Unexpectedly Stable (Chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane, and Related Compounds That Model the Zumach-Weiss-Kühle (ZWK) Reaction for Synthesis of 1,2,4-Dithiazolidine-3,5-diones

Article abstract of DOI:10.1021/acs.joc.5b01826

The Zumach-Weiss-Kühle (ZWK) reaction provides 1,2,4-dithiazolidine-3,5-diones [dithiasuccinoyl (Dts)-amines] by the rapid reaction of O-ethyl thiocarbamates plus (chlorocarbonyl)sulfenyl chloride, with ethyl chloride and hydrogen chloride being formed as coproducts, and carbamoyl chlorides or isocyanates generated as yield-diminishing byproducts. However, when the ZWK reaction is applied with (N-ethoxythiocarbonyl)urethane as the starting material, heterocyclization to the putative "Dts-urethane"? does not occur. Instead, the reaction directly provides (chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane, a reasonably stable crystalline compound; modified conditions stop at the (chlorocarbonyl)[1-ethoxy-(N-ethoxycarbonyl)formimidoyl]disulfane intermediate. The title (chlorocarbonyl)(carbamoyl)disulfane cannot be converted to the elusive Dts derivative, but rather gives (N-ethoxycarbonyl)carbamoyl chloride upon thermolysis, or (N-ethoxycarbonyl)isocyanate upon treatment with tertiary amines. Additional transformations of these compounds have been discovered, providing entries to both known and novel species. X-ray crystallographic structures are reported for the title (chlorocarbonyl)(carbamoyl)disulfane; for (methoxycarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane, which is the corresponding adduct after quenching in methanol; for [1-ethoxy-(N-ethoxycarbonyl)formimidoyl](N′-methyl-N′-phenylcarbamoyl)disulfane, which is obtained by trapping the title intermediate with N-methylaniline; and for (N-ethoxycarbonylcarbamoyl)(N′-methyl-N′-phenylcarbamoyl)disulfane, which is a short-lived intermediate in the reaction of the title (chlorocarbonyl)(carbamoyl)disulfane with excess N-methylaniline. The new chemistry and structural information reported herein is expected to contribute to accurate modeling of the ZWK reaction trajectory.

Full text of DOI:10.1021/acs.joc.5b01826

Article
pubs.acs.org/joc
Unexpectedly Stable (Chlorocarbonyl)(N-ethoxycarbonyl-
carbamoyl)disulfane, and Related Compounds That Model
̈
the Zumach−Weiss−Kuhle (ZWK) Reaction for Synthesis
of 1,2,4-Dithiazolidine-3,5-diones
George Barany,* Doyle Britton,† Lin Chen,‡ Robert P. Hammer,§ Matthew J. Henley,∥ Alex M. Schrader,⊥
and Victor G. Young, Jr.
Department of Chemistry, University of Minnesota, Minneapolis, Minnesota 55455, United States
S
* Supporting Information
ABSTRACT: The Zumach−Weiss−Kuhle (ZWK) reaction provides 1,2,4-dithiazolidine-3,5-diones [dithiasuccinoyl (Dts)-
̈
amines] by the rapid reaction of O-ethyl thiocarbamates plus (chlorocarbonyl)sulfenyl chloride, with ethyl chloride and hydrogen
chloride being formed as coproducts, and carbamoyl chlorides or isocyanates generated as yield-diminishing byproducts.
However, when the ZWK reaction is applied with (N-ethoxythiocarbonyl)urethane as the starting material, heterocyclization to
the putative “Dts-urethane” does not occur. Instead, the reaction directly provides (chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)-
disulfane, a reasonably stable crystalline compound; modified conditions stop at the (chlorocarbonyl)[1-ethoxy-(N-
ethoxycarbonyl)formimidoyl]disulfane intermediate. The title (chlorocarbonyl)(carbamoyl)disulfane cannot be converted to
the elusive Dts derivative, but rather gives (N-ethoxycarbonyl)carbamoyl chloride upon thermolysis, or (N-ethoxycarbonyl)-
isocyanate upon treatment with tertiary amines. Additional transformations of these compounds have been discovered, providing
entries to both known and novel species. X-ray crystallographic structures are reported for the title (chlorocarbonyl)-
(carbamoyl)disulfane; for (methoxycarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane, which is the corresponding adduct after
quenching in methanol; for [1-ethoxy-(N-ethoxycarbonyl)formimidoyl](N′-methyl-N′-phenylcarbamoyl)disulfane, which is
obtained by trapping the title intermediate with N-methylaniline; and for (N-ethoxycarbonylcarbamoyl)(N′-methyl-N′-
phenylcarbamoyl)disulfane, which is a short-lived intermediate in the reaction of the title (chlorocarbonyl)(carbamoyl)disulfane
with excess N-methylaniline. The new chemistry and structural information reported herein is expected to contribute to accurate
modeling of the ZWK reaction trajectory.
and the discoveries reported herein about its structure and
further transformations, set the stage for a fundamental rethink-
ing of past assumptions and the development of new mechanistic
perspectives.
INTRODUCTION
■
A 1966 patent by Zumach, Weiss, and Kuhle1−3 described a
̈
general method (ZWK reaction) for preparation of 1,2,4-
dithiazolidine-3,5-diones (1) by the facile and rapid reaction of
O-ethyl thiocarbamates (2) plus (chlorocarbonyl)sulfenyl
chloride (3) (Scheme 1). The heterocyclic system 14−10 was
subsequently adopted as the basis of the orthogonally removable
dithiasuccinoyl (Dts) amino protecting group for peptide
synthesis,5,11,12 and can be exploited for a myriad of additional
applications.8,12−18
Our interest in developing reliable routes to 1 provided the
impetus for an extensive series of studies regarding the mecha-
nism of the ZWK reaction, and related chemistry.5−10,19,20 The
focus of the present work is on the unique urethane-derived
family (R = CO2Et, series e) in which analogues corresponding
to proposed intermediates in the ZWK reaction mechanism can
be isolated and characterized, in some cases by X-ray crystal-
lography. Nevertheless, in this particular system, the desired final
1e is not accessible. Our observation that the unexpectedly stable
title (chlorocarbonyl)(carbamoyl)disulfane 6e forms instead,
RESULTS AND DISCUSSION
■
Reaction of (N-Ethoxythiocarbonyl)urethane (2e) with
(Chlorocarbonyl)sulfenyl Chloride (3). An initial goal of
these studies was to obtain the putative structure 1e, which would
be the Dts analogue of the Nefkens reagent, (N-ethoxycarbonyl)-
phthalimide;24,25 the latter compound reacts smoothly with
α-amino acids in aqueous bicarbonate to form the appropriate
N-phthaloyl derivatives. The starting material for our purposes,
(N-ethoxythiocarbonyl)urethane (2e), is prepared in modest
yields by an ancient method, due to Delitsch,26 that involves
reaction of ammonium thiocyanate with ethyl chloroformate in
Received: August 6, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Formation of Dts-Amines (1) by the ZWK Reaction, Along with the Mechanistic “Conventional Wisdom” and Related
a b c d e f
, , , , ,
Pathways
a
Literature citations to 1,2,4-dithiazolidine-3,5-dione (Dts-amine) derivatives (1) are not comprehensive, and there are a handful of routes to 1
beyond the robust “classic” ZWK reaction [2 plus 3] depicted within the large box on the top line of this scheme. A wide range of primary and
secondary alkyl groups can be used in place of the ethyl group shown for starting thiocarbamates 2. Species in brackets are only inferred or
postulated, while those not in brackets are stable enough that their presence in the reaction mixtures can be demonstrated (and all, except 8 and 9,
can be isolated after standard workup and purification). In particular, the conversion of putative 6 to furnish 1, while plausible, has never been established
b
experimentally. The ZWK reaction (refs 1−3) was reported originally in the presence of 1 equiv of a tertiary amine base (e.g., pyridine, Et3N, etc.) as
a hydrogen chloride acceptor, but our own experiments going back four decades (vide infra) have shown that the rapid formation of heterocycle 1
occurs just as well in the absence of base. Therefore, this scheme shows structures of proposed intermediate species 4 through 8 as they would
c
appear in the presence of HCl. As indicated in the preceding footnote, these reactions are also generally successful in the presence of bases, e.g.,
pyridine, Et3N, etc., as HCl acceptors; in a useful variation (ref 6), the Et group in 2 is replaced by dimethylaminoethyl, which serves as a “built-in”
base and also allows ready removal of byproducts that contain the O-alkyl group. In the presence of base, intermediate 4 should be drawn as the
d
neutral (deprotonated) species, a (chlorocarbonyl)(formimidoyl)disulfane. Carbamoyl chlorides (8) (absence of base) or isocyanates (9) are yield-
diminishing byproducts of the ZWK reaction, and 1,2,4-thiadiazolidine-3,5-diones (10) are formed as well (with or without base present), as has
been documented (refs 4 and 6). Note that the formation of 10 is consistent with the earlier presence of a (carbamoyl)sulfenyl chloride (7)-type
intermediate, which can be generated independently by careful chlorination of 2 (alone), following previous precedents (refs 6 and 23) and shown
e
on the upper left of the scheme. For series a (R = H), both 4a and 5a (11, i.e., 3-ethoxy-1,2,4-dithiazoline-5-one (EDITH), as drawn in the small
f
box in the lower right of the scheme) exist as the neutral species (without HCl) (more details in ref 9). The primary experimental focus of the
present paper is on series e (R = CO2Et). Structures 1e and 5e were erroneously claimed earlier (footnote 13 in ref 5); based on our current
understanding, these substances should have been assigned respectively to structures 6e and 4e (this latter compound forms as the neutral species,
without HCl).
ethanol. Alternatively, thiocarbamate 2e is obtained by the
quantitative addition of ethanol (solvent) to (N-ethoxycarbonyl)-
isothiocyanate (12) (a likely intermediate in the Delitsch
procedure, and commercially available for this purpose); the
new route is rapid and does not require a basic catalyst.
When the reaction of thiocarbamate 2e with (chlorocarbonyl)-
sulfenyl chloride (3) is carried out in the presence of water
droplets heterogeneously dispersed throughout the medium,
or in the presence of pyridine or triethylamine to absorb the
hydrogen chloride coproduct, the process can be arrested at
the initial adduct, 4e. Straightforward workup gives quantitatively
an oil that is pure by 1H and 13C NMR, and can be characterized
further by IR and mass spectrometry. When anhydrous hydrogen
chloride gas is passed through a solution of substrate 4e in CDCl3,
(chlorocarbonyl)(carbamoyl)disulfane 6e forms, together with an
equivalent amount of ethyl chloride.
Further Transformations of Chlorocarbonyl Disulfanes
4e and 6e (Schemes 2 and 3). Following earlier precedents on
simpler substrates,27,28 formimidoyl disulfane 4e is readily carried
forward to the corresponding N-methylanilide 13, which shows 1H
and 13C NMR, IR, mass spectra, and elemental analysis consistent
with the anticipated structure; X-ray crystallographic analysis (see
Figure 1, later) definitively established the structure of 13. More-
over, 4e, as an oil in open atmosphere at 25 °C, decomposes to an
approximately equimolar mixture of 6e (from loss of ethyl chloride)
In pilot work, the reaction of thiocarbamate 2e with (chloro-
carbonyl)sulfenyl chloride (3)2,27 was carried out in CDCl3, and
monitored by 1H NMR (Scheme 2) at 25 °C. Starting material 2e
is replaced immediately by an initial adduct with altered shifts
for the two ethyl groups; this intermediate can be assigned to
(formimidoyl)disulfane structure 4e (see paragraph that
follows). Then, with a half-life of 5 min to 2 h, depending on
the concentrations of reactants, intermediate 4e is converted
cleanly to ethyl chloride plus a product with a single ethyl group.
The reaction is readily scaled up, and the final product can be
isolated and crystallized in overall 75% yield; elemental analysis,
1H and 13C NMR, IR, and mass spectrometry all support the
novel and unexpected (chlorocarbonyl)(carbamoyl)disulfane
structure 6e. As described later (Figure 1), unambiguous proof
for the structure of 6e comes from X-ray crystallography.
B
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 2. Chemistry to Create (Chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane (6e) and Related Derivatives, and Some
ab c d e
, , , ,
Further Transformations
a
The straightforward reaction of 2e with 3 (box on top of scheme), carried out in CDCl3, shows 4e as a spectroscopically detectable intermediate,
but then 6e is the sole isolated product. Under modified conditions (details in text), the reaction can be arrested at isolable, characterizable 4e.
b
Interestingly enough, the reaction conditions used to transform (chlorocarbonyl)(carbamoyl)disulfane 6e to O-ethyl carbamate (15) are nominally
the same ones suggested by Nefkens (ref 24) for reaction of α-amino acids with (N-ethoxycarbonyl)phthalimide to introduce N-phthaloyl
protection. In the present case, substrate 6e is not soluble in water, but as the reaction takes place, and insoluble 6e is consumed over a 1 h period at
c
d
25 °C, the product mixture gradually becomes homogeneous (by the end point, elemental sulfur precipitates). Base is pyridine or Et3N. The
e
adduct 16 is quite stable under some conditions, but can also lose COS and elemental sulfur to produce derivative 19 (details in text). Treatment of
pure 16 with base gives 19 plus COS and elemental sulfur, in a reaction that is probably mechanistically similar to the treatment of 6e with base to
give acyl isocyanate 9e (see top of Scheme 3). These base-catalyzed conditions also produce O-ethyl carbamate (15) (plus two COS and methanol),
presumably due to reaction with residual water, either atmospheric or in the solvent.
plus the hydrolysis product (N-ethoxycarbonyl)urethane (14).29
However, 4e remains unchanged when stored as a solution in
CDCl3 for several months at −20 °C.
sole organic product. This result suggests that the imide N−H
proton of 6e is particularly acidic, and that its ready abstraction
drives the expulsion of choride ion, COS gas, and elemental
sulfur. Moreover, brief heating of 6e as a neat melt gives the
water-sensitive carbamoyl chloride 8e,32−35 again with concom-
itant loss of COS and elemental sulfur (Scheme 3).
When (methoxycarbonyl)(carbamoyl)disulfane 16, in which a
chlorine atom of 6e is replaced by a methoxy moiety, is treated
with base, a mixture of (N-methoxycarbonyl)urethane (19)36
and O-ethyl carbamate (15) is produced (Scheme 2, especially
note e); this reaction is essentially instantaneous when the base is
(Chlorocarbonyl)(carbamoyl)disulfane 6e hydrolyzes quanti-
tatively in aqueous bicarbonate to give O-ethyl carbamate
(urethane 15), along with elemental sulfur, carbon dioxide
(CO2), and carbonyl sulfide (COS) (Scheme 2; see especially
note b). However, the essential skeleton of 6e is maintained
when this compound is quenched in methanol to provide mixed
(methoxycarbonyl)(carbamoyl)disulfane 16, which showed
1H and 13C NMR, IR, mass spectra, and elemental analysis
consistent with the structural assignment; X-ray crystallographic
analysis (see Figure 1, later) gave definitive proof of its structure.
Carbamoyl disulfane 16 can also be made independently
(Scheme 2) by the Harris reaction2−4,23,27,28,30 of thiocarbamate
2e plus (methoxycarbonyl)sulfenyl chloride (17)2,27 in the
absence of base. The corresponding Harris reaction9,30 (pre-
cedented with simpler thiocarbamates and sulfenyl chlorides) in
the presence of a tertiary amine base (pyridine or triethylamine)
gives the novel (methoxycarbonyl)(formimidoyl)disulfane 18,
which is structurally analogous to compounds 4e and 13 that have
been discussed already (see Scheme 2 and previous paragraph).
Treatment of (chlorocarbonyl)(carbamoyl)disulfane 6e with
base cleanly gives (N-ethoxycarbonyl)isocyanate (9e)31 as the
triethylamine, and still fast, albeit with measurable kinetics (t1/2
∼
40 min), when the weaker base pyridine is used. Given that 6e
is quite stable when stored for several years under ambient
conditions, it is surprising to find that carbamoyl disulfane 16
transforms over a period of months to acyl urethane 19, all while
remaining in the solid state. To accentuate this observation, pure
16, when heated for 20 min at 100 °C, changes quantitatively to
19, with expulsion of COS and elemental sulfur.
The conversion of formimidoyl disulfane 4e to N-methyl-
anilide 13, demonstrated earlier in this work (Scheme 2), is a
prime example of the reliable and robust N-methylaniline
derivatization paradigm.27,28 In contrast, the analogous con-
version of carbamoyl disulfane 6e to N-methylanilide 22 proved
C
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 3. Further Transformations, Both Expected and Surprising, of (Chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane
a b c
, ,
(6e), and an Indirect Route to N-Methylanilide 22
a
Just as reaction of acyl isocyanate 9e plus MeOH (solvent) provides derivative 19 quantitatively (as shown here), reaction of 9e plus EtOH
b
c
provides 14 (see Scheme 2 for a different way to prepare this compound). Base is pyridine or Et3N. Reaction of (chorocarbonyl)-
(carbamoyl)disulfane 6e plus N-methylaniline, followed by standard aqueous workup, does not give the expected N-methylanilide 22; instead,
carbamoyl urea 20 is isolated. However, 22 does form in situ, and then reacts further with excess N-methylaniline, leading to 20, along with COS and
elemental sulfur. This scenario was confirmed by reacting pure 22 (made by the alternate route) with N-methylaniline, resulting in the same
outcome. Finally, upon prolonged storage at 5 °C, a substantial portion of 22 decomposes to 20.
Figure 1. ORTEP representations of compounds 6e, 13, 16, and 22, with 50% displacement ellipsoids and all non-H atoms labeled and numbered.
Compounds are shown in comparable orientations so as to emphasize their structural similarities. The structure of 16 shows disorder in the “urethane”
half of the disulfane. The structure of 22 has considerably smaller displacement ellipsoids due to the lower temperature at which data was collected (see
Supporting Information Table S1 for further details). Selected structural parameters are listed in Table 1 of the main text.
difficult to achieve. Thus, the expected derivative 22 forms
initially as the primary product (and can even be isolated when
workup is carried out within 5 min), but reacts further to provide
carbamoyl urea 2037 as the sole isolated product from standard
protocols. Reinforcing this idea, a sample of pure 22 (made by
the method detailed in the next paragraph), when treated with
N-methylaniline, is almost completely converted to 20 within
3.5 h at 25 °C. These results demonstrate that the propensity of
N-methylaniline to react rapidly with the electrophilic carbonyl
chloride of 6e takes precedence over acid−base chemistry with
the imide N−H of this substrate. However, N-methylaniline is
still basic enough that any excess will abstract the acidic N−H
of 22 and initiate the loss of COS and elemental sulfur that
generates as an intermediate acyl isocyanate 9e, which can react
further with excess N-methylaniline to provide the eventual
product 20.
For completeness, the desired N-methylanilide of 6e, i.e.,
compound 22 (not accessible by direct conversion, for reasons
already covered in Scheme 3 and the preceding paragraph), can
be obtained by an alternative, indirect multistep route (Scheme 3,
lower portion). First, thiocarbamate 2e is converted to (carbamoyl)-
sulfenyl chloride 7e by reaction with sulfuryl chloride, following
precedents from our earlier research.6,23 Next, an in situ Harris
reaction2−4,23,27,28,30 traps intermediate 7e with O-isopropyl-N-
methyl-N-phenylthiocarbamate (21),23 and the desired 22 is
obtained in a facile, rapid process that also generates isopropyl
chloride. While the crude product forms in high yield with
reasonable purity, further crystallization provides a purified
product that satisfies all of the required analytical characteristics
(1H and 13C NMR, IR, HRMS, and elemental analysis). Unam-
biguous proof for the structure of 22 was provided by X-ray
crystallography (see Figure 1, next section).
D
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. Selected Bond Lengths (Å), Bond Angles (°), and Torsion Angles (°)
6e
13
16
22
C6−O6 (Å)
N5−C6 (Å)
O6−C11 (Å)
C6−S7 (Å)
C6−N5−C4 (°)
C6−S7−S8−C9 (°)
1.208 (2)
1.365 (2)
1.318 (3)
1.282 (4)
1.456 (4)
1.780 (3)
117.8 (3)
82.6 (2)
1.209 (2)
1.374 (2)
1.200 (2)
1.374 (2)
1.814 (2)
126.1 (1)
−81.8 (1)
1.806 (2)
125.9 (2)
87.5 (1)
1.821 (1)
126.1 (1)
82.6 (1)
X-ray Crystallographic Structures of (Chlorocarbonyl)-
(N-ethoxycarbonylcarbamoyl)disulfane (6e), [1-Ethoxy-
(N-ethoxycarbonyl)formimidoyl](N′-methyl-N′-
phenylcarbamoyl)disulfane (13), (Methoxycarbonyl)-
(N-ethoxycarbonylcarbamoyl)disulfane (16), and
(N-Ethoxycarbonylcarbamoyl)(N′-methyl-N′-phenyl-
carbamoyl)disulfane (22). Four of the acylcarbamoyl
disulfane derivatives encountered in this work were amenable
to X-ray crystallographic analysis. These were the title
(chlorocarbonyl)(carbamoyl)disulfane 6e, its methyl ester 16,
its N-methylanilide 22 that had been created by an indirect route,
and the N-methylanilide 13 that is derived from intermediate
formimidoyl disulfane 4e. Key geometric parameters have been
compiled (Table 1, excerpted from more comprehensive listings
in the Supporting Information), with the twin goals to map the
structural changes in the transformations from R′SSC(OEt)NR
to R′SS(CO)NHR + EtCl, and to further understand the
circumstances under which the linear carbamoyl disulfane
intermediate does (or does not) heterocyclize.
All atoms that are equivalent in structures 6e, 16, and 22 are
essentially superimposable, with the exception that the torsion
angle of the S−S bond of 6e is opposite in sign to that of 16 and
22; this means that the C9−O9 carbonyl is pointed in an
opposite direction in 6e, by comparison to 16 and 22. In all
four structures, the acidic carbamoyl N−H’s, or the lone pair
electrons of the formimidoyl nitrogen, are anti to the disulfane
(notwithstanding that, in Schemes 2 and 3, these conformations
are depicted as syn so as to mimic the mechanistic sequence in
Scheme 1). Even allowing for reasonable rotations around
various single bonds, the anti conformation is inconsistent with
attack of a nucleophilic electron pair on nitrogen onto the
chlorocarbonyl group of 4e (as modeled by 13) or 6e (as
arranged in its crystal structure, and further modeled by deriva-
tives 16 and 22). While we recognize that solid state con-
formations do not necessarily represent the behavior of the
molecules in solution, our observations about the preference for
anti conformations may offer a partial explanation for why, in this
special case, neither 4e nor 6e cyclizes to the corresponding Dts-
carbamate.
(carbamoyl)disulfane 6e that models a plausible intermediate
(6) previously proposed in the mechanism of the ZWK reac-
tion.2,3 The fact that the intermediate does not cyclize to form
the corresponding 1,2,4-dithiazolidine-3,5-dione (1) means that
earlier mechanistic assumptions will require reassessment. The
developing evidence from this work, and from related studies
(both published with extensive details,5−7,9 as well as some that
are yet preliminary19,20,43,44), is consistent with a different and
more nuanced view of the ZWK reaction: an initial adduct 4 (also
modeled herein, with formimidoyl disulfane 4e) must cyclize
first, prior to loss of ethyl chloride. The information developed
here might also explain why reactions of amines with bis-
(chlorocarbonyl)disulfane27 fail to provide 1, whereas the same
reagent with bis(trimethylsilyl)amines smoothly gives10 1 plus 2
equiv of TMS-Cl.
EXPERIMENTAL SECTION
■
General. 1H NMR spectra were recorded primarily in CDCl3 at 300
MHz (mostly, and assumed if not specified otherwise) or 500 MHz
(some), with CDCl3 normalized to 7.27 ppm. Exchangeable protons,
which give rise to broad peaks at variable chemical shifts, were not
tabulated. Coupling constants were ∼7.2 Hz for adjacent aliphatic C−H
and are not further reported. 13C NMR spectra were recorded in CDCl3
at 75 or 125 MHz, with CDCl3 normalized to δ 77.0 ppm. Whenever a
solvent other than CDCl3 was used, this is stated. Fourier transform IR
spectra were recorded in CH2Cl2 or CDCl3 solutions placed in NaCl
cells. In most cases, high-resolution mass spectra were acquired with
electrospray ionization and a time-of-flight (TOF) mass analyzer. Some
chlorine-containing compounds were analyzed by methane chemical
ionization mass spectrometry on an instrument that had a solids probe.
Elemental compositions were determined by combustion analysis
(for C, H, N, S) and ion chromatography (for Cl). Many of the starting
materials and reagents were made by published procedures that are
referenced appropriately. Unless specifically indicated otherwise, all
reactions were carried out under ambient conditions, i.e., 25 °C
(notwithstanding occasional spontaneous exotherms, which tended to
be relatively small). All workup procedures were carried out at 25 °C as
well, and all solvent evaporations were conducted under aspirator
vacuum (∼10 mm).
X-ray Data Collection, Solution, and Refinement. Data
collection was carried out using Cu Kα (22) or Mo Kα (6e, 13, 16)
radiation. Crystal structures were solved using SHELXS-97 (6e, 16, 22)
or Texsan (13), and refined using SHELXL-97 (6e, 16), SHELXTL
(13), or SHELXL-2014/6 (22).45 All non-hydrogen atoms were refined
with anisotropic displacement parameters. All hydrogen atoms were
placed in ideal positions and refined as riding atoms with relative
isotropic displacement parameters. Further relevant information is in
the Supporting Information (Table S1). CIF files for the X-ray
diffraction crystal structures of 6e, 13, 16, and 22 have been deposited at
the Cambridge Crystallographic Data Center (CCDC) under accession
codes 1430178, 1430179, 1430180, and 1430181, respectively.
(N-Ethoxythiocarbonyl)urethane (2e). Method A. Following
Delitsch,26 ethyl chloroformate (50 mL, 0.52 mol) was added all at
once to a stirred solution of ammonium thiocyanate (38 g, 0.5 mol) in
absolute EtOH (150 mL). The reaction mixture turned light orange;
within 30 min, it spontaneously reached 35 °C while a precipitate of
ammonium chloride formed. After 3 h, the temperature had subsided,
The four structures described in the current work can be
compared to previously reported structures for the starting
thiocarbamate 2e,38 for several other compounds39 that model
(chlorocarbonyl)(carbamoyl)disulfanes by replacement of the
acid chloride with a trichloromethyl moiety, for a cyclized inter-
mediate (11)9 that has not yet lost ethyl chloride (created by
ZWK chemistry with R = H), and for several Dts-amines
(1).9,40−42 Taken together, these should provide the basis for a
complete structural analysis of all of the participants and possible
intermediates of the ZWK reaction.
SUMMARY AND CONCLUSIONS
■
In summary, structural parameters and chemical reactivities have
been elucidated for the surprisingly stable (chlorocarbonyl)-
E
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
and the salt (24.9 g, 97%) was removed by filtration. Concentration of
the filtrate in vacuo gave the title product as a reasonably NMR
pure gold-yellow oil (48 g, ∼55%) which was taken up in hot hexanes
(∼350 mL), filtered, and cooled to 4 °C. Yield: 21.2 g (24%), white
needles, mp 41−43 °C (lit.26,46 mp 43−44 °C). 1H NMR (300 MHz):
δ 4.61 (q, 2 H), 4.22 (q, 2 H), 1.43 (t, 3 H), 1.30 (t, 3 H). 13C NMR
(75 MHz): δ 188.6, 148.9, 69.0, 62.4, 14.1, 13.6. HRMS (ESI): m/z
[M + Na+] calcd for C6H11NO3S: 200.0352. Found: 200.0353; IR
(CDCl3) 2986 (m), 2938 (w), 1768 (s), 1506 (vs) cm−1; Anal. Calcd for
C6H11NO3S (mol wt 177.22): C, 40.66; H, 6.26; N, 7.90; S, 18.09.
Found: C, 40.62; H, 6.14; N, 8.05; S, 18.10.
in CH2Cl2 (50 mL) was added over 20 min to a solution of (N-ethoxy-
thiocarbonyl)urethane (2e) (8.7 g, 50 mmol) in CH2Cl2 (50 mL) at
5 °C. (A separate NMR kinetics experiment in which both 3 and 2e
were 0.1 M in CDCl3 at 25 °C revealed that intermediate 4e formed
essentially instantaneously, and then EtCl plus product 6e formed with
t1/2 ∼ 20 min.) The mixture was stirred an additional 30 min at 5 °C and
40 min at 25 °C, and then concentrated to give the crude title product,
an oil that was triturated with hexanes and became a light yellow solid
(9.6 g, 79%). Analytically pure white needles (5.3 g), mp 91−93 °C, were
obtained by recrystallization from minimal hot hexanes (∼10 mL/g).
An additional fraction of slightly yellow needles (3.8 g), mp 88−91 °C,
was obtained from the mother liquor. Total yield: 9.1 g (75%). 1H NMR
(CDCl3): δ 4.32 (q, 2H), 1.35 (t, 3H). 13C NMR (CDCl3): δ 163.8, 163.6,
152.3, 63.9, 14.1. IR (CDCl3): 3385 (w), 2987 (w), 1776 (m), 1750 (s),
1713 (s), 1481 (m), 1203 (s), 1056 (m), 812 (s) cm−1. Positive methane
CIMS (source 160 °C, solid probe 90 °C, 0.1 mm): m/z 244 [(M + 1)+,
31%], 216 [(M + 1)+ − CO, 8%], 148 [EtO(CO)NH(CO)S+,
62%], 116 [EtO(CO)NH(CO)+, 100%], 88 [EtO(CO)NH+,
86%]. EIMS (source 200 °C, solid probe 50 °C): m/z 243 (M+•, 0.1%),
155 [Cl(CO)SS(CO)+, 1%], 148 [EtO(CO)NH(CO)S+,
2%], 128 [Cl(CO)SSH+, 6%], 116 [EtO(CO)NH(CO)+, 4%],
88 [EtO(CO)NH+, 26%], 70 [N(CO)2+, 100%], 64 (S2+, 18%),
60 (COS+•, 68%), 45 (EtO+, 24%). Anal. Calcd for C5H6NO4S2Cl
(mol wt 243.68): C, 24.65; H, 2.48; N, 5.75; Cl, 14.55; S, 26.31. Found:
C, 24.79; H, 2.57; N, 5.89; Cl, 14.46; S, 26.40. This material was best
stored at −20 °C, although material stored at 5 °C was still useable after
several years. The structure of this compound was also proved by single
crystal X-ray analysis (see Figure 1).
Method B. Working on the same scale, the crude product (∼70%
1
pure by H NMR, with no specific impurity >5%; none identified
further) was distilled directly, bp 85−92 °C (0.5 mm) [lit. bp47 135 °C
(13 mm)], to provide a pure amorphous white solid (14.1 g, 18%), that
was recrystallized further from hot hexanes (200 mL) to give the pure
title product as white needles (8.1 g, 11% overall), mp 37−38 °C; a
second crop was a white fluffy solid (2.5 g, 3% more), mp 35−37 °C.
Method C. (N-Ethoxycarbonyl)isothiocyanate (12) (1.6 g, 12 mmol)
[1H NMR (CDCl3) δ 4.27 (q, 2H); 1.34 (t, 3H); 13C NMR (CDCl3) δ
150.8, 149.4, 65.2, 13.9] was dissolved in absolute ethanol (10 mL), with
a slight spontaneous exotherm (maximum 32 °C). After stirring at 25 °C
for 1 h, solvent was removed in vacuo to produce an oil (2.1 g, 98%),
which was placed under hexanes at −20 °C to produce off-white needles
1
(1.6 g, 75%), mp 48−52 °C, H and 13C NMR identical to material
prepared by method A.
(Chlorocarbonyl)[1-ethoxy-(N-ethoxycarbonyl)formimidoyl]-
disulfane (4e). Method A. Over a period of 15 min, (chlorocarbonyl)-
sulfenyl chloride (3)27 (0.46 mL, 5.6 mmol) in CH2Cl2 (25 mL) was
added dropwise at 5 °C to a well-stirred heterogeneous mixture of water
(1.0 mL, 56 mmol) with a solution of (N-ethoxythiocarbonyl)urethane
(2e) (1.0 g, 5.6 mmol) in CH2Cl2 (25 mL). Stirring continued at 25 °C
for 30 min, and workup by washing with equal volumes of 1 N aqueous
HCl and water, drying (MgSO4), and concentration in vacuo gave the 1H
Hydrolysis of (Chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)-
disulfane (6e). The title substrate 6e (206 mg, 0.8 mmol) was sus-
pended in 0.1 M aqueous sodium bicarbonate (30 mL), with vigorous
magnetic stirring. After 1 h, everything became soluble, and gases
evolved. The reaction was continued for a further 2 h, and a precipitate
of elemental sulfur formed. The mixture was filtered, partially concen-
trated (∼5 mL), extracted with CHCl3 (3 × 10 mL), dried (MgSO4),
and concentrated in vacuo to provide a white solid (62 mg, 87%), mp
45−48 °C (lit.48 mp 48−50 °C), that by 1H NMR [δ 4.12 (q, 2H); 1.26
(t, 3H)] and IR [3350−3500 (br s); 1713 (s); 1596 (m)] was
indistinguishable from commercial O-ethyl carbamate (urethane) (15).
The same experiment (0.4 mmol scale) was carried out in D2O. The
1
NMR pure title product, a viscous yellow oil (1.46 g, 94%). H NMR
(CDCl3): δ 4.52 (q, 2H), 4.26 (q, 2H), 1.39 (t, 3H), 1.36 (t, 3H). 13
C
NMR (CDCl3): δ 169.6, 163.3, 161.0, 69.8, 63.2, 14.2, 13.7; IR (CDCl3)
2984 (m), 1775 (s), 1745 (s), 1712 (s), 1681 (s), 1574 (vs), 1471 (m),
1445 (m), 1391 (m), 1371 (s), 1312 (s), 1278 (s), 1206 (s), 1096 (m),
1057 (m), 1006 (m) cm−1. Positive methane CIMS (source 160 °C,
solid probe 65 °C): m/z 272 [(M + 1)+, 51%], 226 [(M + 1)+ − EtOH,
10%], 176 [EtO(CO)NC(OEt)S+, 100%], 116 [EtO(C
O)NH(CO)+, 16%].
1
aqueous phase was examined after 3 h by H NMR (D2O), which
established that O-ethyl carbamate-d2 had formed quantitatively. The
reaction mixture (in D2O) was extracted into CDCl3, revealing a
1H NMR spectrum that was superimposable on that of commercial 15,
except that the broad NH signal was absent. Furthermore, the IR
spectrum showed the absence of the amide II band at 1596 cm−1 and a
shift of the amide (N−H vs N-D) stretching frequency to lower energy,
i.e., from 1729 to 1709 cm−1.
Base-Catalyzed Decomposition of (Chlorocarbonyl)-
(N-ethoxycarbonylcarbamoyl)disulfane (6e). A solution of substrate
6e (150 mg, 0.6 mmol) in CDCl3 (3 mL) was treated with Et3N (85 μL,
0.6 mmol). The reaction mixture immediately became yellow, and a
white precipitate (presumably Et3N·HCl, admixed with elemental
sulfur) formed. The filtrate from the reaction was examined within
5 min by 1H NMR, which completely matched a standard of
(N-ethoxycarbonyl)isocyanate (9e) and also showed the expected
peaks due to the Et3N salt. 13C NMR also matched 9e and Et3N·HCl,
and showed an additional peak at 152.7 ppm corresponding to COS.
Similarly, the IR spectrum included the strong characteristic peaks for 9e
at 2249, 1741, and 1222 cm−1, as well as 2043 cm−1 for COS.
The experiment was repeated on the same scale using pyridine
(48 μL, 0.6 mmol) as the base, and corresponding results were obtained
(end point observed within 30 min).
The title compound, when stored in open atmosphere for a few
days at 25 °C, decomposed and/or reacted further to produce an
approximately equimolar mixture of (chlorocarbonyl)(carbamoyl)-
disulfane 6e (presumably with loss of EtCl) plus the hydrolysis product
(N-ethoxycarbonyl)urethane (14). Consequently, 4e was stored as a
solution (1.8 M) in CDCl3. At −20 °C, there was no change for several
months, whereas at 25 °C, there was a clean transformation to 6e plus
EtCl, t1/2 ∼ 6 days.
Method B. A solution of (chlorocarbonyl)sulfenyl chloride (3)
(0.42 mL, 5.0 mmol) in CH2Cl2 (25 mL) was added over 15 min to
a solution of thiocarbamate 2e (885 mg, 5.0 mmol) plus pyridine
(0.40 mL, 5.0 mmol) in CH2Cl2 (25 mL) at 5 °C. The homogeneous
reaction mixture was stirred for an additional 30 min at 25 °C, and then
washed with equal volumes of water (3×), dried (MgSO4), and concen-
trated in vacuo to give the title product as a viscous yellow oil (1.1 g,
81%), with 1H and 13C NMR identical to material prepared by
method A. Furthermore, the identical reaction but substituting Et3N
(0.70 mL, 5.0 mmol) for pyridine gave the same overall yield and
spectral characteristics.
Treatment of (Chlorocarbonyl)[1-ethoxy-(N-ethoxycarbonyl)-
formimidoyl]disulfane (4e) with Hydrogen Chloride. A solution
(1.8 M) of title substrate in CDCl3 (0.5 mL) was treated with a slow
stream of HCl gas for 10 min at 5 °C; 1H NMR examination revealed
quantitative formation of EtCl [1H NMR (CDCl3) δ 3.56 (q, 2H); 1.48
(t, 3H)] plus (chlorocarbonyl)(carbamoyl)disulfane 6e.
Thermolysis of (Chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)-
disulfane (6e). Substrate 6e (100 mg, 0.4 mmol) was heated at
100 °C in a sealed tube for 1 h. A yellow solid (11 mg, 85% for elemental
sulfur) remained at the bottom of the tube, and a clear white solid
sublimed to the top. The tube was cooled and vented carefully in order
to allow gases to escape. By working rapidly, the white solid was
removed from the tube, weighed (∼50 mg, ∼80%), and dissolved in
(Chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane (6e). A
solution of (chlorocarbonyl)sulfenyl chloride (3)27 (4.2 mL, 50 mmol)
F
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
CDCl3 (∼1 mL) for examination by IR and 1H and 13C NMR, wherein
it matched (N-ethoxycarbonyl)carbamoyl chloride (8e) in all regards.
Even 5 min exposure of the solid product (8e) to atmospheric humidity
resulted in ∼50% hydrolysis to O-ethyl carbamate (15).
Treatment of (Chlorocarbonyl)(N-ethoxycarbonylcarbamoyl)-
disulfane (6e) with N-Methylaniline. Method A. With no special
precautions for external cooling, a solution of N-methylaniline (2.2 mL,
20 mmol) in CH2Cl2 (20 mL) was added all at once to a solution of
substrate 6e (1.0 g, 4.1 mmol) in CH2Cl2 (20 mL). After 70 min at
25 °C, the homogeneous reaction mixture was washed with 2 N aqueous
HCl (2 × 50 mL) and H2O (50 mL), dried (MgSO4), and concentrated
in vacuo to provide a sticky yellow solid (0.6 g, 66%), which upon
standing under hexanes at −20 °C gave white needles (0.43 g, 47%),
mp 65−67 °C, that on the basis of its 1H NMR and IR was concluded to
be carbamoyl urea 20.
Method B. A solution of N-methylaniline (246 mg, 2.3 mmol) in
CDCl3 (5 mL) was slowly added to a solution of substrate 6e (243 mg,
1.0 mmol) in CDCl3 (5 mL) at 5 °C. After 10 min, 1H NMR revealed
that the reaction mixture comprised primarily 22, along with N-methyl-
aniline hydrochloride, although some 20 (about 12% compared to 22)
was already present. A second time point, 20 min into the reaction,
showed a 1:1 ratio of 22 to 20. After 45 min, the homogeneous mixture
was washed with 1 N aqueous HCl (3 × 10 mL) and brine (10 mL),
dried (MgSO4), and concentrated in vacuo to give a yellow solid
(256 mg) that comprised 22 and 20 in a 1:4 ratio.
297 [(M + 1)+ − EtOH, 31%], 178 [EtO(CO)NC(OEt)SH2+,
33%], 176 [EtO(CO)NC(OEt)S+, 16%], 168 (18%), 166
[PhMeN(CO)S+, 38%], 150 (28%), 134 ([PhMeN(CO)+,
100%], 108 (29%), 107 (11%), 106 (10%). Anal. Calcd for
C14H18N2O4S2 (mol wt 342.43): C, 49.11; H, 5.30; N, 8.18; S, 18.72.
Found: C, 48.91; H, 5.18; N, 8.15; S, 18.61. The structure of this com-
pound was also proved by single crystal X-ray analysis (see Figure 1).
(N-Ethoxycarbonyl)urethane (14). Method A. Essentially from a
procedure by Tompkins and Degering,29 sodium (2.6 g, 0.11 mol) was
added to O-ethyl carbamate (15) (10.0 g, 0.11 mol) in dry xylenes
(100 mL), and the mixture was refluxed (140 °C) until a thick white
suspension appeared that was indicative of complete reaction of the
sodium. After cooling to 90 °C, ethyl chloroformate (10.7 mL, 0.11 mol)
was added dropwise, and stirring continued at 90 °C for 20 h. The
reaction mixture was then cooled, filtered, and concentrated in vacuo.
The 1H NMR pure crude product (16.8 g, 94%) was distilled, bp 107 °C
(1.5 mm) [lit.29 bp 142−145 °C (10 mm)] to furnish white crystals
(12.0 g, 68%), mp 47−48 °C (lit.29 mp 49−50 °C). 1H NMR (300 MHz):
δ 4.24 (q, 4 H), 1.30 (t, 6 H). 13C NMR (75 MHz) δ 150.7, 62.3, 14.2.
Method B. Neat (N-ethoxycarbonyl)isocyanate (9e) (1.0 g, 87
mmol) was added to EtOH (10 mL). After 30 min, the reaction solution
was concentrated in vacuo to give the 1H NMR pure title product as a
1
white crystalline solid (1.39 g, 99%) mp 45−46 °C, with H and 13C
NMR identical to those of material prepared by method A. HRMS (ESI):
m/z [M + Na+] calcd for C6H11NO4: 184.0580. Found: 184.0593.
(Methoxycarbonyl)(N-ethoxycarbonylcarbamoyl)disulfane (16).
Method A. The corresponding chlorocarbonyl substrate 6e (255 mg,
1.1 mmol) dissolved smoothly in methanol (5 mL). After 15 min, the
reaction mixture was concentrated in vacuo to provide the title product
as a fine white powder (250 mg, 96%) that was pure by 1H NMR. This
material was dissolved in minimal CH2Cl2 (∼0.5 mL), hexanes (15 mL)
were added, and storage at −20 °C for 2 days resulted in small white
Method C. A solution of N-methylaniline (680 mg, 6.4 mmol) in
CDCl3 (15 mL) was slowly added to a solution of substrate 6e (729 mg,
3.0 mmol) in CDCl3 (15 mL) at 5 °C. After stirring for 5 min, the
solution was washed with 1 N aqueous HCl (3 × 30 mL) and brine
(30 mL), dried (MgSO4), and concentrated in vacuo to give an off-white
solid (774 mg, 2.46 mmol, 82%), mp 75−77 °C, with 1H and 13C NMR
1
data matching 22. However, after 2 days of storage at 4 °C, H NMR
1
indicated that the solid had mostly (∼80%) decomposed to 20
[remainder untransformed 22].
needles (183 mg, 73%), mp 82−84 °C. H NMR (300 MHz): δ 4.30
(q, 2 H), 3.91 (s, 3 H), 1.34 (t, 3 H). 13C NMR (75 MHz): δ 167.6,
165.8, 152.6, 63.2, 55.8, 14.0. HRMS (ESI) m/z [M + Na+] calcd for
C6H9NO5S2: 261.9814. Found: 261.9819. IR (CH2Cl2): 3054 (s), 2986
(s), 1792 (w), 1747 (m), 1706 (m) cm−1. The structure of the title
compound was proved unambiguously by single crystal X-ray analysis
(see Figure 1).
(N-Ethoxycarbonyl)carbamoyl Chloride (8e). For spectroscopic
characterization, a solution of the title compound was generated in situ
by gently bubbling HCl through a 0.8 M solution of (N-ethoxy-
1
carbonyl)isocyanate (9e) in CDCl3 for 10 min. H NMR (CDCl3): δ
4.27 (q, 2 H), 1.30 (t, 3 H); 13C NMR (CDCl3): δ 149.2, 143.2, 63.6,
13.9; IR (CDCl3): 3420 (w), 2987 (w), 1822 (s), 1753 (m),
1481 (s), 1231 (m), 1142 (m) cm−1.
Method B. A solution of (methoxycarbonyl)sulfenyl chloride (17)27
(0.45 mL, 5.0 mmol) in CH2Cl2 (10 mL) was added slowly to a chilled
and stirred solution of 2e (885 mg, 5.0 mmol) in CH2Cl2 (10 mL).
After 10 min at 5 °C and an additional 2 h at 25 °C, the homogeneous
reaction mixture was concentrated in vacuo to give a yellow oil (1.39 g,
quantitative) that crystallized partially upon standing. Further storage
under hexanes (15 mL) at −20 °C provided the title product as a white
solid (1.06 g, 89%), mp 76−78 °C, with 1H and 13C NMR data identical
to those of the material prepared by method A. Anal. Calcd for
C6H9NO5S2 (mol wt 239.27): C, 30.15; H, 3.79; N, 5.86; S, 26.75.
Found: C, 30.84; H, 3.84; N, 5.86; S, 26.76 (this result required shipping
the sample for elemental analysis when packed in dry ice).
(N-Ethoxycarbonyl)isocyanate (9e). Modifying a procedure due to
Lamon,31 oxalyl chloride (8.6 mL, 100 mmol) was added rapidly to a
solution of O-ethyl carbamate (15) (6.5 g, 74 mmol) in CHCl3 (40 mL).
The homogeneous reaction mixture was refluxed overnight, then
brought to 5 °C for 15 min, and filtered to remove a solid byproduct
(1.5 g) that had formed upon cooling. Concentration in vacuo gave a
clear liquid which was purified further by distillation, bp 61−65 °C
(100 mm) [lit.31 bp 54−60 °C (80 mm)]. Yield: 3.8 g (45%). 1H NMR
(CDCl3): δ 4.25 (q, 2H), 1.33 (t, 3H). 13C NMR (CDCl3): δ 148.9,
129.8, 64.7, 13.5. IR (CDCl3): 2942 (w), 2254 (s), 1745 (s), 1430 (m),
1222 (s) cm−1.
The title compound, regardless of the method by which it was
prepared, was not entirely stable even when kept cold. For example, a
solid sample kept at 5 °C for 6 months, when re-examined by 1H NMR,
had decomposed partially to comprise a 2:1 ratio of unchanged starting
16 and urethane 19.
[1-Ethoxy-(N-ethoxycarbonyl)formimidoyl](N′-methyl-N′-
phenylcarbamoyl)disulfane (13). The viscous yellow oil 4e (1.46 g,
6.0 mmol), obtained as already described, was dissolved in CHCl3
(30 mL) and treated with a solution of N-methylaniline (2.2 mL,
20 mmol) in CHCl3 (20 mL) at 5 °C. External cooling was removed, and
after 2 h reaction at 25 °C, the organic layer was washed with 1 N
aqueous HCl (50 mL) and water (50 mL), dried (MgSO4), and
Base-Catalyzed Decomposition of (Methoxycarbonyl)(N-
ethoxycarbonylcarbamoyl)disulfane (16). Pyridine (7 μL, 0.09 mmol)
was added to a solution of 16 (18 mg, 0.08 mmol) in CDCl3 (∼0.8 mL).
1
1
concentrated to provide H NMR pure 13 as an oil (1.4 g, 68%) that
Monitoring by H NMR revealed that the substrate transformed to a
solidified under hexanes at −20 °C. Crystallization by dissolving in
minimal CHCl3 under ambient conditions, and then adding a layer of
hexanes, provided after cooling to −20 °C, off-white prisms (0.68 g,
33%), mp 75−77 °C. 1H NMR (300 MHz): δ 7.0−7.5 (m, 5 H), 4.46
(q, 2 H), 4.26 (q, 2 H), 3.37 (s, 3 H), 1.36 (t, 3 H), 1.35 (t, 3 H). 13C
NMR (75 MHz): δ 171.0, 167.7, 160.8, 140.9, 129.7, 129.0, 128.5, 68.5,
62.8, 39.3, 14.1, 13.7. HRMS (ESI): m/z [M + Na+] calcd for
C14H18N2O4S2: 365.0600. Found: 365.0595. IR (CH2Cl2): 3018 (vs),
2987 (m), 2939 (w), 1803 (w), 1730 (w), 1679 (s) cm−1. Methane CIMS
(source 160 °C, solid probe 120 °C, 0.1 mm): m/z 343 [(M + 1)+, 14%],
mixture of (N-methoxycarbonyl)urethane (19) and O-ethyl carbamate
(15) (along with methanol in an amount equivalent to 15) in a 4:1 ratio
with t1/2 ∼ 40 min (end point within 4 h). When the same procedure was
repeated using Et3N (13 μL, 0.09 mmol) as the base, results were similar,
but the reaction had reached completion by the time the first 1H NMR
time point was taken (∼5 min). It is important to use bases that were
dried over 4 Å molecular sieves; otherwise, the amount of 15 observed is
2- to 3-fold higher than in the described experiment.
Thermolysis of (Methoxycarbonyl)(N-ethoxycarbonylcarbamoyl)-
disulfane (16). Substrate 16 (220 mg, 0.92 mmol) was heated at 100 °C,
G
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
i.e., above its melting point, in a sealed culture tube for 20 min. Next, the
tube was cooled, and vented to allow gases to escape. A portion of the
resultant yellow solid material was dissolved in CDCl3. 1H and 13C NMR
shifts, as well as the observation of an insoluble yellow solid, were con-
sistent with quantitative transformation of starting 16 to (N-methoxy-
carbonyl)urethane (19), along with elemental sulfur.
In contrast to the above results, a solution (∼0.1 M) of 16 in CDCl3
was heated to 60 °C for 2 weeks. 1H NMR analysis revealed that 16 did
not decompose at all over this time span.
[1-Ethoxy-(N-ethoxycarbonyl)formimidoyl](methoxycarbonyl)-
disulfane (18). A solution of (methoxycarbonyl)sulfenyl chloride
(17)27 (0.45 mL, 5.0 mmol) in CH2Cl2 was added slowly to a chilled
(5 °C) and stirred solution of 2e (885 mg, 5.0 mmol) plus pyridine
(0.40 mL, 5.0 mmol) in CH2Cl2 (10 mL). After an additional 10 min at
5 °C and 2 h at 25 °C, the reaction mixture was diluted with CH2Cl2
(10 mL), and washed with water (3 × 15 mL), and the organic phase was
dried (MgSO4) and concentrated in vacuo to provide an orange oil
(1.17 g, 87%). 1H NMR (300 MHz): δ 4.47 (q, 2 H), 4.25 (q, 2 H), 3.92
(s), 1.36 (t, 3 H), 1.35 (t, 3 H). 13C NMR (75 MHz): δ 170.3, 167.3,
160.3, 68.8, 62.8, 55.7, 14.1, 13.6. HRMS (ESI): m/z [M + Na+] calcd
for C8H13NO5S2: 290.0127. Found: 290.0110. IR (CH2Cl2) 3018 (vs),
2986 (m), 2957 (w), 1757 (m), 1738 (s), 1680 (s) cm−1.
(mol wt 314.38): C, 45.88; H, 4.49; N, 8.91; S, 20.36. Found: C, 45.60; H,
4.63; N, 8.95; S, 20.12. The structure of 22 was further proved by single
crystal X-ray analysis (see Figure 1).
Similar to the case with mixed carbamoyl disulfane 16, the title
compound 22 did not have long-term storage stability. Thus, after 6
months in the solid state at 5 °C, a sample that had previously been pure
22 had decomposed partially to a mixture of 22 and 20, in a ratio of ∼1:1.
Treatment of (N-Ethoxycarbonylcarbamoyl)(N′-methyl-N′-phe-
nylcarbamoyl) disulfane (22) with N-Methylaniline. Neat N-methyl-
aniline (6 μL, 0.06 mmol) was added to a solution of 22 (16 mg,
0.05 mmol) in CDCl3 (∼0.8 mL). 1H NMR revealed the conversion of
22 to carbamoyl urea 20 (t1/2 ∼ 30 min); starting 22 was completely
absent after 3.5 h.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b01826.
Crystallographic data (CIF)
Table of X-ray crystallographic and geometric parameters;
table of diagnostic 1H NMR shifts for important
compounds; a comparison of 1H NMR and IR spectra of
deuterated and non-deuterated 15 formed upon hydrol-
When Et3N (0.70 mL, 5.0 mmol) was used as the base in place of
pyridine, similar yields and purities were noted.
(N-Methoxycarbonyl)urethane (19). Neat (N-ethoxycarbonyl)-
isocyanate (9e) (1.0 g, 8.7 mmol) was added to MeOH (10 mL). After
30 min, the homogeneous reaction mixture was concentrated in vacuo to
give the title product as a white crystalline solid (1.26 g, 98%), mp 65−
68 °C (lit.36 mp 68−73 °C). 1H NMR (300 MHz): δ 4.25 (q, 3 H), 3.80
(s, 3 H), 1.31 (t, 6 H). 13C NMR (75 MHz): δ 151.4, 150.7, 62.4, 53.1, 14.2.
HRMS (ESI): m/z [M + Na+] calcd for C5H9NO4: 170.0424. Found:
170.0440. IR (CDCl3) 2985 (m), 2960 (w), 1805 (vs), 1732 (s), 1505 (s) cm−1.
(N-Ethoxycarbonyl)(N′-methyl-N′-phenyl)urea (20). A solution of
freshly prepared (N-ethoxycarbonyl)isocyanate (9e) (1.5 g, 13 mmol)
in CDCl3 (2 mL) was added over 5 min to a solution of N-methylaniline
(2.3 mL, 2.3 g, 21 mmol) in CDCl3 (10 mL), with no appreciable
spontaneous exotherm. While the reaction was fast (complete after
10 min based on 1H NMR examination), the homogeneous mixture was
maintained for 2 h, and then washed with 1 N aqueous HCl (3 × 20 mL)
and brine (20 mL). The organic phase was dried (MgSO4), concen-
trated, and placed under hexanes at −20 °C to provide a solid product
(1.8 g, 62%). A portion of the crude product (0.50 g) was recrystallized
from hot hexanes (6 mL) to provide, after cooling to −20 °C, white
needles (88% recovery), mp 65−67 °C. 1H NMR (300 MHz): δ 7.2−7.5
(m, 5H), 4.16 (q, 2 H), 3.30 (s, 3H), 1.24 (t, 3 H). 13C NMR (75 MHz):
δ 151.2, 150.6, 141.6, 130.5, 128.5, 127.1, 61.9, 37.5, 14.2. HRMS (ESI):
m/z [M + Na+] calcd for C11H14N2O3: 245.0897. Found: 245.0881. IR
(CH2Cl2): 2982 (vs), 2939 (m), 1779 (vs), 1682 (vs) cm−1; Anal. Calcd
for C11H14N2O3 (mol wt 222.24): C, 59.45; H, 6.35; N, 12.60. Found: C,
59.16; H, 6.34; N, 12.67.
1
ysis of 6e; H and 13C NMR spectra of new compounds
(PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: barany@umn.edu.
■
Present Addresses
‡(L.C.) CordenPharma, Boulder, Colorado 80301, United States.
§(R.P.H.) Ra Pharmaceuticals, Cambridge, Massachusetts
02139, United States.
∥(M.J.H.) Program in Chemical Biology, University of Michigan,
Ann Arbor, Michigan 48109, United States.
⊥(A.M.S.) Department of Chemical Engineering, University of
California, Santa Barbara, California 93106, United States.
Notes
The authors declare no competing financial interest.
†(D.B.) Deceased (July 24, 2015).
ACKNOWLEDGMENTS
■
We thank Charles S. Barrett, William W. Brennessel, David K.
Ford, Erik S. Goebel, Phillip T. Goldblatt, David A. Halsrud,
Michael C. Hanson, Azra Kovacevic, Isaac D. Mitchell, Matthew J.
Turcotte, and Robert L. Walsky for syntheses of starting materials
or other experimental contributions, Jed Fisher for critical reading
and discussion, and Charles S. Barrett, Christie L. Martin, Kara A.
Meyers, and David M. Ungs for help with various aspects of
manuscript preparation. National Science Foundation (NSF)
MRI grant 1229400 provided funds for a Bruker PHOTON-100
diffractometer that was used to obtain some of the crystal
structures presented in this work. Research in the Barany lab has
been supported over the years by the Searle Scholars program and
the National Institutes of Health (NIH) [GM 28934 and 42722].
Fellowships to MJH and AMS were provided by the Heisig/
Gleysteen Summer Research Program, and by a research fund
(N-Ethoxycarbonylcarbamoyl)(N′-methyl-N′-phenylcarbamoyl)-
disulfane (22). A solution of thiocarbamate 2e (885 mg, 5.0 mmol) in
CDCl3 (10 mL) was added dropwise over 5 min to a solution of SO2Cl2
(0.70 mg, 5.2 mmol) in CDCl3 (5 mL) at 5 °C. The reaction mixture was
next stirred for 2 h at 25 °C, and then O-isopropyl-N-methyl-N-
phenylthiocarbamate23 (21) (1.05 g, 5.0 mmol) in CDCl3 (5 mL) was
added over 1 min. The solution turned green immediately, and then
became yellow over time. After overnight stirring, the mixture was
1
concentrated in vacuo to give a reasonably H NMR pure yellow solid
(1.42 g, 90%), which was recrystallized from hot CHCl3 (5 mL) to give
a white solid (239 mg, 15% recovery), mp 102−103 °C. The residue
from the mother liquor was taken up in hot CHCl3 (4 mL), hexanes
(2 mL) were added, and a second crop (346 mg, 22% recovery),
a light yellow solid, mp 90−93 °C, was obtained. 1H NMR (300 MHz):
δ 7.4−7.5 (m, 5H), 4.28 (q, 2H), 3.38 (s, 3H), 1.33 (t, 3H).
13C NMR (75 MHz): δ 166.2, 163.7, 152.0, 140.8, 129.9, 129.3, 128.7,
128.6, 63.2, 39.5, 14.2. HRMS (ESI): m/z [M + Na+] calcd for
C12H14N2O4S2: 337.0287. Found: 337.0286. IR (CDCl3): 3390 (w),
2989 (w), 1787 (w), 1751 (s), 1685 (s), 1595 (w), 1495 (s), 1481 (m),
1348 (w), 1271 (w), 1202 (s) cm−1; Anal. Calcd for C12H14N2O4S2
dedicated to the memories of Kate and Michael Bar
́ ́
any.
DEDICATION
■
́ ́
Dedicated to Professors Michael Barany (October 29, 1921−July
24, 2011) and Doyle Britton (March 6, 1930−July 7, 2015),
H
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(30) Harris, J. F., Jr. J. Am. Chem. Soc. 1960, 82, 155−158. This paper
provides important chemical precedents, and was the point of departure
for some of the nomenclature used in the present report (albeit,
modernized somewhat, e.g., “disulfide” to “disulfane”).
(31) The product observed in our work had spectroscopic properties
matching those of authentic (N-ethoxycarbonyl)isocyanate prepared
according to: Lamon, R. W. J. Heterocycl. Chem. 1969, 6, 261−264.
More recently, this compound can be obtained commercially, albeit at a
relatively high price and only 76% purity (90% purity claimed by
manufacturer).
(32) To the best of our knowledge (see references that follow), acyl
carbamoyl chloride 8e described herein is a new compound; diagnostic
peaks of 8e arising from thermolysis of 6e overlapped with diagnostic
peaks for 8e generated in situ in CDCl3 solution by addition of HCl to
acyl isocyanate 9e. The closest precedent to the latter reaction was
reported by: Akteries, B.; Jochims, J. C. Chem. Ber. 1986, 119, 669−682.
(33) A simpler precedent involves preparation of N-methylcarbamoyl
chloride by addition of HCl to methyl isocyanate: D'Silva, T. D. J.;
Lopes, A.; Jones, R. L.; Singhawangcha, S.; Chan, J. K. J. Org. Chem.
1986, 51, 3781−3788 and references cited therein.
(34) Acyl carbamoyl chloride 8e was implied, but not characterized
spectroscopically, from the reaction of (N-chlorocarbonyl)isocyanate
with EtOH, as included in a review by: Hagemann, H. Angew. Chem., Int.
Ed. Engl. 1977, 16, 743−750.
(35) Acyl carbamoyl chloride 8e was also implied from the reaction of
O-ethylcarbamate (15) with phosgene, as described in: Folin, O. Am.
Chem. J. 1897, 19, 323−352.
(36) N-Methoxycarbonyl)urethane (19) was first made by two
independent methods, either by forming the sodium salt of O-
ethylcarbamate (15), which was then acylated with methyl chlor-
oformate, or alternatively, by acylating the sodium salt of bis(N-
methoxycarbonyl)amine with ethyl chloroformate, as described by:
Diels, O.; Nawiasky, P. Ber. Dtsch. Chem. Ges. 1904, 37, 3672−3683.
(37) Carbamoyl urea 20 described herein is a new compound. For the
preparation of other carbamoyl ureas by reactions of acyl isocyanates
with N-methylaniline, see: Schweim, H. Arch. Pharm. 1987, 320, 430−
437.
whose enthusiasm for science throughout their lives was an
inspiration to all of us.
REFERENCES
■
(1) Zumach, G.; Weiss, W.; Kuhle, E. Belgian Patent 682,991 (June 23,
1966); British Patent 1,136,737 (June 21, 1966). Chem. Abst. 70, 1969,
̈
77951p.
(2) Review: Zumach, G.; Kuhle, E. Angew. Chem., Int. Ed. Engl. 1970, 9,
̈
54−63.
(3) Monograph: Kuhle, E. The Chemistry of the Sulfenic Acids; Georg
̈
Thieme: Stuttgart, 1973; see especially pp 68−70 and 107−108.
(4) Barany, G. Ph.D. Thesis, The Rockefeller University, Dissertation
Abstracts, 1977, 38, 5893-B.
(5) Barany, G.; Merrifield, R. B. J. Am. Chem. Soc. 1977, 99, 7363−
7365.
(6) Słomczynska, U.; Barany, G. J. Heterocycl. Chem. 1984, 21, 241−
̃
246.
(7) Zalipsky, S.; Albericio, F.; Słomczynska, U.; Barany, G. Int. J. Pept.
̃
Protein Res. 1987, 30, 740−783 and references cited therein.
(8) Hammer, R. P.; Albericio, F.; Gera, L.; Barany, G. Int. J. Pept. Protein
Res. 1990, 36, 31−45.
(9) Chen, L.; Thompson, T. R.; Hammer, R. P.; Barany, G. J. Org.
Chem. 1996, 61, 6639−6645 and references cited therein. This paper
also reports the X-ray crystallographic structures of 1,2,4-dithiazolidine-
3,5-dione (1a) and 3-ethoxy-1,2−4-dithiazoline-5-one (11).
(10) Barany, M. J.; Hammer, R. P.; Merrifield, R. B.; Barany, G. J. Am.
Chem. Soc. 2005, 127, 508−509 and references cited therein.
(11) Barany, G.; Albericio, F. J. Am. Chem. Soc. 1985, 107, 4936−4942.
(12) Albericio, F.; Barany, G. Int. J. Pept. Protein Res. 1987, 30, 177−
205.
(13) Jensen, K. J.; Hansen, P. R.; Venugopal, D.; Barany, G. J. Am.
Chem. Soc. 1996, 118, 3148−3155.
(14) Xu, Q.; Musier-Forsyth, K.; Hammer, R. P.; Barany, G. Nucleic
Acids Res. 1996, 24, 1602−1607.
(15) Chen, L.; Barany, G. Lett. Pept. Sci. 1996, 3, 283−292.
(16) Planas, M.; Bardaji, E.; Jensen, K. J.; Barany, G. J. Org. Chem. 1999,
64, 7281−7289.
(38) Henley, M. J.; Schrader, A. M.; Young, V. G.; Barany, G. Acta
Cryst. Sect. E 2015, 71, o782−o783.
(17) Wood, M. E.; Cane-Honeysett, D. J.; Dowle, M. D.; Coles, S. J.;
Hursthouse, M. B. Org. Biomol. Chem. 2003, 1, 3015−3023.
(18) Wood, M. E. 1,2,4-Dithiazolidine-3,5-dione. e-EROS Encyclopedia
of Reagents for Organic Synthesis 2007, and references cited therein
10.1002/9780470842898.rn00717.
(39) Goldenberg, B. L.; Young, V. G., Jr.; Barany, G. Acta Cryst. Sect. E
2015, 71, 1169−1173.
(40) Nα-Dithiasuccinoyl (Dts)-L-phenylalanine: Barany, G. Cryst.
Struct. Commun. 1982, 11, 913−928.
(19) Barany, G.; Słomczynska, U.; Mott, A. W. Abstracts of Papers,
̃
(41) 4,4′-p-Xylenebis(1,2,4-dithiazolidine-3,5-dione): Blencowe, A.;
Clarke, A.; Drew, M. G. B.; Hayes, W.; Slark, A.; Woodward, P. React.
Funct. Polym. 2006, 66, 1284−1295.
187th National Meeting of the American Chemical Society, St. Louis, MO,
April 8−13, 1984; American Chemical Society: Washington, DC, 1984;
ORG 32.
(42) 4-(3-Nitrophenyl)-1,2,4-dithiazolidine-3,5-dione: Ponomarov,
O.; Padelkova, Z.; Hanusek, J. J. Phys. Org. Chem. 2013, 26, 560−564.
(43) Barany, G.; Mott, A. W.; Larka, E. A. In Proceedings of 32nd Annual
Conference on Mass Spectrometry and Allied Topics, San Antonio, TX, May
27−June 1, 1984; 1984, pp 481−482.
(20) Barany, M. J.; Corey, M. M.; Hanson, M. C.; Majerle, R. S.;
Hammer, R. P.; Barany, G. In Understanding Biology Using Peptides,
Proceedings of the 19th American Peptide Symposium, San Diego, CA, USA,
June 18−23, 2005; Blondelle, S. E., Ed.; Springer: New York, 2006; pp
196−197.
(44) Henley, M. J.; Barany, M. J.; Chen, L.; Hammer, R. P.; Schrader, A.
M.; Young, V. G.; Barany, G. In Proceedings of the 24th American Peptide
Symposium, Orlando, FL, USA, June 20−25, 2015; Lebl, M., Srivastava,
V., Yudin, A., Eds.; Prompt Scientific Publishing: San Diego, CA, 2015;
pp 266−267.
(45) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008,
64, 112−122.
(46) The present paper provides two mechanistically related methods
to produce thiocarbamate 2e, essentially by addition of EtOH to
EtO(CO)NCS (12). Compound 2e can also be created by
unrelated chemistry, i.e., reaction of EtO(CO)C(Cl)NOH with
thiourea in EtOH, as described by: Dornow, A.; Fischer, K. Chem. Ber.
1966, 99, 72−80.
(47) Wheeler, H. L.; Sanders, W. M. J. Am. Chem. Soc. 1900, 22, 365−
378.
(48) The Merck Index, 12th ed.; Merck & Co.: Whitehouse Station, NJ,
(21) Gessner, R. K.; Chibale, K. Synlett 2009, 17, 2839−2843.
(22) Martinez, A.; Alonso, M.; Castro, A.; Dorronsoro, I.; Gelpí, J. L.;
́
Luque, F. J.; Perez, C.; Moreno, F. J. J. Med. Chem. 2005, 48, 7103−7112.
(23) Schrader, A. M.; Schroll, A. L.; Barany, G. J. Org. Chem. 2011, 76,
7882−7892.
(24) Nefkens, G. H. L.; Tesser, G. I.; Nivard, R. J. F. Rec. Trav. Chim.
Pays-Bays 1960, 79, 688−698.
(25) For related chemistry to form N-maleoyl derivatives, see: Keller,
O.; Rudinger, J. Helv. Chim. Acta 1975, 58, 531−541.
(26) Delitsch, G. J. Prakt. Chem. 1874, 10, 116−128.
(27) Barany, G.; Schroll, A. L.; Mott, A. W.; Halsrud, D. A. J. Org. Chem.
1983, 48, 4750−4761 and references cited therein.
(28) Schroll, A. L.; Barany, G. J. Org. Chem. 1986, 51, 1866−1881.
(29) The product observed in our work had spectroscopic properties
matching those of an authentic sample prepared according to:
Tompkins, L. G.; Degering, E. F. J. Am. Chem. Soc. 1947, 69, 2616−
2618 as well as by an alternative method developed in this work (see
Scheme 3, note a).
1996; entry 10013.
I
DOI: 10.1021/acs.joc.5b01826
J. Org. Chem. XXXX, XXX, XXX−XXX