Facile, one-step production of niacin (vitamin B3) and other nitrogen-containing pharmaceutical chemicals with a single-site heterogeneous catalyst

Article abstract of DOI:10.1002/chem.200701679

Niacin (3-picolinic acid), which is extensively used as vitamin B ₃ in foodstuffs and as a cholesterol-lowering agent, along with other oxygenated products of the picolines, 4-methylquinoline, and a variety of pyrimidines and pyridazines, may be produced in a single-step, environmentally benign fashion by combining single-site, open-structure, heterogeneous catalysts witha solid source of active oxygen, namely acetyl peroxyborate (APB), in the absence of an organic solvent. The high activities, selectivities, and the relatively mild conditions employed with this single-site heterogeneous catalyst, coupled with ease of transport, storage, and stability of the solid oxidant, augurs well for the future use of APB in conjunction with other open-structure, single-site catalysts for fine-chemical, pharmaceutical, and agrochemical applications.

Full text of DOI:10.1002/chem.200701679

DOI: 10.1002/chem.200701679
Facile, One-Step Production ofNiacin (Vitamin B ) and Other Nitrogen-
3
Containing Pharmaceutical Chemicals with a Single-Site Heterogeneous
Catalyst
Robert Raja,*[a] John Meurig Thomas,*[b] Michael Greenhill-Hooper,[c]
Steven V. Ley,[d] and Filipe A. Almeida Paz[e]
Abstract: Niacin (3-picolinic acid),
which is extensively used as vitamin B3
in foodstuffs and as a cholesterol-low-
ering agent, along with other oxygenat-
ed products of the picolines, 4-methyl-
quinoline, and a variety of pyrimidines
and pyridazines, may be produced in a
single-step, environmentally benign
fashion by combining single-site, open-
structure, heterogeneous catalysts with
a solid source of active oxygen, namely
acetyl peroxyborate (APB), in the ab-
sence of an organic solvent. The high
activities, selectivities, and the relative-
ly mild conditions employed with this
single-site heterogeneous catalyst, cou-
pled with ease of transport, storage,
and stability of the solid oxidant,
augurs well for the future use of APB
in conjunction with other open-struc-
ture, single-site catalysts for fine-chem-
ical, pharmaceutical, and agrochemical
applications.
Keywords: acetyl peroxyborate
·
green chemistry heterogeneous
·
catalysis · mesoporous materials ·
oxidation
Introduction
and the incoming reactant.[1] Because of their high porosity,
such catalysts facilitate ingress of reactants to, and the
egress of products from, the active centers, which are so de-
signed, inter alia, as to liberate intermediate active species
that attack a reactant bound within a microcavity in a
shape-selective or regioselective manner. This yields a prod-
uct which, in general, would otherwise be difficult to synthe-
size. In addition to their regioselectivity, SSHCs possess the
advantage of combining the merits of conventional, single-
site homogeneous catalysts (and enzymes) and those of het-
erogeneous ones, in that separation of product from reactant
is generally facilitated.
Single-site heterogeneous catalysts (SSHCs) are those in
which the active centers are spatially isolated from one an-
other, and uniformly distributed over a large three-dimen-
sional (internal) surface of a porous, high-area solid such
that each site has the same energy of interaction between it
[a]Dr. R. Raja
School of Chemistry, University of Southampton
Highfield, Southampton SO17 1BJ (UK)
Fax : (+44)2380-593781
E-mail: R.Raja@soton.ac.uk
Niacin (3-picolinic acid, also known as Vitamin B3 or nic-
otinic acid) and other oxygenated products of the picolines,
4-methyl quinoline, 4-methyl derivatives of pyrimidine and
pyridazine, and 2-methyl pyrazine, are of considerable value
in pharmaceutical, agrochemical, and fine-chemical applica-
tions (see Schemes 1 and 2). Niacin is used extensively as vi-
tamin B3 in foodstuffs and as a cholesterol-lowering agent;
isonicotinic acid (4-picolinic acid) and its derivatives have
applications as an antibacterial drug for treating tuberculo-
sis, psoriasis, and arthritis and is also used as a plant growth
regulator, as well as a herbicide, pesticide, and as a corro-
sion regulator. Derivatives of quinoline carboxylic acids are
used as biocides, pesticides, anti-bacterial agents, cancer
drugs, seed disinfectants, herbicides (e.g. “Quinclorac”, “Im-
[b]Prof. Sir J. M. Thomas
Department of Materials Science
University of Cambridge, Pembroke Street
Cambridge CB2 3QZ (UK)
Fax : (+44)1223-740360
E-mail: jmt2@cam.ac.uk
[c]Dr. M. Greenhill-Hooper
Rio Tinto Minerals, 2 Place E Bouilleres
31036 Toulouse (France)
[d]Prof. S. V. Ley
Department of Chemistry, University of Cambridge
Cambridge CB2 1EW (UK)
[e]Dr. F. A. Almeida Paz
Department of Chemistry, CICECO, University of Aveiro
3810–193 Aveiro (Portugal)
2340
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2340 – 2348

FULL PAPER
ful nature of some of these processes, 2.8 tonnes of inorganic
waste is produced for every tonne of niacin generated. In
terms of environmental pollution, these methods also ac-
count for 1.5 tonnes of CO2 and 0.37 tonnes of NOx.
As part of a general program[3–6] that entails the design of
“green”, one-step production of desirable chemicals, both
thermally and photochemically,[7] we have evolved efficient,
straight-forward methods of producing all the above-named
fine-chemical compounds using a combination of a mono-
functional, single-site (open-structure) heterogeneous cata-
lyst that operates under mild conditions in association with
a dissolved, solid source of active oxygen, acetyl peroxybo-
rate (APB).[8] Earlier, we showed[9] that this stable source of
active oxygen—which may be indefinitely stored and trans-
ported with impunity, unlike hydrogen peroxide (H2O2) or
peracetic acid, for which it substitutes (see below)—is an ex-
tremely powerful selective oxidation system when dissolved
in aqueous solutions in the presence of a single-site micro-
porous catalyst containing redox active centers. Thus dis-
solved APB, in combination with ferric ions located inside a
microporous aluminophosphate (AlPO) catalyst (e.g.
FeIIIAlPO-31)[10] converts cyclohexane, in a single-step, sol-
vent-free fashion, with unprecedented efficiency (ca. 88%)
and exceptionally high selectivity (ca. 81%) to adipic acid
(an important precursor in the manufacture of nylon 6,6 and
polyurethanes) at 383 K.[9] More importantly, this approach
circumvents the need to use corrosive oxidants, such as
nitric acid, that are favored industrially, and is consequently
environmentally benign, as it does not produce undesirable
greenhouse gases such as N2O that contribute to ozone de-
pletion. This combination of dissolved APB and an open-
structure SSHC also readily converts styrene to styrene
oxide and a-pinene and (+)-limonene to their correspond-
ing epoxides (that are important intermediates in the fra-
grance industry).
Scheme 1.
We have also shown[11–13] that a wide range of primary,
secondary, benzylic, and other unsaturated alcohols, substi-
tuted olefins as well as phenols, anisoles and other substitut-
ed aromatics may also be selectively oxidized to desirable
products and specialty organics with APB and an appropri-
ately designed open-structure SSHC. Here, however, we
focus on a range of nitrogen-containing heterocyclic com-
pounds that are feedstocks for generating many desirable
pharmaceutical, agrochemical, and fine-chemical intermedi-
ates.
Scheme 2.
azaquin”), plant growth regulators, antibiotics, anti-fungal
agents for plants, trypsin inhibitor, charge control agent for
toners (copiers), and as metal-ion chelators (application in
plating baths). Pyridazine carboxylic acids are used widely
as plant growth regulators (e.g. “Clofencet MON21200”—
Monsanto). Almost all these pharmaceutical and agrochemi-
cal intermediates are usually produced from environmental-
ly aggressive oxidizing agents, such as chromic acid, perman-
ganates, selenium dioxide (SeO2), and tert-butylhydroperox-
ide + O2 and may involve two (or more) separate steps.
More recently, liquid-phase oxidation processes that involve
reacting 3-picoline with homogeneous cobalt and manganese
acetates with hydrobromic acid at relatively high tempera-
tures (483 K) and high pressures (100 atm) have been re-
ported.[2] The yields and selectivities are, however, quite low
(32% conversion with 19% selectivity for nicotinic acid).
Notwithstanding the aggressive and environmentally harm-
Results and Discussion
From prior 13C, 1H, and 11B NMR studies (solid and
liquid),[9] it has been established that, after dissolution of the
APB (which is not totally single-phase in the solid-state) in
water, the system establishes the equilibrium, given in Equa-
tion (1), with a half-life of approximately 2 h at 658C.
CH3COOH þ H2O2 Ð CH3COOOH þ H2O
ð1Þ
Chem. Eur. J. 2008, 14, 2340 – 2348
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2341

J. M. Thomas, R. Raja et al.
Figure 1. Top (left) and side (right) perspective views showing the accessibility of nitrogen-containing heterocyclic compound (3-picoline represented as
example) to the internal, high-area catalytic-active surface of the single-site MnAlPO-5 catalyst (the MnIII active sites are shown in yellow). The structur-
al drawings were performed by using a series of software packages, which include Crystal Impact Diamond,[20] Mercury,[21] BABEL for file conversion,[22]
GaussView, Chimera,[23] and POV-Ray[24] in tandem. The computation of zeolitic surfaces has been performed by selecting a probe radius of 1.4 Š.
Moreover, trigonal B(O)3 species are also present in solu-
tion, representing the average over an equilibrium involving
different species of this type including boric acid [B(OH)3].
Whilst the precise nature of the released “active” oxygen
and the critical reaction intermediate that effects the selec-
tive oxidations described below, are at present unclear, the
key feature of the use of this novel and readily preparable
solid oxidant (APB) is that the peroxyacetic acid releases its
active oxygen at the locus of the single-site redox active
center within an open structure that accommodates the reac-
tant. This facilitates shape-selective and regioselective oxi-
dation in the reactants described herein; for example, the
methyl group of the 4-picoline (see Figure 1) is more liable
to oxidative attack than the ring nitrogen (which is rapidly
oxidized in the absence of the solid single-site catalyst),
thereby yielding the desired product with high selectivity.
The single-site, open-structure, solid catalyst used in this
work, MnIIIAlPO-5, belongs to a large family of micropo-
rous aluminophosphates that have been fully characterized
in previous studies[14–16] and notably in our single-step syn-
thesis of e-caprolactam from cyclohexanone.[17] AlPO-5,
dants, such as peracetic acid (PAA), as well as conventional
aqueous peroxides, such as hydrogen peroxide (H2O2). To
rule out the possibility that changes in pH might affect the
observed activities or selectivities with neat PAA, additional
experiments were carried out by adjusting the pH to be the
same as that in the experiments with APB.
Selective oxidation of4-picoline to isonicotinic acid : The re-
sults are summarized in Figure 2 and Table 1 which show
5À
which is composed of corner-linked, alternating AlO4 and
3À
PO4
tetrahedra, has pores of diameter approximately
7.3 Š, large enough to allow ingress of all the reactant, ni-
trogen-containing heterocyclic compounds studied here, to
reach the isolated, active centers at which MnIII replaces
AlIII ions tetrahedrally coordinated to oxygen. Figure 1 illus-
trates the ease with which the reactants may gain access to
the interior of the open structure and hence to the active
site.
The performance of APB as an oxidant in contact with
the open-structure redox SSHC was compared with a
number of other known powerful and popular peroxidic oxi-
Figure 2. The effect of temperature in the catalytic, solvent-free oxidation
of 4-picoline using MnIIIAlPO-5 as the catalyst.
the influence of temperature and the nature and effect of
the different types of oxidant, respectively. It is clear that
the solid APB in conjunction with the MnIIIAlPO-5 catalyst
affords good conversions, and exceptional selectivities
(100% in a few cases, see Figure 2) are achieved in the pro-
duction of isonicotinic acid (4-picolinic acid). The effect of
2342
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2340 – 2348

Heterogeneous Catalysis
FULL PAPER
Table 1. The influence of oxidant and role of the borate component in
the solvent-free, selective oxidation of 4-picoline at 368 K.[a]
vestigated for comparison (Table 1). It is seen that both the
conversion and selectivity, even under optimized conditions
of temperature and contact time are distinctly inferior to
those of MnIIIAlPO-5, in part due to the narrower diameter
of the pores (ca. 5.5 Š) in TS-1 compared with those of
MnIIIAlPO-5 (7.3 Š).
With the robotic set-up (see Experimental Section) used
to follow the course of the liquid-phase oxidation in contact
with MnIIIAlPO-5, the kinetics of the process (Figure 3)
Oxidant
Microporous
Catalyst[b]
Conv
[mol%] 2[c]
Product selectivity
[mol%]
A
3[d]
4[e]
others
APB (solid)
PAA (neat)
PAA + Neobor Mn0.10Al0.90PO4
PAA + NaOAc Mn0.10Al0.90PO4
Mn0.10Al0.90PO4
Mn0.10Al0.90PO4
72.0
74.5
66.7
81.5
92.0
3.0
–
4.8
44.1 20.5 25.9 9.3
89.5 4.2 1.5 4.7
33.7 45.3 15.5 5.4
45.0 35.0 11.3 8.5
APB (solid)
TS-1 (2 wt% Ti) 61.9
[a]Reaction conditions: 1) With APB solid freshly prepared (PAA
22.4%): APB solid (ffi3.49 g) dissolved in double-distilled water (20.5 g),
4-picoline (ffi2.8 g), catalyst (ffi0.30 g); 2) With PAA (neat peracetic
acid): peracetic acid (25%) solution in acetic acid (4.2 g), double-distilled
water (20.5 g), 4-picoline (ffi2.8 g), catalyst (ffi0.30 g); 3) With PAA+
Neoborꢁ: peracetic acid (25%) solution in acetic acid (4.2 g), double-dis-
tilled water (20.5 g), Neoborꢁ (1 g), NaOH (1 g; pH ffi5.1), 4-picoline
(ffi2.8 g), catalyst (ffi0.30 g); 4) With PAA+NaOAc: peracetic acid
(25%) solution in acetic acid (4.2 g), double-distilled water (20.5 g),
NaOAc (0.934 g), NaOH (1 g, pH ffi5.1), 4-picoline (ffi2.8 g), catalyst
(ffi0.30 g). In all cases adamantane was the internal standard (ffi0.5 g);
Tꢀ368 K; t=4 h. [b]The absolute values of the elemental composition
have error limits of Æ3”10À3. [c] 2: Isonicotinic acid (4-picolinic acid).
[d] 3: 4-Picoline N-oxide. [e] 4: Isonicotinic acid N-oxide.
temperature (Figure 2) shows that at lower temperatures
(338 and 348 K) virtually no N-oxides (4-picoline N-oxide or
isonicotinic acid N-oxide) are formed. Conversion improves
significantly on increasing the temperature attaining reason-
able yields and selectivities at 368 K, but the selectivity for
the isonicotinic acid sharply decreases beyond this tempera-
ture.
For the purpose of comparison, the oxidation of 4-picoline
was also carried out with the same catalyst (MnIIIAlPO-5)
with neat peroxyacetic acid (PAA) rather than APB as the
oxidant. Although the conversions observed in both cases
were comparable, the selectivity for isonicotinic acid is re-
duced drastically when PAA was used—the propensity for
N-oxidation of the ring and of the para-methyl group was
equally favored with PAA. In one set of experiments, a com-
bination of peroxyacetic acid, sodium acetate trihydrate, and
NaOH was used to bring the pH close to that for APB,
while in another set, Neoborꢁ borax pentahydrate
(Na2B4O7·5H2O; ex Borax Europe Limited), an important
component that is present in APB, was used. The above sets
of experiments were particularly designed to exclude the ef-
fects of pH, if any. The addition of Neoborꢁ significantly en-
hances the selectivity for 4-picolinic acid, whereas in the ab-
sence of Neoborꢁ, N-oxidation is simultaneously favored (as
with the results for neat PAA). However, this series of ex-
periments shows that the borate component contained in
Neoborꢁ is critical for the high activities and selectivities ob-
served when APB is used as the oxidant. Thus, while APB
serves as a source of “active” oxygen for these catalytic oxi-
dation processes, the role of key borate-containing species
that arise from dissolution of APB also seems necessary for
the high performance and high selectivity.
Figure 3. Typical kinetic plots contrasting the performance and selectivity
of the MnIIIAlPO-5 catalyst at 348 (top) and 378 K (bottom) for the sol-
vent-free oxidation of 4-picoline.
were followed. It is seen that up to 4 h contact time, all the
conversion of the 4-picoline is to the desired isonicotinic
acid at 348 K (Figure 3, top), and this is further oxidized to
the isonicotinic acid N-oxide at prolonged contact times. At
higher temperatures (378 K), a small amount of isonicotinic
acid N-oxide is produced right from the outset along with
isonicotinic acid (Figure 3, bottom), but there is a steep in-
crease in the concentration of the isonicotinic acid N-oxide
at prolonged contact times (beyond 4 h). Interestingly, in
both cases, 4-picoline N-oxide is not generated at lower con-
tact times (up to 4 h), irrespective of the temperature used,
and is produced only in small amounts beyond 6 h. The in-
fluence of substrate/oxidant mole ratio on the activity and
selectivity was also studied (Figure 4). There is an optimal
A titanosilicate catalyst, known as TS-1[18,19] and which a
popular single-site, open-structure solid catalyst, was also in-
Chem. Eur. J. 2008, 14, 2340 – 2348
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2343

J. M. Thomas, R. Raja et al.
4-picoline compared with the 3-picoline (Figure 5), and the
total percentage selectivity of the desired acids amounts to a
value expected from the non-competitive (separate) experi-
Figure 4. The influence of substrate:oxidant mole ratios, showing that an
optimal ratio is required for enhancing the activity and selectivity, in the
oxidation of 4-picoline after 4 h at 368 K.
Figure 5. The competitive oxidation of an equimolar mixture of 3- and 4-
picoline at 368 K substantiates the fact that, the single-site MnIIIAlPO-5
catalyst plays a crucial role in influencing the shape- and regioselectivity
of the reaction, with respect to the position of the methyl group and that
of the nitrogen moiety within the picolines.
ratio (3:1) for enhanced performance and selectivity for the
MnIIIAlPO-5 catalyst.
Selective oxidation of3-picoline to nicotinic acid : As with 4-
picoline there are good conversions (reaching a maximum
of ca. 87% at 378 K) and high selectivities (ca 78% at
348 K); but, interestingly, more of the 3-picoline N-oxide is
produced in sharp contrast to what was observed with 4-pi-
coline, under comparable conditions of temperature and
contact time (Table 2, Scheme 1). Kinetic experiments fur-
ments. Again, consistent with earlier observation we see
that, 3-picoline N-oxide is generated in much larger excess
(ca. 4.5 times) than the corresponding 4-picoline N-oxide de-
rivative.
Selective oxidation of4-methylquinoline : From Table 3 (see
also Scheme 3) it is seen that significantly higher tempera-
Table 2. The influence of temperature in the one-step, solvent-free pro-
duction of nicotinic acid using a single-site MnIIIAlPO-5 catalyst.[a]
T
[K]
Oxidant
Conv
[mol%]
Product selectivity [mol%]
G
2A[b]
3A[b]
4A[b]
others
Table 3. The influence of temperature and comparative performance of
other oxidants in the selective oxidation of 4-methylquinoline to cincho-
ninic acid.[a]
338
348
358
368
378
348
APB
APB
APB
APB
APB
PAA
37.0
46.8
58.3
71.2
86.9
54.5
78.5
67.3
60.0
45.7
42.6
40.6
21.4
25.0
19.5
18.0
15.0
19.3
–
7.5
20.3
28.8
27.8
24.5
–
–
–
7.5
14.5
15.5
Oxidant
T
[K]
Conv
[mol%]
Product selectivity [mol%]
G
2Q[b]
3Q[b]
4Q[b]
others
APB
APB
APB
APB
APB
APB
PAA
H2O2
338
348
368
398
423
448
398
398
2.5
3.3
7.2
19.8
31.2
46.9
19.5
17.6
80.2
79.8
83.5
88.5
65.1
32.1
35.5
21.5
19.9
20.1
16.5
9.0
15.9
25.7
21.2
45.0
–
–
–
2.6
9.0
15.3
15.4
18.0
–
–
–
–
10.2
27.0
28.0
15.6
[a]Reaction conditions: 3-picoline ( ffi2.8 g), catalyst Mn0.10Al0.90PO4
(ffi0.30 g), adamantane (internal standard; ffi0.5 g), t=4 h; for more de-
tails see the footnotes in Table 1. [b]For the products see Scheme 1.
ther confirm that 3-picoline N-oxide is produced right at the
outset, again, in contrast to the observations in Figure 3.
The position of the methyl group, in relation to the nitrogen
within the picolines, clearly plays a role in influencing the
shape- and regioselectivity of the reaction, when APB is
used in combination with MnIIIAlPO-5.
[a]Reaction
conditions:
4-methylquinoline
( ffi4.25 g),
catalyst
Mn0.10Al0.90PO4 (ffi0.50 g); H2O2 (25% 1.5 g), double-distilled water
(20.5 g), adamantane (internal standard, ffi0.5 g), t=4 h; for more details
see the footnotes in Table 1. [b]For the products see Scheme 3.
In a competitive experiment, an equimolar mixture of 3-
and 4-picoline was oxidized with APB with MnIIIAlPO-5 as
the catalyst. Not surprisingly, in view of its cylindrical sym-
metry, there is a slight excess (ca. 25%) of conversion of the
tures are required to achieve reasonable conversions (e.g.,
47% at 448 K), but this consequently results in a lower se-
lectivity (ca. 32%) for the desired cinchoninic acid. Much
higher selectivities and moderate conversions (88 and 20%,
2344
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2340 – 2348

Heterogeneous Catalysis
FULL PAPER
with those of peracetic acid and
H2O2. We again note that APB
is superior not only in overall
performance compared with
these oxidants, but especially in
regard to the desired product
selectivity. As in the case of 4-
picoline, the use of APB in
combination
with
the
MnIIIAlPO-5 SSHC did not
result in the oxidation of the ni-
trogen within the ring, both for
4-methylpyrimidine
and
4-
methylpyridazine, after 5 h at
348 K.
Scheme 3.
respectively) are obtained at 398 K (Table 3). The relatively
large size of the 4-methylquinoline, in comparison to the 3-
and 4-picolines, is doubtless responsible for the sluggish per-
formance of this substrate, due to the diffusion limitations
presented within the microporous AlPO-5 cavity (Figure 1).
However, it is worth noting that the solid APB in conjunc-
tion with the microporous MnIIIAlPO-5 catalyst displayed a
much higher selectivity for the cinchoninic acid compared to
peracetic acid (ca. 2.5 times) or H2O2 (ca. 4 times) under
identical reaction conditions (Table 3).
Selective oxidation of4-methylpyrimidine, 2-methylpyra-
zine, and 4-methylpyridazine: Table 4 (see also Schemes 2,
4, and 5) highlights the salient features of these catalyzed
conversions and compares the performance of APB oxidant
Table 4. Comparative performance of APB, PAA, and H2O2 in the sol-
vent-free oxidation of 2-methylpyrazine, 4-methylpyrimidine, and 4-meth-
ylpyridazine.
Substrate Oxidant
T
[K]
Conv
Product selectivity [mol%]
A
6[b]
7[b]
8[b]
2-methyl
pyrazine
APB
APB
PAA
H2O2
338 38.9
368 67.3
368 61.2
368 59.8
41.2
4.5
–
28.5
56.3
19.2
16.9
21.5
23.0
60.6
49.5
9.0
Scheme 4.
16.2
20.0
8.5
25.0
5 A[b]
6 A[b]
7 A[b]
8 A[b]
Kinetic plots (Figure 6) confirm the above observation. In
the case of 4-methylpyrimidine, pyrimidine-4-carboxalde-
hyde and pyrimidine-4-carboxylic acid were the prime prod-
ucts of conversion up to about 5 h of contact time (Figure 6,
top). Interestingly, after 5 h, small amounts of pyrimidine-4-
carboxylic acid-1-oxide and pyrimidine-4-carboxylic acid-3-
oxide started appearing in the product stream. 4-Methylpyri-
midine-1-oxide and 4-methylpyrimidine-3-oxide were not
generated, in line with the results with 4-picoline. The kinet-
ic plot of 2-methylpyrazine (Figure 6, bottom) was very dif-
ferent though—2-methylpyrazine-1-oxide and 2-methylpyra-
zine-4-oxide were simultaneously produced along with pyra-
zine-2-carboxaldehyde and pyrazine-2-carboxylic acid, and
products resulting from the oxidation of the nitrogen group
4-methyl
pyrimidine
APB
338
368
368
368
51.5
81.2
65.4
45.2
45.3
–
–
54.5
69.5
25.3
21.2
–
–
45.0
29.8
–
APB
PAA
H2O2
30.5
29.5
14.2
34.7
6B[b]
100
86.6
82.3
35.0
41.2
7B[b]
8B[b]
4-methyl
pyridazine
APB
APB
APB
PAA
H2O2
338
32.5
–
–
–
49.3
48.6
–
338
368
368
368
78.4[c]
77.2
46.5
26.3
13.5
17.9
15.4
10.3
[a]Reaction conditions: substrate ( ffi2.8 g), catalyst Mn0.10Al0.90PO4
(ffi0.30 g), adamantane (internal standard, ffi0.5 g), t=5 h; for more de-
tails see the footnotes in Tables 1 and 3. [b]For the products see
Schemes 2, 4, and 5. [c] t=8 h.
Chem. Eur. J. 2008, 14, 2340 – 2348
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2345

J. M. Thomas, R. Raja et al.
Scheme 5.
in the 2-methylpyrazine start to dominate at prolonged con-
tact time (i.e., beyond 5 h). The use of the single-site micro-
porous catalyst, in conjunction with APB, coupled with the
position of the methyl group with respect to that of the two
nitrogen atoms in these types of substrates, predisposes
shape- and regioselective behavior.
Figure 6. Kinetic plots contrasting the shape- and regioselective behavior
of the single-site MnIIIAlPO-5 catalyst in the solvent-free oxidation of 4-
methylpyrimidine (top) and 2-methylpyrazine (bottom) at 368 K.
Conclusion
pure MnIIIAlPO-5 catalyst at 1023 K for 24 h in air: the re-
sulting material was X-ray amorphous.
Is the open-structure solid host for the catalytically active
site essential?: To answer this question we have compared
the performance, for a given conversion (in the oxidation of
4- and 3-picoline), using catalysts of identical composition,
one being crystalline and the other amorphous so that the
porosity of the microcrystalline parent is essentially de-
stroyed. This was achieved by heating the crystalline, phase-
The results for the oxidation of 4- and 3-picoline with
these catalysts are shown in Table 5 (Scheme 1). In the case
of 4-picoline, the overall conversion falls by an order of
magnitude (ca. 2.5 times), but, more significantly, the selec-
tivity for the target (desired) product (isonicotinic acid) was
greatly diminished. Interestingly, substantial amounts
(almost tenfold) of 4-picoline N-oxide and isonicotinic acid
Table 5. The influence of microporosity, crystallinity, and redox-active site on the activity and selectivity, in the oxidation of 3- and 4-picoline using
APB.[a]
Oxidant
Substrate
Catalyst
T
[K]
Conv
[mol%]
Product selectivity [mol%]
G
2
3
4
others
APB (solid)
APB (solid)
4-picoline
4-picoline
microporous Mn0.10Al0.90PO4
amorphous Mn0.10Al0.90PO4
368
368
72.0
25.9
92.0
16.0
3.0
35.8
–
40.5
4.8
7.9
2A
3A
4A
others
APB (solid)
APB (solid)
3-picoline
3-picoline
microporous Mn0.10Al0.90PO4
amorphous Mn0.10Al0.90PO4
338
338
37.0
7.4
78.5
–
21.4
41.5
–
50.0
–
8.7
[a]Reaction conditions: 4-picoline ( ffi2.8 g), 3-picoline (ffi2.8 g), catalyst (ffi0.30 g), adamantane (internal standard; ffi0.5 g), t=4 h; for more details see
the footnotes in Table 1. [b]For products see Table 1 and Scheme 1.
2346
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2340 – 2348

Heterogeneous Catalysis
FULL PAPER
mation. It also minimizes the contact time of the oxidant with the catalyst
thereby enhancing the “peroxide efficiency” and restricting the rapid de-
compostion of the peroxide.
N-oxide were formed, whereas with the open-structure mi-
croporous host, containing the redox-active site in high oxi-
dation (MnIII) state (15–17), 92% selectivity for the desired
isonicotinic acid was maintained. In the case of 3-picoline,
the results were even more pronounced: there was almost a
fivefold decrease in the conversion and virtually no nicotinic
acid was produced.
In summary, we have demonstrated that niacin (Vitamin
B3), quinoline-, pyridazine-, pyrazine-, and other heterocy-
clic carboxylic acids can be produced in a single-step, envi-
ronmentally benign fashion by combining single-site, open-
structure, heterogeneous catalysts with a solid source of
active oxygen, acetyl peroxyborate (APB), in the absence of
an organic solvent. The crystallinity and microporosity of
the MnIII-framework-substituted aluminophosphate catalyst
plays a major role in influencing the shape-selectivity and
regiospecificity of the reaction, when APB is used in combi-
nation with this catalyst. The high activities, selectivities and
the relatively mild conditions employed with this single-site
heterogeneous catalyst, coupled with ease of transport, stor-
age, and stability of the solid oxidant, augurs well for the
future use of APB in conjunction with other single-site cata-
lysts for fine-chemical, pharmaceutical, and agrochemical
applications.
At the end of the reaction, the heating was turned off and the contents
of the reactor were cooled (quenched). A mass-balance calculation was
performed at this stage to check for handling and mass losses. Where ki-
netic and rate effects were studied, a mini-robot liquid sampling valve
was employed to remove small aliquots (0.1 mL) of the sample (aqueous
and organic phases) during the course of the reaction. The products were
analyzed either online [by using a robotically-controlled unit with an
online computer-controlled system which is linked to a gas chromatogra-
phy (GC) and/or liquid chromatography/mass spectrometry (LCMS) ap-
paratus]or offline (with a suitable internal standard) by GC (Varian,
Model 3400 CX) employing a HP-1 capillary column (25 m”0.32 mm)
and flame ionisation detector using a variable-ramp-temperature pro-
gram (from 65 to 3008C). The identities of the products were first con-
firmed by using authenticated standards and their individual response
factors were determined from a suitable internal standard (adamantane)
by the calibration method. The overall yields were normalized with re-
spect to the (GC) response factors obtained as above and the conversions
and selectivities were determined by Equations (2) and (3):
ꢀ
ꢁ
ðmol initial substrateÞÀðmol residual substrateÞ
Conv: % ¼
ꢁ 100 ð2Þ
mol initial substrate
ꢂ
ꢃ
mol individual product
mol total products
Sel: % ¼
ꢁ 100
ð3Þ
For the internal standard GC method, the response factor (RF) and
mol% of individual products were calculated from Equations (4) and (5).
Experimental Section
APB was freshly prepared according to US 5462692.[8] Briefly, concen-
trated (24.5%) peracetic acid (40 g) was reacted with of partially dehy-
drated borax (Na2B4O7·xH2O 0<x<1; 20 g) for 3 h at 358C to form a
paste. The resulting paste was washed twice on a suction filter with etha-
nol (50 mL) and subsequently dried for 2.5 h at 508C in a circulating air
drying chamber to yield the white powdered product.
ꢂ
ꢃ ꢂ
ꢃ
mol product
mol standard
area standard
area product
RF ¼
ð4Þ
ꢂ
ꢃ ꢂ
ꢃ
RF ꢁ mol product
area product
area standard
The catalytic reactions were carried out in a stainless-steel catalytic reac-
tor (100 mL, Parr) lined with poly ether ether ketone (PEEK). The sub-
strates (4-picoline, 3-picoline, 4-methylquinoline, 4-methylpyrimidine, 4-
methylpyridazine, and methylpyrazine), a suitable internal standard (ada-
mantane), and the catalyst (MnIIIAlPO-5) were then introduced into the
reactor, which was subsequently sealed. The reactor and the inlet and
outlet ports were purged thrice with dry nitrogen prior to reaction. The
contents were stirred at 1200 rpm and the reactor was heated to the de-
sired temperature under autogeneous pressure (N2).
mol % product ¼
ꢁ 100
ð5Þ
mol sample
The identity of the products was further confirmed by using LCMS (Shi-
madzu LCMS-QP8000), which was again employed either online or off-
line. Hot filtration experiments and inductively coupled plasma (ICP)
measurements of the aqueous and organic mixtures were independently
(and regularly) carried out to rule out the possibility of leaching. In most
cases, the catalysts have been re-used three times without appreciable
loss in catalytic activity or selectivity. Further, experiments analogous
those reported earlier,[12,13] were carried out to rule out the possibility of
leaching, and analysis of the resulting filtrate at the end of reaction by
ICP and AAS revealed only trace amounts (<5 ppb) of dissolved metal
ions (Mn, Cr).
The oxidants used in our experiments one of the following:
1) APB (comprising 3.49 g of solid APB, prepared according to US
5462692,[8] and shown by titration studies to liberate 0.701 g of per-
oxyacetic acid and 0.045 g of hydrogen peroxide immediately upon
dissolution).
2) Peroxyacetic acid (comprising 4.2 g of 25% peroxyacetic acid solu-
tion in acetic acid).
3) Peroxyacetic acid+Neoborꢁ (comprising 4.2 g of 25% peroxyacetic
acid solution in acetic acid+1 g Neoborꢁ +1 g NaOH)
4) Peroxyacetic acid+NaOAc (comprising 4.2 g of 25% peroxyacetic
acid solution in acetic acid+0.934 g sodium acetate trihydrate+1 g
NaOH).
Acknowledgement
We wish to thank Borax Europe Limited (Rio Tinto Minerals) for finan-
cial support.
In each case, the oxidant was dissolved in double-distilled water (20.5 g)
and the resulting solution was fed over the course of the reaction, em-
ploying a syringe pump (Harvard “33”) to the stirred contents of the re-
actor. The slow addition of the oxidant to the catalyst/substrate mixture
improves the overall selectivity and also helps minimize byproduct for-
[1]J. M. Thomas, R. Raja, D. W. Lewis, Angew. Chem. 2005, 117, 6614–
6641; Angew. Chem. Int. Ed. 2005, 44, 6456–6482.
[2]M. Hatanaka, N. Tanaka (Nissan Chemical Ind. Ltd.), WO 9305022,
1993.
Chem. Eur. J. 2008, 14, 2340 – 2348
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2347

J. M. Thomas, R. Raja et al.
[3]J. M. Thomas, R. Raja, G. Sankar, R. G. Bell, Nature 1999, 398, 227–
230.
[15]J. M. Thomas, Angew. Chem. 1999, 111, 3800–3843; Angew. Chem.
Int. Ed. 1999, 38, 3588–3628.
[16]J. M. Thomas, R. Raja, Chem. Commun. 2001, 675–687.
[17]J. M. Thomas, R. Raja, Proc. Natl. Acad. Sci. USA 2005, 102, 13732–
13736.
[18]B. Notari, Adv. Catal. 1996, 41, 253–334.
[19]M. Taramasso, G. Perego, B. Notari (Snam Progetti), US4410501,
1983.
[20]K. Brandenburg, DIAMOND, Version 3.1d, Crystal Impact GbR,
Bonn (Germany), 2006.
[21]C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock, G. P. Shields,
R. Taylor, M. Towler, J. van De Streek, J. Appl. Crystallogr. 2006,
39, 453–457.
[4]J. M. Thomas, R. Raja, G. Sankar, R. G. Bell, D. W. Lewis, Pure
Appl. Chem. 2001, 73, 1087–1101.
[5]J. M. Thomas, R. Raja, Annu. Rev. Mater. Res. 2005, 35, 315–350.
[6]J. M. Thomas, R. Raja, Solid State Sciences, 2006, 8, 326–331.
[7]M. Anpo, J. M. Thomas, Chem. Commun. 2006, 3273–3278.
[8]R. Roesler, S. Schelle, M. Gnann, W. Zeiss (Peroxid Chemie
GmbH), US 5462692, 1995.
[9]R. Raja, J. M. Thomas, M. Xu, K. D. M. Harris, M. Greenhill-
Hooper, K. Quill, Chem. Commun. 2006, 448–450.
[10]M. Dugal, G. Sankar, R. Raja, J. M. Thomas, Angew. Chem. 2000,
112, 2399–2402; Angew. Chem. Int. Ed. 2000, 39, 2310–2313.
[11]M. Greenhill-Hooper, R. Raja, J. M. Thomas (US Borax Inc.), Brit-
ish Patent Application No. 0423586.7, 2004.
[22]P. Walters, M. Stahl, BABEL Version 1.3, 1996.
[23]E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch, D. M.
Greenblatt, E. C. Meng, T. E. Ferrin, J. Comput. Chem. 2004, 25,
1605–1612.
[12]M. Greenhill-Hooper, R. Raja, J. M. Thomas (US Borax Inc.), WO
Patent Application No. 2006043075 A1, 2006.
[13]R. Raja, J. M. Thomas, M. Greenhill-Hooper, V. Doukova, Chem.
Commun. 2007, 1924–1926.
[24]C. J. Cason, POV-RAY for Windows, Version 3.5, 2002.
Received: October 24, 2007
Published online: January 28, 2008
[14]J. H. Yu, R. R. Xu, Chem. Soc. Rev. 2006, 35 593–604.
2348
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2340 – 2348