|Table of Contents|
Table of Contents
Chapter 1. Introduction
Chapter 2. Alcohol-free water-in-oil microemulsions
Chapter 3. Alcohol-free middle-phase microemulsions
Chapter 4. Excluded volume effect of anionic polymers on anionic surfactant solutions
Chapter 5. Thermal conductivity of microemulsions
Chapter 6. Stability of methanol-isooctane-toluene mixtures at -25 C
Chapter 7. Conclusions and recommendations
Appendix . Solution procedure for the heat conduction of a sphere
SOLUBILIZATION AND THERMAL PROPERTIES
HYEON KOOK LEE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA 1988
My parents, my wife, and my daughter, Boram
This work gave me a great opportunity to go one step further on my endless quest. I gratefully acknowledge those who have assisted me in this work.
My appreciation should first go to my advisor, Professor D. 0. Shah, a man of character and knowledge. He has been not only an
academic advisor but also an advisor in life to me. I would also like to thank Professor Gar B. Hoflund, Professor Gerald B. Westermann-Clark, Professor Gerasimos K. Lyberatos, and Professor Paul W. Chun as members of the supervisory committee for their time and advice.
I thank my research group colleagues for being supportive and encouraging. My gratitude is extended to Tracy, Ron, Jim, Nancy, Shirley, and Jill for their help and cooperation.
I cannot express enough my love and appreciation to my family. My thanks and my expression of regard for my parents are only a small token of my deep-felt gratitude to them for all the love, care and understanding which they have given me. My father has given me a philosophy for living while my mother has taught me love between family members. I should also give my wife, Wha Sun, my special thanks for her infinite support, continuous encouragement and selfsacrifice. Also, I thank my daughter, Boram, who has given me great iii
joy. Finally, my appreciation extends to Hyeon-Deok, who is not only my brother but also my best friend, for his encouragement during this work.
TABLE OF CONTENTS
ACKNOWLEDGMENTS ................................................ iii
LIST OF TABLES ................................................. V
ABSTRACT ...................................................... viii
INTRODUCTION .......................................... 1
1.1 Surfactants, Micelles, and Microemulsions ......... 1 1.2 Solubilization in Microemulsions .................. 14
1.3 Oil Displacement by Surfactants ................... 20
1.3.1 The Importance of Low Interfacial
Tension .................................... 22
1.3.2 Application of Microemulsions in
Tertiary Oil Recovery ...................... 24
1.3.3 Optimization of Microemulsions for
Tertiary Oil Recovery ...................... 24
1.3.4 Effect of Alcohols ......................... 32
1.4 Interaction between Surfactants and Polymers ..... 33 1.5 Thermal Conductivity of Liquids ................... 39
1.6 Phase Stability of the Gasoline-Methanol
Mixture ........................................... 44
1.7 Scope ............................................. 48
2 ALCOHOL-FREE WATER-IN-OIL MICROEMULSIONS .............. 52
2.1 Introduction ...................................... 52
2.2 Materials and Methods ............................. 54
2.2.1 Materials .................................. 54
2.2.2 Methods .................................... 57
2.3 Solubility of Water in the Alcohol-Free Waterin-Oil Microemulsions ............................. 59
2.4 Phase Behavior of the Water-in-Oil
Microemulsions .................................... 62
2.5 Proposed Mechanism of Oil Displacement by
Water-in-Oil Microemulsions ....................... 64
2.6 Experimental Results of Oil Displacements ......... 68
2.6.1 Effect of Composition of the W/O
Microemulsions on Oil Displacement ......... 69
2.6.2 Effect of the Amount of Water in the W/O Microemulsions on the Oil Displacement ........................... 74
2.6.3 Effect of Polymer Flooding and Oil as
a Slug on the Oil Displacement...............76
2.6.4 Effect of the Concentration of Salt
and Surfactant on Oil Displacement...........82
3 ALCOHOL-FREE MIDDLE-PHASE MICROEMULSIONS................89
3.2 The Concept of Solubilization.......................90
3.3 Materials and Methods...............................93
3.4 Results and Discussion..............................100
3.4.1 Solubilization and Optimal Salinity
of Middle-Phase Microemulsions...............100
3.4.2 Interfacial Tension..........................104
3.4.3 Oil Displacement Efficiency..................107
4 EXCLUDED VOLUME EFFECT OF ANIONIC POLYMERS
OH ANIONIC SURFACTANT SOLUTIONS.........................114
4.2 Materials and Methods...............................117
4.3 Effect of Addition of Polymers on
Interfacial and Surface Tension.....................120
4.4 The Excluded Volume Effect of Anionic Polymers ..130 4.5 Effective Concentration of the Surfactant...........134
4.6 Oil Displacement Efficiency.........................144
5 THERMAL CONDUCTIVITY OF MICROEMULSIONS..................150
5.2 Theoretical Model for the Measurement of the
Thermal Conductivity of Liquids.....................151
5.3 Materials and Methods...............................153
5.4 Effect of the Microstructure of the Liquid..........164
5.5 Thermal Conductivity of Microemulsions..............170
6 STABILITY OF METHANOL-ISOOCTANE-TOLUENE
MIXTURES AT -25 'C......................................186
6.2 Materials and Methods...............................187
6.3 Effect of the Chain Length of Alcohols..............192
6.4 Effect of the Structure of Alcohols.................194
6.5 Effect of Surfactants...............................206
7 CONCLUSIONS AND RECOMMENDATIONS.........................229
7.1 Effect of Nonionic Surfactants on Solubilization
Capacity, Interfacial Tension, and Oil
Displacement Efficiency of Nicroemulsions...........229
7.2 Excluded Volume Effect of Anionic Polymers
on Anionic Surfactant Solutions.....................231
7.3 Thermal Conductivity of Nicroemulsions..............232
7.4 Stability of Hydrocarbon and Alcobol
Mixtures at -25C ...................................233
7.5 Recommendations for Further Studies.................234
APPENDIX SOLUTION PROCEDURE FOR THE NEAT CONDUCTION OF
Abstract of Dissertation Presented to the Graduate School of
the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
SOLUBILIZATION AND THERMAL PROPERTIES OF MICROEMULSIONS
Hyeon Kook Lee
Chairman: Professor Dinesh 0. Shah Major Department: Chemical Engineering
The objectives of this study were to investigate the
solubilization, interfacial tension and thermal properties of microemulsions as well as the excluded volume effect of a polymer in surfactant solutions.
Alcohol-free water-in-oil microemulsions using nonionic surfactants in place of short-chain alcohol were successfully formulated using Aerosol-OT, Arlacel 20, mineral oil and brine. The solubility of water in the microemulsion depended on the weight ratio of two surfactants. The oil displacement efficiency of the microemulsion was related to solubilization of the microemulsion and was greatest for the microemulsion having the largest solubilization resulting from the more piston-like displacement due to lower interfacial tension. Further, alcohol-free middle-phase
microemulsion was successfully formulated using TRS10-80, Tween81, mineral oil and brine. Its solubilization and interfacial tension viii
were much larger and lower, respectively, compared to the corresponding middle-phase microemulsion containing n-butanol. The
greater degree of solubilization, using the nonionic surfactant in place of the short-chain alcohol, was explained using the modified Winsor's R. theory.
The mechanism of the excluded volume effect of anionic polymer on an anionic surfactant solution was studied using both static and dynamic light scattering techniques. The interfacial and surface tensions calculated by utilizing the light scattering data were in agreement with the experimental values supporting the excluded volume effect of polymer for the similarly charged polymersurfactant solutions.
The effect of the structure of liquids on their thermal properties was studied using a transient method for thermal diffusivity developed in this study. An isotropic liquid showed 20% larger thermal diffusivity than a birefringent liquid, implying the significant effect of the microstructure of liquids on their thermal properties. As an extension, thermal conductivities of
microemulsions were examined and were compared with the known equations for liquid mixtures. The results suggested that the microstructure of microemulsions influences the thermal properties of these systems.
The study of the phase stability of a methanol-isooctanetoluene mixture at -25 OC showed that alcohol and surfactants were effective in stabilizing the mixture as a single phase at -25 OC and ix
their effectiveness decreased in the following order: mixture of anionic surfactant and alcohol > anionic surfactant > alcohols.
Since 1943, when Hoar and Schulman (1) first described a microemulaion aa a transparent or translucent system formed spontaneously upon mixing oil and water with a relatively large amoun-t of ionic surfactant together with a cosurfactant such as an alcohol of medium chain length (C4 to C7), extensive studies have been done on the surfactant solutions and the microemulsions because of their usefulness in numerous applications such as tertiary oil recovery, detergency, catalysis, drug delivery, lubrication, and an on. Numerous books, symposium proceedings and review articles have been published during the past decade (2-15), and it is virtually impossible to review all such studies here. Hence, this chapter will focus only on the fundamental concepts and findings which are related to the subjects of this dissertation, such as solubilization, interfacial tension, oil displacement, interaction between surfactants and polymers, thermal conductivity, and phase stability.
1.1 Suifactanta, Micelles and Microemulsions
Surfactants are amphiphatic molecules having distinct hydrophobic and hydrophilic groups (16). Depending on the chemical structure of the polar group, surfactants can be nonionic, cationic,
anionic, or zwitterionic. The nonpolar part of surfactants can be a saturated or unsaturated, straight, branched or cyclic hydrocarbon or can be one or more hydrocarbon chains. Functional groups can also be incorporated into surfactants. The structures of typical surfactants are shown in Figure 1-1.
When surfactant molecules are dissolved in water, the presence of the hydrophobic group of a surfactant molecule in the interior of water causes a distortion of water structures increasing the free energy of the system. This means that less work is needed to bring a surfactant molecule than a water molecule to the surface, since the presence of the surfactant decreased the work needed to create a unit area of surface (16).
On the other hand, the presence of the hydrophilic group of a surfactant molecule prevents the surfactant from being expelled completely from water as a separate phase, because that would require desolation of the hydrophilic group. Therefore, the
amphiphatic structure of a surfactant molecule causes not only concentration of the surfactant at the surface and reduction of the surface tension of water, but also orientation of a surfactant molecule at the surface with its hydrophilic group in the aqueous phase and its hydrophobic group oriented away from the aqueous phase (17).
The properties of a surfactant are determined by the balance between its hydrophobic and hydrophilic groups. If the hydrophilic group is dominant, the surfactant is called a water soluble surfactant because the hydrophilic group drags the entire surfactant
Anionic +CH S a
Surfactant 0 6H S 3Na
Cationic R N H+ C F~
(Salt of a Long-Chain Amine)
SuractntR N (C H 3)2C H2CH2S 03
Surf actant 4X
Figure 1-1. Structure of typical surfactants. R represents
alkyl chain of C 12 to C18 ethylene groups.
molecule in water. In contrast, if the hydrophobic group is dominant, it does not dissolve in water. The lowering in surface (or interfacial) tension as mentioned earlier can be described by the well-known Gibbs adsorption isotherm for a multiple-component system as (18)
dY = -Z T. d l -Z T. RT d(ln ai)
where y is the surface tension (or interfacial tension), ri is the surface excess of component i (amount of component i adsorbed per unit area), p i is the chemical potential of component i, and ai is the activity of the component i. Equation 1-1 basically states that the increase of surfactant activity ai in the solution would result in a decrease of surface tension (or interfacial tension) if the surface excess of the surfactant is positive.
As the concentration of a soluble surfactant in water gradually increases, the monomer concentration in water and the surface concentration of adsorbed surfactant increase. At a critical bulk concentration of surfactant, known as the critical micelle
concentration (cmc), the surfactant molecules start to form the aggregates in water called micelles. In general, the micelles are
spherical aggregates of surfactant molecules 40-100 A in diameter and are in equilibrium with single surfactant molecules or monomers
(19). At cmc, the solution properties such as osmotic pressure, surface tension, viscosity, electrical conductivity, and density abruptly change (20).
Properties of aqueous micelles, in aquesous solutions have been discussed in detail in several books sod reviews (8, 21-33). The formation of micelles io an aqueous solution creates a local nonpolar environment at the cores of micelles within the aqueous phase as shown in Figure 1-2 (a) because the hydrophobic group of a surfactant molecule tends to move away from water. Oil soluble molecules can be dissolved within the micelles (19, 29, 34). As shown in Figure 1-2 (b), the structure of the micelle in a nonpolar solution is similar to that of the micelle in an aqueous solution but reversed, with the hydrophilic groups and water comprising the interior region surrounded by an outer region containing the hydrophobic groups and solvent molecules (35).
It should be noted that the exact molecular structure of the micelles is still disputed even though the structure of the micelle in an aqueous solution, at concentrations not too far above the cmc, is generally considered to be roughly spherical (17, 36). For example, Rouviere et al. (37) proposed a spherical shape of the reversed micelle to explain the observed effects of electrolytes on solubility of water, relative viscosity, and enthalpy of hydration in Aerosol-OT-n-decane system while Ekwall et al. (38) showed evidence for prolated ellipsoidal Aerosol-CT reversed micelles. Further, as shown in Figure 1-3, the structure of the micelle changes from spherical through rod- or disklike to lamellar in shape with changes in temperature, concentration of surf actant, and appropriate additives (19, 39). It should be also noted that the
(a) Core Water
Organic Solvent Figure 1-2. Two typical micelle types. (a) Normal
micelle; (b) reverse micelle.
Monomers Micelle Micelle Packing Of
Hexagonal Packing Microernoision Of Water Cylinder
Figure 1-3. A schematic presentation of structure formation
in surfactant solution upon the concentration
of surfactant as well as physico-chemical
equilibrium between single surfactant molecules (monomers) and micelles at the concentrations beyond the cmc is not static but dynamic as shown in Figure 1-4. The dynamic aspect of the
equilibrium has been reviewed in the previous two review articles on amphiphile aggregation in water (32) and in nonpolar solvent (40) in addition to a recent Ph.D. dissertation (33). The chain packing in mecelles have been also studied in recent years using the
statistical thermodynamics of molecular organization and the
conformational statistics of the hydrocarbon chains within the hydrophobic core (25, 26, 27). The results may be summarized as follows: Micellar aggregates are charcterized by motions occurring on distinctively different time scales. The fast motions are
characterized by correlation times of the order of 20 ps (slightly different at different positions along the chain). These
correlation times are close to those of the corresponding liquid alkanes. The slow motion for spherical micelles can be accounted for by micelle rotation and surfactant molecule lateral diffusion. The experimental results by the measurements of free energies of transfer of nonpolar molecules and surfactant molecules into
micelles, viscosity and diffusion measurements, nmr relaxation measurements and neutron diffraction studies suggest minimal water penetration in micellar cores (27, 28).
Many micellar solutions can spontaneously solubilize large
amounts of oil (or water) to form microemulsions (41). At present, there is not a precise, or indeed agreed upon, definition of microemulsions (8, 10, 42-44). Recently, Shinoda and Lindman wrote
Mi cellesa Figure 1-4. A schematic presentation of the dynamic
equilibrium between surfactant monomers, micelles
sod adsorbed monolayer.
in a review article: "In a strict sense, a microemulsion may be defined as a system of water, oil, and amphiphile(s), which is a single phase, and a thermodynamically stable isotropic solution. In a wide sense, it has been used to include fairly stable dispersions which are transparent, translucent (opalescent), and opaque, but it appears that serious confusion may arise without a clear distinction between thermodynamically stable and unstable systems (even if the difference in free energy may be extremely small)" (15, p.136).
In spite of the controversy about the definition of
microemulsions mentioned above, the designation of a clear isotropic single-phase region in a phase diagram as microemulsions does offer practical convenience in terminology (33). Another controversy regarding the nature of microemulsions is a distinction between
micelles and microemulsions. One group of investigators (13, 45) believes microemulsions to be special types of micellar solutions, that is, microemulsions to be solutions of swollen micelles of emulsifying agent in the interior of which the dispersed liquid is solubilized.
In contrast, a second group of researchers (46, 47) believes that micelles and microemulsions are fundamentally different. They feel that the relatively small size of micelles precludes them from having properties and characteristics of the dispersed aggregates in microemulsions and consider microemulsions to consist of 10-200 nm diameter droplets of the dispersed liquid surrounded by an
interfacial film in which the emulsifying agent molecules outnumber
oil molecules (48). In either case, the emulsifying agent is oriented in such a fashion that the hydrophobic groups face the oil phase and the hydrophilic groups face the aqueous phase. Water-inoil (w/o) microemulsions may be considered either as micellar solutions containing reverse micelles with water solubilized in the inner core or as emulsions containing tiny droplets of water surrounded by an interfacial film. Oil-in-water (o/w)
microemulsions may be thought of either as micellar solutions containing normal micelles with solubilized oil in their inner cores or as tiny droplets of oil surrounded by an interfacial film
dispersed in water.
In addition to the o/w and w/o type microemulsions, a
bicontinuous structure has been proposed which does not have a distinct dispersed or dispersing phase (49, 50). Whether the
microemulsion will be w/o or o/w depends on the direction of
curvature of the interfacial region as determined by such factors as the difference in interfacial tension (51), or interfacial pressure
(52), on the two sides of the interface, and/or the difference in volumes and compressibilities of the hydrophilic and hydrophobic moieties of the surfactants (53). For example, if the oilhydrophobic end tension were greater (or interfacial pressure lower) than the water-hydrophilic end tension, then the former side would be shortened, causing the interfacial film to be concave toward the oil, resulting in the enclosure of the oil by the water and
therefore forming an o/w microemulsion. A preferentially oil-soluble surfactant would produce a lower interfacial tension (or greater
interfacial pressure) at the oil interface, yielding a w/o aicroemulsion while a preferentially water-soluble surfactant would produce a lower interfacial tension (or greater interfacial pressure) at the water interface, yielding an o/w microemulsion
Such an explanation of the formation of the o1w- or w/o microemulsion is consistent with the geometrical packing considerations of surfactants through the direction of curvature of the interfacial film. The latter describes the geometrical packing of surfactants at the interface by the packing ratio of crosssectional area of hydrocarbon chain to that of polar head of a surfactant molecule at the interface, V/a0 1c, where V is the volume of hydrocarbon chain of the surfactant, a. is the optimal crosssectional area per polar bead in a planar interface, and lc is approximately 80-90% of fully extended length of the surf actant chain (54). As shown in Figure 1-5, it is intuitively clear that a smaller cross-sectional area of tail than that of head (V/a0 Ic < 1) will favor the formation of the o/w microemulsion (Figure 1-5a), while a greater cross-sectional area of tail than that of head (V/a0 l> 1) would favor the w/o microemulsion (Figure 1-5c). A planar interface requires V/a0 lc = 1 which leads to the formation of lamellar structure (Figure 1-5b). This geometrical packing consideration has been well reviewed in a previous publication (33).
V / aolC <1 (a)
V / no = (b)
V / aolc > 1 c)
Figure 1-5. A schematic diagram representing the effect of
the geometrical packing of surfactants on the structure of microemulsions. (a) Oil-in-water
microemulsion (head area> chain area); (b) lamellar structure (head area= chain area)
(c) water-in-oil microemulsion (head area < chain
1.2 Solubilization in Microemulsions
One of the important properties of micelles and microemulsions is the ability to solubilize a solvent-insoluble substance,
important from a practical point of view because most industrial applications originate from this property (e.g. paints, inks, etc.).
Correlations between the chemical structure of a surfactant and its solubilization capacity are complicated by the fact that both phases, oil and water, can have variable compositions. Moreover, the concentration at which the surfactant is used determines its solubilization capacity as well as the type of microemulsion (o/w or w/o) formed. Consequently, such complexity makes it difficult to select the proper surfactant for a given purpose. However, there are some general guidelines (17) that can be helpful in the
selection of surfactants: I) Surfactants must show surface activity and produce a low interfacial tension in the particular system in which it is to be used; 2) surfactants must form, at the interface, either by itself or with other adsorbed molecules that are present there, an interfacial film that is condensed because of lateral interactions between the molecules comprising the interfacial film; 3) surfactants must migrate to the interface at a rate such that the interfacial tension is reduced to a low value in the time during which the microemulsion is being produced; 4) preferentially water soluble surfactants form o/w microemulsions; 5) preferentially oil soluble surfactants form w/o microemulsions; 6) the more polar the oil phase, the more hydrophilic the surfactant should be; 7) the
more nonpolar the oil to be solubilized, the more hydrophobic the surfactant should be.
A practical way to select the proper surfactant is the HLB (Hydrophile-Lipophile Balance) method. In the HLB method (55), a number between 0 and 40 is assigned to a surfactant depending on its solubilization behavior and is related to the balance between the hydrophilic and lipophilic (hydrophobic) portions of the surfactant molecule. The smaller the HLB value is, the more hydrophobic the surfactant is. An HLB value of 10 is assigned to the balanced state between the hydrophilic and hydrophobic portions of the surfactant molecule. In addition, a similar range of numbers has been assigned to various substances that are frequently solubilized such as oils, paraffin wax, xylene, and so on (56). Thus, the surfactant is
selected whose HLB number is approximately the same as that of the ingredient to be solubilized. However, since the chosen HLB value can be obtained by a combination of the surfactants using the weighted average HLB number, various combinations of surfactants with the same weighted average HLB number must then be tried to determine the optimum surfactant combination.
Winsor's R-theory of solubilization (57) is also helpful in selecting the proper surfactants for maximum solubilization under the given conditions. Since this will be explained in detail in a later chapter, it will be sufficient here to mention that this
theory qualitatively considers the interaction energies between the surfactant and oil and between the surfactant and water. As
described later, the magnitudes of the two interaction energies are related to the solubilization capacity of the microemulsion.
So far, the discussion has focused on the macroscopic view of solubilization. Therefore, considering the microscopic view of solubilization will further help us to understand the solubilization in the microemulsions. From studies on the solubilizate before and after solubilization using X-ray diffraction (58, 59), ultraviolet spectroscopy (60) and nmr spectrometry (61, 62), it has been known that the site of the micelle at which solubilization occurs varies with the nature of the material solubilized. Figure 1-6 shows the five sites available for solubilization based on these findings: I) on the surface of the micelle, at the micelle-solvent interface; 2) between the hydrophilic head groups; 3) in the so-called palisade layer of the micelle between the hydrophilic groups and the first few carbon atoms of the hydrophilic groups that comprise the outer core of the micelle; 4) more deeply in the palisade layer; and 5) in the inner core of the micelle.
In aqueous solvents, nonpolar hydrocarbons are solubilized in the inner core of the micelle, between the ends of the hydrophobic groups of the surfactant molecules. Polarizable hydrocarbons have been shown to be solubilized in quaternary ammonium solutions
initially by adsorption at the micelle-water interface, replacing water molecules that may have penetrated into the outer core of the micelle close to the polar heads, but solubilization of additional material is either deep in the palisade layer or located in the inner core of the micelle (63). In aqueous medium, large polar
Figure 1-6. Sites of solubilization of material in a
surfactant micelle. (1) On surface of micelle;
(2) between the head groups; (3) in palisade layer; (4) more deeply in palisade layer; (5)
in the inner core of the micelle.
molecules are believed to be solubilized between the individual molecules of surfactant in the palisade layer with the polar groups of the solubilizate oriented toward the polar groups of the surfactants and the nonpolar portions oriented toward the interior of the micelle. Small polar molecules, in aqueous medium, are generally solubilized close to the surface in the palisade layer or by adsorption at the micelle-water interface. Among the sites
mentioned above, the solubilization in the inner core of the micelle is relevant to the formation of microemulsions (17) even though
other sites may be considered (64).
It is easy to see the relationship between solubilization and the drop size of a microemulsion by considering a simple geometric calculation for a sphere:
V = AtR / 3 (1-2)
where At is the total interfacial area, R is the radius of the droplet, and V is the total solubilization volume in a microemulsion drop. Since At is related to the total emulsifier concentration in a system, the solubilization is directly related to the droplet radius and hence the curvature of the interface (65-70). The radius of the microemulsion droplet is related to the thermodynamic
stability of microemulsions. This is evident from the observation that all microemulsions exhibit phase separation beyond a critical radius of droplets depending on the composition of the system.
It has been known that the free energy of a microemulsion is primarily determined by the entropic contribution of the dispersion of the droplets in the continuous phase, the interfacial contribution (including the curvature effect) and the interaction between droplets (69, 71-73). Among these three contributions to the free energy of a microemulsion, the curvature of interface and the interaction between droplets are known to be strongly influenced by the molecular structures of the interfacial layer and the continuous phase (64, 74). These two effects on the stability of the w/o microemulsion droplets were recently studied by Hou and Shah
(75). Their theoretical and experimental results can be summarized as follows: 1) From the consideration of the stability of
microemulsion systems, the growth of microemulsion droplets during the solubilization process has to be limited either by the radius of the spontaneous curvature of interface (R.) as the result of curvature effect, or by a critical radius of droplets (Rc) due to the attractive interaction among the droplets; 2) for the systems where solubilization capacity of water is limited by the spontaneous curvature of the layer, the solubilization capacity can be increased by any modification of the molecular structure of either interface or continuous phase so that the spontaneous curvature of the layer is decreased (i.e., Ro is increased); 3) for the systems where the solubilization capacity is limited by critical droplet radius, the reduction of attractive interaction among droplets (i.e., increase of Rc) increases the solubilization capacity of water; 4) both Ro
and Rcare strongly influenced by the variation of the molecular
structure of interface and continuous phase; 5) the maximum solubilization can be obtained by changing the molecular structure of interface or continuous phase, as a result of the compromise between the two opposite effects of the curvature of interface and the attraction among droplets.
From the above mentioned macro- and microscopic view, it is clear that the two important factors for the large solubilization in the microemulsion are the hydrophilic and lipophilic balance of the surfactant and the stability of the microemulsion droplets. Both factors are strongly affected by the structure of the surfactant and the solvent.
1.3 Oil Displacement by Surfactants
When a new oil reservoir is discovered, the reservoir usually possesses the natural pressure to move the oil to the production wells by expansion of volatile components. This phase is called
primary oil recovery. When the natural pressure decreases, it is necessary to increase the pressure by injection of water. The injection of water, called secondary oil recovery or water flooding, is stopped when the production cost becomes significant compared to the value of the produced oil. The total oil recovery by the primary and secondary oil recovery is less than 40% of the original oil in place. Therefore, the remaining oil, 60%, is the target of the enhanced oil recovery techniques called tertiary oil recovery. There are several proven techniques for such tertiary oil recovery:
surfactant-polymer flooding (76-78), foam flooding (78-80), CO2 flooding (81), caustic solution flooding (82), steam injection (83), thermal combustion (84), and microbial methods (85). This section focuses on the fundamental aspects of the surfactant-polymer flooding (or microemulsion flooding) as well as related phenomena.
[The use of a microemulsion to displace the residual oil in the reservoir was not initiated until representatives of the Marathon Oil Company presented four technical papers at the annual fall meeting of the Society of Petroleum Engineers in 1967. Since then many researchers have proposed optimal microemulsion systems for given reservoir conditions. Today, enhanced oil recovery by microemulsions has become one of the most studied applications of microemulsion technology.
Because of multicomponents and multiphases involved in tertiary oil recovery by microemulsions, most of the studies have been experimental rather than theoretical. Further, most of the studies have been done on the laboratory scale, because a study on a pilot plant scale requires considerable expense and several years to obtain data. Nevertheless, many important findings of the laboratory scale studies help us to understand the mechanism and the effect of several variables on the enhanced oil recovery by
microemulsions. A few field tests for surfactant-polymer flooding have been reported (76, 77). The result showed that surfactantpolymer flooding increased tertiary oil recovery by 4 times compared to polymer flooding (77).
1.3.1. The Importance of Low Interfacial Tension
The fundamental question of why a microemulsion is effective for the tertiary oil recovery can be answered by considering the forces acting on the residual oil drops and the concept of capillary number.
After the water flooding, the residual oil drops are captured in the porous medium by the capillary force. In a dynamic situation, the residual oil drop is also subjected to a viscous force. The
ratio of viscous force to capillary force is defined as the capillary number, even though the detailed variables of the capillary number are somewhat different depending on the investigators (86). Foster (87) has defined the capillary number as
Nca = T1 V/Y (1-3)
where n and V are the viscosity and velocity of the displacing fluid, respectively. Gamma is the interfacial tension and is the porosity of the porous medium. He found, as shown in Figure 1-7, that the level of the residual oil decreased as the capillary number increased. Others also have shown similar results (86). Since the capillary number after the water flooding is around 10-6, it should be increased by 3 to 4 orders of magnitude for enhanced oil recovery (88-91). Thus, it is obvious from the definition of the capillary number that a decrease in the interfacial tension increases the capillary number. Therefore, the lowering of the interfacial
0 20 40 60 80Residual Oil (Percent Pore Volume) Figure 1-7. Dependence of residual oil saturation on
tension is believed to be a major role of a microemulsion for tertiary oil recovery. A microemulsion can decrease the interfacial tension from 20 or 30 dynes/cm (mN/m) to 10-3 dynes/cm (mN/m), resulting in the mobilization of the oil drops trapped after water flooding.
1.3.2. Application of Microemulsions in Tertiary Oil Recovery
The application of microemulsions to tertiary oil recovery was well described by Healy and Reed (92). In this process, a
microemulsion or a surfactant slug is injected into a porous medium after the water flooding. The slug is then displaced by thickened or viscous water (usually a polymer solution). As long as the slug has compositions within the single-phase region, oil is displaced miscibly and all of it is recovered. However, for economic reasons, the small size of the slug is mixed with the residual oil and brine resulting in the two or more phases. From this point, the
displacement is immiscible and less efficient, and some of the residual oil remains trapped in the pores. The phase change of the microemulsion slug during tertiary oil recovery is correlated to the phase behavior studies in the test tubes (93). Several investigators (94-99) showed that tertiary oil recovery could be predicted using a chemical flooding compositional simulator based on the phase behavior of a surfactant, an alcohol, oil and brine.
1.3.3. Optimization of Microemulsions for Tertiary Oil Recovery
Several controlling factors are involved in tertiary oil
recovery by microemulsions, namely, interfacial tension,
solubilization, coalescence and mobility control (100). Therefore, these factors should be considered to optimize microemulsions for tertiary oil recovery. \Furthermore, a middle-phase microemulsion has received more attention than lower- and upper-phase microemulsions due to its greater solubilization capacity and lower interracial tension at both microemulsion/oil and
microemulsion/water interfaces.1 Therefore, this section will be devoted to the phenomena occurring in such a middle-phase
The Concept of the Optimal Salinity. When a microemulsion is formed by mixing oil and brine with a surfactant and an alcohol in a tube, the location of the microemulsion phase in a vertical tube varies depending on several parameters as shown in Figure 1-8. The typical way to see the phase behavior is to increase the
concentration of salt while other parameters are fixed. As the
concentration of salt increases, the location of the microemulsion phase moves from the bottom to the top through the middle of the vertical tube; consequently, a lower-phase, an upper-phase, and a middle-phase microemulsion have been named. The lower-phase and the upper-phase microemulsions are o/w- and w/o type, respectively, while the structure of the middle-phase microemulsion has not yet been clearly known (100). A typical definition of the optimal salinity is the salinity at which the volumes of oil and brine solubilized in the middle-phase microemulsion are equal (101).
Oil- -M *..:. U .
The Transition L--4M--U Occurs By:
1. Increasing Salinity
2. Decreasing Oil Chain Length
3. Increasing Alcohol Concentration (C 4- C6)
4. Decreasing Temperature
5. Increasing Total Surfactant Concentration
6. Increase Brine/Oil ratio
7. Increasing Surfactant Solution/Oil Ratio
8. Increasing Molecular Weight Of Surfactant
Figure 1-8. Schematic illustration of the factors influencing the transition from the lower- to the middle- to the
upper phase microemulsion.
Another definition is the salinity at which the least amount of surfactant and/or alcohol is required to form a single phase (the ratio of surfactant to alcohol is fixed) (102). The optimal
salinities obtained by these two different definitions are similar. The importance of the optimal salinity lies in the fact that the interfacial tension is lowest for a microemulsion with excess oil or brine phase at the optimal salinity (103-107). Thus, the optimal salinity can serve as a parameter to formulate an optimal system which produces ultralow interfacial tension.
An important relationship was proposed by researchers (71,
108): the interfacial tension and the solubilization parameter are inversely related at optimal salinity as
= constant (1-4)
Cos T / 4
where 7 is the interfacial tension at optimal salinity and is the solubilization parameter at optimal salinity (the solubilization parameter is the volume of oil or brine solubilized in the microemulsion phase per unit volume of surfactant). Equation 1-4 implies that a low interfacial tension requires a large
solubilization of oil or brine in the microemulsion phase. Bourrel and Chambu (102) discussed the rules to achieve high solubilization by considering interaction energies among molecules of oil, brine and surfactants (and alcohols). One of their conclusions was that the interaction energy between brine and surfactants should be
equally balanced with the interaction energy between oil and aurfactanta for an optimal ayatem. Further, they showed that aolubilization depends on the magnitude of interaction energies for an optimal system. Hsieh and Shah (109) showed the effect of chain length of oil and alcohols, and the surfactant to alcohol ratio on the optimal salinity. They found that for the same concentration and type of petroleum aulfonate and alcohol, the optimal salinity of the microemulsion system can be predicted from the density of the nalkane. The phenomena occurring at optimal salinity are summarized in Figure 1-9 (107).
Coalescence Process. While ultralow interfacial tension is required to mobilize oil drops trapped in porous media, coalescence of the mobilized oil drops is necessary to form an oil bank. Hence very low interfacial viscosity is desirable to form an oil bank (107). Figure 1-10 shows the role of low interfacial viscosity in microemulsion flooding. A previous study (110) showed that tertiary oil recovery efficiency was almost directly correlated with coalescence rate for the system studied. Vijayso et al.(lll) showed that minimal phase separation time or the fastest coalescence rate of emulsions occurred at optimal salinity. Figure 1-11 shows a correlation between the coalescence rate in emulsions and apparent viscosity in the flow through porous media. The minimal apparent
viscosity for the flow of emulsions in porous media coincides with minimal phase separation time at the optimal salinity. This
Oil Recovery Efficiency
Apparent Viscosity of Emulsions in Porous Media Coalescence or PhaseSeparation Time
Surfactant Loss in Porous Vo Media
Solubilization of Oil And Brine in Middle-Phase V W Microemulsions
mW Interfacial Tension of Microemulsion with Excess Oil or I mo Brine Phase
ow I Interfacial Tension between
Excess Oil and Brine Phase in 3-Phase System
Optimal Salinity Salinity
Figure 1-9. The phenomena occurring at the optimal salinity.
A Very Low Interfacial Viscosity Is
Desirable for Coalescence of Displaced
Oil Ganglia to Form an Oil Bank
Figure 1-10. The role of low interfacial viscosity in promoting coalescence of oil ganglia.
Over 10 Days (0% NaCl) T = 250C
System:Sonicated Emulsion Containing
7 Days TRS10-40+IBA(5:3)+Water+NaC1
and Equal Volume of Dodecane 30
S870 5504 5/ 0
a.. -I- - e
0 I 45
0 1 I 1 1 1 I
0 2 4 6 8
NaCl Concentration (Wt.%)
Figure 1-11. Correlation between the apparent viscosity and
coalescence rate of sonicated emulsions.
correlation between the phenomena occurring in porous media and outside the porous medium allows us to use coalescence measurement as a screening criterion for many surfactant formulations in predicting their possible behavior in porous media.
Mobility Control. As mentioned earlier, only a small amount of the microemulsion slug is injected in microemulsion flooding due to the cost of surfactant. Usually a polymer solution is injected after the slug injection. The viscosity of the surf actant slug should be higher than that of oil, and the viscosity of the polymer solution should be higher than that of the surfactant slug to control the mobility during the oil recovery process. Again due to the economic reason, water or brine is injected after the small amount of polymer solution (usually 0.5 to 1.0 pore volume). The viscosity of the microemulsion can be controlled by choosing a proper alcohol (112). Isopropyl alcohol (113) and tertiary amyl alcohol (114) usually cause a viscosity reduction. A polymer may be added to the microemulsion to increase its viscosity. However, it
may cause a problem, namely, polymer-surfactant incompatibility resulting in the surfactant separation and the high interfacial tension (107, 115, 116). Several researchers (117-120) have discussed how to determine oil/water bank mobility in microemulsion flooding.
1.3.4. Effect of Alcohols
In general, a microemulsion formulation for enhanced oil recovery includes a short-chain alcohol as a cosurfactant (112).
The possible effects of such an alcohol include changing viscosity, lowering interfacial tension, reducing interfacial viscosity or changing surfactant partitioning and modifying the solubility of a surfactant in oil or brine phase (107, 112, 121). A previous study (122) suggested that alcohol promotes the mass transfer to the interface and a rapid reduction in the magnitude of the interfacial tension. A mass transfer of an alcohol from the surfactant slug to the oil ganglia in the porous media was shown by Pithapurwala et al. (123).
Iskanderani (124) showed that the separation of an alcohol from a surfactant during the flow through a porous medium led to poor oil recovery. Even though the change in the composition of the surfactant slug cannot be avoided during the flooding due to the dilution and chromatographic effects of a porous medium, the separation of alcohol from surfactant is minimized at optimal salinity.
Lam (125) did a microvisual study of the residual oil mobilization using a relatively simple chemical system. It was
found that under certain circumstances the capillary number required to mobilize the oil drop could be reduced greatly if the displacing phase contained more alcohol than it would in equilibrium with the trapped phase.
1.4 Interaction between Surfactants and Polymers
It has been shown that the solubilization properties of ionic surfactants in water are changed by addition of nonionic polymers
even if these polymers alone show hardly any solubilization effect (126-129). Polyvinyl formal (PVF) and polyvinyl butyral (PVB) are insoluble in water and in the usual organic solvents. However,
these water-insoluble polymers may be dissolved in highly
concentrated surfactant solution (130). Isemura et al. (131) assumed that polymer molecules are solubilized by oriented
adsorption of surfactant molecules in which the nonpolar tail is directed to the polymer and the polar head to the aqueous phase and that the complex would, then, behave like a polyelectrolyte in
contrast to the assumption that the nonionic polymer may be solubilized in the micelles. Experimental results from an electron micrograph, viscosity and electrophoretic mobility supported the former assumption (132).
In constrast to the hydrophobic bonding between nonionic polymers and anionic surfactants, Scholtan (133) suggested the iondipole interaction between nonionic polymers and anionic surfactants. This was partially supported by Saito (134). He experimentally showed that the ionic group of an anionic surfactant might play a significant role in addition to the hydrophobic bonding by the nonpolar parts between the surfactant and the polymer. Further, he showed that the oxyethylene group of surfactants interfered with the binding of the surfactants to the polymer.
Figure 1-12 shows the surface tension as a function of the concentration of sodium dodecyl sulfate (SDS) at a fixed concentration of the water-soluble polymer, polyethylene oxide (PEO)
0 With Polymer
0 Without Polymer
3x10-3 5x10-3 1xl0-2 1.5x10-2
Concentration of SDS (mol/1)
Figure 1-12. Surface tension depending on the concentration of sodium n-dodecyl sulfate (SDS) with and
without polymer (polyethylene glycol) (0.02 M).
(135). There are two transitions compared to only one transition for a surfactant solution alone at CMC. Jones (135) interpreted the transition as follows: 1) Below the first transition, no interaction between the polymer and surfactant occurred. The surface tension and conductance changed in the same manner as expected in the absence of polymer; 2) at the first transition, surfactants started to be adsorbed onto the polymer in similar fashion to micelle
formation; 3) between the first and the second transitions, surfactants were continuously adsorbed onto the polymer until the binding sites were filled; 4) at the second transition, two species are possible, ordinary SDS micelles and polymer-surfactant complex which differ from those below the second transition.
Saito (136) studied the solubilization of aliphatic and aromatic compounds in the surfactant-polymer complex. He concluded that solubilization depends mainly on the compatibility of the polymer with the solubilizate. He also concluded that the function of bound surfactants in solubilization by the complexes was to give the polymers a more hydrophobic character by replacing water layers around the polymers with surfactants and thus to promote contact between the polymers and the solubilizates. This conclusion is consistent with his earlier study (137) in the spectroscopic
investigation of a dye solubilized in the complex.
Fishman and Eirich (138) did equilibrium dialysis measurements for an SDS and poly (N-vinylpyrrolidone) (PVP) system. They suggested that PVP acted as a nucleating agent for surfactant
micelles and stabilized surfactant clusters larger than dimers but smaller than observable in the absence of polymer, that is, the formation of a mixed polymer-surfactant micelle.
It is known that cationic surfactants interact weakly with nonionic polymers (127, 139, 140) while anionic surfactants interact strongly with such polymers (126-128, 135, 139, 141). While the complex formation with anionic surfactants increased by addition of salts, no complex was formed with cationic surfactants (126, 127). These differences between interactions of cationic and anionic surfactants with nonionic polymers (poly glycol esters) were studied by Schwuger (142). He measured surface tension, conductivity, dye solubilization and the effects of charge on pH of medium. On the basis of these measurements, he concluded as follows: 1) For anionic surfactants, the interactions between the surfactants and the polymer in water should be partly chemical and partly physical in nature. It appears that electrical interaction is significant, too. The oxygen atom of the ether linkage with its lone electron pair could possibly become slightly positively charged, which would favor an adsorption of the negatively charged surfactant anion. This effect is enhanced by the hydrophobic interaction between the hydrophobic parts of the surfactant and the polymer; 2) for cationic surfactants, there is a repulsion between the positively charged cationic surfactant ion and the ether oxygen atom. In this case, not only is no electrical attraction operative, but, as a
consequence, there is a weakening of the hydrophobic bonding between the hydrophobic sections of the two species. His conclusions
essentially agreed with those proposed for the interaction of SDS with polyvinyl acetate (PVA) (141) and PVP (143).
Smith and Muller (144) studied the chemical shift of the anionic surfactant and nonionic polymer by nmr and dialysis analysis. They showed that the shape of the binding isotherm is not consistent with a Langmuirian binding process. This does not agree with the previous study (136) which showed that the adsorption of surfactant on the polymer molecule could be expressed in terms of the Langmuir adsorption formula.
While there is evidence that cationic polymers do not interact with nonionic surfactants (145, 146), Goddard and Hannan (146) found that cationic polymers interacted with anionic surfactant by the formation of highly surface active polymer-surfacrant complex resulting from head to head adsorption of the surfactant onto the polymer. The observed phenomena were similar to the previous studies for the oppositely charged polymer-surfactant system (147-150): As the concentration of the surfactant increased, turbidity and then precipitation were observed; resolubilization of the precipitate occurred at still greater surf actant concentrations. They also observed the two transitions in the surface tension versus concentration of surfactant similar to those represented in Figure 1-12. They explained the first plateau as head-to-head adsorption of the surfactant onto the polymer while the resolubilization at the second plateau occurred by tail to tail adsorption of a second layer of surf actant ions to form a polyanion. Their film balance study
provided evidence of the reaction between polymer and surfactant.
Recently, two different models were proposed to explain the interactions between the polymers and the surfactants by Nagarajan (151) and Ruckenstein et al. (152). In Nagarajan's model, surfactant aggregates are adsorbed into the free space of the coiled macromolecules. The latter participates in the formation of aggregates via segments that penetrate the interfacial region of the micelles. In contrast, Ruckenstein's model does not involve the penetration of the micellar interface by segments of the polymer. Instead, his model involves the adsorption of micellar aggregates in the free space of the coiled macromolecules.
In summary, most of the studies were done for the nonionic polymer with the anionic or nonionic surfactants and for the oppositely charged polymer-.surfacrsnt systems. The different properties found in these systems from those in surfactant solution alone were explained by complex formation or the formation of micelles associated with polymer.
Few references are found for similarly charged systems. These systems are reviewed in the later chapter related to the surfactantpolymer interaction between the similarly charged systems.
1.5 Thermal Conductivity of Liquids
Thermal conductivity is a fundamental thermophysical property that is essential for almost all heat transfer application. Reliable thermal conductivity data are needed to interpret many physical and chemical processes as well as to design the equipment
for thermal processes. For example, if the thermal conductivity of the medium with its other properties such as density, viscosity sod heat capacity is known, heat transfer rates of a medium flowing through the tubes can be estimated using the dimensionless relationship (153) shown as
C / 2/3 P 0.8 n-~0.467 = hi Di0.2/0.023 Via .8 (1-5)
where C p= specific heat
K = thermal conductivity
p = density
n = viscosity
hi = heat transfer coefficient
Di = inside diameter of the tube
The thermal conductivity of a material is the ability of the material to conduct heat and is defined by the Fourier equation (154):
qx/A =-K (dT/dX) (1-6)
where qx heat transfer rate in the x direction
A = area normal to the direction of heat flow
K = thermal conductivity
dTldx = temperature gradient in the x direction.
Energy transfer by conduction is accomplished in two ways (154). The first mechanism is that of molecular interaction, in which the greater motion of a molecule at a higher energy level imparts energy to adjacent molecules at lower energy levels. This type of transfer is present, to some degree, in all systems in which a temperature gradient exists and in which molecules of a solid, a liquid, or a gas are present. The second mechanism of conduction heat transfer is by free electrons. The free-electron mechanism is significant primarily in pure-metallic solids.
Therefore, the ability of solids to conduct heat varies directly with the concentration of free electrons. Since the mechanism of conduction heat transfer is one of molecular
interaction and the molecular behavior of gas, especially dilute gas, is well known, it is relatively easy to develop an equation to estimate thermal conductivity using the result of the kinetic theory of gases with the energy transfer equation (154). In contrast, the thermal conductivity of a liquid as well as a dense gas cannot be easily predicted because the molecular behavior of the liquid phase (as well as the dense gas) is not clearly understood (155). The fact that the mechanism of energy transfer between gas and liquid phases is different can be understood by recognizing the difference between transport-property values in the gas phase and those in the liquid phase: Kl/Kg = 10 to 100, nl/ I = 10 to 100, Dl/D 9 10-4 where K, and K. are thermal conductivity of liquid and gas respectively, n I and Tig are viscosity of liquid and gas
respectively, and D, and Dg are diffusivity of liquid and gas respectively.
In the gas phase, the molecules are relatively free to move about and transfer momentum and energy by a collision mechanism. The intermolecular force does not drastically affect the values of K, T1, or D (156). In a liquid, however, the close proximity of molecules to one another influences strongly the intermolecular forces. Energy and momentum are primarily exchanged by oscillation of molecules in the shared force fields surrounding each molecule (157-159). The use of the information about the intermolecular forces and the structure of the liquid in developing the theory of thermal conductivity of the liquid (160-166) indicates a significant effect of the molecular structure of the liquid on thermal
conductivity. Even though a great effort has been made to develop the theory for thermal conductivity of liquids and some progress has been made, a rigorous theory has yet to be proposed.
The prediction of the thermal conductivity of the liquid mixture is even more difficult than the prediction of the thermal conductivity of pure liquids. Even though a number of correlations have been proposed (156, 167-170), only two of them, the Li (168) and power law (156) equations, can be applied to ternary systems. In addition, the experimental data for ternary systems are not widely available (171). The local composition model (172) treats deviations from the ideal method (a mass fraction average of the pure component values: K = E Wi Ki Where Wi is mass fraction of
component i and Ki is pure component thermal conductivity of component i) in terms of deviations of the local composition from random mixing considering binary interactions. However, a recent study (171) showed that further experiments were necessary to check the importance of the ternary interactions because the model performed poorly on some systems. Again, it is expected that the interaction between molecules and the molecular structure of the liquid mixture affect the thermal conductivity of the liquid mixture.
It should be mentioned that the corresponding states techniques for prediction of thermal conductivity have been developed for both liquids (152) and liquid mixtures (172). These methods have a wide range of temperature-pressure applicability, but lack accuracy in predicting composition dependency unless adjustable parameters are used in the mixing rules. In the corresponding states are also implied the effect of the molecular interactions and structure of the liquid or the liquid mixtures on thermal conductivity of the liquid or liquid mixtures.
In summary, thermal conductivity of a liquid depends on the molecular interactions and the molecular structure of the liquid, and that these factors should be taken into account to develop the theory for thermal conductivity of liquids. Because of the
complexity and lack of knowledge of the molecular behavior of liquids, theories to date have not led to a simple estimating technique for liquid thermal conductivity (173). Therefore,
approximate or empirical techniques must be used for engineering applications (156).
1.6 Phase Stability of the Gasoline-Methanol Mixture
The use of alcohol in motor vehicles is not a new technology. An exhaustive international bibliography prepared on the subject can be found before 1920 (174-176). Alcohol fuels have been used in both wartime and peacetime in the United States, and are currently attracting renewed attention as petroleum prices increase and supplies remain uncertain (177). Even though alcohols other than methanol have been considered for blending with gasoline, the current spotlight is on methanol (177, 178). It is known that up to 15% methanol can be added to gasoline in current cars without modification of the engine (178). The addition of methanol to gasoline is advantageous not only in economy but in exhaust quality and performance (179-185). Methanol has an octane rating of 106, compared to typical gasoline ratings of 85 to 100 (186). It prevents knocking and alleviates dieseling when the ignition is switched off, The results for the emission tests are not consistent: While Paul (179) reported that both nitrogen oxides (NOx) and carbon monoxide emissions were decreased considerably with methanol or with
gasoline-methanol blends, two other studies (184, 185) reported an overall reduction in carbon monoxide but no consistent and overall change in the NOX and hydrocarbon emissions. Table 1-1 summarizes the improvement in various performance properties achieved by blending 5 to 30% methanol to 95 to 70% gasoline (187).
Table 1-1. Improved Performance by Blending 5 to 30 Percent
Methanol to 95 to 70 Percent Gasoline
Fuel Economy Increase 5-10
CO Emission Decrease 14-72
Exhaust Temperature Decrease 1-9
Acceleration Increase 7
While the gasoline-methanol mixture offers several advantages, It also has disadvantages compared to gasoline (182): hesitation on warmup, corrosion and phase separation due to lower temperatures or due to absorption of water. Among these, phase separation at low temperatures is found to be most serious problem in practice while the phase separation resulting from the absorption of water is less of a problem in practice because motor fuels are not freely exposed to the atmosphere (186, 188).
In practice, the chemical stability of fuel in storage is a function of the structure of its components, together with its purity. With the hydrocarbons, for example, stability is provided by saturation of the carbon-carbon bonding in the straight-chain paraffins and in the close-chain naphthenes (above C5). The alcohols are stable compounds but, as polar molecules, are active solvents. The alcohols are, however, less soluble in the non-aromatic (aliphatic) hydrocarbons; therefore its solubility in commercial petroleum fuels depends largely on the hydrocarbon constituents of the latter. Solubility also increases with temperature. Ethanol and the higher alcohols dissolve more readily in hydrocarbons, and their presence in an alcohol-hydrocarbon blend increases the solubility of methanol (189, 190). These features are shown in Figure 1-13 for methanol and fur ethanol blended separately with kerosene using isobotanol as a stabilizer. As mentioned earlier, the presence of even small quantities of water in alcohol-hydrocarbon mixtures, however, leads to direct hydrogen bonding between the alcohol and the water
Ethanol Methanol I\\ Miscible "' Isobutanol Kerosene
Figure 1-13. Solubility of alcohols in kerosene.
resulting in phase separation, particularly at low temperatures
(191). Effect of water on the phase separation of gasoline-alcohol mixture is shown in Figure 1-14. The problem of phase separation at low temperatures due to the small amount of water has been investigated by several researchers (191-196) using various surfactants to form the w/o microemulsions.
In summary, the gasoline-methanol mixture has several advantages compared to gasoline as an automobile fuel. However, the phase separation of the mixture, especially at low temperatures, due to the small amount of water in the mixture presents a serious problem. Even though several researchers showed that the formation of w/o microemulsion using several surfactants was effective in preventing the mixture from phase separation, no systematic study has been done to date.
The major goal of this dissertation is to investigate the properties of microemulsions such as solubilization, interfacial tension, thermal conductivity and phase stability as well as the excluded volume effect of polymer in the surfactant solution. The solubilization and interfacial tensions of microemulsions which used nonionic surfactants as cosurfactants in place of short-chain alcohols were examined. Microemulsions prepared by replacing a
short-chain alcohol with a proper nonionic surfactant exhibited better properties in relation to solubilization and oil displacement in porous media. Thermal conductivity studies of microemulsions
Figure 1-14. Solubility of alcohols in gasoline in the
presence of alcohol.
indicated the effect of microstructure on thermal conductivity. The possible use of the microemulsions as lubricant oil, heat transfer fluid, etc. is discussed. The phase stability of the
methanol-isooctane-toluene mixture was studied at -25 0C using alcohols and/or surfactants to determine the structural effect of the additive(s). The results are relevant to the the development of alternative fuels. The mechanism of the decrease of interfacial tension of polymer solution by the addition of a small amount of a surfactant was investigated using static and dynamic light
Following the brief review of microemulsions, enhanced oil recovery and thermal properties of liquids, Chapter 2 reports the relationship between the solubilities of water in alcohol-free w/o microemulsions and the efficiency of the oil displacement by the microemulsions. Further, the displacement mechanism for the alcohol-free w/o microemulsions is proposed which was in qualitative agreement with the oil displacement experiments. In Chapter 3, the advantages of using a nonionic surfactant as a cosurfactant instead of short-chain alcohols for the displacement of heavy oils are dicussed.
Chapter 4 focuses on the effect of polymer molecules on
surfactant solutions. The results indicate that the mechanism of decrease in interfacial tension of anionic polymer solutions upon addition of a small amount of anionic surfactants is much different from that of nonionic polymer-ionic surfactant solutions. The
excluded volume of polymer was obtained by light scattering experiment to support the proposed mechanism.
Chapter 5 experimentally demonstrates the effect of the microstructure of a liquid on thermal diffusivity using two liquids having much different microstructures. As an extension, thermal conductivities of microemulsions are measured and are compared with predicted values using existing equations for liquid mixtures. A simple transient method to measure the thermal diffusivity of a liquid is developed for the study and is described.
Chapter 6 reports the effect of surfactants and/or alcohols on the phase stability of a methanol-isooctane-toluene mixture at -250 and further delineates how the molecular structures of various surf actant and/or alcohols influence the phase stability of the mixture at -25 C. The possibility of the formation of a waterless microemulsion is discussed.
Finally, Chapter 7 summarizes the conclusions from the entire study and provides recommendations for future research in areas related to this dissertation.
ALCOHOL-FREE WATER-IN-OIL MICROEMULSIONS
Microemulsions for enhanced oil recovery have received much attention because of their ability to lower the interfacial tension between residual oil and reservoir water (or brine), resulting in enhanced oil recovery. All three microemulsion structures, namely, the lower-, middle-, and upper-phase microemulsions, have been studied for enhanced oil recovery (76-78, 93, 101, 102, 104, 109, 114, 197-204). The lower- and upper-phase microemulsions are o/wand w/o microemulsions, respectively, while several microstructures of the middle phase microemulsions have been proposed. Scriven (50) proposed that the middle-phase microemulsion may have a bicontinuous structure of oil and water with intervening surfactant layers. Shah et al. (205) and Miller et al. (73) believe that the middle phase microemulsion may be coacervated droplets. Sinoda and Friberg (13) proposed that the middle-phase microemulsion is an essentially lamellar structure with alternating layers of oil and brine. The effect of the structures of the lower- and the upper-phase
microemulsions on the oil recovery is not clearly understood due to difficulty in conducting experiments because a change in the structure of the microemulsion always accompanies changes in other 52
parameters such as the concentrations of the components, viscosity, interfacial tension, and so on (206). However, regardless of their structural differences, these microemulsions commonly employ shortchain alcohols as cosurfactants (112). While a short-chain alcohol is necessary to form the microemulsion by increasing the fluidity of the interface, it has less solubilization capacity than a
surfactant (102). Even though short-chain alcohols as a cosurfactant in a microemulsion are successful for light oil recovery, they are less effective for heavy oil recovery due to poor solubilization capacity of heavy-oil microemulsions. The interfacial tension of the microemulsion with excess oil or brine is high due to the small solubilization of heavy oil in the
microemulsion containing a short-chain alcohol. It is known that the solubilization of brine in a w/o microemulsions decreases as the chain length of oil increases at corresponding optimal salinity (207).
An alcohol-free w/o microemulsion was formulated by replacing a short-chain alcohol with a nonionic surfactant. By using a nonionic surfactant instead of a short-chain alcohol, the solubilization of brine in heavy oil containing microemulsion increased. After testing several nonionic surfactants with an anionic surfactant, Aerosol-OT (AOT), it was found that the system consisting of AOT and Arlacel 20 was best for solubilizing water in the heavy oil
microemulsion. The formulated alcohol-free w/o microemulsion showed that its solubility of water depended on the weight ratio of the two surfactants which, in turn, influenced oil recovery. The oil
displacement mechanisms by the w/o microemulsions which were
proposed were qualitatively supported by oil displacement
experiments. In addition, the effects of concentration of the electrolytes and that of surfactants on the oil recovery by the w/o microemulsion were studied.
2.2 Materials and Methods
Mineral oil supplied by Witco Chemical Corporation was used to simulate a heavy oil. Specific gravity and viscosity of the mineral oil were 29.2 API and 240 cP at 60 OF (160 cP at room temperature), respectively. An anionic surfactant, AOT (sodium dioctyl sulfosuccinate), was purchased from American Cyanamid and a nonionic surfactant, Arlacel 20 (sorbitan monolaurate), was a gift from ICI Americas, Inc. Their chemical structures are shown in Figure 2-1. De-ionized, distilled water was used in all experiments. Brine was prepared by adding NaCl into de-ionized, distilled water. Sodium chloride was purchased from Fisher Scientific Company and was of certified grade. The oil-soluble dye, Sudan III was purchased from Eastman Kodak Corporation and the polymer Pusher 700 (polycrcylamide) was purchased from Dow Chemical Company. All chemicals were used as received without further purification.
Unconsolidated sand packs were used for oil recovery
experiments. Figure 2-2 schematically shows the experimental set-up for the oil displacement experiments. Each sand pack was prepared
CH C"3 0
H3CI-12 ---Cl-12 CH3 ,: CH2
I C'-12 CI-12 I
CH27CII, ': CH-CI-I2
-- C H2
"CH2 Cll: CH2,
0--, _CH2 cl-0 --CH2
0 WATER qH2ood CH2
S03 H OH
Na + H H---]
Aerosul-O'P Olq 0
(Sodium Dioctyl SuLfosuccinate)li OH
Figure 2-1. Structures of aerosol-OT and arlacel 20.
Pump Valve Sand Pack
Transducer |Transducerr Collector
Figure 2-2. A schematic diagram of the experimental set-up for oil displacement
by filling a cylindrical glass tube with sand while continuously shaking and tapping the tube. The dimension of a sand pack was 0.95 inch in diameter and 2 feet in length. The average porosity of a sand pack was 0.39. Each sand pack was used for a single
Solubility of Water. To measure the maximum solubility of
water, the alcohol-free water-in-oil microemulsion was prepared by first dissolving the surfactants into mineral oil. The resulting solutions contained 0.4g surfactants/ml mineral oil and the weight ratios of AOT to Arlacel 20 varied from 0 to 3. These solutions were titrated with water at room temperature ( 25 C 1 C). Water was added drop-by-drop into the oil-AOT-Arlacel 20 solution and the solution was stirred. After each addition of water, the solution was checked for clarity and then for birefringence using polarizing plates. The amount of water required to make the solution turbid and/or birefringent was considered the end point for maximum solubilization.
Phase Behavior. Phase behavior was studied at the weight ratio, 13:27, of AOT to Arlacel 20 at which the solubility of water in the microemulsion was maximum. Nine samples having different weight ratios of surfactants to mineral oil were first prepared and then water was added. The amount of water in each addition was 0.1 ml. After each addition, the samples were placed in a constant
temperature container (300C) for 3 days and were then checked to see whether these showed one or two phases. In the study of phase behavior, an oil soluble dye was added into mineral oil to check the continuous phase of the microemulsion.
Viscosity. Viscosity of the microemulsions was measured as a function of water content using a Brookfield LVT type viscometer at shear rate of 5.75 sec-I.
Electrical Resistance. Electrical resistance of the alcoholfree w/o microemulsions was measured as a function of water content using electrodes connected to Beckman Conductivity Bridge Model RC 16B2. Two glass-sealed silver wires (diameter = 0.16 cm) were used as electrodes. A one cemtimeter length at the end of each wire was exposed outside the glass tube, and the Ag-AgCi electrodes were separated by 0.8 cm. The electrodes were electro plated in dilute HCI to produce Ag-Cl electrodes. The electrical resistance of the microemulsions was measured by dipping the electrodes into the microemulsions.
Oil Recovery. The entrapped air in a sand pack was displaced by the flow of carbon dioxide. Then the pore volume of a sand pack was measured by flowing water. Next, oil was pumped in until no water came out. Saturated oil was then displaced by water until the production ratio of oil to water was less than 0.01. Residual oil
after water flooding was displaced by the 0.1 pore volume of the microemulsion followed by the half pore volumes of a polymer aqueous solution followed by water injection until no oil came out. During oil recovery experiments, the flow rates of fluids were 1 ft/day except during the initial water and oil saturation. To test the effect of salt on the oil recovery, brine (2.6% NaCl, w/v) was used except for tertiary oil recovery in which only de-ionized, distilled water was used.
2.3 Solubility of Water in the Alcohol-Free Water-in-Oil Microemulsions
The combination of Arlacel 20 and AOT was chosen from the
several combinations of nonionic and anionic surfactants because this combination had been found to be the best system for the
solubilization of water in the w/o microemulsions in a previous study (208). The solubility of brine in the w/o microemulsion should be large enough to produce the low interfacial tension
between water and the microemulsion. The previous study had also shown that the amount of water solubilized in the alcohol-free w/o microemulsion depended on the weight ratio of these two chosen surfactants and the kind of oil used. It was shown that the
interfacial tension depended on the solubility of water in the w/o microemulsion. Therefore the solubility of water in the w/o microemulsion was first investigated using the mineral oil.
The results are shown in Table 2-1 and Figure 2-3. For mineral oil, the maximum solubility of water was obtained at the weight
Table 2-1. Solubility of Water in the Alcohol-Free W/O
Ratio of AOTa 0 10 11 13 14 15 20 25 30
by Weight A20b 40 30 29 27 26 25 20 15 10
to Oil Ratio 0.04 0.23 0.28 0.83 0.81 0.76 0.54 0.22 0.06
by Volume _I_
a. AOT means an anionic surfactant, Aerosol-OT.
b. A20 means a nonionic surfactant, Arlacel 20.
c. The system consisted of 4 g of total surfactant in 10 ml of
mineral oil as the initial solution. The water was added to this
solution untiil it became turbid.
- 0.8 8180
0. 6 c 60
8 0 A2
0 0 / 60Cl
S 0.4 C 1.
.0 0.2 20
0 10 20 30 40 AOT
40 30 20 10 0 Arlacel 20
Initial Concentration Of Surfactants In Oil (%,W/V)
Figure 2-3. Solubility of water in the water-in-oil microemulsion,
Compositions of the slugs Al, A2, B1 and Cl are
represented in Table 2-3.
ratio, 13:27, of AOT to Arlacel 20 and the volume ratio of water to oil was 0.83. It had been shown that the ratio of surfactants for the maximum solubility of water had been shifted towards more Arlacel 20 as the chain length of oil increases (208). Therefore the weight ratio of the surfactants for mineral oil was reasonable compared to the weight ratio, 20:20, for hexadecane in the previous study (208). It should be noted that the solubility sharply
decreased in both directions from the ratio having the maximum solubility. In other words, the capability of the w/o microemulsion to accommodate water can be dramatically changed by a small change in the composition of the chosen surfactants. This fact emphasizes the necessity of selecting a proper combination and composition of surfactants for the high water solubilization in the w/o
microemulsion, because solubility is related to the efficiency of oil recovery as shown in Figure 2-3: the oil recovery was maximum (76.2%) when the solubilization of water in the w/o microemulsion was largest.
2.4 Phase Behavior of the Water-In-Oil Microemulsions
Figure 2-4 shows the pseudo-ternary diagram. One phase region was large and appeared to be oil continuous microemulsion determined as follows. The oil soluble dye, Sudan III was added to the microemulsion. The dye quickly dissolved in the external oil phase of the microemulsion. In contrast, in the water-external system, the uptake of the dye was very slow or none at all. This was
supported by measurements of viscosity and electrical resistance as
AOT + Arlacel 20
Water Mineral Oil
Figure 2-4. Phase diagram. P represents the typical
composition of the water-in-oil microemulsion
a function of the amount of water in the microemulsion. Figure 2-5
shows the results of viscosity measurements. As the amount of water in the microemulsion increased, viscosity also increased. If phase inversion from oil to water phases had occurred, there should have been an abrupt change in viscosity (209, 210). As represented in Table 2-2, the values of electrical resistance were greater than 7.8 x 105 ohms, indicating oil continuous phase in the microemulsion (210). Up to the ratio, 0.6, of water to oil, electrical resistances were out of range indicating that their electrical resistances were greater than 2.5 x 106 ohms. Based on the results from three independent techniques, namely, dye solubilization, viscosity and electrical resistance measurements, it was concluded that the microemulsions formed were oil-external.
2.5 Proposed Mechanism of Oil Displacement by Water-in-Oil Microemulsions
The slug can be of any composition in the one phase region around the boundary line in Figure 2-4. The region above the broken line was not of interest due to the economic considerations. Typical composition (P) of the slug is shown in Figure 2-4. Tn Figure 2-6, (a) shows the relative volumes of water and oil in the porous medium just after the secondary oil recovery by water, while
(b) shows that after slug injection. After injection of the slug, two phases are formed, namely, oil-external microemulsions and excess water. Since the w/n microemulsion is oil continuous, it can be miscible with residual oil while it is immiscible with residual
0 0.3 0.6 0.9
Water To Oil Ratio (V/V)
Figure 2-5. Viscosity of the water-in-oil microemulsion as
a function of the amount of water added to the oil plus surfactants mixture. Surfactant ratio
= aersol-OT/arlacel 20 = 13/27 (w/w).
Table 2-2. Electrical Resistance of the Microemulsions Depending on
the Amount of Water in the Microemulsions Ratio of Water to
Oil by Volume 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Resistance (ohms) Greater than 2.5xi06 9.0x105 7.8xi05
I Residual Oil Out
Water Left Behind The Slug
Figure 2-6. A schematic presentation of the relative volume
of oil and water in a porous medium during oil
displacement by the water-in-oil microemulsions.
(a) after water flooding; (b) surfactant slug
injection; (c) oil bank formation; (d) the injection of polymer solution. X-axis represents the distance
along the porous medium and Y-axis represents the
relative volumes of oil and water.
water. Therefore the residual oil contacted by the slug is miscibly displaced and starts to form the oil bank. Most of residual water is displaced immiscibly by the slug, but some is carried by the front part of the slug in the form of win macroemulsion (211), while a fraction of water and a fraction of the slug are left behind, depending on the interfacial tension produced by the w/o microemulsion as shown in (c). Then, the water left behind the slug is miscibly displaced by polymer solution and starts to form a water bank. If this water bank takes over the front part of the slug, then the fractional flow of oil in the stabilized oil bank is reduced as shown in (d). Further, if the water bank takes over even the front part of oil bank, the high fractional flow of oil bank (or oil cut) cannot be obtained, resulting in poor oil recovery. This will result in the leaky-piston mechanism (d) of displacement (211). However, if the w/o microemulsion can produce a sufficiently low interfacial tension, the amounts of water left behind the slug can be reduced and the formation of water bank is delayed, resulting in good oil recovery. In this case, the mechanism of displacement is more piston-like (c).
2.6 Experimental Results of Oil Displacements
Oil recovery experiments were designed to elucidate the effects of the following variables: 1) effect of compositions of the w/n microemulsions, 2) effect of the amount of water in the w/o microemulsions, 3) effects of polymer flooding, 4) effect of oil as
the slug, and 5) effects of the concentration of salt and surfactant in the slug on tertiary oil recovery.
2.6.1 Effect of the Composition of the W/O Microemulsions
on Oil Displacement
Four different compositions of the microemulsions shown as
AI,BI,C1, and A2 in Figure 2-3 and Table 2-3 were used as the microemulsion slugs for tertiary oil recovery. Now consider three microemulsion slugs, Al, BI, and Cl. Each w/o microemulsion slug contained 80% of its maximum solubility of water. Figure 2-3 and Table 2-3 show the experimental results. Tertiary oil recovery was the best for composition B1. This was expected because the better solubilization capacity of water in the w/o microemulsion would result in lower interfacial tension at microemulsion/water or polymer solution interface which, in turn, resulted in the better efficiency of oil recovery. These experimental results can be viewed in another way by investigating the production histories as shown in Figure 2-7. Every curve consists of three linear parts with different slopes. The first linear part with the lower slope up to 0.4 pore volumes of injected fluids was due to the formation of oil bank and the second linear part with the higher slope was due to the flow of the stabilized oil bank. The final linear part with the zero slope was due to surfactant breakthrough. Since the residual oil was viscous, the fractional flow of oil in the stabilized oil bank (called oil cut) was high as shown in Figure 2-8. The values of oil cut in the stabilized oil bank were almost same for the three
Table 2-3. Effect of Composition of the Microemulsion Slugs
on Oil Displacement Efficiency
Slug Symbol Al B1 C1 A2
Mineral Oil 10 ml 10 ml 10 ml lOml
of AOTa 0.5 g 1.3 g 2.5 g 0.5 g
Slug A20b 3.5 g 2.7 g 1.5 g 3.5 g
Water 0.8 ml 6.64 ml 1.76 ml 0.8 ml
Volume % of Surfactants
in microemulsion Slug 23.9 17.0 24.6 23.9
WORmaxc 0.1 0.83 0.22 0.1
Size of Slug injected 0.1 0.1 0.1 0.076
Pore Volume of Sand
Pack (PV) III ml 105 ml 103 ml 104 ml
Porosity of Sand Pack 0.405 0.383 0.376 0.379
Oil Saturation (% PV) 82.0 88.6 85.4 88.5
Oil Recovery (%) 46.2 43.0 50.0 47.8
Residual Oild( %) 53.8 57.0 50.0 52.2
Recoverye (%) 64.1 76.2 43.1 56.4
Recovery (%) 79.3 85.8 69.6 76.1
Emulsion or Surfactant
Break-through (PV) 0.68 0.81 0.57 0.66
a. AOT means an anionic surfactant, Aerosol-OT. b. A20 means a nonionic surfactant, Arlacel 20. c. is the maximum water to oil ratio by volume in the microemulsion. d. Calculated based on the saturated oil. e. Calculated based on the residual oil.
0 0.5 1.0
Pore Volumes Of Injected Fluids(PV)
Figure 2-7. Effect of composition of the water-in-oil microemulsion on oil displacement
efficiency. The compositions of the microemulsion slugs Al, A2, B1 and Cl
are represented in Table 2-3.
- 0.50 4-1 4
0 0.5 1.0 0 0.5 1.
Throughput (PV) Throughput (PV)
1 () fr, ~.
0 0.5 1.0
Throughput (PV) Figure 2-8. Oil production histogram. The compositions
of the microemulsion slugs Al, B1 and C1
are represented in Table 2-3.
different microemulsion slugs; thus the second linear part of every curve has nearly the same slope as seen in Figure 2-7. Figure 2-8 shows that, for the microemulsion slug Cl, the water bank was formed and took over the microemulsion slug at the early stage of tertiary oil recovery, resulting in poor oil recovery with a lower oil cut. A previous study (211) showed that the emulsion breakthrough became faster as the amount of water which could not be displaced by the slug increased. As shown in Table 2-3, Emulsion breakthrough was slowest for Bl. Table 2-3, Figure 2-7, and Figure 2-8 show that the efficiency of oil recovery leveled off after the emulsion came out. It should be noted that the microemulsion slugs Al and C1 contained more surfactants than the microemulsion slug BI because, while the sizes of the slugs were same for all three (0.1 pore volume), the amounts of water in the slugs Al and Cl were much less than that in the slug Bl.
The microemulsion slug, A2, in Table 2-3, and Figure 2-7 had the same composition as the microemulsion slug Al but the size of the slug was reduced from 0.1 to 0.076 pore volumes so that the amount of surfactants in the slug A2 was the same as that in the microemulsion slug BI. This further decreased the efficiency of tertiary oil recovery from 76.2 to 56.4% as shown in Table 2-3 and Figure 2-7. This suggests that the ratio of surfactants, and hence the solubilization capacity is more important than the total amount of the surfactants. Figure 2-8 also shows how much faster the water bank took over slug Cl than slug Bl.
Now, let's try to qualitatively prove the proposed mechanism by using the experimental data. Consider the typical cases, slugs Bl and Cl. The microemulsion at the composition Bl had the best capability to accommodate water while the microemulsion at the composition Cl had the small solubilization of water. Therefore, El produced the lowest interfacial tension while Cl produced the high interfacial tension against excess water phase. If our proposed mechanism is correct, the formation of water bank for the slug Cl should occur earlier than that for the slug Bl because of the larger amount of water left behind the slug Cl due to higher interfacial tension of the 'microemulsion against the residual water at Cl. Indeed, as shown in Figure 2-8, the formation of the water bank occurred at about 0.8 pore volumes of the throughput for the slug El while it occurred at about 0.55 pore volume of the throughput for the slug Cl. Therefore, the lower interfacial tension produced by the slug El resulted in the good oil recovery by the more pistonlike displacement while the higher interfacial tension of slug Cl resulted in the poor oil recovery by the leaky piston displacement.
2.6.2 Effect of the Amount of Water in the W/O Microemulsions
on the Oil Displacement
Additional experiments were performed using the compositions B2 and B3 as the microemulsion slugs represented in Table 2-4. The sizes of the slugs were adjusted so that the amounts of the surfactants in slugs B2 and B3 were the same as that in the slug El. The sizes of slugs B2 and B3 were 0.087 and 0.067 pore volumes,
Table 2-4. Effect of the Amount of Water in the WIO Microemulsion
Slugs with the Same Composition on Oil Displacement Slug Symbol B1 B2 B3
Mineral Oil lOml 10 ml 10 ml
Composition AOTa 1.3 g 1.3 g 1.3 g
Slug A20b 2.7 g 2.7 g 2.7 g
Water 6.4 ml 4.00 ml 0.00 ml
WORmaxc 0.83 0.83 0.83
Total Amount of Surfactant
in the Slug 2.1 g 2.1 g 2.1 g
Size of Slug (PV) 0.100 0.087 0.067
Vol. % of Surfactants
in Slug 17.0 18.9 25.5
Pore Volume of Sand
Pack (PV) 105 ml 104 ml 103 ml
Porosity of Sand Pack 0.383 0.379 0.376
Oil Saturation (% PV) 88.6 84.6 85.4
Recovery (%) 43.0 47.7 47.7
Residual Oild(%) 57.0 52.3 52.3
Tertiary Oil Recoverye (%) 76.2 69.3 67.3
Total OHl Recoveryd (%) 85.8 83.1 82.0
Emulsion or Surfactant
Break-through (PV) 0.81 0.80 0.79
a. AOT means an anionic surfactant, Aerosol-OT. b. A20 means a nonionic surfactant, Arlacel 20. c. is the maximum water to oil ratio by volume in the microemulsion. d. calculated based on the saturated oil. e. calculated based on the residual oil.
respectively. The effect of the initial amount of water in the w/o microemulsion on the oil recovery had a small effect on oil displacement efficiency as shown in Figure 2-9 and Figure 2-10. The small difference in the efficiency of oil recovery seemed to be caused by the different sizes of the slugs because the smaller size of the slug resulted in a little earlier formation of the water bank even though the effect was not significant in this case.
2.6.3 Effect of Polymer Flooding and Oil as a Slug on
the Oil Displacement
To evaluate the better efficiency of oil recovery by the w/o microemulsion flooding than that of polymer flooding, a half pore volume of polymer solution was injected right after the water flooding (polymer flooding). The results are shown in Table 2-5, Figure 2-11, and Figure 2-12. The recovered oil represented about 48.8% which was comparable to the efficiency of the slug Cl ( 43.1% as represented in Table 2-3). This again indicates that the slug Cl did not have an interfacial tension low enough to displace much of water and hence permitted the early formation of the water bank which eventually took over the slug and the oil bank. However when the slug exhibited a low interfacial tension as is the case for BI, it could recover an 27 percent more oil than the polymer flooding by displacing much of water and delaying the formation of the water bank.
Since the w/o microemulsion had an advantage of forming an oil bank due to its compatibility with the residual oil, it was
.,.0- 0 /
0 0.5 1.0
Pore Volumes Of Injected Fluids (PV) Figure 2-9. Effect of water in the water-in-oil microemulsion on oil displacement
efficiency. The compositions of the microemulsion slugs B1, B2 and
B3 are represented in Table 2-4.
0 0.5 1.0
0 0.5 1.0
Throughput (PV) Figure 2-10. Oil production histogram. The compositions
of the microemulsion slugs BI and B2 are
represented in Table 2-4.
Table 2-5. Effect of Polymer Flooding and Oil as Slug on
Polymer Mineral Oil MicroFlooding as the Slug emulsion Pore volume of Sand Pack(PV) 100 ml 105 ml 105 ml
Porosity of Sand Pack 0.365 0.383 0.383
Oil Saturation (% PV) 91.0 85.7 88.6
Secondary Oil Recoverya(%) 49.5 52.2 43.0
Residual Oila(%) 50.5 47.8 57.0
Tertiary Oil Recoveryb(%) 48.8 52.5c 76.2c
Total Oil Recoverya(%) 74.1 74.7 85.8
a. Calculated based on the saturated oil. b. Calculated based on the residual oil. c. Size of oil as the slug was 0.1 PV and this was taken into
account for calculation of oil recovery.
o0 Polymer Flooding
0 0.5 1.0
Pore Volume of Injected Fluids (PV)
Figure 2-11. Effect of polymer flooding and oil as the slug on oil displacement efficiency. The composition of the slug BI is represented in
Polymer flooding Oil As The Slug
0 0.5 1.0 0 0.5 1.0
Microemulsion as the slug
0 0.5 1.0
Throughput(PV) Figure 2-12. Oil production histogram. The composition
of the microemulsion slug Bl is represented
in Table 2-3.
interesting to examine that effect of injected oil instead of the microemulsion slug. Thus, after water flooding, oil alone was injected as the slug followed by aqueous polymer solution. The results are shown in Table 2-5, Figure 2-11, and Figure 2-12. The oil recovery (52.5%) was almost the same as that for the polymer flooding (48.8%). It should be noticed that we did not obtain better oil recovery using oil as the slug rather than by polymer flooding because the interfacial tension between oil and water was high enough so that oils was re-entrapped giving the same value of the residual oil.
2.6.4 Effect of the Concentration of Salt and Surfactant
on Oil Displacement
Effect of Salt Concentration. So far the oil recovery experiments were done using water containing no salt, as shown in Figure 2-13 (a). However, in reality, reservoir water is brine. Since water-in-oil microemulsion, sometimes called soluble oil, decreases its solubilization capacity when water contains salts, previous studies (212) suggested the use of fresh water for tertiary oil recovery. To see the effect of salt on oil displacement, the experiment was modified slightly as shown in Figure 2-13 (b) :brine (2.63% NaCl,W/V) was used initially and for secondary oil recovery. When fresh water was used only for the tertiary oil recovery as suggested by the previous study (212), the oil recovery (72%) was almost equal to that when only fresh water was used for secondary as well as tertiary oil recovery (76.2%), as shown in Figure 2-14.
(a) Water Slug Oil Water With Water
76 2% Oil
Polymer Water Recovery)
(b) Slug Oil Brine With Brine
(72,0? Oil Recovery)
Figure 2-13. A schematic presentation oil displacement
experiments. (a) only water was used; (b) brine was used initially and for secondary flooding. The polymer injection employed
water rather than brine.
m Water Brine
E, 0---~0 0.5 1.0 1.5
Pore Volumes Of Injected Fluids (PV)
Figure 2-14. Effect of salt on the oil displacement efficiency
by the water-in-oil microemulsion slug. The
concentration of NaCl in brine was 2.63% (w/v).
Effect of Concentration of Surfactant. When the concentration of the surfactants in the microemulsion slug was reduced by half, oil recovery decreased by about 5%, as shown in Figure 2-15 indicating that reduction in the concentration of surfactants at the optimum weight ratio did not significantly reduced the oil recovery in the range of surfactant concentration employed here.
Oil production histograms were similar for all three
experiments except for the oil cut for the experiment with brine as shown in Figure 2-16. A little smaller oil cut was probably a result of a little different relative permeability of oil and brine from that of oil and fresh water.
1. Alcohol-free microemulsions produced by Aerosol OT, Arlacel 20,
mineral oil and water were water-in-oil microemulsions which could solubilize a large amount of water (water-to-oil ratio = 0.83) at the optimal weight ratio of surfactants (AOT : Arlacel
20 = 13 : 27).
2. The oil displacement efficiency was also maximum when the weight
ratio of the surfactants was optimum for solubility of water in
the w/o microemulsion.
3. The mechanism of oil displacement by the w/o microemulsion is
proposed. The surfactants at the optimum ratio exhibited a large solubilization capacity of water which, in turn, produced low interfacial tension and high oil displacement efficiency.
Other compositions exhibited a poor solubilization capacity of
m Slug Bl
L Reduced Concentration of Slug Bl
a) 0 E~
0 0.5 1.0 1.5
Pore Volumes Of Injected Fluids (PV)
Figure 2-15. Effect of concentration of surfactants in the
water-in-oil microemulsion slug on the oil displacement efficiency. Slug Bl contained
0.4 g/ml surfactants and reduced concentration
of slug Bl was 0.2 g/ml.
0 0.5 1.0
0 0.5 1.0 0 0.5 i.
Figure 2-16. Oil production histogram. The concentrations of
surfactants in the water-in-oil microemulsion slugs were 0.4 g/ml for (a) and (b) and 0.2 g/ml for (c).
The concentrations of salt in the water-in-oil microemulsion slugs were 0% for (a) and (c) and
2.63% NaCI (w/v) for (b).
water and, hence, produced high interfacial tension and poor oil
4. The proposed mechanism was in agreement with the results of the
oil displacement experiments.
5. The water-in-oil microemulsion consisting of the optimum ratio
of the surfactants was very effective with a viscous oil
(mineral oil) of which viscosity was 160 cP.
ALCOHOL-FREE MIDDLE-PHASE MICROEMULSIONS
Microemulsions for enhanced oil recovery employ short-chain alcohols as cosurfactants (101, 111, 112, 122, 123, 204, 213, 214). Such short-chain alcohols are believed to introduce fluidity into the system to promote the formation of a microemulsion phase (102, 213). As the salinity of brine increases, the surfactant and the short-chain alcohol in the system solubilize more oil and less brine, resulting in the transition of the microemulsion phase from a lower-phase to an upper-phase through a middle-phase. Among the three different types of the microemulsions, the middle-phase microemulsions have received much attention because of its ultralow interracial tension with excess oil and brine and high
solubilization capacity for both oil and brine (102).
Microemulsions containing short-chain alcohols are not very effective for heavy oil recovery. When long-chain oils (> C14)
are used, the middle-phase microemulsion containing the short-chain alcohol does not solubilize much oil or brine (215) as mentioned earlier. Therefore, the volume of the middle-phase microemulsion is relatively small for heavy oil (109). The reduced solubilization of heavy oil in the middle-phase microemulsion does not lead to
ultralow interfacial tension (216). In addition to high interfacial tension, the reduced solubilization of heavy oil in the middle-phase results in the low viscosity of the middle-phase microemulsion. The low viscosity, then, generates the problem of the mobility control of the slug. A polymer may be added to increase the viscosity of the middle-phase microemulsion. However, the addition of a polymer may cause another problem, incompatibility between a polymer and the surfactant in a slug due to the
interaction between polymer and surfactant (101, 217). Further, the phase behavior of the system becomes more complex (217).
The purpose of this study was to show that, by substituting an alcohol with a proper nonionic surfactant, the alcohol-free middlephase microemulsions could solubilize more oil and produce lower interfacial tension, resulting in a significant improvement in heavy oil recovery. By measuring middle-phase volume and interracial tension and by performing oil displacement experiments, a comparison can be made between two different middle-phase microemulsion
systems, an alcohol-free middle-phase microemulsion and a middlephase microemulsion containing a short-chain alcohol.
3.2 The Concept of Solubilization
The basic idea of using a nonionic surfactant in the
microemulsion arose from Winsor's R theory (102, 218, 219). The
theory considered the relative magnitude of the interaction energies of the amphiphile (C) with oil (0) and water (W). The molecular interactions promoting miscibility between C and 0 and between C and