## Summary

The first-ever field pilot on Alaska North Slope (ANS) to validate using polymer floods for heavy-oil enhanced oil recovery is currently ongoing. One of the major concerns of the operator is the effect of polymer on oil/water‐separation efficiency after polymer breakthrough. This work investigates the influence of polymer on the separation behavior of heavy-oil emulsions and evaluates the performance of emulsion breakers (EBs). In this study, two types of heavy-oil emulsions were prepared and tested at 20 and 50% water cut (WC), respectively. The bottle test method was used in the experiments, in which the separated water volume with time, the separated water quality, and the volume fraction of phases were recorded. Results showed that polymer accelerated the oil/water separation acting as an emulsion inhibitor at 20% WC but tended to impede the water separation at 50% WC. Regardless of WC, polymer resulted in poor water quality and the formation of a stable intermediate oil in water (o/w) emulsion, because of the increased viscosity of the water phase. The performance of EBs showed a complex dependency on the WC, the type of demulsifier and dosage, and the polymer concentration. Despite the varied conditions encountered in the heavy-oil/water/polymer/demulsifier system, a compound EB achieved satisfactory demulsification performance, showing the highest potential for deployment in the current ANS polymer flooding pilot. In this paper, we systematically studied the potential influence of polymer breakthrough on the separation behavior of heavy-oil emulsion on ANS for the first time. The findings of this study will provide practical guidance in advance for produced fluid treatment of the ongoing first-ever polymer flooding pilot on ANS.

## Introduction

ANS contains vast viscous and heavy-oil resources ranging between 20 and 25 billion barrels, which are primarily concentrated in West Sak (also called Schrader Bluff) and Ugnu reservoirs (Targac et al. 2005). The development pace of these resources has been slow due to various factors, such as high costs, low oil recovery using conventional techniques, and the inapplicability of thermal methods given the presence of continuous permafrost. Polymer flooding has a great potential to improve oil recovery from ANS's heavy-oil reservoirs due to better mobility control and higher oil sweep efficiency based on initial scoping studies. Currently, the US Department of Energy and Hilcorp is sponsoring the first‐ever advanced polymer flooding pilot test in the Schrader Bluff viscous oil reservoir with an in‐situ oil viscosity of 330 cp, and the current concentration of the injected polymer is 1,700 ppm. However, using polymers in heavy-oil enhanced oil recovery presents a potential problem because this could upset the separation facilities and disrupt existing field operations by negatively influencing the efficiency of oil/water separation after polymer breakthrough into production systems.

With the worldwide application of polymer flooding, the oil/water separation characteristics after the polymer breakthrough have been extensively addressed by field engineers and researchers (Feng et al. 1994; Deng et al. 2002; Wang et al. 2009; Zheng et al. 2011; Wylde et al. 2013; Al Kalbani et al. 2014; Chen et al. 2015a; Liu et al. 2015). Two common emulsions, both water in oil (w/o) and o/w emulsions, have been studied to illustrate the potential problems of polymer flooding. As for the o/w emulsion, it is generally believed that the presence of polymer can result in highly stable emulsion by increasing produced water viscosity, electronegativity, and the strength of interfacial film between oil and water, thus preventing coalescence of oil droplets (Wu et al. 1999; Argillier et al. 2014; Chen et al. 2015b; Liu et al. 2015). Moreover, polymer molecules and the suspended particles, such as clay, carried by polymer flooding can synergistically absorb on the oil/water interface to further stabilize the o/w emulsion (Wang et al. 2012; Li et al. 2014; Liu et al. 2014). However, the influence of polymer on the stability of w/o emulsion is much more complicated, depending on various factors, such as the crude oil composition, property of polymer, and the presence of surfactant. For example, the studies of Kang et al. (2011) and Liu et al. (2015) showed that polymer enhances the stability of w/o (light oil) emulsion by forming a more rigid oil/water interfacial film. In addition, Dalmazzone et al. (2012), Argillier et al. (2013), and Sjöblom et al. (2017) found the polymer itself had no significant influence on the stability of w/o (heavy-oil) emulsion because they determined that the polymer had no interaction with the surface‐active agents at the oil/water interface. It was also found that in the presence of surfactant, the polymer can contribute to the stable intermediate w/o emulsion at low WC (Dalmazzone et al. 2012; Argillier et al. 2013). Some studies reported that especially the hydrophilic polymer could favor oil/water separation by reducing the stability of the emulsion (Wu et al. 1999; Lin et al. 2008; Argillier et al. 2014; Al‐Kayiem and Khan 2017). The unpredictable impact of polymer on w/o emulsion brings a significant challenge in evaluating the influence of polymer on actual oil/water separation in oilfield production systems because the actual produced fluids are composed of both w/o and o/w emulsions. Furthermore, in some studies (Wu et al. 1999; Lin et al. 2008; Argillier et al. 2013), using solvent‐diluted crude oil or synthetic water excluding some chemical additives used in oil fields, such as drilling fluid additive or corrosion inhibitor, may not represent the separation behavior of actual crude oil emulsion. Therefore, the real oil/water separation issues should be investigated as a whole using the on‐site crude oil, produced water, and the injected polymer under specific oilfield conditions to obtain reliable conclusions.

Treating produced liquid from polymer flooding has become a great challenge in oil fields. Many new techniques, such as the electrical dehydrator coupled with gravity settling (Liu et al. 2009, 2015), novel crossflow oil/water separator (Deng et al. 2002), new hydrocyclone (Liu et al. 2007), and new flotation device (Chen et al. 2015a), have been developed to solve the oil/water separation problems. However, chemical methods (that is, demulsifiers and water treatment chemistries) are still the most popular techniques because of their effectiveness, ease of implementation, and low cost. Fang et al. (2014) evaluated the demulsification efficiency of cationic surfactants (that is, alkyltrimethylammonium bromides), which were applied to treat produced fluids directly obtained from a heavy-oil polymer flooding project in China. They found that C14TAB and C16TAB can effectively break both w/o and o/w emulsions existing in the produced fluids because they were found to be able to reduce interfacial film strength and neutralize the surface charge. Also, researchers have been making an effort to develop new demulsifiers to treat the chemical enhanced‐oil‐recovery‐induced emulsions. For example, Duan and his group (Duan et al. 2014; Zhang et al. 2015) developed a new type of nonionic demulsifier that could effectively treat the o/w emulsion directly produced from an offshore polymer flooding project. Li et al. (2016) found a novel polyether demulsifier, TPEA19920, which exhibited an excellent performance to treat the produced heavy-oil emulsions. Even though the development of demulsifiers has made significant progress, there are still no demulsifiers with universal applicability to process the produced fluids. In general, targeted demulsifiers need to be developed or selected through extensive evaluation tests because they are highly sensitive to the properties of the crude oil, produced water, applied chemicals, and the actual operating conditions. However, most of the studies focused on investigating the same types of demulsifiers and mainly used separation efficiency or water clarity to evaluate the demulsification performance (Duan et al. 2014; Fang et al. 2014; Li et al. 2016).

Extensive studies have been conducted to investigate the emulsification and demulsification phenomenon. However, few studies are reported to investigate the effect of polymer on the separation behavior of heavy-oil emulsion from ANS. Herein lies the purpose of this work. In this paper, we used the traditional bottle test method to investigate the influence of polymer on the separation of produced liquid prepared in the laboratory using actual heavy oil, produced water, and the injected polymer obtained from the pilot test site on ANS and evaluated the performance of several commercial demulsifiers. Various influence factors, including WC, polymer concentration, demulsifier type and concentration, and compound demulsifier, have been investigated. In addition, multiple parameter indicators are developed to evaluate the performance of different types of demulsifiers. A radar chart composed of demulsification efficiency, separation speed, water clarity, and optimized dosage has been developed to evaluate the demulsification performance. Finally, the most applicable demulsifier has been proposed for field applications.

## Experimental Method

### Materials

The produced fluid was sampled directly from the wellhead of a production well on a polymer flooding pilot site, which was initially separated by thermal settling. The upper heavy oil was further dehydrated by centrifuge until no water separated, whereas the bottom‐produced water was filtered by 0.2-µm water‐wet filter paper through the vacuum filtration method to remove the suspended oil and particulate matter. The properties of heavy oil and produced water are listed in Tables 1 and 2, respectively. The polymer used was Flopaam™ 3630, having a molecular weight of 18 to 20 MM Dalton and 30% hydrolysis, Commercial EBs (product codes E12085A, E18276A, N1691, and R01319) were used in this study. The names of the four EBs are also abbreviated as E12, E18, N16, and R13, respectively. All chemical reagents, such as petroleum ether, methanol, and xylene were used as received.

Table 1

Properties of heavy oil.

Table 1

Properties of heavy oil.

Table 2

Properties of produced water.

Table 2

Properties of produced water.

### Polymer Solution Preparation

A polymer mother solution of 1,000 ppm concentration was prepared by dissolving a certain amount of polymer powder into the filtered produced water under gentle stirring at room temperature for 24 hours until no fish eyes were remaining. The mother solution was diluted with the filtered produced water to obtain the desired polymer solution with concentrations of 150, 400, and 800 ppm. The polymer concentrations were determined to cover the potential viscosity range of produced water, which would be produced from the pilot well. The apparent viscosity of each polymer solution was measured by a commercial viscometer as a function of shear rate at 130°F.

### Demulsifier Solution Preparation

A 10,000 mg/L demulsifier original solution was prepared by dissolving a certain amount of demulsifier into a specific solvent. Oil‐soluble demulsifiers, E12085A, E18276A, and R01319, were dissolved into xylene, and the water‐soluble demulsifier, N1691, was dissolved into methanol.

### Emulsion Preparation

The separated heavy oil and the aqueous phase (produced water or polymer solution) were added to a beaker according to designed ratios (20 and 50% WC) and preheated to 130°F in the water bath to mimic actual field conditions. The WC is defined as the volume ratio of the aqueous phase to the total volume of emulsion. A commercial digital homogenizer was used to stir the mixture at 5,000 rev/min for 3 minutes to prepare the emulsion.

### Microscopic Observation

As for the oil‐continuous emulsion, to obtain high‐quality microscopic images, the prepared emulsion was first diluted with white oil by 50 times. Subsequently, a small drop of the diluted emulsion was immediately transferred by a syringe onto a glass slide for taking the microscopic image without using a cover slide. As for the water‐continuous emulsion, a small drop of the prepared emulsion was immediately transferred by a syringe onto a glass slide, and a cover slide was placed on the top of the emulsion before taking the microscopic image to avoid evaporation of the aqueous phase. A commercial upright compound microscope with fluorescence was used to observe the micromorphology of the prepared sample. Several images from each slide were captured to ensure uniformity.

### Bottle Test Method

In gravitational separation tests, the emulsion was transferred to a graduated tube immediately after preparation. The tube was placed into the water bath to observe the separation characteristics of emulsion under the action of gravity. In chemical demulsification tests, after transferring the prepared emulsion, a certain amount of demulsifier solution was added to the test tube based on the volume of the emulsion. The tube was shaken by hand for 3 minutes to mix the emulsion and demulsifier thoroughly and then kept in the water bath for the chemical separation test. All the separation tests were performed at the temperature of 130°F. During the separation process, the separated water volume and the volume fraction of each phase (oil, emulsion, and aqueous phase) were recorded with respect to time. In addition, the sharpness of the interface between the oil phase and the separated water phase and the clarity of the separated water was observed. Eventually, the oil content in separated water (OIW) was measured after 24 hours using a commercial ultraviolet visible spectrophotometer.

## Results and Discussion

### Viscosity of Polymer Solutions

The viscosity of the prepared polymer solutions as a function of shear rate is shown in Fig. 1. The polymer solution at each concentration can be characterized as non‐Newtonian fluid showing shearing‐thinning behavior as the viscosity of the polymer solution decreased with the shear rate. The viscosity/shear‐rate decline trend clearly complies with the power‐law model, which is consistent with the study of Gao (2013). It is also obvious that higher polymer concentration led to higher viscosity.

Fig. 1

Viscosity of polymer solutions as a function of shear rate.

Fig. 1

Viscosity of polymer solutions as a function of shear rate.

### Gravitational Separation Behavior

The micromorphology and separation behavior of heavy-oil emulsions were studied. During the microscopic observation, the impact of the WC was evaluated. In the study of separation behavior, water‐separation kinetics, OIW, and volume fraction of the separated phases were analyzed at various WC and polymer concentrations.

#### Micromorphology of Emulsion

Fig. 2  shows the micromorphology of the emulsion generated at 20 and 50% WC with no addition of polymer. At 20% WC, the w/o emulsion was observed with water droplets dispersed in the oil phase as seen from Fig. 2a. At 50% WC, the droplets in forms of w/o emulsion were trapped in the water phase, forming complex water in oil in water emulsions as indicated by Fig. 2b.

Fig. 2

Microscopic images of emulsion at (a) 20% WC, and (b) 50% WC.

Fig. 2

Microscopic images of emulsion at (a) 20% WC, and (b) 50% WC.

#### Effect of Polymer on Separation Kinetics

Water‐separation kinetics can be used to describe the emulsion stability; the faster the water can separate, the less stable the emulsion is (Nguyen and Sadeghi 2011). Fig. 3  shows the impact of polymer concentration on separation kinetics at 20 and 50% WC, in which the markers are the measured data points, and the corresponding regression curves represent the separation trend. To better characterize the separation behavior, the influence of polymer on OIW and volume fraction of the separated phases are shown in Figs. 4 and 5, respectively.

Fig. 3

Effect of polymer on separation kinetics at (a) 20% WC, and (b) 50% WC.

Fig. 3

Effect of polymer on separation kinetics at (a) 20% WC, and (b) 50% WC.

Fig. 4

Effect of polymer on water quality at (a) 20% WC, and (b) 50% WC.

Fig. 4

Effect of polymer on water quality at (a) 20% WC, and (b) 50% WC.

Fig. 5

Effect of polymer on the volume fraction of phases at (a) 20% WC, and (b) 50% WC.

Fig. 5

Effect of polymer on the volume fraction of phases at (a) 20% WC, and (b) 50% WC.

With no addition of polymer (blank sample), the emulsion generated at 20% WC was much more stable than that at 50% WC. The w/o emulsion generated at 20% WC was stable for at least 24 hours (not shown in the plot but was observed); however, the water in oil in water emulsion generated at 50% WC was quite unstable, separating into two layers in less than 5 minutes.

For emulsions at 20% WC, as shown in Fig. 3a, the water separation was accelerated in the presence of polymer compared with the blank sample. The destabilization effect of polymer for the w/o emulsion has been previously reported due to the increased coherent energy of polymer molecules (Wu et al. 1999). It could also be attributed to the increased density of polymer solution (Sjöblom et al. 2017). According to the Stokes equation
$ν=Δρgd218η,$
(1)
where ν is the sedimentation or creaming rate, Δρ is the density difference between oil and water, d is the diameter of the dispersed droplets, and η is the viscosity of the continuous phase. Because the viscosity for the continuous oil phase is constant, the increased density of the polymer solution contributes to a larger Δρ, facilitating the sedimentation of water droplets. To be noted in Fig. 3a is the water separation in excess of 100% at 400 and 800 ppm polymer concentrations, which is attributed to the intermediate layer (it is believed to be an o/w emulsion layer, see Fig. 5) that was included in the separated water volume because the interface between the water layer and the intermediate layer cannot be well defined.

It is also noteworthy that although the water separation was enhanced with the addition of polymer at 20% WC compared with the blank sample, the amount of separated water at first increased with increasing polymer concentration until a critical point of 400 ppm was reached and then decreased. Even though such a critical point has not been reported previously, this trend may be due to the phase inversion (the w/o emulsion could be converted to o/w emulsion), resulting from the increase of polymer concentration (Preziosi et al. 2013; Liu et al. 2015). Thus, in this study, the water‐continuous emulsion could be generated starting from the polymer concentration of 400 ppm at 20% WC. Then, the reduced water separation after the critical point could be attributed to the increased viscosity of the continuous phase (i.e., water phase).

For the water‐continuous emulsion at 50% WC, as shown in Fig. 3b, the presence of polymer impeded the phase separation, which could be attributed to the increasing viscosity of the polymer solution as described in Fig. 1. Whereas, there was not much difference in the final water‐separation volume as the residence time extended to 30 minutes, implying extending the settlement time is one potential method to minimize the stabilization effect of polymer. To be noted, the final separation volume (i.e., at 30 minutes) with the presence of polymer was slightly higher than that of the emulsion without polymer. It is because of the massive oil droplets in the separated water, which could affect the measurement of the separated water volume.

#### Effect of Polymer on Separated Water Quality

The effect of polymer on the separated water quality after 24 hours is displayed in Fig. 4. In general, the polymer had a negative effect on the separated water quality, and the OIW significantly increased with increasing polymer concentration. That is because adding polymer could generate more stable o/w emulsion in the separated water phase (Liu et al. 2015). When the OIW at tested polymer concentrations is compared, the oil content at 50% WC is nearly double than at 20% WC, which implies that the water treatment would be much more challenging at higher WC. That is because at higher WC, the distance between oil droplets becomes larger, which makes it harder to collide and coalesce.

#### Effect of Polymer on the Volume Fraction of Phases

It is worth mentioning that the addition of polymer resulted in a concentrated o/w emulsion layer sandwiched between the top oil layer (w/o emulsion) and the bottom separated water layer during the oil/water separation process, as shown in Fig. 5. According to our observation, the interfaces between this intermediate layer and the top layer or bottom layer became clearer with prolonged settling time. After 24 hours, this intermediate emulsion layer still existed, and the intermediate layer at 20% WC was thicker than that at 50% WC. At the same WC, the more the separated water, the thicker the intermediate layer, which was more difficult to eliminate. The thickness change of this layer with respect to time was also related to the WC. The intermediate layer became thinner with time at 20% WC but slightly thicker at 50% WC. The formation of the middle emulsion layer may result from the higher viscosity of the aqueous phase, generating a concentrated stable o/w emulsion.

The positive effect of polymer for a w/o emulsion by promoting phase separation acting as an EB has been reported by Lin et al. (2008) and Argillier et al. (2013). Different from their findings, the negative effect of polymer on the separated water quality and the formation of intermediate o/w emulsion layer were observed in our study. Those phenomena, which were not previously reported, may be due to using solvent‐diluted heavy oil or synthetic water rather than the actual crude oil or produced water. It has been reported that the addition of solvent could improve the quality of the separated oil and water (Opawale 2009).

### Chemical Demulsification without Polymer

In most oil fields, the typical requirement for the total separation process is to produce dry oil and clean water. Dry oil should contain no more than 0.3% water by volume, and clean water should have an oil content of less than 100 ppm, preferably 50 ppm (Hirasaki et al. 2010). The separation process on ANS consists of a slug catcher (130°F), a heater, and a separator (170°F). In our experiment, we mimicked the first‐stage separation in the slug catcher, the demulsification efficiency of more than 90%, and oil content of less than 100 ppm*.

#### Performance of Individual Demulsifiers

Fig. 6  shows the performance of four types of demulsifiers with a dosage of 100 ppm at 20% WC. As can be seen, three oil‐soluble demulsifiers, E12085A, E18276A, and R01319, promoted oil/water separation, and clear separated water was obtained with an oil content less than 20 ppm. The demulsification efficiency of the three oil‐soluble demulsifiers at 12 hours in decreasing order was E18276A, E12085A, and R01319. It also can be seen that, as for E18276A, water separated rapidly in the first 3 hours and then slowly until it reached the plateau after 6 hours. However, as for E12085A and R01319, the time needed to reach the plateau was 10 hours, indicating a slower separation rate. In terms of OIW shown in Fig. 6b, E18276A yielded the lowest value. Therefore, among the four evaluated demulsifiers, E18276A is the most effective EB from the perspective of demulsification efficiency, separation rate, and OIW. Whereas, in the case of water‐soluble demulsifier, N1691, no water separation was observed because N1691 can rarely be soluble in the continuous oil phase, causing the barriers to the diffusion and adsorption process of N1691 molecules on the oil/water interface (Kang et al. 2018). Note that even though E18276A had the most efficient performance, it did not render complete separation of the water phase in the measured time scale.

Fig. 6

Performance of demulsifiers on oil/water separation at 20% WC: (a) separation kinetics, and (b) oil content in the separated water.

Fig. 6

Performance of demulsifiers on oil/water separation at 20% WC: (a) separation kinetics, and (b) oil content in the separated water.

The performance of the four demulsifiers at a dosage of 100 ppm for emulsion generated at 50% WC is shown in Fig. 7. As can be seen from Fig. 3b, the water‐separation efficiency of the blank sample could achieve 90% within 5 minutes and eventually could reach approximately 93% at 30 minutes. However, as seen from Fig. 7a, with the addition of four EBs, all the water‐separation efficiencies of the four tested samples were lower than 85% at 30 minutes. Thus, the addition of four EBs seemed to impede the separation. By comparing Fig. 4b (i.e., the oil content at 0 ppm polymer concentration) and Fig. 7b, it can be seen that the addition of the four demulsifiers could help to reduce OIW. However, N1691 had an advantage over the other three oil‐soluble demulsifiers in terms of separation speed, resulting in higher separation efficiency within 60 minutes because the unstable water in oil in water emulsion was converted to a stable w/o emulsion after the addition of three oil‐soluble demulsifiers, as shown in Fig. 8. Thus, the slower separation occurred due to the high viscosity of the continuous heavy-oil phase. This conversion might be attributed to the strong lipophilicity of the oil‐soluble demulsifiers (Lv et al. 2014). However, the lack of knowledge of the properties of the tested demulsifiers does not allow us to pinpoint the precise underlying reasons. Comparing Figs. 6a and 7a, it can be seen that although R01319 was not as efficient as E18276A for emulsion generated at 20% WC, R01319 had a higher separation speed than E18276A at 50% WC, implying R01319 has poor diffusivity in oil but could disrupt the interfacial film effectively. Note that if the settlement time was 4 hours, E12085A performed better than N1691 and R01319 with the highest efficiency and lowest oil content in water. Generally, no EB can perform well for emulsion at 50% WC from the standpoint of the three criteria.

Fig. 7

Performance of demulsifiers on oil/water separation at 50% WC: (a) separation kinetics, and (b) oil content in the separated water.

Fig. 7

Performance of demulsifiers on oil/water separation at 50% WC: (a) separation kinetics, and (b) oil content in the separated water.

#### Effect of Demulsifier Dosage

The demulsifier dosage is one important factor affecting the demulsification performance and thus in determining the overall cost of demulsification in the application. Besides, overdosage of demulsifiers may lead to counterproductive effects, such as resulting in more stable w/o emulsion and producing stable reverse o/w emulsion (Manning and Thompson 1995). As shown in Fig. 6, the most efficient demulsifier, E18276A, did not yield a separation efficiency of more than 90% at the dosage of 100 ppm. Therefore, the influence of the demulsifier dosage of E18276A on oil/water separation was investigated for which the results are shown in Fig. 9. Although increased dosage enhanced the demulsification efficiency, it was offset by higher OIW.

Fig. 8

Microscopic images of emulsions: (a) without demulsifier, (b) with E12085A, (c) with E18276A, and (d) with R01319.

Fig. 8

Microscopic images of emulsions: (a) without demulsifier, (b) with E12085A, (c) with E18276A, and (d) with R01319.

Fig. 9

Effect of dosage on the performance of E18276A for emulsions at 20% WC: (a) separation kinetics, and (b) oil content in the separated water.

Fig. 9

Effect of dosage on the performance of E18276A for emulsions at 20% WC: (a) separation kinetics, and (b) oil content in the separated water.

#### Performance of the Compound Demulsifiers

Multiple compound EBs were also proposed and evaluated to find the most efficient combination, which can make use of the advantage of individual EBs to a maximum extent. Based on the analysis for each EB performance in previous tests, E12085A produced the best water quality and the highest separation efficiency if the residence time was long enough; E18276A had the optimum performance for w/o emulsion; N1691 was capable of breaking the emulsion at 50% WC in a short residence time at low dosage; and R01319 had better ability to disrupt the interfacial film. Thus, the performance of compound EBs, E12 + N16, E12 + R13, and E12 + E18, were investigated at 100 ppm, as shown in Fig. 10. E12 + E18 is the most promising compound EB, which achieved a faster and more‐efficient separation than either E12085A or E18276A at the same dosage. E12 + N16 increased the separation efficiency from 65 to 80% at 2 hours compared with E12085A itself and yielded better water quality than either E12085A (Fig. 10b) or N1691 (Fig. 7b). Moreover, the separation efficiency of compound demulsifiers, both E12 + E18 and E12 + N16, achieved 90% at the end of the tests. However, the separation efficiency of the compound demulsifier of E12 + R13 was approximately 80% at the end of the test. Thus, it is concluded that the demulsification performance of E12 + R13 was not as efficient as that of E12 + E18 and E12 + N16.

Fig. 10

Performance of compound demulsifiers at 50% WC: (a) separation kinetics, and (b) oil content in the separated water.

Fig. 10

Performance of compound demulsifiers at 50% WC: (a) separation kinetics, and (b) oil content in the separated water.

To provide a comprehensive analysis, the performances of the qualified EBs for emulsion at 50% WC without polymer are compared in the radar chart (Fig. 11) in terms of demulsification efficiency, water clarity, separation speed, and dosage. These four parameters are defined as the dimensionless numbers in the following equations:
$Efficiency=VsVt,$
(2)
$Water clarity=Ca−CmCa,$
(3)
$Separation speed=Tm−TeTm,$
(4)
$Dosage=Cmax−CCmax,$
(5)
where $Vs$ is the separated water volume at 4 hours; $Vt$ is the total water volume; $Ca$ is the allowed oil content in water, typically less than 100 ppm; $Cm$ is the measured oil content in water after 24 hours; $Tm$ is the measurement time for separation, which is 4 hours; $Te$ is the time required to reach the separation equilibrium; and $Cmax$ is the maximum dosage of EB used in the experiment.
Fig. 11

Radar chart for the performance of demulsifiers at 50% WC.

Fig. 11

Radar chart for the performance of demulsifiers at 50% WC.

According to the definition, the larger the enclosed area in the radar chart, the better the performance of the corresponding EB. It is obviously seen from Fig. 11 that E12 + E18, E12 + N16, and N1691 are the better‐performing demulsifiers. E12 + E18 is the most efficient EB in terms of dosage and separation efficiency, E12 + N16 is the most efficient EB in terms of water clarity, and N1691 is the most efficient EB in terms of dosage and separation speed.

### Chemical Demulsification in the Presence of Polymer

In the presence of polymer, the performance of the four individual EBs and the aforementioned compound EBs were evaluated, and the EB with the optimum performance was proposed. The effect of polymer concentration on the proposed EB was also investigated.

#### Performance of Individual Demulsifiers

The performance of the four demulsifiers with the dosage of 100 ppm was evaluated for the emulsion with 150 ppm polymer at 20% WC. Fig. 12a shows that all three oil‐soluble demulsifiers performed better than the water‐soluble demulsifier. According to the observation during the separation process, the three oil‐soluble demulsifiers could effectively eliminate the intermediate layer described in Fig. 5 and obtain a clear water phase. As can be seen from Fig. 12, E18276A had the most efficient performance in terms of separation efficiency, separation speed, and OIW; however, the dosage appeared to be insufficient to achieve a separation efficiency more than 90%.

Fig. 12

Performance of demulsifiers for emulsion with 150 ppm polymer at 20% WC: (a) separation kinetics, and (b) oil content in the separated water.

Fig. 12

Performance of demulsifiers for emulsion with 150 ppm polymer at 20% WC: (a) separation kinetics, and (b) oil content in the separated water.

The performance of the EBs with the dosage of 50 ppm for the emulsion with 800 ppm polymer at 50% WC is shown in Fig. 13. All the four EBs achieved a separation efficiency of more than 90% in less than 15 minutes, and the three oil‐soluble EBs achieved an OIW lower than 50 ppm. However, the water‐soluble EB, N1691, yielded an unsatisfactory OIW of 160 ppm. Therefore, the application of N1691 after polymer breakthrough may be limited due to its poor performance. E12085A and R01319 are highly competent for emulsions in the presence of polymer at 50% WC.

Fig. 13

Performance of demulsifiers for emulsion with 800 ppm polymer at 50% WC: (a) separation kinetics, and (b) oil content in the separated water.

Fig. 13

Performance of demulsifiers for emulsion with 800 ppm polymer at 50% WC: (a) separation kinetics, and (b) oil content in the separated water.

#### Performance of the Compound Demulsifier

The compound EBs E12 + E18 and E12 + N16 were also evaluated for emulsion with 150 ppm polymer at 20% WC. The performance of the compound EBs is shown in Fig. 14. Because there was no water separation for E12 + N16, Fig. 14 only shows the performances of E12 + E18 in comparison with E18276A. At a dosage of 100 ppm, the separation efficiency of E12 + E18 was slightly lower than that of E18276A. However, when the dosage was more than 100 ppm, E12 + E18 yielded a similar separation efficiency and lower OIW.

Fig. 14

Performance of compound demulsifiers for emulsion with 150 ppm polymer at 20% WC.

Fig. 14

Performance of compound demulsifiers for emulsion with 150 ppm polymer at 20% WC.

As mentioned in the previous analysis (Fig. 13), two individual EBs E12085A and R01319 are most applicable to treat emulsion with 800 ppm polymer at 50% WC. The performances of compound EBs E12 + E18 and E12 + N16 are compared with those of E12085A and R01319 by the radar chart in Fig. 15. It can be seen that E12 + E18 exhibited the optimum performance.

Fig. 15

Performance of demulsifiers for emulsion with 800 ppm polymer at 50% WC.

Fig. 15

Performance of demulsifiers for emulsion with 800 ppm polymer at 50% WC.

#### Effect of Polymer Concentration

Based on the preceding analysis, compound EB E12 + E18 has the potential to be used for emulsions generated at both 20 and 50% WC. Because the varying WC and polymer concentration after breakthrough can change the demulsifier demand, the influence of polymer concentration on the performance of demulsification was also investigated at the two WCs. The effect of polymer concentration on the demulsification performance of E12 + E18 for emulsion at 20% WC is shown in Fig. 16. In this case, the polymer concentration had no obvious effect on the demulsification performance. E12 + E18 with a dosage of 500 ppm could help to achieve the separation requirements at varying polymer concentrations. It may be expected that the required EB dosage be increased or a new demulsifier may be needed when the polymer concentration is more than 800 ppm due to the decrease of demulsification efficiency and increase of OIW.

Fig. 16

Effect of polymer concentration on demulsification for emulsion at 20% WC.

Fig. 16

Effect of polymer concentration on demulsification for emulsion at 20% WC.

For the emulsion generated at 50% WC, the effect of polymer concentration on the demulsification performance of E12 + E18 is shown in Fig. 17. It is obvious that at the same dosage of 50 ppm, the demulsification efficiency was higher in the presence of polymer. In other words, a higher demulsifier dosage would be required in the case of no polymer.

Fig. 17

Effect of polymer concentration on demulsification for emulsion at 50% WC.

Fig. 17

Effect of polymer concentration on demulsification for emulsion at 50% WC.

## Conclusions

In the framework of this study, the gravitational and chemical separation behavior of heavy-oil emulsion prepared with actual heavy oil and produced water from the ANS polymer flooding pilot site was investigated by the bottle test method against various influence factors. The main conclusions are as follows:

For the gravitational separation, polymer generally favored the oil/water separation for emulsion at 20% WC but in a complex manner. The maximum water volume was separated when the polymer concentration reached 400 ppm, but further increasing polymer concentration reduced the water separation. For emulsion at 50% WC, the emulsion became more stable with increasing polymer concentration, which resulted from the increased viscosity of polymer solution. It is also observed that polymer resulted in poor water quality and contributed to the formation of a stable intermediate o/w emulsion layer regardless of WC, which was also attributed to the increased viscosity of the aqueous phase.

For the chemical demulsification, oil‐soluble demulsifiers exhibited better performance than water‐soluble demulsifiers for emulsions both with and without polymer at 20% WC. E18276A was found to be the most effective EB at the dosage of 500 ppm in terms of separation efficiency, separation speed, and water clarity. As for emulsion at 50% WC, no individual EB could perform well from the standpoint of all three criteria. In this case, the water‐soluble demulsifier N1691 exhibited the fastest separation in the absence of polymer; whereas the presence of polymer reduced the separation speed of N1691 and resulted in higher OIW by preventing the coalescence of oil droplets. E12085A yielded better performance in terms of separation efficiency and water quality for emulsion at 50% WC, irrespective of the polymer; while the disadvantage of E12085A lay in the slowest separation when the polymer was absent.

In spite of the complicated interactions involved in the system of heavy oil, produced water, polymer, and demulsifier, the compound EB E12 + E18 was found to be the most effective demulsifier, which could be potentially applied to the ANS polymer flooding pilot. Implicitly, we consider the satisfactory demulsification performance at intricate operational conditions as one of the key supporting contributors to the primary objective of improving heavy-oil recovery by polymer flooding.

As a novelty in this study, the radar chart, which is composed of demulsification efficiency, separation speed, water clarity, and optimized dosage, has been proposed to evaluate the performance of different types of demulsifiers in an intuitive but comprehensive way. This evaluation method establishes an excellent example for future similar studies. More importantly, the findings of this study have implications for the treatment of produced fluid from the first‐ever polymer flooding pilot on ANS.

## Disclaimer

This report was prepared as an account of work sponsored by an agency of the United States Government under DOE Cooperative Agreement DE‐FE0031606. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

## Nomenclature

• $Ca$

allowed oil content in water, µg/g, ppm

•
• $Cm$

measured oil content in water after 24 hours, µg/g, ppm

•
• $Cmax$

maximum dosage of EB used in the experiment, µg/g, ppm

•
• d

diameter of the dispersed droplets, m

•
• g

acceleration of gravity, m/s2

•
• OIW

oil content in the separated water, µg/g, ppm

•
• ppm

parts per million, mg/L or µg/g

•
• $Te$

time required to reach the separation equilibrium, h

•
• $Tm$

measurement time for separation, h

•
• v

sedimentation or creaming rate, m/s

•
• $Vs$

separated water volume at 4 hours, cm3

•
• $Vt$

total water volume, cm3

•
• WC

water cut in percentage, dimensionless

•
• Δρ

density difference between oil and water, kg/m3

•
• η

viscosity of the continuous phase, kg/(m·s), Pa·s

*

Personal communication with Samson Ning, 5 July 2019. Prudhoe Bay, Alaska, USA: Hilcorp.

## Acknowledgments

This material is based upon work supported by the Department of Energy, Office of Fossil Energy, administered by the National Energy Technology Laboratory, under Award Number DE‐FE0031606. We would also like to thank Hilcorp Alaska, LLC and BP Exploration (Alaska) Inc. for cosponsoring this project. We also thank Baojun Bai, Randy Seright, Brent Sheets, and the rest of the team associated with this project for their valuable suggestions.

## References

Al Kalbani
,
H.
,
Mandhari
,
M. S.
,
,
H.
et al. .
2014
.
Impact on Crude Dehydration due to Back Production of Polymer
.
Paper presented at the SPE EOR Conference at Oil and Gas West Asia
,
Muscat, Oman
,
31 March–2 April
.
SPE-169718-MS
. https://doi.org/10.2118/169718-MS.
Al-Kayiem
,
H. H.
and
Khan
,
J. A.
2017
.
Evaluation of Alkali/Surfactant/Polymer Flooding on Separation and Stabilization of Water/Oil Emulsion by Statistical Modeling
.
Energy Fuels
31
(
9
):
9290
9301
. https://doi.org/10.1021/acs.energyfuels.7b01662.
Argillier
,
J. F.
,
Dalmazzone
,
C.
,
Henaut
,
I.
et al. .
2013
.
Methodological Approach for Analyzing the Impact of Chemical EOR on Surface Processes
.
Paper presented at the SPE International Symposium on Oilfield Chemistry
,
The Woodlands, Texas, USA
,
8–10 April
.
SPE-164098-MS
. https://doi.org/10.2118/164098-MS.
Argillier
,
J.-F.
,
Henaut
,
I.
,
Noik
,
C.
et al. .
2014
.
Influence of Chemical EOR on Topside Produced Water Management
.
Paper presented at the SPE Improved Oil Recovery Symposium
,
Tulsa, Oklahoma, USA
,
12–16 April
.
SPE-169067-MS
. https://doi.org/10.2118/169067-MS.
Chen
,
H.-X.
,
Tang
,
H.-M.
,
Duan
,
M.
et al. .
2015a
.
Oil-Water Separation Property of Polymer-Contained Wastewater from Polymer-Flooding Oilfields in Bohai Bay, China
.
Environ Technol
36
(
11
):
1373
1380
. https://doi.org/10.1080/09593330.2014.990522.
Chen
,
H.-X.
,
Tang
,
H.-M.
,
Gong
,
X.-P.
et al. .
2015b
.
Effect of Partially Hydrolyzed Polyacrylamide on Emulsification Stability of Wastewater Produced from Polymer Flooding
.
J Pet Sci Eng
133
:
431
439
. https://doi.org/10.1016/j.petrol.2015.06.031.
Dalmazzone
,
C.
,
Noik
,
C.
, and
Argillier
,
J.-F.
2012
.
Impact of Chemical Enhanced Oil Recovery on the Separation of Diluted Heavy Oil Emulsions
.
Energy Fuels
26
(
6
):
3462
3469
. https://doi.org/10.1021/ef300083z.
Deng
,
S.
,
Bai
,
R.
,
Chen
,
J. P.
et al. .
2002
.
Produced Water from Polymer Flooding Process in Crude Oil Extraction: Characterization and Treatment by a Novel Crossflow Oil-Water Separator
.
Sep Purif Technol
29
(
3
):
207
216
. https://doi.org/10.1016/S1383-5866(02)00082-5.
Duan
,
M.
,
Ma
,
Y.
,
Fang
,
S.
et al. .
2014
.
Treatment of Wastewater Produced from Polymer Flooding Using Polyoxyalkylated Polyethyleneimine
.
Sep Purif Technol
133
:
160
167
. https://doi.org/10.1016/j.seppur.2014.06.058.
Fang
,
S.
,
Duan
,
M.
,
Long
,
W.
et al. .
2014
.
Treatment of Produced Liquid from Polymer Flooding by Using Cationic Surfactant
.
Sep Sci Technol
49
(
3
):
432
437
. https://doi.org/10.1080/01496395.2013.819580.
Feng
,
J.
,
Zhao
,
J.
, and
Chan
,
S.
1994
.
Water/Oil Separation Characteristics of Daqing Oilfield Polymer Flooding Production Fluid
.
Paper presented at the SPE Annual Technical Conference and Exhibition
,
New Orleans, Louisiana, USA
,
25–28 September
.
SPE-28540-MS
. https://doi.org/10.2118/28540-MS.
Gao
,
C.
2013
.
Viscosity of Partially Hydrolyzed Polyacrylamide Under Shearing and Heat
.
J Pet Explor Prod Technol
3
:
203
206
. https://doi.org/10.1007/s13202-013-0051-4.
Hirasaki
,
G. J.
,
Miller
,
C. A.
,
Raney
,
O. G.
et al. .
2010
.
Separation of Produced Emulsions from Surfactant Enhanced Oil Recovery Processes
.
Energy Fuels
25
(
2
):
555
561
. https://doi.org/10.1021/ef101087u.
Kang
,
W.
,
Bin
,
X.
,
Wang
,
Y.
et al. .
2011
.
Stability Mechanism of W/O Crude Oil Emulsion Stabilized by Polymer and Surfactant
.
Colloid Surf A
384
(
1–3
):
555
560
. https://doi.org/10.1016/j.colsurfa.2011.05.017.
Kang
,
W.
,
Yin
,
X.
,
Yang
,
H.
et al. .
2018
.
Demulsification Performance, Behavior and Mechanism of Different Demulsifiers on the Light Crude Oil Emulsions
.
Colloid Surf A
545
:
197
204
. https://doi.org/10.1016/j.colsurfa.2018.02.055.
Li
,
F.
,
Sun
,
L.
,
Wang
,
Y.
et al. .
2014
.
Effect of Laponite Particles on the Emulsion Stability of Produced Water from Polymer Flooding
.
J Pet Sci Eng
124
:
155
160
. https://doi.org/10.1016/j.petrol.2014.10.010.
Li
,
Z.
,
Yin
,
S.
,
Tan
,
G.
et al. .
2016
.
Synthesis and Properties of Novel Branched Polyether as Demulsifiers for Polymer Flooding
.
Colloid Polym Sci
294
:
1943
1958
. https://doi.org/10.1007/s00396-016-3956-x.
Lin
,
M.
,
Zhang
,
C.
,
Zong
,
H.
et al. .
2008
.
Influence of Polymers on the Stability of Gudao Crude Oil Emulsions
.
Pet Sci
5
:
159
162
. https://doi.org/10.1007/s12182-008-0024-2.
Liu
,
H.
,
Li
,
J.
,
Yan
,
J.
et al. .
2009
.
Successful Practices and Development of Polymer Flooding in Daqing Oilfield
.
Paper presented at the Asia Pacific Oil and Gas Conference & Exhibition
,
Jakarta, Indonesia
,
4–6 August
.
SPE-123975-MS
. https://doi.org/10.2118/123975-MS.
Liu
,
S.
,
Zhao
,
X.
,
Dong
,
X.
et al. .
2007
.
Experimental Research on Treatment of Produced Water from a Polymer-Flooding Process Using a Double-Cone Air-Sparged Hydrocyclone
.
SPE Proj Fac & Const
2
(
3
):
1
5
.
SPE-95343-PA
. https://doi.org/10.2118/95343-PA.
Liu
,
Y.
,
Wang
,
Z.
,
Li
,
X.
et al. .
2014
.
ASP Flooding Produced Water Management: Evaluation, Disposal and Reuse
.
Paper presented at the SPE Middle East Health, Safety, Environment & Sustainable Development Conference and Exhibition
,
Doha, Qatar
,
22–24 September
.
SPE-170396-MS
. https://doi.org/10.2118/170396-MS.
Liu
,
Y.
,
Wang
,
Z.
,
Zhuge
,
X.
et al. .
2015
.
Study on Emulsification Behavior and Optimized Separation Technology of High Concentration Polymer Flooding Produced Liquid in Daqing Oilfield
.
Paper presented at the SPE Middle East Oil & Gas Show and Conference
,
Manama, Bahrain
,
8–11 March
.
SPE-172768-MS
. https://doi.org/10.2118/172768-MS.
Lv
,
G.
,
Wang
,
F.
,
Cai
,
W.
et al. .
2014
.
Characterization of the Addition of Lipophilic Span 80 to the Hydrophilic Tween 80-Stabilized Emulsions
.
Colloid Surf A
447
:
8
13
. https://doi.org/10.1016/j.colsurfa.2014.01.066.
Manning
,
F. S.
and
Thompson
,
R. E.
1995
.
Oilfield Processing of Petroleum: Crude Oil
, Vol. 2.
Tulsa, Oklahoma, USA: PennWell Books
.
Nguyen
,
D. T.
and
,
N.
2011
.
Selection of the Right Demulsifier for Chemical Enhanced Oil Recovery
.
Paper presented at the SPE International Symposium on Oilfield Chemistry
,
The Woodlands, Texas, USA
,
11–13 April
.
SPE-140860-MS
. https://doi.org/10.2118/140860-MS.
Opawale
,
A.
2009
.
Oilfield Emulsion Control: A Major Issue during Heavy Crude Oil Production
.
Paper presented at the Nigeria Annual International Conference and Exhibition
,
Abuja, Nigeria
,
3–5 August
.
SPE-128352-MS
. https://doi.org/10.2118/128352-MS.
Preziosi
,
V.
,
Perazzo
,
A.
,
Caserta
,
S.
et al. .
2013
.
Phase Inversion Emulsification
.
Chem Eng Trans
32
:
1585
1590
. https://doi.org/10.3303/CET1332265.
Sjöblom
,
J.
,
Dagsgard
,
C.
,
Simon
,
S.
et al. .
2017
.
Influence of HPAM on W/O Emulsion Separation Properties
.
J Dispers Sci Technol
38
(
2
):
206
215
. https://doi.org/10.1080/01932691.2016.1154864.
Targac
,
G. W.
,
Redman
,
R. S.
,
Davis
,
E. R.
et al. .
2005
.
Unlocking the Value in West Sak Heavy Oil
.
Paper presented at the SPE International Thermal Operations and Heavy Oil Symposium
,
,
1–3 November
.
SPE-97856-MS
. https://doi.org/10.2118/97856-MS.
Wang
,
D.
,
Dong
,
H.
,
Lv
,
C.
et al. .
2009
.
Review of Practical Experience by Polymer Flooding at Daqing
.
SPE Res Eval & Eng
12
(
3
):
470
476
.
SPE-114342-PA
. https://doi.org/10.2118/114342-PA.
Wang
,
Y.
,
Lu
,
F.
,
Li
,
Y.
et al. .
2012
.
Effects of Na–Montmorillonite Particles on the Emulsification Stability of Polymer Flooding Produced Water
.
Colloid Surf A
410
:
125
129
. https://doi.org/10.1016/j.colsurfa.2012.06.030.
Wu
,
D.
,
Ai
,
G.
,
Chen
,
Y.
et al. .
1999
.
Rheology and Stability of IOR Produced Liquid in Daqing Oilfield
.
Paper presented at the SPE Asia Pacific Improved Oil Recovery Conference
,
Kuala Lumpur, Malaysia
,
25–26 October
.
SPE-57318-MS
. https://doi.org/10.2118/57318-MS.
Wylde
,
J. J.
,
Slayer
,
J. L.
, and
Barbu
,
V.
2013
.
Polymeric and Alkali-Surfactant Polymer Enhanced Oil Recovery Chemical Treatment: Chemistry and Strategies Required After Breakthrough into the Process
.
Paper presented at the SPE International Symposium on Oilfield Chemistry
,
The Woodlands, Texas, USA
,
8–10 April
.
SPE-164078-MS
. https://doi.org/10.2118/164078-MS.
Zhang
,
J.
,
Jing
,
B.
,
Fang
,
S.
et al. .
2015
.
Synthesis and Performances for Treating Oily Wastewater Produced from Polymer Flooding of New Demulsifiers Based on Polyoxyalkylated N,N-Dimethylethanolamine
.
26
(
2
):
190
197
. https://doi.org/10.1002/pat.3433.
Zheng
,
F. F.
,
Quiroga
,
P.
, and
Sams
,
G.
2011
.
Challenges in Processing Produced Emulsion from Chemical Enhanced Oil Recovery-Polymer Flood Using Polyacrylamide
.
Paper presented at the SPE Enhanced Oil Recovery Conference
,
Kuala Lumpur, Malaysia
,
19–21 July
.
SPE-144322-MS
. https://doi.org/10.2118/144322-MS.