Study of the carbon dioxide chemical fixation-activation by guanidines

Article abstract of DOI:10.1016/j.tet.2008.08.008

Fixation of CO₂ is one of the most important priorities of the scientific community dedicated to reduce global warming. In this work, we propose new methods for the fixation of CO₂ using the guanidine bases tetramethylguanidine (TMG) and 1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]-pyrimidine (TBD). In order to understand the reactions occurring during the CO₂ fixation and release processes, we employed several experimental methods, including solution and solid-state NMR, FTIR, and coupled TGA-FTIR. Quantum mechanical NMR calculations were also carried out. Based on the results obtained, we concluded that CO₂ fixation with both TMG and TBD guanidines is a kinetically reversible process, and the corresponding fixation products have proved to be useful as transcarboxylating compounds. Afterward, CO₂ thermal releasing from this fixation product with TBD was found to be an interesting process for CO₂ capture and isolation purposes.

Full text of DOI:10.1016/j.tet.2008.08.008

Tetrahedron 64 (2008) 10097–10106
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Study of the carbon dioxide chemical fixationdactivation by guanidines
a
b
c
b
´
´
Fernanda Stuani Pereira , Eduardo Ribeiro deAzevedo , Eirik F. da Silva , Tito Jose Bonagamba ,
a
b
a
a,
*
´
´
˜
´
Deuber L. da Silva Agostıni , Alvicler Magalhaes , Aldo Eloizo Job , Eduardo R. Perez Gonzalez
a
´
´
ˆ
Departamento de Fısica, Quımica e Biologia, Faculdade de Ciencias e Tecnologia, UNESP, C.P. 467, Presidente Prudente, 19060-080 SP, Brazil
b Instituto de F´ısica de S˜ao Carlos, Universidade de S˜ao Paulo, C.P. 369, S˜ao Carlos, 13560-970 SP, Brazil
c SINTEF Materials and Chemistry, R. Birkelands vei 2B, NO-7465 Trondheim, Norway
a r t i c l e i n f o
a b s t r a c t
Article history:
Fixation of CO2 is one of the most important priorities of the scientific community dedicated to reduce
global warming. In this work, we propose new methods for the fixation of CO2 using the guanidine bases
tetramethylguanidine (TMG) and 1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]-pyrimidine (TBD). In order
to understand the reactions occurring during the CO2 fixation and release processes, we employed
several experimental methods, including solution and solid-state NMR, FTIR, and coupled TGA–FTIR.
Quantum mechanical NMR calculations were also carried out. Based on the results obtained, we con-
cluded that CO2 fixation with both TMG and TBD guanidines is a kinetically reversible process, and the
corresponding fixation products have proved to be useful as transcarboxylating compounds. Afterward,
CO2 thermal releasing from this fixation product with TBD was found to be an interesting process for CO2
capture and isolation purposes.
Received 22 July 2008
Accepted 1 August 2008
Available online 9 August 2008
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
a carbamic (zwitterionic) complex DBU–CO2 from the reaction of
CO2 with 1,8-diazabicycloundecene (DBU) in fresh acetonitrile has
There is at present great concern surrounding anthropogenic CO2
emissions.1 One of the options actively being pursued to reduce CO2
emissions is Carbon Capture and Storage technology (CCS).2 A key
step in many CCS technologies is reversible binding of the CO2
molecule. Identification of new mechanisms and compounds for
reversibly binding CO2 is therefore of considerable interest. One of
the goals of the present work is to explore new compounds that
may bind CO2.
While CO2 presents an environmental issue as a greenhouse
gas, it is also environmentally benign in comparison with many
other chemical substrates. The development of suitable methods,
at laboratory scale for the preparation of interesting CO2 con-
taining compounds, like organic carbonates and urethanes, by
respective reaction of alcohols3 or amines4 with CO2, could be an
alternative to avoid the use of the highly toxic phosgene and its
derivatives (e.g., chloroformates).5,6 The use of phosgene requires
secure conditions.6 For this reason, industries that prepare pesti-
cides of the carbamate class using phosgene in the process need
special adjacent storage areas to avoid transportation and long-
term stock.
recently been proposed.7a In the same work, the transcarboxylating
activity of this adduct toward some amines to yield ethyl carba-
mates upon O-alkylation with ethyl iodide has also been demon-
strated. Subsequently, more detailed theoretical and experimental
studies of the CO2-fixation products with DBU and pentam-
ethyldiazabicyclodecene (PMDBD) amidines by solution NMR,
thermogravimetric analyses (TGA), and X-ray diffraction methods
have shown the formation of other products characterized as
dimeric crystalline bicarbonates.7b
More recently, the formation of carbamic intermediates by CO2
reaction with amidines and pyridinic bases and the participation
(intermediates or catalysts) of these carbamic adducts in reactions
involving carbon dioxide has been also proposed.8,9 The formation
of a zwitterionic carbamic complex by the reaction of CO2 with DBU
was investigated and this class of product was not detected.10
However, the possible formation (traces) of the carmabic product
was observed. On the other hand, bicarbonates formed as CO2-
fixation products have been found to be interesting products (e.g.,
ionic solvents)11 or useful intermediates as transcarboxylating
compounds.7b,12
Nature of CO2 fixation by nitrogenated bases. CO2 transfer by
transcarboxylation of amines. A new process for the formation of
In the present work, we describe a study of carbon dioxide
fixation and release by guanidines and the nature of the guanidine–
CO2 complexes. Fixation of carbon dioxide can be a reversible
process under moderate conditions (low or moderate temperature,
normal pressure, etc.). At the same time, releasing CO2 can involve
a transcarboxylation process leading to several nucleophiles.7 For
* Corresponding author. Tel.: þ55 18 3229 5355; fax: þ55 18 3221 5682.
´
´
E-mail address: eperez@fct.unesp.br (E.R. Perez Gonzalez).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.08.008

10098
F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
O
C
O
C
NH
O
HN
NH2
N(CH3)2
N
O TMGH
N(CH3)2
CO2/ACN
+
+
(CH3)2 N
N(CH3)2
(CH3)2 N
N(CH3)2
(CH3)2 N
{H2O traces}
(CH3)2 N
TMG (tetramethyl guanidine)
TMG-CO2
Ia
TMGH HCO3
Ib
TMG-CO2
Ic
I
Scheme 1.
Table 1
synthesis purpose, a CO2 fixation-release occurring as a reaction
process (transcarboxylation) is more interesting because it is a ki-
netically reversible fixation. On the other hand, for other purposes,
such as industrial uses, a more thermodynamically stable fixation
could be more suitable with the CO2 being fixed and released as the
unique product.
Fixation of CO2 with the guanidine bases TMG and TBD. The
structures proposed for products formed by the fixation of CO2 with
the guanidine TMG I are illustrated in Scheme 1. The carbamic and
the bicarbonate products are represented by structures Ia–Ic.
It is possible to speculate that three products could be expected
from the reaction of the guanidine TBD represented by the struc-
ture II with carbon dioxide, two tautomers presenting carbamic
groups as shown in IIa and IIb and bicarbonate represented by IIc.
All representative structures are depicted in Scheme 2.
Gas phase 13C NMR chemical shifts calculated at B3LYP/6-311þþG(d,p) level of
theory (unit is ppm)
C1a
C2–C5b
CO2/HCO3
TMG
TMGH
TMG–CO2 (Ia)
TMG–COꢀ2 (Ic)c
TMGH–HCO3(1)d (Ib)
TMGH–HCO3(2)d (Ib)
173.7
166.2
168.2
152.3
175.1
173.4
41.0 (38.9, 42.5, 41.8, 40.9)
41.3 (39.8, 42.7, 42.7, 39.8)
40.8 (42.4, 41.5, 38.8, 40.7)
41.6 (42.2, 42.6, 41.7, 40.1)
40.9 (39.4, 40.7, 42.8, 40.5)
40.5 (40.1, 41.6, 42.1, 38.5)
145.4
167.9
168.2
164.4
C4e
C1e
C2 e
C3e
CO2/HCO3
TBD
TBDH
TBD–CO2 (IIa)
TBD–CO2 (IIb)
TBDH–HCO3 (IIc)
152.9
155.0
159.1
153.6
155.5
48.9, 44.4
43.9,43.9
42.9, 41.1
49.9, 49.2
40.9, 40.9
27.2, 28.7
24.1, 24.1
25.7, 25.7
24.3, 21.4
25.9, 25.8
51.6, 52.4
50.9, 50.9
53.0, 51.0
47.8, 51.0
50.8, 50.8
151.6
148.4
168.8
a
Carbon bond to all nitrogens.
CH3 groups. Average shown, followed in parenthesis by values for each atom.
Species not stable in gas phase, results obtained with N–C(CO2) bond
b
c
N
constrained to 1.6 Å.
N
O
N
O
CO2/ACN
d
N
N
H
N
H
Two different bicarbonate geometries were studied, geometries shown in
+
+
Supplementary data.
{H2O traces}
N
N
N
H
N
C
OH
N
H
N
e
H
O
O
Groups numbered as in Figure 3a.
O
O
TBD (tri-azabicycledecene)
II
TBD-CO2
IIa
TBD-CO2
IIb
TBDH HCO3
IIc
3. Results and discussion
Scheme 2.
Fixation of CO2 with TMG and TBD guanidines. The reactions were
carried out under two sets of conditions, in the presence or absence
of organic solvent (fresh acetonitrile). This was not difficult for
TMG, but for TBD, previously melted TBD was not found to be an
efficient method for CO2 fixation. Thus, a minimal amount of ace-
tonitrile was added to ensure a better contact of the CO2 with the
guanidine. In the case of TMG, no significant differences were ob-
served under solvent or dry conditions. Hence, fixation of CO2 with
TMG can be proposed as a ‘green’ fixation-activation procedure
with a high yield of the fixation products (90%). TBD–CO2 products
were obtained in acetonitrile with an 85% yield. Both reactions
were carried out under normal pressure CO2 (10 mL/min) and
5 mmol of guanidine was used. Reaction was carried out without
any special experimental requirements. Solvent was not dried in
order to achieve the CO2 fixation by an easy and low cost. Products
were characterized without any previous purification.
2. Quantum mechanical calculations
All chemical species were optimized in vacuum at B3LYP/
6-311þþG(d,p) level of theory.11 The NMR nuclear shielding was
computed by using gauge-invariant atomic orbitals (GIAOs) at this
level of theory. The chemical shifts were calculated with tetrame-
thylsilane as reference. To test the sensitivity of the results to the
level of theory some calculations were carried out at the
MPW1PW91 level. This method has been reported to perform well
in 13C NMR calculations.13
To attempt to capture effects of the environment on the chem-
ical shifts and reaction energies some calculations were carried
out with the IEFPCM continuum solvation model.14 We carried
out calculations with the acetonitrile parameters in IEFPCM. All
quantum mechanical calculations were performed in Gaussian 03.15
In Table 1, we show the calculated 13C NMR chemical shifts for
different TMG and TBD species at B3LYP level of theory. Results
with acetonitrile as solvent and at the MPW1PW91 level of theory
(shown in Supplementary data) did not differ at a qualitative level
from the gas phase B3LYP results.
From B3LYP level energy calculations all reactions were found to
be endothermic (results shown in Supplementary data). While
these results are qualitative in nature they do suggest that the CO2
bound complexes are not stable in vacuum. This would again sug-
gest that the species formed are to some extent stabilized by
intramolecular bonding. In gas phase it was found that IIb was
27 kcal/mol less stable than IIa, suggesting that this species is
unlikely to be formed in significant amounts.
3.1. NMR analysis
13C and 1H NMR of the TMG–CO2. In the first attempt to charac-
terize the reaction, products of the CO2 fixation by TMG, solution 1H
and 13C NMR spectra were acquired. However, the TMG–CO2 re-
action product was not completely soluble in CDCl3 and a white
powder precipitation was observed. In addition, free CO2 was re-
leased from the solution. Despite that 1H and 13C NMR experiments
were carried out in the liquid phase to characterize the soluble part
of the product, resulting in the spectra shown in Figure 1a and b.
These spectra can be assigned considering the TMG structure
(without the CO2 fixation) as shown in Figure 1a. Therefore, these
results indicate that interaction with the solvent produced the

F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
10099
carbons in NCH3 groups, while the line in the 160–163 ppm region
could be either from guanidinic carbon or from carbonyl carbons in
HCOꢀ3 groups. Some results for various bicarbonate structures have
shown that the typical chemical shift value of the carbonyl carbons
in HCOꢀ3 groups is about 162 ppm.16 Thus, the line at 162.2 ppm (4)
can be attributed to this group, while the line at 161.3 ppm is
assigned to the guanidine carbon (3). The broader line width ob-
served for the 161.3 ppm line is further a evidence that it can be
1
attributed to the guanidine carbon. It is well known that spin ⁄2
nuclei in a non-symmetric environment attached to quadrupole
moments (such as 14N) present a strong non-homogeneous line
broadening in the solid state, which is consistent with the spectrum
observed in Figure 1c. These results should be understood bearing
in mind that the ionic bicarbonate structure Ib does not dissolve in
CDCl3, while the carbamate Ic and zwitterionic carbamate Ia
dissolves and, because they are not stable in the apolar media, CO2
is released.
Comparing the experimental spectrum with the calculated
values in Table 1, we see that the calculations predict different
relative positions between the HCOꢀ3 carbon and the guanidine
carbon (3). The experimental shift for the guanidine carbon is,
however, consistent with the calculated value for protonated TMG.
This would suggest that the TMGH–HCOꢀ3 geometry utilized in the
calculations is not representative of the geometry of the complex in
experiment.
Although the solution spectra were quite elucidating in terms of
understanding the products formation, the main interest in this
work is to characterize them as they were obtained, i.e., in the solid
phase. The 13C CP/MAS spectrum of the TMG–CO2 solid product is
shown in Figure 2a. In order to avoid chemical decomposition after
the preparation, the product was kept at ꢀ30 ꢁC, which is also the
measuring temperature. As it can be observed, the spectrum is
composed of six distinguishable lines, with chemical shift values of
39.5 (1,2), 41.2 (5), 43.1 (6), 153.1 (8), 159.1 (7), 161.7 (3), and 162.3
(4) ppm. Taking into consideration the evidence that two structures
were formed, see Scheme 1, one can evaluate the consistency of
the spectrum based on such structures. The NMR lines within the
39–44 ppm region have typical chemical shift values of methyl
carbons in NCH3 groups, while the lines at 150–165 ppm region can
Figure 1. Solution 13C (a) and 1H (b) NMR spectra of the CDCl3 soluble portion of the
TMG–CO2 reaction product (only TMG is detected; structure I in Scheme 1). Both
spectra (a and b) are not quantitative because they were not performed in the fully
relaxed conditions. (c) 13C CP/MAS spectrum of the non-CDCl3 soluble portion of the
TMG–CO2 reaction product (represented by Ib in Scheme 1). The inset shows a zoom of
the 158–166 ppm region. Spinning side bands are indicated by asterisk (
spectrum.
*
) in the
dissociation of one of the reaction products resulting in the release
of CO2 while the TMG remained dissolved in CDCl3. However, there
was also the precipitation of another reaction product, which was
characterized using solid-state 13C Cross-Polarization Magic-Angle
Spinning (13C CP/MAS) NMR experiments (Fig.1c). Four lines at 38.9
(1), 40.5 (2), 161.3 (3), and 162.2 ppm (4) can be observed in
the spectrum shown in Figure 1c. The 13C NMR lines within the
39–44 ppm region have typical chemical shift values of methyl
Figure 2. 13C CP/MAS spectra of TMG–CO2 at ꢀ30 ꢁC. (a) Spectrum of the reaction
mixture at the end of the reaction (Ia and Ib in Scheme 1). The inset shows a zoom of
the 150–170 ppm region. (b) Spectrum after 30 min of drying under reduced pressure.
The inset shows a zoom of the 150–170 ppm region. Spinning side bands are indicated
by asterisk ( ) in the spectra.
*

10100
F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
be either from guanidinic carbon or from carbonyl carbons in the
HCOꢀ3 groups. From the proposed chemical structures shown in
Scheme 1 several non-NMR equivalent NCH3 should be expect (the
methyl groups may be non-equivalent due to restricted confor-
mational rotation on the NMR time scale, with one pair of methyl
groups being closer to the NCOOꢀ group than the other). Since the
main effect of the positive charge is to change the CH3 carbons to
higher parts per million values, it is possible to attribute it to the
CH3 lines as indicated in Figure 2a. The assignment of these groups
as methyl groups was confirmed by measuring the local residual CH
dipolar coupling using 1D-DIPSHIFT experiments (not shown).17
The measured values were 7.3 kHz for the three carbons, which is
typical of CH3 end groups that execute a fast rotation around the
local C3 axis. Based on the previous discussion, the line at
162.3 ppm can be attributed to the HCOꢀ3 carbon in structure Ib. The
assignment of the remaining lines can be made considering that
chemical shift for non-charged guanidine structures, the guanidinic
carbon, was observed at about 157–160 ppm.18 Because of the
proposed structure with which the carbons are positively charged,
the lines at 159.1 and 161.7 ppm should be consistent with the
guanidinic carbons in the structures Ia and Ib, respectively. Finally,
the line at 153.1 ppm is attributed to the carbonyl carbons bonded
to 14N in structure Ia. With these assumptions, a possible assign-
ment to the 13C CP/MAS spectrum is shown in Figure 2a. The 13C CP/
MAS NMR spectra are consistent with the simultaneous presence of
carbamate and bicarbonate solid-state products.
uncertainty in calculations it is hard to draw any confident
conclusions on this point.
An interesting behavior was also observed when the sample was
dried under reduced pressure for 30 min. The 13C CP/MAS spec-
trum, after this treatment, is shown in Figure 2b. As can be
observed, lines attributed to carbons numbered as 5, 6, 7, and 8
disappeared from the spectrum, leaving the lines from carbons
numbered as 1, 2, 3, and 4. Figure 2b also shows that after drying
the sample the line width of the signal corresponding to the methyl
groups is larger than before, which indicates that this signal can be
attributed to methyl groups with slightly different chemical shifts.
It can be also observed that the spectrum shown in Figure 2b is
almost identical to that in Figure 1b, which shows that the final
product obtained either by dissolving the TMG–CO2 in CDCl3 or by
vacuum drying are indeed the same. This confirms the formation of
a highly stable ionic bicarbonate structure Ib and a species (Ia or Ic)
that releases CO2 under a proper treatment.
13C and 1H NMR of TBD–CO2. Solution 1H and 13C NMR spectra
were acquired in order to identify the reaction products from fix-
ation of CO2 by TBD guanidine. In this case, because the expected
structures are more complex than in the case of TMG–CO2, HSQC
and HBMC experiments19 were also carried out. Figure 3a shows
the 13C solution NMR spectrum of the TBD and CO2 reaction
product. The lines at 115.92 ppm and within the 75.72–78.47 ppm
region are due to the presence of CDCl3 and residual acetonitrile
(ACN) in the solution and will be disregarded. The presence of only
three non-equivalent CH2 carbons at 20.00, 37.98, and 45.95 ppm is
not consistent with a fixed C]Nþ bond as shown in structures
IIa–IIc (Scheme 2), but it suggests that the CDCl3 soluble part of the
product assumes a resonant structure as shown in the inset of
Figure 3a. Hence, the basic structure shown in the inset of Figure 3a
Comparison between experimental and the quantum mechan-
ical chemical shifts (Table 1) do in general support the conclusions
drawn on speciation. The main uncertainty is perhaps if a zwitter-
ionic carbamate (Ia) or carbamate (Ic) is formed. The calculated
shifts for Ic are closer to the experimental values, given the
Figure 3. (a) 13C 1D solution, (b) HSQC, and (c) HMBC spectra of TBD–CO2 reaction product.

F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
10101
will be used for the interpretation of the 2D HSQC and HMBC
spectra. Figure 3b shows the region of the HSQC spectra where the
cross-peaks indicate the 1H nuclei directly attached to 13C. Thus,
from the spectrum shown in Figure 3b the following set of CH2
pairs can be assigned: 13C at 21.00 ppm and 1H at 1.95 ppm (C3–
H3); 13C at 38.00 ppm and 1H at 3.25 ppm (C2–H2); 13C at
45.95 ppm and 1H at 3.33 ppm (C1–H1). Further assignments can
be obtained from the HMBC spectrum shown in Figure 3c. The long-
range HMBC spectrum is shown through bond correlation between
1H and 13C nuclei. As can be observed, there are no long-range
correlations between the 1H line at 1.95 ppm and the 13C line at
150.53 ppm (attributed to the chemical shift value of carbamic
carbon C4). Thus, this 1H nucleus must be four bonds away from C4,
which means that this line can be attributed to H in the position 2 of
Figure 3a. Therefore, the pair 13C at 38.00 ppm and 1H at 3.25 ppm
can be assigned to the C2–H2 group. Both 1H lines at 3.25 and
3.33 ppm show three bond correlations with C4, but the correlation
is clearly weaker for the line at 3.33 ppm, which suggests that it
occurs through the resonant double bond. In addition, the broad-
ening of this line also suggests a higher proximity to a positive
charge. Hence, the line at 3.33 ppm is attributed to the 1H nuclei in
the position 1 of Figure 3a and, consequently, the pair 13C at
38.00 ppm and 1H at 3.25 ppm is assigned to the C1–H1 group.
These assignments are in agreement with the two bond correla-
tions shown in the HMBC spectrum. Consequently, these results are
consistent with the formation of a structure that contains the basic
unit shown in Figure 3a, and the presence of extra peaks in the
150–170 ppm region of the 13C NMR spectrum suggests that all
three structures could be formed.
The solution NMR analysis of the TBD and CO2 reaction products
was quite elucidating in terms of obtaining the initial line assign-
ments. However, the reaction products are solid powders and it is
important to characterize them in the solid phase. The 13C CP/MAS
spectrum of the TBD–CO2 product at ꢀ30 ꢁC is shown in Figure 4. As
with TMG–CO2, it is possible to observe the presence of NMR lines
from 150 to 170 ppm and from 20 to 50 ppm. However, in this case,
there are at least six lines at 150 ppm region (165.4, 162.3, 157.0,
155.0, 154.3, and 151.5 ppm) and several lines from 20 to 50 ppm,
which can be divided in three groups with chemical shifts within
the 42–50, 30–42, and 15–30 ppm regions. This is in agreement
with the 13C solution spectrum, making it possible to attribute
them to the CH2 ring carbons. However, in the solid state, each of
these lines is composed by several lines, which can be due either to
the fact that the lines on the right side, numbered 1, 2, and 3, and on
the left side, 10, 20, and 30, are no longer NMR equivalents in the solid
state or due to the presence of near structurally related compounds
(from Scheme 1) as shown in the inset of Figure 4. More specific
information about the formation of different compounds can be
obtained from the 150 to 170 ppm region. As in the case of the
TMG–CO2 reaction product, the line at 162.3 pm can be attributed
to the carbon nucleus of HCO3ꢀ groups in a bicarbonate structure,16a
as represented by structure IIc in Scheme 2 (carbon 5IIc). The four
lines at 151.5, 155.0, 154.3, and 157 ppm are in a chemical shift
region (150–160 ppm) that could be assigned either to guanidinic
carbons or carbamic carbonyls.18 Since the intensity of the 151.5 and
162.3 ppm lines is comparable, the line at 151.5 ppm can most likely
be attributed to the guanidinic carbon in a bicarbonate structure,
such as IIc (carbon 4IIc). Besides, given the possibility of the for-
mation of a carbamate structure such as IIa, and the chemical shift
calculation shown in Table 1, one can suggest that the 157.0 and
155.0 ppm lines are attributable to a guanidinic carbon and a car-
bamic carbonyl. Note that these two lines have similar intensities,
which are in agreement with the above suggestion. Concerning the
line at 165.4 ppm (50IIc), it appears in a chemical shift region that is
neither typical of a guanidinic carbon nor typical of a carbamic
carbonyl. However, it would be consistent with a bicarbonate car-
bonyl with a slightly different chemical environment from the one
responsible for the 162.3 ppm line. This is consistent with the
presence of water as suggested in the literature,16a where
a deshielding of the carbonyl carbon (shift to higher frequencies) is
expected. Note also that in the presence of solvated bicarbonate, the
guanidinic carbon signal is expected to be a little up-field, which is
consistent with the line at 154.3 ppm (note also that the line in-
tensity is consistent). Thus, the CPMAS spectrum is consistent with
the formation of a carbamate structure (<20%) and a bicarbonate
structure (>80%), as shown in the assignments of Figure 4. How-
ever, it should be pointed out that the assignment was made con-
sidering the formation of only these structures; the formation of
structures with similar chemical composition cannot be ruled out.
In fact, assigning peaks in these solid-state systems is not a trivial
task. As a result of the inherent broadening of the 1H signal due to
the H–H dipolar interaction, the C–H correlation experiments are of
less use in differentiating similar structures. Another approach
would be C–N correlation spectra. Due to the strong quadrupolar
interaction of the abundant 14N nuclei, 15N enrichment would be
required in such experiments. Another issue is the non-quantitative
character of the CPMAS experiments, which could compromise the
analysis made in terms of the line intensities. However, since for
analyzed non-protonated carbons the 1H–13C CP transfer between
non-bonded C–H pairs, which are at similar distances in all possible
structures, it is expected to be similar. This allows the association of
the line intensity with the relative amount of carbons of a given
chemical species and validated the above analysis. In addition, in
our experiments a radiofrequency ramp was used, which also
makes the line intensities in the CPMAS more quantitative.18
The quantum mechanical chemical shifts are in general in good
agreement with the interpretation given of the experimental
spectrum. In Supplementary data, we provide an 1H–13C plot from
calculations showing good overall agreement with Figure 3b. The
calculations suggest that carbon 4 on IIa and IIc and the cabamate
carbonyl in IIa are likely to be the ones corresponding to the
experimental lines in the 151–157 ppm range. Moreover, carbon 5
in bicarbonate (water solvated IIc0 and non-solvated IIc) is expected
to appear in the 162–170 ppm range. In fact, these features were
considered in the attribution of the experimental spectrum of
Figure 4. Calculations on the stability of IIb (Supplementary data)
suggest that this species is perhaps too unstable to show up in the
experimental spectrum. The five lines at 150 ppm may be all
attributable to II, IIa and IIc, but there could also be some complex
formed that we have not considered (perhaps a different form of
bicarbonate complex).
Figure 4. 13C CP/MAS spectrum of TBD–CO2 at ꢀ30 ꢁC. The NMR lines were attributed
on the basis of the typical chemical shifts values assigned in the solution spectrum.
Spinning side bands are indicated by asterisk ( ) in the spectrum.
*

10102
F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
Characterization of TMG–CO2 and TBD–CO2 by FTIR. The charac-
terization by FTIR analyses of the products from the fixation of CO2
with the bases TMG and TBD was limited to identify the bands that
would indicate the presence of carbamate and/or carbonate groups.
Hence, for TMG–CO2 the spectrum showed two bands at w1640
and w1670 cmꢀ1 that can be attributed to bicarbonate and carba-
mate carbonyl groups, respectively. However, bands at w1600 to
1700 cmꢀ1 could be associated to the presence of differently sta-
bilized bicarbonates or carbonates compounds, e.g., guanidinium
12b,e
bicarbonate.
For the TBD–CO2 products the FTIR spectrum
showed a band at w1717 cmꢀ1, which can be attributed to the
carbamate carbonyl group. Nevertheless, the characteristic bands of
bicarbonate can be superposed with the guanidine bands at w1440
to 1650 cmꢀ1
.
3.2. TGA–FTIR study of the CO2 fixation with TMG and TBD
guanidines
TGA–FTIR of the product TMG–CO2. An interesting result was
obtained when the product or a mixture of products TMG–CO2 was
analyzed by TGA. Figure 5 shows losses of CO2 in both temperature
ranges 15–38 and 38–74 ꢁC.
Figure 6. FTIR spectrum of the products from the thermal decomposition of TMG–CO2.
These mass losses can be associated to the bands with higher
values of absorbance at 27 and 59 ꢁC attributed to free CO2 (w2300
to 2400 cmꢀ1). In the spectrum shown in Figure 6, it can be ob-
served more intense signals of CO2 at 27 ꢁC and less intense at
38 ꢁC.
From the results of the TGA–FTIR study it can be proposed that
there are at least two CO2 containing compounds where the CO2
molecule is bonded to guanidine TMG by two different bond types.
One of these products presents a covalent bond N–CO2 to form
a carbamate and the other could be the ionic bicarbonate. This
provides further support to the proposed TMG–CO2 products Ia and
Ib (shown in Scheme 1, Fig. 7).
The bicarbonate represented by Ib should release CO2 at higher
temperature than the carbamic form Ia. Thus, observing the TGA–
FTIR in Figure 5 it is also reasonable to suppose that tetrame-
thylguanidinium bicarbonate Ib was formed in a relatively higher
amount than carbamate Ia by comparing the relative intensities of
the CO2 at different temperatures.
Fixation of CO2 with guanidine TMG occurred by the formation
of two products characterized as carbamate Ia and bicarbonate Ib.
The CO2 release from the TMG–CO2 compounds has been observed
Figure 7. (a) FTIR spectra showing the bands attributed to CO2 at w2320 to 2360 cmꢀ1
(55 ꢁC). (b) Material recovered only from TMG thermal decomposition indicating that
at this temperature all CO2 was released (74 ꢁC).
not to be a significant process and it seems that compounds can be
sublimated at moderated temperature at atmospheric pressure
instead of a CO2–TMG rupture. Thus, fixation and release of the CO2
with TMG could be more important for transcarboxylation (kinetic
reversible fixation) reactions and biological studies.
TGA–FTIR of TBD–CO2. Figure 8 has shown at least three CO2
losses at temperatures between 80 and 140 ꢁC. These CO2 losses can
Figure 5. TGA–FTIR 3D-Diagram for TMG–CO2.

F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
10103
98 and 139 ꢁC. The explanation for this behavior could be the
presence of at least three products formed from the fixation of CO2
with the guanidine TBD.
From the TGA–FTIR analyses of the products of the CO2 fixation
with guanidine TBD, it is possible to observe that at least three CO2
losses were identified by typical bands of free CO2 that have
appeared with different intensities at temperatures between 80
and 140 ꢁC.
This is in agreement with the NMR data that suggested the
presence of three CO2 containing products, with their structures as
represented in Scheme 2. An interesting CO2 releasing process from
the TBD–CO2 products was observed. On the other hand, TBD has
a higher melting point (130–132 ꢁC) than liquid TMG and com-
pounds from the fixation of CO2 with TBD can release CO2 as the
almost unique product from w80 to 135 ꢁC. Hence, TBD could be
more useful for CO2 selective fixation and releasing from a gas
mixture. Therefore, TBD could be a promising model compound for
industrial CO2 separation.
Figure 8. TGA–FTIR 3D diagram for TBD–CO2.
3.3. Time-dependent isothermal absorption of CO2 by TMG
and TBD guanidines
be associated with the presence of three compounds and the CO2
molecule could be bound in different ways to TBD moiety.
Figure 9 shows the FTIR bands corresponding to free CO2 at 80
and 110 ꢁC. These bands were also observed with less intensity at
The CO2 fixation with the guanidines TMG and TBD was con-
ducted in a TG instrument under a CO2 flow (15 mL/min). The
guanidines were placed on a sample pan in the TG instrument at
three different temperatures and the experiments were carried out
in an isothermal mode for a determined period of time or until the
weight variation stopped.
Figure 10 shows the CO2 absorption behavior of TMG at 22, 26,
and 30 ꢁC. The CO2-fixation efficiency at 30 ꢁC is lower than that at
26 ꢁC. This could be related to the minor thermal stability of the
TMG–CO2 products as showed in Figure 6. A lower temperature
resulted in higher amounts of CO2 fixation owing to the entropic
advantage.
The CO2 absorption by TBD was also investigated at three dif-
ferent temperatures, at 30, 60, and 90 ꢁC as showed in Figure 11.
Mass losses of TBD at the beginning could be associated with some
volatile solvent impurity (commercial TBD purity is around 98%).
Hence, at 90 ꢁC no mass loss was observed at the starting
absorption experiment. At 90 ꢁC the CO2 absorption by TBD is in
according with this temperature value being between 80 and
98 ꢁC where CO2 mass losses were observed from TBD–CO2
Figure 9. TGA–FTIR spectra of TBD–CO2 versus temperature.
Figure 10. Isothermal CO2 absorption by TMG.

10104
F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
Figure 11. Isothermal CO2 absorption by TBD.
Table 2
products as displayed in Figure 9. On the other hand, after 150 min
the mass absorption starts to decrease, which indicates a possible
TBD–CO2 product degradation or that no more CO2 absorption is
achieved at this time.
Results from transcarboxylation of amines with Guanidine–CO2 products
Carbamate
R
R1X
Yield of carbamates
obtained by transcarboxylation
The guanidines were reacted with CO2 and mass increases were
observed. The mass increases at lower temperatures were larger
than with higher temperatures, probably due to the better CO2-
fixation ability of guanidines at lower temperatures.
These results demonstrate the potential of the guanidines as
recyclable CO2 adsorption materials, which can fix CO2 reversibly
and release it at higher temperatures.
TMG–CO2 (%)
TBD–CO2 (%)
3
4
5
6
7
C6H11
EtI
EtI
EtI
BuBr
BuBr
77
80
74
76
78
68
72
69
70
70
C5H5O
C7H5Cl2
C5H5O
C7H5Cl2
transcarboxylating compounds. Selective transcarboxylation from
carbamate or bicarbonate intermediates is under current investi-
gation and the results will be opportunely communicated.
This method for preparation of carbamates is limited to trans-
carboxylation of compound with higher nucleophilicity than the
initial CO2 trap (TMG and TBD), because reaction of the corre-
sponding TMG and TBD carbon dioxide complexes did not transfer
the CO2 moiety toward alcohols.
3.4. Transcarboxylation experiments
Kinetic reversibility. For the purpose of comparison, the trans-
carboxylation of cyclohexylamine, as a model nucleophilic amine,
was performed at ꢀ5 ꢁC with both TMG–CO2 and TBD–CO2, and the
corresponding ethyl carbamates were obtained by a subsequent
reaction of the carbamate intermediates with ethyl iodide for 6 h at
10 ꢁC. At low temperatures alkylation of the amine or carbamate is
minimal. General pathway for formation of carbamates from
transference of CO2 is illustrated in Scheme 3.
4. Conclusions
Under these conditions, the yields of the isolated N-(cyclohexyl)
ethyl carbamate were 77 and 68%, respectively. Another ethyl and
butyl carbamates were also prepared by transcarboxylation of
furfuryl- and 2,4-dichlorobenzylamine. Structures were confirmed
by NMR analyses and GC conversion to carbamates was w70 to 80%
(no optimized reactions). Table 2 summarizes the approximated
yields of isolated alkyl carbamates from each guanidine–CO2
products.
Guanidines TMG and TBD can be used as model compounds by
capture and activation of the CO2 molecule. These bases form bi-
carbonates presumably by previous formation of water–solvated
carbamic intermediate. The guanidine–CO2 products can trans-
carboxylate nucleophilic amines. This behavior suggests a mimetic
transcarboxylase activity. On the other hand, bicarbonate formation
in the presence of the investigated guanidines, CO2, and catalytic
water is indicative of the non-metal carbonic anhydrase mimetism.
This is very interesting for biochemical study of the guanidine re-
lated compound, which can be used for CO2 and water fixation.
Finally, TBD–CO2 products have shown an interesting thermal
From these results, we can say that CO2 fixation with both
guanidines, TMG and TBD, is a kinetically reversible process as the
corresponding fixation products have shown to be useful as
-5 °C
(eq. 1)
[Guanidine-H] RNHCOO
Guanidine-CO2
+
RNH2
~ 2h
10 °C
~ 6h
RNHCOOR1
68 - 80%
(eq. 2)
R1X
[Guanidine-H] X
+
[Guanidine-H] RNHCOO
+
Scheme 3.

F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
10105
stability. Releasing of CO2 from the TBD–CO2 products has occurred
at moderated temperature. This could be useful for selective sep-
aration of CO2 from complex gas mixtures by TBD or TBD related
compounds.
be attributed to bicarbonate and carbamate carbonyl groups,
respectively.
5.2.3. 1,1,3,3-Tetramethylguanidinium bicarbonate (Ib)
Assigned based on the 1H liquid state by 1H NMR peaks of the
soluble TMG compound (as described above) and in the solid-state
5. Experimental section
13C CPMAS NMR
d
41.2 (CH3), 161.7 (C]N), 162.3 (HCOꢀ3 ). IR bands
5.1. Fixation of CO2 with 1,1,3,3-tetramethylguanidine (TMG)
and 1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]-pyrimidine
(TBD), compounds I and II, respectively: general procedure
at w1600 to 1700 cmꢀ1 could be associated to the presence of
differently stabilized bicarbonates or carbonate compounds, e.g.,
guanidinium bicarbonate.12b,e
The CO2-fixation experiments with the guanidine bases have
been carried out using pure commercially available TMG (99%, GC)
and TBD (98%) without any previous purification. Reactions were
carried out in the presence or absence of organic solvent (fresh
acetonitrile) for liquid TMG or using minimal acetonitrile for TBD.
Experiments using melted TBD were not successful for CO2 capture,
presumably because of low CO2 retention into the reaction media.
No significant differences were observed under solvent or dry
conditions for TMG. Guanidines (5 mmol) were placed in a 50-mL
two-neck round flask and acetonitrile was added if needed, then
CO2 was bubbled under normal pressure (w10 mL/min) at 5 ꢁC
with magnetic stirring for 2 h. A high yield of a white powder
(w90%) was obtained for both TMG and TBD–CO2-captured prod-
ucts. These crude products were stored at 0 ꢁC for liquid-state NMR
(1H and 13C), solid-state NMR, 13C, FTIR, and TGA–FTIR measure-
ments. It should be noted that while the intention has been to carry
out experiments in a water-free environment both TMG and TBD
are highly hygroscopic species and water contamination cannot be
ruled out. Even small quantities of water may have a significant
effect on the overall reactivity of the systems. On the other hand,
experiments carried out under no especially anhydrous condition
can be interesting for future applications.
5.2.4. 1,3,4,6,7,8,9-Hexahydro-2H-pyrimido[1,2-a]-pyrimidin-
1-ium-1-carboxylate (IIa)
Assigned based on the 13C solid-state NMR
d
165.4 (–COOꢀ),
157.5 (–N–C]Nþ). IR w1717 cmꢀ1, which can be attributed to the
carbamic C]O group.
5.2.5. 1,3,4,6,7,8-Hexahydro-2H-pyrimido[1,2-a]-pyrimidinium
bicarbonate (IIc)
Assigned based on the 13C solid-state NMR
d
162.3 (HOCOOꢀ),
152 (–N–C]Nþ) indicates that compound IIc was also formed. The
lines at the 20–50 ppm region are consistent either with IIa or IIc.
Compound IIc was also identified by liquid-state NMR using 1H,
13C{1H}, HSQC, and HMBC correlation spectroscopy and the as-
signment of these signals was 13C at 21.00 ppm and 1H at 1.95 ppm
(C3–H3); 13C at 38.00 ppm and 1H at 3.25 ppm (C2–H2); 13C at
45.95 ppm and 1H at 3.33 ppm (C1–H1). IR w1440 to 1650 cmꢀ1, in
this region characteristic band of bicarbonate can be superposed
with the guanidine bands.
5.3. Transcarboxylation experiments for N-alkyl carbamates
by amine transcarboxylation
5.2. NMR studies of the TMG–CO2 and TBD–CO2 products
In a typical procedure, 2 mmol of amine was slowly added to the
suspension of guanidine–CO2 (2.4 mmol) in 10 mL of anhydrous
acetonitrile, with stirring for 1.5 h at 5 ꢁC. Then, alkyl halide
(3 mmol) was poured into the solution and the reaction mixture
was stirred at ambient temperature for an additional 6 h. Evapo-
ration of the solvent and extraction with ethyl ether (3ꢂ10 mL) or
alternative column chromatography (6:1 hexanes/EtOAc) gave the
corresponding N-alkyl carbamates.
Solid-state NMR experiments were performed at 13C and 1H
frequencies of 100.5 and 400.0 MHz, respectively. A VARIAN 7-mm
MAS double-resonance probe head with variable temperature (VT)
was used. Typical p/2 pulse lengths of 3.5 and 4.0 ms were applied
for 13C and 1H, respectively. Time Proportional Phase Modulated
(TPPM) 1H decoupling with field strength of 70 kHz, cross-polari-
zation with radiofrequency ramp20 with duration of 1 ms, MAS
spinning frequency of 6 kHz, and recycle delays of 5 s were used.
Liquid-state NMR experiments were performed at 13C and 1H
5.3.1. N-(Cyclohexyl) ethyl carbamate (3)
White solid; mp 122–124 ꢁC (from ether); 1H NMR (200.13 MHz,
frequencies of 75 and 300.0 MHz, respectively. Typical
p
/2 pulse
CDCl3)
d
1.27–1.09 (m, 9H, CH2 cyclohexylþCH3), 1.60 (m, 2H, CH2
lengths of 7 and 10
m
s were applied for 13C and 1H, respectively.
cyclohexyl), 1.91 (m, 2H, CH2 cyclohexyl), 3.41 (m, 1H, CH cyclo-
Recycle delays of 0.1 and 1 s and 16 and 64 K acquisition points were
used. The 2D experiments were carried out using 4 K points in the
direct dimension and256t1/u1 incrementsin theindirectdimension.
TGA–FTIR study for TMG–CO2 and TBD–CO2. The study was car-
ried out with a coupled TG–FTIR, in which all gas evaporated at the
TG chamber is channeled to the FTIR spectrometer. The samples
(5.0 mg) were heated in alumina crucible using purified N2 gas
(15 mL/min) and heating rate of 10 ꢁC/min from 0 ꢁC to 140 ꢁC. FTIR
spectra were recorded in the wavenumber range 400–4000 cmꢀ1
with 4 cmꢀ1 spectral resolution, 32 scans, and a DTGS detector.
hexyl), 4.07 (q, 2H, CH2), 4.53 (s, 1H, NH); 13C NMR (50.32 MHz,
CDCl3)
d 14.5, 24.7, 25.4, 33.4, 49.6, 66.0, 155.8 (C]O); MS (70 eV)
m/z 56 (100%), 128, 142, 171 (Mþ). IR (KBr) 3340, 3223, 1680 cmꢀ1
.
5.3.2. N-(2-Furfuryl) ethyl carbamate (4)
Yellow oil; 1H NMR (200.13 MHz, CDCl3)
d 1.24 (t, 3H, methyl),
4.11 (q, 2H, CH2), 4.29 (d, 2H, CH2 furyl), 5.05 (t, 1H, amino), 6.20 (d,
1H, 3-furyl), 6.29 (2d, 1H, 4-furyl), 7.32 (d, 1H, 5-furyl). 13C NMR
(50.32 MHz, CDCl3)
(C]O). MS (70 eV) m/z 69, 81, 96 (100%), 140, 169 (Mþ). IR (thin
film) 3336, 1679, 1312 cmꢀ1
d 15.3, 37.9, 61.0, 107.0, 109.5, 142.1, 151.6, 156.3
.
5.2.1. 1,1,3,3-Tetramethylguanidine (I)
1H NMR (CDCl3)
d
2.65 (s, 12H, 4CH3), 4.94 (s, 1H, NH); 13C
5.3.3. N-(2-Furfuryl) butyl carbamate (5)
(CDCl3)
d
39.19 (CH3), 167.8 (C]N), all spectra were carried out at
Yellow oil; 1H NMR (200.13 MHz, CDCl3)
d 0.94 (t, 3H, CH3), 1.50
room temperature.
(m, 2H, CH2), 1.62 (m, 2H, CH2 furyl), 4.09 (t, 2H, CH2), 5.01 (t, 1H,
NH), 6.23 (d, 1H, 3-furyl), 6.31 (2d, 1H, 4-furyl), 7.34 (d, 1H, 5-furyl);
5.2.2. 1,1,3,3-Tetramethylguanidine-N-carboxylate (Ia)
13C NMR (50.32 MHz, CDCl3)
d
13.9, 20.1, 27.3, 36.0, 61.7, 107.2,
109.2, 151.4, 156.1 (C]O); MS (70 eV) m/z 57, 69, 81, 96 (100%), 136,
193 (Mþ). IR (thin film) 3332, 1677, 1311 cmꢀ1
Assigned by solid-state 13C{1H} NMR
d
41.2 (CH3), 43.5 (CH3),
159.1 (C]N), 153.8 (–NCOO); IR 1640, 1670 cmꢀ1, these bands can
.

10106
F.S. Pereira et al. / Tetrahedron 64 (2008) 10097–10106
O., de Coninck, H. C., Loos, M., Meyer, L. A., Eds.; Cambridge University Press:
5.3.4. N-(2,4-Dichlorobenzyl) ethyl carbamate (6)
Cambridge, United Kingdom and New York, NY, USA, 2005.
3. Mcghee, W.; Riley, D. J. Org. Chem. 1995, 60, 6205–6507.
Yellow oil; 1H NMR (250 MHz, DMSO-d6)
d 1.20 (t, 3H, CH3), 4.08
(q, 2H, CH2), 4.62 (2d, 2H, CH2 benzyl), 6.55 (t, 1H, NH), 7.00 (d,
4. (a) Casadei, M. A.; Inesi, A.; Moracci, F. M.; Rossi, I. Chem. Commun. 1996, 2575;
(b) Stastny, V.; Rudkeviich, D. M. J. Am. Chem. Soc. 2007, 129, 1018–1019.
5. (a) The Merck Index, 12th ed.; Whitehouse Station: New Jersey, NJ, 1996; (b)
Hazardous Chemicals Desk Reference, 4th ed.; John Wiley & Sons: New York, NY,
1998.
J¼8.4 Hz, 1H, ArH), 7.19 (2d, J¼2.0, 8.4 Hz, 1H, ArH), 7.40 (d,
J¼2.2 Hz, 1H, ArH). 13C NMR (62.50 MHz, DMSO-d6)
d 14.4, 44.7,
61.6, 127.0, 129.6, 133.2, 134.0, 136.4, 136.6, 157.6 (C]O); MS (EI,
70 eV) m/z 45, 87, 159, 161, 201 (100%), 203 (64%), 247 (Mþ), 249
6. Delseth, R. Chimia 1998, 52, 698–701.
(Mþþ2); IR (thin film) 3340, 1680 (C]O), 1582–1540 (Ar) cmꢀ1
.
7. (a) Pe´rez, E. R.; Odnicki da Silva, M.; Costa, V. C.; Rodrigues-Filho, U. P.; Franco,
D. W. Tetrahedron Lett. 2002, 43, 4091–4093; (b) Pe´rez, 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.
5.3.5. N-(2,4-Dichlorobenzyl) butyl carbamate (7)
Yellow oil; 1H NMR (250 MHz, DMSO-d6)
0.92 (t, 3H, CH3), 1.55
8. Endo, T.; Nagai, D.; Monma, T.; Yamaguchi, H.; Ochiai, B. Macromolecules 2004,
37, 2007–2009.
9. Darensbourg, D. J.; Mackiewicz, R. M. J. Am. Chem. Soc. 2005, 127, 14026–14038.
10. Heldebrant, D. J.; Jessop, P. G.; Thomas, C. A.; Eckert, C. A.; Liotta, C. L. J. Org.
Chem. 2005, 70, 5335–5338.
d
(m, 2H, CH2), 1.68 (m, 2H, CH2), 4.10 (t, 2H, CH2), 4.59 (2d, 2H, CH2
benzyl), 6.58 (t, 1H, NH), 7.01 (d, J¼8.4 Hz, 1H, ArH), 7.21 (2d, J¼2.0,
8.4 Hz, 1H, ArH), 7.41 (d, J¼2.2 Hz, 1H, ArH); 13C NMR (62.50 MHz,
11. (a) Jessop, P. G.; Heldebrant, D. J.; Li, X.; Eckert, C. A.; Liotta, C. L. Nature 2005,
436, 1102; (b) Yamada, T.; Lukac, P. J.; George, M.; Weiss, R. G. Chem. Mater. 2007,
19, 967–969.
DMSO-d6)
d 13.6, 19.2, 31.9, 44.7, 64.6, 126.8, 129.5, 133.0, 133.8,
136.5, 137.0, 158.0 (C]O); MS (EI, 70 eV) m/z 57, 73 (100%), 159, 161,
201, 203, 275 (Mþ), 277 (Mþþ2); IR (thin film) 3343, 1681 (C]O),
12. (a) Amatore, C.; Savea´nt, J.-M. J. Am. Chem. Soc. 1981, 103, 5021–5023; (b)
Christensen, P. A.; Hamnett, A.; Muir, A. V. G. J. Electroanal. Chem. 1990, 288,
197–215; (c) Genaro, A.; Isse, A. A.; Severin, M. A.; Vianello, E.; Bhugun, I.;
Savea´nt, J.-M. J. Chem. Soc., Faraday Trans. 1996, 92, 3963–3968; (d) Casadei, M.
A.; Inesi, A.; Rossi, L. Tetrahedron Lett. 1997, 38, 3565–3568; (e) Perez, E. R.;
Garcia, J. R.; Cardoso, D. R.; McGarvey, B. R.; Batista, E. A.; Rodrigues-Filho, U. P.;
Vielstich, W.; Franco, D. W. J. Electroanal. Chem. 2005, 578, 87–94; (f) Feroci, M.;
Orsini, M.; Rossi, L.; Sotgiu, G.; Inesi, A. J. Org. Chem. 2007, 72, 200–203.
13. Wiberg, K. B. J. Comput. Chem. 1999, 20, 1299–1333.
14. Cances, M. T.; Mennucci, V.; Tomasi, J. Chem. Phys. 1997, 107, 3032–3041.
15. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Fores-
man, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision C.02; Gaussian: Wallingford, CT, 2004.
1580–1540 (Ar) cmꢀ1
.
Acknowledgements
´
Authors acknowledge Prof. Jose A. Giacometti (FCT–UNESP) for
his fundamental support to this research on chemistry of carbon
dioxide. They also acknowledge Dr. Paulo J.M. Cordeiro and Dr.
Mauro R. Fernandes (IQSC–USP) for GC-MS and FTIR analyses, Prof.
´
Claudio F. Tormena (NMR Laboratory, IQ-UNICAMP) for providing
the NMR spectrometer for the solution analyses. Finally, authors are
grateful to the Brazilian Agencies FAPESP, CAPES, and CNPq.
Supplementary data
Quantum mechanical chemical shift data and reaction energies.
Supplementary data associated with this article can be found in the
online version, at doi:10.1016/j.tet.2008.08.008.
16. (a) Park, J.-Y.; Yoon, S. J.; Lee, H. Environ. Sci. Technol. 2003, 37, 1670–1675;
(b) Hajduk, P. J.; Meadows, R. P.; Fesik, S. W. Q. Rev. Biophys. 1999, 32, 211–240.
17. (a) Hester, R. K.; Ackerman, J. L.; Neff, B. L.; Waugh, J. S. Phys. Rev. Lett. 1976, 36,
1081–1083; (b) Munowitz, M. G.; Griffin, R. G.; Bodenhausen, G.; Huang, T. H.
J. Am. Chem. Soc. 1981, 103, 2529–2533.
References and notes
1. Climate Change 2007: The Physical Science Basis; Solomon, S., Qin, D., Manning,
M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H. L., Eds.; Contribution
of Working Group I to the Fourth Assessment Report of the Intergovernmental
Panel on Climate Change; Cambridge University Press: Cambridge, United
Kingdom and New York, NY, USA, 2007.
18. Huang, M.-J.; Lee, K. S. Mol. Phys. 2005, 103, 2229–2237.
19. (a) Bodenhausen, G.; Ruben, D. J. Chem. Phys. Lett. 1980, 69, 185–189; (b) Bax, A.;
Davis, D. G. J. Magn. Reson. 1985, 63, 207–213; (c) Bax, A.; Griffey, R. H.;
Hawkins, B. L. J. Magn. Reson. 1983, 55, 301–315.
2. IPCC Special Report on Carbon Dioxide Capture and StoragePrepared by Working
Group III of the Intergovernmental Panel on Climate Change; Metz, B., Davidson,
20. Novotny, E. H.; Hayes, M. H. B.; DeAzevedo, E. R.; Bonagamba, T. J. Natur-
wissenschaften 2006, 93, 447–450.