Direct Photoexcitation of Ethynylbenziodoxolones: An Alternative to Photocatalysis for Alkynylation Reactions**

Article abstract of DOI:10.1002/anie.202110257

Ethynylbenziodoxolones (EBXs) are commonly used as radical traps in photocatalytic alkynylations. Herein, we report that aryl-substituted EBX reagents can be directly activated by visible light irradiation. They act as both oxidants and radical traps, alleviating the need for a photocatalyst in several reported EBX-mediated processes, including decarboxylative and deboronative alkynylations, the oxyalkynylation of enamides and the C?H alkynylation of THF. Furthermore, the method could be applied to the synthesis of alkynylated quaternary centers from tertiary alcohols via stable oxalate salts and from tertiary amines via aryl imines. A photocatalytic process using 4CzIPN as an organic dye was also developed for the deoxyalkynylation of oxalates.

Full text of DOI:10.1002/anie.202110257

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Direct Photoexcitation of EthynylBenziodoXolones: An Alternative
to Photocatalysis for Alkynylation Reactions
Authors: Stephanie G. E. Amos, Diana Cavalli, Franck Le Vaillant, and
Jerome Waser
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202110257
Link to VoR: https://doi.org/10.1002/anie.202110257

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
Direct Photoexcitation of EthynylBenziodoXolones: An
Alternative to Photocatalysis for Alkynylation Reactions
Stephanie G. E. Amos+,[a] Diana Cavalli+,[a] Franck Le Vaillant,[b] Jerome Waser*[a]
[a]
Prof. Dr. Jerome Waser
Laboratory of Catalysis and Organic Synthesis,
Institut des Sciences et Ingénierie Chimique,
Ecole Polytechnique Fédérale de Lausanne, CH-1015, Lausanne, Switzerland Department
E-mail: jerome.waser@epfl.ch
[b]
[+]
Dr. Franck Le Vaillant
Max-Planck-Institut fꢀr Kohlenforschung,
Mꢀlheim an der Ruhr 45470, Germany
These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of the document.
Abstract: EthynylBenziodoXolones (EBXs) are commonly used as
radical traps in photocatalytic alkynylations. Herein, we report that
aryl-substituted EBX reagents can be directly activated by visible light
irradiation. They act as both oxidants and radical traps, alleviating the
give the ground state photocatalyst and benzoate II. When
considering that several steps in the catalytic cycle involve
reactive species present in low concentration, it is not surprising
that fine-tuning of both catalyst structure and reaction conditions
need for a photocatalyst in several reported EBX-mediated processes, is required for success.
including decarboxylative and deboronative alkynylations, the
oxyalkynylation of enamides and the C-H alkynylation of THF. In
addition, the method could be applied to the synthesis of alkynylated
quaternary centers from tertiary alcohols via stable oxalate salts and
from tertiary amines via aryl imines. A photocatalytic process using
4CzIPN as an organic dye was also developed for the
deoxyalkynylation of oxalates.
Introduction
Alkynes have found broad applications in synthetic and medicinal
chemistry, chemical biology, and materials science.[1] They can
be used either as an inert and rigid connecting element or as a
reactive unit.[2] Therefore, it is not surprising that synthetic
methods for accessing alkynes are the focus of intensive research.
In addition to alkynylations of nucleophiles and electrophiles, the
alkynylation of carbon radicals has emerged as an attractive
complementary route for the synthesis of alkynes, enabling in
particular the synthesis of highly sterically hindered systems, such
as quaternary centers.[3,4] These compounds have found
numerous applications in the total synthesis of natural products
and medicinal chemistry.[5] Recently, particular attention has been
placed on visible light-mediated alkynylations with a photocatalyst
as they enable the generation of highly reactive open-shell
species such as radicals avoiding the use of strong UV irradiation
or toxic precursors (Scheme 1A). However, fine-tuning of the
photocatalyst is required for each new alkynylation process.
Among possible radical traps, EthynylBenziodoXolones (EBXs, 1)
hypervalent iodine reagents have been especially successful.[3,6]
Photomediated alkynylations with EBXs (1) follow usually a
reductive quenching mechanism,[7] in which an excited state
photocatalyst (PC*) is required to generate a carbon radical by
oxidation of the substrate (Scheme 1B). The reaction of the
radical with EBXs (1) gives then the alkynylation product and
Scheme 1. Photomediated alkynylation and use of EBX reagents as radical
traps with or without photocatalyst. PC = photocatalyst, AG = activating group.
iodanyl radical I. For an efficient catalytic process, the reduced
photocatalyst PC•- needs to be reoxidized by iodanyl radical I to
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
Herein, we report the serendipitous discovery of a different
alkynylation approach via the visible light photoexcitation of aryl-
EBX reagents, alleviating the need for a fine-tuned photocatalyst
(Scheme 1C). Visible light irradiation can promote the excitation
of a variety of hypervalent iodine reagents through spin forbidden
transitions.[8] Nevertheless, to the best of our knowledge, this
activation mode has never been reported for EBXs. We now
demonstrate that the excited state ArEBX* (1*) of ArEBX (1) can
be used as a photooxidant to activate a variety of oxidizable
Discovery and optimization of the photocatalyst-free
deoxyalkynylation. When attempting the photocatalytic
deoxyalkynylation of cesium oxalate 3a (3a•/3a- = 1.3 V vs
SCE)[10a] with PhEBX (1a) and 4CzIPN (2a) as photocatalyst, we
discovered in
a
control experiment that the desired
deoxyalkynylated product 4a could be obtained in 50% yield In
DCM with 1.5 equiv of 1a in absence of 2a (Table 1, entry 1). We
also observed the formation of ketones 5a and 5b in 17 and 12%
yield respectively (entry 1). For this experiment, we used 2 high
intensity Kessil lamps (40 W each) with a broad bandwidth
irradiation around 440 nm as the light source. When frequently
used blue LED strips with a lower intensity (8 W) and an irradiation
centered around 460 nm were applied, the formation of 4a was
not observed (entry 2). This is in agreement with previous reports
using blue LED strips in which no background reaction is
observed in absence of a photocatalyst.[6b-h] With blue LED strips,
we still did see conversion of cesium oxalate 3a and PhEBX (1a)
to compounds 5a and 5b (entry 2). When the reaction was
performed in the dark at 50 °C, ketones 5a and 5b were obtained
in 45% and 20% yield (entry 3). This suggests a thermal pathway
for the formation of 5a and 5b. These ketones originated from the
formal hydration of 1a and the incorporation of a nucleophile
(iodobenzoate or oxalate). Using a larger excess of PhEBX (1a)
resulted in an increase of yield to 57% (entry 4). We then turned
functional
groups,
allowing
deboronative[6b]
and
decarboxylative[6c-e] alkynylations, as well as the decarboxylative
fragmentation of oximes,[6f] the difunctionalisation of enamides[6h]
and the alkynylation of C-H ether bonds.[9] All these processes
were reported only in the presence of a photocatalyst previously.
In addition, this direct excitation strategy allowed the
deoxygenation/deamination-alkynylation of broadly available
tertiary alcohols and amines via cesium oxalate salts[10] or aryl
imines[11] respectively, giving access to valuable alkynes
connected to quaternary centers. These radical precursors have
not yet been used to access alkynes. Only one approach used
alcohols as precursors, exploiting a reductive substrate activation
strategy with unstable and non-isolable N-phthalimidoyl oxalates
as precursors.[12] We especially focused on easily available
oxalate salts and developed in addition for these substrates a
photocatalytic method with the organic dye 4CzIPN (2,4,5,6-
tetrakis(9H-carbazol-9-yl) isophthalonitrile, 2a). Whereas the
direct photoactivation method stands out for its operational
simplicity, the photocatalytic approach usually proceeded in
higher yields and tolerate a broader range of alkynes.
to screening the irradiation wavelength using a single lamp.[13]
A
drop in yield was detected at 440 nm yielding 4a in 41% despite
a longer reaction time (entry 5). Irradiation centered at 427 nm led
to no significant change (entry 6), whereas a lower conversion of
3a and PhEBX (1a) and a lower yield of product 4a were observed
at 390 nm (entry 7). Finally, 2 lamps at 440 nm and a longer
reaction time afforded 4a in 63% yield. However, a more careful
monitoring of the reaction over time showed that no further
conversion was observed after 8 hours, and the observed small
increase is not significant.[14]
Results and Discussion
Table 1. Optimization of the direct excitation deoxyalkynylation
Entry[a]
1a (equiv)
1.5
reaction time
18 h
Light source
2 Kessil lamps
LED strips
λ (nm)
440
460
dark
440
440
427
390
440
residual 3a (equiv)
residual 1a (equiv)
Yield 4a[b]
Yield 5a[b]
Yield 5b[b]
1
2]
3[c]
4
0.25
0.70
0.12
0.66
0.60
0.40
0.40
0.40
0.60
0.40
50
nd
nd
57
41
43
34
63
17
36
45
50
43
25
23
33
12
23
23
23
19
16
17
18
1.5
24 h
1.5
24 h
none
0.75
2.5
18 h
2 Kessil lamps
1 Kessil lamp
1 Kessil lamp
1 Kessil lamp
2 Kessil lamps
0.05
5
2.5
24 h
0.25
6
2.5
24 h
0.15
7
2.5
24 h
0.30 - 0.40
0.08
8
2.5
24 h
[a] 3a (0.1 mmol) and 1a were dissolved in DCM [3a] = 0.1 M and irradiated with two lamps (40 W, 440 nm) or LED strips (8 W, 460 nm) at a temperature of 30-
35 °C. [b] 1H NMR yield was determined using 1 equiv of CH2Br2 as internal standard. [c] Reaction was run at 50 °C. nd = not detected.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
The irradiation of a solution of PhEBX (1a) led to non-negligible
degradation (60% in 16 h) with formation of diyne 6 in 40% yield
(Scheme 2A), whereas cesium salt 3a did not show any
degradation when irradiated separately (Scheme 2B). These
experiments confirmed a possible direct light excitation of PhEBX
(1a) to PhEBX* (1a*) independent of the cesium oxalate.
Scope of functional group activation. With the optimized
conditions in hand, we first investigated the scope of the
alkynylation of cesium oxalates (Scheme 3A). The model
substrate 3a could be converted into 4a in 60% isolated yield,
while the tert-butanol derivative 3b afforded 4b with 57% yield.
The same reaction conditions were applied to 5-, 6-, and 12-
membered rings 3c-e, delivering the products 4c-e in 54, 61 and
37% yield. Alkynylated heterocyclic 4f was obtained in 62% yield.
With pTolEBX (1b) as an alkynylating reagent, 4g was obtained
in 70% yield. However, the use of halogen-substituted aryl EBX
reagents lead to low yields.[15] We then wondered if this direct
excitation strategy could be extended to other decarboxylation
processes. We first investigated the decarboxylative alkynylation
of carboxylic acids 7 (7b•/7b- = 1.2 V vs SCE).[6c-e] By simply
increasing the reagent and base loading when compared to the
photocatalytic procedure, the decarboxylation proceeded in 18
hours affording the desired alkynylation products 8 (Scheme 3B).
This gave access to ynone 8a in 81% yield, as well as aliphatic
alkynes 8b and 8c in 51% and 41% yield respectively.
Scheme 2. Control experiments supporting the direct photoactivation of PhEBX
(1a) and the stability of cesium oxalate 3a in absence of 1a. The reactions were
performed at 0.1 mmol or 0.2 mmol scale and yields were determined by 1H
NMR by addition of 1 equiv of CH2Br2 as an internal standard.
Scheme 3. Scope of functional group activation. For A, B, C D, and F: Reactions were performed on 0.3 mmol, E: the reaction was performed on 0.1 mmol scale,
the yield was determined by 1H NMR using CH2Br2 (1 equiv) as an internal standard and G: Reaction was performed on 0.2 mmol scale, the yield was determined
by 1H NMR using CH2Br2 (1 equiv) as an internal standard.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
We then examined the potential of PhEBX* (1a*) in
a
decarboxylative
oxime
fragmentation-alkynylation
(oxime•/oxime- ≈ 1.5 V vs SCE).[6f] In this case, 2.5 equivalents of
PhEBX (1a) instead of the reported 2.0 equivalents gave the
desired fragmentation products 10a-c in 74, 69 and 53% yield with
no other changes to the reaction conditions (Scheme 3C).
With these results, it appeared that the PhEBX* was a potent
oxidant activating substrates with potentials up to 1.5 V vs SCE.
It should be therefore able to oxidize other functional groups
beside carboxylates. We turned first to the deboronative
alkynylation of alkyl trifluoroborate salts 11 (11c•/11c- = 1.5 V vs
SCE).[6b,16] In the reported procedure, 0.5 equiv of a hypervalent
iodine additive (hydroxybenziodoxolone, BIOH, 19a) were
required. The direct light activation of PhEBX (1a) in excess
alleviated the need for both the photocatalyst and the additive
(Scheme 3D). The alkynylated products 12a-c were obtained in
72%, 61% and 47% yield, respectively. Furthermore, the direct
activation of PhEBX could be used for the difunctionalization of
enamide 13 (enamide•+/enamide ≈ 1.3 V vs SCE) without any
change in the reaction conditions, leading to the formation of 14
in 35% yield (Scheme 3E).[6h] A new alkynylation reaction was
attempted next. Inspired by the deamination-alkylation of Rovis,
we wondered if we could perform a deaminative alkynylation
reaction of imine 15 (imine•+/imine ≈ 1.4 vs SCE).[11] Indeed, the
conversion of 15 into 16 occurred in 57% yield via direct
photoexcitation of PhEBX (1a) in the presence of Cs2CO3
(Scheme 3F). Finally, the C-H alkynylation of THF (17) gave
alkyne 18 in 80% 1H NMR yield (Scheme 3G).
Scheme 4. Alternative mechanisms for the oxidative activation of substrates 3-
8 to give radical I.
To gain support for PhEBX* (1a*) as photooxidant, further
experiments were performed (Table 2). Interestingly, when TIPS-
EBX (TIPS = triisopropylsilyl, 1c) was used as an alkynylation
reagent in absence of photocatalyst, no product was observed
and both 1c and cesium oxalate 3a remained untouched (Table
2, entry 1). 1c has been reported to not undergo direct excitation
at 400 nm,[9c] but is known to work as a radical trap.[6d] When we
performed the reaction with 3a and 1c with 4CzIPN (2a) as a
photocatalyst, we obtained 25% of alkynylation product 4h
confirming that 1c is able to react with the tertiary radical formed
from cesium oxalates, even if the overall reaction is not very
efficient (entry 2). This result confirmed that the aryl substituent
on PhEBX (1a) was required for the reaction to proceed in
absence of photocatalyst, but it still did not allow us to distinguish
between our possible reaction pathways.
We then turned to the role of the side product ketone 5a. We first
explored the possibility of 5a acting as photocatalyst and
performed the alkynylation with 0.2 equivalents of 5a and
TIPS-EBX (1c) (entry 3). Only traces of alkynylation product were
observed. In presence of one equivalent of 5a, no degradation of
cesium oxalate 3a or 5a was observed under irradiation (entry 4).
These results showed that 5a was not competent to catalyze the
alkynylation process. 5b was also subjected to the same control
experiments with no alkynylation products detected (See
Supporting Information).[19] We then investigated the potential
effect of traces of hypervalent iodine species 19a and 19b. When
the reaction was performed with 0.2 equivalents of either additive,
nearly no product formation was obtained with 1c (entries 5 and
6). Even with 1.5 equivalents of additive in absence of 1c, very
little degradation of the starting material was observed upon
irradiation (entries 7 and 8). Finally, we performed the alkynylation
with 0.2 equivalents of PhEBX (1a) and 1.5 equivalents of TIPS-
EBX (1c). In this case, 20% conversion of the cesium salt was
observed with 8-13% deoxyalkynylation with phenyl and TIPS
alkyne products 4a and 4h formed in a 1:1 ratio based on 1H NMR
analysis (entry 9), giving support for 1a only acting as
photooxidant, whereas both 1a and 1c can act as radical traps.
Mechanistic investigation. Having demonstrated that
alkynylation was possible for a broad scope of oxidizable
substrates, we attempted to gain a better understanding of the
transformation. To rationalize our results, we envisaged three
main reaction pathways. First, based on the photoinduced
degradation of PhEBX (1a, Scheme 2A) we postulated the direct
excitation of PhEBX (1a) to give a strong oxidant 1a* able to
oxidize the substrates to give radical intermediate I (Scheme 4A).
Then, based on the broad use of aromatic ketones as
photosensitizers,[17] we considered the possibility that the
aromatic ketone 5a formed during the reaction (Table 1) could act
as a photocatalyst and/or a photooxidant (Scheme 4B). Finally, a
residual HIR(III) species BIOX from the synthesis of PhEBX (1a)
could be the oxidant (Scheme 4C). For the special case of
decarboxylative procedures, the formation of a covalent adduct II
between the carboxylate and BIOX, such as BIOH (19a) and
BIOAc (19b), could be expected based on literature
precedence.[4,18] Especially 19a could be present in small
amounts as impurity in PhEBX (1a), as it is used for its synthesis.
Under visible light irradiation, homolytic cleavage of the
hypervalent iodine IIII-O would form the O-centered radical, which
then undergoes double decarboxylation to give radical I.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
Table 2. Control experiments for the determination of the oxidative species in
the deoxyalkynylation.
Entry[a]
Reagent
Additive (equiv)
-
Residual 3a[b] (%)
Yield[b] (%)
1
2
3
4
5
6
7
8
9
1c
1c
1c
-
100
30
nd
2a (0.05)
5a (0.2)
5a (1.0)
19a (0.2)
19b (0.2)
19a (1.5)
19b (1.5)
1a (0.2)
25
98
2
100
92
-
1c
1c
-
5
Figure 1. A. Normalized absorption of 1a (blue plain line), emission (red dashed
line, excitation at λ = 390 nm, λmax = 485 nm) and fluorescence excitation (grey
dotted line, for emission at 485 nm, λmax = 362 nm). B. Beer Lambert linear
regression for 420 nm (red plain line) and 440 nm (blue dotted line). A = εl[1a],
l = 1 cm), ε420nm = 0.54 L.mol-1.cm-1 and ε440 nm = 0.33 L.mol-1.cm-1.
92
5
90
-
-
-
100
80
1c
8-13[c]
Having confirmed that PhEBX (1a) was absorbing under our
irradiation conditions, it was important to estimate its strength as
an oxidant in the excited state, in particular considering the broad
scope of substrates that could be oxidized with 1a*. First, cyclic
voltammetry allowed us to determine the redox potential of the
ground state E1/2(1a/1a•-) = -0.87 V vs SCE (Figure 2). We could
then calculate an estimate of the excited state E1/2(1a*/1a•-) = +1.8
V vs SCE,[22] thus confirming the thermodynamic feasibility of the
SET oxidation of the substrates (potentials ranging from +1.3 to
+1.5 V, see above).
[a] 3a (0.1 mmol), 1c (1.5 equiv) and the additive were dissolved in DCM [3a] =
0.1 M and irradiated with two lamps (40 W, 440 nm) for 18 h. [b] 1H NMR yield
was determined using 1 equiv of CH2Br2 as internal standard. [c] Overall yield
of deoxy-alkynylation, 4a:4h = 1:1.
With these results in hand, we turned to UV-Vis absorption and
fluorescence spectroscopy to have further support for the
photoactivity of PhEBX (1a) under our reaction conditions (Figure
1). We observed absorption until 460 nm (plain blue line) and
fluorescence at 485 nm (dashed red line) (Figure 1A).
Fluorescence excitation spectroscopy (dotted grey line) showed
that irradiation of 1a from 300 nm to 440 nm was responsible for
the emission at 485 nm. Specifically, we observed two excitation
bands (λmax,1 = 380 nm, λmax,2 = 430 nm) confirming the possibility
of the photoexcitation of 1a with a broad band light source with
emission centered around 440 nm. To identify the molar extinction
coefficient ε of 1a, we performed a Beer-Lambert linear regression
at 420 nm and 440 nm providing ε420nm = 0.54 L.mol-1.cm-1 and
ε440nm = 0.33 L.mol-1.cm-1 (Figure 1B).[20] This is coherent with the
weak absorption we observe in the 390 nm – 460 nm range, even
at high concentrations. These low molar extinction coefficients
suggest that the absorption at 420 nm and 440 nm could result
from a spin forbidden electronic transition.[21]
Figure 2. Cyclic voltammogram of 1a (50 mV.s-1, 1.0 μM in MeCN).
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
We then performed UV-Vis of three other EBXs (Figure 3).
Interestingly, pTolEBX (1b) absorbed more in the region of
irradiation than PhEBX (1a). We believe that this could explain the
higher yield of deoxyalkynylated 4g (70% instead of 60% for
PhEBX). TIPS-EBX (1c) absorbed less than the ArEBXs although
the absorption band did still tail off into the visible light region.
Finally, mFPhEBX (1d) absorbed similarly to PhEBX (1a).
mFPhEBX (1d) did indeed undergo degradation under visible light
irradiation. However, the alkynylation was less efficient (30% by
1H NMR).[15]
Figure 3. Absorption spectra of 1a, 1b, 1c and 1d at 0.1 M in DMSO.
Overall Mechanistic Proposal. With the results obtained in our
work together with literature precedence,[4,6,9,16] we propose the
following speculative mechanism for our alkynylation method
(Scheme 5). Our experimental data (Figures 1, 2 and 3) suggest
that ArEBX (1) undergoes direct photoexcitation to generate a
highly oxidant excited state 1* (+1.8 V vs SCE), the latter can then
perform a SET oxidation of cesium salt 3 (+1.3 V vs SCE). The
resulting O-centered radical I fragments to the acyl radical II and
finally the tertiary alkyl radical III releasing two molecules of
carbon dioxide. The latter can then add to a second EBX reagent
1 affording the final product 4 and the iodanyl radical IV. IV (+0.25
V vs SCE)[6e] would most likely not be capable of performing the
oxidation of the cesium salt as it is thermodynamically unfavored.
Following the oxidation of oxalate 3, we suspect that the reduced
1a•- would be highly unstable and degrade resulting in side
products such as the observed 1,4-diphenylbutadiyne (6).
Additionally, we cannot exclude the possibility of diyne formation
from the excited state 1*. The formation of ketones 5a and 5b
seems to be a background process occurring under thermal
conditions. These ketones impact the yield slightly due to the
consumption of the starting materials, however, they do not seem
to play a role in the reaction mechanism.
Scheme 5. Speculative mechanism.
Photocatalyzed deoxyalkynylation. Considering the synthetic
utility of a photomediated method for accessing alkynylated
quaternary centers, we turned back to photocatalysis to improve
the yields of our deoxyalkynylation strategy. Using 5 mol% of
4CzIPN (2a) (2a*/2a- = +1.35 V vs SCE),[6f] under light irradiation
at 440 nm with only 1.5 equivalent of PhEBX (1a) in DCM, the
desired product (4a) was observed in 75% NMR yield (94% based
on remaining starting material, Scheme 6).
Scheme 6. Optimized conditions for the photocatalyzed deoxyakynylation.
With the optimized reaction conditions established, we proceeded
to explore the scope of cesium oxalates. The model substrate 3a
afforded the desired alkyne 4a in 75% yield (compared to 60% for
the direct photoexcitation) (Scheme 7).
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
Alkynylation of Natural Products. Finally, we were delighted to
see that both the direct excitation and photocatalytic methods
could be applied for the diastereoselective deoxyalkynylation of
(-)-cedrol oxalate 3w (Scheme 8A). Both the direct excitation and
the photocatalytic methods provided products in over 50% yield
and 20:1 diastereoselectivity based on NMR analysis. NOESY
analysis supported that the isomer obtained is of (S) configuration
at C8.
Interestingly, when (-)-terpinen-4-ol-derived oxalate 3x was used,
a different outcome was observed: the 5-exo-trig cyclization of the
intermediate acyl radical II onto the double bond was observed
followed by alkynylation of the formed tertiary carbon radical
(Scheme 8B). Both methods resulted in the formation of product
4x in 47% and 53% yield, respectively. This indicated that the
decarboxylations are stepwise and that the acyl radical II formed
from the oxalate radical after the release of CO2 is long-lived
enough to undergo cyclization before the second decarboxylation.
The remotely alkynylated product 4x was also obtained in over
20:1 diastereoselectivity.
Scheme 7. Scope of the photocatalytic deoxyalkynylation. Reactions were
performed on 0.3 mmol scale using the corresponding cesium oxalate 3 (1
equiv) and ArEBX 2 (1.5 equiv) with 4CzIPN (2a, 5 mol%) in DCM (0.1 M). [a]
Reaction was performed on 0.24 mmol scale with 1.9 equiv of 2a.
Scheme 8. Alkynylation of A. (-)-Cedrol oxalate 3w and B. (-)-Terpinen-4-ol
oxalate 3x. Reactions were performed on 0.3 mmol scale under blue LED
irradiation. Method A: 3w or 3x (1 equiv), 1a (1.5 equiv), 2a (5 mol%) in DCM
(0.1 M), 50 °C. Method B: 3w or 3x (1 equiv), 1a (2.5 equiv) in DCM (0.1 M).
Substrates 3b-f already examined for the direct photoexcitation
gave the products 4b-f in improved yields (63-91%).
Cycloheptane-derived alkyne 4i was isolated in 74% yield. The
adamantyl alkyne 4j was obtained with an expectable drop in yield
when considering the unfavored bridged carbon radical.[23]
Aliphatic alkynes 4k and 4l were isolated in 72% yield.
Homobenzylic scaffolds yielded compounds 4m and 4n in 55 and
30% yield, respectively. A variety of benzyl and silyl protected
alcohols afforded alkynes 4o-q in up to 61% yield. Having
established the scope of alcohols, we turned to explore different
ArEBX reagents. pTolEBX afforded the desired product in 64%
yield, which was slightly lower than using the direct excitation
approach (70%). Nevertheless, the catalytic method tolerated a
greater panel of reagents than the direct excitation approach:
electron-poor fluorinated reagents afforded the corresponding
alkynes 4r and 4s in 56% and 58% yield. Brominated and
chlorinated aryl alkynes 4t-v were obtained in 45 to 78% yield.
Silyl- and alkyl- EBX reagents however gave the product in only
very low yield.
Conclusion
In summary, we have discovered the direct photoexcitation of aryl
EBX reagents in the context of the deoxyalkynylation of cesium
oxalates.[24] The broad applicability of the direct excitation of
ArEBXs was then exemplified in alkynylation processes requiring
a
photocatalyst before, including decarboxylative and
deboronative alkynylations, the oxyalkynylation of enamides and
the C-H alkynylation of THF. The direct excitation of ArEBXs has
also enabled a first example of deaminative alkynylation via an
aryl imine. In the case of the deoxyalkynylation of oxalates, we
also developed a photocatalytic approach using 4CzIPN (2a) as
organophotocatalyst and accessed an extended scope of
alkynylated quaternary centers.[25] The direct excitation approach
discovered in our work results in simplified reaction design and
will therefore facilitate the discovery of new alkynylation reactions
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
[7]
[8]
An oxidative quenching catalytic cycle has also been proposed, see ref
6b.
using ArEBX reagents, as demonstrated in the case of deoxy- and
deamino-alkynylation.[26]
M. Nakajima, S. Nagasawa, K. Matsumoto, T. Kuribara, A. Muranaka, M.
Uchiyama, T. Nemoto, Angew. Chem. Int. Ed. 2020, 59, 6847 –6852;
Angew. Chem. 2020, 132, 6914 –6919.
Acknowledgements
[9]
Selected examples: a) R.-Y. Zhang, L.-Y. Xi, L. Zhang, S. Liang, S.-Y.
Chen, X.-Q. Yu, RSC Adv. 2014, 4, 54349–54353; b) E. Voutyritsa, M.
Garreau, M. G. Kokotou, I. Triandafillidi, J. Waser, C. G. Kokotos, Chem.
– Eur. J. 2020, 26, 14453–14460; c) K. Matsumoto, M. Nakajima, T.
Nemoto, J. Org. Chem. 2020, 85, 11802–11811.
We thank the ERC (European Research Council, Starting Grant
iTools4MC, number 334840) and EPFL for financial support. This
publication was created as part of NCCR Catalysis, a National
Centre of Competence in Research funded by the Swiss National
Science Foundation. We thank Cristian Pezzato and Cesare
Berton from the Laboratory of Supramolecular Chemistry, Ecole
Polytechnique Fédérale de Lausanne for enabling us to run our
absorption and fluorescence studies and Ophélie Planes from the
same laboratory for her help in the cyclic voltammetry studies.
[10] a) C. C. Nawrat, C. R. Jamison, Y. Slutskyy, D. W. C. MacMillan, L. E.
Overman, J. Am. Chem. Soc. 2015, 137, 11270–11273; b) X. Zhang, D.
W. C. MacMillan, J. Am. Chem. Soc. 2016, 138, 13862–13865; c) S. Y.
Abbas, P. Zhao, L. E. Overman, Org. Lett. 2018, 20, 868–871; d) J. Y.
Su, D. C. Grünenfelder, K. Takeuchi, S. E. Reisman, Org. Lett. 2018, 20,
4912–4916.
[11] M. A. Ashley, T. Rovis, J. Am. Chem. Soc. 2020, 142, 18310–18316.
[12] C. Gao, J. Li, J. Yu, H. Yang, H. Fu, Chem. Commun. 2016, 52, 7292–
7294.
Keywords: Alkynes; Hypervalent Iodine; Photochemistry;
[13] This was due for practical reasons, as only a single lamp was available
at other wavelengths in our laboratory.
Quaternary centers; Synthetic methods.
[14] See supporting information Table S3 and Figure S2 for details
[15] 15 - 30% by NMR. See supporting information for details.
[16] Converted from V vs Fc to V vs SCE, taken from: Y. Yasu, T. Koike, M.
Akita, Adv. Synth. Catal. 2012, 354, 3414–3420.
[1]
a) Acetylene Chemistry: Chemistry, Biology, and Material Science; F.
Diederich, P. J. Stang, R. R. Tykwinski, Eds.; Wiley-VCH: Weinheim,
2005; b) I. A. Maretina, J. C. Tebby, Alkynes in Cycloadditions; Wiley:
Chichester, West Sussex, 2014; c) Modern Alkyne Chemistry: Catalytic
and Atom-Economic Transformations; B. M. Trost, C.-J. Li, Eds.; Wiley-
VCH: Weinheim, 2015; d) Y. Jiang, R. E. McNamee, P. J. Smith, A.
Sozanschi, Z. Tong, E. A. Anderson, Chem. Soc. Rev. 2021, 50, 58–71.
a) P. Thirumurugan, D. Matosiuk, K. Jozwiak, Chem. Rev. 2013, 113,
4905–4979; b) T. T. Talele, J. Med. Chem. 2020, 63, 5625–5663.
F. Le Vaillant, J. Waser, Chem. Sci. 2019, 10, 8909–8923.
[17] J. Pérez-Prieto, R. E. Galian, M. A. Miranda, Diaryl Ketones as
Photoactivators. In Mini-Reviews in Organic Chemistry; 2006; Vol. 3, pp
117–135.
[18] K. Jia, Y. Chen, Photochemistry of Hypervalent Iodine Compounds. In
PATAI’S Chemistry of Functional Groups; American Cancer Society,
2018; pp 1–42.
[2]
[19] Irradiation of 3a (1 equiv) and 5b (1 equiv) in DCM led to no observable
degradation. Irradiation of 3a (1 equiv), 5b (0.7 equiv) with 1c (1.5 equiv)
led to less than 5% product formation. Irradiation of 5b (1 equiv) and 1a
(1.5 equiv) in DCM led to no product formation confirming that 5b is most
likely not a reaction intermediate.
[3]
[4]
Selected examples: Radical generated from alkenes: a) Y. Shen, B.
Huang, J. Zheng, C. Lin, Y. Liu, S. Cui, Org. Lett. 2017, 19, 1744–1747;
b) S. Zhou, T. Song, H. Chen, Z. Liu, H. Shen, C. Li, Org. Lett. 2017, 19,
698–701; c) H. Jiang, Y. He, Y. Cheng, S. Yu, Org. Lett. 2017, 19, 1240–
1243; d) Y. Li, R. Lu, S. Sun, L. Liu, Org. Lett. 2018, 20, 6836–6839; e)
H. Jiang, A. Studer, Chem. – Eur. J. 2019, 25, 516–520; Radical
generated via HAT: f) T. Hoshikawa, S. Kamijo, M. Inoue, Org. Biomol.
Chem. 2012, 11, 164–169; g) L. Wang, Y. Xia, K. Bergander, A. Studer,
Org. Lett. 2018, 20, 5817–5820; h) S. P. Morcillo, E. M. Dauncey, J. H.
Kim, J. J. Douglas, N. S. Sheikh, D. Leonori, Angew. Chem. Int. Ed. 2018,
57, 12945–12949; Angew. Chem. 2018, 130, 13127–13131; Other
approaches: i) D. Wang, L. Zhang, S. Luo, Org. Lett. 2017, 19, 4924–
4927; j) S. Tang, Y. Liu, X. Gao, P. Wang, P. Huang, A. Lei, J. Am. Chem.
Soc. 2018, 140 , 6006–6013; k) Y. Xia, A. Studer, Angew. Chem. Int. Ed.
2019, 58, 9836–9840. Angew. Chem. 2019, 131, 9941–9945.
[20] The reproducibility of the measurement was verified by 3 repetitions of
the sample.
[21] Photochemistry of Organic Compounds, 1st ed.; John Wiley & Sons, Ltd,
2009.
[22] E1/2(1a*/1a•-) = E0-0 + E1/2(1a/1a•-). E0-0 was determined experimentally
from the position of the long wavelength tail of the absorption spectrum
at 460 nm, as described in: L. Buzzetti, A. Prieto, S. R. Roy, P. Melchiorre,
Angew. Chem. Int. Ed. 2017, 56, 15039–15043. Angew. Chem. 2017,
129, 15235–15239.
[23] G. L. Lackner, K. W. Quasdorf, L. E. Overman, J. Am. Chem. Soc. 2013,
135, 15342–15345.
[24] Part of this work has appeared as a preprint: S. G. E. Amos, D. Cavalli,
F. Le Vaillant, J. Waser, ChemRxiv 2021, DOI 10.33774/chemrxiv-2021-
56f12.
[5]
Selected examples: a) T. J. A. Graham, E. E. Gray, J. M. Burgess, B. C.
Goess, J. Org. Chem. 2010, 75, 226–228; b) S. Zheng, Y. R. Santosh
Laxmi, E. David, A. T. Dinkova-Kostova, K. H. Shiavoni, Y. Ren, Y.
Zheng, I. Trevino, R. Bumeister, I. Ojima, W. C. Wigley, J. B. Bliska, D.
F. Mierke, T. Honda, J. Med. Chem. 2012, 55, 4837–4846; c) K.
Sugimoto, K. Toyoshima, S. Nonaka, K. Kotaki, H. Ueda, H. Tokuyama,
Angew. Chem. Int. Ed. 2013, 52, 7168–7171; Angew. Chem. 2013, 125,
7309–7312; d) S. Yoshioka, M. Nagatomo, M. Inoue, Org. Lett. 2015, 17,
90–93.
[25] During the revision process of this manuscript, Xie and co-workers also
reported a photocatalytic method for the deoxyalkynylation of cesium
oxalates: M. Li, T. Liu, J. Li, H. He, H. Dai, J. Xie, J. Org. Chem 2021,
DOI: 10.1021/acs.joc.1c01356.
[26] Raw data for NMR, IR, and MS are available at zenodo.org, DOI:
https://doi.org/10.5281/zenodo.5205584.
[6]
Selected examples: a) X. Liu, Z. Wang, X. Cheng, C. Li, J. Am. Chem.
Soc. 2012, 134, 14330–14333; b) H. Huang, G. Zhang, L. Gong, S.
Zhang, Y. Chen, J. Am. Chem. Soc. 2014, 136, 2280–2283; c) Q. Zhou,
W. Guo, W. Ding, X. Wu, X. Chen, L. Lu, W. Xiao, Angew. Chem. Int. Ed.
2015, 54, 11196–11199; Angew. Chem. 2015, 127, 11348–11351; d) F.
Le Vaillant, T. Courant, J. Waser, Angew. Chem. Int. Ed. 2015, 54,
11200–11204; Angew. Chem. 2015, 127, 11352–11356; e) F. Le Vaillant,
M. D. Wodrich, J. Waser, Chem. Sci. 2017, 8, 1790–1800; f) F. Le
Vaillant, M. Garreau, S. Nicolai, G. Gryn’ova, C. Corminboeuf, J. Waser,
Chem. Sci. 2018, 9, 5883–5889; g) M. Garreau, F. Le Vaillant, J. Waser,
Angew. Chem. Int. Ed. 2019, 58, 8182–8186; Angew. Chem. 2019, 131,
8266–8270; h) S. G. E. Amos, S. Nicolai, J. Waser, Chem. Sci. 2020, 11,
11274–11279.
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202110257
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
We report that Aryl EthynylBenziodoXolones (ArEBXs) can be directly activated by visible light irradiation. In the excited state,
ArEBXs act as strong oxidants, alleviating the need for a photocatalyst in decarboxylative and deboronative alkynylations, the
oxyalkynylation of enamides and the C-H alkynylation of THF. This approach could be applied to the synthesis of alkynylated
quaternary centers from tertiary alcohols via oxalates and from tertiary amines via aryl imines.
Institute and/or researcher Twitter usernames: @LcsoLab @EPFL_CHEM_Tweet
9
This article is protected by copyright. All rights reserved.