Formation of aromatic amidoximes with hydroxylamine using microreactor technology

Article abstract of DOI:10.1021/op300113w

In this contribution, the application of microreactor technology for amidoxime formation is demonstrated, with different aromatic nitriles reacted with a 50% solution of hydroxylamine (HA) under homogeneous conditions to reach full conversion. All reactio

Full text of DOI:10.1021/op300113w

Article
pubs.acs.org/OPRD
Formation of Aromatic Amidoximes with Hydroxylamine using
Microreactor Technology
Attila Voros,*,†,‡ Zoltan Baan,† Peter Mizsey,‡ and Zoltan Finta†
́
́
́
́
̈
̈
†Sanofi R&D, Hungary, 1045 Budapest, To
́
u. 1-5
‡Department of Chemical and Environmental Process Engineering, Budapest University of Technology and Economics, Hungary,
1111 Budapest, Budafoki ut 8
́
ABSTRACT: In this contribution, the application of microreactor technology for amidoxime formation is demonstrated, with
different aromatic nitriles reacted with a 50% solution of hydroxylamine (HA) under homogeneous conditions to reach full
conversion. All reactions were executed in the presence of an excess of HA at an elevated temperature near to Tonset of HA, but in
safe mode. Upscaling was performed with success by the adaptation of an increased volume microreactor and by numbering-up.
well-studied core structural unit of various muscarinic agonists,7
INTRODUCTION
■
benzodiazepine receptor partial agonists,8 dopamine trans-
Microflow Technology. In the past decade, microreactor
porters,9 antirhinovirals,10 a growth hormone secretagogue,11
technology has gained in importance in the field of
and 5-HT agonists,12 as well as an urea bioisostere in β3-
pharmaceutical and fine chemical products due to its attractive
adrenergic receptor agonists.13 Among oxadiazoles, 1,2,4-
properties compared to those of batch technology. Generally,
oxadiazole derivatives have gained more importance in
this technology has highly efficient mixing properties combined
medicinal chemistry, suffice it to mention prenoxdiazole,
with superior heat exchange ability, which renders this
oxolamine, proxazole, or butalamine (Scheme 1). Several
technology attractive. Its high surface to volume ratio allows
for running highly exothermic and hazardous reactions in a safe
manner. Microreactors are preferred to macro-, meso-, or
Scheme 1. Biologically active oxadiazoles
microstructured systems for performing reactions with the risk
of runaway. Additionally, microreactors also facilitate safe
handling of highly toxic and dangerous materials under inert
conditions.
Refined conditions after optimization at 10 μL scale can, as a
rule, be used for the production of larger quantities of material
very easily, providing multigrams or multikilograms, or even on
an industrial scale in a very smooth way. Two options of
upscaling are generally used: either by applying an increased
volume of reactor space combined with flow rate enhancement
or by multiplying the number of the same type and volume
reactors.1,2 Both options lead to fast transfer of the process to
an industrial setting when compared to the conventional scale
up of batch processes.
methods have been reported in the literature for the synthesis
of 1,2,4-oxadiazoles,14−20 with one of the most common
methods being the addition of hydroxylamine to a nitrile,
followed by a ring closure with a carboxylic acid derivative
(Scheme 2).
As the upscaling of highly exothermic reactions using
hazardous and toxic reagents remains of high concern for
chemical manufacturing from the point of safety, productivity,
and cost, the possibilities offered by microreactor technology
are also of interest for pharma chemists. This is becoming
increasingly evident as more and more key opinion leaders
publish their results achieved in microreactors and disclose
their experiences about possible applications.
Scheme 2. General preparation method for 1,2,4-oxadiazoles
In this contribution, we describe our results on aromatic
amidoxime formation reactions achieved in commercially
available microflow reactors and our observations concerning
upscaling possibilities.
Amidoximes. In pharmaceutical chemistry, amidoximes
provide one of the shortest ways to reach certain heterocycles,
such as oxadiazoles, which are well-known as bioactive
scaffolds.3−6 As a pharmacophoric scaffold, oxadiazole is a
Received: May 4, 2012
Published: October 25, 2012
© 2012 American Chemical Society
1717
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
Hydroxylamine. Hydroxylamine (HA) is a widely used
reagent in the research of semiconductors, apart from its use in
the chemical and pharmaceutical industries. HA is classified as a
corrosive substance (Class 8) by the United Nations (UN)
recommendations on the transport of dangerous goods (12th
ed.). HA itself is a crystalline material with a highly explosive
property. Its water solution above a concentration of 70% is
also very explosive.21,22
Scheme 3. Typical decomposition of hydroxylamine
To avoid heterogeneous conditions and strong heat
accumulation during the reaction (in situ HA liberation), the
application of HA in water solution (up to 50%) is increasingly
common. In this case, metal-free conditions during storage and
handling are highly recommended, too.24
For successful reactions with HA, elevated temperatures are
required in most of the cases, which are around or above the
decomposition temperature of HA measured in the presence of
metal impurities. For the full conversion of reactants, an excess
of HA is to be used, which is generally 1−4 equivalents.30,31
This, at a large scale, makes the situation more problematic in
batch mode.
The final reaction mixture usually contains a considerable
amount of HA, which needs further attention during processing
or storage before effecting any further treatment. After the
separation of the final product, the waste should be treated with
special care. Acidifying is one of the possible solutions before
incineration. All solutions that can decrease the risk of the
application of HA are welcome in the industry, on the basis of
the facts mentioned above.
HA is also considered to be toxic and from the point of view
of process safety a problematic reagent that is mainly available
in a stable salt form or, recently in water solution, as bulk
material containing stabilizing agents. Water solutions of HA
are very advantageous from a chemical engineering point of
view, but they need very cautious handling and storage due to
their thermal instability and catalytic decomposition potential,
which are well described in the literature.23 Heating onset
temperature (Tonset) and the heat of decomposition are
dependent on the concentration of the solution and can be
catalyzed by metal ions. The Tonset in DTA gold cell is around
136
5 °C,23 which is almost constant irrespective of
concentration. However, this value is much lower in stainless
steel cells (Tonset = 63 3 °C), which decreases as a function of
increasing concentration of HA between 10 and 50%.23
Nevertheless, dependence was found in all cases to be a
function of metal ions, such as iron, copper, nickel, chromium,
titanium, and manganese ions, available in the system.24 All of
these ions decrease Tonset dramatically below 100 °C with
titanium ions, for example, reducing the Tonset of HA to room
temperature.23
The heat of the decomposition of 100% pure HA is 4.3−4.6
kJ/g, the data of which is comparable to the value of TNT (5.1
kJ/g).23,25,26 Highly concentrated solutions (>70%) or HA
itself has a detonation ability, too.23 To avoid a runaway
reaction during a process, less than a 50 °C increase of adiabatic
reaction heat has to be ensured during the reaction by
definition. When applying organic solvents, approximately 20
times dilution is necessary for the reaction with HA 50%, which
is found to be very unfavorable from the point of view of
productivity.
During the chemical process, the most frequently encoun-
tered trace metal ions are the different ions of iron. Since many
reactors in the industry are made of metal (such as stainless
steel or Hastelloy) or at least contain metal parts as a potential
source of iron or rust, no metal-free conditions can be ensured.
On the basis of the above-mentioned properties of HA and
its water solution, thermal behavior of HA in 50% solution has
been extensively examined in the presence on iron ions, such as
rust.24
In order to avoid storage and handling of thermally instable
HA, salt formation is used most frequently. Chloride salt and
sulfate, even more so, bear sufficient stability for long-term
storage.27 However, amidoxime formation reactions work with
liberated HA; thus for a good yield, liberation from the salt
needs to be effected before the reaction or in situ. During salt
liberation, Tonset can be decreased by the presence and
concentration of KOH or KCl. Tonset of HA 50% solution (6
mL) in the presence of 2 mL of 8 N KOH is 46 2 °C.28 HA
solutions treated by acids are more likely to increase
decomposition heat than alkaline solutions.28 On the basis of
the decomposition equations, which are different in the
presence of acidic or alkaline solutions, considerable gas
evolution can be observed during the reaction, which increases
the pressure in the reactor (Scheme 3).29
RESULTS AND DISCUSSION
■
The following section outlines the adaptation to continuous
flow technology of aromatic amidoxime formation using HA. At
first, all model reactions were performed in batch mode under
homogeneous condition using HA in water solution, which was
followed by adaptation to microflow technology. All reactions
were performed in 50% concentrated HA water solution in
order to obtain homogeneous conditions during the reaction.
Selecting the optimal reaction solvent was an essential step to
avoid thermally unsafe HA crystal formation during the
reaction and to avoid the plugging of the microflow system.
At first, we analyzed amidoxime formation via the reaction of
benzonitrile (1a) and HA in batch mode. In this case, isopropyl
alcohol proved to be an appropriate solvent, which ensured
homogeneity during the full process. In batch mode, we
examined the effects of reaction temperature, excess of HA, and
reaction time on conversion. The analysis of the results showed
that reaction temperature and excess of HA proved to be
influencing factors of conversion (Figure 1.).
Thus, reaction time needed for full conversion is highly
influenced by the temperature and the stoichiometry of HA.
The limit of reaction temperature to be considered in this case
is Tonset temperature of HA (136 5 °C) and the boiling point
of isopropanol (82 °C).23 Increasing the proportion of HA is
also a limiting factor with a view to safety considerations and
waste handling. Since these factors have a strong effect on the
reaction, it has been expected that full conversion is achievable
through adaptation of the reaction using microflow technology.
For fast screening in microflow conditions, the Labtrix S1
reactor by Chemtrix BV (NL) was selected. This is a
commercially available microflow reactor, which allows screen-
ing and reaction optimization on the microliter scale in a very
economical way.32
Factors selected for screening were the same as in batch
mode, but we defined values closer to the critical region (this
raises safety concerns). Maximum reaction temperature was set
slightly below Tonset 136 5 °C of HA, and we also applied a
greater excess of HA as a maximum value.23 As the reactive
1718
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
The coefficient and contour plots of the results showed that
all factors had a positive effect on the conversion within the
residence time range we selected (Figures 2 and 3.).
Figure 1. Reaction of benzonitrile (1a) in batch.
mass (volume) we worked with could only be at the microliter
scale in a microreactor, which was supported by very effective
heat exchange, no risk of runaway was predicted in the flow
system. Furthermore, the use of glass and polymeric wetted
parts meant that no metal contamination could be foreseen;
thus, no risk of Tonset decrease was predicted.
For precise screening, we employed the Design of Experi-
ment (DOE) approach (Central Composite Design) using
factors of reaction temperature (75−125 °C), HA stoichiom-
etry (1−7 equiv), and residence time (1−9 min) (Table 1).
Figure 2. Coefficient plot of benzamidoxime formation.
The results clearly indicate that more than one factor setting
is possible to achieve full conversion.
The two extreme values are as follows:
(1) A highly elevated temperature (125 °C) with moderate
excess (4 equiv) of HA can be chosen if the compound
has no temperature sensitivity.
(2) A reduced temperature (100 °C) with greater excess (7
equiv) of HA can be chosen if there is certain
temperature sensitivity.
Table 1. DOE plan and results of benzamidoxime formation
(1a) in Labtrix S1
HPLC (relative
area %)
Alternatively, we can also use other combinations along the
value chain between the extreme values, as required.
To extend the validity of our model, benzonitrile derivatives
with groups containing different electron-donating properties
were selected for testing, which was followed by some
heteroaromatic derivatives as illustrated in Scheme 4 and
Table 2.
In order to achieve homogeneous conditions in batch mode,
the first step in all cases was the selection of the solvent, which
was followed by continuous mode processes. During microflow
experiments, we used the two extreme values for testing, based
on the reaction conditions previously determined for
benzonitrile (1a) (Table 2).
The comparison of the conversion data shows that
benzonitrile derivatives with electron-withdrawing substituents
did not react at unanimous reaction rates. Fast reactions were
observed in the case of para-halogeno substituted benzonitriles,
irrespective of the type of halogen involved. For ortho
derivatives and for dihalo derivatives, a 5-min reaction time
was not sufficient for full conversion (except R = F), a
phenomenon which suggests a steric substituent effect.
temperature
HA exess
(equiv)
residence time
(min)
(°C)
amidoxime nitrile
75
125
100
75
1
1
1
1
1
4
4
4
4
4
4
4
7
7
7
7
7
1
1
5
9
9
1
5
5
5
5
5
9
1
1
5
9
9
14
32
86
68
40
52
22
44
3
60
48
125
100
100
75
78
56
97
74
26
12
0
100
125
100
100
75
88
100
91
9
98
2
37
63
11
0
125
100
75
89
100
96
4
125
100
0
1719
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
Figure 3. Contour plot of benzamidoxime formation.
Scheme 4. Reaction of nitrile derivatives with hydroxylamine
Scheme 5. Reaction of pyridine derivatives with
hydroxylamine
Electron-donating groups also had a negative effect on
conversion time under the same conditions. These observations
are in line with reactions performed in batch mode.
In the case of pyridine derivatives (Scheme 5, Table 3),
reactions were very fast; probably less than 5 min is enough for
full conversion.
Table 3. Results of pyridine derivatives (5 min residence
time, 20 bar)
HPLC (relative area HPLC (relative area
pyridine derivatives
X1 = X2 =
%) 100 °C, 7×
%) 125 °C, 4×
solvent
amidoxime nitrile amidoxime nitrile
In order to complete conversion, the best approach was to
extend residence time under the same condition as before.
Further elevation of temperature (above 125 °C, Tonset = 136
5 °C), or further excess of HA can very easily cause safety
concerns.
The application of 20 min as residence time instead of 5 min
drives all reactions to full conversion without any byproduct
formation (Table 4). In certain examples (Tables 2, 4, 5, 7), we
prepared products by quenching the final reaction mixture. In
all cases, yields were in line with the HPLC results of the
reaction mixtures. This result is appropriate for analyzing
3a
3b
H
N
N
H
MeOH
MeOH
100
100
0
0
100
100
0
0
upscaling possibilities in a microflow system, which constitutes
the next step of our study.
Scalability. We examined the upscaling possibilities of the
reactions, applying the rapid scalability theory of microflow
technologies. For this reason, a Corning system was selected,
which is also a commercially available flow system with larger
a
Table 2. Results using nitrile derivatives
benzonitrile derivatives
R =
HPLC (relative area %) 100 °C, 7×
HPLC (relative area %) 125 °C, 4×
yield (%)
solvent
amidoxime
nitrile
amidoxime
nitrile
amidoxime
1a
H
IPA
100
100
100
31
0
0
100
100
100
28
0
0
90
91
−
−
−
−
−
−
−
−
−
1b
1c
1d
1e
1f
4-F
IPA
4-Cl
EtOH
IPA
0
0
2-Cl
69
69
0
72
64
0
2,4-Cl
4-Cl-2-F
2-Cl-4-F
4-Me
2-Me
4-MeO
2-MeO
IPA
31
36
MeOH
MeOH
EtOH
BuOH
IPA
100
67
100
66
1g
1h
1i
33
15
69
27
52
34
41
77
48
54
85
59
31
23
1j
73
52
1k
IPA
48
46
a
5 min residence time, 20 bar.
1720
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
a
Table 4. Results of nitrile derivatives
HPLC (relative area %)
nitrile derivatives
R =
100 °C, 7×, 20 min
yield (%)
solvent
amidoxime
nitrile
amidoxime
1d
1e
1g
1h
1i
2-Cl
IPA
100
100
100
100
100
100
100
0
0
0
0
0
0
0
−
−
−
90
89
−
2,4-Cl
2-Cl-4-F
4-Me
IPA
MeOH
EtOH
BuOH
IPA
2-Me
1j
4-MeO
2-MeO
1k
IPA
−
a
Figure 4. Labtrix S1 microreactor (10 μL)30 A: nitrile compound in
20 min residence time, 20 bar.
solvent, B: HA, C: quenching by solvent.
volumes than Labtrix (but below the microflow region), and
allows parallelization as an additional option.
The Corning Low Flow microreactor features reaction plates
made of glass, which was important for constantly providing
metal contamination-free conditions. A further advantage of
this Corning system is the upscaling possibility of the system to
industrial level with no change in the critical parameters of
scalability such as geometrical similarity of reaction plates.
The Labtrix S1 microflow system used for screening had a 10
μL volume reaction microreactor, equipped with a T mixing
and Peltier-based temperature control system (Figure 4.
For comparison, the Corning Low Flow system was set with
reactor plates at 0.45 mL scale/plate (see Figure 5). This
system can be used with, at a maximum, nine plates at a time
and is equipped with T mixing with a jacketed heat exchange.
From the point of view of upscaling, 45-fold scale-up was
performed by the definition of reactor plate, and one Corning
plate was used. Since Labtrix S1 and the Corning plates have
different configurations, no geometrical dependence was
established during the adaptation of the process to the Corning
system. In similar cases, mixing and dispersion abilities
influenced the reaction, so we predicted further optimization
would be required.
Figure 5. Corning reactorplates.33,34
The same results were obtained as previously, so no further
optimization was necessary to reach full conversion. The
results’ independence from the applied microflow systems
suggests that the amidoxime formation reaction can easily be
adapted in different microflow systems. Furthermore, the
geometrical differences do not seem to be a critical parameter
of scalability in the case of the flow rates we used.
The second upscaling approach was the numbering up of the
plates in order to increase productivity. By setting up four
Corning plates (0.45 mL/plate) in parallel, productivity could
be increased by 4 times (Figure 6).
All the reactions of the previously examined aromatic nitrile
derivatives were also performed in the Corning system under
the same conditions as the ones optimal for the Labtrix system
(Tables 5 and 6).
The conditions were the same as previously, without any
further refinement (Table 7).
a
Table 5. Results in Corning LF reactor I
benzonitrile derivatives
HPLC (relative area %)
yield (%)
R =
solvent
temperature (°C)
HA excess (equiv)
residence time (min)
amidoxime
nitrile
amidoxime
1a
1a
1b
1b
1c
1c
1d
1e
1f
H
H
IPA
100
125
100
125
100
125
100
100
100
125
100
100
100
100
100
7
4
7
4
7
4
7
7
7
4
7
7
7
7
7
5
5
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
−
−
−
−
−
−
−
−
−
−
−
93
97
−
−
IPA
4-F
IPA
5
4-F
IPA
5
4-Cl
EtOH
EtOH
IPA
5
4-Cl
5
2-Cl
20
20
5
2,4-Cl
4-Cl-2-F
4-Cl-2-F
2-Cl-4-F
4-Me
IPA
MeOH
MeOH
MeOH
EtOH
BuOH
IPA
1f
5
1g
1h
1i
20
20
20
20
20
2-Me
1j
4-MeO
2-MeO
1k
IPA
a
0.45 mL reactor volume, 2.5−3 bar.
1721
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
a
Table 6. Results in Corning LF reactor II
pyridine derivatives
HPLC (relative area %)
X1 =
X2 =
solvent
temperature (°C)
HA exess (equiv)
residence time (min)
amidoxime
nitrile
1l
H
H
N
N
N
N
H
H
MeOH
MeOH
MeOH
MeOH
100
125
100
125
7
4
7
4
5
5
5
5
100
100
100
100
0
0
0
0
1l
1m
1m
a
0.45 mL reactor volume, 2.5−3 bar.
conditions using commercially available HA water solution.
Metal-free conditions were easily ensured in order to avoid
runaway decomposition of reaction mixtures at reaction
temperature.
DESCRIPTION OF THE EXPERIMENTS
■
All chemicals were purchased from Sigma-Aldrich except for 4-
chloro-2-fluorobenzonitrile, which was bought from Acros
Organics. 4-Chlorobenzonitrile and 2,4-dichlorobenzonitrile
were purchased from Merck, and the 50% water solution of
hydroxylamine was purchased from BASF.
All the reactions were tracked using appropriate HPLC
methods on Merck Hitachi Elite Lachrom equipment, column
temperature was 30 °C in each of the four methods in Table 8.
1H and 13C NMR spectra were obtained at room temperature
as solutions in DMSO using tetramethylsilane as an internal
standard. The spectra were recorded using a Bruker Avance 200
MHz spectrometer.
The HRMS spectra were recorded on a Micromass Q-TOF
Premier spectrometer.
FT IR spectra were recorded on the solid state as KBr
dispersion using PerkinElmer Spectrum 100 FT-IR spectropho-
tometer.
Figure 6. Corning LF reactor with four plates.31 (A) Nitrile compound
in solvent. (B) HA, MP: mixing plate. BPR: back pressure regulator.
The selected three model reactions gave the same
conversions as the conversion previously observed in the case
of one plate. In all the cases, final reaction mixtures were
quenched with icy HCl solution after the reaction mixtures had
been collected. The precipitated product was easily separated
from the residues. The mother liquor was contaminated by the
remaining part of the HA.HCl salt, which was found safe
enough for further handling, such as incineration.
Benzamidoxime 2a. 6.00 mL of 1a (6.00 g, 58 mmol) was
heated in 60 mL of isopropyl alcohol to boiling, and 25.2 mL (7
equiv) of 50% water solution of HA was added dropwise. The
reaction was stirred under reflux to full conversion and checked
by HPLC (method No.1, 215 nm, RT: 1a: 3.5, 2a: 1.2). The
reaction mixture was added to iced water, the product was
filtered off, washed with water (2 × 5 mL), and dried for
analysis. Analytical data of 2a was identical to that described in
the literature.35
4-Fluoro-benzamidoxime 2b. 0.50 g of 1b (4.13 mmol)
in 5 mL isopropyl alcohol was heated to boiling and 1.00 mL
(0.55 g, 16.52 mmol) of 50% water solution of HA was added
dropwise. The reaction was stirred under reflux to full
conversion and checked by HPLC (method No.1, 230 nm,
RT: 1b: 1.1, 2b: 0.6). The reaction mixture was added to icy
water, the product was filtered off, washed with water (2 × 5
mL), and dried for analysis. Analytical data of 2b was identical
to that reported in the literature.35
CONCLUSION
■
Amidoxime formation with HA can be safely performed in
commercially available microflow reactors. This technique has
numerous advantages compared to batch reactions, including
safe reaction management, full conversion with a short reaction
time and effective increase in productivity without an upscaling
effect. Results from selected models showed a wide range of
application possibilities for aromatic nitriles, with no limitation
for sterically hindered or electron rich aromatic nitriles. The
method is also useful for reactions under homogeneous
Table 7. Results of the parallel Corning LF reactor (4 × 0.45 mL, 2.5−3 bar)
benzonitrile
compound
HPLC (relative area %)
yield (%)
R =
solvent
temperature (°C)
HA excess (equiv)
residence time (min)
amidoxime
nitrile
amidoxime
1a
1a
1b
1b
1i
H
IPA
100
125
100
125
100
7
4
7
4
7
5
5
100
100
100
100
100
0
0
0
0
0
97
−
96
H
IPA
4-F
4-F
2-Me
IPA
5
IPA
5
−
96
BuOH
20
1722
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
Table 8. HPLC methods
method
no.
flow rate
(ml/min)
column
Chromolit
Performance RP-
18e
parameters
buffer
eluent
gradient
1
4.6 × 100 mm,
2 μm
25 mmol NaH2PO4/
Na2HPO4
MeOH
3
10% MeOH, 1 min 10 → 20% MeOH, 1 min 20% MeOH, 3 min
20 → 10% MeOH, 1 min 10% MeOH, 3 min
2
3
4
Chromolit
Performance RP-
18e
4.6 × 100 mm,
2 μm
10 mmol KH2PO4/
H3PO4 (+5% MeCN)
MeOH
MeCN
MeCN
3
30 → 70% MeOH, 10 min 70 → 30% MeOH, 0.5 min 30% MeOH,
1.5 min
Chromolit
Performance RP-
18e
4.6 × 100 mm,
2 μm
10 mmol KH2PO4/
H3PO4 (+5% MeCN)
3
30 → 70% MeCN, 10 min 70 → 30% MeCN, 0.5 min 30% MeCN,
1.5 min
Xbridge C18
2.1 × 100 mm,
3.5 μm
10 mmol KH2PO4/
H3PO4 (+5% MeCN)
0.4
10% MeCN, 1 min 10 → 70% MeCN, 9 min 70 → 10% MeCN, 1
min 10% MeCN, 2 min
4-Chloro-benzamidoxime 2c. 0.50 g of 1c (3.64 mmol)
was heated in 8 mL of ethyl alcohol to boiling, and 0.89 mL
(0.96 g, 14.56 mmol)of water solution of HA was added
dropwise. The reaction was stirred under reflux to full
conversion and checked by HPLC (method No.3, 205 nm,
RT: 1c: 1.6, 2c: 0.7). The reaction mixture was added to icy
water, the product was filtered off, washed with water (2 × 5
mL), and dried for analysis. All analytical data of 2c was
identical to that described in the literature.36
2-Chloro-benzamidoxime 2d. 0.50 g of 1d (3.64 mmol)
was heated in 8 mL of ethyl alcohol to boiling, and 0.89 mL
(0.96 g, 14.56 mmol) of water solution of HA was added
dropwise. The reaction was stirred under reflux to full
conversion and checked by HPLC (method No.3, 205 nm,
RT: 1d: 1.8, 2d: 0.6). The reaction mixture was added to icy
water, the product was filtered off, washed with water (2 × 5
mL), and dried for analysis. All other analytical data of 2d are
identical to the ones described in the literature.37
2,4-Dichloro-benzamidoxime 2e. 1.00 g of 1e (5.81
mmol) was heated in 30 mL of isopropyl alcohol to boiling, and
1.42 mL (1.54 g, 23.24 mmol) of water solution of HA was
added drop by drop. The reaction was stirred under reflux to
full conversion and checked by HPLC (method No.3, 230 nm,
RT: 1e: 2.4, 2e: 0.7). The reaction mixture was added to icy
water, the product was filtered off, washed with water (2 × 5
mL), and dried for analysis. All other analytical data of 2e are
identical to the ones described in the literature.38
RT: 1g: 1.7, 2g: 0.7). The reaction mixture was added to icy
water, the product was filtered off, washed with water (2 × 5
mL), and dried for analysis. H NMR (DMSO-d6) δ: 9.48 (s,
1
1H), 7.44 (m, 2H), 7.23 (m, 1H), 5.83 (s, 2H). 13C NMR
(DMSO-d6) δ: 114.4 (d, J = 21.4 Hz, CH(5)), 117.1 (d, J =
25.1 Hz, CH(3)), 130.7 (d, J = 3.3 Hz, C(1)), 133.0 (d, J = 9.2
Hz, CH(6)), 133.7 (d, J = 11.1 Hz, C−Cl(2)), 150.2 (s, C(C-
amidoxime)), 162.2 (d, J = 248.8 Hz, C−F(4)). IR (KBr,
cm−1): 3489, 3365, 3061, 2821, 1662, 1588, 1511, 1404, 1373,
1270, 1220, 1045, 958, 946, 907, 826, 680. HRMS (ESI): Calcd
for C7H6N2OClF (M+ + H) 189.0231, Found 189.0231.
HPLC−MS: (m/z) (MH+) = 189.
4-Methylbenzamidoxime 2h. 0.76 g of 1h (6.52 mmol)
was heated in 15 mL of ethyl alcohol to boiling, and 1.60 mL
(1.72 g, 26.08 mmol) of water solution of HA was added drop
by drop. The reaction was stirred under reflux to full conversion
and checked by HPLC (method No.3, 230 nm, RT: 1h: 0.6,
2h: 1.1). The reaction mixture was added to icy water, the
product was filtered off, washed with water (2 × 5 mL), and
dried for analysis. All other analytical data of 2e are identical to
the ones described in the literature.39
2-Methylbenzamidoxime 2i. 0.327 mL of 1i (0.50 g, 4.27
mmol) was heated in 15 mL of butyl alcohol to boiling, and
1.05 mL (1.13 g, 17.08 mmol) of water solution of HA was
added drop by drop. The reaction was stirred under reflux to
full conversion and checked by HPLC (method No.3, 205 nm,
RT: 1i: 1.2, 2i: 0.6) The reaction mixture was added to icy
water, the product was filtered off, washed with water (2 × 5
mL), and dried for analysis. All other analytical data of 2e are
identical to the ones described in the literature.40
4-Methoxybenzamidoxime 2j. 1.00 g of 1j (7.51 mmol)
was heated in 10 mL of isopropyl alcohol to boiling, and 1.84
mL (1.98 g, 30.04 mmol) of water solution of HA was added
drop by drop. The reaction was stirred under reflux to full
conversion and checked by HPLC (method No.4, 205 nm, RT:
1j: 10.2, 2j: 1.2). The reaction mixture was added to icy water,
the product was filtered off, washed with water (2 × 5 mL), and
dried for analysis. All other analytical data of 2e are identical to
the ones described in the literature.35
4-Chloro-2-fluoro-benzamidoxime 2f. 0.50 g of 1f (3.21
mmol) was heated in 8 mL of methyl alcohol to boiling, and
0.79 mL (0.85 g, 12.84 mmol) of water solution of HA was
added drop by drop. The reaction was stirred under reflux to
full conversion and checked by HPLC (method No.3, 205 nm,
RT: 1f: 1.8, 2f: 0.8). The reaction mixture was added to icy
water, the product was filtered off, washed with water (2 × 5
1
mL), and dried for analysis. H NMR (DMSO-d6) δ: 9.71 (s,
1H), 7.49 (m, 2H), 7.30 (m, 1H), 5.85 (s, 2H). 13C NMR
(DMSO-d6) δ: 116.9 (d, J = 26.5 Hz, CH(3)), 121.4 (d, J =
13.3 Hz, C(1)), 124.8 (d, J = 3.0 Hz, CH(6)), 131.5 (d, J = 4.1
Hz, CH(5)), 134.5 (d, J = 10.3 Hz, C−Cl(4)), 147.9 (s, C(C-
amidoxime)), 160.1 (d, J = 253.6 Hz, C−F(2)). IR (KBr,
cm−1): 3492, 3371, 1673, 1596, 1570, 1496, 1413, 1213, 1073,
951, 904, 859, 832, 719. HRMS (ESI): Calcd for C7H6N2OClF
(M+ + H) 189.0231, Found 189.0223. HPLC−MS: (m/z)
(MH+) = 189.
2-Methoxybenzamidoxime 2k. 0.915 mL of 1k (1.00 g,
7.51 mmol) was heated in 10 mL of isopropyl alcohol to
boiling, and 1.84 mL (1.98 g, 30.04 mmol) of water solution of
HA was added drop by drop. The reaction was stirred under
reflux to full conversion and checked by HPLC (method No.4,
205 nm, RT: 1k: 8.8, 2k: 1.0). The reaction mixture was added
to icy water, the product was filtered off, washed with water (2
2-Chloro-4-fluoro-benzamidoxime 2g. 0.50 g of 1g
(3.21 mmol) was heated in 8 mL of methyl alcohol to boiling,
and 0.79 mL (0.85 g, 12.84 mmol) of water solution of HA was
added drop by drop. The reaction was stirred under reflux to
full conversion and checked by HPLC (method No.3, 205 nm,
1
× 5 mL), and dried for analysis. H NMR (DMSO-d6) δ: 9.37
(s, 1H), 7.35 (m, 2H), 6.98(m, 2H), 5.58 (s, 2H). 13C NMR
(DMSO-d6) δ: 55.6 (s, CH3(methoxy)), 112.1 (s, CH(3)),
1723
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
Table 9. Flow rates in Labtrix S1
benzonitrile derivatives
flow rate (μL/min)
R =
temperature (°C)
HA excess (equiv)
residence time (min)
solvent
nitrile
HA
quench
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1b
1c
1c
1d
1d
1d
1e
1e
1e
1f
H
H
75
1
1
1
1
1
4
4
4
4
4
7
7
7
7
7
4
7
4
7
4
7
7
4
7
7
4
7
4
7
7
4
7
7
4
7
7
4
7
7
4
7
7
4
7
4
7
1
1
IPA
9.49
9.49
1.90
1.05
1.05
8.22
1.64
1.64
1.64
0.91
7.25
7.25
1.45
0.81
0.81
1.69
1.51
1.81
1.69
1.81
1.69
0.42
1.91
1.85
0.46
1.83
1.72
1.83
1.72
0.43
1.70
1.82
0.45
1.79
1.87
0.45
1.71
1.55
0.38
1.71
1.55
0.38
1.80
1.67
1.83
1.73
0.51
0.51
0.10
0.06
0.06
1.78
0.36
0.36
0.36
0.2
20
20
10
6
125
100
75
IPA
H
5
IPA
H
9
IPA
H
125
100
100
75
9
IPA
6
H
1
IPA
20
10
10
10
6
H
5
IPA
H
5
IPA
H
125
100
75
5
IPA
H
9
IPA
H
1
IPA
2.75
2.75
0.55
0.31
0.31
0.31
0.49
0.19
0.31
0.19
0.31
0.08
0.09
0.15
0.04
0.17
0.28
0.17
0.28
0.07
0.30
0.18
0.05
0.21
0.13
0.03
0.29
0.46
0.11
0.29
0.46
0.11
0.20
0.33
0.17
0.27
20
20
10
6
H
125
100
75
1
IPA
H
5
IPA
H
9
IPA
H
125
125
100
125
100
125
100
100
125
100
100
125
100
125
100
100
125
100
100
125
100
100
125
100
100
125
100
100
125
100
125
100
9
IPA
6
4-F
5
IPA
10
10
10
10
10
10
3
4-F
5
IPA
4-Cl
4-Cl
2-Cl
2-Cl
2-Cl
2,4-Cl
2,4-Cl
2,4-Cl
4-Cl-2-F
4-Cl-2-F
2-Cl-4-F
2-Cl-4-F
2-Cl-4-F
4-Me
4-Me
4-Me
2-Me
2-Me
2-Me
4-MeO
4-MeO
4-MeO
2-MeO
2-MeO
2-MeO
4-Py
4-Py
2-Py
2-Py
5
EtOH
EtOH
IPA
5
5
5
IPA
20
5
IPA
IPA
10
10
3
5
IPA
20
5
IPA
MeOH
MeOH
MeOH
MeOH
MeOH
EtOH
EtOH
EtOH
BuOH
BuOH
BuOH
IPA
10
10
10
10
3
1f
5
1g
1g
1g
1h
1h
1h
1i
5
5
20
5
10
10
3
5
20
5
10
10
3
1i
5
1i
20
5
1j
10
10
3
1j
5
IPA
1j
20
5
IPA
1k
1k
1k
3a
3a
3b
3b
IPA
10
10
3
5
IPA
20
5
IPA
MeOH
MeOH
MeOH
MeOH
10
10
10
10
5
5
5
120.5 (s, CH(5)), 122.9 (s, CH(2)), 129.9 (s, CH(4)), 130.5
(s, CH(6)), 151.2 (s, C(1)), 157.4 (s, C(C-amidoxime)). IR
(KBr, cm−1): 3470, 3358, 2939, 1675, 1596, 1500, 1463, 1437,
1385, 1282, 1246, 1017, 930, 835, 781, 758. HRMS (ESI):
Calcd for C7H6N2O2 (M+ + H) 167.0785, Found 167.0782.
HPLC-MS: (m/z) (MH+) = 167.
4-Pyridine-benzamidoxime 4a. 0.50 g 3a (4.71 mmol)
was heated in 10 mL of methyl alcohol to boiling, and 1.15 mL
(1.24 g, 18.84 mmol) of water solution of HA was added drop
by drop. The reaction was stirred under reflux to full conversion
and checked by HPLC (method No.2, 230 nm, RT: 3a: 1.0, 4a:
0.7). The reaction mixture was added to icy water, the product
was filtered off, washed with water (2 × 5 mL), and dried for
analysis. All other analytical data of 2e are identical to the ones
described in the literature.40
2-Pyridine-benzamidoxime 4b. 0.40 g 3b (3.84 mmol)
was heated in 9 mL of methyl alcohol to boiling and 0.94 mL
(1.01 g, 15.36 mmol) of water solution of HA was added drop
by drop. The reaction was stirred under reflux to full conversion
and checked by HPLC (method No.2, 210 nm, RT: 3b: 0.9, 4b:
0.6). The reaction mixture was added to icy water, the product
was filtered off, washed with water (2 × 5 mL), and dried for
analysis. All other analytical data of 2e are identical to the ones
described in the literature.40
1724
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
Table 10. Flow rates in Corning LF
nitrile derivatives
R =
temperature (°C)
HA excess (equiv)
residence time (min)
solvent
nitrile (μL/min)
HA (μL/min)
1a
1a
1b
1b
1c
1c
1d
1e
1f
H
H
100
125
100
125
125
100
100
100
125
100
100
100
100
100
100
125
100
125
100
7
4
7
4
4
7
7
7
4
7
7
7
7
7
7
4
7
4
7
5
5
IPA
65
74
68
76
81
76
19
21
82
77
19
20
20
17
17
81
75
82
78
25
16
22
14
9
IPA
4-F
5
IPA
4-F
5
IPA
4-Cl
5
EtOH
EtOH
IPA
4-Cl
5
14
4
2-Cl
20
20
5
2,4-Cl
4-Cl-2-F
4-Cl-2-F
2-Cl-4-F
4-Me
2-Me
4-MeO
2-MeO
4-Py
IPA
2
MeOH
MeOH
MeOH
EtOH
BuOH
IPA
8
1f
5
13
3
1g
1h
1i
20
20
20
20
20
5
2
1
1j
5
1k
3a
3a
3b
3b
IPA
5
MeOH
MeOH
MeOH
MeOH
9
4-Py
5
15
8
2-Py
5
2-Py
5
12
Table 11. Flow rates in Corning LF parallel
Nitrile derivatives
R =
temperature (°C)
HA excess (equiv)
residence time (min)
solvent
nitrile (μL/min)
HA (μL/min)
1a
1a
1b
1b
1i
H
100
125
100
125
100
7
4
7
4
7
5
5
IPA
261
296
272
304
80
99
64
88
56
10
H
IPA
4-F
4-F
2-Me
5
IPA
5
IPA
20
BuOH
Reactions in the Microreactor. Labtrix S1 (Chemtrix BV,
NL): Flow system was fitted with a glass microreactor (3023,
reactor volume =10 μL + 1.5 μL quench volume). Channel
width: 300 μm, channel depth: 60 μm and effective channel
length: 60.8 cm. Reactant solutions were introduced into the
reactor through three 1 mL gastight syringes (SGE, UK)
capable of delivering three solutions at flow rates between 0.1
and 25 μL/min. The system was maintained at 20 bar of back
pressure by means of a preset ultralow dead-volume back-
pressure regulator (Upchurch Scientific, USA) in order to
prevent the reactants and solvent system from boiling when
temperatures above the atmospheric boiling point were
employed. The system was controlled through the Labtrix S1
software, which enables the control of reactant flow rate (total
flow rate ≤80 μL/min, reactant residence time (7.5 s to 50 min
(for a 10 μL reactor)), reactor temperature (−15 to 195 °C),
equilibration time, and sample collection into one of 29 2-mL
sample vials. The software also archives system parameters such
as the set and actual temperatures, system pressure, reactor
type, and the programmed flow rates along with the sample
collection time and vessel, all of which can be reviewed both
during and after the experiment.41 Each respective solution of
benzonitrile derivative (1a−1k, 3a and 3b) was pumped into
the reactor from inlet A, the 50% water solution of HA was
introduced from inlet B, and the solvent was introduced as a
diluent from inlet C. After the system volume had passed
through the reactor three times, the reaction reached a steady
state, and a sample was then collected and analyzed offline by
HPLC (Table 9).
Corning LF reactor: The system was fitted with a glass
microreactor (reactor volume = 0.45 mL). Channel height:
3.34−3.66 mm, the smallest point in the microreactor: 0.4 mm
in width. Reactant solutions were introduced into the reactor
through two lines of Encynova Novasync pump (Car-May
LLC, USA) capable of delivering three solutions at flow rates
between 1 μL/min and 120 mL/min. The system was
maintained at maximum 13 bar of back pressure in order to
prevent the reactants and solvent system from boiling when
temperatures above the atmospheric boiling point were
employed. The solution of nitrile derivatives (1a−1k, 3a and
3b) was pumped into the reactor from inlet A, and the 50%
water solution of HA was introduced from inlet B. After the
system volume had passed through the reactor three times, the
reaction reached a steady state. A sample was then collected
and analyzed offline by HPLC (see Tables 10 and 11).
The DOE plan, contour plot, and coefficient plot were made
with MODDE 9.0 software.
AUTHOR INFORMATION
■
Corresponding Author
*E-mails: attila.voros@sanofi.com, vorosattila85@gmail.com.
Telephone: +36-1-505-17-56.
Notes
The authors declare no competing financial interest.
1725
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726

Organic Process Research & Development
Article
(28) Wei, C.; Saraf, S. R.; Rogers, W. J.; Mannan, S. M. Thermochim.
Acta 2004, 421, 1−9.
ACKNOWLEDGMENTS
■
This paper is dedicated to the memory of Istvan
́
Hermecz
(29) Stephenson, L.; Warbutron, W. K.; Wilson, M. J. J. Chem. Soc. C.
1969, 861−864.
(1944−2011). Special acknowledgement goes to: Las
́
zlo
́
́
Drahos, Las
́
zlon
́
́
e
́
Grosz, Arpad
́
́
Illar
́
, Ilona Koleszar
́
, Zsolt
(30) Godovikova, T. I.; Vorontsova, S. K.; Konyushkin, L. D.;
Firgang, S. I.; Rakitin, O. A. Russ. Chem. Bull. Int. Ed. 2008, 57, 2440−
2442.
(31) Lee, Wai Mun; Novel nitrile and amidoxime compounds and
methods preparation. WO/2009/058277, 2009.
(32) http://www.chemtrix.com.
Regenyi, Gab
́
or Vlad, Charlotte Wiles and Chemtrix BV for
making the use of Labtrix S1 equipment. The author expresses
his gratitude to Sanofi for contributing to his Ph.D. scholarship
and is grateful to Pro Progressio and to Jozsef Varga
́
Foundations for their financial support.
(33) http://www.corning.com.
(34) Barthe, P.; Guermeur, C.; Lobet, O.; Moreno, M.; Woehl, P.;
Roberge, D. M.; Bieler, N.; Zimmermann, B. Chem. Eng. Chem. Eng.
Technol. 2008, 31, 1146−1154.
REFERENCES
■
(1) Wiles, C.; Watts, P. Chim. Oggi 2010, 28, 3−5.
(35) Yang, X.; Li, H.; Zhang, Y.; Li, C.; Liu, B.; Liang, W.; Zhao, G.;
Liu, G.; Song, D.; Wang, R.; Jing, Y. J. Med. Chem. 2010, 53, 1015−
1022.
(2) Nieuwland, P. J.; Segers, R.; Koch, K.; van Hest, J. C. M; Rutjes,
F. P. J. T. Org. Process Res. Dev. 2011, 15, 783−787.
(3) Nicolaides, D. N.; Varella, E. A.; Patai, S., The Chemistry of Acid
Derivatives; Interscience: New York, 1992, 2, 875−966.
(4) Clapp, L.B.; Katritzky, A.R.; Boulton, A.J. Advances in Heterocyclic
Chemistry; Academic Press Inc.: New York, 1976; Vol. 20, pp 65−116.
(5) Clapp, L. B.; Potts, K. T. Comprehensive Heterocyclic Chemistry;
Pergamon Press: Oxford, 1984; Vol. 6, pp 365−392.
(6) Jochims, J. C.; Katritzky, A. R.; Rees, C. W.; Scriven, E. F. V.
Comprehensive Heterocyclic Chemistry II; Pergamon Press: Oxford,
1996; Vol. 4, pp 179−228.
(36) Holsworth, D.; Waaler, J.; Machon, O.; Krauss, S.; Golding, L.
Azole derivatives as WTN pathway inhibitors. WO/2010/139966,
2010
(37) Cottrell, D. M.; Capers, J.; Salem, M. M.; Deluca-Fradley, K.;
Croft, S. L.; Werbovetz, K. A. Bioorg. Med. Chem. 2004, 12, 2815−
2824.
(38) Cai, S. X.; Zhang, H.-Z.; Drewe, J. A.; Reddy, P. S.; Kasibhatla,
S.; Kuemmerle, J. D.; Ollis, K. P. Preparation of 3,5-diaryl-1,2,4-
oxadiazoles and analogs as activators of caspases and inducers of
apoptosis. U.S. Pat. Appl. 2003/45546, 2003
(39) Hirono, S.; Shiozawa, S.; Chaki, H.; Kotsubo, H.; Tanaka, T.;
Aikawa, Y. Preparation of benzophenone derivatives as AP-1 inhibitors
for treatment of arthritis. EP1445249, 2004
(7) Orlek, B. S.; Blaney, F. E.; Brown, F.; Clark, M. S. G.; Hadley, M.
S.; Hatcher, J.; Riley, G. J.; Rosenberg, H. E.; Wadsworth, H. J.;
Wyman, P. J. Med. Chem. 1991, 34, 2726−2735.
(8) Watjen, F.; Baker, R.; Engelstoff, M.; Herbert, R.; MacLeod, A.;
Knight, A.; Merchant, K.; Moseley, J.; Saunders, J.; Swain, C. J.; Wong,
E.; Springer, J. P. J. Med. Chem. 1989, 32, 2282−2291.
(9) Carroll, F. I.; Gray, J. L.; Abrahm, P.; Kuzemko, M. A.; Lewin, A.
H.; Boja, J. W.; Kuhar, M. J. J. Med. Chem. 1993, 36, 2886−2890.
(10) Diana, G. D.; Volkots, D. L.; Nitz, T. J.; Bailey, T. R.; Long, M.
A.; Vescio, N.; Aldous, S.; Pevear, D. C.; Dutko, F. J. J. Med. Chem.
1994, 37, 2421−2436.
(40) Lessel, J. Arch. Pharm. (Weinheim, Ger.) 1993, 326, 383−390.
(41) Wiles, C.; Watts, P. Beilstein J. Org. Chem. 2011, 7, 1360−1371.
(11) Ankersen, M.; Peschke, B.; Hansen, B. S.; Hansen, T. K. Bioorg.
Med. Chem. Lett. 1997, 7, 1293−1298.
(12) Chen, C.; Senanayake, C. H.; Bill, T. J.; Larsen, R. D.;
Verhoeven, T. R.; Reider, P. J. J. Org. Chem. 1994, 59, 3738−3741.
(13) Mathvink, R. J.; Barritta, A. M.; Candelore, M. R.; Cascieri, M.
A.; Deng, L.; Tota, L.; Strader, C. D.; Wyvratt, M. J.; Fisher, M. H.;
Weber, A. E. Bioorg. Med. Chem. Lett. 1999, 9, 1869−1874.
(14) LaMattina, J. L.; Mularski, C. J. J. Org. Chem. 1984, 49, 4800−
4805.
(15) Liang, G. B.; Qian, X. Bioorg. Med. Chem. Lett. 1999, 9, 2101−
2104.
(16) Liang, G. B.; Feng, D. D. Tetrahedron Lett. 1996, 37, 6627−
6630.
(17) Tyrkov, A. G. Khim. Khimich. Tekhnol. 2000, 43, 73−77.
(18) Young, J. R.; DeVita, R. J. Tetrahedron Lett. 1998, 39, 3931−
3934.
(19) Neidlein, R.; Sheng, L. Synth. Commun. 1995, 25, 2379−2394.
(20) Neidlein, R.; Sheng, L. J. Heterocycl. Chem. 1996, 33, 1943−
1949.
(21) The Explosion at Concept Sciences: Hazards of Hydroxylamine
U.S. Chemical Safety and Hazard Investigation Board2002, 1999-13-
C-PA.
(22) Report of Investigation Explosive Fire of Hydroxylamine at a
Chemical Plant in Gunma Prefecture, Hazardous Materials Safety
Techniques AssociationMarch 2001 (in Japanese).
(23) Iwata, Y.; Koseki, H. Process Saf. Prog. 2002, 21, 136−141.
(24) Iwata, Y.; Koseki, H. J. Hazard. Mater. 2003, 104, 39−49.
(25) Bretherick, L. Bretherick’s Handbook of Reactive Chemical
Hazards, 5th ed.; Butterworth-Heinemann: Woburn, MA, 1995.
(26) Iwata, Y.; Koseki, H.; Hosoya, F. J. Loss Prev. Process Ind. 2003,
16, 41−53.
(27) Cisneros, L. O.; Rogers, W. J.; Mannan, S. M. Thermochim. Acta
2004, 414, 177−183.
1726
dx.doi.org/10.1021/op300113w | Org. Process Res. Dev. 2012, 16, 1717−1726