Extended Pummerer fragmentation mediated by carbon dioxide and cyanide

Article abstract of DOI:10.1016/j.tet.2020.131633

Pummerer rearrangement reactions generate sulfur (II) oxidation state from sulfur (IV) starting materials in the presence of activating reagents. We found unprecedented transformation of vinyl sulfoxide; disulfide formation reactions mediated by atmospheric pressure of carbon dioxide in extended Pummerer rearrangement reactions. Only under CO₂ atmosphere, we observed moderate to high yields of disulfide starting from sulfur (IV) starting materials. Investigations on the reaction mechanism revealed that the degradation of the starting materials and the products was significant in the absence of CO₂. Further evidence for the suggested reaction mechanism was obtained by a cross-over experiment and a radical trapping reagent.

Full text of DOI:10.1016/j.tet.2020.131633

Journal Pre-proof
Extended Pummerer Fragmentation Mediated by Carbon Dioxide and Cyanide
Jian Liu, Rasmus R. Kragh, Fadhil S. Kamounah, Ji-Woong Lee
PII:
S0040-4020(20)30840-1
https://doi.org/10.1016/j.tet.2020.131633
TET 131633
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 19 July 2020
Revised Date: 4 September 2020
Accepted Date: 8 September 2020
Please cite this article as: Liu J, Kragh RR, Kamounah FS, Lee J-W, Extended Pummerer
Fragmentation Mediated by Carbon Dioxide and Cyanide, Tetrahedron, https://doi.org/10.1016/
j.tet.2020.131633.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Elsevier Ltd. All rights reserved.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Extended Pummerer Fragmentation
Leave this area blank for abstract info.
Mediated by Carbon Dioxide and Cyanide
Jian Liu, Rasmus R. Kragh, Fadhil S. Kamounah, Ji-Woong Lee
Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen Ø, Denmark

1
Tetrahedron
journal homepage: www.elsevier.com
Extended Pummerer Fragmentation Mediated by Carbon Dioxide and Cyanide
Jian Liua, Rasmus R. Kragha, Fadhil S. Kamounaha, and Ji-Woong Leea
aDepartment of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen Ø, Denmark
Dedicated to the Special Issue of 2020 Tetrahedron Young Investigator Awards for Professor Nuno Maulide
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
Pummerer rearrangement reactions generate sulfur (II) oxidation state from sulfur (IV) starting
materials in the presence of activating reagents. We found unprecedented transformation of
vinyl sulfoxide; disulfide formation reactions mediated by atmospheric pressure of carbon
dioxide in extended Pummerer rearrangement reactions. Only under CO2 atmosphere, we
observed moderate to high yields of disulfide starting from sulfur (IV) starting materials.
Investigations on the reaction mechanism revealed that the degradation of the starting materials
and the products was significant in the absence of CO2. Further evidence for the suggested
reaction mechanism was obtained by a cross-over experiment and a radical trapping reagent.
Available online
Keywords:
Pummerer rearrangement
Carbon dioxide
disulfides
2009 Elsevier Ltd. All rights reserved
.
cyanides
sulfoxides
thiosulfate via a sulfur redox process (Scheme 1C).9 This method
provides un-symmetric disulfides with alkyl halides in the
presence of over-stoichiometric amounts of thiosulfates.
1. Introduction
Sulfoxides and sulfones have a wide range of reactivity
profiles due to the ability of the sulfur atom to accommodate
various oxidation states.1 Various types of sulfur-based organic
reactions have evolved and showcased their application potentials
in organic synthesis. For example, the Pummerer rearrangement2,
3 can provide facile relocation of heteroatoms to access functional
group decorated hydrocarbon scaffolds (Scheme 1A). The
reaction is in general promoted by activating reagents, i.e. acetic
anhydride, to activate α-carbon of the sulfoxide. Based on this
traditional reaction, in 2018, Kaldre et al. reported a synthetically
useful stereodivergent strategy to access disubstituted 1,4-
dicarbonyl compounds from enantioenriched vinyl sulfoxides and
ynamides in the presence of a strong acid (Tf2NH) to initiate the
process (Scheme 1B).4 The characteristic sulfur(IV)-based
transformation showed elegant bond connectivity with high
enantio- and diasteroselectivity of the desired products, while the
corresponding thiol was presumably generated as a byproduct.
Nonetheless, most of sulfur (IV)-based transformation requires
strong activating reagents, therefore, often limiting the utility of
these reactions.
The formation of disulfides is a key biological transformation
that requires i) an oxidation process from the corresponding
thiols or ii) a substitution reaction with disulfides and derivatives
or iii) equilibrium controlled exchange reactions of disulfides.5-7
Considering the significant interests in dynamic disulfide
exchange reactions for supramolecular chemistry,8 we noticed
that a synthetic application of sulfur(IV)-based transformation to
disulfides is surprisingly rare. Recently, a metal-free approach to
disulfides was demonstrated starting from sulfinate and
Scheme 1. (A) Pummerer rearrangement (B) Sulfonium ion
rearrangement with ynamides mediated by a strong Brønsted
acid, (C) Comproportionation reactions of sulfinates and

2
thiosulfates, and (D) our CO2-mediated acid-free sulfoxide
activation to access symmetric disulfides.
Table 1. Optimization of reaction conditions.1
Entry Deviation from standard reaction Conv. (%)2
conditions
Our continuing investigations on CO2 utilization10-13 and
reported CO2-mediated organic reactions14 for organic synthesis
led us to attempt an oxidant-free formation of sulfur-sulfur
bonds, which can be applied in biologically active entities. A
protected sulfur (IV) would be an ideal starting point, where a
deprotecting process can trigger disulfide formation reactions, in
this case with CO2 and a metal cyanide.
1
None
98% (76%)
trace
12%
trace
31%
12%
64%
trace
95%
7%
2
N2 or H2 or O2 or air instead of CO2
3
DMF as a solvent
4
Other solvents3
5
at room temperature
30 min reaction time
12 h reaction time
KCN (1 equiv)
2. Hypothesis
6
We recently reported that carbon dioxide can be utilized as a
molecular catalyst promoting a 1,4-cyanation reaction, which
showed no reactivity in the absence of carbon dioxide
atmosphere.15 It was postulated that CO2 may act as a “acid
catalyst” in nucleophilic addition reactions using a cyanide,
where CO2 would reduce nucleophilicity of the cyanide by
forming cyanoformate.16 Nevertheless, the observed rate
acceleration and improved selectivity toward desired products
indicated the positive effects of CO2 when it was compared to
reactions under inert atmosphere (N2 or Ar atmosphere). We have
shown that vinyl sulfones and cyanides generate soluble
sulfinates, which undergo nucleophilic substitution reactions
affording functionalized sulfones and sulfonamides.17 The
reaction was driven by the formation of acrylonitrile, which can
undergo another cyanide additions step generating succinic
nitrile. Based on this study, we speculated that the employment
of vinyl sulfoxide would provide new ways to in situ generate
versatile sulfenate or thiolate nucleophiles for various organic
transformation. The use of CO2 as reaction medium can promote
chemical reactions by increasing availability of insoluble metal
cyanides in organic solvents. In this context, we sought that a
cyanide addition reaction is a good candidate since CO2 can
increase the solubility of insoluble metal cyanide (NaCN and
KCN) toward vinyl sulfoxide to induce Pummerer
rearrangement-like transformation to access disulfides. The
nucleophilic cyanide would react with an electrophilic vinyl
sulfoxide (extended Pummerer), and a following fragmentation,
generating thiolates or sulfonates, would lead to a disulfide
formation reaction under ambient reaction conditions..
7
8
9
KCN (3 equiv)
10
11
12
13
14
15
16
17
NEt4CN (2 equiv)
TMSCN (2 equiv)
additive NaOH (1 equiv)
additive K2CO3 (1 equiv)
additive NaHCO3 (1 equiv)
additive NEt3 (1 equiv)
acid additives**
3%
54%
94%
90%
92%
trace
84%
p-TSA (0.1 equiv)
1 Reactions were performed in a glass vial charged with KCN and
DMSO (3 ml). The solution was degassed for 1 min using CO2
stream, and the vial was sealed. Phenyl vinyl sulfoxide was
added through a rubber septum, and the reaction was stirred at 50
2
°C, under a CO2 atmosphere (1 bar). Conversion was recorded
by analyzing GCMS traces and compared with isolated yield.
3Tested solvents are; diethylether, 1,4-dioxane, ethylacetate,
water, n-heptane, toluene, THF, acetonitrile, chloroform,
tetrachloromethane,
1,2-dichloroethane
and
dichloromethane.**Tested acids: sulfuric acid, acetic acid, HCl.
It is noteworthy here that the stoichiometry of cyanide was
important presumably due to the formation of succinonirile
(detected by GC-MS under unoptimized reaction conditions)
from acrylonitrile. When only 1 equivalent of KCN was used
(entry 8), we detected only trace amounts of the product, while
almost full conversion to the product was observed with 3
equivalents of KCN (entry 9). The use of soluble cyanide sources
(NEt4CN and TMSCN) was unsuccessful to provide the disulfide,
highlighting the importance of insoluble cyanide source, KCN,
and its interactions with CO2 (entries 10 and 11). Acid and base
additives were tested: a similar result was detected with various
base additives (NaOH, K2CO3, NaHCO3 and NEt3) however
accompanying with significant amount of reduced starting
material 1a’ (entries 12-15). On the other hand, acid additives
were entirely detrimental for the desired transformation (entry
16), except p-toluenesulfonic acid displayed good conversion in
catalytic amounts (10 mol%, entry 17).
3. Results and Discussion
We initiated optimization of reaction conditions using
commercially-available phenyl vinyl sulfoxide under ambient
conditions. The starting material (1a) was smoothly converted to
the corresponding disulfide 2a in good conversion and isolated
yield (up to 76% yield after silica gel column chromatography).
We detected moderate conversion of the starting material to the
product at an earlier stage of the reaction process (3 h); under
atmospheric pressure of O2 (24%), H2 (50%), N2 (47%), open to
air (51%). However, the disulfide product (2a) was presumably
decomposed to other unrecognizable side products after
prolonged reaction time, indicating the labile nature of disulfides.
Therefore, no product was observed after 18h without CO2 (entry
2). Under CO2 atmosphere, on the other hand, quantitative
conversion (98%, GC-MS) to the product was observed in good
isolated yield (76%). Further optimization of reaction solvents
(entries 3, 4), temperatures (entry 5), reaction time (entries 6, 7)
showed no significant improvement.
Based on the time-dependent conversion profile (Figure 1) the
reaction reached maximum conversion after 5-6 hours, while
showing the lagging period (0-100 minutes) due to the low
solubility of KCN in organic solvents. We recently confirmed
that CO2 can improve the solubility of metal cyanide in organic
solvents.18

3
reaction conditions, two different substrates were converted to a
statistical mixture (1:2:1, 2b:2bg:2g) of three different disulfides,
which were detected by GC-MS and confirmed by isolating each
disulfide products by column chromatography. This experiment
confirms that our method renders an opportunity to form sulfur-
based nucleophiles or equivalents, where the reaction can reach
the equilibrium regardless of electronic state of the substrates to
100
80
60
40
20
0
1a
2a
KCN (2 equiv)
CO2 (1 atm)
O
S
S
R
R
S
DMSO, 50 o
C
R
Me
S
Me
S
S
S
S
S
Me
Me
0
100
200
300
400
, 49% yield
, 44% yield
, 49% yield
Time (min)
(20 mmol scale)
Figure 1. Time-dependent conversion of the starting material
(1a) and the formation of disulfide 2a.
tBu
OMe
S
S
S
S
MeO
S
S
OMe
Br
OMe
To quickly probe the reaction mechanism, we tested the
involvement of a radical reaction mechanism. An addition of
TEMPO showed complicated reaction mixture including a
trapped molecule of thiophenol (3) while generating disulfide
product 2a (Scheme 2A). This result indicates that radical
inhibition was not sufficient to shut down disulfide formation
process in the presence of 3 equivalent of TEMPO. Therefore, a
plausible reaction mechanism should involve two-electron
processes to generate thiophenol, sulfenate and the final product
disulfide (2a). A reduced product, phenyl vinyl sulfide 1a’ and
thiol adduct (4) derived from the vinyl sulfide were detected in
trace amounts by GC-MS analysis. Further mechanistic
experiments were performed using cyanide adduct of vinyl
sulfoxide 5. We isolated disulfide product 2a in 21% yield under
CO2 atmosphere under modified reaction conditions (NaOH
instead of KCN), whereas no desired product was obtained under
nitrogen atmosphere. Therefore, we concluded that the cyanide
adduct (5) is responsible for the formation of disulfide in the
tBu
, 37% yield
, 47% yield
, 38% yield
F
Cl
S
S
S
S
S
S
Br
F
Cl
, 55% yield
, 70% yield
, 65% yield
S
S
S
n-C12H25
S
n-C12H25
, 34% yield
, 31% yield
produce a statistical mixture of disulfides.
Scheme 3. Substrate scope
Scheme 4. Cross-over experiment.
presence of CO2.
Based on our observation and mechanistic studies, a plausible
reaction mechanism is illustrated in Figure 1. CO2 is undoubtedly
critical to initiate the reaction with vinyl sulfoxide 1 under the
reaction conditions. Therefore we presumed that CO2 can be
involved before, after and during the cyanide addition step,
mediating extended Pummerer reaction sequences. The
generation of carbonate leaving group at the additive Pummerer
reaction can generate sulfur (II) species from the starting
material. In the presence of excess amounts of KCN, it is
expected that the cyanide nucleophile can exchange
succinonitrile to generate PhS-CN, which can undergo
substitution reaction with thiol to produce disulfides. Also, the
base-mediated reaction of β-nitrile sulfoxide, which was
confirmed by a control experiment (Scheme 2B), can afford
sulfenate intermediate (B). The last step towards disulfides can
be completed by generating water or cyanide as a byproduct. The
formation of the carboxylic nitrile intermediate (A) is still
elusive. Although CO2 is thermodynamically stable, it has been
demonstrated that polymerization reactions of activated olefins
Scheme 2. Mechanistic studies with A) TEMPO and B) isolated
sulfoxide nitrile (5).
With optimized reaction conditions and preliminary
understanding of the reaction mechanism, we tested the scope of
the reaction with various vinyl sulfoxides. Our practical reaction
conditions enabled larger scale disulfide formation reactions from
substrates 1 to disulfides 2 in reasonable isolated yields without
sophisticated purification steps. It is noteworthy here that we
observed a potential reaction intermediate, PhS-CN, by GC-MS
analysis while optimizing large scale reaction. Electron rich and
deficient substrates were tolerated under reaction conditions,
affording symmetric disulfide in moderate to good yields. It is
important to note that CO2 atmosphere is essential to maintain
high conversion of the starting materials and purity of the
products.
Finally, to shed light on the reaction mechanism, we
conducted
a
competition experiment with electronically
differentiated substrates 1b and 1g. Under otherwise identical

4
5. Experimental section
(acrylate derivatives) can be controlled by CO2 where the
cyanocarboxylation prevents oligomerization.19 Under our
reaction conditions, insoluble polymer formation was observed,
which was water-soluble and showed aromatic and aliphatic
signals, indicating that the formation of sulfide, sulfone or
sulfoxide-based polymers. On the other hand, we found
acrylonitrile and maleic acid dinitrile leaving group under GC-
MS conditions, supporting the postulated reaction mechanism.
The formation of the thiol was indirectly confirmed by
conducting a set of control experiments with aromatic thiols
under CO2 atmosphere in the presence of KCN affording
disulfides in quantitative yield.
5.1. General method for disulfide formation reactions
A mixture of vinyl sulfoxide (1, 0.5-20 mmol) and 2.1
equivalents of potassium cyanide was stirred in DMSO (75 ml),
under a carbon dioxide atmosphere (1 atm) at 50 °C for 4 h. The
reaction was diluted with water and extracted with diethylether
three times. The combined organic phase was washed with water
and dried over MgSO4, and concentrated under reduced pressure.
The crude product was dissolved in n-heptane, and filtered
through a silica plug to afford disulfide 2 after evaporating
organic solvents.
CO2
Acknowledgments
O
CO2, CN
IV
S
RSSR
R
Generous support from the Department of Chemistry, University
of Copenhagen, Novo Nordisk Fonden (NNF17OC0027598),
Villum Fonden (00019062) is gratefully acknowledged.
H2O
extended
Pummerer
cyanocarboxylation
+ H+, base
CN
References and notes
O
C
O
S
OH
-
1. Kaiser, D.; Klose, I.; Oost, R.; Neuhaus, J.; Maulide, N., Chem.
Rev. 2019, 119, 8701-8780.
R
S
CO2
O
O
R
CN
CN
CN
S
R
2. Pummerer, R., Chem. Ber. 1909, 42, 2282-2291.
3. Bur, S. K.; Padwa, A., Chem. Rev. 2004, 104, 2401-2432.
4. Kaldre, D.; Klose, I.; Maulide, N., Science 2018, 361, 664.
5. Wang, M.; Jiang, X., Topics Curr. Chem. 2018, 376, 14.
6. Witt, D., Synthesis 2008, 2491-2509.
7. Wu, Z.; Pratt, D. A., J. Am. Chem. Soc. 2020, 142, 10284-10290.
8. Black, S. P.; Sanders, J. K. M.; Stefankiewicz, A. R., Chem. Soc.
Rev. 2014, 43, 1861-1872.
CN
R
CN
CN
NC
II
additive
Pummerer
cyanide
or base
S
CN
CN
CN
or
RS CN
NC
RSH
CN
2-
no CO2
required
-
9. Xiao, X.; Feng, M.; Jiang, X., Chem. Commun. 2015, 51, 4208-
4211.
10. Yang, Y.; Lee, J.-W., Chem. Sci. 2019, 10, 3905-3926.
11. Valera Lauridsen, J. M.; Cho, S. Y.; Bae, H. Y.; Lee, J.-W.,
Organometallics 2020, 39, 1652-1657.
12. Juhl, M.; Kim, M. J.; Lee, H.-Y.; Baik, M.-H.; Lee, J.-W., Synlett
2019, 30, 987-996.
Figure 1. A plausible reaction mechanism for disulfide formation
reactions mediated by CO2.
13. Juhl, M.; Lee, J.-W., Angew. Chem. Int. Ed. 2018, 57, 12318-
12322.
4. Conclusion
14. Schilling, W.; Das, S., Tetrahedron Lett. 2018, 59, 3821-3828.
15. Roy, T.; Kim, M. J.; Yang, Y.; Kim, S.; Kang, G.; Ren, X.;
Kadziola, A.; Lee, H.-Y.; Baik, M.-H.; Lee, J.-W., ACS Catal.
2019, 9, 6006-6011.
16. Murphy, L. J.; Robertson, K. N.; Harroun, S. G.; Brosseau, C. L.;
Werner-Zwanziger, U.; Moilanen, J.; Tuononen, H. M.; Clyburne,
J. A. C., Science 2014, 344, 75.
17. Roy, T.; Lee, J.-W., Synlett 2020, 31, 455-458.
18. Juhl, M.; Petersen, R. A.; Lee, J.-W., Chem. Eur. J. DOI:
10.1002/chem.202003623.
19. White, D. A., J. Chem. Soc., Perkin Trans. 1 1976, 1926-1930.
20. Wang, S.; Cheng, B.-Y.; Sršen, M.; König, B., J. Am. Chem. Soc.
2020, 142, 7524-7531.
21. Ruan, H.; Meng, L.-G.; Zhu, L.; Wang, L., Adv. Synth. Catal.
2019, 361, 3217-3222.
22. Kapoor, M.; Chand-Thakuri, P.; Young, M. C., J. Am. Chem. Soc.
2019, 141, 7980-7989.
23. Kapoor, M.; Liu, D.; Young, M. C., J. Am. Chem. Soc. 2018, 140,
6818-6822.
24. Pupo, G.; Properzi, R.; List, B., Angew. Chem. Int. Ed. 2016, 55,
6099-6102.
We developed an oxidant-free disulfide formation reaction
starting from sulfur (IV) vinyl sulfoxides mediated by CO2 and
potassium cyanide. Under our reaction conditions, we can not
21
exclude the formation of thiyl radical.20,
However, control
experiments and competition experiments showed a hint of
potential reaction mechanism where CO2 is necessary for the
desired reactivity of vinyl sulfoxides. Various aryl disulfide and
an alkyl disulfide were prepared using the optimized reaction
conditions partially owing to the fact that the decomposition of
substrates and products was significantly retarded under CO2.
This preliminary study showed that the disulfide bond formation
can be triggered by a cyanide, necessarily under CO2 atmosphere.
The positive CO2-effects in new organic transformations is
intriguing since this unprecedented reaction is not attainable
under inert atmosphere. The origin of CO2 effect can simply be
ascribed to the solubilized CO2, altering polarity of the solvent
system. However, we reasoned that the chemical reactivity of
CO2 can be significantly important when reversible formation of
CO2-adducts can be postulated. Based on recent developments,22-
25. Riemer, D.; Mandaviya, B.; Schilling, W.; Götz, A. C.; Kühl, T.;
Finger, M.; Das, S., ACS Catal. 2018, 8, 3030-3034.
26. Ye, J.; Kalvet, I.; Schoenebeck, F.; Rovis, T., Nat. Chem. 2018,
10, 1037-1041.
26
CO2 can be used as a temporary protecting group or an
activating reagent, considering electrophilic nature of the sp
carbon of CO2. Therefore, the utility of CO2 will be further
expanded not only for the synthesis, but also for equilibrium
mediated processes, specially where stimuli-triggered disulfide
bond formation reactions are needed.
Supplementary Material
Supplementary material that may be helpful in the review
process should be prepared and provided as a separate electronic
file. That file can then be transformed into PDF format and

5
submitted along with the manuscript and graphic files to the
appropriate editorial office.
Click here to remove instruction text...

•
•
•
•
CO2 dramatically changes the course of Pummerer-type reaction
Higher yield and selectivity under CO2 compared to other inert gases
Mechanistic studies to reveal the novel transformation
CO2 can replace highly reactive electrophilic activating reagents

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests: