Study of the Pyrolysis Process of the 60-95 °C Fraction of Luginets Gas Condensate Using the Simplex Design Method
Abstract Generated abstract
This study investigates laboratory pyrolysis of the 60 to 95 °C fraction of gas condensate from the Luginets field in a flow reactor, with the aim of identifying an ethylene oriented operating regime. A simplex experimental design was applied to vary reactor outlet temperature, steam dilution, and feed rate, successively reflecting the least favorable points to approach improved ethylene yields. The experiments indicated an optimum region with appreciable curvature in the response surface, and statistical comparison supported the selected conditions. The reported optimal regime was 835 °C, 1.45 ml/min feed rate, 15 wt. percent steam dilution, and about 0.4 s contact time, giving an ethylene yield of 35.0 wt. percent.
Full Text
STUDY OF THE PYROLYSIS PROCESS OF THE 60–95°C FRACTION OF LUGINETS GAS CONDENSATE USING THE SIMPLEX DESIGN METHOD
M. D. MEDVEDEV, S. I. SMOLYANINOV, A. V. KRAVTSOV
(Presented by the scientific-methodological seminar of the organic departments of the Faculty of Chemical Technology)
A study of the pyrolysis of the 60–95°C fraction of condensate from the Luginets field was carried out on a laboratory unit with a flow-type reactor. The aim of the investigations was to find an “ethylene” regime and to test the simplex design method in studies of this type.
The simplex design procedure was selected from the literature [1, 2]. Table 1 shows the initial levels and the intervals of their variation. On the basis of the initial levels and variation intervals, the first series of experiments was planned. These are experiments 1, 2, 3, and 4 in Table 2.
Table 1.
Basic level and variation intervals of the planned experiments
| Variable parameters | Basic level | Variation intervals |
|---|---|---|
| Temperature at the reactor outlet, °C | 750 | 30 |
| Dilution with steam, wt. % | 30 | 20 |
| Feed rate of raw material into the reactor, ml/min | 1.3 | 1 |
After carrying out the initial series of experiments, it was found that the worst value of the quality criterion was obtained in the third experiment (13.2 wt. %). The worst point of the simplex is discarded, and the coordinates of the new—fifth—point, which is the reflection of the third, are found by the formulas [3]. The fifth point together with the remaining ones forms a new simplex (1, 2, 5, 4), whose center is shifted toward improvement of the process quality. In the fifth experiment, the ethylene yield was 27.2 wt. %.
Reflection of the worst points continued up to the fifteenth experiment, in which the numerical value of the quality criterion reached 35.0 wt. %. Then follow experiments 16, 17, 18, and 19, in which the ethylene yields are lower; moreover, it is evident that the simplex has begun to rotate, and this confirms the proposition that the optimum region has been reached. To verify this, four experiments were carried out at the center of the last simplex (simplex 12, 8, 11, 15). From the results of experiments 20, 21, 22, and 23, the aver-
Table 2
Planning of experiments in the study of the 60–95°C fraction of condensate from the Luginets field
| Experiment vertex | Initial simplex | Reflected point | Feed rate, ml/min | Dilution with steam, wt. % | Temperature, °C | Ethylene yield, wt. % |
|---|---|---|---|---|---|---|
| 1 | 1,2,3,4 | — | 1,50 | 40 | 760 | 20,1 |
| 2 | 1,2,3,4 | — | 1,50 | 20 | 760 | 20,4 |
| 3 | 1,2,3,4 | — | 1,50 | 30 | 735 | 13,2 |
| 4 | 1,2,3,4 | — | 0,70 | 30 | 750 | 18,7 |
| 5 | 1,2,5,4 | 3 | 0,96 | 30 | 780 | 27,2 |
| 6 | 1,2,5,6 | 4 | 2,00 | 30 | 785 | 25,6 |
| 7 | 7,2,5,6 | 1 | 1,50 | 15 | 790 | 24,0 |
| 8 | 7,8,5,6 | 2 | 1,50 | 30 | 810 | 27,6 |
| 9 | 9,8,5,6 | 7 | 1,50 | 45 | 795 | 26,0 |
| 10 | 9,8,5,10 | 6 | 0,70 | 40 | 805 | 30,2 |
| 11 | 11,8,5,10 | 9 | 0,65 | 20 | 810 | 28,2 |
| 12 | 11,8,12,10 | 5 | 1,00 | 30 | 840 | 31,2 |
| 13 | 11,13,12,10 | 8 | 0,06 | technological limitation | technological limitation | technological limitation |
| 14 | 12,8,10,14 | 11 | 1,50 | 45 | 825 | 27,9 |
| 15 | 12,8,11,15 | 10 | 1,45 | 15 | 835 | 35,0 |
| 16 | 12,16,11,15 | 8 | 0,60 | 15 | 845 | 27,0 |
| 17 | 12,8,15,17 | 11 | 1,40 | 30 | 845 | 29,6 |
| 18 | 18,8,11,15 | 12 | 1,40 | 15 | 795 | 21,1 |
| 19 | 11,8,19,12 | 15 | 0,62 | 40 | 800 | 28,6 |
| 20 | 12,8,11,15 | — | 1,20 | 25 | 825 | 27,4 |
| 21 | 12,8,11,15 | — | 1,20 | 25 | 825 | 26,8 |
| 22 | 12,8,11,15 | — | 1,20 | 25 | 825 | 27,2 |
| 23 | 12,8,11,15 | — | 1,20 | 25 | 825 | 27,1 |
the arithmetic value of the ethylene yields \(y_0=27,1\%\) wt. In this case the standard deviation is \(\sigma(y)=0,25\). Thus, the value of the response function at the center of the simplex is
\[ y_0=(27,1+0,25)\ \%\ \text{wt.} \]
The expected value of it, determined as the arithmetic mean over the points of the last simplex, is \(y_c=30,5\%\) wt. According to the method [3, 4], the difference \((y_c-y_0)\) is a measure of the nonlinearity of the response surface. In the present case \(y_c-y_0=30,5-27,1=3,4\). The value found is much greater than the value of the standard deviation indicated above; this indicates that the response surface has significant curvature. The Student coefficient, determined from the table [5], is \(t_\alpha=3,182\), while the calculated \(t=27,2\); the second value is much greater than the first, which once again confirms the correctness of the optimal parameters found.
As a result of the studies carried out, the parameters of the optimal “ethylene” regime for the 60–95°C fraction of Luginets condensate were found (experiment 15, Table 2):
Temperature at the reactor outlet — 835°C.
Feed rate — 1,45 ml/min.
Dilution with steam — 15% wt.
Calculated contact time is — 0,4 sec.
Conclusions
- A study has been carried out of the pyrolysis of the 60–95°C fraction of condensate from the Luginets field.
-
It has been shown that the simplex design method gives good results in studies of this type.
-
The optimal parameters for conducting the “ethylene” regime have been found; at the specified severity, the ethylene yield was 35.0% by weight.
REFERENCES
-
Nalimov V. V., Chernova N. A. Statistical Methods for Planning Extreme Experiments. Moscow, “Nauka,” 1965.
-
New Ideas in the Design of Experiments. Edited by V. V. Nalimov. Moscow, “Nauka,” 1969.
-
Gorsky V. G., Brodsky V. V. “Zavodskaya laboratoriya,” 7, 831, 1965.
-
Nalimov V. V. “Zavodskaya laboratoriya,” 7, 854, 1966.
-
Rumshisky L. V. Mathematical Processing of Experimental Results. Moscow, “Nauka,” 1971.