Improved Claus plant catalysts f - DOC

Document Sample
Improved Claus plant catalysts f - DOC Powered By Docstoc
					The Role of Claus Catalyst
in Sulfur Recovery Unit Performance

Terry McHugh and Ed Luinstra
Porocel Corporation
10777 Westheimer, Suite 1100
Houston, Texas 77042
Phone 713-260-9626, 888-328-4500, fax 713-260-9627

Peter Clark
Alberta Sulphur Research Ltd.
2500 University Drive NW
Calgary, Alberta, Canada T2N 1N4
Phone 403-220-5346, fax 403-284-2054

                              Presented at Sulphur 98
                                1-4 November, 1998
                                  Tucson, Arizona
                                 The Role of Claus Catalyst
                            in Sulfur Recovery Unit Performance

    Today's sulfur recovery plants must operate at very high efficiency. To achieve the targets, high performance
    Claus catalysts are required. The best Claus catalysts have a high surface area and, in particular, a high
    macroporosity to overcome diffusional limitations. Two new catalysts were tested in the laboratory for their
    ability to convert carbon disulfide, a troublesome side product of the Claus process. The catalysts were
    Porocel Maxcel 727, an activated alumina, and an experimental promoted catalyst. Both catalysts displayed
    excellent activity for carbon disulfide decomposition.

Claus plant operators are under constant pressure to keep operating costs down while maintaining
performance and working within their licensed emission limits. To meet the targets, many factors
have to be taken into consideration. Of particular significance is the type and quality of the catalyst
in the reaction vessels.
Sulfur recovery plants convert noxious hydrogen sulfide to sulfur, a valuable product. In almost all
cases, the Claus process is used. These plants are a mainstay at refineries and natural gas plants
where large amounts of H2S are produced. While many who are involved with the Claus process
might take it for granted, it is in fact a remarkable process and of tremendous value to the oil and
gas industry.

Claus Catalyst
Claus catalyst must deliver good performance at a variety of process conditions. While the main
task is to convert H2S and SO2 to sulfur, the catalyst must also have good activity for conversion of
CS2 and COS.
CS2 is a side product that forms in the Claus reaction furnace from hydrocarbon contaminants
present in acid gas. It is a significant problem in many locations that have relatively high hydro-
carbon concentrations.2 CS2 must be destroyed to maximize sulfur recovery. This is done catalyti-
cally by hydrolysis in the first converter. Since CS2 is difficult to convert, high temperatures are
required to obtain the required reaction rates.
COS forms from CO2, CO, and possibly from conversion of CS2. Although it can be a problem, this
component contains only half as much sulfur as CS2, and is easier to destroy at first converter
conditions. In almost all cases, if CS2 can be satisfactorily reduced, COS will not be an issue. CS2
and COS both form H2S on decomposition.
H2S and SO2 conversion normally proceeds quite fast in all Claus converters, but is incomplete
because of equilibrium limitations. Low temperatures favor better conversion. Progressively lower
inlet temperatures are used in the reactors to maximize sulfur production.
A high catalyst surface area is required for the highest activity. High surface area is not sufficient
by itself, however, because the reactions are all diffusion limited. The reactant molecules must
have good access to the catalytically active sites on the surface. This is provided by the larger
pores, called macropores.
Macroporosity is usually considered to include all pores with diameters greater than 70 or 75 nm
(700-750Å). A catalyst with good macroporosity will have at least 0.15 cc/g porosity in the macro

A well operated Claus plant with three catalytic stages and with fresh catalyst can achieve an
overall sulfur recovery of 96-98.5%.

Catalyst deactivation
The catalysts used in the Claus process are susceptible to deactivation, as is common with all
catalysts. Four types of deactivation are commonly recognized in Claus catalysis: hydrothermal
aging, sulfation, carbon deposition, and liquid sulfur deposition.

Claus catalyst owes its activity to its very high surface area, generally over 300 m2/g for alumina.
Over time, the surface area declines due to the adverse effects of temperature and water vapor,
and the activity correspondingly declines. For most Claus plants, the rate of surface area loss is
quite slow, and the catalysts can be used for years.

The great enemy of Claus catalyst is sulfation.4 In the presence of SO2, the alumina catalyst sur-
face quickly builds up a layer of sulfur oxyanions. Over time, a significant sulfate layer builds up,
reducing the amount of alumina surface that is available to catalyze the reactions.

The extent to which the catalyst will sulfate depends on several factors. Temperature is important,
with lower temperatures favoring sulfation. Thus deactivation due to sulfation is normally a less
serious problem in the first converter, where high temperatures are maintained to decompose CS2
and COS. The greatest opportunity for sulfation exists in the last Claus converter, where low tem-
peratures are maintained to maximize the sulfur yield.

Sulfation is increased significantly by the presence of oxygen. Even small concentrations of oxy-
gen (for instance, due to incomplete combustion in burners) will increase sulfation of downstream
catalyst beds. The opposite effect is seen for H2S. Since it is a chemical reducing agent, H2S will
lower the amount of sulfate. Concentrations of H2S are low in the final converter, and this can
aggravate sulfation there.

The chemical nature of the catalyst affects sulfation. Compared to alumina, titania sulfates much
less.5 This is most likely due to the weak bond that titanium forms with the sulfate ion. Titania is an
excellent catalyst for the Claus process, but is 5 to 10 times more expensive than activated

Any catalyst can be deactivated by coke deposition if the acid gas feed contains heavy carbon-
containing impurities. This can be a problem in specific locations. Liquid sulfur deposition also has
a general effect on catalysts. It occurs when the reactor temperature is low, leading to filling of the
catalyst pores by capillary condensation. This happens even at temperatures several degrees
above the sulfur dew point. Raising the temperature reverses this type of deactivation.

In any case of catalyst deactivation due to pore blockage, a catalyst with a more enhanced pore
structure will minimize the damage.

Introducing Porocel
Porocel was established in Little Rock, Arkansas in the late 1930s. The company has changed
ownership several times. In May of 1996, Porocel Corporation was acquired from Engelhard by a
private group of investors. Porocel is now independently owned and completely divorced from
In January 1998, Porocel finished construction of a new state-of-the-art activated alumina plant.
New and innovative activated alumina catalyst, adsorbent and bed support products were devel-
oped and introduced to the market in 1998.
This new plant, located in Little Rock, can produce a
full line of activated alumina products. The diverse
product line includes Claus catalysts, desiccants, cata-
lyst bed supports and specialty adsorbents.
The flagship of the Porocel Claus catalyst line is Max-
cel 727, a pure activated alumina product. It has high
surface area, good strength, and excellent macropo-
rosity (Table 1). The macroporosity was conferred by a
                                                             Figure 1: Aerial view of plant
proprietary Porocel technique.
Porocel Maxcel 727 has been commercial since the beginning of 1998, when the new activated
alumina plant was started up. As of July 1998, this product is installed in seven commercial sulfur
plants. Preliminary performance data indicates excellent conversions and overall performance.
Besides Maxcel 727, Porocel is developing promoted Claus catalysts. An experimental version,
here called XPC, was produced and subjected to tests. The promoter in XPC is titania.

Performance tests
As a part of Porocel’s ongoing product development program, laboratory tests were conducted to
prove the capabilities of the catalysts. The test method and results are described here. The testing
program was carried out by Alberta Sulphur Research Limited (ASRL), Calgary, Alberta, Canada.
A high performance Claus catalyst must have high activity for conversion of H2S and SO2 as well
as for hydrolysis of CS2 and COS. The most demanding of these is conversion of CS2 at first con-
verter conditions, because it proceeds more slowly than the other reactions. If a catalyst has good

activity for CS2 conversion, it will also display good activity for the other important reactions.
Accordingly, the tests conducted at ASRL focused on CS2 conversion at first converter conditions.
This type of testing has been common to the Claus catalyst industry for many years.6

                                           Table 1. Catalyst Properties
                                                 Competitive           Porocel          Porocel
                                                  Alumina             Maxcel 727         XPC
           Surface area, m /g                          385                380             312
           Macroporosity (>75 nm) mL/g                 0.14               0.19            0.17
           Bulk density, g/mL                          0.700              0.630          0.736
           Loss on ignition (250-1000°C)                6.4                7.4            5.0
           Crush strength, lb.                          24                27.3            37.6

One objective of the lab tests was to find the effect of macroporosity on CS2 conversion. Maxcel
727 and the reference catalyst had significantly different macroporosity. Another objective was to
measure the improvement that could be provided by including titania in the catalyst.

In service, Claus catalysts experience significant initial deactivation because of hydrothermal aging
and sulfation. Since the initial activity is not characteristic of performance after a few months in
service, the catalysts were artificially aged to simulate actual plant aging. This was done by heat-
ing each catalyst at 630°C in 80% steam-20% nitrogen for 14 hours. The aging step reduced the
surface area of the catalysts to roughly what it might be after several years of service. It should be
mentioned, however, that aging varies significantly from plant to plant.
Following the aging step, each catalyst was loaded into the reactor (capacity about 70 cc) of
ASRL's equipment. The tests were conducted at typical Claus first stage conditions, and were
similar to those described previously by others . The feed gas comprised 6% H2S, 4% SO2, 1%
CS2, 28% water, balance nitrogen. This composition gives an H2S/SO2 ratio of 2 after hydrolysis of
CS2. Oxygen was included in the feed at a level of 190 ppm. The space velocity was 1200 NL/L/h,
giving a residence time (calculated at standard conditions) of 3 seconds. The reactor was isother-
mal at 320°C. Data was also obtained at 300 and 350°C.
Porocel Maxcel 727 and the experimental catalyst XPC were pitted against a competitor commer-
cial alumina, one of the best known Claus catalysts in the industry. The competitor catalyst was a
normal commercial batch; the lot was randomly chosen. Properties of the catalysts are shown in
Table 1.
While the tests were under way, the effluent gas composition was measured at frequent intervals
by gas chromatography.

Results                                                          100
The catalysts exhibited very high                                                                               Reactor temperature 320°C
                                                                                                                Space velocity ~1200 NL/L/h
initial activity with rapid deactivation.                         90
This behavior is believed to be

                                             CS2 Conversion, %
caused by initial sulfation, and is                               80
normal. At about 48 hours on line,
                                                                                                        Maxcel 727
the activity of the catalysts had                                 70
virtually leveled out. Figures 2 and 3
depict the performance as a                                       60                              Competitive
function of on-line time for the two
catalysts.                                                        50

Maxcel 727 showed a higher
conversion of CS2 compared to the
                                                   0     10       20        30       40 50 60
competitive reference through the                                    Hours on stream
entire test period. At 48 hours on
line, the difference in performance Figure 2: Maxcel 727 vs. competitive alumina
was about 3.5 percentage points.
Since there was a slight difference in space velocity, a correction was made using first order
reaction kinetics. With this correction, the difference was 2.8 points.
The experimental promoted catalyst also showed superior performance, but the advantage over
the competitive alumina diminished with hours on stream. At 42-48 hours, after correcting for
space velocity, an advantage of about 2.4 points persisted. The results are summarized in Table 2.
CS2 conversion data for the three                                100
catalysts was also determined at                                                                                 Reactor temperature 320°C
                                                                                                                Space velocity ~1200 NL/L/h
300 and 350°C. These meas-                                        90
urements were made after 48 hours
                                             CS2 Conversion, %

on line. The results are concisely                                80
presented in the form of an
Arrhenius plot (Figure 4), in which                               70
the activity for CS2 decomposition is
depicted as a first order rate                                    60
coefficient. Ideally, a logarithmic plot                                       Porocel Maxcel 727
                                                                               Porocel XPC
of the rate coefficient against                                   50
                                                                               Competitive alumina
reciprocal temperature should be
linear, and we see that this is                                   40
                                                                       0           10        20          30         40         50         60
approached in the results. Scatter
about the straight lines gives some                                                               Hours on stream

idea of experimental uncertainty.           Figure 3: XPC vs. competitive alumina
                                    Table 2. Results of laboratory comparison tests

                                                                             Competitive                                   Porocel           Porocel
                                                                              Alumina                                     Maxcel 727          XPC
                            Surface area, m /g after severe aging                   147                                       130              109
                            Reactor conditions
                                  Temperature, °C                                   320                                       320              320
                                  Space velocity, NL/L/h                        1237                                          1217            1218
                            Catalyst charge, g                                  44.4                                          40.6             47.4
                            CS2 conversion at 320°C, %
                                  Average, 42-48 h on line                      59.9                                          63.4             62.9
                                  Corrected to GHSV 1200                        61.1                                          63.9             63.5
                                  Relative improvement                              -                                          2.8              2.4

The Arrhenius plot shows that Maxcel 727 has better CS2 conversion at all temperatures tested.
The experimental catalyst XPC also has good performance. It shows a slight tendency to gain
more activity with temperature than the alumina catalysts. The competitive alumina shows more
departure from linearity than the other catalysts, and a lower slope with temperature. This is
probably partly due to experimental error, but may also reflect greater diffusional limitations due to
lower macroporosity of this catalyst. At higher temperatures, inherent reaction rates increase
                                                                  significantly, much more than
     0.0                                                          diffusion rates.

                                                                               90                                      Conclusions
                   -0.5                                                                                                Porocel's new pure alumina
                                                                               80                                      Claus catalyst, Maxcel 727, was
                                                                                     CS2 conversion, %, at 1200 GHSV

                                                                                                                       tested in the laboratory against
                                                                                                                       the leading competitive Claus
 Activity (ln k)

                                                                                                                       catalyst. Both catalysts had been
                                                                                                                       subjected to severe artificial
                                                                                                                       aging. CS2 conversion at typical
                                                                                                                       first converter conditions was
                                  Maxcel 727
                                                                                                                       measured. Maxcel 727 was
                                  Porocel XPC                                                                          found to have about 2.8
                   -2.0           Competitive alumina                                                                  percentage points more activity
                                                                                                                       than the reference catalyst. The
                          350°C                         320°C        300°C                                             improvement is attributed to the
                   -2.5                                                                                                advantage that Maxcel 727 has
                    0.00160               0.00165          0.00170   0.00175
                                                                                                                       in macroporosity.
                                                 Temperature, 1/K
                                                                                                                       An     experimental     promoted
                                                                                                                       Porocel catalyst containing titania
Figure 4. Comparison of activity at 300 to 350°
                                                                                                                       also displayed excellent activity,

but the advantage was not as great as the pure alumina product. While titania evidently performs
very well, the improvement with this particular preparation would not warrant the increased
manufacturing costs. Efforts are under way to make better promoted catalysts.

This paper was adapted from the article “The Claus Sulfur Recovery Process”, Hydrocarbon Engi-
neering, October 1998.

1   Clark, P.D., Dowling, N.I., and Huang, M., "Understanding Claus furnace chemistry: development of a
    modified Claus for low H2S-content acid gases", 48th Annual Laurance Reid Gas Conditioning
    Conference, Norman, OK, March 1-4, 1998.

2   Luinstra, E.A., and d’Haêne, P.E., Hydrocarbon Processing, July 1989, 53-57.

3   Pearson, M.J., "The role of the catalyst in the Claus process", Discovery Chemicals Claus Symposium,
    Houston, Texas, Feb. 28, 1991.

4   Hyne, J.B., Ho, K., ASRL Quart. Bull., XV (3,4), 49-52 (1978-79).

5   Kettner & Lübcke, "Experience in the commercial use of a new Claus catalyst: the importance of
    COS/CS2 in Claus plants", Sulphur 82.

6   Nédez, C., and Ray, J-L., Catalysis Today 27, 49-53, 1996.


Shared By: