Nitrogen fixation to cyanide at a molybdenum center

Article abstract of DOI:10.1039/c0dt01326a

Facile methoxymethylation of N₂-derived nitride NMo(N[ ^tBu]Ar)₃ provided the imido cation [MeOCH ₂NMo(N[^tBu]Ar)₃]⁺ as its triflate salt in 88% yield. Treatment of the latter with LiN(SiMe₃) ₂ provided blue methoxyketimide complex MeO(H)CNMo(N[ ^tBu]Ar)₃ in 95% yield. Conversion of the latter to the terminal cyanide complex NCMo(N[^tBu]Ar)₃, which was the subject of a single-crystal X-ray diffraction study, was accomplished in 51% yield upon treatment with a combination of SnCl₂ and Me ₂NSiMe₃.

Full text of DOI:10.1039/c0dt01326a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 2429
www.rsc.org/dalton
COMMUNICATION
Nitrogen fixation to cyanide at a molybdenum center†
John J. Curley, Anthony F. Cozzolino and Christopher C. Cummins*
Received 2nd October 2010, Accepted 23rd December 2010
DOI: 10.1039/c0dt01326a
Facile methoxymethylation of N2-derived nitride NMo(N-
[tBu]Ar)3 provided the imido cation [MeOCH2NMo(N-
[tBu]Ar)3]+ as its triflate salt in 88% yield. Treatment of the
latter with LiN(SiMe3)2 provided blue methoxyketimide com-
plex MeO(H)CNMo(N[tBu]Ar)3 in 95% yield. Conversion of
the latter to the terminal cyanide complex NCMo(N[tBu]Ar)3,
which was the subject of a single-crystal X-ray diffraction
study, was accomplished in 51% yield upon treatment with a
combination of SnCl2 and Me2NSiMe3.
The terminal nitride, 2, undergoes a reaction with 1 equiv
MeOCH2I (MOMI) in CDCl3 to rapidly form the cationic imido,
[MeOCH2NMo(N[tBu]Ar)3][I], [3]I. This reaction is more facile
than the analogous reaction of 2 with MeI, which was best
effected by stirring 2 in neat MeI for several hours.9 However,
the oily [3]I proved difficult to isolate as a solid. Therefore,
we sought to synthesize a salt of cation 3 with a counter-
anion that would impart crystallinity. This was achieved by the
treatment of 2 with a combination of MeOCH2Cl and TIPSOTf
(TIPS = tri-isopropylsilyl) to form a red solution containing
[MeOCH2NMo(N[tBu]Ar)3][OTf], [3]OTf, and TIPSCl. Such a
synthetic procedure makes use of the steric bulk of the TIPS group
to preclude a direct reaction with 3 that would unproductively
yield a silyl imido cation.8 The TIPSCl byproduct is soluble in n-
pentane and may be washed away from [3]OTf, which is isolated in
88% yield. This complex salt is a thermally robust, yellow powder
that forms orange solutions in CHCl3. The methoxymethyl group
is easily displaced by a variety of nucleophiles and the complex
decomposes in the coordinating solvents pyridine and THF. The
15N NMR resonance for [3][OTf], d = 466 ppm, is nearly identical
to that found for [MeNMo(N[tBu]Ar)3][I], d = 462 ppm; the latter
complex has been characterized both crystallographically and by
solid-state 15N NMR spectroscopy.9
An early commercialized nitrogen fixation process used barium
carbide at temperatures of 700–800 ◦C, and produced barium
cyanide exclusively; the process had been discovered in search
of a source of cyanide for precious metal recovery.1 Soon there-
after, Frank and Caro discovered the related calcium cyanamide
process,2 which supplanted that for barium cyanide due to the
myriad chemical applications of cyanamide including use as a
fertilizer. Cyanide as a primary product of nitrogen fixation also
appeared in the work of Bucher according to eqn (1).
(1)
Na2CO3 + 4C + N2 = 2NaCN + 3CO
The Bucher process is noteworthy for its exceptional simplicity
and lack of requirement for purified nitrogen, working even in air
and in non-specialized apparatus at near 950 ◦C, by virtue of the
presence of iron filings as catalyst.3 Ammonia is available from
this method by hydrolysis of sodium cyanide.4,5
Crystals of [3][OTf] suitable for X-ray diffraction were grown
from CHCl3 at -35 ◦C. The solid-state structure of [3][OTf] exhibits
˚
a Mo–N bond length of 1.7119(14) A that compares well with
t
9,10
˚
that of 1.708(9) A found for [EtNMo(N[ Bu]Ar)3][I] (Fig. 1).
In connection with our interest in the chemistry of N2 cleavage
by transition-metal complexes,6 and in the use of nitride complexes
derived therefrom in the assembly of carbon-nitrogen triple bonds
(organic nitriles),7,8 the present work illustrates a method for net
carbon-atom addition to a terminal nitride resulting in the cyanide
function. As illustrated in Scheme 1, the overall process can be
divided into three conceptual steps: (i) dinitrogen cleavage to form
the terminal nitride, (ii) transfer of methoxycarbene to the terminal
nitride with carbon-nitrogen double bond formation, and (iii)
removal of the elements of methanol, delivering bound cyanide.
Accordingly, the process described herein is offered as a chemically
distinct, low-temperature alternative to the earliest techniques for
deriving cyanide from N2.
Interestingly, in [3][OTf] the Mo–N–C bond angle is 172.24(12)◦
whereas in [EtNMo(N[tBu]Ar)3][I] this angle is 180◦, coincident
with a crystallographic C3 axis. While imido systems are plentiful
and important,11 imido complexes bearing a –CH2OR substituent
appear not to have been described previously. Hence, the structure
presented here for the [3]+ cation gives us the first glimpse of the
MOM-imido moiety.
Addition of LiN(SiMe3)2 to yellow solutions of [3][OTf] in Et2O
rapidly generates a dark blue solution. Evaporation of the solvent,
followed by extraction into n-pentane to remove LiOTf affords
MeO(H)CNMo(N[tBu]Ar)3, 4, in 95% yield. The dark blue color
of 4 is congruent with two absorption bands in the visible region:
l(e) = 501 (1900), 615 (2500) nm (M-1 cm-1). The proton bound
to the ketimide, HC N, is shifted downfield from 6.1 ppm for
[3][OTf] to 7.4 ppm for 4. Labeling 4 with 15N splits the NMR
Department of Chemistry, Massachusetts Institute of Technology, Cam-
bridge, MA, 02139-4307. E-mail: ccummins@mit.edu
† Electronic supplementary information (ESI) available. CCDC reference
numbers 795137 and 805491. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0dt01326a
2
signal for this proton into a doublet for which JNH = 2.5 Hz.
A doublet with the same coupling was located in the 15N NMR
spectrum at d = 382 ppm. In contrast to [3][OTf], which maintains
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2429–2432 | 2429
©

View Article Online
Scheme 1 Sequence of (i) dinitrogen cleavage to form terminal nitride, (ii) two-step methoxycarbene addition with C N bond formation, and
(iii) removal of the elements of methanol revealing the cyanide function.
its integrity when stored at 20 ◦C for weeks, compound 4 is
thermally sensitive and decomposes at ambient temperature within
12 h to form a myriad of products including the terminal cyanide
NCMo(N[tBu]Ar)3, 5.12 Because complex 4 gradually decomposes
in the solid state at -35 ◦C, it was important to work with freshly
prepared material.
We previously reported that complexes of the formula
R1O(R2)CNMo(N[tBu]Ar)3(R1 = Me3Si; R2 = Me, Ph, tBu)
extrude the organic nitriles, R2CN, and form known chloride
ClMo(N[tBu]Ar)313 when treated with SnCl2 or ZnCl2.8 Therefore
we attempted to liberate HCN by treatment of 4 with SnCl2 under
dynamic vacuum. This protocol did result in the formation of
ClMo(N[tBu]Ar)3; however, the reaction was not clean, as both the
nitride complex 2 and free ligand HN(tBu)Ar were also formed.
Moreover, attempts to collect HCN by vacuum transfer were
unsuccessful.
proton base to the reaction mixture might intercept an acidic
proton and afford a cleaner reaction. It was found that treatment
of 4 with SnCl2 in the presence of 10 equiv of Me3SiNMe2
afforded cyanide 5 as the major product. In an important control
experiment, it was determined that 4 exhibits no reaction with
Me3SiNMe2 on its own, in the absence of SnCl2. The formation
of 5 according to this protocol was confirmed by the appearance
of characteristic 1H NMR resonances and by MALDI-TOF mass
spectrometry. In addition, cyanide 5 could be isolated in 51% yield
by crystallization from the reaction mixture.
While an alternative synthesis of cyanide complex 5 has been
reported previously,12 we now present also a single-crystal X-ray
structure analysis of this system; see Fig. 2.10 The key structural
features observed are (i) linearity at cyanide carbon C4, (ii) a
coordination geometry at Mo that is midway between tetrahedral
and trigonal monopyramidal with C4–Mo–N(1–3) angles of
approximately 100◦, and a short cyano C–N distance, 1.122(2)
The observed formation of chloride ClMo(N[tBu]Ar)3 was
consistent with the initial hypothesis that 4 may react with SnCl2
in a fashion similar to previously reported ketimide complexes
R1O(R2)CNMo(N[tBu]Ar)3. As a result, we postulated that when
R2 = H, an acidic proton is generated during the treatment of 4 with
SnCl2. Accordingly, we speculated that the addition of a suitable
˚
A. While there is nothing unusual about the structure of 5, it is a
fact that we have been unable to observe a nCN band in the infrared
spectrum of this substance (Figure S8†). On the other hand,
examples are known of molybdenum mono-cyanide complexes
having a weak or absent nCN infrared feature.14–16
2430 | Dalton Trans., 2011, 40, 2429–2432
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
decays away while an absorbance due to cyanide 5, = 425 nm,
gains in intensity (Fig. 3). The reaction by which 5 is formed
from 4 poses a mechanistic question because the cyanide function
may either dissociate from the metal or be retained on a single
molybdenum center through the course of the reaction. For
example, this reaction may proceed via the initial production
of Me3SiCN and ClMo(N[tBu]Ar)3 prior to the formation of 5.
Alternatively, a pathway in which the cyanide moiety does not
leave the metal center may be operative. To test the first hypothesis,
ClMo(N[tBu]Ar)3 was treated with Me3SiCN. This protocol was
found to afford compound 5 in 97% yield. However, this protocol
for synthesizing 5 was found to require stirring for several hours
at 20 ◦C to reach completion. The reaction is only about 33%
complete after 1 h when carried out at concentrations similar to
those employed for the conversion of 4 to 5. As the latter reaction
is complete within 30 min, it seems unlikely that a pathway that
requires formation of Me3SiCN predominates.
Fig. 1 Thermal ellipsoid plot (50% probability level) of imido cation [3]+
(CHCl3 of crystallization, hydrogen atoms, and triflate counter-ion not
shown).
Fig. 3 UV-Vis spectral changes accompanying the 4→5 conversion with
spectra shown at 5 min intervals.
Additional mechanistic information ensued from an isotope-
labeling study. If trimethylsilyl cyanide were an intermediate, then
adding this reagent to the reaction mixture containing 15N-4 would
be expected to result in crossover of the nitrogen isotopomers.
Accordingly, 15N-4, prepared from 15N2, was treated with SnCl2
and 10 equiv Me3SiNMe2 in the presence of 1 equiv Me3SiCN.
Analysis of the crude reaction mixture by FT-IR showed a
band at 2195 cm-1 corresponding to the added Me3SiCN. No
band consistent with Me3SiC15N was observed in the spectrum,
and all other bands were reproduced in the crude reaction
mixture obtained without added Me3SiCN. These observations
are consistent with a mechanism in which the CN unit does
not leave the metal. Although the metal isocyanide complex
CNMo(N[tBu]Ar)3 (5-iso, Scheme 1) is not required to be an
intermediate, linkage isomerism of the cyanide ligand has been
studied.17 These studies indicated a rapid conversion of the M–
Fig. 2 Thermal ellipsoid plot (50% probability level) of cyanide complex
NCMo(N[ Bu]Ar)3, 5. Selected interatomic distances (A) and angles ( ):
C4 N41 1.122(2), C4 Mo1 2.0918(17), N1 Mo1 1.9574(12), N2 Mo1
1.9530(12), N3 Mo1 1.9637(12), N41 C4 Mo1 178.95(15), N2 Mo1 N1
116.51(5), N2 Mo1 N3 117.41(5), N1 Mo1 N3 118.12(5), N2 Mo1 C4
100.43(5), N1 Mo1 C4 98.51(6), N3 Mo1 C4 99.48(6).
◦
t
˚
N
C to M–C N via an irreversible, first-order process.
Herein we have reported a surprising transformation of ketimide
complex 4 into a corresponding terminal cyanide complex, 5. As 4
is derived from simple alkylation and deprotonation of 2 (formal
carbene transfer) the overall process converts a terminal N3- ligand
into a CN- ligand that proceeds through three triply bonded atom
pairs: N N → Mo N → C N. The net transformation may
be viewed as the formal insertion of a carbon atom into the
The reaction that converts 4 to 5 (Scheme 1) proceeds over
the course of approximately 30 min with a color change from
a dark blue to red brown. When the heterogeneous reaction
mixture is monitored by UV-vis spectroscopy, the intense band
attributed to the metal-to-ligand charge transfer of 4, = 615 nm,
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2429–2432 | 2431
©

View Article Online
Mo N triple bond. This work delivers a new synthetic method
for terminal nitride functionalization via conversion to the rare
methoxymethyl imido unit.
We thank the National Science Foundation (CHE-719157) and
NSERC (postdoctoral fellowship to AFC) for support of this
work.
7 (a) J. S. Figueroa, N. A. Piro, C. R. Clough and C. C. Cummins, J. Am.
Chem. Soc., 2006, 128, 940–950; (b) J. S. Figueroa and C. C. Cummins,
Dalton Trans., 2006, 2161–2168.
8 J. J. Curley, E. L. Sceats and C. C. Cummins, J. Am. Chem. Soc., 2006,
128, 14036–14037.
9 E. L. Sceats, J. S. Figueroa, C. C. Cummins, N. M. Loening,
P. Van der Wel and R. G. Griffin, Polyhedron, 2004, 23, 2751–
2768.
10 The structure data for [3][OTf] and 5 have been deposited with the
Cambridge Crystallographic Data Centre and assigned respective
CCDC numbers 795137 and 805491†.
References
11 D. E. Wigley, Prog. Inorg. Chem., 42, 239–482.
12 J. C. Peters, L. M. Baraldo, T. A. Baker, A. R. Johnson and C. C.
Cummins, J. Organomet. Chem., 1999, 591, 24–35.
13 (a) A. Fu¨rstner, C. Mathes and C. W. Lehmann, J. Am. Chem. Soc.,
1999, 121, 9453–9454; (b) A. Fu¨rstner, C. Mathes and C. W. Lehmann,
Chem.–Eur. J., 2001, 7, 5299–5317.
14 G. E. Greco, M. B. O’Donoghue, S. W. Seidel, W. M. Davis and R. R.
Schrock, Organometallics, 2000, 19, 1132–1149.
15 A. Hills, D. L. Hughes, C. J. Macdonald, M. Y. Mohammed and C. J.
Pickett, J. Chem. Soc., Dalton Trans., 1991, 121–129.
16 H. Arzoumanian, C. Carao, H. Krentzien, R. Lopez and H. Teruel, J.
Chem. Soc., Chem. Commun., 1992, 856–858.
17 (a) J. Halpern and S. Nakamura, J. Am. Chem. Soc., 1965, 87, 3002–
3003; (b) J. H. Espenson and J. P. Birk, J. Am. Chem. Soc., 1965, 87,
3280–3281; (c) J. P. Birk and J. H. Espenson, J. Am. Chem. Soc., 1968,
90, 1153–1162.
1 M. L. Kastens and W. G. McBurney, Ind. Eng. Chem., 1951, 43, 1020–
1033.
2 F. Aftalion, A history of the international chemical industry, Chemical
Heritage Foundation, Philadelphia, PA, 2001.
3 J. E. Bucher, Ind. Eng. Chem., 1917, 9, 233–253.
4 G. W. Heise and H. E. Foote, Ind. Eng. Chem., 1920, 12, 331–336.
5 F. E. Bartell, Ind. Eng. Chem., 1922, 14, 516–520.
6 (a) C. E. Laplaza and C. C. Cummins, Science, 1995, 268, 861–863;
(b) C. E. Laplaza, M. J. A. Johnson, J. C. Peters, A. L. Odom, E. Kim,
C. C. Cummins, G. N. George and I. J. Pickering, J. Am. Chem. Soc.,
1996, 118, 8623–8638; (c) J. C. Peters, J.-P. F. Cherry, J. C. Thomas,
L. Baraldo, D. J. Mindiola, W. M. Davis and C. C. Cummins, J. Am.
Chem. Soc., 1999, 121, 10053–10067; (d) Y.-C. Tsai and C. C. Cummins,
Inorg. Chim. Acta, 2003, 345, 63–69; (e) J. J. Curley, T. R. Cook, S. Y.
Reece, P. Mu¨ller and C. C. Cummins, J. Am. Chem. Soc., 2008, 130,
9394–9405.
2432 | Dalton Trans., 2011, 40, 2429–2432
This journal is
The Royal Society of Chemistry 2011
©