Oligomerization of 1,2-ethanedithiol: An expedient approach to oligothiaethylenethioglycols

Article abstract of DOI:10.1002/chem.200903397

Reactions of ethylenedithioglycol (ETG) with Na₂CO₃, K₂CO₃, and Cs₂CO₃ provided the oligothiaethylenethioglycols (nETG): di(DETG), tri- (TrETG), tetra- (TETG), and pentathiaethylenethioglycol (PETG), along with higher polymers. The most efficient carbonate was K₂CO₃ and reactions using DETG and TrETG as starting materials-or their mixtures-were also found to afford similar species. This largely unknown oligomerization process was thoroughly explored and potential pathways were put forward. A convenient conversion of ETG to laboratory quantities of the otherwise scarce and/or expensive DETG, TrETG, TETG, and PETG oligomers, in organic or aqueous media was achieved. Notably, this straightforward reaction takes place without the addition of expensive or toxic metal catalysts and with pure water as the solvent, thereby highlighting its potential as a green chemical reaction. Moreover, the process opens up new approaches to dynamic combinatorial libraries (DCLs) of oligomers and macrocycles with manifolded nETG [(SCH₂CH₂)_nS] bridges.

Full text of DOI:10.1002/chem.200903397

FULL PAPER
DOI: 10.1002/chem.200903397
Oligomerization of 1,2-Ethanedithiol: An Expedient Approach to
Oligothiaethylenethioglycols
Dvora Berkovich-Berger,[a] N. Gabriel Lemcoff,*[a] Sarah Abramson,[b]
Mikhail Grabarnik,[b, c] Sarah Weinman,[b] and Benzion Fuchs*[b]
Abstract: Reactions of ethylenedithio-
glycol (ETG) with Na2CO3, K2CO3,
and Cs2CO3 provided the oligothiae-
oligomerization process was thoroughly
explored and potential pathways were
put forward. A convenient conversion
of ETG to laboratory quantities of the
otherwise scarce and/or expensive
DETG, TrETG, TETG, and PETG
oligomers, in organic or aqueous media
was achieved. Notably, this straightfor-
ward reaction takes place without the
addition of expensive or toxic metal
catalysts and with pure water as the
solvent, thereby highlighting its poten-
tial as a green chemical reaction. More-
over, the process opens up new ap-
proaches to dynamic combinatorial li-
braries (DCLs) of oligomers and mac-
rocycles with manifolded nETG
[(SCH2CH2)nS] bridges.
thylenethioglycols
(DETG), tri- (TrETG), tetra- (TETG),
and pentathiaethylenethioglycol
(nETG):
di-
(PETG), along with higher polymers.
The most efficient carbonate was
K2CO3 and reactions using DETG and
TrETG as starting materials—or their
mixtures—were also found to afford
similar species. This largely unknown
Keywords: ethylenedithioglycol
·
green chemistry · macrocycles · mi-
crowave chemistry · oligomeriza-
tion
Introduction
CH2-S)3H), tetra- (TETG=HS-(CH2-CH2-S)4H), and penta-
thiaethylenethioglycol (PETG=HS-(CH2-CH2-S)5H).
Efforts to secure the latter, seemingly simple, reagents
In studies aimed to secure novel macrocyclic host systems
with chiral core units, we used various known approaches of
grafting polyheterochain bridges on the termini of chiral po-
dands to produce 1+1, 2+2, 3+3, and larger macrocyclic
products, as shown for the diacetal type (cis-1,3,5,7-tetraoxa-
decalin (TOD)) core system (Scheme 1, top).[1] In this
framework we recently described[2] new classes of chiral
macrocycles consisting of (SCH2CH2)nS-bridged cis-1,3,5,7-
tetraoxadecalin chiral core molecules. These have been pre-
pared from the starting dibromide (1) with ethylenedithio-
glycol (ETG=HS-CH2-CH2-SH) or its (nETG) oligomers:
di- (DETG=HS-(CH2-CH2-S)2H), tri- (TrETG=HS-(CH2-
span
a century of diverse, usually laborious synthetic
work.[3,4] It should also be stressed that those available syn-
thetic literature procedures usually require dihalogenated
reagents, invariably giving rise to hazardous (e.g., vesicant,
mustard agents) intermediates or byproducts. As to cost and
commercial availability, when getting involved in processes
in which the higher nETG dithioglycols are needed, one re-
alizes that their cost increases exponentially as n is augment-
ed from 1 (ETGꢀ$30 per 100 g) to 2 (DETGꢀ$300 per
100 g) or 3 (TrETG: milligram quantities quoted only on
demand), whereas the higher ones are not commercially
available at all.
Crown thioethers are widely and well known[5] and recog-
nized as excellent ligands for heavy- and transition-metal
ions.[6] Most of the synthetic methods for such ligands are
based on the chain flanking reaction of 1,2-dihalides with
terminal dithiols[7] (e.g., ETG or its higher oligomers as sul-
fide- or a,w-dithiolate salts, over a century old technique),[3]
often using high-dilution techniques.[8] Although early in the
game such reactions had yielded crown thioethers in low
(up to 10%) yields, significant improvements have been
made by Kellogg and co-workers by using Cs2CO3 as a base
in DMF or sometimes K2CO3 in THF.[4] A large variety of
[a] D. Berkovich-Berger, Dr. N. G. Lemcoff
Department of Chemistry, Ben-Gurion University
Beer-Sheva, 84105 (Israel)
Fax : (+972)8-6472943
E-mail: lemcoff@bgu.ac.il
[b] Dr. S. Abramson, Dr. M. Grabarnik, S. Weinman, Prof. B. Fuchs
School of Chemistry, Tel-Aviv University
Ramat-Aviv, Tel-Aviv, 69978 (Israel)
Fax : (+972)3-6409293
E-mail: bfuchs@post.tau.ac.il
[c] Dr. M. Grabarnik
Present address: R&D, Makhteshim Ltd., Beer-Sheva (Israel)
Chem. Eur. J. 2010, 16, 6365 – 6373
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6365

before their interaction with
their dihalo partners, we exam-
ined the basic reaction of ETG
and Cs+, K+, and Na+ carbo-
nates in acetonitrile, with strik-
ing results. The primary process
(i.e., the reaction of ETG itself
with Cs2CO3, K2CO3, and
Na2CO3 in boiling MeCN at di-
verse conditions), provided the
higher
nETGs
H-
A
3
(TrETG),
4
(TETG),
5
(PETG); Figure 1). These solu-
ble and volatile products were
separated, analyzed by GLC,
and isolated accordingly, there-
by leaving traces of hexathia-
AHCTUNGTERGeNNUN thylenethioglycol (HETG) and
insoluble polymers in the resi-
due.
Scheme 1. Formation of ETG-bridged TOD macrocycles (M=Cs or K).
It may be recalled that in the
well-documented substitution
macrocyclic ligands have been synthesized using such tech-
niques, as up to recently copiously reviewed,[5–7] and in most
of those cases the dithiol starting materials were indeed
ETG or some of the higher oligothiaethylenethioglycols.
Thus, when rac-1 was reacted (Cs2CO3 or K2CO3 in
MeCN heated to reflux) with equimolar quantities of ETG[2]
(Scheme 1, bottom), the corresponding dithiacrown-TOD
product (1+1) was indeed obtained, but surprisingly, consis-
tent and rather intriguing occurrences of low yields of
higher-order polythiacrown byproducts (1+2, 1+3, and
traces of higher ones) were observed. Furthermore, process-
es starting with higher nETGs turned out to suffer from
analogous, inexplicable byproducts.[2] This frustrating, but
also challenging behavior was taken to indicate that a con-
comitant reaction was occurring and raised the possibility of
a potential dynamic combinatorial library (DCL) tool.
Herein, we report the study of this intriguing reaction of
ETG and carbonates in more detail.
reactions,[4] the Cs+ ion reigned
at the top, possibly due to the peerless character of the po-
larizable, poorly solvated, and tight-contact ion-pair-forming
Cs+ ion, as Kellogg and co-workers suggested[4] (see also
ref. [10]). A closer look, however, at the time dependence of
the conversion and formation of these dithiols (Figure 1)
made it clear that the picture is more complicated and inter-
esting, as is, for example, the relatively more attractive be-
havior of K2CO3 in the series. Here, the basicity of the alkali
carbonates (vide infra) is clearly accompanied by extra fea-
tures such as the significant difference between these bases
with regard to their solubility in MeCN, with Cs2CO3 being
more soluble in it than the other two,[10b] whereas the other
processes are markedly heterogeneous, thereby making the
exact end results dependent on crystal morphology, stirring
efficiency, and so on.
And yet (Figure 1) the Cs+2-induced process is “too
good,” being mainly plagued by enhanced formation of
polymers with resultant yield impairment. These manifestly
insoluble polymers were of less interest in this investiga-
tionꢁs framework, because inter alia, in addition to the fact
that they had been rather well looked into,[11] they provided
no clues to our problem.
We have attributed the formation of the higher oligothia-
crown byproducts (Scheme 1, bottom) from the reaction of
1 with ETG to a relatively slow thioetherification process,
which consists of a thiolate-initiated oligothiachain forma-
tion. Recognizing the significant “carbon basicity” of thioal-
koxide,[12a] and in some analogy with the Whitesides et al.[12]
mechanism for the thiol–disulfide interchange, it was com-
pelling to invoke the involvement of a primary thiolate
anion: HSCH2CH2SÀ to start the entire chain of events
(Scheme 2), which made available variable amounts of the
higher oligothia-a,w-dithiols products/reactants.
Results and Discussion
A search for existing related information indicated that, not-
withstanding the vast literature on sulfur-containing macro-
cycles, oligomers, and polymers, one could find rather scarce
or largely ignored observations on similar results. The most
relevant are those of Newkome et al.,[9] who reported three
instances of reactions of dihalo substrates with DETG, from
which they isolated TrETG products in addition to the ex-
pected ones, with no further comments. In these processes,
the bases/solvents ranged from NaH/xylene to tertiary
amines.
Conjecturing that the process is one in which the starting
dithiol reagents are affected by the reaction conditions
6366
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6365 – 6373

Oligothiaethylenethioglycols
FULL PAPER
ed to reactions with the choice base, K2CO3, in MeCN
(stirred under Ar) with salient results, as shown in Tables 1,
2, 3, 4, and 5, along with the respective interpretative
Schemes 3 and 4, in analogy with the mechanism proposed
in Scheme 2.
Table 1. Reaction of ETG with K2CO3 in acetonitrile.
Entry
t
ETG [%] DETG [%] TrETG [%] TETG [%]
1
2
3
4
5
6
7
30 min[a] 18
63
58
53
27
46
41
45
18
33
36
–
–
–
1
8
11
–
–
–
60 min[a]
120 min[a]
1
0
96 min[b] 73
168 min[b] 53
288 min[b] 59
3 weeks[b]
0
55% residue
Conversions include corresponding disulfides as minor byproducts.
[a] Stirred during heating at reflux under argon: ETG (1.1 mmol) and
K2CO3 (3 mmol) in MeCN (6 mL). [b] Stirred at RT under argon: ETG
(1.8 mmol) and K2CO3 (5 mmol) in MeCN (10 mL).
Table 2. Reaction of DETG with K2CO3 in acetonitrile.
Entry
t
DETG [%] TrETG [%] TETG [%] PETG [%]
1
2
3
4
5
6
7
8
30 min[a]
99
0.5
–
2.5
2
7
11
–
–
–
–
–
–
0.5
36
–
–
–
–
–
–
–
9
30 min[b] 100
60 min[a]
60 min[b]
97
98
93
89
78
16
120 min[a]
120 min[b]
240 min[b]
24 h[a]
22
39
Conversions include corresponding disulfides as minor byproducts.
Stirred during heating at reflux under argon. [a] DETG (6 mmol) and
K2CO3 (12 mmol) in MeCN (32 mL). [b] DETG (1.1 mmol) and K2CO3
(3.3 mmol) in MeCN (20 mL).
Table 3. Reaction of TrETG with K2CO3 in acetonitrile.
Entry
t [min]
60[a]
60[b]
DETG [%]
TrETG [%]
TETG [%]
Figure 1. Reactions of ETG/M2CO3 in MeCN heated at reflux for
a)Na2CO3, b) K2CO3, and c) Cs2CO3; the consumption/formation of ETG
1
2
3
4
5
6
7
8
6
5
7
6
9
8
10
13
93
93
91
90
87
90
86
82
<1
1.4
2
1.1
4
2
3
5
~
&
^
*
*
(
), DETG ( ), TrETG ( ), TETG ( ), PETG ( ), and polyETG (&)
120[a]
120[b]
180[a,c]
180[b]
240[b]
300[b,d]
are shown. The sum of mole fractions may fall below 100% due to either
volatile products (e.g., dithiane, thiirane) and/or insoluble higher poly-
thiaethylenethioglycols.
In a further, more detailed series of experiments, ETG as
well as the higher dithiols DETG and TrETG were subject-
Conversions include corresponding disulfides as minor byproducts.
Stirred during heating at reflux under argon. [a] TrETG (1 mmol) and
K2CO3 (2.5 mmol) in MeCN (5 mL). [b] TrETG (0.8 mmol) and K2CO3
(2 mmol) in MeCN (5 mL). [c] At the end of the reaction, 58% polymer
was recovered. [d] At the end of the reaction, 17% polymer was recov-
ered.
In all the above analytical trials, the formation of higher
nETG products was our main concern, but variable amounts
of byproducts that are not listed were detected in many
cases, namely, small amounts of the well-known 1,4-dithiane
(2), its less-known isomer 3,6-dithiahexene (3), and the
ETG2-, DETG-, DETG2-, TrETG-, and TrETG2-disulfides
Scheme 2. Proposed mechanism for the formation of nETG oligomers.
Chem. Eur. J. 2010, 16, 6365 – 6373
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6367

N. G. Lemcoff, B. Fuchs et al.
Table 4. Soluble products from the reaction of ETG/DETG mixtures
with K2CO3 in acetonitrile.[a]
Entry
t [min]
ETG [%]
DETG [%]
TrETG [%]
TETG [%]
1
2
3
4
5
6
7
8
9
0
10
20
30
45
60
90
120
180
19
15
9
5
2
1
–
–
–
81
79
80
80
78
76
75
73
71
–
5
–
–
–
–
<1
1
1.7
2.3
3
10
15
20
22
24
25
25
Scheme 3. Proposed mechanism for the disproportionation of DETG.
[a] Conversions include corresponding disulfides as minor byproducts.
Stirred during heating at reflux under argon: ETG (1 mmol), DETG
(2.5 mmol), and K2CO3 (5 mmol) in MeCN (5 mL).
Table 5. Soluble products from the reaction of DETG/TrETG mixtures
with K2CO3 in acetonitrile.[a]
Scheme 4. Proposed mechanism for the disproportionation of TrETG.
Entry
t [min]
ETG [%]
DETG [%]
TrETG [%]
TETG [%]
2
4
7
10
12
14
16
17
0
15
30
60
90
120
180
300
–
–
–
–
–
–
–
–
28
27
28
28
28
31
29
32
70
72
71
70
69
66
68
64
1
1
1
2
2
3
3
4
versatile formation and behavior,[12b] and more recent works
including DCL formation.[14] They can be easily reduced to
the parent dithiols by any of the known methods, as we did
using NaBH4.
Concerning 1,4-dithiane (2), it should be noted that
Ochrymowycz et al. had reported long ago[15] its isolation as
a byproduct along with 12S4 (1,4,7,10-tetrathiacyclodode-
cane) in the preparation of the 18S6 macrocycle from the re-
action of the PETG salt with dibromoethane (Scheme 5,
top) and had rationalized it by postulation of an initial chain
lengthening to an intermediate terminal monohalide, fol-
lowed by an intrachain thia displacement, accompanied by a
[a] Conversions include corresponding disulfides as minor byproducts.
Stirred during heating at reflux under argon: DETG (1.8 mmol), TrETG
(3.2 mmol), and K2CO3 (5 mmol) in MeCN (5 mL).
(4, 5, 6, 7, 8, and 9, respectively). Although the detailed
mechanisms of the above-described processes are not yet
entirely solved, we observed (MS) signs of the -SCH2CH2-
moiety, perhaps as thiirane (consistent with its known role
in the transition-metal-[5j,13a–c] or acid[13d]-catalyzed buildup
of polythiaethylene macrocycles), and of 1,2-dithietane, both
easily rationalized in the above transformations as extrusion
products along with HSÀ.
À
thiolate-assisted C S bond cleavage and 1,4-dithiane (2) for-
mation. Our system differs by having no halogen–carbon-ac-
tivated bond. However, in the course of the oligomerization
process, PETG, for example, may follow a chain-shortening
side reaction by displacement of an adjacent sulfide entity,
thereby yielding TrETG+dithiane (Scheme 5, bottom). This
The disulfide species in the reaction products were actual-
ly anticipated, especially following earlier studies of their
Scheme 5. Mechanistic explanation for the production of dithiane.
6368
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6365 – 6373

Oligothiaethylenethioglycols
FULL PAPER
Figure 2. Reactions of ETG/M2CO3 in water a) heated at reflux (top: K2CO3, bottom: Cs2CO3) and b) MW (1058C) (top: Na2CO3, middle: K2CO3,
~
&
^
*
*
bottom: Cs2CO3). The consumption or formation of ETG ( ), DETG ( ), TrETG ( ), TETG ( ), and PETG ( ) is presented. Initial ETG concentra-
tion 0.4 molLÀ1. Shown products analyzed by GC–MS.
is also how we interpreted the peculiar absence of
(SCH2CH2)n macrocycles in all these processes.
steady and much less harmed. The straight DETG/K2CO3
reaction (Table 2, Scheme 3) is indeed much slower, both in
its decay and in the higher products formation (it is proba-
ble that ETG is initially created, albeit unseen because it is
used up faster than its formation). Similarly, the TrETG/
K2CO3 reaction (Table 3, Scheme 4) is no less sluggish in its
disproportionation process to DETG and TETG and the
two intermixed runs, ETG+DETG and DETG+TrETG/
K2CO3, (Tables 4 and 5) nicely corroborate all of the above.
At this point, juxtaposition and interpretation of the data
in all the tables above is called for, in conjunction with the
accompanying schemes. It is clear that in the canonic ETG/
K2CO3 reactions (Table 1 and Figure 1, middle, and
Scheme 2) the decay of ETG is the most rapid in the series.
It is only when it is depleted that the produced DETG starts
decaying itself, thus resupplying some ETG for its part.
TrETG and TETG show much slower buildups, but they are
Chem. Eur. J. 2010, 16, 6365 – 6373
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6369

N. G. Lemcoff, B. Fuchs et al.
In the course of checking various variables in our above
investigation, small amounts of water were added to the re-
action medium (MeCN) and found not to harm or alter ap-
preciably the outcome of the experiments. Due to the obvi-
ous benefits, we decided to investigate the effects of running
the reactions in pure water as the solvent. Fortunately, we
found that the entire process occurs equally well in water, in
homogeneous carbonate solutions heated at reflux (Fig-
ure 2a) and, even better, with microwave (MW) accelera-
tion[16] of the reaction (Figure 2b).
To be sure, the picture obtained in water (Figure 2) is to
some extent different, but instructive. First, the fact that all
three carbonates behave with only slight variations (viz.,
Cs2CO3 not rushing off and Na2CO3 not falling behind ex-
cessively) is in line with the above-suggested thiolate inter-
mediacy, the pKa of dithiols, for example, ETG (pKa1 =8.85
and pKa2 =10.43) matching closely those of the carbonates
(pKa1 =6.35 and pKa2 =10.33).[17]
In this context, it should be noted that this combination
bestows buffer conditions to the aqueous medium, but the
behavior under the two conditions is influenced also by the
fact that, contrary to the MW experiment, when heated at
reflux the CO2 formed is continuously carried off, which
may clarify the sometimes random ratios of nETG products
(cf. the Experimental Section). Furthermore, attempts to
use alkali hydroxide bases in these reactions were largely
unsuccessful, which conforms to this line of reasoning and
strengthens the above-suggested pathway, since a double thi-
olate anion should be inert towards its neighbors. Moreover,
practically no polymers were isolated in these conditions.
This peculiar reactivity provided the most valuable results
of our study in preparative runs by securing from the simple
and readily available ETG its higher, rare, and pricey nETG
oligomers in straightforward reactions, as shown in
Scheme 6.
tion) with largely comparable results, thereby making readi-
ly available substantial quantities of these oligo-ETG start-
ing materials for the synthesis of a variety of thioether mac-
rocycles.[2] Moreover, one can direct the proportionality of
the reaction products by judicious choice of carbonate, tem-
perature, or reaction time.
The full significance of the above approach and its results
are bound to be greatly appreciated by those actively in-
volved in processes in which the nETG dithioglycols are
needed. It is then realized how these costly or altogether un-
procurable reagents become readily accessible, not to men-
tion the avoidance of health hazards in the available experi-
mental procedures for their preparation (imputable to the
dangerous—vesicant mustard agents—intermediates or by-
products). Admittedly, the ostensible simplicity and attrac-
tiveness of the above-described approach to hitherto expen-
sive or laboriously made nETG oligomers is not tallied by
the clarity of its mechanistic pathway. The exact mechanism
of this variegated reaction along with the reactivity of the
homologous dithiols is expected to lead to highly interesting
consequences. These, along with ion-templating effects, the
addition of other thiol-containing molecules, and the poten-
tial of applying this new approach to DCLs containing
(SCH2CH2)nS units will be reported in due course. It is also
anticipated that the ready availability of such high nETG
oligomers will spur the conception of further interesting poly-
thiamacrocyclic systems.
Conclusion
The vexing byproducts in the preparation of polythiomacro-
cycles were found to be due to alkali-carbonate-catalyzed
thioetherifying oligomerization reactions of ethylenedithio-
glycol (ETG) to higher polythiaethylenethioglycols. In tar-
geted reactions, M2CO3-induced oligomerization of ETG
was established, the most practical being K2CO3; similar cat-
alyzed chain increases of DETG and TrETG—or their mix-
tures—were also found to occur. Thus, ETG could be used
in facile and inexpensive preparations of laboratory quanti-
ties of the otherwise scarce and/or expensive DETG,
TrETG, TETG, and PETG oligomers, thus also avoiding
hazardous intermediates in the process. The use of only
water as a solvent coincides with the safer procedures and
opens the doorway to clean ꢂgreenꢁ chemistry protocols for
the synthesis of the thiaoligomers and their subproducts.
Such procedures were performed repeatedly in up to 50 g
runs in acetonitrile or in water (see the Experimental Sec-
Experimental Section
General: The NMR spectra were obtained using Bruker AMX 360, ARX
500, or DPX 500 spectrometers. All 1H chemical shifts are in ppm and d
values are relative to TMS as internal standard. 13C chemical shifts are
reported in ppm relative to CDCl3 (center of triplet d=77.0 ppm), except
for those taken in D2O, which are reported in ppm relative to TMS salt
as internal standard. Mass spectra (desorption electron ionization (DEI),
desorption chemical ionization (DCI), and fast atom bombardment
(FAB)) were recorded using a VG Autospec 250 mass spectrometer. UV
Scheme 6. Preparative reactions in water and acetonitrile heated to
reflux. Yields include respective disulfides.
6370
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6365 – 6373

Oligothiaethylenethioglycols
FULL PAPER
spectra were taken using a Uvikon 931 spectrophotometer. Chromato-
graphic analyses of experiments in MeCN were run using a Varian 3400
instrument with Megabor column 15 mꢃ0.53 mm, phase DB5, He gas
flow 30 mLminÀ1, column temperature from 50 to 3008C with initial hold
time 1.5 min, gradient 208minÀ1, and hold time 5 min. In experiments in
water the GC–MS columns were alternately: 1) Agilent HP5-MS, column
temperature from 80 to 3008C; 2) Restek Rxi-5ms column, column tem-
perature from 80 to 3008C. All nETG yields include those of their re-
spective disulfide byproducts (which could easily be reduced to their di-
thiol precursors; see below).
dried, and showed a composition (GLC) of 1% ETG, 70% DETG+di-
sulfides, 26% TrETG+disulfide, and 3% TETG. After removal of the
solvent and low fractions at atmospheric pressure, a yellowish oil (8.7 g)
was obtained. This was thoroughly triturated with diethyl ether (3ꢃ
30 mL) and filtered to provide TETG (1.3 g, 15%) as a white solid. The
solvent was evaporated and the residue was distilled under high vacuum,
at 0.02 millibar: the first fraction (608C) consisted of DETG (3.3 g,
29%), whereas the second fraction (688C) was a mixture (0.34 g) and the
residual white semisolid consisted of TrETG (3.0 g, 25%) including ap-
proximately 5% of its cyclic disulfide. The first, low fractions consisted of
small amounts of ETG and the following byproducts: 0.01% dithiane (9)
and approximately 1% of 3,6-dithiahexene. The solid residue contained
variable amounts of PETG. Such procedures were performed repeatedly
in up to fivefold-scaled-up runs with similar results.
Analytical runs
In acetonitrile (Figure 1, Table 1): M2CO3 (2.8 mmol) was suspended in a
solution of ETG (0.096 g, 1 mmol) in acetonitrile (5 mL) and the mixture
was heated at reflux under Ar. At given intervals, chloroform (5 mL) and
1n HCl (10 mL) were added to each individual reaction mixture, and the
organic phase underwent GLC analysis (Megabore column 15 mꢃ
0.53 mm, DB5, He gas flow 30 mLminÀ1, column temperature from 50 to
3008C). Retention times: ETG 1.5 min, DETG 5.7 min, TrETG 8.7 min,
TETG 11.2 min, and PETG 13.5 min. Mass balance calculations were
based on calibration curves. Starting relationships between peak areas S
and reagent concentrations C in acetonitrile were in the range of 9ꢃ10À3
to 0.2 molLÀ1. For ETG it was C=(5.53Æ0.09)ꢃ10À7 ꢃS, r=0.999, n=
12; for DETG it was C=(3.47Æ0.04)ꢃ10À7 ꢃS, r=0.999, n=14; for
TrETG it was C=(3.05Æ0.14)ꢃ10À7 ꢃS, r=0.994, n=8.
In acetonitrile, process 2: After 4 h in the initial product mixture, a scale-
up experiment starting with ETG (47 g) provided (GC–MS): 3.5% ETG,
51% DETG, 35% TrETG, 9% TETG, and 0.3% PETG (including the
respective disulfides); and after isolation as described above: 34%
DETG, 27% TrETG, and 12% TETG (and disulfides). Isolation of the
residue at higher temperatures provided variable small amounts of
PETG and HETG. Part of the 1,4-dithiane and 3,6-dithiahexene formed
in the reaction, as well as ETG, are removed in the first, aqueous washing
step of the workup (from which they can be recovered by chloroform ex-
traction). The rest are obtained in the preliminary distillation step at
1348C/21 millibar.
Product concentrations were based on relationships between reagent con-
centrations Cr in MeCN, in the range 3ꢃ10À3 to 0.2 molLÀ1, and peak
areas Sr from the GLC analysis of the organic phases after workup as
above. For ETG it was Cr =(4.41Æ0.16)ꢃ10À7 ꢃSr, r=0.992, n=14; for
DETG it was Cr =(2.73Æ0.04)ꢃ10À7 ꢃSr, r=0.999, n=11; for TrETG it
was Cr =(2.59Æ0.09)ꢃ10À7 ꢃSr, r=0.999, n=11. For TETG and PETG,
Cr/Sr was estimated to be 2.5ꢃ10À7. All other analytical runs (Tables 2, 3,
4, and 5, and respective schemes; vide supra) were run similarly and re-
peatedly, sometimes using also NMR spectroscopic techniques for analy-
sis, with proper calibration curves. The formation of H2S was proven on
N2 bubbling of the initial crude reaction product mixture after acidifica-
tion.
In water:
A mixture of fresh ETG (9.67 g, 0.103 mol) and K2CO3
(28.44 g, 0.206 mol) in deionized water (86 mL) was swept with Ar for
20 min, after which it underwent microwave radiation for 45 min at
1058C (Discover equipped with Explorer robot, CEM corporation,
200 W, maximum pressure of 249 psi), by a sequence of 6 microwave
tubes, each containing the reaction mixture (14.3 mL). After cooling the
mixture, combining the tubes, and the addition of chloroform (100 mL)
and HCl (2.5m, 200 mL) with stirring, the organic layer was separated.
Additional chloroform (50 mL) was added and again separated, and the
combined organic phases were washed, dried, and showed a composition
(GC–MS) of 1.2% ETG, 18.4% DETG+disulfide, 40.6% TrETG+di-
sulfide, 32.7% TETG+disulfide, and 6.6% PETG disulfide. After re-
moval of the solvent and low fractions (mainly ETG) at atmospheric
pressure by distillation at 618C under nitrogen, a white oil (7 g, 94%
yield) was obtained. This was thoroughly triturated with diethyl ether
(3ꢃ30 mL), to provide a mixture (3.5 g) that contained 15% DETG,
47.3% TrETG, 35% TETG, and 2.6% PETG disulfide as a white solid.
The filtrate was evaporated and the residue (46.5% DETG, 53.5%
TrETG) was distilled under high vacuum, at 0.05 mmHg: one fraction
was distilled at 668C and consisted of DETG with almost no disulfide
(0.3 g,) as a clear oil, and the residual white semisolid consisted of
TrETG with approximately 5% of its disulfide (0.75 g).
In water (Figure 2): A mixture consisting of a suspension of M2CO3
(6 mmol) in a solution of ETG (168 mL, 2 mmol) in deionized water
(5 mL) was purged with N2 for 10 min. The mixture then underwent
either microwave radiation at 1058C (Discover equipped with Explorer
robot, CEM corporation, 200 W, maximum pressure of 249 psi) or heating
to reflux under N2. At given intervals, HCl (600 mL, 32%) was added to
each individual reaction mixture and the neutralized solution was extract-
ed 3 times with CH2Cl2 (2.5 mL). The organic phase underwent GC–MS
analysis (Agilent 6850 GC equipped with an Agilent 5973 MSD and an
Agilent HP5-MS column, heated from 80 to 3008C). Retention times:
ETG 2.5 min, DETG 6 min, TrETG 8.2 min, TETG 10 min, and PETG
12.6 min. Mass balance calculations were based on diglyme, which was
used as an internal standard. Thus, for the MW reaction with 3 equiv
K2CO3 (pH 10.61), a statistical analysis of 14 samples gave mean yields
[%] of (standard errors are given in parentheses): DETG 20.57 (1.34),
TrETG 53.5 (1.29), TETG 26.0 (1.44), PETG<1. As the K2CO3 in-
creased, the yields become rather erratic, and for 15 equiv of K2CO3
(pH 12.56) the results become (3 samples; means) ETG 6, DETG 90,
TrETG 4. Similar runs at reflux were conducted at the same concentra-
tions but different volumes, for example, in Figure 2a, a 25 mL sample of
water was taken.
Reduction of oxidized species: An equimolar amount of sodium borohy-
dride was added to a mixture of DETG, TrETG, and TETG with their
oxidized species in water, and the mixture was stirred at 808C for 1 h.
After acidification (HCl) and extraction (CH2Cl2), only the dithiols and
no oxidized species were observed by GC–MS.
In situ acidification using a Dowex-50W-X2 instrument and CH3CN (no
need for extraction) gave similarly good results.
Tables 6 and 7 provide the NMR spectroscopic and MS data of the prod-
ucts. ETG, DETG, and 1,4-dithiane (2) are commercial and well docu-
mented, but were included for the sake of comparison. All the others are
either unknown or scarcely reported.
Preparative runs: The yields shown in Scheme 6 are approximate because
they are quite dependent on the reaction and isolation conditions,
namely, time and temperature (see below).
In acetonitrile, process 1: A mixture of fresh ETG (11.2 g, 0.12 mol) and
K2CO3 (34.3 g, 0.25 mol) in CH3CN (100 mL) was mechanically stirred
and swept with Ar for 20 min, after which it was heated at reflux under
Ar for 4 h. GLC aliquot sampling showed the following after 1 h: 55%
ETG, 44% DETG+disulfides, 1% TrETG; and after 2 h: 20% ETG,
66% DETG+disulfides, 12% TrETG, and 1% TETG. After cooling the
mixture and the addition of chloroform (100 mL) and HCl (2.5m,
200 mL) with stirring, the organic layer was separated, washed, and
Acknowledgements
We gratefully acknowledge support by research grants from the Israel
Science Foundation, the Ministry of Science, and the Edmond J. Safra
Foundation, as well as the valuable assistance of Mr. Shimon Hauptman
with mass spectrometry.
Chem. Eur. J. 2010, 16, 6365 – 6373
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6371

N. G. Lemcoff, B. Fuchs et al.
Table 6. 1H and 13C NMR spectral data of nETG oligomers and disulfides (ds).[a]
1933, 55, 775–781; e) J. R.
Meadow, E. E. Reid, J. Am.
Chem. Soc. 1934, 56, 2177–2180;
f) W. Rosen, D. H. Busch, J. Am.
Chem. Soc. 1969, 91, 4694–4697;
g) W. Rosen, D. H. Busch, Inorg.
Chem. 1970, 9, 262–265; h) E. N.
Deryagina, N. V. Russavskaya,
O. V. Alekminskaya, N. A. Korch-
evin, B. A. Trofimov, Sulfur Lett.
1999, 22, 85–88; i) E. Sekido, A.
Nakabayashi, Anal. Chim. Acta
1989, 221, 99.
nETG
2,2’ d [ppm]
3,3’ d [ppm]
5,5’ d [ppm]
6,6’ d [ppm]
SH d [ppm]
ETG
1H
13C
1H
13C
1H
13C
1H
13C
1H
13C
1H
1H
1H
13C
1H
13C
1H
13C
1H
13C
2.76 (m, 4H)
27.7
2.77 (brm, 8H)
24.7
2.77 (brm, 12H)
24.6
2.75 (brm, 16H)
24.6
2.9 (m, 8H)
29.1
2.62 (m)
3.17 (brm, 8H)
3.20 (brm, 8H)
41.1
3.20 (brm, 8H)
34.7
–
–
–
36.0
–
36.1
–
36.2
–
–
–
–
–
32.1
–
32.2
–
–
–
–
–
–
–
–
1.76 (t)
–
1.74 (t)
–
1.74 (t)
–
1.73 (t)
–
DETG
TrETG[b]
TETG[c]
2
–
–
–
–
–
–
–
–
3
4
5
2.90 (m)
–
–
30.7
–
250.7
–
32.3
–
32.4
6.33 (m)
–
–
–
–
–
–
31.6
–
31.8
4.90, 5.25
1.70 (t)
[4] a) J. Buter, R. M. Kellogg, J. Org.
Chem. 1981, 46, 4481–4485;
b) J. J. H. Edema, H. T. Stock, J.
Buter, R. M. Kellogg, W. J. J.
Smeets, A. L. Spek, F. Van Bol-
huis, Angew. Chem. Int. Ed. Engl.
1993, 32, 436–439; c) J. J. H.
Edema, J. Buter, R. M. Kellogg,
Tetrahedron 1991, 50, 2095–2098;
d) H. T. Stock, R. M. Kellogg, J.
Org. Chem. 1996, 61, 3093–3105.
[5] a) J. Breitenbach, J. Boosfeld, F.
Vçgtle in Comprehensive Supra-
molecular Chemistry, Vol. 2 (Ed.:
F. Vçgtle), Pergamon, Oxford,
1996, p. 29; b) J. S. Bradshaw,
R. M. Izatt, A. V. Bordunov, C. Y.
Zhu, J. K. Hathaway in Compre-
hensive Supramolecular Chemis-
try, Vol. 1 (Ed.: G. W. Gokel),
Pergamon, Oxford, 1996, pp. 35–
95; c) E. Weber, J. L. Toner, I.
Goldberg, F. Vçgtle, D. A. Lai-
dler, J. F. Stoddart, R. A. Bartsch,
C. L. Liotta, in Crown Ethers and
Analogs (Eds.: S. Patai, Z. Rap-
poport), Wiley, New York, 1989,
p. 20; d) J. S. Bradshaw, J. Y. K.
Hui, J. Heterocycl. Chem. 1974,
11, 649–673; e) S. G. Murray,
F. R. Hartley, Chem. Rev. 1981,
81, 365–414; f) V. V. Litvinova,
A. V. Anisimov, Chem. Hetero-
cycl. Compd. 1999, 35, 1385–
1414; g) G. Reid, M. Schrçder,
Chem. Soc. Rev. 1990, 19, 239–
269; h) G. Brieger, T. J. Nestrick,
Chem. Rev. 1974, 74, 567–580;
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
6
7
8
2.85 (brm, 12H)
39.3
2.8, 2.9 (brm, 12H)
38.7
[a] CDCl3, d [ppm]; the primed atoms are related by C2 symmetry; t=triplet, m=multiplet. [b] Also cf.
ref. [18].
Table 7. MS data of nETG oligomers, their cyclic disulfides (nETGds), and other products.
Molecule
Peaks (relative abundance [%])[a]
Mr
Formula
ETG
94 (82), 61 (50), 60 (77), 47 (100), 45 (50)
94.20
154.32
214.44
C2H6S2
C4H10S3
C6H14S4
DETG
TrETG
154 (18), 152 (45), 120 (16), 94 (32), 61
(100), 60 (25)
214 (100), 180 (20), 154 (50), 152 (47), 120 (15), 61 (76);
CI: 215 (55), 181 (10), 153 (23), 121 (100)
274 (4), 154 (30), 120 (46), 61 (100)
TETG
274.55
334.67
394.79
120.24
120.24
184.37
152.30
304.60
212.42
424.84
272.54
332.66
392.77
C8H18S5
C10H22S6
C12H26S7
C4H8S2
C4H8S2
C4H8S4
PETG
334 (5), 214 (30), 120 (40), 61 (100)
HETG
394 (8)
2
120 (100), 92 (22), 61 (70), 60 (20), 47 (80)
120 (80); CI: 121 (90)
184 (100), 124 (55), 92 (25), 87 (35), 60 (40)
152 (100), 124 (26), 106 (18), 87 (33), 78 (20), 60 (42)
304 (12)
3[b]
4[c]
5[c]
C4H8S3
6
C8H16S6
C6H12S4
C12H24S8
C8H16S5
C10H20S6
C12H24S7
7
8
9
212 (56), 184 (15), 152 (100), 124 (98), 87 (38), 60 (58)
CI: 425 (10), 269 (20), 213 (90), 153 (95), 121 (100)
272 (12); CI: 273 (12), 215 (20), 153 (36), 74 (100)
332 (12), 272 (15), 212 (20), 184 (18), 152 (28), 124 (100)
392 (28), 332 (85), 272 (65), 244 (26), 212 (100), 184 (86)
PETGds
HETGds
[a] Main fragments are listed; isotopic distribution was ascertained and is not exhibited; the higher systems
often carry lower homologue impurities. [b] Also cf. ref. [19] [c] Also cf. ref. [20].
i) R. A. W.
Johnstone,
A. H.
Wilby, I. D. Entwistle, Chem. Rev.
1985, 85, 129–170; j) R. D. Adams, Aldrichimica Acta 2000, 33, 39.
[6] a) R. M. Kellogg in Crown Compounds: Toward Future Applica-
tions, VCH, Weinheim, 1992, pp. 261–284; b) S. R. Cooper in Crown
Compounds: Toward Future Applications, VCH, Weinheim, 1992,
Chapter 15; c) S. R. Cooper, S. C. Rawle, Acc. Chem. Res. 1988, 21,
141–146; d) S. R. Cooper, S. C. Rawle, Struct. Bonding (Berlin)
1990, 72, 1–72.
[7] a) C. M. Rayner, Contemp. Org. Synth. 1995, 2, 409–440 and C. M.
Rayner, Contemp. Org. Synth. 1994, 1, 191–203; b) I. V. Koval, Rus-
sian Chem. Rev. 1993, 62, 769–786; c) F. Lamar, Sulfur Rep. 1993,
13, 197–277.
[1] a) H. Jatzke, K. Frische, M. Greenwald, L. Golender, B. Fuchs, Tet-
rahedron 1997, 53, 4821–4834; b) M. Grabarnik, N. G. Lemcoff, R.
Madar, S. Abramson, S. Weinman, B. Fuchs, J. Org. Chem. 2000, 65,
1636–1642; c) B. Fuchs, A. Nelson, A. Star, J. F. Stoddart, S. Vidal,
Angew. Chem. 2003, 115, 4352–4356; Angew. Chem. Int. Ed. 2003,
42, 4220–4224; d) S. Abramson, E. Ashkenazi, K. Frische, I. Gold-
berg, L. Golender, M. Greenwald, N. G. Lemcoff, R. Madar, S.
Weinman, B. Fuchs, Chem. Eur. J. 2003, 9, 6071–6082.
[2] a) S. Abramson, D. Berkovich-Berger, S. Dagan, I. Goldberg, L. Go-
lender, M. Grabarnik, N. G. Lemcoff, S. Weinman, B. Fuchs, Eur. J.
Org. Chem. 2007, 1957–1975; b) M. Grabarnik, I. Goldberg, B.
Fuchs, J. Chem. Soc. 1997, 3123–3126.
[8] P. Knops, N. Sendhoff, H. B. Mekelburger, F. Vçgtle, Top. Curr.
Chem. 1992, 161, 1–36.
[3] a) V. Meyer, Chem. Ber. 1886, 19, 3259–3266; b) H. Fasbender,
Chem. Ber. 1887, 20, 460–465; c) S. P. C. Rꢄy, J. Chem. Soc. 1920,
117, 1090–1092; d) N. B. Tucker, E. E. Reid, J. Am. Chem. Soc.
[9] a) G. R. Newkome, A. Nayak, J. D. Sauer, P. K. Mattschei, S. F. Wat-
kins, F. Fronczek, W. H. Benton, J. Org. Chem. 1979, 44, 3816–3826;
b) G. R. Newkome, F. Danesh-Khoshboo, A. Nayak, W. H. Benton,
6372
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6365 – 6373

Oligothiaethylenethioglycols
FULL PAPER
J. Org. Chem. 1978, 43, 2685–2690; c) G. R. Newkome, A. Nayak, J.
Org. Chem. 1978, 43, 409–415.
[10] a) C. Galli, L. Mandolini, J. Org. Chem. 1991, 56, 3045–3047; b) A.
Ostrowicki, E. Koepp, F. Vçgtle, Top. Curr. Chem. 1992, 161–163,
37–67.
Chem. 2007, 119, 9014–9017; Angew. Chem. Int. Ed. Engl. 2007, 46,
8858–8861; e) R. Pꢈrez-Fernꢅndez, M. Pittelkow, A. M. Belenguer,
J. K. M. Sanders, Chem. Commun. 2008, 2008, 1738–1740.
[15] L. A. Ochrymowycz, C. P. Mak, J. D. Michna, J. Org. Chem. 1974,
39, 2079–2084.
[11] J. Masamoto, in Polymer Data Handbook (Ed.: Mark, J. E.), Oxford
University Press, New York, 1999, p. 553.
[16] D. Dallinger, C. O. Kappe, Chem. Rev. 2007, 107, 2563–2591.
[17] A. E. Martell, R. M. Smith in Critical Stability Constants, Plenum
Press, New York, 1974.
[18] M. Nikac, M. A. Brimble, R. L. Crumbie, Tetrahedron 2007, 63,
5220–5226.
[12] a) M. R. Crampton, J. Chem. Soc. B 1968, 1208–1213; b) R. Singh,
G. M. Whitesides in The Chemistry of Sulfur-Containing Functional
Groups (Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1993.
[13] a) R. D. Adams, S. B. Falloon, J. Am. Chem. Soc. 1994, 116, 10540–
10547; b) R. D. Adams, J. H. Yamamoto, A. Holmes, B. J. Baker, Or-
ganometallics 1997, 16, 1430–1439; c) R. D. Adams, K. M. Brosius,
O. -Sung. Kwon, J. Organomet. Chem. 2002, 652, 51–59; d) A. Fꢅsi,
ꢆ. Gçmçry, I. Pꢅlinkꢇ, I. Kiricsi, Catal. Lett. 2001, 76, 95–98.
[14] a) S. Otto, R. L. E. Furlan, J. K. M. Sanders, J. Am. Chem. Soc. 2000,
122, 12063–12064; b) S. Otto, S. Kubik, J. Am. Chem. Soc. 2003,
125, 7804–7805; c) K. R. West, K. D. Bake, S. Otto, Org. Lett. 2005,
7, 2615–2618; d) P. T. Corbett, J. K. M. Sanders, S. Otto, Angew.
[19] G. Sehrader, U.S. Patent 3065256, 1962.
[20] a) M. H. Goodrow, W. K. Musker, Synthesis 1981, 457–459;
b) M. H. Goodrow, M. M. Olmstead, W. K. Musker, Tetrahedron
Lett. 1982, 23, 3231–3234; c) W. Amaratunga, J. Milne, A. Santagati,
J. Polym. Sci. Polym. Chem. Ed. 1998, 36, 379–390.
Received: December 10, 2009
Published online: April 21, 2010
Chem. Eur. J. 2010, 16, 6365 – 6373
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6373