Titanium-Catalyzed Multicomponent Couplings: Efficient One-Pot

Article pubs.acs.org/accounts

Titanium-Catalyzed Multicomponent Couplings: Efficient One-Pot Syntheses of Nitrogen Heterocycles Aaron L. Odom* and Tanner J. McDaniel Michigan State University, Department of Chemistry, 578 South Shaw Lane, East Lansing, Michigan 48824, United States S Supporting Information *

CONSPECTUS: Nitrogen-based heterocycles are important frameworks for pharmaceuticals, natural products, organic dyes for solar cells, and many other applications. Catalysis for the formation of heterocyclic scaffolds, like many C−C and C−N bond-forming reactions, has focused on the use of rare, late transition metals like palladium and gold. Our group is interested in the use of Earth-abundant catalysts based on titanium to generate heterocycles using multicomponent coupling strategies, often in one-pot reactions. To be of maximal utility, the catalysts need to be easily prepared from inexpensive reagents, and that has been one guiding principle in the research. For this purpose, a series of easily prepared pyrrole-based ligands has been developed. Titanium imido complexes are known to catalyze the hydroamination of alkynes, and this reaction has been used to advantage in the production of α,β-unsaturated imines from 1,3-enynes and pyrroles from 1,4-diynes. Likewise, catalyst design can be used to find complexes applicable to hydrohydrazination, coupling of a hydrazine and alkyne, which is a method for the production of hydrazones. Many of the hydrazones synthesized are converted to indoles through Fischer cyclization by addition of a Lewis acid. However, more complex products are available in a single catalytic cycle through coupling of isonitriles, primary amines, and alkynes to give tautomers of 1,3-diimines, iminoamination (IA). The products of IA are useful intermediates for the one-pot synthesis of pyrazoles, pyrimidines, isoxazoles, quinolines, and 2-amino-3-cyanopyridines. The regioselectivity of the reactions is elucidated in some detail for some of these heterocycles. The 2-amino-3-cyanopyridines are synthesized through isolable intermediates, 1,2-dihydro-2-iminopyridines, which undergo Dimroth rearrangement driven by aromatization of the pyridine ring; the proposed mechanism of the reaction is discussed. The IA-based heterocyclic syntheses can be accomplished start to finish (catalyst generation to heterocyclic synthesis) in a single vessel. The catalyst can be formed in situ from commercially available Ti(NMe2)4 and the protonated form of the ligand. Then, the primary amine, alkyne, and isonitrile are added to the flask, and the IA product is synthesized. The volatiles are removed (if necessary), and the next reagent is added. A brief video showing the process for the simple heterocycle 4-phenylpyrazole from phenylacetylene, cyclohexylamine, tert-butylisonitrile, and hydrazine hydrate is included. Further development in this field will unlock new, efficient reactions for the production of carbon−carbon and carbon−nitrogen bonds. As an example of such a process recently discovered, a catalyst for the regioselective production of pyrazoles in a single step from terminal alkynes, hydrazines, and cyclohexylisonitrile is discussed. Using titanium catalysis, many heterocyclic cores can be accessed easily and efficiently. Further, the early metal chemistry described is often orthogonal to late metal-based reactions, which use substrates like aryl halides, silyl groups, boryl groups, and so forth. As a result, earth-abundant and nontoxic titanium can fulfill important roles in the synthesis of useful classes of compounds like heterocycles.

1. INTRODUCTION

by using titanium (the second most abundant transition metal after iron) to catalyze multicomponent coupling reactions. Many of the most prevalent reactions in homogeneous catalysis (e.g., Suzuki coupling, Heck coupling, asymmetric hydrogenation, CH-borylation, olefin metathesis, asymmetric dihydroxylation, etc.) involve metals that have enormous utility (Pd, Rh, Ir, Ru, Os, etc.) but are exiguous in the Earth’s crust. Although abundant metals like titanium are involved in

Two prominent trends in synthetic chemistry are (1) catalysis with Earth-abundant metals and (2) development of synthetic methods based on multicomponent reactions. With Earth abundance (Figure 1) generally comes catalysts that are nontoxic and inexpensive. The emphasis on multicomponent coupling chemistry brings molecular complexity from simple starting materials in fewer steps. As these trends develop sophistication, the chemistry allows ready access to complex products for a host of applications with lower environmental impact. Here, we describe chemistry that merges these trends © XXXX American Chemical Society

Received: June 2, 2015

A

DOI: 10.1021/acs.accounts.5b00280 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 1. Abundance of the transition elements in crustal rocks.1

Scheme 1. Reported One-Pot Syntheses of Heterocycles Based on Titanium-Catalyzed 3-Component Coupling (Iminoamination) Chemistry

important specialty chemical catalyses (e.g., Sharpless asymmetric epoxidation) and some of the largest scale industrial processes (e.g., olefin polymerization) relative to their less abundant counterparts, there is a current lack of diversity in catalytic chemistry involving these plentiful elements. It is our opinion that this relative lack of diversity is largely due to the

need for further development rather than an inherent lack of utility. In the studies briefly recounted here, the impetus has been the development of titanium-based catalyses for the production of a large variety of heterocycles. Nitrogen heterocycles are used in a large array of applications from pharmaceuticals to agrochemicals to solar cell dyes. Here, we will describe the use B


Article

Accounts of Chemical Research of titanium-catalyzed multicomponent coupling reactions to generate nitrogen heterocycles in one-pot reactions (Scheme 1) along with some related processes. Syntheses for substituted 2aminopyridines, isoxazoles, quinolines, pyrimidines, pyrazoles, and others have been developed thus far, all using titaniumcatalyzed one-pot or one-step procedures.2,3

Chart 1. Some of the Basic Frameworks for Pyrrole-Based Ligands Shown as Their Products from a Reaction with Ti(NMe2)413,14,16−19

2. INTRODUCTION TO TITANIUM IMIDO AND HYDRAZIDO(2−) CATALYSIS Perhaps the earliest example (eq 1) of titanium-catalysis involving an imido complex is found in a one sentence note by

H2NR1, to give the active catalyst, a titanium imido R1N TiX2. Titanium imidos, similar to transition metal alkylidenes in olefin metathesis, undergo [2 + 2]-cycloaddition reactions with unsaturated carbon−carbon bonds like alkynes in a reversible reaction. The slow step in the catalysis is believed to be the protonolysis of the Ti−C bond. The enamine product could then be released by proton migration from a titanium amido to reform the titanium imido.6,7 In 2001, we published that the commercially available reagent Ti(NMe2)4 was an effective and rapid catalyst for the HA of alkynes with primary amines in some cases. For example, 1-hexyne can be hydroaminated with aniline in ∼2 h in 90% yield to give a 3:1 mixture of regioisomers (eq 3). Adding steric

Scheme 2. Bergman Mechanism for Hydroamination (HA) of an Alkyne by a Primary Amine for Titanium

bulk to the amine or the alkyne slows the reaction significantly with yields using tert-butylamine being generally negligible with most alkynes. Even so, this simple precatalyst worked well for many substrates, especially aniline derivatives, with regioselectivities as high as 40:1.8 To improve the catalysis, our ancillary ligands of choice were based on pyrrole. Behind this choice were several assumptions and observations, which to our good fortune have thus far proved more or less correct (or at least useful approximations) over the past decade and a half of work in this area. (1) If one of the main advantages of titanium catalysis is that it is inexpensive, this advantage can be completely obviated by placing expensive ligands on the metal! Pyrroles allow easy access to multidentate ligands through condensation reactions, like the Mannich reaction, in a single step from inexpensive starting materials. (2) Pyrrole-based ligands should be robust and not exceedingly basic at nitrogen, making protolytic removal of the ancillary unlikely. We have seen some cases where pyrrole ligands are removed from the titanium during catalyses related to the one shown in Scheme 1, but generally not with primary amines.

Rothwell in 1990 that a bis(phenylamido) complex, Ti(NHPh)2(OAr)2, “will catalyze the reaction of aniline with 3hexyne to produce the N-phenylimine of 3-hexanone.”4 This reaction was inspired by the report by Bergman, Walsh, and Hollander of a zirconium imido’s reaction with an alkyne to form an azametallacyclobutene (eq 2), the hydrolysis of which provided the ketone.5 The proposed mechanism for titanium hydroamination (HA) of alkynes by primary amines is shown in Scheme 2. This mechanism was originally established for zirconocenebased catalysts but was found to be operative for titanium as well. The precatalysts for these reactions generally involve dianionic or two monoanionic ancillary ligand(s), shown as X2, and two protolytic leaving groups, dimethylamido in our case, i.e., (Me2N)2TiX2. The protolytic leaving groups of the precatalyst (Me2N)2TiX2 react with the primary amine, C


Article


Figure 2. Examples of HA of enynes to give α,β-unsaturated imines with isolated yields (Cy = cyclohexyl). Rhodium catalyzed C−H activation and insertion followed by electrocyclization in one pot gives the dihydropyridine derivative shown. Complex 2 or 3 was used as catalyst.

Figure 3. Synthesis of pyrroles from 1,4-diynes and primary amines catalyzed by 1 or 2. Bn = benzyl.

(3) The catalytic cycle in Scheme 1 should increase in rate for more electron-deficient catalysts. Biology often uses binding of water to a metal center to increase proton acidity, a fact I (A.L.O.) was reminded of by discussions on Photosystem II with the late biophysical chemist Jerry Babcock around the start of this project.9 Making the titanium center in the unit TiX2 more electron deficient should increase the acidity of the protons on the coordinated amine and increase the rate of Ti− C bond protonolysis. Likewise, increasing the metal’s Lewis acidity should also increase amine binding, also increasing protonolysis rates. (4) Finally, perhaps the most dubious assumption made was that pyrroles would provide metal centers that were relatively electron-deficient. It was by no means certain that these linkages would have advantages over aryloxides or other systems in this regard. Since, we have shown that pyrrole is far less donating than most aryloxides and many other ligands using a newly developed methodology for investigating ligands for early transition metal systems.10−12 The first ligand type generated (eq 4) was N,N-(dipyrrolyl-αmethyl)-N-methylamine (H2dpma). Using similar procedures, a host of new ligands were prepared, which have since found a variety of applications; for example, Chang, Long, and co-

Figure 4. An example of titanium-catalyzed hydrohydrazination (HH) of alkynes using 3 or 7, Fischer cyclization to the indole, and other examples with isolated yields.

workers have used tpa derivatives (Chart 1) to produce single molecule magnets.13−15 An alternative class of chelating pyrrolyl ligands is available from the condensation of ketones or aldehydes with pyrroles catalyzed by acid, e.g., 5,5-dimethyldipyrrolylmethane (H2dpm) from pyrrole and acetone (Chart 1).16 The dipyrrolylmethane H2dpm is placed on titanium by transamination with D


Article


1,4- and 1,5-diynes provided substituted pyrroles in modest yields. Figure 3 shows some examples involving 1,4-diynes.24 Concurrent with the amine chemistry above, we attempted what was at the time a little-explored reaction, addition of hydrazines to alkynes. The first successful catalysts for this reaction were pyrrole-based 3 and Ti(SC6F5)2(NMe2)2(NHMe2) (7). “Hydrohydrazination” (HH), hydrazinyl, and hydrogen addition across a C−C unsaturation proceeded smoothly with these two catalysts to give hydrazones provided that the hydrazine was 1,1-disubstituted. These reactions involved, in the place of imido, hydrazido(2−) complexes of titanium, TiN−NR2. If the hydrazine bore an aromatic group, addition of ZnCl2 to the hydrazone gave the indole products (Figure 4).15,25,26 Bubbling acetylene through a solution of hydrazine and 7 gives acetaldehyde-like hydrazines in high yields (eq 6).

Scheme 3. Two Examples of Catalysis with Ti(NMe2)2(enp) (5) to Give Nitrogen Heterocycles

Ti(NMe2)4 to give a structure with one η5-pyrrolyl and one η1pyrrolyl, Ti(NMe2)2(dpm) (2). When primary amines are used as the substrate, H2dpma and H2dpm are the most commonly employed ligands; however, an interesting menagerie of ligand structures has been prepared during various investigations. These ligands can be placed on titanium (Chart 1) in high yield by transamination with Ti(NMe2)4. For example, H2dpma (eq 4) reacts with Ti(NMe2)4 in near quantitative yield to give Ti(NMe2)2(dpma) (1).16−19 Catalyst 1 is relatively mild with a wide substrate scope that, for example, can be used to hydroaminate 1-hexyne with aniline in ∼6 h at 75 °C, providing 90% yield of the imine product as a single regioisomer.20 Catalyst 2 is one of the most reactive catalysts for intermolecular HA reported and is useful for more difficult, bulky substrate pairs.16 With less bulky substrates, the catalyses with 2 can be very rapid; for example, the catalyzed reaction of aniline with 1hexyne by 5 mol % Ti(NMe2)2(dpm) gave 57% yield in 5 min starting at room temperature.16 Unfortunately, as we realized later, the exotherm of the reaction volatilized the 1-hexyne starting material, the lowest boiling component of the solution, leading to the modest yield. Running the reaction under the same conditions but sealing the vessel as quickly as possible gave yields up to 71% (eq 5).

Catalysts 3 and 7 were only applicable to 1,1-disubstituted hydrazines (Scheme 3). Monosubstituted hydrazines, e.g., H2NN(H)Ph, with 3 generally led to free ligand and recovery of starting materials. In an attempt to gain catalysis with these monosubstituted hydrazines, two dap ligands were linked into a single, tetradentate ligand (enp) derived from N,N′-dimethylethylenediamine. The titanium complex bearing this tetradentate ligand, Ti(NMe2)2(enp) (5) in Chart 1, was an effective catalyst for HH of alkynes with monosubstituted hydrazines. With simple alkynes, like 1-phenylpropyne in Scheme 3, the hydrazone formed cleanly, and the addition of ZnCl2 gave NHindoles. In a few cases, other cyclizations occurred with the βNH of the hydrazone, such as the 5-endo trig cyclization in the bottom example of Scheme 3.17 All of the above chemistry relies on relatively simple twocomponent coupling reactions, HA or HH; the catalyst is making a single new C−N bond. Not long after beginning work in this field of titanium catalysis, new reactions involving modification of the catalytic cycle in Scheme 2 were discovered, titanium-catalyzed 3-component coupling (3CC) reactions iminoamination (IA) and iminohydrazination (IH).

3. TITANIUM-CATALYZED MULTICOMPONENT COUPLING REACTIONS The Bergman Mechanism for hydroamination (Scheme 2) involves the generation of a potentially quite reactive titanium− carbon bond after [2 + 2]-cycloaddition between a titanium imido and an alkyne. In the simple HA reaction, this intermediate is trapped in the slow step of the catalysis by reaction with a proton from amine. Can we trap the same intermediate with something other than a proton prior to protonolysis? The natural choice for a trapping agent for the azatitanacyclobutene intermediate (A in Scheme 4) was isonitrile, CNR.4 It was discovered that catalyst 1 would form new C−C and C−N bonds in a single catalytic cycle in the presence of a primary amine, alkyne, and isonitrile. The reaction was proposed to proceed by the mechanism in Scheme 4, where the Ti−C bond is trapped by 1,1-insertion of the isonitrile followed by hydrolysis with primary amine.27 The final product is due to the formal addition of an iminyl

Hydroamination in general, and titanium HA in particular, has grown into a very popular area of study by organic and inorganic chemists alike using a wide variety of ancillary ligands, and the field has been extensively reviewed. We have pursued several applications of this simple reaction chemistry to generate nitrogen-based heterocycles using catalysts like those shown in Chart 1.21 In one study, we used catalysts 2 and 3 for the selective HA of enynes, which produced α,β-unsaturated imines (Figure 2). Further, it was shown that α,β-unsaturated imines produced in this way, activated by Wilkinson’s catalyst, would insert alkynes and undergo 6-electron electrocyclization to a pyridine derivative.22 Ellman, Bergman, and co-workers have developed related cyclizations into an elegant methodology.23 In related chemistry involving simple HA of alkynes, it was shown that titanium-catalyzed additions of primary amines to E


Article


(Scheme 5). The first byproduct in IA is HA of the alkyne. The second byproduct was formamidine, which is formed by titanium-catalyzed addition of the amine to the isonitrile. The third byproduct was of unknown structure but had the mass of two equiv of isonitrile, one equiv of alkyne, and one equiv of aminea 4-component coupling product. Despite trying many relatively large-scale IA reactions with catalyst 1, we were unable to get enough of the 4-component coupling product (Scheme 5) to isolate and characterize due to the small quantity produced. It was reasoned that slowing the protonolysis reaction in Scheme 4 should allow buildup of intermediate B, which could then insert a second equiv of isonitrile; this would provide a product of the same mass as the observed 4-component byproduct. To slow down the protonolysis reaction in Scheme 4 (conversion of C to D), we made the ancillary ligands more donating by using indoles rather than pyrroles (point (3) in the Introduction to Titanium Imido and Hydrazido(2−) Catalysis).11 Making the catalyst more electron-rich should diminish the protonolysis rate and increase the relative abundance of intermediate B, allowing insertion of a second equiv of isonitrile. The catalyst was prepared (eq 7) from 2,3dimethylindole (HIndMe2).30

Scheme 4. Proposed Mechanism for the Iminoamination (IA) of Alkynes Catalyzed by Titanium

Scheme 5. By-products in the IA Reaction with 1 That Were Observed in Small Quantities by GC−MS

With 8 as catalyst, the 4-component coupling (4CC) product became the major species when an excess of isonitrile was employed. The 4CC species were found to be unusual examples of 2,3-diaminopyrroles. The proposed mechanism is shown in Figure 5 along with some examples of these 4CC products.

4. ONE-POT SYNTHESES OF NITROGEN-BASED HETEROCYCLES USING IMINOAMINATION The products of the iminoamination 3-component coupling (IA 3CC) reactions of alkynes are unsymmetrical 1,3-diimines that often would be multiple steps to synthesize using other methods. These 1,3-diimines can be used in many different heterocyclic syntheses. In general, the IA product is not isolated but instead synthesized and used in a one-pot procedure to construct the heterocyclic ring. The reactions allow ready access to several heterocyclic cores in one-pot reactions (Scheme 1). Quite simply, these reactions involve loading in a reaction vessel with solvent, catalyst, amine, alkyne, and isonitrile and heating to form the IA product. Once the IA product is formed, another reagent is added for the cyclization, along with additives. A brief video of one such reaction is shown in the Supporting Information (SI), which includes synthesis of 4phenylpyrazole in a one-pot sequence without the use of a glovebox and with in situ generated catalyst.

(−C(H)NR4) and amino group across the carbon−carbon triple bond, iminoamination (IA) of the alkyne. The products are tautomers of unsymmetrical 1,3-diimines.28 Thermodynamically, the IA reaction to couple alkyne, primary amine, and isonitrile is calculated as quite exothermic with ΔH° ≈ −65 kcal/mol. Entropically, one would expect that the atom-economical multicomponent coupling is quite unfavorable, but iminoamination is still calculated to have a substantial driving force, ΔG° ≈ −41 kcal/mol.29 Typical yields for IA reactions when run using catalyst 1 or 2 are around 70% with most substrates. The reaction is a bit more challenging than simple HA, requiring slightly higher catalyst loadings (typically 10 mol %) and somewhat longer reaction times (12−48 h) depending on the substrates. In both HA and IA, there is substantial control of regioselectivity of addition to the alkyne. The major products in HA and IA do not necessarily come from the same azatitanacyclobutene intermediate; this is likely a reflection of Curtin−Hammett kinetics with two different trapping agents (and different relative rates of trapping) for the metallacycles. In the initial screens for the iminoamination (IA) reaction using 1, three minor but identifiable byproducts were observed

4.1. Overview of Titanium-Catalyzed Iminoamination (IA) in Heterocyclic Synthesis

The reactions developed thus far can be divided into two categories depending on the number of atoms from the IA product present in the heterocycle (Figure 6). The IA product has a 5-atom backbone, N1, C2, C3, C4, and N5, which are F


Article


Figure 5. (left) Proposed mechanism of the 4-component coupling (4CC) of 2 equiv of isonitrile, 1 equiv of alkyne, and 1 equiv of amine by 5 mol % 8. (right) Examples of products prepared using the 4CC: isolated % yield (GC % yield).

Scheme 6. (top) Synthesis of 4- or 5-Alkyl-1-phenylpyrazoles in a One-Pot Procedure Using Catalysts 1 and 2, and (bottom) the Proposed Cause for the 1,5-Substituted Pyrazole Being the Major Product Is the Preferred Initial Attack at the Aldimine Carbon

Figure 6. Synthesis of the heterocyclic products in Scheme 1 divided into two categories dubbed (3 + n) and (4 + 2) heterocyclizations.

colored in red. The reactions will be classified by how many of these 5-atoms in the backbone are in the heterocyclic product (first number) and by how many atoms from the reagent added to the IA product complete the heterocyclic core (second number). (3 + n)-Heterocyclizations. In the first type, the IA product acts as a 3-carbon containing synthon for the frame of the heterocycle. In these reactions, two amines will be lost as byproducts. These will be referred to as (3 + n)heterocyclizations, where the first number (3 in this case) represents the number of atoms in the IA product backbone used in the product and n represents the atoms from the G


Article


Figure 7. Conditions for the synthesis of (left) pyrimidines and (right) isoxazoles in a one-pot synthesis along with some representative examples and isolated yields.

for attack on the aldimine carbon by the NH2 group of the hydrazine, which leads to the observed product. For internal alkynes, regiochemical issues for addition of monosubstituted hydrazines always arise, but initial attack on the aldimine carbon appears to be strongly preferred. For example, if 1-phenylpropyne is used as the alkyne substrate in IA, there is a strong electronic preference to place the phenyl group in the position shown on the left side of (eq 8). Reaction

reactant added to the IA product that are in the heterocyclic frame. (4 + 2)-Heterocyclizations. The second type of reactions discovered are (4 + 2)-heterocyclizations, where 4 atoms from the IA product are added to two atoms of another reagent. In these cases, the only byproduct will be one primary amine, often tert-butylamine from the isonitrile. The pyridine products at the bottom right of Figure 6 are mechanistically related but isolable as single species through control of reaction conditions. 4.2. (3 + n)-Heterocyclizations: Pyrimidines (n = 3), pyrazoles (n = 2), and isoxazoles (n = 2)

In these reactions, the IA product is used as a readily produced 3-carbon backbone for heterocyclic synthesis. Although the IA reaction is 100% atom economical, these heterocyclizations result in the loss of two equiv of amine. Though lower in atomeconomy than their (4 + n) counterparts, controlling the identity of the amines, which do not appear in the product, can be a useful tool for controlling the regioselectivity of the reactions. The chemistry for pyrazole formation will be discussed in some detail as an example. Pyrazoles.31 Pyrazoles can be prepared by a one-pot titanium-catalyzed IA followed by addition of monosubstituted hydrazines or hydrazine hydrate in pyridine (see SI video for an example). The yields for the two-step reactions were modest (28−50% isolated yields on 1 mmol scales), but a large number of pyrazoles are readily accessed from the simple procedure. The favored products are 1,4- or 1,5-disubstituted pyrazoles with terminal alkynes and monosubstituted hydrazines. The regioselectivity of the product can be controlled with catalyst and conditions if the substituent on the terminal alkyne is an alkyl. For example, the butyl group of 1-hexyne can be placed in either the 4- or 5-position with reasonable selectivity by changing the catalyst. Two examples of such reactions are shown in Scheme 6. In the case of the 1,5-disubstituted product, a different regioisomer (1,3-disubstituted pyrazole) was possible but not observed (Scheme 6). There appears to be a strong preference

of this IA product with phenylhydrazine preferentially gives the product (9:1) derived from the NH2 group of the hydrazine substituting first at the aldimine site followed by ring closure to the 1,4,5-trisubstituted pyrazole (c.f., Scheme 6). This one-pot synthesis of pyrazoles was used to generate one of the few known natural products including the pyrazole core, withasomnine, efficiently from commercially available compounds. Pyrimidines 32 and Isoxazoles. 33 In the case of pyrimidines (Figure 7), there were often no regiochemical issues as many of the amidines are C2v symmetric. Isoxazoles were accessible in a regioselective manner as well from a one-pot synthesis (Figure 7). After initial formation of the IA product, hydroxylamine hydrochloride was added in ethanol at room temperature for cases where no regiochemical issues were present. For internal alkynes, it was advantageous to use 3,5-dichlorophenyl as R1 and THF as solvent, which gave a single regioisomer for the isoxazole product. 4.3. (4 + 2)-Heterocyclizations: Quinolines and 2-Aminopyridines

In a second category of heterocyclizations, 4-members of the IA product’s core are included in the final heterocycle. Two classes H


Article


Scheme 7. Reaction of the IA Product with Malononitrile Provides Either 2-Amino-3-cyanopyridines (A, 2Aminonicotinonitrile) or 1,2-Dihydro-2-imino-3pyridinecarbonitriles (B) Depending on the Conditions

Figure 8. Synthesis of quinolines and related heterocycles using IA in a one-pot procedure.

of heterocyclic compounds have been studied thus far in this category: quinolines and 2-aminopyridines. Quinolines.34 The simplest of these, both mechanistically and operationally, is the use of IA to generate quinolines. The reaction is limited to the use of an aniline derivative (or other aromatic amine) with at least one o-hydrogen; however, thousands of such compounds are commercially available.35 In addition, the reaction is applicable to amine-containing heterocycles as well, which produces fused-ring pyridine systems. The reaction simply involves IA with an aromatic amine, alkyne, and isonitrile to produce the 3CC product. Then, acetic acid is added to promote electrophilic cyclization onto the aromatic ring of the amine with further heating to produce quinolines from aniline-derivatives or similar heterocyclic systems (Figure 8). The only byproduct is the amine from which the isonitrile is derived, typically H2NBut. 2-Amino-3-cyanopyridines.36 Reaction of the IA product with malononitrile (Scheme 7) provided two different products, 2-amino-3-cyanopyridines (A, 2-aminonicotinonitriles) and 1,2dihydro-2-imino-3-pyridinecarbonitriles (B), depending on the conditions employed. The production of B is relatively straightforward mechanistically (vide infra). Interestingly, note that both of the C−N bonds of the IA product are broken in

Figure 9. Synthesis of 2-aminophenyl-3-cyanopyridines in a one-pot procedure with isolated yields.

the production of A, and one of those amines is attached to a carbon originating from malononitrile (Scheme 7); the other nitrogen in the IA product is released as the primary amine. As will be shown, B is an intermediate in the synthesis of A. For this study, we chose to focus on the 2-aminonicotinonitriles (A) as products in no small part because there were interesting mechanistic issues to be explained. The chemistry of the 1,2-dihydro-2-imino-3-pyridinecarbonitriles (B) prepared in this way has not yet been explored in detail. I


Article


One amine is a byproduct, but the amine that ends up on the carbon originating with malononitrile is never lost as free amine during the rearrangement. This was shown by the addition of 10 equiv of 3,5-dimethylaniline to the cyclization (eq 11), which showed no incorporation of the external aniline during the reaction.

Scheme 8. Proposed Mechanism for the Synthesis of 2Aminopyridines by Reaction of the IA Product with Malononitrile in the Presence of an Acid and DBUa

The proposed mechanism involves a Dimroth rearrangement driven by aromatization of the pyridine ring (Scheme 8). A likely role of the acid in the rearrangement is to activate the ring for nucleophilic attack, perhaps through protonation of the imine nitrogen. The nucleophilic base, DBU, attacks the heterocyclic ring, possibly at the carbon containing R2.

5. CONCLUDING REMARKS AND PROSPECTS By manipulation of ancillary ligands, conditions, and substrates, titanium-catalyzed 2-component couplings can be used to

a

The acid catalyst for the rearrangement was typically malononitrile, which was used in excess.

If an internal alkyne is used, such as 1-phenylpropyne, with tert-butylisonitrile and various primary amines, a single regioisomer is obtained. The byproduct is tert-butylamine when the IA product is treated with 2 equiv of malononitrile and 50 mol % DBU. Eleven different primary amines were screened in the reaction with this alkyne for the one-pot sequence (eq 9).

If the primary amine is kept as aniline and different alkynes are explored, a large variety of 2-aminophenyl-3-cyanopyridines (A) can be synthesized; some examples are shown in Figure 9. Producing A involved the use of the nucleophilic base DBU in the optimized conditions. Replacing DBU with less nucleophilic NEt3 gave B. Conversion between the two compounds was possible, but only if both an acid and DBU were present (eq 10). Treatment of B with just DBU gives no reaction; treatment with DBU and an acid (malononitrile or phenol) led to rearrangement, providing A in high yield.

Figure 10. Synthesis of pyrazoles in a single step via titanium-catalyzed 3-component coupling (IH).

generate imines, hydrazones, and pyrroles. Similar catalyst systems catalyze 3-component coupling reactions iminoamination (IA) and iminohydrazination (IH). The IA reaction products, tautomers of 1,3-diimines, have been used in one-pot syntheses for a variety of different heterocycles, including pyrazoles, pyrimidines, isoxazoles, 2amino-3-cyanopyridines, and quinolines. Although applications to other heterocycles will be discovered using these one-pot methodologies, in some cases, it is likely possible to increase atom-economy of the reactions to give one-step rather than one-pot reactions, which will eliminate an amine byproduct J


Article

Accounts of Chemical Research from some reactions. However, the new one-step reactions will likely require catalyst development for substrate compatibility. For example, one could produce pyrazoles in a single multicomponent coupling reaction of a monosubstituted hydrazine, alkyne, and isonitrile, improving the atom-economy of the IA pyrazole chemistry above. In order for this chemistry to be accessed, catalysts that are active for IH with monosubstituted hydrazines are needed. Recently, catalyst 6 in Chart 1 bearing 2-pyridylpyrrolyl ligands was found to catalyze IH with monosubstituted hydrazines (Figure 10). Although the scope of the reaction was limited to terminal alkynes, catalysis with 6 provides 1,3disubstituted pyrazoles (c.f., Scheme 6).19 Like all catalytic systems, there are limitations in substrate scope and compatibility with titanium-based systems; however, there is growing diversity in the types of structures that can be accessed using this nontoxic, earth-abundant element. Further, the chemistry offers orthogonality to late metal-based catalysis, e.g., the titanium chemistry ignores many halides, silanes, and so forth that are often-used groups for late metal functionalization. Most importantly, early metal chemistry provides access to new types of bond-forming reactions that are worth developing as new tools for synthetic targets like nitrogen-based heterocycles.

■

inhibitors, and applications of the ligand donor parameter towards high valent transition metal catalysis.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge the support of the National Science Foundation for this work, currently under CHE1265738. A.L.O. would also like to express his deep appreciation for all the outstanding graduate students, undergraduate students, postgraduate scholars, and visiting scientists who have contributed to the chemistry in this Account. The authors also thank Brennan Billow and the talented Matt Bowen for their help in constructing this Account.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.accounts.5b00280. Experimental protocol for 4-phenylpyrazole and hydroamination of 1-hexyne by aniline with 2 (PDF) Video demonstrating in situ catalyst production and synthesis of 4-phenylpyrazole (MOV)

■

REFERENCES

(1) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements, 1st ed.; Pergamon Press: Oxford, 1984. (2) Nakamura, I.; Yamamoto, Y. Transition-metal-catalyzed reactions in heterocyclic synthesis. Chem. Rev. 2004, 104, 2127−2198. (3) Mathew, S.; Yella, A.; Gao, Y.; Humphry-Baker, R.; Curchod, B. F. E.; Ashari-Astani, N.; Tavernelli, I.; Rothlisberger, U.; Nazeeruddin, M. K.; Grätzel, M. Dye-sensitized solar cells with 13% efficiency achieved through the molecular engineering of porphyrin sensitizers. Nat. Chem. 2014, 6, 242−247. (4) Hill, J. E.; Profilet, R. D.; Fanwick, P. E.; Rothwell, I. P. Synthesis, Structure, and Reactivity of Aryloxo(imido)titanium Complexes. Angew. Chem., Int. Ed. Engl. 1990, 29, 664−665. (5) Walsh, P. J.; Hollander, F. J.; Bergman, R. G. Generation, Alkyne Cycloaddition, Arene C−H Activation, N−H Activation, and Dative Ligand Trapping Reactions of the First Monomeric Imidozirconocene (Cp2ZrNR) Complexes. J. Am. Chem. Soc. 1988, 110, 8729−8731. (6) Walsh, P. J.; Baranger, A. M.; Bergman, R. G. Stoichiometric and catalytic hydroamination of alkynes and allene by zirconium bisamides Cp2Zr(NHR)2. J. Am. Chem. Soc. 1992, 114, 1708−1719. (7) Pohlki, F.; Doye, S. The mechanism of the [Cp2TiMe2]-catalyzed intermolecular hydroamination of alkynes. Angew. Chem., Int. Ed. 2001, 40, 2305−2308. (8) Shi, Y.; Ciszewski, J. T.; Odom, A. L. Ti(NMe2)4 as a Precatalyst for Hydroamination of Alkynes with Primary Amines. Organometallics 2001, 20, 3967−3969. (9) (a) Tommos, C.; Babcock, G. T. Proton and hydrogen currents in photosynthetic water oxidation. Biochim. Biophys. Acta, Bioenerg. 2000, 1458, 199−219. (b) Chang, C. J.; Chang, M. C. Y.; Damrauer, N. H.; Nocera, D. G. Proton-coupled electron transfer: a unifying mechanism for biological charge transport, amino acid radical initiation and propagation, and bond making/breaking reactions of water and oxygen. Biochim. Biophys. Acta, Bioenerg. 2004, 1655, 13−28. (10) Bemowski, R. D.; Singh, A. K.; Bajorek, B. J.; DePorre, Y.; Odom, A. L. Effective donor abilities of E-t-Bu and EPh (E = O, S, Se, Te) to a high valent transition metal. Dalton Trans 2014, 43, 12299− 12305. (11) DiFranco, S. A.; Maciulis, N. A.; Staples, R. J.; Batrice, R. J.; Odom, A. L. Inorg. Chem. 2012, 51, 1187−1200. (12) Ciszewski, J. T.; Harrison, J. F.; Odom, A. L. Investigation of transition metal-imido bonding in M(NBut)2(dpma). Inorg. Chem. 2004, 43, 3605−3617. (13) Li, Y.; Turnas, A.; Ciszewski, J. T.; Odom, A. L. Group-4 η1pyrrolyl complexes incorporating N,N-di(pyrrolyl-α-methyl)-N-methylamine. Inorg. Chem. 2002, 41, 6298−6306. (14) Shi, Y.; Cao, C.; Odom, A. L. Synthesis and group-4 complexes of tris(pyrrolyl-α-methyl)amine. Inorg. Chem. 2004, 43, 275−281. (15) Freedman, D. E.; Harman, W. H.; Harris, T. D.; Long, G. J.; Chang, C. J.; Long, J. R. Slow magnetic relaxation in a high-spin iron(II) complex. J. Am. Chem. Soc. 2010, 132, 1224−1225. (16) Cao, C.; Shi, Y.; Odom, A. L. Intermolecular alkyne hydroaminations involving 1,1-disubstituted hydrazines. Org. Lett. 2002, 4, 2853−2856.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies Aaron L. Odom was born in Wellington, Texas and received a B.S. in Chemistry from Texas Tech University in 1993. He obtained his Ph.D. in 1997 under the instruction of Prof. Christopher “Kit” C. Cummins at the Massachusetts Institute of Technology. He accepted a position at Michigan State University as an Assistant Professor while a graduate student but stayed at MIT to complete postdoctoral studies with Prof. Daniel G. Nocera. He began his independent career at MSU in 1999 and has received the Presidential Early Career Award for Scientists and Engineers, Department of Energy−Career Scientist and Engineer Award, Sloan Scholar Award, and has been an Office of Naval Research Young Investigator. Tanner J. McDaniel obtained his B.S. in Chemistry from Winona State University in 2012. During his undergraduate studies, he worked for Dr. Thomas Nalli on the identification of fatty acids in lipid extracts from Mississippi River biota in collaboration with the Upper Midwest Environmental Sciences Center. He then moved to Michigan State University to pursue his doctoral degree. His research has been focused on exploring applications of titanium-catalyzed multicomponent coupling reactions, the synthesis of novel proteasome K


Article

Accounts of Chemical Research (17) Shi, Y.; Hall, C.; Ciszewski, J. T.; Cao, C.; Odom, A. L. Titanium dipyrroylmethane derivatives: Rapid intermolecular alkyne hydroamination. Chem. Commun. 2003, 586−587. (18) Banerjee, S.; Barnea, E.; Odom, A. L. Titanium-catalyzed hydrohydrazination with monosubstituted hydrazines: Catalyst design, synthesis, and reactivity. Organometallics 2008, 27, 1005−1014. (19) Dissanayake, A. A.; Odom, A. L. Single-step synthesis of pyrazoles using titanium catalysis. Chem. Commun. 2012, 48, 440−442. (20) Cao, C.; Ciszewski, J. T.; Odom, A. L. Hydroamination of alkynes catalyzed by a titanium pyrrolyl complex. Organometallics 2001, 20, 5011−5013. (21) (a) Yim, J. C.-H.; Schafer, L. L. Efficient anti-Markovnikovselective catalysts for intermolecular alkyne hydroamination: Recent advances and synthetic applications. Eur. J. Org. Chem. 2014, 2014, 6825−6840. (b) Muller, T. E.; Hultzsch, K. C.; Yus, M.; Foubelo, F.; Tada, M. Hydroamination: Direct addition of amines to alkenes and alkynes. Chem. Rev. 2008, 108, 3795−3892. (c) Severin, R.; Doye, S. The catalytic hydroamination of alkynes. Chem. Soc. Rev. 2007, 36, 1407−1420. (d) Odom, A. L. New C−N and C−C bond forming reactions catalyzed by titanium complexes. Dalton Trans. 2005, 225− 233. (22) Cao, C.; Li, Y.; Shi, Y.; Odom, A. L. α,β-Unsaturated imines form titanium hydroamination and functionalization by rhodium C−H activation. Chem. Commun. 2004, 2002−3. (23) Colby, D. A.; Tsai, A. S.; Bergman, R. G.; Ellman, J. A. Rhodium catalyzed chelation-assisted C−H bond functionalization reactions. Acc. Chem. Res. 2012, 45, 814−825. (24) Ramanathan, B.; Keith, A. J.; Armstrong, D.; Odom, A. L. Pyrrole syntheses based on titanium-catalyzed hydroamination of diynes. Org. Lett. 2004, 6, 2957−2960. (25) Thinking the term to be a mouthful at the time, we did not use “hydrohydrazination” in ref 16. It was perhaps first used in the literature here: Waser, J.; Carreira, E. M. Convenient synthesis of alkylhydrazies by the cobalt-catalyzed hydrohydrazination reaction of olefins and azodicarboxylates. J. Am. Chem. Soc. 2004, 126, 5676− 5677. (26) (a) Patel, S.; Li, Y.; Odom, A. L. Synthesis, structure, and LLCT transitions in terminal hydrazido(2−) bipyridine complexes of titanium. Inorg. Chem. 2007, 46, 6373−6381. (b) Banerjee, S.; Odom, A. L. Synthesis and structure of a titanium hydrazido(2−) complex. Organometallics 2006, 25, 3099−3101. (c) Schofield, A. D.; Nova, A.; Selby, J. D.; Schwarz, A. D.; Clot, E.; Mountford, P. Reaction site diversity in the reactions of titanium hydrazides with organic nitriles, isonitriles and isocyanates: TiNα cycloaddition, TiNα insertion and Nα-Nβ bond cleavage. Chem. - Eur. J. 2011, 17, 265−285. (27) Vujokovic, N.; Fillol, J. L.; Ward, B. D.; Wadepohl, H.; Mountford, P.; Gade, L. H. Insertions into azatitanacyclobutenes: New insights into three component coupling reactions involving imidotitanium intermediates. Organometallics 2008, 27, 2518−2528. (28) Cao, C.; Shi, Y.; Odom, A. L. A titanium-catalyzed threecomponent coupling to generate α,β-unsaturated β-iminoamines. J. Am. Chem. Soc. 2003, 125, 2880−2881. (29) The thermodynamics were calculated using the G3 method of Gaussian09 for the model reaction of acetylene, methylisonitrile, and methylamine. (30) Barnea, E.; Majumder, S.; Staples, R. J.; Odom, A. L. One-step route to 2,3-diaminopyrroles using a titanium-catalyzed fourcomponent coupling. Organometallics 2009, 28, 3876−3881. (31) Majumder, S.; Gipson, K. R.; Staples, R. J.; Odom, A. L. Pyrazole synthesis using a titanium-catalyzed multicomponent coupling reaction and synthesis of withasomnine. Adv. Synth. Catal. 2009, 351, 2013−2023. (32) Majumder, S.; Odom, A. L. Titanium catalyzed one-pot multicomponent coupling reactions for direct access to substituted pyrimidines. Tetrahedron 2010, 66, 3152−3158. (33) Dissanayake, A. A.; Odom, A. L. Regioselective conversion of alkynes to 4-substituted and 3,4-disubstituted isoxazoles using titanium-catalyzed multicomponent coupling reactions. Tetrahedron 2012, 68, 807−812.

(34) Majumder, S.; Gipson, K. R.; Odom, A. L. A multicomponent coupling sequence for direct access to substituted quinolines. Org. Lett. 2009, 11, 4720−4723. (35) SciFinder search on 5/10/15 of aniline derivatives with at least one o-H that are commercially available as a single component with at least one reference gave ∼54,000 compounds. Obviously, some of these will have compatibility issues. (36) Dissanayake, A. A.; Staples, R. J.; Odom, A. L. Titaniumcatalyzed, one-pot synthesis of 2-amino-3-cyanopyridines. Adv. Synth. Catal. 2014, 356, 1811−1822.

L


Titanium-Catalyzed Multicomponent Couplings: Efficient One-Pot

Recommend Documents