Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-1-
HYDROCARBON SYNTHESIS CATALYST ENHANCEMENT WITH
HYDROGEN AND AMMONIA
BACKGROUND OF THE DISCLOSURE
Field of the Invention
The invention relates to enhancing a hydrocarbon synthesis catalyst with
hydrogen and ammonia. More particularly, the invention relates to improving
the performance of a hydrocarbon synthesis catalyst, including a Fischer-
Tropsch type of hydrocarbon synthesis catalyst, by contacting the catalyst
with a
reducing gas comprising a mixture of hydrogen and ammonia, at catalyst
reduction conditions, and to a hydrocarbon synthesis process using the
enhanced
catalyst.
Background of the Disclosure
The synthesis of hydrocarbons, including oxygenated hydrocarbons such
as methanol, from a synthesis gas comprising a mixture of H2 and CO is well
known. The synthesis gas feed is contacted with a Fischer-Tropsch catalyst at
conditions effective for the H2 and CO in the feed gas to react and form
hydrocarbons. The synthesis is known as a Fischer-Tropsch hydrocarbon
synthesis. Depending on the catalyst and conditions, the hydrocarbons may
range from oxygenated compounds such as methanol and higher molecular
weight alcohols, to high molecular weight paraffins which are waxy solids at
room temperature. The process also makes, in lesser amounts, alkenes,
aromatics, organic acids, ketones, aldehydes and esters. The synthesis is
conducted in a fixed or fluidized catalyst bed reactor or in a liquid phase
slurry
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-2-
reactor. Hydrocarbon synthesis catalysts are also well known and typically
include a composite of at least one iron group catalytic metal component
supported on, or composited with, with at least one inorganic refractory metal
oxide support material, such as alumina, amorphous, silica-alumina, zeolites
and
the like. Various catalyst preparation methods have been used to form
hydrocarbon synthesis catalysts, including impregnation, incipient wetness,
compositing, ion exchange and other known techniques, to form a catalyst
precursor. The precursor must be activated to form the catalyst. Typical
activation methods include oxidation or calcination, followed by reduction in
flowing hydrogen, multiple oxidation-reduction cycles and also reduction
without prior oxidation. Examples of catalyst preparation and activation
methods for Fischer-Tropsch hydrocarbon synthesis catalysts are disclosed in,
for example, US patents US 4,086,262 ; 4,492,774 and 5,545,674.
SUMMARY OF THE INVENTION
The invention relates to enhancing or improving the performance of an
active hydrocarbon synthesis catalyst, including a Fischer-Tropsch type of
hydrocarbon synthesis catalyst, by contacting the catalyst with a reducing gas
comprising a mixture of hydrogen and ammonia, at conventional catalyst
reduction conditions effective to form a active catalyst, and to a hydrocarbon
synthesis process using the enhanced catalyst. By catalyst is meant an active
hydrocarbon synthesis catalyst wherein at least a portion, and preferably only
a
portion, of the one or more catalytic metals (e.g., Co, Fe, Ni) is in the
reduced or
catalytically active metal form, as a consequence of reduction achieved by
contacting the catalyst precursor in a hydrogen reducing gas without ammonia,
at conventional hydrogen reducing conditions. Thus, while the catalyst may be
completely reduced and fully active prior to contact with the hydrogen and
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-3-
ammonia mixture in the practice of the invention, it is preferred that it be
partially reduced in hydrogen without ammonia and partially reduced in a
reducing gas comprising hydrogen and ammonia, to achieve complete reduction
and activation. It has been found that contacting the catalyst with a reducing
gas
comprising a mixture of hydrogen and ammonia enhances the hydrocarbon
synthesis properties of the resulting active catalyst, with respect to at
least one
of; increased CS+ selectivity, increased alpha (Schultz-Flory alpha) of the
synthesis reaction, and a reduction in methane make. These benefits are
unexpected, in view of the fact that ammonia is a well known hydrocarbon
synthesis catalyst poison. Thus, by enhanced catalyst, in the context of the
invention, is meant that at least one, preferably at least two, and more
preferably
all three of the following are achieved during hydrocarbon synthesis: (i) the
CS+
selectivity of the catalyst is greater than it otherwise would be if the
catalyst had
not been contacted with a reducing gas comprising a mixture of hydrogen and
ammonia; (ii) the alpha of the synthesis reaction is greater than it would
otherwise be if the catalyst had not been contacted with a reducing gas
comprising a mixture of hydrogen and ammonia, and (iii) the methane make is
reduced to a level lower than it would otherwise be if the catalyst had not
been
contacted with a reducing gas comprising a mixture of hydrogen and ammonia.
The catalyst precursor may or may not be calcined prior to activation by
reduction. Either or both the hydrogen reducing gas, and the reducing gas
comprising the mixture of hydrogen and ammonia, may or may not contain one
or more diluent gasses which do not adversely effect or interfere with the
reduction and concomitant activation of the catalyst. Such gasses include
methane, argon and the like. The amount of ammonia present in the hydrogen
reducing gas will broadly range from 0.01 to 15 mole %, preferably 0.01 to 10
mole %, more preferably from 0.1 to 10 mole % and still more preferably from
0.5 to 7 mole %, based on the total gas composition. The hydrogen to ammonia
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-4-
mole ratio in the gas will range from 1000:1 to 5:1 and preferably from 200:1
to
10:1.
In one embodiment the invention is a process which comprises enhancing
a Fischer-Tropsch type of hydrocarbon synthesis catalyst, comprising at least
one catalytic metal component, and preferably at least one catalytic metal
component and a metal oxide support type of component, by contacting the
catalyst with a reducing gas comprising a mixture of hydrogen and ammonia, at
catalyst reduction conditions effective to reduce the one or more catalytic
metal
components to the reduced, catalytically active metal form, and preferably
wherein a portion of the one or more catalytic metal components is reduced to
the catalytically active metal form as a result of the contacting. In another
embodiment, the invention comprises a process for synthesizing hydrocarbons
from a synthesis gas which comprises a mixture of H2 and CO, wherein the
synthesis gas is contacted with an enhanced Fischer-Tropsch type of
hydrocarbon synthesis catalyst, at reaction conditions effective for the H2
and
CO in the gas to react and form hydrocarbons, wherein the enhanced catalyst
comprises a composite of at least one catalytic metal component and metal
oxide
support component, and has been enhanced by contacting the catalystwith a
reducing gas comprising a mixture of hydrogen and ammonia, at catalyst
reduction conditions effective to reduce the one or more catalytic metal
components to the reduced, catalytically active metal form, and preferably
wherein a portion of the one or more catalytic metal components is reduced to
the catalytically active metal form as a result of the contacting. In a still
further
embodiment, at least a portion of the synthesized hydrocarbons are liquid at
the
reaction conditions. By conventional catalyst reduction conditions is meant
conditions of temperature, pressure, hydrogen partial pressure and the space
velocity of a conventional hydrogen reducing gas, sufficient to reduce the one
or
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-S-
more catalytic metal components of the precursor to the metal and form an
active catalyst.
DETAILED DESCRIPTION
Hydrocarbon synthesis catalysts are well known and a typical Fischer-
Tropsch hydrocarbon synthesis catalyst will comprise, for example,
catalytically
effective amounts of one or more Group VIII metal catalytic components such as
Fe, Ni, Co and Ru. Preferably the catalyst comprises a supported catalyst,
wherein the one or more support components of the catalyst will comprise an
inorganic refractory metal oxide. The metal oxide support component is
preferably one which is difficult to reduce, such an oxide of one or more
metals
of Groups III, IV, V, VI, and VII. The metal Groups referred to herein are
those
found in the Sargent-Welch Periodic Table of the Elements, ~ 1968. Typical
support components include one or more of alumina, silica, and amorphous and
crystalline aluminosilicates, such as zeolites. Particularly preferred support
components are the Group IVB metal oxides, especially those having a surface
area of 100 m2/g or less and even 70 m2/g or less. These support components
may, in turn, be supported on one or more support materials. Titania, and
particularly rutile titanic, is a preferred support component, especially when
the
catalyst contains a cobalt catalytic component. Titanic is a useful component,
particularly when employing a slurry hydrocarbon synthesis process, in which
higher molecular weight, primarily paraffinic liquid hydrocarbon products are
desired. In some cases in which the catalyst comprises catalytically effective
amounts of Co, it will also comprise one or more components or compounds of
Re, Ru, Fe, Ni, Th, Zr, Hf, U, Mg and La, some of which are effective as
promoters. A combination of Co and Ru is often preferred. Useful catalysts and
their preparation are known and illustrative, but nonlimiting examples may be
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-6-
found, for example, in U.S. patents 4,568,663; 4,663,305; 4,542,122; 4,621,072
and 5,545,674.
The catalysts are prepared by any convenient and known method, such as
impregnation, incipient wetness, ion exchange, kneading, precipitation or
coprecipitation, melt deposition or any other compositing techniques to form
the
catalyst precursor. The catalytic metal component is typically applied as a
solution of a compound which decomposes during the subsequent calcination
and/or reduction. For example, a cobalt component is typically applied as a
nitrate salt. It is not uncommon to calcine the precursor after each
application of
reducible catalytic metal compound, to achieve better catalytic metal
dispersion.
After forming and extruding the precursor composite, it is typically pilled
and
dried. The precursor is then reduced or calcined and reduced, to form the
catalyst. The reduction is achieved by contacting the precursor with flowing
hydrogen or a hydrogen reducing gas, at conditions effective to reduce the
catalytically active metal component (e.g., cobalt) to the metal form. A
common
method is known as the R-O-R method, in which the precursor is reduced in
hydrogen, then calcined, followed by reducing again. As mentioned above, the
reducing hydrogen gas can be neat (all hydrogen), or mixed with one or more
diluent gasses (e.g., methane, argon) which are inert towards the reduction.
In
the practice of the invention, the R-O-R method may also be used and a
conventional hydrogen reducing gas used for the first and at least a portion
of the
second reduction, and preferably with a portion of the final reduction
achieved
by a reducing gas comprising a mixture of hydrogen and ammonia. Typical
reducing conditions effective for forming the catalyst comprising the reduced
metal component on the support from the precursor, range from 1/z to 24 hours,
200-500°C, 1-100 bar, and a GHSV of 50-10000. The actual conditions
will
depend on the hydrogen concentration in the reducing gas, as well as the metal
to be reduced and its precursor form (e.g., salt or oxide). In the catalyst
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
_ '7 _
enhancing process of the invention, the at least partially reduced and active
catalyst is contacted with a mixture of a hydrogen reducing gas and ammonia,
at
typical reducing conditions, as set forth above, similar to those used for
normal
reduction and concomitant activation. The catalyst enhancement may be
conducted according to the process of the invention, either prior to loading
it into
the hydrocarbon synthesis reactor or in-situ in the hydrocarbon synthesis
reactor.
The enhanced catalyst may be used in either a fixed bed, fluid bed or
slurry hydrocarbon synthesis processes, for forming hydrocarbons from a
synthesis gas comprising a mixture of H2 and CO. These processes are well
known and documented in the literature. In all of these processes, the
synthesis
gas is contacted with a suitable Fischer-Tropsch type of hydrocarbon synthesis
catalyst, at reaction conditions effective for the H2 and CO in the gas to
react and
form hydrocarbons. Depending on the process, the catalyst and synthesis
reaction variables, some of these hydrocarbons will be liquid, some solid
(e.g.,
wax) and some gas at standard room temperature conditions of temperature and
pressure of 25°C and one atmosphere, particularly if a catalyst having
a catalytic
cobalt component is used. In a fluidized bed hydrocarbon synthesis process,
all
of the products are vapor or gas at the reaction conditions. In fixed bed and
slurry processes, the reaction products will comprise hydrocarbons which are
both liquid and vapor at the reaction conditions. Slurry hydrocarbon synthesis
processes are sometimes preferred, because of their superior heat (and mass)
transfer characteristics for the strongly exothermic synthesis reaction and
because they are able to produce relatively high molecular weight, paraffinic
hydrocarbons when using a cobalt catalyst. In a slurry hydrocarbon synthesis
process, a synthesis gas comprising a mixture of H2 and CO is bubbled up as a
third phase through a slurry in a reactor which comprises a particulate
Fischer-
Tropsch type hydrocarbon synthesis catalyst dispersed and suspended in a
slurry
liquid comprising hydrocarbon products of the synthesis reaction which are
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
_g_
liquid at the reaction conditions. The mole ratio of the hydrogen to the
carbon
monoxide in the synthesis gas may broadly range from about 0.5 to 4, but is
more typically within the range of from about 0.7 to 2.75 and preferably from
about 0.7 to 2.5. The stoichiometric mole ratio for a Fischer-Tropsch
hydrocarbon synthesis reaction is 2.0, but it can be increased to obtain the
amount of hydrogen desired from the synthesis gas for other than the
hydrocarbon synthesis reaction. In a slurry hydrocarbon synthesis process, the
mole ratio of the H2 to CO is typically about 2.1/I. Reaction conditions
effective
for the various hydrocarbon synthesis processes will vary somewhat, depending
on the type of process, catalyst composition and desired products. Typical
conditions effective to form hydrocarbons comprising mostly CS+ paraffins,
(e.g.,
CS+ C2~) and preferably C1~ paraffins, in a slurry process employing a
catalyst
comprising a supported cobalt component include, for example, temperatures,
pressures and hourly gas space velocities in the range of from about 320-
600°F,
80-600 psi and 100-40,000 Vlhr/V, expressed as standard volumes of the
gaseous CO and H2 mixture (0°C, 1 atm) per hour per volume of catalyst,
respectively. These conditions nominally apply to the other processes as well.
Hydrocarbons produced by a hydrocarbon synthesis process according to
the practice of the invention are typically upgraded to more valuable
products,
by subjecting all or a portion of the CS+ hydrocarbons to fractionation and/or
conversion. By conversion is meant one or more operations in which the
molecular structure of at least a portion of the hydrocarbon is changed and
includes both noncatalytic processing (e.g., steam cracking), and catalytic
processing (e.g., catalytic cracking) in which a fraction is contacted with a
suitable catalyst. If hydrogen is present as a reactant, such process steps
are
typically referred to as hydroconversion and include, for example,
hydroisomerization, hydrocracking, hydrodewaxing, hydrorefining and the more
severe hydrorefining referred to as hydrotreating, all conducted at conditions
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-9-
well known in the literature for hydroconversion of hydrocarbon feeds,
including
hydrocarbon feeds rich in paraffins. Illustrative, but nonlimiting examples of
more valuable products formed by conversion include one or more of a synthetic
crude oil, liquid fuel, olefins, solvents, lubricating, industrial or
medicinal oil,
waxy hydrocarbons, nitrogen and oxygen containing compounds, and the like.
Liquid fuel includes one or more of motor gasoline, diesel fuel, jet fuel, and
kerosene, while lubricating oil includes, for example, automotive, jet,
turbine
and metal working oils. Industrial oil includes well drilling fluids,
agricultural
oils, heat transfer fluids and the like.
The invention will be further understood with reference to the examples
below.
EXAMPLES
ExamQle 1 (Silica support)
A commercially available silica gel known as KCKG #4 (manufactured
by Salavat Catalyst Factory of the Salavat Petrochemical Complex, Salavant,
Russia), 2-4 mm diameter, was ground and sieved to obtain a 0.106-0.250 mm
size fraction. This material was then calcined in flowing air at 450°C
for 5
hours, to form the support for the catalysts prepared below.
Example 2 (Catalyst A precursor)
A solution of 5.18 gm of Co(N03)2*6H20 in 15 ml of distilled water was
prepared. This solution was added to 21 ml (8.38 gm) of the calcined silica
support from Example 1, with stirring, to form a catalyst precursor. Then the
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
- 10-
catalyst precursor was dried on a steam bath. At this stage, the catalyst
precursor
contained nominally 11 wt % cobalt and is the catalyst A precursor.
Example 3 (Catalyst B precursor)
An aqueous solution of 4.9 gm of Zr0(N03)2*2H20 was added to 33.2
gm of the calcined silica support of Example 1, in a volume of water
sufficient to
fully wet the silica, which was then dried over a steam bath, followed by
calcining in flowing air for 1 hour at 450°C, to form a first
composite. Then a
solution of 75 gm of Co(N03)2*6H20 in 30 ml of water was added to the
composite which soaked for 2 hours at room temperature. The excess solution
was decanted off and saved. The resulting second composite was dried on a
steam bath and then calcined in flowing air for 2 hours at 450°C. After
cooling,
the saved excess solution was added to the second composite. The steps of
soaking, decanting, drying, and calcining were repeated until all of the
Co(N03)2*6H20 solution was impregnated onto the composite to form the final
catalyst precursor, which is the catalyst B precursor. The so-formed catalyst
B
precursor contained 27 wt. % cobalt and 4.1 wt. % zirconium oxide.
Example 4 (Catalyst B reduction in H2)
The catalyst B precursor of Example 3 (20 ml) was mixed with 80 ml of
1-3 mm quartz particles and the mixture placed into a 25 mm ID quartz reactor.
The catalyst/quartz mixture was held in place with glass wool at the bottom of
the reactor and a layer consisting of 10 ml of the 1-3 mm quartz particles on
top
of the catalyst/quartz mixture. Hydrogen was then passed through the reactor
at
room temperature and atmospheric pressure at a gas hourly space velocity
(GHSV) of 100 hr-' for 15 minutes. Prior to entering the reactor, the hydrogen
was passed through a column of KOH pellets (pellet diameter nominally 3-5
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-11-
mm) for removal of impurities. The reactor temperature was increased to
450°C
over 40-45 minutes. This condition was held for 5 hours. Then the reactor was
allowed to cool to room temperature in flowing hydrogen. After the reactor had
cooled, the hydrogen flow was replaced with a flow of 2:1 HZ:CO synthesis gas
at 100 hr-1 GHSV, for 15 minutes at atmospheric pressure. As with the
hydrogen, the synthesis gas was passed through a column of KOH pellets for
removal of impurities. Then valves were closed at the inlet and outlet of the
reactor, storing the catalyst under the synthesis gas.
Exam~,le 5 (Catalyst B reduction in H2 and then NH3/H2)
A 20 ml sample of the catalyst B precursor of Example 3 was mixed with
80 ml of 1-3 mm quartz particles and the mixture placed into a 25 mm ID quartz
reactor. The catalysdquartz mixture was held in place with glass wool at the
bottom of the reactor and a layer consisting of 10 ml of the 1-3 mm quartz
particles on top of the catalyst/quartz mixture. Hydrogen was then passed
through the reactor at room temperature and atmospheric pressure at a gas
hourly
space velocity (GHSV) of 100 hr'1 for 15 minutes. Prior to entering the
reactor,
the hydrogen was passed through a column of KOH pellets (pellet diameter
nominally 3-5 mm) for removal of impurities and through a 3-necked flask
containing NaOH pellets. The center neck of the 3-necked flask was equipped
with a syringe for addition of 29 wt % NH3/ 71 wt % H20 solution. The NaOH
in the 3-necked flask served to absorb the water from the NH3/H20 solution,
liberating the NH3 vapor, which was then swept out of the flask and into the
reactor. The reactor temperature was increased to 400°C over 40-45
minutes. In
preparing the catalysts, the procedure used in Example 4 (in which the reactor
temperature was increased from room temperature to 450°C over 40-45
minutes
and held at 450°C for 5 hours) was used until the reactor reached the
450°C
temperature. This was followed by 5 hours of reduction (i) with HZ only for
the
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
- 12-
first sample, (ii) with H2 followed a mixture of H2 and NH3 for the next four
samples and (iii) with a mixture of H2 and NH3 only for the fifth sample, as
shown in Table 2 below. Ammonia addition into the H2 gas to provide the
H2/NH3 reducing gas mixture, was achieved by a continuous drop-wise addition
of the 29 wt % NH3/ 71 wt % HZO solution from the syringe. The nominal
concentration of NH3 in the reducing gas during the NH3/H20 solution addition
was 5 mole %. Irrespective of whether the reduction was achieved with (i) all
H2, (ii) H2 followed by the H2/NH3, or (iii) all H2/NH3, the total reduction
time
was 5 hours, as shown in Table 2. Following reduction, the reactor was allowed
to cool to room temperature in flowing hydrogen. After the reactor had cooled,
the hydrogen flow was replaced with a flow of 2:1 HZ:CO synthesis gas at 100
hr-1 GHSV for 15 minutes at atmospheric pressure. As with the hydrogen, the
synthesis gas was passed through a column of KOH pellets, for removal of
impurities. Then valves were closed at the inlet and outlet of the reactor,
storing
the catalyst under the synthesis gas.
Example 6 (Catalyst A reduction with NH3/H2)
A 20 ml sample of the catalyst A precursor of Example 2 was mixed with
80 ml of 1-3 mm quartz particles and the mixture placed into a 25 mm ID quartz
reactor. The catalyst/quartz mixture was held in place with glass wool at the
bottom of the reactor and a layer consisting of 10 ml of the 1-3 mm quartz
particles on top of the catalyst/quartz mixture. Hydrogen was then passed
through the reactor at room temperature and atmospheric pressure at a gas
hourly
space velocity (GHSV) of 3000 hr-I for 15 minutes. Prior to entering the
reactor,
the hydrogen was passed through a column of KOH pellets (pellet diameter
nominally 3-5 mm) for removal of impurities and through a 3-necked flask
containing NaOH pellets. The center neck of the 3-necked flask was equipped
with a syringe for addition of 29 wt % NH3/ 71 wt % H20 solution. The NaOH
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-13-
in the 3-necked flask served to absorb the water from the NH3/H20 solution,
liberating the NH3 vapor, which was then swept out of the flask to the
reactor.
The reactor temperature was increased to 400°C over 40-45 minutes.
After
reaching 400°C, drop-wise addition of the 29 wt % NH3/ 71 wt. % H20
solution
was commenced from the syringe. The addition rate was varied to give a
nominal concentration of NH3 in the reducing gas between 0 (HZ-only) and 3.0
mole %. This condition was held for 1 hour. Then the reactor was allowed to
cool to room temperature in flowing hydrogen. After the reactor had cooled,
the
hydrogen flow was replaced with a flow of 2:1 H2:C0 synthesis gas at 100 hr-1
GHSV, for 15 minutes at atmospheric pressure. As with the hydrogen, the
synthesis gas was passed through a column of KOH pellets for removal of
impurities. Then valves were closed at the inlet and outlet of the reactor,
storing
the catalyst under the synthesis gas.
Example 7 (Testing of Catalyst A)
The flow of synthesis gas into the reactor was resumed with the catalyst
of Example 6 (Catalyst A reduced with H2+NH3) at 100 hr-1 GHSV and 1 atm
pressure. Prior to entering the reactor, the synthesis gas was passed through
a
column of KOH pellets (pellet diameter nominally 3-5 mm) for removal of
impurities. The synthesis gas composition was 2:1 H2:C0 by volume. The
reactor temperature was increased from room temperature to 160°C in
about 40
minutes. This condition was held for 5 hours, after which the reactor was
cooled
down to room temperature in the flowing synthesis gas and the catalyst stored
under the synthesis gas as described in Example 6. Testing was resumed the
next day following the same procedure, except that the test temperature was
10°C higher. This was repeated each day until the optimum operating
temperature was found. The optimum operating temperature was defined as the
temperature where the yield of CS+products is maximized, as measured in gm of
CA 02395682 2002-06-25
WO 01/49809 PCT/~JS00/35329
- 14-
CS+ product per standard cubic meter of synthesis gas fed to the reactor.
Finding
the optimum operating temperature entailed increasing the reactor temperature
in
10°C steps until the CS+ yield decreased from the previous test. The
temperature
from the previous test is the optimum temperature. Catalyst performance was
determined by measuring the gas contraction, product gas composition by gas
chromatography, and C5+ liquid product yield. The CS+ liquid product was
recovered form the reactor effluent using two traps. The first trap was water
cooled and the second was cooled with dry ice/acetone (-80°C). The CS+
product
in the first trap was weighted directly. The liquid product in the second trap
was
first warmed to room temperature, to vaporize C4_ components prior to
weighing.
The combined weights of the hydrocarbon liquid product in both traps was used
to determine the CS+ product yield. The C5+ product from the optimum
temperature was further analyzed, to determine hydrocarbon type and carbon
chain length distribution. From time to time, the CS+ products from the non-
optimum temperature tests were combined and analyzed. The catalyst precursor
was not calcined prior to reduction in these experiments. The results are
shown
in Table 1 below.
Table 1
Yield, Selectivity,
gm/m3 %
of lpha
Gas
Feed
Mole % Optimu CO CH4 CS+ C4_ CS+
NH3 m Conversio
in ReducingTemp., n
Gas C %
0 200 60 26 7 40 60 0.8
8
0.5 210 45 9 76 37 76 0.84
0.75 210 49 24 70 43 57 0.85
1 210 58 21 72 42 58 0.84
2 200 42 10 72 20 80 0.84
3 210 24 4 40 25 75 0.85
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-15-
These results demonstrate the effect on catalyst performance, of the NH3
concentration in the reducing gas used to convert the precursor to the
catalyst.
Up to about 2 mole % NH3 in the reducing gas, both the % CO conversion and
CS+ yields are good, with the CS+ selectivity peaking at about 80%. Although
the
catalyst activity decreased when NH3 was used in the reducing gas, the
reduction
is mainly from the C4_ gas products. The Schultz-Flory alpha also increased
when the catalyst reducing gas contained NH3 and stayed essentially constant
from 0.5 to 3.0 mole % NH3 in the reducing gas. These results show that the
presence of NH3 in the reducing gas during catalyst reduction, improved
catalyst
performance. Thus, in all cases, reduction in the presence of NH3 increased
the
alpha of the hydrocarbon synthesis reaction. The greatest difference between
having and not having NH3 in the reducing gas, is seen for the run in which
the
H2 reducing gas contained 0.5 mole % NH3. This produced a CS+ selectivity of
76 %, with a high yield of CS+ hydrocarbons and only a 9 % methane make. At
2 mole % NH3, the CS+ selectivity was even higher and the CS+ yield, while
lower, was still good.
Example 8 (Catalyst B testing after Reduction with H2 and then with H2+NH3)
The synthesis gas flow was resumed into the reactor with the catalyst of
Example 5 (Catalyst B reduced first with H2 then with H2+NH3) at 100 hr-'
GHSV and 1 atm pressure. Prior to entering the reactor, the synthesis gas was
passed through a column of KOH pellets (pellet diameter nominally 3-5 mm) for
removal of impurities. The synthesis gas composition was 2:1 H2:C0 by
volume. The reactor temperature was increased from room temperature to
160°C in about 40 minutes. This condition was held for 5 hours, after
which the
reactor was cooled to room temperature in the flowing synthesis gas and the
catalyst stored under the synthesis gas. Testing was resumed the next day
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
- 16-
following the same procedure, except that the test temperature was 10°C
higher.
This was repeated each day until the operating temperature was
190°C. At
190°C the catalyst performance was determined by measuring the gas
contraction, product gas composition by gas chromatography, and CS+ liquid
product yield. The CS+ liquid was recovered form the reactor effluent using
two
traps. The first trap was water cooled and the second was cooled with dry
ice/acetone (-80°C). The CS+ product in the first trap was weighted
directly.
The liquid product in the second trap was first warmed to room temperature to
vaporize C4_ components prior to weighing. The combined weights of the
hydrocarbon liquid product in both traps was used to determine the CS+ product
yield. The CS+ product from the optimum temperature was further analyzed to
determine hydrocarbon type and carbon chain length distribution. From time to
time, the CS+ products from the non-optimum temperature tests were combined
and analyzed. The catalyst precursor used in these experiments was calcined
prior to reduction. It should be noted that while the cobalt oxide formed by
the
calcination converts to the metal during the reduction, the zirconium
component
remains as the oxide and is not reduced to the metal. The results at
190°C are
shown in Table 2 below.
CA 02395682 2002-06-25
WO 01/49809 PCT/US00/35329
-17-
Table 2
Reduction Yield, Selectivity,
Time, gm/m %
Hours* of Gas
Feed
HZ H2 + NH3 CO CH4 Cs+ C4_ Cs+ Alpha
Conversion
0 87 34 87 53 47 0.75
4 1 72 10 90 23 77 0.76
3 2 58 8 97 19 81 0.82
2 3 50 8 96 16 84 0.86
1 4 46 9 84 20 80 0.82
0 5 47 11 70 24 76 0.81
* Total reduction time of 5 hours in each case. Number in HZ column is
reducing time under hydrogen, before switching to a mixture of H2 and NH3 (5
mole %).
Table 2 shows how, at a constant reducing time of 5 hours, the
performance of Catalyst B is affected by the presence of NH3 in the H2
reducing
gas, when sequentially reduced with H2 followed by HZ+NH3, and also when
reduced in only H2+NH3. It should be noted, that tests revealed optimum
catalytic properties for the catalyst reduced under hydrogen, at a hydrogen
treat
gas ratio of 100 hr-1 GHSV, were achieved at a reduction time of five hours.
This is why the total reduction time of 5 hours was chosen for this
experiment.
However, a portion of the metal reduction and concomitant activation is
achieved in one hour. Thus, an active catalyst existed prior to contact with
the
hydrogen and ammonia reducing gas, for the runs in Table 2 above, in which the
contact time with the hydrogen reducing gas was, 1, 2, 3 and 4 hours. In the
last
run in the Table, it was a precursor that was contacted with the hydrogen and
ammonia mixture for 5 hours. Thus, this last run and the first run with only
hydrogen reduction, are not within the scope of the invention and are
presented
for comparative purposes.
10-01-W~~ 9~~~ A'02395682'2002-06-25 ~'G~ ~~~Ii0lt7t~~td(,)S0035329
- 18 - [SUBSTITUTE PAGEI
As Table 2 shows, in the second tturough the fifth runs, the preewrsor was
at least ps~rtially reduced prior to contact with the hydrogan and ammonia
mixture. With 4 hours on H~, foDowed by only 1 hour on Hz+NH3, a sharp
increase in the C5~ yidd and selectivity occurred, with a concomitantly sharp
drop in methane make. Thus, Increasing the reduction time with H~+NH3 caused
the Cs~. yi~d, Cs+ selectivity, and Schuh-Flory alpha to peak, between 2 to 4
ho'ars of H2+NH3 reduction. Although the catalyst activity drops with
increasing
H2+NH3 reduction time, this activity drop is taken fronn the Cs. gas, leaving
the
yield of C5~ approximately constant up to three hours of the Hz/NH3 treatment
More than .three hours d~ereased the CS+ yield and selectivity, but the alpha
of
the reaction remained high. This shows that iatroduetion of NH3 into the
reducing gas, during at least a portion of the catalyst reduction, is e$cctive
in
improving catalyst performance.
Vl5-d ll0/800'd eBe-! Bti8E0E~808 'ld~ ~IY7 1190WNOYX3-~~~ ~1,~:80 ZOEZ-Qi-~~f
AMENDED SHEET
