On the mechanism of the oxygen transfer to sulfoxides by (peroxo)[tris(hydroxyalkyl)amine]TiIV complexes - Evidence for a metal-template-assisted process

Article abstract of DOI:10.1002/ejoc.200390087

Oxidation of p-XC₆H₄SOMe (X = NC, F₃C) by (alkylperoxo)-(tris[(2R)-2-hydroxy-2-phenylethyl]amine}titanium (2b) has been shown to follow Michaelis-Menten kinetics demonstrating the occurrence of an intramolecular nucleophilic oxygen transfer to the Ti^IV-coordinated sulfoxide. The reaction of the titanium(IV) precursors 1 with sulfoxides has been studied by ESI-MS techniques together with ab initio calculations. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejoc.200390087

FULL PAPER
On the Mechanism of the Oxygen Transfer to Sulfoxides by
(Peroxo)[tris(hydroxyalkyl)amine]TiIV Complexes ؊ Evidence for a
Metal-Template-Assisted Process
Marcella Bonchio,*[a] Olga Bortolini,[b] Giulia Licini,*[a] Stefano Moro,[c] and
William A. Nugent[d]
Keywords: Homogeneous catalysis / Oxidations / Titanium / Sulfoxides / N,O ligands
Oxidation of p-XC6H4SOMe (X = NC, F3C) by (alkylperoxo)-
{tris[(2R)-2-hydroxy-2-phenylethyl]amine}titanium (2b) has
tion of the titanium(IV) precursors 1 with sulfoxides has been
studied by ESI-MS techniques together with ab initio calcu-
been shown to follow Michaelis−Menten kinetics demon- lations.
strating the occurrence of an intramolecular nucleophilic
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
oxygen transfer to the TiIV-coordinated sulfoxide. The reac- Germany, 2003)
Introduction
ide functionality[5,6] towards sulfoxides, the occurrence of
an intramolecular nucleophilic oxygen transfer step (kIN
)
triggered by coordination of the substrate to the titanium
We have recently reported a highly efficient enantioselec-
tive oxidation of organic sulfur compounds with (alkylper-
oxo)TiIV complexes containing C3-symmetric tris(hydroxy-
alkyl)amines as ligands, reaching an enantiomeric excess of
up to 84% and a turnover of up to 1000.[1,2] The stereoselec-
tion of the process is based on two independent oxidations
cooperating in building up the same enantiomer: namely
the direct asymmetric oxidation of the sulfide to the sulfox-
ide and its subsequent kinetic resolution by further oxida-
tion to the sulfone.[3] In this reaction the title peroxides fol-
low a biphilic mechanism.[4] While they behave as elec-
trophiles towards thioethers, as expected within the class
of d0 transition metal peroxides, an atypical nucleophilic
pathway with regard to sulfoxides emerges from Hammett
LFER studies.[2] For explaining the umpolung of the perox-
center is considered (Scheme 1).[4,7Ϫ12]
[a]
`
Universita di Padova, Dipartimento di Chimica Organica,
CMRO del CNR,
via Marzolo, 35131 Padova, Italy
Fax: (internat.) ϩ 39-049/8275239
E-mail: giulia.licini@unipd.it; marcella.bonchio@unipd.it
[b]
[c]
[d]
`
Universita di Ferrara, Dipartimento di Chimica,
via Borsari 46, 44100 Ferrara, Italy
Fax: (internat.) ϩ 39-0532/240709
E-mail: brl@unife.it
Scheme 1. Mechanism of the [C3-tris(hydroxyalkyl)amine]TiIV-me-
diated oxidation of sulfoxides by alkyl hydroperoxides (R1
ϭ
`
CPhMe2, iPr; R2 ϭ Me, CD3, nBu; R3 ϭ Ar, Me, CD3, nBu)
Universita di Padova, Dipartimento di Scienze Farmaceutiche,
via Marzolo 5, 35131 Padova, Italy
Fax: (internat.) ϩ 39-049/8275366
E-mail: stefano.moro@unipd.it
Bristol-Myers Squibb Company, Process Research and
Development Department,
In particular, assuming the existence of sulfoxide equilib-
ria, the solution speciation of the titanium complexes
should include precursors 1 and 4 and two diverse perox-
ides, namely complexes 2 and 3. Therefore, at least three
P. O. Box 269, Deepwater, NJ 08023, U.S.A.
E-mail: william.nugent@bms.com
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2003, 507Ϫ511  2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/03/0203Ϫ0507 $ 20.00ϩ.50/0
507

M. Bonchio, O. Bortolini, G. Licini, S. Moro, W. A. Nugent
FULL PAPER
different modes of oxidation may be envisaged to account dissociation constant of the reactive complex 3b (R1
ϭ
for the type of reactivity observed: (i) a bimolecular elec- Me2PhC; R2 ϭ p-XC6H4; R3 ϭ Me), respectively. The com-
trophilic oxygen transfer (kE) by the parent peroxide;[13] (ii) parison of Vmax and KM values with the Hammett substitu-
an inner-sphere nucleophilic oxygen transfer (kIN) proceed- ent constant (σ),[19,20] which establishes a relative estimate
ing in complex 3 at the level of the Lewis acid activated of the nucleophilic nature of the substrate, is shown in Fig-
sulfoxide; and (iii) a bimolecular (outer-sphere) electrophilic ure 1.
oxidation (kEE) performed by 3 at non-ligated sulfoxides.
According to this mechanism, the electronic nature of the
sulfoxide and its interaction with the titanium atom play a
key role in tuning the catalytic activity of the peroxide[2,4]
and the global stereoselectivity of the process, as the over-
oxidation of the sulfoxide to the sulfone is pivotal in enhan-
cing the enantiomeric excess of the sulfoxide by kinetic res-
olution. We addressed this issue through a series of kinetic
and mass spectrometric studies in order to verify: (i) the
observation of MichaelisϪMenten behavior in the oxida-
tion of appropriate substrates demonstrating fast reversible
formation of complex 3 prior to the rate-limiting oxygen
transfer;[14] and (ii) the structural modification of 1 induced
by the sulfoxide binding. For this latter purpose, the use
of electrospray ionization mass spectrometry (ESI-MS)[15,16]
has allowed the monitoring of the formation of DMSO ad-
ducts with the chiral titanatrane unit in solution, whose
structure and properties have also been studied employing 1b (R1 ϭ iPr) cumyl hydroperoxide (ref.[17]
Figure 1. Experimental and calculated plot of initial velocities
versus initial sulfoxide concentration for the catalytic oxidation of
p-NC-C6H4-SO-Me (᭹) and p-F3C-C6H4-SO-Me (᭿) performed by
)
ab initio calculations.
Accordingly, it is possible to confirm the nucleophilic
character of the oxygen transfer step as the p-NC-substi-
tuted sulfoxide turns out to be the more reactive substrate.
Results and Discussion
The first positive evidence that MichaelisϪMenten kin- Also, the highest value of binding constant (ca. 1/KM) is
etic behavior might be exhibited in the reaction under study expected for the more basic sulfoxide, as is indeed observed.
was obtained from the oxidation of different para-substi-
Beside the kinetic studies, further evidence concerning
tuted aryl methyl sulfoxides catalyzed by 1b (R1 ϭ iPr) in the coordination of sulfoxides to catalyst 1 has been ob-
the presence of cumyl hydroperoxide.[4] A set of experiments tained by ESI-MS after speciation of the titanatrane com-
designed to evaluate the dependence of the initial rate of plexes in the presence of DMSO. To the best of our know-
oxidation (Ro) on the initial substrate concentration showed ledge, there are only two reports dealing with DMSO coor-
a marked deviation from linearity, thus indicating a plaus- dination to TiIV complexes. Two X-ray structures of cat-
ible saturation behavior of the TiIV peroxide with respect to ionic TiIV complexes O-coordinated to five {as in [TiOCl2-
the sulfoxide. In our previous work, we also realized that (DMSO)5]} or three {as in Ti4O6(DMSO)12]Cl4]} O-bonded
the oxidation of electron-rich aryl methyl sulfoxides, carry- DMSO molecules have been solved,[21] and the formation
ing the electron-donating substituents p-Me2N, p-MeO, or of four mononuclear, octahedral [Ti(OiPr)Cl3(PhCHO)-
p-Me, was complicated by the presence of different concur- (DMSO)] complexes, which rapidly interconvert, has been
1
rent oxidation pathways, as the oxygen transfer in the inner- detected by VT H NMR experiments after addition of 2
sphere (kIN) was accompanied by a residual reactivity invol- equiv. of DMSO to [Ti(OiPr)Cl2(PhCHO)(µ-Cl]2.[22,23]
ving the non-coordinated sulfoxide (kEE).[2,4]
The ESI-MS measurements were initially performed with
Therefore, in order to further examine the relationship complex 1a (R1 ϭ iPr) in acidic chloroform as mobile phase
between the electronic demand of the oxygen transfer step and recording the positive ion mass spectra upon addition
and substrate binding to the TiIV center, more detailed stud- of increasing amounts of DMSO.[24] Under the ESI-MS
ies using MichaelisϪMenten analysis have been undertaken conditions, addition of a large excess of DMSO (Ͼ 1000
for the oxidation of the electron-poor sulfoxides p- equiv.) leads to the appearance of a new peak at m/z ϭ 314
XC6H4SOMe (X ϭ NC, F3C).[17] Figure 1 displays the ini- (ion I) with a relative intensity of 6Ϫ8% of the base peak at
tial velocities for the reaction of these substrates with 1b m/z ϭ 724 of the trinuclear ion {[TiN(CH2CHMeO)3]3Oϩ},
(R1 ϭ iPr)/cumyl hydroperoxide plotted as a function of the associated to the parent catalyst 1a (R1 ϭ iPr).[25] The ex-
initial sulfoxide concentration.[18]
perimental cluster of I corresponds to the mononuclear cat-
The mathematical interpolation of the experimental ionic species [TiN(CH2CHMeO)3(DMSO)]ϩ (Figure 2),
curves, based on the MichaelisϪMenten model, provides likely derived from the neutral precursor 4a (R1 ϭ iPr) by
good fitting of Vmax and KM values associated to the rate loss of the alkoxide ligand and revealed as such under posit-
of the inner-sphere oxidation of the sulfoxide and to the ive ESI-MS mode.[26]
508
Eur. J. Org. Chem. 2003, 507Ϫ511

Oxygen Transfer to Sulfoxides by (Peroxo)[tris(hydroxyalkyl)amine]TiIV Complexes
FULL PAPER
Figure 2. Data for ESI-MS experiments on complexes 1aϪc (R1 ϭ
iPr) after addition of dialkyl sulfoxides (Ͼ 1000 equiv.)[24]
The structure of ion I was further confirmed by labelling
experiments. In fact, addition of [D6]DMSO produced a
peak at m/z ϭ 320 (ion II) with six extra mass units. A
sulfoxide adduct was also obtained in the presence of di-n-
Figure 3. B3LYP/6-311ϩG(d,p)-optimized geometries of ion A and
B; energies are reported in atomic units (1 au ϭ 2612.9 kJ·molϪ1);
butyl sulfoxide (ion III, m/z ϭ 398). Analogous results were
obtained by using complexes 1b (R1 ϭ iPr) and 1c (R1 ϭ total binding energy has been calculated as ∆E(AϪB) ϭ EA Ϫ EB
in kJ·molϪ1
iPr) and DMSO, leading to the observation of ions at
m/z ϭ 500 (IЈ) and m/z ϭ 272 (IЈЈ), respectively (Figure 2).
Tandem MS2 analysis on ions I, IЈ, IЈЈ, II, and III re-
The calculated TiϪO (DMSO) bond length is shorter
(1.98 A) in structure A than that experimentally found for
vealed that, upon collision-induced dissociation (CID), a
common fragmentation pattern could be observed, where
the major peak originates from the loss of one aldehyde
molecule (Mϩ Ϫ RЈCHO)[27] while the sulfoxide remains
coordinated to the titanium atom (Figure 2).[28]
˚
the other Ti(DMSO) complexes by X-ray analysis (Ͼ 2.00
[21,22]
˚
A).
Moreover, we observed a strong SϪO elongation
with respect to free DMSO caused by the coordination to
The binding mode of the sulfoxide[29,30] to the titanatrane
unit has also been addressed using the computational ap-
proach by carrying out ab initio calculations. As already
reported, ab initio studies of sulfur-containing molecules
are difficult and often show large basis-set effects.[31,32]
Large split-valence basis sets, such as 6-311ϩG(d,p) or 6-
311ϩG(2df,2p), were often required for obtaining reliable
equilibrium structures.[33,34] Moreover, density functional
theory methods have been shown recently to provide accur-
ate equilibrium geometries and good harmonic vibrational
frequencies for a number of sulfur-containing molecules
and ions.[34Ϫ36] In our computational studies, we combined
Becke’s hybrid functional (B3LYP) with the 6-311ϩG(d,p)
basis set to obtain reliable equilibrium structures of sulfox-
ide/titanatrane complexes. The B3LYP/6-311ϩG(d,p) relat-
ive energies have also been compared with Hay and Wadt’s
effective core potentials (ECP) applied in DFT calculations.
Both methods gave comparable results (Figure 3). The
optimized geometry obtained for ion IЈЈ with the DMSO
residue through O or S coordination (structures A and B,
respectively) is given in Figure 3, together with the calcu-
lated energies.
the titanatrane cation. In fact, the SϪO distance changes
˚
from an average value of 1.492 A in free sulfoxides to that
[21]
˚
of 1.571 A in the O-coordinated isomer.
This result is
consistent with the peculiar robustness of the sulfoxide li-
gand as observed in the MS2 fragmentation mode of the
titanatrane adducts. On the contrary, the S-coordinated iso-
˚
mer displays a very long TiϪS distance (2.573 A), and con-
˚
sequently the calculated SϪO distance (1.475 A) is found
to be much shorter than in free DMSO.
Conclusion
In summary, we have delineated a unique titanium tem-
plate oxidation mode, which is dominant in the case of elec-
tron-poor aryl sulfoxides. The study of the stereoelectronic
nature of the intermediates involved is warranted for ex-
panding the reaction scope and understanding the mechan-
istic basis for enantioselectivity in the kinetic resolution.
Towards this latter aim, a detailed examination of the bind-
ing and oxidation kinetic of enantiopure sulfoxides will be
pursued in the future.
Interestingly, the calculated energy difference between the
O- and S-coordinated isomers shows that DMSO-S isomer
is less stable by ca. 230 kJ·molϪ1 with respect to the
DMSO-O one, using both B3LYP/6-311ϩG(d,p) and
B3LYP/LANL2DZ calculations. This supports the proposal
that the TiϪS bond trans to the nitrogen atom of the tris-
(hydroxyalkyl)amine moiety is destabilized in favor of O-
bonding of the DMSO-O isomer.
Experimental Section
1
General Remarks: H NMR spectra were recorded with a 200- or
250-MHz instrument. Kinetic runs were monitored by quantitative
GC analysis with an SE-30 15 m ϫ 0.25 mm (i.d.) capillary column,
using 4-methylbenzophenone as internal standard. ESI-MS and 1H
NMR experiments were performed as already reported.[15] HPLC
Eur. J. Org. Chem. 2003, 507Ϫ511
509

M. Bonchio, O. Bortolini, G. Licini, S. Moro, W. A. Nugent
FULL PAPER
[6]
[7]
M. Selke, J. S. Valentine, J. Am. Chem. Soc. 1998, 120,
2652Ϫ2653.
grade solvents were generally used. [D6]DMSO (Fluka) was used
in the labelling experiments. 1,2-Dichloroethane (DCE) was
washed three times with 10% concentrated H2SO4 and several times
with water until pH ϭ 7, dried overnight with CaCl2, distilled from
P2O5, and stored over molecular sieves. Cumyl hydroperoxide (80%,
Fluka) was stored over molecular sieves at 0 °C. Titanium() isop-
ropoxide (Aldrich) was distilled under vacuum (b.p. 60Ϫ63 °C/
0.1 Torr). Racemic methyl p-(trifluoromethyl)phenyl sulfoxide[37]
and p-cyanophenyl methyl sulfoxide[38] were prepared by oxidation
of the corresponding sulfides.[39] Enantiopure tris(hydroxyalkyl)a-
mines 1a,b were prepared according to the literature procedure.[40]
Tris(2-hydroxyethyl)amine (Aldrich) was distilled under reduced
pressure (b.p. 190Ϫ193 °C/5 Torr).
Template oxygen transfer by peroxo d0 complexes has also been
proposed by other authors (see refs.[8Ϫ12]).
[8]
W. A. Herrmann, R. V. Fischer, J. D. G. Correia, J. Mol. Catal.
1994, 94, 213Ϫ223.
M. M. Abu-Omar, J. H. Espenson, Organometallics 1996, 15,
3543Ϫ3549.
[9]
[10]
[11]
[12]
[13]
K. N. Brown, J. H. Espenson, Inorg. Chem. 1996, 35,
7211Ϫ7216.
W. Adam, M. N. Korb, K. J. Roschmann, C. R. Saha-Möller,
J. Org. Chem. 1998, 63, 3423Ϫ3428.
W. Adam, C. M. Mitchell, C. R. Saha-Möller, T. Selvam, O.
Weichold, J. Mol. Catal. A 2000, 154, 251Ϫ255.
Complex 2 has been demonstrated to attack preferentially at
non-coordinated electron-rich sulfur centers as in sulfides, see
ref.[4]
N. S. Issacs, Physical Organic Chemistry, Longman Scientific
Technical, Harlow, 1987, pp. 626Ϫ631.
M. Bonchio, G. Licini, G. Modena, S. Moro, O. Bortolini, P.
Traldi, W. A. Nugent, Chem. Commun. 1997, 869Ϫ870.
M. Bonchio, G. Licini, G. Modena, O. Bortolini, S. Moro, W.
A. Nugent, J. Am. Chem. Soc. 1999, 121, 6258Ϫ6268, and ref-
erences cited therein.
In this work the oxidation of p-NCC6H4SOMe has been rein-
vestigated at lower catalyst and hydroperoxide concentration
(see Exp. Sect.) in order to slow down the process rate and
allow an accurate kinetic analysis in the whole range of sub-
strate concentrations examined.
Computational Study: All calculations were carried out using
Gaussian 98 (rev. A7),[41] with an SGI O2 R10000 workstation.
HartreeϪFock (HF) and density functional theory (DFT) calcula-
tions were run with 3-21G(*) or 6-311ϩϩ basis sets.[42Ϫ45] More-
over, density functional theory calculations were carried out with
Becke’s hybrid 3-parameter functional with LeeϪYangϪParr non-
local correlation (B3LYP).[41,46] These calculations had to be run
with the (99,302) numerical integration grid[41] and with full accu-
racy at all stages in order to achieve SCF convergence. Hay and
Wadt’s effective core potentials (ECP)[47,48] for titanium were used
in DFT calculations. However, TiIV compounds have a 3d0 config-
uration, and therefore ECPs incorporating up to the outermost
core orbitals (3s23p6) will not give satisfactory results. Accordingly,
the ECPs developed specifically for these cases[47,48] (i.e., explicitly
treating 3s23p6 electrons) were used; this basis set is denoted as
LANL2DZ in Gaussian98. The calculated total energies, and the
complete geometries (Cartesian atomic coordinates) are given as
Supporting Information.
[14]
[15]
[16]
[17]
[18]
In the case under study, parallel to enzyme kinetics,[14] the
simple MichaelisϪMenten equation can be applied (see general
scheme at the end of the reference) where Ro ϭ Vmax ϫ [p-
XC6H4SOMe]0/(KM ϩ [p-XC6H4SOMe]0, Vmax ϭ kIN ϫ [2]0,
KM ϭ (kϪ1 ϩ kIN)/k1. Vmax and KM values have been deter-
mined by least-square fitting of the experimental data (Scient-
istTM-MicroMath Scientific Software, P. O. Box 21550, Salt
Lake City, Utah 84121, USA, 1995), according to the above
described model.
General Conditions for the Kinetic Experiments: In a 3-mL volu-
metric flask, complex 1b (0.08 mmol), the internal standard (0.070
mmol), and an appropriate amount of substrate (0.027Ϫ0.405 ,
see Figure 1) were dissolved in DCE. After cooling to Ϫ20 °C,
cumyl hydroperoxide (0.015 mL, 0.081 mmol) was added under
magnetic stirring. At increasing reaction times, samples of the mix-
ture (0.025 mL) were taken out, immediately quenched with an ex-
cess of di-n-butyl sulfide, and analyzed by quantitative GC analysis.
Values of initial rates of the oxidation (Ro) (Figure 1) were
[19]
[20]
[21]
L. P. Hammett, J. Am. Chem. Soc. 1937, 59, 96Ϫ103.
H. H. Jaffe, Chem. Rev. 1953, 53, 191Ϫ261.
S. Rabe, U. Müller, Z. Naturforsch., Teil
1291Ϫ1295.
B 1997, 52,
determined on the basis of the following equation: Ro
k[Sub]0[CumOOH]0, where k is derived from the kinetic data
according to the second-order integrated law.
ϭ
[22]
[23]
T.-Y. Hwang, J.-Y. Cho, M.-K. Jiang, H.-M. Gau, Inorg. Chim.
Acta 2000, 303, 190Ϫ198.
1
VT H NMR experiments run in CDCl3 upon addition of in-
creasing amounts of [D6]DMSO (up to 260 equiv.) showed a
modest broadening of all resonances in the whole range of tem-
perature explored (from Ϫ60 to ϩ50 °C). This behaviour,
which may be ascribed to the fast equilibria in solution, did
not provide any structural evidence of DMSO coordination to
complex 1.
Acknowledgments
Financial support by University of Padova Progetti di Ateneo 1999,
CNR, and DuPont ATE program is gratefully acknowledged. This
project has been carried out in the frame of COST-D12 (D12/
0018/98).
[24]
[25]
[26]
ESI-MS experiments were performed under the same reported
experimental conditions.[16]
Under ESI-MS conditions, catalyst 1 is expected to yield the
trinuclear oxonium ion {[TiN(CH2CHMeO)3]3Oϩ}.[16]
The hexacoordinated complex [TiN(CH2CHPhO)3(MeO)-
(DMSO)Na]ϩ, carrying both alkoxide and sulfoxide residues,
could be detected by ESI-MS at m/z ϭ 555.
The loss of aldehyde, derived from decomposition at the tetra-
dentate trialkoxy ligand, was already observed in MS2 experi-
ments performed on [tris(hydroxyalkyl)amine]TiIV protonated
complexes, and it is indicative of the coordination of the tris-
(hydroxyalkyl)amine to the metal center.[16] In fact, upon CID,
the free protonated tris(hydroxyalkyl)amines lose one, two or
three water molecules.
[1]
F. Di Furia, G. Licini, G. Modena, R. Motterle, W. A. Nugent,
J. Org. Chem. 1996, 61, 5175Ϫ5177.
G. Licini, M. Bonchio, G. Modena, W. A. Nugent, Pure Appl.
[2]
Chem. 1999, 71, 463Ϫ472.
For other stereoselective sulfoxidations involving cooperative
[3]
[27]
asymmetric oxidation and kinetic resolution, see reference 21
in: M. Bonchio, G. Licini, F. Di Furia, S. Mantovani, G. Mod-
ena, W. A. Nugent, J. Org. Chem. 1999, 64, 1326Ϫ1330.
[4]
M. Bonchio, S. Calloni, F. Di Furia, G. Licini, G. Modena, S.
Moro, W. A. Nugent, J. Am. Chem. Soc. 1997, 119, 6935Ϫ6936.
M. Selke, M. F. Sisemore, J. S. Valentine, J. Am. Chem. Soc.
[5]
[28]
1996, 118, 2008Ϫ2012.
Only in the case of ion III another fragment ion [m/z ϭ 342
510
Eur. J. Org. Chem. 2003, 507Ϫ511

Oxygen Transfer to Sulfoxides by (Peroxo)[tris(hydroxyalkyl)amine]TiIV Complexes
FULL PAPER
[39]
(Mϩ Ϫ 56)], corresponding to the loss of a molecule of butene
from the alkyl chain of the di-n-butyl sulfoxide, was detected.
Moreover, in all the spectra another peak at m/z ϭ 254 (MS2
of ions I, IЈ, and IЈЈ), m/z ϭ 440 (MS2 of ion II), and m/z ϭ
121 (MS2 of ion III) at much lower intensity (12Ϫ27%), was
detected. These ions correspond to the formal adducts of water
to the titanatrane cations [TiN(CH2CHRO)3OH2],ϩ and very
likely they originate from ion-molecule exchange processes oc-
curring with adventitious water molecules present inside the
ion trap. Such a hypothesis was proved by running MS2 experi-
ments on ion I with methanol as mobile phase. As expected,
an additional peak at m/z ϭ 268, corresponding to the
[TiN(CH2CHMeO)3MeOH]ϩ cation, was observed.
M. Calligaris, O. Carugo, Coord. Chem. Rev. 1996, 153,
83Ϫ154.
R. Annunziata, M. Cinquini, F. Cozzi, L. Raimondi, G. Licini,
Tetrahedron 1991, 47, 3869Ϫ3886.
L. Favretto, W.A. Nugent, G. Licini, Tetrahedron Lett. 2002,
43, 2581Ϫ2584 and references therein.
[40]
[41]
M. Frisch, G. W. Trucks, H. Schlegel, G. Scuseria, M. Robb, J.
Cheeseman, V. Zakrzewski, J. Montgomery, R. E. Stratmann,
J. C. Burant, S. Dapprich, J. Millam, A. Daniels, K. Kudin, M.
Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi,
B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J.
Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C.
Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W.
Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Re-
plogle, J. A. Pople, Gaussian98 (revision A.7), Gaussian, Inc,
Pittsburgh, PA, 1998.
A. J. H. Wachters, J. Chem. Phys. 1970, 52, 1033Ϫ1036.
P. J. H ay, J. Chem. Phys. 1977, 66, 4377Ϫ4384.
M. J. Frisch, J. A. Pople, J. S. Binkley, J. Chem. Phys. 1984,
80, 3265Ϫ3269.
T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. v. R. Schleyer,
J. Comput. Chem. 1983, 4, 294Ϫ301.
A. D. Becke, J. Chem. Phys. 1993, 98, 5648Ϫ5652.
P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270Ϫ283.
P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299Ϫ310.
Received June 12, 2002
[29]
[30]
[31]
M. Calligaris, Croat. Chim. Acta 1999, 72, 147Ϫ169.
S. S. Xantheas, T. H. Dunning, J. Phys. Chem. 1993, 97,
6616Ϫ6627.
B. A. Smart, C. H. Schiesser, J. Comput. Chem. 1995, 15,
1055Ϫ1066.
[32]
[33]
[42]
[43]
[44]
P. J. A. Ruttink, P. C. Burgers, J. T. Francis, J. K. Terlouw, J.
Phys. Chem. 1996, 100, 9694Ϫ9697.
[34]
[35]
[45]
ˇ
F. Turecek, J. Phys. Chem. 1998, 102, 4703Ϫ4713.
A. P. Scott, L. Random, J. Phys. Chem. 1996, 100,
[46]
[47]
[48]
16502Ϫ16513.
[36]
[37]
M. Stener, J. Mol. Struct. (THEOCHEM) 2000, 497, 91Ϫ104.
D. R. Boyd, N. D. Sharma, S. A. Haughey, M. A. Kennedy, B.
T. McMurray, G. N. Sheldrake, C. C. R. Allen, H. Dalton, K.
Sproule, J. Chem. Soc., Perkin Trans. 1 1998, 1929Ϫ1934.
F. Fringuelli, R. Pellegrino, O. Piermatti, F. Pizzo, Synth. Com-
mun. 1994, 24, 2665Ϫ2674.
[I02322]
[38]
Eur. J. Org. Chem. 2003, 507Ϫ511
511