Silylative Pinacol Coupling Catalyzed by Nitrogen-Doped Carbon-Encapsulated Nickel/Cobalt Nanoparticles: Evidence for a Silyl Radical Pathway

Article abstract of DOI:10.1021/acscatal.7b02788

The silylative pinacol coupling of arylaldehydes catalyzed by an easily accessible, heterogeneous base-metal catalyst is demonstrated. Instead of using the classical combination of catalyst, stoichiometric metal reductants, and chlorosilanes, the develope

Full text of DOI:10.1021/acscatal.7b02788

Subscriber access provided by READING UNIV
Article
Silylative Pinacol Coupling Catalyzed by Nitrogen-
Doped Carbon-Encapsulated Nickel/Cobalt
Nanoparticles: Evidence for a Silyl Radical Pathway
Soren Kramer, Fatima Hejjo, Kristoffer H Rasmussen, and Søren Kegnæs
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b02788 • Publication Date (Web): 08 Dec 2017
Downloaded from http://pubs.acs.org on December 8, 2017
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.
ACS Catalysis 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 32
ACS Catalysis
1
2
3
4
5
6
7
8
9
Silylative Pinacol Coupling Catalyzed by Nitrogenꢀ
Doped CarbonꢀEncapsulated Nickel/Cobalt
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
Nanoparticles: Evidence for a Silyl Radical Pathway
Søren Kramer, Fatima Hejjo, Kristoffer H. Rasmussen, Søren Kegnæs*
Department of Chemistry, Technical University of Denmark, 2800 Kgs. Lyngby (Denmark).
ABSTRACT: The silylative pinacol coupling of arylaldehydes catalyzed by an easily accessible,
heterogeneous baseꢀmetal catalyst is demonstrated. Instead of using the classical combination of
catalyst, stoichiometric metal reductants, and chlorosilanes, the developed reaction only requires
the use of a catalyst and a hydrosilane, which serves as both reductant and silylating agent. A
rare mechanistic investigation in this field focusing on the organic reactants was undertaken
using various techniques including experimental rate orders, kinetic isotope effect, radical
scavengers, and stoichiometric tests. The obtained results provided evidence for a reaction
mechanism which is different from the classical pinacol coupling pathway. We propose that the
heterogeneous catalyst facilitates easy access to silyl radicals thereby circumventing the usual
need for explosive initiators to access these species. In addition, leaching tests and recycling of
the catalyst were performed, clearly supporting the heterogeneous nature of the catalyst.
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 32
1
2
3
4
5
6
Keywords: Heterogeneous catalysis, nanoparticles, nitrogenꢀdoped carbon, mechanism, silyl
radicals
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
INTRODUCTION
The use of catalysis is widespread in the chemical industry due to its facilitation of energyꢀ
efficient production.[1] In terms of sustainability, heterogeneous catalysis often displays
significant advantages over homogeneous catalysis such as robustness, recyclability, and ease of
catalyst separation.[2] However, compared to homogeneous molecular catalysis, the mode of
operation of heterogeneous catalysts is much less wellꢀunderstood. This feature often
complicates rational reaction improvement and development. In this regard, insight into reaction
pathways can be highly valuable.
Recently, significant research effort has demonstrated the use of base metal nanoparticles
supported on nitrogenꢀdoped carbon as powerful heterogeneous catalysts for a variety of organic
transformations.[3,4] In addition to the methodology development, most reports also study the
physical properties of these novel materials.[4] In contrast, very limited knowledge is available
regarding the underlying pathways that these materials are invoking on the organic reactants.
These pathways are rarely studied; more commonly, they are simply presumed to parallel their
counterparts from homogeneous catalysis.[5]
Formation of carbon–carbon bonds is central to organic synthesis for the production of
pharmaceuticals, materials, and fuels.[6] In addition, 1,2ꢀdiols are prevalent intermediates in
synthesis of fine chemicals, such as pharmaceuticals and agrochemicals, as well as starting
points for bulk production of for example polyesters.[7] The pinacol coupling forms a new
ACS Paragon Plus Environment
2

Page 3 of 32
ACS Catalysis
1
2
3
4
5
6
7
8
9
carbon–carbon bond and a 1,2ꢀdiol through an overall reduction of two carbonyl moieties.[8,9]
Due to its reductive nature, the reaction is usually performed with a stoichiometric metal
reductant and a silicon electrophile, in addition to the catalyst (Scheme 1a). In terms of wasteꢀ
minimization, a more direct approach can be envisioned by using a simple hydrosilane as both
reductant and silylating agent. Such a reaction protocol could potentially be achieved by a silyl
radical mechanism. Unfortunately, the controlled formation of silyl radicals often requires
elaborate conditions (Scheme 1b).[10] Radical initiators such as AIBN or peroxides are required.
These compounds are explosive, thus placing demands on handling and storage. In addition to
the singleꢀuse radical initiator, stoichiometric tin hydrides and specialized reagents such as
PhSeSiMe3 or expensive silanes such as HSi(TMS)3 are often needed.[10,11] The development
of a stable and recyclable catalyst circumventing the need for explosive, singleꢀuse initiators
would be highly beneficial, especially, if it is operational with simple silanes such as HSiEt3.
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 1. (a) Classical approach to pinacol coupling. (b) Typical reaction conditions for
accessing silyl radicals
Herein, we report the reaction development of a silylative pinacol coupling of arylaldehydes
using HSiEt3 without the need for an additional stoichiometric reductant or explosive radical
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 32
1
2
3
4
5
6
7
8
9
initiators. The recyclable, heterogeneous catalyst consists of base metal nanoparticles on
nitrogenꢀdoped carbon. The ensuing mechanistic study of the organic reactants provided
experimental evidence which is consistent with initial silyl radical formation. In contrast, an
alternative pathway proceeding through an initial pinacol coupling followed by silylation is
inconsistent with the obtained data.
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
RESULTS AND DISCUSSION
The study was initiated by the development of a catalytic protocol with easily accessible
nickel/cobalt alloy nanoparticles encapsulated in nitrogenꢀdoped carbon (Ni1Co1@NC).[12,13]
Under the optimized conditions, using 8 mol% Ni1Co1@NC (4 mol% of each metal), the desired
product, 1a, was observed in 79% yield (Table 1, entry 1). The product from direct
hydrosilylation, 2, accounted for the remaining mass balance. Control experiments without either
Ni1Co1@NC or HSiEt3 led to minimal conversions, confirming the crucial role of both reaction
components (entries 2–3). Without benzoic acid added to the >99.5% purity benzaldehyde, the
reaction was less selective and more of the direct hydrosilylation product 2 was obtained (entry
4). In contrast, using less pure (92% purity by 1H NMR), undistilled benzaldehyde without added
benzoic acid resulted in comparable yields to the optimized conditions due to the inherent
content of benzoic acid from autoxidation. Accordingly, careful purification of the aldehyde is
not required. When the reaction was performed under air instead of inert atmosphere, a small
decrease in yield was observed (entry 5). The difference between conversion and yield was
determined to be formation of benzoic acid from aldehyde oxidation. However, the catalyst itself
is only moderately sensitive to air.[14] Performing the reaction neat or in heptane instead of
toluene had only a minor influence on the reaction outcome (entries 6–7), but a decrease in
reaction temperature or silane equivalents led to a less efficient reaction (entries 8–9). Increasing
ACS Paragon Plus Environment
4

Page 5 of 32
ACS Catalysis
1
2
3
4
the amount of silane to 2.5 equivalents did not increase the yield of 1a compared to the standard
5
6
conditions (entry 10).
7
8
9
Table 1. Influence of Reaction Parameters on the Reaction Efficiencya
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
8 mol% Ni1Co1@NC
8 mol% PhCO2H
1.5 equiv HSiEt3
O
OSiEt3
H
Et3SiO
H
PhMe (1 M), 90 °C, 24 h
OSiEt3
0.5 mmol
1a
2
Entry
Change from standard cond. conv. [%] Yield 1a [%] Yield 2 [%]
none
>95
5
5
>95
90
>95
>95
63
60
>95
1
2
3
4
5
6
7
8
9
79
0
0
54
66
77
75
33
47
72
20
<4
0
38
13
23
25
7
no Ni1Co1@NC
no silane
no PhCO2H
reaction in air
neat
heptane instead of PhMe
70 °C instead of 90 °C
1.2 equiv HSiEt3
2.5 equiv HSiEt3
13
22
10
a Conversion and yield are based on 1H NMR analysis with the aid of
an internal standard. d.r. of 1a was 53:47 in all cases. 8 mol% Ni1Co1@NC
corresponds to 3.3 mg.
Recycling of the heterogenous catalyst under the standard conditions (Table 1, entry 1) was also
examined. Since Ni1Co1@NC is magnetic, separation of the catalyst from the reaction mixture is
straightforward. Four consecutive reactions with the same catalyst, separated simply by magnetic
filtration, demonstrated that the catalyst could be recycled (first run: 97% conv., 79% yield of 1a;
second run: 98% conv., 81% yield of 1a; third run: 95% conv., 71% yield of 1a; fourth run: 80%
conv., 54% yield of 1a).[15] Further support for the heterogeneous nature of catalysis was
provided by a hot filtration experiment. After seven hours of reaction, the reaction mixture was
rapidly filtered under inert atmosphere and the resulting filtrate allowed to react for additional 17
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 32
1
2
3
4
5
6
hours. Analysis of the reaction mixture clearly showed that the liquid phase contained no
catalytic activity.[16]
7
8
9
Scheme 2. Influence of Functional Groups on the Reaction Outcomea
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
8 mol% Ni1Co1@NC
O
Et3SiO
8 mol% PhCO2H
R
H
1.5 equiv HSiEt3
R
R
OSiEt3
1a j
PhMe (1 M), 90 °C, 24 h
0.5 mmol
t-Bu
Et3SiO
Et3SiO
OSiEt3
1a, 68%
OSiEt3
1b, 79%
t-Bu
i-Pr
Ph
Et3SiO
Et3SiO
OSiEt3
1c, 80%
OSiEt3
1d, 72%
i-Pr
Ph
F
Et3SiO
Et3SiO
F
F
OSiEt3
OSiEt3
1f, 45%
F
1e, 66%
CF3
Et3SiO
Et3SiO
Cl
Cl
OSiEt3
1h, 73%
OSiEt3
1g, 60%
F3C
OMe
Et3SiO
Et3SiO
N
OSiEt3
1i, 10%b
N
OSiEt3
1j, 0%b
MeO
a Isolated yield of purified product. b NMR yield.
ACS Paragon Plus Environment
6

Page 7 of 32
ACS Catalysis
1
2
3
4
5
6
7
8
9
After having established effective conditions for the silylative pinacol coupling of benzaldehyde,
the influence of functional groups was examined (Scheme 2). The unsubstituted benzaldehyde
afforded 68% isolated yield of the desired product 1a. Introducing aliphatic or aromatic
substituents on the substrate led to an increase in yield of the pinacol coupling products (1b–d).
A substrate bearing a fluoride substituent in the meta position also provided the pinacol product
in a good yield, however, using a paraꢀfluoroꢀsubstituted substrate led to a small drop in yield
(1e–f). In contrast, a paraꢀtrifluoromethyl group was wellꢀtolerated (1g). Interestingly, a chloroꢀ
substituted substrate led to 73% yield of pinacol coupling product without any trace of
competing dehalogenation (1h). Finally, the use of paraꢀmethoxybenzaldehyde led to
significantly reduced conversion and yield (1i). Whether this decrease is due to the coordinating
features of the methoxy group or its strong electronꢀdonating ability is unclear at this point,
however, an electron poor substrate with a coordinating group (1j) also inhibited the
reaction.[17]
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
With an efficient catalytic protocol as platform, we turned our attention to investigation of the
reaction mechanism of the carbon–carbon bond forming pinacol reaction. Classically, the
homogeneous reaction proceeds through two initial single electron transfers from the metal
catalyst (Mn) to two aldehydes affording two ketyl radicals bridged by the metal (Mn+2). After
dimerization of the metalꢀbound ketyl radicals, reaction with the chlorosilane releases the
silylated pinacol product from the metal catalyst which subsequently is reduced by a
stoichiometric metal reductant (Scheme 3a).[8] Although our reaction uses only a hydrosilane
instead of a chlorosilane and a stoichiometric reductant, we first considered the possibility of an
analogous classical pathway for our heterogeneous reaction. In this scenario, an initial reaction
between catalyst and aldehyde would oxidize the metal nanoparticles and reduce the aldehyde
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 32
1
2
3
4
5
6
7
8
9
forming a surfaceꢀbound ketyl radical. After dimerization of the ketyl radicals, reaction with the
silane could liberate the silylated pinacol coupling product and a surface hydride. Release of H2
would ultimately reduce the catalyst and close the catalytic cycle (Scheme 3b). Several aspects
of this pathway are very similar to recent reports in bimetallic homogeneous hydrosilylation.[18]
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 3. Classical Catalytic Cycles Proceeding through an Initial Pinacol Coupling
Followed by Silylation
a) Classical homogeneous catalytic cycle:
b) Heterogeneous catalytic cycle by analogy to the
classical homogeneous cycle:
O
O
metalII
M0
2
H2
2
M0
Ph
Ph
Ph
Ph
metal0
Ph
Ph
O
(reductant)
H
H
MII
O
O
O
Ph
O
Ph
O
MII
MII
OSiEt3
Ph
OSiEt3
Ph
O
Ph
O
MII
Ph
Ph
Ph
OSiEt3
MII
2 HSiEt3
2 Et3SiCl
OSiEt3
MII
In line with this possible reaction pathway, Radivoy et al. have previously reported a pinacol
coupling using cobalt nanoparticles in a noncatalytic fashion in the absence of a silane.[19]
Under those conditions, the metal nanoparticles serve as the stoichiometric reductant.
Surprisingly, for the Ni1Co1@NC catalyst used here, the encapsulation of the metal alloy
nanoparticles in nitrogenꢀdoped carbon appears to protect them from such reactivity. In a
stoichiometric experiment, one equivalent Ni1Co1@NC was allowed to react with two
equivalents of aldehyde in the absence of HSiEt3. After 24 hours at 90 °C, essentially no reaction
had occurred and the aldehyde was recovered (Scheme 4a). This result illustrates that reaction
ACS Paragon Plus Environment
8

Page 9 of 32
ACS Catalysis
1
2
3
4
5
6
7
8
9
between aldehyde and Ni1Co1@NC is unlikely as the first step in the catalytic cycle. Comparison
of transmission electron microscopy (TEM) images of the catalyst material before use and after
four consecutive catalytic reactions also
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 1. TEM image of unused catalyst (left) and catalyst after four cycles (right). The content,
size, and distribution of the metal nanoparticles are unaffected by the reaction conditions.
Scheme 4. Stoichiometric Studies
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 32
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
revealed no change in the metal nanoparticles (Figure 1). In addition, at the end of the four
recycles, the effective catalyst loading can be considered 2 mol%, thus the metal cannot be the
ultimate reductant. In combination, these results demonstrate that the metal in Ni1Co1@NC is
protected from the classical reactivity by the encapsulation in nitrogenꢀdoped carbon.
Although the pinacol product was only observed in trace amounts (1%) in the stoichiometric
experiment, we evaluated if leaching of a metal pinacolate could serve as a highly active
homogeneous catalyst. A reaction was performed in the absence of silane allowing for the
potential formation of a soluble metal pinacolate. The reaction mixture was then filtered, to
remove only the solid catalyst material, and HSiEt3 subsequently added. After additional 24
hours under the reaction conditions, no trace of the silylated pinacol coupling product (1a) was
observed.[15]
As a final examination of the pinacol coupling/silylation pathway, we tested the chemical
competency of hydrobenzoin, the authentic pinacol coupling product (Scheme 4b). After
1
subjection to the reaction conditions, 1a could not be detected by H NMR analysis.
ACS Paragon Plus Environment
10

Page 11 of 32
ACS Catalysis
1
2
3
4
5
6
Accordingly, hydrobenzoin does not appear to be a productive intermediate on the reaction
pathway.
7
8
9
Next, we focused on obtaining experimental rate orders for the homogenous reaction
components. A preliminary time course investigation revealed an induction period of
approximately one hour.[15] Accordingly, the initial rate measurements were made between two
and four hours on the linear section after the induction period. Using this approach, it was found
that the reaction is essentially zero order in benzaldehyde and nearly first order in HSiEt3 (Figure
2). These results exclude that reaction between benzaldehyde and silane is the rateꢀdetermining
step and instead indicate that only the silane is involved between the ground state and the rateꢀ
determining transition state. To further support this suggestion, the presence of a kinetic isotope
effect was examined. By comparing the absolute rates in separate reaction vessels using either
HSiEt3 or DSiEt3, a normal deuterium kinetic isotope effect (kH/kD = 1.9) was observed (Scheme
5).[15] The magnitude of the kinetic isotope effect indicates that the H–Si bond is either cleaved
in the rateꢀdetermining step or reversible H–Si cleavage is occurring on the pathway between the
ground state of the catalyst and the turnoverꢀlimiting transition state. No isotope scrambling of
the aldehyde hydrogen was observed in the recovered benzaldehyde, when using DSiEt3.
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
ACS Paragon Plus Environment
11

ACS Catalysis
Page 12 of 32
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 2. Determination of rate orders in HSiEt3 and benzaldehyde (HSiEt3 0.81–2.08 M;
PhCHO 0.74–1.37 M).
Scheme 5. Detection of Kinetic Isotope Effect by Independent Rate Measurements
8 mol% Ni1Co1@NC
O
Et3SiO
8 mol% PhCO2H
1.5 equiv HSiEt3
H
PhMe (1 M), 90 °C, 2-7 h
OSiEt3
1a
0.50 mmol
8 mol% Ni1Co1@NC
8 mol% PhCO2H
1.5 equiv DSiEt3
O
Et3SiO
H
PhMe (1 M), 90 °C, 2-7 h
OSiEt3
1a
0.50 mmol
KIE = 1.9 (initial rates in separate vessels)
Finally, the possibility of a radical pathway was investigated. When a reaction was performed in
the presence of 0.50 equivalents 2,2,6,6ꢀtetramethylpiperidinꢀ1ꢀoxyl (TEMPO), no conversion of
ACS Paragon Plus Environment
12

Page 13 of 32
ACS Catalysis
1
2
3
4
5
6
7
8
9
benzaldehyde had occurred after 5 hours (Scheme 6a). Instead significant amounts of a TEMPO–
SiEt3 adduct was observed. Repeating the reaction in the absence of benzaldehyde also afforded
the TEMPO–SiEt3 adduct in 70% yield (Scheme 6b). In contrast, no conversion of starting
materials or formation of TEMPO–SiEt3 was observed in the absence of Ni1Co1@NC (Scheme
6c).[15] The influence of 2,6ꢀdiꢀtertꢀbutylꢀ4ꢀmethylphenol (BHT), a known radical inhibitor, on
the standard reaction was also examined. The presence of 20 mol% BHT completely prevented
conversion of the aldehyde.[15] Overall, these results support the suggestion that silyl radicals
are formed during the reaction and that a radical pathway is operating.
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. Experiments with TEMPO demonstrated Rapid Formation of TEMPO–SiEt3 in
the Presence of Ni1Co1@NC.
Collectively, the mechanistic investigations above present several inconsistencies with the
pathway in Scheme 3b. A new pathway consistent with the mechanistic evidence presented
above is illustrated in Scheme 7. Here, the Ni1Co1@NC initially cleaves the H–Si bond liberating
a silyl radical. This radical can attack the oxygen atom of the aldehyde forming a highly
stabilized benzylic radical.[10] The benzyl radical can then add to benzaldehyde, possibly, in a
ACS Paragon Plus Environment
13

ACS Catalysis
Page 14 of 32
1
2
3
4
5
6
7
8
9
Brønsted acidꢀactivated form which could explain the influence of benzoic acid on the
selectivity.[20] Alternatively, two benzyl radicals can dimerize affording the overall silylated
pinacol coupling products.[21] We have previously demonstrated that very little silane is
converted in a highly related system, but where no aldehyde or other reagent is present to
consume the silyl radicals.[13] Accordingly, the initial cleavage of the H–Si bond by
Ni1Co1@NC can be reversible under certain conditions. Reversibility or a late transition state
might account for the moderate kinetic isotope effect observed. Finally, a pressure buildꢀup in
the reaction vessel was observed which corresponds to the release of H2 from
Ni1Co1@NC.[22,23]
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 7. Catalytic Cycle Proceeding through an Initial Formation of Silyl Radicals. RDS
= Rate-Determining Step. For the sake of simplicity all steps are drawn as irreversible.
Due to the proposed formation of silyl radicals, hydrosilylation of benzophenone was attempted
for a comparison with the much more elaborate conditions in Scheme 1b. Gratifyingly, when
benzophenone was subjected to our simplified catalytic conditions, the corresponding
hydrosilylation product was observed in a comparable yield (Scheme 8).
ACS Paragon Plus Environment
14

Page 15 of 32
ACS Catalysis
1
2
3
4
5
6
Scheme 8. Radical Hydrosilylation of Benzophenone under Simplified and Catalytic
Conditions
7
8
9
8 mol% Ni1Co1@NC
1.2 equiv HSiEt3
O
OSiEt3
Ph
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
62% NMR-yield
Ph
Ph
Ph
PhH (0.9 M), 90 °C, 24 h
CONCLUSIONS
In summary, we have developed an improved protocol for the carbon–carbon bond forming
pinacol coupling in terms of simplicity, safety, and sustainability. The protocol uses a simple
hydrosilane and a stable, recyclable, heterogeneous catalyst, which is based on cheap and earthꢀ
abundant base metals, thereby circumventing the need for explosive singleꢀuse radical initiators
and stoichiometric tin reagents. Furthermore, a rare mechanistic investigation, in this field of
heterogeneous catalysis, of the operating reaction pathway for the soluble reagents was
undertaken. The experimental data include stoichiometric studies, experimental rate orders,
kinetic isotope effect, and trapping of silyl radicals. Overall the data are consistent with initial
silyl radical formation. This knowledge was applied directly toward the successful radical
hydrosilylation of benzophenone. Filtration tests and recycling of the catalyst were also
performed, clearly supporting the heterogeneous nature of the catalyst.
A better understanding of the modes of operation for catalysts consisting of base metals
supported on nitrogenꢀdoped carbon will be highly valuable for more efficient design of future
transformations in this field. Further inꢀdepth studies of reactions with these novel materials are
underway in our laboratories.
ASSOCIATED CONTENT
ACS Paragon Plus Environment
15

ACS Catalysis
Page 16 of 32
1
2
3
4
5
6
Supporting Information: Experimental details, characterization data, GCꢀMS data, NMR
spectroscopic data of all products.
7
8
9
AUTHOR INFORMATION
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
Corresponding Author
*skk@kemi.dtu.dk
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The authors gratefully acknowledge the support of the Danish Council for Independent Research,
Grant. 6111ꢀ00237 and the support from Villum Fonden research grant (13158). We thank Jerrik
Mielby for assistance with TEM.
REFERENCES
(1) Rothenberg, G. Catalysis: Concepts and Green Applications; Wiley–VCH: Weinheim, 2008.
(2) Wang, F.; Mielby, J.; Richter, F. H.; Wang, G.; Prieto, G.; Kasama, T.; Weidenthaler, C.;
Bongard, H.ꢀJ.; Kegnæs, S.; Fürstner, A.; Schüth, F. Angew. Chem. Int. Ed. 2014, 53, 8645ꢀ8648.
(3) For a recent review, see: He, L.; Weniger, F.; Neumann, H.; Beller, M. Angew. Chem. Int.
Ed. 2016, 55, 12582ꢀ12594.
(4) (a) Groves, M. N.; Chan, A. S. W.; MalardierꢀJugroot, C.; Jugroot, M. Chem. Phys. Lett.
2009, 481, 214ꢀ219. (b) Jagadeesh, R. V.; Surkus, A.ꢀE.; Junge, H.; Pohl, M.ꢀM.; Radnik, J.;
Rabeah, J.; Huan, H.; Schünemann, V.; Brückner, A.; Beller, M. Science 2013, 342, 1073ꢀ1076.
ACS Paragon Plus Environment
16

Page 17 of 32
ACS Catalysis
1
2
3
4
5
6
7
8
9
(c) Zhang, P.; Gong, Y.; Li, H.; Chen, Z.; Wang, Y. Nat. Commun. 2013, 4, 1593. (d) Banerjee,
D.; Jagadeesh, R. V.; Junge, K.; Pohl, M.ꢀM.; Radnik, J.; Brückner, A.; Beller, M. Angew. Chem.
Int. Ed. 2014, 53, 4359ꢀ4363. (e) Zhong, W.; Liu, H.; Bai, C.; Liao, S.; Li, Y. ACS Catal. 2015,
5, 1850ꢀ1856. (f) Iosub, A. V.; Stahl, S. S. Org. Lett. 2015, 17, 4404ꢀ4407. (g) Arrigo, R.;
Schuster, M. E.; Xie, Z.; Yi, Y.; Wowsnick, G.; Sun, L. L.; Hermann, K. E.; Friedrich, M.; Kast,
P.; Hävecker, M.; KnopꢀGericke, A.; Schlögl, R. ACS Catal. 2015, 5, 2740ꢀ2753. (h) Zhang, L.;
Wang, A.; Wang, W.; Huang, Y.; Liu, X.; Miao, S.; Liu, J.; Zhang, T. ACS Catal. 2015, 5, 6563ꢀ
6572. (i) Xia, W. Catal. Sci. Technol. 2016, 6, 630ꢀ644. (j) Jagadeesh, R. V.; Stemmler, T.;
Surkus, A.ꢀE.; Junge, H.; Beller, M. Nat. Protoc. 2015, 10, 548ꢀ557. (k) Liu, L.; Concepcion, P.;
Corma, A. J. Catal. 2016, 340, 1ꢀ9. (l) Liu, W.; Zhang, L.; Yan, W.; Liu, X.; Yang, X.; Miao, S.;
Wang, W.; Wang, A.; Zhang, T. Chem. Sci. 2016, 7, 5758ꢀ5764.
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
(5) For a recent mechanistic investigation, which does include the solution phase reagents, of a
reaction using a Co@NC catalyst, see: Formenti, D.; Ferretti, F.; Topf, C.; Surkus, A.ꢀE.; Pohl,
M.ꢀM.; Radnik, J.; Schneider, M.; Junge, K.; Beller, M.; Ragaini, F. J. Catal. 2017, 351, 79ꢀ89.
(6) (a) Metal-Catalyzed Cross-Coupling Reactions and More; de Meijere, A.; Bräse, S.;
Oestreich, M., Eds.; Wiley–VCH, 2014. (b) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001,
34, 18ꢀ29.
(7) (a) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. Rev. 1994, 94, 2483ꢀ2547.
(b) Döbler, C.; Mehltretter, G. M.; Sundermeier, U.; Beller, M. J. Organomet. Chem. 2001, 621,
70ꢀ76.
(8) Hirao, T. in Metal Catalyzed Reductive C–C bond Formation; Ed. Krische M. J., Ed; Topics
in Current Chemistry; SpringerꢀVerlag: Berlin, 2007; Vol. 279, pp 53–75.
ACS Paragon Plus Environment
17

ACS Catalysis
Page 18 of 32
1
2
3
4
5
6
7
8
9
(9) For leading references and recent examples, see: (a) Hays, D. S.; Fu, G. C. J. Am. Chem. Soc.
1995, 117, 7283ꢀ7284. (b) With, T. Angew. Chem. Int. Ed. Engl. 1996, 35, 61ꢀ63. (c)
Freudenberger, J. H.; Konradi, A. W.; Pedersen, S. F. J. Am. Chem. Soc. 1989, 111, 8014ꢀ8016.
(d) Nakajima, M.; Fava, E.; Loescher, S.; Jiang, Z.; Rueping, M. Angew. Chem. Int. Ed. 2015,
54, 8828ꢀ8832. (e) Okamoto, S.; Kojiyama, K.; Tsujioka, H.; Sudo, A. Chem. Commun. 2016,
52, 11339ꢀ11342.
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
(10) Nishiyama, Y.; Kajimoto, H.; Kotani, K.; Nishida, T.; Sonoda, N. J. Org. Chem. 2002, 67,
5696ꢀ5700.
(11) Prices from SigmaAldrich.com: HSi(TMS)3 is 4238 EUR/mol; HSiEt3 is 116 EUR/mol.
(12) Deng, J.; Ren, R.; Deng, D.; Bao, X. Angew. Chem. Int. Ed. 2015, 54, 2100ꢀ2104.
(13) Kramer, S.; Mielby, J.; Buss, K.; Kasama, T.; Kegnæs, S. ChemCatChem 2017, 9, 2930ꢀ
2934.
(14) When using a catalyst batch that had been stored for six weeks under air at ambient
temperature, a small decrease in activity and selectivity was observed (89% conversion, 54% 1a,
34% 2).
(15) See Supporting Information for additional details.
(16) The amount of metal in the filtrate was examined by ICPꢀOES. The filtrate contained 6 ppm
nickel and 17 ppm cobalt.
(17) Except for paraꢀOMe, the yields and stereoselectivities are comparable to recent reports
using photoredox catalysis (references 9d–e).
ACS Paragon Plus Environment
18

Page 19 of 32
ACS Catalysis
1
2
3
4
5
6
7
8
9
(18) (a) Zhou, W.; Marquard, S. L.; Bezpalko, M. W.; Foxman, B. M.; Thomas, C. M.
Organometallics 2013, 32, 1766ꢀ1772. (b) Mukhopadhyay, T. K.; Rock, C. L.; Hong, M.;
Ashley, D. C.; Groy, T. L.; Baik, M.ꢀH.; Trovitch, R. J. J. Am. Chem. Soc. 2017, 139, 4901ꢀ
4915.
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
(19) Nador, F.; Mascaro, E.; Castro, M.; Vitale, C.; Radivoy, G. Arkivoc 2011, 312ꢀ326.
(20) The exact role of benzoic acid is unclear. The influence on selectivity suggests an
interaction with the substrates or intermediates. A role in the hydrogen evolution is also possible.
(21) To probe for a possible selectivityꢀdetermining step involving the arylaldehyde, a
competitive Hammett study was attempted by independently reacting three different paraꢀ
substituted arylaldehydes (tꢀBu, F, and CF3) in competition with benzaldehyde. While the
correlation with σ• was much better than σpara, the error on the linear correlation was significant
(R2 = 0.22 for σ•; R2 = 0.004 for σpara). These results could indicate that the pathway after
addition of the silyl radical to the arylaldehyde is not a simple dimerization of siloxybenzyl
radicals.
(22) A related mechanism has been proposed for a heterogeneous goldꢀcatalyzed reaction;
however, the reaction mechanism was merely speculative, based only on the observed silylative
pinacol product, and not supported by any mechanistic evidence. The isolated yield of 1a with
this precious metal catalyst was 32% (95% conversion) for the analogous reaction: Raffa, P.;
Evangelisti, C.; Vitulli, G.; Salvadori, P. Tetrahedron Lett. 2008, 49, 3221ꢀ3224.
(23) The reaction mechanism proposed for our heterogeneous system displays many similarities
to a recent homogeneous system for HSiEt3 activation: (a) Toutov, A. A.; Liu, W.ꢀB.; Betz, K.
ACS Paragon Plus Environment
19

ACS Catalysis
Page 20 of 32
1
2
3
4
5
6
7
8
9
N.; Fedorov, A.; Stoltz, B. M.; Grubbs, R. H. Nature 2015, 518, 80ꢀ84. (b) Liu, W.ꢀB.; Schuman,
D. P.; Yang, Y.ꢀF.; Toutov, A. A.; Liang, Y.; Klare, H. F. T.; Nesnas, N.; Oestreich, M.;
Blackmond, D. G.; Virgil, S. C.; Banerjee, S.; Zare, R. N.; Grubbs, R. H.; Houk, K. N.; Stoltz, B.
M. J. Am. Chem. Soc. 2017, 139, 6867ꢀ6879.
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
TOC graphic
Heterogeneous catalysis
magnetic
catalyst separation
O
Et3SiO
cat. Ni1Co1@NC
Ar
Ar
H
Ar
OSiEt3
HSiEt3
Mechanistic Study:
• Experimental rate orders
• Kinetic isotope effect
• Radical scavengers
• Stoichiometric tests
ACS Paragon Plus Environment
20

Page 21 of 32
ACS Catalysis
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
80x43mm (300 x 300 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 32
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
(a) Classical approach to pinacol coupling. (b) Typical reaction conditions for accessing silyl radicals
113x57mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 23 of 32
ACS Catalysis
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
Influence of Functional Groups on the Reaction Outcomea
116x185mm (300 x 300 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 32
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
Classical Catalytic Cycles Proceeding through an Initial Pinacol Coupling Followed by Silylation
166x64mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 25 of 32
ACS Catalysis
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
Stoichiometric Studies
103x68mm (300 x 300 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 32
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
Detection of Kinetic Isotope Effect by Independent Rate Measurements
96x59mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 27 of 32
ACS Catalysis
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
Experiments with TEMPO demonstrated Rapid Formation of TEMPO–SiEt3 in the Presence of Ni1Co1@NC
131x69mm (300 x 300 DPI)
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 32
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
Catalytic Cycle Proceeding through an Initial Formation of Silyl Radicals. RDS = Rate-Determining Step. For
the sake of simplicity all steps are drawn as irreversible
114x56mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 29 of 32
ACS Catalysis
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
Radical Hydrosilylation of Benzophenone under Simplified and Catalytic Conditions
105x18mm (300 x 300 DPI)
ACS Paragon Plus Environment