Synthesis, modeling and optimization of cyanide antidote (3-methylbutyl) nitrite using response surface methodology

Article abstract of DOI:10.1007/s11164-015-2156-5

In this research, synthesis of cyanide antidote (3-methylbutyl) nitrite has been designed and performed by response surface methodology. The antidote was prepared by the reaction of 3-methyl-1-butanol with sodium nitrite, which is the nitrosonium (NO⁺) source, in the presence of sulfonic acid-functionalized Fe₃O₄ (Fe₃O₄@SO₃H) as a nanomagnetic solid acid catalyst and triethylammonium hydrogen sulfate as ionic liquid in aqueous media. Optimization of the reaction conditions was investigated using the response surface method (central composite design). In accordance with analysis of variance, a quadratic polynomial model was able to predict the response. Predicted response values by the obtained model was in good agreement with the experimental results. A clean reaction, easy workup procedure, reusability of the catalyst and a high yield are some advantages of this method.

Full text of DOI:10.1007/s11164-015-2156-5

Res Chem Intermed
DOI 10.1007/s11164-015-2156-5
Synthesis, modeling and optimization of cyanide
antidote (3-methylbutyl) nitrite using response
surface methodology
1
2
•
Vahid Khakyzadeh
Amir Ahmad Salarian
•
Hadi Rezaei Vahidian
3
2
•
Mohammad Ali Zolfigol
Received: 30 December 2014 / Accepted: 27 June 2015
Ó Springer Science+Business Media Dordrecht 2015
Abstract In this research, synthesis of cyanide antidote (3-methylbutyl) nitrite has
been designed and performed by response surface methodology. The antidote was
prepared by the reaction of 3-methyl-1-butanol with sodium nitrite, which is the
?
nitrosonium (NO ) source, in the presence of sulfonic acid-functionalized Fe O
3
4
(
Fe O @SO H) as a nanomagnetic solid acid catalyst and triethylammonium
3 4 3
hydrogen sulfate as ionic liquid in aqueous media. Optimization of the reaction
conditions was investigated using the response surface method (central composite
design). In accordance with analysis of variance, a quadratic polynomial model was
able to predict the response. Predicted response values by the obtained model was in
good agreement with the experimental results. A clean reaction, easy workup
procedure, reusability of the catalyst and a high yield are some advantages of this
method.
Keywords Cyanide antidote Á Nitrosonium Á Experimental design Á
Functionalized sulfonic acid (Fe O @SO H)
3
4
3
&
Vahid Khakyzadeh
v.khakyzadeh@gmail.com; mr.khakyzadeh@yahoo.com
1
Institut fur Organische Chemie, Albert-Ludwigs-Universitat Freiburg, Albertstrasse 21,
79104 Freiburg, Germany
2
3
Faculty of Chemistry, Bu-Ali Sina University, 6517838683 Hamedan, Iran
Department of Toxicology, School of Pharmacy, Shahid Beheshti University of Medical
Science, Tehran, Iran
123

V. Khakyzadeh et al.
Introduction
Cyanide poisoning may be observed in a wide variety of settings. Hydrocyanic acid
and cyanide are used in most common industrial processes such as metal cleaning,
electroplating, precious metal extraction, laboratory examinations, jewelry produc-
tion and photography [1–4]. Moreover, one of the chemical intermediates in the
manufacture of plastics, synthetic fibers and nitriles is hydrogen cyanide [5]. Also,
fire smoke inhalation in enclosed spaces may result in cyanide poisoning [6, 7].
3
?
By binding with the Fe
of mitochondrial cytochrome oxidase, cyanide
produces histotoxic hypoxia and disrupts the ability of cells to utilize O2 in
oxidative phosphorylation and in the normal functioning of the electron transport
chain [1]. (3-methylbutyl) nitrite is one of the important cyanide antidotes [8].
In cyanide poisoning, (3-methylbutyl) nitrite produces methemoglobinemia and
vasodilation. In this case, vasodilation may contribute to adverse effects and require
therapy [9].
Considering the above points and toxicity of cyanide, finding new improvement
paths for synthesis of cyanide antidotes is still of importance.
Since the late 90s, green chemistry research has been a strategy of removing
waste from chemical processes, thereby contributing to environmental protection
[
10, 11]. One aspect of green chemistry involves using ionic liquids as helpful
substitutes for the conventional organic solvents. In fact, room temperature ionic
liquids have various unique properties such as high thermal stability, wide liquid
range, controllable polarity, negligible vapour pressure and excellent solubility for a
wide range of substances [12].
High specific surface area of the active component is an important property of
nanocatalysts, in which the contact between reactants and catalyst enhances
reactions [13, 14]. To reuse nanomaterials, they must be separated from the reaction
media, a difficult undertaking. This problem could be solved by application of
magnetic nanoparticles (MNPs) as a catalyst. MNPs have large specific surface area
and can be easily collected by a magnet for reuse to prevent loss of the catalyst [15–
2
0]. The relationship between a phenomenon and the response of natural process
and the factors which affect them can be categorized in three levels: top level,
slightly low-level and lower level. The top level is deterministic phenomena, and
the second and third levels are stochastic phenomena [21]. Synthesis processes are
in the lower level. This area of knowledge is based on experimental results. A
mathematical model can be used to interpret the experimental results. Modelling
and optimization of the synthesis processes can be done using design of experiment
(
DOE) [22].
Having the above facts in mind, we report a clean, facile and efficient method for
synthesis of cyanide antidote (3-methylbutyl) nitrite from 3-methyl-1-butanol and
?
nitrosonium (NO ) which is generated in situ using sulfonic acid-functionalized
Fe O (Fe O @SO H) as a nanomagnetic solid acid catalyst and triethylammonium
3
4
3
4
3
hydrogen sulfate (TEAHS) as ionic liquid in aqueous media (Scheme 1). The
synthesis was modelled and optimized by experimental design.
1
23

Synthesis, modeling and optimization of cyanide antidote…
Scheme 1 Synthesis of cyanide antidote (3-methylbutyl) nitrite
Experimental
Chemicals and characterization methods
All chemicals were purchased from Merck, Fluka and Dae-Jung Chemical
Companies. The product was characterized by a gas-chromatography (GC) system,
Shimadzu model CLASS-GC14B with a flame ionization detector (FID).
General procedure for the synthesis of cyanide antidote
To a mixture of 3-methyl-1-butanol (0.1 mL, 1 mmol), H O (3 mL), NaNO
2
2
(
0.142 g, 2.09 mmol) and Fe O @SO H [23, 24] (0.005 g) in a test tube at 0 °C,
3
4
3
TEAHS (1.06 mL) was added at a rate of 0.15 mL/s, and the resulting mixture was
stirred for 6 h at 0 °C. The reaction conversion was investigated by GC. After
completion of the reaction, two phases were formed. The product was accumulated
in the upper phase (yellow liquid). The organic layer was separated and washed with
saline solution, then dried with MgSO to give the product with a conversion of
4
8
7 %.
Table 1 Effect of various
solvents and temperatures on the
reaction of 3-methyl-1-butanol
a
Entry
Solvent
–
Conversion (%)
-
10 °C
0 °C
25 °C
(
1 mmol) with NaNO
1.5 mmol), TEAHS (0.5 mL)
2
(
1
2
3
4
5
6
7
8
–
Trace
50
Trace
15
H
2
O
20
EtOH
Trace
–
Trace
Trace
Trace
–
Trace
–
EtOAc
CHCl
3
–
Trace
–
All reactions were carried out
with a constant catalyst amount
THF
Trace
–
n-Hexane
Trace
–
–
(0.01 g)
Acetonitrile
Trace
Trace
a
Reaction time: 6 h
123

V. Khakyzadeh et al.
Results and discussions
Quick screening of the variables
At first, to find the general reaction conditions and the range of variables for the
synthesis of the cyanide antidote before design of experiments, a number of pre-
tests were performed. To this end, the reaction of 3-methyl-1-butanol with different
amounts of NaNO in the presence of TEAHS as an ionic liquid and Fe O @SO H
2
3
4
3
as a catalyst was tested in two general conditions. To compare the efficiency of the
solvent-free versus solvent conditions as well as the suitable reaction temperature,
the reaction was examined in various solvents and solvent-free conditions in a range
of -10 to 25 °C (Table 1). As it is clear from Table 1, trace conversions of the
reaction were obtained in the organic solvents and also in solvent-free conditions
within the investigated temperature range. The best results were observed in
aqueous media at 0 °C (Table 1, entry 2). Thus, selection of aqueous media and
temperature of zero were two general results of variable screening before design of
the experiment.
In continuation, the effect of different amounts of the nanomagnetic catalyst on
the reaction was investigated at 0 °C in water (Table 2). As shown in Table 2,
higher conversion was obtained when the reaction was carried out using 0.017 g of
Fe O @SO H.
3
4
3
Experimental design and statistical analysis
The central composite design (CCD) of the response surface method, with nine
replicates at the central point was employed for design of the experiment. Variables,
such as temperature (A), catalyst amount (B), amount of NaNO (C), amount of
2
H O (D), amount of ionic liquid (E), rate of adding the compounds (F) and rate of
2
mixing the compounds (G) were used as effective parameters of the process. The
levels of the variables are shown in Table 3.
Table 2 Effect of different amounts of nano magnetic catalyst on the reaction of 3-methyl-1-butanol
(1 mmol) with NaNO
2
(1.5 mmol) and TEAHS (0.5 mL)
a
Entry
Catalyst amount (g)
Conversion (%)
1
2
3
4
5
6
0.017
0.015
0.01
52
50
50
0.007
0.005
0.001
48
45
\20
All reactions were carried out at 0 °C
a
Reaction time: 6 h
1
23

Synthesis, modeling and optimization of cyanide antidote…
Table 3 Levels of the experimental variables of the CCD
Variables
Units
-a
Low actual
0
High actual
?a
15
o
C
A
B
C
D
E
F
-15
1
-7.5
3.25
1.5
0.000
5.500
7.5
g 9 1000
eq.
7.75
2.5
10
3
1
2.000
mL
1
2
3.000
4
5
mL 9 100
(mL/s) 9 10
rpm
50
1
75
100.000
1.500
125
1.75
1100
150
2
1.25
900
G
800
1000.000
1200
Table 4 Analysis of variance for the response surface quadratic model for conversion
Source
Sum of squares
df
Mean square
F value
p value prob [ F
Model
A
33,041.91
420.01
210.01
958.24
1012.07
748.24
223.13
110.63
118.2
16
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2065.12
420.01
210.01
958.24
1012.07
748.24
223.13
110.63
118.2
98.03
19.94
9.97
\0.0001
\0.0001
0.002
B
C
45.49
48.04
35.52
10.59
5.25
\0.0001
\0.0001
\0.0001
0.0014
E
G
AC
AD
AF
AG
0.0235
5.61
0.0193
134.07
9176.75
871.18
1766.96
1645.3
1053.75
871.18
828.25
134.07
9176.75
871.18
1766.96
1645.3
1053.75
871.18
828.25
6.36
0.0128
2
A
435.63
41.36
83.88
78.1
\0.0001
\0.0001
\0.0001
\0.0001
\0.0001
\0.0001
\0.0001
2
B
2
C
2
D
2
E
50.02
41.36
39.32
2
F
2
G
Based on the CCD methodology, there were 151 runs, with different operational
conditions, for deriving the objective function for conversion. Table 4 shows the
analysis of variance (ANOVA) for the response surface reduced quadratic model.
In accordance with regression analysis, the quadratic polynomial model based on
significant levels was fitted with the data as follows.
Y ¼ þ 89:02 À 1:76A þ 1:24B þ 2:65C þ 2:73E þ 2:35G À 1:32AC
2
2
þ 0:93AD þ 0:96AF À 1:02AG À 16:26A À 5:01B
ð1Þ
2
2
2
2
2
À 7:13C À 6:88D À 5:51E À 5:01F À 4:88G
where Y is the response value and A, B, C, D, E, F and G are the coded variables,
and the coefficients of them show their effects on the response and interaction
between them [25].
123

V. Khakyzadeh et al.
Fig. 1 Introduced optimal point of variables
Totally, the value of terms show the amount of their effect; plus or minus signs
indicate a positive or negative effect on Y, respectively. For checking the
significance of each model term, the p value was used as a tool (p values B 0.05
indicate that the model terms are significant). The statistical parameter of correlation
2
2
2
coefficient (R ) of 0.92, R -predicted of 0.82, adjusted R of 0.91, lack of fit F value
of 44.74 and adequate precision of 44.52 ([4 is desirable) reveal that the predicted
values are in good agreement with the experimental values. The Model F value and
p value are 98.3 and \0.0001, respectively, which imply the model is significant.
Optimization of operational condition
In this work, the purpose of reaction optimization is to determine the optimal
variable values at which the reaction conversion is maximized. For this aim, the
desired goal for each variable was chosen within the range except temperature
which was set in the target of 0 °C because of easier operation.
The range of variables and associated optimum values were demonstrated in
Fig. 1. In the optimum condition (A = 0 °C, B = 0.00578 g, C = 2.09 eq.,
D = 3 mL, E = 1.06 mL, F = 0.15 mL, G = 1023 rpm), the value of predicted
conversion is 89.95 %, which is a high conversion rate for this reaction.
Validation of the model
Finally, the validity of the obtained model to predict response values was tested in
the optimum condition. A conversion value of 87 % was obtained from
experimental results which is in good agreement with the predicted response
(
89 %). Thus, the model is valid for results prediction.
1
23

Synthesis, modeling and optimization of cyanide antidote…
Conclusions
In conclusion, we have introduced a new protocol for the synthesis of cyanide
1
antidote (3-methylbutyl) nitrite. Using Fe O @SO H as a nanomagnetic solid acid
3
4
3
catalyst in aqueous media wherein the influence of various factors on the process
was investigated. In accordance with regression analysis, a quadratic polynomial
model can be fitted to the process. Statistical parameters showed that theoretical
values were in good agreement with the experimental values. The promising points
for the presented protocol were efficiency, high conversion, a cleaner reaction
profile and simplicity.
Acknowledgments The authors gratefully acknowledge partial support of this work by the Research
Affairs Office of Bu-Ali Sina University (Grant number 32-1716 entitled development of chemical
methods, reagents and molecules), the Center of Excellence in Development of Chemical Method
(CEDCM), Hamedan, I. R. Iran, Shahid Beheshti University of Medical Science, Tehran, Iran and AJJ-A
University of Medical Science, Tehran, Iran.
References
1. A.H. Hall, B.H. Rumack, Ann. Emerg. Med. 15, 1067 (1986)
2. P. Blance, M. Hogan, K. Mallin, D. Hryhorczuk, S. Hessl, B. Bernard, J. Am. Med. Assoc. 253, 367
(1985)
3
. R. Hartung, Cyanides and nitriles, in Patty’s Industrial Hygiene and Toxicology, 4th edn., ed. by G.D.
Clayton, F.E. Clayton (Wiley, New York, 1994)
. A. Krieg, K. Saxena, Ann. Emerg. Med. 16, 582 (1987)
4
5
. G.J. Hathaway, N.H. Proctor, J.P. Hughers, Hydrogen cyanide. In Proctor and Hughers’ Chemical
Hazards of the Workplace, 4th edn. (Van Nostrand-Reinhold, New York 1996)
. G.D. Hart, M.B. Strauss, P.A. Lennon, D.D. Whitcraft, J. Emerg. Med. 3, 211 (1985)
. F.J. Baud, P. Barriot, V. Toffis, B. Riou, E. Vicaut, Y. Lecarpentier, R. Bourdon, A. Astier, C.
Bismuth, N. Engl. J. Med. 325, 1761 (1991)
6
7
8. K.K. Chen, C. Rose, G. Clowes, JAMA 100, 1921 (1933)
9. J.M. Marraffa, V. Cohen, M.A. Howland, Am. J. Health Syst. Pharm. 69(3), 199 (2012)
1
1
1
1
1
0. J.H. Clark, Green chemistry: challenges and opportunities. Green Chem. 1, 1–8 (1999)
1. L. Xuehui, Z. Dongbin, F. Zhaofu, W. Lefu, Sci. China Ser. B Chem. 49, 385 (2006)
2. H. Liu, G.-H. Tao, D.G. Evans, Y. Kou, Carbon 43, 1782 (2005)
3. V. Polshettiwar, R.S. Varma, Green Chem. 12, 743 (2010)
4. M.A. Zolfigol, V. Khakyzadeh, A.R. Moosavi-Zare, A. Zare, P. Arghavani-Hadi, Z. Mohammadi,
M.H. Beyzavi, S. Afr, J. Chem. 65, 280 (2012)
5. Y. Zhu, L.P. Stubbs, F. Ho, R. Liu, C.P. Ship, J.A. Maguire, N.S. Hosmane, Chem. Cat. Chem. 2, 365
1
(2010)
1
1
1
1
2
2
6. V. Polshettiwar, R.S. Varma, Org. Biomol. Chem. 7, 37 (2009)
7. K.S. Lee, M.H. Woo, H.S. Kim, E.Y. Lee, I.S. Lee, Chem. Commun. 25, 3780 (2009)
8. V. Polshettiwar, A. Moln a´ r. Tetrahedron, 63, 6949 (2007)
9. L.M. Rossi, I.M. Nangoi, N.J.S. Costa, Inorg. Chem. 48, 4640 (2009)
0. M. Ma, Q. Zhang, D. Yin, J. Dou, H. Zhang, H. Xu, Catal. Commun. 17, 168 (2012)
1. Z.R. Lazic, Design of Experiments in Chemical Engineering (Wiley-VCH Verlag GmbH and Co.
KGaA, Weinheim, 2004)
´
1
1
Spectral data of (3-methylbutyl) nitrite: boiling point: 99 °C, H NMR (300 MHz, chloroform-d) d ppm
.91–0.98 (m, 7 H) 1.59–1.74 (m, 2 H) 4.70–4.74 (t, J = 6.60 Hz, 2 H). C NMR (101 MHz,
1
3
0
?
chloroform-d) d ppm 24.99 (s, 1 C) 26.02 (s, 1 C) 37.66 (s, 1 C) 66.73 (s, 1 C). MS: m/z = 117 (M ).
123

V. Khakyzadeh et al.
2
2
2
2. R. Leardi, Anal. Chim. Acta 652, 161 (2009)
3. M.B. Gawande, A.K. Rathi, I.D. Nogueira, R.S. Varma, P.S. Branco, Green Chem. 15, 1895 (2013)
4. M.A. Zolfigol, V. Khakyzadeh, A.R. Moosavi-Zare, A. Rostami, A. Zare, N. Iranpoor, M.H. Beyzavi,
R. Luque, Green Chem. 15, 2132 (2013)
25. T.F. Edgar, D.M. Himmelblau, L.S. Lasdon, Optimization of Chemical Processes, (McGraw-Hill,
New York, 2001)
1
23