Ethene complexes of bulky titanocenes, their thermolysis, and their reactivity toward 2-butyne

Article

pubs.acs.org/Organometallics

Ethene Complexes of Bulky Titanocenes, Their Thermolysis, and

Their Reactivity toward 2-Butyne

Jirí Pinkas,^†Ivana Císarova,^‡Robert Gyepes,^§Jirí Kubista,^†Michal Horacek,^†and Karel Mach*^,†

̌

́

̌

́

̌

^†J. Heyrovsky Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, v.v.i., Dolejsk

̌

ova 3, 182 23 Prague 8,

́

Czech Republic

^‡Department of Inorganic Chemistry, Charles University, Hlavova 2030, 128 40 Prague 2, Czech Republic

^§Faculty of Education, J. Selye University, Bratislavska

́ ́

cesta 3322, 945 01 Komarno, Slovak Republic

S

* Supporting Information

ABSTRACT: Ethene complexes of titanocenes [Ti(II)(η²-C₂H₄)(Cp′)₂] for Cp′ = η⁵-C₅Me₅(1), η⁵-C₅Me₄t-Bu (2), η⁵-

C₅Me₄SiMe₃(3), and η⁵-C₅HMe₄(4) were prepared by reduction of corresponding titanocene dichlorides with magnesium in

THF in the presence of ethene. Thermolysis of 1−3 in toluene solution at a maximum of 100 °C resulted in elimination of

ethane, aﬀording cleanly doubly tucked-in titanocene compounds 5−7, respectively. Experiments with 2 and 3 in NMR tubes

proved that symmetrical isomers 6a and 7a were formed ﬁrst, and these thermally isomerized to thermodynamically more stable

asymmetrical isomers 6b and 7b. The energy diﬀerence between 7a and 7b calculated by DFT methods was 15.3 kJ/mol.

Thermolysis of 4 in m-xylene required a temperature of 135 °C, aﬀording a mixture of 8b > 8a and “dimeric dehydro-titanocene”

9 as a concurrent product of hydrogen abstraction. In contrast to thermolysis in solvents, heating of 1 and 2 in high vacuum to

135 °C resulted in sublimation of known titanocenes [Ti(C₅Me₅)₂] (10) and [Ti(η⁵-C₅Me₄t-Bu)₂] (13) (Chirik et al. J. Am.

Chem. Soc. 2004, 126, 14688−14689), respectively. The former isomerized in hexane solution to the tucked-in hydride

[TiH{C₅Me₄(CH₂)}(C₅Me₅)] (10A) as described by Bercaw (J. Am. Chem. Soc. 2004, 126, 14688−14689). A mixture of 10/

10A decayed within days to give major paramagnetic products [TiH(C₅Me₅)₂] (11) and singly tucked-in titanocene

[Ti{C₅Me₄(CH₂)}(C₅Me₅)] (12) and minor diamagnetic 5 and its so far unknown, less stable isomer [Ti{C₅Me₄(CH₂)}₂]

(10B), identiﬁed by NMR spectra and corroborated by DFT calculations. Solid 3 eliminated ethene at only 80 °C, leaving

titanocene 14, whereas compound 4 sublimed at 135 °C mostly without decomposition. Cocrystals of 10 with [TiCl(C₅Me₅)₂]

(1:2) (10C) aﬀorded an X-ray single-crystal structure with linear geometry for 10. The ethene complexes 1−4 diﬀered in their

reactivity toward but-2-yne: compounds 1 and 4 yielded the respective [Ti(IV)(η¹: η¹-CH₂CH₂CMeCMe)(Cp′)₂] 2,3-

dimethyltitanacyclopent-2-ene complexes 15 and 16, whereas 2 and 3 replaced ethene with but-2-yne, aﬀording the [Ti(II)(η²-

MeCCMe)(Cp′)₂] complexes 17 and 18, respectively. Crystal structures of 2, 4, 10C, 15, 17, and 18 have been determined by

X-ray crystallography.

reagents, sometimes aﬀording diﬀerent products depending on

the Lewis acidity of the central titanium atom and its steric

congestion in the titanocene shell.⁴Such behavior also occurred

upon thermolysis of Cp″₂Ti(η²-BTMSE) complexes, showing

the formation of dimeric titanocenes with elimination of

unchanged BTMSE for n = 0−3, whereas for n = 5 the doubly

tucked-in titanocene [Ti(η⁵-C₅Me₅){η⁴:η³-C₅Me₃(CH₂)₂}]^5a

was cleanly formed as a result of transfer of two hydrogen

atoms to leaving BTMSE.^5b

INTRODUCTION

■

It has been well established that group 4 d²metallocenes at the

onset of their formation are stabilized by coordination of

internal acetylenes¹or ethene,²forming well-deﬁned complexes

that can be used to study reactivities of parent metallocenes

toward organic reagents instead of unstable metallocenes.³

Among internal acetylenes bis(trimethylsilyl)ethyne (BTMSE)

has been a superior leaving ligand, in most cases not interfering

with a rather common oxidative addition of the reagents to a

titanocene moiety.^1a,b,3This circumstance allowed us to study

the reactivity of a whole range of titanocene moieties (Cp″₂Ti^II)

(Cp′ = η⁵-C₅H_5‑nMe_n; n = 0−5) toward various nonprotic

Received: May 25, 2012

Published: July 13, 2012

5478

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

Of titanocene−ethene complexes a fundamental study

describing the synthesis of (η⁵-C₅Me₅)₂Ti(η²-C₂H₄) (1), its

crystal structure, and its reactivity was published by J. E. Bercaw

et al. nearly 30 years ago.^2aThe synthesis of 1 was based on

addition of ethene to titanocene [Ti(η⁵-C₅Me₅)₂] in the form

of its nitrogen complex [{Ti(η⁵-C₅Me₅)₂}₂(μ-N₂)],⁶or it was

generated at the onset of titanocene formation during reduction

of [TiCl₂(η⁵-C₅Me₅)₂] with sodium amalgam in toluene. The

crystal structure of 1 revealed the coordination of ethene in a

back-bonded fashion, as predicted earlier by Lauher and

Hoﬀmann on the basis of EHMO calculation.⁷Compound 1

reacted with excessive ethene under pressure to give reversibly

titanacyclopentane (Scheme 1), whose instability at ambient

temperature was established earlier.⁸

[Ti(C₅Me₄R)₂(η²-C₂H₄)] (R = Me (1), t-Bu (2), SiMe₃(3),

and H (4)) were obtained in nearly quantitative yields by

reduction of corresponding titanocene dichlorides with

magnesium in tetrahydrofuran in the presence of excess ethene

(Scheme 3).

Scheme 3

Scheme 1

All the compounds were obtained as green crystals from

greenish-yellow hexane solutions, the appearance previously

reported for 1^2aand 3.^10aThe crystal structures of new

compounds 2 (Figure 1) and 4 (Figure 2) are very similar to

those of known 1^2aand 3,^10aall showing a 2-fold rotation axis

passing through the titanium atom and the center of the ethene

CC bond.

Internal alkynes reacted with 1 either by displacement of

ethene to give [Ti(η⁵-C₅Me₅)₂(η²-R¹CCR²)] products or by

insertion to form Ti(IV) titanacyclopentenes depending on the

nature of substituents R¹and R²(Scheme 2).

Whereas tolane gave only the back-bonded complex [Ti(η⁵-

C₅Me₅)₂(η²-PhCCPh)],^2b,9the reaction of 1 with but-2-yne

aﬀorded titanacyclopentene product. Other back-bonded

ethene complexes were reported for only bulky titanocene

moieties. Compounds [Ti(η⁵-C₅Me₄SiMe₃)₂(η²-C₂H₄)],^10a

[Ti(η⁵-1,3-C₅H₃t-Bu₂)₂(η²-C₂H₄)],^10band [Ti(η⁵-

C₅Me₄SiMe₂t-Bu)₂(η²-C₂H₄)]^10cwere prepared by addition

of ethene to thermally stable titanocenes and characterized by

common spectroscopic methods and single-crystal X-ray

diﬀraction; however, their reactivities were not investigated.

The state of the art of low-valent titanocene chemistry has been

recently reviewed in the most knowledgeable way.^11a

Recent achievements in investigations of electronic and steric

eﬀects of bulky titanocene moieties [Ti(C₅Me₄R)₂] (R = Me,

Et, Ph, i-Pr, t-Bu, SiMe₃, SiMe₂t-Bu, and SiMe₂Ph) on their

reactivity toward molecular nitrogen^11a−cor carbon mon-

oxide^11dhave encouraged us to communicate our results on the

synthesis and structure of bulky titanocene−ethene complexes

[Ti(C₅Me₄R)₂(η²-C₂H₄)] (R = Me, t-Bu, SiMe₃, and H), their

thermolytic behavior in solution and gas phase, and their

reactivity to but-2-yne.

Figure 1. PLATON drawing of 2 at the 30% probability level, with

atom-labeling scheme. Hydrogen atoms are omitted for clarity except

ethene hydrogen atoms H14A and H14B.

The dihedral angle between the least-squares planes of

staggered cyclopentadienyl rings (φ) is virtually equal for 1−3

since the bulky tert-butyl and SiMe₃groups are situated

symmetrically on opposite sides of the Cg, Ti, Cg′ plane (Cg is

the centroid of the cyclopentadienyl ring). For 4, the proton-

bearing ring carbon atoms in hinge positions allow for a

RESULTS AND DISCUSSION

■

Synthesis of Titanocene−Ethene Complexes 1−4 and

Their Structures. A series of titanocene−ethene complexes

Scheme 2

5479

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

Table 2. ¹H and ¹³C NMR Data (ppm) and Infrared ν(C−H)

Vibrations (cm⁻¹) for Coordinated Ethene and λ_max(nm) for

Compounds 1−4

a

compound

δ_H

δ_C

ν(C−H)

λ_max

ref

b

1

2

3

4

2.02

2.08

2.34

2.03

105.1

3042

3051

940

990

925

2a

101.80

104.3

this work

10a

c

n.o.

102.66

3034

this work

a

Chemical shifts for free ethene (benzene-d₆, 25 °C): δ_H= 5.25 ppm

and δ_C= 122.96 ppm.¹²This work. Not observed.

b

c

this low-intensity band is hidden under ν(C−C) and C−H

deformation vibrations of the cyclopentadienyl ligands.^2a

All the compounds display an electronic absorption band in

the near-infrared region with λ_maxin a close range, 925−990 nm

(Table 2). This band is very broad (half-width ∼250 nm), and

since no other absorption bands occur in the visible region, its

extension to the 700−800 nm region is responsible for the

green color of solutions of 1−4. A yellow coloration of highly

diluted solutions arises from a continuous absorption increasing

from about 450 to 300 nm showing a shoulder at about 360

nm. The absorption band in the near-infrared region has been

observed for all compounds where persubstituted titanocene

(Ti^II) coordinates a double bond. Such compounds are

suggested to be unstable for substituted oleﬁns, as the repulsion

between the oleﬁn substituent(s) and η⁵-C₅Me₅ligands

prevents eﬀective oleﬁn π-coordination.^2aHowever, com-

pounds with the double bond singly or doubly tethered to

the cyclopentadienyl ligands are thermally robust. Their

schematic structures, λ_max, and references¹³are given in Chart 1.

The EI-MS spectra of 1−4 do not show molecular ions due

to the loss of ethene either upon evaporation or as a result of

electron impact.¹⁴The titanocene ions form a base peak for 1

and 4, whereas for 2 and 3 they are slightly less abundant. In all

cases the titanocene ions are accompanied by ion patterns for

titanocenes poorer in hydrogen atom(s). For 2, this loss of

hydrogen atoms is accompanied by abundant elimination of

butyl groups. To distinguish the processes proceeding upon

evaporation and subsequent ionization, compounds 1, 2, and 4

were evaporated under high vacuum and the products analyzed

(see below). Compound 4 sublimed largely without decom-

position, which implies an overwhelming ethene elimination

upon electron impact. Compound 1 aﬀorded largely and

compound 2 exclusively the corresponding titanocene,

indicating thermolytic ethene elimination prior to electron

Figure 2. PLATON drawing of 4 at the 30% probability level, with

atom-labeling scheme. Hydrogen atoms are omitted for clarity except

ethene hydrogen atoms H10A and H10B.

remarkably larger angle φ (Table 1). The geometry of the

titanium−ethene moiety characterized by Ti−C and CC

bond lengths is close to that of the parent compound 1 not

allowing to diﬀerentiate the strength of ethene coordination

from shortening of Ti−C and elongation of CC bonds.

Hydrogen atoms of the coordinated ethene molecule were

observed for all compounds to be displaced from the double-

bond plane farther away from the titanium atom, similar to that

found for 1. In distinction from 1, the hydrogen atoms on each

ethene carbon atom were nonequivalent for 2−4. The steric

hindrance between ethene and cyclopentadienyl ring carbon

atoms is largest for 1 (short contacts <3.6 Å)^2asince the Ti−Cg

distances increase signiﬁcantly in the order 1 < 2 < 3 (Table 1),

whereas the shortest Ti−Cg distance of 2.0711(7) Å found for

4 is associated with the smallest Cg−Ti−Cg angle, not inducing

steric hindrance at the coordinated ethene.

In C₆D₆solution, ¹H and ¹³C NMR spectra of 1−4 indicated

C_2vmolecular symmetry, showing the coordinated ethene

resonances as singlets shifted high-ﬁeld from the value for free

ethene.¹²Infrared spectra indicated the presence of back-

bonded ethene, showing its ν(C−H) absorption band of low

intensity except for 3, where it was not observed due to overlap

with a much stronger absorption due to the presence of SiMe₃

groups (see Table 2). The ν(CC) stretching vibration of the

back-bonded ethene is shifted to lower wavenumbers, where

Table 1. Selected Bond Lengths and Bond Angles for 1, 2, 3, and 4

a

b

c

d

1

2

3

4

Bond Lengths (Å)

e

Ti−Cg(1)

2.092(5)

2.1182(15)

2.1134(3)

2.177(4)

2.129(3)

2.125(1)

2.167(4)

1.442(9)

2.0711(7)

2.0644(2)

2.1588(16)

1.446(4)

f

Ti−Pl(1)

g

Ti−Cα

2.160(4)

1.438(5)

g

Cα−Cβ

1.454(9)

Bond Angles (deg)

e

Cg(1)−Ti−Cg(1′)

143.6(4)

38.9(1)

40.6(4)

146.36(7)

39.0(2)

147.0(1)

38.9(2)

39.7(1)

142.34(3)

39.13(10)

46.43(7)

g

Cα−Ti−Cβ

h

φ

41.08(12)

a

b

c

Taken from ref 2a. Symmetry operation used to generate equivalent positions: 1 − x, y, −z + 1/2. Taken from ref 10a. Symmetry operation used

d

e

to generate equivalent positions: − x, y, −z + 1/2. Symmetry operation used to generate equivalent positions: − x + 1, − y + 1, −z + 1/2. Cg(1) is

f

g

the centroid of the C(1−5) cyclopentadienyl ring. Pl(1) is the least-squares plane of the C(1−5) cyclopentadienyl ring. Cα, Cβ are ethene carbon

h

atoms. Dihedral angle between the least-squares planes of cyclopentadienyl rings Pl(1) and Pl(1′).

5480

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

Thermolysis of 2 and 3 proceeds at 95 and 90 °C,

respectively, liberating ethane and yielding a mixture of

isomers. Thermolyses performed in NMR tubes allowed

recognizing that symmetrical isomers (6a and 7a, respectively)

were formed ﬁrst, and these subsequently isomerized to

Chart 1

1

asymmetrical isomers (6b and 7b) (Scheme 4). The H and

¹³C NMR spectra of 6a were observed in a mixture with the

parent ethene complex 2, and those of 7a in the presence of a

minor content of 7b, which allowed their full assignment (see

Supporting Information). The system of two vicinal exo-

methylene groups in symmetrical and asymmetrical isomers

diﬀers in their NMR spectra pattern. Compounds with

symmetrical structure (6a, 7a) have a mirror plane containing

titanium and C_ipsoatoms, bisecting the C−C distance between

the two exomethylene groups. Hence, the spectra of sym-

metrical isomers displayed half the number of signals of

doubled intensity compared to asymmetrical ones. In the ¹³C

NMR spectra, resonances of exomethylene groups occurring in

the narrow range 70−73 ppm are speciﬁc for these compounds

(δ_C6a: 72.88; 7a: 72.01, 6b: 70.11 and 71.66; 7b: 70.38 and

71.88).

The subsequent conversion of symmetrical-to-asymmetrical

isomers indicates that the formation of the symmetrical isomers

is kinetically controlled, whereas their conversion to asym-

metrical ones is governed by thermodynamics. DFT calcu-

lations for 7a and 7b revealed, accordingly, that the overall

energy diﬀerence was 15.3 kJ/mol, with the lower energy

obtained for 7b. Also, the calculated lowest energy electronic

transition at 613 nm for 7a and at 564 nm for 7b are in

reasonable agreement with the experimental λ_max, 605 nm for

7a and 585 nm for 7b.¹⁵In spite of the increased internuclear

repulsion found for the latter molecule (7a: 3618.54 hartree;

7b: 3624.97 hartree), its higher stability is achieved through

stronger Coulomb and exchange interactions. The NBO

delocalization scheme is rather complicated, and the enhanced

stability of 7b cannot be explained on the basis of a single or a

few orbital energies; however, comparing the number of strong

delocalizations and their energies, the asymmetric arrangement

appears to be the favored one.

impact (see below). The absence of ethane ions m/z 30 proves

that the hydrogen transfer to leaving ethene proceeding upon

thermal decomposition in solution (see below) does not apply

to the gas-phase decomposition.

Thermolysis of 1−4 in Solution. Compounds 1−3

decompose with elimination of ethane upon heating in toluene

to a maximum temperature of 100 °C to give doubly tucked-in

titanocenes (Scheme 4). The thermolysis of 1 was particularly

Scheme 4

Compound 7b was previously obtained by thermolysis of

[Ti(C₅Me₄SiMe₃)₂] under high vacuum at 140 °C;¹⁵however,

an attempt to similarly prepare 6b by thermolysis of 2 under

vacuum revealed a high thermal stability of the titanocene

[Ti(η⁵-C₅Me₄t-Bu)₂] (see below).

Thermolysis of compound 4 required a higher temperature

and was therefore carried out in m-xylene. Monitoring of

thermolysis by electronic absorption spectroscopy revealed that

the absorption band of 4 at 925 nm began to decrease at only

130 °C and disappeared after 8 h. In distinction to the above-

mentioned quantitative thermolytic reactions a mixture of

known¹⁶minor 8a and major 8b was obtained in only 48%

yield by extraction of the overall solid thermolytic product with

hexane (Scheme 5). This yield is, however, superior to

previously obtained ﬁgures since the boiling of

[TiCl₂(C₅HMe₄)₂] with LiAlH₄in toluene followed by

sublimation of soluble products aﬀorded a 10% yield,^16aand

the thermolysis of [Ti(C₅HMe₄)₂(η²-BTMSE)] in m-xylene at

150 °C gave an 8% yield of a mixture of 8a ≪ 8b. In the latter

case, the main product was a brown, amorphous, nonvolatile,

and insoluble solid.^16bHere, the other product, insoluble in

hexane, was extracted with toluene to give a brown,

paramagnetic, low-soluble solid, which was identiﬁed by EI-

MS spectra to be the dimeric dehydrotitanocene [Ti(C₅HMe₄)-

clean, aﬀording after heating to 100 °C for 8 h a blue solution

of pure doubly tucked-in (i.e., η³:η⁴-allyldiene) species

[Ti{η³:η⁴-C₅Me₃(CH₂)₂}(η⁵-C₅Me₅)] (5).^5aThis method of

obtaining 5 is quantitative, similar to the above-mentioned

thermolysis of [Ti(C₅Me₅)₂(η²-BTMSE)], which requires,

however, a temperature of 150 °C.^5bNo other product besides

1 and 5 was found after heating to 80 °C for 24 h.

5481

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

1

gray-green layer was dissolved in toluene-d₈and proved by H

Scheme 5

and ¹³C NMR spectra to be compound 1. The H NMR

1

spectrum of the hexane-soluble sublimate in toluene-d₈revealed

a broad paramagnetic signal at δ 63.8 ppm, Δν_1/2≈ 170 Hz (25

°C), and a set of narrow resonances apparently corresponding

to those assigned by Bercaw et al.^6ato paramagnetic titanocene

[Ti(C₅Me₅)₂] (10) and its diamagnetic singly tucked-in

hydride isomer [TiH(C₅Me₅) (C₅Me₄CH₂)] (10A). In our

hands, the resonance of the hydrogen attached to titanium was

also not observed for 10A; however, the ¹³C NMR signal of the

tucked-in methylene was found at 76.7 ppm, between the

values for [TiMe(C₅Me₅)(C₅Me₄CH₂)] (73.9 ppm)²⁰and

[TiBr(C₅Me₅)(C₅Me₄CH₂)] (80.6 ppm).²¹Since these NMR

measurements were carried out within 3−6 h at ambient

temperature after preparation of the 10/10A solution, the

presence of products of their decomposition was already

discernible. They became dominant after one week at room

temperature, when the signals of 10 and 10A decreased to

about 0.5% intensity each. A relatively narrow paramagnetic

resonance at δ_H23.0 ppm, Δν_1/2≈ 170 Hz (25 °C), due to

δ_H(Me) of titanocene hydride [TiH(C₅Me₅)₂] (11)^6a,22and

much broader resonances at δ_H25.4 ppm with a shoulder at ca.

(μ-C₅Me₄)]₂(9), previously obtained by thermolytic demetha-

nation of [TiMe(C₅HMe₄)₂].¹⁷The product assignment to 9

was proved by EPR measurement in toluene glass, showing

electronic triplet-state spectra for the two Ti^IIIatoms identical

to the authentic sample.¹⁸Here, compound 9 can arise from

thermolysis of 4 provided one molecule each of ethane and

ethene are eliminated from two molecules of 4 (Scheme 5).

The isomerization of symmetrical to asymmetrical doubly

tucked-in titanocenes was not observed during thermolyses of

[Ti(C₅Me₄R)₂(η²-BTMSE)] (R = Ph or 4-ﬂuorophenyl)

compounds^19aor analogous ansa-compounds with R = 1/2

O(SiMe₂)₂(ref 19b) or 1/2 (CH₂)₈because the thermolyses

proceeded at temperatures of at least 150 °C.^13cProvided the

initial formation of the symmetrical isomer is a general case, it

completely isomerized to an asymmetrical one at this

temperature, except for 8a, whose minor presence was

20 ppm (15 + 6 H Δν_1/2≈ 1800 Hz), 12.6 ppm (2 H Δν_1/2

≈

440 Hz), and −11.9 ppm (6H, Δν_1/2≈ 460 Hz (25 °C))

assigned to the paramagnetic singly tucked-in titanocene

[Ti(C₅Me₅)(C₅Me₄CH₂)] (12)^6a,21,23dominated the para-

1

magnetic H NMR spectra. Integration of the signals of both

products revealed that 11 prevails over 12 (∼2:1).

Simultaneously, high-resolution ¹H NMR spectra of

diamagnetic species revealed the above-mentioned decay of

initially prevailing resonances of 10A, whereas another two sets

of resonances became well observable. The dominating set of

resonances (∼15% of Ti) consisted of four singlet signals

assigned to nonequivalent methyl groups (δ_H/δ_C[ppm]: 1.09/

10.12; 1.18/9.83; 1.57/11.10; 1.68/14.82) on one cyclo-

pentadienyl ring and two doublets at 1.20 and 1.31 ppm

assigned to diastereotopic methylene protons. The ¹³C

resonance of the methylene carbon was found at 69.24 ppm,

close to values for 5−8 and distinctly high-ﬁeld shifted from the

above-mentioned values for 10A and other diamagnetic singly

tucked-in titanocene derivatives. The spectrum is consistent

with the C₂molecular symmetry and can be interpreted as

16

1

established by H and ¹³C NMR spectra.

Thermolysis of 1−4 under High Vacuum. Compound 1

was completely evaporated under high vacuum at a temperature

of 135 °C, producing a major brown layer of condensed solid in

the air-cooled upper part of a glass ampule accompanied by a

minor gray-green layer in its lower part. The brown sublimate

was dissolved with a cool condensing hexane vapor to give a

solution, turning rapidly green even at −18 °C. The minor

Scheme 6

5482

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

arising from the so far unknown [Ti(C₅Me₄CH₂)₂] (10B). The

minor set of resonances belonged to 5. The content of this

most stable diamagnetic product grew slowly during the next

weeks at room temperature at the expense of 10B, and this

conversion could be accomplished by warming the solution to

80 °C for 3 h.

in intensity to some extent seemed to arise also at the expense

of 12 (see Supporting Information).

In attempts to isolate a single product from the above

vacuum thermolysis of 1, hexane solutions of the sublimate

were submitted for crystallization by slow evaporation of

hexane to a slightly cooler arm of an ampule in a refrigerator at

−5 °C. In several cases very small yields of green needle crystals

displaying the same IR spectra were obtained from repeated

experiments. Their X-ray single-crystal analysis revealed that

these were cocrystals containing titanocene 10 and [TiCl-

(C₅Me₅)₂] in a 1:2 molar ratio (denoted 10C) in spite of only

trace amounts of the chloride in the solution (see below). The

crystals of 10C were obtained in a large amount by

thermolysis/sublimation of a mixture of solid 1 and [TiCl-

(C₅Me₅)₂] (at molar ratio ∼1:1.5) followed by crystallization

from the hexane solution of all the sublimate. The crystals

appeared to be stable at −5 °C for at least half a year and at

ambient temperature for one month; hence, they could be used

as a source of authentic 10. When dissolved in toluene-d₈, the

disproportionation of 10 and 10A to 11 and 12 and other

The DFT-optimized structure of 10B (Figure 3) contains a

2-fold axis passing through the titanium atom and the center of

1

products was established by H NMR to proceed similarly to

that in the virtual absence of [TiCl(C₅Me₅)₂]. A large

paramagnetic signal of [TiCl(C₅Me₅)₂] (δ_HMe 15.0 ppm,

Δν_1/2≈ 1100 Hz, 25 °C) (cf. ref 25) remained constant during

the aging (see Supporting Information). EI-MS spectra of 10C

showed the molecular ions of 10 (m/z 318) and [TiCl-

(C₅Me₅)₂] (m/z 353) and their fragment ions. The presence of

11 could not be identiﬁed because the ionized molecule loses a

molecule of hydrogen to give m/z 317, which is always present

among fragment ions of 10 and its chloride. Infrared spectra in

KBr pellets proved that 10C reacts with the nitrogen

atmosphere of the glovebox upon grinding with KBr, whereas

the crystals remained intact during their handling for X-ray, EI-

MS, NMR, and melting point measurements. The infrared

spectra reproducibly displayed strong absorption bands at 2080

and 1964 cm⁻¹, weak bands at 2049 and 2016 cm⁻¹, and a very

weak band at 1713 cm⁻¹, all observed by Bercaw^6a,cand

Chirik^11bat slightly shifted positions and similar intesity

proportions in pentane solutions of [{Ti(C₅Me₅)₂}₂(μ-η¹:η¹-

N₂)] saturated with nitrogen gas at about 20 °C lower

temperature. They assigned the two strong absorption bands to

[Ti(C₅Me₅)₂(N₂)₂] and the other three bands to [{Ti-

(C₅Me₅)₂(N₂)}₂(μ-η¹:η¹-N₂)] containing end-on-coordinated

nitrogen molecules. In addition to common absorption bands

due to η⁵-C₅Me₅ligands, the absorption band at 438 cm⁻¹

indicates the presence of [TiCl(C₅Me₅)₂] (see Supporting

Information).

Referring to the Bercaw’s observation that 10A rearranges to

10 upon evaporation of a fresh solution of 10/10A and that 10

eliminates hydrogen upon sublimation to give 12,^6awe have

found that a brown residue after evaporation of a green hexane

solution of 10/10A sublimes under high vacuum with a

regularly (fully) broken breakable seal at 110 °C to give

unchanged 10. The reason for this discrepancy was found in a

higher vacuum achieved by cooling charcoal with liquid

nitrogen. When the same experiment was performed with

only a partly broken breakable seal, causing a lower pumping

speed above the thermolyzed material, compound 10

completely converted to a purple sublimate of 12, thus

conﬁrming Bercaw’s observation. In addition, the thermolysis

of 1 performed at conditions of the latter experiment also

aﬀorded compound 12 as the only product. Gaseous products

of this thermolysis trapped in the cooled charcoal were released

Figure 3. DFT-optimized structure of 10B: Ti−Cg 2.005 Å (Cg,

centroid of the cyclopentadienyl ring), Ti−CH₂2.245 Å, C(1)−CH₂

1.449 Å, Cg−Ti−Cg′ 153.84°, Cg−C(1)−CH₂142.40°.

connection between the two methylene carbon atoms, which

are in hinge positions of the bent titanocene skeleton. The

calculated geometric parameters for the Ti(C₅Me₄CH₂)

fragment (Figure 3) agree with the only high-precision X-ray

single-crystal data found for the nondisordered singly tucked-in

titanocene [Ti(C₅Me₄Ph)(C₅Me₃PhCH₂)]²⁴within a maxi-

mum 0.02 Å and 0.3° deviation.

The calculated ¹³C NMR spectrum excellently agreed with

the experimental one (see Supporting Information), and

calculated energies of 10B and isomeric 5 showed that the

latter is lower by 14.1 kJ/mol, consistent with the observed

isomerization of 10B to 5. Compound 10B probably arose

from thermal decomposition of 10A, whose considerable part,

however, disproportionated via the equilibrium with 10 into 11

and 12. The conversion of 10A to 10B requires elimination of

one hydrogen molecule; however, the free hydrogen could not

be observed in the reaction solution since it reacted readily with

12, aﬀording 11. This reaction was used here for the smooth

preparation of the latter compound. The above-mentioned,

about twice higher concentration of 11 with respect to that of

12 in the ﬁnal products of titanocene 10/10A aging is mainly

due to the hydrogen generated in the 10A → 10B conversion.

In order to prove the equilibrium character of the residual

titanocene concentration that persisted during aging for next

months, ¹H NMR spectra of solutions of 11 and 12 in toluene-

d₈were measured ﬁrst to prove the absence of 10 and then after

the solutions were mixed together. The known resonances of

titanocene 10 and 10A corresponding to about 0.5% of total

titanium concentration each were observed and decreased only

negligibly during the next four months. This result contradicts

conclusions by Luinstra and Teuben that thermolysis of

alkyltitanocenes [TiR(C₅Me₅)₂] to give RH and 12 is not

catalyzed by 10 but exclusively by 11.²³In the present systems

11 is in turn the most stable component; a nonassigned weak

paramagnetic resonance at δ 4.3 ppm, Δν_1/2≈ 270 Hz, growing

5483

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

by its warming and condensed to an ampule cooled with liquid

nitrogen, and were dissolved in toluene-d₈. ¹H NMR analysis of

this sample¹²revealed the ratio of ethene:ethane to be 1:1.

Since no compound 5 was formed in this thermolysis, one has

to suggest that the ethene eliminated from conversion 1 → 10

is eﬀectively hydrogenated with hydrogen that is evolved in the

subsequent conversion 10 → 12. Thus, the products of

thermolysis of 1 are dependent on the thermolytic medium

(yielding 5 in solution) and on the pressure in the gas phase

yielding 10 at a very low pressure and 12 at a higher pressure.

Thermolysis of solid 2 under high vacuum upon gradual

heating to 135 °C resulted in elimination of ethene and

sublimation of brown titanocene [Ti(η⁵-C₅Me₄t-Bu)₂] (13)

onto an ampule wall at ambient temperature (Scheme 7).

Compound 13 could be sublimed under the above conditions;

however, upon electron impact in the EI-MS spectrometer it

abundantly loses the butyl group and hydrogen atoms.

pound 14 was shown to be stable at 90 °C for 18 h; however, it

eliminated hydrogen at 140 °C under high vacuum to give 7b

nearly quantitatively.¹⁵

Compound 4 mostly sublimed at 135 °C without

decomposition, giving only an estimate of less than 5% of a

more volatile brown product. This product was highly soluble

in hexane and so far remains unknown.

Crystal Structure of Titanocene 10 in Cocrystal with

Its Chloride 10C. Cocrystals 10C contained four molecules of

10 and eight molecules of [TiCl(C₅Me₅)₂] in the monoclinic

unit cell (see Supporting Information). Whereas the two

equivalent molecules of [TiCl(C₅Me₅)₂] are crystallographi-

cally asymmetrical, possessing irregularly staggered cyclo-

pentadienyl ligands and approximately planar triangular

coordination of the titanium atom, the molecule of 10 contains

a 2-fold rotation axis passing through the titanium atom,

perpendicular to the Cg−Ti−Cg′ interconnection and the

C(4)−C(9) bond (Figure 4).

Scheme 7

Compound 13 crystallized as small brown crystals with high

mosaicity, unsuitable for X-ray investigation. Its paramagnetic

¹H NMR spectra (for temperature dependence of particular

chemical shifts see Supporting Information) agreed with the

data of Chirik et al.^11cCompound 13 did not contain

paramagnetic [Ti(η⁵-C₅Me₄t-Bu)(η⁵:η¹-C₅Me₄CMe₂CH₂)], an

analogue of 12, which arises from [TiMe(η⁵-C₅Me₄t-Bu)₂] by

thermal elimination of methane,²⁶as evidenced by low

abundance of its base peak (M^+•m/z 401) in the EI-MS

spectra and by the absence of its signal (g = 1.960, ΔH = 24.4

G) in solution EPR spectra.²⁶In contrast to the behavior of

10C titanocene 13 did not react with nitrogen during grinding

with KBr. This is in accord with the observation that it reacts

with nitrogen in solution only at temperatures as low as −75

°C.^11b,c

Figure 4. PLATON drawing of 10 at the 30% probability level, with

atom-labeling scheme. Hydrogen atoms are omitted for clarity.

Essential geometric parameters for molecules of 10 and

[TiCl(C₅Me₅)₂] in cocrystal 10C are listed in Table 3. The

linear molecule of 10 adopts approximately D_5dsymmetry with

exactly staggered cyclopentadienyl ligands and parallel least-

squares planes of the cyclopentadienyl rings. The cyclo-

pentadienyl carbon atoms show very small deviations from

their least-squares planes (maximum 0.0041(24) Å for C(5))

Table 3. Selected Bond Lengths and Bond Angles for 10C

a

10

TiCl(C₅Me₅)₂

Bond Lengths (Å)

Compound 3 liberated ethene upon heating to 80 °C under

high vacuum, aﬀording the known titanocene [Ti(η⁵-

C₅Me₄SiMe₃)₂] (14) (Scheme 8) as described for BTMSE

liberation from [Ti(η⁵-C₅Me₄SiMe₃)₂(η²-BTMSE)].^10aCom-

b

Ti−Cg(1)

Ti−Cg(2)

Ti−Pl(1)

Ti−Pl(2)

1.983(2)

2.046(2)

2.036(2)

2.0461(9)

2.0356(9)

2.398(2)

b

1.9813(18)

Ti−Cl

Scheme 8

Bond Angles (deg)

b

Cg(1)−Ti−Cg(2)

Cg(1)−Ti−Cl

Cg(2)−Ti−Cl

180.00(11)

0.00(29)

147.47(10)

106.30(9)

106.21(10)

Dihedral Angles (deg)

c

φ

30.84(26)

a

Symmetry operation used to generate equivalent positions: −x + 2, −

b

y, −z. Cg(1) and Cg(2) are centroids of the cyclopentadienyl rings

with carbon atoms denoted by lower and higher numbers, respectively,

for each molecule. Dihedral angle between the least-squares planes of

c

the cyclopentadienyl rings.

5484

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

and small deviations of methyl carbon atoms from the least-

squares planes of the cyclopentadienyl rings farther away from

the titanium atom (maximum 0.0692(81) Å). The Ti−Cg

distance of 1.983(2) Å for 10 is uniquely the shortest one

found in the neutral decamethyltitanocene derivative.

Shorter Ti−Cg distances 1.9596(14) and 1.9796(14) Å were

found for [10]⁺[BPh₄]⁻, the former value being for the ligand

with two vicinal methyl groups containing one hydrogen each

in agostic interaction with the titanium atom. Due to this

bonding interaction, the titanocene cation was bent, having a

Cg(1)−Ti−Cg(2) angle of 151.00(6)°.²⁷The Ti−Cg distance

in another titanocene, linear [Ti(C₅Me₄SiMe₃)₂] (2.020(2)

Å),^10ais discernibly longer. The geometric parameters of

[TiCl(C₅Me₅)₂] in 10C (Table 3) are comparable with those

found in single crystal of [TiCl(C₅Me₅)₂], where Ti−Cg =

2.06(1) Å and Ti−Cl = 2.363(1) Å were found.²⁵A remarkable

diﬀerence in the Cg(1)−Ti−Cg(2) angle for the single crystal

of [TiCl(C₅Me₅)₂] (143.6(2)°) and the molecule in 10C

(147.47(10)°) could arise from a suggested admixture of

[TiH(C₅Me₅)₂] (see Experimental Section), which aﬀords an

average of 152.2° for this angle.^22a

Figure 6. PLATON drawing of 17 at the 30% probability level, with

atom-labeling scheme. Hydrogen atoms are omitted for clarity.

Reactions of 1−4 with But-2-yne. Reaction of 1 with but-

2-yne as followed by ¹H NMR is known to aﬀord 2,3-

dimethyltitanacyclopent-2-ene (15) (Scheme 2).^2bHere, both

compounds 1 and 4 smoothly inserted but-2-yne to give 15 and

16, respectively (Scheme 9), which were fully characterized by

spectroscopic methods and the former one by X-ray diﬀraction

single-crystal analysis (Figure 6).

titanacyclopentene methyl groups since the proximal to

titanium methyl group (TiC(Me): δ_H= 1.05 ppm) shows a

nearly equally strong homoallylic interaction (⁵J_HH≈ ⁵J_HH= 1.5

Hz) with the distant methyl group (TiC(Me)CMe: δ_H= 1.47

ppm) and the distant methylene group (TiCH₂CH₂: δ_H= 2.17

ppm). This assignment is also in agreement with a 1D NOESY

experiment showing a strong through-space interaction

between the C₅Me₅protons and the protons of TiC(Me) and

TiCH₂(δ_H= 1.09 ppm) (see Supporting Information). ¹H and

¹³C NMR spectra for the titanacyclopentene moiety of 16

displayed patterns similar to those for 15 (TiCH₂: δ_H= 1.41

and δ_C= 57.18 ppm; TiCH₂CH₂: δ_H= 2.17 ppm and δ_C=

37.20 ppm; TiCC: δ_C= 130.23 ppm; TiC: δ_C= 192.70

ppm). Four nonequivalent methyl groups of tetramethylcyclo-

pentadienyl ligands arose from a lower symmetry of 16 with

respect to 4 (δ_H= 1.41, 1.80, 1.90, 2.05 ppm and δ_C= 12.24,

12.46, 13.26, 13.58 ppm), whereas the methine signals occurred

at δ_H= 4.93 ppm and δ_C= 110.83 ppm. Infrared spectra of 15

and 16 registered the 2,3-dimethyltitanacyclopent-2-ene double

bond as a weak absorption band at 1590 cm⁻¹. EI-MS spectra

for the both compounds displayed low-abundant molecular

ions, which eliminated the whole ethene-but-2-yne adduct to

give the base peaks of corresponding titanocene ions.

Scheme 9

The crystal structure of 15 (Figure 6) is of importance as the

only representative for the strained titanacyclopent-2-ene

moiety. Numerous metallacyclopent-2-ene compounds were

obtained from titanocene^28aand zirconocene−ethene com-

plexes at the rise of their formation with alkynes;^28b,chowever,

none of them were characterized by X-ray single-crystal

analysis. The data listed in Table 4 show a shorter Ti−C

bond length for sp²carbon atom C(24) than for sp³carbon

atom C(21) and the valence angle C(21)−Ti−C(24) smaller

than 90° usually found for free alkyl or halide ligands. The ring

strain is most pronounced in deviations of the valence angles

from the regular 108° for sp³carbon atom C(22): C(21)−

C(22)−C(23) 114.36(16)° and H(22A)−C(22)−H(22B)

99.0°. On the other hand, comparison of pertinent data with

those for titanocene−vinyl compounds [Ti(C₅Me₅)₂(CH

CH₂)X (X = F or OC(CH₂)C₆H₁₁)²⁹on one side and

cyclopentadienyl ring-tethered titanium−propenyl compounds

[(C₅Me₅)Ti(III)CR¹CR²CH₂(η⁵-C₅Me₄)³⁰on the other side

Figure 5. PLATON drawing of 15 at the 30% probability level, with

atom-labeling scheme. Hydrogen atoms are omitted for clarity except

ethene hydrogen atoms H21A, H21B, H22A, and H22B.

1

The H and ¹³C NMR spectra of 15 agree with reported

data.^2bWe only suggest an opposite assignment for the

5485

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

Table 4. Selected Bond Lengths and Bond Angles for 15

Bond Lengths (Å)

a

Ti−Cg(1)

Ti−Pl(1)

2.1394(8)

2.1388(3)

2.2065(18)

2.1374(18)

1.521(3)

Ti−Cg(2)

Ti−Pl(2)

2.1329(8)

2.1325(3)

1.533(3)

1.328(3)

1.519(3)

b

Ti−C(21)

Ti−C(24)

C(23)−C(22)

C(23)−C(26)

C(21)−C(22)

C(24)−C(23)

C(24)−C(25)

1.518(3)

Bond Angles (deg)

a

Cg(1)−Ti−Cg(2)

139.75(3)

107.34(13)

114.36(16)

119.85(19)

109.5

C(21)−Ti−C(24)

Ti−C(24)−C(23)

C(22)−C(23)−C(24)

C(24)−C(23)−C(26)

H(22A)−C(22)−H(22B)

81.63(15)

Ti−C(21)−C(22)

114.63(15)

119.75(18)

125.7(2)

99.0

C(21)−C(22)−C(23)

C(23)−C(24)−C(25)

H(21A)−C(21)−H(21B)

Dihedral Angles (deg)

c

d

φ

41.79(5)

20.44(5)

ψ(1)

21.37(5)

e

ψ(2)

a

b

Cg(1) and Cg(2) are centroids of the C(1−5) and C(11−15) cyclopentadienyl rings, respectively. Pl(1) and Pl(2) are least-squares planes of the

c

d

C(1−5) and C(11−15) cyclopentadienyl rings, respectively. Dihedral angle between Pl(1) and Pl(2). Dihedral angle between Pl(1) and the Ti,

C(21), and C(24) plane. Dihedral angle between Pl(2) and the Ti, C(21), and C(24) plane.

e

shows that geometry at the double bond is suprisingly rigid. In

spite of the apparent strain the titanacyclopentene ring is not

planar, having C(22) declined from the C(21), Ti, C(24) plane

by as much as 0.2952(33) Å toward the C(1−5) cyclo-

pentadienyl ring.

In distinction from the behavior of 1 and 4, compounds 2

and 3 reacted cleanly with but-2-yne by replacing the back-

bonded ethene to give green 17 and blue 18 (Scheme 9).

Similarly to the reported thermally unstable [(η⁵-C₅Me₅)₂Ti-

(η²-MeCCMe)] (19), which was prepared from [{Ti-

1

(C₅Me₅)₂}₂(μ-η¹: η¹-N₂)] and but-2-yne and identiﬁed by H

and ¹³C NMR, and IR spectra,^2b,31compounds 17 and 18 are

unreactive to an excess of but-2-yne. Both compounds were

1

stable in benzene-d₆solution and showed C_2vsymmetry in H,

¹³C, and ²⁹Si NMR spectra. The chemical shifts δ_C(CC) and

infrared ν(CC) vibration for coordinated but-2-yne are close

to those for 19 and other known alkyl- and aryl-substituted

alkyne complexes (see Table 5).^2bThe substitution of ethene

Table 5. ¹H and ¹³C NMR Data (δ, ppm), Infrared ν(CC)

Figure 7. PLATON drawing of 18 at the 30% probability level, with

Vibrations (cm⁻¹), and X-ray d(CC) Distances (Å) for the

atom-labeling scheme. Hydrogen atoms are omitted for clarity.

Coordinated Alkyne in 17−19 and Similar [Cp*₂Ti(η²-

EtCCEt)] (20) and [Cp*₂Ti(η²-PhCCPh)] (21)

but-2-yne is coordinated equally by both carbon atoms (see

a

compound

δ_H

δ_C(CC)

ν(CC)

d(CC)

ref

virtually equal Ti−Cα and Ti−Cβ bond lengths and bending in

but-2-yne); however, one cyclopentadienyl ligand is at a

discernibly longer distance from titanium (see longer Ti−

Cg(2) and Ti−Pl(2) distances). In 18, the diﬀerence in

distances Ti−Cα and Ti−Cβ shows that the but-2-yne ligand is

bonded somewhat asymmetrically, whereas the Ti−Cg

distances are virtually equal. The bond of but-2-yne to the

metal as it follows from the Cα−Cβ distance is negligibly

stronger for 17; however, this is balanced out by a weakly

bonded Cg(2) cyclopentadienyl ligand.

17

18

19

20

21

1.77

1.67

1.74

206.1

206.7

200.1

202.0

200.6

1674

1672

1683

1678

1644

1.3007(19)

1.294(4)

this work

2b, 31

9

1.306(4)

2b, 9

a

Chemical shifts for but-2-yne (C₆D₅Cl −35 °C): δ_H= 1.65 (Me) ppm

and δ_C= 3.4 (Me), 74.7 (CMe) ppm.³²

with but-2-yne is accompanied by replacement of the near-

infrared band of 2 at 940 nm and 3 at 990 nm with the lowest

energy band near 700 nm. Assignments of these absorption

bands will be given in a forthcoming paper.

The crystal structures of 17 (Figure 6) and 18 (Figure 7) and

their common geometric parameters (Table 6) show that steric

congestion between the cyclopentadienyl ligands and the

coordinated but-2-yne is released in diﬀerent ways. In 17, the

CONCLUSIONS

■

Molecular geomerty parameters established from X-ray crystal

structures of 1−4 (Table 1) allow only partly to anticipate the

thermal stabilites of the complexes as observed on heating the

solid compounds under high vacuum and in solution. Heating

of solid compounds 1 and 2 to 135 °C under high vacuum

5486

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

observed for 1 and 4. The rotating cyclopentadienyl ligands

bearing the bulky groups apparently destabilize transient

complexes coordinating both ethene and but-2-yne, allowing

their replacement but not the cycloaddition reaction. Similarly,

a tight closing of the titanium(II) ions in a sandwich of rotating

parallel cyclopentadienyl rings bearing the bulky groups

accounts for a high thermal stability of titanocenes 13 and 14

in solution compared to a highly reactive 10. Although

thermally stable in the solid state, in inert organic solvents it

undergoes fast intramolecular isomerization, 10 ⇔ 10A^6a

followed by a slower disproportionation to give pseudoequili-

brium 1 with K_eq≈ 1 × 10².

Table 6. Selected Bond Lengths and Bond Angles for 17 and

18

17

18

Bond Lengths (Å)

a

Ti−Cg(1)

Ti−Cg(2)

Ti−Pl(1)

2.1407(6)

2.1529(6)

2.1395(2)

2.1501(2)

2.0729(12)

2.0757(13)

1.3007(19)

2.1393(11)

2.1357(11)

2.1390(4)

2.1346(4)

2.063(2)

a

b

Ti−Pl(2)

c

Ti−Cα

Ti−Cβ

c

2.080(2)

c

Cα−Cβ

1.294(4)

Bond Angles (deg)

K_eq

a

Cg(1)−Ti−Cg(2)

Cα−Ti−Cβ

Cα−Cβ−C

Cβ−Cα−C

143.50(2)

36.54(5)

141.38(4)

36.40(10)

133.5(2)

134.9(2)

(1)

10/10A HooI 11 + 12

c

135.16(12)

135.67(13)

The instability of eq 1 lies in a slow conversion of 10A to

10B via hydrogen transfer to 12 to give 11, which is followed

by isomerization of 10B to 5. These reactions were observed at

the high initial concentration of 10/10A; however, the

equilibrium was virtually stable when generated by mixing 11

and 12. Further exploration of the thus generated titanocene

for synthetic purposes is of interest.

With regard to stereoelectronic eﬀects of the varied

substituents (CMe₃, SiMe₃, Me, and H) the present results

conform with those drawn from detailed studies on

composition and stability of titanocene−end-on nitrogen and

−carbonyl complexes, where steric eﬀects were concluded to

play a dominant role.^11bThis conclusion is exempliﬁed by

preparation of bis(1,3-di-tert-butylcyclopentadienyl)titanium

and its ethene and nitrogen complexes,^10balthough the

electron donation eﬀect of the tert-butyl group should not

exceed that of methyl.^11bRecent studies on reactivity of

titanocenes with less-substituted cyclopentadienyl ligands

revealing a side-on coordination of the nitrogen molecule³³

may stimulate investigations on corresponding titanocene−

ethene systems.

Dihedral Angles (deg)

d

φ

41.23(6)

20.79(11)

21.42(10)

1.62(16)

39.09(10)

18.99(18)

20.34(21

1.93(25)

e

ψ1

e

ψ2

f

τ

a

Cg(1) and Cg(2) are centroids of the C(1−5) and C(11−15)

b

cyclopentadienyl rings, respectively. Pl(1) and Pl(2) are least-squares

planes of the C(1−5) and C(11−15) cyclopentadienyl rings,

c

respectively. Cα and Cβ are sp carbon atoms of but-2-yne denoted

d

by lower and higher number, respectively. Dihedral angle between

e

the least-squares planes of cyclopentadienyl rings Pl(1) and Pl(2). ψ1

and ψ2 are dihedral angles between the plane deﬁned by Ti, Cα, and

f

Cβ and planes Pl(1) and Pl(2), respectively. Dihedral angle between

the plane of Ti, Cα, and Cβ and the least-squares planes of carbon

atoms of but-2-yne.

yielded quantitatively titanocenes 10 and 13, whereas

compound 4 sublimed practically undecomposed. Compound

3 eliminated ethene at 80 °C, mimicking the dissociation of the

[Ti(η⁵-C₅Me₄SiMe₃)₂(η²-BTMSE)] complex,^10aand the ther-

molysis of solid titanocene 14 at 140 °C cleanly aﬀorded 7b.¹⁵

In toluene-d₈solution, the thermolysis of 1−3 proceeded at

considerably lower temperatures (80−100 °C) and aﬀorded

diﬀerent products, doubly tucked-in titanocenes (5, 6, and 7)

and ethane. As no deuterium was detected in the evolved

ethane by NMR methods, the intramolecular hydrogen transfer

from methyl groups to dissociating ethene is facilitated by long-

enough lifetimes of transient states in the liquid phase. The

primary formation of symmetrical isomers 6a and 7a followed

by their conversion to asymmetrical 6b and 7b implies the

suggestion that the coordinated ethene in 2 and 3 is activated

by rotating cyclopentadienyl ligands bearing bulky CMe₃and

SiMe₃groups. Passing of these substituents close to a hinge

position likely results in approaching opposite sides of

otherwise methylated cyclopentadienyl ligands, which facilitates

the dissociation of the coordinated ethene. Within the formed

transient state hydrogen atoms of vicinal methyl groups are

transferred to leaving ethene. The exclusive initial formation of

symmetrical isomers is in favor of a concerted process;

however, a stepwise mechanism via an ethyl-tucked-in

titanocene intermediate can also be considered. Compound 4

diﬀered from 1−3 by a higher thermal stability and by partial

formation of dititanium complex 9 in addition to a comparable

amount of 8 (8a < 8b) at 130 °C in m-xylene.

EXPERIMENTAL SECTION

■

General Considerations. All reactions leading to low-valent

titanium compounds and their subsequent reactions were carried out

under high vacuum on a vacuum line in sealed all-glass devices

equipped with breakable seals. The ﬁnal high vacuum of (1−2) × 10⁻⁴

Torr was driven by degassed carbon charcoal cooled by liquid

1

nitrogen. H (300 MHz), ¹³C (75 MHz), and ²⁹Si (59.6 MHz) NMR

spectra were recorded on a Varian Mercury 300 spectrometer in

benzene-d₆or toluene-d₈solutions at 25 °C. Chemical shifts (δ/ppm)

are given relative to the residual solvent signals (δ_H7.15 (C₆D₅H),

2.08 (C₆D₅CD₂H) and δ_C128.0 (C₆D₆), 137.48 (C_ipso)). The δ_Si

values are related to external tetramethylsilane. EI-MS spectra were

obtained on a VG-7070E mass spectrometer at 70 eV. Crystalline

samples in sealed capillaries were opened and inserted into the direct

inlet under argon. The spectra are represented by the peaks of relative

abundance higher than 7% and by important peaks of lower intensity.

Crystals for EI-MS measurements and melting point determinations

were placed in glass capillaries, and KBr pellets were prepared in a

Labmaster 130 glovebox (mBraun) under puriﬁed nitrogen (concen-

trations of oxygen and water were lower than 2.0 ppm) and sealed

with a ﬂame. IR spectra were measured in an air-protecting cuvette on

a Nicolet Avatar FT IR spectrometer in the range 400−4000 cm⁻¹. X-

Band EPR spectra were recorded on an ERS-220 spectrometer

(Center for Production of Scientiﬁc Instruments, Academy of Sciences

of GDR, Berlin, Germany) operated by a CU-1 unit (Magnettech,

Berlin, Germany). g-Values were determined by using an Mn²⁺

standard at g = 1.9860 (M_I= −1/2 line). UV−near IR spectra in

the range 300−2000 nm were measured on a Varian Cary 17D

spectrometer in all-sealed quartz cells (Hellma). Elemental analysis

The steric eﬀect of the bulky CMe₃and SiMe₃ligands is well

discernible in reactions of 2 and 3 with but-2-yne, precluding

the formation of titanacyclopent-2-ene adducts, which is in turn

5487

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

were carried out on a Flash EA1112 CHN/O automatic elemental

analyzer (Thermo Scientiﬁc). Melting points were measured on a

Koﬄer block in sealed glass capillaries under nitrogen and are

uncorrected.

Preparation of 3. Compound [TiCl₂(η⁵-C₅Me₄SiMe₃)₂] (1.0 g,

2.0 mmol) was reduced with Mg (0.050 g, 2.06 mmol) in the presence

of excess ethene, as described for 1. The hexane extract of the products

aﬀorded greenish-yellow crystalline 3 (0.77 g, 83%).

¹H and ¹³C NMR spectra in benzene-d₆, EI-MS, IR, and electronic

Chemicals. The solvents tetrahydrofuran (THF), hexane, and

toluene were dried by reﬂuxing over LiAlH₄and stored as solutions of

green dimeric titanocene [(μ-η⁵:η⁵-C₅H₄C₅H₄)(μ-H)₂{Ti(η⁵-

C₅H₅)}₂].³⁴Benzene-d₆(99.5% D) and toluene-d₈(>99% D) were

degassed, distilled on a vacuum line into a storage ampule containing

dimeric titanocene, and stored as its solution. Titanocene dichlorides

[TiCl₂(η⁵-C₅Me₄t-Bu)₂],³⁵[TiCl₂(η⁵-C₅Me₄SiMe₃)₂],³⁶[TiCl₂(η⁵-

C₅HMe₄)₂]³⁷were prepared as reported, and [TiCl₂(η⁵-C₅Me₅)₂]

was purchased from Aldrich. The singly tucked-in titanocene

[Ti(C₅Me₅)(C₅Me₄CH₂)] was the same as used recently,³⁰and

[TiH(C₅Me₅)₂] was obtained by admitting hydrogen to a Ti(C₅Me₅)-

(C₅Me₄CH₂)] in hexane solution and usual workup.^22aBut-2-yne

(Aldrich) was degassed, puriﬁed by standing with dimeric titanocene,

and distilled into ampules. Ethene (polymerization grade) was

obtained from Polymer Institute Brno (Czech Republic). It was

degassed at liquid nitrogen temperature and dosed by distilling on a

vacuum line. Magnesium turnings (purum for Grignard reaction)

(Aldrich) was used as such.

absorption spectra of 3 agreed with the reported data.^10a

Preparation of 4. [TiCl₂(C₅HMe₄)₂] (0.72 g, 2.0 mmol) and

magnesium (0.05 g, 2.06 mmol) were used to follow the procedure for

obtaining 1. Bright pale green crystals were obtained from hexane

solution. Yield: 0.52 g (83%).

Data for 4 are as follows. Mp: 124 °C. ¹H NMR (benzene-d₆): 1.07

(s, 12H, C₅Me₄H); 2.03 (s, 4H, C₂H₄); 2.44 (s, 12H, C₅Me₄H); 4.10

(s, 2H, C₅Me₄H). ¹³C{¹H} NMR (benzene-d₆): 12.58, 13.41

(C₅Me₄H); 102.66 (C₂H₄); 111.88 (CH, C₅Me₄H); 119.01, 123.87

(C_q, C₅Me₄H). EI-MS (80 °C): m/z (relative abundance) 318 (M^+•;

not observed), 292 (12), 291 (26), 290 ([M − C₂H₄]⁺; 100), 289

(29), 288 (19), 287 (11), 169 ([M − C₂H₄− Cp′]⁺; 10), 167 (8), 166

(7), 164 (7), 105 (10), 69 (28), 28 (24). IR (KBr, cm⁻¹): 3092 (vw),

3034 (m), 2974 (s), 2946 (m), 2906 (s,b), 2860 (m), 2722 (vw), 1599

(vw), 1505 (w), 1484 (w), 1439 (m), 1421 (m), 1382 (s), 1368 (m),

1324 (w), 1171 (vw), 1146 (w), 1113 (vw), 1080 (s), 1026 (m), 974

(vw), 941 (vw), 869 (w), 822 (w), 804 (vs), 699 (vw), 661 (w), 630

(vw), 615 (w), 509 (w), 496 (w), 414 (s). UV−near IR (hexane, nm):

355 (sh) ≫ 920. Anal. Calcd for C₂₀H₃₀Ti: C, 75.46; H, 9.50. Found:

C, 75.41; H, 9.47.

Preparation of 1. To a mixture of degassed [TiCl₂(η⁵-C₅Me₅)₂]

(0.78 g, 2.0 mmol) and magnesium (0.05 g, 2.06 mmol) loaded in a 50

mL ampule on a vacuum line were subsequently distilled THF (20

mL) and ethene (0.5 mL, liquid ∼10 mmol) under cooling with liquid

nitrogen, and the ampule was sealed oﬀ. The content was magnetically

stirred at 40 °C until all the magnesium disappeared. Then, a greenish-

yellow solution was transferred to another ampule, and all volatiles

were distilled oﬀ. The residue was extracted with 50 mL of hexane to

give a green solution. After standing overnight a clear solution was

decanted from traces of white sediment and concentrated to

crystallization at 40 °C. Slow cooling to −28 °C aﬀorded green

crystals. These were separated and dried under vacuum. Yield: 0.30 g

(88%).

Thermolysis of 1 in Toluene to Give 5. Compound 1 (0.14 g,

0.40 mmol) was dissolved in 5 mL of toluene, and the solution heated

to 80 °C for 24 h. A dark solution was evaporated under vacuum,

aﬀording green cubic crystals of 1 and blue prisms of known doubly

tucked-in permethyltitanocene (5).^5aThis mixture of products was

dissolved in toluene and heated to 100 °C for 8 h. A clean blue

solution of 5 was obtained. This was evaporated and crystallized from

a minimum of hexane at −28 °C to give blue prisms of 5. Yield: 0.12 g

(94%).

Data for 5 are as follows. ¹H and ¹³C NMR spectra agree with ref 5a

and UV−vis spectra with ref 16a. EI-MS (60 °C): m/z (relative

abundance) 318 (14), 317 (20), 316 (M^+•; 100), 315 (17), 314 (18),

313 (6), 312 (10), 301 ([M − Me]⁺; 6), 299 ([M − Me − 2 H]⁺; 7),

181 ([M − C₅Me₅]⁺; 7), 180 (8), 179 (7), 178 ([M − C₅Me₅− 3

H]⁺; 12). IR (KBr, cm⁻¹): 3045 (w), 3028 (w), 2965 (s), 2940 (s),

2906 (vs), 2855 (s), 2721 (vw), 1486 (m), 1467 (m), 1445 (m,sh),

1432 (s), 1377 (vs), 1356 (m), 1348 (m), 1171 (vw), 1152 (vw), 1075

(w), 1022 (m), 875 (w), 856 (w), 846 (vw), 829 (s), 808 (m), 754

(vw), 726 (w), 617 (vw), 578 (w), 516 (vw), 442 (m), 430 (w). UV−

vis (hexane, nm): 590.

1

Data for 1 agreed with the reported H and ¹³C NMR and infrared

spectra except the reported very strong band at 1077 cm⁻¹, which we

observed as a weak band. This band grew in intensity when exposing

the KBr pellet to air. Single-crystal X-ray diﬀraction revealed the

reported tetragonal space group P42 m.^2aEI-MS (80 °C): m/z

̅

1

(relative abundance) 346 (M^+•; not observed), 320 (12), 319 (31),

318 ([M − C₂H₄]⁺; 100), 317 (47), 316 (17), 315 (8), 182 (6), 181

(11), 180 (9), 179 (6), 178 (11), 177 (7), 135 (6), 119 (9), 105 (8),

91 (8). The molecular peak of ethene m/z 28 was observed before the

temperature of evaporation of the titanocene was achieved. UV− near

IR (hexane, nm): 360 (sh) ≫ 950.

Thermolysis of 2 to Give 6a and 6b. Compound 2 (0.24 g, 0.56

mmol) was dissolved in toluene (3.0 mL), and this was heated to 95

°C for 8 h to give a blue solution. This was evaporated, and the residue

dissolved in 1.5 mL of hexane. The product(s) did not crystallize

under any condition. The solution was evaporated under vacuum to

give a blue oily solid. This was dissolved in toluene-d₈(2.0 mL) to

Preparation of 2. The procedure for obtaining 1 was followed

using [TiCl₂(η⁵-C₅Me₄t-Bu)₂] (0.472 g, 1.0 mmol) and magnesium

(0.025 g, 1.03 mmol). Pale green crystals were separated and dried

under vacuum. Yield: 0.37 g (86%).

Data for 2 are as follows. Mp: 131 °C (138−140 °C bubbling liquid,

1

upon cooling a bright blue color). H NMR (benzene-d₆): 0.66 (s,

1

measure H and ¹³C NMR and UV−vis spectra. The NMR analysis

18H, CMe₃); 1.49 (s, 12H, C₅Me₄); 2.08 (s, 4H, CH₂); 2.43 (s,

12H, C₅Me₄). ¹³C{¹H} NMR (benzene-d₆): 12.34, 16.02 (C₅Me₄);

31.27 (CMe₃); 35.82 (CMe₃); 101.80 (CH₂); 120.45, 121.42

(C₅Me₄); 129.49 (C_ipso, C₅Me₄). 1D NOESY (benzene-d₆): ¹H_irr/¹H_res

found the sample consisted of symmetrical isomer 6a (75%) and

asymmetrical isomer 6b (25%). Then, the solution was heated to 100

°C for 7 h to give pure 6b. This was transferred to hexane and

crystallized by slow solvent evaporation at −5 °C. Yield of blue waxy

solid: 0.16 g (72%).

1

0.66/1.49; 2.08; 2.43 (CMe₃/C₅Me₄; CH₂; C₅Me₄). H_irr/¹H_res

2.08/0.66; 1.49; 2.43 (CH₂/CMe₃; C₅Me₄; C₅Me₄); interaction

with the resonance 2.43 is twice as strong as that with 1.49. EI-MS

(120 °C): m/z (relative abundance) 430 (M^+•; not observed), 403

(11), 402 ([M − C₂H₄]⁺; 27), 344 (11), 343 ([M − C₂H₄− C₄H₉−

H₂]⁺; 36), 342 (11), 341 ([M − C₂H₄− C₄H₉− 2 H₂]⁺; 19), 340

(13), 339 ([M − C₂H₄− C₄H₉− 3 H₂]⁺; 35), 338 (13), 337 ([M −

C₂H₄− C₄H₉− 4 H₂]⁺; 23), 327 (13), 206 (21), 202 (18), 201 (23),

167 (15), 57 ([C₄H₉]⁺; 20), 41 (29), 28 ([C₂H₄]⁺; 100), 27 (46), 26

(39). IR (KBr, cm⁻¹): 3051 (w), 2991 (s), 2957 (vs), 2901 (s), 1472

(m,b), 1446 (w), 1429 (w), 1404 (w), 1378 (s), 1359 (s), 1235 (m),

1198 (w), 1123 (w), 1088 (s), 1025 (m), 863 (w), 669 (w), 651 (w),

479 (w). UV−near IR (hexane, nm): 365 (sh) ≫ 940. Anal. Calcd for

C₂₈H₄₆Ti: C, 78.11; H, 10.77. Found: C, 78.04; H, 10.72.

1

Data for 6a are as follows. H NMR (benzene-d₆): 1.08 (s, 6H,

C₅Me₂); 1.10 (s, 6H, C₅Me₄); 1.30 (s, 18H, CMe₃); 1.49 (d, ²J_HH= 5.4

Hz, 1H, CH₂); 1.70 (s, 6H, C₅Me₄); 1.91 (d, ²J_HH= 5.4 Hz, 1H, 

CH₂). ¹³C{¹H} NMR (C₆D₆, 75 Hz): 10.19, 10.31, 14.14 (C₅Me₂and

C₅Me₄); 32.68 (CMe₃); 36.72 (CMe₃); 72.88 (CH₂); 120.99, 122.24

(C₅Me₄); 125.13, 128.50 (C₅Me₂); 140.22 (C_ipso). ¹H NMR (toluene-

d₈, 95 °C): 1.07 (s, 6H, C₅Me₂); 1.17 (s, 6H, C₅Me₄); 1.24 (s, 18H,

CMe₃); 1.40 (d, ²J_HH= 5.6 Hz, 1H, CH₂); 1.74 (s, 6H, C₅Me₄); 1.80

2

(d, J_HH= 5.6 Hz, 1H, CH₂). ¹³C{¹H} NMR (benzene-d₆): 10.19,

10.31, 14.14 (C₅Me₂and C₅Me₄); 32.68 (CMe₃); 36.72 (CMe₃); 72.88

(CH₂); 120.99, 122.24 (C₅Me₄); 125.13, 128.50 (C₅Me₂); 140.22

(C_ipso).

5488

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

Data for 6b are as follows. ¹H NMR (toluene-d₈): 0.70, 0.96 (2 × d,

²J_HH= 4.0 Hz, 2 × 1H, CH₂); 1.04 (s, 3H, C₅Me₂); 1.06 partially

Data for 8a and 8b are as follows. Mp: 102 °C. EI-MS (80 °C): m/z

(relative abundance) 290 (11), 289 (27), 288 (M^+•; 100), 287 (18),

286 (20), 285 (7), 284 (12), 273 ([M − Me]⁺; 7), 271 ([M − Me − 2

H]⁺; 10), 269 (7), 167 ([M − C₅Me₄H]⁺; 11), 166 (14), 165 (11),

164 ([M − C₅Me₄H − 3 H]⁺; 18), 163 (12), 91 (9), 87 (8), 48 (9), 41

(10). IR (KBr, cm⁻¹): 3069 (w), 3041 (m), 2966 (s), 2941 (s), 2909

(vs), 2858 (s), 2722 (vw), 1495 (m), 1473 (m), 1430 (s), 1379 (s),

1362 (m), 1341 (m), 1173 (w), 1150 (vw), 1121 (vw), 1091 (w),

1028 (m), 976 (w), 876 (m), 831 (vs), 815 (s), 802 (m), 738 (s), 694

(w), 663 (vw), 619 (vw), 609 (w), 580 (m), 566 (w), 507 (vw), 445

(s). UV−vis spectrum (hexane, nm): 355 (sh) ≫ 570.

The brown residue after the hexane extraction was extracted with

warm toluene (20 mL) to form a greenish-brown solution. A 15 mL

amount of the solution was evaporated, and a dark residue aﬀorded at

290 °C EI-MS spectra that showed the pattern of peaks (M^+•m/z 578

and [M/2]⁺m/z 289) corresponding to the dimer [Ti(C₅HMe₄)(μ-

C₅Me₄)]₂(9).¹⁷Furthermore, the remainder of the solution displayed

the UV−vis spectrum (m-xylene, nm) 465 (sh) > 580 (sh) > 820,

typical of puriﬁed 9, and the EPR spectrum of the toluene glass gave

the same parameters for an electronic triplet state due to two coupled

Ti^IIIatoms as the EPR spectrum of the authentic sample.¹⁸

Thermolysis of 1 under High Vacuum to Give 10 and

Reactions of 10 in Solution. Solid compound 1 (0.16 g, 0.46 mmol)

loaded in a three-necked glass ampule was attached to a vacuum line

via a breakable seal (bent tip), and the ampule bottom was immersed

in an oil bath. The breakable seal was fully opened under dynamic

vacuum generated by a charcoal cooled with liquid nitrogen (up to 1 ×

10⁻⁴Torr), and the bath was gradually heated to 135 °C. All

compound 1 evaporated to form a brown solid condensed in the upper

part of the air-cooled ampule and a minor gray-green layer condensed

in the lower part. The brown product was washed down with a

condensing hexane vapor into an arm equipped with a breakable seal,

while the initial yellowish-brown color of the solution turned green.

The minor gray-green layer was dissolved in toluene-d₈and proved by

¹H and ¹³C NMR spectra to be compound 1. The hexane solution was

2

overlapped (d, J_HH= 5.1 Hz, 2 × 1H, CH₂); 1.07 (s, 9H, CMe₃);

1.24 (d, ²J_HH= 5.1 Hz, 2 × 1H, CH₂); 1.39 (s, 9H, CMe₃); 1.42 (s,

3H, C₅Me₄); 1.57 (s, 3H, C₅Me₂); 1.63, 2.01, 2.03 (3 × s, 3 × 3H,

C₅Me₄). ¹³C{¹H} NMR (toluene-d₈): 10.65, 11.47, 12.20, 14.25, 16.42,

16.73 (C₅Me₂and C₅Me₄); 32.61, 32.92 (CMe₃); 34.44, 36.09 (CMe₃);

70.11, 71.66 (CH₂); 118.48, 119.75, 120.31, 120.36 (CMe, C₅Me₄);

123.86, 133.95 (CMe, C₅Me₂); 134.44 (CCMe₃, C₅Me₄); 136.34

(CCMe₃, C₅Me₂); 140.28, 144.97 (CCH₂, C₅Me₂). EI-MS (70 °C):

m/z (relative abundance) 403 (9), 402 (23), 401 (59), 400 (M^+•;

100), 399 (31), 398 (25), 384 ([M − CH₄]⁺; 10), 381 (8), 345 (10),

344 (9), 343 ([M − Bu]⁺; 20), 342 (8), 341 (7), 340 (10), 339 (23),

338 (11), 337 (24), 336 (7), 328 ([M − Bu − Me]⁺; 15), 327 ([M −

Bu − CH₄]⁺; 16), 325 ([M − Bu − CH₄− 2 H]⁺; 25), 324 (7), 321

(8), 303 (14), 301 (17), 246 (20), 244 (9), 241 (9), 208 (19), 207 (7),

206 (10), 205 (8), 204 (9), 202 (15), 201 (15), 200 (7), 192 (9), 191

(7), 190 (7), 188 (7), 187 (8), 181 (7), 180 (8), 167 (8), 164 (7), 121

(8), 107 (12), 91 (15), 79 (8), 57 ([Bu]⁺; 29). IR (KBr, cm⁻¹): 3071

(vw), 3050 (vw), 2976 (s,sh), 2956 (vs), 2908 (s), 2868 (m), 1480

(m), 1453 (s,b), 1379 (s), 1359 (s), 1326 (w), 1235 (s), 1196 (w),

1133 (vw), 1086 (vw), 1037 (m), 1020 (m), 874 (m), 819 (vw), 773

(vw), 758 (vw), 616 (w), 573 (w). UV−vis (toluene, nm): 605.

Thermolysis of 2 in an NMR Probe. Compound 2 (0.08 g, 0.19

mmol) was dissolved in toluene-d₈(0.8 mL), and the solution was

transferred to an NMR tube, which was ﬂame-sealed. The tube was

heated in an NMR probe to 95 °C, and ¹H NMR spectra were

measured at the same temperature repeatedly for a total of 186 min.

The formation of 6a and elimination of ethane to 50% conversion of 2

were monitored (see Supporting Information). Subsequent heating of

the tube to 100 °C for 7 h gave the above ¹H and ¹³C NMR spectra of

6b.

Thermolysis of 3 to Give 7a and 7b. Compound 3 (0.12 g, 0.26

mmol) in hexane (5.0 mL) was heated to 90 °C for 2 h, turning from

yellow to blue. The solution was evaporated under vacuum, the residue

was dissolved in benzene-d₆(1.2 mL), and its electronic absorption

spectrum was measured. Then, the solution was transferred to an

NMR tube, and this was ﬂame-sealed. ¹H and ¹³C NMR spectra

revealed the presence of 7a (60%) and 3 (40%), which allowed

identifying resonances of 7a. Heating the tube to 90 °C for 3 h gave a

sky blue solution containing nearly equal amounts of 7a and 7b.

Intense signals at δ_H0.80 and δ_C6.96 indicated the presence of

ethane.¹²Further heating to 90 °C for a total of 13 h accomplished the

conversion of 7a to 7b (see Supporting Information). The NMR tube

was opened to vacuum, and the product transferred into hexane and

crystallized out. The crystalline blue product showed virtually the same

¹H and ¹³C NMR, EI-MS, UV−vis, and IR spectra as 7b obtained by

thermolysis of solid titanocene with elimination of hydrogen.¹⁵Yield:

0.10 g (89%).

evaporated under vacuum, and the residue was dissolved in toluene-d₈

1

for H NMR investigation of processes induced by aging at room

temperature. The ﬁrst measurement after 3 h at room temperature

revealed the major presence of paramagnetic 10 and its diamagnetic

singly tucked-in titanocene hydride isomer [TiH{C₅Me₄(CH₂)}-

(C₅Me₅)] (10A)^6ain approximately equal integrated intensity and

minor products of their aging. The next measurement after one week

at room temperature (∼20 °C) revealed only 0.5% intensity of 10 and

10A each, whereas the products of aging dominated. These were

[TiH(C₅Me₅)₂] (11)²¹and [Ti{C₅Me₄(CH₂)}(C₅Me₅)] (12),^22,27

1

both displaying partially overlapped paramagnetic H NMR spectra,

and high-resolution measurement revealed two sets of resonances. The

major one was assigned to the new compound [Ti{C₅Me₄(CH₂)}₂)]

(10B) and the minor one to 5. In addition, a broad single resonance at

δ 4.3 ppm, Δν_1/2≈ 270 Hz, remained unassigned. During the next 14

weeks the signal of 10 diminished to about 0.3% of the initial intensity,

the signal at δ 4.3 ppm did not change, and compound 10B very

slowly converted to 5. The latter conversion was accomplished by

warming to 80 °C for 3 h.

Data for 7a are as follows. ¹H NMR (benzene-d₆, 300 Hz): 0.22 (s,

18H, SiMe₃); 1.02 (s, 6H, C₅Me₂); 1.05, 1.68 (2 × s, 2 × 6H, C₅Me₄);

1.61, 1.71 (2 × d, 2 × ²J_HH= 4.8 Hz, 2 × 1H, CH₂). ¹³C{¹H} NMR

(benzene-d₆, 75 Hz): 1.87 (SiMe₃); 9.93, 10.00, 13.93 (C₅Me₂and

C₅Me₄); 72.01 (CH₂); 125.38 (C₅Me₄); 125.60 (C_ipso); 129.24

(C₅Me₂); 132.02 (C₅Me₄); 137.20 (C₅Me₂). ²⁹Si{¹H} NMR (benzene-

d₆): −6.9 (SiMe₃). UV−vis (benzene, nm): 605.

Data for 10 are as follows. ¹H NMR (toluene-d₈): 63.8 (br s, Δν_1/2

≈ 170 Hz, 30H, C₅Me₅).

Data for 10A are as follows. ¹H NMR (toluene-d₈): 1.02 (d, ²J_HH

=

Data for 7b agree with those reported.¹⁵

4.5 Hz, 1H, CH₂); 1.07 (s, 3H, C₅Me₄); 1.35 (s, 3H, C₅Me₄); 1.55

partially overlapped by signal of 10B (d, ²J_HH= 4.5 Hz, 1H, CH₂); 1.62

(s, 3H, C₅Me₄); 1.83 (s, 15H, C₅Me₅); 2.19 (d, J_HH= 1.5 Hz, 3H,

C₅Me₄). ¹³C{¹H} NMR (toluene-d₈): 10.18, 10.31, 17.81, the fourth

signal was not found (C₅Me₄); 12.55 (C₅Me₅); 76.71 (CH₂); 119.42

(C₅Me₅); 120.98, 122.48, 131.35, 131.61, the ﬁfth signal was not found

(C₅Me₄).

Thermolysis of 4. Compound 4 (0.18 g, 0.56 mmol) was

dissolved in 5 mL of m-xylene, and the solution in a 10 mL ampule

attached to a quartz cuvette (d = 0.2 cm) was heated gradually to 130

°C, at which temperature the absorption band of 4 at 925 nm began to

decrease in intensity. It disappeared after heating for 8 h, whereas

some dark solid precipitated from a dark solution. The m-xylene was

evaporated under vacuum, and a brown residue was extracted with

condensing hexane vapors. A blue hexane extract was evaporated to

1

Data for 10B are as follows. H NMR (toluene-d₈): 1.09 (s, 6H,

C(5)Me); 1.18 (s, 6H, C(4)Me); 1.20, 1.31 (2 × d, ²J_HH= 5.1 Hz, 2 ×

2H, CH₂); 1.57 (s, 6H, C(3)Me); 1.68 (s, 6H, C(2)Me). ¹³C{¹H}

NMR (toluene-d₈): 9.83 (C(4)Me); 10.12 (C(5)Me); 11.10 (C(3)

Me); 14.82 (C(2)Me) 69.24 (CH₂); 119.98 (C(4)); 122.20 (C(3));

123.30 (C(5)); 127.44 (C(2)); 133.44 (C(1)). gCOSY (toluene-d₈):

1

give a blue polycrystalline solid. This was identiﬁed from H and ¹³C

NMR spectra in benzene-d₆to be a mixture of 8a (12%) and 8b

(88%), a similar composition to that previously obtained.¹⁶Yield: 83

mg (48%).

5489

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

¹H/¹H 1.20/1.31 (CH₂). gHSQC (toluene-d₈): ¹H/¹³C 1.09/10.12

(C(5)Me); 1.18/9.83 (C(4)Me); 1.20, 1.31/69.24 (CH₂); 1.57/11.10

(C(3)Me); 1.68/14.82 (C(2)Me). For the carbon atom numbering see

Figure 3.

temperature. Conversion of 10/10A to paramagnetic 11 and 12 and

diamagnetic 5 ran similarly to that for pure 10/10A. After 1 week, the

initially measured signal of 10 decreased to about 1% intensity,

whereas the signal of [TiCl(C₅Me₅)₂] remained constant (see

Supporting Information).

Thermolysis of 1 to Give 12. The thermolysis as described above

was modiﬁed by opening the breakable tip in its upper part so that the

hole diameter was about 0.3 mm. The entire sample of 1 thermolyzed

at 110 °C to give a purple sublimate. This was dissolved in toluene-d₈

and proved by ¹H NMR spectra to be compound 12. Gaseous

products of the thermolysis were evolved by warming of the charcoal

on a vacuum line and were condensed to an ampule cooled with liquid

nitrogen for 3 h. Then, the gaseous products were dissolved in 1.0 mL

of toluene-d₈and analyzed by ¹H and ¹³C NMR spectroscopy.

Resonances of ethane and ethene¹²corresponded to an equimolar

ratio of these products.

Thermolysis of 10 to Give 12. A freshly prepared hexane

solution of 10/10A was evaporated in the ampule for thermolysis to

give a brown residue. This was then thermolyzed as above, with the tip

broken in its narrow upper part. The whole sample sublimed at 110

°C, giving a purple sublimate. This was dissolved in 1.0 mL of toluene-

d₈and analyzed by ¹H NMR spectra to be compound 12. The

hydrogen trapped in the charcoal patron was not assessed.

Preparation of 11. Singly tucked-in titanocene 12 (0.24 g, 0.69

mmol) in hexane (10 mL) was reacted with gaseous hydrogen (50 mL,

750 Torr). The initial purple solution turned orange-red after 30 min.

The solution was degassed, and its volume reduced to 2.0 mL. Cooling

to −28 °C aﬀorded orange-red crystals. Yield: 0.19 g (86%).

Data for 11 are as follows. Mp: 150 °C, dec. ¹H NMR (toluene-d₈):

23.0 (br s, Δν_1/2≈ 170 Hz, 30H, C₅Me₅) agrees with reported data.^22a

EI-MS (70 °C): m/z (relative abundance) 319 (M^+•; 10), 318 (49),

317 ([M − H₂]⁺; 100), 316 (26), 315 (38), 314 (8), 313 (20), 312

(7), 311 (14), 182 (8), 181 (14), 180 (7), 179 (12), 178 (14), 177

(16), 135 (6), 119 (10), 105 (13), 91 (14). IR (KBr, cm⁻¹): 2968

(m,sh), 2940 (m,sh), 2904 (vs), 2856 (s), 2721 (vw), 1509 (s), 1489

(m), 1435 (m,b), 1376 (vs), 1158 (vw), 1065 (vw), 1023 (m), 801

(vw), 571 (vw), 463 (m), 435 (w).

Equilibrium of 10 with 11 and 12. Paramagnetic ¹H NMR

spectra of toluene-d₈solutions of 11 and 12³⁰(0.03 mmol in 1.0 mL

each) were measured to prove the absence of 10. Then the sample

tubes were broken under high vacuum, and the solutions were mixed

and ﬁlled into an NMR tube, which was ﬂame-sealed. The ﬁrst

measurement of the ¹H NMR spectrum after 3 h at room temperature

showed the resonance of 10 of very similar intensity to that obtained

after aging of 10 for one week (see Supporting Information). This

signal decreased only negligibly during the next 4 months.

Preparation of Cocrystals of 10 and [TiCl(C₅Me₅)₂] (10C). (A)

Thermolysis of 1 containing trace amounts of [TiCl(C₅Me₅)₂] was

performed at a slightly lower pumping speed (controlled by a valve) in

order to completely dissociate 1. All the sublimed solid was dissolved

with cool condensing hexane vapor, the ﬁnal drops of the extract

showing a blue coloration of the chloride. The solution in one arm of a

three-necked ampule was put into a refrigerator (−5 °C) for

crystallization driven by slow distillation of the solvent to another

slightly cooler arm. After 2 days several needle crystals grew from the

solution. The mother liquor was decanted to the arm containing the

solvent, the crystals were dried by cooling the mother-liquor arm, and

the arm with crystals was sealed oﬀ. The composition of crystals was

established by X-ray analysis of single crystals (see below). (B)

Compound 1 (0.16 g, 0.46 mmol) and [TiCl(C₅Me₅)₂] (0.24 g, 0.68

mmol) were jointly thermolyzed/sublimed at a smaller pumping

speed, as for (A). From the resulting slightly poorer vacuum all 1

thermolyzed to 10, forming an upper brown layer, and the chloride

sublimed to give a lower blue layer of sublimates. All the sublimates

were dissolved in hexane, and the concentrated solution was

crystallized as for (A). Two crops of gray-green needle crystals of

the same appearance were collected; both aﬀorded identical IR spectra

and crystal unit cell. Yield: 0.23 g.

Data for 10C are as follows. Mp: 170 °C dec. EI-MS (90 °C,

evaporation maximum for 10): m/z (relative abundance) [TiCl-

(C₅Me₅)₂]: 353 (M^+•; 50), 218 ([M − C₅Me₅]⁺; 17), 217 ([M −

HC₅Me₅]⁺; 16); [Ti(C₅Me₅)₂]: 318 (M^+•; 100), 317 (67), 316 (44),

315 (17), 181 ([M − C₅Me₅− 2 H]⁺; 14), 180 (13), 179 (8), 178

([M − C₅Me₅− 5 H]⁺; 17), 177 ([M − C₅Me₅− 6 H]⁺; 17), and

common peaks for the both species 135 ([C₅Me₅]⁺; 9), 119 (18), 105

(17), 91 (15). IR (KBr, cm⁻¹): 2968 (m,sh), 2903 (vs), 2856 (s), 2721

(vw), 2080 (s), 2049 (w), 2016 (m), 1964 (s), 1714 (vw), 1488 (m),

1435 (m,b), 1378 (vs), 1164 (vw), 1077 (vw), 1065 (vw), 1023 (s),

803 (vw), 621 (vw), 570 (vw), 438 (s). Upon grinding with KBr under

N₂in a glovebox the initial gray-green color turned blue. The strong

infrared bands at 2080 and 1964 cm⁻¹are close to the bands at 2090

and 1982 cm⁻¹assigned by Bercaw to [Ti(C₅Me₅)₂(N₂)₂].^6aThe

presence of a very weak band at 1714 cm⁻¹together with weak bands

at 2049 and 2016 cm⁻¹indicate an admixture of [{Ti-

(C₅Me₅)₂(N₂)}₂(μ-η¹:η¹-N₂)],^6c,11band the band at 438 cm⁻¹belongs

to [TiCl(C₅Me₅)₂].

Thermolysis of 2 in High Vacuum to Give Titanocene 13. A

hexane solution of 2 (0.15 g, 0.35 mmol) in an ampule sealed to a

vacuum line via a breakable seal was evaporated, and the residue was

heated gradually to 135 °C under dynamic vacuum, as described above

for thermolysis of 1. A brown solid condensed on the cold ampule

walls. The solid was transferred with condensing hexane vapor to an

attached ampule and crystallized from the concentrated solution by

cooling to −28 °C. Brown needle aggregates were dried under

vacuum. Yield: 0.13 g (93%). The thermolytic residue was negligible.

1

Data for 13 are as follows. Mp: 98 °C. H and ¹³C NMR spectra

agree with reported data.^11cEI-MS (direct inlet, 80 °C): m/z relative

abundance) 404 (13), 403 (36), 402 (M^+•; 100), 401 (18), 400 (17),

345 ([M − C₄H₉]⁺; 6), 343 ([M − C₄H₉− H₂]⁺; 8), 341 (7), 340 (7),

339 ([M − C₄H₉− 3 H₂]⁺; 22), 337 (7), 328 (6), 327 ([M − C₄H₉−

H₂− CH₄]⁺; 18), 325 (6), 313 (7), 303 (7), 285 (7), 206 (9), 201

(9). IR (KBr, cm⁻¹): 2954 (vs), 2906 (vs), 2871 (s), 2725 (vw), 1462

(m,b), 1382 (s), 1359 (s), 1234 (m), 1199 (w), 1034 (m), 1022 (m),

667 (vw), 620 (vw), 571 (vw), 487 (w), 435 (m). UV−vis (hexane,

nm): 345 > 382 (sh) ≫ 552.

Elimination of Ethene from 3. A solution of 3 (0.20 g, 0.4

mmol) in hexane was evaporated under vacuum, and the residue was

heated to 80 °C for 30 min while the vacuum was driven by charcoal

cooled with liquid nitrogen. After cooling to room temperature

crystalline 14 remained at the bottom. Its paramagnetic ¹H NMR

spectrum in benzene-d₆was identical with the reported one.^10a

Sublimation of 4. Compound 4 (0.08 g, 0.25 mmol) was heated

to 135 °C under high vacuum as described above for obtaining 10 or

13. All the solid sublimed during 40 min to give a green sublimate

layer, while the upper part of the ampule showed a brownish

coloration. Condensing hexane vapor easily dissolved the brown

product, which after evaporation left an oily spot. The green sublimate

was dissolved in toluene-d₈and identiﬁed by ¹H and ¹³C NMR spectra

to be compound 4.

Preparation of 15. Crystalline 1 (0.21 g, 0.61 mmol) was

dissolved in warm hexane, but-2-yne (1.0 mL, 12.8 mmol) was added,

and the solution was stirred at 60 °C for 3 h, turning color to red-

brown. All volatiles were evaporated under vacuum, and a brown

residue was dissolved in minimum hexane. An orange-red solid

crystallized from a brown mother liquor, which was decanted. The

solid was recrystallized from hexane to give pink-red crystals of 15.

Yield: 0.19 g (79%).

Data for 15 are as follows. Mp: 173 °C dec. ¹H NMR (benzene-d₆):

5

3

1.05 (q, J_HH= 1.5 Hz, 3H, TiC(Me)); 1.09 (t, J_HH= 7.5 Hz, 2H,

5

TiCH₂); 1.47 (q, J_HH= 1.5 Hz, 3H, TiC(Me)C(Me)); 1.79 (s,

Aging of 10C in Solution. Crystalline 10C (0.06 g) was dissolved

in 1.0 mL of toluene-d₈, and the solution was ﬂame-sealed in an NMR

tube. The ﬁrst measurement was performed after 3 h at room

30H, C₅Me₅); 2.17 (t, ³J_HH= 7.5 Hz, 2H, TiCH₂CH₂). ¹³C{¹H} NMR

(benzene-d₆): 12.2 (C₅Me₅); 18.8, 21.6 (TiC(Me)C(Me)); 38.5

(TiCH₂CH₂); 57.2 (TiCH₂); 121.2 (C₅Me₅); 129.7 (TiC(Me)

5490

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

1

C(Me)); 194.3 (TiC(Me)). 1D NOESY (benzene-d₆): H_irr/¹H_res

1.79/1.05; 1.09 (C₅Me₅/TiC(Me); TiCH₂). EI-MS (130 °C): m/z

(relative abundance) 400 (M^+•; 6), 398 ([M − H₂]⁺; 6), 320 (12), 319

(30), 318 ([M − C₆H₁₀]⁺; 100), 317 (20), 316 (15), 264 ([M −

C₅Me₅H]⁺; 3), 262 ([M − H₂− C₅Me₅H]⁺; 4), 237 ([M − C₅Me₅−

C₂H₄]⁺; 16), 236 (10), 235 ([M − H₂− C₅Me₅− C₂H₄]⁺; 28), 233

(8), 183 (8), 182 (18), 181 ([TiC₅Me₃(CH₂)₂]⁺; 49), 180 (17), 179

(14), 178 (15), 177 (10), 136 ([C₅Me₅H]⁺; 29), 135 ([C₅Me₅]⁺; 17),

121 (40), 119 (16), 105 (19), 93 (11), 91 (13), 69 (21), 57 (11), 55

(14). IR (KBr, cm⁻¹): 2988 (s), 2951 (s), 2901 (vs), 2842 (s), 2789

(m), 2720 (vw), 1589 (w) (ν(CC)), 1494 (m), 1434 (s,b), 1379

(s), 1229 (vw), 1064 (vw), 1022 (m), 808 (vw), 645 (vw), 594 (vw),

491 (vw).

hexane, and but-2-yne (1.0 mL, 12.8 mmol) was added. The solution

turned immediately pale blue, and this color remained unchanged for

one week. Then, all volatiles were evaporated under vacuum, and a

blue residue was dissolved in hexane. Crystallization by slow solvent

distillation in a refrigerator aﬀorded blue crystals.

Yield: 0.20 g (76%).

Data for 18 are as follows. Mp: 68 °C. ¹H NMR (benzene-d₆): 0.13

(s, 18H, SiMe₃); 1.61 (s, 12H, C₅Me₄, β-Me); 1.67 (s, 6H, CMe);

2.12 (s, 12H, C₅Me₄, α-Me). ¹³C{¹H} NMR (benzene-d₆): 2.46

(SiMe₃); 12.06, 16.01 (C₅Me₄); 19.90 (CMe); 117.87 (C_ipso);

124.29, 131.82 (C₅Me₄); 206.73 (CMe). ²⁹Si{¹H} NMR (benzene-

d₆): −8.82 (SiMe₃). EI-MS (140 °C): m/z (relative abundance) 489

(8), 488 (M^+•; 14), 437 (10), 436 (32), 435 (48), 434 ([M − C₄H₆]⁺;

100), 433 (50), 432 (43), 431 (14), 418 (7), 417 (8), 416 ([M −

CH₂SiMe₂]⁺; 10), 362 (6), 361 ([M − C₄H₆− SiMe₃]⁺; 13), 360 (7),

359 (6), 287 (8), 208 (6), 207 (5), 167 (6), 73 ([SiMe₃]⁺; 62), 59

([SiMe₂H]⁺; 8), 55 (5), 54 ([C₄H₆]⁺; 17), 53 (9), 45 (8), 41 (15). IR

(KBr, cm⁻¹): 2953 (s), 2898 (vs), 2831 (m), 2722 (vw), 1672 (w),

1488 (w), 1442 (m,b), 1404 (w), 1377 (m), 1350 (w), 1327 (m),

1246 (s), 1129 (w), 1102 (vw), 1086 (vw), 1021 (m), 951 (vw), 837

(vs,b), 755 (s), 685 (m), 634 (m), 591 (w), 574 (vw), 417 (m). UV−

vis (hexane, nm): 320 (sh) > 390 (sh) ≫ 560 < 710. Anal. Calcd for

C₂₈H₄₈Si₂Ti: C, 68.81; H, 9.90. Found: C, 68.86; H, 9.95.

Reaction of 4 with But-2-yne to Give 16. A solution of 4 (0.13

g, 0.41 mmol) in hexane (25 mL) was transferred to an ampule with

attached quartz cells, and after collecting the absorption spectrum in

the range 300−1600 nm, but-2-yne (0.5 mL, 6.4 mmol) was added.

After 2 h at room temperature (∼22 °C) the absorption spectrum of 4

disappeared, being replaced by that of 16. All volatiles were distilled oﬀ

on a vacuum line, and a brownish-red residue was dissolved in 2 mL of

hexane. Crystallization at −28 °C aﬀorded a red crystalline solid. This

was separated from the ocher mother liquor and recrystallized from

hexane to give orange-red crystals. Yield: 0.10 g (67%).

1

X-ray Crystallography. Single crystals or crystal fragments of 2, 4,

10C, 15, 17, and 18 were mounted into Lindemann glass capillaries in

a Labmaster 130 glovebox (mBraun) under puriﬁed nitrogen.

Diﬀraction data for all complexes were collected on a Nonius

KappaCCD diﬀractometer (Mo Kα radiation, λ = 0.71073 Å) and

processed by the HKL program package.³⁸The phase problem was

solved by direct methods (SIR97),³⁹followed by consecutive Fourier

syntheses, and reﬁned by full-matrix least-squares on F²(SHELXL-

97).⁴⁰Relevant crystallographic data are gathered in the Supporting

Information. All non-hydrogen atoms were reﬁned anisotropically. A

partial lack of electron density found for chlorine atoms in 10C was

solved by adding 33% hydrogen atoms, improving signiﬁcantly the

residual factors (wR from 0.241 to 0.198). This appeared to be

Data for 16 are as follows. Mp: 140 °C. H NMR (benzene-d₆):

1.02 (t, ³J_HH= 7.5 Hz, 2H, TiCH₂); 1.16 (pseudo sextet, ⁵J_HH≈ ⁵J_HH

=

0.9 Hz, 3H, TiC(Me)); 1.41 (s, 6H, HC₅Me₄); 1.48 (q, ⁵J_HH= 0.9 Hz,

3H, TiC(Me)C(Me)); 1.80, 1.90, 2.05 (3 × s, 3 × 6H, HC₅Me₄);

2.17 (tq, ³J_HH= 7.5 Hz, ⁵J_HH= 0.9 Hz, 2H, TiCH₂CH₂); 4.93 (s, 2H,

HC₅Me₄). ¹³C{¹H} NMR (benzene-d₆): 12.24, 12.46, 13.26, 13.58

(HC₅Me₄); 18.88 (TiC(Me)C(Me)); 22.00 (TiC(Me)C(Me));

37.20 (TiCH₂CH₂); 57.18 (TiCH₂); 110.83 (CH, HC₅Me₄); 118.66,

119.57, 124.43, 126.27 (C_q, HC₅Me₄); 130.23 (TiC(Me)C(Me));

192.70 (TiC(Me)). EI-MS (120 °C): m/z (relative abundance) 372

(M^+•; 4), 342 ([M − C₂H₆]⁺; 6), 292 (13), 291 (36), 290 ([M −

C₆H₁₀]⁺; 100), 289 (24), 288 (20), 168 (8), 167 ([Cp′Ti − 2H]⁺; 19),

166 (14), 164 (12), 105 (9). IR (KBr, cm⁻¹): 3093 (w) (ν(C−H)

Cp′), 2973 (m), 2955 (m), 2916 (vs), 2858 (s,b), 2832 (s), 2791 (m),

2722 (w), 1590 (w) (ν(CC)), 1500 (w,sh), 1484 (m), 1440 (s),

1378 (s), 1335 (vw), 1229 (w), 1180 (vw), 1151 (vw), 1111 (vw),

1080 (vw), 1048 (vw), 1021 (m), 844 (s), 833 (m), 648 (vw), 605

(vw), 505 (w), 485 (w), 410 (s). UV−vis (hexane, nm): 330(sh) ≫

405(sh) > 480(sh). Anal. Calcd for C₂₄H₃₆Ti: C, 77.40; H, 9.74.

Found: C, 77.36; H, 9.70.

1

plausible since the presence of [TiH(C₅Me₅)₂] was found in the H

NMR spectra of both crystalline 10C and its mother liquor after

dissolution in toluene-d₈. In all structures, the C−H hydrogen atoms

were placed into idealized positions (riding model) and assigned

temperature factors either H_iso(H) = 1.2U_eq(pivot atom) or H_iso(H) =

1.5U_eq(pivot atom), except for H(14A) and H(14B) of 2, H(10A) and

H(10B) of 4, and H(21A), H(21B), H(22A), and H(22B) of 15.

These were reﬁned isotropically without any restraints. Molecular

graphics were carried out with the PLATON program.⁴¹

Reaction of 2 with But-2-yne to Give 17. Compound 2 (0.23 g,

0.53 mmol) was dissolved in hexane (10 mL), but-2-yne (1.0 mL, 12.8

mmol) was added, and the mixture was stirred at room temperature

for 1 h. A pale green solution was evaporated under vacuum, and the

residue was dissolved in hexane. Crystallization from the saturated

solution by subsequent cooling to −28 °C aﬀorded large green crystals

of 17. Yield: 0.21 g (87%).

Computational Details. DFT calculations have been carried out

at the Fermi cluster at the J. Heyrovsky Institute of Physical

́

Chemistry, Academy of Sciences of Czech Republic, vvi, using

Gaussian 09, Revision B.01.⁴²The geometry optimization was done

using the B3LYP functional; the basis set used for all atoms was 6-

311G(d,p) for 7a and 7b; the basis set 6-31G(d,p) was used for 5 and

10B. The Hessian required for the geometry optimization was

computed analytically before the ﬁrst step of any optimization. The

NMR shielding tensor for 10B was calculated using the GIAO

method⁴³carried out on the optimized geometry and using the

HSEh1PBE functional along with the 6-311++G(3d,p) basis set.

Data for 15 are as follows. Mp: 130 °C dec. ¹H NMR (benzene-d₆):

1.01 (s, 18H, CMe₃); 1.68 (s, 12H, C₅Me₄, β-Me); 1.77 (s, 6H, 

CMe); 2.19 (s, 12H, C₅Me₄, α-Me). ¹³C{¹H} NMR (benzene-d₆):

12.14, 16.45 (C₅Me₄); 20.26 (CMe); 32.02 (CMe₃); 35.97 (CMe₃);

120.87, 123.91 (C₅Me₄); 129.86 (C_ipso); 206.13 (CMe). EI-MS (130

°C): m/z (relative abundance) 456 (M^+•; not observed), 404 (14),

403 (27), 402 ([M − C₄H₆]⁺; 64), 401 (16), 400 (14), 343 ([M −

C₄H₆− C₄H₉− H₂]⁺; 36), 341 (10), 339 ([M − C₄H₆− C₄H₉− 3

H₂]⁺; 21), 337 (11), 327 (31), 325 (18), 313 (12), 228 (33), 225 ([M

− C₄H₆− Cp′]⁺; 10), 206 (15), 205 (10), 202 (26), 201 (17), 187

(27), 162 (27), 161 (12), 147 (13), 119 (34), 105 (22), 91 (14), 57

([Bu]⁺; 100), 54 ([butyne]⁺; 75), 41 (60), 39 (55). IR (KBr, cm⁻¹):

3000 (m), 2988 (s), 2954 (vs), 2900 (vs), 2833 (m), 1674 (w), 1481

(m), 1462 (m), 1442 (m), 1413 (vw), 1378 (s), 1358 (m), 1235 (m),

1197 (w), 1124 (vw), 1097 (vw), 1022 (m), 981 (vw), 929 (vw), 743

(vw), 662 (w), 590 (w), 482 (w). UV−vis (hexane, nm): 325 (sh) >

405 ≫ 520 < 695. Anal. Calcd for C₃₀H₄₈Ti: C, 78.92; H, 10.60.

Found: C, 78.86; H, 10.57.

ASSOCIATED CONTENT

■

S

* Supporting Information

CIF ﬁle for structures 2, 4, 10C, 15, 17, and 18. Table of

crystallographic data and data collection and structure reﬁne-

ment details for 2, 4, 10C, 15, 17, and 18. ¹H NMR spectra of

thermolysis of 2 to 6a and conversions of 3 to 7a and 7b.

Correlation of observed ¹³C NMR chemical shifts with

1

calculated ones for 10B. H NMR spectra of products formed

1

on aging of 10C in toluene-d₈at room temperature. H NMR

spectral evidence of equilibrium concentration of 10 generated

Reaction of 3 with But-2-yne to Give 18. Slightly greenish-

yellow crystals of 3 (0.25 g, 0.54 mmol) were dissolved in 20 mL of

1

from 11 and 12. Plot of temperature dependence of H NMR

5491

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

(11) (a) Chirik, P. J. Organometallics 2010, 29, 1500−1517.

(b) Hanna, T. E.; Lobkovsky, E.; Chirik, P. J. Organometallics 2009,

28, 4079−4088. (c) Hanna, T. E.; Lobkovsky, E.; Chirik, P. J. J. Am.

Chem. Soc. 2004, 126, 14688−14689. (d) Hanna, T. E.; Lobkovsky, E.;

Chirik, P. J. J. Am. Chem. Soc. 2006, 128, 6018−6019.

(12) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;

Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I.

Organometallics 2010, 29, 2176−2179.

chemical shifts of 13 in benzene-d₆solution (in the temperature

range 298−328 K). ¹H NMR (bottom) and 1D NOESY

(C₅Me₅irradiation) spectra of 15 in benzene-d₆solution.

Infrared spectrum of 10C in KBr pellet ground with KBr and

pressed under a nitrogen atmosphere in a glovebox. Cartesian

coordinates for DFT-optimized (6-311++G(3d,p)) structure

10B. This material is available free of charge via the Internet at

http://pubs.acs.org.

(13) (a) Pinkas, J.; Lukeso

M.; Kubist

(b) Lukesova,

Horac k, M.; Kubist

2653. (c) Horac k, M.; Step

Petrusova,

̌

va,

́

L.; Gyepes, R.; Císaro

̌

va,

́

I.; Horac

́

e

̌

k,

I.;

̌

a, J.; Mach, K. Organometallics 2010, 29, 5199−5208.

̌

̌ ̌ ́ ̌ ́

nicka, P.; Fejfarova, K.; Gyepes, R.; Císarova,

L.; Step

a, J.; Mach, K. Organometallics 2002, 21, 2639−

AUTHOR INFORMATION

̌

́

■

́

e

̌

Notes

̌

̌ ̌ ̌ ̌ ́

nicka, P.; Kubista, J.; Císarova,

́

e

̌

The authors declare no competing ﬁnancial interest.

́

L.; Mach, K. J. Organomet. Chem. 2003, 667, 154−166.

(d) Pellny, P.-M.; Burlakov, V. V.; Baumann, W.; Spannenberg, A.;

̌

ACKNOWLEDGMENTS

Horac

́

e

̌

k, M.; Step

̌

nick

̌

a, P.; Mach, K.; Rosenthal, U. Organometallics

■

̌

́ ̌ ̌ ̌

ek, M.; Stepnicka, P.; Gyepes, R.;

2000, 19, 2816−2819. (e) Horac

Císarova, I.; Polas k, M.; Mach, K.; Pellny, P.-M.; Burlakov, V. V.;

This research was supported by Grant Agency of the Czech

Republic (Project No. P207/12/2368). I.C. is grateful to

MSM0021620857. J.K. was supported by Grant Agency of the

Czech Republic (Project No. 203/09/P276).

̌

́

́ ̌

e

Baumann, W.; Spannenberg, A.; Rosenthal, U. J. Am. Chem. Soc. 1999,

121, 10638−10639.

(14) Polas

4083.

(15) Pinkas, J.; Lukesova, L.; Gyepes, R.; Císarova, I.; Lonnecke, P.;

́ ̌ ̌

ek, M.; Kubista, J. J. Organomet. Chem. 2007, 692, 4073−

REFERENCES

̌

́

̌

́

̈

■

Kubist

3110.

(16) (a) Mach, K.; Varga, V.; Hanus,

Chem. 1991, 415, 87−95. (b) Vondrak

̌

a, J.; Horac

́

e

̌

k, M.; Mach, K. Organometallics 2007, 26, 3100−

(1) For Ti (review): (a) Rosenthal, U.; Burlakov, V. V.; Arndt, P.;

Baumann, W.; Spannenberg, A. Organometallics 2003, 22, 884−900.

For Ti and Zr (review): (b) Rosenthal, U.; Burlakov, V. V.; Arndt, P.;

Baumann, W.; Spannenberg, A.; Shur, V. B. Eur. J. Inorg. Chem. 2004,

4739−4749. For Hf: (c) Beweries, T.; Burlakov, V. V.; Bach, M. A.;

Arndt, P.; Baumann, W.; Spannenberg, A.; Rosenthal, U. Organo-

̌

V.; Sedmera, P. J. Organomet.

́

, T.; Mach, K.; Varga, V.;

Terpstra, A. J. Organomet. Chem. 1992, 425, 27−39.

(17) de Wolf, J. M.; Blaauw, R.; Meetsma, A.; Teuben, J. H.; Gyepes,

R.; Varga, V.; Mach, K.; Veldman, N.; Spek, A. L. Organometallics

1996, 15, 4977−4983.

̌

metallics 2007, 26, 247−249. (d) Lamac, M.; Spannenberg, A.;

Baumann, W.; Jiao, H.; Fischer, C.; Hansen, S.; Arndt, P.; Rosenthal,

U. J. Am. Chem. Soc. 2010, 132, 4369−4380. (e) Kaleta, K.; Ruhmann,

M.; Theilmann, O.; Beweries, T.; Roy, S.; Arndt, P.; Villinger, A.;

Jemmis, E. D.; Schulz, A.; Rosenthal, U. J. Am. Chem. Soc. 2011, 133,

5463−5473 , and references therein..

(18) Mach, K. Unpublished results.

̌

̌ ́ ̌ ̌

erova, I.; Stepnicka, P.; Gyepes,

(19) (a) Kupfer, V.; Thewalt, U.; Tisl

R.; Kubista, J.; Horac k, M.; Mach, K. J. Organomet. Chem. 2001, 620,

39−50. (b) Horac k, M.; Step

Císarova, I.; Petrusova, L.; Mach, K. J. Organomet. Chem. 2002, 658,

235−241.

̌

́ ̌

e

̌

̌ ̌ ̌

nicka, P.; Kubista, J.; Gyepes, R.;

́

̌

́

(2) For Ti: (a) Cohen, S. A.; Auburn, P. R.; Bercaw, J. E. J. Am.

Chem. Soc. 1983, 105, 1136−1143. (b) Cohen, S. A.; Bercaw, J. E.

Organometallics 1985, 4, 1006−1014.

(20) (a) McDade, C.; Green, J. C.; Bercaw, J. E. Organometallics

1982, 1, 1629−1634. (b) Bercaw, J. E.; Marvich, R. H.; Bell, L. G.;

Brintzinger, H. H. J. Am. Chem. Soc. 1972, 94, 1219−1238. (c) Bercaw,

J. E.; Brintzinger, H. H. J. Am. Chem. Soc. 1971, 93, 2045−2046.

(21) Pattiasina, J. W. Ph.D. Thesis, University of Groningen, The

Netherlands, 1988; pp 35−77.

(3) Rosenthal, U.; Burlakov, V. V. In Titanium and Zirconium in

Organic Synthesis; Marek, I., Ed.; Wiley-VCH: Weinheim, Germany,

2002; pp 355−389.

(4) (a) Rep, M.; Kaagman, J.-W. F.; Elsevier, C. J.; Sedmera, P.;

Hiller, J.; Thewalt, U.; Horac

2000, 597, 146−156. (b) Witte, P. T.; Klein, R.; Kooijman, H.; Spek,

A. L.; Polas k, M.; Varga, V.; Mach, K. J. Organomet. Chem. 1996, 519,

195−204. (c) Gyepes, P.; Witte, P. T.; Horac k, M.; Císarova, I.;

́ ̌

ek, M.; Mach, K. J. Organomet. Chem.

(22) (a) Lukens, W. W., Jr.; Matsunaga, P. T.; Andersen, R. A.

Organometallics 1998, 17, 5240−5247. (b) de Wolf, J. M.; Meetsma,

A.; Teuben, J. H. Organometallics 1995, 14, 5466−5468.

(23) Luinstra, G. A.; Teuben, J. H. J. Am. Chem. Soc. 1992, 114,

3361−3367.

́ ̌

e

́

e

̌

́

Mach, K. J. Organomet. Chem. 1998, 551, 207−213.

(5) (a) Pattiasina, J. W.; Hissink, C. E.; de Boer, J. L.; Meetsma, A.;

(24) Pinkas, J.; Císaro

̌ ́ ́ ̌

va, I.; Conde, A.; Fandos, R.; Horacek, M.;

Teuben, J. H.; Spek, A. L. J. Am. Chem. Soc. 1985, 107, 7758−7759.

(b) Varga, V.; Mach, K.; Polasek, M.; Sedmera, P.; Hiller, J.; Thewalt,

́ ̌

Kubist

̌

a, J.; Mach, K. Inorg. Chem. Commun. 2009, 12, 11−14.

(25) Pattiasina, J. W.; Heeres, H. J.; van Bolhuis, F.; Meetsma, A.;

U.; Troyanov, S. I. J. Organomet. Chem. 1996, 506, 241−251.

(6) (a) Bercaw, J. E. J. Am. Chem. Soc. 1974, 96, 5087−5095.

(b) Sanner, R. D.; Duggan, D. M.; McKenzie, T. C.; Marsh, R. E.;

Bercaw, J. E. J. Am. Chem. Soc. 1976, 98, 8358−8365. (c) 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, 3078−3083.

(7) Lauher, J. W.; Hoffmann, R. J. Am. Chem. Soc. 1976, 98, 1729−

1742.

Teuben, J. H.; Spek, A. L. Organometallics 1987, 6, 1004−1010.

̌ ́ ́ ̌

(26) Pinkas, J.; Císarova, I.; Conde, A.; Fandos, R.; Horacek, M.;

̌

Kubista, J.; Mach, K. J. Organomet. Chem. 2009, 694, 1971−1980.

(27) (a) Bouwkamp, M. W.; de Wolf, J.; del Hierro Morales, I.;

Gercama, J.; Meetsma, A.; Troyanov, S. I.; Hessen, B.; Teuben, J. H. J.

Am. Chem. Soc. 2002, 124, 12956−12957. (b) Bouwkamp, M. W.;

Budzelaar, P. H. M.; Gercama, J.; del Hierro Morales, I.; de Wolf, J.;

Meetsma, A.; Troyanov, S. I.; Teuben, J. H.; Hessen, B. J. Am. Chem.

Soc. 2005, 127, 14310−14319.

(8) (a) Mc Dermott, J. X.; Whitesides, G. M. J. Am. Chem. Soc. 1974,

96, 947−948. (b) Mc Dermott, J. X.; White, J. F.; Whitesides, G. M. J.

Am. Chem. Soc. 1976, 98, 6521−6528. (c) Mc Dermott, J. X.; Wilson,

M. E.; Whitesides, G. M. J. Am. Chem. Soc. 1976, 98, 6529−6536.

(9) Arndt, P.; Burlakov, V. V.; Spannenberg, A.; Rosenthal, U. Inorg.

(28) (a) Sato, K.; Nishihara, Y.; Huo, S.; Xi, Z.; Takahashi, T. J.

Organomet. Chem. 2001, 633, 18−26. (b) Takahashi, T.; Kageyama,

M.; Denisov, V.; Hara, R.; Negishi, E. Tetrahedron Lett. 1993, 34, 687−

690. (c) Takahashi, T.; Xi, Z.; Kageyama, M.; Nishihara, Y.; Huo, S.;

Kasai, K.; Aoyagi, K.; Denisov, V.; Negishi, E. Tetrahedron 1997, 53,

9123−9134.

(29) (a) Beckhaus, R.; Sang, J.; Oster, J.; Wagner, T. J. Organomet.

Chem. 1994, 484, 179−190. (b) Beckhaus, R.; Strauss, I.; Wagner, T. J.

Organomet. Chem. 1994, 464, 155−161.

Chem. Commun. 2007, 10, 792−794.

̌

(10) (a) Horac

́

e

̌

k, M.; Kupfer, V.; Thewalt, U.; Step

̌

nick

̌

a, P.; Polas

́

e

̌

k,

M.; Mach, K. Organometallics 1999, 18, 3572−3578. (b) Walter, M.

D.; Sofield, C. D.; Andersen, R. A. Organometallics 2008, 27, 2959−

2970. (c) Aazam, E. S. Ph.D. Thesis, University of Sussex at Brighton,

United Kingdom, 2001; pp 54−62.

5492

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Organometallics

Article

̌

va,

́ ́ ̌ ̌

(30) Pinkas, J.; Císaro I.; Gyepes, R.; Horacek, M.; Kubista, J.;

̌

Cejka, J.; Gom

́

ez-Ruiz, S.; Hey-Hawkins, E.; Mach, K. Organometallics

2008, 27, 5532−5547.

(31) Threlkel, R. S. Ph.D. Dissertation, California Institute of

Technology, 1980; pp 1−17.

(32) Stoebenau, E. J., III; Jordan, R. F. J. Am. Chem. Soc. 2003, 125,

3222−3223.

(33) (a) Semproni, S. P.; Milsmann, C.; Chirik, P. J. Organometallics

2012, 31, 3672−3682. (b) Hanna, T. E.; Bernskoetter, W. H.;

Bouwkamp, M. W.; Lobkovsky, E.; Chirik, P. J. Organometallics 2007,

26, 2431−2438.

(34) Antropiusova,

Met. Chem. (London) 1981, 6, 90−93.

́ ̌ ̌

(35) Lukesova, L.; Gyepes, R.; Pinkas, J.; Horacek, M.; Kubista, J.;

́ ́ ̌

H.; Dosedlova, A.; Hanus, V.; Mach, K. Transition

̌

́

̌

Cejka, J.; Mach, K. Collect. Czech. Chem. Commun. 2005, 70, 1589−

1603.

(36) Horac

Mach, K. Collect. Czech. Chem. Commun. 1996, 61, 1307−1320.

(37) Mach, K.; Varga, V.; Antropiusova, H.; Polac k, J. J. Organomet.

Chem. 1987, 333, 205−215.

́ ̌ ̌ ́ ́ ̌

ek, M.; Gyepes, R.; Císarova, I.; Polasek, M.; Varga, V.;

́

́ ̌

e

(38) Otwinowski, Z.; Minor, W. HKL Denzo and Scalepack program

package by Nonius. For a reference, see Otwinowski, Z.; Minor, W.

Methods Enzymol. 1997, 276, 307−326.

(39) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.;

Giacovazzo, C.; Guagliardi, A.; Polidori, G. J. Appl. Crystallogr. 1994,

27, 435−436.

(40) Sheldrick, G. M. SHELXL97, Program for Crystal Structure

Reﬁnement from Diﬀraction Data; University of Gottingen: Gottingen,

̈

1997.

(41) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool;

Utrecht University: Utrecht, The Netherlands, 2007.

(42) Frisch, M. J.; et al. Gaussian 09, Revision B.01; Gaussian, Inc.:

Wallingford, CT, 2010.

(43) Ditchfield, R. Mol. Phys. 1974, 27, 789−807.

5493

dx.doi.org/10.1021/om300461k | Organometallics 2012, 31, 5478−5493

Article Doi

Ethene complexes of bulky titanocenes, their thermolysis, and their reactivity toward 2-butyne

DOI: 10.1021/om300461k

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om300461k

Full text of DOI:10.1021/om300461k

Products guided by the article

R&D Labs maybe for 1388860-45-4

Relevant to this article

Hot Product