Section IV-Paper 2

TWO-STAGE PROCESS FOR PRODUCTION OF

NAPHTHALENE AND HIGH OCTANE GASOLINE
FROM CYCLE STOCK
BY

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

R. S. FRIEDMAN* AND S. M. KOVACH**

ABSTRACT. Earlier attempts to produce naphthalene from light cycle stocks by catalytic dealkylation
were unsatisfactory because of heavy loss of non-aromatic liquid through conversion to dry gas, large
consumption of hydrogen and difficult-to-control heats of reaction. One ,solution to this problem
was to operate with aromatic extracts of this cycle stock.
A two-stage operation has been devised, however, which avoids solvent extraction. In Stage I, a
catalytic light cycle stock is processed under mild hydrocracking conditions over chromia-alumina.
This converts the non-aromatics mainly to high quality gasoline, leaving a recycle oil which is highly
aromatic and low in Sulfur. In Stage II, the aromatic recycle oil is retreated under more severe
conditions to dealkylate the alkylnaphthalene components. This produces pure naphthalene and
additional high quality aromatic gasoline. Pure beta-methylnaphthalene may be produced as a by-
product if desired. It is estimated that the two-stage operation will produce overall about 19 wt.%
naphthalene, 54 vol.% gasoline having clear octane numbers of 88 F-2 and 99 F-1, and 5% of tar.
Hydrogen consumption will be about 2.5 wt.% (1450 CF/bbl).

RESUME. Des essais antérieurs sur l’obtention directe du naphtalène à partir de produits aroma-
tiques légers, par déalkylation en présence de ces catalyseurs, ne donnèrent pas satisfaction, en
raison d’une perte importante de liquide non-aromatique converti en gaz, d’une consommation
élevée d’hydrogène et du contrôle délicat des chaleurs de réaction. Une des solutions utilisées est
l’extraction des composés aromatiques de ces produits.
Un procédé à deux étapes a été essayé; il paraît offrir une alternative intéressante à l’extrac-
tion par solvent. Les essais ont été conduits de la façon suivante: 1. On fit passer une matière
première contenant des composés aromatiques légers sur un catalyseur en chrome et à l’aluminium,
dans des conditions d’hydrocracking douces. Cette opération convertit en grande partie les carbures
non-aromatiques en essences de haute qualité et donna une huile recyclée riche en aromatiques et
contenant peu de soufre. 2. L’huile aromatique recyclée fut traitée à nouveau sur le même cataly-
seur dans des conditions plus sévères pour obtenir la déalkylation des alkyl-naphtalènes. Ceci
donna du naphtalène pur, plus une essence aromatique supplémentaire de haute qualité. Il est
Pgalement possible d’obtenir du béta-méthylnaphtalène pur comme squs-produit.
D’après ces essais, on prévoit que ce procédé à deux étapes peut produire, à partir de produits
aromatiques légers catalytiques, 19 70 en poids de naphtalène, 54% d’essences volatiles avec un
nombre octane F-2 de 88 et F-1 de 99, et 5% de goudrons. La consommation d’hydrogène est estimée
à 2.5% en poids (1.450 pieds cubes par barrel).

Introduction ing refinery fractions, and occasionally small amounts

Petroleum has long been viewed as an alternate of durene and naphthalene from the intermediate cuts.
source to coal tar for the simple aromatic hydrocarbons Some samples of naphthalene have been distributed
required in increasing amounts by chemical industry. in the past [ 11 which are presumably produced from
Currently benzene, toluene, xylenes, and ethylbenzene heavier petroleum fractions, but at this writing we are
are being produced commercially from the lower boil- not aware of any commercial plants for this production.
* Research Associate ; ** Research Chemist, The Sinclair Research Laboratories, Inc., Harvey, Illinois, U.S.A.

17
 18 FWTH WORLD PETROLEUM CONORESS

There has been considerable interest over the years alumina base catalysts (Runs 3, 5, and 6 ) were more
in methods for obtaining naphthalene by dealkylation active than those with the cracking catalyst base
of the higher aromatics in cycle stocks derived from (silica/alumina; Run 4). Molybdena was a more active
catalytic cracking. The literature contains references promotor giving almost twice the conversion given by
to dealkylation by treatment with hydrogen under high chromia or tungstic oxide. In ultimate yield of naph-
pressure (hydrocracking) in the presence [2, 4,51 and thalene molybdena was also superior. However,

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

absence [ 6 ] of catalysts. Still another method described chromia exhibited a much lower coke-producing
is the high temperature, non-catalytic treatment in the activity.
presence of low partial pressure of hydrogen-a process
termed hydropyrolysis [3]. Study of Stage I-“Skimming”
The dealkylation reaction requires rather severe con-
ditions which result in non-aromatic components in Aromatic Extract Feed
the catalytic cycle stocks being converted to paraflìnic A few tests were made with the molybdena/alumina
gases, consuming large quantities of hydrogen and catalyst to obtain preliminary information on the possi-
evolving almost unmanageable heat in the reactor. For bility of converting the non-naphthalenic components
this reason most workers [5, 61 have preferred to to gasoline (“skimming” operation), leaving a recycle
operate with aromatic extracts of these stocks. oil suitable for napthalene production. An SO2extract
We wish to report our studies of a two-stage process of a light cycle oil was employed as feed in these initial
which avoids the extraction step. In Stage I, mild exploratory runs. Table II shows the effect of variations
treatment converts the non-aromatics to gasoline in feed rate and temperature.
range products, and in Stage II, a more severe treat- It is noted that fairly good yields of gasoline are
ment dealkylates the residual aromatic oil to naphtha- made in the “skimming” operation. In one case, the
lene and lower boiling aromatics. C,-40O0F gasoline had a F-1octane number of 99 clear.
A preliminary bench-scale study was made of cata- The recycle oil is much more aromatic than the feed
lysts, operating variables and feeds for each stage of (84-91% vs. 63%) which makes it an improved feed
the process. stock for demethylation to produce naphthalene.
The results of the “B” periods must be regarded as
Catalysts indications only, since they were obtained with cata-
A preliminary screening of catalysts was made with lysts already partially deactivated by service in the “A”
methylnaphthalenes as feed. As shown in Table I, the periods. The carbon figures can be considered signifi-

TABLE I
EVALUATION OF HYDROCRACKING CATALYSTS SHOWS THAT CHROMIA/ALUMINA
PR*ODUCES THE LEAST AMOUNT OF COKE
(1020°F, 400 psig, 5 moles HI, 1.0 LHSV, 5-hourruns)
8 4 6 6
Rwa No. 466 12% Chromia
Catalyst 10% Moos on 2% MOOSon on Ahüa 18.9% WO8
Hour on Stream AhOs (6% SiOa) SiOdAlnOa (2% MgO) on AlsOs
YMd8, Wt. 70 of FMd Feed’ second fifth second second second
Gasoline 6.4 6.8 6.4 6.8 6.6
Napthalene 68.1 48.9 28.7 26.0 28.6
(Ultimate) (62.8) (61.2) (48.9) (64.9) (60.2)
o-Methylnaphthalene 66.0 1.7 6.8 16.2 22.3 14.2
8-Methylnaphthalene 86.9 18.0 28.6 80.8 82.2 28.9
Coke for 6-hr. Run 2.2 3.06 0.88 1.81
(Ultimate) (2.6) (6.66) (0.78) (8.17)
B Consumption, CF/bbl
(for 6-hr. Run) 760 260

-
460 600

l R e i U ~Tar and Chemical Corp. Diesel Reference Fuel; %S 1.18: % N = 0.69.

* Carbon op eablsst.
 TWO-STAGE PROCESS FOR PRODUCTION OF NAPHTHALENE AND H I G H OCTANE GASOLINE 19

TABLE II
LOW-SEVERITY HYDROCRACKING ("SKIMMING") OF AN SOI EXTRACT
CONVERTS NON-AROMATICS T O HIGH OCTANE GASOLINE
(400 psig H,;10% Mû08 on Also&% SiOs)
HUn NO. 466 SIA SIB SSA SSB
Temp., OF 920-960 1010-44 . 920-964 1010-30

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

Catalyst Service, H o m 0-0.6 1.0-1.6 0-1.0 1.6-2.6
LHSV 2.0 2.0 3.8 8.8
Hz Mole Ratio 4.1 6.9 6.6 4.0
Prodyck. Wt. % of F e d Ftiedt
-
Gasoline C4 400'F 28.1 31.1 24.6 16.1
F-1 Octane Number CS -400") (99)
Recycle Oil (400660°F) 442-667'F 47.7 42.0 60.2 67.1
Bromine Number 22.9 6.0 6.2 6.6 8.8
% Ol&ns* 24.4 (8)t 6.2 7.7 8.2 10.3
% Aromatics** 62.6 ( I S ) + 91.3 90.3 84.3 86.7
:n 1.641 1.667 1.677 1.666 1.666
% Sulfur 2.81
API/60°F 19.6
Bottoms (>66O0F) 3.6 3.4 2.3 4.1
Ga 11.8 18.6 12.2 14.1
Carbon f o r Total Run 6.4 2.1
Ha Consumption. CF/bbl 680 890 690 640
by Bromine Number.
** absorption in 70/80 €hSoJpnO~(ASTM DlOi9) leas $6 oleíïna by Bromine Number.
t by F.I.A. (ASMT D1319).
$ 5 0 1 extract of catalytic lkht cycle oiL

TABLE III
"SKIMMING" STEP ON UN-EXTRACTED CYCLE OIL YIELDS
HIGH-OCTANE GASOLINE AND AROMATIC RECYCLE OIL
(400psig H,;10% Moo8 on Also&% Sioz)
Run Number 466-7fl 466-100 51W 518-5
Temp., O F 960 940 986 ioaa
* LHSV (1 Hour on Stream) 1.9 s.9 8.9 4
Productd. Wt.% of Feed Feed*+
Gasoline Cr+ 31.7 24.8 81.0 80.1
F-1 Octane Number 98.8
% Aromatics 64.1
Recycle Oil
% 400-460'F
-
163 ( 6.6).
49
18.2 (33).
61.8 62.6 U.6

460-480 16.6 (24.9)* 10.3 (72).

480620 88.4 (46.6). 13.2 (IS)*
620660 17.2 (61.3)' 6.6 (66).
660400 6.3 (88.7). 1.8 (90).
Aromaticst 61.2 88.7
% OldMt 26.8 6.4
Potential Naphthdene 29.4 29.6
% Sulfur 2.88 0.14 0.24 0.88 0.62
Dry Gas 11.6 6.8 10.2 16.9
Carbon on Catpunt 4.7 2.1 3.0 4.0
Snlfur on Catalyst 1.0 0.6 0.6 0.6
concentration of naphthalene derivatives.
** catalytic ï i h t cycle oil, API/60°F 21.8: 2.88% 8 ; i Q 422'F. 10% 462'. 60% 490°, 90% 634*! EP 668'F.
t by Bromine Number.
$absorption in 70/80 HsSOJp20. (ASTM Dl019) less % olefins by Bromine Number.
20 FIFTH WORLD PETROLEUM CONGRESS

cant only as indications because of the variable and 100 L

prolonged operation without regeneration.

Un-Extracted Feed 50 -
Table III gives the results of a short process variable 40 -
study that was made of the "skimming" step with the 30 -

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

molybdena/alumina catalyst, but this time the feed -
20
consisted of an un-extracted catalytic light cycle oil.
I t is seen from the above data that it is not necessary
w
W
IL

to start with an extract to produce a highly aromatic z -

0 IO
recycle oil. Considerable desulfurization takes place n
W
v,
(78 to 94% removal) so that it is not necessary to avoid U
m
feed stocks high in sulfur. It is interesting to note that s 5 -
sulfur removal is less effective at the higher tempera- ii 4-
9
tures, probably because of earlier deactivation (by
6 3-
coke) during the process period.
However, since the carbon make was found to be
+*
-I

2-
I:
rather high with the molybdena/alumina catalyst, the 0
æ
subsequent tests were made with the chromia/alumina O
r 1.0 -
œ
catalyst. The results obtained with a catalytic light i
cycle oil (422-568°F) on a 12% CrzO/AI2O3 (2% w
L>
Z
MgO) catalyst operating at 400 psig and 5 moles Ht O
-
0.5
are plotted in Figure 1.
These results may be discussed in terms of conver-
sions (a) in the 940-1050"/2 LHSV range and (b) at
1100"/0.5 LHSV:
(a) It is seen that as the operating temperature is /
raised from 940" to 1050" there is a gradual increase d
62 CONSUMPTION , I I
in dry gas, catalyst carbon and gasoline, while the yield
980 1020 1060 1100
of recycle oil decreases. (Not shown is the aromatic
TEMPERATURE,*F (05LHsv)
content of the recycle oil; this increases from 58% to
86%). Only a small amount (3% at 1050") of naph- FIG. 1-Resnlts of the "Skimming" Operation with a
Catalytic Light Cycle Oil and a Chromia Catalyst.
thalene is produced. The octane number of the gasoline
does not seem to change much in the operating range about 1050O and 2 LHSV. A thorough investigation
940' to 1020" (86-88 F-1), but at 1050" there was an was not made of the pressure variable but it was reason-
important increase to 95-96 F-l. The consumption of able to expect that lower pressures would increase
H, undergoes a large increase in the range 940" to catalyst coke and decrease hydrocracking of non-
1050' (0.13 vs. 1.4 wt.%). aromatic components, while higher pressures would in-
(b) At 1i0Oo/0.5 LHSV there are striking improve- crease dry gas formation. Changes in H, mole ratio
ments in gasoline quality (108.4 F-1 octane number) would be expected to give similar results.
and in naphthalene yield (8.4%), but these advan- The chromia catalyst deposits much less sulfur than
tages are more than offset by the 2- to %fold increase does the molybdena catalyst (max. of O . l l ~ ovs. 0.5-
in dry gas, catalyst carbon, and H, consumption, as 1.0% on feed). Thus, with respect to regeneration re-
well as 43% decrease in gasoline yield and 56% de- quirements, the molybdena catalyst is inferior to
crease in recycle oil yield over that produced at chromia on three counts :
1050°/2 LHSV. 1. It forms more carbon on the catalyst.
From these data it appeared that optimum condi- 2. I t combines with more sulfur which must be
tions for the Stage I ccskirnmingyy operation would be removed by regeneration.
 TWO-STAGE PROCESS FOR PRODUCTION OF N A P H T H A L E N E AND HIGH OCTANE GASOLINE 21

3. It undergoes valence changes which consume rived from Stage I ) two catalysts were tested, namely,
oxygen and liberate heat during regeneration, the molybdena and chromia catalysts. The feeds were
and which also add to the consumption of the aromatic recycle oils produced in Stage I.
hydrogen; whereas with chromia, only a very The results appear in Table V. As in Stage I, the
small proportion of the metal oxide undergoes molybdena catalyst produced far more carbon than
oxidation and reduction. the chromia catalyst (9-llyo vs. 4.4%) and was there-

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

-~ ~
fore considered not suitable for this operation.
TABLE IV Because of the difficulty of preparing sufficient feed
CATALYTIC "SKIMMING' PRODUCES A HIGHER stock for Stage II, the only variable investigated with
OCTANE GASOLINE AND A MORE VALUABLE BY- the chromia catalyst was that of temperature. As the
PRODUCT OIL THAN DOES THERMAL CRACKING temperature was increased ( 9 9 0 " - + 1 0 5 0 " ~IlO"),
1
Feed Catalytic Light Cycle Oil
APV6O"F 21.6 21.8
the yield of naphthalene increased from 13.8 to 19.5
o/o Sulfur 1.96 2.38 and 25.9y0, and carbon make increased somewhat
% Aromatics 41.9 61.2 more rapidly, from 1.09 to 2.35 and 4.34%, respec-
iBP. O F 444 422
60%. 'F 638 490 tively.
90%. 'F 600 634 The gasoline yield increased at first, but leveled out
EP, O F 637 668
Operating Conditwm at about 21y0. The high degree of aromaticity of this
Temp., "F 900 1000 1060 product is in agreement with the high octane values.
Pressure, psk- 660 400 400
LHSV (t) 2 2
Hz Mole Ratio - 6/1 6/1
Analysis of Stage II Naphthalene Product
Catalyst Thermal** - 12% CraOa/AlnOs (2% MgO) -
Products. W t . 70 By U-V analysis, the naphthalene cut (400-450°F)
Dry Gas 6.4 6.6 13.6 obtained by distillation of the liquid product from
Carbon on Catalyst - 0.6* 1.3.
Gasoline (CI-EP) 29.1 29.8 83.1 Run 20 (chromia catalyst) contained
F-1 Octane Number 73 86 96
Cycle Oil (EP, OF) 26 (600') 61.7 (640') 46.6 (640°)
66.0% naphthalene
API/6OoF 12 17.8 13.1 23.1% methylnaphthalenes
% Sulfur (a 0.73 0.86
Tar 39.6 3.6 3.9 89.1% total naphthalenes.
API/60°F 0.1 - -
- - - The results of refractionation of this cut are shown
100.0 91.1 96.6 in Table VI. Fractions 2 to 6 contained the highest
Ha Consumption
CF/bbl - 61 300
concentration of naphthalene and were blended and
Wt.% 0.08 0.49 analyzed. In Table VI1 the composition of this blend
% S on catalyst is negligible. is compared to various commercial grades of naptha-
**thermal cracking in pilot plant.
t 6/1 recycle ratio.
lene. It is seen that the Stage II product matches the
$ eYcie oil plus tar average 2.68% Sulfur. best of the commercial grades in freedom from nitro-
gen-containing impurities, and surpasses them with
As shown in Table IV, the "skimming" operation respect to Sulfur-containing and benzene-insoluble
over the chromia/alumina catalyst has some definite impurities.
advantages over the thermal cracking of cycle stock.
Although the feed stocks do not have the same boiling Beta-Methylnaphthalene
range, the comparison does indicate that the catalytic It is interesting to note that there is a high degree of
"skimming" operation produces gasoline higher in F-1 selectivity in the demethyIation of methyl derivative
octane number (86 to 95 vs. 73) and by-product liquid of naphthalene. Thus, in Run 518-20 (Table VIII)
of more value as an aromatic solvent and as feed stock whe're the mono-methylnaphthalene fraction consisted
for naphthalene production. of 98Y0 beta and 2% alpha, it should be possible to
produce rather pure beta-methylnaphthalene, by sim-
Study of Stage II-Production of Naphthalene ple distillation.
from Recycle Oil In spite of its greater resistance to demethylatiori,
For Stage II (the demethylation of recycle oil de- no particular difficulty is anticipated in converting
22 FIFTH WORLD PETROLEUM CONGRESS

TABLE V
IN STAGE II T H E RECYCLE OIL FROM STAGE I I S CONVERTED TO
HIGH OCTANE GASOLINE AND NAPHTHALENE
Catalyst 10.14% MOOa/AhOs (5% Sioz) 12% CrzOdAlnOs (2% MgO)
Stage I II II I II II II
Ran Number 79 + 99 8 9 6,13,14 18 19 20

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

Feed Lt. Cycle Oil Recycle Oil Lt. Cycle Oil Recycle Oil
B.P.. OF 422-668 400-600 422-668 400-600
API/60°F 21.8 16 21.8 12.7
nD~
1.6261 1.6620 1.6261 1.6740
Specific Dispersion 19e 268 198 270
% Aromatics 51.2 83.7 61.2 81.3
Temp., " F 950 1025 986 1060 990 1060 1100
Pressure, psig 400 400 400 400 400 400 400
Hz Mole Ratio 4 6.3 6.9 6-6.6 6.3 6.6 6.7
LHSV 1.8 1.1 1 2 0.66 0.64 0.65
Hours on Stream 1 1 1 1 & 2 2 2 e
Products, Wt. $6 of Feed
Liquid 80.4 63.9 73.4 79 89.4 86.6 77.8
C4+ in Gas 3.8 - - 6.6 0.8 0.8 0.6
Dry Gas 11.6 - - 14 9.2 13.8 19.3
Carbon on Catalyst 4.7 9.4 10.9 1.6 1.09 2.36 4.84
Sulfur on Catalyst 1.0 0.7 - O 0.11 - -
Total Recovery - - -
100.6 102.66 102.04
Liquid, Wt. % of Feed
Gasoline (Ca+ -400'F) 16.2 20.6 21.0
F-1 Octane Number ( C6-400°F) # (102*) (104.4**) (106.6%)
Naphthalene 13.8 19.5 26.9
Alkylnaphthalenest 36.1 29.2 17.0
Other Componentst 20.7 9.6 1.6
Bottoms$$ 3.6 6.8 6.3
€is Consumption
CF/bbl 436 664 370 832 680 1000 1280
Wt. % of Feed 0.68 1.07 0.62 14.4 1.11 1.62 2.08

* isooctane + 0.136 ml TEL, ** isooctane + 0.3 mi TEL, 9 isooctane + 0.54 m l TEL, 9 5 distillations were troubled with plugging by naph-
thalene, t recycle oil, # Aviation Gasoline Advisory Committee Chart, :$ distillation bottoms.

TABLE VI
DISTILLATION OF T H E STAGE II 400-450°FPRODUCT REMOVES MOST OF T H E
NON-NAPHTHALENIC IMPURITIES AS FRONT ENDS AND T H E METHYL-
NAPHTHALENES AS BOTTOMS
Number B.P. "C Methulnaphthalenst
Fraction Wt.% O F M.P.* Naphthalenet alpha beta
1 16.9 i420 - 60.8% 0.3 0.7%
2 7.6 421 76.6 93.6 - 0.4
3 10.1 421 17.8 94.1 0.1 0.8
4 10.7 421 17.6** 96.2 0.3 0.3
6 10.8 422 77 94.1 - 0.6
6 9.3 423 77 94.2 - 0.8
7 10.8 433 - 81.4 - 8.0
8 6.7 450 - 61.6 - 37.8
Bottoms 14.8 - - 1.6 8.6 91.7

* from rough determination of time/temp. cooling curve.

"*<0.05,(rJ, S.
t by U-V.
 . .. .. .

TWO-STAGE PROCESS FOR PRODUCTION OF NAPHTHALENE AND HIGH OCTANE GASOLINE 23

TABLE VI1
NAPHTHALENE PRODUCT OF STAGE II IS PURER THAN
MOST COMMERCIAL GRADES
Purity Benzene
Grade M.P.. OC Color** b y F.P. %N %S Insolubles, yo
Comm. 74" Crude 74.6 (600-b) 90.0% 0.11 0.36 0.36
Comm. 78" Crude 78.2 í500+) 96.3 0.018 0.406 0.038

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

79.0 (600+) 97.7 0.044 0.36 0.039
Comm. Refined 79.9 20 99.3 0.001 0.18 0.039
Comm. Recryst. 79.9 20 99.3 0.000 0.231 0.047
Stage II Product* 78.2 20 96.3t 0.000 0.006 0.000

*blend of cuts 2 to 6.
** APHA (American Public Health Association) Standard Methods, 1946 edition, p. 14.
t 0.6% methylnaphthalenes : 0.1% paraffin.

beta-methylnaphthalene to naphthalene, since as indi-

TABLE VI11 cated in Table IX, the demethylation proceeds satis-
ALPHA-METHYLNAPHTHALENE IS MORE EASILY factorily at 1100°F.
DEMETHYLATED THAN BETA-
METHYLNAPHTHALENE Overall Results of Two-Stage Operation
Run 466-3B 466-5B 518-ZO
The results of the two steps, using the chromia
catalyst for both operations, are summarized in Table
Temp., "F 1020 1026 1100
Pressure, psic 400 400 400 X.
LHSV 1 1 0.6 It is seen that the ultimate yield of naphthalene will
Hours 1-2 1-2 0-2
be about 19.2% (of the original feed). Based on the
Feed, grams a-Methylnaphthalene*" Recycle Oil alkylnaphthalenes presented in the cycle stock (see
alpha 66.0 7.1
beta 36.9 27.8 Table III), this is approximately 65y0of the theoreti-
(13.0)* cal, which compares favorably with the figure of
' Catalyst Molybdena/ Chromia/ Chromia/ 62.3% of theoretical for naphthalene obtained from
Alumina Alumina Alumina
a mixture of alpha- and beta-methylnaphthalene
Products, grams (Table I, Run 3).
Naphthalene 63.1 26.0 26.9
alpha-Methylnaphthalene 1.7 22.3 0.34 The overall yield of gasoline from the two stages is
beta-Methylnaphthalene 13.0 32.2 16.63 estimated to be about 54 vol. per cent with clear octane
values of approximately 90 F-2 and 98 F-l. It will be
* polymethylnaphthalenes.
** Reilly Tar and Chemical Corp. Diesel Reference Fuel. noted that the gasoline produced in Stage II (see
Table X I ) consists chiefly of methylated benzenes.

TABLE IX
DEMETHYLATION OF BETA-METHYLNAPHTHALENE TAKES PLACE
OVER THE CHROMIA/ALUMINA CATALYST
(1100°F. 400 psig H,)
Run N o . 518 2iA SlD 2SA 2lD
LHSV 1.0 1.0 0.6 0.6
Hours on Stream 0-1 3-4 0-1 1-4
Products. Wt. of Feed Feed
Gasoline 2.7 2.2 3.4 2.2
Naphthalene ( u l t ) fI.6 41.0(76) 34.8(70) 49.6(66) 49.7(67)
a-Methylnaphthalene 19.0 1.0 1.9 1.0 0.8
&Methylnaphthalene 71.7 44.2 47.1 23.2 26.6
Tar (>460"F) 3.9 1.0 8.2 4.6
Carbon on Catalyst 0.41 0.42
Total Liquid 96.0 93.4 92.2 91.)
24 FIFTH WORLD PETROLEUM CONGRESS

TABLE X
GOOD YIELDS OF HIGH OCTANE GASOLINE AND NAPHTHALENE
RESULT FROM THE TWO-STAGE OPERATION
(Chromia/Alumina Catalyst; 400 psig HB)
Stage I-"Skimming" Staoe II-Demethulation Combined Yields*
Run Number 14 20
-

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

Feed Cat. Lt. Cycle Oil Recycle Oil
Temp., " F 1060 1100 -
LHSV (approx.) 2 0.5 -
Products, W t . % of Feed to Stage
Gasoline C4-400"F (Vol.%) 33.6 (40.7) 21.0 (22.8) 46.0 (64.8)
-
% Aromatics?
70 Sulfurt
68.1
0.05 -
100

Octane Number, F-lf 96.8 106.6 98.7

Octane Number, F-2t 84 94.0 88.0
Heavier Liquids
Recycle Oil 49.8 60.6 -
Naphthalene (uit) - 26.9 (19.2)
Methylnaphthalenes - 17.0 -
Other Components - 7.6 -
Tar 1.8 6.800 6.4
Carbon 1.7 4.8 4.4
Dry Gas 18.6 19.3 26.6
Hz Consumption
CF/bbl 710 1280 1460
Wt. 70of Feed 1.28 2.1 2.6

* estimated yields based on original light cycle oil and assuming recycle of recycle oil to extinction.
** 480" plus.
t Cs-400°F.

Table XII summarizes the composition of the dry A two-stage operation has been devised, however,
gas product from each step. H,S is a by-product of which appears to offer an attractive alternate to solvent
interest, especially from high-Sulfur cycle stocks. The extraction for preparing a suitable feed stock "his was
overall yield of H2S is estimated at 1.8% or roughly tested as follows:
75% of the sulfur present in the initial cycle stock. 1. A catalytic light cycle stock (51% aromatics;
2.38% S) was processed under mild hydro-
Summary cracking conditions over chromia/alumina.
As reported by others, naphthalene can be obtained This converted the non-aromatics largely to
in good yield by dealkylation of methylnaphthalenes in high quality gasoline, leaving a recycle oil
an operation very similar to but somewhat more severe which was highly aromatic (85% aromatics)
than hydroforming. As catalyst for this process, molyb- and low in sulfur (0.86%).
dena/alumina is more active than chromia/alumina; 2. The aromatic recycle oil was retreated under
however, chromia/alumina produces much less catalyst more severe conditions over the same catalyst
coke and sulfur, and is much easier to regenerate. to dealkylate the alkynaphthalenes present
Earlier attempts to produce naphthalene directly therein. This produced additional high quality
from light cycle stocks by dealkylation in the presence aromatic gasoline and a good yield of pure
of these catalysts were unsatisfactory because of heavy naphthalene. Pure beta-methylnaphthalene
loss of non-aromatic liquid through conversion to dry may be produced as a by-product if desired.
gas, large consumption of hydrogen and difficult-to- On the basis of these tests, it is estimated that the
control (in fixed beds) heats of reaction. One solution two-stage operation will produce from the catalytic
to this problem is to operate, as others have, with light cycle oil about 19 wt.% naphthalene, 54 vol.<r,
aromatic extracts of this cycle stock. gasoline having clear octane numbers of 88 F-2and
 TWO-STAGE PROCESS FOR PRODUCTION OF NAPHTHALENE AND HIGH OCTANE GASOLINE 25

TABLE XI
INSPECTION OF GASOLINE (Ca-400°F)
Stage I Staqu I I
Run No. 518 14 20
Yield, Wt. % of Orig. Feed 27.5 10.2
API/GO”F 44.5 24.8

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

(Cr+ in Gases) (5.9) (0.3)
A S ï M Dzatii D a t a t from Fractionation
iB.P. 96 200°F
10% 159 -
20 206 -
30 250 300
40 282 -
50 305 327
60 318 360
70 331 -
80 347 384
90 362 -
EP 380 400
Rec. 98.0
Res. 0.8 Aromatic l’vpe-‘
Loss 1.2 0.6 Benzene
7.0 Toluene
%S 0.03 30.4 Xylenes
P 27.4 39.3 Trimethylbenzene
O 9.7 18.2 Tetramethylbenzene
N 4.8 4.5 Pentamethylbenzene
A 58.1
nD 1.4530 1.5065**
Vapor Pressure, pounds 3.4 -
Octane Number, F-1
Octane Number, F-2
95.8
84.0 :E{ isooctane
plus 0.64 ml TEL

* b y Mass Spectrometer.
** U-V shows 4.3% naphthalene and 0.4% methylnaphthalene. Corrected n i5 = 1.5026 indicates 100% aromatic.
‘itrue boiling points.

99 F-1, and 5% of tar. Hydrogen consumption will be

TABLE XII about 2.5 wt.% (1450 CF/bbl).
INSPECTION OF PROCESS GASES
Stage I Stage I I
Experimental Procedure
Run Number 518 14 20 The apparatus used for these experiments was a 1”
Yield, Wt. % of Orig. Feed i.d. flow reactor fabricated of #306 stainless steel.
íH2) (80 mol %) (12.8 mol yo)
HzS 1.68 0.21 The product flowed through a filter capsule, filled
ci 4.56 5.97 with glass wool held in place with a metal screen, to
C2 4.65 2.82 a knockout pot. The liquid product separating in the
c3 4.4 0.78
- - knockout pot was drawn off manually into a one-liter
Total Ca-Paramnie Gases 13.61 9.57 three-necked flask. The gases liberated from the flask
i-C4 0.98 - and the gas released from the top of the knockout pot
n-C4 1.68 0.21
C4 = 0.57 0.09 through a needle valve were combined and passed
i-Cs 1.34 - through a Dry-Ice condenser and thence to a wet-test
n-Cs 0.56 -
CS = 0.39 - meter. The effluent gas stream was sampled continu-
ce = 0.19 - ously and collected over water. Where it was desired
cs 0.19
- -
- to obtain more accurate figures as to H,S make, the
TOMC b + 6.90 0.8 effluent gas stream was sampled directly at regular
26 FIFTH WORLD PETROLEUM CONGRESS

Normally the catalyst tube was placed in the elec- 3. Beckberger, L. H., U.S. Patent 2,674,635 (April
trically-heated furnace when it was below about 6, 1954).
200°F, connected up and pressure-tested with H 2 at 4. Friedman, B. S., U.S. Patent 2,700,638 (January
400 psig. The pressure was released and the furnace 25, 1955).
heated to about 50°F below operating temperature. 5. Roebuck, Alan K. and Evering, B. L., Ind. Eng.
The reactor was then pressured up with H2 to 400 Chem., 50, 1135 (1958).

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

psig” and the catalyst reduced with H2 (500-850 6. Bethea, S. R., Heinrich, R.L., Souby, A. M. and
GHSV) for one hour. During this period the furnace Yule, L. T., Ind. Eng. Chem., 50, 1245 (1958).
temperature was raised and the catalyst bed lined out 7. Wiley, J. L. and Anderson, H. C., “Bibliography
at the selected operating level. of Pressure Hydrogenation,” US.Bureau of Mines
The liquid feed was started, and the catalyst bed Bulletin T485, Washington, D. C . (1956).
temperatures further lined out. This usually required This paper was presented June 1, 1959, by B. S.
about 5 to 10 minutes for the runs a t 2 to 4 LHSV FRIEDMAN.
(liquid hourly space velocity), and 20-30 minutes for
the 0.5 to 1 LHSV runs. The liquid and gas products Discussion
produced during the lining-out period were discarded. W. J. BORNS(Socony Mobil Oil Co.,Paulsboro, N.J., .
After operation for the selected period, the liquid U.S.A.). This paper discloses a unique and poten-
and trap products were segregated, and purging was tially valuable process sequence for refining aromatic
accomplished by continuing the flow of H, at 400 psig light gas oils. The gasoline raw material is separated
for 20 minutes. The reactor was allowed to cool under from the naphthalene raw material by the conver-
pressure. The purge liquid and gases were discarded. sion technique itself. There are three interesting
The liquid product was combined with the trap points on which the authors might wish to comment
condensate and in most cases dried over K2C03.The in greater detail.
product was distilled through a 18” x 25 mm column First, I wonder if the experimental data show
(stainless steel protruded packing) into the following what happens to the decalins and tetralins in the
fractions : catalytic gas oil.
i - 400°F gasoline Second, I would like to ask Dr. Friedman if he has
400 - 450 naphthalene further available information on naphthalene fonna-
450 - 475 methylnaphthalenes tion from anything in addition to the bicyclics in the
475 - 520 dimethylnaphthalenes input material.
520 - 550 trimethylnaphthalenes Last, I think it has occurred to many of you that
550 - 600 the molybdena/alumina catalyst contained silica, pre-
Bottoms sumably to increase cracking activity. This type of
Reflux ratios employed were 15/1 to 350°F overhead activity is notably deficient in molybdena/alumina
temperature, 50/1 to 475 OF, and 15/1 thereafter. when used for this purpose. However, addition of
Naphthalene and its methylated derivatives were silica increases coke formation substantially. On the
determined by U-V analysis. Gases were analyzed by other hand, it is quite well known that chro-
mass spectrometry. mialahmina has more cracking function than
needed for aromatization and dealkylation under
Ref erencea these conditions. The excess cracking function is cor-
1. Chemical Week, 78,72 (March 31, 1956) ; Chem- rected in these experiments by addition of magnesia.
ical Week, 79,60 (August 25, 1956) ; Chem Eng. Potassia is also used frequently. The alkali has the
News, 34, 1018 (1956). additional advantage of reducing coke formation
2. Beckberger, L. H., U.S. Patent 2,653,176 (Septem- very greatly. The alkali appears to suppress coke
ber 22, 1953). formation selectivity, without destroying cyclization
--
function. Would the authors care to comment on the
*Molydena/alumina pretreated at atmospheric pressure was possibilities of even better balance between the vari-
quite inactive as a demethylation cataiyst. ous activities of a polyfunctional cataiyst.
 TWO-STAGE PROCESS FOR PRODUCTION OF NAPHTHALENE AND HIGH OCTANE GASOLINE 27

B. S. FRIEDMAN
replies. Dr. Borns brings up some would have such a long life. I recall we did some
very interesting points. The decalin and tetralin, if experiments with molybdenum and tried various
present, and they are present in the cycle stock, are methods of precipitation and promotion and ob-
converted very readily to naphthalene. We have only tained no indication that molybdenum would be
one wish, namely, that there were more of them in the catalyst, and we did all our subsequent work on
the cycle stock. chromia.

Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023

Then he asked if naphthalene was derived from H. W. EARHART (Humble Oil and Refining Com-
anything else besides fused-ring bicyclics. We were pany, Baytown, Texas, U.S.A.). Dr. Friedman has
interested in that, also, but could find no evidence shown the composition of the gasoline produced,
that anything but fused-rings were giving us the which he said was all aromatic. Now I was curious
naphthalene. He indicated that on the basis of our whether in the 28% xylenes, the 37% trimethylben-
work with chromia and the general state of knowl- zenes and the 19% tetramethylbenzenes, the isomers
edge on the effect of acidity on the catalyst leading were present in the thermodynamic equilibrium con-
to coke formation and also leading to improved ac- centrations one would predict from the temperature
tivity, that an alkalized molycatalyst ought to be at which he operated.
useful for this reaction. It might make less carbon
B. S. FRIEDMAN replies. I do not recall looking
and still give us a lot of activity.
into that particular problem. We were merely grate-
We, too, were loath to leave molybdenum as a
catalyst, because it had a good background in com- ful that it was so concentrated in aromatics. I think
mercial production. I t had been used for many years I did freeze out some durene from the tetramethyl-
in the reforming operation involving very high pres- benzene fraction, but other than that I can not say
sures, and we were not sure that the chromia catalyst in what proportions the isomers ocdurred.
Downloaded from http://onepetro.org/WPCONGRESS/proceedings-pdf/WPC05/All-WPC05/2083223/wpc-8302.pdf by Universiti Teknologi Malaysia user on 12 August 2023