Note: Descriptions are shown in the official language in which they were submitted.
1~27S83
BACKGROUND OF THE INVENTION ~
This invention relates to a new process of
catalytically reforming naphtha, and a new catalytic composite
to be employed therein.
Catalytic reforming of naphtha is widely employed
in the petroleum refining industry to manufacture motor fuel
blending stock having a greatly increased octane number, relative
to the naphtha feedstock, and to manufacturing aromatics,
especially benzene, toluene and xylenes. The improvement in
octane number is especially important when metal-containing
additives, such as tetraethyl lead (TEL), are not employed for
environmental reasons. The research octane number of a blending '
stock or motor fuel which is free of such additives is determined i
by ASTM test No. D-2699, and is often referred to as the "clear" I
or F-l octane number. Improvements in the manufacture of no-lead !
motor fuel are of particular environmental-importance.
Most reforming catalysts employ platinum as a catalytic
agent. Some aiso employ rhenium together with platinum. Both
are expensive, although platinum is about 7-10 times as expensive i
as rhenium, and hence reforming catalyst is one of the most
.'
.
. . .
11 11275~33
l !expensive catalysts employed in petroleum refining. Any
2 improvements which reduce catalytic cost per barrel of naphtha
3 reformed, such as by increasing the catalyst life (i.e.,
4 'increasing the number of barrels of naphtha which can be
reformed per pound of catalyst before the catalyst becomes
6' so deactivated as to require replacement) is advantageous.
7 ;This is also true in respect of increasing the cycle length
8 of the catalyst, i.e., increasing the number of barrels of
9 naphtha which can be reformed per pound of catalyst before
the catalyst requires regeneration.
11 In addition, Russia and South Africa are the source
12 of more than 95% of the free world's supply of platinum. Any
13 improvements which decrease the amount of platinum required
14 by industry, again, for example, by increasing the life of
platinum-containing reforming catalysts, while still accomplish-
16 ing the industrial purpose is a contribution to national
17 security and also to economic independence.
1 18 ; The original precious commercial catalysts employed
19 ' a platinum-group metal, preferably platinum itself, as the
catalytic agent; see, for example, Haensel's patents Nos.
21 l 2,479,109-110, granted in 1949 and assigned to the Universal
22 Oil Products Company. About 1968, the use of rhenium together
23 1l with platinum was introduced. A number of references refer to
24 Irhenium-platinum catalysts. Kluksdhal's patent No. 3,415,737,
25 ¦assigned to Chevron Research Corporation, teaches that
26 "It is preferred that the rhenium to platinum atom ratio
be from about 0.2 to about 2Ø More particularly, it is
27 preferred that the atom ratio of rhenium to platinum does
not exceed one. Higher ratios (i.e. greater than one) of
28 rhenium to platinum can be used but generally no further
29 ¦ significant improvement is obtained." (Col. 5, lines 51-56)
Within that criterion, the amount of platinum and rhenium
employed may be varied, according to the patent, respectively
-2-
. _. " . ,
Il 1127583
I, .
il :
l ,within the range of 0.01 to 3, and 0.01 to 5, per cent of the
2 catalytic composite. (Col. 5, lines 35-48) The reasons for
31,employing a low ratio of rhenium to platinum are stated in
4 ,column 4 of the patent. The patent also teaches that the
5 lnaphtha feed should be essentially free of sulfur, more
6 1I preferably less than 5 ppm, and still more preferably less
7 ; than 1 ppm. (Col. 7, lines 67-69)
8 (As employed herein, all compositions expressed in
9 percentages or parts per million are by weight unless otherwise
noted. Because the atomic weights of rhenium and platinum
11 differ only slightly, an atomic ratio of 1 is the same as a
12 rhenium to platinum weight ratio of 0.955.)
13 Kluksdhal's patent No. 3,558,477 teaches that:
14 "It is essential for purposes of the present invention
that the atomic ratio of the rhenium to platinum be not
greater than 1Ø That is, the rhenium to platinum atom
ratio should be 1.0 or less than 1Ø More preferably,
16 the atom ratio of rhenium to platinum should be less than
about 0.7. Inasmuch as rhenium and platinum have almost
17 the same atomic weight, the atomic ratio is essentially
the same as the weight ratio." (Col. 3, lines 26-33)
18
The patent also discloses the same amounts of rhenium and
platinum in the catalytic composite, and of sulfur in the
20'
linaphtha, as does patent No. 3,415,737; see col. 1, line 62,
!! and col. 7, lines 48-51.
22
23 The ~eachings in respect of the rhenium to platinum
24 ratio and the total amount of rhenium as stated in the above- ¦
quoted Kluksdahl's patent No. 3,415,737 are either repeated or
26 expressly incorporated by reference in Jacobson and Spurlock's
l patent No. 3,449,237, at col. 3, lines 1-24; in Jacobson and
27 Vanselow's patent No. 3,558,479, at col. 5, lines 50-69; and
28 ¦in Jacobson's pate~t No. 3,578,582, at col. 1, line 45. Patent
29 ¦NO. 3,578,582 also teaches that rhenium-platinum reforming
3
-3-
:,
~1 llZ~583
I I .
1 catalysts may be presulfided, by treating fresh catalyst, before
2 1l use in reforming, with hydrogen sulfide or an alkyl mercaptan in
3 lan amount sufficient to add 0.05 to 2 mols, preferably 0.1 to 1
4 mol, of sulfur per mol of rhenium and platinum; see col. 2,
5 ~line 58 to col. 3, line 12.
6 I Buss' patent No. 3,578,583 teaches the inclusion of a
7 'minor amount, up to 0.1 per cent, or iridium in a catalyst
8 having up to 0.3 per cent e~ach of rhenium and platinum.
9 An article entitled "New Developments In Reforming"
by Messrs. Haensel, Pollitzer & Hayes (of Universal Oil Products
11 Company), Proceedings of the Eighth World Petroleum Congress,
12 Vol. 4, pages 255-261 (1971) teaches that the yield of Cs+
13 liquid product reformate reaches a maximum when the rhenium
14 constitutes 50 per cent of the total catalytic metal (i.e.,
a rhenium to platinum weight ratio of 1), and that thereafter
16 the yield of liquid product reformate declines as the relative
17 weight of rhenium to platinum is either increased or decreased.
18 It states that "The relationship shown [in Fig. 5 of the article]
19 holds true over a fairly wide range of platinum content,
indicating that the modifying effect of rhénium is indeed
21 exerted on the platinum." Id. at pp. 259-60.
22 I Thus the art of reforming with rhenium-platinum
23 Icatalysts has insistently taught that the rhenium to platinum
24 ratio should be less than 2, and preferably about 1 or less.
THE INVENTION
26
It has now been surprisingly discovered that the
27
28 Icycle length of a rhenium-platinum reforming catalyst is
29 substantially increasèd when the rhenium-platinum ratio is in
3o the range of from not less than 2 to about 5, preferably in the
range of from about 2-1/4 to about 4, and more preferably in
the range of from about 2-1/2 to 3-1~2, when employing a naphtha
-4-
.
1127583
., ,
1 feedstock having less than about 0.5 ppm, and preferably not
2 more than about 0.25 ppm, of sulfur. This improvement is not
3 only outside the teachings of the prior art, but is a unique
4 l'optinum since at very high rhenium/platinum ratios, approximately
5 'labove 5, the cycle length also decreases. The new catalyst is a
6 '`catalytic composite comprising rhenium and platinum on a support,
7 wherein the weight ratio of rhenium to platinum is in the range
8 as stated above. The new process for reforming a naphtha
g fraction com?rises subjecting naphtha having a sulfur content
of less than about 0.5 ppm by weight of sulfur, to contact under
11 reforming conditions and in the presence of gaseous hydrogen with
12 a catalytic composite comprising a support, rhenium and platinum,
13 the weight ratio of rhenium to platinum being in the range of
14 from not less than 2 to about 5, whereby the relative cycle
length of the catalytic composite is greater than if the rhenium
16 to platinum ratio were to be outside the range stated. The
17 preferred weight ratio of rhenium to platinum is in the range
18 of from about 2-1/4 to about 3-1/2.
19 Except for the increased rhenium-platinum weight ratio,
the new catalytic composite may be made in the same manner as
21 has been known heretofore for making catalysts having a lower
22 ~Irhenium-platinum ratio. The amount of platinum may be in the
23 ¦'range of from about 0.1 to about 2 per cent, preferably in the
24 ¦range of about 0.1 to 0.4 per cent, with the rhenium content
25 Ibeing adjusted to furnish the desired rhenium-platinum ratio
26 within the stated range. The support is typically eta or gamma
27 ¦,alumina, and may optionally contain silica, magnesia, oxides
28 !f rare earth metals, and synthetic zeolites, which are sometimes
29 ~Ireferred to as molecular sieves. Typically, up to about 1 per
3o l¦cent of the catalytic composite comprises halides, especially
5-.
112758~
i '
1 chloride or fluoride.
2 The catalyst is preferably presulfided prior to use
3 in reforming in order to avoid excessive hydrocracking when the
4 l,catalyst is initially on stream. Tne presulfiding treatment
5 1i is conducted by contacting the catalyst with a gaseous stream
6 j containing hydrogen sulfide, an alkyl mercaptan or carbon
7 disulfide, preferably admixed with gaseous hydrogen, until the
8 catalyst contains in the range of about 0.1 to about 0.5 parts
9 by weight of sulfur per part of rhenium, i.e., in the range of
about 0.6 to about 3 mols of sulfur per mol of rhenium.
11 Preferably not more than about 0.25 parts of sulfur per part
12 of~rhenium is applied to the catalyst in the presulfiding step,
13 e.g., about 0.17 parts of sulfur per part of rhenium.
14 The process of reforming with the new catalyst is
essentially the same as with the prior art catalyst having a
16 lower rhenium-platinum ratio, except that the naphtha feed
17 should have a sulfur contént of less than about 0.5 ppm, and
18 preferably not more than about 0.25 ppm, in order to achieve
19 the superior catalyst life and about an equivalent yield of
C5+ liquid reformate obtainable with a catalyst having a
21 ' rhenium-platinum ratio in the range of about 2-1/4 to about 5.
22 Normally the naphtha feedstock will be hydrotreated or otherwise
23 desulfurized by processes known in the art. Although the prior
24 i art teaches the use of naphthas having less than 1 ppm of
25 Isulfur, the catalysts of this invention require naphthas having
26 Ithe very low sulfur levels identified above.
27 Halides may be added to the reaction zone during
28 i reforming, such as by injecting hydrogen chloride, carbon- ¦
29 jtetrachloride or an alkyl halide into the naphtha feed and/or
3o linto the recycle hydrogen gas stream entering the reaction zone
~ -6-
I . 1.
, ,.,, . I
- 11275~33
of the reformer. The amount of water in the reaction zone
should be adjusted to maintain a molar ratio of water to
chloride in the range of about 20 to 80, preferably about
40:1.
After the activity of the catalytic composite
has declined by reason of an accumulation of carbonaceous
deposits, generally referred to as "coke", the catalyst may
be regenerated by procedures known in the art. However,
it is a characteristic of the catalyst having a rhenium-
platinum ratio within the range described herein that a
greater amount of coke may be accumulated thereon before
the activity declines to an unsatisfactory level, as
indicated by a reduction in the yield of C5~ reformate
and/or the clear octane number of the reformate, than is the
case for catalysts having a rhenium-platinum ratio of less
than 2. Hence, the regeneration procedure will require
somewhat more time, and precautions, known in the art, should
be taken to avoid excessive flame front temperatures during
the regeneration operation, to the end of avoiding damage
to the catalyst be reason of localized overheating.
Therefore, in accordance with the present invention
; there is provided a process for reforming a naphtha fraction
which process comprises subjecting a naphtha having a sulfur
content of less than about 0.5 ppm by weight of sulfur, to
contact at reforming conditions and in the presence of
gaseous hydrogen with a catalytic composite comprising a
support, rhenium and platinum,-the weight ratio of rhenium
to platinum being in the range of from not less than 2 to
about 5, and the amount of platinum is in the range of about
0.1 to about 2 per cent, whereby the relative cycle length
of the catalytic composite is greater than if the rhenium to
platinum ratio is outside the range.
llZ75~3
The remarkable improvement in catalyst cycle
length has been established by laboratory pilot plant data
described in the following examples. The examples are given
to illustrate the preparation of the catalysts to be
utilized in the process of this invention and their use in
the catalytic reforming of naphtha. However, the examples
are not presented for purposes of limiting the scope of
this invention but in order to further illustrate its
various embodiments.
- 7a -
.
'1, llZ7583
"
1, .
1~'
EXAMPLE I
2 1 A series of reforming catalysts were prepared having
3 ¦~varying weight ratios of rhenium to platinum. The catalyst
5 ¦preparation procedure employed is known in the prior art and is
only briefly described hereinafter on a generalized basis. The
6 preparation procedure comprised adding to a beaker the
7 stoichiometrically desired amounts of ammonium rhenate (NH4ReO4)
8 and diammonium chloroplatinate ((NH4)2PtC16), and de-ionized
9 water. Aqueous ammonium hydroxide was added and the reaction
mixture heated with hand stirring to a temperature in the range
11 of 81-83 C and until all of the ingredients were dissolved in
12 solution and a light straw-yellow color appeared. A 10% aqueous
13 HCl solution and more de-ionized water were added. The solution
14 was then heated to 90-94 C. The mixture was poured over
the desired amount of 1~16th inch gamma alumina extrudates
16 disposed in a rapidly rotating dish. After about one minute
17 or less, the alumina extrudates were removed from the dish and
18 covered with a watch glass. The extrudates were maintained for
19 one hour by means of an infrared lamp at a temperature of at
20 , least 40 C and up to a maximum of about 65 C, with occasional
21 ,'stirring by hand. The catalyst was then dried with air at
2? ' 105-115 C. Thereafter the catalyst was calcined in dry air
23 Iflowing at a rate of about 1,000 V/V/hour for two hours at
24 ~210 F and then for two hours at 900 F.
25 I The analyses of Catalysts A-H is shown in Table I.
26 I I,
27 I ;
28 I ,
29 I '
3 i
-8-
___ _ ... . . I ... ..
~: 11275~3
, ,
1' .
TABLE I
2 Composition, wt. % Wt. Ratio,
3 ! Catalyst Pt Re Cl Re/Pt
'j A 0.334 0.362 0.91 1.08
4 ~¦ B 0.3g4 0.510 0.80 1.48
5 , C 0.340 0.849 0.93 2.50
6~
D 0.248 0.516 0.99 2.08
;~ E 0.236 0.860 0.97 3.64
`~ 8 -
F 0.150 0.366 0.98 2.44
G 0.155 0.875 0.98 5.65
11 H 0.05 0.85 1.0 17
The foregoing catalysts were employed in reforming
pilot plant runs under accelerated aging conditions to establish
', the relative cycle length per unit weight of platinum as a
function of the rhenium/platinum weight ratio. A naphtha from
a Mid-Continent crude oil was employed, the naphtha having been
hydrotreated to a sulfur content of 0.21 ppm by weight. The
7 properties of the naphtha are tabulated:
18 Gravity, API 54.8
9 Distillation
IBP, F 230
10% 246
21 50% 267
;! 90% 315
22 95% 325
, CP 370
23 ~ Sulfur, ppm 0.21
24 ¦ Nitrogen, ppm 0.3
Type Analysis (By Mass Spectrometer)
Paraffins 45.4%
26 Naphthenes 42.6%
Aromatics 12.0%
27 ¦ The pilot plant comprised a single tubular reactor of
28
¦nominal one-inch i.d. stainless steel operated isothermally.
29 ¦It was loaded with about 40 grams of catalyst for the reforming
3
! _g_
.
.~ .
Il 1127583
. . .
,
1 j runs. After a freshly-prepared catalyst was loaded into the
2 ¦ reactor, the catalyst was chemically reduced by passing hydrogen
3 lat 900 F through the catalyst bed. The temperature was then
4 llowered to 800 F and the catalyst was presulfided by passing
5 la mixture of 0.7 vol.% hydrogen sulfide in hydrogen through
6 !the catalyst bed, to a constant sulfur level for each sample
7 of about 0.05% by weight on catalyst. The pre-heated naphtha
8 feedstock was introduced initially at a reactor temperature of
9 800 F, after which the temperature was increased to 925 F and
held constant for the duration of the 300-hour aging runs. The
11 reforming run conditions were a temperature of 925 F, a weight
12 hourly space velocity of 4, and hydrogen/hydrocarbon molar
13 ratio of 3, and a pressure of 200 psig. These conditions
14 produced, at the start of the runs, a C5+ reformate of about
100 research octane number without the addition of tetraethyl
16 lead (referred to as RONC), determined by ASTM Procedure No.
17 D-2699. During the reforming runs a mixture of methanol and an
?8 akylchloride was injected into the naphtha feedstock to maintain
19 a constant (i.e., about 1 weight per cent) chloride content on
each catalyst.
21 ~ The data from the reforming runs were processed as
22 , follows. During the coursè of the 300-hour reforming runs,
23 the Cs+ liquid product was collected periodically and tested for
24 its octane number. The measured research octane number was
plotted against time on stream, and the slope of the octane-
26 time curve was adjusted, by means known in the prior art, to
27 icompensate for any difference between the octane number of the
28 ¦! reformate at the start of the run and the intended initial
29 ¦octane number of 100. The compensated research octane number
3o of the C5+ reformate was then plotted against time in hours
I -10-
I
!1 ~ "~"
i!
i .
Ielapsed since the start of the reforming run, often referred
1 to as "time on feed." The slope of the octane-time curve
2 is negative and is the octane decline rate, i.e., the time
3 rate of octane decline. It has the dimensional units of
4 ¦'RONC/hour and for convenience is often referred to as oRONC/100
5 lhours. In catalytic reforming, it is desirable to minimize
6 'the octane decline rate, and therefore a lesser absolute
7 value of the slope of the curve indicates a more desirable
8 catalyst.
g Table II lists the aging rate, in terms of decline
of research octane number per 100 hours, for the catalysts
11 tested. Catalyst A, which is representative of the preferred
12 rhenium/platinum catalyst of the prior art, was arbitrarily
13 assigned a relative cycle length of 1.00 and the relative
14 cycle length of all of the remaining catalysts were compared
1~ to it by dividing the aging rate of Catalyst A by the aging
16 rate of the catalyst in question. The relative cycle length
17 per unit weight of platinu~ was determined by dividing the
18 relative cycle length of a catalyst by the weight fraction
19 of platinum (as stated in Table I) of the catalyst. The latter
computation is significant because it is indicative of the
21 efficiency with which the expensive platinum is being utilized.
22 ~The data of Table II show that employing rheniumtplatinum ratios
23 li of not less than 2 enables a refiner to obtain a longer cycle
24 1l length and/or better efficiency in utilizing platinum, than
25 l¦with catalysts having a lower Re/Pt ratio. The data also
26 ¦¦establlsh that, surprisingly, this invention permits the use
27 il of catalyst containing a lower weight fraction of platinum
28 ~I(about 0.15%, in Catalysts F and G) than has been employed
29 ¦,commercially heretofore.
30 11
1! --11-- .
Il . ,
_ ~ .. . . .. I I .... . .
1i1 ' .
1 TABLE II
, Relative Cycle
Relative, Length Per Unit
Wt. Ratio, Aging Rate, Cycle Weight of
, 4 I,Catalyst Re/Pt ~ONC/100 Hrs. Length Platinum
5 1 A 1.08 2.8 1.00 2.99
6 ! B 1.48 2.1 1.33 3.87
7 C 2.50 1.7 1.65 4.85
8 D 2.08 - 2.2 1.27 5.12
g E 3.64 1.9 1.47 6.23
F 2.44 2.5 1.12 7.47
11 G 5.65 3.4 0.82 5.47
12 H 17 33 0.08 1.60
13 EXAMPLE II
14
Catalyst C was compared with Catalyst A when employing
16 naphtha feedstocks of d1fferent sulfur content. The test
17 conditions were the same as those referred to in Example I,
18 except that the sulfur content of the feedstock naphtha was
19 adjusted by adding thiophene to increase the sulfur content to
1 20 10 ppm by weight. The results of the aging,test employing
21 ,',naphtha feedstocks having sulfur contents of 0.2 and 10 ppm
~ lare stated in Table III. As was illustrated in Example I,
22 ;
23 ICatalyst C of this invention has a slower aging rate than
24 Catalyst A of the prior art. The relative cycle length, and
also the relative cycle length per unit weight of platinum, of
26 Catalyst C of this invetion is greater than that of Catalyst A
27 ¦of the prior art when the sulfur content of the feedstock is
28 10.2 ppm, but conversely if the sulfur content is 10 ppm. These
29 ¦data establish that for optimum utilization of the invent1on,
-12-
. . , :
11275~33
;.
,
1 the sulfur content of the naphtha feedstock should be reduced
2 to less than 0.5, and preferably to less than 0.25, ppm.
i~
~ TABLE III
4 .! __
lli Relative Cycle
5 1 Aging Rate, Relative Length Per
Feedstock aRONC Cycle Unit Weight of
6 ! Catalyst Sulfur, ppm per 100 hrs. Length Platinum
7 A 0.21 2.8 1.00 2.99
8 A 10 4.0 0.70 2.10
g C 0.21 1.7 1.65 4.85
C 10 6.2 0.45 1.32
11
EXAMPLE III
12
Catalysts A, C and G were employed in reforming pilot
13
plant runs which further illustrate the process of this invention.
14
The hydrotreated naphtha from a Mid-Continent crude oil had the
properties tabulated below:
16
Gravity, AP~ 54.7
17
Distillation - -
18 IBP, F 217
10% 242
19 50% - 268
9o% 322
95% 344
EP 366
21
Sulfur 0.57 ppm
22 Nitrogen 0.77 ppm
I
23 1I Type Analysis (by Mass Spectrometer)
I Paraffins 45.5
24 j! Naphthenes 42.8
Il Aromatics 11.7
25 i
. I The reforming pilot plant, and the processes of
26
¦!reducing and pre-sulfiding the fresh catalyst, were the same
27 1
llas described in Example I, except that the amount of sulfur
28 jj
ladded to each catalyst was varied to furnish about 0.17 parts
29 ¦of sulfur per part by weight of rhenium, according to the
il -13-
¦~ !
11 .
1127583
!
1 following table:
2 Sulfur on Wt. Ratio,
Ca~alyst Catalyst, wt. % S/Re
i A 0.06 0.166
C 0.15 0.177
' G 0.15 0.171
Each such catalyst was employed for reforming at a
temperature of 900 F, a pressure of 175 psig, a hydrogen-
hydrocarbon molar ratio of 9, and a space velocity in the
range of 2 to 12, which was varied to adjust the octane level
of the Cs+ reformate to 91 RONC. The runs were terminated after
about two barrels of naphtha per pound of catalyst had been
12
processed, and the used catalyst was analyzed for carbon
content. The pertinent data are tabulated below:
14
Average Yield of Carbon on Used Catalyst
Wt. Ratio, WHSV at Reformate Wt. % of % Per Unit
Catalyst Re/Pt 91 RONC at 91 RONC Catalyst Wt. of Pt
16
A 1.08 7.0 87.6 1.15 3.44
17
C 2.5 6.7 85.8 0.49 1.44
18
G 5.65 6.0 85.6 0.28 1.81
. ' 19
The activities of the catalysts, as indicated by the
weight hour space velocities and the yields of Cs+ reformate,
were about equal for the catalysts although the initial sulfur
llevels differed. In addition, the reduction in the amount of
; llcarbon (often referred to as coke) on the used catalysts as the
¦¦rhenium/platinum ratio is increased is indicative that the
6 ~! rhenium exerts an on-stream cleansing effect, which is consistent
iwith the greater relative cycle length obtainable with catalysts
8 ,Ihaving an increased rhenium/platinum ratio relative to catalysts
llof the prior art.
3o l¦ Having thus described the invention, what is claimed
l! is
1l -14-
..
