Photoredox Allylation Reactions Mediated by Bismuth in Aqueous Conditions

Article abstract of DOI:10.1002/ejoc.202001640

Organometallic allylic reagents are widely used in the construction of C?C bonds by Barbier-type reactions. In this communication, we have described a photoredox Barbier allylation of aldehydes mediated by bismuth, in absence of other metals as co-reductants. Mild reaction conditions, tolerance of oxygen, and use of aqueous solvent make this photoredox methodology attractive for green and sustainable synthesis of homoallylic alcohols.

Full text of DOI:10.1002/ejoc.202001640

Communications
doi.org/10.1002/ejoc.202001640
1
2
3
4
5
6
7
8
9
Photoredox Allylation Reactions Mediated by Bismuth in
Aqueous Conditions
Simone Potenti,[a, b] Andrea Gualandi,*[a] Alessio Puggioli,[a] Andrea Fermi,[a]
Giacomo Bergamini,[a] and Pier Giorgio Cozzi*[a]
This paper is dedicated to Prof. Franco Cozzi on the occasion of his 70th birthday.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
e.g. an amine (DIPEA or TEA) or Hantzsch’s ester. In the case of
nickel, cobalt, and titanium, the catalytic cycle is feasible with
the use of an organic reducing agent. In both cases, the
allylation reaction is performed in the presence of a photo-
catalyst (iridium complex or an organic dye), although recently
Gansäuer pointed out[16] that titanium complexes themselves
can act as photocatalysts under precise conditions (green light
irradiation). The described methodologies are certainly innova-
tive and can be further expanded towards other C–C bond-
forming transformations. Barbier reactions with certain types of
metals (zinc, indium, bismuth, gallium) were developed in
aqueous solvents.[17] Tofurther extend the advantage of photo-
redox allylation reactions, and explore the use of green and
sustainable conditions, we wondered if the mentioned metals
could be employed in photoredox allylation reactions under
aqueous Barbier conditions, and herein we report the successful
endeavor of our investigations.
Bismuth is an inexpensive, safe, and environmentally-benign
metal, that is commonly used in cosmetics, and as a component
of oral gastrointestinal drugs.[18] Bismuth has been used in
allylation reactions under various conditions,[19] and in the
presence of stoichiometric metals as reductants, such as Al, Mg,
Fe, and Zn.[20] Interestingly, also sodium borohydride was a
suitable reductant for Barbier-typeallylation reaction with
bismuth.[21] Based on these reports, we selected bismuth salts as
metal catalysts for the catalytic photoredox allylation of
aldehydes. Starting by employing 4-chlorobenzaldehyde as the
model substrate, we have optimized the allylation reaction
using allyl bromide and Bi(OTf)3 (Table 1). We have avoided the
employment of metal photocatalysts based on iridium and
ruthenium, focusing our investigation on the class of thermally
activated delayed fluorescence (TADF) organic dyes based on
carbazoyl and diphenylamine substituted dicyanoarenes.[22]
Among all the TADF dyes, 3CzClIPN[23] was the dye of choice.
We also varied – in the model reaction – the solvent, finding
that a 1:1 mixture of EtOH/H2O was convenient to satisfyingly
perform the reaction (Table 1, entry 6). Furthermore, the
reaction does not require strictly de-oxygenated solvents and
can be conveniently settled without a freezing-pump thaw
procedure to eliminate traces of oxygen (Table 1, entry 7). The
reaction is not sensitive to the presence of radical scavengers
like (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO, Table 1, en-
try 8). Different bismuth salts were also tested in the model
reactions (see Table S3) and we found that BiBr3 was also a
compelling catalyst for the reaction (Table 1 entry 14). Due to
Organometallic allylic reagents are widely used in the con-
struction of CÀ C bonds by Barbier-type reactions. In this
communication, we have described a photoredox Barbier
allylation of aldehydes mediated by bismuth, in absence of
other metals as co-reductants. Mild reaction conditions, toler-
ance of oxygen, and use of aqueous solvent make this photo-
redox methodology attractive for green and sustainable syn-
thesis of homoallylic alcohols.
In recent years, photoredox catalysis has been developed
towards effective and practical new synthetic methodologies
for CÀ C and CÀ X (X=O, N, S, P) bond-forming reactions.[1]
Metalla photoredox catalysis, developed from the seminal work
of Sanford,[2] Molander,[3] and MacMillan-Doyle,[4] has consider-
ably expanded the repertoire of reactivity and scope, allowing
the
development
of
new,
mild,
and
interesting
transformations.[5] From the application of metalla photoredox
catalysis in the context of cross-coupling reactions[6] the
methodology has evolved to consider radical to polar crossover
mechanism,[7] in which carbanion[8] or carbenium ion[9] are
formed by the reaction of a metal with a radical generated
under photoredox conditions. Recently, allylation reactions
were described in Barbier conditions, with the involvement of
nickel,[10] chromium,[11] titanium,[12] and cobalt.[13] Stereoselective
allylation reactions with chromium[14] and nickel[15] were also
developed, showing the possibilities offered by the use of
photoredox conditions, in the presence of metals, for asymmet-
ric reactions. Essentially, photoredox Barbier conditions avoid
the need of a metal co-reductant (often Mn or Zn). In the
examples reported with chromium, the photoredox cycle does
not need a sacrificial reducing agent such as organic molecules,
[a] S. Potenti, Dr. A. Gualandi, A. Puggioli, Dr. A. Fermi, Prof. G. Bergamini,
Prof. P. G. Cozzi
Dipartimento di Chimica “G. Ciamician”
ALMA MATER STUDIORUM, Università di Bologna
Via Selmi 2, 40126, Bologna, Italy
E-mail: andrea.gualandi10@unibo.it
piergiorgio.cozzi@unibo.it
https://site.unibo.it/stereoselective-metal-photoredox-catalysis-lab
[b] S. Potenti
Laboratorio SMART, Scuola Normale Superiore,
Piazza dei Cavalieri 7, 56126, Pisa, Italy
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/ejoc.202001640
Part of the Franco Cozzi’s 70th Birthday Special Collection
Eur. J. Org. Chem. 2021, 1624–1627
1624
© 2021 Wiley-VCH GmbH

Communications
doi.org/10.1002/ejoc.202001640
Table 1. Screening of reaction conditions for bismuth-mediated catalytic
photoredox allylation of aldehydes.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Entry[a] Deviation from standard con-
ditions
Conv.
[%][b]
3a
3a:4a
[%][c]
[%][c]
1
2
3
4
5
6
7
8
Solvent THF:H2O (1:1)
Solvent DMF:HO (1:1)
Solvent MeCN:H2O (1:1)
Solvent MeOH:H2O (1:1)
Solvent DMSO:H2O (1:1)
–
97
53
81
>99
98
>99
90
>99
51
48
45
79
66
52:48
85:15
56:44
79:21
68:32
83(74) 83:17
70
95
In air
79:21
95:5
In the presence of TEMPO
(20 mol%)
9
No Bi(OTf)3
No 3CzClIPN
No Bi(OTf)3; No 3CzClIPN
No light
Allylchloride instead allylbro-
>99
59
6
0
15
16
59
0
–
7
16:84
>99:1
>1/99
–
Scheme 1. Photoredox allylation of aromatic aldehydes mediated by
bismuth.
10
11
12
13
48:52
mide
14
BiBr3 instead Bi(OTf)3
1 mmol scale
>99
89
87
87:13
15[e]
85(77) 95:5
[a] All the reactions were carried out under irradiation with Kessil® 40 W
blue LED. [b] Conversions were measured by H-NMR. Isolated yields after
chromatographic purification are reported in parenthesis. [c] Yield% of the
allylated product (3a) determined by 1H-NMR. [d] Ratio% of allylated
product (3a) and pinacol coupling (4a). The d.r. for (4a) is ca 1:1 for all the
reactions. [e] The reaction was performed with 8 mol% of Bi(OTf)3 and
3 mol% of 3CzClIPN.
1
the difficulty to manage this deliquescent salt, we decided to
use Bi(OTf)3, for further studies. Interestingly, in absence of the
Bi(OTf)3, but in the presence of 3CzClIPN, the pinacol coupling
of the aldehyde is favored (Table 1, entry 9).[24] If the reaction is
carried out without 3CzClIPN, under irradiation with blue LED
(Table 1, entry 10), we have observed a clean allylation reaction,
although in quite a minor conversion. This was probably due to
the ability of bismuth to form a reactive allylating agent with
the mediation of Hantzsch’s ester under light irradiation. In fact,
when bismuth is not introduced in the reaction mixture, and
the photocatalyst 3CzClIPN is absent (Table 1 entry 11), the
photoredox properties of the Hantzsch’s ester favored the
pinacol coupling.[25] The scale-up of the reaction is possible and
on 1 mmol scale we have reduced to 8mol% the percentage of
bismuth catalyst (entry 15).
The optimal reaction conditions were explored with a large
variety of aromatic and aliphatic aldehydes, and the salient
results are reported in Scheme 1 and Scheme 2. In general, with
aromatic aldehydes, the isolated yields were from moderate to
good, with a variety of different functional groups compatible
with the reaction conditions. Both electron-rich and -poor
aromatic aldehydes are reactive and, in some cases, yields could
be improved by an increase of the reaction time to 72 h. Sterical
Scheme 2. Photoredox allylation of aliphatic aldehydes mediated by bis-
muth.
hindrance in ortho position does not hamper the reaction.
Heteroaromatic aldehydes can be also employed, with some
limitations. We found that electron-rich aldehydes bearing
pyrrole or indole are unstable during chromatographic purifica-
tions due to their attitude to forming elimination products.
Thiophene is a compatible group, but 2-thiophene carboxalde-
hyde 1o was found poorly reactive, probably due to chelation
of sulfur to the bismuth reagent. The reactivity is restored when
3-thiophene carboxaldehyde 1p is employed. When long
reaction times (72 h) were applied to the reactions of the
aldehydes 1k, 1o and 1q, byproducts were observed. We have
detected (and isolated only for 1k) the corresponding ethyl
ether of the homoallylic alcohols (3k,o,q) formed by their
reaction with ethanol (see Table S4). Although Bi(OTf)3 is
described as a catalyst for SN1-type reaction of alcohols[26] the
Eur. J. Org. Chem. 2021, 1624–1627
www.eurjoc.org
1625
© 2021 Wiley-VCH GmbH

Communications
doi.org/10.1002/ejoc.202001640
formation of ethers was determined by the Brønsted acidity of
1
2
3
4
5
the oxidized Hantzsch’s ester. This behavior was confirmed by
dissolving the alcohol 3k in EtOH in the presence of the
pyridinium salt derived from the oxidation of Hantzsch’s ester
(see Table S5). The formation of these ether byproducts was
observed only for alcohols bearing an electron-rich aromatic
ring.
6
7
On the other hand, the reaction is useful with aliphatic
aldehydes 1r–ab. Several functional groups are tolerated in the
reported conditions and the reaction is applicable to linear and
branched aldehydes with a reaction time of 72 h. Interestingly,
cinnamaldehyde (1ab) gave a mixture of E and Z isomer of the
corresponding allylated product (3ab) probably due to photo-
isomerization of cinnamaldehyde in presence of light and
photocatalyst.[27]
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 1. Proposed catalytic cycle for photoredox allylation of aldehydes
mediated by bismuth.
Ketones showed a quite reduced reactivity in this reaction
compared to aldehydes, and the corresponding products were
detected only in traces (see Figure S1).
SCE) species is formed. The high tolerance to oxygen and
TEMPO is ruling out a radical mechanism and formation of
allylic radicals. We have no direct evidence of the actually
reduced bismuth species but, as suggested by the literature,
the formation of Bi(0) could be considered. Multiple single
electron transfer (SET) events could be responsible for the
formation of Bi in low oxidations state from Bi(III) salt. Insertion
of active bismuth to CÀ Br bond could forms allylbismuth(III),
diallylbismuth(III), and triallylbismuth(III) as the organometallic
intermediates, that react with aldehydes to give the corre-
sponding homoallylic alkoxides.[20c] The protonated re-aromat-
ized Hantzsch ester possesses a low pKa and the aqueous
conditions enable a facile protonation of the intermediate
bismuth alkoxide, allowing the recycle of bismuth salts. We
have demonstrated in previous studies that the pyridine,
produced by the oxidation of the Hantzsch’s ester, does not
partecipate in quenching processes.
To conclude, we have reported a photoredox Barbier
allylation reaction that uses green solvents and conditions and
employs a not toxic metal such as bismuth. The use of
expensive transition metal-based photoredox catalysts and
stoichiometric metals (e.g. Mn or Zn) is avoided. However,
further mechanistic studies are needed to understand the
reaction mechanism with the possibility to expand this
chemistry to the use of other electrophiles.
Substituted allyl bromides were tested in the reaction with
poor conversions (see Table S6). Whereas prenyl and crotyl
derivatives led to the exclusive formation of branched products,
cinnamyl bromide gave a complex mixture of products.
In order to evaluate the mechanism of the reaction, we
investigated the quenching of the photocatalyst’s luminescence
by each of the components of the reaction (see SI for details).
As reported in our investigation on titanium photoredox
allylation,[12] isophthalonitrile derivatives represent a class of
remarkably visible emitters, which can show long emission
lifetimes in specific experimental conditions because of their
TADF behavior. Interestingly, 3CzClIPN displays double deacti-
vation kinetics at λem =550 nm even in air-equilibrated ethanol
at r.t. most likely due to a TADF-active regime, featuring a
delayed component with a lifetime around 150 ns (see SI for full
details).
In the presence of increasing amounts of Bi(OTf)3 or allyl
bromide, no appreciable decrease in the emission intensities of
3CzClIPN was detected (see Figure S3 and Figure S4), thus
showing that these two reactants are not involved in quenching
mechanisms even at high concentrations. On the other hand,
the addition of 4-chlorobenzaldehyde barely decreases the
emission intensity of 3CzClIPN (kq =6.2×106 MÀ 1 sÀ 1, see Fig-
ure S5), explaining the observed pinacol coupling in absence of
Bi(OTf)3. Similar behavior is observed when the Hantzsch’s ester
is introduced in solutions of 3CzClIPN, denoting a significantly
more efficient quenching of the emission of the latter(kq =3.8× Acknowledgements
108 MÀ 1 sÀ 1, see Figure S6).
Based on the results of the photophysical investigation, and
from studies reported in the literature,[20c] we can tentatively
suggest that the reductive quenching of 3CzClIPN by HE
National PRIN 2017 project (ID: 20174SYJAF, SURSUMCAT) is
acknowledged for financial support of this research.
*
induces the formation of HE + (Figure 1). The thermodynamic
feasibility of the photoinduced electron transfer (PET) process Conflict of Interest
relies on the reduction potential of the excited state of 3CzClIPN
*
(E1/2
=1.56 V vs SCE)[23] and on the oxidation
The authors declare no conflict of interest.
3CzClIPN*/3CzClIPN À
*
potential of the Hantzsch’s ester (EHE +/HE = +1.0 vs SCE).[27] The
*
reaction is thus producing HE + that can participate in further
Keywords: Allylation
·
Aldehydes
·
Barbier reaction
·
electron transfer events[27] and is a strong reductant. Further-
Photoredox catalysis · Organic dyes
*
*
more, the reducing 3CzClIPN (E1/2
=À 1.16 V vs
À
3CzClIPN/3CzClIPN À
Eur. J. Org. Chem. 2021, 1624–1627
www.eurjoc.org
1626
© 2021 Wiley-VCH GmbH

Communications
doi.org/10.1002/ejoc.202001640
[1] a) C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. 2013, 113,
[17] C. C. K. Keh, C. Wei, C.-J. Li, J. Am. Chem. Soc. 2003, 125, 4062–4063; and
ref. therein.
1
2
3
4
5
6
7
8
9
5322–5363; for a recent review, see: b) R. C. McAtee, E. J. McClein, C. R. J.
Stephenson, Trends in Chem. 2019, 1, 111–125.
[2] D. Kalyani, K. B. McMurtrey, S. R. Neufeldt, M. S. Sanford, J. Am. Chem.
Soc. 2011, 133, 18566–18569.
[3] J. C. Tellis, D. N. Primer, G. A. Molander, Science 2014, 345, 433–436.
[4] Z. Zuo, D. T. Ahneman, L. Chu, J. A. Terrett, A. G. Doyle, D. W. C.
MacMillan, Science 2014, 345, 437–440.
[5] J. Twilton, C. Lee, P. Zhang, M. H. Shaw, R. W. Evans, D. W. C. MacMillan,
Nature Rev. Chem. 2017, 1, 0052.
[18] P. J. Sadler, H. Li, H. Sun, Coord. Chem. Rev. 1999,185, 689–709.
[19] a) M. Wada, K. Akiba, Tetrahedron Lett. 1985, 26, 4211–4214; b) M. Wada,
H. Ohki, K. Akiba, Bull. Chem. Soc. Jpn. 1990, 63, 1738–1747; c) N.
Miyoshi, M. Nishio, S. Murakami, T. Fukuma, M. Wada Bull. Chem. Soc.
Jpn. 2000, 73, 689–692; d) P. C. Andrews, A. C. Peatt, C. L. Raston, Green
Chem. 2001, 3, 313–315; e) D. Miyamoto, N. Daikawa, K. Tanaka,
Tetrahedron Lett. 2003, 44, 6963–6964; f) K. Smith, S. Lock, G. A. El-Hiti,
M. Wada, N. Miyoshi, Org. Biomol. Chem. 2004, 2, 935–938; g) J. H. Dam,
P. Fristrup, R. Madsen, J. Org. Chem. 2008, 73, 3228–3235.
[20] a) M. Wada, H. Ohki, K.-J. Akiba, J. Chem. Soc. Chem. Commun. 1987, 10,
708–709; b) M. Wada, T. Fukuma, M. Morioka, T. Takahashi, N. Miyoshi,
Tetrahedron Lett. 1997, 38 8045–8048; c) B. D. Jadhav, S. K. Pardeshi,
Tetrahedron Lett. 2014, 55, 4948–4952; d) M. Minato, J. Tsuji, Chem. Lett.
1988, 17, 2049–2052.
[21] R. Ping-Da, P. Shi-Feng, D. Ting-Wei, W. Shi-Hui, Chin. J. Chem. 1996, 14,
462–465.
[22] P. Data, Y. Takeda, Chem. Asian J. 2019, 14, 1613–1636.
[23] E. Speckmeier, T. G. Fischer, K. Zeitler, J. Am. Chem. Soc. 2018, 140,
15353–15365.
[24] For other pinacol coupling promoted by organic dyes, see: a) A.
Gualandi, G. Rodeghiero, E. Della Rocca, F. Bertoni, M. Marchini, R.
Perciaccante, T. P. Jansen, P. Ceroni, P. G. Cozzi, Chem. Commun., 2018,
54, 10044–10047; b) A. Gualandi, A. Nenov, M. Marchini, G. Rodeghiero,
I. Conti, E. Paltanin, M. Balletti, P. Ceroni, M. Garavelli, P. G. Cozzi,
ChemCatChem 2020, https://doi.org/10.1002/cctc.202001690, and ref.
therein.
[6] J. A. Milligan, J. P. Phelan, S. O. Badir, G. A. Molander, Angew. Chem.
2019, 131, 6212–6224; Angew. Chem. Int. Ed. 2019, 58, 6152–6163.
[7] L. Pitzer, J. L. Schwarz, F. Glorius, Chem. Sci. 2019, 10, 8285–8291.
[8] K. Donabauer, M. Maity, A. L. Berger, G. S. Huff, S. Crespi, B. König, Chem.
Sci., 2019,10, 5162–5166.
[9] E. W. Webb, J. B. Park, E. L. Cole, D. J. Donnelly, S. J. Bonacorsi, W. R.
Ewing, A. G. Doyle, J. Am. Chem. Soc. 2020, 142, 9493–9500.
[10] a) A. Gualandi, G. Rodeghiero, A. Faraone, F. Patuzzo, M. Marchini, F.
Calogero, R. Perciaccante, T. P. Jansen, P. Ceroni, P. G. Cozzi, Chem.
Commun. 2019, 55, 6838–6841. For a photoredox nickel mediated
crotylation of aldehydes, see: b) Y.-L. Li, W.-D. Li, Z.-Y. Gu, J. Chen, J.-B.
Xia ACS Catal. 2020, 10, 1528–1534.
[11] a) J. L. Schwarz, F. Schäfers, A. Tlahuext-Aca, L. Lückemeier, F. Glorius, J.
Am. Chem. Soc. 2018, 140, 12705–12709; b) H. Mitsunuma, S. Tanabe, H.
Fuse, K. Ohkubo, M. Kanai, Chem. Sci. 2019, 10, 3459–3465; c) L. J.
Schwarz, H.-M. Huang, T. O. Paulisch, F. Glorius, ACS Catal. 2020, 10,
1621–1627.
[12] a) A Gualandi, F. Calogero, M. Mazzarini, S. Guazzi, A. Fermi, G.
Bergamini, P. G. Cozzi, ACS Catal. 2020, 10, 3857–3863; b) A. Fermi, A.
Gualandi, G. Bergamini, P. G. Cozzi, Eur. J. Org. Chem. 2020, 6955–6965.
[13] A. Gualandi, G. Rodeghiero, R. Perciaccante, T. P. Jansen, C. Moreno-
Cabrerizo, C. Foucher, M. Marchini, P. Ceroni, P. G. Cozzi, Adv. Synth.
Catal. 2021, DOI: 10.1002/adsc.202001250.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
[25] P. Z. Wang, J.-R. Chen, W.-J. Xiao, Org. Biomol. Chem. 2019, 17, 6936–
6951.
[26] M. Rueping, B.J Nachtsheim, Top. Curr. Chem. 2012, 311, 115–141.
[27] a) J. B. Metternich, R. Gilmour, J. Am. Chem. Soc. 2015, 137, 11254–
11257; b) W. Cai, H. Fan, D. Ding, Y. Zhang, W. Wang, Chem. Commun.
2017, 53, 12918–12921; c) N. F. Nikitas, I. Triandafillidi, C. G. Kokotos,
Green Chem. 2019, 21, 669–674.
[14] a) J. L. Schwarz, H.-M. Huang, T. O. Paulisch, F. Glorius, ACS Catal. 2020,
10, 1621–1627; b) S. Tanabe, H. Mitsunuma, M. Kanai, J. Am. Chem. Soc.
2020, 142, 12374–12381.
[15] A. Gualandi, F. Calogero, S. Potenti, E. Bassan, P. Ceroni, P. G. Cozzi,
manuscript in preparation.
[16] Z. Zhang, T. Hilche, D. Slak, N. Rietdijk, U. N. Oloyede, R. A. Flowers II, A.
Gansäuer, Angew. Chem. 2020, 132, 9441–9445; Angew. Chem. Int. Ed.
2020, 59, 9355–9359.
Manuscript received: December 18, 2020
Revised manuscript received: January 18, 2021
Accepted manuscript online: January 18, 2021
Eur. J. Org. Chem. 2021, 1624–1627
www.eurjoc.org
1627
© 2021 Wiley-VCH GmbH