EP2771106A1

EP2771106A1 - Process for the removal of contaminants

Info

Publication number: EP2771106A1
Application number: EP12780522.4A
Authority: EP
Inventors: Gary Bevan COMBES; Peter Carnell
Original assignee: Johnson Matthey PLC
Current assignee: Johnson Matthey PLC
Priority date: 2011-10-27
Filing date: 2012-10-03
Publication date: 2014-09-03
Also published as: GB201405476D0; WO2013061027A1; GB2495962A; GB2509028A; GB201118552D0

Abstract

A process is described for the removal of a contaminantselected from a drying agent and a corrosion inhibitorfrom a sorbent selected from afiltration medium, a molecular sieve material, a porous oxide or a chemical sorbent,comprising the step of applying a carbon dioxide stream in which the carbon dioxide is present in the liquid phase orsupercriticalphase, to the contaminated sorbent in a sorbent vessel to remove the contaminant from the sorbent and form a contaminant-laden carbon dioxide stream and a decontaminated sorbent. Preferably the contaminant-laden carbon dioxide stream is recovered from the sorbent vessel and fedto a separation vessel in which the pressure is reduced to cause separation of the contaminant from a depressurised gaseous carbon dioxide. Adepressurised gaseous carbon dioxidestream and a separate contaminant streammay then be recovered from the separation vessel. The recovered depressurised carbon dioxide may be re-pressurised and fed back to the sorbent vessel. The process may be used for the regeneration of contaminated sorbents.

Description

Process for the removal of contaminants

This invention relates to a process for the removal of contaminants using carbon dioxide at high pressure in the liquid or supercritical phase. Sorbents used in industry become contaminated with various compounds that can reduce their effectiveness to capture the target materials. Liquid contaminants can be particularly problematic. Liquid natural gas (LNG) plants have a problem with triethyleneglycol (TEG) or other similar compounds that are introduced into gas pipelines to remove moisture, to inhibit formation of gas hydrates and to a lesser extent, to inhibit corrosion in pipelines. Onshore efforts to remove TEG are not always successful due in part to the enhanced solubility of TEG in pressurised natural gas streams, which will carry over the TEG and deposit it on downstream equipment, and sorbents, particularly the mercury- and sulphur- guard beds that usually follow the TEG removal stages. This is particularly a problem at high pressure fields, leading to blinding of the mercury- and sulphur- guard beds and molecular sieves. The shortened lifetime of the guard beds adds significant cost of higher pressure drop and catalyst replacement costs to the running of the plant and makes handling spent sorbent difficult.

US20080197079 discloses a method for desorbing 2,2,3,3-tetrafluoro-1-propanol off a carbon sorbent using supercritical carbon dioxide. We have found surprisingly that carbon dioxide may be used to desorb hydrophilic drying agent or corrosion inhibitor compounds from molecular sieves, porous metal oxides or chemical sorbents.

Accordingly the invention provides a process for the removal of a contaminant selected from a drying agent and a corrosion inhibitor from a sorbent selected from a filtration medium, a molecular sieve material, a porous oxide or a chemical sorbent, comprising the step of applying a carbon dioxide in which the carbon dioxide is present in the liquid phase or supercritical phase, to the contaminated sorbent in a sorbent vessel to remove the contaminant from the sorbent and form a contaminant-laden carbon dioxide stream and a decontaminated sorbent. Herein by the term "sorbent" we include "adsorbent" and "absorbent". The sorbent may be a filtration medium, such as a particulate solid or semi-permeable membrane, a molecular sieve material such as a zeolite sorbent, a porous oxide such as an alumina, silica or silica-alumina, which acts to physically trap a target substance, or a chemical sorbent in which the sorbent reacts with a target substance. Such chemical sorbents include, but are not limited to, sulphur- guard sorbents comprising copper, manganese or iron compounds, for example sorbents comprising basic copper carbonate and/or zinc oxide, chloride-guard sorbents such as alkalized oxide materials, or sulphided transition metal sorbents, such as sulphided copper sorbents that may be used to capture heavy metals such as mercury or arsenic. The sorbents may also be catalytically active, for example for oxidation or for the hydrogenation of sulphur or nitrogen compounds.

Contaminants that are removed using the process include drying agents and corrosion inhibitors present in the process fluid. In a particularly preferred embodiment the contaminant comprises one or more glycols, such as diethyleneglycol (DEG) and triethyleneglycol (TEG), polyethyleneglycols (PEG), or glycolethers. The contaminant level on the sorbent may be in the range 1-50% wt or higher depending upon when the decontamination of the sorbent is performed.

The carbon dioxide (C02) stream used for the de-contamination may be a liquid carbon dioxide stream or a supercritical carbon dioxide stream. The carbon dioxide stream may comprise one or more inert compounds or co-solvents, but preferably the carbon dioxide content is >75% vol, more preferably >90% vol. Liquid carbon dioxide streams are preferably used at temperatures in the range -10 to 30°C, preferably 5 to 20°C and at pressures at least about 10 bar above the dew point pressure and in the range 50 to 200 bar abs, preferably 60 to 120 bar abs.

Supercritical carbon dioxide streams are preferably used at temperatures >31°C, preferably 35- 65°C, more preferably 40 to 60°C and at pressures in the range 85 to 250 bar abs, preferably 95 to 160 bar abs. The liquid phase carbon dioxide has the advantage that the contaminant solubility may be higher, particularly at temperatures in the range 0-10°C. The density of the carbon dioxide stream is preferably > 600kg/m³, more preferably > 700kg/m³.

Co-solvents may be included with the carbon dioxide if desired to increase the solubility of the contaminant. It has been found that the inclusion of a co-solvent such as water or methanol may enhance the glycol solubility and so enhance glycol removal from the contaminated sorbent. The amount of co-solvent is preferably in the range 0.1 to 5 mol%.

The contaminate-laden carbon dioxide stream may be subjected to separation of the contaminant by recovering the contaminant-laden carbon dioxide stream from the sorbent vessel and feeding it to a separation vessel in which the pressure is reduced to cause separation of the contaminant from a depressurised gaseous carbon dioxide. A depressurised gaseous carbon dioxide stream and a separate contaminant stream may then be recovered from the separation vessel. In the separation vessel, the contaminated carbon dioxide stream is preferably depressurized to a pressure at which the contaminant solubility in carbon dioxide is significantly reduced. Pressures below 80 bar abs may be used. In a preferred embodiment, with liquid or supercritical carbon dioxide at a pressure 100 to 120 bar abs, the pressure is reduced in the separator vessel in one or more stages to between 40-70 bar abs. The depressurization of the contaminated carbon dioxide stream forces the contaminant out of the carbon dioxide stream to form a depressurized gaseous carbon dioxide stream. Liquid contaminants such as glycols may then be readily recovered from the separation vessel. Solid contaminants may collect within the separation vessel and also be recovered once the separation vessel is off-line. The recovered contaminant may be re-used, recovered for disposal, or burnt as a fuel.

The recovered depressurised carbon dioxide may be re-pressurised and fed back to the sorbent vessel. In this way the carbon dioxide may be re-circulated and not vented to atmosphere. Thus the process preferably includes the steps of re-pressurising and optionally adjusting the temperature of the depressurized gaseous carbon dioxide stream to form a carbon dioxide stream in which the carbon dioxide is present in the liquid phase or supercritical phase and feeding the pressurised carbon dioxide stream back to the sorbent vessel. In the case of liquid carbon dioxide, the stream generally does not require heating, whereas for supercritical carbon dioxide some subsequent heating of the re-pressurised carbon dioxide stream may be required.

Alternatively, the contaminate-laden carbon dioxide stream may be used in an enhanced oil recovery process or may be sent to a carbon-capture & storage (CCS) process.

The process may be used for the regeneration of contaminated sorbents.

Accordingly the invention includes a process for the regeneration and re-use of a contaminated sorbent comprising the steps of:

(i) passing a contaminated process fluid contaminated with a contaminant selected from a drying agent and a corrosion inhibitor over a sorbent selected from a filtration medium, a molecular sieve material, a porous oxide or a chemical sorbent disposed in a sorbent vessel for a period to form a contaminated sorbent,

(ii) stopping the flow of contaminated process fluid to the sorbent vessel,

(iii) treating the contaminated sorbent by applying a carbon dioxide stream in which the carbon dioxide is present in the liquid phase or supercritical phase, to the contaminated sorbent in to remove the contaminant from the sorbent and form a contaminant-laden carbon dioxide stream and a decontaminated sorbent, and

(iv) passing the contaminated process fluid over the decontaminated sorbent.

The contaminated process fluid may be a contaminated gas stream such as natural gas, including natural gas streams comprising carbon dioxide. Other contaminated gas streams may also be treated including inert gases such as nitrogen, hydrocarbon gas streams, flue gas streams and even carbon dioxide gas streams. Alternatively the contaminated process fluid may be a liquid. In a preferred embodiment, the contaminated process fluid is a glycol- contaminated natural gas, particularly a TEG-contaminated natural gas stream. Natural gas streams comprising higher hydrocarbons than methane are particularly susceptible to glycol contamination due to the higher solubility in such natural gases at high pressure. Natural gas streams may be passed over sorbents at pressures in the range 50-250 bar abs and at temperatures in the range -20 to +50°C.

The contaminant level in the contaminated process fluid may be in the range 1-10,000 ppmv or higher.

The carbon dioxide is preferably applied in-situ to the sorbent disposed in the sorbent vessel. The decontamination process may be applied to one or more, preferably at least two, guard beds of chemical sorbents. Alternatively, upstream of the guard bed of chemical sorbent may be one or more, preferably at least two, beds containing a sorbent designed specifically to capture the contaminant. A preferred method of operation of either of these is to have at least one sorbent bed on-stream at any one time, and at least one undergoing treatment with liquid or supercritical carbon dioxide. Such lead-lag arrangements allow continual regeneration and re-use of sorbents in heavily contaminated streams. A benefit of the latter process using contaminant capture upstream of the chemical sorbents is that heating can be applied to smaller vessels upon the removal of the contaminant, thus enabling a more efficient removal by running above the critical temperature and pressure of carbon dioxide, and using pressure to separate the contaminant out in a recovery vessel and recycling the carbon dioxide. At supercritical conditions, the preferred envelope for removing TEG from the sorbents is at pressures above 85bar and to 160 bar or the maximum of the plant process; whichever is the greater, and in the temperature range of 40 to 60°C. This minimises the use of carbon dioxide and the amount of recycle. This technique could be extended to removing the TEG from the upstream TEG recovery beds and filters in situ as well, thus providing a quick and facile way for removing TEG from process equipment and materials.

Where the process fluid contains carbon dioxide, a portion of it may be recovered using conventional washing or membrane separation techniques and used as the source of the liquid or supercritical carbon dioxide used to decontaminate the sorbent. Alternatively the pressurised carbon dioxide may be produced from a carbon dioxide stream recovered from a hydrocarbon conversion process such as gasification, partial oxidation or steam reforming of a hydrocarbon. The process may be applied to other systems. The advantages of the process include:

1 ) extending the lifetime of the guard beds.

2) reducing the variable pressure drop over the beds.

3) recovering the glycol for reuse in the gas field. 4) enabling easier changeout of the guard beds free from contamination with a contaminant residue.

5) use of an existing greenhouse gas material for beneficial use.

6) cost savings on process equipment and materials.

7) the process can be operated within existing plant process conditions.

The invention is further described by reference to the accompanying Drawings in which:

Figure 1 depicts a flowsheet of an embodiment utilizing supercritical carbon dioxide, and Figure 2 depicts TEG recovery with time in liquid and supercritical carbon dioxide from TEG- contaminated materials.

In Figure 1 a lead-lag arrangement is depicted with two sorbent vessels 10 and 12 containing particulate sorbent beds 14 and 16 respectively. The vessels 10, 12 are fed with either a TEG contaminated stream via line 18 or a supercritical C02 stream via line 20. Valves A, B, C, D, E, F, G and H control the flow to the vessels 10, 12 such that while one is in use, one is subject to decontamination. Thus, in one embodiment, a triethyleneglycol (TEG)-contaminated natural gas at 5°C/120 bara. is fed via line 18 through valve A and then line 22 into the first vessel 10. The TEG contaminated natural gas passes through a bed of a particulate copper-containing sorbent 14 which captures sulphur compounds and/or mercury present in the natural gas. TEG is also deposited in the pores of the sorbent over time. The natural gas passes from the first vessel 10 via line 24 through valve B and thence via line 26 for further processing. At the same time, a supercritical C02 stream from line 20 and at 40°C/150 bara, passes through valve G and line 28 into the second vessel 12 containing a TEG-contaminated sorbent material 16. The supercritical C02 dissolves the TEG and removes it from the contaminated sorbent. The TEG- contaminated C02 is recovered from the vessel 12 via line 30, passes through valve H and is fed via line 32 to a first separator 34, where it is depressurized to about 50 bara. The C02 is vapourised under these conditions. The liquid TEG, which may contain C02 residues, is collected from the bottom of the first separator 34 and fed via line 36 to a second separator 38 where the pressure is further reduced to 1-10 bara. The liquid TEG is recovered from the second separator 38 via line 40. A small purge of C02 may be collected from the second separator via line 42. The vapourised C02 is recovered from the first separator 34 via line 44 and fed to a condenser 46 before being fed via line 48 to a pump 50 which re-pressurises the C02 to 150 bara. The feed to the pump 50 may be augmented by a stream of imported C02 fed via line 52. The imported C02 may be at 20°C and 50-60 bara. The re pressurized C02 is fed via line 54 to a pre-heater 56 that adjusts the pressurized C02 temperature to 40°C, to provide the supercritical C02 stream 20.

In this embodiment therefore valves A, B, G and H are initially open and valves C, D, E and F are closed. When the sorbent 14 in vessel 10 becomes saturated with TEG such that its performance is no longer satisfactory, the vessel 10 is taken off-line and the sorbent regenerated using the supercritical C02, and at the same time, the sorbent 16 in vessel 12 is used to remove undesired substances from the TEG-contaminated natural gas. Thus valves A, B, G and H are closed and valves C, D, E and F are opened.

It will be understood that the copper containing sorbent in vessels 10 and 12 may be replaced with a variety of sorbents including filtration media, porous oxides, molecular sieves and other chemical sorbents.

The invention is further described by reference to the following Examples. Example 1 : Deactivating effect of TEG on chemical sorbent

Six, 10 ml beds of a Cu/Zn carbonate/alumina sorbent, were charged in to a reactor and a 1 % v/v H2S stream was passed over the material in a nitrogen carrier gas. The flows were controlled by mass flow controllers and the system operated at ca 2 - 12psig. The test was deemed to be over when >10ppm H2S was detected at the exit of the reactor, measured by gas detection tube. The test was run 3 times, firstly with no glycol (TEG) introduced to the system, secondly with TEG introduced to the system in the vapour by passing all the carrier gas through a 6" bubbler to carry TEG vapour forwards and finally by dipping the front portion of the bed into liquid TEG and loading this into the reactor to simulate a gross carry over event or the effect of continual trace carry for a sustained period of time.

The results of this are shown below:

Table 1

The case where liquid is present in the example shows clear deactivation, whereas the case where TEG vapour is present shows minimal deactivation and the some evidence of the S profile been extended down bed to a greater extent before break through takes place. The results show that liquid TEG on the sorbent reduces its effectiveness.

Example 2: Removal of TEG

Two triethylenglycol-contaminated materials were subjected to decontamination using pressurised C02 with or without methanol or water as co-solvent. Carbon analysis was performed using both a combustion technique with LECO apparatus on the decontaminated sorbent and gravimetrically by periodic sampling of the recovered TEG. The materials are described in Table 2.

Table 2

The conditions investigated included:

1 ) 40°C, 1 1 1 bara C02

2) 40°C, 151 bara C02, saturated with water (~0.4mol% water)

3) 40°C, 151 bara C02, 2.5mol% methanol

4) 40°C, 151 bara C02 on material B

5) 40°C, 151 bara C02 on reduced sample mass

6) 5°C, 1 1 1 bara liquid C02

The sample materials were loaded into a 1 " stainless steel tube with settling of the solids achieved by tamping. Material A was 2-4 mm in diameter but irregular platelets, whilst material B was 2-3 mm in diameter and granules. Unless otherwise stated, approximately 40g of the contaminated material was loaded in each vessel with cotton wool placed at either end to hold the solids in place. C02 was supplied by an ISCO 260D pump at constant pressure to a cell heated in an oven to maintain a tube temperature of 40°C. The flowrate was maintained at approximately 3-5ml/min at the pump after 1 hour initial equilibration. The runs with methanol co-solvent were performed at constant flowrate at 150 barg with minor pressure fluctuation. The run with water saturated C02 was performed by passing the pressurised C02 through a vessel containing liquid water. The C02 after exiting the bed was depressurised through a heated valve and passed through a solution of acetone to capture the extracted TEG. The valve was flushed with acetone upon changing of samples. Samples were collected approximately every 30 ml of C02 and placed in an oven at 60°C to remove the acetone. Samples from the methanol run were dried at 80°C. After 1-2 hours in the oven, the samples were cooled and weighed and then returned to the oven. The process was repeated for typically two cycles to ensure the residual mass was unchanging. The mass of TEG recovered, together with the C02 used during the extraction period was used to provide a gravimetric analysis of the extraction process.

The %TEG recovered versus the carbon dioxide pumped is depicted in Figure 2. The results show differences between the extraction conditions and the samples, but all show effective TEG removal.

The LECO results, along with the adjusted gravimetric results are given in Table 3. Table 3

The differences between the LECO and gravimetric analyses are believed to arise from residual carbon dioxide in the TEG on the sorbent. Accordingly TEG, as observed

gravimetrically, was removed from the sorbent in Run 1 even though the %C result from the LECO analysis suggests otherwise. The differences are reduced at higher levels of extraction.

The impact of pressure can be seen in the enhanced recovery rate at 151 bar compared to 1 1 1 bar. In the runs with co-solvents, the beneficial impact of water and methanol can be observed.

The run at lower temperature demonstrates that liquid carbon dioxide can be used effectively to remove TEG from a sorbent. Moreover, liquid C02 could be used on other parts of the plant that are not heated to remove TEG from filters, lines or other materials. The results of Run 5 clearly demonstrate that >95% extraction of TEG can be achieved from the pores of the alumina material.

Claims

Claims.

1. A process for the removal of a contaminant selected from a drying agent and a

corrosion inhibitor from a sorbent selected from a filtration medium, a molecular sieve material, a porous oxide or a chemical sorbent, comprising the step of applying a carbon dioxide stream in which the carbon dioxide is present in the liquid phase or supercritical phase, to the contaminated sorbent in a sorbent vessel to remove the contaminant from the sorbent and form a contaminant-laden carbon dioxide stream and a decontaminated sorbent.

2. A process according to claim 1 wherein the sorbent comprises a chemical sorbent.

3. A process according to claim 2 wherein the chemical sorbent comprises a copper, manganese or iron compound.

4. A process according to any one of claims 1 to 3 wherein the contaminant comprises one or more glycols.

5. A process according to any one of claims 1 to 4 wherein the contaminant level on the sorbent is in the range 1-50% wt.

6. A process according to any one of claims 1 to 5 wherein the carbon dioxide stream used for the de-contamination is a liquid carbon dioxide stream at a temperature in the range -10 to 30°C, preferably 5 to 20°C and at a pressure at least about 10 bar above the dew point pressure and in the range 50 to 200 bar abs, preferably 60 to 120 bar abs.

7. A process according to any one of claims 1 to 5 wherein the carbon dioxide stream used for the de-contamination is a supercritical carbon dioxide stream at a temperature 31°C, preferably 35-65°C, more preferably 40 to 60°C and at a pressure in the range 85 to 250 bar abs, preferably 95 to 160 bar abs.

8. A process according to any one of claims 1 to 7 wherein a co-solvent is included with the carbon dioxide.

9. A process according to claim 8 wherein the co-solvent comprises water or methanol.

10. A process according to any one of claims 1 to 9 wherein the contaminant-laden carbon dioxide stream is recovered from the sorbent vessel and fed to a separation vessel in which the pressure is reduced to cause separation of the contaminant from a depressurised gaseous carbon dioxide, with recovery of the depressurised gaseous carbon dioxide stream from the separation vessel.

1 1. A process according to claim 10 wherein the he recovered depressurised carbon dioxide is re-pressurised and fed back to the sorbent vessel.

12. A process according to claim 1 1 wherein when the carbon dioxide stream is

pressurised to a supercritical state that it is heated to a temperature >30°C, preferably 35-65°C, more preferably 40 to 60°C.

13. A process according to any one of claims 1 to 9 wherein the contaminate-laden carbon dioxide stream is used in an enhanced oil recovery process or is sent to a carbon- capture & storage process.

14. A process for the regeneration and re-use of a contaminated sorbent comprising the steps of:

(ii) stopping the flow of contaminated process fluid to the sorbent vessel,

(iv) passing the contaminated process fluid over the decontaminated sorbent.

15. A process according to claim 14 wherein the contaminated process fluid comprises a natural gas stream at a pressure in the range 50-250 bar abs and at a temperature in the range -20 to +50°C.

16. A process according to claim 14 or claim 15 wherein the contaminated process fluid is a glycol-contaminated natural gas.

17. A process according to any one of claims 14 to 16 wherein the process fluid contains carbon dioxide, and at least a portion of it is recovered and used as the source of the liquid or supercritical carbon dioxide used to decontaminate the sorbent.

18. A process according to any one of claims 14 to 17 comprising two or more sorbent vessels, each containing a sorbent, wherein at least one sorbent is on-stream at any one time, and at least one other is treated with liquid or supercritical carbon dioxide.