Abstract The objective of this study is to evaluate the alternatives for water disposal and management during coalbed methane production in remote arctic conditions. In this paper, we evaluate five methods of water disposal in the arctic environment. These methods are (1) direct discharge onto the surface, (2) controlled discharge into streams, (3) underground injection using wells, (4) concentration of produced saline water in a multiple effect evaporator, and (5) conversion of wastewater into clean water by reverse osmosis, a submerged combustion evaporator, and a pulse combustion dryer. The methods were ranked based on a set of criteria including logistics and engineering challenges of water treatment and processing in the arctic, environmental regulations, costs of installation, operations and maintenance. We conclude that surface discharge by use of ground sprinklers is not acceptable in the arctic; use of injection wells and the freeze-thaw/evaporation processes hold promise for dewatering coalbed methane wells in rural Alaska. Introduction Coalbed methane has become an important source of natural gas production in the Lower 48 States and other places around the world. One current estimate shows that coal basins in the State of Alaska contain over 1,000 to 1,500 trillion standard cubic feet of coalbed methane. Although the actual extent and magnitude of producible gas from Alaska's remote coal basins are unknown, the Alaska Department of Natural Resources postulates that if only 10% on the estimated 1,000-trillion cubic feet of methane gas is recoverable, this coalbed methane resource would triple the State's current proven conventional gas reserves (Clough, 1998). Exploratory drilling is the key to delineating the potential of the coalbed gas resources of Alaska. Coalbeds are naturally fractured, low pressure, water saturated gas reservoirs. In a coalbed methane reservoir, even though some free gas may exist in the coal deposits, the majority of the gas is adsorbed on the surface of the coal matrix. Proper dewatering of the coal seams is a critical factor associated with gas production from coalbed methane reservoirs. The saline water that is trapped within the deeply buried coal seams must be removed in order to reduce the pressure of the system, and allow the gas to desorb from the coals and flow into the wellbore. Large amounts of water production can be anticipated during exploration drilling and early life of the producing field. There are three different mechanisms by which water can be stored in coal seams. First, adherent moisture or bulk moisture refers to the free water contained in the cleat system that has a normal vapor pressure [Berkowitz, 1979; Rogers, 1994]. This is the volume of water that must be removed during the dewatering phase in the coalbed methane production operations. Second, inherent moisture or absorbed moisture is the water in the micropore system that decreases the adsoptive capacity of the coal for methane [Rogers, 1994]. Although this type of water storage is inconsequential when considering water disposal, it is extremely detrimental to the gas concentration of the coal. The other forms of water found in coal seams include chemically bound water and water of hydration.

Abstract High-sulfide gas field water with more than 100mg/L hydrogen sulfide account for 54.5% of gas wells in the Southern and Eastern gas fields of Sichuan, China. Because an effective method of High-Sulfide Natural Gas Field Water (HSNGFW) treatment has not been found, pollution problems related to HSNGFW remain, In recent years, systematic research on treatment technology for HSNGFW has been accomplished, and stripping-coagulation sedimentation process has been selected. In light of economic factors, air stripping is first used to treat HSNGFW. This results in the sulfide content dropping from several hundred to less than 100 mg/L when optimum treatment parameters are used. Then, ferric chloride, polyaluminium chloride (PAC) and newly developed SW-I are utilized for the further coagulation sedimentation. There procedures further lower sulfide content to less than 1 mg/L. Moreover, the treated water does not become yellow even if excessive ferric chloride is used. Field applications show that the results obtained generalize to in vivo settings. Introduction Gas-field water is formation water which is simultaneously brought to the surface when natural gas is produced. Because gas-field water is in contact with high-sulfide natural gas for a long time before the gas field is developed, it sulfide content is very high. Based on statistical data, the average value and maximum value of its sulfide content are 387.0 and 962.0 mg/L respectively in the Eastern gas field of Sichuan. The average value and maximum value of sulfide content are 177.2 and 763.5 mg/L respectively in the Southern gas field of Sichuan. Because gas-field water contains high sulfide and large quantities of it are produced (Sichuan gas field produces 2.0 X 106 m 3 gas field water), severe pollution problems will result if the water isn't treated carefully. A great many research projects that attempting to resolve HSNGFW's polluting problem have been accomplish for many years by institutes of the Sichuan Petroleum Administrate Bureau. The main method which has been used to treat gas field water is an oxidization process. It is difficult for this process to treat gas field water in which sulfide contents are more than 80 mg/L, so our institute has cooperated with Eastern Development Company of Sichuan, and Southern Gas Field of Sichuan in studying HSNGFW's treatment technology. Systematic research has been done and stripping-coagulation sedimentation process has been selected as the treatment of choice. Treatment results aresatisfactory by laboratory test and field application. This paper discusses key experimental results. EXPERIMENT OF RESEARCH The stripping-coagulation sedimentation process is used to treat HSNGFW. First, the wastewater is sent to a sieve-plate stripping tower and contacts adverse air currents in the tower. Air is utilized to reduce the partial pressure of hydrogen sulfide in the upper area of the water, thus making hydrogen sulfide rapidly separated from the water. Then, treated water enters the coagulation reaction unit, and clear water is obtained by the coagulation sedimentation process. The orthogonal device method is used to select the parameters of the stripping process as well as types and amounts of coagulant and decoloration-coagulant aid.

Abstract With the reality of our depleting natural resources, scientists and engineers are working to maximize the potentials of existing oil fields while minimizing ecological impacts. All divisions within the oil patch are currently investigating the diverse and vast potential bacteria have for industry. Current applications include: oil and gas exploration microbial enhanced secondary oil recovery techniques minimization of microbial induced corrosion and reservoir souring, bioprocessing of fossil fuels and environmental applications. This paper will discuss recent advances and reference pertinent sources. Introduction Biological systems have a significant role in the petroleum industry. An understanding of the positive and negative impacts of these systems can drive management decisions in all levels of the industry from exploration through production refining and eventual site closure and reclamation. This document provides an overview of some important impacts biotic systems are having in the oil-patch and how these systems are being harnessed or controlled. In addition several innovative applications for biological systems are reviewed. Some of the topics addressed include: microbial enhanced oil recovery (MEOR), bacterial contribution to reduced fluid injectivity, biological exploration techniques, reservoir souring, sour produced water treatments, microbial induced corrosion bioprocessing of fossil fuels, and environmental remediation and site reclamation applications. Exploration Microbial surveys have been successfully used to predict locations of oil and gas reserves (1.2). In general these techniques measure signature microbial populations which are often associated with anomalous hydrocarbon micro seepage from oil and gas reservoirs. One method used is called the Microbial Oil Survey Technique (MOST) developed by Phillips Petroleum. In this method anomalous hydrocarbon seeps are identified by observing fluctuations in microbial population and distribution within a survey area (I). Injectivity Problems In the late 50's and early 60's a great deal of literature was published on wellbore plugging as a result of bacterial growth. Bacteria produce high molecular weight polysaccharide polymers which are deposited by biofilms growing on the surface of a formation. This polymer deposition can cause a tremendous reduction in well injectivity. Several earl' publications link biofilm growth to reduced fluid iniectivity of wells (3.4.5). Out of this de"e1oped the popular but impractical theory that reservoirs should be formaldehyde flushed to kill the bacteria particularly the Sulfate Reducing Bacteria (SRB). Microbial Enhanced Oil Recovery There are four primary processes that contribute to Microbial Enhanced Oil Recovery (MEOR): 1) Production of Gases - These gases include CO 2 . H 2 , . N 2 and CH 4 Microorganisms down-hole produce these gases as a result of their natural metabolic processes. These gases can enhance oil recovery by increasing the wells internal pressure and decreasing the viscosity of crude oil due to solubility effects (6.7). 2) Organic Acid Production - Low molecular weight fatty acids are one of the byproduct of microbial metabolism. These compounds have application to MEOR b' their capabilities to increase rock dissolution and enhance permeability (predominantly in carbonate formations) (6). 3) Biosurfactant Production - Surfactants are often extracellular organic molecules produced by bacteria.

Abstract A previous 1 paper indicated that phenol transport was controlled by slow molecular diffusion in emulsion as the rate-determining step. Based on the transfer mechanism field dispersive model and a lab CFSTR model were built to model the time-dependent phenol concentration in emulsion and the flowing stream. Both models predicted that the produced phenol concentration monitored at an observation well will rise sharply and then decline gradually as the phenol flux emerging rom the entrapped emulsion decreases with time. For the CFSTR model phenol production is characterized by three process parameters, namely delay time t o . the specific surface area/wellbore dilution term A s , and reduced flow rate. q = v/L, which can be determined by a peak comparison method. The simplicity makes these models extremely valuable in estimating the size configuration and location of the spill, the time dependent phenol spatial distribution and the time required for phenol concentration to declineto an environmentally acceptable level These models are well suited to the water pollution problems associated with any chemical spills. Introduction In response to the bitumen spill at Cold Lake and the associated potential problem of groundwater contamination by phenol a leaching experiment 1 was carried out in the lab using produced emulsion and fresh water. The purpose of the lab test was to identify and measure the critical transport parameters for phenol under well-controlled conditions. In the study, slow molecular diffusion of phenol within the oil phase of the emulsion was identified as the rate-determining step. This transfer mechanismwas adequately described by a de-coupled second-order diffusion equation from which the diffusion coefficient of phenol in emulsion was found to be 2.2 ×10 −8 cm 2 /s at 4 °C. In this paper, a similar unsteady-state formulation is developed to model field phenol response in a flowing aquifer in direct contact With the contaminated zone. Theory Three assumptions are made in building the model: al Emulsion is modelled as a single homogeneous phase with rapid phenol transfer between thebitumen and the emulsified water phases. Phenol diffusion within the emulsion is so slow that the phenol concentration at the water-emulsion. Interface Is zero and thephenol flux Is independent of the low phenol concentration In the leaching water. This "decoupled" condition has been explicitly stated and verified In a previous paper 1 . l An offset observation well Is produced at such a low rate that it does not affect the aquifer flow pattern. In the formation c) For simplicity, the emulsion-aquifer contact is assumed to be in a semi-infinite plane. The emulsion is immobile while the aquifer is flowing at a constant velocity v. A time-dependent finite, continuous line sourcedischarging into a moving 2 dimension infinite sheet with thickness σz The geometry is shown schematically in Fig. 1 The emulsion pad on the x-z plane is in contact with fresh water flowing in the × direction. The emulsion pad has a length L in the × direction and infinite In the z direction, i.e. .. L 1 → ∞.

Abstract A leaching experiment was conducted at 4 °C to simulate the transport of phenoland chloride from a quiescent, viscous Cold Lake W/O emulsion into flowingfresh water. Test results indicated that phenol and chloride were transferredthrough different mechanisms. The rate-determining step for phenol transportwas the slow molecular diffusion of phenol in emulsion. This was successfullymodelled by a de-coupled second order diffusion equation from which thediffusion coefficient of phenol in emulsion was determined to be 2.2×10 −8 cm 2 /s. Chloride was transported through directcontact of the emulsified water droplets with the leaching water in an 82 µmthick coalescence layer. The transport was essentially complete in less thantwo days and no further leaching was observed for the rest of the experiment.The bulk emulsified water was stable with no coalescing or segregation withtime. The phenol/ chloride transport model can be readily scaled to predict themass transfer of these two chemicals from a contaminated zone into a flowingaquifer as shown in a subsequent paper. No appreciable air oxidation norbiological degradation of phenol was detected during the experimental timeframe (61 days). Introduction Similar to coal gasification, the Cold Lake steam stimulation process alsogenerates highly toxic by-products downhole In the formation. Theseby-products. Including various types of Phenols, carboxylic acids and otherhigh molecular weight acidic compounds generically classified as humicacids 1 , are weak acids. Being soluble both in bitumen andespecially In the igh pH stearn condensate these compounds are brought to thesurface by the produced water and bitumen. Cold Lake bitumen tends to form astable highly viscous water-in oil-emulsion with the produced water underminimal stress. The water content In the emulsion can be as high as 40%. Afterseparation, part of the produced water is softened and re-cycled for steamgeneration and the excess produced water, together with other harmful wastestreams are Injected through a disposal well Into an Isolated zone to avoidaquifer contamination. In 1988, a leaky seal in the abandoned well C713 at Cold Lake caused anaccidental spill of bitumen (in the form of a water-In-oil emulsion) andproduced water into a near-surface aquifer zone where the bitumen and theproduced water were in a danger of being contacted by the flowingaqulfer 2 . Although the W/O emulsion was practically immobile at theformation temperature of 4 °C due to its high viscosity, the toxic phenoliccompounds could still be expected to leach slowly into the flowing stream.posing a threat of potential groundwater contamination. To assess theenvironmental impact of phenol contamination on groundwater water was producedfrom a newly drilled well and continuously monitored for chloride and phenollevels. In the meantime, experiments were designed and conducted in the lab tostudy the rate nd mechanism of the release of phenol into fresh water.Experiments were also carried out to quantify the extent of natural degradationof phenol via air oxidation and biological pathways.

Abstract A data base containing information on up to 70 physical and chemical parameters for 118 produced water samples collected from 47 sampling locations at 18 heavy oil recovery sites in Alberta and Saskatchewan was developed by Environment Canada's Wastewater Technology Centre. One use of the data is to determine treatment requirements for produced water recycle. It was found that treatment to reduce oil, total suspended materials, iron and hardness concentrations to make the produced water amenable to recycle would be necessary for all regions sampled. Treatment for silica and total dissolved materials removal would be required for produced waters from the North Cold Lake region and Cold Lake/Lloydminster regions respectively. Introduction Water produced during the in-situ recovery of heavy oil (termed "produced water") contains a variety of dissolved and suspended contaminants, both inorganic and organic. The identification and quantification of these contaminanls are essential in determining the requirements for the treatment of the produced water for disposal, or for recycle to oil field steam generators. Therefore, Environment Canada's Wastewater Technology Centre (WTC) inijiated a sampling program in 1984'. The objective of the sampling program was to develop a comprehensive data base on the characteristics of produced waters from in-situ heavy oil recovery operations in Alberta and Saskatchewan. This paper presents the data from the produced water characterization program, and summarizes the mean concentrations and distribution of selected parameters in the major heavy oil regions of Alberta and Saskatchewan. Particular emphasis is placed on the parameters which are most relevant to produced water recycle (i.e. oil, total suspended solids, total dissolved solids, iron, hardness and silica). Experimental Details Generally, the oil industry in Alberta and Saskatchewan recognizes live distinct heavy oil producing regions: Peace River, Alhabasca, Wabasca, Cold Lake and Lloydminster. For the purpose of this paper, the Cold Lake region has been further classified inlo North, Mid and South Cold Lake regions. Also, the Uoydminster region is considered to have an additional sub-group: Macklin. The produced water characterization program entailed collection of produced water samples from all these heavy oil producing regions followed by analyses in the field and in the WTC laboratory in Burlington. One hundred and eighteen samples of produced water were collected from 47 sampling locations at 18 heavy oil recovery plants in these heavy oil regions. Table 1 shows the number or sampling sites and the total number of samples collected in each of these regions. The 47 sampling locations included 13 skim tanks, 29 well head tanks, 4 lest separators and 1 slop tank. In the case of tanks and test separators, the produced water samples were obtained from the lower section of the vessel. Table 2 lists the heavy oil recovery sites included in the study, and also indicates the sampling locations at each site. Equations (available in full paper) The sarotpling under this study started in August 1984, and continued until January 1986. The skim tanks and wellheads were sampled approximately once per month for 6 to 14 months, depending on the site, while the test separators were sampled only once during the sampling program.

Abstract Current regulations for environmental monitoring in the sour gas industry require annual reporting of soil pH. It is well known that this procedure may produce results with wide scatter, and without a clear trend over time. An alternative method, which overcomes this problem, is proposed. Rather than relying on soil monitoring to indicate the beginning of an irreversible pH drop, the new method allows the time of this occurrence to be calculated if the mean SO 2 or sulphur deposition rate is known or can be estimated. It is also possible, in atleast some cases, to identify the minerals that governinitial pH control of the soil. The method rests on kinetic measurements of soil pH change with time after acidification in the laboratory. It is recommended for monitoring, and especially for environmental recommended for monitoring, and especially for environmental impact assessment submissions to regulatory authorities. Introduction The acidification of soil, as a potential consequence of sour gas processing operations, has been of concern to industry and government in Alberta for some 20 years. While fears of groundwater pollution from trace elements, thought to be released from acidified soils and free to leach into aquifers, appear not to be a serious threat in the part of Alberta overlain by glacial till soils (1,2), there is a residual concern about yield reductions of acid sensitive crops over the long term. During this period, there have been a number of strategies to monitor or control potential acidification effects. Prior to 1980, there were continuing efforts by government and the private sector to increase sulphur recovery at sour gas plants, measure stack emissions directly and continuously, and to improve mathematical predictions of sulphur dioxide deposition; sulphur dusting was monitored by normal dustfall monitoring techniques. With the increase in the price of sulphur around 1980, there was an incentive to rapidly break up sulphur blocks by mechanical means. The attendant problem of sulphur dusting, as well as the analysis of several years of monitoring data by Alberta Environment, showed that the true sulphur dusting near these plants was seriously underestimated by dust fall methods. In some cases the discrepancies were more than an order of magnitude. In 1980, dustfall methods were abandoned in favor of regular soil pH measurement at predetermined sampline sites. In addition to these measurements, the requirement of the determination of a "buffering curve" was introduced (3), The buffering curves were dropped after a few years, when it became apparent that they underestimated the assimilative capacity of most soils for sulphur and sulphur gases. The reason for this may be the finding of Thimm and Weleschuk (1) that, after an initial acid addition, the equilibration of the soil sample requires some 15–30 days, instead of the 3 days specified in the prescribed method. At present, the pH monitoring requirement still stands, but the details of it have been expanded on significantly to include the analysis of soil sulphur, and of carbonate in some cases. Details are given in the Alberta Environment Reference Method (AERM) (4).

Abstract In petroleum production operations a large amount of waste water needs to be treated before disposal to meet environmental regulations. Fundamental mechanisms involved in the treatment of produced water are described for some of the important processes available including carbon adsorption, polymer flotation, wet oxidation and air stripping. These processes are compared in terms of their effectiveness as well as economics. It is shown that traditional waste water treatment processes such as granular activated carbon adsorption, even though effective in reducing the TOC level, is more expensive to operate than the newer processes described in this paper. Introduction Water produced with oil and gas contains various organic components of the crude oil such as paraffins, naphthenes and aromatics. In addition, organics added to protect equipment such as antifreeze, corrosion inhibitors, etc. as well as those added to enhance oil recovery such as polymers and surfactants are also found in produced water. All these organic compounds contribute to an increase in the dissolved total organic carbon (TOC) in produced water. The TOC content of produced water may be as high as 2000 ppm while the maximum TOC level permitted may be as low as 50 ppm. For example, the produced water from Shell's Peace River in-situ project contains 350 ppm of TOC (1) while produced water from Gulf of Mexico offshore platforms contains about 1,244 ppm of TOC (2) . In addition to TOC, total dissolved solids (TDS) also needs to be removed. In many cases a high concentration of TDS may be found. Shell's Peace River in-situ project water contains about 1770 ppm of TDS along with dissolved calcium and magnesium salts (1) . The quantities of water produced varies from field to field and depends on many factors including the type of process. For example different Oil-Shale retorting processes may produce between 0.1 to 22 barrels of contaminated retort water per barrel of oil produced (3) . TOC CAUSING ORGANIC COMPOUNDS Many of the TOC causing organic compounds originate from the crude oil itself. Many low molecular weight volatile organic compounds of crude oil including paraffins such as n-butane, n-pentane, n-hexane etc. and aromatics such as benzene and toluene are moderately soluble in water. Some other organic compounds are added for operational purposes. For example formic and acetic acids may be added as biocides, methanol and ethanol may be added as antifreeze and surfactants or polymeric compounds may be added as mobility control agents. The low molecular weight volatile components are also known as volatile organic carbons (VOC). The VOC's are easier to remove from the waste water than the heavy molecular weight, non volatile compounds. WASTEWATER TREATMENT PROCESSES There are many waste water treatment processes available. Some involve removal of TOC causing organics by one means or another while others involve conversion of organics into CO 2 by oxidation. The type of process to be selected depends on the type and concentration of contaminants and the degree of purity required in the treated water.

Abstract In steam based thermal recovery operations. during production, a knowledge of reservoir temperature and its trend is very important. The current temperature monitoring techniques involve either a temperature sensing element in the well, which is costly and difficult to maintain, or periodic temperature logs, which are expensive and also necessitate suspension and/or interruption of production. Produced water chemistry was successfully used to monitor the reservoir temperature in several steam stimulation tests and pilot operations in Alberta and offers a very cost effective and reliable method after the initial 30 - 45 days of production. This method is outlined and its validation by field data is demonstrated. Introduction In steam based thermal recovery operations, during production, a knowledge of reservoir temperature and its trend is very important. Because, not only it shows if the injected steam slug has been effective, but also it helps to determine if the production is suffering from lack of mobility due to decreasing temperatures or from some other factors. As such, in the cases where the steam is not confined to the pay zone it may confirm a need for remedial measures; or in huff and puff operations it may be the deciding factor for "turnaround" decisions. The current temperature monitoring techniques involve either the installation of some temperature sensing element in the well, which is costly and difficult to maintain, or periodic temperature logs, which are expensive and also necessitate suspension and/or interruption of production. Produced water chemistry (concentration ratio of two easily determinable ions) was successfully used to monitor the reservoir temperature in several steam stimulation tests and pilot operations and offers a very cost effective and reliable method for use after the initial 30 - 45 days of production. This idea of using produced water chemistry for determining reservoir temperature in steam based thermal recovery operations came from the practice in Geothermal Reservoir Engineering. In geothermal reservoir engineering work, geothermometers, based on the concentrations of various ions or silica in the produced hot water, have been in use for about two decades now for determining reservoir temperature (1,2,3,4) . These geothermometers invariably depend on the chemical equilibrium resulting from rock/water interaction at the reservoir temperature. As such, they require a certain residence time of water for the equilibrium to take place. Since, during production the actual travel time through the well is a lot shorter than the residence time during which the equilibrium takes place, the produced water chemistry still carries the equilibrium relationship which directly reflects the temperature of equilibration. In geothermal reservoir engineering, the three most tested and commonly used geothermometers are: Na/K Na-K-Ca SiO 2 The SiO 2 geothermometer is based on the solubility of quartz at different temperatures. Although, it is quite reliable, sampling and analysis of SiO 2 require certain precautions, and, in comparison to the other two, this geothermometer appears less desirable for monitoring just the reservoir temperature in heavy oil thermal recovery applications. Both Na/K and Na-K-Ca geothermometers were initially based on empirical correlations, although some theoretical background recently has been accounted for SiO (2.3,4,5) .

Introduction The recovery of oil from wells is often helped by sea water flushing, especially on offshore works. That sea water must satisfy various physical chemical characteristics to be compatible with the permeability of the source rock and with the composition of the produced water; moreover it must not corrode the steel piping network used for flushing. This implies a multiple stage treatment line; in which, roughly speaking, the first stages provide the bio-physical purification (chlorination, filtration) and the following anti-corrosion treatment (degassing, conditioning). In fact considering all the possible interferences between the treatment steps, it becomes necessary to set up a line of treatment not merely as a sequence of individual actions, but rather as an optimum combination avoiding any harmful interaction. Thus, the following list of treatments applicable offshore describes only a chronological sequence: sea water mechanical filtration or screening fine filtration anti acid conditioning. degassing by gas stripping or vacuum and complete deoxygenation anticorrosion and antifouling condtioning. The severe restrictions related with offshore works, regarding space and operational conditions, have led these last years to look for a reduction in the size of the two bulkiest units in the treatment line, namely the filters and the degassers, without jeopardizing the operation. For onshore works, space is generally not limited, but the coastal waters are often highly loaded in suspended solids, so that an intermediate stage of clarification may be necessary between the two filtration steps. As usual in matters of water treatment, a perfect knowledge of the physical-chemical characteristics of the water is mandatory, hence the considerable development of pi1ot and test plants, which enable people to collect on site all the actual data related to the water, with the advantage when operating in the open sea of being free from all the accidental contamination and pollution changes so common along the shores and in the estuaries. Characteristics of raw water to treat As examples, let us take the characteristics of Mediterranean and north Sea: Table (Available in full paper) Measurements generally show a large dispersion which does not appear on simplified charts as those given on Fig. 1,2,3. The counting of particulates is done with the Coulter Counter, which gives discontinuous measurements for particulates of 2 microns and even 1.25 microns. This device is used on pilot plants and for the commissioning of plants, while a continuous measure of turbidity in J.U. made automically and semi continuously is enough to control the operation of the filtering plants. Sea Water Chlorination Besides its bactericidal and algicidal action, chlorine distributed continuously assumes other important functions: coagulation of the phvtoplancton, an essential condition for efficient filtration downstream dislacement of the hypobromite in the sea water to bromine, enhancing its disinfection. Chlorine doses required range from 1 to 3 mg/l, out of the estuaries where pollution is higher. Chlorine can be generated as sodium hypochlorate by direct electrolysis of salt water containing at least 10 to 12 g/l of sodium chloride.

Abstract In order to raise 80% quality steam at 15 MPa from the produced water at Cold Lake, the water muse be deoiled, desilicized and softened. In this paper we outline our efforts to deoil the produced water by lowering the pH to around 4.5. A mini pilot running on 4L/min produced water was constructed to measure the effects of acidification on the reduction of suspended oil, dissolved organic matter and hydrogen sulfide as indicated in our previous laboratory experiments. It is shown that acidification is a viable and practicable alternative to using a reverse demulsifying agent. The implications of the process are outlined. Introduction It is expected that up to 2500 m 3 /h of produced water will be treated when the Cold Lake commercial plant comes on line in 1986. This treatment is necessary for reuse because a) of the need to conserve fresh water, and the difficulty of disposing of very large volumes of produced water, b) the heat value of the produced fluids can be recovered for use as feed-water to steam generators. A produced water clean-up program suitable for reuse in high pressure steam generation is necessarily comprised of deoiling, desilicizing and softening stages. Since the former operation must of necessity precede the latter, effective means of demulsification become of prime importance. Demulsification of petroleum from aqueous media by pH adjustment dates back at least forty years. One of the early patents (1) describes the process as being successful at aqueous phase pH range of 4.8 to 5.5 "by adding a suitable reagent." Although demulsification is usually attained through the use of surfactants, heat, and electric fields, the use of acids and bases can be cheaper especially when large surfactant concentrations are needed to break an emulsion or when the emulsion is "tighter" than normally encountered. Although a demulsifier chemical was found to work satisfactorily on Cold Lake produced water by destabilizing the oil/water emulsion, it was decided to look for an alternative process for two main reasons: as a back-up technology especially if the amount of dissolved organic matter becomes critical in raising high pressure steam it would assist in reducing hydrogen sulphide concentration in the water. It must always be borne in mind that. aside from suspended oil, the produced water contains dissolved oil and organic matter which may cause deposition problems in the generator if not reduced to sufficiently low levels. ACIDIFICATION – WORKING PRINCIPLES Generally speaking, acidification can be conceived as increasing the ionic strength and compressing the electrical double layer around the oil droplets thus allowing coagulation and facilitating coalescence. It also displaces the ionization equilibrium of carboxylic acids and phenols towards their molecular forms which are less hydrophilic than their ionized states. Acidic species of high molecular weight thus lose their tendency to accumulate at the interface where they strongly stabilize the emulsion. Heavy crudes and bitumens are more susceptible to these effects than paraffinic types for three main reasons: