Removal of dissolved hydrocarbons from production water by Macro Porous Polymer Extraction (MPPE)

Transcription

1 SPE 6577 Removal of dissolved hydrocarbons from production water by Macro Porous Polymer Extraction (MPPE) H.M. Pars / Elf Petroland BV - D.Th. Meijer / Akzo Nobel NV Copyright 1998, Society of Petroleum Engineers, Inc. This paper was prepared for presentation at the 1998 SPE International Conference on Health, Safety and Environment in Oil and Gas Exploration and Production held in Caracas, Venezuela, 7 June This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box , Richardson, TX , U.S.A., fax Abstract This paper presents the successful application of Macro Porous Polymer Extraction (MPPE) technology to remove hydrocarbons from the produced water of a natural gas plant. This water is discharged to the public watertreatment facilities and is subject to a maximum hydrocarbon contents. Elf Petroland together with Akzo Nobel investigated the use of a Macro Porous Polymer filled with an extraction liquid for hydrocarbon removal. A MPPE unit was installed June 9 at Elf Petroland in Harlingen (The Netherlands), downstream the reflux drum of the DEG/TEG regenerators and treated the condensed water stream. The unit proved to be very efficient. The hydrocarbon content was reduced from approx mg/l to less than mg/l. Since May 97 the performance is tested for the treatment of all produced water (process-, formationand wastewater) to a maximum benzene of 0 ug/l. MPPE utilises two proven technologies, liquid-liquid extraction and steam stripping with an innovative medium (the MPPE particles). Hydrocarbon contaminated water is passed through a column packed with MPPE particles. An extraction liquid immobilised within the polymer matrix removes the hydrocarbons from the water. The purified water passes out of the column directly for reuse or discharge. All hydrocarbons with a relatively high affinity for the extraction liquid (compared to water) are removed. Periodical in-situ regeneration of the extraction liquid containing MPPE particles is accomplished with low pressure steam. The steam volatises the hydrocarbons. Volatised hydrocarbons are condensed and then separated by gravity. The hydrocarbon phase is recovered and sold as product, the water phase is recycled to the system. Besides the Harlingen pilot, the unit is proven successful in other branches of industry. At present fourteen units are operational or under construction in the USA and Europe, of which six treat process water and eight are used in the field of groundwater remediation. Removal efficiencies up to % are achieved for a wide varieties and combinations of aliphatic, aromatic and halogenated (e.g. chlorinated) hydrocarbons. Introduction Early 1992 several E&P companies were contacted by Akzo Nobel to investigate their interest in a new technique to remove hydrocarbons from production water. Elf Petroland was interested because of difficulties to meet the applicable legislation, requiring a maximum hydrocarbon of 0 mg/l on the Dutch Continental Shelf and 20 mg/l on onshore locations. For this reason also another technique was in development, steam stripping, and the Akzo proposal could mean an interesting alternative. Mid 1992, a test was conducted at Elf Petroland s onshore treating centre at Harlingen, in the North of the Netherlands. Condensed water from the glycol (DEG/TEG) regeneration units, having a hydrocarbon of 2000 mg/l, was led through columns packed with MPPE particles. The test proved successful and Elf Petroland showed interest in a fullscale test. A full-scale unit was developed with simplified design choices (like manual operation in stead of automatic) and sized to reduce the hydrocarbon contents from 2000 mg/l to mg/l for a stream of m3/hr. This unit was installed in June 9 for testing at field scale. The good results of the test were for Elf the reason to acquire the unit in June 95, automate it in Nov. 96 and extend the application to the full discharged water stream produced by the Harlingen treating centre (formation-, process- and wastewater). This application and the required pre-treatment are currently being tested, with promising results. The success of the pilot unit led for Akzo to further introduction of the MPPE technology to other industrial applications.

2 2 H.M. PARS AND D. TH. MEIJER SPE 6577 MPPE technology MPPE process Volatile hydrocarbons 1. The MPPE particles. A scanning electron microscopic (SEM) photograph of Macro Porous Polymer particles is shown in figure 1. The porous polymer particles have sizes micron, with pore sizes 0.1- micron. The resulting porosity is 70% to 80%. Extraction Water Stripping Condensor Condensate recycle Settler Water Volatile hydrocarbons Volatile hydrocarbons Figure 2: MPPE extraction/ stripping system Figure 1: Internal structure of the Macro Porous Polymer These polymers were initially developed as controlled release media in medical applications. The application in water treatment started in 1991: a test at Orkney Water Test Centre on behalf of Akzo Nobel showed that MPP can be used to remove dissolved and dispersed oil from water by absorption. This was interesting because dissolved hydrocarbons are the main problem in offshore operations (these cannot be separated by gravity). The polymer was further developed and the extraction liquid added, resulting in the actual efficient removal of dissolved and dispersed hydrocarbons. 2. System description. A schematic representation of a typical MPPE unit is shown in figure 2. The hydrocarbon contaminated water is passed through a column packed with MPPE particles. At the end of the extraction phase, the column is regenerated in situ with low pressure steam. The shown two columns allow continuous operation with simultaneous extraction and regeneration. At the end of the cycle, columns are switched: the one used for extraction is regenerated and vice versa. During regeneration the volatile hydrocarbons are removed from the particles by steam stripping, while the immobilised non-volatile extraction liquid is retained in the pores of the polymer. After condensation of the vapour phase an organic and an aqueous phase are obtained in the gravity separator. The aqueous condensate, containing a small amount of the hydrocarbons, is returned to the feed of the extraction column. The organic phase is sold as product. 3. Applicability. The components than can be removed with above described MPPE technique need to have a high affinity (i.e. partition coefficient) for the extraction liquid compared to water in order to allow extraction, and a high volatility, that is boiling point up to 250 C, to allow regeneration by steam stripping. Type of water Components Flow rate GW GW BTEX VCH BTEX VCH MO Influent Effluent [m 3 / h] [ppm] [ppm] < 0.1 GW VCH PW VCH 0.1 PW HCH VCH PW CS < dl < GW VCH < 0.3 GW VCH GW = Ground water PW = Process water BTEX = Benzene, toluene, ethylbenzene, xylene VCH = Volatile Chlorinated hydrocarbons HCH = Hexa chlorinated hydrocarbons MO = Mineral oil CS 2 = Carbon disulfide dl = detection limit Table 1: Pilot scale field tests: examples of components removed Examples of volatile hydrophobic components are aliphatic hydrocarbons, aromatic hydrocarbons (benzene toluene) and chlorinated hydrocarbons. Polychlorobenyls (PCB s) and polyaromatic hydrocarbons (PAH s) can be extracted with MPPE but due to the low volatilities, a different regeneration technique than steam stripping needs to be used.

3 SPE 1 REMOVAL OF DISSOLVED HYDROCARBONS FROM PRODUCTION WATER BY MPPE 3 The removal efficiency that can be realised with MPPE is very high, as a result of the high number of mass transfer stages in the packed bed (high specific area for mass transfer) and in-situ separation of extraction liquid and purified water. Removal efficiencies up to % ( reduction factor of 6 ) have been achieved in both pilot and full-scale units. Some examples of pilot scale field tests are presented in table 1.. System design. For the design of a specific application a computerised mathematical model is used. The model describes extraction and regeneration. All relevant physical processes are taken into account: Liquid/liquid (extraction) and vapour/liquid (regeneration equilibria) Liquid/liquid and vapour/liquid mass transfer Residence time distribution in the MPPE bed Heat transfer (regeneration) Pressure drop over the packed bed (during extraction and regeneration) Properties of the components to be removed are taken from a database containing physical properties of many organic compounds. A summary of the input and output parameters of the model is given in table 2. Input parameters Output parameters Water flow rate Equipment dimensions Effluent requirements Regeneration time Partition coefficients consumption Mass transfer coefficients Amount of MPPE Residence time distribution data Breakthrough curves MPPE particle size Regeneration curves Bed porosity Pressure drops (extraction Specific heats and regeneration) Feed s Extraction time Table 2: Input and output parameters of the MPPE design model Possibly lacking data can be determined experimentally on lab scale for both extraction and regeneration The calculated extraction breakthrough curves (influent and effluent s versus time) are verified once the unit is installed. If necessary, the operating conditions can be adjusted. An important design parameter is the extraction time. This determines the size of the column: a longer extraction time means a larger quantity of MPPE particles to contain the extracted hydrocarbons and results in a larger sized column. It also determines the required steam capacity: a longer extraction time means more time is available for regeneration, resulting in less required steam capacity. In general the extraction time is chosen 1 to 2 hours. 5. Influence parameters Once the design is calculated based on the parameters mentioned above, the influent properties may change (more or less flow, more or less hydrocarbons ). The influence is shown in table 3 below. Case Base 1 1a 1b 2 Flowrate (m³/hr) Feed conc. (ppm) Effluent conc. (ppm) consumption (kg/m³ water) 8.6 Table 3. Influence of varying flowrate and feed Influent. Case 1 shows that a 50 % higher influent will result in a 50 % higher effluent. To still meet the design effluent of 0.1 ppm, only a % flow reduction is required as shown by case 1a. This can for example be achieved in case of varying flows, by installing a buffer tank. If there is no possibility to decrease the flowrate, the regeneration time could be shortened. Case 1b shows that with a 50 % too high feed the effluent requirements can still be met by increasing the annual steam consumption with %. For this a % larger steam capacity is required, because the increased steam capacity is only used during part of the regeneration cycle. Flow rate. Case 2 shows that a 15 % higher flowrate will result in a 0 % higher effluent. To bring the effluent back to design conditions is more difficult. It is therefore recommended to oversize the unit with respect to the maximum expected flow, to allow treating unexpectedly higher flows. Table, cases c. and d. shows the impact of oversizing for flow rate. Case Flow m3/hr Feed ppm Effl. conc. Ppm Invest. *00 USD consump. ton/year Power consump. Kwh/year Sercie contract *00 USD/year a. 2000, , b , , c. 0, ,000 0 d , ,500 7 Table. Cost and other parameters for some design cases Note that during loading of the column, hydrocarbon contents in the effluent gradually increases, see figure 3. This means that if at the end of the loading time the is within the required limit, in average over the loading period a significantly lower is discharged. In general the average in the effluent during extraction appears to be only approximately 1/3 of the measured at the end of the cycle. For calculating the discharged amount of hydrocarbons, this should be taken into account, otherwise too large values will be obtained.

4 H.M. PARS AND D. TH. MEIJER SPE 6577 Extraction Break-Trough Curve DCM First industrial MPPE unit DCM (ppm) Time (hr) Figure 3. Extraction break-through curve for DCM 6. Process cycle. The unit operates in a sequence of steps described below (see figure for the process flow diagram of a 2-column system). a. Extract with column C-01 and regenerate column C-02. This step together with step d. is the main part of the process: one column being loaded and the other column being regenerated. Advance to the next step after the predefined extraction time (e.g. 90 minutes). b. Pass over from column C-02 to column C-01. The influent is routed to the just regenerated column C-02 and passed over to column C-01. This way the heat content of column and MPPE material of the just regenerated column is used to pre-heat the other column before regeneration. Advance to the next step when the temperature at the top of column C-02 drops below a predefined level (e.g. below 60 C). 1. Application description. The first industrial application of the MPPE technology took place at the gas treating centre of Elf Petroland in Harlingen. Harlingen is located in the North of Holland, on the shore of the Waddensea, a shallow tidal pool connected to the North Sea. There, Elf Petroland collects natural gas from one off-shore platform and several onshore fields, with a total gas production of approx. 2,5 MNm3/day (90 MMscf/D). The total water production is approx. 0 m3/day. Because of the main objective being to reduce the amount of dissolved hydrocarbons (benzene, toluene, ethylbenzene, xylene), it was decided initially to treat the most BTEX rich part of the process water: the condensed vapours from the DEG and TEG regeneration units. 2. Unit description The pilot MPPE unit was installed in June 9 downstream the reflux drum of the glycol regenerators. A photograph of the unit is presented in figure 5. Column sizes are 2 m height and 0.8 m diameter and designed for 8 hours extraction time (the switch over from extraction to regeneration was initially performed manually by the plant operator, the extraction time was taken in order to have only one column changeover per 8 hrs shift). The water stream flowrate to be treated ranged from 2 to m 3 /h, with benzene s from 2000 to ppm. For the regeneration a steam generator is installed in the skid, which is fed with demineralized water. The regeneration vapours are condensed by the production water effluent of the unit. The effluent is lined up to the Harlingen TC watertreatment unit and afterwards discharged to public watertreating facilities. Figure : Process flow diagram 2-column system c. Extract with column C-02 and empty column C-01. The steam is routed to the top of column C-01. Liquid recovered from this column is routed to the buffer tank. Advance to the next step when temperature at the bottom of column C-01 exceeds a predefined value (e.g. above 90 C). d. Extract with column C-02 and regenerate column C-01. e. Pass over from column C-01 to column C-02. f. Extract with column C-01 and empty column C-02. Figure 5: MPPE unit installed at Elf Petroland, Harlingen TC, the Netherlands

5 SPE 1 REMOVAL OF DISSOLVED HYDROCARBONS FROM PRODUCTION WATER BY MPPE 5 3. Improvements. Several problems were solved: Loss of MPPE grains. Initially grains were packed loose in column cylinders on a bottom screen. Material was several times lost with effluent and caused blocked lines. A filter cartridge system was introduced to avoid the possibility to loose material and also for easier handling in case of MPPE renewal. Water supply. Demineralisation had to be applied on the water used for steam generation (potable water), because the hardness of the water contents caused problems with the steam generator. condensation. Initially steam condensation blocked the regeneration lines. This was solved by eliminating low points in the steam piping and by trace heating the piping to ensure at least 1 C. Condensate problems. Free condensate, caused for example by carry-over from the DEG or TEG reflux drums, has caused problems with meeting the required effluent, although also influent peaks of, 7000 and,000 ppm hydrocarbons, showing a dispersed phase, have been reduced to 1.0 ppm level in the effluent. The problem was solved by improvement of the level control in the reflux drums. A major improvement was automation. Initially the unit was designed for manual operation, with an 8 hour loading time (one change over from loading to regeneration per shift). Following the success of the test, the unit was automised.. Results. The results of the pilot unit are presented in figure 6. The last six months are presented, before the MPPE unit was hooked up to treat all discharged water of Harlingen TC. The graph shows that very good results are obtained, compared to the aimed discharged hydrocarbon contents of mg/l. Frequently the hydrocarbon contents is reduced from approx mg/l to mg/l. To be noted however that no problems of free condensate being fed to the unit have occurred in that period. Outlet mg/l ---> HC contents in mg/l (NEN 6675) Inlet MPPE Outlet M PPE Figure 6: Last six months treatment of glycol regeneration condensed vapours Inlet mg/l ---> 5. Present test (formation-, process- and wastewater). The initial MPPE application lasted until May 97 when glycol injection on the offshore platform was stopped. Due to the lower field pressure the risk of hydrates no longer existed, and glycol for hydrate prevention was no longer required. The MPPE unit in the glycol regeneration section was not needed any more. Anticipating a new permit with more severe constraint on hydrocarbon (0.1 ppm benzene), the unit is now hooked up as to treat the total discharged water stream (formation-, process- and wastewater). The new flow is approx. 0 m3/ day with pre-treatment in the existing oily-water unit and an additional sandbed filter. The oily water unit is designed to remove hydrocarbons and sediments by gravity separation (coagulation/flocculation & flotation unit), the sandbed filter is an additional treatment to remove solids and flocculated ironhydroxide. The stream contains maximum 60 mg/l benzene, ions (mainly iron, calcium, chlorides, sulfates), sodium hydroxide (for PH stabilisation), corrosion inhibitor (injected off-shore, film forming), flocculant (to improve separation of suspended hydrocarbons and solids in coagulation/flocculation/flotation unit), glycol and other components. Test is still ongoing. An effluent of 0 µg/l hydrocarbons is realised. A problem that was experienced is fouling of the MPPE bed, for which two possible causes have been identified: - a temporary overdosis of corrosion inhibitor, intended to form an hydrophobic layer in the sea-line, may have covered the MPPE particles. Just after the glycol stop a high dose has been injected to form the initial protective film, but at current lower s, this problem seems solved. - iron hydroxide or other complexes that were formed before the MPPE bed. This seems possible with the applied combination of flocculant, sodium hydroxide and ions, for certain s and retention times that may be the case in the actual set-up. At present the system is adapted and the problems are solved. Some more testing is required to demonstrate the reliability of system. Comparison of MPPE with alternatives 1. Comparison of MPPE with steam stripping. On two Elf Petroland s offshore installations, hydrocarbon removal is accomplished with steam-stripping 1. Comparison of MPPE with steam-stripping is presented in table 5. To allow proper comparison, the parameters of an MPPE unit treating 1 m3/h (the design flow of the steam-strippers) have been calculated and are presented, as well as the Harlingen unit designed for m3/hr. stripper and MPPE unit both treat the condensed overhead vapours of glycol regeneration. Compared to steam stripping the MPPE technology is 60% lower in capex and 30% lower in opex for the case considered. For operation on offshore unmanned platforms, with low visit frequency, it should be mentioned that steam stripping operates with the treated water, whereas MPPE requires an external water supply.

6 6 H.M. PARS AND D. TH. MEIJER SPE 6577 stripper MPPE unit MPPE unit (design 1 m3/hr) (design 1 m3/hr) (design m3/hr) Investment $ 550,0000 $ 220,000 $ 20,000 Elec. Power cons. 38 kw 8 kw 20 kw Yearly cost at $0,/kWh Maint. (corr.1.5% preventive 0,5%) $ 33,300 $ 7,000 $ 17,500 $11,000 $,00 $,800 Service contract - $ 18,500 $ 32,500 Tot. Oper cost/year $,300 $ 29,900 $ 5,800 Oper. Costs $/m3 $ 5.06/m3 $ 3,1/m3 $ 1.56 /m3 Dimensions Skid size LxWxH (mtr) 2.55 x 6.25 x x 1.50 x ,00 x 2.50 x.67 Skid weight (kg) 15,500 1,700 3,00 Used exchange rate $ 1 = 2.0 nlg Table 5: Comparison MPPE unit and steam stripper for treating overhead vapours 2. Other comparisons. Several comparisons with other treatment techniques have been made. One comparison has been effectuated on behalf of the Netherlands Oil and Gas Exploration and Production Association (NOGEPA) and the Dutch government 2. An inventory was prepared of techniques for the reduction of benzene and heavy metals emissions to sea and compared on the following criteria: required area for plant construction, plant weight, cross-media aspects (use of energy and chemicals, waste production, emissions), treatment costs, capability to remove benzene and heavy metals and development status of the technique. MPPE is one of the seven (7) best ranked technologies, the only 7 that were studied in more detail, of the 52 technologies that were considered. In 1996 the MPPE technology has been evaluated and verified for treatment of offshore process water by the Orkney Water Technology Centre on request of various oil and gas companies. Tests were carried out on simulated produced water (obtained by contacting seawater with condensate), with a flowrate of approx. 0.8 m³/hr and a regeneration time of 30 minutes. The tests showed that the BTX was consistently reduced to below 5 mg/l (detection limit of the applied measurement technique), with influent s up to approx. 800 mg/l. Analysis with IR spectrometrie of typical results showed that virtually all aromatic component had been removed and much of the aliphatic hydrocarbons. The tests showed also that dispersed oil (added up to approx. 150 mg/l) did not affect the performance of the unit: BTX was still reduced to below 5 mg/l and also the dispersed oil s at the outlet was found to be much less than at the inlet. This suggests the MPPE unit could operate successfully in the presence of dispersed oil, even if the long term operating trends have not been tested. Finally the tests showed that addition of any of several typical chemical components, did not affect the performance regarding BTX removal. Further MPPE developments 1. Oil & gas industry. Examples of other proposals for off shore gas production water treatment that have been submitted are given in table 6 (150 ppm influent range) and table 7 (00 ppm ranges). From these tables it can be concluded that the steam consumption per m 3 water treated ranges from - 11 kg/m 3 for the different cases. In all these cases benzene is taken as the key component for determining the size of the unit. Of the aromatics, benzene is the most difficult component to remove. Aromatics, BTEX: benzene is key component Flow 1 m 3 /h m 3 /h Influent [ppm] Effluent [ppm] Removal efficiency [%] Weight (total) Column size d [m] h [m] cons. kg/m 3 water Table 6: Examples of off shore gas production process water proposals (150 ppm ranges) Flow rate Influent Effluent Colum size 2 m 3 /h benzene m 3 /h benzene Consumption Capex [ppm] [ppm] d [m] h [m] [kg/m 3 water] USD ~ 250 ~ 300 Table 7: Examples of off shore gas production process water proposals (00 ppm ranges); BTEX removal: dissolved and dispersed Flow rate Influent Effluent Colum size 0 m 3 /h n-heptane 150 m 3 /h benzene Consumption Capex [ppm] [ppm] d [m] h [m] [kg/m 3 water] USD ~ ~ 600 Table 8: Examples of off shore oil production process water proposals; aliphatic and aromatic removal: dissolved and dispersed For oil production water, flow rates are ranging from 0 to several hundreds m 3 /h while aliphatics (mainly) and aromatics are in ranges of hundreds of ppm. Due to these lower ranges and the fact that The results of the comparison with other techniques will be available in April 1999.

7 SPE 1 REMOVAL OF DISSOLVED HYDROCARBONS FROM PRODUCTION WATER BY MPPE 7 aliphatics are easier to remove than aromatics the column sizes remain small and the steam consumption per m 3 water is reduced to 0.3 kg/m 3 (see table 8). If the total contaminants is taken as benzene then of course column size is increasing. 2. Other industrial applications. The proven practical performance of the MPPE unit at Elf Petroland in Harlingen formed an important step forwards in further introduction of this technology for water treatment 3,. At this moment fourteen industrial units are running or under construction in the USA and Europe (see table 9). Six of these are treating process water and eight are applied in the ground water remediation area. Unit Characteristics of running commercial MPPE units and units under construction Type of water Flow rate Composition Influent Required effluent [m 3 /h] [ppm] [ppm] The Netherlands Elf Petroland PW Ar esp < 1 (Elf Aquitaine), Harlingen benzene Allied Signal, Weert PW 0.2 CHC 5 < 0.02 Shell / Akzo Nobel Site: GW 62.5 Ar 0.25 < 0.02 Akzo Nobel Fibers, Arnhem Akzo Nobel Industrial Fibers, PW 25 Solvesso 00 < Arnhem Akzo Nobel Pharma (Diosynth), GW Al, Ar, 600 < 1 Oss CHC Solvay Pharmaceuticals, GW 25 CHC 0 < 0.01 Amsterdam Germany Philips Bildröhrenfabrik, Aachen PW 5 Ar 500 < 0.05 Zürich Insurances Site: Akzo Nobel Chemicals, Mannheim Stadtwerke Flensburg GmbH Flensburg Major Chemical Manufacturer in Ruhr area United States of America Akzo Nobel Chemicals, LeMoyne (pilot), Alabama Akzo Nobel Coatings, Lenoir, North Carolina Akzo Nobel Coatings, Resins Louisville, USA Northeast Chemical Manufacturer GW 18 CHC (vinyl chloride) GW Ar (BTEX) PAH GW 0 CHC Ar Units with flows up to 500 m 3 /h have been offered 5 < < < 0.01 GW 5 - CS 2, 30 < 0.2 CHC GW 1 Ar 0 < PW 2.8 Toluene 500 < 1 ethylbenzene 0 PW 16 CHC 225 < 0.2 Abbreviations CHC = Chlorinated hydrocarbons Al = Aliphatics PW = Process water Ar = Aromatics GW = Ground water PAH = Poly Aromatic Hydrocarbons 1 ppm = 1 mg/l available for this new technique, shows MPPE is a competitive alternative to other methods applied in the oil and gas industry.. Application in other branches of industry shows the unit is operationally field proven (9 units running). Further tests and applications will have to demonstrate to what extent the E&P companies can further benefit from the MPPE technology in the reduction of hydrocarbon emissions to the environment. Acknowledgement The authors wish to thank R. Oetsma (Installation Supervisor Treating Centre Harlingen), W. Venema (Environmental Affairs Department), both of Elf Petroland, A.B. v. d Meer (Technology Manager) and F.A.M. Willekens (Engineering Manager), both of AKZO NOBEL MPP Systems, for their contribution to the preparation of this paper. References 1. Kloppenburg, M.F.C. and Venema, W.: De-oiling condensed glycol regenerator overhead vapours by steam stripping, SPE paper nr 3786 presented at the 1997 SPE/UKOOA European Environmental Conference, Aberdeen, April NOGEPA (Netherlands Oil and Gas Exploration and Production Association) and Dutch Government : Inventory of techniques to reduce emissions of benzene/heavy metals from offshore platforms, report nr , 12 November Van der Meer, A.B.: Effective and economical removal of hydrocarbons from water via extraction liquid filled macroporous polymer particles, Symposium proceedings, the 3 rd Major International Conference on Current Developments in Production Separations Systems, Aberdeen, U.K., Van der Meer, A.B. and Rakel, K.: Technical and economical evaluation of ground water remediation via extraction liquid filled macroporous polymer particles. Symposium proceedings, Grundwassersanierung, Berlin, February 1997, IWS Schriftenreihe, Band 28, Seite Table 9: Characteristics of running commercial MPPE units and units under construction Conclusion MPPE is an interesting innovative technique for the removal of hydrocarbons from waste water streams associated with the production of oil and gas. Interesting because: 1. It has proven an average hydrocarbon contents reduction from 1335 to 0,2 mg/l over 6 months in the treatment of condensed overhead vapours resulting from glycol regeneration. 2. A field scale test on treatment of the full water stream discharged by a gas treating centre is progressing successfully. 3. Comparison with other technologies, as far as yet