Electrochemical Synthesis of Organic Polysulfides from Disulfides by Sulfur Insertion from S8 and an Unexpected Solvent Effect on the Product Distribution

Article abstract of DOI:10.1002/chem.202101023

An electrochemical synthesis of organic polysulfides through sulfur insertion from elemental sulfur to disulfides or thiols is introduced. The highly economic, low-sensitive and low-priced reaction gives a mixture of polysulfides, whose distribution can be influenced by the addition of different amounts of carbon disulfide as co-solvent. To describe the variable distribution function of the polysulfides, a novel parameter, the “absorbance average sulfur amount in polysulfides” (SAP) was introduced and defined on the basis of the “number average molar mass” used in polymer chemistry. Various organic polysulfides were synthesized with variable volume fractions of carbon disulfide, and the yield of each polysulfide was determined by quantitative ¹³C NMR. Moreover, by using two symmetrical disulfides or a disulfide and a thiol as starting materials, a mixture of symmetrical and asymmetrical polysulfides could be obtained.

Full text of DOI:10.1002/chem.202101023

Accepted Article
Accepted Article
Title: Electrochemical Synthesis of Organic Polysulfides from Disulfides
via Sulfur Insertion from S8 and an Unexpected Solvent Effect on
the Product Distribution
Authors: Jan Fährmann and Gerhard Hilt
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: Chem. Eur. J. 10.1002/chem.202101023
Link to VoR: https://doi.org/10.1002/chem.202101023
01/2020

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
Electrochemical Synthesis of Organic Polysulfides from Disulfides
via Sulfur Insertion from S8 and an Unexpected Solvent Effect on
the Product Distribution
Jan Fährmann[a] and Gerhard Hilt*[a]
[a]
MSc. J. Fährmann, Prof. Dr. G. Hilt
Institut für Chemie
Universität Oldenburg
Carl-von-Ossietzky-Straße 9-11, 26111 Oldenburg, Germany
E-mail: gerhard.hilt@uni-oldenburg.de
Supporting information for this article is given via a link at the end of the document.
Abstract: An electrochemical synthesis of organic polysulfides via
sulfur insertion from elemental sulfur to disulfides or thiols was
introduced. The highly economic, insensitive and low-priced reaction
gives a mixture of polysulfides, whose distribution can be influenced
by the addition of different amounts of carbon disulfide as co-solvent.
To describe the variable distribution function of the polysulfides, a
novel parameter, the “absorbance average sulfur amount in
polysulfides” (SAP) was introduced and defined on the basis of the
“number average molar mass” used in polymer chemistry. Various
organic polysulfides were synthesized with variable volume fractions
of carbon disulfide and the yield of each polysulfide was determined
by quantitative 13C NMR. Also, by using two symmetrical disulfides or
a disulfide and a thiol as starting materials, a mixture of symmetrical
and unsymmetrical polysulfides was obtained.
Introduction
Organosulfides play an important role in various parts of
everyday life. Thiols flavor food like onion, cheese and garlic.[1]
Amino acids like cysteine contain thiols to crosslink peptides
forming disulfides, e.g. keratin in human hair.[2] Common
stinkhorn’s (phallus impudicus) intense smell is produced by
dimethyl disulfide and dimethyl trisulfide (DMTS).[3] Even higher
polysulfides can be found in nature, lenthionine from shiitake
mushrooms serves as an example.[4] A high level of interest is
currently given to the inorganic lithium-sulfur (Li–S) batteries. Due
to their excellent specific capacity and energy density, Li–S
batteries may play an important role in global energy transition,
specifically in electric cars and portable device-research.[5]
Recent studies from QIAN and WEN showed that the addition of
organic trisulfides like DMTS to Li–S batteries “enhances the
sulfur utilization rate and facilitates capacity performance”.[6] In
the battery, the trisulfide reacts via sulfur insertion to organic
polysulfides, which then form the soluble lithium organo-
polysulfanes and insoluble lithium sulfide – the driver for an
enhanced sulfur utilization rate.[6,7]
Scheme 1. Previous work in organic polysulfide synthesis.
new route for the synthesis of unsymmetrical trisulfides.[8] The
reaction of nucleophilic 9-fluorenylmethyl (Fmoc) disulfides with
electrophilic S-succinimide derivatives generates the desired
mixed trisulfides (Scheme 1). Several unsymmetrical trisulfides
could be synthesized in high yields using this method. However,
the reaction is laborious and not highly atom-economic. For
example, the desired reactants need to be synthesized from thiols
before the coupling can be performed. Also, this synthetic route is
limited to the formation of trisulfides. The use of elemental sulfur
in organic synthesis is rare – especially in polysulfide synthesis.
One single approach was presented by YAMAGUCHI et al. in 2005.
They discovered, that a rhodium-catalyzed reaction exchanges
sulfur atoms between elemental sulfur and organic disulfides for
the synthesis of a range of organic polysulfides.[9] Thereby,
organo-polysulfides (up to heptasulfides were detected) could be
synthesized in a short reaction time (5 min). However, the
expensive transition metal rhodium, albeit used in catalytic
amounts, is needed. The electroorganic synthesis is currently in
the focus of many scientists due to its mild, efficient and mostly
low priced reaction setup.[10] An easy way for the electrochemical
Several publications of the past few years deal with the
synthesis of organic polysulfides. In 2018, XIAN et al. reported a
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
synthesis of unsymmetrical disulfides was reported last year from
our group. Using alternating current electrolysis, we were able to
obtain a statistical distribution of disulfides, starting from two or
more symmetrical disulfides.[11] This fast and highly atom-
economic process produced only disulfides and no trisulfides or
higher organic polysulfides were found.
Table 1. Optimization experiments.
In this article, we present an electrochemical approach for the
synthesis of organic trisulfides and higher sulfides from their
corresponding disulfides or thiols. We developed an easy, robust
and low-priced method without the need for any catalysts or
additives to generate organic polysulfides.
entry
changes from initial conditions
conversion 1a[a]
yield 1b[a]
1
acetonitrile as solvent
none
68%
71%
71%
69%
0%
27%
31%
24%
28%
0%
2
3
nBu4NPF6 as electrolyte
nBu4NBr as electrolyte
nBu4NOTos as electrolyte
Et4NClO4 as electrolyte
0.05 M nBu4NBF4
0.30 M nBu4NBF4
0.50 M nBu4NBF4
stainless steel electrodes
Ni electrodes
Results and Discussion
4
Optimization
5
6
96%
72%
67%
62%
0%
7%
When we initially investigated the electrochemical organo-
polysulfide synthesis, we adopted the reaction conditions
reported for the disulfide metathesis reaction,[11] using acetonitrile
as solvent, tetrabutylammonium tetrafluoroborate (nBu4NBF4) as
supporting electrolyte and platinum electrodes in an undivided cell.
Two atom-equivalents of sulfur-atoms (= 2.0 [S])[12] were added to
di-n-butyl disulfide (DBDS, 1a) and the pH neutral solution was
electrolyzed at 10 mA constant current under non-inert conditions
(Table 1, entry 1). Monitoring the reaction progress via reverse-
phase HPLC with UV-detector (RP-HPLC-UV) indicated 68%
conversion of the disulfide 1a, 85% conversion of sulfur and the
formation of polysulfides up to the undecasulfide 1j in an acidic
product solution. A representative chromatogram is shown in
Figure 1. The determination of the yield via HPLC-UV analysis is
challenging due to the unknown absorption coefficients of the
polysulfides. Separation of the polysulfide mixture is not possible
via simple silica-gel chromatography due to their very similar
polarity and a preparative HPLC was not available for us at that
time. Also, the detection of polysulfides from a sulfur-chain length
of four or more by GC-FID could not be achieved. Luckily, the
distribution of polysulfides was not affected by different reaction
conditions – a circumstance applying to the synthesis route from
YAMAGUCHI as well.[9] That means, the polysulfide ratio was the
same for all optimization experiments, while their absolute yield
changed. Thereby the conversion of the starting material and the
yield of a single polysulfide could be used for the optimization of
the reaction conditions. Since the starting material 1a and
trisulfide 1b can be detected by GC-FID, we chose this fast and
precise method for the optimization. Therefore, it is important to
keep in mind, that the yield of the trisulfide does not reflect the
overall yield of the polysulfides, it rather serves as an indicator for
a higher/lower yield in the polysulfide formation 1b–1j.
7
28%
27%
23%
0%
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
34%
40%
71%
74%
71%
70%
78%
73%
74%
69%
75%
13%
74%
70%
74%
76%
0%
0%
Cu electrodes
0%
graphite electrodes
GC electrodes
5 mA
30%
35%
31%
30%
18%
35%
30%
23%
10%
0%
20 mA
40 mA
0 °C
35 °C
quasi-divided cell
divided cell (anolyte)[b]
divided cell (catholyte)[b]
inert atmosphere
entry 23 + anhydrous solvent
entry 24 + 2.0 eq. H2O
1.0 mmol scale[c]
0.0 F (160 min stirring)
6.5 F
35%
29%
35%
36%
0%
77%
35%
Unless otherwise stated, 0.25 mmol disulfide 1a (1.0 eq.) were used in a total
volume of 10 mL. Dimensions of the electrodes 35x10x0.5 mm, immersion
depth of the electrodes 1.5 cm, electrode spacing 1.5 cm. Changes from
highlighted entries were taken on following experiments. *2 [S] = 2/8 S8. [a]
Determined by GC-FID analysis of the crude reaction mixture with n-dodecane
as internal standard. [b] A 0.3 M electrolyte concentration was used. [c] Total
volume was 20 mL.
1c
1b
ISTD
1d
1e
Since the solubility of sulfur in acetonitrile is rather low, we
varied the solvent in the beginning of the reaction optimization.
We tested various solvents (tetrahydrofuran, dimethoxyethane,
acetone, dimethyl sulfoxide, dimethylformamide, ethanol, 2-
propanol and dichloromethane). Of all tested solvents, only
dichloromethane (DCM) was also suitable for the polysulfide
synthesis. In all other solvents, the sulfur remained insoluble
during the electrolysis or side reactions with the solvent took place.
We decided to use DCM as a significantly better sulfur-solubilizing
S8
1f
1g
1a
1h
15
1j
1i
0
5
10
20
25
retention time / min
Figure 1. Representative RP-HPLC-UV chromatogram of the crude reaction
mixture 1a–1j at 248 nm with methanol as eluent. Benzophenone was used as
internal standard (ISTD).
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
solvent than acetontitrile for the further investigations (entry 2).
Thereby the conversion of the starting material increased slightly
to 71%. In the following, we varied the electrolyte and its
concentration in solution (entries 2-9). Formation of the
polysulfides 1b–1j remained high when nBu4NPF6 was applied as
supporting electrolyte (71% conversion, 24% yield of 1b, entry 3).
Utilizing nBu4NBr as a low-priced supporting electrolyte is a good
alternative to the BF4 salt, since the formation of the trisulfide
remained high (28% of 1b, entry 4). While the application of
nBu4NClO4 resulted in decomposition of the starting material
(entry 5), as no by-products were detected by GC-FID or HPLC-
UV, the use of nEt4NOTs led to a complete collapse of disulfide
conversion (entry 6). Due to the good results and high
conductivity, we decided to use nBu4NBF4 as the electrolyte of
choice, but nBu4NBr is also recommended as less expensive
alternative. The concentration of the supporting electrolyte can be
varied between 0.05 M and 0.3 M without significant reduction of
product formation (entries 7-9). An electrolyte concentration of
0.5 M is accompanied with a slight decrease of product formation
(62% conversion of 1a to 23% 1b, entry 9). For the best results
and a high conductivity, we decided to hold on to the initial
electrolyte concentration of 0.1 M.
When we varied the electrode materials, stainless steel did
not give any conversion of the starting material (entry 10) and
nickel as well as copper electrodes gave a low conversion,
however, no polysulfides were observed by GC-FID and HPLC-
UV (entries 11/12). Thus, we assumed decomposition of the
disulfide when using these electrodes. On the other hand,
graphite electrodes (see supporting information) proved to be a
good alternative to platinum electrodes, whose results in
conversion and yield were almost identical (entry 13). When
glassy-carbon (GC) electrodes were used, 35% trisulfide 1b were
formed at a conversion of 74% 1a, so we decided to use GC
electrodes for further optimization.
the polysulfide synthesis, additionally working under anhydrous
conditions even lowered the yield of the trisulfide 1b to 29%
(entry 24). When two equivalents of water were added to the
anhydrous solvent (entry 25) identical results were obtained as for
the electrolysis under non-inert conditions, implying that small
amounts of water are useful for this reaction.
The reaction could be performed at a larger scale; e.g. when
the scale was quadrupled while doubling the amount of solvent
(entry 26), without loss of efficiency for the formation of organic
polysulfides. The higher concentration and amount of substrates,
associated with a longer reaction time, were well tolerated. Finally,
in entry 27 we set the standard conditions without applying any
current. After 160 min of stirring (= electrolysis time at standard
conditions), no conversion of the starting material and no organic
polysulfides were observed, which in conclusion proves that
electrochemical current is needed to perform this reaction.
Figure 2 (left) portrays the relative distribution of the organic
polysulfides 1a–1j depending on the amount of applied current.
As can be seen, the maximum amount of polysulfides is reached
after 4.0 F. Figure 2 (right) confirms a plateau in conversion of
disulfide 1a and sulfur at 4.0 F and further electrolysis of up to
6.5 F did not change the polysulfide distribution mixture. Only a
very slight decrease in the polysulfide yield was observed due to
decomposition while the polysulfide distribution remained
identical (Figure 2 left, Table 1, entry 28). A possible explanation
for the need of 4.0 F of current is discussed in the mechanistic
studies part.
100
high
6
80
4
60
40
20
0
4.0 F
4.0 F
2
0
nBu2S2
sulfur
6
The applied current was varied between 5 mA and 40 mA
(entries 15-17). Further increase above 40 mA caused too much
evolution of heat, leading to evaporation of the DCM and collapse
of the reaction. A low current between 5 mA and 20 mA did not
affect the conversion at all (entries 15 and 16), however a high
current of 40 mA led to decomposition of the starting material
(78% conversion of 1a, 18% yield of 1b, entry 17). For a
reasonably fast product formation and a high yield, we decided to
keep the applied current at 10 mA, but for an even faster reaction
20 mA can be applied without the risk of significant loss of
polysulfide products.
low
0
1
2
3
4
5
7
2
4
6
8
10
applied current / F
N(S) in nBu2SN
Figure 2. Graphical presentation of (left) the relative amount of formed organic
polysulfides determined by HPLC-UV and (right) conversion of the starting
materials 1a and elemental sulfur in dependence of the applied current.
Solvent effect on the polysulfide distribution
During preliminary tests, we observed a correlation between
the amount of sulfur added to the reaction and the polysulfide
distribution. Additionally, the volume fraction of added carbon
disulfide (Attention, CS2 should be handled with care!) did affect
the distribution as well. A manual adjustment of distributions from
polysulfide synthesis with elemental sulfur has not been reported
so far. Consequently, we decided to investigate the solvent effect
of CS2 on the polysulfide distribution in more detail. For their
analysis, we referenced the HPLC-UV integral of each experiment
to an internal standard (benzophenone), a representative HPLC-
UV chromatogram is shown in Figure 1. With this, we were able
to determine the conversion of sulfur and disulfide 1a as well as
the referenced HPLC-UV integrals of organic polysulfides 1b–1j
(Figure 1). To set up a mathematical comparison of the integral
ratios, we introduced the “absorbance average sulfur amount in
polysulfides at 248 nm” (SAP248) analogous to the “number
average molar mass” used in polymer chemistry,[14,15] given as:
Decreasing the reaction temperature to 0 °C does not have
an impact on the polysulfide synthesis (entry 18), but increasing
it to 35 °C reduced the formation of 1b to 30% (entry 19). In order
to keep the setup as simple as possible, we decided to perform
further electrolysis at ambient temperature (around 21 °C).
Entry 20 shows that varying the cell design does not improve the
formation of organic polysulfides. When the reaction was
performed in a quasi-divided cell[13] the yield of trisulfide 1b
dropped to 23%. To maintain a high conductivity in a divided cell,
an electrolyte concentration of 0.3 M was necessary. In the anode
compartment only 10% trisulfide were detected (entry 21), while
in the cathode compartment no organic polysulfides were formed
(entry 22). Accordingly, the simplest, and, with respect to the
efficiency of the reaction, the best design is an undivided cell. The
optimization experiments were carried out under non-inert
conditions with undried solvents and supporting electrolytes. In
entry 23 we demonstrate that an inert atmosphere did not improve
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
∞
∑
( )
푛푖,248 ∙ 푁 S
Table 2. Dependency of the polysulfide distribution (SAP and dispersity Đ) on
푖
푖=1
SAP
=
(1)
248
∞
∑
the sulfur amount added to the reaction solution and volume fraction of CS2.
푖=1 푛푖,248
푛푖,248: referenced HPLC-integral of polysulfide 푖 at 248 nm
( )
푁 S : number of sulfur equivalents in polysulfide 푖
푖
This value allows the comparison of experiments with different
polysulfide distributions. Similar to the “number average molar
mass”, the SAP reflects the position of the maximum of
polysulfides’ distribution function. But, the SAP is depending on
the absorption coefficient-relations of the polysulfides. Therefore,
the SAP is suitable for comparison within one sulfide species but
does not give information about the actual favored sulfide or
absolute yield. Additionally, we calculated the dispersity Đ as
applied in polymer chemistry (for the calculation, see supporting
information).[15,16] The dispersity is given as a unified number and
is always ≥ 1. A dispersity of 1.0 describes a uniform distribution,
and therefore a single sulfide, e.g. 100% trisulfide formation. The
higher the dispersity, the more unspecific a single sulfide is
formed.
Two atom-equivalents of sulfur (2.0 [S]), as used in the
optimization experiments, gave the highest conversion rate of the
disulfide 1a (74%) and an almost quantitative conversion of sulfur
(95%). The SAP was 5.34 at a dispersity of 1.13 (Table 2, entry 4).
Using up to four atom-equivalents of sulfur did not affect the
conversion of the starting materials (entries 5–6). The SAP
increased slightly up to 5.46, while the absolute consumption of
sulfur stayed the same and excess sulfur simply remained in
solution. The yield and distribution of polysulfides was almost
identical between two and four atom-equivalents of sulfur, further
increase of the sulfur atom-equivalents to eight equivalents,
however, shifted the SAP to 5.95 (Figure 3 left, entry 7). Another
indicator for this is shown by the absolute conversion of sulfur,
which increased by 17% towards two atom-equivalents of sulfur.
Reducing the amount of sulfur to one atom-equivalent led to a
sulfur-deficiency, resulting in a 20% lower conversion of the
disulfide, accompanying with a huge shift of the SAP towards
lower polysulfides (SAP 3.83, entry 3). Further sulfur-deficiency
(0.5 atom equivalents) comes along with an even lower SAP of
3.29 (entry 2). As expected, no polysulfides were formed at all
without the addition of any sulfur (entry 1). In summary, the
average number of sulfur atoms in polysulfides can be influenced
by the amount of added sulfur atom-equivalents. The dispersion
for all these experiments is constant – but it is relatively high at
1.13 to 1.15. This is visualized by a broad maximum in Figure 3,
left.
SAP248
Đ
entry
conditions
conversion of
1a
[S]
1
2
0.0 [S], pure DCM
0.5 [S], pure DCM
46%
–
–
–
44%
54%
74%
74%
73%
72%
84%
84%
84%
86%
80%
85%
88%
93%
0%
98%
98%
95%
63%
47%
28%
–
3.29
3.83
5.34
5.41
5.46
5.95
4.55
4.47
4.43
4.46
4.52
4.41
4.86
5.15
–
1.21
1.13
1.13
1.14
1.15
1.13
1.11
1.06
1.06
1.06
1.08
1.07
1.07
1.07
–
3
1.0 [S], pure DCM
4
2.0 [S], pure DCM
5
3.0 [S], pure DCM
6
4.0 [S], pure DCM
7
8.0 [S], pure DCM
8
0.0 [S], 10% (vol.) CS2
1.0 [S], 10% (vol.) CS2
2.0 [S], 10% (vol.) CS2
4.0 [S], 10% (vol.) CS2
8.0 [S], 10% (vol.) CS2
2.0 [S], 2% (vol.) CS2
2.0 [S], 25% (vol.) CS2
2.0 [S], 50% (vol.) CS2
entry 10 + no current
9
98%
91%
45%
18%
73%
90%
90%
0%
10
11
12
13
14
15
16
Unless otherwise stated, 0.25 mmol disulfide 1a (1.0 eq.) were used in a total
volume of 10 mL. Dimensions of the electrodes 35x10x0.5 mm, immersion
depth of the electrodes 1.5 cm, electrode spacing 1.5 cm. For a graphical
presentation of the referenced integral see Figure 3. *x [S] = x/8 S8. [a]
Determined by HPLC-UV analysis at 248 nm of the crude reaction mixture with
benzophenone as internal standard.
material raised up to 86% (entry 10). Also, the SAP lowers to 4.43
and the dispersity is considerably reduced to 1.06, giving a
significantly narrower distribution of polysulfides. This finding is
impressively visualized in Figure 3, center. For this 10:1 mixture,
varying the atom-equivalents of sulfur between one to eight did
not affect the dispersity, which stayed constant between 1.06 to
1.08 (entries 9-12). Different to the experiments in pure DCM, the
SAP is also quiet constant between 4.43–4.52. Only the absolute
When a solvent-mixture of DCM:CS2 with 10% (vol.) is used
with two atom-equivalents of sulfur, the conversion of the starting
0.0 eq. [S], 10% CS2
2.0 eq. [S], 10% CS2
8.0 eq. [S], 10% CS2
1.0 eq. [S], pure DCM
2.0 eq. [S], pure DCM
8.0 eq. [S], pure DCM
2.0 eq. [S], pure DCM
2.0 eq. [S], 10% CS2
2.0 eq. [S], 50% CS2
2
4
6
8
10
2
4
6
8
10
2
4
6
8
10
N(S) in nBu2SN
N(S) in nBu2SN
N(S) in nBu2SN
Figure 3. Graphical presentation of the referenced HPLC-UV integrals in dependency of (left) sulfur amount added in pure DCM, (centre) sulfur amount added in
DCM with 10% (vol.) CS2 and (right) varied amount of CS2. See Table 2 for further information and the supporting information for all graphs (colorized).
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
yield of the polysulfides changed clearly (see Figure 3). The
maximum yield of polysulfides is between two to four atom-
equivalents sulfur (entries 10/11) as it was the same as for the
experiments in pure DCM. Interestingly, a relatively high
conversion of disulfide 1a to polysulfides was observed even
without addition of any sulfur (entry 8). It seems, that CS2 does
not only act as a pure solvent, but unfortunately we could not
clarify its exact role in this reaction as cyclic voltammetry and
other spectroscopic methods did not give meaningful results in
this respect. To affect the SAP when using a DCM:CS2 mixture,
the amount of added CS2 was varied. Again, the dispersity
remains the same over a wide range. Between a proportion from
2% (vol.) to 50% (vol.) the dispersity was constant at 1.07,
displaying an extreme narrow distribution (entries 13–15). The
SAP however depended highly on the solvent ratio. It appears,
that the higher the amount of CS2 the higher the SAP (Figure 3,
right). While the SAP was 4.41 at 2% (vol.) CS2, it rose gradually
up to 5.15 at 50% (vol.) CS2 solvent content. The latter conditions
gave the highest conversion of the disulfide of all experiments
(93%, entry 15) due to its narrow dispersity at high polysulfides.
used in pure DCM as solvent. As shown in Table 2, a wide
distribution of polysulfides was expected for these reactions. In
setup B, 10% (vol.) CS2 were added, promising a noticeably
narrower dispersity and probably a higher conversion of the
starting material. For the investigation of the substrate scope, we
determined the absolute yield of the organic polysulfides by
integration of the signals of all separated α-carbons (R–[S]N–CH2–
signals) in quantitative 13C NMR utilizing an internal standard.
Unseparated polysulfides were summed up under “further”
polysulfides. For disulfide 1a, we already performed extensive
investigations in Table 2. However, only relative results were
given, depending on the absorption coefficients of the polysulfides.
In 13C NMR, the quantitative amount of the polysulfide mixture
1b–1j was detected (Table 3, entry 1), which confirms that no
decomposition reactions occurred during electrolysis. Also, the
conversion of the disulfide 1a (determined by 13C NMR) matched
the conversion calculated from integration of the HPLC and GC
signals (74% HPLC and GC, 76% NMR in setup A; 84% and 81%
in setup B). The absolute yield of the trisulfide 1b was also
consistent to the GC results within an acceptable margin of error
(35% GC, 30% NMR). Overall, a good comparability between
these analytical methods was determined.
Substrate scope
When 1a was used as starting material, the trisulfide 1b was
the most populated species, followed by the corresponding
tetrasulfide 1c (22%, Table 3, entry 1). The SAP in Table 2
illustrates a population-shift towards lower sulfides when 10% vol.
Due to the strong dependency of the polysulfide distribution
on the solvent mixture, we decided to continue with two different
reaction setups. In setup A two atom-equivalents of sulfur were
Table 3. Substrate Scope.
entry
1
substrate
setup
conversion of
1a–5a / 6[b]
yield of[b]
[S][a]
95%
94%
63%
1b–6b 1c–6c 1d–6d 1e–6e further[c] sum
A
B
A
77%
81%
72 %
30%
41%
9%
22%
24%
28%
8%
3%
4%
2%
9%
12%
8%
–
77%
78%
67%
1a
2
21%
B
A
B
A
70%
99%
52%
50%
83%
29%
36%
49%
17%
10%
24%
28%
35%
3%
22%
–
5%
–
–
–
–
–
79%
13%
34%
46%
2a
3a
3
4
8%
2%
4%
–
12%
2%
B
A
49%
85%
52%
62%
28%
32%
13%
10%
5%
4%
3%
1%
–
–
49%
47%
4a
5
6
B
27%
62%
34%
12%
6%
4%
–
56%
5a
6
A
B
89%
94%
80%[e]
95%[e]
34% 6a[f]
27% 6a[f]
22%
45%
17%
19%
n.r.[c]
n.r.[c]
n.r.[c]
n.r.[c]
6%
4%
79%
95%
Unless otherwise stated, 0.25 mmol disulfide 1a–5a (1.0 eq.) were used in a total volume of 10 mL. Dimensions of the electrodes 35x10x0.5 mm, immersion depth
of the electrodes 1.5 cm, electrode spacing 1.5 cm. *2 [S] = 2/8 S8. [a] Determined by HPLC-UV analysis at 248 nm of the crude reaction mixture with benzophenone
as internal standard. [b] Determined via quantitative 13C NMR by integration of the α-carbons (R–[S]N–CH2– signals) with benzophenone as internal standard. [c]
The integral of unseparated signals in 13C NMR were summed up under “further”. Entries are marked as n.r. (= not resolved). [d] Two equivalents of the thiol were
used in relation to the amount of disulfide under standard conditions. [e] Conversion of the thiol was determined, yield of the disulfide is given on the right. [f] 6a =
di-n-dodecyldisulfide.
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
CS2 were added as co-solvent. In fact, we observed a higher yield
of trisulfide 1b (41%) and lowered yields of the high sulfides 1d–
1j by the same ratio. For dicyclohexyl disulfide (2a), a similar
behavior compared to the n-butyl derivate 1a was observed. A
high conversion of 2a led to a wide polysulfide distribution in
setup A (entry 2). Different to the di-n-butyl disulfide (1a), the
tetrasulfide 2c is the most populated species with 28% followed
by the pentasulfide 2d (21%) and no higher polysulfides (N > 4)
were observed. Again, the conversion of the starting material
increased in setup B (83%). An increased yield of the trisulfide 2b
was observed to an absolute yield of 17%. For the tetrasulfide 2c,
the same increase in yield was observed. Both substrates confirm
a high response of dialkyl disulfides to the electrochemical
polysulfide synthesis and a positive effect of CS2 as co-solvent in
the reaction.
Aromatic disulfides 3a–5a showed in general a lower
conversion compared to the aliphatic disulfides 1a and 2a. In pure
DCM, diphenyl disulfide (3a) formed 10% trisulfide 3b, 3%
tetrasulfide 3c and no higher sulfides (entry 3, setup A). Also,
diphenyl disulfide (3a) seemed to undergo decomposition
reactions due to the loss of 16% of the polysulfide mixture, most
likely caused by deposits which were observed on the cathode
over the course of the electrolysis. Inexplicably, the conversion of
elemental sulfur in this experiment was quantitative. When CS2
was added as co-solvent in setup B, the decomposition of 3a was
prevented completely. The yield of every polysulfide doubled
(3b 24%, 3c 8%) and the conversion of sulfur adjusted to 52%.
The electron-rich aromatic disulfide 4a converted in moderate
50% resulting in 28% yield of trisulfide 4b, 13% of tetrasulfide 4c
and small amounts of further polysulfides. The 4-methoxy-
substituted derivate 5b showed a similar distribution as the p-tolyl
sulfides 4a–4e but a slightly higher conversion of the disulfide
(62%, entry 4). The distribution of both electron-rich diaryl
disulfides 3a and 4a did not change when CS2 was added as co-
solvent. The conversion of the disulfide 3a/4a as well as the yields
of the polysulfides 3b–3e/4b–4e remained nearly stable.
giving a high yield of low polysulfides (45% 6b, 19% 6c) alongside
the corresponding disulfide 6a.
Synthesis of unsymmetrical polysulfides
Since the formation of polysulfides is possible from thiols, an
electrochemical cleavage of the sulfur-sulfur bond is plausible.
The electrocatalytic synthesis of mixed disulfides, reported from
our group last year, illustrated such a sulfur-sulfur bond
metathesis.[11] To prove the possibility of mixed polysulfide-
formation for this reaction, we electrolyzed 0.5 atom-equivalents
of 1a and 2a each (in relation to the symmetrical polysulfide
synthesis) at reaction setup B. The corresponding HPLC-
chromatogram is shown in Scheme 2. An assignment of the
signals in 13C NMR could not be achieved due to the large number
of signals and the determination of the yields was therefore not
possible. However, secure identification of the mixed polysulfides
was possible via HRMS.
In the next step, we added 1.0 equivalent of thiol 6 to
0.5 equivalents of disulfide 2a. Again, we assigned each signal of
the HPLC-chromatogram in Scheme 3. Large amounts of the
mixed polysulfides 8a–8f were supposed and again confirmed by
HRMS. These experiments showed that the formation of mixed
polysulfides can be accomplished by mixing either two disulfides
or a disulfide and a thiol and applying 4.0 F of electricity to form a
large number of symmetrical and unsymmetrical organic
polysulfides. On the one hand, these results confirm a S–S bond
cleavage according to our previous study,[11] for this reaction. On
the other hand, this might be useful to enlarge the application of
polysulfides in dynamic libraries since many products can be
synthesized from few starting materials in a single reaction.
In addition to disulfides, thiols can be used as starting
materials for the synthesis of polysulfides as well (entry 6). For
the non-volatile thiol n-dodecanethiol (6) also applied an improved
formation of the polysulfides when CS2 was added as co-solvent.
When the reaction conditions of setup B were used, nearly full
conversion of the starting material was observed after 4.0 F,
S8
8c
2b
2c
8d
8b
2d
2a
8e
6c
8a
2e
6b
6d
6e
8f
6a
10
50
100 130
retention time / min
7b
2a
Scheme 3. RP-HPLC-UV chromatogram of the crude reaction mixture 8a–8e
with 2a–2f and 6a–6f at 248 nm with methanol as eluent.
2b
S8
1b
1c
1e
7a
2c
Mechanistic studies
1d
7c
1a
5
2d
2e
7d
Mechanistic studies were performed using cyclic voltammetry
(see supporting information), but only a few meaningful results
could be obtained. However, several hints to a possible
mechanism were discovered over time. Experimental studies
from Table 1 confirm, that the reaction can take place in the anode
4
10
15
retention time / min
17
Scheme 2. RP-HPLC-UV chromatogram of the crude reaction mixture 7a–7e
with 1a–1f and 2a–2f at 248 nm with methanol as eluent.
6
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
compartment of a divided cell, but not in the cathode compartment
(entry 21/22). Hence, oxidation of the disulfide is plausible.
Subsequent cleavage of the S–S-bond of A resulting in the
Unfortunately, we were not able to pin-down the solvent effect
of CS2 in this reaction, since no useful results were be obtained
from cyclic votammograms containing carbon disulfide as a co-
solvent (see supporting information).
•
formation of a thiol-radical (RS ) and a thiol-cation, which is
stabilized by another disulfide molecule to form intermediate B is
plausible (Scheme 4).[17] Since the reaction takes place under
anodic conditions only, reductive activation of sulfur seems not to
be absolutely necessary. However, polysulfides were formed
preferably in an undivided cell in a much higher extent as in a
quasi-divided cell[13] with a platinum wire as anode (Table 1,
entry 20). When a solution of sulfur only was electrolyzed,
brownish streaks were observed at the cathode. According to
these observations, we propose that the activation of elemental
sulfur at the cathode takes place under formation of radical
anion C. Then, recombination of the thiol-radical and the sulfur-
radical in D may be possible in solution under formation of
intermediate E. In the next step, we propose the sulfur-sulfur bond
formation of E with B to one sulfur atom within the sulfur-chain in
F. This step determines the length of the latter polysulfide chain,
as all sulfur atoms between the organic substituents will be
included in the polysulfide. The elemental sulfur-side chain,
attached to the polysulfide-structure, donates an electron pair and
eliminates under reversible formation of the desired polysulfide.
We assume that this reaction runs in a dynamic equilibrium
that is reached after 4.0 F applied current. As shown in Figure 2,
polysulfides of all sulfur chain lengths were formed from beginning
of the reaction. After 4.0 F, the conversion of the disulfide reached
its maximum causing a stagnancy in the sulfur conversion and the
polysulfide formation. But, according to the previous study,[11] we
assume a permanent electrochemical S–S polysulfide-bond
metathesis. At the beginning of the reaction, splitting of a disulfide
is highly probable. But the larger polysulfide quantities are formed,
the higher the probability of a S–S-bond cleavage from a higher
sulfide. In theory, catalytic amounts of current are sufficient,
because no changes in oxidation states take place at the disulfide
and sulfur. However, in practice further electrolysis is necessary
due to the constant S–S-bond metathesis of already formed
polysulfides. Finally, after 4.0 F, the disulfide conversion reaches
its maximum so that further electrolysis only causes the S–S-bond
cleavage and recombination of polysulfides, whose dynamic
equilibrium is already reached.
Conclusion
In this article we were able to introduce the first
electrochemical synthesis of polysulfides (up to undecasulfides)
from disulfides or thiols with elemental sulfur. During the
optimization, the reaction proved to be very robust, as it can be
performed in different solvents with a variety of supporting
electrolytes and electrodes under a wide range of reaction
conditions. Also, the reaction is highly tolerant to the sulfur atom-
equivalents added, since excess sulfur simply remains in solution
to a certain degree. The reaction setup does not need any catalyst
and can be performed under air-atmosphere with low priced
materials which makes it highly economic.
We observed an interesting solvent effect when CS2 was
added as co-solvent, affecting the polysulfide distribution. To
describe the change of the polysulfide distribution, we introduced
the “absorbance average sulfur amount in polysulfides at 248 nm”
(SAP248) analogous to the “number average molar mass” used in
polymer chemistry (Equation 1). Additionally, we calculated the
dispersity Đ of the polysulfide mixture. It turned out, that the
addition of CS2 gave
a significantly narrower polysulfide
distribution. The preferred emerging species can be adjusted by
the volume fraction of added CS2 to some extent (Table 2,
Figure 3).
We therefore established two different reaction setups for the
investigation of different substrates, without any CS2 (setup A)
and with 10% (vol.) CS2 (setup B). Aliphatic disulfides responded
well to both setups. However, setup B increased the conversion
of the disulfides to >80%. Also, the yield of the preferably formed
sulfide rose up to >40% (di-n-butyl trisulfide (1b)). The conversion
of the aliphatic thiol 6 was nearly quantitative and gave 45% of
the trisulfide 6b with reaction setup B. Aromatic disulfides showed
in general a lower conversion compared to aliphatic disulfides (30-
60%). Unfortunately, their conversion and polysulfide distribution
could not be affected by the addition of CS2. In addition to the
synthesis of symmetrical polysulfides, we proved that the
synthesis of unsymmetrical polysulfides from two symmetrical
disulfides or a disulfide and a thiol was also possible.
Overall, this electrochemical reaction widens the field of
organic polysulfide synthesis and is a useful addition to the
expensive, but fast, rhodium-catalyzed synthesis reported by
Yamaguchi[9]. This work demonstrates a new way of sulfur
activation and lays the foundation for potential further work in this
research field. As a direct application, dynamic polysulfide
libraries with many products can be created from a single reaction
by the use of two or more symmetrical disulfides or thiols as
starting materials. We found an interesting effect of CS2 as co-
solvent on the polysulfide distribution that reveals new questions
for mechanistic investigations and might be useful for the lithium-
sulfur battery research.
Scheme 4. Proposed reaction mechanism.
7
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
Angew. Chem. Int. Ed. 2018, 57, 5594–5619; Angew. Chem. 2018, 130,
5694–5721; i) J.-i. Yoshida, A. Shimizu, R. Hayashi, Chem. Rev. 2018,
118, 4702–4730; j) M. Yan, Y. Kawamata, P. S. Baran, Chem. Rev. 2017,
117, 13230–13319; k) R. Francke, R. D. Little, Chem. Soc. Rev. 2014, 43,
2492–2521.
Experimental Section
General procedure for the electrochemical polysulfide
synthesis
[11] L. E. Sattler, C. J. Otten, G. Hilt, Chem. Eur. J. 2020, 26, 3129–3136.
[12] Herein, one atom-equivalent [S] is defined to be: 1 [S] = 1/8 S8.
[13] G. Hilt, ChemElectroChem 2020, 7, 395–405.
[14] a) A. Rudin, J. Chem. Educ. 1969, 46, 595–600; b) T. C. Ward, J. Chem.
Educ. 1981, 58, 867–879.
[15] A. Shrivastava in Introduction to Plastics Engineering (Ed.: A.
Shrivastava), Elsevier, Amsterdam, 2018, 17–48.
[16] R. F. T. Stepto, Polym. Int. 2010, 59, 23–24.
An undivided cell was charged with elemental sulfur
(0.0625 mmol, 2.0 eq. [S]), organic disulfide (0.25 mmol, 1.0 eq.)
or thiol (0.5 mmol, 2.0 eq.) and tetrabutylammonium tetrafluoro-
borate (1.0 mmol) in dichloromethane (setup A = 10 mL, setup B
= 9 mL + 1 mL CS2). After complete dissolution of the sulfur, the
electrolysis at glassy carbon electrodes (1.5 cm²) was performed
at 10 mA constant current until 4.0 F were passed through the
solution (160 min). The reaction mixture was filtered through
aluminium oxide (neutral), with dichloromethane as solvent. Then,
benzophenone was added as internal standard for the
determination of the polysulfide distribution in quantitative
13C NMR or HPLC-UV analysis. The solvent was removed under
reduced pressure in a fume hood (due to the possible exposition
of CS2) to obtain the pure polysulfide mixture.
[17] a) A. Bewick, D. E. Coe, J. M. Mellor, W. M. Owton, J. Chem. Soc., Perkin
Trans. 1 1985, 1033–1038; b) K. Yamamoto, E. Tsuchida, H. Nishide, S.
Yoshida, Y.-S. Park, J. Electrochem. Soc. 1992, 139, 2401–2406.
Acknowledgements
Preliminary investigations by BSc. Rebecca Meier are gratefully
acknowledged.
Keywords: sulfur • electrosynthesis • solvent effect • polysulfides
• trisulfide
[1]
a) M. Boelens, P. J. de Valois, H. J. Wobben, A. van der Gen, J. Agric.
Food Chem. 1971, 19, 984–991; b) M. H. Brodnitz, J. V. Pascale, L. van
Derslice, J. Agric. Food Chem. 1971, 19, 273–275; c) P.M.G. Curioni, J.
O. Bosset, Int. Dairy J. 2002, 12, 959–984; d) J. A. Maga, I. Katz, Crit.
Rev. Food Sci. Nutr. 1976, 7, 147–192; e) S. Sablé, G. Cottenceau, J.
Agric. Food Chem. 1999, 47, 4825–4836; f) A. M. Sourabié, H.-E.
Spinnler, P. Bonnarme, A. Saint-Eve, S. Landaud, J. Agric. Food Chem.
2008, 56, 4674–4680; g) Y. Ueda, M. Sakaguchi, K. Hirayama, R.
Miyajima, A. Kimizuka, Agric. Biol. Chem. 1990, 54, 163–169; h) C.
Vermeulen, L. Gijs, S. Collin, Food Rev. Int. 2005, 21, 69–137.
a) H. H. Bragulla, D. G. Homberger, J. Anat. 2009, 214, 516–559; b) J. T.
Brosnan, M. E. Brosnan, J. Nutr. 2006, 136, 1636-1640; c) G. Bulaj,
Biotechnol. Adv. 2005, 23, 87–92; d) P. Fariselli, P. Riccobelli, R. Casadio,
Proteins 1999, 36, 340–346; e) D. Fass, C. Thorpe, Chem. Rev. 2018,
118, 1169–1198; f) R. He, J. Pan, J. P. Mayer, F. Liu, ChemBioChem
2020, 21, 1101–1111; g) T. Kitahara, H. Ogawa, J. Dermatol. Sci. 1991,
2, 402–406.
[2]
[3]
[4]
[5]
a) S. D. Johnson, A. Jürgens, S. Afr. J. Bot. 2010, 76, 796–807; b) P.
Kakumyan, N. Suwannarach, J. Kumla, N. Saichana, S. Lumyong, K.
Matsui, Mycoscience 2020, 61, 65–70.
a) K. Morita, S. Kobayashi, Chem. Pharm. Bull. 1967, 15, 988–993; b) K.
Morita, S. Kobayashi, Tetrahedron Lett. 1966, 7, 573–577; c) M. Hiraide,
A. Kato, T. Nakashima, J. Wood Sci. 2010, 56, 477–482.
a) W. Kang, N. Deng, J. Ju, Q. Li, D. Wu, X. Ma, L. Li, M. Naebe, B.
Cheng, Nanoscale 2016, 8, 16541–16588; b) T. Li, X. Bai, U. Gulzar, Y.
J. Bai, C. Capiglia, W. Deng, X. Zhou, Z. Liu, Z. Feng, R. Proietti Zaccaria,
Adv. Funct. Mater. 2019, 29, 1901730; c) H.-J. Peng, J.-Q. Huang, X.-B.
Cheng, Q. Zhang, Adv. Energy Mater. 2017, 7, 1700260; d) M. Wild, L.
O'Neill, T. Zhang, R. Purkayastha, G. Minton, M. Marinescu, G. J. Offer,
Energy Environ. Sci. 2015, 8, 3477–3494.
[6]
S. Gu, J. Jin, S. Zhuo, R. Qian, Z. Wen, ChemElectroChem 2018, 5,
1717–1723.
[7]
[8]
D.-Y. Wang, W. Guo, Y. Fu, Acc. Chem. Res. 2019, 52, 2290–2300.
S. Xu, Y. Wang, M. N. Radford, A. J. Ferrell, M. Xian, Org. Lett. 2018, 20,
465–468.
[9]
M. Arisawa, K. Tanaka, M. Yamaguchi, Tetrahedron Lett. 2005, 46,
4797–4800.
[10] a) J. Liu, L. Lu, D. Wood, S. Lin, ACS Cent. Sci. 2020, 6, 1317–1340; b)
K.-J. Jiao, Y.-K. Xing, Q.-L. Yang, H. Qiu, T.-S. Mei, Acc. Chem. Res.
2020, 53, 300–310; c) P. Xiong, H.-C. Xu, Acc. Chem. Res. 2019, 52,
3339–3350; d) Y. Yuan, A. Lei, Acc. Chem. Res. 2019, 52, 3309–3324;
e) K. D. Moeller, Chem. Rev. 2018, 118, 4817–4833; f) Y. Jiang, K. Xu,
C.-C. Zeng, Chem. Rev. 2018, 118, 4485–4540; g) S. Möhle, M. Zirbes,
E. Rodrigo, T. Gieshoff, A. Wiebe, S. R. Waldvogel, Angew. Chem. Int.
Ed. 2018, 57, 6018–6041; Angew. Chem. 2018, 130, 6124–6149; h) A.
Wiebe, T. Gieshoff, S. Möhle, E. Rodrigo, M. Zirbes, S. R. Waldvogel,
8
This article is protected by copyright. All rights reserved.

10.1002/chem.202101023
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
The electrolysis of organic disulfides in dichloromethane (DCM) led to the formation of organic polysulfides with a high incorporation of
additional sulfur atoms. The HPLC / HRMS analysis of the reaction mixture could identify products up to undecasulfides. When CS2
was used as co-solvent the distribution of the organic polysulfide mixture was altered in favor of the lower polysulfides, such as tri- and
tetrasulfides.
9
This article is protected by copyright. All rights reserved.