Purification of 2,4,6-trinitrotoluene by digestion with sodium sulfite and determination of its impurities by gas chromatography–electron capture detector (GC-ECD)

Article abstract of DOI:10.1007/s13738-019-01709-z

Unwanted compounds in 2,4,6-trinitrotoluene (TNT) explosive, such as dinitrotoluene (DNT) and TNT isomers, can lead to exudation defects in the ammunition. For this reason, separation and analysis of impurities are valuable. Chemical treatment with sulfit

Full text of DOI:10.1007/s13738-019-01709-z

Journal of the Iranian Chemical Society
https://doi.org/10.1007/s13738-019-01709-z
ORIGINAL PAPER
Purifcation of 2,4,6‑trinitrotoluene by digestion with sodium sulfte
and determination of its impurities by gas chromatography–electron
capture detector (GC‑ECD)
Mohammad Hariri1 · Kamal Ghani1 · Sajjad Damiri1
Received: 14 November 2018 / Accepted: 6 June 2019
© Iranian Chemical Society 2019
Abstract
Unwanted compounds in 2,4,6-trinitrotoluene (TNT) explosive, such as dinitrotoluene (DNT) and TNT isomers, can lead to
exudation defects in the ammunition. For this reason, separation and analysis of impurities are valuable. Chemical treatment
with sulfte solution is an important method for the purifcation of TNT. In this paper, TNT was purifed by sodium sulfte
solution with putting faked TNT in digestion condition to remove existing impurities, and the efect of four main factors,
digestion temperature, digestion time, concentration of sulfte solution and reaction time of sulfte with impurities on this
purifcation process, was statistically investigated and optimized using central composite design of experiment (CCD) by
Minitab software. Also, the qualifcation of fnal product was characterized by gas chromatography–electron capture detector
method. Here, some conventional compounds such as 2,6-DNT, 2,4-DNT, 3,4-DNT, 2,4,5-TNT (γ-TNT), 2,3,4-TNT (β-TNT)
and aniline derivatives were identifed by standard solutions; then their concentrations were quantitatively determined by
internal standard calibration method.
Keywords Trinitrotoluene · Gas chromatography · Purifcation · Digestion method · Design of experiment (DOE)
Introduction
also TNT can be exuded in various munitions with increasing
temperature in storage condition specially [3–5]. For this rea-
son, researchers have severely sought for a qualitative scheme
to separate and identify the impurities present in military
grade of TNT. Because of the signifcant role of impurities on
physical and chemical properties of TNT, these days, analysis
and identifcation of TNT impurities are a major concern
for explosive producers. Treating by sodium sulfte aque-
ous solution (sellite process) is a standard and cost-efective
method for removing impurities [2]. A nitro group in meta-
positions than methyl group in toluene is exposed to nucleo-
philic attack in the unsymmetrical TNT isomers, and sulfte
anion attacks to a little charge position. Arylsulfonic acids are
fnal product that soluble in the water, as a red water, and can
be washed out with water to purify TNT [4].
2,4,6-Trinitrotoluene (TNT), having three standard types
based on its purity and quality [1], is one of the most com-
mon military explosive materials which is usually prepared
by the nitration of toluene. It has some advantages including
good chemical and thermal stability and safety in handling,
and its low melting point is favorable for melt casting opera-
tions. Unfortunately this method has some major drawbacks
that one of them is the production of dinitrotoluene (DNT)
and unsymmetrical trinitrotoluene isomers (2,4,5-, 2,3,4-,
2,3,6-, 2,3,5- and 3,4,5-TNT) which are produced through
nitration in the meta-position of the toluene ring [2]. These
by-products are indeed able to change TNT various properties
and must be removed as completely as possible. These com-
pounds form eutectics with TNT lead to low melting point
(set point) which are undesirable for military applications and
In overall, purifcation methods of TNT are divided in
two groups mainly, including treatment with aqueous sulfte
solution and with solvents. Diferent works were done for
increasing TNT purity with sulfte solutions such as sodium
sulfte. Increasing the set point from 80.20 to 80.50 [6],
eliminating TNT isomers by adjusting pH [7] and reacting
with TNT, beside TNT isomers, sulfte solution can efec-
tively control the reaction until TNT with high efciency
* Kamal Ghani
kghani@mut-es.ac.ir
1
Department of Chemistry, Malek-Ashtar University
of Technology, Shahin-Shahr, Isfahan, Iran
Vol.:(0123456789)
1 3

Journal of the Iranian Chemical Society
and set point 80.58 °C was obtained [8]. Purifcation process
was carried out by other sulfte solutions such as ammonium
and magnesium sulfte. With ammonium sulfte, impurity
and efciency were 0.3% and 95%, respectively [9] and
with magnesium sulfte the freezing point was obtained at
80.53 °C [10]. Treatment with solvents such as the mixture
of ethylene dichloride and carbon tetrachloride [11], acetone
and water [12], nitric acid [13] and the mixture of nitric and
sulfuric acids was studied [14].
near to 99.9% and 98%, respectively. Also, fuming nitric acid
and sodium sulfte were used in industrial grade.
Instruments
All purifcation experiments were carried out in laboratory
scale, a jacketed agitated glass reactor connected to a water
circulated system (HAAKE W13, Germany) for maintaining
the constant temperature. Gas chromatography (GC) analysis
was used for all experiments. GC analyses were carried out
in Agilent 6890 N model system, which was equipped with a
HP-5 column (5% phenyl methyl siloxane), column capillary
(30.0 m×320 µm×0.25 µm nominal) and electron capture
detector (ECD).
Up to now, various identifcation methods developed to
determine nitroaromatic impurities by using nuclear mag-
netic resonance [15, 16], thin-layer chromatography [17],
thermal methods (TGA, DSC), electrochemistry, UV and IR
spectrophotometry [15, 18], X-ray and mass spectroscopy
[18, 19], and chromatography methods (GC, HPLC, TLC,
GC–MS and LC–MS) [20–27] are tedious, time consuming,
and not exactly reliable for complete separation of all iso-
mers. Among these methods, gas chromatography ofers the
greatest potential for separation of all isomers [27] and is by
far the most efcient and quantitatively accurate as efective
as possible [20, 28, 29].
Design of experiment
Response surface methodology (RSM) is an important method
to design of experiments to study of efective interactions
between experiment parameters to achieve optimum response.
In the present work, central composite design (CCD) was uti-
lized to optimize the purifcation conditions. Four efective
factors on purifcation process of TNT: digestion tempera-
ture (X1), digestion time (X2), concentration of sulfte solu-
tion (X3) and reaction time (X4) of sulfte with impurities,
were optimized by Minitab software. The range of factors
was 60≤X1 ≤75 °C, 15≤X2 ≤45 min, 0.2≤X3 ≤1.0 wt.%,
5≤X4≤15 min. Here, the obtained percentage of extraction of
impurities such as 2,6-DNT(Y1), 2,4-DNT(Y2), 3,4-DNT(Y3),
γ-TNT (Y4) and β-TNT (Y5) are responses which were opti-
mized and modeled using multivariate linear regression.
For obtaining extraction models, analysis of variance
(ANOVA) was performed for statistical analysis and used
for evaluation of data. F value and p value were obtained
in a good value. Adequacy of models was tested by lack of
ft and determination coefcient R2 analysis. From data of
models, contour plot (3D) indicated the factors interaction.
In this work, a process of purifcation was carried out by
digestion of impure solid and faked TNT in sodium sulfte
solution at elevated temperature. Digestion of solid powders
can efectively separate various little impurities in military
grade TNT, with a signifcant reduction in the hazardous
red water compound. In the proposed purifcation process,
efects of four main factors, digestion temperature, digestion
time, concentration of sulfte and reaction time on purif-
cation process, were studied by central composite design
(CCD) of experiment with MINITAB software. All experi-
ments were carried out in laboratory scale, and TNT and
its impurities were analyzed and identifed by gas chroma-
tography–electron capture detector (GC-ECD) method. This
method was employed for determining the various impurities
concentrations such as 2,6-DNT, 2,4-DNT, 3,4-DNT, γ-TNT
and β-TNT which were identifed in the compound.
Purifcation method of TNT
Experimental
To perform TNT purifcation with digestion method and
attain maximum impurity extractions (%) and set point,
10 g of TNT was added to 20 mL distillated water for diges-
tion process in the jacketed glass reactor. Then, TNT was
in digestion conditions according to design of experiment
conditions at specifed temperature and time, to migrate
impurities to the surface of faked TNT. After digestion of
TNT, sodium sulfte solution was added dropwise to suspen-
sion and allowed to reaction time fnished. During separation
method with sodium sulfte, red water was produced as waste
water which should be washed by distillated water. Finally
obtained TNT was dried, weighted and analyzed to deter-
mine extractions (%) of impurity with GC method.
Materials
Flake TNT (type 2 [1]) was purchased from Iranian defense
industries with set point 80.37. 2,6-Dinitrotoluene(2,6-DNT),
2,4-dinitrotoluene(2,4-DNT), 3,4-dinitrotoluene(3,4-DNT)
were purchased from Sigma-Aldrich with purity>98 wt%.
2-Methyl-4-nitroaniline, 2,4-dinitroaniline and 1,2-dini-
trobenzene were purchased from Merck with purity
>99 wt%. 2,4,5-TNT (γ-TNT) and 2,3,4-TNT (β-TNT) were
synthesized by nitration 3,4-DNT and 2,3-DNT, respectively,
as described previously [29]. Chromatography grade acetoni-
trile and sulfuric acid were purchased from Merck with purity
1 3

Journal of the Iranian Chemical Society
concentrations were prepared in the range of 0.25–10 ppm
for each analysis.
Gas chromatography method
Injector and detector temperatures were 225 °C and
300 °C, respectively. Inject volume was 2 µL, and split 1:6
was used for inlet of column. Nitrogen was used as car-
rier gas at constant fow rate of 2 mL/min, and also ECD
as detector. Temperature conditions used for separation
of components, and separations were as follows: initial
temperature was 100 °C (hold for 2 min), frst temperature
ramp (100 °C/min to 200 °C), and second temperature
ramp (20 °C/min to 280 °C), fnally hold for 5.5 min at
280 °C.
Synthesis of β and γ isomers of TNT
2,3,4-TNT and 2,4,5-TNT were synthesized by single-step
nitration from 2,3-DNT and 3,4-DNT, respectively. 0.50 g
of DNT was weighed, and a mixture of 4 mL of 98% sul-
furic acid and 1.5 mL of fuming nitric acid was added to
DNT samples; then temperature was risen to 100–120 °C
and 80–90 °C in 15 and 30 min for synthesizing β and γ
isomers, respectively [30].
Preparation of solutions
Results and discussion
TNT analysis was obtained from injection of 2 µL from
500 µL solution with concentration of 2000 ppm in HPLC-
grade acetonitrile. Internal standard method was used for
quantification of each solution, and 50 µL from 2 ppm
1,2-dinitrobenzene (1,2-DNB) was added to all solutions
for this purpose. A series of standard solutions of various
GC analysis
On the basis of relative vapor pressure and afnity for the
stationary bed in column, the components were separated
in gas chromatography method [31]. For isomers having
similar molecular weight, the compounds exited by polarity
from GC column. Thus, high polarity caused later reten-
tion time for the compound [20]. Table 1 indicates some of
impurities in TNT such as 2,6-DNT, 2,4-DNT, 3,4-DNT,
γ-TNT, β-TNT, 2,4-dinitroaniline and 2-methyl-4-nitroani-
line identifed by GC and their retention times of standard
solutions.
Table 1 Components identifed from purifed TNT
Compounds
Retention
times (min)
2,6-DNT
2,4-DNT
8.4
9.3
1,2-DNB was used for quantifcation as internal standard.
Extractions of impurities were optimized to obtain maxi-
mum of TNT purity. Figure 1 shows a typical chromatogram
of crude TNT and its impurities.
3,4-DNT
α-TNT
γ-TNT
β-TNT
2-Methyl-4-nitroaniline
2,4-Dinitroaniline
1,2-DNB(IS)
10.1
11.3
12.2
12.5
13.1
13.6
8.6
Because of the importance of both DNT and TNT
isomers in exudation behavior of TNT, in this work cali-
bration curves were drawn for both compounds. Stand-
ard curves were prepared with at least four standard
Fig. 1 Chromatogram of TNT
and identifed impurities
1 3

Journal of the Iranian Chemical Society
Table 2 Regression parameters for calibration of TNT impurities
Compound
Linear concentra- Slope
tion range (ppm)
Intercept
R2
2,6-DNT
2,4-DNT
3,4-DNT
γ-TNT
0.10–0.25
1.00–6.00
0.025–2.00
0.25–10.00
0.25–2.00
2.8804
2.2171
8.0255
3.5537
6.9069
0.0057
0.6324
0.0261
0.6198
0.0818
0.9939
0.9942
0.9997
0.9971
0.9985
β-TNT
concentrations. The peak area ratio of samples and the
internal standard were obtained from GC, and calibration
curves were drawn.
Finally, calibration curves were obtained with r2 more
than 0.993. Table 2 shows regression parameters for all the
calibration curves. The calibration curves for 2,4-DNT and
β-TNT are shown in Fig. 2.
Purifcation of TNT by digestion method
Precisely and completely purifcation of TNT was carried
out by digestion method. This method leads to migration of
interior impurities trapped into TNT structure to the surface
by increasing the temperature. This is very suitable method
for purifying TNT, because the production of hazardous red
water is signifcantly reduced and also the physical properties
of fake TNT remain without changing. In the conventional
methods, TNT completely melted in water, leading to signif-
cant degradation of useful TNT isomer) 2,4,6-trinitrotolu-
ene) or lowering of purifcation yield, increasing red water,
and also melting-faking process of standard grade of TNT
should be replicated. Considering that polar components
tend to dissolve in polar solvents and vice versa nonpolar
components dissolve in nonpolar solvents, polarity can afect
Fig. 3 Chromatograms of original TNT (a), after digestion (b) and
after adding sodium sulfte (c)
the solubility and migration behavior of impurities. It seems
in this purifcation process, impurities have higher solubil-
ity relative to TNT. Solubility of 2,4-DNT was 975 mg/L,
whereas TNT was 641 mg/L in water at 61 °C. Also, polarity
of 2,6-DNT, 2,4-DNT, γ-TNT and β-TNT were reported 3.48,
5.38, 4.45 and 8.35 Debye, respectively, while for TNT is
0.37 Debye [20, 32, 33]. In this method, impurities of TNT
can be extracted and dissolved in water and sulfte solution.
The performance of proposed purification process
can be observed according to Fig. 3a–c, indicated three
Fig. 2 Calibration curves for a 2,4-DNT and b β-TNT
1 3

Journal of the Iranian Chemical Society
Table 3 Design of experiment
and data for extraction of 2,4-
DNT(Y2) and β-TNT(Y5)
Run order X1 (°C) X2 (min) X3 (%W/W) X4 (min) Exp. Y2
Pred. Y2 Exp. Y5
Pred. Y5
1
75.0
60.0
75.0
60.0
60.0
60.0
75.0
75.0
67.5
60.0
67.5
67.5
67.5
67.5
75.0
67.5
67.5
67.5
75.0
67.5
67.5
67.5
75.0
60.0
67.5
67.5
60.0
75.0
75.0
60.0
60.0
45
15
15
15
15
45
15
15
30
45
30
30
30
30
45
30
15
30
45
45
30
30
30
45
30
30
30
45
15
45
15
1.0
1.0
0.2
1.0
0.2
1.0
1.0
1.0
0.6
1.0
0.6
0.6
0.6
0.6
1.0
0.6
0.6
0.6
0.2
0.6
0.6
1.0
0.6
0.2
0.2
0.6
0.6
0.2
0.2
0.2
0.2
5
15
5
29.8086 23.0330 13.2617 12.8960
25.1503 24.2521 28.2561 26.8832
2
3
2.8253
0.9726 26.8509 25.0334
4
5
6
7
5
13.9021 14.3645 49.5325 46.6212
22.8094 19.9154 27.7350 28.4883
25.4498 27.5324 34.4867 33.2879
15.4149 15.1970 45.7766 44.7480
15
15
15
5
8
5.3395
5.3094 29.4150 31.4433
9
10
5
40.9609 38.9218 41.1902 41.7633
18.0960 17.6448 35.0766 35.8831
32.7233 36.6694 39.9671 42.6536
40.3928 38.9218 42.4141 41.7633
36.2758 38.9218 43.8158 41.7633
39.8381 38.9218 40.3341 41.7633
33.2113 32.9206 41.9459 43.3435
41.5510 38.9218 43.5201 41.7633
34.4025 33.6708 42.9250 44.7990
20.9184 26.7818 34.1262 33.2913
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
15
10
10
10
15
10
10
5
5
17.0952 18.6963
5.9636
6.4860
10
10
10
10
15
10
10
10
15
15
5
44.4843 44.1728 40.3705 38.7277
39.0749 38.9218 43.8158 41.7633
40.6109 41.0902 44.7158 47.3606
20.3287 24.1428 30.2757 31.3013
28.6754 23.1957 34.3679 34.8930
38.7018 36.7534 45.7514 44.9582
41.6561 38.9218 41.1876 41.7633
19.9809 25.9763 33.0364 33.8624
25.1309 28.5839 46.1365 44.9486
10.5615 10.8603 46.9274 46.3532
12.5625 13.3080 29.1397 29.4731
9.5905 10.0277 37.9726 40.2112
5
Table 4 ANOVA for modeling
2,4-DNT
Source
of 2,4-DNT extraction
DF
Seq SS
Adj SS
Adj MS
F
p
Regression
Linear
Square
7
4
2
7199.92
1036.01
2947.38
208.61
242.44
227.47
20.97
4199.92
1036.01
2947.38
208.61
242.44
227.47
20.97
599.00
259.00
1474.18
208.61
10.54
13.03
3.49
56.83
0.000
0.000
0.000
0.000
24.57
139.86
19.79
Interaction
1
Residual error
Lack of ft
Pure error
Total
23
17
6
3.73
0.056
30
4435.43
chromatograms for original TNT, after digestion step and
after adding sodium sulfte step. For both experiments, tem-
perature and time of digestion were 68 °C and 30 min. It can
be seen that the intensity of the peaks in chromatogram (b)
more decrease than (a) which proving digestion process that
can extract the impurities in water without adding of sodium
sulfte. In Fig. 3(c), the highest decrease is observed in peak
intensity of DNT isomers after digestion and adding sodium
sulfte solution.
1 3

Journal of the Iranian Chemical Society
Table 5 ANOVA for modeling
β-TNT
Source
of β-TNT extraction
DF
Seq SS
Adj SS
Adj MS
F
p
Regression
Linear
Square
Interaction
Residual error
Lack-of-ft
Pure error
Total
11
4
3
2770.81
615.80
644.09
1510.92
69.11
56.68
12.43
2770.81
615.80
644.09
1510.92
69.11
56.68
12.43
251.89
153.95
214.70
377.73
3.64
4.36
2.07
69.25
42.32
59.02
0.000
0.000
0.000
0.000
4
103.85
19
13
6
2.11
0.185
30
2839.92
Table 6 Interaction and coefcient of factors for modeling of 2,4-
exudation behavior of TNT, and other impurities are pre-
sented in Table 3.
DNT extraction
ANOVA tables consist of sum of squares of the errors; F
statics and p value are given in Tables 4 and 5 for 2,4-DNT and
β-TNT, respectively. In multivariate linear regression analysis,
the efects of linear terms, squared terms and two interaction
factors of extraction were investigated (see Tables 6, 7).
Equations 1–5 show multivariate linear regression models
for extraction of 2,6-DNT(Y1), 2,4-DNT(Y2), 3,4-DNT(Y3),
γ-TNT(Y4) and β-TNT(Y5), respectively.
2,4-DNT (Y2)
Term
Coef.
38.922
SE Coef.
t
P
Constant
X1
X2
X3
X4
X1X1
X4X4
X1X2
0.940
0.765
0.765
0.765
0.765
1.74
41.43
0.000
0.243
0.000
0.009
0.000
0.000
0.000
0.000
0.917
5.251
2.168
−1.20
6.86
2.83
6.46
4.944
−13.86
−7.20
3.611
−7.95
−4.13
4.45
Y1 = 939.7 − 29.45 X1 + 2.935 X2 + 29.54 X3 − 0.57 X4
+ 0.2257 X1X1 − 0.02289 X2X2 − 0.1898 X4X4
1.74
0.812
(1)
− 0.02265 X1X2 − 0.497X1X3 + 0.0745X1X4
+ 0.3009 X2X3 − 0.02412 X2X4
Table 7 Interaction and coefcient of factors for modeling of β-TNT
extraction
Y2 = − 1063 + 32.18X1 − 1.816X2 + 5.42X3 + 6.75X4
− 0.2464X1X1 − 0.2878X4X4 + 0.03210X1X2
β-TNT (Y5)
(2)
Term
Coef.
SE Coef.
t
P
Constant
X1
X2
X3
X4
X1X1
X3X3
X4X4
X1X2
X1X4
X2X4
X3X4
41.763
−1.281
−3.036
1.201
4.681
−9.18
4.40
−3.79
−1.952
8.261
4.286
−2.004
0.561
0.450
0.450
0.450
0.450
1.13
1.13
1.13
0.477
0.477
0.477
0.477
74.45
0.000
0.010
0.000
0.015
0.000
0.000
0.001
0.003
0.001
0.000
0.000
0.000
Y3 = − 397 + 11.31X1 + 2.376X2 − 57.4X3 + 12.88X4
− 0.0753X1X1 − 0.01718X2X2 + 38.65X3X3 − 0.2819X4X4
− 0.02844X1X2 − 0.0993X1X4 + 0.647X2X3
−2.85
−6.75
2.67
10.41
−8.12
3.89
−3.35
−4.09
17.33
8.99
+ 0.02472X2X4 − 0.716X3X4
(3)
Y4 =647.9 − 18.94X1 + 2.125X2 + 51.2X3 − 10.967X4
+ 0.1402X1X1 − 0.02626X2X2 − 116.82X3X3 + 0.6106X4X4
− 0.01127X1X2 + 1.710X1X3 − 0.3161X2X3
+ 0.03645X2X4 − 2.273X3X4
(4)
−4.20
Y5 = − 576.1 + 20.18X1 + 0.398X2 − 19.95X3 − 12.01X4
− 0.1632X1X1 + 27.48X3X3 − 0.1516X4X4 − 0.01735X1X2
Modeling of purifcation process
Analysis of variance
+ 0.2203X1X4 + 0.05714X2X4 − 1.002X3X4
(5)
where X1 is digestion temperature (°C), X2 digestion time
(min), X3 concentration of sulfte (%W/W), X4 reaction time
Experimental and predicted data for the extraction of 2,4-
DNT(Y2) and β-TNT(Y5), having higher importance in
1 3

Journal of the Iranian Chemical Society
Fig. 4 3D plots of 2,4-DNT
(min), Y1 extraction (%) for 2,6-DNT Y2 extraction (%) for
2,4-DNT, Y3 extraction (%) for 3,4-DNT, Y4 extraction (%)
for γ-TNT Y5; extraction (%) for β-TNT.
temperature and digestion time were obtained at middle
range and initially range, respectively.
In the validation of models, value of R-sq 97.29% and
R-sq(adj) 95.48% for response Y1, value of R-sq 94.53%
and R-sq(adj) 92.87% for response Y2, and R-sq 95.56%,
R-sq(adj) 92.16% for response Y3, R-sq 98.12% and R-sq(aj)
96.69% for Y4 and R-sq 97.57% and R-sq(adj) 96.16% for
Y5 were obtained. p value should be at level p<0.05, and F
value should be greater. A p>0.05 could not be acceptable
for model because that is not insignifcant.
Optimization of responses
Using verified models obtained for extractions of DNT
isomers, optimization was carried out for reaching to
maximum extraction of DNTs. Optimum conditions
were obtained as follow: for 2,6-DNT digestion tem-
perature 75.0 °C, digestion time 21.96 min, concentra-
tion of sulfite 0.20 w/w% and reaction time 11.86 min,
for 2,4-DNT digestion temperature 68.18 °C, digestion
time 45.0 min, sulfite concentration 1.0 w/w% and reac-
tion time 11.76 min, for 3,4-DNT digestion temperature
60.0 °C, digestion time 45.0 min, sulfite concentration
1.0 w/w%, reaction time 12.97 min. For γ-TNT digestion
temperature 75 °C, digestion time 31.36 min, concentra-
tion of sulfite 0.57 w/w%, time of reaction 15 min and for
β-TNT digestion temperature 69.54 °C, time of digestion
45 min, concentration of sulfite 0.2 w/w%, time of reac-
tion 15 min were obtained.
Response surface plot
In Figs. 4 and 5, three-dimensional plots were represented.
Response surface plots showed two variable interaction
variations while the other variables kept a constant mid-
dle value. For 2,4-DNT digestion temperature indicated the
maximum of response at medium range with variation other
factors. Digestion time and concentration of sulfte showed
maximum of response at high level. For β-TNT digestion
1 3

Journal of the Iranian Chemical Society
Fig. 5 3D plots of β-TNT
9. E.E. Gilbert, US Patent 3956409 (1976)
Conclusions
10. E.E. Gilbert, US Patent 4003953 (1977)
11. W.H. Rinkenbach, US Patent 1936607 (1933)
12. D.C. Warren, US Patent 1964792 (1934)
13. M.A. Spalding, L.F. Albright, Ind. Eng. Chem. Process Des. Dev.
24, 1010 (1985)
GC-ECD is a suitable method for analyzing nitroaromatic
compounds in TNT easily and completely. In this work, the
most efective temperature level for immigration impurities
to the surface of TNT structure was found that can improve
the process speed and operability. In the digestion process,
impurities were exited in a good proportion accompanied
by high efciency. DNT and TNT isomers were identifed
in TNT, and the concentrations of impurities were detected
by internal standard for TNT samples. Response surface
method was employed for optimizing of impurities extrac-
tion by sulfte solution. DNTs were decreased in treatment
with sulfte solution. Optimum conditions were obtained for
extraction of impurities with desirable statistically data in
purifcation of TNT.
14. A. Wyler, A. Joseph, US Patent RE20, 926 (1938)
15. D. Gehring, Anal. Chem. 42, 898 (1970)
16. D. Gehring, G.S. Reddy, Anal. Chem. 40, 792 (1968)
17. S.K. Yasuda, J. Chromatogr. A 13, 78 (1964)
18. T.C. Castorina, DTIC Document (1980)
19. T.F. Jenkins, R.P. Murrmann, D.C. Leggett, J. Chem. Eng. Data
18, 438 (1973)
20. D. Gehring, J. Shirk, Anal. Chem. 39, 1315 (1967)
21. R. Dalton, J. Kohlbeck, W. Bolleter, J. Chromatogr. A 50, 219
(1970)
22. J. Kohlbeck, C. Chandler, W. Bolleter, J. Chromatogr. A 46, 173
(1970)
23. X. Zhao, J. Yinon, J. Chromatogr. A 946, 125 (2002)
24. R. Waddell, D.E. Dale, M. Monagle, S.A. Smith, J. Chromatogr.
A 1062, 125 (2005)
25. H. Brust, S. Willemse, T. Zeng, A. van Asten, M. Koeberg, A. van
der Heijden, A. Bolck, P. Schoenmakers, J. Chromatogr. A 1374,
224 (2014)
References
26. M.E. Walsh, Talanta 54, 427 (2001)
27. J.S. Parsons, S. Tsang, M. Digiaimo, R. Feinland, R. Paylor, Anal.
Chem. 33, 1858 (1961)
1. S. Kaye, Encyclopaedia of Explosives and Related Items (Dover,
NJ, 1980)
28. T.L. Chang, Anal. Chim. Acta 53, 445 (1971)
29. M. Rowe, J. Chromatogr. Sci. 4, 420 (1966)
30. W.H. Dennis Jr., D.H. Rosenblatt, W.G. Blucher, C.L. Coon, J.
Chem. Eng. Data 20, 202 (1975)
2. T. Urbański, Chemistry and Technology of Explosives (Pergamon
Press, Oxford, 1984), pp. 160–185
3. J.G. Harold, M. Louis, and K.P. Otis, US Patents 3536544 (1970)
4. R.W. Millar, A.W. Arber, R.M. Endsor, J. Hamid, M.E. Colclough,
J. Energ. Mater. 29, 88 (2011)
31. H.M. McNair, J.M. Miller, Basic Gas Chromatography (Wiley,
Hoboken, 2011), pp. 55–57
5. P. Guo-zhi, Chin. J. Explos. Propell. 6, 12 (2008)
6. J.A. Wyler, R.N. Boyd, US Patent 2297733 (1942)
7. P.D. George, US Patent 1975598 (1934)
32. J.M. Phelan, J.L. Barnett, J. Chem. Eng. Data 46, 375 (2001)
33. P. Chen, W. Lo, K. Hu, J. Mol. Struct. 389, 91 (1997)
8. M.F. Acken, R.M. Cavanaugh, US Patent 2380248 (1945)
1 3