Electron transfer. Part 165: Oxidations of Ti(II)(aq) with ligated iron(III) and ruthenium(III)

Article abstract of DOI:10.1016/j.ica.2007.01.026

Titanium(II) solutions, prepared by dissolving titanium wire in triflic acid + HF, contain equimolar quantities of Ti(IV). Treatment of such solutions with excess Fe(III) or Ru(III) complexes yield Ti(IV), but reactions with Ti(II) in excess give Ti(III). Oxidations by (NH₃)₅Ru(III) complexes, but not by Fe(III) species, are catalyzed by titanium(IV) and by fluoride. Stoichiometry is unchanged. The observed rate law for the Ru(III)-Ti(II)-Ti(IV) reactions in fluoride media points to competing reaction paths differing by a single F^-, with both routes involving a Ti(II)-Ti(IV) complex which is activated by deprotonation. It is suggested that coordination of Ti(IV) to Ti^II(aq) minimizes the mismatch of Jahn-Teller distortions which would be expected to lower the Ti(II,III) self-exchange rate.

Full text of DOI:10.1016/j.ica.2007.01.026

Inorganica Chimica Acta 360 (2007) 3633–3636
www.elsevier.com/locate/ica
Note
Electron transfer. Part 165: Oxidations of Ti(II)(aq)
with ligated iron(III) and ruthenium(III)
Ritam Mukherjee, V. Manivannan, Edwin S. Gould *
Department of Chemistry, Kent State University, Kent, OH 44242, USA
Received 2 January 2007; accepted 12 January 2007
Available online 12 March 2007
Abstract
Titanium(II) solutions, prepared by dissolving titanium wire in triflic acid + HF, contain equimolar quantities of Ti(IV). Treatment of
such solutions with excess Fe(III) or Ru(III) complexes yield Ti(IV), but reactions with Ti(II) in excess give Ti(III). Oxidations by
(
3 5
NH ) Ru(III) complexes, but not by Fe(III) species, are catalyzed by titanium(IV) and by fluoride. Stoichiometry is unchanged. The
observed rate law for the Ru(III)–Ti(II)–Ti(IV) reactions in fluoride media points to competing reaction paths differing by a single
ꢀ
F , with both routes involving a Ti(II)–Ti(IV) complex which is activated by deprotonation. It is suggested that coordination of Ti(IV)
II
to Ti (aq) minimizes the mismatch of Jahn–Teller distortions which would be expected to lower the Ti(II,III) self-exchange rate.
ꢀ
2007 Elsevier B.V. All rights reserved.
Keywords: Electron transfer; Iron complexes; Ruthenium complexes; Titanium(II); Titanium(IV)
1
. Introduction
this novel reductant from changes in kinetic behavior
resulting from variations in ligating species and reaction
medium.
Generation of aqueous titanium(II), as described by
Kolle and Kolle [1] yielded solutions also containing equi-
molar quantities of Ti(IV) but which were unexpectedly
easy to handle [2]. In contrast to earlier preparations
reported by Olver and Ross [3] and Forbes and Hall [4],
solutions 0.1 M in Ti(II) can be kept in triflic acid-HF
media under argon for 12 h at room temperature.
2
. Experimental
2.1. Materials
Solutions were prepared from Millipore-Q-system
Titanium(II) is a versatile reagent, but its action may be
torpid. The difference in standard potentials (0.47 V) [3,5]
documented for Ti(II) and Ti(III) corresponds, in the Mar-
cus treatment for outer-sphere reductants [6], to a rate ratio
deionized water that had been purged with pure argon
for 2 h to remove dissolved oxygen. Titanium metal wire,
ethanediaminetetraacetic acid iron(III) monosodium salt
[NaFe(EDTA)], trifluoromethanesulfonic acid (triflic acid)
and TiCl solutions in HCl were Aldrich or Alfa products
and were used as received. Titanium(II) solutions (0.10 M
6
exceeding 10 . Observed rate differences for the two tita-
3
nium reductants tend, however, to be much less marked,
and in a few cases, Ti(III) reacts more rapidly [7].
The present report deals with the reductions, using
Ti(II), of several mononuclear complexes of iron(III) and
ruthenium(III). We seek mechanistic inferences concerning
II
Ti in 2.0 M triflic acid) were prepared under argon by
the method of Kolle and Kolle [2,7]; these green solutions
were kept in sealed containers and used within 10 h of
preparation.
Titanium(IV) solutions were prepared by air oxidation
of Ti(II); after the oxidation was complete (as indicated
by loss of color), argon was passed through the solution
*
Corresponding author. Tel.: +1 330 672 2267; fax: +1 330 672 3816.
E-mail address: esgould@kent.edu (E.S. Gould).
0
020-1693/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2007.01.026

3634
R. Mukherjee et al. / Inorganica Chimica Acta 360 (2007) 3633–3636
for 2 h to remove dissolved O . Preparation of Fe-
Temperatures were kept at 22.5 ± 0.5 ꢁC. Ionic strength
2
(
o-phen) (ClO ) utilized a modification of the method of
was maintained with CF SO H/CF SO Na. Reactions
3
4 3
3
3
3
3
Burstall [8], in which oxidation of the Fe(II) complex to
Fe(III) was carried out with NaClO/HCl, rather than with
Cl [9]. Sodium triflate was prepared as described by Saha
2
were most often carried out under pseudo-first order condi-
tions with Ti(II) present in greater than tenfold excess and
yielded exponential decay curves. The Fe (o-phen) -Ti
3
III
II
and Stanbury [10]. Ruthenium(III) complexes were pre-
pared as their triflate salts by the method of Gaunder
and Taube [11], as modified by Swavey et al. [12].
reaction was rapid; even under second order conditions;
only a lower limit for its rate constant could be obtained.
Study of Fe (aq) systems, with added halide or NCS ,
was complicated by ligation changes at the Fe(III) center
on a time scale near to that for the redox reaction of
interest.
3
+
ꢀ
2
.2. Stoichiometric studies
Stoichiometries of Fe(III)–Ti(II) reactions were deter-
mined by addition of measured deficiencies of Ti(II) to a
known excess of Fe(III) at fixed acidity, waiting for
3. Results
1
5 min intervals, adding excess NaI or KI solution, and
Stoichiometric experiments (Table 1) with oxidants in
excess indicate very nearly complete conversion to Ti(IV)
in all systems chosen
ꢀ
then measuring the concentration of I3 formed (480–
5
00 nm). Examination of the reactions with bound Fe(III)
was carried out by measuring the decrease in absorbance of
the oxidant at or near the low energy maximum: precipita-
II
III
IV
II
Ti þ 2Fe ! Ti IVþ 2Fe
ð1Þ
ð2Þ
3ꢀ
II
III
II
tion was encountered when ½FeðCNÞ ꢁ exceeded 1.0 mM
6
Ti þ 2Ru ! Ti þ 2Ru
II
and [Ti ] exceeded 0.6 mM. Stoichiometry of the Ti(II)–
Such reactions may be presumed to proceed through
Ti(III). Kinetic profiles from the Ru(III) reactions and
the faster Fe(III) reactions, carried out with Ti(II) in
excess, (a total of 80 individual profiles embracing eight dif-
ferent systems) exhibit no irregularities pointing to the
intervention of an intermediate formed or destroyed on a
time scale comparable to that of the principal conversions.
Reactions in this group then yield preponderantly Ti(III),
in line with the lower reactivity of the trivalent state in a
previous study [13].
Ru(III) reactions was determined at the maximum of the
highly colored Ru(II) product (550–620 nm). Results are
summarized in Table 1.
2
.3. Kinetic studies
Reactions were performed under argon and monitored
at kmax associated with the Fe(III) oxidant or the Ru(II)
product, using a Durrum–Gibson stopped-flow spectro-
photometer interfaced with an OLIS computer system.
Of the more rapid Ti(II)–Fe(III) systems, oxidation by
III
Fe (EDTA) exhibits
a
straightforward monomial
Table 1
dependence
a
Reductions of Fe(III) and Ru(III) complexes using Ti(II): stoichiometries
þ
3
ꢀ1 ꢀ1
II
rate ¼ kFeðEDTAÞ ꢁ½TiðIIÞꢁ; k ¼ ð4:9 ꢂ 0:2Þ ꢃ 10 M
s
Oxidant
k (nm)
[ox]
mM)
[Ti ]
D[Ox]/
D[Ti(II)]
(
(mM)
ð3Þ
Fe(III) (aq)
480,
5.0
4.6
4.6
4.0
1.0–2.0
1.92 ± 0.08
2.01 ± 0.10
1.96 ± 0.08
2.3 ± 0.1
3
ꢀ
whereas reactions with [Fe(CN) ] is strongly accelerated
5
00
480,
00
480,
00
480,
00
6
FeCl þ FeCl2þ
2
þ
b
b
1.0–2.0
1.0–2.0
+
by H (Table S1):
5
3
6
ꢀ
þ
FeBr þ FeBr2þ
2
þ
rate ¼ ½FeðCNÞ ꢁ½TiðIIÞꢁðk
o
þ k
½H ꢁÞ
ð4Þ
o
H
5
4
ꢀ1 ꢀ1
k
¼ ð5:9 ꢂ 0:2Þ ꢃ 10 M s ;
¼ ð8:6 ꢂ 0:4Þ ꢃ 10 M s :
FeF þ FeF2þ
2
þ
c
0.50–1.50
4
ꢀ2 ꢀ1
5
k
H
3
ꢀ
ꢀ
+ d
[
[
[
[
Fe(CN)
6
]
418
340
630
555
620
1.00
1.00
2.0
0.50
0.60
0.10–0.40 2.01 ± 0.09
0.10–0.30 2.03 ± 0.08
0.40–0.80 2.02 ± 0.11
0.10–0.20 2.23 ± 0.22
0.133–0.20 1.96 ± 0.01
Fe(EDTA)]
2
Fe(NCS)]
Fe(o-phen)
3
+e
3
3
]
Table 2
Reductions of iron(III) complexes with Ti(II) and Ti(III): rate constants
a
Ru(NH
4-AcPy)
Ru(NH
3 5
) -
+
(
ꢀ
1
ꢀ1
ꢀ1 ꢀ1
Ti(III) (M s )
Oxidant
[Fe(H O)
[Fe(EDTA)]
kTi(II) (M
s
)
k
3
)
5
-
550
0.50
0.100–0.20 2.18 ± 0.03
3
+
3+
2
2
(4-PyCONH
2
)
2
6
]
(4.2 ± 0.4) · 10
(4.9 ± 0.2) · 10
(5.9 ± 0.2) · 10
(1.7 ± 0.2) · 10
(3.2 ± 0.3) · 10
(1.7 ± 0.2) · 10
+
3
4
4
a
+
[
H ] = 0.50 M (CF
3
SO
3
H); l = 0.50 M unless otherwise noted.
3
ꢀ
5
[
Fe(CN) ]
6
Results of studies at 480 and 500 nm were in agreement.
4
+
b
+
+ (8.6 ± 0.4) · 10 [H ]
[
H ] = 0.50 M, (CF
3
SO
3
H), l = 2.5 M (CF
3
SO
3
H + LiCl or NaBr).
3
+
6
5b
ꢀ
[Fe(o-phen)
3
]
>1.5 · 10
2.6 · 10
[
Hal ] = 2.0 M.
c
+
a
[
H ] = 0.50 M (CF
3
SO
3
H); l = 1.0 M (CF
3
SO
3
H + CF SO Na +
3
3
Reactions were run at 22.5 ± 0.5 ꢁC; l = 0.50 M (CF
SO Na) in most cases.
25 ꢁC, l = 1.0 M (LiCl) (Akinyugha et al. Inorg. Chem. 17 (1978)
218).
3 3
SO H/
ꢀ
ꢀ
NaF); [F ] = 0.25 M.
d
CF
b
3
3
[
[
NCS ] = 0.040 M.
e
+
H ] = 1.75 M (CF
3
SO
3
H); l = 1.80 M.

R. Mukherjee et al. / Inorganica Chimica Acta 360 (2007) 3633–3636
3635
Table 3
Kinetic parameters pertaining to reductions of substituted pyridine derivatives of (NH
a
3
)
5
Ru(III)
3+
b
eq
ꢀ4
ꢀ1 ꢀ1
ꢀ7
ꢀ2 ꢀ1
s )
Oxidant, [Ru(NH
3
)
5
Lig]
K
10
k
1
(M
s
)
10
k
2
(M
Lig = 4-benzoylpyridine
104 ± 35
118 ± 38
100 ± 22
102 ± 19
114 ± 6
ꢀ
3.1 ± 0.1
2.4 ± 0.1
3.1 ± 0.1
2.4 ± 0.1
0.43 ± 0.01
2.8 ± 0.2
1.80 ± 0.11
6.9 ± 0.6
1.58 ± 0.09
0.43 ± 0.03
4
4
3
-Pyridinecarboxamide
-Acetylpyridine
-Acetylpyridine
Pyridine
III
III
II
IV
+ ꢀ1
Rate = ꢀd[Ru ]/dt = Keq[Ru ][Ti ][Ti ][H ] (k
1
+ k
2
[F ]).
a
Reactions at (22.5 ± 0.5) ꢁC. l = 0.50 M (CF
II IV
Formation constant for Ti Ti (OH); Keq ¼ ½TiII
3
SO
3
H + CF
½Ti Ti ðOHÞꢁ
3
SO
3
Na).
II IV
b
IV
þ
ꢀ1
.
ꢁ½Ti
ꢁ½H ꢁ
3
+
For the most rapid oxidant examined. [Fe(o-phen) ]
,
have been documented, it has been estimated to lie well
3
6
ꢀ1 ꢀ1
only a lower limit, 1.5 · 10 M
s
could be evaluated.
Rate laws and kinetic parameters for reductions of
Fe(III) species by Ti(II) and Ti(III) are compared in Table
above 0.1 [14]. A KA value near unity (not unreasonable)
5
ꢀ1 ꢀ1
then implies a rate constant near 10 M
s
for the pro-
tonated path.
2
. Note that these reactions are quite insensitive to added
Reductions of the substitution-inert Ru(III) species are
kinetically more complex. Rate law (5) is consistent with
a competition between two reaction paths differing by a
single fluoride. Both activated complexes feature, in addi-
tion to the redox partners, a unit of Ti(IV) and (from the
halide and Ti(IV).
In contrast, the reductions of Ru(III) derivatives of
substituted pyridines (Tables S3–S7) are markedly acceler-
ated by both added fluoride and added Ti(IV), with the
Ti(IV)-dependence exhibiting kinetic saturation a high con-
centrations of this state (Table S2). Moreover, they are
inhibited by increases in acidity. The fluoride dependences
+
inverse-[H ] term) the loss of a proton. If the intervention
II
IV
of a Ti –Ti (OH) complex is assumed, an estimate of its
stability constant may be carried out by measurements at
ꢀ
IV
point to a kinetic contribution of a F containing path. At
high [Ti ] values, where the approach to kinetic saturation
[
Ti(IV)] well below the saturation region, reactions in this
is most pronounced. Data for the 4-benzoylpyridine oxi-
dant (Table S2) yield K = 104 ± 33 for this species. Analo-
gous treatments for the remaining Ru(III) oxidants give
similar values (Table 3).
group proceed according to rate law (5)
III
rate ¼ ꢀd½Ru ꢁ=dt
III
II
IV
þ ꢀ1
ꢀ
Kinetic data for reductions of Ru(III) complexes,
¼
K½Ru ꢁ½Ti ꢁ½Ti ꢁ½H ꢁ ðk þ k ½F ꢁÞ
ð5Þ
1
2
(Tables S3–S7) when refined in terms of rate law (5),
II
IV
where K pertains to the formation of a 1:1 Ti –Ti com-
plex, active in its deprotonated form. Calculated rates (in
parantheses) are compared to observed values at the right
of these tables. Rate and equilibrium constants obtained
from refinement of the Ti(II)–Ru(III) data are collected
in Table 3.
yield rate and equilibrium constants summarized in Table
3.
The modest rate increases arising from substituting
keto groups or the –CONH2 function on the pyridine
ring are comparable to those reported for the correspond-
ing reductions by Ti(III) [13], and many orders of magni-
tude less marked than those (five to seven powers of 10)
4
. Discussion
in the analogous Co(III)–Cr(II) and Co(III)–Eu(II) series
III
[
15]. The very large accelerations observed in the Co
Our experiments yield no evidence of inner-sphere paths
systems have been taken as evidence of electron transfer
via preliminary reduction of the carbonyl-bearing substit-
uents. Here, however, we see only slight enhancements
reflecting electron withdrawal from the oxidizing Ru(III)
center.
in these reductions by Ti(II). The very rapid Fe(III) oxi-
ꢀ
3þ
3+
dants, Fe(EDTA) , FeðCNÞ , and [Fe(o-phen)3] , are
6
substitution-inert, and the most reactive of this trio, the
phenanthroline derivative (k > 10 M
6
ꢀ1 ꢀ1
s ) is devoid of
bridging ligands. Its high reactivity may be reasonably
attributed to a combination of its very positive oxidation
potential (+1.14 V) and the relatively low Franck–Condon
barrier associated with passage of an electron to a highly
conjugated system.
The most notable feature of this study is the substantial
redox catalysis by titanium(IV). This cannot reflect coordi-
nation of the Ti(IV) center to the carbonyl oxygens of the
oxidants, for it remains (Keq virtually unchanged) with the
unsubstituted pyridine derivative.
3ꢀ
Two paths are evident for reaction of FeðCNÞ . The
Neither can it arise from preliminary comproportiona-
6
+
II
IV
III
[
H ]-proportional contribution pertains to reduction of
tion (Ti + Ti ! 2Ti ), yielding a more reactive Ti(III)
species, since our Ti(II) solutions undergo no perceptible
change when treated with externally prepared Ti(IV) in
excess. More significantly, early measurements dealing
with the reductions of these Ru(III) complexes by inde-
2
+
the protonated oxidant, [HFe(CN)6] . The bimolecular
rate constant for this route may, in principle, be calculated
4
ꢀ2 ꢀ1
by multiplying our kH value (8.6 · 10 M
s ) by K for
A
this dipositive species. Although this KA appears not to

3636
R. Mukherjee et al. / Inorganica Chimica Acta 360 (2007) 3633–3636
pendently prepared Ti(III) [13] yield rates lower than our
Ti(II) rates by several orders of magnitude. The close
intermediate, and the additional enhancement is reflected
in the [F ]-proportional kinetic term appearing in rate
ꢀ
similarity among the k values for the carbonyl-substi-
law (5) with all Ru(III) oxidants.
1
tuted complexes is intriguing but probably coincidental
since it does not persist with the unsubstituted pyridine
species.
We are then dealing with a Ti(II)–Ti(IV) adduct, from
which a proton has been lost (most probably from a bridg-
ing water molecule):
Acknowledgements
We are grateful to Mrs. Arla Dee McPherson for techni-
cal assistance.
Appendix A. Supplementary material
II
IV
II
Ti
IV
O+
+
H
3
Ti OH
+
H O Ti
O
H
Ti
2
2
ð6Þ
Tables S-1–S-7. Detailed kinetic data for redox reac-
tions. Supplementary data associated with this article can
be found, in the online version, at doi:10.1016/j.ica.
In attempting to explain the catalytic action of Ti(IV),
recall that Ti(II) must be considered a sluggish reactant
when its strongly negative potential is taken into account
2
007.01.026.
References
2þ
2
3þ
1
[
7]. Both the TiðH OÞ
ð3d Þ and the TiðH OÞ
ð3d Þ
2
6
2
6
cations are expected to suffer minor Jahn–Teller distor-
tions, but the degree of such distortions and the occupancy
of the non-degenerate orbitals may match poorly, resulting
in slow self-exchange rate for the hexaaquated Ti(II,III)
system. The loss of overall symmetry when the Ti(II) octa-
[1] U. Kolle, P. Kolle, Angew. Chem., Int. Edn. 85 (2003) 8565.
[2] Z. Yang, E.S. Gould, Dalton (2006) 3427.
[
[
[
Eds.), Standard Potentials in Aqueous Solutions, Marcel Dekker,
New York, NY, 1985 (Chapter 18).
[6] R.A. Marcus, Annu. Rev. Phys. Chem. 15 (1964) 155.
[
[
[
3] J.W. Olver, J.W. Ross, J. Am. Chem. Soc. 85 (1963) 8565.
4] G.S. Forbes, L.P. Hall, J. Am. Chem. Soc. 46 (1924) 385.
5] W.J. James, J.W. Johnson, in: A.J. Bard, R. Parsons, J. Jordan
(
II
IV
hedron is converted to Ti –OH–Ti may minimize struc-
tural differences between the Ti(IV)-bound divalent and
trivalent states, thus lowering the Franck–Condon barrier
to the redox process.
7] Z. Yang, E.S. Gould, Dalton (2004) 3601.
8] F.H. Burstall, R.S. Nyholm, J. Chem. Soc. (1952) 3570.
9] M.C. Ghosh, E.S. Gould, J. Am. Chem. Soc. 115 (1993) 3167.
This effect may be intensified when a portion of the
reductant is converted to the doubly bridged
[
10] B. Saha, D.M. Stanbury, Inorg. Chem. 39 (2000) 1294.
[11] R.G. Gaunder, H. Taube, Inorg. Chem. 9 (1970) 2627.
12] S. Swavey, V. Manivannan, E.S. Gould, Inorg. Chem. 40 (2001) 1312.
[13] S.K. Chandra, P.C. Paul, E.S. Gould, Inorg. Chem. 36 (1997) 4684.
[
F
II
IV
[
[
14] J. Jordan, G.J. Ewing, Inorg. Chem. 1 (1962) 587.
15] E.S. Gould, N.A. Johnson, R.B. Morland, Inorg. Chem. 15 (1976)
1929.
Ti
Ti
OH