A comparative analysis of hydrosilative amide reduction catalyzed by first-row transition metal (Mn, Fe, Co, and Ni): N -phosphinoamidinate complexes

Article abstract of DOI:10.1039/c8dt04221g

A comparative study of the performance of (PN)M(N(SiMe₃)₂) (M = Mn, Fe, Co, and Ni) pre-catalysts supported by N-phosphinoamidinate ligation, as well as M(N(SiMe₃)₂)_n (M = Li, Na, K, Mn, Fe, and Co) pre-catalysts, in the hydrosilative reduction of selected tertiary amide test substrates using PhSiH₃ is reported. Encouraged by the performance observed herein for (PN)Ni(N(SiMe₃)₂) in the reduction of both N,N-dibenzylbenzamide and N,N-diisopropylbenzamide, further competitive testing involving the known complex (PN)Ni(NHdipp) (dipp = 2,6-diisopropylphenyl), as well as the new and crystallographically characterized mononuclear complexes (PN)Ni(OR) (R = 2,6-dimethylphenyl or tBu), revealed (PN)Ni(OtBu) to be particularly effective in such reduction chemistry, including transformations involving the secondary amides N-benzylbenzamide and caprolactam.

Full text of DOI:10.1039/c8dt04221g

Dalton
Transactions
View Article Online
View Journal
PAPER
A comparative analysis of hydrosilative amide
reduction catalyzed by first-row transition metal
Mn, Fe, Co, and Ni) N-phosphinoamidinate
complexes†
Cite this: DOI: 10.1039/c8dt04221g
(
a
a
b
Casper M. Macaulay, Takahiko Ogawa, Robert McDonald,
c
a
a
Orson L. Sydora, * Mark Stradiotto * and Laura Turculet
*
A comparative study of the performance of (PN)M(N(SiMe
ported by N-phosphinoamidinate ligation, as well as M(N(SiMe
catalysts, in the hydrosilative reduction of selected tertiary amide test substrates using PhSiH
Encouraged by the performance observed herein for (PN)Ni(N(SiMe ) in the reduction of both N,N-
3
)
2
) (M = Mn, Fe, Co, and Ni) pre-catalysts sup-
(M = Li, Na, K, Mn, Fe, and Co) pre-
is reported.
3 2 n
) )
3
3 2
)
dibenzylbenzamide and N,N-diisopropylbenzamide, further competitive testing involving the known
complex (PN)Ni(NHdipp) (dipp = 2,6-diisopropylphenyl), as well as the new and crystallographically
characterized mononuclear complexes (PN)Ni(OR) (R = 2,6-dimethylphenyl or tBu), revealed (PN)Ni
Received 23rd October 2018,
Accepted 7th November 2018
DOI: 10.1039/c8dt04221g
(
OtBu) to be particularly effective in such reduction chemistry, including transformations involving the
rsc.li/dalton
secondary amides N-benzylbenzamide and caprolactam.
effecting amide reductions of this type have been identified. In
Introduction
7
8
2009, the groups of Beller and Nagashima independently
There is sustained interest in the development of mild reported the first successful hydrosilative reduction of amides
methods for the preparation of amines, given their ubiquitous employing 3d transition metal catalysis, in which Fe carbonyl
nature in sought-after target compounds, including active pre-catalysts were employed (6–30 mol% Fe, 100 °C, 24 h) in
1
2
pharmaceutical ingredients. While amide reduction rep- reductions involving primarily tertiary amide substrates. In the
7
resents a potentially useful route to amines, such protocols aforementioned report from the Beller group, isotopic label-
typically require the use of harsh alkali metal hydrides (e.g., ing studies provided evidence in support of the mechanism
LiAlH ) or related reductants. In this context, the use of more featured in Scheme 1, whereby metal-catalyzed hydrosilation
4
3
mild hydrosilane reducing agents under the influence of tran- of the amide carbonyl group (e.g., depicted for 1a and 1b)
4
5
6
sition metal or other (e.g., boron or zinc ) catalysis, without affords the corresponding O-silylated hemiaminal (e.g., 1a′ and
recourse to photochemical activation, represents an attractive 1b′); hydrosilation of the derived iminium ion in turn affords
alternative to more conventional amide reduction protocols. the corresponding amine. Subsequent reports by the groups of
9
10
11
The application of 3d (i.e., first row) transition metal catalysts Sortais and Darcel, Buitrago and Adolfsson, and Driess
in the hydrosilative reduction of amides is appealing relative document the use of N-heterocyclic carbene-ligated Fe species
to the use of precious metal catalysts given the comparatively for such transformations. Prior work involving the Mn-, Co-, or
low cost and high natural abundance of the former; however, a Ni-catalyzed hydrosilative reduction of amides is much more
relatively small number of such catalysts that are capable of limited. The use of Co (CO) (1 mol% Co, 100 °C, 3–16 h) in
2
8
1
2
such reductions of tertiary amides has been disclosed, and
the application of Ni catalysis in hydrosilative amide
reductions is limited to a report by Mamillapalli and Sekar
a
Department of Chemistry, Dalhousie University, 6274 Coburg Road, P.O. 15000,
13
Halifax, Nova Scotia B3H 4R2, Canada. E-mail: mark.stradiotto@dal.ca,
laura.turculet@dal.ca
regarding the reduction of α-keto amides (5 mol% Ni(OAc)2,
b
X-Ray Crystallography Laboratory, Department of Chemistry, University of Alberta,
10 mol% TMEDA, 10 mol% KOtBu, 25–60 °C, 8–48 h), as well
1
4
Edmonton, Alberta T6G 2G2, Canada
as a report by Garg and co-workers regarding the hydrosila-
c
Research and Technology, Chevron Phillips Chemical Company LP, 1862 Kingwood
tive reduction of amides employing NiCl (dme) (10 mol% Ni,
2
Drive, Kingwood, Texas 77339, USA. E-mail: sydorol@cpchem.com
115 °C, 24 h). The demonstrated ability of these Ni-based cata-
†
Electronic supplementary information (ESI) available. CCDC 1856638 and
856639. For ESI and crystallographic data in CIF or other electronic format see
lysts to effect the hydrosilative reduction of both tertiary and
1
DOI: 10.1039/c8dt04221g
secondary amides with synthetically useful substrate scope is
This journal is © The Royal Society of Chemistry 2018
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Encouraged by the useful catalytic profile exhibited by (PN)
Mn(N(SiMe ) ), we recently initiated a study focussed on evalu-
3
2
ating in a systematic manner the catalytic performance of (PN)
M(N(SiMe ) (M = Mn, Fe, Co, and Ni) and structurally related
N-phosphinoamidinate pre-catalysts, in addition to
3 2
)
M
2
1
(
N(SiMe
the hydrosilative reduction of selected amide test substrates
1a and 1b, Scheme 1). We disclose herein the results of this
3 2 n
) ) (M = Li, Na, K, Mn, Fe, and Co) pre-catalysts, in
(
investigation, which confirmed the superiority of 3d transition
metals over simple alkali salts in this application, as well as
the benefits of N-phosphinoamidinate ligation. Notably, the
new mononuclear three-coordinate Ni pre-catalyst (PN)Ni
(
OtBu) proved optimal, enabling both the room temperature
reduction of N,N-diisopropylbenzamide (1b) in a manner that
had not been achieved previously, and the reduction of
selected secondary amide test substrates.
Results and discussion
We commenced our catalytic reactivity survey by examining the
reduction of N,N-dibenzylbenzamide (1a) to dibenzylamine
Scheme 1 The hydrosilative reduction of amides reported herein.
(
2a)
with
PhSiH
3
employing
the
three-coordinate
N-phosphinoamidinate pre-catalysts (PN)M(N(SiMe ) ) (M =
3
2
Mn, Fe, Co, and Ni) under reasonably mild reaction conditions
1
5
unique amongst reported 3d transition metal catalysts. Prior (2 mol% M, 75 °C, 1 h, Table 1, entries 1–4). In all cases, high
to 2017, only three isolated table entries documenting the Mn- conversion of the amide was achieved, with the Mn and Ni
catalyzed hydrosilative reduction of amides existed in the lit- pre-catalysts offering optimal catalytic performance in terms of
erature, involving the reduction of dimethylformamide or di- conversion to 2a (>80%). We interpret the observation of
ethylformamide by use of CpMn(CO) (5 mol% Mn, 120 °C, HNBn2 as a side-product in these reactions as arising from
3
1
6
2
4 h), or featuring the use of Mn (CO)10 for the reduction of decomposition of the putative O-silylated hemiaminal inter-
2
N-acetylpiperidine (2 mol% Mn, 5 mol% Et
1
2
NH co-catalyst, mediate (1a′) upon workup prior to calibrated GC analysis. The
00 °C, 16 h). In building on our research program targeting reduction of 1a under conditions outlined in entry 4 of Table 1
highly effective 3d transition metal catalysts supported by employing benzene in place of THF provided support for such
1
7
1
8,19
N-phosphinoamidinate ligation,
we disclosed that three- a proposal; while complete consumption of 1a was achieved in
coordinate (PN)Mn(N(SiMe ) ) (depicted in Scheme 1) is an benzene after 1 h accompanied by the formation of 2a (55%)
3
2
effective pre-catalyst for the hydrosilative reduction of various and HNBn (37%), after a total reaction time of 18 h, high con-
2
carbonyl compounds by use of PhSiH
tertiary (but not primary or secondary) amides under reason- were observed.
ably mild conditions (5 mol% Mn, 75 °C, 1 h; 2 mol% Mn,
Under analogous conditions, M(N(SiMe ) ) pre-catalysts
3
(exclusively), including version to 2a (88%) and negligible amounts of HNBn2 (2%)
3
2 2
2
0
2
5 °C, 3–48 h).
(M = Mn, Fe, and Co) proved inferior to their respective
3
Table 1 Pre-catalyst screening for the hydrosilative reduction of 1a (0.2 mmol) with PhSiH (0.4 mmol) in THF at 75 °C for 1 h. Conversion/yields
given on the basis of response-factor calibrated GC data using authentic samples versus dodecane as an internal standard, following workup of the
crude reaction mixture
Entry
Pre-catalyst
Loading (%)
Conversion 1a (%)
Yield NBn
3
(2a, %)
2
Yield HNBn (%)
1
2
3
4
5
6
7
8
9
(PN)Mn(N(SiMe
3
)
2
)
2
2
2
2
2
2
2
30
30
30
>95
>95
>95
>95
67
39
44
84
>95
>95
84
73
61
85
63
33
43
27
40
23
<5
10
<5
<5
<5
<5
<5
27
32
47
(PN)Fe(N(SiMe
(PN)Co(N(SiMe
(PN)Ni(N(SiMe
Mn(N(SiMe
Fe(N(SiMe
Co(N(SiMe
3 2
) )
3
)
2
)
3
)
2
)
3
)
)
2
)
2
)
2
3
2
)
)
3
2
2
Li(N(SiMe
Na(N(SiMe
3
)
2
)
3
)
2
)
10
3 2
K(N(SiMe ) )
Dalton Trans.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
N-phosphinoamidinate counterparts (Table 1, entries 5–7). effecting the reduction of both 1a and 1b to the corres-
Moreover, while high conversion of 1a was achieved by use of ponding tertiary amine (2a and 2b) in >80% yield under the
alkali metal M(N(SiMe
30 mol% Li, Na, or K, Table 1, entries 8–10), poor selectivity opted to examine alternative (PN)NiX pre-catalysts in such
for 2a was achieved, due in part to the apparently inefficient transformations. In addition to comparing the performance
conversion of intermediate 1a′ as evidenced by the substantial of (PN)Ni(N(SiMe ) versus our previously reported (PN)Ni
amounts of HNBn that were detected in the product mixture (NHdipp) (dipp = 2,6-diisopropylphenyl),
3 2
) ) pre-catalysts at rather high loadings test conditions employed. Building on this observation, we
(
3 2
)
1
8f
2
we identified
following workup. Although not conclusively demonstrated, related three-coordinate (PN)Ni(OR) (R = 2,6-dimethylphenyl
the results presented in Table 1 collectively suggest that tran- or tBu) complexes as being potentially interesting targets of
2
2
sition metal catalysis may be involved in facilitating the con- inquiry. Exposure of ((PN)NiCl)
2
to NaOtBu (two equiv.) in
version of 1a′ to 2a. THF enabled the isolation of (PN)Ni(OtBu) upon workup, and
The hydrosilative reduction of N,N-diisopropylbenzamide treatment of (PN)Ni(NHdipp) with 2,6-dimethylphenol (i.e.,
1b) to benzyldiisopropylamine (2b) was examined sub- dmpOH, one equiv.) in benzene upon workup afforded (PN)
(
sequently. A survey of the literature reveals 1b to be a signifi- Ni(Odmp). In each case the targeted (PN)Ni(OR) complex was
cantly more challenging substrate in this chemistry relative to obtained in analytically pure form, and was characterized by
1
a, often requiring forcing reaction conditions to achieve suit- use of both NMR spectroscopic and single-crystal X-ray crys-
4
able levels of conversion. For example, in Beller’s pioneering tallographic techniques. In keeping with (PN)Ni(N(SiMe ) )
report on the use of Fe
lative reduction of amides, only 59% yield of 2b was achieved species were found to be diamagnetic, exhibiting sharp H,
3
2
1
8f
3
(CO)12 as a pre-catalyst for the hydrosi- and (PN)Ni(NHdipp),
the newly prepared (PN)Ni(OR)
1
1
3
1
31
1
even after extended reaction times at elevated temperature
C{ H}, and P{ H} resonances consistent with the proposed
30 mol% Fe, 100 °C, 24 h). In keeping with this trend, the structural formulation. The crystal structures of (PN)Ni(OtBu)
reduction of 1b proved difficult versus 1a for our (PN)M and (PN)Ni(Odmp) are presented in Fig. 1. Complex (PN)Ni
N(SiMe ) ) pre-catalysts (2 mol% M, 75 °C, 1 h, Table 2). (OtBu) exhibits a highly distorted, three-coordinate T-shaped
7
(
(
3
2
Under these reaction conditions, (PN)Co(N(SiMe
Ni(N(SiMe ) each displayed a promising combination of 1b (N(SiMe
consumption and conversion to 2b. While all members of the nation site located trans to phosphorus. Given that crystallo-
M(N(SiMe (M = Mn, Fe, and Co) pre-catalyst series per- graphically characterized examples of mononuclear Ni alkoxo
3 2
) ) and (PN) structure that is qualitatively similar to that of both (PN)Ni
1
8f
3
)
2
3
)
2
) and (PN)Ni(NHdipp),
with an empty coordi-
3 2 2
) )
formed poorly under similar conditions, the observed con- complexes (beyond methoxide) are limited to a small collec-
sumption of 1b when using M(N(SiMe ) ) (M = Li, Na, and K, tion of four-coordinate (PCP)Ni(OR) species recently reported
3
2
3
0 mol% M, 75 °C, 1 h), accompanied by the generation of sig- by Cámpora and co-workers (where R = Et, nBu, iPr, or
2
3
2 2
nificant amounts of 2b (e.g., 37%, M = Na; 49%, M = K), CH CH OH), the solid state structural characterization of
further underscores the potential utility of such alkali metal three-coordinate (PN)Ni(OtBu) warrants discussion. The Ni–O
salts in hydrosilative amide reduction chemistry (Table 2). distance in (PN)Ni(OtBu) (1.7699(11) Å) is markedly shorter
Application of our (PN)M(N(SiMe
longer reaction times and higher catalyst loadings (5 mol% M, Ni(OR) series (1.84–1.89 Å). While it is tempting to invoke Ni–
8 h, Table 2) resulted uniformly in high conversion of 1b, O π-bonding to account for this observation, the lack of
3 2
) ) pre-catalysts employing than the Ni–O contacts observed in the aforementioned (PCP)
1
1
with the Co and Ni members of the series each providing observable hindered rotation phenomena in the H NMR
optimal (83%) formation of 2b.
In evaluating the catalytic screening results featured in
3 2
Tables 1 and 2, only (PN)Ni(N(SiMe ) ) proved capable of
Table 2 Screening for the hydrosilative reduction of 1b (0.2 mmol)
3 2
with PhSiH (0.4 mmol) in THF at 75 °C employing (PN)M(N(SiMe ) )
3
pre-catalysts, reported as % consumption 1b (% conversion to 2b) on
the basis of calibrated GC data
(
PN)M(N(SiMe
3
)
2
) pre-catalysts
Loading
Time (h)
1
(mol% M)
Mn
Fe
Co
Ni
Fig. 1 Crystallographically determined structures of (PN)Ni(OtBu) and
PN)Ni(Odmp). Selected interatomic distances (Å) and angles (°) for (PN)
Ni(OtBu): Ni–P 2.1092(4), Ni–N 1.8589(11), Ni–O 1.7699(11), P–Ni–N
82.97(4), P–Ni–O 114.11(4), N–Ni–O 160.71(6). Selected interatomic dis-
tances (Å) and angles (°) for (PN)Ni(Odmp): Ni–P 2.1273(4), Ni–N 1.8859
a,b
a
a
2
5
>95 (27)
>95 (40)
32 (<5)
92 (56)
72 (47)
>95 (83)
89 (67)
>95 (83)
(
18
a
3 2 2
Using M(N(SiMe ) ) pre-catalyst (M = Mn, Fe, or Co), <10% con-
b
sumption of 1b was observed. Use of alkali metal M(N(SiMe
catalysts at higher loading (30 mol%), afforded high conversion of 1b,
but poor selectivity for 2b (60 (15), M = Li; 83 (37), M = Na; 89 (49), M = 2.061(17), P–Ni–N 83.64(4), P–Ni–O 94.65(4), N–Ni–O 175.61(5), P–Ni–
3 2
) ) pre-
(12), Ni–O 1.8397(11), Ni⋯C8 2.3585(16), Ni⋯H8A 2.023(17), Ni⋯H8B
K).
C8 162.00(5).
This journal is © The Royal Society of Chemistry 2018
Dalton Trans.

View Article Online
Paper
Dalton Transactions
spectrum of at 300 K would suggest that such interactions are
not significant.
Whereas (PN)Ni(OtBu) is apparently a bona fide three-coor-
dinate complex that is devoid of additional coordinative inter-
actions, (PN)Ni(Odmp) is best described as adopting a dis-
torted square-planar geometry, whereby agostic interaction(s)
involving a methyl group on the aryloxo ligand occur at the
position trans to phosphorus. Such agostic interactions appar-
ently do not result in a chemically significant difference in the
Ni–P distance in (PN)Ni(Odmp) (2.1273(4) Å) versus (PN)Ni
(
(
OtBu) (2.1092(4) Å). The Ni–O distance observed in (PN)Ni
Odmp) (1.8397(11) Å), while significantly longer than that employing pre-catalyst (PN)Ni(OtBu).
Scheme 3 Reduction of selected secondary amides (0.2 mmol)
found in (PN)Ni(OtBu) (1.7699(11) Å), is distinctly shorter than
related Ni–O contacts found in some other crystallographically
characterized
square
planar
Ni(aryloxo)
complexes formance of (PN)Ni(OtBu) and (PN)Ni(N(SiMe ) ) can be
3 2
2
4
(1.90–1.92 Å).
attributed in part to the more efficient activation of the
Given that all reported examples of the hydrosilative former. This assertion is qualitatively consistent with stoichio-
reduction of 1b by use of 3d transition metal catalysis feature metric experiments involving treatment of these pre-catalysts
elevated reaction temperatures (≥100 °C), we opted to establish with PhSiH (40 equiv., C D , 25 °C) in the absence of amide
3
6 6
whether our (PN)NiX pre-catalyst may enable room tempera- substrate. Under such conditions, (PN)Ni(OtBu) was comple-
1
31
ture reactivity (5 mol% Ni, 25 °C, 18 h; Scheme 2). With the tely consumed upon mixing ( H and P NMR), and after 4 h
1
8f
exception of (PN)Ni(N(SiMe
examined herein afforded ≥95% conversion of 1b, with (PN)Ni unknown phosphorus-containing species (δ P = 104 ppm) in
OtBu) providing the highest conversion to 2b (84%), followed ca. 2.7 : 1 ratio. By comparison, no reaction was observed
closely by (PN)Ni(Odmp) (80%). In modifying our original con- between the silane and (PN)Ni(N(SiMe ) ) under similar con-
3
)
2
), each of the Ni pre-catalysts afforded a mixture of the known dimer ((PN)NiH)
and an
2
3
1
(
3
2
ditions (5 mol% (PN)Ni(OtBu), 25 °C, 18 h, Scheme 2) only ditions after 20 minutes; after 4 h a mixture of the aforemen-
31
such that the total volume of added liquids/solution was tioned unknown phosphorus-containing species (δ
50 μL rather than 500 μL, >95% conversion of 1b and 90% 104 ppm) and ((PN)NiH) , along with detectable amounts of
P =
2
2
formation of 2b was achieved.
Ph SiH , were observed spectroscopically.
2 2
In monitoring the progress of the reaction (Scheme 2)
Having identified (PN)Ni(OtBu) as being an effective pre-
employing (PN)Ni(OtBu), 46% conversion of 1b and 37% for- catalyst for the hydrosilative reduction of the somewhat chal-
mation of 2b was noted after only 30 min, and 84% conversion lenging tertiary amide 1b, we turned our attention to related
of 1b and 79% formation of 2b was noted after 8 h, suggesting test transformations involving selected secondary amides
that an 18 h reaction time is not required in order to achieve (Scheme 3). Each of N-benzylbenzamide and caprolactam were
high conversions with this particular pre-catalyst. By compari- successfully reduced, under more mild conditions than those
1
4
son, <10% and 31% conversion of 1b was achieved after 2 h reported by Garg and co-workers employing NiCl (dme) as a
2
and 8 h, respectively, when using (PN)Ni(N(SiMe
catalyst under similar conditions. Presuming that each of the (PN)Ni(OtBu) did not prove to be a useful pre-catalyst for the
PN)NiX pre-catalysts afford a common catalytic intermediate reduction of either benzamide or pivalamide under a variety of
3 2
) ) as a pre- pre-catalyst (10 mol% Ni, 115 °C, 24 h). Despite this success,
(
such as (PN)NiH upon reaction with silane, the differing per- reaction conditions, with high conversion of the starting
materials accompanied by the formation of complex reaction
mixtures.
Conclusions
In comparing the performance of (PN)M(N(SiMe ) ) (M = Mn,
3
2
Fe,
Co,
and
Ni)
pre-catalysts
supported
by
N-phosphinoamidinate ligation, as well as M(N(SiMe
3
)
2
)
n
(M =
Li, Na, K, Mn, Fe, and Co) pre-catalysts, in the hydrosilative
reduction of selected tertiary amide test substrates using
PhSiH , only (PN)Ni(N(SiMe ) ) proved effective in the
3 3 2
reduction of both N,N-dibenzylbenzamide and N,N-diiso-
propylbenzamide. The utility of this Ni pre-catalyst in such
reactions prompted our investigation of other related (PN)NiX
pre-catalysts, which included the new crystallographically
Scheme 2 Room temperature reduction of N,N-diisopropylbenzamide
1b) employing (PN)NiX pre-catalysts, reported as % consumption 1b (%
conversion to 2b) on the basis of calibrated GC data.
(
Dalton Trans.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
characterized complexes (PN)Ni(OR) (R = 2,6-dimethylphenyl Galbraith
Laboratories
Inc.
(Knoxville,
Tennessee).
or tBu). Whereas in the solid state (PN)Ni(OtBu) exhibits a Crystallographic data were obtained at or below 193(2) K on a
three-coordinate structure that is devoid of additional coordi- Bruker PLATFORM/APEX II diffractometer equipped with a
native interactions, (PN)Ni(Odmp) was found to adopt a dis- CCD area detector, employing samples that were mounted in
torted square-planar geometry featuring agostic interaction(s) inert oil and transferred to a cold gas stream on the diffract-
involving a methyl group on the aryloxo ligand. The three-coor- ometer. Unit cell parameters were determined and refined on
dinate (PN)Ni(OtBu) complex proved to be particularly all reflections. Data reduction, correction for Lorentz polariz-
effective in such amide reduction chemistry, including room ation, and absorption correction were each performed.
2
temperature transformations involving N,N-diisopropyl- Structure solution and least-squares refinement on F were
benzamide and caprolactam. Future work will involve probing used throughout. All non-hydrogen atoms were refined with
further the mechanistic details of these transformations.
anisotropic displacement parameters. Full crystallographic
solution and refinement details are provided in the deposited
CIFs (CCDC 1856638 and 1856639†).
Experimental section
Synthesis of (PN)Ni(Odmp)
General considerations
A vial equipped with a magnetic stir bar was charged with a
Unless otherwise indicated, all experimental procedures were solution of (PN)Ni(NHdipp) (0.200 g, 0.30 mmol) in benzene
conducted in a nitrogen-filled, inert-atmosphere glovebox (ca. 5 mL), and stirring was initiated. A solution of 2,6-di-
using oven-dried glassware and purified solvents. The follow- methylphenol (0.037 g, 0.30 mmol) in benzene (ca. 2 mL) was
ing solvent purification methods were used: benzene, toluene, added to the stirring solution, which was allowed to stir at
and pentane were deoxygenated by sparging with nitrogen gas room temperature for 18 h. The volatile components of the
followed by passage through a double column solvent purifi- reaction mixture were then removed in vacuo. The remaining
cation system packed with alumina and copper-Q5 reactant residue was first triturated with pentane (3 × 1 mL), then
and storage over activated 4 Å molecular sieves; tetrahydro- washed with 3 × 0.5 mL of cold (−35 °C) pentane to remove
furan was dried over Na/benzophenone followed by distillation residual 2,6-diisopropylaniline. The remaining residue was dis-
under an atmosphere of nitrogen gas. Solvents were stored solved in Et O (ca. 8 mL) and filtered through Celite to remove
2
over activated 4 Å molecular sieves. Phenylsilane and benzene- any insoluble side-products. Solvent removal in vacuo from the
d
6
were degassed via three freeze–pump–thaw cycles and collected eluent, followed by fractional recrystallization in two
stored over 4 Å molecular sieves. N,N-Dibenzylbenzamide and crops of the crude material from a concentrated Et O/pentane
2
N,N-diisopropylbenzamide were prepared as described pre- (1 : 1) solution at −35 °C afforded (PN)Ni(Odmp) (0.082 g,
2
0
49 2
viously, and the latter was further purified by recrystalliza- 45%) as dark orange crystals. Anal. calcd For C35H N NiOP:
2
0
18b
1
tion from THF at −35 °C. (PN)M(N(SiMe ) ) (M = Mn, Fe,
C, 69.66; H, 8.18; N, 4.64. Found: C, 69.71; H, 7.92; N, 4.73. H
and M NMR (500.1 MHz, benzene-d ): δ 7.34 (m, 2H, Harom), 7.08 (m,
3 2 2
) ) (M = Mn, Fe, and Co) were prepared according 1H, Harom), 6.91–6.88 (4H, Harom), 6.83–6.75 (3H, Harom), 6.57
3
2
1
8b
18f
25
18f
Co,
N(SiMe
to literature procedures. All other reagents were purchased (apparent t, 1H, J = 7 Hz, Harom), 4.11 (apparent septet, 2H,
and Ni ), ((PN)NiCl)
2
,
(PN)Ni(NHdipp),
6
2
6
(
3
HH
3
from commercial suppliers and were used without further
2
JHH = 7 Hz, CHMe ), 2.30 (s, 3H, OCCMe), 2.29 (s, 3H,
1
13
31
3
t
3
2
purification. Unless otherwise stated, H, C, and P NMR OCCMe), 1.69 (d, 18H, JPH = 15 Hz, P Bu ), 1.49 (d, 6H, JHH =
3
13
1
characterization data were collected at 300 K on a Bruker 7 Hz, CHMe ), 0.82 (d, 6H, J = 7 Hz, CHMe2). C{ H} NMR
2
HH
AV-500 spectrometer operating at 500.1, 125.7, and 202.5 MHz (125.7 MHz, benzene-d
6
): δ 175.7 (NCN), 165.6 (Carom), 145.1
(
respectively) with chemical shifts reported in parts per (Carom), 134.4 (Carom), 130.8 (CHarom), 129.3 (CHarom), 127.9
1
13
million downfield of SiMe for H and C, and 85% H PO in (CHarom), 127.5 (CHarom), 126.9 (CHarom), 124.8 (CHarom), 124.0
4
3
4
3
1
1
13
1
D
2
O for P. H and C NMR chemical shift assignments are (Carom), 115.1 (CHarom), 39.8 (d,
J
CP = 23 Hz, PCMe
CP = 2 Hz, PCMe ), 24.8 (CHMe
H– C HSQC, and H– C HMBC NMR experiments. Gas (CHMe2), 18.7 (OCCMe2), 18.6 (OCCMe2).
chromatography (GC) data were obtained on a Shimadzu (202.5 MHz, benzene-d ): δ 121.2. Crystals of (PN)Ni(Odmp)
3
), 29.0
), 23.7
P{ H} NMR
1
3
1
1
3
based on data obtained from C-DEPTQ135, H– H COSY, (CHMe
2
), 28.1 (d,
J
3
2
1
13
1
13
31
1
6
instrument equipped with a SGE BP-5 column (30 m, 0.25 mm suitable for X-ray diffraction were grown from a concentrated
i.d.), whereby response-factor calibration was carried out using pentane solution at −35 °C.
authentic samples relative to an internal standard. HPLC-MS
Synthesis of (PN)Ni(OtBu)
data were collected on an Agilent LC-MSD system. The column
used was Agilent Poroshell 120 EC-C-18 (3 × 150 mm, 2.7 µm) A vial equipped with a magnetic stir bar was charged with
−1
and the flow rate was 2 ml min . For the gradient, solvent A NaOtBu (0.059 g, 0.61 mmol). ((PN)NiCl)
was H O with 0.1% formic acid, solvent B was acetonitrile with was added to the vial as a solution in THF (ca. 5 mL). The reac-
.1% formic acid. Gradient started with 90% solvent A (10% tion mixture was stirred at room temperature for 18 h, after
2
(0.300 g, 0.29 mmol)
2
0
solvent B) and in 15 minutes finished with 100% solvent which a color change from dark red-purple to dark green/
B. The mass spectrometer scanned from 100–1000 m/z in posi- brown was observed. Following removal of solvent in vacuo, the
tive ESI mode. Elemental analysis data were obtained from remaining dark green residue was extracted into pentane (ca.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans.

View Article Online
Paper
Dalton Transactions
1
0 mL), and filtered through a Celite pad in a glass-fritted
Acknowledgements
filter funnel. Solvent was removed in vacuo from the collected
eluent, to obtain a dark green powder (0.288 g, 89% crude We thank Chevron Phillips Chemical Company LP for support-
yield). This material was recrystallized from a concentrated ing this research and for permission to publish. NSERC of
solution of pentane at −35 °C to obtain dark green crystals Canada (including a PGS-D for C. M. M.) and Dalhousie
(
8
0.160 g, 50%). Anal. calcd for C31
: 89; N, 5.04. Found: C, 66.89; H, 8.71; N, 4.86. H NMR
49 2
H N NiOP: C, 67.04; H, University are acknowledged for their support of this work.
1
(
(
500.1 MHz, benzene-d ): δ 7.40–7.37 (m, 2H, Harom), 7.13–7.08
m, 1H, Harom), 6.95–6.93 (m, 2H, Harom), 6.86–6.76 (overlap-
6
References
ping resonances, 3H, Harom), 4.16–4.02 (apparent septet, 2H,
3
3
JHH = 7 Hz, CHMe ), 1.80 (d, 6H, J = 7 Hz, CHMe ), 1.67 (d,
1 S. A. Lawrence, Amines: Synthesis, Properties and
Applications, Cambridge University Press, Cambridge, 2006.
2 (a) J. Magano and J. R. Dunetz, Org. Process Res. Dev., 2012,
16, 1156–1184; (b) J. Blanchet, A. Chardon, E. Morisset and
J. Rouden, Synthesis, 2018, 984–997.
3 J. Pesti and G. L. Larson, Org. Process Res. Dev., 2016, 20,
1164–1181.
4 B. Li, J. B. Sortais and C. Darcel, RSC Adv., 2016, 6, 57603–
57625.
2
HH
2
3
t
t
3
1
7
(
(
(
(
(
d
8H, JPH = 15 Hz, P Bu
2
), 1.23 (s, 9H, O Bu), 0.92 (d, 6H, JHH
). C{ H} NMR (125.7 MHz, benzene-d ): δ 175.2
=
1
3
1
Hz, CHMe
2
6
NCN), 146.2 (Carom), 140.7 (Carom), 134.1 (Carom), 130.5
CHarom), 129.2 (CHarom), 127.5 (CHarom), 126.9 (CHarom), 124.6
1
CHarom), 70.1 (OCMe
3
), 38.1 (d,
J
PC = 27 Hz, PCMe
3
), 35.1
2
OCMe ), 29.0 (CHMe ), 28.1 (d, JPC = 2 Hz, PCMe ), 25.2
3 2 3
3
1
1
CHMe
2
), 23.1 (CHMe
2
). P{ H} NMR (202.5 MHz, benzene-
6
): δ 130.4. Crystals suitable for X-ray diffraction were grown
from a concentrated solution of pentane at −35 °C.
5 Selected representative examples and reviews: For repre-
sentative examples and reviews, see: (a) Y. H. Li,
J. A. M. de la Torre, K. Grabow, U. Bentrup, K. Junge,
S. L. Zhou, A. Bruckner and M. Beller, Angew. Chem., Int.
Ed., 2013, 52, 11577–11580; (b) E. Blondiaux and
T. Cantat, Chem. Commun., 2014, 50, 9349–9352;
General catalytic procedure
A 1 dram vial was charged with reaction components in the
following order: amide (1a or 1b, 0.2 mmol), solvent (as
needed such that the total volume of added liquids/solutions
was 500 μL, unless otherwise stated), pre-catalyst (as a stock
solution), and phenylsilane (0.4 mmol). For the reduction of
N-benzylbenzamide or caprolactam, the reaction com-
ponents were added in the following order: amide, solvent,
phenylsilane, then pre-catalyst, with a total volume of
(
2
c) M. Oestreich, J. Hermeke and J. Mohr, Chem. Soc. Rev.,
015, 44, 2202–2220; (d) A. Chardon, T. M. El Dine,
R. Legay, M. De Paolis, J. Rouden and J. Blanchet, Chem. –
Eur. J., 2017, 23, 2005–2009; (e) For a noteworthy recent
example of chemoselective amide reductions enabled by
use of boron Lewis acid catalysis, see: M. T. Peruzzi,
Q. Q. Mei, S. J. Lee and M. R. Gagne, Chem. Commun.,
2
50 μL. Liquids/solutions were dispensed by micropipette.
2
018, 54, 5855–5858.
The vial was then sealed with a cap featuring a PTFE septum.
For reactions requiring temperatures above 25 °C, the vials
were also sealed with Teflon tape and electrical tape, and
then taken out of the glovebox and heated in an aluminum
heating block set to the desired temperature. In some cases,
vigorous bubbling of the reaction mixture was observed.
Upon completion, the reaction vial was opened in air,
dodecane was added as an internal standard for GC analysis,
6
Selected representative examples and reviews: (a) S. Das,
S. L. Zhou, D. Addis, S. Enthaler, K. Junge and M. Beller,
Top. Catal., 2010, 53, 979–984; (b) S. Das, D. Addis, S. Zhou,
K. Junge and M. Beller, J. Am. Chem. Soc., 2010, 132, 1770–
1
771; (c) S. Das, D. Addis, S. L. Zhou, K. Junge and
M. Beller, J. Am. Chem. Soc., 2010, 132, 4971–4971;
d) S. Das, D. Addis, K. Junge and M. Beller, Chem. – Eur. J.,
011, 17, 12186–12192; (e) O. O. Kovalenko, A. Volkov and
(
2
2
and benchtop Et O or THF (ca. 1 mL in either case) was
H. Adolfsson, Org. Lett., 2015, 17, 446–449.
S. Zhou, K. Junge, D. Addis, S. Das and M. Beller, Angew.
Chem., Int. Ed., 2009, 48, 9507–9510.
Y. Sunada, H. Kawakami, T. Imaoka, Y. Motoyama and
H. Nagashima, Angew. Chem., Int. Ed., 2009, 48, 9511–9514.
D. Bezier, G. T. Venkanna, J.-B. Sortais and C. Darcel,
ChemCatChem, 2011, 3, 1747–1750.
added. The mixture was then filtered through a short plug of
silica. The filtrate was then sampled (ca. 100 μL) and diluted
further (ca. 1 mL), and analyzed by use of gas chromato-
graphy. Conversions of amide starting material and yield of
products were determined via response-factor calibration
employing authentic samples versus a dodecane internal
standard. For the reduction of 1a, reaction product identifi-
cation, following workup, was confirmed independently by
use of HPLC-MS techniques.
7
8
9
1
1
1
1
0 A. Volkov, E. Buitrago and H. Adolfsson, Eur. J. Org. Chem.,
013, 2066–2070.
2
1 B. Blom, G. W. Tan, S. Enthaler, S. Inoue, J. D. Epping and
M. Driess, J. Am. Chem. Soc., 2013, 135, 18108–18120.
2 T. Dombray, C. Helleu, C. Darcel and J. B. Sortais, Adv.
Synth. Catal., 2013, 355, 3358–3362.
3 N. C. Mamillapalli and G. Sekar, Chem. Commun., 2014, 50,
7881–7884.
Conflicts of interest
There are no conflicts to declare.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
1
4 B. J. Simmons, M. Hoffmann, J. Hwang, M. K. Jackl and
N. K. Garg, Org. Lett., 2017, 19, 1910–1913.
5 The hydrosilative reduction of primary amides by use of 3d
(b) O. L. Sydora, M. Stradiotto, L. Turculet, B. L. Small,
R. C. Coffin and S. M. Bischof, PCT Int. Appl. US,
009120826B1, 2015.
1
transition metal catalysis is limited to single report: S. Das, 20 C. M. Kelly, R. McDonald, O. L. Sydora, M. Stradiotto and
B. Wendt, K. Moeller, K. Junge and M. Beller, Angew.
Chem., Int. Ed., 2012, 51, 1662–1666.
6 R. Arias-Ugarte, H. K. Sharma, A. L. C. Morris and
K. H. Pannell, J. Am. Chem. Soc., 2012, 134, 848–851.
7 M. Igarashi and T. Fuchikami, Tetrahedron Lett., 2001, 42,
L. Turculet, Angew. Chem., Int. Ed., 2017, 56, 15901–15904.
21 The instability of the analgous Ni complex precludes its
inclusion in this study: M. Faust, A. M. Bryan,
A. Mansikkamaki, P. Vasko, M. M. Olmstead,
H. M. Tuononen, F. Grandjean, G. J. Long and P. P. Power,
Angew. Chem., Int. Ed., 2015, 54, 12914–12917.
1
1
1
1
945–1947.
8 (a) O. L. Sydora, T. C. Jones, B. L. Small, A. J. Nett, 22 For the application of (Cp*∼NHC)Ni(OtBu) pre-catalysts in
A. A. Fischer and M. J. Carney, ACS Catal., 2012, 2, 2452–
455; (b) A. J. Ruddy, C. M. Kelly, S. M. Crawford,
C. A. Wheaton, O. L. Sydora, B. L. Small, M. Stradiotto and
the hydrosilative reduction of aldehydes and ketones, see:
L. Postigo and B. Royo, Adv. Synth. Catal., 2012, 354, 2613–
2618.
2
L. Turculet, Organometallics, 2013, 32, 5581–5588; 23 L. M. Martinez-Prieto, P. Palma, E. Alvarez and J. Cámpora,
c) A. J. Ruddy, O. L. Sydora, B. L. Small, M. Stradiotto and Inorg. Chem., 2017, 56, 13086–13099.
L. Turculet, Chem. – Eur. J., 2014, 20, 13918–13922; 24 (a) J. Campora, J. A. Lopez, C. Maya, P. Palma, E. Carmona
(
(
d) A. J. Ruddy, O. L. Sydora, B. L. Small, M. Stradiotto and
and P. Valerga, J. Organomet. Chem., 2002, 643, 331–341;
(b) A. Kruckenberg, H. Wadepohl and L. H. Gade,
Organometallics, 2013, 32, 5153–5170.
L. Turculet, Chem. – Eur. J., 2016, 22, 12589; (e) T. Ogawa,
A. J. Ruddy, O. L. Sydora, M. Stradiotto and L. Turculet,
Organometallics, 2017, 36, 417–423; (f) C. M. Macaulay, 25 C. M. Kelly, J. T. Fuller, C. M. Macaulay, R. McDonald,
S. J. Gustafson, J. T. F. III, D.-H. Kwon, T. Ogawa,
M. J. Ferguson, R. McDonald, M. D. Lumsden,
S. M. Bischof, O. L. Sydora, D. H. Ess, M. Stradiotto and
L. Turculet, ACS Catal., 2018, 8, 9907–9925.
9 (a) O. L. Sydora, M. Carney, B. L. Small, S. Hutchison and
J. C. Gee, PCT Int. Appl. WO, 082192A1, 2011;
M. J. Ferguson, S. M. Bischof, O. L. Sydora, D. H. Ess,
M. Stradiotto and L. Turculet, Angew. Chem., Int. Ed., 2017,
56, 6312–6316.
26 R. A. Andersen, K. Faegri, J. C. Green, A. Haaland,
M. F. Lappert, W. P. Leung and K. Rypdal, Inorg. Chem.,
1988, 27, 1782–1786.
1
This journal is © The Royal Society of Chemistry 2018
Dalton Trans.