About 60 of the worlds oil proven reserves is found only in carbonate reservoirs while 90 of these are observed to be neutral to oil-wet Akbar et al. 2001. Relating few factors such as reservoir heterogeneity, reservoir management, fluid type and drive mechanisms, experts observed that more than 50% of the oil trapped only in carbonate rocks is yet to be recovered (Sun and Sloan, 2003).
Throughout the life cycle of the field, three stages of oil recovery are typically executed: Primary recovery uses the reservoir natural energy; secondary recovery mainly uses injection of water and gas for maintenance of pressure; while tertiary recovery or enhanced oil recovery (EOR) uses an injectant (Yousef et al. 2010). Till date, different techniques of EOR have been applied to carbonate formations to improve recovery.
For many decades, since the accidental water injection in Pithole City, Pennsylvania, 1865, waterflooding has been the almost widely used oil recovery method (Lewis, 1961). Historically, the quantity of the water was considered more valuable than its quality. While the target of any waterflood reservoir management is to ultimately maximize the oil recovery, researchers later realized that water quality was very crucial to be considered as the quantity. Good quality water ought to be compatible with formation water, chemically inactive, free from solids suspension and organic matters. It has been well documented that seawater and aquifer water seems to be mainly the two sources of water for waterflooding (Alotaibi and Nasr-El-Din, 2009).
Over the years, the role of the chemistry of the injection brine as well as its influence on oil recovery have received less attention because waterflooding has just been viewed simply as a physical process to maintain reservoir pressure and drive oil towards the producing wells. Recently, extensive research, on both carbonate and sandstone reservoirs, has shown favorable effect of tuning salinity and composition of the injected brine on Crude oil/brine/rock (COBR) interactions, rock wettability alteration, microscopic displacement efficiency enhancement, and oil recovery improvement (Yousef et al. 2010). However, owing to the interaction of COBR, brine chemistry is considered as an important factor that may help in improving oil recovery.
The notion of injecting low salinity water into petroleum reservoirs has been addressed since the 1960s where researchers (Martin, 1957; Bernard, 1967) began injecting fresh water into core samples. Bernard (1967) observed an increase in oil recovery when injecting fresh water and ascribed the improvement to enhanced microscopic sweep efficiency which is stimulated by swelling of clay and pore throat plugging (caused by migrating fines). Swelling of clay reduces the pore space disposable to flow of oil and water, whilst by pore throat plugging; new flow channels are generated thereby increasing oil recovery. Nevertheless this work did not capture the attention of the petroleum industry.
Low salinity waterflooding started gaining considerable interest in the 90's by researchers at the University of Wyoming, studying the effect of brine, crude oil, and mineralogy on wettability (Jadhunandan, and Morrow, 1991; Yildiz and Morrow, 1996; Tang. and Morrow, 1997). Tang and Morrow acknowledged that injecting low salinity brine improves oil recovery in only clay rich core but not in clay free cores. They discovered fines in the effluent during successful low salinity waterflooding experiment. Extensive research work in sandstone reservoirs has developed this idea into an emerged trend (Jadhunandan and Morrow, 1995; Tang and Morrow 1999; Tang and Morrow 2002; Zhang and Morrow 2006; Zhang et al., 2007).
The effectiveness of Low salinity water for improving oil recovery has high potential compared with high salinity waterflooding. Laboratory waterflood (Yildiz and Morrow, 1996; Zhang and Morrow, 2006; James et al. 2008; Morrow and Buckley, 2011) and successful field tests (Webb et al. 2004; McGuire et al. 2005) have showed that low salinity waterflooding can improve the oil recovery in sandstone reservoirs. However, low salinity effect has not been thoroughly observed for carbonates. One key argument that low salinity might not work effectively in carbonates is that clay minerals play a primal role in the low salinity effect, and clays are lacking in most carbonates (Lager et al. 2008a). RezaeiDoust et al. (2009) presented another argument that dissimilar chemical mechanism may be involved for difference in the low salinity effect: which is the crude oil adsorption onto positively charged calcite surface and negatively charged quartz surface. The effect of brine chemistry on carbonates has therefore being tagged smart waterflooding.
Even after few research, (Austad et al., 2005; Strand et al., 2008; Yousef et al., 2010; Gupta et al., 2011; Yousef et al. 2011; Yousef et al., 2012; Romanuka et al., 2012; Zahid and Shapiro, 2012; Winoto et al., 2012), smart waterflooding in carbonates remains quite controversial. Great success was achieved in oil recovery by injecting seawater into the highly fractured mixed-wet Ekofisk chalk field, Norway (Sulak, 1991). Zhang et al. (2006) reported the impact of potential determining ions as one of the element for wettability alteration. Another report was made by Yousef et al. (2010) considering the reduction of the salinity of the brine and its impact on recovery although a contrary report was given by Zahid and Shapiro (2012). The mechanism(s) responsible is poorly understood, the reliability of published results and the technology's initiation to the field is questioned. Withal, optimization of waterflooding by manipulating brine salinity and chemistry is not feasible attributable to the lack of comprehending the primary mechanisms and all other factors that may influence the oil recovery.
1.2 Motivation
Chalk being a pure biogenic substance, having much larger surface area compared with limestone (2 m2/g for chalk compared with ≈ 0.3 m2/g for limestone), shows a higher recovery in several experiments conducted. Albeit, chalk and limestone have similar chemical composition, CaCO3, their response to the potential determining ions present in seawater can be different concerning wettability modification.
Subsequently, similar interactions behavior - improvement in oil recovery and reduction in residual oil saturation - have been reported with limestone rocks (Strand et al., 2008). Many reports have shown additional oil recovery by tuning the brine salinity and chemistry of the injection water. The initial results are promising; many recovery mechanisms are proposed, but still many uncertainties remain about the mechanism and the roles of the water chemistry. This work shall therefore contribute to the understandings of effect of potential determining ions on oil recovery by smart waterflooding.
Nonetheless, smart waterflooding is alluring because it can offer much recovery benefit, it's relatively low-cost and relatively simple compared with other EOR techniques. It is always a wonder how the injection of seawater into the Ekofisk chalk is such a tremendous success in oil recovery, which is now estimated to approach 55%. Ekofisk is mixed wet, highly fractured with low matrix permeability of about 2 mD, and the reservoir temperature of 130oC.
Likewise, our field has reached the waterflooding maturity stage and considerable amount of oil is still left in the formation. Critical examination of the field properties with earlier experiments conducted (Strand et al., 2006a; Zhang et al., 2006; Yousef et al., 2010; Gupta et al., 2011; Yousef et al. 2011; Yousef et al., 2012; Romanuka et al., 2012; Zahid and Shapiro, 2012; Winoto et al., 2012) has made smart waterflooding a promising EOR methods for this very tight Carbonate reservoir field.
1.3 Objectives
Several papers have verified that seawater can act as a "smart water" to improve oil recovery from chalk by wettability alteration towards more water-wet conditions by both spontaneous and forced imbibition (Strand et al., 2006a; Zhang and Austad 2006; Zhang et al., 2006) and mechanism suggested (Zhang et al., 2007). In a highly preliminary study, Strand et al. (2008) showed that the surface reactivity of reservoir limestone cores towards Ca2+, Mg2+ and SO42- had a similar trend as that of the chalk surface.
This research study is aimed at evaluating the combine impact of brine salinity and composition to enhance oil recovery. It will also present a systematic procedure for analysis of the potential of smart waterflooding to cogitate the current reservoir practices, encapsulating reservoir temperature and pressure, both the salinity and ionic composition of formation water and the current types of water injected.
The main objectives of this study include:
Investigate the reported mechanism for oil recovery by tuning the chemistry of injected water.
Study the surface chemistry of carbonate core and investigate the attraction of potential determining ions (especially Mg2+ and SO42-) in injected water towards carbonate core at reservoir conditions.
Investigate the potential of diluting injected water salinity whilst varying the concentration of potential determining ions for improving oil recovery in carbonate formation (single and composite) by analyzing coreflooding results, ionic/chromatographic analysis conducting contact angle and IFT measurements.
Building mechanistic reservoir simulation model to access the possible effect of water chemistry on oil recovery.
Correlate the coreflooding experiments with the simulation results.
1.4 Significance of Study
Until lately, low salinity was a phenomenon attributed to only sandstones. Yousef et al., (2010) reported a significance increase in the oil recovery from composite limestone cores by sequentially flooding the cores with Gulf SW and diluted (2, 10, and 20 times) Gulf SW. On the contrary, flooding outcrop chalk cores with diluted SW doesn't affect oil recovery. As a matter of fact, on diluting seawater, the oil recovery was drastically decreased due to reduction in the active ions concentration (Fathi et al., 2010a).
Austad and co-workers have shown tremendous increase in oil recovery on several works done on chalk cores e (Austad et al., 2005; Zhang and Austad, 2006; Zhang et al., 2007; Strand et al., 2008; Puntervold et al., 2009). They observed few active ions (SO42-, Ca2+ and Mg2+) in seawater which act as a potential determining ions towards chalk surface by changing the surface charge (Zhang and Austad, 2006). They also presented that raising the concentration of SO42- in seawater can serve as a wettability modifier. Fathi et al., (2010a) even suggested that depleting seawater in NaCl concentration should be even smarter water than ordinary seawater. Zhang et al. (2007) investigated the effect of Ca2+ and Mg2+ at the chalk surface and noticed that at high temperature, the affinity of Mg2+ was higher than Ca2+. It was observed that without Mg2+ present, the solubility of CaSO4 is drastically decreased and will precipitate at a temperature above 100oC which will block the porous system.
Considering this reports, this study will based on accessing the combined effect of seawater and diluted seawater on oil recovery while tuning the concentration of Mg2+ and SO42- with/without NaCl on a carbonate core at a high temperature, i.e. reservoir temperature. This study is new and will likely add to literature on the mechanism of oil recovery as observed in the laboratory experiments.
1.5 Scope of Work
Coreflood Experiments
Core flood experiments will be performed using reservoir cores and reservoir crude oil under current reservoir practices, encapsulating reservoir temperature and pressure, representative wettability, salinity and ionic composition of formation water and the current types of water injected. Series of coreflooding experiments will be carried out on both single and composite cores in order to quantify the level of oil recovery by waterflooding using brines with different salinities and ionic compositions. Modification of injection water in terms of both salinity and ionic composition would be based on seawater and two scenarios will be combined:
Waterflood with different salinities
Waterflood with different ionic compositions
The results of these tests will quantify the level of oil recovery by using different injection water and help define the most favourable brine salinity and composition for improving oil recovery. Furthermore these tests will also generate data for calculating water/oil relative permeability for the above scenarios, which would be used in simulation study.
Mechanistic study on smart waterflooding study
In order to better understand relevant mechanisms for smart waterflooding in carbonates, ionic composition analysis, chromatographic studies (adsorption analysis), contact angle and IFT measurements will be carried out under full reservoir conditions. In these measurements, both brine salinity and ionic composition will be considered as variables, and the impact of brine salinity and composition would be well decided.
Simulation study
By adopting the Kr data from 1-D simulation match on Coreflood results and reservoir static model, reservoir simulation will be conducted to evaluate the performance of different injection scenarios on laboratory scale. In addition, based on mechanistic study, simulation method could be proposed for smart waterflooding in carbonates.
1.4 Summary of the Dissertation
This thesis consists of six chapters;
Chapter 1 gives an overview of the introduction, motivation and objectives of the research study.
Chapter 2
Chapter 3
Chapter 4
Chapter 5 presents the result from simulation studies and comparison with the experimental result.
Chapter 6 summarizes the entire findings from this research and presents suggestions for future research.
CHAPTER TWO
LITERATURE REVIEW
Carbonate rock
Carbonates are mostly regarded as biogenic. Rocks comprising of about 50 % carbonate minerals are generally classified as a carbonate rock. Carbonate minerals are often distinguished by their composition; few are Calcite [CaCO3], Magnesite [MgCO3], Dolomite [CaMg(CO3)2], Siderite [FeCO3], Ankerite [CaFe(CO3)2]. Calcite and Dolomite are two minerals majorly discover in carbonate rocks (Mazzullo et al., 1992).
Oil recovery in carbonates
The natural energy of the reservoir is sufficiently used to get the oil from the reservoir during the first production stage. The drive mechanisms produced only 10-30 % of the oil originally in place (OOIP), and result in depletion of the reservoir pressure. Additional energy is needed to be supplied to provide a more efficient oil displacement and maintain reservoir pressure, after primary production has matured. The two secondary methods are gas injection and waterflooding and the recovery factor after this stage is usually 30-50 % of OOIP (Castor et al., 1981).
The tertiary recovery stage (EOR) is employed when secondary stage becomes uneconomical and here, there is still about 40-60% OOIP trapped in the reservoir. The EOR methods target mostly the trapped oil left in the reservoir, which can be a substantial amount particularly in carbonate reservoirs. Five categories of EOR processes exist: mobility-control (polymers, foams), chemical (surfactants, alkaline agents), miscible (hydrocarbon solvents, CO2), thermal (steam, in-situ combustion) and other processes etc., (Green and Willhite, 1998).
Chilingar and Yen (1983) conducted an experimental study on 161 carbonate rocks; 8 % were water-wet, 12 % were of intermediate wettability, 65 % were oil-wet, and15 % were strongly oil-wet. Even still, oil recovery from carbonates by convectional waterflooding is quite limited because they are usually found neutral to oil-wet. Carbonate reservoirs hold about 50 % of the present proven oil reserves in the world, and as a result of unfavorable wetting state, oil recovery are below 30 % OOIP (Treiber et al., 1972). The unfavorable wetting condition therefore prevents imbibition of water into the matrix attributable to a negative capillary pressure. Thus, in order to achieve high recoveries from carbonate rocks, the capillary pressure needs to be increased by a wettability alteration of the rock surface towards a more water-wet state, in that way promoting imbibition of water to expel the oil.
Wettability
Initially, an oil reservoir is completely filled with water, a thermodynamic equilibrium has been constituted between the oil, rock and formation water through millions of years and its mineral surfaces are water-wet. When oil enters the reservoir by moving through the pores, water will still continue to wet the pore surfaces. Once water is pushed away from the wall of the pores, oil will began to contact the pore wall. The potential for change in wettability is therefore determined by the firmness of the water film (Kovscek et al. 1993). If the water film can resist the pressure in the oil phase, then the water film is stable but if not, it collapsed. Once the water film collapsed, active components in the oil will adsorb to rock surface and change the wettability. The structure of the oil and the density of adsorption sites will determine the change in wettability towards either strongly oil-wet or more moderately oil wet (Buckley 2001; Thomas et al. 1993a, b). A very thick water film will prevents the active components contained in the oil from adsorbing to the surface. Nevertheless, a high pressure in the oil phase can reduce the thickness of the water film sufficiently to allow active components in the oil phase to adhere to the mineral surface.
Wettability is therefore defined as the ability of a fluid to adhere to the surface of a solid in the vicinity of other immiscible fluids (Anderson, 1986). In a COBR system, classification of pore space's wettability can be done depending on the shape of the pore space, the mineralogy of the pore space and the composition of the crude oil (Buckley 2001).
A rock is water-wet if the aqueous phase is retained by capillary forces in the smaller pores and on the surface of the larger pores while the oleic phase occupies the center of the larger pores and exists as droplets (fig. 2.1). Oil-wet rock is simply the reverse of water-wet rock. A rock is neutrally-wet if there is no preference for one fluid or another. A rock maybe fractionally wet if it is composed of different minerals with different surface properties and wettability variations exist throughout the rock.
However wetting in carbonates are completely different from sandstones. At reservoir conditions, the carbonate surface is positively charged and the active components, carboxylic material, in crude oil, are negatively charged. The bond existing between the positively charged sites on carbonate surface and the negatively charged carboxylic group, -COO-, is very strong, and the large molecules will cover the carbonate surface. While for sandstones, clay minerals are usually negatively charged at the relevant pH range of the formation water which contains a higher concentration of divalent ion, Ca2+. However, the clay minerals are the most strongly adsorbed by polar components from the crude oil in this case.
Figure 2.1: Illustration of oil and water distribution in water-wet pores
The wetting properties of a COBR system have a very strong influence on the fluid flow because they dictate the capillary pressure and the relative permeabilities of the fluids. The ease with which a phase flows through the porous media in the presence of other phases depends on its relative permeability, whilst the ease with which same phase is displaced from the porous rock depends on its capillary pressure with the other phases. The capillary pressure is also a function of wettability and Interfacial Tension, IFT. The wetting phase classifies the reservoir's wettability, as illustrated in Figure 2.2. Mobilizing oil in an oil-wet reservoir proves ineffective due to the imposed negative capillary pressure and formation of high IFT against water. Therefore, oil exhibits very low relative permeability in oil fields with high water cuts. The remaining oil saturation is much higher than the optimal residual oil saturation due to oil or intermediate wetting conditions. When the reservoir's wettability is altered favorably to approach a more water-wet state, the residual oil saturation decreases, and more oil is recovered (Anderson, 1987).
Figure 2.2: Comparison of water wet and oil wet rocks (Borchardt and Yen, 1989)
Early research with Fresh Water
Almost a half century ago, few researches began by injecting fresh water into core samples, hoping to better understand the effect of clay content and the impact of fresh water and brine on oil recovery.
After production of 15% additional oil recovery from injection of brine (compared to fresh water) to Bradford field in Kansas, numerous laboratory tests were conducted (Smith, 1942). The experiments include Bradford sandstone cores (permeability range up to 77 md) and simulated calcium chloride brines (Bradford-field connate waters). Smith concluded that both brines and fresh water recovered equivalent amount of oil. Five years later, Hughes and Pfister stated the advantages of flooding brines as a secondary recovery method, although, they laid more emphasis on fluid physical and chemical characteristics to prevent clay swelling (Hughes and Pfister, 1947).
Martin (1957) injected fresh water into cores samples from Maracaibo Basin and East Texas reservoir to study the effects of clay content on recovery efficiency and relative permeability. Few cores were treated with cycles of toluene to remove clay materials and flooded with fresh water. The pre-treated cores were observed to have lower irreducible water saturations and higher water relative permeabilites. Then the whole cores were flooded with heavy oil and were later displaced with fresh water. The treated and untreated cores exhibit similar residual oil saturations and oil relative permeabilites. Permeability to fresh water decreased over the course of several hours or days after the fresh water injection was initiated. Fine migration was observed which causes pore throat plugging and the original water permeability could be restored easily by reversing the flow direction. Martin proposed that in the clay-rich cores, clay-water dispersion was created with a higher apparent viscosity and lower water relative permeability then the free water.
Bernard (1967) investigated the relative effectiveness of fresh and salt waters in flooding oil from synthetic and natural cores containing hydra-table clays. Distilled water, and sodium chloride solution at 15, 5, 1, and 0.1% were used in the experiments. The results of the sets of experiments revealed that the only significant increase in oil recovery was when NaCl solution was decreased from 1 to 0.1%, and so also was the pressure differential across the core. It was also documented that the increased recovery was always accompanied with an intense increase in pressure drop across the core. These experiments showed that the oil recovery was a function of salinity only in the range of 0 to 1% NaCl concentration. Bernard attributed the increased recovery to improved microscopic sweep efficiency induced by clay swelling and plugging of pore throat by migrating fines. The study concluded that cores containing clays will produce more oil with fresh waterflooding than with brine and if no high pressure drop is developed, then no additional oil will be produced.
Smart Waterflooding
Smart waterflooding is a collection of new promising methods for enhanced oil recovery. Unlike other EOR methods, it improves oil recovery, relatively inexpensive and environmentally friendly. Injection of formation water is termed to be a secondary recovery method, whilst injecting water with composition different from the initial formation water, may change wetting properties and thus serves as a tertiary recovery method. For carbonates and sandstones, the oil recovery can both increase and decrease to formation water while it can actually increase significantly with smart water. The "smart water" strategy is based on optimizing salinity and ion balance of the injected water in such a way that the change in equilibrium of the initial COBR interaction will modify the wetting conditions (McGuire et al. 2005). The change in the wetting condition therefore has a positive impact on the capillary pressure and the relative permeabilities of the fluids resulting in large additional recovery
Based on these earlier findings, researchers began to focus on the brine salinity which seems to be the only variable that can be manipulated in the injection water. This was done in order to better comprehend the factors that determine a crude oil/rock/brine interaction
