REMOVAL OF SULFUR COMPOUNDS FROM GAS STREAMS VIA PRECIPITATION
20200063054 ยท 2020-02-27
Assignee
Inventors
Cpc classification
B01D53/965
PERFORMING OPERATIONS; TRANSPORTING
C10L2290/542
CHEMISTRY; METALLURGY
B01D2257/306
PERFORMING OPERATIONS; TRANSPORTING
B01D53/80
PERFORMING OPERATIONS; TRANSPORTING
B01D53/96
PERFORMING OPERATIONS; TRANSPORTING
International classification
C10L3/10
CHEMISTRY; METALLURGY
Abstract
Techniques for removing sulfur compounds from a sulfur-containing gas stream can include contacting the gas with an absorption solution comprising a metal cation capable of reacting with the sulfur compound to form a metal sulphide precipitate and/or a metal mercaptide precipitate. In addition, the treatment can include controlling a concentration of the precipitates below a threshold to maintain rheological properties; subjecting the precipitate-enriched solution to vitalization; subjecting the precipitate-enriched solution to regeneration including oxidation; and/or other features to enhance the sulfur removal operations.
Claims
1. A process for removing sulfur compounds from a sulfur-containing gas stream, comprising: in an absorption stage, contacting the sulfur-containing gas stream with an absorption solution comprising a metal cation capable of reacting with the sulfur compound to form a metal sulphide precipitate and/or a metal mercaptide precipitate, thereby producing a sulfur-depleted gas stream and a precipitate-enriched solution; recycling the precipitate-enriched solution back to the absorption stage, thereby accumulating metal sulphide precipitates in the absorption solution; and controlling a concentration of the metal sulphide precipitates present in the absorption solution below a threshold in order to maintain rheological properties of the absorption solution in the absorption stage.
2.-94. (canceled)
Description
BRIEF DESCRIPTION OF DRAWINGS
[0113]
[0114]
[0115]
[0116]
[0117]
[0118]
[0119]
[0120]
[0121]
[0122]
[0123]
[0124]
[0125]
[0126]
DETAILED DESCRIPTION
[0127] Various implementations of removing sulfur compounds from gas streams will be described. While various implementations are described with respect to removing H.sub.2S in particular, it should be understood that various techniques described herein can be used or adapted for removing other sulfur compounds, such as COS, mercaptans, and the like.
[0128] In general, the H.sub.2S removal includes an absorption stage in which the H.sub.2S-containing gas is contacted with a solution containing metal cations (e.g., such as Cu.sup.2+) in order to form metal sulphide precipitates (CuS) that are carried with the solution. A precipitate-loaded solution and an H.sub.2S-depleted gas are produced and withdrawn from the absorption stage. The precipitate-loaded solution can then be subjected to various downstream treatment operations, such as vitalization and/or regeneration, such that at least a portion of the solution can be recycled back to the absorption stage and various streams derived from the precipitate-loaded solution, containing sulfur- and metal-containing compounds, can be produced and removed.
[0129] Various implementations that are described herein may be referred to as the Vitrisol process, and enable the removal of H.sub.2S from gas streams (e.g., natural gas, biogas, process gas, etc.). In an absorption stage, H.sub.2S is removed via a precipitation reaction with a metal ion in an acidic, aqueous solution forming a metal sulphide, thus depleting the active compound (metal cations) in the absorption solution. A vitalization stage can be incorporated into the process as well as a regeneration stage that can enable the oxidative regeneration of the precipitate-loaded solution. Based on experimental work and testing, the Vitrisol process can be operated in a variety of manners. Leading factors determining the process design are the quantity of sulphur to be removed, as well as the composition and capacity of the gas to be purified. In addition, special conditions and design criteria imposed or desired by operators can also have an influence on process design.
[0130] Absorption Stage Implementations
[0131] Referring to
[0132] In some implementations, the process design is particularly advantageous and suited for the removal of low amounts of H.sub.2S from gas streams, in which the absorption solution can operate in a stand-alone fashion for several weeks to several months. One advantage of the process is its ability to provide high removal efficiencies, even when most of the active compound in the solution (e.g., Cu.sup.2+) has been converted into CuS. The process can still provide consistent H.sub.2S removal efficiency during operation, with little to no decline in H.sub.2S removal efficiency. In some implementations, for every mole of H.sub.2S removed as a metal sulphide an equimolair amount of H.sub.2SO.sub.4 is formed. For the situation that pH control is desired, e.g. caustic and/or CaCO.sub.3 can be added to the solution to counteract the increase in acidity of the solution.
[0133] It should be noted that the H.sub.2S absorption process can be carried out in various types of single or multi-stage gas-liquid contactors, such as a packed column, a tray column, a plate column, a bubble column, and so on. Each of the contactors of the absorption stage can be operated either co-currently, or counter-currently. In the absorption contactor, a gas flow can be contacted with the aqueous solution containing a pre-selected metal cation, such as Cu.sup.2+. The metal cation reacts with H.sub.2S forming a metal sulphide precipitate, such as CuS, thus depleting the H.sub.2S from the gas flow. The metal cation can be provided by adding a metal salt, which can be selected from the group consisting of metal nitrates, sulphates, phosphates, sulphites, nitrites, chlorides, bromides, iodides, fluorides, pyrophosphates and perchlorates. In addition, in some implementations the absorption process is operated in the pH range of 1.0<pH<14, and preferably 1.0<pH<7.7. The boundaries ensure that substantially (and preferably exclusively) metal sulphide is precipitated, and that the metal cation will tend not to form precipitates with other compounds that could be present in both the gas phase and liquid phase (e.g., metal carbonates, metal oxides, etc.). As mentioned above, preferably Cu.sup.2+ ions are used to precipitate H.sub.2S in the absorption liquid, although other metal ions may be used. In some implementations, metal cations could be used that either convert H.sub.2S to metal sulphides, and/or directly convert H.sub.2S to sulphur, i.e. no metal sulphide precipitate is formed but directly sulphur. Due to the possibility of operating the absorption process at lower pHs, the co-absorption of other acid gases, e.g. CO.sub.2, will be minimized or even not occur. Therefore, the process can be designed and operated as such that H.sub.2S is selectively removed.
[0134]
[0135] Vitalization Stage Implementations
[0136] Referring to
[0137] Referring to
[0138] In the process implementation of
[0139] Implementations of the process that incorporate a vitalization stage are advantageously suited for the removal of higher amounts of H.sub.2S from gas streams compared to batch-wise stand-alone operation illustrated in
[0140] Regeneration Stage Implementations
[0141] The process can also include a regeneration stage that enables the regeneration of the precipitate-loaded solution, for example by oxidation.
[0142] Referring to
[0143]
[0144] Still referring to
[0145] Referring now to
[0146] Still referring to
[0147] Referring to
[0148] Referring now to
[0149] Still referring to
[0150] Additional discussion of the Vitrisol process with a regeneration step will now be presented. The regeneration with O.sub.2 downstream of the absorption generally refers to oxidizing the metal sulphide precipitate formed in the absorber, and returning the active compound (e.g., the metal ion) to the absorption solution. Subsequently, the regenerated absorption solution can be returned to the absorber. To be able to oxidize the metal sulphide, a second metal ion is placed in the Vitrisol solution. This second metal ion is oxidized by O.sub.2, and in turn the second metal ion in its higher valence state is able to oxidize the metal sulphide to release the active metal ion back in solution and produce solid sulphur (which can also be referred to as the regeneration of the Vitrisol liquid).
[0151] Alternatively, the metal sulphide formed in the absorber can be oxidized through electrochemical oxidation. To be able to oxidize the metal sulphide, a second metal ion is placed in the Vitrisol solution. This metal ion is oxidized in the anode compartment of an electrochemical cell. The respective ion, in its higher valence state, in turn oxidizes the metal sulphide. In the electrochemical cell, the anode compartment is separated from the cathode compartment by a membrane. Liquids can flow through each compartment either co-currently or counter-currently. A function of the membrane is to transfer protons and to retain solids and multivalent ions in solution. At the cathode hydrogen gas is formed.
[0152] In general, the Vitrisol process with O.sub.2 regeneration can be operated in at least two different manners with respect to CuS conversion. Subsequently, the process can be operated with or without sulphur removal from the Vitrisol liquid and in at least two different manners when sulphur is extracted with a solvent. Certain possible implementations will be further described below.
[0153] With respect to CuS conversion, operation can be conducted at (I) at or close to 100% CuS conversion in the oxidation reactor(s), or (II) below 100% CuS conversion in the oxidation reactor(s). In some implementations, it is noted that Fe.sup.2+ is the preferred second metal ion in solution.
[0154] In addition, it has been noted that the presence of the first metal ion (e.g., Cu.sup.2+) can enhance oxidation of the second metal ion (e.g., Fe.sup.2+) even when the first metal ion is present in quite low concentrations. In this regard, multiple types of first metal ions can be provided (e.g., a mixture of at least two of Cu2+, Zn2+, Ag2+, or others) such that one of such ions is provided to maximize the enhanced oxidation of the second metal ion while the other of such ions is provided as per the primary function described herein of forming the metal sulphide and/or mercaptide precipitates. In some scenarios, the oxidation-enhancing metal ion may be a more expensive material and is used in smaller quantities compared to the ion forming the majority of the sulphide/mercaptide precipitates.
[0155] For scenarios (I) with at or close to 100% CuS conversion in the oxidation reactor(s), the process may include one or more of the following aspects: [0156] a. After the oxidation step, return the Vitrisol liquid with sulphur to the absorption column (batchwise operation of the Vitrisol process with oxidation of CuS). [0157] b. After the oxidation step, separate sulphur from the Vitrisol liquid via a suitable separation step (e.g. filtration, centrifugation, sedimentation). Afterwards, the Vitrisol liquid is returned to the absorber. [0158] A make-up stream of Vitrisol is required either continuously or continually to counteract the loss of Vitrisol in the sulphur cake. [0159] c. After the oxidation step, separate sulphur from the Vitrisol liquid via a suitable separation step (e.g. filtration, centrifugation, sedimentation). Afterwards, the Vitrisol liquid is returned to the absorber, and the sulphur cake is extracted with a suitable nonpolar sulphur solvent (named S/L extraction). Subsequently, the sulphur is crystallized out in a crystallizer. [0160] i. The process can be operated with a separation step before the crystallizer (e.g. settler) to separate the Vitrisol from the sulphur solvent before the crystallizer. In this manner the Vitrisol slip in the sulphur cake can be returned to the absorber and Vitrisol liquid losses are minimized. [0161] d. After the oxidation step, extract sulphur from the Vitrisol liquid by contacting a suitable nonpolar sulphur solvent with the Vitrisol liquid containing sulphur (named L/L extraction). Optionally either candle filters could be placed in the extraction step, or a candle filter operation is placed after the extraction step. Subsequently [0162] i. Separate the two liquid streams (e.g. settler); the Vitrisol liquid is returned to the absorber, the sulphur solvent is processed downstream in a crystallizer, and afterwards sulphur is separated (e.g. filtration, centrifugation, sedimentation). A make-up stream is required for the sulphur solvent due to losses in the sulphur cake. [0163] The sulphur cake can be further processed to pure sulphur with purities exceeding 99.9% in e.g. a sulphur smelter. The sulphur solvent can be fully reclaimed, making a make-up stream of sulphur solvent redundant. [0164] In case the nonpolar sulphur solvent dissolves in the Vitrisol liquid, and is stripped in the absorber, the gas exiting the absorber could be treated in an adsorber to remove the nonpolar sulphur solvent vapors. Optionally the nonpolar sulphur solvent adsorbed could be reclaimed. [0165] ii. Both streams are processed downstream in a crystallizer and afterwards sulphur is separated (e.g. filtration, centrifugation, sedimentation). Make-up streams are required for both the Vitrisol liquid and the sulphur solvent. Afterwards the two liquid streams are separated (e.g. mixer settler); the Vitrisol liquid is returned to the absorber, the sulphur solvent is returned to the sulphur extraction operation. [0166] The sulphur cake can be further processed to a sulphur higher in purity in e.g. a sulphur smelter. The sulphur solvent and water can be fully reclaimed, making a make-up stream of sulphur solvent redundant. Salts however could exit as impurity. [0167] e. Combine the oxidation of CuS with O.sub.2 and L/L sulphur extraction in one oxidizer/extraction step. Subsequently either point d)i, or d)ii could be performed. [0168] i. The gas exiting the oxidizer/extraction step may contain nonpolar sulphur solvent vapors. These nonpolar sulphur solvent losses could be reclaimed by treating the gas in an adsorber. [0169] ii. In case the nonpolar sulphur solvent dissolves in the Vitrisol liquid, and is stripped in the absorber, the gas exiting the absorber could be treated in an adsorber to reclaim the nonpolar sulphur solvent.
[0170] For scenarios (II) with at or close to 100% CuS conversion in the oxidation reactor(s), the process may include one or more of the following aspects: [0171] a. During the oxidation of CuS in the presence of a second metal ion and oxygen both the CuS and the second metal ion oxidize (e.g., Fe.sup.2+ to Fe.sup.3+). By operating the oxidation step as such that enough Fe.sup.3+ is formed to convert the remainder of CuS, no oxygen is required anymore to oxidize the CuS. From experimental work was observed that the order of reaction in Fe.sup.3+, for the oxidation of CuS by Fe.sup.3+, is zero when enough Fe.sup.3+ is present for full conversion of CuS. Therefore at this point, the further oxidation of CuS can be performed in a subsequent step in the absence of oxygen, and can be operated in two manners: [0172] i. The process can be operated to 100% CuS conversion in a subsequent step [0173] ii. The process can be operated <100% CuS conversion in a subsequent step [0174] iii. The process operated till 100% conversion of CuS in a subsequent oxidizer/extraction step in the absence of oxygen can be operated in different manners. [0175] According to (I)a, (I)b, (I)c or (I)d. [0176] Combine the oxidation of CuS in the absence of O.sub.2 with L/L extraction in one step. Subsequently either point (I)d.i or (I)d.ii could be performed. [0177] Either safeguard filter elements could be installed in the oxidizer/extraction step, or a safeguard candle filter operation subsequent to the oxidizer/extraction step could be installed, to prevent possible slip of CuS and/or sulphur. Further processing could be according to either point (I)d.i or (I)d.ii. [0178] The process can be operated by two liquid streams exiting the oxidizer: a Vitrisol stream exiting the oxidizer and entering the absorber (exit optionally through a candle filter operation), and a split stream exiting the oxidizer to the oxidizer/extraction step. Vitrisol liquid returning to the absorber will contain Fe.sup.3+ which could react with H.sub.2S in the absorber forming sulphur. Safeguard filter elements could be installed in the oxidizer/extraction step, or a safeguard candle filter operation subsequent to the oxidizer/extraction step could be installed, to prevent possible slip of CuS and/or sulphur. Afterwards, the nonpolar sulphur solvent phase is processed according to point (I)d.i. Point (I)d.i.2 will not occur as the nonpolar sulphur solvent is stripped in the oxidizer. Gas exiting the oxidizer could be treated in an adsorber to remove the nonpolar sulphur solvent vapors. Optionally the nonpolar sulphur solvent adsorbed could be reclaimed. [0179] iv. The process operated till <100% conversion of CuS in a subsequent step in the absence of oxygen can be operated in different manners. [0180] According to point (I)a, however the Vitrisol liquid contains both sulphur and CuS. [0181] According to point (I)b, however sulphur and CuS are separated. [0182] According to point (I)c, however sulphur and CuS are separated and the CuS can either be [0183] a. Separated from the Vitrisol/nonpolar solvent before the crystallization step and either returned to the CuS oxidation vessel in the absence of oxygen, or removed from the process. [0184] b. Processed with the liquid stream in the crystallizer, and afterwards be separated together with the sulphur from the liquid stream. [0185] According to point (I)d, however [0186] a. Regarding point (I)d.i: CuS present can separate over both liquid phases. Consequently part of the CuS is returned to the absorber, and part of the CuS is processed downstream with/without a mixer settler (to reclaim CuS) in the crystallizer and sulphur removal step. [0187] b. Regarding point (I)d.ii: the sulphur cake will contain CuS when CuS is not separated before the sulphur removal step. [0188] c. After the last CuS oxidation step any CuS remainder is separated (e.g. filtration, centrifugation, sedimentation) and returned to either the CuS oxidation step in the absence of oxygen, or the CuS oxidation step in the presence of oxygen. Further processing can be performed according to points (I)d.i and (I)d.ii. [0189] Combine the oxidation of CuS in the absence of O.sub.2 with UL extraction as well as the separation of CuS in one step, by e.g. installing a candle filter in the oxidizer/extraction step in the absence of oxygen. Subsequent separation of CuS is made redundant, however probably a backflush is required periodically to remove the CuS cake layer from the filter. [0190] According to point (II).a.i.4.
[0191] The Vitrisol process with H.sub.2O.sub.2 regeneration can be operated in a variety of different manners, similar to those described in the Vitrisol process with O.sub.2 regeneration. However, oxygen and the second metal ion are not required in this H.sub.2O.sub.2 regeneration process, and therefore the process can be operated in one oxidation step. Extraction can be performed through either S/L or L/L extraction. UL extraction can be performed either in the oxidizer, or in a subsequent step. One relevant factor is the water content of peroxide; generally the source of H.sub.2O.sub.2 is an aqueous solution of H.sub.2O.sub.2, therefore the water balance is a relevant parameter to control.
[0192] Optional Implementations and Features
[0193] In some scenarios, the process may have one or more of the additional features and/or steps listed below: [0194] The CuS formed in the absorber is a highly porous material with a particle size of 1<D.sub.P<20 m. [0195] Choose the second polyvalent metal ion (required for the oxidation of the metal sulphide) as such that the oxidation of this ion to its higher valence state is enhanced by the first metal ion in solution, responsible for the removal of H.sub.2S, e.g. Cu.sup.2+ is the metal ion in solution for the removal of H.sub.2S which enhances the oxidation rate of Fe.sup.2+ (the second polyvalent metal ion) with oxygen (Vitrisol with O.sub.2 regeneration). [0196] The second metal salt, required for the oxidation of the metal sulphide, can be selected from the group consisting of metal nitrates, sulphates, phosphates, sulphites, nitrites, chlorides, bromides, iodides, fluorides, pyrophosphates and perchlorates. [0197] For the situation in which Cu.sup.2+ is the metal ion in solution for the removal of H.sub.2S, and Fe.sup.2+ is second polyvalent metal ion in solution, in order to increase the oxidation rate of Fe.sup.2+ with oxygen it is preferred to increase the Fe.sup.2+ concentration instead of the Cu.sup.2+ concentration as Fe.sup.2+ enhances the Fe.sup.2+ oxidation rate more substantially compared to Cu.sup.2+ (Vitrisol with O.sub.2 regeneration). [0198] For the situation of Cu.sup.2+, Fe.sup.2+ and CuS, all the sulphur formed during oxidation with oxygen is formed as a solid, no sulphur dissolves in the Vitrisol liquid (Vitrisol with O.sub.2 regeneration). Therefore, it can be decided to either extract the sulphur by UL extraction, or by first separating the solids and contacting the solids with a suitable nonpolar sulphur solvent. Whichever operation is dependent on the case/mass balance. [0199] For the situation Cu.sup.2+, Fe.sup.2+, CuS and oxygen, the major part of the CuS oxidation (when sufficient k.sub.ia is ensured) follows the initial oxidation rate of Fe.sup.2+ under the same experimental settings, when no Fe.sup.3+ is present (Vitrisol with O.sub.2 regeneration). [0200] When oxidizing CuS with Fe.sup.3+ under a nitrogen blanket, 100% conversion can be achieved. [0201] The H.sub.2S absorption step should be operated as such that preferably a liquid is obtained with CuS solids content of 0<wt %<50, more preferably 0<wt %<10 and more preferably 0<wt %<5. Too high solids content leads to dramatic changes in the rheological behaviour of the Vitrisol liquid. [0202] The absorption process can be operated either cocurrently or countercurrently. Due to the nature of the Cu.sup.2++H.sub.2S reaction being an instantaneous precipitation reaction at the G/L interface, the rate of transfer of H.sub.2S in the liquid is dependent on gas phase mass transfer. [0203] The CuS solid structure, i.e. the particle size, as well as the porosity, provides a substantial solid surface readily available for oxidation by Fe.sup.3+. Therefore high CuS oxidation rates are achieved (Vitrisol with O.sub.2 regeneration). [0204] Fe.sup.3+ inhibits the oxidation of Fe.sup.2+ with oxygen in the presence of Cu.sup.2+ (Vitrisol with O.sub.2 regeneration). [0205] The CuS oxidation reaction in the presence of Fe.sup.2+, Fe.sup.3+, Cu.sup.2+ and oxygen is preferably carried out in the temperature range of 5 to 155 C., more preferably 50 to 120 C. and more preferably 70 to 95 C. (Vitrisol with O.sub.2 regeneration). [0206] Increasing the oxygen partial pressure increases the oxidation rate of CuS. [0207] The CuS oxidation reaction in the presence of Fe.sup.2+, Fe.sup.3+, Cu.sup.2+ and oxygen is preferably carried out in the oxygen partial pressure range of 0.01 to 50.4 bar, and more preferably in the oxygen partial pressure range of 0.1 to 20 bar (Vitrisol with O.sub.2 regeneration). [0208] Liquid oxygen, air, compressed air, and O.sub.2 enriched air can be used to oxidize Fe.sup.2+ and subsequently CuS with Fe.sup.3+ (Vitrisol with O.sub.2 regeneration). [0209] CuS is hydrophobic in nature. However, the CuS produced in the Vitrisol absorption process is hydrophilic of character (also Broekhuis, 1992). Additionally, the density of CuS is lower than that of H.sub.2O due to the porous nature of the solids. However, CuS formed in the absorber forms a (semi-) suspension with Vitrisol. An experiment was performed with CuS from the Vitrisol absorber, that was washed that many times that no metal salts and H.sub.2SO.sub.4 remained in the pores. This CuS did not form a suspension with Vitrisol, it floated on the G/L interface. Therefore the conclusion can be made that salts in the CuS pores create the hydrophilic character of the CuS particles. Consequently it is required to assure that ions are always present at the CuS surface (including the pores). Therefore it is advised that the CuS particles are always kept in the Vitrisol liquid, and that the CuS should be created by contacting H.sub.2S with Cu2+ ions in and aqueous liquid. [0210] The suitable sulphur, and/or sulphur compound, solvent is at least one selected from the (halogenated) aliphatic and/or (halogenated) aromatic compounds group consisting of e.g. naphthalene, gasoline, diesel, N,N-dimethylaniline, benzene, toluene, p-xylene, m-xylene, o-xylene, cyclohexane, ethyl cyclohexane, dimethyl cyclohexane, halogenated alkanes (e.g. tetrachloroethane, chloroform), halogenated alkenes, n-alkanes, n-alkenes, branched alkanes, branched alkenes and carbon disulphide, and mixtures thereof. [0211] The sulphur solvent can possibly partly dissolve in the Vitrisol liquid. Therefore a downstream removal step for the dissolved sulphur solvent could be required. [0212] A unique feature of the H.sub.2S absorption process, when using Cu.sup.2+ for the removal of H.sub.2S, is that deep removal can be obtained, even at low pH, though H.sub.2S is an acid gas. Additionally, adding a second polyvalent metal ion does not negatively influence the removal of H.sub.2S by Cu.sup.2+. [0213] The oxidation of CuS results in the formation of solid sulphur. However, at T=90 C. the decrease in oxidation rate does not appear to be resulting from mass transfer limitations due to the formation of a sulphur barrier on the CuS, but because the surface readily available for reaction decreases. [0214] The Vitrisol process with regeneration (either oxygen or H.sub.2O.sub.2) is flexible with respect to CuS conversion in the oxidation steps (as explained earlier in this document). [0215] Complete CuS conversion can be obtained in the oxidation process of CuS (Vitrisol with O.sub.2 regeneration). [0216] The Vitrisol process with regeneration (either oxygen or H.sub.2O.sub.2) can be operated with heat integration, i.e. the Vitrisol liquid exiting the oxidation steps can heat up the Vitrisol liquid entering the oxidation steps. [0217] The overall oxidation reaction of H.sub.2S is given by:
2H.sub.2S+O.sub.2.fwdarw.2H.sub.2O+S (Vitrisol with O.sub.2 regeneration)
Or
H.sub.2S+H.sub.2O.sub.2.fwdarw.2H.sub.2O+S (Vitrisol with H.sub.2O.sub.2 regeneration)
[0218] For every mole of sulphur formed, two moles of water are formed. Therefore, depending on the case and resulting process design, the water content in the process can either increase, decrease or remain equal (due to e.g. evaporation of water in the CuS).
EXPERIMENTATION & EXAMPLES
Experimentation 1: Vitrisol Absorber
[0219] The following examples of the cocurrent downflow Vitrisol absorber, operated batchwise, illustrate the influence of the formation of CuS on the behaviour of the Vitrisol absorption liquid, the first order behaviour with respect to H.sub.2S removal and the observed characteristics of the CuS particles formed.
[0220] Experimental Setup Description
[0221] H.sub.2S absorption experiments were executed in a cocurrent downflow Vitrisol pilot absorber at a biogas production site. The absorber included a packed column placed on top of a Vitrisol liquid storage vessel. The packing consisted of Pall rings. Gas- and liquid-samples were obtained at 4 points divided over the length of the packing. The inlet gas was sampled. A pump returned the absorption liquid from the storage vessel back to the top of the packed column. A biogas stream entered the column in the top and exited the column in the liquid storage section. The Vitrisol absorption liquid was an aqueous solution containing copper sulphate, ferrous sulphate and sulphuric acid.
[0222] The absorber was operated with superficial gas velocities up to 1.2 m/s and superficial liquid velocities up to 0.012 m/s. H.sub.2S concentrations in the gas phase up to 6000 ppmV were processed. The absorber setup was operated at atmospheric pressure.
[0223] Influence of CuS on the Absorption Liquid Behaviour
[0224] It was observed that foaming occurred at CuS concentrations of the Vitrisol absorption liquid, exceeding 3 wt % to 5 wt %. Because of the rheological nature of the liquid, the absorption liquid could not be easily pumped. This halted the absorption process as no absorption liquid entered the top of the packed column.
[0225] First Order Behaviour with Respect to H.sub.2S Removal
[0226]
[0227] Characteristics of CuS Particles Formed
[0228] The density of solid CuS is much higher than that of water. Based on this fact it was expected that CuS, formed during the absorption of H.sub.2S, would sink. Furthermore, CuS is hydrophobic of nature. Therefore it was expected that, depending on the CuS particle size, a suspension would not be formed.
[0229] However, it was observed that the absorption liquid and CuS formed a homogeneous suspension when in motion. When left in a motionless state after motion, part of the CuS would sink slowly, and part of the CuS would float on the G/L interface. This behaviour is explained by the following analyses and observations.
[0230] The porosity of CuS was analyzed. The porosity of the CuS particles formed during H.sub.2S absorption appeared to have a porosity higher than 50%. The density of the CuS formed during the absorption process was determined to be lower than the density of water.
[0231] The particle size distribution was measured. The particle size distribution appeared to be ranging within 0.1 m to 50 m.
[0232] CuS obtained from the pilot absorber was separated and washed many times such that no Vitrisol liquid was present in the CuS. Afterwards, the CuS was dried. It was observed that the CuS, free of Vitrisol liquid, did not form a suspension with either water or Vitrisol liquid; i.e. the CuS remained afloat on the G/L interface. Therefore, it can be concluded that Vitrisol liquid, within the pores of the CuS solids, provides the hydrophilic character of the CuS formed in the H.sub.2S absorption process.
[0233] Due to high porosity and small particle size, a large CuS area can be readily available for oxidation by Fe.sup.3+. This is a reason behind high leaching rates of CuS. Normally, in the copper ore industry, CuS ore is crushed to smaller particle size to reduce leaching times. However, leaching rates are still smaller compared to processes described herein resulting in higher leaching times. To enhance leaching rates, anions such as Cl and/or Br ions can optionally be added. However, such enhancements are not required for implementations of the present process, as CuS particles produced in such implementations have displayed superior characteristics with respect to leaching.
Experimentation 2: CuS Sedimentation
[0234] The following example with respect to sedimentation of CuS illustrates behaviour, as well as sedimentation velocities, of CuS in Vitrisol.
[0235] Experimental Setup Description
[0236] Sedimentation experiments were performed in a graduated cylinder. A Vitrisol solution containing CuS was added to the graduated cylinder. Vitrisol solution containing CuS was filtrated to obtain solids-free Vitrisol. After agitation, the sedimentation behaviour, as well as sedimentation speeds, were observed.
[0237] Sedimentation of End-of-Life Vitrisol
[0238] It was observed that end-of-life Vitrisol, i.e. Vitrisol containing a maximum amount of CuS at which a clear change in rheological behaviour of the liquid was observed, did not settle. The wt % of CuS in this end-of-life Vitrisol was 3 wt %.
[0239] Sedimentation of Diluted End-of-Life Vitrisol
[0240] In a subsequent experiment, the end-of-life Vitrisol was diluted with the same Vitrisol liquid, however being free of solids. The diluted Vitrisol contained 1 volume part end-of-life Vitrisol and 3 volume parts of solids-free Vitrisol.
[0241] It was observed that CuS, present in diluted end-of-life Vitrisol, settled towards the bottom. However, the CuS did not settle on the bottom. Rather, a bottom phase, with CuS in suspension, was formed.
[0242] It can be concluded that, after dissolution, the suspension formed within the diluted end-of-life Vitrisol, after sufficient sedimentation time, nears the original solids content of the end-of-life Vitrisol. The solids content of the suspension, formed within the diluted end-of-life Vitrisol, was 2.7 wt %.
[0243] Furthermore, initially the sedimentation velocity increases towards a steady sedimentation velocity. After a certain solids content was reached, the sedimentation velocity decreased to zero.
[0244] It has been posited that changes in rheological behavior and the maximum wt % in which the CuS stays in suspension without agitation are linked. It should be noted that such rheological behavior, particle properties, and suspension properties can be used to aid in the design and operation of various units, such as the absorption unit, in order to mitigate potential problems associated with such solid particles. For example, problems associated with particles that tend to float can be mitigated by using a packed column in which the liquid phase flows as a film over the packing and thus prevents floating issues that could be problematic in other types of contactors in which the liquid is the continuous phase. In addition, counter-current gas-liquid flow can be used to mitigate high rates of solids formation that could occur in co-current scenarios. In addition, appropriate agitation can be provided for liquid-continuous contactors so as to keep the particles in suspension, if desired, and the agitation can be configured and operated based on whether the particles tend to float, for example.
Experimentation 3: CuS Separation from Absorption Liquid
[0245] The following examples with respect to the separation of CuS from the absorption liquid via filtration (candle filters) and centrifugation (decanter) will illustrate how CuS is separated based on two different processes.
[0246] Experimental Setup Description
[0247] Filtration experiments were executed in a pilot candle filter setup. The setup was operated such that the absorption liquid suspension, containing CuS formed by removing H.sub.2S in the pilot absorber, was pumped to a candle filter housing. In the candle filter housing was placed a candle filter. With the pump the pressure difference over the candle filter could be adjusted. The filtrate exited the candle filter housing through the candle filter. CuS particles formed a cake on the candle filter cloth. The cake was discharged by blow back of air. The dry matter content of the filter cake was determined by drying overnight at elevated temperatures.
[0248] Centrifugation experiments were executed in a pilot decanter setup. The setup was operated as such that absorption liquid suspension, containing CuS formed by removing H.sub.2S in the pilot absorber, was pumped to the decanter at varying velocities. The decanter was operated at different centrifugal speeds to investigate the solids content of the cake exiting the decanter. The dry matter content of the filter cake was determined by drying overnight at elevated temperatures.
[0249] Filtration
[0250] It was observed that CuS filter cakes could be obtained with a dry matter content ranging from 25 wt % to 40 wt %.
[0251] Centrifugation
[0252] It was observed that CuS decanter cakes could be obtained with a dry matter content ranging from 25 wt % to 50 wt %.
Experimentation 4: Centrifugation of Filter Cake
[0253] Filter cake obtained from filtration experiments described in Experimentation 3 was exposed to varying levels of gravitational force in a centrifuge to further increase the dry matter content of the cake.
[0254] Experimental Setup Description
[0255] A filter tube was placed inside a metal tube that fitted in a centrifuge. A filter paper was place in the filter tube, and afterwards CuS cake that was obtained from filtration experiments was placed in the filter tube on the filter paper. The filled metal tubes were exposed to varying rotational speeds inside the centrifuge, and consequently to varying gravitational forces. Afterwards, the mass of the cake, as well as Vitrisol liquid removed from the cake, were determined. Afterwards, the dry matter content of the CuS cake was determined by drying overnight.
[0256] Effect of Gravitational Force on CuS Filter Cake
[0257] A CuS filter cake with 32 wt % dry matter content was exposed to varying rotational speeds in a centrifuge. Table 1 shows the effect of the varying rotational speeds on Vitrisol liquid removal from the CuS cake and dry matter content.
TABLE-US-00001 TABLE 1 The effect of gravitational force on CuS filter cake Rotational Vitrisol removed from speed G force dry matter content initial cake [rpm] [G] [wt % of resulting cake] [wt % of initial cake] 5043 2417 41.6 31.6 10118 9729 54.3 48.0 20000 38012 54.7 58.0
[0258] It can be concluded from Table 1 that the dry matter content of the CuS cake is not linear with the gravitational force applied to the samples. Moreover, whereas the dry matter content is not varied much at higher rotational speeds, still a relatively large amount of Vitrisol liquid is removed from the cake.
[0259] An explanation of this phenomena is that the Vitrisol liquid itself contains a large amount of salts. Therefore the dry matter content of the Vitrisol liquid is relatively high (when water is evaporated). Consequently, the dry matter content of the resulting cake is not increased more pronounced at higher rotational speeds. However, as additional Vitrisol liquid is removed at higher rotational speeds, the content of CuS, of the dry matter content, increases.
Experimentation 5: Fe.SUP.2+ Oxidation
[0260] Batchwise Fe.sup.2+ oxidation experiments were performed with and without Cu.sup.2+ to evaluate the Fe.sup.2+ oxidation behaviour and factors influencing the oxidation.
[0261] Experimental Setup Description
[0262] Fe.sup.2+ oxidation experiments were executed in a glass autoclave. The reactor was operated batchwise with regard to the liquid phase, and continuous with regard to the gas phase. Nitrogen could be fed in order to be able to remove oxygen from the setup. During the experiment, air was supplied to the reactor through a mass flow controller. A water saturator was used to humidify the gas phase and maintain the water balance. The pressure in the reactor was controlled with a back-pressure controller. A pressure relief valve was attached to the reactor as a safety precaution. The reactor temperature was regulated by a Julabo heater. A Bchi water bath heated the water saturator to the reactor temperature. A Teflon gas entrainment stirrer was used to obtain maximum gas dispersion in the liquid phase. The stirrer was operated with a magnetic drive. Glass baffles were introduced in the reactor to promote mixing. The steel interior of the reactor was protected by Teflon. Sampling was performed through Teflon tubing and Teflon valves. Cuvettes were used to collect samples. A UV-Vis spectrophotometer was used to analyze sample compositions. Temperature in the reactor was monitored with a Teflon-protected PT-100, temperature in the water saturator was registered with a regular PT-100. The pressure inside the experimental setup was measured with a pressure sensor.
[0263] Concentration Profiles
[0264]
[0265] This assumption was verified experimentally:
[0266] Experimentation 6: CuS Oxidation
[0267] A variety of experiments were performed to investigate the oxidation behaviour of CuS. In the experiments, CuS was used obtained from the Vitrisol pilot absorber described in Experimentation 1. One type of experiment involved the batchwise oxidation of CuS in the presence of Fe.sup.2+ and O.sub.2, the gas phase being operated continuously. This type of experiments was executed to investigate the overall oxidation reaction of both Fe.sup.2+ with O.sub.2 and subsequently CuS with Fe.sup.3+, to simulate the oxidizer operation in a continuous process. Additionally, the effect of turning off the gas flow during an oxidation experiment before 100% CuS conversion, and adding a nonpolar sulphur solvent to the reactor, was investigated. It was determined whether this influenced the CuS oxidation behaviour. Another type of experiments involved determining whether the formed sulphur, after oxidation of CuS, was in a (partially) dissolved or (partially) solid state.
[0268] Experimental Setup Description
[0269] CuS oxidation experiments were executed in the experimental setup described in Experimentation 3: Fe.sup.2+ oxidation. However, a different method of sampling was performed. Before sampling, the stirrer was turned off and the pressure relieved. First a syringe was used to obtain a sample, after which the sample was pushed through a filter to obtain a liquid sample without solids. A UV-Vis spectrophotometer was used to analyze sample compositions.
[0270] Sulphur formed, after fully converting CuS, was separated by filtration. Afterwards, both the solid phase and the liquid phase were separately extracted with a nonpolar solvent to dissolve sulphur. Afterwards, the nonpolar solvent was left to evaporate on a petri disk. If sulphur would be present, it would crystallize out.
[0271] CuS Conversion Profiles
[0272] It was observed during the CuS oxidation experiments that both the Cu.sup.2+, as well as the Fe.sup.3+ concentration, increased. Therefore one oxidation experiment was performed by turning off the gas flow at that point that enough Fe.sup.3+ was present to fully convert the remainder of the CuS. P-xylene was added to the reactor to dissolve sulphur when the gas flow was turned off.
[0273]
[0274] Sulphur Extraction
[0275] It was observed that no sulphur was extracted from the liquid phase. All the sulphur was extracted from the solid phase. Therefore it can be concluded that no sulphur dissolves in the liquid phase.
[0276] Experimentation 7: Electrochemical Oxidation of CuS
[0277] A variety of experiments were performed to investigate the indirect electrochemical oxidation behaviour of CuS. Copper sulphide cannot be oxidized directly; instead Fe.sup.2+ is electrochemically oxidized to Fe.sup.3+ at the anode, and subsequently Fe.sup.3+ oxidizes the metal sulphide. Protons are reduced at the cathode forming hydrogen. E.g. copper plating of the cathode, i.e. Cu.sup.2+ reduction at the cathode, is prevented by use of a suitable membrane (e.g. a bipolar membrane, a nanofiltration membrane or an ion-selective membrane).
[0278] Experimental Setup Description
[0279] Electrochemical oxidation experiments were performed in an electrochemical cell. The electrochemical cell consisted of two compartments, i.e. an anode and a cathode compartment, which are separated by a membrane. Liquids can flow through each compartment either cocurrently or countercurrently. The function of the membrane is to transfer protons (H.sub.3O.sup.+) and to retain multivalent ions in solution (Fe.sup.2+, Fe.sup.3+, Cu.sup.2+ and SO.sub.4.sup.2) and solids (CuS, S). Cation exchange membranes, anion exchange membranes and bipolar membranes were tested.
[0280] The electrochemical cell, or electrodialysis cell, was developed and will be described here-below. The membrane area was slightly larger than the electrode area. A spacer was placed between the membrane and the electrode in both the anode, as well as the cathode compartment. Both the anode and cathode electrodes were graphite electrodes. The electrodialysis cell was operated countercurrently with two separate pumps. A bench power supply was used to apply voltage over the anode and cathode electrodes. A UV-Vis spectrophotometer was used to determine solution compositions.
[0281] The anode solution consisted of a ferrous sulphate solution containing sulphuric acid. The cathode solution either contained demineralized water, of a sulphuric acid solution, depending on the membrane. When experiments were performed with ion exchange membranes a sulphuric acid solution was required to counteract the osmotic pressure over the membrane. In case of a bipolar membrane the use of demineralized water was sufficient, because this type of membrane exhibits low permeability. When a potential difference is applied over the membrane, the catalytic active intermediate layer of the bipolar membrane dissociates water to H.sub.3.sup.+ and OH.sup.. The protons produced move to the cathode, the hydroxide ion move to the anode. pH and temperature of the liquid flowing through the anode and cathode compartment were measured. The anode and cathode compartment solutions were in open contact with surroundings to prevent pressure build-up due to possible hydrogen formation. Oxidation of Fe.sup.2+ because of oxygen was prevented with a nitrogen blanket. First the oxidation behaviour of Fe.sup.2+ to Fe.sup.3+ in the electrochemical cell was determined. Afterwards experiments were performed in which CuS was added.
[0282] Fe.sup.2+ Permeability
[0283] It was observed that when no potential difference was applied, anion exchange membranes exhibited a lower permeability for Fe.sup.2+ compared to cation exchange membranes.
[0284] It was observed, under equal experimental settings, that the bipolar membrane exhibited a lower metal ion permeability compared to the anion exchange membrane at equal experimental settings, as is illustrated in
[0285] Fe.sup.2+ Oxidation
[0286] It was observed that when a potential difference was applied, gas was formed at the cathode side of the electrochemical cell for both the bipolar membrane as well as the anion exchange membrane. Moreover, it was observed that the Fe.sup.2+ oxidation rate was higher when using a bipolar membrane compared to an anion exchange membrane at equal experimental settings, as is illustrated in
[0287] CuS Oxidation
[0288] An additional experiment was performed by adding a Vitrisol solution, obtained from the Vitrisol pilot absorber described in Experimentation 1, containing 4 wt % of CuS to the anode section of the electrochemical cell. A bipolar membrane was used, the experiment was performed under ambient conditions, and a potential difference above 15 V was applied. Before the start of the experiment both Fe.sup.2+ and Fe.sup.3+ were present in the anode compartment.
[0289] It was observed that the current density was higher when CuS solids were added to the anode section. Moreover, it was observed that the hydrogen gas production declined after a prolonged period of operation time. However, at this point the current density remained constant. It was clearly observed that solid copper was deposited on the cathode electrode. The potential difference applied was large enough to deposit iron on the electrode, however, solid iron was not visible on the electrode.
Experimentation 8: Sulphur Dissolution
[0290] A variety of experiments were performed to investigate the solubility of sulphur in various nonpolar solvents.
[0291] Experimental Setup Description
[0292] Dissolution experiments were performed in a three-necked round bottom flask. Temperature was controlled inside the round bottom flask. Stirring was performed with a magnetic stirrer. A condenser was connected to the round bottom flask to condense nonpolar solvent vapours. Samples were taken with syringes and ejected on petri disks to evaporate the nonpolar solvent.
[0293] Sulphur solubilities were obtained by increasing the temperature from 30 C. to 90 C., by decreasing the temperature from 90 C. to 30 C. and by fluctuating the temperature between 30 C. and 90 C.
[0294] Sulphur Solubility
[0295] It was observed during the experiments that the sulphur solubility increased with temperature. An example of a sulphur solubility curve with p-xylene as nonpolar solvent is illustrated in
Experimentation 9: Vitrisol Process Simulation
[0296] Conceptual designs of the Vitrisol process were developed for several low pressure gas and high pressure gas cases with varying H.sub.2S concentration. Simulations of these different conceptual designs were performed. Notable outcomes were the following: [0297] The varying of possible unit operations of the oxidizer step for Vitrisol with regeneration. A bubble column as oxidizer appears to be sufficient for the regeneration of low amounts of CuS, whereas a gas-liquid contactor with high intensity stirring is preferred when large amounts of CuS have to be regenerated. Therefore, it can be stated that the most optimal oxidizer design is strongly case dependent. [0298] The varying of possible unit operations for the removal of sulphur from the Vitrisol process with regeneration. Saturating a batch of nonpolar solvent, and periodically crystallizing out sulphur by cooling to surroundings, is one of the preferred alternatives for low amounts of sulphur, whereas a crystallizer is preferred when large amounts of sulphur need to be crystallized continuously. Therefore, it can be stated that the most optimal sulphur recovery steps are also strongly case dependent.
[0299] While the processes and systems were generally described in relation to H.sub.2S removal, it should be understood that various other sulphur contaminants present in a gas stream could be treated and removed using implementations of the processes and systems. In addition, Cu and Fe are described as the preferred metals which in ionic form are used in certain implementations of the processes and systems described herein, but it should be noted that other appropriate metal ions may be used.