[0001] This invention relates to an improved method for operation of a fluid catalytic cracking
system having a plurality of nozzles for injection of charge to a riser type reactor.
More specifically, the present invention contemplate a high degree of preheat of the
oil feed introduced to a riser reactor by multiple injection nozzles.
[0002] Fluid catalytic cracking of petroleum fractions is a well-established refinery operation.
The catalytic cracking apparatus per se comprises a catalyst section which is subdivided
into a reactor section whee catalytic cracking occurs, coupled with a regenerator
section where coke deposited on spent catalyst is burned. The process operates essentially
as follows. Fresh feed, which may be preheated, is mixed with catalyst and undergoes
cracking within the reactor section. Products are removed from the reactor in the
vapor phase and passed to a products recovery section comprising at least one main
fractionator or distillation column for separation of the products into desired fractions.
Spent catalyst, which has been coked by the cracking reaction, is continuously passed
from the reactor to the regenerator by a spent catalyst transfer line. In the regenerator,
the coke is burned by contact with an oxygen containing gas. Flue gas is passed from
the regenerator, and regenerated catalyst is recirculated to the reactor via a standpipe
where it is picked up by the fresh feed hydrocarbon charge stream. The catalyst itself
is finely divided and simulates a fluid in various portions of the catalyst section,
whence the name of the process. In a typical operation, heat generated in the regenerator
is carried by the hot regenerated catalyst to the reactor to supply head for the endothermic
cracking reaction. Typical fluid catalyst cracking systems are disclosed in U.S. Patent
Nos.3,206,393 and 3,261,777.
[0003] The fluid catalytic cracking process has been improved in efficiency over the years.
In particular, the discovery of zeolite catalysts with their greater activity and
reduced coke make, and improvements in design of the reactor section to emphasize
riser-cracking, are cases in point.
[0004] More recently it has been proposed that the charge be injected to a riser reactor
through a plurality of nozzles as described in U.S.Patent Nos.3,152,065 and 3,246,960.Such
multiple injection nozzles are provided with the intent to obtain more uniform and
immediate dispersion of catalyst in the charge and thus afford in large scale equipment
results like those given by small scale pilot plants.
[0005] It has now been found that the effect of multiple injection nozzles in commercial
scale FCC riser reactors is not simply a matter of improved contact between charge
and catalyst nor simply a function of improved uniformity of the dispersion of FCC
catalyst in the charge. The effects from supplying an increased amount of the endothermic
heat of cracking by increased preheat of the charge run counter to phenomena not previously
recognized in the art or may be due to some other factor not now recognizable.
[0006] Plants for practice of FCC riser cracking are typically controlled by setting the
desired temperature for the top of the reactor. A sensor detects temperature at that
point and adjusts recycle of hot catalyst from the regenerator to maintain actual
conditions at the set point. Thus as reactor top temperature tends to rise, as by
increase in feed preheat, the control moves the catalyst recycle valve towards closed
position thus reducing hot catalyst supply to the riser. As will be immediately apparent,
this action reduces the catalyst/oil ratio in the reactor and has the expected result
or reducing conversion of the charge. Past experience has been a consequent reduction
in yield of gasoline based on charge.
[0007] Unexpectedly, a commercial riser FCC using multiple injection nozzles is found to
follow the usual course of reduced conversion at decreased catalyst/oil on increaing
feed preheat over a part of the range described, but at higher gasoline selectivity
than expected. This improved gasoline selectivity is accompanied by reduction in octane
number of the product, conforming to past experience that, at constant charge and
catalyst, increase in gasoline yield is at the expense of lower octane number.
[0008] The unusual effect of increased gasoline yield at decreased conversion provides means
for optimizing gasoline yield and octane number. According to the. present invention,
gasoline yield is enhanced by increasing preheat of the feed to multiple injection
nozzles. Alternatively, octane number may be enhanced by reducing feed preheat in
such systems.
[0009] The present method for obtaining increased gasoline yield at reduced conversion is
described hereinafter with reference to the annexed drawings wherein:
Fig.1 is a schematic representation of a control system for a riser type FCC Unit;
Fig.2 is a diagrammatic sketch in elevation of a riser conversion zone terminating
in an enlarged catalyst separation zone provided with cyclonic separating means, catalyst
stripping and conduit means for passing catalyst to the bottom of the riser and for
withdrawing from the stripping zone;
Fig.3 is a diagrammatic sketch of the bottom portion of the riser conversion zone
with pipe means for contacting reactant fed with catalyst particles;
Fig.4 is a cross-sectional view of the pipe nozzle arrangement shown in elevation
in Fig.2; and
Fig. 5 is a graph depicting the effect of number of nozzles on the average catalyst
concentration in the riser cross-section at a predetermined level.
[0010] A conventional control scheme for a riser reactor FCC Unit is illustrated in Fig.
1. In Fig. 1 the oil feed, which may be preheated by the heater 2 is passed via line
1 to the lower end of riser pipe 3. Heated catalyst from a stand-pipe 4 having a control
valve 5 is combined with the heated oil in the riser 3 such that an oil-catalyst mixture
rises in an ascending dispersed stream until it is discharged into reactor 6 wherein
catalyst is separated as a dense bed 7. In the reactor 6, thee may be other fluidized
contacting between oil and the catalyst particles within the relatively dense fluidized
bed 7. Generally, a major portion of the necessary cracking and contact of the oil
with the catalyst takes place in riser 3. Any catalyst particles remaining suspended
in the vaporous cracked reaction products are separated at the upper end of reactor
6 by centrifugal or setler type means (not shown in Fig.1). Products are then transferred
overhead via line 8 to a products recovery section (not shown) which includes at least
one fractionator. A temperature sensing means 9, such as a thermocouple, is provided
to sense the temperature within reactor 6 and to provide a signal to the temperature
controller 10 indicative of the temperature at the top of the reactor 6. When the
temperature of the reactor deviates from the predetermined temperature defined by
the set point 11 of the temperature controller 10, the temperature controller places
a signal on a line 12 to adjust the valve 5 and thus adjust the temperature of the
catalyst oil mixture in riser 3 in a direction to reduce the deviation of the measured
temperature from the predetermined temperature as defined by the set point 11 of the
controller 10. Generally, it is highly desirable to minimize fluctuations of the temperature
within reactor 6 and of the products issuing from line 8 so as to minimize disturbances
in the main fractionator column. Spent catalyst from the bed 7 is continuously passed
from reactor 6 by spent catalyst transfer line 13 equipped with a valve 14, and is
passed to the lower portion of the regenerator 15 to form a relatively dense bed 16.
Air is passed via air transfer line 17 equipped with a control valve 18 to the regenerator
15, where it passes through distributor grid 19 and serves to maintain bed 16 in a
fluidized state and to pass catalyst continuously through the riser 20 as a dilute
phase 21. In the upper portion of the regenerator 15 flue gas is separated from regenerated
catalyst, and exists via line 22. Regenerated catalyst is separated to form a relatively
dense bed 23. The regeneratd catalyst bed 23 is at a substantially higher temperature
than the spent catalyst from bed 7 by virtue of the coke burning which occurs in the
regenerator 15. Within the regenerator itself, bed 23 and the flue gas in the chamber
are substantially hotter than the spent but partially regenerated catalyst of bed
16. A composition sensor 24, which indicates the carbon monoxide and oxygen content
of the flue gas, generates a signal indicative of that composition. Valve 18 is commonly
controlled by operator intervention to control the flow of air and thus the CO and
oxygen content of the flue gas. Alternatively, the signal generated by composition
sensor 24 is transmitted to the composition controller 25. Controller 25, equipped
with set points 26, places a signal on line 27, which signal is indicative of the
deviation of the carbon monoxide composition of the flue- gas from the set point 26,
to adjust the control valve 18 in a direction to reduce the deviation of the measured
composition from the predetermined composition as defined by the set point 26. In
general, the set point is adjusted to a CO content less than 2000 ppm and the flue
gas, in general, will contain about 2% excess oxygen gas. Although not explicitely
shown, valve 14 is usually coupled with valve 5 in such a manner as to maintain a
fixed ratio of amount of catalyst in the regenerator 15 to amount of catalyst in the
reactor 6.
[0011] The present invention relates to the catalytic conversion of hydrocarbons with finely
divided particles of catalyst. More particularly, the present invention is concerned
with the conversion of hydrocarbons in a riser conversion zone under conditions promoting
a more uniform mixing between hydrocarbon reactant and finely divided catalyst particles.
In a more particular aspect the present invention is particularly concerned with the
manner relied upon for injecting hydrocarbon feed into a stream of catalyst particles
to form a suspension in a riser conversion zone.
[0012] The catalytic cracking of hydrocarbons, such as a gas oil hydrocarbon feed material
is practised at a temperature within the range of 900
oF. up to about 1200
0F. with the temperature more usually restricted to less than 1100
oF. Operating pressures within the range of atmospheric up to 100 psig may be employed,
it being preferred to select conditions assuring good catalyst circulation, product
and reactant flow which will contribute to the overall economics of the operation.
In a riser conversion operation the hydrocarbon reactant is mixed with hot catalyst
particles in the bottom of lower portion of the riser to form a suspension at a desired
cracking temperature and the suspension is passed through the riser conversion zone
under space velocity conditions providing a hydrocarbon residence time in the riser
within the range of 1 to 15 seconds and more usually less than about 10 seconds. The
suspension is discharged from the riser outlet into cyclonic separation zones or the
suspension may be discharged from the riser into an enlarged zone wherein separation
of catalyst particles from gasiform material such as hydrocarbon vapors is accomplished
by a reduction in velocity thereby causing the catalyst particles to settle out. Cyclonic
separating means may also be relied upon to remove particles of catalyst from vaporous
material not removed by the reduced velocity. The hydrocarbon vapors are recovered
and separated in a product fractionator. The catalyst particles are stripped to remove
entrained hydrocarbons and the stripped catalyst may be passed to catalyst regeneration.
[0013] In the operating concept of this invention it is proposed to inject the hydrocarbon
reactant feed to be cracked through a plurality of upwardly extending pipe nozzles
in the lower portion of the riser conversion zone. The upper ends of the pipe nozzles
are preferably at least above the point of intersection of the upper surface portion
of the catalyst standpipe with that of the riser conversion zone. In one arrangement
the number and position of the pipe nozzles with respect to one another is such that
each nozzle is related to substantially an equal cross-sectional area of the riser
conversion zone. In such an arrangement, there are at least five nozzles distributed
in the cross-sectional area of the riser and preferably there are t least 7 nozzles
distributed in the riser cross-section. In yet another embodiment, it is desirable
to locate a majority of the pipe nozzles in a ring about a concentrically positioned
pipe nozzle within the riser cross-sectional area with the area of the ring being
concentrically located and encompassing at least 65% of the riser cross section. Thus
in a particularly preferred embodiment there are at least seven pipe nozzles terminating
from a single oil feed inlet conduit, at least 6 of the pipe nozzles form a ring spaced
inwardly from the riser wall with the seventh concentrically located there within
and the ring comprising the six nozzles will occupy an area which is at least 70%
of the riser cross-sectional area.
[0014] Referring now to Fig. 2, there is shown diagramatically a riser conversion zone 32,
conduit means 34 for introducing catalyst to the bottom lower portion of riser 32,
a separation zone 36 provided with a catalyst stripping zone 38 in the lower portion
theeof and spent catalyst withdrawal conduit 40. Hydrocarbon feed introduced to the
bottom of the riser 32 by conduit 42 is caused to pass upwardly through a plurality
of distributing pipes 44 open at their upper end to form nozzles for dispersing hydrocarbon
feed in contact with catalyst charged to the bottom of the riser by conduit 34 at
an elevated cracking temperature of at least 900
oF. A suspension is formed with the dispersed and vaporized oil and catalyst which
is then moved under selected velocity conditions upwardly through the riser. In the
arrangement of Fig. 2, the suspension passing upwardly through the riser is discharged
through slotted openings 46 in the upper extreme periphery of the riser 32 and into
the enlarged separation zone above a dense fluid bed of catalyst 48. Vaporous products
of conversion and stripping stream pass through cyclonic separators 50 provided with
catalyst diplegs 52. In separator 50, entrained catalyst particles are separated from
vaporous material for return by dipleg 52 to bed 48. Separated vaporous material pass
through conduits 54 to chamber 56 for withdrawal by conduit 58. Stripping gas such
as steam is introduced to the bottom of bed 38 by conduit 60 and stripped catalyst
is removed therefrom by conduit 40 for transfer to a catalyst regeneration zone shown
in Fig.1.
[0015] Fig.3 shows in greater detail the bottom portion of riser 32 and the multiple pipe
inlet for injecting hydrocarbon feed into regenerated catalyst to form a suspension
therewith for passage upwardly through riser 32. Fig.4, on the other hand, shows a
cross-sectional view 4-4 of the pipe arrangement of Fig. 3 with open end nozzles for
injecting hydrocarbon feed into the catalyst. The circle of pipes 44 preferably identify
a cross-sectional area of the riser which is at least 70% of the riser cross-sectional
area. Such an arrangement has been found to provide a more uniform catalyst density
profile across the riser cross section than 5 or less feed injection pipes as depicted
by the curve of Fig. 5. The graph of Fig. 5 shows the improved catalyst density profile
obtained by use of a multiple pipe nozzle arrangement such as shown in Figs. 3 and
4. Level 1 referred to in the graph relates to measurements taken about 4 feet above
the outlet of the pipe nozzles. It is clear from the graph that the seven pipe feed
inlet device is better than a five and each of these are much better than a 1 or 3
pipe nozzle inlet arrangement.
[0016] The effect of the multi-nozzle injection device and response of a unit so equipped
have been confirmed by surveys conducted on a commercial FCC Unit having a seven nozzle
injection conforming to that shown in Figs. 2 - 4. In commercial practice it becomes
highly impractical to operate at different times with precisely the same charge stock
and catalyst, because the character of the catalyst changes with time on stream and
differences in batches of catalyst added for make-up and because the crude available
to a refinery changes at very short intervals. It thus becomes necessary to adjust
raw survey data to compensate for differences in charge stock and catalyst. The data
here reported have been so adjusted by application of factors equal to differences
predicted by a mathematical mode based on pilot plant and commercial data. The conclusions
stated are predicted on that adjusted data.
[0017] The surveys have demonstrated that the seven nozzle mixer increased gasoline yield
by 2.1 volume percent and decreased Research Octane Number without lead by 1.1 at
equal conversion. Pertinent changes in operation attributed to the multi-nozzle device,
adjusted for feed and catalyst differences, are as follows:

[0018] Three surveys of the commercial unit equipped with seven nozzle injector have been
conducted to evaluate the effect of feed preheat.
[0019] A comparison of the survey results, adjusted to the same feedstock, catalyst, and
operating conditions except for feed preheat temperature, is shown below:

[0020] Although the differences in gasoline selectivity shown in the above table for the
three surveys are primarily due to the effect of feed preheat temperature, some of
these differences can also be attributed to variations among the surveys in conversion
level, and the level of carbon on regenerated catalyst which also affect gasoline
selectivity. The effect of conversion can be accounted for by plotting the gasoline
yields for the three surveys versus conversion and then drawing a line through each
point representing the typical gasoline yield versus conversion slope for the Unit.
Even after adjusting for conversion level, higher gasoline selectivity is obtained
at the higher feed preheat temperature.
[0021] However, a better method of determining the inherent effect of feed preheat temperature
on gasoline selectivity , which accounts for the level of carbon on regenerated catalyst,
as well as conversion level, is simply to compare the gasoline yield tuning factors
used in a mathematical model simulation of the actual surveys. For example, in simulating
each survey with a model, predicted gasoline yields had to be "tuned" donward to match
the actual survey yields. The resulting adjustment ("gasoline delta") represents the
deviation of the actual selectivity observed in the survey from the inherent selectivity
predicted by the model for that particular operation. (Generally, the untuned model
selectivity is greater than that observed in commercial units because the reaction
kinetics in the model are based on data from highly efficient pilot units.) A comparison
of the "gasoline delta" adjustments for the three surveys shows that observed selectivity
deviates most from inherent model selectivity in the low preheat temperature case
and least at the highest feed preheat. The increase in selectivity is greater from
260
oF. to 400
0F. and appears to level out between 400°F. and 500°F. This indicates that the commercial
unit's gasoline selectivity not only increases as feed preheat temperature increases,
but approaches the inherent selectivity predicted by the model at high feed preheat
temperatures. In other words, at low preheat temperatures the commercial unit operation
is non-optimum but that at higher feed temperatures it begins to resemble the more
efficient pilot plant operation.
[0022] Operating conditions and yields for the three surveys are reported in Table 1.

[0023] Table 1 shows the actual reactor effluent yields for the three surveys. Before the
surveys can be compared to determine the effect of feed preheat, yields had to be
adjusted to the same feedstock, catalyst, and operating conditions, (except for preheat
temperature). Election was made to adjust the two extreme preheat cases to the middle
preheat case. Once the three surveys were adjusted to the same basis, any remaining
differences in yields between them could be attributed to the changes in feed preheat
temperature. These final adjusted survey yields are shown in Table 2.

[0024] A comparison of the adjusted survey results shows that when feed preheat on the Unit
is reduced from 514
oF. to 374
oF. conversion increase gradually due to an increase in catalyst-to-oil ratio required
to heat balance the unit. However, contrary to expectation, when feed preheat is further
reduced to 265
0F. conversion drops sharply instead of continuing to increase, despite a further increase
in catalyst-to-oil ratio. Gasoline yield, on the other hand, increases only slightly
when feed preheat temperaure is lowered from 514
oF. to 374°F. , then drops even more sharply then conversion between 374
0F. and 265
0F. For the same temperature change, conversion dropped 1.5% which is much less than
the 3.7% gasoline loss, showing a large loss in selectivity.
[0025] Maximum gasoline selectivity is obtained at the highest feed preheat temperature;
it decreases gradually between 514
oF. and 374°F. and then decreases sharply between 374
0F. and 265
0F.
[0026] Although, at this time no offer can be made of a fundamental explanation of why gasoline
selectivity increases with increasing feed preheat temperaure, it is felt this is
related to changes in the heat and mass transfer and kinetic processes, which occur
at the bottom of the riser, as feed preheat temperature is increased. Temperature
traverses of the riser, obtained at the Unit during the surveys at the three different
preheat levels, provided some evidence that such changes do indeed take place as feed
preheat is increased. For example, whereas the calculated mix temperature of the regenerated
catalyst and oil feed are about the same for the three surveys, the actual mix temperature
observed in the temperature traverses increases as feed preheat temperaure is reduced.
The deviation between the actual and calculated mix temperatures increases from 0°F,
at the high feed preheat case (514°F.) to 85
0F. at the low feed preheat case.
[0027] At the low feed preheat temperature the deviation between actual and calculated mix
temperature was actually greater than the deviation observed at a different commercial
FCC unit earlier (53°F.), which unit had a single nozzle. This implies that as feed
preheat temperature is reduced the multi-nozzle feed injection system is behaving
like a single nozzle system--at least with respect to whatever phenomenon is causing
this mix temperature difference
[0028] Preheat temperatures according to the invention will generally be above 475
0F, preferably about 500
oF. or higher, up to about 750-800°F.