Kinetic modelling of synthesis of N-aminopiperidine from hydroxylamine-o-sulfonique acid and piperidine

Article abstract of DOI:10.1134/S0023158412010041

A new route to synthesize N-aminopiperidine (NAPP) from hydroxylamine-O-sulfonique acid (HOSA) and piperidine was described. Kinetics of the reaction was investigated to optimize the conditions of the synthesis. Since the reaction is fast, this study was carried out in a diluted medium (10^-3 to 10^-2 mol/l). To determine the concentration of the reaction product, NAPP was allowed to react with formaldehyde and the product was analysed by UV and HPLC techniques. The formation of NAPP is consistent with the first-order reaction to two reagents, governed by the nucleophilic substitution via SN2 mechanism. Oxidation of NAPP by HOSA was identified as the main secondary reaction which consistently reduced the yield of NAPP. A number of differential equations were elaborated and solution of these equations serves to predict the behavior of the system as a function of the reagent concentrations, pH and temperature. From the corresponding mathematical treatment a unique implicit expression was derived that characterizes the reaction medium. It was found that the [PP] ₀/[HOSA]₀ molar ratio (p), the initial concentrations of [PP]₀ and [HOSA]₀, the ratio of rate constants k ₂/k₁ and temperature are the only parameters that affect the yield of NAPP from HOSA. The results calculated from this model are in good agreement with the experimental data and they can be used to determine the optimal conditions of the reaction. Pleiades Publishing, Ltd., 2012.

Full text of DOI:10.1134/S0023158412010041

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2012, Vol. 53, No. 1, pp. 25–35. © Pleiades Publishing, Ltd., 2012.
Kinetic Modelling of Synthesis of NꢀAminopiperidine
from HydroxylamineꢀOꢀSulfonique Acid and Piperidine1
E. Labarthe, A.ꢀJ. Bougrine, V. Pasquet, and H. Delalu
Laboratoire Hydrazines et Composés Energétiques Polyazotés, Université Claude Bernard Lyon 1, France
eꢀmail: veronique.pasquet@univꢀlyon1.fr
Received March 15, 2011
Abstract—A new route to synthesize Nꢀaminopiperidine (NAPP) from hydroxylamineꢀOꢀsulfonique acid
(HOSA) and piperidine was described. Kinetics of the reaction was investigated to optimize the conditions of
the synthesis. Since the reaction is fast, this study was carried out in a diluted medium (10–3 to 10–2 mol/l).
To determine the concentration of the reaction product, NAPP was allowed to react with formaldehyde and
the product was analysed by UV and HPLC techniques. The formation of NAPP is consistent with the firstꢀ
order reaction to two reagents, governed by the nucleophilic substitution via SN2 mechanism. Oxidation of
NAPP by HOSA was identified as the main secondary reaction which consistently reduced the yield of NAPP.
A number of differential equations were elaborated and solution of these equations serves to predict the
behavior of the system as a function of the reagent concentrations, pH and temperature. From the correꢀ
sponding mathematical treatment a unique implicit expression was derived that characterizes the reaction
medium. It was found that the [PP]0/[HOSA]0 molar ratio (p), the initial concentrations of [PP]0 and
[HOSA]0, the ratio of rate constants k2 k1 and temperature are the only parameters that affect the yield of
/
NAPP from HOSA. The results calculated from this model are in good agreement with the experimental data
and they can be used to determine the optimal conditions of the reaction.
DOI: 10.1134/S0023158412010041
1
Nꢀaminopiperidine (NAPP) is a precursor for the
Here we disclose the information on kinetic
pharmaceutical industry; it is used for plant protecꢀ parameters of the reaction HOSA with piperidine that
tion, in cosmetology and photographic chemistry. is important to optimize this process.
NAPP is used to prepare derivatives from tetrazoliꢀ
none in manufacturing weed killers. Other applicaꢀ
tions are the fabrication of paper and transparent
EXPERIMENTAL PART
recording films, the formulation of inhibitors and subꢀ
stances resistant to amine based solvents.
Reagents
Water was passed through an ionꢀexchange resin. All
the salts used (NaOH, Na2SO4, Na2SO4 10H2O, BaCl2,
Na2HPO4 and KH2PO4) were products of reagent grade
(99%) purity (Chimieꢀplus, Merck and Acros).
Samples of HydroxylamineꢀOꢀSulfonique Acid
HOSA) were of reagent (97%) grade purity supplied
by ISOCHEM and Aldrich. Piperidine (PP) and Nꢀ
aminopiperidine (NAPP) were supplied by Aldrich,
with reagent grade purities of 99 and 97% respectively.
Azopiperidine (TTZ) was produced by oxidation of
⋅
NAPP synthesis by the Raschig process has already
been optimized in our laboratory [1, 2]. It involves the
selective synthesis based on monochloramine that can
be conducted with any organic solvent at low cost and
without polluting reagents. However, monochloramꢀ
ine is unstable and must be used immediately.
(
Accordingly, a new synthesis route was introduced
with hydroxylamineꢀOꢀsulfonique acid (HOSA). This
synthesis process has many advantages: it is as clean as
the Raschig process (water chemistry and no pollutant
reagents), HOSA is more stable than monochloramꢀ
ine, it is easier to store, it can be processed in more
concentrated solutions and finely, the synthesis can be
operated batchwise or continuously.
NAPP by HOSA. It was prepared at 20 C with a
°
pH 13.7 ([NaOH] = 0.5 mol/l) by mixing two 100 mL
aqueous solutions of HOSA and NAPP. A white prodꢀ
uct precipitated, which was filtered, washed with cold
water (
5°C
), and then dried under low pressure.
Due to its flexibility to fluctuations in the internaꢀ
tional market, a commercialization of the process can
be expected. The commercial route to hydrazines can
be upscaled for industrial production.
Analyses
NaOH and piperidine concentrations were deterꢀ
mined, by using an automatic titration apparatus,
Metrohm Titrino 702 SM, with a glass electrode (refꢀ
erence: calomel electrode, KCl saturated solution).
1
The article is published in the original.
25

26
LABARTHE et al.
2−
O4
RESULTS AND DISCUSSION
The concentration of
S
ion was determined by
HOSA Stability in Basic Medium
conductometric titration using freshly prepared soluꢀ
tions of barium chloride.
To study kinetics of the reaction between HOSA
and piperidine, the stability of HOSA must be deterꢀ
mined as a function of time. Several works have been
carried out in basic medium: Steinmetz, Robinson
and Ackermann [4] studied the kinetics of HOSA
NAPP concentration was determined using the UV
spectrometric method developed in our laboratory [3].
The method is based on the reaction of NAPP with
formaldehyde to give hydrazone chromophore (methꢀ
yleneꢀpiperidinꢀ1ꢀylꢀamine: FNAPP) according to
the following scheme:
alkaline hydrolysis at 26.6 C, the concentration of
°
NaOH between 0.1 and 1.5 mol/l and constant ionic
strength ([NaCl] = 1.5 mol/l).
The first stage is a limiting step, corresponding to
the formation of hydroxylamine that reacts with
HOSA (NH2OSO3H) via the following fast consecuꢀ
tive parallel reactions:
+ CH2O
N NH2
N N CH2 + H2O.
ka
NH2OSO3− + OH− ⎯⎯⎯→NH2OH + SO24−,
FNAPP
kb
2NH2OSO3− + 2NH2OH + 2OH− ⎯⎯⎯→N2
The spectrometer used was a VARIAN CARY 100
Scan apparatus with double beam equipped with a
data acquisition system.
+ N2H4 + 2SO24− + 4H2O,
kc
2NH2OSO3− + N2H4 + 2OH− ⎯⎯⎯→N2 + 2NH3
Gaz chromatography (GC) was used to monitor
concentrations of PP and NAPP in the course of the
synthesis in concentrated medium. The samples were
diluted in methanol before injection into the chroꢀ
+ 2SO24− + 2H2O,
−
with the rate
[M] where at M = OH–
v = k[NH2OSO3 ]
ka = 0.11
b = 9.6 1 mol–1 min–1 and at M = N2H4 kc
6.0 l mol–1 min–1.
Global stoichiometry and the reaction rate can
then be deduced (kb a):
×
10–2 1 mol–1, min–1, at M = NH2OH
matograph (injected amount:
1 μl). Triglyme was used
k
=
as the internal standard. The apparatus was a Hewlett
Packard 6890A equipped with EPC modules (Elecꢀ
tronic Pneumatic Control) to control and measure all
flow rates and gas pressures in different compartments
of the chromatograph. System data acquisition was
managed by Chemstation HP 3365 software. The colꢀ
umn used was a 30 m long HPꢀInnowax capillary with
>
k
c ӷ
k
3NH2OSO3− + 3OH− → N2 + NH3 + 3SO42− + 3H2O,
d[NH2OSO3−]
−
= ka[NH2OSO3−][OH−].
an internal diameter of 250
packed with polyethylene glycol.
μm and a film thickness of
dt
1 μm
Another work, carried out by Matsuguma and
Audrieth [5] the stability of NaOH (0.12 mol/l) and
HOSA (0.1 mol/l) mixtures at temperatures of 6 and
High performance liquid chromatography (HPLC
)
analyses were performed to determine NAPP concenꢀ
tration during degradation kinetics. The apparatus was
a Hewlett Packard 1100 chromatograph combined
with a UV detector. The system data acquisition was
managed by the same software (Chemstation HP
3365). The column was an Eclipse ODS XDBꢀC8 colꢀ
umn (length = 150 mm, diameter 3 mm), equipped by
a preꢀcolumn for the basic injections. The mobile
phase was a mixture of H2O/CH3CN and the loop
23
describing the results led to a rate constant of the same
order (ka 0.14
10–2 l mol–1 min–1). These results
°
C as a function of time was reported. The curves
≈
×
were confirmed by a test carried out in our laboratory
(iodometric titration).
Under these conditions and in the presence of pipꢀ
eridine, the rate of disappearance of HOSA includes 2
terms, one being related to its hydrolysis:
injection volume was 20 l, with the flow, composition
μ
d[HOSA]
−
= k1[HOSA][PP] + ka[HOSA][OH−]
and wavelength depending on the constituents to be
analysed. A standard curve was constructed using a
series of mixtures with concentrations close to those
expected.
dt
at t = 0.
During the first moments of the reaction, the precedꢀ
ing reaction can be written as:
Gas chromatography coupled with mass spectromꢀ
etry (GCMS) was used to quantify and characterize
molecules such as PP, NAPP or Tetrazene. The specꢀ
trometer was a 5970 Agilent Technology spectrometer
equipped with a quadripolar analyser. Chemstation
software was used.
d[HOSA]
d[NAPP]
−
=
+ ka[HOSA]0[OH−]0.
(
)
(
)
dt
Measurement of slopes at the outset for the [NAPP] =
curves shows that the second term is negligible
(0.3%) under the experimental conditions studied.
dt
0
0
f(t)
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

KINETIC MODELLING OF SYNTHESIS OF NꢀAMINOPIPERIDINE
27
Kinetic Study of HOSA—Piperidine Reaction
in Basic Medium
lnB/(a0–x0)
250
This reaction requires an SN2 type nucleophilic
substitution mechanism.
200
H O
2
y
= 3.3886
x
+ H2SO4.
N NH2
NH + NH2OSO3H
R2 = 0.999
NAPP formation kinetics was first studied at
T =
20 . As the reaction rate is high, the tests were carꢀ
°
C
150
100
50
ried out in diluted medium with reagent concentraꢀ
tions varying from 10–3 to 10–2 mol/l. The formation
rate of NAPP is expressed by the relation:
d[NAPP]
α
β
(1)
v =
= k1[HOSA] [PP] ,
1
dt
are the rate constant of the reaction
where k1
,
α
and
β
versus NAPP and the partial orders versus the reagents
HOSA and piperidine, respectively.
The reaction of HOSA on piperidine leads to the
formation of 2 protons that are immediately neutralꢀ
ized by NaOH (pK (PP) = 11.123 [6]). Under these
α
conditions, pH remains constant with a value deterꢀ
mined by NaOH concentration. This gives the followꢀ
ing theoretical stoichiometric relations:
0
20
40
60
80
Time, min
Fig. 1. Kinetics of NAPP formation. Determination of
x0(a0 – u)
k .
1
d[HOSA]
d[PP] d[NAPP]
−
= −
=
.
B
=
ꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀ.
a0(x0 – u)
dt
dt
dt
The kinetics of this reaction was studied with excess
piperidine to limit NAPP oxidation by HOSA. Since
HOSA proportioning is not easily and instantaneously
accessible, the rate laws were established by measuring
the time dependence of the NAPP concentration
under optimal reaction rate conditions. The kinetic
parameters were determined by integration, considerꢀ
k1
NH2OSO3H + C5H10NH ⎯⎯⎯→C5H10NNH2
+ H2SO4,
2NaOH + H2SO4 → Na2SO4 + 2H2O,
ing
α
=
β
= 1.
⇒ NH2OSO3H + C5H10NH + 2NaOH
→ C5H10NNH2 + Na2SO4 + 2H2O.
NAPP concentration was determined by UV analꢀ
ysis: after sampling, the reaction was immediately terꢀ
minated by quantitatively transforming NAPP into
hydrazone followed by dilution of the product was in a
buffer solution with pH 6.88. The concentration of
Denoting the concentrations of NAPP, HOSA and
piperidine by x0 and a0 respectively, the evolution of
the system is expressed by relation (4) that is given by
integrating differential Eq. (3):
u,
NAPP was determined from the optical density (
A)
obtained at wave lengths λ1 = 208 and λ2 = 236 nm:
u
t
λ1
λ1
λ1
du
Abs = εCH O l[CH2O] + εFNAPP l[FNAPP],
(3)
=
k1dt,
2
∫
0
∫
0
(2)
(x0 − u)(a0 − u)
λ2
λ2
λ2
Abs = εCH O l[CH2O] + εFNAPP l[FNAPP],
2
⎛x0(a0 − u)⎞
1
where
ficients of molar extinction of CH2O and FNAPP at λ1
and λ2
The instantaneous concentrations of piperidine
and HOSA were determined experimentally from the
NAPP concentration by using the following equation:
l
is: length of optical thickness (1 cm), —coefꢀ
ε
(4)
ln
= k1t.
⎜
⎟
a0 − x0
a (x − u)
0 0
⎝
⎠
.
The experimental results are collected in Table 1. As an
example, Fig. 1 shows the first term of Eq. (4) plotted as
a
p
x
function of time for
= [C5H10NH]0/[NH2OSO3H]0 = 10 ([NH2OSO3H]0 =
0 = 1.23 0 = 1.23
10–3 mol/l, [C5H10NH]0 =
10–2 mol/l, [NaOH]0 = 5.33 10–2 mol/l)
a reaction mixture of mole ratio
that is to say
and
[PP]t = [PP]0 − [NAPP]t
[HOSA]t = [HOSA]0 − [NAPP]t
x = x0 − u.
a = a0 − u,
×
a
.
×
that is to say
×
Linearity of the plot indicates that the firstꢀorder
kinetics in two reagents. A linear section of the plot
The reaction scheme can then be written as:
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

28
LABARTHE et al.
Table 1. Experimental data of NAPP formation kinetics in basic medium
Abs
[NAPP]
mol/l
×
103,
Time, min
lnAbs/(a0
–
x0)
k
1, l mol–1 min–1
208 nm
236 nm
7
14
21
28
35
42
49
56
63
70
0.585
0.703
0.801
0.872
0.909
0.956
0.977
0.988
1.001
0.468
0.641
0.779
0.872
0.938
0.999
1.034
1.058
1.079
1.091
0.32
0.55
0.72
0.84
0.93
1.01
1.06
1.09
1.12
1.12
24.525
48.646
74.165
96.094
120.36
145.29
167.14
188.30
209.07
209.47
3.50
3.48
3.53
3.43
3.44
3.46
3.41
3.36
3.32
2.99
1.023
–3
–2
–2
[NH OSO H] =
x
= 1.23
×
10 mol/l, [C H NH] =
a
= 1.23
×
10 mol/l, [NaOH] = 5.33
×
10 mol/l,
T
= 20°C.
2
3
0
0
5
10
0
0
0
extends to
version
observed beyond a value of 60 min is related to the proꢀ concentrations of HOSA and piperidine, equal to
file of the curve [NAPP] = directly calculated 1.25
10–3 and 10–2 mol/l respectively. The variaꢀ
t
≤
60 min, the point where a degree of conꢀ
Effect of temperature. The influence of temperaꢀ
ε
= (x0
–
x)/x0 is equal to 0.9. The deviation ture was studied between 10 and 25
°C
for constant
f(t)
×
1
×
from the relation (2) (Fig. 2). A straight line obtained tion of k1 with temperature obeys Arrhenius equation:
during the first 60 min of reaction passes close to the
k
=
A
exp(–
E/RT),
origin with slope of 3.39 l mol–1 min–1 and a correlaꢀ
tion coefficient R2 = 0.999. The reaction rate estabꢀ
where
A is the Arrhenius preꢀexponential factor
(l mol–1 min–1),
E—activation energy (J/mol), R—
lished in relation to HOSA (
ϕ
= [NAPP]∞/[HOSA]0
)
perfect gas constant (8.314472 J K–1 mol–1),
T—temꢀ
is close to 91%.
perature (K).
To verify the rate law at 20°C, a similar treatment
Curve ln(
with slope E1
determine the activation energy E1 and the factor A1
This gives 1 = 1.5
1010 exp(–54.1/RT
k
1) =
f
(1/
T) is described by a straight line
was carried out for reagent mixtures with different
compositions. The results are given in Table 2: the valꢀ
ues of k1 are quite constant. This gives
–
/R and intercept point lnA1. It serves to
.
k
×
).
d[NAPP]/dt
= 3.39 0.10 l mol–1 min–1 at 20
C.
°
k1 =
The activation enthalpy and entropy are deterꢀ
mined from the relations:
[HOSA][PP]
ΔH10# = E1 − RT, ΔS10# = R ln(A1 h ekBT),
(5)
[NAPP], mol/l
0.0012
kB and
Planck constant respectively (
= 6.623 = 25 C,
10–34 Js). At
0 # = –58.6 J K–1 mol–1,
h
, representing the Boltzmann constant and the
B = 1.380
10–23 J/K,
0 # = 51.6 kJ/mol,
k
×
0.0010
0.0008
0.0006
0.0004
0.0002
h
×
T
°
ΔH1
0#
= 69.0 kJ/mol.
ΔG1
ΔS1
Kinetics of NAPP Degradation in Basic Medium
The deviation observed at the end of the reaction is
due to a partial oxidation of hydrazine by HOSA. To
quantify this phenomenon, a kinetic and mechanistic
study was carried out under the same conditions as
described above.
0
0
The GC–MS analysis of the reaction medium
showed that degradation essentially leads to a linear
tetrazene [7] or azopiperidine which was identified
and characterized. Tests were performed in basic
medium for concentrations of NAPP and HOSA
20
40
60
80
Time, min
Fig. 2. Kinetics of NAPP formation in basic medium. Evoꢀ
lution of NAPP concentration through time, calculated
from equation (2).
between 2.5
×
10–3 and 1.5
×
10–2 mol/l. Tests perꢀ
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

KINETIC MODELLING OF SYNTHESIS OF NꢀAMINOPIPERIDINE
29
Table 2. Determination of the rate constant of the reaction between HOSA and PP (k1) in basic medium. Influence of the
initial reagent concentrations
[HOSA]0 [PP]0
103, mol/l
×
×
102 , mol/l
[NaOH]0
×
102, mol/l
k
1, l mol–1 min–1
R2
p
1.24
1.24
1.30
1.20
1.23
1.22
1.29
1.25
0.15
0.37
0.78
0.96
1.23
1.46
1.02
1.00
5.45
5.52
5.57
5.55
5.33
5.55
3.16
7.04
1.19
3.26
3.39
3.42
3.32
3.39
3.57
3.27
3.50
0.992
0.997
0.995
0.996
0.999
0.999
0.996
0.996
3
6
8
10
12
8
8
formed in more concentrated medium led to the preꢀ In aqueous medium, diazene can be considered as a
cipitation of azopiperidine.
The first step corresponds to the formation of a diaꢀ
zene:
base, able to capture a proton according to the followꢀ
ing acidꢀbase equation:
+
–
+
+ H O
+ OH–.
N N H
N N
2
NH2OSO3H +
N NH2
Diazene can be in its conjugate form in a weakly basic
medium. In this case, azopiperidine is obtained by
rapid dimerization between the molecular form and
the protonated form of cycloamonitrene:
+
–
+ NH+HSO–.
N N
4
4
+
–
+
–H+
+
+
N N
H N N
N N N N
N N N N
.
H
In this reaction, the yield of tetrazene per NAPP conꢀ
After sampling, the reaction was immediately terꢀ
sumed was between 74 and 80%, providing evidence of minated by quantitatively transforming NAPP into
a competing reaction. The literature indicates the hydrazone followed by dilution of the product was in a
existence of a triazium salt [8–10] whose reaction buffer solution with pH 6.88. Then, the concentration
mechanism corresponds to an SN2 amino reaction.
in the sample was determined by HPLC at wavelength
= 236 nm. Following this, the HOSA concentration
could be evaluated for each moment of the reaction.
Defining the amount of reacted NAPP as , the HOSA
λ
NH2
+
N
NH2OSO3H +
+ HSO4–.
N NH2
δ
concentration can be calculated from the following
relation:
NH2
Since this reaction is secondary, the reaction used for
modelling is that which leads to the formation of diaꢀ
zene.
[HOSA]t = [HOSA]0
− ([NAPP]0 −[NAPP]t) = [HOSA]0 − δ,
Kinetic aspect. As the formation of diazene is stoꢀ
ichiometric (unity), the NAPP degradation rate is
expressed by the relation:
that is to say
x = x0 − (u0 − u) = x0 − δ.
Figure 3 gives the concentration of NAPP as a
function of time for a mixture with following initial
concentrations:
d[NAPP]
α
β
(6)
v = −
= k2[HOSA] [NAPP] ,
[NH2OSO3H]0 = 1.25
×
×
10–2 mol/l, [C5H10NNH2]0
10–2 mol/l.
dt
and
= 2.50
10–3 mol/l, [NaOH]0 = 5.5
×
where k2
,
α
β
are the degradation rate constant of
NAPP and the partial orders to the reagents, HOSA
and NAPP respectively.
As the evolution of the degradation products is
independent of the reagents, the yield depends only on
The kinetics was followed in oxidative and basic the first reaction step. The kinetic parameters were
medium, i.e. in the presence of HOSA taken in 5 and determined by using an integration procedure similar
8 fold excess. The data were monitored as the NAPP to that described above and assuming a partial order of
concentration plotted as a function of time.
1 for each of the reagents.
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

30
LABARTHE et al.
p
= 5 ([NH2OSO3H]0 = x0 = 1.25
[C5H10NNH2]0 = 0 = 2.50
10–3 mol/l, [NaOH]0 =
5.5
10–2 mol/l). We observe that the firstꢀ rate law is
×
10–2 mol/l,
[NAPP], mol/l
0.0030
u
×
×
0.0025
0.0020
0.0015
0.0010
0.0005
obeyed without deviation up to the completion of the
reaction (250 to 300 min). Thus, the reagents are not
involved in a secondary reaction process. A similar
treatment was carried out for a series of mixtures with
different reagent concentrations. The results are given
in Table 3. The values obtained are fairly constant. At
T
= 20°C we obtain:
d[NAPP]/dt
= 1.17 0.04 l mol–1 min–1.
k2 = −
[HOSA][NAPP]
0
100
200
300
400 500
Time, min
Influence of temperature. The effect of temperature
was studied between 20 and 35 for HOSA and
NAPP at constant concentrations of 1.28
10–2 and
2.50
10–3 mol/l respectively. The variation of k2 with
temperature obeys Arrhenius law. This gives 2 = 1.83
1011 exp(–62.8/RT
. The thermodynamic activation
are as follow:
Fig. 3. Kinetics of NAPP oxidation by HOSA. Evolution of
°
C
NAPP concentration through time.
×
×
k
×
ln
B
'/(x0–u0)
)
values at
T
= 25 C
°
y
= 1.1469x
ΔH20 #
ΔS20 #
= 60.4 kJ/mol,
= –38 J K–1 mol–1,
= 71.6 kJ/mol.
R2 = 0.999
250
200
150
100
50
ΔG20 #
Formulation of a Kinetic Model in Basic Medium
Consistent with the foregoing dada, we can suggest
a plausible sequence of reactions for the synthesis
NAPP with HOSA:
NH2OSO3H + OH− → NH2OSO3− + H2O,
(I)
NH2OSO3− + C5H10NH
(II)
k1
⎯⎯⎯→ C5H10NNH2 + HSO4−,
0
50
100
150
200
250 300
Time, min
OH− + HSO4− → SO42− + H2O,
(III)
Fig. 4. Kinetics of NAPP degradation. Determination of
' = xu x u
−
C5H10NNH2 + NH2OSO3
⎯⎯⎯→ C5H10N =N + NH4 + SO24−,
k
.
B
/
0 0
2
(IV)
k2
+
−
+
Concentrations of NAPP at the initial time and at
, are defined as u0 and , respectively, likewise for
HOSA they are x0 and . Thus, by writing a0 u0 x0
and observing that u0 x0 a0, the inteꢀ
gration of Eq. (6) gives the following expressions:
k
−
+
'
2
t
u
(V)
OH + NH4 ⎯⎯⎯→NH3 + H2O,
x
=
– ,
x
=
u
–
+
= u –
+
−
(VI)
2C5H10N =N → C5H10NN=NNC5H10.
u
t
The excess NaOH ensures a stoichiometry of unity
versus the reagents, since neutralisation reactions (III)
and (V) are instantaneous. The useful step of the synꢀ
thesis is reaction (II). This reaction is bimolecular and
first order in the reagents. Reaction (IV) is consecuꢀ
tiveꢀparallel. Indeed, as NAPP is formed, it is partially
−
du
u(a0 + u)
(7)
(8)
−
=
k2dt,
∫
0
∫
0
⎛u0x⎞
1
ln
= k2t.
⎜
⎝
⎟
⎠
x0 − u0
x u
0
oxidised by NH OS to give an intermediate comꢀ
O3
To illustrate the use of this expression, Fig. 4 shows
2
the first term of the Eq. (8) plotted as a function of pound, diazene (IV), which then dimerizes to form
time for a reaction mixture at 20 and a mole ratio azopiperidine (VI).
°C
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

KINETIC MODELLING OF SYNTHESIS OF NꢀAMINOPIPERIDINE
Table 3. Determination of the rate constant of the oxidation of NAPP by HOSA ( 2) at 20
31
k
°C
[NaOH]0 ×
102,
[NH2OSO3H]0
102, mol/l
×
[C5H10NNH2]0
103, mol/l
×
2, l mol–1 min–1
R2
p
k
mol/l
1.25
1.50
1.74
1.25
1.28
2.50
2.50
2.50
2.50
2.50
7.50
7.50
7.50
5.50
9.50
5.1
6
7
5
5.1
1.13
1.20
1.18
1.14
1.18
0.989
0.991
0.996
0.986
0.989
It is now possible to evaluate the instantaneous
Accordingly, we can modify Eqs. (11) and (10)
concentrations of the main products and the predicꢀ dividing limb by limb:
tive calculation of the yield versus the reagent concenꢀ
trations, pH and temperature. To approach calculaꢀ
tions, the system of differential equations defined by
the rate laws of the previous reactions needs to be
solved. By designating the concentrations of HOSA,
k2
k1
du
da
u
a
with
= −1 + r
r =
.
The integration of this equation leads to:
r −1
⎡
⎤
⎥
piperidine, NAPP, NaOH and diazene, at instant
and
t
, by
⎛
⎞
1
r − 1
a
(15)
a0
u =
a 1 −
.
⎢
x,
a,
u,
b
f
respectively, the synthesis is described by
⎜
⎟
a
0
⎝
⎠
⎢
⎣
⎥
⎦
the following system of differential equations:
Equation (15) permits verifying that
u
= 0 when
a
=
.
dx
dt
Substitution of Eq. (15) in Eq. (14) yields a differꢀ
ential equation whose resolution leads to the following
implicit equation:
(9)
= −k1xa − k2xu,
da
dt
(10)
(11)
(12)
= −k1xa,
r
⎡
⎤
⎥
a0
⎛
⎞
2r − 1
a
(16)
= 0.
(x0 − x) − (a0 − a)
+
1 −
⎣
⎢
⎜
⎝
⎟
r − 1 r − 1
a
0
⎠
⎢
⎥
⎦
du
dt
= k1xa − k2xu,
Therefore the final NAPP concentration (u∞) can be
deduced from Eq. (15):
db
= −2(k1xa + k2xu),
r−1
⎡
⎤
⎥
⎛
⎞
a∞
1
r −1
dt
(17)
u∞ =
a 1 −
.
⎢
∞
⎜
⎝
⎟
⎠
a
⎢
⎣
0
⎥
⎦
df
(13)
= k2xu,
The instantaneous diazene concentration is
obtained from the material balance related to NAPP,
combining Eqs. (10), (11) and (13). That gives at t = ∞:
dt
with the initial conditions
= 0.
x
=
x0,
a
=
a0,
b
=
b0 and
u
= f
r −1
⎡
⎤
⎥
⎛
⎞
a∞
a∞
In particular, we note
(18)
),
f∞ =
pi = (a0 − a∞) −
1 −
⎣
.
⎢
∑
i
⎜
⎝
⎟
r − 1
a
⎠
⎢
0
⎥
⎦
db
dt
dx
dt
= 2
.
The intermediate concentrations of piperidine (
a
HOSA ( ) and NAPP ( ) can be related to time by
x
u
resolving the integral obtained from Eqs. (10) and
(16):
This system of equations can be solved numerically by
the Runge–Kutta method. However, a semiꢀalgebraic
resolution was sought that permits evaluation of the
behaviour of the system in time by using a single
implicit equation and analytical relations. It also
allows a direct access to the final yields on the basis of
reagent concentration ratios and temperature ignoring
at
1
k1
da
t =
−
∫
a0
a[x0 − g(a,a0,r)]
r
⎡
⎤
⎥
a0
⎛
⎞
2r − 1
a
a
0
the time factor. To achieve this,
Eqs. (9) and (10):
t
is eliminated between
with
g(a,
a0,
r) =
(a0 − a)
+
1 −
.
⎢
⎜
⎝
⎟
⎠
r − 1 r − 1
⎢
⎣
⎥
⎦
k
dx
da
2 u
(14)
= 1 +
.
Determination of the Parameters Controlling
the Synthesis
k1 a
Ratio k2
/
k1 depends only on ionic strength and temperꢀ
The previous equations may be applied to predict
the evolution of the system under specific experimenꢀ
ature. Equation (14) is integrable, expressing
u
= f(a).
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

32
Yield, %
LABARTHE et al.
r
⎡
⎤
⎥
⎛
⎞
⎛
⎞
a∞
p
a∞
2r − 1
(19)
1 − p 1 −
+
1 −
= 0.
100
90
80
70
60
50
40
30
20
10
⎢
⎜
⎝
⎟
⎜
⎝
⎟
⎠
a0 r − 1 r − 1
a
⎠
⎢
⎣
0
⎥
⎦
Taking into account the reagent material balances at t∞
x0 ∞ + 2 f∞
a0 u∞
:
=
u
,
–
a∞
=
+ f∞.
Then, by eliminating the term related to NAPP oxidaꢀ
tion between the two previous equations and by noting
ϕu
=
u∞
/
x0, one obtains
experimental
theoretical
a∞
ϕu + 1
2p
= 1 −
.
a0
The substitution of this expression in Eq. (19) leads to
the following implicit general equation:
r
⎡
⎤
⎥
p
⎛
ϕu +1⎞
2r −1
2(r −1) r −1
⎢
1 − (ϕu +1)
+
1 − 1 −
= 0.
⎜
⎝
⎟
2 p
⎠
⎢
⎣
⎥
⎦
This equation clearly shows that NAPP yield (ϕu
depends only on a0 x0 and on k2 k1 being
only a function of pH and temperature. We can verify
that when 0 ( 2 = 0), yield ϕu → . The yields
related to amine and diazene are deduced directly:
)
0
5
10
15
p
p
=
/
r
=
/ , r
r
→
k
1
Fig. 5. Evolution of NAPP yield with reagents ratio
centrated medium at [NaOH] = 0.05 mol/l,
k
p
in conꢀ
T
= 20 C,
°
∞
k
–1
–1
–1
–1
u∞
2ϕu
1 − ϕu
= 3.39 l mol min and
= 1.17 l mol min .
1exp
2exp
and
ϕa =
=
ϕf =
.
a0 − a∞ ϕu + 1
2
tal conditions, but they do not refer to the parameters
that affect the yield, especially the initial concentraꢀ
tion ratios defined by the relation:
Validation of the Model and Determination
of Optimal Conditions
The results obtained by the theoretical study were
verified by experiments conducted in a concentrated
medium at a temperature of 20 C for different ratios p
(1 to 12) and for increasing reagent concentrations.
The amount of sodium hydroxide needed to neutralise
all the H+ ions involved in NAPP formation and mainꢀ
tain a final sodium hydroxide concentration equal to
0.05 mol/l was calculated.
C H NH
[
]
5
10
0
p =
.
°
NH OSO H
[
]
2
3
0
Introducing this parameter into the Eq. (16) we have:
Table 4. Influence of the reagents ratio (
in concentrated medium at [NaOH]∞ = 0.1 mol/l and
20°C
p) on NAPP yield
Influence of the
p ratio. The experiments were carꢀ
T
=
ried out at 20 , by varying the concentration of
°
C
HOSA, while maintaining constant temperature and
final NaOH concentration. Piperidine and HOSA volꢀ
umes were equal to 100 ml. Table 4 gives the initial
reagent concentrations and Table 5 gives the molar
Initial reagents concentrations, mol/l
p
= a0/x0
[C5H10NH]0 [NaOH]0
[HOSA]0
composition at
t
= 0 and
t
= ∞, and the NAPP yield ϕu.
Figure 5 shows that the yield increases by 30% when
increases from 1.2 to 12. The theoretical curve agrees
with the experimental points.
p
1.2
1.2
1.2
3
0.10
0.30
0.50
0.50
0.50
0.50
0.50
0.21
0.55
0.89
0.37
0.21
0.17
0.13
0.08
0.25
0.42
0.16
0.08
0.06
0.04
NAPP yield increases with ratio
[PP]0/[HOSA]0. For = 6, the yield reaches 90%.
Beyond this value, doubling the ratio leads only to a
negligible increase in the yield ( = 94.1% for = 6,
= 96.5% for = 12). Thus ratio = 8 was used for
synthesis under optimal conditions.
Effect of temperature. The influence of temperaꢀ
ture was studied between 10 and 80 for a final conꢀ
centration of NaOH equal to 0.05 mol/l and for a ratio
p
=
p
ϕ
p
ϕ
p
p
6
8
°C
12
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

KINETIC MODELLING OF SYNTHESIS OF NꢀAMINOPIPERIDINE
Table 5. Formation of NAPP. Experimental results
Real concentration in the reaction medium, mol/l
= 0
33
Yield ϕu
= u∞/x0, %
p
=
a0/
x0
t
t
=
∞
[NaOH]0
[HOSA]0
[NaOH]∞
[NAPP]∞
exp.
theor.
1.2
1.2
1.2
3
0.11
0.28
0.45
0.19
0.11
0.09
0.07
0.04
0.13
0.21
0.08
0.04
0.03
0.02
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.03
0.09
0.13
0.07
0.04
0.03
0.02
65.9
66.0
64.5
85.9
91.8
93.9
94.6
65.6
65.6
65.6
87.7
94.1
95.6
96.5
6
8
12
p
= 1.2. The initial reagent concentrations were as folꢀ Under these conditions, pH is controlled through time
2−
lows: [C5H10NH]0 = 0.50 mol/l; [NaOH]0 = 0.89 mol/l;
[NH2OSO3H]0 = 0.42 mol/l. For this experiment,
100 ml of aqueous HOSA solution was introduced
into 100 ml of a mixture of piperidine and NaOH.
Table 6 gives the corresponding yields.
The higher the temperature the lower the yield.
Thus NAPP must be synthesized at a low temperature,
by the buffer mixture PP—(PPH+)
Therefore
2SO4 .
PP is consumed by both reaction with HOSA and by
protonation. The latter reaction contributes to a slowꢀ
down in the rate of NAPP formation. The following
theoretical stoichiometric relations can be written:
d[HOSA]
d[PP] d[NAPP]
1
3
−
= −
=
.
i.e. 20 C.
°
dt
dt
dt
The evolution of NAPP at industrial concentration.
Figure 6 presents the kinetic curves obtained by modꢀ
elling the behaviour of the system under conditions
which include reagent concentrations approaching
those used in industry, i.e. the addition of an aqueous
HOSA solution of 32% w/w to a mixture of NaOH and
piperidine at 20
reaction mixture are as follows: [HOSA]0 = 0.69 mol/l,
[PP]0 = 5.55 mol/l, [NaOH]0 = 1.39 mol/l, = 8.
The reaction is almost terminated at = 0.35 min,
By designating the concentrations of NAPP, HOSA
and piperidine at the initial time by x0 and a0 respecꢀ
tively, the integration of the previous relations gives:
u,
u
t
du
=
k3dt,
°
C. The initial concentrations of the
∫
0
∫
0
(x0 − u)(a0 − 3u)
p
⎛x0(a0 − 3u)⎞
1
t
ln
= k3t.
⎜
⎟
corresponding to a NAPP yield close to 96%.
a0 − 3x0
a (x − u)
0 0
⎝
⎠
Kinetics of NAPP Formation in Buffer Medium
Table 6. Influence of temperature on the NAPP yield in
concentrated medium at
[NaOH]∞ = 0.05 mol/l
p
= 1.2 and real concentration
A kinetic study of NAPP formation was carried out
to verify that the results obtained in basic medium can
be used in buffer medium (pH 11–12). The methodolꢀ
ogy is similar to that applied to the data for basic
medium. The reaction scheme is as follows:
ϕu, %
T, °C
exp.
64.5
theor.
NH2OSO3H + C5H10NH
k3
C H NNH + H SO ,
2
⎯⎯⎯→
5
10
2
4
20
50
80
65.6
59.2
53.6
+
2−
2C5H10NH + H2SO4
→
(С5H10NH2)2 SO4 ,
58.9
53.2
⇒ NH2OSO3 + 3C5H10NH → C5H10NHH2
+
+ (C5H10NH2 )2SO24−.
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

34
LABARTHE et al.
Concentration, mol/l
6
Yield, %
100 100
Yied, %
90
80
70
60
50
40
30
20
10
0
5
80
60
40
20
0
2
1
4
5
3
2
2
4
1
1
3
0
0.05 0.10 0.15 0.20 0.25 0.30 0.35
Time, min
0
5
10
15
p
Fig. 6. Evolution of HOSA (
NaOH ( ) and yield (
medium and under industrial conditions.
1
), PP (
2
), NAPP (
3
) and
4
5
) during the reactions in basic
Fig. 7. Evolution of NAPP yield in basic (
) mediums.
1) and in buffer
(
2
The results are presented in Table 7. The values Figure 7 shows the evolution of NAPP yield in basic
obtained are fairly constant. The following expression medium and in buffer medium. The curves have the
can be deduced at
T
= 20
°
C
:
same shape but in buffer medium, the inflexion point
appears later. This is due to partial protonation of the
amine by the H+ ions stemming from the formation of
NAPP in buffer medium.
d[NAPP]/dt
= 2.56 0.08 l mol–1 min–1.
k3 =
[HOSA][PP]
For
p = 8, the yields are very similar to those
The influence of temperature was studied at temperaꢀ
tures between 5 and 25 for constant concentrations
of HOSA and piperidine equal to 1.25
10–3 and
1.5
10–2 mol/l respectively. The variation of k3 with
obtained in basic medium. Under these conditions, it
is possible to synthesize NAPP in buffer medium, with
the optimal parameters being established in basic
medium.
°
C
×
×
temperature obeys Arrhenius law. Thus we can write:
k3 = 5.99 ×1010 exp −
.
58.2
RT
CONCLUSION
)
(
NAPP synthesis using hydroxylamineꢀOꢀsulfonic
acid (HOSA) is a new and promising process. It was
therefore desirable to study reaction mechanisms. The
results indicate that the formation of NAPP is the
firstꢀ order reaction to two reagents, piperidine and
HOSA described by a nucleophilic substitution mechꢀ
The activation parameters at 25
°
C
can be deduced:
0#= 55.7 kJ/mol,
= –47 J K–1mol–1.
ΔH3
ΔS30 #
= 69.7 kJ/mol.
ΔG30#
Table 7. Determination of rate constant of reaction between HOSA and PP (
k
3) at 20°C in buffer medium
3, l mol–1 min–1
R2
[NH2OSO3H]0
×
103, mol/l [C5H10NH]0
×
103, mol/l
k
p
1.25
1.25
1.26
1.25
10.7
12.5
14.7
17.7
8.6
10.0
11.6
14.2
2.41
2.54
2.50
2.76
0.9992
0.9998
0.999
0.9994
KINETICS AND CATALYSIS Vol. 53
No. 1
2012

KINETIC MODELLING OF SYNTHESIS OF NꢀAMINOPIPERIDINE
35
anism of SN2 type. The main secondary reaction idenꢀ carried out in basic medium and then in buffer
tified was NAPP oxidation by HOSA which consisꢀ medium with fairly similar results.
tently decreased NAPP yield. This reaction occurred
in 2 consecutive steps: the formation of a diazene folꢀ
lowed by dimerization to tetrazene.
The study of the different parallel reactions
resulted in a system of differential equations. By solvꢀ
ing these equations it was possible to find expressions
that predict the evolution of the system as a function of
the reagent concentrations, pH and temperature. A
mathematical treatment was used to characterize the
reaction medium from a unique implicit equation.
REFERENCES
1. Jain, S.R. and Chellappa, D., Proc. Indian Acad. Sci.
Chem. Sci., 1985, vol. 95, no. 4, p. 381.
2. Darwich, C., French Thesis Univ. Claude Bernard Lyon 1
2005, vol. 29, p. 2005.
3. Delalu, H. and Marchand, A., Anal. Chim. Acta, 1992,
,
vol. 266, p. 39.
4. Steinmetz, W.E., Robinson, D.H., and Ackermann, M.N.,
Inorg. Chem., 1975, vol. 14, p. 421.
5. Matsuguma, H.J. and Audrieth, L.F., J. Inorg. Nucl.
Chem., 1959, vol. 12, p. 186.
The equation clearly shows that the
initial concentrations [PP]0/[HOSA]0, the rate conꢀ
stant ratio k2 k1 and temperature are the only paramꢀ
p molar ratio, the
/
6. CRC Handbook of Chemistry and Physics, Lide, D.R.,
eters that affect the yield of NAPP from converted
[HOSA]. The curves obtained according to the model
agree well with the experimental curves. This provides
evidence of the validity of the model that makes it posꢀ
sible to determine optimal conditions for the reaction:
Ed., Boca Raton, Fla.: CRC, 2003–2004, 84th ed., p. 8.
7. Stradi, R., Atti Accad. Naz. Lincei Cl. Sci. Fis. Mat. Nat.
Rend., 1947, vol. 43, p. 350.
8. Linke, K.H. and Turley, R., Z. Naturforsch., A: Phys.
Sci., 1973, vol. 28b, p. 781.
9. Sommer, F., Schluz, O.F., and Nassau, M., Z. Anorg.
Allg. Chem., 1925, vol. 147, p. 142.
[HOSA]0 = 0.69 mol/l, [PP]0 = 5.55 mol/l,
pH 11–12, = 20 C.
p = 8,
T
°
Thus NAPP was synthesised by fast and selective
reaction with a yield around 96%. This study was first
10. Freyer, D., Voigt, W., and Kohnke, K., Eur. J. Solid
State Inorg. Chem., 1998, vol. 35, p. 595.
KINETICS AND CATALYSIS Vol. 53
No. 1
2012