The synthetic challenge of thioglycolurils

Article abstract of DOI:10.1002/ejoc.201301672

The very high stability of cucurbiturils under harsh acidic conditions and their reported prolific chemistry over more than three decades pose a question: why has no thiocucurbituril been reported to date? Furthermore, although glycoluril is a highly stable, easily accessible precursor of all cucurbiturils, its sulfur analog represents a yet unmet synthetic challenge. The reaction between glyoxal and thiourea was found to stop at the level of dihydroxyimidazolidine-2-thione, which is quite unstable under various acidic conditions. In an attempt to answer these questions, several stable analogs of thioglycoluril, that is, monothioglycoluril, ditolylthioglycoluril, and its diether derivative were prepared and characterized in the hope that they could be employed as building blocks for the synthesis of thiocucurbiturils. Several side products were also obtained that highlight the complex reactivity of thiourea in these reactions. The crystal structures of the above-mentioned thioglycolurils are dominated by networks of hydrogen-bonding interactions. Attempts to co-oligomerize these compounds with formaldehyde clearly suggest that it is impossible to synthesize thiocucurbiturils by the methods commonly used for the preparation of cucurbiturils. Given that thiocucurbiturils are expected to be stable molecules, alternative synthetic strategies that are different from the thermodynamically controlled approaches must be designed. Although glycoluril is a highly stable, easily accessible precursor of all cucurbiturils, dithioglycoluril is still unknown. To answer the question why thiocucurbituril has never been made, we prepared several thioglycoluril derivatives and found that they do not survive the acidic conditions usually needed for the co-oligomerization reaction with paraformaldehyde. Copyright

Full text of DOI:10.1002/ejoc.201301672

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
FULL PAPER
DOI: 10.1002/ejoc.201301672
The Synthetic Challenge of Thioglycolurils
Mandeep Singh,[a] Galit Parvari,[b] Mark Botoshansky,[b][‡] Ehud Keinan,*[b] and
Ofer Reany*[a]
Keywords: Building blocks / Dicarbonyls / Hydantoin / Macrocycles / Sulfur
The very high stability of cucurbiturils under harsh acidic
conditions and their reported prolific chemistry over more
than three decades pose a question: why has no thiocucurbit-
uril been reported to date? Furthermore, although glycoluril
is a highly stable, easily accessible precursor of all cucur-
biturils, its sulfur analog represents a yet unmet synthetic
challenge. The reaction between glyoxal and thiourea was
found to stop at the level of dihydroxyimidazolidine-2-thione,
which is quite unstable under various acidic conditions. In
an attempt to answer these questions, several stable analogs
of thioglycoluril, that is, monothioglycoluril, ditolylthioglycol-
uril, and its diether derivative were prepared and charac-
terized in the hope that they could be employed as building
blocks for the synthesis of thiocucurbiturils. Several side
products were also obtained that highlight the complex reac-
tivity of thiourea in these reactions. The crystal structures of
the above-mentioned thioglycolurils are dominated by net-
works of hydrogen-bonding interactions. Attempts to co-
oligomerize these compounds with formaldehyde clearly
suggest that it is impossible to synthesize thiocucurbiturils by
the methods commonly used for the preparation of cucur-
biturils. Given that thiocucurbiturils are expected to be stable
molecules, alternative synthetic strategies that are different
from the thermodynamically controlled approaches must be
designed.
oxidation states.[3,4] Thiacalixarenes in which the methylene
bridge of calixarenes is replaced by a sulfur atom or sulfox-
ide and sulfone derivatives can function as useful ion sen-
sors.[5] Calixarenes bearing thiophenols rather than phenols
offer unique conformational preferences and specific bind-
ing to metal ions.[6] Cyclodextrins with thiols at the 6-posi-
tion form self-assembled monolayers on gold surfaces for
various electrochemical applications.[7]
Introduction
Cucurbiturils have been attracting increasing attention
because these exceptional host molecules exhibit useful
binding properties and unique structures.[1] Their rich sup-
ramolecular chemistry and specific binding interactions to
various organic guests and metal ions have led to a broad
variety of applications, including chemical sensing, cataly-
sis, and the ability to function as essential components of
molecular architectures.[2] On these grounds, it is quite sur-
prising that no attempt to prepare thiocucurbiturils has
been reported within more than three decades of intense
research on the chemistry of cucurbiturils. The incorpora-
tion of sulfur atoms within the framework of other macro-
cyclic host molecules has been quite rewarding. For exam-
ple, thiacrown ethers exhibit moderate π acidity that has
allowed them to coordinate transition-metal ions of low
Aside from theoretical studies by Pichierri,[8,9] which pre-
dicted interesting ligation to metal ions and metal surfaces,
there has been only circumstantial evidence on attempts to
prepare thiocucurbiturils. For example, shortly after his dis-
covery of the cucurbituril structure, Mock reported the
preparation of monothioglycoluril.[10] Other attempts to
prepare thioglycolurils probably reflect similar inten-
tions.[11]
This situation suggests that although thiocucurbiturils
are expected to be thermodynamically stable, their synthesis
represents a significant challenge. Herein, we address this
problem and delineate the reasons why it is difficult to syn-
thesize thiocucurbiturils by using the same chemistry that
has proven so useful for the preparation of cucurbiturils
(Scheme 1).
[a] Avinoam Adam Department of Natural Sciences,
The Open University of Israel,
1 University Road, P.O. Box 808, Ra’anana 43537, Israel
E-mail: oferre@openu.ac.il
http://telem.openu.ac.il/nature
[b] Schulich Faculty of Chemistry,
Technion – Israel Institute of Technology, Technion City,
Haifa 32000, Israel
E-mail: keinan@technion.ac.il
http://schulich.technion.ac.il/Ehud_Keinan.htm
[‡] This article is dedicated to Dr. Mark Botoshansky, a devoted Results and Discussion
unabated crystallographer, whose life came to an abrupt end by
an incurable illness shortly before the manuscript was submit-
Considering potential synthetic approaches to thiocucur-
biturils, one could contemplate either direct thionation of
cucurbiturils or acid-catalyzed condensation of thioglycolu-
ted.
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201301672.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
M. Singh, G. Parvari, M. Botoshansky, E. Keinan, O. Reany
FULL PAPER
1
and trans-5 and determined their ratio by H NMR spec-
troscopy ([D6]DMSO) on the basis of their resonances: δ=
8.56/8.83 (NH), 5.81/6.26 (OH), 5.03/4.73 ppm (CH),
respectively. The corresponding 13C NMR spectrum exhib-
ited the cis/trans isomers at δ = 181.6/182.3, 80.0/87.4 ppm,
respectively, in agreement with the generally observed spec-
tral manifestation of internal gauche and anti O–C–C–O
arrangements.[21]
Scheme 1. General synthesis of cucurbiturils and their sulfur ana-
logs. The hexameric CB[6] represents the entire family of CB[n] =
cucurbit[n]uril.
ril 2 or 3 with paraformaldehyde (Scheme 1). Although the
Given that 4,5-dihydroxyimidazolidone, the oxygen ana-
first approach seems quite attractive, it seems to be imprac- log of 5, is a known intermediate in the synthesis of 1 from
tical because of the insolubility of cucurbiturils in organic glyoxal and urea,[22] we attempted to prepare 2 by treating
solvents that are essential for reactions with a phosphetane 5 with thiourea in aqueous hydrochloric acid. However, all
agent (Lawesson’s reagent).[12] Furthermore, it has already reactions of 5 with either thiourea or urea at various pH
been reported that direct thionation of urea derivatives, in- values and different reaction times and temperatures re-
cluding glycolurils, can work only if all of the nitrogen sulted in the above-described product mixture that was usu-
atoms are fully substituted. Otherwise, the reaction pro- ally obtained under harsh acidic conditions. These observa-
ceeds through C–N bond cleavage to produce isothiocya- tions, as well as previous reports on the reactions of thio-
nates or aminoisothiocyanates.[13,14] Indeed, these findings urea and benzil,[23] indicate that 5 is not a suitable precursor
were confirmed by our preliminary attempts.
of 2.
In an attempt to control the reactivity of 5, we converted
The alternative approach, which is based on the classical
employment of glycolurils, depends on the accessibility, sta- it into corresponding diether 6.[24] Thus, 5 was treated with
bility, and reactivity of 2 and 3. Whereas the condensation slightly acidic methanol to form trans-6 in the form of a
reaction of urea with α-dicarbonyl compounds proceeds single stereoisomer (Scheme 2), as was evident by both
quantitatively, the analogous reaction with thiourea is not 1H NMR and 13C NMR spectroscopy. All attempts to react
trivial. For example, the reaction between thiourea and gly- 6 with thiourea under aqueous acidic conditions resulted
oxal in aqueous HCl at 75 °C leads to a complex mixture in the same mixture that was obtained from 5 under these
of products, including 1,4-diaza-3,6-dithiabicyclo[3.3.0]oc- conditions. Nevertheless, partial success was achieved in the
tane-2,5-diimine (4),[15,16] thiohydantoin,[17] urea, 2-thioimi- reaction of 6 with urea instead of thiourea, which led to
dazole, and elemental sulfur (Table 1, entry 6). The reported monothioglycoluril, 3 in 45% yield[25] along with Ͻ5% of
formation of 2 under these conditions was found to be in- glycoluril (1, Table 2).
correct.[18] Yet, one analog of 2, that is, cyclohexanothio-
The observed acid-catalyzed conversion of 6 into
glycoluril, was obtained in 26% yield from cyclohexane-1,2- glycolurils 3 and 1 at a ratio of 10:1 reflects a multistep
dione and thiourea under milder acidic conditions, such as mechanism with a series of protonation/deprotonation of
trifluoroacetic acid in toluene.[11]
heteroatoms, departure of various leaving groups, forma-
We investigated the reaction between thiourea and gly- tion of heteroatom-stabilized carbocations, and nucleophilic
oxal under milder conditions (Table 1), including ultrasonic attacks on these intermediates.[26,27] Given that all of these
irradiation, which was proven useful for the synthesis of reactions are reversible, one would assume that the product
glycoluril and hydantoin derivatives.[19] As can be seen from distribution reflects thermodynamic control. However, the
Table 1, 5 was formed exclusively as a cis/trans mixture with fact that 1, 3, and 6 have limited solubility in acidic water
no evidence for the formation of 2. Although only trans-5 leads to the consequence that the reaction is, at least par-
was previously characterized,[20] we identified both cis-5 tially, kinetically controlled. Considering the pKa values of
Table 1. Reaction between thiourea and glyoxal; all reactions were performed for 2 h.
Entry
Solvent
T [°C]
Thiourea/glyoxal
Catalyst
Yield [%] of 5 (cis/trans)[a]
1[b]
2[b]
3[b]
4
5
6
EtOH
EtOH
EtOH/H2O
Water
Water
55
r.t.
55
65
r.t.
80
2:1
2:1
2:1
2:1
2:1
2–6:1
KOH (2 equiv.)
NaOH (0.5 equiv.)
acetic acid (pH 4)
none
none
HCl (0.5–8 m)
Ͼ95 (15:85)
65 (10:90)
Ͼ95 (15:85)
90 (35:65)
85 (5:95)
trace
Water
[a] Yield of isolated product. [b] Under ultrasonic irradiation (40 kHz).
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
The Synthetic Challenge of Thioglycolurils
the reagents involved in this reaction, their expected order
of nucleophilicity would be: methanolϾ ureaϾ thiourea.
Accordingly, the order of their performance as a good leav-
ing group upon protonation would be thioureaϾ ureaϾ
methanol. For example, intermediate VIII is expected to be
more stable than intermediate IX, which implies that IX is
more reactive than VIII (Scheme 3).
Thus, 6 is susceptible to protonation on either the meth-
oxy group or the thiourea fragment to produce intermediate
I or VI, respectively (Scheme 3). Departure of methanol
from I and attack by urea on the resultant iminium cation
would lead to intermediates II and III. The latter can un-
dergo reversible proton transfer to form IV, which upon
loss of a second molecule of methanol will produce V, a
clear precursor of 3. Similarly, intermediate VI could un-
dergo a similar sequence of transformations along interme-
diates VIII–X, all the way to XI. The latter is a known inter-
mediate on the way to glycoluril (1).[15] Clearly, various
other steps are expected to occur under aqueous acidic con-
ditions, such as the crossover between the upper and lower
pathways in Scheme 3 by the interconversion of intermedi-
ates VIII and IV through intermediates XII and XIII.
The crystal structure of monothioglycoluril 3 (Figure 1)
shows that both rings are nearly planar and that they adopt
an envelope conformation with both the C=S and C=O
groups at the flap position. The crystallographic unit cell
belongs to the monoclinic crystal system with the non-cen-
trosymmetric space group P21 and a single molecule in the
asymmetric unit. The crystal data and structural refinement
results are summarized in Table 3.
Scheme 2.
Table 2. Reaction of 6 with urea under aqueous HCl (0.3%) condi-
tions at 80 °C.
Entry 6/urea
Time [h]
Yield [%]
3[a]
1[b]
1
2
3
4
5
1:1
2
2
5
2
2
10
45
30
40
40
1
4.5
3
4
4
Notably, there is a 50:50 occupancy of oxygen and sulfur
in each of the carbonyl/thiocarbonyl sites, and each mole-
cule exhibits short contacts to the neighboring molecules.
The crystal structure features chains of molecules along the
b axis within the bc plane, and in each chain, the molecules
present their convex face either up or down to form an anti-
1:1.2
1:1.2
1:2
1:10
[a] Yield of isolated product. [b] Determined by 1H NMR spec-
troscopy.
Scheme 3. Protonation of the urea or thiourea fragment could occur on either nitrogen atom, as shown, for example, in intermediate III
or VI, or on their O/S atoms.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
M. Singh, G. Parvari, M. Botoshansky, E. Keinan, O. Reany
FULL PAPER
Figure 1. Crystal structure of 3 showing a layer of antiparallel molecular chains in the bc plane with intermolecular NH···S hydrogen
bonds (red dashed lines), NH···Cl and CH···Cl hydrogen bonds (green dashed lines), and intramolecular C=S···HN and C=O···HN
hydrogen bonds (black dashed lines). The atomic numbering refers to the symmetry operation [x, y, z + 1]. Color code: C, gray; N, blue;
O, red; S, yellow; Cl, green; H, grayish.
Table 3. Crystallographic data and refinement details of 3, 8, and 10.
3
8
10
Formula
C4H6N4SOCl
193.6
2.344
C18H18N4S2·2CH4N2S
506.77
1.305
C22H22N4O2S2
438.58
1.371
Formula weight
Density [gcm–3]
T [K]
293(2)
293(2)
293(2)
Diffractometer
Scan mode
Crystal size [mm]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
APEX2 DUO
ω scans
0.26ϫ 0.20ϫ 0.16
monoclinic
P21
4.188(6)
10.252(2)
7.561(6)
90
KappaCCD
ω and φ scans
0.28ϫ 0.15ϫ 0.10
orthorhombic
Pnma
9.289(2)
16.465(3)
16.864(3)
90
APEX2 DUO
ω scans
0.44ϫ 0.26ϫ 0.19
monoclinic
P21/n
8.638(3)
11.375(4)
21.757(7)
90
β [°]
90.37(4)
90
324.6(5)
2
90
90
2579.2(9)
4
96.27(1)
90
2125.0(13)
4
γ [°]
V [Å3]
Z
F(000)
232
24.97
673
575
1064
25.04
2346
1354
920
25.24
3765
3351
θ max [°]
Unique reflections
IϾ2σ(I)
R factor (all data)
R factor [IϾ2σ(I)]
S
0.0802
0.0752
1.062
0.099
0.051
0.816
0.1165
0.0442
1.008
parallel array of alternating chains. Apparently, the main these two heteroatoms at a short distance (2.26 or 2.08 Å)
driving force for the formation of the antiparallel array of and a long distance (2.83 or 2.96 Å). These unprecedented
chains along the b axis is the complex network of hydrogen contacts between a chloride ion and a carbonyl/thiocar-
bonds, including NH···X (X = O/S), NH···Cl, and CH···Cl, bonyl have not yet been reported in the Cambridge Crystal-
all along the a axis. Remarkably, each chloride ion is en- lographic Data Centre (CCDC). This unusual phenome-
gaged in six strong hydrogen bonds, all of which range be- non, as well as the observed, relatively short carbonyl and
tween 2.65 and 2.80 Å. Every two molecules in the chain thiocarbonyl bonds (1.25/1.46 and 1.49/1.63 Å, respectively)
are connected by a chloride ion through their oxygen/sulfur are currently under investigation in our laboratories.
atoms. Although the structure does not allow for specific
Unfortunately, although 3 was now readily available in
assignment of either oxygen or sulfur on each site, the con- our hands, all attempts to employ it for the synthesis of
necting chloride atom is asymmetrically positioned between thiocucurbiturils (Scheme 1) failed. Under the classical con-
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
The Synthetic Challenge of Thioglycolurils
ditions of cucurbituril synthesis, this compound decom- entries 7 and 8), the ratio was reversed to 85:15 in DMSO
posed to produce the above-mentioned mixture observed and to 93:7 in water.
upon treatment of either 5 or 6 with strong acids.
Ditolyl thioglycoluril (8), which precipitated from the re-
action mixture together with thiourea, was recrystallized
from ethanol to give co-crystals of 8 and thiourea in a 1:2
ratio. As observed with 3, the structure of 8 (Figure 2)
shows that the two five-membered rings assume an envelope
conformation in which both thiocarbonyl groups adopt the
flap position. The crystallographic unit cell belongs to the
orthorhombic crystal system with a space group Pnma
showing half a molecule in the asymmetric unit. The crystal
data and structural refinement results are summarized in
Table 3. Expectedly, the aryl groups exhibit face-to-face
stacking interactions with an angle of 40° between the two
aromatic planes (Figure 2, left). Interestingly, this angle is
smaller than the corresponding angle of 48°, which was ob-
served for diphenyl thioglycoluril.[29] The crystal packing is
dominated by a network of hydrogen-bonding interactions
(Figure 2, right), which generally characterize crystal struc-
tures of thiourea derivatives.[30] Interestingly, the nitrogen
atoms of both thiourea molecules function as both hydro-
gen-bond donors, NH···S, and hydrogen-bond acceptors,
N···HC, with short contacts (2.82 Å) between the nitrogen
atoms and the aromatic hydrogen atoms of 8.
Substituted thioglycoluril 8 was found to be much more
stable than 3 under acidic conditions. For example, 8 sur-
vived in concentrated H2SO4 at 90 °C for 4 h and at room
temperature for 24 h. Nevertheless, no cyclic or acyclic co-
oligomers could be observed in the reaction of 8 with form-
aldehyde under these conditions. Switching from sulfuric
acid to concentrated HCl afforded diether derivative 10 in
82% yield (Scheme 4).
Compound 10, which was crystallized from chloroform
(Figure 3 and Table 3), shows that the two five-membered
rings assume a half-chair conformation. Thus, the two
bridge carbon atoms are significantly withdrawn from the
plane defined by the remaining three ring atoms; the tor-
Considering the limited stability of 2 and 3 under mildly
acidic conditions, we turned our attention to disubstituted
thioglycolurils, because these compounds can be produced
under basic conditions from the corresponding benzil deriv-
atives.[27] Thus, reaction of thiourea with 4,4Ј-dimethylben-
zil at various molar proportions in basic ethanol afforded
7 along with ditolyl thioglycoluril (8; Table 4, entries 1–4).
Remarkably, as long as the reaction and its workup were
performed under basic conditions, the corresponding hy-
dantoin 9 (Scheme 4) was not formed.[28] Furthermore, our
observation that the ratio between 7 and 8 in the product
mixture was independent of the ratio between the two reac-
tants indicates that under the reaction conditions 7 is not a
precursor of 8.[27] Nevertheless, the product ratio was found
to be dependent on the solvent. For example, whereas 7/8
= 40:60 in ethanol, by performing the reaction in either
isopropanol or tert-butyl alcohol, the ratio changed to
25:75 or 12:88, respectively, with negligible changes in the
overall yields. In contrast, by using polar solvents (Table 4,
Table 4. Reaction of thiourea with 4,4Ј-dimethylbenzil. All reac-
tions were performed under basic conditions (KOH, 0.6 equiv.) by
using ultrasonic irradiation (40 kHz, 55 °C, 2 h).
Entry Solvent
Thiourea/4,4Ј-
dimethylbenzil
Yield[a] [%]
7
8
1
2
3
4
5
6
ethanol
ethanol
ethanol
ethanol
iPrOH
tBuOH
water
1:1
2:1
3:1
4:1
4:1
4:1
4:1
4:1
50
40
40
40
25
11
27
66
25
40
57
57
72
78
2
7
8[b]
DMSO
12
[a] Yield of isolated product. [b] Determined by analysis of the
1
crude material by H NMR spectroscopy.
Scheme 4.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
M. Singh, G. Parvari, M. Botoshansky, E. Keinan, O. Reany
FULL PAPER
hydrogen bonds. In the absence of acidic protons on the
nitrogen atoms, the packing is dictated by hydrophobic in-
teractions, such as the interpenetration of the aromatic
groups (Figure 3, right). Yet, there are significantly short
contacts (2.45 and 2.67 Å) between the fully substituted
nitrogen atoms and the ortho-hydrogen atoms on the aro-
matic rings.
It was already demonstrated that the oxygen analogs of
diaryl thioglycoluril diether 10 are useful monomers in step-
wise co-oligomerization reactions,[32] and Isaacs employed
this strategy extensively for the construction of substituted
cucurbiturils.[33] Yet, our attempts to employ the same con-
ditions did not lead to the desired oligomerization. For ex-
ample, treatment of 10 (2 equiv.) with either 8 or 1 (1 equiv.)
in MeSO3H at 50 °C resulted in mixtures of decomposition
products rather than any oligomers.
Figure 2. Molecular structure of 8. (Left) Side view of the molecule
showing the face-to-face interaction between the two aryl groups
to form an angle of 39.8° and several distances between the planes,
which increase along the a axis. (Right) Crystal structure of 8 and
two molecules of thiourea with intermolecular interactions repre-
sented by dashed lines. For color coding, see Figure 1.
Conclusions
The key question at the basis of this study was why thio-
cucurbiturils were not reported during more than three dec-
ades of prolific research on cucurbituril chemistry. Al-
sional angle of N–C–C–N is 11.6°. This distortion results though glycoluril is a highly stable, easily accessible precur-
in a significant increase in the dihedral angle between the sor of all cucurbiturils, its sulfur analog 2 could not be pre-
two tolyl groups (14.7 vs. 0.0° in 8; Figure 3, left). Further- pared to date. We found that the reaction between glyoxal
more, the two rings are no longer parallel to one another. In and thiourea stopped at the level of dihydroxyimidazolid-
8, the N1–C6a–C1aryl–C2aryl and N3–C3a–C1ЈarylЈ–C2ЈarylЈ ine-2-thione (5). The latter was found to be quite unstable
dihedral angles are both 29°, and in 10 they are significantly under various acidic conditions, and it rearranged into
different (35.8 and 16.2°, respectively). As was reported for hydantoin and self-cleavage products. Several stable analogs
the diphenyl analog,[31] the 1,3,5-oxadiazinane rings adopt of 2, including 3, 8, and 10, were prepared in the hope that
a chair conformation, which thus pulls the two sulfur atoms they could be employed as building blocks for the synthesis
closer to one another (5.80 Å in 10 rather than 6.70 Å in of thiocucurbiturils. However, none of these analogs was
8). In contrast to the crystal structures of 3 and 8, the crys- sufficiently stable under the acidic conditions that were re-
tal packing of 10 is no longer dominated by a network of quired for co-oligomerization with formaldehyde. Even pre-
fabricated diether 10 could not stand the conditions for co-
oligomerization with either thioglycoluril 8 or glycoluril (1).
Pichierri already predicted that thiocucurbiturils should
be stable molecules.[9] DFT calculations anticipated that the
portal diameter of thio-CB[6] would be only 0.1 Å broader
and 0.5 Å higher than that of CB[6]. Given that it seems
impossible to synthesize thiocucurbiturils by the same
methods commonly used for the preparation of cucurbitur-
ils, an alternative route must be designed. Thus, the problem
is not how to keep the thiocucurbiturils alive, but how to
make them. Alternative synthetic strategies that are dif-
ferent from the thermodynamically controlled approaches
are currently being pursued in our laboratories. Their ex-
pectedly strong binding properties to specific metal ions
and metal surfaces render the thiocucurbiturils an undoubt-
edly worthy synthetic target.
Experimental Section
Figure 3. Molecular structure of 10. (Left) Side view of the mol-
ecule showing the face-to-face interaction between the two aryl
General Remarks: All commercially available chemicals were pur-
groups forming an angle of 46.7° and several distances between
chased from Aldrich and used without further purification. TLC
was performed on glass sheets precoated on silica gel. 1H NMR
and 13C NMR spectra were performed in CDCl3 and [D6]DMSO
the planes. (Right) Space-filling presentation of two out of the four
molecules of the unit cell, showing the hydrophobic interpenetrat-
ing interactions of the tolyl groups. For color coding, see Figure 1.
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
The Synthetic Challenge of Thioglycolurils
by using the residual solvent signals: CDCl3: δH = 7.26 ppm, δC
=
DMSO): δ = 8.93 (s, 1 H), 7.42 (s, 1 H), 5.41 (s, 1 H) ppm. 13C
NMR ([D6]DMSO): δ = 182.58, 160.8, 68.6 ppm. HRMS [ESI(+)-
TOF]: calcd. for C4H6N4OS [M + H]+ 159.0341; found 159.0332.
77.2 ppm; [D6]DMSO: δH = 2.5 ppm, δC = 39 ppm. NMR spectra
were recorded with a AVIII400 Bruker spectrometer (1H 400 MHz,
13C 100 MHz). IR spectra (KBr pellet) were recorded by using an
FTIR Bruker Alpha spectrometer in the 400–4000 cm–1 range.
Mass spectra and high-resolution mass spectra were recorded with
a Waters LCT Premier microMax spectrometer (ESI-TOF, MeCN/
H2O = 7:3).
General Procedure for the Synthesis of 7–9: A mixture of 4,4Ј-di-
methylbenzil (240 mg, 1 mmol), thiourea (305 mg, 4 mmol), and
KOH (34 mg, 0.6 mmol) in tBuOH (4 mL) was sonicated in an
ultrasonic bath (40 kHz), and the mixture was irradiated at 55 °C
for 2 h. After solvent evaporation, water was added (20 mL) to the
resultant white solid, and the heterogeneous mixture was sonicated
for 15 min and then filtered. The resulting white powder was
washed thoroughly in cold water and dried to afford 3a,6a-bistolyl-
thioglycoluril (8) as a pure white powder (280 mg, 78%); m.p.
Ͼ300 °C (decomp.). The clear aqueous filtrate was concentrated
under reduced pressure to afford a light yellow solid identified as
pure 4,5-dihydroxy-4,5-bistolylimidazolidine-2-thione (7; 35 mg,
11%); m.p. 168–170 °C. Upon neutralizing the filtrate with acetic
acid, precipitation occurred immediately, and the solid was col-
lected. After washing in water and drying under vacuum, hydantoin
9 was isolated as a pure white powder (33 mg, 11%); m.p. 200–
202 °C.
Crystal Structure Analysis: Crystals of 3, 8, and 10 were obtained
by standing a period of time in ethanol solution. The single crystals
were mounted in a Nonius KappaCCD diffractometer (for 8) or
APEX2 DUO diffractometer (for 3 and 10) and data was collected
by using graphite monochromatized Mo-Kα radiation (λ = 0.71073)
at 293 K. The following programs were used for data collection and
reduction: Nonius 1997 Collect,[34] HKL DENZO, and Scale-
pack.[35] Structures were solved by direct methods by using the
maXus[36] program package and refined in the usual way by using
SHELXL97.[37] Non-hydrogen atoms were refined anisotropically
and hydrogen atoms were refined isotropically. CCDC-945330 (for
8), -945332 (for 3), and -945333 (for 10) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from the Cambridge Crystallographic Data Center
via www.ccdc.cam.ac.il/data_request/cif. Pertinent crystallographic
data and refinement parameters for 3, 8, and 10 are summarized
in Table 3.
4,5-Dihydroxy-4,5-bistolyl-imidazolidine-2-thione (7): 1H NMR
([D6]DMSO): δ = 7.20 (d, J = 8 Hz, 4 H), 7.06 (d, J = 8 Hz, 4 H),
2.25 (s, 6 H) ppm. 13C NMR ([D6]DMSO): δ = 183.8, 139.4, 135.6,
128.2, 126.9, 72.7, 20.6 ppm. HRMS [ESI(+)-TOF]: calcd. for
C17H18N2O2S [M + H]+ 315.1167; found 315. 1153.
cis-/trans-4,5-Dihydroxyimidazolidine-2-thione (5): A mixture of
thiourea (3 g, 40 mmol), glyoxal (40% in H2O, 2 mL, 20 mmol),
and water (20 mL) was stirred at room temperature. After 30 min,
the off-white precipitate separated out. The precipitate was filtered
and dried under vacuum. Product trans-5 was obtained as a pure
3a,6a-Bistolylthioglycoluril (8): IR (KBr): ν = 3100–3300 (NH),
˜
1
1543 (CN), 1495 (C=C), 1208, 1085 (C=S) cm–1. H NMR ([D6]-
DMSO): δ = 9.71 (br. s, 4 H), 6.93 (d, J = 8 Hz, 4 H), 6.83 (d, J =
8 Hz, 4 H), 2.12 (s, 6 H) ppm. 13C NMR ([D6]DMSO): δ = 182.3,
137.7, 132.6, 128.2, 126.8, 89.6, 20.5 ppm. HRMS [TOF(+)-ESI]:
calcd. for C18H18N4S2 [M – H]+ 353.0895; found 353.0876.
off-white powder (2.3 g, 85%). IR (KBr): ν = 3300–3400 (OH),
˜
3150–3250 (NH), 1617 (NH), 1474 (CN), 1414 (C=S) cm–1. 1H
NMR ([D6]DMSO): δ = 8.83 (br. s, NH, 2 H), 6.26 (d, J = 7 Hz,
OH, 2 H), 4.72 (d, J = 7 Hz, CH, 2 H) ppm. 13C NMR ([D6]-
DMSO): δ = 182.2, 87 ppm. HRMS [ESI(+)-TOF]: calcd. for
C3H7N2O2S [M + H]+ 135.0228; found 135. 0226.
5,5-Bistolyl-2-thione-4-imidazolidone (9): IR (KBr): ν = 3050–3150
˜
1
(NH), 1737 (C=O), 1524 (CN), 1188, 1165 (C=S) cm–1. H NMR
([D6]DMSO): δ = 12.08 (s, 1 H), 11.24 (s, 1 H), 7.21 (d, J = 8 Hz,
4 H), 7.16 (d, J = 8 Hz, 4 H), 2.29 (s, 6 H) ppm. 13C NMR ([D6]-
DMSO): δ = 181.62, 175.9, 138.2, 136.06, 129.7, 126.9, 73.2,
20.1 ppm. HRMS [TOF(+)-ESI]: calcd. for C18H18N4S2 [M – H]+
295.0905; found 295.0908.
In a similar reaction performed at 65 °C, a mixture of cis/trans
diastereoisomers of 5 was obtained, and the ratio was determined
according to the 1H NMR spectrum of the mixture. cis-5: 1H NMR
([D6]DMSO): δ = 8.56 (br. s, NH, 2 H), 5.81 (d, J = 7.6 Hz, OH,
2 H), 5.03 (d, J = 7.6 Hz, CH, 2 H) ppm. 13C NMR ([D6]DMSO):
δ = 181.6, 80 ppm.
Compound 10: A mixture of 8 (0.7 g, 2 mmol), paraformaldehyde
(0.24 g, 8 mmol), and concentrated HCl (20 mL) was heated at
100 °C for 24 h. After being cooled to room temperature, the prod-
trans-4,5-Dimethoxyimidazolidine-2-thione (6): Concentrated HCl uct precipitated out of the cold solution. The solid was filtered off,
was carefully added (0.2 mL) to a solution of trans-5 (3 g, 40 mmol)
in methanol (15 mL), and the mixture was heated at 50 °C. After
3 h, a white precipitate separated out. After filtration and drying
under vacuum, 6 was obtained as a pure white solid (2.6 g, 40%).
1H NMR ([D6]DMSO): δ = 9.45 (br. s, NH, 2 H), 4.69 (s, CH, 2
washed with water, and finally dried under vacuum. Thus, 10 was
obtained as a white crystalline (0.72 g, 82%); m.p. Ͼ300 °C (de-
1
comp.). H NMR (CDCl3): δ = 6.97 (s, 8 H), 6.22 (d, J = 10.4 Hz,
4 H), 4.77 (d, J = 10.8 Hz, 4 H), 2.21 (s, 6 H) ppm. 13C NMR
(CDCl3): δ = 186.2, 139.8, 129.7, 128.6, 127.5, 86.6, 75.1, 21.0 ppm.
H), 3.24 (s, OMe, 6 H) ppm. 13C NMR ([D6]DMSO): δ = 184.0, HRMS [ESI(+)-TOF]: calcd. for C22H22N4O2S2 [M
+
H]+
91.7, 54.0 ppm. HRMS [ESI(+)-TOF]: calcd. for C5H10N2O2S [M
439.1262; found 439.1250.
+ H]+ 163.0541; found 163.0529.
Supporting Information (see footnote on the first page of this arti-
cle): Copies of the NMR spectra of 3, trans-5, and 6–10.
Monothioglycoluril (3): A mixture of 6 (0.49 g, 3 mmol) and urea
(0.22 g, 3.6 mmol) in H2O (20 mL) was stirred at 80 °C and concen-
trated HCl (0.2 mL) was carefully added. The reaction was kept at
80 °C for 2 h, and a fine, dark powder separated out during the
reaction. Upon completion of the reaction (TLC), the insoluble
black precipitate was filtered off. The clear aqueous solution was
stored at 5 °C in the fridge overnight, which gave a pale brown
crystalline solid as a mixture of 3 (90%) and 1 (10%). Recrystalli-
zation from aqueous HCl afforded 3 as a brownish solid (0.22 g,
Acknowledgments
This study was supported by the Open University of Israel and the
Institute of Catalysis Science and Technology, Technion. E. K. is
the incumbent of the Benno Gitter & Ilana Ben-Ami Chair of Bio-
technology, Technion. M. S. is supported at the Open University
through a fellowship of the Israel Council for Higher Education
45%); m.p. 300–302 °C. IR (KBr): ν = 3200–3450 (NH), 1713
˜
1
(C=O), 1617 (NH), 1512 (CN), 1110 (C=S) cm–1. H NMR ([D6]- (PBC Program). G. P. thanks the Schulich Scholarship Program
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
M. Singh, G. Parvari, M. Botoshansky, E. Keinan, O. Reany
FULL PAPER
[19] a) J. N. Arani, J. Safari, Ultrason. Sonochem. 2011, 18, 640–
643; b) J.-T. Li, X.-R. Liu, M.-X. Sun, Ultrason. Sonochem.
2010, 17, 55–57.
and the Israel Ministry of Science and Technology for the Eshkol
scholarship.
[20] A. N. Kravchenko, V. V. Baranov, Y. V. Nelyubina, G. A. Gazi-
eva, I. V. Svitan’ko, Russ. Chem. Bull. Int. Ed. 2012, 61, 64–73.
[21] Atta-ur-Rahmans, Nuclear Magnetic Resonance, Springer, New
York, 1986, ch. 4.
[22] E. Grillon, R. Gallo, M. Piemot, J. Boileau, E. Wimmer, Tetra-
hedron Lett. 1988, 29, 1015–1016.
[23] a) C. J. Broan, A. R. Butler, J. Chem. Soc. Perkin Trans. 2 1992,
23–28; b) C. J. Broan, A. R. Butler, J. Chem. Soc. Perkin Trans.
2 1991, 1501–1504.
[24] G. A. Gazieva, Y. V. Nelyubina, A. N. Kravchenko, A. S. Siga-
chev, I. V. Glukhov, M. I. Struchkova, K. A. Lyssenko, N. N.
Makhova, Russ. Chem. Bull. Int. Ed. 2009, 58, 1945–1954.
[25] Compound 3 was previously prepared directly from 5 and urea
in 5% yield: V. A. Eres’ko, L. V. Epishina, O. V. Lebedev, M. V.
Povstyanoi, L. I. Khmel’nitskii, S. S. Novikov, Bull. Acad. Sci.
USSR Div. Chem. Sci. (Engl. Transl.) 1980, 29, 1138–1141.
[26] S. Vail, R. Barkeran, D. P. G. Mennitt, J. Org. Chem. 1965, 30,
2179–2182.
[1]
[2]
a) X.-J. Cheng, L.-L. Liang, K. Chen, N.-N. Ji, X. Xiao, J.-X.
Zhang, Y.-Q. Zhang, S.-F. Xue, Q.-J. Zhu, X.-L. Ni, Z. Tao,
Angew. Chem. 2013, 125, 7393–7396; Angew. Chem. Int. Ed.
2013, 52, 7252–7255; b) J. Lagona, P. Mukhopadhyay, S. Chak-
rabarti, L. Isaacs, Angew. Chem. 2005, 117, 4922–4949; Angew.
Chem. Int. Ed. 2005, 44, 4844–4870.
Latest review on applicable properties of cucurbiturils: a) G.
Parvari, O. Reany, E. Keinan, Isr. J. Chem. 2011, 51, 646–663;
more recently: b) M. K. Sinha, O. Reany, M. Yefet, M. Botosh-
ansky, K. Keinan, Chem. Eur. J. 2012, 18, 5589–5605; c) B. C.
Pemberton, R. Raghunthan, S. Volla, J. Sivaguru, Chem. Eur.
J. 2012, 18, 12178–12190; d) M. Cao, d. Wu, S. Gao, R. Cao,
Chem. Eur. J. 2012, 18, 12978–12985; e) Y. H. Ko, I. Hwang,
D.-W. Lee, K. Kim, Isr. J. Chem. 2011, 51, 506–514; f) X. Lu,
E. Masson, Langmuir 2011, 27, 3051–3058; g) C. Kim, S. S.
Agasti, Z. Zhu, L. Isaacs, V. Totello, Nat. Chem. 2010, 2, 962–
966; h) C. R. Thomas, D. P. Ferris, J.-H. Lee, E. Choi, M. H.
Cho, E. S. Kim, J. F. Stoddart, J.-S. Shin, J. Cheon, J. I. Zink,
J. Am. Chem. Soc. 2010, 132, 10623–10625; i) D. Kim, E. Kim,
J. Lee, S. Hong, W. Sung, N. Lim, C. G. Park, K. Kim, J. Am.
Chem. Soc. 2010, 132, 9908–9919.
[27] C. J. Broan, A. R. Butler, J. Chem. Soc. Perkin Trans. 2 1989,
731–740.
[28] a) G. Baccolini, C. Boga, C. Delpivo, G. Micheletti, Tetrahe-
dron Lett. 2011, 52, 1713–1717; b) J. Safari, H. Naeimi, M. M.
Ghanbari, O. Sabzi-Fini, Russ. J. Org. Chem. 2009, 45, 477–
479; c) G. Muccioli, J. Wouters, J. H. Poupaert, B. Norberg, W.
Poppitz, G. K. E. Scriba, D. M. Lambert, Org. Lett. 2003, 5,
3599–3602; d) G. Muccioli, J. H. Poupaert, J. Wouters, B. Nor-
berg, W. Poppitz, G. K. E. Scriba, D. M. Lambert, Tetrahedron
2003, 59, 1301–1307; e) P. Satya, M. Gupta, R. Gupta, A.
Loupy, Synthesis 2002, 1, 75–78.
[3]
[4]
[5]
[6]
S. G. Murray, F. R. Hartley, Chem. Rev. 1981, 81, 365–414.
S. R. Cooper, Acc. Chem. Res. 1988, 21, 141–146.
P. Lhota’k, Eur. J. Org. Chem. 2004, 1675–1692.
F. Hamada, S. G. Bott, G. W. Orr, A. W. Coleman, H. Zhang,
J. L. Atwood, J. Inclusion Phenom. Mol. Recognit. Chem. 1990,
195–206.
M. T. Rojas, R. Königer, J. F. Stoddart, A. E. Kaifer, J. Am.
Chem. Soc. 1995, 117, 336–343.
[7]
[29] L. Cao, J. Ding, G. Ying, M. Gao, Y. Li, A. Wu, Synlett 2009,
1445–1448.
[8]
[9]
F. Pichierri, Chem. Phys. Lett. 2005, 403, 252–256.
F. Pichierri, Chem. Phys. Lett. 2004, 390, 214–219.
[30] A. Masunov, J. J. Dannenberg, J. Phys. Chem. B 2000, 104,
806–810.
[10] W. L. Mock, T. Manimaran, W. A. Freeman, R. M. Kuksuk,
J. E. Maggio, D. H. Williams, J. Org. Chem. 1985, 50, 60–62.
[11] M. W. Heaven, Ph.D. Dissertation: Structural Supramolecular
[31] G. T. W. Gieling, R. J. M. Nolte, J. W. Scheeren, J. M. M.
Smits, P. T. Beurskens, G. Beurskens, J. Chem. Crystallogr.
1995, 25, 421–424.
Assemblies of Spheres and Tubes (supervisor: J. L. Atwood), [32] M. Stancl, M. Hodan, V. Sindelar, Org. Lett. 2009, 11, 4184–
University of Missouri-Columbia, Colombia, USA, May 2006. 4187.
[12] a) B. S. Pedersen, S. Scheibye, N. H. Nilsson, S. O. Lawesson, [33] D. Ma, P. Y. Zavalij, L. Isaacs, J. Org. Chem. 2010, 75, 4786–
Bull. Soc. Chim. Belg. 1978, 87, 223–228; for review on applica-
tions of this reagent, see b) M. P. Cava, M. I. Levinson, Tetra-
hedron 1985, 41, 5061–5087.
4795.
[34] Kappa CCD Server Software, Nonius BV, 1997, Delft, Nether-
lands.
[13] A. A. El-Barbary, S. O. Lawesson, J. Ind. Chem. 1984, 23B,
655–657.
[35] Z. Otwinowski, W. Minor, Methods Enzymol. 1997, 276, 307.
[36] S. Mackay, C. J. Gilmore, C. Edwards, M. Tremayne, N. Stuart,
K. Shankland, maXus: A Computer Program for the Solution
and Refinement of Crystal Structures from Diffraction Data,
University of Glasgow, Scotland, UK, Nonius BV, Delft, The
Netherlands and MacScience Co. Ltd., Yokohama, Japan,
1998.
[14] C. N. Cow, P. H. M. Harrison, J. Org. Chem. 1997, 62, 8834–
8840.
[15] F. B. Slezak, H. Bluestone, T. A. Magee, J. H. Wotiz, J. Org.
Chem. 1962, 27, 2181–2183.
[16] Q. Long, G. Hong, H. Ding, Youji Hauxue 1982, 3, 170–175.
[17] G. Baccolini, C. Boga, C. Delpivo, G. Micheletti, Tetrahedron
Lett. 2011, 52, 1713–1717.
[18] H. Pauly, H. Sauter, Ber. Dtsch. Chem. Ges. 1930, 63B, 2063–
2066.
[37] G. M. Sheldrick, SHELXL-97, University of Göttingen,
Göttingen, Germany, 1997.
Received: November 7, 2013
Published Online: ᭿
8
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O31672
/KAP1
Date: 02-01-14 19:26:17
Pages: 9
The Synthetic Challenge of Thioglycolurils
Thioglycolurils
Although glycoluril is a highly stable, easily
accessible precursor of all cucurbiturils, di-
thioglycoluril is still unknown. To answer
the question why thiocucurbituril has never
been made, we prepared several thioglycol-
uril derivatives and found that they do not
survive the acidic conditions usually
needed for the co-oligomerization reaction
with paraformaldehyde.
M. Singh, G. Parvari, M. Botoshansky,
E. Keinan,* O. Reany* ...................... 1–9
The Synthetic Challenge of Thioglycolurils
Keywords: Building blocks / Dicarbonyls /
Hydantoin / Macrocycles / Sulfur
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
9