Dimolybdenum-μ-cyanide complexes supported by N-tert-butylanilide ligation: In pursuit of cyanide reductive cleavage

Article abstract of DOI:10.1016/S0022-328X(99)00485-4

The red cyanide complex (NC)Mo(N[R]Ar)₃ (R=C(CD₃)₂CH₃, Ar=3,5-C₆H₃Me₂) was prepared in 77% yield by reaction of iodide IMo(N[R]Ar)₃ with tetra-n-butyl ammonium cyanide. By reaction of cyanide (NC)Mo(N[R]Ar)₃ with the three-coordinate molybdenum(III) complex Mo(N[R]Ar)₃ was prepared the dimolybdenum-μ-cyanide complex (μ-CN){Mo(N[R]Ar)₃}₂ as a violet solid in 80% yield. Reduction of μ-cyanide (μ-CN){Mo(N[R]Ar)₃}₂ by one electron would give a cyanide-bridged anion isoelectronic with the known μ-N₂ complex (μ-N₂){Mo(N[R]Ar)₃}₂, an intermediate in dinitrogen cleavage to two equivalents of nitride NMo(N[R]Ar)₃ by Mo(N[R]Ar)₃. Instead of undergoing an analogous cleavage of cyanide upon one-electron reduction, μ-cyanide (μ-CN){Mo(N[R]Ar)₃}₂ was found to undergo expulsion of a ligand C(CD₃)₂CH₃ substituent upon exposure to reducing conditions, the product isolated in 50% yield being imido-μ-cyanide (Ar[R]N)₂(ArN)Mo(μ-NC)Mo(N[R]Ar)₃. By reaction of [NⁿBu₄][CN] with the 1-adamantyl-substituted molybdenum complex Mo(N[Ad]Ar)₃, the blue salt [NⁿBu₄][(NC)Mo(N[Ad]Ar)₃] was obtained in 91% yield. Reaction of ferrocenium triflate or silver triflate with [NⁿBu₄][(NC)Mo(N[Ad]Ar)₃] gave ferrocene or silver along with the neutral cyanide complex (NC)Mo(N[Ad]Ar)₃, isolated in 74% yield. While reaction of (NC)Mo(N[Ad]Ar)₃ with Mo(N[R]Ar)₃ gave in 53% yield a burgundy-colored dimolybdenum-μ-cyanide complex (Ar[Ad]N)₃Mo(μ-CN)Mo(N[R]Ar)₃. the 1-adamantyl-substituted cyanide did not exhibit any reaction with the 1-adamantyl-substituted tricoordinate complex Mo(N[Ad]Ar)₃. The latter results indicate that cyanide is too small to serve as a bridge for two equivalents of the highly sterically encumbered Mo(N[Ad]Ar)₃ fragment. A metathetical route to a heterodinuclear cyanide-bridged complex was explored involving addition of [NⁿBu₄][(NC)Mo(N[Ad]Ar)₃] to the vanadium iodide complex IV(N[R]Ar_F)₂. By this reaction was obtained the orange-brown μ-cyanide complex (Ar_F[R]N)₂V(μ-NC)Mo(N[Ad]Ar)₃ in 30% recrystallized yield. The latter was characterized by X-ray crystallography. The cyanide chemistry reported here is interpreted with the aid of bonding considerations and cyclic voltammetry studies on the new complexes.

Full text of DOI:10.1016/S0022-328X(99)00485-4

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 591 (1999) 24–35
Dimolybdenum–m-cyanide complexes supported by
N-tert-butylanilide ligation: in pursuit of cyanide reductive
cleavage
Jonas C. Peters, Luis M. Baraldo, Thomas A. Baker, Adam R Johnson,
Christopher C. Cummins *
Department of Chemistry, Massachusetts Institute of Technology, Room 2-227, Cambridge, MA 02139-4307, USA
Received 15 June 1999; accepted 6 August 1999
Abstract
The red cyanide complex (NC)Mo(N[R]Ar)3 (R=C(CD3)2CH3, Ar=3,5-C6H3Me2) was prepared in 77% yield by reaction of
iodide IMo(N[R]Ar)3 with tetra-n-butyl ammonium cyanide. By reaction of cyanide (NC)Mo(N[R]Ar)3 with the three-coordinate
molybdenum(III) complex Mo(N[R]Ar)3 was prepared the dimolybdenum-m-cyanide complex (m-CN){Mo(N[R]Ar)3}2 as a violet
solid in 80% yield. Reduction of m-cyanide (m-CN){Mo(N[R]Ar)3}2 by one electron would give a cyanide-bridged anion
isoelectronic with the known m-N2 complex (m-N2){Mo(N[R]Ar)3}2, an intermediate in dinitrogen cleavage to two equivalents of
nitride NMo(N[R]Ar)3 by Mo(N[R]Ar)3. Instead of undergoing an analogous cleavage of cyanide upon one-electron reduction,
m-cyanide (m-CN){Mo(N[R]Ar)3}2 was found to undergo expulsion of a ligand C(CD3)2CH3 substituent upon exposure to
reducing conditions, the product isolated in 50% yield being imido–m-cyanide (Ar[R]N)2(ArN)Mo(m-NC)Mo(N[R]Ar)3. By
reaction of [NnBu4][CN] with the 1-adamantyl-substituted molybdenum complex Mo(N[Ad]Ar)3, the blue salt [NnBu4]-
[(NC)Mo(N[Ad]Ar)3] was obtained in 91% yield. Reaction of ferrocenium triflate or silver triflate with [NnBu4]-
[(NC)Mo(N[Ad]Ar)3] gave ferrocene or silver along with the neutral cyanide complex (NC)Mo(N[Ad]Ar)3, isolated in 74% yield.
While reaction of (NC)Mo(N[Ad]Ar)3 with Mo(N[R]Ar)3 gave in 53% yield a burgundy-colored dimolybdenum–m-cyanide
complex (Ar[Ad]N)3Mo(m-CN)Mo(N[R]Ar)3. the 1-adamantyl-substituted cyanide did not exhibit any reaction with the 1-
adamantyl-substituted tricoordinate complex Mo(N[Ad]Ar)3. The latter results indicate that cyanide is too small to serve as a
bridge for two equivalents of the highly sterically encumbered Mo(N[Ad]Ar)3 fragment. A metathetical route to a heterodinuclear
cyanide-bridged complex was explored involving addition of [NnBu4][(NC)Mo(N[Ad]Ar)3] to the vanadium iodide complex
IV(N[R]ArF)2. By this reaction was obtained the orange–brown m-cyanide complex (ArF[R]N)2V(m-NC)Mo(N[Ad]Ar)3 in 30%
recrystallized yield. The latter was characterized by X-ray crystallography. The cyanide chemistry reported here is interpreted with
the aid of bonding considerations and cyclic voltammetry studies on the new complexes. © 1999 Elsevier Science S.A. All rights
reserved.
Keywords: Molybdenum; Cyano; N-tert-Butylanilide
the substrate, a substantial thermodynamic driving
force can be traced to the formation of two extremely
strong MoꢀN triple bonds for each molecule of dinitro-
gen severed [6–8]. Following this line of reasoning, it
was of interest to determine what chemistry would be
manifested by 1 in conjunction with carbon monoxide,
on the one hand [9], and cyanide ion on the other.
Known carbon monoxide and cyanide reductive cleav-
age reactions are relevant [10–14], and a splitting of
nitric oxide by a d3 metal complex in conjunction with
a d2 metal complex has been documented [15].
1. Introduction
The reductive scission of molecular nitrogen by two
equivalents of d3 Mo(N[R]Ar)3 (1, R=C(CD3)2CH3,
Ar=3,5-C6H3Me2) established a paradigm for the six-
electron reductive cleavage of a ten-valence-electron
diatomic molecule [1–5]. In the case of dinitrogen as
* Corresponding author. Fax: +1-617-258-6989.
E-mail address: ccummins@mit.edu (C.C. Cummins)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00485-4

J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
25
Initial investigations into the carbonylation of 1 re-
vealed a preference for formation of the mononuclear
monocarbonyl (OC)Mo(N[R]Ar)3 (2) [9]. No reaction
was observed between 1 and carbonyl 2, such that no
CO splitting reaction is to be anticipated for this sys-
tem. On the other hand, stepwise reduction and derivi-
tization of carbonyl 2 was found to afford an
Although slow addition of one equivalent of 1 to 0.5
equivalents of [NnBu4][CN] was attempted as a direct
route to the desired bridged species [NnBu4][(m-
CN){Mo(N[R]Ar)3}2] (7), the reaction instead gave an
uncharacterized mixture of products. Spectroscopic
analysis of the crude mixture resulting from the latter
2:1 stoichiometry showed none of the anticipated
NMo(N[R]Ar)3 (4), an expected product of successful
cyanide cleavage.
unprecedented
terminal
carbide
complex
[M][CMo(N[R]Ar)3], isolated as alkali metal salts (3,
M=K(kryptofix) or K(benzo-15-c-5)2) [9]. The anionic
component of [M][CMo(N[R]Ar)3] is a terminal carbide
species isoelectronic with NMo(N[R]Ar)3 (4) [3].
2.2. Synthesis of [NnBu4][(NC)Mo(N[Ad]Ar)3] (8)
If cyanide ion were to undergo reductive scission in
the presence of 1 (two equivalents), then a carbide
product such as 3 is to be expected along with nitride 4.
Cyanide complex chemistry has been reviewed [16,17],
and advances in the activation and functionalization of
bound cyanide have been described [18–20]. The goal
of the present work was to delineate the chemistry of 1
in the presence of cyanide ion. Also utilized in the
course of this study is the more hindered, adamantyl-
substituted analog of 1, namely Mo(N[Ad]Ar)3 (5,
Ad=1-adamantyl, Ar=3,5-C6H3Me2) [21,22], the sup-
porting ligand [23] in this case having been introduced
recently into zirconium [24], vanadium [25], niobium
[26,27], and chromium [28,29] chemistries.
It has been found that a dinitrogen-bridged species
analogous to the well-characterized complex (m-
N2){Mo(N[R]Ar)3}2 (9) [3] is sterically inaccessible for
the more encumbered 1-adamantyl-substituted deriva-
tive Mo(N[Ad]Ar)3 (5) [21,22]. Hence, dinitrogen is not
split by 5. Taking a hint from the dinitrogen results, it
was elected to deliver cyanide ion to the more crowded
Mo(N[Ad]Ar)3 species, 5. This approach was expected
to minimize any side reactions involving
a 2:1
Mo:cyanide stoichiometry.
Addition of a THF solution of [NnBu4][CN] (1 equiv-
alent) to a stirring solution of Mo(N[Ad]Ar)3 (5) at
25°C resulted in a rapid color change from orange to
intense blue. The blue color persisted, and an IR spec-
trum of the crude mixture featured a band at 1929
cm−1, consistent with successful delivery of cyanide ion
to 5. The blue salt [NnBu4][(NC)Mo(N[Ad]Ar)3] (8) was
isolated in 91% yield by removal of the THF solvent
and subsequent thorough washing of the crude salt with
pentane (see Fig. 1).
2. Results and discussion
2.1. Reaction of Mo(N[R]Ar)3 (1) with [NnBu4][CN]
An initial strategy for cyanide cleavage involved the
addition of [NnBu4][CN] to 1 in tetrahydrofuran solu-
tion. Addition of [NnBu4][CN] (one equivalent) to a
solution of 1 gave a rapid reaction in which a dramatic
purple color appeared, the color persisting in the cold
but fading to brown when the mixture was allowed to
age at 25°C.
Probe experiments were performed in order to test
the viability of 8 as a precursor to the unsymmetrical
bridged dimolybdenum species [NnBu4][(Ar[Ad]N)3-
Mo(m-CN)Mo(N[R]Ar)3] (10). Accordingly,
1 was
added to 8 in THF under a dinitrogen atmosphere. The
result was an unexpected nexus with the N2 chemistry
of 1. Infrared spectra of the reaction mixture revealed
the gradual appearance of a band at 1766 cm−1, this
band being a spectroscopic signature for the well-char-
acterized N2-complex anion derived from 1 [22], here
being generated as the salt [NnBu4][(N2)Mo(N[R]Ar)3]
(11). It is known that, under N2, complex 1 is in
equilibrium with its neutral dinitrogen complex
(N2)Mo(N[R]Ar)3 (12), which in turn is readily reduced
to its monoanion [22]. Here the implication is that the
cyanide salt complex 8 acts as the reducing agent,
giving rise to [NnBu4][(N2)Mo(N[R]Ar)3] (11) along
with the new neutral molybdenum(IV) cyanide complex
(NC)Mo(N[Ad]Ar)3 (13). As described below, the inde-
pendent synthesis of the neutral molybdenum(IV)
cyanide 13 was accomplished via chemical oxidation of
[NnBu4][(NC)Mo(N[Ad]Ar)3] (8) with both ferrocenium
triflate [30] and silver triflate.
2
Spectroscopic monitoring by H-NMR reflected the
aforementioned temperature-dependent color changes
in that initial resonances in the 0–7 ppm range were
observed to decay and be replaced by new signals. A
compound exhibiting a resonance at ca. 6 ppm was
tentatively assigned as the salt [NnBu4][(NC)Mo-
(N[R]Ar)3] (6). Repeated pentane extraction of the
crude brown residue (obtained upon solvent removal)
removed pentane-soluble impurities, leaving behind
small amounts of the desired intense blue salt
[NnBu4][(NC)Mo(N[R]Ar)3] (6). The IR spectrum
(THF, KBr) of 6 featured an intense wCN stretch at 1941
2
cm−1, while a H-NMR signal at 6 ppm was observed
for the complex. Efforts to improve the synthesis of 6
by slow addition of 1 to a cold solution of [NnBu4][CN]
were successful in part, but unidentified decomposition
pathways remained problematic.

26
J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
dinuclear cyanide-bridged assemblies have been dis-
cussed [31]. In addition, the relationship between wCN
band intensity and p-bonding in the complexes has
been explored for hexacyanometalates [32].
2.3. Synthesis and characterization of
(NC)Mo(N[R]Ar)3 (14)
In order to achieve a close comparison with the
well-studied N2-cleavage reaction [1–5], the synthesis of
the dinuclear cyanide-bridged species (m-CN){Mo(N-
SQuID magnetometry data (5–300 K) were collected
for 14, yielding a veff of 2.81 vB, consistent with assign-
ment of the molybdenum(IV) complex as a ground
state triplet. This is the expected result if the complex
adopts C3 symmetry, whereupon the d2 configuration is
represented by (e)2 or (dxz, dyz)2 A molecular orbital
analysis focusing on the embedded linear NCMo
framework of 14 gives the same result, with an indi-
cated (p0)4(p1)2(p2)0 configuration (Section 2.10). A
cyclic voltammetry study of cyanide 14 revealed re-
[R]Ar)3}2 (15, R=C(CD3)2CH3) was required.
A
straightforward route to 15 was achieved via the neutral
molybdenum(IV) cyanide complex (NC)Mo(N[R]Ar)3
(14), which in turn was obtained by cyanide treatment
of the green molybdenum(IV) iodide complex
IMo(N[R]Ar)3 (16). Iodide 16 itself is obtained in high
yield upon treatment of 1 with 0.5 equivalents of I2.
Fig. 2 summarizes the reaction chemistry involving
(NC)Mo(N[R]Ar)3 (14).
versible waves at −0.19 and −1.9
V (THF,
Addition of a chilled solution of [NnBu4][CN] to a
cold, brilliant green ether solution of 16 resulted (1 h)
in smooth substitution of iodide by cyanide. Cyanide 14
[NnBu4][PF6], versus [FeCp2]+/0), assigned respectively
as the reversible oxidation and reduction of 14. The
observed reduction potential of −1.9 V for neutral
molybdenum(IV) cyanide 14 can be compared with that
for the neutral molybdenum(III) N2 complex 12 (−1.7
V) [22]. The apparent contradiction, that molybdenum-
(IV) is more difficult to reduce than molybdenum(III),
can be reconciled in terms of the more p-acidic nature
of dinitrogen vis-a`-vis cyanide.
2
exhibited a single H-NMR signal at 31 ppm, readily
distinguished from the resonance for iodide 16 at 6
ppm. One equivalent of [NnBu4][I] precipitated from the
red reaction mixture upon addition of pentane, and the
neutral molybdenum(IV) cyanide 14 was obtained as a
red, semi-crystalline solid in 77% yield. Like iodide 16,
cyanide 14 was sparingly soluble in pentane and ether
when pure. A wCN stretch was not observed for 14 by
IR spectroscopy. Factors influencing the cyanide
stretching frequency in mononuclear complexes and in
That cyanide 14 exhibits an electrochemically re-
versible reduction is taken to be an indication that
[NnBu4][(NC)Mo(N[R]Ar)3] (6) is stable when prepared
in the absence of 1. If this is correct, then the low yield
Fig. 1. Reaction scheme illustrating the synthesis of neutral cyanomolybdenum (IV) complex 13 and its incorporation into dinuclear-m-cyanide
complexes. Isolated yields of new compounds: 8, 91%; 21, 53%; 23, 30%; 13, 74%.

J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
27
Fig. 2. Reaction scheme illustrating the synthesis of dimolybdenum–m-cyanide 15 and its elimination of a C(CD3)2CH3 substituent under reducing
conditions. Isolated yields of new compounds: 14, 77%; 15, 80%; 19, 50%. Salts 17 and 18 are postulated intermediates and not isolated
compounds.
realized in the synthesis of 6 can be traced to the
intervention of dinitrogen chemistry, it being the case
that all reactions described thus far were carried out
under an N2 atmosphere. These suppositions are consis-
tent with the observations put forward at the end of
Section 2.2, and also with the observed reduction po-
tentials for the neutral cyanide and dinitrogen com-
plexes of 1.
are equivalent, exhibits a single resonance at 13 ppm.
The cyanide-bridged neutral complex 15 possesses one
less electron than its N2-bridged counterpart 9, and is
expected to have
a
doublet ground state:
(p0)4(p1)4(p2)1(p3)0 (see Section 2.10 for a discussion of
the nomenclature used here to denote the configura-
tion). Also, given its electronic configuration, dinuclear
cyanide 15 is expected to exhibit a first-order Jahn–
Teller distortion, giving it a preferred symmetry lower
than C3. Compound 15 is a rare example of the di-
molybdenum–m-cyanide motif [33], some examples of
which have originated in conjunction with molybdenum
sulfide cubane cluster chemistry [34,35]. Examples of
polynuclear tungsten [36,37] and chromium [38] m-
cyanides also are known.
2.4. Synthesis of (v-CN){Mo(N[R]Ar)3}2 (15)
Having in hand the neutral cyanide complex 14, it
became possible to assemble a neutral cyanide-bridged
dimolybdenum species by reaction with 1. Thus, addi-
tion of 1 to an ether solution of 14 resulted in a rapid
color change. The desired cyanide-bridged species (m-
CN){Mo(N[R]Ar)3}2 (15) was violet in color, and easily
isolated and purified due to its low solubility in both
ether and pentane.
A UV–vis spectrum of 15 exhibited an absorbance
maximum at 530 nm, comparable to the characteristic
absorbance at 547 nm for purple (m-N2){Mo(N-
[R]Ar)3}2. SQuID magnetometry data for 15 (veff
=
2
1.61 vB, 5–300 K) are consistent with one unpaired
electron, as expected. Cyclic voltammetry showed that
15 is reversibly oxidized at −1.2 V and reversibly
reduced at −2.2 V. The reduction wave can be com-
pared with that exhibited for dinitrogen complex
A H-NMR spectrum of 15 showed two signals in a
1:1 ratio, one at 6.6 ppm, the other at 3.3 ppm,
consistent with a linear cyanide-bridged dimolybdenum
formulation. The related dinuclear dinitrogen complex
(m-N2){Mo(N[R]Ar)3}2 (9), where both metal centers

28
J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
9 at −2.4 V [22]. Since cyanide-bridged 15 is formally
a dimolybdenum(III/IV) complex, while dinitrogen-
bridged 9 is formally a dimolybdenum(III/III) complex,
it is reasonable that 15 should be more readily reduced.
The monoanion derived from one-electron reduction of
15 is isoelectronic with neutral 9.
the molybdenum center bonded to the cyanide nitro-
gen. In one resonance structure for imido–m-cyanide
19, both molybdenum centers are formally molybde-
num(VI), the highly reduced cyanide behaving as an
imido substituent to one Mo center, and as an alkyli-
dyne substituent to the other Mo center.
It is likely that the reduction of m-cyanide 15 by
sodium amalgam is reversible, given the reduction po-
tential of −2.2 V for the complex (see above). A cyclic
voltammetry study of imido–m-cyanide 19 showed the
complex to be highly resistant to reduction, it being the
case that an irreversible wave was observed at −3.2 V.
The latter wave represented the only reduction mani-
fested by the complex. Sodium amalgam (0.5%) is not
competent to effect such a high-potential reduction so it
therefore stands to reason that the putative anion of 18
would lose an electron, along with the sodium counte-
rion, under the conditions of the synthesis of imido–m-
cyanide 19.
2.5. Chemical reduction of (v-CN){Mo(N[R]Ar)3}2
t
(15): Bu-radical ejection instead of cyanide clea6age
Paramagnetic (m-CN){Mo(N[R]Ar)3}2 (15) under-
went reaction upon treatment with a slight excess of
sodium amalgam over 1 h, giving rise to a single
diamagnetic product. Monitoring of the reaction mix-
ture by 2H-NMR spectroscopy over 1 h evinced a
smooth decay of the two signals corresponding to
m-cyanide 15 and concomitant emergence of several
new signals in the 0–10 ppm range, the data being
indicative of one or more reduction products. After the
Since it is postulated that tBu radical is ejected during
the course of formation of 19, an attempt was made to
observe hexamethylethane, isobutylene, and isobutane.
The latter are the conproportionation and dispropor-
1
reaction was deemed complete, a H-NMR spectrum of
a crude aliquot revealed that a single diamagnetic spe-
cies had been formed in approximately 90% yield. This
deep-orange species was pentane soluble and showed
three distinct sets of resonances for aryl groups (Ar=
3,5-C6H3Me2) in a 3:2:1 ratio. That the new complex
exhibited no reaction with 12-crown-4, and that it was
rather lypophilic suggested that a sodium cation was
not likely to be present as part of the complex. In other
words, the new orange product was a neutral species,
not a salt.
t
tionation products, respectively, of the Bu radical [41–
43]. Accordingly, the 13C-NMR spectrum of the
volatiles from the reaction was acquired, but only the
solvent, tetrahydrofuran (THF), was observed. It is
t
likely that Bu radical reacts with THF solvent readily,
thus being diverted from undergoing disproportiona-
tion and conproportionation, and is converted to less
volatile products. To get around this, the reduction of
15 was attempted in C6D6, in the hope of identifying
isobutane, isobutylene, and hexamethylethane. How-
ever, no reaction was observed in this solvent over a 48
h period. Therefore the statement that tBu radical
ejection is a critical step in the mechanism of formation
of 19 is not firmly established, though it remains a
well-documented decomposition pathway for the
ꢂN(R)Ar ligand [39,40].
The first-formed species in the reduction of m-cyanide
15 is logically anticipated to be [Na(THF)x] [(m-
CN){Mo(N[R]Ar)3}2] (17), but the data pertaining to
the isolated material indicate that the anionic compo-
nent of 17 has undergone both desymmetrization and a
loss of counterion. Desymmetrization is explicable in
terms of fragmentation at one of the six N-tBu bonds.
t
Such a fragmentation leads to ejection of a Bu group,
likely as the radical, and to formation of an imidometal
functionality (e.g. MoꢁNAr, where Ar=3,5-C6H3Me2).
If the new imidometal-containing anion, namely
[Na(THF)x][(Ar[R]N)2(ArN)Mo(m-NC)Mo(N[R]Ar)3]
(18), is a highly-reducing species then formal loss of a
sodium atom would convert it to the neutral imido–m-
cyanide complex (Ar[R]N)2(ArN)Mo(m-NC)Mo(N-
[R]Ar)3 (19). Therefore. the orange complex formed in
high yield upon exposure of m-cyanide 15 to sodium
amalgam is formulated as imido–m-cyanide 19. The
overall transformation amounts to sodium-amalgam
catalyzed ejection of a tert-butyl radical from 15.
It has been documented previously that molybde-
num(V) complexes possessing tert-butyl substituents in
a position a to the metal can lose tert-butyl radical,
giving rise to oxomolybdenum(VI) or imidomolybde-
num(VI) functionalities [39,40]. Therefore it follows
that the freshly-formed imido group of 19 resides on
The electrochemistry of imido–m-cyanide 19 was in-
vestigated further, with the result that a reversible wave
was observed at −2.2 V, presumably corresponding to
the oxidation of 19 to its monocation. The extremely
low reduction potential of −3.2 V for 19 is approxi-
mately 1 V more negative than that exhibited by its
precursor, neutral m-cyanide 15. These electrochemical
data allow us to formulate a model illuminating the
observed conversion of 15–19 under reducing condi-
tions. Sodium amalgam (0.5%) readily reduces 15 to its
monoanion, an electrochemically observable species. In
the actual reaction mixture, this species is a salt such as
17, and undergoes a homolytic CꢂN bond cleavage
t
reaction to generate Bu radical and an extremely re-
ducing anion, the anionic component of salt 18. Salt 18
is oxidized by 15 or by mercury. The overall transfor-
mation, as indicated above, is a sodium amalgam cata-
lyzed ejection of a tBu group from neutral m-cyanide 15.

J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
29
2.6. Synthesis of [Li(THF)x][(NC)Mo(N[Ad]Ar)3] (20)
conjunction with the three 1-Ad substituents to form a
shroud encapsulating the cyanide bridge.
A lithium for tetrabutyl ammonium cation replace-
ment was effected by addition of excess lithium chloride
to a blue THF solution of 8. A color change from blue
to intense violet occurred gradually, concomitant with a
change in the IR absorbance of the mixture. Over a
Mixing of neutral adamantyl–cyanide 13 with a stoi-
chiometric amount of 1 in tetrahydrofuran produced an
intense violet color, very similar to the color of the
all-tBu neutral bridged-cyanide complex 15. A 2H-
NMR spectrum of the reaction mixture evinced a single
broad resonance at 6.8 ppm (Dw1/2=60 Hz), reflecting
attachment of the deuterated molybdenum complex 1
to the non-deuterated Ad-substituted species 13, giving
the mixed cyanide-bridged complex (Ar[Ad]N)3Mo(m-
CN)Mo(N[R]Ar)3 (21, R=C(CD3)2CH3). Compound
21, which was found to be only sparingly soluble in
alkane solvents, was obtained in 53% isolated yield.
period of several hours the stretch at 1929 cm−1
,
corresponding to precursor 8, decayed and was re-
placed by two bands appearing at 1934 and 1915 cm−1
.
The product is assigned as lithium molybdenum
cyanide 20.
The empirical observation of two wCN stretches for 20
is explicable in terms of two isomers of the complex, or
in terms of its adopting an unsymmetrical dimeric
structure. In either case, interactions between the
lithium cation and the 3,5-C6H3Me2 substituents may
be an essential component of the structure, as noted in
other instances [9,44]. The increased hydrocarbon solu-
bility of lithium salt 20 as compared with the tetrabutyl
ammonium salt 8 is likewise an indication that the
lithium cation is tightly associated with its counterion.
2.9. A cyanide-bridged 6anadium/molybdenum complex
One of the attractive features of monocyanometal-
lates as in [NnBu4][(NC)Mo(N[Ad]Ar)3] (8) is the poten-
tial for forming heterodinuclear cyanide-bridged
complexes via metathetical reactions. This type of reac-
tivity is illustrated by the reaction of 8 with the previ-
ously described [45] vanadium(III) iodide complex
2.7. Chemical oxidation of the adamantyl-substituted
anionic cyanide complex (8)
IV(N[R]ArF)2
(22,
ArF=2,5-C6H3FMe,
R=
C(CD3)2CH3). The indicated reaction generated the
cyanide-bridged complex (ArF[R]N)2V(m-NC)Mo(N-
[Ad]Ar)3 (23), the reaction being carried out at low
temperature for best results. Vanadium/molybdenum
cyanide 23 was formed in ca. 60% yield, as estimated by
Oxidation of [NnBu4][(NC)Mo(N[Ad]Ar)3] (8) with
ferrocenium triflate [30] resulted in the formation of
ferrocene along with the neutral cyanide complex 13.
Like its Bu-substituted analog 14, neutral cyanide 13 is
red in color and rather insoluble in pentane, the latter
t
a
2H-NMR spectrum of the crude reaction mixture.
property facilitating its separation from ferrocene.
As was the case for Bu-substituted 14, an IR spec-
trum of the neutral Ad-substituted cyanide 13 did not
exhibit a wCN stretch.
The isolated yield of complex 23 was 30%, subsequent
to crystallization from ether. A magnetic susceptibility
measurement by the method of Evans gave a veff value
of 1.70 vB for 23, consistent with an S=1/2 system.
An X-ray structural investigation of 23 revealed a
MoꢂCꢂNꢂV linkage exhibiting a slight bend at the
cyanide nitrogen. Another salient feature of the solid
state structure of 23 is the unusually short V–F(2)
t
2.8. Steric tuning of the cyanide-bridged dimolybdenum
system
,
The results given above confirm that the cyanide
ligand is capable of serving as a bridge to join together
two molecules of Mo(N[R]Ar)3 (1). Also, it has been
shown [22] that the more sterically shielded three-coor-
dinate molybdenum(III) complex Mo(N[Ad]Ar)3 (5)
does not form a dinuclear dinitrogen-bridged complex,
and consequently does not split dinitrogen, in stark
contrast to the results obtained for compound 1. In
keeping with the N2 result with 5, it was found here
that the adamantyl-substituted cyanide 13 exhibits no
reaction with Mo(N[Ad]Ar)3 (5). With the cyanide
chemistry uncovered in the present work, it became
possible to see if the Mo(N[R]Ar)3 (1) fragment could
be linked to the Mo(N[Ad]Ar)3 (5) fragment by a
cyanide bridge. Given the known structural preferences
for complexes of 5 and 1, formation of the indicated
interaction (2.147 A), F(2) being an ortho-fluorine com-
ponent of one of the two ArF substituents. Such chela-
tion of fluoroaryl functional groups in the context of
fluoroaryl amido chemistry has been described in some
recent reports [46,47], and should be viewed as similar
to an agostic interaction.
The MoꢂCꢂNꢂV tetratomic unit in 23 can be viewed
in p-delocalized molecular orbital terms as being a
(p0)4(p1)4(p2)1(p3)0 system (see Section 2.10). This
configuration is consistent with the observed magnetic
moment. Also consistent with the observed magnetic
moment would be a Lewis description for the molecule
in which molybdenum is assigned the +6 oxidation
state and vanadium the +4 oxidation state. In such a
description the molybdenum forms a formal triple bond
to the cyanide carbon, while the vanadium forms an
imido-like linkage to the cyanide nitrogen. That the
t
tBu/1-Ad linked system requires the three Bu groups in

30
J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
cyanide ligand is indeed highly reduced in complex 23
can be inferred from the relevant bond distances found
in the structural investigation (Fig. 3).
For a tetraatomic linear system, for example two
molecules of dinitrogen taken together, or dimolybde-
num–m-cyanide 15, there are four p energy levels, each
being doubly-degenerate and referred to as p0 through
p3 (Fig. 5).
The configuration is arrived at by determining the
number of p-type electrons in the system in question. In
the case of (NC)Mo(N[R]Ar)3 14, it is seen that the
system is formally d2, and that the cyanide ligand
contributes four p-symmetry electrons, for a total of six
p-electrons. Thus the configuration for 14 is
(p0)4(p1)2(p2)0, as indicated in Table 1. A triplet state
therefore is indicated for 14, consistent with its mag-
netic behavior.
A final point to be made is that the most stable
systems are those with the (p0)4(p1)4(p2)0 configurations
for triatomic systems, and those with the
(p0)4(p1)4(p2)0(p3)0 configurations for tetraatomic sys-
tems. This can be understood with reference to CO2 as
a very stable diamagnetic closed-shell representative of
this class.
Similar considerations of electronic structure apply to
dinitrogen complexes in the molybdenum triami-
doamine systems of Schrock and co-workers [49], and
Bercaw has given a particularly lucid description of the
bonding in dinitrogen-bridged dinuclear systems [50].
2.10. Bonding considerations for multiply bonded
systems deri6ed from 1
It has been mentioned previously [5] that
Mo(N[R]Ar)3 (1) is isolobal [48] with a ground state
nitrogen atom. This isolobal analogy permits a first-or-
der analysis of the bonding in linear multiply bonded
systems that have Mo(N[R]Ar)3 (1) as a terminus.
The analysis given here presumes a linear system of
atoms, and stems from the equivalence of the x and y
coordinates in a system (such as 1) that is threefold
symmetric about the z coordinate. Attention is focused
on the p system of molecular orbitals, since this is
where any unpaired electrons are expected to reside.
For a triatomic system such as CO2, or for a system
such as (NC)Mo(N[R]Ar)3 (14) in which a triatomic
NCMo system is embedded, it can be seen (Fig. 4) that
there are three p energy levels. These are each doubly-
degenerate due to the equivalence of the x and y
coordinates, and they are referred to as p0, p1, and p2,
the subscript being indicative of the number of inter-
atomic nodes in each p-type molecular orbital.
Fig. 3. Thermal ellipsoid representation of the structure of (ArF[R]N)2V(m-NC)Mo(N[Ad]Ar)3 (23) from a single-crystal X-ray diffraction study.
,
Selected bond distances (A) and angles (°): MoꢂC, 1.847(13); VꢂN, 1.799(11); CꢂN, 1.260(14); VꢂF(2), 2.148(7); MoꢂN(1), 1.950(11); MoꢂN(3),
1.933(12); MoꢂN(5), 1.971(12); VꢂN(2), 1.962(10); VꢂN(4), 1.905(9); NꢂCꢂMo, 176.1(9); CꢂNꢂV, 169.6(9); CꢂMoꢂN(1),N(3),N(5), avg.=100.8(4);
N(2)ꢂVꢂF(2), 121.9(4); NꢂVꢂN(4), 115.0(5).

J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
31
Table 1
p-Bonding scheme for linear multiply bonded systems a
Compound
Configuration
Spin
state
[Na(THF)x][(m-CN){Mo(N[R]Ar)3}2] (17) (p0)4(p1)4(p2)2(p3)0 Doublet
CO2 (p0)4(p1)4(p2)0
N2O (p0)4(p1)4(p2)0
Singlet
Singlet
Triplet
Doublet
Doublet
Singlet
(NC)Mo(N[R]Ar)3 (14) (p0)4(p1)2(p2)0
(N2) Mo(N[R]Ar)3 (12) (p0)4(p1)3(p2)0
[NnBu4][(NC)Mo(N[Ad]Ar)3] (8) (p0)4(p1)3(p2)0
[NnBu4][(N2)Mo(N[R]Ar)3] (11) (p0)4(p1)4(p2)0
2 N2 (p0)4(p1)4(p2)0(p3)0 Singlet
(m-CN){Mo(N[R]Ar)3}2 (15) (p0)4(p1)4(p2)1(p3)0 Doublet
[Na(THF)x][(m-CN){Mo(N[R]Ar)3}2] (17) (p0)4(p1)4(p2)2(p3)0 Triplet
(m-N2){Mo(N[R]Ar)3}2 (9) (p0)4(p1)4(p2)2(p3)0 Triplet
2 NMo(N[R]Ar)3 (4) (p0)4(p1)4(p2)0(p3)0 Singlet
a Configurations for triatomic systems are derived from Fig. 4, while those
for tetraatomic systems derive from Fig. 5.
Fig. 4. Depiction of the p-symmetry molecular orbitals of the tri-
atomic NCMo moiety embedded within the (NC)Mo(N[R]Ar)3
molecule 14. The designation p0 refers to the p molecular orbital with
zero interatomic nodes, and similarly the designation p1 refers to the
p molecular orbital with one interatomic node, and so on. The
contour diagrams shown are intended to be schematic and are derived
from an extended Hu¨ckel molecular orbital calculation [53] on
(N2)Mo(N[R]Ar)3 (12). Orbitals with the same general features are
anticipated for any system having an electronegative diatomic
molecule linearly coordinated to a transition metal fragment such as
Mo(N[R]Ar)3 (1).
3. Conclusions
Inspired by the dinitrogen chemistry of the three-co-
ordinate molybdenum(III) complex Mo(N[R]Ar)3 (1)
[1–3,5], this work has explored the cyanide chemistry of
1 [51] and its more sterically hindered adamantyl-
substituted analog 5 [21,22]. Cyanomolybdenum and
cyanodimolybdenum complexes have been prepared in
the course of this study, the main goal of which was to
probe the possibility that two equivalents of 1 might act
cooperatively to effect the six-electron reductive cleav-
age of cyanide ion. Such a cleavage reaction, if success-
ful, would generate one equivalent each of the known
nitride [3] and carbide [9] complexes, NMo(N[R]Ar)3
(4) and [M][CMo(N[R]Ar)3] (3), which have been syn-
thesized independently. In the experiments that bear
most strongly on this key issue it was found that a
ligand degradation pathway, namely the ejection of a
t
ligand Bu group, takes precedence over cyanide reduc-
tive cleavage. The observed CꢂN cleavage reaction,
generating an imidometal unit, represents another ex-
ample of a growing class of reactions in which a
metalꢂligand multiple bond is formed via formal radical
expulsion [39,40].
It is difficult to compare directly the known dinitro-
gen cleavage reaction with the cyanide splitting reaction
envisioned here, primarily because the anionic carbide
complex isoelectronic with nitride 4 requires a counter-
ion. The role of the said counterion in a hypothetical
cyanide cleavage reaction is likely to be critical for
stabilization of the developing negative charge on the
carbide carbon with progress along the reaction
coordinate.
Fig. 5. Depiction of the p-symmetry molecular orbitals of the te-
traatomic MoNCMo moiety embedded within the (m-
CN){Mo(N[R]Ar)3}2 molecule 15. The designation p0 refers to the p
molecular orbital with zero interatomic nodes, and similarly the
designation p1 refers to the p molecular orbital with one interatomic
node, and so on. The contour diagrams shown are intended to be
schematic and are derived from an extended Hu¨ckel molecular orbital
calculation [52] on (m-N2){Mo(N[R]Ar)3}2 (9). Orbitals with the same
general features are anticipated for any system having an electronega-
tive diatomic molecule linearly bridging two transition metal frag-
ments such as Mo(N[R]Ar)3 (1).
Methods were developed for the synthesis of mono-
cyanometallate derivatives, potentially useful synthons
for the construction of heterodinuclear cyanide-bridged

32
J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
assemblies. The latter approach was illustrated via the
assembly of a vanadium/molybdenum cyanide-bridged
complex.
0.5 M [NnBu4][PF6]. In a typical procedure 5 mg of the
complex was dissolved in 0.75 ml of clean THF. To this
solution was added 0.75 ml of 1.0 M THF solution of
[NnBu4][PF6]. A platinum disk (1.6 mm diameter, Bio-
analytical systems), a platinum wire, and a silver wire
were employed as the working electrode, the auxiliary,
and the reference, respectively. The electrochemical re-
sponse was collected with the assistance of an Eco-
Chemie Autolab potentiostat (pgstat20) and the GPES
4.3 software. An IR correction drop was always em-
ployed due to the high resistance of the solutions. A
typical resistance value measured with the positive feed-
back technique for these solutions was 975 V. All of the
potentials are reported against the [FeCp2]+/0 couple
measured in the same solution.
4. Experimental
4.1. General considerations
Unless stated otherwise, all operations were per-
formed in a Vacuum Atmospheres dry box under an
atmosphere of purified nitrogen, or using standard
Schlenk techniques under an argon or dinitrogen atmo-
sphere. Anhydrous ether and toluene were purchased
from Mallinckrodt; n-pentane and n-hexane were pur-
chased from EM Science. Ether was purified according
to a published procedure [52]. Aliphatic hydrocarbon
solvents were distilled under a nitrogen atmosphere
from very dark blue to purple sodium benzophenone
ketyl solubilized with a small quantity of tetraglyme.
Distilled solvents were transferred under vacuum into
Teflon-stopcocked glass vessels and stored, prior to use,
in a vacuum atmospheres dry box. C6D6 was degassed
4.3. Synthesis of IMo(N[R]Ar)3 (16)
To a solution of 1 (0.3494 g, 0.5435 mmol) in ether
(10 ml) was added iodine (0.0706 g, 0.2782 mmol, 0.51
equivalents) as a solution in ether (5 ml). The resulting
deep green solution was stirred for 30 min, whereupon
the solvent was removed in vacuo. The green product
was recrystallized from ether (0.3461 g, 0.4496 mmol,
83%). M.p. 122–124°C. 2H-NMR (46 MHz, ether):
,
and dried over activated 4 A molecular sieves and
transferred under vacuum into a storage vessel. Sieves
,
(4 A), Celite and alumina were activated in vacuo
overnight at a temperature above 180°C. Mo(N[R]Ar)3
(1) [3], ferrocenium triflate [30], LiN(Ad)Ar [23], and
Mo(N[Ad]Ar)3 (5) [21,22], were prepared according to
published procedures. Other chemicals were purified
and dried by standard procedures or were used as
received. Infrared spectra were recorded on a Bio-Rad
135 Series FTIR spectrometer. UV–vis spectra were
recorded on a Hewlett–Packard 8453 diode-array spec-
trophotometer. 1H- and 13C-NMR spectra were
recorded on Varian VXR-500, Varian XL-300, or
Varian Unity-300 spectrometers. 1H- and 13C-NMR
chemical shifts are reported with reference to solvent
resonances (residual C6D5H in C6D6, 7.15 ppm; C6D6,
128.0 ppm; CHCl3 in CDCl3, 7.24 ppm; CDCl3, 77.0
1
l=6.36 (Dw1/2=11 Hz). H-NMR (300 MHz, C6D6):
l=6.58 (C(CD3)2CH3), 4.62 (aryl ortho CꢂH). 3.56
(aryl para CꢂH), 1.15 (aryl CH3). veff (300 MHz, C6D6,
25°C): 1.94 vB. Anal. Calc. for C36H36D18MoN3I: C,
56.17; H, 7.07; N, 5.46. Found: C, 56.15; H, 6.90; N,
5.34%.
4.4. Synthesis of (NC)Mo(N[R]Ar)3 (14)
A solution of 494 mg [NnBu4][CN] (1.841 mmol) in 3
ml of THF was chilled to −35°C and added via pipette
to a chilled, brilliant green solution containing 1.181 g
IMo(N[R]Ar)3 (16, 1.534 mmol) in 20 ml of THF. The
reaction mixture was allowed to warm to ambient
temperature and was stirred for 2 h. At this time, 30 ml
of pentane was added to precipitate the salts, and the
slurry was then filtered through a sintered glass frit.
The red filtrate was dried to a solid, and addition of 25
ml of OEt2 partially dissolved the solid. The undis-
solved solid was collected by filtration as a first pure
crop (502 mg), and storage of the ethereal filtrate
afforded a second crop (284.6 mg) for an overall 77%
yield of pure (NC)Mo(N[R]Ar)3 (14). 2H-NMR (76
MHz, THF, 25°C): l=31.0 ppm (s, Dw1/2=13 Hz,
C(CD3)2CH3). SQuID (5–300K): veff=2.81 vB. veff
(300 MHz, 25°C. C6D6)=2.46 vB. An FT-IR spectrum
2
ppm). H-NMR chemical shifts are reported with re-
spect to external C6D6 (7.15 ppm). Solution magnetic
susceptibilities were determined by 1H-NMR at 300
MHz using the method of Evans. Routine coupling
constants are not reported. Combustion analyses (C, H,
and N) were performed by Microlytics, Southdeerfield
MA. X-ray diffraction data were collected on a Siemens
Platform goniometer with a charge coupled device
(CCD) detector. Structures were typically solved by
direct methods (SHELXTL V5.0, G.M Sheldrick and
Siemens Industrial Automation, Inc., 1995) unless oth-
erwise noted.
4.2. Electrochemical measurements
did not show
a wCN stretch. Anal. Calc. for
C37H36D18N4Mo: C. 66.44; H, 8.14; N, 8.37. Found: C,
66.19; H, 8.20; N, 7.99%.
The electrochemical measurements were performed in
THF solution containing the desired compounds and

J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
33
4.5. Synthesis of (v-CN){Mo(N[R]Ar)3}2 (15)
CꢂH), 6.41 (imido aryl ortho CꢂH), 2.23 (aryl CH3),
2.16 (imido aryl CH3), 1.60 (C(CD3)2CH3).
(Ar[R]N)3Mo: 6.70 (aryl para CꢂH), 6.08 (aryl ortho
CꢂH), 2.13 (aryl CH3), 1.62 (C(CD3)2CH3). 13C-NMR
(125.66 MHz, C6D6, 25°C): 153.4, 151.2, 137.8, 137.4,
131.3, 129.1, 127.8, 127.4. 159.8, 122.2, 116.3, 62.7,
61.5, 34.6, 32.7, 22.0, 21.9, 21.8. Anal. Calc. for
C70H69D30N7Mo2: C. 66.69; H, 7.95; N, 7.78. Found: C,
65.99; H, 8.09; N, 7.48%.
Solid orange Mo(N[R]Ar)3 (1, 378 mg, 0.588 mmol)
was added to
a
stirring red solution of
(NC)Mo(N[R]Ar)3 (14, 393.1 mg, 0.588 mmol) in 10 ml
of a 4:1 ether/benzene mixture at 25°C. The mixture,
which attained a brilliant violet color quickly upon
mixing, was stirred for 1 h after which time the solvent
was removed in vacuo. The resulting solid was collected
on a sintered glass frit and washed thoroughly with
pentane to remove a dark brown, pentane-soluble im-
purity. The violet solid which remained was sparingly
soluble in both pentane and in ether. It was dried to a
mass of 618 mg (80% yield) and was found to be
4.7. Synthesis of [NnBu4][(NC)Mo(N[Ad]Ar)3] (8)
A solution of 419 mg [NnBu4][CN] (1.560 mmol) in
15 ml of THF was added to a stirring solution of 1.340
g Mo(N[Ad]Ar)3 (5, 1.560 mmol) in 20 ml of THF. The
reaction mixture turned from the orange color of 5 to
an intense blue upon mixing. Stirring was continued for
10 min after which time tire solvent was removed in
vacuo, leaving a blue residue. Pentane (10 ml) was
added to the crude residue to form a slurry, filtration of
which on a sintered glass frit provided a blue, pentane-
insoluble solid. This solid was washed×2 with 5 ml of
a 1:1 pentane–benzene mixture, and then it was washed
with pentane until the washings were colorless. The
blue powder. [NnBu4][(NC)Mo(N[Ad]Ar)3] (8), was
2
spectroscopically and analytically pure. H-NMR (46
MHz, THF, 25°C): l=6.6 ppm (s, Dw1/2=36 Hz,
C(CD3)2CH3), 3.3 ppm (s, Dw1/2=21 Hz, C(CD3)2-
CH3). SQuID (5–300K): veff=1.61 vB. UV−vis
(THF): u=350 nm, 425 nm, 530 nm. FTIR (THF,
KBr): 1575 cm−1 (overlapping with ligand aryl CC
stretch). Anal. Calc. for C73H72D36N7Mo2: C, 66.84; H,
8.30; N, 7.47. Found: C, 66.53; H, 8.32; N, 7.30%.
4.6. Synthesis of
(Ar[R]N)2(ArN)Mo(v-NC)Mo(N[R]Ar)3 (19)
dried to a weight of 1.605 g (91% yield). FTIR: wCN
=
Solid (m-CN){Mo(N[R]Ar)3}2 (15, 377 mg, 0.287
mmol) was added to a 0.40% w/w stirring sodium
amalgam (9.5 mg Na in 2.38 g Hg, 0.413 mmol Na) in
5 ml of THF under N2. Within 30 min the solution had
turned from the intense red–purple of 15 to a deep
orange colon. An FTIR spectrum after 45 min showed
an intense stretch at 1584 cm−1 likely resulting from
overlapping wCN and wCC vibrations. An aliquot of the
reaction mixture was removed after 1 h, dried in vacuo,
triturated with hexane, and extracted into C6D6 con-
taining hexamethyldisiloxane as an internal standard. A
1H-NMR spectrum of this aliquot showed that
(Ar[R]N)2(ArN)Mo(m-NC)Mo(N[R]Ar)3 had been gen-
erated in approximately 90% yield (based on the start-
ing material 15) by this time. Stirring was continued for
an additional 2 h at 25°C, after which time the solvent
was removed in vacuo and the crude mixture was
triturated with hexane. The resulting product was ex-
tracted into pentane, filtered through Celite, and dried
in vacuo. The product was difficult to recrystallize from
hydrocarbon solvents due to its high lypophilicity, but
was obtained in ca. 50% yield (215 mg) as a spectro-
scopically pure substance by precipitation from hexa-
methyldisiloxane. The orange powder thus obtained
was then recrystallized to afford orange crystals in low
1929 cm−1 (THF, CaF2). Anal. Calc. for
C71H108N5Mo: C, 75.63; H, 9.65; N, 6.21. Found: C,
75.46; H, 9.69; N, 6.15%.
4.8. Reaction of [NnBu4][CN] with Mo(N[R]Ar)3 (1)
Solid white [NnBu4][CN] (181.5 mg, 0.679 mmol) was
dissolved in 10 ml of THF in a scintillation vial.
Crystalline Mo(N[R]Ar)3 (1, 436.7 mg, 0.679 mmol)
was added slowly to the solution in small portions. The
color turned to an inky black–blue quickly, and an
intense purple color was not observed. A crude 2H-
NMR spectrum showed a signal at 6.1 ppm and several
peaks between 0 and 2 ppm. A repetitive pentane
extraction process left behind 100 mg of an insoluble
blue solid which was identified by spectroscopy as
[NnBu4][(NC)Mo(N[R]Ar)3] (6). 2H-NMR (76 MHz,
THF, 25°C): l=6.1 ppm (s, Dw1/2=8 Hz,
C(CD3)2CH3). FTIR: wCN=1941 cm−1 (THF, CaF2).
Elemental analysis was not obtained.
4.9. Synthesis of [Li(THF)x][(NC)Mo(N[Ad]Ar)3] (20)
A slurry of 1.037 g of Mo(N[Ad]Ar)3 (5, 1.2071
mmol) was prepared in 15 ml of THF and cooled to
−35°C. While stirring this mixture, 324.1 mg
[NnBu4][CN] (1.207 mmol) was added to it as a solid, in
one portion. After stirring for 45 min, 51.2 mg of
1
yield. H-NMR (300 MHz, C6D6, 25°C): Assignments
for respective ligands are listed separately for the two
metal centers: (Ar[R]N)2(ArN)Mo: 6.93 (aryl ortho
CꢂH), 6.74 (aryl para CꢂH), 6.45 (imido aryl para

34
J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
LiCl was added to the blue solution, effecting a color
change to blue purple. After 16 h of stirring the mixture
an FTIR spectrum was indicative of incomplete cation
exchange. An additional 375 mg LiCl was added to the
reaction mixture and stirring was continued for an
additional 24 h. The solvent then was removed in
vacuo, and the resulting residue was triturated with
pentane. The powder thereby obtained was extracted
into 25 ml pentane and filtered through celite. Storage
of the pentane filtrate at −35°C afforded 618.1 mg of
0.2157 mmol) was then added. The reaction mixture
turned to an inky violet color rapidly and was stirred
for an additional 15 min at 25°C. Some solid particu-
late material precipitated from solution and 2 ml of
THF was added to achieve homogeneity. Filtration and
storage of the solution at −35°C for 12 h produced
175 mg (53%) of solid burgundy (Ar[Ad]N)3Mo(m-
CN)Mo(N[R]Ar)3 (21). 2H-NMR (76 MHz, THF,
25°C): l=6.8 ppm (s, Dw1/2=60 Hz, C(CD3)2CH3).
FTIR (THF, CaF2): wCN=1575 cm−1 (overlapping
with ligand aryl CC stretch). Anal. Calc. for
C91H108D18N7Mo2: C, 71.53; H, 8.31; N, 6.42. Found:
C, 70.98; H, 8.29; N, 6.40%.
1
red-violet crystals. A H-NMR spectrum showed broad
signals at l=4.45, 3.75, 1.74, 1.50, and 1.40 ppm. An
X-ray crystallographic study of one such crystal was
consistent
with
the
proposed
formulation
[Li(THF)x][(NC)Mo(N[Ad]Ar)3] (20, where x=3). Un-
fortunately, successful refinement of the structure was
hampered by disorder problems involving the tetrahy-
drofuran moieties. Analytical data are presumed to be
indicative of variable amounts of tetrahydrofuran sol-
vent. Anal. Calc. for C67H96N4LiMoO3: C, 72.60; H,
8.73; N, 5.06. Found: C, 70.56; H, 7.61; N, 5.44.
Chemical characterization of 20 was obtained via silver
triflate oxidation (see Section 4.10), which was found
smoothly to produce neutral (NC)Mo(N[Ad]Ar)3 (13).
4.12. Synthesis of (ArF[R]N)2V(v-NC)Mo(N[Ad]Ar)3
(23)
A
chilled (−35°C) green–brown solution of
IV(N[R]ArF)2 (22, 146.4 mg, 0.266 mmol) in ether (3
ml) was added dropwise over ca. 20 s to a chilled slurry
of [NnBu4][(NC)Mo(N[Ad]Ar)3] (8, 300 mg, 0.266
mmol) in 5 ml of a 4:1 THF–ether mixture at −35°C.
The reaction mixture began rapidly to turn orange–
brown and stirring was continued for 1 h at 25°C, after
which time pentane (5 ml) was added to induce salt
precipitation. Filtration of the mixture through celite,
drying in vacuo, and trituration with pentane provided
351 mg of a dark orange–brown powder. The crude
powder thus obtained was ca. 60% pure judging by
2H-NMR spectroscopy. Recrystallization from ether
afforded (ArF[R]N)2V(m-NC)Mo(N[Ad]Ar)3 (23, 105
4.10. Synthesis of (NC)Mo(N[Ad]Ar)3 (13)
Solid blue [NnBu4][(NC)Mo(N[Ad]Ar)3] (8, 24.2 mg,
0.0215 mmol) and orange [FeCp2][O3SCF3] (7.2 mg,
0.0215 mmol) were weighed into a 20 ml scintillation
vial equipped with a magnetic stir bar. Addition of
ether (3.5 ml) with stirring effected a rapid color change
to red. The wCN stretch for starting material 8 decayed
away and no new wCN stretch was observed. Filtration
of the mixture, followed by solvent removal in vacuo,
gave a reddish residue the 1H-NMR spectrum (C6D6) of
which showed Cp2Fe (4.00 ppm) and broad signals for
(NC)Mo(N[Ad]Ar)3 (13) as follows: l=13.5, 8.1, 4.1,
2.5, 1.3,−4.0, −7.8, and −13.6 ppm. The solid mate-
rial was recovered by drying in vacuo. Subsequent
pentane washing of this solid effected the removal of
orange ferrocene and left behind a red powder, that
once dried, was analytically pure (NC)Mo(N[Ad]Ar)3
(13). Anal. Calc. for C55H72N4Mo: C, 74.63; H, 8.20; N,
6.32. Found: C, 74.95; H, 8.38; N, 6.21%.
2
mg, 30%) as a dark orange solid. H-NMR (76 MHz,
OEt2, 25°C): l=17 ppm (s, Dw1/2=600 Hz,
C(CD3)2CH3). veff (300 MHz, 25°C, C6D6)=1.70 vB
FTIR: wCN (tentative)=1615 cm−1 (OEt2, KBr). Anal.
Calc. for C77H90D12F2MoN6V: C, 70.67; H, 7.86; N,
6.42. Found: C, 71.22; H, 7.53; N, 6.16%.
4.13. X-ray structure of
(ArF[R]N)2V(v-NC)Mo(N[Ad]Ar)3 (23)
Deep orange crystals of 23 were grown from an ether
solution at −35°C. The crystals were transferred from
a scintillation vial to a microscope slide and coated
with Paratone N oil (an Exxon product). A dark orange
plate was selected and mounted on a glass fiber using
wax. A total of 11 301 reflections were collected (−85
h515, −18Bk517, −195l519) in the q range of
1.18–20.00° of which 6993 were unique (Rint=0.0626).
The structure was solved by direct methods in conjunc-
tion with standard difference Fourier techniques. A
troublesome molecule of ether was present in the asym-
metric unit and included during refinement (atom labels
C1 C2, C3, C4, O1). This ether molecule did not refine
very well but all atoms in the asymmetric unit were
Compound 13 also was obtained in 74% yield subse-
quent to treatment of the lithium derivative 20 (0.319
mmol) with silver triflate (82 mg, one equivalent) in
ether.
4.11. Synthesis of (Ar[Ad]N)3Mo(v-CN)Mo(N[R]Ar)3
(21)
Red (NC)Mo(N[Ad]Ar)3 (13, 190.9 mg, 0.2157
mmol) was stirred in 10 ml ether and the mixture was
cooled to −35°C. Orange Mo(N[R]Ar)3 (1, 138.6 mg,
t
refined anisotropically. The Bu group connected to

J.C. Peters et al. / Journal of Organometallic Chemistry 591 (1999) 24–35
35
ola, W.M. Davis, C.C. Cummins, J. Am. Chem. Soc. 121 (1999)
accepted for publication.
[23] A.R. Johnson, C.C. Cummins, Inorg. Synth. 32 (1998) 123.
[24] A. Kasani, S. Gambarotta, C. Bensimon, Can. J. Chem. Rev.
Can. Chim. 75 (1997) 1494.
[25] K.B.P. Ruppa, N. Desmangles, S. Gambarotta, G. Yap, A.L.
Rheingold, Inorg. Chem. 36 (1997) 1194.
[26] M. Tayebani, H. Feghali, S. Gambarotta, C. Bensimon,
Organometallics 16 (1997) 5084.
N(4) was disordered and modeled as two positions at
half occupancy. All non-hydrogen atoms were placed in
,
calculated (dCꢂH=0.96 A) positions. The largest peak
and hole in the difference map were 1.043 and −0.453
e A−3, respectively. No absorption correction was ap-
,
plied. The least-squares refinement converged normally
with residuals of R (based on F)=0.1012, ꢀR (based
on F2)=0.2323, and Goodness-of-fit=1.109 based
upon I\2|(I). Crystal data for C81H103F2MoN6V: tri-
[27] K.B.P. Ruppa, S. Gambarotta, G.P.A. Yap, Inorg. Chim. Acta
280 (1998) 143.
(
,
clinic, space group=P1, z=2, a=13.5574(9) A, b=
[28] K.B.P. Ruppa, K. Feghali, I. Kovacs, K. Aparna, S. Gam-
barotta, G.P. A. Yap, C. Bensimon, J. Chem. Soc. Dalton
Trans. (1998) 1595.
[29] D. Reardon, I. Kovacs, K.B.P. Ruppa, K. Feghali, S. Gam-
barotta, J. Petersen, Chem. Eur. J. 3 (1997) 1482.
[30] R.R. Schrock, L.G. Sturgeoff, P.R. Sharp, Inorg. Chem. 22
(1983) 2801.
,
,
16.3838(11) A, c=17.5538(12) A, h=92.6060(10)°,
3
,
i=99.8760(10)°, k=98.0680(10)°. V=3793.7(4) A ,
zcalc. =1.192 g cm−3, F(000)=1444.
[31] B.S. Lim, R.H. Holm, Inorg. Chem. 37 (1998) 4898.
[32] L.H. Jones, Inorg. Chem. 2 (1963) 777.
References
[33] A. Bell, S.J. Lippard, M. Roberts, R.A. Walton, Organometal-
lics 2 (1983) 1562.
[34] K.D. Demadis, S.J. Chen, D. Coucouvanis, Polyhedron 13
(1994) 3147.
[35] K.D. Demadis, D. Coucouvanis, Inorg. Chem. 34 (1995) 436.
[36] H.M. Dawes, M.B. Hursthouse, A.A. Del Paggio, E.L. Muetter-
ties, A.W. Parkins, Polyhedron 4 (1985) 379.
[37] V.G. Albano, L. Busetto, C. Castellari, M. Monari, A. Palazzi,
V. Zanotti, J. Chem. Soc. Dalton Trans. (1993) 3661.
[38] F. Edelmann, U. Behrens, J. Organomet. Chem. 131 (1977) 65.
[39] A.R. Johnson, W.M. Davis, C.C. Cummins, S. Serron, S.P.
Nolan, D.G. Musaev, K. Morokuma, J. Am. Chem. Soc. 120
(1998) 2071.
[1] C.E. Laplaza, C.C. Cummins, Science 268 (1995) 861.
[2] C.E. Laplaza, A.R. Johnson, C.C. Cummins, J. Am. Chem. Soc.
118 (1996) 709.
[3] C.E. Laplaza, M.J.A. Johnson, J.C. Peters, A.L. Odom, E. Kim,
C.C. Cummins, G.N. George, I.J. Pickering, J. Am. Chem. Soc.
118 (1996) 8623.
[4] M.J.A. Johnson, P.M. Lee, A.L. Odom, W.M. Davis, C.C.
Cummins, Angew. Chem. Int. Ed. Engl. 36 (1997) 87.
[5] C.C. Cummins, Chem. Commun. (1998) 1777.
[6] K.M. Neyman, V.A. Nasluzov, J. Hahn, C.R. Landis, N. Ro¨sch,
Organometallics 16 (1997) 995.
[7] J. Hahn, C.R. Landis, V.A. Nasluzov, K.M. Neyman, N. Rosch,
Inorg. Chem. 36 (1997) 3947.
[40] J.C. Peters, A.R. Johnson, A.L. Odom, P.W. Wanandi, W.M.
Davis, C.C. Cummins, J. Am. Chem. Soc. 118 (1996) 10175.
[41] W.A. Pryor, Free Radicals, McGraw-Hill, New York, 1966.
[42] M. Weydert, J.G. Brennan, R.A. Andersen, R.G. Bergman,
Organometallics 14 (1995) 3942.
[43] J.O. Terry, J.H. Futrell, Can. J. Chem. 46 (1968) 664.
[44] M.G. Fickes, A.L. Odom, C.C. Cummins, Chem. Commun.
(Cambridge) (1997) 1993.
[8] Q. Cui, D.G. Musaev, M. Svensson, S. Sieber, K. Morokuma, J.
Am. Chem. Soc. 117 (1995) 12366.
[9] J.C. Peters, A.L. Odom, C.C. Cummins, Chem. Commun. (Cam-
bridge) (1997) 1995.
[10] M.H. Chisholm, J.A. Heppert, J.C. Huffman, W. E. Streib, J.
Chem. Soc. Chem. Commun. (1985) 1771.
[11] M.H. Chisholm, C.E. Hammond, V.J. Johnston, W.E. Streib,
J.C. Huffman, J. Am. Chem. Soc. 114 (1992) 7056.
[12] R.E. LaPointe, P.T. Wolczanski, J.F. Mitchell, J. Am. Chem.
Soc. 108 (1986) 6382.
[45] M.G. Fickes, W.M. Davis, C.C. Cummins, J. Am. Chem. Soc.
117 (1995) 6384.
[46] S.L. Stokes, W.M. Davis, A.L. Odom, C.C. Cummins,
Organometallics 15 (1996) 4521.
[47] H. Memmler, K. Walsh, L.H. Gade, J.W. Lauher, Inorg. Chem.
34 (1995) 4062.
[48] R. Hoffmann, Angew. Chem. Int. Ed. Engl. 21 (1982) 711.
[49] M.B. O’Donoghue, W.M. Davis, R.R. Schrock, W.M. Reiff,
Inorg. Chem. 38 (1999) 243.
[13] D.R. Neithamer, R.E. LaPointe, R.A. Wheeler, D.S. Richeson,
G.D. Van Duyne, P.T. Wolczanski, J. Am. Chem. Soc. 111
(1989) 9056.
[14] R.L. Miller, P.T. Wolczanski, A.L. Rheingold, J. Am. Chem.
Soc. 115 (1993) 10422.
[15] A.L. Odom, C.C. Cummins, J.D. Protasiewicz, J. Am. Chem.
Soc. 117 (1995) 6613.
[50] R.D. Sanner, J.M. Manriquez, R.E. Marsh, J.E. Bercaw, J. Am.
Chem. Soc. 98 (1976) 8351.
[51] C.E. Laplaza, A.L. Odom, W.M. Davis, C.C. Cummins, J.D.
Protasiewicz, J. Am. Chem. Soc. 117 (1995) 4999.
[52] A.B. Pangborn, M.A. Giardello, R.H. Grubbs, R.K. Rosen, F.J.
Timmers, Organometallics 15 (1996) 1518.
[53] G.A. Landrum, YAeHMOP, Yet another extended Huckel
molecular orbital package. YAeHMOP is freely available on the
WWW at URL: http: //overlap.chem.cornell.edu: 8080/yaeh-
mop.html.
[16] K.R. Dunbar, R.A. Heintz, Prog. Inorg. Chem. 45 (1997) 283.
[17] W.P. Griffith, Coord. Chem. Rev. 17 (1975) 177.
[18] K.H. Trylus, A. Schro¨der, I. Bru¨dgam, R. Thiel, W.P. Fehlham-
mer, Inorg. Chim. Acta 269 (1998) 23.
[19] W.P. Fehlhanner, M. Fritz, Chem. Rev. 93 (1993) 1243.
[20] W.P. Fehlhammer, A. Schroder, J. Fuchs, E.U. Wurthwein,
Angew. Chem. Int. Ed. Engl. 31 (1992) 590.
[21] A.L. Odom, P.L. Arnold, C.C. Cummins, J. Am. Chem. Soc.
120 (1998) 5836.
[22] J.C. Peters, J.-P. Cherry, J.C. Thomas, L. Baraldo, D.J. Mindi-
.