ESTABLISHING
STANDARDS
OF C A U S T I C
SODA
GASES .(LPG)
TO R E M O V E
FOR C O N S U M P T I O N
IN T R E A T I N G
LIQUEFIED
MERCAPTANS
O. V. T u k o v , N. N. I v a n o v a , A . N. S a d y k o v , A . M. P o l o t s k i i , a n d N. A. G l e b o v a
UDC
665.54:665.547.931
Caustic soda is one of the primary reagents used in treating hydrocarbon feedstocks to remove mercaptans, and it is also a reagent that is in e x t r e m e l y short supply. The consumption of this reagent is a controlling factor in the economics of producing hydrocarbon feedstocks in petroleum refineries. In curtailing the consumption of caustic soda and hence the quantity of harmful effluents, correct s t a n d a r d i z a tion of caustic consumption in treating various petroleum products is one of the most important factors. In the petroleum refining industry, two procedures are being used in c a l c u l a t i n g the standard for unit consumption of caustic in treating LPG to remove mercaptans. In one of these procedures, the t h e o r e t i c a l standard is c a l c u lated on the basis of mercaptan contents before and after treating; in the second method of calculation, the input data are the i n i t i a l and final concentrations of NaOH solution and the volume of treating solution, with allowance for the incompleteness of the reaction. Both these methods need to be improved, since refinery operating e x p e r i ence has shown that there are substantial differences between theoretically c a l c u l a t e d norms for caustic consumption and the a c t u a l values. These differences are due to the influence of various factors on the LPG d e m e r c a p t a n i z a tion process, these factors not being taken into account in these two procedures. tt is impossible to c a l c u l a t e the required caustic consumption without some manner of accounting for the m e r captan content and the required degree of removal. In a nonrecycie batch process, the caustic solution c o n c e n t r a tion is of great importance, since the c a p a c i t y of a weight unit of caustic decreases with increasing caustic concentration [1]. During the treating process, the caustic solution concentration decreases, not only because of dilution with water formed through interaction of the mercaptans and NaOH, but even more because of dilution with water present in the product being treated. Hence the final caustic concentration should not be considered in c a l c u l a t i n g the consumption standard, since the drop in caustic concentration due to moisture in the product does not have any significant effect on the absolute quantity of caustic soda in the treating system. In order to define more precisely the conditions for caustic treating and thus arrive at a soundly based procedure for standardizing the consumption of caustic soda in LPG d e m e r e a p t a n i z a t i o n , we have carried out certain e x p e r i m e n t a l studies. These experiments were performed on m o d e l mixtures in thermostated glass reactors under an argon blanket (to prevent oxidation of the mercaptan by atmospheric oxygen). The temperature of experiment was 30~ and the v o l u m e ratio of caustic solution to modeI mixture was 0.5 : 1. The components chosen for the m o d e l mixture were n-hexane and ethyI mercaptan. The choice of ethyl m e r c a p t a n was dictated by the fact that about 95% of the total m e r c a p t a n content in LPG consists of this compound. The conditions required to reach equilibrium (mixing and settling time) were established e x p e r i m e n t a l l y . The e q u i librium phases were analyzed by potentiometric titration with a pH-340 potentiometer. In e a c h experiment, the caustic was used as required to give an ethyl m e r c a p t a n content in the treated mixture on the order of 0.02%. This was based on the requirements of TU 38101498-74 relative to the m e r c a p t a n content in b u t a n e / b u t y l e n e cuts, since more than 30~ of the total consumption of caustic soda used in treating LPG is used in d e m e r c a p t a n i z i n g such cuts. A l l - U n i o n Scientific-Research Institute of Hydrocarbon Raw Materials (VNIIUS). Translated from Khimiya i ' Tekhnologiya Topliv i MaseI, No. 11, pp. B3-36, November, 1975. 9 1976 Plenum Publishing Corporation, 227 West 1 7th Street, New York, N. Y. 10011. No part o f this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission o f the publisher. A copy o f this artiele is available from the publisher for $15.00.
869
g!51
e
-
<
I0
1~,
t8
8
gg
Ig
12
XZ, wt. % Fig. 1
I4 18 18 X 1, wt. %
gg
8g
Fig. 2
Fig. 1. Specific consumption of caustic soda Y' (or YI) as a functionofcausticsolutionconcentration X1, with various initial mercaptan contents (in % by weight): 1, 3) 0.33%; 2, 4) 0.22%; 5, 7) 0.11%; 6, 8) 0.07%. O) "Breakthrough" of treated product with respect to mercaptan content; A) "averaging" of treated product with respect to mercaptan content. Fig. 2. Limiting saturation Yz (or Y~) as a function of caustic solution concentration X v Symbols same as in Fig. 1. TABLE I. Levels and Variation of Factors
Factors Level
xo Axj
+1,41 --1,41
xl
x,
15 5 22
0,22 0,11 0,37 0,07
8
TABLE 2. Design Matrix and Results of Experiments r
No.
x,
1
20 I0 20 10 8
2 3 4 5 6 7 8
9 10 11 12 13
870
22
15 15 15 15 15 15 15
Xz
0,33 0,33 0,11 0,11 0,22 0,22 0,07 I 0,37] 0,22] O,22 I 0,22 0,22
0,22
Ya
23,08 12,01 6,37 3,11 9,42 14,77 3,31 16,53 12,02 12,02 10,17 10,17 11,02
15,39 8,41 3,36 1,72
6,59 8,62 1,74 11,02
7,35 7,35 6,01 5,51 7,35
Ye
Y2
0,158 0,254 0,186 0,255 0,298 0,182 0,167 0,197 0,198 0,199 0,256 0,257 0,222
0,251 0,449 0;335 0,658 0,380 0,290 0,360 0,396 0,341 0,341 0,418 0,454 0,341
TABLE 3. Influence of Degree of Treating on Caustic Sod~ Consumption Mercaptan c o n centration, % initial
I Theoretical Theoretical ]standard for [specific c o n ~'~.'-".~ I NaOH c o n - IsumDtion of solutton tsumption as [NaCIH as calc. c o n c e n - - [calc. by cur-l from Eqs. (1) tration, "/olrentoroce. and (2), ldure; kg/ kg/ton Iton ~,tont~
final
!
0,11 0,tl
0,014 0,020
] ]
1,20 1,125
5,66 2,69
Statistical design and analysis of the experiments [2] was used in the present study. The factors governing the degree of treating were taken to be X1, the initial concentration of NaOH solution (in % by weight). The base levels, the intervals of v a r i ation, and the limits of the region of investigation are listed in Table 1. The intervals of variation of the caustic solution c o n centration and the i n i t i a l ethyl m e r c a p t a n concentration are in line with the values observed in c o m m e r c i a l t r e a t m e n t of LPG.
9
~ o,s7 I o, ao 0
E o, e o, as
o,oz
15 t5
i
o,ooz
i
I
o,o~6
I
1
0,o3o
I
I
gg~
X1, wt. 07o Fig. 3. Saturation of caustic solution Y~ as a function of residual concentration of m e r c a p tans in treated product X. 9 Data obtained in treating butane/butylene cut at Syzran' Refinery (XI= 13%); A) laboratory data (XI= 15%).
As the dependent variables we selected YI, the t h e o r e t i c a l specific consumption of caustic (kg/ton*) for "breakthrough, ~ i.e., with a content of ethyl m e r c a p t a n in the treated portion of n-hexane amounting to 0.02 i 0.002% (or in the entire treated mixture amounting to 0.013-0.014% (Variant I); Yi is the t h e oretical specific consumption of caustic (kg/ton) for "averaging," i.e., with the content of ethyl m e r c a p t a n in the entire treated mixture on the order of 0.02% (Variant II); Y2 is the saturation of the caustic solution for "breakthrough," moles S ' - / m o l e NaOH; Y~ is the saturation of the caustic solution for " a v e r a g ing," moles S " / m o l e NaOH. The design matrix in natural units and the results of the experiments are shown in Table 2. The data of Table 2 were used to calculate the coeffi-
cients of the regression equations and the respective errors. After e l i m i n a t i n g the nonsignificant coefficients and testing the equations by the Fisher criterion with a 5~ level of significance, the following regression equations were o b tained, in natural units, adequately describing the experiment: Y1 = 3.58 - - 0.23X 1 - Yi
= 2,42
-- 0.25X
2.98X~
I -- 0.26X
2+
-}-
3.55X 1 X2
2.43XIX
2
(1) (2)
Y~ = 0.350 - - 0.008X 1
(3)
Y2 z 0.624 - - 0.016X 1
(4)
As is welI known, the regression equations can be used to obtain sections of the response surface in coordinates of Y vs Xj for fixed values of the other factors. Graphical relations obtained in an analogous manner are shown in Fig. 1. It can be seen from these plotsthat, with an i n i t i a l content of mercaptans greater than 0.07 or 0.11% (for variants I and II, respectively), the specific consumption of caustic soda increases with increasing initial concentration of the caustic solution. The reverse trend may be noted when the i n i t i a l mercaptan content is less than 0.11%. With an i n i t i a l mercaptan content of 0.07 or 0.11% (for variants I and II, respectively), the specific consumption of caustic soda is very nearly independent of i n i t i a l caustic solution concentration. In the calculation procedures currently used, changes in the degree of t r e a t i n g have very little influence on the value obtained for the theoretical standard for caustic soda consumption. The relationships obtained in the present 9 Metric tons, here and throughout a r t i c l e - Translator.
871
work demonstrate that this influence is highly significant and that it cannot be accounted for on the basis of the difference between initial and final mercaptan concentrations (Table 3). These experiments, together with refinery tests on LPG demercaptanization units, have confirmed that the level of NaOH solution saturation with mercaptans is the most important criterion of the suitability of the solution for further use. These studies have shown that the saturation value, expressed in moles S-" per mole NaOH, does not depend on the initia[ content of mercaptans in the product being treated. This saturation value decreases with increasing initial caustic solution concentration; and, for a given treating level and NaOH solution concentration, the saturation value is constant (Fig. 2). In Fig, 8 we present results obtained in treating a butane/butylene cut at the $yzran' Refinery (caustic solution concentration 18% by weight), these data having been worked up by a least-squares method. Comparative data were obtained in laboratory treating with a caustic solution concentration of 15% by weight. The identical nature of these relationships confirms the correctness of the correlations that have been developed. SUMMARY 1. On the basis of experimental data, regression equations have been obtained, giving an adequate description of the specific consumption of caustic soda as a function of mercaptan content in the original feedstock and the initial concentration of the NaOH solution, under specific conditions of temperature, treating ratio (caustic solution/ feedstock), and degree of treating. 2. The specific consumption of NaOH (kg of 100% NaOH per metric ton of feedstock), for a given degree of treating, is directly related to the initial caustic solution concentration and to the initial mercaptan concentration in the feedstock; hence, both these factors must be taken into account in calculating standards for caustic soda consumption. 3. The limiting saturation of the caustic solution with mercaptans for a given degree of treating (this saturation representing the discard point for the caustic) does not depend on the initial mercaptan content in the feedstock, but is a function of the initial concentration of the caustic solution. 4. With the aim of providing stable quality for LPG meeting the requirements of TU 88101498-74, the existing method for monitoring the depletion level of caustic soda in treating mercaptanTcontaining LPG should be reviewed, on the basis that the primary criterion for depletion is the degree of saturation of the caustic, expressed in moles S-" per mole NaOH. 8. In reviewing the requirements on LPG quality with respect to content of sulfur compounds, from the standpoint of imposing more severe requirements, it must be considered that this will involve a considerable increase in specific consumption of caustic soda. Hence we see a great need for the rapid commercialization of catalytic processes for demercaptanization. LITERATURE CITED 1,
2.
872
Ya. D. Zel'venskii, S. F. Shakhova, and I. V. Dedova, Tr. Gos. Nauchno-Issled. Proekt. Inst. Azot. Prom. Produktov Org. Sinteza (Goskhimizdat, Moscow), 6, 185 (1956). L. P. Ruzinov, Statistical Methods in Optimizing Chemical Processes [in Russian], Khimiya, Moscow (1972).