Metal-Mediated Production of Isocyanates, R3EN=C=O from Dinitrogen, Carbon Dioxide, and R3ECl

Article abstract of DOI:10.1002/anie.201502293

A highly efficient and versatile chemical cycle has been developed for the production of isocyanates through the molecular fixation of N₂, CO₂ and R₃ECl (E=C, Si, and Ge). Key steps include a 'one-pot' photolytic N-N bond cleavage of a Group6 dinuclear dinitrogen complex with insitu trapping by R₃ECl to provide a metal terminal imido complex that can engage in simultaneous nitrene-group transfer and oxygen-atom transfer to generate an intermediate metal terminal oxo complex with release of the isocyanate product. Reaction of the oxo complex with additional equivalents of R₃ECl regenerates a metal dichloride that is the precursor for dinuclear dinitrogen starting material.

Full text of DOI:10.1002/anie.201502293

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201502293
German Edition: DOI: 10.1002/ange.201502293
Nitrogen Fixation
Hot Paper
= =
Metal-Mediated Production of Isocyanates, R3EN C O from
Dinitrogen, Carbon Dioxide, and R3ECl**
Andrew J. Keane, Wesley S. Farrell, Brendan L. Yonke, Peter Y. Zavalij, and Lawrence R. Sita*
Abstract: A highly efficient and versatile chemical cycle has
been developed for the production of isocyanates through the
molecular fixation of N2, CO2 and R3ECl (E = C, Si, and Ge).
À
Key steps include a ꢀone-potꢁ photolytic N N bond cleavage of
a Group 6 dinuclear dinitrogen complex with in situ trapping
by R3ECl to provide a metal terminal imido complex that can
engage in simultaneous nitrene-group transfer and oxygen-
atom transfer to generate an intermediate metal terminal oxo
complex with release of the isocyanate product. Reaction of the
oxo complex with additional equivalents of R3ECl regenerates
a metal dichloride that is the precursor for dinuclear dinitrogen
starting material.
O
ne of the ultimate objectives of exploring metal-mediated
nitrogen (N2) fixation is to develop more energy-efficient and
atom-economical routes to high-value nitrogen-based chem-
icals in a manner that generates little or no waste by-
products.[1] In this regard, heterocumulenes, comprising iso-
= =
= =
cyanates (RN C O) and carbodiimides (RN C NR’), are
ideal synthetic targets since the overall free-energy penalty
associated with breaking the exceptionally strong triple bond
of N2 can be partially offset through the compensating
Scheme 1. Complete N2 fixation cycle for production of Me3SiNCO.
See text for descriptions of Steps A–D.
À
formation of N C double bonds within these products. To
date, some progress has been made towards the development
of N2 fixation cycles that generate small molecules containing
À
Me3SiNCO generation through simultaneous nitrene group
transfer (NGT) and oxygen-atom transfer (OAT) from
carbon dioxide (CO2) to produce an intermediate metal
terminal oxide (3!4), and Step D) regeneration of the
starting metal dichloride through OAT with concomitant
formation of Me3SiOSiMe3 (4!1).
N C single and multiple bonds; however, the challenge of
coupling high chemical efficiency with synthetic versatility
has not yet been demonstrated.[2–6] Herein, we report
a versatile and atom-economical chemical cycle that formally
converts N2, carbon dioxide (CO2), and a Group 14 element
alkyl- or aryl-substituted chloride, R3ECl, into a set of
For Step A in Scheme 1, we have previously reported that
chemical reduction of the Group 6 MIV dichlorides, [Cp*M{N-
(iPr)C(Me)N(iPr)}Cl2] (Cp* = h5-C5Me5), 1a (M = Mo) and
1b (M = W), using sodium amalgam (0.5% NaHg) provides
excellent yields of the corresponding diamagnetic dinuclear
= =
isocyanate derivatives, R3EN C O, for E = C, Si, and Ge.
Scheme 1 provides a summary of the new N2 fixation
cycle, exemplified using Me3SiCl, that comprises four sepa-
rate chemical processes involving: Step A) dinuclear coordi-
nation of N2 through reduction of a metal dichloride (1!2),
Step B) photolytic N ꢀ N bond cleavage and N-atom func-
tionalization to form a terminal metal imido (2!3), Step C)
ꢀend-on-bridgedꢁ
dinitrogen
complexes,
[{Cp*M[N-
(iPr)C(Me)N(iPr)]}2(m-h1:h1-N2)] 2a (M = Mo) and 2b (M =
W), respectively.[7] Importantly, while both 2a and 2b were
determined to be thermally robust in hydrocarbon solution up
to temperatures of at least 1008C, during the course of
subsequent investigations, it was discovered that both of these
compounds are light sensitive in hydrocarbon (e.g. benzene or
methylcyclohexane) solution. Following this lead, preliminary
photolysis experiments were conducted with a [D6]benzene
solution of 2 that was sealed under Ar within a Pyrex J-
Young-NMR tube and irradiated using a Rayonet carousel of
[*] A. J. Keane, W. S. Farrell, B. L. Yonke, P. Y. Zavalij, Prof. L. R. Sita
Department of Chemistry and Biochemistry
University of Maryland
College Park, MD 20742 (USA)
E-mail: lsita@umd.edu
[**] Funding for this work was partially provided by the Department of
Energy, Basic Energy Sciences (DE-SC0002217) and the National
Science Foundation (CHE-1361716). E=C, Si, and Ge.
1
medium-pressure Hg lamps.[8] A series of H NMR spectra
taken at timed intervals then revealed the complete disap-
pearance of 2a after 48 h of irradiation with concomitant
formation of new paramagnetic species. In the case of 2b,
a similar set of 1H NMR spectra showed that photoconversion
Supporting information (complete experimental details) for this
article is available on the WWW under http://dx.doi.org/10.1002/
anie.201502293.
10220
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10220 –10224

Angewandte
Chemie
to a single new diamagnetic product occurred quite cleanly up
to about 50% conversion. However, the much longer time
required for complete conversion of 2b (ca. 72 h) resulted in
a subsequent small degree of photodegradation of this initial
product. Following the course of photolysis of a solution of 2a
in methylcyclohexane (c = 0.86 mm) by UV/Vis spectroscopy
did not produce a well-defined isosbestic point from an
overlay of electronic spectra taken at timed intervals, thereby
indicating formation of more than one photoproduct. On the
other hand, a similar UV/Vis spectroscopic investigation of
the photolysis of 2b under identical conditions now produced
well-resolved isosbestic points at 430, 583, and 761 nm after
a relatively short period of time (e.g. < 9 h). However,
Scheme 3. Alternative synthesis of 5 and 7.
trimethylsiloxy complexes, [Cp*M{N(iPr)C(Me)N(iPr)}(N)-
(OSiMe3)] (M = Mo and W).[8,11]
1
consistent with the photodegradation observed by H NMR
spectroscopy, these isosbestic points became less resolved
As presented in Scheme 3, chemical reduction of 8a and
8b in diethyl ether (Et2O) solution using a slight excess of
0.5% NaHg provided good yields of the dinuclear bis(m-
nitrido) complexes 5 and 7 as analytically pure crystalline
materials. As Figure 1 presents, single-crystal X-ray analyses
upon continued irradiation of 2b.[8]
Fortunately, large-scale photolysis of 2a provided crystal-
line material directly from the reaction mixture on two
different occasions, and when subjected to single-crystal X-
ray analyses, two different photoproducts were identified,
with these being the formal MoV (d1), MoV (d1) bis(m-nitrido)
complex, [{Cp*Mo[N(iPr)C(Me)N(iPr)](N)}2] (5), and the
formal MoIII (d3), MoIV (d2) mono(m-nitrido) complex,
[{Cp*Mo[N(iPr)C(Me)N(iPr)]}2(N)] (6), that are shown in
Scheme 2.[8] The occurrence of both 5 and 6 as photolysis
Scheme 2. Photolysis of compounds 2a and 2b.
products is consistent with reported observations of compet-
ing N ꢀ N bond cleavage and N2 extrusion reaction pathways
in the photolysis of dimolybdenum m-N2 complexes.[9,10]
Finally, although crystalline material could not be isolated
in the case of large-scale photolysis of 2b, the structure of the
Figure 1. Molecular structures (thermal ellipsoids set at 30% proba-
bility) of a) 5 and b) 7. Hydrogen atoms have been removed for the
sake of clarity. Selected geometric parameters for the M2N2 cores of 5
and 7 are provided in (c) and (d) respectively.
1
new diamagnetic photoproduct observed by H NMR spec-
troscopy was tentatively assigned as being the corresponding
dinuclear bis(m-nitrido) complex [{Cp*W[N(iPr)C(Me)N-
(iPr)](N)}2] (7).
To have adequate supplies of 5 and 7 for more extensive
characterization and chemical reactivity studies, an alterna-
tive synthetic route to these photoproducts was devised. Thus,
as Scheme 3 shows, reaction of the MIV dichlorides 1a and 1b
with a slight excess of trimethylsilylazide, N3SiMe3, in toluene
solution at room temperature provided excellent yields of the
corresponding MVI terminal nitrido, chlorides, [Cp*M{N-
(iPr)C(Me)N(iPr)}(N)Cl] 8a (M = Mo) and 8b (M = W).[8,10]
The molecular structures of 8a and 8b were confirmed by
single-crystal X-ray analyses which revealed solid-state struc-
tural similarities with the reported pair of MVI nitrido,
of 5 and 7 served to confirm the dinuclear bis(m-nitrido)
molecular structures of these two compounds in the solid
state.[8] For the Group 6 metals, only one experimental
example of the formal MV (d1), MV (d1) ground state
electronic configuration that is suggested by the {M(m-N)}2
four-membered ring core of 5 and 7 has been reported for
M = Cr.[12] For [{(iPr2N)2Cr(m-N)}2], variable temperature
NMR spectroscopy and SQUID data support the conclusion
that the two CrV metal centers are antiferromagnetically
coupled in solution and the solid-state, while a crystallo-
graphic analysis reveals a rhomboid structure for the {Cr(m-
Angew. Chem. Int. Ed. 2015, 54, 10220 –10224
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10221

.
Angewandte
Communications
À
N)}2 core with a pair of ꢀlongꢁ and ꢀshortꢁ Cr N bond
lengths of 1.743(3) Š and 1.730(3) Š, respectively. For
5 and 7, the exact nature of the ground-state
electronic configuration for each compound is far
less certain. The detailed crystallographic analyses of
the data for 5 and 7 establish, with a very high degree
of confidence, that the two bridging first-row elements
are indeed nitrogen in each case—thereby ruling out
any possible structural artefact that might be intro-
duced with partial or full occupancy by adventitious
inclusion of oxygen in these positions.[8] Variable
temperature 1H NMR (500 MHz, [D8]toluene, 213–
354 K) spectra establish that while 7 is diamagnetic in
Scheme 4. Functionalization of 5 and 7. Yields are based on durene as an
solution, it is engaged in a dynamic structural internal standard.
process.[8] Geometric parameters for the {W(m-N)}2
À
core of 7 are also consistent with a W W bonding
interaction in the solid-state (see Figure 1). In contrast, the
corresponding {Mo(m-N)}2 core of 5 displays a striking
À
asymmetry with a pair of ꢀlongꢁ Mo(1) N(1A) bonds of
1.9636(13) Š being associated with the Mo(1) center and
À
a pair of ꢀshortꢁ Mo(2) N(1A) bonds of 1.8495(13) Š being
assigned to the Mo(2) center. Preliminary magnetic data
obtained using a SQUID magnetometer strongly supports
a diamagnetic closed-shell electronic configuration in the
solid-state.[8] On the other hand, variable temperature NMR
and EPR spectra recorded for solutions of 5 are indicative of
paramagnetic character. Additional detailed experimental
and computational investigations of the molecular and
electronic structures of 5 and 7 are clearly warranted and
the results of these studies will be reported in due course.
A preliminary screen of the chemical reactivity of 5 and 7
revealed that addition of a slight excess of Me3SiCl to
a benzene solution of these compounds rapidly produced an
excellent yield of a 1:1 mixture of the corresponding mono-
nuclear nitrido, chlorides, 8a and 8b, respectively, along with
the previously reported MoIV terminal imido complexes,
[Cp*M{N(iPr)C(Me)N(iPr)}(NSiMe3)] 3a (M = Mo) and 3b
Scheme 5. Photolysis of 2 in the presence of excess equivalents of
(M = W),[13] according to Scheme 4.[8,14,15] More interestingly,
as Scheme 4 further presents, replacement of Me3SiCl with
the trisubstituted Group 14 chlorides, Ph3SiCl, Me3GeCl, and
Me3CCl provided the corresponding MoIV imido complexes,
3c, 10, and previously reported 11[13] in modest to excellent
yields. Thus, with the isolation of analytically pure 10 and
solid-state characterization achieved through a single-crystal
X-ray analysis,[8] the collection of compounds, 11, 3a, and 10,
now establish an unprecedented Group 14 vertical series of
isostructural metal imido congeners.[10,16] Finally, in support of
the hypothesis that reaction of 5 with Me3SiCl proceeds
through chloride-atom abstraction and capture of a trimethyl-
silyl radical to produce 8a and 3a, respectively, use of one
equivalent of bis(trimethylsilyl)mercury, Hg(SiMe3)2,[17] now
led to a near quantitative yield of 3a (Scheme 4).
With respect to the development of Step B in Scheme 1,
the chemical transformations presented in Scheme 2 and
Scheme 4 raised the possibility of being able to directly
convert 2!3 by photolyzing 2 in the presence of excess
equivalents of Me3SiCl.[6] In practice, while this hypothesis
proved to be correct, its experimental verification yielded yet
another surprise. More specifically, in addition to the
R3ECl and subsequent nitrene-group transfer from 3, 10, and 11 to CO.
Yields of products are relative to 2 and based on durene as an internal
standard.[8]
expected terminal imido product 3, this ꢀone-potꢁ photo-
conversion of 2 now provided the corresponding metal
dichloride 1 as the co-product with no 8 being observed
(Scheme 5). Importantly, results obtained from five separate
experiments conducted with 2b firmly established a reprodu-
cible yield for 3b of 63 Æ 1% and for 1b of 33 Æ 1%. On the
strength of these data in which up to 96% of the metal can be
accounted for based on the stoichiometry of Scheme 5, it is
reasonable to propose a formal mechanism in which one
equivalent of the dinuclear m-N2 starting material 2 is
efficiently acting as an internal trap for the total of four
chlorine atoms that are being generated. Finally, one of the
most exciting consequences of Scheme 5 is that we had
already reported NGT using CO as a substrate. Namely,
treatment of the terminal imido complexes 3a and 3b with
CO (10 psi) as shown in this scheme, provides an excellent
yield of Me3SiNCO and the corresponding Group 6 MII
10222 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10220 –10224

Angewandte
Chemie
bis(carbonyl) complexes, [Cp*M{N(iPr)C(Me)N(iPr)}(CO)2]
9a (M = Mo) and 9b (M = W), respectively.[13]
The versatility of the N2 fixation process established by
the chemistry in Scheme 5 was explored further. To begin,
chemical reduction of 1a under an atmosphere of isotopically
labeled 15N2 (98%) first provided (15N2)-2a that was then
photolyzed to first provide (15N)-3a, which was then reacted
with isotopically labelled 13CO (99%) to produce double
isotopically labelled [15N, 13C]-Me3SiNCO in high yield.[8] The
series of heteronuclear (13C, 15N, and 29Si) NMR spectra
obtained for this product (Figure 2) serve to establish the
Scheme 6. Synthesis of Group 6 metal dichlorides by oxygen atom
transfer.[8]
Figure 2. Partial heteronuclear NMR spectra for [15N, 13C]-Me3SiNCO.
1
a) 13C{1H} NMR (125.76 MHz, [D6]benzene): d=124.41 ppm; J(13C-
15N)=42.8 Hz, b) 15N NMR (50.68 MHz, [D6]benzene, neat MeNO2 as
1
external standard): d=33.48 ppm; J(15N-13C)=42.8 Hz., c) 29Si DEPT-
35 NMR (99.36 MHz, 4.67% natural abundance 29Si, [D6]benzene, neat
2
Me4Si as external standard): d=3.91 ppm; 1J(29Si-15N)=14.7 Hz, J-
(29Si-13C)=8.5 Hz.
Scheme 7. Synthesis of Group 6 metal dichlorides by nitrene-group
transfer/oxygen-atom transfer with CO2 and Me3ECl. Yields are based
on durene as an internal standard.[8]
efficiency and extent of double isotopic incorporation.[8]
Similar results were obtained upon substitution of Me3SiCl
with the other R3ECl derivatives Ph3SiCl, Me3GeCl, and
Me3CCl in the photolysis of 2a (Scheme 5) to respectively
provide the corresponding terminal imido complexes 3c, 10,
and 11.[8,13] Finally, treatment of 3c and (15N)-10, and (15N)-11
with 13CO yielded [15N, 13C]-Ph3SiNCO and [15N, 13C]-
Me3GeNCO, and [15N, 13C]-Me3CNCO respectively
(Scheme 5), and the structural identities of these heterocu-
Although Scheme 5 and Scheme 6 comprise a formal N2
fixation chemical cycle for production of Group 14 isocya-
nates, the overall process requires both CO and CO2, as well
as photolysis. Accordingly, to provide a more efficient path,
the direct reaction of 3a, 3b, and 10 with CO2 and Me3ECl
(E = Si and Ge) was investigated. As Scheme 7 reveals, when
a solution of 3a in [D6]benzene was heated under CO2
pressure (70 psi) at 758C for 14 days within a thick-walled
sealed NMR tube, conversion into 4a (76% yield) and
Me3SiNCO (73% yield) occurred as established by 1H NMR
spectroscopy and an internal standard (durene).[8] Under
lower CO2 pressure (20 psi), compound 10 was observed to
convert even more efficiently at the same temperature into 4a
and Me3GeNCO in 82% and 80% respective yields after
a total of 14 days. To our knowledge, these results provide the
first documentation of simultaneous NGTand OAT involving
CO2 and a Group 6 metal imido as substrates.[21] A final
consideration was to determine if 3a and 10 could be directly
converted into 1a by simply heating under CO2 pressure in
the presence of excess equivalents of Me3SiCl and Me3GeCl,
respectively. Fortunately, as shown in Scheme 7, this goal was
achieved in a surprisingly easy manner and in high yield.[8]
To test the validity of combining the NGT and OAT
processes of Scheme 5–7 to provide complete N2 fixation
cycles for production of isocyanates, fixed amounts of 2a and
2b were carried through in one-pot fashion to provide these
1
mulene products were confirmed by H, 13C, and 15N NMR
spectra and comparison with literature values.[8,18]
To complete the formal catalytic cycle for N2 fixation
presented in Scheme 1, a step is required to recycle the metal
bis(carbonyl) complexes 9 back to the starting metal dichlor-
ides 1. Gratifyingly, this goal could be realized by relying on
our reported catalytic OAT process involving CO2 and
a Group 6 MII/MIV redox couple. More specifically, as
presented in Scheme 6, photolysis of 9a and 9b in the
presence of CO2 cleanly provides the corresponding mono-
nuclear terminal oxo complexes 4a (M = Mo) and 4b (M =
W).[19] In the present work, it was determined that both 4a
and 4b react with excess equivalents of Me3SiCl in benzene
solution to produce one equivalent of hexamethyldisiloxane,
Me3Si-O-SiMe3, and the desired respective products 1a and
1b (see Scheme 6).[8,20] Even more satisfying, the 9a!1a
transformation could be achieved by simply photolyzing 9a in
the presence of CO2 and excess equivalents of Me3SiCl, to
directly produce 1a in high yield and without the need to
isolate the intermediate terminal oxo complex 4a.
Angew. Chem. Int. Ed. 2015, 54, 10220 –10224
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10223

.
Angewandte
Communications
40, 3907 – 3909; Angew. Chem. 2001, 113, 4025 – 4027; b) J. J.
Curley, T. R. Cook, S. Y. Reece, P. Müller, C. C. Cummins, J. Am.
Chem. Soc. 2008, 130, 9394 – 9405; c) H. Kunkely, A. Vogler,
Angew. Chem. Int. Ed. 2010, 49, 1591 – 1593; Angew. Chem.
2010, 122, 1636 – 1638; d) A. S. Huss, J. J. Curly, C. C. Cummins,
D. A. Blank, J. Phys. Chem. B 2013, 117, 1429 – 1436; e) C.
Rebreyend, B. de Bruin, Angew. Chem. Int. Ed. 2015, 54, 42 – 44;
Angew. Chem. 2015, 127, 42 – 44; f) T. Miyazaki, H. Tanaka, Y.
Tanabe, M. Yuki, K. Nakajima, K. Yoshizawa, Y. Nishibayashi,
Angew. Chem. Int. Ed. 2014, 53, 11488 – 11492; Angew. Chem.
2014, 126, 11672 – 11676.
dinuclear m-N2 starting materials in 50% recovered yields
when CO was employed (Scheme 6) and a 82% yield of the
dichloride 1a was isolated when CO2 was used for simulta-
neous NGT and OAT according to Scheme 7.[8]
In summary, the present work establishes a highly efficient
and versatile chemical cycle for the production of isocyanates
from N2, CO2, and R3ECl, and one that proceeds under
relatively mild conditions. We are currently investigating
further refinements to this chemical cycle that would permit it
to proceed in true catalytic fashion.
[10] R. A. Eikey, M. M. Abu-Omar, Coord. Chem. Rev. 2003, 243,
83 – 124.
[11] B. L. Yonke, J. P. Reeds, P. Y. Zavalij, L. R. Sita, Z. Anorg. Allg.
Chem. 2015, 641, 61 – 64.
Keywords: nitrene-group transfer · nitrogen fixation ·
oxygen-atom transfer · small-molecule activation
[12] A. L. Odom, C. C. Cummins, Organometallics 1996, 15, 898 –
900.
[13] B. L. Yonke, J. P. Reeds, P. P. Fontaine, P. Y. Zavalij, L. R. Sita,
Organometallics 2014, 33, 3239 – 3242.
How to cite: Angew. Chem. Int. Ed. 2015, 54, 10220–10224
Angew. Chem. 2015, 127, 10358–10362
[14] Z. J. Tonzetich, R. R. Schrock, K. M. Wampler, B. C. Bailey,
C. C. Cummins, P. Müller, Inorg. Chem. 2008, 47, 1560 – 1567.
[15] For recent examples of nitride-functionalization, see: a) C. R.
Clough, P. Müller, C. C. Cummins, Dalton Trans. 2008, 4458 –
4463; b) J. S. Silvia, C. C. Cummins, J. Am. Chem. Soc. 2010, 132,
2169 – 2171; c) J. J. Scepaniak, R. P. Bontchev, D. L. Johnson,
J. M. Smith, Angew. Chem. Int. Ed. 2011, 50, 6630 – 6633; Angew.
Chem. 2011, 123, 6760 – 6763; d) P. J. J. Curley, A. F. Cozzolino,
C. C. Cummins, Dalton. Trans. 2011, 40, 2429 – 2432; e) P. A.
Cleaves, D. M. King, C. E. Kefalidis, L. Maron, F. Tuna, E. J. L.
McInnes, J. McMaster, W. Lewis, A. J. Blake, S. T. Liddle,
Angew. Chem. Int. Ed. 2014, 53, 10412 – 10415; Angew. Chem.
2014, 126, 10580 – 10583; f) I. Klopsch, M. Finger, C. Würtele, B.
Milde, D. B. Werz, S. Schneider, J. Am. Chem. Soc. 2014, 136,
6881 – 6883.
[1] a) K. C. MacLeod, P. L. Holland, Nat. Chem. 2013, 5, 559 – 565;
b) N. Khoenkhoen, B. de Bruin, J. N. H. Reek, W. I. Dzik, Eur. J.
Inorg. Chem. 2015, 567 – 598.
[2] For formation of a carbodiimide product from N2 and CO
through a series of transformations that do not constitute
a chemical cycle, see: S. P. Semproni, P. J. Chirik, J. Am. Chem.
Soc. 2013, 135, 11373 – 11383.
[3] For a chemical cycle for formation of KNCO from N2 and CO,
see: a) Y. Ishida, H. Kawaguchi, J. Am. Chem. Soc. 2014, 136,
16990 – 16993; For related work, see: b) J. S. Silvia, C. C.
Cummins, J. Am. Chem. Soc. 2009, 131, 446 – 447; c) A. F.
Cozzolino, J. S. Silvia, N. Lopez, C. C. Cummins, Dalton Trans.
2014, 43, 4639 – 4652.
[4] For development of a chemical cycle for formation of organo-
nitriles from N2 and acid chlorides, see: a) J. J. Curley, E. L.
Sceats, C. C. Cummins, J. Am. Chem. Soc. 2006, 128, 14036 –
14037; b) J. S. Figueroa, N. A. Piro, C. R. Clough, C. C. Cum-
mins, J. Am. Chem. Soc. 2006, 128, 940 – 950.
[16] K. Weber, K. Korn, A. Schorm, J. Kipke, M. Lemke, A.
Khvorost, K. Harms, J. Sundermeyer, Z. Anorg. Allg. Chem.
2003, 629, 744 – 754.
[17] J. Arnold, T. D. Tilley, A. L. Rheingold, S. J. Geib, Organo-
metallics 1987, 6, 473 – 479.
[5] For metal-mediated N2 fixation leading to other small molecule
products, see: a) P. Sobota, Z. Janas, J. Organomet. Chem. 1984,
276, 171 – 176; b) W. H. Bernskoetter, E. Lobkovsky, P. J. Chirik,
Angew. Chem. Int. Ed. 2007, 46, 2858 – 2861; Angew. Chem. 2007,
119, 2916 – 2919; c) D. J. Knobloch, H. E. Toomey, P. J. Chirik, J.
Am. Chem. Soc. 2008, 130, 4248 – 4249; d) D. J. Knobloch, E.
Lobkovsky, P. J. Chirik, J. Am. Chem. Soc. 2010, 132, 10553 –
10564; e) D. J. Knobloch, E. Lobkovsky, P. J. Chirik, J. Am.
Chem. Soc. 2010, 132, 15340 – 15350; f) S. P. Semproni, E.
Lobkovsky, P. J. Chirik, J. Am. Chem. Soc. 2011, 133, 10406 –
10409.
[6] For catalytic formation of N(SiMe3)3 from N2 and Me3SiCl, see:
a) K. Shiina, J. Am. Chem. Soc. 1972, 94, 9266 – 9267; b) H.
Tanaka, K. Arashiba, S. Kuriyama, A. Sasada, K. Nakajima, K.
Yoshizawa, Y. Nishibayashi, Nat. Commun. 2014, 5, 3737 – 3747.
[7] P. P. Fontaine, B. L. Yonke, P. Y. Zavalij, L. R. Sita, J. Am. Chem.
Soc. 2010, 132, 12273 – 12285.
[18] M. H. Palmer, J. Mol. Struct. 2004, 692, 43 – 56.
[19] a) B. L. Yonke, J. P. Reeds, P. Y. Zavalij, L. R. Sita, Angew.
Chem. Int. Ed. 2011, 50, 12342 – 12346; Angew. Chem. 2011, 123,
12550 – 12554; b) J. P. Reeds, B. L. Yonke, P. Y. Zavalij, L. R.
Sita, J. Am. Chem. Soc. 2011, 133, 18602 – 18605.
[20] J. H. Shin, W. Savage, V. J. Murphy, J. B. Bonanno, D. G.
Churchill, G. Parkin, J. Chem. Soc. Dalton Trans. 2001, 1732 –
1753.
[21] a) A. J. Blake, J. M. McInnes, P. Mountford, G. I. Nikonov, D.
Swallow, D. J. Watkin, J. Chem. Soc. Dalton Trans. 1999, 379 –
392; b) P. Royo, J. Sµnchez-Nieves, J. Organomet. Chem. 2000,
597, 61 – 68; c) U. J. Kilgore, F. Basuli, J. C. Huffman, D. J.
Mindiola, Inorg. Chem. 2006, 45, 487 – 489; d) Z. J. Tonzetich,
R. R. Schrock, P. Müller, Organometallics 2006, 25, 4301 – 4306.
[8] Experimental details are provided in the Supporting Informa-
tion.
[9] a) E. Solari, C. Da Silva, B. Iacono, J. Hesschenbrouck, C.
Rizzoli, R. Scopelliti, C. Floriani, Angew. Chem. Int. Ed. 2001,
Received: March 11, 2015
Published online: June 26, 2015
10224 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10220 –10224