CH LORINATED PLAT1NUM-A LUMINA LOW

TEMPERATURE ISOMERIZATION CATALYSTS

J O S E P H P. G I A N N E T T I A N D RAYNOR T. SEBULSKY
Gulf Research & Development Co., Pittsburgh, Pa. 15230

Low temperature paraffin isomerization catalysts can be prepared by reaction

of a platinum-alumina with various chlorinating agents. Thionyl chloride, sulfuryl
chloride, sulfur monochloride, and a mixture of sulfur dioxide and chlorine can
be employed a t elevated temperatures to chlorinate the catalyst. Increased catalyst
activity is obtained if the platinum-alumina is treated with hydrogen chloride
prior to the sulfur chloride treatment. These chlorinated platinum-aluminas are
capable of isomerizing butane, pentane, and hexane to near their equilibrium
isomer distributions a t temperatures below 3503 F. In addition to the chlorination
sequence, the alumina and its platinum content play major roles in determining
catalyst activity.

PARAFFIN
isomerization is an excellent means of obtaining from the Eastman Kodak Co. and used as received. The
high octane gasoline components. Hexane can be isom- sulfur dioxide, hydrogen chloride, and chlorine were
erized to the higher octane dimethylbutanes and methyl- obtained from the Matheson Chemical Co. and also used
pentanes, while butane can be isomerized to isobutane as received. The hexane and butane were Phillips Petro-
(2-methylpropane), which can be alkylated with olefins leum Co. technical grade and pure grade, respectively.
to give the highly branched, high octane C6 to C, com- The butane contained 0.35 isobutane and 0.3': pentanes,
pounds. while the hexane contained 0.45 methylpentanes and 2.1%
Thermodynamic equilibrium favors branched paraffins methylcyclopentane. Both were dried and desulfurized
a t low temperatures. Accordingly, maximum isomerization over Linde 13X molecular sieves. Hydrogen was supplied
catalyst activity is desired to permit isomerization pro- by the Air Reduction Co. and was deoxygenated and
cesses to operate a t minimum temperatures. The use of dried prior to use.
Friedel-Crafts catalysts such as anhydrous aluminum chlo- With the exception of catalysts specifically made to
ride and aluminum bromide, singly or in combination with study different supporting aluminas, the catalysts were
other halides, a t low temperatures, has been known for a prepared from Sinclair Baker platinum-alumina. The alu-
long time (Egloff et cd.,1942; Thomas, 1941). However, mina used as the support in the study of platinum level
these catalysts are corrosive, easily deactivated, difficult to was also obtained from Sinclair Baker. Other aluminas
handle, and unselective in the reactions which they cata- and platinum-aluminas were obtained from various manu-
lyze. In addition, the aluminum halides are often employed facturers. A silica-alumina and a palladium-containing
as slurries or sludges which add mechanical problems. silica-alumina were also evaluated. Where the metal was
Within the past few years several active solid low tem- not present on the support, platinum was added by
perature isomerization catalysts have been reported by slurrying the support with an aqueous solution of chloro-
workers a t Texaco (Riordan and Estes, 1966) and British platinic acid and oven drying a t 250°F.
Petroleum (Goble and Haresnape, 1964; Goble and Catalyst Preparation. All the supports (12- to 20-mesh
Lawrence, 1964; O'May, 1963). The key to their activity size, U. S. sieve series) were calcined in air a t 500°F.
is the introduction of chlorine into platinum-alumina by overnight and then a t 900°F. for 2 hours. They were
reaction with carbon chlorides. Carbon tetrachloride is reduced in H 2 a t 900" F. and atmospheric pressure employ-
particularly effective. I n addition, active catalysts have ing 1.6-standard cubic foot H 2 per hour per 100 grams
been prepared by reaction of thionyl chloride with of support for 2 hours.
platinum-alumina (Goble and Haresnape, 1965). A 170-ml. charge of the calcined platinum-alumina was
Our studies have shown that there is a chlorinating placed in an electrically heated 1-inch-i.d. glass reactor
sequence for platinum-aluminas which gives catalysts with equipped with a glass thermowell. The temperature was
greatly enhanced low temperature true isomerization raised to the 900" F . hydrogen reduction temperature under
activity. In addition, we have extended previous work a small flow of nitrogen. Following reduction with hydro-
with thionyl chloride to other chlorine-containing sulfur gen, they were brought to the treatment temperature
compounds. in nitrogen. Where the hydrogen chloride pretreatment
and posttreatment were employed, it was added downflow
Experimental over the platinum-alumina. Hydrogen (1.20-standard
cubic foot H ? per hour per 100 grams of platinum-alumina
Materials. The thionyl chloride (SOCL), sulfuryl chloride in most cases) was used as a carrier gas in the hydrogen
(S02C12),and sulfur monochloride (S2C12)were obtained chloride pretreatment. The liquid activating agents

356 I & E C PRODUCT RESEARCH A N D D E V E L O P M E N T

(SOCl,, SO?Cl,, and SKL) were added dropwise from R e s u l t s and D i s c u s s i o n
a buret and allowed to vaporize into the approprihte carrier
stream (air, nitrogen, or hydrogen a t usually 0.40 standard Reaction of Thionyl Chloride with Platinum-Alumina.
cubic foot per hour per 100 grams of platinum-alumina). Platinum-alumina reacted with thionyl chloride (SOClJ
When the activating agents were sulfur dioxide and chlo- a t 350", 560", 800", and 1050" F. with a nitrogen carrier
rine, the two gases were combined prior to the reactor gas. I n all but the 1050°F. preparation, a hot spot of
and added as a single gas stream. from 75. to 150°F. traveled down through the bed as
Apparatus. The isomerization experiments were con- the catalyst was activated. The catalysts were evaluated
ducted in a 1-inch i.d. carbon steel fixed-bed high pressure for hexane isomerization. Figure 1 compares the activity
reactor. Butane or hexane was combined with hydrogen, of the catalysts, expressed as hexane conversion and 2,2-
preheated, and passed downflow through a 100-ml. bed dimethylbutane selectivity, with their surface areas and
of catalyst. I n the butane runs the total gaseous products chlorine contents. For comparison, equilibrium hexane con-
leaving the reactor were depressurized, passed through version and 2,2-dimethylbutane selectivity a t 225" F. are
a dry gas meter, and. collected in a vessel by salt water about 91 and 40 weight ' C , respectively (Evans et a l ,
displacement. For the hexane runs the effluent from the 1962; Ridgeway et al.. 1959). The catalyst prepared a t
reactor was divided into gas and liquid portions in a 1050" F. has the highest activity, but rather surprisingly,
high pressure separat'or. The liquid product was collected the lowest surface area and chlorine content. The highest
in a vessel, while the gaseous product was passed through surface area and chlorine content were obtained on the
a wet gas meter. catalyst prepared a t 350" F., yet this catalyst is inactive
Analyses. The chlorine content of the catalyst was deter- for hexane isomerization. Therefore, merely adding chlo-
mined by an automatic potentiometric procedure. The rine to the platinum-alumina is not sufficient for activity.
sample was first slurried in water and the hydrochloric Analysis of the catalysts prepared a t 800" and 350'F.
acid resulting from the hydrolysis of the chlorine- by infrared and nuclear magnetic resonance techniques
containing catalyst titrated potentiometrically with a (Basila, 1965) provides a partial explanation for this
standard base to p H 7. All the chlorine from the catalyst activity difference. For the active catalyst prepared a t
was removed. The chlorine results reported were corrected 800"F., all the surface hydroxyl groups of the alumina
for the chlorine initially present on the platinum-alumina- were replaced by chlorine. For the inactive catalyst pre-
0.50 weight 'C chlorine on the Sinclair Baker platinum- pared a t 350°F., only about 6OCcof the surface hydroxyl
alumina, for example--and thus represent only the chlorine groups were replaced by chlorine.
added in the activation procedures. Surface areas were The effect of varying the amount of thionyl chloride
determined by nitrogen adsorption. used in the reaction with the platinum-alumina was deter-
The sulfur content of the catalyst was determined by mined. Thionyl chloride in quantities of 0.09, 0.18, 0.35,
a high temperature combustion method. The catalyst was
heated in oxygen to convert the sulfur into sulfur dioxide.
This stream was passed into an acidic solution of potassium
iodide and starch indicator. A small amount of potassium
iodate was added to develop a blue color. As the SO,
stream reacted with this solution, potassium iodate was
added to maintain the blue color. T h e sulfur content
was calculated from the amount of potassium iodate added.
Certain catalysts were analyzed for hydroxyl group con-
tent by both infrared (Perkin Elmer 421 dual gradient
spectrophotometer) and nuclear magnetic resonance
(Varian Associates V4500 wide-line N M R spectrometer)
techniques.
The catalysts were evaluated for hexane isomerization
a t 225" or 280"F., 2!50 or 400 p.s.i.g., 1 LHSV (liquid
hourly space velocity defined as milliliters of hydrocarbon
9 1
feed per milliliter of catalyst per hour), and 2.5 to 1 350
hydrogen-hexane mole ratio. Butane runs were conducted 8 - u
' N
a t 350" F., 800 p.s.i.g., 1 LHSV, and 2.5 to 1 hydrogen- I'
7 - 300 3
butane mole ratio. The hydrocarbon samples were P;
analyzed on a Fisher-Gulf Partitioner using a column 6 -

packed with 10% isoquinoline on Gas Chrom R , 45- to 250 fa

5 -
60-mesh, obtained from the Applied Sciences Laboratory.
For isomerizing both hexane and butane, the reaction 4 - 200
was over 98% selective to branched hexanes and isobutane, 300 550 800 1050
respectively. I n the case of hexane, the selectivity to CATALYST PREPARATION TEMPERATURE: "F
the 2,2-dimethylbutane isomer is also presented. This was
calculated by div:iding the amount of the 2,2-
Figure 1. Effect of temperature of thionyl chloride
dimethylbutane in the product by the hexane conversion. reaction with platinum-alumina on catalyst prop-
This particular isomer has a high octane rating (106 erties and activity
Research octane nurnber + 3 ml. of tetraethyllead) and 0.36 mole SOClr/100 g. platinum-alumino over 45 min.
is present in high concentrations in an equilibrium hexane with Nr diluent. Hexane isomerization at 225' F., 250 p.s.i.g.,
mix a t low temperatures. 1 LHSV, 2.5:l Hz-hexane mole ratio

VOL. 8 N O . 4 D E C E M B E R 1 9 6 9 357
and 0.54 mole per 100 grams of platinum-alumina reacted
over a 45-minute period a t 560°F. The chlorine content Table I. Carrier Gas Type during Thionyl Chloride Reaction
and surface area of the catalysts and their activity for with Platinum-Alumina on Catalyst Properties and Activity
hexane conversion and 2,2-dimethylbutane selectivity are 0.36 mole SOCl,/lOO g. platinum-alumina a t 56W F. over 45 min.
presented in Figure 2. Results show an increase in the Hexane isomerization a t 225" F., 250 p.s.i.g., 1 LHSV, 2.5:l H2-
hexane conversion and 2,2-dimethylbutane selectivity with hexane mole ratio.
increasing amounts of thionyl chloride, matched by an Carrier Gas Type
increase in the chlorine content and a decrease in the
surface area of the catalysts. Therefore, the most active Catalyst analyses Nz 02 H2
catalyst employed the greatest amount of thionyl chloride Surface area, sq. m./g. 298 301 325
in the preparation. However, as the amount of thionyl Chlorine, wt. % 8.2 8.8 4.5
Hexane conversion, wt. 70 83.0 81.2 7.3
chloride increased, the efficiency-i.e., fraction of available 2,2-Dimethylbutane selectivity, wt. % 24.1 26.6 ...
chlorine retained by the catalyst-decreased, as shown
by the numbers in parentheses next to the chlorine content.
Continued increases in available chlorine would probably
not produce a catalyst containing appreciably more chlo- can affect their catalytic properties. Generally, the lower
rine and having greater activity. the state of hydration the more active the alumina for
Nitrogen was used as the carrier gas during these reac- the usual acid-catalyzed reactions. T o see the effect of
tions between thionyl chloride and platinum-alumina, T o low hydration states of the alumina prior t o thionyl chlo-
see if the type of carrier gas influences the final catalyst, ride reaction on the resultant catalyst properties, the
the reaction was also performed with oxygen and hydrogen platinum-alumina was given a 1050"F. treatment with
(Table I). Oxygen and nitrogen produce catalysts with hydrogen chloride, followed by the 560" F. thionyl chloride
generally comparable properties and activities. The treatment. The resultant catalyst had lower chlorine con-
catalyst made with a hydrogen carrier was low in chlorine, tent and lower surface area than the catalyst prepared
high in surface area, and very low in activity. This is with thionyl chloride alone. But, surprisingly, this catalyst
in agreement with reported results showing that a reducing had considerably more activity for butane isomerization
atmosphere causes the loss of carbon chlorides (Goble than the nonpretreated catalyst. T o show that this was
and Haresnape, 1964) and thionyl chloride (Goble and not an additive effect, a hydrogen chloride-treated
Haresnape, 1965) by converting to hydrogen chloride. platinum-alumina without subsequent reaction with thio-
Hydrogen Chloride Treatment of Platinum-Alumina Prior nyl chloride was tested and found to be inactive (Table
to Thionyl Chloride Reaction. Maclver, Tobin, and Barth 11). The exact function of the hydrogen chloride pretreat-
(1963) have shown that the state of hydration of aluminas ment is not known. However, infrared and N M R studies
(Basila, 1965) have shown that the hydrogen chloride
treatment removes about 30% of the hydroxyl groups

c
on the alumina surface. The thionyl chloride then reacts

a-@I
loo
* with the remaining hydroxyl groups.
80
/ONVERSION 40 s The temperature of the reaction with hydrogen chloride
f I-
has a pronounced effect on the activity obtained after
a P the subsequent thionyl chloride reaction (Figure 3). The
L greatest activity for the hydrogen chloride-thionyl chloride
r z

8 $ catalyst occurs when the hydrogen chloride pretreatment

temperature is 950°F. and above. Thus, 46 to 47 weight
z >
c
6 % butane conversion is obtained for the 950" and 1050" F.
20 14
I /
,o-
1 lo
1
xI hydrogen chloride-treated catalysts, while 36 to 37 weight
% conversion results from the catalysts treated a t 400"
l
o
' 0 and 800" F. Since a catalyst activated with only the thionyl
chloride (hydrogen chloride pretreatment omitted) also
10 I

I 127 WT.%I -1
Table 11. Hydrogen Chloride Pretreatment Prior to Thionyl
*i Chloride Reaction with Platinum-Alumina on Catalyst
z 7 - Properties and Activity
IL
s 0.30 mole HCl/hr./100 g. platinum-alumina at 1050' F. for 3 hr.
u 6 -
with H, diluent
0.36 mole SOCL/lOO g. platinum-alumina at 560" F. over 45 min.
200 with diluent
I I I I Butane isomerization at 350" F., 800 p.s.i.g., 1 LHSV, 2.5:l Hz-
0.W 0.18 0.27 0.36 0.45 0.54 butane mole ratio
MOLE SOCIz/lW 0 . PLATINUM-ALUMINA
HC1 +
HCI SOCI? SOCl,
f:igure 2. Effect of amount of thionyl chloride Catalyst analyses
n reaction with platinum-alumina on catalyst Surface area, sq. m.1 g. 175 298 145
iroperties and activity Chlorine, wt. 5 1.3 8.2 3.9
i 0 C h a t 5 6 0 ° F . over 4 5 min. with N2 diluent. Hexane
Butane conversion, wt. 0 36 47
somerizatian a t 225" F., 2 5 0 p.s.i.g., 1 LHSV, 2.5:l-hexane "Equilibrium butam conversion to isobutane at 350" F . , 61 ut. 5'1.
mole ratio

358 l & E C PRODUCT RESEARCH A N D DEVELOPMENT

2 5 0 w l , l , , l 1

I l l 1 J

-U
After HCI + SOClz
5 2 -
E 6 0-@-After HCI
' ' I
0-@
1 ' i ' ' I ! I
0
1
0

I l l
4w M)O 800 1wo Figure 4. Effect of contact time of hydrogen chloride pretreat-
HYDROGEN CHLORIDE ADDITION TEMPERATURE OF ment prior to thionyl chloride reaction with platinum-alumina
on catalyst properties and activity
Figure 3. Effect of temperature of hydrogen chloride pretreat- 00.90 mole HC1/100 g. platinum-alumina at 1050" F. with H:, diluent
ment prior to thionyl chloride reaction with platinum-alumina (amount constant, rate varied)
on catalyst properties cind activity 00.30 mole HCl/hr./100 g. platinum-alumina a t 1050" F. with HL diluent
0.30 mole HCl/hr./100 g. platinum-alumina for 3 hr. with Hz diluent (rate constant, amount varied)
0.36 mole SOCI:/lOO g. platinum-alumina at 560" F. over 45 min. with 0 0 . 3 6 mole SOClr/100 g. platinum-alumina at 560" F. over 45 min. with
NL diluent N! diluent following either of above treatments
Butane isomerization at 350" F., 800 p.s.i.g., 1 LHSV, 2.5:l Hi-butane Butane isomerization at 35OCF. 800 p.s.i.g., 1 LHSV, 2.5:l H r b u t a n e
mole ratio mole ratio

gives 36 weight 'C conversion, no significant activity in- the hydrogen chloride pretreated platinum-alumina. First
crease is derived from the 800°F.and lower hydrogen the platinum-alumina was given a 3-hour, 1050" F.hydro-
chloride treatments. The 400" and 800" F. pretreatments gen chloride pretreatment. Then the thionyl chloride reac-
result in more chlorine on the catalyst after both the tion temperature, amount, and addition time were studied.
hydrogen chloride treatments and thionyl chloride reaction There were no significant differences in catalyst properties
than the 950" F.+pretreatments. Thus the chlorine added or activity over the ranges studied. These results differ con-
by the low temperature hydrogen chloride treatments had siderably from the thionyl chloride variations with the
no promotional effect. Also, the catalysts prepared with unpretreated platinum-alumina, where there were effects
the lower temperature pretreatments have higher surface of reaction temperature and amounts of thionyl chloride
areas than those prepared with the higher temperature employed. The hydrogen chloride pretreatment, although
treatments. in itself ineffective for making an active low temperature
The hydrogen chloride contact time has also a pro- isomerization catalyst, apparently mo6ifies the surface
nounced effect on the catalyst properties and activity. such that all subsequent thionyl chloride treatments
This was tested by adding both different amounts of hydro- resulted in maximum activity.
gen chloride a t a constant rate and a constant amount Other Sulfur-Chlorine A4ctivatingAgents. In addition to
of hydrogen chloride at different addition rates. The effect thionyl chloride, other activating agents were tested:
of the changes in the contact time by these two procedures sulfuryl chloride, sulfur monochloride, and a mixture of
is shown in Figure 4 . Generally, both procedures give sulfur dioxide and chlorine. The platinum-alumina was
the same results, with the shorter contact times being first pretreated with the hydrogen chloride. A few changes
the more effective pretreatments. When followed by the were made from the thionyl chloride preparations. The
standard thionyl chloride reaction, the catalysts given a catalysts were purged with nitrogen a t 800" to 850°F.
hydrogen chloride pretreatment of an hour or less give immediately after reaction with the sulfur-chlorine com-
about 6 weight ' C higher conversion than those with the pounds. This treatment lessens the possibility of unreacted
longer treatments. The shorter hydrogen chloride treat- material remaining on the catalyst surface. Also, the
ments also do not reduce the surface area of the platinum- catalysts were treated with hydrogen chloride a t 400"F.
alumina as much. Best results are obtained when the following the nitrogen purge. This step increased the initial
catalyst is not held a t the high pretreatment temperature catalyst activity of thionyl chloride-activated catalysts
any longer than necessary. by about 3 weight 5 butane conversion. The results
The variables studied for the reaction of thionyl chloride obtained with these catalysts are presented in Table 111.
with the platinum-alumina were again investigated with The catalysts contained from 3.4t o 3.9 weight '-c chlorine,

VOL. 8 NO. 4 DECEMBER 1969 359

 ~~~~ ~ ~ ~ ~ ~~

Table 111. Properties and Activity of Catalysts Prepared with Various Sulfur-Chlorine Compounds
0.30 mole HCl/hr./100 g. platinum-alumina for 3 hr. with H, diluent a t 1050" F. prior to and 400" F. following sulfur-chlorine compound
addition
Hexane isomerization at 280" F., 400 p.s.i.g., 1 LHSV, 2.5:1 H,-hexane mole ratio
Butane isomerization at 350" F.. 800 p.s.i.g., 1 LHSV, 2.5:l H2-butane mole ratio
CatalyTt Anal,yses
Chlorine. Surface area,
Sulfur-Chlorine Compounds ut ( I ).sq m g Catal)).stA C ~ Z L Z ~ Y
SOKL (0.23 mole SOX1, added over 3 hr. 3.9 155 89.2'~ hexane conversiono
a t 600" F.) 32.4'1, 2,2-dimethylbutane selectivity
SO! + Clz (2.1 moles SO, + 2.1 moles C1, added 3.9 155 89.05 hexane conversion"
over 10 hr. at 600" F.) 31.95 2,2-dimethylbutane selectivity
SX1, (0.35 mole SZCL added over 3 hr. 3.4 161 46.0% butane conversion
at 600" F.)
'Equilibrium hexane conuersion at 28'0" F. is about S I C c (Ridgeway a d Schoen, 1959)

had surface areas from 155 to 161 sq. meters per gram, is likely that this sulfur is present as sulfate from the
and were active for either butane or hexane isomerization. reaction of a portion of the sulfur trioxide liberated from
The catalyst inspections and activities are what would these activating agents and the alumina. I n the case of
generally be expected for the corresponding thionyl chlo- the thionyl chloride, sulfur dioxide is liberated, which
ride catalysts. does not react with the alumina.
The above catalysts were also analyzed for their sulfur Variations in Platinum Content. Although we believe the
content (Massoth, 1967). Where thionyl chloride was used main function of the platinum is to hydrogenate any
in the catalyst preparation, no sulfur in any form was coke precursors, it also has an effect on activity. Catalysts
found. This was also true with the sulfur monochloride- were prepared with platinum contents ranging from 0
treated catalyst, provided oxygen was employed either to 0.58 weight 70. Analyses showed the catalysts all con-
during or after the sulfur monochloride treatment or a tained from 3.9 to 4.6 weight % chlorine. However, the
high temperature nitrogen purge-e.g., 900" F.-followed activities of the catalysts varied from 37 to 46 weight
the sulfur monochloride treatment. The catalysts prepared ;'C (Figure 5 ) . Platinum levels of about 0.3 weight %
with sulfuryl chloride or a mixture of sulfur dioxide and or higher give the same butane conversion.
chlorine resulted in from 0.7 to 1.5 weight % sulfur. I t Variations in Platinum-Alumina. A series of commercial
aluminas containing from 0.35 to 0.71 weight % platinum
was treated with hydrogen chloride and thionyl chloride.
The results (Table IV) indicate a wide range of activities
and catalyst properties. Obviously, variations in the manu-
46
4.1 WT.% CI 0-0 0 facture of the aluminas play a major role in the properties

/
3.9 WT.% CI 3.9 W.%C I
of the chlorinated catalysts. No attempt has been made
to correlate alumina preparation and properties t o the
44
activity of the chlorinated platinum-aluminas. We merely
0 4.6 WT.% C I
wish to illustrate the variation in catalyst properties and
f
2 42
activity by varying the alumina source.
In addition to these aluminas, two silica-aluminas were
Ln
similarly treated and tested. Although, after treatment,
:
8
2 40
I?
m Table IV. Properties and Activity of Catalysts Prepared
with Aluminas from Other Manufacturers
38 0.30 mole HCl/hr./100 g. platinum-alumina for 3 hr. with H2
diluent
0 0 4.4 WT.% CI
0.36 mole SOClz/1O0 g. platinum-alumina a t 560" F. over 45 min.
4.1 WT.% CI

36
with N2diluent
Butane isomerization a t 350" F., 800 p.s.i.g., 1 LHSV, 2.5:1 Hz-
1 1 1 l butane mole ratio
0 0.2 0.4 0.6
Butane
PLATINUM CONTENT WT.%
Chlorine, Surface Area, Conversion,
Manufacturer Wt. cc Sq. M . i G . wt.
%
Figure 5. Effect of platinum level on catalyst activity
A 3.4 145 37
0.30 mole HCl/hr./100 g. platinum-alumina at 1050°F. for 3 hr.
B 3.9 169 22
with H2 diluent C 3.8 145 20
0.36 mole SOCI2/100 g. platinum-alumina a t 560" F. over 45 min. D 5.1 176 18
with Nz diluent E 4.5 175 9
Butane isomerization at 350°F., 800 p.s.i.g., 1 LHSV, 2.5:1 H2- F 4.6 171 2
butane mole ratio

360 l & E C PRODUCT RESEARCH A N D DEVELOPMENT

these catalysts contained 19.8 and 5.4 weight % chlorine, Egloff, G., Hulla, G., Komarewsky, V. I., “Isomerization
respectively, they were inactive for isomerization a t our of Pure Hydrocarbons,” p. 28, Reinhold, Kew York,
conditions. A probable explanation for their inactivity 1942.
is that most of the h,ydroxyl groups in silica-aluminas Evans, H. D., Fountain, E. B., Ross, W. E., Preprints
are attached to silicon rather than aluminum atoms (Basila 27th Midyear Meeting, API Division of Refining, San
Francisco, 1962.
et al., 1964). Chlorination occurs, with the replacement Goble, A. G., Haresnape, J. N. (to British Petroleum
of these hydroxyl groups leading to silicon-chlorine bonds, Co.), Brit. Patent 953,187 (March 25, 1964).
instead of aluminum-chlorine bonds obtained when alu- Goble, A. G., Haresnape, J. N. (to British Petroleum
mina is chlorinated. Goble and Lawrence (1964) suggest Co.), U. S. Patent 3,218,267 (Nov. 16, 1965).
that activity stems largely from the Lewis acid sites on Goble, A. G., Lawrence, P. A., Preprints, Proceedings,
the chlorinated aluminas which have extreme electron- Third International Congress on Catalysis, Amsterdam,
attracting capacity. 1964.
MacIver, D. S., Tobin, H. H., Barth, R . T., Catalysis
Acknowledgment
2, 485 (1963).
Massoth, F. E., Gulf Research & Development Co., Pitts-
burgh, Pa., private communication, 1967.
We acknowledge the valuable assistance of M. R . Basila, O’May, T. C., Sixth World Petroleum Conference, Section
F. E. Lutinski, and ’W.L. Kehl in the instrumental 111, Paper 4, FrankfurtlMain, 1963.
analyses and of K. I. Miller in the catalyst preparation. Ridgeway, J. A., Schoen, W., Preprints of Division of
Petroleum Chemistry, 4, 2A, Boston, 1959.
Riordan, M. D., Estes, J. H. (to Texaco), U. S. Patent
literature Cited 3,242,228 (March 22, 1966).
Thomas, C. A., “Anhydrous Aluminum Chloride in
Basila, M. R., Gulf Research & Development Co., Pitts- Organic Chemistry,” p. 787, Reinhold, New York, 1941.
burgh, Pa., private communication, 1965.
Basila, M. R., Kantner, T. R., Rhee, K . H., J . Phys. RECEIVEDfor review May 9, 1969
Chem. 68, No. 11, 31917 (1964). ACCEPTED September 17, 1969

HYDROCRACKING OF n-HEXANE AND

CYCLOHEXANE OVER ZEOLITES

ALEXIS VO O R H I E S , JR., A N D WI L L I A M J. HATCHER, JR.’

Louisiana State University, Baton Rouge, La. 70821

The lhydrocracking reaction, scission of carbon-to-carbon bonds plus hydrogenation,

w a s investigated, using n-hexane and cyclohexane as reactants. Two crystal-
line aluminosilicate catalysts were studied: the hydrogen, or acid, form of synthetic
faujasite and mordenite. Both catalysts were extremely active for hydrocracking.
Integral reactor data were correlated using a conventional Langmuir-Hinshelwood
model. The reaction rate constant w a s found to be a n Arrhenius function of
temperature with activation energies comparable to catalysts with much larger
pore diameters. The model was first-order with respect to the hydrocarbon,
and the effect of increasing the hydrogen partial pressure w a s compatible with
a “dual-site” mechanism. Typical reaction conditions were 500’ to 800’ F., 750
p.s.i.g., 10 moles of hydrogen per mole of hydrocarbon reactant, and liquid
hourlly space velocities of 1 to 8.

THE,hydrocracking of low-octane naphtha is a scheme sulfides on conventional supports such as silica-alumina.

being considered for producing liquefied petroleum gas This paper describes the results of an investigation of
(LPG) in areas lacking; natural gas. Commercial plants two zeolite catalysts for hydrocracking: the hydrogen,
have been installed in some areas. Typical hydrocracking or acid, form of synthetic faujasite and synthetic morden-
catalysts for this type of operation, and other feedstocks ite. Both catalysts were impregnated with palladium. Pure
as well, have been composed of dispersed metals or metal hydrocarbons, n-hexane and cyclohexane, were used as
’ Present address, Chemical Engineering Department, University reactants, chosen since they represent some typical molecu-
of Alabama, Tuscaloosa, Ala. 35486 lar structures found in naphtha.

VOL. 8 N O . 4 DECEMBER 1969 361