH H H H N S N O N O O O H HO O N COOH rivastigmine amoxicillin (Alzheimer) (antibiotic) (S)-repaglinide (hypoglycemic agent) SO2NH2 HO MeO O N H OEt N Cl O N H N H H HCl HO H2N OMe CONH 2 (R)-tamsulsin hydrochloride (S)-zacopride Labetalol (benign prostatic hyperplasia) (5-HT3 agonist) blocker) (β− O Ph O OH NH AcO COOMe H Ph O N H Ph HO O O AcOH Ritalin HO OBz (treatment of hyperdeficit disorder) Taxol (anticancer drug)

Figure 1.7. Examples of Drugs with α-Chiral Amines. 21

1.8. α-Chiral Amine Synthesis Different Methodologies

As mentioned previously chiral amines are key components of different pharmaceutical and agrochemical compounds. Over the last fifty years different methodologies have been developed for their synthesis. Some of the methodologies are industrially viable and others are better suited for pilot studies. Of course for a methodology to be applicable on industrial; scale it must fulfil certain features e.g. should be cost effective and waste generation should be low. Some processes are highly efficient in preparing chiral amines in high yield and stereoselectivity. Despite their efficiency they suffer mainly from major drawbacks as lengthy multistep procedures which hinder their applications on industrial scale. Among the versatile strategies employed is the hydrogenation of enamine esters (diastereo and enantioselective),[61] hydrogenation of α- or β-N-acetylenamide esters,[62] 1,4-addition of amines to enones,[63] Chemical,[64] and enzymatic[65] reductive amination of α- ketoacids, remote amination via C-H insertion[66] and hydroamination of olefins.[67]

Reduction of unfunctionalized and aldehydes is one of the major strategies for the synthesis of chiral amines. This strategy can be subdivided into various subdivisions which includes the following.

1) N-acetylenamide reduction. 2).Transfer hydrogenation or hydrogenation of imines 3). Reductive amination of ketones 4) Carbanion addition to aldimine and ketimine derivatives. 5) Sequential aminationalkylation of aldehydes.

The first three methodologies are closely related as they use hydrogen from different hydride sources for the reduction of prochiral carbonyl compounds. The asymmetric version of these methodologies has been developed extensively over the past few years. N-acetylenamide reduction and transfer hydrogenation or hydrogenation of imines will be discussed in details trying to shed light on their advantages and disadvantages and their applications. Reductive amination as the core of my work will be discussed showing its historical development over the last century and the major breakthroughs in the field during the last two decades. The application of reductive amination in pharmaceutical industry will be summarized in the 22

fourth chapter showing different drug and natural product categories prepared utilizing reductive amination.

1.8.1. Imine and Enamide Synthesis

A discussion about the preparation of imine and enamide are necessary as most of the examples in scientific journals focus mainly on the manipulation of imines (N- phosphinoylimines) or N-acyl enamines as starting materials without a clear picture about their preparations. The overall yield of the chiral amine products is very rarely discussed and therefore a perspective in this regard needs to be established.

Fe powder NHAc O NH2OH HCl NOH R' R' R' Ac2O R R MeOH R o AcOH. Toluene, 75 C

Scheme 1.10 Synthesis of N-Acyl Enamide from

The commonly used method of enamide[68] synthesis is the one in which the desired compound is synthesized from different substituted ketones in two steps as carried out by Burk (scheme 1.10).[69]

In the first step of this synthesis the ketone is converted to an oxime with hydroxylamine hydrochloride in MeOH, the yield of the ketoxime is generally >90%. The next step is the interaction of the resultant ketoxime with Iron powder and acetic anhydride with AcOH in toluene at a temperature of 75 °C. The yield of the enamide is generally between 30-60% in this step.[70]

In general the enamide synthesis methodology is low yielding process. Besides the possibility of diacetyl formation which is considered another drawback. The E/Z mixtures which are obtained with R’ -as non-hydrogen atom- are difficult to separate.

1.8.2. Enantioselective Reduction of Enamides

23

Enantioselective reduction of enamide is very interesting approach for the synthesis of chiral amines from the enantioselectivity prospective. On the other hand, it is a four step procedure to get the final primary amine product. Two step process N-acyl enamide synthesis and a further two steps are involved, reduction of the enamides and hydrolysis of enamide, before the primary amine is obtained. The overall yield is rarely mentioned in literature. Through calculating each step yield and estimating the overall all yield it is obvious that the yield is low (usually below 50%). Enamides generally are obtained as E and Z mixtures, but this does not seem to affect enantioselectivity. Specific substrate categories can be only utilized in enamide reduction protocol, pinacolone and aryl-alkyl ketones.[71]

We will focus on enamide methods allowing alkyl-alkyl and aryl-alkyl substituted α-chiral amine synthesis. Ding immobilizes Feringa’s MonoPhos/Rh catalyst for the asymmetric hydrogenation of dehydro-α-amino acid esters and enamides. Treatment of the ditopic

MonoPhos ligand with [Rh(cod)]BF4 in DCM/toluene resulted in an immediate precipitation of an amorphous Rh-containing polymer, which were demonstrated to be effective catalysts for the asymmetric hydrogenation. Secondary amines were prepared in excellent enantioselectivity (scheme 1.11).[72]

N P O [Rh(cod)2]BF4 linker O O CH Cl /Toluene O P 2 2 N

linker a: linker N O O P [Rh] P O O b: N

n c: single bond O O 12a-c, 1 mol% R OCH3 R OCH3 NHAc H2, 40 atm, toluene NHAc R=H, CH3, Ph Full conversion 94-96% ee

12a-c, 1 mol%

Ph NHAc H2, 40 atm, toluene Ph NHAc

Scheme 1.11 Heterogeneous Catalysis with Self Supported Rh Catalysts.

24

Burk was successful in reducing aryl-alkyl or alkyl-alkyl enamides with high ee (>95%) utilizing Rh[Me-DUPHOS] or Rh[Me-BPE] catalysts (Figure 1.7). The substrates are acyclic and benzocyclic aryl-alkyl ketones, and only two examples for alkyl-methyl ketones with sterically encumbered groups such as t-Bu (pinacolone) or adamantly groups as alkyl substituents. As described before there is no information about the yield from the starting ketone up to the final product. Another issue regarding this work is the limited substrate breadth.[73]

P P P P

Figure 1.7 Examples of Chiral Ligands Used by Burk.

Noyori has reported a general and straightforward method for synthesizing enantiomerically pure tetrahydroisoquinoline alkaloids through reduction of enamide. Ru-(S)-BINAP and Ru- (S)-BIPHEMP complexes. These complexes resulted in almost perfect enantioselectivities in hydrogenation for a wide array of tetrahydroquinolines. The present reaction provides access to a wide variety of alkaloids as morphinic and synthetic morphinans and benzomorphans analogues (scheme 1.12).[74]

1-4 bar H2 MeO Ru-(S)-BINAP MeO MeOH NCHO NCHO MeO MeO

R R ee>99% R=H, OMe

OMe OMe

Scheme 1.12. Noyori Catalyst for Enamide.

25

Another methodology for the synthesis of N-Boc-(R)-3-amino-2,3,4,5-tetrahydro-1 H- [1]benzazepin-2-one, which is an important intermediate for the preparation of an angiotensin converting enzyme inhibitor, based on asymmetric acyclic enamide hydrogenation has been reported by Merck (scheme 1.13).[75]

3.4 bar H2 NHBoc NHBoc (S)-BINAP RuCl 2 N N 80% yield H O H O 82% ee

Scheme 1.13. Merck Based Methodology for Enamide Hydrogenation.

Zhang worked extensively on the enantioselective reduction of N-acetyl enamides. Various types of chiral ligands were tested with rhodium catalysts showing extremely high enantioselectivities for aryl-alkyl, benzocyclic and ortho substituted aryl ketones using acceptable catalyst loading (0.1-1 mol %) (figure 1.8).[76]

P P P P

(R,S,R,S)-MePennPhos (R,R)-Binaphane

OCH3 PPh2 H CO Ph Ph2P R 3 H O P N R= Et or Me H3CO PPh2 O H R H3CO PPh2 PPh2 (R,R)-BICP PPh2 H3CO Ph Phosphine-Phosphoramide ligand OCH3 (S)-o-Ph-hexMeO-BIPHEP

Figure 1.8. Chiral Ligands Used by Zhang in Enamide Reduction.

Rhodium catalyst proved to be the catalyst of choice for N-acetyl enamide hydrogenation the following table summarizes the latest findings in this field (table 1.1).[77]

26

Table 1.1 Rhodium Catalyzed Reduction of Enamide

Substrate Ligand H2(bar) ee Configuration

R = H, Ar = Ph Manniphos R= 10 99.5 (R) Me

R = H, Ar = p-CF3Ph Tangphos 1.4 99 (R)

R = H, Ar = p-NO2Ph DiSquare P* 2 99 (R)

R = H, Ar = p-ClPh MorPhos 55 99 (R)

R = H, Ar = Ph Aaphos 10 87 (R)

R = H, Ar = Ph (17) 10 93 (R)

R = H, Ar = m-CO2MePh t-Bu-BisP* 3 97 (S)

R = H, Ar = Ph (18) 20.6 96.5 (S)

F3C CF3 Ph O

O OR P CF3 O P P O P N O O

Ph O PPh2 CF3 DiSquare P* Fe ManniPHOS 17 H O N P PR2 O

O PPh2 P N O PPh2 O Re CO Fe OC CO

AaPHOS (R = Cy) MorPHOS 18

Figure 1.9. Examples of Different Ligands Used in Enamide Reduction.

27

1.9. Conclusion

Chirality plays an important role in nature and almost all biological reactions are highly affected by chirality. Pharmaceutical drugs which were sold as racemate proved to have lower therapeutic activity and more adverse effects compared to their single isomeric analogues. Catastrophic incidence of misuse of drug isomers forced drug regulatory agencies and pharmaceutical companies to focus on developing new drug entities in a single isomeric form. Despite the fact that agrochemicals and other fine chemicals are still marketed as racemate, many alerts suggest that selling these products in a single isomeric form will dramatically reduce the cost and toxicity. The significant importance of chiral agents derived chemists to develop various strategies for their preparation. Asymmetric synthesis is a powerful convenient way for developing new entities of chiral agents. Developing new chiral ligands for organometallic catalyst and new organocatalysts forms the core of organic chemistry research in the last three decades. Chiral amines synthesis is one of the ultimate goals for asymmetric synthesis. Chemoselective and bioselective methodologies have been developed for their synthesis. N-acetyl enamaide hydrogenation has been extensively investigated in the last two decades for α-chiral amine synthesis. It is four steps procedures for the final product with an overall yield not exceeding 50%. Several chiral ligands have been tested with rhodium catalyst for their hydrogenation resulting in 99% enantioselectivity.

1.10. References:

[1] M. Simonyi, Med. Res. Rev. 1984, 4, 359. [2] E. J. Ariens, E. W. Wuis, Clin. Pharmacol. Ther. 1987, 42, 361. [3] K. Mislow,Topics Stereochem. 1999, 22, 1. [4] J. Gal, Enantiomer, 1998, 3, 263. [5] J. B. Biot, Bull. Soc. Philamoth. Paris 1815, 190. [6] a) L. Pasteur, C. R. Acad. Sci. 1848, 26, 535; b)Pasteur, L. Ann. Chim. Physique 1848, 24, 442. [7] a) J. H. van’t Hoff, Arch. Neerl. Sci. Extracts Nat. 1974, 8, 445; b) E. W. Meijer, Angew. Chem. Int. Ed. 2001, 40, 3783-3789.

28

[8] E. L. Eliel, S. H. Wilen, Stereochemistry of organic compounds, Wiley, New York, 1994, pp. 12. [9] M. Nógrádi, Stereoselective Synthesis, Wiley-VCH, Weinheim, Germany, 1995, pp. 1-3. [10] U. Meierhenrich, Amino Acids and the Asymmetry of Life, Springer-Verlag, Berlin Heidelberg, 2008, pp. 17-39. [11] R. A. Sheldon. Chirotechnology, Industrial synthesis of optically active compounds. Marcel Dekker, New York, 1993. [12] M. M. Islam, J. G. Mahdi, I. D. Bowen, Drug Safety 1997, 17, 149. [13] I. Wainer, Drug Stereochemistry: Analytical Methods and Pharmacology, 2nd Ed.; Marcel Dekker: New York, 1993. [14] a)D. Drayer, Clin. Pharmacol. Ther. 1986, 40, 125;b) K. M. Midha, G. McKay, M. J. Rawson, J. W. Hubbard, J. Pharm. Sci. 1998, 87, 797. [15] P. Knightley, H. Evans, E. Potter, M. Wallace, Suffer The Children: The Story of Thalidomide, The Viking Press, New York, 1979. [16] D. D. Leipold, D. Kantoci, Jr. Murray, E. D., D. D. Quiggle, W. J. Wechter, Chirality 2004, 16, 379. [17] C. D. Siebert, A. Hänsicke, T. Nagel, Chirality 2008, 20, 103. [18] S. Stinson, Chem. Eng. News 1998, 76, 83. [19] R. DiCicco, The future of S-(þ)-ibuprofen and other racemic switches, In Chiral’ 93 USA, Spring Innovations: Stockport, UK, 1993. [20] a) S. Houlton, Manufacturing Chemist 2002, 73, 28; b) A. M. Rouhi, Chem & Eng News 2003, 5, 56. [21] E. Francotte , W. Lindner, Chirality in Drug Research, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2006, pp.3-24. [22] F. Leusen, G. Engel, Pharm. Pharmacol. 1999, 51, 1. [23] a) J. Caldwell, Modern Drug Discov. 1999, 2, 51; b) I. Agranat, H. Caner, J. Caldwell, Nat. Rev. Drug Discov. 2002, 21, 753. [24] A.M. Rouhi, Chem. Eng. News 2003, 81, 45. [25] V. A. Davankov, Chirality 1997, 9, 99. [26] R. A. Sheldon, Chirotechnology: Industrial Synthesis of Optically Active Compounds, 2nd Ed., Marcel Dekker, New York, 1993. [26] L. Pasteur, C. R. Hebd. Seances Acad. Sci. 1853, 37, 162. [27] J. F. Larrow, E. N. Jacobsen, Org. Synth. 1998, 75, 1.

29

[28] H. B. Kagan, J. C. Fiaud, Topics in Stereochemistry, E. L. Eliel and S. H. Wilen (Eds.) Vol. 18, Wiley, New York, 1988, pp. 249. [29] Y. Kim, E. T. Choi, M. H. Lee, Y. S. Park, Tetrahedron Lett. 2007, 48, 2833. [30] J. B. Crawford, R. T. Skerlj, G. J. Bridger, J. Org. Chem., 2007, 72, 669. [31] R. Noyori, T. Ikeda, T. Ohkuma, M. Widhalm, M. Kitamura, H. Takaya, S. Akutagawa, N. Sayo, T. Saito, J. Am. Chem. Soc. 1989, 111, 9134. [32] V. Sunjic, F. Kajfez, I. Stromar, N. Blazevic, D. Kolbah, J. Heterocycl. Chem. 1973, 10, 591. [33] Z. -Y. Wei, E. E. Knaus, Tetrahedron 1994, 50, 5569. [34] R. M. Scott, G. D. Armitage, E. Haddock, Optically active 4-phenoxyphenoxypropionic acid derivatives (Shell Internationale Research Maatschappij B. V., Neth.) 1980, 2, 946, 652, pp. 23. [35] A. G. Myers, B. H. Yang, H. Chen, J. L. Gleason, J. Am. Chem. Soc. 1994, 116, 9361. [36] W. S. Knowles, M. J. Sabacky, B. D. Vineyard, D. J. Weinkauf, J. Am. Chem. Soc. 1975, 97, 2567; b) W. S. Knowles, Angew. Chem. Int. Ed. 2002, 41, 1998. [37] R. Noyori, Adv. Synth. Catal. 2003, 345, 15. [38] a) T. Ohta, H. Takaya, M. Kitamura, K. Nagai, R. Noyori, J. Org. Chem. 1987, 52, 3174; b) M. Kitamura, M. Yoshimura, M. Tsukamoto, R. Noyori, Enantiomer 1996, 1, 281. [39] For other synthetic applications, see reviews: a) G. M. Ramos Tombo, G. Bellus, Angew. Chem. Int. Ed. Engl. 1991, 30, 1193; b) A. Börner, J. Holz in Transition Metals for Organic Synthesis, Vol. 2 (Eds.: M. Beller, C. Bolm), Wiley- VCH, Weinheim, 1998, pp.3; c) J. M. Brown in Comprehensive Asymmetric Catalysis, Vol. 1 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999, pp. 121; d) T. Ohkuma, M. Kitamura, R. Noyori in Catalytic Asymmetric Synthesis, 2nd ed. (Ed.: I. Ojima), Wiley-VCH, New York, 2000. [40] a) M. Kitamura, M. Tokunaga, T. Ohkuma, R. Noyori, Tetrahedron Lett. 1991, 32, 4163; b) M. Kitamura, M. Tokunaga, T. Ohkuma, R. Noyori, Org. Synth. 1993, 71, 1; c) K. Mashima, K. Kusano, N. Sato, Y. Matsumura, K. Nozaki, H. Kumobayashi, N. Sayo, Y. Hori, T. Ishizaki, S. Akutagawa, H. Takaya, J. Org. Chem. 1994, 59, 3064. [41] a)T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102, 5974; b) R. M. Hanson, K. B. Sharpless (1986) J. Org. Chem. 1986, 51,1922; c) E. N. Jacobsen, I. Marko , W. S. Mungall, G. Schröder, K. B. Sharpless, (1988) J. Am. Chem. Soc. 1988,110, 1968; d) G. Li, H. T. Chang, K. B. Sharpless, Angew. Chem. Int. Ed. 1996, 35, 451; e) R. A. Johnson, K. B. Sharpless, in Catalytic Asymmetric Synthesis (Ed.: I. Ojima) VCH, Weinheim, 1993. 30

[42] a) J. Blacker, Innovations in Pharmaceutical Technology 2001, 1, 77; b) H. -U. Blaser, F. Spindler, A. Studer, App. Catal. Gen. 2001, 221, 119; c) H. -U. Blaser, M. Eissen, P. F. Fauquex. K. Hungerbuhler, E. Schmidt, G. Sedelmeier, M. Studer, Asymmetric Catalysis on Industrial Scale: Challenges, Approaches, and Solutions; H.-U. Blaser, E. Schmidt Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2004, pp. 91. [43] a) J. Blacker, Innovations in Pharmaceutical Technology 2001, 1, 77. [44] a) P. Kukula, R. Prins, Topics in Catalysis 2003, 25, 29; b) M. Bissen, C. Pinel, Topics in Catalysis 2003, 25, 43; c) T. Valaivan, W. Bhanthumnavin, Y. S. -Anant, Curr. Org. Chem. 2005, 9, 1315. [45] a) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, Adv. Synth. Catal. 2006, 348, 1289; b) T. C. Nugent, Chiral Amine Synthesis - Strategies, Examples, and Limitations. In Process Chemistry in the Pharmaceutical Industry, Second Edition: Challenges in an Ever-Changing Climate, T. F. Braish, K. Gadamasetti Eds.; CRC Press- Taylor and Francis Group: New York, 2008; c) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, WO2006030017, 2006; d) T. C. Nugent, V. N. Wakchaure, A. K. Ghosh, R. R. Mohanty, Org. lett. 2005, 7, 4967; e) T. C. Nugent, A. K. Ghosh, Eur. J. Org. Chem. 2007, 3863. [46] J. Jacques, A. Collet, S. H. Wilen, Enantiomers, Racemates and Resolutions, John Wiley and Sons: New York, 1981. [47] J. K. Whitesell, Chem. Rev. 1989, 89, 1581. [48] E. M. Vogl, H. Groger, M. Shibasaki, Angew. Chem. Int. Ed. 1999, 38, 1570. [49] a) M. Nogradi, Stereoselective Synthsis, 2nd Ed. , VCH: Weinheim, Germany, 1995; b) S. Arrasate, E. Lete, N. Sotomayor, Tetrahedron: Asymmetric 2001, 12, 2077; c) G. M.Coppola, Asymmetric Synthesis: Construction of Chiral Molecules using Amino Acids, Wiley, New York, 1987; d) G. Bringmann, In The Alkaloids, (Ed.: A. Brossi) Acadamic Press, New York, 1986, 141; e) D. Lucet, T. Le Gall, C. Mioskowski, Angew. Chem., Int. Ed. 1998, 37, 2580; f) J. E. Clifton, I. Collins, P. Hallett, D. Hartley, L. H. C. Lunts, P. D. Wicks J. Med. Chem. 1982, 25, 670; g) R. M. Barmore, S. R. Logan, B. C. Van Wagenen, Tetrahedron Lett. 1998, 39, 3451. [50] J. W. Canary, C. S. Allen, J. M. Castagnetto, Y. Wang, J. Am. Chem. Soc. 1995, 117, 8484.

31

[51] a) T. Soe, N. Fukui, T. Hino, M. Nakagawa, Heterocycles 1996, 42, 347; b) D. Xu, K. Prasad, O. Repic, T. J. Blacklock, T. J. Tetrahedron: Asymmetry 1997, 9, 1445; c) C. A. Parrodi, E. Juaristi, L. Quintero, L., A. Clara-Sosa, Tetrahedron: Asymmetry 1997, 7, 1072. [52] J. E. Clifton, I. Collins, P. Hallett, D. Hartley, L. H. C. Lunts, P. D. Wicks, J. Med. Chem. 1982, 25, 670. [53] a) G. Xu, S. R. Gilbertson, Tetrahedron Lett. 2003, 44, 953; b) Y. B. Kim, M. K. Kim, S. H. Kang, T. H. Kim, Synlett 2005, 1995. [54] B. List, R. A. Lerner, C. F. Barbas, J. Am. Chem. Soc. 2000, 122, 2395. [58] M. Yamaguchi, N. Yokota, T. Minami, J. Chem. Soc., Chem. Commun. 1991, 1088. [56] B. List, J. Am. Chem. Soc. 2000, 122, 9336. [57] a) T. C. Nugent, R. Seemeyer, Org. Process. Res. Dev. 2006, 10, 144; b) M. Langlois, C. Meyer, J. L. Soulier, Synth. Commun. 1992, 22, 1895; c) H. Parnes, E. J. Shelton, J. Labelled Compd. Radiopharm. 1996, 38, 19. [58] E. T. Michalson, J. Szmuszkovicz, Prog. Drug. Res. 1989, 33, 135. [59] D. Lucet, T. Le Gall, C. Mioskowski, Angew. Chem., Int. Ed. 1998, 37, 2580. [60] G. M. Coppola, Asymmetric Synthesis: Construction of Chiral Molecules using Amino Acids, Wiley, New York, 1987. [61] a) T. Bunlaksananusorn, F. Rampf, Synlett 2005, 17, 2682; b) N. Ikemoto, D. M. Tellers, S. D. Dreher, J. Liu, A. Huang, N. R. Rivera, E. Njolito, Y. Hsiao, J. C. McWilliams, J. M. Williams, J. D. Armstrong, Y. Sun, D. Mathre, E. J. J. Grabowski, R. D. Tillyer, J. Am. Chem. Soc. 2004, 126, 3048; c) Y. Hsiao, N. R. Rivera, T. Rosner, S. W. Krska, E. Njolito, F. Wang, Y. Sun, J. D. Armstrong, E. J. J. Grabowski, R. D. Tillyer, F. Spindler, C. Malan, J. Am. Chem. Soc. 2004, 126, 9918. [62] a) X. –P. Hu, Z. Zheng, Org. Lett. 2005, 7, 419; b) Y. J. Zhang, K. Y. Kim, J. H. Park, C. E. Song, K. Lee, M. S. Lah, S.-G. Lee, Adv. Synth. Catal. 2005, 347, 563; c) J. You, H.-J. Drexler, S. Zhang, C. Fischer, D. Heller, Angew. Chem., Int. Ed. 2003, 42, 913. [63] P. H. Phau, J. G. de Vries, K. K. Hiia, Adv. Synth. Catal. 2005, 347, 1775. [64] R. Kadyrov, T. H. Reirmeier, U. Dingerdissen, V. I. Tararov, A. Borner, J. Org. Chem. 2003, 68, 4067. [68] A. Menzel, H. Werner, J. Altenbuchner, H. Groeger, Eng. in Life Sci. 2004, 4, 573. [66] a) H. Lebel, K. Huard, Org. Lett. 2007, 9, 639; b) M. Kim, J. V. Mulcahy, C. G. Espino, J. Du Bois, Org. Lett. 2005, 7, 4685; c) C. G. Espino, K. W. Fiori, M. Kim, J. Du Boisn, J. Am. Chem. Soc. 2004, 126, 15378. 32

[67] a) H. Qin, N. Yamagiwa, S. Matsunaga, M. Shibasaki, J. Am. Chem. Soc. 2006, 128, 1611; b) A. Zulys, M. Dochnahl, D. Hollmann, K. Loehnwitz, J.-S. Herrmann, P. W. Roesky, S. Blechert, Angew. Chem., Int. Ed. 2005, 44, 7794. [68] For other methods of enamide preparation, see a) H. B. Kagan, N. Langlois, T. P. Dang, J. Organomet. Chem. 1975, 90, 353; b) D. Sinou, H. B. Kagan, J. Organomet. Chem. 1976, 114, 325; c) T. Morimoto, M. Chiba, K. Achiwa, Chem. Pharm. Bull. 1992, 40, 2894; d) G. R. Lenz, Synthesis 1978, 489; e) D. M. Tsachen, L. Abramson, D, Chai, R. Desmond, U. -H. Dolling, L. Frey, S. Karady, Y. -J. Shi, T. R. Verhoeven, J. Org. Chem. 1995, 60, 4324. [69] a) M. J. Burk, Y. M. Wang, J. R. Lee, J. Am. Chem. Soc. 1996, 118, 5142; b) M. J. Burk, G. Casy, N. B. Johnson, J. Org. Chem. 1998, 63, 6084. [70] a) M. J. Burk, Y. M. Wang, J. R. Lee, J. Am. Chem. Soc. 1996, 118, 5142; b) M. J. Burk, G. Casy, N. B. Johnson, J. Org. Chem. 1998, 63, 6084. [71] a) M. J. Burk, G. Casy, N. B. Johnson, J. Org. Chem. 1998, 63, 6084. b) H. Bernsmann, M. van den Berg, R. Hoen, A. J. Minnaard, G. Mehler, M. T. Reetz, J. G. de Vries, B. L. Feringa, J. Org. Chem. 2005, 70, 943; c) R. Hoen, M. van den Berg, H. Bernsmann, A. J. Minnaard, J. G. de Vries, B. L. Feringa, Org. lett. 2004, 6, 1433. [72] K. Ding, Y. Uozumi, Handbook of Asymmetric Heterogeneous Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA, 2008. [73] a) M. J. Burk, Y. M. Wang, J. R. Lee, J. Am. Chem. Soc. 1996, 118, 5142; b) M. J. Burk, G. Casy, N. B. Johnson, J. Org. Chem. 1998, 63, 6084. [74] M. Kitamura , Y. Hsiao , M. Ohta , M. Tsukamoto , T. Ohta , H. Takaya, R. Noyori, J. Org. Chem. 1994, 59, 297 . [75] J. D. Armstrong, K. K. Eng, J. L. Keller, R. M. Purick, F. W. Hartner, Jr. , W.B. Choi, D. Askin, R. P. Volante, Tetrahedron Lett. 1994, 35, 3239. [76] a) Z. Guoxin, X. Zhang, J. Org. Chem. 1998, 63, 5871; b) W. Tang, X. Zhang, Angew. Chem., Int. Ed. 2002, 41, 1612; c) W. Tang, X. Zhang, Angew. Chem., Int. Ed. 2002, 41, 1612. [77] P.G. Andersson, I. J. Munslow, Modern Reduction Methods, Wiley-VCH, Weinheim, 2008.

33

Chapter 2 Imine Reduction

2.1. Historical View

Reduction of imines with chiral catalysts and hydride source to prepare α-chiral amines with high yield and enantioselectivity represents an important achievement in organic and pharmaceutical chemistry over the last two decades. Different catalytic systems have been developed, heterogeneous and homogenous systems and recently organocatalysis were investigated. Few of these systems were applicable on industrial scale.[1] Historically heterogeneous systems were developed first. Many attempts to attach chiral auxiliary to heterogeneous catalysts (Pt/C, Pd/C and R-Ni) were not successful to attain high enantioselectivity.

Homogenous systems proved to be the systems of choice to reduce imines on the laboratory and on the industrial scales. In general all of these systems require the use of activated imines as substrates in which nitrogen atom is attached to a bulk group (phenyl, phosphinoyl, chiral auxiliary).[2,3] These groups have to be removed in the final step which adds to the total number of steps from the prochiral ketone to the α-chiral amine. Removal of these groups requires harsh conditions which is not compatible with many sensitive groups.

In general cyclic imines are easier in reduction with higher enantioselectivity as they do not have anit/syn conformation.[4] They are considered important intermediates for many pharmaceutical drugs. Acyclic aryl imines were successfully hydrogenated with high enantioselectivities and yields. Metals as Rh, Ru, Ir and Ti were useful in this process. Ir was the best metal for imine reduction with different chiral auxiliaries. Ru catalysts which were developed by Noyori and proved to be highly effective in the reduction of ketones showed limited success. Titanium system which was developed by Buchwald in the nineties gave superior results in terms of yield and enantioselectivity but their industrial application was not that successful.[1] In this section of the thesis, I will focus on the achievments in the field 34

of imine reduction in the past eight years. Of course in the nineties great achievements were accomplished for complete picture please refer to the following review.[2]

2.2. Asymmetric Reduction of N-Phosphinoyl Imines

Of the useful imine substrates examined to date, N-phosphinoyl imines hold the advantage of being reduced with high yields and ees. The steric bulk of the diphenylphosphine group affects the geometric form of imine (only anti isomer is obtained).[5] To access N-phosphinoyl imines researchers universally begin with a ketone and convert it to an oxime (high yield).

Oximes are readily prepared from ketones and HCl.H2NOH, pyridine in ethanol by mixing 1.0 equiv of ketone and 1.1 hydroxyl amine hydrochloride and 1.1 equiv of pyridine.[6]

Treatment of the oxime with chlorodiphenylphosphine [Ph2P(O)Cl] at –45-78 °C provides the N-phosphinoyl imine in mediocre to good yields. For example aryl alkyl N-phosphinoyl imines provide yield in the range of 40-70%, while alkyl alkyl N-phosphinoyl imines provide yields in the range of 50-70%. The product is purified with column chromatography (scheme 2.1).[7]

2.2.1. Synthesis of N-Phosphinoyl Imines.

OH high N Ph P(O)Cl 2 yield Et3N, -45 °C R1 R2 CH2Cl2

P(O)Ph2 P(O)Ph O O Ph N HN 2 NH + reduction high 2 H N P R1 R2 2 Ph R R high yield yield 1 2 R1 * R2 R1 * R2

R1 = aryl, alkyl, heterocylic R2 = alkyl

NOH H NOH O 1- Ph2P(O)Cl, CH2Cl2,-78°C 2 2- hydrolytic work-up HCl

Scheme 2.1. General Strategies for Synthesis of N-Phosphinoyl Imines.

35

Figure 2.1: General Substrates Categories.

P(O)Ph P(O)Ph2 P(O)Ph N 2 N N 2

R 2 1 3

Figure 2.2: General Catalysts Categories.

H3C H t-Bu PR2 O OMe Fe PR2

O P t-Bu N N 2 josiphos type Co O P t-Bu O O O O [Rh(nbd)2]BF4 O OMe (R)-(S)-R2PF-PR2 Catalyst 2 t-Bu t R = cycohexyl, Bu2 2 Catalyst 1 CuCl (R)-(-)-DTBM-SEGPHOS Catalyst 3 R O O N Cl Ph N Ph N NC Re Ir Rh N Cl N OPPh Ph Cl N O 3 H2 Ph Cl H2 R Catalyst 4 Catalyst 5 R= 4-tBu-ph

Catalyst 6 Ph Ph O

N N O PPh2 H H NH HN Ph Ph O PPh2 S S ZnEt2 O Catalyst 7 n n Pd(CF CO ) n=2 3 2 2 S S L-5 (S)-SEGPHOS ZnEt2 Catalyst 8 Catalyst 9

2.2.2. Different Substrates Categories.

Phenyl alkyl N-phosphinoyl imines (Structure 1, 2, 3, figure 2.1) have been extensively investigated over the last few years. We will focus our investigation on the results for the last 36

8 years beginning from the year 2000.Blaser tested Rh-ferrocenyl-catalyst which he used 1.0 mol % of this catalyst (catalyst 1, figure 2.2), 70 bar (1015 psi) of H2, CH3OH at 60 °C over 21 h, the ee was 99% with full conversion (structure 1, figure 2.1).[8] He tested also his system for different substituted phenyl alkyl N-phosphinoyl imines. p-OMe phenyl (62% ee), p-CH3 phenyl (97% ee), p-CF3 phenyl (93% ee) were successfully reduced. For p-Cl phenyl derivative, the ee was only 28 % and improved to 67% with another chiral ligand (structure 3, figure 2.1).

Yamada developed the use of 1.0 mol % of cobalt based catalyst (catalyst 2, figure 2.2), 1.5 equiv NaBH4 in CH3Cl, 0 °C, 4 h, providing 97% isolated yield with 90% ee (structure 1, figure 2.1).[9] Lipshutz developed the use of the DTBM-SEGPHOS ligand with CuCl (catalyst 3, figure 2.2).[10] He used 6.0 mol % of the catalyst, 3.0 equiv tetramethyldisiloxane (TMDS), 6.0 mol % NaOMe, 3.3 equiv t-BuOH, toluene, 25 °C, 17 h, the ee for (structure 1, figure 2.1) was 96% with 99% isolated yield. Cooling the reaction to -25 °C increased the ee to 99% with slightly lower yield (94%) for (structure 2, figure 2.1). Different substituted phenyl alkyl N- phosphinoyl imines were tested. p-Br phenyl (96% ee, 95% yield), p-C3F phenyl (97% ee, 94% yield), p-OMe phenyl (94% ee, 98% yield) were reduced successfully (structure 3, figure 2.1). They were able to reduce sterically hindered imine (phenyl iso-propyl n- phosphinoyl imine) with 94% ee with 90% yield. The ee was improved to 97% ee with 93% yield at -25 °C.

Avecia Limited reported the use of CATHyTM (Catalytic Asymmetric Transfer [11] Hydrogenation) catalysts (catalyst 4-5, figure 2.1). They utilized 24 equiv of Et3N/HCO2H (2:5 ratio) for reduction of phenyl methyl N-phosphinoyl imine (structure 1, figure 2.1) with

86% ee, for 1-acetyl naphthalene derivative the ee was 99% and for 2-octanone derived N- phosphinoyl imine the ee was 95%.

Toste and coworkers developed a highly efficient chiral ligand for rhenium metal.[12] The use of this ligand eliminates the need of restrictive inert condition (open flask technique). Using 3.0 mol % of the catalyst (catalyst 6, figure 2.2), 2.0 equiv of diphenylmethylsilane (DPMS-

H), CH2Cl2, 25 °C over 72 h product ee was provided in >99% albeit in mediocre yield (51%) (structure 1, figure 2.1). They tested other substituted phenyl alkyl N-phosphinoyl imines. 37

Phenyl n-propyl N-suilphinyol imine was reduced with 68% yield and >99% ee. p-OMe phenyl (98% ee, 61% yield), p-CF3 phenyl (98% ee, 78% yield), p-I phenyl (99% ee, 71% yield) methyl N-phosphinoyl imines were reduced. The system was also applicable for heterocyclic derivatives.

The use of Zn/diamine catalyst was reported by Yun.[13] One of the problems related to the use of Zn for the catalytic enantioselective reduction of imines is the strong Zn-N bond formed between Zn and amine product. The source of hydride should affect this bond without affecting the bond between the metal and the diamine. They thought that the choice of the substituent attached to the imine nitrogen will be crucial, so they selected diphenylphoshinoyl moiety. Using 5.0 mol % of the catalyst (catalyst 7, figure 2.2), 3.0 equiv of polymethylhydrosiloxane (PMHS), THF/MeOH, 25 °C, 12 h, the ee was 97% and 86% isolated yield for phenyl methyl N-phosphinoyl imine (structure 1, figure 2.1). For the phenyl ethyl N-phosphinoyl imine, they achieved 96% ee with 82% yield (structure 2, figure 2.1). p- Br phenyl (97% ee, 77% yield) and p-OMe phenyl (96% ee, 83% yield) methyl N- phosphinoyl imines were reduced (structure 3, figure 2.1)

[14] Zhou used Pd(CF3CO2)2/(S)-SEGPHOS for reduction of this category of imines. Using 2.0 mol % of the catalyst (catalyst 8, figure 2.2), 69 bar (1015 psi) of H2, 2,2,2 trifluroethanol, 25 °C, 8-12 h, the ee was 96% with 98% yield for phenyl methyl N-phosphinoyl imine (structure 1, figure 2.1). His catalyst proved to be highly efficient for the reduction of different substituted phenyl methyl N-phosphinoyl imines. p-CH3 phenyl (97% ee, 93% yield), p-F phenyl (94% ee, 87% yield), p-Cl phenyl (94% ee, 90% yield), p-OMe phenyl (96% ee, 96% yield), m-OMe phenyl (96% ee, 97% yield), o-OMe phenyl (99% ee, 80% yield) methyl N- phosphinoyl imines were tested (structure 3, figure 2.1).

Zn-diamino-bis(tert-thiophene) catalyst was tested by Ronchi.[15] Using 5.0 mol % of the catalyst (catalyst 9, frigure 2.2), 5.0 equiv of PMHS, THF/MeOH, 0 °C, 3 h, the ee was 97% and 70% yield for phenyl methyl N-Phosphinoyl imine (structure 2, figure 2.1).

38

2.2.3. Nguyen Special Substrates.

Apart from the classical substrates (structure 1-3, figure 2.1) investigated, substrates which were tested by Nguyen were unique. By today’s standards the use of stoichiometric quantities of a chiral reducing agent are not acceptable, but in this case Nguyen has developed a system capable of accepting a much broader substrate scope and therein lies the significance of his research. Using stoichiometric amounts of (S)-BINOL/AlMe3 with isopropanol as a source of hydrogen to reduce different N-phosphinoyl imines. Subtle difference between small alkyl groups could be distinguished. They reported 93% ee with 85% yield with imine derived from 3-octanone. This class of substrates is synthesized utilizing carbanion chemistry because hydrogen reduction gives low ee (15%). As far as we know this is the only example for reduction of 3-octanonene utilizing hydrogen with such high enantioselectivity. He tested his system for other N-phosphinoyl imines and reported high yields and ees (table 2.1).[16]

Table 2.1. Different substrate Categories introduced by Nguyen

entry imine yield(product) ee(%)

Aryl Alkyl 1 1 Ph Me 85% 96% 2 2 Ph Et 85% 95% 3 3 Ph nPr 84% 94% 4 4 Ph iPr 79% 96% 5 5 1-naphthyl Me 80% 98% 6 6 2-naphthyl Me 84% 96%

P(O)Ph2 7 N 7 84% 94% Ph Me

8 8 80% 94%

39

9 9 84% 94%

10 10 85% 93%

2.3.Asymmetric Reduction of N-aryl Imines

2.3.1. Synthesis of N-Aryl Imines.

N-aryl imines are synthesized from their corresponding ketones and N-aryl amines. They are mixed in anhydrous toluene in the presence of NaHCO3 and 4Å ctivated molecular sieves. The mixture is heated for 12 h at 80 °C. The product is purified by crystallization and distillation (scheme 2.2).[17]

NH 2 R3 O NaHCO3,4ÅMS + N R3 toluene, 80 °C,12h R1 R2 ( mediocre-good yield) R1 R2

R3 reduction HN cerium ammonium nitrate NH2

high yield MeOH/H2O, 0 °C, 6h R * R R1 * R2 1 2 (good-high yield)

Scheme 2.2 General Strategy for Synthesis of N-Aryl Imines.

Figure 2.3: General Imine Structures:

R R R N N N

R1 1 2 3

40

Figure 2.4 General Catalyst structures:

+

O P R Fe - PH N BARF P Ir [{Ir(cod)Cl}2] R Catalyst 2

R= H Catalyst 1 OCH R 3

R R R R P OCH Ir Ph tBuSiO 3 O 2 O O O Ph2P O P O PPh2 P O O R O OSitBuPh O O O O 2 P O O O R R O H3CO [Ir(COD)2]BF4

R=tBu Catalyst 3 OCH3

Catalyst 4

R R Ph P P NH3 H2N S = = O * * R2 BARF H2N Ph P P NH3 P N R R R2 DPEN Ir 1 R DuPHOS RuCl 2 P Cl H2N 1 i 2 R = Pr, R =Ph * Ru * P Cl H N Catalyst 5 2 Catalyst 6

+ Ph2 O BARF- + Ph Ph P N P Ir Ir N BARF- CF SO- N Ir 3 N H O P Ph2 O R R=iPr Catalyst 841 Catalyst 9 (S,R)-15 or (S,S)-15 Catalyst 7 tBu PAr2 + O O Ir Me O P - S N PPh2 P P BARF O t R1 Bu t [{Ir(cod)Cl }] Bu R2 2 Me tBu 1 2 i Catalyst 12 R =Ph,R = Bu 11a, Ar = xyl Catalyst 10 11b, Ar = ph, o-OMe-ph Catalyst 11 BARF + + Ir(COD)

Ph P O 2 OR Ir PF - R 6 Ph2P N N PPh2 - O BARF O P Ir Fe H Ar H Ar R=CH Catalyst 13 3 Catalyst 15 R=Bn,Ar=3,5-DiMe-Ph Catalyst 14

H N CONH Ph N O * CHO Me N N N Cl SiH O 3 O Cl3SiH Catalyst 16 H Cl3SiH Catalyst 18 Catalyst 17

Me H H Me N O N O n PH2 H3C N Me N O n=3 O Me C6F13 O Cl3SiH Me O Cl3SiH H H Catalyst 20 Catalyst 19 t SO2(p- BuPh)

H N N Ph O OH H H H N S N N N N n N O N Ph N H H O AcO Ph O n=2 O O O O H O Cl SiH Cl3SiH H H Cl3SiH 3 Cl3SiH F Catalyst 21 Catalyst 22 Catalyst 23 Catalyst 24 O O S S N n N H H n=5 Cl SiH 3 Catalyst 25

42

2.3.2. Different Substrates Categories

Several organometallic and organocatalytic systems were developed for the reduction of phenyl methyl (ethyl) N-aryl imines. The hydrogen source is either molecular hydrogen or hydride reagents. Transition metals having chiral ligands on Ir, Ru, Rh and Ti were used and resulted in high yield and selectivity.

Pfaltz and Leitner used cationic Ir complexes with chiral phosphinodihydrooxazoles modified with perfluroalkyl groups.[18] Using 0.09 mol % of the catalyst (catalyst 1, figure 2.4), 30 bar

(435 psi) H2, supercritical carbon dioxide (scCO2) at 40 °C an ee of 80% was accomplished with complete conversion (structure 1, figure 2.2). The choice of counter ion dramatically influenced selectivity with tetrakis-3,5-bis(trifluoromethyl)phenylborate anion (BARF), resulting in the highest selectivity. They later developed the use of scCO2 with ionic liquids and obtained the same result.[19]

Zhang and Xiao reported one of the earliest examples for the efficient reduction of aryl methyl N-aryl imines utilizing iridium. They introduced the use of air stable Ir- bisphospahnoferrocene catalyst (catalyst 2, figure 2.4).[20] Using 2.0 mol % of the catalyst

(catalyst 2, figure 2.4), 70 bar (1015 psi) of H2, CH2Cl2, 25 °C over 44 h, 99% ee with 77% conversion for phenyl methyl N-aryl imines (structure 1, figure 2.3). For p-OMe-phenyl methyl N-aryl imines the ee was 98% with 77% conversion and for p-CF3-phenyl methyl N- aryl imine the ee was 99% with 80% conversion (structure 3, figure 2.3).

Claver and Castillón introduced the use of sugar derived diphosphite ligands.[21] Using 1.0 mol % of the iridium catalyst (catalyst 3, figure 2.4), 10 bar (145 psi) H2, CH2Cl2, 25 °C, 18 h an ee was 57% with 83% conversion for phenyl methyl N-aryl imine (structure 1, figure 2.3).

The use of 4.0 mol % Bu4NI improved conversion (100%) but lowered the ee (46%) at 70 bar

(1015 psi) of H2. Later they reported the use of other diphosphinite ligands (catalyst 4, figure

2.4). Using 1.0 mol % of the catalyst, 70 bar ( 1015 psi) H2, CH2Cl2, 25 °C, 16 h, the ee was 70% with complete conversion.[22]

43

Cozzi et al. developed the use of phosphino derived ligands.[23] Using 0.1 mol % of the catalyst (catalyst 5, figure 2.4), 50 bar (725 psi) H2, CH2Cl2, 25 °C, 4 h, the ee was 86% with complete conversion for phenyl methyl N-aryl imine (structure 1, figure 2.3).

Ruthenium catalysts earlier developed by Noyori for ketone reduction were useful for imine [24] reduction which was tested by Cobley. Using 1.0 mol % of RuCl2 (diphosphine) (diamine)

(catalyst 6, figure 2.4), 15 bar (218 psi) H2, 100 mol % of t-BuOK in t-BuOH for in situ activation of the catalyst, 65 °C, 20 h, the ee was 91% with complete conversion for phenyl methyl N-aryl imine (structure 1, figure 2.3).

Grützmacher was successful in using mixed phosphane olefin ligand for imine reduction.[25]

Using 1.0 mol % of the iridium catalyst (catalyst 7, figure 2.4), 50 bar (725 psi) H2, CHCl3, 50 °C, 2 h an ee of 86% with >98% yield for phenyl methyl N-aryl imine was reported (structure 1, figure 2.3).

Niedercorn was able to reduce N-aryl imines with Ir-aminophosphine- derived catalyst.[26] Using 2.0 mol % of the catalyst (catalyst 8, figure 2.4), 20-50 bar (290-725 psi)

H2, CH2Cl2, 12 h an ee was 90% with full conversion for phenyl methyl N-aryl imine was reported (structure 1, figure 2.3).

Andersson developed a new class of chiral phosphine-oxazoline ligands for iridium imine [27] reduction. Using 0.5 mol % of the catalyst (catalyst 9, figure 2.4), 20 bar (290 psi) of H2,

CH2Cl2, 25 °C over 2 h an ee was 90% with 98% conversion for phenyl, methyl N-aryl imines (structure 1, figure 2.3). He also tested his catalyst for reducing p-fluoro phenyl methyl N-aryl imine and reported 89% ee in 2 h, for p-OMe phenyl methyl N-aryl imine the ee was 86% within 2-3 h, for p-chloro phenyl methyl N-aryl imine the ee was 89% within 1.5 h with full conversion. In case of o-Me phenyl methyl N-aryl imine the ee was lower (83%) and the conversion was much lower (52%) after even 12 h (structure 3, Figure 2.3). Later they reported 78% ee for phenyl ethyl N-aryl imines (structure 2, figure 2.3). 2-naphthyl methyl N-aryl imine was reduced with 91% ee.[28]

Blom prepared a new class of diphenylphosphanyl sulfoximines ligands.[29] Using 1.1 mol % of the Ir-Sulxoimine catalyst (catalyst 10, figure 2.4), 2.0 mol % of iodine, 20 bar (290 psi) of 44

H2, toluene, 25 °C, 4 h an ee of 96% with full conversion was reported for phenyl methyl N- aryl imine and 92% ee for phenyl ethyl N-aryl imnes (structure 1,2 , figure 2.3). Reducing the catalyst loading to 0.5 mol % resulted in the same enantioselectivity. Lower catalyst loading (0.1 mol %) the hydrogen pressure had to increase to 50 bar to achieve the same ee with full conversion. He tested his system for substituted phenyl methyl N-aryl imines. The ee was 96% for p-Me phenyl methyl N-aryl imine, for m-Me phenyl methyl N-aryl imine the ee was 93% and for o-Me phenyl methyl N-aryl imine the ee was 94%. For o-OMe phenyl methyl N- aryl imine the ee was 90%, for p-OMe phenyl methyl N-aryl imine the ee was 94% and for m-OMe phenyl methyl N-aryl imine the ee was 96% with full conversion in all cases (structure 3, figure 2.3).

Pizzano tested Ir-phosphine–phosphites based catalysts for reduction of N-aryl imines.[30]

Using 1.0 mol % of the catalyst (catalyst 11a, figure 2.4), 30 bar (436 psi) of H2, CH2Cl2, 25 °C, 24 h an ee of 82% with complete conversion for phenyl methyl N-aryl imine was achieved (structure 1, figure 2.3). Later he used another derivative of the catalyst (catalyst 11b, figure 2.3) for reduction of substituted phenyl methyl N-aryl imine. For p-methyl phenyl methyl N-aryl imine the ee was 72%, for p-OMe phenyl methyl N-aryl imine the ee was 85%, for p-fluoro phenyl methyl N-aryl imine the ee was 79% and for p-chloro phenyl methyl N- aryl imine the ee was 82%.[31]

Imamoto reduced N-aryl imines utilizing Ir-phosphine catalyst.[32] Using 0.5 mol % of the catalyst (catalyst 12, figure 2.4), 1 bar (14.5 psi) of H2, CH2Cl2, 25 °C, 1.5 h, the ee was 99% with 95% isolated yield for phenyl methyl N-aryl imine (structure 1, figure 2.3). They tested reduction of p-OMe phenyl methyl N-aryl imine imines with 83% ee and 98% yield within 2 h. For p-fluoro phenyl methyl N-aryl imine the ee was 84% with 92% yield within 1.5 h (structure 3, figure 2.3).

Dervisi synthesized a new Ir diphosphine catalyst {[(Ir (ddppm)-(COD)]PF6} and tested it for the N-aryl imine reduction.[32] Using 1.0 mol % of the catalyst (catalyst 13, figure 2.4), 1 bar

(14.5 psi) of H2, ClCH2CH2Cl, 25 °C, 24 h, the ee was 84% with 99% yield for phenyl methyl N-aryl imine (structure 1, figure 2.3). Operating at atmospheric pressure allowed the hydrogenation to be carried using Schlenk technique instead of high pressure autoclaves. They expanded their investigation to include p-chloro phenyl methyl N-aryl imines which 45

were reduced with 80% ee with 99% yield. For p-OMe phenyl methyl N-aryl imines the ee was 81% with 100% yield (structure 3, figure 2.3).

New chiral phosphine oxazoline ligands was prepared by Zhou for the Ir reduction of imines [33] (Ir-SIPHOX) (catalyst 14, figure 2.4). Using 1.0 mol % of the catalyst, 1 bar (14.5) of H2, t-butyl methyl ether (TBME), 4 Å MS, 10 °C over 20 h, the ee was 93% with complete conversion for phenyl methyl N-aryl imine (structure 1, figure 2.3). He investigated the application of his catalyst on the reduction of different substituted phenyl methyl N-aryl imine, for p-Me the (94% ee), p-Cl (90% ee), p-Br (91% ee), m-Cl (93% ee), m-Br (92% ee) phenyl methyl N-aryl imine derivatives with full conversion in all cases. For 3,4- Di-Me phenyl methyl N-aryl imine the ee was 94% (structure 3, figure 2.3).

New ferrocenyl P,N-ligands was introduced by Knochel and used iridium for the reduction of N-aryl imines.[34] Using 1.0 mol % of the catalyst (catalyst 15, figure 2.4), 10 bar (145 psi) of

H2, toluene/ MeOH (4:1) at 25 °C over 2 h, the ee was 94% with full conversion for phenyl methyl N-aryl imine(structure 1, figure 2.3). He also tested his catalyst for the reduction of substituted phenyl methyl N-aryl imine. p-Ph and p-Cl (92% ee), m-Me (93% ee), o-Me (94% ee), m-F (93% ee) and p-CF3 (89% ee) phenyl methyl N-aryl imine were reduced with high ees (structure 3, figure 2.3). 2-naphthyl methyl N-aryl imine was reduced with 93% ee.

For the reduction of this category of chiral imines organocatalytic methods have proved to be highly effective. Different organocatalysts have been developed utilizing various silane derivatives or Hantzsch as a source of hydride. Although these sources of hydrides are not atom economic, they are commercially available in large quantities at rather reasonable prices and offer the potential of chemoselectivity not possible in the presence of H2.

In 2001 Matsumura and coworkers developed the use of proline (catalyst 16, figure 2.4) derivatives for the hydrosilylation of imines. They achieved mediocre enantioselectivity reporting 66% ee with 52% yield for phenyl methyl N-aryl imine (structure 1, figure 2.3).[35] Inspired by the research of Matsumura, Kočovský and coworkers developed the use of valine derived (formamide/ based) catalysts for the hydrosilylation of N-aryl imine.[36] Using

10 mol % of the catalyst (catalyst 17, figure 2.4), 1.5 equiv of Cl3SiH, CHCl3, -20 °C, 16 h, the ee was 92% with 94% isolated yield for phenyl methyl N-aryl imine (Structure 1, Figure 46

1). They described the role of each in the catalyst and its importance in controlling stereoselectivity. They also marked the important structural features in the imine which controls the total outcome of the reaction. They tested their system also for the reduction of substituted phenyl methyl N-aryl imines. p-OMe (85% ee, 86% yield), p-CF3 (89% ee, 86% yield), o-Me (92% ee, 90% yield) phenyl, methyl N-aryl imines were successfully reduced (structure 3, figure 2.3).

In 2007 he reported the use of his catalyst with fluorous tag for imine reduction.[37] Fluorous tags are used to enable recycling of the catalyst. He was able to reuse the catalyst 4-5 times without significant loss of enantioselectivity and yield. Using 10 mol % of valine derived catalyst (catalyst 19, figure 2.4), 2.0 equiv of HSiCl3, toluene, 18 °C, 16 h, the ee was 90% with 98% yield for phenyl methyl N-aryl imine (structure 1, figure 2.3). p-CF3 (92% ee, 72% yield, 10 °C) and p-OMe (84% ee, 84% yield) phenyl methyl N-aryl imines were reduced with high yields and ees (structure 3, figure 2.3). 2-naphthyl methyl N-aryl imine was reduced with 92% ee and 93% yield.

In 2008 he reported the use of his catalyst with polymer support, which can be used up to 5 times without significant loss of catalyst activity.[38] Using 15 mol % of the catalyst (catalyst

20, figure 2.4), 2.0 equiv Cl3SiH, CH3Cl, 25 °C, 16 h, the ee was 82% with 84% yield for phenyl methyl N-aryl imine (structure 1, figure 2.3). He expanded his study to cover other substituted phenyl methyl N-aryl imines. p-OMe (77% ee, 63% yield) and p-CF3 (81% ee, 67% yield) phenyl methyl N-aryl imines were reduced. Also 2,5 Me-3-furyl phenyl, methyl N-aryl imine was reduced with 78% ee and 67% yield (structure 3, figure 2.3).

In 2006 he introduced the use of oxazoline catalyst for reduction of N-aryl imines.[39] Using

20 mol % of the catalyst (catalyst 18, figure 2.4), 2.0 equiv of HSiCl3, CHCl3, –20 °C, 24 h, resulted in 87% ee and 65% yield phenyl methyl N-aryl imine (structure 1, figure 2.3). p-

OMe (87% ee, 51% yield) and p-CF3 (87% ee, 65% yield) phenyl methyl N-aryl imines were reduced successfully (structure 3, figure 2.3).

The group of Sun also prepared several organocatalysts for the hydrosilylation of imines. In 2006, they tested pipecolinic acid derived formamides catalyst.[40] Using 10 mol % of the catalyst (catalyst 21, figure 2.4), 2.0 equiv of Cl3SiH, CH2Cl2, 0 °C, 16 h, the ee was 95% 47

with 97% yield for phenyl methyl N-aryl imine (structure 1, figure 2.3). They also tested their catalyst for different substituted phenyl methyl N-aryl imines. p-OMe (93% ee, 95% yield), p-Br (95% ee, 98% yield), m-Br (94% ee, 82% yield), p-CF3 (96% ee, 85% yield), para-NO2 (95% ee, 96% yield) phenyl methyl N-aryl imines were reduced successfully (structure 3, figure 2.3). 2-naphthyl methyl N-aryl was reduced with (93% ee, 92% yield) and p-methoxy substituted naphthyl methyl N-aryl imine (91% ee, 97% yield).

They also reported the use of formamide derivative of piperazine .[41] Using 10 mol % of the catalyst (catalyst 22, figure 2.4), 2.0 equiv Cl3SiH, CH2Cl2, -20 °C, 48 h, resulted in 89% ee with 95% yield for phenyl methyl N-aryl imine (structure 1, figure 2.3) and for phenyl ethyl N-aryl imine the ee was 94% with 92% yield (structure 2, figure 2.3). p-

NO2 (90% ee, 99% yield), p-Br (89% ee, 81% yield) and p-Me (85% ee, 71% yield) phenyl methyl N-aryl were successfully reduced. 2-naphthyl (88% ee, 63% yield) and 6-OMe 2- naphthyl methyl N-aryl imine (85% ee, 64% yield) were reduced. p-F (95% ee, 87% yield), p-Cl (94% ee, 83% yield), p-Br (95% ee, 89% yield), p-Me (88% ee, 87% yield) and p-OMe (90% ee, 83% yield) phenyl ethyl N-aryl imine were reduced with high yields and ees. Phenyl n-propyl N-aryl imine was reduced with 90% ee and 88% yield and phenyl n-butyl N- aryl imine was reduced with 89% ee and 84% yield.

They were also successful in using chiral sulfinamides based on Ellman auxiliary [(R)-tert- butansulfinamide.[42] Using of 20 mol % of the catalyst (catalyst 23, figure 2.4), 2.0 equiv

Cl3SiH, CH2Cl2, -20 °C, 24 h, the ee was 92% with 92% yield for phenyl methyl N-aryl imine (structure 1, figure 2.3). p-OMe (93% ee, 98% yield), p-Br (92% ee, 92% yield), p-NO2

(90% ee, 94% yield) and p-CF3 (92% ee, 93% yield) phenyl methyl N-aryl imines were reduced (structure 3, figure 2.3).

In 2007 they reported the use of proline derived tetramide catalyst for imine reduction.[43]

Using 10 mol % of the catalyst (catalyst 24, figure 2.4), 2.0 equiv Cl3SiH, CH2Cl2, 0 °C, 16 h, the ee was 77% with 93% yield for phenyl methyl N-aryl imines. For substituted phenyl methyl imines the ees were lower compared with his other catalytic systems (structure 1, figure 2.3)

48

More recently, 2008, he described the use of S-Chiral Bissulfinamide catalyst for imine [44] reduction. Using 10 mol % of the catalyst (catalyst 25, figure 2.4), 2.0 equiv Cl3SiH, 0.3 equiv 2,6-lutidine, CH2Cl2, -20 °C, 24 h, the ee was 96% with 91% yield for phenyl methyl N-aryl imine. (structure 1, figure 2.3). p-OMe (95% ee, 83% yield), p-Br (95% ee, 92% yield), p-NO2 (93% ee, 90% yield) and p-CF3 (95% ee, 95% yield) phenyl methyl N-aryl imines were reduced (structure 3, figure 2.3).

2.4. Reduction of Miscellaneous Imines:

Other imine derivatives were reduced as N-benzyl and N-tosyl imines. These substrates were less investigated compared to the previous discussed substrates in the last 10 years. For further reading please consult the following literatures.[45] Oximes and were also tested but with fewer examples over the last 10 years. For further reading please consult the following literatures.[46] Imines with different chiral auxiliaries were reduced in good to high enantioselectivity. For further reading please consult the following references.[47] Cyclic imines were extensively investigated and extensively reviewed in book chapters and published reviews. For further reading please consult the following references.[48]

2.5 Conclusion.

Imine reduction has been extensively investigated by many groups over the last forty years. During the last two decades several milestones have been achieved in asymmetric imine reduction. Imines are reduced with high enantioselectivity and yield. This methodology is highly efficient for obtaining α-chiral amines in 99% enantioselectivity but the over all yield from the starting material to the final product is usually low and below 50%. Removal of the auxiliary usually requires harsh acidic conditions which may not compatible with different acid sensitive groups. Despite these drawbacks the high enantioselectivity obtained makes this method attractive for further improvements.

49

2.6. References.

[1] a) P. G. Andersson, I. J. Munslow in Modern Reduction Methods, Wiley-VCH, Weinheim, 2008; b) J. G. De Vries, C. J. Elsevier in The Handbook of Homogeneous Hydrogenation, Wiley-VCH, Weinheim, 2008. [2] S. Kobayashi, H. Ishitani, Chem. Rev. 1999, 99, 1069. [3] J. A. Ellman, T. D. Owens, T. P. Tang Acc. Chem. Res. 2002, 35, 984. [4] E. N. Jacobsen, A. Pfaltz, H. Yamamoto, Comprehensive Asymmetric Catalysis I–III, Springer-Verlag Berlin Heidelberg, 2000. [5] F. Spindler, H. -U. Blaser, Adv. Synth. Catal. 2001, 343, 68. [6] B. Krzyzanowska, W. J. Stec, Syntheis 1982, 270. [7] a) T. Yamada, T. Nagata, K. D. Sugi, K. Yorozu, T. Ikeno, T. Ohtsuka, D. Miyazaki, T. Mukaiyama, Chem. Eur. J. 2003, 9, 4485. [8] F. Spindler, H. -U. Blaser, Adv. Synth. Catal. 2001, 343, 68. [9] Yamada, T. Nagata, K. D. Sugi, K. Yorozu, T. Ikeno, Y. Ohtsuka, D. Miyazaki, T. Mukaiyama, Chem. Eur. J. 2003, 9, 4485. [10] H. Lipshutz, H. Shimizu, Angew. Chem. Int. Ed. 2004, 43, 2228. [11] a) Transfer hydrogenation process. Martin, J.; Campbell; L. A.; Avecia Limited, Great Britain. Patent publication number WO0112574, 2001; b) Transfer hydrogenation process. Martin, J.; Campbell, L. A.; Avecia Limited, Great Britain. Patent publication number US6696608, 2004. [12] K. A. Nolin, R. W. Ahn, F. D. Toste, J. Am. Chem. Soc. 2005, 127, 12462. [13] B. M. Park, S. Mun, J. Yun, Adv. Synth. Catal. 2006, 348, 1029. [14] a)Y. Q. wang, Y. G. Zhou, Synlett 2006, 8, 1189; b) Y. Q. Wang, S. M. Lu, Y. G. Zhou, J. Org. Chem. 2007, 72, 3729. [15] M. Bandini, M. M. Melucci, F. Piccinelli, R. Sinisi, S. Tommasi, A. U. Ronchi, Chem. Commun. 2007, 4519. [16] C. R. Graves, K. A. Scheidt, S. T. Nguyen, Org. Lett. 2006, 8, 1229. [17] Z. Wang, X. Ye, S. Wei, P. Wu, A. Zhang, J. Sun, Org. Lett. 2006, 8, 999. [18] S. Kainz, A. Brinkmann, W. Leitner, A. Pfaltz, J. Am. Chem. Soc. 1999, 121, 6421. [19] M. Solinas, A. Pfaltz, P. G. Cozzi, W. Leitner, J. Am. Chem. Soc. 2004, 126, 16142. [20] D. Xiao, X. Zhang, Angew. Chem. Int. Ed. 2001, 40, 3425. [21] E. Guiu, B. Munoz, S. Castillón, C. Claver, Adv. Synth. Catal. 2003, 345, 169. 50

[22] E. Guiu, M. Aghmiz, Y. Diaz, C. Claver, B. Meseguer, C. militzer, S. Castillón. Eur. J. Org. Chem. 2006, 627. [23] P. G. Cozzi, F. Menges, S. Kaiser, Synlett 2003, 6, 833. [24] C. J. Cobley, J. P. Henschke, Adv. Synth. Catal. 2003, 345, 195. [25] P. Maire, S. Deblon, F. Breher, J. Geier, C. Böhler, H. Rüegger, H. Schönberg, H. Grützmacher, Chem. Eur. J. 2004, 10, 4198. [26] C. Blanc, F. A. Niedercorn, G. Nowogrocki, Tetrahedron: Asymmetry 2004, 15, 2159. [27] A. Trifonova, J. S. Diesen, C. J. Chapman, P. G. Andersson, Org. Lett. 2004, 6, 3825. [28] A. Trifonova, J. S. Diesen, P. G. Andresson, Chem. Eur. J. 2006, 12, 2318. [29] C. Moessner, C. Bolm, Angew. Chem. Int. Ed. 2005, 44, 7564. [30] S. Vargas, M. Rubio, A. Suárez, A. Pizzano, Tetrahedron Lett. 2005, 46, 2049. [31] S. Vargas, M. Rubio, A. Suárez, D. del Río, E. Álvarez, A. Pizzano, Organometallics 2006, 25, 961. [32] T. Imamoto, N. Iwadate, K. Yoshida, Org. lett. 2006, 8, 2289. [32] A. Dervisi, C. Carcedo, L. L. Ooi, Adv. Synth. Catal. 2006, 348, 175. [33] S. F. Zhu, J. B. Xie, Y. Z. Zhang, S. Li, Q. L. Zhou, J. Am. Chem. Soc. 2006, 128, 12886. [34] M. N. Cheemala, P. Knochel, Org. Lett. 2007, 9, 3089. [35] F. Iwasaki, O. Onomura, K. Mishima, T. Kanematsu, T. Makib, Y. Matsumura, Tetrahedron Lett. 2001, 42, 2525. [36] a) A. V. Malkov, A. Mariani, K. N. MacDougall, P. Kočovský, Org. Lett. 2004, 6, 2253; b) A. V. Malkov, S. Stoncius, K. N. MacDougall, A. Mariani, G. D. McGeoch, P. Kočovský, Tetrahedron 2006, 62, 264. [37] A. V. Malkov, M. Figlus, S. Stoncius, P. Kočovský, J. Org. Chem. 2007, 72, 1315. [38] A. V. Malkov, M. Figlus, P. Kočovský, J. Org. Chem. 2008, 73, 3985. [39] A. V. Malkov, A. J. P. S. Liddon, P. R. Lopez, L. Bendova, D. Haigh, P. Kočovský, Angew. Chem. Int. Ed. 2006, 45, 1432. [40] L. Zhou, Z. Wang, S. Wei, J. Sun, Chem. Commun., 2007, 2977. [41] Z. Wang, M. Cheng, P. Wu, S. Wei, J. Sun, Org. Lett. 2006, 8, 3045. [42] D. Pei, Z. Wang, S. Wei, Y. Zhang, J. Sun, Org. Lett. 2006, 8, 5913. [43] Z. Wang, S. Wei, C. Wang. J. Sun, Tetrahedron: Asymmetry 2007, 18, 705. [44] D. Pei, Y. Zhang, S. Wei, M. Wang, J. Sun, Adv. Synth. Catal. 2008, 350, 619.

51

[45] a) A. Ros, A. Magriz, H. Dietrich, Mark Ford, R. Fernández, J. M. Lassalettaa, Adv. Synth. Catal. 2005, 347, 1917; b) J. B. Åberg, J. S. M. Samec, J.E. Bäckvall, Chem. Commun. 2006, 2771; c) M. T. Reetz, O. Bondarev, Angew. Chem. Int. Ed. 2007, 46, 4523; d) Q. Yang, G. Shang, W. Gao, J. Deng, and X. Zhang, Angew. Chem. Int. Ed. 2006, 45, 3832. [46] a) I. Takei, Y. Nishibayashi, Y. Ishii, Y. Mizobe, S. Uemura, M. Hidai, Chem Commun. 2001, 2360; b) E. Fontaine, C. Namane, J. Meneyrol, M. Geslin, L. Serva, E. Roussey, S. Tissandié, M. Maftouh, P. Roger, Tetrahedron: Asymmetry 2001, 12, 2185; c) M. P. Krzemiński, M. Zaidlewicz, Tetrahedron: Asymmetry 2003, 14, 1463; d) X. Huang, M. O.- Marciales, K. Huang, V. Stepanenko, F. G. Merced, A. M. Ayala, W. Correa, M. De Jesús, Org. Lett. 2007, 9, 1793; e) M. Sugiura, S. Kobayashi, Angew. Chem. Int. Ed. 2005, 44, 5176. [47] a) G. Chelucci, S. Baldino, R. Solinasa, W. Barattab, Tetrahedron Lett. 2005, 46, 5555; b) X. Xiao, H. Wang, Z. Huang, J. Yang, X. Bian, Y. Qin, Org. lett. 2006, 8, 139; c) C. Cimarelli, G. Palmieri, Tetrahedron: Asymmetry 2000, 11 2555; d) B. TöröK, G. K. S. Prakash, Adv. Synth. Catal. 2003, 345, 165. [48] a) M. Rueping, A. P. Antonchick, T. Theissmann, Angew. Chem. Int. Ed. 2006, 45, 6751; b) M. Rueping, A. P. Antonchick, T. Theissmann, Angew. Chem. Int. Ed. 2006, 45, 3683; c)M. Rueping, A. P. Antonchick, Angew. Chem. Int. Ed. 2007, 46, 4562; d) P. Roszkowski, Z. Czarnockia, Mini-Reviews inorganic Chemistry 2007, 4, 190; e) Y.-Q. Wang, C. –B. Yu, D.-W. Wang, X. –B. Wang, Y. –G. Zhou, Org. Lett. 2008, 10, 2071.f) J. Li, Y. Zhang, D. Han, Q. Gao, C. Li, Journal of Molecular Catalysis A: Chemical 2009 298, 31.

52

Chapter 3 Reductive Amination

3.1. Historical View:

3.1.1. Reductive Amination Utilizing Heterogeneous Catalyst:

Reductive amination can be defined as the reaction of or ketone with ammonia or with a primary or a secondary amine to give alkylated amines in the presence of a catalyst under hydrogenation conditions. Reductive amination was first described in the early days of twentieth century by Mignonac.[1] Since then it was widely applied for the preparation of different types of amines. The process could be described as reductive of ammonia or reductive amination of aldehydes or ketones.[2]

Reductive amination is a powerful methodology for the synthesis of primary, secondary or tertiary amines. Its main advantage is being a stepwise efficient methodology. It is only two steps from the carbonyl compound to the primary amine. Due to the significance of this transformation it was subjected to an intensive investigation by different research groups. Several obstacles appeared on the surface during developing this methodology. Formation of alcohol as a main by product resulting in lower yield was one of the major problem. Over alkylation of the amine was an additional problem. Also side reactions, as aldol condensation which leads to the formation of side products and lower the overall yield was another major concern. Despite the fact that most of these problems were solved over the last three decades, I will try to describe briefly the historical development of this methodology and the problems associated with its development.

The introduction of catalysis in organic chemistry was the most important breackthrough in the field in the twentieth century. The use of minute quantity of a catalyst to accelerate the required reaction, accumulate the product and inhibit side reactions were the main objectives of the catalyst use. Historically heterogeneous catalysts were first introduced in the main

53

stream of organic chemistry. Hydrogenation catalysts as nickel, palladium platinum and other transition metal catalysts received the maximum attention. These catalysts were initially introduced for alkene hydrogenation and they were later tested for other transformations as reductive amination.[3]

Reductive alkylation of ammonia was one of the earliest examples described in literatures. It generally proceeds under mild conditions using heterogeneous catalyst. The reductive alkylation of ammonia with carbonyl compounds may produce primary, secondary, and tertiary amines, as well as alcohol as a side product. The origin of product selectivity depends primarily on the molar ratio of carbonyl compound to ammonia, the nature of catalyst and structure of the carbonyl compound. The reaction of benzaldehyde in the presence of 1.0 equivalent of ammonia in ethanol over Raney Ni gave benzylamine in an 89.4% yield while with 0.5 equivalent of ammonia dibenzylamine was obtained in an 80.8% yield.[4]

Reductive amination of aliphatic aldehydes having α-hydrogen atoms, especially of the type

RCH2CHO, usually results in lower yields due to the formation of by products through aldol or other condensation reactions. Also lower aliphatic aldehydes usually produce mixture of primary, secondary, and tertiary amines. The reaction of butyraldehyde with 0.5 equivalent of ammonia over Raney Ni also resulted in a mixture of 31% of butylamine, 17% of dibutylamine, and 8% tributylamine.[5] Higher aldehydes usually react selectively with ammonia producing less by products.[6]

The reductive alkylation of ammonia with ketones is performed under conditions similar to those for aldehydes, but appears to proceed with more difficulty. Initially, reductive amination of ketones with ammonia was tested without any additives resulting in lower yields.[7,8] Primary amines are considered better neocluophihes compared to ammonia. Despite their higher nucleophilicity they are more sterically hindered. They were tested in reductive amination of carbonyl compounds utilizing different heterogeneous catalysts as nickel, platinum oxide, platinum sulphide and nickel sulphide. The following examples are some early trials for the preparation of secondary amines from primary amines (scheme 3.1).[9-12]

Ni-kieselguhr NH2 CH(CH2)2CHO o NHC4H9 125 C,100 bar H2,1h (0.5mol) (0.55mol) 54 (91%)

Pt oxide* CH CO(CH ) CH H NCH CH OH 3 2 5 3 2 2 2 C6H13CH(CH3)NHCH2CH2OH 100ml EtOH (1.3mol) (1mol) (96%) RT, 1-2 bar H2,7h *Prereduced in 50ml EtOH at 1 bar H2 O

Ni sulfide* NH2 o . 180 C,100-120 bar H2,14h NH *Supported on montmorillonite (15% Ni) (1.0mol) (1.90mol) (95.5%)

Pt sulfide-C PhHN NH2 MeCOCH2CHMe2 PhHN NHCH(CH3)CH2CHMe2 o 175-180 C, 30-40 bar H2,4.5h (0.86mol) (0.95mol) (99%)

Scheme 3.1. Preparation of Secondary Amines

In their attempts to overcome the problems associated with reductive amination, scientists tested the effect of additives on this transformation. It was found that the addition of a small quantity of Brønsted acid improved the yield dramatically. Dialkyl ketones, especially sterically hindered ones, tended to produce the corresponding alcohols to significant extents under the conditions of reductive amination decreasing the overall yield of the amine. The addition of a small amount of acetic acid or ammonium acetate is effective in suppressing alcohol formation. Thus, the formation of 2-nonanol could be depressed effectively in the presence of ammonium acetate in the reductive amination of 2-nonanone (scheme 3.2).[3] CH OC H 10 mL EtOH/0.8 g (0.047mol) NH3 3 7 15 o CH3CH(NH2)C7H15 CH3CHOHC7H15 50 C, 80 bar H2,Ra-Ni,Brønstedacid 100% 0%

Scheme 3.2. Reductive Amination of 2-nonanone

3.1.2. Reductive Amination Utilizing Homogenous Catalysis:

As mentioned before heterogeneous catalysts were first utilized for reductive amination. After the introduction of Wilkinson catalyst which opened the door for the use of homogenous catalysts in organic synthesis. Interest has been expressed in the use of homogeneous catalysts for reductive amination. Bakos was the first to utilize homogenous

55

catalysts for reductive amination in 1974. He tested rhodium and cobalt based catalysts for the reductive alkylation of ammonia and aniline derivatives. He found that the activity of cobalt catalyst is highly influenced by the structure of phosphine ligand. Also he recognized that using basic aliphatic amines led to poising of the cobalt catalyst and no product was formed. On the other hand rhodium was used successfully in the reductive amination of these basic amines.[13] Despite these interesting results the reaction conditions were harsh (100–300 bar H2, 110–200 °C) and no turnover number were reported. In 2000, Börner described more practical system for homogenous reductive amination.[14] Benzaldehyde and piperidine could be reductively aminated using [Rh(dppb)(COD)]BF4 or [Rh(1,2-bis- diphenylphosphinitoethane)(COD)]BF4 under milder conditions (50 bar H2, room temperature) with 500 TON (scheme 3.3).

O H 0.2% Rh cat H N N PhCH2OH H2 (50 bar) rt, <20 h

Scheme 3.3. Reductive Amination of Benzaldehyde with Piperidine.

Beller in 2002 reported the first use of soluble rhodium catalyst in a biphasic system for the reductive alkylation of ammonia (scheme 3.4).[15] The use of iridium catalyst resulted in the formation of 8% of the required product on the other hand the use of [Rh(cod)Cl]2/ trisulfonated triphenylphosphine (TPPTS) was successful in promoting reductive amination of different substituted benzaldehydes. He was able to reuse the catalyst after recovery without any reduction in activity.

O NH3 (aq.), NH4Cl (50 mol%) o H THF, H2 (60 bar), 135 C,2h PhCH2NH2 PhCH2OH 0.05 % [Rh(COD)Cl]2,1.3%TPPTS 86% 3%

Scheme 3.4. Reductive Amination of Benzaldehyde with Ammonia.

Fernández introduced the use of supported rhodium catalyst for the reductive amination of acetone with 4-anilino-aniline to give the commercial product 3-IPPD. He reported that both the homogeneous phase or the immobilized catalyst on Montmorillonite K10 clay was

56

superior in terms of conversion (89-92%) to the commercially applied Pt/C catalyst (74% conversion) (scheme 3.5).[16]

O 10 bar H2 PhHN NH PhHN NH2 5h,120°C

[Rh(COD)(PPh3)2]BF4,TON=1060 [Rh(COD)(PPh ) ]BF on MM-K10, TON=1010 3 2 4

Scheme 3.5. Reductive Amination with Rhodium Supported Catalyst.

3.2. Reductive Amination the Current State of Art:

Reductive amination is a one-pot process in which the formation and the isolation of imines or enamines are avoided. Over the last three decades several research groups studied this transformation and factors affecting it. It was proved that pH has important influence on the progress of the reaction.[17] It was proposed that reductive amination passes through reduction of imine or iminium ion. As it is shown in (scheme 3.6), a carbonyl compound combines with a primary or secondary amine to form a hemiaminal species which forms an iminium ion. This iminium ion loses hydrogen resulting in the formation of imine which is then reduced to the amine product. The most critical factor in reductive amination is the good choice of conditions which favours intermediate reduction over ketone reduction to suppress alcohol formation.[18]

H R3 +H+ O N H R3 R4 -H2O N N NH2R3 R R R1 R2 +H O 1 2 2 R R R R HO -H+ 1 2 1 2 R3 H2 HN

R1 R2

Scheme 3.6. Mechanism of Reductive Amination.

The most common strategies in reductive amination representing the current state of art can be subdivided into three main strategies. In the first strategy reduction is carried out using molecular hydrogen with heterogeneous catalysts (palladium, platinum or nickel catalysts).

57

This is a straight forward, environmentally friendly with easy procedures, but incompatible with the coexisting functional groups such as nitro, cyano and C-C multiple bonds.[19] The second strategy is based on the transfer hydrogenation conditions utilizing formic acid or one of its derivative (Leuckart-Wallach type).[20] The third strategy uses hydride reductants e.g. [21] [22] [23] [24] [25] NaBH3CN, LiBH3CN, NaBH3CN-ZnCl2, NaBH3CNMg(ClO4)2, NaBH4-NiCl2, [26] [27] [28] [29] NaBH4-ZnCl2, borohydride exchange resin, ZnBH4, ZnBH4-ZnCl2, pyridine- borane,[30] picoline-borane,[31] etc. Recently organocatalysts were used in reductive amination utilizing Hantzsch esters or silanes as hydride sources with organocatalysts as chiral phosphoric acid and its derivatives.[32]

Reviewing the literature of the last 50 years it is obvious that among the different reductive amination strategies discussed above, hydride reduction with NaBH3CN which was introduced by Borch[33] has been used extensively. This may be due to the ease of use of these hydride sources. Borohydride salts are furthermore cheap, available in kg quantities and do not require special precautions in handling. Despite these advantages borohydride salts suffer from other drawbacks.

This reductant is used in excess quantity, toxic and produces toxic byproducts such as HCN and NaCN upon workup which limits its applications according to the new environmental standards. Abdel-Magid aimed to avoid this toxicity by using NaBH(OAc)3, (introduced by Gribble) as a mild reductant.[34,35] The mild nature of this reductant is due to the steric and electronic effects of the acetoxy groups which stabilize the B-H bond. His system was applicable for different types of aldehydes and unhindered aliphatic ketones. In spite of the significant applications of the above reductants they are not free from limitations regarding functional group tolerance and side reactions.[36] Also the formation of tertiary or secondary amine from primary amine (the desired product) due to over alkylation represents another limitation.[37]

Bhattacharyya and coworkers developed a highly efficient mild system for reductive i amination utilizing Ti(O Pr)4, and NaBH4 as the hydride donor and an amine source e.g. ammonia, ammonium chloride or methylamine. He was able to obtain high yields for different aldehydes, cyclic ketones and ketone with acid labile groups (scheme 3.7).[38]

58

O NHR3R4 i R4R3N OTi(O Pr)3 NaBH3CN R4R3N H i R1 R2 Ti(O Pr)4 R1 R1 R2 R2 i Scheme 3.7. Reductive Amination in the Presence of Ti(O Pr)4.

i Ti(O Pr)4 is considered as a mild and effective Lewis acid for suppressing alcohol formation in the reductive amination of ketones and aldehydes. It is compatible with most of acid- sensitive functional groups (e.g. acetonides, silyl ethers, esters, amides etc.).[39]

In order to understand the role of titaiunm isopropoxide in reductive amination, Matson i mixed equimolar ratios of amine and ketone with excess amount of Ti(O Pr)4. He tried to isolate and detect the intermediates. Neither imine nor enamine could be detected (by IR measurements) or isolated. Therefore, he predicted the formation of hemiaminal titanate intermediate (Scheme 3.8) which is an unstable complex and is reduced with NaBH3CN to form the amine product. Earlier findings by other scientists supported this proposal. Reetz has also predicted a similar titanium intermediate in his reaction between ketone and titanium amides with diisobutyl aluminium hydride (DIBAL-H) as a reductant.[40]

Also reductive amination was tested under solvent free conditions. The aldehyde or the ketone is mixed with the amine and the mixture is mixed in a mortar with NaBH4 or α- picoline borane until TLC showed disappearance of starting material (scheme 3.8).[41]

O NHPh NaBH .H BO (1:1) PhNH 4 3 3 H + 2 H grinding X X a: x = COMe d: x = CO 2Me b: x = CN e: x = NHCOMe c: x =CO2Hf:x=NO2

Scheme 3.8. Solvent Free Reductive Amination.

Baba developed the use of dibutylchlorotin hydride-HMPA complex as a mild hydride source for the reductive amination of various ketones and aldehydes. Various aromatic aldehydes with para or ortho electron withdrawing and electron donating groups were tested producing

59

high yields of secondary amines (81-99%). Cyanao, nitro and halogens substituents were tolerated and the amines were prepared in good to high yields (70-99%). Utilization of aliphatic amines as n-propyl amine resulted in poor yields (3-56%) compared with aniline derivatives (70-99%). Aromatic ketones as acetophenone and it substituted derivatives were also tested showing lower yields (35-69%) compared to aromatic aldehydes. Also aliphatic ketones showed variable results, benzyl acetone was an excellent substrates affording 91% yield. Other aliphatic 2-alakonones showed lower yields (<50%). In all cases, different labile substitutes were tolerated which is one of the major advantages of this system. Also he tested the use of ammonium salts for reductive amination resulting in high yields for benzyl acetone (99%) and acetophenone (71%) and benzaldehyde (99%).[42]

3.3. Asymmetric Reductive Amination:

3.3.1. Asymmetric Reductive Amination Utilizing Chiral Catalysts:

The progress in the preparation of chiral ligands and the success of their applications in many asymmetric transformations had led scientists to test these ligands in reductive amination. The asymmetric version of reductive amination represents an interesting approach for synthesis of enantiopure chiral amines in a one-pot procedure. Unfortunately, fewer examples of successful applications of chiral ligands for reductive amination compared to other organic transformations. The interaction of a prochiral ketone with an amine source to produce an enantioenriched chiral amine in the presence of a catalytic amount of a chiral entity is the most desired pathway. In one such enantioselective reductive amination approach, Blaser tested the chiral Ir-xyliPhos catalyst which was utilized in the synthesis of (S)-Metolachlor through catalytic imine reduction. Unfortunately the results for reductive amination was inferior compared to the imine reduction and only 77% ee was reported (scheme 3.9).[43]

60

Scheme 3.9. Synthesis of (S)-Metolachlor

One of the significant examples for the asymmetric reductive amination was developed by Zhang. The substituted aromatic and heteroaromatic ketones (acetophenone and substituted acetophenone) were used as examples and produced ee in the range of 89-96% and >99% yield with 1.0 mol % of Ir-(S,S)-f- Binaphane catalyst (scheme 3.10). According to the report,

Ti(OiPr)4 did not have any effect on the enantioselectivity but facilitates producing imine in situ from a hemiaminal titanate intermediate (as discussed earlier). He found that the presence I of iodine is essential for the reaction to proceed as the oxidative addition of I2 to the Ir precursor generates IrIII complex which is then coordinated with hydrogen forming IrIII-H complex to which imine is coordinated and starting the catalytic cycle. Despite these fascinating ees and yields for the aromatic ketones, this system failed to reductively aminate aliphatic ketones. High catalyst loading and high hydrogen pressure (69 bar) are other limitations of this methodology.[44]

OMe Ir-(S,S)-f-Binaphane (1.0 mol%) O 10% I2/Ti(OiPr)4 (1.5 equiv.) Ar R p-anisidine (1.2 equiv.) HN ∗ H (69 bar), RT 2 Ar R

P Fe P

Ir-(S,S)-f-Binaphane

Scheme 3.10. Binaphane Iridium Catalyst In Asymmetric Reductive Amination.

61

Pérez reported the use of BINAP derived palladium catalyst for the reductive amination of alkyl and cycloaliphatic ketones (scheme 3.11). He used o-, m- and p-substituted aniline derivatives as the source of nitrogen, molecular hydrogen at 800 psi (55 bar), in CHCl3 at 70 °C for 24 h. The use of molecular sieves was crucial for obtaining the product in high yields. He observed that the presence of substituents on the aniline improved stereoselectivity with a little effect on reactivity as they increase the steric bulk of the nitrogen source improving the interaction. When isobutyl methyl ketone reacted with aniline the ee was 51% but when p- anisidine was used the ee jumped to 90%. One of the remarkable examples was the success in reductive amination of 3-alkanones (2-heptanone). Although the ee was mediocre (49-59%) but it is known that these substrates are only accessible through carbanion chemistry. 2,3- butanedione underwent chemoselective reductive amination in good yields (83-85%) and low enantioselectivity (2-20% ee). Aryl ketones were also tested resulting in poor to mediocre ee (35-43%). One of the draw backs of this methodology is the harsh conditions needed for the removal of phenyl ring to obtain the primary amine.[45]

P Benzene (MeCN)2PdBr2 + P P PdBr2 rt, overnight P

Me PPh2 P P(p-Tolyl) PPh2 2 = Me PPh2 PPh P(p-Tolyl)2 P 2

R3 O NH 2 Catalyst (2.5 mol%) HN + R3 R1 R2 5Åms,CHCl3 R1 * R2 H2 (800 psi), 70 °C, 24 h

Scheme 3.11. BINAP Derived Palladium Catalyst in Asymmetric Reductive Amination

Xiao extended his work on imine reduction and tested his iridium based catalysts for reductive amination (scheme 3.12). He reported high ees and high yields for different acetophenone derivatives with para-anisidine as nitrogen source. The catalyst loading was 1.0 mol% with 5.0 mol% of phosphoric acid derivative as Lewis acid. The addition of molecular sieves was crucial for faster reaction. Using aniline with electron withdrawing 62

groups decreased both the enantioselectivities and the yields. Ortho substituted acetophenone derivatives(o-Cl, o-Me,o-OMe, o-F) which are known to be difficult substrates for reductive amination were reduced with high yields and high enantioselectivities using less sterically hindered catalysts. Aliphatic ketones were also tested and showed high ees and high yields with para-anisidine and with aniline.[46]

O NH2 [Ir] (1.0 mol %) HN OMe OH 5barH2 toluene 35oC, 12h OMe Ar Me Ar O S O O S O O Ph N O Ph N P Ir Ir O OH

Ph N X Ph N X H2 H2 Ar Ar = 2,4,6-(2-C3H7)3C6H2 X=7-H a: = Cl; b: X = 7-H c: Ar = 4-CH3C6H4,X=7-H d: Ar = 2,3,4,5,6-(CH3)5C6,X=7-H

OMe OMe OMe

HN HN HN O O2N O 92 % yield, 95 % ee 93 % yield, 95 % ee OMe 88 % yield, 81 % ee Br HN OMe HN HN F 92 % yield, 96 % ee 85 % yield, 85 % ee 89 % yield, 95 % ee

Scheme 3.12. Reductive Amination Utilizing Iridium Catalyst and Phosphoric Acid.

63

3.3.2. Reductive Amination Utilizing Chiral Auxiliary.

Ellman was the first to introduce the use of t-butylsulfinylamide as a chiral auxiliary for the asymmetric reductive amination (Scheme 3.13).[47] The imine was generated in situ with

Ti(OEt)4 at 60-70 °C which was then reduced with NaBH4 at -48 °C. Ti(OEt)4 serves as a desiccant, facilitates imine formation and even helps to improve the yield and diastereoselectivity of the t-butylsulfinyl protected-amines. He demonstrated that his methodology is applicable for both aryl-alkyl and alkyl-alkyl ketones in 66-86% yield and 80-94% de. Deprotection step is carried out under acidic conditions to obtain the primary amine without any compromise in the yield or ee. In his attempt to broaden the scope of the reaction he tested L-Selectride as a hydride donor under similar conditions. The opposite enantiomer of the primary amine was obtained. The aliphatic ketone substrates gave similar yield and de of the protected amine but benzocyclic ketones showed an improved de with similar yields.[48]

O R1 NaBH4 S N R2 THF H -48 °C O O O R R R 1 S 1 2 S NH2 N R2 Ti(OEt) 4 THF O R1 L-Selectr ide S N R2 THF H -48 °C

Scheme 3.13. Reductive Amination of Ketones using tert-butylsulfinylamide as Auxiliary.

Pannecoucke used the chiral auxiliary developed by Ellman for the reductive amination of α- fluoro α,β-unsaturated ketones. He reported mediocre to good yields (46-86%) and high des (96-99%) for aryl, aliphatic and sterically aliphatic substrates. The imine was prepared in situ through combining Ti(OEt)4 (2.0 equiv), (S)- tert-butylsulfinylamine (2.0 equiv), and 2- fluoroenone (1.0 equiv) in dry THF under argon and heated to reflux for 2 h. The mixture was allowed to cool to room temperature and then cooled to -78 °C. DIBAL-H (1M in toluene, 4.0 equiv) was then added dropwise, and the mixture was stirred for 1 h and the

64

progress of the reaction was monitored with NMR. The use of DIBAL-H resulted in the (S) conformation and the use of L-Selectride resulted in (R) conformation (scheme 3.14).[49]

O O F O 1) Ti(OEt)4,THF,reflux S R2 HN tBu 2) Reducing agent, THF H2N tBu F R1 H R2

R1 H Scheme 3.14. Reductive Amination of α-Fluoro α,β-Unsaturated Ketones.

3.3.3. Reductive Amination Utilizing Molecular Hydrogen:

Alexakis, utilized a combination of Ti(OiPr)4/Pd-C/H2 to synthesize C2 symmetric secondary amines with 70-82% de and 88-92% yield. His system was only applicable for the aromatic substrates.[50] Nugent and Seemayer developed a highly efficient methodology for the synthesis of quinuclidine.[51] Through utilization molecular hydrogen, heterogeneous hydrogenation catalysts Pt-C or Pd-C and Ti(OiPr)4 for the reductive amination of a labile α- chiral quinuclidinone they were successful in incorporation the amine without epimerization, a feat not previously accomplished (scheme 3.15).

NHCH Ph NHCH2Ph O 2 Ph NH2 N N N Ti(OiPr) /Pt-C/H (4.2 bar) Ph Ph Ph 4 2 Ph Ph 25oC, 15h, 80% de Ph

Scheme 3.15. Reductive Amination of a Quinuclidinone

Nugent group recently developed a two-step methodology relying on the asymmetric reductive amination of prochiral ketones with the chiral ammonia equivalent (R)- or (S)-α- methylbenzylamine for α-chiral primary amine synthesis. They found that the use of i (Ti(O Pr)4 (1.2 equiv)) with ((R)- or (S)-α- methylbenzylamine (α-MBA) (1.1 equiv)) and heterogeneous catalyst (Ra-Ni, Pd-C, Pt-C) produced the secondary amine of different 2- alkanones, cyclic ketones, and aryl alkyl ketones in high yields and diastereoselectivities. The secondary amine is produced in a single step without the need for the tedious process of imine isolation and purification. Simple acid base work is usually enough to purify the 65

secondary amine product and any impurities like α-MBA (3-5%) can be removed easily through washing with NH4Cl. The primary amine is produced in high ee and yield through hydrogenolysis using Pd-C in MeOH and the ee can be further enhanced through simple crystallization. Category 1: Raney-Ni Category 2: Pt-C

HN Ph HN Ph HN Ph HN Ph

76% yld,87% de 94% yld, 74% de 79% yld, 87% de 82% yld, 92% de 1,2-dichloroethane THF hexane ethanol

Category 3: Pd-C

H N NH 2 HN Ph 2 HN Ph

Ph Ph 92% yld, 94% de 89% yld, 80% de 92% ee, over all yld 76% 76% ee, overall yld 64% ethylacetate methylenechloride ethylacetate methyl-t-butylether

Figure 3.1. Correlation of Heterogeneous Hydrogenation Catalysts with Ketone Structure and Product example.

Nugent group also investigated other commercially available Lewis acids and they found that i i i B(O Pr)3 or Al(O Pr)3, can be used to replace Ti(O Pr)4. These Lewis acids are cheaper than i Ti(O Pr)4 but must be used in greater quantities. These Lewis acids hold the advantage over i i Ti(O Pr)4 due to their easier work up procedures. On the work-up of Ti(O Pr)4 reaction a finely dispersed TiO2 can be formed forcing a celite filtration onscale. If no Lewis acid or the wrong one is present, large quantities of the alcohol by product can be expected.[52]

3.3.4. Asymmetric Reductive Amination Utilizing Transfer Hydrogenation Conditions (the Leuchart–Wallach Reaction):

As mentioned before, the source of hydrogen can be molecular hydrogen, hydride or through utilizing transfer hydrogenation conditions. Transfer hydrogenation conditions were applied successfully for the reduction of ketones to alcohols. The method is highly successful in terms of obtaining high ees and high yields.[53] As it is a highly desirable goal several

66

attempts were directed for the asymmetric reductive amination of ketones under transfer hydrogenation conditions but with limited success compared to ketone reduction. Nevertheless, some useful recent breakthroughs have been achieved.[54] A remarkably effective asymmetric version was reported by Kadyrov, Riermeier and Börner.[55] They reported the use of several ruthenium and rhodium catalysts for the conversion of acetophenone derivatives to the enantiomerically enriched amines (Scheme 3.17 ). According to their strategy different aryl-alkyl ketones can be utilized as substrates producing the primary amines in high yields (74-92%) and enantioselectivity (89-95%) in a one step reaction. HCO2NH4 was used as the hydride source with NH3 as the nitrogen source and different BINAP ligands were tested showing the best catalyst was [Ru-(R)-(TolBINAP)Cl2]. In the reaction along with the primary amine a formyl derivative (RHNC(O)H) is produced and in order to improve the yield the crude product is treated with HCl (EtOH/H2O) at reflux to obtain the desired amine in a good to excellent yield. Despite these encouraging results the application of this method is limited to the aromatic substrates in which other substrates as 1- indanone (6% yield, no reported ee, chiral Ru catalyst) and aliphatic ketones, e.g. 2-octanone (44% yield, 24% ee, using chiral Ru catalyst) showed unsatisfactory results (scheme 3.16).

O (i)1 mol% [Ru-(R)-TolBINAP(Cl)2] H NH2 R' NH HCO ,NH/MeOH (15-20%), 85oC R' R 4 2 3 R (ii) 6N HCl, reflux, 1h

Aryl group R Yield and ee

Ph Me 92, 95% ee R Ph Et 89, 95% ee R 3-MeC6H4 Me 74, 89% ee R 4-MeC6H4 Me 93, 93% ee R 4-ClC6H4 Me 93, 92% ee R 4-(NO2)C 6H4 Me 92, 95% ee R

Scheme 3.16. Transfer Hydrogenation of Acetophenone Derivatives.

3.4.5. Organocatalytic Asymmetric Reductive Amination:

The use of organocatalysts was slowly introduced to organic chemistry over the last two decades. There were earlier trials of using organic compounds to catalyze organic reactions but the low yields and were discouraging. Recent advances in spectroscopic and asymmetric techniques have opened the door for the synthesis of a big 67

library of organocatalysts which were used efficiently for different organic transformations. Several research groups focused their efforts on the synthesis of novel organocatalysts for asymmetric imine reduction and reductive amination. Among these transformations are imine reduction and reductive amination. At this time, development in the application of organocatalysis for reductive amination is still in its infantile stage compared to other transformations.[56]

H X +2 [H] X ∗ R R chiral 1 2 catalyst R1 R2 X=CR2,O,NR

Scheme 3.17. Asymmetric Reduction of prochiral Compounds.

Inspired by nature and how living organisms reduce imino group through the employment of organic dihydropyridine cofactors such as nicotinamide adenine dinucleotide (NADH) in combination with enzyme catalysts (figure 3.2).[57]

O H H NH2 N

O - H H . O P O O O H H O OH OH H HAdenine O P O O O- H H OH OH reduced nicotinamide adenine dinucleotide (NADH) Figure 3.2. Reduced Nicotinamide Adenine Dinucleotide

Scientists started to think of NADH analogues and they found that the best analogues would be Hantzsch esters. These hydrogen sources in the presence of achiral Lewis or Brønsted acid catalysts proved to be efficient in imine reduction.[58] List investigated the catalytic cycle of reductive amination utilizing Hantzsch esters. He proposed that reductive amination of ketones is initiated by protonation of the in situ generated ketimine from a chiral Brønsted acid catalyst (Scheme 3.18). The resulting iminium ion pair, which may be stabilized by

68

hydrogen bonding, is chiral and its reaction with the Hantzsch dihydropyridine could give an enantiomerically enriched amine and pyridine. After screening different phosphoric acid catalysts, catalyst 9 was found to be the best catalyst for this reaction and 1.0 mol% of the catalyst resulted in 93% ee for the product with an excellent yield of 96% (scheme 3.19).[59]

OMe OMe

HN EtOOC COOEt N O X* +PMPNH2 N H -H O 2

HX* PMP OMe HN H EtOOC COOEt

H2N H X* N

Scheme 3.18. Mechanism of Chiral Brønsted Acid Catalysed Reductive Amination.

O o PMP 9(1mol%),toluene,35C,71h, 98% HN

EtOOC COOEt (1.4 equiv) 93% ee N H i-Pr i-Pr

i-Pr O P O O i-Pr OH

i-Pr i-Pr 9 Scheme 3.19. Organocatalytic Reductive Amination Developed by List.

List also investigated the reductive amination of aldehydes. He proposed that under the conditions of reductive amination an α-branched aldehyde substrate would undergo a fast racemization in the presence of the amine and acid catalyst via an imine/enamine tautomerization. The reductive amination of one of the two imine enantiomers would then

69

have to be faster than that of the other, resulting in an enantiomerically enriched product which is another successful example of a dynamic kinetic resolution process.[60]

He reported high enantioselectivities for the reductive amination of hydratopicaldehyde with p-anisidine in the presence of Hantzsch ester and phosphoric acid catalyst 9.[61] List demonstrated that racemic aldehydes could be successfully converted to branched-chain secondary amines in an excellent enantiomeric excess. He found that the use of a highly hindered phosphate catalyst was essential, and intriguingly the best results required the very specific use of a particular Hantzsch ester, in this case (Scheme 3.20). O H NR 2 3 NHR3 α-branched chiral amines ∗ [H] R1 R2 R R 1 2 O

R1 H H2NR3 R1 ∗ β-branched chiral amines NHR3 R2 [H] R2

R3 R3 N R3 N HN O +H NR R1 R1 2 3 H H R1 R1 H -H2O H R2 R2 R2 R2 racemization

H R3 N HX* R1 H

R2 R1 NHR3 R2 EtOOC COOEt EtOOC COOEt

N N H

o R1 CHO 9(5mol%),MS5Å,C6H6,6 C, 72 h R1 NHR3 R2 EtOOC COOEt R2 (1.2 equiv) N H F

NHPMP NHPMP NHPMP NHPMP

87%, 96% ee 88%, 98% ee 89%, 94% ee 92%, 98% ee

NHPh NHPMP

78%, 94% ee 77%, 80% ee Scheme 3.20. Asymmetric Reductive Amination of Aldehydes. 70

Later he utilized this methodology for the synthesis of important pharmaceutical building blocks. He reported the enantioselective synthesis of pharmaceutically relevant 3-substituted cyclohexylamines from 2,6-diketones via an aldolization-dehydrationconjugate reduction- reductive amination cascade that is catalyzed by a chiral Brønsted acid and accelerated by the achiral amine substrate, which is ultimately incorporated into the product. 2,6-diketone was treated with 1.0 equiv of an achiral amine, 2.0 equiv of a Hantzsch ester, and 10 mol % of a chiral Brønsted acid resulted in the formation of the corresponding cyclohexylamines with mediocre to good yield (35-79%) and with good to high diastereoselectivity (82-96%). Alkyl, aryl and sterically congested aryl substituted 2,6-diketones were reductively aminated with high stereoselectivities.[62]

+NHR NHR2 2 +NHR - 2 X - X X O R1 R1 R1

i-Pr i-Pr

i-Pr O O P EtO2C CO2Et O OH i-Pr N O H 3(2.2equiv) R2 HN i-Pr i-Pr R2NH2 (1.5 equiv) Y O (R)-TRIP (10mol%) MS 5Å, cyclohexane, 50 oC Y R1 R1

Scheme 3.21. Synthesis of Pharmaceutical Building Blocks Utilizing Reductive Amination.

Menche have also demonstrated that thiourea acts as an efficient organocatalyst for the reductive amination of aldehydes using aniline derivative and the ethyl Hantzsch ester providing the corresponding achiral N-benzylanilines in good to excellent yields (72-93%). Using 1.1 equiv of Hantzsch ester and 1.0 equiv of thiourea with molecular sieves in toluene at 70 °C for 24 h substituted benzaldehydes as well as two aliphatic aldehydes were reacted with p-anisidine forming the secondary amines in good isolated yields (scheme 3.22).[63]

71

H H EtO2C CO 2Et OMe OMe N O H (1.1 equiv) + R N R H H N H 2 S

H2N NH2 (1.0 equiv) 5 Å MS, toluene, 70 °C

Scheme 3.22. Reductive Amination of Aldehydes Utilizing Thiourea.

One of the milestones in this field was developed by MacMillan, who investigated the reductive amination of acetophenone with p-anisidine utilizing ethyl Hantzsch ester, and BINOL-derived phosphoric acids. The catalyst showed high catalytic activity and excellent enantiocontrol in the reductive coupling reaction. Removal of water by the addition of 5Å molecular sieves proved to be important for achieving high catalytic activity and selectivity. Aryl-alkyl ketones were reduced with good yields (70-87%) and high enantioselectivity (85- 97%) and 2-alkanones were also reduced with mediocre to good yields (49-75%) with good to high enantioselectivities (81-94%). Different aniline derivatives were tested producing the secondary amine with good to high yields (55-92%) and high enantioselectivities (90-95%). The overall diversity of ketone substrates makes his work fascinating from all aspects.[64]

O o PMP 15 (10 mol%),PMPNH2(1 equiv), MS 5A,C6H6,50 C, 96h, 87% HN

EtOOC COOEt

(1.2 equiv) 87% yield, 94 % ee N H

SiPh3

O O P O OH

SiPh3 15

OMe OMe OMe

HN HN F HN F

79 % yield, 91 % ee 60 % yield, 83 % ee

72

OMe OMe O HN HN HN

O2N 71 % yield, 95 % ee 75 % yield, 85 % ee 92 % yield, 91 % ee

OMe OMe OMe HN HN HN

2 71 % yield, 83 % ee 60 % yield, 90 % ee 73 % yield, 96 % ee

Scheme 3.23. Asymmetric Reductive Amination System Developed by MacMillan

3.4. Green Chemistry and Reductive Amination:

3.4.1. Green Chemistry Basic principles.

Reductive amination is a one pot process for chiral amine synthesis. As described previously, it has many advantages compared to the other available methodologies. One of the significant advantages of this methodology is that it is considered an environmentally friendly process. To understand on which bases scientists made such assumption, we have first to know more about green chemistry and environmentally friendly process.

It is widely acknowledged that there is a growing need for more environmentally acceptable processes in the chemical industry and pharmaceutical industry. This trend has led to the concept of ‘Green Chemistry’.[65]

The new trend differs dramatically from the old traditional concepts focused only on process efficiency and chemical yield. The new trend assigns the economic value of the process depending on its ability to eliminate waste at source and avoid the use of toxic and/or

73

hazardous substances. Green chemistry can be defined as the new trend in chemistry which efficiently utilizes (preferably renewable) raw materials, eliminates waste and avoids the use of toxic and/or hazardous reagents and solvents in the manufacture and application of chemical products.[65]

Recently scientists proposed the basic principles of Green Chemistry which can be paraphrased as:[65,66]

1. Waste prevention instead of disposal. 2. Atom efficiency and atom economy. 3. Less hazardous/toxic chemicals 4. Design safer process and safer product. 5. Safe solvents and auxiliaries 6. Design efficient energy manipulation. 7. Use of renewable raw materials. 8. Step wise efficient methodologies. 9. Catalytic rather than stoichiometric reagents 10. Biodegradable products. 11. Desing analytical techniques for pollution control. 12. Inherently safer working environment.

Green chemistry main concern is the environmental impact of both chemical products and the processes by which they are produced. It is well known that prevention is better than cure. Green chemistry eliminates waste at source, it focuses on primary pollution prevention rather than waste remediation (end-of-pipe solutions).

To be able to classify any chemical or pharmaceutical process as green process, two important measures of the potential environmental acceptability of the process have been introduced the E factor and the Atom Efficiency.[65]

The E factor is defined as the mass ratio of the waste to the desired product. A higher E factor means more waste and, consequently, greater negative environmental impact. The ideal E factor is zero. It was found that the pharmaceutical industry has the highest E factor 74

compared to the oil refinery industry which has the lowest E factor. Atom efficiency is calculated by dividing the molecular weight of the desired product by the sum of the molecular weights of all the substances produced in the stoichiometric equation.

3.4.2. Hydrogenation and Green Chemistry:

Catalytic hydrogenation–utilizing hydrogen gas and heterogeneous catalysts–can be considered as the most important catalytic method in synthetic organic chemistry on both laboratory and production scales. Hydrogen is, without doubt, the cleanest reducing agent and the heterogeneous robust catalysts have been routinely employed. Catalytic hydrogenation has distinctive advantages over other methodologies. Key advantages of this technique are:[67]

1. Broad scope, many functional groups can be hydrogenated with high selectivity. 2. High conversions are usually obtained under relatively mild conditions in the liquid phase. 3. The large body of experience with this technique makes it possible to predict the catalyst of choice for a particular problem. 4. The process technology is well established and scale-up is therefore usually straightforward.

The field of hydrogenation is also the area where catalysis was first widely applied in the fine chemical industry. It is a key example of green technology, due to the low amounts of catalysts required, in combination with the use of hydrogen (100% atom efficient!) as the reductant. In general, if chirality is not required, heterogeneous supported catalysts can be used in combination with hydrogen. Catalytic hydrogenation is considered as the green route for the synthesis of different functional compounds as amines, alcohols, and amino acids. Once selectivity and chirality is called for, homogeneous catalysts and biocatalysts are applied. The use and the application of chiral Ru, Rh and Ir catalysts has become a well developed technology. Homogenous catalytic hydrogenation gives access to a large variety of asymmetric transformations: imines and functionalized ketones and alkenes can be converted with high selectivity in most cases.

75

3.5. Conclusion:

Different methodologies have been introduced for the synthesis of α- chiral amines. One of the most important strategy introduced for this purpose is reductive amination. Reductive amination is a stepwise efficient methodology starting from the prochiral ketone to the α- chiral amine. Earlier reports described the use of classical heterogeneous catalysts for this transformation. Homogeneous catalysts were also introduced in the seventies and marked a significant breakthrough in the field. Asymmetric version was introduced utilizing chiral catalyst, chiral auxiliaries or chiral organocatalysts. The use of Brønsted or Lewis acids was important in most of the methodologies. Earlier reports suggested the role of the acid as an efficient desiccant but recent reports suggested more complicated function as forcing hemiaminal formation. Over the last two decades several methodologies were evolved allowing the synthesis of α- chiral amines from aryl-alkyl ketone and 2-alaknones in good to high yields and enantioselectivities.[68]

3.6. References:

[1] 1. Mignonac, G. Compt. Rend. 1921, 172, 223. [2] For review articles, see (a) W. S. Emerson, Org. React. 1948, 4, 174; (b) F. Möller, R. Schröter, in Methoden der Organischen Chemie (Houben-Weyl), Thieme: Stuttgart; Bd. XI/1, 1957, 602–673. [3] S. Nishimura, Handbook of Heterogeneous Catalytic Hydrogenation for Organic Synthesis, John Wiley &Sons, Inc., New York, 2001. [4] C. F. Winans, J. Am. Chem. Soc. 1939, 61, 3566. [5] B. M. Vanderbilt, U.S. Pat. 2,219,879, 1941. [6] E. H. Pryde, D. E. Anders, J. C. Cowan, J. Am. Oil Chem. Soc. 1969, 46, 67. [7] A. Skita, F. Keil, Ber. Dtsch. Chem. Ges. 1928, 61, 1682. [8] R. Cantarel, Compt. Rend. 1940, 210, 403. [9] Winans, C. F.; Adkins, H. J. Am. Chem. Soc. 1932, 54, 306 [10] A. C. Cope, E. M. Hancock, J. Am. Chem. Soc. 1942, 64, 1503 [11] Dovell, F. S.; Greenfield, H. J. Org. Chem. 1964, 29, 1265 [12] F. S. Dovell, H. Greenfield, J. Am. Chem. Soc. 1965, 87, 2767 [13] L. Marko, J. Bakos, J. Organomet. Chem. 1974, 81, 411 76

[14] V. I. Tararov, R. Kadyrov, T.H. Riermeier, A. Borner, J. Chem. Soc. Chem. Commun. 2000, 1867. [15] T. Gross, A.M. Seayad, M. Ahmad, M. Beller, Org. Lett. 2002, 4, 2055. [16] R. Margalef-Catala, C. Claver, P. Salagre, E. Fernandez, Tetrahedron Lett. 2000, 41, 6583. [17] R. F. Borch, M. D. Bernstein, H. D. Durst, J. Am. Chem. Soc. 1971, 93, 2897. R. F. Borch, A. I. Hassid, J. Org. Chem. 1972, 37, 1673. [18] A. F. Abdel-Majid, K. G. Carson, B. D. Harris, C. A. Maryanoff, R. D. Shah, J. Org. Chem.1996, 61, 3849. [19] a) P. N. Rylander, Catalytic hydrogenation in organic synthesis; Academic: New York, 1979; p 165; b) M. V. Klyuev, M. L. Khidekel, Russ. Chem. Rev. 1980, 49, 14; c) T. A. Tarasevich, N. G. Kozlov, Russ. Chem. Rev., 1999, 68, 55. [20] M. Kitamura, M. Tsukamoto, Y. Bessho, M. Yoshimura, U. Kobs, M. Widhalm, R. Noyori, J. Am. Chem. Soc. 2002, 124, 6649. [21] R. F. Borch, M. D. Bernstein, H. D. Durst, J. Am. Chem. Soc. 1971, 93, 2897. [22] R. F. Borch, H. D. Durst, J. Am. Chem. Soc. 1969, 91, 3996. [23] S. Kim, C. H. Oh, J. S. Ko, K. H. Ahn, Y. J. Kim, J. Org. Chem.1985, 50, 1927. [24] J. Brussee, R. A. T. M. van Benthem, C. G. Kruse, A. van der Gen, Tetrahedron: Asymmetry 1990, 1, 163. [25] I. Saxena, R. Borah J. C. Sarma, J. Chem. Soc., Perkin Trans. 1, 2000, 503. [26] S. Bhattacharyya, J. Org. Chem. 1995, 60, 4928. [27] N. M. Yoon, E. G. Kim, H. S. Son, J. Choi, Synth. Commun. 1993, 23, 1595. [28] H. Kotsuki, N. Yoshimura, I. Kadota, Y. Ushio, M. Ochi, Synthesis 1990, 401. [29] S. Bhattacharyya, A. Chatterjee, J. S. Williamson, Synth. Commun. 1997, 27, 4265. [30] a) A. Pelter, R. M. Rosser, S. Mills, J. Chem. Soc., Perkin Trans. 1, 1984, 717; b) M. D. Bomann, I. C. Guch, M. DiMare, J. Org. Chem. 1995, 60, 5995. [31] S. Sato, T. Sakamoto, E. Miyazawa, Y. Kikugawa, Tetrahedron 2004, 60, 7899. [32] D. Menche, J. Hassfeld, J. Li, G. Menche, A. Ritter, S. Rudolph, Org. lett.2006, 8, 741. [33] R. F. Borch, M. D. Bernstein, H. D. Durst, J. Am. Chem. Soc. 1971, 93, 2897. [34] A. F. Abdel-Majid, K. G. Carson, B. D. Harris, C. A. Maryanoff, R. D. Shah, J. Org. Chem.1996, 61, 3849; [35] a) A. F. Abdel-Majid, S. J. Mehrman, Org. Process Res. Dev. 2006, 10, 971; b) G. W. Gribble, Org. Process Res. Dev. 2006, 10, 1062. 77

[36] E. W. Baxter, A. B. Reitz, Organic reactions; Wiley: New York, 2002; Vol. 59. [37] M. B. Smith, J. March, March’s advanced organic chemistry; Wiley: New York, 2001. [38] a) S. Bhattacharyya, Tetrahedron Lett. 1994, 35, 2401; b) S. Bhattacharyya, J. Org. Chem. 1995, 60, 4928; c) K. A. Neidigh, M. A. Avery, J. S. Williamson, S. Bhattacharyya, J. Chem. Soc., Perkin Trans. 1, 1998, 2527. [39] R. J. Mattson, K. M. Pham, D. J. Leuck, K. A. Cowen, J. Org. Chem. 1990, 55, 2552. [40] M. T. Reetz, Organotitanium reagents in organic synthesis; Springer-Verlag: Berlin, 1986; p 107. [41] a) B.T. Cho, S. K. Kang, Tetrahedron 2005, 61, 5725 ; b) S. Sato, T. Sakamoto, E. Miyazawa, Y. Kikugawa, Tetrahedron 2004, 60, 7899. [42] a) T. Suwa, E. Sugiyama, I. Shibata, A. Baba, Synthesis 2000, 789; b) T. Suwa, E. Sugiyama, I. Shibata, A. Baba, Synlett 2000, 556. [43] H. -U. Blaser, H. -P. Buser, H. -P. Jalett, B. Pugin, F. Spindler, Synlett 1999, 867. [44] Y. Chi, Y.-G. Zhou, X. Zhang, J. Org. Chem. 2003, 68, 4120. [45] L. R. Pérez, F. J. P.-Flores, P. Sharma, L. Velasco, A. Cabrera Org. Lett. 2009, 11, 265. [46] C. Li, B. V.-Marcos, J. Xiao, J. Am. Chem. Soc. 2009, 131, 6967. [47] G. Borg, D. A. Cogan, J. A. Ellman, Tetrahedron Lett. 1999, 40, 6709. [48] J. Tanuwidjaja, H. M. Peltier, J. A. Ellman, J. Org. Chem. 2007, 72, 626. [49] G. Dutheuil, S. C. -Bonnaire, X. Pannecoucke, Angew. Chem. Int. Ed. 2007, 46, 1290. [50] A. Alexakis, S. Gille, F. Prian, S. Rosset, K. Ditrich, Tetrahedron Lett. 2004, 45, 1449. [51] T. C. Nugent, R. Seemeyer, WO2004035575, 2004. [52] a) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, Adv. Synth. Catal. 2006, 348, 1289; b) T. C. Nugent, Chiral Amine Synthesis - Strategies, Examples, and Limitations. In Process Chemistry in the Pharmaceutical Industry, Second Edition: Challenges in an Ever-Changing Climate, T. F. Braish, K. Gadamasetti Eds.; CRC Press- Taylor and Francis Group: New York, 2008; c) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, WO2006030017, 2006; d) T. C. Nugent, V. N. Wakchaure, A. K. Ghosh, R. R. Mohanty, Org. lett. 2005, 7, 4967; e) T. C. Nugent, A. K. Ghosh, Eur. J. Org. Chem. 2007, 3863. [53] R.Noyori, Adv. Synth. Catal. 2003, 345, 15 [54] M. Kitamura, D. Lee, S. Hayashi, S. Tanaka, M. Yoshimura, J. Org. Chem. 2002, 7, 8685

78

[55] a) R. Kadyrov, T. H. Riermeier, Angew Chem., Int. Ed. 2003, 42, 5472; b) A. Börner, U. Dingerdissen R. Kadyrov, T. H. Riermeier, V. Tararov, US2004267051, 2004; c) T. H. Riermeier, K.-J. Haack, U. Dingerdissen, A. Börner, V. Tararov, R. Kadyrov, US6884887, 2005. [56] a) T. Ohkuma, M. Kitamura, R. Noyori, Asymmetric hydrogenation. In: Ojima I (ed) Catalytic asymmetric synthesis, 2nd edn. Wiley-VCH, New York, 2000; b)T. Ohkuma, R. Noyori Hydrogenation of imino groups. In: Jacobsen EN, Pfaltz A, Yamamoto H (eds) Comprehensive asymmetric catalysis, suppl 1. Springer, New York, 2004; c) V. I. Taratov, A. Börner A, Synlett 2005, 203. [57] B. D. Sanwal, M. W. Zink, Arch Biochem Biophys, 1961, 94, 430. [58] T. Itoh, K. Nagata, M. Miyazaki, H. Ishikawa, A. Kurihara, A. Ohsawa Tetrahedron 2004, 60, 6649. [59] S. Hoffmann, A. M. Seayad, B. List, Angew. Chem. Int. Ed. 2005, 44, 7424. [60] a) Noyori, M. Tokunaga, M. Kitamura, Bull Chem Soc Jpn 1995, 68, 36; b) R. S. Ward, Tetrahedron Asymmetry 1995, 6, 1475; c) H. Stecher, K. Faber, Synthesis 1997, 1; d) H. Perllissier, Tetrahedron 2003, 59, 8291; e) F.F. Huerta, A. B. E. Minidis, J. E. Bäckvall, Chem Soc Rev 2001, 30, 321; f) S. Caddick , K. Jenkins, Chem Soc Rev 1996, 25, 447. [61] S. Hoffmann, M. Nicoletti, B. List, J. Am. Chem. Soc. 2006, 128, 13074. [62] J. Zhou, B. List, J. Am. Chem. Soc. 2007, 129, 7498. [63] D. Menche, F. Arikan, Synlett 2006, 841. [64] R. I. Storer, D. E. Carrera, Y. Ni, D. W. C. MacMillan, J. Am. Chem. Soc. 2006, 128, 84. [65] a) P. Anastas, J. C. Warner, Green Chemistry: Theory and Practice, Oxford University Press, Oxford, 1998; b) P.T. Anastas, M.M. Kirchhoff, Acc. Chem. Res. 2002, 35, 686; c) P.T. Anastas, L.G. Heine, T. C. Williamson, Green Chemical Syntheses and Processes, American Chemical Society, Washington DC, 2000; c) P.T. Anastas, C. A. Farris, Benign by Design: Alternative Synthetic Design for Pollution Prevention, ACS Symp. Ser. nr. 577, American Chemical Society, Washington DC, 1994; d) J.H. Clark, D.J. Macquarrie, Handbook of Green Chemistry and Technology, Blackwell, Abingdon, 2002; e) A. S. Matlack, Introduction to Green Chemistry, Marcel Dekker, New York, 2001; f) M. Lancaster, Green Chemistry: An Introductory Text, Royal Society of Chemistry, Cambridge, 2002; g) J.H. Clark (Ed.), The Chemistry of Waste Minimization, Blackie, London, 1995. [66] R. A. Sheldon, C. R. Acad. Sci. Paris, IIc, Chimie/Chemistry, 2000, 3, 541. 79

[67] a) R. A. Sheldon, Chemtech 1994, 38; b) R. A. Sheldon, J. Chem. Technol. Biotechnol. 1997, 68, 381; c) R. A. Sheldon, J. Mol. Catal. A: Chemical 1996, 107, 75; d) R. A. Sheldon, Pure Appl. Chem. 2000, 72, 1233. [68] a) D. A. Cogan, G. Liu, J. Ellman, Tetrahedron Lett. 1999, 55, 8883; b) D. A. Cogan, J. A. Ellman, J. Am. Chem. Soc. 1999, 121, 268; c) L. C. Akullian, M. L. Snapper, A. H. Hoveyda, Angew. Chem., Int. Ed. 2003, 42, 4244; d) L. C. Akullian, J. R. Porter, J. F. Traverse, M. L. Snapper, A. H. Hoveyda, Adv. Synth. Catal. 2005, 347, 417; e) A. Côté, A. B. Charette, J. Org. Chem. 2005, 70, 10864.

80

Chapter 4 Drugs and Reductive Amination

4.1 Reductive Amination in the Synthesis of Drugs and Natural Products:

Different natural products and pharmaceutical drugs contain amino group as an important part of their structure. Of course some drugs are still sold as racemic compounds but the latest trend over the past three decades is to design, develop and market new drug entities as a single isomeric form. Amino group can be introduced in the drug entity through different strategies one of these strategies is reductive amination. Older reports describing the synthesis of natural products and pharmaceutical drugs utilizing reductive amination did not involve any source of chirality resulting in the synthesis of racemic product. Recent literatures focused on utilizing the asymmetric versions of reductive amination for the synthesis of enantiopure compounds. I will try to give a brief overview on the potential applications of this powerful methodology in the synthesis of natural products and pharmaceutical drugs. We will try also to show the relevance of our developed strategy for the efficient synthesis of these entities.

4.1.1. Synthesis of Delavirdine:

This compound is a member of nonnucleoside HIV-1 reverse transcriptase inhibitors.[1] This class of compounds was discovered by Upjohn scientists from a computer-directed dissimilarity analysis of the Pharmacia & Upjohn chemical library to select compounds for screening against HIV-1 reverse transcriptase. Synthesis of this compound starts with the addition of piperazine (20) to chloropyridine (21). The nitro group is reduced to the amino group and the resultant amine undergoes reductive amination with acetone to provide pyridylpiperazine (23). Coupling of (23) with 6-nitroindole-2-carboxylic acid (24) is accomplished using either 1-ethyl-3-(dimethylamino) propylcarbodiimide (EDC) or 1,10-

81

carbonyldiimidazole (CDI) to give amide (25). The nitro group is reduced under hydrogenation conditions using Pd-C. The resulting amine is then sulfonylated with methanesulfonyl chloride to provide delavirdine, which is then transformed to delavirdine

1) Pd/C, H2,78% H NO2 1) CH Cl 96% O N N 2 2 2 2) acetone Cl 2) (Boc)2O96% HN NaCNBH3 91% BocN N N N N HN N H 22 23 N 20 21 O N 2 CDI or EDC mesylate (3). COOH 74% N H 24

1) H2, Pd/C, 66% HN HN 2) CH3SO2Cl, N N pyr, CH2Cl2 N 3) CH SO H N 3 3 H O2N N N N S N O O CH3SO3H O N O H 25 H 3

Scheme 4.1. Synthesis of Delavirdine.

4.1.2. Synthesis of Muraglitazar:

Muraglitazar is developed to treat hyperglycemia and dyslipidemia through decrease triglycerides and increase HDL cholesterol with minimal effects on LDL cholesterol.[2] Several syntheses have been developed for its efficient preparation. Synthesis starts with the alkylation of 4-hydroxybenzaldehyde with phenyloxazolemesylate (23), which can be easily synthesized from commercially available alcohol (22), resulting in the aldehyde (24) synthesis. The aldehyde is then treated with glycine methyl ester under reductive amination conditions to provide secondary amine (25) in an excellent yield. Reaction of amine (25) with 4-methoxyphenyl chloroformate followed by hydrolysis of the methyl ester afforded Muraglitazar in 94% yield.

82

Scheme 4.2. Synthesis of Muraglitazar.

4.1.3. Synthesis of Amphetamine:

Amphetamines synthesis is one of the well known classical examples of drugs synthesised utilising reductive amination.[2] Other methodologies were also developed for there synthesis as the direct displacement of a leaving group by an amine, nitro alkane addition followed by reduction of the nitro group and metal-promoted amination of an unsaturated carbon compound. The reductive amination of methyl benzyl ketone was one of the earliest strategies used in amphetamine synthesis. Despite being developed in the thirties of last century it continues to be one the best developed methodologies because of its simple elegance and the availability of cheap starting materials.

According to the developed methodology, oxime is formed as a mixture of isomers upon exposure of the ketone to hydroxylamine hydrochloride under mildly basic conditions. 83

Reduction of the oxime can be accomplished using a variety of reducing agents. The initial report employed sodium in methanol for converting the oxime to the target amphetamine. Other modifications of this approach have been described in recent literatures.

Scheme 4.3.Synthesis of Amphetamine:

The asymmetric synthesis of was developed in the seventies of the last century. The synthesis starts with methyl benzyl which is reductively aminated using readily available chiral α-methyl benzyl amine producing the imine intermediate and the synthesis is driven to completion by removing H2O with Dean–Stark trap. The resulting imine was reduced with Raney nickel. The product was isolated and crystallized as HCl salt which is then hydrogenolyzed with Pd-C producing the primary amine in high overall optical purity.

4.1.4. Synthesis of Sertraline:

Sertraline is an anti-depressant drug that affects serotonin levels in the brain. Initially the active isomer was not known when both diastereoisomers were prepared through an unselective route.[3,4] Synthesis starts with Friedel-Crafts reaction between 1,2- dichlorobenzene and succinic anhydride forming the starting material presented in the following scheme.

Scheme 4.4. Synthesis of Sertraline:

84

Studies showed that the active isomer is the syn diastereomer. Reductive amination of the ketone in the final step could be controlled to give 70% syn diastereomer.

4.1.5. Synthesis of Emitine:

Emitine is a natural compound which is extracted from ipecacuanha plant (Brazilian root) and used as the primary drug for treating amebiasis, leishaniasis, and dysentery.[5,6] It has a direct amebicidal effect against trophozoites E. histolytica in tissues, and it is not active against cysts in either the lumen or intestinal walls, or in other organs. It blocks protein synthesis in eukaryotic (but not in prokaryotic) cells. Protein synthesis is inhibited in parasite and mammalian cells, but not in bacteria. Several synthetic routes have been developed for its synthesis.

Synthesis starts with reductive amination of 2-(3,4- imethoxyphenyl)ethylamine and ethyl ester of β–(α-cyano)propylglutaric acid, followed with intramolecular cyclization. The lactam produced is reacted with phosphorus oxychloride leading to heterocyclization into the derivative of benzoquinolizine. Subsequent reaction of the product with homoveratrylamine produces the corresponding amide. Upon reaction with phosphorus oxychloride, this compound cyclizes to an isoquinoline derivative and the pyridine ring is then hydrogenated to a racemic mixture of the products, from which the desired emetine is isolated.

85

H3CO NC C2H5 H2/PtO HN O O H3CO NH2 H3CO C H OOC C H C2H5O OC2H5 2 5 2 5 H3CO COOC2H5

H3CO N H CO POCl 3 H3CO 3 O N C H Adams Catalyst 2 5 H3CO

COOC2H5 C2H5

COOC2H5 H CO 3 NH2 2.POCl3 1. 3.H2/PtO H3CO

H3CO N H3CO

C2H5 CH2 H CO 3 NH

H3CO Scheme 4.5. Synthesis of Emitine

4.1.6. Synthesis of Taltobulin:

Taltobulin is an anticancer drug which interferes with tubulin function inhibiting the formation of microtubules that form the microskeleton of cells. This process has provided some valuable antitumor activity.[7]

The synthetic strategy depends on the separate synthesis of two intermediates and combining them at the final step. One arm of the synthesis begins with the construction of acrylate- containing moiety through condensation of the t-BOC protected α-aminoaldehyde derived from valine with the arbethoxymethylene phosporane resulting in the amino ester. Removal of the protecting group is carried out under acidic condition producing the free amine. The other arm of this synthetic strategy starts with condensation of that tertiary butyl-substituted

86

aminoacid producing the protected amide which can be deprotected under acidic condition. The second arm of this synthetic strategy starts with the addition of a pair of methyl groups to the benzylic position of pyruvate through addition of methyl iodide to ketoacid in the presence of hydroxide. The addition of methylamine and diborane results in the reductive amination of the carbonyl group, and thus formation of α-aminoacid as a mixture of the two isomers. Condensation of this moiety with dipeptide formed previously under peptide forming condition resulted in the formation of amide product which is separated by column chromatography affording the desired isomer of taltobulin.

N CHO N CO C H H+ t-BOC 2 2 5 HN CO2C2H5 (C6H5)3P CO2C2H5 t-Boc

CO H CO2H 2 CO H CH3I 2 NaOH O O DCC NHCH3 t-BOC N CO2H H

O

N CO2R R N CO2C2H5 N N H H O NHCH3 O Scheme 4.6. Synthesis of Taltobulin

4.1.7. Synthesis of Perzinfote:

Perzinfote is a nonaddictive opiate alternative which is currently used to treat chronic pain.[7] Preparation starts with the reductive amination of the acetaldehyde derivative with monocarbobenzyloxy ethylenediamine leading to the disubstituted ethylenediamine (116). The amine is reacted with the commercially available cyclobutenedione derivative (117) resulting in the replacement of one of the ethoxy groups in (117) by the free amino group in (116) to afford the coupled product (118). Transfer hydrogenation of the (118) with 1,4- cyclohexadiene/Pd leads to the loss of the carbobenzoxy group and the formation of the transient primary amine (119) which is then cyclised to form eight-membered ring (120).

87

Removal of the ethyl group on the phosphorous is done by treating with trimethylsilyl bromide resulting in the formation of free phosphonic acid and thus perzinfotel (121).

NaCNBH 3 HN NHCO2CH2C5H6 H2N NHCO2CH2C5H6 115 CHO O OC2H5 P O O C2H5O P OC2H5 OC H O OC2H5 C2H5O 2 5 116 117

OC2H5 O OC2H5 O

N NHCO2CH2C5H6 Pt NH2 O N O

C H O P O 2 5 P O OC2H5 C2H5O OC2H5 118 119

HN HN O O (CH3)3SiBr N N O O

P O P O HO C2H5O OH OC2H5

120 121 Scheme 4.7.Synthesis of Perzinfote:

4.1.8. Synthesis of Namindinil:

Namindinil is used to treat hair loss in males. It has a vasodilator action improving blood circulation in hair follicle capillaries.[7] The convergent synthesis of this drug involves preparation of the complex alkyl group as a single enantiomer. The process involves the preparation of imine utilizing p-toluene sulfonic acid. The imine is then reduced with borane- THF complex yielding the secondary amine as a mixture of diastereomers. The two diastereomers were separated by column chromatography. Hydrogenolysis of the chiral auxiliary leads to the formation of the primary amine. The primary amine is added to thiourea derivative forming the required compound. 88

O 1.Column 1. p-TsOH/toluene Chromatograpgy NH2

2. BH3-THF/THF HN 2.10 mol% Pd-C/EtOH NH2 NC S N OH C NC N NH N N NH N NH 2 NaCN S NH H N

NH2 CN CN CN

Scheme 4.8. Synthesis of Namindinil.

4.1.9. Synthesis of Ezlopipant:

Ezlopipant is an antiemetic drug which is prescribed for severe nausea and vomiting associated with chemotherapy.[7] Preparation starts with the condensation of acetonitrile with ester derivative of piperidine. Nitrile group is converted to the corresponding acid under acid hydrolysis. The carboxylic acid undergoes spontaneous decarboxylation forming the corresponding ketone which is reacted with bromine yielding bromoketone. This unstable intermediate undergoes spontaneous internal displacement forming quinuclidine (172) as a quaternary salt. Debenzylation using palladium leads to the formation of quinuclidone. Reductive amination 2-methoxy-4-isopropylbenzylamine (174) affords ezlopipant (175).

89

CO2C2H5 CN CN O O + base H N

N N

167 168 170 169 Br2

O Br O N H2 N O

173 N

172

OCH3 171 174 NH2

NH OCH3 N

175 Scheme 4.9. Synthesis of Ezlopipant.

4.1.10. Synthesis of Monomorine:

Monomorine alkaloid is a trail pheromone of the widespread pharaoh’s ant Monomorium pharaonis.[8] Synthesis starts by treating amine (125) with 20% titanocene trichloride forming the imidotitanium complex (126). This compound then underwent a [2+2]-cycloaddition with the alkyne to afford intermediate (127). Ring opening and subsequent metathesis. Imine was reduced with diisobutylaluminum hydride providing amine intermediate followed by deprotection and intramolecular reductive amination resulting in the formation of the alkaloid monomorine. 90

O O O O O O a NH2 93% N Cl Ti Cp Ti N Bu Bu Cp Cl 126 125 127 H H

OO b c,d,e N 95% OO HN 72% N Bu 128 129 Bu

(a) Et3N, CpTiCl3 (20% mol); (b) DIBALH; (c) HCl; (d) K2CO 3; (e) NaCNBH3

Scheme 4.10. Synthesis of Monomorine:

4.1.11. Synthesis of Ontazolast:

Ontazolast is an antiasthmatic drug which acts as leukotrienes synthesis inhibitors.[8] Synthesis starts with the condensation of pyridine 2-aldehyde with the cyclohexylmethyl magnesium bromide forming carbinol which is oxidized with manganese dioxide to afford the ketone. Reductive amination utilizing ammonium formate/formic acid system converts the carbonyl group into the primary amine. The other half of the final product is formed through the reaction of aminotoluol with carbon disulfide in the presence of a base proceeds to the addition product, benzoxazole. The thiol at the 2 position is then replaced by halogen by reaction with phosphorus oxychloride. Combing this part with the primary amine leads to the formation of alkylated product of ontazolast.

MgBr

1. HCO2NH4 N N 2. MnO2 HCO2H OHC N O NH2

POCl N NH2 SCS N 3 Cl SH NaOH O OH O

Scheme 4.11. Synthesis of Ontazolast N HN N 91 O

4.1.12. Synthesis of Pamaquine:

Pamaquine is an antimalarial quinoline derivative. It is very effective against the erythrocytic stages of all four human malarias.[8] The synthesis of pamaquine is achieved through starting quinoline (3) which is synthesized through reaction of substituted aniline (1) with glycerol in sulfuric acid and Nitrobenzene. The resulting nitro group is then reduced by molecular hydrogen giving the free amine which is reductively aminated forming pamaquine

OH H2SO4 O OH O HO OH OH

H3CO H3CO CHO H2SO4 H3CO H3CO

NH2 N H H N 2 N NO2 NO2 NO2 NH2 1 2 3 4 O Et2N

H3CO

N HN NEt 2

5 Scheme 4.12. Synthesis of Pamaquine

4.1.13. Synthesis of Torcetrapib:

Torcetrapib is a cholesterol lowering agent.[8] Dietary cholesterol needs be esterified in order to be absorbed from the gut. The enzyme, cholesteryl ester transfer protein (CETP), then completes the absorption of cholesterol. Torcetrapib inhibit this enzyme helping to lower LDL (Low density lipoproteins) cholesterol and VLDL (Very Low Density Lipoproteins) cholesterol. It also increases the high density–good- lipoprotein cholesterol. Preparation of torcetrapib starts with the reaction of the trifluoromethylaniline (1) with propanal in the presence of benzotriazole (2) which affords aminal (3). The aminal is condensed with vinyl carbamate forming tetrahydroquinoline ring. The nitrogen group on the ring is protected as 92

ethyl carbamate by with ethyl chloroformate. Benzyl carbamate group on the nitrogen at 4 position is hydrogenolyzed using ammonium formate over palladium forming the primary amine. The chiral amine is resolved as debenzyl tartarate salt to afford (2R, 4S) isomer. Bis-trifuoromethyl benzaldehyde and sodiumtriacetoxy borohydride are used for the reductive amination of the pure isomeric amine which is then acylated with chloroformate forming the final product.

N N N H N F3C 2 CHO F3C N N vinyl carbamate NH2 1 N H 3 O O CH2C5H5 HN O CH2C5H5 HN O ClCO2C2H5 - + F3C 1. HCO NH F C 2 4 3 Pd N H N 2.resolve 5 CF3 C2H5 O O NH2 1. 6 O F3C CF3 F3C CHO O N N F C 2. CH3OCOCl 3 C2H5 O O CF3 N 7 C2H5 O O 8

Scheme 4.13.Synthesis of Torcetrapib:

4.1.14. Synthesis of Polyaminocholestanol derivatives:

Polyaminocholestanol derivatives are used as potent antibiotics for highly resistant bacterial strains (superbugs).[9] The key step in their synthesis is reductive amination. Different titanium sources were tested in different solvents. The highest des were obtained when i Ti(O Pr)4 was used with MeOH and NaBH as hydride source. Different amine and diamaines were tested and the des were higher than 95% but the yields were poor to mediocre (6-77%).

93

i 1) Ti(O Pr)4,MeOH 20 oC, 5-6 h o 2) NaBH4,-78 C,2 h RNH 2 3) K 2CO3, MeOH/THF (1:1) AcO AcO NH2R O rt, 12 h H H

Scheme 4.14. Synthesis of Polyaminocholestanol Derivatives.

4.1.15. Synthesis of Piperazinylpropylisoxazoline Analogues:

Piperazinylpropylisoxazoline analogues are potent ligands for dopamine receptors which have strong influence on the general psychological condition.[10] Synthesis is accomplished through the reductive amination of enantiomerically pure (S) or (R) aldehyde (1) with different piprazine derived amines. The (R) isomer showed higher potent activity for dopamine receptors.

O

H R1 O (S)-1 N

R2 N (S)-1 N or N R1 R2 O (R)-1 N N or H N N R R1 2 O N Scheme 4.15. Synthesis of piperazinylpropylisoxazoline analogues.

94

4.1.16. Synthesis of Ritonavir and Lopinavir:

Ritonavir and lopinavir are HIV-protease inhibitors. Their synthesis depends on the synthesis of chiral aminoalcohols.[11] Titanium isopropoxide with polymethylhydrosiloxane (hydride source) and p-anisidine were utilized for the synthesis of aminoalcohols under reductive amination conditions. Aliphatic, cyclic, as well as aromatic and heteroaromatic hydroxyketones were tested showing good to high yields (76-89%) with good stereoselectivities (de 72-86%). H+ OH O NR3H2 TiLn OH NR3H i O R1 R2 Ti(O Pr)4 R1 R1 R2 PMHS N R3 R2

S CH O 3 H O N N HN N N NH OH H NH OH H O N O S O N O Ph O N Ph O Ph Ph

Ritonavir Lopinavir

Scheme 4.16. Synthesis of Aminoalcohols the Core of Ritonavir and Lopinavir:

4.1.17. Synthesis of Tetrahydrocarbazoles:

Tetrahydrocarbazoles is an efficient compound for the treatment of human papillomaviruses (HPVs).[12] HPV infection is considered the most common sexually transmitted disease throughout the world. There are over 5.5 million new cases of sexually transmitted HPV in the United States each year, with at least 20 million people currently infected. The chiral centre in the molecule is the α-chiral amine in which its synthesis represents the key step in the synthesis of this tetrahydrocarbazoles. In their initial attempts for building this chiral moiety they tested chemical resolution. The best results were obtained utilizing dibenzoyl-D- tartaric acid leading to 86% ee with only 13% yield. This low yield encouraged them to shift to the asymmetric synthesis for building the chiral centre. Noyori catalyst was tested for the

95

reductive amination of the prochiral ketone resulting in 80% ee with 60% yield. They also tested reduction of isolated imine resulting in the same enantioselectivity and with unacceptable chemical purity. Higher ees are usually required for pharmaceutical development levels. Later they tested the chiral auxiliary approach for the synthesis of chiral amine moiety. They tested different derivatives of MBA and phenylglycinaol. They reported 86% yield with 90% de which was improved by crystallization. The instability of substituents under hydrogenolysis standard conditions was another challenge. BCl3 and BBr3 gave the cleanest N-debenzylation without affecting the chloro substituent and tetrahydrocarbazole moiety.

o HCO2NH4,MeOH,60 C Cl Cl N NH2 N O H H 80% ee, yield 60%

SO2 NH3 N (7N in MeOH) Ru TsOH N Cl H2

[RuCl2(benzene)]2 Cl Cl o HCO2NH4,MeOH,65 C N NH2 N NH H H 81% ee PPh2 PPh2 X X 1. NaBH4 p-TsOH, or conc.HCl EtOH, -30 oCtoRT Cl Cl toluene, reflux Cl 2. HCl (R) N O X N HN H N N H Cl H H2N Y Y

Y OMe Cl Cl O 1. BCl ,DCM,0oC Cl 3 N NH2 T3P (50% in EtOAc) N HN H o H N i-Pr2NEt, DCM, 0 C N HN N COOH,i-PrOH N COOH 60-87% H 2. Pr O O 80-92% T3P= P P Tetrahydrocarbazoles ee 99.2% O Pr ee >99.5% O O P Scheme 4.17. Synthesis of Tetrahydrocarbazoles: Pr O

96

4.2. Conclusion

Different important pharmaceutical and natural products are prepared industrially utilizing reductive amination as a key step in their preparation. Reductive amination is the method of choice for incorporating amino group in the drug entity as it is a single step process which is highly preferable from the industrial point of view. Most of the developed methodologies for the reductive amination utilized boran as a reducing agent which suffers from many drawbacks as the large toxic waste production. In the last ten years scientists focused their efforts on developing an asymmetric version of reductive amination utilizing environmentally friendly hydride source as molecular hydrogen.

4.3. References:

1] D. L. Romero, R. A. Morge, C. Biles, N. Berrios-Pena, P. D. May, J. R. Palmer, P. D. Johnson, H. W.Smith, M. Busso, C. -K Tan,R. L. Voorman, F. Reusser, I.W. Althaus, K. M. Downey,A. G. So, L. Resnick, W.G. Tarpley, P. A. Aristoff, J. Med. Chem. 1994, 37, 999. [2] D. S. Johnson, J.J. Li, The Art of Drug Synthesis, Wiley & Sons, New Jersey, 2007. [3] M. Lautens and T. Rovis, J. Org. Chem. 1997, 62, 5246; [4] E. J. Corey and T. G. Gant, Tetrahedron Lett. 1994, 35, 5373. [5] S. Takano, M. Sasaki, H. Kanno, K. Shishido, K. Ogasawara, J. Org. Chem.1978, 43, 4169. [6] T. Fujii, S. Yoshifuji, Tetrahedron 1980, 36, 1539 . [7] D. Lednicer, The Organic Chemistry of Drug Synthesis, Wiley & Sons, New Jersey, 2008. [8] D. Lednicer, Strategies for Organic Drug Synthesis and Design, Wiley & Sons, New Jersy, 2009. [9] C. Loncle, C. Salmi, Y. Letourneux, J. M. Brunel, Tetrahedron 2007, 63, 12968. [10] J. Y. Jung, S. H. Jung a, H. Y. Kohb, European Journal of Medicinal Chemistry 2007, 42, 1044. [11] D. Menche, F. Arikan, J. Li, S. Rudolph, Org. Lett. 2007, 9, 267.

97

[12] S. D. Boggs, J. D. Cobb, K. S. Gudmundsson, L. A. Jones, R. T. Matsuoka, A. Millar, D. E. Patterson, V. Samano, M. D. Trone, S. Xie, X. –M, Zhou, Org. Process Res. Dev. 2007, 11, 539.

98

Chapter 5 Stoichiometric Use of Ytterbium Acetate in Reductive Amination.

5.1. Introduction.

The general lack of literatures describing the synthesis of alkyl-alkyl chiral amines has encouraged us to try developing a new methodology for their synthesis. Of the commonly explored strategies for the α-chiral amine synthesis (described earlier), reductive amination has the advantage of being a stepwise efficient methodology with low waste generation.[1]

The use of atom economic environmentally friendly hydride source (molecular hydrogen) is routinely practiced by the pharmaceutical industries for achiral C-N bond formations but not often reported in literatures. (R)- and (S)-α-methylbenzylamine (α-MBA) were chosen as the chiral amine auxiliary among. Of course there are other available auxiliaries such as (R)-and (S)- phenylglycinol, (R)- and (S)- phenylglycine amide or (R)- or (S)- t-butylsulfinylamide. α- MBA was chosen for three reasons: it is inexpensive, already in use by the pharmaceutical industries and the cleavage (hydrogenolysis) of this auxiliary is well established high yielding process.[2]

Table 5.1. Sigma-Aldrich Quote may 2006 for Two Common Chiral Ammonia Equivalents. [3]

Chemical Name Quantity(kg) Price (US dollars)

(S)-N-tert-butansulfinamide 1.0 13.125.00

(R)-N-tert-butansulfinamide 1.0 13.125.00

99

(S)-α-methylbenzylamine[(S)-α-MBA] 1.0 798.00

(R)-α-methylbenzylamine[(S)-α-MBA] 1.0 2.220.00

This auxiliary has been previously explored in literature for the synthesis of chiral amines. The previous methods were based on the reduction of N-α-methylbenzyl ketimines, not on reductive amination.[4]

For reductive amination, the principal side-reaction is the formation of an alcohol from the competing hydrogenation of the carbonyl starting material. It could be stated that a desirable attributes of an effective reductive amination method is the one capable of using atom efficient reducing reagents and ideally do not allow alcohol formation. In that sense, molecular hydrogen is an ideal source of hydride and the reduction of imine or iminium ion intermediates must be fast relative to the reduction of the starting carbonyl compounds. Thus, in reductive amination the correct combination of hydride source, catalyst and additives is an important prerequisite for its success. Few reports are available and they require further development for optimum reactions. It is a two step process for the final α-chiral primary amine inhibiting any possibility of imine isolation which is time consuming and low yielding process. The prochiral ketone is converted to the corresponding secondary amine in a good to high diastereoselectivity and yield in the presence of Lewis acid/(R)- or (S)-α-MBA/H2. The absolute configuration of the major and minor diastereomer depends on the absolute configuration of the chiral amine source. This amine can then be purified by column chromatography or by crystallization. The chiral primary amine is produced from the secondary amine through hydrogenolysis.

As motioned before, Nugent et al have succeeded to establish a new methodology for synthesizing α-chiral amines. Reaction conditions as hydrogen pressure, temperature and time were milder compared to other available methodologies. Reaction time is short compared to other methodologies adding to the advantages of this strategy. The use of minute quantities of metal catalysts (Pd and Pt) is another advantage. Another factor which makes this reaction attractive is the low hydrogen pressure needed; almost 80% of all ketones are hydrogenated at 8.0 bar and also at room temperature. Usually other methodologies require the use of higher (up to 100 Bar).

100

The unique feature of this methodology is the ability to use a wide variety of structurally different substrates. Aliphatic and aromatic ketones with different steric and electronic environment are successfully reductively aminated in this reaction. The diversity of the used ketones will help to broad the scope of the reaction applications.[5]

The success in synthesizing a wide variety of amines by this methodology encouraged us to investigate other commercially available Lewis acids. I thought that changing the Lewis acid will have an effect on the reaction based on our previous results. The use of titanium isopropoxide as a Lewis acid has many advantages and some disadvantages. It is cheap and available in kg quantities and it is already in use for in industrial processes. On the other hand it is moisture sensitive and it should be used in stoichiometric quantities which complicates the work up process.

Different Lewis acids were tested and some of the tested Lewis acids showed promising results. The most pronounced results obtained through using stoichiometric quantity of ytterbium acetate. The use of this Lewis acid improved the de of the reaction tremendously. The effect of this Lewis acid was significant when aliphatic ketones were used as substrates. The increase in the de reached in certain cases more than 10%. Ytterbium acetate is a solid Lewis acid and less moisture sensitive compared to titanium isopropoxide which makes it easier in handling. In the following paragraphs I will try to give a brief overview about ytterbium and its applications in organic synthesis.

5.1.1. Ytterbium.

5.1.1.1. Electronic Overview:

Ytterbium is one of the Lanthanides rare earth metals. Symbol Yb; atomic number 70; atomic weight 173.04; valence +2, +3; atomic radius 1.945Å; ionic radius, Yb3+ 0.868Å and 0.98Å for CN 6 and 8; respectively; seven naturally occurring stable isotope: Yb-170 (3.05%), Yb- 171 (14.32%), Yb-172 (21.93%), Yb-173 (16.12%), Yb-174 (31.84%), Yb-176 (12.72%); twenty-three artificial radioactive isotopes in the mass range 151-167, 169, 175, 177-180; the

101

longest-lived radioisotope Yb-169, t1/2 32.03 days; shortest-lived radioisotope Yb-154, 0.40 second.

5.1.1.2. Ytterbium Discovery:

Ytterbium was discovered in 1878 by J. C. G. de Marignac. The element got its name from the Swedish village Ytterby where this rare earth first was discovered. In 1907, Urbain separated ytterbia into two components, neoytterbia (oxides of ytterbium) and lutecia (lutecium). The first preparation of metallic ytterbium was achieved by Klemm and Bommer through reduction with potassium metal resulting in an impure ytterbium metal (mixed with potassium chloride). Daane, Dennison, and Spedding were the first to prepare the pure metal in 1953 in gram quantities. Abundance of ytterbium in the earth’s crust is estimated to be 3.2 mg/kg. Up till now the metal had showed very little applications on the commercial level. In elemental form it can be used as a laser source, a portable x-ray source, and as a dopant in garnets. When added to stainless steel, it improves grain refinement, strength, and other properties. Some other applications, particularly used as oxides mixed with other rare earths, include carbon rods for industrial lighting, in insulated capacitor and in glass industry. Its radioactive isoptope is used in detection of metal perfection.

5.1.1.3. Ytterbium Reactions:

Ytterbium metal reacts with oxygen above 200°C forming two oxides, the monoxide, YbO, and more stable sesquioxide, Yb2O3. The metal dissolves in dilute and concentrated mineral acids. At ordinary temperatures, ytterbium, similar to other rare earth metals, is corroded slowly by caustic alkalies, ammonium hydroxide, and sodium nitrate solutions. The metal dissolves in liquid ammonia forming a deep blue solution. It can react slowly with halogens at room temperature but progress rapidly above 200°C forming ytterbium trihalides. All the trihalides; namely, the YbCl3, YbBr3, and YbI3 with the exception of trifluoride, YbF3, are hygroscopic and soluble in water. Ytterbium forms many binary, metalloid, and intermetallic compounds with a number of elements when heated at elevated temperatures. It can form salt with organic acid triflic or acetic acid. These salts and the salts with halogens are used as Lewis acids in different organic transformations.[6]

102

Ytterbium triflate is the most commonly used form of ytterbium in organic synthesis. It is used in Michael addition of β-Ketoester in water,[7] synthesis of ethyl arylacetates,[8] hydrolysis of tritylamines and trityloxy compounds to the corresponding amines and alcohols,[9] electrophilic substitution of indoles for the synthesis of unnatural tryptophan derivatives,[10] Friedel–Crafts reaction of arylidenecyclopropanes,[11] synthesis of polyhydroquinoline derivatives,[12] and synthesis of substituted imidazoles.[13]

5.1.1.4. Ytterbium Acetate

Ytterbium acetate is a moderately water soluble crystalline ytterbium source that decomposes to ytterbium oxide on heating. Acetates are excellent precursors for the production of ultra high purity compounds and certain catalysts and nanoscale (nanoparticles and nanopowders) materials. All metallic acetates are inorganic salts of a metal cation and the acetate anion. The acetate anion is a univalent (-1 charge) polyatomic ion composed of two carbon atoms ionically bound to three hydrogen and two oxygen atoms (Symbol: CH3COO) for a total formula weight of 59.05. Ytterbium acetate is applied to fibre amplifier and fibre optic technologies and in lasing applications. It has a single dominant absorption band at 985 in the infra-red useful in silicon photocells to convert radiant energy to electricity.

Two examples were reported in literatures for the applications of ytterbium acetate in organic chemistry. Fujiwara reported the use of ytterbium acetate for the synthesis of acetic acid in water. The method depends on the carboxylation of methane with carbon monoxide using ytterbium acetate. Sodium hypochlorite or hydrogen peroxide was used as the oxidant in this reaction. The catalytic activity was improved by the addition of transition-metal salts such as manganese acetate. The best result was achieved at a ratio of manganese acetate to ytterbium acetate of 1:10.[14]

Oshima reported the use of ytterbium acetate as additive in oxidation of alcohols to aldehydes and ketones utilizing iodosylbenzene as oxidant. Mixing ytterbium salt with isodosylbenzene and alcohol in 1,2-dichloroethane and heating the mixture at 80 °C for 3.5 hours provided the ketone products in good to excellent yields.[15]

103

From the previous discussion it is obvious that ytterbium acetate was rarely reported in organic chemistry. In our initial studies I tested different available Lewis acids with benzyl i acetone as the ketone substrate. The original reported de was 80% using Ti (O Pr)4 / Ra-Ni, α-

MBA, H2 (120 psi), DCM. Benzyl acetone is an excellent substrate for screening as it is considerably cheap, available in large quantities with high purity from chemical suppliers and has high molecular weight facilitating its work up. Some of the tested Lewis acids and results obtained are summarized in (table 5.2).

Table 5.2.a Different Lewis Acids Tested for the Reductive Amination of Benzyl Actone.

Lewis Acid Ketone Left Alc. Formed Imine Left de%

Dysprosium(III) acetate - - 1.36 77.5 hydrate

Lanthanum(III) acetate - - 1.3 76.44 hydrate

Lanthanum(III) 90 - - NA trifluoromethanesulfonate

Cerium 90 - - NA trifluoromethanesulfonate

Cerium(III) acetate 1.6 - 1.29 76.9 hydrate

Indium(III) acetate 29.9 - 13.1 79.03

Indium(III) 90 - - NA trifluoromethanesulfonate

104

Ruthenium(III) chloride 31.7 4 4.2 71.29

Cerium(IV) sulfate 90 - - NA tetrahydrate

Cerium(III) chloride 90 - - NA heptahydrate

Copper(II) 90 - - NA trifluoromethanesulfonate

Zinc 2.8 - - 66.5 trifluoromethanesulfonate

Yttrium(III) 68.17 - - 70.7 trifluoromethanesulfonate

Bismuth(III) 90 - - NA trifluoromethanesulfonate

Dysprosium(III) 63.5 - - 75.5 trifluoromethanesulfonate

Scandium(III) triflate 90 - - NA

Iron(III) bromide 90 - - NA

Aluminum chloride 90 - - NA

105

Bismuth(III) acetate 34.8 - - 68.49

Yttrium(III) 2.86 - - 76.79 trifluoroacetate hydrate

Ytterbium(III) acetate - - - 85 hydrate

Ytterbium(III) acetate 1.5 - - 84 tetrahydrate a All reactions performed using 1.0 mmol of Benzylacetone, 1.1 mmol of (S)-(−)-α-Methylbenzylamine, 1.1 equiv of the indicated Lewis acid, room temperature, 120 psi (8.3 bar) of H2, 100 wt % Raney Nickel, and methanol as a solvent. All components (except the Raney Ni and H2) are added together and pre-stirred for 30 min. The heterogeneous hydrogenation catalyst (Raney Ni) is then added and the system pressurized with 8 bar of H2. The indicated data is at 12 h of reaction from the onset of hydrogenation.

The data from the table indicates that most of the used Lewis acids showed inferior results i compared to Ti(O Pr)4. The reaction did not even proceed using certain Lewis acids. Only ytterbium acetate hydrate and ytterbium acetate tetrahydrate showed improvement in the de. These results were encouraging to test ytterbium acetate hydrate or tetrahydrate under different reaction conditions aiming for further improvement in the de. Different solvents were tested. The best solvent in terms of reaction rate was methanol. This may be due to partial solubility of ytterbium in methanol. The results of solvent screening are summarized in the (table 5.3).

Table 5.3. Solvent Screeing with Ytterbium Acetate Hydrate.a

Solvent Ketone Left% Alc. Formed Imine Left de%

Methylene chloride 77 - - 78

Isopropanol 15 - - 76

Toluene 23 - - 80

106

t-Butyl methyl 15 - - 82 ether

Hexane 60 - - 75

THF 20 - - 86

Methanol - - - 81 a All reactions performed using 1.0 mmol of Benzylacetone, 1.1 mmol of (S)-(−)-α-Methylbenzylamine, 1.1 equiv of Yb(OAc)3, room temperature, 120 psi (8.3 bar) of H2, 100 wt % Raney Nickel, and solvent as indicated. All components (except the Raney Ni and H2) are added together and pre-stirred for 30 min. The heterogeneous hydrogenation catalyst (Raney Ni) is then added and the system pressurized with 8 bar of H2. The indicated data is at 12 h of reaction from the onset of hydrogenation.

The use of THF improved the de but the reaction was slower. Other solvents showed slower reaction rate with low de. The high de resulting from the use of THF and the fast reaction rate resulting from the use of methanol was the driving force to test the solvent combination of THF-MeOH (1:1). The rate of the reaction was acceptable with high de (87-89%). Other solvent combinations also were tested (table 5.3).

Table 5.3. Screening of Different Solvent Combinations with Ytterbium Acetate Hydrate.a

Solvent Ketone Left% Alc. Formed Imine Left de%

THF-DMF >90 - - NA

DCM-MeOH 77 - - 78

Isopropanol- 15 - - 76 MeOH

Toluene-MeOH 24 - - 82

EtOAc-MeOH 6 - - 86

THF-MeOH 2 - - 89

DME-MeOH 50 - - 85

DCE-MeOH 51 - - 80

DEE-MeOH 55 - - 84 a All reactions performed using 1.0 mmol of Benzylacetone, 1.1 mmol of (S)-(−)-α-Methylbenzylamine, 1.1 equiv of Yb(OAc)3, room temperature, 120 psi (8.3 bar) of H2, 100 wt % Raney Nickel, and solvent as 107

indicated. All components (except the Raney Ni and H2) are added together and pre-stirred for 30 min. The heterogeneous hydrogenation catalyst (Raney Ni) is then added and the system pressurized with 8 bar of H2. The indicated data is at 12 h of reaction from the onset of hydrogenation.

Ethyl acetate-MeOH and THF-MeOH combinations showed the highest possible de with acceptable reaction time (12 h). The use of THF-MeOH resulted in slightly higher de compared to EtOAc-MeOH encouraging us to proceed using this solvent combination for the rest of the study. The addition of an anhydrous MgSO4 or NaSO4 as desiccants to the reaction mixture before hydrogenation did not have any effect on the reaction profile.

After using several bottles of Yb(OAc)3, which is sold and described as a semihydrated form (Sigma-Aldrich catalogue no. 544973), I noted that it was sometimes free flowing while other bottles from the same lot were not. In our attempt to eliminate any variation of results and to get consistent reaction profile for all substrates, I decided to dry Yb(OAc)3 powder obtained from the commercial supplier. The powder was high vacuum dried to constant weight at 80 °C (12 h). The dried powder was kept in airtight container and used for further reactions. Through this drying procedures all reactions results were reproducible and constant reaction profile was obtained. In the rest of this study the term “dry Yb(OAc)3” means dried as just stated. The dried Yb(OAc)3 could be stored in a dry screw cap glass bottle at room temperature. The container could be repeatedly opened to the atmosphere (at least 6 times without detrimental effect) and the desired quantity of Yb(OAc)3 weighed out without the need for a glovebox. This is one of the significant advantage of Yb(OAc)3 use in reductive amination compared to other air sensitive Lewis acids. Moisture and air stability of Yb(OAc)3 will open the door for its applications on industrial scale.

After initial optimization of the reaction using benzylacetone as the substrate other ketone substrates were also tested. 2-octanone is another example of 2-alkanones which has been i reductively aminated with Ti(O Pr)4 system with a de of (72%), was one of the interesting substrates to be tested with Yb(OAc)3 system. Solvent screening was carried also to this substrate utilizing all findings obtained from benzylacetone results. For example 2-octanone, in MeOH, was fully consumed within 8 h providing the secondary amine in 82% de in the presence of Raney-Ni (Scheme 5.1). When the solvent was changed to THF, the stereoselectivity increased to 87-88% de, but 24 h were required to completely consume the

108

2-octanone starting material. When the binary solvent system of MeOH-THF (1:1) was examined, an 86% de was consistently achieved with a fast reaction time of 10-12 h. The same solvent systems which proved to be efficient in the reductive amination of benzylacetone were also useful in the reductive amination of 2-octanone. From solvent screening studies it was obvious that the presence of MeOH in the solvent mixture is essential for the fast reaction rate. Replacement of THF in THF-MeOH mixture with other solvents resulted in lower de or/and prolonged reaction time. The same de was obtained through replacing THF in the THF-MeOH system with toluene, Et2O, or 1,3-dioxolane but with moderately longer reaction times. Replacing MeOH with EtOH in THF-MeOH system resulted in the same de but the reaction time was longer (24h). To ensure that this high diastereoselectivity is maintained and no racemization is occurring, I hydrogenolyzed the reductive amination product to ensure that the enantiopurity of the primary amine is preserved (scheme 5.1.). This level of diastereoselectivity represents a 15-16% increase in the de over the best previously reported for 2-octanone and α-MBA.

O (S,S)-2d HN Ph NH 1d Yb(OAc)3,MeOH-THF Pd-C (S)-3d 2 + H2N Ph H (60 psi) Raney-Ni, H2 (120 psi) 2 (S)-α-MBA 86% de 85% ee Scheme 5.1. Two-Step Procedure for Producing (2S)-Aminooctane in High ee.

The high de obtained in reductive amination of benzylacetone and 2-ocatnone utilizing

Yb(OAc)3 encouraged us to test and evaluate other ytterbium salts present commercially. As mentioned before Yb(OTf)3 is the most common derivative of ytterbium used in organic synthesis. When Yb(OTf)3 was used in reductive amination of 2-ocatnone alcohol was the major product. No amine was detected after 10 h or the reaction. YbCl3 provided the product in 66% de, but in only 23 area % (GC) after 24 h (Table 5.4, entries 2 and 3). The use of i highly expensive salt of ytterbium which is Yb(O Pr)3 resulted also in alcohol formation and no secondary amine was detected. From previous findings it is obvious that Yb(OAc)3 is the best form of ytterbium to be used in reductive amination. Of course to eliminate any doubts regarding free acetate in solution modifying the heterogeneous metal surface of the catalyst, acetate salts were tested alone in the reaction. The addition of NaOAc was examined (Table 5.4, entry 9). In the event, gross quantities of the alcohol by-product resulted, making this

109

simplified scenario less likely. It is clear that ytterbium and the acetate ligand together are needed for the efficient reductive amination process.

Historically the use of acetic acid as Brønsted acid in reductive amination is well established on laboratory and industrial scales. It is cheap, available in kilogram quantities, easy to handle and usually used in catalytic quantities. Testing acetic acid in our reaction will help to i establish another reference point to our work after the comparing with Ti(O Pr)4 system. The use of (0.2, 0.5 and 1.0 equiv) of acetic acid inhibited alcohol formation but did not show any de improvement. The addition of acetic acid to Yb(OAc)3 reaction resulted in the reduction of diastereoselectivity of the secondary amine (72%). All these findings proved that Yb(OAc)3 is a unique Lewis acid for reductive amination.

As mentioned previously that reductive amination does not require imine separation and purification which is achieved by the addition of proper Lewis acid. Some older reports simplify the role of Lewis acids in reductive amination to the level of an efficient desiccant promoting in situ imine formation in high yield. Extensive studies on the mechanism of reductive amination and intermediates structures contradicted these simplified speculations and proved that Lewis acids have greater role than efficient desiccants. To study this effect more closely, I examined some traditional desiccants. When Yb(OAc)3 (1.1 equiv) is replaced by MgSO4 (5 equiv) or 4Å molecular sieves (4 wt equiv), all vacuum oven dried at 150 °C for 15 h before use, not only low diastereoselectivities were observed (Table 5.4, compare entries 1, 6, 7), but gross amounts of 2-octanone were reduced to the alcohol by- product. In relation to these results, alcohol by-product formation could be significantly i suppressed when Ti(O Pr)4 (1.25 equiv) was used, but the de remained low. These combined findings clearly establish Yb(OAc)3 as fulfilling a greater role than that of a simple desiccant.

a a Table 5.4. Initial Study of the Role of Yb(OAc)3 in the Reductive Amination of 2-Octanone.

Amine 2d entry additive time (h) (S)-α-MBA (%)b 2-octanol (%) yield (%) de (%)

c 1 Yb(OAc)3 10 0.7 0.3 90.4 86.1

d 2 YbCl3 23 77.4 0.0 22.5 66.2

110

d [f] 3 Yb(OTf)3 9 - 96.4 3.5 -

i e 4 Ti(O Pr)4 10 2.8 11.4 82.3 67.0

i e 5 B(O Pr)3 10 14.7 24.4 59.6 71.0

6 MgSO4 12 19 29.3 52.0 70.5

7 4 Å M.S. 12 18.7 28.8 52.5 69.6

8 none 12 24.2 34.6 41.1 70.8

9 NaOAc 23 26.5 43.3 29.7 70.8

10 HOAc 12 4.3 1.0 94.1 72 a (S)-α-MBA (2.5 mmol, 1.0 equiv), 2-octanone (1.2 equiv), and an additive (entries 1-5, 9, and 10, 1.1 equiv of additive, 4Å molecular sieves (4 wt equiv), or MgSO4 (5.0 equiv)) are stirred in MeOH (1.0 M) for 30 min at rt, then THF (final molarity 0.5 M) and Raney-Ni are added, and the reaction pressurized with H2 (120 psi). All data is based on GC area % analysis. b Sum of (S)-α-MBA and imine remaining at the indicated time. c This d result is when Yb(OAc)3 has not dried, 2-aminooctanone was noted in 6.5 area %. These Yb salts were only e i examined in MeOH. Under the optimal Yb(OAc)3 conditions used here, the Ti(O Pr)4 and B(O-i-Pr)3 Lewis acids did not provide optimal results. f Not integrated, to emphasize the presence of the dominant alcohol by- product.

5.1.2. Commercial Yb(OAc)3 vs Dried Yb(OAc)3:

Initial studies were performed using the commercially available Yb(OAc)3 which had a different physical appearance (powder flowability) in each bottle. Initial results for the reductive amination of 2-octanone with Yb(OAc)3 in THF-MeOH were promising in terms of high de, but the isolated (chromatographic) yield of secondary amine was 75%. This yield is considered mediocre for a single step process. GC analysis revealed the presence of unknown peak with 10-12 area %. Reaction was performed at larger scale (8 mmol) allowing the chromatographic separation of this compounds. The isolated compound was analyzed using (1H and 13C NMR) which revealed that this compound may be as 2-aminooctane. The conditions under which this primary amine by-product formed were when (S)-α-MBA was used as the limiting reagent [(ketone 1.2 equiv and undried Yb(OAc)3 (1.1 equiv)], 2-aminooctanone is consistently observed at 6-7 area % (GC).

Surprisingly the vacuum drying of Yb(OAc)3, reduced the amount of 2-aminooctane dramatically to 1-3 area % (GC). Of course the isolated yield of the secondary amine increased 111

to 86%, and the de was consistently reported between 87-88%. Aiming to understand the concept behind this effect, H2O or AcOH (50 or 100 mol %) was added to the reaction mixture containing dried Yb(OAc)3. The amount of 2-aminooctane formed increased also to ~10 area %

(GC). For this reason I decided to conduct all experiments using only dried Yb(OAc)3. Another significant finding from this observation was the importance of using ketone as the limiting reagent instead of α-MBA (1.1 equiv) to obtain the optimal de and yield. The use of ketone as a limiting reagent adds to the advantages of our methodology as it reduces the expenses of starting materials especially when using expensive ketones on larger scale.

Regarding the formation of 2-aminooctane, our initial speculation was that the reductive amination products (S,S)- and (R,S)-secondary amine (scheme 5.1.) were slowly being hydrogenolyzed under the reaction conditions, but chiral GC analysis (trifluoroacetamide derivative) established the primary amine side product as a racemate. Based on our prior i experience, regarding the stereoinducing capabilities of (S)- or (R)-α-MBA with Ti(O Pr)4 and 2-octanonone, I considered it unlikely that a racemate would form if a sequential reductive amination-hydrogenolysis scenario had occurred. This led us to examine the possibility that (S)-

α-MBA was being hydrogenolyzed by Raney-Ni, in the presence of Yb(OAc)3, in a small but significant amount, and thereby producing ammonia and ethylbenzene in situ. The subsequent, but non-productive, reductive amination of ammonia with 2-octanone would then account for racemic 2-aminooctane formation. In an effort to support this hypothesis, several reactions were examined by GC in an effort to identify ethylbenzene (relative to an authentic reference standard), but none was ever observed. To further corroborate those findings, I treated α-MBA

(in the absence of 2-octanone) with Raney-Ni/H2. Several repetitions of this experiment failed to allow the identification of ethylbenzene. The major obstacle for identification of ethylbenzene was its low boiling point (136 °C). Any low quantities produced could potentially evaporate on quenching the reaction aliquot at room temperature (work-up: drop aliquot into sat. aq NaHCO3/EtOAc). Hoping to reduce ethyl benzene evaporation the quenching solution of aqueous NaHCO3/EtOAc was cooled to 0 °C. Unfortunately, this did not allow the observation of ethylbenzene by GC.

While [1,3]-proton shift of imine is known, it is to our knowledge only accomplished under the presence of a strong base.[16]

112

If a [1,3]-proton shift of the initially formed imine occurred, followed by hydrolysis, 2- aminooctane and acetophenone would result. When 2-octanone (1.0 equiv), (S)-α-MBA (1.1 equiv), and Yb(OAc)3 (1.1 equiv) were added to THF-MeOH (standard reaction conditions), and stirred in the absence of Raney-Ni and H2, a very small quantity of new compound (<2 area %) with very similar retention time to acetophenone (as compared to an authentic sample) was formed. Addition of water to the reaction, which allowed more of the 2-aminooctane to form when Raney-Ni and H2 was present, did not increase the amount of the -new compound- believed to be acetophenone. No further studies regarding the origin of 2-aminooctane formation were pursued and as stated earlier high vacuum drying of Yb(OAc)3 suppressed this by-product (1-3 area %, GC) from forming.

5.1.3. Optimized Conditions and Useful Substrate Range:

After optimizing reaction conditions, different substrates were tested to understand the scope and limitation of our developed methodology. The following reaction conditions were applied to all further ketone substrates in (table 5.5). The ketone (1.0 equiv), α-MBA (1.1 equiv), and dried Yb(OAc)3 (1.1 equiv), are prestirred in MeOH (1.0 M) for 20-30 min before the addition of Raney-Ni, as a slurry in THF (final molarity 0.5 M), and pressurization with hydrogen (120 psi) at 22 oC.

Table 5.5. Examples of Ketones Showing Amplified Diastereoselectivity.a

entry 2-alkanone 1 amine product 2 yield de change in de (%)b (%)c (%)d

O 1 1a HN Ph 2a 78 94 1

2 1b O 2b HN Ph 77 93 6e

3 1c O 2c HN Ph 87 89 9

113

4 1d O 2d HN Ph 86 87 15

5 1e O 2e HN Ph 82 85 14

6 1f O 2f HN Ph 80 79 5

a Ketone (2.5 mmol, 1.0 equiv), dried Yb(OAc)3 (1.1 equiv), (S)-α-methylbenzylamine (1.1 equiv) equiv), b Raney-Ni, H2 (120 psi), MeOH-THF (1:1) 0.50 M, 22 °C, 12 h. Isolated yield of both diastereomers after chromatography. c determined by GC analysis of crude product 2. d Compared to the best previously reported results. Pinacolone is a Pt-C substrate and requires a T= 50 °C over 22 h. The 6% increase in de only represents an increase over the best reported reductive amination procedure, vs a previously reported stepwise method there is no change in de.

Through examining the results presented in table 5.5 It seems that our methodology is highly efficient for reductive amination of ketones having the general class RSC(O)CH3, linear 2- alkanones. The subscript serves as a generic reference to the steric bulk of the substituent:

RS= small (any straight chain alkyl substituent, but not a methyl group); RM= medium, e.g. –

CH2CH2Ph or -CH2CH(CH3)2; RL= e.g. -Ar, -i-Pr, -c-hexyl.

For example, the longer straight chain 2-alkanones, e.g. 2-octanone (1d) and 2-hexanone (1e), showed dramatic improvements in de, 15% and 14% respectively, with good isolated yield (table 5.5, entries 4 and 5). As mentioned before the highest reported de for the i reductive amination of 2-octanone (1d) with α-MBA in the presence of Ti(O Pr)4/Raney-

Ni/H2, was 72%. Aiming to prove that the addition of Yb(OAC)3 had a dramatic effect on the de of amine product, I synthesized the ketimine of 2-octanone. This ketimine was reduced with Raney-Ni/H2 in THF-MeOH (1:1) providing 2d in only 64% de.

As the chain of 2-alkanone gets shorter as the steric bulkiness gets smaller reducing the enhancement effect of Yb(OAc)3 addition. The short chain 2-butanone (1f) showed a small but consistent and significant 5% increase in de vs the best previously reported result i (Ti(O Pr)4/CH2Cl2/Raney-Ni: 74% de). Shifting to substrates having medium sized R substituent residing on 2-alkanone, RMC(O)CH3helps to define the boundary substrates for

114

enhanced stereoselectivity. For example when the γ-branched benzylacetone (1c) was examined a 9% increase in de was observed vs the previous best reported methods. The use of Yb(OAc)3 for reductive amination of β-branched i-butyl methyl ketone (1a), did show any significant improvement in stereoselectivity (1% increase) (table 5.5, entry 1). Also α- Branched 2-alkanones, e.g. i-propyl methyl ketone or cyclohexyl methyl ketone, which have i been reductively aminated with a very high diastereoselectivity using Ti(O Pr)4 (>98% de), did not show any improvement when using Yb(OAc)3.

Examination of alkyl-aryl ketones, e.g. acyclic acetophenone or cyclic benzosuberone, or a non-2-alkanone, e.g. i-propyl n-propyl ketone, proved problematic. Acetophenone required higher temperature, 60 °C with 120 psi of H2, and produced the product in 92% de, but with large amounts of the corresponding alcohol noted (~20 area %, GC). For benzosuberone and i-propyl n-propyl ketone repeated attempts to obtain the intended product by heating and/or increasing the hydrogen pressure failed.

i The previous substrates were successfully reductively amianted using Ti(O Pr)4 with good yield and de. As mentioned previously the heterogeneous catalyst used was Raney-Ni. Pt-C was tested for the reductive amination of i-propyl n-propyl ketone, a 35 area % of the product was noted in 76% de (GC). This ketone has been previously reductively aminated with i Ti(O Pr)4/Raney-Ni providing the desired product in 76% yield and 87% de.

From previous studies done in our group, it was noted that ketone substrates with an α- tertiary carbon cannot be reductively aminated using Raney-Ni, even under forcing conditions, instead Pt-C is the catalyst of choice for this class of prochiral ketones.

Examination of pinacolone (1b) with Yb(OAc)3/Pt-C/H2, provided a consistent 93% de, which is 6% greater than the previously best reported result reductive amination result (87% i Ti(O Pr)4/Pt-C/H2), but is the same as compared to a previously reported stepwise approach.[17]

5.2. Conclusion:

115

Reductive amination is a step powerful methodology for the synthesis of α-chiral amines. The correct choice of Lewis acid is critical for efficient suppression of alcohols. The diastereoselectivity obtained with or without the use of Lewis acids was the same. No precedent of increasing diastereoselectivity in reductive amination with the use of achiral

Lewis acid was ever reported. The use of Yb(OAc)3 in reductive amination resulted in a significant increase in diastereoselectivities for different 2-alaknones. Diastereoselectivity of 2-octanone increased 15% compared to the highest reported result. Aromatic and cyclic i ketones were not successfully reductively aminated using Yb(OAc)3. Ti(O Pr)4 proved to be the best Lewis acid in reductive amination of these substrates.

5.3. References:

[1] a) H. Lebel, K. Huard, Org. Lett. 2007, 9, 639; b) M. Kim, J. V. Mulcahy, C. G. Espino, J. Du Bois, Org. Lett. 2005, 7, 4685; c) C. G. Espino, K. W, Fiori, M. Kim, J. Du Boisn, J. Am. Chem. Soc. 2004, 126, 15378. [2] a) J. Blacker, Innovations in Pharmaceutical Technology 2001, 1, 77; b) H. -U. Blaser, F. Spindler, A. Studer, App. Catal. Gen. 2001, 221, 119; c) H. -U. Blaser, M. Eissen, P. F. Fauquex. K. Hungerbuhler, E. Schmidt, G. Sedelmeier, M. Studer, Asymmetric Catalysis on Industrial Scale: Challenges, Approaches, and Solutions; H.-U. Blaser, E. Schmidt Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2004. [3] T. C. Nugent, Chiral Amine Synthesis - Strategies, Examples, and Limitations. In Process Chemistry in the Pharmaceutical Industry, Second Edition: Challenges in an Ever-Changing Climate, T. F. Braish, K. Gadamasetti Eds.; CRC Press-Taylor and Francis Group: New York, 2008. [4] a) L. Storace, L. Anzalone, P. N. Confalone, W. P. Davis, J. M. Fortunak, M. Giangiordano, J. J. Haley, Jr., K. Kamholz, H.-Y. Li, P. Ma, W. A. Nugent, R. L. Parsons, Jr., P. J. Sheeran, C. E. Silverman, R. E. Waltermire, C. C. Wood, Org. Process Res. Dev. 2002, 6, 54; b) E. Juaristi, J. L. León-Romo, A. Reyes, J. Escalante, Tetrahedron: Asymmetry 1999, 10, 2441; c) G. Lauktien, F. -J. Volk, A. W. Frahm, Tetrahedron: Asymmetry 1997, 8, 3457; d) B. Speckenback, P. Bisel, A. W. Frahm, Synthesis 1997, 1325. [5] a) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, Adv. Synth. Catal. 2006, 348, 1289; b) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty,

116

WO2006030017, 2006; c) T. C. Nugent, V. N. Wakchaure, A. K. Ghosh, R. R. Mohanty, Org. lett. 2005, 7, 4967; d) T. C. Nugent, A. K. Ghosh, Eur. J. Org. Chem. 2007, 3863. [6] P. Patnaik, Handbook of Inorganic Chemicals, McGraw-Hill Companies, 2003, PP 973- 975. [7] E. Keller, B. L. Feringa, Tetrahedron Lett. 1996, 37,1882. [8] S. Sinha, B. Mandal, S. Chandrasekaran, Tetrahedron Lett. 2000, 41, 9109. [9] R. J. Lu, D. Liu, R. W. Giese, Tetrahedron Lett. 2000, 41, 2817. [10] A. Janczuk, W. Zhang, W. Xie, S. Lou, J. P. Chengb, P. G. Wang, Tetrahedron Lett. 2002, 43, 4271. [11] I. Nakamura, M. Kamada, Y. Yamamoto, Tetrahedron Lett. 2004, 45, 2903. [12] L.-M. Wang, J. Sheng, L. Zhang, J.-W. Han, Z.-Y. Fan, H. Tiana, C.-T. Qianb, Tetrahedron 2005,61,1539. [13] L.-M. Wang, Y.-H.Wang, H. Tian, Y.-F. Yao, J.-H. Shao, B. Liu, J. Fluorine Chem. 2006, 127, 1570. [14] M. Asadullah, Y.Taniguchi, T. Kitamura, Y. Fujiwara, Appl. Organometal. Chem. 1998, 12, 277. [15] T. Yokko, K. Matsumoto, K. Oshima, K. Utimoto, Chemistry Letters 1993, 571. [16] (a) G. Cainelli, D. Giacomini, A. Trerè, P. P. Boyl, J. Org. Chem. 1996, 61, 5139. (b) J. G. H. Willems, J. G. de Vries, R. J. M. Nolte, B. Zwanenburg, Tetrahedron Lett. 1995, 36, 3917. (c) V. A. Soloshonok, A. G. Kirilenko, S. V. Galushko, V. P. Kukhar, Tetrahedron Lett. 1994, 35, 5063. [17] N. Moss, J. Gauthier, J. –M. Ferland, Synlett 1995, 142.

117

Chapter 6 Catalytic Lewis Acids in Reductive Amination.

6.1. Introduction.

Historically Lewis acids were used in reductive amination in stoichiometric quantities. Through extensive literature search I did not find any previous example for the catalytic use of Lewis acids in reductive amination. This was one of the limitations for scaling up the use of Lewis acids in reductive amination. The high diastereoselectivity obtained with the use of

Yb(OAc)3 in reductive amination motivated us to investigate deeply for the possibility of using Lewis acids in catalytic quantities. Table 6.1 shows the effect of slowly decreasing the mol % of dried Yb(OAc)3 from 110 mol % to 10 mol %. The data shows that 80 mol %

Yb(OAc)3 produces the same results as 110 mol %, and interestingly 50 mol % Yb(OAc)3 is capable of maintaining very similar de (1% less) as compared to 110 mol %. Rapid deterioration in the diastereoselectivity was noticed when the Yb(OAc)3 loading was reduced below 40%. When the Yb(OAC)3 loading was reduced to 10 mol % no improvement in the de was noticed the de’s observed were similar to the previously reported with non-Yb(OAc)3 based methods with α-MBA.[1]

Table 6.1. Relationship Between Mol % of Yb(OAc)3 and Diastereoselectivity of Amine Product.a entry Yb(OAC)3 mol% de% 1 110 87 2 100 87 3 80 87

118

4 60 86 5 50 86 6 40 84 7 20 79 8 20 80b 9 10 72 a Ketone (2.5 mmol, 1 equiv), dried Yb(OAc)3 (1.1 equiv), (S)-α-methylbenzylamine (1.1 equiv), Raney-Ni, H2 (120 psi), MeOH-THF (1:1) 0.50 M, 22 °C, 12 h. b 4 Å molecular sieves (4 wt equiv) were also added.

Further reduction in the Yb(OAc)3 loading below 10 mol % led to increase the formation of alcohol byproduct. When 5 mol % of Yb(OAc)3 was used, alcohol by product was detected in aquantity greater than 2 area % (GC). For this detailed study I reached to the conclusion that the least Yb(OAc)3 loading is 10 mol % which should be used the rest of the study. Encouraged by these results I investigated a variety of transition metal, lanthanide, and metalloid halide, acetates, alkoxides, and sulfonates, for their ability to allow fast and high yielding reductive amination reactions to occur. From this extensive study Ce(OAc)3 (15 mol

%) and Y(OAc)3 (15 mol %) emerged as useful Lewis acids inhibiting alcohol formation. The results obtained through using these two Lewis acids were consistently similar to the results of Yb(OAc)3 (10 mol %), regarding reaction times, yield, and diastereoselectivity for the reductive amination of 2-octanone with (S)-α-MBA. Stoichiometric use of Ce(OAc)3 or

Y(OAc)3, did not provide enhanced de or any other added benefit over those reactions examined at the 15 mol % level.

Other transition metal acetate salts showed interesting results in terms of promoting reductive amination in an acceptable time frame. On the other hand, the use of these Lewis acids resulted in formation of alcohol byproduct in concentration typically 5-15 area % by GC analysis. These Lewis acids are : In(OAc)3, Sc(OAc)3, CuOAc, Er(OAc)3, Gd(OAc)3,

Dy(OAc)3, AgOAc, Zn(OAc)2, and Cd(OAc)2, they were tested at 15 mol % concentration. Other commercially available salts of these elements were tested but none of them proved useful as the alcohol byproduct concentration was always above 25 area % by GC analysis. From this study it is obvious that all interesting Lewis acids have acetate as counter ion.

Exceptions to this general observation were noted for Bi(OTf)3, AgCl, ScCl3, and scandium

119

hexafluoroacetylacetone which provided 5-15 area % of the alcohol by-product and/or observably longer reaction times than Yb(OAc)3 (10 mol %), Y(OAc)3 (15 mol %), or

Ce(OAc)3 (15 mol %).

Thionylchloride which was tested industrially for promoting ketimine formation at 5 mol % concentration was also tested in our study.[2] Contradicting to their findings, the replacement of the Lewis acid by thionylchloride, led to the amine product in 50% de, which is even lower than the normally achieved 72% de for 2-octanone. Phosphorous oxychloride which has not been previously reported for reductive amination was also tested. Its use at a concentration 10 mol % was beneficial in obtaining 72% de for 2-octanone but the alcohol by-product was observed in ~4 area % (GC).

The above listed Lewis acids were available commercially in their semi-hydrated forms, and used without further purification for the catalytic screening studies. According to the findings from the stoichiometric use Yb(OAc)3, drying Yb(OAc)3 was extremely important for consistent results. In the initial catalytic screening studies, metalloids Bi(OAc)3 and

Sb(OAc)3 were included with Yb(OAc)3, Y(OAc)3, and Ce(OAc)3 as useful in inhibiting alcohol formation. During the optimization stage of the catalytic study I recognized that the purchased Bi(OAc)3 and Sb(OAc)3 smelled strongly of AcOH. This raised the question of what was catalyzing the reductive amination, the Lewis acid or co-existing acetic acid. The

Lewis acids [Yb(OAc)3, Y(OAc)3, Ce(OAc)3, Bi(OAc)3, and Sb(OAc)3] were then dried until each maintained a constant weight. Reexamination of these dried salts showed Bi(OAc)3 and

Sb(OAc)3 were no longer efficient catalysts for reductive amination, high alcohol by-product formation (>15 area %, GC) was noted. In stark contrast, Yb(OAc)3, Y(OAc)3, and

Ce(OAc)3, were as effective as before, although their solubility in the binary reaction solvent system, THF-MeOH, was visibly reduced. The effect of adding H2O (1.0 equiv) to the reactions with dried Yb(OAc)3, Y(OAc)3, or Ce(OAc)3, resulted in increased alcohol by- product formation. While this intentional addition of water was clearly not beneficial, the indicated dried Lewis acids were routinely weighed without precaution for atmospheric moisture and had no ill effect on the reaction profile and reaction time, but dry solvents are always used for the reactions.

120

The reactions described above all used 100 wt % Raney-Ni, based on the limiting ketone reagent, and this quantity is not untypical for the use of this heterogeneous hydrogenation catalyst regarding reductive amination and imine reduction.[3]

I tried to study the effect of reducing heterogeneous catalyst loading on the rate of the reaction and the de. I found that using 50 wt % of Raney-Ni in combination with dried

Yb(OAc)3 did not allow complete consumption of the ketone (8 area % remained unreacted) within the standard room temperature and 12 h reaction time. It is known that increasing the temperature increases reaction rate, so I increased the temperature (40 oC) and pressure (20 bar) simultaneously (table 6.2, compare entries 2 and 3), the reaction could be completed before 12 h without compromising the diastereoselectivity of the secondary amine. When 25 wt % of Raney-Ni was used 50% of unreacted ketone was detected after 12 h.

Table 6.2. Raney-Ni loading: Reductive Amination of 2-Octanone for 12 ha Entry Raney-Ni (wt %) Ketone (area %)[b] de[b]

1 100 0 72

2 50 8 72

3[c] 50 0 71

4 25 47 71 a Reaction conditions: 2-octanone (2.5 mmol), Yb(OAc)3 (15 mol %), (S)-α-methylbenzylamine (1.1 o b c o equiv), Raney-Ni, H2 (8.3 bar), 0.5 M, 12 h, 22 C. GC analysis. H2 (20 bar), T= 40 C.

After finalizing the different useful Lewis acids categories, I tested the use of 10 mol % of

Yb(OAc)3 for reductive amination of different ketone substrates. The following table summarizes the results obtained from our study.

a Table 6.3. 10 mol % Yb(OAc)3 Catalyzed Reductive Amination: Substrate Breadth. entry ketone 1 secondary amine 2 yield %b de %c

121

1 1g O 81 98 2g HN Ph

2 1h O 78 98 2h HN Ph

3 1i O 63 94 2i HN Ph

d 4 1b O 78 92 2b HN Ph

O 5 1a 79 92 HN Ph 2a

6 1c O 87 80 2c HN Ph

e 7 1f O 82 79 2f HN Ph

8 1d O 83 72 2d HN Ph

9 1e O 82 71 2e HN Ph

a Ketone (2.5 mmol), dried Yb(OAc)3 (10 mol %), (S)-α-methylbenzylamine (1.1 equiv), Raney Ni, 0.50 M o o (MeOH-THF, 1:1), 12 h. Entries 1-4: T= 50 C, H2 pressure = 290 psi (20 bar); entry 5: T= 50 C, H2 pressure = o b 120 psi (8.3 bar), entries 6-9: T= 22 C, H2 pressure = 120 psi (8.3 bar). Isolated yield of both diastereomers after chromatography. c Determined by GC analysis of the crude product. d Pt-C was used instead of Raney Ni, o e T= 50 C, H2 pressure = 120 psi. Non-Yb(OAc)3 based methods provide 74% de.

Table 6.3 shows the breadth of prochiral ketones that serve as good substrates for our catalytic method. As might be expected, based on the steric considerations, the diastereoselectivity of the reductive amination product increases in a fairly undisturbed and linear progression (72–98%), on changing the R substituent of the 2-alkanone, RC(O)CH3, from a straight chain alkyl group to those having –γ, -β, and finally -α branching (table 6.3,

122

e.g. compare entries 1, 5, 6, and 9). This general trend is interrupted only when the R substituent is t-butyl, e.g. pinacolone (table 6.3, entry 4). Pinacolone again fails to react under the standard Raney-Ni catalyst conditions, but the desired product is produced when using Pt- C as the hydrogenation catalyst. It is interesting to note that unlike our earlier findings with i [3] Raney-Ni/Ti(O Pr)4, which allow 2-alkanones with α-branching (table 6.3, entries 1-4) to o be reductively aminated at 22 C and 120 psi H2 in 12 h, the use of Raney-Ni/10 mol % o Yb(OAc)3 requires the more forcing conditions of 50 C and 290 psi (20 bar) for 12 h reaction times to be accomplished with complete consumption of the starting ketone.

Regarding aryl-alkyl ketones, acetophenone (table 6.3, entry 3) was sluggish to react even at 50 oC and 432 psi (30 bar) of hydrogen, with isolated yields varying between 60-65% and concomitant alcohol by-product formation always noted. Examination of 1-phenylbutanone at 50 oC and 432 psi (30 bar) of hydrogen only allowed ~20 area % (GC) of the expected product to form after 24 h. Benzosuberone (cyclic aryl-alkyl ketone) and i-propyl n-propyl o ketone, under similar forcing conditions (50 C, 580 psi (40 bar) H2, >24 h), showed that these sterically challenging substrates could not be reductively aminated.

6.2. Stoichiometric and Catalytic Brønsted Acid Promoted Reductive Amination

Table 6.4. Brønsted Acid Based Reductive Amination of 2-Octanone with (S)-α-MBA.a Entry Brønsted acid ( 20 2-octanone alcohol formed (area deb mol %) remaining (area %) %)b

1 Acetic acid 1 2 72 2 Trifluoroacetic acid - 22.34 71 3 Trichloroacetic acid 1 2.1 72 4 Formic acid - 2.67 71 5 Oxalic acid 1.65 24 73 6 Thiophenol 51.76 3.47 26 7 Phenol - 29 70

123

a 2-Octanone (2.5 mmol), Brønsted acid (20 mol %), (S)-α-methylbenzylamine (2.75 mmol), Raney Ni, 0.50 o b M (MeOH), 12 h, T= 22 C, H2 pressure = 8.3 bar. Determined by GC analysis at 12 h.

As mentioned previously, the main byproduct of reductive amination is alcohol. The use of optimum Brønsted or Lewis acids inhibit byproduct formation. Brønsted acids were used historically on industrial scale for reductive amination. Despite their importance there is a great lack of detailed study for useful Brønsted acids in primary literatures.

I decided to test the effect of using different Brønsted and mineral acids under different loadings in reductive amination of 2-ocatnone. Acetic acid, trichloroacetic acid, or formic acid at 20 mol % successfully catalyzed reductive amination of 2-octanone with α-MBA. Reducing the loading of AcOH (5 mol %) had detrimental effect of allowing significant alcohol by-product formation (> 5 area %, GC). When the loading of acetic acid, trichloroacetic acid, or formic acid was increased to stoichiometric quantities no improvement of de was noticed compared to the use of stoichiometric quantities of

Yb(OAC)3 (de 87%).

Strong mineral and organic acids were also tested showing different reaction profile compared to other Brønsted and Lewis acids. The use of stoichiometric or catalytic (5 or 10 mol %) quantities of 12 N HCl or 18 M H2SO4, which were diluted in MeOH, p-TsOH, or trifluoroacetic acid, resulted in high alcohol by-product formation (15-30 area %, GC). Despite the lower yields of secondary amine of 2-octanone, the de was always consistent (70- 72%) and no reduction was noted. In all cases the use of weak and strong Brønsted acids for reductive amination of 2-octanone shows them be a minimum of 15% lower than when using

110 mol % Yb(OAc)3.

Solvent screening was also needed for choosing the best solvent for acetic acid promoted reductive amination. Protic solvents as MeOH and EtOH were optimal solvents allowing completing the reaction within 8 h. The use of THF-MeOH (which was optimal for stoichiometric Yb(OAc)3 study) slowed down the reaction (12 h). When THF was used as a sole solvent with 20 mol % AcOH for reductive amination of 2-octanone the reaction rate was extremely slow showing 30-45% of the starting ketone after 24 h. Despite this slow rate reaction, no alcohol was detected after 24 h only starting ketone. The use of THF as sole 124

solvent in the stoichiometric and catalytic Lewis acid promoted reactions resulted in complete reaction within 24 h under identical reaction conditions. This different reactivity profile for different Brønsted and Lewis acids in protic vs aprotic reaction solvent may allow future substrates with acid labile functional groups or restricted solubility to be reductively aminated.

Acetic acid (20 mol %) was used as the Brønsted acid of choice to be tested for the reductive amination. The solvent of choice was dry MeOH and different ketones were reductively o aminated as 2-octanone (83% isolated yield, 72% de, T= 22 C, 120 psi (8.0 bar) H2), i-butyl o methyl ketone (80% isolated yield, 92% de, T= 50 C, 120 psi (8.0 bar) H2), cyclohexyl o methyl ketone (82% isolated yield, 98% de, T= 50 C, 290 psi (20 bar) H2), and o acetophenone (55% isolated yield, 93% de, T= 50 C, 435 psi (30 bar) H2). The reaction rate, isolated yield and de were almost the same as the optimal catalytic Lewis acids [Yb(OAc)3

(10 mol %), Y(OAc)3 (15 mol %), and Ce(OAc)3 (15 mol %)] results, but never showed the enhanced diastereoselectivity possible when using 50-110 mol % Yb(OAc)3.

The use of 20 mol % of AcOH failed to promote reductive amination of sterically hindered ketone substrates benzosuberone, 1-phenylbutanone, or i-propyl n-propyl ketone, Even under forcing conditions of high temperatures (22-50 oC) and hydrogen pressure [120-580 psi (8-40 bar)]. 1-phenylbutanone provided the desired product only in very low yield (20-25 area %, GC) after >24 h of reaction. These results add to the general conclusion that for such sterically congested substrates the only effective system for their reductive amination is the i [3] use of stoichiometric quantities of Ti(O Pr)4.

i Ti(O Pr)4 system has another major advantage as it allows α-branched and β-branched methyl ketones to be reductively aminated at mild condition ( 22 oC and 120 psi within 12 h), on the other hand, the same ketones under acetic acid promoted reductive amination require harsher conditions of elevated temperature (50 oC and 120 psi) for β-branched 2-alkanones or elevated temperature and pressure (50 oC and 290 psi) for α-branched 2-alkanones. Acetic acid provides a mediocre yield for acetophenone (55%) and no improvement even under harsher conditions.

125

6.3. Conclusion:

I developed the use of catalytic quantities of Lewis acids in reductive amination for the first time. The use of catalytic quantities of Yb(OAc)3 or Y(OAc)3 or Ce(OAc)3 proved to be efficient in reducing the alcohol by product formation and to produce the secondary amine in good yield and normal de. No enhancement of the de resulted from the use of catalytic quantities of Lewis acids. Other Lanthanide salts were also successful in suppressing alcohol formation but with lower efficiency. Brønsted acids were used historically in reductive amination but without enough reports on their role in reductive amination. I have conducted extensive study on the use of different Brønsted and mineral acids in reductive amination. 20 mol % of acetic acid and formic acid proved to be efficient in suppressing alcohol formation in reductive amination. The use of mineral acids resulted in more alcohol formation (20-30 area % GC). Reductive amination of sterically congested ketones and also aromatic ketones i are best performed using Ti(O Pr)4 not catalytic amount of Lewis acids nor Brønsted acids.

6.4. References:

[1] For advances in the diastereoselective reduction of (R)- or (S)-α-MBA ketimines, see: (a) Nichols, D. E.; Barfknecht, C. F.; Rusterholz, D. B. J. Med. Chem. 1973, 16, 480. (b) Clifton, J. E.; Collins, I.; Hallett, P.; Hartley, D.; Lunts, L. H. C.; Wicks, P. D. J. Med. Chem. 1982, 25, 670. (c) Eleveld, M. B.; Hogeveen, H.; Schudde, E. P. J. Org. Chem. 1986, 51, 3635- 3642. (d) Bringmann, G.; Geisler, J.-P. Synthesis 1989, 608. (e) Marx, E.; El Bouz, M.; Célérier, J. P.; Lhommet, G. Tetrahedron Lett. 1992, 33, 4307. (f) Moss, N.; Gauthier, J.; Ferland, J.-M. Synlett 1995, 142. (g) Lauktien, G.; Volk, F.-J.; Frahm, A. W. Tetrahedron: Asymmetry 1997, 8, 3457. (h) Bisel, P.; Breitling, E.; Frahm, A. W. Eur. J. Org. Chem 1998, 729. (i) Gutman, A. L.; Etinger, M.; Nisnevich, G.; Polyak, F. Tetrahedron: Asymmetry 1998, 9, 4369. (j) Cimarelli, C.; Palmieri, G. Tetrahedron: Asymmetry 2000, 11, 2555. (k) Storace, L.; Anzalone, L.; Confalone, P. N.; Davis, W. P.; Fortunak, J. M.; Giangiordano, M.; Haley, J. J. Jr.; Kamholz, K.; Li, H.-Y.; Ma, P.; Nugent, W. A.; Parsons, R. L. Jr.; Sheeran, P. J.; Silverman, C. E.; Waltermire, R. E.; Wood, C. C. Org. Process Res. Dev. 2002, 6, 54.

126

[2] Farina, V.; Grozinger, K.; Müller-Bötticher, H.; Roth, G. P. Ontazolast: The Evolution of a Process. In Process Chemistry in the Pharmaceutical Industry; K. G. Gadamasetti, Ed.; Marcel Dekker, Inc.: New York, 1999, pp 107–124. [3] a) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, Adv. Synth. Catal. 2006, 348, 1289; b) T. C. Nugent, Chiral Amine Synthesis - Strategies, Examples, and Limitations. In Process Chemistry in the Pharmaceutical Industry, Second Edition: Challenges in an Ever-Changing Climate, T. F. Braish, K. Gadamasetti Eds.; CRC Press-Taylor and Francis Group: New York, 2008; c) T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, WO2006030017, 2006; d) T. C. Nugent, V. N. Wakchaure, A. K. Ghosh, R. R. Mohanty, Org. lett. 2005, 7, 4967; e) T. C. Nugent, A. K. Ghosh, Eur. J. Org. Chem. 2007, 3863.

127

Chapter 7 Stereochemical Considerations of Proposed Mechanistic Models

7.1. Introduction.

The outstanding diastereoselectivities reported for different amines using stoichiometric quantities of Yb(OAc)3 and the application of catalytic quantities of Lewis acids for first time in reductive amination of prochiral ketones were the driving force for investigating the mechanistic aspects behind this effect. It is known historically that the imine conformation plays the major role in determining the stereochemical outcome of reductive amination. Analyzing transition state structures for trans-and cis imines which controls the facial selectivity during addition of hydrogen to the imine allows the prediction of which diastereomer will be formed in excess (figure 7.1). Diastereomeric excess of the amine product depends on which enantiomer of α-MBA is used, the well known concept of allylic 1,3-strain[1] and on the earlier proposed models.[2]

Although predictions obtained from this model agree with the reaction outcome, other models were also introduced aiming to describe the reason behind the improved diastereoselectivity. It was suggested previously that the reductive amination of an α-ketoester with α-MBA may involve a rotamer (about the nitrogen-benzylic carbon bond) with the phenyl ring of α-MBA coplanar to the imine double bond.[3] This proposed idea agrees with the known fact of π bonds affinity for heterogeneous hydrogenation catalyst surfaces.[4]

Despite this fact, close examination of other two possible trans-ketimine rotamers, having in a coplanar conformation with the imine double bond reveals that one rotamer suffers from high allylic 1,3-strain resulting from steric crowding of phenyl group with methyl group connected to imine carbonyl carbon the other rotamor is less sterically

128

congested which should be favored. Unfortunately this rotamer leads to the formation of the wrong diastereomer. It can be stated that phenyl group is not adsorbed on the heterogeneous surface of the catalyst during hydrogen addition step.

Ph Ph Ph H R H N H NH N CH3 CH3 R R Re-face addition of hydrogen cis-(R)-ketimine (S,R)-2 to the cis-(R)-ketimine H2

H2 Si-face addition of hydrogen Ph to the trans-(R)-ketimine CH3 Ph R N H HN H N CH3 H R R Ph trans-(R)-ketimine (R,R)-2 Figure 7.1. Nitrogen-Benzylic Carbon Bond Rotamers Responsible for Hydrogen Addition to cis- and trans-N-α-MBA Ketimines.

7.1.1. Mechanism Behind Enhanced Stereoselectivity with Yb(OAc)3:

Through consulting literatures regarding reductive amination, it can be stated clearly that there is no precedent for Lewis acid enhanced diastereoselectivity during reductive amination has been reported. Also with respect to the recent literatures reporting the addition of Lewis acid to an N-α-MBA ketimine no enhanced diastereoselectivity was noted.

To be able to understand the origin of Yb(OAc)3 effect on diastereoselectivity, different experiments have been conducted. In these experiments I tried to focus on detecting or isolating the ketimine intermediates. As stated previously the ketimine intermediate conformation is the key factor in determining the stereochemical outcome of the reaction. Also I tried to test the reaction before completion after certain time intervals and compare the i results with results obtained from Ti(O Pr)4 reaction. Reductive amination of 2-octanone with i Ti(O Pr)4 or Yb(OAc)3 requires 12 h for complete consumption of the ketone to occur, examination of the reaction at 1 h and 3 h showed the diastereoselectivity of the product to be

129

fully consistent with that found at the end of the reaction (table 7.1). The consistent diastereoselectivity throughout the whole reaction suggests that one mechanism is operating from the first minute till the end of the reaction. This conclusion is applicable on Yb(OAc)3 i ,Ti(O Pr)3 system and even when no Lewis acid was used. Despite the importance of this conclusion the origin of enhanced diastereoselectivity was not clarified.

Table 7.1. Closer Examination of the Reductive Amination of 2-Octanone.a Amine 2d entry additive time (S)-α- 2-octanone 2-octanol (%) yield (%) de (%) (h) MBA (%) (%)b

c 1 Yb(OAc)3 1 18.1 6.5 0.5 73.9 87.1

3d 12.5 2.5 1.0 81.5 87.2

i 2 Ti(O Pr)4 1 43.9 7.9 0.63 47.6 67.0

3 8.5 4.0 2.1 85.4 67.0

3 none 1 47.1 10.5 16.6 25.8 72.0

3 29.2 1.1 27.6 42.2 72.1

a i 2-octanone (2.5 mmol, 1.0 equiv), (S)-α-MBA (1.1 equiv), Yb(OAc)3 or Ti(O Pr)4 (1.1 equiv) are stirred in MeOH (1.0 M) for 90 min at rt, then THF (final molarity 0.5 M) and Ra-Ni (100 wt %) are added, and the b reaction pressurized with H2 (8.3 bar/120 psi). All data is based on GC area % analysis. Sum of (S)-α-MBA and imine remaining. c 1.0 area % of (±)-2-aminooctane was noted. d 2.5 area % (±)-2-aminooctane was noted.

In the process of searching for the origin of enhanced diastereoselectivity I tried to collect information about the in situ imine formation. Samples were taken from mixture of 2- i octanone and (S)-α-MBA after 30 min with no additive or with Ti(O Pr)4 (1.25 equiv) and analyzed by GC. In both reactions the area % of the imine was almost similar. Comparing these results with results obtained from mixing 2-octanone and (S)-α-MBA after 30 min with

Yb(OAc)3 (1.1 equiv) showed that no appreciable amounts of imine (< 3 area %) was detected when Yb(OAc)3 was used. Extending the reaction time upto 12 h aiming to force the imine formation did not show any success (table 7.2).

Table 7.2. In Situ Imine Formation Study: 2-Octanone and (S)-α-MBA.a

130

imine area % (GC analysis)

i time (min) no additive Ti(O Pr)4 (1.25 equiv) Yb(OAc)3 (1.1 equiv)

30 16 38 <3

90 33 40 <3

a 2-octanone (1.0 equiv), (S)-α-MBA (1.1 equiv), and the indicated additive were added to anhydrous MeOH (1.0 M). All aliquots were worked-up with saturated aq NaHCO3/EtOAc at the indicated times.

Previous reports about reductive amination stated that imine in the key intermediate in this reaction. It is also stated that imine is highly labile to hydrolysis but in general under mild work up conditions it can be detected. Failure in detecting appreciable amounts of imine when Yb(OAc)3 was used suggests that imine is formed but it is extremely labile. Lewis acid-base pair formation with Yb(OAc)3 made the imine more susceptible to hydrolysis upon work-up, even under the mild conditions of aq

NaHCO3/EtOAc. For the sake of confirming this finding, N-α-MBA ketimine of 2- octanone was prepared under standard reaction conditions using Dean-Stark trap. To the isolated imine, Yb(OAc)3 (1.1 equiv) in anhydrous MeOH was added and stirred for 30 min. Sample was taken from the reaction and worked up using (aq

NaHCO3/EtOAc) and analyzed by GC. No appreciable amounts of imine was detected (< 6 area %) only starting amine and ketone. Also our attempts to analyze the ketimine using NMR did not show any success. Mixing ketone, α-MBA, Yb(OAc)3 in 1 CD3OH and measuring H NMR proved unhelpful.

No clear resonance patterns could be detected due the active paramagnetic isotopes of Yb or the slight turbidity of the solution. All these findings strengths our hypothesis that imine formation under Yb(OAc)3 conditions is extremely labile and cannot be isolated nor analyzed. Despite the importance of this piece of information it ruled out the possibility for easily determining the origin of the enhanced de through direct i comparison of the trans/cis imine ratios when no additive or Ti(O Pr)4 is present vs when Yb(OAc)3 is present.

Combining all previous finding, it can proposed that Yb(OAc)3 enhanced diastereoselectivity comes from its effect on changing the proportion of in situ

131

generated trans- and cis-imine mixture. This means that the addition of Yb(OAc)3 may results in enrichment of trans- over cis- imines before hydrogenation. This idea is only applicable if Yb(OAc)3 was capable of isomerizing some of the cis-imine to the trans-imine.

To test this proposed idea practically I synthesized N-(S)-MBA ketimine of 2- octanone using Dean-Stark trap. To the isolated imine, Yb(OAc)3 (1.1 equiv) in anhydrous MeOH was added and stirred for half an hour at 22 °C before the addition of Raney-Ni slurry in THF and the then the whole mixture was hydrogenated at 8.1 bar (120 psi) for 12h. GC analysis after 12 h showed that all imine was consumed and the product de was 86%. Repeating the same reaction but with extended stirring time for the imine with Yb(OAc)3 resulted in similar de (86%). These results correlate with the earlier obtained results of enhanced diastereoselectivity utilizing Yb(OAc)3 in the reductive amination of 2-octanone. Further examination of effect of changing reaction conditions on the reaction out come as changing the solvent during prestirring period was also tested. THF was used instead of MeOH in the prestirring period resulting in lowering the de of the formed amine to 77%. This may be a direct result of low solubility of Yb(OAc)3 in THF allowing the back round reaction to occur and showing the importance of having MeOH as the prestirring solvent. All these findings support the proposed hypothesis of Yb(OAc)3 promoted isomerization of the in situ formed imine.

In our attempt to clarify the role of Yb(OAc)3 in reductive amination of prochiral ketones, I proposed a mechanism for the in situ imine isomerization (scheme 7.1). The reaction starts with the formation of a Lewis acid-base pair between the imine and

Yb(OAc)3 in the cis conformation. In this adduct, acetate ligand of ytterbium attacks the electrophilic carbonyl carbon of iminium ion via a six-membered transition state forming an oxygen-acetylated hemi-aminal in gauche conformation. This conformation suffers from steric crowding because of the gauche relationship between the “α-methylbenzyl” substituent on the nitrogen atom and the “R” substituent of the former carbonyl carbon. To relief this steric strain the molecule undergoes pyrimidal inversion at nitrogen atom. This process is a low energy process which readily occurs at room temperature. 132

Inversion at the nitrogen atom of the gauche conformation results in formation of anti conformation allowing an anti-relationship to exist between the “α-methylbenzyl” substituent of nitrogen and the “R” substituent of the former carbonyl carbon; this conformation also allows an antiperiplanar arrangement between the nitrogen lone pair and the acetate leaving group, allowing facile elimination of acetate and trans- imine formation. Through this proposed mechanism I can understand the role of ytterbium acetate imine isomerization.

OAc OAc Yb O Yb Ph N O N R O R O Ph cis-5

higher energy cis-ketimine pathway

R CH3 R CH3 N pyrimidal inversion N Yb Yb

Ph O O at nitrogen O O Ph gauche-6 anti-6

lower energy trans-ketimine pathway

Ph OAc OAc N Yb Yb O Ph N O R O R O trans-5

Scheme 7.1. Proposed Mechanism for In Situ Isomerization of the Ketimine during Reductive Amination

7.1.2. Reasons Behind Enhanced Diastereoselectivity for Different 133

Substrate Categories.

I found that the biggest jump for de is associated with straight-chain 2-alkanones (e.g. 2-octanone) and γ-branched 2-alkanones (e.g. benzylacetone). These groups of substrates are good substrates for this methodology in terms of high diastereoselectivity. α- and β-branched 2-alkanones did not show any significant increase in the de. This effect can be rationalized through understanding the nature of R group attached to of the two Newman projections in gauche and anti conformations illustrated in (scheme 7.2).

When α- and β-branched 2-alkanones (when Rα or Rβ = alkyl respectively) are used with Yb(OAc)3 , no improvement of diastereoselectivity was noticed as Illustrated from the scheme. This finding implies that the energy difference between the gauche and the anti conformations is not significant. This means that steric crowding of the “α-methylbenzyl” substituent on the nitrogen atom and the “R” substituent of the former carbonyl carbon in the gauche conformation vs the steric crowding experienced when the “ytterbium” substituent of the nitrogen atom is gauche the “R” substituent of the former carbonyl carbon of the anti conformation has almost the same energy. On the other hand, examination of the energy difference of the gauche and anti conformations of γ-branched 2-alkanones and straight chain 2- alkanones shows that the anti congregation is lower in energy compared to the gauche conformation. The effect is more pronounced in straight chain 2-alkanones compared to γ-branched 2-alkanones. This difference in energy can be rationalized by noting that the “α-methylbenzyl” substituent of nitrogen is sterically very crowded α to the nitrogen, while the ytterbium-nitrogen bond will be expected to be longer than the benzylic carbon-nitrogen bond of the “α-methylbenzyl” substituent and thereby reduce the immediate steric volume next to nitrogen.

The conclusion is that regardless of the steric bulk of the “R” substituent, there is always a high degree of steric crowding when the “α-methylbenzyl” substituent on nitrogen is gauche to it (Figure 2, gauche-6). To account for the observed enhancement in de, “R” substituents (Scheme 3) having only γ-branching (Figure 3,

134

anti-6) would be expected to have medium steric crowding with a gauche ytterbium atom, while non-branched “R” substituents would have low steric crowding in relation to a gauche positioned ytterbium atom. These considerations would thus favor cis-imine to trans-imine isomerization for straight-chain 2-alkanones and γ-branched 2-alkanones, but exclude isomerization for α- and β-branched 2-alkanone substrates.

Rγ Rα Rγ Rα CH3 CH R R 3 N N ~ Rβ Yb = RβYb Ph O O O O Ph gauche-6 anti-6 Rα or Rβ = alkyl

Rγ Rγ

CH3 CH3 R R N N Yb > Yb Ph O O O O Ph gauche-6 anti-6 Rα and Rβ = H

CH3 CH R R 3 N N Yb >> Yb Ph O O O O Ph gauche-6 anti-6 Rα, Rβ, and Rγ = H

Scheme 7.2. Newman Projections Showing Interactions of Ytterbium Acetate with R Group.

Because no other lanthanide acetates were identified as capable of providing enhanced de, ytterbium would appear to have unique coordination sphere attributes (Lewis acidity and high coordination number) allowing the proposed in situ isomerization to occur. Furthermore a unique and critical role is indicated for the acetate ligand. The mechanism presented here is more likely than those in which simple ligation of Yb(OAc)3 to the in situ formed trans- and cis-imine mixture allows constructive influence of the rotamer environment, about the benzylic carbon-nitrogen bond of the ketimine, and thereby improved facial selectivity during reduction. This is 135

supported by the fact that other ytterbium salts, e.g. YbCl3 and Yb(OTf)3, were ineffective at providing efficient product formation and could not do so with enhanced diastereoselectivity. Furthermore none of the other Lanthanides or transition metals examined allowed enhanced diastereoselectivity.

Lastly, Yb(OTf)3 could in theory undergo a similar imine isomerization process via its sulfonate oxygen, but as noted earlier only provided the alcohol by-product under the reductive amination conditions noted here, implying it is too Lewis acidic. To further probe this, I preformed the N-α-MBA ketimine of 2-octanone and then added

Yb(OTf)3 (110 mol%) to it. After 30 min Raney Ni (THF slurry) was added followed by the onset of hydrogenation, the desired product was observed in 50% de.

7.1.3. Key Findings for Reductive Amination with α-MBA

These studies, and our earlier ones, complete a body of research regarding the reductive amination of prochiral ketones with (R)- or (S)-α-MBA (1.1 equiv) under the influence of an optimal Lewis acid or Brønsted acid. The findings show that prochiral alkyl-alkyl' and aryl-alkyl ketones can be readily reductively aminated, and by doing so higher yields and much shorter reaction times are achieved compared to the previously practiced two-step strategy via preformed (R)- or (S)-α-methylbenzyl ketimines.[5]

Furthermore, the reducing system of choice is a heterogeneous hydrogenation catalyst in the presence of hydrogen, due to the superior diastereoselectivity afforded with good isolated yield and reaction time vs all other reducing systems examined to date.[2]

Use of the correct acid (Lewis or Brønsted) catalyst is crucial for a successful outcome when reductively aminating a prochiral ketone with α-MBA in the presence of Raney-Ni (generally the most useful heterogeneous catalyst) and hydrogen (120 psi). Failure to have the optimal Lewis acid or Brønsted acid, or no acid at all, results in gross alcohol by-product formation. Adding catalytic quantities of Yb(OAc)3,

Y(OAc)3, Ce(OAc)3, or catalytic or stoichiometric quantities of a weak Brønsted acid, 136

e.g. AcOH, suppresses alcohol by-product formation for 2-alkanones below 2%, providing the desired amine product in good yield. Application of the catalytic Lewis acid or Brønsted acid systems to aryl-alkyl ketones reveals that only acetophenone will react and then only under forcing conditions (50 °C, 30 bar) with low yield (63% and 55 % respectively) of the desired product.

i When stoichiometric quantities of Ti(O Pr)4, a Lewis acid, are used for reductive amination the reactions are complete within the same reaction time and again alcohol by-product formation is suppressed below 2%; but unlike the above mentioned ° systems, which require elevated temperature (50 C) and/or H2 pressure (290 psi) for i α-branched (RLC(O)CH3) and β-branched (RMC(O)CH3) ketones, Ti(O Pr)4 only requires 22 °C and 120 psi for these hindered 2-alkanones. Additionally, aryl-alkyl ketones and more sterically demanding alkyl-alkyl' ketones, e.g. i-propyl n-propyl i ketone, can be reductively aminated in good yield and de when using Ti(O Pr)4.

When comparing the de of the reductive amination products that are common to i Ti(O Pr)4, Brønsted acids (catalytic or stoichiometric, e.g. AcOH), Yb(OAc)3 (10 mol

%), Y(OAc)3 (15 mol %), and Ce(OAc)3 (15 mol %), the de of the amine product is the same. Furthermore, if preformed (R)- or (S)-α-MBA ketimines are reductively aminated the same de is observed as when the above noted Lewis or Brønsted acids catalysts are used for reductive amination of the corresponding ketone. In stark contrast to these stereoselectivity trends, 2-alkanones without branching at the α- or β-carbons, e.g. 2-octanone or benzylacetone, can be reductively aminated with dramatically increased diastereoselectivity when using as little as 50 mol %

Yb(OAc)3, again alcohol by-product formation is suppressed below 2% and good yields are always realized. These combined findings are summarized in table 7.3.

Table 7.3. Useful Substrate Classes, Optimal Acid Catalysts, and Trends for α- MBA Reductive Aminationa

137

ketone class subclass examples de acid catalystb comment

i O RL= i-Pr or c-hexyl 98 Ti(O Pr)4 viable alternative AcOHc R L CH3 i RL= Ph 95 Ti(O Pr)4 other catatalysts - low yield

i O RL= Ph; RS= n-Pr 90 Ti(O Pr)4 other catalysts - no product R L RS i RL= i-Pr; RS= n-Pr, 87 Ti(O Pr)4 other catalysts - no n-Bu product

i O RM= i-Bu 93 Ti(O Pr)4 viable alternative AcOHd R M CH3 RM= -CH2CH2Ph 89 Yb(OAc)3 other catalysts - low de

O RS= n-hexyl 87 Yb(OAc)3 other catalysts - low de R CH S 3 RS= n-butyl 85 Yb(OAc)3 other catalysts - low de

a o Unless otherwise noted, all reactions performed at 22 C and 120 psi H2. The indicated ketone classes and subclasses provide a starting point for assessing near optimal conditions for similar substrates, see reference 1c and this manuscript for specific details. b For optimal yield and de, i c Ti(O Pr)4 is always used in stoichiometric quantities while Yb(OAc)3 can be used in 50-110 mol %. o d The use of 20 mol% AcOH allows very similar results, but only at 50 C and 290 psi (H2). The use of 20 mol% AcOH allows very similar results, but only at elevated temperature (50 oC).

7.2. Conclusion:

Strategies for α-chiral amine synthesis employing preformed imines or enamines are stepwise long and can suffer from lower overall yield because of mediocre imine or enamine yield forming steps, as previously commented on. This problems can be alleviated by using a reductive amination strategy, as outlined here, and avoids the normally stepwise excessive procedures of chiral auxiliary approaches by simultaneously incorporating a nitrogen atom (from the auxiliary) and a new stereogenic center at the carbonyl carbon during step one (reductive amination). A second step, hydrogenolysis, allows the enantioenriched primary amine to be isolated in good to high overall yield.

The ytterbium acetate method expounded on here unequivocally demonstrates the first example of constructive interference, by any additive, during the asymmetric 138

reductive amination of a prochiral ketone or an N-α-MBA ketimine. It also represents the first documented use of ytterbium for reductive amination. The initial mechanistic investigations elaborated on here suggest an imine isomerization pathway promoted by Yb(OAc)3 allowing enhanced diastereoselectivity during reductive amination. A future study will be required to elaborate on these initial mechanistic proposals, would likely require computational analysis and include the investigation of chiral and other achiral Lewis acid derivatives. Investigation of heterogeneous hydrogenation catalysts prepared by different methods or supported on different materials (e.g. carbon nanostructures, alumina, etc.), could provide further beneficial insights. Finally, the general phenomenon of in situ promoted imine isomerization, with Yb(OAc)3, would be expected to have a beneficial impact on the study of imine/enamine chemistry in general, e.g. in conjunction with enantioselective organocatalysis.

7.3. References:

[1] (a) R. W. Hoffmann, Chem Rev. 1989, 89, 1841. (b) K. W. Lee, S. Y. Hwang, C. R. Kim, D. H. Nam, J. H. Chang, S. C. Choi, B. S. Choi, H. -W. Choi, K. K. Lee, B. So, S.W. Cho, H. Shin, Org. Process Res. Dev. 2003, 7, 839. [2] (a) M. B. Eleveld, H. Hogeveen, E.P. Schudde, J. Org. Chem. 1986, 51, 3635; (b) A. L. Gutman, M. Etinger, G. Nisnevich, F. Polyak, Tetrahedron: Asymmetry 1998, 9, 4369. [3] G. Siedlaczek, M. Schwickardi, U. Kolb, B. Bogdanovic, D. G. Blackmond, Catal. Lett. 1998, 55, 67. [4] (a) A. Kraynov, A. Suchopar, L. D'Souza, R. Richards, Physical Chemistry Chemical Physics, 2006, 8, 1321. (b) T. Bürgi, A. Baiker Acc. Chem. Res. 2004, 37, 909. (c) M. Studer, H. –U. Blaser, C. Exner, Adv. Synth. Catal. 2003, 345, 45. [5] For advances in the diastereoselective reduction of (R)- or (S)-α-MBA ketimines, see: (a) D. E. Nichols,C. F. Barfknecht, D. B. Rusterholz, J. Med. Chem. 1973, 16, 480. (b) J. E.Clifton, I. Collins, P. Hallett, D. Hartley, L. H. C. Lunts, P.D. Wicks, J. Med. Chem. 1982, 25, 670. (c) M. B. Eleveld, H. Hogeveen, E. P. Schudde, J. Org. Chem. 1986, 51, 3635-3642. (d) G. Bringmann, J. –P. Geisler, Synthesis 1989, 608. (e) E. Marx, M. El Bouz,J. P. Célérier, G. Lhommet, Tetrahedron Lett. 1992, 33, 4307. (f) N. Moss, J. Gauthier, J. –M. Ferland, Synlett 1995, 142. (g) G. Lauktien, F. – 139

J. Volk, A. W. Frahm, Tetrahedron: Asymmetry 1997, 8, 3457. (h) P. Bisel, E. Breitling, A. W. Frahm, Eur. J. Org. Chem 1998, 729. (i) A. L. Gutman, M. Etinger, G. Nisnevich, F. Polyak, Tetrahedron: Asymmetry 1998, 9, 4369. (j) C. Cimarelli, G. Palmieri, Tetrahedron: Asymmetry 2000, 11, 2555. (k) L. Storace, L. Anzalone, P. N. Confalone, W. P. Davis, J. M. Fortunak, M. Giangiordano, J. J. Jr. Haley, K. Kamholz, H. –Y. Li, P. Ma, W. A. Nugent, R. L. Parsons, Jr.; P.J. Sheeran, C. E. Silverman, R. E. Waltermire, C. C. Wood, Org. Process Res. Dev. 2002, 6, 54.

140

Appendix

Experimental Section

General Remarks

NMR spectra were recorded on a JEOL ECX 400 spectrometer, operating at 400 MHz (1H) and 100 MHz (13C) respectively. Chemical shifts (δ) were reported in parts per million (ppm) 1 downfield from TMS (= 0) or relative to CHCl3 (7.26 ppm) or D2O (4.79 ppm) for H NMR. 13 For C NMR, chemical shifts were reported in the scale relative to CHCl3 (77.0 ppm) or D2O

(CH3OH internal reference, δ= 49.5 ppm) as an internal reference. Multiplicities are abbreviated as: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br, broad. The coupling constants are expressed in Hz. FTIR spectra were obtained on Nicolet Avatar 370 spectrometer. Mass spectra were recorded on a Finnigan MAT 95 (EI) with an ionization potential of 70 eV. Elemental analyses were performed by an external vendor in Lindlar, Germany on an Elementar Vario EL III instrument. For amine products 2, reaction progress and diastereomeric excess measurements were obtained using a Shimadzu GC-2010 instrument with a Rtx-5 amine column (Restec, 30 m x 0.25 mm); Tinj = 300 °C and Tdet = 300 °C, and carrier gas He @ 24 psi were always constant. Program A: 50 °C (1 min), then 14 °C/min to 280 °C (hold 2 min); Program B: 50 °C (1 min), then 14 °C/min to 130 °C (hold 9 min), then 20 °C/min to 280 °C (hold 2 min); Program C: 50 °C (1 min); then 14 °C/min to 280 °C (hold 1 min); Program D: 50 °C (1 min); then 14 °C/min to 280 °C (hold 5 min). For hydrogenolyzed product primary amine 4d the enantiomeric excess of the trifluoroacetamide derivative was determined by gas chromatography using a Shimadzu GC-2010 instrument on a Chiraldex B-DP column (Astec, 30 m x 0.25mm); Tinj = 200 °C, Tdet = 200 °C, and carrier gas He @ 24 psi were constant. Program E: 130 °C (20 min), then 20 °C/min to 180 °C (hold 10 min), split ratio 60:1. Column chromatography was performed using silica gel 60 (0.040- 0.063 mm). Thin-layer chromatography (TLC) was performed using precoated plates of silica gel 60 F254 and visualized under ultraviolet irradiation (254 nm). All reactions were performed under an inert atmosphere (nitrogen). All reagents were obtained from Sigma-Aldrich (except cyclohexyl methyl ketone, obtained from ABCR GmbH &Co) and used without further purification. Before using the commercially purchased

141

Yb(OAc)3 (Aldrich catalog number 544973, 99.999% grade), Y(OAc)3 (Aldrich catalog number, 326046, 99.9% grade), and Ce(OAc)3 (Aldrich catalog number, 529559, 99.999% grade ), each was dried at 80 °C under high vacuum until a constant weight was achieved (12 h). The dried Lewis acids could be stored in dry screw cap glass bottles at room temperature, and these containers could be repeatedly opened to the atmosphere (at least 6 times without detrimental effect) without special precaution or need for a glovebox. In this way constant and repeatable results were always observed. The (S)-α-methylbenzylamine (Aldrich catalog number, 115568) was of 98% chemical purity and 98% ee. The Raney-Nickel (in water) was purchased from Fluka (Catalog number, 83440). Pd(OH)2/C [≤ 50% water, 20 wt % loading (dry basis)] was purchased from Aldrich (catalog number, 212911). Pt/C (1-4% water, 5 wt % loading) was purchased from Aldrich (Catalog number, 205931).

Experimental Section

Synthesis of N-(S)-α-MBA ketimine of 2-octanone p-Toluene sulphonic acid (2 mol %, 80 mg) was added to a double neck 100 mL round bottom flask, 60 mL of toluene was added, 2-octanone (22 mmol, 1.00 equiv, 3.45 mL), and (S)-α-methylbenzylamine (24.2 mmol, 1.10 equiv, 3.08 mL) were added to the flask. The flask was connected to a Dean-Stark trap which was connected to a refluxing condenser. The mixture was refluxed for 24 h at 120 °C. The mixture was allowed to cool and then the toluene was evaporated under vacuum. The residue was dissolved in hexane (~30 mL), the solution was passed through filter paper and into a separatory funnel. Aqueous NaHCO3 (1.0 M, 40 mL) was added and after very brief mixing, the hexane layer was separated, washed with brine, dried over MgSO4 and filtered. The organic layer was concentrated under rotary evaporation then under high vacuum at 40 °C (with stirring) for 24 h (3.7 g, 57% yield). GC analysis showed 96 area % of the imine and 4 area % of the starting ketone and amine. This imine was used for the further experiments.

General procedure for the reduction of the N-(S)-α-MBA ketimine of 2-octanone

The imine (2.0 mmol, 462 mg) was added to a hydrogenation vessel, and then anhydrous

MeOH (2.5 mL) or THF (2.5 mL) was added. Dried Yb(OAc)3 or Yb(OTf)3 was then added, 142

and the mixture was stirred for 30 min. A THF slurry of Raney-Ni (100 wt % based on the imine, pre-triturated with EtOH (×3) and then with anhydrous THF (×3) before addition) (final reaction molarity 0.4 M) was then added and the reaction vessel pressurized at 120 psi

(8.3 bar) of hydrogen. GC samples were worked-up using NaHCO3/ EtOAc.

General procedure: Stoichiometric Yb(OAc)3 (enhanced de)

In a dry reaction vessel Yb(OAc)3 (0.96 g, 2.75 mmol, 1.1 equiv) was added and subsequently evacuated under high vacuum for 5 min before flooding with nitrogen, anhydrous MeOH (2.5 mL, 1.0 M) was then added. To this solution a prochiral ketone 1 (2.5 mmol, 1.0 equiv) and (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv) were added and subsequently stirred at room temperature for 20-30 min. A THF slurry of Raney-Ni (100 wt % based on the ketone, pre-triturated with EtOH (×3) and then with anhydrous THF (×3) before addition) was transferred to the reaction mixture using 2.5 mL of anhydrous THF (final molarity of reaction solution is 0.5 M) and the reaction vessel pressurized at 120 psi (8.3 bar) of hydrogen. After 12 h at 22 oC, <3 area % of the ketone and imine intermediate remained, GC), and the reaction mixture was then diluted with MeOH (10 mL), filter through a bed of celite, and the celite subsequently washed with MeOH (3 × 10 mL). The combined filtrate was then evaporated to dryness (rotary evaporate at ≤25 °C for short periods due to there semi-volatility), and CHCl3 (20 mL) and aqueous NaOH (15 mL, 1.0 M) were added and this mixture stirred for 90 min. After transferring to a separatory funnel, the CHCl3 layer was removed, and the aqueous layer further extracted with CHCl3 (3 x 15 mL). The combined CHCl3 extracts were filtered through a small bed of celite (removes turbidity) and the celite subsequently washed with CHCl3 (2 × 15 mL). The filtrate was then washed with saturated NH4Cl (2 × 20 mL) [removes residual α-MBA], then with brine (1 × 20 mL), dried over MgSO4, filtered, and evaporated to dryness (rotary evaporate at ≤25 °C for short periods) to obtain the crude product (this material is used to determine the de). Purification by silica gel flash chromatography provides the mixture of diastereomers as a colorless viscous liquid (rotary evaporate at ≤25 °C for short periods), treatment with etheral HCl allows hydrochloride salt formation. Note that amine products 2 are considered semi-volatile and converted to HCl salts to enable their high vacuum drying (defined as 0.5 – 2.0 mm Hg) to constant weight (generally 12-24 h) for yield determination.

143

(2S)-4-Methyl-N-((S)-1-phenylethyl)pentan-2-amine (2a)

Reaction details: Yb(OAc)3 (1.1 equiv), 4-methyl-2-pentanone (0.31 mL, 2.5 mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv). Reaction time: 12 h; 94% de. Purification by silica gel flash chromatography (hexanes/EtOAc/NH4OH, 83:15:2) gave the mixture of diastereomers as a colorless viscous liquid, which was then treated with etheral HCl to obtain the hydrochloride salt (0.467 g, 78% yield) after high vacuum drying. GC (program A, see: Experimental section (general remarks)): retention time [min]: major (S,S)- 2a isomer, 10.9; minor (R,S)-2a isomer, 10.6. The NMR data of (S,S)-2a (free base) matches that reported in the literature.[1] 1 Major (S,S)-2a: H NMR (400 MHz, CDCl3): δ 7.36-7.20 (m, 5H), 3.89 (q, J = 6.8 Hz, 1H), 2.60-2.52 (m, 1H), 1.63-1.56 (m, 1H), 1.43-1.36 (m 1H), 1.32 (d, J = 6.8 Hz, 3H), 1.13-1.07 (m, 1H), 0.94 (d, J = 6.4 Hz, 3H), 0.89 (d, J = 6.8 Hz, 3H), 0.76 (d, J = 6.8 Hz, 3H). 13C

NMR (100 MHz, CDCl3): δ 146.5, 128.3, 126.7, 126.5, 55.2, 48.4, 46.6, 25.1, 24.4, 23.6, 22.3, 21.6.

(2S)-4-Phenyl-N-((S)-1-phenylethyl)butan-2-amine (2c)

Reaction details: Yb(OAc)3 (1.1 equiv), 4-phenyl-2-butanone (0.37 mL, 2.5 mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv). Reaction time: 12 h; 89% de. Purification by silica gel flash chromatography (hexanes/EtOAc/NH4OH, 78:20:2) gave the mixture of diastereomers as a colorless viscous liquid, which was then treated with etheral HCl to obtain the hydrochloride salt (0.625 g, 87% yield) after high vacuum drying. GC (program D, see Experimental section (general remarks)): retention time [min]: major (S,S)- 2c isomer, 15.5; minor (R,S)-2c isomer, 15.4. The NMR data of (S,S)-2c (free base) matches that reported in the literature.[1] Major (S,S)-2c: 1H NMR (CDCl3, 400 MHz): δ 7.30-7.14 (m, 10 H), 3.85 (q, J = 6.4 Hz, 1H), 2.69-2.49 (m, 3 H), 1.85-1.83 (m, 1H), 1.63-1.55 (m, 1H), 1.29 (d, J = 6.4 Hz, 3H), 1.02 (d, J = 6.4 Hz, 3H). 13C NMR (CDCl3, 100 MHz): δ 146.3, 142.5, 128.3, 128.2, 126.6, 126.4, 125.6, 54.9, 49.6, 37.9, 31.9, 24.4, 21.2.

(2S)-N-((S)-1-Phenylethyl)octan-2-amine (2d) 144

Reaction details: Yb(OAc)3 (1.1 equiv), 2-octanone (0.39 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv). Reaction time: 12 h; 87% de.

Purification by silica gel flash chromatography (hexanes/EtOAc/NH4OH, 78:20:2) gave the mixture of diastereomers as a colorless viscous liquid, which was then treated with etheral HCl to obtain the hydrochloride salt (0.580 g, 86% yield) after high vacuum drying. GC (program A, see Experimental section (general remarks)): retention time [min]: major (S,S)- 2d isomer, 12.9; minor (R,S)-2d isomer, 13.1. The NMR data of (S,S)-2d (free base) matches that reported in the literature.[1] 1 Major (S,S)-2d: H NMR (400 MHz, CDCl3): δ 7.33-7.20 (m, 5H), 3.88 (q, J = 6.4 Hz, 1H), 2.53-2.46 (m, 1H), 1.34-1.20 (m, 14H), 0.94 (d, J = 6.4 Hz, 3H), 0.88 (t, J = 6.4 Hz, 3H). 13C

NMR (100 MHz, CDCl3): δ 146.5, 128.3, 126.6, 126.5, 55.1, 50.1, 36.4, 31.8, 29.5, 25.7, 24.6, 22.6, 21.3, 14.1.

(2S)-N-((S)-1-Phenylethyl)hexan-2-amine (2e)

Reaction details: Yb(OAc)3 (1.1 equiv), 2-hexanone (0.31 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv). Reaction time: 12 h; 85% de.

Purification by silica gel flash chromatography (hexanes/EtOAc/NH4OH, 88:8:2) gave the mixture of diastereomers as a colorless viscous liquid, which was then treated with etheral HCl to obtain the hydrochloride salt (0.480 g, 80% yield) after high vacuum drying. GC (program A, see Experimental section (general remarks)): retention time [min]: major (S,S)- 2e isomer, 11.3; minor (R,S)-2e isomer, 11.1. The NMR data of (S,S)-2e (free base) matches that reported in the literature.[2] Major (S,S)-2e: 1H NMR (CDCl3, 400 MHz): δ 7.33-7.19 (m, 5H), 3.88 (q, J = 6.5 Hz, 1H), 2.51-2.45 (m, 1H),1.52-1.46 (m, 1H), 1.32 (d, J = 6.5 Hz, 3H), 1.28-1.15 (m, 6H), 0.94 (d, J = 6.34 Hz, 3H), 0.88 (t, J = 6.95 Hz, 3H). 13C NMR (CDCl3, 100 MHz): δ 146.4, 128.3, 126.6, 126.5, 55.1, 50.1, 36.0, 27.9, 24.6, 22.9, 21.3, 14.1

(2S)-N-((S)-1-Phenylethyl)butan-2-amine (2f)

Reaction details: Yb(OAc)3 (1.1 equiv), 2-butanone (0.22 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv). Reaction time: 12 h; 79% de. After 145

stopping the hydrogenation, further MeOH was added, and this heterogeneous solution was filtered to remove the Raney-Ni, and excess ethereal HCl was added. This was concentrated to dryness (rotary evaporation) and then aqueous HCl (2.0 M) and ether were added. The acidic aqueous layer was removed and the Et2O was extracted with further extracted with aqueous HCl (2.0 M, 2 × 15 mL). The aqueous acidic layer was basified with NaOH (4.0 M) to a pH= 12-14 and the free amine extracted with CH2Cl2 (4 x 20 mL). The combined organic extracts were dried over Na2SO4, filtered and to the filtrate ethereal HCl (2.0 M, 4.0 mL) was added. This solution was concentrated on rotary evaporator and after high vacuum drying (≥24 h) afforded the HCl salt (0.42 g, 79% yield) after high vacuum drying. GC (program B, see Experimental section (general remarks)): retention time [min] for the free base: major (S,S)-2f isomer, 13.0; minor (R,S)-2f isomer, 12.7. The NMR data of (S,S)-2f (free base) matches that reported in the literature.[2] Major (S,S)-2f: 1H NMR (CDCl3, 400 MHz,): δ 7.38-7.20 (m, 5H), 3.87 (q, J = 6.4 Hz, 1H), 2.49-2.41 (m, 1H), 1.56-1.49 (m, 1H), 1.34-1.24 (m, 5 H), 0.95 (d, J = 6.4 Hz, 3H), 0.84 (t, J = 7.2 Hz, 3H). 13C NMR (CDCl3, 100 MHz): δ 146.4, 128.3, 126.7, 126.5, 55.0, 51.3, 28.6, 24.7, 20.7, 9.8.

(2S)-3,3-Dimethyl-N-((S)-1-phenylethyl)butan-2-amine (2b) (Pt substrate)

In a reaction vessel dry Yb(OAc)3 (0.96 g, 2.75 mmol, 1.1 equiv) was added and subsequently evacuated under high vacuum for 5 min before flooding with nitrogen,, anhydrous MeOH (2.5 mL, 1.0 M) was then added. To this solution 3,3-Dimethyl-2-butanone (0.31 mL, 2.5mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv) were added. The reaction was then stirred at 50 ºC for 2 h. Pt-C (98.0 mg, 1.0 mol %) [The 98 mg of Pt/C was added in four equal portions, thus 24.5 mg at t= 0 h, t= 2 h, t= 4 h, and finally at t= 6 h] and THF (2.5 mL, final molarity of reaction vessel 0.5 M) was then added and the reaction vessel pressurized at 120 psi (8.3 bar) of hydrogen. The reaction was then stirred at 50 ºC. After 22 h (<11 area % of the ketone remained by GC), the reaction mixture was diluted with MeOH (10 mL), and the general procedure then followed to provide the crude product (92% de). Purification by silica gel flash chromatography provided the mixture of diastereomers as a colorless viscous liquid (rotary evaporator at ≤25 °C for short periods), which was then treated with etheral HCl to obtain the hydrochloride salt (0.395 g, 77% yield) after high vacuum drying. GC (program A, see Experimental section (general remarks)): 146

retention time [min]: major (S,S)-2b isomer, 10.9; minor (R,S)-2b isomer, 10.6. The NMR data of (S,S)-2b (free base) matches that reported in the literature.[2] 1 Major (S,S)-2k: H NMR (400 MHz, CDCl3): δ 7.35-7.20 (m, 5H), 3.77 (q, J = 6.4 Hz, 1H), 2.29 (q, J = 6.4 Hz, 1H), 1.27 (d, J = 6.4 Hz, 3H), 0.89-0.84 (m, 12H). 13C NMR (100 MHz,

CDCl3): δ 147.6, 128.2, 126.7, 126.6, 59.5, 57.0, 34.7, 26.5, 23.7, 16.0.

(S)-2-aminooctane (4d)

The diastereomeric amine mixture (2d) (0.466 g, 2.0 mmol, 86% de) was dissolved in EtOH

(5.0 mL, 0.4 M) and hydrogenolysis was carried out in presence of Pd(OH)2/C (0.196 g, 7.0 mol %) at 8.3 bar (120 psi) of hydrogen pressure at room temperature. After 10 h, the catalyst was filtered through filter paper and which was subsequently washed with EtOH (2 × 10 mL). 2.0 M etheral HCl (4.0 mL) was then added to the filtrate, and this solution was evaporated to dryness to obtain an oil. The oil was triturated with hexane (4 × 10 mL) and the residual hexane evaporated, this was repeated 3-4 times to obtain a white solid. Further drying for 15 h under high vacuum provided a white solid in qualitative purity (0.25 g, 76% yield). The trifluoroacetyl derivative of 4d had an ee of 85% (chiral GC program E, see Experimental section (general remarks) and Supporting Information chromatograms). GC retention time [min]: major (S)-4d trifluoroacetamide isomer, 15.3; minor (R)-4d trifluoroacetamide isomer, 16.5. 1 4d-HCl salt: H NMR (400 MHz, CDCl3): δ 8.32 (br s, 3H), 3.32-3.29 (m, 1H), 1.82-1.56 13 (m, 2H), 1.41-1.28 (m, 11H), 0.87 (t, J = 6.4 Hz, 3H). C NMR (100 MHz, CDCl3): δ 46.9, 40.2, 31.8, 29.4, 26.3, 23.9, 22.6, 14.0. 4d-oxalate salt: The reported literature data for this compound is that of the oxalate salt for which the 1H NMR is reported. I also formed this salt and found the 1H NMR data for this 1 oxalate salt of (S)-4d matched that reported: H NMR (400 MHz, D2O): δ 3.32-3.27 (m, 1H), 1.64-1.47 (m, 2H), 1.28-1.22 (m, 11H), 0.81 (t, J = 6.4 Hz, 3H).[3] I additionally recorded the 13 C NMR (100 MHz, D2O, CH3OH was used as the internal reference, δ= 49.5 ppm): δ 164.9, 48.6, 34.7, 31.5, 28.8, 25.2, 22.5, 18.3, 14.0.[4]

General procedure: Catalytic Yb(OAc)3, Y(OAc)3, or Ce(OAc)3 (normal de)

147

In a reaction vessel the Lewis acid [Yb(OAc)3 10 mol %, Y(OAc)3 (15 mol %) or Ce(OAc)3 (15 mol %)] was added and subsequently evacuated under high vacuum for 5 min before flooding with nitrogen. To the vessel, anhydrous methanol (2.5 mL, 1.0 M), ketone (2.5 mmol, 1.0 equiv), and (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv) were added and stirred for 20-30 min at the temperature at which the hydrogenation was performed at (22 or 50 oC). A THF slurry of Raney-Ni (100 wt % based on the ketone, pre-triturated with EtOH (×3) and then with anhydrous THF (×3) before addition) was transferred to the reaction mixture using 2.5 mL of anhydrous THF (final molarity of reaction solution is 0.5 M). The vessel was then pressurized to the indicated pressure 120-290 psi (8-20 bar) of hydrogen and stirred at room temperature or at 50 °C as indicated. At 12 h (< 3 area % of ketone and intermediate imine by GC) the reaction mixture was worked-up as delineated in the section entitled: “General procedure: Stoichiometric Yb(OAc)3 (enhanced de).”

(2S)-4-methyl-N-((S)-1-phenylethyl)pentan-2-amine (2a)

Reaction details: Yb(OAc)3 (10 mol %), 4-methyl-2-pentanone (0.31 mL, 2.5 mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at 50 °C, and then hydrogenated at 50 oC and 120 psi (8.3 bar). Reaction time: 12 h; 92% de.

Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 87:9:4) gave the mixture of diastereomers as viscous colorless oil, which then treated with etheral HCl to obtain the hydrochloride salt (0.60 g, 79 yield %). GC (program D, see Experimental section (general remarks)) retention time [min]: major (S,S)-2a isomer, 11.8; minor (R,S)-2a isomer, 11.6, matched those reported in the literature.[1]

(2S)-4-phenyl-N-((S)-1-phenylethyl)butan-2-amine (2c)

Reaction details: Yb(OAc)3 (10 mol %), 4-phenyl-2-butanone (0.37 mL, 2.5 mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at room temperature; hydrogen pressure 8.3 bar (120 psi); hydrogenation was performed at room temperature; reaction time: 12 h; 80% de. Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 83:15:2) gave the mixture of diastereomers as viscous colorless oil, which then treated with etheral HCl to obtain the hydrochloride salt (0.63 g, 87 yield %). GC (program D, see Experimental section (general remarks)) retention 148

time [min]: major (S,S)-2c isomer, 16.5 minor (R,S)-2c isomer, 16.4, matches that reported in the literature.[1]

(2S)-N-((S)-1-phenylethyl)octan-2-amine (2d)

Reaction details: Yb(OAc)3 (10 mol %), 2-octanone (0.39 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at room temperature; hydrogen pressure 120 psi (8.3 bar); hydrogenation performed at room temperature. Reaction time: 12 h; 72% de. Purification by silica gel chromatography (Hexane/EtOAc/NH4OH = 58:40:2) gave the mixture of diastereomers as viscous colorless oil, which then treated with etheral HCl to obtain the hydrochloride salt (0.67 g, 83 yield %). GC (program A, see Experimental section (general remarks) retention time [min]: major (S,S)-2d isomer, 10.9; minor (R,S)-2d isomer, 10.8 match those reported in the literature.[1]

(2S)-N-((S)1-phenylethyl)hexan-2-amine (2e)

Reaction details: Yb(OAc)3 (10 mol %), 2-hexanone (0.31 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at room temperature; hydrogen pressure 8.3 bar (120 psi); hydrogenation performed at room temperature. Reaction time: 12 h; 71% de. Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 89.5:5.5:5) gave the mixture of diastereomers as viscous colorless oil, which then treated with etheral HCl to obtain the hydrochloride salt (0.61 g, 82 yield %). GC (program A, see Experimental section (general remarks)) retention time [min]: major (S,S)-2e isomer, 9.7; minor (R,S)-2e isomer, 9.6, match those reported in the literature.[2]

(2S)-N-((S)-1-phenylethyl)butan-2-amine (2f)

Reaction details: Yb(OAc)3 (10 mol %), 2-butanone (0.22 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at room temperature; hydrogen pressure 8.3 bar (120 psi); hydrogenation done at room temperature. Reaction time:

12 h; 79% de. Purification by silica gel chromatography (Hexane/EtOAc/NH4OH = 91:4:5) gave the mixture of diastereomers as viscous colorless oil, which then treated with etheral 149

HCl to obtain the hydrochloride salt (0.54 g, 82 yield %). GC (program A, see Experimental section (general remarks)) retention time [min]: major (S,S)-2f isomer, 8.5; minor (R,S)-2f isomer, 8.4, matches that reported in the literature.[2]

(1S)-N((S)-1-cyclohexylethyl)-1-phenylethanamine (2g)

Reaction details: Yb(OAc)3 (10 mol %), cyclohexyl methyl ketone (0.34 mL, 2.5 mmol, 1.0 equiv), (S)-α-MBA (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at 50 °C; hydrogen pressure 20 bar (290 psi); hydrogenation performed at 50 °C. Reaction time: 12 h; 98% de.

Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 83:15:2) gave the mixture of diastereomers as a viscous colorless oil, which when treated with etheral HCl provided the hydrochloride salt (0.60 g, 81 yield %). GC (program D, see Experimental section (general remarks)) retention time [min]: major (S,S)-2g isomer, 14.9; minor (R,S)-2g isomer, 14.7, matches that reported in the literature.[1]

(2S)-3-Methyl-N- ((S)-1-phenylethyl) butan-2-amine (2h)

Reaction details: Yb(OAc)3 (10 mol %), 3-methyl-2-butanone (0.27 mL, 2.5 mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at 50 °C; hydrogen pressure 20 bar (290 psi); hydrogenation performed at 50 °C. Reaction time: 12 h; 98% de. Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 92.5:3.5:4) gave the mixture of diastereomers as a viscous colorless oil, which then treated with etheral HCl provided the hydrochloride salt (0.54 g, 78 yield %). GC (program D, see Experimental section (general remarks)) retention time [min]: major (S,S)-2h isomer, 11.2; minor (R,S)-2h isomer, 11.1, matches that reported in the literature.[1]

Bis((S)-1-phenylethyl)amine (2i)

Reaction details: Yb(OAc)3 (10 mol %), acetophenone (0.29 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 30 min at 50 °C; hydrogen pressure 20 bar (290 psi); hydrogenation performed at 50 °C. Reaction time: 12 h; 94% de.

Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 74:25:1) gave the mixture of diastereomers as a viscous colorless oil, which then treated with etheral HCl to 150

obtain the hydrochloride salt (0.41 g, 63 yield %). GC (program D, see Experimental section (general remarks)) retention time [min]: major (S,S)-2i isomer, 14.4; minor (R,S)-2i isomer, 14.7, matches that reported in the literature.[2]

(2S)-3,3-dimethyl-N-((S)-1-phenylethyl)butan-2-amine (2b)

Reaction details: Yb(OAc)3 (10 mol %), 3,3-dimethyl-2-butanone (0.31 mL, 2.5mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), pre-stirred 4h at 50 °C; then adding Pt/C (instead of Raney Ni) in four equal portions at t= 0, 6, 12, 20 h (total added Pt equals 1.0 mol %), with a total hydrogenation time of 30 h at 8.3 bar (120 psi) and at 50 o C. 92% de. Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 94.5:1.5:4) gave the mixture of diastereomers as a viscous colorless oil, which when treated with etheral HCl to obtain the hydrochloride salt (0.58 g, 78 yield %). GC (program C, see Experimental section (general remark)) retention time [min]: major (S,S)-2b isomer, 11.8; minor (R,S)-2b isomer, 11.6, matches that reported in the literature.[2]

General Procedure: Brønsted acids (normal de)

The reaction vessel was evacuated under high vacuum for 5 min before flooding with nitrogen. To the vessel, anhydrous methanol (2.5 mL, 1.0 M), acetic acid (20 mol %), ketone (2.5 mmol, 1.0 equiv) (1), and (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv) were added and stirred for 20-30 min at the temperature at which the hydrogenation was performed at (22 or 50 oC). The remaining procedural details should be followed as in the section entitled: “General procedure: Catalytic Yb(OAc)3, Y(OAc)3, or Ce(OAc)3 (normal de).”

(2S)-4-methyl-N-((S)-1-phenylethyl)pentan-2-amine (2a)

Reaction details: 4-Methyl-2-pentanone (0.31 mL, 2.5 mmol, 1.0 equiv), (S)-α- methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), acetic acid (0.028 mL, 20 mol %), pre- stirred 30 min at 50 °C; hydrogen pressure 8.3 bar (120 psi); hydrogenation performed at 50 °C. Reaction time: 12 h; 92% de. Purification by silica gel flash chromatography

(Hexane/EtOAc/NH4OH = 87:9:4) gave the mixture of diastereomers as a viscous colorless 151

oil, which when treated with etheral HCl provided the hydrochloride salt (0.60 g, 80 yield %). GC (program D, see Experimental section (general remarks)) retention time [min]: major (S,S)-2a isomer, 11.8; minor (R,S)-2a isomer, 11.6, matches that reported in the literature.[1]

(2S)-N-((S)-1-phenylethyl)octan-2-amine (2d)

Reaction details: 2-Octanone (0.39 mL, 2.5 mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), acetic acid (0.028 mL, 20 mol %), pre-stirred 30 min at room temperature; hydrogen pressure 8.3 bar (120 psi); hydrogenation performed at room temperature. Reaction time: 12 h; 72% de. Purification by silica gel chromatography

(Hexane/EtOAc/NH4OH = 58:40:2) gave the mixture of diastereomers as viscous colorless oil, which when treated with etheral HCl provided the hydrochloride salt (0.68 g, 83 yield %). GC (program A, see Experimental sectuib (general remarks)) retention time [min]: major (S,S)-2d isomer, 10.9; minor (R,S)-2d isomer, 10.8, matches that reported in the literature.[1]

(1S)-N((S)-1-cyclohexylethyl)-1-phenylethanamine (2g):

Reaction details: Cyclohexyl methyl ketone (0.34 mL, 2.5 mmol, 1.0 equiv), (S)-α-MBA (0.35 mL, 2.75 mmol, 1.1 equiv), acetic acid (0.028 mL, 20 mol%), pre-stirred 30 min at 50 °C ; hydrogen pressure 20 bar (290 psi); hydrogenation done at 50 °C. Reaction time: 12 h;

98% de. Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 83:15:2) gave the mixture of diastereomers as a viscous colorless oil, which when treated with etheral HCl provided the hydrochloride salt (0.62 g, 82 yield %). GC (program D, see Experimental section (general remarks)) retention time [min]: major (S,S)-2g isomer, 14.9; minor (R,S)-2g isomer, 14.7, matches that reported in the literature.[1]

Bis((S)-1-phenylethyl)amine (2i)

Reaction details: Acetophenone (0.29 mL, 2.5 mmol, 1.0 equiv), (S)-α-methylbenzylamine (0.35 mL, 2.75 mmol, 1.1 equiv), acetic acid (0.028 mL, 20 mol %), pre-stirred 30 min at 50 °C; hydrogen pressure 20 bar (290 psi); hydrogenation performed at 50 °C. Reaction time: 12 h; 93% de. Purification by silica gel flash chromatography (Hexane/EtOAc/NH4OH = 74:25:1) gave the mixture of diastereomers as a viscous colorless oil, which when treated 152

with etheral HCl provided the hydrochloride salt (0.37 g, 55 yield %). GC (program D, see Experimental section (general remarks)) retention time [min]: major (S,S)-2i isomer, 14.4; minor (R,S)-2i isomer, 14.7, matches that reported in the literature.[2]

References and Notes [1]T. C. Nugent,V. N. Wakchaure, A. K. Ghosh, R. R. Mohanty, Org. Lett. 2005, 7, 4967. [2] T. C. Nugent, A. K. Ghosh, V. N. Wakchaure, R. R. Mohanty, R. R. Adv. Synth. Catal. 2006, 348, 1289. [3] B. A. Davis, D. A. Durden, Synth. Commun. 2001, 31, 569. [4] H. E. Gottlieb,V. Kotlyar, A. Nudelman, A. J. Org. Chem. 1997, 62, 7512.

153

Mohamed Mahmoud El-Shazly Curriculum Vitae

Date of Birth: 09.April.1977 Mailing Address: Research III, Room 132 School of Engineering and Science Jacobs University, 28759, Bremen, Germany. Nationality: Egyptian E-mail: [email protected], [email protected] Phone: +49 160 557 9099

Academic Qualifications: Degree Month/Year University

B.Sc. Pharmaceutical Science Sep. 1995-Jun. 2000 Ain-Shams University, Cairo, Egypt. (Excellent with Honor, GPA 1.0) Post Graduate Diploma Sep. 2000-Jul. 2004 Ain-Shams University, Cairo, Egypt. (Excellent, GPA 1.33) M.Sc. Nanomolecular science Aug. 2004-Aug. Jacobs University, Bremen, Germany. (Very Good, GPA 1.67) 2006 PhD Organic Chemistry Sep. 2006-Aug. 2009Jacobs University, Bremen, Germany.

Academic awards/Honors:

ƒ Teaching and Research Assistantship, Department of Natural Product Chemistry, Faculty of Pharmacy, Ain Shams University, Cairo, Egypt. ƒ Best Teaching Award, Ain Shams University, Cairo, Egypt. ƒ Best Presentation Skills, Ain Shams workshop for presentation skills. ƒ Graduate Student Fellowship, Jacobs University for both MSc and PhD, Bremen, Germany.

Membership:

ƒ American Chemical Association (ACS). ƒ The Egyptian Federation of Red Cross and Red Crescent Societies. ƒ International Pharmaceutical Federation (FIP). ƒ Egyptian Pharmacists Syndicate (EPS).

154

Research experience: Date Project

Phytochemical and Pharmacological Investigation of Natural Products Isolated from Stipagrostis scoparia Family Graminea.* Sep. 2000- Jul. 2004 Project description: Isolation of biologically active fractions with antihypertensive effect (dieresis and vasodilating effect) from Stipagrostis scoparia for the first time. Study the Application of Transfer Hydrogenation in Reductive Amination.

Dec. 2004- Project description: Using chiral auxiliary (α-Methyl Benzyl Amine) as chiral Jun. 2005 nitrogen source with different hydrogen donors ( isopropanol, formic acid,..etc) and ruthenium or rhodium catalysts for reductive amination of different ketones. Conversion < 50% with 40% diastereoselectivity. Study the Effect of Additives on Asymmetric Reductive Amination.

Jun. 2005- Sep. 2005 Project description: Testing the effect of different acidic and basic additives on the reductive amination of ketones. In general acidic compounds improved diastereoselectivity and vice versa for basic compounds. Synthesis of Different Carbenes and Testing their Application.

Sep. 2005- Jan. 2006 Project description: Synthesizing different chiral amines and utilizing them as building blocks for chiral carbenes. Testing carbenes in enantioselective epoxide ring opening reactions. Study the Effect of Different Lewis Acids on Asymmetric Reductive Amination.

Jan. 2006- Sep. 2006 Project description: Testing the effect of different available Lewis acids on diastereoselectivity of reductive amination. The use of Yb(OAc)3 resulted in great enhancement of diastereoselectivity of 2-alaknones. Developing New Chiral Modifiers for Pt or Pd Metal Surface.

Oct. 2006- Feb. 2007 Project description: Developing chiral cinchonidine analogues (chiral urea and thiourea derivatives) and testing their applications as modifiers for heterogeneous catalysts in asymmetric ketone reduction reactions. Synthesis of New Thiourea Organocatalysts and Testing their Application. Feb. 2007- Aug. 2007 Project description: Preparing different chiral thiourea derivatives and testing their applications in meso desymmetrization through using different acylating agents. Synthesis of New Formamide Organocatalysts and Testing their Application. Aug. 2007- Jun. 2008 Project description: Synthesizing different chiral formamide and testing their applications in the asymmetric allylation of aldehydes with allyltrichlorosilane. Synthesis of Novel Catalysts for Enantioselective Reductive Amination. Jun. 2008- Nov. 2008 Project description: Developing new air stable iridium chiral ligands and testing their applications in one pot enantioselective reductive amination without glovebox. Synthesis of Novel Chiral Amines on Multi-gram Scale.

Nov. 2008- Feb. 2009 Project description: Preparing different α-chiral primary amines with >98% chemical purity and >98% enantioselectivity on multi gram scale (50 gram) utilizing large scale crystallization eliminating the need of chromatographic purification. * This project was carried out at Ain-Shams University Cairo Egypt. The rest of projects were carried out at 155

Jacobs University, Bremen, Germany.

Conferences and Oral Presentation:

ƒ Nugent, T. C.; El-Shazly, M.; Wakchaure, V. N. ‘Ytterbium acetate promoted asymmetric reductive amination: Significantly enhanced stereoselectivity’, Abstracts of Papers, 235th ACS National Meeting, New Orleans, LA, United States, April 6-10, 2008.

Internship:

ƒ El Fatooh Pharmaceutical Corporation, Cairo, Egypt (Marketing and Product Management section, Jun-Sep. 1996).

ƒ Memphis Pharmaceutical Company, Cairo, Egypt (Sterile Products Section Jun.-Sep. 1997). ƒ Faculty of Pharmacy, Sofia, Bulgaria (Natural Product Department Jul.-Aug. 1998). ƒ Drug Analysis and Development Unit (DAAU), Ain-Shams University Cairo, Egypt. (Herbal Drugs Quality Control Section Jun.-Sep. 1999).

Computer Skills:

ƒ Literature search software MDL Cross Fire, SciFinder and ISI Web of Knowledge. ƒ ISIS draw, Chem-sketch, and ChemDraw. ƒ Windows, Linux, Mac OS operative systems, Microsoft Office, Adobe Photoshop, Adobe Illustrator, LATEx.

Languages: ƒ English: Excellently written and spoken ƒ German: Working knowledge written and spoken ƒ Arabic: Mother tongue, excellently written and spoken

156