Javad Amani1, Gary A Molander1. 1. Roy and Diana Vagelos Laboratories, Department of Chemistry, University of Pennsylvania , 231 South 34th Street, Philadelphia, Pennsylvania 19104-6323, United States.
Abstract
Visible light photoredox/nickel dual catalysis has been employed in the cross-coupling of acyl chlorides with potassium alkyltrifluoroborates. This protocol, based on single-electron-mediated alkyl transfer, circumvents the restriction of using reactive alkylmetallic nucleophiles in transition-metal-catalyzed acylation and achieves a mild and efficient method for the synthesis of unsymmetrical alkyl ketones. In this approach, a variety of acyl chlorides have been successfully coupled with structurally diverse potassium alkyltrifluoroborates, generating the corresponding ketones with good yields.
Visible light photoredox/nickel dual catalysis has been employed in the cross-coupling of acyl chlorides with potassium alkyltrifluoroborates. This protocol, based on single-electron-mediated alkyl transfer, circumvents the restriction of using reactive alkylmetallic nucleophiles in transition-metal-catalyzed acylation and achieves a mild and efficient method for the synthesis of unsymmetrical alkyl ketones. In this approach, a variety of acyl chlorides have been successfully coupled with structurally diverse potassium alkyltrifluoroborates, generating the corresponding ketones with good yields.
Ketones,
widespread in nature
and in pharmaceuticals,[1] are among the
most important and versatile functional groups in organic molecules
and have long been used as important intermediates for the synthesis
of complex structures. Because of their ubiquity, the development
of diverse synthetic methods to prepare these compounds has been the
subject of intense research.In principle, one of the most straightforward
methods for the synthesis
of alkyl ketones is the reaction of acyl halides with alkylmetallic
reagents.[2] However, the addition of highly
reactive alkylmetallic reagents (e.g., alkyllithium or Grignard reagents)
typically generates tertiary alcohols, and the enolization of the
produced ketones in these reactions cannot be repressed completely.
The use of low temperatures or less reactive nucleophiles such as
alkylcopper, alkylmanganese, alkylzinc, or alkylzirconium reagents
can be used to generate the ketones without undesired over-reaction,
but many of these reagents are prepared from alkyllithium or Grignard
reagents and thus are often minimally functionalized.[3] Furthermore, these reagents are not shelf-stable, and many
alkylzinc and alkylstannane reagents lack atom economy because only
one alkyl group on the metal is transferred.To overcome these
problems, significant progress has been made
in the development of transition-metal-catalyzed acylation of alkyl
nucleophiles using acyl halides.[4] In general,
however, metal-catalyzed ketone formation employing aliphatic nucleophiles
with carboxylic acid chlorides has been the least explored, perhaps
owing to the slow transmetalation of the alkylmetallic nucleophiles
and the susceptibility of the resulting transition metal intermediates
to undergo β-hydride elimination.[5]Recently, our group has developed an efficient photoredox/Ni
dual
catalysis paradigm for transition-metal-catalyzed cross-couplings
of alkylboron reagents with halide electrophiles based on a single-electron
transfer (SET) transmetalation pathway.[6] This synergistic catalysis approach via the simultaneous activation
and engagement of two separate coupling partners with two distinct
catalysts through low-barrier, open-shell processes, has had a profound
impact on C(sp3)–C(sp2) bond formations
under mild conditions. In this protocol, complications associated
with the transmetalation of alkylmetallic reagents in conventional
cross-coupling methods, which are mainly a consequence of the two-electron
nature of the transmetalation step, are avoided. The Doyle group has
employed a related strategy for acylation using anhydride electrophiles
(eq ).[7] This method was restricted to α-amino radical precursors,
and in fact acyl chlorides could not be used in the protocol developed.
Our group subsequently investigated the photoredox/Ni dual catalyzed
approach to α-alkoxy ketones using α-alkoxyalkyltrifluoroborates
as the radical precursor in conjunction with carboxylic acid chloride
electrophiles (eq 2).[8]To achieve an efficient protocol for the general synthesis
of aliphatic
ketones and in an effort to broaden the range of electrophiles to
which the photoredox/Ni dual catalysis activation mode could be applied,
we investigated the potential of this method for the cross-coupling
of acyl chlorides with potassium secondary alkyltrifluoroborates to
generate alkyl–alkyl and alkyl–aryl ketones (Figure ).
Figure 1
Photoredox/Ni dual catalysis
cross-coupling of acyl chlorides with
secondary R–BF3K.
Photoredox/Ni dual catalysis
cross-coupling of acyl chlorides with
secondary R–BF3K.To validate the strategy outlined in Figure through the conversion of acyl chlorides
to aliphatic ketones, the cross-coupling of hydrocinnamoyl chloride
with i-PrBF3K was evaluated as a model
reaction under a variety of conditions (e.g., varying solvents, Ni
catalysts, ligands, and additives). The investigation commenced with
the evaluation of an extensive array of Ni sources and ligands (L1–L8) by means of microscale high-throughput
experimentation (HTE, Figure ).[9] After addition of an internal
standard to the reaction mixtures in these screens, the results were
analyzed by UPLC. Then, to determine the relative amount of product
formed in each microscale reaction, the product to internal standard
(P/IS) ratios were calculated. The P/IS ratios of the first screen
are shown in Figure .
Figure 2
Graphical chart of P/IS ratios of the cross-coupling reactions
using different Ni sources and ligands. Reaction conditions: 0.01
mmol of hydrocinnamoyl chloride, 1.5 equiv of i-PrBF3K, 3 mol % Ir[dF(CF3)ppy]2(bpy)PF6, 6 mol % Ni source, 6 mol % ligand, 1.0 equiv of lutidine,
rt, blue LEDs, 24 h in THF (0.1 M).
Graphical chart of P/IS ratios of the cross-coupling reactions
using different Ni sources and ligands. Reaction conditions: 0.01
mmol of hydrocinnamoyl chloride, 1.5 equiv of i-PrBF3K, 3 mol % Ir[dF(CF3)ppy]2(bpy)PF6, 6 mol % Ni source, 6 mol % ligand, 1.0 equiv of lutidine,
rt, blue LEDs, 24 h in THF (0.1 M).In this screen, with hydrocinnamoyl chloride and i-PrBF3K (1.5 equiv) as the coupling partners,
using Ir[dF(CF3)ppy]2(bpy)PF6 (1) [dF(CF3)ppy = 2-(2,4-difluorophenyl)-5-(trifluoromethyl)pyridine,
bpy = 2,2′-bipyridine] as the photocatalyst, NiCl2·dme (6 mol %)/4,4′-di-tert-butyl-2,2′-dipyridyl, L1, (6 mol %) as the cross-coupling catalyst, and lutidine
(1 equiv) as a base in THF at room temperature provided the best yield.Control experiments revealed that all of the reaction parameters
(photocatalyst, Ni source, and ligand) are essential for efficient
coupling, in agreement with our mechanistic proposal. Additionally,
without added base, the product to internal standard ratio noticeably
decreased. Considering the important role of the base, which is ascribed
to sequestration of the BF3 generated upon oxidation of
the alkyltrifluoroborate, an extensive screening of various bases
in different solvents was carried out (Figure ). The investigations revealed that using
KF instead of lutidine and DME instead of THF provided a substantial
improvement and afforded the highest P/IS ratio.
Figure 3
Graphical chart of P/IS
ratios of the cross-coupling reactions
using different bases and solvents. Reaction conditions: 0.01 mmol
of hydrocinnamoyl chloride, 1.5 equiv of i-PrBF3K, 3 mol % Ir[dF(CF3)ppy]2(bpy)PF6, 6 mol % NiCl2·dme, 6 mol % L1, 1.0 equiv of base, rt, blue LEDs, 24 h (0.1 M).
Graphical chart of P/IS
ratios of the cross-coupling reactions
using different bases and solvents. Reaction conditions: 0.01 mmol
of hydrocinnamoyl chloride, 1.5 equiv of i-PrBF3K, 3 mol % Ir[dF(CF3)ppy]2(bpy)PF6, 6 mol % NiCl2·dme, 6 mol % L1, 1.0 equiv of base, rt, blue LEDs, 24 h (0.1 M).Further investigations demonstrated that lower
concentrations (0.05
M) increased the yield of the reaction, while higher molarities (0.2
M) resulted in diminished yields. To explore the generality of this
transformation, the developed conditions were applied to the cross-coupling
of hydrocinnamoyl chloride with a variety of structurally diverse,
secondary alkyltrifluoroborates. As demonstrated in Table , all the secondary alkyltrifluoroborates
function efficiently in this transition-metal-catalyzed acylation
protocol. Isopropyl- and sec-butyltrifluoroborates
were acylated in 78% and 77% isolated yields, respectively (entries
1 and 2). In these reactions only the desired regioisomers were obtained.
Although alkyltrifluoroborates possessing various ring sizes were
coupled in good yields (entries 3–5), the reaction with potassium
cyclopropyltrifluoroborate did not produce the desired product
owing to the high s character of the hybrid orbitals, which destabilizes
the incipient radical and engenders an unfavorable redox potential
on the trifluoroborate. The products derived from sterically hindered
structures such as 2-methylcyclopentyl and pinanyl substrates were
also isolated in good yields (entries 6 and 8). In contrast to the
reported procedures using Pd catalysis,[10] no isomerization was observed for these substrates, and trans-diastereomers were the major products formed in these
cross-coupling reactions. The reaction with aliphatic, heterocyclic
compounds containing oxygen or protected nitrogen atoms produced the
coupled products with moderate to good yields under this protocol
(entries 9–12), illustrating the versatility of this method.
Moreover, β-trifluoroboratoketones and -esters were
successfully coupled to generate the corresponding 1,4-diketones (entries
13–15), which are useful precursors to generate substituted
furans, pyrroles, and thiophenes via the Paal–Knorr synthesis.[11] The scalable nature of this coupling was demonstrated
by performing the reaction on 6.0 mmol scale (entry 3). In this reaction,
using half of the catalyst loadings (1.5 mol % of I and
3 mol % of NiCl2·dme/L1) afforded the
desired product in 69% yield.
Table 1
Scope of Secondary
R–BF3K in Cross-Coupling with Hydrocinnamoyl Chloride
Reaction completed
on 6.0 mmol scale
with 1.5 mol % Ir photocatalyst 1 and 3 mol % NiCl2·dme/L1.
Reaction completed
on 6.0 mmol scale
with 1.5 mol % Ir photocatalyst 1 and 3 mol % NiCl2·dme/L1.To investigate the applicability of this method to
other electrophiles,
potassium cyclohexyltrifluoroborate was coupled with various
acyl halides (Table ). We found that a broad range of aliphatic and aromatic acyl chlorides
participated in the cross-coupling reaction with great efficiency.
In these reactions, the desired products 2a–o were obtained with yields up to 88%. The reaction proceeded
well with cyclopropane- and cyclobutanecarbonyl chlorides, resulting
in 78% and 79% yields, respectively (entries 1 and 2). Both electron-rich
and electron-poor aromatic acyl chlorides were coupled using the optimal
conditions with 53–88% yields (entries 8–10). Remarkably,
the sterically hindered 1-adamantanecarbonyl chloride provided
the desired cross-coupled product with 60% yield (entry 6). The success
of the reaction with 4-cyanobenzoyl chloride (entry 9) and 1-chlorocarbonyl-1-methylethyl
acetate (entry 14) further demonstrates the functional group compatibility
of this method. Finally, the reaction with 4-morpholinecarbonyl chloride
to generate cyclohexyl(morpholino)methanone in 74% yield
(entry 15) shows the extension of this protocol to carbamoyl chloride
cross-coupling.
Table 2
Scope of the Cross-Coupling with Acyl
Chlorides
It is important to note that primary alkyltrifluoroborates
are not suitable substrates for these processes owing to an unfavorable
redox potential. Additionally, preliminary results indicate that a
different set of conditions are required for reaction of tertiary
alkyltrifluoroborates, and research along this line of investigation
continues.In conclusion, the first visible light photoredox
cross-coupling
of acyl chlorides with potassium secondary alkyltrifluoroborates
has been developed to access a variety of dialkyl- and alkyl-aryl
ketones in high yields. This efficient protocol establishes one of
the few catalytic methods to generate 2° alkyl ketones from acyl
chlorides. It extends photoredox/Ni dual catalytic cross-coupling
of secondary alkylboron reagents to catalytic acylation approaches.
Finally, the ability to employ bench-stable, storable nucleophiles,
the mild conditions, high functional group tolerance, and the operational
simplicity of this protocol lend themselves favorably to the facile
synthesis of unsymmetrical alkyl ketones.
Experimental
Section
General Considerations
All reactions were carried out
under an inert atmosphere of nitrogen or argon unless otherwise noted.
THF was dried over activated alumina. DME (99.5%, anhydrous) was used
as received. IrCl3·xH2O and NiCl2·dme were purchased from commercial sources.
All other reagents were purchased commercially and used as received.
Photoredox reactions were irradiated with two or three standard 26
W compact fluorescent light bulbs. Melting points (°C) are uncorrected.
NMR spectra were recorded on a 400 or 500 MHz spectrometer. Data are
presented as follows: chemical shift (ppm), multiplicity (s = singlet,
d = doublet, t = triplet, m = multiplet, br = broad), coupling constant J (Hz), and integration. Analytical thin-layer chromatography
(TLC) was performed on TLC silica gel plates (0.25 mm) precoated with
a fluorescent indicator. Visualization of the TLC plates was effected
with ultraviolet light. HRMS data were obtained by either ESI or CI
using a TOF mass spectrometer.
Synthesis of Potassium Secondary Alkyltrifluoroborates
Most of the potassium alkyltrifluoroborates were purchased
commercially. In cases where the desired potassium organotrifluoroborate
was not available, the corresponding boronic acid derivative was converted
to the trifluoroborate by the following procedure.
General Procedure
for Conversion of Boronic Acids to Trifluoroborates
To a
solution of boronic acid derivative in MeOH (0.1 M) at 0 °C
saturated aq KHF2 (4.5 M) was added dropwise over 30 min.
After completion of the reaction, followed by 11B NMR,
the resulting suspension was concentrated under reduced pressure.
H2O was removed by lyophilizer. The remaining solid was
suspended in hot acetone (3 × 100 mL) and filtered. The filtrate
was concentrated to a minimal volume (5–20 mL), and hexane
or Et2O (∼200 mL) was added to yield a white precipitate.
The precipitate was isolated by filtration, washing with hexanes (∼30
mL) and CH2Cl2 (∼30 mL), to afford the
desired secondary alkyltrifluoroborate.
Synthesis of Ir[dF(CF3)ppy]2(bpy)PF6 as the Photocatalyst 1
Photocatalyst 1 was synthesized according
to the literature procedure.[6]
High-Throughput
Experiments in the Design and Optimization of
the Photoredox Cross-Coupling Reaction of Hydrocinnamoyl Chloride
with Potassium Isoropyltrifluoroborate as Model Coupling
Partners
High Throughput Experimentation (HTE) was performed
at the Penn/Merck Center for High Throughput Experimentation at the
University of Pennsylvania. The screens were performed on a 10 μmol
scale. To reaction vials equipped with a Teflon coated magnetic stir
bar in a glovebox was added a solution of the Ni source and ligand
[1:1] dissolved in THF. The solvent was removed in vacuo under an inert atmosphere. Then a solution of a desired additive,
potassium isopropyltrifluoroborate, hydrocinnamoyl chloride,
and photocatalyst 1 in a desired solvent was added to
each vial. The vials were sealed and stirred over blue LED lights.
After 24 h the reactions were opened to air, 1 μmol of 4,4′-di-tert-butylbiphenyl (500 μL of a 0.002 μM solution
in MeCN) was added to each vial as an internal standard, and the reaction
mixtures were diluted with MeCN. The reaction mixtures were then analyzed
by UPLC to obtain product-to-internal standard (P/IS) ratios for different
conditions.
General Procedure for the Photoredox Cross-Coupling Reaction
of Acyl Chlorides with Potassium Alkyltrifluoroborates
To a long, thin glass vial (∼20 mL) equipped with a Teflon-coated
magnetic stir bar was added NiCl2·dme (6.6 mg, 0.03
mmol) and 4,4′-di-tert-butyl-2,2′-bipyridine L1 (8.1 mg, 0.03 mmol). The vial was sealed, evacuated under
vacuum, and purged with Ar three times. Anhydrous degassed THF (∼1
mL) was added by syringe under an inert atmosphere, and the resulting
mixture was stirred until it appeared as a pale green suspension.
Then, the solvent was removed under vacuum. Once dry, potassium alkyltrifluoroborate
(0.75 mmol, 1.5 equiv), Ir[dFCF3ppy]2(bpy)PF6 (15.1 mg, 0.015 mmol), and KF (29.1 mg, 0.5 mmol) were added.
Next, the vial was sealed, evacuated, and purged four times. Anhydrous
degassed DME (10 mL) was then added by syringe under an inert atmosphere
followed by the corresponding acyl halide (0.5 mmol). The resulting
mixture was stirred for 24 h in the presence of two 26 W fluorescent
light bulbs while a fan was blown across the reaction setup to suppress
the heat and maintain an ambient temperature of 24 °C. After
completion, the crude reaction mixture was filtered through a plug
of silica and rinsed with EtOAc (20 mL). The resulting solution was
concentrated, and the residue was purified by column chromatography
on silica gel, eluting with EtOAc and hexanes, to obtain products
in pure form.
Gram Scale Reaction
1-Cyclohexyl-3-phenylpropan-1-one
(1c)
To a long, thin-walled vacuum flask equipped
with a Teflon-coated
magnetic stir bar was added NiCl2·dme (39.5 mg, 0.18
mmol, 3.0 mol %) and 4,4′-di-tert-butyl-2,2′-bipyridine L1 (48.3 mg, 0.18, 3.0 mol %). Anhydrous degassed THF (∼5
mL) was added by syringe under Ar, and the resulting mixture was stirred
until it appeared as a pale green suspension. The solvent was then
removed under vacuum. Once dry, potassium cyclohexyltrifluoroborate
(1.71 g, 9.0 mmol, 1.5 equiv), Ir[dFCF3ppy]2(bpy)PF61 (90.8 mg, 0.09 mmol, 1.5
mol %), and anhydrous KF (348.6 mg, 6.0 mmol, 1.0 equiv) were added.
The vial was then capped with a rubber septum and purged and evacuated
four times. Under an inert atmosphere, DME (120 mL, 0.05 M) was introduced
followed by hydrocinnamoyl chloride (1.012 g, 6.0 mmol). The resulting
mixture was stirred vigorously for 48 h in the presence of three 26
W fluorescent light bulbs while a fan was blown across the reaction
setup to maintain an ambient temperature of 24 °C. After completion,
the crude reaction mixture was filtered through an approximately 4
cm × 2 cm cylindrical plug of silica, washing with EtOAc (60
mL). The resulting solution was concentrated, and the residue was
purified by column chromatography on silica gel, eluting with EtOAc
and hexanes, to obtain 895.5 mg (69% yield) of the product as a colorless
liquid.
Authors: Jennifer K Matsui; Álvaro Gutiérrez-Bonet; Madeline Rotella; Rauful Alam; Osvaldo Gutierrez; Gary A Molander Journal: Angew Chem Int Ed Engl Date: 2018-11-07 Impact factor: 15.336