Catalysis Letters Vol. 72, No. 3-4, 2001 233
N-oxidation of pyridines by hydrogen peroxide in the
presence of TS-1
Denis J.
Robinson a, Paul McMorn a, Donald Bethell b, Philip C. Bulman-Page c, C. Sly d, Frank King c,
Frederick E.
Hancock e and
G.J. Hutchings a
a Department of Chemistry, Cardiff University, Cardiff
CF10 3TB, UK
b Leverhulme Centre for Innovative Catalysis,
Department of Chemistry, University of Liverpool, Liverpool L69 3BX, UK
c Department of Chemistry, Loughborough University,
Leicestershire LE11 3TU, UK
d Robinson Brothers Ltd., West Midlands B70 0AH, UK
e R,T & E Division, Synetix, Billingham TS23 1IB,
UK
Received 14
November 2000; accepted 4 January 2001
In the production of aromatic N-oxides
using the oxidation
of N-containing heterocyclic aromatic substrates with H2O2 as oxidant,
the non-catalysed homogeneous oxidation
is found to play an important part in the overall reaction. In addition, when TS-1 is used as a catalyst, there are many potential
competitive interactions between the catalyst, the reactants and the products,
which limit the effectiveness of the catalyst.
It is concluded that the use of TS-1 and other microporous catalysts for the heterogeneous N-oxidation of pyridine and substituted pyridines needs to be interpreted with caution.
KEY WORDS: N-oxidation of
pyridine; hydrogen peroxide; TS-1; heterogeneous catalysis; uncatalysed oxygen
transfer
1. Introduction
A recent publication [1] describes
the use of micro- and mesoporous titanium-containing materials as catalysts for
the efficient conversion of substituted pyridines and related heterocyclic
compounds into their N-oxides by hydrogen peroxide. No control experiments, carried out in the absence of the catalysts, were reported
and the possible contribution of uncatalysed oxidation was ignored. In our investigation of the oxidation of sulfides by hydrogen peroxide in the pres-
ence of one of these catalysts, titanium silicalite (TS-1), to form first the
corresponding sulfoxide which was then con- verted into the sulfone [2] we
showed that uncatalysed oxy- gen transfer dominated sulfoxide formation, but
that TS-1 catalysis was essential for production of the sulfone. This paper reports
that, in the presence of TS-1, uncatalysed oxy- gen transfer from hydrogen peroxide to pyridine and the
pi- colines does indeed contribute substantially to the yield of the N-oxides,
and, from preliminary studies, we show that the efficiency of the catalytic reaction
can best be under- stood in terms of competitive binding effects involving the
reactants and the N-oxide produced.
2. Experimental
TS-1
was prepared according to the method
of Taramasso et al. [3] and calcined at 550 ◦C immediately
prior to use. Reactions were carried out using the heterocyclic substrate (0.05 mol), aqueous hydrogen
peroxide (30 vol%; 0.05 mol) and TS-1 (200 mg). Two procedures were investigated; in
method A, the reaction was initiated by adding the premixed
reactants to the catalyst, while in method B the
catalyst and heterocycle were stirred together for 5 min after which ad- dition
of the oxidant started the reaction. The reaction mix- tures, consisting of the
catalyst suspended in a single aque- ous organic liquid phase, were stirred at
60 ◦C
for a period of 24 h, samples of the liquid being removed at intervals,
quenched, and then analysed by GC. Product identification was by NMR
spectroscopy, GC/MS (Fisons Trio 1000) and
by comparison of GC retention times with those of authentic specimens. We have demonstrated [2] that, under these
re- action conditions, decomposition of hydrogen peroxide
cor- responds with the extent of oxygen transfer to the organic
substrate; other decomposition pathways of H2O2 are negli- gible.
3. Results and discussion
Control experiments were carried out
on N-oxidation of pyridine, 2-, 3- and 4-picolines in the absence of TS-1. The
yields of N-oxide after 24 h are in table 1 and indicate that substantial oxidation
occurs, the proportion being greater for
Table 1
|
Oxidation of pyridine and picolines using H2O2 as oxidant.a Catalyst/method N-oxide product yield (%)
|
|
a Substrate (0.05 mol), H2O2 (30 vol%, 0.05
mol) reacted at 60 ◦C for 24 h.
1011-372X/01/0400-0233$19.50/0
Ó 2001 Plenum Publishing Corporation
234 D.J. Robinson et al. / N -oxidation of pyridines
Table 2
Oxidation of substrate
mixtures.a
Substrates N-oxide product
yield (%) Pyridine 2-picoline 4-picoline
Pyridine and 2-picoline 80.2 30.3 –
2-picoline and 4-picoline – 42.6 17.0
a Each substrate (0.025 mol) reacted with H2O2 (30 vol%, 0.05 mol)
at 60 ◦C with TS-1 (0.2 g) using
method A for 24 h.
pyridine (pKa 5.27 at 20 ◦C)
[4] than for the more basic pi- colines (pKa-values: 2- 6.05, 3-
5.75 and 4- 6.10). Since the pH of 30
vol% aqueous H2O2 is ca. 5.5, the results sug-
gest that N-protonation is an inhibiting factor in these reac- tions.
The
presence of TS-1 enhances the extent of N-oxidation
which was essentially complete in the case of pyridine after 24 h (table 1).
Again the picolines were apparently less re- active, and reactions conducted using method B gave higher conversions than by method A. This
difference was particu- larly noticeable for 3- and 4-picoline and may point
to partial control of the
early stages of reaction by mass transport in these cases. The initial
rates of formation
of the N-oxides re- flected the 24 h conversions, but the increasing production of
pyridine N-oxide with time did not fit a simple second-order
kinetic law. We suggest that this is a consequence of equilib- ria, rapidly
established in this case, in which pyridine,
H2O2 and the product pyridine N-oxide compete for occupancy of the
intracrystalline volume of the TS-1, reaction occurring only when H2O2, activated by interaction with framework ti- tanium, is in the vicinity of a molecule
of free pyridine.
This situation declines in probability as the amount
of N-oxide in the
reaction system builds
up. An analogous situation has al-
ready been described in the acetylation of anisole by acetic anhydride in the
presence of zeolite H-β [5].
Further light on this
situation is shed by the results of re- actions
in which 1 : 1 mixtures of pyridine (0.025 mol) and 2-picoline
(0.025 mol) and of 2- and 4-picolines (0.025 mol of
each) were allowed to compete for H2O2 (0.05 mol) us- ing
method A. The yields of the two N-oxides produced are given
in table 2 and again these reflect the initial reaction rates.
Clearly, 2-picoline reduces the conversion of pyri- dine to the N-oxide somewhat, but the presence of pyridine inhibits the formation of 2-picoline N-oxide almost com- pletely, i.e., to the background level. In
the case of the mix- ture
of 2- and 4-picolines, which have closely similar pKa- values, formation of both
products is reduced by about two thirds.
Again the pattern of results is consistent with com- petitive binding
of the heterocycles within the pores of TS-1. We conclude
that heterogeneous N-oxidation of pyridine
and substituted pyridines using H2O2 in
the presence of TS-1 and other microporous catalysts needs to be interpreted with caution. The uncatalysed process can make control
of the reaction difficult.
In addition, reactivity cannot be simply predicted
on the basis of the electronic characteristics of the heterocycle, but results from a complex
interplay of compet- itive
interactions between the catalyst, the reactants and the
products.
References
[1] M.R. Prasad, G. Kamalakar, G. Madhavi, S.J. Kulkarni and
K.V. Ragharan, J. Chem. Soc. Chem. Commun. (2000) 1577.
[2] D.J. Robinson, P. McMorn, D. Bethell, P.C. Bulman-Page,
C. Sly, F. King, F.E. Hancock and G.J. Hutchings, Phys. Chem. Chem. Phys.
(2000) 1523.
[3] M. Taramasso and B.
Notari, US Patent 4410501 (1983).
[4] H.H. Perkampus and G. Prescher, Ber. Bunsenges. Physik.
Chem. 72 (1968) 429.
[5] E.G. Derouane, C.J. Dillon, D. Bethell and S.B.
Derouane-Abd Hamid, J. Catal. 187 (1999) 209.
Hiç yorum yok:
Yorum Gönder