Catalytic dinitrogen reduction at the molybdenum center promoted by a bulky tetradentate phosphine ligand

Article abstract of DOI:10.1002/anie.201408664

Stoichiometric reduction of N₂ at a Mo center stabilized by a bulky tetradentate phosphine ligand (PP₃^Cy) allowed isolation of Mo-imidoamine and Mo-imido complexes. Both complexes as well as the Mo^II precursor are equally suitable catalysts for the synthesis of NTMS₃ (TMS= trimethylsilyl) from N₂, TMSCl, and electron sources. Mechanistic studies prove the involvement of a TMS radical at least in one of the catalytic steps.

Full text of DOI:10.1002/anie.201408664

Angewandte
Chemie
DOI: 10.1002/anie.201408664
Dinitrogen Conversion
Catalytic Dinitrogen Reduction at the Molybdenum Center Promoted
by a Bulky Tetradentate Phosphine Ligand**
Qian Liao, Nathalie Saffon-Merceron, and Nicolas Mꢀzailles*
In memory of Pascal Le Floch
Abstract: Stoichiometric reduction of N2 at a Mo center
stabilized by a bulky tetradentate phosphine ligand (PPC3 y)
allowed isolation of Mo–imidoamine and Mo–imido com-
plexes. Both complexes as well as the MoII precursor are
equally suitable catalysts for the synthesis of NTMS3 (TMS =
trimethylsilyl) from N2, TMSCl, and electron sources. Mech-
anistic studies prove the involvement of a TMS radical at least
in one of the catalytic steps.
intermediates that may take part in the catalytic cycle (N2
to NH3) through sequences of protonation/one-electron
reduction of N2 at the metal center,[14–20] whereas Hidai and
co-workers proposed the intermediacy of SiMe3 radical.[3,21]
Nishibayashi, Yoshizawa, and co-workers have performed
a DFT study[4] which corroborates that the mechanism could
follow a radical pathway (in terms of energy requirements),
but none of the postulated intermediates could be isolated (or
observed), most likely because of the efficiency of the
catalyst. In the overall process, they propose that the final
I
n the challenging domain of N2 functionalization using
transition metal complexes,[1] catalyzed processes under
ambient conditions are very rare. A common feature with
nitrogenases is that, apart from the first example by Shiina,[2]
they rely on either Mo or Fe complexes. Indeed, in 1989,
Hidai reported the catalyzed transformation of N2 into
silylamines (N(SiMe3)3) using a Mo0 complex bearing mono-
dentate phosphines as precursor.[3] This groundbreaking
strategy was not followed for more than twenty years until
2011 when Nishibayashi, Yoshizawa, and co-workers reported
that a carefully designed diphosphine/Mo0 catalytic system
could promote the same transformation with greatly
increased efficiency (TON of ca. 200).[4,5] The same authors
reported recently the same reaction with several Fe com-
plexes, yet with lower efficiency (TON up to 34).[6] In 2003,
Schrock designed a trisamidoamine/MoIII precursor that is
able to catalytically transform N2 into NH3.[7] Nishibayashi
and co-workers reported in 2010 that a pincer PNP/Mo0
precursor could also effect this process[8–11] followed most
recently by Peters and co-workers with the use of a tetraden-
tate ligand/Fe0 precursor.[12,13] Although leading to related
compounds, NH3 and N(SiMe3)3, using a similar approach
(addition of sources of electrons and electrophiles) the
catalyzed processes are proposed to follow different mecha-
nisms. Indeed, Schrock has isolated a great number of
N N bond is cleaved once the (SiMe3)2N N(SiMe3) anion is
decoordinated from the metal center and a low energy path
À
À
À
involving necessary P Mo decoordination/coordination
(hemilabile character of the diphosphine). It is therefore of
utmost importance to obtain mechanistic information on the
nature of intermediates in this process in order to further
develop catalysts for N2 functionalization.
Of interest to us is the “(PPC3 y)Mo” fragment, which
features a neutral tetradentate phosphine, PPC3 y, that should
accommodate a mononuclear low-valent Mo center. Here we
report the syntheses of two aminoimide complexes
Cy
Cy
=
=
[(PP3 )Mo NNR2] and an imido complex [(PP3 )Mo
NTMS] by stepwise reduction of N2. We also show that
these two MoII complexes are equally suitable catalysts for the
transformation of N2 into N(SiMe3)3, and prove that the
overall catalytic transformation requires at least one step
involving a silicon-based radical.
When stoichiometric amounts of PPC3 y and [MoCl3(THF)3]
were mixed in THF, a novel paramagnetic complex was
formed, as shown by the complete disappearance of the
signals of the free ligand in the 31P spectrum. This complex is
the expected [(PPC3 y)MoCl3], 1 (Scheme 1), as shown by
elemental analysis and X-ray diffraction (see the Supporting
Information, SI). Interestingly, one of the phosphine arms is
not coordinated to the MoIII center which possesses a pseudo-
octahedral geometry. Reduction of complex 1 by one
equivalent of Na/Hg in the presence of NaBPh4 resulted in
the formation of a blue diamagnetic complex, 2. This complex
is characterized in the 31P{1H} spectrum by a quartet/doublet
at 159 and 137 ppm, respectively, pointing to a symmetrical
structure in which all three phosphine arms are bound to the
MoII center. The diamagnetism of the complex strongly
suggested an ML5 geometry and thus a cationic complex. An
X-ray structure confirmed the trigonal bipyramid (TBP)
geometry (Figure 1). Further reduction of complex 2 under N2
by two equivalents of Na/Hg resulted again in the loss of
signals in the 31P{1H} spectrum. An IR spectrum of the crude
mixture showed two different uNN stretches at 2016 and
[*] Dr. Q. Liao, Dr. N. Mꢀzailles
Laboratoire Hꢀtꢀrochimie Fondamentale et Appliquꢀe
Universitꢀ Paul Sabatier, CNRS
118 Route de Narbonne, 31062 Toulouse (France)
E-mail: mezailles@chimie.ups-tlse.fr
Dr. N. Saffon-Merceron
Institut de Chimie de Toulouse ICT-FR2599
Universitꢀ Paul Sabatier, 31062 Toulouse Cedex (France)
[**] The “Rꢀgion Midi Pyrꢀnꢀes” is gratefully acknowledged for sup-
porting this work. CNRS and the “Rꢀgion Midi-Pyrꢀnꢀes” are
gratefully acknowledged for a postdoctoral fellowship (Q. L.). N.M.
thanks the “Rꢀgion Midi-Pyrꢀnꢀes” for funding.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201408664.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
positions) is strongly affected by a change in the ligand in cis
position (upfield shift of ca. 20 ppm between 2 and 4 or 5),
whereas the chemical shift of the P (axial position) is almost
unchanged by the ligand in trans position. In complex 4, the
29Si spectrum shows a single resonance at 2.6 ppm, consistent
with the formation of the expected five-membered ring
À
“NSiCH2CH2Si” featuring two Si N bonds. The correspond-
ing spectrum for complex 5 showed a singlet at À0.3 ppm.
Crystals of complex 4 could be grown, and its structure
analysis corroborated the NMR data (Figure 1). Such a sily-
laminoimide fragment obtained by direct reduction of N2 at
the Mo center is unknown, and has rarely been isolated,
except at Fe centers.[22–24] Notable is the linear arrangement at
the proximal N atom in this structure (Mo1-N1-N2 angle of
À
177.1(3)8). The Mo N bond of 1.828(3) ꢀ lies on the high side
=
ꢀ
of the reported Mo N (between 1.64 and 1.83 ꢀ) and Mo N
(between 1.64 and 1.79 ꢀ) bonds in the CCDC database. A
related ML5 complex is calculated to be a high-energy
intermediate (square pyramid) in the system modeled by
Nishibayashi, Yoshizawa, and co-workers, leading to phos-
phine dissociation.[4] In the present case, the PPC3 y ligand
strongly stabilizes the TBP geometry.
Scheme 1. Molybdenum complexes and the amine syntheses.
Alternatively, direct syntheses of these complexes can be
realized from either the MoIII complex 1 or the MoII complex
2 with the appropriate amount of reducing agent Na/Hg under
N2 in the presence of the chlorosilane derivatives (Scheme 1).
One of the key questions in the formation of amine (either
NH3 or NTMS3) from N2 is the sequence of addition of the H
or Si atoms. In the Chatt mechanism, H atoms (proton +
electron) are sequentially added to the distal N, resulting in
Figure 1. POV-Ray plots of complexes 2 (cationic part only) and 4
(50% thermal ellipsoids are shown). Hydrogen atoms have been
omitted and only C atoms of the Cy groups have been kept for clarity
in complex 4. Selected bond lengths [ꢁ] and angles [8]: 2: Mo1–Cl1
2.3468(13), Mo1–P1 2.4201(13), Mo1–P2 2.4439(14), Mo1–P3
2.4420(14), Mo1–P4 2.4398(14); Cl1-Mo1-P1 179.47(5), P1-Mo1-P2
75.39(5), P1-Mo1-P3 75.76(5), P1-Mo1-P4 75.42(5), P2-Mo1-P3
113.66(5), P2-Mo1-P4 114.28(5), P3-Mo1-P4 113.98(5). 4: Mo1–N1
1.828(3), N1–N2 1.387(4), N2–Si1 1.756(4), N2–Si2 1.752(4), Mo1–P1
2.4548(11), N2–N1-Mo1 177.1(3); N1-N2-Si2 123.5(3), N1-N2-Si1
124.5(3), Si2-N2-Si1 111.99(19), N1-Mo1-P1 170.48(12).
ꢀ
the formation of one equivalent of NH3 and an M N
intermediate which is then further reduced. On the other
hand, following Hidaiꢁs proposal, Nishibayashi, Yoshizawa,
and co-workers calculated that a SiMe3 radical could add onto
the proximal N atom of the Mo(NN(SiMe3)2) intermediate,
À
À
leading to the corresponding Mo N(SiMe3) N(SiMe3)2 com-
plex. This latter intermediate would then liberate the anionic
tris(trimethylsilyl) hydrazide upon one-electron reduction.
Nevertheless, they were not successful in isolating or at least
observing any intermediate in this reaction. We therefore
studied the reaction of complexes 4 and 5 with two
equivalents of Na/Hg and TMSCl. A transformation of both
complexes into the same, diamagnetic purple complex 9 was
observed. This complex is characterized also by a quartet and
a doublet at 159 and 135 ppm, respectively. Here again, the
signal for the P atom trans to the N atom is only marginally
affected upon the transformation, whereas the signal for the
three equivalent phosphines of the equatorial plane are
shifted downfield by ca. 20 ppm. The concomitant formation
of the silylamines 6 and 7 was proven by 29Si NMR spectros-
copy as well as GC-MS. Together this data suggested complex
9 to be a Mo–imido complex. Alternatively, complex 9 was
1945 cmÀ1. The synthesis of the plausible octahedral complex
[(PPC3 y)Mo(N2)2] can be ruled out as it would be diamagnetic.
It thus appears that the PP3Cy ligand strongly stabilizes ML5
geometries at the metal center as initially envisioned to form
[(PPC3 y)Mo(N2)] complexes (mixture of TBP and square-
based pyramid, both paramagnetic d4 complexes). Unexpect-
edly, it further appeared that the dinitrogen is only weakly
bound to the Mo0 center, as the mixture evolved when
submitted to vacuum (loss of uNN stretch by IR). Nevertheless,
this complex is sufficiently stable in solution and reaction with
a slight excess of Na/Hg in the presence of 1,2-bis(chlorodi-
methylsilyl)ethane or two equivalents of SiMe3Cl under 1 atm
N2 lead to the formation of the dark pink aminoimide
complexes 4 and 5, respectively. Both complexes are dia-
magnetic, as shown by the quartet/doublet pattern in the
31P{1H} NMR spectrum, at 158 and 118 ppm for complex 4 and
at 157 and 116 ppm for complex 5, again suggesting ML5
geometry at the MoII centers. Interestingly, in these ML5
complexes, the chemical shift of the PCy2 moieties (equatorial
also obtained cleanly from the stoichiometric reaction
0
À
between the non-isolated Mo N2 intermediate presented
above and TMSN3. An X-ray structure analysis confirmed the
postulated structure (Figure 2), which has a linear arrange-
ment at the N atom (Mo-N-Si angle of 178.3(3)8) and
À
possesses a C3 axis that runs through the Si Mo axis.
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
upon heating at 608C for 48 h, only TMSCl was observed in
the reaction mixture. This experiment clearly proved that
under these conditions the TMS radical is not formed. In
a second stage, the Mo-catalyzed transformation of N2 into
N(TMS)3 was studied (Table 1). The formation of N(TMS)3 as
Table 1: Catalytic investigations.
[a]
Entry Reductant Additive Catalyst
T
NTMS3
Yield of
NTMS3
[b]
[8C] [equiv]
Figure 2. POV-Ray plots of complexes 9 and 10 (50% thermal ellip-
soids are shown). Hydrogen atoms have been omitted and only iPr
fragments of the Cy have been kept for clarity. Selected bond lengths
[ꢁ] and angles [8]: 9: Mo–N 1.798(5), N–Si 1.740(5), Mo–P1
2.4319(18), Mo–P2 2.4822(17), Mo–P3 2.4170(18), Mo–P4 2.4271(18);
N-Mo-P2 179.59(18), Mo-N-Si 178.3(3), P1-Mo-P2 75.45(6), P1-Mo-P3
113.03(7), P1-Mo-P4 114.85(6), P2-Mo-P3 75.62(6), P2-Mo-P4
74.76(6), P3-Mo-P4 113.43(6). 10: Mo–N 1.839(2), N–Si 1.718(2), Mo–
P1 2.4911(7), Mo–P2 2.4281(6), Mo–P3 2.4342(6), Mo–P4 2.4221(7);
N-Mo-P1 178.06(7), Mo-N-Si 177.01(15), P1-Mo-P2 74.33(2), P1-Mo-P3
75.02(2), P1-Mo-P4 75.71(2), P2-Mo-P3 114.82(2), P2-Mo-P4
113.41(2), P3-Mo-P4 112.44(2).
1
2
3
4
5
6
7
8
Na sand
Na sand
Na/Hg
Na/Hg
Na/Hg
Na/K
Na/K
K
K
K
K
K
K
K
K
–
NaI
–
–
–
–
NaI
–
–
–
–
–
NaI
–
–
5
5
5
5
9
5
5
5
5
5
5
5
5
9
2
25
50
25
50
50
50
90
25
25
25
50
90
50
50
50
5.8
2.8
2.1
2.4
1.4
11.7
10.2
10.5
11.4
4.4
15.0
8.3
13.3
11.8
11.7
8.7
4.2
3.2
3.6
2.1
17.6
15.3
15.8
17.1
6.6
22.5
12.5
20.0
17.7
17.6
9[c]
10[d]
11
12
13
14
15
À
The Mo N bond of 1.798(5) ꢀ is shorter than in complex
=
ꢀ
4 but still lies on the high side of the reported Mo N and Mo
N bonds. A related complex 10 was synthesized in a one-pot
procedure from complex 2 by subsequent reductions in the
presence of TMSCl (to form 5 in situ) and then PhSiMe2Cl.
The corresponding amine 8 was observed by GC-MS analysis.
Complex 10, characterized also by a quartet and a doublet at
158 and 133 ppm, respectively, in the 31P{1H} NMR spectrum,
was crystallized (Figure 2). The structure presents very similar
features to that of 9, such as the linear arrangement of Mo-N-
[a] Equivalents of NTMS3 =equiv amine/cat. [b] Yield of NTMS3 based
on initial TMSCl. [c] The reaction time was 4 days. [d] TMSBr was used
instead of TMSCl.
the sole amine was verified by 29Si NMR spectroscopy, and
quantified by the indophenol method.[25,26] Complex 5 was
used to optimize the experimental conditions. The nature of
the reducing agent was probed at 258C (entries 1, 3, and 8),
showing that Na/Hg did not allow a catalytic reaction to occur
(2.1 equiv of NTMS3: the amount of NH3 measured by the
indophenol method results from the hydrolysis of complex 5,
which already contains two N atoms). This was corroborated
by the reaction with complex 9, which only provided 1.4 equiv
of NTMS3 (entry 5). Potassium provided the best results
(10.5 equiv of NTMS3, entry 8) of the reducing agents. When
the reaction was performed for a longer time (4 days, entry 9),
the outcome was very similar (11.4 versus 10.5 equiv NTMS3),
which indicated catalyst decomposition after 2 days. Most
importantly, these experiments together with the above-
mentioned catalyst-free reactions prove the involvement of
a TMS radical in the catalytic process. As the reactions
appeared quite slow, the effect of temperature (entries 8, 11,
and 12) as well as the nature of the Si derivative were probed
(entries 2, 7, 10, and 13). The best results were obtained at
508C [15.0 equiv of NTMS3, entry 11 versus 10.5 equiv
(entry 8) and 8.3 equiv (entry 12) at 258C and 908C, respec-
tively]. In contrast to Hidaiꢁs observation, the addition of
neither TMSBr nor TMSI (generated in situ by the reaction
between TMSCl and NaI) improved the catalysis. The results
with TMSI were at best similar to those with TMSCl (13.3
versus 15.0 equiv NTMS3, entry 13 versus 11, respectively).
Finally, with the optimized conditions (508C, TMSCl, and K),
the three complexes 2, 5, and 9 could be compared, and
appeared to be equally suitable for the catalytic transforma-
tion of N2 to N(TMS)3.
=
Si and the TBP geometry, yet with a slightly elongated Mo N
bond (1.839(2) ꢀ in 10 versus 1.798(5) ꢀ in 9). These
complexes are the first examples, in which the Mo–imido
bond is generated through N2 functionalization. It also proves
that the chosen ligand, despite its bulkiness, favors amine
À
elimination through in-sphere N N splitting rather than
hydrazine derivative elimination.
With a series of Mo complexes in hand that are able to
transform N2 into N(SiMe3)3 or N(SiMe3)(SiMe2CH2-
CH2SiMe2) by the stoichiometric reduction/functionalization
of N2, their performance in the catalytic formation of
N(SiMe3)3 from N2, TMSX (X = Cl, Br, and I), and various
sources of reducing agents was studied. Most importantly,
their relative efficiency in the catalytic process will provide
information on their involvement in the catalytic cycle versus
being dead-end species. In a first series of experiments,
“blank” reactions (without Mo complex) were carried out.
These were crucial to prove the involvement of Si-centered
radicals during the process. Indeed, when TMSCl was reacted
with K or NaK in THF at room temperature, TMS–TMS was
formed as expected from the efficient and fast generation of
SiMe3 radical. When the reaction was carried out with Na,
Me3Si(CH2)4OSiMe3 was formed as a result of the reaction of
the radical with THF, in addition to TMS–TMS as previously
observed by Nishibayashi, Yoshizawa, and co-workers.[4]
On the other hand, although the redox potential of Na/Hg
is low enough to allow the effective reduction of TMSCl, the
kinetics of this reduction appeared very slow. Indeed, even
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
volume was reduced to 0.1 mL, and crystals were obtained by cooling
the concentrated solution to À808C for 2 days. The mother liquor was
filtered and the crystals were collected and dried. Yield of the isolated
product: 27 mg, 55%. 1H NMR (300 MHz, [D8]toluene): d = 2.19–
2.15 (m, 6H, CH2arm), 1.95–1.91 (m, 6H, CH2arm), 1.83–1.77 (m, 12H,
CyH), 1.68–1.45 (m, 24H, CyH), 1.34–1.10 (m, 30H, CyH), 0.81 (s,
4H, SiCH2CH2Si), 0.37 ppm (s, 12H, Si(CH3)2); 31P{H} NMR
(121.5 MHz, [D8]toluene): d = 158 (q, 2J = 28 Hz, 1P), 118 ppm (d,
2J = 28 Hz, 3P); 29Si{H} NMR (59.6 MHz, THF): d = 2.6 ppm.
Indeed, they provided very similar results (11.7, 15.0, and
11.8 equiv NTMS3, entries 15, 11, and 14, respectively). Thus,
not only is complex 2 a competent precatalyst, as expected
from the stoichiometric reactions presented above, but also
both complexes 5 and 9 are intermediates in the catalytic
cycle. Overall, the stoichiometric and catalytic reactions
between N2, TMSCl, reducing agents, and the three com-
plexes 2, 5, and 9 support the catalytic cycle presented in
Scheme 2, involving a TMS radical at least in the step
between complex 9 and the “Mo(N2)” intermediate.
Received: August 29, 2014
Published online: && &&, &&&&
Keywords: amine formation · dinitrogen conversion ·
.
dinitrogen reduction · Mo complexes · silicon radicals
[1] For reviews, see: a) Y. Tanabe, Y. Nishibayashi, Coord. Chem.
Rev. 2013, 257, 2551; b) K. C. MacLeod, P. L. Holland, Nat.
Chem. 2013, 5, 559; c) Y. Nishibayashi, Dalton Trans. 2012, 41,
7447; d) N. Hazari, Chem. Soc. Rev. 2010, 39, 4044; e) J. L.
Crossland, D. R. Tyler, Coord. Chem. Rev. 2010, 254, 1883; f) M.
Mori, Heterocycles 2009, 78, 281; g) M. Mori, J. Organomet.
Chem. 2004, 689, 4210; h) P. J. Chirik, Dalton Trans. 2007, 16;
i) M. Hidai, Y. Mizobe, Pure Appl. Chem. 2001, 73, 261; j) M.
Hidai, Y. Ishii, J. Mol. Catal. A 1996, 107, 105; k) S. Gambarotta,
J. Organomet. Chem. 1995, 500, 117; l) T. A. Bazhenova, A. E.
Shilov, Coord. Chem. Rev. 1995, 144, 69.
[2] K. Shiina, J. Am. Chem. Soc. 1972, 94, 9266.
[3] K. Komori, H. Oshita, Y. Mizobe, M. Hidai, J. Am. Chem. Soc.
1989, 111, 1939.
[4] H. Tanaka, A. Sasada, T. Kouno, M. Yuki, Y. Miyake, H.
Nakanishi, Y. Nishibayashi, K. Yoshizawa, J. Am. Chem. Soc.
2011, 133, 3498.
[5] Very low TON of silylamine from N2 was recently reported: T.
Ogawa, Y. Kajita, Y. Wasada-Tsutsui, H. Wasada, H. Masuda,
Inorg. Chem. 2013, 52, 182.
[6] M. Yuki, H. Tanaka, K. Sasaki, Y. Miyake, K. Yoshizawa, Y.
Nishibayashi, Nat. Commun. 2012, 3, 2261.
Scheme 2. Operative mechanism for the catalytic transformation of N2
to N(TMS)3 using the “(PPC3y)Mo” fragment.
[7] D. V. Yandulov, R. R. Schrock, Science 2003, 301, 76.
[8] K. Arashiba, Y. Miyake, Y. Nishibayashi, Nat. Chem. 2011, 3,
120.
[9] E. Kinoshita, K. Arashiba, S. Kuriyama, Y. Miyake, R. Shima-
zaki, H. Nakanishi, Y. Nishibayashi, Organometallics 2012, 31,
8437.
[10] S. Kuriyama, K. Arashiba, K. Nakajima, H. Tanaka, N. Kamaru,
K. Yoshizawa, Y. Nishibayashi, J. Am. Chem. Soc. 2014, 136,
9719.
[11] H. Tanaka, K. Arashiba, S. Kuriyama, K. Nakajima, Y.
Nishibayashi, A. Sasada, K. Yoshizawa, Nat. Commun. 2014, 5,
3737.
In conclusion, we show here that a tetradentate phosphine
ligand “PPC3 y”, featuring strongly donating moieties favors the
TBP geometry for the MoII center. This enforced geometry
allowed us to isolate three intermediates in the stepwise
reduction of N2: two imidoamine complexes, 4 and 5, and an
imido complex, 9. The latter imido complex 9 as well as
=
complex 10 are the first examples of a M NTMS fragment
generated upon N2 reduction at a metal center. Finally,
catalytic investigations with various sources of reducing agent
prove the involvement of a Si-centered radical in the N2 to
=
[12] J. S. Anderson, J. Rittle, J. C. Peters, Nature 2013, 501, 84.
[13] S. E. Creutz, J. C. Peters, J. Am. Chem. Soc. 2014, 136, 1105.
[14] D. V. Yandulov, R. R. Schrock, A. L. Rheingold, C. Ceccarelli,
W. M. Davis, Inorg. Chem. 2003, 42, 796.
NTMS3 reduction process, at least in the Mo NTMS to
Mo(N2) step.
[15] R. R. Schrock, Angew. Chem. Int. Ed. 2008, 47, 5512; Angew.
Chem. 2008, 120, 5594.
[16] R. R. Schrock, Acc. Chem. Res. 2005, 38, 955.
[17] D. V. Yandulov, R. R. Schrock, J. Am. Chem. Soc. 2002, 124,
6252.
[18] D. V. Yandulov, R. R. Schrock, Inorg. Chem. 2005, 44, 1103.
[19] V. Ritleng, D. V. Yandulov, W. W. Weare, R. R. Schrock, A. S.
Hock, W. M. Davis, J. Am. Chem. Soc. 2004, 126, 6150.
[20] M. R. Reithofer, R. R. Schrock, P. Mꢂller, J. Am. Chem. Soc.
2010, 132, 8349.
Experimental Section
Synthesis of [(PPC3 y)Mo(NN(SiMe2CH2)2)] (4) as a representative
example: To a suspension of [Mo(PPC3 y)Cl]BPh4 (0.05 mmol, prepared
from [Mo(PPC3 y)Cl3]) in THF (2.5 mL) was added freshly prepared
sodium amalgam (5.8 mg, 0.25 mmol Na; 600 mg Hg) and bis(chlor-
odimethylsilyl)ethane (12 mg, 0.055 mmol). The mixture was stirred
vigorously for 12 h under N2, forming a dark red-purple solution. A
31P{H} NMR spectrum of the crude reaction mixture was recorded.
The NMR yield (70%) was measured by integration versus an
internal standard (PPh3) in a capillary. THF was then removed under
vacuum, and the residue was extracted with pentane ( ꢁ 1 mL). The
[21] M. Hidai, Y. Mizobe, Chem. Rev. 1995, 95, 1115.
[22] D. L. M. Suess, J. C. Peters, J. Am. Chem. Soc. 2013, 135, 4938.
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[23] M. Marc-Etienne, J. C. Peters, J. Am. Chem. Soc. 2011, 133,
18118.
[25] A. L. Chaney, E. P. Marbach, Clin. Chem. 1962, 8, 130.
[26] N. M. Tzollas, G. A. Zachariadis, A. N. Anthemidis, J. A. Stratis,
[24] P. A. Rudd, N. Planas, E. Bill, L. Gagliardi, C. C. Lu, Eur. J.
Inorg. Chem. 2013, 3898.
Int. J. Environ. Anal. Chem. 2010, 90, 115.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Dinitrogen Conversion
Stoichiometric N2 reduction at a Mo
center stabilized by bulky tetradentate
phosphine ligand allowed the isolation of
Mo–aminoimide and Mo–imido com-
plexes. Both are suitable catalysts for the
synthesis of NTMS3 from N2 and TMSCl
through a mechanism involving a TMS
radical.
Q. Liao, N. Saffon-Merceron,
N. Mꢁzailles*
&&&& — &&&&
Catalytic Dinitrogen Reduction at the
Molybdenum Center Promoted by a Bulky
Tetradentate Phosphine Ligand
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!