Hypovalent titanium and Ti(II)-Ti(III) interconversions

Article abstract of DOI:10.1039/b914989a

Treatment of pink titanium(iii) triflate (0.045 M) with HF in triflic acid (CH₃SO₃H) converts Ti(iii) rapidly to a 1:1 mixture of Ti^IV and green Ti^II: (2 Ti^III + 4 HF → TiF₄ + Ti^II + 4 H⁺). This disproportionation is half complete when [HF] added is 0.027 M. Substituted 1,4-benzoquinones are reduced rapidly by Ti(iii) in the absence of fluoride, yielding straightforward logarithmic curves, but reactions of the same quinones with Ti(ii) in fluoride media exhibit more complex profiles, the major portions of which are zero order in oxidant. These reactions are strongly catalyzed by added Ti(iv). Analyses of complex curves are consistent with a reaction sequence initiated by Ti(ii)-Ti(iv) disproportionation, forming Ti(iii), which reacts with the quinone, yielding the quinhydrone, QH. The latter is rapidly reduced by Ti(ii). Values of rate constants obtained from these analyses are in agreement with those for reductions of quinones by Ti(iii), in the absence of fluoride. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/b914989a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Hypovalent titanium and Ti(II)–Ti(III) interconversions†‡
Basab Bijayi Dhar* and Edwin S. Gould*
Received 24th July 2009, Accepted 6th November 2009
First published as an Advance Article on the web 14th December 2009
DOI: 10.1039/b914989a
Treatment of pink titanium(III) triflate (0.045 M) with HF in triflic acid (CH3SO3H) converts Ti(III)
rapidly to a 1 : 1 mixture of TiIV and green TiII: (2 TiIII + 4 HF → TiF4 + TiII + 4 H+). This
disproportionation is half complete when [HF] added is 0.027 M. Substituted 1,4-benzoquinones are
reduced rapidly by Ti(III) in the absence of fluoride, yielding straightforward logarithmic curves, but
reactions of the same quinones with Ti(II) in fluoride media exhibit more complex profiles, the major
portions of which are zero order in oxidant. These reactions are strongly catalyzed by added Ti(IV).
Analyses of complex curves are consistent with a reaction sequence initiated by Ti(II)–Ti(IV)
disproportionation, forming Ti(III), which reacts with the quinone, yielding the quinhydrone, QH∑. The
latter is rapidly reduced by Ti(II). Values of rate constants obtained from these analyses are in
agreement with those for reductions of quinones by Ti(III), in the absence of fluoride.
potentials is rare and stands in contrast to other cation pairs
Introduction
such as V(II)/V(III) (with the lower state much more active) and
s-centers such as In(I)/In(II) (with the higher state much more
reactive).10
For over 160 years after the discovery of titanium in 1791, the
chemistry of that element was concerned mainly with the metal
itself and its 3+ and 4+ states. The report, in 1955,2 of titanium-
based Ziegler catalysts for the polymerization of alkenes triggered
a remarkable growth of low-valent titanium chemistry, and an im-
pressive array of organotitanium compounds have been reported
during the intervening half century,3 many of these derived from
the dipositive state. In addition, titanium(II) cyclopentadienyls and
alkoxides have exhibited versatility in organic conversions.4
Nevertheless, relatively little attention has been devoted to
the “parent” Ti(II) entity, the aquatitanium(II) cation, Ti2+(aq).
Reports by Forbes and Hall5 and Olver and Ross6 indicated that
this cation readily reduces aqueous solutions to H2 at low pH,
thus complicating study using conventional methods of handling.
Among the 3d-block dipositive ions, Ti(II) is undoubtedly the most
elusive.
An unusually wide variety of catalyses has been reported for
reductions by Ti(II). Reductions of iron(III) and ruthenium(III)
complexes are accelerated by halide, with the effects of added
fluoride much more marked than those by the heavier halides.11
Dramatic catalysis by added Ti(IV) has been reported for re-
ductions of quinones,12 and Ti(II) reductions of triiodide and
pentapositive vanadium are accelerated by added molybdic acid.13
Ti(II) is of particular interest because it can be induced to
undergo oxidations at rates independent of oxidant. This unusual
kinetic picture, first observed with very reactive Co(III) oxidants,14
implies the slow generation of an activated reducing species.
We here present evidence that this behavior reflects the initial
formation of Ti(III) by comproportionation (1):
Ti(II) + Ti(IV) → 2Ti(III)
(1)
However, facile preparations of this state in HF-triflic acid
mixtures were reported in 2003 by Kolle and Kolle,7 and Yang
has noted that 0.1 M Ti(II) solutions so generated survive for 12 h
at room temperature under argon.8
Experimental
Titanium(II) solutions prepared in this manner contain equiva-
lent concentrations of Ti(IV) and a HF/Ti(II) ratio approximately
6 : 1. Although such preparations are strongly acidic, it is likely
Materials
Solutions were prepared from Millipore-Q System deionized water
that had been boiled for 2 h and then purged with argon for a fur-
ther 2 h to removed traces of dissolved O2. Titanium(II) solutions
(ca 0.10 M in reductant in 2.0 M triflic acid) were prepared under
argon by a modification of the method of Kolle and Kolle.7,8 Those
green solutions were kept in sealed containers and used within 10 h
of preparation. The Ti(II) concentration was determined by adding
a known volume to a solution of [Co(NH3)5F](ClO4)2 under argon,
waiting 20 min, diluting ten-fold with con HCl, and determining
[CoCl4]2- at 692 nm (e = 560 M-1cm-1). Titanium(III) chloride (10%
in 25% HCl) (Aldrich) was used as received. Titanium(IV) solutions
were prepared by air oxidation of Ti(II) solutions; after oxidation
to Ti(IV) was complete (total loss of color), argon was passed
through the solution for 2 h to remove dissolved O2. Sodium triflate
was prepared by the method of Stanbury.15 Solutions of Ti(III) free
-
that Ti(IV) is present in part as TiF62- and/or TiF5 . Conversion of
Ti(II) (a d2 cation) to a fluoro complex is slight.9 Yang’s solutions
absorb preferentially at 430 and 660 nm, but extinction coefficients
(6.0 and 3.5 M-1 cm-1) are greater than those reported by Kolle
and Kolle.7
Dipositive titanium presents features which are seldom observed
among the other elements. The occurrence of two adjacent
oxidation states exhibiting similar reducing reactivities (similar
patterns of rate constants) despite substantial differences in formal
Department of Chemistry, Kent State University, Kent, Ohio, 44242, USA
† Electron transfer part 168. For part 167, see ref. 1.
‡ Electronic supplementary information (ESI) available: Detailed kinetic
data for redox reactions (Tables S1–S3). See DOI: 10.1039/b914989a
1616 | Dalton Trans., 2010, 39, 1616–1619
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
from Cl- in 1 M triflic acid were prepared by a modification of the
method of Martin.16
For the system at hand, a simplified equilibrium quotient (3)
applies since [H+] is kept at 0.50 M
[Ti(II)]2
[TI(III)]2[HF]4
Spectroscopic studies
K3,2
=
,
(3)
Spectral studies with TiII and TiIII solutions were carried out under
argon using a Shimadzu 1601 UV-visible spectrophotometer. The
single lmax for Ti(III) in 0.5 M triflic acid is at 502 nm, whereas Ti(II)
exhibits peaks at 540 and 650 nm. When treated with a five-fold
excess of fluoride in 0.5 M triflic acid, Ti(III) is converted rapidly
and very nearly completely to a 1 : 1 mixture of Ti(II) and Ti(IV).
and [TiII] = [TiIV].
Ti(IV) does not contribute to visible absorbance. Hence, the
absorbance of these solutions reflects only the two lower states:
Abs = [TiIII]e3 + [TiII]e2
Treatment of absorbances at different [HF] values, in conjunc-
tion with the summation relationship (4), generates (Appendix) a
series of simultaneous equations, one for each portion of added
HF. Least squares refinement of data at 500 nm yields
(4)
Kinetic studies
Reactions were run under argon. Rates of reduction of quinones
were obtained from absorbance decreases at lmax values of the
oxidant, using a Durrum-Gibson stopped-flow spectrophotome-
ter. Concentrations of reagents were generally adjusted so that no
more than 10% of the reagent in excess was consumed in a single
run. Temperatures were maintained at 22.0 0.5 ◦C during all
experiments. All reductions by Ti(III) exhibited straightforward
exponential decay curves which yielded pseudo-first order rate
constants proportional to the concentrations of excess reductant.
Reductions by excess Ti(II), on the other hand, gave more complex
patterns (e.g. Fig. 3) from which several contributing rate constants
could be obtained by computerized procedures.
e3 = 4.1 M-1 cm-1, e2 = 4.5 M-1 cm-1 and K3,2 = (2.4 0.2)
¥ 105 M-4.
Calculated and observed absorbances are compared in Fig. 2
and Table S-1.‡ Alternate treatments using 3 HF and 5 HF in
(2) gave much poorer fittings. Analogous treatment of absorbance
data, at 430 and 650 nm, yields K3,2 values of (2.3 0.1) ¥ 105 and
(2.3 0.2) ¥ 105 M-4.
Results and discussion
TiIII–TiII interconversions
Treatment of pink Ti(III) solutions in 0.5 M triflic acid with
successive portions of HF results in fading of the peak at 500 nm
and the growth of peaks at 430 and 650 nm typifying Ti(II).7,8
Isosbestic points appear at 490 and 600 nm (Fig. 1). The reaction
is a disproportionation which is favored by the strong affinity of
fluoride for tetrapositive titanium.17
2TiIII + 4HF→ TiF4 + TiII + 4H+.
(2)
Fig. 2 Observed and calculated absorbance changes at 500 and 650 nm
as HF is added to Ti(III). Solid lines were obtained from eqn (4) in text
and eqn (g) in Appendix. Extinction coefficients (in M-1 cm-1) for Ti(III)
(e3) and Ti(II) (e2) are tabulated below. Reaction quotient K3,2 (eqn (3) in
text) is taken as 2.4 ¥ 105 M-4
500 nm
650 nm
e3
e2
4.10
4.50
0.74
4.9
Reduction of quinones
An earlier study8 showed that reductions of 1,4-benzoquinone in
aqueous acid, both by Ti(II) and Ti(III), yield the corresponding
hydroquinones (1,4-dihydroxybenzenes). It is reasonable to sus-
pect that action by both reductants is initiated by single electron
Fig. 1 Conversion of 0.045 M Ti(III) to Ti(II) + Ti(IV) in 0.50 M CF3SO3H
by addition of HF. Curve a—no HF; curve b—0.0375 M HF; curve
c—0.0675 M HF. Vertical line denotes the change in absorbance at 500 nm
as HF is added.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 1616–1619 | 1617
©

View Article Online
Table 1 Titanium(III) reductions of 1,4-benzoquinones. Kinetic
curvature becomes important during the final stages as the oxidant
becomes nearly consumed.
parametersab
The striking kinetic behavior, in conjunction with the recognized
ease of the Ti(II)/Ti(III) interconversion in fluoride-containing
media, points strongly to sequence (6)–(8) for the Ti(II)-quinone
reactions.
kobs = ko + k-1/[H+]
Quinone
ko, M-1s-1
k-1, s-1
1,4-Benzoquinone
(1.07 0.01) ¥ 102 23
1
Chloranilic acidc
(7.0 0.1) ¥ 103
(1.0 0. 1) ¥ 102
k
TiII + TiF
2TiIII + 4F−
1
ꢀꢀꢀꢁ
(6)
(7)
(8)
4 ꢂꢀꢀꢀ
Tetrahydroxy-1,4-benzoquinone
k
−1
a Reactions at 22.0 0.5 ◦C. Acidity was maintained using triflic acid, m =
0.5 M (CF3SO3H + CF3SO3Na). b Parameters were obtained by refinement
of data in Table S-3. c 2,5-Dichloro-3,6-dihydroxy-1,4-benzoquinone.
+
k
,H
TiIII + Q ⎯⎯⎯→TiIV + QHi
2
+
QHi + TiII ⎯H⎯→QH2 + TiIII (rapid)
steps, but distortion of decay curves reflecting semiquinone
intervention is not generally observed, indicating that reduction
of this intermediate is rapid in this medium and, hence, kinetically
silent.
Reductions by Ti(III) are retarded by boosts in acidity (Table
S-3).‡ The observed dependency for the benzoquinone reduction
follows rate-law (5), whereas with the ring-substituted quinones,
the acid-independent component, ko, is negligible (Table 1).
Incorporation of sequence (6)–(8) into a variant of the KINSIM
treatment19 yielded calculated absorbance curves which were
compared with the observed kinetic profiles. Kinetic parameters
giving optimal agreement with the run pictured in Fig. 3 are k1 =
1.2 ¥ 102 M-1 s-1, k-1 = 5.0 M-1 s-1, and k2 = 150 M-1 s-1. Parallel
treatments of runs at different acidities indicate that k1 and k-1
do not vary significantly with [H+], whereas k2 falls from 150 to
90 M-1 s-1 between 0.250 mM and 7.5 mM HF. Note that the
value of k2 is in reasonable agreement with the second order rate
constant (0.15/0.0010 = 150 M-1 s-1) obtained by direct kinetic
measurement of the Ti(III)-quinone system (Table S-2).‡
Kinetic parameters contributing to the Ti(II) reductions of
several quinones under similar conditions are summarized in
Table 2. As expected, values of k1 and k-1 are independent of
the oxidant, which is waiting to enter the picture. A conditional
equilibrium quotient for the comproportionation20 step may be
estimated as k1/k-1 = 110/5.0 = 22. The modest value of this
kobs = (ko + k-1/[H+])[TiIII]
(5)
The inverse-acid dependence, k-1/[H+], often appears in Ti(III)
reductions in aqueous media16,18 and is commonly attributed to
participation of the conjugate base, [TiIII(OH)]2+.
Kinetic profiles for the reductions of these quinones by Ti(II)
are much more complex than the exponential patterns obtained
with Ti(III). As shown in Fig. 3, which is typical, a major portion
of the curve is linear, during which the rate of reaction is nearly
independent of the concentration of the quinone. The slope of this
section is nearly proportional to the concentration of Ti(II) taken,
and also to the concentration of Ti(IV) present. In addition, slight
acceleration occurs during the first few percent of reaction, and
Table 2 Titanium(II) reductions of 1,4-benzoquinones. Kinetic
parametersa
k
TiII + TiIV
2Ti
1
III
ꢀꢀꢀꢁ
ꢂꢀꢀꢀ
k
−1
+
k
,H
TiIII + Q ⎯⎯⎯→TiIV + QHi
2
b
b
b
Quinone
[HF], mM
k1
k-1
k2
Chloranilic acide
0.25
2.5
5.0
7.5
7.5
7.5
110
115
105
105
110
120
5.0
5.0
3.9
5.0
5.0
5.0
2300
1400
640
360
370c
380d
1,4-Benzoquinone
0.25
2.5
5.0
7.5
120
104
110
106
5.0
4.0
5.0
4.4
150
120
95
90
Tetrahydroxy-1,4-
benzoquinone
0.25
100
5.0
280
2.5
5.0
7.5
104
110
106
5.0
5.0
4.4
200
180
180
Fig. 3 Kinetic profile at 246 nm for reaction of 1,4-benzoquinone with
Ti(II) in 0.50 M CF3SO3H. [TiII] = 1.0 mM; [HF] = 0.25 mM, [quinone] =
0.05 mM. Circles represent the experimental data, whereas the solid line
was obtained by KINSIM simulation (ref. 19) of the proposed mechanism
(6)–(8), taking k1 as 120 M-1 s-1, k-1 as 5.0 M-1 s-1, and k2 as 150 M-1 s-1.
[TiIV] = 1.00 mM; molar absorbances (e values) were 2.60 ¥ 104 M-1 cm-1
for this oxidant and 1.40 ¥ 104 M-1 cm-1 for the product. Path length =
1.00 cm.
a Component rate constants (22 ◦C) obtained by KINSIM treatment (ref.
19) of kinetic data for Ti(II) reductions of benzoquinones in 0.50 M
CF3SO3H in the presence of 2.0 mM Ti(IV) and added HF. [Ox] = 0.01 mM;
[TiII] = 1.0 mM unless otherwise indicated. b Second order rate constants
giving optimal fit of calculated and observed kinetic curves. c [TiII] =
0.25 mM. d [TiII] = 0.50 mM. e 2,5-Dichloro-3,6-dihydroxybenzoquinone.
1618 | Dalton Trans., 2010, 39, 1616–1619
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
quotient reminds us that in fluoride-rich media the effective
oxidation potentials of Ti(II) and Ti(III) approach each other and
that this trend arises not from complexation of Ti(II), but rather
from ligation of Ti(IV), the product from Ti(III).
OD = e3[TiIII] + e2[TiII] (Beer’s law) (h)
e3[Ti]tot
1+ 2(K3,2
)
1/ 2[HF]2
e2(K3,2 )
1/ 2[HF]2[Ti]tot
Rates for the TiIII-quinone step (k2) are seen to diminish
perceptibly with increases of (HF), presumably due to the partial
loss of the more active reducing species, [TiIIIOH]2+ (eqn (5)).
In short, at high concentrations of HF, direct electron transfer
from Ti(II) to quinone is so slow that virtually all loss of oxidant
is triggered instead by reaction with Ti(III). We suspect that
the sluggish action of Ti(II), may reflect, at least in part, an
unusually low Ti(II,III) self-exchange rate. Both TiII (3d2) and
TiIII(3d1) should be subject to minor Jahn–Teller distortions, but
the degree of such distortion and the occupancy of the non-
degenerate orbitals (one short, two long) vs. (two short, one
long) need not match. The two states may have quite different
geometries, and this mismatch would be expected to result in a
higher Franck–Condon barrier to exchange.21 Beyond this, there
is the possibility of reversible donor–acceptor interaction between
the d2 center and the highly conjugated quinone molecule, yielding
a complex which geometrically disfavors internal electron transfer.
Such intervention need not persist when we are dealing with TiIII(a
d1 center), or with a less markedly conjugated semiquinone (QH∑)
transient.
+
(i)
1+ 2(K3,2 )
1/ 2[HF]2
(e3+e2(K 3,2
)
1/ 2[HF]2 )[Ti]tot
OD =
(j)
1+2(K 3,2 )
1/ 2[HF]2
Acknowledgements
We are grateful to Dr Yuriy Tolmachev, Dr Paul Sampson, and
Mr Robert Hoover for helpful discussions and to Mrs Arla Dee
McPherson for technical assistance.
References
1 Part 167.B. B. Dhar, R. Mukherjee and E. S. Gould, Dalton Trans.,
2009, 868.
2 K. Ziegler, E. Holzkamp, H. Breil and H. Martin, Angew. Chem., 1955,
67, 541.
3 (a) Reviews: E. A. Mints, in Encyclopedia of Inorganic Chemistry,
R. B. King, ed. Vol. 8, Wiley, New York, 1994, 4206-4224.; (b) S. B.
Smith and D. W. Stephan, in Comprehensive Coordination Chemistry,
J. A. McCleverty and T. J. Meter, Ed., v. 4, Elsevier, Amsterdam, 2004,
Chapt. 4.2.
4 Reviews: F. Sato and H. Urabe, Titanium and Zirconium in Organic
Synthesis, I. Marek, ed., Wiley, New York, 2002, pp. 281-329.
5 G. S. Forbes and L. P. Hall, J. Am. Chem. Soc., 1924, 46, 385.
6 J. W. Olver and J. W. Ross, Jr., J. Am. Chem. Soc., 1963, 85, 2565.
7 U. Kolle and P. Kolle, Angew. Chem., Int. Ed., 2003, 42, 4540.
8 Z. Yang and E. S. Gould, Dalton Trans., 2005, 1781.
9 R. M. Smith and A. E. Martell, Critical Stability Constants, v. 4,
Plenum, New York, 1971, p. 99.
Appendix
Treatment of absorbance data for Ti(II), Ti(III), Ti(IV) systems
2TiIII + 4HF ꢀ TiII + TiIVF4 + 4H+ (a)
10 S. K. Chandra and E. S. Gould, Inorg. Chem., 1996, 35, 3381.
11 R. Mukherjee, V. Mannivanan and E. S. Gould, Inorg. Chim. Acta,
2007, 360, 3633.
12 B. B. Dhar and E. S. Gould, Dalton Trans., 2007, 1416.
13 Z. Yang and E. S. Gould, Dalton Trans., 2006, 396.
14 R. Mukherjee, Z. Yang and E. S. Gould, Dalton Trans., 2006, 772.
15 B. Saha and D. Stanbury, Inorg. Chem., 2000, 39, 1294.
16 A. H. Martin and E. S. Gould, Inorg. Chem., 1975, 14, 873.
17 L. Ciavatta and A. Pirozzi, Polyhedron, 1983, 2, 769.
18 J. P. Birk, Inorg. Chem., 1975, 14, 1724.
[TiIII] + [TiII] + [TiIV] = [Ti]tot (b)
In these systems, [TiII] = [TiIV]; hence [TiIII] + 2[TiII] = [Ti]tot (c)
For our measurements [H+] is very nearly constant (0.50 M
[Ti(II)]2
CF3SO3H)
(d)
,
K3,2
=
[TI(III)]2[HF]4
Whence [TiII] = (K3,2
)
1/2[TiIII][HF]2 (e)
19 B. A. Barshop, R. F. Wrenn and C. Frieden, Anal. Biochem., 1983, 130,
134.
Substituting (e) into (c):
20 This quotient differs from K3,2 derived from spectral studies under a
much wider range of (HF) and (F-). During the kinetic runs treated
with KINSIM, these parameters are held nearly constant.
21 Note that an unusually low self-exchange rate has been calculated
for the analogous d1, d2 pair,[VIVO(H2O)4]2+, [VIII(OH)2(H2O)3]+. See:
M. C. Ghosh and E. S. Gould, J. Am. Chem. Soc., 1993, 115, 3167.
[TiIII] + 2(K3,2
)
1/2[TiIII][HF]2 = [Ti]tot (f)
[Ti]tot
III
Solving for [TiIII],
(g)
[Ti ]=
1+2(K3,2 )
1/2[HF]2
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 1616–1619 | 1619
©