Unprecedented Spectroscopic and Computational Evidence for Allenyl and Propargyl Titanocene(IV) Complexes: Electrophilic Quenching of Their Metallotropic Equilibrium

Article abstract of DOI:10.1002/chem.201504281

The synthesis and structural characterization of allenyl titanocene(IV) [TiClCp₂(CH=C=CH₂)] 3 and propargyl titanocene(IV) [TiClCp₂(CH₂-C≡C-(CH₂)₄CH₃)] 9 have been described for the first time. Advanced NMR methods including diffusion NMR methods (diffusion pulsed field gradient stimulated spin echo (PFG-STE) and DOSY) have been applied and established that these organometallics are monomers in THF solution with hydrodynamic radii (from the Stokes-Einstein equation) of 3.5 and 4.1 ? for 3 and 9, respectively. Full ¹H, ¹³C, Δ¹H, and Δ¹³C NMR data are given, and through the analysis of the Ramsey equation, the first electronic insights into these derivatives are provided. In solution, they are involved in their respective metallotropic allenyl-propargyl equilibria that, after quenching experiments with aromatic and aliphatic aldehydes, ketones, and protonating agents, always give the propargyl products P (when carbonyls are employed), or allenyl products A (when a proton source is added) as the major isomers. In all the cases assayed, the ratio of products suggests that the metallotropic equilibrium should be faster than the reactions of 3 and 9 with electrophiles. Indeed, DFT calculations predict lower Gibbs energy barriers for the metallotropic equilibrium, thus confirming dynamic kinetic resolution.

Full text of DOI:10.1002/chem.201504281

DOI: 10.1002/chem.201504281
Full Paper
&
Titanocene Complexes
Unprecedented Spectroscopic and Computational Evidence for
Allenyl and Propargyl Titanocene(IV) Complexes: Electrophilic
Quenching of Their Metallotropic Equilibrium
[
a]
[a]
[b, e]
[c]
Ana BelØn Ruiz-Muelle, Pascual OÇa-Burgos, Manuel A. OrtuÇo,
J. Enrique Oltra,
[a]
[a, d]
Ignacio Rodríguez-García, and Ignacio Fernµndez*
Abstract: The synthesis and structural characterization of al-
lenyl titanocene(IV) [TiClCp (CH=C=CH )] 3 and propargyl ti-
lution, they are involved in their respective metallotropic al-
lenyl–propargyl equilibria that, after quenching experiments
with aromatic and aliphatic aldehydes, ketones, and pro-
tonating agents, always give the propargyl products P
(when carbonyls are employed), or allenyl products A (when
a proton source is added) as the major isomers. In all the
cases assayed, the ratio of products suggests that the metal-
lotropic equilibrium should be faster than the reactions of 3
and 9 with electrophiles. Indeed, DFT calculations predict
lower Gibbs energy barriers for the metallotropic equilibri-
um, thus confirming dynamic kinetic resolution.
2
2
tanocene(IV) [TiClCp (CH ÀCꢀCÀ(CH ) CH )] 9 have been de-
2
2
2 4
3
scribed for the first time. Advanced NMR methods including
diffusion NMR methods (diffusion pulsed field gradient
stimulated spin echo (PFG-STE) and DOSY) have been ap-
plied and established that these organometallics are mono-
mers in THF solution with hydrodynamic radii (from the
Stokes–Einstein equation) of 3.5 and 4.1 Š for 3 and 9, re-
1
13
1
13
spectively. Full H, C, D H, and D C NMR data are given,
and through the analysis of the Ramsey equation, the first
electronic insights into these derivatives are provided. In so-
Introduction
offer unique properties owing to their stabilization by cyclo-
pentadienyl ligands. Allyl and propargyl titanocenes are gener-
ally nucleophilic, and can interact with a variety of electro-
philes. Derivatives of titanocene(III) and titanocene(IV) have
been well characterized over the past decades but, interesting-
ly, there has been no spectroscopic evidence of allenyl or
propargyl derivatives described so far.
Organotitanium reagents have found, in the last few decades,
broad applicability in organic synthesis. They can perform
a wide variety of transformations, including aldol reactions,
asymmetric epoxidations, olefinations, cyclopropanations, hy-
[
1]
drogenations, reductive couplings, and polymerizations.
Many of these powerful methods are now established as indis-
pensable tools to the synthetic chemist, as titanocene species
Titanocene(III) derivatives are well known to be effective
inner-sphere single-electron transfer reagents, and can reduce
epoxides, ozonides, carbonyls, and NÀO bonds. The distinct re-
activity of each of titanocene’s oxidation state has allowed for
[
1]
[
a] A. B. Ruiz-Muelle, Dr. P. OÇa-Burgos, Prof. Dr. I. Rodríguez-García,
Prof. Dr. I. Fernµndez
its mechanistic flexibility. These versatile complexes have fre-
quently been studied as reagents for organic and polymer syn-
thesis, and even as biologically active pharmaceutical agents
Department of Chemistry and Physics, ceiA3, Universidad de Almería
Carretera de Sacramento s/n, Almería, 04120 (Spain)
E-mail: ifernan@ual.es
[
2]
themselves. Structurally, titanocene(III) chloride exists in solu-
tion as a mixture between the monomer and dimer, depend-
ing on the conditions employed. Mechanistically, titanocene(III)
complexes act to activate the halide of the electrophile by se-
quential electron-transfer processes to generate transient allyl
or propargyl titanocene(IV) species that have never been char-
[
b] Dr. M. A. OrtuÇo
Departament de Química, Universitat Autònoma de Barcelona, Edifici C.n
Barcelona, Cerdanyola del Vall›s, 08193 (Spain)
[
c] Prof. Dr. J. E. Oltra
Departamento de Química Orgµnica, Facultad de Ciencias
Universidad de Granada, Campus Fuentenueva s/n
Granada, 18071 (Spain)
[
3–7]
acterized.
[
d] Prof. Dr. I. Fernµndez
With respect to the application of NMR methodologies to-
wards the unravelling of these titanocene derivatives, whether
in catalytic or stoichiometric processes, there are no literature
precedents and the data reported previously by Oltra and co-
BITAL, Research Centre for Agricultural and Food Biotechnology
Carretera de Sacramento s/n, Almería (Spain)
[
e] Dr. M. A. OrtuÇo
Present address: Department of Chemistry, University of Minnesota
Minneapolis, Minnesota 55455-0431 (USA)
[
13]
workers turned out to be inaccurate, as discussed below.
In 1982, Yamamoto and co-workers described the synthesis
of a-allenic and b-acetylenic alcohols through a propargyl tita-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201504281.
Chem. Eur. J. 2016, 22, 2427 – 2439
2427
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
er-type method to the synthesis of exocyclic allenes under
mild conditions and have even applied the procedure to the
[
14]
synthesis of the natural alkaloid stemoamide. In that contri-
IV
bution, a metallotropic equilibrium between a propargyl–Ti
IV
and an allenyl–Ti species is proposed. On the other hand,
Chemla and Ferreira have compared the stabilities of titanium
carbenoids with the corresponding lithium and zinc deriva-
[
15]
tives.
As in the synthesis of complex 1, the corresponding titanium
organometallics were prepared by lithiation with nBuLi at
À958C in the presence of tetramethylethylenediamine
[
16]
(
TMEDA) and subsequent transmetallation with Ti(iPrO)4.
Scheme 1. Yamamoto’s work and the first evidence of the metallotropic
equilibrium of titanium complexes.
These authors postulate the existence of the metallotropic
equilibrium between propargyl and allenyl complexes, show-
ing their configurational stability at certain conditions, but un-
nium intermediate (1) obtained after transmetalation of the
fortunately attempts to characterize the metallocarbenoids by
[
8]
[15]
corresponding lithiated derivative with [Ti(iPrO) ] (Scheme 1).
low-temperature NMR experiments were unsuccessful.
In
4
The condensation with cyclohexanecarboxaldehyde pro-
duced the allene or acetylenic derivatives when terminal or in-
ternal alkynes were employed, respectively. The explanation
given by these authors, supported by IR data, was based on an
allylic (metallotropic) rearrangement of the organometallic spe-
cies. The authors did not provide additional details about
these titanium isopropoxide intermediates, so caution must be
taken when considering these. The high regioselectivity ob-
served appeared to be general for a range of acetylenes and
this case, the reactive intermediate proved to react in both its
propargylic and allenic forms, affording mixtures of a-allenic
and b-acetylenic alcohols with modest regioselectivities.
[
17]
As mentioned before, previous works in this area have in-
1
1
cluded the characterization of h -propargyl and/or h -allenyl
[
18]
[19]
[18]
[20]
[21]
[22]
[23]
complexes of Fe, Co, Mo, W, Ru, Pd, and Pt.
For comparison purposes regarding the complexes synthesized
herein, we provide in Scheme 3 and Scheme 4 the most rele-
1
13
vant H and C NMR data for all these complexes.
[
9]
[10]
aldehydes, including optically active ones, and also for the
[11]
intramolecular version to afford cyclic allenyl alcohols.
Some features that may shed light on the structural empti-
ness regarding the metallotropic rearrangement could be de-
duced from complex [Cp TiÀCH CCCH ÀTiCp ] (2; Cp=cyclo-
2
2
2
2
[
12]
pentadienyl), which is identified by the resonance structures
of a butynediyl-bridged s-propargylic complex, a butatriene-
bridged p-complex, and a bimetallic dititanabicycle species.
[8–11]
According to Yamamoto and Sato,
the participation of
the aldehyde or ketone group occurs after the oxidation of ti-
tanium(III) to titanium(IV). Later, Oltra and co-workers rational-
ized that once the allenyl radical is formed by means of
III
III
a [Ti ClCp ], it is trapped by a second molecule of [Ti ClCp ]
2
2
to give the corresponding organometallic species (3*),
which eventually attacks the added carbonyl compound
1
13
[
13]
13
Scheme 3. Diagnostic H and C NMR data for allenic (A) metal complexes
4
(
Scheme 2). These authors provided C NMR chemical shifts
–7 described in the literature.
for the intermediate species 3* as the only supporting data for
the intermediates involved in this process, albeit in an incon-
clusive manner. Some of us have recently expanded this Barbi-
For some time now, some of us have been interested in the
reaction of titanocene chlorides with terminal and internal
propargyl halides as a way to synthesize more elaborate prod-
ucts. We provide herein definitive proof of allenyl and proparg-
yl titanocene structures in solution by means of advanced
NMR methods, infrared spectroscopy, electrophilic quenching
reactions, and DFT calculations.
Results and Discussion
As already seen, it is known that many allenyl metals can un-
dergo a metallotropic rearrangement resulting in either the ho-
mopropargylic alcohol or the allenic alcohol upon reaction
Scheme 2. Selective formation of terminal homopropargylic alcohols. The
C NMR data can be found in reference [13].
1
3
Chem. Eur. J. 2016, 22, 2427 – 2439 www.chemeurj.org
2428
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Allenyl titanium species 3
First, [TiClCp ] was generated in situ at room temperature by
2
stirring commercially available [TiCl Cp ] (2 equiv) with Mn dust
2
2
(
1.0 equiv) in deoxygenated THF for 2 h. The resulting solution
was filtered through a Millipore filter and then propargyl chlo-
ride (1.0 equiv) was added, forming an equimolecular mixture
of the allenyl titanium species and [TiCl Cp ] (Scheme 5). The
2
2
latter precipitated out of the solution during the course of the
NMR measurements.
2
Scheme 5. In situ preparation of [TiCpCl ] and its reaction with propargylic
chloride in THF at room temperature.
1
13
Scheme 4. Diagnostic H and C NMR data for propargylic (P) metal com-
plexes described in the literature.
1
The H NMR spectrum (Figure 1a) essentially exhibits signals
corresponding to the new titanium species and to [TiCl Cp ].
2
2
In complex 3, three signals are observed at d =3.75 (J=
H
with a carbonyl substrate. As with other metals, it has been
found that with titanium intermediates, the relative ratio of
products depends on the solvent, the nature of the propargyl
5.7 Hz), 5.99 (J=5.7 Hz), and 6.33 ppm, which are assigned to
the CH (H3), CH (H1), and Cp (H4) groups, respectively. The in-
2
tegration data were in perfect agreement with the presence of
two cyclopentadiene groups. Particularly diagnostic is a rela-
[13]
halide, or the carbonyl substrate. In fact, Oltra and co-work-
ers have studied the reactivity of organotitanium species
under various conditions, finding that terminal propargyl ha-
lides with aldehydes and ketones exclusively afford homopro-
4
tively large J coupling between the methylene and methine
protons (J=5.7 Hz), which is very similar to that observed pre-
viously for the tungsten compound 4A (Scheme 3) and that
[13]
[18]
1
pargylic alcohols in all cases assayed.
for the iron complex. In addition, the H chemical shifts for
both H1 and H3 are consistent with the data reported for
other allenyl metals such as those illustrated in Scheme 4. Se-
lective gradient total correlation spectroscopy (gTOCSY) irradia-
tion of the allenylic H1 clearly corroborates its coupling, with
NMR studies
Owing to the weak spectroscopic support for titanium allenyl
or propargyl species, we attempted the structural characteriza-
tion of these titanium organometallics by monitoring their for-
mation reactions with NMR spectroscopy. The NMR studies
were carried out on a spectrometer operating at a proton fre-
quency of 500.13 MHz. The information collected was comple-
mented by diffusion pulsed field gradient stimulated spin echo
the doublet located at d =3.75 ppm (Figure 1b).
H
A deeper examination of the spectrum showed some mech-
anism-related signals that are now highlighted. Resonances as-
signable to potential propargyl titanium species were not ob-
(
PFG-STE) and DOSY experiments. The NMR samples of 80 mm
in [D ]THF were prepared under conditions analogous to those
8
used in the bulk (see the Experimental Section). The only dif-
ference introduced is that in a glovebox, the reaction vial was
placed in the freezer at À308C for 30 min to help the precipi-
tation of MnCl . Upon filtration through a Millipore filter, the
2
clear solution was transferred into a J. Young 5 mm NMR tube
at room temperature. The sample was then transferred to the
1
13
spectrometer and then a series of H, COSY, C, DEPT-135,
gHMQC and gHMBC NMR spectra were sequentially measured
at increasing time. The spectra obtained are shown below and
in the Supporting Information. We first tackled the NMR struc-
tural study of 3 and secondly the same arsenal of experiments
was applied to 9. The sample preparation was analogous for
both of them.
1
Figure 1. a) H NMR spectrum (500.13 MHz) of 3 measured in [D
8
]THF at
room temperature. b) 1D gTOCSY spectrum with selective excitation of the
triplet at d
H
=5.99 ppm (denoted with #) with a spin-lock duration of 90 ms.
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2429
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
served, although a tiny amount (1% and 3% with respect to 3)
of its protonation product, propyne, was detected through the
signals at d =1.75 and 2.02 ppm. The product formed by the
H
protonation of 3, propadiene, was also detected through just
one singlet at d =4.66 ppm (3% and 8% with respect to 3).
H
The presence of these two species could confirm the existence
of the metallotropic equilibrium between the allenic and prop-
argylic forms of the organotitanium, although at room temper-
ature and under the conditions assayed, only the allenyl titani-
um species 3 is observed. The rest of signals detected in the
1
H NMR spectrum were assigned to toluene and propargyl
chloride. Prior to undertaking the full spectral assignment of
compound 3, we cooled down the temperature of the sample
from room temperature to À1008C in steps of 208C and mea-
1
sured the H NMR spectra at the new temperatures. This tem-
perature modification was implemented for one main reason:
the metallotropic equilibrium between species 3 and its prop-
argylic counterpart could be altered and therefore directly
identify its existence. Unfortunately, we were not able to ob-
serve the propargylic titanocene derived from complex 3, and
only the progressive broadening of the spectrum was clearly
seen (see the Supporting Information).
Figure 2. Expansion of the 2D NMR gHMBC spectrum ([D
complex 3, showing the selected interactions used for structural elucidation.
8
]THF, 295 K) of
13
The C NMR spectrum helped to clarify the type of inter-
mediate formed during the process. As expected, the spectrum
is dominated by the signals of 3 [d =61.2 (=CH ), 120.1
C
2
1
13
Table 1. Selection of H and C NMR data (d in ppm, J in Hz) of 3 and 9
(
=CH), 123.2 (Cp), and 198.3 ppm (=C=)] and [TiCl Cp ] (d =
2 2 C
1
13
8
measured in [D ]THF at room temperature. H and C coordination shifts
119.9 ppm). Nevertheless, signals for propyne, propadiene,
1
13
[a]
(
D Hcoord and D Ccoord in ppm) are in parenthesis.
propargyl chloride, and toluene were also detected, as men-
tioned before. Particularly diagnostic is the broad quaternary
1
13
Compounds
Label
H
C
singlet at d =198.3 ppm, which nicely proves the sp-hybridized
1
2
3
4
5.99 (+1.33)
121.1 (+47.4)
198.3 (À14.2)
61.2 (À11.5)
123.2
C
allenic nature of 3. The downfield chemical shift of C1 found
3.75 (À0.91)
at d =120.1 ppm is due to the contribution of the paramag-
C
6.33
netic term s to the screening constant (the diamagnetic com-
p
ponent s is not necessary small, but may not change much)
d
[
24]
as already documented for other titanium organometallics.
1
2
3
4
5
6
7
8
9
1.86 (+0.1)
38.6 (+35.2)
88.5 (+13.2)
83.2 (+3.8)
19.2 (+0.3)
29.7 (+0.7)
31.2 (À0.1)
22.2 (À0.2)
13.5 (À0.6)
117.2
Except for the unusual chemical shift of the carbon directly
attached to the titanium, the rest of signals are comparable to
1
2.22 (+0.1)
1.47 (+0.07)
1.47 (+0.07)
1.47 (+0.07)
0.93 (+0.03)
6.36
those found in others metal h -allenyl complexes such as 4A–
7
A (Scheme 3). The complete structural assignment of 3 was
based on the analysis of the bidimensional spectra. The
gHMQC allowed us to identify the most relevant CÀH pairs of
nuclei and the gHMBC (Figure 2) permitted us to establish the
connectivity in the core fragment, such as determining the
1
13
[a] The H and C NMR coordination shifts were determined with respect
to free propadiene and oct-2-yne for 3 and 9, respectively.
quaternary carbon at d =198.3 ppm to be part of the allenic
H
moiety (Table 1).
The IR spectrum of the allenyl titanium 3 showed a weak to
very weak absorption for the (asymmetric) stretching vibration
complex 3. The resulting solution was filtered through a Milli-
pore filter and then 1-chlorooct-2-yne (1 equiv) was added
dropwise, forming an equimolecular mixture of the desired
propargyl titanium species 9 and [TiCl Cp ] (Scheme 6). The
À1
of the allenyl moiety at 1895 cm . Generally, allenic hydrocar-
À1 [18a,23a]
bons absorb in the region around 1950 cm .
Our value
1
is in good agreement, however, with values reported for h -al-
2
2
lenyl complexes derived from other transition metals such as
latter also precipitated out of the solution, but in contrast to
the synthesis of 3, no protonation adducts were detected from
the propargyl or allenyl derivatives, which is probably associat-
ed with the higher stability of 9 against hydrolysis.
4
A–7A (Scheme 3).
Propargyl titanium species 9
Regarding the structural characterization of 9, it was again
readily achieved with the combined aid of IR and NMR spec-
At room temperature, we first generated [TiClCp ] (2.0 equiv)
2
À1
with Mn dust (1.0 equiv) in deoxygenated THF, the same as for
troscopy. The IR spectrum showed a weak band at 2161 cm
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2430
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
broadest width of W1/2 =2.2 Hz compared with the rest (aver-
age W1/2 =0.8 Hz). This could be explained by the proximity of
4
7/49
the quadrupolar
transverse relaxation rate.
This feature indirectly supports the h -coordination fashion
Ti nucleus, which causes an increase in the
1
13
of the propargyl moiety. Similar C chemical shifts for alpha ti-
tanocenic carbons have been previously observed for
Scheme 6. In situ preparation of [TiCl
2
Cp] and its reaction with 1-chlorooct-
[
TiClCp Me] and [TiCp Me ] with d =49.3 and 45.6 ppm, re-
2 2 2 C
2
-yne in THF at room temperature.
[
25,26]
spectively.
The signals of the alkyne carbons C2 and C3
(
dC =88.5 and 83.2 ppm, respectively), absent in the DEPT-135
attributable to the CꢀC stretching frequency. This band is in
NMR spectrum (Figure 4b), fit again reasonably well with other
propargyl organometallics (Scheme 4).
1
the range of coordinated h -propargyls such as 4P–8P. The
1
H NMR spectrum essentially exhibits six signals, which confirm
The complete structural assignment was supported by the
analysis of the 2D spectra. Figure 5 shows an interesting cross
peak observed between H9 and C1, which supports not only
the titanocenic nature of the complex but also that both the
the expected presence of a propargyl ligand (Figure 3a).
propargylic moiety and the Cp rings are linked together by
3
a through-metal J coupling constant. This feature, as will be
CH
shown below, is supported by the diffusion NMR studies in
both complexes 9 and 3.
1
13
The full H and C NMR data for 3 and 9 are given in
Table 1. To the best of our knowledge, these are the first full
1
13
spectroscopic H and C data sets for allenic and propargylic
organotitaniums reported so far. In addition to the chemical
1
13
1
13
shifts, H and C coordination shifts (D H
, D Ccoord) are also
coord
provided in Table 1.
1
The acquisition of H NMR spectra as a function of tempera-
ture was performed from room temperature to À1008C in
steps of 208C. As in the case of complex 3, we could not find
any signals attributable to the allenyl species derived from 9
and only the progressive broadening of the spectrum was ob-
served (see the Supporting Information).
1
Figure 3. a) H NMR spectrum (500.13 MHz) of 9 measured in [D
8
]THF at
room temperature. b) 1D gTOCSY spectrum with selective excitation of the
triplet at d =2.22 ppm (denoted with #) with a spin-lock duration of 90 ms.
H
The sharp signal at d =6.36 ppm in the Cp region integrat-
H
ed to give ten protons with respect to the rest of signals, con-
firming its titanocene nature. The singlet signal (W1/2 =6.0 Hz)
at d =1.86 ppm was assigned to H1, and the triplet found at
H
higher frequency (d =2.22 ppm) corresponded to the methyl-
H
ene H4 of the pentyl chain. Interestingly, the 1D gTOCSY re-
sulting from the selective excitation of this signal revealed the
whole spin system as illustrated in Figure 3b.
13
In the C NMR spectrum (Figure 4a), the resonance C1
occurs considerably downfield at d =38.6 ppm and with the
C
13
Figure 4. Sections of C (a) and DEPT-135 (b) NMR (125.76 MHz) spectra of 9
measured in [D ]THF at room temperature. Note the absence of the quater-
8
8
Figure 5. Section of the 2D NMR gHMBC spectrum ([D ]THF, room tempera-
nary alkyne carbons C2 and C3 and the negative phase of the methylenes in
the polarization transfer experiment.
ture) of complex 9, showing the key interactions used for structural elucida-
tion.
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2431
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Diffusion studies
For the diffusion studies presented, two model systems with
different steric hindrance were employed, that is, the partially
deuterated THF as an internal standard, and the titanocene di-
chloride precursor as an external one. From the measured D
values for 3 and 9 (entries 1 and 3, Table 2), we estimate (from
The aggregation and solvation numbers of organometallic in-
termediates play a dominant role in reaction mechanisms and
[
27]
product formation.
Although we could characterize com-
pounds 3 and 9, we were also interested in the study of their
diffusive properties through molecular diffusion NMR spectros-
copy. This methodology has gained increasing importance in
this area and is routinely employed to estimate molecular vol-
the Stokes–Einstein equation) the hydrodynamic radii r to be
H
3.5 and 4.1 Š, respectively, which are in reasonable agreement
with the values derived from the crystallographic data for
monomeric cyclopentadienyl titanium species (see the Sup-
porting Information), given that solvent molecules cannot be
[
28]
[29]
umes, complex mixtures, ion-pairing, host–guest interac-
[
30]
[31]
[12,35]
tions, and catalysis. In fact, it represents an extraordinary
tool if uncertain whether solid-state structures are retained in
excluded from the calculation.
Solid-state radii can be obtained by simply dividing the
volume of the crystallographic cells by the number of asym-
metric units contained within and calculating the radii by the
[
32]
solution. The physical observable that can be derived from
the diffusion NMR experiment is the diffusion coefficient, D,
which is sensitive to the size and shape of the molecular spe-
3
assumption of spherical shape (so that follows V=4/3pr ). The
[33]
cies. Often, the Stokes–Einstein equation and its modifica-
D values obtained for the unfunctionalized alkynes (entries 6
and 7, Table 2) provide hydrodynamic radii of 1.6 and 2.7 Š,
significantly smaller than their corresponding titanocene deriv-
atives. For known dinuclear titanium organometallics, the cal-
culated crystallographic radii are in the range of 4.9–5.3 Š
(Scheme S1 in the Supporting Information). Comparison of
these values with those experimentally obtained for allenyl (3)
and propargyl (9) titanocenes confirms for the first time their
monomeric nature.
[
28a,b]
tions
are useful and enable molecular size estimation of
large particles that are much larger than the solvent. These cal-
culated hydrodynamic radii, r , assume spherical shapes;
H
hence, they do not represent the real shape of the molecules.
Nevertheless, their use is well established for comparisons as
they offer a rapid and easy method to recognize ion pairing
and/or aggregation.
To the best of our knowledge, the challenge of applying dif-
fusion NMR methods to allenic and propargylic organometal-
lics has never been achieved and, therefore, their size and/or
aggregation numbers have never been studied. Table 2 shows
Electrophilic reactions
To get more detailed reactivity knowledge, titanocenes 3 and
9 were prepared in situ in THF and reacted with one molar
equivalent of an electrophile at room temperature. The result-
ing brownish solution was subsequently treated with dilute
HCl to yield compounds A and P in good to excellent isolated
yields (Scheme 7).
À9
2
À1
Table 2. Diffusion coefficients, D (”10 m s ), and hydrodynamic radii,
(Š), values for titanocenes 3 and 9 at 80 mm at room temperature. The
D and r values for their corresponding precursors under the same condi-
tions are also provided.
r
H
H
9
2
À1 [a]
[b]
Entry
d [ppm]
Compound
D [10 m s
]
H
r [Š]
The electrophiles used to study the scope of the reactivity
of the intermediates involved in the metallotropic equilibrium
included aldehydes, ketones, and several protonating agents
of variable steric hindrance such as water, methanol, phenol, 2-
tert-butyl-4-methylphenol, and 2,6-di-tert-butyl-4-methylphe-
1
2
3
4
5
6
7
6.34, 5.99, 3.74
3.63
6.36, 2.22, 0.93
3.63
6.58
4.22
3
THF
9
THF
2 2
[TiCl Cp ]
1.295
2.593
1.087
2.582
1.402
2.712
1.656
3.5
1.7
4.1
1.7
3.2
1.6
2.7
ClCH
ClCH
2
CCH
CCPn
1
4.23, 2.26, 0.95
2
nol. The evaluation of the effect of the substituent R linked to
the triple bond was also monitored through the use of two
substrates.
[
a] Experimental error in D values is Æ2%. [b] The viscosity, h, used in the
À3 À1 À1
Stokes–Einstein equation was 0.4809”10 kgm
from www.knovel.com.
s . Viscosity was taken
The new derivatives 10–14 (Table 3) were isolated by
column chromatography and identified through their NMR
the pulsed field gradient stimulated spin echo (PFG-STE) diffu-
sion data for 80 mm [D ]THF samples of 3 and 9.
8
In the traditional Stejskal–Tanner plots (see the Supporting
[
34]
Information), the lower the attenuation, the lower the diffu-
sion coefficient, and the larger the molecular size. Although
the viscosity of the THF solutions varies with concentration
and type of analyte dissolved, we have used the viscosity of
the pure solvent for radii calculations using the Stokes–Einstein
equation. For comparison purposes, samples containing the ti-
tanium precursor [TiCl Cp ], propargyl chloride, and 1-chlor-
2
2
ooct-2-yne were also prepared at the same concentration and
their results are shown in Table 2 as well.
Scheme 7. Metallotropic rearrangement providing either alkyne (P) or allenic
(A) products after electrophilic quenching.
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2432
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
were performed to check if the metallotropic equilibrium
could be shifted towards the allenic derivative by just altering
the temperature. Unfortunately, the pinacol reaction is much
favored at these low temperatures, which inhibits further eval-
uation. In titanocene-based pinacol processes, it is generally
presumed that reduction of [TiCl Cp ] produces an odd-num-
Table 3. Stoichiometric reactions of the in situ formed 3 and 9 with car-
bonyl electrophiles and protonation reagents.
[
a]
1
2
3
4
[b]
Entry
R
R
R
R
Conv. [%]
Product Ratio (P/A)
1
2
3
4
5
6
7
8
9
H
H
H
H
H
H
–
–
–
–
–
–
–
–
–
H
H
H
3-F-Ph
3-F-Ph
3-F-Ph
>99
10P
–
–
11 P
–
–
12
13
14
15
15
15
15
15
>99
–
–
>99
–
–
50:50
78:22
75:25
26:74
28:72
31:69
28:72
–
[
c]
0
0
2
2
[
c]
bered-electron titanium(III) species, which is the active reduc-
tive coupling agent. This species, upon association with the
carbonyl substrate, induces CÀC bond formation (through di-
merization) to give a pinacolate, which is then hydrolyzed (see
Me 3-F-Ph
Me 3-F-Ph
Me 3-F-Ph
>99
[
d]
0
0
[
d]
Pn
Pn
Pn
Pn -H
Pn Me
Pn Ph
H
3-F-Ph
>99
>99
>99
>99
>99
>99
>99
0
Me 3-F-Ph
Me Et
[37]
the Supporting Information).
1
For internal alkynes (R =(CH ) CH ), the addition of 3-fluoro-
1
0
–
–
–
–
–
–
–
–
–
–
2 4
3
1
1
benzaldehyde yielded a conversion of the alkyne to the two al-
cohol products 12P and 12A with no regiochemical prefer-
ence for either product (entry 7, Table 3), which may suggest
an equilibrium and/or isomerization between the organotitani-
1
1
1
2
3
4
[
e]
Pn Ar
Pn Ar
[f]
[
[
a] Conditions: Deoxygenated, dry THF (10 mL) was added to a mixture of
TiCl Cp ] (2.0 mmol) and Mn dust (1.0 mmol) under a N atmosphere.
um intermediates, which is driven by k (Scheme 7). The use of
e
2
2
2
ketones, whether aliphatic (2-butanone) or aromatic (3-fluoro-
acetophenone), produced a significant change in the regiose-
lectivity of the process, again with a preference toward forma-
tion of the homopropargylic alcohols 13P and 14P (entries 8
and 9, Table 3). The unexpected difference in behavior be-
tween aldehydes and ketones was intriguing and encouraged
us to pursue reactions with protonating agents of variable
steric hindrance (entries 10–14, Table 3). Interestingly, the re-
gioselectivity outcome was reversed, obtaining in all cases the
allene product 15A as the major one. Interestingly, no reaction
was observed with the most hindered phenol (entry 14,
Table 3), which was probably due to the lack of coordination
of the hydroxyl group to titanium. This behavior is discussed
later in the DFT section. All the alkyne compounds 12–14P
were identified through the same set of NMR spectra men-
tioned above plus an interesting ABCD higher-order spin
The suspension was stirred at room temperature until it turned lime
1
green (about 2 h). Then, propargyl chloride (R =H, 1.0 mmol) or 1-chloro-
1
oct-2-yne (R =Pn=(CH
2
)
4
CH
3
, 1.0 mmol) was slowly added, and the was
mixture stirred for 30 min. The reaction was then quenched with the cor-
responding electrophile (1.1 mmol for aldehydes and ketones, and
1
19
5
mmol for ROH). [b] Determined by H and F NMR spectroscopy. [c] Re-
action performed at À208C. [d] Reaction performed at À808C. [e] Ar=2-
tBu-4-Me-Ph. [f] Ar=2,6-tBu-4-Me-Ph.
[
36]
data. Volatiles 15P and 15A (Table 3) were characterized di-
rectly in the crude reaction mixture through their most charac-
teristics NMR signals (see the Supporting Information). For ter-
1
minal alkynes (R =H) and by using 3-fluorobenzaldehyde as
the electrophile, the process induced >99% conversion of the
alkyne to just one alcohol product, which was identified as
a secondary homopropargylic alcohol 10P (entry 1, Table 3).
Next, the effect of using ketones was examined, as it has been
shown to be effective in promoting different reaction out-
comes in Barbier-type processes. Surprisingly, the process re-
mained regiospecific, providing the alkyne derivative 11 P with-
out the presence of any allene byproduct (entry 4, Table 3). Be-
cause both starting material and products contain fluorine, it
system observed for H1 and H4 at d =2.62 and 2.18 for 12P,
H
d =2.46 and 1.89 for 13P, and d =2.26 and 2.01 for 14P. In
H
H
13
addition, the C NMR data of the quaternary alkyne carbons in
the range d =75.7–84.36 ppm were diagnostic in their assign-
C
ment. On the other hand, allenic derivatives were easily identi-
fied by their carbon signals at d =204.3 (12A), 205.0 (13A),
C
19
was relatively easy to follow the reaction progress by F NMR
and 203.99 ppm (14A) together with the analysis of their
DEPT-135 NMR spectra.
spectroscopy. In this sense, the appearance of signals at d =
F
À112.62 and À112.75 ppm for 10P and 11 P helped in moni-
As mentioned before, the protonated derivatives 15A/15P
ratios (Table 3) were obtained by protonating reactions in J.
toring the reactions and quantifying the conversion before
1
13
1
work-up. A cursory inspection of the H and C NMR spectra
Young NMR tubes and their progress was analyzed by H NMR
1
allowed us to establish the expected structures. In the H NMR
spectroscopy (see the Supporting Information). All the alcohol-
ic sources provided similar ratios ranging from 26:74 to 31:69
for 15A/15P, respectively. When water, methanol, or phenol
were employed (entries 10–12, Table 3), the protonation pro-
ceeded instantly; however, when 2-tert-butyl-4-methylphenol
was the reagent of choice, the reaction needed about 10 min
(see the Supporting Information), and when the more hindered
2,6-di-tert-butyl-4-methylphenol (entry 14, Table 3) was em-
ployed, no reaction was observed after two days of stirring.
Mechanistically speaking, even though products P (when
carbonyls are employed), or products A (when a proton source
is added) are always formed as the major isomers, a variation
in the P/A ratio was observed depending on the electrophile.
spectrum, 10P was identified by a clear triplet at d =
H
3
4
.91 ppm ( J=6.3 Hz), assigned to the hydroxyl methine H1,
whereas 11 P showed a 3H integral singlet at d =1.31 ppm,
H
corresponding to the methyl attached to the same carbon as
1
3
the hydroxyl group. In the C NMR spectra, the quaternary sp-
hybridized alkyne carbons for 10P and 11 P appeared at d =
C
8
0.3–71.7 ppm, whereas the incorporation of the arylhydroxy-
methyl substituent was confirmed by the methine signal at
dC =71.4 and 71.9 ppm for 10P and 11 P, respectively.
When the reaction temperature was reduced to À208C or
À808C (entries 2–3 and 5–6, Table 3, respectively), full pinacoli-
zation of the aldehyde or ketone was observed. Such reactions
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2433
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
This variation is a good indication that the Curtin–Hammett
principle applies. This means that the metallotropic equilibrium
is faster than the reactions of 3 and 9 with electrophiles so
that P/A=k /k . Interestingly, to explain both the variation in
P
A
+
the P/A ratio and the predominant formation of P (E =
3
4
+
2
C(O)R R ) or A (E =R OH), the rate constants should be such
that in the first case k @k >k , whereas in the second case
e
P
A
k @k >k at room temperature in THF solution. Only when 3-
e
A
P
fluorobenzaldehyde is added in the case of the internal alkyne
entry 7, Table 3), are both rate constants k and k of similar
(
A
P
magnitude and, therefore, generate an equal amount of both
regioisomers. Based on these results, we propose that a dynam-
ic kinetic resolution is taking place, where the electrophilic
quenching is slower than the interconversion of the regioiso-
meric complexes. These results are in agreement with those
previously reported for 1-chloro-3-(trimethylsilyl)propargyl and
Scheme 8. Calculated metallotropic equilibrium of Ti complexes with R=H
À1
and Pn. DGTHF in kcalmol
.
-allenyl titanium derivatives, which also showed that these spe-
cies are configurationally stable at temperatures below
agreement with species 3. The conformer H-A’ is found at
[
15]
À1
À408C. There can, of course, be a combination of both ther-
modynamic and kinetic factors that contribute to the ratio of
the products, depending on the relative rates of these process-
1.9 kcalmol and the propargyl derivatives H-P and H-P’ are
À1
placed at 6.6 and 4.8 kcalmol above H-A (3), respectively. For
the rearrangement between the allenyl and propargyl forms,
fast interconversion between the prime and non-prime con-
formers (blue arrows in Scheme 8) is assumed.
[38]
es. However, low temperatures produced full pinacolization
of the carbonyl substrates, precluding the evaluation of such
experimental conditions. To fully confirm the dynamic kinetic
resolution hypothesis, DFT calculations have been carried out
and are discussed below.
À1
Accordingly, the prime route via H-TS-2’ (15.4 kcalmol ) is
slightly favored over the non-prime one via H-TS-2 (17.7 kcal
À1
mol ). The estimated Gibbs energy barrier for this process
À1
(
from H-A (3) to H-TS-2’) is 15.4 kcalmol . Contrary to R=H,
the most stable complex for R=Pn is the propargyl form Pn-P’
in agreement with species 9. Pn-A was set as the zero energy
point, with the conformer Pn-A’ placed at 1.1 kcalmol , but
DFT calculations
À1
For a better understanding of the stability and reactivity of the
previously discussed titanocene compounds, theoretical calcu-
both propargyl Pn-P and Pn-P’ (9) are now found at 2.2 and
[39]
À1
lations were performed at the DFT level
using the PBE0
4.3 kcalmol below Pn-A. Assuming fast interconversion be-
[
40]
[41]
À1
functional
supplemented with dispersion D3
(see the
tween conformers, the prime route via Pn-TS-2’ (6.8 kcalmol )
Computational Details). Firstly, the equilibrium between the al-
lenyl and propargyl species was studied to account for the ex-
perimentally observed allenyl titanium 3 (Scheme 5) and prop-
argyl titanium 9 (Scheme 6). Then, the reactivity of 3 and 9
with electrophiles was addressed. For the sake of comparison,
is hardly favored over the non-prime one via Pn-TS-1 (7.4 kcal
À1
mol ). The estimated Gibbs energy barrier for this process
À1
(from Pn-P’ to Pn-TS-2’) is 11.1 kcalmol .
Overall, the computational results for the metallotropic equi-
librium (Scheme 8) are in line with the allenyl 3 (Figure 6, left)
and propargyl 9 (Figure 6, right) detected experimentally.
3
-fluorobenzaldehyde has been selected as the electrophile as
1
13
it had been found to react with both 3 and 9 (entries 1 and 7,
Finally, the H and C NMR chemical shifts for H-A (3) and
Pn-P’ (9) have been calculated to fully characterize these struc-
Table 3). All energies reported herein correspond to Gibbs en-
À1
ergies in THF in kcalmol .
Metallotropic equilibrium
[42]
The computed metallotropic equilibrium between the allenic
and propargylic forms of the titanium complexes is shown in
Scheme 8. The prefixes H- and Pn- stand for H and (CH ) CH
3
2
4
(
n-pentyl) substituents, respectively. Prime suffixes indicate
that the substituent R is placed away from the chloride ligand;
otherwise, R points toward it. The allenyl species A has been
taken as the zero energy point for comparison with both set
of compounds.
From either allenyl A or A’, the formation of the correspond-
3
Figure 6. DFT-optimized structures of the most stable species for R=H, al-
ing propargyl P and P’ takes place via h -allenyl/propargyl in-
1
13
lenyl H-A (3), and R=Pn, propargyl Pn-P’ (9). H and C NMR chemical shifts
in ppm are shown in blue and black, respectively. Calculated values in italics,
experimental values underlined.
[43]
termediates I and I’ (Scheme 8). For R=H, the most stable
structure corresponds to the allenyl H-A (zero energy point) in
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2434
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
44]
À1
tures (see the Computational Details). These values (Figure 6)
agree with those experimentally reported in Table 1, showing
absolute errors in the range 0.19–1.21 and 1.0–8.9 ppm for the
rise to H-PE at 29.9 kcalmol , which evolves through the CÀC
À1
bond forming transition state H-TS-PE at 32.7 kcalmol (Fig-
ure S22) to form H-10A (Scheme 10, bottom). This trend favors
the formation of 10P as experimentally reported (entry 1,
Table 3). The relatively large Gibbs energy barriers are overesti-
1
13
[45,46]
H and C NMR chemical shifts, respectively.
[
50,51]
mated by the entropic contributions.
As a matter of fact,
when inspecting enthalpies, more accessible barriers of 14.4
Reactions with electrophiles
À1
and 19.0 kcalmol are found for H-TS-AE and H-TS-PE, respec-
Concerning the reaction of 3 and 9 with electrophiles (Table 3),
it is suggested that the allenyl/propargyl species may be in
tively. For R=Pn, similar intermediate and transition-state geo-
metries were found (Scheme 10, R=Pn). However, contrary to
R=H, the allenyl product Pn-12A (via Pn-TS-PE at 23.5 kcal
[47]
equilibrium prior to the electrophilic quenching. Although
the Gibbs energy barriers for the tautomerization could be low
À1
mol , Scheme 10, bottom) is favored over the propargyl deriv-
enough to invoke the Curtin–Hammett principle (15.4 and
À1
ative Pn-12P (via Pn-TS-AE at 27.2 kcalmol , Scheme 10, top).
À1
1
1.1 kcalmol for R=H and Pn, respectively), the reaction be-
The corresponding Gibbs energy barriers are 27.8 and 31.5 kcal
tween allenyl and propargyl Ti complexes with 3-fluorobenzal-
dehyde, E, has also been computed.
According to the experiments, only the propargyl product
À1
mol , respectively.
To check the sensitivity of the calculated mechanism to the
density functional, selected minima and all transition states
10P is observed for R=H (entry 1, Table 3), whereas a 50:50
[52,53]
(
R=H, Pn) were optimized by using M06,
and similar re-
mixture of allenyl 12A and propargyl 12P is obtained for R=
Pn (entry 7, Table 3). For ketones, the allenyl derivatives 13A
and 14A are found as major products. Acting through a S 2’
sults were obtained (Table S4 in the Supporting Information).
The feasibility of cationic intermediates from chloride re-
[54]
E
lease has also been considered. The reaction has been stud-
mechanism, the carbonyl group reacts with propargyl and al-
lenyl precursors, yielding allenyl and propargyl products, re-
ied with and without the participation of [MnCl ] by using the
2
PBE0-D3 and M06 functionals. See the Supporting Information
for further discussion. According to the above-mentioned sce-
nario (Scheme 10), the coordination of the electrophile to the
Ti center is required. Extrapolating this conclusion to protonat-
ing agents, the lack of reactivity of 2,6-tert-butyl-4-methylphe-
nol (entry 14, Table 3) can be attributed to the fact that the hy-
droxyl group cannot coordinate to Ti owing to the bulkiness of
the tert-butyl substituents. Indeed, no structures could be
found for such coordination for either the neutral and cationic
derivatives. All these results are summarized as Gibbs energy
profiles in Figure 7.
[13,48]
spectively (Scheme 9).
This mechanism has been studied
for both H and Pn substituents and is discussed next.
For R=H, the coordination of the aldehyde to H-A (3) gen-
À1
erates the intermediate H-AE at 23.8 kcalmol and the CÀC
À1
bond formation takes place via H-TS-AE at 28.4 kcalmol (Fig-
ure S22 in the Supporting Information) to form H-10P
[
49]
(
Scheme 10, top). The aldehyde coordination to H-P’ gives
For both systems, the allenyl-to-propargyl interconversion
À1
(
15.4 and 11.1 kcalmol for R=H and Pn, respectively) is
À1
faster than their reaction with E (28–32 kcalmol ). These re-
sults clearly support a dynamic kinetic resolution situation,
which is consistent with the observed configurational stability
of these titanium carbenoids.
Scheme 9. General S
carbonyl electrophiles and allenyl and propargyl ligands.
E
2’ mechanism for the CÀC bond formation between
Scheme 10. CÀC bond formation between allenyl/propargyl titanocenes and 3-fluorobenzaldehyde. DGTHF in kcalmolÀ1
.
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2435
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ferred onto 3 Š molecular sieves prior to use. Benzene-d6 and
CDCl were dried over CaH and vacuum transferred onto 3 Š mo-
3
2
lecular sieves prior to use. All other reagents and solvents were of
commercial quality and were used without further purification. TLC
was performed on Merck plates with aluminum backing and silica
gel 60 F254. For column chromatography, silica gel 60 (40–63 mm)
from Scharlau was used. NMR spectra were measured on a Bruker
1
13
19
Avance III 300 ( H, 300.13 MHz; C, 75.47 MHz; F, 282.40 MHz) or
a Bruker Avance III 500 spectrometer equipped with a third radio-
1
13
frequency channel ( H, 500.13 MHz; C, 125.76 MHz) using a 5 mm
BBFO H/BB( F) probe and an indirect 5 mm TBI 1H/ P/BB triple
probe, respectively. The spectral references used were internal tet-
ramethylsilane for H and C and an external reference of CF CO H
for F. Infrared spectra were recorded with a Bruker Alpha FTIR
spectrophotometer.
1
19
31
Figure 7. Simplified Gibbs energy profile for the equilibria between allenyl A
and propargyl P species and the subsequent reaction with 3-fluorobenzalde-
hyde E via neutral species.
1
13
3
2
19
Conclusions
Typical procedure for the reaction of titanium organometal-
lics with aldehydes and ketones
We have synthesized and structurally characterized for the first
time an allenyl titanocene(IV) [TiClCp (CH=C=CH )] (3) and
2
2
Deoxygenated THF (10 mL) was added to a mixture of commercial
[TiCl Cp ] (2.50 mmol) and Mn dust (1.25 mmol) under a nitrogen
a propargyl titanocene(IV) [TiClCp (CH ÀCꢀCÀ(CH ) CH )] (9).
2
2
2 4
3
2
2
These complexes were prepared by utilizing [TiCl Cp ], manga-
atmosphere and the suspension was stirred at room temperature
until it turned lime green (about 2 h). Then, the corresponding
alkyne derivative (1.25 mmol) was slowly added and the mixture
stirred for 30 min. Finally, the aldehyde or ketone (1.25 mmol) was
added and stirred at room temperature for 48 h. The reaction was
filtered and the volatiles were partially removed in vacuo. The re-
sulting solution was then quenched with 2N HCl and extracted
with Et O, washed with brine, dried (anhyd. Na SO ), and the sol-
2
2
nese as a reducing agent, and propargyl chloride or 1-chloro-
oct-2-yne for 3 and 9, respectively. Advanced NMR methods in-
cluding PFG-STE and DOSY experiments have been applied
and established that these organometallics are monomers in
1
13
1
THF solution. Full experimental and simulated H, C, D H, and
13
D C data are given providing, through the analysis of the
Ramsey equation, the first electronic insights into these deriva-
tives. Quenching experiments have proved they are involved
in their respective metallotropic equilibria, producing proparg-
yl products P (when carbonyls are employed) or allenyl prod-
ucts A (when a proton source is added) as the major isomers.
In all the cases assayed, the ratio of products suggested that
the metallotropic equilibrium is faster than the reactions of 3
and 9 with electrophiles, which has been supported by theory.
DFT calculations estimate Gibbs energy barriers of 11–15 kcal
2
2
4
vent removed in vacuo. Products were purified by flash chroma-
tography of the residues on silica gel (n-hexane/EtOAc mixtures).
Typical procedure for the reaction of titanium organometal-
lics with proton sources
An oven-dried Schlenk flask was charged with [TiCl Cp ]
2
2
(
0.25 mmol) and Mn dust (0.125 mmol) and evacuated in vacuo for
at least 3 h. The Schlenk flask was then transferred to a four-glove
drybox and deoxygenated [D ]THF (1 mL) added. The suspension
8
À1
À1
mol for the metallotropic equilibrium and 28–32 kcalmol
was stirred at room temperature until it turned lime green (about
2 h). Then, 1-chlorooct-2-yne (0.125 mmol) was added and the
whole mixture was stirred for 30 min. The resulting suspension was
filtered through a Millipore filter into a scintillation vial and cooled
at À358C for 1 h. The solution was filtered again through a Milli-
for the electrophilic quenching; thus, confirming a dynamic ki-
netic resolution of the configurationally stable intermediates.
We are currently investigating the synthesis of new mixed bi-
metallic complexes containing MnCl as new species to open
2
pore filter to get rid of some more MnCl and transferred into an
2
up new reactions.
oven-dried 5 mm J. Young NMR tube. The electrophilic quenching
was performed by adding the corresponding protonating agent
(
0.625 mmol) as specified in Table 3.
Experimental Section
Propyne and propadiene are commercially available and have
been previously characterized. Oct-2-yne (15P) and oct-1,2-diene
[55]
All manipulations were performed under an atmosphere of dry,
(15A) are commercially available and have been previously charac-
oxygen-free N by means of standard Schlenk or glovebox tech-
2
[13]
terized.
niques (Innovative Technology glovebox equipped with a À358C
freezer). Pentane, tetrahydrofuran, dichloromethane, and toluene
were dried by using an Innovative Technology solvent purification
system and degassed prior to use. Propargyl chloride and 1-chloro-
oct-2-yne were purchased from Sigma Aldrich and distilled and
stored over 3 Š molecular sieves prior to use. [TiCl Cp ] was pur-
1-(3-Fluorophenyl)but-3-yn-1-ol (10P):Yield after chromatography
(ethyl acetate/hexane 1:1) 79% (179 mg). Pale yellow oil. IR (film):
2
n=3430 (nOH, vs), 3300 (nCꢀC–H, s), 3077 (nCsp –H, m), 2982,
3
2928 (nCsp –H, s), 2119 (nCꢀC, w), 2060, 1942, 1863, 1589, 1486,
À1
1
1438, 1269 (nC–O, w), 1133, 882, 829, 788, 647 cm
; H NMR
2
2
chased from Sigma Aldrich and used without further purification.
-Fluorobenzaldehyde and 3-fluoroacetophenone were purchased
(300 MHz, CDCl ): d=7.39–7.26 (m, 1H, H9), 7.20–7.14 (m, 1H, H8),
3
3
7.14–7.12 (m, 1H, H10), 7.05–6.96 (m, 1H, H6), 4.91 (t, J=6.3 Hz,
1H, H1), 2.67 (m, 2H, H2), 2.36 (s, OH), 2.11 ppm (t, J=2.7 Hz, 1H,
from Alfa Aesar and distilled and stored over 3 Š molecular sieves
prior to use. Tetrahydrofuran-d ([D ]THF), [D ]benzene (C D ), and
1
3
1
H4); C NMR (75 MHz, CDCl ): d=163.1 (C, JC,F =250.1 Hz, C7),
8
8
6
6
6
3
3
3
chloroform-d (CDCl ) were purchased from Eurisotop. [D ]THF was
145.1 (C, J =6.9 Hz, C5), 130.0 (CH, J =8.3 Hz, C9), 121.3 (CH,
C,F C,F
JC,F =3.3 Hz, C10), 114.9 (CH, J =21.2 Hz, C6), 112.9 (CH, J =
C,F C,F
1
3
8
4
2
2
dried over K/benzophenone, heated at reflux, and vacuum trans-
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org 2436
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
3
2
2.1 Hz, C8), 80.3 (C, C3), 71.7 (CH, C1), 71.4 (CH, C4), 29.6 ppm
(CH, J =246.8 Hz, C12), 150.1 (CH, J =7.1 Hz, C10), 129.4 (CH,
C,F C,F
1
9
3
4
2
(
(
CH , C2); F NMR (282 MHz, CDCl ): d=À112.62 ppm (s); HRMS
JC,F =8.0 Hz, C14), 120.5 (CH, J =2.8 Hz, C15), 113.4 (CH, J
20.7 Hz, C11), 112.3 (CH, J =21.8 Hz, C13), 84.1 (C, C4), 75.7 (C,
C3), 72.7 (C, C1), 34.9 (CH , C7), 30.8 (CH , C6), 28.8 (CH , C16), 27.4
CH , C2), 22.1 (CH , C8), 18.4 (CH , C5), 13.7 ppm (CH , C9);
2 2 2 3
=
2
3
C,F
C,F
+
2
ES): m/z calcd for C H F: 147.0610 [MÀH O+H] ; found:
1
0
8
2
C,F
1
47.0601.
-(3-Fluorophenyl)pent-4-yn-2-ol (11P): Yield after chromatogra-
phy (ethyl acetate/hexane 1:1) 90% (222 mg). Colorless oil. IR
2
2
3
(
2
1
9
F NMR (282 MHz, C D ): d=À113.07 ppm (s); HRMS (ES): m/z
6
6
+
2
calcd for C H FO: 249.1649 [M+H] ; found: 249.1650. Identified
(
2
1
film): n=3392 (nOH, vs), 3298 (nCꢀC–H, s), 3071 (nCsp –H, m),
16 22
3
from a 7:93 mixture of 13P and 13A.
915 (nCsp –H, s), 2120 (nCꢀC, w), 1942, 1865, 1764, 1597, 1484,
À1
1
444, 1250 (nC–O, w), 1052, 878, 786, 646 cm ; H NMR (300 MHz,
2-(3-Fluorophenyl)-3-vinyliden octan-2-ol (13A): Yield after chro-
CDCl ): d=7.23–7.16 (m, 1H, H9), 7.02–6.97 (m, 1H, H8), 6.96–6.88
matography (ethyl acetate/hexane 1:1) 82% (271 mg) as a 22:78
3
1
(
m, 1H, H10), 6.77–6.69 (m, 1H, H6), 2.36–2.31 (m, 2H, H2), 2.29 (s,
mixture of 13P and 13A. Colorless oil. H NMR (300 MHz, C
6
D ):
6
13
OH), 1.66 (t, J=2.7 Hz, 1H, H4), 1.31 ppm (s, 3H, H12); C NMR
(
7
d=7.41–7.36 (m, 1H, H14), 7.21–7.14 (m, 1H, H13), 6.98–6.92 (m,
1H, H15), 6.78–6.72 (m, 1H, H11), 4.70 (t, J=3.8 Hz, 2H, H9), 1.92
(m, 1H, H2), 1.64 (m, 1H, H2), 1.46 (s, 3H, H16), 1.36 (m, 2H, H5),
1.17 (m, 2H, H4), 1.15 (m, 2H, H6), 0.80 ppm (t, J=7.1 Hz, 3H, H7);
1
3
75 MHz, CDCl ): d=163.1 (C, J =243.1 Hz, C7), 149.6 (C, J =
3 C,F C,F
3
4
.2 Hz, C5), 129.5 (CH, JC,F =8.2 Hz, C9), 120.4 (CH, JC,F =2.8 Hz,
2
2
C10), 113.6 (CH, JC,F =20.9 Hz, C6), 112.3 (CH, JC,F =22.0 Hz, C8),
0.2 (C, C3), 72.6 (C, C1), 71.9 (CH, C4), 34.3 (CH , C2), 28.6 ppm
1
3
1
8
(
(
C NMR (75 MHz, C
6
D
6
): d=205.0 (C, C8), 165.8 (CH,
J
J
C,F
=
=
2
1
9
3
3
CH , C12); F NMR (282 MHz, CDCl ): d=À112.75 ppm (s); HRMS
246.1 Hz, C12), 149.8 (CH, JC,F =6.5 Hz, C10), 129.4 (CH,
8.0 Hz, C14), 120.9 (CH, JC,F =3.1 Hz, C15), 113.4 (CH, JC,F =20.7 Hz,
C11), 112.6 (CH, JC,F =24.1 Hz, C13), 111.6 (C, C2), 78.9 (CH
C,F
3
3
+
4
2
ES): m/z calcd for C H F: 161.0767 [MÀH O+H] ; found:
11
10
2
2
1
61.0755.
, C9),
2
7
4.6 (C, C1), 31.4 (CH , C5), 30.7 (CH , C16), 27.5 (CH , C3), 26.6
2
3
2
1
-(3-Fluorophenyl)non-3-yne-1-ol (12P): Yield after chromatogra-
19
(
CH , C4), 22.5 (CH , C6), 14.2 ppm (CH , C7); F NMR (282 MHz,
2 2 3
phy (ethyl acetate/hexane 1:12) 28% (81 mg). The low isolated
yield was due to decomposition of the product during chromo-
graphic separation. Colorless oil. IR (film): n=3428 (nOH, vs), 3076
C D ): d=À113.05 ppm (s); HRMS (ES): m/z calcd for
6
6
+
C H FO:249.1649 [M+H] ; found: 249.1650. Identified from
16
22
2
3
a 7:93 mixture of 13P and 13A.
(
nCsp –H, m), 2928, 2859 (nCsp –H, s), 2289 (nCꢀC, w), 1980, 1932,
À1
1
869, 1814, 1592, 1449, 1254 (nC–O, w), 1045, 878, 787 cm
;
3-Methyl undec-5-yne-3-ol (14P): Yield after chromatography
(ethyl acetate/hexane 1:8) 67% (163 mg). Colorless oil. IR (film): n=
3398 (nOH, vs), 2964 (nCsp –H, s), 2930 (nCsp –H, s), 2866 (nCsp –
1
H NMR (300 MHz, C D ): d=7.37–7.30 (m, 1H, H14), 7.09–7.01 (m,
6
6
3
3
3
2
H, H13, H15), 7.02–6.95 (m, 1H, H11), 4.82 (s, OH), 2.62 (m, 2H,
À1
H2), 2.18 (m, 2H, H5), 1.50 (m, 2H, H6), 1.33 (m, 2H, H7), 1.34 (m,
2
d=161.5 (CH, J =247.8 Hz, C12), 145.7 (CH, J =7.4 Hz, C10),
H, m), 2228 (nCꢀC, w), 1459, 1375, 1150 (nC–O, m), 929 cm
;
1
3
1
H, H8), 0.91 ppm (t, J=6.9 Hz, 3H, H9); C NMR (75 MHz, C D ):
H NMR (300 MHz, C D ): d=2.26 (m, 2H, H4), 2.05–1.98 (m, 2H,
6 6
6
6
1
3
H7), 1.60–1.46 (m, 2H, H2), 1.45–1.31 (m, 2H, H8), 1.30–1.17 (m,
4H, H9, H10), 1.16 (s, 3H, H12), 0.86 (t, J=7.4 Hz, 3H, H11),
C,F
C,F
3
4
1
30.0 (CH, J =8.6 Hz, C14), 121.4 (CH, J =3.4 Hz, C15), 114.7
C,F
C,F
2
2
13
(
CH, J =20.7 Hz, C13), 112.8 (CH, J =22.1 Hz, C11), 84.3 (C, C4),
0.81 ppm (t, J=7.3 Hz, 3H, H1); C NMR (75 MHz, C
6
D
6
): d=83.0
, C4), 31.5
, C10), 14.0 (CH
, C1); HRMS (ES): m/z calcd for C12 23O: 183.1743
[M+H] ; found: 183.1744.
-Methyl-4-vinyliden nonan-3-ol (14A): Yield after chromatogra-
C,F
C,F
7
5.7 (C, C3), 72.0 (CH, C1), 31.1 (CH , C7), 30.2 (CH , C6), 28.9 (CH ,
(C, C5), 77.1 (C, C6), 71.6 (C, C3), 33.8 (CH
(CH , C9), 29.0 (CH , C8), 25.6 (CH , C12), 22.4 (CH
C11), 8.3 ppm (CH
2
, C2), 32.4 (CH
2
2
2
2
19
C2), 22.3 (CH , C8), 18.9 (CH , C5), 14.2 ppm (CH , C9); F NMR
2
2
3
2
,
3
2
2
3
(
282 MHz, C D ): d=À112.97 ppm (s); HRMS (ES): m/z calcd for
3
H
6
6
+
+
C H FO: 235.1493 [M+H] ; found: 235.1473.
15
20
3
1
-(3-Fluorophenyl)-2-vinyliden heptan-1-ol (12A): Yield after
phy (ethyl acetate/hexane 1:8) 23% (54 mg). Colorless oil. IR (film):
chromatography (ethyl acetate/hexane 1:12) 27% (80 mg). The low
isolated yield was due to decomposition of the product during the
chromographic separation. Colorless oil. IR (film): n=3428 (nOH,
n=3429 (nOH, vs), 3051 (nC=C=C–H, s), 1982 (nC=C=C, s), 2964,
À1
2
930 (nC–H, s), 1458, 1264 (nC–O, w), 1095, 1020, 802, 738 cm
;
1
3
H NMR (300 MHz, C D ): d=4.70 (t, J=3.7 Hz, 2H, H11), 1.95–1.86
vs), 2927, 2860 (nCsp –H, s), 1955 (nC=C=C, w), 1645, 1450, 1263
6
6
À1
1
(
2
m, 2H, H5), 1.63–1.52 (m, 2H, H2), 1.52–1.47 (m, 2H, H6), 1.29 (m,
H, H7), 1.28 (m, 2H, H8), 1.25 (s, 3H, H12), 0.89 (t, J=6.8 Hz, 3H,
(
nC–O, w), 1026, 739 cm ; H NMR (300 MHz, C D ): d=7.37–7.30
6 6
(
5
m, 1H, H14), 7.18–7.01 (m, 2H, H13, H15), 7.03–6.94 (m, 1H, H11),
.12 (s, 1H, H1), 5.02 (c, J=3.1, 5.2 Hz, 2H, H9), 1.91–7.72 (m, 2H,
H3), 1.42 (m, 2H, H4), 1.40 (m, 2H, H5), 1.38 (m, 2H, H6), 0.86 ppm
13
H9), 0.87 ppm (t, J=7.4 Hz, 3H, H1); C NMR (75 MHz, C D ): d=
6
6
2
03.9 (C, C10), 111.2 (C, C4), 78.8 (CH , C11), 72.5 (C, C3), 33.5 (CH ,
2 2
1
3
C2), 31.8 (CH , C7), 30.0 (CH , C6), 27.9 (CH , C1), 26.5 (CH , C5),
(
1
t, J=6.8 Hz, 3H, H7); C NMR (75 MHz, C D ): d=204.3 (C, C8),
2
2
3
2
6
6
1
3
2
2.6 (CH , C8), 14.0 (CH , C9), 8.1 ppm (CH , C1); HRMS (ES): m/z
2 3 3
63.3 (CH, J =253.8 Hz, C12), 145.0 (CH, J =6.8 Hz, C10), 129.6
C,F
C,F
+
3
4
calcd for C H O: 183.1743 [M+H] ; found: 183.1745.
(
CH, JC,F =7.9 Hz, C14), 122.3 (CH, JC,F =2.6 Hz, C15), 114.5 (CH,
JC,F =21.1 Hz, C11), 113.5 (CH, JC,F =21.1 Hz, C13), 108.3 (C, C2),
0.0 (CH , C9), 73.7 (CH, C1), 31.6 (CH , C5), 27.9 (CH , C3), 27.0
12 23
2
2
8
(
2
2
2
1
9
CH , C4), 22.3 (CH , C6), 14.1 ppm (CH , C7); F NMR (282 MHz,
2 2 3
C D ): d=À113.11 ppm (s); HRMS (ES): m/z calcd for C H FO:
6
6
15 20
+
Diffusion NMR details studies
2
35.1493 [M+H] ; found: 235.1483.
2
-(3-Fluorophenyl)dec-4-yne-2-ol (13P): Yield after chromatogra-
Pulsed gradient spin echo (PGSE) NMR diffusion measurements
were carried out by using the stimulated echo sequence and mo-
nopolar pairs pulses (stegp). A sine shape was used for the gra-
dient pulses and their strength varied automatically in the course
phy (ethyl acetate/hexane 1:1) 82% (271 mg) as a 22:78 mixture of
3P and 13A. Colorless oil. IR (film): n=3446 (nOH, s), 3067 (nC=
[56]
1
3
3
C=C–H, s), 2929 (nCsp –H, s), 2861 (nCsp –H, s), 2248 (nCꢀC, w),
1
1
7
6
3
0
954 (nC=C=C, w), 1593, 1442, 1370, 1334, 1241 (nC–O, w), 1114,
060, 933, 877, 785, 700 cm ; H NMR (300 MHz, C D ): d=7.30–
.26 (m, 1H, H14), 7.08–7.05 (m, 1H, H13), 6.98–6.92 (m, 1H, H15),
.78–6.72 (m, 1H, H11), 2.46 (m, 2H, H2), 1.89 (m, 2H, H5), 1.44 (s,
H, H16), 1.37 (m, 2H, H7), 1.27 (m, 2H, H6), 1.14 (m, 2H, H8),
of the experiments. The D values were determined from the slope
À1
1
2
of the regression line ln(I/I ) versus G , according to the following
6
6
0
[57]
equation, and by employing the DiffAtOnce package. I/I =ob-
0
served spin echo intensity/intensity without gradients, G=gradient
strength, D=delay between the midpoints of the gradients, D=
diffusion coefficient, d=gradient length.
1
3
.81 ppm (t, J=7.1 Hz 3H, H9); C NMR (75 MHz, C D ): d=165.8
6
6
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2437
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
I
d
4
p2
were computed according to the following formula, where the gp
2
2
Lnð Þ ¼ ÀðgdÞ ðD À ÞD
G
subscript indicates the gas-phase: DGTHF =DETHF +DG ÀDE
I
4
gp
gp
0
For the computation of NMR chemical shifts, the studied species
The measurements were carried out without spinning. Gradient
calibration was carried out by means of a diffusion measurement
and the reference SiMe
the gas phase by using the basis set 6–311+ +G(2d,2p)
atoms and the above-mentioned pseudopotential and associated
4
were optimized at the PBE0-D3 level in
[64,66]
for all
À9
2
À1 [58]
of HDO in D O (D(HDO)=1.902”10 m s ). To check reprodu-
2
[61,62]
cibility, three different measurements with different diffusion pa-
rameters (d and/or D) were always carried out. The experimental
error in the D values was estimated to be smaller than Æ2%. All of
the data leading to the reported D values afforded lines the corre-
lation coefficients of which were above 0.999, and 8 to 12 points
were used for regression analysis. To check reproducibility, three
different measurements with different diffusion parameters (d and/
or D) were always carried out. The gradient strength was incre-
mented in 4–8% steps from 2–10% to 98% so that, depending on
the signal-to-noise ratio, 12–25 points could be used for the re-
basis set for Ti.
Then, the chemical shielding tensors were com-
[67]
puted by using the GIAO method as implemented in Gaussian
[60]
[68]
09 by means of the B97–2 functional in THF solvent using
[65]
1
SMD. B97–2 is recommended for the computation of H and
1
3
[46]
C NMR chemical shifts. These chemical shifts were calculated as
d=s(SiMe
)Às(species), where s stands for the isotropic chemical
1 13
4
shielding of the H and C nuclei.
Acknowledgments
gression analysis. The recovery delay was set to 5 times T . All dif-
1
fusion processing and molecular size estimations were performed
by using the DiffAtOnce software package available at www.diffa-
tonce.com.
This work was supported by the Junta de Andalucía (Spain)
under the project number P12-FQM-2668 and by the Ministerio
de Economía y Competitividad under the project number
CTQ2011-24443. I.F. thanks the Spanish specialized group of
NMR (GERMN) for a travel grant. We are deeply grateful to
Prof. Agustí Lledós and Prof. Gregori Ujaque (Universitat Au-
tònoma de Barcelona) for the generous allocation of computer
time.
DOSY NMR data were acquired by using the stimulated echo se-
quence with bipolar pair pulses and longitudinal eddy current
[59]
delay (ledgpbp2s) with a total diffusion encoding pulse dura-
tion, d, of 2–4 ms, a diffusion delay, D, of 100–200 ms, and 12–24
nominal gradient amplitudes ranging from 5 to 95% of G to give
equal steps in gradient lineal. Experiments were carried out with-
out active temperature regulation, at the probe quiescent temper-
ature of 25Æ18C. DOSY spectra were constructed by standard
methods, by using fitting to a modified Stejskal–Tanner equation
parameterized to take into account the effects of pulsed field gra-
dient non-uniformity. DOSY spectra were processed and evaluated
by using the exponential fit implemented in Topspin 2.0. The diffu-
sion dimension was zero-filled to 1 K, sine window function with
line broadening of 0.5 Hz, and sine bell shift equal to zero. This di-
mension was exponentially fitted according to a preset window
Keywords: density functional calculations
·
DOSY
·
metallotropic equilibrium · NMR spectroscopy · titanocenes
[1] a) N. Petasis, Y. Hu, Curr. Org. Chem. 1997, 1, 249; b) D. J. Ramón, M. Yus,
Chem. Rev. 2006, 106, 2126; c) S. P. Morcillo, D. Miguel, A. G. CampaÇa,
L. lvarez de Cienfuegos, J. Justicia, J. M. Cuerva, Org. Chem. Front.
2014, 1, 15; d) A. Rosales, I. Rodríguez-García, J. MuÇoz-Bascón, E.
(
logD=À12.0 to À8.0). Gaussian line shapes were chosen to opti-
Roldan-Molina, N. M. Padial, L. Pozo-Morales, M. García-OcaÇa, J. E.
Oltra, Eur. J. Org. Chem. 2015, 4567.
mize the resolution of neighboring signals in F2 and no reference
deconvolution was used.
[
[
2] J. Murray, M. Harding, J. Med. Chem. 1994, 37, 1936.
3] A. Rosales, J. Oller-López, J. Justicia, A. Gansäuer, J. E. Oltra, J. M. Cuerva,
Chem. Commun. 2004, 2628.
[
[
[
4] J. Justicia, I. Sancho-Sanz, E. lvarez-Manzaneda, J. E. Oltra, J. M. Cuerva,
Adv. Synth. Catal. 2009, 351, 2295.
5] L. Sgreccia, M. Bandini, S. Morganti, A. Quintavalla, A. Umani-Ronchi,
P. G. Cozzi, J. Organomet. Chem. 2007, 692, 3191.
Computational details
[39]
All calculations were performed at the DFT level by using the
[40]
[41]
PBE0 functional supplemented with dispersion D3 as imple-
6] A. Gansäuer, M. Pierobon, H. Bluhm, Angew. Chem. Int. Ed. 2002, 41,
[
60]
mented in Gaussian09.
functional
For selected species, the M06 density
with an ultrafine grid was also employed. The Ti
atom was described by using the effective core potential SDD for
3206; Angew. Chem. 2002, 114, 3341.
[52,53]
[7] A. Gansäuer, A. Fleckhaus, M. A. Lafont, A. Okkel, K. Kotsis, A. Anoop, F.
Neese, J. Am. Chem. Soc. 2009, 131, 16989.
[8] M. Ishiguro, N. Ikeda, H. Yamamoto, J. Org. Chem. 1982, 47, 2225.
[9] a) T. Nakagawa, A. Kasatkin, F. Sato, Tetrahedron Lett. 1995, 36, 3207;
b) A. Kasatkin, F. Sato, Angew. Chem. Int. Ed. Engl. 1996, 35, 2848;
Angew. Chem. 1996, 108, 3024; c) F. Sato, S. Okamoto, Adv. Synth. Catal.
[61]
the inner electrons and its associated double-z basis set for the
[
62]
outer ones, complemented with a f-polarization function. The 6–
[63]
3
1G** basis set was used for the H, C, O, F, and Cl atoms. Diffuse
[64]
functions were added for O, F, and Cl in all calculations. Tetrahy-
drofuran (THF) solvent effects (e=7.426) were calculated by means
of single-point calculations on gas-phase optimized geometries by
2001, 343, 759.
[
10] a) S. Okamoto, D. K. An, F. Sato, Tetrahedron Lett. 1998, 39, 4551; b) S.
Okamoto, Y. Gosei, K. Kyokaishi, J. Synth. Org. Chem. 2001, 59, 1204;
c) C. Delas, S. Okamoto, F. Sato, Tetrahedron Lett. 2002, 43, 4373.
[11] Y. Yoshida, T. Nakagawa, F. Sato, Synlett 1996, 437.
[12] V. V. Burlakov, P. Arndt, W. Baumann, A. Spannenberg, U. Rosenthal, P.
Parameswaran, E. D. Jemmis, Chem. Commun. 2004, 2074.
13] J. MuÇoz-Bascón, I. Sancho-Sanz, E. lvarez-Manzaneda, A. Rosales, J. E.
Oltra, Chem. Eur. J. 2012, 18, 14479.
14] J. MuÇoz-Bascón, C. Hernµndez-Cervantes, N. M. Padial, M. lvarez-
Corral, A. Rosales, I. Rodríguez-García, J. E. Oltra, Chem. Eur. J. 2014, 20,
[65]
using the continuum model SMD. The structures of the reac-
tants, intermediates, transition states, and products were fully opti-
mized without any symmetry restriction. Transition states were
identified by having one imaginary frequency in the Hessian
matrix. It was confirmed that the transition states connect with the
corresponding intermediates by application of an eigenvector cor-
responding to the imaginary frequency and subsequent optimiza-
tion of the resulting structures.
[
[
801.
All reported values in the text correspond to Gibbs energies
[15] J. Bejjani, C. Botuha, F. Chemla, F. Ferreira, S. Magnus, Organometallics
2012, 31, 4876.
À1
(
298 K) in THF in kcalmol . These Gibbs energies in THF solution
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2438
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3
IV
[
16] As pointed out by one of the reviewers, it is worth mentioning that the
[43] For related h -allenyl/propargyl Zr derivatives, see: a) P. W. Blosser, J. C.
Gallucci, A. Wojcicki, J. Am. Chem. Soc. 1993, 115, 2994; b) P. W. Blosser,
J. C. Gallucci, A. Wojcicki, J. Organomet. Chem. 2000, 597, 125; c) D. R.
Denomme, S. M. Dumbris, I. F. D. Hyatt, K. A. Abboud, I. Ghiviriga, L.
McElwee-White, Organometallics 2010, 29, 5252.
III
Ti /propargyl combination reported herein is a more convenient way of
IV
preparing allenyl Ti complexes than previous procedures such as
those described by Chemla and Ferrerira in reference [15].
1
[
[
17] For a review of h -propargyl/allenyl chemistry, see: A. Wojcicki, C. E.
Shuchart, Coord. Chem. Rev. 1990, 105, 35.
[44] M. W. Lodewyk, M. R. Siebert, D. J. Tantillo, Chem. Rev. 2012, 112, 1839.
[45] The larger errors come from the allenic moieties, which are known to
be difficult to predict (see, for instance, reference [49]).
13
18] Unfortunately, no C NMR data have been found in the literature for
1
such complexes. For H NMR chemical shifts and synthetic procedures,
see: a) J. E. Thomasson, P. W. Robinson, D. A. Ross, A. Wojcicki, Inorg.
Chem. 1971, 10, 2130; b) J. L. Roustan, P. Cadiot, C. R. Seances Acad. Sci.
Ser. C 1969, 268, 734.
[46] D. Flaig, M. Maurer, M. Hanni, K. Braunger, L. Kick, M. Thubauville, C.
Ochsenfeld, J. Chem. Theory Comput. 2014, 10, 572.
[47] H. M. Wisniewska, E. R. Jarvo, J. Org. Chem. 2013, 78, 11629.
[48] G. Fan, X. Xie, Y. Liu, Y. Li, Organometallics 2013, 32, 1636.
[49] Another conformation can be described in which the allenyl ligand is
placed between chloride and aldehyde ligands, but the resulting transi-
[
[
[
19] F. Ungvµry, A. Wojcicki, J. Organomet. Chem. 1990, 396, 95.
20] R.-S. Keng, Y.-C. Lin, Organometallics 1990, 9, 289.
21] C. E. Shuchart, R. R. Willis, A. Wojcicki, J. Organomet. Chem. 1992, 424,
À1
1
85.
22] C. J. Elsevier, H. Kleijn, J. Boersma, P. Vermeer, Organometallics 1986, 5,
16.
tion state H-TS-AE’ is found at higher energies (32.9 kcalmol ).
[
[
[50] S. P. Morcillo, A. Martínez-Peragón, V. Jakoby, A. J. Mota, C. Kube, J. Justi-
cia, J. M. Cuerva, A. Gansäuer, Chem. Commun. 2014, 50, 2211.
[51] For a detailed discussion, see: R. E. Plata, D. A. Singleton, J. Am. Chem.
Soc. 2015, 137, 3811.
7
23] a) B. E. Mann, B. L. Shaw, N. I. Tucker, J. Chem. Soc. A 1971, 2667; b) P. W.
Blosser, D. G. Schimpff, J. C. Gallucci, A. Wojcicki, Organometallics 1993,
1
2, 1993; c) M. W. Baize, P. W. Blosser, V. Plantevin, D. G. Schimpff, J. C.
[52] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215.
[53] a) Y. Zhao, D. G. Truhlar, Acc. Chem. Res. 2008, 41, 157; b) Y. Zhao, D. G.
Truhlar, Chem. Phys. Lett. 2011, 502, 1.
[54] A. Cangçnül, M. Behlendorf, A. Gansäuer, M. van Gastel, Inorg. Chem.
2013, 52, 11859.
Gallucci, A. Wojcicki, Organometallics 1996, 15, 164.
[
24] a) R. Benn, in Transition Metal Nuclear Magnetic Resonance, (Ed.: P. S. Pre-
gosin), Elsevier, New York, 1991; b) P. S. Pregosin, NMR in Organometallic
Chemistry, Wiley-VCH, Weinheim, 2012.
[
[
25] K. P. Bryliakov, E. P. Talsi, M. Bochmann, Organometallics 2004, 23, 149.
26] J. F. Payack, D. L. Hughes, D. Cai, I. F. Cottrell, T. R. Verhoeven, Org. Synth.
[55] N. J. R. van Eikema Hommes, T. Clark, J. Mol. Model. 2005, 11, 175.
[56] D. H. Wu, A. D. Chen, C. S. Johnson, J. Magn. Reson. Ser. A 1995, 115,
260.
2002, 79, 19.
[
[
27] P. F. Godenschwager, D. B. Collum, J. Am. Chem. Soc. 2008, 130, 8726.
28] a) A. Macchioni, G. Ciancaleoni, C. Zuccaccia, D. Zuccaccia, Chem. Soc.
Rev. 2008, 37, 479; b) R. Evans, Z. Deng, A. K. Rogerson, A. S. McLachlan,
J. J. Richards, M. Nilsson, G. A. Morris, Angew. Chem. Int. Ed. 2013, 52,
[57] DiffAtOnceꢁ is a registered program developed by I. Fernµndez and
F. M. Arrabal-Campos at the University of Almería, 2013.
[58] H. J. V. Tyrrell, K. R. Harris, Diffusion in Liquids, Butterworths, London,
1984.
3
2
199; Angew. Chem. 2013, 125, 3281; c) R. Neufeld, D. Stalke, Chem. Sci.
015, 6, 3354.
[59] S. J. Gibbs, C. S. Johnson, J. Magn. Reson. 1991, 93, 395.
[60] Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr. , J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian, Inc. Wallingford CT, 2009.
[
[
29] a) A. Macchioni, Chem. Rev. 2005, 105, 2039; b) P. S. Pregosin, P. G. A.
Kumar, I. Fernµndez, Chem. Rev. 2005, 105, 2977; c) A. Moreno, P. S. Pre-
gosin, L. F. Veiros, A. Albinati, S. Rizzato, Chem. Eur. J. 2009, 15, 6848.
30] a) Y. Cohen, L. Avram, L. Frish, Angew. Chem. Int. Ed. 2005, 44, 520;
Angew. Chem. 2005, 117, 524; b) A. Pastor, E. Martinez-Viviente, Coord.
Chem. Rev. 2008, 252, 2314; c) S. Han, Z. Ma, R. Hopson, Y. Wei, D. Budil,
S. Gulla, B. Moulton, Inorg. Chem. Commun. 2012, 15, 78.
31] a) K. Schober, E. Hartmann, H. Zhang, R. M. Gschwind, Angew. Chem. Int.
Ed. 2010, 49, 2794; Angew. Chem. 2010, 122, 2855; b) P. Horeglad, A. Lit-
winska, G. Z. Zukowska, D. Kubicki, G. Szczepaniak, M. Dranka, J. Za-
chara, Appl. Organomet. Chem. 2013, 27, 328.
[
[
[
32] D. T. Thielemann, A. T. Wagner, Y. Lan, P. OÇa-Burgos, I. Fernµndez, E.
Rçsch, D. K. Kçlmel, A. K. Powell, S. Bräse, P. W. Roesky, Chem. Eur. J.
2
015, 21, 2813.
[61] D. Andrae, U. Haeussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim.
Acta 1990, 77, 123.
33] D=kT/(6phr ), where r is the hydrodynamic radius, D is the diffusion
H
H
coefficient, k is the Boltzmann constant, T is the absolute temperature,
and h is the solvent viscosity.
34] E. O. Stejskal, J. E. Tanner, J. Chem. Phys. 1965, 42, 288.
[62] A. W. Ehlers, M. Bohme, S. Dapprich, A. Gobbi, A. Hollwarth, V. Jonas,
K. F. Kohler, R. Stegmann, A. Veldkamp, G. Frenking, Chem. Phys. Lett.
1993, 208, 111.
[
[
35] a) A. Spannenberg, F. G. Kirchbauer, V. V. Burlakov, U. Rosenthal, Z. Kris-
tallogr. New Cryst. Struct. 2001, 216, 619; b) G. Fachinetti, C. Floriani, H.
Stoeckli-Evans, J. Chem. Soc. Dalton Trans. 1977, 2297; c) G. Erker, R.
Schlund, C. Krueger, J. Organomet. Chem. 1980, 338, C4; d) J. E. John,
C. R. Karen, W. Stefan, K. Carl, Organometallics 1991, 10, 3417; e) A. B.
Nicholas, B. K. Alan, C. R. Ian, G. P. Robin, E. W. John, Organometallics
[63] a) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56, 2257;
b) M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D. J.
DeFrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654.
[64] T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. v. R. Schleyer, J. Comput.
Chem. 1983, 4, 294.
[65] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378.
2
004, 23, 2680.
1
13
[
36] The structural assignment was based on the analysis of the 1D ( H, C,
DEPT-135, selective TOCSY) and 2D (gHMQC and gHMBC) NMR spectra.
37] Y. Handa, J. Inanaga, Tetrahedron Lett. 1987, 28, 5717.
[66] a) R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980,
72, 650; b) A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639;
c) M. J. Frisch, J. A. Pople, J. S. Binkley, J. Chem. Phys. 1984, 80, 3265.
[67] a) F. London, J. Phys. Radium 1937, 8, 397; b) R. McWeeny, Phys. Rev.
1962, 126, 1028; c) R. Ditchfield, Mol. Phys. 1974, 27, 789; d) K. Wolinski,
J. F. Hilton, P. Pulay, J. Am. Chem. Soc. 1990, 112, 8251; e) J. R. Cheese-
man, G. W. Trucks, T. A. Keith, M. J. Frisch, J. Chem. Phys. 1996, 104,
5497.
[
[
38] P. Beak, D. R. Anderson, M. D. Curtis, J. M. Laumer, D. J. Pippel, G. A. Wei-
senburger, Acc. Chem. Res. 2000, 33, 715.
39] C. J. Cramer, D. G. Truhlar, Phys. Chem. Chem. Phys. 2009, 11, 10757.
40] a) J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865;
b) J. P. Perdew, M. Ernzerhof, K. Burke, J. Chem. Phys. 1996, 105, 9982;
c) C. Adamo, V. Barone, J. Chem. Phys. 1999, 110, 6158.
[
[
[68] P. J. Wilson, T. J. Bradley, D. J. Tozer, J. Chem. Phys. 2001, 115, 9233.
[
[
41] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
1
54104.
42] H. Zhang, X. Fu, J. Chen, E. Wang, Y. Liu, Y. Li, J. Org. Chem. 2009, 74,
351.
Received: October 25, 2015
Published online on January 20, 2016
9
Chem. Eur. J. 2016, 22, 2427 – 2439
www.chemeurj.org
2439
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim