CO2 (De)Activation in Carboxylation Reactions: A Case Study Using Grignard Reagents and Nucleophilic Bases

Article abstract of DOI:10.1021/acs.organomet.9b00838

Carbon dioxide (CO₂) is an intrinsically stable molecule. However, its reactivity toward nucleophilic bases has constituted an appealing characteristic for applications such as CO₂ capture and functionalization. To shed light on the role of nucleophilic bases in CO₂ functionalization, we performed some mechanistic studies using nitrogen-containing bases as an additive-in catalytic amounts-for carboxylation reactions of Grignard reagents. Our kinetic analysis and in situ infrared spectroscopy revealed the role of nucleophilic bases, particularly that of DBU (1,8-diazabicycloundec-7-ene), in CO₂ (de)activation for carboxylation reactions.

Full text of DOI:10.1021/acs.organomet.9b00838

pubs.acs.org/Organometallics
Article
CO2 (De)Activation in Carboxylation Reactions: A Case Study Using
Grignard Reagents and Nucleophilic Bases
Jerik Mathew Valera Lauridsen, Sung Yeon Cho, Han Yong Bae,* and Ji-Woong Lee*
Cite This: https://dx.doi.org/10.1021/acs.organomet.9b00838
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Carbon dioxide (CO2) is an intrinsically stable molecule. However,
its reactivity toward nucleophilic bases has constituted an appealing characteristic
for applications such as CO2 capture and functionalization. To shed light on the
role of nucleophilic bases in CO2 functionalization, we performed some
mechanistic studies using nitrogen-containing bases as an additivein catalytic
amountsfor carboxylation reactions of Grignard reagents. Our kinetic analysis
and in situ infrared spectroscopy revealed the role of nucleophilic bases,
particularly that of DBU (1,8-diazabicycloundec-7-ene), in CO2 (de)activation for
carboxylation reactions.
INTRODUCTION
■
The Grignard reaction with carbon dioxide (CO2) has
constituted a landmark C−C bond-forming reaction for
probing the chemical reactivity of organomagnesium com-
pounds, while accessing the useful carboxylic acid functional
group from alkyl and aryl halides.1 The high reduction
potential of Grignard reagents allows the formation of C−C
bonds to spontaneously afford carboxylic acids from gaseous or
solid-state CO2. In 1912, Victor Grignard proposed that an
organomagnesium compound might play a role in CO2 fixation
in chlorophyllwhere carbon fixation occurs.2 Although the
mode of action on carbon fixation is distinct from a typical
Grignard reaction, the Calvyn−Benson−Bassham cycle in-
corporates a carbon nucleophile for carboxylation reaction with
CO2.3 It has been verified that the magnesium(II) Lewis acidic
center was crucial for the carboxylation reaction by capturing
CO2 within the active site with the aid of a nucleophilic
nitrogen basebiotin for pyruvate carboxylase4 and lysine for
the Rubisco enzyme, respectively (Figure 1A,B).5 Although the
formation of carbamate with a base may reduce electrophilicity
for the carboxylation reactions, these Lewis acid/base adducts
have been commonly found in enzymes for carbon fixation as
key intermediates. Moreover, the change in the bond angle of
CO2 (O−C−O) decreases the LUMO energy of gaseous
CO2,6 implying that the chemical processes involving CO2
utilization can be facilitated catalytically.
Conceptually, frustrated Lewis pairs (FLPs) are recognized
for their capabilities in the catalytic “activation” of inert
molecules (e.g., dihydrogen via heterolytic cleavage).7,8
However, the catalytic activation of carbon dioxide with
FLPs is still not clear because of the ambiguity of the
Figure 1. Active sites for carbon dioxide fixation. (A) Pyruvate
carboxylase. (B) Rubisco enzyme. (C) Frustrated Lewis pair (FLP)
for CO2 activation. (D) This work: Schematic representation of CO2
(de)activation for carboxylation reactions of Grignard reagents.
electrophilicity of the CO2 base adducts (Figure 1C).9 For
example, FLPs have been evaluated as catalysts in hydro-
Special Issue: Organometallic Chemistry for Enabling
Carbon Dioxide Utilization
Received: December 12, 2019
https://dx.doi.org/10.1021/acs.organomet.9b00838
© XXXX American Chemical Society
Organometallics XXXX, XXX, XXX−XXX
A

Organometallics
pubs.acs.org/Organometallics
Article
boration reactions of CO2 and both positive and negative
effects of the “catalysts” were observed, depending on the
reaction mechanisms.10 The proposed enigmatic intermediate
that may increase reaction rates in CO2 functionalization is still
elusive, and the roles of nucleophilic bases and Lewis acids in
such reactions have been postulated only as catalysts for CO2.
An ideal chemical CO2 functionalization reaction is yet to be
developed because of the high thermodynamic stability and
kinetic inertness of carbon dioxide.11 To overcome this
difficulty, a CO2-activation catalystwhich would increase
the effective concentration of CO2 in the reaction medium
without forming stabilized adducts (poisoning)would be
desirable. However, it is not trivial to design a differentiate
binding modestabilizing CO2 with a base while inducing a
superior binding of the transition state and, thus, induce an
observable rate enhancement. This paper describes our model
study for probing into this problem: Grignard reactions with
carbon dioxide in the presence of nucleophilic bases, in
particular, DBU (1,8-diazabicyclo(5.4.0)undec-7-ene), which
is prone to interact with CO2 (Figure 1D). Considering the
high reaction rate of Grignard reagent carboxylation reactions,
we postulated that the nucleophilic interactions of bases with
CO2 would be expressed in the kinetic analysis as a pre-
equilibrium state or catalysis, rendering opportunities to
experimentally evaluate these hypotheses. In addition, the
resulting strong C−C bond (Ar−CO2H) would, in principle,
prevent the reverse decarboxylation reaction, allowing us to
pinpoint the action of the nucleophilic bases in the CO2
functionalization reaction. To achieve this, we conducted a
competitive kinetic analysisbecause of the early rate-limiting
step of the Grignard carboxylation reactionsto shed light on
the role of CO2 nucleophilic base adducts in the organo-
metallic carboxylation reaction.
Table 1. Screening of the Nucleophilic Base for the
Carboxylation Reaction of 2,6-Dimethylphenylmagnesium
a
Bromide
b
entry
base
X mol %
time (min) conv (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
no base
no base
triethylamine
triphenylphosphine
DABCO
-
-
5
30
5
17%
49%
6%
25 mol %
20 mol %
20 mol %
50 mol %
50 mol %
50 mol %
50 mol %
50 mol %
20 mol %
10 mol %
50 mol %
10 mol %
10 mol %
10 mol %
30
30
30
5
30
30
30
30
30
30
30
30
30
3%
11%
47%
64%
77%
64%
31%
50%
48%
52%
48%
50%
47%
2,6-lutidine
DBU
DBU
TMEDA
Et4NCN
piperidine
N,N-diisopropylamine
Hunig’s base
̈
DMAP
TBD
imidazole
a
Reaction conditions: The Grignard reagent was freshly prepared and
directly injected into a reaction vessel containing THF (pre-CO2
saturated; see the Supporting Information) and a base at room
b
temperature, and stirred at room temperature for T minutes. The
conversion was determined by 1H NMR analysis in the presence of an
internal standard. DABCO = 1,4-diazabicyclo[2.2.2]octane; TMEDA
= tetramethylethylenediamine; Hunig’s base = N,N-diisopropylethyl-
̈
amine; DMAP = 4-(dimethylamino)pyridine; TBD = triazabicyclo-
decene.
RESULTS AND DISCUSSION
■
This indicates that the concentration of CO2 under
presaturated THF solution was sufficient for the initial
conversion of the Grignard reagents (entries 1 and 2). We
also noticed that the nature of the base was critical to the
outcome of the carboxylation reactions. Common nucleophilic
bases such as triethylamine, triphenylphosphine, and DABCO
showed negative effects on the conversion (entries 3 to 5).
Surprisingly, we noticed that DBU consistently showed notable
differences when compared with the reactions without
additives (entries 7 and 8). It has been postulated that DBU
can increase the solubility of CO2, perhaps by forming a 1:1
CO2-base adduct.14 Under our reaction conditions, we ruled
out the effect of increased CO2 concentration, given that the
amount of dissolved CO2 in all reactions was identical. Apart
from DBU, most bases showed negative effects, highlighting
the unique property of DBU in promoting the Grignard
carboxylation. In the cases of enzymatic carboxylation with
pyruvate carboxylate and Rubisco, biotin and lysine coordinate
with CO2, with a high proximity between the Mg center and
nitrogen atoms. The coordination of CO2 may be productive
for biocatalytic carboxylation reactions by either (1) increasing
the local concentration of CO2 in the active sites or (2)
increasing the reactivity of CO2 for incoming nucleophiles.
As further investigation, we sought to pinpoint the role of
DBU in the carboxylation by in situ infrared spectroscopy. For
example, the concentration of DBU in THF is critical to
significantly increase CO2(soln.) from CO2(g) (Figure 1D).
Furthermore, on the basis of the experiments conducted by
At the outset, we commenced our investigation by screening
various nitrogen-containing bases for Grignard carboxylation
reactions in THF, which was presaturated with CO2. This
should preclude the other external effects of nucleophilic bases
for CO2 absorption and CO2 solubility with different bases.
Sterically demanding 2,6-dimethylphenyl magnesium bromide
was selected in this study as a means to mitigate the fast
carboxylation reaction rate of the Grignard reagents (Table 1).
We also expected reduced interactions of the Grignard
reagents with the nucleophilic bases, which would affect
Schlenk equilibrium prior to the carboxylation reaction step.12
In addition, we postulated that the effect of the nitrogen-
containing bases in coordinating to Grignard reagents is
negligible in THF, as indicated by the higher affinity of
Grignard reagents to THF, as compared with DBU.13
We first confirmed that the decarboxylation of the carboxylic
acid was negligible in the presence of nitrogen-containing bases
with and without additional Grignard reagent. Additionally, we
verified that the isolated yield of the obtained carboxylic acid
was accurate, by comparison with the corresponding 1H NMR
yields. Our experiments showed that the carboxylation reaction
occurred spontaneously at room temperature; almost 50%
conversion was detected after 30 min in CO2-saturated THF
(entry 2 in Table 1). We observed that an additional CO2
balloon afforded only minor changes in the carboxylation
reaction outcome in the absence of bases (see the Supporting
Information).
B
https://dx.doi.org/10.1021/acs.organomet.9b00838
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
varying the DBU concentration in THF with a CO2 balloon
(Figure 2, top), the relative CO2 concentration was maximized
However, because of the low activation barrier of the
carboxylation of Grignard reagents (ca. 2.3 kcal/mol),15 the
rate limiting step will be the diffusion or mixing rate of the
reagents (See Supporting Information for the full equation).
Under conventional reaction conditions with a large value of
kr[RMgBr], we would observe apparent pseudozero order
kinetics, with the rate of product formation following the CO2
absorption rate in the solution, as implied by the steady-state
approximation for the CO2 concentration (eq 1).
d[CO2]
= 0
dt
= −kr[RMgBr][CO2(soln.)] + kabs[CO2(g)] − kdis[CO2(soln.)
]
d[P]
dt
= kr[RMgBr][CO2(soln.)] = kabs[CO2(g)
]
(1)
when kdis ≪ kr[RMgBr] [where R = 2,6-dimethylphenyl].
Therefore, we designed a comparative kinetic analysis
(Figure 3A) for the carboxylation reaction, in which the
Figure 2. (Top) Relative CO2 concentration measured by in situ FT-
IR at 2340 cm−1 in solution (THF), for different DBU concentrations.
The intensity of the CO2 absorption peak is normalized by the highest
CO2 absorption peak at 0.1 M DBU, the red line. (Bottom)
Comparison of the infrared absorption peaks obtained in situ,
measured in the presence (solid lines) and in the absence (dashed
lines) of DBU (CO2: 2240 cm−1), substrate (761 cm−1), product
(776 cm−1), and DBU-CO2 (1652 cm−1) as functions of time.
Substrate = 2,6-dimethylphenylmagnesium bromide; Product = 2,6-
dimethylbenzoic acid.
at 0.1 M of DBU. Higher concentrations of DBU showed
negligible effects in changing the relative CO2 absorption and
the initial absorption rate, which can be ascribed to the
formation of insoluble DBU-CO2 adducts in THF (See Figure
S2). Therefore, the key question seems to be are CO2(soln.) and
DBU-CO2 adducts relevant for the rate of the carboxylation
reaction?
Figure 3. (A) Simplified reaction kinetics for the Grignard reagent
carboxylation reaction. (B) Kinetic expression of the comparative
Grignard carboxylation reaction.
influence of DBU can be determined by the variation of the
reactions of the two reagents. We realized that the kinetic
analysis of a single Grignard reagent is superfluous, because the
measurement of product formation as a function of time would
provide insufficient kinetic information. After testing a few
additional Grignard reagents (Tables S1 and S2), we employed
fluorinated substrates (4-fluorophenylmagnesium bromide as
starting material) to ease the observation of the corresponding
products using 19F NMR spectroscopy, after quenching the
reaction mixtures in CD3OD (see the Supporting Information
for the experimental details). Based on the fast equilibrium
approximation of “activated” CO2 by DBU and the excess
amounts of Grignard reactants compared to CO2, we derived
an eq (Figure 3B) capable of providing information on the
susceptibility of the organometallic reagents toward DBU-CO2
adducts (see the Supporting Information for the derivation of
eq 6). The simplified constants a and b are the ratios of
catalyzed and uncatalyzed Grignard reactions (k1c/k1 and k2c/
k2, respectively, eq 4. According to eq 6, plotting the ratio
between the apparent rate ratio constant Kr and the ratio of
To answer this question, we performed in situ carboxylation
reactions with DBU (solid lines of Figure 2, bottom) and
without DBU (dotted lines of Figure 2, bottom). Under
identical conditions, the consumption of the Grignard reagent
showed pseudozero order kinetics, and was marginally faster in
the presence of DBU. Moreover, the signature peak of
CO2(soln.) (2630 cm−1) was sharply intensified after con-
sumption of the Grignard reagent (ca. 150 s, blue lines)
whereas the reaction without DBU showed a gradual increase
of CO2(soln.) at a lower rate. Compared with the peak
corresponding to CO2, a peak observed at 1652 cm−1
attributed to an adduct between CO2 and DBU (solid light
blue line)showed fast kinetics unrelated to product
formation. Besides the observed DBU-CO2 adduct, DBU was
shown to be helpful in increasing CO2(soln.) in the solution from
the beginning of the reaction (as results from the comparison
with the reaction without DBU), indicating a potential role of
DBU in the carboxylation reaction.
C
https://dx.doi.org/10.1021/acs.organomet.9b00838
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
rate constants K1,2(Kr/K1,2), versus (1 − Kr/K1,2)/[DBU]
should provide information regarding k1c/k2c, based on the
obtained y-intercept (a/b) and the background experiments
without DBU (k1/k2) (eq 7). Therefore, based on the
competitive kinetic experiments, we can determine the relative
susceptibility of two different organometallic species (R1MgBr
and R2MgBr) toward the DBU-CO2 adduct (KDBU).
Having determined the kinetic model to use, we conducted
competitive Grignard carboxylation reactions with a 1:3 ratio
of differently substituted Grignard reagents, such as R1MgBr
(4-fluorophenylmagnesium bromide) and R2MgBr (4-trifluor-
omethoxyphenylmagnesium bromide), to compensate for the
difference of reaction rates. The remaining substrate
concentration was determined by integrating deuterated
Grignard reagents based on 19F NMR spectroscopy. The
ratio of the Grignard carboxylation reaction was determined
(k1/k2 = 3.20 0.09) by conducting five individual identical
experiments, which showed, with reliable correlation, the
presence of a constant k1/k2 value (Figure 4).
Figure 5. Determination of a/b (y-intercept).
On the basis of this linear fitting procedure, the y-intercept,
a/b, was determined to be 0.94 0.06, which leads to a value
of k1c/k2c = 3.02
0.2. Therefore, we concluded that the
“catalysis” of the DBU-CO2 adduct is almost identicalor
slightly favoredfor R2MgBr, which is more electron-deficient
and, therefore, more susceptible to the catalytic role of DBU.
However, due to experimental errors, this can also be explained
by (1) almost no participation of DBU as a catalyst in the rate-
limiting carboxylation step; or (2) the effect of the catalyst
being too small to be measured, because of the already low
activation barrier of the Grignard addition reactions to CO2.
Additionally, the slope of the plot in Figure 5 is negative, which
would require that the b value must be negative (KDBU is a
positive quantity). A plausible explanation would be that k2c is
an apparent rate constant of a more complex underlying
mechanism. Therefore, further kinetic experiments with a more
complex kinetic expression would be required to uncover these
mechanisms.
A different view on the effect of the DBU-CO2 adduct can
still be obtained be rewriting eq 3 as,
k1c
[CO2(soln.)] + [DBUCO2]
Kr
K1,2
k1
=
k2c
[CO2(soln.)] + [DBUCO2]
k2
(8)
from where we immediately conclude that
Figure 4. Determination of k1/k2 in the absence of DBU.
l
k1c
o
o
o
Kr
k1c
k2c
k1
k2
k1
k2c
k2
o
o
o
o
o
o
> 1 ⇔
= 1 ⇔
< 1 ⇔
> 1 ⇔
= 1 ⇔
< 1 ⇔
>
=
<
On the basis of the reliability and reproducibility of our
model reactions, we then conducted carboxylation reactions in
the presence of different concentrations of DBU (0.067−0.37
M). The set of comparative Grignard carboxylation reactions
(12 points) provided the value of Kr (([P1][R2])/([P2][R1])).
Considering the variation of Kr/K1,2 as a function of the
variation of a function of [DBU], in this case (1 − Kr/K1,2)/
[DBU], a linear fitting procedure was performed, as shown in
Figure 5, resulting in a value of R2 = 0.70. According to eq 6,
the obtained slope corresponds to 1/(KDBUb), and the y-
intercept is the value of a/b. Using the obtained values for a/b
and K1,2, the value of k1c/k2c can be obtained with eq 7. A
problem associated with this model lies in the need for large
DBU concentrations or large differences between the Kr and
K1,2. In our experimental setup, measuring points at lower
DBU concentrations might lead to a larger weighting of those
points and, similarly, small differences between Kr and K1,2
would lead to small changes.
K
1,2
o
o
o
o
o
o
o
k1c
k1
k2c
k2
o
o
o
o
Kr
k1c
k2c
k1
k2
o
m
o
o
o
o
K
1,2
o
o
o
o
o
o
k1c
k1
k2c
k2
o
o
o
Kr
K1,2
k1c
k2c
k1
k2
o
o
o
o
o
o
o
o
(9)
n
Therefore, by simply looking at the behavior of Kr/K1,2 as a
function of [DBU], we can qualitatively evaluate the behavior
of k1c/k2c relative to k1/k2 and, thus, the effect of the DBU-CO2
adduct. As shown in Figure 6, and despite the existence of
some outliers (to be expected considering the difficulties in
reproducing organometallic reagents and their reactions), no
visible trend is seen in Kr/K1,2 as [DBU] increases, with the
D
https://dx.doi.org/10.1021/acs.organomet.9b00838
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
General Procedure for In Situ FT-IR Measurements. The FT-
IR spectra measured in situ were recorded using a Mettler Toledo
ReactIR 15 and processed with iC IR software (version 7.0.297.0).
The IR spectra were sampled in the spectral window between 3000 to
650 cm−1 with a DiComp (Diamond) probe and Silver Halide fiber.
The FT-IR spectra were recorded every 15 s with 50 scans. A blank IR
spectrum for the empty reaction vessel was recorded as background,
and the subsequent spectra of the solution, recorded before the start
addition of carbon dioxide, were used for solvent subtraction.
The React-IR probe was immersed into the solution for the same
length for each experiment, and the reaction vessel was placed at the
same height from the same stir plate in each measurement, to keep
the stirring rate consistent. The stirring rate was set to 900 rpm on the
stir plate and kept constant in all the experiments.
Procedure for the Grignard Reaction Competition Reaction
to Determine the Kinetic Rate Ratio between 4-Fluorophe-
nylmagnesium Bromide and 4-(Trifluoromethoxy)-
phenylmagnesium Bromide. A 5 mL vial equipped with a septum
and a magnetic stir bar was flame-dried and flushed with nitrogen five
times. Then, 0.25 mL of 4-fluorophenylmagnesium bromide in THF
solution (1.0 M) and 0.75 mL of 4-(trifluoromethoxy)-
phenylmagenesium bromide in THF solution (0.5 M) were charged
into the vial. In a separated flame-dried vial flushed five times with
CO2, 0.5 mL of CO2 (1 bar) was transferred and charged into the
reaction vial. The solution was stirred for 30 min before an aliquot
was quenched directly into MeOD-d4.
Figure 6. Kr/K1,2 as a function of [DBU].
observed values clustering around the ordinate Kr/K1,2= 1.
Considering eq 8, this again supports the notion that the
“catalysis” of the DBU-CO2 adduct is negligible.
Procedure for the Competition Reaction to Determine the
y-Intercept Using Equation 6 between 4-Fluorophenylmag-
nesium Bromide and 4-(Trifluoromethoxy)phenylmagnesium
Bromide. A 5 mL vial equipped with a cap with septum and a
magnetic stir bar was flame-dried and flushed with nitrogen five times.
A varying volume of DBU (5 to 55 μL) was charged to the reaction
vial before 0.25 mL of 4-fluorophenylmagnesium bromide in THF
solution (1.0 M) and 0.75 mL of 4-(trifluoromethoxy)-
phenylmagensium bromide in THF solution (0.5 M) was charged
into the vial. In a separated flame-dried vial flushed five times with
CO2, 0.5 mL of CO2 (1 bar) were transferred and charged into the
reaction vial. The solution was stirred for 30 min before an aliquot
was quenched directly into MeOD-d4.
CONCLUSIONS
■
Several aspects of the effect of nucleophilic bases, particularly
DBU, on the carboxylation reaction between Grignard reagents
and CO2 have been studied. Screening reactions with DBU as
an additive consistently showed higher yields of carboxylation
product in comparison to the absence of DBU and other
screened nucleophilic bases. In situ FT-IR experiments show
the fast reaction between CO2 and Grignard reagents and
illuminate the diffusion limitations of the reaction. CO2-
absorption measurements have shown that the diffusion of
CO2 into the solution improves significantly at specific
concentrations of DBU. The attempts to use a simplified
competitive kinetic analysis to determine the CO2 (de)-
activation using DBU illustrated the low energy barrier of the
Grignard carboxylation reaction.
Nonetheless, the observed effects of nucleophilic bases on
chemical yields and CO2 solubility indicate that the
nucleophilic bases for CO2 absorption are an important factor
in carboxylation reactions, particularly in the presence of
organometallic species and for enzymatic carboxylation.
Although the effect of DBU in the Grignard carboxylation in
the rate-determining or rate-limiting step does not seem to be
significant, the competitive kinetic analysis can be applied in
other catalytic carboxylation reactions with different types of
organometallic reagents. On the basis of this study, we are
currently evaluating other organometallic reagents, to further
investigate potential catalysis in carboxylation reactions. Our
model showed high reliability, considering the difficulties in
reproducing organometallic reagents and their reactions. On
this basis, further investigations on different Grignard reagents,
transition metal catalysts (e.g., TiIV),16 and organometallic
species are underway, to explain the true role of base catalysts
in CO2 (de)activation.
ASSOCIATED CONTENT
* Supporting Information
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.organomet.9b00838.
1H, 13C NMR, IR spectra and equation expressions
(PDF)
AUTHOR INFORMATION
Corresponding Authors
■
Han Yong Bae − Department of Chemistry, Sungkyunkwan
University, Jangan-gu, Suwon-si, Gyeonggi-do 16410, Korea;
Email: hybae@skku.edu
Ji-Woong Lee − Department of Chemistry, University of
Copenhagen, Copenhagen Ø 2100, Denmark; orcid.org/
0000-0001-6177-4569; Email: jiwoong.lee@chem.ku.dk
Authors
Jerik Mathew Valera Lauridsen − Department of Chemistry,
University of Copenhagen, Copenhagen Ø 2100, Denmark
Sung Yeon Cho − Department of Chemistry, Sungkyunkwan
University, Jangan-gu, Suwon-si, Gyeonggi-do 16410, Korea
EXPERIMENTAL SECTION
■
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.9b00838
Procedure for CO2 Saturation of THF. A flame-dried 20 mL vial
equipped with a magnetic stir bar and organic base, was flushed with
analytically pure CO2 three times, and was charged with THF. A CO2
balloon was attached, and the solution was left to stir for at least 30
min before usage.
Notes
The authors declare no competing financial interest.
E
https://dx.doi.org/10.1021/acs.organomet.9b00838
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
Hypothetical Role of CO2 in the Origin of Life. Synlett 2019, 30,
987−996.
ACKNOWLEDGMENTS
■
The generous support of the Department of Chemistry,
University of Copenhagen, the Novo Nordisk Fonden (Grant
No. NNF17OC0027598), the Villum Fonden (00019062),
and the Carlsberg Foundation (in situ infrared spectroscopy) is
gratefully acknowledged. H.Y.B and S.Y.C are grateful for the
financial support provided by the research programs through
the National Research Foundation of the Korean government
(2019R1F1A1046943 and 2019R1A6A1A10073079). The
authors appreciate Yang Yang for initial experimental
contributions and Paulo Monica for proofreading and
numerical analysis of kinetic models.
REFERENCES
■
(1) Seyferth, D. The Grignard Reagents. Organometallics 2009, 28,
1598−1605.
(2) Grignard, V. Nobel Lecture; NobelPrize.org. Nobel Media AB,
2019. See the following: https://www.nobelprize.org/prizes/
chemistry/1912/grignard/lecture/ (accessed 2019-07-24).
(3) Calvin, M.; Benson, A. A. The Path of Carbon in Photosynthesis.
Science 1948, 107, 476.
(4) Attwood, P. V. The Structure and the Mechanism of Action of
Pyruvate Carboxylase. Int. J. Biochem. Cell Biol. 1995, 27, 231−249.
(5) Stec, B. Structural mechanism of RuBisCO activation by
carbamylation of the active site lysine. Proc. Natl. Acad. Sci. U. S. A.
2012, 109, 18785.
(6) Indrakanti, V. P.; Kubicki, J. D.; Schobert, H. H. Photoinduced
activation of CO2 on Ti-based heterogeneous catalysts: Current state,
chemical physics-based insights and outlook. Energy Environ. Sci.
2009, 2, 745−758.
(7) Stephan, D. W.; Erker, G. Frustrated Lewis Pair Chemistry:
Development and Perspectives. Angew. Chem., Int. Ed. 2015, 54,
6400−6441.
(8) Ashley, A. E.; O’Hare, D. FLP-Mediated Activations and
Reductions of CO2 and CO. In Frustrated Lewis Pairs II: Expanding
the Scope; Erker, G., Stephan, D. W., Eds.; Springer Berlin Heidelberg:
Berlin, Heidelberg, 2013; pp 191−217.
́
́
(9) Courtemanche, M.-A.; Legare, M.-A.; Maron, L.; Fontaine, F.-G.
Reducing CO2 to Methanol Using Frustrated Lewis Pairs: On the
Mechanism of Phosphine−Borane-Mediated Hydroboration of CO2.
J. Am. Chem. Soc. 2014, 136, 10708−10717.
̀
́
(10) von Wolff, N.; Lefevre, G.; Berthet, J. C.; Thuery, P.; Can-Tat,
T. Implications of CO2 Activation by Frustrated Lewis Pairs in the
Catalytic Hydroboration of CO2: A View Using N/Si+ Frustrated
Lewis Pairs. ACS Catal. 2016, 6, 4526−4535.
(11) For a summary of recent reviews, see Table 1 in Yang, Y.; Lee,
J.-W. Toward ideal carbon dioxide functionalization. Chem. Sci. 2019,
10, 3905−3926.
(12) Peltzer, R. M.; Gauss, J.; Eisenstein, O.; Cascella, M. The
Grignard Reaction − Unraveling a Chemical Puzzle. J. Am. Chem. Soc.
2020, 142, 2984−2994.
(13) Soai, K.; Machida, H.; Yokota, N. Asymmetric conjugate
addition to unsaturated chiral amido alcohols using Grignard reagents
co-ordinated with tertiary amines or DBU. Preparation of optically
active 3-substituted carboxylic acids and (S)-(−)-citronellol. J. Chem.
Soc., Perkin Trans. 1 1987, 1, 1909−1914.
́
(14) Villiers, C.; Dognon, J.-P.; Pollet, R.; Thuery, P.; Ephritikhine,
M. An Isolated CO2 Adduct of a Nitrogen Base: Crystal and
Electronic Structures. Angew. Chem., Int. Ed. 2010, 49, 3465−3468.
(15) Rogers, H. R.; Hill, C. L.; Fujiwara, Y.; Rogers, R. J.; Mitchell,
H. L.; Whitesides, G. M. Mechanism of formation of Grignard
reagents. Kinetics of reaction of alkyl halides in diethyl ether with
magnesium. J. Am. Chem. Soc. 1980, 102, 217−226.
(16) (a) Juhl, M.; Lee, J.-W. Umpolung Reactivity of Aldehydes
toward Carbon Dioxide. Angew. Chem., Int. Ed. 2018, 57, 12318−
12322. (b) Juhl, M.; Kim, M. J.; Lee, H.-Y.; Baik, M.-H.; Lee, J.-W.
Aldehyde Carboxylation: A Concise DFT Mechanistic Study and a
F
https://dx.doi.org/10.1021/acs.organomet.9b00838
Organometallics XXXX, XXX, XXX−XXX