Note: Descriptions are shown in the official language in which they were submitted.
127854~
~ACKGROUND OF THE INVENTION AND PRIOR ART
I. Field of the Invention
Thi3 invention rela~e3 to a process for re~orming
with sulfur-sensitive, polymetallic platinum-containing
catalysts wherein sulfur is added to an intermediate or
final reactor, or reaction zone, or zones, and excluded from
the initial reactor, or reaction zone, of the series.
II. Background _nd_Problems
Catalytic reforming, or hydroforming. is a well
establi3hed industrial process employed by the petroleum
industry for improving the octane quality of naphthas or
straight run gasolines. In reforming, a multi-functional
catalyst is employed which contains an acid component and a
metal hydrogenation-dehydrogenation (hydrogen transfer) com-
ponent, or components, substantially atomically dispersed
upon the 3ur~ace of a porous-, inorganic oxide support,
notably alumina. Nobie metal catalysts, notably of the
platinum type, are currently employed as metal hydrogena-
tion-dehydrogenation components, reforming being defined as
the total effect of the molecular changes, or hydrocarbon
reactions, produced by dehydrogenation of cyclohexanes and
dehydroisomerization of alkylcyclopentanes to yield
aromatics; dehydrogenation of paraffins to yield olefins;
dehydrocyclization of paraffins and olefins to yield
aromatics; isomerization of n-paraffins; isomerization of
alkylcycloparaffins to yield cyclohexanes; isomerization of
s-ubstituted aromatics; and hydrocracking of paraffins which
produces gas, and inevitably coke, the latter being progres-
sively deposited on the catalyst as reforming is continued.
In a reforming operation, one or a series of reac-
tors, or a series of reaction zones, are employed. Typi-
cally, a series of reactors are employed, e.g., three or
four reactors, these constituting the heart of the reformin3
unit. Each reforming reactor is generally provided with a
fixed bed, or beds, of the catalyst which receive downflow
~;~'7~5~
feed, and each i~ provided with a preheater or interstags
heater. A naphtha feed, with hydrogen, or recycle hydrogen
gas, is concurrently pa~sed through a preheat furnace and
reactor, and then in sequence through sub~equent interstage
heaters and reactors of the series. The product from the
last reactor is separated into a liquid fraction, and a
vaporous effluent. The former is recovered as a C5+ liquid
product. The latter is a gas rich in hydrogen, and u~ually
contains ~mall amounts of normally gaseous hydrocarbons,
from which hydrogen is separated and recycled as "recycle
sas" to the proces~ to minimize coke production. In conven-
~ onal operations, the recycle gas, which generally contains
moisture and hydrogen sulfide impurities, i3 passed through
a recycle gas drier which removes much of the moisture and
hydrogen sulfide prior to the introduction of the recycle
gas into the first reactor of the series.
The ~um-total of the reforming reactions, supra,
occurs as a continuum between the fir~t and last reactor o~
the series, i.e., as the feed enter~ and passes over the
~irst fixed catalyst bed of the first reactor and exits from
the last fixed catalyst bed of the last reactor of the
series. The reactions which predominate between the several
reactors differ dependent principally upon the nature of the
feed, and the temperature employed within the individual
reactors. In the initial reaction zone, or first reactor,
which is maintained at a relatively low temperature, the
primary reaction involves the dehydrogenation of naphthenes
to produce aromatic~. The isomerization of naphthenes,
notably C5 and C6 naphthene , also occurs to a considerable
extent. Most of the other reforming reactions also occur,
but only to a lesser, or smaller extent. There is relative-
ly little hydrocracking, and very little olefin or paraf.in
dehydrocyclization occurs in the first reactor. Naphthene
dehydrogenation is an endothermic reaction, and consequently
the reactions in the first reactor are extremely endo-
thermic, generally accounting for as much as 2/5 to 3/5 o~
the total observed temperature difference (~T) across the
12'7~3~;44
several catalyst beds contaLned in the several reactors of
the series. Within the intermediate reactor(s), or reaction
zone(s), the temperature 19 maintained somewhat higher than
in the first, or lead reactor of the series, and it is
believed that the primary reactions in the intermediate
reactor, or reactors, involve the i30merization of naph-
thenes and paraffin~. Where, e.g., there are two reactors
disposed between the first and last reactor of the series,
it is believed that the principal reaction involves the
isomerization of naphthenes, normal paraf~ins and isopara~-
fins. Some dehydrogenation of naphthenes may, and usually
does occur, at least within the first of the intermediate
reactors. There is usually some hydrocracking, at ieast
more than in the lead reactor of the series, and there is
more olefin and paraffin dehydrocyclization. The net effect
of the reaction~ which occur in this reactor are endother-
mic, and though the temperature drop between the feed inlet
and feed outlet i9 not a3 large as in that of the initial
reactor (even though the second reactor generally contains a
larger catalyst charge than the initial reactor), it is
nonetheless considerable. The third reactor of the series,
or second intermediate reactor, is generally operated at a
somewhat higher temperature than the second reactor of the
series. It i3 believed that the naphthene and paraffin
isomerization reactions continue as the primary reaction in
this reactor, but there is very little naphthene dehydro-
genation. There is a further increase in paraffin dehydro-
cyclization, and more hydrocracking. The net effect of the
reactions which occur in this reactor is also endothermic,
though the temperature drop between the feed inlet and feed
outlet is smaller than in the first two reactors. In the
final reaction zone, or final reactor, which is typically
operated at the highest temperature of the series, it is
believed that paraffin dehydrocyclization, particularly the
dehydrocyclization of the short chain, notably C6 and C7
par~ffins, is the primary reaction. The isomerization reac-
tions continue, and there is more hydrocracking and coke
1278S44
formation in this reactor than in any o~ the other reactors
of the series. The net e~fect of the reactions which occur
in this reactor is al90 generally endothermic.
Platinum is widely commercially used in the pro-
duction of reforming catalysts, and platinum-on-alumina
catalysts have been commercially employed in refineries ~or
the last ~ew decades. In more recent years polymetallic
cata~yats have been used. These are catalySts wherein addi-
tional metallic components have been added to platLnum as
promoters to further improve the activity or selectivity, or
both, of the basic platinum catalyst, e.g., iridium,
rhenium, tln, and the like. Such catalysts po~sess superior
activity, or selectivity, ar both, as contrasted with the
basic platinum catalyst. Platinum-rhenium catalyst3 by way
of example possess admirable selectivity as contrasted with
platinum catalyst~, selectivity being defined as the ability
of the catalyst to produce high yields of C5 liquid pro-
ducts with concurrant low production of normally gaseous
hydrocarbons, i.e., methane and other gaseous hydrocarbons,
and coke.
The activity of the catalyst gradually declines
due to the build-up of coke. Co~e formation is believed to
result from the deposition of coke precursors such as
anthracene, coronene, ovalene, and other condensed ring
aromatic molecules on the catalyst, these polymerizing to
form coke. During operation, the temperature o~ the process
is gradually raised to compensate ~or the activity 1099
caused by the coke deposition. Eventually, however, temper-
ature increages cannot compensate for tne loss in catalytic
activity and hence it becomes necessary to reactivate the
catalyst. Consequently, in all processes of thi~ type the
catalyst must necessarily be periodically regenerated by
burning off the coke at controlled conditions.
Two major types of reforming are generally prac-
ticed in the multi reactor units, both o~ which necessitate
periodic reactivation of the catalyst, the initial sequence
of which requires regeneration, i.e., burning the coke from
1278S~4
the catalyst. Reactivation of the catalyst Ls then com-
pleted in a sequence of steps wherein the agglomerated metal
hydrogenation-dehydrogenation component~ are atomLcally
redispersed. In the semi-regenerative proce3s, a proceas of
the rirst type, the entire unit is operated by gradually and
progressively Lncreasing the temperature to maintain the
actlvity of the catalyst caused by the coke deposition,
until finally the entire unit is shut down for regeneration,
and reactivation, of the cataly3t. In the second, or cyclic
type of proce~s, the reactors are individually isolated, or
in effect ~wung out of line by various manifolding arrange-
ments, motor operated valving and the like. The catalyst is
regenerated to remove the coke depo~its, and then reacti-
vated while the other reactors of the series remain on
stream. A "3wing reactor" temporarily replaces a reactor
which is removed from the series for regeneration and
reactivation of the catalyst, until it is put back in
series.
Change in the total, or overall, aT is a good
indication of changing performance in the reactors during an
operating run, and correlates well with the ability of the
reaction system to produce reformate octane value; which
normally decrease~ throughout the run. In cyclic opera-
tions, in particular, the decline in the temperature drop
across a catalyst bed is sometimes used as a criterion for
selecting the next reactor candidate for regeneration of its
catalyst charge.
Essentially all petroleum naphtha or synthetically
derived naphtha feeds contain sulfur, a well known catalyst
poison which can gradually accumulate upon and poison
reforming catalysts. Most of the sulfur, because of this
adverse effect, is removed from feed naphthas, e.g., by
hydrofining and subsequent contact with guard beds packed or
filled with sulfur ad30rbents. The polymetallic refor~ing
catalysts are particularly sulfur-sensitive, and particular-
ly susceptible to sulfur poisoning. The presence of even
small and virtually infinitesimal amount3 of sulfur in
~2t7~S4~
-- 6
either the naphtha feed or hydrogen recycle gas, or both, lt
has been observed, adversely affects 5he process, and per-
formance of the catalysts. Wherea3 various improvements
have been made in adsorbents, and in the operation of guar~
beds to eliminate sulfur ~rom the naphtha feed and hydrogen
recycle gas, the complete elimination of sulfur from the
naphtha does not appear practical, if indeed poqsible, and
sulfur inevita~ly appear3 in the process. The effect of
sulfur in the naphtha, even in concentration ranging only a
few parts per million is that, in the overall reforming
operation, the yield of hydrogen, aromatics, and C5l liquid
yield decreases as sulfur builds up and increases in the
system, there is an increase in the rate of cataly3t deacti-
vation, and in the total production of C1-C4 light gases.
III. Objects
It iq, accordingly, the primary objective of this
invention to provide a new and improved proces_ useful in
the operation of reforming unit3 which employ highly active
sulfur-sensitive polymetallic platinum-containin~ catalyst
to produce high octane gasolines.
A specific object is to provide a novel process as
characterized, but particularly a star~-up procedure which
will provide improved hydrogen purity and aromatics produc-
tion, increased C5~ liquid yield, and decreased C4- light
gas production in the operation of a reforming unit.
I~. The Invention
Theqe objects and others are achieved in accord-
ance with the present invention, embodying a process
whereln, in a series of reforming zones, or reactors, each
of which contains a bed, or beds of catalyst, the catalyst
in each of which i9 constituted of a sulfur-3ensitive,
polymetallic platinum-containing catalyst which contains
little or no coke, and naphtha, and hydrogen, are introduced
into the lead reactor, pa3sed in series from one reactor to
another, and reacted at reforming conditions, sulfur is
introduced into the final or tail reactor of the series to
provide and maintain sulfur in concentration within the
8~4~
naphtha to said final reactor ranging from about 0.5 part3
per milllon, based on the weight of the naphtha ~eed (wppm),
to about 20 wppm, preferably from about 0.5 wppm to about ô
wppm, while excluding sul~ur from the laad reactor of the
series. The sulfur can be added to an intermediate reactor
Or the series, or added directly to the ~inal reactor, pref-
erably the latter. It ha~ been ~ound that the introduction,
or addition of qulfur to the final reactor, or reaction zone
of the series, during ~tart-up, or that early portion of the
operating cycle when the cataly~t contains lLttle or no
coke, relative to the total operating cycle throughout which
coke gradually, and progressively builds up and accumulates
on the catalyst, provide~ improved hydrogen, aromatics, and
C5~ liquid yields, and reduced C1-C4 light gas make.
In accordance with thi~ invention, sulfur is
excluded, or its presence minimized, during start-up with a
fresh or regenerated catalyst from the lead reactor of the
series wherein naphthene dehydrogenation is the predominant
reaction. A higher level of sul~ur is maintained during
this period in the final reactor of the series wherein
paraffin and olefin dehydrocyclization are the predominant
reactions. In carrying out such operation, sulfur is
injected during this period into an intermediate or the tail
reactor, preferably the latter, as hydrogen sulfide, or
compound decomposable in 3itu to form hydrogen sulfide, ir.
amount suf~icient to provide sulfur in concentration ranging
from about 0.5 wppm to about 20 wppm, prererably 0.5 wppm to
about 8 wppm, based on naphtha feed. Sulfur is then removed
~rom the product hydrogen gas from the last reac~or of the
~erLes, suitably by passage o~ the gas through a sulfur
adsorbent, and the gas is recycled. The gas i~ recycled to
the lead reactor where it is added with fre~h essentially
sulfur-free naphtha. The essentially sulfur-free naphtha
feed, and hydrogen, entering the lead reactor will provide a
concentration of less than 0.5 wppm within the naphtha feed
to the first reactor, and preferably no more than about 0.1
wppm sul~ur. It is found that hydrogen, aromatics and C5~
~2~85~
-- 8
liquid yields are improved, and reduced Cl-C4 light sas ma~e
up to such point in time that the catalyst in the final
reactor, or reaction zone, contains no more than about 10
percent coke, based on the total weight of the catalyst,
deposited thereon; and preferably no more than about 5 per-
cent coke deposited upon the catalyst. This period corre-
sponds generally from about 5 to about 60 percent of the
operating cycle, and preferably from about 5 percent to
about 30 percent o~ the total operating cycle which begins
at start-up, or when the catalyst i3 first placed on-oil.
The catalyst employed in accordance with this
invention is necessarily constituted of composite particles
which contain, besideY a carrier or support material, a
hydrogenation-dehydrogenation component, or components, a
halide component and, pre~erably, the catalyst is sulfid-
ed. The support material i~ constituted of a porous,
refractory inorganic oxide, particularly alumina. The sup-
port can contain, e.g., one or more of alumina, bentonLte,
clay, diatomaceous earth, zeolite, silica, activated carbon,
magnesia, zirconia, thoria, and the like; though the most
preferred support is alumina to which, if desired, can be
added a suitable amount of other refractory carrier mate-
rials such as silica, zirconia, magnesia, titania, etc.,
usually in a range of about 1 to 20 percent, based on the
weight of the support. A preferred support for the practice
of the present invention is one having a surface area of
more than 50 m2/g, pre~erably from about 100 to about 300
m2~g, a bulk density of about 0.3 to 1.0 g/ml, preferably
about 0.4 to 0.8 g/ml, an average pore volume of about 0.2
to 1.1 ml/g, preferably about 0.3 to 0.8 ml/g, and an
average pore diameter of about 30 to 300~.
The metal hydrogenation-dehydrogenation component,
or components, includes platinum, and one or more of
iridium, rhenium, palladium, rhodium, tin, and tungsten.
Preferably, the hydrogenation-dehydrogenation component, or
components, are platinum and iridium or rhenium, or platinum
and both iridium and rhenium. The hydrogenation-dehydroge-
1278544
nation component, or components, can be composited with orotherwise intimately associated with the porous inorganic
oxide support or carrier by variou~ techniqueq known to the
art such as ion-exchange, coprecipitation with the alumina
in the 901 or gel ~orm, and the like. For example, the
catalyst composite can be formed by adding together suitable
reagents such as a salt o~ platinum and a salt of rhenium
and ammonium hydroxide or carbonate, and a 3alt of aluminum
such as aluminum chloride or aluminum sulfate to form
aluminum hydroxide. The aluminum hydroxide containing the
salts of platinum and rhenium can then be heated, dried,
formed into pellets or extruded, and then calcined in
nitrogen or other nonagglomerating atmosphere. The metal
hydrogenation components can also be added to the catalyst
by impregnation, typically via an "incipient wetness" tech-
nique which requires a minimum of solution 90 that the total
solution is absorbed, initially or after some evaporation.
It is preferred to deposit the platinum and
iridium or rhenium metals, or both, and additional metals
used as promoters, if any, on a previously pilled, pelleted,
beaded, extruded, or sieved particulate support material by
the impregnation method. Pursuant to the impregnation
method, porous refractory inorganic oxides in dry or
solvated state are contacted, either alone or admixed, or
otherwise incorporated with a metal or metals-containing
solution, or solution~, and thereby impregnated by either
the "incipient wetness" technique, or a technique embodying
absorption ~rom a dilute or concentrated solution, or 301u-
tions, with subsequent filtration or evaporation to effect
total uptake of the metallic components.
Platinum in absolute amount is usually supported
on the carrier within the range of from about 0.05 to 3
percent, preferably from about 0.2 to 1 percent, based on
the weight of the catalyst (dry basis). Rhenium, in abso-
lute amount, is also usually supported on the carrier in
concentration ranging from about 0.05 to about 3 percent,
preferably from about 0.3 to about 1 percent, based on the
1278~
-- 1 o -
weight of the catalyst (dry basi3). Iridium, or metal other
than platinum and rhenium, when employed, is alao added in
concentration ranglng from about 0.05 to about 3 percent,
preferably ~rom about 0.2 to about 1 percent, based on the
weight of the catalyst (dry basis). The absolute concentra-
tion of each, of course, is preselected to provide the
desired ratio of rhenium:platinum for a respective reactor
of the unit, a~ heretofore expre~sed.
In compositing the metal~ with ehe carrier, e3sen-
tially any soluble compound can be used, but a soluble com-
pound which can be easily ~ubjected to thermal decomposition
and reduction is preferred, for example, inorganic salts
such as halide, nitrate, inorganic complex compounds, or
organic salts such as the complex salt of acetylacetone,
amine salt, and the like. When, e.g., platinum i9 deposited
on the carrier, platinum chloride, platinum nitrate, chloro-
platinic acid, ammonium chloroplatinate, potassium
chloroplatinate, platinum polyamine, platinum
acetylacetonate, and the like, are preferably used.
To enhance catalyst performance in reforming
operations, it is also required to add a halogen component
to the catalysts, fluorine and chlorine being preferred
halogen components. The halogen is contained on the cata-
lyst within the range of 0.1 to 3 percent, preferably within
the range of about 0.3 to about 1.5 percent, based on the
weight of the catalyst. When using chlorine as halogen com-
ponent, it is added to the catalyst within the range of
about 0.2 to 2 percent, preferably within the range of about
0.5 to 1.5 percent, based on the weight of the catalyst.
The introduction of halogen into catalyst can be carried out
by any method at any time. It can be added to the catalyst
during catalyst preparation, ~or example, prior to, follow-
ing or simultaneously with the incorporation of the metal
hydrogenation-dehydrogenation component, or components. It
can also be introduced by contacting a carrier material in a
vapor phase or liquid phase with a halogen compound such as
hydrogen fluoride, hydrogen chloride, carbon tstrachloride,
or the like.
12t78~4
- 1 1 --
The catalyst i9 dried by heating at a temperature
above about 80F, preferably between about 150~F and 300F,
in the presence of nitrogen or o~cygen, or both, in an air
stream or under vacuum. The catalyst is calcined at a tem-
perature between about 500F to 1200F, preIerably about
500F to 1000F, either in the presence of' oxygen in an air
stream or in the presence of an inert gas ~uch a3 nitro&en.
Sulfur i~ a highly preferred component of the
catalysts, the sulfur content of the catalyst senerally
ranging to abou'c 0.2 percent, preferably from about 0.03
percent to about 0.15 percent, based on the weight of the
catalyst (dry basis). A fresh reforming cataly~t is
generally ~ulfided prior to its being placed on-oil, and
3ince sul~ur i~ lost during reforming, the catalyst i~ again
sulf-ided during catalyst regeneration. The sulf'ur can be
added to the catalyst by conventional methods, 3uitably by
breakthrough sulf'iding of a bed of the cataly3t with a
sulrur-containing gaseous stream, e.g., hydrogen sulfide in
hydrogen, performed at temperatures ranging from about 350F
to about 1050F and at pre~sures ranging from about 1 to
about 40 atmo3pheres for the time necessary to achieve
breakthrough, or the desired sulf ur level.
The feed or charge stock can be a virgin naphtha
cracked naphtha, a naphtha f`rom a coal liquefaction proces~,
a Fischer-Tropsch naphtha, or the like. Typical feeds are
those hydrocarbons containing from about 5 to 12 carbon
atom3, or more preferably from about 6 to about 9 carbon
atoms. Naphthas, or petroleum fractions boiling within the
r~nge of from about 80F to about 450F, and preferably from
about 125F to about 375F, contain hydrocarbons of carbon
numbers within these ranges. Typical fractions thus usually
contain from about 15 to about 80 vol. ,~, paraffin3, both
normal and branched, which fall in the range of about C5 to
C~2, from about 10 to 80 vol. % of naphthenes falling within
the range of from about C6 to C12, and from 5 through 20
vol. ~ of the desirable aromatics falling within the range
of from about C6 to C12.
~2'7~S~
- 12 -
The rerorm$ng run~ are inltlated by adJustlng the
hydrogen and feed rate3, and the temperature and pressure to
operatlng condltion~. The run $s contlnued at optimum
re~ormlng condltlons by ad~u-~tment o~ the ma~or process
variables, with$n the ranges de~crlbed below:
MaJor Operating Typical Process Preferred Process
Variables Conditions Conditions
Pressure, psig 50-750 100-500
Reactor Temp., F 800-1200 850-1000
Recycle Gas Rate, SCF/B 1000-10,000 1500-5000
Feed Rate, W/Hr/W 0.5-10 1-5
The invention will be ~ore rully understood by
re~erence to the following comparative data illustratin~ its
more salient reatures. All parts are given in terms of
weight except as otherwi~e specified.
A rhenium promoted platLnum catalyst (0.3 wt. ~
Pt/0.3 wt. % Re) obtained from a catalyst manufacturer was
employed in the several reactors of the reforming units used
in conducting the following runs. The catalyst was reduced
and pre ulrided by contact with an admixture of 300-500 vppm
(ppm by volume) H2S in hydrogen at 700-8500F until H2S
breakthrough at the reactor outlet, a treatment suf~icient
to deposit 0.08-0.10 wt. S sul~ur on the cataly~t initially
charged to the unit.
In~pections on the petroleum virgin naphtha
~eedstock used in the te~t for conducting the run described
under Example 1, and the naphtha feedstock employed for
conductlng the runs described under Examples 2, 3 and 4 are
g$ven in the table i~mediately below.
Figures 1 and 2 of the attached drawings respectively
illustrate the results of the pilot plant run in Example 2 and the
maintenance of the hydrogen yield after increase in the sulfur level
during such run.
~78544
-- 13
Table
.
Example 1 Examples 2,3 & 4
Feedstock Feedstock
API Gravity 56.2 59.7
Average Mol. Wt. 108 108
Nitrogen, wppm <1 <1
ASTM DistLllation
IBPF 180+2 181 +2
207 196
211 204
243 241
297 310
309+5 328
F9P 323+10 340+10
.
The following demonstration run, Example 1, illu-
3trates the detrimental effects of feed-borne ~ulfur on
catalyst in the lead reactor position and net overall re-
forming unit performance.
EXAMPLE 1
A commercial ref'orming unit with three onstream
reactors in serie~, each containing a platinum-rhenium
catalyst (0.3 wt. ~ Pti3.3 wt. % Re) in which less than
eleven percent of the total onstream catalyst charge was
contained in the first of the three reactors, was employed
to process eighteen thousand barrels per day of naphtha of
average molecular weight ~ 108 and specific gravity
(60/600F) ~ 0.754. During the run, after continuous opera-
tion to a point at which the catalyst in the tail reactor
was estimated to have accumulated from 7 to 8 weight percent
coke, the feed sulfur level was increased from an initial
level of less than 0.3 wppm sulfur in feed to 1.0 wppm
sulfur in feed. Although operating at the lowest average
temperature of the three reactors, and containing the
smallest amount of catalyst, performance of the initial
reactor, or reactor in which the incoming hydrocarbon feed-
stock is first contacted with catalyst, was observed to
undergo a decrease in the temperature drop between the
1'~78544
- 14 -
reactor Lnlet and outlet, indicating a change in the nature
or extent Or reactions taking place wlthin the catalyst
bed. Within three days Or the increase in sulfur level,
temperature drop of the lead reactor was observed to
decrease from 97F to 77~F while average temperature in the
catalyst bed increased 10F. Thi~, as would be expected,
resulted in an increased rate of catalyst deactivation,
which i3 highly temperature dependent. Temperature differ-
ences between the inlet and outlet Or the second and third
reactors during this time remained essentially unchanged at
84 + 1F and 16 ~ 1F, respectively. Overall unit perfor-
mance, as indicated by the total temperature drop acro3s all
reactors, deteriorated from 197F temperature drop to 177F
temperature drop. This apparent decrease in unit perfor-
mance was corroborated by an observed drop in reforming unit
hydrogen yield from 1.70 weight percent Or feed before the
sulrur increase to 1.60 weight percent hydrogen yield a~ter-
ward, and an approximate 10 percent decrease in unit activ-
ity, or a drop in reformate octane of 0.8 RONC at the con-
ditions at which the unit was operating.
The following Example 2 substantiates the large
sensitivity of overall unit performance to deteriorations in
the lead reactor catalyst, showing that the harmful effects
of feed-borne sulfur can be prevented if damage to the cata-
lyst in the lead reactor position is avoided.
EXAMPLE 2
A first pilot plant run was conducted in which a
coke-free catalyst of the same composition employed in the
p-receding demonstration was contained in a re~orming unit
employing three onstream reactors in series in the same
proportions as described in the preceding demonstration.
Process conditions of the pilot plant operation were the
same as those of the preceding run, 360 + lO psig operating
pressure and 3200 + 200 SCF/B recycle gas ratq. The
reforming unit in which this evaluation was made permitted
adiabatic operation in each of the individual reactors, or
the use of an adjustable external heat control means such
1;~7~3544
15 -
that a temperature profile could be lmposed. After startup
the adiabatic temperature profiles were permitted to develop
in all reactors whlle the sulrur level Or the incoming
naphtha reedstock was maintained at a constant level of 0.3
wppm. Arter eignt days, at which time the catalyst in all
reactors had accumulated less than 4 weight percent coke, as
determined from previous tests conducted at the same condi-
tions, reedstock sulrur level was increased to 1.5 wppm. To
examine the er~ects o~ thi~ change without the perturba-
tional ef~ects Or increased averaga temperature in the rirst
reactor catalyst bed, the adiabatic temperature drop which
exlsted in the rirst reactor cataly3t bed before the ~ulrur
level increase was maintained after the increase by means cr
an ad~ustable external heat control. This prevented an
increased average bed temperature and accelerated rate Or
deactivation in the lead reactor, while permitting the
changed sulrur level to manirest its er~ects on the catalyst
charge contained in the second and third onstream reactors.
- The results of the run are graphically depicted as
circular points by rererence to Figure 1. As in the preced-
lng demonstration, there was no observed change in the
temperature drop between the inlet and outlet Or the second
and third reactors. Unlike the preceding demonstration,
however, there was a substantial improvement in the purity
Or the hydrogen-containing recycle gas. In less than
twenty-four hours after the sulfur level increase, the
hydrogen content of gas rrom the reactor erfluent product
separator vessel, maintained at constant temperature and
pressure, was observed to increase by 2 volume percent. The
yield of methane, the principal impurity which remains in
the excess separator drum gas after downstream puri~ication
ror use in other hydrocarbon processes in an integrated
refinery, decreased by roughly one-sixth, as also shown in
Figure 1. There was no loss in hydrogen yield arter the
sulfur level increase. This is illustrated by the data
shown in Figure 2. That the additional sulrur added to the
system was present throughout both the second and third
8544
- l6 -
reactors, and not, in view o~ the small total mass involved,
merely adsorbed on a portion of the catalyst bed, was con-
~irmed by careful measurements of the hydrogen sul~ide con-
tent of the separator drum overhead gas.
The following Example 3, when compared with the
preceding examples, 3how3 that the presence of feed-borne
sulfur early in the unit operatlng cycle, when the catalyst
contains no more than about lO weight percent carbon, can in
~act be advantageous when damage to the lead position cata-
lyst is avoided.
EXAMPLE 3
A 3econd pilot plant run identical to that de-
scribed in Example 2 was carried out as in the first pilot
plant run, the only difference between the two runs being
that the level of sulfur in this second, 3imultaneous
control test was maintained at 0.3 wppm throughout. The
results obtained Ln thi~ second pilot plant run are
~uperLmposed on the graphs depicted in Figures 1 and 2 as
dashed lines, and reference is made to these figures. A
comparison of results from these two evaluations reveals
that the incremental sulfur added to the second and third
reactors of Example 2 while preventing damage to catalyst in
the lead reactor is, in fact, advantageous. Figure l
illustrates the higher hydrogen purity and lower methane
yield of the performance obtained in the first pilot plant
run. These benefltq, as shown by reference to Figure 2,
have been realized without any disadvantageous 1099 of
hydrogen yield. The overall unit actirities, or ability to
produce high-octane reformate at a given set of conditions,
of Examples 2 and 3 were equal both at the time o~
incremental sulfur addition in Example 2 and for at least
ten days afterward.
The following Example 4 confirms that these
benefits are enjoyed during the early portions of the
operating cycle between regenerations, up to the time that
the catalyst contains about lO weight percent carbon.
~'~78~i44
- 17 -
EXAMPLE 4
The first and second pilot plant runs (Example3 2
and 3) were continued past the times graphically depicted .n
the figures. The inlet temperatures Or the reactors in each
evaluation were increased, as necessary, to compensate for
catalyst deactivation and maintain constant reformate pro-
duct octane value. The temperature drop across the first
reactor cataly3t beds was kept equal to ensure that any
observed differences between the two run~ would be manifes-
tations of the effects of sulfur addition on the second and
third onstream reactors. It was found that the performance
advantages obtained in the first p$10t plant run did not
extend much beyond the times shown in the illustratLng
figures. After about twenty days on oil, a slight activity
difference between the two pilot plant runs began to be
observed. This difference grew, of course, as temperatures
were increased and the rates of deactivation accelerated.
After about forty days on oil, the hydrogen purities of
separator drum overhead gas were the 3ame. After about
fifty-~ive days on oil, a difference in the weight percent
hydrogen yields of the two runs began to develop. To
provide assurances of data integrity and userul extended
performance results, the runs were continued until catalyst
activity had in each case declined to one-fourth of it3
initial, start-of-run value. The runs begun as described in
Examples 2 and 3 were terminated after about three and
slightly greater than five months, respectively. From the
multitude of measurements made over the course of these
evaluations, it was found that the benefits obtained in the
first pilot plant run are enjoyed during the early portions
of the operating cycle, when the reforming catalyst contains
up to about tO weight percent carbon, or coke.
