peri-Xanthenoxanthene (PXX): a Versatile Organic Photocatalyst in Organic Synthesis

Article abstract of DOI:10.1002/adsc.202100030

Recent years have witnessed a continuous development of photocatalysts to satisfy the growing demand of photophysical and redox properties in photoredox catalysis, with complex structures or alternative strategies devised to access highly reducing or oxidising systems. We report herein the use of peri-xanthenoxanthene (PXX), a simple and inexpensive dye, as an efficient photocatalyst. Its highly reducing excited state allows activation of a wide range of substrates, thus triggering useful radical reactions. Benchmark transformations such as the addition of organic radicals, generated by photoreduction of organic halides, to radical traps are initially demonstrated. More complex dual catalytic manifolds are also shown to be accessible: the β-arylation of cyclic ketones is successful when using a secondary amine as organocatalyst, while cross-coupling reactions of aryl halides with amines and thiols are obtained when using a Ni co-catalyst. Application to the efficient two-step synthesis of the expensive fluoro-tetrahydro-1H-pyrido[4,3-b]indole, a crucial synthetic intermediate for the investigational drug setipiprant, has been also demonstrated. (Figure presented.).

Full text of DOI:10.1002/adsc.202100030

FULL PAPER
DOI: 10.1002/adsc.202100030
Very Important Publication
peri-Xanthenoxanthene (PXX): a Versatile Organic Photocatalyst
in Organic Synthesis
Cristofer Pezzetta,a, b Andrea Folli,a Oliwia Matuszewska,a Damien Murphy,a
Robert W. M. Davidson,b,* and Davide Bonifazia, c,*
a
School of Chemistry, Cardiff University,
Park Place, Cardiff CF10 3AT United Kingdom
Dr. Reddy’s Laboratories (EU),
b
410 Science Park, Milton Road, Cambridge,
CB4 0PE, United Kingdom
Phone: +44 1223651443
E-mail: rdavidson@drreddys.com
Institute of Organic Chemistry, Faculty of Chemistry,
c
University of Vienna, Währinger Strasse 38,
1090, Vienna, Austria
Phone: +43-1-4277-52158
E-mail: davide.bonifazi@univie.ac.at
Manuscript received: January 26, 2021; Revised manuscript received: March 12, 2021;
Version of record online: April 7, 2021
Supporting information for this article is available on the WWW under https://doi.org/10.1002/adsc.202100030
© 2021 The Authors. Advanced Synthesis & Catalysis published by Wiley-VCH GmbH. This is an open access article under the
terms of the Creative Commons Attribution Non-Commercial NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
Abstract: Recent years have witnessed a continuous development of photocatalysts to satisfy the growing
demand of photophysical and redox properties in photoredox catalysis, with complex structures or alternative
strategies devised to access highly reducing or oxidising systems. We report herein the use of peri-
xanthenoxanthene (PXX), a simple and inexpensive dye, as an efficient photocatalyst. Its highly reducing
excited state allows activation of a wide range of substrates, thus triggering useful radical reactions.
Benchmark transformations such as the addition of organic radicals, generated by photoreduction of organic
halides, to radical traps are initially demonstrated. More complex dual catalytic manifolds are also shown to be
accessible: the β-arylation of cyclic ketones is successful when using a secondary amine as organocatalyst,
while cross-coupling reactions of aryl halides with amines and thiols are obtained when using a Ni co-catalyst.
Application to the efficient two-step synthesis of the expensive fluoro-tetrahydro-1H-pyrido[4,3-b]indole, a
crucial synthetic intermediate for the investigational drug setipiprant, has been also demonstrated.
Keywords: PXX; xanthenoxanthene; photocatalysis; CÀ C; CÀ S; CÀ N; β arylation; mechanism; EPR; photo-
redox; highly reducing photocatalyst
Introduction
ate reactive radical species in very mild conditions is
surely at the root of this popularity. Ru(II)- and Ir(III)-
The field of photoredox catalysis has seen a huge based complexes are the work-horses in the field:
increase in popularity in the last decade, with numer- chemical stability, long excited state lifetimes and
ous useful transformation being discovered that were tunability of the photoredox properties through ligand
not possible, or were much more difficult to obtain, modifications are among the reasons for this prefer-
with non-photoredox methods.[1] The ability to gener- ence in the community.[1b] Early works revealed that
Adv. Synth. Catal. 2021, 363, 1–15
1
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
fac-[Ir(ppy)3], a homoleptic complex of Ir(III) (Fig- photocatalysts for CÀ N and CÀ S bond-forming cross-
ure 1), was particularly interesting for the strong coupling reactions as well.[5] Benzotriazole 3 was
reductive properties of its excited state.[2] With a redox designed for the activation of aryl iodides and
potential of À 1.73 V vs SCE, it can be used effectively subsequent radical addition to furans.[6] Carbazole 4
for the activation of alkyl and aryl halides to form showed a very low reduction potential, allowing
organic radical intermediates.[2b–d] The peculiar reac- activation of aryl chlorides and fluorides alike.[7]
A
tivity found with fac-[Ir(ppy)3] inspired further work different strategy involved the photoexcitation of an
towards the search of alternative photocatalytic sys- electron-rich transient species, being it an anion, a
tems with comparable or higher reducing power (Fig- photocatalytically-formed radical anion, an electro-
ure 1), especially to trigger photoredox activation of chemically-formed radical anion, a neutral radical or
non-accessible substrates because of their lower reduc- an excited state.[8] In these cases the excited state of the
tion potentials. The interest focused mainly on the transient species is extremely electron-rich, often
development of organic dyes, so that reaction mixtures capable of reducing electron-rich aryl halides to their
could be devoid of metallic impurities, and to avoid corresponding aryl radicals.[8f,g,i] Two-photon excitation
the use of expensive and supply-dependent precious of a water soluble fac-Ir(ppy)3 analogue also allowed
metals.[1c,f] The reduction of organic halides has the formation and synthetic utilisation of highly
typically been used as a benchmark reaction, either for reducing hydrated electrons.[8j] Other strategies in-
dehalogenation and formation of the corresponding volved the sensitisation of pyrene with [Ru(bpy)3]2+ as
hydrocarbon, or for addition to electron-rich radical visible light photocatalyst,[9] and triplet-triplet annihila-
traps (benzene, pyrrole and similar).[3] This is the case tion for the generation of a high-energy singlet excited
of phenothiazine (PTH)-based photocatalysts,[4] whose state.[10] In most of the aforementioned cases, demon-
use was also recently demonstrated for the activation strations of photocatalytic activity have been restricted
of in situ-formed arylsulfonium ions.[4c] Similarly, to a few benchmark reactions (dehalogenation, addition
dihydrophenazine 1 and phenoxazine 2 were exploited to radical traps) and limited efforts have been
for the synthesis of CF3-bearing heterocycles, and as
Figure 1. Selection of highly reducing photocatalysts and photocatalytic systems from the literature, along with their reported
photophysical properties. Reduction potentials reported vs SCE in CH3CN.[1b,4a,5a,6,7,8b,f]
Adv. Synth. Catal. 2021, 363, 1–15
2
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
dedicated to expand the chemical space of their good choice for applications as organic semiconductors
reactivity.[4c,5]
in transistors,[12] as conducting polymers[13] or prepar-
Considering all the efforts gone to the development ing p-type graphenoid structures.[11,14] PXX itself is a
of highly reducing organic photocatalysts, it is surpris- bright yellow solid, showing absorption in the UV and
ing that a molecule as simple as PXX (peri-xantheno- visible spectrum, with a lower energy peak at 439 nm
xanthene, Figure 2a), the O-doped analogue of anthan- (blue light) in CH3CN (Figure 2b,c).[14a] Besides visible
threne, has been so far disregarded. Structural light absorption, of uttermost interest are the electro-
simplicity, ease of synthesis (see SI) and low cost (raw chemical properties of both PXX ground state and
material cost estimated at £2.30/mmol, see SI) would singlet excited state (Figure 2c). In the S1 state
*
make it an appealing alternative to the more expensive (PXX*), a remarkable reduction potential E1/2(PXX /
+
and structurally complex photocatalysts previously PXX*) of À 2.00 V vs SCE has been calculated,
mentioned.
suggesting the possibility of an efficient SET reduction
The use of PXX is inherited from organic semi- of a number of organic substrates (Figure 3). The
conductor chemistry, as substituted PXX derivatives resulting PXX radical cation has a discrete potential
show excellent carrier-transport and electron injection (+0.77 V), allowing turnover of the photocatalytic
properties, easy processability, chemical inertness and cycle through oxidation of electron-rich reagents. In an
exceptional thermal stability.[11] This made them a effort to probe the photocatalytic properties of PXX,
Figure 2. Properties of PXX: (a) structure and appearance; (b) absorption and emission (λexc at 412 nm) spectra of a 10À 4 M solution
in CH3CN; (c) summary of photophysical and electrochemical properties.[14a] Photophysical data determined in CH3CN,
electrochemical data determined vs SCE in CH3CN.
Figure 3. Reduction potentials of selected organic compounds compared to ground and excited state reduction potentials of PXX.
Values given vs SCE in CH3CN.[15,16]
Adv. Synth. Catal. 2021, 363, 1–15
3
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
°
our group showed that the dehalogenation of a set of carried out at 50 C (entry 1), while a reduced 19%
°
electron-poor aryl and alkyl halides could be achieved yield was obtained at 25 C (entry 2). Extended
upon irradiation with blue light (450 nm) in the reaction times allow improvement in yields only when
presence of DIPEA as electron donor.[14a] These an additional loading of PXX is added (entries 3–4),
preliminary investigations prompted us to further suggesting degradation of the photocatalyst during the
investigate the potentiality of PXX as a more general, reaction.
highly reducing photocatalyst, capable of promoting a
The arylation of N-methylpyrrole proved more
variety of different transformations.[17] The possibility successful: two sets of conditions were found to be
to employ dual-catalytic platforms, wherein PXX productive (Scheme 1). Method A involved the use of
°
photocatalysis could be coupled to another catalytic Cs2CO3 as base in CH3CN at 50 C (see SI for more
system, was also considered in our investigations.
details), and allowed the efficient arylation of 4’-
Herein, we demonstrate the use of PXX as photo- bromoacetophenone and 4-bromobenzaldehyde in 71%
catalyst for the formation of CÀ C bonds via activation and 75% isolated yields, respectively (5b and 5c).
of organic halides and subsequent reaction with radical These conditions performed poorly with other aryl
traps. The β-arylation of cyclic ketones is also shown halide substrates. However, when carrying out the
°
when a secondary amine is introduced as organo- reaction in DMSO with DIPEA as base at 25 C
catalyst. The formation of CÀ N and CÀ S bonds when (Method B),[9] a number of aryl bromide and chloride
coupled with nickel catalysis is also tackled, and the derivatives could be reacted, including those bearing
application to a synthetic intermediate of setipiprant ester (5d) and nitrile (5e, 5h) groups. The activation
through preparation of the corresponding N-aryl of bromopentafluorobenzene was also successful, and
hydrazone precursor demonstrated. The mechanism of compound 5f obtained in 61% yield in a 2.4:1 isomeric
the transformations were further studied by photo- ratio.[18] Heterocyclic bromides were not amenable of
physical and EPR studies to confirm the role of PXX efficient activation, with compounds 5g and 5i
as photoreducing catalyst.
obtained in limited yields. Other radical traps were
also tested: while thiophene did not provide 5j in
useful yields, 1,3,5-trimethoxybenzene proved an
efficient trap affording 5k in 58% yield. The formation
of CÀ Heteroatom bonds could be also probed, with the
reaction with triethylphosphite affording 5l in 66%
yield.[19] The borylation with B2pin2 gave 5m in low
Results and Discussion
Activation of Organic (Pseudo)halides and Addition
to Radical Traps
Our investigations started with an attempt to extend yield (25%);[20] further attempts with modified con-
the observed activation of aryl halides to the formation ditions did not provide any significant improvements.
of CÀ C bonds. Radical traps such as benzene and N-
The activation of electron-poor alkyl halides is also
methylpyrrole were introduced to reaction mixtures feasible with PXX. For example, when diethyl 2-
containing the given aryl halide, PXX and a base. The bromo-2-methylmalonate was reacted with indole in
arylation of benzene with 4’-bromoacetophenone was the presence of lutidine as base, the coupling product
obtained when benzene was used as solvent in the 5n was obtained in 66% yield (Scheme 2a).[21] Analo-
presence of DBU as base (Table 1, see SI for more gously, MacMillan’s arene trifluoromethylation[22]
details). The reaction afforded a 42% yield when could be readily obtained (Scheme 2b) using triflyl
chloride. In the reaction with N-methylpyrrole, both
mono- and bis-trifluoromethylation were observed: 5o
in 33% yield and 5p in 57% yield, respectively. With
3-methylbenzofuran, the main product was 5q (59%,
some was lost during purification due to the high
volatility). Pleasingly, when the same conditions were
applied to the perfluoroalkylation of 3-meth-
ylbenzofuran with perfluorohexyl iodide, an 82% yield
of coupled product, 5r, was obtained.
Table 1. The arylation of benzene with 4’-bromoacetophenone.
#
Modification
–
5a %[a]
Conv. %[a]
6%[a]
The mechanism of these transformations was
subsequently examined. Electrochemical data and
results from fluorescence quenching experiments are
reported in Table 2. Building on the calculated reduc-
1
2
3
4
42
19
40
53
54
26
60
75
8
°
Run at 25 C
< 5
17
18
Run for 40 h
*
+
tive and oxidative properties of PXX* (E1/2(PXX /
Run for 64 h[b]
*
À
PXX*)=À 2.00 V vs SCE; E1/2(PXX*/PXX )=
+0.61 V vs SCE),[14a] we conjectured that PXX* is
capable of photoreducing all the aryl halide substrates
(Table 2) and at the same time to oxidise DIPEA
1
[a] Determined by H-NMR analysis of reaction mixtures, using
1,3,5-trimethoxybenzene or CH3NO2 as internal standard.
[b] Additional 2 mol% PXX added after 40 h.
Adv. Synth. Catal. 2021, 363, 1–15
4
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
Scheme 1. Scope of aryl halides and radical traps in the PXX-catalysed arylation reaction. Method A: radical trap (20 equiv.), PXX
°
(2 mol%), Cs2CO3 (1.2 equiv.), Acetonitrile 0.1 M, 50 C. Method B: radical trap (10 equiv.), PXX (2 mol%), DIPEA (1.4 equiv.),
[a]
[b]
°
DMSO 0.2 M, 25 C. Isolated yields reported unless otherwise stated. Notes: 20 equiv. of 1,3,5-trimethoxybenzene were used.
5 equiv. of triethyl phosphite were used. [c] 3 equiv. of B2pin2 were used, along with a DMSO:H2O 9:1 solvent system.
(from 108 to 1010 MÀ 1 sÀ 1, entries 1–3) are considerably
higher than those determined in CH3CN with DIPEA
(3.5×107 MÀ 1 sÀ 1, entry 4) and N-methylpyrrole, for
which no quenching was observed (entry 5). This
suggests that, under illumination, aryl radicals are
Table 2. Redox potentials and quenching studies on PXX (4×
10À 5 M in CH3CN, intensity of fluorescence recorded at
477 nm).
#
Quencher
E
red (V)[a]
kQ (MÀ 1sÀ 1)[b]
1.3×1010
1.8×1010
3.2×108
3.5×107
N.D.[c]
*
+
1
2
3
4
5
4-Bromoacetophenone
4-Bromobenzaldehyde
2-Chlorobenzonitrile
DIPEA
À 1.89 (r)
À 1.76 (r)
À 1.80 (r)
+0.52 (o)
+1.04 (o)
formed together with PXX , as also reported in a
previous work by us.[14a] Based on the information
gathered, the most likely mechanistic pathway is
outlined in Scheme 3.[3] After photoreduction of the
aryl halide and formation of the corresponding aryl
radical (7), this reacts with suitable unsaturated
systems (radical traps) to form the new CÀ C bond.
N-Methylpyrrole
[a] Determined in DMF (entry 3) or CH3CN (other entries) vs
SCE; (r) refers to a reduction of the quencher, (o) to an
oxidation of the quencher. Obtained from references[15]
(entries 1–2, 4–5) and[24] (entry 3).
*
+
Oxidation of the resulting radical adduct 8 by PXX
and subsequent deprotonation of 9 close the catalytic
cycle.
[b] Entries 1, 2, 4: obtained from reference[14a]. Entries 3, 5:
calculated from Stern Volmer plots as kQ =KSV/τ0 where τ0 is
the lifetime of PXX* without quencher (5.1 ns in
CH3CN).[14a]
Merging Organo- and Photocatalysis: β-Arylation
of Carbonyl Compounds
[c] No quenching observed.
The merging of photoredox catalysis with organo-
catalysis has been among one of the main initiators of
(E1/2ox = +0.52 V vs SCE, entry 5), but not N-methy- the field, with the α-alkylation of aldehydes from
pyrrole (E1/2ox = +1.04 V vs SCE, entry 6). These Nicewicz and MacMillan being the seminal
considerations mirror Stern-Volmer investigations[23] example.[25] Typical organocatalysis intermediates such
which showed that the fluorescence quenching rates as enamines and iminium ions can readily participate
(kq, Table 2) observed for PXX with all aryl halides in redox and radical reactions, either in conjunction
Adv. Synth. Catal. 2021, 363, 1–15
5
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
saturated carbonyl compounds (aldehydes and cyclic
ketones).[27] While the α position of carbonyl deriva-
tives is easily accessible from the formation of enolate
or enamine intermediates, limited examples were
reported for the β functionalisation without using α,β-
unsaturated compounds in the first place. In particular,
a β-arylation reaction was carried out with fac-
[Ir(ppy)3] as photocatalyst and electron-deficient ben-
zonitrile derivatives as reactants.[27a] When the reaction
of cyclohexanone with 1,4-dicyanobenzene was tested
in the same conditions using PXX as photocatalyst
(Scheme 4), β-arylated product 10a was obtained in
79% yield.
Applying the reaction conditions to other ketones,
β-arylated derivatives could be obtained in 60–72%
yields with different cyclohexanone precursors (10b–
c). Cyclopentanone 10d was also obtained in 54%
yield, while cycloheptanone 10e could not be recov-
ered in good amounts (less than 10% yield by NMR).
When turning to other cyanoarenes, only 4-cyanoph-
thalide afforded 10g in a modest 35% yield. Other
substrates provided limited or no conversion (10f, 10i)
or complex reaction mixtures (10h).
Scheme 2. (a) Reaction of a bromomalonate substrate with
indole as radical trap. (b) Perfluoroalkylation of heteroarenes.
19F-NMR yields (using C6F6 as internal standard) are provided
for the trifluoromethylation products due to volatility of the
products, isolated yield reported for 5r.
Scheme 3. Postulated mechanism for the PXX-photocatalysed
addition of aryl radicals to radical traps (in this case, a 5-
membered heterocycle).
Scheme 4. Scope of the PXX-catalysed β-arylation of cyclic
ketones with cyanoarenes. Isolated yields reported unless
otherwise stated.
with a photocatalyst or by exploitation of their intrinsic
photoreactivity.[26] In 2013–2015, the group of Mac-
Millan reported a series of β-functionalisations of
Adv. Synth. Catal. 2021, 363, 1–15
6
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
The mechanism of this transformation was further reduce the cyanoarene to its radical anion 13, while the
*
+
studied by photophysical and EPR studies. Stern- resulting PXX cation (E= +0.77 V vs SCE) would
Volmer fluorescence quenching experiments (Table 3) be capable to oxidise enamine 12 to its radical cation
revealed that all four electron-poor cyanoarenes exam- 14, reforming PXX. In the latter cycle the parts would
ined (entries 1–4) to be efficient quenchers of PXX*, be inverted, with PXX* oxidising enamine 12 to 14,
*
À
with quenching constant values in the order of and the resulting highly reducing PXX (E=À 2.16 V
1010 MÀ 1 sÀ 1. This result comes as no surprise, as all vs SCE) reducing 11a to 13. In either case, the so-
their reduction potentials are accessible by PXX* (E1/2 formed enamine radical cation 14 would undergo
*
+
(PXX /PXX*)=À 2.0 V vs SCE). With cyclohexa- deprotonation by DABCO in the β position to form 5π-
none (entry 5) and azepane (entry 6) no fluorescence enaminyl radical 15. Subsequent steps include radical-
quenching was observed. However, a limited quench- radical coupling between 15 and 13 to give 16, loss of
ing was detected when cyclohexanone, azepane and a cyanide to give enamine 17 and final hydrolysis to
acetic acid where combined in a single solution. We the product (10a) while regenerating the
thus prepared and analysed enamine 12 as quencher organocatalyst.[27a]
(entry 7). Notably, a high quenching constant was
To further support the mechanistic propositions,
obtained (1.5×1010 MÀ 1 sÀ 1), of the same order of EPR measurements were performed to identify the
magnitude as that obtained with the cyanoarenes (1.2– presence of the relevant radicals (Figure 4). The
1.6×1010 MÀ 1 sÀ 1). DABCO is also a possible quench- reaction with 1,4-dicyanobenzene 11a and cyclohex-
er, but with a two orders of magnitude lower rate (2.2× anone yielding product 10a was performed in the
108 MÀ 1 sÀ 1, entry 8).
Taken all together, these results suggest both an
oxidative photocatalytic mechanism (Scheme 5a, right
side), reliant on the photoreduction of cyanoarenes,[27a]
and a reductive photocatalytic cycle (Scheme 5a, left
side), based on the photooxidation of 12 and/or
DABCO.[27d] In the former mechanism, PXX* would
Table 3. Redox potentials and quenching studies on PXX (4×
10À 5 M in CH3CN, intensity of fluorescence recorded at
477 nm).
#
Quencher
Ered (V)[a]
kQ (MÀ 1sÀ 1)[d]
1
2
3
4
5
6
7
8
11a
11b
11c
11d
Cyclohexanone
Azepane
12
À 1.64 (r)
1.6×1010
1.2×1010
1.6×1010
1.5×1010
N.D.[e]
À 1.90 (r)
À 1.59 (r)
À 1.53 (r)
À 2.33 (r)
ffi+0.9 (o)[b]
+0.3–+0.5 (o)[c]
+0.69 (o)
N.D.[e]
1.4×1010
2.2×108
DABCO
[a] Determined in CH3CN; (r) refers to a reduction of the
quencher, (o) to an oxidation of the quencher. Entries 3 and 4:
experimentally determined (see SI). Other entries: obtained
from reference.[15]
Figure 4. Continuous wave EPR spectra of a reaction medium
composed of 1,4-dicyanobenzene, cyclohexanone, azepane,
PXX, AcOH and CH3CN with the addition of PBN spin trap
measured at rt and (a) before light irradiation, (b) after 30
minutes of irradiation at 450 nm; (c) reaction medium without
1,4-dicyanobenzene after ca. 2 hours of irradiation at 450 nm;
(d) reaction medium without cyclohexanone, azepane and acetic
acid after 30 minutes of irradiation at 450 nm. Below: structure
of PBN-trapped radicals 18a and 18b.
[b] Typical values for cyclic secondary amines.
[c] Typical values for enamines of cyclic secondary amines.
[d] Calculated from Stern Volmer plots as kQ =KSV/τ0 where τ0 is
the lifetime of PXX* without quencher (5.1 ns in
CH3CN).[14a]
[e] No quenching observed.
Adv. Synth. Catal. 2021, 363, 1–15
7
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
Scheme 5. Proposed mechanisms for the PXX-photocatalysed β-arylation of cyclic ketones: (a) reductive and oxidative quenching
mechanistic cycles; (b) downstream pathway with formation of the product and regeneration of the organocatalyst.
presence of N-tert-butyl-α-phenylnitrone (PBN) as a same trapped radical species (Figure 4c). On the
radical spin trap.
contrary, when cyclohexanone and azepane were
The reaction mixture was measured before and after removed from the reaction medium, no paramagnetic
30 minutes of irradiation at 450 nm (Figure 4). Whilst species were detected in continuous wave mode (Fig-
before irradiation no detectable EPR signal was ure 4d).
observed (Figure 4a), after 30 minutes of irradiation a
This evidence seems to indicate that the PBN-
clear triplet of doublets pattern arose (Figure 4b). trapped radical species (signals R1- and R2-PBN) both
Simulation of the signal in Figure 4b highlighted the arise from the combination of azepane and cyclo-
presence of two PBN-trapped radical species (R1-PBN hexanone. Given the systematic lack in the literature of
and R2-PBN), characterized by the following spin comparative examples about PBN-trapped cyclohexa-
Hamiltonian parameters: R1-PBN: giso =2.006, aiso- noyl radicals in CH3CN, we resort to DFT to assign
(14N)=1.47 mT, aiso(1Hβ)=0.36 mT (spectral contribu- the signals of radicals R1-PBN and R2-PBN.The
tion 23 %) and R2-PBN: giso =2.006, aiso(14N)= structure of 3-cyclohexanoyl PBN adduct 18b as well
1.51 mT, aiso(1Hβ)=0.18 mT (spectral contribution 77 as its enamino form 18a, derived from addition of
%). Cross checks have been performed by removing PBN to 5π-enaminyl radical 15 (Figure 4), were first
each of the reaction substrates (i.e., either 1,4- geometry optimized at the B3LYP/def2-TZVP level of
dicyanobenzene or cyclohexanone + azepane) to prove theory (SI). Spin Hamiltonian parameters were calcu-
that the trapped radical species observed do indeed lated on the optimized structures at the PBE0/EPR-II
come from the reaction and are not byproducts from level of theory, i.e. a commonly used combination of
PBN decomposition under irradiation or other species functional and basis set for the light elements such as
present in the reaction medium. When 1,4-dicyanoben- H and B–F. These calculations returned the following
zene was removed from the reaction medium, another spin Hamiltonian parameters; PBN-trapped cyclohex-
spectrum like that in Figure 4b was detected, with the anoyl adduct 18b: giso =2.006, aiso(14N)=1.12 mT,
Adv. Synth. Catal. 2021, 363, 1–15
8
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
aiso(1Hβ)=0.37 mT; PBN-trapped cyclohexanoyl ad- including BET, are much faster than the trapping
duct enamino form 18a: giso =2.006, aiso(14N)= reaction with PBN and that none of the dicyanoben-
1
1.03 mT, aiso(1Hβ)=0.11 mT. Theoretical Hβ isotropic zene radicals live long enough to be seen as a free, not
hyperfine coupling constants are significantly different trapped species over the time scales of the continuous
between the two optimized structures, and almost wave EPR experiment.
completely match the values measured for the two
experimental signals, R1-PBN and R2-PBN, of
Nickel-PXX Dual Catalysis: Cross-Coupling Reac-
0.36 mT and 0.18 mT. As to the 14N isotropic hyperfine
tions
coupling constants, the theoretical values are generally
significantly lower than the experimental values shown In recent years, nickel catalysis has seen renewed
by radicals R1-PBN and R2-PBN; however, the differ- interest for the possibility to replace catalysts based on
ence between the two values is much more contained, more precious metals and the development of new
i.e. 0.09 mT, resembling the 0.04 mT separating the cross-couplings reactions,[29] often in combination with
1
experimental values (on the contrary the Hβ are much photoredox catalytic manifolds.[30] Examples include
further apart, i.e. 0.26 mT for the theoretical values cross-couplings using Ni-photoredox dual catalysis to
and 0.18 mT for the experimental values). Based on form C-heteroatom bonds:[31] CÀ N,[5,32] CÀ O,[33]
the comparison between theory and experiments, one CÀ S[5a,16,34] and C-P[35]. Considering the chemical
could assign the R1-PBN signal to the PBN-trapped stability and low cost of PXX, we envisaged to use it
cyclohexanoyl radical (18b) and the R2-PBN signal to as the photocatalyst partner in the Ni-catalysed cross-
the enamino form of the PBN-trapped cyclohexanoyl coupling reaction to form C-heteroatom bonds. At first,
radical (18a), unambiguously providing evidence for we have explored the reaction of aryl bromides with
the formation of radical 15 in the reaction mixture.
secondary amines to form CÀ N bonds. Applying a
EPR also allowed to ascertain further aspects of the standard set of conditions[32a] with NiBr2·DME as Ni
reaction mechanism. Fluorescence quenching studies source, DABCO as base and PXX as photocatalyst
suggest that both oxidative and reductive mechanisms revealed a very efficient reaction, providing quantita-
(Scheme 5a) are possible, and in the case of the tive yields of 19a in the reaction of pyrrolidine with 4-
reaction with all the components (Figure 4b) both bromobenzotrifluoride (Table 4).
mechanisms fit the EPR results. Considering the high
value of the fluorescence quenching constant and the
high concentration of the cyanoarenes in solution, the
Table 4. Screening of conditions for the PXX-photocatalysed
oxidative photocatalytic cycle seems favoured. How-
Ni-photoredox cross-coupling of pyrrolidine and 4-bromoben-
ever, the fact that an almost identical EPR spectrum is
zotrifluoride.
obtained in the absence of 1,4-dicyanobenzene (Fig-
ure 4c) would suggest that a reductive quenching cycle
is in operation. While reduction of cyanoarenes by
PXX* is possible, the fate of the resulting radical ion
pair might not include reaction with enamine 12 and
eventual delivery of the product, as other processes can
occur. For example, upon irradiation cyanoarenes are
known to form complexes with electron donor mole-
cules (D) such as pyrenes.[28] These complexes are
#
Modification
Yield %[a] Conv. %[a]
usually in equilibrium with the spin-correlated radical
*
*
À
ion pair [D + ·cyanoarene ] following single electron
transfer. Back electron transfer (BET) could then
occur, with the result of quenching of the radicals, a
process that is facilitated by CH3CN when used as a
solvent.[28] A similar behaviour can be postulated for
the interaction of cyanoarenes with PXX as an electron
donor. Further experiments involving time-resolved
EPR and other spin-chemistry techniques, which are
beyond the scope of this paper, will be used to quantify
kinetic aspects of the interaction of PXX with
cyanoarenes. Nonetheless, under our reaction condi-
tions the complete absence of EPR signals when
cyclohexanone and azepane are not added to the
reaction medium (Figure 4d) seems to suggest that the
kinetics involving the dicyanobenzene radical anion,
1
2
3
4
5
6
7
8
9
4 mol% PXX
none (2 mol% PXX)
1 mol% PXX
0.5 mol% PXX
0.2 mol% PXX
0.1 mol% PXX
0.2 mol% PXX, 10 mol% Ni 99
0.2 mol% PXX, 1 mol% Ni
NiCl2 DME as Ni source
97
94
99
99
99 (87)
96
100
100
100
100
100
100
100
97
96
95
95
79
13
100
100
98
10 DMF as solvent
11 Acetonitrile as solvent
12 No PXX
13
[a] Determined by H NMR analysis of crude products, using
1,3,5-trimethoxybenzene as internal standard. Isolated yield
in parentheses.
1
Adv. Synth. Catal. 2021, 363, 1–15
9
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
The reaction was robust as small variations were reaction is more difficult, and 19j was obtained in
well tolerated. For example, the PXX loading could be 26% yield only by using 4-anisyl iodide in place of the
varied from 2 mol% to 0.1 mol% without losses in bromide congener.
efficiency (entries 1–6). Similarly, Ni loading can also
The good reactivity found with benzophenone
be reduced to 1 mol% with only a minor reduction in hydrazone was particularly intriguing. Products such as
yield (entry 8). Alternative Ni salts and solvents also 19f can be regarded as masked hydrazines, important
provided efficient reactions (entries 9–10) with the substrates for the formation of indole rings through a
exception being CH3CN, still giving 79% yield Fisher indole synthesis reaction.[37] When coupling
(entry 11). Lastly, in absence of PXX only minor conditions were applied to 4-bromofluorobenzene and
reactivity is observed (13% yield, entry 12), in line benzophenone hydrazone, N-arylhydrazone 19k was
with the hypothesis of background photoactivity of the obtained with 95% NMR yield (Scheme 7). While 19k
Ni complexes advanced by Miyake and co-workers.[36] was found to be prone to hydrolysis (both through
The conditions found above were thus applied to a aqueous washings and column chromatography on
range of aryl halides and amine reaction partners silica gel), the reaction mixture after evaporation of the
(Scheme 6). Secondary amines efficiently deliver solvent can be directly subjected to Fisher indole
coupled products 19a–c (79–87% yield), with reac- synthesis conditions with an enolisable ketone to form
tions almost devoid of impurities. Unfortunately, fluoro-tetrahydro-1H-pyrido[4,3-b]indole 20 in 68%
primary amines showed only limited reactivity (19d–e, yield (over two steps). Indole derivative 20 is a crucial
20% yield by NMR), and every attempt to increase the intermediate for the investigational drug setipiprant,
yield (for example, by heating the reaction mixture to under study for treating asthma and scalp hair loss.[38]
°
50 C) did not provide any improvement. Pleasingly, Considering that the reaction was performed on a
benzophenone hydrazone is also a competent substrate, synthetically useful mmol scale, it demonstrates the
and derivative 19f was obtained in 72% yield. possibility to scale this reaction up in a laboratory
Guanidine 19g was obtained in limited yields (11%), setting.
and also in this case attempts to improve them were
The attention was then turned to the CÀ S bond-
not fruitful. Examining other aryl bromides, electron- forming cross-coupling reaction. Under conditions
poor substrates provide products in good yields (19h– analogous to those suggested by Oderinde and co-
i, 69–77% yield), while for electron-rich ones the workers,[16] the reaction between 4-tolylbromide and
octanethiol was found to be difficult (Table 5, entries
1–2). By replacing the former with 4-tolyl iodide the
reactivity changed dramatically, and the coupled
product 21a could be obtained quantitatively (entry 3).
It was possible to reduce the loading of PXX and Ni
pre-catalyst to 0.2 mol% and 5 mol% respectively
Scheme 6. Scope of the Ni-photoredox CÀ N cross-coupling.
1
Isolated yields are reported, unless otherwise stated. H NMR
yields are determined using 1,3,5-trimethoxybenzene as internal
[a]
[b]
standard. Notes: Reaction time: 64 h. 4-Anisyl iodide used
as substrate.
Scheme 7. Application of the Ni-photoredox CÀ N cross-cou-
pling to 20, a synthetic intermediate of setipiprant.
Adv. Synth. Catal. 2021, 363, 1–15
10
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
12c was obtained in 93% yield using 4-bromobenzoni-
trile. Examining other thiols, a more complex system
as found in a protected cysteine is still amenable to
reaction: a 85% yield of 21d was obtained after 40 h.
An aryl thiol in thiophenol is also a viable substrate,
and 21e was isolated in 92% yield.
Table 5. Screening of conditions for the PXX-photocatalysed
Ni-photoredox cross-coupling of 4-halotoluenes and octane-
thiol.
The mechanistic understanding of these cross-
couplings typically rely on the possibility to control
the oxidation state of the Ni centre by the photo-
catalyst, which would circumvent difficult steps such
as the reductive elimination of C-heteroatom bonds
from a catalytic Ni(II) complex intermediate.[16,32a,39]
Further studies have revealed that energy transfer
mechanistic manifolds are also possible, extending the
range of photocatalysts that can be applied to these
transformations as different photophysical properties
are relevant depending on the mechanism.[5b,32c,33b,40]
When considering the CÀ N cross-coupling, electro-
chemical considerations make us believe that PXX acts
in controlling the oxidation state of the metal. Recent
studies from MacMillan and co-workers revealed a
Ni(I)/Ni(III) pathway to be operating, where the photo-
catalyst acts as an initiator by reducing key Ni(II)-
amino complexes (resting state) to Ni(I).[39c] The high
#
X
Modification
Yield %[a]
1
2
3
4
5
6
7
8
9
10
Br
Br
I
I
I
I
I
I
I
none
0.2 mol% PXX
none
0.4 mol% PXX
0.2 mol% PXX
0.2 mol% PXX, 5 mol% Ni/L
NiBr2 DME as Ni source
DMF as solvent
Acetonitrile as solvent
No PXX
23
0
> 99
> 99
98
> 99 (93)
99
> 99
> 99
0
I
[a] Determined by H NMR analysis of crude products, using
1,3,5-trimethoxybenzene as internal standard. Isolated yield
in parentheses.
1
*
À
reducing power of PXX* and PXX (À 2.00 V and
À 2.16 V, respectively) well matches the requirements
found by MacMillan for the iridium photocatalysts
(entries 4–6). Changing the Ni pre-catalyst or solvent (E<À 1.26 V vs SCE), whereas the oxidising poten-
*
+
(DMF, or CH3CN) did not provide any loss in tials of PXX* and PXX
(+0.61 and +0.77,
reactivity (entries 7–9). In the absence of PXX, the respectively) are high enough to oxidise DABCO, as
reaction does not proceed at all (entry 10). suggested by our quenching studies and electrochem-
With the conditions found in entry 6, a brief scope ical considerations (Table 3, entry 8). Taking all this
of the transformation was examined (Scheme 8). Aryl together, we hypothesized the catalytic cycle depicted
iodides bearing elecron-rich groups worked well (21a– in Scheme 9. Once the Ni(I) species 23 is formed by
b, 93–98% yield) and in the case of an electron-poor reduction of Ni(II) (22), oxidative addition of the aryl
system the aryl bromide could be used as substrate: bromide generates an aryl-Ni(III) species 24 that can
Scheme 8. Scope of the Ni-photoredox CÀ S cross-coupling.
Isolated yields are reported. Notes: [a] 4-Bromobenzonitrile used
as substrate. [b] Reaction time: 40 h.
Scheme 9. Postulated mechanism for the PXX- and Ni-cata-
lysed CÀ N cross-coupling reaction.
Adv. Synth. Catal. 2021, 363, 1–15
11
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
organic radicals can then add to suitable radical traps
to form new CÀ C bonds. Turning the attention to more
complex catalytic manifolds, PXX photocatalysis has
been successfully coupled with both organocatalysis
and metallocatalysis. In the first instance, a β-arylation
of cyclic ketones with cyanoarenes was shown to be
productive. A range of mechanistic experiments have
revealed that a reductive mechanism cycle seems more
probable than the normally-accepted oxidative catalytic
cycle. It was also possible to identify adducts of the
radical trap PBN with potential radical intermediates.
Some Ni-catalysed cross-couplings were also promoted
by PXX-based photocatalysis, allowing the formation
of CÀ N and CÀ S bonds from aryl halides and
demonstrating effective partnership with metallocatal-
ysis. To the best of our knowledge, this work is one of
the few examples of wide demonstrations of reactivity
for a new highly reducing organic photocatalyst, thus
adding another precious tool to the array of photo-
catalysts to be chosen for expanding the chemical
space of photoredox reactions. Compared to other
photocatalysts, PXX stands out for its very low
reduction potential of the singlet excited state (À 2.0 V
vs SCE) coupled to structural simplicity, low cost and
ease of preparation (see SI). Further studies are
Scheme 10. Postulated mechanism for the PXX- and Ni-
catalysed CÀ S cross-coupling reaction.
undergo reductive elimination to furnish the CÀ N bond underway to expand the application of PXX-based
(product 19). Turnover of the photocatalyst can be photocatalysis to the preparation of other classes of
achieved by oxidation of DABCO. A similar reductive compounds.
quenching cycle with initial oxidation of DABCO by
PXX* could also be considered.
Experimental Section
PXX might take a similar role in the mechanism of
Experimental procedures and characterisation data are reported
on the supporting information (PDF). Geometry optimisation
data for adduct 18a (XYZ).Geometry optimisation data for
adduct 18b (XYZ). EPR computational data for adduct 18a
the CÀ S cross-coupling reaction as well. Building on
the studies performed by Oderinde and co-workers,[16]
we postulate the possibility of reduction of Ni(II) to
catalytically-competent Ni(I) by PXX*. Moreover,
*
+
(OUT). EPR computational data for adduct 18b (OUT).
oxidation of the thiol is a likely step, which PXX
(+0.77 V vs SCE) seems to be able to achieve (e.g.,
benzylthiol has an oxidation potential of +0.83 V vs
Acknowledgements
SCE in CH3CN).[16] Based on these considerations, a
The authors thank Hugo Morren (Dr. Reddy’s) and Tommaso
Battisti (Cardiff University) for important analytical support.
D.B. and R.D. thank the EU and H2020-MSCA-ITN-2016
possible mechanism is depicted in Scheme 10. PXX*
effects the necessary reduction of Ni(II) complexes
such as 25 to Ni(I) (26), enabling the oxidative
addition of the aryl bromide to form an aryl-Ni(III)
complex 27, which undergoes reductive elimination to
form the aryl sulfide 21. Generated PXX oxidises
the thiol, forming after deprotonation thiyl radical 29,
which can be intercepted by the Ni(I) centre (28 to 25).
°
(project n 722591–PHOTOTRAIN) for the generous funding.
C.P. and O.M. thank the EU through the PHOTOTRAIN project
for a pre-doctoral fellowship. D.B. thanks the University of
Vienna for financial support.
*
+
References
Conclusion
[1] a) J. W. Tucker, C. R. J. Stephenson, J. Org. Chem.
2012, 77, 1617–1622; b) C. K. Prier, D. A. Rankic,
D. W. C. MacMillan, Chem. Rev. 2013, 113, 5322–5363;
c) D. Ravelli, M. Fagnoni, A. Albini, Chem. Soc. Rev.
2013, 42, 97–113; d) T. Koike, M. Akita, Inorg. Chem.
Front. 2014, 1, 562–576; e) M. H. Shaw, J. Twilton,
D. W. C. MacMillan, J. Org. Chem. 2016, 81, 6898–
6926; f) N. A. Romero, D. A. Nicewicz, Chem. Rev.
In conclusion, the ability of PXX as photocatalyst to
promote a wide variety of radical reactivities was
demonstrated, in particular in reactions typical of the
more common yet expensive Ir(III) complexes. The
relatively high reduction potential of its excited state
allows reduction of aryl, alkyl and perfluoroalkyl
halides (and chlorides in particular); the resulting
Adv. Synth. Catal. 2021, 363, 1–15
12
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
2016, 116, 10075–10166; g) J.-P. Goddard, C. Ollivier, [11] T. Miletić, A. Fermi, I. Orfanos, A. Avramopoulos, F.
L. Fensterbank, Acc. Chem. Res. 2016, 49, 1924–1936.
[2] a) K. A. King, P. J. Spellane, R. J. Watts, J. Am. Chem.
Soc. 1985, 107, 1431–1432; b) J. D. Nguyen, E. M. D.
De Leo, N. Demitri, G. Bergamini, P. Ceroni, M. G.
Papadopoulos, S. Couris, D. Bonifazi, Chem. Eur. J.
2017, 23, 2363–2378.
Amato, J. M. R. Narayanam, C. R. J. Stephenson, Nat. [12] a) N. Kobayashi, M. Sasaki, K. Nomoto, Chem. Mater.
Chem. 2012, 4, 854–859; c) Y. Cheng, X. Gu, P. Li, Org.
Lett. 2013, 15, 2664–2667; d) H. W. Shih, M. N.
Vander Wal, R. L. Grange, D. W. C. MacMillan, J. Am.
Chem. Soc. 2010, 132, 13600–13603.
2009, 21, 552–556; b) N. Kobayashi, M. Sasaki, T. Ohe,
US Patent 2012/0025173A1, 2012; c) H. Li, F. Zhang, S.
Qiu, N. Lv, Z. Zhao, Q. Li, Z. Cui, Chem. Commun.
2013, 49, 10492–10494; d) N. Lv, M. Xie, W. Gu, H.
Ruan, S. Qiu, C. Zhou, Z. Cui, Org. Lett. 2013, 15,
2382–2385.
[3] a) I. Ghosh, L. Marzo, A. Das, R. Shaikh, B. König, Acc.
Chem. Res. 2016, 49, 1566–1577; b) M. Majek, A. J.
Von Wangelin, Acc. Chem. Res. 2016, 49, 2316–2327.
[4] a) E. H. Discekici, N. J. Treat, S. O. Poelma, K. M.
[13] C. Song, T. M. Swager, Macromolecules 2009, 42, 1472–
1475.
Mattson, Z. M. Hudson, Y. Luo, C. J. Hawker, J. R. [14] a) A. Sciutto, A. Fermi, A. Folli, T. Battisti, J. M.
De Alaniz, Chem. Commun. 2015, 51, 11705–11708;
b) S. O. Poelma, G. L. Burnett, E. H. Discekici, K. M.
Mattson, N. J. Treat, Y. Luo, Z. M. Hudson, S. L.
Shankel, P. G. Clark, J. W. Kramer, C. J. Hawker, J. R.
De Alaniz, J. Org. Chem. 2016, 81, 7155–7160; c) M. H.
Aukland, M. Šiaučiulis, A. West, G. J. P. Perry, D. J.
Procter, Nat. Catal. 2020, 3, 163–169.
Beames, D. M. Murphy, D. Bonifazi, Chem. Eur. J. 2018,
24, 4382–4389; b) A. Sciutto, A. Berenzin, M. Lo Ci-
cero, T. Miletic, A. Stopin, D. Bonifazi, J. Org. Chem.
2018, 83, 13787–13798; c) D. Stassen, N. Demitri, D.
Bonifazi, Angew. Chem. Int. Ed. 2016, 55, 5947–5951;
Angew. Chem. 2016, 128, 6051–6055. d) A. Berezin, N.
Biot, T. Battisti, D. Bonifazi, Angew. Chem. Int. Ed.
2018, 57, 8942–8946; Angew. Chem. 2018, 130, 9080–
9084.
[5] a) Y. Du, R. M. Pearson, C.-H. Lim, S. M. Sartor, M. D.
Ryan, H. Yang, N. H. Damrauer, G. M. Miyake, Chem.
Eur. J. 2017, 23, 10962–10968; b) M. Kudisch, C.-H. [15] H. G. Roth, N. A. Romero, D. A. Nicewicz, Synlett 2016,
Lim, P. Thordarson, G. M. Miyake, J. Am. Chem. Soc.
2019, 141, 19479–19486.
[6] L. Wang, J. Byun, R. Li, W. Huang, K. A. I. Zhang, Adv.
Synth. Catal. 2018, 360, 4312–4318.
27, 714–723.
[16] M. S. Oderinde, M. Frenette, D. W. Robbins, B. Aquila,
J. W. Johannes, J. Am. Chem. Soc. 2016, 138, 1760–
1763.
[7] R. Matsubara, T. Yabuta, U. Md Idros, M. Hayashi, F. [17] During the preparation of this manuscript, Dilman and
Ema, Y. Kobori, K. Sakata, J. Org. Chem. 2018, 83,
9381–9390.
[8] a) I. Ghosh, T. Ghosh, J. I. Bardagi, B. König, Science
2014, 346, 725–728; b) I. Ghosh, B. König, Angew.
co-workers reported the use of PXX as photocatalyst for
the iododifluoromethylation of alkenes: A. L. Trifonov,
L. I. Panferova, V. V. Levin, V. A. Kokorekin, A. D.
Dilman, Org. Lett. 2020, 22, 2409–2413.
Chem. Int. Ed. 2016, 55, 7676–7679; Angew. Chem. [18] A. U. Meyer, T. Slanina, C.-J. Yao, B. König, ACS Catal.
2016, 128, 7806–7810; c) J. I. Bardagi, I. Ghosh, M. 2016, 6, 369–375.
Schmalzbauer, T. Ghosh, B. König, Eur. J. Org. Chem. [19] R. S. Shaikh, S. J. S. Düsel, B. König, ACS Catal. 2016,
2018, 34–40; d) M. Neumeier, D. Sampedro, M. Májek, 6, 8410–8414.
V. A. de la Peña O’Shea, A. J. von Wangelin, R. Pérez- [20] a) M. Jiang, H. Yang, H. Fu, Org. Lett. 2016, 18, 5248–
Ruiz, Chem. Eur. J. 2018, 24, 105–108; e) M. Schmalz-
bauer, I. Ghosh, B. König, Faraday Discuss. 2019, 215,
364–378; f) N. G. W. Cowper, C. P. Chernowsky, O. P.
5251; b) A. Nitelet, D. Thevenet, B. Schiavi, C.
Hardouin, J. Fournier, R. Tamion, X. Pannecoucke, P.
Jubault, T. Poisson, Chem. Eur. J. 2019, 25, 3262–3266.
Williams, Z. K. Wickens, J. Am. Chem. Soc. 2020, 142, [21] E. C. Swift, T. M. Williams, C. R. J. Stephenson, Synlett
2093–2099; g) I. A. MacKenzie, L. Wang, N. P. R. 2016, 27, 754–758.
Onuska, O. F. Williams, K. Begam, A. M. Moran, B. D. [22] D. A. Nagib, D. W. C. MacMillan, Nature 2011, 480,
Dunietz, D. A. Nicewicz, Nature 2020, 580, 76–80; h) A. 224–228.
Graml, T. Neveselý, R. J. Kutta, R. Cibulka, B. König, [23] S. P. Pitre, C. D. McTiernan, J. C. Scaiano, Acc. Chem.
Nat. Commun. 2020, 11, Article number: 3174; i) H. Res. 2016, 49, 1320–1330.
Kim, H. Kim, T. H. Lambert, S. Lin, J. Am. Chem. Soc. [24] D. E. Bartak, K. J. Houser, B. C. Rudy, M. D. Hawley, J.
2020, 142, 2087–2092; j) C. Kerzig, X. Guo, O. S.
Wenger, J. Am. Chem. Soc. 2019, 141, 2122–2127.
[9] I. Ghosh, R. S. Shaikh, B. König, Angew. Chem. Int. Ed.
Am. Chem. Soc. 1972, 94, 7526–7530.
[25] D. A. Nicewicz, D. W. C. MacMillan, Science 2008, 322,
77–80.
2017, 56, 8544–8549; Angew. Chem. 2017, 129, 8664– [26] M. Silvi, P. Melchiorre, Nature 2018, 554, 41–49.
8669.
[27] a) M. T. Pirnot, D. A. Rankic, D. B. C. Martin, D. W. C.
MacMillan, Science 2013, 339, 1593–1596; b) F. R.
Petronijevic, M. Nappi, D. W. C. MacMillan, J. Am.
Chem. Soc. 2013, 135, 18323; c) J. A. Terrett, M. D.
Clift, D. W. C. MacMillan, J. Am. Chem. Soc. 2014, 136,
[10] a) M. Majek, U. Faltermeier, B. Dick, R. Pérez-Ruiz,
A. J. von Wangelin, Chem. Eur. J. 2015, 21, 15496–
15501; b) B. D. Ravetz, A. B. Pun, E. M. Churchill,
D. N. Congreve, T. Rovis, L. M. Campos, Nature 2019,
565, 343–346.
Adv. Synth. Catal. 2021, 363, 1–15
13
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
asc.wiley-vch.de
6858–6861; d) J. L. Jeffrey, F. R. Petronijevic, D. W. C.
MacMillan, J. Am. Chem. Soc. 2015, 137, 8404–8407.
[28] S. Narra, Y. Nishimura, H. A. Witek, S. Shigeto,
b) E. R. Welin, C. Le, D. M. Arias-Rotondo, J. K.
McCusker, D. W. C. MacMillan, Science 2017, 355,
380–385.
ChemPhysChem 2014, 15, 2945–2950 and references [34] M. Jouffroy, C. B. Kelly, G. A. Molander, Org. Lett.
cited therein. 2016, 18, 876–879.
[29] C. C. C. Johansson Seechurn, M. O. Kitching, T. J. Cola- [35] J. Xuan, T. T. Zeng, J. R. Chen, L. Q. Lu, W. J. Xiao,
cot, V. Snieckus, Angew. Chem. Int. Ed. 2012, 51, 5062–
5085; Angew. Chem. 2012, 124, 5150–5174.
[30] a) J. Twilton, C. C. Le, P. Zhang, M. H. Shaw, R. W.
Chem. Eur. J. 2015, 21, 4962–4965.
[36] C. H. Lim, M. Kudisch, B. Liu, G. M. Miyake, J. Am.
Chem. Soc. 2018, 140, 7667–7673.
Evans, D. W. C. Macmillan, Nat. Rev. 2017, 1, Article [37] S. Wagaw, B. H. Yang, S. L. Buchwald, J. Am. Chem.
number: 0052; b) L. N. Cavalcanti, G. A. Molander, Top.
Curr. Chem. 2016, 374, 39.
[31] C. Cavedon, P. H. Seeberger, B. Pieber, Eur. J. Org.
Chem. 2020, 1379–1392.
Soc. 1998, 120, 6621–6622.
[38] H. Fretz, A. Valdenaire, J. Pothier, K. Hilpert, C. Gnerre,
O. Peter, X. Leroy, M. A. Riederer, J. Med. Chem. 2013,
56, 4899–4911.
[32] a) E. B. Corcoran, M. T. Pirnot, S. Lin, S. D. Dreher, [39] a) Z.-H. Qi, J. Ma, ACS Catal. 2018, 8, 1456–1463;
D. A. DiRocco, I. W. Davies, S. L. Buchwald, D. W. C.
Macmillan, Science 2016, 353, 279–283; b) M. S. Oder-
inde, N. H. Jones, A. Juneau, M. Frenette, B. Aquila, S.
Tentarelli, D. W. Robbins, J. W. Johannes, Angew. Chem.
b) R. Sun, Y. Qin, S. Ruccolo, C. Schnedermann, C.
Costentin, D. G. Nocera, J. Am. Chem. Soc. 2019, 141,
89–93; c) N. A. Till, L. Tian, G. D. Scholes, D. W. C.
MacMillan, J. Am. Chem. Soc. 2020, 142, 15830–15841.
Int. Ed. 2016, 55, 13219–13223; Angew. Chem. 2016, [40] a) L. Tian, N. A. Till, B. Kudisch, D. W. C. MacMillan,
128, 13413–13417; c) T. Kim, S. McCarver, C. Lee,
D. W. C. MacMillan, Angew. Chem. Int. Ed. 2018, 57,
3488–3492; Angew. Chem. 2018, 130, 3546–3550.
G. D. Scholes, J. Am. Chem. Soc. 2020, 142, 4555–4559;
b) P. Ma, S. Wang, H. Chen, ACS Catal. 2020, 10, 1–6;
c) J. Lu, B. Pattengale, Q. Liu, S. Yang, W. Shi, S. Li, J.
Huang, J. Zhang, J. Am. Chem. Soc. 2018, 140, 13719–
13725.
[33] a) J. A. Terrett, J. D. Cuthbertson, V. W. Shurtleff,
D. W. C. Macmillan, Nature 2015, 524, 330–334;
Adv. Synth. Catal. 2021, 363, 1–15
14
© 2021 The Authors. Advanced Synthesis & Catalysis
published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
peri-Xanthenoxanthene (PXX): a Versatile Organic Photoca-
talyst in Organic Synthesis
Adv. Synth. Catal. 2021, 363, 1–15
C. Pezzetta, A. Folli, O. Matuszewska, D. Murphy, R. W. M.
Davidson*, D. Bonifazi*