Water Softening of High TDS Produced Water

Recycle of high hardness, high TDS(total dissolved solids) waters (hardness>1000 parts per million (ppm) as CaCO3 and TDS>10,000 ppm) water for steam generation or other reuse such as irrigation or drinking water is very expensive. Silica content is usually above 250 ppm in such waters which can cause problems in steam generation and with desalination. Many of the these high hardness waters are oil field produced waters but there are other processes which generate waters which require additional treatment. For those high hardness waters, hot lime or hot caustic, followed with strong acid/weak acid resin or just weak acid resin softeners are used in conjunction with oil field steam generation. Treatment of these waters for irrigation or drinking waters involves thermal desalination or reverse osmotic(RO) treatment with biological control. Silica removal is not required for normal wet steam genera tion. However for desalination operations, whenever the feed water concentrate is to be used for the steam generator feed and the fresh water sold, silica removal is required to aid in scale formation in desalination plus silica level in the concen trate water. The known limits on the concentrate feed water to a wet steam generator are 500 ppm silica and 25,000 - 30,000 ppm TDS of total water ions, based on soluble salts solubility. Silica removal is also required in other waters such as 210,000 ppm TDS water containing sodium carbonate where silica removal is required prior to crystallization of sodium carbonate crystals. In the softening test work, the high solids addition and disposal associated with a hot lime system was not desired so alternatives were investigated. In addition, better silica removal than silica absorption on magnesium hydroxide or alumina or aluminum was required. Steam stripping of the high pH water and removal of the precipitates using a ceramic crossflow filter for which a special crossflow filter back pulse unit for cleaning was developed. Temperature and pH were increased prior to the steam stripping to decrease steam condensation and drive the reaction. When silica removal was required, a bed of alumina or aluminum was used at the high pH and temperature to put aluminum into solution so aluminum silicates were removed with the hardness precipitates. The solids often contain some oil when using oil field waters so an odor chemical was also developed for the microbiological soil remediation site. The steam stripping was tested first in a countercurrent mode in a stainless steel column clad with a Hastelloy C 22, packed with stainless steel packing. The second test was by injecting the steam on the outside circumference of a crossflow ceramic microfilter in a cocurrent mode with flashing in a exit vessel. In the countercurrent tower operation, the control of the equili brium of the carbon dioxide, the carbonate and bicarbonate at the top of the tower was more difficult than when contacting with the microfilter. With the microfilter, the equilibrium approach was not a large concern as fresh steam was contacting the water and was then flashed. However the exact control of the steam to water ratio was more difficult in the second case. Both thermal desalination and RO were pilot tested with waters from 10,000 to 36,000 ppm TDS to produce potable water with a quartz ultraviolet light for biological (disinfection) control. For wet steam generation, the field produced waters(10,000-24,000 TDS) were tested using strong acid/weak acid resin softening with no silica removal in a 1 MM BTU/Hr wet steam generator.1 The overall operational costs were less than normal sequence of processes mentioned in the literature while the capital costs were in the same range. Patents were obtained on the(1) steam stripping softening, (2)silica removal,(3) back pulse on the micro filter and(4) the odor chemical. A patent on the sour gas treatment is pending.

High pH dissolution of quartz, aluminum-bearing minerals and their mixtures is important in several oil production processes. Rock consumption of caustic is a major limiting factor for successful alkaline waterf looding. Chemical dissolution of the gravel pack and near well-bore rock are problems in both steam and caustic flooding. Mineral dissolution is probably a contributing mechanism for the reported success of potassium hydroxide treatments in effectively and permanently stabilizing clays. The presence of even small amounts of soluble aluminum can greatly reduce the kinetics of silica dissolution in near neutral and high pH solutions. Earlier research has shown that clay minerals frequently dissolve more slowly than quartz or amorphous silica. Although these clay minerals may not themselves contribute substantially to caustic consumption, they release aluminum ions which may slow dissolution rates of other silaceous minerals. Most previous dissolution experiments have examined dissolution of single minerals. In this paper we report experimental dissolution results for quartz, kaolinite, and α-alumina and their binary mixtures in 0.1 N NaOH solutions at 70°C. Results for quartz and kaolinite in 0.1 N KOH are also described. When soluble aluminum or aluminum-bearing minerals are present, the dissolution rate of quartz is significantly reduced. Adding soluble silica can have a similar affect on kaolinite dissolution rates as measured by production of aluminum and silicon. In-congruent dissolution occurred when kaolinite solutions were shaken.

ABSTRACTA pilot waste water recycling project was conducted at Conoco's Cat Canyon Field, Santa Barbara County, California. The project developed a process to treat the field's high silica produced water for reuse in steam generation.The treatment of produced water was found to be more complex than conventional fresh water treatment techniques currently employed in the field. Oil, solids, and iron were each removed from the process stream before softening the water. Softening was then conducted in a two-stage process consisting of strong acid primary softeners followed by a weak acid polisher. Continuous operation of this pilot recycling plant virtually eliminated scale deposition inside a test steam generator.

The high temperature and pressures encountered in steam flooding have necessitated the use of premium and, in many cases, unique equipment. Field results from three steam flood installations give an insight as to the performance of high-pressure equipment when subjected to the accompanying high temperatures. Performance of the equipment and factors entering into the selection of equipment for steam flood installations are discussed. Introduction Recovery of low-gravity, high-viscosity crude oil from relatively shallow reservoirs is becoming feasible through the application of steam flooding. Pan American Petroleum Corp. initiated a pilot steam flood with a 5.36 million Btu/hour, 1,500-psi steam generator at the Winkleman Dome field in west central Wyoming in March, 1964. After one year of operation, this steamer was replaced with a larger unit capable of 12 million Btu/hour at 2,500 psi. Two other pilot steam flood projects were started at that time using 12 million Btu/hour, 2,500-psi steam generators, one at the Salt Creek Shannon field and another at the Fourbear field, both in Wyoming. This paper discusses the equipment used in high-pressure steam flooding and reviews some of the problems encountered in the application of the equipment. Where determined, a suggested solution is presented. The discussion follows the conventional flow of water and steam: water treating. steam generation, steam control and transmission, wellhead and subsurface equipment, and instruments and data collection (Fig. 1). The steam generation and transmission equipment discussed in this paper has been designed for 2,500-psi saturated steam service. Maximum operating conditions have been 1,875 psi and 626F. Water Treating Successful operation of steam generation equipment depends primarily upon a good source of water combined with an effective water treating system. Experience gained in the past two years of pilot steam flood operation indicates the majority of steamer down-time is caused by water treating problems. The quality of raw water dictates the amount of treating required; therefore, it is imperative that the best water available should be used. Some criteria for a good quality raw water are: the water should be free of oil or filming amines; dissolved gases such as O2, C2 and H2S, should be absent or at least present only in trace amounts; total hardness should be low; and suspended solids concentration should be low. Since the quality requirements of water used in a single-pass steam generator are extremely stringent, it is unlikely that the available raw water can be used without some form of treatment. JPT P. 1525ˆ

Summary A laboratory study was undertaken to investigate the potential of improving tertiary oil recovery of intermediate to heavy oils by use of caustic soda as a chemical additive in waterflooding and steam flooding. Seven aspects of this study were examined. In all cases, the process was started with a fresh and similarly water-and oil-saturated sand pack. The sand pack was obtained each time by packing sieved glass beads in the range of 35 to 48 mesh size. The cases include:weak caustic soda flooding (0.001 g/cm3) of previously waterflooded sand pack followed by steam flood,caustic steam flooding immediately after waterflooding the saturated sand pack,a process similar to Case 2 followed by a second cycle flooding with a higher concentration (0.0015 g/cm3) of caustic steam,waterflooding a saturated sand pack followed by conventional steam flood,caustic flooding of water-saturated sand pack followed by caustic steam flood.cyclic steam/caustic flooding of water-saturated sand pack, anddetermination of the optimal temperature for caustic flood. The results of these experiments showed that as a chemical additive for tertiary waterflooding and steam flooding, caustic soda substantially improved oil recovery of mildly acidic 18 deg. API gravity oil over conventional waterflood and steam flood. The performance of cool caustic flood as a tertiary recovery mechanism was good at high residual oil saturation and poor at low residual oil saturation. Caustic hot water and caustic steam flooding recovered 14.5% more original oil in place (OOIP) than conventional hot water and steam flooding recovery under similar reservoir conditions. Introduction Steam Flooding has been demonstrated to have considerable potential for improving oil recovery from fields containing high viscosity oils. A troublesome aspect of steam flooding, however, is the tendency of steam to override the bottom half of the formation, which becomes essentially flooded by hot to warm water (condensate). It has been observed in most steam floods that the top half of the formation swept by steam has its residual oil reduced almost to near zero, while substantial residual oil saturation is observed in the water-swept portion of the reservoir. Some investigators have suggested that oil/steam ratio could be reduced significantly if suitable additives were found that would reduce oil saturation in the lower portion of the formation overridden by steam and mostly swept by hot to warm condensate. The wettability of petroleum reservoir rocks and its effect on the displacement of oil by water are still controversial subjects of a large body of literature. Craig gave an excellent review of developments in this field. Many recent investigators generally agree that preferred wettability may not be a discrete value function, oil-wet or water-wet, but may span a continuum between these extremes. Recently it has been noted that a large number of reservoirs are more oil-wet than water-wet and that causing a reservoir to become more water-wet by chemical means during the course of waterflood results in an increase of oil recovery compared with an altered displacement by water alone. Leach et at. demonstrated this using a refined oil containing an amine to stimulate an oil-wet system, and an aqueous acid solution to reverse wettability. Mungan, to using NaOH solution, was able to alter the wettability of each crude-oil/brine/sand system. Under certain circumstances, therefore, caustic waterflooding has been found to increase oil production byflow interfacial tension (IFT) displacement,rigid film breaking, andfavorable wettability alteration. As early as 1920, Flyeman patented a process using sodium carbonate for separating bitumen from sands. JPT P. 1817^

A chemical equilibrium flow model that incorporates conductive and convective heat transfer is presented for continuous alkaline steam flooding of acidic oils in linear systems. The chemical aspect of the model consists of the chemistry of acid hydrolysis at elevated temperatures, resulting in in-situ generation of surfactants and the interaction of the alkali with reservoir rock at high temperatures, leading to alkali consumption. The adsorption and desorption of active species at the interfaces are expressed as ionic processes using the Gouy-Chapman theory of diffuse double layers. The model predicts that the surfactant slug generated in-situ in alkaline flooding increases with temperature and that, while adsorption of surfactant increases with temperature, there is a net increase in surfactant slug size to recover otherwise immobile oil, leading to higher oil recovery.

Field testing has confirmed the newly-discovered Modified Hot Lime Process (patent pending) as a significant improvement over existing precipitation softening options. A Permian Basin produced oilfield water containing 2000 ppm hardness, 500 ppm sulfides, 10,000 TDS and 200 ppm oil is successfully being converted to steam generator quality feedwater. Demonstrated success of this new process may open-up steam injection projects in other areas where available water supplies are not economically treatable using existing technologies. Alkali consumption and sludge production have been reduced by 50% with the Modified Process when compared to conventional Hot Lime. In addition, alkali consumption by entrained CO2 is eliminated. Many hot lime softeners currently in service can be inexpensively converted to this more efficient process. This paper also discusses use of two alkalis to improve performance of the precipitation process and effectiveness of CO2 as the stripping gas for sulfide removal. Introduction Economic treating of water is one of the most critical obstacles to achieving a successful steam injection project. Because of strict environmental regulations and lack of available fresh water, produced oilfield water is typically used as source water for steam generation. For most steamfloods, quality of the produced water is fair - sufficient treating can be accomplished using typical oil removal and softening techniques. In many west Texas oil fields, however, produced water quality is much worse than that currently used for steam generation. Hardness and sulfide levels are ten times the average for California steamflood source waters. In fact, this pilot's water source may be the poorest ever used for conversion to steam generator feed. This, along with the general absence of typical heavy oil steam targets, accounts for the historic reluctance to pursue steam projects in the Permian Basin. However, Marathon has identified a unique thermal opportunity in a large west Texas oil field. Realizing the significant reserve and economic potential of a thermal project in this area as well as potential pitfalls associated with treating poor quality water, Marathon Oil Company embarked on an operational steam field test in 1995. The test consisted of installing a 5,000 BWPD oilfield water purification facility, three single pass waste heat steam generators and drilling one steam injection well. According to published guidelines, all oil, suspended solids, hardness and sulfides should be removed from the oilfield water in order to prevent damage to steam generator tubes. Since primary challenges associated with this water deal with sulfide and hardness removal, the majority of this paper will focus on removal of those two constituents.Hardness Removal - Many different commercial hardness removal systems were explored for the project: Strong Acid/Strong Acid Ion Exchange, Strong Acid/Weak Acid Ion Exchange, Weak Acid/Weak Acid Ion Exchange, Hot Lime Soda, Cold Lime Soda and Thermosoft. A detailed review of each of these processes is given by R.J. Jan and T.G. Reed, Jr. Economic analysis of the above processes identified hot lime softening as the most attractive of the existing technologies for the high hardness produced water to be used. Following additional lab and field testing, a new, more efficient process was discovered and commercially demonstrated. The process is a variation of the standard hot lime process and will be discussed in more detail herein.Sulfide Removal - To effectively remove sulfides, water's pH must be reduced to convert ionic sulfide to hydrogen sulfide (H2S) gas. A packed column is then used to strip the H2S. This paper discusses an uncommon technique of using CO2 for both pH reduction and stripping instead of the typical approach of using acid for pH reduction. The benefit to downstream softening is explained. P. 491^

This paper reviews the application of a 50MM Btu/hr oil field steam generator operating on low Btu produced "waste" gas. The generator is the first 50MM Btu/hr oil field steam generator permitted in the Los Angeles Basin since the 1980's. Stack emissions of less than 7 ppm NO × corrected to 3%O2 were achieved using a Low NO × burner without selective catalytic reduction with ammonium injection (SCR) and without flue gas recirculation (FGR). (Future references to NO × ppm will be corrected to 3% O2). Introduction Tidelands Oil Production Company is an operator of 922 wells consisting of mature water flood and steam flood wells. These wells are in the Wilmington Oil Field, Los Angeles County, California. The steam generator is on the Parcel "A" Lease in the Port of Long Beach (fig. 1). In the 1960's, wells on the Parcel "A" Lease were steam stimulated using the huff and puff method with portable steam. Results of steam stimulation were encouraging and showed that a steam flood should be economical. The installation of a steam source is part of a pilot program to develop a steam flood on the Parcel "A" Lease. Tidelands produces approximately 2000 MCFD of gas with a higher heating value of 550 to 1000 Btu/scf yielding approximately 62 MM Btu/hr of fuel. The fuel contains 40% to 60% CO2 and has no commercial value. Varying composition, varying delivery pressure, and varying volume of the produced gas posed unique design and operational considerations that will be discussed later in this paper. A steam requirement and the availability of waste fuel were a good fit for a S50MM Btu/hr steam generator. A used steam generator was rebuilt and retrofitted with a low NO × burner designed to burn fuel with a higher heating value of 550 Btu/scf. The generator was rated at 58,500 lb/hr with a maximum working pressure of 1700 psi. The South Coast Air Quality Management District (SCAQMD) is the environmental regulatory agency with jurisdiction in the Los Angeles Basin. The SCAQMD is considered the toughest district to permit a major emitting source. The SCAQMD permit to construct required that the generator stack be monitored real time with a continuous emission monitoring system(CEMS). Maximum NO × allowed by the permit is 15 ppm. The steam generator was delivered in February 1996 and system check out started in June 1996. The emissions data presented in this paper were taken in October 1996. Data was independently tested and verified by World Environmental with the results of these tests presented to SCAQMD as required by the SCAQMD permit. Air Quality Regulations and Permit Requirements The applicable emission compliance requirements are defined in the permit to construct that is issued by the SCAQMD. The governing regulation that sets the permissible NO × level is Rule 11.46 of the Rules and Regulations for SCAQMD. The SCAQMD permit to construct allows this generator installation a maximum permissible level of 1 5 ppm of NO ×. Regulation 13 of the Rules and Regulations for SCAQMD are the guidelines for Best Available Control Technology (BACT) and cross references the SCAQMD document titled BACT Guidelines. These documents set guidelines for what type of pollution control equipment must be installed. The SCAQMD also takes into account past permit requirements for the same type of equipment and current pollution control technology. For natural gas combustion, SCR has been established as the BACT on many systems and is required for many permits. A SCR system requires ammonium injection. Installation is costly, operating costs are high, and it requires the use of hazardous substance(s). For our project, a Low NO × burner with FGR was allowed by the SCAQMD. This was the first steam generator of its type to be reviewed by the SCAQMD since the 1980's and establishes BACT in the Los Angeles Basin. Rule 20.12 Appendix A Chapter 2 of the Rules and Regulations for SCAQMD requires that stack emissions data be measured real time and daily reports be sent to the SCAQMD via a modem using a continuous emission monitoring system (CEMS). NO x, wet O2, dry O2, stack gas flow rate, steam output, fuel flow rate, and FOR recirculation rates are measured as required by Rule 20.12. In the past, CEMS were costly sensitive analyzers that required frequent cleaning and repair. Dependability was poor and equipment failure was common. The equipment required constant attention by field personnel and the manufacturer. Going into this project, I expected a long CEMS start up and debugging. P. 237^

The influence of phosphate treatment, low-AVT and high-AVT on the corrosion resistance of the heating tubes of steam generators (SG) with U-shaped tube bundles is discussed. Phosphate acts against possible local pH-value shifts due to the ingress of salts into the SG water. This inhibits denting in steam generators with rigid tube support plates made of carbon steel. Laboratory investigations show that phosphate has an inhibiting effect on pitting. In addition, phosphate acts against the formation of copper ions, which are discussed as the oxidizing agent leading to pitting phenomena in SGs of pressurized water reactors. The oxygen content of water in the SG is influenced by phosphate only as a function of the differences in the solubility behaviour of demineralized and saline water. Because of uniform corrosion in the area of tube scales containing phosphate (so-called "wastage corrosion"), numerous plants outside the Federal Republic of Germany have been converted for operation using AVT. This reduces the safety margin with respect to the aforementioned corrosion phenomena. Such plants must therefore be operated strictly under reducing water conditions. Hydrazine acts as a reducing agent to inhibit the oxidizing effect of O2. At high temperatures, hydrazine also combines directly with oxygen. With high-AVT treatment, the SG water contains N2H4 concentrations greater than 100μ/kg. A prerequisite for the use of high-AVT is that there are no copper materials in the secondary circuit so that copper ions as oxidants can be ruled out. With low-AVT, the SG water usually contains only a very low concentration of N2H4. Low-AVT therefore has neither the advantages of phosphate treatment nor those of high-AVT and exhibits the lowest safety margin with respect to anodic potential shifts in the SG water. The redox potential of the fluid due in each case to the interaction of the various oxidizing and reducing agents present in high-temperature water and the resulting corrosion potential of the tubing cannot be determined using the methods of water analysis. To determine such potentials, it is suggested that, regardless of the water treatment, potential probes be installed in the component concerned. In this way an increased tubing corrosion hazard caused by oxidizing water conditions in the SG can be detected in good time. This technique is described here.

Champlin Petroleum Co.'s pilot test of the CO/sub 2/ immiscible-drive oil recovery process has been in continuous operation for 50 months in the Wilmington field, Tar zone, Fault Block III Unit reservoir. First CO/sub 2/ injection was in March 1981. The pilot, containing 1,700 acre-ft (2.1 x 10/sup 6/ m/sup 3/) includes four injection wells and three producers. The reservoir is an unconsolidated sandstone at 2,500 ft (762 m) that contained 920 bbl/acre-ft (0.12 m/sup 3//m/sup 3/) of 14/sup 0/API (0.97-g/cm/sup 3/) crude oil at the start of the CO/sub 2/ pilot. This is the first test of immiscible CO/sub 2/ tertiary oil recovery in a late-life waterflood reservoir. Cumulative water injection before start of the CO/sub 2/ pilot was three PV's. Through May 1, 1985, 2.1 Bcf (60 x 10/sup 6/ m/sup 3/) of produced and purchased CO/sub 2/ had been injected intermittently with water. Cumulative purchased CO/sub 2/ through May 1, 1985, was 1.5 Bcf (42.5 x 10/sup 6/ m/sup 3/). Each of the three producing wells has shown stimulated oil response with production rates increasing an average of seven-fold. One producing well, converted from a former water-injection well, produced 100% water for 5 months after first CO/submore » 2/ injection. The well has since steadily increased in production to over 40 B/D (6.4 m/sup 3//d) of oil. All produced oil is considered incremental oil since the pre-CO/sub 2/-flood oil rates were essentially at the economic limit and the wells would soon have been plugged and abandoned.« less

American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc. Abstract A critical review of steamflood mechanisms is made based on published laboratory and field data. Qualitative and quantitative discussions are presented on the major steamflood mechanisms presented on the major steamflood mechanisms including steam drive and in-situ solvent drive; viscosity reduction and thermal permeability variation; and thermal permeability variation; and thermal expansion and gravity segregation. Interaction of these mechanisms is responsible for high steamflood displacement efficiency, enhanced oil production rate, and poor vertical steam conformance. A basic poor vertical steam conformance. A basic understanding of these mechanisms should assist petroleum engineers in the design, analysis, and evaluation of steamflood operation and mathematical simulation. Introduction Steamflooding is a proven secondary or tertiary method for recovering oil economically from some heavy oil reservoirs. It has potential applications as a secondary or tertiary method to recover light oils. Therefore, it is conceivable that steamflood mechanisms responsible for displacing oils or helping to displace oil from the reservoirs are of great interest and importance to the industry. Steamflooding is a complex oil displacement process involving simultaneous heat, mass, and fluid transports. Although steamflood mechanisms associated with these complex transports may be well known to some people, reports on these mechanisms have been scattered in many publications over more than fifteen years and have not been readily available for most petroleum engineers. Willman, et al. reported experimental studies on the effect of some steamflood mechanisms, such as viscosity reduction, thermal expansion, and steam distillation, on the oil recovery. Johnson, et al. Quinones and Wu and Brown reported steam distillation of crude oils in porous media. Farouq Ali and Alikhan and Farouq Ali reported the effect of solvent slug on the oil recovery by steam and hot-water floods. Gates and Ramey, Dingley, and Braden presented the thermal viscosity reduction of some crude oils. Several articles reported the temperature effects on the oil-water relative permeabilities and on the absolute permeability.

The Wilmington Steamflood of Union Pacific Resources Co. (UPRC) at Long Beach, CA was initiated in 1989, in a previously waterflooded reservoir. Average initial reservoir oil saturation, at the start of the steamflood, was 35%. Field production data were studied, to derive an overall energy balance for the steamflood, to calculate the steamflood capture efficiency and predict further steamflood performance. Heat-losses due to produced fluids were calculated. Predicted production schedules from the model were history-matched with field production data.' All steamflood calculations were carried out using a PC-based spreadsheet program. The major results were as follows: The capture efficiency of the Wilmington steamflood was calculated at 60%. This is an acceptable value, taking into account the fact that the reservoir had previously been waterflooded to a low oil saturation of 35%.The calculated heat balance showed a high heat-loss, not only to adjacent formations, but also through produced fluids. Of the cumulative heat injected up to the time of the study, 21% had been lost to vertical conduction and 21% through produced fluids.Predicted production schedules indicated that up to 43% of the oil in place (at steamflood initiation) would be recovered by the steamflood.

Oilfield "produced" waters usually contain high hardness, and high dissolved solids along with some alkalinity. The problem of disposing of these waters and the need for huge volumes of water for alkaline water flooding and steam generation for steam floods, necessitate the softening and reinjection of softened produced water. Water for alkaline and steam flooding needs to be softened to almost zero hardness to prevent plugging during injection and to prevent scaling on boiler tubes. Softening of high TDS and high hardness waters requires a very selective resin with high operating capacity such as a weak carboxylic acid-type ion exchange resin. Conventional softening with strong acid resins would not work under these conditions. This paper discusses the different processes in softening oilfield produced water. Data obtained in softening varieties of produced waters from different oilfields in California will be presented. Performance characteristics of weak cation exchange resins and their chemical regenerant requirements will also be discussed.

In Trinidad, oil production started just over 100 years ago and steam flood operations started just under 50 years ago. In steam flood operations, oil recovery by steam distillation can be in the range of 5 to 60 % and therefore requires separate experimental and mathematical studies for accurate steam flood predictions. The steam distillation apparatus required for the experimental study can be quite costly. In order to conduct steam distillation studies in the Petroleum Studies Unit in Trinidad, a steam distillation apparatus was designed, fabricated and tested to perform these studies, as an integral part of the experimental steam flood studies on Trinidad crude oils. The apparatus consist of a positive displacement pump, a steam generator, a steam distillation cell, a temperature measurement and control system, a back pressure valve and a condensing and liquid collection system. The steam generator and steam distillation cell were fabricated in-house, from stainless steel and were designed to conduct steam distillation studies at a safe working pressure of 4.654 MPa and temperature of 260 °C. From the operational procedures outlined in this study and from repeat test runs conducted at 100 °C and 260 °C steam distillation results were reproduced with differences of less than ± 4.0 % between the original and repeat runs. Details of the apparatus design and operational procedures from this study can provide a useful guide for other Researchers on crude oil steam distillation studies. Key words : Steam distillation apparatus; Design; Operational procedures; Oil; Steam; Temperature; Pressure; Trinidad

The Shannon Reservoir at the Naval Petroleum Reserve No. 3, Natrona County, Wyoming, has some six years of Thermal Steam EOR history and has been expanded to six patterns with four 50 MMBTU/hr natural gas fired steam generators in operation. During Pilot Steam Flood operations, scale formation at producing wellbores and tubulars became excessive. One pulling unit was employed full-time to mechanically cleanout tubulars and replace scaled pumps in 21 original Pilot producers. Acid Cleanouts and scale inhibitor squeeze jobs were expensive and short lived. Due to planned fieldwide expansion of the Steamflood Pilot, the magnitude of costs associated with production side scale treatment would adversely impact project economics. A means to inhibit scale formation insitu from the injection to production wellbore was investigated. Based on scale inhibitor application in steam generator tubulars in California, a successful chemical formulation was known to be available. Research determined that the product could be injected with the softened generator feed water and remain in the liquid phase of the injected steam. Sufficient residual was expected to be seen at the producing wellbore to prevent scale formation. A new analytical method had recently been developed to allow measurement of the chemical residual in the produced water. The chemical formulation had not been used in this type of application prior to use at NPR-3. This paper details the successful application of scale inhibitor injected in the steam generator feed water, transported through the reservoir in sufficient concentration to inhibit scale and to provide measurable residual for confirmation. The economic feasibility is shown to be dramatic with long term reduction in pulling costs, elimination of ineffective acid cleanouts and scale squeezes, and minimal chemical cost to maintain residual at producing wellbores.