Application of proazaphosphatranes in stoichiometric reactions
The synthesis of proazaphosphatranes follows such a general procedure4: the reaction of tris(2-ethylamino)amine (‘tren’) with aldehydes to give corresponding alkylated tetramine, which is then converted into protonated type azaphosphatrane 2. Subsequently, 2 is deprotonated to produce a proazaphosphatrane of type 1 by strong ionic base, normally potassium t-butoxide (Figure 1.2).
Proazaphosphatrane 1b proves to be an efficient promoter for acylation of hindered alcohols giving desired products in high yields,5 and only a very slight excess of the acid anhydride is required compared to other commonly used DMAP and tributylphosphine.5 Moreover, 1b could be easily recovered through deprotonation by t-BuOK. The excellent performance can be attributed to efficient delocalization of positive charge among the nitrogens in 1b, making deprotonation of alcohol more easily by its anion, and further the acyl part becomes more attackable by RO- (Figure 1.3).
Selective epoxidation is an important and challenging transformation in synthetic chemistry 6 especially when substrates bear sensitive functional groups. Proazaphosphatrane 1b acts as a highly selective reagent which gives rises to epoxides with trans/cis ratio up to 99:1 when compared to its acyclic analogue triaminophosphine P(NMe2)3 (trans/cis ratios: 72/28-51/49) (Figure 1.4).7 Besides, in the same conditions, the reactions in the presence of 1b are faster (95% conversion) which could be attributed to the fact that 1b is more nucleophilic than P(NMe2)3 (< 5% conversion).
Oxazoles are important intermediates for preparing pharmaceutically valuable α-C-acyl amino acids, which are useful precursors in the preparation of β-hydroxy amino acids, like β-aryl serine, amino alcohols and sympathomimetic agents such as ephedrine and epinephrine.8 However, long reaction times and a large excess of TEA or DBU are typically required. It is reported that in the presence of 1 equiv. of proazaphosphatrane 1b (Figure 1.5), reactions of isocyanoacetates with acyl chlorides give oxazoles in nearly quantitative yields, which could further afford α-C-acyl amino acids through acid hydrolysis in high yields. 9 The resulted azaphosphatrane 2b could be easily separated from reaction mixtures in high yield for regenerating 1b by deprotonation with t-BuOK.
With a stoichiometric amount of proazaphosphatrane 1b, direct synthesis of E-α,β-unsaturated esters in excellent selectivity can be achieved by reacting ethyl acetate or methyl propionate with a variety of aromatic aldehydes (Figure 1.6). 10 This methodology is advantageous over most commonly used methods for preparing E-α,β-unsaturated esters such as the Wittig and the Wittig – Horner – Wadsworth, Peterson and Julia-Lythgoe olefinations, and the Perkin reaction, which present the following limitations: low selectivity for the E isomers, high temperatures and/or toxic reagents, low yields and requirements for preparing necessary intermediates.
Application of proazaphosphatranes as organocatalysts
Silylation of alcohols is one of most commonly used methods for protecting alcoholic OH groups in synthetic chemistry. t-Butyldimethylsilyl chloride (TBDMSCl) and t-butyldiphenylsilyl chloride (TBDPSCl) are the most popular silylation agents. 15 However, under conventional conditions such as TBDMSCl/ imidazole, 18-crown-6, Et3N/TMG, Et3N/DBU, DBU/potassium carbonate, i-Pr2NEt, Et3N/DMAP, the reactions of sterically hindered alcohols and phenols with TBDMSCl has been difficult to accomplish.16 Nonionic superbase 1b was reported to be a very effective catalyst for the preparation of TBDMS ethers of tertiary alcohols and hindered phenols in good yields (80-99%) under a nitrogen atmosphere in CH3CN or in DMF (Figure 1.8).
Interestingly, an excellent selectivity of this methodology in protecting secondary alcohols in the presence of primary alcohols has been observed, which prefers to silylate secondary position rather than the primary one (Figure 1.9).
Transesterification is an important organic transformation, because an ester can be transformed into a more valuable one though this procedure.18 Traditionally, strong acids and ionic bases are used as catalysts for this purpose.19 Proazaphosphatrane 1b is found to effectively catalyze the transesterification of esters with high selectivity and in excellent yields at room temperature in the presence of a catalytic amount of proazaphosphatrane 1b (Figure 1.10). 20 This methodology has a large substrates tolerance ranging from aromatic esters to amino acid esters which are very useful protecting groups in peptide synthesis.
Allylation of aromatic aldehydes by the addition of allylsilanes to aldehydes is an important transformation in organic synthesis because of its importance in the formation of C-C bonds.21 Extensive studies have been performed based on Lewis acid mediated reactions. Lewis base proazaphosphatrane 1c was found to be effective to catalyze the reaction of allylsilanes with aldehydes at room temperature or 40 °C with 20 mol % catalyst loading (Figure 1.11). 22 Moderate yields were obtained, which suggested that an allylic anion is generated during the reaction, then reacts with the aldehyde to form the silyl ether product.
Species having acidic α-hydrogen atoms such as nitriles can be deprotonated by proazaphosphatranes to generate an equilibrium mixture of both the corresponding protonated base and the free base.
Proazaphosphatrane 1b was found to be a catalytic nonionic base for direct synthesis of α,β-unsaturated nitriles by the condensation of benzyl cyanide with a variety of functionalized aromatic and aliphatic aldehydes in excellent yields (Figure 1.12).23 However, this methodology does not work for primary and secondary aliphatic aldehydes and ketones to produce the corresponding α,β-unsaturated nitriles with benzyl cyanide.23 Proazaphosphatrane 1b has advantages over traditional methods utilizing an alkali metal alkoxide or some other ionic base as a catalyst which lead to undesired side reactions such as self-condensation of the nitrile, aldol condensation of the carbonyl compound.
β-Nitroalkanols are useful and significant building blocks in organic transformations, which act as the precursors of nitroalkenes, 2-amino alcohols, and α-nitro ketones. Conventional methods for preparing β-nitroalkanols comprise condensation of carbonyl compounds with a nitroalkane in the presence of an ionic base, which brings about a lot of disadvantages. For example, it suffers a base-catalyzed elimination of water, and often the work-up is tedious. The strong nonionic bases proazaphosphatranes 1b and 1c have been shown to be a superior promoter for nitroaldol reaction, giving rise to the corresponding β-nitroalkanols at room temperature in the presence of MgSO4 in generally excellent yields.25 A large variety of substrates are tolerated ranging from different substituted aldehydes and ketones (Figure 1.13).
These are just several selected applications of proazaphosphatranes as organocatalysts in various organic transformations. Other examples such as deprotection of alcohols, synthesis of 3-substituted coumarins, oxa-Michael addition of alcohols have also been promoted successfully in the presence of proazaphosphatranes.
Application of proazaphosphatranes as ligands in metallic catalysis
α-Aryl-substituted nitriles are important intermediates for synthesizing carboxylic acids, primary amines, aldehydes, esters, and biologically active compounds as well. 26 However, direct synthesis of α-aryl nitriles by arylation of nitriles has proved to be very difficult, not to mention much less reactive aryl chlorides.27 An efficient catalyst system combining the commercially available proazaphosphatrane 1d as a ligand with Pd(OAc)2 in the presence of a base NaN(SiMe3)2 leads to successful coupling of a variety of nitriles with a broad range of aryl chlorides including electron-rich, electron-poor, electron-neutral, and sterically hindered examples in generally high yields (Figure 1.14).
Besides, another catalytic system 1d/Pd2(dba)3 (dba = dibenzylideneacetone) was found to be an efficient and general method for amination of aryl chlorides.29 Different aryl chlorides substrates ranging from electron-poor to electron-rich ones can be coupled to a variety of amines. Good to excellent yields were observed (Figure 1.15). The authors attributed the effectiveness of ligand 1d to the following facts that: (a) the possible transannulation of the bridgehead nitrogen lone pair to the phosphorus enriches the basicity of these proazaphosphatranes, and also the stability of reaction intermediates formed with them, thus enhancing the rate of oxidative addition step, and
(b) steric hindrance stemming from the iso-butyl groups promotes the reductive elimination step.
The Suzuki-Miyaura cross-coupling of halides with arylboronic acids to form C-C bonds is an important reaction in organic transformation for preparation of biaryls. Proazaphosphatrane 1d has proved to be an excellent ligand for this reaction using aryl bromides and chlorides as substrates (Figure 1.16).30 Compared to acyclic analogue triaminophosphine P(NMe2)3, the catalytic system 1d/Pd(OAc)2 is much more efficient and has a large tolerance for different substrates. For example, in the presence of 1d/Pd(OAc)2, the reaction of 2-chlorotoluene with phenylboronic acid gave the desired product in 92% yield, in contrast, no detectable product was observed with P(NMe2)3/Pd(OAc)2 even after 36h.
In the past two decades, proazaphosphatranes have been broadly explored in organic synthesis methodology as stoichiometric bases and as catalysts. However, it is still worth continuing to discover their new applications, tuning their activities by modifying substitution patterns, especially on the nitrogens adjacent to phosphorus. And also, their derivatives such as protonated azaphosphatranes analogues, oxidized or chalcogenated proazaphosphatranes may bring more novel catalytic applications. Besides, chiral proazaphosphatranes may yet prove fruitful in asymmetric synthesis, which will arouse considerable interest.
The complexity and remarkable tasks achieved by biological systems arouse an increasing interest. The non-covalent interactions and the resulting pre-organization account for the efficiency and selectivity of these systems. For instance, the folding of the protein chain in enzymes induce the formation of a well-defined cavity around the reactive center that can impose specific orientation and conformation of the incoming substrate, hence high catalytic activity and selectivity can be reached. Chemists have thus designed nano-reactors presenting a molecular cavity surrounding both the active site and the substrate, in order to mimic such efficient systems. 31 Nevertheless, endohedral functionalization is hard to achieve and has been rarely reported. Furthermore, such supramolecular catalysts often suffer from product inhibition: low turnover numbers are obtained when the product exhibits a high affinity for the cavity and remains in the confined space of the molecular cage, preventing any catalytic cycle.32 Molecular receptors presenting a cavity just above a catalytic center, can fall into two main classes: covalent or self-assembled cages. These latter are obtained from smaller subcomponents, allowing the access to sophisticated structures in only few steps of synthesis. Following the pioneering work of M. Fujita, other remarkable examples of self-assembled cages have been reported by this group and those of K. N. Raymond, J. Rebeck, Jr., P. Ballester, J. N. Reek, P. J. Stang, M. Hardie and J. N. Nitschke, to only cite a few.33 Here, only covalent cages presenting both endohedral functionalization of their inner cavity and activity as metal- or organocatalysts will be described in detail. The synthesis of such architectures and their use as catalysts are highly challenging and as a consequence only few examples have been reported to date.
ions encapsulated in calix-pyrrole based cages, hemicryptophanes or cyclodextrines, respectively. However, the catalytic activity associated with these promising structures were not investigated, hence these systems will not be described herein.34,35,36 Moreover, we will only focus on examples where the catalytic activity of the cage complex has been compared either with that of a model catalyst, which lacks cavity, or with another cage catalyst built from a ligand presenting similar stereoelectronic properties in the vicinity of the metal center. Covalent molecular structures encaging a metallic active site will be firstly presented, then organocatalysts confined in a molecular cavity will be described. Direct comparisons with the model catalysts, without cavity, will allow emphasizing the gains in activity and selectivity induced by the confinement of the catalytic center.
Supramolecular systems combining a well-defined cavity with a metallic center are mainly based on resorcinarenes, cylclodextrins (CD), or calixarenes scaffolds. In most of these structures, the metal ion is located at the rim of the molecular cavity, and their remarkable binding properties allow for an increase of the concentration of the guest substrate near the active site, and good catalytic activities can be observed. However, true endohedral functionalization of a host molecule, i.e. a metal trapped in the heart of the cavity, is hard to achieve. Moreover, once such challenging structures are obtained, no catalytic activity is usually observed because of ligand degradations, under the reaction conditions, or product inhibitions. 37 As a consequence, very few true endohedral functionalized covalent cages and their applications as catalysts have been reported.
A FeII porphyrin complex 45 sandwiched by two cyclodextrins was reported in 1990, by Kuroda et al. (Figure 2.1). 38 This compound mimics the catalytic activity of cytochrome P-450 and acts as an efficient catalyst in the epoxidation of cyclohexene, whereas its model parent, which lacks cavity, displays no catalytic activity for this reaction (55% and less than 2% yields for 45 and 46, respectively).
Table of contents :
Part I: Introduction
Chapter 1. Proazaphosphatranes: discovery, synthesis and applications
1.2 Application of proazaphosphatranes in stoichiometric reactions
1.3 Application of proazaphosphatranes as organocatalysts
1.4 Application of proazaphosphatranes as ligands in metallic catalysis
Chapter 2. Covalent cages with inwardly directed reactive centers as confined organo- and metal catalysts
2.2 Confined metal-catalysts
2.3 Confined organocatalysts
Chapter 3. Halogen bonding: an emergent non-covalent interaction
3.2 Halogen bonding in anion recognition
3.3 Halogen bonding in organocatalysis
Part II: New applications of proazaphosphatranes in catalysis
Chapter 4. Verkade’s superbase as an organocatalyst for Strecker Reaction
4.2 Results and discussion
4.2.1 Initial experiments
4.2.2 Optimization of reaction conditions
4.2.3 Scope of substrates
4.2.4 Mechanism study
Chapter 5. Endohedral functionalized cage as a tool to create Frustrated Lewis Pairs
5.2 Results and discussion
5.2.1 Initial experiments
5.2.2 Scope of substrates
5.2.3 Mechanism study
5.2.4 MBH reaction in the presence of enantiopure M-74/P-74 and TiCl4
5.2.5 Synthesis of enantiopure encaged Verkade’s superbase
5.2.6 Attempts to obtain enantiomeric excess for reactions in the presence of M-74/P-74
5.2.7 Synthesis of enantiopure BINOL-based hemicryptophanes
5.2.8 Assignment of the absolute configuration
Part III: Beyond proazaphosphatranes: haloazaphosphatrane – from organocatalysis to halogen bonding
Chapter 6 Halogenated azaphosphatrane: a new member of halogen-bond donor
6.2 Results and discussion
6.2.1 Application of azaphosphatranes as hydrogen-bonding organocatalysts
6.2.2 Recognition properties
6.2.3 Halogen bonding in a cage
6.2.4 From halogen bonding to chalcogen bonding