A Divergent Strategy for Site-Selective Radical Disulfuration of Carboxylic Acids with Trisulfide-1,1-Dioxides

Article abstract of DOI:10.1002/anie.202104595

The direct conversion of carboxylic acids into disulfides is described. The approach employs oxidative photocatalysis for base-promoted decarboxylation of the substrate, which yields an alkyl radical that reacts with a trisulfide dioxide through homolytic substitution. The trisulfide dioxides are easily prepared by a newly described approach. 1°, 2°, and 3° carboxylic acids with varied substitution are good substrates, including amino acids and substrates with highly activated C?H bonds. Trisulfide dioxides are also used to achieve the γ-C(sp³)?H disulfuration of amides through a radical relay sequence. In both reactions, the sulfonyl radical that results from substitution propagates the reaction. Factors governing the selectivity of substitution at S2 versus S3 of the trisulfide dioxides have been explored.

Full text of DOI:10.1002/anie.202104595

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: A Divergent Strategy for Site-Selective Radical Disulfuration of
Carboxylic Acids with Trisulfide-1,1-Dioxides
Authors: Zijun Wu and Derek Andrew Pratt
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.202104595
Link to VoR: https://doi.org/10.1002/anie.202104595

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
A Divergent Strategy for Site-Selective Radical Disulfuration of
Carboxylic Acids with Trisulfide 1,1-Dioxides
Zijun Wu and Derek A. Pratt*[a]
Dedicated to Professor Ilhyong Ryu, in recognition of his contributions to radical chemistry, on the occasion of his 70th birthday.
Abstract: The direct conversion of carboxylic acids to disulfides is
described. The approach employs oxidative photocatalysis for the
base-promoted decarboxylation of the substrate, which yields an alkyl
radical that engages a trisulfide dioxide by homolytic substitution. The
trisulfide dioxides are easily prepared via a newly described approach.
Each of 1, 2 and 3 carboxylic acids with varied substitution are good
substrates, including amino acids and substrates with highly activated
C-H bonds. Trisulfide dioxides are also used to achieve -C(sp3)–H
disulfuration of amides via radical relay. In both reactions, the sulfonyl
radical that results from substitution propagates the reaction. Factors
governing the selectivity of substitution at S2 vs S3 of the trisulfide
dioxides are explored.
Disulfide moieties, which frequently occur in both natural and
synthetic products, have significant roles in biological and
pharmaceutical contexts due to their unique redox equilibrium
with thiols.1 Indeed, this equilibrium is generally exploited in their
synthesis; a symmetric disulfide is prepared via oxidation of a thiol
to a symmetric disulfide and then one half is exchanged for
another to access the disulfide.2 Of course, given this equilibrium,
achieving a single unsymmetric disulfide product directly can be
challenging and/or wasteful (Figure 1A). The emerging role of
disulfides in redox-sensitive therapeutics, imaging agents,
chemical probes and various other applications3 has therefore
prompted the development of strategies for direct disulfuration (i.e.
where both sulfur atoms are introduced via the same reagent).4
These approaches generally involve nucleophilic substitution
between appropriately activated disulfuration reagents and
electrophilic substrates4a,b,f-i or transition metal-catalyzed cross-
coupling to aryl boronic acids (Figure 1B, left).4b-d
We recently demonstrated that disulfides could be accessed
quite conveniently by homolytic substitution on tetrasulfides.5 This
Figure 1. (A) Thiol oxidation and thiol-disulfide exchange are typical means
approach relies on the weak central S-S bond in tetrasulfides and
of accessing unsymmetric disulfides – by chemistry and in nature. (B) Despite
recent advances in selective disulfuration, direct functionalization of readily
available subtrates remains elusive. (C) A summary of the divergent, site-
selective radical disulfuration of carboxylic acids presented in this work.
the fact that the highly stabilized perthiyl radical (RSS) which
results from substitution simply combines with another under the
reaction conditions to yield more starting tetrasulfide.6 In principle,
any alkyl or aryl radical source could be utilized, but in practice,
Hunsdiecker-type conditions9 and more contemporary oxidative
we found the scope and utility of the approach to be most easily
photocatalysis conditions10 yielded only trace disulfide. We
illustrated using radicals obtained by photolysis of Barton esters
surmised that the perthiyl radical liberated upon substitution on
the tetrasulfide under the photocatalytic conditions was not
capable of re-oxidizing the photocatalyst and wondered if a
disulfurating reagent RSS-X that would yield a more oxidizing
radical X upon substitution might enable the direct conversion of
or energy transfer photocatalysis7 utilizing O-acyl oximes (Figure
1B, right). From the outset we sought to use carboxylic acids
directly as substrates,8 but the tetrasulfide was destroyed under
carboxylic acids to disulfides (Figure 1C). Of course, even if a
[a]
Dr. Z. Wu, Prof. Dr. D. A. Pratt
suitable reagent could be identified, both it and the product must
be stable to the conditions of decarboxylative radical
generation,8a,10 which is a challenge in of itself given the lability of
disulfides to oxidants, reductants, nucleophiles and electrophiles.
In earlier work on organosulfur antioxidants, we had found
that trisulfide-1-oxides undergo homolytic substitution with
Department of Chemistry and Biomolecular Sciences,
University of Ottawa
10 Marie Curie Pvt., Ottawa, Ontario, K1N 6N5 (Canada)
E-mail: dpratt@uottawa.ca
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under: [link].
This article is protected by copyright. All rights reserved.

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
peroxyl radicals,6a so we initially considered them as disulfuration
reagents. However, we found them to be unstable to heat or light.
Although the weak S-S bond in trisulfide-1-oxides is stronger than
that in tetrasulfides (e.g. 36.3 kcal/mol in t-BuS(O)SSt-Bu6b vs.
30.0 kcal/mol in t-BuSSSSt-Bu6a), thermal or photochemical
cleavage of the central S-S bond in tetrasulfides is fully reversible
(even in the presence of O26b). In contrast, the trisulfide-1-oxides
photolyze/thermolyze irreversibly to form tetrasulfide and
thiosulfonate.6b Given that sulfonyl radicals are less stable than
sulfinyl radicals (the O-H BDEs in sulfinic and sulfenic acids are
ca. 78 and 70 kcal/mol, respectively11) we anticipated that
trisulfide dioxides would be more stable. Indeed, CBS-QB3
calculations12 predict that the t-BuS(O)2SSt-Bu BDE is 12
kcal/mol stronger than in t-BuS(O)SSt-Bu. Moreover, sulfonyl
excited
state
(e.g.
[Ir(dF(CF3)ppy)2(dtbbpy)]PF6
and
[Ru(bpy)3]PF6) were found to be either inferior or ineffective.
Although the reaction could be carried out in a wide variety of
solvents (entries 3-7), EtOAc proved optimal. The identity of the
base was found to have a dramatic effect on the yield. Exchanging
Cs2CO3 for less basic options (KHCO3, K2HPO4, NaOAc) resulted
in a progressive decrease in yield (entries 8-10). The stronger, but
t
also nucleophilic BuOK also gave a poor yield (entry 11, 32%).
Overall, CsF proved to be most effective, presumably due to the
high stability towards oxidation of the fluorine anion, delivering the
product in 78% yield (entry 12). Boosting the concentration of 2a
from 1.2 to 1.5 equivalents improved the yield to 83% (entry 13)
and decreasing the reaction concentration to 0.1 M improved
things further to 85% yield (entry 14), presumably due to improved
solubility of the cyclohexanoate. Control experiments confirmed
the essentiality of each of the base, photocatalyst and irradiation
to successful decarboxylative disulfuration (entries 15-17).
With optimal reaction conditions identified, we sought to
explore the substrate scope in both carboxylic acid and trisulfide
dioxide. Trisulfide dioxides, known more precisely as sulfenic
sulfonic thioanhydrides,16a are generally prepared from the
reaction of a thiosulfonate salt and a sulfenyl chloride.16 The
sulfenyl chloride can be obtained by halogenation of a thiol or
treating a disulfide with sulfonyl chloride.17 The latter route was
recently taken by Dong in a one-pot preparation of pTolS(O)2SSt-
Bu, which was subsequently utilized to produce unsymmetric
disulfides ArSSt-Bu in Cu-catalyzed couplings with aryl boronic
acids.4b We were able to prepare several simple alkyl- and phenyl-
substituted trisulfide dioxides 2a−2f using this method (Figure
2A). However, we anticipated that the requirement of SO2Cl2 to
convert disulfide to sulfenyl chloride may limit the application of
this method and designed a new strategy involving substitution of
PhSO2Na on easily accessible phthalimidyl persulfides
(sometimes referred to as the Harpp reagent),4g,18 exemplified by
red
radicals are capable oxidants (e.g. PhSO2•/PhSO2Na, E1/2 = +
0.50 V vs SCE13), implying that they may recycle reduced
photocatalyst in an oxidative photoredox cycle. Indeed, vinyl
sulfones have been used by MacMillan to achieve photoredox
α‑vinylation of α‑amino acids.10g As such, we prepared the
trisulfide dioxide 2a and were delighted to see that when
combined with carboxylic acid 5 (E1/2red(hexanoate) = +1.16 V vs
SCE14) in the presence of an acridinium salt (Mes-Acr-PhBF4,
E1/2red*(Mes-Acr-Ph+*/Mes-Acr-Ph+) = +2.15 V vs SCE15) and
base (Cs2CO3) in EtOAc the desired unsymmetric disulfide 6 was
Table 1. Optimization of the reaction conditionsa
produced in 69% yield.
Entry
Deviation from above conditions
tBuSSSStBu as disulfuration reagent
none
Yield (%)b
trace
1
2
69
3
4
5
6
7
8
with tBuOMe as solvent
with TBME as solvent
with CH3CN as solvent
with DCE as solvent
with DMF as solvent
with KHCO3 as base
54
64
37
27
38
60
9
with K2HPO4 as base
with NaOAc as base
49
10
trace
aReaction conditions: 5 (0.2 mmol), 2a (0.24 mmol), photocatalyst (5 mol%)
11
with tBuOK as base
32
and base (0.24 mmol), blue LEDs, N2 atmosphere, room temperature, 14 h.
12
with CsF as base
78
bYields were determined by 1H NMR using 1,3,5-trimethoxybenzene as
internal standard. c2a (0.30 mmol), CsF as base. dEtOAc (0.1 M). eIsolated
13c
yield.
14c,d
with CsF as base
2a (0.30 mmol), CsF as base
Without base
83
85 (80)e
NP
Figure 2. Construction of benzenesulfonodithioperoxoates (trisulfide
dioxides).
15 c,d
16 c,d
Without light
NP
those containing methyl isobutyl ketone (2g), propiophenone (2h)
and pulegone (2i) scaffolds (Figure 2B).
17c,d
Without Mes-Acr-PhBF4
NP
Equipped with optimized reaction conditions and a set of
trisulfide dioxides, the generality of this direct decarboxylative
disulfuration of carboxylic acids was explored. A range of primary
(7−18), secondary (19, 20, 22) and tertiary (21, 26) carboxylic
acids were efficiently transformed into the corresponding (7−18)
In an effort to boost the yield of the reaction, a variety of
reaction conditions were surveyed (Table 1 and Supporting
Information). Unsurprisingly, photocatalysts with a less oxidizing
This article is protected by copyright. All rights reserved.

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
Figure 3. Scope of the decarboxylative disulfuration of carboxylic acids. aReaction conditions: carboxylic acid (0.2 mmol), 2 (0.3 mmol), photocatalyst
(Mes-Acr-PhBF4, 5 mol%) and CsF (0.24 mmol), blue LEDs, N2 atmosphere, room temperature, 14 h. Isolated yields are reported. bVarious solvents (DMF,
DMSO, CH3CN/H2O, EtOAc/DMF), bases (CsF and Cs2CO3) and photocatalysts gave the same result. c10 h. dDioxane was used as solvent.
e[Ir(dF(CF3)ppy)2(dtbbpy)]PF6 (1 mol%) was used as the photocatalyst.
disulfides. Substrates possessing bromide (10), azide (11),
alkene (14), alkyne (15), ether (19) and protected amine (16, 20)
functionalities were all well accommodated (Figure 3A).
Tolerance of azides, bromides and alkynes are expected to be
particularly useful in applications of this chemistry in
(bio)conjugation chemistry, but are versatile functional handles, in
general. Although alkenes did not interfere in the radical chemistry
per se, internal olefins (e.g. oleic acid) were found to isomerize
under the reaction conditions (9).19 Disulfides from amino acids
aspartate (17) and glutamate (18), derivatives of lithocholic acid,
chenodeoxycholic acid and pregnenolone (23−25), menthylformic
acid (22), and dehydroabietic acid (26) were all easily accessed.
We were excited to try biotin as a substrate since disulfides are
often used as redox sensitive linkers to biotinylated conjugates,
but were dismayed to find no disulfide 27. Extensive screening of
reaction conditions proved fruitless. Interestingly, when biotin was
added to the prototype reaction of 2a and 5 none of disulfide 6
was observed,20 suggesting that biotin quenches the excited state
of the photocatalyst, precluding decarboxylation of the substrate.
Further investigations revealed that a variety of α-amino
acids could undergo this transformation, giving the corresponding
disulfides 28-32 (Figure 3B). It should be noted that Boc-
protected amines consistently afforded lower yields than when
amines were protected as phthalimides – presumably due to the
greater lability of the of α-amino disulfide products. Improved
yields for these substrates (28, 29) were obtained using a lower
polarity solvent (dioxane). It is of note that α-O-silyl-disulfides
have been described to release reactive sulfur species by pH
control,21 and as such, these α-amino disulfides have the potential
to be developed as hydropersulfide precursors. Similarly, α-ether
acids, such as 2-tetrahydrofuroic acid and a ribosic acid derivative,
were good substrates (88% for 33, 86% for 34). Although messy
mixtures were observed on oxidizable peptide substrates under
the
standard
reaction
conditions,
by
adopting
[Ir(dF(CF3)ppy)2(dtbbpy)]PF6 as photocatalyst (E1/2red* ((IrIV/IrIII) =
This article is protected by copyright. All rights reserved.

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
Figure 4. Scope of the remote radical C(sp3)–H disulfuration. aReaction conditions: N-allylsulfonamide (0.2 mmol), 2 (0.4 mmol), dilauroyl peroxide (20
mol%) and CHCl3 (0.5 mL), N2 atmosphere, 80 °C, 24 h. Isolated yield reported. bN-(allylsulfonyl)-N-hexylbenzamide was used instead of an N-allylsulfonamide.
+1.69 V vs SCE in CH3CN22), synthetically useful yields could be
obtained (i.e. 51% for 35). Trisulfide dioxides with longer alkyl
chains (2b), phenyl (2c) or phenone (2h) substitution could be
exchanged for 2a, affording the desired products 36−39 in
76%−82% yield (Figure 3C).
4, the disulfide moiety could be selectively installed on long
aliphatic chains (41, 42, 44) and carbocycles (43, 49), tolerating a
range of functionalities, including azide (42), protected amine (45,
46) and ether (48, from chenodeoxycholic acid). Again,
exchanging trisulfide dioxide 2a with those incorporating
pulegone, propiophenone and longer sidechains were competent
reagents, providing the -C(sp3)–H disulfuration products in good
yield (50–52). Finally, and perhaps most notably, alkylamine-
derived sulfonamides were selectively disulfurated at the -
position of the amine (53).
During the course of our studies, we surprisingly obtained
sulfides 56 and 57 when trisulfide dioxides featuring n-propyl- and
phenyl substitution on the terminal sulfur were reacted with 54
under the standard decarboxylative reaction conditions (Figure
5A). This result suggested that steric hinderance was responsible
for the successful disulfuration chemistry observed until that point
(the corresponding t-butyl trisulfide dioxide gave the disulfide 16
exclusively in 85% yield upon reaction with 54). Given that no
sulfide products were observed in radical substitutions on
tetrasulfides regardless of their structure, we investigated further.
Upon moving to a trisulfide dioxide with an i-propyl substituent at
the terminal sulfur atom as the reaction partner for 54, a mixture
of disulfide and sulfide was isolated (55, Figure 5A). The
contribution of steric effects to the selectivity of substitution was
further evident from the results of a series of reactions with the
tertiary carboxylic acid 58; the phenyl substrate yielded sulfide,
the n-propyl substrate yielded a mixture of sulfide and disulfide
whereas the i-propyl and t-butyl substrates yielded disulfides
Late-stage C–H functionalization has emerged as
a
powerful tool for the diversification of medicinal compound
libraries.23 Although there are now many radical-mediated C–H
functionalization strategies (for oxidation, amination, sulphuration,
fluorination, chlorination, bromination, cyanation, alkenylation and
alkylation, etc.)24, there is no protocol for the disulfuration of
unactivated C(sp3)–H bonds, which would be very useful given
the limitations of previous reported disulfide formation strategies.
Inspired by Studer’s remote site-selective C–H functionalization24j
using amidyl radicals formed from N-allylsulfonamides,25 we
surmised that the sulfonyl radicals formed upon substitution on
the trisulfide dioxides could be engaged in a chain reaction to
achieve a remote site-selective C-H disulfuration. Thus, the
sulfonyl radical would add to the termination carbon of an N-
allylsulfonamide, liberating an amidyl radical that could undergo a
1,5–hydrogen atom transfer to generate an alkyl radical at the -
site of the amide.24b,26 Since the N-allylsulfonamide can be readily
prepared from a carboxylic acid,24j,25 this would enable access to
another selective disulfuration from the same starting materials as
above. Indeed, following a short screening of reaction conditions
(see the Supporting Information for details), the envisioned -
C(sp3)–H disulfuration was feasible as a radical chain reaction
using lauroyl peroxide as the initiator. As summarized in Figure
This article is protected by copyright. All rights reserved.

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
Figure 5. Steric effects on homolytic substitution, competition kinetics and use of additional radical precursors. (A and B) Steric effects on
disulfide/sulfide product distribution. (C) Lowest energy transition state structures and corresponding reaction energetics determined by CBS-QB3 for homolytic
substitution by a model alkyl (methyl) radical on phenyl trisulfide dioxides bearing differing substitution on the terminal sulfur atom. (D) Radical clock experiment
provides ksub = 6.5 ×105 M-1s-1 for the reaction of a primary alkyl radical with trisulfide dioxide 2c. (E, F) Reaction of trisulfide dioxides with cyclohexyl radicals
generated from either an acyl oxime using energy transfer photocatalysis or from an alkyltrifluoroborate using oxidative photocatalysis.
‡
exclusively. Computations12 revealed that the barrier to
substitution at S2 to afford the disulfide is essentially independent
of the substituent on the terminal sulfur (G ~ 12-13 kcal/mol),
strongly dependent on its substituent, with G increasing from
9.3 to 18.0 kcal/mol along the series Ph < n-Pr < i-Pr < t-Bu. We
were surprised that sulfide formation was preferred to disulfide
formation for the less hindered substrates since the driving force
‡
whereas the barrier to substitution at S3 to afford the sulfide is
This article is protected by copyright. All rights reserved.

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
for disulfide formation is significantly greater than for sulfide
formation (ca. -17 kcal/mol vs -5 kcal/mol) owing to the greater
stability of the departing sulfonyl radical compared to the
thiosulfonyloxyl radical (CBS-QB3 predicts the S-H BDE in
PhS(O)2SH to be 86.9 kcal/mol and the O-H BDE in PhSO2H to
be 77.2 kcal/mol11). Evidently, a significant polar effect in the
transition state for substitution at S3 to yield the sulfide must be
at play. Indeed, although the pKa of phenylthiosulfonic acid has
not been reported, its gas phase acidity is calculated (by CBS-
QB3) to be 14 kcal/mol lower than that of phenylsulfinic acid.
Since reactions at S1 and S2 of the tetrasulfides have
similar thermodynamics (the displaced RSS and RSSS have
similar stabilities6a), and the polar effects are expected to be
similar in each of the transition states, the preference for disulfide
formation observed in our previous work5 must originate from
small steric effects on substitution. Indeed, we calculate
increasing barriers for substitution of a methyl radical on S1 of n-
PrSSSSn-Pr (11.3 kcal/mol), i-PrSSSSi-Pr (14.0 kcal/mol) and t-
BuSSSSt-Bu (17.1 kcal/mol), while the barrier to substitution at
S2 is comparatively invariant (10.3-11.7 kcal/mol) (see
Supporting Information).32 Thus, the same steric effects that
drive the inherent preference for substitution at S2 in tetrasulfides
must be leveraged to overcome the inherent preference for
substitution at S1 of the trisulfide dioxides in order for efficient
disulfuration to take place.
To round out our investigations, we also evaluated the
capacity of the trisulfide dioxides to deliver disulfides in
conjunction with other radicals sources.31 Use of 2a with oxime
ester 75 under the energy transfer photocatalytic conditions which
we employed in our report on tetrasulfides yielded disulfide 6 in
94% yield – besting the 87% yield obtained with t-BuSSSSt-Bu
(Figure 5E). Likewise, alkyltrifluoroborates reacted with the
trisulfide dioxides shown in Figure 5F by oxidative photocatalysis
to afford the corresponding disulfide products in good yield. It is
noteworthy that the base-free deboronative disulfuration
conditions enabled us to access the -perthioketone products 77
and 78, which cannot be obtained from the (basic)
decarboxylative disulfuration due to the elimination of perthiolate.
In summary, the use of trisulfide dioxides in place of
tetrasulfides has enabled the development of
a practical
photocatalytic protocol for the direct disulfuration of (aliphatic)
carboxylic acids. In addition, we have also presented the first
C(sp3)–H bond disulfuration strategy to install disulfide moieties
onto the
-site of amides. These strategies should find
widespread applications in late-stage disulfuration given their
operationally simple procedures, broad substrate scope and wide
functional group tolerance.
Acknowledgements
To provide some insight on the absolute kinetics of radical
substitution on the trisulfide dioxides and enable comparison to
the same process on the tetrasulfide, we carried out radical clock
experiments using the cyclization of the 5-hexenyl radical derived
from 72 as the calibrated unimolecular rearrangement (kcyc = 2 ×
105 s−1).27 Thus, the ratio of the linear product 73 derived from
substitution by the unrearranged 5-hexenyl radical and the
cyclized product 74 formed from substitution by the rearranged
cyclopentylcarbinyl radical rearrangement were determined by
GC-MS as a function of the concentration of 2c (see Figure 5D
and the Supporting Information). The data yield a rate constant
of 6.5 × 105 M-1 s−1 for substitution of the primary alkyl radical on
trisulfide dioxide 2c. The fact that this rate constant is slightly
larger than the value we determined for t-BuSSSSt-Bu using the
same methodology (5.8 × 105 M-1 s−1)5 despite the fact that the
reaction is ~6 kcal/mol less exergonic again speaks to the greater
polarization in the transition state for substitution for the sulfonyl
radical over the perthiyl radical; consistent with the lower pKa of
PhSO2H (~3)28 relative to t-BuSSH (~7).29 The strikingly similar
rate constants prompted us to reconsider our proposed
mechanism; could the trisulfide dioxide simply be a source of
tetrsulfide that reacts with alkyl radicals? Indeed, we found that
the decomposition of trisulfide dioxide 2c could deliver the
tetrasulfide under the standard disulfuration reactions conditions.
However, after 14 hours (the time for which the foregoing
preparative reactions have been run), less than 0.15 equivalents
of tetrasulfide were observed in the reaction mixture,30 indicating
that it cannot be the principal disulfuration reagent in the reactions
of the trisulfide dioxides. This was reinforced by a direct
competition experiment utilizing differently substituted tetrasulfide
and trisulfide dioxide that give rise to unique disulfide products
upon substitution. In each case, the predominant product
observed was derived from alkyl radical substitution on the
trisulfide dioxide (see the Supporting Information for the details).
This work was supported by grants from the Natural Sciences and
Engineering Research Council of Canada and the Canada
Foundation for Innovation.
Keywords: disulfides • alkyl radicals • homolytic substitution •
photocatalysis • C-H functionalization
[1]. a) A. Mirzahosseini, B. Noszál, Sci Rep 2016, 6, 37596; b) A. Hiniker, J. C. A.
Bardwell, Biochemistry 2003, 42, 1179–1185; c) M. Góngora-Benítez, J. Tulla-
Puche, F. Albericio, Chem. Rev. 2014, 114, 901–926. d) W. T. Lowther, N. Brot,
H. Weissbach, J. F. Honek, B. W. Matthews, Proc. Natl. Acad. Sci. U. S. A. 2000,
97, 6463−6468; e) R. D. Bach, O. Dmitrenko, C. Thorpe, J. Org. Chem. 2008, 73,
12−21; f) C. Nagy, Antioxid. Redox Signaling 2013, 18, 1623−1641; g) M. Putzu,
F. Gräter, M. Elstner, T. Kubař, Phys. Chem. Chem. Phys. 2018, 20,
16222−16230
[2]. For recent reviews, see: a) D. Witt, Synthesis 2008, 2491−2509; b) Wang, M.;
Jiang, X. Top. Curr. Chem. 2018, 376, 14.
[3]. For selected recent reviews, see: a) Jiang, C.-S.; Mꢀller, W. E. G.; Schrꢁder, H.
C.; Guo, Y.-W. Chem. Rev. 2012, 112, 2179−2207. (b) Scott, K. A.; Njardarson,
J. T. Top. Curr. Chem. 2018, 376, 5–39. (c) Feng, M.; Tang, B.; Liang, S. H.;
Jiang, X. Curr. Top. Med. Chem. 2016, 16, 1200−1216.
[4]. a) C.-M. Park, B. A. Johnson, J. Duan, J.-J. Park, J. J. Day, D. Gang, W.-J. Qian,
M. Xian, Org. Lett. 2016, 18, 904−907; b) W. Wang, Y. Lin, Y. Ma, C.-H. Tung,
Z. Xu, Org. Lett. 2018, 20, 3829−3832; c) J. Zou, J. Chen, T. Shi, Y. Hou, F. Cao,
Y. Wang, X. Wang, Z. Jia, Q. Zhao, Z. Wang, ACS Catal. 2019, 9, 11426−11430;
d) X. Xiao, M. Feng, X. Jiang, Angew. Chem. Int. Ed. 2016, 55, 14121−14125;
e) Z. Dai, X. Xiao, X. Jiang, Tetrahedron 2017, 73, 3702−3706. f) X. Xiao, J.
Xue, X. Jiang, Nat. Commun. 2018, 9, 2191−2200; g) W. Gao, J. Tian, Y. Shang,
X. Jiang, Chem. Sci. 2020, 11, 3903-3908; h) J. Xue, X. Jiang, Nature. Commun.
2020, 11, 4170; i) J. Xue, X. Jiang, Org. Lett. 2020, 22, 8044-8048.
[5]. Z. Wu, D. A. Pratt, J. Am. Chem. Soc. 2020, 142, 10284−10290.
[6]. a) J.-P. R. Chauvin, E. A. Haidasz, M. Griesser, D. A. Pratt, Chem. Sci. 2016, 7,
6347−6356. b) J.-P. R. Chauvin, M. Griesser, D. A. Pratt, Chem. Sci. 2019, 10,
4999−5010.
[7]. Recent reviews on decarboxylative functionalization: a) J. Xuan, Z.-G. Zhang,
W.-J. Xiao, Angew. Chem. Int. Ed. 2015, 54, 15632−15641; b) R. Cannalire, S.
Pelliccia, L. Sancineto, E. Novellino, G. C. Tron, M. Giustiniano, Chem. Soc. Rev.
2021, 50, 766−897; c) L. J. Goossen, F. Collet, K. Goossen, Isr. J. Chem. 2010,
50, 617– 629; d) T. Patra, D. Maiti, Chem. – Eur. J. 2017, 23, 7382– 7401; e) N.
Rodríguez, L. J. Goossen, Chem. Soc. Rev. 2011, 40, 5030– 5048.
[8]. Recent reviews on energy transfer photocatalysis: a) F. Strieth-Kalthoff, F.
Glorius, Chem 2020, 6, 1888−1903; b) R. C. McAtee, E. J. McClain, C. R.J.
This article is protected by copyright. All rights reserved.

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
Stephenson, Trends in Chemistry 2019, 1, 1 111−125; c) F. Strieth-Kalthoff, M.
J. James, M. Teders, L. Pitzer, F. Glorius, Chem. Soc. Rev. 2018, 47, 7190-7202.
[9]. a) H. Hunsdiecker, C. Hunsdiecker, Ber. Dtsch. Chem. Ges. B 1942, 75, 291–
297; b) R. G. Johnson, R. K. Ingham, Chem. Rev. 1956, 56, 219– 269; c) F. Yin,
Z. Wang, Z. Li, C. Li, J. Am. Chem. Soc. 2012, 134, 10401−10404; d) X. Liu, Z.
Wang, X. Cheng, C. Li, J. Am. Chem. Soc. 2012, 134, 14330−14333; e) F. Hu, X.
Shao, D. Zhu, L. Lu, Q. Shen, Angew. Chem. Int. Ed. 2014, 53, 6105– 6109.
[10]. Selected recent examples: c) M.-C. Fu, R. Shang, B. Zhao, B. Wang, Y. Fu,
Science 2019, 363, 1429−1434; d) A. Noble, R. S. Mega, D. Pflästerer, E. L.
Myers, V. K. Aggarwal, Angew. Chem. Int. Ed. 2018, 57, 2155–2159; e) S. Ventre,
F. R. Petronijević, D. W. C. MacMillan, J. Am. Chem. Soc. 2015, 137,
5654−5657; f) A. Fawcett, J. Pradeilles, Y. Wang, T. Mutsuga, E. L. Myers, V.
K. Aggarwal, Science 2017, 357, 283−286; g) A. Noble, D. W. C. MacMillan, J.
Am. Chem. Soc. 2014, 136, 11602−11605.
[11]. M. Griesser, J.-P. R. Chauvin, D. A. Pratt, Chem. Sci. 2018, 9,7218–7229.
[12]. J. A. Montgomery Jr, M. J. Frisch, J. W. Ochterski, G. A. Petersson, J. Chem.
Phys. 1999, 110, 2822−2827. The calculations were carried out as implemented
in the Gaussian 16 (revision A.03) suite of programs.
[13]. B. Persson, Acta Chem. Scand. 1977, 31B, 88.
[14]. M. Galicia, F. J. J. Gonzalez, Electrochem. Soc. 2002, 149, D46–D50.
[15]. N. A. Romero, K. A. Margrey, N. E. Tay, D. A. Nicewicz, Science 2015, 349,
1326–1330.
[16]. a) L. G. S. Brooker, R. Child, S. Smiles, J. Chem. Soc. 1927, 1384–1388; b) D.
N. Harpp, D. K. Ash, R. A. Smith, J. Org. Chem. 1979, 44, 4135–4140; c) G.
Derbesy, D. N. Harpp, J. Org. Chem. 1995, 60, 1044–1052.
[17]. Y. Gui, L. Qiu, Y. Li, H. Li, S. J. Dong, J. Am. Chem. Soc. 2016, 138, 4890−4899.
[18]. a) D. N. Harpp, D. K. Ash, Int. J. Sulfur Chem. Part A 1971, 1, 57−59; b) D. N.
Harpp, K. Steliou, T. H. Chan, J. Am. Chem. Soc. 1978, 100, 1222–1228; c) K.
C. Nicolaou, R. Li, Z. Lu, E. N. Pitsinos, L. B. Alemany, M. Aujay, C. Lee, J.
Sandoval, J. Gavrilyuk, J. Am. Chem. Soc. 2018, 140, 12120–12136.
[19]. When the reaction was halted after 6 hours, the unreacted material was found to
consist of predominantly elaidaic acid (the Z-isomer of oleic acid).
[20]. See the Supporting Information for details of our attempts to identify suitable
reaction conditions for the decarboxylative disulfuration of biotin.
[21]. J. Kang, S. Xu, M. N. Radford, W. Zhang, S. S. Kelly, J. J. Day, M. Xian, Angew.
Chem. Int. Ed. 2018, 57, 5893–5897.
[22]. M. S. Lowry, J. L. Goldsmith, J. D. Slinker, R. Rohl, R. A. Pascal, G. G. Malliaras,
S. Bernhard, Chem. Mater. 2005, 17, 5712–5719.
[23]. Selected recent reviews on late-stage functionalization: a) J. Börger, T. Ritter,
Chem 2020, 6, 1877–1887; b) B. Hong, T. Luo, X. Lei, ACS Cent. Sci. 2020, 6,
5, 622–635; c) M. C. White, J. Zhao, J. Am. Chem. Soc. 2018, 140, 13988−14009;
d) D. J. Abrams, P. A. Provencher, E. J. Sorensen, Chem. Soc. Rev. 2018, 47,
8925–8967; e) C. S. Yeung, V. M. Dong, Chem. Rev. 2011, 111, 1215–1292; f)
J. C. K. Chu, T. Rovis, Angew. Chem. Int. Ed. 2018, 57, 62–101.
[24]. Selected recent reviews: a) L. Capaldo, L. L. Quadri, D. Ravelli, Green Chem.
2020, 22, 3376−3396; b) L. M. Stateman, K. M. Nakafuku, D. A. Nagib, Synthesis
2018, 50, 1569−1586; c) L. Revathi, L. Ravindar, W.-Y. Fang, K. P. Rakes, H.-
L. Qin, Adv. Synth. Catal. 2018, 360, 4652−4698; selected recent examples, d) K.
A. Margrey, W. L. Czaplyski, D. A. Nicewicz, E. J. Alexanian, J. Am. Chem. Soc.
2018, 140, 4213−4217; e) M. Ueda, K. Kamikawa, T. Fukuyama, Y.-T. Wang,
Y.-K. Wu, I. Ryu, Angew. Chem. Int. Ed. 2021, 60, 3545–3550; f) H.-B. Yang,
A. Feceu, D. B. C. Martin, ACS Catal. 2019, 9, 5708−5715; d) Q. An, Z. Wang,
Y. Chen, X. Wang, K. Zhang, H. Pan, W. Liu, Z. Zuo, J. Am. Chem. Soc. 2020,
142, 6216–6226; g) C. G. Na, D. Ravelli, E. J. Alexanian, J. Am. Chem. Soc. 2020,
142, 44–49; h) A. F. Prusinowski, R. K. Twumasi, E. A. Wappes, D. A. Nagib, J.
Am. Chem. Soc. 2020, 142, 5429–5438; i) S. Mukherjee, B. Maji, A. Tlahuext-
Aca, F. Glorius, J. Am. Chem. Soc. 2016, 138, 16200−16203; j) Y. Xia, L. Wang,
A. Studer, Angew. Chem. Int. Ed. 2018, 57, 12940 –12944.
[25]. C. Moutrille, S. Z. Zard, Chem. Commun. 2004, 1848–1849..
[26]. For pioneering work on remote functionalization using amidyl radicals, see: a) D.
H. R. Barton, A. L. J. Beckwith, A. Goosen, J. Chem. Soc. 1965, 181–190, and
for selected recent examples, see: b)G. Kumar, S. Pradhan, I. Chatterjee, Chem
Asian J. 2020, 15, 651–672; c) D.-F. Chen, J. C. K. Chu, T. Rovis, J. Am. Chem.
Soc. 2017, 139, 14897−14900; d) J. C. K. Chu, T. Rovis, Nature 2016, 539,
272−275; e) G. J. Choi, Q. Zhu, D. C. Miller, C. J. Gu, R. R. Knowles, Nature
2016, 539, 268−271.
[27]. D. Griller, K. U. Ingold, Acc. Chem. Res. 1980, 13, 317–323.
[28]. D. De Filippo, F. Momicchioli, Tetrahedron 1969, 25, 5733–5744.
[29]. J.-P. Chauvin, M. Griesser, D. A. Pratt, J. Am. Chem. Soc. 2017, 139, 6484–6493.
[30]. We also observed the formation of PhSO2St-Bu under the reaction conditions.
For previous reports on trisulfide dioxide decomposition, see: G. Derbesy, D. N.
Harpp, J. Org. Chem. 1996, 61, 991–997 and ref. 11b.
[31]. J. K. Matsui, S. B. Lang, D. R. Heitz, G. A. Molander, ACS Catal. 2017, 7,
2563−2575.
[32]. Importantly, inclusion of Grimme’s D3 empirical dispersion corrections in the
CBS-QB3 calculations did not lead to significant changes in these barrier (see
Supporting Information).
This article is protected by copyright. All rights reserved.

10.1002/anie.202104595
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 1:
COMMUNICATION
Trisulfide-1,1-dioxides are shown to
undergo rapid homolytic substitution
by alkyl radicals, which has enabled
the development of an oxidative
photocatalytic approach for the direct
conversion of carboxylic acids to
disulfides and a related approach for
installation of a disulfide moiety by
remote C-H functionalization.
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
This article is protected by copyright. All rights reserved.