Selective Reduction of CO2 to CH4 by Tandem Hydrosilylation with Mixed Al/B Catalysts

Article abstract of DOI:10.1021/jacs.6b01497

This contribution reports the first example of highly selective reduction of CO₂ into CH₄ via tandem hydrosilylation with mixed main-group organo-Lewis acid (LA) catalysts [Al(C₆F₅)₃ + B(C₆F₅)₃] {[Al] + [B]}. As shown by this comprehensive experimental and computational study, in this unique tandem catalytic process, [Al] effectively mediates the first step of the overall reduction cycle, namely the fixation of CO₂ into HCOOSiEt₃ (1) via the LA-mediated C - O activation, while [B] is incapable of promoting the same transformation. On the other hand, [B] is shown to be an excellent catalyst for the subsequent reduction steps 2-4, namely the hydrosilylation of the more basic intermediates [1 to H₂C(OSiEt₃)₂ (2) to H₃COSiEt₃ (3) and finally to CH₄] through the frustrated Lewis pair (FLP)-type Si-H activation. Hence, with the required combination of [Al] and [B], a highly selective hydrosilylative reduction of CO₂ system has been developed, achieving high CH₄ production yield up to 94%. The remarkably different catalytic behaviors between [Al] and [B] are attributed to the higher overall Lewis acidity of [Al] derived from two conflicting factors (electronic and steric effects), which renders the higher tendency of [Al] to form stable [Al]-substrate (intermediate) adducts with CO₂ as well as subsequent intermediates 1, 2, and 3. Overall, the roles of [Al] and [B] are not only complementary but also synergistic in the total reduction of CO₂, which render both [Al]-mediated first reduction step and [B]-mediated subsequent steps catalytic.

Full text of DOI:10.1021/jacs.6b01497

Subscriber access provided by TULANE UNIVERSITY
Article
Selective Reduction of CO2 to CH4 by Tandem
Hydrosilylation with Mixed Al/B Catalysts
Jiawei Chen, Laura Falivene, Lucia Caporaso, Luigi Cavallo, and Eugene Y.-X. Chen
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b01497 • Publication Date (Web): 04 Apr 2016
Downloaded from http://pubs.acs.org on April 4, 2016
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 free 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 accessible to all
readers and 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.
Journal of the American Chemical Society 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 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Selective Reduction of CO to CH by Tandem Hydrosilylation with Mixed Al/B Catalysts
2
4
†
*,‡
§
‡
*,†
Jiawei Chen, Laura Falivene, Lucia Caporaso, Luigi Cavallo, and Eugene Y.ꢀX. Chen
†
Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523-1872, USA
‡
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
King Abdullah University of Science and Technology (KAUST), Physical Sciences and Engineering
Division, Kaust Catalysis Center (KCC), Thuwal 23955-6900, Saudi Arabia
§
Dipartimento di Chimica e Biologia, Università di Salerno, Via Papa Paolo Giovanni II, I-84084,
Fisciano, Italy
2
ABSTRACT: This contribution reports the first example of highly selective reduction of CO into CH4
via tandem hydrosilylation with mixed mainꢀgroup organoꢀLewis acid (LA) catalysts [Al(C F ) +
6
5 3
B(C F ) ] {[Al] + [B]}. As shown by this comprehensive experimental and computational study, in this
6
5 3
unique tandem catalytic process, [Al] effectively mediates the first step of the overall reduction cycle,
namely the fixation of CO into HCOOSiEt (1) via the LAꢀmediated C=O activation, while [B] is
2
3
incapable of promoting the same transformation. On the other hand, [B] is shown to be an excellent
catalyst for the subsequent reduction steps 2–4, namely the hydrosilylation of the more basic
intermediates [1 to H C(OSiEt ) (2) to H COSiEt (3) and finally to CH ] through the frustratedꢀLewisꢀ
2
3 2
3
3
4
pair (FLP)ꢀtype Si–H activation. Hence, with the required combination of [Al] and [B], a highly selective
hydrosilylative reduction of CO system has been developed, achieving high CH production yield up to
2
4
9
4%. The remarkably different catalytic behaviors between [Al] and [B] are attributed to the higher
overall Lewis acidity of [Al] derived from two conflicting factors (electronic and steric effects), which
renders the higher tendency of [Al] to form stable [Al]–substrate (intermediate) adducts with CO as well
2
as subsequent intermediates 1, 2 and 3. Overall, the roles of [Al] and [B] are not only complementary but
also synergistic in the total reduction of CO , which render both [Al]ꢀmediated first reduction step and
2
[
B]ꢀmediated subsequent steps catalytic.
*
Corresponding author. Eꢀmails: laura.falivene@kaust.edu.sa; eugene.chen@colostate.edu
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
Achieving efficient direct and catalytic reduction of CO into CH will have significant impact on
2
4
addressing two currently biggest issues facing humanity: global warming largely due to the rising level of
the greenhouse gas CO and increasing demand on clean energy such as solar energy and natural gas CH .
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
Although CO is considered to be an ideal renewable C feedstock for chemicals, materials, and fuels as it
2
1
1
is renewable, abundant, nonflammable, and inexpensive, it is a highly stable, inert molecule so that it has
been a challenge to develop technologically feasible and economically competitive methods to convert
2
CO into fuels, especially high energy density, deoxygenated fuels such as methane. Catalytic
2
hydrosilylation of CO , although with silanes as an expensive hydrogen source, is a thermodynamically
2
3
favored process with formation of stronger Si–O bonds, when compared to the catalytic hydrogenation of
4
5
6
7
8
9
10
CO . Currently, transitionꢀmetal catalysts based on metals such as Ru, Co, Rh, Ir, Ni, Cu and mainꢀ
2
11
group Zn have been utilized but achieved limited success in the catalytic hydrosilylation of CO to
2
2
lower oxidation state species with low selectivity for methane formation.
An emerging approach to enhancing catalytic performance of the CO hydrosilylation systems is to
2
1
2
13
couple the strong organoꢀLewis acid B(C F ) , which, when used alone, is incapable of reducing CO ,
6
5 3
2
13
14
15
16
with transition metal (TM) complexes of Zr, Pt, Pd, Re and Sc for fixation of CO (by TM
2
complexes) and subsequent reduction of the more basic formate, aldehyde and methanol to CH (by the
4
borane). More recently, there is emerging interest in the development of TMꢀfree catalysis spotlighted by
17
frustrated Lewis pairs (FLPs). O’Hare et al. reported CO can be hydrogenated to CH OH with the
2
3
1
8
19
2,2,6,6ꢀtetramethylpiperidine (TMP)/B(C F ) FLP system. Alternatively, Stephan et al. demonstrated
6
5 3
the combination of PMes /AlX (Mes = mesityl, X = Cl, Br) effects the reduction of CO to CH OH with
3
3
2
3
H NBH as a hydrogen source. However, both of these transformations require a stoichiometric amount
3
3
of the FLP reagents. In this regard, Fontaine et al. discovered an ambiphilic Lewis pair system with a less
Lewis acidic catecholboryl unit which can promote the release of the reduced products from the catalyst
2
0
and thus render the hydroboration of CO catalytic. Ever since, major experimental and computational
2
2
ACS Paragon Plus Environment

Page 3 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
21,22
efforts have been directed to the study of the CO reduction utilizing the FLP chemistry.
On the other
2
hand, only a few examples of catalytic hydrosilylation of CO have surfaced in the literature by utilizing
2
either the separate reactivity of Lewis acid (LA)/Lewis base (LB) or the combination of them. For
2
3
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
instance, Ying et al. reported NHCꢀcatalyzed CO reduction with H SiPh , while Piers utilized B(C F )
2 2 2 6 5 3
2
4
(in excess) and TMP for deoxygenative hydrosilylation of CO . Müller et al. demonstrated that silyl
2
cations are effective in promoting conversion of CO into benzoic acid, formic acid and methanol, albeit
2
25
not in a catalytic fashion. More recently, Wehmschulte et al. reported the LAꢀcatalyzed hydrosilylation
+
26
of CO by cationic aluminum species AlR (R = Et or OAr). However, the detailed mechanism of such
2
2
LAꢀcatalyzed CO reduction remains unclear and thus further enhancing catalytic performance through
2
modification of catalyst structure seems challenging. In this context, it is of great interest and importance
to survey suitable main group LA candidates for effective CO reduction. Inspired by the dual activation
2
of CO and silane R SiH {[SiH]} by the above overviewed TM or TM/B(C F ) systems, we envisioned
2
3
6 5 3
27
that a combination of the more oxophilic and higher Lewis acidic Al(C F ) {[Al]}, which could be
6
5 3
sufficiently potent to render CO fixation into silyl formate, with the less oxophilic B(C F ) {[B]}, which
2
6 5 3
2
8
favors FLPꢀtype [SiH] activation for subsequent steps of reduction, could serve as a mainꢀgroup tandem
FLP system that effectively converts CO to CH . The tunability of [Al]/[B] catalysts should allow us to
2
4
develop more efficient, economical and recyclable tandem LA catalysts based on interꢀ or intramolecular
and heterogeneous platforms. Accordingly, this contribution reports the first such mainꢀgroup tandem LA
catalytic system for highly effective and selective CO reduction into CH through hydrosilylation using a
2
4
pair of [Al] and [B] LAs (Chart 1) which, when each used only, is ineffective (in case of [B]) or only
marginally effective (in case of [Al]) for catalyzing this transformation. In this context, we present herein
a full account of our combined experimental and theoretical/computation investigations into this novel
mainꢀgroup [Al]/[B] tandem catalyst system.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Tandem Catalysis: [Al] = Al(C
F
)
; [B] = B(C
F )
6 5 3
6
5
3
Et SiH
3
[
Al]-catalyzed
Step 1: CO -[Al]
2
(
C=O activation)
HCOOSiEt (1)
3
Step 2:
Step 3:
Step 4:
Si-H-[B]
Si-H-[B]
Si-H-[B]
H C(OSiEt ) (2)
2
3 2
[
B]-catalyzed
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
Si-H activation)
H COSiEt (3)
3
3
CH (4)
4
Chart 1. Proposed fundamental steps involved in the mixed main group [Al]/[B] LA system for tandem
catalytic hydrosilylation of CO into CH .
2
4
Results and Discussion
Fixation of CO with {Et SiH + [Al]} and Selective Reduction of CO to CH by {Et SiH +
2
3
2
4
3
29
[
Al]/[B]}. As [B] was shown to be ineffective in catalyzing reduction of CO via hydrosilylation, we
2
turned our attention to [Al], a stronger LA based on several lines of experimental and theoretical
2
7, 30
evidence.
Most recently, we have revealed that the Al center of the even unsolvated Al(C F ) is not
6 5 3
truly tricoordinated, but it adopts a dimeric structure with double Al···F(ortho) interactions in solid
31
state. Such weak interactions are readily destroyed by addition of more electronꢀdonating substrates than
3
0a, 31ꢀ32
[
Al] itself, such as toluene, ferrocene, or even fluorosilane and hydrosilane.
For instance, the
unsolvated [Al] forms a stable and isolable adduct with Et SiH in hexanes, while the interaction between
3
2
8h
[B] and Et SiH could only be detected by indirect spectroscopic evidence. These observations suggest
3
that the activation of the Si–H bond by [Al] is stronger than by [B]. However, the hydrosilylation of
ketones by Et SiH proceeds more efficiently with [B] than with [Al], due to the high Lewis acidity and
3
oxophilicity of [Al] that favor its complexation with the ketone substrates, greatly suppressing the Si–H
bond activation. On the other hand, the feasible dissociation between the carbonyl substrates and [B]
(
even at a small ratio) enables the activation of the Si–H bond by the free [B] and the subsequent
2
8j
hydrosilylation. Notably, Stephan et al. showed that [Al] reacts with CO at 90 °C to form a dimeric Al
2
33
compound [(C F )C(O)OAl(C F ) ] by insertion of CO into the Al–C F bond, while Müller reported
6
5
6
5 2 2
2
6 5
4
ACS Paragon Plus Environment

Page 5 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
that a [Si–H···Al] system with a coordinatively saturated, less Lewis acidic aluminum center is inactive
34
for CO fixation.
2
Excitingly, the stoichiometric reaction between [Et SiH·Al(C F ) ] and CO at room temperature
3
6
5 3
2
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
(RT) (entry 1,Table 1), revealed the rapid disappearance of the silane Si–H signal at 3.75 ppm, and
1
13
concomitant appearance of new signals at 8.17 ppm in H NMR, at 172.8 ppm in C NMR, and at 46.0
2
9
ppm (downfield shifted from 0.0 ppm from the silane) in Si NMR (Figure 1). These spectroscopic
features are consistent with the generation of a complexed silyl formate species, which was identified as
the complex HCOOSiEt ꢀ[Al] (1-[Al]), the structure of which was confirmed by singleꢀcrystal Xꢀray
3
diffraction (SCꢀXRD) analysis (Chart 2). No further reduction to lower oxidation state species was
detected. With 10 mol% of [Al], the generation of 1-[Al] (10% based on Et SiH) was observed but no
3
i
further silane conversion was achieved. Other silanes such as Pr SiH and PhMe SiH also led to the
3
2
formation of the corresponding silyl formateꢀ[Al] complexes (see the Supporting Information). These
observations suggest that the activation of CO by [Al] likely occurs through the typical LAꢀcarbonyl
2
activation and the LA is not released from the resulting product under these conditions, thus requiring a
stoichiometric amount of LA. To render this reaction catalytic and possibly promote further
hydrosilylation of the formed intermediates, we heated the above mixture to 80 °C for 24 h, achieving a
moderate conversion of 54% (calculated from the silane consumption) with a 16% CH yield (entry 4b).
4
These results imply that the catalytic performance of [Al] decreases as the basicity of the subsequent
reaction intermediates [HCOOSiEt , H C(OSiEt ) and H COSiEt , vide infra] increases. Intriguingly,
3
2
3 2
3
3
when we employed a mixed LA system containing 5 mol% [Al] + 5 mol% [B], quantitative conversion of
the silane with 82% CH yield was achieved at 80 °C in 5 h (entry 6, Table 1; Figure 2).
4
Table 1. Selected Results of CO Reduction via Hydrosilylation by [Al] and [Al]/[B]
2
a
b
Entry
Silane
Cat.
Conditions
12 h, 80 °C
12 h, 80 °C
1 h, RT
10 h, 80 °C
24 h, 80 °C
[SiH] Conv. (%)
CH Yield (%)
4
1
2
3
Et SiH
ꢀꢀ
0
2
98
39
54
0
0
0
14
16
3
Et SiH
5% [B]
100% [Al]
10% [Al]
10% [Al]
3
Et SiH
3
4a
Et SiH
3
4b
Et SiH
3
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
Et SiH
5.0% [Al] + 5.0% [B] 1 h, RT
15
100
7
82
3
6
Et SiH
5.0% [Al] + 5.0% [B] 5 h, 80 °C
3
a
b
Based on silane consumption. See the SI for yield calculation details.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chart 2. CO insertion into Si–H bond with the [Et SiH + Al(C F ) ] system to form the [Al]ꢀ
2
3
6 5 3
coordinated silyl formate and the corresponding solid state structure determined by SCꢀXRD.
1
13
29
Figure 1. Comparisons of H, C and Si NMR spectra before (bottom) and after (top) CO fixation by
2
[
Et SiH + Al(C F ) ] in C D Br at 25 °C.
3
6
5 3
6
5
6
ACS Paragon Plus Environment

Page 7 of 31
Journal of the American Chemical Society
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
1
Figure 2. H NMR (25 °C, C D Br) spectra of hydrosilylation of CO catalyzed by the mixed Al/B
6
5
2
system (Entries 3 and 4, Table 1) at different time intervals (•: HCOOSiEt ꢀ[Al], ‡: H C(OSiEt ) ꢀ[Al], †:
3
2
3 2
free H C(OSiEt ) and *: toluene from [Al] and NMR solvent residue).
2
3 2
Isolation of Reaction Intermediates. To elucidate the possible mechanism for this unique [Al]/[B]
tandem catalyst system, we synthesized each of the intermediates or byproducts involved in the above
reduction scheme, including HCOOSiEt , H C(OSiEt ) , H COSiEt and Et SiOSiEt (Scheme 1).
3
2
3 2
3
3
3
3
Compounds HCOOSiEt and H COSiEt were readily prepared from the dehydrogenative coupling of the
3
3
3
26b
corresponding precursors by Pd/C or [B].
On the other hand, successful examples or selective
5
a, 15ꢀ16
formation and isolation of H C(OSiEt ) remain scarce.
Piers et al noted that the sequestration of the
2
3 2
1
6a
[B] catalyst to prevent further hydrosilylation is necessary. Interestingly, in our attempts to reduce
HCOOSiEt with 1 equiv of Et SiH, we observed selective formation of H C(OSiEt ) in 5 min, after
3
3
2
3 2
which the product gradually underwent further rearrangement to form 1,3,5ꢀtrioxane (a formaldehyde
equivalent) and hexaethyldisiloxane Et SiOSiEt . This process is believed to proceed through
3
3
coordination of the catalytic amount of [B] to the substrate, as sequestration of [B] with DABCO (1,4ꢀ
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
diazabicyclo[2.2.2]octane) prevented the aforementioned rearrangement reaction and enabled the
isolation of H C(OSiEt ) . In a separate NMR scale experiment, H C(OSiEt ) was left for 1 day in
2
3 2
2
3 2
C D Br and remained intact, but addition of 5 mol% of [B] converted H C(OSiEt ) into 1,3,5ꢀtrioxane
6
5
2
3 2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
and Et SiOSiEt in 1 h (Figures S21 and S22). Indeed, the formation of the formaldehyde intermediate
3 3
even without a catalyst is energetically feasible for bis(phenylsilyl)acetal H C(OSiH Ph) , via
2
2
2
35
intramolecular elimination of PhH SiOSiH Ph, as predicted by calculations reported by Wang et al. Our
2
2
proposed mechanism through coordination of H C(OSiEt ) to [B] will be discussed in the following
2
3 2
sections. The final disiloxane product of CO reduction, Et SiOSiEt , can also be isolated as the
2
3
3
byproduct by vacuum distillation.
Pd/C
1) [B]; 2) DABCO
HCOOH + Et SiH
HCOOSiEt (1)
3
3
1
+ Et SiH
H C(OSiEt ) (2)
2 3 2
3
[B]
CH OH + Et SiH
3
3
H COSiEt (3)
3 3
O
O
C
[B]
[
H CO]
2
+
2
O
O
H
H
Et SiOSiEt3
3
Scheme 1. Preparation of the Reaction Intermediates Involved in the Overall CO Reduction Cycle.
2
Binding Interaction between the Intermediates and [Al] or [B]. With all three intermediates in
hand, we studied their interaction with [Al] and [B] (Scheme 2). As expected, the less sterically hindered
HCOOSiEt forms a stable adduct with both [Al] and [B]. Crystalline 1-[Al] and 1-[B] adducts can be
3
isolated in pure form by crystallization of equimolar 1 and [Al] or [B] in hexanes at –35 °C. Further
supporting evidence came from multiꢀnuclear spectroscopic data (Figures S1ꢀ9) chiefly: 1) the aldehyde
1
19
proton appears at 8.17 and 7.95 ppm in H NMR for 1-[Al] and 1-[B], respectively; 2) the F NMR
spectra also exhibit typical patterns for tetracoordinate aluminum and boron species; 3) both the formate
13
29
and the LA signals are present in the C NMR and assigned; and 4) the Si NMR signals are
significantly downfield shifted for 1-[Al] ( 46.0 ppm) and 1-[B] ( 46.4 ppm) when compared to the free
δ
δ
8
ACS Paragon Plus Environment

Page 9 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
(
δ
26.6 ppm). Worth noting here is that the spectroscopic data of 1-[Al] is consistent with those
generated by the stoichiometric reaction between CO and [Et SiH·Al(C F ) ].
2
3
6
5 3
[
Al]
[B]
O
C
O
C
O
C
O
C
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
[Al]
OSiEt3
[B]
OSiEt3
H
H
OSiEt
3
H
H
OSiEt3
[B]
1-[Al]
1-[B]
1
1
[
Al]
OSiEt3
C
OSiEt3
C
OSiEt3
C
OSiEt3
C
[B]
[Al]
H
OSiEt3
H
OSiEt3
H
H
OSiEt3
H
H
OSiEt
3
H
H
2
2
2-[Al]
2-[B]
[
Al]
[B]
OSiEt3
OSiEt3
OSiEt3
OSiEt3
C
[B]
[
Al]
C
C
C
H
H
H
H
H
H
H
H
H
H
H
H
3
3
3-[Al]
3-[B]
[Al]
[B]
3
Et SiOSiEt3
Et SiOSiEt3
3
Scheme 2. Binding Interaction of each Intermediates with [Al] and [B] Catalysts.
The direct structural evidence for the 1:1 complexation in 1-[Al] and 1-[B] was derived from the
SCꢀXRD analysis (Figure 3). In the structure of 1-[Al], the Al–O2 bond is rather strong, as indicated from
the short bond length of 1.8532(16) Å. Notably, the C1–O2 and C1–O1 distances are 1.242(2) and
1.277(2) Å. This specific alternation suggests that C1–O2 features a more double bond character while
C1–O1 features a more single bond character. In another word, if 1-[Al] is viewed as a model in which
+
+
[
Al] and [Si] are competing for the middle formate anion, then [Si] exhibits stronger interaction with the
anion. Nonetheless, both C1–O2 and C1–O1 distances are shorter than the typical C–O single bond (c.f.
1.43 Å) and longer than the C=O double bond (c.f. 1.23 Å), indicative of a certain degree of conjugated
resonance structure along the –O1–C1–O2– moiety. The middle C1 carries a partial positive charge,
accounting for the further secondary interactions with two paraꢀF atoms [C1···F contact: 2.908(3) and
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
.935(3) Å] from two neighboring molecules. The structural parameters for 1-[B] are overall similar to
those of 1-[Al]: C1–O1 1.276(2), C1–O2 1.245(2), B1–O2 1.621(2), except that secondary C···F contacts
2.748(2) and 2.838(3) Å] were derived from an intramolecular fashion with the adjacent ꢀC F ring
[
6
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
instead, presumably due to the smaller radius of boron and shorter bond lengths around it. Further
evidence for activation of the carbonyl moieties by LAs was deduced from FTꢀIR analysis of the solid
ꢀ1
samples of 1-[Al] and 1-[B] (Figure S32). When compared with the uncoordinated 1 (c.f. 1798 cm ), both
ꢀ1
ꢀ1
1-[Al] (1604 cm ) and 1-[B] (1597 cm ) exhibit significant shifts of the C=O stretching mode to lower
ꢀ
1
frequencies by 204 and 211 cm , respectively.
Figure 3. Xꢀray crystal structures of 1-[Al] and 1-[B] with thermal ellipsoids drawn at the 50%
probability. Hydrogen atoms except H1 were omitted for clarity. Selected bond lengths (Å) and angles (º)
for 1-[Al]: C1–O1 1.277(2), C1–O2 1.242(2), Al1–O2 1.8532(16), Si1–O1 1.7717(15), C1···F8’ 2.935(3)
and C1···F13’ 2.908(3); for 1-[B]: C1–O1 1.276(2), C1–O2 1.245(2), B1–O2 1.621(2), Si1–O1
1.7499(14), C1···F5 2.838(3) and C1···F11 2.748(2).
In contrast, the interaction of [Al] and [B] with the bulkier and less basic H C(OSiEt ) and
2
3 2
H COSiEt is different (see Scheme 2 & SI). In the case of the more sterically hindered and less acidic [B]
3
3
1
(
relative to [Al]), there was no spectral change from the H NMR signals for its 1:1 mixture with
1
0
ACS Paragon Plus Environment

Page 11 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
H C(OSiEt ) or H COSiEt . However, based on the fact that H C(OSiEt ) readily undergoes the
2
3 2
3
3
2
3 2
rearrangement reaction in the presence of [B] in a prolonged period, it is likely that [B] can bind to
H C(OSiEt ) , albeit in a very weak fashion that favors the dissociated form. Given that the bulkiness of
2
3 2
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
H COSiEt is less than that of H C(OSiEt ) , it is plausible that [B] can also bind to H COSiEt reversibly,
3 3 2 3 2 3 3
1
even though the H data again indicated that this equilibrium strongly favors the dissociated form. To the
contrary, the NMR scale reaction between [Al] and H C(OSiEt ) or H COSiEt results in spectral shifts
2
3 2
3
3
1
in the H signals, corresponding to complexation between [Al] and H C(OSiEt ) or H COSiEt . In
2
3 2
3
3
particular, [Al] forms isolable crystalline complex with H COSiEt . We tried to perform SCꢀXRD
3
3
analysis of the complex but failed to obtain satisfactory results after several attempts due to poor crystal
quality. Nonetheless, the overall molecular skeleton and atom connectivity of this adduct could be
recognized from the crude data (Figure S34). Collectively, these observations further support that in the
subsequent reduction of lowerꢀoxidationꢀstate species HCOOSiEt , H C(OSiEt ) and H COSiEt , [B]
3
2
3 2
3
3
should serve as a better catalyst than [Al] due to the weaker substrateꢀcatalyst interaction, which
facilitates the FLPꢀtype Si–H activation. In addition, the byproduct Et SiOSiEt does not form a
3
3
detectable adduct with either [B] or [Al], which enables [Al] and/or [B] to reenter the catalytic cycle after
the last step (CH generation).
4
Kinetic and Mechanistic Studies of Hydrosilylation of CO . To further address our hypothesis
2
on the catalytic roles of each LA in the current mixed tandem [Al]/[B] system, we carried out a kinetic
study on each step of the hydrosilylation (Scheme 3), coupled with a computational investigation (vide
infra). As described previously, in step 1, the fixation of CO is mediated by [Al] to form 1-[Al]. The
2
following hydrosilylation steps 2-4 proceed more efficiently with [B] than with [Al]. Among them, the
reactions of HCOOSiEt or H COSiEt with 1.0 equiv of Et SiH in the presence of 5 mol% [B] were
3
3
3
3
completed within 5 min at RT, while the slowest step, the hydrosilylation of H C(OSiEt ) under similar
2
3 2
conditions generated a mixture of unreacted H C(OSiEt ) , CH , and a trace amount of H COSiEt .
2
3 2
4
3
3
Nonetheless, treatment of HCOOSiEt with 3.0 equiv of Et SiH and 5 mol% [B] resulted in full
3
3
conversion to CH in 4 h, during which H C(OSiEt ) accumulates as the detectable intermediate. In
4
2
3 2
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
addition, the reduction of 1,3,5ꢀtrioxane with 1.0 equiv of Et SiH and 5 mol% [B] led to exclusive
3
formation of H COSiEt , indicating that the hydrosilylation of trioxane is even faster than that of
3
3
H COSiEt . In comparison, the hydrosilylation of the related substrates with [Al] at RT is not effective at
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
all and only provided a trace amount of the reduction products, but at 80 ºC after longer times some
hydrosilylation products were observed (Scheme 3)
O
C
OSiEt3
C
O
5
% [B]
5% [B]
Et SiH
Et SiOSiEt3
1/3
3
3
5
min, RT
1 h, RT
O
O
H
H
OSiEt3
H
OSiEt3
H
1
2
O
C
5
% [B]
3
Et SiH
CH4 (95%)
2 Et SiOSiEt3 (only 2 observed as intermediate)
3
3
4 h, RT
OSiEt3
1
O
C
OSiEt3
C
5
% [Al]
Et SiH
3
o
H3COSiEt3
1
(1 : 2 : 3 = 1 : 1.04 : 1.26)
1
6 h, 80 C
H
OSiEt3
H
OSiEt3
H
3
1
2
O
OSiEt3
C
5% [B]
5% [B]
3
Et SiH
3 CH OSiEt3
2 Et SiH
CH4
2 Et SiOSiEt
3 3
3
3
3
5 min, RT
O
O
4 h, RT
3
H
OSiEt3
H
2
OSiEt3
OSiEt3
5% [B]
5
% [Al]
Et SiH
CH4
Et3SiOSiEt3
CH4 (7%)
Et SiOSiEt3
C
H
3
C
Et SiH
3
3
5 min, RT
o
H
H
H
1
0 h, 80 C
H
H
3
3
Scheme 3. Catalytic Hydrosilylation Reaction of Each Fundamental Step.
To gain further insights of such tandem hydrosilylation of CO , we carried out a kinetic study of the
2
overall reduction process. Under our standard conditions (entry 6, Table 1), the reaction was complete in
5 h. Et SiH consumption and CH yield at the early stage of the reaction showed a linear relationship with
3
4
time (Figure 4). The concentration of detectable intermediate H C(OSiEt ) (2) reached a plateau at 1.5 h
2
3 2
and then decreased gradually. The maximum turnꢀover frequency (TOF) of the consumption of [SiH]
ꢀ1
using the first 5 data point from the initial 2 h was calculated to be 6.6 h , and the rate for the [SiH]
1
2
ACS Paragon Plus Environment

Page 13 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
ꢀ1
conversion was rSiH = 0.13 M h . Variation of the [Al] catalyst concentration (from 1.0 mol% to 10
mol%, Table 2) greatly impacted the [SiH] consumption rate. Overall, the reduction rate of CO , as
2
monitored by the consumption of [SiH] signal, was proportional to the [Al] loading. The only observed
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
species at δ 4.75 ppm after complete [SiH] conversion was the 2-[Al] adduct, which is consistent with the
stoichiometric reaction between 2 and [Al]. As expected, lowering the [Al] loading reduced the amount of
the final 2-[Al] residue and hence improved the yield of CH . In addition, decreasing the [B]
4
concentration also resulted in a decrease of CO reduction rate. Remarkably, when only 1.0 mol% of [Al]
2
was employed, the highest CH yield of 94% was achieved (Table 2).
4
Figure 4. Kinetic plots under current standard conditions (left, 80 °C, 5 mol% [Al] and 5 mol% [B]) and
varied loading of [Al] and [B] (right).
Table 2. Selected Results of CO Hydrosilylation at 80 °C under Different Tandem Catalyst Loadings
2
Catalyst
composition
Reaction
time (h)
Max TOF
rSiH
(M h )
Conv.
(%)
CH Yield
4
a
ꢀ1 b
ꢀ1 c
d
e
(h )
(%)
10% [Al] + 5.0% [B]
5.0% [Al] + 5.0% [B]
3
5
10
30
5.0
6.6
7.9
6.03
7.85
0.20
0.13
0.079
0.024
0.0785
100
100
100
99
77
82
91
94
80
2
1
.5% [Al] + 5.0% [B]
.0% [Al] + 5.0% [B]
.0% [Al] + 2.5% [B]
5
9
100
a
1
b
Based on [SiH] consumption from H NMR measurement. based on the steepest slope from the linear
c
fit of conversion against time at the early stage of hydrosilylation, see Figures S27ꢀS31. Rate of [SiH]
consumption, calculated based on Max TOF. Based on [SiH] consumption. See the SI for yield
calculation details.
d
e
1
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Collectively, the above results showed that CO fixation is promoted by [Al], while the subsequent
2
hydrosilylation reactions are catalyzed by [B]. To further clarify the preferred pathway of each step,
namely conventional carbonyl activation versus FLPꢀtype silane activation, and to explain the observed
activity for the CO reduction into CH by [Al] alone but requiring 80 °C for 24 h (entries 2aꢀb, Table 1),
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
albeit being much less effective than the mixed [Al]/[B] tandem catalyst system, we performed the
following additional experiments. We first mixed equimolar 1-[B] and Et SiHꢀ[Al], which resulted in the
3
instantaneous replacement of Et SiH by HCOOSiEt to form 1-[Al]. However, the subsequent
3
3
hydrosilylation was hindered, as only a trace amount of 2-[Al] was detected up to 1 h, in contrast to the
rapid reduction of 1-[B] by Et SiH in the absence of [Al]. This observation is in line with the Si–H
3
activation mechanism, in which the free carbonyl has to be formed to initiate the attack at the Si center of
a transient Si–H···[B] intermediate. If the carbonyl is coordinated to an additional LA, [Al], such
nucleophilic attack becomes less plausible. With this premise, we gratifyingly found that heating the
mixture to 80 °C led to the formation of 2-[Al], due to thermally induced dissociation of HCOOSiEt3
from [Al], enabling its reduction to H C(OSiEt ) that recombines with [Al] to form 2-[Al]. Addition of
2
3 2
an excess amount of Et SiH into such a mixture further converted H C(OSiEt ) to CH at 80 °C. In a
3
2
3 2
4
similar fashion, addition of [Al] to a mixture of H COSiEt , [B] and Et SiH reduced the hydrosilylation
3
3
3
rate of H COSiEt to CH . These experiments further confirm that the reduction steps 2-4 occur through
3
3
4
the FLP Si–H activation mechanism, and a higher temperature is necessary to facilitate the release of
substrates (1, 2 and 3) from [Al], which also renders [Al] catalytic by reentering the catalytic cycle for the
step 1 reduction.
In order to investigate the catalyst recyclability, the catalytic system consisting of 10 mol% [Al]
st
and 5.0 mol% [B] was tested for 3 cycles. In the 1 cycle, the hydrosilylation was complete in 3 h (Figure
5). After careful removal of excess CO and CH under vacuum, the system was recharged with the same
2
4
amount of Et SiH and CO as in the first cycle. The hydrosilylation of the second and third cycles was
3
2
complete in 6 and 14 h, respectively. This recycling experiments indicated that the [Al]/[B] catalysts
survived during catalytic cycle and are recyclable. Although we detected the decrease in catalyst
1
4
ACS Paragon Plus Environment

Page 15 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
efficiency from cycle to cycle (presumably due to partial hydrolysis of the catalyst during the reloading of
rd
CO ), quantitative conversion can still be achieved on the 3 load of Et SiH and CO .
2
3
2
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 5. Recyclability of CO reduction with 10 mol% [Al] and 5.0 mol% [B]. Reaction time for
2
complete SiH consumption in each cycle: 3 h, 6 h, and 14 h.
3
6
Computational Studies of Fundamental Steps 1–4. In this section we discuss each reduction
step separately starting from the hydrosilylation of CO to formate and ending with the hydrosilylation of
2
methyl silyl ether to methane. In Schemes 4, 5 and 6 numbers in green refer to [Al] while numbers in blue
refer to [B]. Further, to strengthen connections with experiments, we labeled DFT structures using letters,
such as A, B, C, etc., while for structures involving intermediates 1–3 we kept the same labeling scheme
as that used in the description of the experimental results.
Step 1: Hydrosilylation of CO to silylformate 1-[X] (X = Al, B). The two possible reaction
2
pathways we investigated to generate 1 from CO are reported in Scheme 4.
2
1
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 4. Hydrosilylation of CO to the Silylformate Complex 1-[X]. Free Energies (kcal/mol) in nꢀ
2
Hexane are Reported for X = Al (numbers in green) and B (numbers in blue) (n.l. = not located)
Along pathway 1 the individual LA directly activates the CO molecule towards the Hꢀtransfer from
2
Et SiꢀH. Pathway 2, instead, starts with a FLPꢀtype Si–H activation with coordination of Et SiH to the
3
3
LA, followed by the Hꢀtransfer to the CO molecule through an insertion mechanism or a SN transition
2
2
+
ꢀ
state and by rearrangement of the initially formed [OCOSiEt ] /[HꢀLA] ion pair. Focusing on pathway 1,
3
in the case of [B] we were not able to locate a CO ꢀB(C F ) adduct, B. Nevertheless, we found two
2
6 5 3
possible fourꢀmembered transition states, B/C and B/1-[X] in Scheme 4, where the silicon attacks one
oxygen of CO favoring the simultaneous Hꢀtransfer from Et SiꢀH to the CO carbon atom. Transition
2
3
2
state B/C collapses into intermediate C before reaching 1-[X], favored by 9.4 kcal/mol, while transition
state B/1-[X] collapses directly into product 1-[X]. The energy cost to reach transition states B/1-[X] and
B/C, around 35ꢀ39 kcal/mol, and the fact that these transition states should be formed by the encounter of
three molecules due to the instability of the preformed B intermediate explain the experimentally
observed inability of [B] to activate CO . Conversely, the CO ꢀalane adduct B was located 6.5 kcal/mol
2
2
1
6
ACS Paragon Plus Environment

Page 17 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
below separated Al(C F ) and CO . This interaction weakens the C=O bond of the CO moiety, as shown
6
5 3
2
2
by the slightly elongated C=O bond distance (1.19 Å in B vs 1.17 Å in the free CO ). [Al] coordination
2
results in an increased positive charge on the C atom of the CO moiety (1.07e in B vs 0.99e in the free
2
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
CO ), promoting the Hꢀtransfer from Et SiH to the electrophilic C center. This Hꢀtransfer occurs via the
2
3
four membered transition state B/C, with the [Al] coordinated O atom attacking the Si atom and an
energy barrier of 22.0 kcal/mol. Transition state B/C collapses into intermediate C, which can further
precipitate in the final product 1-[X] (ꢁG = –28.8 kcal/mol) through the almost barrierꢀless transfer of the
[
Al] to the carbonyl oxygen, via transition state C/1-[X]. The alternative transition state with the
uncoordinated O atom of B attacking the Si atom, B/1-[X], and leading to 1-[X] directly, was located 3.5
kcal/mol above B/C. The stability of the CO ꢀ[Al] adduct B and the low energy barrier of 22.0 kcal/mol
2
explain the experimentally observed capability of [Al] to activate CO2.
As for the competitive mechanism via Si–H activation, pathway 2 in Scheme 4, it involved the
formation of a silaneꢀLA adduct. Formation of such adduct was found to be favored by 7.2 kcal/mol with
[
Al], whereas in the case of [B] it is disfavored by 2.7 kcal/mol. The transition state for the following
insertion of CO into the Si–H bond, D1/1-[X] in Scheme 4, lies at 32.7 kcal/mol for [Al] and 33.1
2
kcal/mol for [B]. Finally, we also tried to promote Si–H activation via the SN type transition state D2/1-
2
[
X], corresponding to the concerted attack of the CO oxygen to the silicon atom and transfer of the H
2
atom to the [X] moiety. However, we were not able to locate this kind of transition state for both [Al] and
[B]. Overall, calculations confirmed the experimental inefficiency of [B] in the first step of CO2
hydrosilylation, since [B] is not able to effectively activate CO through either direct coordination to CO2
2
or FLPꢀtype silane activation. As far as [Al] is concerned, calculations suggest that the reduction of CO2
to silylformate occurs via CO activation, whereas the pathway involving the FLPꢀtype Si–H activation
2
was ruled out.
Steps 2 and 3a: Hydrosilylation of silylformate 1-[X] to H C(OSiEt ) (step 2) and formaldehyde
2
3 2
(step 3a). Starting from the LAꢀsilylformate adduct 1-[X], we studied the mechanisms most likely
1
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
operative in the second hydrosilylation step, namely from the silylformate adduct 1-[X] to fomaldehyde.
The pathways considered with the corresponding energetics are reported in Scheme 5. We considered the
Hꢀtransfer step from a second Et SiH molecule to the starting 1-[X] via the same pathways considered
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
above for hydrosilylation of CO . Specifically, we considered two transition states along the C=O
2
activation pathway, i.e. 1-[X]/E and 1-[X]/2-[X], where the Et SiH molecule attacks the oxygen of the
3
LA–O bond or the oxygen of the Si–O bond of 1-[X]. Transition state 1-[X]/E, leading directly to
Et SiOSiEt and the LAꢀaldehyde adduct, E in Scheme 5, is favored by roughly 6ꢀ7 kcal/mol relative to
3
3
transition state 1-[X]/2-[X], leading to adduct 2-[X], with the experimentally characterized CH (OSiEt3)2
2
intermediate 2 bound to the LA. In this conversion, the two LAs behave similarly, with an overall energy
barrier through the favored transition state 1-[X]/E of roughly 29 kcal/mol. This relatively high energy
barrier can be ascribed both to the high steric pressure in this second hydrosilylation transition state (two
molecules of silane plus one LA) and to the reduced electrophilicity of the formate carbon center with
respect to that of CO , as revealed by the charges of these two C centers, i.e. 1.07e in CO vs 0.74e in 1-
2
2
[
X]. Finally, formation of E from 1-[X] + Et SiH is exergonic by almost 4.0 kcal/mol for both LAs, which
3
drives aldehyde formation. As for conversion of the experimentally characterized intermediate 2-[X] to
the formaldehyde adduct E, it can proceed either in a single step through transition state 2-[X]/E, or
through a twoꢀstep pathway via dissociation of [X] from 2-[X], followed by the direct reductive
elimination of Et SiOSiEt from 2 to release the aldehyde via transition state 2/E, following a similar
3
3
35
mechanism reported in literature. According to our calculations, the oneꢀstep mechanism via transition
state 2-[X]/E, with an overall energy barrier around 20 kcal/mol, is clearly favored over the direct
reductive elimination from 2. Nevertheless, dissociation of the LA from 2-[X] releasing intermediate 2 is
energetically favored with [B], actually representing the thermodynamic product of the reaction.
As for the Si–H activation pathway, it starts with the conversion of 1-[X] into 2-[X], with the steps
from 2-[X] to E already being discussed above. Thus, we focus here on the pathway from 1-[X] to 2-[X].
We calculated first the energetics involved in the release of the LA from the silylformate adduct 1-[X]. As
1
8
ACS Paragon Plus Environment

Page 19 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
expected, dissociation of [Al], with a ꢁG of 25.2 kcal/mol, is more expensive than dissociation of [B],
with a ꢁG of only 5.8 kcal/mol. Starting from the LA free species 1 and a preꢀformed Et SiHꢀ[X] adduct,
3
we located both the transition states for the direct formate insertion into the Si–H bond, F1/2-[X], and the
direct Hꢀtransfer between Et SiH and the LA via SN ꢀtype reactivity, F2/2-[X]. Based on the energy of
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
3
2
the H transfer transition states, both the SiꢀH activation pathways we examined can be ruled out in the
case of [Al], since transition states F1/2-[X] and F2/2-[X] are more than 35 kcal/mol above intermediate
1-[X], respectively. Conversely, the insertion Si–H activation pathway is isoenergetic with the C=O
activation pathway in the case of [B] and, more relevantly, the SN type Hꢀtransfer mechanism via
2
transition state F2/2-[X] shows a barrier of only 13.2 kcal/mol.
H
[X]
Et SiO C O
3
Et Si
3
H
1
-[X]/E
0.7
9.9
-
1
Et Si
3
O
Et Si
3
X= Al
X = B
Et
3
SiO
Et SiO SiEt
3 3
+[X]
+
2
H C OSiEt3
H C
H
O
[X]
-
-
17.0
17.0
O
H
[X]
2/E
20.6
0.6
Pathway 1:
C=O activation
C
H
H
-
2
E
-32.5
-13.3
Et SiO
3
[X]
[
X]
Et SiO
Et SiO SiEt3
3
3
[
X]
H
C
O
H C
H
O
H C
H
O
O
C
H
SiEt3
SiEt3
[X]
H
OSiEt3
1-[X]/2-[X]
6.1
2
-[X]
2-[X]/E
-14.4
1
-
-[X]
28.8
9.4
-34.1
-8.6
26.1
10.1
-
Et
Si H [X]
Et
H
O
OSiEt3
C
-[X]
Pathway 2:
O
Et Si
3
H
[X]
HC
Et
F2/2-[X]
Si-H activation
F1/2-[X]
Et SiO
3
9.4
7.4
3.8
Insertion
SN2
19.2
O
C
O
C
+
[X] +SiEt H
3
+
Et Si H
3
[X]
H
OSiEt3
H
OSiEt3
F
1
-3.6
-3.6
-
10.7
0.9
-
Scheme 5. Hydrosilylation of silylformate 1-[X] to H C(OSiEt ) /formaldehyde. Free Energies
2
3 2
(
kcal/mol) in nꢀHexane are Reported for X = Al (numbers in green) and B (numbers in blue).
1
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Focusing on the thermodynamic scenario, the stability of the intermediates shown in Scheme 5
seems to correlate with the different catalytic behavior in the presence of [Al] or [B]. In fact, for [Al] the
thermodynamic product is the H C(OSiEt ) ꢀ[Al] adduct 2-[X], which is favored by 1.6, 5.3 and 17.1
2
3 2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
kcal/mol relative to E, 1-[X] and 2, respectively. As consequence, liberation of [Al] to promote further
reactivity is compromised. On the contrary, intermediate 2 is the most stable species in the presence of
[
B], followed by E, with 1-[X] and 2-[X] being clearly much less stable. The different thermodynamic
stability of 2 and 2-[X] with [Al] and [B], together with the much lower energy barrier via the Si–H
activation pathway, renders the [B] catalyst active in the reduction of silylformate to formaldehyde, in
agreement with the experimental results.
Steps 3b and 4: Hydrosilylation of formaldehyde to silyl methanol (step 3b) and then to methane
(step 4). The reaction pathways we considered for the last two steps to form methane from the aldehydeꢀ
LA adduct E are reported in Scheme 6 with the related energies. Since no favored hydrosilylation
pathway was located starting from 2 of Scheme 5, we focused on the aldehydeꢀLA adduct E that turns out
to be the crucial species for the course of the reaction. Intermediate E can react with the third equivalent
of Et SiH along the C=O activation pathway via the low energy Hꢀtransfer transition state E/3-[X] with
3
both the LAs considered, with an energy barrier around 5 kcal/mol. Alternatively, the key intermediate 3-
[
X] can also be reached from E along the Si–H activation pathway after dissociation of the LA to reach
intermediate G. As calculated in the previous reaction sequences, dissociation of [Al] from E is clearly
endergonic, whereas it is favored with [B]. The liberated formaldehyde can react with a Et SiHꢀLA
3
adduct, intermediate H, via the already considered insertion or SN ꢀtype pathways, through transition
2
states H1/3-[X] and H2/3-[X], respectively. The insertion mechanism can be excluded since the H1/3-[X]
transition state is more than 20 kcal/mol higher in energy than transition state E/3-[X] for both [Al] and
[
B]. As for the SN ꢀtype pathway, it is competitive for [B], since transition state H2/3-[X] is almost
2
isoenergetic with transition state E/3-[X] along the C=O activation pathway. Differently, it can be
2
0
ACS Paragon Plus Environment

Page 21 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
excluded for [Al] since the transition state H2/3-[Al] is almost 25 kcal/mol higher in energy than
transition state E/3-[Al].
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
Scheme 6. Hydrosilylation of Formaldehyde to Silyl Methanol and then to Methane. Free Energy
(
kcal/mol) in nꢀHexane are Reported for X = Al (numbers in green) and B (numbers in blue).
2
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Two different pathways were considered to generate methane from 3-[X]. The first pathway is the
oneꢀstep Hꢀtransfer from the fourth equivalent of Et SiH molecule to the carbon of the activated adduct 3-
3
[
X] (C–O activation pathway) via transition state 3-[X]/K. This pathway can be excluded, since it
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
requires the overcome of an energy barrier greater than 60 kcal/mol (Scheme 6). The second pathway is a
multiꢀstep process and starts with release of the LA from 3-[X], which is again strongly endergonic for
[
Al], whereas it is slightly exergonic for [B] (3-[X] vs 3 for [Al] and [B] in Scheme 6). After releasing LA
from 3-[X], silyl methanol 3 proceeds to the last hydrosilylation step, via SN ꢀtype transition state I/J
2
involving 3 and a Et SiHꢀLA adduct. This Hꢀtransfer step is rate determining, with a barrier of 16.0
3
kcal/mol for [B] (calculated from the most stable 3 + Et SiH + LA species) and of almost 40 kcal/mol for
3
[
Al] (calculated from the most stable 3-[X] + Et SiH species). As a consequence and in agreement with
3
the experimental results, the [Al] catalyst is inactive also in this last hydrosilylation step, which proceeds
+
-
smoothly with the [B] catalyst. The last Hꢀtransfer within the formal [CH O(Et Si) ] /[HꢀLA] adduct J to
3
3
2
release methane via transition state J/K is almost barrierꢀless. As expected, formation of methane is
thermodynamically favored by roughly 50 kcal/mol with respect to the silyl methanol intermediate 3. For
the sake of simplicity, the insertion pathway is not reported for the conversion of 3 to methane, since it
involves clearly unfeasible barriers, as discussed in the previous sections.
Alane vs Borane: Electronic and Steric Analysis. The results reported in the previous sections
showed that the experimentally observed catalytic difference between [Al] and [B] seems to be due to the
much higher stability of the LAꢀadducts with [Al] relative to [B], in intermediates 1-[X], 2-[X], E and 3-
[X] (see Scheme 5 and 6), which reflects the difficult release of the LA from the substrates and the high
barrier for the Si–H bond activation. Focusing on the named intermediates, the relative stability of 1-[X],
E and 3-[X] is more than 10 kcal/mol higher for [Al] than [B], respectively. This result reflects in the key
transition states for Si–H activation (i.e. F2/2-[X] and H2/3-[X]) lying much high in energy for [Al].
Moreover, 2-[X] is favored by –17.1 kcal/mol with [Al] and disfavored by 8.4 kcal/mol with [B] relative
2
2
ACS Paragon Plus Environment

Page 23 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
to 2, and 3-[X] is favored by –17.6 kcal/mol with [Al] and disfavored by 3.5 kcal/mol with [B] with
respect to 3 (the total difference between [Al] and [B] is of 25.5 and 21.1 kcal/mol, respectively).
To better understand the large difference in the relative stability of these intermediates, we analyzed
both electronic and steric effects. From an electronic point of view, the calculated electrophilicity of the
two LAs (0.24 for [Al] and 0.38 for [B]) appears to indicate that the formation of [B]ꢀadducts should be
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
favored relative to the formation of [Al]ꢀadducts. Indeed, the relative energies of the SN Hꢀtransfer
2
transition states relative to the preceding intermediates are in line with the electrophilicity results. Moving
to sterics, we tested the relative stability of 2 and 3 with respect to 2-[X] and 3-[X] with XH (X = Al and
3
B) as LA. The difference in the binding energy of 2 and 3 to AlH and BH is only 2.7 and 4.5 kcal/mol,
3
3
relative to a difference greater than 20 kcal/mol with [Al] and [B]. Incidentally, the same analysis
performed on the LAꢀHSiEt adducts shows that the 10 kcal/mol of difference in the relative stability of
3
the [Al]···H–SiEt adduct respect to the [B]···H–SiEt one decreased to –2 kcal/mol in favor of the borane
3
3
when AlH and BH are used as the LA. This suggests that steric effects strongly impact the binding
3
3
ability of [B].
As a further test, we decomposed the gasꢀphase binding energy of [Al] and [B] to formaldehyde
into a preparation and an interaction energy term. The first contribution is the energy paid to deform the
LA and the aldehyde from their ideal conformations in the unbound state to the geometries they assume in
the LAꢀadducts E. The interaction energy term, instead, corresponds to the energy gain due to the rigid
interaction between the LA and the aldehyde frozen in the same conformation as in E. We found that the
deformation of the free [B] to the geometry it has in E is 9.5 kcal/mol more expensive than the
deformation of [Al], while the interaction energy between the deformed LA and formaldehyde is only 6.5
kcal/mol stronger with [Al]. The two terms cumulate in an [Al]ꢀaldehyde bond being 16.0 kcal/mol
stronger than the [B]ꢀaldehyde bond in the gasꢀphase.
2
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Overall, these results indicated that the different behavior of [Al] and [B] can be ascribed to the
significantly smaller size of boron that causes unfavorable steric repulsions between the C F rings, as
6
5
well as between the C F rings and the substituents on the silane atoms, in the adducts.
6
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In Scheme 7 we summarized the whole catalytic cycle in the presence of both [Al] and [B] LAs,
while the calculated energy profiles of intermediates and transition states for [Al] and [B] pathways along
the reaction coordinate were plotted in Figure 6. As the first step, hydrosilylation of CO to silylformate
2
takes place via the [Al] catalyzed C=O activation pathway. Next, the reduction of HCOOSiEt to
3
formaldehyde proceeds through the [B] catalyzed SN type Si–H activation mechanism. In the following
2
reduction step to achieve intermediate Et SiOCH , the C=O and the Si–H activation pathways turn out as
3
3
competitive with [B] since the corresponding determining barriers are almost the same in energy. Finally,
in the last reduction step to methane, the SN Si–H activation mechanism is again favored. The sterics of
2
the substrate seems to play a key role, with the C=O activation pathway suffering the steric hindrance
37
more than the SN Si–H one due to a more crowded geometry of the corresponding transition state. The
2
thermodynamics of the adducts favors [Al]; however, in agreement with the experimental results, only
[
Al] is active in the first step, while [B] is active in the subsequent steps. As a consequence an exchange
between the two Lewis acids is likely to take place during the four reduction steps.
Step 4
~16 kcal/mol
Step 3b
5 kcal/mol
[Al]
H
Si-H-[B]
[
B]
SiOSi + [B] + CH4
~
SiO
C
Si-H-[B]
[Al]
[Al]
H
[Al]
H
CO , Si-H
2
[Al]
O
C
Si-H
Step 1
~22 kcal/mol
H
H
[B]
[
Al]
[Al]
[
Al] [B]
SiO
C
O
C
[Al]
SiOSi
H
OSi
OSi
H
H
Step 3a
~20 kcal/mol
Si-H-[B]
Step 2
[B]
~13 kcal/mol
2
4
ACS Paragon Plus Environment

Page 25 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Scheme 7. Four Fundamental Steps and Calculated Energy Barriers in the Proposed Complete Catalytic
Cycle for the Hydrosilylation of CO into CH in the Presence of Mixed [Al] and [B] LAs.
2
4
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. Calculated energy profiles of intermediates and transition states for [B] and [Al] pathways
along the reaction coordinate.
Conclusions
In summary, we have developed the first example of highly selective reduction of CO into CH via
2
4
tandem catalytic hydrosilylation by a mixed main group B(C F ) /Al(C F ) catalyst system. The results,
6
5 3
6 5 3
obtained from our comprehensive study involving the detection, characterization and independent
synthesis of each reaction intermediate, reactions under catalytic conditions, computational calculations,
as well as kinetic and mechanistic investigations, have demonstrated that [Al] is responsible for the first
step of the reduction that converts CO into HCOOSiEt , while the subsequent reduction steps of
2
3
2
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
HCOOSiEt to H C(OSiEt ) to H COSiEt and finally to CH are catalyzed by [B]. The hydrosilylation
3
2
3 2
3
3
4
of H C(OSiEt ) , the rateꢀlimiting step in the [B]ꢀcatalyzed FLP reduction sequences, is proposed to
2
3 2
proceed through a twoꢀstep pathway involving the formation and reduction of formaldehyde. Our
computational results further address the fixation of CO into HCOOSiEt by [Al] via the classical LAꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
mediated C=O activation, the subsequent transformations into CH by [B] through the FLPꢀtype Si–H
4
activation, as well as the H C(OSiEt ) reduction via the formaldehyde cycle, all of which are consistent
2
3 2
with the experimental results.
We attribute this remarkably different catalytic behaviors between Al(C F ) and B(C F ) to the
6
5 3
6 5 3
higher overall Lewis acidity of [Al] derived from two conflicting factors, electronic and steric effects.
While the study of the electronic term indicates that [B] has a higher electronic affinity, the steric term
suggests that [B] pays much higher reorganization energy penalty due to both a smaller radius of boron
and repulsion between the orthoꢀfluorine atoms. This stronger overall Lewis acidity of [Al], when
compared to [B], renders its higher tendency to form stable [Al]–substrate (intermediate) adducts with
CO as well as intermediates 1, 2 and 3, hence accounting for its distinct yet complementary catalytic
2
behaviors in the CO ꢀtoꢀCH hydrosilylative reduction cycle. Overall, the roles of [Al] and [B] are not
2
4
only complementary but also synergistic in the total reduction of CO , which render both [Al]ꢀmediated
2
first reduction step (which, when carried out alone, is a stoichiometric reaction) and [B]ꢀmediated
subsequent steps catalytic. With an optimized loading and [Al]/[B] ratio of 1.0% : 5.0%, a high CH4
production yield of 94% has been achieved. Such a catalytic system is also shown to be recyclable, based
on three cycling experiments. The tunability of [Al]/[B] catalysts should allow one to develop more
efficient, economical and recyclable tandem LA catalysts based on interꢀ or intramolecular and/or
heterogeneous catalysts.
Supplementary Information. Full experimental details and additional figures (80 pages). This
material is available free of charge via the Internet at http://pubs.acs.org.
2
6
ACS Paragon Plus Environment

Page 27 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Acknowledgments. This work was supported by the US National Science Foundation (NSFꢀ CHEꢀ
1
507702) for the study carried out at Colorado State University and by the funding from King Abdullah
University of Science and Technology (KAUST) for the study performed at KAUST. We thank Boulder
Scientific Co. for the research gift of B(C F ) .
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
6 5 3
References
1
. (a) Carbon Dioxide as Chemical Feedstock. Aresta, M. Ed.; WileyꢀVCH, Weinheim, 2010. (b) Olah,
G. A. Angew. Chem., Int. Ed. 2005, 44, 2636ꢀ2639.
2
. For recent reviews of carbon dioxide conversion and reduction, see: (a) Maeda, C.; Miyazaki, Y.;
Ema, T. Catal. Sci. Technol. 2014, 4, 1482ꢀ1497. (b) Oh, Y.; Hu, X. Chem. Soc. Rev. 2013, 42, 2253ꢀ
2261. (c) Costentin, C.; Robert, M.; Savéant, J. M. Chem. Soc. Rev. 2013, 42, 2423ꢀ2436. (d) Centi,
G.; Quadrelli, E. A.; Perathoner, S. Energy Environ. Sci. 2013, 6, 1711ꢀ1731. (e) Peters, M.; Köhler,
B.; Kuckshinrichs, W.; Leitner, W.; Markewitz, P.; Müller, T. E. ChemSusChem 2011, 4, 1216ꢀ1240.
(
f) Cokoja, M.; Bruckmeier, C.; Rieger, B.; Herrmann, W. A.; Kühn, F. E. Angew. Chem. Int. Ed.
011, 50, 8510ꢀ37. (g) Riduan, S. N.; Zhang, Y. Dalton Trans. 2010, 39, 3347ꢀ3357. (h) Benson, E.
E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem. Soc. Rev. 2009, 38, 89ꢀ99.
2
3. For a recent review of catalytic carbon dioxide hydrosilylation, see: FernándezꢀAlvarez, F. J.; Aitani,
A. M.; Oro, L. A. Catal. Sci. Technol. 2014, 4, 611ꢀ624.
4
. For recent reviews of catalytic carbon dioxide hydrogenation, see: (a) Li, Y.ꢀN.; Ma, R.; He, L.ꢀN.;
Diao, Z.ꢀF. Catal. Sci. Technol. 2014, 4, 1498ꢀ1512. (b) FernándezꢀAlvarez, F. J.; Iglesias, M.; Oro,
L. A.; Polo, V. ChemCatchem 2013, 5, 3481ꢀ3494. (c) Wang, W.; Wang, S.; Ma, X.; Gong, J. Chem.
Soc. Rev. 2011, 40, 3703ꢀ3727. (d) Federsel, C.; Jackstell, R.; Beller, M. Angew. Chem. Int. Ed. 2010,
49, 6254ꢀ6257. (e) Jessop, P. G.; Joó, F.; Tai, C.ꢀC. Coord. Chem. Rev. 2004, 248, 2425ꢀ2442. (f)
Leitner, W. Angew. Chem., Int. Ed. Engl. 1995, 34, 2207ꢀ2221.
5. (a) Metsanen, T. T.; Oestreich, M. Organometallics 2015, 34, 543ꢀ546. (b) Deglmann, P.; Ember, E.;
Hofmann, P.; Pitter, S.; Walter, O. Chem. - Eur. J. 2007, 13, 2864ꢀ2879. (c) Jansen, A.; Görls, H.;
2
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 31
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Pitter, S. Organometallics 2000, 19, 135ꢀ138. (d) SüssꢀFink, G.; Reiner, J. J. Organomet. Chem.
1
981, 221, C36ꢀC38. (e) Koinuma, H.; Kawakami, F.; Kato, H.; Hirai, H. J. Chem. Soc. Chem.
Comm. 1981, 213ꢀ214.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6. Scheuermann, M. L.; Semproni, S. P.; Pappas, I.; Chirik, P. J. Inorg. Chem. 2014, 53, 9463ꢀ9465.
7. Itagaki, S.; Yamaguchi, K.; Mizuno, N. J. Mol. Catal., A: Chem. 2013, 366, 347ꢀ352.
8
. (a) Lalrempuia, R.; Iglesias, M.; Polo, V.; Miguel, P. J. S.; FernándezꢀAlvarez, F. J.; PerézꢀTorrente,
J. J.; Oro, L. A. Angew. Chem., Int. Ed. 2012, 51, 12824ꢀ12827. (b) Park, S.; Bézier, D.; Brookhart,
M. J. Am. Chem. Soc. 2012, 134, 11404ꢀ11407. (c) Eisenschmid, T. C.; Eisenberg, R.
Organometallics 1989, 8, 1822ꢀ1824.
9. GonzálezꢀSebastián, L.; FloresꢀAlamo, M.; Garcıa, J. J. Organometallics 2013, 32, 7186ꢀ7194.
́
10. (a) Zhang, L.; Cheng, J. H.; Hou, Z. M. Chem. Commun. 2013, 49, 4782ꢀ4784. (b) Motokura, K.;
Kashiwame, D.; Takahashi, N.; Miyaji, A.; Baba, T. Chem. - Eur. J. 2013, 19, 10030ꢀ10037. (c)
Motokura, K.; Kashiwame, D.; Miyaji, A.; Baba, T. Org. Lett. 2012, 14, 2642ꢀ2645.
1
1. Sattler, W.; Parkin, G. J. Am. Chem. Soc. 2012, 134, 17462ꢀ17465.
2. (a) Piers, W. E. Adv. Organomet. Chem. 2005, 52, 1ꢀ76. (b) Chen, E. Y.ꢀX.; Marks, T. J. Chem. Rev.
2000, 100, 1391ꢀ1434.
1
13. Matsuo, T.; Kawaguchi, H. J. Am. Chem. Soc. 2006, 128, 12362ꢀ12363.
1
1
1
4. Mitton, S. J.; Turculet, L. Chem. - Eur. J. 2012, 18, 15258ꢀ15262.
5. Jiang, Y. F.; Blacque, O.; Fox, T.; Berke, H. J. Am. Chem. Soc. 2013, 135, 7751ꢀ7760.
6. (a) LeBlanc, F. A.; Piers, W. E.; Parvez, M. Angew. Chem., Int. Ed. 2014, 53, 789ꢀ792. (b) Berkefeld,
A.; Piers, W. E.; Parvez, M.; Castro, L.; Maron, L.; Eisenstein, O. Chem. Sci. 2013, 4, 2152ꢀ2162.
17. For selected reviews of the FLP chemistry, see: (a) Stephan, D. W.; Erker, G. Angew. Chem., Int. Ed.
2015, 54, 6400ꢀ6441. (b) Stephan, D. W.; Erker, G. Eds.; Frustrated Lewis Pairs I & II, Topics in
Current Chemistry. SpringerꢀVerlag: Berlin, Germany, 2013; Vol. 332 & 334. (c) Stephan, D. W.;
Erker, G. Angew. Chem., Int. Ed. 2010, 49, 46ꢀ76.
1
8. Ashley, A. E.; Thompson, A. L.; O'Hare, D. Angew. Chem. Int. Ed. 2009, 48, 9839ꢀ9843.
2
8
ACS Paragon Plus Environment

Page 29 of 31
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
9. Menard, G.; Stephan, D. W. J. Am. Chem. Soc. 2010, 132, 1796ꢀ1797.
2
0. Courtemanche, M.ꢀA.; Légaré, M.ꢀA.; Maron, L.; Fontaine, F.ꢀG. J. Am. Chem. Soc. 2013, 135, 9326ꢀ
9
329.
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
21. For recent reviews of carbon dioxide reduction ultilizing the FLP chemistry, see: (a) Bontemps, S.
Coord. Chem. Rev. 2016, 308, 117ꢀ130. (b) Fontaine, F.ꢀG.; Courtemanche, M.ꢀA.; Légaré, M.ꢀA.
Chem. - Eur. J. 2014, 20, 2990ꢀ2996.
2
2. (a) Declercq, R.; Bouhadir, G.; Bourissou, D.; Légaré, M.ꢀA.; Courtemanche, M.ꢀA.; Nahi, K. S.;
Bouchard, N.; Fontaine, F.ꢀG.; Maron, L. ACS Catal. 2015, 5, 2513ꢀ2520. (b) Courtemanche, M.ꢀA.;
Pulis, A. P.; Rochette, E.; Légaré, M.ꢀA.; Stephan, D. W.; Fontaine, F.ꢀG. Chem. Commun. 2015, 51,
9797ꢀ9800. (c) Wang, T.; Stephan, D. W. Chem. - Eur. J. 2014, 20, 3036ꢀ3039. (d) Gomes, C. D.;
Blondiaux, E.; Thuéry, P.; Cantat, T. Chem. - Eur. J. 2014, 20, 7098ꢀ7106. (e) Courtemanche, M.ꢀA.;
Légaré, M.ꢀA.; Maron, L.; Fontaine, F.ꢀG. J. Am. Chem. Soc. 2014, 136, 10708ꢀ10717. (f) Menard,
G.; Stephan, D. W. Dalton Trans. 2013, 42, 5447ꢀ5453. (g) Lim, C. H.; Holder, A. M.; Hynes, J. T.;
Musgrave, C. B. Inorg. Chem. 2013, 52, 10062ꢀ10066. (h) Courtemanche, M.ꢀA.; Larouche, J.;
Légaré, M.ꢀA.; Bi, W.; Maron, L.; Fontaine, F.ꢀG. Organometallics 2013, 32, 6804ꢀ6811. (i) Roy, L.;
Zimmerman, P. M.; Paul, A. Chem. - Eur. J. 2011, 17, 435ꢀ439. (j) Kwon, H. J.; Kim, H. W.; Rhee,
Y. M. Chem. - Eur. J. 2011, 17, 6501ꢀ6507. (k) Zimmerman, P. M.; Zhang, Z. Y.; Musgrave, C. B.
Inorg. Chem. 2010, 49, 8724ꢀ8728.
2
2
2
3. Riduan, S. N.; Zhang, Y. G.; Ying, J. Y. Angew. Chem., Int. Ed. 2009, 48, 3322ꢀ3325.
4. Berkefeld, A.; Piers, W. E.; Parvez, M. J. Am. Chem. Soc. 2010, 132, 10660ꢀ10661.
5. Schäfer, A.; Saak, W.; Haase, D.; Müller, T. Angew. Chem., Int. Ed. 2012, 51, 2981ꢀ2984.
26. (a) Wehmschulte, R. J.; Saleh, M.; Powell, D. R. Organometallics 2013, 32, 6812ꢀ6819. (b)
Khandelwal, M.; Wehmschulte, R. J. Angew. Chem., Int. Ed. 2012, 51, 7323ꢀ7326.
27. Chen, E. Y.ꢀX., Tris(pentafluorophenyl)alane, In Encyclopedia of Reagents for Organic Synthesis,
John Wiley & Sons, Ltd, 2012, DOI:10.1002/047084289X.rn01382.
2
9
ACS Paragon Plus Environment