[(DPEPhos)(bcp)Cu]PF6: A General and Broadly Applicable Copper-Based Photoredox Catalyst

Summary

Detailed and general protocols are presented for the synthesis of [(DPEPhos)(bcp)Cu]PF₆, a general copper-based photoredox catalyst, and for its use in synthetic chemistry for the direct arylation of C-H bonds in (hetero)arenes and radical cyclization of organic halides.

Abstract

Our group recently reported the use of [(DPEPhos)(bcp)Cu]PF₆ as a general copper-based photoredox catalyst which proved efficient to promote the activation of a broad variety of organic halides, including unactivated ones. These can then participate in various radical transformations such as reduction and cyclization reactions, as well as in the direct arylation of several (hetero)arenes. These transformations provide a straightforward access to a range of small molecules of interest in synthetic chemistry, as well as to biologically active natural products. Altogether, [(DPEPhos)(bcp)Cu]PF₆ acts as a convenient photoredox catalyst which appears to be an attractive, cheap and complementary alternative to the state-of-the-art iridium- and ruthenium-based photoredox catalysts. Here, we report a detailed protocol for the synthesis of [(DPEPhos)(bcp)Cu]PF₆, as well as NMR and spectroscopic characterizations, and we illustrate its use in synthetic chemistry for the direct arylation of (hetero)arenes and radical cyclization of organic halides. In particular, the direct arylation of N-methylpyrrole with 4-iodobenzonitrile to afford 4-(1-methyl-1H-pyrrol-2-yl)benzonitrile and the radical cyclization of N-benzoyl-N-[(2-iodoquinolin-3-yl)methyl]cyanamide to afford natural product luotonin A are detailed. The scope and limitations of this copper-based photoredox catalyst are also briefly discussed.

Introduction

Radical transformations have been known for decades to provide remarkably efficient pathways in synthetic chemistry which are often complementary to transformations based on cationic, anionic or pericyclic processes¹. While particularly promising for various types of transformations, radical-based chemistry has however long been underexploited, mainly because of the need for highly toxic reagents which considerably limits its attractiveness. Moreover, radical processes have long been considered as transformations associated with poor levels of control in terms of regio- and/or stereoselectivity, or leading to extensive dimerization and/or polymerization issues.

Alternative strategies have recently been developed in order to facilitate the generation and better control the reactivity of radical species. Among them, photoredox catalysis has become one of the most powerful methods as it allows the convenient generation of radical species using a light-responsive compound, namely the photoredox catalyst, and visible light irradiation^2,3. Visible light itself is indeed able to promote population of the excited state of the photoredox catalyst which becomes, consequently, both a stronger reductant and oxidant than in its corresponding ground state. These enhanced redox properties make single-electron transfer processes, not feasible in the ground state, possible under mild conditions from the excited state. Over the past decade, visible light photoredox catalysis has become an attractive and powerful technique in organic synthesis and has allowed the development of numerous remarkably efficient and selective transformations based on radical intermediates generated under sustainable, mild and user-friendly conditions.

While most photoredox processes reported to date are dominated by the use of iridium- and ruthenium-based photoredox catalysts, as well as by some organic dyes such as pyrylium and acridinium derivatives⁴, cheaper alternatives are still highly demanded for the development of complementary processes of interest for industrial applications. In this regard, the use of copper-based photoredox catalysts appears particularly appealing as they are not only cheaper but also provide opportunities to activate a broader and/or different range of substrates, which therefore opens new perspectives in photoredox catalysis^5,6,7,8. Despite some promising early works reported by the Kutal⁹, Mitani¹⁰ and Sauvage¹¹ groups, photoactivatable copper complexes have, however, only been scarcely used in photoredox catalysis, most probably because of their short-lived excited states compared to their ruthenium- and iridium-based congeners. More recently, recent remarkable contributions by Peters and Fu^12,13,14,15, Reiser^{16,17,18,19,20} and other groups^{21,22,23,24,25} have clearly brought attention back to copper-based photoredox catalysts and demonstrated their unique potential.

As part of our recent interest in copper-catalyzed radical processes^26,27, we recently reported a general and broadly applicable copper-based photoredox catalyst, [(DPEPhos)(bcp)Cu]PF₆ (DPEPhos: bis[(2-diphenylphosphino)phenyl] ether; bcp: bathocuproine), which turned out to be particularly efficient for the activation of organic halides under visible light irradiation (Figure 1A)^28,29,30. Upon irradiation with visible light and in the presence of an amine as sacrificial reductant, a wide range of unactivated aryl and alkyl halides was shown to be easily activated by catalytic amounts of [(DPEPhos)(bcp)Cu]PF₆ and therefore to participate in various radical transformations including reductions, cyclizations and direct arylation of several electron-rich (hetero)arenes. Furthermore, [(DPEPhos)(bcp)Cu]PF₆ has also proven successful at promoting photoinduced radical domino cyclizations of ynamides and cyanamides, providing an efficient and straightforward access to complex tri-, tetra- and pentacyclic nitrogen heterocycles at the core structures of various natural products. This strategy permitted the efficient synthesis of rosettacin, luotonin A, and deoxyvasicinone, natural products that exhibit anticancer, antimicrobial, anti-inflammatory and antidepressant activities. These transformations are illustrated in Figure 1C. From a mechanistic standpoint, the photoinduced activation of organic halides with [(DPEPhos)(bcp)Cu]PF₆ proceeds through a rare Cu(I)/Cu(I)*/Cu(0) catalytic cycle, which has been confirmed by extensive mechanistic and photophysical studies. In particular, excitation of the ground state [(DPEPhos)(bcp)Cu]PF₆ [Cu(I)] upon irradiation by visible light leads to the formation of the corresponding excited complex [(DPEPhos)(bcp)Cu]PF₆* [Cu(I)*] which is then reduced by the sacrificial amine to generate the corresponding [(DPEPhos)(bcp)Cu]PF₆ [Cu(0)] species. This Cu(0) intermediate is reductive enough to reduce the carbon–halogen bond of various organic halides to generate the corresponding radicals, which can then participate in the aforementioned transformations, together with regeneration of the starting catalyst (Figure 1B).

In the following section, we first describe the protocol to synthesize the photoactivatable [(DPEPhos)(bcp)Cu]PF₆ (whose NMR and spectroscopic characterizations are presented in the representative results section). The synthesis is straightforward and particularly convenient, and simply requires addition of 1 equivalent of DPEPhos and 1 equivalent of bcp to a solution of tetrakisacetonitrile copper(I) hexafluorophosphate in dichloromethane. The desired [(DPEPhos)(bcp)Cu]PF₆ is then isolated by precipitation from diethyl ether and can be easily obtained on a multigram scale (Figure 2A). Importantly, the isolated copper complex is not particularly sensitive to oxygen and moisture and can therefore be conveniently handled with no specific precautions other than being stored away from light.

Secondly, we describe the protocols to activate organic halides using [(DPEPhos)(bcp)Cu]PF₆ under visible light irradiation by focusing on two different transformations. The first reaction is the direct arylation of N-methylpyrrole with 4-iodobenzonitrile using catalytic amounts of [(DPEPhos)(bcp)Cu]PF₆ as photoredox catalyst, dicyclohexylisobutylamine as the sacrificial reductant and potassium carbonate as the base under irradiation at 420 nm (Figure 2B). The second reaction is the radical cyclization of N-benzoyl-N-[(2-iodoquinolin-3-yl)methyl]cyanamide, using the same catalyst and sacrificial reductant, whose cyclization directly leads to luotonin A, a natural product displaying interesting anticancer activities (Figure 2C). Detailed protocols are provided for both transformations.

Protocol

1. Synthesis of [(DPEPhos)(bcp)Cu]PF₆

1. Add 3.73 g (10.00 mmol) of tetrakisacetonitrile copper(I) hexafluorophosphate and 5.39 g (10.00 mmol) of DPEPhos to a 2 L round bottom flask equipped with a magnetic stir bar.
2. Fit the round bottom flask with a three neck vacuum adapter connected to a vacuum line and an argon line.
3. Evacuate the flask under vacuum and backfill with argon three times. Replace the three neck vacuum adapter by a rubber septum.
   NOTE: The reaction can be performed under air with slightly reduced efficiency (see discussion section for details).
4. Add 800 mL of dry dichloromethane.
   NOTE: Dichloromethane is freshly distilled from CaH₂. The reaction can also be performed in regular dichloromethane (99.8%) with similar efficiency (see discussion section for details).
5. Stir the reaction mixture for 2 h in the dark (reaction flask covered with aluminum foil) at room temperature (23-25 ºC) under an argon atmosphere.
6. Add 3.60 g (10.00 mmol) of bcp to a 500 mL round bottom flask equipped with a magnetic stir bar.
7. Fit the round bottom flask with a three neck vacuum adapter connected to a vacuum line and an argon line.
8. Evacuate the flask under vacuum and backfill with argon three times. Replace the three neck vacuum adapter by a rubber septum.
9. Add 200 mL of dry dichloromethane and gently stir the suspension until complete dissolution of the bcp.
10. Add the solution of bcp in dichloromethane to the reaction mixture using a cannula.
11. Stir for an additional hour in the dark (reaction flask covered with aluminum foil) at room temperature (23-25 ºC) under an argon atmosphere.
12. Filter the mixture through a pad of Celite, wash with ca. 100 mL of dichloromethane and concentrate the filtrate to ca. 50-100 mL under reduced pressure.
13. Add the concentrate dropwise to 1 L of diethyl ether using an addition funnel and with vigorous stirring to induce precipitation of the desired complex.
14. Collect the precipitate by filtration through a fritted glass (pore size 3) and wash the precipitate with ca. 100 mL of diethyl ether.
15. Dry the bright yellow precipitate under vacuum at room temperature (23-25 ºC) for 5 h to recover 10.1 g (91% yield) of the copper complex.
    NOTE: A 75% yield was obtained when the reaction was performed under air using distilled dichloromethane; a 89% yield was obtained when the reaction was performed under argon using regular dichloromethane; when performing the reaction under air with regular dichloromethane, the reaction was found to be less efficient and led to much lower purity (ca 70%, as estimated by ¹H NMR with an internal standard).
16. Characterize [(DPEPhos)(bcp)Cu]PF₆ as previously reported³¹.

2. Direct arylation of N-methylpyrrole with 4-iodobenzonitrile

1. Add 55 mg (0.05 mmol) of [(DPEPhos)(bcp)Cu]PF₆, 59 mg (0.25 mmol) of dicyclohexylisobutylamine (Cy₂NiBu), 138 mg (1.0 mmol) of K₂CO₃ and 114 mg (0.50 mmol) of 4-iodobenzonitrile to an oven-dried 10 mL vial.
2. Add a magnetic stir bar, seal the vial with a rubber septum, evacuate the vial under vacuum and backfill with argon three times.
3. Add 5 mL of acetonitrile and 890 μL (10.00 mmol) of N-methylpyrrole. Replace the rubber septum by a screw cap.
   NOTE: Acetonitrile is freshly distilled from CaH₂ and degassed using freeze-pump-thaw cycles prior to use to ensure high yields and reproducibility.
4. Stir the reaction mixture for 3 days at room temperature (23-25 ºC) in a photoreactor under 420 nm wavelength irradiation.
   NOTE: As an alternative to the use of a photoreactor, the reaction can also be conveniently performed using blue LEDs strips or a photochemistry device with a blue LED lamp (440 nm, 34 W). These experimental setups are shown in Figure 3 (see Representative results) and a discussion on their respective results is provided in the "Discussion" section.
5. Filter the reaction mixture through a pad of Celite, wash with ca. 5 mL of diethyl ether and concentrate the filtrate under reduced pressure.
6. Purify the crude residue by flash column chromatography over silica gel (eluent system petroleum ether/EtOAc: 90/10).
7. Dry the pure compound under vacuum at room temperature (23-25 ºC) for 3 h to recover 65 mg (72% yield) of the desired C2-arylated pyrrole.
   NOTE: The use of blue LEDs strips afforded a 76% yield (69 mg) while the use of a photochemistry device with a blue LED lamp (440 nm, 34 W) afforded an 86% yield (78 mg).
8. Characterize the pure compound as previously reported³².

3. Cyclization of N-benzoyl-N-[(2-iodoquinolin-3-yl)methyl]cyanamide to luotonin A

1. Add 37 mg (0.09 mmol) of N-benzoyl-N-[(2-iodoquinolin-3-yl)methyl]cyanamide, 9 mg (9.0 μmol) of [(DPEPhos)(bcp)Cu]PF₆, 11 mg (0.04 mmol) of Cy₂NiBu and 25 mg (0.18 mmol) of K₂CO₃ to an oven-dried 7 mL vial equipped with a magnetic stir bar.
2. Seal the vial with a rubber septum, evacuate the vial under vacuum and backfill with argon three times.
3. Add 2 mL of acetonitrile. Replace the rubber septum by a screw cap.
   NOTE: Acetonitrile is distilled from CaH₂ and degassed using freeze-pump-thaw cycles prior to use to ensure high yields and reproducibility.
4. Stir the reaction mixture for 5 days at room temperature (23-25 ºC) in a photoreactor under 420 nm wavelength irradiation.
5. Filter the reaction mixture through a pad of Celite, wash with ca. 2 mL of dichloromethane and concentrate the filtrate under reduced pressure.
6. Purify the crude residue by flash column chromatography over silica gel (eluent system: petroleum ether/EtOAc 60:40).
7. Dry the pure compound under vacuum at room temperature (23-25 ºC) for 3 h to recover 20 mg (79% yield) of the desired luotonin A.
8. Characterize the desired luotonin A as previously reported²⁹.

Results

Synthesis of [(DPEPhos)(bcp)Cu]PF₆
As shown by the protocol described in the above section, the synthesis of [(DPEPhos)(bcp)Cu]PF₆ is particularly convenient and can be easily performed on a multigram scale. The ¹H and ¹³C NMR spectra indicate formation of the pure complex (Figure 4A,B). The spectroscopic data correspond to those previously reported³¹.

^...

Discussion

Synthesis of [(DPEPhos)(bcp)Cu]PF₆
The synthesis of [(DPEPhos)(bcp)Cu]PF₆ is typically performed using dry dichloromethane (distilled prior to use) and under argon to ensure the highest yield, purity and good reproducibility. As mentioned in the protocol, the synthesis of [(DPEPhos)(bcp)Cu]PF₆ can be performed with regular dichloromethane (99.8%) and/or under air with variable efficiencies. Indeed, while the use of regular dichloromethane under argon afforded the sa...

Materials

Name	Company	Catalog Number	Comments
Material
Bathocuproine (bcp)	Acros	161340010
Acetonitrile, 99.9+	Acros	326811000
Celite 545	Acros	349670025
Bis[(2-diphenylphosphino)phenyl] ether (DPEphos)	Acros	383370050
Calcium hydride	Acros	C/1620/48
Dichloromethane, 99.8%	Fisher Chemical	D/1852/25
Dietyl ether, >= 99%	Fisher Chemical	D/2400/MS21
Ethyl acetate	Fisher Chemical	E/0900/25
N-Methylpyrrole, 99%	Sigma Aldrich	M78801
4-Iodobenzonitrile, 98%	Combi-Blocks	OR-3151
Petroleum ether (40-60 °)	Fisher Chemical	P/1760/25
Potassium carbonate, anhydrous	Fisher Chemical	P/4120/60
Tetrakisacetonitrile copper(I) hexafluorophosphate, 97%	Sigma Aldrich	346276
Equipment
1H and 13C NMR spectrometer	Bruker	Avance 300 Spectrometer
1H and 13C NMR spectrometer	Varian	VNMRS 400 Spectrometer
420 nm light tubes	Luzchem	LZC-420
Blue LEDs lamp	Kessil	H150-Blue
Blue LEDs strips	Eglo	92065
Photochemistry Device PhotoRedOx Box	Hepatochem	HCK1006-01-016
Photoreactor	Luzchem	CCP-4V
Spectrofluorimeter	Shimadzu	RF-5301PC
UV/Vis spectrometer	Perkin Elmer	Lambda 40