A Photoredox Coupling Reaction of Benzylboronic Esters and Carbonyl Compounds in Batch and Flow

Article abstract of DOI:10.1021/acs.orglett.9b02307

Mild cross-coupling reaction between benzylboronic esters with carbonyl compounds and some imines was achieved under visible-light-induced iridium-catalyzed photoredox conditions. Functional group tolerance was demonstrated by 51 examples, including 13 he

Full text of DOI:10.1021/acs.orglett.9b02307

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
A Photoredox Coupling Reaction of Benzylboronic Esters and
Carbonyl Compounds in Batch and Flow
Yiding Chen,† Oliver May,† David C. Blakemore,‡ and Steven V. Ley*,†
†Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, United Kingdom
‡Medicine Design, Pfizer, Inc., Eastern Point Road, Groton, Connecticut 06340, United States
S
* Supporting Information
ABSTRACT: Mild cross-coupling reaction between benzylboronic esters with carbonyl compounds and some imines was
achieved under visible-light-induced iridium-catalyzed photoredox conditions. Functional group tolerance was demonstrated by
51 examples, including 13 heterocyclic compounds. Gram-scale reaction was realized through the use of computer-controlled
continuous flow photoreactors.
Table 1. Reaction Optimization
arbonyl compounds serve as fundamentally important
Celectrophilic building blocks during carbon−carbon
bond-forming reactions commonly using transition-metal
catalysis1 or organometallic coupling processes.2 While wide
functional group tolerance can be achieved in these reactions,3
there is interest in moving away from using strongly basic
conditions or highly reactive substrates.4 In this content, the
current trend in the use of visible light photoredox methods
has led to impressive discoveries, especially to bring about
carbon−carbon bond formation with single electron-transfer
processes.5
Previously we have noted the addition of Lewis bases to
activate photoredox-mediated coupling reactions between
normally unreactive benzylic boronic esters and olefinic
acceptors.6 Others have also reported useful processes whereby
selective reductive addition to carbonyl compounds can be
achieved.7 More specifically, Rueping described the dimeriza-
tion of aldehydes and ketones via intermediate photoredox-
generated ketyl radicals.8 Several other reports9 concerning
reactions with alkenes,10 intramolecular cyclization,11 or
photocatalytic Barbier-type reactions12 via putative radical−
radical coupling pathways are known as well. In view of the
opportunities that arise through these photoredox coupling
processes, we commenced the study below employing benzylic
boron esters as precursors during this potential direct coupling
with carbonyl compounds and imines, thus avoiding strongly
nucleophilic organometallic reagents.
a
b
entry
variation from above conditions
yield (%)
c
1
2
3
4
5
6
7
8
9
no variation
86 (84)
24
2 mol % Mes-Acr-BF4 (ii) instead of Ir
2 mol % [Ru(bpy)3]Cl2 (iii) instead of Ir
50 mol % DMAP as base
additional 20 mol % Sc(OTf)2
acetonitrile (0.1 M) as solvent
0.1 M concentration instead of 0.05 M
without base
without photocatalyst
without light
green LEDs (540 nm, 14 W)
0
44
70
19
54
43
0
10
11
0
16
a
Reaction conditions: 0.1 mmol boronic acid pinacol ester, 0.15
mmol aldehyde, 1 mL acetone, and 1 mL methanol. NMR yield
using 1,3,5-trimethoxybenzene as an internal standard. Isolated yield.
b
c
From a test reaction based on the conditions we had
explored previously,6 we were pleased to observe an optimal
yield of 84% for the specific cross-coupling event of the p-
methoxybenzylboronic acid pinacol ester 1a with p-chlor-
obenzaldehyde 2a. This was achieved by employing iridium
photoredox catalyst i with 3-quinuclidinol in a 0.05 M 1:1
methanol and acetone solvent system under a blue LED light
for 8 h to give the addition product 3a (Table 1, entry 1). For
full details of the optimization and screening, see the
Supporting Information. Although these conditions bear a
similarity to recent procedures,6 the reaction progressed rather
differently over the optimization process. Commonly used
Received: July 4, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b02307
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
photocatalysts mesityl acridinium salt ii and the ruthenium
catalyst iii failed to deliver coupled products (entries 2 and 3).
On the other hand, using 4-dimethylaminopyridine (DMAP)
as additive, a moderate 44% of product 3a was realized (entry
4), while addition of the Lewis acid Sc(OTf)3 together with 20
mol % equivalent of the quinuclidine base did not significantly
enhance the overall reactivity (entry 5). Solvent changes were
usually detrimental; for example, acetonitrile leads to a yield of
only 19% of the coupled product (entry 6). However, it is
worth noting that the concentration of the reaction mixture
does have a noticeable impact, in that a 0.1 M reaction mixture
caused a drop in yield to 54% (entry 7). More difficult to
understand in this specific coupling example was the 43% yield
of product 3a when no base was added to the mixture (entry
8), which is discussed later. Without either photocatalyst or a
light source, the reaction was completely shut down (entries 9
and 10). Switching to a green light led to a decreased yield of
16% (entry 11). The reaction was accompanied by minor
byproducts which were removed on workup by normal silica
gel chromatography.
With the conditions from entry 1 in hand, we directed our
attention to examine a series of benzyl boronic esters 1 and
coupling partners 2 (Figure 1). Generally, electron-donating
Figure 2. Scope of aryl and heteroaryl aldehydes and imines.
substrates. Moderate yields were obtained from pyridyl
substrates (3aa−3ad). The nitrogen-containing heterocycles
pyrrole, pyrimidine, and quinoline (3ae−3ag) as well as sulfur-
containing heterocyclic compounds such as thiazole (3ah) and
thiophene (3ai and 3aj), all reacted satisfactorily. For
completeness, listed in red are a few aldehyde coupling
partners that failed to yield useful products.
We further expanded the scope of the reactions to include
ketone derivatives (Figure 3). Acetophenone derivatives (4a−
4g) were tolerated, while better yields were observed with
benzophenone substrates (5h−5j). Photochemically active
reagents such as xanthenone (5k), thioxanthenone (5l), and
fluorenone (5m) were particularly reactive, which supports our
later mechanistic proposals. Acetohexamide (4n)13 successfully
delivered the coupled product in 59% yield (5n). In this
example, a reversed ratio of 1.5:1 boronic ester and ketone was
employed to aid with the isolation of the benzylated product
from acetohexamide starting material. In cases where there is
significant enolization in the coupling partner (shown in red),
very poor product yields are realized.
Figure 1. Scope of boronic acid pinacol esters.
group (EDG) substituents on the boronic ester component
worked well, as exemplified by high yields from the optimized
4-methoxyphenyl substrate (1a) as well as p-methylbenzyl
(1b), o-methylbenzyl (1c), and 1-nathphylmethyl (1d)
boronic esters. The parent benzyl boronic ester (1e) also
provided a moderate yield of coupled product (3e), although
electron-withdrawing groups (EWGs) such as the p-trifluor-
omethoxybenzyl derivative (1f), p-chlorobenzyl (1g), m-
fluorobenzyl (1h), and p-methylbenzoate (1i) all showed
lower efficiencies. Likewise, for the phenylsulfanylmethyl
boronic ester (1j), only 16% of the corresponding coupled
product (3j) was obtained. Similarly, with other long-chain
aliphatic boronic esters and aryl boronic esters, no useful
products were observed.
To further investigate the scope with respect to aldehydes 2
(Figure 2), we examined a wide range of aromatic aldehydes
with various ring substituents (2k−z). Among these, a number
contain challenging features, such as bromo- (2l), boc-
protected amino- (2m), photosensitive nitrile (2n), boron
pinacol ester (2o), hydroxyl (2p), and pentafluoro (2y)
groups. Also interestingly, given a choice between a ketone and
aldehyde as coupling partners, we saw complete aldehyde
specificity (3s). In three examples, addition to imines occurs
readily to afford coupled products (3v−x). Particularly worth
noting is that the method is compatible with many heterocyclic
The use of continuous methods (Scheme 1)14 has been
shown to offer immediate improvements over equivalent batch
photoredox procedures15 when scaling reactions, particularly
when new reactor designs lead to improved photon flux
densities16 and material throughput.17
Reaction mixtures were prepared in a normal fashion and
pumped through a photoreactor (Vapourtec UV-150) fitted
with a blue LED module (17 W at λ = 420 nm) and a 10 mL
coil (FEP, 1/16 in.). With a residence time of 100 min, 100%
B
DOI: 10.1021/acs.orglett.9b02307
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
whereas in this new work, DMAP did not promote the reaction
to any equivalent level (Table 1, entry 4). These distinctive
differences, though not direct evidence, suggested a mecha-
nism that differed from a Lewis base-catalyzed photoredox
activation of the boron species.6
In other control experiments (Table 1, entries 9 and 10, see
more in Supporting Information) suggested the iridium
complex i was the only functional photoredox catalyst, while
no conclusion can be drawn about the absolute necessity of
quinuclidinol additive (Table 1, entry 8). This observation
contrasts with previous mechanistic proposals,18 since the
reduction potential of most carbonyl compounds presented are
beyond the range of this particular photocatalyst.19 For
red
1/2
example, the reductive potential of benzophenone is (E
=
−1.87 V vs SCE), whereas the reductive potential of the
catalyst i is (Er1e/d2 = −1.37 V vs SCE).20 Hence a further control
experiment was performed (Scheme 2a). Using 4-methylbenzyl
a
Scheme 2. Additional Control Experiments
Figure 3. Scope of ketones.
conversion was obtained giving a 70% 3a isolated yield
(Scheme 1a). Encouraged by this early success, a larger scale
experiment was carried out on the recently commercialized
Uniqsis Photosyn reactor, which was equipped with a more
powerful (420 W at λ = 450 nm) lamp and a bigger 20 mL
(PFA, 1/16 in.) flow coil, which could further reduce the
residence time to 60 min. The reaction was controlled with the
LabVIEW program and was monitored with Uniqsis Flow-UV.
A 10 h run was carried out to produce 1.69 g of product (3ak).
(See Supporting Information for detailed reaction setup; no
attempt at this stage was made to scale up the reaction any
further.)
As mentioned above, this reaction was performed under
similar catalytic conditions to our previous work.6 However,
two main differences in reactivity were noticed. First, while
aromatic boronic acid pinacol esters and aliphatic boronic
acids were viable precursors in our earlier study,14 only
benzylic Bpin esters were successful in the current work.
Additionally, previous studies clearly demonstrated that other
Lewis bases (e.g., DMAP) could be as efficient as quinuclidine,
a
(a) Coupling of p-methylbenzylboronic ester with p-chlorobenzalde-
hyde in the absence of quinuclidine. (b) Competition reaction
between benzaldehyde and ethyl acrylate with benzylboronic ester.
(c) Radical trap experiment.
Scheme 1. Continuous Flow Synthesis with (a) Vapourtec UV-150 and (b) Uniqsis Photosyn controlled by LabVIEW
C
DOI: 10.1021/acs.orglett.9b02307
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
boronic pinacol ester 1b as a precursor, only 6% of desired
product 3b was detected in the absence of a base additive. This
contrasts with a 77% yield when a 20 mol % 3-quinuclidinol
was added (Figure 1, 3b), indicating that the quinuclidine base
plays a crucial role. These results might suggest that the
then oxidized by radical cation A, which terminates the radical
sequence and gives borate D and the quinuclidine catalyst.
Finally, D is hydrolyzed to the targeted alcohol product.
In summary, we describe above a cross-coupling reaction of
benzylic organoboron reagents with carbonyl compounds via
photoredox catalysis, displaying a wide functional group
tolerance. Preliminary results also show the reactions could
be scaled using more powerful light sources coupled to
computer-controlled flow chemistry devices.
methoxyboronic ester could initiate reaction activation, where
21
̈
a similar observation was reported by Konig et al.
Competition reactions between chlorobenzaldehyde and
ethyl acrylate as coupling partners were also briefly investigated
(Scheme 2b). In the presence of quinuclidine base, 61% of the
coupled boronic ester 7 was realized, and only 8% 3a was
isolated. Lastly, the product yield dropped dramatically with
the addition of radical scavenger TEMPO (Scheme 2c), further
proving the existence of a radical pathway. Stern−Volmer
experiments showed 3-quinuclidinol was able to quench the
photocatalyst under reaction conditions, while Bpin ester and
aldehyde presented no such effect (see Supporting Informa-
tion).22
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b02307.
Details of experimental procedure, compound character-
ization data and NMR spectra (PDF)
The reaction analysis may suggest a ketyl radical pathway
could also be operating. In most cases, a byproduct formed in
the reaction which was thought to be a diol from aldehyde/
ketone dimerization, as expected from a ketyl radical
mechanism. To the best of our knowledge, although radical
trapping with carbonyl compounds is known,7e,23 the
generation of a thermodynamically unfavorable alkoxy radical
often lead to a reversable process24 or homolytic C−C β-
scission.25
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: svl1000@cam.ac.uk.
ORCID
Yiding Chen: 0000-0001-6932-9005
Steven V. Ley: 0000-0002-7816-0042
Notes
We therefore propose the mechanism shown in Figure 4.
Initially, the Ir(III) catalyst i, when exposed to blue light, is
The authors declare the following competing financial
interest(s): D.C.B. is an employee and stockholder of Pfizer,
Inc.
ACKNOWLEDGMENTS
■
We thank Uniqsis Ltd and Mark Ladlow for the generous loan
of a Photosyn reactor. Y.C. thanks Pfizer for funding the
postdoctoral fellowship. The authors also gratefully acknowl-
edge financial support from H2020-FETOPEN-2016-2017
program of European commission (S.V.L.; grant agreement
no.: 737266-ONE FLOW).
REFERENCES
■
(1) (a) Kuriyama, M.; Shimazawa, R.; Shirai, R. J. Org. Chem. 2008,
73, 1597−1600. (b) Gois, P. M. P.; Trindade, A. F.; Veiros, L. F.;
́
Andre, V.; Duarte, M. T.; Afonso, C. A. M.; Caddick, S.; Cloke, F. G.
N. Angew. Chem., Int. Ed. 2007, 46, 5750−5753.
(2) Hollerbach, M. R.; Barker, T. J. Organometallics 2018, 37, 1425−
1427.
(3) Sakai, M.; Ueda, M.; Miyaura, N. Angew. Chem., Int. Ed. 1998,
37, 3279−3281.
Figure 4. The proposed mechanism.
(4) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207−2293.
(5) (a) Campos, K. R.; Coleman, P. J.; Alvarez, J. C.; Dreher, S. D.;
Garbaccio, R. M.; Terrett, N. K.; Tillyer, R. D.; Truppo, M. D.;
Parmee, E. R. Science 2019, 363, No. eaat0805. (b) Banerjee, A.; Lei,
Z.; Ngai, M.-Y. Synthesis 2019, 51, 303−333. (c) Shu, C.; Noble, A.;
Aggarwal, V. K. Angew. Chem., Int. Ed. 2019, 58, 3870−3874.
(d) Shaw, M. H.; Shurtleff, V. W.; Terrett, J. A.; Cuthbertson, J. D.;
MacMillan, D. W. C. Science 2016, 352, 1304−1308. (e) Le, C.;
Liang, Y.; Evans, R. W.; Li, X.; MacMillan, D. W. C. Nature 2017,
547, 79.
excited to a long-lifetime excited-state Ir(III)*.20 The
quinuclidine base (E = +1.00 V vs SCE)26 then undergoes
ox
1/2
a single electron oxidation by a photoexcited Ir(III)* catalyst
(E = +1.21 V vs SCE),20 leading to the initiation of radical
ox
1/2
cation A and the reduced Ir(II) complex.27 The aldehyde/
ketone activated by boronic ester21 is then reduced by a Ir(II)
red
1/2
complex (E
= −1.37 V vs SCE) via a single electron
reduction, thus completing the photoredox cycle. The resulting
ketyl radical anion B subsequently attacks the benzylboronic
acid pinacol ester to assemble the borate radical anion C.
Alternatively, the aldehyde/ketone could be activated by a
boronic ester through a boron−oxygen interaction, thus the
single electron reduction creates radical anion C directly. C is
(6) Lima, F.; Sharma, U. K.; Grunenberg, L.; Saha, D.; Johannsen,
S.; Sedelmeier, J.; Van der Eycken, E. V.; Ley, S. V. Angew. Chem., Int.
Ed. 2017, 56, 15136−15140.
̈
̈
(7) (a) Schwarz, J. L.; Schafers, F.; Tlahuext-Aca, A.; Luckemeier, L.;
Glorius, F. J. Am. Chem. Soc. 2018, 140, 12705−12709. (b) Capaldo,
L.; Riccardi, R.; Ravelli, D.; Fagnoni, M. ACS Catal. 2018, 8, 304−
D
DOI: 10.1021/acs.orglett.9b02307
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
309. (c) Vu, M. D.; Das, M.; Liu, X.-W. Chem. - Eur. J. 2017, 23,
́
́
15899−15902. (d) Matos, J. L. M.; Vasquez-Cespedes, S.; Gu, J.;
Oguma, T.; Shenvi, R. A. J. Am. Chem. Soc. 2018, 140, 16976−16981.
(e) Pitzer, L.; Sandfort, F.; Strieth-Kalthoff, F.; Glorius, F. J. Am.
Chem. Soc. 2017, 139, 13652−13655.
(8) Nakajima, M.; Fava, E.; Loescher, S.; Jiang, Z.; Rueping, M.
Angew. Chem., Int. Ed. 2015, 54, 8828−8832.
(9) Wang, R.; Ma, M.; Gong, X.; Fan, X.; Walsh, P. J. Org. Lett.
2019, 21, 27−31.
(10) (a) Fava, E.; Nakajima, M.; Nguyen, A. L. P.; Rueping, M. J.
Org. Chem. 2016, 81, 6959−6964. (b) Lee, K. N.; Lei, Z.; Ngai, M.-Y.
J. Am. Chem. Soc. 2017, 139, 5003−5006. (c) Qi, L.; Chen, Y. Angew.
Chem. 2016, 128, 13506−13509.
(11) Li, W.; Duan, Y.; Zhang, M.; Cheng, J.; Zhu, C. Chem.
Commun. 2016, 52, 7596−7599.
̈
(12) Berger, A. L.; Donabauer, K.; Konig, B. Chem. Sci. 2018, 9,
7230−7235.
(13) Field, J. B.; Ohta, M.; Boyle, C.; Remer, A. N. Engl. J. Med.
1967, 277, 889−894.
(14) Lima, F.; Grunenberg, L.; Rahman, H. B. A.; Labes, R.;
Sedelmeier, J.; Ley, S. V. Chem. Commun. 2018, 54, 5606−5609.
́
(15) Cambie, D.; Bottecchia, C.; Straathof, N. J. W.; Hessel, V.;
̈
Noel, T. Chem. Rev. 2016, 116, 10276−10341.
(16) Roibu, A.; Fransen, S.; Leblebici, M. E.; Meir, G.; Van Gerven,
T.; Kuhn, S. Sci. Rep. 2018, 8, 5421.
̀
̈
(17) (a) Loubiere, K.; Oelgemoller, M.; Aillet, T.; Dechy-Cabaret,
O.; Prat, L. Chem. Eng. Process. 2016, 104, 120−132. (b) Su, Y.;
̈
Straathof, N. J. W.; Hessel, V.; Noel, T. Chem. - Eur. J. 2014, 20,
10562−10589.
(18) Xia, Q.; Dong, J.; Song, H.; Wang, Q. Chem. - Eur. J. 2018, 25,
2949−2961.
(19) Wagner, P. J.; Truman, R. J.; Puchalski, A. E.; Wake, R. J. Am.
Chem. Soc. 1986, 108, 7727−7738.
(20) 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.
̈
(21) Wang, S.; Lokesh, N.; Hioe, J.; Gschwind, R. M.; Konig, B.
Chem. Sci. 2019, 10, 4580−4587.
(22) (a) Proctor, R. S. J.; Davis, H. J.; Phipps, R. J. Science 2018, 360,
419−422. (b) Ashley, M. A.; Yamauchi, C.; Chu, J. C. K.; Otsuka, S.;
Yorimitsu, H.; Rovis, T. Angew. Chem., Int. Ed. 2019, 58, 4002−4006.
(c) Tian, H.; Xia, Q.; Wang, Q.; Dong, J.; Liu, Y.; Wang, Q. Org. Lett.
2019, 21, 4585−4589.
(23) (a) Dowd, P.; Zhang, W. Chem. Rev. 1993, 93, 2091−2115.
(b) Saladrigas, M.; Bosch, C.; Saborit, G. V.; Bonjoch, J.; Bradshaw, B.
Angew. Chem., Int. Ed. 2018, 57, 182−186. (c) Saladrigas, M.; Loren,
G.; Bonjoch, J.; Bradshaw, B. ACS Catal. 2018, 8, 11699−11703.
(24) Wilsey, S.; Dowd, P.; Houk, K. N. J. Org. Chem. 1999, 64,
8801−8811.
(25) Salamone, M.; Bietti, M. Synlett 2014, 25, 1803−1816.
(26) Smith, J. R. L.; Masheder, D. J. Chem. Soc., Perkin Trans. 2
1977, 1732−1736.
(27) (a) Dimakos, V.; Su, H. Y.; Garrett, G. E.; Taylor, M. S. J. Am.
Chem. Soc. 2019, 141, 5149−5153. (b) Ye, J.; Kalvet, I.; Schoenebeck,
F.; Rovis, T. Nat. Chem. 2018, 10, 1037−1041.
E
DOI: 10.1021/acs.orglett.9b02307
Org. Lett. XXXX, XXX, XXX−XXX