Carbonic acid diester activation by polymer-bound DBU and its relevance to catalytic N-carbonylation of N-heteroaromatics: Direct evidence for an elusive N-carboxy-substituted amidinium cation intermediate

Article abstract of DOI:10.1021/cs400661q

Polymer-bound DBU (PS-DBU, DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene) is an effective and selective catalyst for solventless N-phenoxycarbonylation of N-heteroaromatics (pyrrole (1), indole (2), and carbazole (3)) with diphenyl carbonate (DPC), used as an

Full text of DOI:10.1021/cs400661q

Research Article
pubs.acs.org/acscatalysis
Carbonic Acid Diester Activation by Polymer-Bound DBU and Its
Relevance to Catalytic N‑Carbonylation of N‑Heteroaromatics: Direct
Evidence for an Elusive N‑Carboxy-Substituted Amidinium Cation
Intermediate
Eugenio Quaranta,*,†,‡,§ Antonella Angelini,§ Marianna Carafa,† Angela Dibenedetto,§
and Valentina Mele†
†Dipartimento di Chimica, Universita
̀
degli Studi “Aldo Moro” di Bari, Campus Universitario, Via E. Orabona 4, 70126 Bari, Italy
‡Centro Interdipartimentale di Ricerca su Metodologie e Tecnologie Ambientali (METEA), via Celso Ulpiani 27, 70126 Bari, Italy
§Consorzio Interuniversitario “Reattivita
̀
Chimica e Catalisi”, via Celso Ulpiani 27, 70126 Bari, Italy
S
* Supporting Information
ABSTRACT: Polymer-bound DBU (PS-DBU, DBU = 1,8-diazabicyclo[5.4.0]-
undec-7-ene) is an effective and selective catalyst for solventless N-
phenoxycarbonylation of N-heteroaromatics (pyrrole (1), indole (2), and
carbazole (3)) with diphenyl carbonate (DPC), used as an eco-friendly active
carbonyl species in place of phosgene-derivatives. The immobilized catalyst is
less active than unsupported DBU but can be recovered easily at the end of
catalytic run and recycled effectively. Dedicated studies have demonstrated that
DBU can react as a nucleophile with DPC even when it is anchored on the
polymeric matrix, and provided the first direct evidence (FTIR and solid-state
13C NMR) of formation of a 8-carboxy-substituted 1,8-diazabicyclo[5.4.0]undec-7-enium cation in DBU-promoted nucleophilic
activation of carbonic acid diesters.
KEYWORDS: carbamate, carbonylation, DBU, diphenyl carbonate, N-heteroaromatics, organic carbonate, organocatalysis,
phosgene substitution
chloroformates.35 The developed synthetic methodology does
1. INTRODUCTION
As a part of our studies devoted to exploring new ways of
not require severe temperature conditions (333−433 K,
activation of carbonic acid diesters,1,2 currently obtainable
depending on catalyst load and used substrate) and is
through phosgene-free routes,3−10 and utilizing these com-
solventless, as the organic carbonate, which was reacted in
excess with respect to HetNH, served as reactant and solvent.
pounds as active carbonyl species succedaneous of phosgene in
carbonylation processes,11−18 we have recently reported on a
The excess of DPC can be recovered from the reaction mixture
in high yield.18 However, catalyst recycling was not as effective,
new approach to N-carbonylation of N-heteroaromatics
because DBU catalyst was recovered from reaction mixture in
moderate yield and not in a pure form.
(HetNH) such as pyrrole (1), indole (2), and carbazole (3),
which is based on the use of diphenyl carbonate (DPC) as a
carbonylating agent (eq 1).18 In the presence of catalytic
In the process (eq 1, cat. = DBU), DBU can act not only as a
base catalyst, by activating the HetNH substrate to the relevant
more nucleophilic HetN− anion (Scheme 1),18 but also as a
nucleophile,31,36−41 by activating DPC through the formation
of ketene aminal 7a (Scheme 2).2,18
We have documented experimentally that ketene aminals 7
behave as effective “CO2R” carriers toward NuH pronucleo-
philes (eq 2) and are more active carbonylating agents not only
than the organic carbonate from which they can be generated
but also than the relevant alkoxycarbonyl amidinium cation 8
from which they can plausibly form by deprotonation at C6
(Scheme 3).2,18 DBU itself can act as a proton acceptor, as
amounts of DBU (≥1 mol % vs HetNH), reaction 1 provides a
straightforward selective high-yield entry into heterocarbamates
HetNCO2Ph (4−6)19 through a fully halogen-free synthetic
route, which is a safe ecofriendly alternative to classic protocols
implying the use of phosgene-derivatives,32 such as dicarbona-
tes,23,33a alkylazidoformates,33b 1,1′-carbonyldiimidazole,34 and
Received: August 9, 2013
Revised: October 27, 2013
© XXXX American Chemical Society
195
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202

ACS Catalysis
Research Article
ROC(O)OPh mixtures at the equilibrium with PhOH, the
amidine reactant and the relevant carbonic acid diester
(Scheme 3), cation 8 still remains a spectroscopically elusive
intermediate, in accordance with the fact that 8, once formed
from DBU and the organic carbonate, easily converts into 7 in
the presence of free DBU.2
Scheme 1. Base Catalysis
As well-known, catalyst anchoring on a suitable polymeric
support42 may facilitate recovery of the catalyst from the
reaction mixture and, in principle, may provide a suitable
strategy to trap and detect intermediate species bound at the
polymeric matrix. We have, therefore, focused on commercially
available polystyrene-supported DBU43−46 (PS-DBU, Chart 1)
Scheme 2. Nucleophilic Catalysis
Chart 1. Polymer-Bound DBU (PS-DBU) and its N-
Phenoxycarbonyl Derivatives A and B
as a potential catalyst of reaction 1. Herein, we report on the
catalytic activity of PS-DBU in the carbonylation process (eq
1). We also describe the reactivity of PS-DBU toward DPC,
documenting that the amidine catalyst can nucleophilically
activate the organic carbonate even when it is immobilized on
the polymeric matrix, and, for the first time, provide direct
evidence of the formation of a N-carboxy-substituted
amidinium cation, like A, in the amidine-promoted activation
of carbonic acid diesters.
Scheme 3. Reactivity of DBU with Carbonic Acid Diesters2
2. EXPERIMENTAL SECTION
2.1. General Methods. Unless otherwise stated, all
manipulations were carried out under an inert atmosphere by
using vacuum line techniques. All solvents were dried according
to conventional methods (P2O5; Na/benzophenone)47 and
stored under N2. DPC, 1, 2, and 3 were commercial products. 1
was dried over CaH2, filtered, distilled in vacuo over fresh
CaH2, stored and manipulated under N2. PS-DBU (50−100
mesh; 1.83 mmol N/g; 1% cross-linked with divinylbenzene)
was from Aldrich and was stored under an inert atmosphere to
prevent contamination by atmospheric CO2 and/or mois-
ture.44,48 In the supported catalyst, DBU is anchored through
C6 at the polystyrene matrix (Chart 1). Products 4−6 can be
isolated as described elsewhere,28 after removing the catalyst by
filtration (section 2.3).
supported by the reactivity of a few 8 chloride salts toward the
amidine base. In fact, we have isolated and fully characterized
both salts (8a)Cl and (8b)Cl·H2O (synthesized by reaction of
DBU with the relevant chloroformate), and shown that they
can easily convert, respectively, into 7a and 7b upon reaction
with DBU (Scheme 4).2,41 However, while ketene aminal 7 can
be easily detected spectroscopically (FTIR, NMR) in DBU/
Scheme 4. Synthesis of (8)Cl Salts and Their Reactivity with
DBU2
GC analyses were performed with a HP 5890 Series II gas-
chromatograph (capillary column: Heliflex AT-5, 30 m × 0.25
mm, 0.25 μm film thickness). GC-MS analyses were carried out
with a Shimadzu GC-17A linked to a Shimadzu GC-MS
QP5050 selective mass detector (capillary column: Supelco
MDN-5S, 30 m × 0.25 mm, 0.25 μm film thickness). IR spectra
were taken on a Shimadzu FTIR Prestige 21 spectropho-
tometer or a Perkin-Elmer FTIR 1710 instrument. 13C cross-
polarization magic angle scanning (CP MAS) measurements
were performed at 151 MHz on a Bruker AVANCE 600
apparatus equipped with a MAS II Pneumatic Unit, using total
suppression of sideband (TOSS) technique. Solid sample was
196
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202

ACS Catalysis
Research Article
packed in a 4 mm zirconia rotor with a Kel-F cap and spinned
at 11 kHz during the analysis (delay time = 4 s).
(0.1074 g, 0.90 mmol) was added. The mixture was stirred at
393 K for 3 h, cooled to ambient temperature and suspended in
THF. The suspension was filtered and the solid material was
repeatedly washed on filter with small volumes of THF to
remove unreacted indole, dried in vacuo and analyzed by FTIR
spectroscopy (spectrum b, Figure 8). The filtered solution,
analyzed by GC-MS, contained, in addition to the excess of 2,
also 5, (HetN)2CO (HetN, indolyl; MS (70 eV), m/z 260
[M+], 144 [M+ − HetN], 116 [M+ − HetNCO, (base peak)],
89, 69, 57, 44), PhOH, and DPC.
2.5. Reactivity of DPC-Pretreated PS-DBU with HetNH
(Pyrrole). To 94.3 mg of DPC-pretreated PS-DBU (spectrum
a, Figure 9), isolated as reported in section 2.4, an excess of 1
(0.200 mL, 2.88 mmol) was added. The mixture was stirred at
333 K for 4 h, cooled to ambient temperature and suspended in
THF. The suspension was filtered and the solid material, after
washing with more THF, was dried in vacuo and analyzed by
FTIR spectroscopy (spectrum b, Figure 9). The GC-MS
analysis of the filtered solution showed the formation of 4,
(HetN)2CO (HetN, pyrryl; MS (70 eV), m/z 160 [M+], 94
[M+ − HetN], 66 [M+ − HetNCO], 39),28 PhOH, and DPC.
In an analogous experiment, 46.8 mg of an aged sample of
DPC-pretreated PS DBU (spectrum b, Figure 7) were reacted
with an excess of 1 (0.200 mL, 2.88 mmol), at 333 K for 4 h.
The reaction mixture was worked up as described above. The
solid material isolated by filtration did not show any more
significant absorptions between 1700 and 1750 cm−1 (spectrum
d, Figure 7). Also in this case the GC-MS analysis of the filtered
solution showed the formation of 4, (HetN)2CO (HetN,
pyrryl), PhOH, and DPC.
2.2. PS-DBU Catalyzed Carbonylation of HetNH with
DPC: General Procedure. Typically, into a 30 mL Schlenk
tube, containing the catalyst (PS-DBU),49 DPC and the
substrate (HetNH) were added. The reaction mixture was
heated in an oil bath to the working temperature, allowed to
react for a given time (for further details see Tables 1−4 and
Table 1. Carbonylation of HetNH (1−3) to HetNCO2Ph
(4−6) with DPC in the Presence of PS-DBU Catalyst at 393
K
HetNH
(mmol)
DPC/HetNH
(mol/mol)
catalyst
a
t
HetNH
entry
load (%)
(h) conversion (%)
1
2
3
4
1 (1.20)
2 (1.26)
2 (0.62)
3 (1.13)
3.9
3.7
4.2
4.0
9.8
8.9
9.6
9.9
24
3
90
96
3
98
3
∼100
a
Catalyst load = [mol of supported DBU/mol of HetNH]100.
3. RESULTS AND DISCUSSION
3.1. Carbonylation of 1−3 with Diphenyl Carbonate:
Catalytic Activity of PS-DBU. In this study the activity of PS-
DBU as the catalyst of reaction 1 was investigated under
solventless conditions (see Introduction). At the working
temperature, low-melting DPC (mp = 353 K) acted as reactant
and solvent.
Table 1 shows that PS-DBU is an active catalyst for N-
phenoxycarbonylation of 1−3 with DPC, albeit less active
(Table 2) than unsupported DBU. Under the working
conditions (Table 1), both 2 and 3 were carbonylated
quantitatively and selectively (≥99%) within a short time (3
h), while selective (∼99%) high yield carbonylation of less
reactive pyrrole required longer times (entry 1, Table 1).
Immobilization of the amidine catalyst on the polymeric
support did not produce any significant decrease of carbamate
Figure 1. Reaction of HetNH (3) with DPC in the presence of PS-
DBU: influence of DPC/3 molar ratio.50 Reaction conditions: catalyst
load, 4.7
0.2 mol %; T, 393 K; reaction time, 2 h. Carbamate
selectivity was ≥99% in all the runs.
the legends50 of Figure 1 and 2), and then cooled to ambient
temperature. The solidified mixture was treated with diethyl
ether or THF and the liquid phase analyzed by GC (internal
standard = n-dodecane) or GC-MS.
2.3. Catalyst Recovery and Recycling. At the end of the
catalytic run (section 2.2) the reaction mixture was cooled to
room temperature, suspended in diethyl ether or THF, and
filtered. The catalyst, which was not soluble in the solvent used,
was repeatedly washed with more solvent, and, after drying in
vacuo, was reusable for a new run (Figure 3 and Table 5).
2.4. Reaction of PS-DBU with DPC: Typical Procedure.
A solventless mixture of PS-DBU and DPC (excess) was stirred
at a given temperature (393−443 K) for a measured time,
cooled to room temperature and, then, suspended in diethyl
ether. The suspension was filtered and the solid material was
repeatedly washed on filter with small volumes of diethyl ether,
dried in vacuo and analyzed by spectroscopic methods (Figures
5−7). The GC analysis of the combined mother and washing
solutions showed the formation of phenol. For further details,
see the legends of Figures 5−7.
Table 2. Carbonylation of HetNH (1−3) to HetNCO2Ph
(4−6) with DPC at 393 K: PS-DBU vs DBU Catalytic
Activity
DPC/
HetNH
(mol/mol)
catalyst
load
(%)
HetNH
conversion
(%)
HetNH
entry (mmol)
catalyst
t (h)
a
1
2
3
4
5
6
1 (1.20) PS-DBU
1 (3.60) DBU
2 (1.30) PS-DBU
2 (1.24) DBU
3 (1.20) PS-DBU
3 (1.24) DBU
4.0
3.9
3.8
3.8
3.9
3.8
9.9
8
61
83
56
82
54
87
b
9.3
8
a
9.2
0.5
0.5
0.5
0.5
b
9.7
a
9.2
b
9.2
2.5. Reactivity of DPC-Pretreated PS-DBU with HetNH
(Indole). To 87.2 mg of DPC-pretreated PS-DBU (spectrum a,
Figure 8), isolated as reported in section 2.4, an excess of 2
a
b
[mol of supported DBU/mol of HetNH]100. [mol of DBU/mol of
HetNH]100.
197
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202

ACS Catalysis
Research Article
selectivity relative to unsupported DBU. In fact, ureidic species
(HetN)2CO, if any, formed only in trace amounts.
Table 3). Under analogous experimental conditions, 1, the least
reactive of the substrates investigated, converted into 4 with
very modest yield (entry 7, Table 3). Higher conversions of 1
into 4 were achieved by increasing temperature to 473 K (entry
8, Table 3). Under more moderate temperature conditions,
good to high conversion of pyrrole into N-phenoxycarbonyl
pyrrole required either longer times (entry 1, Table 4) or
higher catalyst loadings (entries 2−4, Table 4).
Using 3 as the reference substrate we have investigated the
influence of a few reaction parameters on the productivity of
the process. Figure 1 illustrates the change of carbazole
conversion against the DPC/HetNH (HetNH = 3) molar ratio,
when the reaction was carried out at 393 K, for 2 h, with a DBU
load close to 5 mol %. The use of a DPC/HetNH molar ratio
higher than 4 did not offer any substantial benefit as it effected
only a very modest variation of substrate conversion to
carbamate. Conversely, substrate conversion diminished
sensibly when using DPC/HetNH molar ratios lower than 4.
The observed trend justifies the use of DPC/HetNH molar
ratios close to 4 in the present work.
Table 4. N-Phenoxycarbonylation of Pyrrole (1) with DPC
Promoted by PS-DBU
HetNH
conversion
(%)
b
1
DFC/1
catalyst
T
(K)
t
a
entry (mmol) (mol/mol) load (%)
(h)
Carbonylation of 3 can be effectively promoted by a catalyst
load as low as 1 mol %. Curve a in Figure 2 illustrates the
1
2
3
4
1.84
1.05
1.22
1.20
4.0
4.0
4.0
3.9
1.0
3.4
3.4
9.8
443
443
443
393
48
24
36
24
77
73
97
90
a
b
[mol of supported DBU/mol of HetNH]100. Selectivity ≥ 99%.
3.1.1. Catalyst Recycling. At the end of the reaction, the
catalyst can be easily recovered quantitatively (section 2.3). The
recovered catalyst, when reused, displayed a catalytic activity
and selectivity comparable with that of fresh PS-DBU, as shown
in Figure 3.
Figure 2. Reaction of HetNH (3) with DPC at 393 K.50 (a) catalyst:
PS-DBU; DPC/3 = 4.0 0.1 mol/mol; catalyst load = 1.0 0.1 mol
%. (b) catalyst: DBU; DPC/3 = 3.9 0.1 mol/mol; catalyst load =1.1
0.1 mol %. Carbamate selectivity: ≥ 99% in all the runs.
conversion of 3 as a function of time, at 393 K, when using a
DPC/HetNH/(supported)DBU molar ratio close to 4:1:0.01.
Under the working conditions, carbazole was carbonylated
selectively (>99%) and with high yield (95%) within a time as
long as 24 h. The carbonylation process proceeded more slowly
than with unsupported DBU (curve b in Figure 2).
At the same temperature (393 K), under otherwise similar
reaction conditions, indole exhibited a behavior close to that of
carbazole (entry 2, Table 3). However, at a markedly lower
temperature, carbonylation of 2 (mp: 326 K) or 3 (mp: 520 K)
was excessively slow (see, for instance, entry 1 in Table 3, for
the specific case of indole). We, therefore, explored the use of
higher temperatures. At 433 K, both 2 and 3 were selectively
and quantitatively carbonylated within 4 h (entry 4 and 6,
Figure 3. Catalytic N-phenoxycarbonylation of 1-3 with DPC: catalyst
recycling. Reaction conditions: 393 K, DPC/HetNH ≈ 4 mol/mol,
catalyst load = 9−10 mol % vs HetNH. Cycles 1−8: Used HetNH was
3, reaction time = 3 h. Cycles 9−12: Used HetNH was 2; cycles 9 and
10, reaction time = 3 h. Cycle 11: Reaction time = 2 h. Cycle 12:
Reaction time =0 .5 h. Cycle 13: Used HetNH was 1, reaction time 24
h. Carbamate selectivity ≳ 99%.
Figure 4 shows the FTIR spectrum of fresh PS-DBU and
those of catalyst recovered from a few catalytic runs (see entry
3 in Table 4 and entry 2 in Table 5). The fresh catalyst
(spectrum a) displays, in addition to the characteristic bands at
696, 754, and around 1600 cm−1 due to the phenyl rings of the
polymeric matrix, a poorly resolved absorption around 1616
cm−1 assigned to CN stretching of the anchored DBU
moieties. The latter band is found at 1618 cm−1 in the FTIR
spectrum of neat DBU. Relative to that of fresh PS-DBU, the
spectra of the recovered catalysts show a few evident alterations
in the region of CO stretching (1700−1750 cm−1) and in the
range 1230−1160 cm−1. The appearance of these new bands,
which are related with a structural modification of the catalytic
system, and the persistence of the absorption around 1616
cm−1 suggest that a few of the anchored DBU molecules may
have undergone a phenoxycarbonylation reaction at the iminic
Table 3. N-Phenoxycarbonylation of HetNH (1−3) with
DPC Promoted by PS-DBU: Influence of Temperature
HetNH
conversion
(%)
b
HetNH DFC/HetNH
catalyst
T
t
a
entry (mmol)
(mol/mol)
load (%) (K)
(h)
1
2
3
4
5
6
7
8
2 (1.22)
2 (2.48)
2 (1.56)
2 (1.22)
3 (1.51)
3 (1.53)
1 (1.80)
1(4.86)
3.9
4.0
4.0
3.9
4.3
3.8
4.0
4.0
0.89
0.97
1.0
343
393
433
433
433
433
433
473
24
24
2
34
87
91
1.3
4
>99
93
1.1
2
0.99
0.99
1.0
4
100
37
24
24
c
81
a
b
[mol of supported DBU/mol of HetNH]100. Selectivity ≥99%.
c
The reaction was carried out in a steel autoclave.
198
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202

ACS Catalysis
Research Article
Figure 4. FTIR spectra (2200−500 cm−1, KBr) of (a) PS-DBU, (b)
catalyst recovered from run 3 in Table 4, and (c) Catalyst Recovered
from Run 2 in Table 5
Figure 5. FTIR spectra (1800−650 cm−1, KBr) of (a) PS-DBU, (b)
PS-DBU pretreated with DPC at 393 K for 2 h (PS-DBU 0.1285 g,
0.116 mmol of supported DBU; DPC 0.5007 g, 2.33 mmol), (c) PS-
DBU pretreated with DPC at 393 K for 16 h (PS-DBU 0.2286 g, 0.208
mmol of supported DBU; DPC 0.900 g, 4.20 mmol), and (d) PS-DBU
pretreated with DPC at 438 K for 12 h (PS-DBU 0.4219 g, 0.384
mmol of supported DBU; DPC 1.6518 g, 7.71 mmol).
Table 5. Catalytic N-Phenoxycarbonylation of pyrrole (1)
a
with DPC
catalyst (g)
b
1
DPC
(mmol)
recovered
catalyst
1 conversion
run (mmol)
PS-DBU
(%)
c
For longer reaction times at the same (393 K) or higher (438
K) temperature an absorption can be noted at higher frequency
(spectrum c and d, respectively, in Figure 5). The latter
absorption falls in the region wherein ν(CO) was found in
(8a)Cl (1764 cm−1, with unresolved shoulder at lower
wavenumbers) and (8b)Cl (1748 cm−1) salts and is compatible
with the formation of (A)OPh units, which are the reasonable
intermediates through which (B) moieties may form. There-
fore, solid state CP MAS 13C NMR spectroscopy was used as a
diagnostic tool to further substantiate the above assignment.
Accordingly, the solid state CP MAS 13C NMR-TOSS
spectrum (Figure 6) of PS-DBU pretreated with DPC at 438
K for 12 h (FTIR spectrum: d in Figure 5) showed a signal at
171.8 ppm,51 which is absent in the spectrum of PS-DBU and
can be assigned to C7 carbon of (A) moieties by comparison
1
2
5.81
5.81
23.2
23.1
0.2151
97
95
d
0.2164
a
Reaction conditions: DPC/HetNH ≈ 4 mol/mol; T = 443 K;
b
c
reaction time = 36 h. Selectivity to carbamate ≥ 99%. Catalyst load
d
([mol of supported DBU/mol of HetNH]100): 3.3 mol %. Mass of
catalyst recovered from run 1 and reused.
N8 atom by DPC. This prompted us to investigate this issue
more deeply, in order to ascertain whether also PS-DBU can
nucleophilically activate the organic carbonate as unsupported
DBU does.
3.2. PS-DBU-Promoted Activation of DPC. PS-DBU was
reacted with an excess of DPC under solventless conditions
comparable with those employed in the catalytic runs. The solid
material isolated as described in section 2.3 was studied both
spectroscopically and from the reactivity point of view.
The FTIR spectrum of the polymeric material isolated after a
reaction time of 2 h at 393 K (spectrum b, Figure 5) shows a
few evident changes between 1150 and 1200 cm−1 and new
ν(CO) bands at 1733 cm−1 (poorly resolved) and around
1718 cm−1. The latter absorptions agree well with those
observed for 7a (ν(CO) = 1717 and 1732 cm−1 (shoulder))
or 7b (ν(CO) = 1702 and 1717 cm−1 (shoulder))2 and
indicate that, likewise unsupported DBU, also polymer-
anchored DBU can be N-phenoxycarbonylated by reaction
with DPC (eq 3). The reaction cogenerates phenol, which was
detected by GC and GC-MS.
Figure 6. CP MAS 13C NMR-TOSS spectrum (11 kHz) of (a) PS-
DBU and (b) PS-DBU pretreated with DPC at 438 K for 12 h (see
also IR spectrum d in Figure 5).
199
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202

ACS Catalysis
Research Article
with the 13C NMR (100 MHz, CDCl3) spectrum of (8a)Cl (δ
= 20.85, 21.54, 24.86, 27.85, 34.04 (C6), 44.56, 52.36, 57.84,
120.72 (C-ortho), 126.88 (C-para), 129.64 (C-meta), 149.61
(C-ipso), 150.89 (COO), 172.62 (C7) ppm).2 The above
spectroscopic results provide the first sound experimental
evidence of the formation of a N-carboxy-substituted
amidinium intermediate, like A, in the reaction of DBU with
an organic carbonate.
Inspection of Figure 5 shows that A moieties can accumulate
on the polymeric matrix for longer reaction times. This suggests
that conversion of A into B moieties may become less and less
facile as reaction 3 proceeds. Nevertheless, A → B conversion
can occur, albeit very slowly, even at ambient temperature, as it
can be inferred by comparing the IR spectrum of a freshly
isolated sample of DPC-pretreated PS-DBU (spectrum a,
Figure 7; bands at 1750 and 1719 cm−1 due to A and B groups,
respectively) with that of the same material aged for a few
months at room temperature (spectrum b, Figure 7), which, in
the carbonyl region, does not show any more the band at 1750
cm−1, but exhibits only absorptions in the range 1710 −1734
cm−1 assignable to B groups.
Figure 8. FTIR spectra (1800−650 cm−1, KBr) of (a) PS-DBU
pretreated with DPC at 393 K for 16 h (PS-DBU 0.2286 g, 0.208
mmol of supported DBU; DPC 0.900 g, 4.20 mmol), (b) solid
material isolated after reacting DPC-pretreated PS-DBU (see a) with
indole at 393 K for 3 h (section 2.5), and (c) PS-DBU.
7 and 9). In this case, the GC-MS analysis of the reaction
mixtures showed the formation of 4, in addition to 1,1′-
carbonyldipyrrole, PhOH and DPC.
Figure 7. (a) FTIR spectrum (1850−650 cm−1, KBr) of a freshly
isolated sample of PS-DBU pretreated with DPC at 443 K for 12 h
(PS-DBU 0.271 g, 0.244 mmol of supported DBU; DPC 1.046 g, 4.88
mmol); (b) FTIR spectrum (KBr) of the same material (see above a),
after aging for a few months at r.t.; (c) FTIR spectrum (KBr) of the
solid material isolated after reacting aged DPC-pretreated PS-DBU
(see b) with pyrrole at 333 K for 4 h (section 2.6); and (d) FTIR
spectrum (KBr) of PS-DBU.
Figure 9. FTIR spectra (1850−600 cm−1, KBr) of (a) PS-DBU
pretreated with DPC at 438 K for 12 h (PS-DBU 0.4219 g, 0.384
mmol of supported DBU; DPC 1.6518 g, 7.71 mmol) and (b) solid
material isolated after reacting DPC-pretreated PS-DBU (see a) with
pyrrole at 333 K for 4 h (section 2.6).
The reactivity of DPC-pretreated PS-DBU with HetNH (1,
2) is consistent with the above-described chemistry.
For instance, DPC-pretreated PS-DBU (spectrum b, Figure
8) was reacted with an excess of 2 at 393 K for 3 h (section
2.5). The solid material isolated by filtration after suspending
the reaction mixture in THF did not show any more significant
absorptions in the carbonyl region (spectrum a, Figure 8) and,
furthermore, was practically superimposable with that of PS-
DBU (spectrum c, Figure 8). Moreover, the GC-MS analysis of
the THF solution showed the formation of 5, together with
(HetN)2CO (HetN, indolyl), phenol, and DPC.
Reaction 4, implying direct N-carbonylation of HetNH by
(B)-moieties, as well as reaction 5 (which may be promoted by
Analogous experiments were carried out using pyrrole in
place of indole, but under milder conditions (333 K, 4 h, Figure
200
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202

ACS Catalysis
Research Article
supported DBU molecules28) and the reactivity summarized in
Scheme 3, describe plausible ways to the formation of the
observed products. We note that reaction 4 is reminiscent of
step (II) in Scheme 2 and finds sound experimental support in
reaction 2, which we have documented elsewhere.2,18
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: eugenio.quaranta@uniba.it. Phone: +39 080 5442100.
Notes
The authors declare no competing financial interest.
HetNC(O)OPh + HetNH
(PS−DBU)
ACKNOWLEDGMENTS
■
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ (HetN)2CO + PhOH
(5)
This work was supported by Universita
Moro”, Bari (Fondi di Ateneo).
̀
degli Studi “Aldo
Figure 9 deserves an additional comment. In the spectrum of
the solid isolated after reaction with pyrrole (spectrum b) the
relative intensities of the residual absorptions assigned to A
(1751 cm−1) and B moieties (1700−1734 cm−1) changed with
respect to those observed in the starting material (spectrum a),
suggesting that, under the mild (333 K) working conditions
used, B residues react more easily than A moieties. This agrees
well with our previous findings showing that ketene aminal 7 is
a more active carbonylating agent of pyrrole than the
corresponding N-alkoxycarbonyl amidinium cation (8).2,18
As a whole, the body of the above results demonstrates
unambiguously that also PS-DBU, likewise unsupported DBU,
can nucleophilically activate the organic carbonate and can act
as a nucleophile catalyst in reaction 1 (see Scheme 2).
However, considering the well-known ability of PS-DBU to act
as a proton acceptor,42 we cannot exclude, also in the present
case, the co-occurrence of a basic mechanism involving the
activation of HetNH substrate, as described in Scheme 1 for the
unsupported catalyst. It is reasonable to expect that DBU
anchorage at the polymeric matrix may hamper, for steric
reasons,42−45 the nucleophilic pathway (Scheme 2) more
markedly than the proton transfer reaction between the HetNH
substrate and the immobilized base. This may offer a plausible
rationale for the lower activity exhibited by PS-DBU compared
to the unsupported catalyst.
REFERENCES
■
(1) Distaso, M.; Quaranta, E. J. Catal. 2008, 253, 278−288.
(2) Carafa, M.; Mesto, E.; Quaranta, E. Eur. J. Org. Chem. 2011,
2458−2465.
(3) Cshihony, S.; Mika, L. T.; Vlad, G.; Barta, K.; Mehnert, C. P.;
Horvath, I. T. Collect. Czech. Chem. Commun. 2007, 72, 1094−1106.
(4) Ballivet-Tkatchenko, D.; Bernard, F.; Demoisson, F.; Plasseraud,
L.; Sanapureddy, S. R. ChemSusChem 2011, 4, 1316−1322.
(5) Gong, J.; Ma, X.; Wang, S. Appl. Catal., A 2007, 316, 1−21.
(6) Keller, N.; Rebmann, G.; Keller, V. J. Mol. Catal. A 2010, 317, 1−
18.
(7) Dibenedetto, A.; Aresta, M.; Angelini, A.; Ethiraj, J.; Aresta, M. B.
Chem.Eur. J. 2012, 18, 10324−10334.
(8) Leino, E.; Maki-Arvela, P.; Eta, V.; Murzin, D.Yu.; Salmi, T.;
̈
Mikkola, J.-P. Appl. Catal., A 2010, 383, 1−13.
(9) Honda, M.; Kuno, S.; Sonehara, S.; Fujimoto, K.; Suzuki, K.;
Nakagawa, Y.; Tomishige, K. ChemCatChem 2011, 3, 365−370.
(10) Tian, D.; Liu, B.; Gan, Q.; Li, H.; Darensbourg, D. J. ACS Catal.
2012, 2, 2029−2035.
(11) Carafa, M.; Quaranta, E. Mini-Rev. Org. Chem. 2009, 6, 168−183
and references therein.
(12) Chaturvedi, D.; Chaturvedi, A. K.; Mishra, V. Curr. Org. Chem.
2012, 16, 1609−1635.
(13) Juarez, R.; Corma, A.; Garcia, H. Pure Appl. Chem. 2012, 84,
685−694.
(14) Grego, S.; Arico, F.; Tundo, P. Org. Process Res. Dev. 2013, 17,
679−683.
4. CONCLUSIONS
The activity of polymer-bound DBU as catalyst for carbon-
ylation of N-heteroaromatics, such as pyrrole, indole and
carbazole, with diphenyl carbonate has been investigated under
solventless conditions. PS-DBU is an efficient and selective
catalyst for N-phenoxycarbonylation of 1−3 but is less active
than unsupported DBU. However, immobilization of DBU
allows the quantitative recovery of the catalytic system which
can be reused without any substantial decrease both of activity
and selectivity. The easy quantitative recovery of the catalyst, as
well as its effective reutilization, markedly improve the
greenness of the synthetic method, and make profitable the
use of more massive catalyst loadings, which allow to conjugate
the achievement of very high conversions within reasonable
reaction times with more moderate energy inputs.
Ad hoc studies have demonstrated that DBU can nucleophili-
cally activate DPC even when the amidine base is supported on
the polymeric matrix and provided, for the first time, direct
evidence of the intermediacy of a N-carboxy-substituted
amidinium cation in the DBU-promoted activation of a
carbonic acid diester.
(15) Selva, M.; Noe, M.; Perosa, A.; Gottardo, M. Org. Biomol. Chem.
2012, 10, 6569−6578.
(16) Distaso, M.; Quaranta, E. J. Catal. 2004, 228, 36−42.
(17) Distaso, M.; Quaranta, E. Appl. Catal., B 2006, 66, 72−80.
(18) Carafa, M.; Iannone, F.; Mele, V.; Quaranta, E. Green Chem.
2012, 14, 3377−3385.
(19) HetNCO2R (R = alkyl, aryl) carbamates are important synthetic
intermediates for the preparation of a variety of chemicals, including
pharmaceutically relevant substances (epibatidine and derivatives;
epiboxidine and other 7-azabicyclo[2.2.1]heptanes), biologically active
compounds (alkaloids) and more complex heterocyclic systems.20−26
Under suitable conditions, N-phenoxycarbonyl derivatives HetN-
CO2Ph (HetNH = 1−3) can act as carbonylating agents and are
27
converted easily into unsymmetrical ureas HetNC(O)NR2 or, by
transesterification with alcohols, into HetNCO2R compounds,28 which
are useful precursors of the corresponding N-alkyl derivatives HetNR
(R = alkyl).29−31
(20) Natsume, M.; Muratake, H. Tetrahedron Lett. 1979, 36, 3477−
3480.
(21) Seerden, J.-P. G.; Tulp, M. Th. M.; Scheeren, H. W.; Kruse, C.
G. Bioorg. Med. Chem. 1998, 6, 2103−2110.
(22) Pandey, G.; Tiwari, S. K.; Singh, R. S.; Mali, R. S. Tetrahedron
Lett. 2001, 42, 3947−3949.
(23) Paparin, J.-L.; Crev
3909−3918.
́ ́
isy, C.; Gree, L. Eur. J. Org. Chem. 2000,
ASSOCIATED CONTENT
■
S
* Supporting Information
(24) Oldroyd, D. L.; Payne, N. C.; Vittal, J. J.; Weedon, A. C.; Zhang,
B. Tetrahedron Lett. 1993, 34, 1087−1090.
(25) Hastings, D. J.; Weedon, A. C. Can. J. Chem. 1991, 69, 1171−
1181.
Spectroscopic characterization of compounds 4−6. This
material is available free of charge via the Internet at http://
pubs.acs.org.
201
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202

ACS Catalysis
Research Article
(26) Albini, F. M.; Albini, E.; Bandiera, T.; Caramella, P. J. Chem. Res.,
Synop. 1984, 36−37.
PS-DBU, the resonance at 158.8 ppm is assigned to C7 of unreacted
polymer-bound DBU molecules, the presence of which is also
supported by the absorption around 1616 cm−1 (section 3.1.1) in
the FTIR spectrum of the isolated material. The strong signals at
145.9, 128.2, ∼50, 41.2, and 40.6 ppm are due to the carbons
(aromatic and aliphatic) of the polymeric matrix (PS), while the
resonances at 65.9 and 15.3 ppm (masking signals formerly present in
the spectrum of PS-DBU) may be due to residual amounts of diethyl
ether, used in the workup of the reaction mixture (section 2.4).
(27) Carafa, M.; Mele, V.; Quaranta, E. Green Chem. 2012, 14, 217−
225.
(28) Carafa, M.; Distaso, M.; Mele, V.; Trani, F.; Quaranta, E.
Tetrahedron Lett. 2008, 49, 3691−3696.
(29) Becherer, J.; Koch, P.; Cassella, A. G. German Patent DE
4,324,707, 1994.
(30) Shieh, W.-C.; Dell, S.; Bach, A.; Repic, O.; Blacklock, T. J. Org.
Chem. 2003, 68, 1954−1957.
(31) Quaranta, E.; Carafa, M.; Trani, F. Appl. Catal., B 2009, 91,
380−388.
(32) Cotarca, L.; Eckert, H. Phosgenations: A Handbook; Wiley-VCH:
Weinheim, Germany, 2004.
(33) (a) Grehn, L.; Ragnarsson, U. Angew. Chem., Int. Ed. 1984, 23,
296−301. (b) Carpino, L. A.; Barr, D. E. J. Org. Chem. 1966, 31, 764−
767.
(34) Macor, J. E.; Cuff, A.; Cornelius, L. Tetrahedron Lett. 1999, 40,
2733−2736.
(35) (a) Handy, S. T.; Sabatini, J. J.; Zhang, Y.; Vulfova, I.
Tetrahedron Lett. 2004, 45, 5057−5060. (b) Wang, N. C.; Anderson,
H. J. Can. J. Chem. 1977, 55, 4103−4111. (c) Gabel, N. J. Org. Chem.
1962, 27, 301−303. (d) Acheson, R. M.; Vernon, J. M. J. Chem. Soc.
1961, 457−459. (e) Jacquemard, U.; Ben
́ ́
eteau, V.; Lefoix, M.; Routier,
S.; Mer
́
our, J.-Y.; Coudert, G. Tetrahedron 2004, 60, 10039−10047.
(f) Illi, V. O. Synthesis 1979, 397−388. (g) Itahara, T. Heterocycles
1986, 24, 2557−2562.
(36) Ishikawa, T.; Kumamoto, T. In Amidines in Organic Synthesis, in
Superbases for Organic Synthesis: Guanidines, Amidines, Phosphazenes
and Related Organocatalysts; Ishikawa, T., Ed.; John Wiley & Sons, Ltd:
Chichester, U.K., 2009; pp 49−92.
(37) Larriv e-Aboussafy, C.; Jones, B. P.; Price, K. E.; Hardink, M. A.;
McLaughlin, R. W.; Lillie, B. M.; Hawkins, J. M.; Vaidyanathan, R. Org.
Lett. 2010, 12, 324−327.
(38) Aggarwal, V. K.; Mereu, A. Chem. Commun. 1999, 2311−2312.
(39) Heller, S. T.; Schultz, E. E.; Sarpong, R. Angew. Chem., Int. Ed.
2012, 51, 8304 −8308.
(40) Shieh, W.-C.; Dell, S.; Repic, O.; Blacklock, T. J. Org. Chem.
2002, 67, 2188−2191.
(41) Mesto, E.; Quaranta, E. Acta Crystallogr. 2013, C69, 444−447.
(42) (a) Haag, R.; Roller, S. Top. Curr. Chem. 2004, 242, 1−42.
(b) Dioos, B. M. L.; Vankelekom, I. F. J.; Jacobs, P. A. Adv. Synth.
Catal. 2006, 348, 1413−1446.
(43) (a) Tomoi, M.; Kato, Y.; Kakiuchi, H. Makromol. Chem. 1984,
185, 2117−2124. (b) Iijima, K.; Fukuda, W.; Tomoi, M. Pure Appl.
Chem. 1992, A29 (3), 249−261.
(44) Ochiai, B.; Yokota, K.; Fujii, A.; Nagai, D.; Endo, T.
Macromolecules 2008, 41, 1229−1236.
(45) Tamura, Y.; Fukuda, W.; Tomoi, M. Synth. Commun. 1994, 24,
2907−2914.
(46) Chiara, J. L.; Encinas, L.; Diaz, B. Tetrahedron Lett. 2005, 46,
2445−2448.
(47) Perrin, D. D., Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals; Pergamon: Oxford, U.K., 1986.
(48) Perez, E. R.; Santos, R. H. A.; Gambardella, M. T. P.; de
Macedo, L. G. M.; Rodrigues-Filho, U. P.; Launay, J. C.; Franco, D. W.
J. Org. Chem. 2004, 69, 8005−8011.
(49) mmol of supported DBU = [(m(g)PS‑DBU)•(1.83 mmol N/g)]/
2.
(50) The experimental points in Figure 1 and 2 have been obtained
according to the protocol described in section 2.2. The legends of the
figures report the mean values of the DPC/HetNH molar ratio or the
catalyst load ([mol of (supported) DBU/mol of HetNH]100).
(51) The signal at 151.8 ppm (Figure 6) can be attributed to
carbonyl carbon of A or B moieties. Herein, we report, as a
comparison, also the 13C NMR (100 MHz, CDCl3) spectrum of 7a: δ
= 23.38, 24.46, 25.88, 28.26, 43.75, 48.97, 52.88, 106.04 (C6), 121.31
(C-ortho), 124.71 (C-para), 128.76 (C-meta), 142.83 (C7), 151.25 (C-
ipso), 153.18 (COO) ppm.2 By comparison with the 13C spectrum of
202
dx.doi.org/10.1021/cs400661q | ACS Catal. 2014, 4, 195−202