Mechanisms for the oxidation of secondary alcohols by dioxoruthenium(VI) complexes

Article abstract of DOI:10.1139/v98-087

Possible mechanisms for the oxidation of alcohols by dioxoruthenium(VI) complexes are critically evaluated. Rate constants for the reduction of trans-[(TMC)Ru(VI)(O)₂]⁺⁺ (TMC = 1,4,8,11-tetramethyl-1,4,8,11- tetraazacyclotetradecane) by substituted benzhydrols are correlated more satisfactorily with Hammett σ substituent constants (p = -1.44 ± 0.08, r² = 0.98) than with σ⁺ substituent constants (ρ = -0.72 ± 0.11, r² = 0.83). Similar observations for the oxidation of substituted benzyl alcohols have recently been reported, confirming that the transition state for these reactions is not carbocation-like. Primary deuterium isotope effects indicate that cleavage of the α-C-H bond is rate-limiting. The lack of an observable O-D isotope effect and the ease of oxidation of ethers indicates that the presence of a hydroxyl is not essential. The previously reported observation that cyclobutanol is quantitatively converted into cyclobutanone by dioxoruthenium(VI) complexes eliminates free-radical intermediates from consideration as part of the mechanism, and negative entropies of activation (-ΔS(+) = 96-137 J mol^-1 K^-1) suggest a structured transition state. Only two of eight possible reaction mechanisms considered were found to be consistent with the available data. A critical analysis of the available data indicates that a 2 + 2 (C - H + Ru = O) addition and a reaction initiated by ligand formation through the interaction of the reductant's HOMO with the oxidant's LUMO are the most likely reaction mechanisms.

Full text of DOI:10.1139/v98-087

9
19
Mechanisms for the oxidation of secondary
alcohols by dioxoruthenium(VI) complexes
Zhao Wang, W. David Chandler, and Donald G. Lee
Abstract: Possible mechanisms for the oxidation of alcohols by dioxoruthenium(VI) complexes are critically evaluated. Rate
VI
++
constants for the reduction of trans-[(TMC)Ru (O) ] (TMC = 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane) by
2
2
substituted benzhydrols are correlated more satisfactorily with Hammett σ substituent constants (ρ = –1.44 ± 0.08, r = 0.98)
+
2
than with σ substituent constants (ρ = –0.72 ± 0.11, r = 0.83). Similar observations for the oxidation of substituted benzyl
alcohols have recently been reported, confirming that the transition state for these reactions is not carbocation-like. Primary
deuterium isotope effects indicate that cleavage of the α-C–H bond is rate-limiting. The lack of an observable O-D isotope
effect and the ease of oxidation of ethers indicates that the presence of a hydroxyl is not essential. The previously reported
observation that cyclobutanol is quantitatively converted into cyclobutanone by dioxoruthenium(VI) complexes eliminates
‡
free-radical intermediates from consideration as part of the mechanism, and negative entropies of activation (–∆S = 96–137 J
–1
–1
mol K ) suggest a structured transition state. Only two of eight possible reaction mechanisms considered were found to be
consistent with the available data. A critical analysis of the available data indicates that a 2 + 2 (C—H + RuTO) addition and
a reaction initiated by ligand formation through the interaction of the reductant’s HOMO with the oxidant’s LUMO are the
most likely reaction mechanisms.
Key words: oxidation, alcohols, ruthenium(VI), mechanism, substituent effects.
Résumé : On évalue d’une façon critique divers mécanismes possibles pour l’oxydation des alcools par des complexes de
VI
++
dioxoruthénium(VI). La corrélation entre les constantes de vitesse de réduction du trans-[(TMC)Ru (O) ] (TMC =
2
1
,4,8,11-tétraméthyl-1,4,8,11-tétraazacyclotétradécane) par des benzhydrols substitués est meilleure avec des constantes de
2
+
2
+
substituants σ de Hammett (ρ = –1,44 ± 0,08; r = 0,98) qu’avec des constantes de substituants σ (ρ = –0,72 ± 0,11; r =
,83). Des observations semblables ont été rapportées récemment pour l’oxydation d’alcools benzyliques substitués; elles
0
confirment que l’état de transition de ces réactions ne ressemble pas à un carbocation. Les effets isotopiques primaires du
deutérium indiquent que le clivage de la liaison C-H en α limite la vitesse. Le fait que l’on n’observe pas d’effet isotopique
O-D et la facilité d’oxydation des éthers indiquent que la présence d’un groupe hydroxyle n’est pas essentielle. L’observation
rapportée antérieurement à l’effet que le cyclobutanol peut être transformé quantitativement en cyclobutanone par les
complexes de dioxoruthénium(VI) élimine la possibilité d’un mécanisme impliquant des intermédiaires radicalaires; par
‡
–1 –1
ailleurs, les entropies d’activation négatives (–∆S = 96 à 137 J mol K ) suggèrent la présence d’un état de transition
structuré. On a trouvé que seulement deux des huit mécanismes réactionnels possibles sont en accord avec les données
disponibles. Une analyse critique des données disponibles indique qu’une addition 2 + 2 (C-H + RuTO) et une réaction
initiée par la formation de ligand par interaction des OMHO du réducteur avec les OMBV de l’oxydant sont les mécanismes
réactionnels les plus probables.
Mots clés : oxydation, alcools, ruthénium(VI), mécanisme, effets de substituants.
[
Traduit par la rédaction]
Introduction
these complexes, which can exist as either cis or trans isomers,
have been studied in several laboratories including those of
Che (1), Meyer (2), Drago (3), James (4), Groves (5),
Morishima (6), Hirobe (7), and Jorgensen (8).
A knowledge of the redox properties of dioxoruthenium(VI)
complexes, [LRu(O) ] , where L is a neutral macrocyclic li-
gand, is important because of their potential use as stoichiomet-
ric or catalytic oxidants. The chemical and physical properties of
++
2
It has been observed in these studies that dioxo-
ruthenium(VI) complexes can be used as oxidants for alcohols,
alkenes, arenes, sulfides, amines, and phosphines. The prod-
ucts obtained from the oxidation of primary and secondary
alcohols are aldehydes (7, 9–13) and ketones (9–14) respec-
tively; from alkanes the products are alcohols (3, 10, 12) or
ketones (10, 12); from alkenes the products are α,β-unsatu-
rated ketones (10–13, 15), epoxides (5, 7, 10–13, 16), or prod-
ucts derived from the oxidative cleavage of the double bond
(10, 11, 16); from arenes the products are α-phenyl alcohols
(10, 12, 13, 16), aldehydes (11, 16, 17), ketones (10, 12, 16),
or carboxylic acids (10), depending on the nature of the side
This paper is dedicated to Professor Erwin Buncel in
recognition of his contributions to Canadian chemistry.
Received November 3, 1997.
1
Z. Wang, W.D. Chandler, and D.G. Lee. Department of
Chemistry, University of Regina, Regina, SK S4S 0A2, Canada.
1
Author to whom correspondence may be addressed.
Telephone: (306) 585-4247. Fax: (306) 585-4894. E-mail:
dglee@meena.cc.uregina.ca
Can. J. Chem. 76: 919–928 (1998)
© 1998 NRC Canada

9
20
Can. J. Chem. Vol. 76, 1998
Scheme 1.
Single
Fig. 1. Sequential scans for the oxidation of benzhydrol (0.485 M)
by trans-[(TMC)Ru(O)2] (1.68 × 10 M) at 40.0 ± 0.1°C. Scans
were taken at intervals of 5500 s.
++
–3
-
e
lectron transfer:
. +
+
+
+
VI
V
R CHOH + [LRu (O) ]
R CHOH + [LRu (O) ]
2 2
2
2
Hydrogen atom transfers:
.
+
+
++
++
VI
V
R CHOH + [LRu (O) ]
R CHO + [LRu (OH)O]
2
2
2
.
+
+
VI
V
R CHOH + [LRu (O) ]
R COH + [LRu (OH)O]
2
2
2
Hydride transfers:
+
+
+
+
+
VI
IV
R CHOH + [LRu (O) ]
R CHO + [LRu (OH)O]
2
2
2
+
+
+
VI
IV
R CHOH + [LRu (O) ]
R COH + [LRu (OH)O]
2
2
2
2
+2 (C-H + Ru=O) addition:
HO
+
+
++
VI
VI
R CHOH + [LRu (O) ]
[L(R C)Ru (OH)O]
2
2
2
Experimental section
2
+2 (O-H + Ru=O) addition:
+
+
+
+
+
Materials
VI
VI
R CHOH + [LRu (O) ]
[L(R CHO)Ru (OH)O]
2
2
2
Benzhydrol (α-phenylbenzenemethanol), 4-methylbenzhy-
drol, 4-chlorobenzhydrol, and 4,4′-dichlorobenzhydrol were
obtained commercially (Aldrich) and purified by repeated
crystallizations from a mixture of diethyl ether and hexane
until sharp melting points in agreement with literature values
were obtained. 2-Propanol was purified by distillation from
calcium oxide. A middle fraction (bp 80.5–81.0°C) was col-
lected and analyzed by use of GLC. The results indicated a
purity of at least 99.9%.
Ligand Formation:
+
+
+
VI
VI
R CHOH + [LRu (O) ]
[L(R CHO)Ru (O) ]
2
2
2
2
H
chain; from cyclic ethers the products are lactones (10); from
sulfides the products are the corresponding sulfoxides (6, 7) or
sulfones (7); from amines, dehydrogenated products (primar-
ily nitriles and imines) are obtained (18); and from phosphines
the products are the corresponding phosphine oxides (2, 6, 19,
4-Bromobenzhydrol, 4,4′-dimethylbenzhydrol, 4-fluoro-
benzhydrol, and 4-methoxybenzhydrol were prepared by re-
duction of the corresponding benzophenones with lithium
aluminum hydride (25). The benzophenones were obtained
from Aldrich. In a typical procedure, a flame-dried 300 mL
three-necked flask, fitted with a magnetic stirrer and a drop-
ping funnel, was charged with 0.3 g (8 mmol) of lithium alu-
minum hydride (J.T. Baker) and 100 mL of anhydrous ether.
A solution of benzophenone (24 mmol) in anhydrous ether
2
0).
This paper reports the results obtained from a variety of
experiments designed to provide a better understanding of the
mechanism of the reaction between secondary alcohols and a
typical dioxoruthenium(VI) complex, trans-[(TMC)Ru (O) ]
2
TMC = 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetrade-
cane). Eight theoretically possible reaction mechanisms have
been considered: direct single-electron transfer from the alco-
hol to the oxidant, hydrogen atom or hydride ion transfer from
either oxygen or carbon, 2 + 2 addition of a ruthenium oxo
RuTO) to either the O—H or the α-C—H bond of the alcohol
and ligand formation followed by a cyclic transfer of hydro-
gen. These possibilities, derived from the discussion by March
VI
++
(50 mL) was added dropwise. The mixture was stirred under
(
gentle reflux until the reaction was complete. The solution was
then treated with 50 mL of ice water while cooling the reaction
flask in an ice bath. The organic layer was separated and the
aqueous layer extracted with ether (2 × 30 mL). The combined
organic layers were dried over CaCl , filtered, and evaporated
2
(
to give the corresponding benzhydrol in 85–93% yield. The
products were repeatedly recrystallized from a solution of di-
ethyl ether and hexane until sharp melting points were ob-
(
21a) and the theoretical work of Deng and Ziegler (22),
tained and then identified from physical and spectral data.
1
Mayer and co-workers (23), and Rappe and Goddard (24) on
the reactions of high valent transition metal compounds, have
been summarized in Scheme 1.
4
-Bromobenzhydrol, mp 62–63°C, lit. (26) 63–65°C; H NMR
13
(CDCl ), δ: 2.12 (d, 1H), 5.74 (d, 1H), 7.36 (m, 9H); C NMR
3
(
CDCl ), δ: 75.6, 121.4, 126.6, 127.9, 128.3, 128.7, 131.6,
3
Each of the mechanistic possibilities outlined in Scheme 1
has been examined in light of the rather large amount of ex-
perimental evidence currently available with respect to the re-
actions of dioxoruthenium(VI) complexes and alcohols. From
such an analysis it is possible to eliminate several potential
processes and to arrive at the “most likely” mechanisms.
142.8, 143.4. 4,4′-Dimethylbenzhydrol, mp 68.5–69.0°C, lit.
1
(26) 69°C; H NMR (CDCl ), δ: 2.22 (d, 1H), 2.36 (s, 6H),
3
1
3
5.72 (s, 1H), 7.38 (M, 8H); C NMR (CDCl ), δ: 21.2, 76.0,
3
126.5, 129.2, 137.2, 141.2. 4-Fluorobenzhydrol, mp 47–48°C;
1
H NMR (CDCl ), δ: 2.41 (d, 1H), 5.78 (d, 1H), 6.99 (t, 2H),
3
1
3
7.28 (m, 7H); C NMR (CDCl ), δ: 75.6, 115.5, 126.5, 127.8,
3
©
1998 NRC Canada

Wang et al.
921
Fig. 2. Typical pseudo first-order plot for the oxidation of
Table 1. Rate constants for the oxidation of 2-propanol and
benzhydrol by [trans-(TMC)RuO2] in acetonitrile at 40°C.
++
++
benzhydrol (0.670 M) by trans-[(TMC)Ru(O) ] . Slope = 9.67 ×
2
–5
–1
2
1
0
s , r = 0.999.
–
1
5 a
–1 –1
s
5 b
Reductant Concentration (M) k1 (s × 10 ) k2 (M
× 10 )
2
2
2
2
2
-Propanol
-Propanol
-Propanol
-Propanol
-Propanol
2.62
6.6 ± 0.6
7.0 ± 0.3
7.3 ± 0.1
8.1 ± 0.1
9.1 ± 0.1
6.8 ± 0.3
7.2 ± 0.5
8.2 ± 0.1
9.0 ± 0.6
9.6 ± 0.2
2.5 ± 0.2
2.5 ± 0.1
2.4 ± 0.1
2.5 ± 0.1
2.6 ± 0.1
14 ± 1
15 ± 1
15 ± 1
15 ± 1
14 ± 1
2.84
3.01
3.26
3.56
Benzhydrol
Benzhydrol
Benzhydrol
Benzhydrol
Benzhydrol
a
0.485
0.492
0.550
0.661
0.670
Pseudo first-order rate constants.
b
k = k /[reductant].
2
1
filtered hot, and the yellow filtrate evaporated to dryness. This
product was redissolved in a minimum of hot 3 M HCl. Yellow
III
crystals of trans-[(TMC)Ru (Cl) ]Cl formed when the solu-
2
tion was allowed to slowly cool. After filtration and washing
with cold 3 M HCl, a yellow crystalline product (0.70 g,
1
6
28.2, 128.3, 128.6, 139.5, 143.0. 4-Methoxybenzhydrol, mp
1
1.64 mmol, 58%) was obtained; λmax = 370 nm, with a shoul-
5–66°C; H NMR (CDCl ), δ: 2.65 (d, 1H), 3.81 (s, 3H), 5.74
3
der at 315 nm when dissolved in acetonitrile. A portion of this
product (0.43 g, 0.95 mmol) and silver p-toluenesulfonate
1
3
(
d, 1H), 6.88 (d, 2H), 7.33 (M, 7H); C NMR (CDCl ), δ: 55.3,
3
5.8, 113.9, 126.5, 128.0, 128.5, 136.0, 144.5, 159.5.
7
(0.77 g, 2.77 mmol) were heated at 60°C for 0.5 h in 35 mL
Benzhydrol-α-d and 2-propanol-α-d were prepared by the
reduction of benzophenone and acetone with lithium alu-
of distilled water. A grey–white precipitate (AgCl), which
formed immediately, was removed by filtration, and 50%
H O (9 mL) was added dropwise to the remaining yellow
minium deuteride (Aldrich). Benzhydrol-α-d, mp
2
2
6
7.5–68.0°C, was identified by use of NMR: 1H NMR
solution at 60°C. After 30 min the solution was cooled in an
ice–water bath and an excess of NaClO added. The compound
VI
1
3
3 3
(
CDCl ), δ: 2.39 (s, 1H), 7.38 (m, 10H); C NMR (CDCl ), δ:
4
trans-[(TMC)Ru (O) ](ClO ) , which slowly precipitated,
126.6, 127.6, 128.5, 144.5. 2-Propanol-α-d was purified by
2
4 2
was collected and recrystallized from hot (60°C) 0.1 M
distillation from calcium oxide, bp 80.0–80.5°C, and identified
1
by NMR: H NMR (CDCl ), δ: 1.11 (s, 6H), 2.66 (br, 1H).
3
HClO . Crystalline product (0.23 g, 43%) was obtained. The
4
Only one peak was detected in a GLC analysis indicating at
least 99.9% purity. From the NMR integrals, the incorporation
of deuterium was calculated to be 98%. 2-Propanol-O-d was
spectrum of this product dissolved in acetonitrile was similar
to that reported in the literature for other dioxoruthenium(VI)
complexes: λmax = 228, 257, and 388 nm; lit. (13, 27) 223, 255,
and 388 nm.
prepared by repeatedly treating 2-propanol with D O. The
1
2
product was identified by NMR; the H NMR integrals indi-
VI
The compound trans-[(TMC)Ru (O) ](BF ) was pre-
2
4 2
pared by first treating trans-[(TMC)Ru (Cl) ]Cl (0.43 g,
cated 96.7% incorporation of deuterium.
The oxidant was prepared in two forms, [(TMC)-
III
2
.95 mmol) with silver p-toluenesulfonate (0.77 g,
0
VI
VI
Ru (O) ](ClO ) and [(TMC)Ru (O) ](BF ) using proce-
4 2
2
4 2
2
2.77 mmol) dissolved in 35 mL of distilled water and heating
the mixture at 60°C for 0.5 h. The solution was filtered to
remove AgCl, and 50% H O (9 mL) was added dropwise to
dures previously described in the literature (27). Both salts
reacted with alcohols at the same rate under any particular set
of conditions, indicating that the identity of the anion has little
influence on the reaction rate. The initial work was done with
the perchlorate salt; however, on two occasions small explo-
sions occurred when it was being transferred as a dry solid
from the reaction flask in which it had been prepared to a
storage vial. The explosions were sufficiently violent to shatter
glassware and cause minor cuts to the hand of one of us (Z.W.).
The fluoroborate salt proved to be much safer to work with.
2
2
the yellow filtrate at 60°C. This solution was concentrated to
two thirds of its volume and cooled in an ice-water bath. When
an excess of saturated NaBF solution was added, a precipitate
4
formed slowly over several hours. Collection of the precipitate
and recrystallization from hot (60°C) 0.1 M HBF yielded
4
VI
trans-[(TMC)Ru (O) ](BF ) (0.12 g, 34%).
2
4 2
Plots of the concentration of this product in acetonitrile
solutions vs. absorbance at 388 nm were precisely linear, as
required by Beer’s Law, with a slope of ε = 539 ± 5.
III
Both salts were prepared starting from K [Ru (Cl) H O],
2
5
2
obtained from either Johnson Matthey or Strem Chemicals.
The solvent, acetonitrile, used in these experiments was
purified by stirring with KMnO overnight and simple distil-
III
Initially, 1.056 g (2.82 mmol) of K [Ru (Cl) (H O)] was sus-
2
5
2
4
pended in 150 mL of absolute ethanol and heated under reflux
with vigorous stirring for 15 min. 1,4,8,11-Tetramethyl-
lation. The distillate was then carefully redistilled over CaH2
in a nitrogen atmosphere.
1
3
,4,8,11-tetraazacyclotetradecane, TMC (Aldrich, 0.92 g,
.59 mmol), dissolved in 280 mL of ethanol was added drop-
wise to the refluxing suspension over a period of 6 h. After
refluxing for an additional 12 h (overnight), the solution was
Kinetic method
Reaction rates were determined by monitoring spectral
©
1998 NRC Canada

9
22
Can. J. Chem. Vol. 76, 1998
Table 2. Kinetic deuterium isotope effects for the oxidation of alcohols by trans-dioxoruthenium(VI) complexes.
a
Reductant
Oxidant
Solvent
Temp. (°C)
k /k
H
Reference no.
D
CH CDOHCH
3
trans-[LRu(O)2]++
trans-[LRu(O)2]++
CH CN
3
40
40
25
25
25
25
40
25
25
7.8 ± 0.5
21 ± 2
19 ± 2
8 ± 1
8 ± 1
This work
3
Ph CDOH
2
CH CN
3
This work
+
+
+
+
+
PhCD OH
2
trans-[L′Ru(O)2]
trans-[L′Ru(O)2]
trans-[L′Ru(O)2]
trans-[L′Ru(O)2]
trans-[LRu(O)2]
trans-[L′Ru(O)2]
trans-[L′Ru(O)2]
CH CN
3
9
+
+
+
CD OD
3
CH CN
3
9
CD CD OD
3
CH CN
3
9
2
CD CDODCD
3
CH CN
3
11 ± 1
9
3
++
CH CHODCH
3
CH CN
3
1.0 ± 0.1
1.0 ± 0.1
1.0 ± 0.1
This work
3
+
+
+
CH CH OD
3
CH CN
3
9
9
2
+
CH CH OD
3
D O
2
2
a
L is 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane (TMC). L′ is N,N′-dimethyl-6,7,8,9,10,11,17,18-octahydro-5H-
dibenzo[en][1,4,8,12]dioxadiazacyclopentadecane.
+
Fig. 3. Hammett correlation, using σ substituent constants, for the
Fig. 4. Attempted Hammett correlation, using σ substituent
oxidation of substituted benzhydrols by trans-[(TMC)Ru(O)2]++ at
constants, for the oxidation of substituted benzhydrols by
2
trans-[(TMC)Ru(O)2]++ at 40.0 ± 0.1°C. Slope = –0.72 ± 0.11, r =
0.834. Substituent constants were taken from March (21b). The
effects when substituents are present in both rings were taken to be
additive.
2
4
0.0 ± 0.1°C. Slope = –1.44 ± 0.08, r = 0.978. Substituent
constants were taken from March (21b). The effects when
substituents are present in both rings were taken to be additive.
a
Table 3. Hammett correlations for the oxidation of alcohols by dioxoruthenium(VI) complexes in acetonitrile.
b
2 c
+
2 d
Oxidant
Reductant
Temp. (°C)
ρ(r )
ρ (r )
Reference
This work
trans-[LRu(O)2]++
Ar CHOH
2
40
25
25
–1.44 (0.98)
–1.51 (0.96)
–1.24 (0.99)
–0.72 (0.83)
–0.91 (0.87)
–0.75 (0.90)
+
+
e
trans-[L′Ru(O) ]
cis-[L′′Ru(O)2]
ArCH OH
2
9
2
+
+
e
ArCH OH
2
16
a
Substituent constants were taken from ref. 21b.
b
L is 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane (TMC). L′ is N,N′-dimethyl-6,7,8,9,10,11,17,18-octahydro-5H-
dibenzo[en][1,4,8,12]dioxadiazacyclopentadecane. L′′ is N,N,N′,N′-tetramethyl-3,6-dimethyl-3,6-diazaoctane-1,8-diamine.
c
d
e
2
Hammett ρ value obtained using σ substituent constants (r is the correlation coefficient).
+
2
Hammett ρ value obtained using σ substituent constants (r is the correlation coefficient).
Correlations were recalculated using rate-constant data reported in these references.
changes as the reaction progressed. All measurements were
made under pseudo first-order conditions with alcohol concen-
trations at least 100 times that of the oxidant. Each rate
constant reported is the average of at least three determinations
that agreed within ± 10% or less. In a typical kinetic experi-
ment, a quartz cuvette containing a 2.00 mL aliquot of
©
1998 NRC Canada

Wang et al.
923
Fig. 5. Eyring plot for the oxidation of benzhydrol (0.504 M) by
Table 4. Activation parameters for the oxidation of alcohols by
dioxoruthenium(VI) complexes in acetonitrile.
++
[
(TMC)Ru(O) ] . Slope = –6920 ± 370 K, intercept = 7.5 ± 1.2,
2
2
r = 0.989.
‡
‡
∆
H
∆S
(kJ/mol) (J/mol K)
Reference
no.
b
Reductant
Oxidant
Benzhydrol
trans-[LRu(O) ]++ 58 ± 9 –137 ± 8 This work
2
2
2
2
-Propanol
-Propanol
-Propanol
trans-[LRu(O) ]++ 62 ± 5 –135 ± 2 This work
2
cis-[L′Ru(O) ]++
54 ± 4 –109 ± 17
50 ± 4 –117 ± 13
59 ± 4 –92 ± 13
50 ± 4 –100 ± 13
16
2
+
+
trans-[L′′Ru(O) ]
9
2
Cyclobutanol cis-[L′Ru(O)2]++
16
16
10
9
++
Benzyl alcohol cis-[L′Ru(O) ]
Benzyl alcohol trans-[L′′′Ru(O) ] 35 ± 4 –96 ± 13
Benzyl alcohol trans-[L′′Ru(O)2]++ 42 ± 4 –109 ± 13
2
a
++
2
a
Solvent: CF SO H.
3
3
b
L is 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane (TMC). L′ is
N,N,N′,N′-tetramethyl-3,6-dimethyl-3,6-diazaoctane-1,8-diamine. L′′ is
N,N′-dimethyl-6,7,8,9,10,11,17,18-octahydro-5H-dibenzao[en][1,4,8,112]
dioxadiazacyclopentadecane. L′′′ is 5,5-dimethyl-2,2′-bipyridine.
that the transition state is not carbocation-like. Table 3 con-
tains a comparison of the Hammett correlations for similar
reactions of dioxoruthenium(VI) complexes that have been re-
ported from other laboratories.
–
3
VI
++
approximately 1 × 10 M [(TMC)Ru (O) ] in acetonitrile
2
was placed in the thermostated cell compartment of a Hewlett
Packard 8450A diode array spectrophotometer and held there
until thermal equilibrium had been established. An excess of
the alcohol, also dissolved in acetonitrile and thermostated at
the same temperature, was added from a microliter syringe
and, after thorough mixing, spectra recorded at equal time in-
tervals. Sequential scans exhibited sharp isosbestic points at
Activation parameters
The enthalpies and entropies of activation for the oxidation of
2-propanol and benzhydrol were determined from a study of
the reaction rates at different temperatures. A typical Eyring
plot has been reproduced in Fig. 5. The activation parameters
that have been reported for other similar reactions are included
in Table 4 for comparison purposes.
334 nm and 404 nm (Fig. 1). The final spectrum was identical
to that reported for [(TMC)Ru O(CH CN)] (13).
IV
++
3
Results
Discussion
There are at least two reasons why chemists wish to have a
good understanding of reaction mechanisms: (i) a knowledge
of the mechanism of a reaction provides an indication of the
conditions that should be used to maximize the yield and rate
of formation of a specific reaction product, and (ii) insights
into the details of a particular reaction mechanism are often
transferable to other reactions; i.e., there is a likelihood that all
chemistry will ultimately be understandable in terms of a finite
number of fundamental principles.
The reactions between high valent transition metal com-
pounds (oxidants) and organic reductants are examples of im-
portant, often used processes for which the chemistry is not
particularly well understood, despite a voluminous literature.
The products obtained, in common with most redox reactions,
could be produced via several possible mechanisms (21a), as
summarized in Scheme 1 for the reaction under consideration.
With such a multiplicity of viable mechanistic options avail-
able, it will likely be very difficult to “prove” that a particular
mechanism is correct; the best that can be expected at this time
is a critical evaluation of each possible mechanism in light of
the available experimental evidence and the most current theo-
retical calculations. Using this approach several possibilities
can be eliminated and the “most likely” mechanisms proposed.
Ultimately, it is reasonable to expect that it may be possible to
arrive at a unified mechanistic consensus that is applicable to
a large number of redox reactions (28).
Rate law
Plots of ln(absorbance – final absorbance) at 388 nm were
linear (Fig. 2), indicating that the reaction is first order in oxi-
dant. Second-order rate constants, calculated by division of the
pseudo first-order rate constants by alcohol concentration (Ta-
ble 1), were observed to be constant, indicating that the reac-
tion is also first order with respect to the reductant. The rate
law is therefore given by eq. [1].
VI
VI
[
1]
–d[Ru ]/dt = k [Ru ][alcohol]
2
Isotope effects
Kinetic isotope effects, determined for the oxidation of 2-
propanol-2-d, 2-propanol-O-d and benzhydrol-α-d are sum-
marized in Table 2. The results reported from other
laboratories for the oxidation of isotopically substituted alco-
hols by dioxoruthenium(VI) complexes have also been in-
cluded in Table 2 for comparison purposes.
Substituent effects
The second-order rate constants for the oxidation of substi-
tuted benzhydrols by [(TMC)Ru (O) ] were found to corre-
2
late well with Hammett σ values (Fig. 3), but not with the
corresponding σ values (Fig. 4). The negative ρ value (–1.44)
VI
++
+
indicates that the electron density on the α-carbon has de-
creased in the transition state; however, the lack of correlation
+
with σ values, observed in this and other studies, indicates
The observation made by Che and co-workers (12, 29), that
©
1998 NRC Canada

9
24
Can. J. Chem. Vol. 76, 1998
the only product obtained from the oxidation of cyclobutanol
by dioxoruthenium(VI) complexes is cyclobutanone, indicates
that free-radical intermediates are not involved. As discussed
in detail elsewhere (16, 30–36), it is known that the internal
strain of the cyclobutyl system causes it to open if free radical
intermediates are formed. For example, acyclic products are
obtained when cyclobutanol is oxidized by typical one-elec-
tron oxidants such as vanadium(V) (30, 31), manganese(III)
is replaced by deuterium (Table 2). This would include both
hydrogen atom and hydride ion transfers from oxygen, as well
as 2 + 2 (O-H + RuTO) additions. The report that ethers, such
as tetrahydrofuran, are readily oxidized by diox-
oruthenium(VI) complexes (10, 29, 37) indicates that the pres-
ence of O-H groups in the structure of the reductant is not
essential, in contrast to chromic acid oxidations (for example
see ref. 38).
(
(
31), chromium(IV) (32), cerium(IV) (33), and ruthenium(VII)
34, 36), while cyclobutanone is obtained from typical two-elec-
tron oxidants such as ruthenium tetroxide (35), ruthenium(VI)
35, 36), chromium(VI) (32), and chromium(V) (32). The ex-
A mechanism involving direct transfer of a hydride ion
from the α-carbon would be in agreement with the observed
isotope effects, but not with the Hammett studies reported here
and elsewhere. The rate constants for such reactions (eq. [2])
(
+
pected products from oxidations involving single-electron or
hydrogen atom transfers would therefore be acyclic com-
pounds (Scheme 2), while two-electron transfer reactions
would result in the formation of cyclobutanone (Scheme 3).
Reaction mechanisms that involve cleavage of the O—H
bond in the rate-limiting step are also eliminated from further
consideration by the observation that there is no measurable
kinetic deuterium isotope effect when the hydroxyl hydrogen
should correlate better with σ values (39), contrary to the
experimental observations summarized in Table 3. In fact, any
2
reaction in which the α-carbon is carbonyl-like (sp hybrid-
+
ized) in the transition state should correlate better with σ val-
ues. The correlation reported herein and those found in the
literature are consistently better when σ values are used, the
clear implication being that direct hydride transfer is not a
likely mechanism.
[
2]
A mechanism in which the reaction is initiated by a 2 + 2
addition of a ruthenium oxo to the α-C—H bond, as has been
proposed for several other oxidation reactions (40–42), is in
agreement with the isotope effects reported in Table 2 and the
‡
effect and activation parameters (k /k = 20 ± 2, ∆H = 59 ± 4
H
D
–1
‡
–1 –1
kJ mol , ∆S = –117 ± 13 J mol K ) similar to those for the
corresponding alcohol reactions (Tables 2 and 4) is also con-
sistent with a possible 2 + 2 (C—H + RuTO) addition mecha-
nism. The α-hydrogens of ethers and alcohols are
electronically similar and would therefore be expected to re-
spond similarly to oxidants that cause C—H bond cleavage in
the rate-limiting step.
Hammett studies summarized in Table 3. In this reaction
3
(
possibility of a direct resonance contribution from the substi-
eq. [3]) the α-carbon remains sp hybridized, eliminating the
+
tuents and therefore a lack of correlation with σ values.
It is also possible to consider a mechanism for conversion
of the organometallic intermediate, 1, into products. Two pos-
sibilities exist: (i) homolytic cleavage of the Ru—C bond
would give a ruthenium(V) complex and a free radical that
could subsequently undergo a hydrogen atom transfer as in
eqs. [4] and [5], or (ii) heterolytic cleavage of the Ru—C bond
to give a ruthenium(IV) complex and a cation that could trans-
fer a proton to give the observed products as in eqs. [6] and
[7].
[
3]
The observation (37) that the oxidation of tetrahydrofuran by
dioxoruthenium(VI) complexes exhibits a primary isotope
[
4]
[
[
5]
6]
[
7]
The reaction sequence in eqs. [4] and [5] involves forma-
tion of free-radical intermediates, which are considered to be
unlikely because of the previously described observation that
oxidation of cyclobutanol does not result in ring opening. It is,
©
1998 NRC Canada

Wang et al.
925
Scheme 2.
Single- electron transfer :
.
. H
H+
H
O
.O
H
++
+
VI
V
+
[(TMC)Ru (O) ]
+
[(TMC)Ru (O) ]
2
2
.
H+
H
O
OH+
CH CH CH CH
O
.
.
2
H+ + CH CH CH CH
Acyclic products
2
2
2
2
2
Hydrogen atom transfer from O-H:
O
H
H
.O
++
H
++
VI
V
+
[(TMC)Ru (O) ]
+
[(TMC)Ru (OH)O]
2
.
O
O
H
.
CH CH CH CH
Acyclic products
2
2
2
Hydrogen atom transfer from C-H:
O
H
H
O
.
H
O
++
++
VI
V
+
[(TMC)Ru (O) ]
+
[(TMC)Ru (OH)O]
2
.
O
O
.
H
.
2
H
CH CH CH CH
Acyclic products
2
2
however, possible that the transfer of the hydrogen atom
eq. [5]) could occur more rapidly than ring opening, thereby
one translational degree of freedom. The low enthalpy of ac-
tivation is also consistent with a 2 + 2 addition of a ruthenium
oxo bond to a C—H bond because the energy required for bond
cleavage would be partly compensated for by the concurrent
formation of new O—H and Ru—C bonds.
There is, however, a theoretical difficulty associated with
the 2 + 2 (C—H + RuTO) mechanism. Frontier molecular
orbital theory suggests that oxidation reactions of this type
should involve an initial interaction between the reductant’s
HOMO and the oxidant’s LUMO. The oxidant’s LUMO in this
case would be an RuTO π* antibonding orbital (28), while
the reductant’s HOMO would be an oxygen 2p-nonbonding
orbital. Since the LUMO would be centered primarily on the
metal, it is predicted on the basis of this theory that the initial
reaction would involve ligand formation as in eq. [8].
(
precluding the formation of acyclic products. In such an event,
the hydrogen atom transfer would have to take place rapidly
before the free radical could diffuse out of the solvent cage
enclosing it and the ruthenium(V) complex; diffusion out of
the solvent cage would provide more than sufficient time for
ring opening to occur.
A mechanism consisting of reaction [3], followed by either
4] and [5] or [6] and [7], is consistent with the observed rate
[
law, the products from the oxidation of cyclobutanol, the iso-
tope effects, and the Hammett study. It is also consistent with
the observed activation parameters (Table 4). A 2 + 2 addition
(
eq. [3]) would be expected to exhibit a negative entropy of
activation in common with most combination reactions that
require an organized transition state and result in the loss of
[
8]
©
1998 NRC Canada

9
26
Can. J. Chem. Vol. 76, 1998
Scheme 3.
Hydride transfer from C-H:
O
H
H
O
H
O
+
++
+
+
VI
IV
+
[(TMC)Ru (O) ]
+
[(TMC)Ru (OH)O]
2
OH
OH
+
+
+
H
Hydride transfer from O-H:
+
O
H
H
O
++
H
+
VI
IV
+
[(TMC)Ru (O) ]
+
[(TMC)Ru (OH)O]
2
+
O
O
H
+
+
H
C-H + Ru=O addition followed by a transfer of 2 electrons :
OH
OH
+
+
VI
H
++
Ru (TMC)(OH)O]
+
[(TMC)Ru(O)2]
OH
OH
+
+
VI
+
Ru (TMC)(OH)O]
+
IV
+
[(TMC)Ru (OH)O]
OH
+
OH
O
+
+
+
H
O-H + Ru=O addition followed by a transfer of 2 electrons :
+
+
VI
O
H
H
ORu (TMC)(OH)O]
++
H
VI
+
[(TMC)Ru (O) ]
2
+
+
+
O
Ru(TMC)(OH)O]
H
O
H
+
IV
+
[(TMC)Ru (OH)O]
+
O
O
H
+
+
H
Ligand formation would be followed by proton transfer as in
eq. [8] to give 4, which is formally the 2 + 2 (O—H + RuTO)
addition product. The amounts of 3 and 4 present would be
complexes also indicates that a proton transfer would not be an
essential part of the reaction if this is the correct mechanism.
The intermediates 3 and 5 for alcohol and ether oxidation,
respectively, would decompose to yield oxidized products by
a cyclic transfer of hydrogen as depicted in eqs. [9] or [10] and
[11]. This mechanism is similar to one recently proposed by
Espenson and co-workers (43) for the oxidation of alcohols by
controlled by the relative pKBH values for the two compounds.
+
However, the lack of an observable O-D isotope effect (Ta-
ble 2) suggests that 4 may not be part of the reaction sequence
leading to product formation. The aforementioned fact that
tetrahydrofuran is readily oxidized by dioxoruthenium(VI)
++
[Cr T O] . It is also in general agreement with the theoretical
©
1998 NRC Canada

Wang et al.
927
description of the reaction between methanol and oxidants
containing chromium or ruthenium oxos recently published by
Deng and Ziegler (22).
[
[
9]
10]
[
11]
The electron flow in these reactions has been indicated as if it
involved hydride transfer; however, for cyclic redox mecha-
nisms it is not possible to specify the direction of electron flow,
nor whether it occurs by single- or double-electron shifts. The
assumption that the oxygen bears a positive charge in the tran-
sition state is consistent with the negative Hammett ρ value
and with the observation that electronic substituent effects cor-
relate with σ rather than σ values. Since the positive charge is
not located adjacent to the aromatic ring in intermediate 3,
direct resonance interactions are not possible.
Conclusions
The experimental and theoretical information available with
respect to the oxidation of alcohols by dioxoruthenium(VI)
complexes is not compatible with several possible redox
mechanisms including a single-electron transfer and direct hy-
drogen atom or hydride ion transfers from either carbon or
oxygen. The available information is consistent with both a
+
2
+ 2 (C—H + RuTO) mechanism (eqs. [3]–[7]) or attach-
ment of the alcohol as a ligand on Ru, followed by a cyclic
transfer of hydrogen from the reductant to the oxidant as in
eqs. [8] and [9]. Currently available information does not per-
mit a choice between these two possibilities; however, the lat-
ter appears to be more consistent with theoretical
considerations.
In common with the 2 + 2 (C—H + RuTO) reaction mecha-
nism discussed previously, the mechanism defined by eqs. [8]
and [9] would be consistent with the unfavorable entropies of
activation and the low enthalpies of activation summarized in
Table 4. The cyclic, well-organized transition state would be
expected to result in an unfavorable entropy of activation and
the simultaneous bond making and breaking would be consis-
tent with a low enthalpy of activation. The significant C-D
isotope effect and the lack of an observable O-D isotope effect
are also consistent with this mechanism.
Acknowledgment
Financial assistance from the Natural Sciences and Engineer-
ing Research Council of Canada is gratefully acknowledged.
Both the 2 + 2 (C—H + RuTO) addition mechanism
eq. [3]) and the ligand association mechanism (eq. [8]) re-
References
(
quire the introduction of a new ruthenium ligand into the oc-
1. C.M. Che and W.V.V. Yam. Adv. Inorg. Chem. 39, 233 (1992),
and refs. therein.
VI
++
tahedral complex, [(TMC)Ru (O) ] . In eq. [3] the new
2
ligand is attached through carbon to give an organometallic
intermediate, while in eq. [8] the new ligand is attached
through oxygen giving a ruthenate ester intermediate. Whether
these new ligands would be introduced by associative or dis-
sociative mechanisms (44) cannot be determined conclusively
on the basis of the experimental evidence currently available.
However, the observation made by Che et al. (9) that com-
plexes with more flexible acyclic tetraaza ligands react faster
by several orders of magnitude than the corresponding cyclic
tetraaza ligands suggests that a dissociative mechanism with
loss of an N-donor ligand may be most likely (45).
2. A. Dovletoglou and T.J. Meyer. J. Am. Chem. Soc. 116, 215
(
1994), and refs. therein.
. A.S. Goldstein, R.H. Beer, and R.S. Drago. J. Am. Chem. Soc.
16, 2424 (1994), and refs. therein.
. T. Mlodnicka and B.R. James. In Metallorporphyrins catalyzed
oxidations. Edited by F. Montanari and L. Casella. Kluwer Aca-
demic Publishers, Amsterdam. 1994. pp. 121–148, and
refs. therein.
3
4
1
5
. J.T. Groves and R. Quinn. J. Am. Chem. Soc. 107, 5790 (1985),
and refs. therein.
6. Y. Tokita, K. Yamguchi, Y. Watanabe, and I. Morishima. Inorg.
Chem. 32, 329 (1993).
©
1998 NRC Canada

9
28
Can. J. Chem. Vol. 76, 1998
7
8
. T. Higuchi, H. Ohtake, and M. Hirobe. Tetrahedron Lett. 32,
435 (1991); 30, 6545 (1989).
. K.A. Jorgensen. In Metalloporphyrins catalyzed oxidations. Ed-
ited by F. Montanari and L. Casella. Kluwer Academic Publish-
ers, Amsterdam. 1994. pp. 207–238.
25. L.A. Paquette. Encyclopedia of reagents for organic synthesis.
Vol. 5. Edited by L.A. Paquette. Wiley, New York. 1995.
p. 3009; W.F. Gannon and H.O. House. Organic syntheses. Coll.
vol. 5. Wiley, New York. 1973. p. 294.
7
26. C.W.N. Cumper and A.P. Thurston. J. Chem. Soc. Perkin Trans.
2, 106 (1972).
9
. C.M. Che, W.T. Tang, W.O. Lee, K.Y. Wong, and T.C. Lau. J.
Chem. Soc. Dalton Trans. 1551 (1992).
2
7. C.M. Che, K.Y. Wong, and C.K. Poon. Inorg. Chem. 24, 1797
1985); C.M. Che, T.W. Tang, and C.K. Poon. J. Chem. Soc.
Chem. Commun. 641 (1984).
1
1
1
1
1
1
1
1
1
1
0. C.M. Che, W.H. Leung, C.K. Li, and C.K. Poon. J. Chem. Soc.
Dalton Trans. 379 (1991).
(
1. C.M. Che, W.T. Tang, W.O. Lee, W.T. Wong, and T.F. Lai. J.
Chem. Soc. Dalton Trans. 2011 (1989).
2. C.M. Che, W.T. Tang, W.T. Wong, and T.F. Lai. J. Am. Chem.
Soc. 111, 9048 (1989).
2
2
8. R.H. Holm. Chem. Rev. 87, 1401 (1987).
9. C.M. Che, W.T. Tang, and C.K. Li. J. Chem. Soc. Dalton Trans.
3735 (1990).
30. J. Rocek and D.E. Aylward. J. Am. Chem. Soc. 97, 5452 (1975).
31. J. Rocek and A.E. Radkowsky. J. Org. Chem. 38, 89 (1973).
32. J. Rocek and A.E. Radkowsky. J. Am. Chem. Soc. 95, 7123
3. C.M. Che, T.F. Lai, and K.Y. Wong. Inorg. Chem. 26, 2289
(
1987).
4. S.Y.S. Cheng, N. Rajapakse, S.J. Rettig, and B.R. James. J.
Chem. Soc. Chem. Commun. 2669 (1994).
5. C.K. Li, C.M. Che, W.F. Tong, W.T. Tang, K.Y. Wong, and T.F.
Lai. J. Chem. Soc. Dalton Trans. 2109 (1992).
(
1973).
3
3. K. Meyer and J. Rocek. J. Am. Chem. Soc. 94, 1209 (1972).
34. D.G. Lee and L.N. Congson. Can. J. Chem. 68, 1774 (1990).
35. D.G. Lee, U.A. Spitzer, J. Cleland, and M.E. Olson. Can. J.
Chem. 54, 2124 (1976).
36. D.G. Lee, L.N. Congson, U.A. Spitzer, and M.E. Olson. Can. J.
Chem. 62, 1835 (1984).
6. W.C. Cheng, W.Y. Yu, C.K. Li, and C.M. Che. J. Org. Chem. 60,
6
840 (1995).
7. N. Rajapakse, B.R. James, and D. Dolphin. Catal. Lett. 2, 219
1989).
8. A.J. Bailey and B.R. James. J. Chem. Soc. Chem. Commun.
343 (1996).
9. B.A. Moyer, B.K. Sipe, and T.J. Meyer. Inorg. Chem. 20, 1475
1981).
(
3
7. C.M. Che, W.T. Tang, K.Y. Wong, and C.K. Li. J. Chem. Soc.
Dalton Trans. 3277 (1991).
2
3
3
8. F.H. Westheimer. Chem. Rev. 45, 419 (1948).
9. T.H. Lowry and K.S. Richardson. Mechanism and theory in or-
ganic chemistry. Harper and Row, New York. 1987.
pp. 143–151.
(
2
2
0. S.Y.S. Cheng and B.R. James. J. Mol. Catal. 117, 91 (1997).
1. (a) J. March. Advanced organic chemistry. 4th ed. Wiley, New
York. 1992. pp. 1158–1161, and refs. therein; (b) Advanced or-
ganic chemistry. 4th ed. Wiley, New York. 1992. p. 280.
2. L. Deng and T. Ziegler. Organometallics, 16, 716 (1997), and
refs. therein.
40. D.G. Lee and T. Chen. J. Am. Chem. Soc. 115, 11231 (1993).
41. D.G. Lee and T. Chen. J. Org. Chem. 56, 5341 (1991).
42. D.G. Lee and H. Gai. Can. J. Chem. 71, 1394 (1993).
43. S.L. Scott, A. Bakac, and J.H. Espenson. J. Am. Chem. Soc. 114,
2
2
3. K.A. Gardner, L.L. Kuehnert, and J.M. Mayer. Inorg. Chem. 36,
4205 (1992).
2
1
069 (1997); G.K. Cook and J.M. Mayer. J. Am. Chem. Soc.
17, 7139 (1995); J. Am. Chem. Soc. 116, 1855 (1994); J. Am.
44. J.E. Huheey, E.A. Keiter, and R.L. Keiter. Inorganic chemistry
principles of structure and reactivity. Harper Collins, New York.
1993. p. 540.
Chem. Soc. 116, 1868 (1994); K.A. Gardner and J.M. Mayer.
Science (Washington, D.C.) 269, 1849 (1995).
2
4. A.K. Rappe and W.A. Goddard. J. Am. Chem. Soc. 104, 3287
45. R.G. Wilkins. Kinetics and mechanisms of reactions of transition
metal complexes. 2nd ed. VCH, New York. 1991. p. 224.
(
1982).
©
1998 NRC Canada