Download PDF
Synlett 2025; 36(11): 1435-1440
DOI: 10.1055/a-2538-2165
synpacts

Cyclic Imine-BF3 Complexes as Precursors for Functionalized Azacycles

Kamal Bhatt
,
Daniel Seidel 

Funding

Financial support from the National Institute of General Medical Sciences, National Institutes of Health (NIH-NIGMS) (R35GM149246) is gratefully acknowledged.

› Further Information
 
 PDF Download Permissions and Reprints All articles of this category


Abstract

Due to the relative instability and low electrophilicity of enolizable alicyclic imines, their functionalization commonly requires cryogenic temperatures and highly reactive nucleophiles such as organolithium compounds. Stable BF3 adducts of these imines streamline the synthesis of functionalized amines and obviate the need for cryogenic temperatures. In favorable cases, these adducts can be stored for over a year. The compatibility of cyclic imine-BF3 complexes with organometallic and radical-centered nucleophiles makes them ideal building blocks for functionalized azacycles.

1 Introduction

2 Synthesis of Cyclic Imine-BF3 Complexes

3 Reactions of Imine-BF3 Complexes with Organometallic Nucleophiles

4 Radical Additions to Imine-BF3 Complexes

5 Conclusions


#

Key words

imine-BF3 complexes - alicyclic amines - acridines - photocatalysis - radical additions
1

Introduction

Zoom Image
Daniel Seidel studied chemistry at the Friedrich-Schiller-Universität Jena, Germany and at the University of Texas at Austin, USA (Diplom 1998). He performed his graduate studies in the lab of Prof. Jonathan L. Sessler, obtaining his Ph.D. in 2002. From 2002 to 2005, he was an Ernst Schering Postdoctoral Fellow in the group of Prof. David A. Evans at Harvard University. He started his independent career at Rutgers University in 2005 and was promoted to associate professor in 2011 and full professor in 2014. In the summer of 2017, his research group moved to the University of Florida, where he is currently the Katritzky Term Professor in Heterocyclic Chemistry.

Lewis acids such as boron trifluoride–diethyl etherate (BF3·OEt2) have been extensively used to activate imines toward nucleophilic addition, finding utility in traditional organometallic[1] as well as radical chemistry.[2] However, the use of imine-borane complexes has been largely limited to either linear imines or non-enolizable cyclic imines. Piers and co-workers[3] first isolated and characterized (including by X-ray crystallography) the adducts of simple benzophenone- and benzaldehyde-derived imines with the bulky Lewis acid tris(pentafluorophenyl)borane B(C6F5)3, establishing the stability of such complexes (Scheme [1a]). Collum et al. later demonstrated that even simple enolizable imine-BF3 adducts are stable when limiting air exposure.[4] These adducts have been utilized in stereoselective Mannich reactions, highlighting the utility of such complexes (Scheme [1b]).[5] Given the medicinal importance of functionalized azacycles,[6] we considered the possibility of utilizing previously unknown BF3 adducts of cyclic imines as building blocks (Scheme [1c]). The stability of such compounds was unclear, as BF3 coordination increases the acidity of β-hydrogens, potentially promoting deprotonation/oligomerization.[7] However, if successful, such compounds could emerge as versatile building blocks to access functionalized nitrogen heterocycles, eliminating the additional step of having to prepare unstable cyclic imines for the synthesis of every new desired product. Here we summarize our successful attempts to synthesize and utilize cyclic imine-BF3 adducts as precursors for functionalized alicyclic amines.

Zoom Image
Scheme 1 Borane-imine complexes

# 2

Synthesis of Cyclic Imine-BF3 Complexes

In earlier work, we showed that cyclic imines generated upon the oxidation of lithium amides with ketones could be activated by the addition of BF3·OEt2, presumably due to the formation of cyclic imine-BF3 adducts.[8] We were curious to learn whether these imine-BF3 adducts could be isolated. Gratifyingly, trapping the in situ generated 1-piperideine with boron trifluoride–diethyl etherate gave a 40% yield (unoptimized) of the desired 1-piperideine-BF3 complex 1a (Scheme [2a]).[9] Complex 1a is a column-stable crystalline solid that can be stored at –20 °C for over a year without any noticeable degradation. Despite the short reaction time, the disadvantage of this method is the need for cryogenic temperatures. To avoid this requirement, we attempted to access 1a from the well-known 1-piperideine trimer.[10] Previous reports had noted that detrimerization of 1-pyrroline and 1-piperideine trimers could be achieved under Brønsted acidic conditions, suggesting that the same might be possible with Lewis acids.[11] Indeed, detrimerization with BF3·OEt2 proceeded smoothly at room temperature, furnishing 1a in 81% yield (Scheme [2b]). This method is applicable to gram-scale production without significantly affecting the reaction yield. However, a limitation is the need for isolated trimers, species that can only reliably be obtained from piperidine and pyrrolidine.

We thus desired to establish a more general procedure for converting various alicyclic amines into their corresponding imine-BF3 adducts under mild conditions. The reported N-halogenation/dehydrohalogenation of piperidine to access 1-piperideine requires polar protic solvents, potentially hampering the in situ formation of imine-BF3 adducts.[10] To avoid this incompatibility, we investigated a phase-transfer approach for the dehydrohalogenation step, using toluene as the solvent (Scheme [2c]). Using cesium hydroxide monohydrate as the base and 18-crown-6 as a phase-transfer catalyst furnished the desired imine-BF3 adducts in a single operation. This method is applicable to cyclic amines of different ring sizes and those bearing ring substituents. Despite its advantages, the heterogeneous nature of the phase-transfer approach requires stir-rate optimization to obtain reproducible results, which can render scaleup challenging. To develop a readily scalable procedure, we explored a telescoped approach where N-halogenation/dehydrohalogenation was conducted in ethanol as the solvent (Scheme [2d]). After a quick workup and solvent change to either diethyl ether or dichloromethane, the crude imine (presumably a mixture of monomer and trimer) was exposed to BF3·OEt2 to furnish the desired imine-BF3 adducts.[12] This is easily scalable to 10 mmol. However, larger ring sizes (e.g., 7- and 8-membered amines) provide unfavorable results with this approach.

Zoom Image
Scheme 2 Different strategies to access cyclic imine-BF3 complexes

# 3

Reactions of Imine-BF3 Complexes with Organometallic Nucleophiles

We first investigated the possibility of using imine-BF3 adducts as substrates for the addition of organometallic nucleophiles at ambient temperatures (Scheme [3]).[8] To our delight, a range of Knochel zinc reagents (products 2a, 2b, and 2d) and Grignard reagents (product 2e) performed well. Organolithium reagents also proved to be viable but typically require a reaction temperature of –78 °C to obtain good yields (products 2f–h). Imine-BF3 complexes with existing substituents furnished products with excellent diastereoselectivity (2f and 2g). Due to the instability of morpholine and piperazine-derived imine-BF3 complexes, an in situ addition protocol was established to directly access products such as 3a–d. It is worth mentioning that our previous attempts to functionalize morpholine via lithium amide chemistry failed to provide any meaningful quantities of the desired products.[13]

Zoom Image
Scheme 3 Selected scope for the reaction of imine-BF3 complexes with organometallic nucleophiles

# 4

Radical Additions to Imine-BF3 Complexes

4.1

Overview of Decarboxylative Additions to Imine Derivatives

Nucleophilic radical addition to imine derivatives is an attractive route to substituted amines.[14] Advancements in photochemistry have opened multiple avenues of radical generation under ambient conditions. Of specific interest are photo-decarboxylative transformations that utilize commercially abundant carboxylic acids as feedstock chemicals.[15] N-Aryl imines and imines containing an electron-withdrawing group on nitrogen have been extensively used in radical additions (Scheme [4a]).[14] [15] In contrast, there are still only a few reports on N-alkyl imines as radical acceptors,[2b,16] a fact that is undoubtedly due to the low electrophilicity of N-alkyl imines. Prior to our work, radical additions to enolizable alicyclic imines had remained elusive. In addition to their poor stability and electrophilicity, the fact that the amine products are easily oxidizable represents another significant challenge. We envisioned that imine-BF3 adducts would not only be sufficiently electrophilic to engage alkyl radicals, but that BF3 coordination to the amine product would serve the dual role of preventing undesirable oxidation (Scheme [4b]). A higher order variant of this methodology involving [1.1.1]propellane as an additional reagent would allow the formation of valuable α-amino bicyclopentanes (BCPs).

Zoom Image
Scheme 4 Overview and concept

# 4.2

Decarboxylative Addition to Cyclic Imine-BF3 Complexes

Initially discovered by the Nozaki[17a] [b] and Oda[17c] groups, and later popularized by Larionov et al.,[17d,e] acridines enable a base-free decarboxylation of alkyl carboxylic acids to generate alkyl radicals under mild conditions. The Larionov group has also shown that a catalytic amount of copper salts can improve the catalytic turnover of acridine photocatalysts.[17e] Drawing inspiration from these reports, we established an acridine-copper dual catalytic system for functionalizing cyclic imine-BF3 complexes.[12] After extensive optimization, we found that irradiation (395 nm LED) of the substrates in dichloromethane solution at room temperature, in the presence of an acridine photocatalyst, a copper source, and triethylamine, produced the best results for secondary and tertiary carboxylic acids. On the other hand, running the reaction in dichloroethane (DCE) at an elevated temperature of 70 °C worked best for primary carboxylic acids. Catalytic amounts of triethylamine proved essential. This additive is likely responsible for stabilizing different copper intermediates. The use of acridine photocatalyst 6, previously shown by us to offer enhanced stability over frequently used acridines,[18] is crucial to achieving high yields. A selected scope is shown in Scheme [5]. A range of secondary (products 5a–g and 5j), tertiary (products 5h and 5k), and primary carboxylic acids (products 5i and 5l–n) were found to be viable substrates. Functional groups such as esters, NH-carbamates, and free alcohols were tolerated. Late-stage modification of bioactive molecules is also possible due to the mild reaction conditions. Although some tertiary acids (products 5h and 5k) furnished the desired products, more sterically encumbered tertiary-acid-derived radicals failed to engage with imine-BF3 complexes. Less nucleophilic benzylic radicals also failed to add to imine-BF3 complexes under our optimized conditions.

Zoom Image
Scheme 5 Selected scope of decarboxylative additions to imine-BF3 complexes

# 4.3

Synthesis of BCP-Substituted Azacycles

1,3-Disubstituted bicyclo[1.1.1]pentanes (BCPs) are sought-after bioisosteres for para-substituted phenyl rings.[19] The increased three-dimensionality of BCP can improve the efficacy of a drug candidate by improving its solubility and metabolic stability. α-Amino BCPs are benzylamine analogs, representing a relatively underdeveloped class of compounds that are present in several bioactive cores.[20] A recent report from the Molander group disclosed a three-component reaction of N-tosyl/aryl imines, [1.1.1]propellane, and alkyl radicals derived from alkyltrifluoroborates to access BCP-alkylamines.[21] Remarkably, the authors exclusively obtained three-component products, highlighting that the rate of alkyl radical addition to [1.1.1]propellane outcompetes the direct radical addition to imines. Motivated by this report and our ability to prepare product 5h, we developed a three-component variant to access BCP-substituted azacycles (Scheme [6]). Irradiating the substrates with three equivalents of [1.1.1]propellane and removing the copper source provided optimal results.[12] Interestingly, carboxylic acid substrates that failed in the two-component variant proved suitable for the three-component version. The reaction can tolerate esters, amides, thioethers, carbamates, and halides. Modification of medicinally relevant compounds like bezafibrate (product 8f) and gemfibrozil (products 8h and 8i) is readily achieved.

Zoom Image
Scheme 6 Selected scope for BCP-substituted azacycles

# 4.4

Plausible Reaction Mechanisms

An interesting feature of the two-component variant is the formation of a copper mirror that appears within 30 minutes of irradiation. To investigate the potential involvement of copper(0) in the reaction mechanism, we performed the model reaction with copper(0) nanopowder (previously shown to be a productive co-catalyst in a different reaction).[17e] However, this provided a diminished yield of 35%, substantially lower than the 57% yield obtained without any copper catalyst.[12] This indicates that the copper mirror is most likely produced by the disproportionation of Cu(I) and has little to no bearing on the reaction mechanism other than providing a source of Cu(II). A plausible mechanism for the direct radical addition is shown in Scheme [7a]. The hydrogen-bonded acridine complex, upon photoexcitation, undergoes proton-coupled electron transfer (PCET) generating CO2, the cyclohexyl radical, and acridinyl radical 6H• . The cyclohexyl radical is then captured by the imine-BF3 complex 1a along with Cu(I) complex 10 to form intermediate 11. Oxidation of acridinyl radical 6H• by Cu(II) species 9 generates acridinium 6H+ and Cu(I) complex 10. Acridinium 6H+ protonates intermediate 11 to furnish the amine-BF3 complex 12 and acridine 6. In the three-component version (Scheme [7b]), the cyclohexyl radical first reacts with [1.1.1]propellane (7) to form BCP radical intermediate 13, which then reacts with the imine-BF3 complex 1a, generating intermediate 14. Aminyl radical 14 is then reduced by the acridinyl radical 6H• to form acridinium 6H+ and intermediate 15 which, upon protonation, forms the product-BF3 complex 16, regenerating the photocatalyst.


#
# 5

Conclusions

In summary, we have established reliable procedures to access imine-BF3 complexes derived from cyclic amines. The benefits and challenges of the different methods have been discussed to enable practitioners to select the methodology that best suits their requirements. This novel class of compounds provides expedient access to functionalized amines via organometallic and radical approaches. Ambient photochemical conditions offer significantly improved functional group compatibility and allow for late-stage modifications. We expect that these transformations and anticipated future improvements will establish cyclic imine-BF3 complexes as routine building blocks for the synthesis of substituted azacycles.

Zoom Image
Scheme 7 Plausible reaction mechanisms

#
#

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

We are grateful to our co-workers who contributed to the development of the chemistry described in this account.


Corresponding Author

Daniel Seidel
Center for Heterocyclic Compounds, Department of Chemistry, University of Florida
Gainesville, Florida 32611
USA   

Publication History

Received: 05 January 2025

Accepted after revision: 12 February 2025

Accepted Manuscript online:
12 February 2025

Article published online:
05 May 2025

© 2025. Thieme. All rights reserved

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany


   Permissions and Reprints All articles of this category
Zoom Image
Daniel Seidel studied chemistry at the Friedrich-Schiller-Universität Jena, Germany and at the University of Texas at Austin, USA (Diplom 1998). He performed his graduate studies in the lab of Prof. Jonathan L. Sessler, obtaining his Ph.D. in 2002. From 2002 to 2005, he was an Ernst Schering Postdoctoral Fellow in the group of Prof. David A. Evans at Harvard University. He started his independent career at Rutgers University in 2005 and was promoted to associate professor in 2011 and full professor in 2014. In the summer of 2017, his research group moved to the University of Florida, where he is currently the Katritzky Term Professor in Heterocyclic Chemistry.
Zoom Image
Scheme 1 Borane-imine complexes
Zoom Image
Scheme 2 Different strategies to access cyclic imine-BF3 complexes
Zoom Image
Scheme 3 Selected scope for the reaction of imine-BF3 complexes with organometallic nucleophiles
Zoom Image
Scheme 4 Overview and concept
Zoom Image
Scheme 5 Selected scope of decarboxylative additions to imine-BF3 complexes
Zoom Image
Scheme 6 Selected scope for BCP-substituted azacycles
Zoom Image
Scheme 7 Plausible reaction mechanisms