Preparation of simulated emulsion: A simulated emulsion of crude oil-water with fracturing was prepared in the laboratory by mixing of 10 mL oil phase and 10 mL water phase. The oil phase was a solution of crude oil or crude oil compositions in kerosene, and the water phase was simulated water (as shown in Table 1 with or without fracture additives. The mixture was sheared by an IKA T18 basic shear apparatus (IKA Corp., Germany) at 15,500 rpm for 5 min and then it was poured into a graduated test tube with a stopper. After shaking vigorously 200 times by hand, a stable simulated emulsion was formed [19J. Zhang, F. Yan, L. Li, Z. Fang, Z. Song, M. Yang, and J. Li, "Study of the interationship between structure of polyether and the demulsification of fractured emulsion", Acta Petrol. Sin., vol. 30, no. 3, pp. 548-554. [Petroleum Processing Section]., 21Y. Xie, F. Yan, Z. Yan, J. Zhang, and J. Li, "Demulsification and interfacial properties of crosslinking phenol-amine resin block polyether demulsifiers", J. Dispers. Sci. Technol., vol. 33, no. 12, pp. 1674-1681. [https://doi.org/10.1080/01932691.2011.635500].
[http://dx.doi.org/10.1080/01932691.2011.635500] ].

Table 1
Composition of simulated water.

Bottle Testing [19J. Zhang, F. Yan, L. Li, Z. Fang, Z. Song, M. Yang, and J. Li, "Study of the interationship between structure of polyether and the demulsification of fractured emulsion", Acta Petrol. Sin., vol. 30, no. 3, pp. 548-554. [Petroleum Processing Section]., 22M. Nikkhah, T. Tohidian, M.R. Rahimpour, and A. Jahanmiri, "Efficient demulsification of water-in-oil emulsion by a novel nano-titania modified chemical demulsifier", Chem. Eng. Res. Des., vol. 94, pp. 164-172.
[http://dx.doi.org/10.1016/j.cherd.2014.07.021] , 23J.D.L.C. Clavel, J.C. Navarro, and R. Martínez-Palou, "Demulsification of water-in-heavy crude oil emulsion using amphiphilic ammonium salts as demulsifiers", Tenside Surfact. Det., vol. 54, no. 4, pp. 361-364. [https://doi.org/10.3139/113.110510].
[http://dx.doi.org/10.3139/113.110510] ]: The separation of water/oil from the emulsions was evaluated by bottle testing (SY/T 5281-2000) at room temperature and the percentage of water/oil separation was visually quantified. The dehydration rate (DH) and deoiling rate (DO) were defined as equation (1):

(1)

Where X is the dehydration rate or deoiling rate which evaluates the stability of the emulsion. V is the volume of the water or oil separated; V₀ is the original volume of water or oil contained in the emulsion.

The microscope is a particularly intuitive approach for calibration of droplet size distributions [24J.A. Boxall, C.A. Koh, E.D. Sloan, A.K. Sum, and D.T. Wu, "Measurement and calibration of droplet size distributions in water-in-oil emulsions by particle video microscope and a focused beam reflectance method", Ind. Eng. Chem. Res., vol. 49, no. 3, pp. 1412-1418.
[http://dx.doi.org/10.1021/ie901228e] ]. The size distribution of water droplets in the emulsions (W/O) was measured using a BK-POLR polarizing microscope (Chongqing Ott Optical Instrument Co., Ltd.) at room temperature. A freshly prepared emulsion drop was put on a glass slide, which was placed on the objective table of microscope rapidly. The images of the emulsion on the glass plate were taken. Then the particle size distribution of the droplets was determined.

The interfacial tension is the basic parameter to evaluate the interfacial activity of the surface-active substance [25M. Bourrel, A. Graciaa, R.S. Schechter, and W.H. Wade, "The relation of emulsion stability to phase behavior and interfacial tension of surfactant systems", J. Colloid Interface Sci., vol. 72, no. 1, pp. 161-163.
[http://dx.doi.org/10.1016/0021-9797(79)90198-X] , 26F.O. Opawale, and D.J. Burgess, "Influence of interfacial properties of lipophilic surfactants on water-in-oil emulsion stability", J. Colloid Interface Sci., vol. 197, no. 1, pp. 142-150. [https://doi.org/10.1006/jcis.1997.5222].
[http://dx.doi.org/10.1006/jcis.1997.5222] ]. The dynamic and equilibrium interfacial tensions (IFT) were determined by a high precision spinning drop interfacial tension meter (JJ2000B, Power each Ltd., Shanghai China). A little drop of oil phase was injected into a rotating capillary tube filled with the aqueous phase. The diameter of the elongated oil drop at a fixed temperature (T=30 °C) and speed of rotation (r=5000 rpm) was monitored as functions of time until it reached a constant value. From the drop diameter’s value, the dynamic and equilibrium IFTs were calculated by the software attached.

The background of interfacial dilational viscoelasticity has been described in the literatures [27H. Zhou, Q. Luo, Q-T. Gong, Z-Y. Liu, M. Liu, L. Zhang, L. Zhang, and S. Zhao, "Interfacial dilational properties of di-substituted alkyl benzene sulfonates at kerosene/water and crude oil/water interfaces", Colloids Surf. Physicochem. Eng. Aspects, vol. 520, pp. 561-569.
[http://dx.doi.org/10.1016/j.colsurfa.2017.02.011] -29S.S. Hu, L. Zhang, X.L. Cao, L.L. Guo, Y.W. Zhu, L. Zhang, and S. Zhao, "Influence of crude oil components on interfacial dilational properties of hydrophobically modified polyacrylamide", Energy Fuels, vol. 29, no. 3, pp. 1564-1573.
[http://dx.doi.org/10.1021/ef5027407] ]. The interfacial viscoelasticity or interfacial dilational modulus (ε) is defined by

(2)

where γ is the interfacial tension, A is the drop area and dA is the deformation.

In this study, the interfacial dilational viscoelasticity was measured by a JMP 2000A interface expansion rheometer (Power each Ltd., Shanghai, China). It includes a Langmuir trough with two symmetrically oscillating barriers for changing the interfacial area and a Wilhelmy plate for measuring the interfacial tension [21Y. Xie, F. Yan, Z. Yan, J. Zhang, and J. Li, "Demulsification and interfacial properties of crosslinking phenol-amine resin block polyether demulsifiers", J. Dispers. Sci. Technol., vol. 33, no. 12, pp. 1674-1681. [https://doi.org/10.1080/01932691.2011.635500].
[http://dx.doi.org/10.1080/01932691.2011.635500] , 30T. Sun, L. Zhang, Y. Wang, S. Zhao, B. Peng, M. Li, and J. Yu, "Influence of demulsifiers of different structures on interfacial dilational properties of an oil–water interface containing surface-active fractions from crude oil", J. Colloid Interface Sci., vol. 255, no. 2, pp. 241-247. [https://doi.org/10.1006/jcis.2002.8661].
[http://dx.doi.org/10.1006/jcis.2002.8661] ]. The Langmuir trough was filled with 90 mL water phase and 60 mL oil phase (solution of crude oil in kerosene, 5 wt %). The dilational viscoelasticity experiment began after 6 h of pre-equilibrium. Then, the barriers were expanded and compressed sinusoidally with small amplitude in the frequency range of 0.0033 to 0.1 Hz. All experiments were performed at 30 °C.

Different oil has different characteristic and offers different problems [31M. Ali, and M. Alqam, "The role of asphaltenes, resins and other solids in the stabilization of water in oil emulsions and its effects on oil production in Saudi oil fields", Fuel, vol. 79, no. 11, pp. 1309-1316.
[http://dx.doi.org/10.1016/S0016-2361(99)00268-9] ]. So the influences of oil components on emulsion stability were investigated. The oil components were extracted via column chromatography. The influences of oil components on the stability of emulsion were shown in Fig. (1). As shown in Fig. (1), the emulsions prepared from saturates-water and aromatics-water showed poor stability, because both emulsions achieved complete demulsification in a short time (less than 30 min). The emulsion prepared from resins-water was more stable than those from saturates-water and aromatics-water. 88.5% deoiling rate was observed, and no free water was obtained in more than 2 h. The emulsion prepared from asphaltenes was the most stable one. Less than 70% deoiling rate was achieved, and also no free water was obtained in near 3 h. Therefore, it can be concluded that the stability of emulsion is strongly correlated to asphaltenes and resins, namely, the resin and asphaltene make an important contribution to the emulsion stability among the oil components. These results were consistent with the previous work of Xia et al. [32L. Xia, S. Lu, and G. Cao, "Stability and demulsification of emulsions stabilized by asphaltenes or resins", J. Colloid Interface Sci., vol. 271, no. 2, p. 504.
[http://dx.doi.org/10.1016/j.jcis.2003.11.027] ]. They studied the influence of asphaltenes and resins on the stability and demulsification of emulsions and found that asphaltenes and the resins stabilize the emulsions, while in the absence of asphaltenes and resins, no stable emulsions can be formed in the water-jet kerosene system. The influence of asphaltenes and resins on emulsion stability from Ref [32L. Xia, S. Lu, and G. Cao, "Stability and demulsification of emulsions stabilized by asphaltenes or resins", J. Colloid Interface Sci., vol. 271, no. 2, p. 504.
[http://dx.doi.org/10.1016/j.jcis.2003.11.027] ]. were shown in Fig. (2). This is because that asphaltenes and resins adsorbed on the water-oil interface to form a viscoelastic interfacial film with high mechanical strength.

Fig. (1) The emulsion stability of different oil components. Water phase: Simulated water; oil phase: crude oil components (a) deoiling rate, (b) dehydration rate.

The influences of fracture additives on the stability of emulsion were studied in the present work. Fracture additives such as hydroxypropyl guar, friction reducers (SL-P), clay stabilizer (FP-2) and pH adjuster (Na₂CO₃) were dissolved in water. Then the water phase was mixed with oil phase (5% crude oil in kerosene) by an IKA T18 basic shear apparatus.

Fig. (2) The influence of asphaltenes and resins on emulsion stability (comparing with Ref [32L. Xia, S. Lu, and G. Cao, "Stability and demulsification of emulsions stabilized by asphaltenes or resins", J. Colloid Interface Sci., vol. 271, no. 2, p. 504. [http://dx.doi.org/10.1016/j.jcis.2003.11.027] ].)

The effects of fracture additives on emulsion stability were illustrated in Fig. (3). As shown in Fig. (3b), there was no free water observed for each emulsion, that is the dehydration rates were all 0% for emulsions prepared from water phases containing fracture additives and crude oil. However, there were some differences in deoiling rates. As shown in Fig. (3a), the deoiling rates at 120 min were 61%, 60%, 51% and 35% for emulsions with clay stabilizer, friction reducer, pH adjuster and hydroxypropyl guar, respectively. The results suggested that the emulsion with hydroxypropyl guar was the most stable one. Ferrer et al. [33I. Ferrer, and E.M. Thurman, "Analysis of hydraulic fracturing additives by LC/Q-TOF-MS", Anal. Bioanal. Chem., vol. 407, no. 21, pp. 6417-6428.
[http://dx.doi.org/10.1007/s00216-015-8780-5] ] traced the chemical additives used in fracturing fluids by a high-resolution mass spectrometry coupled with liquid chromatography. Fragmentation of guar gum was detected in the flow back fluid. They supposed that guar gum may form a stable emulsion when dissolved in methanol or water, which was consistent with the present work. This may be due to enhancing the viscosity of the emulsion, and the amphiphilic properties of the hydroxypropyl guar. The improved viscosity of emulsion led to high frictional resistance and then prevented the collision between droplets of water droplets and slowing sinking speed to a large extent [17G.E. King, "Hydraulic fracturing 101: what every representative, environmentalist, regulator, reporter, investor, university researcher, neighbor, and engineer should know about hydraulic fracturing risk", J. Pet. Technol., vol. 64, no. 04, pp. 34-42.
[http://dx.doi.org/10.2118/0412-0034-JPT] ].

Fig. (3) The effect of fracture additives on the emulsion stability. Water phase: Simulated water with fracture additive; oil phase: 5% crude oil in kerosene (a) deoiling rate, (b) dehydration rate.

At the end of fracturing, the fracture fluid was broken by a breaker to reduce molecular weight and fluid viscosity. Therefore, the fluid that remains underground may mostly be the gel breaking liquid. Thus, it is important to investigate the influence of gel breaking liquid on the stability of emulsion. Herein, the influences of gel breaking liquid on the stability of emulsion prepared from oil components were explored as shown in Fig. (4). Similarly, the emulsions prepared from saturates-gel breaking liquid and aromatics-gel breaking liquid showed poor stability, because there were hardly any interfacial active substances in saturates and aromatics. The emulsions prepared from resins-gel breaking liquid and asphaltenes-gel breaking liquid were more stable. As shown in Fig. (4b), there were no free water phase separated for resins and asphaltenes in the experiment time range, and the deoiling rates for resins and asphaltenes were 79% and 58% in Fig. (4a), respectively. This result indicated that resin, asphaltene and gel breaking liquid, as well as their interaction, play important roles for the stability of emulsion. Zhang et al. [34T. Zhang, and K. Chen, "Effect of flowback fluid on crude oil demulsification in low permeability oil field and countermeasure", Oil-Gasfield Surface Engineering, vol. 35, no. 2, pp. 4-7. [http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yqtdmgc201602002].] studied the demulsification of flow back fluid produced after fracturing in a low permeability oil field in Bohai China. They found that the flow back fluid was hard to be broken due to the serious emulsion, which was consistent with our observations, and can be illustrated according to the conclusion of this study.

Fig. (4) The interactions between gel breaking liquid and ingredient of crude oil on emulsion stability. (a) deoiling rate, (b) dehydration rate

Fig. (5) The particle size distribution for various emulsion obtained by crude components. (a) saturates, (b) aromatics, (c) resins and (d) asphaltenes

The initial droplet size is an indicator of emulsion stability. Generally, the smaller the initial droplet size is, the more stable the emulsion is. In this study, the effect of crude oil compositions on the stability of the emulsion was studied from the perspective of visual particle size under a polarizing microscope.

The influences of components of crude oil on the particle size of emulsion were shown in Fig. (5). As shown, the saturate emulsifier system had large particle sizes which mainly in 44μm~130 μm and the corresponding emulsion is unstable. The droplets sizes were mainly in 22 μm~93 μm for emulsion prepared from aromatics. The large distributes of the droplets size indicated unstable emulsions of the saturate and the aromatic. Furthermore, a smaller-sized droplet was obtained in resin emulsion, and the particle sizes mainly ranged from 19 μm to 49 μm. The smallest-sized droplet was obtained in asphaltenes emulsion as shown in Fig. (5d). There were well-distributed particles, and the particle sizes were mainly distributed in 9 μm~27 μm, which suggested that the asphaltene emulsion was very stable. These results agreed with the results reported in emulsion stability from a macroscopic view (Section 3.1.3).

The effects of fracture additives on the particle size of emulsion were shown in Fig. (6). It can be clearly seen that well-distributed droplets were obtained in hydroxypropyl guar and pH adjuster (Na₂CO₃) emulsion, and the droplets sizes for both emulsions were mainly in 9 μm~32 μm. It is not strange to get a stable emulsion with pH adjuster (Na₂CO₃). Na₂CO₃ can react with petroleum acid in crude oil to present a kind of surfactant, which is the main substance for stable emulsion. On contrary, the particle sizes of emulsions prepared from friction reducer (SL-P) and clay stabilizer (FP-2) were much larger. The droplets sizes for each were 9 μm~51 μm and 17 μm~56 μm, respectively. Therefore, hydroxypropyl guar and pH adjuster (Na₂CO₃) in fracture additives play the main role for the stability of the emulsion.

Fig. (6) The particle size distribution for emulsion obtained by different fracture additives. (a) hydroxypropyl guar, (b) pH adjuster (Na2CO3), (c) friction reducer (SL-P) and (d) clay stabilizer (FP-2).

The effects of fracture additives on interfacial tension between crude oil (5% in kerosene) and simulation water were shown in Fig. (7). Each fracture additive was dissolved in simulation water at a concentration of 1000 mg/L in advance. Then the water phase with one of the fracture additives was poured into a rotating capillary tube. Subsequently, a little drop of oil phase (20 μL) was injected into the middle of the same tube. After sealed, the two liquid system rotated with the tube at a speed of 5000 rpm, and the oil drop was stretched. Finally, the diameter of the elongated oil drop was monitored as functions of time until it reached a constant value, and then the interfacial tensions were calculated by the software attached.

Fig. (7) Effects of fracture additives on interfacial tension.

It can be observed from Fig. (7), the control equilibrium IFT of oil-water was about 24.0 mN/m, and the equilibrium IFT was a little lower when hydroxypropyl guar was added in the water. The lowest interfacial tension was obtained by friction reducer, and the value of equilibrium IFT was 1.9 mN/m. This is because the friction reducer used in Shengli Oil Field is a kind of surfactant named SL-P. The friction reducer reduced the interface energy of oil-water, which is favor for the formation of emulsion. The pH adjuster (Na₂CO₃) reacts with petroleum acid to present petroleum carbonate, which is a natural surfactant. Therefore, the IFT was decreased to 6.4 mN/m after Na₂CO₃ was added into the water phase.

Dilational rheology are the main characteristics of the equilibrium and dynamic properties of a liquid film and has been proved to be a powerful tool to investigate the properties of adsorption films. Interfacial dilational properties play crucial roles in the formation and brake of crude oil emulsion, and the relationship between dilational data and emulsion stability had been discussed by Zhang et al. [35H.Q. Sun, L. Zhang, Z.Q. Li, L. Zhang, L. Luo, and S. Zhao, "Interfacial dilational rheology related to enhance oil recovery", Soft Matter, vol. 7, no. 17, pp. 7601-7611.
[http://dx.doi.org/10.1039/c1sm05234a] ].

In the present work, the influences of fracture additives on the interfacial film of crude oil and simulated water were studied by the dilational rheological method. The dilational modulus (ε) for fracture additives, such as hydroxypropyl guar, pH adjuster (Na₂CO₃), friction reducer (SL-P) and clay stabilizer (FP-2) were shown in Fig. (8). As shown, the dilational modulus (ε) increased gradually with time except for friction reducer. For the control film from crude oil and simulated water, the dilational modulus increased from around 6 mN/m to 20 mN/m within 100 min. This is because the active ingredients of oil (such as resins and asphaltenes) gradually adsorbed on the oil-water interface, and then formed a viscoelastic film. As shown in Fig. (8), the curve was not changed too much when clay stabilizer was added into the water phase. The reason is that clay stabilizer is usually ammonium or inorganic salt, such as tetramethyl ammonium chloride, potassium chloride or sodium chloride, which has little influence on the film of emulsion. However, the dilation modulus was increased significantly when hydroxypropyl guar was added into simulated water. The ε value increased from around 10 mN/m to 32 mN/m in 60 min, which indicated that the polymer of hydroxypropyl guar was adsorbed on the oil-water interface, and a viscoelastic film was formed. As referred by Zhang et al. [35H.Q. Sun, L. Zhang, Z.Q. Li, L. Zhang, L. Luo, and S. Zhao, "Interfacial dilational rheology related to enhance oil recovery", Soft Matter, vol. 7, no. 17, pp. 7601-7611.
[http://dx.doi.org/10.1039/c1sm05234a] ] that interfacial dilational rheological data can reflect some aspects of the nature of the film. A direct, reliable and qualitative link may exist between dilational data and the emulsion stability. Generally, coalescence can be described as the formation of a hole in a thin film. Interfacial dilational elasticity are thought to resist the deformation of the film. Therefore, a higher dilational modulus always leads to the formation of a more stable emulsion [36E. Santini, F. Ravera, M. Ferrari, M. Alfè, A. Ciajolo, and L. Liggieri, "Interfacial properties of carbon particulate-laden liquid interfaces and stability of related foams and emulsions", Colloids Surf. Physicochem. Eng. Aspects, vol. 365, no. 1-3, pp. 189-198.
[http://dx.doi.org/10.1016/j.colsurfa.2010.01.041] ]. From this point of view, hydroxypropyl guar may be the most important stabilizer among the fracture additives for the emulsion. As for pH adjuster, the dilational modulus of which was a little higher than that of control. This is because Na₂CO₃ reacted with –COOH on resins or asphaltenes in crude oil, and resulted in the formation of hydrophilic –COONa. The saponified resins or asphaltenes adsorbed onto the interface and resulted in the increase of dilational modulus of the film. On the contrary, the dilational modulus was decreased significantly when friction reducer was added into the water phase. This is because that friction reducer is a kind of low molecular weight surfactant. Generally, low molecular weight surfactants and surface active components in crude oil will dramatically reduce the dilational parameters of the interfacial film, which results from the fast diffusion-exchange process of molecules between the bulk and the interface.

Fig. (8) Dilational rheology of fracture additives. Oil phase =5% crude oil; Water phase= fracture additives