Dinitrogen complexes of bis(cyclopentadienyl) titanium derivatives: Structural diversity arising from substituent manipulation

Article abstract of DOI:10.1021/om900282u

Dinitrogen coordination to a family of bis(cyclopentadienyl)titanium sandwich complexes, (η⁵-C₅Me₄R) ₂Ti (R = alkyl, aryl, silyl), has been systematically evaluated by in situ solution infrared spectroscopy. The

Full text of DOI:10.1021/om900282u

Organometallics 2009, 28, 4079–4088 4079
DOI: 10.1021/om900282u
Dinitrogen Complexes of Bis(cyclopentadienyl) Titanium Derivatives:
Structural Diversity Arising from Substituent Manipulation
Tamara E. Hanna,† Emil Lobkovsky, and Paul J. Chirik*
Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York
14853. †Present address: Department of Chemistry and Biochemistry, Texas Tech University, Lubbock,
TX 79409-1061
Received April 14, 2009
Dinitrogen coordination to a family of bis(cyclopentadienyl)titanium sandwich complexes,
(η5-C5Me4R)2Ti (R=alkyl, aryl, silyl), has been systematically evaluated by in situ solution infrared
spectroscopy. The maximum temperature of N2 coordination (Tmax) has been determined as a
function of cyclopentadienyl substituent and has been correlated with the structural type (monomer
or dimer) of the resulting dinitrogen complex. The electronic properties of each titanium sandwich
complex were assessed by infrared spectroscopy of the dicarbonyl derivatives and by the oxidation
potentials of the corresponding ferrocenes, (η5-C5Me4R)2Fe. From these data, the thermodynamic
preferences for N2 coordination have been established and they increase with smaller, electropositive
substituents.
Introduction
Titanium dinitrogen complexes5,6 have received consider-
able attention given their historical role in solution ammonia
synthesis7 and the more recently discovered ability to use N2
as the nitrogen source for complex molecular targets such as
monomorine I, (()-lycopodine, and pumiliotoxin C.8,9 De-
spite this interest, one of the potentially simplest chemical
transformations in the sequence-namely, the N2 coordina-
tion event-is not well understood. This lack of insight is
primarily a result of how early transition metal dinitrogen
complexes are synthesized. Typically a metal halide precur-
sor is reduced with an excess of a strong reducing agent such
as sodium amalgam or potassium graphite, and often the
resulting dinitrogen complex is merely a transient intermedi-
ate. Even when isolable N2 complexes result, studying dini-
trogen coordination is challenging given the ill-defined
nature of the alkali metal reduction procedures.
The fixation of molecular nitrogen into more value-added
nitrogen compounds is an area of continued interest given
the utility of these molecules as fuels, fertilizers, pharmaceu-
ticals, and other fine chemicals.1 The advent of well-defined,
soluble metal compounds2 that coordinate, activate, and
even functionalize3 the otherwise inert N2 molecule has
provided insight into a number of pathways by which
dinitrogen can be reduced. Significantly, homogeneous dini-
trogen compounds present the unique opportunity to estab-
lish structure-reactivity relationships for nitrogen fixation
and offer the possibility for new transformations by manip-
ulation of the ancillary ligands.4
*Corresponding author. E-mail: pc92@cornell.edu.
(1) Smil, V. Enriching the Earth: Fritz Haber, Carl Bosch, and the
Transformation of World Food Production; MIT Press, 2001.
(2) For reviews see: (a) Gambarotta, S.; Scott, J. Angew. Chem., Int.
Ed. 2004, 43, 5298. (b) Kozak, C. M.; Mountford, P. M. Angew. Chem.,
Int. Ed. 2004, 43, 1186. (c) MacKay, B. A.; Fryzuk, M. D. Chem. Rev.
2004, 104, 385. (d) Shaver, M. P.; Fryzuk, M. D. Adv. Synth. Catal. 2003,
345, 1061.
(3) (a) Hidai, M.; Mizobe, Y. Can. J. Chem. 2005, 83, 358. (b) Schrock,
R. R. Chem. Commun. 2003, 2389. (c) Fryzuk, M. D. Chem. Rec. 2003, 3, 2.
(d) Korobkov, I.; Gambarotta, S.; Yap, G. P. A. Angew. Chem., Int. Ed.
2003, 42, 4958. (e) Spencer, L. P.; MacKay, B. A.; Patrick, B. O.; Fryzuk,
M. D. Proc. Natl. Acad. Sci. 2006, 103, 17094. (f) Chirik, P. J. Dalton Trans.
2007, 16. (g) Fryzuk, M. D. Acc. Chem. Rev. 2009, 42, 127.
The discovery of thermally stable, isolable monomeric
bis(cyclopentadienyl)titanium sandwich complexes, (η5-
C5Me4R)2Ti (R=silyl), by Lawless10 and Mach11 has pro-
vided a unique opportunity to study dinitrogen coordination
at a divalent titanium center. Of particular interest is the
influence of cyclopentadienyl substitution on the azophili-
city of the titanium as well as the molecularity (monomer
versus dimer) of the resulting dinitrogen complex. Depend-
ing on the size of the ring substituents, a range of structural
types has been observed and, in many cases, crystallogra-
phically characterized. For the purpose of this study, each of
(4) Pool, J. A.; Chirik, P. J. Can. J. Chem. 2005, 83, 286.
(5) For representative examples see: (a) Knobloch, D. J.; Toomey,
H. E.; Chirik, P. J. J. Am. Chem. Soc. 2008, 130, 4248. (b) Scherer, A.;
Kollak, K.; Luetzen, A.; Friedemann, M.; Haase, D.; Saak, W.;
Beckhaus, R. Eur. J. Inorg. Chem. 2005, 6, 1003. (c) Hagadorn, J. R.;
Arnold, J. J. Am. Chem. Soc. 1996, 118, 893. (d)Beydoun, N.;Duchateau,
R.; Gambarotta, S. Chem. Commun. 1992, 244. (e) Duchateau, R.;
Gambarotta, S.; Beydoun, N.; Bensimon, C. J. Am. Chem. Soc. 1991,
113, 8986. (f) Baumann, R.; Stumpf, W. M.; Davis, W. M.; Liang, L. -C.;
Schrock, R. R. J. Am. Chem. Soc. 1999, 121, 7822.
(7) Vol’pin, M. E.; Shur, V. B. Nature 1966, 209, 1236.
(8) Mori, M. J. Organomet. Chem. 2004, 689, 4210.
(9) Mori, M. Heterocycles 2009, 78, 281.
(10) Hitchcock, P. B.; Kerton, F.; Lawless, G. A. J. Am. Chem. Soc.
1998, 120, 10264.
ꢀ
ꢀ
(11) (a) Lukesova, L.; Horacek, M.; Stepnicka, P.; Fejfarova, K.;
Gyepes, R.; Cisorova, I.; Kubista, J.; Mach, K. J. Organomet. Chem.
ꢀ
ꢀ
ꢀ
2002, 663, 134. (b) Horacek, M.; Kupfer, V.; Thewalt, U.; Stepnicka, P.;
Polasek, M.; Mach, K. Organometallics 1999, 18, 3572.
(6) (a) Bercaw, J. E.; Brintzinger, H. H. J. Am. Chem. Soc. 1971, 93,
2045. (b) Bercaw, J. E. J. Am. Chem. Soc. 1974, 96, 5087.
r
2009 American Chemical Society
Published on Web 06/23/2009
pubs.acs.org/Organometallics

4080 Organometallics, Vol. 28, No. 14, 2009
Hanna et al.
Figure 1. Structurally characterized examples of substituted bis(cyclopentadienyl)titanium dinitrogen complexes.
the major structural types is presented in Figure 1 and
classified as type I, II, III, IV, or V. One unusual example
not depicted in Figure 1 has been observed with the parent
cyclopentadienyl complex where a μ3-N2 ligand is coordi-
nated between three titanium centers.12
Our laboratory has recently added to the known structural
types of titanocene dinitrogen complexes. Exposure of (η5-
C5Mei4Pr)2Ti to 1 atm of N2 at temperatures below -30 °C
results in coordination of 2 equiv of dinitrogen, as judged by
in situ infrared spectroscopy.13 X-ray diffraction established
a monomeric titanocene bis(dinitrogen) complex, (η5-
C5Mei4Pr)2Ti(N2)2, isoelectronic and isostructural to more
common dicarbonyl compounds. Elucidation of the solid
state structure of (η5-C5Me4i Pr)2Ti(N2)2 verified formation
of (η5-C5Me5)2Ti(N2)2, initially proposed by Bercaw on the
basis of low temperature heptane solution spectroscopic
studies.6b Introduction of larger, electron-withdrawing silyl
substituents into the tetramethylated cyclopentadienyl rings
has allowed spectroscopic observation13 and ultimately crys-
tallographic characterization of monomeric titanocene
mono(dinitrogen) compounds, (η5-C5Me4SiMe2Ph)2Ti(N2)
(Figure 1).14 These compounds served as an inspiration for
the synthesis of related titanocene monocarbonyl deriva-
tives, (η5-C5Me4R)2Ti(CO) (R=iPr, SiMe2Ph, SiMe3), and
mixed carbonyl-dinitrogen species, (η5-C5Me4R)2Ti(CO)-
(N2).14 By reducing the size and number of the cyclopenta-
dienyl substituents, μ2, η2, η2-N2 “side-on” coordination
can be favored, as evidenced by the synthesis and crys-
tallographic characterization of [(η5-C5H2-1,2,4-Me3)2Ti]2-
(μ2,η2,η2-N2).15
Figure 2. Bis(cyclopentadienyl)titanium sandwich complexes,
(η5-C5Me4R)2Ti, used in this study.
symmetric and asymmetric NtN bands arising from the
dinitrogen complex. Of the structural types presented in
Figure 1, three contain infrared active N2 ligands. One
limitation of this technique is the detection of centrosym-
metric, bridging end-on compounds (types III and V), where
the μ2-N2 ligand is not infrared active.
The bis(cyclopentadienyl)titanium compounds used in
this study and corresponding numbering scheme are pre-
sented in Figure 2. Many of these compounds have been
previously reported by Brintzinger,6a Lawless,10 Mach,11
and our laboratory.13,14 New additions include R=Et (2),
Ph (5), and 3,5-Me2-C6H3 (6). The majority of the titanium
sandwiches used in this study were prepared by sodium
amalgam reduction of the appropriate monochloride
precursor, (η5-C5Me4R)2TiCl, in toluene solution (eq 1).
Exceptions are 1 and 7, where the titanocene dichloride,
(η5-C5Me4R)2TiCl2, was used.
In this article, we report a more comprehensive synthetic
and spectroscopic study aimed at elucidating the influence of
cyclopentadienyl substitution on the thermodynamics of
dinitrogen coordination by titanium sandwich compounds.
This work also highlights how manipulation of ring sub-
stitution governs the structure of the resulting dinitrogen
complex.
Results and Discussion
Synthesis of Bis(cyclopentadienyl)titanium Sandwich Com-
pounds and Dinitrogen Coordination Studies. As we have
described previously,14,13 in situ solution infrared spec-
troscopy is a convenient tool to assay N2 coordination in
titanium sandwich compounds, a consequence of the intense
(12) Pez, G. P.; Apgar, P.; Crissey, R. K. J. Am. Chem. Soc. 1982, 104,
482.
(13) Hanna, T. E.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc.
2004, 126, 14688.
(14) Hanna, T. E.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc.
2006, 128, 6018.
(15) Hanna, T. E.; Bernskoetter, W. H.; Bouwkamp, M. W.;
Lobkovsky, E.; Chirik, P. J. Organometallics 2007, 26, 2431.
The majority of the desired titanium sandwiches were
isolated as burgundy solids following recrystallization from
pentane at -35 °C. Sandwich compounds 2 and 5 were
isolated as green solids. In certain instances (1-3, 5, 6),
low temperature isolation furnished a dinitrogen complex.

Article
Table 1. 1H NMR Chemical Shifts of (η5-C5Me4R)2Ti
Organometallics, Vol. 28, No. 14, 2009 4081
Compounds Recorded in Benzene-d6 at 23 °C
com-
pound
δ(CH3)
(ppm)
δ(CH3)
(ppm)
R
δ(R) (ppm) reference
1
2
Me
Et
62.0
57.06
13
2.48 (CH2CH3) this work
14.48 (CH2CH3)
64.82
62.70
3
iPr
45.37 3.19 (CHMe2) 12
4.87 (CHMe2)
34.11 4.03
47.14 -8.59 (Ph)
3.72 (m-Ph)
9.47 (Ph)
4
5
tBu
Ph
62.87
65.17
12
this work
6
3,5-Me2-
C6H3
66.73
49.02 -8.94 (o-Ph)
this work
Figure 3. In situ infrared spectrum of a 22 mM solution of 2
under an atmosphere of dinitrogen at -78 °C.
1.40 (m-Me)
5.19 (p-Ph)
26.25 6.95
23.95 2.11 (SiMe2)
6.71 (CMe3)
25.12 7.52 (Ph)
8.78 (Ph)
9.98 (Ph)
and 2024 cm-1 and one weaker band centered at 1719 cm-1
.
7
8
SiMe3
SiMet2Bu
79.86
82.62
10
9
Stretching frequencies in this region of the IR spectrum are
similar to those reported for dimeric, end-on dinitrogen com-
plexes with two terminal and one bridging N2 ligand (type IV).
9
SiMe2Ph
78.67
10
term
For example, [(η5-C5Me5)2Ti(N2)]2(μ2,η1,η1-N2) has ν
=
N2
bridging
2058, 2020 and ν
= 1711 cm-1 (Nujol), similar to the
N2
values reported for crystallographically characterized [(η5-
Conversion to the bis(cyclopentadienyl) titanium sandwich
was accomplished by stirring in pentane solution at ambient
temperature. One exception was 1, where competing cyclo-
melation pathways prevented isolation in the solid state, in
agreement with the literature report.6
term
bridging
N2
C5Me5)2Zr(N2)]2(μ2,η1,η1-N2) (ν
=2040, 2003, ν
=
N2
1578 cm-1 (Nujol)).18 Based on these data, the solution
structure of the new compound formed at higher titanium
concentrations is [(η5-C5Me4Et)2Ti(N2)]2(μ2,η1,η1-N2) ([2-(N2)]2-
N2) (eq 2). Recall that the type III complex [(η5-C5Me4Et)2Ti]2-
(μ2 η1,η1-N2) ([2]2N2) may also be present in these experiments
(vide infra) but cannot be detected by infrared spectroscopy.
As expected for linear, d2 metallocenes,16 1-9 are para-
magnetic, with benzene-d6 solution magnetic moments
(Evans method17) consistent with S=1 compounds (see
Experimental Section). Compiled in Table 1 are the para-
magnetically broadened and shifted 1H NMR resonances for
each compound. The purely hydrocarbyl-substituted deri-
vatives, 1-6, exhibit two inequivalent cyclopentadienyl
methyl resonances in the vicinity of 60-70 and 34-57 ppm
at 23 °C in benzene-d6. This chemical shift dispersion is more
pronounced in the silylated derivatives, 7-9, where one
methyl group appears around 80 ppm while the other is
located between 20 and 26 ppm.
With a series of bis(cyclopentadienyl) titanium sandwich
compounds in hand, the coordination of dinitrogen was
monitored in pentane solution by in situ infrared spectros-
copy. Because the dinitrogen complex derived from 1 has
been isolated as a dimeric, end-on dinitrogen complex, [(η5-
C5Me5)2Ti]2(μ2, η1, η1-N2) ([1]2N2, type III), the ethylated
titanocene, 2, was initially studied. Cooling a 4.2 mM
pentane solution of 2 under an atmosphere of dinitrogen to
temperatures below 0 °C produced two intense bands cen-
tered at 2090 and 1984 cm-1, assigned to the symmetric and
asymmetric NtN stretches of the monomeric titanocene
bis(dinitrogen) complex (η5-C5Me4Et)2Ti(N2)2 (2-(N2)2, type
II). Increasing the titanium concentration to 22 mM produced
the infrared spectrum shown in Figure 3, where a new titanium
dinitrogen complex in addition to 2-(N2)2 was detected. Three
new bands were observed, two strong stretches centered at 2057
On a preparative scale, cooling a ∼50 mM pentane solu-
tion of 2 to -35 °C under an atmosphere of dinitrogen
yielded deep blue crystals identified as [2-(N2)]2N2. The solid
state structure (Figure 4) established a molecular geometry
that resembles [(η5-C5Me5)2Zr(N2)]2(μ2,η1,η1-N2)19 with one
bridging and two terminal dinitrogen ligands. To our knowl-
edge, [2-(N2)]2N2 is the first example of a crystallographi-
cally characterized titanocene dinitrogen complex of this
type (type IV). The more common structural motif in
(18) Manriquez, J. M.; McAlister, D. R.; Rosenberg, E.; Shiller, A.
M.; Williamson, K. L.; Chan, S. I.; Bercaw, J. E. J. Am. Chem. Soc. 1978,
100, 3078.
(16) Freitag, M. A.; Gordon, M. S. J. Phys. Chem. A 2002, 106, 7921.
(17) Sur, S. K. J. Magn. Reson. 1989, 82, 169.
(19) Sanner, R. D.; Manriquez, J. M.; Marsh, R. E.; Bercaw, J. E. J.
Am. Chem. Soc. 1976, 98, 8351.

4082 Organometallics, Vol. 28, No. 14, 2009
Hanna et al.
Previously our laboratory reported that cooling concen-
trated pentane solutions of 3 to -35 °C under an atmosphere
of dinitrogen produced X-ray quality crystals of the bis
(dinitrogen) complex 3-(N2)2. Because concentration-depen-
dent solution behavior was observed with 1 and 2, concen-
trated (50 mM) pentane solutions of 3 were prepared and
cooled below -35 °C. Under these conditions, no evidence
for dimeric titanium dinitrogen complexes was obtained by
IR spectroscopy, suggesting that an isopropyl group on each
ring is sufficiently large to prevent dimerization. Toepler
pump experiments were also consistent with liberation of
2 equiv of N2 per titanium, demonstrating that type III
complexes are not formed in an appreciable amount.
titanocene chemistry is monobridging end-on (type III), as
exemplified by [(η5-C5Me4R)2Ti]2(μ2,η1,η1-N2) (R = H,20
Me21) and [(η5-C5H3-1,3-(SiMe3)2)2Ti]2(μ2,η1,η1-N2).22 As
mentioned above, the direct analogue of [2-(N2)]2N2, [(η5-
C5Me5)2Ti(N2)]2(μ2,η1,η1-N2), has been observed only by
infrared spectroscopy.18
The solid state structure of [2-(N2)]2N2 has idealized C2
symmetry with the principal axis bisecting the N(1)-N(1A)
bond (Figure 4). The dihedral angle between the planes
defined by the metal and the titanium centroids is 78.6°.
Overall the bond lengths in the core of the molecule are
consistent with weakly activated dinitrogen ligands. The
bridging dinitrogen ligand has an N(1)-N(1A) bond dis-
˚
The pentane solution dinitrogen coordination chemistry
of the previously unpublished members of the series 5, 6, and
8 was also examined as a function of temperature by in situ
IR spectroscopy. Cooling the phenyl-substituted titanocene,
5, to -25 °C produced six new bands in the region of the
infrared spectrum typically associated with terminal NtN
stretches (Figure 5). Four nearly equal intensity bands were
observed at 2107, 2088, 2016, and 1984 cm-1, while two
weaker bands were observed at 2099 and 1995 cm-1. One
additional very weak stretch was located at 1603 cm-1 and
was assigned to a μ2-N2 of a type IV complex, establishing
formation of [5-(N2)]2N2. Curiously this band appears at a
significantly lower frequency than that observed for [2-
(N2)]2N2. N2 coordination is reversible from all species, as
warming the sample above -25 °C quantitatively regener-
ated 5. The three sets of peaks appeared simultaneously and
changes in concentration did not alter the relative intensities.
Similar infrared spectroscopic behavior was observed with
6. Once again a total of six NtN stretches (four major, two
minor) were observed in the terminal N2 stretching region.
tance of 1.150(2) A, slightly elongated from the cor-
˚
responding distance of 1.1111(17) A for the terminal N2
ligands. The values are similar to those reported for [(η5-
C5Me5)2Zr(N2)]2(μ2,η1,η1-N2), where the N-N distances of
˚
˚
1.182(5) A and 1.116(8), 1.114(7) A are observed for the
bridging and terminal dinitrogen ligands, respectively.19
Dinitrogen coordination in [2-(N2)]2N2 is reversible. Stirring
the compound in pentane solution at ambient temperature
produced 2 equiv of noncombustible gas (Toepler pump),
consistent with formation of [2]2N2 (eq 3). Continued stirring
under these conditions did not liberate additional N2 gas. The
third equivalent of dinitrogen was released only upon addition
of diphenylacetylene to yield the corresponding titanocene
diphenylacetylene compound, (η5-C5Me4Et)2Ti(η2-PhCCPh).
An additional weak NtN band was observed at 1603 cm-1
,
signaling formation of [6-(N2)]2N2. As with 5, the infrared
spectrum of a 0.993 mM solution, the most dilute conditions
experimentally possible, was indistinguishable from those
recorded at more typical concentrations of 5.0 mM.
The observation of concentration-dependent dinitrogen
complex formation prompted reinvestigation of the low tem-
perature solution behavior of 1. Bercaw previously reported
the detection of transient 1-(N2)2 by infrared spectroscopy with
the observation of two strong NtN bands centered at 2086
and 1980 cm-1 in heptane solution.6b However, conditions
were not reported where 1-(N2)2 was most likely the principal
species in solution. In our hands, we found that cooling a
4.5 mM pentane solution of 1 to below 11 °C produced two in-
tense N2 bands centered at 2090 and 1982 cm-1 in the pentane
solution in situ IR spectrum assigned to 1-(N2)2 (eq 4). Under
these conditions no other titanocene dinitrogen complexes
were detected, although it is possible that [1]2N2 was present.
Comparing these values to other known monomeric titano-
cene bis(dinitrogen) complexes provides conclusive evidence
for Bercaw’s spectral assignment for 1-(N2)2.
On the basis of the spectroscopic data for 5 and 6, we
tentatively conclude that one type IV and two type II
dinitrogen complexes form upon cooling the aryl-substituted
titanocenes. Detection of two different monomeric, type II
titanocene dinitrogen complexes is most likely a result of
population of different rotamers of the titanocene dinitrogen
complex. Importantly, the different orientations of the cy-
clopentadienyl rings or substituents induce a measurable
change in the frequency of the N2 vibrations. Similar ob-
servations have been reported for bis(indenyl)zirconocene25
and hafnocene23 dicarbonyl complexes. Because this beha-
vior is observed only in the aryl-substituted cases, the
orientation of the aromatic ring relative to the cyclopenta-
dienyl plane may be the origin of the spectroscopically
distinct dinitrogen complexes. Attempts to favor one rota-
mer with the preparation of the titanocene derivative where
R=2,6-Me2-C6H3 have been unsuccessful. Another possi-
bility is that one type II dinitrogen complex forms along with
two type IV compounds. However, the observation of a
single, weak μ2-N2 stretch argues against this possibility. It is
also curious that unlike the N2 coordination chemistry of 2,
changes in titanium concentration do not change the ratio
of dinitrogen complexes formed. The origin of this behavior
(20) deWolf, J. M.; Blaauw, R.; Meetsma, A.; Teuben, J. H.; Gyepes,
R.; Varga, V.; Mach, K.; Veldman, N.; Spek, A. L. Organometallics
1996, 15, 4977.
(21) Sanner, R. D.; Duggan, D. M.; McKenzie, T. C.; Marsh, R. E.;
Bercaw, J. E. J. Am. Chem. Soc. 1976, 98, 8358.
(22) Hanna, T. E.; Keresztes, I.; Lobkovsky, E.; Bernskoetter, W. H.;
Chirik, P. J. Organometallics 2004, 23, 3448.
(23) Pun, D.; Leopold, S. M.; Bradley, C. A.; Lobkovsky, E.; Chirik,
P. J. Organometallics 2009, 28, 2471.

Article
Organometallics, Vol. 28, No. 14, 2009 4083
Figure 4. Molecular structure of [2-(N2)]2N2 at 30% probability ellipsoids (left) and a view of the core of the molecule (right).
Hydrogen atoms are omitted for clarity.
observed in the pentane solution infrared spectrum down to
temperatures as low as -90 °C. This compound is the only
titanium sandwich in the series that did not coordinate
dinitrogen at temperatures conveniently accessed in the
laboratory.
The infrared stretching frequencies and maximum tempera-
tures (Tmax) at which coordination was first observed for all of
the monomeric bis(dinitrogen) titanocene complexes (type II)
are compiled in Table 2. Because the band assignments for
5 and 6 are not definitive, the data for the type IV dinitrogen
complex are also included. Within the series of purely alky-
lated bis(cyclopentadienyl) titanium sandwiches, 1-4, Tmax
steadily decreases: R=Me (11 °C)>Et (0 °C)>iPr (-30 °C)>
tBu (-75 °C). These compounds have very similar N2 stretch-
ing frequencies with ν(NtN)sym=2090 cm-1 for each com-
pound and ν(NtN)asym varying by only 2 cm-1. Only two
silylated titanium sandwiches coordinate dinitrogen, 7 and 9,
with values of Tmax generally lower than the alkylated com-
Figure 5. Pentane solution infrared spectrum of 5 under an
atmosphere of dinitrogen as a function of temperature.
pounds. Recall thatbothof thesecompounds formobservable
titanocene mono(dinitrogen) compounds (type I) that
when cooled to lower temperatures or exposed to higher N2
pressures convert to 7-(N2)2 and 9-(N2)2.13,14 In cases where
dimeric structures are observed, appreciable amounts of the
monomeric compounds remain in the infrared spectrum.
One potential complication in drawing conclusions from
the maximum temperatures of dinitrogen coordination con-
cerns the equilibrium constants for the formation of the
monomeric titanocene bis(dinitrogen) complexes. Determin-
ing these values by in situ infrared spectroscopy is challen-
ging, as the starting bis(cyclopentadienyl) titanium sandwich
compounds do not have a distinct spectroscopic feature that
allows reliable quantitation. To address this issue, a near
equimolar mixture of 3 and the corresponding titanocene
dicarbonyl complex, 3-(CO)2, was prepared in pentane
solution and cooled to -78 °C under an atmosphere of
dinitrogen. The resulting infrared spectrum of 3-(N2)2 and
3-(CO)2 (Figure 7a) indicates a near equal concentration of
titanocene bis(dinitrogen) and dicarbonyl complexes, sug-
gesting that N2 coordination to this particular titanocene
is essentially quantitative. Previous work from our labora-
tory on mixed titanocene carbonyl dinitrogen complexes,
(η5-C5Me4R)2Ti(CO)(N2),14 established that the extinction
coefficients for N2 and CO bands are similar.
for the aryl-substituted titanocenes is not understood at
this time.
On a preparative scale, cooling a pentane solution of 6
to -35 °C under a dinitrogen atmosphere produced dark
purple crystals suitable for X-ray diffraction. The solid state
structure (Figure 6) establishes a type III end-on bridging
dinitrogen complex, [6]2N2. The overall geometry of the
molecule is best described as idealized C2 symmetric with
the principal axis on the Ti-N2-Ti vector. As is typically
observed with this structural motif, a weakly activated
dinitrogen ligand with a short N(1)-N(2) bond distance of
˚
1.1574(11) A is observed.
The crystallographic characterization of infrared-silent
[6]2N2 suggested that type III dinitrogen complexes must be
considered in N2 coordination studies. To determine if [6]2N2,
not 6, was actually the product isolated from the reduction of
6-Cl, the titanium product was treated with PhCtCPh and
the progress of the reaction monitored with a Toepler pump.
No detectable quantity of noncombustible gas was collected
during the formation of 6-(PhCtCPh). This experiment, in
conjunction with a satisfactory combustion analysis, defini-
tively established the identity of the reduction product as the
bis(cyclopentadienyl)titanium sandwich, 6. Therefore, [6]2N2
forms only upon cooling to temperatures below -31 °C.
For the silylated bis(cyclopentadienyl) titanium sandwich,
8, first reported by Lawless and co-workers,10 no change was
Performing the same experiment with an equimolar mix-
ture of 7 and 7-(CO)2 at -78 °C revealed diminished quan-
tities of 7-(N2)2 (Figure 7b), suggesting a lower equilibrium

4084 Organometallics, Vol. 28, No. 14, 2009
Hanna et al.
Figure 6. Molecular structure (left) of [6]2N2 at 30% probability ellipsoids. Hydrogen atoms are omitted for clarity. The core of the
molecule is presented on the right.
and subsequently our laboratory,25,26 pentane solution in-
Table 2. Pentane Solution Infrared Stretching Frequencies and
Temperature of Dinitrogen Coordination for a Series of Bis
frared spectroscopy of group 4 metallocene dicarbonyl com-
pounds has proven to be the most reliable method for
assessing the electron density at the metal center as a function
of cyclopentadienyl (indenyl)25 substituent. To extend this
approach to titanocene chemistry, dicarbonyl derivatives of
each titanium sandwich presented in Figure 2 were prepared.
Several examples, 1-(CO)2,6a 2-(CO)2,27 3-(CO)2,13 4-(CO)2,13
(cyclopentadienyl) Titanium Complexes, (η5-C5Me4R)2Ti
compound
R
Tmax (°C)a
ν(N2)sym
ν(N2)asym
1
2
3
4
5
Me
Et
11
0
-30
-75
-24
2090
2090
2090
2090
2107c
2099
2088c
2105c
2099
2086c
2098
1982
1984
1986
1982
2016c
1995
1984c
2013c
1993
1982c
2002
iPr
tBu
Ph
14
7-(CO)2,13 and 9-(CO)2 have been reported previously;
however the pentane solution infrared spectrum of each
compound was recorded in our laboratory to ensure proper
comparison. The new members of the series, 5-(CO)2, 6-
(CO)2, and 8-(CO)2, were prepared either by addition of
carbon monoxide to the titanium sandwich or by reduction
of the titanocene monochloride with sodium amalgam in the
presence of carbon monoxide (eq 5).
6
3,5-Me2-C6H3
-31
7
8
9
SiMe3
SiMet2Bu
SiMe2Ph
-50 (-45)b
-90d
-60 (-36)b
2096
2003
a Highest temperature ((2 °C) at which coordination of N2 was
detected by in situ IR spectroscopy. b Tmax for mono(dinitrogen) com-
plex formation. c Denotes major band. d No N2 coordination detected at
this temperature.
constant for N2 coordination. Also evident in Figure 7b is a
significant quantity of the titanocene mono(dinitrogen)
complex, 7-N2. In the case of 9 and 9-(CO)2 at -78 °C, the
titanocene mono(dinitrogen) complex, 9-N2, is the dominant
N2 complex in solution, demonstrating a reduced concentra-
tion of the bis(dinitrogen) complex. Although in the silylated
compounds dinitrogen coordination is not quantitative,
The pentane solution infrared stretching frequencies
of each titanocene dicarbonyl compound are reported in
Table 3. In agreement with trends established in related
zirconocene chemistry,24 alkyl-substituted cyclopentadienyl
rings are more electron donating than their silylated counter-
parts, as evidenced by the shifting of the average νCO values
to higher frequency for the latter class of compounds. Within
the series of purely alkylated titanocenes, 1-(CO)2-4-(CO)2,
the tert-butyl-substituted ring is the most electron donating
(νaCvO=1894 cm-1), while the electronic properties of R=Me,
qualitatively the equilibrium constants track with Tmax
.
Thus, the titanium sandwich compounds that coordinate
N2 at higher temperatures also have larger equilibrium
constants than those with lower values of Tmax. As a result,
Tmax can be used as a qualitative assessment of the thermo-
dynamic parameters of dinitrogen coordination as a function
of cyclopentadienyl substituent.
Synthesis and Infrared Spectroscopy of Titanocene Dicar-
bonyl Compounds. To differentiate between steric and elec-
tronic contributions of each cyclopentadienyl substituent, a
series of spectroscopic and electrochemical studies were
carried out. As demonstrated by Parkin and co-workers24
i
Et, Pr derivatives are indistinguishable by IR spectro-
scopy. Comparison of the CO stretching frequencies of the
~
(25) Bradley, C. A.; Flores-Torres, S.; Lobkovsky, E.; Abruna, H. D.;
Chirik, P. J. Organometallics 2004, 23, 5332.
(26) Pool, J. A.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2003,
125, 2241.
(27) Bettinger, B.; Scherer, O. J.; Heckmann, G. J. Organomet. Chem.
1991, 405, C19.
(24) Zachmanoglou, C. E.; Docrat, A.; Bridgewater, B. M.; Parkin,
G. E.; Brandow, C. G.; Bercaw, J. E.; Jardine, C. N.; Lyall, M.; Green,
J. C.; Keister, J. B. J. Am. Chem. Soc. 2002, 124, 9525.

Article
Organometallics, Vol. 28, No. 14, 2009 4085
Figure 7. Pentane solution infrared spectrum of an equimolar mixture of (a) 3 and 3-(CO)2, (b) 7 and 7-(CO)2, and (c) 9 and 9-(CO)2.
All spectra recorded at -78 °C under 1 atm of N2.
Table 3. Pentane Solution Infrared Carbonyl Stretching
Frequencies for a Series of Bis(cyclopentadienyl) Titanium
Dicarbonyl Complexes, (η5-C5Me4R)2Ti(CO)2
aryl-substituted titanocene dicarbonyls, 5-(CO)2 and 6-
(CO)2, reveals a slight reduction of ν(CO) as methyl groups
are added to the phenyl rings, further underscoring the
electron-donating ability of alkyl-substituted cyclopentadie-
nyls. Notably, both 5-(CO)2 and 6-(CO)2 exhibit only two
carbonyl bands, suggesting that if different rotamers are
responsible for the multiple bands observed with the corre-
sponding type II dinitrogen complexes, they are not distin-
guishable by IR spectroscopy for the dicarbonyl derivatives.
The family of silyl-substituted titanocene dicarbonyls
exhibit carbonyl bands at higher frequencies than the corre-
sponding alkylated compounds, consistent with a less elec-
tron rich titanium center resulting from the inductively
withdrawing nature of the silyl groups.24-26 The data within
this series display little variance and suggest that attempts to
distinguish the electronic properties of these compounds
based on infrared spectroscopy are not possible. Therefore
from these studies, the relative electronic donation of the
cyclopentadienyl substituents is R=tBu>iPr ∼ Et ∼ Me>
3,5-Me2-C6H3 >Ph>SiR3 (e.g., SiMe3 ∼ SiMet2Bu ∼ Si-
Me2Ph).
Electrochemical Studies on the Corresponding Ferrocene
Derivatives. Because infrared spectroscopy was unable to
distinguish the subtle differences between like substituents
within a given series (e.g., alkyl, silyl), another independent
measure of the electron density of the metal center as a
function of the cyclopentadienyl substituent was sought.
Ideally, electrochemical studies would confirm the general
conclusions from the infrared spectroscopic studies on the
titanocene dicarbonyls and provide better resolution within
the series of alkyl and silyl groups. The corresponding
compound
R
ν(CO)sym
ν(CO)asym
ν(CO)av
1
2
3
4
5
6
7
8
9
Me
Et
1939
1939
1938
1933
1941
1940
1941
1942
1941
1859
1859
1859
1855
1862
1860
1865
1868
1865
1899
1899
1898.5
1894
1901.5
1900
1903
1905
1903
iPr
tBu
Ph
3,5-Me2-C6H3
SiMe3
SiMet2Bu
SiMe2Ph
ferrocenes, (η5-C5Me4R)2Fe, were initially targeted due to
their ease of synthesis and well-defined voltammetric re-
sponses. It should be noted that these experiments were
conducted to differentiate the electronic properties of cyclo-
pentadienyl substituents, not to directly evaluate N2 coordi-
nation by a 14-electron titanium sandwich compound. The
numbering scheme for the bis(cyclopentadienyl) iron com-
pounds is retained from the previous section but is slightly
modified by the addition of “Fe” to each shorthand desig-
nator. Thus, (η5-C5Me4Et)2Fe will be referred to as 2-Fe.
While 1-Fe is commercially available, 2-Fe,28 3-Fe,29 5-Fe,30
and 8-Fe31 were reported previously but were prepared in
our laboratory for the subsequent study. The remaining
(28) Feitler, D.; Whitesides, G. M. Inorg. Chem. 1976, 15, 466.
(29) Bensley, D. M.; Mintz, E. A. J. Organomet. Chem. 1988, 353, 93.
(30) Threlkel, R. S.; Bercaw, J. E. J. Organomet. Chem. 1977, 136, 1.
(31) Constantine, S. P.; Hitchcock, P. B.; Lawless, G. A.; De Lima, G.
M. Chem. Commun. 1996, 1101.

4086 Organometallics, Vol. 28, No. 14, 2009
Hanna et al.
compounds were prepared by straightforward salt meta-
thesis of the lithium cyclopentadienide with ferrous di-
chloride and isolated as air-stable orange solids in excellent
yield (eq 6).
Table 4. Formal Potentials for a Series of (η5-C5Me4R)2Fe
Compounds Recorded in THF Solution
compound
R
E1/2(mV)a
1-Fe
2-Fe
3-Fe
4-Fe
5-Fe
7-Fe
8-Fe
9-Fe
Me
Et
-404
-438
-444
-409
-251
-324
-358
-274
iPr
tBu
Ph
SiMe3
SiMet2Bu
SiMe2Ph
a Values reported relative to ferrocene/ferrocenium.
electron-donating ability of the substituents in (η5-
C5Me4R)2Fe complexes is as follows:
The redox chemistry of each of the ferrocenes was studied
in THF solution by cyclic voltammetry using a platinum
counter electrode and a glassy carbon working electrode.
A silver wire was used as a quasi reference electrode, and its
potential was calibrated versus ferrocene, (η5-C5H5)2Fe.
As expected, all of the compounds studied exhibited well-
behaved cyclic voltammetric responses, and the voltam-
metric profiles are as anticipated for a freely diffusing
redox couple that is chemically reversible. As reported
previously,25 the peak potential differences are larger than
the expected 59 mV due to ohmic losses in solution rather
than sluggish kinetics. Ferrocene itself exhibited a similar
response due to the relatively high resistance of the THF
solvent.
The formal potentials (Table 4) are negatively shifted
relative to ferrocene, as expected from the introduction of
electron-donating, methyl-substituted cyclopentadienyl
rings. In agreement with the infrared spectroscopic data,
the purely alkylated rings produce more electron rich iron
centers, as evidenced by the more negatively shifted poten-
tials. Similar trends have been established in other substi-
tuted ferrocenes32 and dibenzoferrocenes.25
Within the series of alkyl-substituted ferrocenes, the for-
mal oxidation potentials distinguish each of the hydrocarbyl
groups, although some caution must be exercised when
interpreting the data. As the size of the alkyl is steadily
increased, R=Me<Et<iPr, the oxidation potential becomes
steadily more negative, thereby demonstrating that 3-Fe has
a more electron rich iron center than 1-Fe. One exception is
4-Fe, where a more positive potential than expected
was obtained. The origin of this behavior is most likely steric.
As established previously,25,32 sterically hindered ferrocenes
exhibit artificially positive oxidative potentials arising from
alleviation of transannular interactions upon oxidation to
the less congested ferrocenium ion.
The electrochemical studies also produced different oxida-
tion potentials within the series of silylated ferrocenes. As
expected from the trend established in the alkylated cases,
introduction of more substituted silyl substituents produces
more electron rich metal centers. Thus, the formal oxidation
potential of [SiMet2Bu]-substituted 8-Fe is more negative
than [SiMe3]-substituted 7-Fe. Introduction of an induc-
tively withdrawing phenyl ring onto the silyl produces a
relatively electron poor iron center. The oxidation potential
of 9-Fe is shifted 50 mV more positive than 7-Fe. Notably,
the phenyl-substituted ferrocene, 5-Fe, has the most electron
poor iron center in the series. From these data, the relative
t
R ¼Ph<SiMe2Ph<SiMe3<SiMe2 Bu<Me<Et<iPr
f
e-donating
For a more relevant comparison, electrochemical mea-
surements were also conducted on the bis(cyclopentadienyl)
titanium sandwich complexes. However all of these com-
pounds suffered from irreversible voltammetric responses in
THF solution using [NBu4][PF6] as a supporting electrolyte
and a scan rate of 100 mV/s.
Interpretation of Cyclopentadienyl Substituent Effects on
Dinitrogen Coordination. From the combination of infrared
spectroscopic data and electrochemical results, a general trend
has emerged that is well established in the literature:24-26,33
alkylated cyclopentadienyls are more electron donating than
silylated cyclopentadienyls and as a result engender more
electron rich titanium centers. These results allow interpreta-
tion of the thermodynamics of dinitrogen coordination as a
function of cyclopentadienyl substituent. In the series of
alkyl-substituted titanium sandwiches, 1-4, the Tmax and
hence the equilibrium constants for dinitrogen coordination
steadily increase as the size of the substituent decreases (e.g.,
Keq(tBu)<Keq(iPr)<Keq(Et)<Keq(Me)). The origin of this
effect is most likely dominated by steric effects, as smaller
cyclopentadienyl substituents favor dinitrogen coordination
over larger ones. However, the infrared carbonyl stretching
frequencies in combination with the electrochemical studies
also suggest that increasing the alkyl substitution yields a
more electron rich metal center. Thus, 1 is the most electro-
philic alkylated titanocene and coordinates dinitrogen with
the largest equilibrium constant. From these data, it appears
that the steric and electronic effects work in concert, where
the most electrophilic titanium sandwiches bearing the smal-
lest cyclopentadienyl substituents have the highest affinity
for dinitrogen. This effect would be expected to be predomi-
nant if Ti-N2 bonding was largely dictated by σ rather than
π effects. The infrared stretching frequencies of the N2 bands
in the monomeric titanocene bis(dinitrogen) compounds
appear at relatively high frequency, consistent with little
back-bonding from the titanium center.
This trend is also obeyed with the silylated titanocenes
(7-9). While more electrophilic than 1-4, these compounds
have the largest substituents and hence are least likely
to coordinate N2. For example, the largest member of the
series, 8, does not coordinate dinitrogen to any detectable
(33) (a) Gassman, P. G.; Campbell, W. H.; Macomber, D. W.
Organometallics 1984, 3, 385. (b) Benn, R.; Rufinska, A. J. Organomet.
Chem. 1984, 273, C51.
(32) Okuda, J.; Albach, R. W.; Herdtweck, E. Polyhedron 1991, 10,
1741.

Article
Organometallics, Vol. 28, No. 14, 2009 4087
concentration, while reducing the size of the silyl substituent
to [SiMe3], as is the case with 7, likewise increases the
azophilicity. Again, steric and electronic effects work in
concert, as the largest, most electron donating compound
has the lowest affinity to coordinate N2.
In cases where the steric protection around the titanocene is
not sufficiently large, dimeric structures result. Specifically for
the (η5-C5Me4R)2Ti family of compounds, the cases where R=
Me, Et, Ph, and 3,5-Me2-C6H3 form dimeric dinitrogen com-
plexes, while the derivatives where R>iPr form only mono-
meric N2 compounds. No silyl-substituted (η5-C5Me4R)2Ti
sandwich formed a dimeric dinitrogen compound.
analyses were performed at Robertson Microlit Laboratories,
Inc., in Madison, NJ.
Cyclic voltammograms (CVs) were collected using 30 mL
beakers as electrochemical cells with a 3 mm glassy carbon
working electrode, Pt wire as a counter electrode, and Ag wire as
a reference in a drybox equipped with electrochemical outlets.
CVs were recorded using a Bioanalytical Systems CV-27 vol-
tammograph. All CVs were run at a scan rate of 100 mV/s.
Solutions of the individual compounds were prepared by char-
ging 2 mg of compound (4.5 mmol) and 0.200 g of [n-Bu4N][PF6]
in a vial and dissolving the solids in 5 mL of THF. This produced
solutions of approximately 0.1 mM in compound and 0.1 M in
electrolyte. After recording the baseline of a standard 0.1 M
solution of electrolyte, CVs were collected for each compound,
including (η5-C5H5)2Fe. Oxidation potentials were then refer-
enced to the formal potential of (η5-C5H5)2Fe/(η5-C5H5)2Fe+.
Solution infrared spectra were recorded with an in situ IR
spectrometer fitted with a 30-bounce, silicon-tipped probe
optimized for sensitivity. The spectra were acquired in 16 scans
(30 s intervals) at a gain of 1 and a resolution of 4. A represen-
tative reaction was carried out as follows: The IR probe was
inserted through a nylon adapter and O-ring seal into a flame-
dried, cylindrical flask fitted with a magnetic stir bar and T-
joint. The T-joint was capped by a septum for injections and
a nitrogen line. Following evacuations under full vacuum
and flushing with nitrogen, the flask was charged with pentane
and a background was recorded at ambient temperature and at -
78 °C. The flask was then charged with a pentane solution of
titanocene to make the final reaction volume 10.0 mL with
approximately 5.0 mM concentration. The samples were cooled
using ethanol/dry ice or acetone/dry ice baths, and the tempera-
tures were monitored using an external, low-temperature thermo-
meter. The reactions were recorded between 20 and 40 intervals.
The maximum temperature for N2 coordination was deter-
mined by monitoring the IR spectrum as a function of tempera-
ture. The temperature of the sample was raised in five-degree
increments until the N2 stretches were no longer observed. Once
the approximate binding temperature was recorded, the tem-
perature of the sample was then changed in approximately one-
degree increments until a more precise value was measured. The
error associated with the measurements is (2 °C.
Experimental Section
General Considerations. All air- and moisture-sensitive ma-
nipulations were carried out using standard high-vacuum-line,
Schlenk, or cannula techniques or in an M. Braun inert atmo-
sphere drybox containing an atmosphere of purified nitrogen.
The M. Braun drybox was equipped with a cold well designed
for freezing samples in liquid nitrogen. Solvents for air- and
moisture-sensitive manipulations were dried and deoxygenated
using literature procedures. Deuterated solvents for NMR
spectroscopy were purchased from Cambridge Isotope Labora-
tories and were distilled from sodium metal under an atmo-
˚
sphere of argon and stored over 4 A molecular sieves. Hydrogen
and argon gas were purchased from Airgas Incorporated and
passed through a column containing manganese oxide sup-
˚
ported on vermiculite and 4 A molecular sieves before admission
to the high-vacuum line. Carbon monoxide was passed through
a liquid nitrogen cooled trap immediately before use. 1, 1-Cl2,34
1-(CO)2,6a 2-Cl,35 2-(CO)2,27 2-Fe,28 3-Cl, 3-(N2)2,13 3-(CO)2,13
3-Fe,29 4-Cl, 4, 4-(CO)2, 4-(N2)2,13 5-Cl,36 5-Fe,30 7,11b 7-Cl2,37
7-(CO)2,14 7-(N2)2,13 8-Cl,10 8,10 9-Cl,38 9,11a 9-N2,14 and
9-(CO)214 were prepared according to literature procedures.
1H NMR spectra were recorded on Varian Mercury 300 and
Inova 400 and 500 spectrometers operating at 299.763, 399.780,
and 500.62 MHz, respectively. All chemical shifts are reported
relative to SiMe4 using 1H (residual) chemical shifts of the solvent
as a secondary standard. For paramagnetic compounds, 1H
NMR data are reported with the chemical shift followed by
the peak width at half-height in hertz, followed by integration
value and, where possible, peak assignment. Unless stated other-
wise, magnetic moments were measured at 22 °C by the method
originally described by Evans17 with stock and experimental
solutions containing a known amount of a ferrocene standard.
Single crystals suitable for X-ray diffraction were coated with
polyisobutylene oil in a drybox, transferred to a nylon loop, and
then quickly transferred to the goniometer head of a Bruker X8
APEX2 diffractometer equipped with a molybdenum X-ray
Preparation of [(η5-C5Me4Et)2Ti]2(μ2,η1,η1-N2) ([2]2N2). A
100 mL round-bottomed flask was charged with 15.48 g of
0.5% sodium amalgam and approximately 15 mL of toluene.
With vigorous stirring, 0.255 g (0.668 mmol) of 2-Cl was added
as a purple-blue toluene solution, and the resulting reaction
mixture was stirred for three days at ambient temperature. The
green solution was filtered through a pad of Celite, and the
toluene was removed in vacuo, leaving a green solid. Recrystal-
lization from pentane yielded 0.121 g (50%) of a green crystal-
line solution identified as [2]2N2. 1H NMR (benzene-d6): δ 14.71
(br s, 6H, C5Me4Et), 55.88 (br s, 12H, C5Me4Et), 64.12 (br s,
12H, C5Me4Et).
˚
tube (λ=0.71073 A). Preliminary data revealed the crystal
system. A hemisphere routine was used for data collection
and determination of lattice constants. The space group was
identified, and the data were processed using the Bruker
SAINT+ program and corrected for absorption using SA-
DABS. The structures were solved using direct methods
(SHELXS) completed by subsequent Fourier synthesis and
refined by full-matrix least-squares procedures. Elemental
Preparation of [(η5-C5Me4Et)2Ti(N2)]2(μ2,η1,η1-N2) ([2-
(N2)]2N2). A concentrated solution of [(η5-C5Me4Et)2Ti]2-
(μ2,η1,η1-N2) was recrystallized from pentane at -35 °C to yield
crystallographically pure [(η5-C5Me4Et)2Ti(N2)]2(μ2,η1,η1-N2)
as dark blue blocks. In addition, a React IR cell was charged
with 0.105 g (0.146 mmol) of [(η5-C5Me4Et)2Ti]2(μ2,η1,η1-N2) in
6.75 mL of pentane against a flow of dinitrogen. The cell was
cooled to -78 °C, and a spectrum was recorded under a flow of
dinitrogen. The temperature of binding was determined on the
basis of recording spectra at different temperatures and looking
for the N-N stretches. IR (pentane, -78 °C): ν, 1719 (bridging
N2), 2024, 2057 cm-1 (terminal N2).
(34) Bercaw, J. E.; Marvich, R. H.; Bell, L. G.; Brintzinger, H. H.
J. Am. Chem. Soc. 1972, 94, 1219.
(35) Mach, K.; Antropiusova, H.; Varga, V.; Hanus, V. J. Organo-
met. Chem. 1988, 358, 123.
(36) de Wolf, J. M.; Meetsma, A.; Teuben, J. H. Organometallics
1995, 14, 5466.
(37) Horacek, M.; Gyepes, R.; Cisarova, I.; Polesek, M.; Varga, V.;
Mach, K. Collect. Czech. Commun. 1996, 61, 1307.
(38) Lukesova, L.; Horacek, M.; Stepnicka, P.; Fejfarova, K.;
Gyepes, R.; Cisarova, I.; Kubista, J.; Mach, K. J. Organomet. Chem.
2002, 663, 134.
Preparation of (η5-C5Me4t Bu)2Fe (4-Fe). A 20 mL scintillation
vial was charged with 0.115 g (0.624 mmol) of Li[C5Me4CMe3],
0.040 g (0.312 mmol) of FeCl2, and approximately 5 mL Et2O.
The resulting reaction mixture was stirred overnight, changing
to a bright orange color. The solution was filtered through a pad

4088 Organometallics, Vol. 28, No. 14, 2009
Hanna et al.
of Celite, the solvent removed in vacuo, and the residue recrys-
tallized from cold pentane to yield 0.114 g (89%) of an orange
solid identified as 4-Fe. 1H NMR (benzene-d6): δ 1.40 (s, 18H,
C5Me4CMe3), 1.65 (s, 12H, C5Me4CMe3), 1.87 (s, 12H,
C5Me4CMe3). 13C NMR (benzene-d6): δ 10.77, 14.28 (Me),
32.60 (CMe3), 33.59 (CMe3), 76.85, 80.01, 92.95 (Cp). Mass
spectrum (direct detect, 70 eV), m/z (%): calc, 410.2636 (100);
found, 410.2633 (100).
6 and dissolved in approximately 0.5 mL of benzene-d6. On the
high-vacuum line, the tube was submerged in liquid nitro-
gen and degassed, and 1 atm of carbon monoxide was added
at -196 °C. The contents of the tube were thawed, producing a
color change to bright red, signaling formation of 6-(CO)2.
Anal. Calcd for C36H42O2Ti: C, 77.96; H, 7.63. Found: C,
1
77.94; H, 7.83. H NMR (benzene-d6): δ 1.78 (s, 12H, C5Me4-
3,5-Me2C6H3), 1.82 (s, 12H, C5Me4-3,5-Me2C6H3), 2.19 (s, 12H,
C5Me4-3,5-Me2C6H3), 6.72 (s, 2H, C5Me4-3,5-Me2C6H3), 6.89
(s, 4H, C5Me4-3,5-Me2C6H3). 13C NMR (benzene-d6): δ 11.47,
12.31 (Me), 21.36 (3,5-Me2), 104.64, 106.03, 112.07, 129.61,
136.02, 137.04 (Cp/Ph), 264.86 (CO). One Ph resonance not
located. IR (pentane): ν, 1860, 1940 cm-1 (CO).
Preparation of (η5-C5Me4Ph)2Ti (5). A 100 mL round-bot-
tomed flask was charged with 15.68 g of 0.5% sodium amalgam
and approximately 15 mL of toluene. With vigorous stirring,
0.321 g (0.672 mmol) of 5-Cl was added as a green toluene
solution, and the resulting reaction mixture was stirred for three
days at ambient temperature. The green-red solution was filtered
through a pad of Celite, and the toluene was removed in vacuo,
leaving a green-red solid. Recrystallization from pentane yielded
0.176 g (59%) of (η5-C5Me4Ph)2Ti. Anal. Calcd for C30H34Ti: C,
81.43; H, 7.74. Found: C, 81.08; H, 7.39. 1H NMR (benzene-d6):
δ -8.59 (br s, 4H, C5Me4Ph), 3.72 (s, 2H, C5Me4Ph), 9.47 (s, 4H,
C5Me4Ph), 47.14 (s, 12H, C5Me4Ph), 65.17 (s, 12H, C5Me4Ph).
Magnetic susceptibility (benzene-d6): μeff =2.7 μB.
Preparation of [(η5-C5Me4-3,5-Me2C6H3)2Ti]2(μ2,η1,η1-N2)
([62]N2). A 20 mL scintillation vial was charged with 0.160 g
(0.320 mmol) of 6 and approximately 5 mL of pentane. The
solution was cooled to -35 °C, resulting in the precipitation of
an analytically pure blue-green solid identified as [62]N2. Anal.
Calcd for C68H84N2Ti: C, 79.67; H, 8.26; N, 2.73. Found: C,
80.08; H, 8.41; N, 2.51.
Preparation of (η5-C5Me4SiMe3)2Fe (7-Fe). This molecule
was prepared in an identical manner to 4-Fe with 0.253 g
(1.26 mmol) of Li[C5Me4SiMe3] and 0.080 g (0.632 mmol) of
FeCl2 to yield 0.189 g (67.5%) of an orange solid identified as
7-Fe. 1H NMR (benzene-d6): δ 0.35 (s, 18H, C5Me4SiMe3), 1.70
(s, 12H, C5Me4SiMe3), 1.83 (s, 12H, C5Me4SiMe3). 13C NMR
(benzene-d6): δ 2.32 (SiMe3), 10.62, 13.18 (Me), 68.42, 83.53,
84.45 (Cp). Mass spectrum (direct detect, 70 eV), m/z (%): calc,
442.2175 (100); found, 442.2181 (100).
Preparation of (η5-C5Me4Ph)2Ti(CO)2 (5-(CO)2). A thick-
walled glass vessel was charged with 3.62 g of 0.5% sodium
amalgam and approximately 10 mL of pentane. With vigorous
stirring, 0.075 g (0.157 mmol) of 5-Cl was added as a pentane
solution and the vessel was brought out of the drybox and was
immediately submerged in liquid nitrogen. On the high-vacuum
line, the vessel was degassed and 1 atm of carbon monoxide was
added at -196 °C. The contents of the vessel were thawed, and
the reaction mixture was stirred for 18 h at ambient temperature.
The solvent was removed in vacuo, and the vessel was brought
into the drybox, filtered through a pad of Celite, and recrystal-
lized from pentane to yield 0.056 g (72%) of a red solid identified
as 5-(CO)2. 1H NMR (benzene-d6): δ 1.70 (s, 12H, C5Me4Ph),
1.71 (s, 12H, C5Me4Ph), 7.04 (s, 2H, C5Me4Ph), 7.17 (m, 8H,
C5Me4Ph). 13C NMR (benzene-d6): δ 11.38, 12.29 (Me), 104.79,
106.05, 111.44 (Cp), 126.38, 131.39, 136.15 (Ph), 264.59 (CO).
IR (pentane): ν, 1862, 1941 cm-1 (CO).
Preparation of (η5-C5Me4SiMe2t Bu)2Ti(CO)2 (8-(CO)2). A J.
Young NMR tube was charged with 0.015 g (0.029 mmol) of 8
and approximately 0.5 mL of benzene-d6. On the high-vacuum
line, the tube was degassed and 1 atm of carbon monoxide was
added at -196 °C. The tube was thawed and shaken, resulting in
a color change from purple-gray to orange. The solvent and
excess carbon monoxide were removed in vacuo, and the
compound was recrystallized from pentane to yield 0.012 g
(71%) of an orange solid identified as 8-(CO)2. 1H NMR
(benzene-d6): δ 0.38 (s, 12H, C5Me4SiMe2CMe3), 0.91 (s, 18H,
C5Me4SiMe2CMe3), 1.63 (s, 12H, C5Me4SiMe2CMe3), 1.80
(s, 12H, C5Me4SiMe2CMe3). 13C NMR (benzene-d6): δ -1.79
(SiMe2), 11.47, 14.64 (Me), 20.93 (CMe3), 27.88 (CMe3),
93.33, 109.99, 111.67 (Cp), 263.18 (CO). IR (pentane): ν, 1868,
1942 cm-1 (CO).
Preparation of (η5-C5Me4-3,5-Me2C6H3)2TiCl (6-Cl). A
100 mL round-bottomed flask was charged with 1.005 g (4.33 mmol)
of Li[C5Me4-3,5-Me2C6H3] and 0.802
g (2.16 mmol) of
TiCl3(THF)3. A reflux condenser and 180° needle valve were
then attached. On the high-vacuum line, 50 mL of tetrahydro-
furan was added by vacuum transfer at -78 °C, and the resulting
mixture warmed to ambient temperature while stirring. The
reaction mixture was then refluxed for 60 h, after which time the
solvent was removed in vacuo, leaving a dark solid. The residue
was brought into the drybox, dissolved in pentane, and filtered
through a pad of Celite. The solvent was removed in vacuo,
leaving a green solid. Recrystallization from pentane resulted in
0.736 g (64%) of 6-Cl. Anal. Calcd for C24H38ClTi: C, 70.33; H,
9.34. Found: C, 69.88; H, 9.18. 1H NMR (benzene-d6): δ -4.85
(Δν1/2=624 Hz), 2.12 (Δν1/2=37.6 Hz), 6.01 (Δν1/2=34.3 Hz).
Magnetic susceptibility (benzene-d6): μeff=1.5 μB.
Preparation of (η5-C5Me4SiMe2Ph)2Fe (9-Fe). This molecule
was prepared in an identical manner to 4-Fe with 0.429 g
(1.63 mmol) of Li[C5Me4SiMe2Ph] and 0.104 g (0.817 mmol)
of FeCl2 to yield 0.379 g (81.8%) of an orange solid identified as
1
9-Fe. H NMR (benzene-d6): δ 0.71 (s, 12H, C5Me4SiMe2Ph),
1.73 (s, 12H, C5Me4SiMe2Ph), 1.82 (s, 12H, C5Me4SiMe2Ph),
7.18 (m, 6H, C5Me4SiMe2Ph), 7.42 (d, J=8 Hz, 4H, C5Me4-
SiMe2Ph). 13C NMR (benzene-d6): δ 1.69 (SiMe2), 10.66, 13.30
(Me), 66.48, 83.99, 85.31 (Cp), 128.75, 133.92, 141.50 (Ph). One
Ph resonance not located. Mass spectrum (direct detect, 70 eV),
m/z (%): calc, 566.2488 (100); found, 566.2474 (100).
Preparation of (η5-C5Me4-3,5-Me2C6H3)2Ti (6). A 100 mL
round-bottomed flask was charged with 8.24 g of 0.5% sodium
amalgam and approximately 15 mL of toluene. With vigorous
stirring, 0.190 g (0.356 mmol) of 6-Cl was added as a green
toluene solution, and the resulting reaction mixture was stirred
for three days at ambient temperature. The red solution was
filtered through a pad of Celite, and the toluene was removed
in vacuo, leaving a red solid. Recrystallization of the residue
from pentane yielded 0.160 g (90%) of 6. Anal. Calcd for
C34H42Ti: C, 81.91; H, 8.49. Found: C, 81.74; H, 8.29. 1H
NMR (benzene-d6): δ -8.97 (br s, 4H, o-Ph), 1.40 (s, 12H,
C5Me4-3,5-Me2C6H3), 5.19 (s, 2H, p-Ph), 49.02 (br s, 12H,
C5Me4-3,5-Me2C6H3), 66.73 (br s, 12H, C5Me4-3,5-Me2C6H3).
Magnetic susceptibility (benzene-d6): μeff =2.5 μB.
Acknowledgment. We thank the Director, Office of
Basic Energy Sciences, Chemical Sciences Division, of
the U.S. Department of Energy (DE-FG02-05ER15659)
for financial support. P.J.C. is a Cottrell Scholar spon-
sored by the Research Corporation and a David and
Lucile Packard fellow in science and engineering. We also
thank the Collum group for access to a React IR spectro-
meter. We also would like to thank Professor Bruce
Bursten (Tennessee) for helpful discussions.
Supporting Information Available: Crystallographic data as
CIF files. This material is available free of charge via the Internet
at http://pubs.acs.org.
Preparation of (η5-C5Me4-3,5-Me2C6H3)2Ti(CO)2 (6-(CO)2).
A J. Young NMR tube was charged with 0.025 g (0.05 mmol) of