Decamethyltitanocene hydride intermediates in the hydrogenation of the corresponding titanocene-(η2-ethene) or (η2-alkyne) complexes and the effects of bulkier auxiliary ligands

Article abstract of DOI:10.1039/c7dt01545c

¹H NMR studies of reactions of titanocene [Cp?₂Ti] (Cp? = η⁵-C₅Me₅) and its derivatives [Cp?(η⁵:η¹-C₅Me₄CH₂)TiMe] and [Cp?₂Ti(η²-CH₂CH₂)] with excess dihydrogen at room temperature and pressures lower than 1 bar revealed the formation of dihydride [Cp?₂TiH₂] (1) and the concurrent liberation of either methane or ethane, depending on the organometallic reactant. The subsequent slow decay of 1 yielding [Cp?₂TiH] (2) was mediated by titanocene formed in situ and controlled by hydrogen pressure. The crystalline products obtained by evaporating a hexane solution of fresh [Cp?₂Ti] in the presence of hydrogen contained crystals having either two independent molecules of 1 in the asymmetric part of the unit cell or cocrystals consisting of 1 and [Cp?₂Ti] in a 2:1 ratio. Hydrogenation of alkyne complexes [Cp?₂Ti(η²-R¹CCR²)] (R¹ = R² = Me or Et) performed at room temperature afforded alkanes R¹CH₂CH₂R², and after removing hydrogen, 2 was formed in quantitative yields. For alkyne complexes containing bulkier substituent(s) R¹ = Me or Ph, R² = SiMe₃, and R¹ = R² = Ph or SiMe₃, successful hydrogenation required the application of increased temperatures (70-80 °C) and prolonged reaction times, in particular for bis(trimethylsilyl)acetylene. Under these conditions, no transient 1 was detected during the formation of 2. The bulkier auxiliary ligands η⁵-C₅Me₄^tBu and η⁵-C₅Me₄SiMe₃ did not hinder the addition of dihydrogen to the corresponding titanocenes [(η⁵-C₅Me₄^tBu)₂Ti] and [(η⁵-C₅Me₄SiMe₃)₂Ti] yielding [(η⁵-C₅Me₄^tBu)₂TiH₂] (3) and [(η⁵-C₅Me₄SiMe₃)₂TiH₂] (4), respectively. In contrast to 1, the dihydride 4 did not decay with the formation of titanocene monohydride, but dissociated to titanocene upon dihydrogen removal. The monohydrides [(η⁵-C₅Me₄^tBu)₂TiH] (5) and [(η⁵-C₅Me₄SiMe₃)₂TiH] (6) were obtained by insertion of dihydrogen into the intramolecular titanium-methylene σ-bond in compounds [(η⁵-C₅Me₄^tBu)(η⁵:η¹-C₅Me₄CMe₂CH₂)Ti] and [(η⁵-C₅Me₄SiMe₃)(η⁵:η¹-C₅Me₄SiMe₂CH₂)Ti], respectively. The steric influence of the auxiliary ligands became clear from the nature of the products obtained by reacting 5 and 6 with butadiene. They appeared to be the exclusively σ-bonded η¹-but-2-enyl titanocenes (7) and (8), instead of the common π-bonded derivatives formed for the sterically less congested titanocenes, including [Cp?₂Ti(η³-(1-methylallyl))] (9). The molecular structure optimized by DFT for compound 1 acquired a distinctly lower total energy than the analogously optimized complex with a coordinated dihydrogen [Cp?₂Ti(η²-H₂)]. The stabilization energies of binding the hydride ligands to the bent titanocenes were estimated from counterpoise computations; they showed a decrease in the order 1 (-132.70 kJ mol^-1), 3 (-121.11 kJ mol^-1), and 4 (-112.35 kJ mol^-1), in accordance with the more facile dihydrogen dissociation.

Full text of DOI:10.1039/c7dt01545c

Dalton
Transactions
View Article Online
View Journal
PAPER
Decamethyltitanocene hydride intermediates in
the hydrogenation of the corresponding
Cite this: DOI: 10.1039/c7dt01545c
titanocene-(η2-ethene) or (η2-alkyne) complexes
and the effects of bulkier auxiliary ligands†
Jiří Pinkas,a Róbert Gyepes,b,c Ivana Císařová,c Jiří Kubišta,a Michal Horáčeka and
a
Karel Mach
*
1H NMR studies of reactions of titanocene [Cp*2Ti] (Cp* = η5-C5Me5) and its derivatives [Cp*(η5:η1-C5Me4CH2)
TiMe] and [Cp*2Ti(η2-CH2vCH2)] with excess dihydrogen at room temperature and pressures lower than
1 bar revealed the formation of dihydride [Cp*2TiH2] (1) and the concurrent liberation of either methane or
ethane, depending on the organometallic reactant. The subsequent slow decay of 1 yielding [Cp*2TiH] (2) was
mediated by titanocene formed in situ and controlled by hydrogen pressure. The crystalline products obtained
by evaporating a hexane solution of fresh [Cp*2Ti] in the presence of hydrogen contained crystals having
either two independent molecules of 1 in the asymmetric part of the unit cell or cocrystals consisting of 1 and
[Cp*2Ti] in a 2 : 1 ratio. Hydrogenation of alkyne complexes [Cp*2Ti(η2-R1CuCR2)] (R1 = R2 = Me or Et) per-
formed at room temperature afforded alkanes R1CH2CH2R2, and after removing hydrogen, 2 was formed in
quantitative yields. For alkyne complexes containing bulkier substituent(s) R1 = Me or Ph, R2 = SiMe3, and R1 =
R2 = Ph or SiMe3, successful hydrogenation required the application of increased temperatures (70–80 °C)
and prolonged reaction times, in particular for bis(trimethylsilyl)acetylene. Under these conditions, no transient
1 was detected during the formation of 2. The bulkier auxiliary ligands η5-C5Me4 Bu and η5-C5Me4SiMe3 did
t
not hinder the addition of dihydrogen to the corresponding titanocenes [(η5-C5Me4 Bu)2Ti] and [(η5-
t
t
C5Me4SiMe3)2Ti] yielding [(η5-C5Me4 Bu)2TiH2] (3) and [(η5-C5Me4SiMe3)2TiH2] (4), respectively. In contrast to 1,
the dihydride 4 did not decay with the formation of titanocene monohydride, but dissociated to titanocene
upon dihydrogen removal. The monohydrides [(η5-C5Me4 Bu)2TiH] (5) and [(η5-C5Me4SiMe3)2TiH] (6) were
t
obtained by insertion of dihydrogen into the intramolecular titanium-methylene σ-bond in compounds [(η5-
C5Me4 Bu)(η5:η1-C5Me4CMe2CH2)Ti] and [(η5-C5Me4SiMe3)(η5:η1-C5Me4SiMe2CH2)Ti], respectively. The steric
t
influence of the auxiliary ligands became clear from the nature of the products obtained by reacting 5 and 6
with butadiene. They appeared to be the exclusively σ-bonded η1-but-2-enyl titanocenes (7) and (8), instead
of the common π-bonded derivatives formed for the sterically less congested titanocenes, including [Cp*2Ti
(η3-(1-methylallyl))] (9). The molecular structure optimized by DFT for compound 1 acquired a distinctly lower
total energy than the analogously optimized complex with a coordinated dihydrogen [Cp*2Ti(η2-H2)]. The
stabilization energies of binding the hydride ligands to the bent titanocenes were estimated from counterpoise
computations; they showed a decrease in the order 1 (−132.70 kJ mol−1), 3 (−121.11 kJ mol−1), and 4
(−112.35 kJ mol−1), in accordance with the more facile dihydrogen dissociation.
Received 27th April 2017,
Accepted 30th May 2017
DOI: 10.1039/c7dt01545c
rsc.li/dalton
aJ. Heyrovský Institute of Physical Chemistry, Czech Academy of Sciences, v.v.i.,
Dolejškova 3, 182 23 Prague 8, Czech Republic. E-mail: karel.mach@jh-inst.cas.cz
bDepartment of Chemistry, Faculty of Education, J. Selye University, Bratislavská
cesta 3322, 945 01 Komárno, Slovak Republic
cDepartment of Inorganic Chemistry, Charles University, Hlavova 2030,
128 40 Prague 2, Czech Republic
Introduction
The gradual development of opinions on dihydrogen activation
and on various modes of its interaction with transition metals,
as well as the role of transition metal hydrides in the hydro-
genation of organic substrates during the past 50 years, has
been reviewed by G. J. Kubas1 and by the recent state-of-art
article by R. H. Crabtree2 in a special issue of Chemical Reviews
devoted to various aspects of dihydrogen activation. The
notable absence of group IVB early transition metal (Ti, Zr,
†Electronic supplementary information (ESI) available: A table of crystallo-
graphic data and data collection and structure refinement details for 1A, 1B, 5,
and 8. Experimental and characterization details, including spectroscopic data
(NMR, UV, EPR), and X-ray crystallographic data. Cartesian coordinates and
energy data for calculated structures. CCDC 1545964–1545967. For ESI and crys-
tallographic data in CIF or other electronic format see DOI: 10.1039/c7dt01545c
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
and Hf) metallocenes in considering the binding of molecular
hydrogen might reflect both the difficulties in obtaining the d2
metallocenes containing empty dσ and filled dπ suitable to
combine with H2(σ) and H2(σ*) orbitals,3 and/or the high elec-
tropositivity of these metals with their Pauling electronegativity
indexes of 1.5, 1.3 and 1.3,4 preferring the formation of
hydrides. Actually, the thermally stable Zr and Hf decamethyl-
metallocene dihydrides [Cp*2ZrH2]5 and [Cp*2HfH2]6 are
already known and have been properly characterized. Their
reactivity towards olefins and alkynes is also clearly under-
stood. The subsequent regular insertion of double or triple
bonds into their M–H bonds afforded clean metallocene
hydride-alkyls or -alkenyls and dialkyls or dialkenyls.7 For
titanium, various synthetic pathways to titanocene and their
hydrides as depending on the substituents of titanocene
auxiliary ligands were mentioned by J. P. Chirik in an overview
on low-valent nitrogen titanocene complexes,8 and a full
account on early transition metal hydride complexes was
reported in 2002 by A. J. Hoskin and D. W. Stephan.9 In
contrast to the above mentioned Zr and Hf dihydrides, the
decamethyltitanocene analogue [Cp*2TiH2] (1) was found to be
unstable in its solution at ambient temperatures; its decompo-
sition to titanocene [Cp*2Ti] and dihydrogen was sugges-
ted.10a,b The subsequent study of the hydrogenation of the
singly tucked-in titanocene methyl [Cp*(C5Me4CH2)TiMe]
revealed that a rapid uptake of 2 molecules of hydrogen result-
ing in elimination of methane was followed by a slower liber-
ation of a half equivalent of dihydrogen. This explanation
suggested the formation of 1 (eqn (1)) followed by its partial
dehydrogenation yielding titanocene [Cp*2Ti] (eqn (2)), which
comproportionated with the remaining 1 to yield the final
[Cp*2TiH] (2) (eqn (3)).10c
Scheme 1 Tautomerization of decamethyltitanocene and its decay.12
transient diamagnetic symmetrical doubly tucked-in titano-
cene [(η5:η1-C5Me4CH2)2Ti] which further isomerizes to final
doubly tucked-in titanocene having vicinal exo-methylene
groups [Cp*(η3:η4-C5Me3(CH2)2)Ti] (Scheme 1).12,13
The monohydride 2 has been shown to be an excellent
hydrogen transfer agent in hydrogenations of unsaturated
hydrocarbons acting in a cycle with [Cp*(η5:η1-C5Me4CH2)Ti]
(Scheme 2)14 and taking part in thermal decomposition of
alkyltitanocenes.15
Among other known titanocenes, bis(1,3-di-tert-butylcyclo-
pentadienyl)titanium [(1,3-tBu2C5H3)2Ti] and [(C5Me4 Pr)2Ti]
i
were reported to react with dihydrogen at low temperatures to
give titanocene dihydrides [(1,3-tBu2C5H3)2TiH2]16 and
[(C5Me4 Pr)2TiH2]17 which differed by thermal stability and
i
reactivity. On warming their solutions to ambient temperature
and degassing, the former cleanly dissociated to dihydrogen
and the titanocene whereas the latter gave rise to the monohy-
dride [(C5Me4 Pr)2TiH],17 similarly to 1. The former dihydride
i
[(1,3-tBu2C5H3)2TiH2] was obtained from hexane solution at
−80 °C as dark red crystals that were stable under vacuum and
melted at 152–154 °C.16 For both dihydrides their 1H NMR
spectra were determined at low temperature in the absence of
their parent titanocenes. Following these achievements a
thorough investigation of the formation of 1 and its properties
½Cp*ðC5Me4CH2ÞTiMeꢀ þ 2H2 ! ½Cp*2TiH2ꢀ þ CH4 ð1Þ
½Cp*2TiH2ꢀ , ½Cp*2Tiꢀ þ H2
ð2Þ
ð3Þ
½Cp*2TiH2ꢀ þ ½Cp*2Tiꢀ , 2½Cp*2TiHꢀ
All the species appearing in this system have been assigned
their 1H NMR chemical shifts, and X-ray single crystal struc-
tures were determined for 211 and for decamethyltitanocene in
the co-crystal with its monochloride [Cp*2Ti/Cp*2TiCl] (1 : 1).12
The above system is actually more complicated due to the
instability of [Cp*2Ti], which disproportionates to paramag-
netic 2 and singly tucked-in titanocene [Cp*(η5:η1-C5Me4CH2)
Ti] (equilibrium (4))12 and rapidly tautomerizes to the diamag-
netic singly tucked-in titanocene hydride [Cp*(η5:η1-
C5Me4CH2)TiH].10b Upon aging, the tautomer spontaneously
dehydrogenates to give
ð4Þ
Scheme 2 Hydrogenation of internal alkynes mediated by [Cp*2TiH].14
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
and the effects of its yet bulkier cyclopentadienyl auxiliary intensity of the methane signal at δH 0.17 ppm (cf. ref. 18)
ligands on the properties of titanocene hydride is to be proved that methane elimination was already accomplished
1
sought.
before acquiring the first H NMR spectrum 30 minutes after
Herein we report on the detailed studies of dihydrogen reac- adding hydrogen, as the signal intensity showed no increase
tions carried out under mild conditions (22–70 °C, <1 bar) with later measurements. The progress of hydrogenating [Cp*
with [Cp*2Ti], [Cp*(η5:η1-C5Me4CH2)TiMe], 2, [Cp*2Ti(η2- (C5Me4CH2)TiMe] was also followed by measuring its elec-
ethene)] and a number of titanocene-(η2-alkyne) complexes, tronic absorption spectra in the visible region, taking advan-
with particular attention paid to the formation of transient tage of its thermal stability in comparison with that of
titanocene dihydride species. The effects of bulky auxiliary [Cp*2Ti].12 The dark green color of the reagent (λmax
=
t
ligands η5-C5Me4 Bu or η5-C5Me4SiMe3 on the stability of the 607 nm)19 turned brownish-yellow after about 30 min, which
corresponding titanocene dihydrides and on the structure of allowed assigning the new absorption band at 483 nm to 1.
the products of butadiene addition to the titanocene mono- After degassing and aging for 5 days, the color of the solution
hydrides are also investigated.
turned reddish-ochre, displaying an absorption band at
477 nm and a shoulder at 550 nm, which was in agreement
with data of Andersen et al.11 for 2 (for visible spectra see the
ESI†).
Results and discussion
1H NMR monitoring of the hydrogenation of [Cp*2Ti(η2-
C2H4)]20 revealed that reaction (5) was accomplished in
10 minutes upon warming the frozen solution to room temp-
Reaction of hydrogen with [Cp*2Ti], [Cp*(η5:η1-C5Me4CH2)
TiMe], and [Cp*2Ti(η2-C2H4)]
Admission of dihydrogen (∼4.2 mmol) to titanocene [Cp*2Ti] erature (see Fig. S3 and S4 in the ESI†).
(0.2 mmol) in toluene-d8 (1.0 mL) caused an immediate
½Cp*2Tiðη2-CH2vCH2Þꢀ þ 2H2 ! ½Cp*2TiH2ꢀ þ C2H6
ð5Þ
change of its light green color (λmax ∼ 630 nm) to yellow-
brown, while the complete disappearance of its paramagnetic
1
resonance at 63.8 ppm (Δν1/2 = 170 Hz) in the broad-range H
1H NMR signals of the reagent disappeared and that of
NMR spectrum was observed. The latter signal was replaced by ethane at δH 0.81 ppm (cf. ref. 18) did not grow further. The
a dominant broad resonance at δH 1.90 ppm (Δν1/2 = 60 Hz) approximate composition of titanocene products – [Cp*2Ti]
accompanied by a weaker one at δH 3.37 ppm (Δν1/2 ∼ 70 Hz) 10%, 1 88%, and 2 2% showed that the dihydrogen in solution
and a weak resonance at δH 23.0 ppm (Δν1/2 = 170 Hz) due to was substituted by ethane, since the titanocene resonance
2.10c,11,14 The dominant resonance occurred at the chemical could not be observed in the presence of the dihydrogen. After
shift found by Brintzinger et al.10b in the same reacting system 40 min the titanocene signal disappeared, whereas that of 2
and was assigned to the Cp* methyls of 1. However, we could grew in intensity at the expense of the signals of 1 appearing at
not find any signal at 0.28 ppm in our spectra, which was 1.88 ppm and 3.37 ppm. With continuing conversion of 1 to 2
assigned by them to the hydride nuclei. Since the integral the former signal shifted to 1.91 ppm and turned narrower
intensity ratio of the signals at 1.90 ppm and 3.37 ppm was (Δν1/2 = 10 Hz), whereas the latter broadened so that it became
close to 15 and the signals displayed only marginal changes in imperceptible. The latter effect can be attributed to an exten-
their chemical shifts upon decreasing the sample temperature sive exchange of the both hydrides with dihydrogen, whose
to −45 °C, we have assigned the resonance at 3.37 ppm to the concentration should be increasing upon conversion of 1 to 2
hydride moieties of diamagnetic [Cp*2TiH2]. Concurrently, the (via eqn (2) followed by eqn (3) or eqn (4)). After 3 months this
signal at 1.90 ppm narrowed down to give Δν1/2 = 16 Hz, which system yielded an approximate equilibrium composition of 2
allowed the measurement of the 13C NMR resonance signals at (65%) and 1 (35%), similar to the above mentioned systems. In
δC (C5Me5) 13.34 ppm and δC (C5Me5) at 119.79 ppm (see all these hydrogenated systems the subsequent formation of 2
Fig. S1 and S2 in the ESI†). On aging, the above solution was accelerated by removing free dihydrogen under vacuum,
showed a steady, slow increase in the signal intensity of 2, at apparently shifting the equilibrium (2) to the right side, thus
the expense of both resonances of 1, giving approximately supplying more [Cp*2Ti] for comproportionation with
1
60% of 2 and 40% of 1 after 3 months at room temperature. (eqn (3)) and/or its disproportionation (eqn (4)). A rapid degas-
Unfortunately, the expected increase in dihydrogen concen- sing of the reaction solution resulted even in transient recovery
tration could not be monitored by 1H NMR spectroscopy, of [Cp*2Ti], whose signal at 63.8 ppm was observed for
because the resonance of dihydrogen at δH 4.50 ppm (cf. ref. approximately 10 min. The full conversion of 1 to 2 was
18) was quenched by traces of both hydrides, apparently due achieved after complete dihydrogen removal by twice repeating
to exchanges causing a large signal broadening. In contrast, the freezing/thawing cycle under vacuum.
the paramagnetic [Cp*2TiCl], used as a signal intensity stan-
dard in most of the investigated systems, did not affect the the signal of 1 at 1.91 ppm occurred after adding dihydrogen
dihydrogen signal. to a toluene-d8 solution of 2, its area amounting to about 10%
The above results were complemented by the discovery that
The analogous hydrogenation of thermally stable [Cp* of the signal area of 2 at 23.0 ppm. Measurements of this
(C5Me4CH2)TiMe] afforded virtually identical resonances for sample between −45 °C and +65 °C showed that the signals at
both the major 1 and the minor 2 products. In addition, the 3.37 ppm and 1.90 ppm were clearly observed at the lowest
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
temperature; however, the former broadened with increasing as suggested by eqn (3).10c The steric hindrance for this
temperature, becoming invisible at 25 °C and the latter shifted bimolecular reaction could be partially relieved by reacting
slightly at this temperature to 1.91 ppm. Upon increasing the with the bent titanocene tautomer. The viability of the compro-
temperature further, the latter signal broadened strongly and portionation reaction was supported by obtaining evidence
moved to lower magnetic fields to coalesce finally into one that a similar sterically demanding reaction between excess
broad signal at about 2.6 ppm at 65 °C (see Fig. S5 and S6 in [Cp*2Ti] and [Cp*2Ti(η2-C2H4)] gave rise to [Cp*(C5Me4CH2)Ti]
the ESI†). This coalescence apparently resulted from exchanges and ethane, until all the ethene complex was consumed.
involving the hydrides, the Cp* methyl hydrogen atoms and Only then the remaining [Cp*2Ti] disproportionated (eqn (4))
dihydrogen, which are known to proceed effectively during to give [Cp*(C5Me4CH2)Ti] and 2, whose formation was clearly
deuteration of 2.11
detected by the δH 23.0 ppm signal (see Fig. S7 and S8 in the
An overview of the reactions of titanocene reagents with ESI†).
excess dihydrogen followed by 1H NMR is depicted in
The addition of dihydrogen to 2 which generates 1 is evi-
Scheme 3. Concerning the reaction mechanism, the bent tita- dently a retro-process involving equilibrium (4) and a high
nocene derivatives [Cp*(C5Me4CH2)TiMe] and [Cp*2Ti(η2- affinity of titanocene to dihydrogen. The retro-reaction (3) is
C2H4)] should coordinate dihydrogen which upon insertion difficult to realize since the disproportionation of two mole-
into the Ti–C bond(s) would yield titanocene and methane or cules of 2 should be initiated by dihydrogen addition. There
ethane, respectively. In particular, titanocene itself reacted are no signs of dimerization of 2 alone11 and no evidence of
with hydrogen immediately and completely as evident from interaction of 2 with dihydrogen from 1H NMR spectra; the
the absence of its signal at 63.8 ppm and the absence of its paramagnetic shift of its resonance at 23.0 ppm (25 °C) and
tautomer resonance at 2.19 ppm. Apparently, both the bent linewidth did not change in the presence of dihydrogen in the
titanocene tautomer and the parallel titanocene are capable of range from −45 to +65 °C (cf. Fig. S9 and S10 with S5 and S6 in
coordinating dihydrogen and then accomplishing the oxidative the ESI†).
addition rearrangement to 1. A slow conversion of 1 to 2 which
The surprising thermal stability of 1 in the solid state was
runs to equilibrium composition depending on hydrogen established in the ocher-orange crystalline material obtained
pressure should be mediated by titanocene although its equili- after admission of hydrogen to a hexane solution of [Cp*2Ti]
1
brium concentration was either very low or its H NMR signal followed by slow distillation of the solvent under a hydrogen
was unobservable due to its broadening, e.g. due to exchanges atmosphere to a trap cooled by liquid nitrogen. The X-ray
between the titanocene containing the coordinated dihydrogen single crystal analysis of selected orange plates revealed two
[Cp*2Ti(H2)] and dihydrogen. Under consideration is the crystal entities. The first fraction of crystals, obtained when
disproportionation of titanocene according to equilibrium (4) maintaining the evaporated solution at −5 °C, contained two
which in the presence of dihydrogen would yield only 2 due independent molecules of 1 in the asymmetric part of the unit
to dihydrogen addition to [Cp*(C5Me4CH2)Ti].14 Such an cell (crystal A). The second fraction (crystal B), which was
equilibrium for pure 2 contained trace amounts of [Cp*2Ti] obtained by evaporation at room temperature, yielded cocrys-
and [Cp*(C5Me4CH2)Ti] detectable in amplified 1H NMR tals of 1 and [Cp*2Ti] in a 2 : 1 ratio. The important geometric
spectra, however, only in the strict absence of dihydrogen. The parameters of the different molecules of 1 and DFT optimized
other possibility is the comproportionation of [Cp*2Ti] with 1 structures of the dihydride and the titanocene-dihydrogen
complex are listed in Table 1.
Although the data for the hydride atoms – which all were
localized on Fourier maps – show a mediocre precision in
determining their position, their comparison with the data of
molecules optimized by DFT methods nonetheless excludes
the possibility that these hydrogen atoms might belong to a
titanocene-dihydrogen complex. All the dihydride molecules
are contaminated with an admixture of [Cp*2TiCl] (see the
Experimental section). The titanocene molecule in crystal 1B
showed parallel cyclopentadienyl rings with the Ti–Cg distance
of 2.0021(11) Å, which is slightly longer than the distance of
1.983(2) Å found in the co-crystal of [Cp*2Ti] and [Cp*2TiCl]
(1 : 1) (Fig. 1).12
Crystalline samples of the same appearance as the above
single crystals were subjected for melting point determination
in sealed capillaries under argon. They softened at 155 °C (1A)
and at 165 °C (1B) and both slowly sublimed onto the capillary
walls. 1H NMR spectra of average crystalline samples (1A/1B)
in toluene-d8 revealed the following composition: [Cp*2Ti]
together with its tautomer 12%/24%, 2 13%/9%, and 1 75%/
Scheme 3 Reactions of decamethyltitanocene and its derivatives with
hydrogen.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Table 1 Selected interatomic distances (Å) and bond angles (°) for crystals 1A and 1B and DFT-computed optimized Cp*2TiH2 and Cp*2Ti(η2-H2)
molecules
1A
DFT optimized structure
Cp*2TiH2
Crystal
Molecule
1B
1
2
1a
Cp*2Ti(η2-H2)
Ti–Cg(1)b
Ti–Cg(2)b
Ti–Pl(1)c
2.0414(10)
2.0391(9)
2.0414(8)
2.0391(8)
1.63(3)
1.68(3)
2.31(4)
149.24(4)
88.9(14)
29.93(9)
2.0426(10)
2.0372(11)
2.0391(8)
2.0424(8)
1.59(4)
1.66(4)
2.13(5)
148.56(4)
82(2)
30.3(8)
2.0392(12)
2.0337(13)
2.0388(11)
2.0334(12)
1.88(4)
1.76(4)
1.58(9)
149.22(5)
51(3)
28.7(2)
2.024
2.028
—
1.979
1.981
—
Ti–Pl(2)c
—
—
Ti–H(lA)
Ti–H(1B)
H(1A)–H(1B)
Cg(1)–Ti–Cg(2)b
H(1A)–Ti–H(1B)
φd
1.730
1.727
2.467
147.909
91.066
—
1.799
1.803
0.911
155.265
29.289
—
a Symmetry operation used to generate equivalent positions: 1 − x, −y + 2, −z + 1. b Cg(1) and Cg(2) denote the centre of gravity of the cyclopenta-
dienyl rings. c Cyclopentadienyl ring least-squares planes containing the center of gravity having the same numeral label. d Angle between the
least-squares planes of cyclopentadienyl rings.
a decreased content of 1 indicated both by the crystal structure
and by 1H NMR measurement.
Hydrogenation of Cp*2Ti(η2-R1CuCR2) complexes
Decamethyltitanocene-alkyne
complexes
[Cp*2Ti(η2-
R1CuCR2)]21 (0.3–0.5 mmol) in toluene-d8 (4.0 mL) were
hydrogenated with at least 4.2 mmoles of gaseous dihydrogen
in a constant volume. This hydrogen amount thus always sur-
passed the stoichiometry for the conversion of organic ligands
attached to titanium to corresponding alkanes and formation
of 1 (eqn (6)).
½Cp*2Tiðη2-R1CuCR2Þꢀ þ 3H2 ! ½Cp*2TiH2ꢀ þ R1CH2CH2R2
ð6Þ
The kinetics of hydrogen consumption was not followed;
the hydrogenation was terminated by vacuum distillation of
all volatiles into a liquid nitrogen-cooled trap after the typical
ocher-red color of 2 appeared and did not change further
visually for the last third of the total reaction time. Organic
products were separated by vacuum distillation at tempera-
tures never exceeding 80 °C and were analyzed by GC-MS
and/or 1H and 13C NMR spectroscopy (see the ESI†). The
residue was dissolved in toluene-d8 and analyzed by broad
range 1H NMR spectroscopy always displaying the overwhelm-
ing broad resonance of paramagnetic 2 at 23.0 ppm at 25 °C.
The course of hydrogenation was found to depend strongly
on the nature of the alkyne substituents, presumably on their
bulkiness. Whereas the 2-butyne and 3-hexyne complexes
were hydrogenated at room temperature within 8 h, the ana-
Fig. 1 PLATON drawing of [Cp*2TiH2] (crystal 1A, molecule 1) at the
30% probability level and with the atom labeling scheme. Hydrogen
atoms except the hydrides and the disordered chloride atom are
omitted for clarity.
67% after 40 min of aging at room temperature, the period logues containing phenyl, and especially trimethylsilyl substi-
required for sample preparation. The infrared spectra of the tuent(s) required heating to 70 °C for a prolonged reaction
average crystalline samples in KBr pellet were very similar, time.
However,
the
hydrogenation
of
[Cp*2Ti(η2-
showing an absorption band at 1560 cm−1 and a shoulder at Me3SiCuCSiMe3)]21c was not accomplished even after 112 h
1586 cm−1 of medium intensity, in general accordance with at 80 °C (Table 2).
the initial report for [Cp*2TiH2].10b Their lower intensity com-
The hydrogenation of [Cp*2Ti(η2-PhCuCPh)] and [Cp*2Ti
pared with the IR spectrum of 2, which showed a very strong (η2-Me3SiCuCSiMe3)] was also followed by 1H NMR spectra.
band at 1509 cm−1 for the valence Ti–H vibration accounts for About one half of the former complex was converted after 1 h
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 2 Hydrogenation of Cp*2Ti(R1CuCR2) complexesa
R1
R2
Temperature (°C)
Time (h)
Conversion to 2 (%)
Organic productb
Conversion (%)
Me
Et
Ph
Me
Et
Ph
22
22
22
70
70
70
80
4
8
12
32
32
48
112
100
100
5
100
100
100
87
Me(CH2)2Me
Me(CH2)4Me
Not determined
Ph(CH2)2Ph
Ph(CH2)2SiMe3
Me(CH2)2SiMe3
Me3Si(CH2)2SiMe3
100
100
—
100
100
100
45
Ph
Me
SiMe3
SiMe3
SiMe3
SiMe3
c
a Excess dihydrogen (∼4.2 mmol) was admitted to the reagent (0.3–0.5 mmol) in toluene-d8 (1.0 mL) in a constant volume. b Volatile organic pro-
ducts were vacuum-distilled together with toluene-d8 and analyzed by 1H and 13C NMR spectra and the GC-MS method. c A mixture with 1,2-bis
(trimethylsilyl)ethene ((E-) 50%, (Z-) 5%).
at room temperature into 1 (δH 1.88 ppm (Δν1/2 = 65 Hz) and Hydrogenation of doubly tucked-in titanocenes
3.37 ppm (Δν1/2 = 46 Hz)) and cis-stilbene (δH 6.43 ppm),
It is of practical interest to know whether [Cp*Ti(η3:η4-
whereas only a trace of 2 was observed. After 72 h at room
C5Me3(CH2)2)] as the main, thermally stable decay product of
temperature the reagent hydrogenation was accomplished
[Cp*2Ti]12 can be converted by reacting with dihydrogen to cat-
yielding overwhelming amounts of 2 and cis-stilbene. Heating
alytically active 2. Two doubly tucked-in compounds
the mixture to 100 °C for 1 h resulted in hydrogenation of the
[Cp*Ti(η3:η4-C5Me3(CH2)2)]13 and [Ti{η3:η4-C5Me3(CH2)2}{η5-
majority of cis-stilbene to 1,2-diphenylethane whereas no
C5Me3(CH2CHtBuCHvCHCHtBuCH2)}]22 (C1) in toluene-d8 solu-
decay was noticed for 2 (see Fig. S11 and S12 in the ESI†). The
tions showed no visible change of their blue color after introdu-
cing dihydrogen at room temperature within 3 days. This proved
absence of a [Cp*(C5Me4CH2)Ti] signal at δH −11.7 ppm indi-
cated that the hydrogenation was run with dihydrogen and 2
that the dehydrogenation of the titanocene tautomer followed by
participated only as a hydrogen transfer agent.14
rearrangement to [Cp*Ti(η3:η4-C5Me3(CH2)2)] (Scheme 1) is practi-
An attempt to shed light on the mechanism of the dipheny-
cally irreversible at room temperature. However, heating to 70 °C
lacetylene ligand hydrogenation by using dideuterium failed,
for 26 h resulted in converting [Cp*Ti(η3:η4-C5Me3(CH2)2)] to 2
as the resonance of cis-stilbene at 6.43 ppm did not reveal a
and compound C1 to [Cp*Ti{η5:η2-C5Me3(CH2CHtBuCHv
recognizable presence of deuterated species. However, the
CHCHtBuCH2)}] (C2).23a The latter compound was previously
obtained by hydrogenating the intramolecular acetylenic
signal of [Cp*2TiH2] at 3.45 ppm appeared to be overlapped
with the triplet 1 : 1 : 1 signal of [Cp*2Ti(H)D] showing the JHD
complex [Cp*Ti{η5:η2-C5Me3(CH2CHtBuCuCCHtBuCH2)}] (C3)
coupling of ∼3.4 Hz that is a typical value of the metal
hydride-deuteride species1 (see Fig. S13 in the ESI†). These
results indicate that a considerable part of deuterium under-
went rapid exchanges with the Cp* ligands; an assessment of
deuterium distribution over the system by a combined GC-MS
and NMR study has not yet been attempted.
(Scheme 4).23b The structure of product C2 indicates that the tita-
nocene [Cp*Ti{η5-C5Me3(CH2CHtBuCHvCHCHtBuCH2)}] was a
transient reaction product which subsequently stabilized by coor-
dinating the intramolecular double bond. Further hydrogenation
of C2 at 80 °C brought about an increased elimination of 2,7-
tetramethyloct-4-ene isomers, indicating the hydrogenolysis of the
bicyclic ligand.
The [Cp*2Ti(η2-Me3SiCuCSiMe3)] complex remained intact
after exposure to hydrogen at room temperature for 3 days. At
80 °C the signal of 2 began to grow gradually in time while no
signals of 1 were detected. The hydrogenation of the reagent
was not accomplished even after 112 h at 80 °C, when the
experiment was terminated. The acetylene ligand was hydro-
genated to cis-alkene, the latter concurrently isomerized to
trans-alkene, and both were more slowly hydrogenated to the
alkane (Table 2).
Effects of bulky auxiliary ligands η5-C5Me4 Bu or η5-
t
C5Me4SiMe3 on the stability of titanocene dihydrides and on
the structure of the products of addition of butadiene to tita-
nocene monohydrides
Replacement of one methyl group on each Cp* ligand by the
more bulky t-butyl or trimethylsilyl group was previously
Scheme 4 Hydrogenation of doubly tucked-in titanocene C1 and the other precursor C3 to C2.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
shown to change the reactivity of titanocenes towards dinitro-
gen,24 CO, NH3, and Me3SiN3,17 and the titanocene-ethene
complexes to but-2-yne,12 apparently due to the increased steri-
cal congestion in the bent titanocene moiety. Both titano-
t
cenes, the brown [(C5Me4 Bu)2Ti]12,24 and the green
[(C5Me4SiMe3)2Ti]25a indicated their reaction with hydrogen by
the immediate change of their colors to ocher. 1H NMR
spectra revealed the presence of minor signals of the titano-
cene monohydrides as impurities (see below) and broad
signals at about 0–3 ppm likely due to expected dihydrides
t
[(C5Me4 Bu)2TiH2] (3) and [(C5Me4SiMe3)2TiH2] (4). These
resolved at −35 °C into single resonances for the TiH2 moiety,
one pair of singlets for two different methyl groups on the
cyclopentadienyl ring C5Me4 and one singlet due to the CMe3
or SiMe3 group in the order of decreasing δH and in appropri-
ate intensity ratios. Their 13C NMR spectra were also obtained
as seven and six singlets, respectively (see Fig. S14–S16 in the
ESI†).
Fig. 3 An expanded region of Fig. 2 demonstrates broadening and sub-
sequent coalescence of 1H NMR signals of 4 with increasing tempera-
ture. The signal highlighted by blue bar corresponds to the SiMe3 group
of [(C5Me4SiMe3)2TiH] (6). (*) denotes the solvent signal.
Simultaneously with
4 the considerable presence of
[(C5Me4SiMe3)2Ti] was also established from its outmost low
magnetic field resonance at 99.6 ppm (Fig. 2) implying an
equilibrium (7) whose composition depended on the hydrogen [(η5-C5Me4SiMe3)(η5:η1-C5Me4SiMe2CH2)Ti] or similar com-
pressure.
pounds which resulted from thermal treatment of the titanoce-
ne,25b and not from the decay of 4, as known for the Cp*
system above.
½ðC5Me4SiMe3Þ2Tiꢀ þ H2 $ ½ðC5Me4SiMe3Þ2TiH2ꢀ
ð7Þ
An attempted preparation of [(C5Me4SiMe3)2TiH(D)] by
reacting the titanocene with HD apparently yielded the
required compound contaminated with the dihydride, since
the integral intensity of the distorted signal at 2.78 ppm was
1.4 instead of the theoretical value 1.0, with respect to the inte-
gral value of 12 for each of the two δH (C5Me4) signals (see
Fig. S17 in the ESI†). The distorted signal was composed of
singlet dihydride and triplet 1 : 1 : 1 for hydride–deuteride
showing the JHD coupling of ∼6 Hz. This value is compatible
An extensive hydrogen exchange within this equilibrium
apparently gave rise to the broad signal at 1.6 ppm observed at
25 °C (Fig. 3).
The composition of this ocher solution did not change even
after aging at room temperature for 6 months. Degassing of
this solution restored the original brownish-green color, and
the
1H
NMR
spectrum
showed
the
dominant
[(C5Me4SiMe3)2Ti] and an admixture of [(C5Me4SiMe3)2TiH]
only. The latter compound thus arose from hydrogenation of
1
with the hydride–deuteride structure since the JHD coupling
for coordinated molecular HD ranges between 15–35 Hz.1,2
This JHD coupling is, however, larger than that for
[Cp*2TiH(D)] (3.4 Hz, see above) reflecting thus a weaker calcu-
lated Ti–H bonding for 4 with respect to 1 (see below). The
signal for 4, of correct intensity, was obtained after aging the
sample for 6 months (see Fig. S18 in the ESI†), which indicates
that exchanges between hydride/deuteride and methyl hydro-
gen atoms are markedly slower than e.g., for [Cp*2TiH].11 This
can be caused by the virtual absence of the tucked-in titano-
cene hydride tautomer for [(C5Me4SiMe3)2Ti] which likely
mediates such exchanges in the [Cp*2Ti] systems.
The chemical shifts for titanocene dihydrides so far gath-
ered fit into the order of decreasing δH (TiH2) with increasing
sterical bulk of substituents on cyclopentadienyl ligands: Me
(Cp*) 3.42 (−45 °C) > Pr 3.25 (−75 °C)17 > Bu 3.04 (−35 °C) >
SiMe3 2.74 ppm (−35 °C). The value of δH (TiH2) 2.62 ppm for
[(1,3-tBu2C5H3)2TiH2]16 can fit this series with regard to the
bulkiness of the cyclopentadienyl substituents, however, little
is known on the influence of the electron donation ability of
auxiliary ligands which should be lower for the less alkylated
compound. An analogous effect of the bulkiness of cyclopenta-
i
t
Fig. 2 1H NMR spectra observed after addition of hydrogen to
a
mixture of [(C5Me4SiMe3)2Ti] and [(C5Me4SiMe3)Ti(C5Me4SiMe2CH2)] in
toluene-d8 measured at temperatures from −35 to 25 °C. Signals of
[(C5Me4SiMe3)2Ti] and [(C5Me4SiMe3)2TiH] (6) as depending on tempera-
ture are highlighted by red and blue bars, respectively; signals of dia-
magnetic 4 fall into 0–3 ppm range.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 3 EPR and UV-vis spectra of titanocene hydrides 2, 5, and 6 and products of their reaction with butadiene 7 and 8
EPR
g1
g2
g3
Compound no.
giso
ΔH (G)
g┴
g
gav
UV-vis (nm)
Ref.
k
a
a
2
5
6
7
8
—
—
—
—
—
—
1.975
1.980
1.981
1.997
2.000
1.769
1.776
1.806
1.895
1.900
1.906
1.912
1.922
1.959
1.962
550b
580b
580b
605b
620b
476
475
475
500
485
—
11
a
a
375b
375b
380b
—
This work
This work
This work
This work
a
a
1.956
1.962
38
24
1.985
1.985
a Only EPR spectra of impurities are observable as single lines for Cp*2TiCl (giso ∼ 1.956, ΔH ∼ 30 G) and/or oxo-impurity Cp*2TiOR (giso ∼ 1.977,
ΔH = 3–5 G, a(Ti) = 8–10 G). These are observable in frozen toluene glass at common g1 ∼ 1.999, g2 ∼ 1.982 and specific g3(Cp*2TiCl) ∼ 1.888 (ref.
30 and 31) and g3(Cp*2TiOR) ∼ 1.952 (see the ESI).32 The electronic triplet state impurity [(Cp*2Ti)2O]33 has been observed in some samples
showing the outmost feature ΔHzz = 473 G, gzz = 1.985 and ΔMS = 2 at g ∼ 3.970 (see the ESI). b Shoulder.
dienyl auxiliary ligands inducing a decrease in equilibrium ation of hydrogen molecule(s). The [M − H2]+ ion was the base
constants for dinitrogen coordination to titanocenes has been peak for 6 whereas [M − nH2]+ ions (n = 1–3) were highly abun-
1
reported by Chirik et al.26
dant for 5, the base peak for n = 3. H NMR spectra of 5 and 6
DFT computations for the presently prepared titanocene revealed a set of three broad signals in an integral ratio
dihydrides and titanocene dihydrogen complexes indicated 18 : 12 : 12. The assignment of high field signals at 3.5 ppm to
that the dihydride [Cp*2TiH2] is by 34.31 kJ mol−1 more stable CMe3 of 5 and 2.00 ppm to the SiMe3 group of 6 was compati-
than the analogous dihydrogen complex [Cp*2Ti(η2-H2)] ble with their intensities. The signals of C5Me4 moieties of 5
whereas for the [(C5Me4SiMe3)2Ti] analogues only one half of occurred at 18.3 ppm and 22.7 ppm and those of 6 at
the energy difference (17.11 kJ mol−1) was found. The counter- 13.7 ppm and 27.91 ppm; the signals of hydrides (TiH) were
poise energies computed between the titanocene fragment and not found similarly to 2 (see Fig. S19 and S20 in the ESI†). The
the dihydride atoms correspondingly showed a decrease in the infrared ν(Ti–H) vibration occurred as a broad medium-inten-
t
order of auxiliary ligands Cp* > C5Me4 Bu > C5Me4SiMe3 (see sity absorption band at 1521 cm−1 for 5 and at 1525 cm−1 for
Table 1 in the ESI†); the difference between edge cases of 20 6. These bands diminished after exposing the KBr pellets to
kJ mol−1 is in accord with the reversible dissociation of hydro- air, whereas the absorption bands of oxygenated products at
gen from an equilibrium mixture of 4 and [(C5Me4SiMe3)2Ti] 571 cm−1 and 577 cm−1, respectively grew stronger in intensity.
upon degassing (eqn (7)).
The electronic absorption spectra of 5 and 6 were nearly identi-
Owing to the low sterical demand of the hydrogen mole- cal and exhibited similarity to the visible spectrum of 2 (476
cule, no sterical hindrance was encountered in preparing tita- and 547 nm).11 EPR spectra of 2, 5, and 6 in toluene glass dis-
nocene monohydrides by hydrogenation of intramolecular played extremely anisotropic spectra differing distinctly in
t
σ-titanium-methylene
compounds
[(η5-C5Me4 Bu)(η5:η1- their high magnetic field g-tensor component. The shape of
C5Me4CMe2CH2)Ti] (ref. 27) and [(η5-C5Me4SiMe3)(η5:η1- anisotropic spectra could be interpreted as arising from the
C5Me4SiMe2CH2)Ti] (ref. 28) yielding [(η5-C5Me4 Bu)2TiH] (5) g-tensor of cylindrical symmetry, however, its g┴ value could
t
and [(η5-C5Me4SiMe3)2TiH] (6), respectively (Scheme 5). This not be exactly identified because this feature coincided with g1
method was previously used to obtain lanthanide(III) hydride and g2 tensor values of Cp*2TiL impurities (L mostly Cl or OR).
complexes [(η5-C5Me4SiMe3)2LnH(THF)] (Ln = Y, Nd, Sm, Dy, The EPR and UV-vis data for 2, 5, and 6 are gathered in Table 3
Lu,29a and Ce29b) and [(Cp*2CeH)2].29c
(for relevant EPR records see the ESI†); the crystal structure of
The EI-MS spectra of the brown crystalline hydrides 5 and 6 5 is shown in Fig. 4.
showed molecular ions of low intensity due to the easy elimin-
All these data and crystallographic Ti–Cg and Ti–H para-
meters are in agreement with those for 211 and the other well-
characterized titanocene hydride [(C5Me4Ph)2TiH] (including
the description of EPR spectra for (C5Me4Ph)2TiOR impurity in
solution).34 The low precision of determination of the Ti–H
bond length from X-ray single crystal diffraction is exemplified
on these compounds and 1, ranging from 1.59(4) Å for mole-
cule 2 of 1A to 1.88(4) Å for molecule 1 of 1B (Table 1).
Introducing excess butadiene to both hydrides 5 and 6
induced a nearly immediate color change from ocher to
brown. This indicated at first sight that common π-bonded
methylallyl derivatives which are typically purple colored for
the sterically less congested titanocenes, including [Cp*2Ti(η3-
Scheme 5 Synthesis of 5, 6, 7, and 8.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
ESI†). This complexity together with the presence of paramag-
netic impurities established by EPR measurements precluded
the assignment of spectra. The infrared spectra of both 7 and
8 displayed a weak absorption band at 1629 cm−1 apparently
due to ν(CvC) valence vibration. The σ-bonding of 2-butenyl
group was established by X-ray single crystal analysis of 8
(Fig. 4). Comparing the basic parameters Cg–Ti–Cg angle
143.19(4)° and Ti–Cg distances 2.0985(10) Å and 2.0854(10) Å
with those for [Cp*2Ti(η3-(1-methylallyl))] (Cg–Ti–Cg 138.15
(10)–138.46(11) ° and Ti–Cg 2.120–2.134(3) Å for three inde-
pendent molecules) shows that η1- and η3-bonding modes of
butadiene addition products are controlled by the space in the
titanocene shell available for coordination.
Butadiene in excess was also added to toluene-d8 solutions
t
of titanocenes [(C5Me4 Bu)2Ti] and [(C5Me4SiMe3)2Ti] at room
temperature for one day; however, no reaction products were
1
detected by H NMR spectra after evaporation of butadiene to
dryness. This is in sharp contrast with the whole class of zirco-
nocene(II) complexes containing the η4-1,3-butadiene ligand
(largely
s-trans)37a–d
including
the
persubstituted
[(C5Me4Ph)2Zr(s-trans-η4-buta-1,3-diene)].37e Their formation is
facilitated by a larger covalent radius of Zr with respect to Ti by
Fig. 4 PLATON drawing of 5 at the 30% probability level, with atom
labeling scheme. Hydrogen atoms except the hydride are omitted for
clarity. Selected bond distances (Å) and bond angles (°) for molecule 1:
Ti–Cg(1) 2.0296(8), Ti–Cg(1) 2.0360(7), Ti–H(1) 1.865(17), Cg(1)–Ti–
Cg(2) 152.29(3), Cg(1)–Ti–H(1) 104.9(5), Cg(2)–Ti–H(1) 102.8(5).
(1-methylallyl))] (ref. 14) were not formed. Instead, the σ-
t
bonded
CH2CHvCHMe)]
η1-but-2-enyl
derivatives
[(C5Me4 Bu)2Ti(η1-
(7)
and
[(C5Me4SiMe3)2Ti(η1-
CH2CHvCHMe)] (8) were obtained in high yields and were
characterized by their broad EPR signals at giso = 1.958, ΔH =
35 G and giso = 1.962, ΔH = 22 G, respectively, which resembled
those of the corresponding titanocene alkyl30 or alkenyl com-
pounds.35 They also displayed electronic absorption spectra
typical of (Cp′2TiL) compounds (transitions 1a1 → 2a1 and
1a1 → b1).30 In contrast, (η3-1-methylallyl)titanocenes exhibit
narrower EPR signals (ΔH ∼ 6 G) at giso ∼ 1.990 and a single
absorption band in the visible region (518–530 nm).14,36 EI-MS
spectra of 7 did not show the molecular ion (m/z 457), which
eliminated C4H7 to give a [Cp′2Ti]+ fragment, subsequently
losing hydrogen and butyl groups similarly to 5. At variance,
EI-MS spectra of 8 displayed a weak molecular ion (m/z 489)
Fig. 5 PLATON drawing of 8 at the 30% probability level, with atom
losing C4H8 to give the [Cp′2Ti − H]+ base peak. 1H NMR labeling scheme. Hydrogen atoms except those at C25A–C28A are
omitted for clarity. Selected bond distances (Å) and angles (°) for mole-
cule 1: Ti–Cg(1) 2.0985(10), Ti–Cg(2) 2.0854(10), Ti–C(25A) 2.225(4),
C(25A)–C(26A) 1.504(6), C(26A)–C(27A) 1.326(4), C(27A)–C(28A)
1.489(4), Cg(1)–Ti–Cg(2) 143.19(4), Cg(1)–Ti–C(25A) 106.92(8), Cg(2)–
spectra of 7 and 8 span the range −15 to 15 ppm where three
broad signals for methyl groups, two of the tetramethyl-
cyclopentadienyl ring and one of the t-Bu or SiMe3 groups and
signals of 2-butenyl groups of much lower intensity and highly
Ti–C(25A) 109.61(9), Ti–C(25A)–C(26A) 107.5(2), C(25A)–C(26A)–C(27A)
varying in linewidth should occur (see Fig. S21 and S22 in the 129.0(3), (C26A)–C(27A)–C(28A) 125.8(3).
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
∼0.15 Å as well as by a lower Pauling’s electronegativity (1.3 for spectra were obtained on a VG-7070E mass spectrometer at
Zr versus 1.5 for Ti) (Fig. 5).4
70 eV. Crystalline samples in sealed capillaries were opened and
inserted into the direct inlet under argon. The spectra are rep-
resented by the peaks of relative abundance higher than 10%
and by important peaks of lower intensity. Crystalline samples
for EI-MS measurements and melting point determinations
Conclusions
1H NMR investigations of hydrogenation of decamethyl- were placed in glass capillaries in a glovebox Labmaster 130
titanocene [Cp*2Ti], [Cp*(C5Me4CH2)TiMe], and [Cp*2Ti(η2- (mBraun) under purified argon (oxygen and water vapor con-
ethene)] revealed that the titanocene dihydride 1 is the centrations were below 2.0 ppm) and sealed with flame. IR
primary product which in solution loses hydrogen to give the spectra of samples in KBr pellets prepared in the glovebox
stable monohydride 2. The role of titanocene as an intermedi- were measured in an air-protecting cuvette on a Nicolet Avatar
ate during the latter process is inevitable upon assuming both FT IR spectrometer in the range 400–4000 cm−1. EPR spectra
possible pathways (a) comproportionation of titanocene with were recorded on a MiniScope MS400 (Magnettech GmbH,
the dihydride as previously suggested by Bercaw10c or (b) the Berlin, Germany) equipped with
a microwave frequency
titanocene disproportionation according to eqn (4) in the pres- counter FC 400 and a temperature controller TC H03. UV–near
ence of dihydrogen. The conversion of 1 to 2 is accomplished IR spectra in the range 300–2000 nm were measured on a
after removal of all dihydrogen; however, admission of di- Varian Cary 17D spectrometer in all-sealed quartz cells
hydrogen induces a reversible process converting Ti(III) 2 to Ti(IV) 1, (Hellma) and electronically saved. GC-MS analysis was carried
apparently via a retro-pathway. Unlike in solutions, 1 is out on a Thermo Focus DSQ using a capillary column Thermo
stable in crystalline form and its X-ray single crystal structure TR-5MS (15 m × 0.25 mm × ID 0.25 μm). Elemental analyses
was determined successfully. Although the precision of were carried out on a FLASH EA1112 CHN/O Automatic
hydride atom position determination is low, with the H–H dis- Elemental Analyzer (Thermo Scientific). Melting points were
tances appearing in the range 1.58(9)–2.31(4) Å, these were measured on a Koffler block in sealed glass capillaries under
nevertheless incompatible with the H–H distance of 0.911 Å argon and were not corrected.
obtained for the DFT-optimized structure of Cp*2Ti(II) having
Chemicals
its dihydrogen molecule η2-coordinated. A larger steric conges-
tion due to η5-C5Me4 Bu and η5-C5Me4SiMe3 auxiliary ligands The solvents tetrahydrofuran (THF), hexane, and toluene were
t
had no effect on the instant reaction of titanocenes dried by refluxing over LiAlH4 and stored as solutions of green
t
[(C5Me4 Bu)2Ti] and [(C5Me4SiMe3)2Ti] with hydrogen yielding dimeric titanocene [(μ-η5:η5-C5H4C5H4)(μ-H)2{Ti(η5-C5H5)}2].38
dihydrides 3 and 4 which were well characterized by 1H and Toluene-d8 C7D8 (99.5% D) (Sigma Aldrich) was degassed, dis-
13C NMR spectra at low temperature. The latter compound tilled under vacuum on the singly tucked-in permethyl-
formed an equilibrium with [(C5Me4SiMe3)2Ti] depending on titanocene [Ti(C5Me5)(C5Me4CH2)],10b and stored as its solu-
hydrogen pressure while not showing any decay, e.g., to tion on a vacuum line. The ethene complexes [Cp*2Ti(II)(η2-
t
[(C5Me4SiMe3)2TiH]. Monohydrides 5 and 6 were obtained C2H4)],
[(η5-C5Me4 Bu)2Ti(II)(η2-C2H4)],
and
[(η5-
from hydrogenation of intramolecular σ-titanium-methylene C5Me4SiMe3)2Ti(II)(η2-C2H4)] were obtained and thermolyzed
t
t
compounds [(η5-C5Me4 Bu)(η5:η1-C5Me4CMe2CH2)Ti] and [(η5- under vacuum to titanocenes [Cp*2Ti], [(η5-C5Me4 Bu)2Ti] and
C5Me4SiMe3)(η5:η1-C5Me4SiMe2CH2)Ti] and they were used to [(η5-C5Me4SiMe3)2Ti] as reported.12 Complexes of decamethyl-
react with butadiene. The steric effect of auxiliary ligands was titanocene with internal alkynes but-2-yne, hex-3-yne, 1,2-
manifested by the formation of titanocene (η1-but-2-en-1-yl) diphenylethyne, 1-phenyl-2-(trimethylsilyl)ethyne, 1-methyl-2-
compounds 7 and 8, whereas decamethyltitanocene binds η3- (trimethylsilyl)ethyne, and 1,2-bis(trimethylsilyl)ethyne were
1-methylallyl, which is a ligand common for sterically less identical to those obtained from [Cp*2Ti(II)(η2-C2H4)] and the
demanding titanocenes.
alkynes via the ligand exchange.21a Compounds [Cp*
(C5Me4CH2)TiMe],19
[{η3:η4-C5Me3(CH2)2}Ti],13,39
[(η5-
t
C5Me4 Bu)(C5Me4CMe2CH2)Ti],27
[(η5-C5Me4SiMe3)
(C5Me4SiMe2CH2)Ti],25b,28 and [{η3:η4-C5Me3(CH2)2}{η5-C5Me3
(CH2CHtBuCHvCHCHtBuCH2)}Ti]22 were obtained according
to published procedures. A mixture of [Cp*2Ti] and [Cp*2Ti(II)
Experimental section
General considerations
Synthesis, treatment, and subsequent reactions of low-valent (η2-C2H4)] was obtained unintentionally by thermolyzing of the
titanium compounds were carried out on a vacuum line in latter at 123 °C using a high pumping speed allowing for
sealed all-glass devices equipped with magnetically breakable partial sublimation of non-thermolyzed [Cp*2Ti(II)(η2-C2H4)].
seals. H (300 MHz), 13C (75 MHz) and 29Si (59.6 MHz) NMR The impurity of Cp*2TiCl was present in twice crystallized
1
spectra were recorded on a Varian Mercury 300 spectrometer [Cp*2Ti(II)(η2-C2H4)].12 It was established by EPR (Table 3) and
in toluene-d8 solutions at 25 °C. Chemical shifts (δ/ppm) are 1H NMR spectra (14.4 ppm) and used as a spectrum intensity
given relative to the residual solvent signal (CD2H: δH standard since it did not react with hydrogen. Cp*2TiOR impu-
2.08 ppm) and the solvent resonance (Cipso: δC 137.48 ppm). rities were detected by EPR spectra in very low concentrations
The δSi values are related to external tetramethylsilane. EI-MS (<1% Ti) (Table 3), not reacting with hydrogen. Commercial
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
hydrogen (anal.3.0) (Messer) and deuterium (Sigma Algrich) NMR tube. A broad-range 1H NMR spectrum revealed 2
were used from steel cylinders without purification. HD was (23.0 ppm, Δν1/2 ∼ 170 Hz) contaminated with a trace of
prepared by adding D2O to LiAlH4 (see the ESI†).
[(Cp*2Ti)2O] (δ 4.3 ppm, Δν1/2 ∼ 270 Hz).33 The eliminated
methane was also observed at 0.17 ppm.18 Volatiles were evap-
orated to dryness and the residue was extracted with a
Reactions of titanocene derivatives with dihydrogen
A three-necked ampule (96 mL) equipped with two magneti- minimum of hexane and crystallized in a refrigerator. A yield
cally breakable glass seals and a Teflon valve was evacuated to of 0.12 g (95%) reddish-ocher crystalline solid was obtained.
2 × 10−4 Torr and filled with hydrogen (anal. 3.0) from a steel
Analytical data for compound 2. EI-MS, 1H NMR, EPR, IR,
cylinder after flushing the space around the Teflon valve and UV-vis data generally agree with ref. 11 and 12.
piston at slightly higher than atmospheric pressure (content of
Hydrogenation of [Cp*2Ti(η2-C2H4)] to 2
H2 ≥ 4.2 mmol). This ampule was attached via a breakable
seal to an all-glass T-shape device bearing a three-arm empty
(A) Preparative: toluene (4 mL) was added to crystalline [Cp*2Ti
ampule (90 mL) and an ampule containing a degassed solu-
(η2-C2H4)] (0.17 g, 0.49 mmol) and hydrogen (∼4.2 mmol)
tion of the titanocene reactant in toluene-d8, typically
from the ampule (96 mL) was admitted via an evacuated
0.3–0.5 mmol/1.0–1.5 mL. After evacuating on a vacuum line
device. The green color of [Cp*2Ti(η2-C2H4)] began to change
the T-shape device was sealed off, the solution was transferred
to yellow immediately, turning reddish-ocher after 2 h. After
to the empty ampule, hydrogen was admitted, and the hydro-
stirring for another 3 h the reaction mixture was slowly evapor-
genation ampule was kept at 25–80 °C in a water bath for a
ated to dryness, the residue was dissolved in hexane (3 mL)
required time. The hydrogenation ampule could also be torch-
and crystallized by slow evaporation of the solvent to an
sealed off from the T-shape device, containing then at least
ampule arm placed at the cooling wall of a refrigerator.
1.9 mmol of hydrogen available for hydrogenation. No
Reddish-ocher crystalline solid was washed out with conden-
attempts were made to follow the kinetics of hydrogen con-
sing hexane vapor and dried under vacuum. Yield 0.14 g
sumption. The hydrogenation was terminated by pumping off
(89%).
free hydrogen from a liquid nitrogen-cooled solution. The
Broad-range 1H NMR spectra revealed that the dominant
solution was transferred to an NMR tube under vacuum and
broad resonance at δ 23.0 ppm of 2 was accompanied inher-
the tube was sealed off with a torch. After the NMR analysis
ently by the titanocene [Cp*2Ti] resonance at δ 63.8 ppm of
was carried out the NMR tube was magnetically broken at the
very low intensity (≪1%) and a yet weaker resonance at
sealed tip under vacuum and the content was collected in a
−11.7 ppm typical of [Cp*(C5Me4CH2)Ti].10b,15 In addition a
three-necked ampule for vacuum distillation of volatiles for
weak signal of [(Cp*2Ti)2O] at 4.3 ppm (Δν1/2 = 270 Hz)33 was
1
GC-MS and/or H and 13C NMR analysis and the isolation of
also observable. IR (KBr), EPR and UV-vis spectra are in
the titanocene product. Hydrogenations as described above
general accord with published data.11,12
were also performed in torch-sealed NMR tubes; a maximum
(B) NMR tube experiment: green crystals of [Cp*2Ti(C2H4)]
of 0.2 mmol of the reagent in 1.0 mL of toluene-d8 was taken
(0.07 g, 0.2 mmol) were dissolved in toluene-d8 (0.8 mL),
and sealed off, and cooled with liquid nitrogen.
hydrogen was admitted as above, and a slightly cooled NMR
tube was sealed off with a flame. 1H NMR after 10 min at room
Reaction of [Cp*2Ti] with dihydrogen to give 1
temperature: [Cp*2Ti] δ 63.8 ppm (10%), 2 δ 23.0 ppm (2%),
The titanocene obtained by thermolysis of [Cp*2Ti(η2-C2H4)] at
1 δ 3.37 and 1.88 (Δν1/2 60 Hz) (88%), C2H6 δH 0.81 ppm (ref. 16)
125 °C under high vacuum12 was extracted with hexane vapor
(95%); after 120 min: [Cp*2Ti] (0%), 2 (23%), 1 δ 3.37 ppm very
and the solution was evaporated under vacuum. Crystalline
broad, 1.90 (Δν1/2 12 Hz) (77%), C2H6 (100%). After 3 months:
solid (0.07 g, 0.22 mmol) was dissolved in toluene-d8 (1.0 mL)
2 (60%), 1 δ 1.91 ppm (40%). The sample after twice repeated
and poured into an NMR tube; hydrogen was admitted, and
degassing 2 (100%), 1 (0%).
the tube cooled with liquid nitrogen was sealed off with flame.
After warming to room temperature for 10 min the 1H NMR
Hydrogenation of [Cp*2Ti(η2-MeCuCMe)] and Cp*2Ti
(EtCuCEt)
spectrum shown in Fig. S1 of the ESI† was recorded.
Analytical data for compound 1. 1H NMR (toluene-d8, 25 °C):
1.90 (br s, Δν1/2 = 60 Hz, 30H, C5Me5); 3.37 (br s, Δν1/2 = 65 Hz, Admission of gaseous hydrogen (∼4.2 mmol) to a dirty green
2H, TiH). 1H NMR (toluene-d8, −45 °C): 1.87 (br s, Δν1/2
=
solution of the reactant complex14 (0.3 mmol) in toluene-d8
16 Hz, 30H, C5Me5); 3.42 (br s, Δν1/2 = 45 Hz, 2H, TiH). 13C{1H} (1.0 mL) resulted in a nearly immediate change of its color to
NMR (toluene-d8, −45 °C): 13.34 (C5Me5); 119.79 (C5Me5).
reddish-ocher. After stirring for 6 h the reaction solution was
degassed, the volatiles were distilled into an NMR tube for ana-
lysis, and the residue was dissolved in toluene-d8, and trans-
ferred to an NMR tube. Paramagnetic 1H NMR spectra revealed
Preparation of 2 by hydrogenation of [Cp*(η5:η1-C5Me4CH2)
TiMe]
A turquoise solution of the crystalline reagent19 (0.14 g, the presence of
2 (δH 23.0 ppm) with [Cp*2TiCl] (δH
0.4 mmol) in toluene-d8 (1.2 mL) was reacted with hydrogen 14.4 ppm)30 and [(Cp*2Ti)2O] (δH 4.3 ppm)33 impurities. 1H
(∼4.2 mmol) at room temperature for 8 h. The resulting and 13C NMR spectra and GC-MS analysis of the volatiles
reddish-ocher solution was degassed and transferred to an revealed only butane and hexane (see the ESI†).
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Hydrogenation of [Cp*2Ti(η2-PhCuCPh)]
toluene-d8 and 1H NMR spectra revealed that the solution
contained only 2.
The reagent preparation, its stoichiometry to hydrogen, and
other conditions were the same as in the preceding experi-
ments. The original green color changed to greenish-yellow
overnight. The solution was transferred to an NMR tube and
assessed to contain only 5% of [Cp*2TiH]. It was then trans-
ferred to an ampule and hydrogenated with a new portion of
hydrogen (4.2 mmol) at 70 °C for 32 h. The reddish-ocher
solution was degassed through a trap cooled with liquid
nitrogen and distilled under high vacuum at 70 °C. Volatiles
collected in the trap were analyzed by 1H and 13C NMR spectra
to contain 1,2-diphenylethane and a trace of cis-stilbene in
toluene-d8. The toluene-d8 solution of the residue showed only
the signal of 2 at 23.0 ppm.
Hydrogenation of C1
Crystalline C122 (0.12 g, 0.25 mmol) in toluene-d8 (1.2 mL) in
the presence of hydrogen (∼4.2 mmol) did not change its blue
color at room temperature overnight. After heating to 70 °C for
26 h its initial blue color turned clear yellow. The volatiles
were vacuum-distilled to an NMR tube and the residue was
extracted with hexane and crystallized in a freezer by slow dis-
tillation off the solvent. The yield of yellow finely crystalline
solid was 0.11 g (91%). 1H and 13C NMR spectra of the product
agreed with the data for compound C223a previously obtained
by hydrogenation of the acetylenic precursor C323b (Scheme 4).
The presence of about 1% amount of 2,7-tetramethyloct-4-ene
isomers in the volatiles indicated that the hydrogenation
induced also the retrosynthesis of the bicyclic ligand; therefore
the further hydrogenation of C2 at higher temperature was not
investigated.
Hydrogenation of [Cp*2Ti(η2-PhCuCSiMe3)]
A brown crystalline complex (0.15 g, 0.3 mmol) was dissolved
in toluene-d8 (1.0 mL), and hydrogen (∼4.2 mmol) was
admitted. After stirring overnight at room temperature, the
solution color turned more yellow. The solution was further
kept at 70 °C for 32 h to give a reddish-ocher solution.
Volatiles were distilled under vacuum to a cooled trap and ana-
lyzed to contain only 1-phenyl-2-(trimethylsilyl)ethane (for
t
Hydrogenation of [(C5Me4 Bu)2Ti]
Crystalline titanocene12,24 (0.11 g, 0.3 mmol) in toluene-d8
(1.2 mL) was reacted with hydrogen overnight and the NMR
tube was sealed off with a torch under hydrogen. An ocher
solution revealed the 1H NMR spectrum of 5 (see below) and a
1
GC-MS and H and 13C NMR data see the ESI†). The residue
1
was 2 (δH 23.0 ppm).
very broad signal at about 2 ppm attributed to 3. At −35 °C H
and 13C NMR spectra of 3 were resolved and the 1H NMR spec-
trum of 5 and the outmost resonance at low magnetic field of
Hydrogenation of [Cp*2Ti(η2-MeCuCSiMe3)]
t
[(C5Me4 Bu)2Ti] were observed.
A crystalline complex (0.13 g, 0.3 mmol) was dissolved in
toluene-d8 (1.0 mL), and hydrogen (∼4.2 mmol) was admitted.
After stirring overnight at room temperature, the solution
color turned more yellow. Heating to 70 °C for 48 h was
required to obtain the reddish-ocher solution of 2 containing
propyltrimethylsilane (for 1H NMR and GC-MS data see the
ESI†).
Analytical data for compound 3. 1H NMR (toluene-d8,
−35 °C): 1.45 (s, 12H, C5Me4); 1.48 (s, 18H, CMe3); 2.10 partly
overlapped by a solvent signal (s, 12H, C5Me4); 3.04 (s, 2H, Ti–
H2). 13C NMR (toluene-d8, −35 °C): 11.56, 16.98 (C5Me4); 33.16
(CMe3); 36.28 (CMe3); 117.78, 119.01 (C5Me4, Cq); 134.42
(C5Me4, Cipso). The resonance at δH 2.10 ppm was recognized
using a gHMQC experiment.
Hydrogenation of [Cp*2Ti(η2-Me3SiCuCSiMe3)]
Hydrogenation of [(C5Me4SiMe3)2Ti]
A
solution of [Cp*2Ti(η2-Me3SiCuCSiMe3)] in toluene-d8
Crystalline titanocene (0.13 g, 0.3 mmol) in toluene-d8
(0.5 mmol/1.5 mL) after addition of hydrogen remained yellow
after 3 days, indicating no visible reaction. Then the solution
was heated to 70 °C for 34 h yielding, according to 1H NMR
spectra, 2 (80%) and initial reagent (20%) and olefins. This
experiment was repeated with heating to 80 °C for 112 h. The
content of the initial reagent was 13% and that of 2 was 87%,
and the composition of hydrocarbon products was (Z)-
Me3SiCHvCHSiMe3 5%, (E)-Me3SiCHvCHSiMe3 50%, and
Me3SiCH2CH2SiMe3 45% (see the ESI†).
(1.2 mL) was reacted with hydrogen overnight and the NMR
tube was sealed off with a torch under hydrogen. The H NMR
1
spectrum of the ocher solution revealed a very broad signal at
1.6 ppm due to mainly 4, weak resonances of 6 (see below) and
a very broad and extremely weak outmost resonance at low
magnetic field of [Ti(C5Me4SiMe3)2] (see Fig. S14 in the ESI†).
At −35 °C 1H and 13C NMR spectra of 4 were resolved (see
Fig. S15 and S16 in the ESI†), signals of 6 showed a para-
magnetic shift and the outmost resonance at low magnetic
field of [(C5Me4SiMe3)2Ti] turned narrow and paramagnetically
shifted to 99.6 ppm (Fig. 2 and 3). 1H NMR resonance of di-
Hydrogenation of [Cp*Ti(C5Me3(CH2)2)]
The doubly tucked-in titanocene crystalline reagent (0.13 g, hydrogen at 4.50 ppm was not observed. After degassing by
1
0.4 mmol) was dissolved in toluene-d8 (1.2 mL), and hydrogen two freezing/thawing cycles the H NMR spectra of the major
(∼4.2 mmol) was admitted. The blue color of the reagent [(C5Me4SiMe3)2Ti] and minor 6 were only observed in a brown-
remained unchanged at room temperature over 24 h, however, ish green solution (see Fig. S14 in the ESI†).
after heating to 70 °C for 26 h its color turned ocher. After
evaporation of volatiles to dryness the residue was dissolved in −35 °C): 0.36 (s, 18H, SiMe3); 1.40, 1.96 (2 × s, 2 × 12H, C5Me4);
Analytical data for compound 4. 1H NMR (toluene-d8,
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
2.74 (s, 2H, TiH2). 13C NMR (toluene-d8, −35 °C): 2.76 (SiMe3); 12H, C5Me4); 27.9 (br s, Δν1/2 ∼ 290 Hz, 12H, C5Me4). EPR
11.23, 15.88 (C5Me4); 116.60 (C5Me4, Cipso); 121.72, 125.93 (toluene-d8, −160 °C): g┴ = 1.980, g = 1.808, gav = 1.922. UV-vis
k
(C5Me4, Cq). Interaction of TiH2 and SiMe3 was proved by 1D (toluene-d8, nm): 318 > 375(sh) ≫ 470 > 580(sh). Found (%):
NOESY.
C, 66.12; H, 9.92. C24H43Si2Ti requires (%): C, 66.17; H, 9.95.
t
Hydrogenation of [(C5Me4 Bu)Ti (C5Me4CMe2CH2)] to give 5
Reaction of 5 with butadiene to give 7
A crystalline reagent27 (0.4 g, 1.0 mmol) was dissolved in Crystalline hydride (0.23 g, 0.6 mmol) was dissolved in hexane
toluene-d8 (1.0 mL) and hydrogen was admitted to the solu- (3.0 mL) and butadiene (∼0.2 mL, >2 mmol) was added. After
tion. The initial dark color turned immediately to ocher. After 30 min at room temperature all volatiles were evaporated
stirring for 30 min the unreacted hydrogen was pumped off under vacuum and a brown residue was dissolved in hexane
and a reddish-ocher solution was investigated by 1H NMR, and crystallized by slow solvent distillation in a refrigerator. A
EPR and UV-vis spectra. Then, toluene-d8 was replaced with brown finely crystalline powder was obtained after separation
hexane and the product was crystallized in a refrigerator. Dark of ca. 0.2 mL of mother liquor. The product was recrystallized
red-ocher crystals were recrystallized from hexane and dried from hexane at −28 °C in a freezer. Yield 0.23 g, 85%.
under vacuum. Yield 0.33 g, 82%.
Analytical data for compound 7. Mp. 103 °C, dec. EI-MS
Analytical data for compound 5. Mp. 111 °C dec. EI-MS (direct inlet, 70 eV, 100 °C): m/z (relative abundance) 457 (M•+
(direct inlet, 70 eV, 100 °C): m/z (relative abundance) 404 (10), not observed), 404 (22), 403 (57), 402 ([M − C4H7]+; 100), 401
403 (M•+; 36), 402 (68), 401 ([M − H2]+; 89), 400 (60), 399 ([M − (65), 400 ([M − C4H7 − H2]+; 63), 399 (67), 398 (25), 397 (57),
2H2]+; 95), 398 (31), 397 ([M − 3H2]+; 100), 396 (26), 395 ([M − 396 (15), 345 (16), 344 ([M − C4H7 − C4H10]+; 41), 343 (14),
4H2]+; 43), 394 (19), 393 (41), 381 (26), 346 ([M − C4H9]+; 17), 341 (13), 339 (25), 327 (37), 325 (22), 226 (25), 223 (14),
345 (17), 344 ([M − H2 − C4H9]+; 41), 339 ([M − 2H2 − C4H9]+; 206 (26), 202 (27), 201 (37), 187 (18), 119 (10), 105 (13), 91 (14),
20), 327 (37), 325 (26), 226 [M − Cp′]+; (25), 223 (19), 206 (26), 57 (68), 55 (48), 43 (40), 41 (73), 39 (29). IR (KBr, cm−1):
202 (27), 201 (42), 119 (10), 105 (13), 91 (14), 57 (38). IR (KBr, 2991 (s), 2953 (vs), 2906 (vs), 2867 (s, sh), 2722 (vw), 1629 (w),
cm−1): 2987 (s, sh), 2954 (vs), 2907 (vs), 2867 (s), 2722 (vw), 1455 (s, b), 1380 (s), 1358 (vs), 1234 (s), 1198 (m), 1150 (w),
1521 (m, b), 1480 (m), 1457 (m, b), 1381 (s), 1373 (s), 1358 (s), 1122 (m), 1063 (w), 1035 (m), 1022 (s), 975 (w), 958 (m),
1234 (m), 1200 (w), 1123 (vw), 1038 (w), 1023 (m), 829 (vw), 746 866 (vw), 836 (vw), 818 (vw), 761 (vw), 711 (vw), 664 (w),
(vw), 712 (vw), 670 (vw), 620 (w), 571 (w), 505 (w), 434 (m). 642 (w), 618 (m), 572 (w), 492 (w), 473 (w), 453 (vw), 423 (s).
1H NMR (toluene-d8): 3.5 (br s, Δν1/2 ∼ 80 Hz, 18H, CMe3); 18.3 EPR (toluene-d8, 22 °C): giso = 1.956, ΔH = 38 G; (toluene-d8,
(br s, Δν1/2 ∼ 230 Hz, 12H, C5Me4); 22.7 (br s, Δν1/2 ∼ 390 Hz, −140 °C): g1 = 1.997, g2 = 1.985, g3 = 1.895, gav = 1.959. UV-vis
1
12H, C5Me4). EPR (toluene-d8, −140 °C): g┴ ∼ 1.980, g = 1.776, (toluene-d8, nm): 380(sh) ≫ 500 > 605(sh). The H NMR spec-
k
gav = 1.912. UV-vis (toluene-d8, nm): 310 > 375(sh) ≫ 475 > 580 trum attributed to 7 is shown in Fig. S21 in the ESI.† Found:
(sh). Found (%): C, 77.35; H, 10.72. C26H43Ti requires (%): C, 77.77; H, 10.81. C30H49Ti requires (%): C, 78.74; H, 10.79.
C, 77.39; H, 10.74.
Reaction of 6 with butadiene to give 8
Hydrogenation of [(C5Me4SiMe3)Ti (C5Me4SiMe2CH2)] to give 6
A crystalline reagent (0.20 g, 0.46 mmol) was dissolved in
A crystalline reagent24,28 (0.22 g, 0.5 mmol) was hydrogenated hexane (3.0 mL) and butadiene (∼0.2 mL, >2 mmol) was
in hexane (10 mL). After stirring overnight the initial green added. After 30 min at room temperature all volatiles were
color of the solution turned light brown. The unreacted hydro- evaporated under vacuum and a brown residue was dissolved
gen was removed under vacuum and the solvent was evapor- in hexane and crystallized by a slow solvent distillation in a
ated to get the saturated solution. Crystallization in a three- refrigerator. A brown finely crystalline powder was obtained
necked ampule by slow distillation of the solvent into an arm after separation of ca. 0.2 mL of mother liquor.
close to the cooling side of a refrigerator afforded pale brown Recrystallization from the same hexane afforded pale greenish-
needle aggregates unsuitable for X-ray investigation. Yield brown crystals. Yield 0.19 g, 84%.
0.19 g, 87%.
Analytical data for compound 8. EI-MS (130 °C): m/z (relative
Analytical data for compound 6. Mp. 65 °C dec. EI-MS abundance) 489 (M•+; 3), 474 ([M − Me]+; 6), 437 (19), 436 (65),
(80 °C): m/z (relative abundance) 436 (10), 435 (M•+; 36), 434 435 (70), 434 (83), 433 ([M − C4H8]+; 100), 432 (81), 431 (9),
(68), 433 ([M − 2H]+; 100), 432 (26), 431 (15), 361 ([M − 430 (12), 429 (14), 419 (12), 418 ([M − C4H8 − Me]+; 21),
SiMe3H]+; 11), 73 ([SiMe3]+; 60), 59 ([SiMe2H]+; 12). IR (KBr, 362 (12), 361 (31), 360 ([M − C4H8 − Me − SiMe3]+; 30),
cm−1): 2952 (vs), 2909 (vs), 2865 (s), 2722 (vw), 1525 (m) ν(Ti– 359 (20), 358 (11), 287 ([M − C4H8 − Me − 2SiMe3]+; 17), 285 (13),
H), 1480 (w), 1448 (m), 1404 (w), 1378 (s), 1348 (w), 1329 (m), 217 (16), 208 (18), 73 (87), 59 (17), 57 (13), 56 (48), 55 (39),
1245 (s), 1128 (w), 1083 (vw), 1022 (m), 992 (vw), 948 (vw), 846 45 (21). IR (KBr, cm−1): 3015 (w), 2970 (s, sh), 2950 (s), 2908 (s,
(vs), 782 (m), 756 (m), 687 (m), 669 (w), 633 (m), 576 (w), 511 b), 2850 (m, sh), 2720 (vw), 1629 (w), 1480 (w), 1450 (m),
(w), 469 (m), 429 (m). Upon exposure of the KBr pellet to air 1432 (w), 1416 (w), 1407 (w), 1378 (m), 1347 (m), 1246 (vs),
the absorption band at 1525 cm−1 disappeared and new bands 1146 (w), 1127 (m), 1059 (w), 1020 (m), 982 (vw), 953 (w), 847 (vs),
1
at 780 a 577 cm−1 grew strong in intensity. H NMR (toluene- 835 (vs), 755 (s), 686 (m), 659 (w), 634 (m), 576 (vw), 417 (m). The
d8): 2.0 (br s, Δν1/2 ∼ 50 Hz, 18H, SiMe3); 13.7 (br s, Δν1/2 ∼ 150 Hz, band at 576 cm−1 grew in intensity upon exposure of the KBr
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
pellet to air. EPR (toluene-d8, 22 °C): giso = 1.962, ΔH = 24 G; asymmetric part of the unit cell; however, both molecules were
(toluene-d8, −140 °C): g1 = 2.000, g2 = 1.985, g3 = 1.901, gav struck with a similar disorder (with the monochloride ana-
=
1.962 (impurity Cp′2Ti-OR: giso = 1.976, ΔH = 3.5 G, aTi = 7.8 G; logue present) as for 1A. The treatment of the disorder was
(−140 °C): g1 = 2.000, g2 = 1.983, g3 = 1.953, gav = 1.978). UV-vis analogous with the former case.
(toluene-d8, nm): 485 > 620(sh). 1H NMR spectrum of 8
For complex 5 there were two symmetrically independent
showed mainly very broad signals (6000–10 000 Hz) centered at molecules present in the asymmetric part of the unit cell. One
ca. 8–12 ppm (see Fig. S22 in the ESI†). Found (%): C, 68.64; of the two molecules had, together with the monohydride
H, 9.97. C28H49Si2Ti requires (%): C, 68.67; H, 10.01.
complex discernible on the Fourier maps, the analogous oxo-
complex also. The refinement of this disorder was done by
constraining the respective site occupation factors. In complex
8, the disorder of the aliphatic ligand overlapping with a chlor-
Attempted reaction of titanocenes [(C5Me4tBu)2Ti] and
[(C5Me4SiMe3)2Ti] with butadiene
Solutions from the above described hydrogenation of titano- ide was treated analogously with the above cases.
cenes [(C5Me4tBu)2Ti] and [(C5Me4SiMe3)2Ti] yielding equili-
bria with their dihydrides and showing the monohydrides 5
and 6 as impurities were degassed by two freezing/thawing
Computational details
DFT calculations were carried out at the Fermi cluster at the
J. Heyrovský Institute of Physical Chemistry, Czech Academy of
Sciences, v.v.i. using Gaussian 09, Revision D.01.42 Geometry
optimizations and NBO analyses were done using the M06
functional and the 6-31G(d,p) basis set used for all atoms. The
Hessians were computed analytically before starting the first
geometry optimization steps. All calculations were performed
using ultrafine integration grids.
cycles and ca. 0.2 mL of liquid butadiene was added to each
solution afterwards. After standing for one day at room temp-
erature all volatiles were evaporated to dryness and the
residues were dissolved in toluene-d8. Both samples revealed
the 1H NMR spectra of titanocenes and 2-buten-1-yl derivatives
7 and 8 (see Fig. S21 and S22 in the ESI†) whereas the
volatiles contained unreacted butadiene. The inertness of
these titanocenes towards butadiene was thus proved.
X-ray crystallography
Suitable samples of 1A (CCDC 1545965), 1B (CCDC 1545966),
5 (CCDC 1545967), and 8 (CCDC 1545964) were filled into
Lindemann glass capillaries under a purified argon atmo-
sphere in a Labmaster 130 glovebox (mBraun). Diffraction data
were collected using MoKα radiation (λ = 0.71073 Å) either on
a Nonius Kappa diffractometer equipped with a Bruker
ApexII-CCD detector (5, 8) or on a Bruker D8 Venture diffracto-
meter (1A, 1B). For the measurement of crystal samples having
larger dimensions, a diffractometer employing a 0.6 mm colli-
mator and no focussing of the primary beam was employed.
The phase problem was solved by intrinsic phasing
(SHELXT)40 and the structure models were refined by full
matrix least squares on F2 (SHELXL-2014).40 All non-hydrogen
atoms were refined anisotropically. Unless otherwise given,
hydrogen atoms were placed in idealized positions and refined
isotropically using the riding model. Molecular graphics was
carried out using PLATON.41
In the solid-state structure of 1A, apart from the dihydride
complex its parent monochloride analogue occupied the same
position in the unit cell. The two hydrogen atoms bonded to
titanium were refined with coordinates left free to vary and
with site occupation factors constrained with that of the chlor-
ide atom. The site occupation factors were constrained to yield
one after summing; their final values became 0.32 for the
chloride and 0.68 for both hydride atoms. In addition to this
disorder on the bent titanocene complex, the asymmetric part
of the unit cell included an additional one-half of a permethyl-
titanocene molecule, whose heavy atoms were refined with no
constraints, while its hydrogen atoms were refined only isotro-
pically and using the riding model. In the case of 1B there
were two symmetrically independent molecules present in the
Acknowledgements
This research was supported by a Grant Agency from the Czech
Republic (Project No. P207/12/2368). I. C. is grateful to the
Ministry of Education, Youth and Sport of the Czech Republic
(Project No. 0021620857). J. K. was supported by the Grant
Agency of the Czech Republic (Project No. 203/09/P276). R. G.
is thankful to the OP VVV “Excellent Research Teams” (Project
No. CZ.02.1.01/0.0/0.0/15_003/0000417 – CUCAM).
Notes and references
1 G. J. Kubas, J. Organomet. Chem., 2014, 751, 33–49.
2 R. H. Crabtree, Chem. Rev., 2016, 116, 8750–8769.
3 F. Maseras, A. Lledos, E. Clot and O. Eisenstein, Chem.
Rev., 2000, 100, 601–636.
4 L. Pauling, The Nature of the Chemical Bond, Ithaca, 3rd
edn, 1960.
5 (a) J. M. Manriquez, D. R. McAlister, R. D. Sanner and
J. E. Bercaw, J. Am. Chem. Soc., 1976, 98, 6733–6735;
(b) J. M. Manriquez, D. R. McAlister, R. D. Sanner and
J. E. Bercaw, J. Am. Chem. Soc., 1978, 100, 2716–2724;
(c) J. M. Manriquez, D. R. McAlister, E. Rosenberg,
A. M. Shiller, K. L. Williamson, S. I. Chan and J. E. Bercaw,
J. Am. Chem. Soc., 1978, 100, 3078–3083.
6 (a) D. M. Roddick, M. D. Fryzuk, P. F. Seidler,
G. L. Hillhouse and J. E. Bercaw, Organometallics, 1985, 4,
97–104; (b) P. J. Chirik, M. W. Day and J. E. Bercaw,
Organometallics, 1999, 18, 1873–1881.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
7 (a) P. J. Chirik and J. E. Bercaw, Organometallics, 2005, 24,
5407–5423; (b) C. McDade and J. E. Bercaw, J. Organomet.
Chem., 1985, 279, 281–315.
3578; (b) J. Pinkas, L. Lukešová, R. Gyepes, I. Císařová,
P. Lönnecke, J. Kubišta, M. Horáček and K. Mach,
Organometallics, 2007, 26, 3100–3110.
8 P. J. Chirik, Organometallics, 2010, 29, 1500–1517.
9 A. J. Hoskin and D. W. Stephan, Coord. Chem. Rev., 2002,
233–234, 107–129.
10 (a) J. E. Bercaw and H. H. Brintzinger, J. Am. Chem. Soc.,
1971, 93, 2045–2046; (b) J. E. Bercaw, R. H. Marvich,
L. G. Bell and H. H. Brintzinger, J. Am. Chem. Soc., 1972,
26 (a) T. E. Hanna, E. Lobkovsky and P. J. Chirik,
Organometallics, 2009, 28, 4079–4088; (b) T. E. Hanna,
W. H. Bernskoetter, M. W. Bouwkamp, E. Lobkovsky and
P. J. Chirik, Organometallics, 2007, 26, 2431–2438;
(c) S. P. Semproni, C. Milsmann and P. J. Chirik,
Organometallics, 2012, 31, 3672–3682.
94, 1219–1238; (c) J. E. Bercaw, J. Am. Chem. Soc., 1974, 96, 27 J. Pinkas, I. Císařová, A. Conde, R. Fandos, M. Horáček,
5087–5095.
J. Kubišta and K. Mach, J. Organomet. Chem., 2009, 694,
1971–1980.
28 M. Horáček, J. Hiller, U. Thewalt, M. Polášek and K. Mach,
Organometallics, 1997, 16, 4185–4191.
11 W. W. Lukens Jr., P. T. Matsunaga and R. A. Andersen,
Organometallics, 1998, 17, 5240–5247.
12 J. Pinkas, I. Císařová, R. Gyepes, J. Kubišta,
M. Horáček and K. Mach, Organometallics, 2012, 31, 5478– 29 (a) Y. Takenaka and Z. Hou, Organometallics, 2009, 28,
5493.
5196–5203; (b) O. T. Summerscales, E. R. Batista,
B. L. Scott, M. P. Wilkerson and A. D. Sutton, Eur. J. Inorg.
Chem., 2016, 4551–4556; (c) H. J. Heeres, J. Renkema,
M. Booij, A. Meetsma and J. H. Teuben, Organometallics,
1988, 7, 2495–2502.
13 J. W. Pattiasina, C. E. Hissink, J. L. de Boer, A. Meetsma,
J. H. Teuben and A. L. Spek, J. Am. Chem. Soc., 1985, 107,
7758–7759.
14 J. Pinkas, R. Gyepes, I. Císařová, J. Kubišta, M. Horáček
and K. Mach, Organometallics, 2014, 33, 3399–3413.
15 G. A. Luinstra and J. H. Teuben, J. Am. Chem. Soc., 1992,
114, 3361–3367.
16 M. D. Walter, C. D. Sofield and R. A. Andersen,
Organometallics, 2008, 27, 2959–2970.
17 T. E. Hanna, E. Lobkovsky and P. J. Chirik, Eur. J. Inorg.
Chem., 2007, 2677–2685.
30 W. W. Lukens Jr., M. R. Smith III and R. A. Andersen, J. Am.
Chem. Soc., 1996, 118, 1719–1728.
31 K. Mach and B. Raynor, J. Chem. Soc., Dalton Trans., 1992,
683–688.
32 (a) J. W. Pattiasina, H. J. Heeres, F. van Bolhuis,
A. Meetsma, J. H. Teuben and A. L. Spek, Organometallics,
1987, 6, 1004–1010; (b) R. Gyepes, V. Varga, M. Horáček,
J. Kubišta, J. Pinkas and K. Mach, Organometallics, 2010,
29, 3780–3789; (c) V. Varga, I. Císařová, R. Gyepes,
M. Horáček, J. Kubišta and K. Mach, Organometallics,
2009, 28, 1748–1757; (d) M. Horáček, R. Gyepes, J. Kubišta
and K. Mach, Inorg. Chem. Commun., 2004, 7,
155–159.
18 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb,
A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg,
Organometallics, 2010, 29, 2176–2179.
19 K. Mach, V. Varga, V. Hanuš and P. Sedmera, J. Organomet.
Chem., 1991, 415, 87–95.
20 Hydrogenation of [Cp*2Ti(η2-C2H4)] was assumed to give
[Cp*2TiH2]: S. A. Cohen, P. R. Auburn and J. E. Bercaw, 33 J. Pinkas, I. Císařová, R. Gyepes, J. Kubišta, M. Horáček
J. Am. Chem. Soc., 1983, 105, 1136–1143. and K. Mach, Organometallics, 2013, 32, 6306–6314.
21 (a) J. Pinkas, R. Gyepes, I. Císařová, J. Kubišta, M. Horáček 34 J. M. de Wolf, A. Meetsma and J. H. Teuben,
and K. Mach, Dalton Trans., 2015, 44, 7276–7291; Organometallics, 1995, 14, 5466–5468.
(b) P. Arndt, V. V. Burlakov, A. Spannenberg and 35 J. Pinkas, I. Císařová, R. Gyepes, M. Horáček, J. Kubišta,
U. Rosenthal, Inorg. Chem. Commun., 2007, 10, 792–794;
(c) V. V. Burlakov, A. V. Polyakov, A. I. Yanovsky,
J. Čejka, S. Gómez-Ruiz, E. Hey-Hawkins and K. Mach,
Organometallics, 2008, 27, 5532–5547.
Yu. T. Struchkov, V. B. Shur, M. E. Vol’pin, U. Rosenthal 36 (a) K. Mach, H. Antropiusová, V. Varga and V. Hanuš,
and H. Görls, J. Organomet. Chem., 1994, 476, 197–206.
22 M. Horáček, P. Štěpnička, R. Gyepes, I. Císařová,
M. Polášek, K. Mach, P.-M. Pellny, V. V. Burlakov,
J. Organomet. Chem., 1988, 358, 123–133; (b) K. Mach,
H. Antropiusová and V. Hanuš, Transition Met. Chem.,
1985, 10, 302–307.
W. Baumann, A. Spannenberg and U. Rosenthal, J. Am. 37 (a) G. Erker, G. Kehr and R. Froehlich, J. Organomet. Chem.,
Chem. Soc., 1999, 121, 10638–10639.
2005, 690, 6254–6252; (b) G. Erker, G. Kehr and
R. Froehlich, J. Organomet. Chem., 2004, 689, 4305–4318;
(c) G. Erker, J. Wicher, K. Engel, F. Rosenfeldt and
W. Dietrich, J. Am. Chem. Soc., 1980, 102, 6344–6346;
(d) M. Dahlmann, G. Erker, R. Froehlich and O. Meyer,
Organometallics, 1999, 18, 4459–4461; (e) M. Horáček,
J. Pinkas, J. Kubišta, I. Císařová, R. Gyepes and
P. Štěpnička, Collect. Czech. Chem. Commun., 2007, 72, 679–
696.
23 (a) P.-M. Pellny, V. V. Burlakov, W. Baumann,
A. Spannenberg, M. Horáček, P. Štěpnička, K. Mach and
U. Rosenthal, Organometallics, 2000, 19, 2816–2819;
(b) P.-M. Pellny, F. G. Kirchbauer, V. V. Burlakov,
W. Baumann, A. Spannenberg and U. Rosenthal, Chem. –
Eur. J., 2000, 6, 81–90.
24 T. E. Hanna, E. Lobkovsky and P. J. Chirik, J. Am. Chem.
Soc., 2004, 126, 14688–14689.
25 (a) M. Horáček, V. Kupfer, U. Thewalt, P. Štěpnička, 38 H. Antropiusová, A. Dosedlová, V. Hanuš and K. Mach,
M. Polášek and K. Mach, Organometallics, 1999, 18, 3572–
Transition Met. Chem., 1981, 6, 90–93.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
39 V. Varga, K. Mach, M. Polášek, P. Sedmera, J. Hiller, 41 A. L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2009,
U. Thewalt and S. I. Troyanov, J. Organomet. Chem., 1996,
506, 241–251.
40 G. M. Sheldrick, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 2015, 71, 3–8.
65, 148–155.
42 M. J. Frisch, et al., Gaussian 09, Revision D.01, Gaussian,
Inc., Wallingford CT, 2013. (Full reference in
the ESI.†).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017