Václav Havel1, Tereza Sadilová1, Vladimír Šindelář1. 1. Department of Chemistry, Faculty of Science, and RECETOX, Faculty of Science, Masaryk University, Kamenice 5, 625 00 Brno, Czech Republic.
Abstract
A new bambusuril derivative, (H)BU[6], lacking substituents on the ureidic nitrogen atoms, has been isolated and characterized. This macrocycle was prepared by the deprotection of bambusuril (PMB)BU[6]. (H)BU[6] is attractive for use as a starting compound for the preparation of other bambusuril derivatives, which was demonstrated via propargylation and the copper-catalyzed click reaction performed on the macrocycle.
A new bambusuril derivative, (H)BU[6], lacking substituents on the ureidic nitrogen atoms, has been isolated and characterized. This macrocycle was prepared by the deprotection of bambusuril(PMB)BU[6]. (H)BU[6] is attractive for use as a starting compound for the preparation of other bambusuril derivatives, which was demonstrated via propargylation and the copper-catalyzed click reaction performed on the macrocycle.
Bambusurilsare a family
of macrocyclic compounds consisting of n glycoluril
units connected by n methylene
bridges (where n = 4 and 6).[6]Uril. Angew. Chem.,
Int. Ed.. 2010 ">1,2] Four-membered
bambusurils (BU[4]) are not able to include any guests
inside of their small cavity. On the contrary, six-membered bambusurils
(BU[6]) are prized for their excellent affinity for inorganic
anions in water and many organic solvents.[3−6] The high affinity of six-membered
bambusurils toward anions is further shared among other hemicucurbituril
derivatives,[7−9] including biotinurils.[10,11] Recently,
bambusuril derivatives, semithio- and semiaza-bambusurils, have been
prepared by replacing oxygen atoms on the portals of the bambusurils
by sulfur and nitrogen, respectively.[12,13] The bambusuril
macrocycle has been used as parts of anion recognition systems,[14] electromembrane anion extraction,[15] photosensitive materials,[16] and systems for anion transmembrane transport.[17]Both the solubility and binding properties
of bambusurilscan be
influenced by the type of substituents attached to the nitrogen atoms
on the opposed macrocycle portals. Until now, there has been only
one synthetic approach enabling the synthesis of bambusurils and thus
also only one synthetic approach enabling the modification of substituents
on their framework. This approach is based on the preparation of desired
derivatives of glycoluril (Scheme ) and its subsequent acid-catalyzed macrocyclization
reaction with formaldehyde yielding a desired bambusuril. However,
some glycolurils bearing certain types of substituents have proved
to be unstable under these conditions, and no macrocycle could be
isolated. Herein, we present a new synthetic protocol which allows
for the preparation of bambusurils which are not achievable by the
traditional synthetic approach.
Scheme 1
Classical and Newly Designed Approach
for the Synthesis of Bambusurils, n = 2 and 3
Results and Discussion
Our novel synthetic protocol is based on the preparation of a bambusuril
bearing protecting groups on the nitrogen atoms of the glycoluril
units (Scheme ). The
protecting group would subsequently be removed yielding bambusuril (H)BU[6]/(H)BU[4], with nitrogen atoms on the
macrocycle portals available for further reactions. Several protecting
groups were tested with this approach. Our first choice, a p-methoxyphenyl substituent (Figure ), which is removable by oxidizing agents,[18] proved to be unsuitable because of the low solubility
of the corresponding glycoluril. In addition, steric hindrance by
the rigid substituent prevented the glycolurilbuilding blocks from
assembling into a macrocyclic structure. We also investigated deprotection
of the benzyl substituent on the already available macrocycles (Bn)BU[6] or (Bn)BU[4] (see Figure for the structure). Unfortunately,
deprotection of the macrocycles via hydrogenation or their treatment
with a solution of sodium metal in liquid ammonia resulted either
in only partial deprotection (Figure S28) or decomposition of the macrocycle. Finally, we prepared bambusurils(PMB)BU[4] and (PMB)BU[6] bearing a p-methoxybenzyl protecting group (PMB, Scheme ), which we successfully converted to the
bambusurils (H)BU[6] and (H)BU[4]. The synthetic
procedure for PMB derivatives started with the quantitative synthesis
of the corresponding urea, which can then undergo an acid-catalyzed
reaction with 4,5-dihydroxyimidazolidin-2-one to give the glycoluril,
with a yield of 81%. The macrocyclization reaction was complicated
by the unwanted reaction between formaldehyde and the PMB protecting
groups on the glycolurils, which takes place under the acidic conditions
required for the reaction.
Figure 1
Compounds unsuccessfully tested for bambusuril
deprotection.
Scheme 2
Preparation of New
Bambusuril Derivatives
Compounds unsuccessfully tested for bambusuril
deprotection.Thus, moderate formation of the desired macrocycles was
only observed
when we limited the amount of formaldehyde to an amount equivalent
with regard to the monomer unit. Even a slight excess of formaldehyde
leads to a complete degradation of the macrocycles formed, and no
macrocycle could then be isolated.The main focus was on the
six-membered bambusuril homologue because
this macrocycle, unlike its four-membered homologue, acts as an excellent
receptor for a number of anions. The best conversion of the glycoluril
to (PMB)BU[6] was achieved when 1,4-dioxane was used
as the solvent, and the reaction was carried out with only a catalytic
amount of H2SO4 under a longer reaction time
(48 h) and at an elevated temperature (Scheme ). Given the acid lability of the PMB group,
we were unable to access an anion-free derivative by the use of a
HSO4– template as recently published;[5] instead, an iodide template was employed. (PMB)BU[6] was isolated in the pure form by column chromatography
as the TBAI complex, with a 19% yield. The removal of the iodide template
by oxidation[4,19] was not possible because of the
instability of PMB groups toward the oxidizing reagents. The (PMB)BU[6] complex with TBAI was used for the deprotection
reaction.The first attempts were made with cerium(IV) ammonium
nitrate (CAN)
in either a CH3CN/H2O mixture[20] or DMSO/H2O. Unfortunately, because of the low
solubility of the partially deprotected macrocycle (in CH3CN/H2O) and difficult separation (in DMSO/H2O), these attempts did not lead to the desired compound. Strong acids,
such as CF3COOH, have also been reported in the literature
as an alternative for the removal of this protecting group at elevated
temperatures.[21,22] Neat CF3COOH, or its
50% solution in CHCl3, proved to function both as a deprotecting
agent and also as a solvent. Complete deprotection was achieved after
1 h at 100 °C in a 50% CF3COOH/CHCl3 mixture
or 15 min in neat CF3COOH under microwave irradiation.
If a chromatographically purified sample of (PMB)BU[6] was used for the deprotection, it was possible to obtain pure (H)BU[6], with a yield of 58%.We also investigated
if the deprotection of a crude mixture containing (PMB)BU[6] without its chromatography purification could be
used to obtain (H)BU[6]. Given the poor solubility of
this unprotected macrocycle, we expected that the majority of linear
oligomers, and other side products present in the crude material,
could be separated because of their better solubilities. Indeed, after
deprotection, most of the unwanted oligomers were washed out by MeOH
and CH3COOH, and (H)BU[6] was isolated with
a purity sufficient for further reactions. Unfortunately, the atom
efficiency of the deprotection reactions is low, as the removal of
12 protecting groups radically decreases the molecular weight of the
compound. Furthermore, the isolated (PMB)BU[6]·TBAI
complex is, in the presence of strong acid, transformed into the (H)BU[6]·HI complex, and the molecular weight of the
starting material decreases from 2736 to 1053 g mol–1 in the product. In addition, (PMB)BU[4] was prepared
in 1,4-dioxane with p-toluenesulfonic acid (PTSA)
used as a catalyst, and in the absence of a template, with the low
yield of 6%. Deprotection of (PMB)BU[4] was successfully
achieved by its reaction with CAN in a DMSO/MeOH/CH3CN
mixture.[20] Given the low amount of the
isolated macrocycle as well as its inability to bind anions, its modification
was not studied further and (H)BU[4] was only characterized
by the 1H NMR spectrum in DMSO-d6 (Figure S3). The 1H and 13C NMR spectra of (H)BU[6] retain the high symmetry
exhibited by all bambusuril derivatives (Figure ). The presence of ureidic NH groups on the
macrocycle was supported by the measured 1H–15N heteronuclear single quantum coherence (HSQC) spectrum
(Figure S11). The inclusion complexes between
six-membered bambusurils and anions are usually observed by 1H NMR spectroscopy.
1H–n class="Chemical">13C HSQC NMR spectrum
(DMSO-d6, 500/126 MHz, 30 °C) of (H)BU[6].
A chemical shift of anion-free
bambusuril protons (particularly
methine proton c) differs significantly from that
of anion-free macrocycles. We tried to follow the chemical shift of
an anion-free (H)BU[6], generated inside a NMR tube by
the addition of an excess of anion-free (Bn)BU[6] receptor
into the solution of the (H)BU[6] complex with HI. New
signals corresponding to an iodide complex with (Bn)BU[6] indicated that the ion was transferred from (H)BU[6] to (Bn)BU[6]. Interestingly, we only observed a negligible
change in the chemical shift of the NH and CH signals of (H)BU[6], which would correspond to the removal of I– from
the complex (Figure S11). This unexpected
feature is currently under investigation in our laboratory.We decided to illustrate the use of (H)BU[6] as a
precursor for other bambusuril derivatives. Free NH positions on a
bambusuril scaffold can conveniently be alkylated in DMSO in the presence
of KOH. As reported,[23] powdered KOH in
DMSO acts as a strong base which is just strong enough to deprotonate
the ureidic NH groups. (H)BU[6] prepared from a crude (PMB)BU[6] mixture was propargylated with an excess of propargyl
bromide (Scheme ).
A small amount of insoluble oligomeric impurities were removed after
the alkylation. According to MS analysis (Figure S32), the majority of the NH groups on bambusuril reacted and
only traces of a macrocycle with 11 substituents were present.Alkynes readily react with organic azides under copper(I) catalysis
(azide–alkyne cycloaddition), and we used (propargyl)BU[6] (Scheme ) as an
alkyne component in this reaction. After the reaction, we observed
the disappearance of the NMR signal of the starting alkyne compound;
however, the characteristic signals of product 1 were
only observed after the addition of an excess of TBAI.The alkylation
reaction was also attempted on2,4-dimethylglycoluril,
which is structurally similar to the (H)BU[6] macrocycle.
The propargyl derivative of glycoluril as well as compound 2 (Figure ) was used
to further prove the structures of the novel bambusurils ((propargyl)BU[6] and 1). Unfortunately, no crystals suitable for X-ray
analysis were obtained from any of the novel macrocycles ((H)BU[6], (propargyl)BU[6], or 1). To validate
the structure of 1 and to observe the spatial arrangement
of its bulky substituents on the glycoluril monomer, we attempted
to crystallize compound 2.
Figure 3
Glycoluril derivatives
used in the optimization of bambusuril derivatization.
Glycoluril derivatives
used in the optimization of n class="Chemical">bambusuril derivatization.
This compound was prepared prior the modification
of the macrocycle
as a test compound (see the Supporting Information). Slow diffusion of diethyl ether vapors into CH2Cl2 solution of 1 produced monocrystals suitable
for the X-ray diffraction analysis. The crystal structure (Figure ) showed that the
bulky substituents are oriented in such a way that the aromatic ring
can possibly engage in π–π stacking interactions.
Figure 4
Front
and side views of the crystal structure of glycoluril 2. Color coding: O, red; C, gray; N, blue; and H, white.
Front
and side views of the crystal structure of glycoluril 2. Color coding: O, red; C, gray; N, blue; and H, wn class="Chemical">hite.
Conclusions
In conclusion, for the
first time, four- and six-membered bambusurils
lacking substituents on the nitrogen atoms of their portals were synthesized
and characterized. This was achieved by careful selection of the protecting
group as well as the conditions of macrocyclization and the deprotection
reaction. A deprotected bambusuril represents a versatile bambusuril
intermediate, which will allow for the preparation of derivatives
that are not achievable by classical acid-catalyzed macrocyclization.
This post-macrocyclization derivatization of bambusurils was demonstrated
by the preparation of the propargyl derivative, which was later used
in azide–alkyne cycloaddition.
Experimental Section
General
Methods
NMR spectra were recorded on a Bruker
AVANCE III 300 spectrometer with working frequencies of 300.15 MHz
for 1H and 75.47 MHz for 13C or on a Bruker
AVANCE III 500 spectrometer with working frequencies of 500.11 MHz
for 1H and 125.75 for 13C. Both spectrometers
were equipped with a BBFO probe. All experiments were recorded at
303.15 K and were processed using a MestReNova v. 9.1.0 program. NMR
chemical shifts (δ) are reported in parts per million (ppm)
using the residual solvent signal as a reference for the measured
spectra CDCl3 (1H = 7.26, 13C = 77.16)
and DMSO-d6 (1H = 2.50, 13C = 39.52). The proton resonances are annotated in text as:
chemical shift, multiplicity (s, singlet; d, doublet; t, triplet;
q, quartet; and m, multiplet), coupling constant [J (Hz)], and integration. In several instances, the position of carbon
signal for reported compounds was deduced from HSQC and heteronuclear
multiple bond correlation (HMBC) experiments as 13CNMR
produced only weak signals for quaternary groups. HRMS analysis was
recorded on an Agilent 6224 Accurate-Mass TOF LC–MS. Samples
were ionized by electrospray ionization (ESI). The MALDI-TOF mass
spectrum was recorded on a MALDI-TOF Axima CFR spectrometer. The sample
was ionized with the aid of a nitrogen laser (a wavelength of 337
nm and a maximum power of 6 MW); α-cyano-4-hydroxycinnamic acid
or 2,5-dihydroxybenzoic acid was used as a matrix. To improve the
accuracy of the measurements, a calibration experiment was performed
by mixing the sample with synthetic peptides. The mixture was analyzed,
and the obtained calibration function was applied as a correction
for the analysis of a pure sample of macrocycle. All column chromatography
procedures were performed on columns packed with a silica gel (40–60
μm). Thin-layer chromatography (TLC) was performed using silica
gel plates Silica Gel 60 F254 (0.2 mm thickness, Merck) and visualized
under a UV lamp (254 and 366 nm) or with CAM stain. All solvents and
chemicals were used as purchased or purified by standard procedures
when necessary.
Synthetic Procedures
1,3-Bis(4-methoxyphenyl)urea
4-Methoxyaniline (3.7
g, 30 mmol) and n class="Chemical">urea (0.60 g, 10 mmol) were dissolved in H2O (4.7 mL) acidified with 35% HCl (2.6 mL). The reaction mixture
was heated to reflux for 26 h. During the reaction, solid precipitates
from the mixture. After cooling, the solid was collected by filtration,
washed with cold H2O and EtOH, and dried in vacuum. The
product was isolated as a solid, slightly colored to violet by traces
of starting compound (2.4 g, 88%). 1H NMR (300 MHz, DMSO-d6): δ 8.33 (s, 2H), 7.33 (d, J = 8.2 Hz, 4H), 6.85 (d, J = 8.6 Hz, 4H), 3.71 (s,
6H), spectral data correspond to the literature.[24]
2,4-Bis(4-methoxyphenyl)glycoluril
1,3-Bis(4-methoxyphenyl)urea
(1.0 g, 3.7 mmol) and n class="Chemical">DHI (1.3 g, 11 mmol) were dissolved in boiling
EtOH (100 mL) acidified with 35% HCl (0.42 mL). The mixture was refluxed
for 20.5 h. A precipitated solid material was collected by filtration
from hot mixture, washed with water, and dried in vacuum. Beige solid
was obtained in yield (0.5 g, 38%). 1H NMR (300 MHz, DMSO-d6): δ 7.87 (s, 2H), 7.44 (d, J = 8.9 Hz, 4H), 6.94 (d, J = 9.0 Hz, 4H), 5.88 (s,
2H), 3.76 (s, 6H). 13C NMR (75 MHz, DMSO): δ 161.16,
155.99, 153.58, 130.95, 122.76, 114.00, 66.53, 55.24, 39.52. HRMS
(APCI+): calcd for [C18H18N4O4 + H]+, theoretical: 355.1401; experimental:
355.1399.
1,3-Bis(4-methoxybenzyl)urea
4-Methoxybenzylamine (1.0
mL, 7.7 mmol) and n class="Chemical">diphenyl carbonate (0.80 g, 3.7 mmol) were mixed
in triethylamine (2.7 mL, 19 mmol). The reaction mixture was heated
to reflux for 5 h. After cooling down, 1 M NaOH solution (10 mL) was
added and the suspension was thoroughly stirred for 30 min. A solid
precipitate was collected by filtration and thoroughly washed with
water (3 × 5 mL). A white solid product was dried in vacuum (1.1
g, quantitative). 1H NMR (300 MHz, DMSO-d6): δ 7.17 (d, J = 8.6 Hz, 4H),
6.86 (d, J = 8.6 Hz, 4H), 6.28 (t, J = 6.0 Hz, 2H), 4.14 (d, J = 5.9 Hz, 4H), 3.72 (s,
6H), spectral data correspond to the literature.[25]
2,4-Bis(4-methoxybenzyl)glycoluril
1,3-Bis(4-methoxybenzyl)urea
(6.3 g, 21 mmol) and DHI (4.9 g, 42 mmol) were suspended in MeOH (84
mL) and acidified with 35% HCl (0.42 mL). The mixture was refluxed
for 8 h during which all starting materials dissolved. The mixture
was filtered while hot to remove small amount of unsubstituted glycoluril
and after cooling to room temperature (RT), it was left to crystallize
in fridge overnight. A colorless crystalline product was collected
by filtration and washed with water (3 × 15 mL) and acetone (2 ×
10 mL). Yield (6.5 g, 81%). 1H NMR (300 MHz, DMSO-d6): δ 7.60 (s, 2H), 7.21 (d, J = 8.6 Hz, 4H), 6.90 (d, J = 8.6 Hz, 4H), 5.00 (s,
2H), 4.58 (d, J = 15.2 Hz, 2H), 3.96 (d, J = 15.2 Hz, 2H), 3.73 (s, 6H). 13C NMR (75 MHz,
DMSO): δ 161.0, 158.5, 157.2, 129.2, 113.9, 65.0, 55.0, 43.4.
HRMS (APCI+): calcd for [C20H22N4O4 + H]+, theoretical: 383.1714; experimental:
383.1713.
2,4-Bis(4-methoxybenzyl) glycoluril
(0.5 g, 1.3 mmol),
n class="Chemical">paraformaldehyde (39 mg, 1.3 mmol), and PTSA (0.12 g, 0.65 mmol) were
heated in dioxane (2.5 mL) at 80 °C for 22 h. After cooling to
RT, the solution was diluted with water to a volume of 25 mL. A precipitated
solid was collected by filtration and washed with water (2 ×
25 mL). The crude material was filtered through silica column (10%
acetone in DCM) and later purified by PTLC (15% acetone in DCM). The
solid material (46 mg) was washed with MeOH (1 mL) and collected by
centrifugation to obtain pure (PMB)BU[4] as a white solid
(33 mg, 6%). 1H NMR (500 MHz, CDCl3): δ
7.12 (d, J = 8.2 Hz, 16H), 6.82 (d, J = 8.3 Hz, 16H), 5.38 (s, 8H), 4.40 (d, J = 15.7
Hz, 8H), 4.33 (d, J = 15.7 Hz, 8H), 3.88 (s, 8H),
3.78 (s, 24H). 13C NMR (126 MHz, CDCl3): δ
159.35, 158.90, 158.75, 129.15, 129.11, 114.29, 77.41, 77.36, 77.16,
76.98, 76.91, 71.33, 55.42, 50.79, 46.65. HRMS (MALDI-TOF): calcd
for [C84H88N16O16 + Na]+, theoretical: 1599.646; experimental: 1599.645 ± 0.003.
2,4-Bis(4-methoxybenzyl)glycoluril (1 g, 2.6 mmol), para-formaldehyde (78.6 mg, 2.62 mmol), and TBAI (0.16 g,
0.44 mmol) were heated in dioxane (4 mL) at 80 °C (temperature
of the heating adapter). After 5 min, H2SO4 (22
μL) was added into the warm mixture. The heating was stopped
after 48 h. After cooling to RT, the solution was diluted with water
to a volume of 40 mL. A precipitated solid was collected by filtration
and washed with water (2 × 25 mL). The wet solid was dissolved
in acetone/DCM (1:1) concentrated in vacuum and azeotropically dried
with toluene (20 mL). A crude material was prepurified by column chromatography
on silica (15% acetone in DCM, removal of oligomers) and by PTLC [pure
EA, separation of (PMB)BU[4] and traces of oligomers
from (PMB)BU[6]]. The product was obtained as a complex
of TBAI (0.23 g, 19%). 1H NMR (500 MHz, CDCl3): δ 7.25 (d, J = 9.1 Hz, 24H), 6.76 (d, J = 8.7 Hz, 24H), 5.74 (s, 12H), 4.84 (d, J = 15.8 Hz, 12H), 4.57 (d, J = 15.9 Hz, 12H), 4.30
(s, 12H), 3.20–3.02 (m, 8H), 1.58–1.44 (m, 8H), 1.34
(q, J = 7.3 Hz, 8H), 0.96 (t, J =
7.3 Hz, 12H). 13C NMR (126 MHz, CDCl3): δ
160.41, 159.29, 158.70, 131.43, 128.07, 114.00, 70.18, 59.00, 55.33,
47.66, 47.23, 24.12, 19.85, 13.77. HRMS (MALDI-TOF): calcd for [C84H88N16O16 + Na]+, theoretical: 2387.974; experimental: 2387.976 ± 0.009.
Unsubstituted Bambus[6]uril [(H)BU[6]]
(PMB)BU[6]·n class="Chemical">TBAI (25 mg, 9.1 μmol) was dissolved
in CHCl3 (1 mL) and CF3COOH (1 mL) in a 10 mL
microwave pressure tube. The reaction mixture was subjected to microwave
irradiation (100 W, 100 °C) for 1 h. After cooling to RT, the
dark mixture was precipitated with MeOH (25 mL) and the solid was
separated by centrifugation. The liquid was decanted, and the crude
product was washed with MeOH (2 × 5 mL), CH2Cl2/MeOH 1:1 (5 mL), acetone (5 mL), and H2O (5 mL).
In all washing steps, the isolation was done with centrifugation and
decantation. After drying in vacuum at 50 °C, the material was
obtained as a gray solid (5.6 mg, 58%). After deprotection, the macrocycle
was obtained in the form of (H)BU[6]·HI complex. 1H NMR (500 MHz, DMSO-d6): δ
7.66 (s, 1H), 5.32 (s, 1H), 4.52 (s, 1H). 13C NMR (126
MHz, DMSO): δ 160.61, 157.77, 65.96, 46.66. HRMS (MALDI-TOF):
calcd for [C30H36N24O12 + H]+, theoretical: 925.3017; experimental: 925.3023
± 0.0013; calcd for [C30H36N24O12 + Na]+, theoretical: 947.2837; experimental:
947.2825 ± 0.0013.
Dodecapropargylbambus[6]uril
[(propargyl)BU[6]]
Crude (H)BU[6]·HI (200 mg) was dissolved in DMSO
(dried 4 Å MS, 10 mL) and mixed with powdered KOH (0.58 g, 10
mmol). Afterward, propargyl bromide (1.1 mL, 10 mmol) was added, and
the suspension was stirred at RT for 68 h. The product was precipitated
with water (40 mL), collected by filtration, and washed with H2O (2 × 15 mL). After drying in vacuum, the solid was
washed with small portions of MeOH (total 50 mL). The MeOH solution
was evaporated to dryness, and the residue was washed with Et2O (2 × 15 mL). The product was obtained as a dark brown
solid (0.18 g). 1H NMR (500 MHz, CDCl3): δ
5.93 (s, 12H), 5.19 (s, 12H), 4.53 (dd, J = 17.9,
2.3 Hz, 12H), 4.46 (dd, J = 17.9, 2.4 Hz, 12H), 2.14
(t, J = 2.3 Hz, 12H). 13C NMR (126 MHz,
CDCl3): δ 159.26, 159.10, 80.44, 71.85, 67.93, 59.53,
47.93, 34.21, 24.47, 20.09, 13.93. HRMS (MALDI-TOF): calcd for [C66H60N24O12 + H]+, theoretical: 1381.490; experimental: 1381.490 ± 0.002.
(propargyl)BU[6] (10 mg, 6.5 μmol) and dimethyl-5-azidobenzene-1,3-dicarboxylate
(23 mg, 97 μmol) and DIPEA (70 μL, 0.40 mmol) were mixed
in CH3CN (0.9 mL). The mixture was bubbled with Ar for
10 min, and solid CuI (5.3 mg, 28 μmol) was added. The mixture
was stirred for 20 h; afterward, volatiles were evaporated and the
residue was washed with Et2O (5 mL). The crude product
was separated by centrifugation, decanted, and mixed with CH2Cl2 (5 mL). The organic phase was washed with H2O (3 × 10 mL), phases were separated by centrifugation, and
the aqueous part was discarded. The organic solution was azeotropically
dried by coevaporation with CH3CN and dried in vacuum.
The residue of azide was removed from the product by washing with
MeOH and centrifugation. Product 1 was obtained as a
brown solid (16 mg, 54%). 1H NMR signals of compound were
extremely broad before excess of TBAI was added, even when the reaction
was performed already on the I– complex. 1H NMR (500 MHz, CDCl3): δ 8.55 (t, J = 1.6 Hz, 12H), 8.44 (d, J = 1.5 Hz, 24H), 8.18
(s, 12H), 6.45 (s, 12H), 5.31 (s, 16H), 5.12 (d, J = 15.7 Hz, 13H), 4.85 (d, J = 15.7 Hz, 13H), 3.91
(s, 67H). 13C NMR (126 MHz, CDCl3): δ
165.01, 160.89, 159.72, 146.94, 137.52, 132.34, 130.01, 124.83, 121.37,
70.12, 59.40, 52.83, 47.71, 39.96, 24.36, 19.97, 13.85. MS (MALDI-TOF):
calcd for [C186H168N60O60 + H]+, theoretical: 4202.202; experimental: 4202.562.
2,4-Dimethyl-6,8-dipropargylglycoluril
2,4-Dimethylglycoluril
(0.10 g, 0.59 mmol) and small KOH flakes (130 mg, 2.3 mmol) were mixed
with DMSO (dried 4 Å MS, 0.5 mL), and the suspension was stirred
at RT in a closed flask. After 0.5 h, propargyl bromide (80% solution
in toluene, 190 μL, 1.8 mmol) was added. After 22 h, the mixture
was diluted with CH2Cl2 (5 mL) and the organic
phase was washed with H2O (3 × 10 mL). The phase separation
was accelerated by centrifugation. The CH2Cl2 solution was dried with Na2SO4, filtered (desiccant
was washed with a fresh solvent), and evaporated. The oily residue
was washed with Et2O (5 mL) and decanted. Volatile impurities
were removed by coevaporation with toluene (10 mL), and the oily product
was finally evaporated three times from small amount of CHCl3. After drying in vacuum, the product was obtained as brown oil (75
mg, 52%). 1H NMR (500 MHz, CDCl3): δ 5.30
(s, 2H), 4.55 (dd, J = 18.0, 2.6 Hz, 2H), 3.88 (dd, J = 18.0, 2.5 Hz, 2H), 3.01 (s, 6H), 2.36 (t, J = 2.5 Hz, 2H). 13C NMR (126 MHz, CDCl3): δ
159.02, 156.62, 74.08, 69.31, 32.93, 30.65. HRMS (APCI+): calcd for [C12H14N4O2 + H]+, theoretical: 247.1190; experimental: 247.1189;
calcd for [2C12H14N4O2 + H]+, theoretical: 493.2306; experimental: 493.2307.
2,4-Dimethyl-6,8-dipropargylglycoluril (69 mg, 0.28 mmol)
and dimethyl-5-azidobenzene-1,3-dicarboxylate (69 mg, 0.29 mmol) were
dissolved in CH3CN (0.5 mL). DIPEA (54 μL, 0.31 mmol)
was added, and the reaction mixture was bubbled with Ar for 10 min.
Solid CuI (5.3 mg, 28 μmol) was added, and the mixture stirred
for 22 h. Volatiles were evaporated, and the residue was dissolved
in CH2Cl2 (5 mL) and H2O (10 mL +
ammonia solution 0.1 mL, 26% w/w). Phases were separated by centrifugation,
and the aqueous part was discarded. The organic solution was then
washed with 1 M HCl (10 mL) and H2O (10 mL), dried with
Na2SO4, and evaporated to dryness. According
to NMR, 20% of propargyl groups remained and so the reaction was repeated
with the crude mixture and one-half amounts of other reagents. Isolation
was repeated, and after evaporation, the residue was washed with Et2O (2 × 5 mL), collected by centrifugation, and dried
in vacuum. Product 2 was obtained as beige solid (0.11
g, 53%). 1H NMR (500 MHz, CDCl3): δ 8.71
(t, J = 1.4 Hz, 2H), 8.57 (d, J =
1.5 Hz, 4H), 8.18 (s, 2H), 5.28 (s, 2H), 4.91 (d, J = 15.9 Hz, 2H), 4.60 (d, J = 16.0 Hz, 2H), 3.98
(s, 12H), 3.03 (s, 6H). 13C NMR (126 MHz, CDCl3): δ 159.02, 156.62, 74.08, 69.31, 32.93, 30.65. HRMS (APCI+): calcd for [C32H32N10O10 + H]+, theoretical: 717.2376; experimental: 717.2372.