Oxidation of 2,3-butanediol by alkaline hexacyanoferrate(III) using Ru(III) or Ru(VI) as catalyst

Article abstract of DOI:10.1002/poc.1090

The reactions of 2,3-butanediol by hexacyanoferrate(III) in alkaline medium using ruthenium compounds as catalysts have been studied spectrophotometrically. The effect on the reaction rate of concentration of substrate, oxidant, catalyst and basicity of the medium leads to similar experimental rate equations for both catalysts, Ru(III) and Ru(VI). The reaction mechanism involves the formation of a catalyst-substrate complex that yields a carbocation for Ru(VI) or a radical for Ru(III) oxidation. Hexacyanoferrate(III) 's role is the catalyst regeneration. The rate constants of complex decomposition and catalyst regeneration have been determined. Copyright

Full text of DOI:10.1002/poc.1090

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2006; 19: 597–602
Published online 2 August 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.1090
Oxidation of 2,3-butanediol by alkaline
hexacyanoferrate(III) using Ru(III) or Ru(VI) as catalysty
*
´
F. J. Poblete, A. E. Mucientes, S. Villarreal, F. Santiago, B. Caban˜ as and R. E. Gabaldon
´
´
´
Departamento de Quımica Fısica, Facultad de Ciencias Quımicas, Universidad de Castilla-La Mancha,
´
Avda Camilo Jose Cela, Ciudad Real, Spain
Received 30 September 2005; revised 6 April 2006; accepted 18 April 2006
ABSTRACT: The reactions of 2,3-butanediol by hexacyanoferrate(III) in alkaline medium using ruthenium
compounds as catalysts have been studied spectrophotometrically. The effect on the reaction rate of concentration
of substrate, oxidant, catalyst and basicity of the medium leads to similar experimental rate equations for both
catalysts, Ru(III) and Ru(VI). The reaction mechanism involves the formation of a catalyst–substrate complex that
yields a carbocation for Ru(VI) or a radical for Ru(III) oxidation. Hexacyanoferrate(III)’s role is the catalyst
regeneration. The rate constants of complex decomposition and catalyst regeneration have been determined.
Copyright # 2006 John Wiley & Sons, Ltd.
KEYWORDS: alcohol oxidation; catalysed; Ruthenium(III) and (VI) and hexacyanoferrate(III).
INTRODUCTION
ate, were purchased from Merck (A.R. grade) and
ruthenium trichloride (Johnson–Matthey). The solutions
were prepared using water from an OSMO BL-6
deionizer from SETA. A stock solution of ruthenium
trichloride (0.0024 M) was prepared by dissolving the
sample in very dilute hydrochloric acid 0.1638 M.
Sodium ruthenate solution was prepared following the
Lalitha–Sethuram procedure.8 The purity of ruthenate
stock solutions was assessed by taking into account that
the ratio between the absorbance at 465 nm and 386 nm
should be equal to 2.07 for pure ruthenate.11
All kinetic runs were initiated by the addition of
substrate to a mixture containing the other reagents. The
oxidation kinetics of 2,3-butanediol were followed by
measuring the absorbance of hexacyanoferrate(III) at
420 nm (e ¼ 1000 Mꢀ1 cmꢀ1) on a Simadzu UV-160
spectrophotometer. The initial rates method was used
for kinetic analysis.12 The ionic strength was kept
constant at 0.5 M by the addition of sodium perchlorate.
The only organic reaction product detected for the
oxidation of 2,3-butanediol was 3-hydroxy-2-butanone,
which was identified using a Hewlett-Packard 5890
Series II gas chromatograph equipped with a BP-21
polyethylene glycol column (50 m long ꢁ 0.22 mm i.d.,
25 mm film thickness). The stoichiometry obtained
showed that one mole of diol consumed two moles of
hexacyanoferrate(III):
The oxidation of organic compounds such as alcohols and
organic acids is a topic of great interest, especially if the
organic substrates are not easily oxidised by common
oxidants.1–3 Thus an alternative to solve this problem is
the addition of catalytic quantities of transition-metal ions
and a soft cooxidant to the reaction.4 The catalytic activity
of these ions is attributed to their capacity to exist in more
than one oxidation state, their capacity to form complexes
and their capacity to change their coordination number.5
These facts justify that actually the ruthenium complexes
were employed in the homogeneous alcohol oxidations.
Moreover, the use of a catalyst allows the existence of
sensitive linkages in the alcohol molecule.6 Although
numerous kinetic studies have been made using Ru(III) or
Ru(VI) as catalysts,7–10 the catalytic behaviour has rarely
been compared. Thus, this work focuses on the
comparative study of both catalysts in the oxidation of
2,3-butanediol with alkaline hexacyanoferrate(III) in
order to predict which of the two is the best catalyst.
EXPERIMENTAL
All the reagents used, that is, hexacyanoferrate(III),
sodium hydroxide, 2,3-butanediol and sodium perchlor-
3ꢀ
R-CHOH-R0 þ 2FeðCNÞ6 þ 2OHꢀ ! R-CO-R0
´
*Correspondence to: F. J. Poblete, Departamento de Quımica Fısica,
Universidad de Castilla-La Mancha, Avda. Camilo Jose Cela, 10,
´
(1)
þ 2FeðCNÞ46ꢀ þ 2H2O
where
R ¼ CH3—CHOH— and R0 ¼ CH3—
´
13071 Ciudad Real, Spain.
E-mail: fcojavier.poblete@uclm.es
ySelected paper presented at the 10th European Symposium on Organic
Reactivity, 25?30 July 2005, Rome, Italy.’
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 597–602

598
F. J. POBLETE ET AL.
Table 1. Kinetic parameters obtained by non-linear
RESULTS
regression fitting from Eqn 2
Figure 1 shows the variation of initial rate with respect to
k0/min
k00/102 min
k000/103 min Mꢀ1
½FeðCNÞ36ꢀꢂ (Fig. 1A) and [diol]0 (Fig. 1B) for both
0
catalysts. These results have led to the following
expression at constant concentrations of catalyst and
hydroxide ions:
For Ru(III)
For Ru(VI)
2.83 ꢃ 0.09
7.01 ꢃ 0.10
1.25 ꢃ 0.06
3.02 ꢃ 0.04
3.03 ꢃ 0.09
1.72 ꢃ 0.20
[Ru] ¼ 1.0 ꢁ 10-6 M, [NaOH] ¼ 0.15 M, I ¼ 0.5 M and T ¼ 30 8C.
½FeðCNÞ36ꢀꢂ0½diolꢂ0
The possible formation of free radicals as intermediates
was investigated by adding radical scavengers to the
reaction mixture. In Ru(III) case, polymeric species has
been obtained by the addition of 0.7 M acrylonitrile
indicating that radicals are present in the reaction
medium. However, for Ru(VI), no effect has been
observed when 0.01 M acrylonitrile or 1.6 ꢁ 10ꢀ4 M
2,4,6-tri-tert-butylphenol (a stronger radical scavenger)
has been added.
The oxidation of cyclobutanol was carried out because
the nature of its oxidation products depends on the reaction
mechanism. One-electron oxidation produces acyclic four-
carbon compounds, which appear to be derived from the
primary free radical ꢄCH2CH2CH2CHO, whereas two-
electron oxidation produces cyclobutanone directly.13,14
Under the kinetic conditions (cyclobutanol] ¼ 0.08 M,
[hexacyanoferrate(III)] ¼ 2.0 ꢁ 10ꢀ3 M, [catalyst] ¼ 2.5 ꢁ
10ꢀ6 M, [OHꢀ] ¼ 0.1 M, I ¼ 0.5 M and T ¼ 308C) the
reaction yields butanal as the major product in the case of
Ru(III) and cyclobutanone in the case of Ru(VI).
v0 ¼
3ꢀ
k0½diolꢂ0 þ k00½FeðCNÞ6 ꢂ0 þ k000½diolꢂ0½FeðCNÞ63ꢀꢂ0
(2)
where the parameters obtained by non-linear regression
fit are shown in Table 1.
This equation justifies the change of order from one to
zero for both hexacyanoferrate(III) and diol species upon
increasing their concentrations.
The variation of v0 with [catalyst]0 was linear (Fig. 2),
being negligible the uncatalysed process. Thus, the rate
equation would be:
v0 ¼ kc½catalystꢂ0
(3)
where the values of kc are 50.57 ꢃ 0.02 minꢀ1 for Ru(III)
and 62.47 ꢃ 0.9 minꢀ1 for Ru(VI).
As shown in Fig. 3, the plots of v0 versus [OHꢀ] reach a
maximum for both catalysts. In the case of Ru(VI), v0 does
not tend to be zero at very low [OHꢀ], whereas it does for
Ru(III). The variation of v0 with the basicity of the medium
is complicated and obeys the following equation:
The presence of a hydrogen on the a-carbon of the
alcohol is necessary for the reaction progress15 because
tertiary alcohols (0.1 M tert-butanol) do not react under
kinetic conditions16 for both catalysts.
The observed oxidation rate of CD3—CDOD—CD
was compared with that of CH3—CHOH—CH3 in order
2
A0 þ A1½OHꢀꢂ þ A2½OHꢀꢂ
v0 ¼
(4)
2
1 þ B1½OHꢀꢂ þ B2½OHꢀꢂ
The best average error was obtained for Ru(III) when
A0 ¼ 0 and for Ru(VI) when A2 ¼ 0.
Figure 1. Plots of v0 versus [Fe(CN)36ꢀ]0, [2,3butanediol] ¼ 0.03 M (A) and plots of v0 versus [2,3-butanediol]0, [Fe(CN)63ꢀ]0 ¼
8.0 ꢁ 10ꢀ4 M (B). [Ru]0 ¼ 1.0 ꢁ 10ꢀ6 M, [NaOH] ¼ 0.15 M, I ¼ 0.5 M and T ¼ 30 8C. Ru(III) (a) and Ru(VI) (b).
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 597–602

OXIDATION OF 2,3-BUTANEDIOL BY ALKALINE HEXACYANOFERRATE(III)
599
Figure 2. Effect of [catalyst]0 on the initial rate. [2,3-
butanediol]0 ¼ 0.05 M, [Fe(CN)36ꢀ]0 ¼ 8.0 ꢁ 10ꢀ4 M, [OHꢀ] ¼
0.15 M, T ¼ 30 8C, I ¼ 0.5 M. Ru(III) (a) and Ru(VI) (b).
Figure 3. Variation of v0 with respect to [NaOH].
[Fe(CN)36ꢀ]0 ¼ 8.0 ꢁ 10ꢀ3 M, [2,3-butanediol]0 ¼ 0.02 M,
[Ru]0 ¼ 1.0 ꢁ 10ꢀ6 M, I ¼ 0.5 M and T ¼ 30 8C. Ru(III)
(a) and Ru(VI) (b).
k1
2þ
RuIIIðH2OÞ5ðOHÞ þ OHꢀ ÐRuIIIðH2OÞ4ðOHÞ2þþH2O
(5a)
to verify the existence of a kinetic isotope effect. A
substantial primary kinetic isotope effect was indeed
observed [(v0,H/v0,D) ¼ 5.9] for both catalysts under the
following kinetic conditions: [catalyst] ¼ 2.0 ꢁ 10ꢀꢀ6 M,
k2
RuIIIðH2OÞ4ðOHÞ2þþOHꢀ ÐRuIIIðH2OÞ3ðOHÞ3 þ H2O
(6a)
3ꢀ
½FeðCNÞ6 ꢂ ¼ 1:2 ꢁ 10ꢀ3 M, [alcohol] ¼ 0.5 M, [OH ] ¼
0.2 M, I ¼ 0.5 M and T ¼ 308C.
And, for Ru(VI) process, RuO24ꢀ and RuO4(OH)3ꢀ are the
catalytic species18:
k10
DISCUSSION
3ꢀ
RuVIO42ꢀ þ OHꢀ Ð RuVIO4ðOHÞ
(5b)
(6b)
Before formulating the probable oxidation mechanism, it
may be helpful to select the ruthenium species that may
act as catalyst in the reaction.
The dependence of v0 on [OHꢀ] in the process
catalysed by Ru may be justified by assuming the
existence of two active species of catalyst with similar
reactivity in equilibrium.
Thus, in the process catalysed by Ru(III) the active
catalytic species are Ru(H2O)4(OH)þ2 and Ru(H2O)3
(OH)3.17
k20
3ꢀ
4ꢀ
RuVIO4ðOHÞ þ OHꢀ Ð RuVIO4ðOHÞ2
Both in the case of the oxidation catalysed by
Ru(III) as in the case of the process catalysed by
Ru(VI), the dependence of initial rate on substrate fits
Michaelis–Menten model. Therefore, we suggest the
existence of an intermediate complex formed from the
organic substrate and the respective active catalytic
species.
ð7aÞ
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 597–602

600
F. J. POBLETE ET AL.
ð7bÞ
Both complexes, C1þ and C21ꢀ, decompose slowly but
in a different way.
RuðH2OÞ5OH2þ þ OHꢀ ! RuðH2OÞ4ðOHÞ2þ þ H2O
In the case of Ru(III), the decomposition of Cþ1 yields a
ketyl radical and Ru(II) and it takes place by means of a
homolitic rupture and a hydrogen atom transfer from the
a—CH bond of the alcohol to the oxygen of the hydroxo
ligand of ruthenium. This transfer is experimentally
guaranteed by the presence of free radicals in the reaction
mixture and the presence of butanal as major product in
the oxidation of cyclobutanol.
(10a)
ꢄ
ꢅ
3ꢀ
4ꢀ
RR0 C OH þ FeðCNÞ6 ! RR0 C OH þ FeðCNÞ6
(11a)
ꢅ
RR0 C OH þ OHꢀ ! RR0CO þ H2O
(12a)
And for Ru (VI):
ꢄ
k2
Cþ1 ꢀ! RR0 C OH þ RuIIðH2OÞ5OHþ
(8a)
k30
3ꢀ
4ꢀ
RuIVO3OH3ꢀ þ FeðCNÞ6 ꢀ! RuVO3OH2ꢀþ FeðCNÞ6
However, in the case of catalyst Ru(VI), the complex
C21ꢀ decomposes by means of a heterolitic rupture and the
subsequent hydride transfer from the a—CH bond of the
substrate to the oxoligand of ruthenium as follows:
(9b)
3ꢀ
4ꢀ
RuVO3OH2ꢀ þ FeðCNÞ6 ! RuVIO3OHꢀ þ FeðCNÞ6
(10b)
k20
ꢅ
C21ꢀ ꢀ! RR0 C OH þ RuIVO3OH3ꢀ
(8b)
ꢅ
RR0 C OH þ OHꢀ ! RR0CO þ H2O
(11b)
This process is favoured by the prior coordination of
the organic substrate to the metal through the oxygen of
the hydroxyl group and supported by the following
experimental results: (a) a moderate kinetic isotope effect,
which indicates cleavage of a C—H bond in the absence
of free radicals in the reaction mixture, (b) the presence of
cyclobutanone as the unique product in the oxidation of
cyclobutanol and (c) the negative value of the Hammett
reaction constant found for the oxidation of benzyl
alcohol.19
From the species involved in these decompositions,
two new processes take place: the regeneration of the
catalyst and the formation of final products.
The dependence of initial rate v0 on [Fe(CN)36ꢀ], can be
justified by the oxidation of the reduced species of
catalyst precisely through hexacyanoferrate(III).
For Ru (III):
In this way, the role of Fe(CN)36ꢀ is limited to
regenerate catalysts, at least, in the case of Ru(VI).
A similar mechanism can be used for the other active
catalyst species, Ru(H2O)3 (OH)3 and RuO4(OH)3ꢀ
.
Thus, the experimental rate equation can be written as:
d½FeðCNÞ36ꢀꢂ
ð0Þ
ꢀ
¼ 2k3 ½RuðII or IVÞꢂ½FeðCNÞ36ꢀꢂ
þ 2k3 ½Ru0ðII or IVÞꢂ½FeðCNÞ36ꢀꢂ
dt
ð0Þ
(13)
where the concentrations of reduced species of the active
catalytic species of ruthenium appear. Therefore, it was
supposed that the constant of catalyst regeneration is the
same for the two active species, k3.
By the application of the steady-state conditions to the
reduced species of catalyst and complexes, C, the
theoretical rate equation of disappearance of Fe(CN)63ꢀ
can be obtained.
k3
3ꢀ
RuIIðH2OÞ5OHþ þ FeðCNÞ6 ꢀ! RuIIIðH2OÞ5OH2þ
4ꢀ
þ FeðCNÞ6
(9a)
For Ru(III):
d½FeðCNÞ36ꢀꢂ
dt
2 k1k2k3kA½FeðCNÞ3ꢀꢂ½diolꢂ½RuðIIIÞꢂT
6
ꢀ
¼
(14a)
k1k2kA½diolꢂ þ k3ðkꢀ1 þ k2ÞkB½FeðCNÞ3ꢀꢂ þ k1k3kA½FeðCNÞ3ꢀꢂ½diolꢂ
6
6
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 597–602

OXIDATION OF 2,3-BUTANEDIOL BY ALKALINE HEXACYANOFERRATE(III)
601
also show a first order with respect to catalyst
concentrations. The reaction mechanism proposed
involves oxidation of diol by Ru(VI) or Ru(III) through
the formation of a substrate–catalyst complex, which
subsequently decomposes to give Ru(IV) or Ru(II)
species. Intermediate complex decomposition involves
a hydrogen transfer from the a—C—H bond of the
alcohol to the oxoligand of ruthenium in the case of
Ru(VI), and a hydride transfer for the Ru(III) case. The
role of Fe(CN)36ꢀ is the regeneration of the catalyst. For
both oxidations the rate constants of complex decompo-
sition and catalyst regeneration have been obtained
(Scheme 1).
Table 2. Rate constants of complex decomposition and
catalyst regeneration
Ru(VI)
Ru(III)
k2 (minꢀ1
)
2.9 ꢃ 0.1 ꢁ 102
1.66 ꢃ 0.03 ꢁ 102
1.76 ꢃ 0.07 ꢁ 105
k3 (l ꢄ molꢀ1 ꢄ min-1)
7.1 ꢃ 0.8 ꢁ 104
[NaOH] ¼ 0.15 M, I ¼ 0.5 M and T ¼ 30 8C.
The [Ru(III)]T takes in account [Ru(III)] and [Ru(II)]
and kA ¼ k1[OHꢀ] þ k1k2[OHꢀ]2, kB ¼ kA þ 1.
For Ru(VI):
d½FeðCNÞ36ꢀꢂ
dt
2 k1k2k3kA½FeðCNÞ3ꢀꢂ½diolꢂ½RuðVIÞꢂT
6
ꢀ
¼
(14b)
k1k2kA½diolꢂ þ k3ðkꢀ1 þ k2ÞkB½FeðCNÞ3ꢀꢂ þ k1k3kA½FeðCNÞ3ꢀꢂ½diolꢂ
6
6
[Ru(VI)]T is the sum of [Ru(VI)] and [Ru(IV)]
since [Ru(V)] is negligible because such species are
only involved in fast steps and kA ¼ 1 þ k1[OHꢀ],
kB ¼ kA þ k1k2[OHꢀ]2.
Both rate equations are consistent with experimental
results obtained and justify the dependence of v0 on
[OHꢀ], [Fe(CN)36ꢀ], [Substrate] and [Catalyst].
Since the parallelism that shows the dependence of v0
on catalyst, substrate and hexacyanoferrate(III) in the
process catalysed by Ru(VI) and Ru(III), both theoretical
rate equations are very similar. The only difference in the
equations is due to the slightly different behaviour of v0
with respect to [OHꢀ]. The equation that summarizes the
catalytic process is:
d½FeðCNÞ36ꢀꢂ
ꢀ
dt
A½FeðCNÞ36ꢀꢂ½diolꢂ½RuðVIÞꢂT
¼
(15)
B½diolꢂ þ C½FeðCNÞ36ꢀꢂ þ D½FeðCNÞ63ꢀꢂ½diolꢂ
By comparing both Eqns 14a and b with the respective
experimental rate equations it is possible to estimate the
rate constants for intermediate complex decomposition,
k2, and catalyst regeneration, k3. These results are shown
in Table 2.
CONCLUSION
The kinetics for the oxidation of 2,3-butanediol by
Fe(CN)36ꢀ, using Ru(VI) or Ru(III) as catalysts, are
governed by similar experimental rate equations. These
equations show a change of order from one to zero for
both hexacyanoferrate(III) and diol concentrations. They
Scheme 1
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 597–602

602
F. J. POBLETE ET AL.
8. Lalitha TV, Sethuram B. Transit. Met. Chem. 1992; 17: 29.
9. Dwivedi RK, Verma M, Kumar P, Behari K. Tetrahedron 1983; 39:
The fact that Ru(III) reaches the maximum rate at
lower concentrations of substrate and hexacyanoferrate(III)
suggests that the catalysis of Ru(III) is better than Ru(VI).
815.
10. Singh HS, Singh RK, Singh SM, Sisodia AK. J. Phys. Chem. 1977;
81: 1044.
11. Larsen RP, Ross LE. Anal. Chem. 1959; 31: 176.
´
´
12. Mucientes AE, Poblete FJ, Gabaldon RE, Rodrıguez MA, Santiago
REFERENCES
F. J. Chem. Res. (S). 1998; 6: 286.
13. Rocek J, Radkowsky AE. J. Am. Chem. Soc. 1973; 95: 7123.
14. Meyer K, Rocek J. J. Am. Chem. Soc. 1972; 94: 1209.
15. Marmion ME, Takeuchi KJ. J. Chem. Soc. Dalton Trans. 1988:
284.
1. Hiremath GA, Timmanagoudar PL, Nandibewoor ST. J. Phys. Org.
Chem. 1998; 11: 31.
2. Panari RG, Chougale RB, Nandibewoor ST. J. Phys. Org. Chem.
1998; 11: 448.
3. Griffith WP, Suriaatmaja M. Can. J. Chem. 2001; 79: 598.
4. Sheldon RA. Bull. Soc. Chim. Belg. 1985; 94: 651.
5. Sheturam B. Topics in Chemistry Series 1: Chemical Kinetics and
Reaction Mechanism. RBSA Publishers: Jaipur, 1991.
6. Griffith WP. Chem. Soc. Rev. 1992; 21: 179.
´
16. Mucientes AE, Poblete FJ, Rodrıguez MA, Santiago F. J. Phys.
Org. Chem. 1999; 13: 901.
17. Mucientes AE, Poblete FJ, Rodrıguez MA, Santiago F. J. Phys.
´
Org. Chem. 1997; 10: 662.
18. Mucientes AE, Poblete FJ, Rodrıguez D, Rodrıguez MA, Santiago
´
F. Int. J. Chem. Kinet. 1999; 31: 1.
´
19. Mucientes AE, Santiago F, Almena MC, Poblete FJ, Rodriguez-
Cervantes AM. Int. J. Chem. Kinet. 2002; 34: 421.
´
7. Mucientes AE, Poblete FJ, Santiago F, Rodrıguez MA. J. Chim
Phys. 1997; 94: 1642.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 597–602