Metal-free, cooperative asymmetric organophotoredox catalysis with visible light

Article abstract of DOI:10.1002/anie.201002992

The dawn of old stars: Classic xanthene dyes like eosin Y (gr. I=goddess of dawn) and green-light irradiation can replace precious metal complexes for the organocatalytic asymmetric α-alkylation of aldehydes, thus rendering the process purely organic.

Full text of DOI:10.1002/anie.201002992

DOI: 10.1002/anie.201002992
Photocatalysis
Metal-Free, Cooperative Asymmetric Organophotoredox Catalysis
with Visible Light**
Matthias Neumann, Stefan Fꢀldner, Burkhard Kꢁnig, and Kirsten Zeitler*
In the last decade organocatalysis has developed into an
essential third branch of asymmetric catalysis that now
complements the fields of metal and enzyme catalysis and
provides widely applicable methods for efficient organic
synthesis.[1,2] Especially the combination and integration in
cooperative catalysis such as domino reactions[3] and the
recent efforts in combining organocatalysis with metal
activation[4] demonstrate that the potential of organocatalysis
for the development of new activation modes in selective
organic synthesis is still not fully uncovered. Moreover,
photocatalysis with visible light[5] is undoubtedly one of the
emerging strategies to meet the increasing demand for more
sustainable chemical processes. Building on seminal results
employing photoinduced electron-transfer processes,[6] which
often required UV light, a number
and the choice of appropriate reaction conditions would
additionally allow for the cooperative merging with asym-
metric organocatalysis.
Herein, we present a versatile metal-free, purely organic
photoredox catalysis with visible light. As a first example of
our strategy we demonstrate the successful application of
simple, inexpensive organic dyes as effective photocatalysts
for the cooperative organocatalytic asymmetric intermolecu-
lar a-alkylation of aldehydes.[11]
Initial studies began with the screening of a number of red
and orange dyes (Scheme 1) for the photocatalytic reductive
dehalogenation of a-bromoacetophenone (E0 = À0.49 V
vs. SCE)[12] as a test reaction (Table 1).[6c,13] Following the
observation that classic organic dyes show striking similarities
of powerful methods have been
developed recently applying organ-
ometallic complexes such as [Ru-
(bpy)3]2+ and [Ir(ppy)2(dtb-bpy)]+
(bpy = bipyridine, ppy = 2-phenyl-
pyridine, dtb-bpy =
4,4’-di-tert-
butyl-2,2’-bipyridine).[5,7] Of partic-
ular note is the cooperative combi-
nation of photocatalysis with an
organocatalytic cycle[8] offering one
of the rare catalytic methods for the
enantioselective a-alkylation of
aldehydes.[9,10]
However, the high cost and
potential toxicity of the ruthenium
and iridium salts as well as their
limited availability in the future are
disadvantages of these metal-based
methods. Stimulated by the attrac-
tiveness of using green light, the
most abundant part of solar light, we
speculated that a number of red to
orange dyes could also be used
successfully in photoredox catalysis,
Scheme 1. Absorption and redox properties of red and orange organic dyes used as photoredox
catalysts (lmax (CH3CN) in nm; 3 in CH2Cl2; E0 (dye/dyeCÀ) in V vs. SCE)[16] in comparison with
common organometallic photocatalysts. SCE: saturated calomel electrode.
[*] M. Neumann, S. Fꢀldner, Prof. Dr. B. Kꢁnig, Dr. K. Zeitler
Institut fꢀr Organische Chemie, Universitꢂt Regensburg
Universitꢂtsstrasse 31, 93053 Regensburg (Germany)
Fax: (+49)941-943-4121
to the widely employed organometallic ruthenium- and
iridium-containing photosensitizers, we chose our test candi-
dates based on their lmax, their redox potential E0, and their
precedented use as photosensitizers for semiconductor-based
photocatalysis or dye solar cells.[14,15]
To achieve this desired transformation we investigated the
conditions reported by Stephenson and co-workers for the
photocatalytic dehalogenation of activated benzylic halides in
the presence of [Ru(bpy)3]2+. In accordance with their results
E-mail: kirsten.zeitler@chemie.uni-regensburg.de
[**] This work was supported by the GRK 1626 (“Chemical Photo-
catalysis”) of the DFG. S.F. gratefully acknowledges a fellowship of
the Bayerische Elitefꢁrderung.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002992.
Angew. Chem. Int. Ed. 2011, 50, 951 –954
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
951

Communications
Table 1: Dehalogenation of a-bromoacetophenone.
Table 2: Photocatalytic reductive dehalogenation with eosin Y using
Hantzsch ester 7 as a reduction equivalent.
Entry Substrate
1
Product
Yield [%][a]
100[b]
Entry[a]
Dye catalyst
Yield [%][b]
1
2
3
4
5
6
7
8
9
10
none
40
100
36
[Ru(bpy)32+] (8)[c]
alizarin red S (4)
perylene 3
100
100
100
100
3[d]
nile red (5)
fluorescein (1)
eosin Y (2)
eosin Y (2)
eosin Y (2)
2
3
4
83
78 (78)[c]
89 (88)[c]
80[e]
80
rhodamine B (6)
[a] Standard conditions as described above. [b] GC yield determined
using calibrated internal standard. [c] A blue high-power LED
a
(lꢀ455 nm) was used instead. [d] Reaction was performed in the dark.
[e] Reaction was conducted in sunlight; full conversion was reached after
1 h of irradiation.
[a] Yield of isolated products. [b] Yield determined by GC and NMR using
appropriate calibrated internal standards. [c] Yields in brackets as
reported in Ref. [13].
we noticed that also for our a-carbonyl bromide substrate the
use of 1.1 equiv of Hantzsch ester 7 as a hydride source was
beneficial to avoid potential side reactions.
proving the effectiveness of our operationally simple, inex-
pensive conditions.[20] It also should be noted that the
irradiation power of the employed LEDs and therefore the
applied energy to the reaction system is drastically less than
that of sunlight or typically applied fluorescent lightbulbs.[17]
Next we turned our attention to the application of organic
While under these conditions a slow background reaction
also leads to detectable amounts of the debrominated product
(Table 1, entry 1), most of the simple organic dyes were
effective for this transformation under optimized conditions,
albeit with different yields. Whereas light proved essential for
this transformation (Table 1, entry 8), the reaction can be
conducted using different light sources. Fast conversion is
observed in ambient sunlight (Table 1, entry 9), however with
a slight decrease in product yield, potentially because of side
reactions that may occur at the higher reaction temperature
and at the UV portion of the solar spectrum.
dyes as photoredox catalysts in the asymmetric organocata-
[8]
À
lytic C C bond formations developed by MacMillan et al.
As highlighted in Table 3 the transformations were found to
be both high-yielding and enantioselective when a combina-
tion of eosin Y (2) and MacMillanꢀs imidazolidinone catalyst
17 were applied. Even though our organic-dye-sensitized
conditions require somewhat longer reaction times,[21] we did
not observe product racemization, which further illustrates
the previously elucidated strict differentiation of the trans-
substituted catalyst between a-methylene aldehydes and a-
substituted products.[22] The enantioselectivity depends on the
reaction temperature (Table 3, entries 1, 4, and 5) and À58C
was found to be optimal. Performing the reaction under direct
sunlight led to faster conversion, albeit with a slight erosion in
enantioselectivity presumably because of the increased reac-
tion temperature (roughly 308C).
Upon irradiation with green light[17] from high-power
LEDs with an emission of l ꢀ 530 nm, bleaching of the dyes
was minimized but still observable for alizarin 4, nile red (5),
and rhodamine B (6) indicating the slow degradation of the
photosensitizer. However, perylene 3 and the xanthene-based
dyes 1 and especially eosin Y (2) proved to be sufficiently
stable under the reaction conditions. Using eosin 2 as the
photocatalyst affords the defunctionalized product in a very
clean, high-yielding reaction as determined by both GC and
NMR studies using appropriate internal standards.[18] Owing
to its simplicity and favorable redox and photochemical
properties eosin Y (2) was selected as the photocatalyst for
our subsequent studies.[19]
A number of dehalogenations (Table 2) under our opti-
mized conditions showed that the reaction is also tolerant to
aromatic residues with electron-withdrawing substituents
(Table 2, entry 2). Polar functional groups such as esters are
tolerated and exclusive chemoselectivity for a-activated
substrates over aryl halides was observed for the defunction-
alization (Table 2, entries 3 and 4). In all cases the obtained
yields of the isolated products are equal or better than those
for the reported transition-metal-catalyzed counterpart[13]
Our methodology is also compatible with the stereospe-
cific incorporation of polyfluorinated alkyl substituents
(Table 3, compound 21), which are important elements in
drug design to modulate specific properties.[23]
At present, the mechanistic picture of this reaction is not
complete. It is evident, however, that eosin Y acts as a
photoredox catalyst after its excitation with visible light and
population of its more stable triplet state finally enabling
single-electron transfer (SET; Scheme 2).[24] Similar to the
chemistry of Ru2+* both reductive and oxidative quenching
3
are known for excited eosin Y EY*.[25] Because our results
are comparable to those of MacMillan et al. we presume that
952
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 951 –954

Table 3: Purely organocatalytic enantioselective a-alkylation/perfluoro-
alkylation of aliphatic aldehydes.
eosin Y acts a reductant—relying on the sacrificial oxidation
of a catalytic amount of the enamine as the initial electron
reservoir[26]—to furnish the electron-deficient alkyl radical by
means of SET with an alkyl halide. Addition of this radical to
the electron-rich olefin of the enamine that is simultaneously
generated within the organocatalytic cycle merges both
activation pathways. In the catalytic cycle the subsequent
oxidation of the amino radical to the iminium species
provides the electron for the reductive quenching of the
dyeꢀs excited state 3EY*.[27]
Having successfully demonstrated the versatility of simple
organic dyes for photoredox catalysis we directed our efforts
to the determination of the quantum yield of the reaction to
gain further information on its efficiency.[28] We reproducibly
found values in the range of 6 to 9% indicating a more
complex reaction course than the proposed simplified mech-
anistic platform. To further prove this assumption we
conducted an additional GC-based yield determination after
keeping the initially irradiated sample in the dark for 3 h and
6 h. Here we found a significant increase of the yield which
might stem from the involvement of an amplifying “dark
reaction”.
Entry Variation from the standard conditions[a]
Yield [%][b] ee [%][c]
1
2
3
none
63
78
77
80
76
23W fluorescent bulb instead of LED
23W fluorescent bulb instead of LED and 75
[Ru(bpy)3]Cl2 (8)
4
5
6
T=08C
70
85
77
81
88
76
T=À58C
sunlight (Tꢀ308C)[d]
In summary, we have developed a metal-free method
using inexpensive eosin Y as a powerful photocatalyst for
various photoredox transformations with a performance
comparable to that of noble-metal catalysts. The discovery
of a purely organic asymmetric cooperative photoredox
organocatalysis will facilitate applications of these useful
reactions in organic synthesis significantly as xanthene dyes
are readily accessible, cheap, and less toxic than transition-
metal complexes. This extension of highly versatile photo-
redox catalysis to classic organic dyes is expected to find
broadly useful across many applications.
[a] Standard conditions as described above. [b] Yield of isolated
products. [c] Enantiomeric excess was determined as reported in
Ref. [8a]. [d] Full conversion was reached after approximately 4 h.
[e] Reaction was performed at +58C; p-NO2-phenacyl bromide was
used. [f] Phenylpropionaldehyde was used instead of octanal. [g] Reac-
tion was performed at À158C; 1-iodoperfluorobutane was used instead
of diethyl bromomalonate.
Received: May 17, 2010
Revised: June 10, 2010
Published online: September 28, 2010
Keywords: asymmetric alkylation · cooperative catalysis ·
.
organocatalysis · photocatalysis · xanthene dyes
[1] For recent reviews on organocatalysis, see: a) A. Dondoni, A.
Massi, Angew. Chem. 2008, 120, 4716 – 4739; Angew. Chem. Int.
Ed. 2008, 47, 4638 – 4660; b) D. W. C. MacMillan, Nature 2004,
346, 1007 – 1249; f) J. Seayad, B. List, Org. Biomol. Chem. 2005,
3, 719 – 724; g) A. Berkessel, H. Grꢁger, Asymmetric Organo-
catalysis, Wiley-VCH, Weinheim, 2005; h) P. I. Dalko, Enantio-
selective Organocatalysis, Wiley-VCH, Weinheim, 2007.
[2] For some selected recent applications, see: a) H. Ishikawa, T.
Suzuki, Y. Hayashi, Angew. Chem. 2009, 121, 1330 – 1333;
Angew. Chem. Int. Ed. 2009, 48, 1304 – 1307; b) B.-C. Hong, P.
Kotame, C.-W. Tsai, J.-H. Liao, Org. Lett. 2010, 12, 776 – 779;
c) P. Jakubec, D. M. Cockfield, D. J. Dixon, J. Am. Chem. Soc.
2009, 131, 16632 – 16633; Review: d) R. Marcia de Figueiredo,
M. Christmann, Eur. J. Org. Chem. 2007, 2575 – 2600.
[3] a) C. Grondal, M. Jeanty, D. Enders, Nat. Chem. 2010, 2, 167 –
178; b) D. Enders, C. Grondal, M. R. M. Hꢂttl, Angew. Chem.
2007, 119, 1590 – 1601; Angew. Chem. Int. Ed. 2007, 46, 1570 –
1581; c) L. F. Tietze, G. Brasche, K. Gerike, Domino Reactions in
Organic Chemistry, Wiley-VCH, Weinheim, 2006.
Scheme 2. Proposed mechanism and comparison of the photoredox
cycles of RuII and eosin Y.
Angew. Chem. Int. Ed. 2011, 50, 951 –954
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
953

Communications
[4] a) J. Zhou, Chem. Asian J. 2010, 5, 422 – 434; b) C. Zhong, X. Shi,
converted according to: V. V. Pavlishchuk, A. W. Addison,
Inorg. Chim. Acta 2000, 298, 97 – 102; see the Supporting
Information for details and references.
Eur. J. Org. Chem. 2010, 2999 – 3025; c) Z. Shao, H. Zhang,
Chem. Soc. Rev. 2009, 38, 2745 – 2755.
[5] a) K. Zeitler, Angew. Chem. 2009, 121, 9969 – 9974; Angew.
Chem. Int. Ed. 2009, 48, 9785 – 9789; b) D. Ravelli, D. Dondi, M.
Fagnoni, A. Albini, Chem. Soc. Rev. 2009, 38, 1999 – 2011.
[6] a) M. Ishikawa, S. Fukuzumi, J. Am. Chem. Soc. 1990, 112, 8864 –
8870; b) S. Fukuzumi, S. Mochizuki, T. Tanaka, J. Chem. Soc.
Perkin Trans. 2 1989, 1583 – 1589; c) S. Fukuzumi, S. Mochizuki,
T. Tanaka, J. Phys. Chem. 1990, 94, 722 – 726; d) G. Pandey, S.
Hajra, Angew. Chem. 1994, 106, 1217 – 1218; Angew. Chem. Int.
Ed. Engl. 1994, 33, 1169 – 1171; e) G. Pandey, M. K. Ghorai, S.
Hajra, Pure Appl. Chem. 1996, 68, 653 – 658.
[7] a) A. G. Condie, J. C. Gonzꢃlez-Gꢄmez, C. R. J. Stephenson, J.
Am. Chem. Soc. 2010, 132, 1464 – 1465; b) J. W. Tucker, J. M. R.
Narayanam, S. W. Krabbe, C. R. J. Stephenson, Org. Lett. 2010,
12, 368 – 371; c) J. Du, T. P. Yoon, J. Am. Chem. Soc. 2009, 131,
14604 – 14605; d) M. A. Ischay, M. E. Anzovino, J. Du, T. P.
Yoon, J. Am. Chem. Soc. 2008, 130, 12886 – 12887; e) T. Koike,
M. Akita, Chem. Lett. 2009, 38, 166 – 167.
[17] The reaction can also be conducted under the light of a 23 W
fluorescent bulb; see also Ref. [21]. High-power LEDs applied
for photocatalysis (e.g. Philips LUXEON Rebel 1W) show high
color fidelity (l = 530 Æ 10 nm) and a radiometric power of
approximately 145 lm.
[18] See the Supporting Information for details.
[19] In a number of comparative studies using xanthene dyes as
photoinitiators eosin Y has been shown to counterbalance high
reactivity with sufficient stability. For selected examples, see:
a) T. Lazarides, T. McCormick, P. Du, G. Luo, B. Lindley, R.
Eisenberg, J. Am. Chem. Soc. 2009, 131, 9192 – 9194; b) M. V.
Encinas, A. M. Rufs, S. G. Bertolotti, C. M. Previtali, Polymer
2009, 50, 2762 – 2767; c) S. H. Lee, D. H. Nam, C. B. Park, Adv.
Synth. Catal. 2009, 351, 2589 – 2594.
[20] Price per mmol and molecular weight of different photocata-
lysts: [Ru(bpy)3]Cl2 ꢇ 6H2O (MW= 748.62 gmmolÀ1): $62.50;
[Ir(ppy)2(dtb-bpy)]PF6 (MW= 1072.09 gmmolÀ1): ca. $630
(single-step synthesis from commercial [{Ir(ppy)2Cl}2] with
2 equiv dtb-bpy); eosin Y (MW= 647.89 gmmolÀ1): $2.66
(based on Sigma–Aldrich and Acros prices for 2010).
[21] In our hands using household fluorescent bulbs available in
Germany (e.g. OSRAM, 23 W, 6500 K, 1470 lm) we also could
not reach full conversion in the time span reported by
MacMillan and co-workers[8a] with their optimized [Ru(bpy)3]2+
conditions (see Table 3, entry 3).
[22] M. Amatore, T. D. Beeson, S. P. Brown, D. W. C. MacMillan,
Angew. Chem. 2009, 121, 5223 – 5226; Angew. Chem. Int. Ed.
2009, 48, 5121 – 5124.
[23] For the seminal report on enantioselective trifluormethylation of
aldehydes by photoredox catalysis see Ref. [8b]. A mechanisti-
cally different nonphotolytic (closed-shell) approach merging
organocatalysis with iodonium salts was published only recently:
A. E. Allen, D. W. C. MacMillan, J. Am. Chem. Soc. 2010, 132,
4986 – 4987.
[8] a) D. A. Nicewicz, D. W. C. MacMillan, Science 2008, 322, 77 –
80; b) D. A. Nagib, M. E. Scott, D. W. C. MacMillan, J. Am.
Chem. Soc. 2009, 131, 10875 – 10877.
[9] For a pioneering example in intramolecular organocatalytic
asymmetric alkylation: a) N. Vignola, B. List, J. Am. Chem. Soc.
2004, 126, 450—451; for intramolecular examples within domino
processes, see: b) D. Enders, C. Wang, J. W. Bats, Angew. Chem.
2008, 120, 7649 – 7653; Angew. Chem. Int. Ed. 2008, 47, 7539 –
7542; c) H. Xie, L. Zu, H. Li, J. Wang, W. Wang, J. Am. Chem.
Soc. 2007, 129, 10886 – 10894; d) R. Rios, H. Sundꢅn, J. Vesely,
G.-L. Zhao, P. Dziedzic, A. Cꢄrdova, Adv. Synth. Catal. 2007,
349, 1028 – 1032.
[10] For recent examples relying on the use of stabilized carbocations
as compatible electrophiles: a) R. R. Shaikh, A. Mazzanti, M.
Petrini, G. Bartoli, P. Melchiorre, Angew. Chem. 2008, 120,
8835 – 8838; Angew. Chem. Int. Ed. 2008, 47, 8707 – 8710; b) P. G.
Cozzi, F. Benfatti, L. Zoli, Angew. Chem. 2009, 121, 1339 – 1342;
Angew. Chem. Int. Ed. 2009, 48, 1313 – 1316; c) L. Zhang, L. Cui,
X. Li, J. Li, S. Luo, J.-P. Cheng, Chem. Eur. J. 2010, 16, 2045 –
2049.
[24] The efficient formation of long-lived triplet states following
photoexcitation through ISC is facilitated by the heavy-atom
effect (increased spin–orbit mixing; Br substituents): a) T.
Shimidzu, T. Iyoda, Y. Koide, J. Am. Chem. Soc. 1985, 107,
35 – 41; b) D. C. Neckers, O. M. Valdes-Aguilera, Adv. Photo-
chem. 1993, 18, 315 – 394.
[11] For a recent review on catalytic asymmetric a-alkylations, see:
A.-N. Alba, M. Viciano, R. Rios, ChemCatChem 2009, 1, 437 –
439.
[12] If one considers the reductive quenching of the dyeꢀs photo-
excited state Dye* to DyeCÀ by an available electron-donating
species (vide infra), all the displayed dyes apart from Alizarin
red S (4) have been shown to be potent reductants (cf. redox
potentials E0), which should facilitate cleavage of the C-Hal
bond to furnish the electron-deficient radical by SET to the
activated a-bromocarbonyl compound.
[25] For a discussion on reductive vs. oxidative properties of the
3
excited EY*, see Ref. [19a]. For selected recent examples on
reductive and oxidative quenching, see: a) F. Labat, I. Ciofini,
H. P. Hratchian, M. Frisch, K. Raghavachari, C. Adamo, J. Am.
Chem. Soc. 2009, 131, 14290 – 14298; b) M. A. Jhonsi, A.
Kathiravan, R. Renganathan, J. Mol. Struct. 2009, 921, 279 – 284.
3
[26] The high-energy intermediate of eosin Y EY* can function as
[13] J. M. R. Narayanam, J. W. Tucker, C. R. J. Stephenson, J. Am.
Chem. Soc. 2009, 131, 8756 – 8757.
an oxidant, showing similar redox properties as photoexcited
[Ru(bpy)3]2+ *: cf. E0(3EY*/EYCÀ) = + 0.83 V vs. E0(Ru2+*/
Ru+) = + 0.79 V vs. SCE; see the Supporting information for
details.
[14] M. Zhang, C. Chen, W. Ma, J. Zhao, Angew. Chem. 2008, 120,
9876 – 9879; Angew. Chem. Int. Ed. 2008, 47, 9730 – 9733.
[15] a) J. Moser, M. Grꢆtzel, J. Am. Chem. Soc. 1984, 106, 6557—
6564; for recent reviews, see: b) Y. Ooyama, Y. Harima, Eur. J.
Org. Chem. 2009, 2903 – 2934; c) A. Mishra, M. K. R. Fischer, P.
Bꢆuerle, Angew. Chem. 2009, 121, 2510 – 2536; Angew. Chem.
Int. Ed. 2009, 48, 2474 – 2499.
[27] Remarkably, in contrast to most photocatalytic processes
MacMillanꢀs system does not require any sacrificial oxidant or
reductant by its design; both oxidation and reduction steps are
productive and lead to the formation of the desired product.
[28] For a recent discussion on the normalization of photocatalytic
reactions, see: T. Maschmeyer, M. Che, Angew. Chem. 2010, 122,
1578 – 1582; Angew. Chem. Int. Ed. 2010, 49, 1536 – 1539.
[16] For a better comparison all values are reported in reference to
the SCE electrode. If necessary the original values were
954
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 951 –954