Kinetics and mechanism of the reaction of Cr(II) aqua ions with benzoylpyridine N-oxide

Article abstract of DOI:10.1039/b701236e

Aqueous chromium(II) ions, Cr_aq²⁺, react with benzoylpyridine oxide (BPO) much more rapidly than with other pyridine N-oxides previously explored. The kinetics were studied under pseudo-first order conditions with either reagent in excess. Under both sets of conditions, the major kinetic term exhibits first order dependence on limiting reagent, and second order dependence on excess reagent, i.e. k_Cr = k₂^Cr[BPO][Cr_aq²⁺]² (excess Cr_aq²⁺), and k_BPO = k₂^BPO[Cr_aq²⁺][BPO]² (excess BPO), where k₂^Cr = (6.90 +/- 0.27) x 10⁴ M^-2s^-1 and k₂^BPO = (3.32 +/- 0.28) x 10⁵ M^-2s^-1 in 0.10 M HClO4. The rate constant k₂^Cr contains terms corresponding to [H⁺]-independent and [H⁺]-catalyzed paths. In the proposed mechanism, the initially formed Cr_aq(BPO)²⁺ engages in parallel oxidation of Cr_aq²⁺ and reduction of BPO. The latter reaction provides the basis for a convenient new preparative route for the BPO complex of Cr(III).

Full text of DOI:10.1039/b701236e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Kinetics and mechanism of the reaction of Cr(II) aqua ions with
benzoylpyridine N-oxide†
Mingming Cheng and Andreja Bakac*
Received 26th January 2007, Accepted 12th March 2007
First published as an Advance Article on the web 2nd April 2007
DOI: 10.1039/b701236e
Aqueous chromium(II) ions, Craq2+, react with benzoylpyridine oxide (BPO) much more rapidly than
with other pyridine N-oxides previously explored. The kinetics were studied under pseudo-first order
conditions with either reagent in excess. Under both sets of conditions, the major kinetic term exhibits
first order dependence on limiting reagent, and second order dependence on excess reagent, i.e. kCr
=
k2Cr[BPO][Craq2+]2 (excess Craq2+), and kBPO = k2BPO[Craq2+][BPO]2 (excess BPO), where k2 = (6.90
Cr
0.27) × 104 M−2 s−1 and k2 = (3.32 0.28) × 105 M−2 s−1 in 0.10 M HClO4. The rate constant k2
BPO
Cr
contains terms corresponding to [H+]-independent and [H+]-catalyzed paths. In the proposed
mechanism, the initially formed Craq(BPO)2+ engages in parallel oxidation of Craq and reduction of
BPO. The latter reaction provides the basis for a convenient new preparative route for the BPO complex
of Cr(III).
2+
3+
In contrast, anation reactions of Craq are typically very slow
and produce mixtures of polysubstituted complexes that require
difficult separations by ion exchange.25
Introduction
Pyridine N-oxides (PyO) are an important class of chemical
compounds. They serve as catalysts in organic synthesis1 and
in various substitution2 and rearrangement3,4 reactions, and as
promoters of epoxidations and oxidations by metal-oxo complexes
We have now extended this work to a more complex benzoylpyri-
dine oxide, PhC(O)PyO (hereafter abbreviated as BPO), and have
made some unusual observations that led to the development of a
novel method for the preparation of chromium(III) complexes of
these redox-active pyridine oxides. The details of the preparative
procedure and mechanistic studies are presented herein. The
structure of BPO and its 2-e and 4-e reduction products are shown
in Chart 1.
5–10
=
(LM O).
The chemical inertness in oxidizing environments,
and ready coordination to metal complexes are essential factors in
the latter type of reactions. Pyridine oxides can themselves act as
oxidants as well, most often as oxygen atom donors,11–13 although
one-electron processes have also been observed.14,15
The reduction of PyO by reducing transition metal complexes
2+
3+
such as Craq and Tiaq in acidic aqueous solution involves
two one-electron transfer steps and radical intermediates.14,15
3+
Surprisingly, the reduction by Tiaq (Ered = 0.1 V) under typical
conditions is faster than the reduction by the much more strongly
2+
reducing Craq (Ered = −0.41 V), and has been adopted as an
analytical, “titanometric”, method for the determination of the
N-oxide group.16 Clearly, these reactions do not take place by
simple electron transfer, but involve coordination of the oxidant
to the metal.16
Our own work with pyridine oxides has focused mainly on
the mechanistic studies of both free bases and their metal
complexes.17,18 As part of this effort, we have recently prepared a se-
ries of monosubstituted chromium complexes, Craq(X–C5H4NO)3+
(X = H, 3-CH3, 4-CH3, 4-OCH3, 4-NO2)18 by an adaptation
of the “peroxide method” that we originally developed for the
preparation of chromium(III) pyridine complexes.19 Although a
number of metal–PyO complexes are known,20–24 and many can
be prepared in simple anation reactions,25,26 the development
of the new method provides some obvious advantages. Most
notably, this method is fast and generates only 1 : 1 complexes.
Chart 1 Benzoylpyridine oxide and reduction products.
Experimental
Materials
Department of Chemistry, Iowa State University, Ames, IA, 50011. E-mail:
bakac@ameslab.gov
† Electronic supplementary information (ESI) available: Spectral and
kinetic data, Fig. S1–S6. See DOI: 10.1039/b701236e
4-Benzoylpyridine oxide (BPO) was prepared as described in the
literature.27 1H NMR (CDCl3): d 7.50–7.56 (m, 2 H), 7.65 (t, 1 H,
J = 7.38 Hz), 7.71–7.78 (m, 4 H), 8.26 (d, 2 H, J = 7.26 Hz). IR
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 2077–2082 | 2077
©

View Article Online
(photoacoustic): v(C O) 1657 cm−1, v(N–O) 1261 cm−1, d(N–O) 871 cm−1.
free conditions in an argon atmosphere. UV-Vis spectra and kinetic
data were obtained with a Shimadzu 3101 PC spectrophotometer
in a 1-cm cell. Photoacoustic spectra were obtained on a Digilab
FTS 7000 FT-IR spectrometer with Win-IR Pro (V. 3.2) software
and MTEC Photoacoustics Model 300 photoacoustic detection.
Before each data collection, the PA cell was purged with dry helium
=
UV: kmax 292 nm, e = 1.97 × 104 M−1 cm−1.
4-(a-Hydroxybenzyl)pyridine N-oxide (B(OH)PO) was pre-
pared by a modification of a published procedure28 for the
reduction of ketones. Sodium borohydride (9.5 mg, 0.25 mmol)
was added to 100 mL of an aqueous solution of BPO (2 mM) and
the pH adjusted to 10 with sodium hydroxide. The mixture was
stirred for 1 h, after which the acidity was adjusted with HClO4
to pH 2.0. The solvent was removed under reduced pressure, the
residue washed with 20 mL acetonitrile and dried in air to give
32 mg (80%) of a pale yellow solid. 1H NMR (CDCl3): d 1.25 (s, 1
H), 5.78 (s, 1 H), 7.31–7.38 (m, 7 H), 8.09 (d, 2 H, J = 7.08 Hz).
UV: kmax 262 nm, e = 1.82 × 104 M−1 cm−1.
1
for 3 min. H NMR spectra were recorded at room temperature
(20
1
◦C inside the NMR probe) with a Varian VXR 400
spectrometer. CDCl3 was employed as the solvent. The program
Kinsim31 was used to simulate kinetic traces.
Results
4-Benzoylpyridine (BP) from Avocado Research Chemicals Ltd.
was used as received. 1H NMR (CDCl3): d 7.49–7.54 (m, 2 H), 7.58
(dd, 2 H, J = 1.64 Hz, J = 4.36 Hz), 7.65 (t, 1 H, J = 7.42 Hz),
7.82 (dd, 2 H, J = 1.34 Hz, J = 8.43 Hz), 8.81 (dd, 2 H, J =
1.63 Hz, J = 4.38 Hz).
Preliminary observations
Attempts to prepare Craq(BPO)3+ by the anation reaction25 were
not successful. Most likely, the reaction was too slow at the low
experimental concentrations limited by the solubility (ca 3 mM)
2+
of BPO in aqueous solutions. When Craq was added to the
4-(a-Hydroxybenzyl)pyridine (B(OH)P) was synthesized from
BP by a modification of a literature procedure.29 Sodium borohy-
dride (47.2 mg, 1.25 mmol) was added to an ice-cold solution of
BP (183 mg, 1 mmol) in MeOH (10 mL). The mixture was stirred
for 3 h. After addition of 5 mL of water, the solution was placed
into boiling water for 10 min, then cooled and extracted with
ethyl acetate (3 × 4 mL). The solvent was removed under reduced
pressure, the residue washed with 20 mL hexane and dried in air
to give 163 mg (89%) of a white solid. 1H NMR (CDCl3): d 4.25 (s,
1 H), 5.77 (s, 1 H), 7.28–7.34 (m, 7 H), 8.09 (d, 2 H, J = 6.05 Hz).
reaction mixture in an effort to speed up the substitution process,
the unexpectedly rapid reduction of BPO ensued and yielded
Craq(BPO)3+ and Craq as major chromium products, along with
3+
trace amounts of Cr(III) dimer, (H2O)4Cr(OH)2Cr(H2O)44+. This
result prompted us to examine the Craq2+/BPO reaction in more
detail, and to optimize the preparation of Craq(BPO)3+ by this
method which, for this pyridine oxide, turned out to be superior
to the peroxide method.
Products and stoichiometry
Preparation of Craq(BPO)3+ by the peroxide method
The UV-visible spectrum of the ion exchanged product of the
Craq2+/BPO reaction was identical to the spectrum of the complex
obtained by the peroxide method, see Experimental, as shown in
Fig. 1 and S1.† The peroxide method has been used successfully
in the past to prepare Cr(III) complexes of both pyridines and
pyridine oxides.18,19 The spectrum of Craq(BPO)3+ is independent
of [H+] in the range 0.10 ≤ [H+] ≤ 1.0 M, and exhibits the following
features: kmax 584 nm (e = 40.2 M−1 cm−1), ksh 405 nm (e =
57 M−1 cm−1), kmax 260 nm (e = 2.25 × 104 M−1 cm−1).
To a stirred solution of 0.2 g BPO (1 mmol) and 0.10 g CrO3
(1 mmol) in 2 ml ice cold 1 : 2 H2O/CH3CN (v/v) was added
0.20 mL of 30% H2O2 (2 mmol). A blue paste formed immediately,
and was treated with an ice-cold H2O/CH3CN solution containing
7 equivalents of Fe2+ and 11 equivalents of HClO4 for each mole
of H2O2. The solution was loaded onto a column of Dowex-50W-
X8 cation-exchange resin. The resin was treated with a solution of
0.1 M NaSCN/1 M HClO4 followed by 1 M HClO4. The unreacted
BPO was removed with 30% aqueous CH3CN. The Cr(III)
species (Craq3+, Craq(BPO)3+, and (H2O)4Cr(OH)2Cr(H2O)44+) were
separated and eluted with 3 M HClO4.
Preparation of Craq(BPO)3+ by the Craq2+/BPO method
2+
1 mL of aqueous 0.25 M Craq was injected into 70 mL of an
aqueous solution of argon-saturated 3.6 mM BPO. The reaction
was allowed to proceed for 45 min. After dilution with 70 mL
of distilled water, the solution was loaded onto a column of
Sephadex C-25 cation-exchange resin. The separation and removal
of various species, see below, was accomplished by elution with
increasing concentrations of HClO4. The green BPO complex was
eluted with 0.40 M HClO4. The chromium content of the purified
complexes was determined spectrophotometrically after oxidizing
the chromium to chromate with alkaline peroxide.18,30
Fig. 1 UV-Vis spectrum of Craq(BPO)3+. Molar absorptivity in the main
2+
figure was divided by 104.
Solutions of Craq in dilute perchloric acid were prepared by
zinc amalgam reduction of Craq3+. Laboratory distilled water was
further purified by passage through a Barnstead EASYpure III
system. All the kinetics experiments were carried out under air-
The identity of Craq(BPO)3+ was established unequivocally by
allowing the ion-exchanged product to aquate for several weeks in
2078 | Dalton Trans., 2007, 2077–2082
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
acidic solutions, or several minutes in alkaline solutions followed
by reacidification. The final UV spectrum matched precisely that
of BPO. None of the BPO reduction products absorbs significantly
in the 292 nm region where BPO exhibits a maximum, see Fig. S2.†
The amount of BPO generated during hydrolysis was within 2% of
the concentration of chromium determined on the same sample,
which establishes a 1 : 1 ratio of Cr(III) to BPO in the complex.
The results of a spectrophotometric titration at 250 nm are
displayed in Fig. 2. The stoichiometry clearly changes throughout
the titration, and only an approximate ratio of 2 : 1 [Craq2+]/[BPO]
can be estimated from this experiment. In the early stages, when
BPO is in excess, the reaction appears to require ∼1.6 equivalents
of Craq2+ per mole of BPO. As the proportion of Craq2+ increases, so
does its consumption. Close to 3 equivalents of Craq2+ are required
for complete reduction. Similar behavior was observed at 292 nm,
Fig. S3.† At 275 nm, two breaks were observed, one at 1 : 1, and the
other at 3 : 1 ratio of [Craq2+] to [BPO], supporting the notion that
the chemistry changes with changing concentrations. The changes
in absorbance even after the addition of three full equivalents of
Craq2+ are caused by further, much slower reduction of the initially
formed B(OH)PO.
Once the spectral properties of all the relevant species were
determined, the yields of Craq(BPO)3+, and/or the loss of BPO in
redox processes was determined in 0.10 M HClO4 under kinetic
conditions with two sets of concentrations. In one experiment,
equimolar amounts of Craq2+ and BPO (0.10 mM each) were used.
After the completion of the reaction, the solution was basified to
pH 11 to induce hydrolysis of the chromium complex(es). After
reacidification, 80% ofthe initialBPO was recoveredas determined
from the UV spectrum, Fig. S4.† Owing to the approximate 2 :
1 stoichiometry, only about 50% of BPO actually reacted with
Craq2+, so that the yield of Craq(BPO)3+ was about 60% of the
reacting BPO, in good agreement with the data in ion-exchange
experiment.
2+
In another experiment, using Craq (0.50 mM) in excess over
BPO (0.070 mM), only about 20% of BPO was recovered at the
end. Moreover, there were additional features in the spectrum
suggesting that some B(OH)PO may also be generated in the
hydrolysis step. To check whether the chromium complex of
B(OH)PO was involved, a large-scale preparation was carried
2+
out using a two-fold excess of Craq over BPO, followed by ion-
exchange on Sephadex C25. The early fractions of Craq(BPO)3+
exhibited UV spectra that were identical to that in Fig. S1†
(kmax 262 nm), and yielded quantitative amounts of BPO upon
hydrolysis. The UV maximum of the later, more dilute fractions,
however, shifted to shorter wavelengths (∼258 nm), and hydrolysis
yielded both BPO and B(OH)PO, as shown for one such fraction
in Fig. S5†. We estimate that at least 20% of the total chromium
complex was present as Craq(B(OH)PO)3+.
Kinetics
Most of the measurements were carried out with a large (> 10-
fold) excess of Craq2+. Kinetic traces were exponential and yielded
pseudo-first order rate constants kCr. The dependence of kCr on
[Craq2+] was close to second order as shown by a linear plot of
kCr vs [Craq2+]2, or of kCr/[Craq2+] vs [Craq2+]. The latter is displayed
in Fig. 3 for a data set obtained in 0.10 M HClO4 at [BPO] =
0.020 mM. The line has a slope k2 = (6.90 0.27) × 104 M−2 s−1
and a small intercept, k1 = 5.35 1.71 M−1 s−1, indicative of a
Cr
Fig. 2 Plot of absorbance vs concentration ratio [Craq2+]/[BPO] during
a spectrophotometric titration of 0.136 mM BPO with 5.0 mM Craq at
minor parallel pathway that is first order in each reagent. The rate
2+
250 nm.
Detailed product analysis was carried out for a preparation
that started with equimolar amounts (0.25 mmol) of the two
reagents in 0.10 M HClO4, as described in the Experimental.
The following species were separated on a column of Sephadex
C-25 and identified by UV-visible spectroscopy: unreacted BPO
(0.091 mmol), and BPO reduction products that could not be
completely separated, but exhibited UV spectra consistent with
a mixture of 4-benzoylpyridine (BP), 4-(a-hydroxybenzyl)pyridine
(B(OH)P) and 4-(a-hydroxybenzyl)pyridine N-oxide (B(OH)PO).
This was followed by Craq3+ (0.08 mmol), Craq(BPO)3+ (0.08 mmol),
4+
and (H2O)4Cr(OH)2Cr(H2O)4 (0.01 mmol Cr). In total, ∼70%
of the chromium was recovered. The missing portion was most
probably present as the weakly absorbing monomeric and dimeric
aquachromium(III) ions. The strong UV features of BPO and
its various reduced forms makes chromium complexes of these
ligands more easily detectable even at low concentrations.
2+
Fig. 3 Plot of kCr/[Craq2+] against the concentration of Craq for the
reaction with BPO (2.0 × 10−5 M) in 0.10 M HClO4.
Dalton Trans., 2007, 2077–2082 | 2079
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
law applicable under these conditions is given in eqn (1)
−d[BPO]/dt = kCr[BPO] = (k1Cr[Craq2+] + k2Cr[Craq2+]2)[BPO] (1)
The apparent pseudo-first order rate constant measured under
such conditions is a mild function of the concentration of BPO,
the limiting reagent. At 1.0 mM Craq2+, the rate constant kCr
=
0.0662 s−1 ([BPO] = 0.015 mM), and 0.0739 s−1 (0.040 mM). As
shown later, this result is consistent with the involvement of an
2+
intermediate that reacts with both Craq and BPO.
Several experiments were also carried out with BPO in excess
over Craq2+. Under these conditions, the concentration ranges of
both reactants were quite limited owing to the low solubility and
strong UV absorption of BPO. The kinetics were monitored at
315 nm to reduce the background absorbance of BPO. Within
these restrictions, the fits to first-order kinetics were excellent,
and the first-order rate constants independent of [Craq2+] (0.020–
0.060 mM), establishing first-order dependence on [Craq2+]. The
plot of kBPO/[BPO] vs [BPO] is linear, as shown in Fig. 4, although
the long extrapolation and narrow concentration range make the
intercept quite uncertain. The line in Fig. 4 is described by eqn
Fig. 5 Plot of kCr/[Cr2+]2 vs [H+] at 1.0 M ionic strength.
in eqn (4), where eOH and eH represent molar absorptivities of
the unprotonated and protonated forms, respectively, eobs is the
experimental value at a given [H+], and Ka is defined in eqn (5).
The fit, shown in Fig. 6, yielded eOH = 2.05 × 104 M−1 cm−1, eH =
9.64 × 103 M−1 cm−1, and Ka = 1.15 0.26 M.
eOHKa + eH[H+]
(2), where k1 = 18.8 4.1 M−1 s−1 and k2 = (3.32 0.28) ×
BPO
BPO
eobs
=
(4)
Ka + [H+]
105 M−2 s−1.
BPOH+ ꢀ BPO + H+ Ka
(5)
−d[Cr2+]/dt = kBPO[Craq2+] = (k1BPO[BPO] +
2+
k2BPO[BPO]2)[Craq
]
(2)
The values of Ka obtained by the two approaches, 1.39 M and
1.15 M, are reasonably close, although the direct determination
in Fig. 6 resulted in a more precise number. The data in Fig. 5
were refitted to eqn (3) with c fixed at 1.15 M to yield a = (6.83
0.63) × 104 M−1 s−1 and b = (4.34 0.18) × 105 M−2 s−1.
Fig. 4 Plot of kBPO/[BPO] against the concentration of BPO for the
2+
reaction with 20–60 lM Craq in 0.10 M HClO4.
H+-dependence was examined in a series of experiments under
pseudo-first order conditions using excess Craq (1.0 × 10−3 M)
2+
Fig. 6 Plot of the observed molar absorptivity of BPO at 292 nm vs [H+]
at 1.0 M ionic strength. The line is the fit to eqn (4).
over BPO (2.0 × 10−5 M) at 1.0 M ionic strength. Independent
experiments at both the high and low end of [H+] confirmed that
the dependence on [Craq2+] remained predominantly second-order
with only a minor first order component that contributed no more
than 3% to the measured rate constants in Fig. 5. The data were
fitted to eqn (3) and yielded the following parameters: a = (9.2
3.9) × 104 M−1 s−1, b = (4.35 0.98) × 105 M−2 s−1, and c = 1.39
0.53 M.
Related chemistry
Several experiments were carried out to check whether side
reactions influenced the Craq2+/BPO reaction. In one experi-
2+
ment, Craq (1 mM) was added to an ion-exchanged sample of
Craq(BPO)3+ (0.10 mM) in 0.10 M HClO4, and the absorbance was
monitored at 260 nm. The decrease was much faster than in the
absence of Craq2+, but the apparent rate seemed to decrease much
kCr/[Cr2+]2 = (a + b[H+])/(c + [H+])
As discussed later, parameter c in the proposed mechanism
is the acidity constant of BPO. This parameter was determined
independently from the UV spectral changes shown in Fig. S6.†
The absorbance data at 292 nm were fitted to the expression
(3)
2+
faster than predicted by first-order kinetics. Most likely, Craq
catalyzes the aquation of the complex to generate BPO, which
then reacts with excess Craq2+ to form Craq(B(OH)PO)3+ and some
Craq(BPO)3+, both of which absorb at 260 nm. From the initial
2080 | Dalton Trans., 2007, 2077–2082
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
rates, we estimate a rate constant of ca 2 × 10−4 s−1 for the Craq
-
organochromium intermediate, would complete the sequence in
both cases.
2+
catalyzed aquation of Craq(BPO)3+ under these conditions.
2+
We also briefly checked for a reaction between Craq and the
products of reduction of BPO. Neither B(OH)PO nor BP reacted
measurably on the time scale of the Craq2+/BPO reaction. Also,
•
Craq(BPO)2+ + H2O → Craq + BPO − k9
(9)
3+
k9K6[Cr2+][BPO]T
−d[BPO]/dt =
2+
no reaction was observed between Craq and benzophenone,
(10)
(11)
1 + K6[Cr2+]
3+
(C6H5)2CO.
•
Craq + BPO → Craq + BPO − k11
2+
−d[BPO]/dt = k11 [Cr2+][BPO]
(12)
Discussion
The preparation of Craq(BPO)3+ by the Craq2+ reduction of BPO is
much simpler and faster than the peroxide method. Only a single
Cr(III) complex is generated provided the ratio [BPO]/[Craq2+] ≥
1. At lower ratios, some Craq(B(OH)PO)3+ is also produced.
•
BPO − + Craq (+ 2H+) → BOHPO + Craq fast
(13)
2+
3+
2+
2
The acid dependence of the dominant kinetic term, k2[Craq
]
[BPO], provides important insight into the mechanism. The
complex formation in eqn (6) is almost certainly not acid catalyzed,
and is probably impeded by the protonation at the PyO oxygen.
The source of the positive effect of [H+] on the reaction thus
resides in the electron transfer step of eqn (7), and perhaps also
in reactions 9 and 11, although the small contribution of these
pathways makes the acid dependence quite uncertain.
The reaction in eqn (7) requires H+ in the stoichiometric sense,
and it is reasonable to expect the proton to be involved in
the transition state as well. A strong catalytic effect of H+ was
also observed in our earlier study on the reduction of pyridine
oxide with hydroxyisopropyl radicals.17 In the current work, the
protonation is viewed as taking place at the coordinated BPO,
as shown in eqn (14), followed by intramolecular transfer of
two electrons, one from each chromium(II), and protonation at
carbon. In a parallel, acid-independent path (k0), represented
by the intercept in Fig. 5, the protonation at oxygen apparently
takes place after the rate-determining electron transfer from the
carbonyl oxygen-bound chromium(II).
2+
The Craq reduction of BPO is remarkably fast in comparison
with that of other pyridine oxides (X-PyO) for which the rate
constants span a range 9.6 × 10−5 M−1 s−1 (X = H) to 1.2 M−1 s−1
(X = 4-COOH) in 1.0 M HClO4.14 The different rate law observed
in the previous study14 rule out strict comparisons, but pseudo-first
order rate constants in 1.0 M HClO4 differ widely. The reaction
2+
of 1 mM Craq with limiting [BPO] has k = 0.22 s−1, that with
unsubstituted pyO has k = 9.6 × 10−8 s−1, and even 4-COOH-
PyO, the fastest among the previously studied pyridine oxides,14
still reacts ∼200 times more slowly (k = 1.2 × 10−3 s−1) than
BPO. At higher [Craq2+], the differences are even greater because
2+
of the second-order dependence on Craq in the BPO reaction.
The obvious difference between BPO and the pyridine oxides
studied earlier14 is the presence of the keto group. In fact, it
is this functionality, and not the pyridine oxide in BPO that
2+
Craq reduces. Both sites are critical however, as shown by the
immeasurably slow reduction of B(OH)PO and of benzophenone,
each of which lacks one of the two BPO functionalities. As
shown in the mechanism below, the pyO site is critical because
it allows coordination to chromium, while the keto group presents
a reducible site.
This mechanism leads to the acid dependence in eqn (15), where
k2 describes the major kinetic term in eqn (1). Eqn (15) has the
same form as eqn (3), with a = k0K6Ka, b = kHK6Ka, and c = Ka =
1.15 M. In deriving eqn (15), we assumed that the protonation
equilibrium in the first step in reaction 14 lies far to the left.
Cr
At a constant [H+], the kinetic data in the presence of excess
2+
Craq suggests the mechanism in eqn (6) and (7) which yields the
rate law of eqn (8), where [BPO]T represents the sum {[BPO] +
[Craq(BPO)2+]}.
Craq + BPO ꢀ Craq (BPO)2+ + H2O K6
(6)
(7)
2+
2+
3+
Craq(BPO)2+ + Craq (+ 2H+) → 2 Craq + B(OH)PO k7
(14)
k7K6[Cr2+]2[BPO]T
−d[BPO]/dt =
(k0 + kH[H+])K6Ka
(8)
1 + K6[Cr2+]
kC2 r
=
(15)
Ka + [H+]
In the limit K6[Cr2+] ꢀ 1, eqn (8) reduces to the second term in
In the absence of an independent value of the equilibrium
constant K6 for the formation of Craq(BPO)2+, the absolute values
of the rate constants kH and k0 cannot be calculated.
Cr
the observed rate law of eqn (1), where k2 = k7K6. The lack of
any spectral evidence for Craq(BPO)2+ in the visible or UV range
under our experimental conditions is consistent with a small value
of K6.
2+
The site of Craq binding in eqn (14) is shown to be the
carbonyl oxygen, although we cannot rule out the possibility
that it takes place at the carbonyl carbon to generate a transient
organochromium complex. The final chromium products shown
in eqn (14) are consistent with the finding that, in addition to
BPO, some B(OH)PO was generated by base hydrolysis of late
fractions of the ion exchanged chromium complex. The amount of
The first, minor term in eqn (1) is first order in each reagent,
consistent with either a unimolecular decay of Craq(BPO)2+, eqn (9)
and (10), or a parallel reduction of BPO by Craq2+, eqn (11)
and (12). Rapid follow-up reactions of the radical BPO•−, such
2+
as the reduction with Craq in eqn (13), perhaps involving an
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 2077–2082 | 2081
©

View Article Online
Craq(B(OH)PO)3+ was, however, small, suggesting that a portion of
the binuclear intermediate had hydrolyzed under the experimental
Acknowledgements
3+
2+
We are grateful to Dr Roger Jones for help in obtaining the
photoacoustic spectra and to Dr E. Szajna-Fuller for help with
the NMR spectra. This work was supported by a grant from the
National Science Foundation, CHE 0602183. Some of the work
was conducted with the use of facilities at the Ames Laboratory.
conditions to give Craq and free B(OH)PO, perhaps in a Craq
catalyzed reaction similar to that observed with Craq(BPO)3+.
-
None of the steps invoked so far generates Craq(BPO)3+, a
product that was observed in a small yield even in the experiments
using excess Craq2+. Ignoring for the moment the acid dependence,
the composition of the transition states for the two kinetic terms
are {2Cr2+, BPO} and {Cr2+, BPO}. Clearly, neither one of these
will lead to Craq(BPO)3+ without further oxidation. The source of
Craq(BPO)3+ in the experiments using excess Craq2+ will be discussed
later.
References
1 H. Hagiwara, H. Inoguchi, M. Fukushima, T. Hoshi and T. Suzuki,
Synlett, 2005, 2388–2390.
2 Y. S. Chang and J.-J. Jwo, J. Mol. Catal. A: Chem., 2000, 160, 357–366.
3 K. Harano, I. Shinohara, M. Murase and T. Hisano, Heterocycles,
1987, 26, 2583–2586.
4 K. Harano, H. Nakagawa, K. Kamei, H. Kiyonaga and T. Hisano,
Chem. Pharm. Bull., 1992, 40, 1675–1682.
In the presence of excess BPO, the major kinetic term is
second order in [BPO] and first order in [Craq2+], suggesting
that Craq(BPO)2+ is now the major reductant for BPO, eqn (16).
5 E. G. Samsel, K. Srinivasan and J. K. Kochi, J. Am. Chem. Soc., 1985,
107, 7606–7617.
The inequality K6[BPO] ꢀ 1 would simplify the rate law of
BPO
eqn (17) to the second (major) term of eqn (2), with k2
=
6 J. P. Collman, L. Zeng and J. I. Brauman, Inorg. Chem., 2004, 43,
k16K6. Detailed acid dependence was not explored under these
conditions, but qualitatively the reaction rate did increase at
higher [H+], consistent with the greater reactivity of the protonated
form,17 BPOH+, in the redox step.
2672–2679.
7 N. S. Venkataramanan, S. Premsingh, S. Rajagopal and K. J. Pitchu-
mani, J. Org. Chem., 2003, 68, 7460–7470.
8 Z. Gross and S. Ini, Inorg. Chem., 1999, 38, 1446–1449.
9 R. I. Kureshy, I. Ahmad, N.-u. H. Khan, S. H. R. Abdi, K. Pathak and
R. V. Jasra, J. Catal., 2006, 238, 134–141.
10 R. I. Kureshy, N.-u. H. Khan, S. H. R. Abdi, S. T. Patel, P. K. Iyer and
R. V. Jasra, Tetrahedron Lett., 2002, 43, 2665–2668.
11 R. R. Conry and J. M. Mayer, Inorg. Chem., 1990, 29, 4862–4867.
12 J. H. Espenson, Adv. Inorg. Chem., 2003, 54, 157–202.
13 R. Ito, N. Umezawa and T. J. Higuchi, J. Am. Chem. Soc., 2005, 127,
834–835.
Craq(BPO)2+ + BPO → Craq(BPO)3+ + BPO•−k16
(16)
k16K6[Cr2+]T[BPO]2
−d[Cr2+]/dt =
(17)
1 + K6[BPO]
Reaction 16 is also the most likely source of the small yields of
Craq(BPO)3+ under conditions of excess Craq2+. In 0.10 M HClO4,
the rate constant for reaction 16 is about five times larger than
that for reaction 7, which makes even limiting [BPO] competitive
with excess [Craq2+], especially in the early stages of the reaction
before [BPO] is significantly depleted. The observation that the
rate constants show a mild dependence on the limiting reagent
when Craq2+ is in excess, but not when BPO is in excess, is also fully
consistent with the proposed scheme and derived rate constants.
Kinetic simulations with Kinsim31 also confirm these dependences,
as well as the fact that the appearance of kinetic traces remains
exponential under all of the conditions employed.
14 A. P. Zipp and R. O. Ragsdale, J. Chem. Soc., Dalton Trans., 1976,
2452–2455.
15 T. Everton, A. P. Zipp and R. O. Ragsdale, J. Chem. Soc., Dalton Trans.,
1976, 2449–2452.
16 R. T. Brooks and P. D. Sternglanz, Anal. Chem., 1959, 31, 561–565.
17 A. Bakac, V. Butkovic, J. H. Espenson, J. Lovric and M. Orhanovic,
Inorg. Chem., 1996, 35, 5168–5172.
18 M. Kotowski, R. Marcec, V. Butkovic, A. Bakac and M. Orhanovic,
Eur. J. Inorg. Chem., 2006, 2894–2899.
19 A. Bakac, R. Marcec and M. Orhanovic, Inorg. Chem., 1974, 13, 57–60.
20 W. L. Purcell, Inorg. Chem., 1989, 28, 2312–2315.
21 D. W. Meek, R. S. Drago and T. S. Piper, Inorg. Chem., 1962, 1, 285–289.
22 R. E. Kohrman, P. G. Phadtare and D. X. West, J. Inorg. Nucl. Chem.,
1975, 37, 301–303.
Cr
BPO
The intercepts in Fig. 3 (k1 = 5.35 M−1 s−1) and Fig. 4 (k1
=
23 Y. K. S. Kakiuti and J. V. Quagliano, Spectrochim. Acta, 1963, 19,
18.8 M−1 s−1) represent an overall second order term, as described
by eqn (9) and/or (11). If eqn (13) correctly describes the rapid
follow-up step under both sets of conditions, the rate law in eqn
(18) applies, where k stands for either k9K6 or k11.
201–211.
24 D. A. Edwards and S. C. Jennison, Transition Met. Chem. (Dordrecht),
1981, 6, 235–238.
25 T. J., Jr. Weeks and E. L. King, J. Am. Chem. Soc., 1968, 90, 2545–2550.
26 R. L. Carlin, J. Am. Chem. Soc., 1961, 83, 3773–3775.
27 C. W. Muth, R. S. Darlak and J. C. Patton, J. Heterocycl. Chem., 1972,
9, 1003–1007.
28 O. O. Blumenfeld and P. M. Gallop, Proc. Natl. Acad. Sci. USA, 1966,
56, 1260–1267.
29 S. M. N. Efange, R. H. Michelson, R. P. Remmel, R. J. Boudreau, A. K.
Dutta and A. J. Freshler, J. Med. Chem., 1990, 33, 3133–3138.
30 G. W. Haupt, J. Res. Natl. Bur. Stand. (U. S.), 1952, 48, 2331.
31 B. A. Barshop, R. F. Wrenn and C. Frieden, Anal. Biochem., 1983, 130,
134–145.
−d[Craq2+]/dt = −2d[BPO]/dt = k[Craq2+][BPO]
(18)
The observed rate constants should differ by a factor of
BPO
Cr
two depending on which reagent is limiting, i.e. k1
= 2k1
.
The experimentally observed factor that is somewhat greater
than 3 seems acceptable in view of the low precision of both
intercepts.
2082 | Dalton Trans., 2007, 2077–2082
This journal is
The Royal Society of Chemistry 2007
©