Kinetics and mechanism of oxygen atom abstraction from sulfenatocobalt(III) complexes by hydrated methyldioxorhenium(V)

Article abstract of DOI:10.1021/ic990425q

Cationic sulfenatocobalt(III) complexes, with the remaining ligands amines, are deoxygenated upon reaction with hydrated methyldioxorhenium(V). The reactions of (Am)₅Co-S(O)R²⁺ yield the thiolato complexes (Am)₅Co-SR²⁺ and methyltrioxorhenium(VII), MTO. A kinetic study of these reactions was carried out, using the known reaction between MTO and hypophosphorous acid to prepare the Re(V) reagent in solution. The second-order rate constants between (Am)₅Co-S(O)R²⁺ and MeReO₂(aq), after correction for protonation, fall in the range 47-455 L mol^-1 s^-1 in aqueous solution at 25.0 °C and 1.0 M ionic strength. In the analysis of the pH effects, allowance was made for the formation of the protonated species (Am)₅Co-S(OH)R³⁺ at high [H₃O⁺]. The values of pK_a are in the range 0.49-0.96, as determined from fitting the rate-pH profiles.

Full text of DOI:10.1021/ic990425q

5230
Inorg. Chem. 1999, 38, 5230-5234
Kinetics and Mechanism of Oxygen Atom Abstraction from Sulfenatocobalt(III) Complexes
by Hydrated Methyldioxorhenium(V)
David W. Lahti and James H. Espenson*
Ames Laboratory and Department of Chemistry, Iowa State University of Science and Technology,
Ames, Iowa 50011
ReceiVed April 20, 1999
Cationic sulfenatocobalt(III) complexes, with the remaining ligands amines, are deoxygenated upon reaction with
hydrated methyldioxorhenium(V). The reactions of (Am)5Co-S(O)R2+ yield the thiolato complexes (Am)5Co-
SR2+ and methyltrioxorhenium(VII), MTO. A kinetic study of these reactions was carried out, using the known
reaction between MTO and hypophosphorous acid to prepare the Re(V) reagent in solution. The second-order
rate constants between (Am)5Co-S(O)R2+ and MeReO2(aq), after correction for protonation, fall in the range
47-455 L mol-1 s-1 in aqueous solution at 25.0 °C and 1.0 M ionic strength. In the analysis of the pH effects,
allowance was made for the formation of the protonated species (Am)5Co-S(OH)R3+ at high [H3O+]. The values
of pKa are in the range 0.49-0.96, as determined from fitting the rate-pH profiles.
Introduction
are the analogues of organic sulfoxides. Thiols (e.g., from
cysteine) are oxidized in a protein environment.5 It should be
noted that considerable research on the oxidation of nickel and
palladium sulfenates and sulfinates has been reported.6 We have
examined cobalt(III) complexes, (Am)5Co-S(O)nR2+ (Am )
amine, n ) 1, 2).5 The structural formulas of the sulfenato (n
) 1) complexes are displayed in Chart 1 designated with the
suffix b; the suffix a is used for thiolate (n ) 0) and c for
sulfinato (n ) 2) complexes, whose structures are not explicitly
shown.
The rhenium(V) complex CH3ReO2, presumably the diaqua
complex CH3ReO2(OH2)2 (aqua ligands hereinafter omitted),
has been shown to abstract oxygen atoms from certain reagents,
as in the reaction
XO + CH3ReO2 f X + CH3ReO3
(1)
Examples of the XO reagents for this reaction include C5H5-
NO, R2SO, Ar3AsO, Ar3SbO, etc.1 The Re(V) reagent is so
kinetically aggressive that even perchlorate ions react in a
cascade of similar steps that eventually forms chloride ions:2
Results and Discussion
The UV-vis spectral data of the thiolato- and sulfenatocobalt
complexes are given in Table S-1 (Supporting Information). The
thiolates have a characteristic LMCT band near 285 nm. The
sulfenates have an intense peak near 365 nm. Certain special
cases deserve mention. Complex 4 can exist as either of
two isomers. The geometry for complex 4b depicted in Chart 1
presents the structure determined from the 13C NMR
spectrum.7 The trien complex 5a was also made in the course
of this study. It was fully characterized and analyzed. Since it
was otherwise difficult to spectroscopically ascertain with
certainty the geometry, its structure was determined by crystal-
lography. 8 The trien ligand adopts the â mode of coordination
in which three of the nitrogen atoms of the ligand are in the
equatorial plane of the complex,8 as displayed in Chart 1.
Efforts then were directed toward the sulfenato series. The
conditions for preparing solutions of CH3ReO2 useful for
kinetics are demanding. The acid concentration must be high;
we chose 0.1-1.0 M H3O+ for the most part. Without that, the
colorless solutions of CH3ReO2 turned a blue color that strongly
CH3ReO2
CH3ReO2
8 f f Cl-
(2)
-
-
ClO4
8 ClO3
The driving force for these reactions is the formation of a
strong Re-O bond. The value of ∆G° for reaction 3 was
CH3ReO3(aq) f CH3ReO2(aq) + O(g)
(3)
established by kinetic methods to be 111 kcal mol-1.1 This value
places a limit on the feasibility of substrates for reaction 2;
phosphine oxides in place of perchlorate ions, for example, are
unable to participate. On the other hand, the thermodynamic
values tell only what is feasible. Dinitrogen monoxide, for
example, does not react with CH3ReO2, despite the favorable
value ∆G° ) -30 kcal mol-1, which clearly shows the reaction
is spontaneous.
Since oxygen atom transfers hold biological and industrial
relevance,3,4 and are of fundamental interest as well, it seemed
useful to extend the essence of reaction 1 to a broader and
potentially more demanding range of substrates. In this study
we consider the reactions of certain sulfenato complexes that
(5) (a) Adzamli, I. K.; Libson, K.; Lydon, J. D.; Elder, R. C.; Deutsch, E.
Inorg. Chem. 1979, 18, 303-311. (b) Adzamli, I. K.; Deutsch, E.
Inorg. Chem. 1980, 19, 1366.
(1) Abu-Omar, M. M.; Appleman, E. H.; Espenson, J. H. Inorg. Chem.
1996, 35, 7751-7757.
(2) Abu-Omar, M. M.; Espenson, J. H. Inorg. Chem. 1995, 34, 6239-
(6) (a) Buonomo, R. M.; Font, I.; Maguire, M. J.; Reibenspies, J. H.;
Tuntulani, T.; Darensbourg, M. Y. J. Am. Chem. Soc. 1995, 117, 963-
973. (b) Darensbourg, M. Y.; Tuntulani, T.; Reibenspies, J. H. Inorg.
Chem. 1995, 34, 6287-6294.
6240.
(3) Holm, R. H. Chem. ReV. 1987, 87, 1401-1449.
(4) Woo, L. K. Chem. ReV. 1993, 93, 4515-4523.
(7) Jackson, W. G.; Sargeson, A. M. Inorg. Chem. 1978, 17, 2165.
10.1021/ic990425q CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/21/1999

O Atom Abstraction from (Am)5Co-S(O)nR2+
Inorganic Chemistry, Vol. 38, No. 23, 1999 5231
Chart 1
Figure 1. Absorbance-time recording of the reaction between (1b)-
Cl2 (1.0 mM) and H2P(O)OH (50 mM) catalyzed by CH3ReO3 (84 µM).
Conditions: 0.13 M HCl in 1:1 aqueous acetonitrile at 298 K.
mol-1 s-1 at 298 K. This agrees with the accepted value, k4 )
1
2.8(2) × 10-2 L mol-1 s-1
.
The rate constant of reaction 4
was insensitive to the value of [H+] in the range 0.20-1.0 M.
The rate constant of reaction 5 was also determined in aqueous
acetonitrile (1:1): k4 ) 3.7(3) × 10-3 L mol-1 s-1 at 298 K,
some 7 times lower than in aqueous solution. This is not simply
the effect of a presumably innocent solvent change. It appears
as if, to an appreciable but undetermined extent, that acetonitrile
has replaced the solvating water molecules coordinated to the
rhenium atom of CH3ReO2.
intensified with time, ultimately depositing a blue-black solid.
This material contains an intact Re-CH3 group, since its
treatment with hydrogen peroxide generates the well-known
peroxo complex CH3Re(O)(η2-O2)2(OH2).1 Another requirement
for the kinetic studies is that the total rhenium concentration
be kept quite low (usually <0.25 mM) for the same reason.
We presume the blue substance to be [CH3ReO]2(µ-O)2, by
analogy to the derivative with Cp* in place of CH3.9 Solutions
of CH3ReO2 were made by reducing CH3ReO3 with aqueous
hypophosphorous acid:
It followed that a catalytic system could be constructed,
featuring the repetitive cycling of the two rhenium species
through reactions 4 and 5 in sequence, with the anticipated result
H2P(O)OH + (Am)5Co-S(O)R2+
f
CH3ReO3 + H2P(O)OH f CH3ReO2 + HP(O)(OH)2
(4)
HP(O)(OH)2 + (Am)5Co-SR2+ (6)
The reaction is not particularly rapid (k4 ) 2.8 × 10-2 L mol-1
s-1 at 298 K), but that causes no difficulty. Solutions of CH3-
ReO2 so prepared are stable for 1-3 h.
The general procedure for the kinetics studies was to generate
CH3ReO2 first and then add the sulfenato complex. Its concen-
tration could be set much smaller than that of CH3ReO2, given
the strong absorption bands; thus, first-order kinetics were
closely followed. The reactions occur as in
Figure 1 shows data for one such experiment for (1b)Cl2. It
was carried out with 84 µM MTO, 1.0 mM (1b)Cl2, and 50
mM H2P(O)OH. The absorbance of species 1b fell to nearly
zero, consistent with reaction 6 going to completion. During
most of the course of the reaction, the data follow zero-order
kinetics. This is consistent with reaction 4 being rate-controlling,
since the concentration of H2P(O)OH remains effectively
constant because it is present in large excess and because it is
the catalyst and is continuously regenerated.
For the study of reaction 5, solutions of CH3ReO2 were
prepared by the prior reaction of CH3ReO3 with excess H2P-
(O)OH. After the necessary time had passed (5 half-lives for
reaction 4), the sulfenatocobalt complex was then added. Its
concentration was chosen to be e 10[MeReO2]0. The reaction
was followed at one or more appropriate wavelengths. The data
at each wavelength were fitted to pseudo-first-order kinetics.
To do so, a nonlinear least-squares fit to the rate law in eq 7
CH3ReO2 + (Am)5Co-S(O)R2+
f
CH3ReO3 + (Am)5Co-SR2+ (5)
Kinetic data were obtained by monitoring the loss of the
sulfenato complex or occasionally the buildup of the thiolato
product. In some instances, CH3ReO3, H2P(O)OH, and (Am)5-
Co-S(O)R2+ were mixed concurrently. The two reaction stages,
reaction 4 followed by 5, occurred in sequence. The rate of
reaction 4 is much lower than that of reaction 5, and the rate of
conversion of the sulfenate to the thiolate complex was governed
by it. The rates from these experiments were proportional to
each of the concentrations of H2P(O)OH and CH3ReO3, and
Abst ) Abs∞ + (Abs0 - Abs∞) exp(-kψt)
(7)
gave rate constants in the range k4 ) (2.4(1)-2.7(1)) × 10-2
L
was carried out. For each complex, the plot of kψ against the
total rhenium concentration of the cobalt complex was linear.
This treatment establishes that reaction 5 follows mixed second-
order kinetics.
The sulfenatocobalt complexes are known to be weak
Brønsted bases:
(8) X-ray crystal data for C8H24Cl2CoN5O8S: monoclinic, P21/n, a )
8.900(4) Å, b ) 12.458(5) Å, c ) 15.974(5) Å, b ) 90.00(1), V )
1771.2(12) Å3, Z ) 4, T ) 173(2) K, Dcalcd ) 1.801 g/cm3, R(F) )
14.78% [I ) 2σ(I)]. Owing to the poor crystal quality only the atom
connectivity was established with certainty. The software and sources
of the scattering factors are contained in the SHELXTL (version 5.1)
program library (G. Sheldrick, Bruker Analytical X-ray Systems,
Madison, WI).
(Am)5Co-S(OH)R3+ a H+ +
(9) Herrmann, W. A.; Flo¨el, M.; Kulpe, J.; Felixberger, J.; Herdtweck,
E. J. Organomet. Chem. 1988, 355, 297-313.
(Am)5Co-S(O)R2+
(Ka) (8)

5232 Inorganic Chemistry, Vol. 38, No. 23, 1999
Lahti and Espenson
UV-vis spectrum that accompanies the shift in position of
reaction 7. Figure S-3 shows a family of juxtaposed spectra of
complex 1b taken over the [H+] range of interest.
Table 1 presents both sets of pKa values. In some cases the
agreement is excellent, in others only fair. Considering the
extrapolations being made, the overall adherence to the proposed
reaction scheme seems acceptable. Certainly there is no simple
alternative that would suffice. In all cases pKa(kinetic) < pKa-
(equil). Equation 9 could be brought into agreement with the
independent value of pKa by having an added numerator term,
as in
kapp ) (k5 + k5′[H+]-1)/(1 + [H+]/Ka)
(10)
in which the term k5′ represents a conjugate base (CB)
enhancement. This alternative is not without a certain appeal
in the molecular mechanism, but no independent evidence for
it could be obtained. Moreover, there is a case in which the
two pKa values agree well. There is no reason to admit the
possibility of the CB interpretation in certain cases, only to
disallow it in others. Furthermore, the k5′ value needed to
accommodate the fits is unreasonably large. For these reasons
we return to the simpler scheme.
Figure 2. pH profiles for selected sulfenato complexes. The lines
represent least-squares fitting to eq 9. Data are shown for complexes
1b (open triangles), 4b (filled triangles), and 6b (squares).
Table 1. pKa Values for Sulfenatocobalt(III) Complexes and
Kinetic Data for Oxygen Atom Transfer Reactions from
Sulfenatocobalt(III) and CH3ReO2 Complexesa
entry
k5/(L mol-1 s-1
)
pKa (kinetics)
pKa (equil)
The noteworthy features of the data are these: the sulfenato
complexes transfer an oxygen atom to CH3ReO2 readily enough,
whereas their O-protonated forms do not. The sulfinato com-
plexes do not suffer oxygen loss upon treatment with CH3ReO2.
That feature is not unique with these Co(III) complexes:
sulfones themselves, R2SO2, do not transfer an oxygen atom to
CH3ReO2, whereas sulfoxides do so readily. Clearly, coordina-
tion, ionic charge, etc. do not provide the difference. The failure
of sulfones and sulfinatocobalt(III) complexes must also be
examined in light of the comparative bond energies. It is well-
known that sulfur-oxygen bond free energies of sulfones are
greater than those of sulfoxides. This amounts to a difference
of 25 kcal. For example, the family of reactions R2SOn f
R2SOn-1 + O(g) has the following values of bond dissociation
energies (kcal mol-1): 112 (R ) Me) and 113 (R ) Ph) for
sulfones (n ) 2), as compared to 86.6 (R ) Me) and 89.3 (R
) Ph) for sulfoxides (n ) 1).10
We assert, however, that the higher value of the driving force
is (a) thermodynamically insufficient to prevent the reaction of
sulfones and sulfinates, given ∆G° ) 111 kcal for reaction 3,
(b) incidental, as bond enthalpy values, although providing a
prerequisite for reaction, seem largely immaterial to the question
of reactivity (it should be noted that even N2Oaq, with ∆G° )
82 kcal, the weakest oxygen bond of all those considered, fails
to react with CH3ReO2), and (c) not at all reflected in the rate
constants for the transfer of an oxygen atom to a sulfur of an
organic sulfide or sulfoxide,11 as shown in
1b
2b
3b
4b
5b
6b
7b
8b
305(37)
355(24)
220(46)
116(13)
47(2)
0.74
0.49
0.72
0.96
0.60
0.92
0.92
0.72
-1.5c
∼0.1
0.4-0.5
0.1-0.13
430(20)
455(20)
312(16)
15.2b
0.05
Me2SO
PhS(O)Me
p-TolS(O)Me
17b
21b
a At 298 K and ionic strength 1.0 M. b Reference 1. c Reference 20.
Consequently, the reaction rates will likely show a pH effect,
regardless of any involvement of H+ (or lack thereof) in the
mechanism, except in the case where the two cobalt species
fortuitously have the same reactivity. The predominant species
of the sulfenatocobalt reagents changes with the hydrogen ion
concentration. Experiments were carried out in which [H+] was
varied in the range (usually) 0.10-1.00 M. In every case the
apparent rate constant decreased with increasing [H+], from
which one infers that only (Am)5Co-S(O)R2+, and not its
conjugate acid, reacts with CH3ReO2. This postulate seems quite
reasonable because the proton of the conjugate acid would block
the point of attack of the rhenium (or, would block the basic
site that would reasonably be the atom that attacks the Lewis
acid). To explore the validity of this postulate, the data at a
given [H+] were represented by an apparent second-order rate
constant, kapp. Its value would vary with [H+] according to
CH3Re(O)2(η2-O2) + R2SOn-1 f CH3ReO3 + R2SOn (11)
kapp ) k5/(1 + [H+]/Ka)
(9)
for which k(Ph2S)11 ) 1.2 × 102 L mol-1 s-1 and k(Ph2SO) )
0.91 L mol-1 s-1. The analogous trend has been established
for the sequence 1a f 1b f 1c.12 The analogous rate constants
with these sulfur compounds are k(1a) ) 4.3 × 103 L mol-1
s-1 and k(1b) ) 3.05 × 102 L mol-1 s-1 at 298 K. With and
without cobalt, the gross trend is much the same.
These values are instructive, illustrating that the stronger S-O
bonds of the sulfones and sulfinatocobalt(III) complexes form
more slowly. The rate constants are sufficiently similar that it
where k5 represents the actual second-order rate constant for
the species (Am)5Co-S(O)R2+ reacting with CH3ReO2. The pH
profiles for several sulfenato complexes are presented in Figure
2. The smooth curves through the points are the nonlinear least-
squares fits to eq 9. The values of pKa are given in Table 1.
This treatment allows a comparison between values of the
kinetically determined parameter pKa and those found by specific
equilibrium determinations. One such value has been reported
previously: pKa ) 0.15(3) for 1b.5a Three other values of pKa
were determined directly during the course of these studies. The
basis for these determinations is the significant change in the
(10) Benson, S. W. Chem. ReV. 1978, 23, 23-35.
(11) Vassell, K. A.; Espenson, J. H. Inorg. Chem. 1994, 33, 5491.
(12) Huston, P.; Espenson, J. H.; Bakac, A. Inorg. Chem. 1993, 32, 4517.

O Atom Abstraction from (Am)5Co-S(O)nR2+
Inorganic Chemistry, Vol. 38, No. 23, 1999 5233
Cl2,5,17 (2a)(ClO4)2,17 (3a)ClO4,18 (4a)(ClO4)2,7 and (5a)(ClO4)2.18
Compound (6a)(ClO4)2 was prepared by a variant of the literature
method,19 in which 6 M HOTf (CF3SO3H) was added instead of
concentrated HClO4 and cobalt triflate used in place of the perchlorate.
Argon was passed through to evaporate the remaining THF, giving a
precipitate that was taken up in the minimum amount of water. Cooling
gave a tarry residue, which was again dissolved in the minimum amount
of water and filtered. Excess sodium perchlorate was then added;
crystals formed on allowing this preparation to stand overnight at 10
°C. The following compounds were prepared analogously: (7a)(ClO4)2,
(8a)(ClO4)2. The following analyses were carried out:
seems as if ionic charge on the cobalt participant is nearly
without effect, or at least not of a constant influence. What is
important in determining the rates of the first step of reaction
11 relative to the second is, or appears to be, the existence or
absence of an electronegative oxygen atom on the sulfur to
which the oxygen is being transferred. These reactions, best
regarded as the attack of a sulfur nucleophile on an oxygen of
the η2-peroxide group, are highly sensitive to the extra oxygen
atom, the reactivity being reduced accordingly.
The nature of the S-O bond in sulfoxides has been debated.13
The issue centers on whether a true (full) double bond exists.
In view of the large dipole moments of dialkyl- and diarylsul-
foxides, little π-bonding may exist.13 Consequently these bonds
are weaker than in, for example, phosphine and arsine oxides.
Moreover, the typical bond strength of a sulfoxide is ca. 25
kcal mol-1 weaker than that of a sulfone. An exception arises
when an electronegative atom or group is placed on the sulfur,
which strengthens the S-O bond; for example, F2SO and
(MeO)2SO have bond strengths of ca. 118 kcal, compared to
87 kcal for M.14
The stability of the sulfenatocobalt(III) complexes has been
attributed, at least in part, to π back-bonding of the (t2g)6
configuration into a vacant π orbital on sulfur,5b in competition
with the O f S π bonding. This competition for vacant orbitals
on sulfur makes it seem reasonable that the sulfenatocobalt
complexes are more reactive than organic sulfoxides. This effect
manifests itself elsewhere. Sulfenatochromium(III) complexes,
not stabilized in this fashion, are not formed upon oxidation of
the thiolato complexes.5b In addition, this orbital picture explains
the basicity of the oxygen atom of the sulfenato complexes,
reaction 8, despite the overall 2+ ionic charge.
found (calcd)
compound
C
H
N
(5a)(ClO4)2 (C8H24N5SCoCl2O8) 19.75 (20.02) 4.83 (5.04) 14.22 (14.59)
(6a)(ClO4)2 (C10H22N5SCoCl2O8) 24.34 (23.92) 4.23 (4.42) 13.69 (13.95)
(7a)(ClO4)2 (C10H21N5SCoCl3O8) 22.17 (22.38) 3.54 (3.94) 12.74 (13.05)
(8a)(ClO4)2 (C11H24N4SCoCl2O8) 25.70 (25.59) 4.46 (4.65) 12.87 (13.56)
Perchlorate complexes are convenient for isolation, but are entirely
unsuited for these studies because reaction 2 occurs readily. The thiolato
complexes were stirred with Dowex 1-X8 anion-exchange resin in the
chloride ion form for at least 10 min. Sulfenato complexes are so soluble
that, except for (1b)Cl2, crystallizing them from solution is difficult.
They were formed in solution by oxidation with hydrogen peroxide5
once the resin had been removed by filtration. In the case of solution
complexes 2b, 3b, 4b, and 5b, the ca. 2 mM solution was treated with
hydrogen peroxide; it was added in 80% of the requisite quantity to
avoid overoxidation. Complexes 6b, 7b, and 8b were converted at the
ca. 10 mM level to the sulfenato complex in solution, which was diluted
to 2 mM and converted to the chloride form by the ion-exchange
procedure described previously.
Kinetic Studies. Rate constants were determined spectrophotometri-
cally at 25.0 ( 0.1 °C. In all the kinetics experiments in aqueous
solution the ionic strength was adjusted to 1.00 M with HCl and LiCl.
Stock solutions of MTO in 1.00 M HCl were prepared daily and used
within 24 h. The MTO concentration was checked by its UV
With that in mind, we return to the principal subject. Oxygen
transfer from the SO/SO2 group, attached to cobalt(III) or not,
can also be depicted as a nucleophilic reaction in which the
oxygen attacks the d2 rhenium(V) center. The rhenium(V) center
is surely electropositive; oxygen attack on it will be more
favored for the sulfenate (lacking a second electronegative
oxygen atom) over the sulfinate.
spectrum: λmax ) 239 and 270 nm (ꢀ ) 1900 and 1300 L mol-1 cm-1
,
respectively). The UV-vis spectra of the thiolato and sulfenato
complexes are given in Table S-1.
Control experiments were performed to show that there is no
uncatalyzed reaction between the sulfenato complexes and hypophos-
phorous acid. In a few reactions, the sulfenatocobalt(III) complex was
used in excess. In such cases either MTO or H2P(O)OH was added
last, and the disappearance of the cobalt complex monitored at 360
nm, where it has an intense absorption band (Table S-1). More
commonly, the solution of CH3ReO2 was prepared first, monitoring
the solution at 290 nm, and waiting about 4 half-times for reaction 4.
The sulfenate complex was then added, and the kinetics monitored at
λmax ) 360-372 nm. Concentrations used were 50 mM H2P(O)OH,
0.22 mM total rhenium, and 0.010-0.015 mM cobalt. In a few
experiments buildup of the thiolate complex was monitored at 282 nm.
Data were analyzed according to first-order kinetics, eq 7. In the
case of 3b the initial rate method was used. The absorbance values
were converted to concentrations, and the values fit to a ninth-order
polynomial, which gives the initial rate Vi. The initial rate was obtained
by dividing Vi by [3b]0. The Vi of a polynomial function of 4-9 terms
proved largely immaterial, although the scatter of the initial rates so
calculated was ca. 8-10%.
Experimental Section
Materials. High-purity water was obtained by passing laboratory-
distilled water through a Millipore-Q water purification system. Many
substances were obtained from commercial sources, although certain
disulfides, 2-amino-4-chlorophenyl disulfide and 2-amino-4-methylphe-
nyl disulfide, were prepared from a literature procedure in which
dinitrotrisulfides were first obtained from 4-chloro-3-nitrotoluene or
2,5-dichloronitrobenzene.15 The substituted dinitrodisulfides bis(2-
amino-4-methylphenyl) trisulfide and bis(2-amino-4-chlorophenyl)
trisulfide were reduced with LiAlH4.16 The reaction in ether was
quenched with a minimum amount of water and then filtered. The
substituted disulfides were recovered once the ether was removed, and
the thiol was oxidized by air. No IR bands were observed in the S-H
region, 2550-2600 cm-1, indicating complete oxidation.
Thiolates and sulfenates were prepared according to literature
procedures or slight modifications thereof. Caution; some of the
compounds were isolated as perchlorate salts; although no problems
were encountered, the possibility of explosion hazard should be noted.
Chart 1 shows the sulfenate series; an analogous series of thiolates (a)
and sulfinates (c) can be considered. The following compounds were
synthesized by literature methods5,17 and isolated: (1a)Cl2,5,17 (1b)-
Spectrophotometric Titrations. The UV-vis spectra of selected
sulfenatocobalt complexes were recorded at different H3O+ concentra-
tions in the range 0.1-1.0 M, with the ionic strength adjusted to 1.00
M. The data, mostly in the region 360-367 nm, were fit to an equation
that describes the equilibrium in reaction 8 in terms of the acid
concentration, the total cobalt concentration, and the molar absorptivities
(13) Oae, S. Organic Sulfur Chemistry; Structure and Mechanism; CRC
Press: Boca Raton, FL, 1991.
(14) Sargeson, A. M.; Searle, G. H. Inorg. Chem. 1967, 6, 787.
(15) Gupta, R. R.; Ojha, K. G.; Kalwania, G. S.; Kumar, M. Heterocycles
1980, 14, 1145.
(16) Palmer, P. J.; Trigg, R. B.; Warrington, J. V. J. Med. Chem. 1971,
14, 248.
(18) Lane, R. H.; Sedor, F. A.; Gilroy, M. J.; Eisenhardt, P. F.; Bennett, J.
P. J.; Ewall, R. X.; Bennett, L. E. Inorg. Chem. 1977, 16, 93-101.
(19) Dickman, M. H.; Doedens, R. J.; Deutsch, E. Inorg. Chem. 1980, 19,
945-950.
(20) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry; Harper Collins: New York, 1987; p 297.
(17) Nosco, D. L.; Deutsch, E. Inorg. Synth. 1982, 21, 19.

5234 Inorganic Chemistry, Vol. 38, No. 23, 1999
Lahti and Espenson
of acid and base forms. Least-squares fitting gave the value of the acid
ionization constant Ka.
the Ames Laboratory. We are grateful to Dr. Ilia Guzei for the
crystallographic data and interpretation.
ꢀBKa + ꢀA[H+]
Ka + [H+]
Abs
l[Co]T
Supporting Information Available: Table giving UV-vis spectral
data for thiolato- and sulfenatocobalt(III) compounds, figure showing
juxtaposed UV-vis spectra for 1b, and plots of kinetic data to illustrate
agreement with selected mathematical forms and to evaluate numerical
parameters. This material is available free of charge via the Internet at
http://pubs.acs.org.
)
(12)
In one case the sulfenato was titrated with small aliquots of sodium
hydroxide. The same data analysis applies.
Acknowledgment. This research was supported by a grant
from the National Science Foundation (CHE-9625349). Some
experiments were conducted with the use of the facilities of
IC990425Q