GB1578230A - Process for removal of hydrogen sulphide and hydrogen polysulphide from liquid sulphur - Google Patents

Description

(54) PROCESS FOR REMOVAL OF HYDROGEN SULFIDE AND HYDROGEN POLYSULFIDE FROM LIQUID SULFUR (71) We, STANDARD OIL COM PANY, a corporation organized and existing - under the laws of the State of Indiana, United States of America of 200 East Randolph Drive, Chicago, Illinois 60601, United States of America, do hereby declare the invention for which we pray that a patent may be granted to us and the method by which it is to be performed, to be particularly described in and by the following Statement:- The present invention relates to an improved process for catalytic conversion of hydrogen polysulfides to hydrogen sulfide in liquid sulphur and degasification of the liquid sulphur. More specifically, it is concerned with the removal of hydrogen polysulfides and hydrogen sulfide present in liquid sulfur produced by the Claus process.

The toxicity and combustion hazards associated with gaseous hydrogen sulfide are well recognized and documented in literature. Further, the presence of hydrogen sulfide dissolved in sulfur (usually from 200 to 600 ppm by weight, particularly in sulfur produced in a Claus plant or sulfur from certain natural sources), and its associated slow release during subsequent handling and transportation are equally recognized commercially as serious health hazards.

Normally in a gas/liquid system the adsorption rate of the gas is lower at higher temperatures. Thus in principle, the hot liquid sulfur stream in contact with a gaseous phase containing hydrogen sulfide, as found in a Claus plant, should not represent a serious problem if the dissolution is the only adsorption process.

However, the hydrogen sulfide is known to combine with the sulfur to form hydrogen polysulfides according to the following reaction: SX+H2SH2SX+l The formation of the polysulfides are favored at the high temperatures associated with the Claus plant. This is particularly true during the initial oxidation step in the furnace and boiler where the major portion of the sulfur is also produced.

Unfortunately, the kinetics of the reverse reaction at lower temperatures characteristic of hydrogen sulfide removal are extremely slow. Thus, the polysulfides are inherently produced in the Claus process, and once formed are extremely slow in decomposing. Consequently, the apparent solubility of hydrogen sulfide in liquid sulfur is unexpectedly high due to the formation of polysulfides, and the subsequent release or removal of hydrogen sulfide is slow and difficult, frequently involving significant quantities of hydrogen sulfide being released days and even weeks after formation.

In response to this problem, commercial specifications have been suggested and adopted setting from 5 to 10 ppm by weight as the maximum desired H2S content for safe handling, storage and transportation of bulk quantities of liquid sulfur. To comply with these conditions, it has been recommended that any time the H2S content exceeds 15 ppm a H2S removal process should be employed.

Various techniques and methods have been proposed to accomplish the removal of H2S dissolved in sulfur. In British Patent 1,067,815 a degasification process for removal of sulfur is proposed. The liquid sulfur containing hydrogen sulfide is atomized by forcing it through a jet or nozzle and then the resulting spray is directed against an obstacle, thus promoting the removal of the gaseous H2S.

It was further disclosed that the presence of ammonia (100 ppm) promoted the removal of H2S. In the absence of the use of ammonia the H2S reduction is extremely slow, involving long time spans. The use of ammonia inherently results in a contaminated product.

Alternative methods for removal of H2S reminiscent ofthe Claus reaction have been proposed in U.S. Patent 3,447,903 and Canadian Patent 964,040. In 3,447,903 a catalytic process for producing elemental sulfur from H2S and SO2 in liquid sulfur is disclosed. The catalyst involved is described generically as a basic nitrogen compound having a KB value (in water) greater than 10'0 and a solubility in molten sulfur of at least one part per million. This process, as taught, can be practiced for the purpose of controlling purity of liquid sulfur containing small concentrations of H2S. Canadian Patent 964,040 involves injecting liquid SO2 and a nitrogen containing compound, which complexes with the SO2 to form an adduct, into the molten sulfur for the expressed purpose of having the SO2-nitrogen adduct react with the undesirable polysulfide dissolved in sulfur. Hence, it is known that certain nitrogen compounds in combination with SO2 will catalytically reduce the H2S and H2SX concentration found in liquid sulfur.

Such processes again inherently involve soluble nitrogen containing species being present in the sulfur after degradation of the sulfide and polysulfides; i.e., the processes merely replace one contaminant for another contaminant.

In a more recent U.S. Patent 3,807,141 an apparatus for reducing the H25 and H2SX content of liquid sulfur without the addition of other contaminants such as ammonia or hydrogen sulfide reacting amines is disclosed. The apparatus involves a vertical liquid sulfur scrubbing tower, wherein, the liquid sulfur flows downward through the tower passing from one of a series of Lshaped baffle plates to another which tends to agitate and increase the surface area of the liquid sulfur allowing the dissolved H25 to escape. Although the sulfur recovered from the apparatus is free of nitrogen contaminants, the use of this device will involve either 1 to 9 days of continuous recycle or 1 to 8 days of storage prior to passing the sulfur over the series of baffles to insure breakdown of the polysulfides.

Such time spans are impractical with respect to contemporary large scale commercial operations.

As summarized in an article entitled "H2S Removal from Liquid Sulphur" by F.

W. King presented at the November, 1973 meeting of Canadian Natural Gas Processing Association and published in the Energy Processing/Canada, March-April, 1974, the liberation of H25 from liquid sulfur takes place in two ways, i.e., through a drop in temperature and through physical agitation. As implied in this article as well as the previously mentioned patents, the extremely slow conversion of hydrogen polysulfide back to hydrogen sulfide prior to degasificntion is the overall rate limiting step and the primary source of major conccrn. In addition to the known use of soluble amines and ammonia to catalyze the decomposition of H2SX, an article published by W. J. Rennie entitled "The Removal of H25 'Dissolved' in Liquid Sulphur" in the Alberta Sulphur Research LTD. Quarterly Bulletin, v IX, No. 4, January March, 1973, discloses on a laboratory scale the use of alumina, bauxite and PbS supported on alumina as a solid catalyst for the conversion of H2SX to H25 and suggests that they may be useful on a commercial plant scale.

In view of the aforementioned limitations and problems, we have developed a process for catalytic degradation of hydrogen polysulfide to hydrogen sulfide in liquid sulfur and removal of hydrogen sulfide from the liquid sulfur, which comprises contacting the liquid sulfur with a solid degradation catalyst selected from alumina and alumina impregnated with a cobalt-molybdenum catalyst at a temperature from 250"F to 3200 F, simultaneously purging the liquid sulfur in contact with said catalyst with a molecular oxygen-containing purge gas preferably selected from air and oxygen enriched air, and separating the resulting hydrogen sulfide laden purge gas from the liquid sulfur. The process according to the present invention is particularly advantageous in that not only is the rate of reaction for degradation of H2X to H2S and rate of removal of H25 from liquid sulfur far greater than that predicted from the sum of corresponding rates of the catalyst by itself and the degasification step by itself but also this combination significantly exceeds the rates corresponding to identical conditions with the catalyst and an inert gas such as nitrogen being employed. In fact, the use of undried, air, instead of dry nitrogen, has been observed to result in an incremental increase of 49 percent in the rate constant for conversion of H2SX to H25 and 31 percent in a rate constant descriptive of the removal of H2S. Thus, the increased rates will correspond to more rapid removal of both H2SX and H2S.

The process of this invention thus provides a simple method for catalytically promoting the degradation of H2SX and removal of H25 from liquid sulfur in a time frame consistent with contemporary commercial scale operations without introducing additional contaminant.

The process of the invention will now be described in more detail with particular reference to the accompanying drawing which is a plot of the concentration of H2SX (circles) and H25 (triangles) expressed in parts per million by weight as determined by infrared analysis as a function of reaction time at 3000F and 18.5 cc/minute sweep gas in the presence of alumina during an experimental run involving the catalytic removal of H2S and H2SX dissolved in liquid sulfur. As illustrated, approximately the first 1.75 hours involved a dry nitrogen sweep gas followed by 30 minutes of no gas flow. Then a final 1.75 hours of wet air purge was applied. During all three phases, all other parameters were held essentially identical. The apparent rate constants are presented on the appropriate portions of the curve.

As previously indicated, the specific improvement of the present invention involves the steps of contacting liquid sulfur containing dissolved H2S and H2SX with a solid catalyst while simultaneously sweeping the liquid sulfur with a purge gas containing oxygen. The novel aspect of this invention lies in the synergistic effect of the oxygen containing purge gas in combination with the solid catalyst upon the rates of conversion of H2SX to H2S and the removal of the H2S. Although the explicit mechanism involved is not fully understood, it has been empirically observed that both rate constants for the degradation step and degasification step are significantly altered when oxygen is present in the purge gas and both changes favor more rapid removal of H2S.

Since the improved process of this invention is to be performed on liquid sulfur, the limits of the acceptable temperature range will be determined by two pragmatic considerations. The- lower limit corresponds to the melting point of the highest melting form of elemental sulfur (approximately 2500 F). The upper limit corresponds to the known viscosity increase (approximately 3200 F) above which the molten sulfur is relatively unmanageable. As recognized in the art, lower temperatures favor the decomposition of H2SX, yet too low a temperature increases the risk of the sulfur solidifying. Therefore, a temperature range of about 265 to 3000 F is preferred with the range 270 to 2750F being particularly suitable for purposes of this invention.

The liquid sulfur to be processed by this invention comes from a variety of sources.

Categorically, they involve sources of elemental sulfur contaminated with both H2S and H2SX. Usually, this involves sulfur which during or after contact with H2S has experienced temperatures in excess of 300"F, promoting the formation of polysulfides. Various natural deposits are known to have both H2S and H2SX present and many types of commercial plants produce sulfur of this nature. For purposes of this invention, the sulfur that is generated in the Claus type plant is of particular interest.

The concentration range of total sulfides (H2S plus H2SX) dissolved in the sulfur is frequently quoted in the literature as being from 200 to 700 ppm H2S by weight with even higher concentrations being known.

The range of 200 ppm down to 5 ppm by weight and the kinetics of the decomposition reaction and degasification process in this range are of particular concern. Typically, a contemporary commercial scale Claus plant will produce sulfides at the upper end of this range, but present toxicity and combustion safety limits favor the lower end of the range.

Since the present improvement involves a liquid sulfur phase, a solid catalyst and a sweep gas, the basic configuration of a plant utilizing this improvement can be any of the processes known in the art involving commingling of three phases. This would include, but is not limited to, processes such as a cocurrent flow or countercurrent flow through a fixed bed or fluidized bed with or without vacuum assist and the like.

This improvement is also consistent with the processes disclosed in the previously mentioned patents including the atomizing spray technique and the baffled scrubbing tower method. The preferred commercial embodiment involves a countercurrent flow arrangement, wherein, the molten sulfur flow down through a packed catalyst bed while the oxygen containing purge gas flows up through the bed.

The preferred solid catalysts to be used in this invention are alumina or more specifically activated (porous) aluminum oxide and alumina impregnated with a cobalt-molybdenum catalyst recognized in the petroleum refining art as desulfurization catalysts. The specific form, shape and size of the solid catalyst to be used depends on the particular process into which the improvement is to be incorporated. Thus, processes such as disclosed in the above mentioned patents may advantageously use alumina as a coating or structural component, while the preferred packed bed would employ an alumina particle of about 2 to 9 mesh (U.S.

Sieve) and fluidized beds may employ a much finer powdered catalyst form. A 4 to 9 mesh size alumina particle is adequate for most commercial scale packed bed operations.

The purge gas employed in this improvement is essentially any oxygen containing gas including air, oxygen enriched air, and the like. Various other inert gases can be present including water vapor. Thus, if air is to be employed, no drying step is necessary.

In order to demonstrate the specific advantages associated with our improved process, a test using first dry nitrogen and then wet air as the stripping gas was made with all other conditions held constant. The experiment was performed in a test cell constructed from an aluminum block specifically designed for this purpose. The test cell contained internal provisions for saturation and stripping of the confined liquid sulfur with gases from an external source. It also contained internal provisions for continuous external infrared analysis of the composition of the liquid sulfur.

Basically there were two cavities within the aluminum block each performing one of the aforementioned functions. One cylindrical cavity served as an infrared cell having a path length of 150 mm with zinc selenide windows at each end. A second vertical cylindrical cavity was equipped with a gas inlet sparge at the bottom and a gas outlet at the top. Positioned between the inlet and outlet was a basket capable of holding approximately one cubic inch of catalyst such as to simulate a packed column configuration. The test block was further designed for continuous circulation of the confined liquid sulfur. The liquid sulfur being lifted vertically in the catalyst containing cavity would flow from the top of this stripping chamber to one end of the infrared cell, through the infrared cell, out the other end and return to the bottom of the chamber containing the catalyst. The temperature of the aluminum block was controlled by pair of appropriately sized and monitored electrical heating elements.

During the experimental run the entire aluminum block was placed directly in the infrared beam of a commercially available single beam IR analyser with a variable filter supplied by Wilks Scientific Corporation of South Norwalk, Connecticut under the tradename of "Miran I" System. The block was positioned such that the IR beam passed directly through the first chamber via the zinc selenide windows permitting the measurement of light absorption as a function of frequency and reaction time.

The basic operation of the experimental equipment involves loading the catalyst basket with the selected catalyst and placing it in the vertical chamber between the gas inlet and gas outlet and filling the remaining portion of the internal chambers of the test block with liquid sulfur. The desired gas is pumped through the inlet at the bottom of the catalyst chamber. As it rises up through the catalyst bed and out the top, an internal circulation of the liquid sulfur through the IR cell chamber is induced. In this manner, the concentration of H2S and H2Sx impurities in the liquid sulfur can be monitored continuously and the effect of various catalysts and purge gases on the decomposition of H2Sx and removal of H2S can be studied.

In this specific case one cubic inch, approximately twelve grams, of 0.3 cm diameter spheres of an alumina catalyst, supplied commercially under the tradename Kaiser S-201 alumina, was placed in a catalyst basket made from 16x 18 mesh aluminum screen and placed in the previously described catalyst chamber of the aluminum block. The remainder of the interior of the test block was filled with approximately 130 grams of liquid sulfur which was then intentionally contaminated to approximately 70 ppm H2S and 100 ppm H2SX by bubbling a mixture of N2 and H2S through the test cell for approximately 27 hours. A purge stream of dry nitrogen was then applied to the lower end of catalyst chamber at a controlled flow rate of 18.52 cc/min. During the first hour and forty minutes of the experiment. During this time the temperature was maintained at 300"F and the IR analyzer repeatedly scanned and recorded the infrared sDectrum from approximately 2.5 ,u to 4.5 ,u. All adsorption measurements were made using the "Miran I" System with a slit width setting of 0.25 mm, gain selector at 10x, and a time constant of 0.25 sec. At the end of 1 hour and 40 minutes the dry nitrogen flow was stopped and the test cell was maintained in a static condition for approximately 34 minutes while periodic IR scans were continued. A purge of wet air was commenced at a rate of 18.56 cc/min.

and maintained for the next hour and forty minutes, again with repeated IR scans.

Quantitative interpretation of the IR spectrums was performed using the base line technique with adsorptions at 3.9 /1 and 4.0 ,fl corresponding to H2S and H2SX respectively, similar to work reported in T.

K. Wiewiorowski and F. J. Tuoro, "The Sulfur-Hydrogen Sulfide System", The Journal of Phys. Chem., Vol. 70, No. 1, January, 1966, p. 234. Both the decomposition of H2SX to H2S and the removal of the H2S from liquid sulfur were modeled by an overall Ist order rate law corresponding to In (C/Co)=kt, where CO is the initial concentration of the species at time=o, C is the concentration at any time t > o and k is the rate constant.

The data from the above described test are presented in the drawing. It should be readily apparent from the respective portions of the curves and their associated rate constants that the use of wet air in combination with a solid alumina catalyst involves more favorable kinetics for the decomposition of polysulfides as well as more rapid removal of hydrogen sulfide than the use of dry nitrogen with the same catalyst and conditions. The practice of this specific improvement on a commercial scale will possess the advantage of greater removal of H2S in a shorter period of time.

It should also be appreciated that in the absence of a catalyst the rate constants, measured in a manner analogous to the illustrated data, are significantly smaller indicative of much slower reaction rates.

The incremental differences being observed here are associated with the choice of stripping gas beirig used in combination with the catalyst.

Further, the numerical values of the rate constants should not be interpreted as being unduly limiting for they are intended to establish relative effectiveness of the stripping gases and not absolute rates characteristic of a full scale commercial plant. Thus, for example, the rate of removal of H2S and H2SX from liquid sulfur would be expected to vary according to which particular overall process (cocurrent, countercurrent, etc.) is selected and according to such variables as residence time, catalyst surface area and the like.

However, it can be stated categorically that a solid catalyst greatly improves the hydrogen polysulfide decomposition rate and when used in combination with wet air or the like as a stripping gas, an additional significant incremental increase in the rates will result. Additionally, the rate of decomposition and degasification of this process as a whole tend to increase with a decrease in temperature and alumina with wet air seem to be more effective than the cobalt-molybdena with wet air.

WHAT WE CLAIM IS: 1. A process for catalytic degradation of hydrogen polysulfide to hydrogen sulfide in liquid sulfur and removal of hydrogen sulfide from said liquid sulfur, which comprises contacting said liquid sulfur with a solid degradation catalyst selected from alumina and alumina impregnated with a cobalt-molybdenum catalyst at a temperature fr:om 250"F to 3200 F, simultaneously purging said liquid sulfur in contact with said catalyst with a molecular oxygen-containing purge gas, and separating the resulting hydrogen sulfide laden purge gas from said liquid sulfur.

2. A process according to Claim 1 wherein said purge gas is selected from air and oxygen-enriched air.

3. A process according to Claim 1 or Claim 2 involving a countercurrent flow wherein said liquid sulfur flows down through a packed bed of said catalyst and said purge gas flows up through said packed catalyst bed.

4. A process according to any preceding claim wherein said catalyst is alumina.

5. A process according to any preceding claim wherein said temperature is from 270"F to 275"F.

6. A process according to Claim 1 or Claim 2 involving a cocurrent flow wherein said liquid sulfur and said purge gas flow up through a packed bed of said catalyst.

7. A process according to Claim 6 wherein said catalyst is alumina.

8. A process according to Claim 6 or Claim 7 wherein said temperature is from 270"F to 2750F.

9. A process for reducing the hydrogen poly-sulfide content of liquid sulfur according to Claim 1 substantially as hereinbefore described.

10. Sulfur whenever purified by a process according to any preceding claim.

**WARNING** end of DESC field may overlap start of CLMS **.

Claims (10)

**WARNING** start of CLMS field may overlap end of DESC **. than the use of dry nitrogen with the same catalyst and conditions. The practice of this specific improvement on a commercial scale will possess the advantage of greater removal of H2S in a shorter period of time. It should also be appreciated that in the absence of a catalyst the rate constants, measured in a manner analogous to the illustrated data, are significantly smaller indicative of much slower reaction rates. The incremental differences being observed here are associated with the choice of stripping gas beirig used in combination with the catalyst. Further, the numerical values of the rate constants should not be interpreted as being unduly limiting for they are intended to establish relative effectiveness of the stripping gases and not absolute rates characteristic of a full scale commercial plant. Thus, for example, the rate of removal of H2S and H2SX from liquid sulfur would be expected to vary according to which particular overall process (cocurrent, countercurrent, etc.) is selected and according to such variables as residence time, catalyst surface area and the like. However, it can be stated categorically that a solid catalyst greatly improves the hydrogen polysulfide decomposition rate and when used in combination with wet air or the like as a stripping gas, an additional significant incremental increase in the rates will result. Additionally, the rate of decomposition and degasification of this process as a whole tend to increase with a decrease in temperature and alumina with wet air seem to be more effective than the cobalt-molybdena with wet air. WHAT WE CLAIM IS:

1. A process for catalytic degradation of hydrogen polysulfide to hydrogen sulfide in liquid sulfur and removal of hydrogen sulfide from said liquid sulfur, which comprises contacting said liquid sulfur with a solid degradation catalyst selected from alumina and alumina impregnated with a cobalt-molybdenum catalyst at a temperature fr:om 250"F to 3200 F, simultaneously purging said liquid sulfur in contact with said catalyst with a molecular oxygen-containing purge gas, and separating the resulting hydrogen sulfide laden purge gas from said liquid sulfur.

2. A process according to Claim 1 wherein said purge gas is selected from air and oxygen-enriched air.

3. A process according to Claim 1 or Claim 2 involving a countercurrent flow wherein said liquid sulfur flows down through a packed bed of said catalyst and said purge gas flows up through said packed catalyst bed.

4. A process according to any preceding claim wherein said catalyst is alumina.

5. A process according to any preceding claim wherein said temperature is from 270"F to 275"F.

6. A process according to Claim 1 or Claim 2 involving a cocurrent flow wherein said liquid sulfur and said purge gas flow up through a packed bed of said catalyst.

7. A process according to Claim 6 wherein said catalyst is alumina.

8. A process according to Claim 6 or Claim 7 wherein said temperature is from 270"F to 2750F.

9. A process for reducing the hydrogen poly-sulfide content of liquid sulfur according to Claim 1 substantially as hereinbefore described.

10. Sulfur whenever purified by a process according to any preceding claim.