Mechanistic Study of the N-Formylation of Amines with Carbon Dioxide and Hydrosilanes

Article abstract of DOI:10.1021/acscatal.8b03274

N-formylation of amines with CO₂ and hydrosilane reducing agents proceeds via fast and complex chemical equilibria, which hinder easy analysis of the reaction pathways. In situ reaction monitoring and kinetic studies reveal that three proposed pathways, via direct- and amine-assisted formoxysilane formation (pathways 1 and 2, respectively) and via a silylcarbamate intermediate (pathway 3), are possible depending on the reaction conditions and the substrates. While pathway 1 is favored for non-nucleophilic amines in the absence of a catalyst, a base catalyst results in noninnocent behavior of the amine in the CO₂ reduction step toward the formoxysilane intermediate. The reaction pathway is altered by strongly nucleophilic amines, which form stable adducts with CO₂. Silylcarbamate intermediates form, which can be directly reduced to the N-formylated products by excess hydrosilane. Nevertheless, without excess hydrosilane, the silylcarbamate is an additional intermediate en route to formoxysilanes along pathway 2. Exchange NMR spectroscopy (EXSY) revealed extensive substituent exchange around the hydrosilane silicon center, which confirms its activation during the reaction and supports the proposed reaction mechanisms. Numerous side reactions were also identified, which help to establish the reaction equilibria in the N-formylation reactions.

Full text of DOI:10.1021/acscatal.8b03274

Subscriber access provided by Kaohsiung Medical University
Article
Mechanistic Study of the N-Formylation of
Amines with Carbon Dioxide and Hydrosilanes
Martin Hulla, Gabor Laurenczy, and Paul J. Dyson
ACS Catal., Just Accepted Manuscript • Publication Date (Web): 04 Oct 2018
Downloaded from http://pubs.acs.org on October 4, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 14
ACS Catalysis
1
2
3
4
5
6
7
8
9
Mechanistic Study of the N-Formylation of Amines with Carbon
Dioxide and Hydrosilanes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Martin Hulla, Gabor Laurenczy and Paul J. Dyson*
Institut des Sciences et Ingénierie Chimiques, École Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lau-
sanne, Switzerland
ABSTRACT: N-formylation of amines with CO2 and hydrosilane reducing agents proceeds via fast and complex chemical equilib-
ria, which hinder easy analysis of the reaction pathways. In-situ reaction monitoring and kinetic studies reveal that the three
proposed pathways, direct- or amine-assisted formoxysilane formation (pathways 1 and 2, respectively) or via a silylcarbamate
intermediate (pathway 3), are possible depending on the reaction conditions and the substrates. While pathway 1 is favored for
non-nucleophilic amines in the absence of a catalyst, a base catalyst results in non-innocent behavior of the amine in the CO2
reduction step towards the formoxysilane intermediate. The reaction pathway is altered by strongly nucleophilic amines, which
form stable adducts with CO2. Silylcarbamate intermediates form, which can be directly reduced to the N-formylated products by
excess hydrosilane. Nevertheless, without excess hydrosilane, the silylcarbamate is an additional intermediate en-route to
formoxysilanes along pathway 2. Exchange NMR spectroscopy (EXSY) revealed extensive substituent exchange around the hy-
drosilane silicon center, which confirms its activation during the reaction and supports the proposed reaction mechanisms. Nu-
merous side reactions were also identified, which help to establish the reaction equilibria in the N-formylation reactions.
Key words: Carbon dioxide, Reduction, N-formylation, Hydrosilane, Ionic liquid, Catalysis
form the formamide product. Hydrosilane activation by a nu-
Introduction
cleophilic catalyst and CO2 reduction by a hypervalent silicon
was proposed as the transition state (TS).22,24,26,31 The pathway
Amine reactions with CO2 and hydrosilanes have attracted
is supported by DFT calculations32 and the isolation and reac-
tivity of the formoxysilane intermediate.33 However, more re-
cent DFT calculations suggest that this pathway is favored in
the absence of a catalyst,34 but otherwise is likely to be slow.19
Pathway 2 also involves a formoxysilane intermediate and its
subsequent reaction with an amine to form the formamide
product. However, its formation is dependent on a catalyst
stabilized carbamate salt, which activates the hydrosilane in a
similar manner to that proposed for pathway 1 with a compa-
rable TS.35 DFT calculations of the reaction mechanism with
NHC-carbene and ZnCl2 catalysts indicate that the energy of
pathway 2 is lower than that for pathway 1.35,36 Nevertheless,
based on the N-formylation of morpholine with phenylsilane
and 1,1,3,3-tetramethylguanidine catalyst, stable carbamate
salts react with phenylsilane to form a silylcarbamate, rather
than the formoxysilane previously reported, pathway 3.19 The
silylcarbamate intermediate is then transformed to the forma-
mide product in the presence of CO2, either via the
formoxysilane or directly by a different mechanism. An alter-
native pathway, via water sensitive silylamines, towards the si-
lylcarbamate was also proposed,30 but has since been ruled out
as unfeasible under the reaction conditions normally em-
ployed and due to its extreme moisture sensitivity.19
considerable attention in recent years.1–3 Formamide, N-me-
thylamine and aminal products are highly valuable building
blocks in chemical synthesis.2,4 Moreover, CO2 is a cheap, non-
toxic and abundant C1 source5 and hydrosilanes are conven-
ient, mild and easy to use reducing agents (polymethylhy-
drosiloxane, PMHS, for example, is a waste product of the sil-
icon industry and is inexpensive).6 Consequently, a large num-
ber of catalysts have been developed for this reaction includ-
ing simple metal salts, complexes based on Rh,7 Ru,8 Re,9 W,10
Cu,11,12 Zn,13,14 and Fe,15 and transition metal-free catalysts such
as carbenes,6,16–18 guanidines,19 zwitterionic P-ylide-CO2 ad-
ducts,20 1,3,2-diazophospholenes,21 carboxylates,22,23 car-
bonates,24 ionic liquids,25,26 triazabicyclodecene (TBD),27,28
and γ-valerolactone.29 In addition, the solvents typically used,
i.e. DMSO and DMF, are catalytically active.30 Interestingly,
the organocatalysts can catalyze multiple CO2-hydrosilane re-
actions leading to more than one of the abovementioned
products or even all of them depending on the conditions em-
ployed.6,18,27,28,31
The observed selectivity, or lack of, generated interest in the
reaction pathways that lead to the individual formamide, N-
methylamine and aminal products.3 With N-formylated prod-
ucts, most recent evidence suggests that three possible path-
ways can take place involving two different intermediates
(Scheme 1).19 Pathway 1 involves direct reduction of CO2 by the
hydrosilane, to form a formoxysilane intermediate, followed
by a simple addition-elimination reaction with the amine to
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 14
ine31 and various carboxylates22 (Scheme 2). Clearly, the mech-
Scheme 1: N-formylation of amines with CO2 and hy-
drosilanes
1
2
3
4
5
6
7
8
anism of CO2 reduction and the formation of the
formoxysilane intermediate is key to understanding pathway
1. In addition, all N-formylation catalysts should catalyze the
reduction of CO2 to formoxysilane (or formic acid after hy-
drolysis) in the absence of an amine. Notably, direct reduction
of CO2 to formoxysilane was demonstrated with NHC-
carbenes37 and [TBA][F]38 catalysts, which provided part of the
basis for pathway 1.
Scheme 2: Pathway 1 for the N-formylation of amines
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a) General reaction scheme. b) Previously proposed reaction
pathways.
Irrespective of the reaction pathway, it became apparent that
the reaction proceeds by a series of fast and complex chemical
equilibria.19 This complexity, combined with the instability of
the intermediates, hinders identification of the reaction path-
way. Difficulties with the gas-liquid binary system largely re-
stricted mechanistic investigations to ex-situ analysis of the
1
reaction intermediates by H, 13C and 29Si NMR spectroscopy
and FTIR spectroscopy or stepwise addition experi-
ments.19,25,30,31 Hence, we decided to investigate and establish
the chemical equilibria as well as the reaction kinetics in-
volved in the N-formylation reaction in-situ using NMR spec-
troscopic techniques. The effects of the amine and hydrosilane
on the equilibria as well as the reaction pathway were investi-
gated and, finally, a number of conditions governing the reac-
tion pathway and observed intermediates were identified.
a) Pathway 1 reaction scheme b) Proposed mechanistic cycle
for the rate determining CO2 reduction step.23,35 X- represents
a general nucleophilic anion such as OAc-, F- or CO32- pro-
posed to follow pathway 1.
Experiments with [TBA][OAc] (10 mol%) showed that the salt
catalyzes CO2 reduction with phenylsilane, decreasing the
reaction time from over 13 h to < 6 h (Figures S2 and S9), to
afford a mixture of formoxysilanes, formic acid and, in
extremelly low yield, bis(silylacetals) and methoxysilane side-
products.37 Post-reaction analysis of the reduction of CO2 with
triethoxysilane by GC-MS also indicated the presence of
tetraethoxysilane (vide infra) and various siloxanes, which
could not be identified with certainty by NMR spectroscopy
due to extensive overlap of the peaks. Formic acid probably
results from formoxysilane hydrolysis due to trace water in the
reaction solvent as its concentration does not change after the
first 30 minutes of reaction. Bis(silylacetals), methoxysilanes
and siloxanes are products from the reduction of
formoxysilane with the hydrosilane.37 Silanols potentially
produced by formoxysilane hydrolysis were not detected
(Figure 1) indicating that they react further with
formoxysilanes to generate formic acid and siloxanes (Scheme
3). It should be noted that siloxanes can also be formed from
the reaction of silanols with (unreacted) hydrosilane
generating hydrogen as the by-product.39 Considering that
water is usually not rigourously excluded from N-formylation
reactions it is highly probable that formic acid is formed and
reduces the overall yield or at least the amount of hydrosilane
available for reaction.
Reactions were primarily carried out in DMSO-d6 with N-
methylaniline and phenylsilane at 25°C and 20 bar CO2 pres-
sure. However, other anilines as well as benzylamine and tri-
ethoxysilane were used to provide further insights. Low tem-
peratures and high pressures were used to suppress the for-
mation of N-methylamines and aminal side products, which
require higher temperatures and, based on the pathways pro-
posed, sequential reduction of CO2 by the hydrosilane.19,22,31
Tetrabutylammonium acetate ([TBA][OAc]) was selected as a
representative oxygen-containing nucleophilic,20,22–24 ionic
liquid-like,25,26 basic19,27,28 catalyst.
Results and discussion
Pathway 1 and CO2-hydrosilane chemical equilibria
In the absence of a catalyst, reduction of CO2 and the for-
mation of the formoxysilane intermediate was identified by
DFT calculations as the rate-determining step in the N-
formylation reaction of N-methylaniline with phenylsilane.34
Simultaneously, hydrosilane activation by nucleophilic cata-
lysts was proposed as the mode-operandi for a number of N-
formylation catalysts including NHC-carbenes,32 tetrabu-
tylammonium fluoride ([TBA][F]),26 Cs2CO3,24 glycine beta-
2
ACS Paragon Plus Environment

Page 3 of 14
ACS Catalysis
Scheme 3: CO2 reduction with phenylsilane at 20 bar pressure
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
a) CO2 reduction with phenylsilane reaction scheme. b) Observed side reactions.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Larger quantities of bis(silylacetals), methoxysilanes and
only broadening of the peaks in the 29Si NMR spectra is ob-
served, indicating close proximity of multiple signals and/or
continuous rapid exchange of the Si substituents. Incomplete
conversion of phenylsilane to PhSiH(OCHO)2 and
PhSiH2(OCHO) was ruled out by the absence of any Si-H sig-
nals in both the 1H and 29Si NMR spectra (Figure S7), indicat-
ing the formation of other formyl species in solution. Large
changes in the 1H NMR spectra were also observed in the aro-
matic region and for the signals corresponding to the
[TBA][OAc] catalyst (vide infra). Exchange spectroscopy
(EXSY) revealed interchange of the formyl groups between
formic acid and the three other –CHO containing compounds
as well as the exchange of phenyl groups between various spe-
cies in solution (Figure S5). Even at 218 K the -CHO signals
were broad and not fully resolved, confirming exchange on the
NMR time scale and limiting the analysis. The full extent of
the exchange was revealed with triethoxysilane (Scheme 4).
Triethoxysilane can reduce one CO2 molecule to
formoxysilane EtO3SiOCHO. However, in DMSO-d6 at room
temperature, five peaks attributable to the –CHO moiety at
8.13, 8.25, 8.31, 8.34 and 8.35 ppm gradually form, of which the
one at 8.35 ppm was consumed during the course of the reac-
tion and the one at 8.13 ppm corresponds to the C-H proton in
formic acid. In DMF-d7 upon cooling to 218 K the remaining
peaks were attributed to three different formoxysilanes (with
peaks at 8.44, 8.54 and 8.59 ppm) as well as to formic acid (at
8.32 ppm). EXSY at 258 K showed exchange of the formyl
group between formic acid and the formoxysilanes and also
exchange among the various formoxysilanes (Figure 2). Due to
extensive overlap of the peaks from the ethoxy substituents,
exchange of these groups could not be unambiguously con-
firmed, but is implied by the large number of ethoxy signals
and the detection of tetraethoxysilane (which gives peaks at
3.78 and 1.17 ppm in the 1H NMR spectrum and at -82 ppm in
the 29Si NMR spectrum). The presence of tetraethoxysilane
was further confirmed by GC-MS at the end of the reaction.
Finally, the formyl signals were assigned to formic acid, R3Si-
OCHO, R2Si(OCHO)2, RSi(OCHO)3 and Si(OCHO)4, where R
= Ph or EtO. The formation of these four formoxysilanes is ac-
companied by the simultaneous formation of SiR4 supported
by the observed exchange of phenyl groups and confirmed by
the formation of tetraethoxysilane.
siloxanes could also be expected due to lower pressures
usually employed, decreasing the rate of formation of
formoxysilane and increasing the rate of its reduction.19
The presence of formic acid in the reaction hinders accurate
analysis of the formoxysilane(s) formed as these are usually
identified by a distinct formyl (-CHO) proton signal in their
1H NMR spectra. In the reaction of phenylsilane at 25°C this
signal overlaps with the signal from formic acid resulting in a
broad peak (Figure 1), which, however, moves from 8.24 to 8.16
ppm as the reaction progresses. Hence, at 25°C in DMSO-d6
the presence of formoxysilane is only indicated by the uneven
integration of the acidic signal at 12.76 ppm and the broad –
CHO signal between 8.24 and 8.16 ppm.
1
Figure 1: H NMR spectrum of the reduction of CO2 with
phenylsilane after 13 h (i.e. at the end of the reaction).
Since broad signals are indicative of substituent exchange, the
reaction was gradually cooled to 218 K in DMF-d7 confirming
the presence of a dynamic exchange process between formic
acid and formoxysilane(s). Surprisingly, the broad singlet at-
tributed to –CHO splits into four different signals at 8.51, 8.69,
8.70 and 8.74 ppm rather than the expected two for formic
acid and the formoxysilane PhSi(OCHO)3 (Figures S3 and S4).
Similar changes were observed in the 13C NMR spectra, but
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 14
Scheme 4: Exchange reactions that take place during the reduction of CO2 with hydrosilanes
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Substituent exchange has been observed in the related hy-
drosilane reduction of carbonyls catalyzed by fluoride salts
and bases.40 The reaction proceeds by hydrosilane activation
with a nucleophilic catalyst and a five-membered hypervalent
silicon intermediate. Similarly, a five-membered silicon inter-
mediate was proposed for the hydrosilane reduction of CO2.26
DFT calculations demonstrated that a related five-membered
TS, rather than an intermediate, is energetically favored in
CO2 reductions.35 However, similar exchange behavior is still
expected and confirmed by the EXSY experiments further sup-
porting the proposed reaction mechanism. Hence, a binuclear
associative mechanism,40 possibly assisted by a bridging car-
boxylate anion may be involved in the exchange process. In
addition, the exchange can potentially be exploited for the
preparation of highly reactive formoxysilanes from hy-
drosilanes of inherently low reactivity, similarly to the re-
cently reported preparation of gaseous hydrosilanes from
alkoxyhydrosilane surrogates in the reduction of alkenes.41
of the acetate anion with the formate anion is expected with
the [TBA][OAc] catalyst. Indeed, with phenylsilane the ace-
tate peak broadened and shifted to higher frequency during
the reaction. Upon cooling to 218 K three partially overlapping
peaks are observed (Figure S4). In the reactions of triethox-
ysilane the acetate 1H NMR signal splits into two distinct peaks
at 1.91 and 2.09 ppm. The former corresponds to [TBA][OAc]
and the latter may be assigned to the silane bound acetate,
R3SiOAc, the assignment of which is supported by comparison
with separately prepared EtO3SiOAc.42 Dynamic exchange be-
tween the two species was confirmed by EXSY NMR experi-
ments (Figure S17). Incorporation of the anion in the substit-
uent exchange process was confirmed by the use tetrabu-
tylammonium formate ([TBA][HCO2]) instead [TBA][OAc].
With [TBA][HCO2], upon addition of CO2, the peak corre-
sponding to the formate anion disappears and only peaks at-
tributable to formoxysilanes and the tetrabutylammonium
cation are observed (Figures S21 and S22), confirming that the
anion of the catalyst is involved in the exchange equilibrium.
Proton exchange was also detected between the formyl –CHO
protons and the formic acid –OH protons, which cannot be
explained by a five-membered silicon mediated transfer. How-
ever, such exchanges are known for reversible reductions of
CO2 to formic acid with dihydrogen43 indicating, at least in
part, that formation of the formoxysilane is reversible, i.e. the
hydrosilane and CO2 can reform.
Finally, N-methylaniline was added to the reaction mixture
and the consumption of phenylsilane (Figure 3) as well as the
formation of N-methylformanilide was monitored. In the ab-
sence of the [TBA][OAc] catalyst, phenylsilane was consumed
at the same rate irrespective of whether N-methylaniline is
present or not. In addition, N-methylformanilide forms at a
comparable rate to formoxysilane. The rate of phenylsilane
consumption in the presence of the [TBA][OAc] catalyst is
considerably faster in the presence of N-methylaniline (com-
pared to the reaction without N-methylaniline). The
phenylsilane was completely consumed within 1 hour in con-
trast to 6 hours in the absence of N-methylaniline. Consider-
ing that formoxysilane formation was identified as the rate-
determining step of pathway 1,34 the absence of any interme-
diates is as expected. The equivalent CO2 reduction rates with-
out the [TBA][OAc] catalyst concur with DFT calculations
that pathway 1 is the preferred N-formylation route for N-
methylaniline in the absence of a catalyst.
Figure 2: EXSY NMR spectrum of the products from the reac-
tion of EtO3SiH and CO2 at 258 K. For full Spectrum Figure
S17.
DFT calculations suggest that an off-cycle species may be de-
rived en-route to formoxysilane in which the acetate anion at-
taches to the hydrosilane to eliminate a formate anion (INT1',
Scheme 2).23 This behavior was observed with iridium tri-
fluoroacetate complexes, where detachment of the ligand and
its reaction with hydrosilane and CO2 resulted in partial cata-
lyst deactivation.39 However, considering the substituent ex-
change process around the silicon center, a dynamic exchange
4
ACS Paragon Plus Environment

Page 5 of 14
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Figure 3: a) Reaction of phenylsilane with CO2 in the presence and absence of N-methylaniline. b) Example of the 1H NMR spectra
of the N-formylation of N-methylaniline with PhSiH3 under CO2 (20 bar) over a period of 13 h. For full spectra Figure S28. c)
Consumption of phenylsilane over the course of the reactions – see inset for details. General reaction conditions: phenylsilane (0.5
mmol), N-methylaniline (0.5 mmol), DMSO-d6 (0.5 ml), [TBA][OAc] (10 mol%), CO2 (20 bar), 298 K.
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
However, results with [TBA][OAc] demonstrate that N-
methylaniline provides an alternative, much faster pathway
for CO2 reduction. Similar observations were made when tri-
ethoxysilane was employed as the reducing agent (Figures S11
and S31).
tween the amine and the formed formoxysilane.35 Interest-
ingly, all the nucleophilic catalysts including [TBA][OAc],23
fluorides,26 carbonates24 and NHC-carbenes6,16–18,45 are also
basic. Furthermore, non-nucleophilic bases such as TBD27,28
and guanidines19 catalyze the N-formylation reaction. A gen-
eral base catalyzed reaction mechanism would thus encom-
pass the majority of organic and salt catalysts.
Overall, the reduction of CO2 with hydrosilanes results in a
mixture of formoxysilanes with continuous substituent ex-
change around the silicon center. An off-cycle species, R3Si-
OAc, forms during the reaction, resulting in the partial trans-
formation of the catalyst to [TBA][HCO2]. Formic acid and
various siloxanes are the main reaction side products with the
formation of bis(silylacetals) and methoxysilane in less than
2% yield. The formoxysilanes formed react with amines to
form the formamide products.33 However, pathway 1 is only
accessible for N-methylaniline in the absence of a catalyst as
proposed by DFT calculations.34 Although [TBA][OAc] can ac-
tivate hydrosilanes for direct CO2 reduction, N-formylation of
amines proceeds along a different, much faster pathway in the
presence of this catalyst.
Scheme 5: Pathway 2 for the N-formylation of amines
Pathway 2 and role of the amine
Pathway 2 was proposed based on computational studies and
the observation of the formoxysilane intermediate.35,36 In con-
trast to pathway 1, the catalyst acts to stabilize the formation
of a carbamate salt formed from the reaction of the amine with
CO2, rather than activate the hydrosilane. The reaction of
amines with CO2 to form carbamate salts is well established
and used on industrial scale to scrub CO2 from various gas
streams.44 However, not all amines form stable carbamate
salts, including N-methylaniline. Base catalysts help to stabi-
lize the salt, which then activates the hydrosilane in a similar
manner to that proposed in pathway 1 (Scheme 5). In addition,
a base catalyst can accelerate the subsequent reaction be-
a) Pathway 2 reaction scheme. b) Proposed mechanistic cycle
for the rate determining CO2 reduction step.35
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: a) N-formylation of N-methylaniline with CO2 and triethoxysilane and its side reactions and by-products. b) 1H NMR
spectra of N-methylaniline N-formylation with EtO3SiH under CO2 (20 bar) over a period of 13 h. For full spectra Figure S30. c)
Kinetic curves for the formylation of N-methylaniline using triethoxysilane. Reaction conditions: triethoxysilane (0.5 mmol), N-
methylaniline (0.5 mmol), DMSO-d6 (0.5 ml), [TBA][OAc] (10 mol%), CO2 (20 bar), 298 K.
Kinetic studies of N-methylaniline formylation with
phenylsilane and the [TBA][OAc] catalyst reveal that the
amine influences the rate of CO2 reduction with hydrosilanes
(Figure 3). In addition, the N-methylformanilide product
forms at the same rate as the phenylsilane is consumed and no
intermediates were detected. The only noticeable changes in
the NMR spectra during the course of the reaction is a broad-
ening of the –NH signal of the N-methylaniline and loss of its
coupling to the methyl group upon addition of CO2, indicating
partial formation of a carbamate salt. However, a peak corre-
sponding to the carbamate carbon was not detected by 13C
NMR spectroscopy, which supports the transient nature of
such a salt even in the presence of a base catalyst. Both path-
ways 2 and 3 were proposed to proceed by the formation of a
carbamate salt, with different intermediates derived from the
carbamate salt proposed for each route.19,35,36 Identifying the
intermediate, either a formoxysilane or a silylcarbamate, that
leads to the product is key for distinguishing between path-
ways 2 and 3. However, the only compounds identified were
phenylsilane, N-methylaniline, N-methylformanilide, formic
acid and some minor side products, i.e. bis(silylacetals) and
methoxysilane. The presence of formic acid slightly favors
pathway 2 over pathway 3 as the intermediate in formation of
formic acid is formoxysilane.46 However, the side products
and by-products observed during the formation of a silylcar-
bamate have not been previously reported and, as such, the
presence of formic acid does not distinguish between the two
pathways.
Ex situ reaction of the formoxysilane, EtO3SiOCHO, with N-
methylanililne to N-methylformamide has been reported.33
Monitoring the reaction of N-formylation with triethoxysilane
revealed the formation of at least three different
formoxysilanes, identified by the appropriate -CHO signals in
1
13
the H, C and 29Si NMR spectra (Figures S33-S35) and facili-
tated by comparison to spectra recorded for CO2 reduction re-
actions in the absence of N-methylaniline. Two of the three
formoxysilanes were gradually consumed during the reaction
at 298 K, whereas the remaining formoxysilane reacted rapidly
only after heating to 378 K. Simultaneous N-formanilide for-
mation was observed further supporting pathway 2 (Figure 4).
However, when all of the starting triethoxysilane was con-
sumed it became apparent that two equivalents of
formoxysilane are required to obtain N-methylformanilide as
the formoxysilanes were consumed at twice the rate at which
the N-methylformanilide was formed. As formic acid forms at
an equivalent rate to the N-methylformanilide and no silanol
by-product was observed, it seems likely that the silanol pro-
duced rapidly reacts with formoxysilane to generate various
siloxanes and formic acid, as observed in the earlier reactions.
6
ACS Paragon Plus Environment

Page 7 of 14
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Figure 5: a) Reaction of p-substituted anilines with CO2 and triethoxysilane. b) Example of the 1H NMR spectra of the N-formylation
of aniline with EtO3SiH under CO2 (20 bar) over a period of 13 h. For full spectra Figure S51. c) Consumption of triethoxysilane in
the N-formylation reactions of methoxyaniline, aniline, bromoaniline, trifluoromethylaniline and 4-aminobenzonitrile. Reaction
conditions: triethoxysilane (0.6 mmol), aniline (0.5 mmol), DMSO-d6 (0.5 ml), [TBA][OAc] (10 mol%), CO2 (20 bar), 298 K
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The presence of EtO3SiOSiOEt3 as well as higher weight silox-
anes was confirmed by GC-MS at the end of the reaction. Ex-
cess formic acid, in comparison to N-methylformanilide, may
be attributed to the rapid initial reaction of formoxysilane
with trace water present in the solvent. Overall, two equiva-
lents of hydrosilane and CO2 are required to produce one
equivalent of N-methylformanilide with the simultaneous for-
mation of formic acid and a siloxane (Figure 4a).
amine and the stability of the carbamate salt increases an in-
duction period for the N-formylation reaction is observed, fol-
lowed by a sudden and rapid depletion of the triethoxysilane
and amine, but without the formation of the product (Figures
S52 and S56). Instead, an intermediate presumed to be silyl-
carbamate,19 from which the product forms, was observed
(vide infra). Formoxysilanes were also detected, and as the ba-
sicity of the amine increases, they are consumed more rapidly.
Only 4-aminobenzonitrile and to some extent 4-trifluoro-
methylaniline undergo direct observable conversion of
formoxysilanes to their respective N-formylated products
(Figures S40 and S44). In all other cases, formation of a new
silylcarbamate intermediate dominates the reaction. Appar-
ently, as the basicity of the amine increases, pathway 2 is grad-
ually suppressed and pathway 3 becomes the dominate reac-
tion pathway.
The formation of formoxysilanes and their reaction with N-
methylaniline confirms that the N-methylformanilide is
formed via pathway 2 in the presence of the [TBA][OAc] cata-
lyst. In addition, pathway 2 proceeds via a similar five-mem-
bered hypervalent silicon TS to pathway 1 (Schemes 2 and 5),35
which allows for exchange around the silicon center, substan-
tiated by the multiple formoxysilanes observed rather than
EtO3SiOCHO alone. The substituent exchange process was
confirmed by EXSY experiments (Figure S36). Interestingly,
exchange between the acetate anion (observed at 1.93 ppm)
and silane bound acetate (observed at 2.09 ppm) also takes
place indicating that pathway 1 may compete with pathway 2
during N-methylaniline formylation or that the [TBA][OAc]
catalyst is, at least in part, transformed to [TBA][HCO2] or de-
activated following reaction of the acetate with the silicon
center.
Thus, N-methylaniline formylation proceeds via pathway 2, as
previously proposed by DFT calculations.35,36 A mixture of
formoxysilanes with different reactivities is formed, which
then react with N-methylaniline to form the desired product.
However, two equivalents of Si-H and CO2 are required for the
N-formylation reaction due to simultaneous formation of a si-
loxane and formic acid. Nevertheless, the reaction pathway is
not universal for all amines, but appears to operate only for
amines of low basicity, which do not form stable carbamate
salts. An alternative reaction pathway with an additional in-
termediate is responsible for the N-formylation of more basic
amines. The transition from pathway 2 to pathway 3 does not
appear immediately, but rather as a gradual transition, where
both pathways compete with each other with amines of inter-
mediate basicity.
To ascertain the role of the amine and the related carbamate
salt on the reaction a series of p-substituted anilines were N-
formylated under the standardized reaction conditions using
triethoxysilane. The progress of the N-formylation reactions
1
was monitored by H NMR spectroscopy and the rate of hy-
drosilane consumption recorded (Figure 5). Both the reaction
rate and the entire reaction kinetics demonstrated a strong
dependence on the nature of the amine. As the basicity of the
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6: a) N-formylation of benzylamine with triethoxysilane. b) 1H NMR spectra of benzylamine N-formylation with EtO3SiH
under CO2 (20 bar) over a period of 13 h. For full spectra Figure S58. c) Kinetic curves for the N-formylation of benzylamine with
triethoxysilane and CO2. Reaction conditions: triethoxysilane (0.5 mmol), benzylamine (0.5 mmol), DMSO-d6 (0.5 ml),
[TBA][OAc] (10 mol%), CO2 (20 bar), 298 K.
period, only 8% of the triethoxysilane reacts to form formic
Pathway 3
acid. Notably, traces of ethanol are also detected. The second
Pathway 3, involving a silylcarbamate intermediate, was pro-
phase comprises an extremely rapid reaction between trieth-
posed from the N-formylation of morpholine.19,30 Morpholine,
oxysilane and benzylcarbamic acid and, within 10 minutes, all
like some of the other anilines studied, forms a stable carba-
of the hydrosilane and 65% of benzylcarbamabic acid are con-
mate salt even in the absence of a base. Moreover, stabiliza-
sumed. However, the N-formylated product is not observed
tion of the carbamate salt was demonstrated as non-essential
and, instead, an intermediate identified as silylcarba-
and potentially degradatory for the N-formylation of morpho-
mate19,30,47,48 appears, supporting the proposed pathway 3.
line.19 Nevertheless, base catalysts including guanidines,19
Simultaneously, a large increase in the quantity of formic acid
24
TBD,27,28 Cs2CO3 or carboxylates22 catalyze the N-formyla-
is observed. The third stage of the reaction involves the grad-
tion of morpholine. However, the role of the base in pathway
ual conversion of the silylcarbamate to the N-benzylforma-
3 remains unclear. Indeed, little is known about pathway 3 in-
mide product.
cluding the mechanism by which the silylcarbamate is formed
Scheme 6: Proposed transformations of silylcarbmate
to N-formylated compounds along pathway 3
and its subsequent transformation into the N-formylated
product. Two possibilities were proposed for its transfor-
mation to the product (Scheme 6).19 One is direct and the
other includes the other known formoxysilane intermediate.
In both cases, CO2 was required for the transformation to pro-
ceed in the presence of the tetramethylguanidine catalyst.19
Understanding the silylcarbamate intermediate, its formation
and reactivity as well as the formoxysilane role and the reac-
tion delay is essential to complete pathway 3. For this purpose,
the reaction of benzylamine with triethoxysilane was studied,
with each step of the reaction analyzed.
The reaction of benzylamine with CO2 and triethoxysilane
may be divided into three phases (Figure 6). First, the induc-
tion period; upon addition of CO2 benzylamine is directly con-
verted to benzylcarbamic acid, rather than the previously pro-
posed benzylcarbamate, but then remains essentially un-
changed for the next 100 minutes of the reaction. During this
X and Y represent previously unreported reaction side prod-
ucts. Scheme adapted from reference 19.
8
ACS Paragon Plus Environment

Page 9 of 14
ACS Catalysis
Scheme 7: Proposed reactions involved in the water induced delay to the N-formylation reaction
1
2
3
4
5
6
7
8
9
10
11
12
13
14
a) Proposed series of reactions responsible for benzylamine N-formylation reaction delay. b) Overall observable reaction.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The initial reaction kinetics resemble product catalyzed reac-
tions, but upon addition of 20 mol% of benzylformanilide to
the reaction mixture no changes were observed (Figures S72
and S73). Addition of formic acid slightly accelerates the reac-
tion (Figures S74 and S75). However, the acceleration is insuf-
ficient to explain the observed 'step' change during N-formyla-
tion of benzylamine. One way in which formic acid can form
in the reaction is from the hydrolysis of formoxysilane with
trace water.38,46 Hence, we hypothesized that water causes a
reaction delay, which is the opposite to product catalyzed re-
actions with a similar kinetic profile. Indeed, addition of trace
amounts of water increases the induction time from 100
minutes to 300, slightly larger quantities to almost 400
minutes and further increases result in the slow, but complete
transformation of triethoxysilane to formic acid (Figures S76
– S81). In the reaction with excess water, N-formylation was
not observed and the reaction stopped after depletion of the
triethoxysilane. Large quantities of water also prevent the N-
formylation reaction when phenylsilane is used as the reduct-
ant. However, trace water and even quantities as large as 10
mol% do not affect the reaction kinetics19 as in the case of tri-
ethoxysilane. In the presence of the catalyst, phenylsilane rap-
idly reacts with water, removing it prior to reaction with
CO2.49,50 As a result, trace water only decreases the initial
quantity of phenylsilane in the reaction rather than causing a
reaction delay.
Scheme 9: Pathway 3 for the N-formylation of amines
Scheme 8: Calculated transition states and intermedi-
ates along pathway 2 for CO2 reduction with hy-
drosilanes35
a) Pathway 3 reaction scheme. b) Proposed mechanistic cycle
for the rate determining CO2 reduction step.
A second way in which formic acid can potentially form is
from the reaction of carbamic acid with the hydrosilane and
CO2, resulting in the simultaneous formation of the silylcar-
bamate. This reaction is actually observed in the second phase
of the N-formylation reaction, where silylcarbamate forms
along with large quantities of formic acid. However, in the
presence of water silylcarbamates are readily hydrolyzed back
to carbamic acids and silanols,51 which can react further with
silylcarbamates to form the corresponding carbamic acids and
EtO3SiOSiOEt3 (Scheme 7).52
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 14
Scheme 10: Proposed pathway 3' for the N-fomylation of amines
1
2
3
4
5
Considering that carbamic acids are required for the reaction
of the hydrosilane on the reaction rate is usually assigned to
delay, it appears that the reaction of carbamic acid with hy-
drosilane and CO2 followed by hydrolysis of the resulting si-
lylcarbamate is responsible for the induction period observed
during benzylamine N-formylation with triethoxysilane.
Thus, when all water and silanols are consumed the desired
N-formylation reaction can proceed, which is characterized by
rapid consumption of triethoxysilane, benzylcarbamic acid
and the concomitant formation of formoxysilanes and silylcar-
bamates.
the SN2-type TS formed during reduction of CO2 and, hence,
steric hindrance around the silicon center.26 However,
phenylsilane also has three Si-H bonds, which can act as a hy-
dride source, whereas triethoxysilane only has one. In effect,
with phenylsilane an excess of reducing agent is normally
used. Similarly, excess of triethoxysilane results in the com-
plete N-formylation of benzylamine within 4 hours in compar-
ison to incomplete reaction after 13 hours with only one equiv-
alent of the hydrosilane (Figure S82 – S83). Five equivalents of
triethoxysilane were used for the experiment, rather than
three, in order to compensate for the formation of
formoxysilane. Surprisingly, the excess triethoxysilane only
slightly decreases the initial reaction delay to 60 minutes and
does not affect the rate of silylcarbamate formation. However,
the excess hydrosilane strongly affects the rate of the transfor-
mation of the silylcarbamate to the N-formylated product.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The reactions are accompanied by characteristic substituent
exchange around the silicon center identified by the presence
of ethanol in the initial phases of the reaction and confirmed
by EXSY NMR experiments as well as the detection of tetra-
ethoxysilane (Figure S64). Hence, similar five-membered TSs
are expected for pathway 3 en-route to silylcarbamate as for
pathway 2 towards formoxysilanes. In the presence of an
amine, two such TSs as well as one high energy intermediate
were proposed by DFT calculations during formoxysilane for-
mation (Scheme 8).35 Moreover, the intermediate strongly re-
sembles the silylcarbamate observed during the N-formyla-
tion of benzylamine. The only difference is incomplete elimi-
nation of formate/formic acid from the intermediate. How-
ever, formic acid is also observed during the N-formylation of
benzylamine. Hence, it appears that stabilization of the carba-
mate (R2NCOO-) with basic amines results in the complete
transformation of this calculated intermediate to the experi-
mentally observed silylcarbamate and formic acid. The silyl-
carbamates then react with formic acid to form
formoxysilane(s). Hence, almost the same mechanism can be
proposed for pathway 3 as for pathway 2 (Scheme 9). The ob-
served difference between pathway 2 and 3 is dependent on
the relative stabilities of silylcarbamates and formoxysilanes,
which defines the major intermediate detected in solution.
For benzylamine and morpholine it is the silylcarbamate. A
mixture of silylcarbamates and formoxysilanes is formed with
p-substitued anilines. Finally, in the case of N-methylaniline
only formoxysilanes are detected. Nevertheless, the pathways
do not seem to substantially differ, at least during the reduc-
tion of CO2, and only the most stable intermediate along the
N-formylation pathway varies.
Direct conversion of silylcarbamates to the product has been
proposed.19,30 In addition, silylcarbamate cleavage with
chlorosilanes to siloxanes and isocyanates is known.51 The for-
mation of siloxanes also appears as the final reaction by-prod-
uct of the N-formylation reaction, which suggests that direct
reduction of silycarbamates with excess hydrosilane to N-
formylated compounds and siloxanes is possible. In the ab-
sence of excess hydrosilane the silylcarbamate is gradually
converted to formoxysilane leading to the slow formaton of
the N-formylated product. Whereas in the presence of excess
hydrosilane, the silylcarbamate is directly reduced to the
formamide, which corresponds to pathway 3' (Scheme 10).
Thus, pathway 2 and 3 share the same mechanism for the CO2
reduction step, but differ according to the relative stabilities
of the silylcarbamate and formoxysilane intermediates, which
defines the major intermediate observed in solution. Due to
the low stability of the carbamate linkage (NCOO-) with N-
methylaniline, no silylcarbamates are observed and only
formoxysilanes are detected. However, as the stability of car-
bamate increases, silylcarbamate formation becomes gradu-
ally favored until it becomes the dominant species in solution.
A mixture of silylcarbamates and formoxysilanes is observed
in-between the two extreme cases. The N-formylated product
forms by gradual conversion of silylcarbamate to
formoxysilane, which then reacts with the amine to form the
desired product. Formate or formic acid is required for silyl-
carbamate transformation to formoxysilane rather than CO2
as previously proposed.19 Nevertheless, in the presence of ex-
cess hydrosilane the silylcarbamate can be directly reduced to
the final N-formylated product without the need for
formoxysilane formation (Scheme 11).
In the final stage of the reaction, benzylformamide gradually
forms from the reaction of benzylamine with formoxysilane.
However, the rate of the reaction is dependant on the stability
of the silylcarbamate, which hinders formoxysilane formation.
Overall, benzylamine N-formylation with triethoxysilane pro-
ceeds at comparable rate to the N-formylation of N-methyl-
aniline. Nevertheless, with phenylsilane, N-formylation of
benzylamine is ~3.3 times faster (Figures S84 - 85). The effect
10
ACS Paragon Plus Environment

Page 11 of 14
ACS Catalysis
Scheme 11: Overview of N-formylation reaction pathways in relation to the conditions
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Formoxysilane and silylcarbamate side reactions with silanols are not shown for clarity.
water, which hydrolyses the formoxysilanes to silanols, silox-
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusions
anes and formic acid. In the case of silylcarbamate, water in-
duces a reaction delay as the silylcarbamate intermediate un-
dergoes hydrolysis back to the starting carbamate and si-
lanols, which further hinder the reaction. However, as
phenylsilane is the most commonly used hydrosilane the
amine to Si-H ratio is automatically augmented to 1 : 3 result-
ing in little significance of water induced side reactions.
Three possible pathways have been proposed for the N-
formylation of amines with CO2 and hydrosilanes. Pathway 1,
based on the direct insertion of CO2 into the hydrosilane, was
demonstrated to only operate for the N-formylation of low ba-
sicity amines in the absence of a catalyst. However, in the
presence of a base catalyst, N-formylation of N-methylaniline
follows pathway 2, where base stabilized carbamate salts acti-
vate the hydrosilane prior to CO2 insertion. Some evidence
suggests that direct hydrosilane activation with nucleophilic
base catalysts simultaneously takes place. However, as the ba-
sicity of the amine and the related carbamate stability in-
creases, pathway 1 is overshadowed by pathway 2, which in
turn is overtaken by pathway 3 in the case of highly basic
amines such as morpholine or benzylamine.
Finally, this study demonstrates that organic and salt N-
formylation catalysts act as bases rather than nucleophiles. A
general base-catalyzed mechanism indicates that a large num-
ber of bases will be catalytically active for the reaction. It is our
hope that these findings will facilitate the search for optimal
base catalysts and the development of even more efficient N-
formylation reaction with CO2.
Pathway 3 appears to follow a similar CO2 reduction mecha-
nism as pathway 2, but with increased stability of the carba-
mate linkage (NCOO-). As a result, stable silylcarbamates are
observed en-route to formoxysilanes and are detected in solu-
tion instead (or together) with formoxysilanes. The final N-
formylated compounds form either from the reaction of the
amine with formoxysilane or by the direct reduction of the si-
lylcarbamates with excess hydrosilane (pathway 3'). Conse-
quently, the N-formylation pathway is dependent on the
amine and the quantity of hydrosilane present. It is likely that
the substrate dependency might also explain the discrepancy
in proposed pathways towards N-methylamines and aminals,
which in the first step also originate from the reduction of CO2
to formoxysilanes, but may vary according to the stability of
the silylcarbamate.
AUTHOR INFORMATION
Corresponding Author
*pjd@epfl.ch
Notes
The authors declare no competing financial interests.
ASSOCIATED CONTENT
Supporting Information
Experimental procedures and NMR spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org.
ACKNOWLEDGMENT
For all CO2 reduction reactions with hydrosilanes, substituent
exchange is observed around the silicon center of the hy-
drosilane supporting the formation of a five-membered hyper-
valent silicon species during the reaction, either as an inter-
mediate or, more likely, as an SN2-type transition state.35 Si-
lanols, the primary reaction by-products, further react with
the formoxysilane intermediates resulting in formic acid and
siloxanes and a amine:hydrosilane (Si-H bond) ratio of at least
1 : 2 is required. The ratio needed increases in the presence of
We thank the EPFL and the Swiss Competence Center for En-
ergy research (SCCER) for funding.
REFERENCES
(1) Li, Y.; Cui, X.; Dong, K.; Junge, K.; Beller, M. Utilization of CO 2 as
a C1 Building Block for Catalytic Methylation Reactions. ACS Catal.
2017, 7, 1077–1086.
(2) Tlili, A.; Blondiaux, E.; Frogneux, X.; Cantat, T. Reductive
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 14
Functionalization of CO 2 with Amines: An Entry to Formamide,
Formamidine and Methylamine Derivatives. Green Chem. 2015, 17,
157–168.
Catalyzed N -Formylation of Amines. Angew. Chem. Int. Ed. 2015, 54,
12116–12120.
(22) Liu, X.-F.; Qiao, C.; Li, X.-Y.; He, L.-N. Carboxylate-Promoted
1
2
3
4
5
6
7
8
9
(3) Liu, X.-F.; Li, X.-Y.; Qiao, C.; He, L.-N. Transition-Metal-Free
Catalysis for the Reductive Functionalization of CO2 with Amines.
Synlett 2018, 29, 548–555.
(4) Klankermayer, J.; Wesselbaum, S.; Beydoun, K.; Leitner, W.
Selective Catalytic Synthesis Using the Combination of Carbon
Dioxide and Hydrogen: Catalytic Chess at the Interface of Energy and
Chemistry. Angew. Chem. Int. Ed. 2016, 55, 7296–7343.
Reductive Functionalization of CO with Amines and Hydrosilanes
2
under Mild Conditions. Green Chem. 2017, 19, 1726–1731.
(23) Li, X.-Y.; Zheng, S.-S.; Liu, X.-F.; Yang, Z.-W.; Tan, T.-Y.; Yu, A.;
He, L.-N. Waste Recycling: Ionic Liquid-Catalyzed 4-Electron
Reduction of CO 2 with Amines and Polymethylhydrosiloxane
Combining Experimental and Theoretical Study. ACS Sustain. Chem.
Eng. 2018, 6, 8130–8135.
(24) Fang, C.; Lu, C.; Liu, M.; Zhu, Y.; Fu, Y.; Lin, B.-L. Selective
Formylation and Methylation of Amines Using Carbon Dioxide and
Hydrosilane Catalyzed by Alkali-Metal Carbonates. ACS Catal. 2016,
6, 7876–7881.
(25) Hao, L.; Zhao, Y.; Yu, B.; Yang, Z.; Zhang, H.; Han, B.; Gao, X.; Liu,
Z. Imidazolium-Based Ionic Liquids Catalyzed Formylation of Amines
Using Carbon Dioxide and Phenylsilane at Room Temperature. ACS
Catal. 2015, 5, 4989–4993.
(26) Hulla, M.; Bobbink, F. D.; Das, S.; Dyson, P. J. Carbon Dioxide
Based N-Formylation of Amines Catalyzed by Fluoride and Hydroxide
Anions. ChemCatChem 2016, 8, 3338–3342.
(27) Das Neves Gomes, C.; Jacquet, O.; Villiers, C.; Thuéry, P.;
Ephritikhine, M.; Cantat, T. A Diagonal Approach to Chemical
Recycling of Carbon Dioxide: Organocatalytic Transformation for the
Reductive Functionalization of CO2. Angew. Chemie - Int. Ed. 2012, 51,
187–190.
(28) Frogneux, X.; Blondiaux, E.; Thuéry, P.; Cantat, T. Bridging
Amines with CO 2ꢀ: Organocatalyzed Reduction of CO 2 to Aminals.
ACS Catal. 2015, 5, 3983–3987.
(29) Song, J.; Zhou, B.; Liu, H.; Xie, C.; Meng, Q.; Zhang, Z.; Han, B.
Biomass-Derived γ-Valerolactone as an Efficient Solvent and Catalyst
for the Transformation of CO 2 to Formamides. Green Chem. 2016, 18,
3956–3961.
(30) Lv, H.; Xing, Q.; Yue, C.; Lei, Z.; Li, F. Solvent-Promoted Catalyst-
Free N-Formylation of Amines Using Carbon Dioxide under Ambient
Conditions. Chem. Commun. 2016, 52, 6545–6548.
(5) Aresta, M. Carbon Dioxide as Chemical Feedstock, First edit.;
Wiley-VCH Verlag GmBh: Weinheim, 2010.
(6) Jacquet, O.; Das Neves Gomes, C.; Ephritikhine, M.; Cantat, T.
Recycling of Carbon and Silicon Wastes: Room Temperature
Formylation of N-H Bonds Using Carbon Dioxide and
Polymethylhydrosiloxane. J. Am. Chem. Soc. 2012, 134, 2934–2937.
(7) Nguyen, T. V. Q.; Yoo, W.-J.; Kobayashi, S. Effective Formylation
of Amines with Carbon Dioxide and Diphenylsilane Catalyzed by
Chelating Bis( Tz NHC) Rhodium Complexes. Angew. Chem. Int. Ed.
2015, 54, 9209–9212.
(8) Li, Y.; Fang, X.; Junge, K.; Beller, M. A General Catalytic
Methylation of Amines Using Carbon Dioxide. Angew. Chem. Int. Ed.
2013, 52, 9568–9571.
(9) Morris, D. S.; Weetman, C.; Wennmacher, J. T. C.; Cokoja, M.;
Drees, M.; Kühn, F. E.; Love, J. B. Reduction of Carbon Dioxide and
Organic Carbonyls by Hydrosilanes Catalysed by the Perrhenate
Anion. Catal. Sci. Technol. 2017, 7, 2838–2845.
(10) Wang, M.-Y.; Wang, N.; Liu, X.-F.; Qiao, C.; He, L.-N. Tungstate
Catalysis: Pressure-Switched 2- and 6-Electron Reductive
Functionalization of CO 2 with Amines and Phenylsilane. Green Chem.
2018, 20, 1564–1570.
(11) Motokura, K.; Takahashi, N.; Kashiwame, D.; Yamaguchi, S.;
Miyaji, A.; Baba, T. Copper-Diphosphine Complex Catalysts for N-
Formylation of Amines under 1 Atm of Carbon Dioxide with
Polymethylhydrosiloxane. Catal. Sci. Technol. 2013, 3, 2392–2396.
(12) Zhang, S.; Mei, Q.; Liu, H.; Liu, H.; Zhang, Z.; Han, B. Copper-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Catalyzed N-Formylation of Amines with CO
under Ambient
(31) Liu, X.-F.; Li, X.-Y.; Qiao, C.; Fu, H.-C.; He, L.-N. Betaine Catalysis
for Hierarchical Reduction of CO 2 with Amines and Hydrosilane To
Form Formamides, Aminals, and Methylamines. Angew. Chem. Int.
Ed. 2017, 56, 7425–7429.
(32) Wang, B.; Cao, Z. Sequential Covalent Bonding Activation and
General Base Catalysis: Insight into N-Heterocyclic Carbene
Catalyzed Formylation of N–H Bonds Using Carbon Dioxide and
Silane. RSC Adv. 2013, 3, 14007.
2
Conditions. RSC Adv. 2016, 6, 32370–32373.
(13) Jacquet, O.; Frogneux, X.; Das Neves Gomes, C.; Cantat, T. CO2 as
a C1-Building Block for the Catalytic Methylation of Amines. Chem.
Sci. 2013, 4, 2127–2131.
(14) Luo, R.; Lin, X.; Chen, Y.; Zhang, W.; Zhou, X.; Ji, H. Cooperative
Catalytic Activation of Si−H Bonds: CO
-Based Synthesis of
2
Formamides from Amines and Hydrosilanes under Mild Conditions.
ChemSusChem 2017, 10, 1224–1232.
(15) Frogneux, X.; Jacquet, O.; Cantat, T. Iron-Catalyzed
(33) Liu, X.-F.; Ma, R.; Qiao, C.; Cao, H.; He, L.-N. Fluoride-Catalyzed
Methylation of Amines by Reductive Functionalization of CO with
2
Hydrosilylation of CO 2ꢀ: CO
Conversion to Formamides and
Hydrosilanes. Chem. - A Eur. J. 2016, 22, 16489–16493.
(34) Li, W.; Kim, C. K. A Theoretical Study on the Mechanism of
2
Methylamines. Catal. Sci. Technol. 2014, 4, 1529–1533.
(16) Riduan, S. N.; Ying, J. Y.; Zhang, Y. Solid Poly-N-Heterocyclic
Carbene Catalyzed CO2 Reduction with Hydrosilanes. J. Catal. 2016,
343, 46–51.
Methylation of N -Methylaniline with CO and Silyl Hydrides. Bull.
2
Korean Chem. Soc. 2017, 38, 12–18.
(35) Zhou, Q.; Li, Y. The Real Role of N-Heterocyclic Carbene in
Reductive Functionalization of CO2: An Alternative Understanding
from Density Functional Theory Study. J. Am. Chem. Soc. 2015, 137,
10182–10189.
(36) Li, W.; Kim, C. K. Theoretical Investigations on the Methylation
of NH Bond Using CO2 and Hydrosilane Catalyzed by ZincII
Complexes: Mechanism and Ligand Effect. J. CO2 Util. 2017, 20, 178–
189.
(37) Riduan, S. N.; Zhang, Y.; Ying, J. Y. Conversion of Carbon Dioxide
into Methanol with Silanes over N-Heterocyclic Carbene Catalysts.
Angew. Chem. Int. Ed. 2009, 48, 3322–3325.
(38) Motokura, K.; Naijo, M.; Yamaguchi, S.; Miyaji, A.; Baba, T.
Reductive Transformation of CO2 with Hydrosilanes Catalyzed by
Simple Fluoride and Carbonate Salts. Chem. Lett. 2015, 44, 1217–1219.
(39) Julián, A.; Guzmán, J.; Jaseer, E. A.; Fernández-Alvarez, F. J.; Royo,
R.; Polo, V.; García-Orduña, P.; Lahoz, F. J.; Oro, L. A. Mechanistic
(17) Jacquet, O.; Das Neves Gomes, C.; Ephritikhine, M.; Cantat, T.
Complete Catalytic Deoxygenation of CO
2 into Formamidine
Derivatives. ChemCatChem 2013, 5, 117–120.
(18) Das, S.; Bobbink, F. D.; Laurenczy, G.; Dyson, P. J. Metal-Free
Catalyst for the Chemoselective Methylation of Amines Using Carbon
Dioxide as a Carbon Source. Angew. Chem. Int. Ed. Engl. 2014, 53,
12876–12879.
(19) Nicholls, R. L.; McManus, J. A.; Rayner, C. M.; Morales-Serna, J.
A.; White, A. J. P.; Nguyen, B. N. Guanidine-Catalyzed Reductive
Amination of Carbon Dioxide with Silanes: Switching between
Pathways and Suppressing Catalyst Deactivation. ACS Catal. 2018, 8,
3678–3687.
(20) Zhou, H.; Wang, G.-X.; Zhang, W.-Z.; Lu, X.-B. CO 2 Adducts of
Phosphorus Ylides: Highly Active Organocatalysts for Carbon Dioxide
Transformation. ACS Catal. 2015, 5, 6773–6779.
(21) Chong, C. C.; Kinjo, R. Hydrophosphination of CO
and
Insights on the Reduction of CO to Silylformates Catalyzed by Ir-
2
2
Subsequent Formate Transfer in the 1,3,2-Diazaphospholene-
NSiN Species. Chem. - A Eur. J. 2017, 23, 11898–11907.
12
ACS Paragon Plus Environment

Page 13 of 14
ACS Catalysis
(40) Revunova, K.; Nikonov, G. I. Base-Catalyzed Hydrosilylation of
Ketones and Esters and Insight into the Mechanism. Chem. - A Eur. J.
2014, 20, 839–845.
(41) Buslov, I.; Keller, S. C.; Hu, X. Alkoxy Hydrosilanes As Surrogates
of Gaseous Silanes for Hydrosilylation of Alkenes. Org. Lett. 2016, 18,
1928–1931.
(42) Moni Chauhan; Bhanu P. S. Chauhan, A.; Boudjouk, P. An
Efficient Pd-Catalyzed Route to Silyl Esters. Org. Lett. 2000, 2, 1027–
1029.
(43) Mellmann, D.; Sponholz, P.; Junge, H.; Beller, M. Formic Acid as
a Hydrogen Storage Material – Development of Homogeneous
Catalysts for Selective Hydrogen Release. Chem. Soc. Rev. 2016, 45,
3954–3988.
347–352.
(47) Knausz, D.; Meszticzky, A.; Szak′acs, L.; Cs′akv′ari, B.; Ujsz′aszy,
K. Trimethylsilylated N- Alkyl-Substituted Carbamates I.Preparation
and Some Reactions. J. Organomet. Chem. 1983, 256, 11–21.
(48) Rohonczy, J.; Knausz, D.; Csákvári, B.; Sohár, P.; Pelczer, I.;
Párkányi, L. Trimethylsilylated N-Alkyl-Substituted Carbamates: III.
Structure of Trimethylsilyl-N,N-Diisopropyl Carbamate in Solid Phase
and in Solution. J. Organomet. Chem. 1988, 340, 293–302.
(49) Baines, J. E.; Eaborn, C. Organosilicon Compounds. Part XVI. The
Alkaline Hydrolysis of Triorganomonosilanes. J. Chem. Soc. 1955, 0,
1436–1441.
1
2
3
4
5
6
7
8
9
(50) Kaplan, L.; Wilzbach, K. E. Hydrogen Isotope Effects in the
Alkaline Cleavage of Triorganosilanes 1. J. Am. Chem. Soc. 1955, 77,
1297–1302.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(44) Dutcher, B.; Fan, M.; Russell, A. G. Amine-Based CO Capture
2
Technology Development from the Beginning of 2013—A Review. ACS
Appl. Mater. Interfaces 2015, 7, 2137–2148.
(45) Das, S.; Bobbink, F. D.; Bulut, S.; Soudani, M.; Dyson, P. J.
Thiazolium Carbene Catalysts for the Fixation of CO2 onto Amines.
Chem. Commun. 2016, 52, 2497–2500.
(46) Itagaki, S.; Yamaguchi, K.; Mizuno, N. Catalytic Synthesis of Silyl
Formates with 1 Atm of CO2 and Their Utilization for Synthesis of
Formyl Compounds and Formic Acid. J. Mol. Catal. A Chem. 2013, 366,
(51) Kraushaar, K.; Schmidt, D.; Schwarzer, A.; Kroke, E. Reactions of
CO2 and CO2 Analogs (CXY with X, Y = O, S, NR) with Reagents
Containing Si–H and Si–N Units. Adv. Inorg. Chem. 2014, 66, 117–162.
(52) Knausz, D.; Meszticzky, A.; Szakács, L.; Csákvári, B.
Trimethylsilylated N-Alkyl-Substituted Carbamates: II. Alcoholysis of
Trimethylsilyl N,N-Dimethylcarbamate. J. Organomet. Chem. 1984,
268, 207–211.
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 14
Insert Table of Contents artwork here
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
14
ACS Paragon Plus Environment