Photocatalytic and Chemoselective Transfer Hydrogenation of Diarylimines in Batch and Continuous Flow

Article abstract of DOI:10.1021/acs.orglett.7b03565

A visible-light photocalytic method for the chemoselective transfer hydrogenation of imines in batch and continuous flow is described. The reaction utilizes Et₃N as both hydrogen source and single-electron donor, enabling the selective reduction of imines derived from diarylketimines containing other reducible functional groups including nitriles, halides, esters, and ketones. The dual role of Et₃N was confirmed by fluorescence quenching measurements, transient absorption spectroscopy, and deuterium-labeling studies. Continuous-flow processing facilitates straightforward scale-up of the reaction.

Full text of DOI:10.1021/acs.orglett.7b03565

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Photocatalytic and Chemoselective Transfer Hydrogenation of
Diarylimines in Batch and Continuous Flow
Dean J. van As,† Timothy U. Connell,‡ Martin Brzozowski,†,§ Andrew D. Scully,§
and Anastasios Polyzos*,†,§
†School of Chemistry, The University of Melbourne, Parkville, VIC 3010, Australia
‡School of Science, RMIT University, Melbourne, VIC 3000, Australia
§CSIRO Manufacturing, Clayton, VIC 3068, Australia
S
* Supporting Information
ABSTRACT: A visible-light photocalytic method for the chemo-
selective transfer hydrogenation of imines in batch and continuous
flow is described. The reaction utilizes Et3N as both hydrogen source
and single-electron donor, enabling the selective reduction of imines
derived from diarylketimines containing other reducible functional
groups including nitriles, halides, esters, and ketones. The dual role of Et3N was confirmed by fluorescence quenching
measurements, transient absorption spectroscopy, and deuterium-labeling studies. Continuous-flow processing facilitates
straightforward scale-up of the reaction.
he ubiquity of aliphatic amines in nature has engendered
this structural motif with prominence throughout families
manifold. Such a transformation would putatively exhibit high
chemoselectivity for imines in the presence of functional groups
typically susceptible to reduction under traditional hydro-
genation conditions. Given the propensity of tertiary amines to
T
of biologically active molecules. The importance of this structural
class to human health is evidenced by a wealth of
pharmaceuticals, natural products, and agrochemicals exhibiting
a broad spectrum of bioactivities including antibiotic, anticancer,
and antiviral properties.1 The prevalence of amines in bio-
logically relevant small molecules necessitates the development
of new, chemoselective methods for their preparation.
undergo single-electron oxidation with mild oxidants (E1/2
=
+0.6−0.8 V),10 together with the ability of the aminium radical
cation to undergo facile hydrogen atom transfer (BDE = ∼42
kcal/mol),11 we postulated that simple amine additives could
serve the dual role of hydrogen source and single-electron
oxidant (Scheme 1). Herein, we describe the development of a
Photoredox catalysis has contributed to a renaissance in
radical-based methods in organic synthesis. Building upon
seminal studies in photocatalysis,2 Yoon, MacMillan, and
Stephenson demonstrated that visible light could be harnessed
to access a range of new catalytic transformations.3 These
pioneering studies have led to a host of novel amine α-C(sp3)−H
functionalization reactions based on the oxidative formation of
iminium ions and their subsequent trapping with (pro)-
nucleophiles.4 Alternatively, a reverse polarity reaction manifold
is also accessible by intercepting the transient α-amino radical
intermediate prior to iminium formation and coupling this
species with electron-deficient olefins.5 Examples of such
umpolung amine functionalization are less common in
comparison to the oxidative iminium approach due to the
propensity of the α-amino radical to undergo overoxidation to
the iminium ion. Within this context, single-electron reduction of
imines has been investigated as an alternative means of
generating α-amino radicals in a controlled fashion. Several
groups have recently disclosed that imine substrates can be
engaged in photoredox chemistry to affect dimerization,6
alkylation,7 allylation,8 and aminomethylation9 transformations.
Given the propensity for imines to undergo selective single-
electron transfer (SET) reduction, we hypothesized that a
visible-light-mediated, chemoselective imine transfer hydro-
genation could be achieved using a photoredox catalysis
Scheme 1. Mechanistic Hypothesis for Photoredox-Transfer
Hydrogenation Reaction
mild, chemoselective transfer hydrogenation reaction of diary-
limines under visible-light irradiation in batch and continuous
flow. The resulting benzhydrylamines are key structural motifs in
δ opioid receptor agonists,12 dopamine transporter ligands,13 and
histamine H1 antagonists.14
Preliminary optimization studies were conducted using
ketimine 1a and Et3N in DMF (Table 1) with photocatalyst
Received: November 16, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b03565
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a b
,
a
Table 1. Selected Optimization Experiments
Scheme 2. Evaluation of N-Aryl and N-Alkyl Substrates
c
entry
deviation from above conditions
yield (%)
1
2
3
4
5
6
7
8
9
[Ru(bpy)3]Cl2.6H2O
39
26
a
Reaction conditions: imine (0.45 mmol), photocatalyst (1.5 mol %),
fluorescein
b
Et3N (5 equiv), MeCN, 3 h, rt, 14 W blue LEDs. Reaction time of 5
min.
Eosin Y
25
Ir(ppy)3
25
[Ir{dF(CF3)ppy}2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6, 2 equiv Et3N
[Ir(ppy)2(dtbbpy)]PF6, 1 equiv Et3N
[Ir(ppy)2(dtbbpy)]PF6, MeCN, 3 h
52
91% yield with careful reaction monitoring. 1H NMR spectros-
copy studies revealed that imine 1h was converted quantitatively
to amine 2h within 5 min and then protodeiodination occurred
slowly over 100 min. Excellent chemoselectivity for imine
reduction was observed in the presence of ketone (2j) and ester
(2k) functional groups as well as nitrile group (2f), which
typically undergo reduction when subjected to transfer hydro-
genation or hydrosilylation conditions.21 N-Alkylimines were
also competent substrates, giving access to N-benzyl- or N-allyl-
protected amines (2l−m) as well as saturated N-alkylamine
products (2n).
97
79
39
>98
a
Reactions performed with imine (0.45 mmol), photocatalyst (1.5 mol
%), Et3N (5 equiv), DMF, 16 h, 14 W blue LEDs unless specified
b
otherwise. bpy = 2,2′-bipyridine; ppy = 2-phenylpyridine; dF(CF3)-
ppy = 2-(2,4-difluorophenyl)-5-(trifluoromethyl)pyridine; dtbbpy =
c
4,4′-di-tert-butyl-2,2′-bipyridine. Yields were determined by 1H NMR
analysis using 2,5-dimethylfuran as internal standard.
Further substrate evaluation focused on differing the
substitution pattern of the diaryl ketimine group (Scheme 3).
selection guided by analysis of the electrochemical reduction
potentials. Pleasingly, our design plan was promptly verified
when irradiation of the reaction components and archetypal
photocatalyst [Ru(bpy)3]Cl2 with 14 W blue LEDs for 16 h
afforded the desired product 2a in 39% yield (entry 1). Organic
dyes fluorescein and Eosin Y exhibit similar oxidizing potential
(E0 [*PCn/PCn−1] ≈ + 0.8 V vs SCE)15 relative to [Ru(bpy)3]Cl2
(E0 [*RuII/RuI] = +0.77 V vs SCE in MeCN)16 and should
undergo reductive quenching in the presence of Et3N (E0 = +0.66
V vs AgNO3 in MeCN). Both fluorescein and Eosin Y gave low
yields of 2a (entries 2 and 3), presumably due to diminished
reducing power from the PCn−1 state (−1.22 and −1.08 V,
respectively). The strongly reducing photocatalyst Ir(ppy)3 (E0
[IrII/IrIII] = −2.19 V)17 also gave a low yield of 25% (entry 4),
which is rationalized by the excited state species being a
comparatively weaker oxidant (E0 [*IrIII/IrII] = +0.31 V) than
[Ru(bpy)3]Cl2.
a
Scheme 3. Evaluation of Diarylketimine Groups
The strongly oxidizing and reducing heteroleptic iridium
complex [Ir{dF(CF3)ppy}2(dtbbpy)]PF6 furnished a 52% yield
of 2a (entry 5). Near-quantitative yields were achieved using the
analogous desfluorinated complex [Ir(ppy)2(dtbbpy)]PF6
(entry 6), which is also a potent reductant from the Ir(II) state
(E0 [IrII/IrIII] = −1.51 V).18 The reaction tolerated a range of
different amine donors, with n-Bu3N, i-Pr2NEt, and triethanol-
amine each facilitating effective reduction of 1a.19 Stoichiometry
investigations revealed that lowering the Et3N loading from 5 to
1 equiv resulted in a significant decrease in reaction efficiency
(entries 7 and 8). Polar, aprotic solvents were most effective, with
MeCN giving quantitative conversion by 1H NMR spectroscopic
analysis within 90 min.20 The optimal reaction conditions were
selected as 1.5 mol % [Ir(ppy)2(dtbbpy)]PF6 photocatalyst and 5
equiv of Et3N in MeCN with 3 h of visible light irradiation (entry
9). Control experiments showed that the reaction product was
not formed in the absence of photocatalyst, visible light, or Et3N.
With optimal reaction conditions established, we sought to
evaluate the scope of N-aryl and N-alkyl substrates amenable to
the transfer hydrogenation reaction (Scheme 2). The reaction
proceeded in high yield with both electron-rich (1a−e) and
electron-deficient (1f−k) N-aryl substrates. Unprotected phenol
groups (2e) were viable, as were chloro and trifluoromethyl
substituents (2g,i). Aryl iodide product 2h could be isolated in
a
Reaction conditions: imine (0.45 mmol), photocatalyst (1.5 mol %),
Et3N (5 equiv), MeCN, 3 h, rt, 14 W blue LEDs. PMP = p-
methoxyphenyl.
The transfer hydrogenation reaction was found to be insensitive
to electronic effects with both electron-rich and electron-
deficient systems proceeding in high yields. Notably, pyridine-
substituted ketimines (2s,t) were readily reduced, as were N-
PMP-protected imines (2p,r,t).
We next turned to flow chemistry as a means of enabling
effective scale-up and improving the efficiency of light
penetration of the liquid medium.22 The continuous flow setup
consisted of a 5 mL reactor coil irradiated in a custom built 14 W
blue LED photoreactor (Figure S1). A selection of diaryl
ketimines were submitted to transfer hydrogenation under
continuous-flow conditions (Scheme 4). The continuous-flow
process afforded the corresponding amines in comparable yields
to the batch methods with a significant reduction in reaction time
and inherent scalability offered by flow approaches. To
demonstrate the scalability of this protocol, 1 g of imine 1a
afforded 96% isolated yield of 2a without modification of the
B
DOI: 10.1021/acs.orglett.7b03565
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Continuous Flow Photoredox Transfer
Hydrogenation
a
Reaction conditions: imine (0.09 mmol), photocatalyst (1.5 mol %),
Et3N (5 equiv), MeCN, rt, 12 min, 14 W blue LEDs. b1a (3.89 mmol),
photocatalyst (1.5 mol %), Et3N (5 equiv), MeCN, rt, 12 min, 14 W
blue LEDs.
method, highlighting the robustness and practicality of the flow
process.
We performed mechanistic and control experiments to probe
the reaction mechanism. Deuterium labeling studies confirmed
that Et3N was indeed functioning in the dual role of hydrogen
atom source and single electron donor. When the reaction was
conducted under standard conditions with Et3N-d15, 75%
deuterium incorporation was observed at the amine α-position
(Scheme 5). No deuterium incorporation was observed with
Figure 1. Mechanistic elucidation experiments. A) Stern−Volmer plot
for quenching of [Ir(ppy)2(dtbbpy)]PF6 phosphorescence intensity by
■
●
triethylamine ( ) and imine 1a ( ). B) Transient absorption decay
profiles (λmeas = 380 nm) of [Ir(ppy)2(dtbbpy)]PF6 (0.1 mM, blue line)
and in the presence of triethylamine (50 mM, orange line). Solutions
were prepared using acetonitrile, and sparged with argon (6 min) before
measurement, λex = 355 nm (Figure S7).
Scheme 5. Deuterium Incorporation Studies
the reported excited state absorption spectrum of similar
heteroleptic iridium complexes.23 Consequently, the transient
absorbing species is assigned to a triplet metal-to-ligand charge-
transfer (3MLCT) excited state in which an electron is
transferred from iridium to the dtbbpy ancillary ligand. The
lifetime of this 3MLCT excited state, [IrIV(ppy)2(dtbbpy•−)]3,*
was determined to be 0.58 μs from the transient absorption decay
profile at 380 nm shown in Figure 1B. The decay profile
measured at 380 nm is very different when measured in the
presence of Et3N (50 mM); in this case, the decay is much
shorter, consistent with the steady-state phosphorescence
quenching measurements, together with an absorption signal
that persists for the time scale of this measurement.
MeCN-d3 as the reaction solvent. Radical-trapping experiments
with radical scavengers (TEMPO, BHT, 1,1-diphenylethylene)
did not afford the expected α-amino radical derived adducts
presumably due to steric hindrance at the diphenylmethylene
position.
Further measurements indicate that this absorption displays
no sign of decay even on a 100 μs time scale (Figure S7). This
result confirms the generation of a new chemically distinct
species upon excitation of the photocatalyst in the presence of
Et3N. Based on the persistence of the characteristic dtbbpy•‑
absorption at 380 nm, this new species is assigned as the reduced
photocatalyst [IrIII(ppy)2(dtbbpy•−)]0. These results provide
compelling evidence for a reductive quenching photocatalyst
cycle which is consistent with the proposed electron-transfer
mechanism. Further evidence for formation of the long-lived
reduced photocatalyst and its properties are the subject of
ongoing investigation.
In summary, a method for the facile and chemoselective
photoredox transfer hydrogenation of imines in batch and flow is
reported. With Et3N in the dual role of hydrogen source and
single-electron donor, high chemoselectivity for imines in the
presence of other reducible functional groups was achieved.
Comprehensive spectroscopic investigations into the reaction
Steady-state luminescence quenching experiments showed
that phosphorescence intensity from the photocatalyst
[IrIII(ppy)2(dtbbpy)]PF6 was quenched in the presence of
Et3N. The linearity of the Stern−Volmer plot (Figure 1A)
indicates the absence of static quenching, and that quenching
involves only a reaction between the emitting state of the
photocatalyst and Et3N. Negligible quenching of photocatalyst
phosphorescence was observed in the presence of imine 1a.
Transient absorption spectroscopy measurements were con-
ducted to elucidate the nature of the quenching reaction. The
differential absorption spectrum of [IrIII(ppy)2(dtbbpy)]PF6 in
solution displays strong transient absorption bands centered at
around 380, 480, and 800 nm (Figure S6) and closely resembles
C
DOI: 10.1021/acs.orglett.7b03565
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(14) Opalka, C. J.; D’Ambra, T. E.; Faccone, J. J.; Bodson, G.;
Cossement, E. Synthesis 1995, 1995, 766−768.
mechanism are currently underway and will be reported in due
course.
(15) Romero, N. A.; Nicewicz, D. A. Chem. Rev. 2016, 116, 10075−
10166.
ASSOCIATED CONTENT
* Supporting Information
■
(16) Juris, A.; Balzani, V.; Belser, P.; von Zelewsky, A. Helv. Chim. Acta
1981, 64, 2175−2181.
(17) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti, F.
Top. Curr. Chem. 2007, 281, 143−203.
(18) Lowry, M. S.; Goldsmith, J. I.; Slinker, J. D.; Rohl, R.; Pascal, R. A.;
Malliaras, G. G.; Bernhard, S. Chem. Mater. 2005, 17, 5712−5719.
(19) The reduction potentials of these tertiary amines are favourable
for exergonic electron transfer to the *[Ir] species. For tributylamine, E0
= + 0.62 V vs AgNO3 in MeCN. See: ref 9. For N,N-diisopropylethyl-
amine, E0 = + 0.65 V vs SCE in MeCN; see: Chatterjee, T.; Iqbal, T.;
You, Y.; Cho, E. J. Acc. Chem. Res. 2016, 49, 2284−2294. For
triethanolamine, E0 = + 0.52 V vs SCE in MeCN. Value vs SCE
calculated from literature value reported vs normal hydrogen electrode
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b03565.
Experimental procedures, mechanistic studies, compound
characterization data, and NMR spectra for all compounds
(PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: anastasios.polyzos@unimelb.edu.au.
ORCID
(NHE); see: Kalyanasundaram, K.; kiwi, J.; Gratzel, M. Helv. Chim. Acta
̈
1978, 61, 2720−2730.
(20) See the Supporting Information for further details.
(21) For reviews of transfer hydrogenation see: (a) Brieger, G.;
Nestrick, T. Chem. Rev. 1974, 74, 567−580. (b) Wang, D.; Astruc, D.
Chem. Rev. 2015, 115, 6621−6686 For examples of nitrile and ketone
hydrosilylation, see:. (c) Gutsulyak, D. V.; Nikonov, G. I. Angew. Chem.,
Anastasios Polyzos: 0000-0003-1063-4990
Notes
The authors declare no competing financial interest.
́
Int. Ed. 2010, 49, 7553−7556. (d) Perez, M.; Qu, Z.-W.; Caputo, C. B.;
ACKNOWLEDGMENTS
■
Podgorny, V.; Hounjet, L. J.; Hansen, A.; Dobrovetsky, R.; Grimme, S.;
Stephan, D. W. Chem. - Eur. J. 2015, 21, 6491−6500.
(22) Tucker, J. W.; Zhang, Y.; Jamison, T. F.; Stephenson, C. R. J.
Angew. Chem., Int. Ed. 2012, 51, 4144−4147.
D.J.v.A. acknowledges the University of Melbourne, Melbourne
Research Scholarship (MRS). A.P. acknowledges the University
of Melbourne and CSIRO for the joint Establishment Grant.
(23) Ichimura, K.; Kobayashi, T.; King, K. A.; Watts, R. J. J. Phys. Chem.
1987, 91, 6104−6106.
REFERENCES
■
(1) Vitaku, E.; Smith, D. T.; Njardarson, J. T. J. Med. Chem. 2014, 57,
10257−10274.
(2) Hedstrand, D. M.; Kruizinga, W. M.; Kellogg, R. M. Tetrahedron
Lett. 1978, 19, 1255−1258.
(3) (a) Ischay, M. A.; Anzovino, M. E.; Du, J.; Yoon, T. P. J. Am. Chem.
Soc. 2008, 130, 12886−12887. (b) Nicewicz, D. A.; MacMillan, D. W. C.
Science 2008, 322, 77−80. (c) Narayanam, J. M. R.; Tucker, J. W.;
Stephenson, C. R. J. J. Am. Chem. Soc. 2009, 131, 8756−8757.
́ ́
(4) (a) Condie, A. G.; Gonzalez-Gomez, J. C.; Stephenson, C. R. J. J.
Am. Chem. Soc. 2010, 132, 1464−1465. (b) Rueping, M.; Zhu, S.;
Koenigs, R. M. Chem. Commun. 2011, 47, 12709−12711. (c) DiRocco,
D. A.; Rovis, T. J. Am. Chem. Soc. 2012, 134, 8094−8097.
(5) Kohls, P.; Jadhav, D.; Pandey, G.; Reiser, O. Org. Lett. 2012, 14,
672−675.
(6) Nakajima, M.; Fava, E.; Loescher, S.; Jiang, Z.; Rueping, M. Angew.
Chem., Int. Ed. 2015, 54, 8828−8832.
(7) (a) Hager, M.; MacMillan, D. W. C. J. Am. Chem. Soc. 2014, 136,
́
16986−16989. (b) Jeffrey, J.; Petronijevic, F. R.; MacMillan, D. W. C. J.
Am. Chem. Soc. 2015, 137, 8404−8407. (c) Patel, N. R.; Kelly, C. B.;
Siegenfeld, A. P.; Molander, G. A. ACS Catal. 2017, 7, 1766−1770.
(8) (a) Qi, L.; Chen, Y. Angew. Chem., Int. Ed. 2016, 55, 13312−13315.
(b) Fuentes de Arriba, A. L.; Urbitsch, F.; Dixon, D. Chem. Commun.
2016, 52, 14434−14437.
(9) (a) Uraguchi, D.; Kinoshita, N.; Kizu, T.; Ooi, T. J. Am. Chem. Soc.
2015, 137, 13768−13771. (b) Fava, E.; Millet, A.; Nakajima, M.;
Loescher, S.; Rueping, M. Angew. Chem., Int. Ed. 2016, 55, 6776−6779.
(10) Chow, Y. L.; Danen, W. C.; Nelsen, S. F.; Rosenblatt, D. H. Chem.
Rev. 1978, 78, 243−274.
(11) Beatty, J. W.; Stephenson, C. R. J. Acc. Chem. Res. 2015, 48, 1474−
1484.
(12) (a) Barn, D. R.; Bom, A.; Cottney, J.; Caulfield, W. L.; Morphy, J.
R. Bioorg. Med. Chem. Lett. 1999, 9, 1329−1334. (b) Delorme, D.;
Berthelette, C.; Lavoie, R.; Roberts, E. Tetrahedron: Asymmetry 1998, 9,
3963−3966.
(13) (a) Dutta, A. K.; Xu, C.; Reith, M. E. A. J. Med. Chem. 1998, 41,
3293−3297. (b) Zhang, Y.; Joseph, D. B.; Bowen, W. D.; Flippen-
Anderson, J. L.; Dersch, C. M.; Rothman, R. B.; Jacobson, A. E.; Rice, K.
C. J. Med. Chem. 2001, 44, 3937−3945.
D
DOI: 10.1021/acs.orglett.7b03565
Org. Lett. XXXX, XXX, XXX−XXX