Alkylation Studies

1

Alkylation

Studies

G. M. KRAMER

Downloaded by UNIV OF TASMANIA on May 3, 2015 | http://pubs.acs.org Publication Date: June 1, 1977 | doi: 10.1021/bk-1977-0055.ch001

Exxon Research and Engineering Co., P.O. Box 45, Linden, NJ 07036

The a l k y l a t i o n of C , C and C olefins with isobutane i n s u l f u r i c acid i s a large scale commercial process. The reaction has been extensively studied and i t s general features are well understood and documented i n several l u c i d reviews (1-3). In addition to being of commercial i n t e r e s t , the reaction has stimulated much fundamental research that i s of enormous value i n carbonium ion theory. The o v e r a l l reaction i s usually viewed as proceeding through an ionic chain mechanism i n which a t - b u t y l ion adds to an o l e f i n forming a larger cation that subsequently abstracts a hydride ion from isobutane forming product and a new t - b u t y l i o n . Through the use of isotopic l a b e l i n g , much has been learned about the behavior of the ions while i n the acid (4-11). Nevertheless, there are a number of important questions about the a l k y l a t i o n reaction which have not been dealt with as extensively as might be desired and some of these will be discussed i n this paper. One of these i s the question of where does the reaction occur? I t i s often assumed that a l k y l a t i o n proceeds i n the bulk acid phase (12a), but one of the aims of t h i s report i s to show that a l k y l a t i o n must proceed i n at least two phases and that the reaction occurring at the hydrocarbon-acid interface i s by f a r the most important i n c o n t r o l l i n g the quality or s e l e c t i v i t y of alkylate; i . e . the formation of a C3 f r a c t i o n from isobutane plus butènes while minimizing the production of side products. The fact that a l k y l a t i o n does not occur uniformly throughout the acid has been recently suggested by Doshi and Albright (12b). A second question i s what might be done to improve s e l e c t i v i t y beyond the usual practice of r e f i n e r s to maximize mixing, maximize the isobutane to o l e f i n r a t i o , lower the temperature and reduce the o l e f i n space v e l o c i t y . One approach i s to decide what's rate determining and then to develop a chemical solution. This paper w i l l be concerned with developing evidence that hydride transfer from a t e r t i a r y p a r a f f i n i s generally slow and may be considered to be the rate determining step. The fact that a cation abstracts H® from isobutane r e l a t i v e l y slowly compared to 3

4

5

1

In Industrial and Laboratory Alkylations; Albright, L., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1977.


2

INDUSTRIAL

A N DLABORATORY

ALKYLATIONS

other reactions i t may undergo (deprotonation, racemization and c e r t a i n isomerizations), can also be deduced from the exchange studies (4-11). Doshi and Albright, and e a r l i e r Hoffmann, Schriesheim and t h i s author (13) have recognized that a l k y l a t i o n performance i s related to the presence of o i l soluble hydrocarbons, commonly c a l l e d red o i l or conjunct polymers. These species are usually considered to be saturated and unsaturated cations which can function as intermediates i n the transfer of hydride ions from isobutane to other a l k y l cations. Assuming that hydride transfer i s a l i m i t i n g factor, the discovery of means to augment the rate should r e s u l t i n improved a l k y l a t i o n . This report deals with research which has led to the successful application of c a t i o n i c surfactants for t h i s purpose i n commercial plants. The report f i r s t summarizes hydride transfer information determined during: a. the reaction of t-butyl chloride with methylcyclopentane, b. the reaction of 2,3,4-trimethylpentane i n t r i t i a t e d s u l f u r i c acid, and c. the isomerization of 3-methylpentane i n t r i t i a t e d βμΙίυΓίΰ acid. Some of the reactions have been conducted with "acid modifiers present and t h e i r r o l e i s believed to be primarily to change the s t a b i l i t y and hence r e a c t i v i t y of the c a t i o n i c i n t e r mediates. Some additives have a marked a b i l i t y to increase the "steady state" hydride transfer rate and these have proved useful i n improving the commercial a l k y l a t i o n reaction. The routine use of the additives has followed p i l o t plant and commercial studies which indicate improved a l k y l a t i o n s e l e c t i v i t y under well mixed or poorly mixed conditions and a simultaneous reduction i n acid consuming side reactions. Typical data from these tests w i l l also be presented. 11

The Reaction of t-Butyl Chloride with Methylcyclopentane The major products of the commercial a l k y l a t i o n of isobutane with butènes are trimethylpentanes. This indicates that the products of a l k y l a t i o n are k i n e t i c a l l y controlled because thermodynamics would predict a minor proportion of trimethylpentanes i f the octanes were to isomerize to equilibrium. The p r o b a b i l i t y that hydride transfer from isobutane to carbonium ion intermediates i s the k i n e t i c a l l y slow step a f f e c t i n g product quality was raised by an experiment i n which a stream of 96 percent H2SO4 was rapidly mixed with a stream of isobutane plus isobutylene i n a mixing tee, a f t e r which the products were immediately quenched i n a large vessel containing cold caustic (13). The C3 f r a c t i o n of the product contained trimethylpentenes but no trimethylpentanes.


1.

KRAMER

Alkyhtion

Studies

3

In view of these results a study of the reaction of t-butyl chloride,which should act as an ion source,with methylcyclopentane which should be an e f f e c t i v e hydride donor was undertaken. The reaction was followed manometrically by observing the pressure generated a f t e r the halide was added to a s t i r r i n g emulsion of methylcyclopentane and 96 percent H SO^ (14). By using high s t i r r i n g rates i t i s possible to operate under conditions i n which the k i n e t i c s of the reaction can be studied i n a regime not controlled by mass transfer l i m i t a t i o n s . Figure 1 shows a t y p i c a l curve of the generation of pressure with time i n a well mixed system. In t h i s experiment t-butyl chloride was added to an 0.1 M emulsion of methylcyclopentane and the acid being s t i r r e d at 1000 rpm. Three d i s t i n c t regions of pressure generation were detected. F i r s t there i s an immediate r i s e when t-butyl chloride contacts the emulsion. This i s followed by a t r a n s i t i o n period leading to the t h i r d region i n which there i s a nearly l i n e a r pressure r i s e . The t r a n s i t i o n l a s t s about 0.5 minutes and i s probably strongly associated with HC1 evolution. By connecting the apparatus d i r e c t l y to a time of f l i g h t mass spectrometer, i t was found that the i n i t i a l hydrocarbon product was isobutylene, not isobutane. After about 10 seconds, however, the gas was mainly isobutane, and o l e f i n was no longer detected. The other major gaseous product i s HC1 but adsorption on the spectrometer walls made i t impossible to quantitatively determine the HC1 concentration. The slope of region I I I extending from about 0.75 to 2.0 minutes i n this experiment i s taken as the rate of the hydride transfer reaction. A conventional k i n e t i c analysis showed the rate to be f i r s t order i n t-butyl chloride and f i r s t order i n the concentration of methylcyclopentane i n the emulsion, Figures 2 and 3. It was also found that the methylcyclopentane concentration i n the acid phase was about 60 ppm. An order of magnitude calcul a t i o n indicates that the d i f f u s i o n of methylcyclopentane into the bulk acid phase occurs much faster than the rate of formation of isobutane. Thus the acid phase should be considered to be saturated with methylcyclopentane throughout the reaction i n a l l the k i n e t i c experiments. The fact that hydride transfer shows a dependence on the methylcyclopentane concentration i n the emulsion i s consistent with a reaction occurring at the acid-hydrocarbon interface. It would be inconsistent with reaction occurring only i n the bulk acid phase since there the methylcyclopentane concentration i s constant. The hydride transfer reaction was found to have an a c t i v a t i o n energy of 9.9 Kcal/mole. Transfer from methylcyclopentane occurred much more e a s i l y than from methylcyclohexane where an Ea of 21.3 Kcal/mole was obtained. The difference between the two hydride donors can be r a t i o n a l i z e d by noting that s o l v o l y s i s of the corresponding 1-chloro-l-methylcycloalkanes favors the cyclopentyl


2


INDUSTRIAL

A N DLABORATORY

ALKYLATIONS

120 100 III Hydride Transfer Region ^

βομ 60 h

20

H S 0 = 25000., 1 0 ° C . 2


Proton Elimination Butylène Forms I

4

MCP, m/l t-C.H CI, m/l , 0.1 * *0.2 0

I

L

t

TIME, min. Figure 1. Hydride transfer appears in a distinct region of a pressure-time plot. The concentrations of MCP and t-C^HgCl in the H SO (96% ) emulsion are 0.1M and 0.2M. 2

h

ο 5

0.04

0.08

0.16

0.24 0.32 0.4

[ t - C H C l ] , m/l 4

Figure 2.

9

Hydride transfer isfirstorder in [t-C^HgCl]


1.

KRAMER

Alkylation

5

Studies

system by about 4 Kcal/mole and that i n a bimolecular t r a n s i t i o n state between t-C^Hg* and methylcyclohexane strong s t e r i c i n t e r actions between the 3-5 hydrogens and the t-butyl ion develop that should raise the t r a n s i t i o n state energy considerably. The l a t t e r interactions are not present i n the methylcyclopentane reaction. The preceding data are consistent with hydride transfer being predominantly an i n t e r f a c i a l reaction i n s u l f u r i c acid. This view i s supported by studies of the e f f e c t of acid modifiers on the reaction i n 96% H2SO4. Table I shows the e f f e c t of halogenated acetic acids, methanesulfonic acid and water upon the hydride transfer rate. The rates were estimated i n two ways and a l l rates are r e l a t i v e to that i n 96 percent H^O^.


Table I Medium Effect on Relative Hydride Transfer Rates Vol. %

Vol. %

A

Β

Additive

CH3COOH

0 10 20 30

1.00 0.74 0.34 0

1.0 0.49 0.31 0.13

CF C00H( )

0 2 6 20

1.00 2.31 2.47 2.39

CH C1C00H

0 4 12 20

1.00 1.23 1.18 1.31

1.00 1.28 1.18 1.05

CH3SO3H

0 2 10 20

1.00 1.15 1.44 0.73

1.00 1.04 1.09 0.52

CHC1 C00H

0 4 12 20

1.00 1.06

1.00 1.26 1.87 1.90

H S 0 wt % 96.6 96 90* 85*

1.31 1.00 0.62 0.37

1.28 1.00 0.41 0.26

80*

0.15

0.19

Additive

2

2

CCI3COOH

0 4 12 20

1.00 1.68 1.63 1.85

a

3

2

A

4

A Β

Estimated from dp/dt slopes. Estimated from t o t a l gas evolution i n 1 minute.

(a) (*)

14°C. Product i s o l e f i n i c .



INDUSTRIAL

0.0016

0.004

Figure 3.

"0

ι

ALKYLATIONS

0.04 [MCP],

>l

A N D LABORATORY

0.4

m/l

Hydride transfer is first order in MCP

I

20

ι

I

ι

1

40

1

60

Wt. % H S 0 2

1

80

1

1

100

4

Figure 4. Typical acidity functions for modified sulfuric acid, (a) H. H. Hyman and R. A. Garber, J. Am. Chem. Soc, 81, 1847 (1959); (b) N. F. Hall and W. F. Spengeman, J. Am. Chem. Soc., 62, 2487 (1940); (c) L. P. Hammett and A. /. Deyrup, J. Am. Chem. Soc, 54, 2721 (1932); L. P. Hammett and M. A. Paul, J. Am. Chem Soc, 56, 827 (1934).


1.

KRAMER

Alkyhtion

Studies

7

The r e l a t i v e rates were estimated both from dp/dt slopes and from the t o t a l gas evolved a f t e r 1 minute. The data indicate that as the acid i s weakened by addition, the hydride transfer reaction slows down. On the other hand, the addition of many of the halogenated acetic acids and methanesulfonic acid r e s u l t s , at least i n i t i a l l y , i n a s l i g h t rate increase. The rate changes should r e f l e c t a change i n the concentration and s t a b i l i t y of the cationic intermediates i n the a c i d . In a l l cases the a c i d i t y of modified I^SO^ decreases as the additive concentration i s increased, t y p i c a l relationships being shown i n Figure 4. The change i n a c i d i t y must be reflected i n the steady state R concentration which therefore always decreases i n these systems. The concentration of methylcyclopentane i n the acid depends strongly on the modifier, Figure 5. The data were obtained by infrared analyses of CCI4 extracts of saturated solutions of methylcyclopentane i n H2SO4 a f t e r standing 24 hours. There i s c l e a r l y no simple relationship between the s o l u b i l i t y of methylcyclopentane and the r e l a t i v e hydride transfer rates. 2% Methanesulf onic acid f o r example, increases the s o l u b i l i t y about 10-fold but y i e l d s only a s l i g h t rate increase. The fact that some increased rates are observed suggests that we have increased cationic r e a c t i v i t y by weakening the acid and d e s t a b i l i z i n g carbonium ion intermediates. The lack of c o r r e l a t i o n with s o l u b i l i t y of the hydride donor again indicates that s i g n i f i c a n t hydride transfer does not occur i n the bulk acid phase. Rather, the data are more e a s i l y understood i f the reaction occurs primarily at the acid-hydrocarbon interface under well mixed conditions.


+

2,3,4 Trimethylpentane i n T r i t i a t e d S u l f u r i c Acid The reaction of 2,3,4 trimethylpentane i n t r i t i a t e d s u l f u r i c acid provides additional evidence indicating the importance of an i n t e r f a c i a l hydride transfer i n the formation of products during alkylation. This compound reacts readily i n concentrated acid. Its behavior was studied i n 88.92 to 98.53 percent acid (15). The reaction i s preceded by an induction period whose duration i s inversely related to the acid strength. In t h i s period one can detect the formation of small amounts of SO2 due to the reduction of the acid as the hydrocarbon i s being oxidized to a carbonium ion. I t i s possible to r e l a t e the conversion of 2,3,4-trimethylpentane to the production of ions and a chain length of 860 has been determined f o r this reaction (16). Once a steady state ion concentration i s obtained, the reaction exhibits k i n e t i c s which are f i r s t order i n the 2,3,4trimethylpentane concentration i n the emulsion and second order i n proton a c t i v i t y measured on the H scale, Table II and Figure 7. 0


8

INDUSTRIAL


800 j

fjl

1

ι

1

1

I

ι

10

1

1

I

1

20

1

1

1

1

30

V O L . % ADDITIVE

Figure 5.

A N DLABORATORY

1

1

1

1

40

ALKYLATIONS

Γ

1— 50

IN H S 0 (96%) 2

4

Solubility of methylcyclopentane in modified sulfuric acid, 25°

T I M E , min.

Figure 6.

Reaction of 2,3,4-trimethylpentane in HTSO

k

exhibited on induction period


Alkylation


KRAMER

•6| 8.0

Studies

ι 8.4

I 8.8

I 9.2

I 9.6

I 10.0

I 10.4

J 10.8

Figure 7. First-order rate constants for the reaction 2,3,4trimethylpentane in concentrated sulfuric acid

0

5

10

15

20

TIME, min. Figure

8.

Reaction products obtained from 2,3,4-trimethylpentane in HTSO at 94% lt


INDUSTRIAL

10

A N D LABORATORY

ALKYLATIONS

Table I I Reaction Rate Constants As A Function Of A c i d i t y


Acidity wt %

(a)

k 34 TMP - H o s e c 2

1

88.92

8.72

1.77 X 10"

6

90.64

9.05

1.05 X i o -

5

92.36

9.35

5.00 X 10-5

94.17

9.70

1.95 X 10"

96.06

10.05

1.26 X 10-3

97.78

10.40

5.00 X 10-3

98.53

10.58

7.81 X 10-3

4

Jorgenson and Hartter, JACS 85^ 878 (1963).

Of s p e c i a l interest with respect to a l k y l a t i o n are the product d i s t r i b u t i o n s shown as a function of time i n Figures 8, 9, 10 and 11. The e a r l i e s t product i s isobutane i n both 94 and 98.5% H2SO4. A l l other products including the C5, and Cy paraffins and the C8 f r a c t i o n are secondary. The data can be r a t i o n a l i z e d by assuming that 2,3,4-trimethylpentane i s i n i t i a l l y converted to an ion by hydride transfer to either a carbonium ion or an oxidizing agent i n the acid. The TMP ion then rearranges and cleaves to a t-C^Hg"*" ion and 1C4H3. Rapid proton exchange with the acid equilibrates the ion with isobutylene before the cation extracts a hydride ion from another 2,3,4-trimethylpentane molecule or an o l e f i n i n the acid. As the ion concentration i n the acid grows toward a steady state value, a l k y l a t i o n and polymerization-cracking reactions occur which generate a d i s t r i b u t i o n of 05+ to C3"*" ions i n the acid. An estimate of this "homogeneous" alkylate d i s t r i b u t i o n i n the bulk acid can be made from the product d i s t r i b u t i o n at the e a r l i e s t times shown i n Figures 8 to 11. The "homogeneous" alkylate i s d i s t i n c t l y d i f f e r e n t from t y p i c a l alkylate made i n a well s t i r r e d p i l o t unit or commercial reactor. In p a r t i c u l a r i t exhibits a very high r a t i o of dimethylhexanes to trimethylpentanes, and t h i s r a t i o i s higher i n weaker acid than i n stronger, Table I I I . +


KRAMER

Alkyhtion

Studies

I Ο

(g Figure 9.

TIME, min. Reaction products obtained from 2,3,4-trimethylpentane in HTSO at 94%


k

10

20

30

40

50

60

TIME, SECONDS Figure 10. Reaction products obtained from 2,3,4-trimethylpen tane in HTSO at 98.5% k

Figure 11. Reaction products obtained from 2,3,4-trimethylpentane in HTSO4

at

98.5%


12

INDUSTRIAL

AND

LABORATORY

ALKYLATIONS

Table I I I Octane D i s t r i b u t i o n Vs 2,3,4-Trimethylpentane Conversion (a) 2,3,4

% H S0A

DMH/TMP

3.2 9.5 12.0 20.3

94.17

0.62 0.58 0.55 0.53

6.0 11.7 14.2 27.0

98.53

0.33 0.37 0.36 0.31

wt


TMP Conv. %

9

(a) T y p i c a l a l k y l a t e , DMH/TMP = 0.15/1. The high DMH/TMP r a t i o i s probably mainly due to reactions 1 to 3.

^

+ ^

+

>

+

(1)

+ —> J^^

^

tff

>

DMH

(3)

Production of dimethylhexanes i n t h i s way seems reasonable since chain branching rearrangements are not f a c i l e i n H2SO4, i . e . methylpentanes w i l l e q u i l i b r a t e without forming dimethylbutane. On the other hand 2,5-dimethyl-l,5-hexadiene can be converted to dimethylhexanes, 0v50% s e l e c t i v i t y ) , and c y c l i c alkanes when reacted with a good hydride donor i n concentrated H2SO4 indicating that multiple protonation and hydride transfer occurs faster than the rearrangement to trimethylpentane. Other contributing routes to the dimethylhexanes and the cracking products may include a polymerization-rearrangementcracking sequence involving C^2 25 (iZ)> unlikely rearrangement of t-C^Hg"*" to S-C4H9+ followed by i t s addition to isobutylene, or by the addition of t-C4H9+ to 1-butene. The important point i s that i f the dimethylhexanes are produced as a result of a sequence of reactions i n the bulk acid, then t h i s reaction should become more important as the ion and o l e f i n l e v e l s i n the acid increase. This should lead to H

+ i

o

n

s

a

r

a

t

n

e

r



1.

Alkyfotion

KRAMER

Studies

13

increasing DMH/TMP ratios as reaction proceeds, but i n fact the DMH/TMP r a t i o decreases as the 2,3,4-trimethylpentane conversion increases. The drop i n the DMH/TMP r a t i o suggests that a second reaction s i t e which p r e f e r e n t i a l l y y i e l d s trimethylpentanes develops with time. This should be the i n t e r f a c i a l reaction. A rough measure of the r e l a t i v e importance of the second s i t e can be made by assuming that bulk acid yields a DMH/TMP r a t i o of 0.33 and that the interface y i e l d s pure ΤΜΡ. In order to obtain a t y p i c a l product with DMH/TMP = 0.15/1 the r a t i o of product formed at the interface to that formed i n the bulk acid i s about 1:1. This r a t i o w i l l be much higher i f the i n t e r f a c i a l reaction produces some of the dimethylhexanes as seems l i k e l y . The behavior of 2,3,4-trimethylpentane reinforces the conclu sions of the study of the t-butylchloride reaction with methyl cyclopentane i n suggesting that a means of increasing the hydride transfer rate i n H2SO4 could lead to improved s e l e c t i v i t y i n a l k y l a t i o n . The next reaction was chosen to investigate t h i s possibility. 3-Methylpentane Isomerization i n T r i t i a t e d S u l f u r i c Acid. Controlling The Rate with The Triphenylmethyl Cation. The isomerization of 3-methylpentane occurs slowly i n concentrated H2SO4. When using t r i t i a t e d acid at tracer l e v e l s , i t i s r e a d i l y observed that the t e r t i a r y 2 and 3-methylpentyl ions equilibrate quickly once an ion forms. Each ion i s assumed to undergo fast reversible exchange of a l l protons adjacent to the cationic center so that to a good approximation the radioactive methylpentanes w i l l contain 13 exchanged protons.

1Î* H

+

Atf

+ C H 6

1 2

H

+

+ C H 6

1 2

(6)

The r e l a t i v e rates indicated i n equations 4, 5 and 6 show the chain carrying hydride transfer reaction as the rate determining step. Hence determination of means to control the rate of incorporation of r a d i o a c t i v i t y i n the methylpentanes i s synonomous with being able to control hydride transfer. If the hydride transfer reaction involves a bimolecular encounter between a cation and a donor, one way of increasing i t s



14

INDUSTRIAL

AND

LABORATORY

ALKYLATION S

rate would be to d e s t a b i l i z e the cation. In the case of an i n t e r f a c i a l reaction, t h i s might i n p r i n c i p l e be accomplished by adding c a t i o n i c surfactants. In the bulk acid the addition of large cations that might cause a reorganization of solvent structure might also change the s t a b i l i t y of the c a t i o n i c i n t e r mediates. I d e a l l y , one might add a large cation that i n addition to the preceding might be a bonafide intermediate i n accepting hydride ions from paraffins and passing them along to other cations i n the acid system, and such a r o l e has occasionally been postulated for "red o i l " i n a l k y l a t i o n (12b,13). The u t i l i t y of the triphenylmethyl cation i n c o n t r o l l i n g hydride transfer which might be due to several of these reasons i s the topic of t h i s section. Before discussing the " t r i t y l " ion i t i s important to note that the methylpentane isomerization i s slow i n fresh acid (there's an induction period as with 2,3,4-trimethylpentane), but may be accelerated by the addition of small amounts of an o l e f i n l i k e 2-methyl-l-butene. However, the reaction i n i t i a t e d by the o l e f i n i s rapidly quenched as the acid reduces a momentarily high intermediate ion concentration to a steady state l e v e l . The reduced rate i s now a measure of the hydride transfer rate i n the a c i d . The data to be discussed were obtained at 23°C with emulsions containing equal volumes of methylpentane and 95 percent H S0 . Control of the 3-methylpentane reaction was found to be a complex function of the amount of o l e f i n used to i n i t i a t e reaction and the amount of triphenylmethyl ion present. Figures 12 and 13 show t y p i c a l data over a range of compositions and the r e l a t i v e hydride transfer rates as a function of the reagent concentration i s shown i n Figure 14. ^ One mole percent triphenylmethyl i n the acid quenches the "background" exchange or isomerization of 3-methylpentane i n the acid. Taking the steady state rate of the isomerization i n i t i a t e d with 0.05 mole percent o l e f i n as unity one finds that i t may be doubled with the large cation i n the acid. The accelerated rate has been found to be stable throughout an 8 hour day i n a properly conditioned acid. S t i l l higher controlled rates are attainable as shown i n Figure 14. Phenomenologically, as the o l e f i n concentration i s increased at a given triphenylmethy]® concentration, the rates maximize and then descend slowly. At high o l e f i n concentrations the rates decrease toward the lined out value with o l e f i n alone. The deleterious e f f e c t of too high o l e f i n concentrations may be p a r t i a l l y o f f s e t by increasing the triphenylmethyl^ l e v e l . These results unambiguously show that i t i s possible to s i g n i f i c a n t l y increase hydride transfer rates i n H2SO4. Why does t h i s cation work and can increased hydride transfer be r e a l i z e d in steady state a l k y l a t i o n i n a commercial unit? The triphenylmethyl ion exerts i t s e f f e c t at a r e l a t i v e l y high concentration i n the acid. With a ring balance i t has not 2

4


1.

Alkyhtion Studies

KRAMER

15

18Γ

Mole % Additive *3C-C λ ,

Key

10 =

8

•

10

•

Δ

1 1

A

0.7

•

0.5 0.3 0.2 0.1

•

• Downloaded by UNIV OF TASMANIA on May 3, 2015 | http://pubs.acs.org Publication Date: June 1, 1977 | doi: 10.1021/bk-1977-0055.ch001

^

c ο

4

TIME, Figure 12.

0.05 _ 0.05 _ 0.08 0.05 " 0.07 " 0.05 0.05

1

30

20

10

-

MINUTES

Effect of 0.1-1.0 mol % triphenylmethylcation on controlling olefinic initiation of 3-methylpentane exchange

16 — 14 12 10 -

^

8 -

ADDITIVE, MOLE % Ο 0 C-CI, 2.0+J^ , 0.1 3

6 -

•

4 -

Δ 03C-CI, 2 . 0 + λ / ,

2 -

• ^ C - C I , 3 . 5 + λ / , 0.3

03C-CI, 1.5+Js^, 0.2

_

• # 3 Ç - Ç ' / 2.0+,Αχ, 0.25

o|—

40 TIME, min.

Figure 13.

"

0.18-

Using C* to control olefin initiation of 3-MC exchange 3

5


INDUSTRIAL


16

AND

LABORATORY

ALKYLATIONS

been possible to find a surface excess of the ion at either the a c i d - a i r or acid-hydrocarbon interface so that the ion i s uniformly distributed throughout the acid. It may be causing s t r u c t u r a l changes i n the acid and may be an intermediate i n hydride transfer, but i t i s not an e f f e c t i v e surfactant and should not be very e f f i c i e n t at influencing the surface reaction. Cationic surfactants which should a f f e c t t h i s reaction are however, readily available. Figure 15 shows t y p i c a l graphs of surface tensions of the hexadecyltrimethylammonium ion and an isotridecyldimethyl benzylammonium ion at a i r - a c i d and hexaneacid interfaces. The surfactants behave much as i n water and form films containing about one ion/130 A^. The cationic surfactants would be expected to control carbonium ion behavior at the a c i d hydrocarbon interface at much lower concentrations than the t r i phenylmethyl ion. A complete comparison of exchange i n the 3methylpentane system i s not available but the surface active cations have been found to markedly improve the a l k y l a t i o n reaction i n continuous p i l o t plant and commercial apparatus. Cationic Surfactants i n A l k y l a t i o n The u t i l i t y of cationic surfactants i n increasing hydride transfer would be expected to be shown by an increased y i e l d of octanes during butene a l k y l a t i o n . This follows i f alkylate s e l e c t i v i t y i s decided by the r a t i o of the rate at which i n t e r mediate ions are captured by hydride transfer to the rate at which they add to o l e f i n s and polymerize, and i f the e f f e c t of the additives i s to s e l e c t i v e l y raise the s p e c i f i c rate constant for hydride transfer, k^ . -

k -(iC H )(R+) H

Selectivity

=

K P

4

1Q

(R+)(R-)

To study t h i s e f f e c t a small microalkylation p i l o t unit (MAPU) was constructed. The unit contained a p o s i t i v e displacement pump for accurately passing feed into the bottom of a %" diameter glass reactor containing 5 ml of acid, 98% H2SO4 fresh. The reactor was mixed with a high speed rotary "Magna Drive" unit that reached within 1/16" of the feed i n l e t . Temperature was controlled with a c i r c u l a t i n g water bath. The acid-hydrocarbon emulsion would v i s u a l l y separate while the system was s t i r r e d and the products l e f t the reactor through a back pressure regulator. A l l the products were vaporized and passed through a gas sampling valve that automatically drew a sample into a gas chromatograph about once an hour throughout the run. The analyses were used to calculate a motor octane number assuming the MON was a l i n e a r function of i t s components. A l l runs were of the "dying acid" type i n that a single i n i t i a l acid charge was made. The runs lasted about a week and the acid was then measured for i t s carbon


1.

Alkyhtion Studies

KRAMER

17

= Relative Hydride Transfer Rates

Ο

2.2 Φ

3h Ο 2.6-3

Ο 2.0

IP


2.8 Ο

2.1o

2h

1.7

O - ^ O - 2.3

1.2

(§> 1 . 0

2.7
a

18.0

89.0

3.6

14.4

91.5

3.3

19.2

93.3

2.8

10.6

94.9

2.2

19.1

92.7

2.2

13.3

93.3

2.9

(a) Cumulative volume of o l e f i n fed per volume of acid. (b) In the run with mixed o l e f i n s , 0.05% was used.


Alkylation Studies

KRAMER


100,

0

0.10

0.20

OLEFIN

Figure 18.

SPACE

0.30 VELOCITY

V/HR/V

Alkylation selectivity improved with refinery olefins

2500

B/O REACTOR 8

400

G/HR

75% Acid-Emulsion 7 5 % i C4 in Product .08 SV 50°F H

2

S

0

4

98%

2500

B/D REACTOR 7

400

Figure 19.

G/HR

Plant test conditions


22

INDUSTRIAL

A N D LABORATORY

ALKYLATIONS

Table VI shows that the t i t r a t a b l e a c i d i t y of used acid i s higher and the carbon content lower i n the runs containing additives with a 2-butene feed. The difference i s somewhat less but s t i l l substantial i n the experiments with the mixed o l e f i n feed. For comparison one can estimate the a c i d i t y of the blank a f t e r feeding 10.6 volumes of olefin/volume acid as 93.2 percent. The reactions involved i n consuming acid are not well under stood at t h i s time and the simplest r a t i o n a l i z a t i o n of the data i s to note that as a l k y l a t i o n conditions improve, less of the feed has the opportunity of entering side reactions that deplete the acid.


Commercial Test One of the c a t i o n i c surfactants was evaluated i n a commercial a l k y l a t i o n unit at the Baytown r e f i n e r y . A p a r a l l e l test was conducted i n which two reactors received the same feed and fresh acid, Figure 19. The modifier was injected into an acid recycle l i n e on reactor 8, to rapidly bring i t s concentration to working strength and then the rate was lowered to maintain the concentra t i o n . After 11 days the additive concentration was doubled and a f t e r 19 days i t s addition was switched from reactor 8 to reactor 7. The additive concentration i n reactors 7 and 8 i s shown i n Figure 20· Note that a f t e r the switch the concentration i n reactor 8 depletes i n accord with the acid replacement rate of the unit and so this reactor w i l l continue to receive the benefit of the additive u n t i l the concentration drops to a l e v e l estimated as about 0.005 wt percent. After this time a l l improvements due to the additive should be seen as changes developing i n reactor 7. The difference i n t i t r a t a b l e a c i d i t y of the recycle acid from the reactors was taken as the best parameter for measuring the effectiveness of the additive i n the t e s t . I t can be shown that the t i t r a t a b l e a c i d i t y , C, (wt % H2SO4) w i l l respond according to equation 7. c

C

- A " (C -C )eA

k t

(7)

0

Here k i s a constant, t i s the time, C i s the steady state a c i d i t y i n the presence of the additive and C i s the steady state a c i d i t y i n the absence of additive. The rate constant k i s determined by the acid make up rate and the acid inventory i n the system and has a value of 0.12 per day i n this study. The difference i n t i t r a t a b l e a c i d i t y , Δ, between reactors 7 and 8 should follow equations 8, 9 and 10. A

0


Alkylation Studies

KRAMER

1 .02

1

-

Reactor 8


1

/

Reactor

1 0

7

10

ι

ι

ι

20

30

40

TIME

Figure 20.

τ

_l

Ο

DAYS

Additive concentration in commercial test

1

I

10

1

I

20

1

I

30

r

ί-

40

TIME, DAYS

Figure 21.

Additive is effective


24

INDUSTRIAL

A N DLABORATORY

ALKYLATIONS

Δ

=

(c

A

- cJQ. - e" t], 0-20 days

(8)

Δ

=

[c

A

- cJËe-^t-

(9)

Δ

=

k

2 0

)^- ^),

20-31 days

1

(?A - c J ^ - k C t ^ D - e - k i t ^ O ) - ^ ,

31-43 days

During the f i r s t twenty days the additive was i n reactor 8 and not reactor 7 and the difference i n t i t r a t a b l e a c i d i t y should follow equation 8. During the next eleven days the additive i s in both reactors and the Δ decreases according to equation 9. On the 3 1 day the additive i s below i t s e f f e c t i v e concentration l e v e l , (^0.005 wt % ) , and the difference i n a c i d i t y decreases at a faster rate. The s o l i d curve i n Figure 21 was calculated from these equations using a value of C - C of 0.7. The c i r c l e s i n Figure 21 are experimental points, each representing a minimum of 36 t i t r a t i o n s . The t i t r a t a b l e a c i d i t y of each reactor was measured i n t r i p l i c a t e every eight hours and the points average two-day periods. These data show an uncorrected decrease of 14 percent i n acid consumption (derived from Δ developed a f t e r 18 to 20 dayg). However the average temperature i n reactor 8 was 4 to 5 F higher than i n reactor 7 during the entire test and we estimate that the difference i n acid consumption i s closer to 20 percent at constant temperature. The plant test was unfortunately terminated before the f u l l Δ i n the opposite d i r e c t i o n could be obtained but the data c l e a r l y shows the additive to be strongly b e n e f i c i a l i n a l k y l a t i o n . In addition to saving acid the additive appeared to improve the octane number by more than 0.1 MON as was indicated by a few spot checks of alkylate during the run. The improvements generally arose from a s l i g h t increase i n the C3 f r a c t i o n , a r i s e i n the trimethylpentane concentration and changes of the t r i methylpentane d i s t r i b u t i o n . The octane analyses are not nearly as extensive as the t i t r a t a b l e a c i d i t y determinations and the improvements are noted as being consistent with what would be estimated from plant correlations and the observed reduction i n acid composition. The additive used i n the commercial test i s being used i n nearly a l l of Exxon's a l k y l a t i o n units. No operating problems have been encountered and i t generally has been found to reduce acid consumption by 15 to 20 percent and to generate s l i g h t l y higher octane number product. The cost of the additive i s small r e l a t i v e to the acid savings alone and i t i s available f o r license. The a l k y l a t i o n model developed i n this work i s one i n which the reaction i s viewed as occurring i n the acid phase and at the acid-hydrocarbon interface. The formation of Cg s and trimethyl pentanes occurs p r e f e r e n t i a l l y at the interface. Adding c a t i o n i c surfactants reduces the s t a b i l i t y of the carbonium ion s t


(10)

A

Q

r


1.

Alkyfotion Studies

KRAMER

C

25

Feed

4

C4 Feed

hC -Cglc -C i-C ^

K -N-C 1N-C -N^

Acid

Acid

4

4

4

4

4

8

4

Figure 22. An alkyhtion model. Cationic surfactants should block sur face reactions and destabilize reaction intermediates.


intermediates and causes them to abstract hydride ions more rapidly from isobutane or any other p o t e n t i a l donor. Increased hydride transfer converts more of the carbonium ions at the acid interface to saturates faster, y i e l d i n g product while minimizing polymerization and side reactions. I t i s also l i k e l y that the surfactants p h y s i c a l l y block a l k y l ions from one another i n the surface f i l m and thus impede ion + o l e f i n polymerization. In such a f i l m the carbonium ion concentration must also be lower than i n the absence of surfactant and mass law e f f e c t s w i l l therefore also lead to less polymerization and cracking. The fact that steady state hydride transfer rates i n H S0 are subject to control through the use of acid modifiers which act i n the bulk acid and at the acid-hydrocarbon interface i s the key to the control of s u l f u r i c acid a l k y l a t i o n . 2

4

Literature Cited 1. 2. 3.

4. 5. 6. J. 7. 8. 9. J.

Kennedy, R. M. i n "Catalysis," V I , ed. P. H. Emmett, Chapter 1, 1, Reinhold (1958). Condon, F . E . i n "Catalysis," V I , ed. P. H. Emmett, Chapter 2, 43, Reinhold (1958). Schmerling, L . i n "The Chemistry of Petroleum Hydrocarbons," I I I , ed. Β. T. Brooks, S. S. Kurtz, Jr., C. E . Board, L . Schmerling, Chapter 34, 363, Reinhold (1955). Burwell, R. L . and Gordon, G. S., III, J. Am. Chem. Soc. (1948) 70, 3128. Burwell, R. L . Jr., Maury, L . G. and Scott, R. B . , J. Am. Chem. Soc. (1954) 76, 5828. Burwell, R. L., Scott, R. B., Maury, L . G. and Hussey, A. S . , Am. Chem. Soc. (1954) 76, 5822. Gordon, G. S . , III and Burwell, R. L., J. Am. Chem. Soc. (1949) 71, 2355. Ingold, C. K., R a i s i n , C. G. and Wilson, C. L., J. Am. Chem. Soc. (1936) 58, 1643. Otvos, J. W., Stevenson, D. P., Wagner, C. D. and Beeck, O., Am. Chem. Soc. (1951) 73, 5741.


26

INDUSTRIAL AND LABORATORY ALKYLATIONS

10.

J.


J.

Stekina, V. N., Jursanov, D. Ν., Sterligov, O. D. and Liberaan, A. L., Doklady Akad. Nauk. S.S.S.R. (1952) 85, 1045. 11. Stevenson, D. P., Wagner, C. D . , Beeck, O. and Otvos, J. W., Am. Chem. Soc. (1952) 74, 3269. 12a. Thomas, C. L., "Catalytic Processes and Proven Catalysts," Chapter 9, 87 Acad. Press (1970). 12b. Doshi, B. and A l b r i g h t , L . F., Ind. Eng. Chem., Proc. Des. Dev. (1976) 15, 53. 13. Unpublished results of G. M. Kramer. Mentioned i n Hoffmann, E . and Schriesheim, Α . , J. Am. Chem. Soc. (1962) 84, 953. 14. Kramer, G. M., J. Org. Chem. (1965) 30, 2671. 15. 16. 17.

Kramer, G. M., J. Org. Chem. (1967) 32, 920. Kramer, G. M., J. Org. Chem. (1967) 32, 1916. Hoffmann, J. E . and Schriesheim, Α . , J. Am. Chem. Soc., (1962) 84, 957.


Alkylation Studies

Recommend Documents